SHORT COMMUNICATION Natural Vertical Transmission of Ndumu Virus in Culex pipiens (Diptera: Culicidae) Mosquitoes Collected as Larvae

1,2 3 4 1 JOEL LUTOMIAH, JULIETTE ONGUS, KENNETH J. LINTHICUM, AND ROSEMARY SANG

J. Med. Entomol. 51(5): 1091Ð1095 (2014); DOI: http://dx.doi.org/10.1603/ME14064 ABSTRACT Ndumu virus (NDUV) is a member of the family Togaviridae and genus Alphavirus.In Downloaded from https://academic.oup.com/jme/article/51/5/1091/884145 by guest on 23 September 2021 Kenya, the virus has been isolated from a range of mosquito species but has not been associated with human or animal morbidity. Little is know about the transmission dynamics or vertebrate reservoirs of this virus. NDUV was isolated from two pools of female Culex pipiens mosquitoes, IJR37 (n ϭ 18) and IJR73 (n ϭ 3), which were collected as larvae on 15 April 2013 from two dambos near the village of Marey, Ijara District, Garissa County, Kenya, and reared to adults and identiÞed to species. These results represent the Þrst Þeld evidence of vertical transmission of NDUV among mosquitoes.

KEY WORDS Culex pipiens, Ijara, Ndumu virus, vertical transmission

Ndumu virus (NDUV) is a member of the family habitats. There is little evidence to suggest that ver- Torgaviridae and genus Alphavirus, which comprises tical transmission of alphaviruses does occur (Tesh Ͼ31 viruses (Forrester et al. 2012). NDUV was Þrst 1984). Therefore, the Þnding of vertical transmission isolated in South Africa from Mansonia uniformis here provides important information regarding the (Theobald) in 1959 (Kokernot et al. 1961) and has natural maintenance of NDUV. since been described as widespread in most of Africa. It was recently documented in Kenya for the Þrst time Materials and Methods circulating among mosquitoes (Crabtree et al. 2009; Ochieng et al. 2013). Study Sites. Mosquitoes were sampled as larvae In general, infection with alphaviruses such as chi- from dambos in the villages of Marey and Wakab- kungunya, Ross River, Mayaro, and Sindbis causes Harey sublocations in Sangailu Division, Ijara District, clinical manifestations characterized by fever, rash, Garissa County, Kenya (Fig. 1), during the long rainy and arthralgia, although seroconversion in the ab- season of AprilÐMay and short rainy season of No- sence of clinical disease is uncommon with all alpha- vemberÐDecember in 2013. This site was selected viruses (Schmaljohn and McClain 1996). Although based on prior RVFV activity during the 2006Ð2007 mice experimentally infected with NDUV do not sur- RVF epizootic in Kenya. vive the infection (Kokernot et al. 1961), NDUV has Larval Sampling, Rearing, and Identification. Lar- not been associated with human or animal morbidity. val sampling was conducted using standard larval dip- This is because of lack of affordable diagnostic kits to pers (350 ml) (BioQuip, Inc., Compton, CA), from the test febrile patients for infection with this virus and the beginning to the end of the rainy season. Larvae col- lack of studies looking for evidence that NDUV causes lected from the dambos were transferred, in well la- diseases in humans. beled (date of collection and number collected) There are no studies that have been conducted Whirl-Pak bags (Universal Medical, Norwood, MA), speciÞcally to determine whether the virus can be to a Þeld laboratory where they were reared to adults maintained by vertical transmission in mosquito vec- in 4-liter capacity plastic cages. The emerging adults tors. Even this study did not target NDUV in particular were immobilized using 99.5% triethylamine (Sigma- but was primarily designed to explore vertical trans- Aldrich, St. Louis, MO) and identiÞed to species using mission of Rift Valley fever virus (RVFV), and other the keys of Edwards (1941), Gillies and Coetzee arboivruses occurring in mosquito species in natural (1987), and Jupp et al. (1976) and pooled up to 25 mosquitoes per pool. Pooled mosquitoes were pre- This paper was published with the permission from the Director, served in liquid nitrogen for transportation to the Kenya Medical Research Institute, KEMRI. KEMRI laboratories in Nairobi for virus isolation at- 1 Centre for Virus Research, Kenya Medical Research Institute, PO Box 54628, Nairobi, Kenya. tempts. 2 Corresponding author, e-mail: [email protected]. Sample Preparation and Virus Isolation in Cell Cul- 3 Department of Human Pathology and Laboratory Medicine, Jomo ture. Mosquito pools were homogenized using cop- Kenyatta University of Agriculture and Technology, PO Box 62000, per beads in 1 ml of homogenizing media comprising 00200-Nairobi, Kenya. 4 USDA/ARS Center for Medical, Agricultural and Veterinary En- Minimum Essential Medium Eagle (MEM), with tomology, 1600/1700 SW 23rd Dr., Gainesville, FL 32608. EarleÕs salts and reduced NaHCO3 (Sigma-Aldrich,

0022-2585/14/1091Ð1095$04.00/0 ᭧ 2014 Entomological Society of America 1092 JOURNAL OF MEDICAL ENTOMOLOGY Vol. 51, no. 5 Downloaded from https://academic.oup.com/jme/article/51/5/1091/884145 by guest on 23 September 2021

Fig. 1. Map of Kenya showing the study sites in Marey and Wakab-Harey sublocations in Sangailu Division, Ijara District, Garissa County, where larval sampling was conducted.

St. Louis, MO) supplemented with 15% heat-inac- (Invitrogen, Carlsbad, CA) using random hexamers tivated fetal bovine serum (FBS; Sigma-Aldrich), 2% followed by RT-PCR using AmpliTag Gold PCR Mas- L-glutamine (Sigma-Aldrich), and 2% antibioticÐan- ter Mix (Applied Biosystems). The cDNA was tested timycotic solution (Sigma-Aldrich) with 10,000 U using a panel of genera (alphavirus, bunyavirus, and penicillin, 10 mg streptomycin, and 25 ␮g ampho- flavivirus) and speciÞc primers for RVFV, NDUV, tericin B per milliliter. The homogenates were clar- babanki (BBKV), and sindbis (SINV) virus (Table 1). iÞed by centrifugation at 12,000 rpm (Eppendorf A positive control cDNA and a no-template negative Њ centrifuge 5417R) for 10 min at 4 C and the super- control were included during the setting up of all PCR natants transferred into 1.5-ml cryogenic vials. Each mos- reactions. AmpliÞcation products were resolved in ␮ quito pool supernatant (50 l) was inoculated per well 1.5% agarose gel in tris-borate-EDTA buffer stained onto conßuent monolayers of Vero cells (CCL81) in with ethidium bromide. 24-well plates grown in MEM, which was supplemented Sequencing. The ampliÞed PCR products were with 10% FBS. The inoculated cultures were incubated puriÞed using ExoSAP-IT (Affymetrix, Inc.). The pu- for 45 min to allow for virus adsorption, and 1 ml of maintenance media comprising MEM supplemented riÞed DNA fragments were sequenced using the Big- with 2% FBS was added to each well. The cultures were Dye 3.1 kit (PE Applied Biosystems, Foster City, CA) Њ and analyzed using ABI 3500xL Genetic analyzer (Ap- incubated at 37 Cin5%CO2 and monitored daily, through day 10, for cytopathic effects (CPE) as an in- plied Biosystems, Foster City, CA). The sequences dication of virus infection. were compared with available sequences of other Total RNA Isolation and Virus Identification by members of the alphavirus genus in the GenBank Reverse Transcription PCR. Total RNA was isolated database (http://www.ncbi.nlm.nih.gov/genbank/) from the supernatant of each culture exhibiting CPE using Basic Local Alignment Search Tool (BLAST, by Trizol-LS-Chloroform method (Chomczynski and http://blast.ncbi.nlm.nih.gov/Blast.cgi) to conÞrm the Sacchi 1987). Extracted RNA was reverse transcribed identity of the virus isolates. The sequences were to cDNA using SuperScriptIII reverse transcriptase aligned using ClustalW (Larkin et al. 2007). September 2014 LUTOMIAH ET AL.: VERTICAL TRANSMISSION OF NDUMU VIRUS IN Culex MOSQUITOES 1093

Results In total, 4,683 adult mosquitoes emerged from larvae (MareyÑ2,565; Wakab-HareyÑ2,118) and were identiÞed into three genera, representing 14 species, and pooled. Overall, Culex pipiens was the most pre- dominant and constituted 59.7% (n ϭ 2,797) of the total collection and 144 pools, followed by Culex uni- vittatus (21.8%; n ϭ 1,022; Table 2). Two pools (IJR37 and IJR73) out of 325 inoculated in Vero cell cultures caused CPE. The Vero cells in- oculated with IJR37 (n ϭ 18) and IJR73 (n ϭ 3), which

were both female Cx. pipiens, developed CPE on day Downloaded from https://academic.oup.com/jme/article/51/5/1091/884145 by guest on 23 September 2021 777Ð798 Kuno et al. (1996) 1309Ð1327 3 and 2 post infection (pi), respectively. Analysis of the two positive samples by RT-PCR conÞrmed the virus agent to be NDUV. IJR37 and IJR73 were collected on 15 April 2013, from dambo 9 (01Њ 14Ј46.3Љ S, 040Њ ) 86Ð114 Kuno et al. (1996)

Ј 39Ј31.6Љ E) and dambo 2 (01Њ 14Ј45.5Љ S, 040Њ 39Ј29.6Љ E), ) 6971Ð6997 Eshoo et al. (2007) Ј respectively, in the village of Marey. BLAST sequence ) 7086Ð7109 ) 9308Ð9283 Ј Ј for the envelope (E1) gene fragment used to identify ) 9007Ð9032 Bryant et al. (2005) Ј the isolates was 99% identical to previously reported ) 4184Ð4203 ) 3368Ð3388 Bryant et al. (2005) Ј Ј Ј

) 309Ð329 sequences (HM147989.1; AF339487.1; AF398375.1; Ј )-3 Ј JN989958.1). ) 5194Ð5213 Bryant et al. (2005) Ј ) 615Ð632 ) 6482Ð6500 ) 124Ð141 Bryant et al. (2005) Ј Ј Ј Ј Discussion Vertical transmission is a fairly common phenom- enon among arboviruses and is thought to play a role in the maintenance of a number of arboviruses in nature, thus ensuring survival during adverse condi- tions (Tesh 1984). Until this study, vertical transmis- sion of NDUV had not been demonstrated. This Þrst evidence of natural vertical transmission of NDUV -TAC GAT GTT GTC GTC GCC GAT GAA-3 -GCT CAC GAT ATG GTC AGC AGG-3 -TGG CGC TAT GAT GAA ATC TGG AAT GTT-3 -CAG CAG ATT GCG CGA CTG ACC-3 Ј Ј Ј Ј -ATT GCA GAT GGG ATA CCG-3 -TGT TCC TGT TGC CAG GAA AAT-3 -CAC CCT AAA AGT GAC GTT-3 therefore suggests that it is one of the mechanisms by Ј Ј Ј -ATG ACT GAG TTG GAG TTT CAT GAT GTC GC-3 Ј -TTT AGC GGA TCG GAC AAT TC-3 -GCG GTG ACG AAC TCA GTA G-3 -GTG TCC CAG CCG GCG GTG TCA TCA GC-3 Ј Ј

Ј which the virus is maintained in this semiarid region -TAC AAC ATG ATG GGA AAG AGA GAG AA-3 -TGT GAA CAA TAG GCA TTG G-3 -GAC TAC CAG TCA GCT CAT TAC C-3 Ј Ј Ј that is characterized by prolonged drought conditions and limited breeding habitats for the mosquitoes dur- CFD2; (5 RVF2: 5 BCS332V; (5 VIR 2052R; (5 ND 615R; (5 Bab 4203R; (5 SINV2; (5 ing the dry season. However, Culex mosquitoes do not lay drought-resistant eggs, and therefore this signiÞ- cant Þnding may suggest that in areas where breeding sites are constantly present, transmission of NDUV may be a constant phenomenon. Because of the lim- ited breeding sites in Ijara, it is also difÞcult to un- derstand how the few Culex mosquitoes available could sustain transmission of the virus between them- selves and the available vertebrate hosts. Because no work involving previously isolated NDUV was being conducted in the laboratory at the time of processing our mosquito samples, and because all necessary precautions were taken to avoid cross contamination between pools of mosquitoes during processing, we can only conclude that these isolations are not as a result of contamination of the specimens. In contrast to other arbovirus groups, there is little Þeld or experimental evidence for vertical transmis- ““ ““ ““ ““ ““ ““ sion of alphaviruses in general (Tesh 1984). The few alphaviruses for which vertical transmission has been

Virus or genus Gene targetdemonstrated Primer sequence include eastern equine encephalitis Position vi- Reference rus (EEEV) from Culiseta melanura larvae and a pool of male mosquitoes of the same species (Chamberlain

Table 1. DNA sequences of the primers used, their target genes and proteins and positions and Sudia 1961), and Ross River virus in experimen- 2. Flavivirus NS5 FU 1; (5 3. Bunyavirus Nucleocapsid protein BCS82C; (5 1. RVFV M Segment RVF1: 5 4. Alphavirus NSP4 VIR 2052F; (5 5. Ndumu Envelope (E1) gene ND 124F; (5 6. Babanki E1 envelope glycoprotein Bab 3368F; (5 7. Sindbistally infected Nonstructural proteinAedes vigilax SINV1; (5 mosquitoes (Kay 1982). 1094 JOURNAL OF MEDICAL ENTOMOLOGY Vol. 51, no. 5

Table 2. Mosquito species and their abundance, sampled as larvae in Sangailu, Ijara

Mosquito collection and virus isolation in the two sites Mosquito Marey Wakab-Harey species Total No. of pools Virus isolates (ϩve pools) Total No. of pools Virus isolates Ae. mcintoshi 00 Ð 154Ð Ae. ochraceus (F) 220 16 Ð 65 12 Ð Ae. sudanensis 18 3 Ð 10 5 Ð Ae. trihcolabis 00 Ð 627Ð Aedes sp. 0 0 Ð 1 1 Ð Anopheles coustani 41 Ð 00 Ð An. gambiae 00 Ð 6912Ð An. maculipalpis 51 Ð 00 Ð

Cx. pipiens (F) 1,813 80 NDUV (2) 653 42 Ð Downloaded from https://academic.oup.com/jme/article/51/5/1091/884145 by guest on 23 September 2021 Cx. pipiens (M) 93 8 Ð 238 14 Ð Cx. poicilipes 00 Ð 53 Ð Cx. theileri 00 Ð 11 Ð Cx. tigripes 00 Ð 84 Ð Cx. univittatus 65 6 Ð 957 72 Ð Cx. vansomereni 44 4 Ð 12 3 Ð Cx. zombaensis 294 17 Ð 0 0 Ð Culex sp. 9 4 Ð 22 5 Ð Total 2,565 140 3 2,118 185 0

F, female; M, male.

Also, Buggy Creek Virus (family: Togaviridae, genus: natural maintenance and ampliÞcation of NDUV (Lu- Alphavirus) has been isolated from Þeld-collected tomiah et al., 2014). This signiÞes the potential of eggs of Oeciacus vicarius (family: Hemiptera, genus: NDUV maintenance by livestock as well as wild ani- Cimicidae) (Brown et al. 2009). Therefore, the isola- mals. tion of NDUV from Cx. pipiens mosquitoes is an im- This Þnding of natural vertical transmission of portant Þnding regarding its natural maintenance. NDUV is the Þrst and adds to the existing knowledge This is also an interesting observation, as NDUV has of vertical transmission that has been demonstrated not been associated with Cx. pipiens before. for other alphaviruses. However, more studies need to NDUV was Þrst isolated in South Africa from Ma. be conducted especially to determine if it is an efÞ- uniformis (Kokernot et al. 1961) and recently in Kenya cient and therefore important component of main- from Ͼ10 species including Ae. mcintoshi, Ae. ochra- taining this virus in nature. The evidence that NDUV ceus, Ae. tricholabis, Ae. sudanensis, and Cx. rubinotus does not cause human illness is also not well sup- (Crabtree et al. 2009, Ochieng et al. 2013; Lutomiah et ported, and more studies need to be conducted to al. 2014. This means that the (potential) vectors of determine if human infection with NDUV actually NDUV are more than is currently recognized. Isola- causes disease or not. tion of NDUV from Þeld specimens collected as larvae provides evidence that Cx. pipiens may play a role in its maintenance in nature. Based on the high number of NDUV isolates (6) Acknowledgments from Cx. rubinotus (Ochieng et al. 2013), it is thought We acknowledge the contributions of Dunstone Beti, that this mosquito, occurring in abundance in estuar- James Mutisya, John Gachoya, Reuben Lugalia, Gilbert Ro- ies and riverine lakes (Jupp et al. 1976) may be im- tich, and Betty ChelangÕat of Centre for Virus Research, portant in the maintenance and transmission of this Kenya Medical Research Institute (KEMRI), for technical virus. Because Cx. rubinotus feeds preferentially on assistance in Þeld collection, identiÞcation, and preparation rodents (Jupp et al. 1976) and Cx. pipiens on avians, it of samples. This project was funded by the U.S. Department is possible that these vertebrate hosts may play a role of Agriculture (USDA). This paper was published with the in the circulation of NDUV especially if the mosquito permission from the Director, KEMRI. species lack the salivary gland infection and escape barriers. However, NDUV has been isolated from nu- merous mosquito species that feed generously on a References Cited wide range of vertebrate hosts which could also be involved in its circulation. Masembe et al. (2012) dem- Brown, C. R., A. T. Moore, G. R. Young, A. Padhi, and N. onstrated the domestic pig (Sus scrofa) as a potential Komar. 2009. Isolation of Buggy Creek Virus (Togaviri- dae: Alphavirus) From Field-Collected Eggs of Oeciacus vertebrate host for NDUV while Lwande et al. (2013) vicarius (Hemiptera: Cimicidae). J. Med. Entomol. 46: isolated NDUV from Rhipicephalus pulchellus col- 375Ð379. lected from cattle (Bos taurus) and warthog (Phaco- Bryant, J. E., M. B. Crabtree, V. S. Nam, N. T. Yen, H. M. Duc, choerus africanus), although it is possible that this virus and B. R. Miller. 2005. Isolation of arboviruses from was in the blood meals from infected animals. It has mosquitoes collected in northern Vietnam. Am J. Trop. also been suggested that goats are involved in the Med. Hyg. 73: 470Ð473. September 2014 LUTOMIAH ET AL.: VERTICAL TRANSMISSION OF NDUMU VIRUS IN Culex MOSQUITOES 1095

Chamberlain, R. W., and W. D. Sudia. 1961. Mechanism of nia serogroups by a PCR technique. J. Clin. Microbiol. 34: transmission of viruses by mosquitoes. Annu. Rev. Ento- 1184Ð1188. mol. 6: 371Ð390. Larkin, M. A., G. Blackshields, N. P. Brown, R. Chenna, P. A. Chomczynski, P., and N. Sacchi. 1987. Single Step Method McGettigan, H. McWilliam, F. Valentin, I. M. Wallace, A. of RNA Isolation by Acid Guanidinium Thiocyanate-Phe- Wilm, R. Lopez, et al. 2007. Clustal W, Clustal X: version nol-Chloroform Extraction. Anal. Biochem. 162: 156Ð159. 2.0. Bioinformatics 23: 2947Ð2948. Crabtree, M., R. Sang, J. Lutomiah, J. Richardson, and B. Lutomiah, J., D. Omondi, D. Masiga, C. Mutai, P. O. Mireji, Miller. 2009. Arbovirus surveillance of mosquitoes col- J. Ongus, K. J. Linthicum, and R. Sang. 2014. Bloodmeal lected at sites of active Rift Valley fever virus transmis- analysis and virus detection in bloodfed mosquitoes col- sion: Kenya, 2006Ð2007. J. Med. Entomol. 46: 961Ð964. lected during the 2006Ð2007 Rift Valley fever outbreak in Edwards, F. W. 1941. Mosquitoes of the Ethiopian Region. Kenya. Vector-Borne and Zoonotic Dis. (in press). III. Culicine Adults and Pupae. London British Museum Lwande, O. W., J. Lutomiah, V. Obanda., F. Gakuya, J. Muti- (Natural History), United Kingdom. sya, F. Mulwa, G. Michuki, E. Chepkorir, A. Fischer, M.

Eshoo, M. W., C. A. Whitehouse, S. T. Zoll, C. Massire, T. D. Venter, and R. Sang. 2013. Isolation of tick and mosquito- Downloaded from https://academic.oup.com/jme/article/51/5/1091/884145 by guest on 23 September 2021 Pennella, L. B. Blyn, R. Sampath, T. A. Hall, J. A. Ecker, borne arboviruses from ticks sampled from livestock and A. Desai, et al. 2007. Direct broad-range detection of wild animal hosts in ijara district, Kenya. Vector-Borne alphaviruses in mosquito extracts. Virology 368: 286Ð295. and Zoonotic Dis. 13: 637Ð642. Forrester, N. L., G. Palacios, R. B. Tesh, N. Savji, H. Guzman, Masembe, C., G. Michuki, M. Onyango, C. Rumberia, M. M. Sherman, S. C. Weaver, and W. I. Lipkin. 2012. Ge- Norling, R. P. Bishop, A. Djikeng, S. J. Kemp, A. Orth, nome-Scale Phylogeny of the Alphavirus Genus Suggests R. A. Skilton, et al. 2012. Viral metagenomics demon- a Marine Origin. J. Virol. 86: 2729Ð2738. strates that domestic pigs are a potential reservoir for Gillies, M. T., and M. Coetzee. 1987. A supplement to the Ndumu virus. Virol. J. 9: 218. Anophelinae of Africa South of the Sahara (Afrotropical Ochieng, C., J. Lutomiah, A. Makio, H. Koka, E. Chepkorir, Region), p. 143. Publications of the South African Insti- S. Yalwala, J. Mutisya, L. Musila, S. Khamadi, J. Richard- tute of Medical Research No. 55, Johannesburg, South son, et al. 2013. Mosquito-borne arbovirus surveillance Africa. at selected sites in diverse ecological zones of Kenya; Jupp, P. G., B. M. McIntosh, and D. Anderson. 1976. Culex 2007Ð2012. Virol. J. 10: 140. (http://www.virologyj.com/ (Eumelanomyia) rubinotus Theobald as vector of Banzi, content/10/1/140). Germiston and Witwatersrand viruses. IV. Observations Schmaljohn, A. L., and D. McClain. 1996. Alphaviruses on the biology of C. rubinotus. J. Med. Entomol. 12: (Togaviridae) and Flaviviruses (Flaviviridae) In: S. Baron 647Ð651. (ed.), Medical Microbiology. 4th ed. Chapter 54, Univer- Kay, B. H. 1982. Three modes of transmission of Ross River sity of Texas Medical Branch at Galveston, Galveston, TX. virus by Aedes vigilax. Aust. J. Exp. Biol. Med. 60: 339Ð344. (http://www.ncbi.nlm.nih.gov/books/NBK7633/). Kokernot, R. H., B. M. McIntoshi, and C. B. Worth. 1961. Tesh, R. B. 1984. Current Topics In Vector Research, Kerry Ndumu virus, a hitherto unknown agent, isolated from F. Harris, vol. 2, pp. 57Ð76. Praeger Publishers, New York, culicine mosquitoes collected in northern Natal, Union of NY. South Africa. Am. J. Trop. Med. Hyg. 10: 383Ð386. Kuno, G., C. J. Mitchell, G.-J.J. Chang, and G. C. Smith. 1996. Detecting bunyaviruses of the Bunyamwera and Califor- Received 16 April 2014; accepted 16 July 2014.