https://orcid.org/0000-0001-7973-088X
Figures
Abstract
The mosquito Aedes aegypti is the primary vector of dengue virus, and the release of mosquitoes infected with the bacterium Wolbachia is an increasingly used strategy to reduce dengue transmission. This approach relies on Wolbachia’s ability to suppress virus replication within the mosquito, but the effectiveness of virus suppression and the associated fitness costs may depend on interactions among mosquito genetic background, Wolbachia strain, and dengue virus type. Here, we examined these interactions using mosquito populations from Brazil, Paraguay, and Peru infected with two Wolbachia strains, wMelM and wAlbB, and exposed to dengue virus serotype 1 or serotype 2. We measured dengue virus infection in transmission-relevant tissues and quantified mosquito survival, development, and reproduction under controlled laboratory conditions. Wolbachia infection reduced dengue virus infection overall, but the strength of virus suppression varied among mosquito populations, with the strongest blocking observed in Brazilian mosquitoes and the weakest in Peruvian mosquitoes. Across all populations, the wMelM strain consistently produced stronger dengue virus suppression than wAlbB, whereas mosquitoes infected with wAlbB generally exhibited higher fitness, with fitness effects varying among mosquito populations and life-history traits. Together, these results show that genetic interactions among the mosquito, Wolbachia, and dengue virus shape both virus blocking and mosquito fitness, highlighting trade-offs that may influence the establishment and long-term effectiveness of Wolbachia-based dengue control programs.
Author summary
Dengue virus is transmitted by the mosquito Aedes aegypti, causing large outbreaks in many parts of the world. One promising way to reduce dengue virus transmission is to release mosquitoes infected with a naturally occurring bacterium called Wolbachia, which interferes with the virus inside the mosquito. However, this strategy has not worked equally well in all locations, and the reasons for this variability are not fully understood. In this study, we tested whether differences in mosquito populations, Wolbachia strains, and dengue virus types influence how well this approach works. We compared mosquitoes from Brazil, Paraguay, and Peru infected with two commonly used Wolbachia strains and exposed them to two dengue virus types. We found that one Wolbachia strain was better at suppressing dengue virus, while the other allowed mosquitoes to survive and reproduce more effectively. Importantly, these effects depended strongly on the mosquito population being tested, with mosquitoes from Brazil showing the strongest virus suppression and those from Peru showing the weakest. Our results show that there is no single Wolbachia strain that performs best in the settings we studied and that matching Wolbachia strains to local mosquito populations may be important for improving the long-term success of dengue control programs.
Citation: Ser SL, Dennington NL, Engler HI, Harris ML, Jones MJ, Avila FW, et al. (2026) Interactions between mosquito genetic background and Wolbachia strain affect dengue virus blocking and fitness in South American populations of Aedes aegypti. PLoS Negl Trop Dis 20(5): e0014403. https://doi.org/10.1371/journal.pntd.0014403
Editor: Sassan Asgari, University of Queensland, AUSTRALIA
Received: January 27, 2026; Accepted: May 22, 2026; Published: May 27, 2026
Copyright: © 2026 Ser et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data are available at figshare https://doi.org/10.6084/m9.figshare.29840471.
Funding: This project was supported by NIAID R56 AI155573-01A1 to EM. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Dengue fever has become an escalating public health crisis across the Americas, with cases surging to record levels. In 2024, over 14 million cases were reported in the region [1]. Notably, dengue is now being reported in new areas, suggesting a geographic expansion of virus transmission driven by changing environmental conditions and increased habitat suitability for the mosquito vectors. The virus, now endemic in most Latin American and Caribbean countries, is posing a persistent threat to regional health systems. Limited antiviral treatments and restricted access to effective vaccines have left vector control as the primary strategy for curbing dengue virus (DENV) transmission [2]. However, traditional methods such as insecticide spraying and larval habitat reduction have shown declining effectiveness due to the evolution of insecticide resistance [3] in Aedes aegypti (Ae. aegypti) populations and inconsistent implementation of these methods, often hindered by social and logistical challenges that allow mosquito densities to rebound [4]. These challenges underline the need for innovative, resilient vector control strategies that work under real-world ecological and operational conditions. The release of the insect endosymbiont, Wolbachia, into mosquito populations is one such promising strategy [5].
In one approach using Wolbachia, the microbe is released into mosquito populations where it spreads, ‘replacing’ the local Wolbachia-free population. Wolbachia is maternally transmitted and spreads through mosquito populations via cytoplasmic incompatibility, an effect that gives infected females a reproductive advantage [6]. Crucially, Wolbachia also inhibits arbovirus replication within the mosquito host, thereby reducing the transmission potential of coinfecting RNA viruses, including dengue, Zika, chikungunya, and Yellow Fever, once Wolbachia has established [7]. This ‘pathogen blocking’ trait is believed to arise through multiple mechanisms, including immune activation [8], oxidative stress [9], and disruptions to cellular processes such as lipid transport [10], cell adhesion [11], and amino acid availability [12] in the vector. These physiological changes collectively create a hostile intracellular environment that impairs viral replication and dissemination to transmission-relevant tissues, including, the salivary glands.
Although field deployments of Wolbachia-infected mosquitoes have been encouraging—both in terms of Wolbachia establishment and subsequent reductions in dengue transmission—these outcomes have depended on the site where releases have taken place. In Yogyakarta, Indonesia [13], and North Queensland, Australia [14], large-scale releases have led to reductions in dengue incidence ranging from 77% to 99% in targeted regions. However, outcomes have been less consistent in other locations. In Rio de Janeiro, Brazil [15,16], and Vietnam [17], and more recently Medellín, Colombia [18], Wolbachia establishment has been variable or declined following release cessation, with recent surveys reporting low and highly heterogenous prevalence (~32% or less) in local Ae. aegypti populations. More recently, a field trial of wAlbB-infected mosquitoes in Malaysia reported a 62% reduction in dengue cases, highlighting promising but variable success [19]. Wolbachia strain wAlbB, originally derived from Aedes albopictus, is known for its relatively high thermal stability and ability to persist under field-relevant temperature conditions, though its pathogen-blocking efficacy can be more variable [20]. In contrast, wMelM, a variant of the wMel strain originating from Drosophila melanogaster, has been associated with strong dengue virus blocking, and improved thermal tolerance compared to the original wMel strain, although environmental conditions can still influence its performance [7,21]. These differences in release outcomes across geographic locations suggest that Wolbachia-based strategies might depend on a combination of genetic, ecological, and operational factors that remain insufficiently understood.
Environmental factors, such as temperature and humidity, can influence Wolbachia’s density within mosquitoes, which in turn impacts both pathogen-blocking efficiency and invasion potential [22,23]. In addition to these abiotic factors, the genetic backgrounds of both the mosquito and the Wolbachia strain have been shown to modulate the strength of virus blocking. For instance, in one study comparing Ae. aegypti populations from Mexico and Singapore infected with the same wAlbB strain, there were differences in the strength of DENV blocking, suggesting a role for mosquito genotype [24]. In another study with an Australian Ae. aegypti population infected with either wMelM or wAlbB, differences in DENV susceptibility were detected, indicating a role for Wolbachia genotype [25]. Finally, in a study of a single genetic background with one strain of Wolbachia, there was variation in blocking efficacy across different DENV serotypes (genetically and antigenically distinct variants of the virus), indicating the importance of virus genotype [26]. Importantly, Wolbachia infection can also reduce mosquito fitness, via such measured traits as fecundity [27], lifespan [28] and fertility [29], affecting its ability to invade and persist in the field. These fitness effects also vary by host and symbiont genotype [30]. This body of work highlights the ability for mosquito:virus:Wolbachia genotypes to make or break the efficacy of the biocontrol agent, which in turn also depends on environmental conditions.
Given the rising urgency of dengue control in the Americas and the scale-up of Wolbachia release programs across Brazil, Colombia, Honduras, and Puerto Rico, there is a clear need to understand how local genetic variation might influence intervention success and possibly explain past failures. This study uses a genotype-by-genotype-by-genotype (G × G × G) framework to evaluate how mosquito genetic background, Wolbachia strain, and DENV serotype interact to shape virus blocking and mosquito fitness- factors that will either promote or limit the success of Wolbachia-based interventions. We have selected mosquito populations from across Brazil, Paraguay, and Peru, the two leading Wolbachia release strains (wMelM and wAlbB), and two of the commonly circulating DENV serotypes that are also highly divergent from one another genetically [26,31]. This work provides a basis for understanding how interacting genetic factors influence both the efficacy and sustainability of Wolbachia-based dengue control, offering insights that can help refine and localize intervention strategies across the globe.
Methods
Mosquitoes and colony maintenance
Ae. aegypti mosquitoes were collected from field sites across three South American countries: Brazil (Rio de Janeiro), Paraguay (Ciudad del Este), and Peru (Campo Verde). Eggs were collected using 60 ovitraps homogeneously spread in dengue endemic neighborhoods of each location, and paddles were replaced weekly until obtaining a minimum of 10,000 eggs per site to capture local genetic diversity. The wild mosquito populations were screened for Wolbachia infection using diagnostic PCR assays targeting Wolbachia-specific genes to confirm the absence of infection, following established protocols [30,32]. Only Wolbachia-free populations were used to establish experimental lines. The wMelM donor line was obtained from Ary Hoffmann (University of Melbourne) [33], and the wAlbB donor line was obtained from Zhiyong Xi (Michigan State University), that had previously been established in Aedes aegypti through backcrossing into a Wolbachia-free line from Mérida, Mexico [24].Wolbachia strains, wMelM [34], and wAlbB [33], were first introgressed into each country line via repeated backcrossing using infected donor lines. For each population, infected females were backcrossed to males from the corresponding Wolbachia-free country line for six consecutive generations. This approach preserves the maternally inherited Wolbachia infection while progressively replacing the nuclear genetic background of the donor line. After six generations, the nuclear genome (>98%) is expected to reflect the recipient population, minimizing donor genetic carryover while maintaining colony viability. Both wMelM and wAlbB strains were originally transinfected into Ae. aegypti in previous studies [24,33] and have since been maintained in laboratory colonies for multiple generations prior to their use in this study. Following backcrossing, qPCR assays targeting strain-specific markers were performed to confirm the presence and identity of the respective Wolbachia strain in each experimental population [30,32]. A Wolbachia-free strain was maintained as a control for each population, resulting in nine experimental populations. All mosquito lines were reared under standard laboratory conditions: 26°C, 75% relative humidity, and a 12-hour light/dark photoperiod. Larvae were provided fish food (TetraMin) ad libitum throughout their development. Adult mosquitoes were maintained on 10% sucrose and were not blood-fed except where explicitly stated for experimental assays.
Vector competence assay
The DENV serotypes/strains used for this experiment were DENV-1 strain FR-50 (GenBank accession number FJ432734.1) and DENV-2 strain ET-300 (GenBank accession number EF440433.1). The viruses were cultured in Ae. albopictus C6/36 cells (Sigma), following previously established protocols [35]. C6/36 cells were maintained in RPMI 1640 medium (Life Technologies) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 20mM HEPES buffer (Sigma-Aldrich), and 1% penicillin-streptomycin (Life Technologies). Cells were grown to 80% confluence in T75 flasks before inoculation with DENV-1 or DENV-2. Infected cultures were incubated at 27°C for 7 days, and the supernatant was harvested and titrated to a final viral load (as per below) of 107 DENV copies per ml. The supernatant was mixed in a 1:1 ratio with blood from anonymous human donors (BioIVT) before feeding. Donor blood was not screened for preexisting immunity to DENV or other flaviviruses. Adult female mosquitoes at 7 days post-eclosion were deprived of sucrose for 24 hours before the infectious feed. The mosquitoes were fed using a Hemotek artificial feeder (Hemotek Ltd., UK) at 37 ∘C using pig intestine (sausage casing) as the membrane. The final virus concentration in the blood was 1.20e7 DENV copies/ml for DENV-1 and 2.40e7 DENV copies/ml for DENV-2. After feeding, mosquitoes were anesthetized on ice, and only the fully engorged were selected for subsequent experiments. Engorged mosquitoes were then placed into 32 oz paper soup cups with mesh lids, with each cup containing 12 individuals. A total of 54 groups were established, with each treatment represented by a single cup; all groups were set up simultaneously as part of a single experimental infection assay. All groups were provided with 10% sucrose-soaked cotton balls, which were refreshed daily. All mosquitoes were maintained under the same environmental conditions as the stock lines. At 5, 10, and 15 days post-infection, mosquitoes were anesthetized on ice, and specific tissues (salivary gland and carcass) were dissected under a microscope. Dissected tissues and body parts were stored separately in 1.5 ml microcentrifuge tubes (Sarstedt, Nümbrecht, Germany) containing 300μl of DNA/RNA shield and a 2.8mm ceramic bead. Samples were homogenized using a Bead Ruptor Elite (Omni International, USA) and kept frozen at -80°C until further processing.
Mosquito nucleic acid extraction
Total RNA was extracted using the Zymo Quick-DNA/RNA Pathogen MagBead Kit (Zymo Research, Cat.No R2146) according to the manufacturer’s protocol. The extraction was performed using a MagMax Express 96 (Applied Biosystems), and DNA/RNA was eluted in 50μl ZymoBIOMICS DNAase/RNase-free water at room temperature. RNA was eluted in 50μl RNase-free water and then treated with 5 units of DNase I (Sigma-Aldrich) at room temperature for 15 mins, followed by inactivation with 50mM EDTA at 70°C for 10 mins. RNA and DNA extractions were performed using the column-based Direct-zol DNA/RNA miniprep kit. RNA was eluted into 50μl RNase-free water, followed by DNA elution in 50μl Direct-zol DNA elution buffer. All RNA quantification assays used the same endogenous control (18S ribosomal RNA) across all mosquito populations and treatments to ensure consistency in normalization.
Dengue virus quantification
DENV was quantified using the TaqMan Fast Virus 1-step Master Mix (Thermo Fisher Scientific) in 10 µl reaction volumes, with DENV-2 specific primers and probes for qRT-PCR, following previously described methods [11]. The protocol consisted of a reverse transcription step at 50°C for 5 minutes, an initial denaturation at 95°C for 20 seconds, and amplification cycles of 95°C for 3 seconds followed by 60°C for 30 seconds. Absolute quantification was achieved using a standard curve generated from a known concentration of a DENV-2 genomic fragment. This fragment was cloned into a plasmid, transformed into Escherichia coli, and prepared as described previously [11]. The linearized and purified fragment was serially diluted from 106 to 102 copies to create a standard curve for DNA amplification. Each 96-well plate included the standard curve and negative controls, both run as duplicates.
Survival and fecundity
Female adult mosquitoes from the nine experimental populations were maintained under standard conditions in 5 groups of 20 individuals each per cup. Survival was monitored daily until all individuals died. Mosquitoes in the survival assay were maintained with cotton balls soaked in 10% sucrose solution, which were replaced daily, and were never blood fed during the experiment. For fecundity, adult female mosquitoes 7 days post-eclosion (after mating) were sorted by experimental populations and divided into groups. Each population group consisted of 30 females housed in three BugDorm-4S1515 (5.4-liter) insect-rearing cages (10 individuals per cage). Each cage had a small urine specimen cup (180ml) lined with filter paper and filled with deionized water to keep the paper moist. Throughout the assay, mosquitoes had access to 10% sucrose water except for 24 hours prior to each blood meal consisting of human blood. At 7 days post-eclosion, females were deprived of sucrose for 24 hours before receiving a noninfectious blood meal, as above. Females were allowed to oviposit over three days per cycle, during which the filter paper was replaced every three days. This process was repeated at 14 and 21 days post-eclosion to capture the second and third gonotrophic cycles, respectively, and eggs were manually counted each time. After each blood meal, the number of surviving females was recorded, and the average fecundity for each gonotrophic cycle was calculated by dividing the total number of eggs by the number of live females per replicate and experimental population.
Larval development
Eggs from each of the nine experimental populations were hatched under standard laboratory conditions as described above. After 24 hours, 150 first-instar larvae from each population were randomly selected and divided into five 250 ml plastic cups, with 30 larvae per cup. Each cup contained deionized water and was provided with fish food (TetraMin) ad libitum. Larvae were continuously fed until the onset of pupation. Once pupation began, the amount of food provided was adjusted according to the remaining larvae in each cup to prevent overfeeding. Pupae, both live and dead, were removed daily, counted, and placed in individual 30 ml plastic cups containing water from their original larval environment to facilitate eclosion. These cups were added to respective BugDorm-4S1515 (5.4-liter) insect-rearing cages (Megaview Science Co., Ltd., Taiwan) with continuous access to 10% sucrose solution. The number of adults emerging each day was recorded to assess development time.
Model development for the prediction of mosquito fitness (r) fitness
To calculate population-level fitness, we used a stage-structured matrix projection model, which we parameterized from our data as previously described [11,36,37]. We describe the change in population over time:
𝑁𝑡+1 = 𝑀𝑁𝑡
N is the abundance in the stage at a given time (t), and M is the population projection matrix. Fecundity is the first matrix row, populated by fecundity or the average number of offspring produced per female at a given age, of which we included two gonotrophic cycles. The sub-diagonal of the matrix consists of the survival probability from a given age to age i + 1. The stage-structured population dynamics are the product of the transition matrix and the stage-structured population size vector across time. The stable stage distribution of the abundance vector is reached after repeated multiplication, and the dominant eigenvalue of the system is the finite population rate of increase (𝜆). The intrinsic rate of population growth is
𝑟𝑚𝑎𝑥 = log(𝜆)
This is the population fitness representing the population’s ability to reproduce, and as such, negative rmax indicates a population that is in decline, while positive rmax indicates population growth.
For each trait, replicate values were resampled with replacement with 10,000 iterations per treatment to account for variation in the number of replicates between trait measurements. Note that we maintained the order of survival probability across time for resampling in adult survival. We incorporated our resampled iterations for each trait with a Bayesian inference framework to estimate the posterior distribution of rmax for each treatment. The model assumed that each estimate of rmax has a normal distribution with a mean of 𝜇𝑟𝑚𝑎𝑥 and a precision parameter 𝜏𝑟𝑚𝑎𝑥 where 𝜏 is the inverse of the variance. We assigned weekly informative priors with a mean (𝜇) of 0, a precision (𝜏) of 0.001, and a uniform prior between 0 and 10 for the standard deviation (𝜎𝑟𝑚𝑎𝑥) allowing our data to inform the posterior distribution.
We fit the models using a Markov Chain Monte Carlo (MCMC) sampling as implemented in JAGS, using the R2jags along with the popbio package [38–40]. For each treatment group we ran three MCMC chains with 10,000 iterations and a 3,000 step burn-in. We used these samples for the posterior distribution. We then summarized the rmax by calculating the mean and 95% highest posterior density (HPD) interval. The HPD includes the smallest continuous range, containing 95% of the probability, which is the credible interval implemented by the coda package [41]. We then used pairwise comparisons of the posterior difference of samples to understand the probability that one treatment’s rmax was more significant than another.
Parameterizing of life-history traits
Individuals’ survival and life stage duration (larva to pupa to adult stages) were measured through pupation and adult emergence time. The first instar larval emergence was assumed to be one day. Adult survival elements were populated using a continuous survival proportion from a Kaplan-Meier survival function in the R survival package survival and survminer [42]. Fecundity measurements were the average eggs per female per gonotrophic cycle with two stages for the first two gonotrophic cycles. The assumption of unlimited resources, density, and constant laboratory conditions constrains these estimates.
Statistical analysis
All statistical analyses were performed using R (version 4.3.2) or JMP Pro (version 18.1.0, SAS Institute Inc., Cary, NC, 1989–2025), depending on the assay. For the vector competence assay (dengue prevalence), analyses were conducted in R. A chi-squared binomial generalized linear model (GLM) with a logit link was used to compare infection prevalence among treatments within specific mosquito tissues. Infection prevalence was analyzed as the number infected over the total number of individuals, with ‘Dengue serotype,’ ‘Mosquito strain,’ ‘Wolbachia strain,’ and ‘Days Post-Infection (DPI)’ included as categorical fixed effects. Wolbachia infection status (wAlbB, wMelM, and Wolbachia-free) was modeled as a single factor, with comparisons to the Wolbachia-free group representing overall blocking effects, and direct comparisons between wAlbB and wMelM reflecting strain-specific differences. Differences between models including or excluding interactions were assessed using ANOVA. For the survival assay, Kaplan-Meier survival curves were generated in R, stratified by Wolbachia strain, to visualize survival differences. Pairwise comparisons were conducted using the log-rank test to identify significant differences between populations or treatments. For the fecundity assay, ANOVA was performed in R to assess the effects of mosquito population (country), Wolbachia infection status, number of blood meals (gonotrophic cycle), and their interactions on egg production. For the larval development assay, ANOVA was performed to assess the effects of mosquito population (country), Wolbachia infection status, and their interaction on development time. All factors were treated as fixed effects, and pairwise comparisons were conducted using Tukey’s post hoc tests where appropriate. The statistical analysis of dengue viral loads in mosquitoes from the vector competence assay was conducted in JMP Pro. Mixed-effects models were fit by maximum likelihood and statistically compared using Tukey’s post-hoc tests to evaluate interactions among population and treatment groups.
Results
wMelM Exhibits stronger DENV blocking than wAlbB in South American Ae. aegypti
To investigate the effects of Wolbachia strain and mosquito genetic background on DENV blocking, we generated multiple experimental populations through backcrossing. Specifically, we introduced wAlbB and wMelM strains into Ae. aegypti populations from Brazil, Paraguay, and Peru. The Wolbachia-infected and wildtype pair-matched control lines were then fed dengue virus serotype 1 (DENV-1) or serotype 2 (DENV-2) to assess the impact on Wolbachia-mediated pathogen blocking (Fig 1).
Three Wolbachia conditions were tested: wAlbB, wMelM, and Wolbachia-free controls. Each Wolbachia strain was backcrossed into Ae. aegypti populations from each country-based population and then exposed to either dengue virus serotype 1 (DENV-1) or dengue virus serotype 2 (DENV-2). Viral infection was assessed at 5,10, and 15 days post-infection (dpi) in two tissues: salivary glands and carcass.
We assessed both DENV prevalence and viral load in several tissues for the different mosquito populations at three time points post-infection. We hypothesized that mosquito genetic background, Wolbachia strain, and DENV strain would influence the extent of Wolbachia-mediated pathogen blocking. For both prevalence and load, the three main effects of population, Wolbachia status, and DENV serotype were all significant, (S1–S3 Tables). The three-way interactions for prevalence (χ2 = 594.0, df = 4, p < 0.0001) and viral load (F(4,1188) =7.39, p < 0.0001), were also significant, revealing the presence of GxGxG effects.
Our results show that Wolbachia infection significantly reduced both DENV prevalence and viral load, with wMelM-infected mosquitoes exhibiting the strongest blocking effect, regardless of mosquito population. Specifically, wMelM mosquitoes had lower DENV prevalence in the salivary glands (Fig 2: Generalized Linear Model: “Wolbachia”: χ2 = 87.51, df = 2, p < 0.0001) compared to wAlbB-infected and Wolbachia-free controls. This pattern was mirrored in viral load, where wMelM-infected mosquitoes had significantly lower DENV loads in the salivary glands (Fig 3: Mixed-Effects Model: “Wolbachia”: F = 697.62, df = 2, p < 0.0001). A similar trend was observed in the carcass for both DENV prevalence and viral load (S1 and S2 Fig), reinforcing the robust blocking effect of wMelM across tissues. Additionally, Wolbachia infection was more effective at blocking DENV-2 than DENV-1, as measured by prevalence (“DENV serotype”: χ2 = 25.01, df = 1, p < 0.0001) and viral load (“DENV serotype”: F(1,1188) =27.73, p < 0.0001) across mosquito populations. Across both Wolbachia strains (wMelM and wAlbB), the strength of Wolbachia-mediated blocking varied by country-based population, with mosquitoes from the Brazil line frequently exhibiting stronger blocking and Peruvian lines comparatively weaker blocking across strains, serotypes, and tissues. This population-level trend was consistent across both salivary glands (Figs 2–3) and carcasses (S1 and S2 Figs) for both DENV prevalence and viral load. Full GLM results are provided in S1 and S2 Tables. Detailed pairwise comparisons are provided in S1 Data.
Each population was exposed to a) DENV-1 or b) DENV-2, and mosquitoes were subsequently collected at 5-, 10-, and 15-days post-infection (DPI) to assess dengue prevalence in the salivary glands. Bars represent the proportion of infected mosquitoes per treatment group (n = 12 individuals per treatment group at each time point), with error bars indicating standard error.
Each population was exposed to DENV-1 or DENV-2, and dengue viral loads were assessed in salivary glands collected at 5-, 10-, and 15- days post-infection (DPI). Bars represent median viral load and whiskers indicating the 95% confidence intervals. Each dot represents salivary gland tissue dissected from a single individual. n = 12 individuals per treatment group at each time point.
Mosquito fecundity varies independently of Wolbachia strain and genetic background
We then evaluated the effect of Wolbachia strain and the mosquito’s genetic background on Ae. aegypti fecundity measured over three gonotrophic cycles. There was no significant interaction between Wolbachia strain and mosquito population, nor between gonotrophic cycle and either factor, indicating independent effects on fecundity. Our analysis revealed that only gonotrophic cycle was significant (ANOVA: F(1,63) = 12.55, p = 0.00075), while neither Wolbachia strain (F(2,63) = 2.33, p = 0.105) nor mosquito population (F(2,63) = 0.321, p = 0.726) had an effect. As shown in Fig 4, fecundity declined across gonotrophic cycles, with the number of eggs per female generally decreasing from the first to the third cycle. However, there was no clear trend associated with Wolbachia infection or geographic background, indicating that reproductive output is primarily influenced by the gonotrophic cycle rather than Wolbachia presence or country of origin (see S3 Table for full ANOVA analysis), with no evidence of improved fecundity in either wMelM or wAlbB-infected mosquitoes relative to wildtype.
The average number of eggs per female was recorded over three gonotrophic cycles to assess reproductive output across different Wolbachia strains and geographic backgrounds. Data are presented as a dot plot, with each point representing a biological replicate (one cage of 10 females). Egg counts were pooled within each replicate and normalized by the number of surviving females. Each treatment group consisted of three replicate cages (n = 3).
Wolbachia reduces Ae. aegypti survival, with strongest effects in Brazil
To analyze the effect of Wolbachia infection on mosquito survival, we performed survival analysis using a pairwise log-rank test across the different South American mosquito populations in the absence of DENV infection. Among Wolbachia-free control mosquitoes, there were significant differences in baseline survival between populations, with the Brazilian line exhibiting the highest survival, followed by Peru, and then Paraguay (Fig 5). Overall, our results show that Wolbachia infection significantly reduced overall survival in most populations, but the magnitude of these effects varied across populations. The greatest reduction was observed in the Brazilian population, where Wolbachia-infected mosquitoes exhibited the shortest lifespan, reaching 50% death sooner than their uninfected counterparts, with median survival reduced by approximately 1.25-1.3-fold. This trend was consistent across both wMelM and wAlbB strains (log-rank; χ2 = 20.31, df = 1, p < 0.05). While both Wolbachia strains reduced the survival compared to Wolbachia-free controls, differences between wMelM and wAlbB were dependent on the mosquito population. In Brazil, there was no significant difference in survival between mosquitoes infected with wMelM and those infected with wAlbB (log-rank; χ2 = 0.00015, df = 1, p = 0.99). In contrast, in Paraguay, wMelM-infected mosquitoes tended to have a slightly longer lifespan than those infected with wAlbB (log-rank; χ2 = 7.51, df = 1, p < 0.05). A similar trend was also observed in Peru, where wAlbB-infected mosquitoes had a shorter lifespan compared to those infected with wMelM (log-rank; χ2 = 38.98, df = 1, p < 0.05) (see S4 Table for details on individual pairwise comparisons).
The colored lines represent the probability of survival for 100 individual mosquitoes per treatment group (five independent biological replicates of 20 mosquitoes each), with shaded areas representing 95% confidence intervals. The dotted line indicates the time point at which 50% of the individuals in each group survived.
Effect of Wolbachia infection on mosquito development rate depends on mosquito’s genetic background
We examined mosquito development rate, calculated as the inverse of the time (in days) it took for mosquitoes to develop from the first instar larval stage to the adult stage across different South American mosquito populations. Here, we show that neither Wolbachia status (ANOVA: F(2,36) = 2.80, p = 0.074) nor the mosquito’s genetic background (F(2,36) = 0.77, p = 0.47) alone significantly influenced mosquito development rate. However, there was a significant interaction between Wolbachia status and population (F(4,36) = 3.38, p = 0.019), indicating that the effect of Wolbachia infection on development rate depends on the mosquito’s genetic background. As shown in Fig 6, development rate varied inconsistently across populations and Wolbachia strains. In the Brazil line, mosquitoes infected with wAlbB and wMelM exhibited slightly higher development rates compared to Wolbachia-free controls. In the Paraguay line, development rates were more consistent across all Wolbachia statuses, with no clear pattern of suppression or enhancement. In Peru, however, wMelM-infected mosquitoes tended to have slower development rates than Wolbachia-free controls. This variation suggests that the influence of Wolbachia on mosquito development rate is background-dependent.
The development rate, calculated as the inverse of the time taken (days) for first-instar larvae to reach the adult stage, is presented for each group. Data are displayed as boxplots, with bars representing median development rate and whiskers indicating the 95% confidence intervals. Each dot represents a replicate group. The sample size is n = 5 replicates per treatment, with 30 individual larvae included at the start of each experiment.
wAlbB-infected mosquitoes exhibit higher population growth rates across genetic backgrounds
We compiled these life-history traits into a composite fitness model to assess how the interaction between mosquito genetic background and Wolbachia infection influences population growth rate (rmax). Our results show a consistent pattern across populations: mosquitoes infected with wAlbB exhibited the highest population growth (Fig 7). In Brazil, wAlbB-infected mosquitoes had a 2.3-fold higher rmax compared to wMelM-infected mosquitoes. In Peru and Paraguay, the difference was smaller but still notable—approximately 1.8-fold and 1.5-fold, respectively. Although some similarities were observed between populations, treatment groups were significantly different from each other (posterior probability, Δ > 0; see S5 Table). It is also important to note that in this model, not all life-history traits contribute equally to population fitness. For example, depending on the dataset, survival may have a stronger influence on rmax than development rate or fecundity. To explore this further, we examined the elasticity of each trait to determine how much each one contributed to overall fitness; details of the elasticity analysis are provided in the supplementary materials. Fecundity juvenile survival showed the highest elasticity, indicating that small changes in these traits would have the greatest impact on rmax. However, because juvenile survival was uniformly high (~100%) across populations, variation in fecundity was more likely to explain the observed differences in population fitness.
The point indicates the mean model fit, and the error bars indicate the 95% credible intervals.
Discussion
Using a G x G x G experimental design, this study revealed that DENV blocking by Wolbachia in Ae. aegypti mosquitoes is influenced by complex interactions between the mosquito’s genetic background, Wolbachia strain, and viral serotype. While both wAlbB and wMelM strains effectively suppressed DENV infection overall, the extent of blocking varied depending on these interacting factors, with wMelM conferring stronger and more consistent blocking than wAlbB across diverse Ae. aegypti populations. This effect was especially pronounced in mosquitoes from Brazil, where wMelM showed the greatest reduction in both DENV prevalence and viral load. We also note that the Peruvian population exhibited consistently weaker blocking across strains and serotypes, highlighting a population background in which Wolbachia-mediated suppression may be least effective. Blocking was also more effective against DENV-2 than DENV-1 across all Wolbachia strains and mosquito populations, suggesting that DENV serotype influences the degree of Wolbachia-mediated pathogen blocking. Previous studies have shown that both mosquito genotype and Wolbachia strain influence the strength of virus blocking [25]. However, the relative blocking performance of different Wolbachia strains appears to be context dependent. For example, in this particular study, stronger inhibition of dengue transmission by wAlbB than by wMelCS was reported under their experiment, particularly when patient-derived viremic blood meals were used [25]. In our study, by contrast, wMelM generally showed stronger blocking than wAlbB across several South American mosquito populations. These differences may reflect variation in experimental design, including blood-meal source (patient-derived versus laboratory-spiked), viral isolate, serotype, and titer, mosquito genetic background, and the specific Wolbachia variant evaluated. In addition, comparative studies across all four DENV serotypes have found DENV-1 to be more difficult to inhibit than DENV-2 [43], supporting our observation of serotype-dependent variation in blocking. These prior findings align with the trends observed across our South American mosquito populations.
The observed variability in DENV blocking may be driven, in part, by differences in host–symbiont genetic compatibility between Wolbachia strains and their mosquito hosts. Genetic compatibility refers to how effectively a Wolbachia strain can replicate, localize, and exert pathogen-blocking effects within a given mosquito genotype, shaped by interactions between both the nuclear and mitochondrial genomes. The host cellular environment can support or constrain Wolbachia persistence and its ability to modulate host biology in ways that ultimately interfere with DENV replication and dissemination [11,26,44]. Because Wolbachia and mitochondria are maternally inherited, introgression and backcrossing strategies used to localize Wolbachia strains into field-adapted mosquito populations may generate mismatched nuclear–mitochondrial combinations, potentially affecting symbiont performance and mosquito viability [45–47]. One plausible mechanism underlying these background-dependent fitness effects is cyto-nuclear incompatibility between mosquito nuclear genomes, mitochondrial haplotypes, and Wolbachia strains. Publicly available nuclear and mitochondrial genomic datasets from South American Ae. aegypti populations could be leveraged in future studies to identify polymorphisms or haplotypes associated with differential Wolbachia compatibility. Integrating such genomic information with phenotypic fitness measurements would help clarify the mechanistic basis of host–strain interactions and improve predictions of Wolbachia performance in diverse field settings.
Understanding the role of host–symbiont genetic compatibility has important implications for the global scalability of Wolbachia-based interventions. Despite being a single global species, given its relatively recent expansion out of the African continent, Ae. aegypti populations show substantial genetic differentiation across geographic regions. For example, pairwise Fst values are 0.119 for African to South America, and 0.114 for South America to Asian populations [47]. Within the South American continent, the intrapopulation Fst value is as high as 0.181 [43]. This underlying genetic structure across the continent likely reflects invasion history and population-specific adaptations to local environments. In terms of the latter, traits such as dengue virus susceptibility [48], thermal tolerance [49], and insecticide resistance [50] are known to vary among Ae. aegypti populations due to local selection pressures. As such, the success of Wolbachia interventions may depend heavily on how well released strains interact with these locally adapted genetic backgrounds.
Importantly, the variations in these interactions were observed not only in blocking efficacy but also in mosquito fitness. While both Wolbachia strains showed a significant fitness cost effect in our experimental settings, variation among host backgrounds suggests fitness outcomes may be background-dependent. For instance, survival outcomes and mosquito development rates differed depending on both the Wolbachia strain and the host population. In some populations like Brazil, both Wolbachia strains significantly reduced lifespan, with no difference in the scale of that effect between the strains. However, in both Peru and Paraguay, being infected with wMelM provided these populations a longer lifespan than those infected with wAlbB. These variations can also be seen when we look at the mosquito development rate. Comparisons of just the Wolbachia-free populations alone showed differences in their development rate, suggesting that mosquito genetic background on its own influences how quickly these mosquitoes can develop. Wolbachia infection, in some cases, can either slow development or, have no effect. The direction and magnitude of change depend on the Wolbachia strain and mosquito population. Using a composite fitness model, based on the various fitness parameters that were measured, wAlbB-infected mosquitoes had better population fitness relative to both wMelM-infected and Wolbachia-free mosquitoes. Consistent with previous studies, wAlbB infection has been reported to retain several traits favorable for invasion in Ae. aegypti, including strong cytoplasmic incompatibility, perfect maternal transmission, and limited effects on some fitness traits [51]. It is important to note that the stage-structured population growth model used here assumes constant laboratory conditions and does not incorporate density dependence or environmental heterogeneity. As a result, estimates of rmax reflect relative fitness differences under controlled conditions rather than absolute predictions of population growth in the field. In natural settings, factors such as fluctuating temperature, resource limitation, larval competition, and spatial structure are likely to modify both mosquito demography and Wolbachia dynamics. Despite these limitations, laboratory-based estimates of population growth remain valuable for comparative assessments of fitness trade-offs among treatments and provide a useful baseline for informing more complex, field-relevant models.
The results from our study align well with previous work that have shown that mosquito and Wolbachia genetic background play an important role in mosquito fitness. For instance, one study that looked at the fecundity of both wAlbB and wMelM-infected Ae. aegypti (Brazilian genetic background) found that wMelM-infected mosquitoes laid significantly fewer eggs than wAlbB-infected mosquitoes [30]. Notably our findings may help explain field observations in countries like Brazil (Jurujuba) [52] and Colombia (Medellín) [18] where, wMel Wolbachia invasion has been slower or less stable than expected. In particular, the fitness costs we observed in Brazilian (Rio de Janeiro) populations infected with either Wolbachia strain could contribute to reduced establishment and spread in the field. In this context, the reduced population fitness observed for wMelM in the Brazilian genetic background provides a plausible explanation for the slower or less stable invasion dynamics. If similar host-Wolbachia compatibility effects occur in the field, even modest fitness costs could limit the rate of spread or long-term persistence of specific Wolbachia strains, despite strong virus-blocking efficacy.
The presence of Wolbachia may alter key metabolic and physiological processes in Ae. aegypti, leading to trade-offs that impact host fitness, although the magnitude of these effects may vary across mosquito populations. These alterations include increased energetic demands to support symbiont replication and maintenance, competition for essential cellular resources, and modulation of host immune and stress response pathways [9,12,53]. For instance, Wolbachia can influence lipid metabolism and reactive oxygen species levels, potentially diverting energy away from other vital functions such as reproduction and longevity [10,54]. These physiological shifts might reduce host fitness in a context-dependent manner, as evidenced by variation in survival across mosquito genetic backgrounds in our study.
Together, these findings support the idea that genetic interactions between host, symbiont, and virus shape the outcomes of Wolbachia-mediated pathogen blocking, invasiveness and ultimately the epidemiological outcome of Wolbachia population replacement as a dengue mitigation strategy. Overall, our study has shown wMelM blocked DENV more effectively than wAlbB, but the strength of blocking varied depending on the mosquito population and DENV serotype. The Brazilian population tended to exhibit the strongest blocking, and the Peruvian, the worst. In terms of fitness, wAlbB-infected mosquitoes consistently showed higher fitness compared to wMelM, but fitness effects were not uniform — they depended on the mosquito population and fitness trait measured. To further investigate the mechanisms underlying these interactions, future studies could incorporate genomic and transcriptomic analyses. Sequencing the genomes of the mosquito populations used in this study could identify key variants that mediate Wolbachia compatibility or viral susceptibility. In parallel, transcriptomic profiling could reveal how gene expression patterns shift across host-symbiont-virus combinations, offering insights into host pathways that regulate blocking or fitness outcomes. Ultimately, such mechanistic work will be valuable for complementing phenotypic studies and refining deployment strategies. In particular, genotype-specific differences in virus blocking and fitness could be used to inform strain-matching approaches in which Wolbachia strains are selected based on their compatibility with local mosquito genetic backgrounds to balance virus suppression and population persistence. These interaction effects could also be incorporated into predictive models of Wolbachia invasion and impact, improving forecasts of establishment success and long-term efficacy across tropical regions. Ideally, Wolbachia interventions could be tailored to specific ecological and genetic contexts—matching mosquito strains with compatible Wolbachia infections to maximize blocking while minimizing fitness costs in target environments. However, achieving that level of precision requires significant time, resources, and localized data. Given that Wolbachia deployment and introgression can take several months, it may not be suitable as a rapid response during active dengue outbreaks. Instead, these efforts could be viewed as part of long-term, regionally informed preparedness strategies aimed at reducing the risk and severity of future outbreaks.
Supporting information
S1 Fig. Dengue prevalence in the carcass of mosquito populations from Brazil, Paraguay, and Peru tested with Wolbachia strains wAlbB, wMelM, and Wolbachia -free controls.
Each population was exposed to a) DENV-1 or b) DENV-2, and mosquitoes were subsequently collected at 5-, 10-, and 15-days post-infection (DPI) to assess dengue prevalence in the carcasses. N = 12 individuals per treatment.
https://doi.org/10.1371/journal.pntd.0014403.s001
(TIF)
S2 Fig. Dengue viral load in the carcass of mosquito populations from Brazil, Paraguay, and Peru tested with Wolbachia strains wAlbB, wMelM, and Wolbachia - free controls.
Each population was exposed to DENV-1 or DENV-2, and dengue viral loads were assessed in carcasses collected at 5-, 10-, and 15- days post-infection (DPI). Bars represent median viral load and whiskers indicating the 95% confidence intervals. Each dot represents carcass tissue dissected from a single individual. N = 12 individuals per treatment.
https://doi.org/10.1371/journal.pntd.0014403.s002
(TIF)
S1 Table. Results of binomial generalized linear models (GLMs) testing the effects of mosquito population (country), Wolbachia infection status, dengue virus (DENV) serotype, and days post-infection (DPI), including all interaction terms, on dengue infection prevalence in Aedes aegypti.
Infection prevalence was modeled as the number of infected individuals out of the total sampled per treatment group. Statistical significance was assessed using likelihood ratio tests.
https://doi.org/10.1371/journal.pntd.0014403.s003
(DOCX)
S2 Table. Summary of ANOVA results from mixed-effects models testing the effects of mosquito population (country), Wolbachia infection status, dengue virus (DENV) serotype, and days post-infection (DPI), as well as their interactions, on dengue viral load in Ae. aegypti.
https://doi.org/10.1371/journal.pntd.0014403.s004
(DOCX)
S1 Data. 5. Full pairwise comparisons of dengue viral load across mosquito populations, Wolbachia infection states, dengue virus (DENV) serotypes, and days post-infection (DPI).
Pairwise differences, standard errors, test statistics, and adjusted p-values are reported for all comparisons.
https://doi.org/10.1371/journal.pntd.0014403.s005
(XLSX)
S3 Table. Summary of ANOVA results testing the effects of Wolbachia strain, mosquito population, and gonotrophic cycle, as well as their interactions, on fecundity.
https://doi.org/10.1371/journal.pntd.0014403.s006
(DOCX)
S4 Table. Log-rank p-values from pairwise comparison analysis of survival in South American Ae. aegypti populations with three Wolbachia infectious status- wAlbB, wMelM, and Wolbachia -free controls.
https://doi.org/10.1371/journal.pntd.0014403.s007
(DOCX)
S5 Table. Fitness (rmax) values derived from life-history traits measured in mosquitoes reared from three populations (Brazil, Paraguay, Peru) and three Wolbachia infection statuses (Wolbachia-free, wMelM, wAlbB).
The 95% credible intervals are derived from the posterior distribution.
https://doi.org/10.1371/journal.pntd.0014403.s008
(DOCX)
References
- 1. Report on the epidemiological situation of dengue in the Americas. Pan American Health Organization. 2025. https://www.paho.org/sites/default/files/2025-05/2025-cde-dengue-sitrep-americas-epi-week-15-apr.pdf
- 2. Obi J, Gutiérrez-Barbosa H, Chua J, Deredge D. Current trends and limitations in dengue antiviral research. Tropical Med. 2021;6(4):180.
- 3. Zulfa R, Lo W-C, Cheng P-C, Martini M, Chuang T-W. Updating the insecticide resistance status of Aedes aegypti and aedes albopictus in asia: a systematic review and meta-analysis. Trop Med Infect Dis. 2022;7(10):306. pmid:36288047
- 4. Bouzid M, Brainard J, Hooper L, Hunter PR. Public health interventions for Aedes control in the time of Zikavirus– a meta-review on effectiveness of vector control strategies. PLoS Negl Trop Dis. 2016;10(12):e0005176.
- 5. Flores HA, O’Neill SL. Controlling vector-borne diseases by releasing modified mosquitoes. Nat Rev Microbiol. 2018;16(8):508–18. pmid:29777177
- 6. Kaur R, Meier CJ, McGraw EA, Hillyer JF, Bordenstein SR. The mechanism of cytoplasmic incompatibility is conserved in Wolbachia-infected Aedes aegypti mosquitoes deployed for arbovirus control. PLoS Biol. 2024;22(3):e3002573.
- 7. Walker T, Johnson PH, Moreira LA, Iturbe-Ormaetxe I, Frentiu FD, McMeniman CJ, et al. The wMel Wolbachia strain blocks dengue and invades caged Aedes aegypti populations. Nature. 2011;476(7361):450–3. pmid:21866159
- 8. Pan X, Pike A, Joshi D, Bian G, McFadden MJ, Lu P, et al. The bacterium Wolbachia exploits host innate immunity to establish a symbiotic relationship with the dengue vector mosquito Aedes aegypti. ISME J. 2018;12(1):277–88. pmid:29099491
- 9. Pan X, Zhou G, Wu J, Bian G, Lu P, Raikhel AS. Wolbachia induces reactive oxygen species (ROS)-dependent activation of the Toll pathway to control dengue virus in the mosquito Aedes aegypti. Proc Natl Acad Sci USA. 2012;109(1).
- 10. Geoghegan V, Stainton K, Rainey SM, Ant TH, Dowle AA, Larson T, et al. Perturbed cholesterol and vesicular trafficking associated with dengue blocking in Wolbachia-infected Aedes aegypti cells. Nat Commun. 2017;8(1):526. pmid:28904344
- 11. Ford SA, Allen SL, Ohm JR, Sigle LT, Sebastian A, Albert I, et al. Selection on Aedes aegypti alters Wolbachia-mediated dengue virus blocking and fitness. Nat Microbiol. 2019;4(11):1832–9. pmid:31451771
- 12. Caragata EP, Rancès E, O’Neill SL, McGraw EA. Competition for amino acids between Wolbachia and the mosquito host, Aedes aegypti. Microb Ecol. 2014;67(1):205–18. pmid:24337107
- 13. Utarini A, Indriani C, Ahmad RA, Tantowijoyo W, Arguni E, Ansari MR. Efficacy of Wolbachia-infected mosquito deployments for the control of dengue. N Engl J Med. 2021;384(23):2177–86.
- 14. Schmidt TL, Filipović I, Hoffmann AA, Rašić G. Fine-scale landscape genomics helps explain the slow spatial spread of Wolbachia through the Aedes aegypti population in Cairns, Australia. Heredity. 2018;120(5):386–95.
- 15. Kotsakiozi P, Gloria-Soria A, Caccone A, Evans B, Schama R, Martins AJ. Tracking the return of Aedes aegypti to Brazil, the major vector of the dengue, chikungunya and Zika viruses. PLoS Negl Trop Dis. 2017;11(7):e0005653.
- 16. Ribeiro Dos Santos G, Durovni B, Saraceni V, Souza Riback TI, Pinto SB, Anders KL, et al. Estimating the effect of the wMel release programme on the incidence of dengue and chikungunya in Rio de Janeiro, Brazil: a spatiotemporal modelling study. The Lancet Infectious Diseases. 2022;22(11):1587–95.
- 17. Hien NT, Anh DD, Le NH, Yen NT, Phong TV, Nam VS, et al. Environmental factors influence the local establishment of Wolbachia in Aedes aegypti mosquitoes in two small communities in central Vietnam. Gates Open Res. 2022;5:147. pmid:35602266
- 18. Calle-Tobón A, Rojo-Ospina R, Zuluaga S, Giraldo-Muñoz JF, Cadavid JM. Evaluation of Wolbachia infection in Aedes aegypti suggests low prevalence and highly heterogeneous distribution in Medellín, Colombia. Acta Trop. 2024;260:107423. pmid:39366500
- 19. Hoffmann AA, Ahmad NW, Keong WM, Ling CY, Ahmad NA, Golding N, et al. Introduction of Aedes aegypti mosquitoes carrying wAlbB Wolbachia sharply decreases dengue incidence in disease hotspots. iScience. 2024;27(2):108942. pmid:38327789
- 20. Xi Z, Dean JL, Khoo C, Dobson SL. Generation of a novel Wolbachia infection in Aedes albopictus (Asian tiger mosquito) via embryonic microinjection. Insect Biochem Mol Biol. 2005;35(8):903–10. pmid:15944085
- 21. Ross PA, Wiwatanaratanabutr I, Axford JK, White VL, Endersby-Harshman NM, Hoffmann AA. Wolbachia infections in Aedes aegypti differ markedly in their response to cyclical heat stress. McGraw EA, editor. PLoS Pathog. 2017;13(1):e1006006.
- 22. Murdock CC, Blanford S, Hughes GL, Rasgon JL, Thomas MB. Temperature alters Plasmodium blocking by Wolbachia. Sci Rep. 2014;4(1).
- 23. Ye YH, Carrasco AM, Dong Y, Sgrò CM, McGraw EA. The effect of temperature on Wolbachia-mediated dengue virus blocking in Aedes aegypti. Am J Trop Med Hyg. 2016;94(4):812–9. pmid:26856916
- 24. Liang X, Tan CH, Sun Q, Zhang M, Wong PSJ, Li MI, et al. Wolbachia wAlbB remains stable in Aedes aegypti over 15 years but exhibits genetic background-dependent variation in virus blocking. PNAS Nexus. 2022;1(4):pgac203. pmid:36714832
- 25. Flores HA, Taneja de Bruyne J, O’Donnell TB, Tuyet Nhu V, Thi Giang N, Thi Xuan Trang H, et al. Multiple Wolbachia strains provide comparative levels of protection against dengue virus infection in Aedes aegypti. PLoS Pathog. 2020;16(4):e1008433. pmid:32282862
- 26. Ferguson NM, Hue Kien DT, Clapham H, Aguas R, Trung VT, Bich Chau TN, et al. Modeling the impact on virus transmission of Wolbachia -mediated blocking of dengue virus infection of Aedes aegypti. Sci Transl Med. 2015;7(279).
- 27. Lau MJ, Ross PA, Hoffmann AA. Infertility and fecundity loss of Wolbachia-infected Aedes aegypti hatched from quiescent eggs is expected to alter invasion dynamics. Gillespie JJ, editor. PLoS Negl Trop Dis. 2021;15(2):e0009179.
- 28. Nguyen TH, Nguyen HL, Nguyen TY, Vu SN, Tran ND, Le TN, et al. Field evaluation of the establishment potential of wMelPop Wolbachia in Australia and Vietnam for dengue control. Parasit Vectors. 2015;8:563. pmid:26510523
- 29. Petersen MT, Couto-Lima D, Garcia GA, Pavan MG, David MR, Maciel-de-Freitas R. Dengue exposure and Wolbachia wMel strain affects the fertility of quiescent eggs of Aedes aegypti. Viruses. 2023;15(4):952. pmid:37112932
- 30. Maciel-de-Freitas R, Sauer FG, Kliemke K, Garcia GA, Pavan MG, David MR, et al. Wolbachia strains wMel and wAlbB differentially affect Aedes aegypti traits related to fecundity. Microbiol Spectr. 2024;12(4):e0012824. pmid:38483475
- 31. Bell SM, Katzelnick L, Bedford T. Dengue genetic divergence generates within-serotype antigenic variation, but serotypes dominate evolutionary dynamics. eLife. 2019.
- 32. Allman MJ, Lin Y-H, Joubert DA, Addley-Cook J, Mejía-Torres MC, Simmons CP, et al. Enhancing the scalability of Wolbachia-based vector-borne disease management: time and temperature limits for storage and transport of Wolbachia-infected Aedes aegypti eggs for field releases. Parasit Vectors. 2023;16(1):108. pmid:36934294
- 33. Ross PA, Gu X, Robinson KL, Yang Q, Cottingham E, Zhang Y. A w AlbB Wolbachia transinfection displays stable phenotypic effects across divergent Aedes aegypti mosquito backgrounds. Dozois CM, editor. Applied and Environmental Microbiology. 2021;87(20):e01264-21.
- 34. Gu X, Ross PA, Rodriguez-Andres J, Robinson KL, Yang Q, Lau M-J, et al. A wMel Wolbachia variant in Aedes aegypti from field-collected Drosophila melanogaster with increased phenotypic stability under heat stress. Environ Microbiol. 2022;24(4):2119–35. pmid:35319146
- 35. Terradas G, Joubert DA, McGraw EA. The RNAi pathway plays a small part in Wolbachia-mediated blocking of dengue virus in mosquito cells. Sci Rep. 2017;7(1):43847.
- 36. Ohm JR, Teeple J, Nelson WA, Thomas MB, Read AF, Cator LJ. Fitness consequences of altered feeding behavior in immune-challenged mosquitoes. Parasit Vectors. 2016;9:113. pmid:26927687
- 37. Huxley PJ, Murray KA, Pawar S, Cator LJ. The effect of resource limitation on the temperature dependence of mosquito population fitness. N/A. 2023.
- 38.
Su YS, Yajima M, Baio G. R2jags: A package for running jags from R. 2012.
- 39.
Plummer M, Stukalov A, Denwood M. rjags: Bayesian graphical models using MCMC. 2014.
- 40.
R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing; 2019.
- 41. Plummer M, Cowles K, Best N. CODA: convergence diagnosis and output analysis for MCMC. RNews. 2006;6:7–11.
- 42.
Therneau T, Lumley T. Package survival: a package for survival analysis in R. 2015.
- 43. Carrington LB, Tran BCN, Le NTH, Luong TTH, Nguyen TT, Nguyen PT, et al. Field- and clinically derived estimates of Wolbachia-mediated blocking of dengue virus transmission potential in Aedes aegypti mosquitoes. Proceedings of the National Academy of Sciences of the United States of America. 2018;115(2):361–6.
- 44. Terradas G, Allen SL, Chenoweth SF, McGraw EA. Family level variation in Wolbachia-mediated dengue virus blocking in Aedes aegypti. Parasit Vectors. 2017;10(1):622. pmid:29282144
- 45. Yeap HL, Rašić G, Endersby-Harshman NM, Lee SF, Arguni E, Le Nguyen H, et al. Mitochondrial DNA variants help monitor the dynamics of Wolbachia invasion into host populations. Heredity (Edinb). 2016;116(3):265–76. pmid:26531251
- 46. Rasgon JL, Cornel AJ, Scott TW. Evolutionary history of a mosquito endosymbiont revealed through mitochondrial hitchhiking. Proc Biol Sci. 2006;273(1594):1603–11. pmid:16769630
- 47. Gloria-Soria A, Ayala D, Bheecarry A, Calderon-Arguedas O, Chadee DD, Chiappero M, et al. Global genetic diversity of Aedes aegypti. Mol Ecol. 2016;25(21):5377–95. pmid:27671732
- 48. Fansiri T, Fontaine A, Diancourt L, Caro V, Thaisomboonsuk B, Richardson JH, et al. Genetic mapping of specific interactions between Aedes aegypti mosquitoes and dengue viruses. PLoS Genet. 2013;9(8):e1003621. pmid:23935524
- 49. Gloria-Soria A, Armstrong PM, Powell JR, Turner PE. Infection rate of Aedes aegypti mosquitoes with dengue virus depends on the interaction between temperature and mosquito genotype. Proceedings of the Royal Society B. 2017;284(1864):20171506.
- 50. Salgueiro P, Restrepo-Zabaleta J, Costa M, Galardo AKR, Pinto J, Gaborit P, et al. Liaisons dangereuses: cross-border gene flow and dispersal of insecticide resistance-associated genes in the mosquito Aedes aegypti from Brazil and French Guiana. Mem Inst Oswaldo Cruz. 2019;114:e190120. pmid:31553370
- 51. Axford JK, Ross PA, Yeap HL, Callahan AG, Hoffmann AA. Fitness of wAlbB Wolbachia infection in Aedes aegypti: parameter estimates in an outcrossed background and potential for population invasion. Am J Trop Med Hyg. 2016;94(3):507–16. pmid:26711515
- 52. Pinto SB, Riback TIS, Sylvestre G, Costa G, Peixoto J, Dias FBS, et al. Effectiveness of Wolbachia-infected mosquito deployments in reducing the incidence of dengue and other Aedes-borne diseases in Niterói, Brazil: a quasi-experimental study. PLoS Negl Trop Dis. 2021;15(7):e0009556. pmid:34252106
- 53. Kambris Z, Cook PE, Phuc HK, Sinkins SP. Immune activation by life-shortening Wolbachia and reduced filarial competence in mosquitoes. Science. 2009;326(5949):134–6. pmid:19797660
- 54. Molloy JC, Sommer U, Viant MR, Sinkins SP. Wolbachia modulates lipid metabolism in Aedes albopictus mosquito cells. Applied and Environmental Microbiology. 2016;82(10):3109–20.