SciELO - Scientific Electronic Library Online

vol.43 issue5Polymerase chain reaction (PCR) for the detection of Salmonella in artificially inoculated chicken meatExperimental infection and horizontal transmission of Bartonella henselae in domestic cats author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista do Instituto de Medicina Tropical de São Paulo

On-line version ISSN 1678-9946

Rev. Inst. Med. trop. S. Paulo vol.43 no.5 São Paulo Sept./Oct. 2001 



Adriana Marques JOPPERT(1), Mitika Kuribayashi HAGIWARA(2) & Natalino Hajime YOSHINARI(3)




Dogs sera samples collected from Cotia County, São Paulo were tested using indirect immunoenzymatic test (ELISA) in order to study Lyme disease serology in dogs. ELISA method was standardized and G39/40 North American strain of Borrelia burgdorferi was used as antigen. Positive results were confirmed employing the Western blotting technique. Because of the possibility of cross-reactions, sera were also tested for different serological strains of Leptospira interrogans and L. biflexa using microscopic sera agglutination test. Twenty-three of 237 (9.7%) serum samples were positive in the ELISA; 20 of them (86.9%) were confirmed by the Western blotting, what suggests that Cotia may be a risk area for Lyme disease. Although 4 samples (1.7%) were positive for Lyme disease and leptospirosis, no correlation was found between the results (X2 = 0.725; p = 0.394) what suggests absence of serological cross reactivity.

KEYWORDS: Lyme disease; Borrelia burgdorferi; Dogs; ELISA; Brazil.




Lyme disease is a multisystemic zoonotic disease caused by a tick-borne spirochete Borrelia burgdorferi16. The disease in humans typically begins with a skin rash named erythema migrans, and is often associated with flu-like symptoms that may be followed weeks to months later by cardiac, neurological, and joint disease22.

Many Borrelia strains have been isolated from Ixodes sp. ticks, reservoir animals and Lyme disease patients in many parts of the world. B. burgdorferi sensu latu, that was first thought to be a homogeneous species, is classified into new genospecies: B. burgdorferi sensu strictu isolated in North America and Europe, B. garinii and B. afzelli in Europe, B. japonica in Japan, and B. andersonii in North America2.

Borrelia burgdorferi has been isolated and identified in many different animal species10, yet clinical manifestations of the disease have been reported in dogs, horse and cattle6,7,12. Affected dogs may exhibit a variety of clinical signals including fever, lethargy, lymphoadenopathy, kidney disorders, heart block and recurrent polyarthritis, the latter being the most frequent manifestation in seropositive animals12. Due to a lack of pathognomonic clinical signals, serology has been widely used for the diagnosis of canine borreliosis, and has been a helpful tool for epidemiological surveys. Indirect fluorescent antibody test (IFA) and enzyme immunosorbent assay (ELISA) are used in humans and dogs to test for exposure to Borrelia burgdorferi, and for confirmation, Western blotting assay8.

In areas where Lyme disease is endemic, the prevalence of antibodies to B. burgdorferi in dogs is higher than in co-resident human population9. This probably is because dogs are more exposed to the vectors, which remain longer on the host making easier the transmission of the spirochetes. For this reason, dogs are considered as sentinel animals, as a means of determining the prevalence of Lyme disease in a geographic area.

In Brazil, there are reports of clinical cases in humans patients,24,26,27,28 and Cotia County in São Paulo State was the first region where disease was identified in brothers, and the diagnosis was confirmed by the presence of B. burgdorferi antibodies revealed by ELISA and Western blotting tests. Seroepidemiological study performed in this area25 revealed that about 7.5% of the people were seropositive, similar prevalence observed at risk areas in North America and Europe15 .

Despite of the known existence of human infection, no data was available for dogs disease in the same area, so to fulfil this gap, the seroprevalence of B. burgdorferi in dogs was studied, using indirect ELISA technique, and confirmation of positive samples by the Western blotting method.



Sera: During anti-rabies vaccination campaign in Cotia County (São Paulo, Brazil), 237 sera samples from dogs of any age, gender and breed, owned near the woods were collected, from August 18th to September 1st, 1992. Informations related to previous or recent contact with ticks were obtained for all dogs.



Antigen: For antigen preparation it was used Borrelia burgdorferi G39/40 strain of North American origin, maintained at Laboratório de Investigação em Reumatologia da Faculdade de Medicina da Universidade de São Paulo. The preparation of whole sonicated spirochetes suspension was made as previously described3. Briefly, microorganisms were grown in Kelly's modified medium for approximately one week at 33 °C. At late-log-phase growth, organisms were pelleted by centrifugation at 12000 g for 20 min at 4 °C, washed with a solution of cold PBS 0.001M MgCl2 6H2O PBS (pH 7.4) and sonicated on ice for 3 min with 15 sec intervals with a cell sonicator (Sonic Dismembrator Model 300, Dynatech Lab, Inc).

The suspension was filtered (0.45 mm diameter pore) and the protein concentration was determined by the method of Folin. The antigen preparation was stored in aliquots at -70 °C until use.

Positive control serum: To obtain a positive control serum, a dog without serum antibodies against Borrelia burgdorferi (Canine B. burgdorferi antibody test kit, Cite) and L. interrogans (Microscopic sera agglutination test) was inoculated twice, with a 7 days interval, with a heat killed (56 °C – 30 min) B. burgdorferi suspension at a dose of 0.1 mg/Kg of body weight.

The antigen suspension was prepared as above described but was not submitted to sonication, and 0.5 ml of the suspension was homogenizated with 0.5 ml of complete Freund's adjuvant and inoculated subcutaneously.

Serum samples were collected 4, 7, 10, 14, 21, 36, 44, 58 and 64 days after the first inoculation and tested for the presence of antibodies to B. burgdorferi using ELISA and Western blotting methods. Sample collected on day 36 was chosen as the standard positive control. Antibodies against L. interrogans were tested by microscopic sera agglutination assay.

Negative control sera: Sera from eight dogs, free for ticks and negative for Borrelia (Canine B. burgdorferi antibody test kit, Cite) and Leptospira antibodies (microscopic agglutination test) were used as negative controls.

ELISA: The ELISA was performed by standard method23 with modifications. Briefly, the microtitrations plates (EIA/RIA plate, Costar) were coated with 200 ml of a 0.015 mg/ml solution of the spirochete proteins diluted in 0.05M sodium carbonate (pH 9.6) and incubated overnight at 4 °C. The plates were then washed three times with PBS containing 0.05% Tween 20 and blocked with PBS Tween 20 and 5% skimmed milk (pH 7.4). The wash procedure was repeated and test samples and control sera were diluted in blocking solution and plated in duplicate (200 ml/well). The samples and negative control sera were diluted 1/400 and the positive control serum was diluted using serial two ratio dilutions starting on 1/400. After incubation at room temperature for 1 hour, the wash procedure was repeated and the conjugate was added (alkaline phosphatase conjugated rabbit serum anti-dog IgG, Sigma) diluted 1/1000 in blocking solution. After incubation and wash procedure was added the substrate p-nitrophenyl sodium phosphate substrate (Sigma) diluted at 1 mg/ml in glycine buffer (pH 10.5). The plates were read at 405 nm in a Titertek Multiscan MCC/340 (Flow Laboratories), when the first dilution of the positive control serum reached the optical density value near 1.0.

Cut off value was obtained by considering the mean optical density plus three standard deviations of the eight negative controls sera. Optical densities higher than cut off values were considered positive and the titers were estimated by regression curve.

Western blotting (WB): Electrophoresis and immunoblotting were done as previously described with modifications14. Briefly, 750 mg of spirochete proteins, reduced with SDS, was electrophoresed (Mini-Protean II System, Bio Rad) on a 10% acrylamide gel. After running, gel proteins were transferred to nitrocellulose paper over night (Mini-Trans Blot System, Bio Rad), and the paper cut in strips, which were washed in distilled water. One strip containing Borrelia burgdorferi antigens and other with molecular weight markers were separated and stained with a solution 1:1 of colloidal gold (Bio Rad). The others strips were blocked with 0.1% TBS Tween 20 and 5% skimmed milk for 1 hour at room temperature and then washed five times with 0.1% TBS Tween 20. The sera samples to be tested (positive and negative controls, and test sera), were diluted 1:100 in blocking solution and added to incubate with strips for 1 hour at room temperature. After washing, the strips were incubated for 1 hour with alkaline phosphatase-conjugated rabbit serum anti-dog IgG (Sigma Chemical) diluted at 1/1000 with blocking solution. Washing was repeated and substrate consisting of NBT/BCIP diluted in bicarbonate buffer pH 9.8 was added. The reaction was blocked when positive control developed color.

Microscopic sera agglutination reaction: Study on antibodies to the main sorovars of Leptospira interrogans and L. biflexa was performed in the Laboratório de Zoonoses do Departamento de Medicina Veterinária Preventiva e Saúde Animal da Faculdade de Medicina Veterinária e Zootecnia da Universidade de São Paulo, using the microscopic sera agglutination technique11.

Statistical analysis: The chi-square test was used in order to evaluate the correlation between positive results for Lyme disease and positive results for leptospirosis and to verify the presence of significant differences between distribution of positive animals, using the indirect ELISA test, according to gender, age and history of contact with ticks.



A total of 23 (9.7%) of the 237 serum samples analysed by ELISA had antibodies to Borrelia burgdorferi at a dilution of 1/400 or higher (Table 1). Although a higher frequency (69.6%) was found on males when compared to females (30.4%) no difference among them could be find (p = 0.417). Also, no difference on the frequency of reagents was observed related to age of dogs (p = 0.256) (Table 2). On the other hand, positive results were found mainly in dogs with history of previous contact with ticks (p = 0.034) (Table 3).







When ELISA positive sera were submitted to Western blotting, all samples except three (86.9%) presented reactivity to more than five bands, according to GREENE et al., 199113 standardization confirming the presence of Borrelia burgdorferi antibodies (Fig. 1).



The microscopic seroagglutination test for leptospira antibodies, showed that 24 of 237 sera (10.1%) had positive reaction, with titer ranging from 1/100 and 1/400 for several serovars of Leptospira interrogans. Only four samples (1.7%) of the 237 gave positive serology for Lyme disease and leptospirosis. No correlation (p = 0.394) between both diseases regarding cross reactivity could be found.



Lyme disease is an emerging disease in Brazil, and seems to be endemic in some areas where ticks and wild animals are present, close to human residents.

Cotia county is located near from São Paulo city and has a vast wooded area named Atlantic Forest. Epidemiological studies performed at this region, where first cases of Lyme disease in humans had been described25, revealed presence of potential vectors4 such as ticks of species Ixodes loricatus, I. didelphidis and Amblyomma cajennense. Spirochete like organisms were identified in the blood culture of wild animals and ticks, however these microorganisms did not grow longer in BSK medium1.

Seroepidemiological surveillance done at Cotia county revealed a prevalence of 7.5% of positive results for Borrelia burgdorferi antibodies, suggesting that residents of the area had previous contact with spirochetes25 .

The prevalence found here in dogs (9.7%), slightly higher than the human's one, confirmed the possibility of transmission of B. burgdorferi sensu latu to dogs, although no clinical case of Lyme disease had been observed. The low prevalence of reagents, when compared to some endemic areas in USA5, where as high as 53% of dogs were found to be positive for B. burgdorferi, might be due to a low rate of infected ticks. Also, in areas where Lyme disease is endemic, the seroprevalence for Lyme disease in dogs is higher than the one found in co-resident human population9, and low prevalence in dogs was found around the world, where human cases of disease are rarely described.

The incidence of B. burgdorferi infection in dogs20 is closely related to the ticks density in a given area, the rate of tick infection and outdoor or indoor habits of dogs. Also, to the sensitivity and specificity of the surveillance method employed17.

A decrease in the sensitivity of the method may have occurred because of the antigen employed. It is known that there are antigenic differences among B. burgdorferi sensu latu strains isolated around the world19. In this study G39/40 strain of B. burgdoferi of north American origin was used, because native isolate was not isolated in Brazil yet.

During introduction of ELISA test for B. burgdorferi antibodies search in dogs, some difficulties were found. Because of not existence of positive control serum, a dog previously known to be free of Borrelia sp antibodies tested by commercial ELISA was inoculated with killed microorganism plus Freund's adjuvant twice. The use of alive spirochetes was avoided to prevent any risk of bacterial dissemination to the environment, despite of employment of a non virulent strain of Borrelia burgdorferi.

The successful production of specific antibodies was confirmed by ELISA and WB methods, and curiously the immunoblotting showed different humoral reactivity between the serum obtained from sensitized dog and positive test samples from the region of Cotia. This data suggested the presence of a different etiological agent in Brazil.

As WB is a more specific method for Borrelia burgdorferi antibodies detection, it was used to confirm the results of ELISA test of samples and controls. It was considered positive WB, when the assay showed reactivity to at least 5 bands13. Twenty of 23 (86.9%) positive test samples at ELISA were confirmed by Western blotting assay, revealing good correlation between these methods. The remaining three samples and eight negative controls sera were considered negative for B. burgdorferi antibodies by WB analysis.

The possibility of cross reaction between Borrelia burgdorferi and others spirochetes is likely18. At the present work four sera (1.7%) of the dogs had presented simultaneous antibodies against both infections, but no correlation could be found.

SOARES in 199821 studied presence of IgG class antibodies to B. burgdorferi and Babesia canis in 150 dogs sera from Itaguaí, Rio de Janeiro, Brazil, and using ELISA test standardized at same way, obtained similar results, when it was found 30 positive samples (20.0%) for borrelia antigens, and did not find serological cross-reactivity between these diseases.

Interestingly, low titers of antibodies to known borrelias such us B. burgdorferi, B. garinii and B. afzelli are obtained with sera of Brazilian patients with Lyme disease, becoming difficult the diagnosis of this illness in our country26,27,28. WB pattern is also quite different, suggesting the presence of another etiological agent. Spirochete like microorganisms are seen in peripheral blood and cerebrospinal fluid of patients with borreliosis, however are uncultivable and not identified by PCR assays1,28. The present work done in dogs from risk area for Lyme disease, confirms data obtained in humans, supporting the idea of existence of a slightly different clinical entity in Brazil, which has been named as Lyme like disease.




Pesquisa de anticorpos anti- Borrelia burgdorferi em cães da região de Cotia, São Paulo, Brasil

Com a finalidade de estudar a ocorrência da doença de Lyme em cães no Brasil, o teste imunoenzimático (ELISA) indireto, utilizado para o diagnóstico da doença no homem, foi padronizado para a espécie canina. Utilizou-se como antígeno a cepa americana de Borrelia burgdorferi G 39/40. Os soros de cães procedentes da região de Cotia, área de risco para ocorrência da doença de Lyme, foram testados e os resultados positivos foram confirmados através da técnica de "Western blotting". Para investigação de possíveis reações cruzadas, os soros foram também testados para diferentes variantes sorológicas de Leptospira interrogans e L. biflexa pela técnica de soroaglutinação microscópica. Dos 237 soros testados pela técnica de ELISA, 23 (9,7%) foram positivos, sendo que 20 (86,9%) destes soros foram também positivos pela técnica de "Western blotting", sugerindo que a região de Cotia seja área de risco para doença de Lyme. Embora 4 (1,7%) dos soros testados apresentassem positividade para doença de Lyme e para leptospirose concomitantemente, não foi observada associação entre estes resultados (X2 = 0,725 e p = 0,394), sugerindo ausência de reatividade sorológica cruzada entre as duas enfermidades.




We would like to thank Dr. Allen C. Steere, from the Department of Rheumatology and Immunology at TUFTS School of Medicine, Boston, USA, for the G39/40 strain of Borrelia burgdorferi; Dr. Leonardo José Ritchtzenhain, from the Laboratório de Imunologia do Departamento de Medicina Preventiva e Saúde Animal da Faculdade de Medicina Veterinária e Zootecnia da Universidade de São Paulo for helping to prepare the antigen to be inoculated and Dr. Silvio de Arruda Vasconcelos from the Laboratório de Zoonoses do Departamento de Medicina Preventiva e Saúde Animal da Faculdade de Medicina Veterinária e Zootecnia da Universidade de São Paulo for performing the microscopic seragglutination test.

This study was supported by Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP)



1. ABEL, I.S.; MARZAGÃO, G.; YOSHINARI, N.H. & SCHUMAKE, T.T.S. - Borrelia-like spirochetes recovery from ticks and small mammals (Acari:Ixodidae) collected in the Atlantic Forest Reserve, Cotia County, State of São Paulo, Brazil. Mem. Inst. Oswaldo Cruz., 95: 621-624, 2000.        [ Links ]

2. BARANTON, G.; POSTIC, D.; SAINT-GIRONS, I. et al - Delineation of Borrelia burgdorferi sensu stricto, Borrelia garinii sp nov., and group VS461 associated with Lyme borreliosis. Int. J. system. Bact., 42: 378-383, 1992.        [ Links ]

3. BARBOUR, A.G. - Isolation and cultivation of Lyme disease spirochetes. Yale J. Biol. Med., 57: 521-525, 1984.        [ Links ]

4. BARROS-BATTESTI, D.M.; YOSHINARI, N.H.; BONOLDI, V.L.N. & GOMES, A.C. - Parasitism by Ixodes didelphidis and I. loricatus (Acari:Ixodidae) on small wild mammals from an Atlantic Forest in State of São Paulo, Brazil. J. med. Ent., 37: 820-827, 2000.        [ Links ]

5. BURGESS, E.C. - Natural exposure of Wisconsin dogs to the Lyme disease spirochete (Borrelia burgdorferi). Lab. Anim. Sci., 36: 288-290, 1986.        [ Links ]

6. BURGESS, E.C.; GENDRON-FITZPATRICK, A. & WRIGHT, W.O. - Arthritis and systemic disease caused by Borrelia burgdorferi infection in a cow. J. Amer. vet. med. Ass., 191: 1468-1470, 1987.        [ Links ]

7. COHEN, N.D. & COHEN, D. - Borreliosis in horses: a comparative review. Comp. contin. Educ. [Equine], 12: 1449-1458, 1990.        [ Links ]

8. CRAFT, J.E.; GRODZICKI, R.L. & STEERE, A.C. - Antibody response in Lyme disease: evaluation of diagnostic tests. J. infect. Dis., 149: 789-795, 1984.        [ Links ]

9. ENG, T.R.; WILSON, M.L.; SPIELMAN, A. & LASTAVICA, C.C. - Greater risk of Borrelia burgdorferi infection in dogs than in people. J. infect. Dis., 158: 1410-1411, 1988.        [ Links ]

10. EUZÉBY, J.P. - Borrelia burgdorferi et la maladie di Lyme chez les animaux. Revue générale. Rev. Méd. vét., 140: 371-388, 1989.        [ Links ]

11. GALTON, M.M.; SULZER, C.R.; SANTA ROSA, C.A. & FIELDS, M. - Application of a microtechnique to the agglutination test for leptospiral antibodies. Appl. Microbiol., 13: 81-85, 1965.        [ Links ]

12. GREENE, R.T. - Canine Lyme borreliosis. Vet. Clin. N. Amer. small Anim. Pract., 21: 51-64, 1991.        [ Links ]

13. GREENE, R.T.; HIRSCH, D.A.; ROTTMAN, P.L. & GERIG, T.M. - Interlaboratory comparison of titer of antibody to Borrelia burgdorferi and evaluation of a commercial assay using canine sera. J. clin. Microbiol., 29: 16-20, 1991.        [ Links ]

14. GRODZICK, R.L. & STEERE, A.C. - Comparison of immunoblotting and indirect enzyme-linked immunosorbent assay using different antigen preparations for diagnosing early Lyme disease. J. infect. Dis., 157: 790-797, l988.        [ Links ]

15. HANRAHAN, J.P.; BENACH, J.L.; COLEMAN, J.L et al. - Incidence and cumulative frequency of endemic Lyme disease in a community. J. infect. Dis., 150: 489-496, 1984.        [ Links ]

16. JOHNSON, R.C.; SCHMID, G.P.; HYDE, F.W.; STEINGERWALT, A.G. & BRENNER, D.J. - Borrelia burgdorferi sp. nov.: etiologic agent of Lyme disease. Int. J. system. Bact., 34: 496-497, 1984.        [ Links ]

17. LINDENMAYER, J.M.; WEBER, M.; BRYANT, J.; MARQUEZ, E. & ONDERDONK, A.B. - Comparison of indirect immunoflurescent-antibody assay, enzyme-linked immunosorbent assay, and Western immunoblot for the diagnosis of Lyme disease in dogs. J. clin. Microbiol., 28: 92-96, 1990.        [ Links ]

18. MAGNARELLI, L.A.; ANDERSON, J.F. & JOHNSON, R.C. - Cross-reactivity in serological test for Lyme disease and other spirochetal infections. J. infect. Dis., 156: 183-188, 1987.        [ Links ]

19. MAGNARELLI, L.A.; ANDERSON, J.F.; JOHNSON, R.C.; NADELMAN, R.B. & WORMSER, G.P. - Comparison of different strains of Borrelia burgdorferi sensu lato used as antigens in enzyme-linked immunosorbent assays. J. clin. Microbiol., 32: 1154-1158, 1994.        [ Links ]

20. RAND, P.W.; SMITH, R.P. & LACOMBE, E.H. - Canine seroprevalence and the distribution of Ixodes dammini in an area of emerging Lyme disease. Amer. J. pub. Hlth., 81: 1331-1334, 1991.        [ Links ]

21. SOARES, C.O. - Estudo da borreliose canina: imunodiagnóstico, seroepidemiologia e interação imunológica com Babesia canis. Rio de Janeiro, 1998. (Dissertação de Mestrado - Faculdade de Medicina Vaterinária da Universidade Federal Rural do Rio de Janeiro).        [ Links ]

22. STEERE, A.C. - Lyme disease. New Engl. J. Med., 321: 586-596, 1989.        [ Links ]

23. VOLLER, A.; BIDWELL, D.E. & BARTLETT, A. – Enzyme immunoassays in diagnostic medicine: theory and practice. Bull. Wld. Hlth. Org., 53: 55-65, 1976.        [ Links ]

24. YOSHINARI, N.H.; BARROS, P.J.L.; CRUZ, F.C.M. et al. - Clínica e sorologia de doença de Lyme no Brasil. Rev. bras. Reum., 32 (supl. 4): 57, 1992.        [ Links ]

25. YOSHINARI, N.H.; BARROS, P.J.L.; STEERE, A.C. et al. - Epidemiological study of Lyme disease in Cotia (São Paulo, Brasil). Rev. esp. Reum., 20 (supl. 6): 1993.        [ Links ]

26. YOSHINARI, N.H.; BARROS, P.J.L.; BONOLDI, V.L.N. et al. - Perfil da borreliose de Lyme no Brasil. Rev. Hosp. Clín. Fac. Med. S. Paulo, 52: 111-117, l997.        [ Links ]

27. YOSHINARI, N.H.; BONOLDI, V.L.N. & BARROS-BATTESTI, D.M. - Doença de Lyme–símile no Brasil. Rev. bras. Reum., 39: 57-58, l999.        [ Links ]

28. YOSHINARI, N.H.; BARROS, P.J.L. & GAUDITANO, G. - Report of 57 cases of Lyme-like disease (LLD) in Brazil. Arthr. and Rheum., 43 (suppl.): S188, 2000.        [ Links ]


Received: 23/11/1999
Accepted: 10/09/2001



(1) Veterinarian, Divisão Técnica de Medicina Veterinária e Biologia da Fauna, Secretaria Municipal do Meio Ambiente da Prefeitura Municipal de São Paulo, São Paulo, SP, Brasil.
(2) Full Professor, Departamento de Clínica Médica. da Faculdade de Medicina Veterinária e Zootecnia da Universidade de São Paulo, São Paulo, SP, Brasil.
(3) Associate Professor, Departamento de Clínica Médica da Faculdade de Medicina da Universidade de São Paulo, São Paulo, SP, Brasil.

Correspondence to: Natalino Hajime Yoshinari, MD, PhD; Faculdade de Medicina da Universidade de São Paulo, Disciplina de Reumatologia, Av. Dr. Arnaldo 455, 3° andar, 01246-903 São Paulo, SP, Brazil.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License