Transmission of Japanese Encephalitis Virus from the Black Flying Fox, Pteropus Alecto , to Culex Annulirostris Mosquitoes, Despite the Absence of Detectable Viremia

Home , Culex annulirostris

Transmission of Japanese Encephalitis Virus from the Black Flying Fox, Pteropus alecto , to Culex annulirostris Mosquitoes, Despite the Absence of Detectable Viremia

Andrew F. van den Hurk , * Craig S. Smith , Hume E. Field , Ina L. Smith , Judith A. Northill , Carmel T. Taylor , Cassie C. Jansen , Greg A. Smith , and John S. Mackenzie Virology, Queensland Health Forensic and Scientific Services, Coopers Plains, Queensland, Australia; School of Chemical and Molecular Biosciences, University of Queensland, St Lucia, Queensland, Australia; Department of Primary Industries and Fisheries, Animal Research Institute, Yeerongpilly, Queensland, Australia; Australian Army Malaria Institute, Gallipoli Barracks, Enoggera, Queensland, Australia; Australian Biosecurity Cooperative Research Centre, Curtin University of Technology, Perth, Western Australia, Australia

Abstract. To determine the potential role of flying foxes in transmission cycles of Japanese encephalitis virus (JEV) in Australia, we exposed Pteropus alecto (Megachiroptera: Pteropididae) to JEV via infected Culex annulirostris mosquitoes or inoculation. No flying foxes developed symptoms consistent with JEV infection. Anti-JEV IgG antibodies developed in 6/10 flying foxes exposed to infected Cx. annulirostris and in 5/5 inoculated flying foxes. Low-level viremia was detected by real-time reverse transcriptase polymerase chain reaction in 1/5 inoculated flying foxes and this animal was able to infect recipient mosquitoes. Although viremia was not detected in any of the 10 flying foxes that were exposed to JEV by mosquito bite, two animals infected recipient mosquitoes. Likewise, an inoculated flying fox without detectable viremia infected recipient mosquitoes. Although infection rates in recipient mosquitoes were low, the high population densities in roosting camps, coupled with migratory behavior indicate that flying foxes could play a role in the dispersal of JEV.

INTRODUCTION in eastern Australia. Certainly, the behavioral ecology of flying foxes deems them potential dispersal hosts of mammalian Japanese encephalitis virus (JEV) is a mosquito-borne fla- viruses, and several recently emerged zoonotic viruses, includ- vivirus of Southeast Asia responsible for over 40,000 clinical ing Hendra virus (HeV) and Australian bat lyssavirus, have 1 cases annually, with a 25% fatality rate. Japanese encephali- been described in flying foxes in Australia. 12 Furthermore, tis virus is an emerging virus in the Australasian region with both experimental and natural infection studies have sug- 2 potentially serious public and animal health implications. gested a possible role for bats in the transmission cycles of Widespread JEV activity was first detected in Australia in 1995 a number of arboviruses, including JEV (reviewed by Sulkin on Badu Island in Torres Strait, when it caused an outbreak and Allen). 13 However, the majority of these experiments have 3 consisting of three human cases and two deaths. Subsequently, examined insectivorous bats of the order Microchiroptera. except for 1999, JEV activity was recognized every year in the The only experimental infection of Megachiropteran bats Torres Strait between 1995 and 2005, when the sentinel pig with JEV has been undertaken in India. In these earlier stud- program was discontinued. In 1998, human and pig infections ies, viremia developed in both Rousettus leschenaulti and were identified for the first time on the Australian mainland Cynopterus sphinx , with the latter species able to infect recip- 4 at the mouth of the Mitchell River on Cape York Peninsula. ient mosquitoes. 14,15 Entomologic studies implicated Culex annulirostris as the pri- We undertook laboratory-based infection and transmis- 5,6 mary mosquito vector of JEV in northern Australia. sion experiments to determine whether flying foxes can Pigs and ardeid wading birds, such as herons and egrets are act as amplifying hosts for JEV. Specifically, we examined the primary amplifying hosts of JEV, as they produce viremia 1) the ability of flying foxes to develop clinical illness, vire- 7–10 sufficient to infect mosquitoes. Although humans and horses mia, and specific antibodies after being exposed to JEV via can develop fatal encephalitis, they are considered to be dead- infected mosquitoes or by inoculation; and 2) the likelihood 11 end hosts of the virus, developing only low-level viremia. of recipient mosquitoes becoming infected after feeding on Should JEV become established on the Australian main- flying foxes that had been previously exposed to JEV. The land, the potential importance of Australian fauna species Black flying fox, Pteropus alecto, was chosen because of acting as vertebrate amplifying hosts is largely unknown. its widespread geographic distribution throughout north- Experimental studies at the Commonwealth Scientific and ern Australia, including areas where JEV has previously Industrial Research Organization Australian Animal Health occurred.16 Laboratory previously examined the possible role of a number of Australian native species (such as wallabies and pos- sums) as reservoir or amplifying hosts.2 However, flying MATERIALS AND METHODS foxes (Megachiroptera: Pteropididae) were not included in All animal experimentation complied with the Queensland these studies, and mosquitoes were not used as the route of Animal Care and Protection Act 2001 and was approved by virus exposure or to show transmission of JEV. Flying foxes the Queensland Department of Primary Industries Animal are of particular interest as a potential reservoir or amplify- Ethics Committee (AEC; approval no. 33/08/02), University ing host of JEV as a result of their nomadic behavior, and of Queensland AEC (approval no. MICRO/PARA/463/02/), because of their increasing abundance in major urban areas and Queensland Health Forensic and Scientific Services AEC (QHFSS; approval no. VIR 1/03/32). In addition, the collection of wild flying foxes was approved by the * Address correspondence to Andrew F. van den Hurk, Virology, Queensland Parks and Wildlife Service (Scientific Purposes Queensland Health Forensic and Scientific Services, 39 Kessels Rd, Coopers Plains, Queensland 4108, Australia. E-mail: andrew_hurk@ permit no. WISP01419503). As JEV is considered an exotic health.qld.gov.au virus in Australia, all infection and transmission experiments 457 458 VAN DEN HURK AND OTHERS were undertaken at the Physical Containment Level 3 the commencement and cessation of feeding, samples of the insectary/animal house at QHFSS, Brisbane. blood/virus suspension were diluted 1:20 in growth media Animals. Pteropus alecto were collected from flying fox (GM; Opti-MEM [Invitrogen, Grand Island, NY] with 3% colonies located in the Brisbane suburbs of East Brisbane fetal bovine serum, antibiotics and fungizone) and stored at (27°28′S, 153°03′E) and Indooroopilly (27°30′S, 152°58′E), −70°C for later titration. which also contained the Grey-headed flying fox, Pteropus After 18 hr, mosquitoes were anesthetized with CO2 and poliocephalus and the Little red flying fox, Pteropus scapulatus . blood engorged mosquitoes were placed into 1 L contain- Flying foxes returning from overnight foraging were captured ers within an environmental growth cabinet (Sanyo Electric, using mist nets for a period of approximately 90 min before Gunma, Japan) maintained at 28°C, 70–75% RH, and 12:12 sunrise. Flying foxes were anaesthetized using the inhalation (L:D), and offered 10% sucrose as a nutrient source. Mosqui- anesthetic Isoflurane (Isoflurane Inhalation Anesthetic, Laser toes were held under such conditions for an extrinsic incuba- Animal Health, Salisbury, Australia) administered by a portable tion period of 13 d. To determine if the mosquitoes intended Ohmeda Isotec 3 vaporizer (BOC Health Care, West Yorkshire, for the infection of the flying foxes were indeed infected, a England) at a concentration of 5% for induction and 1.5% sample of ≤ 6 individual mosquitoes were removed on day 12 during maintenance with an oxygen flow rate of 1 L/min.17 Each post virus exposure and tested for the presence of virus. bat was assessed for age, weight, forearm length, and female Flying fox husbandry. On the afternoon before the infection bats for pregnancy status by abdomen palpitation. Pregnant experiments, flying foxes were moved from ARI to QHFSS females were excluded from the study and released. To aid and placed in individual cages in a room within the PC3 in future identification, each flying fox was injected with a animal house. The dimensions of the cages were 900 × 900 × microchip containing a unique identification number (LifeChip, 900 mm, and were of wire mesh construction (25 × 25 × 25 Destron Fearing, South St. Paul, MN). A 1–2 mL blood sample mm aperture), elevated 1 m above the ground, with 10–15 cm was obtained from the propatagial vein using a 3 mL syringe between cages. Seasonal fruits (including rockmelon, bananas, (Terumo [Philippines] Corporation, Laguna, Philippines), a mangoes, apples, etc.), supplemented by full cream milk 0.50 × 16 mm needle (Terumo [Philippines] Corporation), powder, were fed to the flying foxes between 4 and 6 pm daily; and lithium heparin (Heparin Injection BP, David Bull water was provided ad libatum via drip bottles. Artificial light Laboratories, Mulgrave North, Australia) as an anticoagulant. was provided for the flying foxes each day between the hours The serum fraction was tested for JEV-specific antibodies of usual sunrise and sunset. using a neutralization assay.18 As HeV circulates within wild Exposure of flying foxes to JEV. Individual flying foxes flying fox populations in Brisbane, 19 the serum samples also were restrained, placed in calico bags, and transferred from were tested for antibodies to this virus, using the HeV Serum the animal room to the insectary, located within the QHFSS Neutralization Test developed and performed at the CSIRO PC3 insectary/animal house. Before infection and transmission Australian Animal Health Laboratory, Geelong.20 procedures, flying foxes were anesthetized as described pre- Flying foxes were held at the Queensland Department of viously. Ten flying foxes were exposed to JEV via mosquito Primary Industries and Fisheries, Animal Research Institute bite, during two separate trials, each involving five flying foxes. (ARI), Yeerongpilly, for ≤ 14 d to allow for a second serum Fifteen donor mosquitoes (previously exposed to JEV), in 30 × sample to be obtained for the completion of baseline serology. 40 mm containers with gauze covering the open end, were Any animals showing evidence of current or previous infec- allowed to feed on the upper leg for ≤ 15 min. At the same time, tion with either JEV or HeV were excluded from the study. 0.5 mL blood samples were obtained from the propatagial vein Virus strain. All experiments used the JEV TS3306 strain and 50 μL diluted 1/10 in GM for virus detection. The serum isolated from Aedes vigilax collected from Badu Island in fraction of the remaining 450 μL was retained for serology. February 1998. The virus had been passaged once in C6/36 Animals were also weighed and rectal temperatures obtained ( Ae. albopictus salivary gland) cells and twice in porcine stable- during each flying fox/mosquito interaction. After recovering equine kidney (PS-EK) cells. The titer of virus stock was 10 8 from the anesthesia (usually within 5 min), flying foxes were

PS-EK TCID50 (tissue culture infectious dose)50 /mL. returned to their cages. To determine the virus titer of donor Mosquito strain. The mosquitoes used for all infection and mosquitoes at the time of exposure, mosquitoes were killed transmission studies were Culex annulirostris from colonies with CO 2 and blood engorged mosquitoes stored at −70°C to housed at the Australian Army Malaria Institute, Brisbane. await analysis. Five additional animals were exposed to JEV The Cx. annulirostris colony was established from adults by being inoculated subcutaneously in the upper thigh area 4.8 collected from the Boondall Wetlands, Brisbane in 1998. with 0.1 mL of 10 PS-EK TCID50 /mL of virus. Vector competence experiments had previously showed that Flying foxes were observed at least twice daily for clini- this strain of Cx. annulirostris had infection and transmission cal symptoms of encephalitis (including fever, reduced appe- rates of 100% and 83%, respectively, for the same strain of tite, ruffled fur, decreased grooming, increased vocalization, JEV used in the current study.6 ataxia, eye closure, muscular rigidity, etc.) caused by infection Infection of mosquitoes with JEV. The 2- to 4-day-old with JEV. mosquitoes were starved for 24 hr before feeding for 30 min To determine if infected bats could produce a viremia capa- to 1 hr on a virus suspension using the hanging drop method ble of infecting mosquitoes, batches of recipient mosquitoes of Goddard and others. 21 The virus suspension contained were allowed to feed on the upper leg, interfemoral mem- JEV stock diluted in heparinized rabbit blood and 1% brane, or the plagiopatagium of each flying fox on days 1–5, sucrose. Two to three milliliter (mL) of this suspension (first 7, 9, 14, 20/21, 23/28 post exposure for the mosquito-exposed heated to 37 ± 1.0°C) was dropped onto the gauze covering animals, and days 3–5 for the inoculated animals. Because of the open end of 700 mL plastic containers that housed the unforseen external circumstances independent of this study, mosquitoes. To determine the virus titer of the blood meal at the first series of infections were concluded on day 23 instead JAPANESE ENCEPHALITIS VIRUS INFECTION OF FLYING FOXES 459 of day 28 post exposure. During each mosquito/flying fox five times in phosphate buffered saline (PBS)-Tween. Sera interaction, blood samples, temperatures, and weights for each collected from the 15 flying foxes during the experiments were animal were obtained while the animal was anaesthetized. added at a 1/100 dilution to the wells and incubated for 1 hr Batches of recipient mosquitoes were maintained for 10 d as at 37°C before a second wash. The presence of JEV-specific described previously, before being killed with CO2 and frozen antibody was detected by the addition of ImmunoPure at −70°C to await virus assay. Protein A/G–peroxidase conjugate (Pierce, Rockford, IL) at At the conclusion of the experiments, flying foxes were a dilution of 1/160,000. The conjugate was incubated for 1 hr euthanized while under anesthesia with a lethal intravenous at 37°C. The strips were again washed before the addition injection of Pentobarbitone sodium (Lethabarb Euthenasia of KBlue substrate (Neogen, Lexington, KY) for 10 min.

Injection; Virbac [Australia] Pty. Ltd., Peakhurst, Australia). The reaction was stopped by the addition of 1N H2 SO4 . The carcasses were necropsied, during which the brain, heart, Absorbances were measured at 450 nm with a reference liver, lung, spleen, and kidneys removed and retained, before wavelength of 650 nm. autoclaving and disposal. Virus assay. The blood/virus mixture used for mosquito RESULTS infection was titrated as 10-fold dilutions in a 96-well microtiter plate containing confluent PS-EK cell monolayers. Plates were Virus titer of donor mosquitoes. After 13 d extrinsic incubated at 37°C with 5% CO 2 and checked daily for any incubation, 1–7 infected donor mosquitoes fed on each of cytopathic effect (CPE) for ≤ 7 d. The amount of virus ingested the flying foxes. The virus titer of donor mosquitoes ranged by individual mosquitoes was estimated based on an average 1.0 6.1 between 10 and 10 TCID50 per mosquito ( Table 1 ). blood meal volume of approximately 3 μL. 22 The heads and Clinical symptoms of JEV infection in the flying foxes. bodies of the representative sample of donor mosquitoes None of the flying foxes displayed any clinical symptoms of were homogenized separately in 1 mL of GM using a SPEX encephalitis consistent with infection by JEV. 8000 mixer/mill (Spex Industries, Edison, NJ). Viral RNA was Viremia. None of the flying foxes exposed to JEV by extracted using the QIAamp Viral RNA kit (Qiagen, Clifton mosquito bite developed a detectable viremia. Only one Hill, Australia) before being tested for JEV using a real time inoculated flying fox (microchip no. 3003) developed a viremia TaqMan reverse transcriptase-polymerase chain reaction on the fourth day post inoculation. However, this viremia was 23 procedure (RT-PCR). In this assay, a cycle threshold (C t ) of a low level, as evidenced by the high cycle threshold score of value of 40 corresponded to < 0.001 of one plaque-forming 36 cycles in the TaqMan RT-PCR and a lack of detectable viral unit (PFU) of virus, and was deemed to be not detected. To replication when inoculated onto PS-EK cells. determine viral titers of donor mosquitoes, whole individual Infection of recipient mosquitoes. A total of 2,464 recipient mosquitoes were homogenized in 1 mL of tissue culture medium mosquitoes in 305 pools were processed for detection of and filtered using a 0.2 μm filter, before being titrated as 10-fold JEV RNA. Of these mosquitoes, 2,191 had fed on the flying dilutions on confluent PS-EK cell monolayers. Recipient foxes that had been infected by mosquito bite and 273 on the mosquitoes were pooled in batches of ≤ 12 according to bat inoculated flying foxes. JEV RNA was detected in 7 pools of number and day post exposure. Pools of recipient mosquitoes recipient mosquitoes that had fed on 4 flying foxes, indicating were then homogenized in 1 mL of GM. Viral RNA was virus transmission ( Table 2 ). Recipient mosquitoes were extracted from the supernatant using the QIAamp Viral RNA infected on either Days 3, 4, or 5 post exposure and infection kit before being tested using the TaqMan RT-PCR. Infection rates for these pools ranged between 38.5 and 90.9 per 1,000 rates were calculated for the pools of recipient mosquitoes mosquitoes. using the PooledInfRate statistical software package.24 Antibody response to infection. Sixty percent (6/10) of To test for viremia, all blood samples collected from the first the flying foxes that were exposed to JEV via mosquito had five flying foxes exposed to JEV by mosquito bite, were ini- detectable IgG responses in the Protein A/G ELISA, indicating tially titrated as serial 10-fold dilutions on PS-EK monolayers a seroconversion to JEV (Figure 1). In the experiments using as described previously. When virus was not detected using the inoculation as the source of infection, 100% of the flying foxes cell culture system, the blood samples from this trial and from seroconverted to JEV ( Figure 2 ). the remaining flying foxes exposed to JEV by mosquito bite and by inoculation were screened using the TaqMan RT-PCR. Any positive blood samples in the RT-PCR were titrated as Table 1 10-fold dilutions on PS-EK cell monolayers as described Number and virus titer of donor mosquitoes used to infect flying above. foxes No. of infected donor Mean (range) titer Serology. To establish if seroconversion to JEV had Flying fox no.* mosquitoes (no. fed) of infected mosquitoes† occurred in flying foxes after JEV exposure, an enzyme- 3924 1 (4) 1.1 linked immunosorbent assay (ELISA) was developed using 9919 2 (6) 5.5 (4.9–6.1) peroxidase conjugated Protein A/G to monitor IgG response. 8309 2 (4) 4.8 (4.4–5.1) Porcine sera, which had previously been shown to contain 2258 2 (5) 3.1 (1.0–5.1)‡ JEV-specific antibodies, were used for assay optimization and 1459 6 (9) 4.1 (1.8–5.5) 25 3023 6 (6) 3.6 (1.0–5.9) as positive controls. Negative controls consisted of porcine 3466 4 (7) 1.8 (1.0–2.9) sera, which had previously tested negative for flavivirus 6594 3 (4) 3.4 (2.9–3.9) antibodies. The ELISA was performed using Maxisorp 3332 5 (7) 4.0 (1.5–5.1) strips (NUNC A/S, Roskilde, Denmark), which were coated 2355 7 (11) 4.2 (2.5–6.0) with inactivated JEV antigen in carbonate buffer (pH 9.6). 26 * Flying fox number refers to the microchip number. † Titer expressed as log10 PS-EK TCID50 per mosquito. After overnight incubation at 4°C, the strips were washed ‡ Threshold of detection was 1.0 log10 PS-EK TCID50 per mosquito. 460 VAN DEN HURK AND OTHERS

Table 2 Infection rates (IRs) of pools of infected recipient mosquitoes that were allowed to feed on Japanese encephalitis virus (JEV) exposed flying foxes* Source of Day post No. mosquitoes No. PCR positive pools IR per 1,000 Bat no. exposure exposure tested (no. pools tested) mosquitoes 1459 Mosquito 3 24 2 (3) 83.3 4 22 1 (3) 38.5 3466 Mosquito 4 26 2 (2) 90.9 3003 Inoculation 4 16 1 (4) 62.5 0752 Inoculation 5 16 1 (4) 62.5 * PCR = polymerase chain reaction.

DISCUSSION As shown by serologic studies, numerous animals are natu- rally infected with JEV,27,28 although humans and horses are Figure 2. IgG antibody response of flying foxes exposed to the only animals that develop severe and fatal disease, char- Japanese encephalitis virus (JEV) by inoculation, as detected by acterized by acute encephalitis. In some instances, infection in enzyme-linked immunosorbent assay (ELISA). pigs can result in abortion and stillbirth in sows, and reduced 29,30 sperm production in boars. In the current study, infection This cryptic viremia may be explained by the replication of with JEV did not result in overt clinical illness in any of the virus in skin or dendritic cells at the site of inoculation of the fly- flying foxes. These results support previous studies of both ing fox, with or without entering the bloodstream. In our experi- Microchiropteran and Megachiropteran bats, in which exper- ments, some recipient mosquitoes fed at the same site (i.e., the imental infection with JEV failed to produce any signs of upper leg) as the donor mosquitoes and may have subsequently 13–15 encephalitis, even if inoculated directly into the brain. imbibed virus liberated from the cells and tissues damaged by Common opinion considers that virus circulating in the the action of the probing mouthparts. Indeed, West Nile virus blood is essential to facilitate ingestion by a mosquito and that (WNV), a member of the JEV serologic group, has been shown a threshold level of viremia must be reached before infections to replicate in the skin cells of eastern grey squirrels (Sciurus 31 of vectors can occur. Therefore, infection of recipient mos- carolinensis) and mice at the site of inoculation, with viral loads quitoes after feeding on flying foxes without detectable vire- in the latter species increasing from an undetectable level on mia was unexpected. However, this is not a novel phenomenon, the first day post exposure to a titer of ≤ 10 5 PFU/g at day 7 as earlier experiments have shown that recipient mosquitoes post exposure. 34,35 Initial infection of skin tissue is possibly can be infected after feeding on horses and Agile wallabies caused by the deposition of virions extravascularly in tissue sur- ( Macropus agilis ) without detectable viremia that had previ- rounding capillaries and not directly into the vascular system ously been exposed to JEV and Murray Valley encephalitis while mosquitoes are probing and locating blood vessels.36,37 32,33 virus, respectively. This is a different concept to purported nonviremic trans- In our study, we used a sensitive real-time RT-PCR, which is mission (NVT) of arboviruses, where infection of recipient 23 able to detect viral RNA from 0.005 PFU of virus. Therefore, mosquitoes with WNV occurs when they feed simultaneously it should have been more likely to detect viral RNA than with infected donor mosquitoes and, importantly, before viral earlier methods of virus assay used in vertebrate studies. replication in the host. 38 When McGee and others 39 assessed Importantly, the results of this study emphasize the need to the effects of time and space on NVT, recipient mosquitoes use highly susceptible recipient arthropods to show transmis- could be infected ≤ 45 min after the donor mosquitoes had fed. sion when assessing the relative ability of vertebrate species to However, in our experiments, transmission to recipient mos- serve as amplifying hosts of arboviruses. quitoes occurred 3 to 5 d after exposure, providing adequate time for the virus to replicate at the site of inoculation. Further experiments are required to elucidate whether recipient mosquitoes can be infected while probing cells and tissues before cannulation of the blood vessel. Although infection rates in recipient mosquitoes were rel- atively low in this experiment, the large number of animals in P. alecto colonies, in which populations can reach tens of thousands, 40 may provide sufficient numbers of hosts to infect local mosquitoes. Indeed, large flying fox colonies are pres- ent on some Torres Strait islands where JEV activity has been observed, including Badu Island, the island that has yielded the majority of JEV isolates. Furthermore, analysis of mosquito host feeding patterns has revealed that Cx. annulirostris does feed on flying foxes on Badu Island (van den Hurk AF, Hall-Mendelin S, and Cheah WY, unpublished data). Only a small number of infected flying foxes would be Figure 1. IgG antibody response of flying foxes exposed to Japanese encephalitis virus (JEV) by infected mosquitoes, as detected required to facilitate spillover of JEV into local bird and/or by enzyme-linked immunosorbent assay (ELISA). pig populations via mosquito bite and thus initiate epizootic JAPANESE ENCEPHALITIS VIRUS INFECTION OF FLYING FOXES 461 activity. Importantly, because there is considerable movement 6. van den Hurk AF, Nisbet DJ, Hall RA, Kay BH, Mackenzie JS, of P. alecto between Cape York Peninsula, the Torres Strait, Ritchie SA, 2003. Vector competence of Australian mosquitoes and Papua New Guinea, it is conceivable that flying foxes (Diptera: Culicidae) for Japanese encephalitis virus. J Med Entomol 40: 82–90. may have introduced JEV into northern Australia from the 7. Buescher EL, Scherer WF, Rosenberg MZ, McClure HE, 1959. New Guinea landmass, as suggested by Mackenzie and oth- Immunologic studies of Japanese encephalitis virus in Japan. ers.2 Flying foxes can travel > 50 km in a single night, so that it III. Infection and antibody responses of birds. J Immunol 83: would take approximately 2 and 4 d for a flying fox to traverse 605–613. 8. Buescher EL, Scherer WF, McClure HE, Moyer JT, Rosenberg the distance between southern Papua New Guinea and Badu MZ, Yoshii M, Okada Y, 1959. Ecologic studies of Japanese 41 Island, or Cape York Peninsula, respectively. This is well encephalitis virus in Japan. IV. Avian infection. Am J Trop Med within the 3–5 d time period after JEV exposure that flying Hyg 8: 678–688. foxes were able to infect recipient mosquitoes in our experi- 9. Gresser I, Hardy JL, Hu SMK, Scherer WF, 1958. Factors influenc- ments. However, whereas flying foxes undoubtedly undertake ing transmission of Japanese B encephalitis virus by a colonized strain of Culex tritaeniorhynchus Giles, from infected pigs and foraging flights to different islands, such sporadic activity could chicks to susceptible pigs and birds. Am J Trop Med Hyg 7: not account for the synchronicity of the widespread appear- 365–373. ance of JEV on eastern Torres Strait islands. This phenomenon 10. Scherer WF, Moyer JT, Izumi T, Gresser I, McCown J, 1959. suggests an alternative mechanism of incursion, such as wind- Ecologic studies of Japanese encephalitis virus in Japan. VI. blown mosquitoes, may be more likely. 42 Swine infection. Am J Trop Med Hyg 8: 698–706. 11. Endy TP, Nisalak A, 2002. Japanese encephalitis virus: ecology and Received October 27, 2008. Accepted for publication May 28, 2009. epidemiology. Curr Top Microbiol Immunol 267: 11–48. 12. Mackenzie JS, Field HE, Guyatt KJ, 2003. Managing emerging dis- Acknowledgments: We thank Bruce Harrower for undertaking the eases borne by fruit bats (flying foxes), with particular refer- post mortems on the flying foxes; Amanda McLaughlin, Carol de Jong, ence to henipaviruses and Australian bat lyssavirus. J Appl and Janine Barrett for assistance with flying fox collection; and Sadet Microbiol 94: 59S–69S. Davis, Teck Chuan, Petrina Johnson, Donna Mackenzie, and Russell 13. Sulkin SE, Allen R, 1974. Virus infections in bats. Melnick JL, ed. Simmons for technical assistance. We also thank Les Hall for discus- Monographs in Virology , Vol. 8. Basel: S. Karger, 1–103. sions on flying fox movements in the Torres Strait and Laura Kramer, 14. Banerjee K, Ilkal MA, Bhat HR, Sreenivasan MA, 1979. Scott Ritchie, and Roy Hall for their comments on the manuscript. Experimental viraemia with Japanese encephalitis virus in cer- tain domestic and peridomestic vertebrates. Indian J Med Res Financial support: This study was funded by the Australian Health 70: 364–368. Minister’s Advisory Council (AHMAC) Priority Driven Research 15. Banerjee K, Ilkal MA, Deshmukh PK, 1984. Susceptibility of Program and Queensland Health. Cynopterus sphinx (frugivorus bat) and Suncus murinus (house Authors’ addresses: Andrew F. van den Hurk, Ina L. Smith, Judith A. shrew) to Japanese encephalitis virus. Indian J Med Res 79: Northill, Carmel T. Taylor, and Greg A. Smith, Virology, Queensland 8–12. Health Forensic and Scientific Services, 39 Kessels Rd., Coopers Plains, 16. Webb NJ, Tidemann CR, 1996. Mobility of Australian flying foxes, Queensland 4108, Australia, Tel: 617-3274-9135, Fax: 617-3000-9186, Pteropus spp. (Megachiroptera): evidence from genetic varia- E-mails: [email protected], [email protected] tion. Proc R Soc Lond B Biol Sci 263: 497–502. .au, [email protected], [email protected], 17. Jonsson NN, Johnston SD, Field H, de Jong C, Smith C, 2004. Field and [email protected]. Craig S. Smith and Hume E. Field, anaesthesia of three Australian species of flying fox. Vet Rec Department of Primary Industries and Fisheries, Animal Research 154: 664. Institute, 665 Fairfield Road, Yeerongpilly, Queensland 4105, Australia, 18. Beaty BJ, Calisher CH, Shope RE, 1995. Arboviruses. Lennette E-mails: [email protected] and [email protected] EH, Lennette DA, Lennette ET, eds. Diagnostic Procedures for .au. Cassie C. Jansen, formerly Australian Army Malaria Institute, cur- Viral, Rickettsial, and Chlamydial Infections . Washington, DC: rently Australian Biosecurity Cooperative Research Centre, University American Public Health Association, 189–212. of Queensland, St. Lucia, Queensland 4072, Australia, E-mail: cassie_ 19. Halpin K, Young PL, Field HE, Mackenzie JS, 2000. Isolation of [email protected]. John S. Mackenzie, Australian Biosecurity Hendra virus from pteropid bats: a natural reservoir of Hendra Cooperative Research Centre, Centre for International Health, Curtin virus. J Gen Virol 81: 1927–1932. University of Technology, GPO U1987, Perth, Western Australia, 6845, 20. Crameri G, Wang LF, Morrissy C, White J, Eaton BT, 2002. A rapid Australia, E-mail: [email protected]. immune plaque assay for the detection of Hendra and Nipah viruses and anti-virus antibodies. J Virol Methods 99: 41–51. 21. Goddard LB, Roth AE, Reisen WK, Scott TW, 2002. Vector com- REFERENCES petence of California mosquitoes for West Nile virus. Emerg Infect Dis 8: 1385–1391. 1. Mackenzie JS, Williams DT, Smith DW, 2007. Japanese encephali- 22. Turell MJ, Kay BH, 1998. Susceptibility of selected strains of tis virus: the geographic distribution, incidence, and spread of a Australian mosquitoes (Diptera: Culicidae) to Rift Valley fever virus with a propensity to emerge in new areas. E. Tabor, ed. virus. J Med Entomol 35: 132–135. Emerging Viruses in Human Populations . Amsterdam: Elsevier, 23. Pyke AT, Smith IL, van den Hurk AF, Northill JA, Chuan TF, 201–268. Westacott AJ, Smith GA, 2004. Detection of Australasian flavi- 2. Mackenzie JS, Johansen CA, Ritchie SA, van den Hurk AF, Hall virus encephalitic viruses using rapid fluorogenic TaqMan RA, 2002. Japanese encephalitis as an emerging virus: the emer- RT-PCR assays. J Virol Methods 117: 161–167. gence and spread of Japanese encephalitis virus in Australasia. 24. Biggerstaff BJ, 2003. PooledInfRate: a Microsoft Excel add-in to Curr Top Microbiol Immunol 267: 49–73. compute prevalence estimates from pooled samples. Fort 3. Hanna JN, Ritchie SA, Phillips DA, Shield J, Bailey MC, Mackenzie Collins, CO: Centers for Disease Control and Prevention. JS, Poidinger M, McCall BJ, Mills PJ, 1996. An outbreak of 25. Smith IL, Westacott AJ, Smith GA, 2005. Development and imple- Japanese encephalitis in the Torres Strait, Australia, 1995. Med mentation of a porcine IgM capture enzyme linked immuno- J Aust 165: 256–260. sorbent assay for the detection of Japanese encephalitis virus 4. Hanna JN, Ritchie SA, Phillips DA, Lee JM, Hills SL, van den antibodies in sentinel pigs. Arbovirus Res Aus 9: 352–356. Hurk AF, Pyke AT, Johansen CA, Mackenzie JS, 1999. Japanese 26. Pyke AT, Phillips DA, Chuan TF, Smith GA, 2004. Sucrose density encephalitis in north Queensland, Australia, 1998. Med J Aust gradient centrifugation and cross-flow filtration methods for 170: 533–536. the production of arbovirus antigens inactivated by binary eth- 5. Ritchie SA, Phillips D, Broom A, Mackenzie J, Poidinger M, van ylenimine. BMC Microbiol 4: 3–10. den Hurk A, 1997. Isolation of Japanese encephalitis virus from 27. Johnsen DO, Edelman R, Grossman RA, Muangman D, Pomsdhit Culex annulirostris in Australia. Am J Trop Med Hyg 56: 80–84. J, Gould DJ, 1974. Study of Japanese encephalitis virus in 462 VAN DEN HURK AND OTHERS

Chiangmai Valley, Thailand. V. Animal infections. Am J 35. Brown AN, Kent KA, Bennett CJ, Bernard KA, 2007. Tissue tropism Epidemiol 100: 57–68. and neuroinvasion of West Nile virus do not differ for two mouse 28. Oda K, Igarashi A, Keong CT, Hong CC, Vijayamalar B, Sinniah strains with different survival rates. Virology 368: 422–430. M, Hassan SS, Tanaka H, 1996. Cross-sectional serosurvey for 36. Turell MJ, Tammariello RF, Spielman A, 1995. Nonvascular deliv- Japanese encephalitis specific antibody from animal sera in ery of St. Louis encephalitis and Venezuelan equine encephali- Malaysia 1993. Southeast Asian J Trop Med Public Health 27: tis viruses by infected mosquitoes (Diptera: Culicidae) feeding 463–470. on a vertebrate host. J Med Entomol 32: 563–568. 29. Takashima I, Watanabe T, Ouchi N, Hashimoto N, 1988. Ecologic 37. Turell MJ, Spielman A, 1992. Nonvascular delivery of Rift Valley studies of Japanese encephalitis virus in Hokkaido: interepi- fever virus by infected mosquitoes. Am J Trop Med Hyg 47: demic outbreaks of swine abortion and evidence for the virus to 190–194. overwinter locally. Am J Trop Med Hyg 38: 420–427. 38. Higgs S, Schneider BS, Vanlandingham DL, Klingler KA, Gould 30. Burns KF, 1950. Congenital Japanese B encephalitis infection of EA, 2005. Nonviremic transmission of West Nile virus. Proc swine. Proc Soc Exp Biol Med 75: 621–625. Natl Acad Sci USA 102: 8871–8874. 31. Lord CC, Rutledge CR, Tabachnick WJ, 2006. Relationships 39. McGee CE, Schneider BS, Girard YA, Vanlandingham DL, Higgs between host viremia and vector susceptibility for arboviruses. S, 2007. Nonviremic transmission of West Nile virus: evaluation J Med Entomol 43: 623–630. of the effects of space, time, and mosquito species. Am J Trop 32. Gould DJ, Byrne RJ, Hayes DE, 1964. Experimental infection of Med Hyg 76: 424–430. horses with Japanese encephalitis virus by mosquito bite. Am J 40. Markus N, Hall L, 2004. Foraging behaviour of the black flying Trop Med Hyg 13: 742–746. fox ( Pteropus alecto) in the urban landscape of Brisbane, 33. Kay BH, Young PL, Hall RA, Fanning ID, 1985. Experimental Queensland. Wildl Res 31: 345–355. infection with Murray Valley encephalitis virus. Pigs, cattle, 41. Breed AC, Smith CS, Epstein JH, 2006. Winged wanderers: long sheep, dogs, rabbits, macropods and chickens. Aust J Exp Biol distance movements of flying foxes. MacDonald DW, ed. The Med Sci 63: 109–126. Encyclopedia of Mammals. Oxford: Oxford University Press, 34. Gómez A, Kramer LD, Dupuis AP, Kilpatrick AM, Davis LJ, Jones 474–475. MJ, Daszak P, Aguirre AA, 2008. Experimental infection of 42. Ritchie SA, Rochester W, 2001. Wind-blown mosquitoes and intro- eastern gray squirrels ( Sciurus carolinensis) with West Nile duction of Japanese encephalitis into Australia. Emerg Infect virus. Am J Trop Med Hyg 79: 447–451. Dis 7: 900–903.