Dengue Virus Isolation in Mosquito Aedes albopictus Captured During an Outbreak in Tokyo,2014,by a Method Relying on Antibody-Dependent Enhancement Mechanism Using FcγR-Expressing BHK Cells

Introduction

It is estimated that dengue virus (DENV) infects up to 4 billion people living in tropical and subtropical areas of the world every year (Bhatt et al. 2013). The virus is transmitted by mosquitoes, mainly Aedes aegypti and Aedes albopictus mosquitoes. Isolation and detection of viruses in mosquitoes are important in the determination of viral transmission cycles.

Molecular tools such as the RT-PCR and virus isolation are widely used for virus detection (Gubler et al. 1984, Chow et al. 1998, Yamada et al. 2002). However, confirmation of infectious virus in transmission cycles and virus characterization require virus isolation. In contrast to virus detection by using RT-PCR, conventional virus isolation methods generally possess lower sensitivity.

In our previous study, we have applied the antibody-dependent enhancement (ADE) mechanism to isolate virus in serum samples of dengue patients by using FcγR-expressing baby hamster kidney (BHK) cells and a mouse monoclonal antibody with ADE activity, mAb 4G2 (Moi et al. 2011). The novel method efficiently isolated DENV from patient serum samples, in comparison with other cell lines, the BHK, Vero, and C6/36 cell lines, which are used in conventional virus isolation assays (Moi et al. 2011). In this study, we applied the ADE mechanism in virus isolation by using A. albopictus mosquitoes that were captured during a DENV outbreak in Tokyo in 2014 (Kutsuna et al. 2015, Seki et al. 2015).

Materials and Methods

Mosquitoes identified as A. albopictus were divided into 48 pools with an average of 21 mosquitoes per pool (21 ± 4). The mosquitoes were collected in sweeping nets at various parks located in Shinjuku and Shibuya wards, Tokyo, during the 2014 autochthonous dengue outbreak using previously described methods (Kobayashi et al. 2014). The pooled mosquitoes were then stored at −80°C until virus detection using virus isolation and conventional RT-PCR (Ejiri et al. 2015). The supernatant of the homogenates was then passed through sterile filters and inoculated onto the C6/36 cells and FcγR-expressing BHK cells. C6/36 cells, BHK cells, and FcγR-expressing BHK cells were cultured as previously described (Moi et al. 2011, Ejiri et al. 2015). C6/36 cells and FcγR-expressing BHK cells were used for virus isolation from mosquito homogenates and BHK cells were used for plaque titration.

For virus isolation using C6/36 cells, 100 μL of supernatant was added onto monolayers in 24 wells and incubated at 28°C for 2 h for virus adsorption. After 2 h incubation, 500 μL of fresh medium was added to the cells and the cells were further incubated at 28°C with 5% CO₂ for 7 days. For virus isolation using FcγR-expressing BHK cells, 5 μL of mAb 4G2 (40 μg/mL) was added to 95 μL mosquito homogenate to make up a final antibody concentration of 2 μg/mL. A concentration of 2 μg/mL has been shown to demonstrate ADE activity to DENV (Moi et al. 2011). After incubation at 37°C for 30 min, 100 μL of mosquito homogenate–antibody mixture was inoculated onto FcγR-expressing BHK monolayers in 24-well plates. The plates were then incubated for 60 min at 37°C in 5% CO₂. After incubation, 0.5 mL of fresh medium was added. Culture supernatant fluids were collected after 5 days of incubation at 37°C in 5% CO₂. Viruses were titrated using quantitative RT-PCR and plaque assay as previously described (Ito et al. 2004, Moi et al. 2011).

Results and Discussion

A total of 998 A. albopictus mosquitoes collected during the autochthonous dengue outbreak in Tokyo were divided into 48 pools of mosquitoes. Of the 48 pools, 16 pools of mosquitoes that were positive for DENV genome, as determined by RT-PCR (Ejiri et al. 2015), were used in this study. Of the 9 pools tested positive for virus genome, only 1 mosquito pool was positive for plaques in BHK cells (pool No. 2, 1.7 log₁₀ PFU/mL) (Table 1).

Of the 16 pools of mosquitoes used for virus isolation, DENV genome was detected in all 16 samples of cell culture supernatant fluids (5.0–10.6 log₁₀ genome copies/mL) when FcγR-expressing BHK cells and mAb 4G2 were used. In comparison, DENV genome was detected in 15 samples of cell culture supernatant fluids (15/16) when C6/36 cells were used as assay cells (4.1–7.6 log₁₀ genome copies/mL, Table 1). Taken together, the levels of virus genome in cell culture supernatant fluids of FcγR-expressing BHK cells (9.1 ± 2.1 log₁₀ genome copies/mL) were consistently higher than those of C6/36 cells (6.9 ± 0.9 log₁₀ genome copies/mL; p < 0.01). In addition, DENV was isolated from a mosquito pool (No. 14) only when FcγR-expressing BHK cells were used as assay cells.

Plaque assay was then used to determine infectious virus titers. Using this method, higher virus yields of up to 10,000 times were detected in cell culture supernatant fluids of FcγR-expressing BHK cells (5.5–8.3 log₁₀ PFU/mL) but not in C6/36 cell culture supernatant fluids (2.5–3.6 log₁₀ PFU/mL). In addition, by using the plaque titration method, DENV isolation was confirmed in six samples (Nos. 1, 4, 5, 9, 13, and 16) only when FcγR-expressing BHK cells and ADE antibody were used (Table 1). Although virus titration by the plaque method is specific for the detection of infectious virus, the method is less sensitive than real-time PCR.

Consequently, because of lower DENV yields in the cell culture supernatant fluids of C6/36 cells, the number of samples that were positive by using the plaque titration method was also lower than those by using the virus isolation method by using the FcγR-expressing BHK cells. In concordance with findings of a previous study using human serum samples, DENV was exclusively isolated in a sample (mosquito pool No. 14) when FcγR-expressing BHK cells and an antibody with ADE activity were used (Moi et al. 2011). The comparison of efficiency of virus isolation from mosquito pools using the cell lines with that of other virus isolation methods should be addressed in a next series of studies, by using a larger number of samples and different assay conditions. Overall, the results suggest that the method using FcγR-expressing BHK cells and an antibody with ADE activity offers an alternative method for the isolation of infectious virus in field-caught A. albopictus mosquitoes, particularly in samples with low virus concentration or those that were negative for virus isolation by using the C6/36 cell lines.

In this study, higher virus yield was confirmed when FcγR-expressing cells and an antibody with ADE activity were used. The results confirm that infectious viruses were present in A. albopictus mosquitoes during the dengue outbreak in Tokyo in 2014, and suggest that the novel virus isolation method using the FcγR-expressing BHK cells as assay cells is useful for confirmation of infectious virus in mosquito vector and for the characterization of virus transmission in vectors during an outbreak.