Virus Research 116 (2006) 1–10 Molecular epidemiological study of Arctic rabies virus isolates from Greenland and comparison with isolates from throughout the Arctic and Baltic regions K.L. Mansfield a , V. Racloz a,b , L.M. McElhinney a , D.A. Marston a , N. Johnson a , L. Rønsholt c , L.S. Christensen c , E. Neuvonen d , A.D. Botvinkin e , C.E. Rupprecht f , A.R. Fooks a,∗ a Rabies Research and Diagnostic Group, Veterinary Laboratories Agency (VLA, Weybridge), WHO Collaborating Centre for the Characterisation of Rabies and Rabies-Related Viruses, New Haw, Addlestone, Surrey KT15 3NB, UK b The Royal Veterinary College, University of London, London, UK c Danish Institute for Food and Veterinary Research, Lindholm, DK-4771 Kalvehave, Denmark d National Veterinary and Food Research Institute, Department of Virology, P.O. Box 45, Hameentie 57, FIN-00581 Helsinki, Finland e Irkutsk State Medical University, 1 Krasnogo Vosstania, Irkutsk 664003, Russia f Rabies Section, Centers for Disease Control and Prevention, 1600 Clifton Rd., Atlanta, GA 30333, USA Received 2 May 2005; received in revised form 12 August 2005; accepted 12 August 2005 Available online 28 September 2005 Abstract We report a molecular epidemiological study of rabies in Arctic countries by comparing a panel of novel Greenland isolates to a larger cohort of viral sequences from both Arctic and Baltic regions. Rabies virus isolates originating from wildlife (Arctic/red foxes, raccoon-dogs and reindeer), from domestic animals (dogs/cats) and from two human cases were investigated. The resulting 400 bp N-gene sequences were compared with isolates representing neighbouring Arctic or Baltic countries from North America, the former Soviet Union and Europe. Phylogenetic analysis demonstrated similarities between sequences from the Arctic and Arctic-like viruses, which were distinct from rabies isolates originating in the Baltic region of Europe, the Steppes in Russia and from North America. The Arctic-like group consist of isolates from India, Pakistan, southeast Siberia and Japan. The Arctic group was differentiated into two lineages, Arctic 1 and Arctic 2, with good bootstrap support. Arctic 1 is mainly comprised of Canadian isolates with a single fox isolate from Maine in the USA. Arctic 2 was further divided into sub-lineages: 2a/2b. Arctic 2a comprises isolates from the Arctic regions of Yakutia in northeast Siberia and Alaska. Arctic 2b isolates represent a biotype, which is dispersed throughout the Arctic region. The broad distribution of rabies in the Arctic regions including Greenland, Canada and Alaska provides evidence for the movement of rabies across borders. © 2005 Elsevier B.V. All rights reserved. Keywords: Rabies; Greenland; Fox; Arctic; Baltic; Zoonosis; Wildlife 1. Introduction Classical rabies virus (RABV) is one of seven recognised genotypes of the genus Lyssavirus within the family Rhabdoviridae. Members of this genus have a negative-strand RNA genome encoding five proteins: nucleoprotein, phosphoprotein, matrix protein, glycoprotein and an RNA polymerase. The nucleoprotein gene, although highly conserved, allows viral strains to be accurately differentiated by analysing genetic differences that ∗ Corresponding author. Tel.: +44 1932 357 840; fax: +44 1932 357 239. E-mail address: t.fooks@vla.defra.gsi.gov.uk (A.R. Fooks). 0168-1702/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.virusres.2005.08.007 are present within the gene (Johnson et al., 2002). Given the circumpolar nature of rabies, genetic sequencing may provide an insight as to the maintenance of the disease and its possible translocation throughout the Arctic. Considering its size and location, rabies virus circulation in Greenland should be comparable to other polar locations. In this study, sequencing and phylogenetic analysis of a partial region of the nucleoprotein gene was undertaken, for a panel of rabies virus isolates representing both Arctic and Baltic countries. Rabies is pathognomonic and has been detected in a number of wild species throughout the Arctic (Mørk and Prestrud, 2004); in the seal (Pusa hispida) population (Ødegaard and Krogsrud, 1981), the raccoon-dog (Nyctereutes procyonoides) population 2 K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 in Finland (Nyberg et al., 1992), whilst the first confirmed case of rabies in a polar bear (Ursus maritimus) was detected in the Northwest Territories of Canada in 1989 (Taylor et al., 1991). In addition, rabies often occurs in domestic dogs (Canis familiaris) and sledge dogs (Sikes, 1968) and is occasionally detected in sheep (Ovis aries) (Leisner, 2002). Throughout the Arctic, principally due to harsh polar weather conditions, the biodiversity is limited and contains nine terrestrial mammals: wolf (Canis lupus), hare (Lepus arcticus), lemming (Dicrostonyx torquatus), polar bear (Ursus maritimus), caribou (Rangifer tarandus), musk-oxen (Ovibos moschatus), stoat (Mustela erminea), wolverine (Gulo gulo) and Arctic fox (Alopex lagopus) (Leisner, 2002). Greenland is the largest island in the world, with 77% of its 2,175,600 km2 covered by an ice cap up to 3 km thick with animal and plant life restricted to the 44,000 km of coastal areas. The principal reservoir for rabies is the Arctic fox (WHO, 1990). Rabies epizootics have been reported periodically in Greenland and coincide with population peaks and migration patterns of this species (Follmann et al., 1992). The first description of rabies-like signs was recorded in 1859 (WHO, 1990) during an outbreak of ‘Eskimo dog disease’. A more detailed description of rabies in Arctic regions was made by Plummer in 1947 who identified ‘Arctic dog disease’ (Plummer, 1947a,b). This was followed by the first laboratory confirmation of rabies in 1959 (Jenkins and Wamberg, 1960), where it was detected in two dogs and two foxes. The interaction between wildlife reservoirs (principally the Arctic fox) and the domestic and sledge dog population suggests that transfer of rabies virus between wildlife reservoirs and domestic animals occurs periodically. Canine distemper virus has also been regularly reported in the sledge dog population of northern Greenland and the spread of the disease, connected with Canadian outbreaks, is communicated by wildlife, particularly foxes. Once established, further spread of the disease was mediated by travelling dog teams (Bohm et al., 1989). Between 1957 and 1960, four outbreaks of rabies among dogs in northern Greenland almost caused a famine among the Eskimo population, who rely heavily upon their dogs (Lassen, 1962). Following a mass epidemic during 1959–1960, which killed over 1000 dogs in the district of Egedesminde (Crandell, 1991), the government of Greenland introduced a rabies vaccination programme for dogs in 1969 (WHO, 1990). With the introduction of the vaccination programme, the number of cases in dogs dramatically decreased. However, from 1964 to 1975, a number of rabies cases were still being confirmed in various animal species (Arctic foxes, dogs, horses, sheep and caribou), particularly in the southwestern coastal district of Disko Bay, and Thule district (WHO, 1990). During the period from 1975 to 1989, 16% of suspect dogs and 67% of suspect Arctic foxes in Greenland were reported to be rabid (WHO, 1990), and most of the cases were located along the western coast (Fig. 1). In 1987, rabies was detected for the first time in Arctic foxes in Godthab, southwest Greenland (WHO, 1990), whereas in 1988, rabies was only found in Arctic foxes from Thule, Upernavik, Sukkertoppen and Godthab districts (Fig. 1), and was not found in dogs or other animals (WHO, 1990). Fig. 1. Map showing the location in Greenland, where positive samples of Arctic rabies isolates were recorded. Official statistics provided by the Office International des Epizooties (OIE) demonstrate that the total number of recorded rabies cases in Greenland has increased from five cases in 1996 to 21 in 2002 (OIE, 2004). Rabies is considered endemic among Arctic foxes in Greenland, especially in the northwestern districts from Sisimiut to Avanersuaq (OIE, 2002). The presence of this disease presents a continuous threat to wildlife, domestic animals and human health, particularly as attacks from aggressive dogs are common in Greenland (Crandell, 1991). However, only one human rabies case has been documented in Greenland. In 1960, a 4-year-old Eskimo child from Egedesminde died after being savagely attacked by a pack of dogs during a period when a rabies epizootic was prevalent throughout the dog population (Lassen, 1962). Although ‘atypical’ Negri bodies were first reported, both clinical and laboratory diagnosis later confirmed the diagnosis of rabies. The low incidence of rabies in humans may be as a result of under-reporting, due to a lack of surveillance and diagnostic facilities in some Arctic regions. The relatively small numbers of human inhabitants and the limited contact with arctic fox populations may also be responsible for the rarity of human cases, despite epizootics of rabies in wildlife and transmission to domestic species. Although controversial, it has also been proposed that the Arctic strain of rabies virus is less virulent in man and that through the induction of acquired immunity (particularly virus neutralising antibodies), individuals can effectively clear the virus, especially from a minor peripheral exposure. The human case of rabies in Greenland followed a category III exposure in which the girl was heavily bitten ‘to the bone’ around the face (Lassen, 1962). Similarly, a rare human case of arctic fox rabies was reported in 1998 in Norolisk, in K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 the Arctic region of Russian Siberia, following an attack by a rabid wolf. The man received extensive wounds to the head, face, shin and hand and died 31 days after the attack (Kuzmin, 1999). In Alaska, human exposures to RABV are typically not from wildlife but from domestic dogs that have had contact with Arctic foxes or other infected wild animals (Follmann et al., 1996). In 1989, a single sledge dog from Station Nord (northeast Greenland) was infected with rabies (WHO, 1990), along with a number of Arctic foxes from the same location. Greenland’s Home Rule Government made the vaccination of dogs and cats mandatory in 1997, in accordance with Home Rule Order No. 22 (OIE, 2002). At present, Greenland has still not achieved 100% vaccine coverage, and a recent study suggested that antibody titres measured among the dogs were low (Leisner, 2002). Due to the low level of human habitation in many areas and the lack of geographic symmetry of rabies outbreaks, re-introductions of rabies may go unnoticed or become endemic for long periods of time before detection (Follmann et al., 1992). In the Norwegian islands of Svalbard, several cases in different species of animals have been detected since the 1980s (Prestrud et al., 1992; Westerling et al., 2004), but the overall probability of detecting rabid animals is low due to the sparse population (about 80% of all detected rabies cases were less than 20 km from permanent settlements). This study represents the first molecular epidemiological study of rabies virus isolates circulating in Greenland and other countries within both Arctic and Baltic regions. 3 2.3. RT-PCR and sequencing Reverse transcription and polymerase chain reaction were performed as described previously (Johnson et al., 2004) using the combination of primers JW12, JW6 DPL, JW6 E and JW6 M, although the amount of cDNA added to each reaction was doubled. PCR products were purified using the QIAquickTM PCR Purification Kit, and sequenced in both directions using the Big Dye Kit (Applied Biosystems, Warrington, UK) with primers at 3.2 pmol/␮l. Forward sequences were produced with primer JW12, whereas reverse sequences were obtained using either JW6 DPL or JW6 E. 2.4. Phylogenetic analysis 2. Materials and methods Nucleotide sequences were edited to 400 bp using the DNAstar program (DNAstar Inc. Madison, USA). Multiple sequence alignments were generated using the ClustalW programme 1.83 (Thompson et al., 1994) and transition-transversion ratios were estimated by the Puzzle 32 programme (Strimmer and Von Haeseler, 1996). Each data set was analysed using the maximum likelihood method within the DNAdist programme of the PHYLIP package (Felsenstein, 1989). Bootstrap re-sampling with 100 replicates using the Seqboot, DNAdist and Neighbour programmes provided confidence limits for the constructed phylogenies, and consensus trees were generated with the Consense programme. Phylogenetic trees were generated using the Drawtree and Drawgram programmes (Phylip), and bootstrap values were visualised using Treeview (Page, 1996). Additional sequences were obtained from GenBank and from samples previously submitted to the VLA rabies archive. 2.1. Isolates 3. Results Seventeen samples from various locations along the western coastline of Greenland (Fig. 1) were collected between 1990 and 2002 from Arctic foxes (Alopex lagopus), dogs (Canis familiaris), and a reindeer (Rangifer tarandus), and were cultivated in MNA monolayer cell cultures according to standard diagnostic procedures before analysis. Of the 17 submitted samples, six were included in phylogenetic analysis. Previously published RABV sequences of isolates originating from North America (USA and Canada), Russia and Europe were also included in the phylogenetic comparison. All isolates included in phylogenetic analysis are detailed in Table 1, with the reference to the origin of isolates denoted after the accession numbers as follows: (a) Bourhy et al., 1999, (b) Nadin-Davis et al., 1993, (c) Kissi et al., 1995, (d) Nadin-Davis et al., 1999, (e) Nadin-Davis et al., 1994, (f) Arai, 2004, (g) Vanaga et al., 2003, (h) Kuzmin et al., 2004. The data for rabies cases in Greenland between 1975 and 2001 demonstrate that the principal vector for rabies in Greenland is the Arctic Fox (Fig. 2) and suggests the periodic nature of rabies epizootics are possibly as a result of re-introduction of the disease over large areas, complicated by surveillance bias and lapses in precise surveillance and reporting. Of 27 samples submitted to the Danish Institute for Food and Veterinary Research from locations on the west and southern coastal regions of Greenland (Fig. 1), 17 were sequenced. The remaining 10 2.2. Extraction of RNA RNA was extracted using TRIzol® (Invitrogen, Paisley, UK) following the manufacturer’s instructions, and resuspended in HPLC purified water (Sigma–Aldrich). The RNA was stored at −80 ◦ C and used undiluted in RT-PCR. Fig. 2. Reported cases of rabies in Greenland from 1975 to 2001 (Source: Rabies Bulletin Europe). 4 K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 Table 1 Viral isolates included in phylogenetic analysis Sample ID 9447AUT RV441 9352BEL RABN0783 RABN1578 RABN2756 RABN9196 9105CAN 6199 8480FX 2244 RV163 1090DG 4055DG RV437 9339EST 9342EST RV118 9147GSFRA RV313 8684GRO RV1391 RV1396 RV1407 RV1413 RV1419 RV1420 9387HON Komatsugawa Takamen RV259 LAT02 LAT04 LAT12 LAT18 9126MEX RV277 RV245 RV248 RV255 RV294 RV303 RV304 RV439 RV440 RV443 RV1334 RV1336 RV1338 857r RVHK 3683c RV61 1420 1421 1422 4795 RV51 RV53 8658YOU Sender’s reference 5 Ontario T3 Ontario T1 Ontario T2 Ontario T4 Ontario T2/4 Ontario T5 Ontario T2/4 5367-N85 Arctic 1 HB G1455 Gra 2/90 Gra 7/92 Gra 23/94 Gra 7/97 Gra 36/02 Gra 37/02 360 11 SHEG 209 745 994 768 158 196 210 43 1 N26 N25 MIFOX86 MEFOX26 Species Location Date GenBank® accession no. Red fox Fox Red fox Arctic fox Arctic fox Arctic fox Arctic fox Red fox Red fox Red fox Red fox Arctic fox Dog Dog Raccoon dog Raccoon dog Raccoon dog Dog Red fox Red fox Arctic fox Arctic fox Arctic fox Arctic fox Arctic fox Arctic fox Arctic fox Cat Dog Human Red fox Dog Raccoon dog Fox Dog Dog Goat Human Red fox Arctic fox Polar fox Raccoon dog Human Red fox Dog Horse Wolf Arctic fox Arctic fox Raccoon dog Human Steppe fox Human Red fox Red fox Arctic fox Dog Fox Fox Cattle Austria Belarus Belgium Canada (Ontario) Canada (Ontario) Canada (Ontario) Canada (Ontario) Canada (Ontario) Canada (Ontario) Canada (Ontario) Canada (Ontario) Canada (Arctic) Canada (Arctic) Canada (Hudson Bay) Estonia Estonia Estonia Finland France Germany Greenland Greenland Thule Greenland Sisimiut Greenland Upernavik Greenland Grønnedal Greenland Ilulissat Greenland Kangerlussuaq Hungary Japan Japan Kazakhstan Latvia Latvia Latvia Latvia Mexico Pakistan Russia (Pskov) Russia (Chita) Russia (Yakutia) Russia (Yakutia) Russia (Spassk) Russia (Byrobidzhan) Russia (Pskov) Russia (Pskov) Russia (Yakutia) Russia (Krasnoyarsk) Russia (Yakutia) Russia (Yakutia) Russia (Chabarovsk) Russia (Norilsk) Russia (Omsk) UK ex India USA (Alaska) USA (Alaska) USA (Alaska) USA (Alaska) USA (Michigan) USA (Maine) F.R. Yugoslavia 1994 U42708 (a) DQ010126 U42709 (a) L20675 (b) L20673 (b) L20674 (b) L20676 (b) U22655 (c) U11734 (d) U03768 (e) U11735 (d) DQ010124 U03769 (e) U03770 (e) AY091627 U42707 (a) U43432 (a) AY751534 U22474 (c) AY062070 U22654 (c) DQ010132 DQ010136 DQ010141 DQ010143 DQ010147 DQ010148 U43001 (a) AB178890 (f) AB178891 (f) AY352491 AY277571 (g) AY277572 (g) AY277574 (g) AY277576 (g) U22477 (c) AY062069 AY352475 AY352460 DQ010125 AY352514 AY352505 AY352502 AY352474 AY751535 DQ010127 DQ010128 DQ010129 DQ010131 AY352458 (h) AY352462 (h) AY352469 (h) AY102993 AY352499 (h) AY352500 (h) AY352501 (h) AY352498 (h) DQ010122 DQ010123 U42705 (a) 1992 1991 1991 1991 1990 1990 1991 1993 1993 1989 1993 1992 1988 1991 1990 1981 1990 1991 1994 1997 2002 2002 1993 1940’s 1940’s 1988 1999 1999 1999 1999 1991 1990 1977 1988 1990 1980 1988 1990 1990 1990 2002 2002 2002 1998 1988 1986 1981 K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 Fig. 3. (a) Radial tree of Greenland isolates with Arctic and Baltic isolates. (b) Phenogram of Arctic lineages. 5 6 K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 samples did not produce a detectable PCR product, despite a suspicion of rabies and were further confirmed negative by in vitro analysis in a rabies tissue culture inoculation test. Of the 17 sequences generated, a high degree of homology was demonstrated among the Greenland isolates within the Arctic group based upon phylogenetic analysis of the 400 base pair nucleoprotein region (Fig. 3a and b). Many values were close to 100%, with a number of sequences demonstrating 100% homology, such as those obtained for two Arctic foxes from Kangerlussuaq (RV1418-not shown and RV1420). A second group of identical sequences comprised isolates originating from two Arctic foxes from Sisimiut, dogs from Christianshåb and Narssaq, Arctic foxes from Julianehåb and Kangerlussuaq, and a dog and Arctic fox from Nuuk. This group contained both dog and Arctic fox isolates, which originated from a wide area on the southwestern coast of Greenland. The panel of Greenland isolates yielded six unique sequences, which were included in the phylogenetic analysis. Without exception, all of the submitted Greenland isolates converged within Arctic group 2b, one of three discrete groups among the Arctic isolates. This group also contained a number of isolates from Canada (dog, red fox and Arctic fox), Alaska (red and Arctic foxes) and an Arctic fox isolate from Siberia (RV1338). Arctic 2b comprises isolates from throughout the Arctic region and may represent a rabies virus biotype (Fig. 3b). Further RABV sequences were categorised into groups that are defined by geographical location: Baltic group (European part of the Former Soviet Union); Steppes Group (Russia); European Group (western, central and eastern Europe); Arctic 1 group (North America); Arctic 2a Group (North East Siberia/ Alaska); Arctic 2b Group (North America/Siberia/Greenland) and the Arctic-like Group (Russia/Japan/Pakistan/India/Siberia) (Fig. 3a and b). Isolates from the Baltic regions formed a single group. The Baltic group comprised isolates from dogs and terrestrial wildlife from Latvia, Finland, Estonia and North West Russia. This group contained a human isolate from Pskov in Russia (RV245), which demonstrated 100% homology with a raccoon dog isolate originating from the St Petersburg region (RV309, not shown). Within this group is also a red fox isolate (RV439) and a dog isolate (RV440) originating from Pskov in Russia. Grouping closely to the Baltic group was a smaller group, Steppes Group, comprising isolates from a wolf from Siberia (RV1334), a red fox from Kazakhstan (RV259) and a steppe fox from Omsk, Russia (3683c). A cat isolate from Hungary and red fox isolates from Austria, Belgium, France and Germany form a discrete European group. The Arctic and Arctic-like viruses are distinct from rabies isolates originating in the Baltic region of Europe, the Steppe in Russia and from North America (Fig. 3a). In addition, a bootstrap value of 100% separates the Arctic and Arctic-like groups suggesting that the virus clades from both groups are distinctive. The Arctic group is further differentiated into two lineages (Arctic 1 and Arctic 2) with 100% bootstrap support. Arctic 1 is principally comprised of Canadian isolates with two exceptions, RV53, and an earlier Arctic fox isolate originating from Greenland in 1981 (8684GRO). The virus, 8684GRO is most closely related to RV53 (98.2%), both divergent isolates forming a separate sub-lineage within the group. The isolate RV53 was a fox isolate from Maine in the northeast region of the USA on the border with Canada. The possibility exists that 8684GRO was a previous introduction of Arctic rabies from North America. The second group may be further divided into sub-lineages: Arctic 2a and 2b, however, the bootstrap values are low and cannot be considered significant (Fig. 3b). Arctic 2a comprises isolates from Russia and Alaska although no details are available for the exact location of the Alaskan samples. The Russian isolates were from the Arctic region of Yakutia in northeast Siberia. There is also an isolate from Belarus (RV441), which has been shown to be positive using a monoclonal antibody (P41) that is characteristic for epitopes on the N-protein of specific Arctic isolates. Interestingly, there are no Canadian isolates present in the Arctic 2a group (Fig. 3b). A distinct lineage, known as the Arctic-like Group, comprised diverse isolates from raccoon dogs, red foxes, humans, a goat and a dog from South East Siberia, India, Pakistan and Japan. This lineage was distinct from the Arctic lineages with good bootstrap support (100%). 4. Discussion The Arctic-like group consists of isolates encompassing an area from southeast Siberia, India and Pakistan to Japan in the Far East (Bunge, 1888; Arai, 2004; Kuzmin et al., 2004). Although only speculative, it is highly likely that the Arctic-like group would include isolates from the Middle East including those from Iran (Nadin-Davis et al., 2003). The Arctic RABV variant has also been detected in areas of the world distant from the Arctic region. Indeed, viruses of the Arctic lineage are known to circulate in dogs and wild canids in northern temperate, subArctic and Arctic regions of Canada, Russia, northern India, Korea and Nepal (Rausch, 1958; Nadin-Davis et al., 1993, 2003; Webster et al., 1986; Kissi et al., 1995; Mørk and Prestrud, 2004). Recently, Arctic RABV was detected from Iran in a sheep and a dog (Nadin-Davis et al., 2003). Two isolates were recovered in a highly restricted area in northeastern Iran. This study demonstrated that these two viruses were closely related to members of the Arctic lineage, and in particular to viruses recovered from dogs in Nepal, and suggested that this Arctic-like strain was recently transmitted to Iran from neighbouring countries to the east. At certain times of the year, periods of cold allow Greenland and other countries within the Arctic region to be connected by large moving masses of ice. Parts of Russia, Norway, United States, Canada and Greenland are all connected by ice, and are accessible to animal migration. The potential exists for migrating wildlife to transmit viruses between neighbouring Arctic regions, where they may spill-over into other wild or domesticated species. In Arctic regions, the occurrence of rabies epizootics have not been regularly recorded nor the public health risks fully appreciated, due mainly to the sparse human populations in these areas (Follmann et al., 1992). The isolates analysed in this study originated mainly from the western coastal regions of Greenland (Fig. 1). These areas K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 are the most populated in terms of animal reservoirs and human settlements. This does not imply that rabies is restricted to this area only, as there may be isolated cases in sparsely populated areas where sampling is difficult. The high level of homology among the Greenland isolates demonstrates that there is much interaction between the animal populations from which these isolates originated. The isolates that displayed 100% homology originated from along the western coastline of Greenland. Phylogenetic analysis based upon a 400 bp sequence of the nucleoprotein, demonstrated genetic similarities between isolates from Greenland and those from North America. We have shown a distinction between samples from Arctic regions (Greenland, Canada, USA) and samples from Baltic regions (Latvia, Estonia and Finland). However, isolates from Russia and Siberia appear in both the Arctic and Baltic groups, which demonstrates that regions such as these may have both Baltic and Arctic variants of RABV circulating, and suggests movement of virus between the different Arctic regions. Therefore, although there is a distinction between Arctic, Arctic-like and Baltic variants of RABV, all groups are associated with specific geographic regions and there is a continual interaction between them as wildlife migrates throughout the regions. In Greenland and North America the host range is relatively small with fewer indigenous mammals capable of acting as a RABV reservoir. The host range in the Baltic and Russian groups include raccoon dogs, red foxes and domestic dogs in the former and Arctic fox, red fox and domestic dogs in the latter. Although the isolates originating from Greenland were the entire Arctic lineage, regions such as Russia and Siberia contain both the Arctic and Baltic variants, which may be a consequence of their accessibility to the European landmass. The Baltic group contained a red fox isolate (RV439) from Pskov in Russia, which is approximately 700 km outside the Arctic Circle, suggesting that the terrain for Baltic variants of RABV is broad (Botvinkin and Kosenko, 2004). In another study (Metlin et al., 2004), antigenic differences between isolates from domestic animals and wildlife circulating in Russia, Finland and Estonia suggested that similar Arctic viruses were present in both species (variants I, III and IV). However, the presence of one variant (variant II) isolated from wildlife was rare and reported to circulate in Baltic regions only (Metlin et al., 2004). In Russia, Arctic foxes have been found as far south as 55◦ N latitude (Crandell, 1991), and in 1998 the Arctic fox RABV strain was responsible for a human rabies case in Siberia following a wolf bite (Kuzmin, 1999). Virus isolation and staining with monoclonal antibody P-41 suggested that this virus was of Arctic fox origin. This isolate (RVHK) is present in the Arctic 2b group and like the Yakutian Arctic fox isolate (RV1338) is distinct from the Arctic 2a isolates circulating in North East Siberia. Previous data have demonstrated that the human isolate from Pskov and a raccoon dog isolate from St Petersburg (Baltic group) shared 100% homology in a 400 bp region of the nucleoprotein gene (Johnson and Fooks, 2005). Therefore, in this region both the Arctic and Baltic variants of RABV pose a risk in the further dissemination of rabies to terrestrial wildlife, particularly from migrating Arctic foxes (Selimov et al., 1994). 7 The presence of a red fox from Canada (8480FX) in the Greenland group suggests a link between the Arctic RABV variants of Greenland and North America although the direction of virus transfer has not been confirmed. A study on isolates obtained from an outbreak in central Ontario, Canada, has suggested that Arctic rabies may have spread southwards from the Arctic and around both sides of Hudson Bay as two distinct fronts (Nadin-Davis et al., 1994). Conversely, Arctic foxes from the Northwest Territories of Canada are thought to migrate from the ice-covered straight on Ellesmere Island to Greenland (Webster et al., 1985). The most likely origin of rabies in Svalbard, Norway, is thought to be via the migration of Arctic foxes from Greenland or the Siberian islands (Prestrud et al., 1992). There is much evidence to suggest the movement of RABV between the Arctic countries, via migration of Arctic foxes and other wildlife across the ice, although the involvement of polar bears in the movement of RABV is unlikely. All of the evidence suggests that there is no distinct pattern to the movement of rabies virus between the Arctic countries; a fact, which may hinder any, attempt to control or eliminate the virus across the Arctic (Mørk and Prestrud, 2004). One isolate, RV51, a fox virus from Michigan, is genetically distant to all other isolates. Although RV51 differs genetically, the isolate was detected in the geographical vicinity of the Great Lakes and may therefore represent a sporadic introduction from a different region of the USA (Fig. 3a). Arctic 2b comprises isolates from throughout the Arctic region and may represent a biotype which is dispersed as a result of the mass ice movements or human movements, e.g. the Yakutian isolate (Arctic fox RV1338) groups with these isolates in Arctic 2b instead of the other Arctic fox isolates from Yakutia (Arctic 2a). Surprisingly, the previously published Greenland isolate (8684GRO) from 1981 (Kissi et al., 1995) is the only Greenland isolate to fall outside of Arctic 2b. None of the isolates used in this study that are representative of Arctic 2b were isolated before 1990 and the inclusion of 8684GRO in Arctic 1 may suggest that it resulted from a sporadic import from Canada or represents a biotype that no longer exists. Programmes to control rabies in red foxes, such as that in Ontario, Canada (MacInnes et al., 2001) have already been effective, and vaccination campaigns of Arctic foxes in Finland have ensured that the last outbreak occurred there in 1989 (WHO, 1990). Due to the small number of host species, Greenland should be one of the most amenable Arctic regions in which a targeted vaccination strategy for companion animals (sledge and domestic dogs only) would succeed. However, wildlife vaccination (either oral-baiting vaccination or trap-vaccinate-release strategies) presents some distinct challenges in a region as large as Greenland, although Arctic rabies has been controlled in other countries within a limited area (WHO, 1990). As a direct result of the interactions between wildlife in the Arctic region, and between wildlife and domestic animals, however, rabies elimination is problematic. Previous vaccination trials in northern Greenland failed to fully control the disease (Hansen, 1996). Furthermore, the probability of virus re-introductions from migrating wildlife across the frozen ice plains further complicates any rabies control strategy. Arctic foxes live in small groups and have 8 K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 territories, which can measure from 8.5 to 72.8 km2 depending upon the amount of available food. Although Arctic foxes are known to follow polar bears onto the drift ice to feed on the remains of their seal catches (Leisner, 2002), their principal food source are rodents, along with certain bird species (Follmann et al., 1992). However, with the onset of the harsh Arctic winter, numbers of these rodents decline and the birds migrate south, requiring the Arctic fox to increase its territory to find sufficient food. Healthy Arctic foxes have been known to travel distances of up to 2300 km during a lifetime (Prestrud et al., 1992). Reports in Greenland have suggested that rabid Arctic foxes may become tame, lose their natural shyness, and therefore facilitate the transfer of virus to the sledge dog, which is the second highest infected animal species in Greenland (Leisner, 2002). The number of rabies cases observed in Greenland each year is variable (Fig. 2), according to the population patterns of the Arctic fox, and most diseased foxes are observed in March and April during the hunting season (Crandell, 1991). Moreover, there is a continual interaction between animals that are infected with either the Arctic or Baltic RABV variants, as wildlife migrate throughout the Arctic region. Future rabies control strategies in Greenland should therefore be planned based on our understanding of the epidemiology of rabies in the Arctic and Baltic regions (Mørk and Prestrud, 2004). Although, our knowledge of the principal reservoir, the Arctic fox, has improved, little information is reported on other possible reservoir species. These include the possibility of spillover infections from bats (Daoust et al., 1996) and other terrestrial mammals such as raccoon-dogs, skunks, racoons and red foxes. Unfavourable factors for wildlife vaccination include the cost-benefit of any control strategy for wildlife, topography, remoteness of the terrain, daily/seasonal and annual temperature variation, seasonal and annual sea ice occurrence and ecology of other susceptible hosts (WHO, 1990). The maintenance of rabies virus in populations of Arctic foxes at low densities is still not known; the possibility for re-introductions exists through oral transmission as a result of consuming a previously preserved carcass from the permafrost and from animal movements between regions (Mørk and Prestrud, 2004). One suggestion is that the virulence of some Arctic fox strains is lower than strains of rabies virus circulating in other continents, although this hypothesis remains unproven (Sikes, 1962; Crandell, 1965). Previous studies that support this theory report a proportion of experimental animals that survived a challenge with an Arctic fox isolate (Mørk and Prestrud, 2004) and that antibodies to rabies virus have been detected in Arctic foxes without any demonstrable active infection being apparent (Ballard et al., 2001). One study (Follmann et al., 1994) reported the first incidence of an unvaccinated person, a fox trapper in northern Alaska, who had never received pre- or post-exposure rabies vaccination but had acquired a rabies-specific serum neutralising antibody titre of 2.3 IU/ml, probably as a result of exposure to wildlife Arctic rabies virus. These data further support the hypothesis that strains of Arctic fox rabies virus are of lower virulence compared with certain ‘street’ strains of rabies virus circulating in other parts of the world, although this is a highly contentious issue as all strains of rabies virus are considered highly virulent (Kuzmin, 1999). If the Arctic rabies virus strain is indeed lower virulence, the incubation period for the Arctic fox strain in susceptible animals may therefore be longer and this might be a plausible reason to explain why the virus can be sustained in the animal population from one year to the next leading to sporadic outbreaks in previously rabies-free areas. Other workers have shown that Arctic fox isolates are biologically and serologically related to classical rabies virus, but that differences related to pathogenesis and virulence exist (Kantorovich, 1957). Similarly, bat variants of rabies virus may exist in a state of ‘equilibrium’ with their natural host. It is likely that host adaptation by the virus under specific environmental conditions during a period of many years has generated a lower virulent bat variant. It has been previously postulated that rabies virus persists in host populations for long periods of time, analogous to billions of virus generations, resulting in a low level of random viral mutation, which is offset by genetic selection (Wandeler et al., 1994). Similar reports of rabies virus strains with lower virulence have been reported previously, in domestic dogs; ‘oulou fato’, a strain of canine rabies from sub-Saharan Africa (Bisseru, 1972) and in wildlife (Mebatsion et al., 1992). Substantial numbers of bats in the USA and in Europe have been shown to have demonstrable antibody titres to bat rabies variants without any evidence for an active infection (Shankar et al., 2004; Brookes et al., 2005). These data suggest that virus adaptation has resulted in a strain of rabies virus with lower virulence compared to other strains of rabies virus, which through repeated exposure might lead to a subclinical viral dose and as a result generate high antibody levels to the virus, which will effectively clear the virus, with most probably ‘sterilising immunity’. Our data suggest that the range of species and geographical locations observed within the Arctic groups demonstrate the similarity of viruses found great distances apart, which could indicate rapid transmission of rabies over vast distances both within and outside of the Arctic Circle. In addition, the spread of RABV by the Arctic fox and its interaction with the domestic and sledge dog population suggests that transfer of rabies virus from sylvatic (wildlife) reservoirs to domestic animals occurs regularly. However, the direction of transfer between Arctic foxes and domestic dogs remains unknown, although the direction is most probably to be from wildlife (Arctic foxes) to the dog population. Acknowledgements The authors thank Drs. Neil Stoker and Janice Bridger (Royal Veterinary College, London, UK) in providing guidance and educational support to VR during her secondment to the VLA. VR performed this study in partial completion of an MSc in ‘Control of Infectious Disease in Animals’ and was financially supported by the Royal Veterinary College. We thank Mr. Colin Black for excellent technical support. This work was supported by an international grant from The Royal Society and from the Department for Environment, Food and Rural Affairs (Defra), UK (ROAME SE0417). K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 References Arai, Y.T., 2004. Phylogenetic analysis of two rabies viruses, Takamen and Komatsugawa strains isolated in Japan in the 1940s. Kansenshogaku Zasshi 78 (9), 815–822. Ballard, W.B., Follmann, E.H., Ritter, D.G., Robards, M.D., Cronin, M.A., 2001. Rabies and canine distemper in an Arctic fox population in Alaska. J. Wildl. Dis. 37 (1), 133–137. Bisseru, B., 1972. Rabies. William Heinemann Medical Books, London, pp. 34–35. Bohm, J., Blixenkrone-Moller, M., Lund, E., 1989. A serious outbreak of canine distemper among sledge-dogs in northern Greenland. Arctic Med. Res. 48 (4), 195–203. Bourhy, H., Kissi, B., Audry, L., Smreczak, M., Sadkowska-Todys, M., Kulonen, K., Tordo, N., Zmudzinski, J.F., Holmes, E.C., 1999. Ecology and evolution of rabies virus in Europe. J. Gen. Virol. 80, 2545–2557. Botvinkin, A., Kosenko, M., 2004. Chapter 5—rabies in the European parts of Russia, Belarus and Ukraine. In: King, A.A., Fooks, A.R., Aubert, M., Wanderler, A.I. (Eds.), Historical Perspective of Rabies in Europe and the Mediterranean Basin. OIE Publications, pp. 47–63. Brookes, S.M., Aegerter, J.N., Smith, G.C., Healy, D.M., Jolliffe, T., Swift, S.M., Mackie, I., Pritchard, J.S., Racey, P.A., Moore, N.P., Fooks, A.R., 2005. Prevalence of antibodies to European Bat Lyssavirus type-2 in Scottish bats. Emerg. Infect. Dis. 11, 572–578. Bunge, A., 1888. Medicinische Beobachtungen in Nord-Siberian/St. Petersburg Medicinische Wochenschrift 7, pp. 59–61. Crandell, R.A., 1965. Laboratory investigation of Arctic strains of rabies virus. Acta Pathol. Microbiol. Scand. 63, 587–596. Crandell, R.A., 1991. Arctic fox rabies. In: Baer, GM. (Ed.), The Natural History of Rabies, second ed. CRC Press, pp. 291–306. Daoust, P.-Y., Wanderler, A.I., Casey, G.A., 1996. Cluster of rabies cases of probable bat origin among red foxes in Prince Edward Island, Canada. J. Wildl. Dis. 32, 403–406. Felsenstein, J., 1989. PHYLIP-Phylogeny Inference Package Version 3.2. Cladistics 5, pp. 164–166. Follmann, E.H., Ritter, D.G., Baer, G.M., 1992. Oral rabies vaccination of Arctic foxes (Alopex lagopus) with an attenuated vaccine. Vaccine 10, 305–308. Follmann, E.H., Ritter, D.G., Beller, M., 1994. Survey of fox trappers in northern Alaska for rabies antibody. Epidemiol. Infect. 113, 137–141. Follmann, E.H., Ritter, D.G., Baer, G.M., 1996. Evaluation of the safety of two attenuated oral rabies vaccines, SAG1 and SAG2, in six Arctic mammals. Vaccine 14, 270–273. Hansen, E.H., 1996. Oral vaccination of Arctic foxes in Greenland: a field trial. VII Annual International Meeting: Advances Towards Rabies Control in the Americas, Atlanta, Georgia, p. 7. Jenkins, M., Wamberg, K., 1960. Rabies discovered in Greenland. J. Am. Vet. Med. Assoc. 137, 183–185. Johnson, N., McElhinney, L.M., Smith, J., Lowings, P., Fooks, A.R., 2002. Phylogenetic comparison of the genus Lyssavirus using distal coding sequences of the glycoprotein and nucleoprotein genes. Arch. Virol. 147, 2111–2123. Johnson, N., Letshwenyo, M., Baipoledi, E.K., Thobokwe, G., Fooks, A.R., 2004. Molecular epidemiology of rabies in Botswana: a comparison between antibody typing and nucleotide sequence phylogeny. Vet. Microbiol. 101, 31–38. Johnson, N., Fooks, A.R., 2005. Archival study of a Finnish isolate from the 1988/89 rabies outbreak. Arch. Virol. 150, 1407–1414. Kantorovich, R.A., 1957. The etiology of ‘madness’ in polar animals. Acta Virol. (Prague) Engl. Ed. 1, 220. Kissi, B., Tordo, N., Bourhy, H., 1995. Genetic polymorphism in the rabies virus nucleoprotein gene. Virology 209, 526–537. Kuzmin, I.V., 1999. An Arctic fox rabies virus strain as the cause of human rabies in Russian Siberia. Arch. Virol. 144, 627–629. Kuzmin, I.V., Botvinkin, A.D., McElhinney, L.M., Smith, J.S., Orciari, L.A., Hughes, G.J., Fooks, A.R., Rupprecht, C.E., 2004. Molecular epidemiology of terrestrial rabies in the former soviet union. J. Wildl. Dis. 40, 617–631. 9 Lassen, H.C.A., 1962. Paralytic human rabies in Greenland. Lancet 1, 247–249. Leisner, K., 2002. Rabies in Greenland 1975–2001. Rabies Bull. Eur. 26 (2), 10–14. MacInnes, C.D., Smith, S.M., Tinline, R.R., Ayers, N.R., Bachmann, P., Ball, D.G.A., Calder, L.A., Crosgrey, S.J., Fielding, C., Hauschildt, P., Honig, J.M., Johnston, D.H., Lawson, K.F., Nunan, C.P., Pedde, M.A., Pond, B., Stewart, R.B., Voigt, D.R., 2001. Elimination of rabies from red foxes in eastern Ontario. J. Wildl. Dis. 37, 119–132. Mebatsion, T., Sillero-Zubiri, C., Gottelli, D., Cox, J.H., 1992. Detection of rabies antibody by ELISA and RFFIT in unvaccinated dogs and in the endangered Simien jackal (Canis simensis) of Ethiopia. Zentralbl Veterinarmed B 39, 233–235. Metlin, A.E., Cox, J., Rybakov, S.S., Huovilainen, A., Grouzdev, K.N., Neuvonen, E., 2004. Monoclonal antibody characterization of rabies virus isolates from Russia, Finland and Estonia. J. Vet. Med. B Infect. Dis. Vet. Public Health 51, 94–96. Mørk, T., Prestrud, P, 2004. Arctic rabies—a review. Acta Vet. Scand. 45, 1–9. Nadin-Davis, S.A., Casey, G.A., Wandeler, A.L., 1993. Identification of regional variants of the rabies virus within the Canadian province of Ontario. J. Gen. Virol. 74, 829–837. Nadin-Davis, S.A., Casey, G.A., Wandeler, A.I., 1994. A molecular epidemiological study of rabies virus in central Ontario and western Quebec. J. Gen. Virol. 75, 2575–2583. Nadin-Davis, S.A., Sampath, M.I., Casey, G.A., Tinline, R.R., Wandeler, A.I., 1999. Phylogeographic patterns exhibited by Ontario rabies virus variants. Epidemiol. Infect. 123, 325–336. Nadin-Davis, S.A., Simani, S., Armstrong, J., Fayaz, A., Wandeler, A.I., 2003. Molecular and antigenic characterization of rabies viruses from Iran identifies variants with distinct epidemiological origins. Epidemiol. Infect. 131, 777–790. Nyberg, M., Kulonen, K., Neuvonen, E., Ek-Kommonen, C., Nuorgam, M., Westerling, B., 1992. An epidemic of sylvatic rabies in Finland—descriptive epidemiology and results of oral vaccination. Acta Vet. Scand. 33 (1), 43–57. Ødegaard, Ø.A., Krogsrud, J., 1981. Rabies in Svalbard: infection diagnosed in Arctic fox, reindeer and seal. Vet. Rec. 109, 141–142. Office Internationale Epizooties (OIE), 2002. Greenland. II. Comments on selected list B diseases. World Animal Health 1. Office Internationale Epizooties (OIE), 2004. Greenland/Rabies: multiannual animal disease status. Page, R., 1996. Treeview: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12, 357–358. Plummer, J.P.G., 1947a. Preliminary note on Arctic dog disease and it relationship to rabies. Can. J. Comp. Med. 11, 154. Plummer, J.P.G., 1947b. Further notes on Arctic dog disease and it relationship to rabies. Can. J. Comp. Med. 11, 330. Prestrud, P., Krogsrud, J., Gjertz, I., 1992. The occurrence of rabies in the Svalbard Islands of Norway. J. Wildl. Dis. 28, 57–63. Rausch, R.L., 1958. Some observations on rabies in Alaska, with special reference to wild canidae. J. Wildl. Manag. 22, 246–260. Selimov, M.A., Botvinkin, A.D., Khozinsky, V.V., Gribanova, L.Ya., 1994. New data on the spread of the P41 positive strains of rabies virus in Arctic and non-Arctic regions. J. Microbiol. Epidemiol. Immunobiol. 2, 53–57 (in Russian). Shankar, V., Bowen, R.A., Davis, A.D., Rupprecht, C.E., O’Shea, T.J., 2004. Rabies in a captive colony of big brown bats (Eptesicus fuscus). J. Wildl. Dis. 40 (3), 403–413. Sikes, R.K., 1962. Pathogenesis of rabies in wildlife. I. Comparative effect of varying doses of rabies virus inoculated into foxes and skunks. Am. J. Vet. Res. 23, 1041–1047. Sikes, R.K., 1968. Arctic rabies. Arch. Environ. Health 17, 622–626. Strimmer, K., Von Haeseler, A., 1996. Quartet puzzling: a quartet maximum likelihood method for reconstructing tree topologies. Mol. Biol. E vol. 13, 964–969. Taylor, M., Elkin, B., Maier, N., Bradley, M., 1991. Observation of a polar bear with rabies. J. Wildl. Dis. 27, 337–339. 10 K.L. Mansfield et al. / Virus Research 116 (2006) 1–10 Thompson, J.D., Higgins, D.G., Gibson, T., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680. Vanaga, S., van der Heide, R., Joffe, R., van der Poel, W.H.M., 2003. Rabies in Wildlife in Latvia. Vector-Borne Zoonotic Dis. 3, 117–124. Wandeler, A.I., Nadin-Davis, S.A., Tinline, R.R., Rupprecht, C.E., 1994. Rabies epidemiology: some ecological and evolutionary perspectives. Curr. Top. Microbiol. Immunol. 187, 297–324. Webster, W.A., Casey, G.A., Charlton, K.M., Wiktor, T.J., 1985. Antigenic variants of rabies virus in isolates from eastern, central and northern Canada. Can. J. Comp. Med. 49, 186–188. Webster, W.A., Casey, G.A., Charlton, K.M., 1986. Major antigenic groups of rabies virus in Canada determined by anti-nucleocapsid monoclonal antibodies. Comp. Immunol. Microbiol. Infect. Dis. 9, 59– 69. World Health Organization (WHO), 1990. Report of a WHO/NVI workshop on Arctic Rabies, WHO/Rab. Res/90.35, Uppsala, Sweden. Westerling, B., Andersons, Z., Rimeicans, J., Lukauskas, K., Dranseika, A., 2004. Chapter 4–rabies in the Baltics. In: King, A.A., Fooks, A.R., Aubert, M., Wanderler, A.I. (Eds.), Historical perspective of Rabies in Europe and the Mediterranean Basin. OIE Publications, pp. 33–46.