Vertical transmission of Zika virus in Culex quinquefasciatus Say and Aedes aegypti (L.) mosquitoes

Several mosquito species have been described as vectors for the Zika virus (ZIKV), such as those in the Aedes, Anopheles, Mansonia and Culex genera. Our previous survey studies were found the ZIKV RNA positive in both male, female and larvae of Culex quinquefasciatus Say and Aedes aegypti (L.) mosquitoes collected from active ZIKV infected patients’ homes in Thailand. Therefore, the aims of this study were to investigate whether ZIKV could be vertically transmitted in Cx. quinquefasciatus, Ae. aegypti and Ae. albopictus. Laboratory and field colonies of these mosquito species were maintained and artificially fed with ZIKV in human blood. Fully engorged mosquitoes (F0) were selected and reared for the vertical transmission study. The subsequent mosquito generations were fed with human blood without the virus. ZIKV in the mosquitoes was detected by hemi-nested RT-PCR and sequencing. C6/36 cells were used to isolate ZIKV from samples that tested positive by hemi-nested RT-PCR. Moreover, ZIKV was identified by immunocytochemical staining 7 days after infection in several organs of infected F0 females, including the salivary glands, midguts, yoke granules and facet cells of the eye. The localization of the ZIKV antigen was identified by the presence of the specific antibody in the salivary glands, midguts, yoke granules and facet cells. ZIKV was detected in female and male Cx. quinquefasciatus until the F6 and F2 generations, respectively. The isolated virus showed cytopathic effects in C6/36 cells by 5 days postinfection. The results suggested that the vertical transmission of ZIKV occurs in Cx. quinquefasciatus in the laboratory. However, we were able to detect the presence of ZIKV in Ae. aegypti in only the F1 generation in both male and female mosquitoes, and Ae. albopictus mosquitoes were not able to vertically transmit the virus at all. Data obtained from this study could be valuable for developing a better understanding of the role of Cx. quinquefasciatus as a potential vector for ZIKV transmission in Thailand and may be useful in creating more effective mosquito vector control strategies in the future.

. In Southeast Asia, ZIKV was isolated from wild-caught Ae. aegypti in Malaysia 11 ; moreover, Ae. aegypti 12 and Ae. albopictus 13 were reported as potential vectors of ZIKV transmission in Singapore. However, knowledge of the vectors of ZIKV transmission in Thailand is limited. In Thailand a total of 686 of confirmed Zika case had been reported between January and November 2016 14 , by the Ministry of Public Health (MoPH), especially in Phitsanulok and Chanthaburi provinces where Cx. quinquefasciatus were found naturally ZIKV infections. In addition, the MoPH also reported the first two indigenous cases of ZIKV-related microcephaly in Thailand 15 . In 2018, the Bureau of Epidemiology (BoE) of the MoPH revealed that 306 Zika case were reported from 22 provinces in January-August 2018 16 . Our previous studies, ZIKV RNA have been found in 1.85% female, 1.66% male, and 0.29% larva of Cx. quinquefasciatus mosquitoes collected from active ZIKV infected patients' homes are infected with ZIKV in Thailand 17 . Therefore, in this study, we determined the potential for the Thai Cx. quinquefasciatus mosquito to vertically transmit ZIKV. The hemi-nested RT-PCR (hnRT-PCR) developed for this study is able to effectively detect ZIKV in mosquitoes. Information obtained from this study provides fundamental data regarding whether Ae. aegypti, Ae. albopictus and Cx. quinquefasciatus are capable of vertically transmitting ZIKV in the laboratory. There are no vaccines or specific therapies against ZIKV. Data regarding ZIKV infection in these mosquitoes will be valuable in developing effective control strategies for ZIKV infection in Thailand.

Results
Vertical transmission of ZIKV in Ae. aegypti, Ae. albopictus and Cx. quinquefasciatus. The mosquitoes were maintained and artificially fed with 1.7 × 10 5 florescent focus units (FFU)/ml of ZIKV in human blood. The experiments were performed in triplicate. Progeny of Ae. aegypti, Ae. albopictus and Cx. quinquefasciatus exposed to ZIKV were reared to subsequent generations. ZIKV RNA was detected in each pooled generation of mosquitoes.
In field strains of Ae. aegypti, a total of 120 pool consisting of 2,400 F 1 female adults and 30 pool consisting of 900 F 1 male adults were tested in triplicate. We found that the male and female mosquitoes exposed to ZIKV can maintain the virus for only one generation and were showed infected at rates of 3.3% and 3.3%, respectively. In the laboratory strain, two pools of female F1 progeny were positive with infection rate of 1.7%, and ZIKV was not detected in the F1 generation of the male mosquito (Table 1). In both the laboratory and field strains of Ae. albopictus, the results showed that ZIKV was not detected in the offspring.
In Cx. quinquefasciatus, ZIKV RNA could be detected until the F 6 generation in females and the F 2 generation in males. The F 1 generation had the highest level of ZIKV infection (29.2%), which decreased to 24.2%, 8.3%, 7.5%, 5.8%, 0.83% and 0% in the F 2 to F 7 generations of female mosquitoes, respectively. However, in male mosquitoes, the transmission of ZIKV was only found by using hnRT-PCR until the F 2 generation, decreasing from 20% in F 1 to 16.7% in F 2 and then undetectable levels in F 3 ( Table 2). The nucleotide sequences of all positive samples showed 99-100% similarity to the Zika virus/H. sapiens-tc/THA/2014/SV0127-14 (accession number KU681081) that was used to infect the F 0 generation; this strain belongs to the Asian lineage of the ZIKV. The entire sequences were assigned GenBank numbers (GenBank: MK028538-MK028557).   www.nature.com/scientificreports www.nature.com/scientificreports/ Isolation of positive ZIKV from each generation of mosquitoes in C6/36 cells. Samples that were positive for ZIKV RNA in each generation of mosquitoes were used to isolate the virus by inoculating C6/36 cells, and the morphological changes of the infected cells were compared with the morphology of uninfected cells (Fig. 1A). The ZIKV infection in C6/36 was monitored microscopically for cytopathic effects (CPEs) at days 3 ( Immunocytochemistry (ICC) staining of ZIKV in mosquitoes. The F 0 generation of female Cx. quinquefasciatus mosquitoes that were infected with ZIKV were dissected 7 days postinfection to obtain their salivary glands, embryos, midguts, and heads. ZIKV antigens were detected in the salivary glands, midguts, yoke granules and facet cells of the eyes by ICC assay (Fig. 2). The dissection of the heads of the mosquitoes allowed investigation of the salivary glands, midguts and compound eyes. Positive staining for ZIKV is shown as distinct brown staining caused by the oxidation of 3,3′-diaminobenzidine (DAB) by horseradish peroxidase (HRP) within the organs of the mosquitoes ( Fig. 2A). In the salivary glands, which are essential for transmission, ZIKV antigens were detected in the three lobes, especially in the distal lateral lobes, which play a major role in the blood feeding process (Fig. 2C). ZIKV can replicate in the midguts (Fig. 2E), yoke granules (Fig. 2G) and facet cells (Fig. 2I), all of which displayed localization of the ZIKV antigen-specific ZIKV-NS1 protein antibody, seen as brownish-red staining ( Fig. 2A,C ZIKV infection, dissemination and transmission rates in the F 0 generations of Ae. aegypti, Ae. albopictus, and Cx. quinquefasciatus mosquitoes. The ICC results in F 0 mosquitoes were used to calculate the infection, dissemination and transmission rates in the F 0 generations of Ae. aegypti, Ae. albopictus, and Cx. quinquefasciatus mosquitoes. Mosquitoes that died before 7 days postinfection were excluded from the study. The Ae. aegypti laboratory strain and field strain had infection, dissemination and transmission rates of 88.2% and 90.5%, 71.1% and 67.4% and 60.8% and 60.0%, respectively. The Ae. albopictus laboratory strain and field strain had infection, dissemination and transmission rates of 53.9% and 51.0%, 41.8% and 42.0% and 21.8% and 22.0%, respectively. The Cx. quinquefasciatus laboratory strain had infection, dissemination and transmission rates of 87.5%, 72.4% and 63.3%, respectively (Table 3). www.nature.com/scientificreports www.nature.com/scientificreports/

Calculation of the filial infection rate (FIR) of ZIKV in progeny.
The FIR of ZIKV in the progeny was tested using pooled populations of mosquitoes; these pools consisted of 2,400 and 900 filial female and male adult mosquitoes, respectively. To calculate the FIR, the total number of pools was divided by the number of positive pools. Each positive pool indicated that one or more of the filial progeny in the pool were infected with ZIKV. In www.nature.com/scientificreports www.nature.com/scientificreports/ this study, female Cx. quinquefasciatus mosquitoes had an FIR for ZIKV of 1:66 for F1, which decreased to 1:2,400 in F6. Male Cx. quinquefasciatus mosquitoes had an FIR for ZIKV of 1:150 for F1, which decreased to 1:180 for F2. In the Ae. aegypti laboratory strain, the female mosquitoes had an FIR for F1 of 1:1,200, while the Ae. aegypti field strain female and male mosquitoes had FIRs for F1 of 1:600 and 1:900, respectively. The FIR results for ZIKV infection for the progeny are shown in Table 4.

Discussion
Our results demonstrated that vertical transmission of ZIKV occurs in the Cx. quinquefasciatus mosquito. The study was based on a molecular technique (hnRT-PCR) for ZIKV RNA detection, the isolation of viruses in cell culture and an ICC assay for the localization of a ZIKV-specific antigen. Culex mosquito species have been found to be able to vertically transmit West Nile (WNV) 18 and Japanese encephalitis (JE) 19 viruses. The role of the vertical transmission of ZIKV is still under investigation. Several reports have suggested that ZIKV could be transovarially transmitted to progeny in both laboratory and field experiments. Guo et al. (2016) revealed that Cx. pipiens quinquefasciatus clearly demonstrates the potential to be a vector for ZIKV in Southern China 20 . The Cx. quinquefasciatus laboratory colonies had detectable ZIKV in the midgut and salivary glands after artificial blood feeding with ZIKV; moreover, field-caught Cx. quinquefasciatus tested positive for ZIKV by RT-qPCR in Brazil 21 . In addition, Cx. quinquefasciatus had detectable ZIKV in the salivary glands at 7 and 15 days postfeeding in Northeast Brazil 22 .
In contrast, several reports have shown evidence of a lack of competence of Culex species for ZIKV 23,24 . For example, ZIKV isolated from Cx. pipiens and Cx. quinquefasciatus mosquitoes from the United States were unable to replicate, as determined by a plaque assay 25,26 . Similar reports showed that C. pipiens and other Culex species from Brazil 27 , Germany 28 , Tunisia 29 , Italy 30 and Australia 31 had no detectable ZIKV transmission. Factors that affect the competence of mosquito vectors include the following. (1) Differences in geographic regions mean that mosquito colonies from various areas have unique genetic backgrounds; therefore, mosquitoes collected from different areas may have varying degrees of viral competence [32][33][34] . This genetic variation may affect the morphology of mosquito organs and processes that are involved in virus replication and dissemination, such as mosquito immune responses, small RNA-based interferon (RNAi) pathways [35][36][37] and the midgut and salivary gland barriers 29,38 . (2) Mosquitoes collected from different geographic regions also have different microbiomes and microviromes 39 . Microbiomes have been shown to interfere with viral replication in mosquito vectors; therefore, different microbiomes would affect the competency of the mosquitoes for ZIKV [40][41][42] . With regard to intracellular bacteria in mosquitoes, Wolbachia is another factor that affects viral replication in mosquitoes. A novel strategy to interfere with arbovirus transmission in mosquitoes using Wolbachia pipientis (wPip) has been proposed 43,44 . However, a study by Lourenço-de-Oliveira, et al. (2018) showed that no ZIKV was found in Cx. quinquefasciatus lines whether or not they contained Wolbachia 38 . The effects of the microbiome and Wolbachia in Cx. quinquefasciatus lines in Thailand could be investigated in the future. (3) The differences in the ZIKV strains used for the experiments, including the genotype, titer, and number of passages, could affect the ability of the ZIKV to enter mosquito organs, resulting in differences in replication and dissemination in the mosquitoes [32][33][34][35][36][37][38][39][40][41][42][43][44] . (4) The techniques used in the experiments, such as the mode of ZIKV infection of the mosquitoes (oral, intrathoracic) and the ZIKV detection method (plague assay, molecular techniques, immunological techniques), which have different sensitivities and specificities 32-44 , might also result in different findings.   www.nature.com/scientificreports www.nature.com/scientificreports/ This study is the first report of vertically transmitted ZIKV in Cx. quinquefasciatus and Ae. aegypti mosquitoes in Thailand. We also successfully isolated infectious ZIKV from a C6/36 cell line that had been infected with ZIKV from Cx. quinquefasciatus and and Ae. aegypti. Cx. quinquefasciatus mosquitoes are widely distributed in tropical and subtropical areas 45 . In Thailand, there have been reports of JE virus isolation from Cx. quinquefasciatus using C6/36 cells 46 . However, ZIKV in Cx. quinquefasciatus has never been studied in Thailand. In this study, we investigated the localization of ZIKV antigens in the organs of mosquitoes by ICC. The salivary gland, midguts, yoke granules and facet cells showed a reaction between the ZIKV antigen and its specific antibody. The salivary glands of mosquitoes play an important role in the transmission of pathogens and an essential role in the transmission cycle 47 . This study demonstrated that ZIKV also infects the salivary glands and midguts; therefore, we hypothesized that ZIKV was replicating in the salivary glands and midguts at 7 days postinfection. Several reports have shown the presence of ZIKV in the salivary glands and midguts of female mosquitoes such as Ae. aegypti 48 , Ae. albopictus 49 , Cx. quinquefasciatus 21 , Cx. coronator, and Cx. tarsalis 50 . The ICC assay was first performed on the facet cells and yoke granules of mosquitoes orally infected with ZIKV. The facet cells of the mosquito eyes showed brownish-red staining in the ICC assay. The ZIKV infection, dissemination and transmission rates in the F 0 generations of Ae. aegypti, Ae. albopictus, and Cx. quinquefasciatus mosquitoes were calculated as shown in Table 3. Based on these results, there were no differences in the infection, dissemination or transmission rates in the F 0 generations of the laboratory and field strains in both Ae. aegypti and Ae. albopictus. However, Ae. aegypti has higher infection, dissemination and transmission rates than Ae. albopictus, which is a similar finding to that in a recent report by Liu et al. (2017). High infection, dissemination and transmission rates in Ae. aegypti were also reported by Liu et al. (2017) and Main et al. 24,51 . In this study, we found lower infection, dissemination and transmission rates in Ae. albopictus than in Ae. aegypti, which is similar to the results of the study by Liu et al. (2017). Interestingly, the infection, dissemination and transmission rates of ZIKV in Cx. quinquefasciatus mosquitoes were higher in this study than in previous reports 24,51 . As discussed previously, several factors could affect the competency of mosquitoes for ZIKV, and future studies should be conducted to investigate the factors influencing these results. However, in this study we determined the transmission rates by examining virus infection in salivary glands, not in the expectorated saliva; therefore, the transmission rates in this study may over estimated. Further investigation for transmission rate in expectorated saliva should be perform for obtaining more accurate data.
The FIRs for ZIKV in the F1 generation of Ae. aegypti in both the laboratory and field strains in this study were lower than those in the previous reports by Thangamani et al. 52 and Ciota et al. 53 , which had FIRs of 1:290 and 1:84, respectively. However, for Cx. quinquefasciatus mosquitoes, our results showed that the FIR for ZIKV in the F1 generation was high (1:66) and that it decreased to 1:2,400 in the F6 generation. The high FIR for ZIKV is related to the vertical transmission phenomenon.
The results of this study provide more information about the transmission dynamics of ZIKV in mosquitoes and could be used to explain the natural pathogenesis of ZIKV infection in wild mosquitoes. The presence of ZIKV antigens in the yoke granules may be associated with the vertical transmission of ZIKV, as ZIKV may infect the germinal tissues of the female mosquito and may occur in the fully formed egg during oviposition 54,55 . However, the mechanism of transovarian transmission is still unclear. ZIKV detected in the facet cells could imply that ZIKV also infects the nervous system organs of the mosquitoes.
Unlike Cx. quinquefasciatus, Ae. aegypti had detectable ZIKV in only the F 1 generation, and ZIKV was not detected in the F 2 generation in either the field strain or the laboratory strain (Table 1). ZIKV was detected in only the F 0 generation of Ae. albopictus mosquitoes in both the field and laboratory strains. Vertical transmission of ZIKV in Ae. aegypti mosquitoes was reported by Thangamani et al. (2016); they found that ZIKV was transmitted to the F 1 generation. The current study also demonstrated that ZIKV was not found in any F 1 Ae. albopictus 52 . We therefore conclude that vertical transmission occurred in both Cx. quinquefasciatus and Ae. aegypti and not in Ae. albopictus. Further research should be conducted to explore the factors that might affect these results.
Cx. quinquefasciatus prefers to feed on animal blood, and some studies have suggested that the blood found in Cx. quinquefasciatus is 50% human, 32% bird, 12% dog, and less than 3% cat 56 . In Thailand, Cx. quinquefasciatus is found in urban and suburban areas. Feeding behavior studies in Koh Chang, Thailand showed that only 0.98% of the blood meals ingested by Cx. quinquefasciatus were human 57 . Data regarding the vertical transmission of ZIKV obtained from this study together with the feeding pattern of Cx. quinquefasciatus in Thailand indicate that Cx. quinquefasciatus mosquitoes may be a potential vector for ZIKV transmission in Thailand. Therefore, vector control strategies for addressing ZIKV outbreaks by managing mosquitoes should not focus only on Aedes mosquitoes; in particular, the larval control strategies should also focus on the breeding sites of Cx. quinquefasciatus. The development of vector control measures for ZIKV outbreaks in Thailand should consider both Aedes and Culex mosquitoes. www.nature.com/scientificreports www.nature.com/scientificreports/ Mosquitoes. Laboratory and field colonies of Ae. aegypti, Ae. albopictus and Cx. quinquefasciatus mosquitoes were maintained in the Biology and Ecology Laboratory of the National Institute of Health (NIH), Department of Medical Sciences, Nonthaburi Province, Thailand, under standard conditions as follows: 28 ± 2 °C, 65-85% relative humidity, and a 12/12-hour light/dark cycle. These mosquitoes were originally obtained from eggs laid in Nonthaburi Province in Central Thailand in 2007. The populations of Cx. quinquefasciatus, Ae. aegypti and Ae. albopictus had been reared for 274, 232 and 215 generations, respectively. Field strains of Ae. aegypti and Ae. albopictus were obtained from eggs laid in Nonthaburi Province, and the F4 of those lines were used in this study. Adult mosquitoes were maintained ad libitum with a mixture of 5% sucrose and 5% vitamin B complex (w/v), while larvae were maintained in plastic trays and fed on minced commercial mouse food until reaching the pupal stage.

Materials and Methods
Virus strain. The virus was provided by the National Science and Technology Development Agency of Thailand. This strain was named Zika virus/H. sapiens-tc/THA/2014/SV0127-14, and it was isolated from the blood of a patient in Thailand in 2014 and used to infect the Toxorhynchites splendens mosquito (1 passage). Following isolation, the virus was passed once in Ae. albopictus C6/36 cells (1 passage). This strain is from an Asian lineage. The complete genome is accession number KU681081. Viral stocks were then produced in C6/36 cells and stored at −80 °C until further use.

Mosquito infection.
Prior to artificial oral infection, the viral titer of the virus stock was calculated by fluorescent focus assay in C6/36 cells and was determined to be 1.7 × 10 6 FFU/ml. Four-to five-day-old female Ae. aegypti, Ae. albopictus and Cx. quinquefasciatus mosquitoes were deprived of food for 24 hours prior to being provided with a blood meal. Starved females were fed with expired human blood from deidentified donors 58 , which was obtained from the National Blood Center, Thai Red Cross Society, Bangkok, Thailand, with ZIKV added at a concentration of 1.7 × 10 5 FFU/ml; the mosquitoes were fed via artificial blood feeding. The female mosquitoes were allowed to feed for 45 minutes. Non-engorged females were removed, and engorged females were reared in a cage and maintained with a 5% sucrose and 5% vitamin B complex (w/v). Three days after receiving the blood meal, water containing a black plastic bow was placed into the mosquito cage for 3 days for oviposition. Mosquitoes were collected for virus detection 7 days after blood feeding (F 0 ). Eggs were collected and allowed to hatch in a plastic tray. Larvae were reared to the pupal stage and then adulthood to obtain the subsequent progeny (F) and for the detection of ZIKV. Starting with the F 1 generation, the mosquitoes were fed blood without ZIKV under the same conditions described previously.

Viral detection in mosquitoes.
There were forty pools per replicate of female (20 adults/pool) and ten pools per replicate of male (30 adults/pool) Ae. aegypti, Ae. albopictus and Cx. quinquefasciatus mosquitoes. The experiments were performed in triplicate. Each pool was ground in 1X phosphate-buffered saline (PBS) and centrifuged at 14,000 g at 4 °C for 10 minutes. The supernatant was transferred to minimum essential media (MEM) (Gibco, US) containing 2% fetal bovine serum (FBS) (Gibco, US) and maintained at −80 °C. Viral RNA was extracted from the pellets of the pooled mosquito samples, ground in a lysis solution (provided with the kit) and centrifuged; then, the supernatant was processed for RNA extraction using the Invisorb Spin Virus RNA Mini Kit (Invitec GmbH, Germany). The viral RNA samples from the mosquitoes were amplified to test for ZIKV infection by hnRT-PCR. Primers for the NS5 gene were modified from Moureau et al. 59 . The RT-PCR amplification reaction was set up in a final volume of 25 μl using the Superscript III One-Step RT-PCR kit, and the nested PCR was performed with 2 μl from the first reaction and 1 unit of Taq DNA polymerase (Fermentas, USA). Ae. aegypti, Ae. albopictus and Cx. quinquefasciatus mosquitoes that had fed on human blood without ZIKV were used as negative controls. The PCR products were analyzed via 2% agarose gel electrophoresis, stained with ethidium bromide, and visualized with Quantity One Quantification Analysis Software version 4.5.2 (Gel DocEQ System; Bio-Rad, Hercules, CA). The positive PCR products were recovered from the gel and purified using the Agarose Gel DNA Purification Kit: Invisorb Fragment CleanUp (STRATEC Molecular GmbH, Germany) following the manufacturer's instructions. The purified DNA was sent to Macrogen, Inc. (Macrogen, South Korea) for direct DNA sequencing for confirmation of the identification of ZIKV.
ZIKV isolation and propagation. The supernatants of the samples that tested positive by hnRT-PCR were filtered through a 0.2 µm sieve and spread in a 12-well plate with a monolayer of Ae. albopictus C6/36 cells (ATCC CRL-1660) for 1 hour. After discarding the supernatant and refreshing with 2 ml of MEM (Gibco, US) containing 10% FBS (Gibco, US), 1% penicillin (100 U/ml; Sigma-Aldrich, US), and streptomycin (100 μg/ml; Sigma-Aldrich, US) (P/S), the cells were maintained at 37 °C with 5% CO 2 . The cultures were incubated for 7 days. CPEs were checked every 24 hours for 7 days after the initial 24-hour incubation period, with observations made under an inverted microscope (Olympus, Japan).
Mosquito salivary glands, yoke granules and eye samples. The salivary glands, midguts, yoke granules and eyes were collected at 7 days postinfection from females orally exposed to ZIKV. Anesthetized individual mosquitoes (30-40 mosquitoes per replicate) were dissected in a drop of 1X PBS on a glass slide under a stereomicroscope (Olympus, Japan). The salivary glands, embryos and head were transferred onto SuperFrost Plus microscope slides (Thermo Scientific, USA). ICC staining. The salivary glands, embryos and heads of infected and uninfected mosquitoes were transferred to SuperFrost Plus microscope slides (Thermo Scientific, USA), which were then air dried and fixed in 100% cold acetone before being rehydrated in graded absolute ethanol. The slides were stained with the primary rabbit-Zika virus NS1 protein antibody (GeneTex, USA) and the HRP-conjugated anti-rabbit IgG secondary antibody (Abcam, MA). The color was developed using DAB and counterstained with hematoxylin (Dako, CA), and the specimens were then examined under a light microscope (Olympus, Japan) at 100X magnification.