Venezuelan equine encephalitis complex viruses (VEEV) are mosquito-borne alphaviruses (Togaviridae family). Epidemiologically, VEEV complex is classified in enzootic (subtypes ID, IE, IF, II-VI) and epizootic (subtypes IAB and IC) strains. Viruses of the first group are avirulent for equines and generally produce only low-titre viraemia and little or no illness. However, some of these viruses can be pathogenic for humans (Weaver et al. 2004). Epizootic strains periodically produce outbreaks involving horses and humans, with mortality rates of 19–83% for equines and human neurological disease appearing in 4–14% of cases (Weaver et al. 2004). Epizootic VEE is considered an important emerging disease threat as well as a potential biological weapon (Hawley & Eitzen 2001).
In Argentina, circulation of some enzootic VEE complex viruses in temperate regions of the northern part of the country is well recognized. Rio Negro Virus (RNV, VEEV subtype VI) and Pixuna Virus (PIXV, VEEV subtype IV) have been detected in mosquitoes from Chaco province (Mitchell et al. 1985; Pisano et al. 2010), and RNV has been isolated from rodents of Formosa province (Contigiani et al. 1999) (Figure 1). Serological studies indicate the presence of human antibodies against more than one enzootic subtype (I and VI) in subtropical regions of Argentina (Cámara et al. 2003). In 1989, an outbreak of acute febrile illness caused symptoms compatible with Dengue virus to residents of General Belgrano Island (Formosa province). Serological tests were negative for Dengue virus, but positive for RNV, pointing it as the causative agent of that outbreak (Contigiani et al. 1993).
There are no previous reports of the detection of VEE complex viruses in other regions of the country, including Tucumán province. For this reason, we conducted surveillance studies to obtain data about other potential sites for movement of VEE members in this region.
Mosquitoes were collected in March, April, November and December of 2005 and in February 2006, using CDC light traps supplemented with CO2, in the city of San Miguel de Tucumán (26°48′83″ S, 65°13′32″ W) (urban area). Collected mosquitoes were transported alive under cold conditions to the laboratory, pooled by species, place of collection and date, in lots of 1–58 unengorged individuals, and stored at −70° C until processed. Each pool for viral testing was triturated in a cold mortar with sterile minimum essential medium supplemented with 10% foetal bovine serum and 1% gentamicin and clarified by centrifugation at 11 400 g for 30 min. The supernatant was poured into individual screw-cap vials and frozen at −70° C, until viral RNA detection by RT-nested PCR assay. For RNA extraction, Trizol reagent (Invitrogen BRL, Life Technologies, Rockville, MD, USA) was used according to the manufacturer’s instructions. RT-nested PCR for alphaviruses was performed, amplifying a 195-bp fragment of the nsP4 gene (Sánchez-Seco et al. 2001). Positive samples were purified using QIAquick Gel Extraction Kit (Quiagen, Valencia, CA, USA) and submitted to direct nucleotide sequencing reaction in both directions. Obtained sequences were submitted to GenBank, with accession numbers FJ002857, GQ885141 and GU002046. Analysis with blast 2.2.19 basic local alignment search tool (http://blast.ncbi.nlm.nih.gov) (Zhang et al. 2000) was carried out to find the identity of amplified sequences. Phylogenetic analyses were performed on nucleotide sequences using Mega software version 4 (Tamura et al. 2007). A phylogenetic tree was constructed with neighbour-joining method and p-distance parameter. Minimum infection rates (MIR) were calculated as follows: MIR = number of virus isolations by species/total number of that species tested from that site ×1000.
A total of 4700 mosquitoes belonging to 16 species and grouped in 139 pools was analysed by RT-nested PCR for alphavirus detection. Four pools tested positive: 2 from Aedes (Ae.) (Oc.) scapullaris, with MIR of 2.2; 1 from Culex mollis(Cx.), with MIR of 18.5; and 1 from Ae. aegypti, with MIR of 5.7. MIRs obtained showed more viral activity in Cx. mollis than in other species. Three pools were sequenced: 2 (ArgSMT121, collected on November 2005, and ArgSMT71, collected on March 2005; Ae. (Oc.) scapullaris and Cx. mollis, respectively), grouped with RNV, strain AG80-663 (subtype VI), and 1 (ArgSMT90, collected on December 2005, Ae. aegypti) with PIXV (subtype IV) (Figure 2). The remaining sample (ArgSMT73, collected on April 2005, Ae. (Oc.) scapullaris) was not sequenced because it did not produce a high-quality amplicon.
Previous Argentine studies carried out in Chaco province have postulated Culex (Mel.) delpontei as the main vector of RNV (Mitchell et al. 1985). Recent investigations have demonstrated the presence of this virus in Culex coronator, Culex maxi, Psorophora cingulata and Psorophora spp. in the same region (Pisano et al. 2010). Culex (Mel.) delpontei has not been detected in San Miguel de Tucumán yet (Darsie & Mitchell 1985). In this work, we did not collect mosquitoes of Cx. (Mel.) subgenus. This finding indicates that other mosquito genus or species could be involved in the enzootic cycle of RNV.
Detection of PIXV confirms its circulation in the country. This virus was isolated for the first time in Belem, Brazil, in 1961 (Shope et al. 1964) and is still poorly characterized.
Our results showed for the first time the presence of viruses from the VEEV complex in mosquitoes from Tucumán province, indicating a wider distribution in the North temperate region of Argentina. Although it is not possible to make a large and detailed phylogenetic analysis because the sequenced fragment is very small, we observed that some strains detected in mosquitoes from Chaco (previously sequenced) and Tucumán provinces and in rodents from Formosa province (previously sequenced) grouped together (Figure 2), despite the geographic and ecological range of dispersion between those provinces (San Miguel de Tucumán is located 630 km west of Resistencia city, Chaco province; and 720 km southwest of Formosa city, Formosa province) (Figure 1). This suggests a probable common origin. On the other hand, the PIXV clade showed two groups, where Argentinean strains grouped together separately from Brazilian strain (BeAr35645) (Figure 2). More detailed phylogenetic studies of larger and more variable portions of the viral genome (E1, E2 genes) are needed to obtain a better understanding of the genetic diversity and dispersion routes of VEEV complex in Argentina. Therefore, attempts are been made to isolate these viruses.
Emergence of epizootic strains via mutation of enzootic strains has been observed among other members of the VEE complex (Powers et al. 1997). While this has not been described for RNV or PIXV, we do not exclude this possibility, which is why we consider the circulation of 2 enzootic subtypes (IV and VI) at the same place and time very important. In summary, our data show the importance of developing or intensifying surveillance measures in the region to understand the importance of these viruses as potential human and animal pathogens, as well as for vector control.