The antiviral JNJ-A07 significantly reduces dengue virus transmission by Aedes aegypti mosquitoes when delivered via blood-feeding

Abstract

Dengue virus (DENV) is the most widespread mosquito-borne virus worldwide, but no antiviral therapies are available yet. The pan-serotype DENV inhibitor JNJ-A07 has shown potent activity in a mouse model. It remains unknown whether an antiviral drug ingested by mosquitoes could inhibit virus replication and thus reduce transmission to other hosts. Here, we investigated the antiviral activity of JNJ-A07 when administered in the blood meal to Aedes aegypti mosquitoes. JNJ-A07 blocked DENV-2 transmission by the mosquitoes in both pre-exposure and post-exposure settings. In addition, JNJ-A07 remained in the mosquito bodies for 7 days after blood meal. Reductions of DENV systemic infection in the mosquitoes suggested a potential for decreased proportions of DENV outbreaks in a simulated environment when the mosquitoes ingested JNJ-A07 via the blood meal.

Fig. 1.. In vitro activity of JNJ-A07 against DENV-2 and cell toxicity in mosquito cell lines.

Intracellular viral RNA compared to the virus control (normalized and depicted as percentage of control) was measured by qRT-PCR to evaluate the antiviral activity of JNJ-A07 against DENV-2 in Ae. albopictus (C6/36) (A) and in Ae. aegypti (Aag2-AF5) (B) derived cells. The effect of JNJ-A07 on the viability of C6/36 and Aag2-AF5 cells was measured by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay. The CC₅₀ value is indicated by a dotted line perpendicular to the x axis. The EC₅₀ and CC₅₀ were calculated by fitting a nonlinear regression on the empirical data corresponding to two (antiviral activity) and four (toxicity) independent replicates, each with three biological replicates per concentration. Each dot represents the median, and the error bars correspond to the 95% confidence interval.

Fig. 2.. Antiviral activity of JNJ-A07 in ex vivo cultured guts of Ae. aegypti.

(A) DENV RNA levels in guts when incubated in the presence of JNJ-A07 for 5 days, measured at day 5 pi by qRT-PCR. Vehicle, virus control in presence of DMSO (untreated); dpi, days post infection. DMSO concentration was kept at 0.005% in both control and compound conditions. Statistical significance was evaluated with a Kruskal-Wallis test (P < 0.0001) and Dunn’s multiple comparison test (vehicle versus 0.2 μM, ***P = 0.0001; vehicle versus 2 μM, ****P < 0.0001). (B) DENV RNA levels in guts exposed to JNJ-A07 (2 μM) starting at 0 (time of infection), 1, and 3 days pi, measured at day 5 pi by qRT-PCR. Statistical significance was evaluated with a Kruskal-Wallis test (P < 0.0001) and Dunn’s multiple comparison test (vehicle versus 0 dpi, ****P < 0.0001; vehicle versus 1 dpi, ***P = 0.0001; vehicle versus 3 dpi, *P = 0.0145). Each dot represents a mosquito gut. Black lines represent the median, and the colored lines correspond to the 95% confidence interval per condition. Dotted lines depict the limit of detection (LOD) of the assay. Data shown correspond to two independent experiments.

Fig. 3.. Effect of JNJ-A07 on survival and fecundity of Ae. aegypti mosquitoes.

(A) Mosquito longevity after feeding on a blood meal with JNJ-A07 at 25 and 100 μM, and the vehicle (DMSO, control group) is depicted as percentage of survival. The blood meal for all conditions contained a final concentration of 1% DMSO. Mosquito sample size is indicated per condition by n. Statistical significance was assessed by two-sided log-rank (Mantel-Cox) test, P = 0.2034. ns, not significant. (B) Number of eggs developed in the ovaries of mosquitoes that fed on a blood meal with JNJ-A07 at 2 (n = 85) and 25 μM (n = 96), and the vehicle (n = 61) and PBS (n = 122) as control groups. The blood meal for all conditions contained a final concentration of 0.5% DMSO. Each dot represents data from one mosquito, and the lines indicate the median. Statistical significance was assessed with a Kruskal-Wallis test (P = 0.0622) and Dunn’s multiple comparison test (vehicle versus PBS, P = 0.0610; vehicle versus 2 μM, P = 0.2738; vehicle versus 25 μM, P = 0.0864; PBS versus 2 μM, P > 0.9999; PBS versus 25 μM, P > 0.9999). Data shown correspond to two (survival) and three (fecundity) independent experiments.

Fig. 4.. Dose-response effect of JNJ-A07 on the vector competence of Ae. aegypti mosquitoes for DENV-2.

DENV infection (A), disseminated infection (B), and transmission (C) rates for mosquitoes that fed on a DENV-infectious blood meal spiked with JNJ-A07 at 0.02, 0.05, 0.2, 2, and 25 μM, and the vehicle (control group) were calculated on the basis of the detection of infectious virus at days 3 and 7 pi by plaque assay or focus-forming assay. Sample size across all blood-feeding experiments is depicted for each condition and time point in (D). Each symbol represents an independent blood-feeding experiment where said condition was tested, and the bar height corresponds to the mean rate. Percentages appear above the bar when it is different than 0%. Statistical significance was evaluated with a Fisher’s exact test (IRs at 3 and 7 dpi, ****P < 0.0001; DIRs at 3 dpi, P = 0.0677, and 7 dpi, ****P < 0.0001). ns, not significant.

Fig. 5.. Efficacy of JNJ-A07 on DENV-2 infection in mosquitoes when ingested via the blood meal in a prophylactic setting.

(A) Timeline depicting the design of the experiment, in which mosquitoes received a blood meal spiked with JNJ-A07 at 2 μM or the vehicle (control group), and 6 days later, they received a second blood meal containing DENV-2. Mosquitoes were euthanized at day 7 pi (i.e., 13 days after exposure). Infectious virus was detected by plaque assay. DENV infection (B), disseminated infection (C), and transmission (D) rates for mosquitoes that were exposed to JNJ-A07 before the infectious blood meal. Each dot or square represents an independent blood-feeding experiment in (B) to (D), and the bar height corresponds to the mean. Mosquito sample size is indicated per condition by n. Percentages appear above the bar. Statistical significance was evaluated with a Fisher’s exact test (IRs, ****P < 0.0001; DIRs, *P = 0.0191; TRs, ***P = 0.0004). Infectious virus (E) and DENV RNA levels (F) in the mosquito bodies when pre-exposed to the vehicle and 2 μM of JNJ-A07, measured by plaque assay and qRT-PCR, respectively. Each circle or square represents a biological replicate (one mosquito) in (E) and (F), and the black lines correspond to the median. The dotted line represents the LOD of the assay. Mosquito sample size is indicated per condition by n. Statistical significance in (E) and (F) was assessed with a Mann-Whitney test (****P < 0.0001). Data shown correspond to three independent blood-feeding experiments. gc, genome copies.

Fig. 6.. Efficacy of JNJ-A07 on DENV-2 kinetics when ingested via the blood meal in a therapeutic setting.

(A) Timeline depicts the design of the experiment, in which mosquitoes received a blood meal containing DENV-2, and 6 days later, they received a second blood meal spiked with JNJ-A07 at 2 μM or the vehicle (control group). Mosquitoes were euthanized at day 7 post-exposure (pe) (i.e., 13 dpi). Infectious virus was detected by plaque assay. DENV infection (B), disseminated infection (C), and transmission (D) rates for mosquitoes that were exposed to JNJ-A07 after the infectious blood meal. Each dot or square represents an independent blood-feeding experiment in (B) to (D), and the bar height corresponds to the mean. Mosquito sample size is indicated per condition by n. Percentages appear above the bar. Statistical significance was evaluated with a Fisher’s exact test (IRs, P = 0.1295; DIRs, ****P < 0.0001; TRs, P = 0.5503). ns, not significant. Infectious virus (E) and DENV RNA levels (F) in the bodies of mosquitoes exposed after infection to the vehicle and JNJ-A07 measured by plaque assay and qRT-PCR, respectively. (G) DENV RNA levels in the heads, wings, and legs of mosquitoes exposed after infection to the vehicle and JNJ-A07 measured by qRT-PCR. Each circle or square represents a biological replicate (one mosquito) in (E) to (G), and the black lines correspond to the median. The dotted line represents the LOD of the assay. Mosquito sample size is indicated per condition by n. Statistical significance in (E) to (G) was assessed with a Mann-Whitney test (****P < 0.0001). Data shown correspond to two independent experiments. H+W+L, head, wings, and legs. LOD, limit of detection of the corresponding assay.

Fig. 7.. Bodyweight– and concentration-time course of JNJ-A07 in Ae. aegypti mosquitoes when delivered via the blood meal.

(A) The bodyweight of each mosquito that fed on a blood meal spiked with the vehicle (control group) or JNJ-A07 at 2 μM was monitored for 7 days after feeding. The black lines represent the median. Each dot or square represents a single mosquito (JNJ-A07, n = 100; vehicle, n = 100). (B) The concentration of JNJ-A07 was measured in 10 mosquitoes per time point for 7 days after feeding by liquid chromatography–mass spectrometry. The line connects the medians for each time point. The dotted line represents the limit of quantification of the assay. Data shown correspond to one blood-feeding experiment.

Fig. 8.. Modeled effect of the reductions in DENV systemic infection, as a result of JNJ-A07 exposure, on the magnitude of DENV outbreaks.

DENV infection (A) and disseminated infection (B) rates for mosquitoes that fed on a DENV-infectious blood meal spiked with JNJ-A07 at 0.02, 0.05, and 2 μM, and the vehicle (control group). Infectious virus was detected at days 3, 5, 7, 11, and 14 pi by plaque assay or focus-forming assay. Each symbol represents an independent blood-feeding experiment. Data for the 2 μM dose were only collected for days 3 and 7 pi; data for days 5, 10, and 14 were extrapolated from available time points. The bar height corresponds to the mean. (C) The cumulative prevalence of systemic infection over time after exposure (and after infection) is displayed for each condition. A logistic model was fitted to all exposure conditions based on three parameters. For each condition, dots represent the empirical data. The size of the dots represents the sample size. The dashes correspond to the 95% confidence interval of the prevalence. The colored line represents the logistic fit of the dataset. (D) Proportion of simulations that resulted in ≥100, <100, or no secondary human infections using the K, B, and M values empirically measured. Statistical significance was assessed by the test of Equal or Given proportions (0.02 μM versus 0.05 μM, *P = 0.001; 0.02 μM versus vehicle, P = 0.182; 0.05 μM versus vehicle, *P = 0).