Red Foxes (Vulpes vulpes) Are Exposed to High Diversity of Borrelia burgdorferi Sensu Lato Species Infecting Fox-Derived Ixodes Ticks in West-Central Poland

The role of red fox, Vulpes vulpes, and its associated ticks in maintaining Borrelia burgdorferi sensu lato (s.l.) was studied. A total of 1583 ticks were removed from ears of 120 infested animals and were identified as species using a nested PCR targeting the ITS2 and coxI fragments of Ixodes DNA. Ixodes kaiseri prevailed (76%), followed by I. canisuga, I. ricinus, and I. hexagonus. In total, 32.4% of 943 ticks revealed Borrelia DNA and 10 species of B. burgdorferi s.l. complex were identified. Borrelia garinii and B. afzelii comprised 70% of all infections. The other eight species included B. americana, B. bissettiae, B. burgdorferi sensu stricto (s.s.), B. californiensis, B. carolinensis, B. lanei, B. spielmanii, and B. valaisiana. Analysis of tissues from 243 foxes showed that 23.5% were infected with B. burgdorferi s.l. Borrelia garinii was detected in 91% of the infected animals, including 31% of mixed infections with B. afzelii, the second most prevalent species, followed by B. spielmanii. The predominance of B. garinii in PCR-positive animals and infected larval ticks (38.1%), suggests that this spirochete and B. afzelii are preferentially associated with foxes. Although red foxes are exposed to a high diversity of B. burgdorferi s.l. species found in engorged Ixodes ticks, their reservoir competence for most of them appears to be low.


Introduction
The red fox Vulpes vulpes is the most widely distributed terrestrial carnivore species, with its native range including temperate and subarctic regions of the Northern Hemisphere [1]. The implementation of anti-rabies vaccination during the past three decades increased population density of foxes in Central Europe, including Poland [2]. In contrast to cervids or wild boars, the red fox is a game animal hunted predominantly as a pest, mostly in farmland areas without the emphasis on trophy hunting [1]. In Poland, according to the Polish Hunters Association, the red fox is the most abundant carnivore, with an annual harvest estimated at approximately 150,000 animals.
Foxes are hosts for hematophagous ectoparasites, primarily fleas and ixodid ticks, transmitting a wide variety of blood-borne bacterial agents [3][4][5][6]. European eco-epidemiological studies on the identification of potentially zoonotic vector-borne bacteria in foxes, documented the presence of Anaplasma phagocytophilum [7,8], A. platys, Ehrlichia canis [9,10], and Candidatus Neoehrlichia sp. (FU98) [11]. Moreover, Bartonella rochalimae was confirmed in red foxes sampled in Spain [12,13]. However, the role of the red fox in the epidemiology of these vector-borne agents is still poorly understood. This concerns the spirochete species from the complex of Borrelia burgdorferi sensu lato (s.l.), some of which are the causative agents of Lyme disease (LD). To date, there are only a few European reports In Poland most of the hunted foxes are left in the hunting ground and buried so acquired material was limited only to blood and biopsies of skin and liver collected by the hunters. In total, 558 tissue samples were obtained from 243 foxes. EDTA-whole blood was collected from the body cavity out of 216 (88.9%) of the animals tested. Furthermore, biopsies of skin (from ear) and liver were obtained from 243 and 99 of the foxes, respectively. These three types of tissue samples were collected within one hour after each animal was culled and stored at −20 °C until extraction.
Both ears of 243 red foxes were cut off basally by hunters and placed into a separate plastic bag, and stored at −20 °C. In the laboratory, after thawing, ears were inspected for ticks under a stereoscopic microscope. Ticks were counted and stored in plastic vials containing 75% ethanol. A total of 1583 Ixodes ticks were removed from 120 infested animals. Since morphological identification of adult female ticks to the species level is easier than immature ticks, females were determined by using taxonomic keys by Siuda In brackets: number of PCR positive hosts vs. tested. Black circles denote collection sites with infected foxes. Animals were harvested during fox-hunting seasons (1 June to 31 March) from 2009 to 2011. Borrelia garinii was found in fox-derived ticks in sites no. 1-7 and 9, B. afzelii in sites 1-4 and 6-9, B. burgdorferi s.s. in sites no. 1, 3, 4, 6, and 7, B. valaisiana in sites no. 6, 7, and 9, B. spielmanii in sites no. 1, 6, 7, and 9, B. bissettiae in sites no. 1, 6, and 7; B. carolinensis and B. californiensis in sites no. 6, 7, and 9; B. lanei in sites no. 6, and 9; B. americana in site no. 7, and B. turcica in site no. 6.

DNA Extraction
DNA extraction from animal tissues and engorged ticks was performed with a phenolchloroform protocol [23]. To avoid any contamination only undamaged ticks rinsed with 75% ethanol before DNA extraction were selected for bacterial DNA detection. DNA samples were stored at −70 • C before PCR analyses.

Molecular Identification of Ixodes Tick Species
To confirm the accuracy of morphological tick identification according to taxonomic keys, nested PCR assays based on two molecular markers and primer sets targeting (i) the ITS2 (internal transcribed spacer 2) fragment from the nuclear genome of Ixodidae [21], and (ii) the mitochondrial cytochrome c oxidase subunit I (coxI) were used (Table 1). In each PCR run, DNA isolates of four reference Ixodes spp. female ticks-I. hexagonus from the European hedgehog Erinaceus europaeus, I. canisuga and I. kaiseri from the racoon dog, and I. ricinus from the vegetation-were used. The identities of these ticks had been previously validated with morphological taxonomic keys and molecular studies [18,21,22]. The PCR products were separated on 1.5% agarose gel (Bioshop, Burlington, ON, Canada) and archived, as described elsewhere [21].
In the first stage, including all tick specimens, molecular identification was performed, based on PCR-restriction fragment length polymorphism analysis (PCR-RFLP). PCR-amplified sequences of the ITS2 gene generated with primers SP2-100f and SP2-1274r were digested with enzyme AluI (Thermo Fisher Scientific, Waltham, MA, USA) to obtain RFLP patterns for different Ixodes species according to the protocol previously described [21]. In the second stage, to validate identification done by PCR-RFLP analysis of the ITS2 gene, partial sequencing of Ixodes DNA of ITS2 fragments amplified with inner primer sets SP-100f/SP-1274r was performed for subset of amplicons representing different restriction patterns. Furthermore, partial sequencing of the coxI gene fragments of Ixodes obtained with inner primer sets CO1-375f/CO1-1086r was performed (Table 1). Sequencing was conducted in Macrogen Europe (The Netherlands). Representative partial sequences (n = 78) were deposited in GenBank. The

Detection of Borrelia DNA by Nested PCR
Altogether, 943 undamaged, fully, or partially, engorged ticks (76 females, 166 nymphs, and 701 larvae) and the 558 tissues from foxes were selected. Both groups of samples were tested for the presence of Borrelia DNA using a nested PCR with two primer sets amplifying a fragment of the flaB gene; the protocol has been previously described by Wodecka et al. [24]. All positive samples were rerun using nested PCR assays with two independent sets of primers targeting parts of the p66 gene and intergenic spacer between 3-methyladenine glycosylase (mag) gene and trnI gene encoding tRNA for isoleucine (Table 1). To confirm the specificity and sensitivity of implemented protocols in each PCR run, randomly selected DNA isolated from one of the 11 reference strains of Borrelia spp. was used as a positive control and TE buffer was used as a negative control. In total, 11 reference strains were included: B. burgdorferi sensu stricto (s. The PCR products were separated on 1.5% agarose gel and the results were written, as described elsewhere [21].

Identification of Borrelia Species by PCR-RFLP and Sequencing
The flaB gene fragments amplified with primers 220f and 823 r were digested with enzymes HpyF3I and Ecl136II (Thermo Fisher Scientific, Waltham, MA, USA) to obtain RFLP patterns of different Borrelia species, according to the protocol previously described [25]. To confirm Borrelia species identification based on PCR-RFLP analysis, partial sequencing of the flaB gene products amplified with primers 220f and 823r or FL120F and FL908R, p66 gene fragments obtained with primers P66-487F and P66-1087R and fragments of intergenic spacer between mag gene and trnI gene generated with primers glz435F and ile65R (Table 1)

Contamination Control in DNA Analysis
To minimize contamination, the processes of DNA isolation, the reaction mixture preparation and electrophoresis were carried out in separate rooms.

DNA Sequence Analysis
Aligned sequences representing 43 nuclear ITS2 gene fragments and 35 mitochondrial coxI gene fragments of Ixodes strains as well as 112 flaB gene, 33 p66 gene, and 44 mag-trnI intergenic spacer fragments of Borrelia strains were examined with MEGAX software (Molecular Evolutionary Genetics Analysis, version X) [26]. Relationships between individuals were assessed by distance estimation between sequences as a measure of the number of allelic substitutions on selected loci, as described earlier [21].

Statistical Analyses
We analysed Borrelia prevalence in ticks using chi-squared test with Yates' correction. Differences in mean intensity of tick infestation by Mann-Whitney U test with p < 0.05 were considered statistically significant. All calculations were done using Statistica 8.0 software (StatSoft Inc., Palo Alto, CA, USA).

Tick Species Identification
Of the 243 red foxes, 120 (49.4%) hosted on their ears 1583 Ixodes ticks (Table 2). Using morphological criteria for tick females, four Ixodes species were identified: I. ricinus, I. kaiseri, I. canisuga, and I. hexagonus. Analysis of ITS2 DNA fragments of these females according to the method described by Wodecka et al. [21], confirmed morphological identification. Analysis of 43 ITS2 sequences obtained from ticks representing different PCR products or restriction patterns, revealed also four distinct groups represented by I. kaiseri (n = 9), I. canisuga (n = 11), I. hexagonus (n = 8), and I. ricinus (n = 15). Comparative analysis of these sequences generated mean distance values within each species as follows: 0.0 for I. canisuga, I. kaiseri, and I. hexagonus, and 0.023 for I. ricinus. The analysis of mean distance values between species showed the highest value for I. ricinus and the three nidicolous tick Pathogens 2022, 11, 696 7 of 20 species: I. canisuga, I. kaiseri, and I. hexagonus (0.262, 0.387, and 0.332, respectively). The distance among these ticks reached: 0.118 for I. kaiseri and I. hexagonus, 0.126 for I. canisuga and I. hexagonus, and 0.185 for I. canisuga and I. kaiseri.
To confirm the ITS2 identification of tick species and its conformity with morphological characterization according to a key by Hornok et al. [18], the analysis of 35 coxI sequences was carried out. This analysis confirmed four distinct groups represented by I. kaiseri (n = 5), I. canisuga (n = 5), I. hexagonus (n = 8), and I. ricinus (n = 17). Comparative analysis of these sequences revealed different mean distance values within each species as represented ITS2 sequences analysis: 0.004 for I. canisuga, 0.002 for I. kaiseri, 0.001 for I. hexagonus, and 0.005 for I. ricinus. The analysis of mean distance values between species confirmed their molecular distinctiveness and the highest value for the exophilic I. ricinus and two nidicolous tick species I. kaiseri and I. hexagonus (0.306 and 0.316, respectively). Surprisingly the distance value for I. ricinus and I. canisuga (0.178) was lower than those for I. canisuga and I. kaiseri (0.212) and I. canisuga and I. hexagonus (0.226). The distance for I. kaiseri and I. hexagonus was the lowest (0.142).

Occurrence of Tick Species on Ears of Red Foxes
Out of 1583 Ixodes specimens collected from 120 foxes, 1341 were larvae, 166 nymphs, and 76 females. On an average, one infested animal hosted 13.2 ticks ( Table 2). Larvae distinctly prevailed over nymphs and females (84.7% vs. 10.5% and 4.8% of total tick numbers, respectively). Larvae infested 34% on animals, nymphs 30%, and females 17%. In total, 55 (45.8%) of the infested animals were parasitized concurrently by at least two of the three different tick developmental stages.
Most I. kaiseri (92%, n = 1109) and I. canisuga (81%, n = 152) ticks were larvae which contributed to the larval predominance on the infested foxes. Except for the rare I. hexagonus, loads of nymphs were comparatively low for each of the other three tick species as sparser than larvae intermediate stage (range: 0.1-0.3 ticks per animal). Female ticks were represented mostly by I. ricinus and I. kaiseri, which constituted 84% of the 76 females.
Each of the four Ixodes species occurred on foxes year-round (Table S1). Ixodes ricinus demonstrated clear seasonality, with intensity and prevalence values decreasing steadily from spring-summer to winter. Ixodes canisuga tended to be more prevalent and abundant in winter, whereas the remaining ticks were collected mostly in autumn and winter months. Ixodes kaiseri was more abundant in spring-summer and winter.

Detection of Borrelia DNA by Nested PCR in Foxes and Their Ticks
Of the 558 tissue samples obtained from the 243 red foxes, 67 (12%) displayed Borrelia spp. specific DNA ( Table 3). The bacterium was identified in all three isolate types, but it occurred more frequently in blood samples (23.6%) than in liver (6.1%) or skin (4.1%) biopsies. The infected isolates were derived from 57 (23.5%) of the animals tested. A total of 47 (82.5%) of the infected foxes revealed the bacterium in one isolate type. In this group, blood PCR positive animals (n = 42) prevailed over those carrying the pathogen in liver (n = 3) or skin (n = 2) biopsies. A total of 10 animals (17.5%) harboured spirochetes concurrently in two isolate types: in blood and skin samples (n = 6), in blood and liver  Table S2). The infected foxes were detected in three of the nine sampling sites (Figure 1) with the highest prevalence in Margonin (34/44, 77%) and Czerniejewo (22/36, 61%), and the lowest in Zielonka (1/23, 4.3%). Borrelia DNA was identified in 306 (32.4%) of the 943 ticks tested and occurred in all parasitic life stages (Table 4). Of the 120 infested animals, 81 (67.5%) hosted at least one infected tick. Of the 81 hosts, 46 (56.8%) carried infected larvae. A total of 12 (14.8%) of the foxes carrying infected ticks proved to harbour concurrently Borrelia spirochetes. Female and nymphal ticks were infected with the bacterium more frequently than larvae (47.4% and 43.9% vs. 28.1%, respectively; p = 0.014). The overall prevalence of infection in I. ricinus ticks was comparable to those recorded in three nidicolous tick species.

Identification of Borrelia Species
Altogether, PCR positive for flaB gene tissue samples from red foxes and ticks revealed 11 unique RFLP patterns corresponding with different Borrelia species. Of the 11 species, 10 were attributed to the B. burgdorferi s.l. complex i.e., B. garinii, B. afzelii, B. spielmanii, B. burgdorferi s. s., B. valaisiana, B. bissettiae, B. carolinensis, B. californiensis, B. lanei, and B. americana. The first three species were found in the infected foxes, whereas PCR positive ticks revealed all 10 spirochete species (Tables 3 and 5). Furthermore B. turcica of the reptile-associated borreliae (REP) group was found in a single larva of I. kaiseri.   (Table 3). A total of 10 infected hosts revealed spirochetes concurrently in two different tissue isolate types (Table S2).
The analysis of the 44 sequences of mag-trnI intergenic spacer confirmed all 11 identified spirochete species (Table S7). Comparative analysis of these sequences revealed not only different mean distance values within each species, as represented by the flaB and p66 gene sequences analysis, but also differences in length of sequence, depending on Borrelia species (Table S7) (Table S8).

Discussion
Based on morphological and molecular methods, we documented the occurrence of four Ixodes tick species on the ears of red foxes. Furthermore, we provide evidence of Borrelia spp. infections in tissue samples of red foxes and in their Ixodes ticks collected from the ears.

Ixodes Tick Species on Red Foxes
We confirmed the presence of three nidicolous Ixodes species of the subgenus Pholeoixodes, namely I. kaiseri, I. canisuga, I. hexagonus, and the generalist exophilic I. ricinus on red foxes. The first three species are considered nonhuman-biting and are associated mainly with Canidae and Mustelidae and reproduce inside the burrows of their hosts [22,27]. Ixodes hexagonus infests hedgehogs also those inhabiting sub-and urban areas [28]. A taxonomic key published by Hornok et al. [18] allowed us to identify Pholeoixodes females of the three tick species collected in our study. Furthermore, comparison of the coxI sequences obtained from the nidicolous tick species with those published by Hornok et al. [18] confirmed high reliability of this molecular marker in accuracy of species identification of ticks. Therefore, using the coxI gene, we re-examined feeding ticks derived from raccoon dogs and badgers tested in our previous study [21]. This new analysis showed that engorged females of I. kaiseri have been misnamed on the basis of morphological analysis and ITS2 reference sequences deposited in GenBank database as I. canisuga and vice versa. As it turned out, I. kaiseri, and not I. canisuga, dominated both on raccoon dogs and badgers (44.8% vs. 14% and 48.2% vs. 18.5%, respectively). Our present study confirmed the importance of I. kaiseri as the most prevalent tick species associated with wild carnivores in west-central Poland. Moreover, we suggest that in Central Europe, I. kaiseri parasitizes these mammals more frequently than previously thought and its former distribution is not restricted to Romania, the Republic of Moldavia, and Ukraine [22,29]. This confirms the report by Hornok et al. [18] in which the presence of I. kaiseri was documented on carnivores from Germany, Hungary, Romania, and Serbia, expanding its known geographical range in Europe. In our opinion, its hitherto infrequent occurrence on European carnivore species results from misidentification of the three nidicolous Ixodes species using only the morphological approach, therefore, molecular verification is necessary to avoid any misidentification between morphologically similar Ixodes species of the subgenus Pholeoixodes.

Borrelia Infections in Red Foxes
We found that overall, 23.5% of red foxes revealed DNA of Borrelia spp. Interestingly, over 92% of the 51 blood PCR positive hosts (including 21.3% double infections with B. afzelii) carried B. garinii. Infections caused by B. garinii were also found in skin and liver biopsies. This finding indicates that red foxes develop a disseminated infection with B. garinii more frequently than with B. afzelii, known to exhibit the host specificity for mammals [30]. We also found a predominance of B. garinii over B. afzelii (62.5% vs. 25%) in PCR-positive raccoon dogs [21]. Co-occurrence of both spirochete species has been reported in one European hedgehog Erinaceus europaeus [31] and one Siberian chipmunk Tamias sibiricus barberi [32]. Dumitrache et al. [16] found only B. afzelii (n = 4) and B. burgdorferi s.s. (n = 1) in 1.4% of 353 fox heart isolates tested in Romania. In a German study 24% of skin samples of red foxes had only B. garinii [14]. Furthermore, evidence of B. garinii in blood of a dog from the Czech Republic [33] and in two dogs in Japan was reported [34]. Therefore, we suggest that this avian-adapted spirochete seems to be often associated with canids, including the red fox. However, to clarify its status as a potential reservoir for B. garinii, xenodiagnostic experiments are necessary.

Borrelia Infections in Ixodes Ticks
The overall infection prevalence found in the four Ixodes tick species collected from ears of red foxes was 32.4% and ranged from 30% to 37%, depending on tick species. The PCR-positive ticks were found year-round. A total of 10 species of the B. burgdorferi s.l. complex were identified, and I. ricinus was infected with eight species, whereas their number in the burrow-dwelling ticks ranged from four (I. hexagonus) to 10 (I. canisuga, and I. kaiseri). So far, there is only one report from Switzerland in I. ricinus ticks infected with eight B. burgdorferi s.l. species [35]. To our knowledge, none of the European mammals has been found to be exposed to such diversity of B. burgdorferi s.l. species as we report here for fox-derived ticks.
Similarly, as in the case of the PCR-positive foxes, B. garinii and B. afzelii comprised most (70%) of the total infections, and reached comparable infection prevalence in each of the four tick species. The distinct predominance of B. garinii and B. afzelii in infected larvae, as well as in foxes, suggests that larvae could have acquired infection while feeding on animals infected with these spirochete species, especially that larval ticks are rarely infected transovarially with B. burgdorferi s.l. Interestingly, the most abundant B. afzelii and B. garinii are associated with rodents and birds, respectively [30], the vertebrates serving as the main food source for red foxes [1]. Therefore, their abundance in our study is not a surprise, especially in the case of I. ricinus that may be shared by the fox and its prey. The presence of the remaining eight species in fox-derived ticks and their absence in hosts, might be explained by co-feeding transmission between spatially clustered infected and uninfected ticks [36]. The observed high aggregation of larval ticks on the ears of the foxes, and the fact that nearly half (46%) of the infested animals were parasitized concurrently by larvae and nymphs or females could greatly enhance the efficiency of this localized non-systemic transmission ( Figure 2). Furthermore, the finding that in the case of 11 non-infected foxes, larvae co-feeding with nymphs or females carried the same spirochete species supports our assumption (Table 6). This mode of transmission may facilitate contact and exchange between Borrelia species adapted to different vertebrate host species [37].
Surprisingly, four B. burgdorferi s.l. species previously believed to be restricted only to North America: B. carolinensis, B. californiensis, B. americana, and B. lanei were identified. The first two species were the most prevalent in PCR-positive ticks after B. garinii and B. afzelii and constituted almost 20% of all infections. To date, there are only two European reports documenting the presence of B. carolinensis in ticks: first in one (2.9%) questing I. ricinus from France [38], and the second case together with B. lanei detected for the first time in Europe in three Ixodes tick species associated with European bats from Poland and Romania [39]. In South Carolina and California enzootic cycles of B. carolinensis involve small rodents and a nidicolous I. minor [40,41]. Borrelia californiensis is maintained by Cali-fornia kangaroo rats Dipodomys californicus and two non-human (or rarely) biting I. jellisoni and I. spinipalipis as well as I. pacificus, the key vector of LD species in the western United States [42]. Importantly, in our study B. carolinensis and, for the first time in Europe, B. californiensis were also detected in PCR-positive I. ricinus ticks. Unexpectedly, we detected sequences almost identical to California strains of B. americana in two females of I. ricinus and individual nymphs of I. cansiuga and I. kaiseri. This spirochete was first identified in I. minor from birds in South Carolina as well as from I. pacificus in California [43]. Outside the United States, it was detected only in I. persulcatus ticks in China [44] and, recently, in two I. ricinus specimens in Poland [45]. Thus, our study reaffirms that B. americana occurs in Ixodes ticks in three continents, but its vertebrate host(s) in Europe remains unknown. We identified B. lanei in I. kaiseri and I. canisuga. Considering the study by Michalik et al. [39], it is the second report documenting B. lanei in the group of European Ixodes ticks inhabiting the burrows or caves of their hosts. In the USA this spirochete perpetuates in enzootic cycles involving lagomorphs and rodents and the two tick species, of which I. spinipalpis is nidicolous [46,47]. eight B. burgdorferi s.l. species [35]. To our knowledge, none of the European mammals has been found to be exposed to such diversity of B. burgdorferi s.l. species as we report here for fox-derived ticks.
Similarly, as in the case of the PCR-positive foxes, B. garinii and B. afzelii comprised most (70%) of the total infections, and reached comparable infection prevalence in each of the four tick species. The distinct predominance of B. garinii and B. afzelii in infected larvae, as well as in foxes, suggests that larvae could have acquired infection while feeding on animals infected with these spirochete species, especially that larval ticks are rarely infected transovarially with B. burgdorferi s.l. Interestingly, the most abundant B. afzelii and B. garinii are associated with rodents and birds, respectively [30], the vertebrates serving as the main food source for red foxes [1]. Therefore, their abundance in our study is not a surprise, especially in the case of I. ricinus that may be shared by the fox and its prey. The presence of the remaining eight species in fox-derived ticks and their absence in hosts, might be explained by co-feeding transmission between spatially clustered infected and uninfected ticks [36]. The observed high aggregation of larval ticks on the ears of the foxes, and the fact that nearly half (46%) of the infested animals were parasitized concurrently by larvae and nymphs or females could greatly enhance the efficiency of this localized non-systemic transmission ( Figure 2). Furthermore, the finding that in the case of 11 noninfected foxes, larvae co-feeding with nymphs or females carried the same spirochete species supports our assumption ( Table 6). This mode of transmission may facilitate contact and exchange between Borrelia species adapted to different vertebrate host species [37].  Borrelia bissettiae like B. burgdorferi s.s., occurs in both the Old and the New World and is considered a human pathogen in Europe and a likely one in the United States [48,49]. In the United States transmission cycles of this species are driven by I. pacificus, and three nidicolous species, I. spinipalpis, I. minor, and I. affinis, and various rodent species acting as reservoirs [42,50,51], and has been detected in the blood of several bird species and I. pacificus feeding on them [52]. Our findings suggest that three burrow-dwelling tick species associated with the red fox could similarly serve as enzootic maintenance vectors of this rarely recorded spirochete in Europe, for which prevalence in I. ricinus is very low [53][54][55]. Furthermore, the finding that B. bissettiae infected ticks were collected only in three collection sites, confirms its highly focal distribution. Although mixed infections of B. bissettiae and B. carolinensis were found in tissues from the European hedgehog, and the Eurasian red squirrel Sciurus vulgaris from the Czech Republic [56], any association of this spirochete with particular vertebrate host(s) in Europe is speculative, at best. Borrelia spielmanii, with pathogenic potential to humans [57], was detected with a low 1% total prevalence, including six larvae of I. kaiseri, two I. canisuga, and a single I. ricinus female. Together with B. afzelii and B. bavariensis it is considered as maintained by small-and medium-sized mammals [58,59]. Interestingly, it prevailed in bat specific Ixodes ticks [39] and was also reported in bird-derived ticks [60,61].
Enigmatic is evidence of B. turcica in a single larva of I. kaiseri. This spirochete was originally isolated from Hyalomma aegyptium ticks associated with Mediterranean Testudo tortoises in Turkey [62] and is a member of the reptile-associated borreliae (REP), representing the third major group of spirochetes, distinct from LD and relapsing fever (RF) borreliae [63]. Hyalomma aegyptium ticks occur in northern Africa, western Asia, and southern Europe and, apart from their preferential hosts, tortoises, the immature stages may alternatively feed on lizards, birds, and small-or medium-sized mammals [64,65]. Presumably, these vertebrates might influence expansion some pathogens associated with H. aegyptium beyond typical geographical areas. Since the ecology of B. turcica is not sufficiently known [66], the definite cause of detection of this bacterium in a I. kaiseri larva requires further studies.
Of the 14 B. burgdorferi s.l. species so far detected in Europe ( [39,61,67],), 10 were detected in red-fox derived ticks, including three identified in tissues of these hosts. Among the 10 species found in ticks, six were discovered for the first time in North America, namely: B. burgdorferi s.s., B. bissettiae, B. americana, B. carolinensis, B. californiensis, and B. lanei. These species together with B. kurtenbachii, detected so far only in North America [68], and B. finlandensis, known solely from Europe [69], are the most closely related in the whole LD stock. To confirm and determine their proper identification we used three molecular markers, i.e., flaB and p66 gene fragments and mag-trnI intergenic spacer. Regardless of the marker used, their examination confirmed less genetic distance value between the six species stated in North America and Europe than between the other four LD species detected in this study. A possible explanation why so many closely related species can exist on continents separated by the Atlantic Ocean may be the hypothesis of Borrelia evolution proposed by Estrada-Peña et al. [70]. The authors assume common evolution of the ancestors of all modern ticks and Borrelia species and species level-differentiation at the time of existence of the Pangea supercontinent (between 300 and 180 million years ago). Their contemporary occurrence is the effect of species expansion within Pangea and then separation of individual Borrelia species or populations transmitted by competent tick vectors as the result of the land masses moving apart and then forming the present continents. According to this hypothesis, it is not possible to delineate a "European" or "American" origin of the B. burgdorferi s.l. stock. Their common ancestor might evolve from the relapsing fever group before the Pangea supercontinent splitting and forming of Laurasia and Gondwana [70].
Borrelia garinii identified in red foxes, as well as in their ticks, showed a high level of diversity, which is characteristic of this species. However, unlike birds, carnivore mammals are not recognized as its reservoir. Studies of seabird colonies within the North Atlantic region revealed alongside local genetic variants of B. garinii, also the genotypes similar to those obtained from some human clinical samples in Europe [71]. The North Atlantic region is a particular habitat for B. garinii with the seabird tick Ixodes uriae [72]. Similarly, carnivore mammals, including foxes with their specific tick fauna species, might create conditions favorable to new genetic variant of this bacterium and the remaining LD Borrelia species detected in this study.
Except for the three most prevalent B. burgdorferi s.l. species (B. garinii, B. afzelii, and B. burgdorferi s.s.), the other seven species displayed a highly patchy distribution and were found in three, two or one of our sampling sites. Notably, the highest diversity of the rare spirochetes (i.e., B. bissettiae, B. carolinensis, B. californiensis, and B. lanei) was recorded in Niemierzewo Forest Range. It is located within Sierakowski Landscape Park with diverse ecosystems, including moraine hills, river valleys, dunes, lakes, arable fields, and various forest types. These heterogeneous ecosystems may account for the observed high diversity of Borrelia species, and the park may act as their biodiversity hotspot.

Conclusions
Our findings underscore the importance of burrow-dwelling Ixodes species in the ecology of B. burgdorferi s.l. in west-central Poland. The presence of 10 species of this complex in carnivore-associated I. kaiseri and I. canisuga indicates that these non-human biting ticks may serve as maintenance vectors in silent enzootic transmission cycles. Although their vectorial capacity remains unknown, they could potentially increase diversity of Borrelia species which may be transferred to the local population of the generalist tick I. ricinus. Detection in fox-derived samples, only three of the 10 spirochete species infecting engorged ticks, implies that these carnivores appear to be rather incompetent, or exhibit reduced and/or short-lived infectivity for most of them. It cannot be excluded that red foxes in the absence of widespread and long-term infections could serve as so-called "co-feeding transmission hosts" [73], especially that almost in half of the tick-infested animals the co-feeding events were observed. Nevertheless, competent reservoir species for these spirochete species remain unknown and we cannot come to any final conclusions. On the other hand, a distinct predominance of B. garinii in PCR-positive animals, as well as in infected fox-derived larval ticks, suggests that this species, along with B. afzelii, the second most prevalent species, may be preferentially associated with canids. It means that avian and mammalian-adapted spirochete species occupying different ecological niches might be concurrently maintained by red foxes.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/pathogens11060696/s1, Table S1. Seasonal occurrence of four Ixodes species collected from the ears of 243 red foxes harvested during the fox-hunting seasons (1 June to 31 March) between 2009 and 2011 in west-central Poland. Table S2. Detection of LD Borrelia species concurrently in two isolate types derived from red foxes. BA-B. afzelii, BG-B. garinii. Table  S3. Comparison of the selected 112 partial flaB gene sequences of Borrelia species amplified from red foxes (blood, and skin samples: n = 13, and 2, respectively) and their engorged Ixodes ticks (n = 97) sampled in this study. Table S4. MEGA X results of mean distance between Borrelia species obtained on the basis of flaB gene sequence fragment comparison. Table S5. Comparison of the selected 33 partial p66 gene sequences of B. burgdorferi s.l. species amplified from engorged Ixodes ticks removed from red foxes sampled in this study. Table S6. MEGA X results of mean distance between B. burgdorferi s.l. species obtained on the basis of p66 gene sequence fragment comparison. Table S7. Comparison of the selected 44 partial intergenic spacer (IGS) of 3-methyladenine glycosylase (mag) and tRNA-Ile (trnI) genes sequences of Borrelia species amplified from engorged Ixodes ticks removed from red foxes sampled in this study. Table S8. MEGA X results of mean distance between Borrelia species obtained on the basis of intergenic spacer (IGS) of 3-methyladenine glycosylase (mag) and tRNA-Ile genes sequence fragment comparison.

Institutional Review Broad Statement:
The study was not conducted on live animals and none of the foxes was mortified for the purpose of the study. All foxes used for the study came from hunting conducted as planned wildlife management and were already dead while the tissues and blood were collected. Therefore, the Ethics Committee or Institutional Review Board approval is not required.

Informed Consent Statement: Not applicable.
Data Availability Statement: All data generated or analysed during this study are included in this published article and its supplementary information file. The accession numbers of DNA sequences obtained for ticks and bacteria are mentioned in Material and Methods and are available in the GenBank (https://ww.ncbi.nlm.nih.gov/nuccore, accessed on 23 March 2018).