Abstract
The presence of viruses in arthropods in Spain has been studied over 5 years. Flaviviruses similar to cell-fusing agent, sequences of a flavivirus related to those transmitted by mosquitoes, and a phlebovirus similar to Naples and Toscana viruses were detected. Their potential human or animal pathogenicity should be studied.
With the aim of obtaining more data about arboviruses a multidisciplinary study was conducted in Spain and arboviruses were searched for within their vectors. We used a multiplex and generic RT-nested-PCR that is the result of the combination of three simple generic reactions for the detection of alphaviruses, phleboviruses, and flaviviruses (Sánchez-Seco et al. 2001, 2003, 2005). The method has been assayed with Chikungunya, Ross River, Venezuelan Equine Encephalitis (alphaviruses), TOSV (phleboviruses), West Nile virus, Dengue 1–4, Saint Louis, and Tick-borne encephalitis (flaviviruses) viruses. The limit of detection was determined using plasmids containing DNA target sequences for a member of each of the amplified genera and was about 100 copies for phlebovirus detection and 10 for alphavirus and also for flavivirus amplification.
Mosquitoes and sandflies from Catalonia, Northeastern Spain, from 2001 to 2004, and Andalusia, Southwestern Spain, from 2002 to 2005, were captured (Aranda et al. 2008). After classification they were pooled in a lysis buffer, and nucleic acids were extracted and analyzed. At the end of 2005, minor modifications, mainly using EMEM (Eagle's Minimum Essential Medium; Sigma, St. Louis, MO), in collecting samples in Southwestern Spain were carried out to try viral isolation in Vero and C6–36 cells. A total of 4300 pools were analyzed and 107 tested positive for different arboviruses. Of them, 102 (94 pools of mosquitoes and 8 of sandflies) rendered a band corresponding to that observed in flavivirus amplification. Positive amplification for phleboviruses was obtained in five pools of sandflies from Catalonia.
The topology of the tree obtained with the sequences shows that these viruses are distributed among four main groups (mosquito-borne and tick-borne viruses, viruses with unknown vector, and mosquito viruses) (Fig. 1A), as previously known, and analysis of almost the complete NS5 gene is in agreement with this result (Dr. Ana Vázquez, personal communication). The sequences of two pools (HU566/03 and HU205/05), from Andalusia, are of special interest because they seem to group with the arthropod-borne flaviviruses (Fig. 1E).
Identification of sequences. Bootstrap consensus trees showing the relationships between the sequences of nucleotides detected in this paper and those available from databases were constructed using the p-distance option with the Neighbour Joining method in the MEGA (Molecular Evolutionary Genetic Analysis) 4 software. Sequences from databases are shown with the accession number following their names. Those obtained in this study are named with an internal code corresponding to the Wetland (HU: Andalucia, GI; B and T: Catalonia), the year of capture and the species (Ae, Aedes; Oc, Ochlerotatos; C, Culiseta; Cx, Culex). (
The other positive pools are related to mosquito flaviviruses (cell-fusing agent [CFA], Kamiti River virus [KRV], and Culex Flavivirus) (Fig. 1B–E). Within them, a group more related to CFA and KRV, mainly present in Aedes vexans, has been detected (Fig. 1B). The most abundant and heterogeneous group, with a wide geographic and temporal distribution, is mainly formed by Culex mosquitoes and is related to Culex Flavivirus (Fig. 1E). Ochlerotatos caspius–positive pools are present in every group although several pools of Oc. caspius form a specific cluster (Fig. 1D). The last group is formed by some species of mosquitoes and also sandflies (Fig. 1C). Flaviviruses in Aedes mosquitoes have been described in Africa (CFA and KRV), South America (Culebra strain, a variant of CFA), and Europe (this article) (Sang et al. 2003, Cook et al. 2006). Recently, Hoshino et al. (2007) have described a variant of CFA obtained in Japan from Culex mosquitoes, and now we have another variant in Europe (this article). This close association between flaviviruses and their vectors requires deeper studies to establish potential biological advantages for the viruses.
Along with the growing number of new mosquito flaviviruses, the presence of CFA-related sequences integrated in the genome of Aedes mosquitoes (CSA; Cell Silent Agent) has recently been described (Crochu et al. 2004). To assure that positive pools were the result of RNA amplification, we used RNAsa A (Sigma) in some extracts that were directly amplified without the previous retro-transcription step. The results for flaviviruses show that some sequences correspond to RNA and other to DNA. In the Culex group, six of seven assayed pools were RNA, in the Ochlerotatus three of nine were DNA, in Phlebotomus five of five were RNA, and in the Aedes group five out of five were DNA. We also assayed the pools related to mosquito-borne viruses, and the amplification was due to an RNA target. So, most of the sequences we obtained do not correspond to DNA amplification although we amplified DNA sequences when Aedes mosquitoes (where the integration phenomenon seems to be more frequent) were the vectors. In Ochlerotatus (previously, part of the Aedes genus) DNA, as well as RNA, amplification has also been detected. Taken together, it seems that flaviviruses, not CSA-like sequences, are present in most of our mosquitoes. However, isolation in cultured cells is needed to confirm viral presence; therefore, with this aim three positive pools from the most abundant Culex group were used for viral isolation. Three to five blond passes were carried out. Cytopathic effect was observed at 5–6 dpi (days post-infection) in every pass in C6–36 but not in Vero cells. To assure that the effect was due to infection, the supernatants were assayed by RT-nested-PCR, and positive amplification from 1 dpi was obtained in all the passes in C6–36 and not in Vero cells; so, viruses from pools of Culex mosquitoes have been isolated.
The sequences of the detected phleboviruses were also different from those available in databases with the exception of a phlebovirus, named Massilia virus, recently sequenced by Charrel and coworkers (2008), which has been detected in sandflies captured in France (Fig. 1F) and whose implications for public health are unknown. They belong to one cluster that is related to the viruses belonging to the Naples complex (TOSV and Naples serotypes). It seems that the new virus has a wide distribution, because it has been detected in Catalonia and France and it has also been detected and isolated in the Southeastern region of Spain (Dr. Sara Sambonmatsu, personal communication).
Further studies are being carried out to obtain more data and determine if the new viruses infect animals or humans, and if so whether they represent a potential threat to human or veterinary health, as well as their role in the evolution of these genera.
Footnotes
Acknowledgments
The authors thank Domingo Estévez and Teresa de Miguel for their technical assistance. Dr. Inmaculada Casas and all the members of the Red EVITAR (Enfermedades Víricas Transmitidas por Artrópodos y Roedores), a multidisciplinary group founded by the Fondo de Investigaciones Sanitarias (FIS) (Spanish Ministry of Health) (G03/059), provided invaluable advice and help. All the authors were members of this network. This project has been partially funded by the projects of the FIS numbered 98/0229 and G03/059 and by grants from the Spanish Ministry of Defence.