Mosquitoes Associated with

1
Mosquitoes Associated with Equine West Nile
Virus Cases in Southeast Georgia Calvin W.
Hancock William S. Irby Dept. of Biology,
Georgia Southern University, Statesboro, GA
INTRODUCTION West Nile Virus is a
mosquito-borne flavivirus first encountered in
Africa in 1937. Since its introduction into
Queens, New York in 1999, the virus has spread
exponentially across the United States and viral
activity (infected mosquito pools, birds,
equines, or humans) has been reported in 48
states. It appeared in Georgia during the summer
of 2001 and since then, there have been more than
50 equine infections in the southeastern part of
the state. In 2003, we began a two year study
monitoring mosquito populations associated with
West Nile Virus (WNV) equine infection case
sites. Mosquitoes were obtained through use of
light traps as well as aspiration. Samples are
tested for the presence of the virus and sources
of mosquito blood meals will be identified in
hopes of establishing the significance of each
vector species to the enzootic and epizootic
transmission cycles. The main objectives of the
research are summarized as follows to determine
which mosquito species are associated with equine
WNV infections in Southeastern Georgia, determine
which mosquito species are positive for WNV, and
ultimately to determine which species may serve
as bridge vectors between avian reservoirs and
mammal (equine) hosts.
PCR-Heteroduplex Assays PCR products of
cytochrome B amplification are made from each
test animal and mixed with either northern
cardinal or Carolina chickadee driver
cytochrome B PCR products. The mixture is then
heated to 99oC to denature DNA, followed by slow
cooling to promote heteroduplex formation.
Homologous and heterologous duplexes form, where
the number of duplexes 2n, where n number of
different cytochrome B genes in mix. The
resulting products are distinguished by relative
mobilities by Polyacrylamide Gel Electrophoresis
(PAGE) and any products not matching standards
are sequenced. See figure 2 for example. These
tests will be concluded in the fall of 2005.
Table 2 Mosquito pools tested for WNV, EEEV and
SLEV from equine WNV sites, 2003-4
Mosquito Species 2003 2003 2004 2004
Mosquito Species Pools Total Specimens Pools Total Specimens
Ae. albopictus 7 14 18 78
Ae. vexans 15 32 72 2184
Cx. melanura 1 1
Cq. perturbans 6 7 12 359
Cx. erraticus 24 211 52 1535
Cx. quinquefasciatus 1 6
Cx. nigripalpus 27 503 65 2288
Cx. restuans 5 7 1 6
Cx. tarsalis 3 3
Cx. territans 4 7
Cx. salinarius 20 255 12 53
Oc. atlanticus 5 102 28 740
Oc. canadensis 5 23
Oc. fulvus pallens 3 8
Oc. triseriatus 1 1 6 12
Ps. ciliata 8 32
Ps. columbiae 4 10 15 254
Ps. ferox 1 1 10 81
Ur. sapphirina 3 7
Totals 124 1160 310 7660
MATERIALS AND METHODS Collection of
mosquitoes Adult mosquitoes were collected,
starting in August, 2003 as follows By
regular sampling using modified-CDC light traps
mosquitoes were collected from 22 sites, in a 5
county area, where avian or equine cases of West
Nile Virus had occurred in 2002 or 2003 (above
right). If notification of a case was received
within 3 weeks, trapping was initiated as soon
as possible. For some cases, trapping was
conducted at positive sites one year later, in
order to capture mosquitoes that were
representative of populations that would have
been present during the time of infection. For
avian cases, modified CDC light traps (right)
baited with dry ice were placed at the site were
the infected bird was located. For equine
cases, light traps were similarly placed.
Trapping in 2003 was conducted from 29-VIII-03 to
16-X-03, and in 2004 from 3-IX-04 to 23-X-04.
Trapping seasons were chosen based on
encompassing all onset dates. By
regular sampling using modified-CDC backpack
aspirator resting mosquitoes were collected by
aspiration of light trapping sites. Vertical
surfaces of structures which may serve as
potential mosquito resting sites (barns, feed
sheds, etc.) were sampled by aspiration using a
battery-powered modified CDC backpack aspirator
(right). Collection by aspirator was conducted
from 30-VIII-03 to 24-X-04. Processing of
mosquitoes Testing for the presence of West
Nile Virus upon collection, mosquitoes were
placed on ice until they could be returned to the
laboratory and frozen. A chill plate was used to
keep the mosquitoes frozen while they were
separated by gonotrophic level and keyed to
species. Unfed females were pooled by species,
site, date, and collection method in pools of 50
individuals. These pools are analyzed using the
antigen assay test, VecTest. Specimens of
Anopheles spp. and males were not tested due to
their low likelihood of involvement in the
transmission cycle. Blood meal analyses
blood meals are first identified to vertebrate
class (mammal, avian, reptile/amphibian) using
screening antisera produced in New Zealand White
Rabbits through the use of Enzyme-Linked
Immunosorbent Assays (ELISAs). This antisera is
produced by intranodal and interdermal injection
of a mix of sera from representatives of major
groups within each class and Freuds adjuvant to
induce an immune response. Blood meals
determined to be of mammalian or
reptile/amphibian sources will be analyzed by use
of ELISAs and species-specific antisera produced
in the same manner as the screening antisera
described above. Blood meals determined to be of
avian source will be analyzed using PCR-based
Heteroduplex Assays (HDAs) described by Lee and
others (2002) and further outlined in the
following section.
RESULTS DISCUSSION To date, we have
collected over 10, 500 mosquitoes throughout the
14 month project. Representative data (Table 1)
shows that Culex nigripalpus, Aedes vexans, and
Culex erraticus were the most abundant species
encountered in our area. However, the majority
of the Aedes vexans collected were from light
traps during the 2004 season only, and those were
limited to a few single traps. Culex
quinquefasciatus, the mosquito species most
commonly associated with West Nile Virus by
surveillance data throughout the United States,
was rarely encountered and the majority of the
specimens were collected in two samples. Culex
nigripalpus and Culex erraticus were collected
more evenly and in high abundance, indicating
that these species may have a more significant
role in transmission than previously assumed. To
date, we have tested 7,660 specimens from 2003
and 2004 (Table 2) for the presence of WNV, but
all results have been negative. Temporal data
for each collection method are shown in Figures
1, 2, and 3. Searching for a commonality between
populations encountered at different sites,
diversity was analyzed with the Shannon-Weaver
Index (Table 3). However, it indicated that
diversity was uncommon among trapping sites. We
hope that complete testing of samples will reveal
which species are most associated with arbovirus
transmission in our area. In light of the lack
of abundance of Culex quinquefasciatus, we
suspect that Culex nigripalpus or other species
may have a more significant role in local virus
transmission. Blood meal analyses will be
conducted in fall 2005 and collection data is
summarized in Table 4.
ACKNOWLEDGMENTS I would like to thank Dr
William Irby for his support and guidance
throughout the project. This work was partially
funded by a grant from the CDC to Michael Womack
and partially through a Georgia Southern
University faculty grant.