411

JUUfl(Il of %Vil(flsfe J)h s s :34(2). PP 411-414© \‘ildlit,’ I)is’a.se’ Asssiami,m 199

Serologic Survey for Antibodies to Borrelia burgdorferi inWhite-tailed Deer in Ontario

G. James GalIlvan,’ Ian K. Barker,27 Harvey Artsob,3 Louis A. Magnarelli,4 Jeffrey T. Robinson,5 and DennisR. Mgt,61i 19 Guelph Street, Apartment 5, Guelph, Ontario, Canada NiH 5 ; 2 Department of Pathology, OntarioVeterinary College, University of Guelph, Guelph, Ontario, Canada Ni G 2W1; 3Zoonotic Diseases, National Lab-oratory for Special Pathogens, Bureau of Microbiology, Laboratory Centre for Disease Control, Tunney’s Pasture,Ottawa, Ontario, Canada K1A 0L2; Department of Entomology, The Connecticut Agricultural Experimental Station,P.O. Box 1106, New Haven, Connecticut 06504, USA; Canadian Wildlife Service, 465 Gideon Drive, P.O. Box490, Lambeth Station, London, Ontario, Canada N6P 1 Ri; 6Ontano Ministry of Natural Resources, 300 WaterStreet, Peterborough, Ontario, Canada K9J 8M5; 7Corresponding author for reprint requests.

ABSTRACT: Serum samples collected from 623white-tailed deer (Odocoilens virginianus) insouthern Ontario (Canada) from 1985 to 1989were tested for antibodies to Borrelia burgdor-I en using an indirect fluorescent antibody(IFA) staining method. Samples from 150 ofthe deer were also tested using an enzyme-linked immunosorbent assay (ELISA). At IFAtiters of 1 :64 and 1: 128 deer with antibodies toB. burgdoiferi appeared to be widespreadthroughout southern Ontario, with an apparentprevalence ranging from 3 to 47%. At IFA ti-tres 1:256 and ELISA titres 1:160 deer withantibodies to B. burgdorfrri were only presenton Long Point which is the only known endem-ic focus of Ixodes scapularis, the primary vectorfor B. burgdorferi, in southern Ontario. Atthese titres the apparent prevalence of antibod-ies to B. burgdorferi on Long Point was only 5to 7%, even though the mean intensity of in-festation of adult I. scapularis on deer was>180, and 60% of the adult ticks are infectedwith B. burgdo feri. Based on these results,white-tailed deer do not appear to be a goodsentinel species for the distribution of B. burg-dorferi.

Key words: Borrelia burgdorferi, Lyme dis-ease, Odocoileus virginianus, sentinel species,seroprevalence, survey, white-tailed deer.

White-tailed deer (Odocoileus virgini-anus) are the definitive host of Ixodes sca-pularis which is the primary vector forBorrelia burgdoiferi, the etiologic agent ofLyme disease in eastern North America.Several studies (Magnarelli et at., 1984a,1984b, 1986, 1991; Gill et at., 1993; Mahn-ke et at., 1993) have reported antibodiesto B. burgdorferi in white-tailed deer fromthe eastern United States, and it has beensuggested that serological surveys in white-tailed deer may be useful for determiningthe distribution of B. burgdo7feri (Gill etat., 1993; Mahnke et al., 1993). In the

present paper, we report the results of aserological survey for antibodies to B.burgdorferi in white-tailed deer from thesouthern part of the province of Ontario(Canada). The only known endemic focusfor I. scapularis in southern Ontario is onLong Point (44#{176}34’N; 80#{176}10’W) on thenorth shore of Lake Erie (Barker et at.,1992). However, Lyme borreliosis hasbeen reported in Ontario residents whohave not travelled to areas where I. sca-pularis is endemic (C. LeBer, pers. com-mun.), and seropositive dogs which havenot travelled to I. scapularis endemic areashave also been reported (Artsob et at.,1993). Our aim was to determine if anti-bodies to B. burgdotferi were widespreadin white-tailed deer in southern Ontario.

Blood samples were collected fromwhite-tailed deer at check stations duringthe fall hunting season, from animals im-mobilized during the winter months forstudies on deer ecology, and from animalseuthanized following motor vehicle colli-sions, in various locations throughoutsouthern Ontario (Fig. 1) from 1985 to1989. Serum was frozen at -70 C and lat-er analyzed. All of the samples were ana-lyzed using an indirect fluorescent anti-body assay (IFA) starting at a dilution of1:32 as described by Artsob et at. (1993).One hundred and fifty samples were alsoanalyzed using an enzyme-linked immu-nosorbent assay (ELISA) (Magnarelli etat., 1991). One hundred and twelve of

these samples were from Long Point. Theremaining 38 samples, which included allof the samples with IFA titres 1:64, werefrom other locations.

412 JOURNAL OF WILDLIFE DISEASES, VOL. 34, NO. 2, APRIL 1998

St’t’ Fig. I Immr locatiomus.Nmmmmmlxr(ml (leer.

FI(;URE 1 . NIap of southern Oustano sho ving thelx.’ations Irons which sera ere collected fromim vhite-taile(l (leer. The loc’ations are ( 1 ) Long Point National\Vildlife Area (44#{176}:34’N; 80 l0’\V), (2) Rondeamu Pro-viuscial Park (42#{176}17’N: 81#{176}51’\\’), (3) 1 oint Pelee Na-tional Park (41#{176}57’N; 82#{176}3l‘\\‘), (4) \ ‘ingharn l)istrict(4:3#{176}53’N; 8l0l9\ ), (5) Maple l)istrict (43#{176}51’N;7903 1 ‘\V). (6) 1 Immronia l)istrict (44#{176}27’N; 79#{176}44’W).(7) Somuth (: tm10mmt() Township (45#{176}05’N: 76 50’\\’), (8)

North Camuonto Township (45#{176}10’ N : 76#{176}5:3’W). and

(9) L()rimug I)istrict (45#{176}56’N; 80#{176}00’\\).

The results from the IFA at three titrationend-points are shown in Table 1. The de-creasing prevalence with increasing end-points is similar to the pattern reported byother authors (Magnarelli et at., 1984a,1984b, 1986; Mahnke et at., 1993). PointPelee (Fig. 1) had the highest prevalence atan endpoint of 1:64, but Long Point was theonly site with seropositive deer at an end-point of 1:256. Six of the 112 (5%) samplesfrom Long Point were sero positive using

ELISA, with titres 1:640. Five of the sam-ples which were seropositive using ELISAhad IFA litres 1:256, as did three of theELISA negative samples (titres <1:160).None of the 38 samples from other locationswere seropositive using ELISA, although 20had IFA titres 1:64.

At IFA titres <1:256, antibody to B.burgdorferi appears to be widespreadthroughout southern Ontario, but it is onlycommon at Point Pelee and Long Point.The occurrence of deer with low antibodytitres in areas outside of the known focusof I. scapularis may be caused by trans-mission from nymphs disseminated on mi-grating birds (Battaly et at., 1987). It mayalso be caused by cross-reaction with otheragents, such as Leptospira interrogans.Antibodies to B. burgdotferi will cross-re-act with other spirochetes in both IFA andELISA (Magnarelli et at., 1986, 1987), andmay also cross-react with other bacterialproteins (Hansen et at., 1988). L. interro-gans serovars pomona, grippotyphosa, andicteroherno rrhagiae, which cross-reactwith B. burgdoiferi at dilutions of 1:64 and1 : 128 (Magnarelli et at., 1986), have beenreported from white-tailed deer in south-ern Ontario (Abdulla et at. , 1962).

At an IFA titre 1:256 and ELISA titre 1:160, the only seropositive deer were onLong Point. This correlates with the ob-servation that Long Point is the only en-

TALSI.E I . Prevalence (‘ / ) of titres to tluc aumtil)odies to Borrelia burg(lo7fcrm deterunined b indirect ininitu-nofliuorescemmt assay fromuu white-tailed (leer at (lifferent locations in southern Ontario’.

IAmemtiomms \‘ear Momuths mmt’

Titre

1:64 1: 12S 1:2.56

Long Point 1987-89 Oct/Nov 116 15 11 7Rommdeamm 1986-87 Jan/Feb 37 14 0 0Point Pelee 1989 Jan-May 19 37 11 0\\‘inghammm 1985-87 Nov-Feb 40 5 5 0Maple 1985-86 Feb-Jrulv 9 11 11 0Ilmmromuia 1985-87 Nov-Apr 70 7 4 0Sommtlu Camsomito 1985-86 Nov 33 3 0 0North (: trmmmto 1985-86 Nov 84 5 2 0Lorimig 1986-87 Oct-Aug 215 4 1 0

SHORT COMMUNICATIONS 413

demic focus of 1. scapularis in southernOntario (Barker et at., 1992). The low

prevalence (5 to 7%) on Long Point is alsoconsistent with the results reported in oth-er studies. At an IFA titre 1:256, the ap-parent prevalence of antibodies to B. burg-dorferi ranged from 6 to 18% in Lyme dis-ease endemic areas in Connecticut be-tween 1980 and 1984 (Magnarelli et at.,1984a, 1984b, 1986). Mahnke et at. (1993)reported a prevalence of 4% in the barrierislands off Georgia at the highest end-point of their ELISA test, and Gill et at.(1993) reported a prevalence of 9% in anI. scapularis endemic area in Minnesotausing ELISA, although they later reporteda prevalence of 71% in the same area (Gillet at., 1994). Gill et at. (1993, 1994) re-

ported prevatences up to 5% in areas with-

out endemic tick populations, and Mahnkeet at. (1993) also reported a prevalence of4% in areas without endemic tick popula-tions at their lowest end-point.

The low apparent prevalence of anti-bodies to B. burgdoiferi in white-taileddeer on Long Point is somewhat surprising

given that all of the deer are infested withadult 1. scapularis from October to De-cember with a mean intensity of infesta-tion >180 ticks (Watson and Anderson,1976; Lindsay, 1995), and that approxi-mately 60% of the unfed adult ticks areinfected with B. burgdotferi (Lindsay etat., 1991). Experimentally-infected deer

develop an antibody response to B. burg-dorferi (Gill et at., 1993; Mahnke et at.,1993; Lane et at., 1994; Luttrell et at.,1994), but the intensity of the antibody re-sponse appears to vary between the JD-1and SH-2 strains (Luttrell et at., 1994),and the response to the same strain mayvary widely between deer (Lane et at.,1994). Thus, the low apparent prevalenceof antibodies in deer on Long Point maybe caused in part by the strains of B. burg-dorferi which are present.

Another factor causing the low apparentprevalence of antibodies to B. burgdorferiin white-tailed deer on Long Point may bethe timing of the antibody response. Mag-

narelli et al. (1995) reported the highestprevalence of antibodies from February to

April. The antibody response to B. burg-dorferi in experimentally-infected deertakes 3 wk to develop and appears to bedecreasing by 10 wk (Lane et at., 1994;Luttrell et al. , 1994). White-tailed deer onLong Point support large numbers of lar-vae and adult I. scapularis, but fewnymphs (Watson and Anderson, 1976;Lindsay, 1995). The prevalence ofB. burg-do,feri is <0.2% in larvae (Lindsay et a!.,1991) which are found on deer fromMarch to August (Watson and Anderson,1976), and the prevalence in nymphs,which are found on deer from April to Au-gust, is 17%. Adult ticks are found on deerfrom October to April. Most of the deerfrom Long Point were collected in Octo-ber (Table 1). Thus, deer which had beenexposed to B. burgdoiferi from the adultticks may not have had time to develop adetectable antibody response. The preva-lence of IFA titres did not differ betweenfawns, yearlings and adults ( 4.34; P= 0.36; Zar, 1974), thus it is doubtful thatthe antibody response resulted from ex-posure the previous year.

Based on the results of this study, white-tailed deer do not appear to be a good sen-tinel species for serologic surveys for B.

burgdoiferi. The occurrence of seroposi-tive deer with IFA titres of 1:64 and 1:128in areas without endemic populations of I.scapularis suggests that the tests are notsufficiently specific when cross-reactingbacteria may be present in the deer herd.The low apparent prevalence at higherend-points on Long Point where deerwere heavily infested with adult ticks, 60%of which are infected with B. burgdorferi,suggests that white-tailed deer do not de-velop a sufficiently strong antibody re-sponse to compensate for the loss of sen-sitivity associated with an increased speci-ficity at higher cutoff values, even whenthey are heavily parasitized by ticks.

Financial support for this study was pro-vided by Health and Welfare Canada, theOntario Ministry of Health and the Uni-

414 JOURNAL OF WILDLIFE DISEASES, VOL. 34, NO. 2, APRIL 1998

Received for publication 24 July 1997.

versity of Guelph Research EnhancementFund. This work would not have been pos-sible without the collaboration of the Ca-nadian Wildlife Service personnel in LongPoint National Wildlife Area, and the co-operation of the hunters who collected andcontributed sera on Long Point and for theOntario Ministry of Natural Resources,

and the OMNR staff who assisted withcapturing deer.

LITERATURE CITED

ABDuLL , I’ K.. L. KARSTAD. AND N. A. FIsu. 1962.Ciultiural an(1 serological evidence of leptospirosisimi (leer in Ontario. Canadian Veterinary Journal3: 71-78.

AHTSOB. 11., I. K. BARKER. R. FISTER, C;. SEPHTON,

1). l)IcK, J. A. LYNCH, AND 1). KEY. 1993. Se-rological stiudies on the infection of dogs in On-tario with Borrelia I)mirg(Iorferz, the etiological

agent of Lvnme disease. Canadian VeterinaryJournal 34: 543-548.

BARKER, 1. K., (;. A. SURCEONER, H. ARTSOB, S. A.

MCEWEN, L. A. ELLIoTT, G. D. CAMPBELL, AND

J. T. RoBINsoN. 1992. i)istribution ofthe Lymedisease vector, Ixo(/c’s (lallz?nini (Acari: Ixodidae)and isolation of Borrelia burgdorfc’ri in Ontario,

(:antda. Jotunmal of Medical Entomology 29:1011-1022.

BATrAIS. C. R., 1). FIsH, AND R. C. I)OwLER. 1987.The seasonal occurremice of Ixodes (iWluhlifli andIXO(/(’.S (1t’flt(ltU.S (Acari: Ixodidae) on 1)irds in aLvrne disease endemic area in sotutheastern NewYork state. Journal of the New York Entomolog-ical Society 95: 461-468.

(;I1.I., J. S., R. C. MCLEAN, D. F NEITzEL, AND R.C. JOHNSON. 1993. Serologic analysis of white-tailed (leer sera for antibodies to Borrelia burg-(IO?fCfl I)y enzyme-linked immiunosorbent assayand vesterim immunoblotting. Journal of ClinicalMicrobiology 31: 318-322.

R. B. SHRINER, AND R. C. JoHNSON.

1994. Serologic surveillance for the Lynie diseasespirochete, Borrelia burgdorferi, in Minnesota by

musing white-tailed deer as sentinel animals. Journal

of Clinical Microbiology 32: 444-451.

hANSEN, K.. J. M. BANGSBORG, H. FJORDVANC, N.

S. PEDERSEN, ANt) P. HINDERSSON. 1988. luru-mtunochemical characterization of and isolation

of the gene for a Borrdia burgdoifrrm iunnmuno-doniinamit 60-kilodaltomi antigen common to awude range of bacteria. Infection and Immunity

56: 2047-2053.LANE, R. S.. I). P BERGER, L. E. CASI-IER, AND V.

BURCDORFER. 1994. Experimental infection ofCol,unmbian black-tailed (leer with the Lvmmme (us-

ease spirochete. Journal of Wildlife Diseases 30:20-28.

LINDSAY, L. R. 1995. Factors limitimig the (liStribu-tion of the blacklegged tick, Iu)des .scapularz.sSay, in Ontario, Canada. Ph.D. Thesis, Universityof Guelph, Gtuelph, Ontario, 219 pp.

, I. K. BARKER, C. A. SURGEONER, S. A.

MCEwEN, L. A. ELLIorr, AND J. KOLAR. 1991.

Apparent incompetence of Dermacentor caria-bilLs (Acan: lxodidae) an(l fleas (Imisecta: Siphomi-aptera) as vectors of Borrelia hurgdorferi in an

IXO(1CS (la?llllZiHi endemic area of Ontario, Can-ada. Journal of Medical Entomology 28: 750-753.

LUTTRELL, M. P., K. NAKACAKI, E. W II0wERTH, D.

E. STALLKNECHT, AND K. A. LEE. 1994. Exper-imental infectiomi of Borrelia hurgdorferi inwhite-tailed deer. Journal of Wildlife Diseases

30: 146-154.MAGNARELLI, L. A., J. F. ANDERSON, AND W. A.

CHAPPELL. 1984a. Antibodies to spirochetes inwhite-tailed (leer and prevalence of infected ticksfrom foci of Lyme disease in Connecticut. Jotur-nal of Wildlife Diseases 20: 2 1-26.

, , W. BURGDORFER, AND W. A. CHAP-

PELL. 1984b. Parasitismn by lxo(le.s (l000nini (Ac-ari: Ixodidae) and antibodies to spirochetes inmammals at Lyme disease foci imi Connecticut,

USA. Journal of Medical Entomology 21: 52-57., , C. S. APPERSON, I). FISH, R. C.

JOHNSON, AND W A. CHAPPELL. 1986. Spiro-chetes in ticks an(l antibodies to Borrelia Imrg-(k)1:ti’ri in white-tailed deer from Connecticut,New York State, and North Carolina. Journal ofWildlife Diseases 22: 178-188.

, , AND R. C. JOHNSON. 1987. Cross-reactivity in serological tests for Lyme diseaseand spirochetal infections. The Journal of Infec-tious Diseases 156: 183-188.

, J. H. OLIVER, JR., H. J. FIuTCIIES0N, ANI) J.F. ANDERSON. 1991. Antibodies to Borreliahurgdorferi iui deer and raccoomis. Journal ofWildlife Diseases 27: 562-568.

, A. DENICOLA, K. C. STAFFORD III, AND J.F. ANDERSON. 1995. Borrelia hurgdorferi in anurban environmeuit: white-tailed deer with in-fected ticks and antibodies. Journal of ClinicalMicrobiology 33: 54 1-544.

MAHNKE, G. L., D. L. STALLKNECHT, C. E. GREEENE,

V. F. NETTLES, AND M. A. MARKS. 1993. Sero-

logic survey for antibodies to Borrelia hurgdor-feri in Georgia. Journal of Wildlife Diseases 29:230-236.

WATSON, T. C;., AND R. C. ANDERSON. 1976. Ixodesscapular-is Say on white-tailed deer (Odocoileuscirginiaiuu.s) from Long Point, Ontario. Journalof Wildlife Diseases 12: 66-7 1.

ZAR. J. H. 1974. Biostatistical analysis. Prentice-Hall,Englewood Cliffs, New Jersey, 620 pp.