Abstract

Introduction

Tick-borne encephalitis (TBE), caused by the tick-borne encephalitis virus (TBEV), is the most important tick-borne zoonotic disease in Europe and Asia from a medical perspective (Süss, 2011). Indeed, even though TBEV infection in humans often causes unspecific febrile symptoms and remain unnoticed, it can result in severe neurological diseases, including encephalitis, meningitis, and meningoencephalitis, with frequent incomplete recovery and, though rarely, death (Bogovic et al., 2010; Kohlmaier et al., 2021). TBEV is a positive-sense, single-stranded RNA virus of the genus Flavivirus, family Flaviviridae that circulates preferably among ticks of the genus Ixodes and small mammals, but large mammals and migrating birds also contribute to the virus geographic distribution and dispersion (Labuda et al., 1997; Randolph et al., 1999; Waldenström et al., 2007; Lindquist and Vapalahti, 2008; Gunnar et al., 2009; Jaenson et al., 2012; Tonteri et al., 2016; Paulsen et al., 2019). TBE most often results from tick bites, however, cases can also result from consumption of unpasteurised milk or dairy products from infected cows, goats, and ewes (Gresíková et al., 1975; Holzmann et al., 2009; Balogh et al., 2010; Cisak et al., 2010; Caini et al., 2012; Hudopisk et al., 2013; Markovinović et al., 2016; Brockmann et al., 2018; Dorko et al., 2018; Ilic et al., 2020; Chitimia-Dobler et al., 2021). Indeed, ruminants are also infected when bitten by an infected tick. Even if they remain asymptomatic, virus excretion in milk has been reported for cows, sheep, and goats (Gresíková et al., 1975; Balogh et al., 2010; Cisak et al., 2010; Caini et al., 2012). Pasteurisation prevents TBEV transmission through inactivation of virus infectivity (Rónai and Egyed, 2020). The increasing geographic range of TBEV and its patchy distribution around local foci puts an increasing number of countries at risk of sporadic occurrence of TBEV infections that can both be difficult to diagnose and prevent (Blaskovic, 1967; Deviatkin et al., 2020).

Tick-borne encephalitis virus (TBEV) is endemic throughout Central and Eastern Europe and has so far been detected in twenty-seven European countries. In the past decades, TBE human incidence has been increasing in several European countries (ECDC, 2012, 2018), with new foci detected in the United Kingdom (Holding et al., 2020) and Netherlands (Jahfari et al., 2017), expanding into previously unaffected areas (Daniel et al., 2004; Lukan et al., 2010; Donoso Mantke et al., 2011; Martello et al., 2014; Brabec et al., 2017; Rieille et al., 2017; Hellenbrand et al., 2019; Alfano et al., 2020). These observations, associated with evidence of a recent increase in TBEV diversity in Europe, suggest that TBEV may be regarded as an emerging disease (Deviatkin et al., 2020). France is located on the border of the known distribution of TBEV in Europe, and very few human cases are usually diagnosed annually, mainly in the Alsace region (about 10–30 cases per year, 0.5/100,000 inhabitants) based on suspect clinical presentation and the detection of TBEV-specific antibodies (Hansmann et al., 2006; Herpe et al., 2007; Donoso Mantke et al., 2011; Velay et al., 2018).

Herein, we report the first outbreak of alimentary TBEV in France and describe the investigation that was conducted within the suspected farm and surrounding area. In April 2020, an outbreak of encephalitis and meningoencephalitis occurred in the Ain department (Eastern France) where TBEV had never been detected. Consequently, a sanitary alert was decreed on May 10. Within a month, epidemiological investigation evidenced that 43 patients with encephalitis, meningoencephalitis, or flu-like symptoms, except for one, had consumed fresh goat cheese made of raw milk named “faisselle” coming from a single local producer. Immediately, the suspected goat flocks were confined into stall, unpasteurised goat cheese production was stopped, raw cheese produced before the alert was withdrawn from commercial markets, and an investigation was conducted within the farm. We confirmed the alimentary transmission by demonstrating the presence of TBEV in a batch of cheese and goat milk. We also monitored the serological status and virus excretion in milk of the animals after control measures had been established and demonstrated their efficacy since no TBEV infection has been reported later than early June. We assessed enzootic hazard for animal exposure through tick bites measured as the density of infected ticks in the pasture and in the neighbour woodland. The full-length virus genome sequence was obtained from milk, cheese, and tick samples, allowing the virus molecular characterisation, and an isolate was obtained. Finally, we assessed TBEV seroprevalence in surrounding farms to evaluate whether TBEV had been silently established in the area or not.

Materials and Methods

An overview of the timeline of sampling collections is provided in Figure 1.

Open in a separate window

FIGURE 1

Timeline of the food, environmental and veterinary investigations, and sampling scheme from April to November 2020.

Results

Tick-Borne Encephalitis Virus Infection and Exposure in Suspected Farm Livestock

IgG and IgM antibodies against TBEV were detected using the ELISA tests (Igs competition and IgM capture) and MNT. On May 27, six goats out of thirty were TBEV-seropositive as assessed with competition ELISA and MNT. One of them had anti-TBEV IgM antibodies. The following weeks, after their placement in a stable to prevent further tick bites, all 56 goats were tested for a complete assessment of the TBEV exposure rate. Eleven goats were TBEV-seropositive on June 2 (20%) (with one seroconversion in the 30 goats collected on May 27) and 13 (23%) on June 9 (with two more seroconversions) (Figure 2). Goat seroconversions were confirmed by MNT (Figure 2, Table 1, and Supplementary Data). Overall, the ELISA results were confirmed by MNT with the exception of a single individual that appeared positive based on ELISA competition but did not develop neutralising antibodies (Figure 2). The seroconversion of three goats between May 27 and June 9 indicates that infection occurred very recently. Yet, none of the animals had high viremia given that all qRT-PCRs conducted on blood samples were negative (collected on June 2, 9, and 19). ELISA competition conducted in 2021, about 1 year after the initial infection, showed a decrease in IgG levels for most seropositive individuals (10/13), but the level of neutralising antibodies remained high. Two individuals that were TBEV-seropositive in 2020 turned negative based on competition ELISA in 2021, and one of those had no remaining neutralising antibodies according to MNT (Figure 2). Within the same farm, one of the three dairy cows and two of the four suckling cows which had grazed in the same or in the adjacent pasture as the goats were TBEV seropositive. No virus genome was detected in the blood of these animals.

Open in a separate window

FIGURE 2

IgG levels (A) and micro-neutralisation antibody levels (B) in goats on four dates in 2020 (May 27, June 2, June 8, and June 19) and on April 2, 2021. The threshold levels applied are indicated as dashed lines. We provide examples of seronegative individuals (green) and show results for the individual that returned seropositive via ELISA competition, but returned negative in neutralising antibody titre (grey), the three individuals that seroconverted after May 27 (yellow), the eight seropositive individuals (red), and the two individuals that returned negative IgG competition results in 2021 (blue), one of which maintains neutralising antibodies.

Seroprevalence Survey in Surrounding Farms

Within a couple of kilometers away from the suspected farm, five farms that were hosting cows and goats were identified, and sera from all individuals were collected and processed as above. Few seropositive animals were found in three of the five farms, with one to five seropositive animals and a TBEV seroprevalence ranging from 5.5 to 25% (Figure 2 and Table 2). Animals from the suspected farm and from farms #1 and #3 within which seropositive animals were found have all grazed in meadows located in close proximity to the same forest (Figure 3). On farm #4, a single 10-year-old cow was TBEV-seropositive, but it was not possible to trace back where the cow had grazed in the past.

TABLE 2

Number of questing ticks collected and mean density per habitat.

Habitat	No. sub-areas	Sampled surface (m2)	No. nymphs	No. female	No. male	Density of nymphs (/100 m2)
						mean	Range
Forest	3	540	382	10	20	71.5	64.2–81.3
Additional collection	/	/	285	7	5	/	/
Pasture
Wooded area	3	300	105	6	3	31.7	20.0–50.0
Wooded edge	5	400	94	2	2	24.0	5–56.7
Hedgerows	7	870	4	0	0	0.3	0–1.1
Meadow between 3 and 10 m from the wooded edge	4	530	35	2	1	4.5	0–8.8
Meadow between 3 and 10 m from the hedgerow	4	400	2	0	0	0.3	0–1.1
Meadow > 10 m from the edge	6	600	0	0	0	0	0

Open in a separate window

Open in a separate window

FIGURE 3

Localisation of the farm responsible for the tick-borne encephalitis cluster through alimentary route (star) and results of the serological survey conducted in surrounding farms. Serum samples were first tested by a competitive ELISA. Then, positive and doubtful ELISA samples were tested by specific tick-borne encephalitis virus (TBEV) micro-neutralisation tests (MNT). A serum sample was considered TBEV-seropositive when MNT was positive. Dairy cows, suckling cows, goats, suckling cows, and goats were found in farms #1 to #5, respectively.

Phylogenetic Analysis

Three RNA samples were extracted from two TBEV-positive pools of nymphs and one TBEV-positive male tick, and RNA samples obtained from three individual infected-goat milk were subjected to sequencing. Our amplicon-based sequencing protocol allowed us to successfully sequence the TBEV genome from RNA extracted from milk and ticks (Figure 4). The full-length genome sequence obtained from milk and ticks was similar to the full-length genome obtained from the contaminated cheese collected during the investigation led by a local authority agency. Sequences obtained from the milk sample, and the virus isolate obtained from ticks were 100% identical except for single nucleotide insertions and deletions that are most likely due to sequencing errors.

Open in a separate window

FIGURE 4

Full-length TBEV genome sequence (GenBank accession number: OL441148) was obtained from sequencing 35 overlapping amplicons in two multiplex reactions. Read coverage varied along the sequence length, but a full-length genome sequence was successfully recovered.

The full-length genome of the new viral strain was sequenced and compared to previously characterised TBEV genome sequences. The TBEV_Ain_France_2020 had limited homology (98.4% identity) to the partial NS5 of the TBEV from the Alsace region (AF0910) (Supplementary Data). The nucleotide sequence presents a 98.8% identity to the same partial NS5 and between 98 and 98.07% identity to the full-length genomes of TBEV isolates detected in Russia, Germany, Austria, Czech Republic, and the Netherlands with the most ancient isolate dating from 1951 in Russia. The polyprotein sequence presented up to 99.30 and 99.24% identity to MK922616 and MK922617 isolated from ticks collected in the German federal state of Lower Saxony in 2018. Our phylogenetic analyses confirmed that TBEV_Ain_2020 (OL441148) belongs to the European subtype (TBEV-Eu3) and is most closely related to TBEV strains recently isolated in bordering countries and Eastern Europe (Figure 5).

Open in a separate window

FIGURE 5

Phylogenetic tree of TBEV complete open reading frames that were available in GenBank and the TBEV-Ain-France-2020 obtained in the present study (marked red) using all sequences (A) or those corresponding to the European subtype only (B). The two trees were inferred in PhyML using the Le Gascuel (LG) substitution model. Branch points indicate that results of Shimodaira-Hasgawa branch test > 0.8. Scale bar shows the number of nucleotide changes. Virus isolate names are given as follows: accession number from NCBI_ country to name_year.

Discussion

Food-borne TBE outbreaks had been reported in several endemic countries in Central Europe at a regular frequency, particularly in countries where the consumption of traditional raw milk products is popular (Kerlik et al., 2018). France is on the border of the TBEV geographic range, and cases resulting from tick bites are reported very occasionally. The first occurrence of TBE-food-borne outbreak in France in 2020 has led to an unexpected high number of cases. Forty-three patients have developed clinical symptoms compatible with TBE, and all but one had consumed raw goat milk and/or cheese products originating from a single producer, while 41 cases could be confirmed by detection of TBEV-specific antibodies in sera and/or in cerebrospinal fluid. A health alert was triggered within a month after the first patient showed symptoms, and an in-depth investigation of TBEV presence was conducted in the farm suspected to be the source of contamination and in animals in surrounding farms. We applied a “one health” approach, combining environmental, veterinary, virology, and food microbiology approaches to identify the origin of infection and provide a range of epidemiological information of significance to public health decision-makers. We characterised the TBEV strain circulating in Ain, detected viral genome presence in a batch of cheese, goat milk, and in questing ticks, demonstrated recent infection of goats, assessed the enzootic hazard for TBEV exposure, and identified a wooded TBEV-focus area. This integrative investigation provides information that would allow the farmer and authorities to assess the risk of infection and develop control measures.

The initial investigation of seroprevalence (23%) in goat flocks at the suspected farm reveals that the animals had been moderately exposed to the virus these past few years. In neighboring areas where TBEV is endemic, such as Switzerland and Germany, seroprevalence varies from 1 to 80% with median range value of 10–20% (calculated from Klaus et al., 2012; Klaus et al., 2014; Rieille et al., 2017; Casati Pagani et al., 2019). Seroconversion results, presence of IgM antibodies, and detection of the virus genome in milk indicate that infection was very recent for half of the animals that had been exposed to the virus (6 animals over 13). IgM was detected in one individual. Three individuals seroconverted after entering the stable, indicating that TBEV infection occurred a few days before they entered the stable. Indeed, seroconversion in goats and sheep occurs in 6–10 days (Van Tongeren, 1955; Klaus et al., 2014; Paulsen et al., 2019). One of these individuals and two others excreted viruses in milk 14 days after entering the stable. These results also support a recent infection given that virus excretion in goat and sheep milk is detected from 2 to 23 days after infection (Gresikova, 1958; Gresíková et al., 1975; Balogh et al., 2010, 2012). So far, information is relatively scarce on the infection dynamics in ruminants, especially in naturally infected animals. Our investigation confirmed a long-lasting excretion of the virus in milk in naturally infected goats since three individuals excreted the virus into milk for over 14 days post-infection. The date of the seroconversion of the additional seven TBEV-seropositive goats and two cows from the same farm is unclear. Previous studies suggest that TBEV-antibodies may persist up to 6 years, but the level of IgG produced decreases over time (Klaus et al., 2014, 2019). In agreement, even though the level of neutralising antibodies remain high a year later, we observed a decrease in the level of IgG detected by ELISA. Thus, the high level of IgG, as measured by ELISA competition, indicates that the remaining animals have been likely infected or re-infected by the virus in 2020. Since no virus excretion has been observed in five experimentally pre-immunised goats (Balogh et al., 2012), only naïve individuals would have excreted TBEV in milk. The high recent infection rate among naïve individuals of the flock would explain the high virus load detected in tank milk in June. Milk and cheese viral load were determined using a genome quantification assay, whereby milk sample concentration was estimated based on cycle threshold (Ct) values derived from measurements of serial dilutions of infectious TBE viral particles with known concentrations (from 10⁸ to 10³ TCID50/ml). We evaluated a minimal detection of infectious viral particles at 10⁴ TCID50/ml of milk and 10⁵ TCID50/ml after dissolution of 2.5 g of cheese. In ticks, the mean viral copy number lies between 2 × 10² and 4.8 × 10³ RNA copies per sample (Liebig et al., 2020, 2021). Our data confirmed the high virus load detected in tank milk. Moreover, the virus was detected in seven cheese (over 91 tested). Among them, one contaminated cheese was produced on April 28. This is outstanding, especially when we consider that cheese was kept for several days at 1–5°C by consumers or producers before being analysed, which may have reduced the persistence of TBEV load and hence its detection.

The high recent infection rate of the goat flock suggests regular contact between TBE-infected ticks and goats in spring 2020, when ticks started to be active. We found infected questing nymphs and adults in the wooded edge of the pasture (in the meadow between the pasture wood and the forest) and in the forest nearby. With an MIR of 0.22% (IC_95%: 0.03–0.80%) in nymphs and an infection rate of 1.8% (IC_95%: 0.1–9.9%) in adults, it appears that TBEV is well established in the forest near the pasture. Indeed, while the densities and prevalence of infected questing ticks may not be deemed exceptionally high compared to what is seen in endemic countries (Süss, 2011), it is similar to the highest values observed in an endemic foci in Alsace region, East France (Bournez et al., 2020). Since we captured a few adult ticks (probably because the dragging method is not well suited to capture this stage), the density of infected adults was estimated with a low precision and hence might be higher than those observed in Alsace. We did not detect TBEV RNA in small mammals trapped in the pasture, but we probably missed infected individuals since we only trapped a few rodents. Indeed, the year 2020 was probably a low phase of the usual yearly fluctuating rodent population. Overall, the results suggest that the presence of a large wooded area within the pasture frequently used by goats as a shelter as well as of a long pasture-forest edge where tick densities are relatively high have likely favoured the contacts between goats and TBEV-infected ticks. The forest may also represent a source of frequent introduction of infected ticks into the pasture through the movements of tick-feeding hosts.

This cluster was surprising by its location. It occurred in the Ain department in France where TBEV human cases had never been reported. Our investigation in surrounding farms also revealed that TBEV was also present in various pastures surrounding the same single forest where we found infected ticks over at least 2 km. In three of these farms, seroprevalence varied between 5.5 and 25%. With the exception of one individual (for which it was not possible to track back the pasture in which it had grazed), all seropositive animals had likely grazed within a pasture adjacent to the same forest. These results indicate that the virus is well established in the area and, probably, in a larger extent in the Ain department. Because of the absence of an active TBEV surveillance system assessing its distribution in France and potential misleading of TBE diagnosis, it is impossible to conclude whether the presence of TBEV in the department is ancient or whether it was recently introduced. In addition, diagnosis of human TBE cases was initially delayed as TBE serology was not systematically researched among patients presenting with meningitis or meningoencephalitis. Similarly, the presence of TBEV in centre France, in Loire, and Haute-Loire departments detected for the first time in 2017 and 2018 (Herpe et al., 2007; Velay et al., 2018) is impossible to date. Given that the life cycle of ticks involves generally one meal per stage per year, we can presume that TBEV-infected ticks have been present in the environment for at least 1 or 2 years before the occurrence of human cases. Evidence suggest that the geographic range of TBEV is currently expanding in Europe, especially in Western and Northern Europe (Csángó et al., 2004; Donoso Mantke et al., 2011; Rieille et al., 2017; Andersen et al., 2019; Casati Pagani et al., 2019; Smura et al., 2019; Alfano et al., 2020; Ullrich et al., 2021). This expansion also have occurred in France. For instance, in a recent expansion of TBEV in higher altitudes, other previously unaffected areas or ancient affected areas have been observed in several France’s neighbouring endemic countries, including Germany (Dobler et al., 2019; Hellenbrand et al., 2019), Switzerland (Rieille et al., 2017; Casati Pagani et al., 2019; Desgrandchamps et al., 2019), and Italy (Alfano et al., 2020). The virus was also detected for the first time in the United Kingdom in 2018 (Holding et al., 2020) and in the Netherlands in 2015 (Jahfari et al., 2017). TBEV expansion is believed to be driven by climate, landscape, and anthropogenic changes (Randolph, 2010; Jaenson et al., 2012; Holding et al., 2020). Moreover, the first occurrence of TBEV in new areas in France since 2017 might be the results of exceptional high densities of infected ticks these past 5 years in endemic areas, which may have led to higher dispersal rates of infected ticks into new areas. Indeed, although TBE human incidence usually exhibits significant oscillations over time, in 2016–2020, it was persistently high in numerous endemic countries of Western and Central Europe (Germany, Switzerland, Czech Republic, Slovakia, and Italy), including France (Dobler et al., 2019). This may partly be caused by high densities of infected ticks during those years. A high density of infected ticks during several years in endemic areas may increase the frequency of medium or long-distance movements of infected ticks on their host (birds, cervids) and led to a more successful introduction of the virus into new areas. Alternatively, it is also possible that TBEV has been circulating in Ain department undetected for decades, and the 2020 outbreak could be the result of suitable conditions for its detection. Indeed, TBE cases are likely underdiagnosed in France given that human clinical cases are rare and local clinicians are unfamiliar with the disease (infection often asymptomatic, not yet notifiable disease in April 2020, low virus circulation, few TBEV foci, few human visits in infected areas) (Dollat et al., 2021). The year of 2020 was exceptional with a very high human incidence observed in numerous endemic European countries. In Switzerland and Germany, the notable increase of TBE cases observed in spring 2020 was partly associated with higher recreational activities in risky endemic areas around people’s homes during the COVID-19 containment measures (Steffen et al., 2020; Ullrich et al., 2021). In France, the COVID-19 containment measures may have also indirectly increased the consumption of local products. Hence, the emergence of a cluster of food-borne TBE cases in a restricted geographical area that was easier to detect. However, this was not the sole factor as an upsurge of infected tick densities in 2020 likely contributed to the rise of human incidence. Indeed, models forecasting the human TBE incidence based on demographic parameters, large-scale atmospheric circulation patterns, the fructification of the European beech (Fagus sylvatica) 2 years prior, and the national TBE vaccination coverage already predicted an increase in TBE incidence in 2020 in Germany, Switzerland, and Austria (Rubel and Brugger, 2020, 2021). In Germany, an unusually high density of infected adult ticks has also been observed in 2020 in some places (PROMED, 2020).

We provide the first TBEV isolate responsible for a source of dietary contamination in France, which will facilitate experimental studies. The genome sequences obtained were used to conduct phylogenetic analyses using the complete open reading frame. We revealed that the TBEV_Ain_France_2020 belonged to the European TBEV subtype, but do not cluster with the strain circulating in the Alsace region and to other isolates from Europe, with the closest relatives being isolates from Germany, Austria, Czech Republic, The Netherlands, and Russia. In all actuality, we do not know enough about the distribution of TBEV in France and the genetic diversity of the TBEV genome in Europe. The distinct genome of the French isolate would tend to indicate that this isolate has not recently spread to France but has remained undetected and evolved locally for some time. Further studies in the department and neighbouring regions are needed and are being set up to further characterise the focus-specific genetic diversity of TBEV and to confirm or disprove the ancient origin and geographical extension of this isolate.

In conclusion, the cluster of TBE alimentary cases in Ain was unexpected. It surprised clinicians and local health authorities given the number of infected patients. The high recent infection rate of the goat flock probably led to a high virus load in tank milk in April and May and facilitated infection by consumption of raw cheese. In addition, the COVID-19 containment measures in spring 2020 may have also indirectly increased the consumption of local products and, hence, the risk of food-borne TBE cases. The Ain outbreak also highlighted the need for improving surveillance, detection, and prevention of TBE in France. Clearly, a better understanding of TBEV distribution in France, of local virus dynamic, and of food-borne contamination with better identification of raw milk products being at risk is needed to better adapt surveillance and prevention measures. This is all more important as France produces each year 40,000 tonnes of raw milk-products and cheese. In addition, the local production and consumption of traditional delicacies is increasing. TBEV is now a notifiable disease in France, facilitating the systematic report of human cases and sensitising medical communities for diagnosis of humans showing encephalitis in France. In endemic regions and those surrounding affected areas, patients with meningitis and encephalitis in the absence of another etiological diagnosis should be more systemically screened for TBEV infection (Herpe et al., 2007; Velay et al., 2018; Botelho-Nevers et al., 2019; Blanchon et al., 2020). Surveillance could also include serological testing for anti-TBEV antibodies in animal blood or in milk tanks to allow early detection of virus foci (Imhoff et al., 2015; Rieille et al., 2017; Wallenhammar et al., 2020; Bauer et al., 2021). Finally, in focus areas, TBE vaccination could provide protection from disease caused by infections from tick bites or consumption of virus-contaminated dairy products (Chernokhaeva et al., 2018; Kubinski et al., 2020; Chitimia-Dobler et al., 2021).

Data Availability Statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

Ethics Statement

Ethical review and approval was not required for the animal study because the species captured during this study are neither protected in France nor included in the International Union for Conservation Nature Red List of threatened species and, therefore, no special authorisation was needed for trapping, according to French law. Animal trapping took place with permission from the landowners. All efforts were made to minimise animal suffering. The traps that we used are designed not to stress or harm the animals. Small mammals were euthanised by authorised experimenters according to French law and to the European guidelines (AVMA Guidelines for the euthanasia of Animals: 2020 editions). Written informed consent was obtained from the owners for the participation of their animals in this study.

Author Contributions

All authors contributed to the article and approved the submitted version.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Acknowledgments

Footnotes

References