VECTOR/PATHOGEN/HOST INTERACTION, TRANSMISSION Japanese Encephalitis in Kerala, South India: Can Mansonia (Diptera: Culicidae) Play a Supplemental Role in Transmission? N. ARUNACHALAM,1, 2 P. PHILIP SAMUEL,1 J. HIRIYAN,1 V. THENMOZHI,1 AND A. GAJANANA3 J. Med. Entomol. 41(3): 456Ð461 (2004) KEY WORDS Japanese encephalitis, secondary vector, Mansonia, Culex, India JAPANESE ENCEPHALITIS (JE) is a mosquito-borne ßavivirus responsible for thousands of clinical cases, mostly children, in Eastern and Southeastern Asia each year (Vaughn and Hoke 1992). JE was clinically diagnosed for the Þrst time in India in 1955 at Vellore in North Arcot District in Tamil Nadu (Webb and Pereira 1956). Several JE outbreaks of varying intensity were reported from different parts of India (Rodrigues 1984). Currently, JE remains a major public health problem in India, where major outbreaks occur (Arunachalam et al. 2002). JE is a zoonosis, affecting many species of animals and birds of which pigs and ardeid birds are known to be important maintenance and amplifying hosts for the virus. Humans are a “dead end” host, playing no role in the maintenance cycle. The mosquito vectors breed in paddy Þelds, irrigation channels, rainwater pools, and seepages (Reuben 1987). The Þrst JE outbreak was reported in Kerala in 1996, when JE virus was isolated from Culex tritaeniorhynchus Giles, and Mansonia indiana and Ma. uniformis (Theobald) were found naturally infected (Dhanda et al. 1997). Previously, JE virus was isolated from Ma. uniformis in Malaysia (Macdonald et al. 1967) and Sri Lanka (Peiris et al.1992), and members of the subgenera Mansonioides and Coquillettidia have been implicated in the epidemiology of several arboviruses in Africa (Theiler and Downs 1973). Isolation of any arbovirus from naturally caught mosquitoes is not sufÞcient evidence to implicate a species as a biological vector, because infected mosquitoes might be ecologically insigniÞcant or dead-end hosts (Scherer et al. 1971). To in1 Centre for Research in Medical Entomology (Indian Council of Medical Research), 4, Sarojini Street, Chinna Chokkikulam, Madurai 625002, Tamil Nadu, India. 2 E-mail: crmeicmr@satyam.net.in. 3 146 11th Main Road, Hanumantha Nagar, Bangalore 560 019, India. criminate a mosquito species as a vector, it is necessary to demonstrate that the species acquire the infection in nature, that it is capable of transmitting the infection by bite, that it feeds on humans, and that it is abundant (Scherer et al. 1971). Some vectors like Cx. gelidus Giles and Cx. fuscocephala Theobald from which isolations of JE virus have been reported from India are highly zoophagic and poorly anthropophagic and therefore may have an important role in amplifying JE virus but not in transmitting virus to humans (Reuben et al.1992). Entomological studies were carried out in 1999 and 2000 in Kuttanadu, Kerala, to determine the seasonal abundance and JE virus infection rates in Cx. tritaeniorhynchus. Suspected Mansonia vectors were studied coincidentally to understand their role in the transmission of JE virus. Materials and Methods The villages of Kavalam, Molagan Thurthy, Neelamperoor, Nehru Trophy Ward, Pulimcunnu, and Veliyanadu in the Kuttanadu region of Kerala state near Lake Vembanad were selected as index villages for this study (Fig. 1). At least one JE case occurred in each village during the 1996 and 1997 epidemic. Kuttanadu is a warm humid region with fairly uniform temperature, which ranges from 21 to 35⬚C throughout the year (Fig. 2). Most villagers work as agricultural laborers. Cattle, goat, pigs, dogs, fowl, and ducks are the most common domestic animals (Alappuzha district Livestock Census). The area receives most of its rainfall (83%) from JuneÐAugust under the inßuence of southwest monsoons and less rainfall from OctoberÐDecember under the inßuence of the northeast monsoons. The annual average rainfall is ⬃300 cm. 0022-2585/04/0456Ð0461$04.00/0 䉷 2004 Entomological Society of America Downloaded from https://academic.oup.com/jme/article/41/3/456/917545 by guest on 13 February 2022 ABSTRACT A 2-yr entomological study was carried out in Kerala, south India, to identify the mosquito vectors of Japanese encephalitis (JE) virus and to determine their seasonal abundance and infection. In total, 150,454 mosquitoes belonging to Þve genera and 18 species were collected from vegetation surrounding cattle sheds and pigsties in villages at dusk. Culex tritaeniorhynchus Giles (66.7%) was the most abundant species, with increases in numbers associated with rice cultivation. JE virus isolations were made from Cx. tritaeniorhynchus and Mansonia indiana Edwards. Based on high abundance and frequent JE virus infection, Cx. tritaeniorhynchus seems to be the most important vector, whereas Ma. indiana is probably a secondary vector. May 2004 ARUNACHALAM ET AL.: JE IN KERALA, SOUTH INDIA 457 Each study village was sampled at monthly intervals during 1999 and 2000. Mosquitoes resting on vegetation and bushes around cattle sheds and pigsties were collected for 1 h after dusk by oral aspirator and transported to the laboratory for identiÞcation and enumeration. Mosquito (only females) abundance was calculated as number collected per man-hour. Male mosquitoes also were collected resting in and around cattle sheds and pigsties. Wild-caught mosquitoes were counted into pools of 25Ð50 and were stored at ⫺80⬚C until processed for JE virus detection and isolation. Two systems were used (Gajanana et al. 1995). (1) Antigen capture ELISA: monoclonal antibody 6B4A-10 (reactive against all viruses in JE/WN/SLE/MVE complex) was used as capture antibody and monoclonal antibody peroxidase conjugate SLE MAB 6B6C-1 (reactive against all ßaviviruses) as detector antibodies (Supplied by Dr. T. F. Sai, Centers for Disease Control and Prevention, Fort Collins, CO). (2) Insect bioassay: Toxorhynchites splendens Wiedemann larvae were inoculated intracerebrally, incubated for 7Ð10 d at 29⬚C, and tested by indirect immunoßuorescence assay (IFA) on head squash preparations (Toxo-IFA). Smears were Þrst screened using an anti-JE virus immune serum raised in rabbits that was broadly reactive against ßaviviruses and detected by ßuorescein isothiocyanate (FITC) conjugated anti-rabbit immunoglobulin (Dakoppats, Glostrup, Denmark). For conÞrmation, duplicate smears were tested with JE virusÐspeciÞc monoclonal antibody, MAB 112 (supplied by Dr. Kimura Kuroda, Tokyo Metropolitan Institute of Neurosciences, Tokyo, Japan), and detected by FITC-conjugated antimouse immunoglobulin (Dakoppats). Results In total, 150,454 female mosquitoes representing 6 anopheline and 12 culicine species were collected. Cx. tritaeniorhynchus was the most abundant species, comprising 66.9% of the total collected. This was followed in decreasing order by Cx. gelidus (11.1%), Ma. uniformis (9.6%), Ma. indiana (8.2%), and Ma. annulifera (Theobald) (3.0%); the remaining species comprised ⬍2% of the mosquitoes collected (Table 1). The abundance of Cx. tritaeniorhynchus was lowest in JuneÐAugust, increased in September, and reached a maximum in DecemberÐMarch (Fig. 3). The increase corresponded with the period of rice cultivation. Monthly abundance of Cx. tritaeniorhynchus was Downloaded from https://academic.oup.com/jme/article/41/3/456/917545 by guest on 13 February 2022 Fig. 1. Map of Alappuzha district showing the study villages in Kuttanadu region. 458 JOURNAL OF MEDICAL ENTOMOLOGY Vol. 41, no. 3 negatively correlated with rainfall (r ⫽ ⫺0.61, df ⫽ 23, P ⬍ 0.05) and was not correlated with temperature or humidity (P ⬎ 0.05). Culex gelidus was least abundant during the monsoon months, but unlike Cx. tritaeniorhynchus, it did not show major seasonal ßuctuations (Fig. 3). Mansonia species (Ma. annulifera, Ma. indiana, and Ma. uniformis) were collected throughout the year. Ma. uniformis was more abundant than the other two species (Fig. 4). Although the abundance of the Mansonioides was not correlated signiÞcantly with any meTable 1. Species composition of mosquitoes collected in Kuttanadu, Kerala Species No. collected Ae. aegypti An. barbirostris An. jamesii An. pallidus An. peditaeniatus An. subpictus An. tesellatus Ar. subalbatus Cx. fuscanus Cx. fuscocephala Cx. gelidus Cx. infula Cx. quinquefasciatus Cx. tritaeniorhynchus Cx. vishnui Ma. annulifera Ma. indiana Ma. uniformis Total 2 602 157 85 133 106 2 362 2 1 16,658 2 333 100,611 3 4,530 12,362 14,503 150,454 % ⬍1 ⬍1 ⬍1 ⬍1 ⬍1 ⬍1 ⬍1 ⬍1 ⬍1 ⬍1 11.1 ⬍1 ⬍1 66.7 ⬍1 3.0 8.2 9.6 teorological parameters, Ma. uniformis was most abundant during the monsoon months (Fig. 4). JE virus was isolated from Cx. tritaeniorhynchus and Ma. indiana. Overall, 146,560 mosquitoes were tested for JE virus in 3,374 pools, of which 64 pools of Cx. tritaeniorhynchus, 10 pools of Cx. gelidus, 3 pools of Ma. annulifera, 12 pools of Ma. indiana, and 5 pools of Ma. uniformis tested positive by ELISA. JE virus infection was conÞrmed by Toxo-IFA only from Cx. tritaeniorhynchus and Ma. indiana. Virus infection in Mansonia was observed in May, July, and August in 1999 and May, August, and October in 2000. Infection in Cx. tritaeniorhynchus was observed during JanuaryÐApril and November in 1999 and JanuaryÐMarch and SeptemberÐDecember in 2000. JE virus also was isolated from male mosquitoes collected resting in and around cattle sheds and pigsties. Of the 4 pools of male Cx. gelidus, 33 of Cx. tritaeniorhynchus, 25 of Ma. indiana, and 35 of Ma. uniformis tested, 3 male poolsÑ2 Ma. indiana and 1 Ma. uniformisÑwere found positive for ßavivirus antigen. Of the three ßavivirus-positive pools, one (of Ma. indiana) was conÞrmed to be positive for JE virus by Toxo-IFA, indicating vertical transmission by Ma. indiana. Discussion The seasonality of JE virus transmission depends on various factors, among which the relative abundance of the vector species is most important (Pant 1979). Our entomologic assessment indicated that Cx. tritaeniorhynchus was the primary vector based on relative Downloaded from https://academic.oup.com/jme/article/41/3/456/917545 by guest on 13 February 2022 Fig. 2. Meteorological data recorded during the study period. May 2004 ARUNACHALAM ET AL.: JE IN KERALA, SOUTH INDIA 459 abundance, widespread distribution, and frequent virus infection. Vector competence of Cx. tritaeniorhynchus has been demonstrated in laboratory studies (Carey et al.1969, Mourya et al.1991). Cx. tritaeniorhynchus accounted for 67% of mosquitoes collected, was the most abundant Culex during the transmission season, and was the most frequently infected species during our study. If vector abundance and JE virus infection rates were indicators of potential spillover of JE virus to humans, then January and April would be the months of greatest risk. Abundance of Cx. tritaeniorhynchus was very high (up to 397 females/manhour) during the rice cultivation season (JanuaryÐ April), which also was the main transmission season Fig. 4. Monthly rainfall and abundance of Mansonia species in Kuttanadu (1999Ð2000). Downloaded from https://academic.oup.com/jme/article/41/3/456/917545 by guest on 13 February 2022 Fig. 3. Abundance of Cx. tritaeniorhynchus and Cx. gelidus and total rainfall per month in Kuttanadu (1999Ð2000). 460 JOURNAL OF MEDICAL ENTOMOLOGY transmission in Japan and Taiwan (Wu et al.1999). The removal of the pigs from within the Badu Island community also has reduced the potential contact between viremic pigs and vectors within the community (Van den Hurk et al. 2001). The relocation of domestic pigs could be adopted as a control strategy in Kerala to prevent/reduce JE transmission to humans. Acknowledgments The authors thank SEARO/WHO New Delhi for Þnancial support. This publication is an outcome of WHO project SN 1094. We thank A. Veerapathiran, V. Kodangi Alagan, and V. Rajamannar of Vector Biology and training division of Centre for Research in Medical Entomology for excellent technical assistance. We appreciate the excellent help rendered by A. Venkatesh (CRME) in preparation of this manuscript, particularly in DTP work. References Cited Arunachalam, N., P. Philip Samuel, J. Hiriyan, V. Thenmozhi, A. Balasubramanian, A. Gajanana, and K. Satyanarayana. 2002. Vertical transmission of Japanese encephalitis virus in Mansonia species, in an epidemicprone area of southern India. Ann. Trop. Med. Parasitol. 96: 419 Ð 420. Burton, G. J. 1959. Studies on the bionomics of mosquito vectors, which transmit Þlariasis in India. I. Attachment of Mansonia annulifera and Mansonia uniformis larvae to host plants occurring in Pistia tanks in Kerala, south India. Ind. J. Malariol. 13: 75. Carey, D. E., R. Reuben, and R. M. Myers. 1969. Japanese encephalitis studies in Vellore, South India. Part V. Experimental infection and transmission, Ind. J. Med. Res. 37: 282. Dhanda, V., D. T. Mourya, A. C. Mishra, M. A. Ilkal, U. Pant, P. George Jacob, and H. R. Bhat. 1989. Japanese encephalitis virus infection in mosquitoes reared from Þeldcollected immatures and in wild-caught males. Am. J. Trop. Med. Hyg. 41: 732Ð736. Dhanda, V., V. Thenmozhi, N. P. Kumar, J. Hiriyan, N. Arunachalam, A. Balasubramanian, A. Ilango, and A. Gajanana. 1997. Virus isolation from wild-caught mosquitoes during a Japanese encephalitis outbreak in Kerala in 1996. Ind. J. Med. Res. 106: 4 Ð 6. Gajanana, A., R. Rajendran, V. Thenmozhi, P. Philip Samuel, T. F. Tsai, and R. Reuben. 1995. Comparative evaluation of bioassay and ELISA for detection of Japanese encephalitis virus in Þeld collected mosquitoes. Southeast Asian J. Trop. Med. Publ. Hlth. 26: 91Ð97. Macdonald, W. W., C.E.G. Smith, P. S. Dawson, A. Ganapathipillai, and S. Mahadevan. 1967. Arbovirus infections in Sarawak: further observations on mosquitoes. J. Med. Entomol. 4: 146 Ð157. Mourya, D. T., A. C. Mishra, and R. S. Soman. 1991. 1991. Transmission of Japanese Encephalitis virus in Culex pseudovishnui and Culex tritaeniorhynchus mosquitoes India. Ind. J. Med. Res. 93: 250 Ð252. Pant, C. P. 1979. Vectors of Japanese encephalitis and their bionomics. WHO/VBC/79.732: 1Ð18. Peiris, J.S.M., F. P. Amarasinghe, P. H. Amarasinghe, C. B. Ratnayaka, S.H.P. Karunatne, and T. F. Tsai. 1992. Japanese encephalitis in Sri Lanka. A study of an epidemic: vector incrimination, porcine infection and human disease. Am. J. Trop. Med. Hyg. 86: 307Ð313. Downloaded from https://academic.oup.com/jme/article/41/3/456/917545 by guest on 13 February 2022 for JE in this area. Ma. indiana and Ma. uniformis abundance (maximum 62 females/man-hour) peaked during the monsoon months (JulyÐNovember), when the paddy Þelds were ßooded, and few Cx. tritaeniorhynchus could be found (collections falling as low as 2.4 females/man-hour). The monsoon rains permit large stands of hydrophytes (such as Pistia, Salvinia, and Eichhornia) to develop, and these plants are essential for the larval development of Mansonia (Burton 1959). Most of our study area was water-logged at an elevation of 0.5Ð2 m below mean sea level; this created swampy areas supporting dense stands of aquatic vegetation, which permited the breeding of Mansonia. Vertical transmission of JE virus was documented by virus isolation from male Ma. indiana (females as well as males were infected). Vertical transmission in which the virus is transmitted from an infected female mosquito to her eggs as they pass through the genital tract is known to support the persistence of some arboviruses in nature (Rosen 1988). There is Þeld evidence of this vertical transmission of JE virus in south India. Dhanda et al. (1989) detected JE virus in wild-caught males of Cx. tritaeniorhynchus, indicating the occurrence of vertical transmission in Culex. Thenmozhi et al. (2001) detected JE virus infection in Þeld-collected immature stages of Cx. tritaeniorhynchus. JE virus could use mosquitoes as reservoir hosts for local survival of the virus during adverse conditions of hot seasons, when vector abundance is low and nonimmune pigs are few in number (Thenmozhi et al.2001). The infected Mansonia mosquitoes were collected toward the end of the monsoon, at a time when few adult Cx. tritaeniorhynchus can be caught. If the larval stages of local Mansonia are sometimes infected with JE virus, the relatively slow development of Mansonia (Srinivasan and Viswam 1986) may help the virus to survive a period when its main vector is scarce. Mansonia is probably a secondary vector, with Cx. tritaeniorhynchus as the primary vector. Blood meals of Cx. tritaeniorhynchus and Mansonia were analyzed against antisera of six hosts, cattle, pigs, ducks, goats, fowl, and humans, to Þnd the preferred hosts of vectors in this region. Cx. tritaeniorhynchus fed predominantly on cattle (76%), with pigs accounting for 5% of blood meals identiÞed. Pigs constituted 6% of Ma. indiana blood meals tested (Centre for Research in Medical Entomology [CRME], unpublished data). A serological survey revealed a high prevalence of (69%) of JE antibodies in pigs (CRME, unpublished data). Pigs were maintained in small groups in backyard pens surrounded by paddy Þelds. Consequently, the primary vector Cx. tritaeniorhynchus and the potential bridge vector Ma. indiana were in close proximity to the amplifying pig hosts and humans, favoring the spillover transmission of JE virus to humans. Similarly, intense transmission of JE on Badu Island, Australia, was attributed to a large domestic pig population housed in small backyard pens throughout the community (Van den Hurk et al. 2001). The relocation of pigs to specialized farms separate from human habitation was useful in reducing pig-mosquito-human JE Vol. 41, no. 3 May 2004 ARUNACHALAM ET AL.: JE IN KERALA, SOUTH INDIA Thenmozhi, V., R. Rajendran, P. Philip Samuel, J. Hiriyan, K. Ayanar, A. Balasubramanian, and A. Gajanana. 2001. Natural vertical transmission of Japanese encephalitis virus in south Indian mosquitoes. Trop. Biomed. 18: 19 Ð27. Van den Hurk, A. F., D. J. Nisbet, C. A. Johansen, P. N. Foley, S. A. Ritchie, and J. S. Mackenzie. 2001. Japanese encephalitis on Badu Island, Australia: the Þrst isolation of Japanese encephalitis virus from Culex gelidus in the Australasian region and the role of mosquito host-feeding patterns in virus transmission cycles. Trans. Roy. Soc. Trop. Med. Hyg. 95: 595Ð 600. Vaughn, D. W., and C. H. Hoke. 1992. The epidemiology of Japanese encephalitis: prospects of prevention. Epidemiol. Rev. 14: 197Ð221. Webb, J.K.G., and S. M. Pereira. 1956. Clinical diagnosis of an arthropod-borne type of encephalitis in North Arcot district, Madras state, India. Ind. J. Med. Res. 10: 583Ð588. Wu, Y. C., Y. S. Huang, L. J. Chien, T. L. Lin, Y. Y. Yueh, W. L. Tseng, K. J. Chang, and G. R. Wang. 1999. The epidemiology of Japanese encephalitis on Taiwan during 1966 Ð 1997. Am. J. Trop. Med. Hyg. 61: 78 Ð 84. Received 23 May 2003; accepted 7 November 2003. Downloaded from https://academic.oup.com/jme/article/41/3/456/917545 by guest on 13 February 2022 Reuben, R. 1987. The epidemiology of Japanese encephalitis in Tamil Nadu, pp. 135Ð142. In Proceedings of the Symposium of Alternatives to Synthetic Insecticides, Madurai, India. Reuben, R., V. Thenmozhi, P. P. Samuel, A. Gajanana, and T. R. Mani. 1992. Mosquito blood feeding patterns as a factor in the epidemiology of Japanese encephalitis in southern India. Am. J. Trop. Med. Hyg. 46: 654 Ð 663. Rodrigues, F. M. 1984. Epidemiology of Japanese encephalitis in India: a brief overview, pp. 1Ð9. In Proceedings of the National Conference of Japanese Encephalitis, 1982, New Delhi, India. Rosen, L. 1988. Further observations on the mechanism of vertical transmission of ßaviviruses by Aedes mosquitoes. Am. J. Trop. Med. Hyg. 39: 123Ð126. Scherer, W. F., R. W. Dickerman, A. Diaz-Najera, B. A. Ward, M. H. Miller, and P. A. Schaffer. 1971. Ecological studies of Venezuelan encephalitis virus in South Eastern Mexico III Infection of mosquitoes. Am. J. Trop. Med. Hyg. 20: 969 Ð979. Srinivasan, R., and K. Viswam. 1986. Laboratory studies on the biology of Mansonia annulifera Theobald 1901. (Diptera:Culicidae) Ind. J. Med. Res. 83: 384 Ð386. Theiler, M., and W. G. Downs. 1973. The arthropod-borne viruses of vertebrates. Yale University Press, New Haven, CT. 461