Possible Introduction of Epizootic Hemorrhagic Disease of Deer Virus (Serotype 2) and Bluetongue Virus (Serotype 11) into British Columbia in 1987 and 1988 by Infected Carried on the Wind

Robert F. Sellers and Abdel R. Maarouf

ABSTRACT RESUME Culicoides infectes auraient vraisem- blablement pu provenir de la Vallie Outbreaks of epizootic hemorrhagic Des eruptions de maladie hemorra- d'Okanagan ainsi que des regions disease of deer and of bluetongue gique du cerf (MHC) et de fievre catar- situ&es aussi au sud que la jonction des began in British Columbia in August rhale du mouton (FCM) apparurent en riviieres Okanagan et Columbia dans and October 1987 respectively and Colombie-Britannique respectivement l'etat americain du Washington. Les recrudescence of infection by both en aouAt et en octobre 1987. Une recru- insectes se seraient diplacis A une hau- viruses was detected the following year descence de l'infection par ces deux teur pouvant atteindre 1,5 km A une in August. Weather records for up to virus fut decelee au mois d'aoiut de tempfrature de 12,6 IC ou plus. 18 days before the initial outbreaks of l'annee suivante. On examina les rele- (Traduit par Dr Richard Drolet) disease, isolation of virus or serocon- ves meteorologiques retrospectivement version were examined to determine if jusqu'a 18 jours prealablement a INTRODUCTION the viruses could have been introduced l'eruption initiale de la maladie, ainsi by infected Culicoides carried on the que les resultats de l'isolement viral ou In 1987 there were outbreaks of wind. Data on temperature, rainfall, la presence de seroconversion, afin de epizootic hemorrhagic disease of deer wind speed and direction and pressure determiner si les virus en cause avaient (EHD) and bluetongue (BT) in the together with backward trajectory pu etre introduits par des moustiques Okanagan Valley, British Columbia analysis showed that there were du genre Culicoides, transportes par le and EHD serotype 2 and BT serotype suitable winds which could have intro- vent. L'analyse des donnees concer- 11 viruses were isolated (1). In the fol- duced Culicoides infected with nant la temperature, la pression baro- lowing year, 1988, seroconversion to epizootic hemorrhagic disease of deer metrique, les precipitations, la velocite BT11 and EHD2 virus serotypes was virus on 13 August 1987 (14 days et la direction des vents ainsi que detected in sentinel and BT virus before disease was observed), Culi- l'etude de leurs mouvements retro- serotype 11 was later isolated (2). The coides infected with bluetongue virus grades, a permis de mettre en evidence present analysis was undertaken to on 1 October 1987 (7 days before virus la presence de vents favorables qui determine if the first outbreaks of each was isolated and 13 days before disease auraient pu introduire des Culicoides disease in 1987 or the seroconversion in ) and Culicoides infected with infectes par le virus de la MHC le in 1988 could have resulted from the bluetongue or epizootic hemorrhagic 13 aofut 1987 (14 jours avant que la introduction of EHD2 or BT11 virus disease of deer viruses on 20 July 1988 maladie soit observee), des Culicoides serotypes by infected Culicoides car- (15 days before seroconversion was infectes par le virus de la FCM le ried on the wind from further south in detected). The arrival on 13 August ier octobre 1987 (7 jours avant que western USA. Carriage on the wind of 1987 coincided with the passage of a l'isolement viral soit effectue et infected Culicoides was shown to be a cold front and rain and that on 13 jours avant l'observation de la possible method of introduction of 1 October 1987 with a fall in tempera- maladie chez les moutons), et des bluetongue into Portugal, Cyprus and ture and calm winds. The source of the Culicoides infectes par l'un ou I'autre Florida (3-5). Culicoides before arrival could have des virus le 20 juillet 1988 (15 jours been the Okanogan Valley as far south avant la detection d'une seroconver- as the junction of the Okanogan and sion). L'arrivee du 13 aofut 1987 coin- MATERIALS AND METHODS Columbia rivers in Washington, USA. cida avec le passage d'un front froid Flight would have been at tempera- et de pluie, tandis que celle du Clinical disease due to EHD virus tures of 12.6°C or higher and at ier octobre 1987, avec une baisse de was first seen near Osoyoos (Fig. 1) in heights up to 1.5 km. temperature et des vents faibles. Les deer on 27 August 1987 (6, L. Karstad,

Agriculture Canada, Health of Laboratory Branch, Halldon House, 2255 Carling Avenue, Ottawa, Ontario KlA 0Y9 (Sellers) and Environ- ment Canada, Canadian Climate Centre, 4905 Dufferin Street, Downsview, Ontario M3H 5T4 (Maarouf). Present address of Dr. R.F. Sellers: 4 Pewley Way, Guildford, Surrey, England GUI 3PY. Reprint requests to: Dr. G.C. Dulac, Diseases Research Institute, P.O. Box 11300, Station H, Nepean, Ontario K2H 8P9. Submitted December 6, 1990.

Can J Vet Res 1991; 55: 367-370 367 A. Peatt, G.C. Dulac, D. Ward, 13 August 1200 Greenwich Mean Time unpublished observations, 1988). (Vernon (GMT) (0500 local time) to 14 August Bluetongue virus serotype 11 was 0600 GMT (13 August 2300 local time) isolated first from the blood of a heifer the sources of winds were to the south. collected on 8 October 1987 near In Fig. 1 can be seen the locations of Okanagan Falls and antibodies to BT backward trajectories of winds at virus were detected in the heifer's 1000/900 mb and 850 mb (correspond- blood collected on 26 October 1987 ing to surface, 1000 m and 1500 m (G.C. Dulac, C. Dubuc, unpublished above sea level respectively) 6 h before observations, 1988). Clinical disease arrival at Osoyoos at 1800 GMT (1100 due to BT virus occurred in sheep on local time). On 12 August a cold front 14 October 1987 at Oliver and anti- lay off the coast of British Columbia; bodies to BT virus were found in sheep it moved eastwards on 13 August blood collected on 19 October 1987 and dissipated as it crossed British (L. Karstad, A. Peatt, G.C. Dulac, Columbia. Showers were reported in C. Dubuc, D. Ward, unpublished the morning and heavy rain in the observations, 1988). In 1988 serocon- evening of 13 August at Penticton. version to BT serotype 11 and EHD Southerly winds were blowing at a serotype 2 viruses was first detected in mean speed of 13 kmh- I and the cattle bloods collected on 4 August at distance of 110 km from the junction Oliver and BT serotype 11 virus was of the Okanogan and Columbia rivers isolated from blood collected in early 0 30 in Washington State to Osoyoos would October (2, G.C. Dulac, unpublished km have taken a flight time of 8.5 h. If the observations, 1989). Periods during Culicoides had commenced the flight August, September and October 1987 on the late evening of 12 August they and July 1988 were examined for days would have arrived on the morning of on which flight of Culicoides could 13 August before the morning showers. have occurred. The possible interval Temperatures in the Okanagan Valley before disease, isolation of virus or were from 15.6°C to 18.1 °C during the seroconversion had to take account of flight. the incubation period of EHD and BT During the period 20 September to in the deer and sheep (5-11 days), the 6 October 1987 backward trajectories minimum time required for develop- y showed sources of winds 12 h previ- ment of EHD and BT viruses in Fig. 1. Location map of the Okanagan Valley, ously to the west (3 days), northwest Culicoides before transmission (7-10 British Columbia, Canada and of the Okanogan (5 days), east (1 day), south (5 days) days), the first appearance of BT Valley, Washington, USA. and southwest (2 days). On one day it viremia in cattle (2 days onwards) and t z: location of backward trajectories of was calm. Southwesterly and southerly winds 6 h before arrival at Osooyos - the period after infection for t - 1000/900 mb, 13 August 1987 winds blew from 20-23 September; on antibodies to either virus to develop 1800 GMT 22 and 23 September the winds were (10 days onward) (3-5, 7-9). w - 850 mb, 13 August 1987 1800 GMT light. There were southerly winds from The six-hourly surface and 12-hourly and 850 mb, 20 July 1988 1800 GMT on 1 October (1100 local 850 mb weather 1800 GMT daily maps of the x - 1000/900 mb, 2 October 1987 time) until 0000 GMT on 3 October Canadian Climate Centre, the surface 0000 GMT (1700 local time 2 October). Winds at daily weather maps of the National y - 850 mb, 2 October 1987 0000 GMT 850 mb arriving at Osoyoos at 0000 Weather Center, USA and the Meteo- z - 1000/900 mb, 20 July 1988 GMT on 2 October (1700 local time on rological Summaries for Penticton Air- 1800 GMT. 1 October) had their origin 6 h previ- port, British Columbia were examined ------: US/Canada border. ously not far from the junction of the for temperature, precipitation (rain), Okanogan and Columbia rivers (Fig. 1). wind speed and direction, pressure and the US-Canadian border at the south- With windspeeds of 19 kmh- I the the position of fronts. In addition ern end of the Okanagan Valley (Fig. 1). distance of 35 km from Osoyoos to backward trajectories of winds were Okanagan Falls would have been computed for every 6 h for 120 h start- covered in 1 h 50 min. Thus Culicoides ing at three levels in the boundary layer RESULTS infected with BT virus could have (1000, 900 and 850 mb) at approximate started their flight near the junction of heights of 0.1, 1.0 and 1.5 km above In the period 10-25 August 1987 the two rivers at 1100 local time on sea level as previously described backward trajectories indicated that 1 October 1987 and landed near (5, 10). Osoyoos, British Columbia apart from 13 August the sources of Okanagan Falls (site of first virus isola- was chosen as the starting point for the winds were to the west (8 days), north- tion) about 1900 local time. Flight trajectories as the area where EHD was west (1 day), north (1 day), northeast would have taken place during the day, first reported. It is the closest town to (3 days) and east (2 days). From when temperatures in the Okanagan

368 Valley were between 130 and 21°C. DISCUSSION Seroconversion to BT and EHD Night temperatures for the beginning viruses in 1988 could have resulted of October were between 4° and 10°C. Bluetongue virus serotype 11 has from the survival of the viruses in adult The minimum temperature for flight been detected in previous years in Culicoides or in domestic or wild of Culicoides was found to be 10°C northern California, Oregon and ruminants during the winter. However (11) and the fall of temperature on the Washington and EHD virus has been seroconversion was not detected until evening of 1 October 1987 together reported in Oregon and Washington the beginning of August, which, with calm conditions from 1900 local (12-15). Antibodies to BT and EHD2 although earlier than the initial out- time onwards would have contributed viruses were found in Washington in break of EHD in 1987 (end of August), to the landing of the Culicoides. cattle bloods collected in 1985 (J.E. is later than the first peak of Culicoides In 1988 backward trajectories were Pearson, A.L. Shafer, G.A. Gustafson, in June (17). Overwintering of BT was examined for the period 13-31 July. E.A. Carbrey, unpublished observa- reported to have occurred in Oregon On six days winds were from the west, tions, 1987). Epizootic hemorrhagic during the winter of 1973-74, as nine days from the northwest, one day disease of deer virus serotype 2 was clinical outbreaks were first seen in from the north, one day from the east, reported in Washington in August June 1974 rather than at the usual time one day from the south and one day 1987 and EHD virus serotype 1 was in late summer (13). It is possible that from the southwest. The day on which isolated in Washington from a deer, the source of infection for the Okanagan winds blew from the south was 20 July that had died of an hemorrhagic Valley in 1988 could have been virus and temperatures during the night of disease on 10 September 1987 (16, that had survived the winter in Oregon 19-20 July were between 12.60 and G.C. Dulac, unpublished observa- or further south and been introduced 19°C. The locations of backward tra- tions, 1988). The junction of the by the movement of infected wild jectories of winds at 1000/900 mb and Okanogan and Columbia rivers was ruminants or by wind carriage of 850 mb 6 h before arrival at Osooyos taken as a possible source because of infected Culicoides or by a combina- on 20 July at 1800 GMT (1100 local its position east of the Cascade Range tion of both methods. time) are shown in Fig. 1. At wind- at the beginning of the Okanogan No Culicoides were trapped for speeds of 12-15 kmh- the flight of Valley leading into Canada. The flight identification and virus isolation dur- infected Culicoides from the junction of Culicoides would have been up the ing the outbreaks. However, the of the Okanogan and Columbia rivers valley on the southerly winds. Epizootic Culicoides involved in transmission to Oliver (130 km) would have taken hemorrhagic disease of deer virus most likely belonged to the C. varii- 8.7 to 10.8 h. serotype 2 and BT virus serotype 11 pennis complex, which consists of the The interval between the suggested were detected in the Okanagan Valley subspecies C. v. variipennis, C. v. day of arrival of Culicoides infected but not EHD virus serotype 1. This sonorensis, C. v. albertensis, C. v. with EHD virus (13 August 1987) and may be because from 2 September to australis and C. v. occidentalis (19). disease in deer (27 August 1987) - 12 September 1987 - the period when Downes (20), on the other hand, con- 14 days - lies within the incubation the deer in Washington was sick - sidered C. variipennis and C. occiden- period for EHD and the time for devel- winds from the south were not talis to be distinct species with sono- opment of EHD virus in the midge. If recorded. rensis, albertensis and australis as BT virus had been introduced by It could be that EHD virus serotype subspecies of C. occidentalis. Individ- infected Colicoides between 20 and 23 2 was endemic in the Okanagan Valley uals of C. variipennis and C. occiden- September 1987, the interval before and that the outbreak in 1987 was a talis species were caught in the virus isolation from the heifer and recrudescence. From 1986 to the begin- Okanagan Valley in 1976 and 1977 (17, before disease in sheep would have ning of the outbreak in 1987 no 20), corresponding to the subspecies been 15-18 and 21-24 days respec- evidence of EHD virus activity was C. v. variipennis, C. v. occidentalis tively. Although flight on these days observed in native Canadian animals and C. v. sonorensis of Wirth and might have occurred, a more likely day (6). Even if it was a recrudescence, the Morris (19). Culicoides v. sonorensis for introduction would have been outbreak would more likely have is considered to be the main vector of 1 October 1987. The interval for started in June, when numbers of BT virus in terms of vector competence arrival on that day was seven days Culicoides are at their first peak (17). (19,21). The distribution of BT and before isolation of BT virus and With BT the number of cattle in the EHD corresponds with the distribution 13 days before disease in sheep. In Valley with antibody to the virus was of C. v. sonorensis, C. v. albertensis 1988, if Culicoides infected with BT or nine in 1982, one in 1983, 1984 and and C. v. australis as shown in the map EHD viruses had been introduced on 1986 and none in 1985 (18). Although of Wirth and Morris (19) and apart 20 July, the interval before seroconver- recrudescence is a possibility, the late from the Fraser Valley in British sion was detected would have been date of BT infection in October 1987 Columbia with that of C. occidentalis 15 days. On 25 July 1988, five days suggests introduction of virus from as shown in the map of Downes before the suggested date of introduc- further south. It is possible that EHD (20,22). McMullen (17) collected tion, antibody to BT and EHD viruses and BT viruses could have been intro- C. occidentalis in mid-April (earliest in was not detected in the cattle (G.C. duced by wildlife moving into the 1977) and at the beginning of October Dulac, unpublished observations, Valley from further south, but evidence (latest in 1976) with peaks in late June 1989). for this is difficult to obtain. and late August in 1976. Culicoides

369 occidentalis was trapped at heights up REFERENCES 12. STOTT JL, ELSE KC, McGOWAN B, WILSON LK, OSBURN BI. Epizootiology to 1820 m. The introducton of EHD of bluetongue virus in Western United virus by Culicoides in August 1987 1. DULAC GC, DUBUC C, AFSHAR A, States. Proc US Anim Health Assoc 1981; could have preceded an increase in the MYERS DJ, BOUFFARD A, SHAPIRO J, 85: 170-180. SHEITrGARA PT, WARD D. Consecutive 13. KISTNER TP, REYNOLDS GE, KOLLER population of Culicoides, whereas at outbreaks of epizootic haemorrhagic disease the time of the introduction of BT LD, TRAINER CE, EASTMAN DL. of deer and bluetongue. Vet Rec 1988; 122: Clinical and serological findings on the virus numbers of Culicoides were 340. distribution of bluetongue and epizootic probably decreasing. Thus BT infec- 2. DULAC GC, DUBUC C, MYERS DJ, hemorrhagic disease viruses in Oregon. Proc tion would have been less extensive. AFSHAR A, TAYLOR EA, WARD D, Am Assoc Vet Lab Diagnost 1975; 18: In conclusion the analysis suggests STERRITT W. Incursion of bluetongue 135-148. virus type 11 and epizootic hemorrhagic 14. PARISH SM, EVERMANN JF, OLCOTT that EHD virus serotype 2 was intro- disease of deer type 2 for two consecutive B, GAY C. A bluetongue epizootic in north- duced by infected Culicoides on the years in the Okanagan Valley (British western United States. J Am Vet Med Assoc wind on 13 August 1987 and BT virus Columbia, Canada). Can Vet J 1989; 30: 1982; 181: 589-591. 351. 15. FOSBERG SA, STAUBER EH, REN- serotype 11 on 1 October 1987. Rein- 3. troduction of both serotypes by SELLERS RF, PEDGLEY DE, TUCKER SHAW HW. Isolation and characterization MR. Possible windborne spread of blue- of epizootic hemorrhagic disease virus from infected Culicoides carried on the wind tongue to Portugal, June-July 1956. J Hyg white-tailed deer (Odocoileus virginianus) in could have occurred on 20 July 1988. Camb 1979; 81: 189-196. Eastern Washington. Am J Vet Res 1977; 38: In 1987 the arrival of the Culicoides 4. SELLERS RF, GIBBS EPJ, HERNIMAN 361-364. coincided with the passage of a cold KAJ, PEDGLEY DE, TUCKER MR. Possi- 16. PEARSON JE. In: Proc US Anim Health ble origin of the bluetongue epidemic in Assoc 1988; 92: 122-123. front and rain in August as has been Cyprus, August 1977. J Hyg Camb 1979; 83: 17. McMULLEN RD. Culicoides (Diptera: found on other occasions (5,10,23) and 547-555. ) of the South Okanagan with a decrease in air temperature and 5. SELLERS RF, MAAROUF AR. Trajectory area of British Columbia. Can Entomol analysis and bluetongue virus serotype 2 in 1978; 110: 1053-1057. change to calm conditions in October; Florida in 1988 no association with a cold front 1982. Can J Vet Res 1989; 53: 18. STERRITT W, DULAC GC. Evolving 100-102. perceptions of bluetongue: a challenge for or rain was found. The air tempera- 6. DULAC GC, DUBUC C, AFSHAR A, government and industry. Can Vet J 1991; tures associated with flight were within MYERS DJ, WARD D, BOUFFARD A, 32: (in press). those previously found and flight SHAPIRO J. Outbreak of epizootic hemor- 19. WIRTH WW, MORRIS C. The taxonomic rhagic disease of deer in the Okanagan complex, Culicoides variipennis. In: Barber would have been at heights up to Valley, British Columbia, 1987: virus isola- 1.5 km (3,5,23). TL, Jochim MM, eds. Bluetongue and tion and serological findings. Can Vet J Related Orbiviruses. New York: Alan R Liss 1991; 32: (in press). Inc., 1985: 165-175. 7. FAY LD, BOYCE AP, YOU4TT WG. An 20. DOWNES JA. The Culicoides variipennis ACKNOWLEDGMENTS epizootic in deer in Michigan. Trans North complex: a necessary re-alignment of Am Wildl Nat Resour Conf 1956; 21: nomenclature (Diptera: Ceratopogonidae). 173-184. Can Entomol 1978; 110: 63-69. The authors are grateful to their col- 8. HOURRIGAN JL, KLINGSPORN AL. 21. JONES RH, FOSTER NM. Heterogeneity leagues in Agriculture Canada and Bluetongue: the disease in cattle. Aust Vet of Culicoides variipennis field populations Environment Canada for their sup- J 1975; 51: 170-174. to oral infection with bluetongue virus. Am 9. BOORMAN J, GIBBS EPJ. Multiplication J Trop Med Hyg 1978; 27: 178-183. port. A. Afshar, C. Dubuc, G.C. of the virus of epizootic haemorrhagic Dulac, L. Karstad and 22. COSTELLO RA. Light trap collections of W.G. Sterritt disease of deer in Culicoides species (Diptera, Culicoides (Diptera: Ceratopogonidae) in the are thanked for information and Ceratopogonidae). Arch ges Virusforsch Fraser Valley of British Columbia. advice on the field and laboratory 1973; 41: 259-266. J Entomol Soc British Columbia 1982; 79: 10. SELLERS RF, MAAROUF AR. Impact of 66-70. aspects of the outbreaks and Joan climate on western Graham for typing the manuscript. equine encephalitis in 23. SELLERS RF, PEDGLEY DE. Possible Manitoba, Minnesota and North Dakota, windborne spread to Western Turkey of The Meteorological Office, Bracknell, 1980-1983. Epidemiol Infect 1988; 101: bluetongue virus in 1977 and of Akabane England was helpful in providing the 511-535. virus in 1979. J Hyg Camb 1985; 95: senior author with access to North 11. NELSON RL, BELLAMY RE. Patterns of 149-158. flight activity of Culicoides variipennis American weather data. (Coquillett) (Diptera: Ceratopogonidae). J Med Entomol 1971; 8: 283-291.

370