Toxoplasma gondii in Ontario Mammals

The Prevalence of Serum Antibodies to Toxoplasma gondii in Ontario Mammals 1. R. Tizard, J. Harmeson and C. H. Lai* ABSTRACT The prevalence of seropositive reactions to Toxoplasma gondii was studied in farm animals, companion animals, wild rodents and birds. Of the animals tested, 17% of cattle, 65% of sheep, 45% of pigs, 9%7, of horses, 33% of dogs and 2% of cats were seropositive by the Sabin-Feldman dye test. In addition 11%7o of mice (Mus musculus), 5% of deer mice (Peromyscus maniculatus), 3% of rats (Rattus norvegicus) and less than 2% of sparrows (Passer domesticus) were seropositive. All samples from short-tailed field mice (Microtus pennsylvanicus), squirrels (Sciurus carolinensis), chipmunks (Tamias striatus), meadow jumping mice (Zapus hudsonius) and starlings (Sturnus vulgaris) were seronegative. The significance of these findings in relation to the epizootiology of toxoplasmosis in Ontario is discussed. RESUMe Les auteurs ont recherche la predominance de reactions serologiques positives, a l'endroit de Toxoplasma gondii, chez des animaux domestiques, des animaux de compagnie, des rongeurs sauvages et des oiseaux. Parmi les animaux qu'ils eprouverent au cours de cette experience, 17% des bovins, 65% des moutons, 45%- des porcs, 9% des chevaux, 33% des chiens et 2% des chats reagirent de fagon positive 'a l'epreuve de Sabin-Feldman. De plus, 11% des souris (Mus musculus), 5% des souris sylvestres (Peromyscus maniculatus), 3% des rats (Rattus norvegicus) et moins de 2% des moineaux (Passer domesticus) donnierent des resultats positifs. Tous *Toxoplasmosis Diagnostic Laboratory, Department of Veterinary Microbiology and Immunology, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada. Submitted May 18, 1977. les echantillons des mulots (Microtus pennsylvanicus), des ecureuils (Sciurus carolinensis), des tamias (Tamias striatus), des souris sauteuses des champs (Zepus hudsonius) et des etourneaux (Sturnus vulgaris) s'avererent negatifs. Les auteurs commentent la signification de leurs resultats par rapport 'a l'epizootiologie de la toxoplasmose, en Ontario. INTRODUCTION Toxoplasmosis is endemic in Ontario (3, 8, 22, 24). Approximately 4% of the human population possess serum antibodies to Toxoplwsna gondii when measured either by the indirect fluorescent antibody test (3) or by the Sabin-Feldman dye test (8, 24). Humans may acquire infection by at least three routes: by transplacental transfer of organisms from pregnant mothers to their fetus, through the consumption of raw or grossly undercooked meat or milk containing cysts, or by the accidental ingestion of oocysts shed in feces from infected cats (15, 18, 27). The congenital route of infection is probably the least important of these three routes in Ontario (22) but the relative importance of the other two is unclear. It is generally assumed that toxoplasmosis must be enzootic in the small rodent population which serves as a reservoir of infection for cats (27). Cats in turn are considered to act as sources of infection for both humans and domestic herbivores. Herbivores may then provide an alternative source of infection for humans through infected meat. This paper reports the results of a survey of sera taken from each of these groups of animals in order to provide basic data for studying the epizootiology of this condition in Southern Ontario. Volume 42 - April, 1978 177

SAMPLES MATERIALS AND METHODS Farm animal samples (cattle, horses, pigs and sheep) were obtained from five major sources (Table I). Samples from the Clinical Pathology Laboratory, Ontario Veterinary College (OVC) represented a "normal hospital population" but also included some samples from nonhospitalized animals. These samples had been taken for clinical reasons but not primarily for toxoplasmosis testing. Horse serum samples obtained from the Large Animal Clinic at OVC were from a "normal hospital population" but were taken for the purpose of this survey. Samples obtained through the courtesy of the Animal Pathology Laboratory, Agriculture Canada, Guelph were originally submitted for brucellosis serology and are representative of a population of normal animals. Samples from the Serology Laboratory of the Veterinary Services Branch (VSB), Ontario Ministry of Agriculture and Food, were submitted to that laboratory by veterinarians for routine serology and are probably comparable to the "normal hospital population" samples obtained from OVC. The bovine samples obtained at an abattoir in Guleph were from feedlot animals from various locations in southwestern Ontario. Companion animal samples were obtained from some of the same sources (Table I). Blood samples taken from cats and dogs at the Guelph Humane Society animal shelter are considered to be representative of a healthy but uncontrolled animal population. Some cats encountered on routine farm visits were also sampled. Small rodents were livetrapped1 at nine locations within and around the city of Guelph. These locations included barns, woodlots, fields, riverbanks, gardens and the city garbage dump. No more than ten rodents were trapped at any one location. Birds were caught with mist nets in a garden in suburban Guelph, at the Large Animal Clinic, OVC, at the Swine Experimental Station of the Ontario Ministry of Agriculture and Food at Arkell, Ontario and on a farm located approximately ten miles southwest of Guelph. IHavahart traps. Havahart Inc., Ossining, New York. 178 TESTING PROCEDURES All serum samples were stored frozen at -2 C until required. Immediately prior to testing they were inactivated by incubation at 56 C for 45 -minutes and then diluted to 1:16 in normal saline. Antibodies to T. gondii were detected by means of the Sabin-Feldman dye test (SFDT) (2). The test was conducted in a standard manner using human accessory factor and the RH strain of T. gondii. In view of published reports that birds although infected with this organism may remain seronegative (12, 26), attempts to isolate organisms were routinely undertaken on all birds. Brain, liver, heart and spleen were removed from each bird, pooled and homogenized in 5 ml normal saline. Each of five mice were inoculated intraperitoneally with.5 ml of this homogenate and examined at regular intervals. After two weeks an attempt was made to aspirate toxoplasms from their peritoneal cavities. Each animal was then exsanguinated, its serum was tested for antitoxoplasma antibodies by SFDT and its brain, liver and spleen examined histologically for the presence of Toxoplasma cysts. DATA ANALYSIS Data from each animal tested (date of testing, species, result, age, sex, source and original location of specimen) were cross tabulated and subjected to statistical analysis using the SPSS 6 program on an IBM 37/155 computer of the Institute of Computer Science at the University of Guelph. CATLE RESULTS Of the bovine samples tested 17% were positive (Table II). The average age of the animals (excluding the abattoir samples) was 4.2 years. In general, the source of the samples appeared to have little effect, although 15% of the samples from the abattoir were positive as com- Can. J. comp. Med.

TABLE I. The Sources and Numbers of Serum Samples Used in this Survey Source Bovine Equine Porcine Ovine Canine Feline A... 75 1 4 197 42 B..218 C... 112 2 152 2 32 34 D 518 267 E... 564 F 5 63 G 1 13 Total... 1759 238 671 273 235 152 A Clinical Pathology Laboratory, Ontario Veterinary College, Guelph B Large Animal Clinic, Ontario Veterinary College, Guelph C Serology Laboratory, Veterinary Services Branch, Ontario Ministry of Agriculture and Food, Guelph D = Animal Pathology Laboratory, Agriculture Canada, Guelph E = A Guelph abattoir F Guelph Humane Society G = Various farms TABLE II. The Prevalence of Antibodies to T. gondii in the Samples Tested No Positive/No % Species Tested Positive Meat Animals Sheep... Pigs... Cattle... Companion Animals Horses... Cats... Dogs... Small Rodents Mouse (Mus musculus) Rat (Ratius norvegicus) Short-tailed field mouse (Microtus pennsylvanicus)... Deer mouse (Peromyscus maniculatus)... Squirrel (Sciurus carolinensis)... Chipmunk (Tamias striatus)... Meadow jumping mouse (Zapus hudsonius)... Birds House sparrow (Passer domesticus)... Common starling (Sturnus vulgaris)... 176/273 3/671 39/1759 65 45 17 bovine samples taken in February were positive while the percentage of seropositive reactions declined over the summer so that only 15 of 118 (12.7%) samples taken in November were positive. PIGS AND SHEEP Positive reactions were obtained in 45% 21/238 9 of the pig samples and 65% of the 3/152 2 sheep samples. There were no significant 111/335 33 differences in prevalence between samples obtained from different sources and in- 4/36 11 sufficient information was available to 1/35 3 draw reliable conclusions on the influence of sex on these results. The average age /2 1/19 5 was not available. /16 /6 /2 1/71 /28 pared with 18-19% in samples from the other sources. There were no significant differences in the prevalence of positive reactions between animals of different sex but age differences were marked. In animals whose age was known, the prevalence of positive reactions climbed from 15% in animals one year of age to 37% in animals eight years of age or older. It was also observed that 3 out of 84 (36%) of the pig population tested was 1.55 years. Information on the age of sheep HORSES Equine samples had a considerably lower rate of infection than the other large domestic animals, only 9% being seropositive. The average age of this population was 5.7 years. DOGS Of the dog sera tested 33% were positive. The average age of the population tested was 4.96 years. No differences in prevalence were observed between sexes nor between samples obtained from different sources. Of 13 one year old dogs, three (8%) were seropositive while 12 of 26 (46%) dogs eight or more years of age were positive. Volume 42 - April, 1978 179

CATS Of the cats tested 2% were seropositive. With regard to sources, 23 % of samples submitted by veterinarians (i.e. to OVC or VSB) were positive while 19% of samples from the humane society were positive. The average age of the population tested was 1.15 years and the incidence rate appeared to be approximately 2%/year (Fig. 1). SMALL RODENTS The prevalence of positive reactions in the rodents tested was relatively low as might be anticipated in animals with a short lifespan. The highest prevalence was found in house mice (Mus musculus) with four of 36 (11%) animals tested positive. These four mice were captured in barns on two farms at which toxoplasmosis appeared to be highly prevalent. On farm A, two of four mice, one of two cats, four of five calves and the one member of the farm family tested were seropositive. On farm B, two of eight mice, neither of two deer mice and 34 out of 52 cattle tested were positive. Of these four house mice, one was male and three were females, two were juvenile while two were adult. The one positive rat was a juvenile female trapped close to the Speed River in central Guelph. No positive reactions were obtained in any of the other rodents tested. 1 - Percentage Positive 8-6 - 4-2 1 2 3 3. 2 2 3 3 *.1 3 4 Age (Years) Fig. 1. The relationship between age of cats and the prevalence of positive reactions to T. gondii. Each value represents the number positive over the number tested. 18 6 WILD BIRDS One male house sparrow, trapped on a farm ten miles south of Guelph, was found to be seropositive. Attempts to isolate the organism from this and other birds trapped were uniformly negative. DISCUSSION When the SFDT was first introduced (2) there was concern as to its specificity and it was eventually concluded that the test, as performed in humans, was essentially specific for T. gondii (7). It has recently been shown however, that the closely related coccidian Hammondia hammondi (1) will cross react in the dye test (4). The test is believed not to detect antibodies to sarcocysts in either humans (2, 14) or sheep (1). It is therefore probable that this test will detect antibodies only to T. gondii or to very closely related organisms (13) although the possibility of other unrecognized cross-reactivity cannot be excluded. The reported prevalence of seropositive reactions in cattle in North America has ranged from zero to 49% (Table III) but the prevalence of 17% reported here is relatively high for a herbivore. There are a number of possible reasons for this. Probably the most important of these is the exposure of cattle to cats over the winter months when housed indoors. Barn cats appear to prefer to defecate in granular feed (I. R. Tizard, unpublished observations) and it is likely that cattle can acquire infection by consuming feed contaminated in this way. The possible importance of this route is supported by the extremely high prevalence of positive reactions in samples taken from cattle in February. The very high prevalence of positive reactions in sheep (65%) is with one exception (7) considerably higher than that reported from elsewhere in North America. This is in contrast to absence of seropositive reactions in 12 sheep in Saskatchewan (17) but is almost identical to the 63 % prevalence of positive reactions recorded in Ontario goats (23). This may Can. J. comp. Med.

TABLE III. The Reported Prevalence (%) of Seropositive Reactions to T. gondii in Farm Animals in North America Species Location Bovine Ovine Caprine Porcine Equine References Ontario... Ontario... Saskatchewan... New York... Connecticut... Kentucky... Tennessee... Iowa... Illinois... Arizona... California... 17-49 7 24-32 suggest that a common mechanism of infection exists although its precise nature is unclear. Both sheep and goats are capable of grazing very close to the ground and so are potentially capable of picking up oocysts but will not normally do so unless no other food is available. Sheep are the one species of domestic animals in Ontario in which clinical toxoplasmosis is known to be a significant problem (11). The prevalence rate in pig sera is very similar to reported figures from the Northeastern United States (43% in New York (7) and 42% in Connecticut (28)), (Table III) and is, in general, comparable with results obtained in other omnivores (Fig. 2). This may be a reflection of the feeding habits of this species. The most likely sources of infection are through the accidental ingestion of oocysts while rooting and the deliberate consumption of dead rodents when available. The percentage of positive reactions in horses is comparable to results obtained by others in North America (Table III) and approximates to the expected infection rate in herbivores. The reasons for this prevalence being lower in horses than in cattle is unknown but may be due to the custom of feeding horses from an elevated manger, so reducing the possibility of oocyst contamination of feed. The level of infection in domestic dogs (33%) is considerably lower than that found in wild canids in this region. For example, coyotes (Canis latrans) are 65% seropositive (21). The difference may be attributable to the inhibition of natural hunting in the dog. This prevalence is within the range of positive reactions reported elsewhere (Table IV). The incidence rate in dogs is probably considerably less than that in cats in view of the relatively 65 56 24 37 5 28 45 9 This paper 63 23 - _- - 17 4-48 43 6 7 42 28 7 19-26 4 6 7 Cited in 25 4 - Cited in 25 - - _- 7 27 27 14 25 high age of the dog population tested compared with that of cats. The prevalence in cats reported here is relatively low by American standards (4, 6, 7) but considerably higher than that in Saskatchewan (17). Given an approximate incidence rate of 2%/year (Fig. 1), it is possible to calculate an upper limit on the percentage of cats shedding oocysts at any one time. For example, if 2% of cats become infected in their first year and shed oocysts for two weeks (9) then at any one time.5% of cats will be shedding oocysts. However, while all cats which acquire infection by eating cysts in muscle will shed oocysts, less than half the animals which acquire infection from oocysts or tachyzoites do so (4). Consequently this figure of.5% represents only an upper limit on the prevalence of shedding cats. Certainly, in the Guelph region, the percentage of cats shedding oocysts is somewhat less than this (Dr. J. P. Lautenslager, personal communication). This calculated percentage is also less than the percentage of cats found to be shedding oocysts in other locations in Northeastern TABLE IV. The Reported Prevalence (%) of Seropositive Reactions to T. gondii in Dogs and Cats in North America Location Dogs Cats Reference Ontario... 33 25 This paper Saskatchewan... 3.4 17 Tennessee... 15-35 4 6 Northern California. 5 25 Iowa... 16 77 New York... 28 31 Arizona... 3 Pennsylvania... 59 7 Massachusetts... 34 Middle Atlantic States... 25 16 Volume 42 - April, 1978 181

the Microtus population (5). Indeed, if an fox animal such as a fox consumes more than 8- a hundred Microtus in a year then it will almost certainly become infected even al- Nercentage though the prevalence in the prey species bsitive 7 is less than 1%. The number of Microtus sheep tested by Quinn et al (19) or by ourselves goat COYt are quite insufficient to exclude this 6 species as a significant source of infection. mink The low prevalence of seropositive reactions in wild birds tends to suggest that they too may be epidemiologically relatively bear unimportant except as prey for carnivores. 4- tman In this case, the food pyramid effect described above could ensure that infection 3- skunk dog occurs at a considerably higher rate in the predator species. In general, the results of this survey to- 2 cat gether with other published data for Onbovin rcoo tario confirm the general association between rodent consumption and the pre- 1 - mrtin valence of toxoplasmosis (Fig. 2). Cattle rabbit have a higher seropositivity rate than would be anticipated from their diet but Herbivores Omnivores Carnivores this may be due to their close association Food Habits with cats over the winter months. Dogs Fig. 2. The relationship between diet and the prevalence have a lower than expected rate possibly of positive reactions to T. gondii. Data taken from Reference 21 and this because of paper. the way they are treated in this society. Humans conform to the general trend since although raw meat consump- America (1%ln Columbus, Ohio, 1.5% in tion is rare it is probably compensated for Urbana, Illinois) (4).b xouet oy,s Toxoplativelymlow levelin wilarstodben and The most significant unexplained feature Toxoplasma infection appears to be at by exposure to oocysts. a relatively low level in wild rodents and of the epidemiology of this condition is the birds. Other North American reports in- extraordinarily high prevalence of positive dicate that 4% of Mn mvscuilss and 3% of reactions in sheep and goats. As pointed Rattus norvegicus in urban Memphis are out earlier, this may be due to their parseropositive (6) while 8% of Rattus rattus ticular grazing habits or perhaps to some and 32% of M. mutscuus are seropositive in unknown epidemiological factor. It is also Hawaii (26). In this study only two possible that sheep and goats may be in- Microtus pennsyrvafuncus were captured fected by an organism which cross-reacts and both were found to be seronegative. with T. gondii in the SFDT. Quinn et at (19) tested twenty animals of In conclusion, it must be pointed out that this species captured in the Guelph region a significant proportion of the meat proand all were seronegative. In spite of these duced in Ontario is infected with T. gondii results we believe that this species may and care should be taken not only to ensure represent a significant source of toxo- that meat is properly cooked but also to plasmosis in carnivores for the following ensure that milk is pasteurized and that reasons. Firstly, Quinn et at (19) found hands are well washed after handling raw 18% of the short-tailed shrew (Blarind meat. Although only a very low percentage brevicauda) to be seropositive and thls of cats are probably active shedders of animal is a carnivore whlch commonly oocysts they almost certainly represent a preys on Microtuls. Secondly, Tizard et at significant source of infection for animals (21) found 84% of foxes in this region in this province. to be seropositive and this animal too preys largely on Microtus. Finally, toxoplasmosis is reported to occur in this ACKNOWLEDGMENTS species in other locations and may at times have a significant influence on the size of This study was supported by the Ontario 182 Can. J. comp. Med.

Ministry of Health. We should like to thank Mr. S. S. Chauhan for technical assistance and the directors of the various laboratories mentioned for the generous provision of serum samples. REFERENCES 1. AWAD, F. I. and R. LAINSON. A note on the serology of sarcosporidiosis and toxoplasmosis. J. clin. Path. 7: 152-156. 1954. 2. BEVERLEY, J. K. A., C. P. BEATTIE and C. ROSEMAN. Human toxoplasma infection. J. Hyg., Camb. 52: 37-46. 1954. 3. DESAVIGNY, D. H. Canadian Public Health Ass. Meeting, Ottawa. 1974. 4. DUBEY, J.. A review of Sarcocystis of domestic animals and other coccidia of cats and dogs. J. Am. vet. med. Ass. 169: 161-178. 1976. 5. ELTON, C. Voles, Mice and Lemmings. Oxford University Press. 1942. 6. EYLES, D. E., C. L. GIBSON, N. COLEMAN, C. S. SMITH, J. R. JUMPER and F. E. JONES. The prevalence of toxoplasmosis in wild and domesticated animals in the Memphis region. Am. J. trop. Med. Hyg. 8: 55-51. 1959. 7. FELDMAN, H. A. and L. T. MILLER. Serological study of toxoplasmosis prevalence. Am. J. Hyg. 64: 32-335. 1956. S. FRENCH, C. E. and N. A. FISH. A survey of toxoplasmosis in an Ontario community. J. Can. med. Ass. 84: 757-767. 1961. 9. FRENKEL, J. K. and J. P. DUBEY. Toxoplasmosis and its prevention in cats and man. J. infect. Dis. 126: 664-673. 1972. 1. FRENKEL, J. K. and J. P. DUBEY. Hammondia hammondi, gen. nov. sp. nov., from domestic cats, a new coccidian related to toxoplasma and sarcocystis. Z. ParasitKde 46: 3-12. 1976. 11. HULLAND, T. J. and S. B. TOBE. Toxoplasmosis as a cause of abortion in Ontario sheep. Can. vet. J. 2: 45-51. 1961. 12. JACOBS, L., M. L. MELTON and M. K. COOK. Experimental toxoplasmosis in pigeons. Expl Parasit. 2: 43-416. 1953. 13. MARKUS, M. B. Serology of toxoplasmosis, isosporosis and sarcosporidiosis. New Engl. J. Med. 289: 98. 1973. 14. MARCUS, M. B. Serology of human sarcosporidiosis. Trans. R. Soc. trop. Med. Hyg. 68: 415-416. 1974. 15. MILLER, N. L., J. K. FRENKEL and J. P. DUBEY. Oral infections with toxoplasma cysts and oocysts in felines, other mammals and in birds. J. Parasit. 58: 928-937. 1972. 16. MORRIS, J. A., C. G. AULISIO and J. M. MC- COWN. Serological evidence of toxoplasmosis in animals. J. infect. Dis. 98: 52-54. 1956. 17. NATION, P. N. and J. R. ALLEN. Antibodies to Toxoplasma gondii in Saskatchewan cats, sheep and cattle. Can. vet. J. 17: 38-31. 1976. 18. QUINN, P. J. and B. M. McCRAW. Current status of Toxoplasma and toxoplasmosis. A review. Can. vet. J. 13: 247-262. 1972. 19. QUINN, P. J., R.. RAMSDEN and D. H. JOHN- SON. Toxoplasmosis: A serological survey in Ontario wildlife. J. Wildl. Dis. 12: 54-55. 1976. 2. SABIN, A. B. and H. A. FELDMAN. Dyes as microchemical indicators of a new immunity phenomenon affecting a protozoan parasite (Toxoplasma). Science 18: 66-663. 1948. 21. TIZARD, I. R., J. B. BILLETT and R.. RAMS- DEN. The prevalence of antibodies against Toxoplasma gondii in some Ontario mammals. J. Wildl. Dis. 12: 322-325. 1976. 22. TIZARD. I. R., N. A. FISH and P. J. QUINN. Some observations on the epidemiology of toxoplasmosis in Canada. J. Hyg., Camb. 77: 11-21. 1976. 23. TIZARD, I. R., M. CARRINGTON and C. H. LAI. Toxoplasmosis in goats in southern Ontario. A public health hazard? Can. vet. J. 18: 274-277. 1977. 24. TIZARD, I. R., S. S. CHAUHAN and C. H. LAI. The prevalenice and epidemiology of toxoplasmosis in Ontario. J. Hyg., Camb. 78: 275-289. 1977. 25. VANDERWAGEN, L. D., D. E. BEHYMER, H. P. RIEMANN and C. E. FRANTI. A survey for Toxoplasma antibodies in northern California livestock and dogs. J. Am. vet. med. Ass. 164: 134-137. 1974. 26. WALLACE, G. D. Intermediate and transport hosts in the natural history of Toxoplasma gondii. Am. J. trop. Med. Hyg. 22: 456-464. 1973. 27. WALLACE, G. D. The role of the cat in the natural history of Toxoplasma gondil. Am. J. trop. Med. Hyg. 22: 313-322. 1973. 28. WEINMAN, D. and A. H. CHANDLER. Toxoplasmosis in man and swine. An investigation into the possible relationship. J. Am. vet. med. Ass. 161: 229-232. 1956. THE BIOLOGY AND MEDICINE OF RABBITS AND RODENTS. J. E. Harkness and J. E. Wagner. Published by Macmillan of Canada, Toronto. 1977. 15 pages. This handy little one hundred and fifty page book is advertised as the answer to the needs of the veterinarian, student and biomedical investigator. It is refreshing in this age of the promotion of so many false expectations, that here is a case where an honest pitch has been made. The organization of the information of this wide ranging field into so few pages and yet managing to include all the relevant material is, in my estimation, rather phenomenal. No claims are made for its being exhaustive. But each subject that is covered is done carefully, emphasizing the common BOOK REVIEW and yet giving much helpful information. For the veterinarian this hand book allows him to extend his already extensive knowledge and understanding of medicine to handle laboratory type species in his practise. The authors have resisted the urge to overwhelm with the unusual but have provided a bibliography which will introduce one to many subjects and persons in the field. The case report method of self-teaching is another example of the thoughtful organization of this book. As you may have guessed I am impressed with the book. I would recommend it highly to those it was supposedly written for and anyone else who wants to learn. - J. D. Schroder. Volume 42 - April, 1978 183