Seasonal and inter-annual variation of TBEV prevalence in ticks and rodents in 2012-2014
Although the variations were low, we detected an annual and seasonal effect on NIP (the TBEV prevalence of questing nymphs) but neither on DIN nor rodent seroprevalence. Our study site in 2012-2014 was characterised by a high inter-annual and inter-seasonal variation in small mammal and questing nymph densities. Annually, the peak of activity for questing larvae and nymphs, of larvae feeding on small mammals, and of small mammal density occurred during the same season, i.e. in June-July. The density of bank voles and yellow-necked mice started to decrease from September. This decrease was greater for yellow-necked mice than for bank voles, only 13% of yellow-necked mice being captured in September-October. This seasonality is similar to field observations in TBEV-infected sites in neighbouring countries (C. Burri et al., 2011; S. E. Randolph et al., 1999; Rosà et al., 2007), but differs from findings in Brittany, western France, where the peak of small mammal densities and of feeding larvae were found to occur in autumn (Perez et al., 2017). Our observations on inter-annual variations comply with previous observations (Kiffner et al., 2011; Krawczyk et al., 2020; Ostfeld et al., 2018; Perez et al., 2017; Rosà et al., 2007): we found that the density of questing nymphs was higher the year following a year of high rodent density and that the intensity of tick aggregation on small mammals was higher when the density of small mammals was low. Indeed, the prevalence of infestation by both stages — larvae and nymphs — and the mean infestation was higher in 2013 (a low-density year for small mammals) than in 2012 (a high-density for small mammals).
NIP was higher in both 2012 and 2014 compared than 2013; this is contrary to the density of questing nymphs, for which we found opposite variations. However, DIN was near-constant over those years. Therefore, our results suggest that the proportion of infected larvae produced was higher in 2013 when more larvae and nymphs fed on the same very few hosts because there were fewer hosts available. This led to a higher proportion of infected questing nymphs the year after. This is consistent with the fact that the proportion of larvae feeding on rodents infested by nymphs was higher in 2013 than 2012. Therefore, the inter-annual difference in NIP can be explained by the variation in aggregation intensity of ticks on rodents along with rodent abundance. However, the overall number of fed larvae produced was probably lower in 2013, the year of a lower abundance of small mammals. Consequently, the DIN stayed constant over the years. Our study was only conducted over a 3-year period and the results cannot be generalised. Further investigations over a longer period would be needed to better understand the relationship between DON, DIN, NIP and small mammal density for TBEV. Few studies have investigated these relationships for other tick-borne pathogens amplified by rodents. In North America, Ostfeld et al. (Ostfeld et al., 2018) studied for 19 years the relationship between the Ixodes scapularis tick, the Borrelia burgdorferi sl. bacteria and the white-footed mouse (Peromyscus leucopus ). They found no effect of white-footed mouse density on NIPBorrelia burgdorferi sl. whereas DIN NIPBorrelia burgdorferi sl was positively but not linearly related to mouse density in the previous year. At low mouse density values, DIN Borreliaburgdorferi sl. was almost constant. In a short-term study in the Netherlands, Krawczyk et al. (Krawczyk et al., 2020) observed that the relationship between NIP, DIN and the rodent density the previous year depended on the transmission mode of the tick-borne pathogens. Contrary to most tick-borne pathogens amplified by rodents, TBEV is mainly transmitted by co-feeding and the nymph-to-larva transmission of TBEV strongly depends on the intensity of tick aggregation on small mammals. These elements probably induced a different relationship between DIN, NIP and rodent density that it would be interesting to investigate more.
Although the difference was small, TBEV prevalence in questing ticks was higher in autumn than in spring. This finding had already been observed in the previous study conducted in Alsace (Perez-Eid et al., 1992) and in studies targeting other tick-and-rodent-borne pathogens in Germany and Luxembourg (May, Jordan, Fingerle, & Strube, 2015; Reye, Hubschen, Sausy, & Muller, 2010). There are no data on the timings of diapause and moulting of I.ricinus stages in the Alsace region. However, if larvae that feed early in the season (in April-May) become nymphs that quest later the same year (in July-October) as observed in Switzerland (Perret et al., 2004), then this could be a potential mechanism to explain the higher prevalence in questing nymphs in autumn. Indeed, TBEV transmission to larvae might be higher in June-early July since this period coincides in our study area with the peak abundance of questing larvae and nymphs, of larvae feeding on small mammals and of small mammals themselves, especially Apodemus sp ., which has been shown experimentally to be more efficient in fostering TBEV transmission than M. glareolus (M. Labuda et al., 1993; Milan Labuda et al., 1997). Moreover, the virus titre may drop over time after the moult and in unfed ticks undergoing a winter diapause, as observed experimentally (Mishaeva and Erofeeva (Mishaeva & Erofeeva, n.d.) cited by Perez-Eid et al. (Perez-Eid, 1990)), which would also induce lower prevalence in early spring.
Surprisingly, from 2012 to 2014 we did not detect any annual or seasonal effects on rodents’ TBEV infection probability in keeping with the variation in the probability of a rodent being infested by ticks, whereas DIN values were constant. A higher seroprevalence could be expected in those years or seasons of low rodent abundance along with the higher prevalence of tick (nymph) infestation on rodents, as observed in other studies (Caroline Burri et al., 2012) and for other tick-borne pathogens amplified by rodents (Perez et al., 2017). Similarly, there was no species effect on the infection probability despite A. flavicollis being found to be more infested by ticks in this study (especially in spring when A. flavicollis is abundant). This apparent contrast may result from a difference in the immune response between species with a lower TBEV antibody titre and persistence in A. flavicollis compared with M. glareolus(Heigl & Von Zeipel, 1966; Knap et al., 2012; Milan Labuda et al., 1996, 1997; Zeipel & Heigl, 1966). We could then have expected to see an annual or seasonal difference in the infection probability per species, but our sample size of rodents was probably too small to detect a significant difference given the low seroprevalence of rodents, the low persistence of the antibodies and the low DIN. Another explanation for the absence of any annual effect on small mammal seroprevalence could be a lower detection of seropositive small mammals the years when they were scarce. Indeed, during those years, small mammals may acquire the infection much earlier in the year given the high tick aggregation level when fewer hosts are available. The exposure to this infection would probably remain undetected since the anti-TBEV antibodies’ half-life is very short in small mammals.
In conclusion, this study shows that the virus was circulating at a very low level in our study site. Despite this very low-level circulation, we were able to observe a significant variation in the inter-annual and inter-seasonal prevalence of TBEV in questing nymphs, indicating that the nymph-to-larva transmission of TBEV varied over time. However, the density of infected questing nymphs showed no detectable variations over time, suggesting that the rate of exposure for humans was probably relatively constant over the period of the study. The seroprevalence of small mammals was constant over time although the prevalence of tick infestation varied on an annual and seasonal basis. More studies are needed to understand the relationship between the density of TBEV-infected questing infected nymphs, the density of questing nymphs, the prevalence of TBEV in questing nymphs, TBEV seroprevalence in small mammals and small mammal density.