Babesia spp. and Anaplasma phagocytophilum in questing ticks, ticks parasitizing rodents and the parasitized rodents – Analyzing the host-pathogen-vector interface in a metropolitan area

Background The aims of this study were to evaluate the host-tick-pathogen interface of Babesia spp. and Anaplasma phagocytophilum in restored areas in both questing and host-attached Ixodes ricinus and Dermacentor reticulatus and their small mammalian hosts. Methods Questing ticks were collected from 5 sites within the city of Leipzig, Germany, in 2009. Small mammals were trapped at 3 of the 5 sites during 2010 and 2011. DNA extracts of questing and host-attached I. ricinus and D. reticulatus and of several tissue types of small mammals (the majority bank voles and yellow-necked mice), were investigated by PCR followed by sequencing for the occurrence of DNA of Babesia spp. and by real-time PCR for A. phagocytophilum. A selected number of samples positive for A. phagocytophilum were further investigated for variants of the partial 16S rRNA gene. Co-infection with Rickettsia spp. in the questing ticks was additionally investigated. Results 4.1% of questing I. ricinus ticks, but no D. reticulatus, were positive for Babesia sp. and 8.7% of I. ricinus for A. phagocytophilum. Sequencing revealed B. microti, B. capreoli and Babesia spp. EU1 in Leipzig and sequence analysis of the partial 16S RNA gene of A. phagocytophilum revealed variants either rarely reported in human cases or associated with cervid hosts. The statistical analysis revealed significantly less ticks infected with A. phagocytophilum in a city park in Leipzig as compared to the other sampling sites. A. phagocytophilum-DNA was detected in 2 bank voles, DNA of B. microti in 1 striped field-mouse and of Babesia sp. EU1 in the skin tissue of a mole. Co-infections were detected. Conclusion Our results show the involvement of small mammals in the natural endemic cycles of tick-borne pathogens. A more thorough understanding of the interactions of ticks, pathogens and hosts is the essential basis for effective preventive control measures.


Background
Tick-borne disease agents form complex life cycles involving an arthropod and vertebrate host. The complexity is enhanced by the diversity of hosts in different biotopes, which depends on factors like type of vegetation, climate and/or human influence, such as restoration of former industrial sites which leads to the development of new biotopes. Thus, on the one hand, new habitats for plants and animals including ticks, and on the other hand, new interfaces for humans and nature are created.
In contrast to Ixodes ricinus, the most common tick in Europe, Dermacentor reticulatus has a focal distribution. I. ricinus is an important vector for pathogens such as Borrelia burgdorferi sensu lato and the tick-borne encephalitis virus, but also for so-called emerging pathogens like Anaplasma phagocytophilum and Babesia spp. [1]. D. reticulatus as well serves as a vector for pathogens of human and veterinary concern, e. g. for Rickettsia raoultii (causing Tick-Borne Lymphadenopathy -TIBOLA) and Babesia canis canis (one of the causative agents of canine babesiosis) [2,3]. Canine babesiosis can be a mild to serious disease in dogs, leading even to organ failure and death [2]. In the USA, human babesiosis is primarily caused by B. microti, in Europe by B. divergens and few cases caused by B. microti and Babesia sp. EU1 have been described [4][5][6][7]. Whereas ticks as well as cattle and roe deer can serve as a reservoir for B. divergens and Babesia sp. EU1 due to transovarial transmission, B. microti is not transmitted transovarially and small mammals are the reservoir [8][9][10][11]. A. phagocytophilum causes granulocytic anaplasmosis in dogs, horses and humans [12]. Transovarial transmission of A. phagocytophilum has not been proven in I. ricinus, and several mammalian species are under discussion as reservoir hosts [12]. Previous infection rates of Babesia spp. and A. phagocytophilum detected in questing I. ricinus in Germany are listed in Tables 1 and 2. Comparison of Babesia prevalences from previous studies shows that infection rates are generally lower, by a factor of 5 to 10, in the South of Germany (Table 1). Both babesiosis and granulocytic anaplasmosis are zoonotic diseases with a natural enzootic cycle involving the agent itself, the tick and the host. For both tick species and disease agents, humans can be considered accidental hosts acquiring the respective infection during interaction with the natural enzootic cycle. In this context, recreationally used urban and suburban green areas are intriguing habitats due to the strong influence that humans exert on them and thus, these areas should be given special consideration with regard to host-pathogenvector interactions. We have previously shown high prevalences of spotted fever group rickettsiae in questing I. ricinus and D. reticulatus in recreationally used renatured areas [13]. The aims of the present study were the evaluation of the tick-host-pathogen interface in recreational areas in Leipzig, Germany. Therefore we attempted (i) to identify Babesia infections in questing and hostattached I. ricinus and D. reticulatus populations, (ii) to identify the infection rate with A. phagocytophilum in questing and host-attached I. ricinus, (iii) to identify possible small mammalian reservoir hosts in the investigated areas and (iv) to evaluate co-infections with Rickettsia spp. in the same questing ticks which had been investigated for Rickettsia spp. in a previous study [13].

Sampling sites and questing tick collection
Questing ticks were collected with the flagging method in 5 sites (E-I) in Leipzig (Saxony, Germany), all located within the city or close to suburban residential areas ( Figure 1). Sites E-G are part of the "Leipziger Neuseenland" (http://www.leipzigerneuseenland.de/), a restored area of former pit holes of a brown coal surface mining site. Brown coal was mined till 1990 and flooding of the area started in 1993. When fully flooded, the area will contain 18 lakes. Ticks were collected around Lake Cospuden (Sites E-G; fully flooded since 2000; 436 ha in size). Site E is located on the Eastern, Site F on the Northern and site G on the Western shore. The area is popular for hiking, jogging, cycling and dog walking. Further ticks were collected in the recreational city park "Lößnig-Dölitz" (site H; (http://www.leipzig.de/de/ buerger/freizeit/leipzig/parks/park_loessnig/), containing several small ponds. This park is located in an area which was used for underground brown coal mining up to 1959 and was also created by renaturation in the 1980s . To the North it borders to the city of Leipzig and to the South, where a renatured waste disposal mountain is located, it merges directly into forest areas. Site I is located in one of the largest riparian forest areas in Germany, within the city of Leipzig (www.leipziger-auwald.de).
For comparison, one site in Bavaria, located 20 km north of the city of Munich in a riparian forest on the Eastern banks of the river Isar was chosen. The area is popular for hiking, biking and dog walking.
Ticks were collected in April and September/October 2009 and were stored as described [13]. Further I. ricinus  ticks were available from a previous study [14] from an area in the Saarland of altogether 150 km 2 west of Saarbrücken (collection from March to September 2008). That previous study was concerned primarily with D. reticulatus and the I. ricinus were collected but kept aside. Usually twenty traps per site were placed along natural structures or directly in front of holes in the ground. The number of traps per night never exceeded 75 and they all were checked twice daily. Captured animals were anesthetized with CO 2 exposure and killed humanely according to the German Animal Protection Act after blood was drawn by heart puncture. After recording trapping site, species, gender, and health status, rodents were dissected under BSL-2 conditions in the laboratory and stored at −80°C. Ticks were collected from the small mammals and stored at −80°C separately for each individual animal until species identification.

DNA extraction
Species identification of ticks was carried out with standard taxonomic keys [15][16][17] and DNA was extracted from the ticks as described [13] and from tissues and body liquids with the Qiagen DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instruction for either animal tissue or blood. Tissue lysis was carried out over night at 56°C in a thermomixer (Eppendorf, Hamburg, Germany). Quality and quantity of extracted DNA were tested with a spectrophotometer (NanoDrop ND-1000, Erlangen, Germany).

PCR methods
Detection of Babesia spp.-DNA in all tick and small mammal samples was carried out with a conventional PCR targeting the 18S rRNA gene followed by gel electrophoresis as described [18,19]. Detection of A. phagocytophilum was carried out in I. ricinus ticks only and in the small mammals with a real-time PCR as described [20,21]. For a subsample of the A. phagocytophilum positive samples, a nested PCR targeting a 497 bp region of the 16S rRNA gene was performed as previously described [22,23]. Genomic DNA from A. phagocytophilum cell culture and from B. microti from continuous culture in BALB/c mice (courtesy of Prof. Dr. E. Siński) were used as positive controls and molecular grade water as negative controls and were included in each PCR run.

Sequence analysis
PCR products of the partial Babesia spp. 18S rRNA gene and of the partial 16S rRNA gene of A. phagocytophilum were purified with the QIAquick PCR Purification Kit (Qiagen, Hilden, Germany) and sequenced bidirectionally at Eurofins MWG Operon (Martinsried, Germany). The analysis of the obtained sequences was performed as described [24].

Co-infections and statistical analysis of questing tick results
Exact confidence intervals of the prevalences (CI 95%) were computed with the Clopper and Pearson method. For pooled samples the confidence interval was calculated for the minimal infection, taking into account the nymphal pools and assuming that only one of the nymphs in one pool was infected.
Differences in the infection rates with Anaplasma phagocytophilum, Babesia spp. and Rickettsia spp. (data in the same questing ticks from a previous study [13]) between sex of ticks, collection sites and region were computed with logistic regression (generalized linear models with binominal distribution and logit link) using R 2.13.1 [25]. Separate models had to be computed to investigate the effect of collection sites and region. Moreover, separate models were used for the two species. However, a variable containing the information about the developmental stage and the sex of ticks was included as covariate in all models. P-values <0.05 were regarded as statistically significant. Co-infections of the questing ticks with Rickettsia spp. were investigated as data on this pathogen in the same ticks was available from a previous study [13]. As the trapping of small mammals took place at irregular intervals, no statistical analysis was attempted for the infestations of small mammals with ticks.

Tick infestation on small mammals
Overall, 365 ticks were collected from 50 of the 80 small mammals. The infested mammals were 15 out of 31 females and 35 out of 49 males. Species identification revealed 232 I. ricinus (44 nymphs, 188 larvae) and 117 D. reticulatus (112 nymphs, 5 larvae) ( Table 5). One I. trianguliceps larva was identified on a bank vole at site I. A further 12 ticks were identified as Ixodes spp., 1 as Dermacentor spp. and for 2, identification was not possible due to damage. Ixodes spp. were obtained on 6 small mammal species (except S. araneus) at all 3 trapping sites during all trapping months (Table 6). Comparable numbers of them had tick infestation (slightly more in yellow-necked mice), and of those, comparable numbers were infested with I. ricinus larvae and nymphs with comparable median numbers of ticks (Table 6). In contrast, D. reticulatus were obtained on bank voles only (apart from 1 nymph on one yellow-necked mouse), exclusively during the month of August and only at sites E and I, even though at the latter no questing D. reticulatus had been detected ( Table 3). The median of D. reticulatus nymphs was higher than that of I. ricinus larvae and nymphs on bank voles. 95.4% (n = 21) out of the 22 tickinfested yellow-necked mice were infested with I. ricinus larvae, 36.4% with I. ricinus nymphs and 4.5% (1 individual) with a D. reticulatus nymph. 80.1% out of the 21 tick-infested bank voles were infested with I. ricinus larvae, 28.6% with I. ricinus nymphs, 4.8% with D. reticulatus larvae and 62% with D. reticulatus nymphs.
Co-infestation and thus co-feeding of I. ricinus nymphs and larvae was observed on 7 yellow-necked mice, 8 bank voles, 1 striped field mouse and 1 shrew (C. russula). One of those yellow-necked mice and 5 bank voles were additionally infested with D. reticulatus nymphs and one bank vole with D. reticulatus larvae. 3 bank voles carried I. ricinus larvae and D. reticulatus nymphs and 1 bank vole I. ricinus larvae and an I. trianguliceps larva. The rest were single infestations of I. ricinus larvae (14 yellownecked mice, 6 bank voles, 1 striped field mouse, 1 great white-toothed shrew, 1 common mole), I. ricinus nymphs (1 yellow-necked mouse, 2 European water voles) and D. reticulatus nymphs (4 bank voles). DNA was extracted individually from all host-attached nymphs.

Small mammals and host-attached ticks
The Babesia spp.-PCR was positive for the kidney and transudate of one striped field mouse from site H (B. microti), for the ear tissue of the European mole (Babesia   sp. EU1) and for the transudates of a bank vole and a shrew from site E (  [26].

Small mammals and host-attached ticks
All blood and tissue samples of the small mammals were negative for A. phagocytophilum-DNA, but it was detected in the transudates of 2 bank voles from site E ( Table 5). From one of them the kidney sample was also positive. One I. ricinus nymph collected from a bank vole from site E and from a yellow-necked mouse from site I were positive for A. phagocytophilum, all other host-attached ticks were negative (Table 6).

Co-infections
Co-infections of Babesia spp. and A. phagocytophilum were not detected in 539 questing I. ricinus in Leipzig and in 128 questing I. ricinus in Bavaria, whereas in the Saarland 1/115 (0.9%; one adult male) was infected with B. microti and A. phagocytophilum.
With regard to co-infections and differences in prevalence patterns we also statistically analysed data from a previous study in the same questing ticks which were investigated for Rickettsia spp. (R. helvetica in I. ricinus and R. raoultii in D. reticulatus) [13]. In Leipzig, 7/539 I. ricinus (1.3%; 3 adult male and one adult female from site I; one adult male and two females from site E) were coinfected with R. helvetica and A. phagocytophilum, and 1/539 (0.2%; one adult female) from site H was coinfected with B. microti and R. helvetica. In Bavaria, 1/128 I. ricinus (0.8%; one adult male) was coinfected with R. helvetica and A. phagocytophilum. In the Saarland, altogether 6 I. ricinus samples (5.2%; 4 males, 1 female, 1 nymphal pool containing two nymphs) were coinfected with R. helvetica and A. phagocytophilum, one adult female with R. helvetica and B. microti and one nymph pool (containing two nymphs) with Babesia spp. and R. helvetica. Triple infections were not detected [13]. The comparison of the single collection sites amongst each other revealed for I. ricinus a significant difference in the infection rates with R. helvetica between site E and I in Leipzig (p = 0.02). For D. reticulatus, the collection sites in the Saarland and Bavaria, as well as the site G revealed a lower prevalence compared to site E which was used as a reference site for the analysis. The comparison of the three regions showed significantly higher prevalences in I. ricinus in comparison to Leipzig for Bavaria (p = 0.001) and the Saarland (p < 0.001), whereas for D. reticulatus, the infection was significantly higher in Leipzig compared to Bavaria and the Saarland (for both p < 0.001).

Discussion
To investigate the host-pathogen-vector interface in urban recreational areas in the city of Leipzig, Germany, questing I. ricinus and D. reticulatus ticks, small mammals and the ticks infesting them were screened by molecular methods for the presence of zoonotic Babesia spp. and A. phagocytophilum. Additionally, questing ticks from Bavaria and the Saarland were collected for comparative purposes. The questing tick collection revealed a coexistence of I. ricinus and D. reticulatus in recreationally used renatured areas in Leipzig. Reports on the existence of D. reticulatus in Eastern Germany date back to the first half of the 20 th century [33,34], but recently reports about their abundance in new or renewed habitats are accumulating [13,[35][36][37][38][39]. Climatic change or changes in landscape structure or land use are being discussed for its geographical expansion and our findings in a newly created renatured area adds evidence to this [13,40].

PCR results for Babesia spp
Average detection rates in questing I. ricinus ticks of DNA of Babesia spp. were 4.1% in Leipzig, 5.5% in Bavaria and 6.1% the Saarland) and the species B. capreoli, B. divergens, B. microti, Babesia sp. EU1 and Babesia spp. were identified by sequencing. We detected DNA of B. microti also in the kidney of only one A. agrarius but not in its other tissues. Previous PCR-detections of B. microti in small mammals in Europe ranged from 1.6% to 42% in microtine voles and from 1% to 11.8% in A. flavicollis [41][42][43][44][45] and generally voles seem to be infected more often with B. microti than mice [46]. Thus, voles and shrews, but not mice are considered as reservoirs [43]. These small mammalian species all live in urban environments, but the abundance of Apodemus spp. is considered to be higher than that of M. glareolus [46,47]. As expected, at site H in the present study (an urban park) more Apodemus spp. were trapped than voles (this was the opposite for sites E and I), nonetheless in questing I. ricinus ticks a high prevalence of B. microti was found at site H. This is a rather surprising result and shows that further systematic investigations in this geographic area are needed, but also that the choice of tissue for investigation has a strong influence on the outcome of a study.
The presence of Babesia sp. EU1 was shown for the first time in questing ticks in the Middle of Germany in the present study. The mammalian reservoir for Babesia sp. EU1 is unknown, but roe deer has also been suggested [8]. This may explain the predominance of this Babesia species around Lake Cospuden where roe deer has found a new habitat. We detected Babesia sp. EU1-DNA in two nymphs collected from rodents and in the ear tissue of a common mole, which is the first detection of this Babesia species in this insectivore. Whether this is a rare coincidence or whether T. europaea, or other rodents and insectivores, play a role in the life cycle of Babesia sp. EU1 remains to be investigated. Apicomplexan parasites as well as trypanosomes have been shown in the red blood cells of the common mole before [48,49]. Isolation of the tick-borne encephalitis virus was achieved from a common mole, showing that this insectivore species may indeed play a role in the life cycle of some tick-borne pathogens [50]. It is interesting in this context that we are not aware of any description of Babesia spp. other than B. microti in small mammals thus far. We cannot tell whether this is because other investigators specifically look for B. microti or do not differentiate the Babesia spp., but the Babesia sp. EU1 detection in the European mole is a new and interesting finding in this regard. The detection of three zoonotic Babesia species in actively questing I. ricinus in recreational areas in Germany has considerable public health implications even though, in Europe, human babesiosis occurs almost exclusively in splenectomised or otherwise immunocompromised individuals [4,11].
In the present study, no evidence was found in the investigated D. reticulatus populations in Leipzig and Bavaria for infection with B. canis even though reports on autochthonous cases of canine babesiosis occur occasionally in Germany [51]. Discussion is ongoing that the import of persistently infected dogs may provide a source and could contribute to the creation of new endemic cycles in previously Babesia-free D. reticulatus populations [52].
It is noteworthy that both a Hepatozoon spp. and a Sarcocystis spp. were identified in a bank vole and a shrew due to two reasons, respectively. Firstly, this shows that in an apparently genus-specific PCR other genera of closely related protozoa may also be amplified and highlights the importance of sequencing in such screenings. In our particular case, this would have resulted in doubling the prevalence of Babesia spp. in the small mammals investigated. Secondly and even more importantly, it also shows that rodents and insectivores are hosts to a very large parasitical and bacterial fauna.

PCR results for Anaplasma phagocytophilum
The average prevalence of A. phagocytophilum in Ixodes ricinus in the investigated areas in Leipzig (8.7%) was considerably higher than most previously detected average infection rates in Germany (see Table 2) and was most similar to average prevalences recorded from city parks in Bavaria [29,53]. The presence of a different reservoir host had been hypothesised as a reason for this in areas with a city park character. However, in Leipzig, a city park (Site H) had the lowest prevalence (2.3%). The highest prevalence (14.2%) was observed close to the game park (site I). The presence of cervids might lead to a higher prevalence of A. phagocytophilum in ticks, as for example roe deer is frequently infested with ticks and is discussed as a potential reservoir host [12,54]. Similarly high prevalences of up to 17.1% have been detected in Norway in sites with a high density of wild cervids [55]. Surprisingly, and also on the contrary to the "city park" hypothesis, an infection rate of 9.4% was recognized in the alluvial forest north of Munich along the river Isar. The prevalence in an earlier study in the same region was remarkably lower (females 4.1%, males 0.75%) [29]. The difference in prevalence rates may also be based on a year-to-year fluctuation, as has been shown in other studies [56,57]. For Borrelia spp., a correlation was observed between peak abundance of rodent populations and an increase of infection in ticks in following years [58].
Significantly less nymphs than adults were infected with A. phagocytophilum in the present study. This result is well explained by the lack of transovarial transmission of A. phagocytophilum and has been shown in previous studies as well [29,53]. Apart from roe deer, small mammals have been suggested as reservoir hosts for A. phagocytophilum. In the present study, transudates of two bank voles were positive for DNA of A. phagocytophilum as well as one kidney sample. In the U.K., where A. phagocytophilum was found in field voles, bank voles and wood mice, the infection rate was generally higher in voles. Higher infection was further observed in sites with occurrence of I. trianguliceps. Rodents were not or less infected in sites where only I. ricinus occurred [59,60]. A separate endemic niche cycle involving voles and I. trianguliceps was suggested for A. phagocytophilum [59][60][61]. When individual voles were investigated several times during one study, A. phagocytophilum infection seemed short-lived whereas rodents were persistently infected with B. microti [45,59]. Only voles were infected with A. phagocytophilum in this study which coincides with this hypothesis. This could be the reason for the low prevalence with A. phagocytophilum in this study.
Larvae are free of A. phagocytophilum as no transovarial transmission occurs and adult ticks do not generally feed on rodents. Consequently, larvae need to acquire the infection from rodents and following this, nymphs have to give it back to the rodents to keep up the endemic cycle. Keeping in mind that small rodents are themselves short-lived animals and that the infection with A. phagocytophilum is transient, the question has to be raised if they would provide a suitable reservoir for keeping an endemic cycle. We observed a high percentage of co-feeding ticks on the trapped small mammals and similar to Borrelia spp. or the TBE-virus, this could be part of the endemic cycle [62,63].

Anaplasma phagocytophilum 16S rRNA gene variants
Four genetic variants of the partial 16S rRNA of A. phagocytophilum were detected in questing ticks in this study, all of which have been previously described. Interestingly, the human pathogenic prototype variant "B" [equalling GenBank U02521 in the amplified part] was not detected. About one-fourth of the A. phagocytophilum positive samples were sequenced and the variant "A" was the most prevalent variant (8 of 13). This variant is very common in ticks in Germany and has also been detected as the most common variant in hedgehogs [29,30,53]. Variant "A" is also known to cause granulocytic anaplasmosis in humans, horses and dogs, but apparently to a lesser extent [24,27,28,31]. Variant "S" has been previously detected in clinical cases of dogs and horses [24,28]. Variants "X" and "Y" seem to be typical for caprines and cervids, and have also been previously detected in ticks [23,27,32,53]. Due to the low occurrence of seemingly pathogenic variants, the high prevalences and the resulting actual risk for human beings to develop clinical disease has to be carefully evaluated. In the USA, human granulocytic anaplasmosis (HGA) is the second most important tick-borne disease after Lyme Borreliosis, whereas in Europe only about 70 cases have been reported [64]. Even though human cases have to date not been reported in Germany, seroprevalence studies have shown that the human population is exposed to the agent [65,66] and granulocytic anaplasmosis cases have been reported from dogs and horses [67][68][69]. Around Germany, HGA cases have been reported, for example from Slovenia, Austria and Poland [70][71][72].

Co-infections
The rates of co-infections detected in this study were less than 1% of investigated ticks and similar to previously detected ones [18]. The lowest co-infection rate seems to occur between Babesia spp. and Rickettsia spp., and the highest between Rickettsia spp. and A. phagocytophilum [18]. The co-infection rate of Rickettsia spp. and A. phagocytophilum was especially high in the Saarland (5.2%), and can be lead back to the fact that 17.4% I. ricinus ticks were infected with A. phagocytophilum. Co-infections between the pathogens investigated in this study have also been detected in other European countries [73][74][75][76][77][78][79][80].

Tick infestation on small mammals
Developmental stages of D. reticulatus were observed almost exclusively on bank voles. Similar results were obtained previously [62,81]. Furthermore, they were only observed during trapping in the month of August. This reflects the seasonal dynamics of this tick species, with developmental stages being active during the months of July and August [16], whereas adults have their activity peak in spring and autumn [15]. The developmental stages have an endophilic life cycle adapted to their small mammalian hosts and are therefore not obtained by the flagging method.
Co-feeding of both developmental stages of I. ricinus and additionally co-infestation with both hard tick species was frequently observed, especially in bank voles. This finding highlights the importance of rodent species in the developmental cycles of ticks and consequently for tick-borne pathogens. Larvae and nymphs of I. ricinus originate most probably from different tick generations, based on the life cycle of this tick species, whereas Dermacentor spp. can complete its entire life cycle within one year [16] which may result in different dynamics of the involved pathogens. I. ricinus is the main vector in Europe for A. phagocytophilum, R. helvetica and the Babesia species detected in this study. Even though I. trianguliceps may also act as a vector for A. phagocytophilum, its occurrence is relatively rare in Central and Eastern Europe, and therefore its importance is less apparent [59]. Furthermore, I. trianguliceps rarely feeds on humans whereas this is regularly the case for I. ricinus. In contrast, the distribution of D. reticulatus is highly focally. Its developmental stages do not attack humans, and the adult stages only rarely do. Therefore the public health risk of I. ricinus-borne pathogens has to be considered way higher by comparison, especially as co-infections with several pathogens can be detected. However, D. reticulatus is a tick frequently observed on dogs, thus posing a health threat for the canine population.

Conclusion
In conclusion, three zoonotic Babesia species and A. phagocytophilum as well as co-infections between the two and with Rickettsia spp. were detected in three areas of Germany -Leipzig, north of Munich and the Saarland. As the areas under investigation are recreational, the detection of A. phagocytophilum has both implications for human and animal health and the detection of the zoonotic Babesia spp. is of importance for humans. Although we provide new data regarding the distribution patterns of the described pathogens, we still lack a thorough understanding of the epidemiology and interplay of the ticks, their feeding sources and pathogens. In particular, knowledge about the role of co-feeding in the transmission cycles, as well as the specific hosts and potential reservoir animals in these areas, is the essential basis for the development of effective preventive control measures.