Potential drivers of vector-borne pathogens in urban environments: European hedgehogs (Erinaceus europaeus) in the spotlight

Vector-borne diseases (VBDs) are considered as (re-)emerging, but information on the transmission cycles and wildlife reservoirs is often incomplete, particularly with regard to urban areas. The present study investigated blood samples from European hedgehogs (Erinaceus europaeus) presented at wildlife rehabilitation centres in the region of Hanover. Past exposure to B. burgdorferi sensu lato (s.l.) and tick-borne encephalitis virus (TBEV) was assessed by serological detection of antibodies, while current infections with Borrelia spp., Anaplasma phagocytophilum, Rickettsia spp., Neoehrlichia mikurensis, Bartonella spp., Babesia spp. and Spiroplasma ixodetis were investigated by (q)PCR. Of 539 hedgehogs tested for anti-Borrelia antibodies, 84.8% (457/539) were seropositive, with a higher seropositivity rate in adult than subadult animals, while anti-TBEV antibodies were detected in one animal only (0.2%; 1/526). By qPCR, 31.2% (168/539) of hedgehog blood samples were positive for Borrelia spp., 49.7% (261/525) for A. phagocytophilum, 13.0% (68/525) for Bartonella spp., 8.2% for S. ixodetis (43/525), 8.0% (42/525) for Rickettsia spp. and 1.3% (7/525) for Babesia spp., while N. mikurensis was not detected. While further differentiation of Borrelia spp. infections was not successful, 63.2% of the A. phagocytophilum infections were assigned to the zoonotic ecotype I and among Rickettsia spp. infections, 50.0% to R. helvetica by ecotype- or species-specific qPCR, respectively. Sequencing revealed the presence of a Rickettsia sp. closely related to Rickettsia felis in addition to a Bartonella sp. previously described from hedgehogs, as well as Babesia microti and Babesia venatorum. These findings show that hedgehogs from rehabilitation centres are valuable sources to identify One Health pathogens in urban areas. The hedgehogs are not only exposed to pathogens from fleas and ticks in urban areas, but they also act as potent amplifiers for these vectors and their pathogens, relevant for citizens and their pets.


Introduction
Urbanization creates complex challenges for human, animal and environmental health including heat stress, increased pollution, habitat fragmentation and altered host communities for pathogens [1][2][3].Within cities, urban green spaces increase human wellbeing by improving air and water quality, regulating the climate and fostering physical and mental health [4], but they also increase urban biodiversity by providing habitats for wild vertebrates and invertebrates, which may entail the risk of zoonotic disease transmission [5,6].Vector-borne diseases (VBDs) are considered as (re-)emerging and of increasing One Health importance in Europe, affecting both human and animal health and involving invertebrate vectors as well as wildlife reservoir hosts [7][8][9][10].Lyme borreliosis (LB), caused by spirochetes of the Borrelia burgdorferi sensu lato (s.l.) complex and primarily transmitted by the widespread hard tick Ixodes ricinus, is the most frequent VBD on the continent, with an estimated 24% of the European population living in areas of high LB incidence [11].Data from different countries show that a substantial share of Borrelia infections leading to LB cases is acquired in urban or peri-urban areas [12,13].The annual economic burden resulting from LB-associated healthcare and indirect costs has been estimated at 0.14-1.36USD per capita in different European countries [14].In addition, a variety of other pathogens relevant for human and/ or veterinary health circulates between I. ricinus and vertebrate reservoir hosts, including the relapsing-fever spirochete Borrelia miyamotoi, bacteria of the order Rickettsiales (Anaplasma phagocytophilum, Rickettsia helvetica, Rickettsia monacensis, Neoehrlichia mikurensis), tick-borne encephalitis virus (TBEV) and various Babesia species [15].Recently, the tick endosymbiont Spiroplasma ixodetis has also been implicated in human disease cases [16].
Often, the natural transmission cycles and involved reservoir hosts of these pathogens are not fully known [17], particularly in urban and periurban areas which are characterized by a high human and pet density, but often a different and less diverse wildlife community composition compared to rural areas [18].The European hedgehog (Erinaceus europaeus) is omnipresent in urban and peri-urban habitats, where its population density is often even higher than in rural landscapes [19,20].Their preference for residential gardens and city parks results in a strong overlap with the environment of humans and their pets [21] and they are often taken into human care if found weak or injured [22][23][24].Hedgehogs show a high ectoparasite prevalence and often a high infestation intensity [24,25].In contrast to smaller mammals like rodents, which usually act as hosts for immature I. ricinus only [26], hedgehogs sustain all life stages of I. ricinus and of the hedgehog tick Ixodes hexagonus [24,25], thereby acting as a potential source of ticks in their local environment.Like I. ricinus, I. hexagonus is a vector of B. burgdorferi s.l.[27,28].Due to its specialized, nest-adapted lifestyle, humans are only occasionally bitten by I. hexagonus [29,30], however, this species represented 5.5% and 1.6% of ticks collected from domestic cats and dogs in Germany, respectively [31].
The reservoir function of hedgehogs for B. burgdorferi s.l. has been proven by transmission to xenodiagnostic I. hexagonus as well as I. ricinus larvae [32].Moreover, previous studies investigating European as well as Northern white-breasted hedgehogs (Erinaceus roumanicus) revealed high infection rates with A. phagocytophilum [33][34][35], indicating that hedgehogs potentially play an important epidemiological role for this pathogen.This is further corroborated by a higher A. phagocytophilum prevalence in engorged I. ricinus collected from European hedgehogs compared to questing ticks [25].Similarly, DNA of R. helvetica, and certain members of the B. burgdorferi s.l.complex (B.afzelii, B. bavariensis and B. spielmanii) was also detected at higher prevalence in engorged ticks from hedgehogs than in questing ticks [25], although presence of DNA in engorged ticks does not necessarily indicate reservoir competence.
In addition, several Bartonella spp., which are primarily transmitted by fleas, have been detected in E. europaeus and E. roumanicus, including species with validated zoonotic potential [36].Further Bartonella isolates from hedgehogs may also be of zoonotic relevance, as the number of Bartonella spp.associated with human disease continues to grow [37].When hedgehogs are taken into human care [22][23][24], they may represent a source of Bartonella infection for humans via transfer of fleas or via bites and scratches.
The objective of the present study was to shed further light on the reservoir role of European hedgehogs by determining past and present infections with vector-borne pathogens as well as seasonal and agerelated patterns in infection rates.Therefore, the present study investigated blood samples from >500 European hedgehogs presented at wildlife rehabilitation centres in or near the city of Hanover, Northern Germany, during 2018-2021.Past exposure to B. burgdorferi s.l. and TBEV was assessed by serological detection of antibodies, while molecular methods were employed to assess current infections with Borrelia spp., A. phagocytophilum, Rickettsia spp., N. mikurensis, Bartonella spp., Babesia spp.and S. ixodetis.The findings help to assess whether hedgehogs are a valuable tool to monitor the presence of One Health pathogens in urbanized areas.If so, then the potential contribution of hedgehogs to the One Health risk of these pathogens in urban areas can be assessed and taken into account by municipalities, citizens, veterinarians, and medical doctors.

Hedgehog sampling
Hedgehogs were examined at three wildlife rehabilitation centres located in or near the city of Hanover, capital of the federal state of Lower Saxony, Germany, between July 2018 and May 2021 [24].Hedgehogs were brought into these rehabilitation centres due to weakness, illness or injury and subjected to a clinical examination and appropriate treatment.Based on a combination of body weight, body condition, teeth condition and season, hedgehogs were classified as juvenile, subadult or adult as described by Schütte et al. [24].
During the clinical examination, up to 2 ml of blood were taken from the saphenous vein or, in case the animal had to be euthanized or died prior to the examination, by puncture of the heart.Blood sampling was conducted in accordance with the German Animal Welfare act as well as the national and international guidelines for animal welfare and was approved by the ethics commission of the Animal Care and Use Committee of the Lower Saxony State Office for Consumer Protection and Food Safety (Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit) under reference number 33.8-42502-05-18A320.The blood samples were centrifuged at 2000 xg for 5 min.Serum and blood clots were stored separately at − 20 • C for antibody ELISA and PCR investigations, respectively (Fig. 1).

Serological assays
The VetLine Borrelia ELISA kit (NovaTec Immundiagnostica GmbH, Dietzenbach, Germany) was used to detect IgG antibodies against B. burgdorferi s.l.according to the manufacturer's instructions, with test results interpreted as follows: > 11 NovaTec units (NTU) were considered positive, 9-11 NTU as borderline and < 9 NTU as negative.
For detection of anti-TBEV antibodies, the Immunozym FSME IgG All Species ELISA kit (Progen, Heidelberg, Germany) was used.Test interpretation followed the manufacturer's recommendations, with <63 Vienna units (VIEU)/ml considered negative, 63-126 VIEU/ml as borderline and > 126 VIEU/ml as positive.Samples with ≥60 VIEU/ml were subjected to serum neutralization test (SNT) at the National Reference Laboratory for TBEV to exclude false-positive results due to cross-reactions with antibodies against other flaviviruses.The SNT was conducted as a micro-neutralization test in 96-well plates as previously described [38].Briefly, sera were inactivated at 56 • C for 30 min, followed by 1:10 and 1:20 dilution using cell culture medium (MEM supplemented with antibiotics, antimycotics and non-essential amino acids; Invitrogen, ThermoFisher Scientific, Darmstadt, Germany).Serum dilutions were incubated in duplicate for 1 h at 37 • C together with 50-100 tissue culture infectious doses of TBEV strain Neudörfl.After addition of 10 4 A549 cells/well, the serum-virus solution was incubated for 5 days at 37 • C. Cells were fixed and stained using a 13% formalin-0.1% crystal violet solution, resulting in a prominent blue well for antibody-positive sera, while antibody-negative sera remained transparent.A positive result at 1:20 dilution was considered anti-TBEV antibody-positive.Sera with a titre of 1:10 were repeated in triplicate to confirm this titre and were then also classified anti-TBEV antibodypositive.Sera with titres <1:10 were classified as anti-TBEV antibodynegative.

DNA extraction and qPCR for vector-borne pathogens
DNA was extracted from blood clots, whereby the weight of each blood clot used for DNA extraction was determined and documented.The NucleoSpin Blood kit (Macherey-Nagel, Düren, Germany) was used according to the manufacturer's instructions.
Conducted quantitative real-time PCR (qPCR) analyses for screening of the samples are summarized in Table 1.For detection of Borrelia spp., the 23S rDNA and intergenic spacer (IGS) region was targeted by probebased qPCR using 10 μl template as described previously [39][40][41].
Plasmid standards with 10 0 to 10 6 copies were included for quantification.Additionally, a B. miyamotoi-specific qPCR [42] was conducted for a subset of 525 samples.These 525 samples were additionally tested for A. phagocytophilum, Rickettsia spp., N. mikurensis, S. ixodetis and Babesia spp.by probe-based qPCR.For A. phagocytophilum, the msp2 gene was targeted [43].Anaplasma-positive samples were subsequently subjected to a qPCR protocol distinguishing between ecotype I and II by use of two different probes [44].Regarding Rickettsia spp., two different qPCRs based on the gltA gene were employed [45,46], while the groEL gene was targeted for N. mikurensis [47], the ssrA gene for Bartonella spp.[48] and the rpoB gene for Spiroplasma spp.[49].For detection of Babesia spp., the 18S gene was amplified [50] and the ITS for B. microti [51].

Species differentiation of Borrelia spp., Rickettsia spp., Bartonella spp. and Babesia spp
Aliquots of Borrelia qPCR-positive samples were sent to the National Reference Center for Borrelia for further differentiation by multi-locus sequence typing (MLST).After DNA extraction using the Maxwell® 16 LEV Blood DNA Kit (Promega, Walldorf, Germany) according to manufacturer's recommendations, DNA concentrations were determined and samples diluted 1:2 or 1:5.All samples were re-tested by qPCR as described above [39].Nested PCR to amplify fragments of the pepX and recG gene were carried out as previously described [52] with following modifications.We used a BioRad C1000 thermocycler and a PhireTaq mastermix (Invitrogen, ThermoFisher Scientific, Darmstadt, Germany).Primers and annealing temperatures were as described.Time for initial denaturation was 30 s and during each cycle 5 s; annealing time 5 s, elongation time 15 s.These conditions were used in first and second amplification rounds.Samples that showed bands of the expected size were sent for Sanger sequencing (GATC, Eurofins Genomics, Ebersberg, Germany).

Statistical analyses
To investigate the influence of temporal and animal-related factors on Borrelia seroprevalence as well as Borrelia spp.and A. phagocytophilum qPCR results, generalized linear models (GLMs) with binomial error structure and logit link were constructed in R v. 4.2.1 [60].Sampling season (spring/summer/autumn/winter) and sampling year as well as animal sex and age category were included as covariates.For the model of Borrelia antibody seroprevalence, the corresponding  Borrelia qPCR result was included as an additional covariate, while samples with a borderline ELISA result were excluded.In the models investigating qPCR results, a possible association between Anaplasma and Borrelia infection status was considered by including the respective other pathogen as a covariate.Furthermore, the weight of the cell pellet used for DNA extraction was included to control for a possible bias due to different sample amounts.For the remaining pathogens, no individual GLMs were calculated due to low prevalence.Full models were compared to null models including only the intercept in a likelihood ratio test.

Serology results
Out of 541 hedgehogs, serological testing was performed on 539 for anti-Borrelia antibodies and 526 for anti-TBEV antibodies due to limited blood volume (Fig. 1).A significant majority, 84.8% (457/539), tested positive for B. burgdorferi s.l.antibodies, whereas 1.5% (8/539) yielded borderline results.Anti-TBEV antibodies were detected in one animal (0.2%) by ELISA and confirmed by SNT with a titre ≥1:20, while two further samples were considered borderline by ELISA but negative by SNT.

PCR results
A Borrelia qPCR analysis was performed on 539 hedgehogs, revealing a prevalence rate of 31.2% (168/539), as detailed in Table 2.The Borrelia 5S-23S IGS copy numbers were low, with a mean of 7.2 copies per reaction (median: 1.2).The 168 positive blood samples were sent to the National Reference Center for Borrelia, where additional DNA isolation and 5S-23S IGS qPCR were performed.A positive qPCR result was observed in 31 samples, with Ct values ranging between 35 and 40.Borrelia spp.differentiation by MLST was attempted, but no Borrelia recG nor pepX sequences could be generated.However, among the 525 samples tested additionally by B. miyamotoi-specific qPCR, one sample (0.2%) was positive for this species.
DNA of A. phagocytophilum was identified in 49.7% (261/525) of the hedgehog samples tested.Among these, 63.2% (165/261) were classified as ecotype I infections.The remaining positive samples did not match either ecotype I or II.
Prevalence of Rickettsia spp.amounted to 8.0% (42/525), with 50.0%(21/42) of infections assigned to R. helvetica by qPCR.Furthermore, a Rickettsia gltA sequence was obtained from four samples, while no gltA amplification was achieved for the remaining samples.All four sequences (GenBank acc.nos.PP104507-PP104510) showed 100% nucleotide identity to a Rickettsia sp.closely related to Rickettsia felis (99% query cover [QC], GenBank acc.nos.MG253006 and MG253007).Three of these samples were also positive for R. helvetica by qPCR.

Coinfections
Among the 525 hedgehogs tested for all pathogens, 70.1% (368/525) were positive for at least one pathogen, including 31.6% (166/525) that were coinfected with more than one pathogen.Single infections with A. phagocytophilum were detected in 22.1% (116/525), single infections with Borrelia spp. in 10.5% (55/525) and double infections with these two pathogens in 9.9% (52/525) of tested animals.Single infections with Rickettsia spp., Bartonella spp., S. ixodetis and Babesia spp. as well as various other coinfections with up to four different pathogens were each noted in <5% of hedgehogs (Table 3).

Temporal and host-related patterns
The seroprevalence of Borrelia antibodies showed a seasonal pattern with a higher proportion of seropositive animals in the summer months.However, the GLM indicated that this was driven by the age distribution of the animals, with a significantly higher probability of seropositivity in adults, which were predominantly sampled during summer (Fig. 2, Table 4).In contrast, animal sex, season, sampling year and Borrelia qPCR result were not significantly associated with Borrelia antibody status.In fact, among Borrelia seropositive animals, 32.6% (149/457) were positive by qPCR, while 24.3% (17/74) of seronegative animals were also qPCR-positive (Table 5).
Regarding Borrelia spp.infection status as determined by qPCR, no significant age, sex or seasonal difference was determined, whereas a decline in prevalence was noted over the course of the study (Fig. 2, Table 6).However, a lower amount of blood was used on average for DNA extraction in 2020 and 2021 than in 2018 and 2019 (Supplementary Fig. S1).Although this covariate did not show a significant influence in the GLM for Borrelia prevalence (Table 6), a higher amount of blood was significantly associated with lower Borrelia copy numbers among positive samples (Supplementary Fig. S1).
In contrast, no decline in prevalence was apparent for A. phagocytophilum.For this pathogen, a seasonal pattern was observed with a significantly higher prevalence in spring, summer and autumn as compared to winter, whereas the model indicated no significant age or sex effect (Fig. 2, Table 6).No significant association between Borrelia and Anaplasma infection status was observed.
For the remaining pathogens, detection rates in the two age groups as well as over the course of the study are visualized in Fig. 3, but no statistical models were calculated due to the low prevalence values.

Discussion
Hedgehogs often show high levels of ectoparasite infestation [24,25].Therefore, the high prevalence and species diversity of vectorborne pathogens detected in the present study was not unexpected.The fact that almost all adult animals, i.e. animals >1 year, had antibodies against B. burgdorferi s.l.confirms a high level of exposure.In contrast, neutralizing antibodies against TBEV were detected in one animal only.Most parts of Northern Germany, including the region of Hanover, are not regarded as TBEV risk areas [61] and the distribution of TBEV is restricted to small foci even within endemic regions [62].This and the rather small home range size of hedgehogs compared to larger wildlife species may explain the low TBEV seroprevalence.
Moreover, hedgehog blood samples were screened for several vectorborne pathogens by qPCR.No tissue samples were taken for ethical reasons, as it was the goal to rehabilitate the hedgehogs which were often weak and/or injured.Blood is not regarded as an ideal sample type for B. burgdorferi s.l.detection because spirochetemia is usually transient and low-level [63,64].For example, B. burgdorferi was cultured approximately ten times more often from the spleen than from blood of white-footed mice (Peromyscus leucopus) [65].Therefore, the rather high rate of 31.2%Borrelia spp.-positive blood samples in the present study was surprising, but may be due to a high sensitivity of the employed probe-based qPCR.In a previous study based on tissue samples of European hedgehogs from the Czech Republic, a 90.0%detection rate of B. burgdorferi s.l.DNA was reported [66], in line with the high seroprevalence rate determined in the present study.In contrast, another study, also based on hedgehog tissue samples, reported only a 13.5% detection rate [67].This discrepancy may be due to differences in the sensitivity of employed PCR techniques, the tissue sample types, regional differences or temporal fluctuations.
Interestingly, the Borrelia spp.detection rate declined significantly over the course of the present study.However, this result should be treated with caution as the amount of blood used for DNA isolation was on average lower in 2020 and 2021 (0.04 g) than in 2018 and 2019 (0.07 g) to avoid clogging of DNA isolation columns.Although no statistically significant effect of the amount of blood on the Borrelia detection rate could be discerned, the Borrelia 5S-23S IGS copy numbers were negatively correlated with the weight of the blood clot, which could mean that a higher amount of blood allowed detection of lower copy numbers.Thus, infections with a low level of spirochetemia may have been missed in 2020 and 2021 due to the lower amount of blood used.
A significant seasonal pattern was neither detected for Borrelia ELISA nor for qPCR results, as the apparently higher antibody detection rate in summer was driven by the high proportion of adult animals sampled during these months.The age difference in the seropositivity rate can be attributed to cumulative exposure combined with a long-lived antibody response.The fact that animals were qPCR positive despite being antibody-positive, without any seasonal pattern, implies that they may be more persistent carriers than e.g.rodents, which show low infection rates during winter [68,69].As hedgehogs also move larger distances than rodents, this indicates a potentially important role for pathogen maintenance and spread, especially in urban areas, where tick density can be low.
To further elucidate the zoonotic potential of the Borrelia spp.harboured by hedgehogs, species differentiation by MLST was attempted but was not successful, probably due to the very low amount of borrelial DNA in the samples, whereas only one sample was B. miyamotoi-positive by qPCR.In previous studies, B. afzelii, B. garinii, B. bavariensis, B. spielmanii, as well as B. miyamotoi, among others, were all detected in European hedgehogs, with most infections attributed to B. afzelii [35,66,67].Moreover, B. afzelii, B. bavariensis and B. spielmanii were detected significantly more often in engorged ticks from European hedgehogs than in questing ticks in a Dutch study, indicating that hedgehogs may serve as a reservoir for these species [25].
Furthermore, hedgehogs are regarded as potential reservoirs for A. phagocytophilum [34,70].In the present study, A. phagocytophilum was the most frequently detected pathogen, with a prevalence of almost 50%.A higher detection rate of >90% was reported in skin biopsies of European hedgehogs from the Czech Republic, while the prevalence in blood samples amounted to 75.0% [70].Furthermore, a study based on repeated blood sampling of a captive, but tick-exposed hedgehog population in Germany determined monthly prevalence values between 41.7 and 84.6% [34].Although it should be kept in mind that the hedgehogs in the mentioned study did not live under completely natural conditions, one interesting fact was that individual animals showed repeated bouts of bacteraemia, with highest detection rates in spring and autumn.A seasonal pattern in A. phagocytophilum prevalence was also apparent in the present study, with higher detection rates from spring through autumn compared to winter, probably related to the main activity period of I. ricinus in central Europe [71].
European hedgehogs may carry several ecotypes of A. phagocyophilum, including zoonotic variants [70,72,73].Approximately two-thirds of the A. phagocytophilum infections in the present study were attributed to ecotype 1.This ecotype, also referred to as cluster 1 based on MLST and ankA phylogenies [72], is associated with I. ricinus ticks and constitutes the most frequent ecotype detected in European as well as Northern white-breasted hedgehogs [35,70,74].Moreover, almost all isolates from hedgehogs form a monophyletic group with zoonotic strains within ecotype 1/cluster 1 [reviewed by 72], emphasizing the reservoir role of this wildlife species for humanpathogenic A. phagocytophilum.Like ecotype 1, ecotype 2 is associated with I. ricinus ticks, but mainly with ruminants as reservoirs [74].This ecotype was not detected in the present study.The remaining undifferentiated A. phagocytophilum infections may have included ecotype 4, which is mainly associated with birds [74], but has recently also been detected in European hedgehogs [70].With a prevalence of 13%, Bartonella spp.were the third most frequently detected pathogens in the present study.An even higher prevalence of 24% was noted in European hedgehogs in the Czech Republic [36].This difference may again be due to the use of blood instead of tissue samples in the present study, as Majerová et al. [36] determined higher infection rates in spleen and ear samples than in blood.Bartonella spp.differentiation was successful in >70% of cases, yielding a gltA sequence previously amplified from European and Northern whitebreasted hedgehogs [36].No other Bartonella spp.were detected, in contrast to previous findings of Bartonella washoensis, Bartonella melophagi and another undescribed species, Bartonella sp.SCIER [36].It remains unclear whether the identified Bartonella sp., which seems closely related to the cat-associated zoonotic pathogen Bartonella clarridgeiae [36], is tick-or flea-transmitted and whether it has any clinical significance for humans or domestic animals.
In contrast to the rather high prevalence values of Borrelia spp., A. phagocytophilum and Bartonella spp., the prevalence of Rickettsia spp.amounted to only 8.0%, which seems especially low in light of high detection rates in questing ticks in Northern Germany, e.g.50.8% in the city of Hanover in 2015 [75] and 36.0% in 2020 [76].In other northwestern German areas, similar values between 22.4 and 36.5% Rickettsia-positive ticks were recorded, with the vast majority of infections attributed to R. helvetica [77].In the present study, R. helvetica was identified in 50.0% of Rickettsia-positive hedgehogs by qPCR.While no comparable studies on Rickettsia infections of European hedgehogs have been published to the authors' knowledge, reports exist of R. helvetica in different rodent species (e.g.[78,79]) and Northern white-breasted hedgehogs [35].However, transmission of this rickettsial species to vertebrates does not seem to be very efficient, or results only in a very short-lived rickettsemia.For example, R. helvetica was detected in approximately 10% of immature ticks collected from 158 rodents in Poland, but not in the rodent blood samples [80].Similarly, R. helvetica prevalence in blood samples of birds amounted to only 4.7%, but was tenfold higher in ticks collected from these birds [81].Like other spotted-fever group rickettsiae, R. helvetica can be regarded as a tick symbiont with efficient vertical transmission in the tick population [82], which may explain its low propensity to cause systemic infection in vertebrates.
Because hedgehogs are frequently infested with fleas, infection with flea-associated Rickettsia spp.seems likely.Rickettsia felis and a closely related Rickettsia sp. have been detected in the hedgehog flea Archaeopsylla erinacei at high prevalence [83,84].Unfortunately, further differentiation of Rickettsia positive samples from the current study by sequencing of the gltA gene was only successful in four cases, probably due to the low amount of Rickettsia DNA indicated by high Ct values.The amplified gltA sequences were 100% identical to a Rickettsia sp.closely related to R. felis, which was previously detected in yellow-necked mice (Apodemus flavicollis) in Slovakia [85].
Neoehrlichia mikurensis is another member of the order Rickettsiales which has previously been detected in Northern white-breasted hedgehogs at a low prevalence of 2.3% [33].It was not detected in the present study, in contrast to high detection rates in blood samples of various rodents in different European countries [86,87].Therefore, European hedgehogs may not be suitable reservoir hosts for this pathogen.This is also supported by the fact that N. mikurensis was not detected at higher prevalence in engorged ticks from hedgehogs than in questing ticks, contrary to different B. burgdorferi s.l.species and A. phagocytophilum [25].
Moreover, DNA of S. ixodetis, an endosymbiont of ticks and other arthropods [88], was detected in 8.2% of hedgehogs and thus at a similar frequency as Rickettsia species.Like Rickettsia spp., S. ixodetis is transmitted vertically in Ixodes populations [89], and transmission to hedgehogs may be incidental and without epidemiological relevance.Nevertheless, the detection in hedgehogs indicates potential exposure of humans in the study region, which is of interest as S. ixodetis is suspected to be human pathogenic under certain conditions [16].
Concerning tick-transmitted protozoa, a 1.3% prevalence of Babesia spp. was noted.Five of seven infections were attributed to rodentassociated B. microti and one to B. venatorum, while the Babesia sp. in the remaining qPCR-positive sample could not be identified.The generated 18S sequences corresponded to B. microti s.s.(clade 1), which  is responsible for most human babesiosis cases [90] and associated with different rodents and shrews as reservoir hosts [91][92][93].To the authors' knowledge, this is the first report of B. microti in European hedgehogs.The detection of B. venatorum in a European hedgehog was even more surprising, as this species is associated with roe deer, although it has also been found in sheep [94].As the molecular assay does not allow to conclude whether the amplified DNA in the present study belonged to viable parasites, it remains questionable whether the hedgehog was truly infected or only carried DNA of B. venatorum in its bloodstream due to the bite of an infected tick.

Conclusions
The present study showed that hedgehogs from wildlife rescue centres provide a valuable source of information on the occurrence of One Health pathogens in (peri-)urban areas.These hedgehogs can be sampled during initial clinical examination, without the necessity of trapping and even killing them.A high species diversity of vector-borne pathogens and a particularly high prevalence of Borrelia spp.and A. phagocytophilum was detected among the European hedgehogs in the present study, which puts this species in the spotlight of potential reservoir hosts for the most important zoonotic VBDs in urban areas.Moreover, the fact that more than a third of hedgehogs carried coinfections indicates that hedgehogs may be an important source of coinfections in ticks, which poses an increased health risk to humans due to increased disease severity and diagnostic challenges [95].As hedgehogs are expected to thrive in urban environments as a consequence of increased urban greening, the risk for VBD transmission may increase.Municipalities should be aware of this risk and act to raise awareness among citizens and veterinary and medical doctors to improve preventive measures and enhance VBD diagnosis.For example, as domestic dogs and particularly free-ranging cats are prone to roam in hedgehog habitats, protecting them against ectoparasites is equally important for the sake of their health and to avoid translocation of ticks and fleas into human households.When hedgehogs are taken into human care, they should be immediately treated against ectoparasites to avoid transfer of potentially infected vectors to human caretakers.Further studies are necessary to quantify hedgehog density in different environments and to compare their VBD infection rates between urban and rural areas, calling for close collaboration between veterinarians, ecologists, public health professionals, and conservationists.Supplementary data to this article can be found online at https://doi.org/10.1016/j.onehlt.2024.100764.
• C, followed by 40/25 cycles of 95 • C for 30 s, 53/56 • C for 30 s, 72 • C for 60 s in the first and the second round, respectively, and final elongation at 72 • C for 10 min.

Fig. 1 .
Fig. 1.Overview of the study design investigating vector-borne pathogens in European hedgehogs from Northern Germany.

Fig. 2 .
Fig. 2. Patterns of (a) B. burgdorferi s.l.ELISA results, (b) Borrelia spp.qPCR results and (c) A. phagocytophilum qPCR results in blood samples of European hedgehogs according to animal age class and over the course of the study.Sample sizes are indicated above the bars.Note that animals with undetermined age class were excluded from the panels on the left.No sampling was done in December 2020 and January 2021.

Fig. 3 .
Fig. 3. Prevalence of (a) Rickettsia spp., (b) Bartonella spp., (c) Spiroplasma ixodetis and (d) Babesia spp. in blood samples of European hedgehogs according to animal age class and over the course of the study.Sample sizes are indicated above the bars of panel A and refer to all panels.Note that animals with undetermined age class were excluded from the panels on the left.No sampling was done in December 2020 and January 2021.

Table 1
Detection and differentiation methods used for different vector-borne pathogens in blood samples of European hedgehogs.

Table 2
Prevalence of vector-borne pathogens in blood samples from European hedgehogs in Northern Germany as determined by PCR.

Table 3
Mono-and coinfections with vector-borne pathogens among 525 European hedgehogs.

Table 4
Results of the generalized linear model investigating the influence of temporal and animal-related factors on Borrelia antibody status in 496* European hedgehogs.The full model was significantly different from a null model containing only the intercept (χ 2 = 109.44,Df = 9, P < 0.001).
* Samples with a borderline ELISA result were excluded.

Table 5
Prevalence of Borrelia spp.DNA in blood samples of European hedgehogs from northern Germany according to their anti-Borrelia antibody status.

Table 6
Results of the generalized linear models investigating the influence of temporal and animal-related factors on Borrelia and Anaplasma infection status as determined by qPCR in 492 European hedgehogs.Significant P-values are printed in bold.The full models were significantly different from null models containing only the intercept (model A: χ 2 = 37.8, Df = 10, P < 0.001; model B: χ 2 = 32.9,Df = 10, P < 0.001).