Characterizing and Quantifying Arbovirus Transmission by Aedes aegypti Using Forced Salivation and Analysis of Bloodmeals

Abstract

Arbovirus transmission studies are dependent on the ability to estimate the titer of virus transmitted from infectious mosquitoes to a host. There are several methods for estimating virus titer in mosquito saliva, including (1) using forced salivation (FS) whereby the infectious mosquito's proboscis is forced into a capillary tube containing media to collect and test their saliva for virus, and (2) by quantifying virus expectorated into host tissues or into the blood contained in an artificial feeder immediately after blood feeding. We studied FS and bloodmeals to estimate and compare titers of Zika virus and chikungunya virus transmitted by the mosquito vector Aedes aegypti. Infectious virus and viral genomes of both viruses were detected more often from individual mosquitoes using immersion oil for the FS media compared to fetal bovine serum (FBS) plus glycerol, but the FS media had no influence on virus quantification from positive samples. FS virus titers were equivalent when comparing individuals or groups of mosquitoes that never received a blood meal compared to those that were blood fed immediately prior, showing that blood feeding does not influence FS. This suggested that performing FS on mosquitoes after blood feeding might be an efficient way to estimate virus transmitted during blood feeding. However, detecting virus from the blood remaining in an artificial feeder post-blood feeding was mostly unsuccessful relative to quantifying virus from FS of the post-blood fed mosquitoes. In contrast, immunocompromised mice always became infected after being fed on by Zika-infected mosquitoes, even when no infectious virus was detected in their saliva by FS post-blood feed. Due to this discrepancy, we tested the ingested bloodmeals of individual mosquitoes that fed on artificial blood feeders for virus, and compared these to virus in their saliva harvested from FS and to virus in their bodies. These experiments revealed ~50-100 times higher virus titers in the dissected bloodmeals compared to those detected in the same mosquitoes' saliva, demonstrating how mosquitoes re-ingest much of their saliva during artificial blood feeding, and highlighting a large increase in virus transmission during Aedes aegypti blood feeding. Both FS and the dissected bloodmeals of artificially blood-fed mosquitoes showed that the quantity of viral RNA expectorated by mosquitoes was 2-5 logs more than the quantity of infectious virus. The results from this study add critical information to understanding and quantifying the transmission of Aedes aegypti arboviruses.

Keywords: Aedes aegypti; arboviruses; saliva; transmission.

Figure 1

Graphical picture of experiments completed in this study. Green ball represents Zika virus (ZIKV) or chikungunya virus (CHIKV). Green thorax mosquitoes = mosquitoes infected by infectious bloodmeal as shown in panel 1.

Figure 2

Quantification of virus and viral RNA from virus-positive forced salivations using oil or FBS-glycerol as the collection media; titer and genome copies of virus-positive saliva samples from individual mosquitoes infected with ZIKV (A,B) or CHIKV (C,D). LOD = limit of detection. Titer LOD = 2 PFU; genome copies LOD = 10. The means of virus titers or genome copies from saliva collected by either method (horizontal bars), regardless of virus, were not significantly different (un-paired t-test; p ≥ 0.05).

Figure 3

Quantification of ZIKV and CHIKV from groups of virus-positive saliva-samples collected from forced salivation Scheme 2 (A,B) or CHIKV (C,D; displayed in Table 3). LOD = limit of detection. Titer LOD = 2 PFU; genome copies LOD = 10. Means of virus titers or genome copies from saliva collected by either method (horizontal bars) were not significantly different (one-way ANOVA and post hoc Tukey test; p ≥ 0.05).

Figure 4

Log transformed difference between ZIKV and CHIKV detected in the bloodmeals and saliva of individual infected mosquitoes after they were given different blood feeding treatments. (A) Infectious ZIKV mean difference: None = 2.21, Artificial feeder = 3.65, Artificial feeder + mixing = 3.69. (B) ZIKV genome copies mean difference: none = 5.05, artificial feeder = 7.06, artificial feeder + mixing = 6.35. (C) Infectious CHIKV mean difference: none = 2.06, artificial feeder = 3.41, artificial feeder + mixing = 3.73. (D) CHIKV genome copies mean difference: none = 5.19, artificial feeder = 6.82, artificial feeder + mixing = 6.65. Graphs show both PFUs determined by plaque assay and RNA determined by qRT-PCR. Titer LOD = 2 PFU; genome copy LOD = 10 genomes. Means are horizontal bars (one-way ANOVA and post hoc Tukey test; * p ≤ 0.05).