Introduction of Aedes albopictus in Gabon: what consequences for dengue and chikungunya transmission?


Corresponding Author Anna-Bella Failloux, Génétique moléculaire des Bunyavirus, Institut Pasteur, 25–28 rue du Dr Roux, 75724 Paris cedex 15, France. E-mail:


The 2007 outbreak of chikungunya in Gabon has indicated the potential of this disease to spread beyond its usual range ensuing from the expansion of the mosquito Aedes albopictus. A few cases of dengue (DEN) infection were also detected. Because little is known about the potential for Gabonese mosquito species to transmit both chikungunya and DEN viruses (DENV), we conducted studies to determine the susceptibility of Ae. albopictus and Aedes aegypti collected in Libreville to both viruses by experimental infections. Disseminated infection rates were high for Ae. albopictus infected with chikungunya virus (CHIKV) (66.7–86%) and low with DENV (13–21.4%). Moreover, Ae. aegypti sp. formosus was a less efficient vector of CHIKV than Ae. albopictus. The recent introduction and dissemination of chikungunya associated with the invasion of Ae. albopictus in Africa illustrates the potential for CHIKV to spread to other parts of the world.


L’épidémie de chikungunya en 2007 au Gabon a révélé le possibilité de cette maladie de se propager au-delà de son secteur habituel résultant de l’expansion du moustique Aedes albopictus. Un petit nombre de cas de dengue a également été détecté. Vu qu’il y a peu connaissance sur la capacité des d’espèces de moustiques gabonais de transmettre les virus du chikungunya et la dengue, nous avons mené des études visant à déterminer la sensibilité d’Ae. albopictus et Aedes aegypti recueillis à Libreville, aux deux virus par des infections expérimentales. Les taux d’infection disséminée étaient élevés pour Ae. albopictus infectées par le virus du chikungunya (66,7% - 86%) et faible par le virus de la dengue (13% - 21,4%). En outre, Ae. aegypti sp. formosus était un vecteur moins efficace du virus de chikungunya que Ae. albopictus. La récente introduction et la dissémination de chikungunya associéà l’invasion de l’Ae. albopictus en Afrique illustre la possibilité du virus du chikungunya de se propager vers d’autres parties du monde.


El brote de chikungunya en el 2007 en Gabón ha hecho patente el potencial que tiene esta enfermedad de propagarse más allá del área usual gracias a la expansión del mosquito Aedes albopictus. También se detectaron unos pocos casos de infección por dengue. Puesto que se conoce poco sobre el potencial que tienen las especies de mosquito Gabonesas para transmitir tanto el virus chikungunya como el del dengue, hemos conducido estudios con el fin de determinar, mediante infección experimental, la susceptibilidad que Ae. albopictus y Aedes aegypti, recolectados en Libreville, tienen a ambos virus. Las tasas de la infección diseminada eran altas para Ae. albopictus infectado con el virus chikungunya virus (66.7% - 86%) y baja con el virus del dengue (13% - 21.4%). Más aún, Ae. aegypti sp. formosus era un vector menos eficiente del virus chikungunya que Ae. albopictus. La reciente introducción y diseminación de chikungunya asociada con la invasión de Ae. albopictus en África, ilustra el potencial del virus chikungunya para diseminarse a otros lugares del mundo.


First isolated in Tanzania in 1952 (Ross 1956), chikungunya virus (CHIKV), an arthropod-borne virus, is believed to be maintained in Africa in a sylvatic cycle involving non-human primates and forest-dwelling Aedes spp. mosquitoes (Jupp & Kemp 1996; Diallo et al. 1999). Outbreaks are heavily dependant upon the sylvatic mosquito densities, members of the Aedes furcifertaylori group (Jupp & McIntosh 1990), increasing during periods of heavy rainfall. In Asia, the urban and anthropophilic mosquito, Aedes aegypti is the most significant vector. This species maintains close associations with humans and thus causes large outbreaks. Another common peridomestic species, the Asian tiger mosquito Aedes albopictus, is abundant in CHIK endemic areas. This species has a broad distribution and can spread CHIKV into many new areas. A CHIK outbreak was recorded in Kenya in 2004 (Chretien et al. 2007), then in the Indian Ocean (Schuffenecker et al. 2006) and later, in India (Arankalle et al. 2007), Europe (Rezza et al. 2007) and other African countries (Peyrefitte et al. 2008). In Kenya and India, the main vector was probably Ae. aegypti. In contrast, in La Reunion Island and Europe, Ae. albopictus was the major vector. Sequence analysis of the virus genome revealed that these recent outbreaks were caused by a new variant characterized by a mutation in the E1 envelop glycoprotein gene (A226V) (Schuffenecker et al. 2006). This mutation has favoured better transmissibility of the virus by Ae. albopictus (Vazeille et al. 2007). Globalization of commercial traffic is probably the main cause of Ae. albopictus expansion from its native region in Southeast Asia. Eggs of Ae. albopictus tolerate long periods of desiccation, allowing long transports between continents in used tire shipments. The species has succeeded in colonizing both tropical and temperate regions throughout the world in Europe, Oceania, the Americas and in some countries of sub-Saharan Africa.

In 2007, a serological survey of patients presenting febrile symptoms, inhabitants of Libreville and of the French military camp ‘De Gaulle’, revealed the presence of CHIKV. Evidence of primary dengue (DEN) infection was also found in some patients. The CHIKV strain exhibited a high degree of identity with strains isolated in Cameroon in 2006 (Peyrefitte et al. 2007). Isolates from Gabon and Cameroon are close to Democratic Republic of Congo strains but different from the Central African Republic isolates (Powers et al. 2000), suggesting the maintenance in the western part of Central Africa of a different CHIKV variant. Both isolates from Gabon and Cameroon harbour the A226V mutation in the E1 glycoprotein. This emergence coincides with the establishment of Ae. albopictus in Gabon (Coffinet et al. 2007) where Ae. aegypti was the only Aedes species present before 2006. This latter species is considered as the main vector of DEN viruses (DENV) in large urban centres in the tropics, whereas Ae. albopictus is known as a secondary vector. Thus, both Ae. aegypti and Ae. albopictus being now sympatric in Gabon, it is interest to evaluate their susceptibility to CHIKV and DENV by measuring disseminated infection rates in the laboratory.

Materials and methods


Mosquito samples ‘Rivière’, ‘Ga’ and ‘Garage’ were collected in 2007 as eggs in ovitraps placed in and near the De Gaulle Military Camp in Libreville (Gabon, Central Africa). For the ‘Rivière’ sample, ovitraps were placed away from habitats in a wetland; both Ae. albopictus and Ae. aegypti sp. formosus were found. For the ‘garage’ sample, ovitraps were placed in a shed where tires and other artificial containers suitable as larval breeding sites were present; only Ae. albopictus was collected. For the ‘Ga’ sample, ovitraps were placed in a shrubbery surrounding cottages; only Ae. albopictus was found.

As a control of susceptibility, we used two strains: (i) the Paea strain of Ae. aegypti provided by Institut Louis Malardé (Tahiti, French Polynesia) and established in Paris since 1994 and (ii) the STDEN-F4 strain of Ae. albopictus which derived from a field-collected population STDEN collected in La Reunion Island in March 2006.

Eggs were hatched and raised to the adult stage (F0 generation) in an insectarium. The adults were sorted by species. They were given a 10% sucrose solution and females were allowed to feed on restrained mice every 2 days. Eggs corresponding to the F1 generation were collected within a month, hatched and larvae reared to adult stage in pans with tap water and yeast tablets. One-week-old F1 females were tested for their susceptibility to CHIK and DENVs.


The CHIKV strain 06.21, provided by the French National Reference Centre for Arbovirus of the Institut Pasteur, was isolated on Ae. albopictus cells C6/36 (Igarashi 1978) from a serum collected in November 2005 from a newborn male from La Reunion Island. CHIKV 06.21 was the major genotype isolated from patients during the 2005–2006 outbreak. It presented an A226V mutation in the gene E1 (Schuffenecker et al. 2006), as did the strains isolated in Gabon in 2007 (Peyrefitte et al. 2008). Viral production and titration are described in Vazeille et al. (2007). CHIKV 06.21 was isolated and amplified on C6/36 cells, the stock used in this study being a third passage.

The DENV was a serotype 2 strain (DENV2) provided by Rosen. It was isolated from a human serum collected in Bangkok (Thailand) in 1974. This virus had been passed only in different mosquito species (Toxorhynchites amboinensis, Ae. albopictus and Ae. aegypti) by intrathoracic inoculation (Rosen & Gubler 1974). Viral production and titration are described by Vazeille-Falcoz et al. (1999a).

Oral infection of mosquitoes

Infections were all performed in one assay with 1-week-old females which were allowed to feed for 15 min through a chicken skin membrane covering the base of a glass feeder supporting the blood–virus mixture maintained at 37 °C. The infectious meal was composed of a virus suspension diluted (1:3) in washed rabbit erythrocytes isolated from arterial blood collected 24 h before the infectious meal (Vazeille-Falcoz et al. 1999a). Adenosine tri phosphate, a phagostimulant, was added at a final concentration of 5 × 10−3 m. Blood meal titres were 107 pfu (plaque-forming unit)/ml for CHIKV and 108 MID50 (mosquito infectious dose)/ml for DENV; titres of meals remained stable during the whole feeding procedure. Titres were chosen according to disseminated infection rates obtained in Ae. aegypti, Paea strain (see Table 1), i.e. a titre of 108 MID50/ml for DENV2 and of 107 pfu/ml for CHIKV 06.21. For CHIKV, 107 pfu/ml corresponds to 108 MID50/ml. Fully engorged females were transferred to small cardboard containers and maintained with 10% sucrose at 28 ± 1 °C for 14 days. To evaluate disseminated infection rates and thus vector competence, surviving females were frozen at −80 °C and tested for the presence of CHIK or DEN antigens in head squashes by indirect immunofluorescent assays. The disseminated infection rate corresponds to the percentage of females whose head were tested positive for viral antigens among surviving females, 14 days after ingestion of the infectious blood-meal. Dissemination occurs when virions disseminate via the midgut epithelium into the haemocele, which spreads the infection to other tissues and organs including salivary glands. Disseminated infection rates were compared using the Fisher’s exact test.

Table 1.   Disseminated infection rates towards CHIKV and DENV of Aedes albopictus and Aedes aegypti samples collected in Gabon
SpeciesSampleDisseminated infection rate (n)
  1. n, number of tested females; –, not determined.

Aedes albopictusGA66.7 (33)
Garage77.8 (9)13.0 (23)
Riviere86.0 (43)21.4 (14)
STDEN96.9 (32)88.9 (18)
Aedes aegyptiRiviere55.0 (40)
Paea100 (76)92.0 (112)


Eggs of Ae. aegypti were only recorded in one ovitrap at Riviere when Ae. albopictus eggs were present in all three locations. When infected with CHIKV 06.21, the three Ae. albopictus samples, GA, Garage and Riviere, collected in and near De Gaulle Camp exhibited high but not significantly different disseminated infection rates ranging from 66.7% to 86.0% (Fisher’s exact test: P = 0.11). Among Ae. albopictus samples infected by CHIKV, Rivière was the only one exhibiting a non-significative difference with the Ae. albopictus STDEN control (P > 0.05).

When comparing disseminated infection rates of the three Ae. albopictus samples (Ga, Garage and Rivière) with the Ae. aegypti sp. formosus sample Rivière (55%), a significant difference was only obtained between the two species collected in the same site, Rivière (P = 0.003). When infected with DENV2, the two Ae. albopictus samples tested, Garage and Rivière, showed similar low disseminated infection rates, respectively 13.0% and 21.4% (P = 0.65), significantly lower to the two control strains, Paea and STDEN (P < 0.05). Controls showed the highest disseminated infection rates whatever the virus used for infection.


Aedes albopictus and Ae. aegypti from Gabon exhibited different susceptibilities to CHIKV, Ae. albopictus showing a higher susceptibility than Ae. aegypti sp. formosus from the same site. Disseminated infection rates of Ae. albopictus collected in Libreville during the CHIK outbreak in 2007 were roughly similar to rates obtained with Ae. albopictus from La Reunion Island in 2006–2007 when infected with the same viral strain CHIKV 06.21 (Vazeille et al. 2007). Thus, Ae. albopictus assumes to be a better vector to CHIKV than Ae. aegypti.

Moreover, as demonstrated by the disseminated infection rates obtained, Gabonese Ae. albopictus have a low susceptibility to DENV2, as in rural areas of DEN-endemic countries in Southeast Asia and Pacific islands (Vazeille et al. 2003). Even if this species has been able to sustain DEN outbreaks in the absence of Ae. aegypti, as in China, the Seychelles, Japan, Hawaii and more recently, in La Réunion Island in 2004, it has never been associated with the severe syndrome, DEN haemorrhagic fever/DEN shock syndrome (Metselaar et al. 1980; Gratz 2004). Therefore, despite its strong invasive capacity, Ae. albopictus should not play a significant role in DEN transmission in Gabon. Although we did not test Gabonese Ae. aegypti sp. formosus samples for their susceptibility to DENV2, our previous studies have demonstrated that their susceptibility was also very low compared with that of the domestic and anthropophilic Ae. aegypti sp. aegypti (Vazeille-Falcoz et al. 1999b; Failloux et al. 2002).

As Ae. albopictus continues to spread, challenging Ae. aegypti as a human pest in some countries in Africa, it is important to go into details about its vectorial role. Thus, in Gabon, Ae. albopictus appears not to be an efficient DEN vector, whereas it seems to be a good vector of CHIK virus as demonstrated recently in La Reunion Island and in Italy.


We thank Félix Rey, director of the virology department at the Institut Pasteur, for his constant support. This project was supported by the Institut Pasteur and the office of the surgeon general.