Mosquitoes from different species endemic to North Florida were collected over a 12-month period, and blood meal analyses were conducted on engorged females to determine the range of vertebrate hosts each species fed on. Thirty-one mosquito blood meals from 6 mosquito species were identified, and blood meal hosts identified included horse, cow, armadillo, deer, raccoon, rabbit, and owl. Several mosquitoes captured in the study should be considered potential bridge vectors for encephalitis viruses considering their ubiquity, their preference for mammalian blood meals, and their competence as viral vectors.
Translation provided by the authors.
Mosquitoes vector a variety of pathogens of medical and veterinary importance. Growing concern over the spread of West Nile virus (WNV) and related encephalitis viruses has prompted extensive investigation into the host-feeding patterns and preferences, as well as vectoral capacities, of pests now viewed mainly as cosmopolitan nuisances (Sardelis et al. 2001; Turell et al. 2005; Rodrigues & Maruniak 2006; Molaei et al. 2008). Host-feeding patterns and preferences vary according to a number of innate, seasonal, and environmental tendencies, including host availability and abundance, flight behavior and feeding periodicity of mosquitoes (Molaei et al. 2008) as well as by region (Turell et al. 2005). The current study sought to collect an array of mosquito species native to North Florida over a 12-month period and identify the range of vertebrate hosts that those mosquitoes fed on through blood meal analysis.
MATERIALS AND METHODS
Mosquito Collection and Identification
Mosquitoes were collected in CO2-baited Center for Disease Control (CDC) light traps (John W. Hock, Gainesville, FL) from May 2006 through Apr 2007 from 11 sites in Alachua County, Florida. The habitats chosen were sampled due to proximity either to aquatic environments or to habitats of suspected vertebrate hosts. Traps were placed on site in late afternoon and collected the next morning, approximately 18 h later. Captured mosquitoes were transferred to the University of Florida while still alive in mesh collection bags from the trap, placed in a -70°C freezer to euthanize them, and stored in the freezer until subsequent identification and processing. Mosquitoes were sorted by species as described in Darsie & Morris (2003) on a -20°C chill-table, and those without apparent blood meals were pooled in groups of up to 25 in sterile microcentrifuge tubes, while female mosquito abdomens with apparent blood meals were homogenized individually.
PCR Amplification and Sequencing of Mosquito Blood Meals
DNA was extracted with the DNAzol® Direct Extraction Kit (MRC, Cincinnati, OH). Polymerase chain reactions (PCR) were conducted on extracted DNA with vertebrate-specific primer sets and cycling conditions described in Cupp et al. (2004) that preferentially amplified a 290-bp region from the cytochrome b gene within the mitochondrial DNA of vertebrates. Briefly, the PCR reactions were conducted in 50 µL reactions with the following reagents: 5.0 µL of 10X reaction buffer (600 mM Tris-HCl pH 8.5, 150 mM (NH4)2SO4, 35 mM MgCl2), 0.2 mM dNTPs, 25 pmol of each primer, 2.5 units of Taq Polymerase (Invitrogen, Carlsbad, CA) and 2.5 µL of DNA template. If 290 bp amplicons were not produced or were not able to be identified, a second PCR was conducted with mammalian-specific primer sets and cycling conditions described in Ngo & Kramer (2003) to preferentially amplify a 772-bp region from the cytochrome b gene. Briefly, the PCR reactions were conducted in 25 µL reactions with the following reagents: 2.5 µL of 10X reaction buffer (200 mM Tris-HCl pH 8.4, 500 mM KCl), 3.0 mM MgCl2, 0.5 mM dNTPs, 5.0 pmol of each primer, 1.25 units of Taq Polymerase, and 2.5 µL of blood meal DNA template. Agarose gel electrophoresis was used to separate amplicons, and DNA was purified from excised bands with the expected size with QIAquick columns (QIAGEN, Valencia, CA). Fragments were sequenced with BigDye Terminator Sequencing Ready Reaction Kits (Applied Biosystems, Foster City, CA) at the Interdisciplinary Center for Biotechnology Research at the University of Florida. Sequences were analyzed and edited with Sequencher™ software (Gene Codes Co., Ann Arbor, MI), and were compared with the GenBank database entries by the BLAST program from NCBI ( http://blast.ncbi.nlm.nih.gov) to identify the vertebrate host on which each mosquito had fed.
RESULTS
Sixty-one of the 45,692 mosquitoes collected showed apparent blood meals. Upon amplification, 57 of those 61 samples (93%) showed amplicons of the expected size. Of those, vertebrate hosts were identified in 33 individuals (58%) (Table 1). Blood meals were most commonly identified from hosts fed on by Coquillettidia perturbons Walker, a particularly opportunistic feeder (Molaei et al. 2008). Coquillettidia perturbans blood meals represented 5 mammalian species (horse, armadillo, deer, rabbit, and raccoon) and 1 avian species (owl). Blood meals were identified from 6 horses and 4 cows in the 10 Mansonia titMans Walker, from 2 horses and 1 cow in 3 Aedes vexans Meigen, from 1 cow and 1 deer in 2 Culex salinarius Coquillett, and from 1 rabbit in Culex erraticus Dyer & Knab, while all 3 Anopheles crucians Wiedemann fed upon horses. The identity of the blood meal match with GenBank sequences ranged from 89–100%, and the best match was shown in Table 1.
DISCUSSION
Of the 6 species of blood-fed mosquitoes captured in the study, 4 species fed from horses, which are particularly susceptible to diseases caused by encephalitis viruses. Preferential feeding on horses by Ae. vexans is a major concern, as Ae. vexans is considered a competent to highly-efficient vector for Saint Louis encephalitis (SLE), western equine encephalomyelitis (WEE), eastern equine encephalomyelitis (EEE), and WNV (Goddard et al. 2002; Turell et al. 2005).
TABLE 1.
VERTEBRATE-DERIVED BLOODMEALS IDENTIFIED FROM MOSQUITO SPECIES CAPTURED IN ALACHUA CO. (FLORIDA) FROM 2006–2007.
Coquillettidia perturbans is considered a moderately competent vector of EEE (Vaidyanathan et al. 1997) and an inefficient but competent vector for WNV (Sardelis et al. 2001) in lab settings. EEE and WNV have been isolated from mosquitoes in nature throughout the Eastern and Southeastern United States (Srihongse et al. 1980; Crans & Schulze 1986; Edman et al. 1993; Godsey et al. 2005; Lukacik et al. 2006). Anopheles crucians and Ma. titillans showed a considerable affinity for horse hosts in a previous study by Cupp et al. (1986) as well, and have been positive for WNV isolations in the field (CDC 2007; Cupp et al. 2007). The paucity of research on An. crucians and Ma. titillans vectoral capacities for encephalitis viruses indicates that neither have historically been considered important disease vectors. However, considering their pervasiveness and their preference for mammalian blood meals in our study, as well as their moderate competence as viral vectors (Molaei et al. 2008), the potential for Cq. perturbans and Ma. titillans to be bridge vectors in the transfer of EEE and WNV from viremic primary hosts to mammals should still be considered.
CO2-baited light traps, used in a series of studies conducted on the mosquitoes captured, are not considered particularly attractive to engorged mosquitoes, thus explaining the low percentage (61 of 45,692, or > 0.002%) of blood-fed mosquitoes captured. Future studies in mosquito blood meal analysis will supplement CO2-baited light traps with mosquito resting boxes.
Of particular interest is the low yield of positive blood meal identifications in An. crucians (3 positive identifications in 12 blood meals). Townzen et al. (2008) had similar complications, reporting that co-amplification of Anopheles genes occurred often with cyt b primer sets, which consequently prevented positive host identification.
ACKNOWLEDGMENTS
We thank Dr. Alejandra Garcia-Maruniak for invaluable contributions and for help with editing the manuscript. This research was funded by USDA-CSREES, T-STAR grant # 00057913.
