VECTOR-BORNE VIRAL DISEASES

Arthropod-borne diseases are interesting because pathogen transmission and mainte­nance is typically more complex than directly transmitted diseases. The ecology of vector- borne diseases necessarily involves the ecology of the vectors, including host-parasite inter­actions, seasonality, patterns of abundance, geographic distribution, habitat preferences, and host preferences.

WEST NILE VIRUS IN NORTH AMERICA West Nile virus (WNV) is a remarkable case study for many reasons: its ability to cause disease in humans, the patterns of the explosive epidemic spread following introduction of an exotic pathogen, the potentially dramatic impacts on wildlife populations and communities, and the potential effects of climate change on the global distribution of this vector-borne disease.

causative agent West Nile virus is a mosquito-borne Arbovirus, meaning that it is an arthropod-borne RNA virus in one of the following families: Togaviridae, Flaviviridae, Bunyaviridae, or Arenaviridae. For histori­cal purposes, other arthropod-borne viruses generally are not discussed as Arboviruses. Specifically, West Nile virus is a flavivirus in the Family Flaviviridae (App. 1: Table 8), which also includes such infamous viruses as yellow fever virus, dengue fever viruses, Japanese encephalitis virus, and a group of related viruses that cause tick-borne encephalitis (Acha and Szyfres 2003b).

Flavivirus particles are spherical, approxi­mately 50 nm in diameter, and are covered by an envelope membrane derived from the host's cellular membranes.

geographic distribution and host range West Nile virus traditionally was found in Africa, southwestern Asia, and the Middle East, but moved into eastern Europe, western Europe, and Southeast Asia during the 1990s. In 1999, WNV was found in New York City, the first occurrence of this virus in the Americas (Acha and Szyfres 2003b, McLean and Ubico 2007). There were many reasons to be con­cerned about this virus becoming established in the New World, but its zoonotic potential is clearly a threat. That said, a serologic survey conducted by the Centers for Disease Control (CDC) during the 1999 epidemic in New York City indicated that as many as 1,250 people were infected, of whom 240 may have suf­fered clinical disease. However, there were only 61 reported human cases, and only seven peo­ple died, resulting in a case fatality rate of less than 0.6% (Foster and Walker 2009). As will be addressed, humans suffer much lower risks of mortality than do many species of wild birds.

It is no surprise that variation occurs in nature among pathogens as well as among hosts, and this is certainly the case with WNV. The viral variant (NY99) that arrived in New York was most similar to a contemporary variant from Israel, suggesting that the new variant evolved in the Middle East (McLean et al. 2002, McLean and Ubico 2007). This variant is more virulent than other strains, and proved to be especially virulent for American crows (Corvus brachyrhynchos) and other species in the fam­ily Corvidae. Once in North America, the virus moved rapidly westward, northward, and southward, so that by 2003 the virus was found throughout much of the United States and parts of southern Canada.

Wild birds are clearly the primary hosts of this virus, and a large variety of avian species serve as reservoirs, amplifying hosts, or inci­dental, dead-end, hosts in different regions of the world. In North America alone, over 300 species of birds have been found dead as a result of WNV infections (Acha and Szyfres 2003b, McLean and Ubico 2007, Bowen 2011a), and the key is likely the habitat overlap of avian hosts and the primary or amplifying ornitho- philic vectors. Between 1999 and 2004, 47,016 birds reportedly died in North America, com­prising a total of 294 species from 57 families and 24 avian orders (McLean and Ubico 2007).

Importantly, WNV also commonly infects many mammals, including people, horses, some carnivores, bats, rabbits, and squirrels. Over 100 species of mammals have been associated with WNV (Root 2013). However, most mammals are fairly resistant and do not develop adequate levels of viremia to facili­tate transmission to mosquitoes. Therefore, most mammals should be considered to be incidental, dead-end, hosts that do not con­tribute to the maintenance of the virus. Some species of mammals, including tree squirrels (Sciurus spp.), eastern chipmunks (Tamias striatus), and eastern cottontail rabbits (Sylvilagusfloridanus), develop moderate levels of viremia adequate to infect some species of vectors. Additionally, mesocarnivores includ­ing raccoons (Procyon lotor), deer (Odocoileus spp.), and tree squirrels have been suggested to be useful species for surveillance and monitor­ing (Blitvich et al. 2009, Docherty et al. 2009, Root 2013). It is important to understand that West Nile virus causes encephalitis and death in horses and humans.

The host range of the virus also includes American alligators (Alligator mississippiensis) in Florida, crocodiles (Crocodylus niloticus) from Israel, and occasionally other reptiles and some amphibians. American alligators have been shown to develop adequate levels of vire­mia to allow them to serve as local amplifying hosts, and they suffer disease associated with WNV infections (Miller et al. 2003, Jacobson et al. 2005, Farfan-Ale et al. 2006, Ariel 2011). Finally, hundreds of species of mosquitoes serve as hosts of the virus worldwide, and these relationships will be covered in the next section.

LIFE CYCLES, RESERVOIRS, AND TRANSMISSION Ecological maintenance of WNV involves cycles among wild birds and ornithophilic (bird-feeding) mosquitoes. Although hundreds of different species of vertebrates have been associated with WNV, maintenance cycles typically involve passerine birds and Culex spp. mosquitoes. Those species of birds that develop the highest levels of viremia also serve as important reservoir hosts. Of course, there is a trade-off between the infectiousness of birds with extremely high viremia that cause severe disease and kill birds quickly, versus species that develop high viremia but survive a bit lon­ger; those that develop moderately high viremia but survive longer, including many finches and sparrows, may serve as better maintenance hosts (Komar et al. 2003, Bowen and Nemeth 2007, Guerrero-Sanchez et al. 2011, Hamer et al. 2011); in contrast, those that develop the highest viremia and die most quickly, includ­ing several of the corvids, may serve as better amplifying hosts. Most species that suffer little disease also fail to develop high enough levels of viremia to be of importance in ecological cycles (Langevin et al.

Reports of experimental exposures of 25 species (from 10 different orders) of birds to mosquitoes infected with the epidemic-type strain of WNV (isolated from a crow in 1999) were reviewed (Komar et al. 2003). Several authors reported considerable variation among species and orders of birds in terms of both the level and duration of viremia as well as mortality. Corvids such as American crows, blue jays (Cyanocitta cristata), and black-billed magpies (Pica pica), as well as common grack- les (Quiscalus quiscula) and house sparrows (Passer domesticus), suffer extremely high levels of viremia and mortality (Komar et al. 2003, Brault et al. 2004, Weingartl et al. 2004, Langevin et al. 2005). Additional species that suffer high levels of viremia and mortality include greater sage-grouse (Centrocercus urophasianus) (Clark et al. 2006), western scrub jays (Aphelocoma coerulescens), and house finches (Carpodacus mexicanus) (Reisen et al. 2005). Other spe­cies, including fish crows (Corvus ossifragus), European starlings (Sturnus vulgaris), American robins (Turdus migratorius), and a variety of raptors (Gancz et al. 2006; Joyner et al. 2006; Nemeth et al. 2006a, 2006b; Saito et al. 2007; Dusek et al. 2010; Ziegler et al. 2013), develop moderate to high levels of viremia and likely serve as amplifying hosts in areas where they are abundant. Finally, some species including domestic chickens and turkeys, doves, pigeons, and quail rarely develop adequate viremia to transmit the virus to mosquitoes (Langevin et al. 2001, Reisen et al. 2005, Guerrero-Sanchez et al. 2011). Young birds tend to suffer higher levels of viremia than do most adults (McLean and Ubico 2007), suggesting that there may be some transmission from young birds of mod­erately resistant species, but it seems unlikely that these low levels of transmission contribute significantly to local reservoirs.

Although most amphibians and rep­tiles have not been found to be infected with WNV, some species, including green iguanas (Iguana iguana), red-ear sliders (Trachymes elegans), garter snakes (Thamnophis sirtalis sirtalis), and bullfrogs (Rana catesbeiana), developed low-level infections following experi­mental inoculations (Bowen and Nemeth 2007). American alligators especially, but also crocodiles, exposed both experimen­tally and naturally, develop adequate levels of viremia to contribute to local reservoirs. (Miller et al. 2003, Steinman et al. 2003, Klenk et al. 2004, Jacobson et al. 2005).

In Europe, Africa, and Asia, migratory birds are considered very important to the spread and maintenance of WNV (Malkinson and Banet 2002, Malkinson et al. 2002), and migratory species have been important in the spread of the virus through Central America, the Caribbean, and South America. However, the role of migratory species has been less clear in the epidemic wave that swept across North America (McLean and Ubico 2007). Some early models suggested a critical role for migratory species (Peterson et al. 2003), but the radia­tion within North America spread steadily and did not show a leap-frog progression along the flyways as would be expected if migratory spe­cies were the key to the epidemic wave (Rappole and Hubalek 2003, Rappole et al. 2006). The multidirectional, steady spread suggests that lateral movements of resident species, coupled with a large variety of potential vectors in a wide range of habitats, as well as the broad host range of the primary vectors, contributed to the observed pattern of spread in North America.

Most transmission of WNV occurs via the bites of vector-competent mosquitoes. However, vector competence and vectorial capacity of mosquito species are as variable as is the res­ervoir competence of different avian species (Brault 2009). Vectorial capacity depends upon a complex set of characteristics involving spe­cies abundance, host feeding preferences, and ability to become infected, remain infected, and to infect naive hosts.

In general, Culex spp. mosquitoes are the primary vectors of West Nile virus among potential reservoir hosts, as well as to many amplification hosts and incidental hosts. Many species in other genera have been found to be infected, and some species in other genera may be essential as bridge vectors or maintenance vectors in regional or specific circumstances; the CDC lists 65 species found associated with WNV in North America between 1999 and 2012 (Centers for Disease Control 2012c). However, Culex spp. provide the key to the large-scale ecological patterns associated with WNV (Acha and Szyfres 2003b, McLean and Ubico 2007, Foster and Walker 2009). The importance of different species varies in different regions. The C. pipiens complex of species occurs broadly over the temperate and tropical zones of all continents, and this is the single most important vector group worldwide. Other spe­cies of primary vectors in this genus include C. tarsalis in much of North America; C. restuans, especially in the eastern and central United States and south-central Canada; C. univittatus in the Middle East and Africa; C. pipiens, C. modestus, and Coquillettidia richiardii in Europe; and C. quinquefasciatus, C. tritaeniorhynchus, and C. vishnui in parts of Asia (Hubalek and Halouzka 1999).

Mosquito communities vary across seasons and habitats at all spatial scales (from continen­tal to fine-scale patterns associated with land use and availability of different types of breeding habitat). The importance of a given vector in any locality depends upon complex interactions of many variables, but the bottom line is the num­ber of infectious mosquitoes that result from feeding upon different species of vertebrates within a community of hosts (Hamer et al. 2011). It is noteworthy that relatively small differences in vector competence within and between popu­lations, or in association with strain variation of the virus, can be associated with significant differences in vector competence among popu­lations or among years for individual species of vectors (Reisen et al. 2005, Vaidyanathan and Scott 2007, Reisen et al. 2008).

In addition to vertebrate hosts and a large number of species of mosquitoes, a few tick spe­cies also have been found infected with WNV. In experimental studies, at least three argasid species (soft ticks), Ornithodoros moubata (Lawrie et al. 2004), Carios capensis (Hutcheson et al. 2005), and Argas arboreus (Abbassy et al. ¹993), were moderately efficient vectors of WNV and may be involved in some local foci. Additionally, Hyalomma marginatum, an ixo- did (hard tick) species from Europe and Asia, has been shown to become infected when feeding on infected rabbits, to maintain WNV through the transstadial molt, and to transmit the virus to susceptible hosts (Formosinho and Santos-Silva 2006). On the other hand, some other human-biting hard ticks, including both the black-legged tick, Ixodes ricinus (Lawrie et al. 2004), and the western black-legged tick, Ixodes pacificus (Reisen et al. 2007), lack vector competence for this virus.

Direct infections also can occur via contact with bodily secretions or infected tissues, and predators may be exposed by consumption of prey. Oral infections have been shown for a variety of vertebrates, including domestic cats (Austgen et al. 2004), fox squirrels (Tiawsirisup et al. 2010), mice (Odelola and Oduye 1977), hamsters (Sbrana et al. 2005), raptors (Nemeth et al. 2006a, 2006b), crows (McLean and Ubico 2007), and alligators (Klenk et al. 2004). Humans have become infected via a variety of routes that are typically unimportant in cycles involving wildlife; WNV has been transmitted transuterine, transmammary, and via blood transfusions (Julander et al. 2006, O'Leary et al. 2006), but these aren't routes of infection that would be available in ecological maintenance cycles. However, the epidemiologic importance of direct transmission among wildlife, either via direct horizontal exposures to secretions or via oral ingestion, remains largely unknown.

population effects Pathogens with broad host ranges are expected to vary in their effects on the health of individuals and populations of different species, and WNV provides a great example of this phenomenon. When WNV first became established in the New World, all popu­lations of potential hosts were naive to this new, exotic pathogen. It quickly became obvious that deaths among some species, including Ameri­can crows, were common, while other species, including mourning doves (Zenaida macroura) and gray catbirds (Dumetella carolin ens is), seemed resistant to infections (Komar et al. 2003, Kilpatrick et al. 2006). Although small­scale mortality events may not necessarily impact populations, WNV has clearly caused large-scale die-offs of some species associated with concern about population declines of both rare and common species. Community struc­ture can also be affected because of differential mortality among sympatric species (LaDeau et al. 2007, Hamer et al. 2011).

American crows suffered considerable population declines, with as high as 45% reductions occurring throughout its range and over two-thirds dying in some populations (Yaremych et al. 2004, LaDeau et al. 2007). LaDeau et al. (2007) used annual Breeding Bird Survey data to show significant range­wide declines of American crows, blue jays, tufted titmice (Baeolophus bicolor), American robins, chickadees, house wrens (Troglodytes aedon), and eastern bluebirds (Sialia sialis) that coincided with WNV expansion. However, even with these species, the impacts seemed to vary between regions, and this should be expected due to vector abundance, vector feeding prefer­ences, host community structure, land use pat­terns, and other environmental variables that influence exposures (LaDeau et al. 2007).

One of the greatest concerns for conser­vation biologists has been that WNV would impact species of conservation concern. Greater sage-grouse have suffered severe losses associ­ated with habitat loss and degeneration, and they are endangered in Canada and a candidate species in the United States under the Federal Endangered Species Act. Habitat availability for this species has been reduced to approximately 56% of the geographic range occupied prior to the westward expansion ofEuropeans (Connelly et al. 2011), and numbers have correspondingly declined by 45-80% (Naugle et al. 2004, Clark et al. 2006) in different populations. On top of these problems, sage grouse have been shown to be highly susceptible to WNV in laboratory studies, in which all exposed grouse died within 6 days of exposure (Clark et al. 2006). Naugle et al. (2004) reported a mean of 25% reduction in late-summer survivorship in four populations due to WNV infections, and none of 112 greater sage-grouse had antibodies show­ing prior exposure to WNV. The lack of sero­logically positive grouse, in this case, is highly suggestive that those that become exposed died too quickly to be sampled. West Nile virus infec­tions have proven to be the greatest disease risk facing greater sage-grouse, and the disease has a significant impact on plans for the species' recovery (Connelly et al. 2011).

Clearly, different mosquito assemblages associated with different host species in differ­ent habitats and regions of the world result in unique impacts on populations and commu­nities. Furthermore, changes in risks for sus­ceptible populations and communities of hosts vary as the virus strains continue to evolve and immunity among host populations waxes and wanes. Most of the literature cited above con­cerns disease that progresses rapidly to death. However, since many species are somewhat resistant, and a variable percentage of infected individuals of susceptible species die, less obvi­ous effects associated with chronic disease should be expected (Ward et al. 2010). Ward et al. (2010) studied survival of a marked popu­lation of cardinals (Cardinalis cardinalis) and reported higher mortality of birds lacking anti­bodies during the peak WNV season but lower mortality for birds lacking antibodies during the rest of the year. Higher mortality during the summer suggests deaths associated with WNV. Higher mortality of exposed birds during the rest of the year suggests negative influences related to chronic disease.

CLINICAL EFFECTS AND IDENTIFICATION Exposure to WNV results in the full range of potential severity in different species of wild­life. In those that become successfully infected, some species tend to develop antibodies with­out showing any signs of disease. In other spe­cies, some, most, or nearly all of the individuals develop disease and die.

Clinical and pathological signs of disease in wild birds vary by species depending, in part, on the level of viremia developed and the dura­tion of the disease. Live birds suffering from WNV infections may be found with any level of body condition, but some are found in poor body condition and may appear emaciated; however, some species (including blue jays and American crows) die so quickly that live sick birds are rarely observed. Clinical signs consis­tent with neurologic problems associated with meningitis and encephalitis are very common among bird species that are susceptible to the virus. Such birds may seem weak and unable to perch, movements may appear uncoordinated (ataxic), and they may show signs of head tilt, body tremors, or the rigid, stiff* necks of animals with meningitis (torticollis). Inflammation of the heart muscle (myocarditis) is a common finding in many studies. Inflammation of the spleen, kidneys, or liver may lead to enlarge­ment of the organs, with focal areas of necrosis and outward signs of disease depending on which organs become most severely involved; however, once such signs become severe, mul­tiple organ shutdown may lead quickly to death. Involvement of the cells lining the blood vas­cular system may cause hemorrhages on organ surfaces (sometimes including the brain), and dead and dying birds may show signs of bleeding in the mouth, nares, or cloaca (Steele et al. 2000; Komar et al. 2003; Weingartl et al. 2004; Gancz et al. 2006; Nemeth et al. 2006a, 2006b, 2009; Ernest et al. 2010; Wheeler et al. 2011; Ziegler et al. 2013).

Mammals are also exposed to WNV during outbreaks, and some species suffer moder­ate levels of viremia, disease, and death (Root 2013). Although most species don't develop ade­quate levels of viremia to contribute to local res­ervoirs, some do; tree squirrels (including fox squirrels (Sciurus niger), eastern gray squirrels (S. carolinensis), and western gray squirrels (S. griseus) often are found sick or dead during out­breaks of WNV. Tree squirrels may suffer high rates of exposure due to the amount of time they are active in tree canopies in proximity to ornithophilic vectors. Root (2013) also noted that eastern cottontails (Sylvilagus floridanus) develop moderately high levels of viremia and may serve as minor amplification hosts in some areas. Like horses, alpacas, humans, squirrels, and carnivores (including domestic dogs and cats), seals also can be dead-end hosts of West Nile virus. In one study, a 12-year-old male harbor seal (Phoca vitulina) showed signs of progressive neurologic dysfunction includ­ing “head tremors, muzzle twitching, clonic spasms, and weakness,” and lesions were con­sistent with polio-encephalomyelitis (Del Piero et al. 2006). In humans, the incubation period is fairly short, at 3 to 6 days, but the onset of dis­ease can be sudden. General signs in humans include fever, headache, lymphadenopathy, and a maculopapular rash. Less commonly, humans develop myocarditis, meningitis, and encephalitis (Acha and Szyfres 2003b).

Alligators can be infected via oral ingestion or mosquito bite and develop signs similar to birds and mammals, with multi-organ necrosis, granulomas, and perivasculitis associated with viremia or associated bacteremia (Miller et al. 2003). They also develop meningoencephalits, necrotizing hepatitis and splenitis, pancreatic necrosis, myocardial degeneration with necro­sis, pneumonia, and necrotizing lesions in their mouths (Klenk et al. 2004). Uninfected juvenile alligators also become infected, pre­sumably by mosquito bite, when tank mates are exposed to the virus (Jacobson et al. 2005).

In summary, the clinical effects vary among the taxonomic groups of vertebrates. In suscep­tible species, most clinical signs result from inflammation of the brain, lining of the CNS, the heart muscle, the blood vascular system, and the organs that become inflamed and necrotic as a result of infection.

control No control procedures have blocked the establishment or spread of WNV throughout much of North America, South America, and Europe. However, local control efforts have likely saved millions of individual humans, horses, and wild animals. Control has been a multifaceted effort based on plan­ning, dissemination of information, use of a variety of sentinel animals (primarily small flocks of chickens), surveillance of dead wild birds to identify outbreaks as early as possible, various types of mosquito control, vaccination, and therapeutic treatment for humans. Many of the general control strategies employed for WNV are similar to control of any spreading, mosquito-borne zoonotic disease. However, the control of WNV also includes vaccination strat­egies unique to this pathogen.

There is no human vaccine currently avail­able, but there may be at some point in the near future. There are several currently marketed vaccines for horses, and several of these have been used “off-label” to protect zoo animals or in trials involving wildlife. Equine vaccines have been used to vaccinate fish crows (Corvus ossifragus) (Turell et al. 2003), American crows (Corvus brachyrhynchos) (Bunning et al. 2007), California condors (Gymnogyps californianus) (Chang et al. 2007), black-footed penguins (Spheniscus demersus), little blue penguins (Eudyptula minor) (Okeson et al. 2007), Hum­boldt penguins (Spheniscus humboldti), mag­ellanic penguins (Spheniscus magellanicus), Gentoo penguins (Pygoscelis papua), and rock­hopper penguins (Eudyptes chrysoscome) (Davis et al. 2008), American flamingos (Phoenicop- terus ruber), Chilean flamingos (Phoenicopterus chilensis), Attwater's prairie-chickens (Tympa- nuchus cupido attwateri) (Okeson et al. 2007), sandhill cranes (Grus canadensis) (Olsen et al.

2009), American robins (Turdus migratorius) (Kilpatrick et al. 2010), western scrub jays (Aphelocoma californica), and island scrub jays (Aphelocoma insularis) (Boyce et al. 2011, Wheeler et al. 2011). Generally, vaccination of wildlife should be considered for captive populations, endangered species, or for spe­cies with small or isolated populations, such as the island scrub jay. Likewise, medical treat­ment to try to save valued individuals may be largely ineffective and may be available only in specific circumstances; however, treatment remains an untenable management strategy even for those species of considerable conser­vation concern.

ORBIVIRUSES CAUSING HEMORRHAGIC DISEASE OF RUMINANTS

causative Agents Viruses of the genus Orbivirus, Family Reoviridae (App. 1: Table 8), are double-stranded RNA viruses with a double­layered protein capsid; these viruses replicate in cell cytoplasm (Calisher 1994, Roy 1996). Epizootic hemorrhagic disease virus (EHDV) and bluetongue virus (BTV) are closely related and are often collectively referred to as hemor­rhagic disease viruses (Howerth et al. 2001, MacLachlan and Osburn 2004, Saif 2011). Tra­ditionally, eight serotypes of EHDV have been identified, labeled serotypes EHDV-1 through EHDV-8. In recent years this has been raised to 10 serotypes (Mertens et al. 2005), with a recent proposal for seven serotypes (Anthony et al. 2009). Traditionally, 24 serotypes of BTV have been recognized (MacLachlan and Osburn 2004, Saif 2011); with two new isolates there are 26, labeled BTV-I through BTV-26 (Maan et al. 2011).

host range and geographic distribution Geographically, hemorrhagic disease viruses are distributed throughout Africa, North America, Australia, East Asia, and the Middle East (Howerth et al. 2001, Alli­son et al. 2010). For EHDV, serotypes 1 and 2 are characteristic of North America, serotypes 3 and 4 are linked to Africa, and serotypes 5, 6, 7, and 8 originally were isolated from Australia.

Both EHDV and BTV infect ruminant wildlife and domestic animals; however, clini­cal disease among wildlife is common only in North America; this is associated with EHDV serotypes 1 and 2, and the five BTV serotypes (2, 10, 11, 13, and 17) (Stallknecht and Howerth 2004). Bluetongue virus causes disease in both domestic and wild ruminants and results in serious losses among wildlife and domes­tic animals. However, while some strains of BTV appear established among wildlife in some regions (Corbiere et al. 2012), there is little evidence that either EHDV or BTV cause significant problems among wildlife outside of North America (Howerth et al. 2001).

Infections, clinical disease, and mortality in North American wildlife have been documented since 1955 (Shope et al. 1960, Nettles and Stallknecht 1992), and have been described in white-tailed deer (Odocoileus virginianus), mule deer (O. hemionus), elk (Cervus elaphus), prong­horn (Antilocapra americana), mountain goats (Oreamnos americanus), bison (Bison bison), and bighorn sheep (Ovis canadensis) (Howerth et al. 2001). Of these, white-tailed deer have been the most severely affected, and mortality has been associated with serotypes 10, 11, 13, and 17 of BTV and serotypes 1 and 2 of EHDV (Stallknecht and Howerth 2004). Recently an EHDV-6 was isolated from moribund and dead white-tailed deer in several states of the United States (Allison et al. 2010).

While there is serological evidence of EHDV infection in domestic cattle, EHDV strains 1 and 2 have not been shown to cause disease in cattle of North America (Abdy et al. 1999). In contrast, other EHDV serotypes have been linked to disease in cattle in other regions of the world (Omori et al. 1969, Breard et al. 2004, Yadin et al. 2008, Temizel et al. 2009).

RESERVOIRS AND TRANSMISSION Both EHDV and BTV are transmitted by biting midges ofthe genus Culicoides (Foster et al. 1977, Wittmann et al. 2002). Proven Culicoides vec­tors for BT viruses include C. imicola in Africa,

C. variipenis and C. insignus in North America, and C. fulvus and C. actoni in Australia. Less is known about vectors for EHD viruses. In North America, EHDV has been isolated only from

C. variipennis, although other species can be experimentally infected (Howerth et al. 2001).

CLINICAL EFFECTS AND DIAGNOSIS Clinical signs among deer often include hyperemia of the skin and mucous membranes, swelling of the face, anorexia, lethargy, nasal discharge, and bloody diarrhea (Howerth et al. 2001). Lesions resulting from the disease often are multisystemic, including severe edema, con­gestion, acute vascular necrosis, and hemor­rhage (Brodie et al. 1998, Howerth et al. 2001).

Several tests are available for serological diagnosis, including agar gel diffusion (Pearson and Jochim 1979) and serum neutralization tests (Howerth et al. 2001). Antibody prevalence for EHDV also can be determined from hunter- killed deer using whole blood dried on paper strips using serum neutralization (Dubay et al. 2006). However, antigenic cross-reactivity between BTV and EHDV results in serologic misdiagnosis with these tests (Mecham and Wilson 2004). Competitive enzyme-linked immunosorbent assays (c-ELISA) are sensi­tive and specific, but are time consuming and require large volumes of cells and reagents; also, the antigens prepared often vary in quality (Mecham and Wilson 2004).

There are several tests to identify the species and serotype of hemorrhagic disease viruses. A multiplex real-time polymerase chain reaction (RT-PCR) test is available (Aradaib et al. 2003). Another rapid real-time PCR test now can be used to detect all eight EHDV serotypes; it does not cross-react with BTV strains (Wilson et al. 2009b). The proteins used for serotyping both EHD-1 and EHD-2 appear to be quite stable over time (Mecham et al. 2003, Murphy et al. 2005). There currently also is a multiplex real-time reverse transcription PCR assay for detection of the various BTV serotypes that differentiates all BTV serotypes and does not cross-react with EHDV (Wilson et al. 2009a).

population effects Hemorrhagic dis­ease occurs in the late summer and early fall (Couvillion et al. 1981) and likely is related to seasonal patterns of the vector (Howerth et al. 2001). Mortality events in the western United States, especially those involving large numbers of animals, tend to be severe but sporadic (Nettles et al. 1992, Howerth et al. 2001). Generally, the frequency of hemorrhagic disease decreases, and the reported mortality increases, with latitude. (Nettles et al. 1992). In the southern United States, most reports of hemorrhagic disease in deer are linked with insignificant mortality (Nettles et al. 1992); many animals survive, as indicated by high antibody prevalences (Couvillion et al. 1981, Stallknecht et al. 1991). In these areas, mortality events occur annually or on a 2-3 year cycle (Couvillion et al. 1981).

Herd immunity can be a factor. There is evidence that infection with EHDV-I and EHDV-2 in white-tailed deer leads to a protec­tive immunity against the other virus (Gaydos et al. 2002).

special problems Other than the some­times substantial mortality from these dis­eases, and economic losses to ranchers, there are no persistent special problems identified for hemorrhagic diseases.

prevention and control There cur­rently are no wildlife management tools or strat­egies to prevent, predict, or minimize potential impacts of hemorrhagic disease viruses in wild populations (Howerth et al. 200I). There is evidence for innate resistance to EHDV-i and EHDV-2 among some white-tailed deer subspe­cies. For example, O. virginianus borealis fawns are very susceptible to both serotypes, whereas O. virginianus texanus experienced mild or non­detectable disease for each viral strain (Gaydos et al. 2002).

Bluetongue vaccines are avai lable for domestic animals, and vaccination is the pri­mary control for bluetongue in sheep. To date, only modified live (attenuated) virus vaccines have been used. Because of the variety of BTV serotypes and variable cross-protection between serotypes, vaccination has inconsistent success. No inactivated or subunit vaccines are currently available, though several experimental vaccine preparations have been studied. No vaccine is available for EHDV.