Risk of dengue in Central Africa: Vector competence studies with Aedes aegypti and Aedes albopictus (Diptera: Culicidae) populations and dengue 2 virus

Introduction Dengue is the most important mosquito-borne diseases worldwide but was considered scarce in West-Central Africa. During the last decade, dengue outbreaks have increasingly been reported in urban foci in this region suggesting major epidemiological changes. However, in Central Africa where both vectors, Aedes aegypti and Aedes albopictus are well established, the role of each species in dengue transmission remains poorly investigated. Methodology/Principal findings Field-collected strains of Ae. aegypti and Ae. albopictus from different ecological settings in Central Africa were experimentally challenged with dengue 2 virus (DENV-2). Mosquitoes were analysed at 14- and 21-days post-infection. Analysis provide evidence that both Ae. aegypti and Ae. albopictus in Central Africa were able to transmit dengue virus with Ae. aegypti exhibiting a higher transmission rate. Unexpectedly, two Ae. aegypti populations from Bénoué and Maroua, in northern Cameroon, were not able to transmit DENV-2. Conclusions/Significance We conclude that both Ae. aegypti and Ae. albopictus are susceptible to DENV-2 and may intervene as active dengue vectors. These findings highlight the urgent need to plan a vector surveillance program and control methods against dengue vectors in Central Africa in order to prevent future outbreaks.


Introduction
Dengue is the most important mosquito-borne diseases worldwide but was considered scarce in West-Central Africa. During the last decade, dengue outbreaks have increasingly been reported in urban foci in this region suggesting major epidemiological changes. However, in Central Africa where both vectors, Aedes aegypti and Aedes albopictus are well established, the role of each species in dengue transmission remains poorly investigated.

Methodology/Principal findings
Field-collected strains of Ae. aegypti and Ae. albopictus from different ecological settings in Central Africa were experimentally challenged with dengue 2 virus (DENV-2). Mosquitoes were analysed at 14-and 21-days post-infection. Analysis provide evidence that both Ae. aegypti and Ae. albopictus in Central Africa were able to transmit dengue virus with Ae. aegypti exhibiting a higher transmission rate. Unexpectedly, two Ae. aegypti populations from Bé noué and Maroua, in northern Cameroon, were not able to transmit DENV-2.

Conclusions/Significance
We conclude that both Ae. aegypti and Ae. albopictus are susceptible to DENV-2 and may intervene as active dengue vectors. These findings highlight the urgent need to plan a vector surveillance program and control methods against dengue vectors in Central Africa in order to prevent future outbreaks. PLOS

Introduction
Dengue is one of the most important arboviral diseases in the world with nearly 390 million annual dengue infections and 96 million (67-136 million) clinical cases [1]. Dengue is caused by a dengue virus (DENV) belonging to the genus Flavivirus (family Flaviviridae). There are four distinct, but closely related serotypes of dengue (DENV-1, DENV-2, DENV-3 and DENV-4). DENV is transmitted to humans through the bite of infected Aedes mosquitoes primarily Aedes aegypti Linneaus 1772 and Aedes albopictus (Skuse 1894). In Africa, the situation of dengue was less critical as human cases were mainly associated with mild symptoms [2,3]. Haemorrhagic syndromes were only reported in East Africa [4,5]. However, dengue outbreaks have been reported recently in some West-Central African countries [6][7][8][9][10] suggesting a switch in the epidemiological dynamics of dengue. The two invasive species, Ae. aegypti and Ae. albopictus are well established in Africa. While Ae. aegypti native from Africa took 400-500 years to invade the tropical belt [11,12], Ae. albopictus originated from Asian forests has colonized all five continents in less than four decades [13,14]. Aedes albopictus has been first reported in Central Africa in early 2000s in Cameroon [15], and since then, this species has invaded almost all countries of the region including the Republic of Congo [16][17][18]. In sympatric areas, Ae. albopictus outcompetes with the native species Ae. aegypti [18][19][20][21]. Coincidentally, the emergence of arboviral diseases such as dengue and chikungunya in Central Africa has coincided with the establishment of Ae. albopictus in this region. Indeed, Ae. albopictus was identified as the main vector during concurrent dengue/chikungunya outbreak in Gabon in 2007 [8,22], and in Cameroon in 2006 [23]. During the last two decades, DENV-1 and DENV-2 mainly, were circulating in Cameroon [24][25][26][27][28][29]. Nationwide surveillance of dengue in 2006/2007 only revealed that seroprevalence (IgG and IgM antibodies) was higher in Douala [29]. In the neighbouring country, the Republic of the Congo, only little information is known about dengue circulation. The vector competence (which refers to the potential of an arthropod to ingest the pathogen, ensure replication, dissemination and transmission) which is one of the main factors required to establish the epidemiological role of mosquitoes in transmission is poorly studied in Central Africa. Previous studies only focused on infection and dissemination rates [8,30,31] and not transmission potential (i.e. virus detection in mosquito saliva). To fill this important gap, we performed a comparative analysis aiming to assess the ability of Ae. aegypti and Ae. albopictus collected in different ecological settings in Central Africa to transmit DENV-2.

Ethics statement
This study was approved by the Cameroonian national ethics committee for human health research N˚2017/05/911/CE/CNERSH/SP. Oral consent to inspect the potential breeding sites was obtained in the field in household or garage owners. The Institut Pasteur animal facility has received accreditation from the French Ministry of Agriculture to perform experiments on live animals in compliance with the French and European regulations on care and protection of laboratory animals (EC Directive 2010/63, French Law 2013-118, February 6th, 2013). All experiments were approved by the Ethics Committee and registered under the reference APA-FIS6573-201606l412077987 v2.

Mosquito sampling
Larvae and pupae were collected from August 2017 to April 2018 in several locations in Central Africa including Brazzaville (Republic of the Congo), Yaoundé, Douala, Tibati and Bénoué National Park (Cameroon, Fig 1). Each of these locations have been previously characterised [18,19]. In each location, mosquitoes were collected in peri-urban (i.e. peripheral area of the city) and downtown (i.e. city centre with high building density) from a minimum of 20 containers per environment. Immature stages of Aedes were transported in the insectary and pooled together according to the city. Larvae were raised until adults and identified morphologically. Adults from same location and species were reared at 28˚±1˚C under 12h dark:12h light cycle and 80% relative humidity. Eggs obtained (Table 1) were transported to the Institut Pasteur Paris, reared to adult stage and used to challenge with DENV-2.

Virus strain
The dengue 2 virus (DENV-2) strain provided by Leon Rosen (Institut Pasteur, Paris, France) was isolated in 1974 from a human sera from Bangkok, Thailand [32]. This virus had been passed only in different mosquito species (Toxorhynchites amboinensis, Ae. albopictus, and Ae. aegypti) by intrathoracic inoculation. Viral stocks were produced by inoculating Ae. albopictus cells (C6/36 clone) with triturated infected mosquitoes.

Challenging mosquitoes with DENV-2
For each sample, six batches of 60 7-10 day old females were challenged with an infectious blood meal (Institut Pasteur, Paris, France) containing 1.4 mL of washed rabbit erythrocytes and 700 μL of viral suspension. The blood meal was supplemented with adenosine 5'-triphosphate (ATP) as a phagostimulant at a final concentration of 1 mM and provided to mosquitoes at a viral titre of 10 7 focus-forming unit (ffu)/mL using a Hemotek membrane feeding system (Hemotek Ltd, Blackburn, United Kingdom). Mosquitoes were allowed to feed for 20 min through a piece of pork intestine (Institut Pasteur, Paris, France) covering the base of a Hemotek feeder maintained at 37˚C. Fully engorged females were transferred in cardboard containers and maintained with 10% sucrose under controlled conditions (28±1˚C, relative humidity of 80%, light: dark cycle of 12 h: 12 h) for up to 21 days with mosquito analysis 14 and 21 days post-infection (dpi). 21-32 mosquitoes were examined at each dpi.

Infection, dissemination and transmission assays
For each mosquito examined, body (abdomen and thorax) and head were tested respectively for infection and dissemination rates at 14 and 21 dpi. For this, each part was ground individually in 300 μL of L15 medium (Invitrogen, CA, USA) supplemented with 2% fetal serum bovine (FBS), and centrifuged at 10,000×g for 5 min at +4˚C. The supernatant was processed for viral titration. Saliva was collected from individual mosquitoes using technique of forced salivation as described previously [33]. Briefly, mosquitoes were cool anesthetized, wings and legs of each mosquito were removed and the proboscis inserted into a tip of 20 μL containing 5 μL of FBS. After 30 min, FBS containing saliva was mixed in 45 μL of L15 medium for titration. Infection rate (IR) refers to the proportion of mosquitoes with infected body (i.e. abdomen and thorax) among tested mosquitoes. Disseminated infection rate (DIR) corresponds to the proportion of mosquitoes with infected head among the previously detected infected mosquitoes (i.e. virus positive abdomen/thorax). Transmission rate (TR) represents the proportion of mosquitoes with infectious saliva among mosquitoes with disseminated infection. Vector competence can be summarized by the transmission efficiency (TE) which was calculated as the proportion of mosquitoes with infectious saliva among total of mosquitoes tested [34].

Viral titration by focus forming assay
Samples were titrated by focus fluorescent assay on Ae. albopictus C6/36 cells [35]. Body, head and saliva suspensions were serially diluted in L15 medium supplemented with 2% of FBS and inoculated onto cells in 96-well plates. After incubation of 5 days at 28˚C, samples were fixed with 0.1mL/well of formaldehyde 3.6% in phosphate buffer saline (PBS) during 20 min at room temperature. Then, plates were stained using antibodies specific to DENV as the primary antibody, and conjugated goat anti-mouse immunoglobulin G (Alexa Fluor 488) as the second antibody (Life Technologies, CA, USA). Titres were expressed as ffu/mL.

Statistical analysis
All statistical analyses were performed with R software v 3.5.2 (R Core Team, Vienna, Austria). Qualitative variables were expressed as proportion and compared using Fisher's exact test (RVAideMemoire package). While quantitative variables were described as mean and compared using non-parametric test of Kruskal-Wallis because of non-normal distribution. Pairwise comparison were performed using Fisher's exact test for proportions and Kruskal-wallis test for means. P-value <0.05 was considered as statistically different.

Infection and disseminated infection rates in Ae. albopictus and Ae. aegypti
To determine if Ae. aegypti (six populations) or Ae. albopictus (four populations) were more likely to sustain DENV outbreak in Central Africa, the ability of the virus to replicate and disseminate in both species was examined at 14 and 21 dpi as well as DENV particles excreted in saliva (only at 21 dpi) (Figs 2 and 3). At 14 dpi, Ae. albopictus infection rate (IR) ranged from 33.3% in Douala population to 68.4% in Yaoundé urban population but no statistical difference was detected (Fig 2A, Fisher'

Discussion
During the past decade, there has been a rise of dengue cases in urban foci in Central Africa notably in Cameroon [26][27][28][29]. Even suspected, vectors were not well identified and characterised. In this study, we assessed for the first time, the ability of Ae. aegypti and Ae. albopictus collected in different ecological settings (Fig 1) in Central Africa to transmit DENV-2, a serotype repeatedly reported in the region [9,29]. We demonstrated that DENV-2 was able to replicate, disseminate and be secreted in saliva of both Ae. aegypti and Ae. albopictus populations from Central Africa, thus enable to transmit DENV. However, infection rates were significantly higher for Ae. aegypti than for Ae. albopictus at 14 and 21 dpi. Disseminated infection rates followed the same trend at 14 dpi. Nevertheless, DENV was detected in saliva of all Ae. albopictus populations tested while for Ae. aegypti, virus was not detected in both populations (2/6) from northern Cameroon, Bénoué and Maroua. These results suggest that vector competence of Ae. aegypti to DENV-2 in Central Africa vary significantly according to geographical population as previously suggested elsewhere [36,37]. This may due to the fact that populations from Bénoué and Maroua exhibited an extrinsic incubation period longer than 21 days; to note, the extrinsic incubation period refers to the duration between the ingestion of an infectious blood meal and the excretion of virus in saliva when the mosquito bites [38]. It depends on the three-way combination of mosquito, virus and environment described under genotypeby-genotype-by-environment (GxGxE) interactions [39]. In addition, low vector competence in these populations would be due to presence of specific refractory genes [40,41]. Indeed, refractoriness of mosquito to dengue virus may be caused by different parameters like microbiome composition as bacterial symbionts of mosquitoes have been shown to alter the vector competence to arboviruses [42] and immune system of mosquito since it was demonstrated that anti-viral immunity in mosquito vectors is critical to prevent virus replication and transmission [43]. Further investigations in this regard are needed to elucidate.
Moreover, the seroprevalence of dengue examined in 2006/2007 in three main cities of Cameroon located in different ecological settings revealed that anti-DENV IgG and IgM antibodies varied significantly with a higher prevalence reported in Douala [29], location where the highest transmission rate and viral load were also detected in Ae. aegypti in this study. Beside the mosquito genetic background, mosquito microbiome can modulate arbovirus transmission [42,44,45]. The transmission rate was higher for Ae. aegypti compared to Ae. albopictus in locations where both species are sympatric. This result is in agreement with the fact that Ae. aegypti is considered as a major dengue vector, and Ae. albopictus, the secondary one [46]. Meanwhile, it would be interesting to highlight that Ae. albopictus can become a major dengue vector in the absence of Ae. aegypti as reported previously in China, the Seychelles, Japan, Hawaii and on La Réunion [47] or when Ae. albopictus becomes the most prevalent species as reported in Gabon [8]. Nevertheless, infection and disseminated infection rates assessed for Ae. albopictus in this study are similar to those reported in previous studies in Africa [8,31] and in Southeast Asia [48]. For Ae. aegypti, infection and disseminated infection rates are higher compared to that previously reported in Cameroon (17.2% to 59.7%) but similar to that often reported outside Africa [37,48]. Albeit Ae. aegypti is more competent than Ae. albopictus to transmit DENV, some parameters can influence DENV transmission in nature, such as vector densities, host preference, virus evolution and proportion of immunologically naive people [49]. Additional studies using a local strain of DENV circulating in Central Africa are needed to validate these results. Regarding vector densities, recent studies in Cameroon and Republic of Congo revealed that Ae. albopictus tends to replace Ae. aegypti in most areas where both species are sympatric [18,19]. It was also demonstrated that in Yaoundé (Cameroon) Ae. albopictus preferentially fed on humans rather than on available domestic animals [50] Data generated in our study demonstrated that both Ae. aegypti and Ae. albopictus can sustain dengue transmission in Central Africa. This could increase the risk of dengue outbreak in the region and urge the need of a vector surveillance program to prevent and preparedness for an intervention in case of outbreaks.