Abstract
Genetic-variant analysis of rabies viruses provides the most sensitive epidemiologic tool for following the spread and persistence of these viruses in their wildlife hosts. Since its introduction by a southern epizootic movement that began in the far north, the arctic fox (AFX) strain of Rabies virus has been enzootic in Ontario for almost 50 y. Prior genetic studies identified 4 principal genetic variants (ONT.T1 to ONT.T4) that were localized to different regions of the province; furthermore, these viruses could be distinguished from the variant circulating in northern regions of Quebec, Newfoundland, and arctic zones, ARC.T5. Despite an intensive provincial control program undertaken over the last decade that involved aerial distribution of baits laden with rabies vaccine to combat fox rabies throughout the enzootic zone of Ontario, pockets of rabies activity persist. Re-evaluation of the genetic characteristics of the viral variants circulating in these areas of persistence has been undertaken. These data demonstrate that the recent outbreaks are, with 1 exception, due to persistence of the regional variant first identified in the area in the early 1990s. In contrast, the disease in the Georgian Bay area is a consequence of the incursion of a variant previously found further south. An outbreak that occurred in northern Ontario north and west of North Bay and in the neighboring border areas of Quebec in 2000–2001 was due to renewed incursion of the ARC.T5 variant from more northerly areas.
Résumé
L’analyse des variants génétiques des virus rabiques fournit probablement l’outil épidémiologique le plus sensible pour suivre la dispersion et la persistance de ces virus chez les animaux de la faune. Depuis son introduction suite à un mouvement épizootique qui a débuté dans le grand nord et s’est dirigé vers le sud, la souche rabique du renard arctique (AFX) est enzootique en Ontario depuis presque 50 ans. Des études génétiques antérieures ont identifié 4 variants génétiques principaux (ONT.T1 à ONT.T4) qui étaient localisés dans différentes régions de la province; de plus, ces virus pouvaient être distingués du variant ARC.T5 circulant dans les régions nordiques du Québec, de Terre-Neuve, et des zones arctiques. Malgré un programme provincial intense pour limiter cette infection entrepris au cours de la dernière décennie et qui comportait la distribution aérienne d’appâts chargés de vaccin rabique pour combattre la rage du renard dans la zone enzootique de l’Ontario, des foyers d’infection persistent. Une évaluation des caractéristiques génétiques des variants viraux qui circulent dans ces zones de persistance a été entreprise. Les données démontrent que les poussées de cas récentes sont, sauf une exception, dues à la persistance du variant régional identifié au départ dans la région au début des années 1990. La maladie dans la région de la Baie Georgienne est une conséquence de l’incursion d’un variant préalablement retrouvé plus au sud. Une poussée de cas s’étant produite dans le nord de l’Ontario à l’ouest de North Bay et dans les régions avoisinantes du Québec en 2000–2001 était due à une nouvelle incursion du variant ARC.T5 des régions plus au nord.
(Traduit par Docteur Serge Messier)
Introduction
Rabies virus causes an invariably fatal encephalitis in a wide range of mammalian species, the transmission of virus from animal to animal usually being via a bite that transfers virus-laden saliva into muscle tissue (1). The virus, the type member of the genus Lyssavirus, of the family Rhabdoviridae, contains a nonsegmented, negative-sense RNA genome of about 12 kb that is organized into 5 genes (2). The 3′ proximal N gene encodes the nucleoprotein that encapsidates the viral genome, which, together with the products of the L and P genes, form the ribonucleoprotein (RNP) core of the viral particle that provides transcription and replicative functions of the viral life cycle (2). The nucleoprotein is also a useful target for antigenic methods of strain discrimination (3). The G gene product, the glycoprotein, is the sole surface protein responsible for attachment of the virus particle to the host cell receptor and host cell penetration (2). An important property of the glycoprotein is its ability to induce neutralizing antibodies and hence its use as an immunogen in vaccination strategies (4).
Although the virus can infect many animal species, each strain is perpetuated within populations of a certain species that acts as a viral reservoir (5). In Ontario, apart from sporadic outbreaks in dogs in the early 20th century, the province was free of rabies until the incursion of the arctic fox (AFX) strain of Rabies virus from northern regions (then the Northwest Territories) in the mid-1950s, as described in detail by Tabel et al (6). For the next 40 y, southern and eastern Ontario reported large numbers of rabies cases in the red fox population, which was considered to be the primary viral reservoir; in addition, significant numbers of cases were recorded in both domestic and other wild animals, events that were considered to represent spillover from the fox reservoir (7).
To improve understanding of rabies epidemiology in Ontario, genetic characterization of rabies viruses recovered from a variety of species throughout the enzootic zone was undertaken and described (8–10). In these studies, as in others (11), small but consistent genetic differences between field strains from different areas were observed. The very limited differences observed between viruses from a specific area are thought to be a consequence of the high level of adaptation required by these viruses in order to persist in their host populations; highly selective pressures apparently operate to limit viral genetic variation. Consequently, these small but consistent genetic differences between field strains can be used to track the origins of isolates responsible for infections in certain areas (9,11).
In Ontario, 4 principal viral variants (N1 to N4) were identified according to N gene variation; as well, a small group of viruses was designated as N2/4 owing to their characterization as intermediate between types 2 and 4 (8). Additional variants were evident upon G gene characterization (10). All variants were regionally localized, apparently independently of the species of origin, and variant distribution patterns appeared to be consistent with the hypothesis that landscape topography, which influences patterns of animal movement, was largely influential in determining variant spread. These observations were interpreted as indicating that conclusions regarding the role of red foxes as the principal reservoir of Rabies virus within the region were essentially correct and thus supported control efforts that targeted this population. The more recent incursion into central Ontario of a 5th variant, N5, which is typical of the viruses circulating in northern areas, was also reported (9).
In the early 1990s, with the intent of eventually eliminating rabies from southern and eastern Ontario (12), field distribution of baits containing the attenuated ERA strain of Rabies virus, which had been shown to be efficacious as an oral rabies vaccine in foxes (13), was begun in selected regions of the province (14). Since then, widespread distribution of baits has been ongoing in areas where the disease has persisted. Despite the long-term nature of these control activities and substantial reductions in numbers of rabies cases in some areas (15), the AFX strain persists in certain restricted areas of the province. Especially notable in certain counties is the high proportion of rabies-positive cases in skunks. To investigate the reasons behind the persistence of rabies in southern Ontario, despite the control strategies aimed at elimination, we have re-examined the nature of the viruses circulating in the control area. Using previously developed genetic methodologies (8), we have characterized representative viruses retrieved from the field over a 12-year period with the aim of discriminating between residual circulating viruses and new incursions of the AFX strain.
Materials and methods
Source of rabies viruses
All rabies viruses included in this study were obtained from diagnostic submissions to the Ottawa Laboratory of the Centre of Expertise for Rabies, Canadian Food Inspection Agency (CIFA). Diagnosis was performed on brain smears with use of the fluorescent antibody test (FAT) as described previously (16). In more recent years, rabies-positive specimens were subjected to antigenic typing analysis by an indirect FAT involving a panel of 16 anti-N monoclonal antibodies that exhibit differential reactivity with the various Rabies virus strains circulating in Canada (17). In this way, specimens infected with the AFX strain were identified. Representative specimens were selected for genetic analysis on the basis of location, as defined by the Universal Transverse Mercator Code (UTMC) accompanying each submission, and species of origin. Specimens collected from 1990 to 1993 have been described previously (8,9) but are included in this report to facilitate comparison with more recent isolates.
RNA extraction and N gene amplification
Total RNA was extracted from 0.1 g of brain tissue infected with Rabies virus with the use of TRIzol reagent, as described by the supplier (Invitrogen Canada, Burlington, Ontario). Final RNA pellets were dissolved in sterile ribonuclease-free water, and the RNA concentration was determined spectrophotometrically. For reverse transcription polymerase chain reaction (RT-PCR) targeting the N gene, we used 2 μg of RNA and the primers Nseq0 (5′-AACACCTCTACAATGGATGCCGAC-3′, positive sense, corresponding to bases 59 to 82 of the Pasteur virus [PV] reference strain) and RabN5 (5′-GGATTGAC[AG]AAGATCTTGCTCAT-3′, negative sense, corresponding to PV bases 1514 to 1536), as described previously (18). In a few instances, we undertook nested PCR using the conditions of the 1st-round PCR and the internal primers RabNfor and RabNrev (18) to generate a 762 base pair (bp) amplicon. The PCR products were purified with use of a Wizard PCR Preps DNA Purification System (Promega, Madison, Wisconsin, USA), according to the instructions from the manufacturer, prior to characterization.
Variant analysis
Viral variant discrimination was performed by 2 approaches: restriction endonuclease analysis (REA) and nucleotide sequencing. A panel of restriction endonucleases was used, as detailed previously (8,9), to score the presence or absence of selected sites within the PCR product of the N gene and thereby assign the isolate to 1 of 6 predefined groupings (N1 to 5 and N2/4). In some cases a reduced panel of 4 enzymes (AvaI, BamHI, BstBI, and EcoO109I) was found to be sufficient for reliable variant discrimination.
In some cases, for which either REA was not possible since the complete N gene could not be amplified or the results of REA were not definitive, direct consensus sequencing of 390 nucleotides of the N gene was undertaken by manual sequencing of the purified PCR product, as described previously (17). Phylogenetic analysis of the partial N gene sequences was thereby generated, together with reference sequences for each of the previously defined Canadian AFX strain variants, by using computer software (Clustal X Multiple Sequence Alignment Program version 1.8; National Center for Biotechnology Information, Bethesda, Maryland, USA [19]; PHYLIP [Phylogeny Inference Package], version 3.61 [20]; and TreeView [21]), as detailed previously (17). This analysis allowed the grouping of all isolates into a particular variant type.
Mapping of cases
The UTMC associated with each specimen was converted to latitude and longitude coordinates, which were used to place the specimens in maps created using computer software (ArcView 8, ArcGIS Desktop product; ESRI Canada, Toronto, Ontario).
Statistical analysis
Analysis of variance (ANOVA) was performed by using computer software (SigmaStat software, version 2.03; Systat Software, Point Richmond, California, USA).
Results
Case reporting for 1990 to 2004
Table I summarizes the numbers of rabies cases reported for eastern Canada by the Ottawa Laboratory of the CIFA Centre of Expertise for Rabies for the 15-year period of 1990 to 2004. These numbers reflect all diagnostic submissions from eastern Canada, almost to the western border of Ontario with Manitoba and including areas of the eastern arctic. However, most submissions originate within the area of Canada traditionally regarded as the enzootic region of Ontario and neighboring border areas of Quebec.
Table I.
Numbers of animal submissions from eastern Canada per year to the Ottawa Laboratory of the Centre of Expertise for Rabies, Canadian Food Inspection Agency, that were positive for rabies by the fluorescent antibody test from 1990 to 2004
| Number of positive cases (and total number of submissions) per year
|
|||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Species | 1990 (10 096) | 1991 (9881) | 1992 (10 372) | 1993 (10 183) | 1994 (7654) | 1995 (6434) | 1996 (5965) | 1997 (9418) | 1998 (6376) | 1999 (7725) | 2000 (14 733) | 2001 (12 610) | 2002 (11 308) | 2003 (11 992) | 2004 (11 165) |
| Bat | 26 | 49 | 28 | 24 | 18 | 22 | 25 | 31 | 31 | 19 | 45 | 67 | 92 | 88 | 63 |
| Red fox | 1148 | 926 | 1157 | 944 | 322 | 136 | 80 | 33 | 17 | 16 | 41 | 42 | 48 | 25 | 8 |
| Striped skunk | 356 | 388 | 286 | 227 | 145 | 101 | 74 | 40 | 34 | 47 | 63 | 62 | 46 | 22 | 25 |
| Other domestic | 539 | 451 | 463 | 409 | 220 | 90 | 54 | 33 | 12 | 17 | 24 | 26 | 18 | 17 | 21 |
| Other wild | 61 | 70 | 72 | 45 | 26 | 4 | 5 | 0 | 1 | 11 | 50 | 95 | 31 | 17 | 8 |
| Total number (and %) positive | 2130 (21.10) | 1884 (19.07) | 2006 (19.34) | 1649 (16.19) | 731 (9.55) | 353 (5.49) | 238 (3.99) | 137 (1.45) | 95 (1.49) | 110 (1.42) | 223 (1.51) | 292 (2.32) | 235 (2.08) | 169 (1.41) | 125 (1.12) |
The yearly number of positive cases has dropped substantially, especially from 1994 onwards, reflecting the apparent success of the provincial oral vaccination program in reducing the incidence of rabies in foxes. Although viral typing regimens have been in routine use only in the later years of this period, the fox was considered the principal terrestrial host of the Rabies virus for most of this timeframe. Cases in bats are due to viral variants that are quite distinct from the viruses that circulate in terrestrial species (17), and these 2 reservoirs are generally considered to be quite independent, although spillover of bat strains into terrestrial animals is occasionally noted (22). The increase in numbers of cases in other wild species from 1999 to 2003 was due primarily to 2 outbreaks in raccoons that crossed into Canada from the United States, first into Ontario in July 1999 (23) and second into New Brunswick in 2000 (24). Also notable are the relative proportions of cases in red foxes and skunks. In the early 1990s, before the oral vaccination program was widely applied, cases in skunks generally made up 14% to 21% of all terrestrial reports. In 1990 to 1992, 3231 (67.9%) out of 4758 diagnostic submissions for foxes were positive, in contrast to 1030 (45.3%) out of 2275 for skunks. In 1997 to 1999, only 66 (1.6%) out of 4163 submissions for foxes were positive, whereas the skunk positivity rate had dropped only to 14.8% (121 out of 816 submissions). A 1-way ANOVA found these differences to be statistically significant (P < 0.05 for all pairwise comparisons). Thus, the ratio of positive cases in red foxes to those in skunks was 3.14 in 1990 to 1992 but only 0.545 in 1997 to 1999, despite the fact that red-fox submissions were similar in number over the 2 periods, whereas the number of skunk submissions dropped by two-thirds. These data indicate that a much higher proportion of cases in skunks were being reported in the late 1990s. This trend has continued up to the present. Although in 2002 to 2003, similar numbers of cases were reported in skunks and red foxes, in 2004 only 8 red foxes but 25 skunks were recorded as rabies-positive. Other species affected, owing to spillover events, include cats, dogs, cattle, sheep, pigs, goats, and horses (domestic) and wolves, coyotes, a bear, buffalo, and red deer (wild).
Variant analysis
To explore whether there was any change in the nature of the viral variants circulating in Ontario over the 12-year period of 1990 to 2001, we genetically characterized isolates, selecting rabies-positive specimens primarily on the basis of host species (to include red foxes, skunks, and other commonly reported wild and domestic animals) and spatial distribution. Table II summarizes data for the 279 AFX isolates characterized with respect to the genetic variant of AFX. Usually isolates were typed by a single method (either REA or nucleotide sequencing), but in all cases in which both methods were used there was complete concordance in the typing assignment.
Table II.
Variants of the arctic fox (AFX) strain of Rabies virus identifieda in 279 rabies-positive specimens over a 12-year period
| Number of isolates per year
|
||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Rabies virus variant | Species group | 1990 | 1991 | 1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | Total numbers of isolates per year |
| N1 | Fox | 2 | 12 | 14 | 2 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 31 |
| Skunk | 2 | 5 | 3 | 1 | 0 | 0 | 1 | 0 | 4 | 0 | 2 | 2 | 20 | |
| Wild | 0 | 1 | 2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 4 | |
| Domestic | 0 | 1 | 2 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | |
| Subtotal | 4 | 19 | 21 | 4 | 0 | 0 | 3 | 0 | 4 | 0 | 2 | 2 | 59 | |
| N2 | Fox | 7 | 6 | 2 | 2 | 0 | 4 | 2 | 0 | 3 | 0 | 2 | 2 | 30 |
| Skunk | 5 | 6 | 2 | 4 | 1 | 3 | 5 | 3 | 17 | 4 | 6 | 10 | 66 | |
| Wild | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | |
| Domestic | 1 | 1 | 0 | 0 | 0 | 2 | 5 | 0 | 7 | 4 | 1 | 4 | 25 | |
| Subtotal | 14 | 14 | 4 | 6 | 1 | 9 | 12 | 3 | 27 | 8 | 9 | 16 | 123 | |
| N3 | Fox | 2 | 5 | 0 | 17 | 0 | 0 | 0 | 0 | 1 | 0 | 2 | 0 | 27 |
| Skunk | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 10 | 1 | 7 | 2 | 24 | |
| Wild | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Domestic | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 4 | 1 | 0 | 0 | 6 | |
| Subtotal | 4 | 7 | 0 | 18 | 0 | 0 | 0 | 0 | 15 | 2 | 9 | 2 | 57 | |
| N4 | Fox | 4 | 4 | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 10 |
| Skunk | 4 | 2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 7 | |
| Wild | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | |
| Domestic | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Subtotal | 9 | 6 | 0 | 2 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 18 | |
| N2/4 | Fox | 1 | 1 | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 |
| Skunk | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | |
| Wild | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Domestic | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Subtotal | 2 | 1 | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 5 | |
| N5 | Fox | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 7 | 4 | 12 |
| Skunk | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Wild | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 3 | |
| Domestic | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 0 | 2 | |
| Subtotal | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 9 | 6 | 17 | |
| All variants | 33 | 49 | 25 | 32 | 1 | 10 | 15 | 3 | 46 | 10 | 29 | 26 | 279 | |
Identified as the AFX strain by a process now routinely undertaken with the use of a panel of monoclonal antibodies that discriminate the major Rabies virus strains circulating in Ontario. Genetic analysis, by either restriction endonuclease analysis or nucleotide sequencing, was used to assign each isolate to a specific AFX variant
To illustrate the phylogenetic relationships between the variants, nucleotide sequencing was performed on 50 AFX isolates, selected as representing the 279 isolates on the basis of the date of recovery over the 12-year period, region of origin, variant type, and species infected (Table III). A 390-bp sequence window (corresponding to bases 287 to 676 of PV [GenBank accession number NC_001542]) was determined for each isolate, and all sequences, including the PV strain as an outgroup, were aligned by means of Clustal X. Bootstrap resampling (1000 replicates) of the resulting output file was performed with use of the seqboot program (PHYLIP, version 3.61), followed by neighbor-joining (NJ) analysis of the resulting file with use of the dnadist, neighbor, and consense programs (PHYLIP). Distance values were reapplied to the final consensus tree by means of the fitch program (PHYLIP), and the final output was converted to a graphical tree by using computer software (TreeView). Six N gene variant reference sequences, ARC.T5 and ONT.T1 to ONT. T5, which were reported previously (9), were included for comparison; all additional isolates are described in Table III.
Table III.
Details of the 50 arctic fox (AFX) isolates examined by nucleotide sequencing and phylogenetic analysis
| Isolate number | Date received (y-mo-d) | Host species | County and province of origin | GenBank accession # |
|---|---|---|---|---|
| 6199.91 | 1991-08-13 | Red fox | Lambton ON | DQ063227 |
| 0173.93 | 1993-01-12 | Striped skunk | Waterloo ON | DQ063228 |
| 2244.93 | 1993-04-14 | Red fox | Perth ON | DQ063229 |
| 3870.95 | 1995-08-10 | Striped skunk | Grey ON | DQ063230 |
| 0507.96 | 1996-02-13 | Striped skunk | Victoria ON | DQ063231 |
| 2790.96 | 1996-06-28 | Striped skunk | Peterborough ON | DQ063232 |
| 5649.97 | 1997-04-29 | Striped skunk | Grey ON | DQ063233 |
| 0337.98 | 1998-02-06 | Bovine | Huron ON | DQ063234 |
| 0713.98 | 1998-03-17 | Striped skunk | Peterborough ON | DQ063235 |
| 1784.98 | 1998-06-08 | Striped skunk | Peterborough ON | DQ063236 |
| 3207.98 | 1998-07-24 | Striped skunk | Bruce ON | DQ063237 |
| 0250.99 | 1999-01-28 | Striped skunk | Grey ON | DQ063238 |
| 0251.99 | 1999-01-28 | Bovine | Grey ON | DQ063239 |
| 0417.00 | 2000-02-08 | Red fox | Cochrane ON | DQ063240 |
| 0466.00 | 2000-02-08 | Striped skunk | Grey ON | DQ063241 |
| 0583.00 | 2000-02-22 | Striped skunk | Dufferin ON | DQ063242 |
| 0934.00 | 2000-03-17 | Red fox | Wellington ON | DQ063243 |
| 1035.00 | 2000-03-28 | Red fox | Territoire-du-Nouveau-Québec QC | DQ063244 |
| 1246.00 | 2000-04-11 | Equine | Grey ON | DQ063245 |
| 1278.00 | 2000-04-13 | Striped skunk | Bruce ON | DQ063246 |
| 1512.00 | 2000-05-02 | Red fox | Cochrane ON | DQ063247 |
| 1774.00 | 2000-05-19 | Red fox | Abitibi QC | DQ063248 |
| 2365.00 | 2000-06-20 | Red fox | Nipissing ON | DQ063249 |
| 2847.00 | 2000-07-11 | Red fox | Perth ON | DQ063250 |
| 3420.00 | 2000-08-03 | Dog | Territoire-du-Nouveau-Québec QC | DQ063251 |
| 4261.00 | 2000-09-06 | Striped skunk | Peterborough ON | DQ063252 |
| 4337.00 | 2000-09-08 | Red fox | Territoire-du-Nouveau-Québec QC | DQ063253 |
| 4456.00 | 2000-09-15 | Red fox | Cochrane ON | DQ063254 |
| 4820.00 | 2000-10-04 | Red fox | Cochrane ON | DQ063255 |
| 4947.00 | 2000-10-11 | Striped skunk | Waterloo ON | DQ063256 |
| 5419.00 | 2000-10-31 | Striped skunk | Peterborough ON | DQ063257 |
| 0053.01 | 2001-01-04 | Striped skunk | Waterloo ON | DQ063258 |
| 0297.01 | 2001-01-25 | Wolf | Cochrane ON | DQ063259 |
| 0522.01 | 2001-02-13 | Coyote | Cochrane ON | DQ063260 |
| 0623.01 | 2001-02-20 | Red fox | Abitibi-Ouest QC | DQ063261 |
| 0644.01 | 2001-02-21 | Striped skunk | Wellington ON | DQ063262 |
| 0668.01 | 2001-02-22 | Bovine | Dufferin ON | DQ063263 |
| 1954.01 | 2001-04-20 | Red fox | Sudbury ON | DQ063264 |
| 3447.01 | 2001-04-25 | Red fox | Minganie–Basse-Côte-Nord QC | DQ063265 |
| 5884.01 | 2001-06-05 | Striped skunk | Perth ON | DQ063266 |
| 7043.01 | 2001-07-04 | Striped skunk | Grey ON | DQ063267 |
| 10248.01 | 2001-09-12 | Striped skunk | Wellington ON | DQ063268 |
| 10294.01 | 2001-09-13 | Red fox | Bruce ON | DQ063269 |
| 10974.01 | 2001-10-18 | Striped skunk | Huron ON | DQ063270 |
| ARC.T5 | 1993-02-26 | Dog | Igloolika NWT | U03769 |
| ONT.T1 | 1991-03-12 | Striped skunk | Renfrew ON | L20673 |
| ONT.T2 | 1991-04-25 | Striped skunk | York ON | L20674 |
| ONT.T3 | 1991-02-06 | Red fox | Kent ON | L20675 |
| ONT.T4 | 1990-11-20 | Red fox | Grey ON | L20676 |
| ONT.T5 | 1991-11-05 | Red fox | La Reine QC | U03768 |
ON — Ontario; QC — Quebec; NWT — Northwest Territories
Community of origin
Five distinct clades are evident in the resulting dendogram (Figure 1); 4 are supported with bootstrap values greater than 700 out of 1000 replicates, a level that is considered significant for analysis of RNA viruses. The most distinct clade is for the N5 variant, typical of the viruses circulating in northern regions of Canada (9), and it clearly is outlying the other AFX isolates in the analysis. The next most closely related variant is N1, which has been found exclusively in eastern Ontario. The other clades define variants N3 and another grouping that includes all N2 and N4 variants and 2 isolates designated as N2/4 because of their intermediate REA pattern. The close relationship of the N2, N4, and N2/4 variants in this analysis explains the difficulties encountered in their discrimination by means of REA. Despite the long period over which these samples were collected, there is no apparent trend showing that the more recent sequences have emerged from sequences representative of earlier isolates. The maximum nucleotide distance between any 2 AFX isolates was 0.042 over the targeted sequence window, and in most cases the value was much lower (data not shown). Thus, the difference between these isolates is very small, clearly a reflection of the high level of constraint that operates on this segment of the viral genome.
Figure 1.
Phylogenetic tree of 50 Canadian specimens of the arctic fox (AFX) strain of Rabies virus with 6 N gene variant reference sequences, ARC.T5 and ONT.T1 to ONT.T5, included for comparison. Bootstrap values for all main groupings are indicated above the branch immediately to the left of the group. The dotted line for the Pasteur virus (PV) reference strain has been shortened to allow better presentation of the AFX groupings. A genetic distance scale is at the bottom left.
Spatial and temporal distribution of Rabies virus variants
The data in Table II summarize, for each of the variants, the number of AFX isolates found in the 4 species groupings. The same data are depicted in the maps in Figure 2 to show the spatial distribution of these viral variants over 4 periods: 1990 to 1993 (a), 1994 to 1997 (b), 1998 to 1999 (c), and 2000 to 2001 (d). The data in Figure 2a are a compilation of the results presented previously (8,9) and are provided here for comparison with our later findings. From 1994 onwards, the reduced number of cases reported in many counties limited coverage of the entire study area. Review of the data provided in these maps and in Table II leads to the following observations.
Figure 2.
Locations of recovery of Rabies virus variants in 1990 to 1993 (a), 1994 to 1997 (b), 1998 to 1999 (c), and 2000 to 2001 (d). Viral variants N1 to N5 are represented by diamonds, triangles, circles, stars, and squares, respectively. Black indicates red fox, white indicates skunk, and grey indicates all other species.
The N4 variant, previously found south of Georgian Bay and down the eastern shore of Lake Huron, was found in only 1 sample in this area in 1995 and not at all thereafter. The N2 variant appears to have spread into the area, including Grey, Bruce, and Simcoe counties, that initially harbored variant N4.
The small numbers of cases in eastern Ontario still represent the N1 variant that has traditionally infected this area, but since 1998 all N1 isolates identified have been in skunks.
In more recent years, a substantial proportion of cases, especially in Perth, Wellington, Dufferin, Huron, and Grey counties, which harbor a mixture of N2 and N3 variants, have been in skunks and species other than red foxes.
The 1990 to 1993 outbreak in the northern part of the study area (north and west of North Bay) was due primarily to westward and northward movement of the N1 and N2 variants from eastern and southern Ontario, respectively. However, 2 isolates from that period were identified as N5 and were thus believed to represent a wave of disease spread from the north. The 2000 to 2001 outbreak in the same area was due exclusively to the N5 variant and again appears to be due to southward movement from more northerly regions. Most of the specimens infected with this variant were from red foxes or other wild species.
Discussion
This genetic-variant analysis of field isolates of the AFX strain of Rabies virus permits the following conclusions regarding the epidemiology of rabies in Ontario between 1990 and 2001. Essentially the variants that were identified in the early 1990s persist in Ontario, albeit over much reduced areas in most cases, despite the intensive, provincially funded oral vaccination program that has targeted the red fox population. The N1 variant, which was distributed throughout eastern Ontario at the start of the study period, now persists in a much smaller region, in the county of Peterborough. By 2000 the N2 and N3 variants still occupied several counties in southern Ontario, most notably Grey, Bruce, Wellington, and Dufferin, but a few reports came from counties further south. The N4 variant had not been found since 1995 and appears to have disappeared from the counties around the Georgian Bay area, where it was widely distributed in the early 1990s. The closely related N2 variant appears to have taken the place of N4 in much of this region. The enzootic area of Ontario has thus retained the virus strains that have been present since the early 1990s, and failure to eliminate rabies was due to failure to break the cycle of disease transmission rather than to any incursion from other areas. Indeed, the only area newly invaded, as indicated by the exclusive presence of the N5 variant, which is typical of northern isolates, was the northern portion of the study area, a region west and north of North Bay, in 2000 to 2001. This observation, together with earlier evidence of southwards movement of the AFX strain into this region (9), emphasizes the ongoing threat of reinfection of southern and eastern Ontario by this rabies strain if surveillance and control activities are not maintained over the long term.
With regard to continued maintenance of the AFX strain in southern Ontario, the observation that skunks played a more prominent role in case numbers in the late 1990s and early 2000s may be significant. In many regions of North America, Rabies virus strains having quite distinct evolutionary origins have adapted to the striped skunk (25,26), and this species is regarded as a significant reservoir for this virus. Furthermore, the relative ease with which new long-term virus:host associations can arise in the skunk was suggested by the persistence and sustained transmission of a bat rabies variant in a local US population of skunks in Flagstaff, Arizona in 2001 (27). The role of skunks in helping to maintain the AFX strain in Ontario had been considered previously, but there had been little solid evidence. The data presented here, which clearly indicate long-term persistence of rabies in Ontario during a period of intense efforts at control in the red fox, coupled with the observation that recent cases are occurring predominantly in skunks in the controlled area, strongly suggest that skunks play a significant role in maintaining the AFX strain in Ontario. Currently, local control of rabies in skunks is possible by trap–vaccinate–release (TVR) activity (28), although this approach is too labor-intensive and costly for application to large areas. Since the oral vaccine currently used in the fox rabies control program is not highly efficacious in skunks, as we (unpublished data) and others (29,30) have observed, elimination of rabies in Ontario may require development of novel vaccines more suitable for oral immunization of the skunk.
Acknowledgments
The authors thank Geoff Turner and Martine Sampath for excellent technical assistance and Andrea Bernhoff for generation of the maps.
References
- 1.Charlton KM. The pathogenesis of rabies. In: Campbell JB, Charlton KM, eds. Rabies. Boston, Massachusetts: Kluwer, 1988:101–150.
- 2.Tordo N, Charlton K, Wandeler A. Rhabdoviruses: rabies. In: Collier LH, ed. Topley & Wilson’s Microbiology and Microbial Infections. London: Arnold Press, 1998:666–692.
- 3.Rupprecht CE, Dietzschold B, Wunner WH, Koprowski H. Antigenic relationships of lyssaviruses. In: Baer GM, ed. The Natural History of Rabies. 2nd ed. Boca Raton, Florida: CRC Press, 1991:69–100.
- 4.Wiktor TJ, Macfarlan RI, Reagan KJ, et al. Protection from rabies by a vaccinia virus recombinant containing the rabies virus glycoprotein gene. Proc Natl Acad Sci USA. 1984;81:7194–7198. doi: 10.1073/pnas.81.22.7194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Niezgoda M, Hanlon CA, Rupprecht CE. Animal rabies. In: Jackson AC, Wunner WH, eds. Rabies. San Diego, California: Academic Press, 2002:163–218.
- 6.Tabel H, Corner AH, Webster WA, Casey GA. History and epizootiology of rabies in Canada. Can Vet J. 1974;15:271–281. [PMC free article] [PubMed] [Google Scholar]
- 7.MacInnes CD. Rabies. In: Novak M, Baker JA, Obbard ME, Malloch B, eds. Wild furbearer management and conservation in North America. North Bay, Ontario: Trappers Association, 1987: 910–929.
- 8.Nadin-Davis SA, Casey GA, Wandeler A. Identification of regional variants of the rabies virus within the Canadian province of Ontario. J Gen Virol. 1993;74:829–837. doi: 10.1099/0022-1317-74-5-829. [DOI] [PubMed] [Google Scholar]
- 9.Nadin-Davis SA, Casey GA, Wandeler AI. A molecular epidemiological study of rabies virus in central Ontario and western Quebec. J Gen Virol. 1994;75:2575–2583. doi: 10.1099/0022-1317-75-10-2575. [DOI] [PubMed] [Google Scholar]
- 10.Nadin-Davis SA, Sampath MI, Casey GA, Tinline RR, Wandeler AI. Phylogeographic patterns exhibited by Ontario rabies virus variants. Epidemiol Infect. 1999;123:325–336. doi: 10.1017/s0950268899002885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Smith JS, Fishbein DB, Rupprecht CE, Clark K. Unexplained rabies in three immigrants in the United States: a virologic investigation. N Engl J Med. 1991;324:205–211. doi: 10.1056/NEJM199101243240401. [DOI] [PubMed] [Google Scholar]
- 12.MacInnes CD. Eliminer la rage du sud de l’Ontario. Med Vet Que. 1998;28:115–118. [Google Scholar]
- 13.Lawson KF, Hertler R, Charlton KM, Campbell JB, Rhodes AJ. Safety and immunogenicity of ERA strain of rabies virus propagated in a BHK-21 cell line. Can J Vet Res. 1989;53:438–444. [PMC free article] [PubMed] [Google Scholar]
- 14.Rosatte RC, MacInnes CD, Power MJ, et al. Tactics for the control of wildlife rabies in Ontario (Canada) Rev Sci Tech. 1993;12:95–98. doi: 10.20506/rst.12.1.670. [DOI] [PubMed] [Google Scholar]
- 15.MacInnes CD, Smith SM, Tinline RR, et al. Elimination of rabies from red foxes in Eastern Ontario. J Wildl Dis. 2001;37:119–132. doi: 10.7589/0090-3558-37.1.119. [DOI] [PubMed] [Google Scholar]
- 16.Dean DJ, Ableseth MK, Atanasiu P. The fluorescent antibody test. In: Meslin FX, Kaplan MM, Koprowski H, eds. Laboratory Techniques in Rabies. 4th ed, Geneva: WHO, 1996:88–95.
- 17.Nadin-Davis SA, Huang W, Armstrong J, et al. Antigenic and genetic divergence of rabies viruses from bat species indigenous to Canada. Virus Res. 2001;74:139–156. doi: 10.1016/s0168-1702(00)00259-8. [DOI] [PubMed] [Google Scholar]
- 18.Nadin-Davis SA. Polymerase chain reaction protocols for rabies virus discrimination. J Virol Methods. 1998;75:1–8. doi: 10.1016/s0166-0934(98)00106-2. [DOI] [PubMed] [Google Scholar]
- 19.Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997;25:4876–4882. doi: 10.1093/nar/25.24.4876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Felsenstein J. PHYLIP: Phylogeny Inference Package. Seattle, Washington: University of Washington, 1993.
- 21.Page RDM. TREEVIEW: an application to display phylogenetic trees on personal computers. Comput Appl Biosci. 1996;12:357–358. doi: 10.1093/bioinformatics/12.4.357. [DOI] [PubMed] [Google Scholar]
- 22.Daoust P-Y, Wandeler AI, Casey GA. Cluster of rabies cases of probable bat origin among red foxes in Prince Edward Island, Canada. J Wildl Dis. 1996;32:403–406. doi: 10.7589/0090-3558-32.2.403. [DOI] [PubMed] [Google Scholar]
- 23.Wandeler AI, Salsberg E. Raccoon rabies in eastern Ontario. Can Vet J. 1999;40:731. [PMC free article] [PubMed] [Google Scholar]
- 24.MacInnes CD. Raccoon rabies in New Brunswick. Rabies Reporter. 2000;11((3)):10. [Google Scholar]
- 25.Smith JS, Orciari LA, Yager PA. Molecular epidemiology of rabies in the United States. Semin Virol. 1995;6:387–400. [Google Scholar]
- 26.Nadin-Davis SA, Abdel-Malik M, Armstrong J, Wandeler AI. Lyssavirus P gene characterisation provides insights into the phylogeny of the genus and identifies structural similarities and diversity within the encoded phosphoprotein. Virology. 2002;298:286–305. doi: 10.1006/viro.2002.1492. [DOI] [PubMed] [Google Scholar]
- 27.Smith JS, Rohde R, Mayes B, Parmely C, Leslie MJ. Molecular evidence for sustained transmission of a bat variant of rabies virus in skunks in Arizona. Presented at the 12th International Meeting on Advances in Rabies Research and Control in the Americas, 2001 Nov 12–16, Peterborough, Ontario.
- 28.Rosatte RC, Power MJ, MacInnes CD, Campbell JB. Trap–vaccinate–release and oral vaccination for rabies control in urban skunks, raccoons and foxes. J Wildl Dis. 1992;28:562–571. doi: 10.7589/0090-3558-28.4.562. [DOI] [PubMed] [Google Scholar]
- 29.Tolson ND, Charlton KM, Lawson KF, Campbell JB, Stewart RB. Studies of ERA/BHK-21 rabies vaccine in skunks and mice. Can J Vet Res. 1988;52:58–62. [PMC free article] [PubMed] [Google Scholar]
- 30.Rupprecht CE, Charlton KM, Artois M, et al. Ineffectiveness and comparative pathogenicity of attenuated rabies virus vaccines for the striped skunk (Mephitis mephitis) J Wildl Dis. 1990;26:99–102. doi: 10.7589/0090-3558-26.1.99. [DOI] [PubMed] [Google Scholar]