The importance of study duration and spatial scale in pathogen detection-evidence from a tick-infested island

Abstract

Ticks (Acari: Ixodoidea) are among the most common vectors of zoonotic pathogens worldwide. While research on tick-borne pathogens is abundant, few studies have thoroughly investigated small-scale spatial differences in their occurrence. Here, we used long-term cloth-dragging data of Ixodes ricinus and its associated, known and putative pathogens (Borrelia burgdorferi s.l., Borrelia miyamotoi, Anaplasma phagocytophilum, Rickettsia spp., Candidatus Neoehrlichia mikurensis, Bartonella spp., Babesia spp., and tick-borne encephalitis virus, TBEV) from a small, well-studied island in southwestern Finland to analyze potential temporal and spatial differences in pathogen prevalence and diversity between and within different biotopes. We found robust evidence indicating significant dissimilarities in B. burgdorferi s.l., A. phagocytophilum, Rickettsia, and Ca. N. mikurensis prevalence, even between proximal study areas on the island. Moreover, during the 6 years of the ongoing study, we witnessed the possible emergence of TBEV and Ca. N. mikurensis on the island. Finally, the stable occurrence of a protozoan pathogen that has not been previously reported in Finland, Babesia venatorum, was observed on the island. Our study underlines the importance of detailed, long-term tick surveys for public health. We propose that by more precisely identifying different environmental factors associated with the emergence and upkeep of enzootic pathogen populations through rigorous longitudinal surveys, we may be able to create more accurate models for both current and future pathogen distributions.

Fig. 1

Estimated probabilities (with 95% confidence limits) of nymph samples being positive for B. burgdorferi s.l. (a) and A. phagocytophilum (b) across biotopes, as predicted by the GLMM. Different biotope classes were assigned matching letters when no statistically significant differences between them could be identified (p > 0.05; multiple pairwise comparisons adjusted by Tukey test). Mismatching letters denote statistically significant differences between biotope classes with different letters (p < 0.05)

Fig. 2

Estimated probabilities (with 95% confidence limits) of nymph samples being positive for B. burgdorferi s.l. (a, c), A. phagocytophilum (b), Rickettsia (d), and Ca. N. mikurensis (e) within biotopes (across transects), as predicted by the GLMM. Different transects were assigned matching letters when no statistically significant differences between them could be identified (p > 0.05; multiple pairwise comparisons adjusted by Tukey test). Mismatching letters denote statistically significant differences between transects with different letters (p < 0.05). C_1-3, transects in coniferous forest; D_1-3, transects in deciduous forest

Fig. 3. Study location.

The location of the study area, Seili Island, in the Archipelago Sea, SW Finland

Fig. 4. Study transects on Seili Island. Three transects were assigned to each biotope type, denoted with letter and number combinations.

C coniferous forest, D deciduous forest, A alder thicket, M meadow, P pasture. The Archipelago Research Institute is located in the middle of the island