Zika virus can be venereally transmitted between Aedes aegypti mosquitoes

Background Alternative transmission routes have been described for Zika virus (ZIKV). Here, we assessed for the first time the venereal transmission of ZIKV between Aedes aegypti under laboratory conditions. Results Orally-infected mosquito females were able to transmit the virus to males venereally, and males inoculated intrathoracically were capable of infecting females during mating. The genome of venereally-transmitted virus recovered from males was identical to that of ZIKV ingested by mated females. Conclusion We conclude that venereal transmission between Aedes mosquitoes might contribute to Zika virus maintenance in nature.


Methods
We experimentally assessed venereal transmission in two populations of Brazilian Ae. aegypti mosquitoes: Urca (URC; F2 generation), Rio de Janeiro, coastal Southeast region, and Goiânia (GOI; F1 generation), Goiás, inland Central-West region. Their F0 generation tested negative for ZIKV by qRT-PCR as described previously [9]. Mosquitoes were reared at 26 ± 1°C, 12 h:12 h light:dark cycle and 70 ± 10% humidity (standardized environmental condition, SEC) as previously described [9,10]. Mosquitoes were sexed at the pupal stage, and the emerged adults of each gender were maintained in distinct cages supplied with a 10% sucrose solution at SEC. Virgin Ae. aegypti females aged 5-7 days post-emergence were orally challenged with two ZIKV strains belonging to the Asian lineage, namely ZIKV-Rio-U1 (GenBank: KU926309) and ZIKV-PE243 (Gen-Bank: KX197192), respectively, isolated from humans in Southeast [11] and Northeast [12] Brazil. These ZIKV strains display 99% nucleotide and amino acid identity and may similarly infect Brazilian Ae. aegypti [9][10][11][12]. Accordingly, females were provided with a mixture containing two parts of washed rabbit erythrocytes and one part of viral suspension at a final titter of 10 6 PFU/ml as described previously [9]. We incubated sets of fully engorged females at SEC in cylindrical carton cages (20 × 16 cm) daily supplied with a 10% sucrose solution. Groups of 10-30 females corresponding to each combination of the mosquito population and virus strain were randomly killed, and body homogenates in culture medium were examined by RT-qPCR for determining infection rates (IR) on the 10th and 14th day postinfection (dpi) as described elsewhere [9]. At this same dpi, virgin males of the same population, generation and age were introduced into the respective female-infected cages at two sex ratios (Table 1). Then, distinct groups of mosquitoes were allowed to freely mate for 30 h devoid of any food source, to avoid oral contamination, or for 5 days provided with a piece of cotton mesh imbibed with 10% sucrose solution on the top of the cage. In the last case, samples of the sucrose solution probed by mosquitoes were examined daily for the presence of ZIKV by RT-qPCR, and have always tested ZIKVnegative. On completion of the above-mentioned contact periods, samples of 30 females were dissected to determine insemination rates, that is, the proportion of females with spermatheca containing spermatozoids [13]. RNA was extracted from male entire body homogenates for determining IR by RT-qPCR and whole viral genome sequencing on the 14th day after mating as described [9,11]. Due to the scanty number of health or surviving males that mated with infected females, we decide to intrathoracically inject virgin males with ZIKV to access venereal transmission from male to female. Thus, the supernatant of the body homogenate of one venereally contaminated URC male was inoculated in other URC virgin males. After 10 and 14 days of incubation at SEC, URC virgin females were left to mate for 30 h (devoid of any food source) at a sex ratio of 1 male:2 females, and incubated for 14 days at SEC. Male and female entire body homogenates were tested by RT-qPCR [9]. Aiming to investigate possible genetic changes in the viral genome occurring during the venereal transmission, we also extracted RNA and sequenced the whole genome from the virus of the infectious meal taken by females as well as from a pool of five infected females that mated with the venereally contaminated males. Nucleotide sequences were aligned, edited and compared as described previously [11].

Results
Insemination rates were 50-90% and 100% in females kept in contact with males for 30 h and 5 days, respectively. IRs in orally challenged females ranged from 46% to 70% in URC mosquitoes challenged with ZIKV-PE243 and ZIKV-Rio-U1, respectively (Table 1). Venereal transmission to males was achieved in mosquito groups whose females were infected with ZIKV-Rio-U1, where IRs in males was 0.9% in URC and 33% in GOI (median of viral loads in bodies of 1.5 × 10 6 RNA copies/ml; CT ≤ 17.2). No venereal transmission was detected in the mosquito groups infected with ZIKV-PE243, whose females coincidently had the lowest insemination rates. IRs of intrathoracically inoculated males were 87.5 and 97.9% at 10 and 14 days, respectively ( Table 2). Viral load in bodies of males at 10 and 14 days after injection reached 3.6 × 10 6 and 9.8 × 10 6 RNA copies/ml (median: 3.3 × 10 6 and 3.6 × 10 6 ; CT ≤ 15.6), respectively. Males with 10 days after inoculation could not venereally transmit the virus. However, the venereal transmission was detected in 50% of females mated by males with 14 days after intrathoracic injection (viral load in bodies of positive females ranging from 3.4 × 10 2 to 9.8 × 10 2 RNA copies/ml; CT ≤ 28.8). The compared ZIKV genomes were identical, with the virus recovered from a venereally contaminated male (URC-ZIKVRioU1) presenting a single nucleotide substitution at position 1115 (protein E) in which a thymine was replaced by a cytosine, without generating amino acid change.

Discussion
Male Aedes mosquitoes have been found naturally infected with ZIKV in Africa [14] and South America [4], providing evidence that inter-mosquito contamination occurs in nature and this phenomenon may play some role in virus maintenance and viral evolution [15]. Being non-hematophagous, male mosquitoes may acquire an  [18][19][20]. We demonstrated that ZIKV could be venereally transmitted between Ae. aegypti of two geographically-distant populations.
Males obtained the infection from females and vice versa. Moreover, the genome of the virus transmitted by the venereal route was identical to those orally absolved as well as infecting females, suggesting that no expressive genetic changes occur in the viral genome during the venereal transmission. The efficiency of the venereal transmission route may be greater than expected. The variation in insemination and infection rates in orally-challenged females suggest that not all of them had sexual contact with males, especially in the groups brought into contact for only 30 h. The higher rate of venereal transmission from infected females to males (33%) was detected in the group of mosquitoes in which contact time was 5 days and insemination rate was of 100%, that is, all females had sexual contact with males. Considering that one male could have inseminated more than one female in such a group [13], it is possible that not all males had sexual contact with infected females, reducing the chance of venereal transmission. These data altogether may explain the large variation in male IRs. Several factors have been proposed to explain the silent endemic/enzootic maintenance of ZIKV in Africa and Asia as well as the recent fast spread of this virus in the West Hemisphere, but they remain poorly understood. Although the main mode of maintenance of ZIKV in nature is supposed to be the primate-mosquitoprimate transmission cycle, evidence of inter-vertebrate (congenital, perinatal, sexual, occupational and others) [1,[5][6][7] and inter-invertebrate contamination routes (vertical and venereal) [8] may help in explaining the high capacity of ZIKV to be transmitted and spread.

Conclusion
We conclude that together with other modes of transmission, the venereal route in Aedes mosquitoes might contribute to ZIKV virus maintenance in nature. and National Institut of Health (grant 1UO1 AI115595-01). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Availability of data and materials
The data used in the present study are available from the corresponding author upon reasonable request.
Authors' contributions SSC and RSF carried out mosquito rearing and experimental infections. SSC and ELT performed mosquito intrathoracic inoculation. SSC, RSF, RMM, MGC and AFB tested mosquito samples by molecular methods. AACS and MCB produced the viral stocks and sequenced and analyzed the whole virus genomes. MCB and ABF participated in producing reagents for mosquito experiments and helped to prepare the text. RLO and SSC conceived the study and analyzed data. RLO wrote the manuscript. All authors read and approved the final manuscript.

Ethics approval
This study was approved by the Institutional Ethics Committee on Animal Use (CEUA-IOC license LW-34/14) at the Instituto Oswaldo Cruz.

Consent for publication
Not applicable.

Competing interests
The authors declare that they have no competing interests.

Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.  Abbreviations: n number of tested mosquitoes, IR Infection rate referring to the proportion of mosquitoes with infected body among tested ones by RT-qPCR