Vector Competence of Aedes caspius and Ae. albopictus Mosquitoes for Zika Virus, Spain

We assessed the vector competence of Aedes caspius and Aedes albopictus mosquitoes in Spain for the transmission of Zika virus. Whereas Ae. albopictus mosquitoes were a competent vector, Ae. caspius mosquitoes were unable to transmit Zika virus. We also identified high levels of vertical transmission of Zika virus in Ae. albopictus mosquitoes.


Z ika virus is an emerging arbovirus of the family
Flaviviridae primarily transmitted by Aedes aegypti mosquitoes, but other Aedes species mosquitoes could be competent vectors (1). Ae. aegypti mosquitoes are absent from most countries in Europe (2), and the invasive Ae. albopictus mosquito and other native species could create novel epidemiologic scenarios for Zika virus. Indeed, Ae. albopictus mosquito populations from Europe are competent vectors for Zika virus (3,4). However, the vector competence for transmission of Zika virus of most mosquito species of Europe is currently unknown and may vary across virus strains and mosquito populations (5).
Although no autochthonous vectorborne Zika virus transmission has been reported in Spain, >316 imported cases of Zika virus have been confirmed (6). The confirmed cases, together with the presence of both the Ae. albopictus mosquito (7) and the native Ae. caspius mosquito (8,9) (a potential vector of chikungunya virus [10] and Rift Valley fever virus [11]), indicate a risk for Zika virus transmission in Spain. Accurately quantifying this risk requires evaluating the competence of these mosquito species for Zika virus.
We determined vector competence at different days postinfection (dpi) by exposing F0 generation of Ae. caspius mosquitoes (collected as larvae in Huelva, Spain, because we were unable to rear it under laboratory conditions) and F2 generation of Ae. albopictus mosquitoes (collected as eggs in Barcelona, Spain) to Zika virus through infectious blood meals. We used F8 generation of colonized populations of Ae. aegypti mosquitoes (collected in Poza Rica, Mexico) as a control population and Zika virus strains CAM (2010 Cambodia; GenBank accession no. JN860885) and PR (2015 Puerto Rico; GenBank accession no. KU501215), passaged 4 times on Vero cells and 2 times on C6/36 cells. We propagated on C6/36 cells for 4 days, and freshly harvested supernatant was mixed 1:1 with sheep blood (Colorado Serum Company, http://www.thepeakofquality.com) and 2.5% sucrose (5).
We offered to 4-to 7-day-old Ae. albopictus and Ae. aegypti female mosquitoes infectious blood meals containing either the CAM or PR strain at a final concentration of 7.6 log 10 PFU/mL. Infection rates were determined by screening mosquitoes' bodies, dissemination rates by screening legs, and transmission rates by screening saliva, at 3 different time points (7, 14 and 21 dpi) using Zikaspecific quantitative reverse transcription PCR including negative controls in each reaction (12) ( Table 1; Appendix, https://wwwnc.cdc.gov/EID/article/25/2/17-1123-App1. pdf). We calculated Zika titers from standard curves on the basis of infectious particle standards created from matched virus stocks (5). We further exposed 4-to 10-day-old Ae. caspius female mosquitoes to the PR strain as described. We conducted 3 independent trials using different Zika virus concentrations at different time points (7, 14, or 21 dpi) for each trial (Table 1; Appendix).
To determine the ability of Ae. albopictus mosquitoes to vertically transmit Zika virus, 4-to 7-day-old females were infected with Zika PR as described, and noninfectious blood meals were offered weekly after the first oviposition. We collected eggs laid in the second oviposition and hatched them for subsequent testing. We grouped second instar larvae in pools of 5 individuals and tested them for Zika virus (13). We estimated vertical transmission rate, measured as filial infection rate using the maximumlikelihood method (PoolInfRate version 4.0, https://www. cdc.gov/westnile/resourcepages/mosqSurvSoft.html) (13).
We performed generalized linear models with binomial error distribution and logit link function to assess the effect of mosquito species, virus strains, and dpi on the infection, dissemination, and transmission rates. We also considered the interactions between virus strain and dpi and between virus strain and mosquito species. We determined differences in mean viral titers between mosquito species, virus strains, and dpi in mosquito body, legs, and saliva using Kruskal-Wallis tests. Analyses were run in JMP version 9 (SAS Institute, http://www.jmp.com). Infection rate was higher in Ae. albopictus than in Ae. aegypti mosquitoes, and Zika PR had a higher infection rate than Zika CAM. Dissemination rate increased with time (dpi) but was similar between mosquito species and Zika strains. Transmission rate also increased with time, and mosquitoes infected with Zika CAM showed a higher transmission rate than those infected with Zika PR. Transmission rate did not differ between Ae. albopictus and Ae. aegypti mosquitoes (Table 1, 2). Mean viral titers in bodies differed between mosquito species and Zika strains, with higher titers in Ae. albopictus compared with Ae. aegypti mosquitoes (χ 2 = 5.09, df = 1; p<0.02) and higher titers for Zika PR compared with Zika CAM (χ 2 = 6.92, df = 1; p<0.009). Mean viral titers in legs were similar for both Zika strains (χ 2 = 0.95, df = 1; p = 0.33), but were higher in Ae. aegypti relative to Ae. albopictus mosquitoes (χ 2 = 9.53, df = 1; p<0.002). Mean viral titers did not differ in saliva secretions between mosquito species (χ 2 = 1.7, df = 1; p = 0.19) or Zika strains (χ 2 = 1.02, df = 1; p = 0.31). We detected Zika virus infection in Ae. caspius mosquitoes at 7, 14, and 21 dpi, but detected no virus dissemination or transmission at any point (Table 1). Five larval pools of Ae. albopictus mosquitoes (29.4%; N = 17) were positive for Zika virus, with a filial infection rate of 72.2 (95% CI 27.6-156.1) and mean viral load of 2.5 log 10 PFU/mL. This value equates to a ratio of 1:14.
Our results suggest Ae. albopictus mosquitoes in Spain are competent vectors of Zika virus at levels similar to Ae. aegypti mosquitoes. We detected Zika CAM in saliva earlier than Zika PR, which suggests that genetically variable strains may have different transmission potential (5). Although a similar transmission rate was found in Ae. albopictus mosquitoes from Spain and Italy (3), lower rates were measured in populations in France (4). In addition, Ae. albopictus mosquitoes from Spain could transmit Zika virus at 7 dpi, 4 days earlier than mosquitoes in Italy (4). These discrepancies may be explained by variation in vector competence between mosquito populations and virus strains (5). Although Zika virus can infect Ae. caspius mosquitoes, it is unable to escape the midgut and be effectively transmitted (14).
Zika virus is vertically transmitted by the population of Ae. albopictus mosquitoes in Spain at substantially higher rates than found in Ae. albopictus mosquitoes from New York and Italy (4,13) and for other flaviviruses (15). These results suggest that the ability of Zika virus to be transmitted vertically is highly population dependent and could contribute to maintenance of the virus in Ae. albopictus mosquitoes in Spain.
Our results confirm that populations of Ae. albopictus mosquitoes increase the risk for Zika virus transmission in  In the second trial, 61 Ae. caspius mosquitoes were exposed to Zika virus (blood meal titer = 7.7 Log10 PFU/mL) and 14 of them were completely engorged. These mosquitoes were used to determine competence of Zika virus at 7 dpi. In the third trial, 65 Ae. caspius mosquitoes were exposed to ZIKV (blood meal titer = 7.6 Log10 PFU/mL) and 32 of them were completely engorged. These mosquitoes were used to determine competence of Zika virus at 21 dpi.
Approximately 2,500 F2 Ae. albopictus and 2,500 F8 Ae. aegypti mosquitoes were exposed to Zika virus PR (blood titer = 7.6 log10 PFU/mL) and Zika virus CAM (blood titer = 7.6 log10 PFU/mL). Approximately 150 mosquitoes of each species were engorged with infected blood. Mosquitoes were housed in independent containers and competence for Zika virus was determined at 7, 14, and 21 dpi.

RT-qPCR Protocol for Virus Detection
We