Aedes aegypti mosquitoes from Guadeloupe (French West Indies) are able to transmit yellow fever virus

The recent yellow fever epidemic in Brazil has raised the concern of outbreaks in neighboring countries, particularly in the Caribbean region where the vector Aedes aegypti is predominant. This threat comes from the past when in the Americas, this disease caused devastating urban epidemics. We report the vector competence of Ae. aegypti from Guadeloupe for yellow fever virus by determining different parameters describing virus infection, dissemination, and transmission. The results indicate that Ae. aegypti Guadeloupe are susceptible to yellow fever virus with viral particles detected in mosquito saliva at 14 and 21 days post-infection. Local authorities and more broadly, international organizations should maintain the active surveillance of Aedes mosquitoes and the spreading of human cases from South America.


Introduction
Yellow fever (YF) is a mosquito-borne viral disease endemic to some countries of South America and sub-Saharan Africa. It can present various clinical features ranging from a self-limited, mild febrile illness to fatal symptoms such as hemorrhages and liver damages. Most of all cases reported annually (80-90%) occur in Africa where YF covers 44 countries [1]. In South America, YF is described in less than 10 countries: Argentina, Bolivia, Brazil, Colombia, Ecuador, Paraguay, Peru, and Venezuela (http://ais.paho.org/phip/viz/ ed_yellowfever.asp). In these locations, YF uses to periodically spread via epizootic outbreaks following the displacements of non-human primates [2]. From July 2017 through March 2018, the states of Rio de Janeiro, Minas Gerais, and São Paulo in Brazil, counted 932 human cases including 300 fatal cases [3]. Alarmingly, human cases were reported near São Paulo city, threatening the initiation of an urban transmission that has not been notified in the country since 1942.
Originally from Africa, YF is believed to be introduced into America via the slave trade in the middle of 18 th century [4]. Approximately 10.7 million slaves were deported to the Caribbean, North and South Americas during four centuries [5]. Likewise Ae. aegypti by finding suitable breeding sites in slave transport ships, was introduced in America at the same period [6]. Deadly YF epidemics devastated the continent for centuries. An eradication campaign targeting Ae. aegypti organized by the Pan American Health Organisation (PAHO) was initiated in 1946 and led to the elimination of the vector from most American countries, and consequently, the disappearance of urban YF [7]. Unfortunately, the eradication campaign was interrupted and most countries were reinfested by the vector [8].
Yellow fever virus (YFV; Flavivirus, Flaviviridae) is primarily transmitted by the mosquitoes Aedes spp. (e.g. Aedes africanus) in Africa and Haemagogus (e.g. Haemagogus janthinomys) in South America [9]. In Brazil, the anthropophilic mosquitoes Ae. aegypti and Aedes albopictus as well as the YFV-enzootic mosquitoes Haemagogus leucocelaenus and Sabethes albiprivus are highly susceptible to YFV [10]. Thus, the widely distributed Ae. aegypti in American countries raises the concern of a re-urbanization of YF if the virus is introduced via viremic vertebrate hosts. In the Caribbean, Guadeloupe Island has experienced several outbreaks caused by arboviruses such as dengue [11], chikungunya [12] and zika [13], all three viruses only be transmitted by Ae. aegypti as Ae. albopictus is absent from the island [14]. To be considered as an epidemic vector of YFV, Ae. aegypti should be experimentally susceptible to the virus (i.e. a competent vector) in addition of being an anthropophilic mosquito [15] in close contacts with humans [16]. In this report, we evaluate the vector competence of Ae. aegypti from Guadeloupe to YFV. These results will help the local health authorities and the decision-makers to anticipate the arrival of YF in the Caribbean.

Virus strain
The strain IEC-4408 (YFV-4408; accession number: KY861728) belonging to the 1E lineage of YFV, was isolated from a Howler monkey in 2008 [10]. The strain was passaged four times on Ae. albopictus C6/36 cells. Viral stocks for mosquito infections were produced on C6/36 cells and stored at -80˚C.

Artificial blood feeding
Six boxes of 60 7-day-old F1 female adults (F0 collected as larvae in June 2017 in Deshaies, Basse Terre, Guadeloupe) were fed on an infectious blood meal containing 1.4 mL of washed rabbit red blood cells and 0.7 mL of virus suspension. The blood meal supplemented with ATP as a phagostimulant at a final concentration of 1 mM was provided to mosquitoes at a titer of 10 6.5 focus-forming unit (ffu)/mL using a Hemotek membrane feeding system. Engorged mosquitoes were transferred into boxes and maintained with 10% sucrose at 28˚C under a photoperiod of 12:12.

Mosquito sampling and processing
Mosquitoes were examined at 7, 14, and 21 days post-infection (dpi). After removing mosquito wings and legs, the proboscis was inserted into a P20 tip filled with 5 μL of fetal bovine serum (FBS) [17]. After 30 min, saliva was expelled from the tip to 45 μL of L-15 medium (Invitrogen, California, USA) and then processed for viral titration to estimate transmission. Then, mosquito head and body were collected and ground individually in 300 μL of L-15 medium supplemented with 2% FBS, for respectively, viral infection and dissemination analysis. 200 μL of homogenates were collected after centrifugation at 10,000 g for 5 min at +4˚C before viral titration. To estimate the vector competence, three parameters were calculated: (i) infection rate (IR) referring to the proportion of mosquitoes with infected body among engorged mosquitoes, (ii) dissemination rate (DR) corresponding to the proportion of mosquitoes with infected head among mosquitoes with infected body, and (iii) transmission rate (TR) representing the proportion of mosquitoes with infectious saliva among mosquitoes with infected head.

Virus titration
Mosquito samples were titrated by focus fluorescent assay on Ae. albopictus C6/36 cells in 96-well plates [18]. After 5 days of incubation at 28˚C, plates were stained using antibodies specific to YFV as the primary antibody and conjugated Alexa Fluor 488 goat anti-mouse IgG as the second antibody (Life Technologies, California, USA).

Statistical analysis
Statistical analyses were performed with Stata software (StataCorp LP, Texas, and USA). P-values<0.05 were considered significant.

Discussion
Except Trinidad and Tobago in 1979 [19], the Caribbean has not suffered from YF since 1960. The ongoing YFV circulation in Brazil raises concern regarding viral importation into the Caribbean. However, the requirement of YF vaccination is not mandatory in many Caribbean islands (e.g. Haiti, Cuba), but restricted to the travelers coming from YF-epidemic countries (http://www.who.int/ith/ITH_country_list.pdf?ua=1). The disease control relies only on the check of vaccination card, which is insufficient for YF prevention. Thus, YF is still a threat for this region where Ae. aegypti is widely distributed. Here, although not based on vertebrate animal transmission model, we demonstrated that Ae. aegypti from Guadeloupe are susceptible to YFV and able to transmit viral particles from 14 days post-infection. Viral infection, dissemination and transmission increased along with dpi. Infection and dissemination reached a peak at 14 dpi (70% and 57% for IR and DR, respectively), suggesting that the midgut has a limited role as barrier to the viral dissemination in the mosquito general cavity. Transmission was only detected from 14 dpi suggesting an extrinsic incubation period (i.e. period between the ingestion of infectious blood meal and the excretion of virus in saliva) is between 7 and 14 days as shown for Ae. aegypti populations from Congo and Brazil [10]. However transmission was quite low (i.e. 8%) suggesting a significant role of salivary glands to retain viral particles. At 21 days post-infection, transmission was more efficient with 12% of mosquitoes presenting disseminated infection and delivering virus in saliva.
It is now widely admitted that vector competence depends on the virus genotype, the mosquito genotype and their interactions, promoting local adaptation of viral lineages to mosquito vector populations [20]. The table below (Table 1) presents the vector competence of several Ae. aegypti populations from Africa, America, Asia and South Pacific region to several lineages/genotypes of YFV, exemplifying the specific outcome to each combination virus-vector.
The pattern of Ae. aegypti Guadeloupe infected with a YFV belonging to the 1E lineage (IEC-4408; [10]) should be close to the profile of mosquitoes from the American continent. Ae. aegypti from Rio and Manaus presented similar IR, DR and TR when compared to Ae. aegypti Guadeloupe (see Table 1, [10]). It has been demonstrated previously that Ae. aegypti from the Caribbean were genetically close to mosquitoes from Brazil [27]. However, other factors should be considered to assess the risk of transmission; while YF is still absent from Asia, the vector competence of Asian Ae. aegypti (see Table 1; Phnom Penh: DR = 64%, Ho Chi Minh city: DR = 48%) was higher than values of Ae. aegypti from Africa. These factors include: vector densities, trophic preference of vectors for humans, proportion of immunologically naïve humans, and environmental conditions favorable to transmission. Surveillance of travelers coming from YFV-endemic regions of Africa or South America able to initiate a local transmission in the Caribbean should be reinforced. Likewise, vaccination coverage should be reexamined as the live-attenuated 17D is one of the most effective vaccines available against this arbovirus.  [24] (Continued)