Determination of mosquito (Diptera: Culicidae) bloodmeal sources in Western Australia: implications for arbovirus transmission.

A double-antibody enzyme-linked immunosorbent assay was used to determine the bloodmeal sources of adult mosquitoes (Diptera: Culicidae) collected in encephalitis vector surveillance mosquito traps in Western Australia between May 1993 and August 2004. In total, 2,606 blood-fed mosquitoes, representing 29 mosquito species, were tested, and 81.7% reacted with one or more of the primary antibodies. Aedes camptorhynchus (Thomson) and Culex annulirostris Skuse were the most common species tested, making up 47.2% (1,234) and 35.6% (930), respectively. These species obtained bloodmeals from a variety of vertebrate hosts but particularly marsupials and cows. In contrast, Culex pullus Theobald (72.7%; 24/33), Culiseta atra (Lee) (70.0%; 7/10), Culex globocoxitus Dobrotworsky (54.5%; 12/22), and Culex quinquefasciatus Say (39.3%; 22/56) often obtained bloodmeals from birds. Although Ae. camptorhynchus and Cx. annulirostris are well established vectors of arboviruses, other mosquitoes also may have a role in enzootic and/ or epizootic transmission.

Impact of dryland salinity on population dynamics of vector mosquitoes (Diptera: Culicidae) of Ross River virus in inland areas of southwestern Western Australia.

Clearing of native vegetation for agriculture since European settlement has left 1.047 million ha of southwestern Australia affected by a severe form of environmental degradation called dryland salinity, characterized by secondary soil salinization and waterlogging. This area may expand by a further 1.7-3.4 million ha if current trends continue. Detailed investigations of seasonal of adult and larval mosquito population dynamics were undertaken in the region to test the hypothesis that the development of dryland salinity and waterlogging in inland southwestern Australia has led to a succession of mosquito species and increased Ross River virus (family Togaviridae, genus Alphavirus, RRV) transmission risk. Aedes (Ochlerotatus) camptorhynchus (Thomson) made up >90% of adult mosquito collections in saline regions. Nonmetric multidimensional scaling and generalized estimating equations modeling demonstrated that it was strongly associated with increasing severity of dryland salinity. This article describes the first detailed investigation of the mosquito fauna of inland southwestern Australia, and it is the first description of the influence of secondary soil salinity on mosquito population dynamics. Despite the dominant presence of Ae. camptorhynchus, RRV disease incidence is not currently a significant population health priority in areas affected by dryland salinity. Potential limiting factors include local climatic impacts on the seasonal mosquito population dynamics, vertebrate host distribution and feeding behavior of Ae. camptorhynchus, and the scarce and uneven distribution of the human population in the region.

Flavivirus isolations from mosquitoes collected from Saibai Island in the Torres Strait, Australia, during an incursion of Japanese encephalitis virus.

Adult mosquitoes (Diptera: Culicidae) were collected in January and February 2000 from Saibai Island in the Torres Strait of northern Australia, and processed for arbovirus isolation during a period of Japanese encephalitis (JE) virus activity on nearby Badu Island. A total of 84 210 mosquitoes were processed for virus isolation, yielding six flavivirus isolates. Viruses obtained were single isolates of JE and Kokobera (KOK) and four of Kunjin (KUN). All virus isolates were from members of the Culex sitiens Weidemann subgroup, which comprised 53.1% of mosquitoes processed. Nucleotide sequencing and phylogenetic analysis of the pre-membrane region of the genome of JE isolate TS5313 indicated that it was closely related to other isolates from a sentinel pig and a pool of Cx. gelidus Theobald from Badu Island during the same period. Also molecular analyses of part of the envelope gene of KUN virus isolates showed that they were closely related to other KUN virus strains from Cape York Peninsula. The results indicate that flaviviruses are dynamic in the area, and suggest patterns of movement south from New Guinea and north from the Australian mainland.

Mosquito host-feeding patterns and implications for Japanese encephalitis virus transmission in northern Australia and Papua New Guinea.

Japanese encephalitis (JE) virus spread to northern Australia during the 1990s, transmitted by Culex annulirostris Skuse and other mosquitoes (Diptera: Culicidae). To determine the relative importance of various hosts for potential vectors of JE virus, we investigated the host-feeding patterns of mosquitoes in northern Australia and Western Province of Papua New Guinea, with particular attention to pigs, Sus scrofa L. - the main amplifying host of JE virus in South-east Asia. Mosquitoes were collected by CDC light traps baited with dry ice and 1-octen-3-ol, run 16.00-08.00 hours, mostly set away from human habitations, if possible in places frequented by feral pigs. Bloodmeals of 2569 mosquitoes, representing 15 species, were identified by gel diffusion assay. All species had fed mostly on mammals: only <10% of bloodmeals were from birds. The predominant species was Cx. annulirostris (88%), with relatively few (4.4%) bloodmeals obtained from humans. From all 12 locations sampled, the mean proportion of Cx. annulirostris fed on pigs (9.1%) was considerably lower than fed on other animals (90.9%). Highest rates of pig-fed mosquitoes (>30%) were trapped where domestic pigs were kept close to human habitation. From seven of eight locations on the Australian mainland, the majority of Cx. annulirostris had obtained their bloodmeals from marsupials, probably the Agile wallaby Macropus agilis (Gould). Overall proportions of mosquito bloodmeals identified as marsupial were 60% from the Gulf Plains region of Australia, 78% from the Cape York Peninsula and 64% from the Daru area of Papua New Guinea. Thus, despite the abundance of feral pigs in northern Australia, our findings suggest that marsupials divert host-seeking Cx. annulirostris away from pigs. As marsupials are poor JE virus hosts, the prevalence of marsupials may impede the establishment of JE virus in Australia.

Collection of wind-borne haematophagous insects in the Torres Strait, Australia.

Circumstantial evidence has implicated wind-borne mosquitoes (Diptera: Culicidae) in the introduction of Japanese encephalitis (JE) virus into Australia from the New Guinea mainland. A study was initiated on Saibai Island in the northern Torres Strait, during January and February 2000, to identify the potential source of insects collected in aerial (kytoon) and surface-level traps. Wind speed and direction were recorded to determine wind profiles during insect sampling. Northerly winds capable of carrying insects from New Guinea to Saibai Island were only present on three out of 18 nights sampled. Only three male mosquitoes, comprising two Verrallina funerea (Theobald) and one Ochlerotatus vigilax (Skuse), were collected in aerial samples, and were most likely of local origin. Culicoides midges were also collected in aerial nets and included gravid/parous C. bundyensis Lee and Reye, and one parous C. histrio Johannsen. Highest densities of arthropods (up to 1562/million m3) were on 30 January 2000 when NW winds, sustained for six hours, probably introduced midges from the New Guinea mainland. Adult mosquitoes (including three female Ve. funerea and a single female Ficalbia) and Culicoides (including two gravid C. bundyensis and one parous C. cordiger Macfie) were also collected in 2 m high mast nets during northerly surface winds. Although the results do not provide evidence that wind-blown mosquitoes introduced JE from New Guinea into Australia, they do not preclude that strong N winds associated with low pressure systems SW of the Torres Strait could have done so. However, results suggest that Culicoides were more likely than mosquitoes to reach high altitude and travel long distances during the light N winds experienced during the study.

Emerging viral diseases of Southeast Asia and the Western Pacific.

Over the past 6 years, a number of zoonotic and vectorborne viral diseases have emerged in Southeast Asia and the Western Pacific. Vectorborne disease agents discussed in this article include Japanese encephalitis, Barmah Forest, Ross River, and Chikungunya viruses. However, most emerging viruses have been zoonotic, with fruit bats, including flying fox species as the probable wildlife hosts, and these will be discussed as well. The first of these disease agents to emerge was Hendra virus, formerly called equine morbillivirus. This was followed by outbreaks caused by a rabies-related virus, Australian bat lyssavirus, and a virus associated with porcine stillbirths and malformations, Menangle virus. Nipah virus caused an outbreak of fatal pneumonia in pigs and encephalitis in humans in the Malay Peninsula. Most recently, Tioman virus has been isolated from flying foxes, but it has not yet been associated with animal or human disease. Of nonzoonotic viruses, the most important regionally have been enterovirus 71 and HIV.

Entomological investigations of an outbreak of Japanese encephalitis virus in the Torres Strait, Australia, in 1998.

Japanese encephalitis (JE) virus first appeared in Australia in 1995, when three clinical cases (two fatal) were diagnosed in residents on Badu Island in the Torres Strait, northern Queensland. More recently, two confirmed human JE cases were reported in the Torres Strait Islands and Cape York Peninsula, in northern Queensland in 1998. Shortly after JE virus activity was detected in humans and sentinel pigs on Badu Island in 1998, adult mosquitoes were collected using CO2 and octenol-baited CDC light traps; 43 isolates of JE virus were recovered. Although Culex sitiens group mosquitoes yielded the majority of JE isolates (42), one isolate was also obtained from Ochlerotatus vigilax (Skuse). Four isolates of Ross River virus and nine isolates of Sindbis (SIN) virus were also recovered from members of the Culex sitiens group collected on Badu Island in 1998. In addition, 3,240 mosquitoes were speciated and pooled after being anesthetized with triethylamine (TEA). There was no significant difference in the minimum infection rate of mosquitoes anesthetized with TEA compared with those sorted on refrigerated tables (2.8 and 1.6 per 1,000 mosquitoes, respectively). Nucleotide analysis of the premembrane region and an overlapping region of the fifth nonstructural protein and 3' untranslated regions of representative 1998 Badu Island isolates of JE virus reveled they were identical to each other. Between 99.1% and 100% identity was observed between 1995 and 1998 isolates of JE from Badu Island, as well as isolates of JE from mosquitoes collected in Papua New Guinea (PNG) in 1997 and 1998. This suggests that the New Guinea mainland is the likely source of incursions of JE virus in Australia.

Flaviviruses isolated from mosquitoes collected during the first recorded outbreak of Japanese encephalitis virus on Cape York Peninsula, Australia.

In response to an outbreak of Japanese encephalitis (JE) virus on Cape York Peninsula, Australia, in 1998, mosquitoes were collected using CO2 and octenol-baited Centers for Disease Control and Prevention light traps. A total of 35,235 adult mosquitoes, comprising 31 species, were processed for virus isolation. No isolates of JE virus were recovered from these mosquitoes. However, 18 isolates of Kokobera virus, another flavivirus were obtained from Culex annulirostris. Twelve isolates were from western Cape York (minimum infection rate (MIR) of 0.61: 1,000 mosquitoes) and 6 were from the Northern Peninsula Area (MIR of 1.0:1,000). Potential explanations for the failure to detect JE virus in mosquitoes collected from Cape York Peninsula include the timing of collections, the presence of alternative bloodmeal hosts, differences in pig husbandry, asynchronous porcine seroconversion, and the presence of other flaviviruses.

Japanese encephalitis on Badu Island, Australia: the first isolation of Japanese encephalitis virus from Culex gelidus in the Australasian region and the role of mosquito host-feeding patterns in virus transmission cycles.

During investigation of an outbreak of Japanese encephalitis (JE) in the Torres Strait, Australia, in 2000, mosquitoes were collected in Badu Island community and at a newly established communal piggery about 3 km from the community. A total of 94,285 mosquitoes, comprising 91,240 (96.8%) unengorged females, 1630 (1.7%) blood-engorged females and 1415 (1.5%) males, were processed for virus isolation. One isolate of JE virus was obtained from Culex gelidus, with a minimum infection rate of 12.4:1000. This is the first isolate of JE virus from Cx. gelidus in the Australasian region. No isolates were obtained from Cx. annulirostris, the primary implicated Australian JE vector. Analysis of mosquito host-feeding patterns, using gel diffusion, demonstrated that Cx. annulirostris and 5 other species fed predominately on mammals. Analysis of blood-fed mosquitoes collected within the community demonstrated that the proportion of Cx. annulirostris feeding on pigs in 2000 (2.3%) was significantly lower than that for the 1995-97 period (31.3%). The removal of the pigs from Badu Island community has limited the contact between potential amplifying hosts and mosquitoes, thus potentially reducing the risk of transmission of JE virus to the human population.

The appearance of a second genotype of Japanese encephalitis virus in the Australasian region.

In mid-January 2000, the reappearance of Japanese encephalitis (JE) virus activity in the Australasian region was first demonstrated by the isolation of JE virus from 3 sentinel pigs on Badu Island in the Torres Strait. Further evidence of JE virus activity was revealed through the isolation of JE virus from Culex gelidus mosquitoes collected on Badu Island and the detection of specific JE virus neutralizing antibodies in 3 pigs from Saint Pauls community on Moa Island. Nucleotide sequencing and phylogenetic analyses of the premembrane and envelope genes were performed which showed that both the pig and mosquito JE virus isolates (TS00 and TS4152, respectively) clustered in genotype I, along with northern Thai, Cambodian, and Korean isolates. All previous Australasian JE virus isolates belong to genotype II, along with Malaysian and Indonesian isolates. Therefore, for the first time, the appearance and transmission of a second genotype of JE virus in the Australasian region has been demonstrated.

Isolation of Japanese encephalitis virus from mosquitoes (Diptera: Culicidae) collected in the Western Province of Papua New Guinea, 1997-1998.

After Japanese encephalitis (JE) virus emerged in the Torres Strait in Australia in 1995, investigations were initiated into the origin of the incursion. New Guinea was considered the most likely source, given its proximity to islands of the Torres Strait. Almost 400,000 adult mosquitoes were processed for virus isolation from 26 locations in the Western Province of Papua New Guinea (PNG) between February 1996 and February 1998, yielding three isolates of JE virus. Two isolates of Murray Valley encephalitis, 17 isolates of Sindbis, and 1 each of Sepik and Ross River viruses were also obtained. Nucleic acid sequences of the PNG JE isolates were determined in the prM region, and in a region overlapping a part of the fifth nonstructural protein and the 3' untranslated region. The PNG isolates belonged to genotype II, and shared > 99.2% identity with isolates from humans and mosquitoes from the Torres Strait, suggesting that PNG is the source of incursions of JE virus into Australia.

Japanese encephalitis in north Queensland, Australia, 1998.

OBJECTIVE: To describe the circumstances of two cases of Japanese encephalitis (JE) in north Queensland in 1998, including one acquired on the Australian mainland. DESIGN: Serological surveillance of sentinel pigs for JE virus activity; serological surveys of humans and pigs and viral cultures of mosquito collections. SETTING: Islands in the Torres Strait and communities in the Northern Peninsula Area (NPA) and near the mouth of the Mitchell River in Cape York, Queensland, in the 1998 wet season (December 1997-May 1998). RESULTS: Sentinel pigs in the Torres Strait began to seroconvert to JE virus in February 1998, just before onset of JE in an unvaccinated 12-year-old boy on Badu island. By mid-April, most sentinel pigs had seroconverted. Numerous JE viruses were isolated from Culex annulirostris mosquitoes collected on Badu. In early March, a person working at the mouth of the Mitchell River developed JE. Serological surveys showed recent JE virus infection in 13 young pigs on a nearby farm, but not in 488 nearby residents. In NPA communities, sentinel pigs seroconverted slowly and JE viruses were isolated from three, but none of 604 residents showed evidence of recent infection. Nucleotide sequencing showed that 1998 JE virus isolates from the Torres Strait were virtually identical not only to the 1998 isolate from an NPA pig, but also to previous (1995) Badu isolates. CONCLUSIONS: JE virus activity was more widespread in north Queensland in the 1998 wet season than in the three previous wet seasons, but ecological circumstances (e.g., less intensive pig husbandry, fewer mosquitoes) appear to have limited transmission on the mainland. Nucleotide sequencing indicated a common source for the 1995 and 1998 JE viruses. Circumstantial evidence suggests that cyclonic winds carried infected mosquitoes from Papua New Guinea.

Identification of Australian arboviruses in inoculated cell cultures using monoclonal antibodies in ELISA.

An ELISA using a panel of specific monoclonal antibodies was developed to identify all alpha and flaviviruses isolated from mosquitoes caught throughout Australia. This technique is sensitive and rapid and is more specific than the traditional methods used to identify flaviviruses. The ability to identify unknown virus isolates from field-caught mosquitoes quickly and accurately improves the efficiency of arbovirus surveillance programs and allows health authorities to give an early warning of an increased health risk from a mosquito-borne virus in a particular region.

Arboviruses in the Australian region, 1990 to 1998.

Arboviruses continue to be major human pathogens in the Australian region. This report provides a summary of the activities of these viruses over the past eight years, and comments on new findings relevant to their respective ecologies. Of particular interest and concern is the propensity of these viruses to spread. The examples discussed include the initiation of dengue epidemics in north Queensland by virus imported in viraemic travellers; the spread of Japanese encephalitis virus to the Australasian region and its probable enzootic establishment in the south-west of Papua New Guinea; the potential spread of Ross River virus to other countries, as demonstrated by the 1979-80 outbreak in the South Pacific, and the recent occurrence in military personnel from the United States of America after an exercise; and the recent spread of Barmah Forest virus into Western Australia.

Two possible mechanisms for survival and initiation of Murray Valley encephalitis virus activity in the Kimberley region of Western Australia.

Two possible mechanisms are described for the initiation of Murray Valley encephalitis (MVE) virus activity in arid, epizootic regions of tropical Australia. Virus isolations were made from mosquitoes trapped shortly after the first heavy wet season rains and flooding in the east Kimberley, which followed approximately nine months of drought. A number of isolates of MVE virus were obtained, including isolates from pools of blood-engorged Culex annulirostris mosquitoes and from a single pool of male Aedes tremulus mosquitoes. The results strongly suggested that MVE virus activity was due both to its introduction in viremic vertebrate hosts, from which first-generation mosquitoes became infected following blood meals, and also to reactivation of vertically transmitted virus from desiccation-resistant eggs of Ae. tremulus. Both mechanisms are discussed with respect to environmental conditions.

An outbreak of Barmah Forest virus disease in the south-west of Western Australia.

OBJECTIVES: To describe the first reported outbreak of Barmah Forest (BF) virus disease in the south-west of Western Australia. DESIGN: Case series correlated with results of arbovirus surveillance. All patients with clinically suspected Ross River (RR) virus infection were serologically tested for antibodies to BF and RR viruses. Home address and date of presentation of patients with serologically confirmed recent infection were recorded. Mosquitoes collected from the districts before and during the BF virus outbreak were identified to species level and tested for virus. RESULTS: Twenty-two cases of BF disease were reported from the region between August 1992 and March 1994. Most occurred in the Peel region in the spring and early summer of 1993. Eighteen isolates of BF virus were obtained from three different species of mosquito trapped between January and October 1993. Fifteen were from mosquitoes in the Peel region and a single isolate was from the Perth metropolitan area. No isolates were obtained from the region before 1993. RR virus was not isolated from mosquitoes trapped in the region during the BF virus outbreak. CONCLUSIONS: Most BF infections were acquired in the Peel region during spring and early summer of 1993. Aedes camptorhynchus mosquitoes were probably the main vectors. The lack of isolations from mosquitoes before 1993 suggests that the virus may have only recently been introduced (or reintroduced) to the region. It was transmitted under conditions that were apparently not conducive to transmission of RR virus.

Ross River virus isolations from mosquitoes in arid regions of Western Australia: implication of vertical transmission as a means of persistence of the virus.

Outbreaks of mosquito-borne Ross River (RR) virus disease (epidemic polyarthritis) occur suddenly in the arid north and interior of the State of Western Australia, often within a few weeks of heavy rainfall. Between outbreaks, these regions may undergo long periods of drought, with little or no mosquito or arbovirus activity. The means by which RR virus is reintroduced or reactivated in these areas when environmental conditions favor mosquito-borne virus activity are unknown. In this paper, we describe isolations of RR virus from eight mosquito species trapped at two different locations, one coastal and one inland, in the arid Pilbara region of Western Australia, prior to outbreaks of epidemic polyarthritis. The isolation of RR virus has not been previously reported for five of these species and the isolations from the other three species are new records for Western Australia. The timing and number of isolations of RR virus from Aedes (Ochlerotatus) vigilax (Skuse, 1889) implicate that species as a vector of the virus on the Pilbara coast. Significantly, RR virus was isolated from pools of male Ae. vigilax and male Ae. (Macleaya) tremulus (Theobald, 1903) mosquitoes. This is the first report of RR virus (or other Australian arbovirus) isolates from wild-caught male mosquitoes. Both Ae. vigilax and Ae. tremulus have desiccation-resistant eggs that can survive long periods of drought, making them ideal candidates for the overwintering of arboviruses. The findings implicate vertical transmission as a means of persistence of RR virus in arid regions of Australia and therefore offer a likely explanation for the sudden recurrence of virus activity after heavy rains.