Antibodies to Aedes aegypti D7L salivary proteins as a new serological tool to estimate human exposure to Aedes mosquitoes

Abstract

Introduction: Aedes spp. are the most prolific mosquito vectors in the world. Found on every continent, they can effectively transmit various arboviruses, including the dengue virus which continues to cause outbreaks worldwide and is spreading into previously non-endemic areas. The lack of widely available dengue vaccines accentuates the importance of targeted vector control strategies to reduce the dengue burden. High-throughput tools to estimate human-mosquito contact and evaluate vector control interventions are lacking. We propose a novel serological tool that allows rapid screening of human cohorts for exposure to potentially infectious mosquitoes.

Methods: We tested 563 serum samples from a longitudinal pediatric cohort study previously conducted in Cambodia. Children enrolled in the study were dengue-naive at baseline and were followed biannually for dengue incidence for two years. We used Western blotting and enzyme-linked immunosorbent assays to identify immunogenic Aedes aegypti salivary proteins and measure total anti-Ae. aegypti IgG.

Results: We found a correlation (rs=0.86) between IgG responses against AeD7L1 and AeD7L2 recombinant proteins and those to whole salivary gland homogenate. We observed seasonal fluctuations of AeD7L1+2 IgG responses and no cross-reactivity with Culex quinquefasciatus and Anopheles dirus mosquitoes. The baseline median AeD7L1+2 IgG responses for young children were higher in those who developed asymptomatic versus symptomatic dengue.

Discussion: The IgG response against AeD7L1+2 recombinant proteins is a highly sensitive and Aedes specific marker of human exposure to Aedes bites that can facilitate standardization of future serosurveys and epidemiological studies by its ability to provide a robust estimation of human-mosquito contact in a high-throughput fashion.

Keywords: Aedes; Cambodia; ELISA; dengue; exposure marker; mosquito saliva.

Figure 1

Sera from children living in dengue-endemic areas in Kampong Speu Province, Cambodia, recognize native salivary gland homogenate (SGH) from Aedes aegypti. (A) Western blotting displaying reactivity of Cambodian serum samples organized by increasing Ae. aegypti SGH ELISA optical density (OD) values, classified as “low” when OD < 0.162 (lower than the first quartile), “high” when OD > 0.4475 (higher than the third quartile), and “medium” if the OD fell in between the first and third quartiles. (B) Coomassie blue stained sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) of the five selected recombinant proteins (AeApyrase, AeD7L1, AeD7L2, NIH-23, and NIH-27). (C) Western blotting with AeApyrase, AeD7L1, AeD7L2, NIH-23, and NIH-27 against a pool of negative (left) and positive (right) sera.

Figure 2

Heat map showing Pearson’s (A) and Spearman’s (B) correlation coefficients (r and r_s, respectively) between antibody responses against salivary gland homogenate (SGH) and against AeD7L1+2, AeD7L1, or AeD7L2, alone or when combined with AeApyrase, AeNIH-27, or AeNIH-23.

Figure 3

Seasonal variation in anti-Aedes aegypti IgG levels. Dot and box plots of standardized ELISA optical density (OD) values of 18 individuals tested at seven different time points (Visits 1 to 7) for IgG against (A) salivary gland homogenate (SGH), (B) AeD7L1+2, (C) AeD7L1, and (D) AeD7L2. Visits 1, 3, 5, and 7 took place in wet seasons from 2018 to 2021; the other visits took place during the dry seasons. The solid black horizontal line within the box plots is the median; the lower and upper borders are, respectively, the first and third quartiles; the vertical bars indicate the minimum and maximum values. The orange smoothed fitted curves represent a loess regression (span α = 0.45) with 95% confidence intervals (shaded area). Results of the overall Friedman’s rank test are provided.

Figure 4

Trends in baseline anti-Aedes aegypti IgG responses. (A) Exponentiated regression coefficient estimates for the generalized linear model fitted to the IgG data per assay. The circles represent the exponentiated mean coefficient estimate (the value shown above the corresponding circle). The horizontal lines represent the 95% confidence intervals of the estimate. The referent group was “asymptomatic” for both disease outcomes and the age × disease outcome interaction term. (B) Scatterplot of standardized ELISA IgG optical density (OD) values with age at study baseline for both SGH and AeD7L1+2 (n = 563). The solid lines visualize the marginal effect of the age × disease outcome interaction from the generalized linear model (GLM) regression fitted to the antibody data. (C) Dot and box plots of IgG OD values for SGH and AeD7L1+2, comparing different dengue disease outcomes for children younger or older than 6 years. The solid black horizontal line within the box plots is the median; the lower and upper borders are, respectively, the first and third quartiles; the vertical bars indicate the minimum and maximum values. Results of the overall Kruskal–Wallis test and the post-hoc Dunn’s test for pairwise comparisons are provided. GLM, generalized linear model; OD, optical density; NE, no event; AS, asymptomatic dengue; S, symptomatic dengue; SGH, salivary gland homogenate. Significance levels **p < 0.01; *p < 0.05; °p < 0.1.

Figure 5

Specificity of AeD7L1+2 recombinant proteins to Aedes mosquitoes. (A) Western blotting showing reactivity of Cambodian sera to salivary gland homogenate (SGH) of Aedes aegypti, Aedes albopictus, Culex quinquefasciatus, and Anopheles dirus without (left) and with (right) pre-incubation with AeD7L1+2 recombinant proteins. (B) Multiple alignments of D7L proteins from Ae. aegypti, Ae. albopictus, and Cx. quinquefasciatus using MUSCLE. The matrix indicates the percentages of identities between D7L proteins of different species. Color shading indicates percentage identity between sequences, going from 100% (black) to 0% identity (white). The evolutionary history was inferred using the maximum likelihood method. Evolutionary analyses were conducted in Geneious using the Paup plugin.