Candidatus Neoehrlichia mikurensis and its co-circulation with Anaplasma phagocytophilum in Ixodes ricinus ticks across ecologically different habitats of Central Europe

Background Candidatus Neoehrlichia mikurensis is a newly emerging tick-borne bacterium from the family Anaplasmataceae. Its presence in Ixodes ricinus ticks was reported from various European countries, however, it’s ecology and co-circulation with another member of the same family, Anaplasma phagocytophilum has not been rigorously studied yet. Findings Candidatus N. mikurensis was detected in all sampling sites. In total, 4.5% of ticks were positive including larvae. The highest positivity was detected in Austria with a prevalence of 23.5%. The probability of Candidatus N. mikurensis occurrence increased with the proportion of ticks infected with Anaplasma phagocytophilum. Conclusion A positive association between the occurrences of Candidatus N. mikurensis and A. phagocytophilum indicates that both bacteria share similar ecology for their natural foci in Central Europe.

In Europe, Candidatus N. mikurensis represents a newly emerging tick-borne zoonotic bacterium from the family Anaplasmataceae. Phylogenetic analyses revealed that it is closely related to the Ehrlichia-like microorganisms previously detected in ticks and rodents from various regions of Europe and Asia [1][2][3][4][5][6]. Recently its pathogenicity was reported, as it was detected in immunosuppressed patients with septicaemia [7][8][9]. Rodents are the competent reservoir hosts since they develop a systemic infection [1,3,4,6] and are able transmit Candidatus N. mikurensis to the xenodiagnostic ticks [10]. The prevalence of Candidatus N. mikurensis in ticks over Europe varies, usually not exceeding 10%. Most reports are from Western Europe [2,4,5,11]. Recently it was reported in questing I. ricinus from Hungary [12] and Austria [13]. Here we report the prevalence of Candidatus N. mikurensis from 11 diverse ecological habitats from three Central European countries and its co-circulation in natural foci with Anaplasma phagocytophilum.
A total of 1535 (755 adults, 614 nymphs, 140 larvae, and 26 individuals for which the developmental stage was not identified) and 1413 (756 adults, 621 nymphs, 10 larvae, and 26 individuals for which the developmental stage was not identified) I. ricinus ticks from three Central European countries (Slovakia, the Czech Republic and Austria) ( Figure 1) were tested for the presence of Candidatus N. mikurensis and A. phagocytophilum, respectively. Ticks were sampled from diverse habitats ( Table 1) by blanket dragging. DNA was extracted from single individuals by DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany). The presence of Candidatus N. mikurensis was detected by RT-PCR of groEL gene as described before [4] or by nested PCR of the specific fragment of 16S rRNA gene [1]. A. phagocytophilum was detected by RT-PCR of msp2 according to a previously described protocol [14] or by nested PCR amplifying the specific 546 bp fragment of 16S rRNA [15].
Candidatus N. mikurensis was detected in all 11 sampling sites. In total, 69 (4.5%) of 1535 ticks were positive. The prevalence of Candidatus N. mikurensis ranged from 1.1% to 23.5% (Table 1). In Austria at the site Kundl, Candidatus N. mikurensis was detected in 4 of 10 questing larvae.
In Total, 1413 ticks were tested for A. phagocytophilum and 53 (3.8%) were positive. It was detected in all sampling sites, but one ( Table 1). None of the tested larvae carried A. phagocytophilum.
Furthermore, we analyzed the relationship between the occurrence probability of Candidatus N. mikurensis and the proportion of ticks infected with A. phagocytophilum with a generalized linear mixed model (GLMM). The number of ticks infected with Candidatus N. mikurensis was entered as a dependent variable and was linked with a binomial error to the number of all ticks from a given  site and tick developmental stage. The proportion of ticks infected with A. phagocytophilum and the developmental stage of ticks were examined as fixed factors. As ticks for each site were examined at two developmental stages (nymphs and adults), site identity was entered as a random factor; three sampling sites from Austria were pooled due to sample size limitation. The probability of tick infection with Candidatus N. mikurensis increased with the proportion of ticks infected with A. phagocytophilum. The occurrence probability of Candidatus N. mikurensis did not differ between adult and nymphal ticks ( Table 2). The solutions of random effects revealed that the occurrence probability of Candidatus N. mikurensis for the Austrian sites was significantly higher than the mean occurrence probability (estimate ± SE = 1.08 ± 0.43, t 6 = 2.53, p = 0.039).
We have confirmed the permanent circulation of Candidatus N. mikurensis and A. phagocytophilum in each of the three examined countries of Central Europe across a wide ecological spectrum of habitats ( Table 1). The highest prevalence of Candidatus N. mikurensis (23.5%) was observed in Austria. A similarly high prevalence (24.2-26.6%) was found for questing ticks from Germany [16]. These are so far the highest prevalence results reported for Europe. Moreover, in Austria we have detected four positive questing larvae. Up to this date, the transovarial transmission has not been reported for Candidatus N. mikurensis. However, to our knowledge the questing larvae were examined for the pathogen only at one site in The Netherlands, by Jahfari et al. [4]. As the mode of pathogen transmission by vectors is of high epidemiological significance [17], possible transovarial transmission of Candidatus N. mikurensis should be elucidated in future studies.
The reservoir competency of rodents for Candidatus N. mikurensis have been recently confirmed [10]. As for A. phagocytophilum, it is unlikely that rodents are important reservoir hosts of the genotypes that are transmitted by I. ricinus. Based on the phylogenetic analyses of several genes, rodents in Europe are infected with distinct genotypes from that found in questing I. ricinus ( [18], unpublished observation). Moreover, recent study showed that rodents infected with A. phagocytophilum were not able to transmit it to xenodiagnostic larvae [10]. The reservoir competence of other hosts for Candidatus N. mikurensis needs to be elucidated, since it was detected in the ticks feeding on red deer, mouflon and wild boar [4].

Conclusions
We have revealed a positive association between the occurrences of Candidatus N. mikurensis and A. phagocytophilum. This finding indicates that both bacteria share similar ecology for their natural foci in Central Europe. This result has an important implication for public health, and patients with a history of tick bite should also be examined for the presence of Candidatus N. mikurensis since it is widespread throughout Central Europe in all regions where I. ricinus is present. The pseudo-likelihood function was used to calculate parameter estimates. The analysis was conducted with SAS (SAS Institute Inc., Cary, NC) and the GLIMMIX macro.