CASE NOTES
A PILOT SURVEY OF THE PREVALENCE OF SERUM ANTIBODIES AGAINST TOXOPLASMA GONDII AND RISK FACTORS FOR TRANSMISSION OF T GONDII TO SHEEP IN THE TUMBARUMBA SHIRE OF NEW SOUTH WALES
Helen McGregor District Veterinarian Hume LHPA, Wagga Wagga and Rosalie Harvey BSc BVSc Hons
Posted Flock & Herd March 2011
Introduction
Toxoplasma gondii (T. gondii)is a zoonotic protozoan parasite that causes toxoplasmosis, a disease particularly damaging if acquired during gestation. Sheep are more commonly infected by T. gondii than other livestock such as cattle or pigs (Munday, 1975; Sharif, et al., 2007). Consequently, sheep are more frequently linked to the transmission of T. gondii to humans compared with other livestock species. Felidae (commonly domestic cats) are the definitive hosts of T. gondii. Environmentally resistant oocytes are passed in the faeces of infected cats (Dubey & Lappin, 2006). There are three infective stages of T. gondii; bradyzoites which are found enclosed in tissue cysts; tachyzoites which are found in host tissue and sporozoites that develop in oocysts (Dubey & Lappin, 2006). Definitive and intermediate hosts become infected with T. gondii by ingestion of bradyzoites, tachyzoites or sporozoites (Dubey & Lappin, 2006). All stages of T. gondii occur in cats whereas only bradyzoites and tachyzoites occur in intermediate hosts (Dubey & Lappin, 2006).
Antibodies against T. gondii have been detected in a number of other species in Australia including pigs, cattle, macropods, cats and dogs (Al-Qassab, et al., 2009; Johnson, Roberts, & Munday, 1988; Munday, 1975). Cats have historically been assumed to be the main transmitter of T. gondii to livestock. However, the increasing detection of T. gondii or antibodies against T. gondii in other animal species poses the question; are species other than cats involved in the transmission of T. gondii. Limited information also exists on the prevalence of T. gondii in Australian cat populations indicating the necessity for further investigation into the role of cats and possibly other feral or wildlife species in the transmission cycle of T. gondii to livestock.
Financial losses in the sheep industry due to T. gondii have the potential to be large during an outbreak. However, under Australian conditions abortion investigations are difficult and unrewarding due to the extensive nature of most sheep production systems and the lack of suitable pathological samples.
Recorded seroprevalence estimates vary considerably worldwide. Values cited ranged from 3% to 95.7% (Dubey, 2009). Within individual countries there is also a wide range of seroprevalence recorded. These have been investigated in relation to host and environmental risk factors (Dubey, 2009). In Australian studies the prevalence of serum antibodies against T. gondii in rams from sheep flocks within New South Wales (NSW) varied between regions from 3.4% (Plains) to 13.4% (Tablelands) with a mean of 9.0% (Plant, et al., 1982). Conversely, seroprevalence of antibodies against T. gondii in sheep tested from a Tasmanian abattoir was higher with 16.9% of lambs and 61.7% of adult sheep demonstrated evidence of exposure (Munday, 1975).
A pilot survey investigating the seroprevalence of antibodies against T. gondii in sheep from the Tumbarumba shire of NSW and risk factors associated with transmission of T. gondii to sheep was conducted in August and September 2009 (Rosalie Harvey, Hons, Vet Science (in press)).
MATERIAL AND METHODS
Properties for inclusion in the trial were selected from a pool of properties identified by NSW State Forests within the Tumbarumba shire. Farms were randomly selected from this list and owners contacted to assess level of interest and if they met the other selection criteria for the trial. Five of the ten properties were identified has having a border with State Forest and/or National Park and a likely higher level of interaction between wildlife/feral species and domestic livestock. Five other properties had borders only with other commercial livestock properties thus a lower expected level of interaction between wildlife/feral species and domestic livestock. Wildlife and feral species observed on the properties were recorded in the questionnaire data. Structured surveys of wildlife populations and density were not conducted but information was also collected from the Sate Forest feral and pest animal records regarding areas of high/low density of these species and where wildlife species had been recorded or numbers assessed/estimated.
Each property selected for inclusion in the trial had at least 100 ewes aged three years or more which had been run on the property for at least 12 months. Fifty sheep were systematically selected from each property for blood sample collection and testing. Selecting ewes that were of three years or older and that had been present on the property for at least 12 months increased the likelihood of a recent or repeated exposure to T. gondii. Breed was not used as a selection criterion but was recorded for each sheep sampled. Blood samples were also collected from 11 domestic cats and one feral cat on four of the ten properties. Questionnaire data were collected from flock owners either face to face or over the telephone.
A commercial indirect enzyme linked immunosorbent assay (ELISA) kit (IDVET innovative diagnostics, ID Screen, Toxoplasmosis Indirect®) was used to detect antibodies against T. gondii.
A linear mixed model was developed for the T. gondii antibody titre data which were log transformed to meet the modelling assumptions. Statistical analysis was performed using GenStat 12.1 (2009) software. Initially the data were analysed using BCS, age, breed and property as fixed effects and ELISA test number and grid number as random effects. The data were then analysed to determine statistical significance of variables between properties in relation to seroprevalence. A biometrician and epidemiologist were consulted in both trial design and analyses of data. For the initial analyses of data in this pilot study, only selected variables from the risk factor/questionnaire data were analysed as the data set as it stands has low power for these variables. Full analyses will be conducted on the complete data set in late 2011.
Results
Serum samples from 489 sheep were tested in ELISA for antibodies against T. gondii. Results are shown in Table 1.
| Property code | No. tested | No. +ve | No. -ve | No. inconclusive. | Seroprev % |
|---|---|---|---|---|---|
| A | 50 | 3 | 47 | 0 | 6.0 |
| B | 50 | 18 | 32 | 0 | 36.0 |
| C | 50 | 0 | 50 | 0 | 0.0 |
| D | 50 | 0 | 50 | 0 | 0.0 |
| E | 50 | 3 | 47 | 0 | 6.0 |
| F | 50 | 0 | 50 | 0 | 0.0 |
| G | 50 | 0 | 50 | 0 | 0.0 |
| J | 50 | 5 | 45 | 0 | 10.0 |
| K | 50 | 2 | 48 | 0 | 4.0 |
| P | 39* | 2 | 36 | 1 | 5.1 |
| Total | 489 | 33 | 455 | 1 | Average 6.7 |
Table 1. Seroprevalence of antibodies against T. gondii in sheep tested
The presence of domestic and feral cats on properties and the results of serological testing of cats and sheep are summarised in Table 2. Each of the four properties with cats tested for antibodies against T. gondii had one cat with a positive titre. One property with a T. gondii seropositive cat did not have any T. gondii seropositive sheep. Domestic and/or feral cats were reported on all six properties that had T. gondii seropositive sheep.
| Property code | No. of domestic cats | Feral cats present | No. tested | No. positive | Sero-prevalence in cats (%) | Sero-prevalence in sheep (%) |
|---|---|---|---|---|---|---|
| A | 0 | Y | 0 | 0 | 0 | 6.0 |
| B | 1 | Y | 1 | 1 | 100 | 36.0 |
| C | 1 | Y | 0 | N/A | N/A | 0.0 |
| D | 0 | N | 0 | 0 | 0 | 0.0 |
| E | 0 | Y | 0 | 0 | 0 | 6.0 |
| F | 3 | Y | 3 | 1 | 33.3 | 0.0 |
| G | 0 | N | 0 | 0 | 0 | 0.0 |
| J | 6 | Y | 0 | N/A | N/A | 10.0 |
| K | 4 | N | 4 | 1 | 25 | 4.0 |
| P | 3 | Y | 4* | 1 | 25 | 5.1 |
| Total | 12 | (total Y) 7 | 12 | 4 | Average 6.7 |
(* 3 domestic and 1 feral cat was tested; Y – yes; N – no; N/A – not applicable)
Table 2. Presence of domestic and feral cats and seroprevalence of antibodies against T. gondii in cats and sheep tested
Domestic cats on five (5/6) properties had access to feed stores for livestock. Wildlife and feral species reported as present on the properties were foxes, cats, wild dogs (dingoes and/or dingo/domestic dog hybrids), pigs, deer, kangaroos, wombats and rabbits. Foxes and cats were reported to be the most common feral/non native species.
Property size ranged from 5.25 hectares (ha) to 890 ha. Three (3/10) property owners operated a sheep only production system while the balance (7) operated a mixed production system. The main enterprise types were wool (3/10) and prime lamb (7/10). Four (4/10) property owners bred their own replacement ewes. All property owners have purchased replacement sheep in the past five years, sourced both locally and from outside the Tumbarumba region.
Flock size ranged from 50 to 3000 head (including all ages and classes of sheep) with eight (8/10) properties having a flock size greater than 1000 head. All properties rely predominantly on dams as watering points for stock. Creeks and/or rivers and/or springs are also accessible on eight properties. Reticulated watering systems were limited to holding yards and small paddocks on all properties. Grazing systems were defined as rotational (5/10) or set stocking (4/10) with one property applying both rotational and set stocking practices. Sheep were supplementary fed with hay and/or grain on nine (9/10) properties. This practice occurred mainly between December and July but was most common in February and March; the time of year when pasture based feed supply was at its lowest. The most common method of feeding was by trail feeding on the ground with some use of feeders.
The average rainfall for the last three years ranged from 483mm to 711mm. Annual winter temperature ranged from -8°C to 20°C. Annual summer temperature ranged from 2°C to 40°C.
STATISTICAL ANALYSIS OF RESULTS
A statistically significant association (p < 0.001) was detected between T. gondii seroprevalence and property of origin. There was a trend suggesting merinos had a higher T. gondii seroprevalence than crossbreds (Table 3). However, this finding was not statistically significant (p = 0.851) when tested in the model. There was no significant effect (p = 0.926) of BCS on T. gondii seroprevalence (Table 4).
| Breed | No. tested | No. seropositive | No. inconclusive | Seroprevalence (%) |
|---|---|---|---|---|
| Cross bred | 368 | 15 | 1 | 4.1 |
| Merino | 121 | 18 | 0 | 14.9 |
| Total | 489 | 33 | 1 |
p=0.851
Table 3. Seroprevalence of antibodies against T. gondii in sheep by breed
| BCS | No. tested | No. seropositive | No. inconclusive | Seroprevalence (%) |
|---|---|---|---|---|
| < 2 | 49 | 3 | 0 | 6.1 |
| ≥ 2 < 3 | 237 | 18 | 0 | 7.6 |
| 3 | 148 | 10 | 0 | 6.8 |
| > 3 | 55 | 2 | 1 | 3.6 |
| Total | 489 | 33 | 1 | Average 6.7 |
p=0.926
Table 4. Seroprevalence of antibodies against T. gondii in sheep by body condition score (BCS)
A number of property factors were also examined. There was no statistically significant effect of any of these variables on seroprevalence of antibodies against T. gondii in sheep. However, the following observations were made in the data. There was a trend for sheep on properties that had domestic cats to have a higher seroprevalence of antibodies against T. gondii compared with sheep on properties that did not have domestic cats (9.3% and 3.0% respectively). Sheep on properties that had T. gondii seropositive domestic cats had a higher T. gondii seroprevalence compared with sheep on properties that did not have T. gondii seropositive domestic cats (11.6% and 3.0% respectively). Sheep on properties where domestic cats had access to feed sheds had a higher T. gondii seroprevalence compared to sheep on properties where domestic cats did not have access to feed sheds (11.3% and 2.4% respectively).
A higher seroprevalence was recorded in sheep on properties with a flock size greater than 1000 (7.7% and 3.0% respectively). Sheep on larger properties (700-1000ha) had higher T. gondii seroprevalence compared to small (1-400ha) and moderate (400-700ha) properties (12.5%, 6.0% and 2.1% respectively). Farmers using set stocking management and those using a combination of rotational grazing and set stocking had lower seroprevalence compared to farmers utilising rotational grazing systems only (4.7%, 2.2% and 9.6% respectively). Seroprevalence was also higher on properties that ran sheep only compared to those that had a combination of sheep and cattle (12.0% and 4.4% respectively). Seroprevalence of antibodies against T. gondii was higher in sheep used to produce wool (15.3%) compared to those used to produce meat (2.9%).
Sheep that received supplementary feed (grain and/or hay) had a higher T. gondii seroprevalence compared to sheep that did not receive supplementary feed (7.5% and 0.0% respectively).
Sheep on properties that shared a border with State Forest and/or National Park and therefore had a potentially higher level of interaction with wildlife had a slightly higher T. gondii seroprevalence compared to other properties (9.6% and 3.8% respectively).
Discussion
This pilot survey demonstrated that there have been variable rates of exposure of sheep to T. gondii in the Tumbarumba shire with 2 of the 10 properties sampled showing high seroprevalences (10 and 36%) to the organism. These rates of exposure are significantly different between properties (p < 0.001) and trends were evident for a number of risk factors investigated. However, no risk factor tested in the survey had a statistically significant association with the seroprevalence of antibodies against T. gondii in sheep. Given the range of seroprevalence and the fact that a high seroprevalence was detected in this region in both sheep and cats, further work is warranted to determine the importance of this parasite and further clarify primary risk factors for transmission under Australian conditions. There is a large, active feral cat population in the Tumbarumba shire of NSW and it is likely that there are regular interactions between wildlife/feral species and sheep in this region. This finding further supports the hypothesis that exposure to Toxoplasmosis is common for sheep in this region and that should be further investigated as a possible cause of infertility or abortion in sheep. This project is on going with the aim to survey and sample a further 10 properties in the Tumbarumba Shire in 2011. Since this project began there have been 2 suspected cases of abortion/infertility in sheep reported in the Hume Authority suspected to involve toxoplasmosis. These clinical investigations will also continue into 2011.
The mean seroprevalence of antibodies against T. gondii in the sheep (489) tested in this study is comparable to the nine percent T. gondii seroprevalence previously reported for other areas of NSW (Plant, et al., 1982). These data markedly contrast the 61.7% T. gondii seroprevalence in adult sheep in Tasmania (Munday, 1975). Based on these reports, the seroprevalence of antibodies against T. gondii in the current survey was lower than expected. Tumbarumba has a similar climatic environment to areas of Tasmania with cold wet winters and mild dry summers, consequently a T. gondii seroprevalence closer to that of the Tasmanian study was expected. The low seroprevalence of antibodies against T. gondii in sheep in the current survey may have been influenced by recent drought during the last eight to ten years and warmer than normal winter and summer temperatures. Differences in T. gondii seroprevalence between regions of NSW and in other parts of the world suggests climate may be a risk factor therefore this should be investigated further to determine the effect climatic conditions have on the survival of oocysts in the environment and consequently the prevalence of antibodies to T. gondii in sheep (Abu Samra, et al., 2007; Plant, et al., 1982).
A number of risk factors for increased seroprevalence of antibodies against T. gondii in sheep were investigated to explore the differences between farms. However, no significant difference between the risk factors investigated and T. gondii seroprevalence in sheep on these 10 farms was found. The small sample size may have reduced the power of the analyses to detect differences if they were present. A higher apparent T. gondii seroprevalence in Merinos has previously been reported in the literature (Hartley, 1961; Plant, et al., 1982; Watson & Beverley, 1971). Further work to investigate the effect of breed is required to determine if particular breeds are more susceptible to infection with T. gondii or whether the management of breeds in different farming systems affects transmission of T. gondii to sheep.
The literature also reports a higher T. gondii seroprevalence in adult sheep compared to lambs (Bahrieni, et al., 2008; Munday, 1975). However, age was not tested in the current model as age was used as a selection criterion. Sheep three years and older have an increased probability of being exposed to T. gondii organisms and may encounter repeated exposure. Repeated exposure may increase the number of antibodies againstT. gondii.
The seroprevalence of antibodies against T. gondii in cats in this survey was 33.3%. The presence of cats has been recorded as associated with the presence of antibodies against T. gondii in sheep (Lopes, et al., 2009; Plant, et al., 1982). However, this is the first report of T. gondii seroprevalence in coexisting cat and sheep populations in Australia. The small sample of cats tested in this survey precludes this work reaching a definitive conclusion. However, this is an avenue of research which justifies further exploration possibly including utilising molecular techniques to identify if it is the same strain of organism in both populations. Data from the current survey did not support previous research demonstrating a significant association between presence of cats on properties and seroprevalence of antibodies against T. gondii in sheep on those properties (Lopes, et al., 2009; Plant, et al., 1982). While there was no significant effect found between presence of cats and seroprevalence of antibodies against T. gondii in sheep on properties tested, all properties with T. gondii antibody positive sheep, reported cats (domestic and/or feral) to be present on the property. Alternatively, the lack of association between T. gondii antibodies in sheep and presence of cats could suggest other sources of infection being involved, including wildlife or feral species. Recent evidence of exposure to T. gondii in species other than cats supports the possibility that interaction of domestic species and wildlife may be an important factor for transmission of T. gondii.
Supplementary feeding of hay and/or grain is a risk factor much discussed as part of the likely transmission of T. gondii between cats and livestock species. Of note in this study was that prevalence of antibodies against T. gondii in sheep on properties that supplementary fed sheep was 7.5% but zero in sheep on properties that did not supplementary feed. This may indicate that farms that supplementary feed are more intensive and have higher stocking densities which may influence T. gondii transmission to sheep. Any future work investigating the importance of flock and property size as risk factors of transmission of T. gondii should also address whether stocking density is the more important factor due to an increased risk of exposure of sheep to T. gondii at water sources and through supplementary feeding practices. Other studies have established that sheep on intensively managed properties have a higher T. gondii seroprevalence (Abu Samra, et al., 2007; Plant, et al., 1982). However, we were not able to show such an association in this study.
Sheep on properties that bordered State Forest and/or National Park are likely to have a relatively high but indirect interaction with wildlife/feral animals. Sheep on these properties had a higher T. gondii seroprevalence compared to sheep on properties that did not border State Forest and/or National Park (9.6% and 3.8% respectively). One interpretation of this data is that wildlife/feral species may contribute to transmission of T. gondii to sheep, as antibodies against T. gondii have been identified in both predatory and non-predatory animals. It would be interesting to include data on T. gondii seroprevalence in both predatory and non-predatory animals in future studies to explore the effect animal type (predator versus non-predator) and species have on the lifecycle ofT. gondii in the wildlife and feral animal reservoirs.
Classification by enterprise type produced a similar outcome to breed (wool sheep compared to meat sheep; 15.3% and 2.9% respectively). It may be that management practices employed on these different enterprises are more important in terms of transmission of T. gondii to sheep than genotype or enterprise focus. It would be expected that sheep used for meat may be managed more intensively compared to sheep used for wool. However, as determined by the questionnaire data collected so far, there has been a high rate of supplementary feeding on most farms (9/10) over the past few years due to drought conditions. These conditions may have also affected 'traditional' husbandry structures on these property types and this may have confounded any real effect of breed or sheep enterprise type.
Cohorts of younger sheep should also be tested to further investigate age in relation to exposure. Collection of samples such as stored feed, water and sediment would be helpful to determine the importance of these factors in transmission to sheep. Further collection of management and inventory information from flocks with both high and low seroprevalence of antibodies against T. gondii may assist to establish possible relationships between inherent infertility reproductive problems and exposure to T. gondii.
References
- Abu Samra N, McCrindle C, Penzhorn B & Cenci-Goga B (2007) Seroprevalence of toxoplasmosis in sheep in South Africa Journal of South Africa Veterinary Association 78(3):116-120
- Al-Qassab S, Reichel M P, Su C, Jenkins D, Hall C, Windsor PA et al. (2009) Isolation of Toxoplasma gondii from the brain of a dog in Australia and its biological and molecular characterization Veterinary Parasitology 164(2-4):335-339
- Bahrieni M, Fasihi Harandi M, Beigzadeh M, Kamyabi H, & Zia-Ali N (2008) Risk factors analysis associated with seropositivity to Toxoplasma gondii in sheep and goats in Southeastern Iran using modified agglutination test (MAT) Iranian Journal of Parasitology 3(1):38-43
- Caballero-Ortega H, Palma J, Garcia-Marquez L, Gildo-Cardenas A, & Correa D (2008) Frequency and risk factors for toxoplasmosis in ovines of various regions of the State of Colima, Mexico Parasitology 135:1385-1389
- Dubey JP (2009) Toxoplasmosis in sheep - The last 20 years Veterinary Parasitology 163(1-2)
- Dubey JP & Lappin MR (2006) Toxoplasmosis and Neosporosis In CE Green (Ed.) Infectious diseases of the dog and cat 3 ed. pp 754-774 USA: Saunders Elsevier
- Hartley WJ (1961) Experimental transmission of toxoplasmosis in sheep New Zealand Veterinary Journal 9:1-6
- Johnson AM, Roberts H & Munday BL (1988) Prevalence of Toxoplasma gondii antibody in wild macropods Australian Veterinary Journal 65(7):199-201
- Lopes WDZ, Santos TRD, da Silva RDS, Rossanese WM, de Souza FA, de Faria Rodrigues JDA et al. (2009) Seroprevalence of and risk factors for Toxoplasma gondii in sheep raised in the Jaboticabal microregion, Sao Paulo State, Brazil Research in Veterinary Science In Press, Corrected Proof
- Munday BL (1975) Prevalence of Toxoplasmosis in Tasmanian meat animals Australian Veterinary Journal 51:315-316
- Plant J, Freeman P, & Saunders E (1982) Serological survey of the prevalence of Toxoplasma gondii antibodies in rams in sheep flocks in New South Wales Australian Veterinary Journal 59:87-89
- Sharif M, Gholami S, Ziaei H, Daryani A, Laktarashi B, Ziapour SP et al. (2007) Seroprevalence of Toxoplasma gondii in cattle, sheep and goats slaughtered for food in Mazandaran province, Iran, during 2005 The Veterinary Journal 174(2):422-424
- Vesco G, Buffolano W, La Chiusa S, Mancuso G, Caracappa S, Chianca A et al. (2007) Toxoplasma gondii infections in sheep in Sicily, southern Italy Veterinary Parasitology 146:3-8
- Watson W & Beverley J (1971) Ovine abortion due to experimental Toxoplasmosis The Veterinary Record 88:42-45