+ Author affiliations
Citation style for this article: Di Luca M, Severini F, Toma L, Boccolini D, Romi R, Remoli ME, Sabbatucci M, Rizzo C, Venturi G, Rezza G, Fortuna C. Experimental studies of susceptibility of Italian Aedes albopictus to Zika virus. Euro Surveill. 2016;21(18):pii=30223. DOI: http://dx.doi.org/10.2807/1560-7917.ES.2016.21.18.30223
Received:22 April 2016; Accepted:04 May 2016
Zika virus (ZIKV) is an emerging mosquito-borne virus (Flaviviridae family) isolated from different Aedes species in the past. In the recent outbreaks that occurred in Latin America, Aedes aegypti is believed to be the main vector. The isolation of ZIKV from this mosquito species in Malaysia , and early experimental studies [2,3] appear to confirm this hypothesis. Recent vector competence studies have also shown that the American Ae. albopictus exhibits similar transmission potential as the American Ae. aegypti .
Ae. albopictus is widespread in Mediterranean countries, in particular in Italy where it caused an outbreak of Chikungunya virus (CHIKV) (Togaviridae family, Alphavirus genus) in 2007 . To assess the risk of ZIKV transmission, we evaluated the vector competence of an Italian Ae. albopictus population for the virus. Potential vertical (transovarial) transmission of ZIKV was also evaluated.
Experimental infection by membrane feeding technique
Oral infection was performed in a BSL-3 laboratory using a ZIKV strain of the Asian genotype (kindly provided by Dr Isabelle Leparc-Goffart of the French National Reference Centre for Arboviruses in Marseille) isolated from a patient returning from French Polynesia in 2013 . Ten-day-old mosquito females from an Italian Ae. albopictus population (collected in Scalea town, Calabria region, in the late summer of 2015) and from a long-established colony of Ae. aegypti (collected in Reynosa, Mexico, in 1998) were allowed to feed for 1 hour through a membrane feeding apparatus. The virus was diluted in rabbit blood (final virus concentration: 6.46 log10 plaque-forming units (PFU)/mL) and maintained at 37 °C by a warm water circulation system. After the blood meal, fully engorged females were transferred to cages and maintained on a 10% sucrose solution in a climatic chamber (26 ± 1 °C; 70% relative humidity; 14 h:10 h light/dark cycle) for 21 days. Ten mosquitoes from either species were individually processed at 0, 3, 4, 7, 11, 14, 18 and 21 days post infection (dpi). To evaluate viral infection, dissemination and transmission, body (head, thorax and abdomen), legs plus wings, and saliva were analysed, as previously described . ZIKV titre was evaluated by quantitative reverse transcription PCR (qRT-PCR). Specific primers ZIKV 1086 and ZIKV 1162c were used, with 5-FAM as the reporter dye for the probe (ZIKV 1107-FAM) . Crossing point values were compared with a standard curve obtained from 10-fold serial dilutions of virus stock of known concentration [8-10].
Mosquito bodies were analysed in order to evaluate the infection rate (IR), calculated as the number of ZIKV-positive bodies with respect to the total number of fed females. Legs plus wings were tested to assess the dissemination rate (DR), calculated as the number of the specimens with ZIKV-positive legs plus wings among the number of specimens with ZIKV-positive bodies. The saliva of the potentially infected females was processed to assess the transmission rate (TR), defined as the number of mosquitoes with ZIKV-positive saliva among the number of specimens with ZIKV-positive bodies . The potential vector competence was expressed as population transmission rate (PTR), calculated as the number of specimens with ZIKV-positive saliva with respect to the total number of fed mosquitoes [9,11].
Vector competence analysis
Mean viral titres and IR, DR, and TR values are shown in Figures 1 and 2.
Mean Zika virus titres in different body parts of Aedes albopictus and Ae. aegypti colonies, at different times post infection, Italy, 2016
PFU: plaque-forming units.
Infection, dissemination and transmission rates in Aedes albopictus and Ae. aegypti colonies, at different times post infection with Zika virus, Italy, 2016
Infection rate: number of positive bodies/number of tested fed females; dissemination rate: number of positive legs plus wings/number of positive bodies; transmission rate: number of positive saliva/number of positive bodies.
All of the Ae. aegypti and Ae. albopictus bodies analysed immediately after the infectious blood meal (day 0) showed positive results, with mean viral titres of 3.85 ± 0.44 log10 PFU/mL and 3.57 ± 0.28 log10 PFU/mL, respectively, confirming the ingestion of infectious viral particles. The viral titres detected in the bodies increased gradually in both mosquito colonies, reaching 5.18 ± 0.16 log10 PFU/mL in Ae. aegypti and 4.88 ± 0.21 log10 PFU/mL in Ae. albopictus at 18 and 14 dpi (Figure 1A). As expected, differences in IR values between the two species were observed (Figure 2A). In particular, Ae. albopictus showed lower IR values than Ae. aegypti at all collection times. Whereas an IR of 40% was already detected at 3 dpi for Ae. aegypti, infected Ae. albopictus specimens were observed starting from 7 dpi, with an IR value of 20%. Cumulative IR values were 43% for Ae. aegypti and 10% for Ae. albopictus (Table).
Cumulative infection, dissemination, transmission and population transmission rates of Aedes albopictus and Ae. aegypti experimentally infected with Zika virus, Italy, 2016
|Population transmission rate
Disseminated infection was observed in Ae. aegypti starting from 3 dpi, with a mean viral titre of 2.74 ± 0.06 log10 PFU/mL (DR 50%), while in Ae. albopictus, the presence of the virus in legs and wings was detected from 11 dpi, with a lower viral titre (1.62 log10 PFU/mL) and an equal value of DR (50%) (Figures 1B and 2B). Starting from 4 dpi, the saliva of Ae. aegypti showed ZIKV particles (titre of 1.99 log10 PFU/mL and TR of 17%) and remained positive throughout all collection times. In particular, the viral titres increased reaching the highest levels after 11 dpi (2.64 ± 0.50 log10 PFU/mL). In contrast, virus was detected in the saliva of Ae. albopictus at 11 and 14 dpi, with TR values of 50% at both collection points, showing a longer extrinsic incubation period (EIP). Cumulative DR and TR values were 73% and 60% for Ae. aegypti and 29% and 29% for Ae. albopictus. Finally, PTR values were 26% for Ae. aegypti and 3% for Ae. albopictus (Table).
After the infectious blood meal, 40 to 50 engorged mosquitoes from each species were kept separate in different cages, under the same laboratory conditions as the ones analysed above, and were allowed to lay eggs (first gonotrophic cycle). Two weeks after the infectious blood meal, a second uninfected blood meal was provided to obtain a second gonotrophic cycle. Pools of 15 to 30 specimens (males and females) from the first and second gonotrophic cycles of both species were processed by qRT-PCR and were negative for ZIKV.
Little is known on ZIKV despite its significant epidemic potential . The introduction and dissemination of this previously neglected flavivirus in Latin America, raised concern in temperate climate countries with established Ae. albopictus populations . In light of the spread of this mosquito species in Italy, proven vector in the 2007 outbreak of CHIK , it is particularly important to evaluate its vector competence for ZIKV and to assess the potential risk transmission in Italy as well as in other Mediterranean countries. Our study shows that the Italian Ae. albopictus population is susceptible to ZIKV, allowing viral replication and dissemination also in the salivary glands. The short persistence of the virus in the mosquito’s saliva, the PTR value of 3% and the long EIP indicate a low transmission efficiency compared with that of Ae. aegypti. In addition, it should be noted that despite the use of a long-established mosquito colony, not representative of a wild population, the vector competence of Ae. aegypti for ZIKV was significant in our experiment. A recent modelling study  that was based on parameters of susceptibility to infection of the Ae. albopictus derived from mosquito populations from the United States and Singapore [4,14], also estimated a low risk of sustained autochthonous transmission of ZIKV in northern Italy. Our results are similar to the above results on American Ae. albopictus and substantially confirm the low epidemic potential of ZIKV in Italy. However, the epidemic potential and the capacity to cause long chains of transmission depends on a series of factors such as the abundance of the mosquito population, the density of the human population, feeding host preferences, biting rates and environmental conditions. High mosquito density, day-biting activity, opportunistic feeding behaviour and climatic and environmental adaptability can affect the efficiency of Ae. albopictus as a vector, favouring its primary role in epidemics, also in the presence of a limited vector competence .
Our results also have important public health implications for preparedness. In fact, the extended EIP, which is consistent with the results of studies using American Ae. albopictus mosquitoes , would allow the implementation of mosquito control measures that are likely to be more efficient than those implemented in areas infested by the tropical mosquito Ae. aegypti. Moreover, our analysis of offspring of both species from the first and second gonotrophic cycle showed no evidence of transovarial transmission of ZIKV; this finding adds knowledge on the bionomics of this vector and may aid the optimisation of vector control management. Finally, ZIKV appears to be less well adapted to the Italian Ae. albopictus than the A226V variant of CHIKV (data not shown), which caused more than 250 cases in Italy after a single introduction from Kerala, India [5,16].
In conclusion, this experimentally infected Italian Ae. albopictus population appeared to be a competent vector for ZIKV, albeit less efficient than the primary vector Ae. aegypti. However, we should not forget the risk posed by CHIKV and dengue virus that remains high in southern European countries, where small outbreaks and clusters of autochthonous cases have been already documented [17,18].