First report on the sero-epidemiology of Toxoplasma gondii infection in German roe deer (Capreolus capreolus)

While the roe deer (Capreolus capeolus) is the most important game species in Germany and its venison is popular, there is limited knowledge about the prevalence of Toxoplasma gondii in this animal population in the country, and in wild ungulates in Germany generally. Between 2013 and 2015, we collected 295 blood samples from roe deer belonging to a central German population. Sera were analysed using a modified agglutination test (MAT, cut-off 1:20), and antibodies were detected in 86 of the 295 samples (29%). Seroprevalence values differed significantly between the different age classes, with antibodies more frequently observed in adults. In contrast, seroprevalence did not differ significantly between the sexes or collection years. Venison is frequently consumed raw or undercooked and may be a potential source of human infection with T. gondii.


Introduction
Toxoplasma gondii is an obligate intracellular protozoan and the causative agent of toxoplasmosis [9]. Unsporulated oocysts are shed into the environment by felids, which are the only definitive hosts [8]. Most mammals can become intermediate hosts after consuming raw or undercooked meat containing T. gondii tissue cysts, or food and drink with oocysts [6,9,21]. Meat-derived products from domestic animals and game species may represent a potential source of human infection with T. gondii and the European Food Safety Authority (EFSA) recommends the monitoring of toxoplasmosis in humans, animals and foodstuffs [11].
The roe deer (Capreolus capreolus) is the most important game species in Germany [32]. According to the German Hunting Federation (DJV), around 1.2 million animals have been harvested annually countrywide in recent years (www.jagdverband.de). Despite the popularity of venison and the associated processed meats, there is currently no surveillance of T. gondii infection in German roe deer populations and little knowledge about the prevalence of the parasite in wild ungulates in Germany generally [17,23,29].
Here, we aim to assess the seroprevalence of T. gondii in a free-living German population of roe deer by sampling carcases that were intended for human consumption.

Ethics
Roe deer are a legal game species in Germany that licensed hunters can harvest outside the closed season. No animals were killed in order to provide samples for this study. All animals were legally shot and the carcases made available to the authors.

Sample collection
The study was performed in the west of the German federal state of Thuringia. The total size of the study area was roughly 1800 km 2 comprising the Eichsfeld, the western part of the Unstrut-Hainich and the northern part of the Wartburg administrative districts. Between 2013 and 2015, local hunters collected blood from the heart of 295 legally hunted roe deer. Animals were sampled in hunting areas across the whole study area. After centrifuging samples for 10 min at 1000g using an EBA 200 (Hettich, Tuttlingen, Germany), the sera were stored at À20°C until analysis. The sex, age and year of sample collection were recorded for each animal. Based on the dentition of the lower jaw, animals were classified as juveniles ( 1 year), yearlings (1-2 years), or adults (!2 years) [19,32].

Determination of antibodies to T. gondii
A commercial kit (Toxo-Screen DA Ò , bioMérieux, Lyon, France) was used to perform a modified agglutination test (MAT) to analyse sera for the presence of T. gondii immunoglobulin G (IgG) antibodies. Positive and negative controls employed formalin-fixed tachyzoites as antigens. Serum samples were tested at dilutions of 1:20, 1:400, 1:1600 and 1:3200. The sensitivity and specificity of the test were maximized by using a cut-off titre of 1:20 [10]. Of all the available serological tests, the MAT is considered to be the most reliable in terms of detecting antibodies to T. gondii, especially in latently infected animals [9].

Statistical analysis
We performed a v 2 -test in SPSS v.22 (SPSS Inc., Chicago, Illinois, USA) to assess the effect of sex, age class and collection year on T. gondii seroprevalence. Odds ratios (ORs) and their 95% confidence intervals (95% Cls) were calculated to assess the strength of the association between the presence of antibodies and the explanatory variables.

Discussion
This is the first study investigating the seroprevalence of T. gondii antibodies in German roe deer. Values reported from other European roe deer populations ranged from 13% to 63% (Table 2). These previous studies used at least six different diagnostic tests (Table 2). In addition to our MAT test, the direct agglutination test (DAT) and the enzyme-linked immunosorbent assay (ELISA) have also been used frequently in this context and it has been shown that the three tests produced congruent and comparable results [13,14,24,34]. Seroprevalences reported using one of these three tests ranged from 13% to 52% (Table 2). Of these, studies performed in Spain and Poland often reported substantially lower prevalence values than the 29.15% reported here, while studies from Belgium and France reported substantially higher figures. Other studies presented estimates that were in line with the estimate from the present study (Table 2).
There are two previous studies that investigated the T. gondii seroprevalence in wildlife from our study region. The values of 38.3% reported for raccoons (Procyon lotor) [16] and of 24.5% reported for the European mouflon (Ovis orientalis musimon) [17] were relatively high compared to values from other European studies in these species. These authors took this as evidence of high environmental contamination with oocysts as, in addition to the presence of feral, stray, and pet cats (Felis sylvestris domesticus), the study region was located within the core distribution area of the wildcat in central Germany [16,17]. Beral et al. [4] found a positive link between higher T. gondii antibody levels in wild boar (Sus scrofa) and the occurrence of wildcats in France. Our results do not contradict this conclusion, as the T. gondii seroprevalence in the roe deer population in the area is comparable to the values observed in the other two species. While the roe deer value obtained here is not particularly high compared to other European results (Table 2), the wildcat also occurs in the study areas in France and Belgium where a high seroprevalence was observed in roe deer. Further research on the environmental factors associated with high T. gondii seroprevalence in European wildlife is clearly needed. Our results suggest that older roe deer had a higher seroprevalence than younger animals. Other studies on roe deer came to a similar conclusion [25,33]. T. gondii antibodies are frequently more prevalent in older animals, since the cumulative likelihood of exposure to T. gondii increases with age and the antibodies persist for a lifetime [1,20]. We did not identify a significant difference in seroprevalence depending on sex and year of sample collection. For at least some part of the year, both sexes have overlapping home ranges [32] and a substantial difference in exposure risks between the two sexes seems unlikely. Seroprevalence did not significantly differ between years, implying that the environmental contamination with infective oocysts remained constant throughout the study, corroborating findings from the mouflon obtained for the same region and study period [17]. It has indeed been suggested that humidity and moderate temperatures promote the survival and sporulation of the oocysts [1,9,13,30].
The high seroprevalence of T. gondii antibodies in a Central German population of roe deer highlights a potential source of human infection. German hunters frequently produce home-made sausages using raw or undercooked meat. Our results suggest that this may lead to an increased risk of food-borne transmission of T. gondii. Additional studies are required to assess infection levels in venison and derived products in order to assess the risk of transmitting T. gondii to humans.

Conclusions
We analysed the sero-epidemiology of T. gondii infection in roe deer from a central German study population. T. gondii antibodies were present in animals of all ages. Raw or undercooked venison and its derived products may be a potential source of human infection with T. gondii.