Tick-borne encephalitis (TBE) is the most important viral tick-borne zoonosis in Europe and causes between 5352 (in 2008) and 12,733 (in 1996) human cases per year in Europe and parts of Asia, especially in the Siberian part of Russia [1]. In Switzerland, where TBE has been treated as a notifiable disease since 1984 [2], about 100–130 cases are reported each year (the maximum was 244 cases in 2006). The reasons for these annual fluctuations in the incidence of TBE are not well understood. Across Europe, geographic variation in the prevalence of TBE in humans is largely dependent on climate factors that influence the questing activity of ticks [3]. In addition, anthropomorphic changes in agriculture and outdoor and leisure activities influence the risk that humans will contract TBE [4]. Theoretical models that examine how climate change will influence tick ecology predict that the TBE virus (TBEV) will spread to the north and to higher altitudes over the next decades [5]. Field studies have confirmed that TBEV has spread northwards in Norway [6] and to higher altitudes in the Czech Republic and Austria [7–10]. The spread of TBEV to higher altitudes is also an important concern in Switzerland.

TBEV is a member of the Flavivirus genus that includes the yellow fever virus and the dengue virus [11]. In Central Europe, the main vector for TBEV is the hard tick Ixodes ricinus, which has three blood-feeding stages: larva, nymph, and adult. The larvae and nymphs maintain TBEV in nature because they feed on the same group of TBEV-competent reservoir hosts, mainly wild rodents [5]. Larval ticks acquire the virus after feeding on an infected rodent, but this mode of transmission is relatively inefficient because the duration of infectivity to ticks is short (2-3 days) [12]. Other studies have shown that TBEV can be found in rodent tissues at 10 to 50 days post-infection [13]. Compared to other tick-borne pathogens, the prevalence of TBEV in I. ricinus populations is generally very low (< 1.0%) [14, 15]. Theoretical models have shown that the short duration of infectivity is the main reason why TBEV has such a low prevalence in nature [16–18]. Larval ticks can also acquire TBEV by co-feeding transmission where they feed in close proximity to an infected nymph on the same reservoir host [12, 18, 19]. Co-feeding transmission is a fragile mode of transmission because it depends on the synchronized questing activity of larval and nymphal ticks, which in turn, depend on a particular set of climatic conditions [5]. This particular set of climatic conditions is one reason why TBEV has a patchy geographic distribution across Europe [5]. Even in areas where TBEV is endemic, the presence of the virus in ticks and reservoir hosts is often highly focal [20, 21]. Pawlovskij pointed out that a natural TBEV focus depends on a number of botanical, zoological, climatical and geo-ecological conditions [22].

In veterinary medicine, clinical cases of TBE are rare, but have been reported in horses [23] and dogs [24, 25]. Other species like goats, sheep and cattle develop antibody titres without exhibiting clinical signs. These species are of high relevance for the so-called alimentary TBE. During viraemia, TBEV is excreted into the milk and can be ingested via consumption of raw milk or raw milk products such as cheese. While TBE in humans is mostly caused by tick bites, cases of alimentary TBE have been reported in recent years from Slovakia [26], Estonia [27], the Czech Republic [28], Austria [9], and Hungary [29].

Many studies have surveyed populations of wild I. ricinus ticks for the prevalence of TBEV [30–32]. The advantage of this approach is the direct detection of the virus in the tick vector. The disadvantage is that testing ticks for TBEV is time-consuming and expensive. Due to the low prevalence of TBEV in ticks (0.1–5.0% [14, 15]), and the high variability of TBEV prevalence in space and time, there is much interest in developing alternative methods to assess the human risk of TBE [33]. One such method is the detection of TBEV-specific antibodies in sentinel vertebrate hosts. Various wild and domestic vertebrates such as rodents [13, 34, 35], roe deer [36], goats [37, 38], sheep [39], dogs [40, 41], and even brown bears [42] have been used as sentinels in endemic areas. A disadvantage of this method is that detection of antibodies does not provide any information on the time and place of infection. An advantage of this method is that vertebrate hosts can feed many ticks and therefore “amplify” the TBEV signal in a given area. Another advantage is that screening vertebrate serum samples for TBEV-specific antibodies is fast and cheap. Goats, sheep and horses may be especially well-suited as sentinel hosts because they graze in meadows over long periods each year and are therefore potentially exposed to many TBEV-infected ticks. In addition, small ruminants and horses are kept in locations that are well known to the owner, which provides information on the location of TBEV infection. These locations can then be sampled for ticks to test for the existence of an endemic TBEV focus [43, 44].

Since 2013, the Swiss Federal Office for Public Health (FOPH) has edited two different maps with respect to TBEV. One map shows the risk of TBE infection and is based on the frequency of human TBE cases. The other map indicates the areas where the FOPH recommends prophylactic vaccination against TBEV. This map describes a risk area by using all relevant information, including human cases of TBE and TBEV-positive ticks (Bull OFSP 18/2013). A recent area of interest in Switzerland with respect to TBEV is the canton of Valais, which is located in the south of Switzerland. In this canton, 17 human cases have been described within the past 10 years, 15 of them since 2010. In 2009, two foci of TBEV were identified in Valais by a large nation-wide survey that screened populations of I. ricinus ticks [30]. Over the following years, the presence of TBEV in these areas was confirmed. A large tick survey that sampled more than 19,000 ticks across the canton of Valais found four new foci close to the two original ones [45, 46]. The aim of the present study was to use goats as sentinels to confirm existing TBEV risk areas and to detect new ones. The value of goats as sentinel animals was confirmed by analyzing ticks collected from areas identified by sero-positive goats. Our study demonstrates that testing antibodies in goats is an effective method for detecting new foci of TBEV.

The most important result was that we were able to detect two new TBEV foci out of three potential sites based on the initial detection of seropositive goats and the subsequent confirmation of TBEV-positive ticks [45, 46]. The identification of TBEV foci in I. ricinus tick populations is difficult for a number of reasons. The geographic distribution of TBEV foci is highly patchy, and the foci are often very small with an area of about 100 m² [20, 21]. At such foci, the percentage of infected ticks is generally very low (<1%). Identification of new TBEV foci in the tick population therefore depends on strong a priori evidence that such a focus exists. Goats are interesting sentinel animals to use for sero-epidemiological surveys of TBEV. In Switzerland, goats are kept in the same two or three enclosed pasture sites year after year. If a TBEV focus is present, the goats will encounter infected ticks over the years and develop an antibody response. In summary, the fact that goats are kept in small, well-defined pasture sites in Switzerland makes them ideal sentinel hosts for screening for TBEV.

From an epidemiological point of view, the method used in the present study should be effective at identifying seropositive flocks without necessarily identifying every seropositive goat in that flock. In our study, we were confronted with some unexplained results. For example, one confirmed positive goat was detected in a flock of 51 goats in the village of Euseigne. Checking the origin of the seropositive goat confirmed that it was born and raised in Euseigne. This observation suggests that there may be a TBEV focus in the pasture sites of Euseigne even though this village is not located in the known endemic area of TBEV. An alternative explanation is that this serum sample produced a false positive on both the ELISA and the SNT, but such a result would be unlikely. As shown in a district in Thuringia, Germany, a single positive goat can provide evidence of a TBE risk area [38]. A sero-survey of the flock in Euseigne should be initiated in the future. Our study also showed the importance of knowing the origin of the goats. In a flock of goats in Reckingen, five sera tested positive, but all of these goats originated from the cantons of Bern and Solothurn, where TBEV is endemic. Thus, the conservative conclusion is that these goats were exposed to TBEV in these other cantons before being transported to the canton of Valais.

Statistical analyses are difficult due to the patchy pattern of TBEV foci and because in the present study, 50% of the goats originated from the known endemic area between Sierre and Visp. Data on sex and breed show that TBEV-positive goats are typical of the ‘average’ goat in the canton of Valais. Most of the goats in the canton of Valais are female (92.0% = 1262/1372) and most belong to the col noir breed (63.1% = 866/1372). Similarly, most of the TBEV-positive goats in the present study were female (51/52) and belonged to the col noir breed (80.8% = 42/52). This breed of goats is specific to the German-speaking region, which includes the known TBEV foci, and we experienced some difficulties in obtaining complete data from some of these goat flocks. It is unlikely that a TBEV infection is associated with a specific breed or sex. In contrast, breed is important in dogs where individuals with long and pale hair were more often infected by TBEV [48]. Seroprevalence increased with age as for many other infectious diseases. The mean age of the 52 positive goats was 3.4 years compared to the mean age of 2.8 years for the entire sample of goats. These findings were in agreement with other studies on sheep [47], where older animals with more than one season on the pasture were more likely to have TBEV antibodies than younger animals. The obvious explanation is that older animals are more likely to have been exposed to TBEV-infected ticks than younger animals.

The design of our ELISA allowed for quick screening of >4000 individual goat serum samples. A subsample of 175 of the most reactive goat serum samples was studied further using the absorption test (AT) and the serum neutralization test (SNT). Of the subsample of goat serum samples that were highly reactive on the ELISA, 26.9–33.1% tested positive on the AT (depending on how the equivocal samples were treated) and 40.4% tested positive on the SNT. Comparison between the AT and the SNT revealed that the AT was highly specific (98.1–100.0%), but with moderate sensitivity (71.4–77.1%; Table 2). The testing of 100 randomly selected serum samples by SNT demonstrated that the results of our ELISA were good. ELISA is the best and easiest assay to screen a large number of sera. Positive sera need to be checked using the gold standard SNT to confirm TBEV-seropositive status. A possible explanation for serum samples that test positive on the ELISA but negative on the SNT is infection with other flaviviruses that cross-react with the antigens of the ELISA assay. Possible flaviviruses that could result in cross-reactivity include the louping ill virus and the “ruminant” TBEV-like viruses that have been identified recently [49, 50].

The tick-borne encephalitis virus is classified into three subtypes that are found in different geographical areas: European (TBEV-Eu), Siberian (TBEV-Sib), and Far Eastern (TBEV-FEa) [11, 51, 52]. The European subtype is the least virulent subtype and is responsible for all human cases of TBE in Western Europe. In Switzerland, TBEV isolates circulating in ticks collected at 39 foci were closely related and all of them belonged to the European subtype [53]. In our previous study, we had shown that all tick-derived TBEV isolates from the canton of Valais belonged to the European subtype [45]. In vivo and in vitro studies of Swiss TBEV isolates suggest a high number of avirulent isolates, which is in agreement with a high proportion of subclinical or mild TBE infections in the Swiss public [53].

Our present study confirmed the utility of goats as sentinel animals in sero-epidemiological surveys of TBEV. Sero-surveys help to identify candidate sites where ticks should be collected to detect new TBEV foci. ELISA is an effective and simple tool to screen large numbers of sentinel hosts for antibodies against TBEV. Confirmation of ELISA-positive results is necessary and the SNT is the recommended gold standard. Knowledge of the origin of seropositive goats is essential to further define the potential presence of a new TBEV focus. Our sero-survey of goats in the canton of Valais revealed several hot spots that should be investigated in the future.