Neurologic Abnormalities in Two Dogs Suspected Lyme Disease

Microbiol. Immunol., 37(4), 325-329, 1993 Neurologic Abnormalities in Two Dogs Suspected Lyme Disease Yuzo Azuma *,1, Katsuzi Kawamura1, Hiroshi Isogai2, and Emiko Isogai3 1 Monami Animal Clinic, Sapporo, Hokkaido 005, Japan, 2Division of Animal Experimentation, Sapporo Medical College, Sapporo, Hokkaido 060, Japan Gakuen University, Hokkaido 061-02, Japan, and 3Department of Preventive Dentistry, Higashi Nippon Received August 26, 1992; in revised form, January 12, 1993. Accepted January 14, 1993 ônh ô Abstract ôns ô: A 2-year-old mongrel dog developed neurological signs following tick bite. These includ edastasia, persistent tonic convulsions and hyper-reflexia. Both serum IgG and IgM antibody titers against ônh ôborrelia burgdorferi ôns ô were positive in enzyme-linked immunosorbent assay (ELISA). The neurological signs subsided after high-dose penicillin and streptomycin treatment. A strain of spirochetes (P427a) was isolated from the midgut of ônh ôixodes persulcatus ôns ô feeding on the dog. Morphological characteristic, immunological property and protein profile revealed that the isolate was ônh ôb. burgdorferi. ôns ô Similarly, a 2-year-old Labrador retriever dog developed neurological signs after tick bite and showed a positive IgG antibody titer against ônh ôb. burgdorferi. ôns ô Antibiotic treat mentwas effective also in this case. These findings suggest that neurological symptoms shown in both dogs were caused by infection with ônh ôb. burgdorferi. ôns ô ô Key words ôns ô: Neuro-Lyme disease, Dog Lyme disease is a tick-borne spirochetal disease with clinical manifestations including dermatitis, myocarditis, neuritis and/or arthritis (19). Initial reports of Lyme disease in dogs described arthritis as the sole clinical manifestation (1, 12-14), but recent reports described skin lesions, lameness, vomiting, abortion and/or renal disease (5, 7, 15). Neurological manifestations have not been report edeither in experimental or natural canine infec tionswith B. burgdorferi in the U.S.A. (7). Recent ly,meningitis in dogs attributed to B. burgdorferi infection has been reported in Europe (Morner, A. P., Olson, P., Carlsson, M., Olersbacken, M.: Clini caland serological studies of borreliosis in Swedish dogs. Presented at the Fourth Interna tionalconference on Lyme Borreliosis; June 18-21, 1990, Stockholm, Sweden). In the two cases of tick infestation reported here, the dogs developed high serum antibody titers to B. burgdorferi and showed neurologic abnormal ities.a bacterial isolate from Ixodes persulcatus found on the dog was morphologically and im munologicallycharacterized as B. burgdorferi. The first case (case 1), a 2-year-old male mongrel dog weighing 10.4kg, was presented to Monami *Address correspondence to Dr. Yuzo Azuma, Monami Animal Clinic, 3-3-4-28, Kawazoe, Minami-ku, Sapporo, Hokkaido 005, Japan. Animal Clinic (Sapporo, Japan) for neurologic abnormalities on 23 April 1992 (on day 1: first medical examination). The dog had suffered from tick bite on several occasions. Two ticks, as adult male and a female Ixodes persulcatus, were judged to have been feeding on the dog for 5 days, and were removed from the dog on day 1 Spirochetes were isolated from the midgut of the adult female tick after culture in BSK II medium. Electron microscopic specimens were prepared by the con ventionalprocedure. The specimens were negative lystained with 1% sodium phosphotungstic acid (ph 7.0). The isolate, designated as P427a, showed similar morphological features to B. burgdorferi (Fig. 1, A and B). It possessed 8-9 flagella, with lengths of 11.1 to 16.6ƒÊm and diameters of 0.22 to 0.29ƒÊm. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblot analysis of the isolate were carried out according to the method descrived by Masuzawa et al (17). The isolate showed a similar SDS-PAGE profile to the strains HP3 and HO14 isolated from Japanese Ixodid ticks (Fig. 2A). The isolate had 60 kda antigen, flagellin (41 kda) and outer Abbreviations: CBC, complete blood cell count; ELISA, enzyme-linked immunosorbent assay; SDS- PAGE, sodium dodecyl sulfate-polyacrylamide gel electro - phoresis. 325

326 Y. AZUMA ET AL Fig. 1. Electron micrographic observation of strain P427a. The cell of P427a was negatively stained with 1% sodium phosphotungstic acid. The bar represents 0.2im (A) and 1.67pm (B). - Fig. 3. Creatine phosphokinase isoenzyme profile of cases 1 and 2. The mean ratio of "BB" fraction in 38.5% in healthy dogs. "BB"=brain type, "MB"=hybrid type, "MM" =muscular type. Fig. 2. SDS-PAGE (A) and immunoblott (B) of strain P427a. Gel was stained with Coomassie brilliant blue. P427a antigens were transferred from the gel strip to polyvinylidene difluoride membranes and reacted with rabbit immune sera against strain H014 and HP3. Numbers on the right indicate the molecular masses (in kilodaltons, kda). surface proteins. P427a reacted with rabbit antiserum against HP3 and H014 (Fig. 2B). Potent reactivity of the strain with the rabbit antisera was found in several bands. The neurological signs displayed by case 1 were astasia, persistent tonic convulsions, and hyperreflexia. Thrombocytopenia and fever (39.4 C) were also seen. Creatinine, glutamic-oxaloacetic transaminase, glutamic-pyruvic transaminase and creatine phosphokinase levels, measured by the automatic super dry system Spotchem TM SP-4410 (Kyoto-Daiichi-Kagaku, Co., Ltd., Kyoto, Japan), were 1.4 mg/dl, 79 IU/L, 81 IU/L, and 1,064 IU/ L, respectively. These values are in the normal to high range. Diazepam (5 mg) was administered intravenously after collection of a blood sample at the first medical examination (day 1). Benzyl penicillin procaine (300,000 IU) and dihydrostreptomycin sulfate (370 mg) were adminstered subcutaneously from day 1 to 6. Amoxicillin was administered orally from day 7 to 16. About 10 hr after the first administration of the abovementioned antibiotics, the dog was able to walk with an ataxic gait. The clinical signs disappeared within 1 week. On day 63 after the first medical examination, the second examination was performed. The dog recovered and the blood plasma profile returned to the normal range, with the exception of the creatinine level (1.6 mg/dl). On cellulose acetate electrophoresis the creatine phosphokinase isoenzyme showed a high ratio of the "BB" fraction (brain type), although the overall enzyme level had returned to normal (Fig. 3) (11). The IgG antibody titers to B. burgdorferi were examined by ELISA using the antigens H014 and

NOTES 327 Fig. 4. ELISA of IgG antibodies to HO14 and HP3 of B. burgdorferi in case 1., HO14 on day 1; œ, HO14 on day 63;, HP3 on day 1;, HP3 on day 63. HP3, both of which had been prepared from B. burgdorferi isolated from Ixodid ticks in Hok kaido,japan (10). As shown in Fig. 4, the serum IgG antibody titer increased from day 1 to 63. For the determination of positive serum, a cut-off value was established by the method previously de scribed(10). The serum specimen obtained on day 63, which gave an absorbance at 492nm greater than the cut-off value, was considered positive to the strain HO14: a positive absorbance of>0.707 for IgG antibody against HO14 and of>0.667 for IgG antibody HP3. IgM antibody titers to B. burgdorferi showed continual high levels. The absorbance for HO14 was 0.763 on day 1, 0.670 on day 63, greater than the cut-off value (HO14: >0.667). Similarly, IgM antibody to HP3 was high (0.323 on day 1; 0.346 on day 63). However, it was less than the cut-off value (HP3: >0.366). Absorbance of the control (5 SPF Beagle dogs) was less than 0.200 (IgG, HO14: 0.200; IgG, HP3; 0.183; IgM, HO14: 0.112; IgM, HP3: 0.052). The second case, a 2-year-old female Labrador retriever dog weighing 22kg, showed neurological signs (2 June 1992), ten days after receiving a tick bite. This dog had suffered from tick bite on Fig. 5. ELISA of IgG antibodies to HO14 and HP3 of B. burgdorferi in case 2., HO14 on day 1; œ, HO14 on day 30;, HP3 on day 1;, HP3 on day 30. several occasions. The clinical signs and blood plasma profiles were almost identical to those seen in case 1. This dog was able to walk at about 5hr after the first administration of the above antibio tics.clinical signs subsided with antibiotic thera pysimilar to that on case 1. On day 30, a second examination was performed. The dog had recover edand the blood plasma profile had returned to the normal range with the exception of the creatinine level. As shown in Fig. 3, the percentage of the " BB" fraction of creatine phosphokinase isoen zymewas high on day 1. The ratio of the "BB" fraction had decreased to be within the normal range on day 30. The serum IgG antibody titers to both strains of B. burgdorferi were positive, levels giving absorbances greater than the cut-off values (Fig. 5). The IgG antibody levels decreased from day 1 to 30. From the serological examination, the dog was considerd to be cured of the disease. The level of IgM antibody to each strains HO14 and

328 Y. AZUMA HP3 was less than the cut-off value at this time. Diagnosis of canine Lyme disease is often based on positive serologic response in conjunction with compatible clinical signs (12-14). However, elevated IgG antibody titers can be found in asymptomatic dogs (10, 12, 13). Therefore, the clinical diagnosis of canine Lyme disease involving the nervous system was done by combination of the following facts: (a) the presence of a compatible neurological abnormality without other cause; (b) a history of tick-bite; (c) serum reactivity against B. burgdorferi; (d) prompt response to antibiotic therapy (1, 4). The clinical signs in the cases reported here were characterized by persistent tonic convulsions and hyper-reflexia such as seen in human neuro-lyme disease (19). Similar clinical signs can be observed in disease such as distemper, tick-borne encephalitis, bacterial meningitis and epilepsy. These diseases, however, do not rapidly respond to antibiotic treatment alone. Furthermore, abnormalities of the complete blood cell count (CBC) profile are present in cases of bacterial infection such as Streptococcal meningitis (4). Abnormalities of CBC were not recognized in our cases. B. burgdorferi was isolated from the midgut of I. persulcatus removed from case 1. Since B. burgdorferi is isolated from the midgut of 16%of I. persulcatus in Hokkaido (18), it is uncertain whether the clinical signs in case I were caused by the isolated strain. As the dog had suffered several episodes of tick bite previously, and IgM antibody to HP3 was less than IgG antibody, we think that a different strain of B. burgdorferi might have been the cause of onset. IgG antibody to HP3 were less than those to HO14. Since the strain from Ixodes persulcatus has the multiplicity of antigen, IgG antibody to HP3 migit be less than HO14 in case of the use of a specific strain from Ixodes persulacatus. It has been reported that dogs with Lyme disease have IgG antibodies to B. burgdorferi (1, 4, 6-8, 15). The level of IgG antibody on day 1 in case 1 was seronegative. This result was due to the fact that IgG antibodies to B. burgdorferi had not increased during the early phase of the disease (1, 4). Measurement of IgM antibody is useful for sero-diagnosis in such cases. We presumed that the neurological abnormalities originated in the central nervous system. The high ratio of the "BB" fraction of creatine phosphokinase isoenzyme suggests that inflammation was present in the brain. ET AL These findings suggest that neurological malities shown in both dogs were caused ection with B. burgdorferi. In Japan, there are reports that antibodies to B. burgdorferi have been measured in dogs (2, 10). The antibody was positive in 27% of 387 clinically healthy dogs with the opportunity for exposure to tick bite (2). It has been suggested that dogs are at high risk to B. burgdorferi infection in Japan (2, 10, 17, 18). There has been no previous report concerning the clinical manifestations of canine Lyme disease in Japan. Since antibiotics are frequently administered to dogs with fever, patients with Lyme disease may have recovered without the cause of disease being determined. In the U.S.A., clinical signs of canine Lyme disease have been reported as arthritis, arthralgia, generalized stiffness, abortion, and renal failure (1, 2, 4, 5, 7, 8, 12-15). Neurological abnormalities in Lyme disease-affected dogs have not been reported in the U.S.A. A number of different strains of B. burgdorferi have been identified on the basis of variation in their proteins (1, 3, 17, 18). Most of the American strains share a similar ospa protein. Strain differences may underlie some of the report= ed clinical differences between American and European human patients with Lyme disease. It has been indicated that the antigenicity of Japanese strains differs from that of strains isolated in the U.S.A. and Europe (17, 18). We, therefore, believe that these could account for differences in clinical signs exhibited by dogs in the U.S.A., Europe and Japan. Morphological characteristics of the isolate in the current case were in accordance with those of B. burgdorferi. The number of flagella isolate P427a was 8-9. It is known that the number of flagella of B. burgdorferi varies among strains, with American isolates having seven or eleven (9, 16). P427a reacted with rabbit-antiserum against strain H014 and HP3 of B. burgdorferi in immunoblot blot. Further characterization of strain P427a by immunoblot using monoclonal antibodies and DNA analysis is being conducted. References 1) Appel, M. 1990. Lyme disease in dogs and cats. Compend.Contin. Educ. Prac. Vet. 12: 617-627. 2) Arashima, Y. 1991.Anti-Borrelia burgdorferi antibody in dogs: Lymediseaseas zoonosis.jpn. J. Clin. Pathol. 39: 869-874(in Japanese). 3) Barbour,A.G., Heiland, R.A., and Howe,T.R. Heterogeneityof major proteins in Borrelia bur

NOTES 329 feri: a molecular analysis of North American and European isolates. J. Infect. Dis. 152: 478-484. 4) Burgess, E.C. 1988. Borreliosis (Lyme disease), p.153-159. In Barlough, J.E.(eds), Manual of small animal infectious disease, Churchill Livingstone, New York. 5) Grauer, G.F., Burgess, E.C., Colley, A.J., and Hagee, J.H. 1988. Renal lesions associated with Borrelia burgdorferi infection in a dog. J. Am. Vet. Med. Assoc. 193: 237-239. 6) Green, R.T., Levine, J.F., Breitschwerdt, E.B., and Berkhoff, H.A. 1988. Antibodies to Borrelia burgdor feriin dogs in North Carolina. Am. J. Vet. Res. 49: 473-476. 7) Green, R.T., Levine, J.F., Breistchwerdt, E.B., Walker, R.L., Berkhoff, H.A., Cullen, J., and Nicholson, W.L. 1988. Clinical and serologic evaluations of induced Borrelia burgdorferi infection in dogs. Am. J. Vet. Res. 49: 752-757. 8) Green, R.T. 1990. Lyme borreliosis, p.508-514. In Green, C.E. (ed), Infectious diseases of the dog and cat, W.B.Saunders Company, Philadelphia. 9) Hovind-Hougen, K., Asbrink, E., Stiernstedt, G., Steere, A.C., and Hovmark, A. 1986. Ultrastructual differences among Spirochetes isolated from patients with Lyme disease and related disorders and from Ixodes vicinus. Zbl. Bakt. Hyg. A263: 103-111. 10) Isogai, E., Isogai, H., Sato, N., Yuzawa, M., and Kawakami, M. 1990. Antibodies to Borrelia burgdor feriin dogs in Hokkaido. Microbiolo. Immunol. 34: 1005-1012. 11) Kikuchi, Y., and Ohnishi, D. 1987. Creatine kinase and isoenzymes in dogs. J. Jpn. Vet. Med. Assoc. 40: 26-30 (in Japanese). 12) Kornblatt, A.N., Urband, P.H., and Steer, A.C. 1985. Arthritis caused by Borrelia burgdorferi in dogs. J. Am. Vet. Med. Assoc. 186: 960-964. 13) Lissman, B.A., Bosler, E.M., Camay, H., Ormiston, B. G., and Benach, J.L. 1984. Spirochete-associated arthritis (Lyme disease) in a dog. J. Am. Vet. Med. Assoc. 185: 219-220. 14) Magnarelli, L.A., Anderson, J.F., Kaufman, A.F., Lieberman, L.L., and Whitney, G.D. 1985. Borreliosis in dogs from southern Connecticut. J. Am. Vet. Assoc. 186: 955-959. 15) Magnarelli, L.A., Anderson, J.F., Schreier, A.B., and Ficke, C.M. 1987. Clinical and Serologic studies of Canine borreliosis. J. Am. Vet. Med. Assoc. 191: 1089-1093. 16) Masuzawa, T., Okada, Y., Beppu, Y., Oku, T., Kawamori, F., and Yanagihara, Y. 1991. Im munologicalproperties of Borrelia burgdorferi iso latedfrom the Ixodes ovatus in Shizuoka, Japan. Microbiol. Immunol. 35: 913-919. 17) Masuzawa, T., Okada, Y., Yanagihara, Y., and Sato, N. 1991. Antigenic properties of Borrelia burgdorferi isolated from Ixodes ovatus and Ixodes persulcatus in Hokkaido, Japan. J. Clin. Microbiol. 29: 1568-1573. 18) Nakao, M., Miyamoto, K., and Fujita, H. 1992. Characterization of Borrelia burgdorferi isolated from Ixodes persulcatus and Ixodes ovatus ticks in Japan. Am. J. Trop. Med. Hyg. 47(4): 505-511. 19) Reik, L., Jr. (eds), 1991. In Lyme disease and the nervous system, p. 52-76. Thieme Medical Publishers, Inc., New York.