Alpha-mannosidase-2 modulates arbovirus infection in a pathogen- and Wolbachia-specific manner in Aedes aegypti mosquitoes

Abstract

Multiple Wolbachia strains can block pathogen infection, replication, and/or transmission in Aedes aegypti mosquitoes under both laboratory and field conditions. However, Wolbachia effects on pathogens can be highly variable across systems and the factors governing this variability are not well understood. It is increasingly clear that the mosquito host is not a passive player in which Wolbachia governs pathogen transmission phenotypes; rather, the genetics of the host can significantly modulate Wolbachia-mediated pathogen blocking. Specifically, previous work linked variation in Wolbachia pathogen blocking to polymorphisms in the mosquito alpha-mannosidase 2 (αMan2) gene. Here we use CRISPR-Cas9 mutagenesis to functionally test this association. We developed αMan2 knockouts and examined effects on both Wolbachia and virus levels, using dengue virus (DENV; Flaviviridae) and Mayaro virus (MAYV; Togaviridae). Wolbachia titers were significantly elevated in αMan2 knockout (KO) mosquitoes, but there were complex interactions with virus infection and replication. In Wolbachia-uninfected mosquitoes, the αMan2 KO mutation was associated with decreased DENV titers, but in a Wolbachia-infected background, the αMan2 KO mutation significantly increased virus titers. In contrast, the αMan2 KO mutation significantly increased MAYV replication in Wolbachia-uninfected mosquitoes and did not affect Wolbachia-mediated virus blocking. These results demonstrate that αMan2 modulates arbovirus infection in Ae. aegypti mosquitoes in a pathogen- and Wolbachia-specific manner, and that Wolbachia-mediated pathogen blocking is a complex phenotype dependent on the mosquito host genotype and the pathogen. These results have significant impact for the design and use of Wolbachia-based strategies to control vector-borne pathogens.

Graphical Abstract

• Mosquito genetic background can modulate Wolbachia pathogen-blocking phenotypes. The Alpha-mannosidase-2 gene has previously been identified as a potential candidate.

• αMan2 CRISPR KO mutagenesis suppresses dengue virus titers but increases Mayaro virus titers, in Wolbachia-uninfected mosquitoes

• In a Wolbachia-infected background, αMan2 KO ablates Wolbachia pathogen blocking but has no effect on Mayaro virus titers.

Introduction

Dengue virus (DENV) (genus Flavivirus, family Flaviviridae) is an important human pathogen that is transmitted primarily by Aedes aegypti mosquitoes (Bhatt et al., 2013). Mayaro virus (MAYV) (genus Alphavirus, family Togaviridae) is an emerging human pathogen that is transmitted mainly by Haemagogus janthinomys mosquitoes (Pujhari et al., 2022); however, Ae. aegypti mosquitoes are also competent vectors for this virus (Pereira et al., 2020). There are no approved vaccines or specific antivirals to prevent and manage disease outbreaks that are caused by either virus and thus novel strategies for disease control are needed to combat arbovirus infections. The use of the intracellular invertebrate-specific bacterium Wolbachia as a biological control agent against Ae. aegypti has emerged as an innovative vector control strategy to reduce arbovirus transmission. Wolbachia bacteria are useful because, when incorporated into Ae. aegypti mosquitoes, it suppresses vector populations via a reproductive manipulation called cytoplasmic incompatibility (CI) (Beckmann et al., 2019; Sicard et al., 2019; Caragata et al., 2021) and also prevents replication of viruses inside mosquitoes, a trait known as pathogen blocking (PB), thereby limiting subsequent virus transmission to humans (Caragata et al., 2021).

Wolbachia-mediated pathogen blocking (PB) phenotypes in mosquitoes depend not just on the infecting Wolbachia strain, but also on many other factors including pathogen, infection type (natural vs. artificial), environmental conditions, and, importantly, host genetics (Ford et al., 2019; Ford et al., 2020; Liang et al., 2022). For example, mosquitoes infected with the wAlbB Wolbachia strain exhibited better arbovirus blocking when the mosquito nuclear genome was derived from Singapore compared to Mexico-derived (Liang et al. 2022). Ford et al. found enough standing genetic variation in Australian Ae. aegypti to select for significant weakening of PB within a few generations of artificial selection, suggesting that the host genetic background can have a strong effect on PB (Ford et al., 2019). Identified candidate mosquito host genes for this modulation were not the canonical suspects of mosquito innate immunity or detoxification; rather, they were primarily related to cell adhesion, Notch signaling, and cell cycle (Ford et al., 2019; Ford et al., 2020) highlighting our current lack of mechanistic understanding of the PB phenomenon.

Ford et al. identified single nucleotide polymorphisms in the non-coding region of the alpha-mannosidase 2 (αMan2) gene that were strongly associated with PB strength in Wolbachia-infected Ae. aegypti mosquitoes selected for high vs. low Wolbachia-mediated PB of DENV (Ford et al., 2019). αMan2 is putatively involved in protein glycosylation (Nemčovičová et al., 2013), and thus could alter PB by modulating viral glycosylation. Protein glycosylation, the enzymatic attachment of oligosaccharide structures to the peptide backbone, is an important post-translational modification for both host cell and viral proteins (Rogers and Heise, 2009; Schiller et al., 2012; Yap et al., 2017). In eukaryotic cells, glycosylation is responsible for many functions, including proper protein folding, trafficking, stability, receptor-ligand recognition, and cell adhesion (Schiller et al., 2012). Viruses do not have their own protein glycosylation machinery and employ host cellular enzymes for this purpose (Yap et al., 2017). Glycosylation of viral proteins plays a crucial role in the lifecycles of dengue and other viruses, influencing virus infectivity, pathogenicity, and host immune responses (Vigerust and Shepherd, 2007; Rogers and Heise, 2009). Enzymes involved in protein glycosylation are important potential targets to control viral replication in eukaryotic cells (Chang et sl., 2013; Pérez-García et al., 2017). However, how specific genes in these pathways affect arboviral replication in mosquito vectors is poorly understood.

We previously attempted to validate the role of αMan2 in wAlbB-mediated Wolbachia pathogen blocking using RNAi but we were unable to knock down expression of this gene in Wolbachia-infected mosquitoes (Sigle et al. 2022). Therefore, in this study we used CRISPR-Cas9 gene editing to ablate the αMan2 gene in Ae. aegypti and examined effects of gene knock-out (KO) on mosquito vector competence for DENV and MAYV in both Wolbachia-infected and uninfected mosquitoes. Results demonstrated complicated interactions between gene KO, Wolbachia infection, and viral pathogen, highlighting the complex nature of Wolbachia PB phenotypes.

Materials and Methods

Cells:

African green monkey kidney (Vero, ATCC CCL-81) cells were obtained from ATCC (Manassas, VA, USA) and maintained in Dulbecco’s Modified Eagle Medium (DMEM) (Gibco/Thermo Fisher, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco/Thermo Fisher), 100 μg/mL of streptomycin (Gibco/Thermo Fisher) and 100 units/mL of penicillin (Gibco/Thermo Fisher) at 37°C in 5% CO₂. Aedes albopictus cells (C6/36) were obtained from Sigma-Aldrich, St. Louis, MO, USA, and maintained in RPMI 1640 medium (Gibco/Thermo Fisher) supplemented with 10% FBS (Gibco/Thermo Fisher), 100 μg/mL of streptomycin (Gibco/Thermo Fisher) and 100 units/mL of penicillin (Gibco/Thermo Fisher) at 28°C.

Viruses:

MAYV strain BEAN343102 (GenBank: KP842802.1) was obtained from BEI Resources, NIAID, NIH (Manassas, VA, USA). To produce MAYV stocks, virus was propagated on Vero cells for 24 hours and stored at − 80°C. DENV serotype 2 strain JAM 1409 (Bennett et al., 2002) was propagated on C6/36 cells for 7 days as previously described (Terradas et al., 2017). MAYV stocks were initially quantified by plaque assay, while DENV stocks were initially quantified by qPCR. For all mosquito infection experiments, viruses were quantified by focus-forming assay (FFAs; see below for specific methods).

Antibodies:

Mouse monoclonal anti-alphavirus antibodies (G77L) (#MA5-18173) were obtained from Thermo Fisher and used in FFAs to detect MAYV at a dilution of 1:40 and incubated at 4 °C overnight. Mouse monoclonal anti-flavivirus group antigen antibodies, clone D1-4G2-4-15 (produced in vitro) (NR-50327) were obtained from BEI Resources, NIAID, NIH (Manassas, VA, USA). These antibodies were used in FFAs for the detection of DENV antigens at a dilution of 1:500 and incubated at 4 °C overnight. Goat anti-mouse IgG (H+L) highly cross-adsorbed secondary antibodies, Alexa Fluor 488 (A-11029) were purchased from Invitrogen. Secondary antibodies were used in FFAs at a dilution of 1:1,000 and incubated at a room temperature for at least 3 hours or at 4 °C overnight.

Plaque assay for the quantification of MAYV stocks:

For quantification of MAYV viral stocks, Vero cells were seeded in 6-well plates at a density of 5×10⁵ cells/well. Ten-fold serial dilutions of virus stocks were prepared in PBS and 200 uL of these dilutions were used for infections. Cells were infected for 1 hour at 37°C, infectious media removed, and cells covered with 1 mL of complete DMEM medium containing 0.5% agarose. Three days post-infection, cells were fixed with 4% formaldehyde (Sigma-Aldrich) in phosphate-buffered saline (PBS) (Gibco/Thermo Fisher) for 25 min, agarose covers were removed, and cells were stained for 5 min using aqueous solution containing 1% crystal violet and 20% ethanol to visualize plaques. For mosquito experiments, virus was quantified by FFA (see below).

qPCR for the quantification of DENV stocks:

Viral RNA was purified using Direct-zol RNA kit (Zymo Research) according to the manufacturer’s instructions and used as template in qPCRs. All primer sequences are in Table S1. qPCRs were set up using TaqMan^™ Fast Virus 1-Step Master Mix (Thermo Fisher) and run on an ABI 7500 Fast Real-time PCR System (Applied Biosystems/Thermo Fisher). The thermocycling conditions were as follows: 50 °C for 5 min; 95 °C for 20 s; 35 cycles of 95 °C for 3 s; 60 °C for 30 s; 72 °C for 1 s; and 40 °C for 10 s. Product was detected by measuring the fluorescence signal from the FAM reporter. A standard reference curve of known quantities of a DENV-2 genomic fragment was used for absolute quantification by qPCR. The DENV-2 genomic fragment was inserted into a plasmid and transformed into E. coli as previously described (Terradas et al., 2017). The linearized and purified fragment was serially diluted ranging from 10⁷- 10² copies and were used to create a standard curve of DENV amplification. The standard curve was run in duplicate on each 96-well plate, and the limits of detection were set at 10² copies. For mosquito experiments, virus was quantified by FFA (see below).

Mosquito rearing:

Ae. aegypti mosquitoes (Liverpool genetic background) expressing Cas9 protein in the germline (AAEL006511-Cas9; Li et al., 2017) were used. Ae. aegypti mosquitoes stably infected with the wAlbB strain of Wolbachia (backcrossed into the Merida, Mexico genetic background; Sigle et al., 2022) were provided by Prof. Zhiyong Xi, Michigan State University. Mosquitoes were reared at the PSU Millennium Sciences Complex insectary under the following environmental conditions: 27±1°C, 12:12 hours light:dark diurnal cycle, 80% relative humidity. For reproduction, mosquitoes were maintained on expired anonymous human blood using a 37 °C water-jacketed membrane feeder. Larvae were fed on koi fish pellets (TetraPond). Adult mosquitoes were maintained on 10% sucrose solution.

Preparation of single guide RNAs (sgRNAs):

The αMan2 Entrez Gene ID 5564678 gene sequence was used as a reference to design sgRNAs using CRISPOR (Concordet et al., 2018). sgRNAs were produced using overlapping nucleotides with the MegaScript T7 (Invitrogen/Thermo Fisher) in vitro transcription system. PCR templates for sgRNAs were produced using Phusion High-Fidelity DNA polymerase. The thermocycling conditions were as follows: 98 °C for 20 s; 35 cycles of 98 °C for 1 min s; 58 °C for 1 min; 72 °C for 1 min; and a final extension of 72°C for 7 min. Oligonucleotide sequences are given in Table S1. PCR products were purified using NucleoSpin Gel and PCR Clean-Up kit (Takara Bio, Kusatsu, Shiga, Japan), and 600 ng-1 μg of DNA templates were added to set up in vitro transcription reactions. Reactions were run for 16 hours at 37°C, treated with Turbo DNAse according to manufacturers’ instructions and purified using the MegaClear column purification kit (Thermo Fisher). The purified sgRNAs were tested with an in vitro cleavage assay. To produce a DNA template, genomic DNA (gDNA) from Ae. aegypti mosquitoes was purified using E.Z.N.A. MicroElute Genomic DNA Kit (Omega Bio-tek, Norcross, GA, USA) and the target region was amplified using Phire Animal Tissue Direct PCR Kit (Thermo Fisher) as described below. Reactions containing DNA template, individual sgRNAs and Cas9 protein in 1X NEB 3.1 buffer (New England Biolabs, Ipswich, MA, USA) were incubated at 37°C for 2 h, and diagnostic bands were visualized by electrophoresis on 1% agarose gel. sgRNAs for Ae. aegypti embryo injections were used at concentrations ranging between 70 ng/uL-180 μg/uL.

Embryo injections and establishment of knock-out (KO) mosquito lines:

Four to 5 days after blood feeding, 5–10 mated Wolbachia-uninfected females were placed into a Drosophila vial with damp cotton and filter paper and placed in the dark for 50 min to stimulate oviposition. To generate heritable mutations in Ae. aegypti mosquitoes, mixtures of selected sgRNAs were injected into pre-blastoderm-stage embryos of Cas9-expressing mosquitoes 1–2 hours after laying. Briefly, embryos were aligned (with posterior poles on one side) on damp filter paper using a paintbrush, transferred on a glass slide using double-sided scotch tape, dried for 1 min and covered with a mixture of Halocarbon 700 oil and Halocarbon 27 oil (1:1) to prevent further desiccation. Embryos were injected into the posterior poles with quartz needles (QF100-70-10, Sutter Instrument, Novato, CA, USA) pulled by a Sutter P2000 needle puller (program 50, HEAT=500, FIL=5, VEL=50, DEL=128, PUL=0), using a Femtojet injector (Eppendorf, Hamburg, Germany) and an InjectMan micromanipulator using the following settings: injection pressure (pi) 1,000 hPa, compensation pressure (pc) 700 hPa, injection time (manual mode) 2–3 sec. After injection, embryos were transferred on the damp filter paper into egg cups with wet cotton, kept in the humid insectary for 4–5 days, and then hatched. Injected embryos (G₀) that hatched and survived until adulthood were crossed individually. Legs of G₁ mosquitoes were individually screened by PCR for the presence of deletions in the target gene as described below. A single heterozygous founder mosquito was outcrossed with wild-type age-matched Ae. aegypti mosquitoes to establish a KO mosquito line (see Results). As the target gene was located in the chromosome 1, the mutation was sex-linked (Hall et al., 2015). As a result, to obtain homozygous mutants of both sexes, the selection process relied on chromosome recombination and identification of recombinant mosquitoes.

Mosquito screenings for mutations:

To screen live mosquitoes for the presence of deletions in the target gene, Phire Animal Tissue Direct PCR Kit (Thermo Fisher) was used according to manufacturers’ instructions. Briefly, mosquitoes were anesthetized on ice, a leg from each mosquito was removed using sharp forceps and immersed into 20 ul of sample dilution buffer supplemented with 0.5 ul of DNA release reagent. Leg samples in dilution buffer were incubated for 3 min at 98°C then used in PCR reactions. Primer sequences are provided in Table S1.

To characterize the αMan2 mutation at the transcript level, total RNA from αMan2 KO and wild-type mosquitoes was purified using E.Z.N.A. Total RNA kit (Omega Bio-tek) and cDNA synthesized using a gene-specific reverse primer and SuperScript III First-Strand Synthesis System (Thermo Fisher) according to manufacturers’ instructions. For cDNA synthesis, negative control gDNA from KO and wild-type mosquitoes was purified using E.Z.N.A MicroElute Genomic DNA Kit (Omega Bio-tek). PCR reactions were performed using Phire Animal Tissue Direct PCR Kit as described above. Information on primer sequences is in Table S1. For the detection of deletions in target genes, PCR products were separated by 2% agarose gel electrophoresis. Samples that separated into multiple bands were considered likely to contain an indel. The presence of αMan2 deletion(s) in both DNA and mRNA was then confirmed by PCR and direct sequencing of the target region.

Generation of Wolbachia-infected lines.

Wolbachia-infected Ae. aegypti females were crossed with KO Wolbachia-negative male mosquitoes. Homozygous mutant Wolbachia-infected males and heterozygous Wolbachia-infected females were isolated and further crossed to obtain homozygous Wolbachia-infected male and female mosquitoes. Homozygous mutant Wolbachia-infected male and female mosquitoes were isolated crossed to establish a pure homozygous Wolbachia-infected αMan2 KO mosquito line. Wolbachia-negative male mosquitoes from the parental Cas9-expressing Ae. aegypti line, which was used for embryo injections, were crossed with Wolbachia-infected Ae. aegypti females to obtain a Wolbachia-infected wild-type control line with comparable genetic background for infection experiments.

Quantification of relative Wolbachia density.

Total DNA was extracted from Wolbachia-infected mosquito homogenates (females, 7 days post-bloodfeed; 11–12 days post-emergence) using a E.Z.N.A. Tissue DNA Kit (Omega Bio-Tek) kit according to the manufacturer’s instructions. qPCR was performed using PerfeCTa SYBR Green FastMix (Quantabio, Beverly, MA, USA) on a Rotor-Gene Q qPCR machine (Qiagen, Hilden, Germany) under the following thermocycling conditions: 95 °C for 2 min for initial denaturation; 40 cycles at 95 °C for 10 s, 60 °C for 40 s, 72 °C for 30 s for DNA amplification and data acquisition; 55–99 °C (5 s per increment) for the melt curve analysis. Relative Wolbachia densities were obtained by normalizing Wolbachia titers to the RpS17 gene levels as described previously (Ford et al., 2019). Primer sequences are provided in Table S1. Crosses to introgress KO mutations into the Wolbachia-infected background are described in Results and Fig 1C.

Figure 1.

Overview of the approach used to generate Ae. aegypti strains used in this study. A) Generation of αMan2 KO mutations in Wolbachia-uninfected mosquitoes using CRISPR-Cas9 mutagenesis. B) Crossing scheme to generate a homozygous αMan2 KO line. C) Crossing scheme to generate a homozygous αMan2 KO line infected with Wolbachia.

Vector competence studies:

Four-to-five-day old female mosquitoes were blood fed for approximately 1 hour on infected human blood containing 10⁷ infectious MAYV particles per mL or 10⁶ infectious DENV particles per mL. After blood feeding, mosquitoes were anesthetized on ice and fully engorged females were transferred into cardboard cages; unfed females were discarded. Seven days post infection, mosquitoes were anesthetized using triethylamine (Sigma-Aldrich) and processed for vector competence assays. Although the KO line was fertile and viable, the mosquitoes were very fragile and died easily, and thus we were not able to examine timepoints beyond 7 days. Mosquitoes were forced to salivate for 30 min into glass capillaries filled with a mix of 50% sucrose solution and FBS (1:1) to collect saliva samples. Body (infection), legs (dissemination), and saliva (transmission) were then separately immersed in diluent solution containing 10% of FBS, 100 μg/mL of streptomycin, 100 units/mL of penicillin, 50 μg/mL gentamicin, and 2.5 μg/mL Amphotericin B in PBS. Body and legs samples were further homogenized by a single zinc-plated, steel, 4.5 mm bead using TissueLyser II (Qiagen) at 30 Hz for 2 min and centrifuged at 3,500 rpm at 4°C for 7 min in a bench top centrifuge to clear the homogenates. Samples were stored at −80°C. Virus titers in collected samples were determined by FFAs.

Focus-forming assay for the quantification of MAYV and DENV:

FFA was used to quantify both DENV and MAYV in all mosquito infction experments. Vero or C6/36 cells were seeded in 96-well plates at a density of 3×10⁴ cells/well or 3×10⁵ cells/well for the titration of MAYV or DENV, respectively. Ten-fold serial dilutions (in serum-free medium) of virus samples obtained from mosquito bodies and legs were prepared and 30 uL of each were used in assays. Saliva samples were not further diluted. Cells were infected for 1 hour at 37°C or 28°C for MAYV and DENV assays, respectively. Infectious solutions were then removed and cells covered with 100 uL of complete growth medium (DMEM or RPMI) containing 0.8% methylcellulose (Sigma-Aldrich) and incubated at their respective temperatures. After 24 hours for MAYV or 3 days for DENV assays, overlay medium was removed, cells were fixed with 4% formalin (Sigma-Aldrich) in PBS (Gibco/Thermo Fisher) for 15 min and permeabilized with 0.2% TritonX in PBS for 15 min. Primary antibodies were diluted in PBS and incubated overnight at 4°C. Secondary antibodies were incubated overnight at 4°C for MAYV assays and 3 hours at room temperature for DENV. After the final wash, cells were dried briefly, and MAYV or DENV foci immediately counted using an Olympus BX41 inverted microscope equipped with an UPlanFI 4X objective and a FITC filter. Dissemination rates only included mosquitoes that became infected. Transmission rates only included mosquitoes that became disseminated.

Statistical analysis:

All experiments were replicated twice. Infection, dissemination, and transmission rates were analyzed using contingency tables. Data on Wolbachia titers were analyzed by Mann-Whitney U tests. Due to violation of the equal variance assumption, data on viral titers were analyzed using the Brown, Forsythe ANOVA method with Welch’s correction for multiple tests.

Results

Generation of Wolbachia-negative and Wolbachia-positive αMan2 KO Ae. aegypti mosquitoes:

To generate a deletion in the αMan2 gene, Wolbachia-uninfected Ae. aegypti embryos expressing Cas9 protein (G₀, N=115) were injected with a mix of four sgRNAs targeting exon 5 of the gene. Surviving G₀ individuals (females N=19, males N=10) were outcrossed to wild-type mates (1 male per 1–2 females), females blood fed, eggs collected, and hatched in small batches for further screening (Fig. 1A). Forty G₁ male mosquitoes were individually screened by PCR for the presence of deletions in the target gene. Three G₁ males with αMan2 deletions were identified: two with an identical 13 nt deletion and one with a double deletion allele consisting of a 46 nt deletion at one sgRNA target site and a 9 nt deletion at another sgRNA target site (55 nt deletion total) (Supplementary Figure 1A). This 55 nt deletion was predicted to result in a 155-amino acid-long truncated protein instead of a 1174-amino acid long functional enzyme. The male mosquito with two deletions (totaling 55 nt) was further crossed with 7 age-matched wild-type females to establish a line (Fig. 1A).

Since αMan2 is located in on chromosome 1, deletions in this gene were expected to be sex-linked [19]. All G₂ male progeny from the selected heterozygous G₁ mutant mosquito that were screened (N=33) carried deletions, while the majority of G₂ females were wild-type. To obtain mutant females, we relied on identification of recombinant mosquitoes. Four out of 122 screened G₂ females (3.3%) were recombinants and carried a deleted copy of αMan2. These G₂ females were further crossed with wild-type males to obtain G₃ males with the mutation on the opposite chromosome. Homozygous mutant males were obtained via crossing G₃ mutant females and the generated G₂ mutant males. Homozygous mutant males and females were screened, selected, and crossed to obtain a homozygous αMan2 KO line (Fig. 1B). The presence of αMan2 deletions in both DNA and mRNA was confirmed by PCR and direct sequencing of the target region (Supplementary Figure 1B).

To generate a Wolbachia-infected αMan2 KO line, Wolbachia-infected Ae. aegypti females we crossed with αMan2 KO Wolbachia-negative male mosquitoes so that CI would not sterilize the cross (Beckmann et al., 2019). Every generation after crossing was checked by PCR for the presence of both the mutation and Wolbachia infection, and heterozygous Wolbachia-infected males and females were crossed. Homozygous Wolbachia-infected males and heterozygous Wolbachia-infected females were selected and further crossed as described above to obtain homozygous Wolbachia-infected male and female mosquitoes. Homozygous Wolbachia-infected male and female mosquitoes were selected using PCR and crossed to establish a pure homozygous Wolbachia-infected αMan2 KO mosquito line (Fig. 1C). Wolbachia-negative male mosquitoes from the parental Cas9-expressing Ae. aegypti line, which was used for embryo injections, were crossed with Wolbachia-infected Ae. aegypti females following similar procedure as described above to obtain a Wolbachia-infected wild-type control line with comparable genetic background for infection experiments.

Effect of αMan2 KO on Wolbachia titers in Ae. aegypti mosquitoes:

Mosquitoes carrying the αMan2 KO mutation had mean Wolbachia levels that were approximately 2-fold increased compared to the wild-type genetic background (Mann-Whitney U test, P = 0.0007) (Fig. 2).

Figure 2.

Wolbachia titers in αMan2 KO (N = 28) and wild-type (N = 36) mosquitoes. Mutant mosquitoes had significantly higher levels of Wolbachia compared to wild-type (Mann-Whitney U test, P = 0.0007).

Effect of Wolbachia and αMan2 KO on DENV infection, dissemination, and transmission rates:

DENV infection and dissemination rates were lower in Wolbachia-infected wild-type mosquitoes compared to Wolbachia-uninfected w wild-type mosquitoes (50% vs. 36% infection) although this trend was not statistically significant (Table 1). In Wolbachia-uninfected mosquitoes, the αMan2 KO mutation was associated with significantly reduced DENV infection rates (16% vs 50%) (Table 1). Although either Wolbachia alone or the αMan2 KO mutation alone both tended to reduce DENV infection and dissemination rates, the two effects were not additive. Rather, there was an interaction between Wolbachia infection status and genotype; when the αMan2 KO mutation was present in a Wolbachia-infected background, infection rates were similar to Wolbachia-uninfected wild-type mosquitoes (50% vs. 46%; P < 0.05) (Table 1). We did not observe DENV transmission in any treatment, possibly due to the timepoint assessed.

Table 1.

Virus infection, dissemination, and transmission rates for experimental treatments 7 days post-infection. Treatments with different letters are significantly different (contingency table analysis; P < 0.05).

Virus	Genotype	Wolbachia	N	Infection	Sig	Dissemination	Sig	Transmission	Sig
DENV	WT	No	36	0.50	a	0.22	a	0.00	a
DENV	WT	Yes	69	0.36	a	0.07	a,b	0.00	a
DENV	Man2 KO	No	37	0.16	b	0.03	a,b	0.00	a
DENV	Man2 KO	Yes	46	0.46	a	0.02	b,c	0.00	a
MAYV	WT	No	71	1.00	a	1.00	a	0.42	a
MAYV	WT	Yes	64	0.45	b	0.19	b	0.00	c
MAYV	Man2 KO	No	43	0.98	a	0.95	a	0.33	b,c
MAYV	Man2 KO	Yes	66	0.27	c	0.08	b	0.00	c

Effect of Wolbachia and αMan2 KO on DENV titers in mosquitoes:

In a wild-type genetic background, DENV titers were significantly lower in Wolbachia-infected mosquitoes compared to uninfected (Fig. 3A); a demonstration of canonical Wolbachia-induced PB. In a Wolbachia-uninfected background, mosquitoes with the αMan2 KO mutation had reduced DENV titers (Fig. 3A). We observed an interaction between Wolbachia infection status and genotype, where the αMan2 KO mutation in a Wolbachia-infected background reduced the ability for Wolbachia to suppress DENV (Fig. 3A). DENV dissemination titers in mosquito legs between treatments did not significantly differ, possibly due to lower dissemination rates and resulting lack of power to detect a statistical difference (Supplementary Fig. 2A).

Figure 3.

A) DENV and B) MAYV body titers in experimental mosquitoes 7 days post-infection. Red = DENV; Blue = MAYV; Squares = Wolbachia-uninfected; Circles = Wolbachia-infected; Filled symbols = wild-type; open symbols = αMan2 KO. Viruses were analyzed separately; treatments with different letters are significantly different (Brown-Forsythe ANOVA with Welch’s correction; P < 0.01). Data are pooled from 2 replicate experiments. N = sample size.

Effect of Wolbachia and αMan2 KO on MAYV infection, dissemination, and transmission rates:

MAYV infection and dissemination rates were higher generally compared to DENV, perhaps due to higher initial viral titers or greater viral permissiveness. 100% of wild-type, Wolbachia-uninfected mosquitoes became infected with and disseminated MAYV. Wolbachia-infected, wild-type mosquitoes had significantly reduced infection (45%) and dissemination (19%) rates as would be expected from Wolbachia-induced PB (Table 1, P < 0.05). In the absence of Wolbachia, infection and dissemination rates were similar in αMan2 KO and wild-type mosquitoes. αMan2 KO, Wolbachia-infected mosquitoes had the lowest infection (27%) and dissemination (8%) rates, opposite to what was observed for DENV (Table 1). We did observe transmission of MAYV in these experiments, where the highest transmission rate (42%) was observed in wild-type Wolbachia-uninfected mosquitoes, and no transmission was observed in Wolbachia-infected mosquitoes, regardless of genotype (Table 1). Wolbachia uninfected KO mosquitoes had intermediate transmission rates (33%) (Table 1).

Effect of Wolbachia and αMan2 KO on MAYV titers in mosquitoes:

In a wild-type genetic background, MAYV titers were significantly lower in Wolbachia-infected mosquitoes compared to uninfected (Fig. 3B); again consistent with canonical Wolbachia-induced PB. However, in a Wolbachia-uninfected background, the αMan2 KO mutation was associated with enhanced MAYV titers compared to wild-type mosquitoes (Fig. 3B). In a Wolbachia-infected background, the αMan2 KO mutation did not affect PB, and MAYV titers were indistinguishable from Wolbachia-infected wild-type mosquitoes (Fig 3B). MAYV dissemination titers in mosquito legs between treatments significantly differed in a similar pattern to body titers (Supplementary Fig. 2B).

Discussion

The role of the mosquito nuclear genome in modulating Wolbachia-induced pathogen blocking has been observed both empirically (Liang et al. 2022) and experimentally (Ford et al., 2019). A recent genetic screen identified that single-nucleotide polymorphisms in the Ae. aegypti αMan2 gene were associated with stronger or weaker Wolbachia (wMel)-mediated PB of DENV (Ford et al., 2019), but the functional role of this gene in DENV blocking remains unclear. Due to the intronic location of the identified polymorphisms, it was hypothesized that they could affect gene expression or splicing; however, no significant differences in αMan2 expression were found between selected low and high blocking mosquito populations (Ford et al., 2019). We recently published a study (Sigle et al., 2022) using RNAi to knock down expression of αMan2 in Wolbachia infected and uninfected Ae. aegypti to examine its effect on PB for DENV and Chikungunya virus (CHIKV); an alphavirus closely related to MAYV (Brustolin et al., 2018, Terradas et al., 2022). RNAi demonstrated some influence of αMan2 in virus infection in Wolbachia-negative mosquitoes, but we were unable to successfully knockdown gene expression in Wolbachia-infected mosquitoes and thus could not directly test the interaction of αMan2 and Wolbachia on virus infection (Sigle et al., 2022). To address this issue, here we used CRISPR-Cas9 gene editing to generate αMan2 KO mutations in Ae. aegypti mosquitoes to functionally investigate the role of this gene in arbovirus replication and found that the αMan2 KO mutation affected arboviruses in a pathogen and Wolbachia infection-specific manner. This is especially interesting as the Wolbachia strain used in the original genetic screen (Ford et al. 2019) was wMel (originally from Drosophila melanogaster), while we performed experiments (here, and previously [Sigle et al., 2022]) using the Wolbachia strain wAlbB (originally from Ae. albopictus). These two Wolbachia strains are not closely related, yet both seem to interact with αMan2, suggesting that candidate genes identified by Ford et al. (2019) may be broadly applicable across different Wolbachia strains.

Differences in viral phenotypes between mutant and wild-type mosquitoes in a Wolbachia-infected background cannot be explained by a direct effect of the KO mutation on Wolbachia titers. Wolbachia levels were approximately twice as high in αMan2 KO mosquitoes compared to wild-type. While Wolbachia-induced suppression of MAYV was similar in both αMan2 KO and wild-type mosquitoes, DENV was not blocked in Wolbachia-infected mutant mosquitoes, highlighting the complex interactions between the mosquito genome, Wolbachia, and the specific viral pathogen. Wolbachia loads can be affected by mosquito immunity (Pan et al., 2018), and the mosquito immune system can be modulated by glycosylation pathways (Bednarska et al., 2017), suggesting a potential explanation for higher Wolbachia titers in αMan2 KO mosquitoes, although this phenomenon requires further study.

For DENV, the αMan2 KO mutation itself conferred some resistance to virus, significantly reducing viral titer. Wolbachia alone also reduced viral titer. However, there was an interaction between αMan2 genotype and Wolbachia infection; when the mutation was coupled with Wolbachia in the mosquito, DENV infections were no longer suppressed. We were not able to make conclusions on the effect of the mutation or Wolbachia on DENV transmission as no mosquitoes transmitted DENV at the studied timepoint of 7 days post-infection (due to the fragile nature of the KO line, mosquitoes generally did not survive long enough to examine later timepoints). We observed a different phenomenon with MAYV. In a Wolbachia-uninfected background, the αMan2 KO mutation did not significantly alter viral infection rates but did significantly enhance viral titers in the mosquito. In a Wolbachia-infected background, the mutation increased the ability for Wolbachia to suppress viral infection rates and did not interfere with the ability for Wolbachia to suppress viral titers, although it did not further enhance Wolbachia PB. As MAYV is transmitted faster than DENV, we were able to measure an effect of both Wolbachia and the mutation on MAYV transmission at 7 days post-infection.

The fact that the αMan2 mutation (in the absence of Wolbachia) had different effects on DENV vs. MAYV is not necessarily surprising, as DENV is a flavivirus, while MAYV is an alphavirus. These two viral families are not closely related, and it has been demonstrated that the mosquito immune system responds differently to diverse viral groups (Samuel et al., 2018). The fact that the αMan2 KO mutation can have different effects on how Wolbachia suppresses different viral families is perhaps also not surprising, as Wolbachia has been shown to differentially suppress different pathogens in other systems (Hughes et al., 2011; Hughes et al., 2012; Fraser et al., 2020). Ultimately, these data demonstrate the complexity of the Wolbachia PB phenotype. In their screens, Ford et al. (2019; 2020) identified dozens of potential candidate genes regulating PB; here we have disrupted one of them. It is likely that disruption of other candidates could have equally complex consequences, to say nothing of multiple stacked mutations.

While our data show that Ae. aegypti αMan2 is a modulator of arbovirus infection, and likely involved in the Wolbachia PB phenotype, the mechanism by which it works, and has variable effects on different viruses, remains unclear. αMan2 is involved in protein glycosylation (Nemčovičová et al., 2013), which may affect viral biogenesis, replication, and infectivity (Rogers and Heise, 2019), so it is logical that disruption of this gene would affect viral infection phenotypes. However, CRISPR is a blunt tool, and further molecular research is necessary to determine the specific mechanism by which αMan2 modulates replication of specific viruses and how it contributes to Wolbachia PB. It also should be noted that although all mosquito treatment strains used in this study were siblings, they differed in mitochondrial background between Wolbachia-infected and uninfected lines, and were not 100% homogeneous in their nuclear backgrounds.

Data availability:

All data are presented in the manuscript figures, tables, and supplementary information.