Different Parasite Faunas in Sympatric Populations of Sister Hedgehog Species in a Secondary Contact Zone

Providing descriptive data on parasite diversity and load in sister species is a first step in addressing the role of host-parasite coevolution in the speciation process. In this study we compare the parasite faunas of the closely related hedgehog species Erinaceus europaeus and E. roumanicus from the Czech Republic where both occur in limited sympatry. We examined 109 hedgehogs from 21 localities within this secondary contact zone. Three species of ectoparasites and nine species of endoparasites were recorded. Significantly higher abundances and prevalences were found for Capillaria spp. and Brachylaemus erinacei in E. europaeus compared to E. roumanicus and higher mean infection rates and prevalences for Hymenolepis erinacei, Physaloptera clausa and Nephridiorhynchus major in E. roumanicus compared to E. europaeus. Divergence in the composition of the parasite fauna, except for Capillaria spp., which seem to be very unspecific, may be related to the complicated demography of their hosts connected with Pleistocene climate oscillations and consequent range dynamics. The fact that all parasite species with different abundances in E. europaeus and E. roumanicus belong to intestinal forms indicates a possible diversification of trophic niches between both sister hedgehog species.


Introduction
The restriction of populations of a wide variety of European species to spatially limited refuge areas during the cyclic climatic changes of the Pleistocene, together Brachylecithum mackoi from the island of Elba [14]. These may play an important role in host morbidity and mortality (see e.g. [15], [16]). However, very little is known about the similarities and differences in the parasite fauna of the three Eurasian hedgehog species. Given the relatively deep split defining these hedgehog species, as well as the cyclic restriction in distributional area and population size, we predict that parasite co-speciation may have occurred and that the parasite community infesting the particular host species may differ. Thus, investigating the composition of the parasite fauna may provide valuable information regarding the phylogenetic differentiation, niche and population dynamics, diversification and species interactions in this group.
In this study we compare the parasite faunas of ER and EE from contact zones in the Czech Republic and interpret this information with regard to the phylogeography of the genus.

Material and Methods
The hedgehogs from this study were either provided by wildlife rescue centers, where the animals died naturally, or were collected as road kills. Therefore an ethical approval by the relevant ethics committee was not required. All specimens were frozen at 220˚C until they were used for dissection. Prior to examination individuals were thawed at room temperature, weighed and the sex was determined. Animals were classified either as hoglets (,100 g), juveniles (,500 g) or adults (.500 g) [16].
The samples originated from 21 different areas in the Czech Republic ( Figure 1, Table S1). Collections were made during 2008-2011. In total we examined 109 animals (ER 527, EE 582). For ER we collected 19 juveniles (female 511, male 58) and seven adults (female 52, male 55). For EE we found eight hoglets (female 53, male 55), 61 juveniles (female 530, male 531) and 12 adults (female 55, male 57). Genetic analysis of mitochondrial DNA and nucelar microsatellites were used for determination of 20 specimens of EE and 14 specimens of ER, which were identical with the study of Bolfíková & Hulva [10]. Considering the presumed absence of hybridization (or very low degree of introgression) between both species in Central Europe ascertained by reciprocal monophyly of respective clades and the absence of intermediate phenotypes in the analysis of more than 200 individuals within the contact zone [10], morphology-based discrimination was used for remaining specimens within the present study.
Fleas and ticks were collected, identified to life history stage, sex (if not immature) as well as to species after Beaucournou & Launay [17] and Arthur [18], respectively, and subsequently quantified. Due to the difficulty of quantifying infestation rates, mites were not included in the examination. The body cavity (peritoneum), connective tissue and the surface of the organs were examined for encysted acanthocephalans. The lung was examined under a binocular microscope (Stemi 2000, Carl Zeiss Mikroskopie, Jena, 07740, Germany) for nematode infections. Crenosoma striatum and Capillaria aerophila from the bronchi and bronchioles were quantified. The stomach and the intestine were stored overnight in tap water in the refrigerator at 4˚C to allow the intestinal parasites from the intestinal wall to move into the water. The next day, the water and the intestinal sections were examined under a binocular microscope (water with transmitted light, intestinal sections with direct light). All parasites found were identified to species after Beck & Pantchev [19] and quantified. Information about the taxonomic status, the habitat preference and the host specificity of the parasites found in this study are listed in Table S2.
All statistical analyses were conducted using IBM SPSS Statistics Version 20. To test for differences in parasite abundance between sex, age and species groups, a Mann-Whitney U-test was used. To test for differences in parasite prevalence between sex, age and species group we used a chi-square test.

Results
In total, twelve parasite species were determined (Table 1). Ectoparasites included one flea species, the hedgehog flea Archaeopsylla erinacei and two tick species, the hedgehog tick Ixodes hexagonus and the castor bean tick I. ricinus. The lungworm C. striatum and C. aerophila were found in the lungs. In the intestines Capillaria spp., Physaloptera clausa (only found in the stomach) and one unidentified nematode, the trematode Brachylaemus erinacei, the cestode Hymenolepis erinacei and the acanthocephalans Nephridiorhynchus major and Plagiorhynchus cylindraceus were found.
Since none of the hoglets were parasitized, they were eliminated from further statistical analysis. We did not find any significant differences in parasitization between the sexes. We therefore pooled the sexes in each age group. We could not find any differences between age groups for ER. For EE we found higher mean abundances of C. striatum (p50.01), B. erinacei (p50.006) and lower abundances of H. erinacei (p50.02) in juveniles compared to adult animals (Table 1). Therefore we treated hedgehog age groups for those parasites separately for further statistical analysis. For all other parasites the age groups were pooled.
When we compared the differences in mean parasite infection rates between the two hedgehog species (Mann-Whitney U-test) we found significant differences for H. erinacei from the intestines for juveniles (p50.009), B. erinacei for juveniles (p,0.001), Capillaria spp. (p50.026), P. clausa (p,0.001) and N. major (p,0.001). The infection rates with B. erinacei and Capillaria spp. were higher in EE, while ER showed higher infection rates with H. erinacei, P. clausa and N. major (Table 1). Similar results were found when we compared the prevalences of the parasite species between EE and ER. EE showed higher prevalences of B. erinacei in juveniles (x 2 1 513.015, p,0.001), while ER had higher prevalences for P. clausa (x 2 1 520.371, p,0.001) and N. major (x 2 1 520.53, p,0.001).

Discussion
Neutral and adaptive changes during parasite-host coevolution are likely to affect the population and community attributes in both the host and the parasite. In the host species, with demography and range history affected by Pleistocene climatic oscillations, parasites may undergo complicated evolution [20]. Neutral evolution might occur during refugial, peripatric isolation and recurrent bottlenecks during population re-expansions, including founder effects causing parasite release, genetic drift and other factors acting on small populations. Site-specific adaptive responses may occur as well. These may culminate in the extinction of particular parasite species in a particular host lineage or in allopatric speciation. Both would contribute to divergence of parasite faunas between sister host species. On the other hand, parasites may affect evolution of the host, for example via parasitemediated selection acting in genes regulating immune defense, and may contribute to the host speciation process. Here we show that although there do not appear to be any major morphological changes in the species infesting the two hedgehog species in their zone of overlap, there are considerable differences in the prevalence and intensity of infestation by a variety of endoparasitic species. In order to determine how significant these changes are it is also necessary to consider completely allopatric population of both host species.

Divergent patterns of parasite diversity and load
Many studies on the parasite fauna of E. erinacei were carried out in the late 1970s and 1980s (e.g. [21][22][23][24]). These are complimented by several more recent studies [16], [25][26][27]. In contrast to this, only limited information is available on the parasites of EC (e.g. [28]) and ER (e.g. [29]). With the exception of the work of [29] using fecal analysis to determine the helminth fauna of EE and ER in the contact zone in Poland, comparative studies for both species are lacking. All parasites found in this study have been found previously in or on hedgehogs (e.g. [16], [21], [26], [30][31][32]), although molecular analysis will be needed to confirm that no sibling species are present. If available, when comparing this study to other studies hedgehog sample size (n) and mean abundances ( x) of the other studies will be provided in brackets.
Abundances and prevalences of the flea A. erinacei from this study are relatively low. Egli [33] reports prevalences from EE (n5135) of 43.7% in Switzerland and Visser et al. [34] of 84.2% in Germany (n576). This also applies for ER from Hungary where Földvari et al. [35] found prevalences between 26.3% and 72.1% (n5247). Beck & Clark [36] and Beck et al. [37] even state that every hedgehog is to a greater or lesser extent infested with the hedgehog flea.
Both I. ricinus and I. hexagonus are also frequently found on hedgehogs. As for the hedgehog flea, tick abundances and prevalence were lower in the present study compared to others. Ixodes ricinus prevalences on hedgehogs from Germany (23.4%, n5133, x52.83 [16]) and Switzerland (11.1%, n5135, [33]) were higher than 0% for EE and 4% for ER in the present study. For I. hexagonus Pfäffle [16] found prevalences of 53.3% (n5133, x547.56) and 40% (n530, x523.77) for EE from Germany and the UK, respectively. Egli [33] determined I. hexagonus prevalences of 58.5% for EE (n5135). However, Földvari et al. [35] found I. hexagonus prevalences of only 1.1% on ER from Hungary (n5247). They indicate that this low prevalence could be related to the high hedgehog density on Margaret Island where the study was conducted. Such high hedgehog population densities in semi-natural environments might lead to higher densities of I. ricinus compared to I. hexagonus [38]. Almost all of the animals from the present study came from wildlife rescue centers where they were treated for injuries or symptoms of disease. During this time both ticks and fleas might have been collected by the caretakers or left and dropped of the host, respectively. Therefore our data for ectoparasite prevalences and abundances might well be imprecise and not express the natural infestation rates of EE and ER in the Czech Republic. However, the results do provide an insight into the species of ectoparasites found on both Czech hedgehog species.
Capillaria aerophila is found in the smaller bronchi and is usually less abundant than the lungworm C. striatum, but can cause similar symptoms, such as weight loss, bronchitis and pulmonary damage [16], [24], [42]. The prevalences of C. aerophila found in EE from the present study are comparable to findings from other locations ( Table 2). The prevalence of C. aerophila infections in ER is comparably lower than the prevalences found in the study of Mizgajska et al. [29]. However, it should be noted that the information about the parasite fauna of ER is very scarce.
In addition to the findings from Edelenyi & Szabo [43], this is the first time that P. cylindraceus has been described for ER, although it has been described from the Czech Republic in EE by Prokopic [44] (1957, syn. Prosthorhynchus jormosus). It is an intestinal parasite of passerine birds which is sporadically found in the intestinal tracts of mammals, causing diarrhea, peritonitis and sometimes increased mortality [32]. Infections seem to occur more often in juveniles than in adult animals, since younger animals also feed on unpalatable prey like woodlice, which are the intermediate hosts for this parasite [32], [45]. However, we were not able to find significant differences in infection rates between adults and juveniles neither for EE nor for ER.
Cestode infections are rare for EE, prevalence is normally low and restricted to certain regions (e.g. [23] EE, n5754, 0.39%; [31] EE, n539, 8%; [46] EE, n5437, 0.7%). In ER and EC infections seem to be more common ( [28] EC, n518, 55%; Pfäffle unpublished data). However, in the present study only one EE and three ER were infected with H. erinacei, while Mizgajska et al. [29] did not find any cestode infections at all. All three infected ER came from Prague, while the infected EE originated from Kocbere. These results indicate that the abundance of H. erinacei might also be restricted to certain regions in the Czech Republic. Nevertheless, the sample size might have been too small to support this hypothesis and it is not possible to draw conclusions as to whether the differences in infection rates are species dependent on or are influenced by other factors.
We found higher abundances and prevalences of B. erinacei in juvenile EE compared to juvenile ER and higher mean infection rates with Capillaria spp. in EE compared to ER in all age groups. Abundances and prevalence of P. clausa and N. major where higher in ER compared to EE (all age groups). Capillaria spp. are common parasites of hedgehogs and can reach high prevalence (up to 90%) and intensities (see [16], [25], [29]). Capillaria spp. can be transmitted directly or indirectly via the ingestion of earthworms. They can have a severe effect on the body condition of hedgehogs, which might be increased during periods of high

Parasite species
Central Europe [16], [21], [23], [33], [41], [62][63][64][65][66] Spain [40], [53] Italy [50], [51], [54], [67] Capillaria stress, for example during the reproductive phase or hibernation, and higher hedgehog population densities might increase the transmission rates between individuals, hence increasing prevalences and abundances [16]. The work of Mizgajska et al. [29] on Capillaria spp. showed infections in EE and ER which were the opposite of those found in the present study, with ER having higher prevalences than EE. It seems that this parasite is neither specific for, nor occurs predominantly in, a certain hedgehog species and that there is a high variability in infection rates dependent on region and habitat. Brachylaemus infections were both higher in prevalence and abundance in juvenile EE than in juvenile ER. This is comparable to the study by Mizgajska et al. [29], where only EE (n515, 33%) were infected with trematodes. Brachylaemus erinacei is host specific [46], transmitted via the ingestion of various intermediate gastropod hosts [47] and can cause diarrhea, hemorrhagic enteritis, inflammation of the bile ducts, anemia and death [39], [42], [48].
In general, the parasite fauna of the best studied species, EE, is relatively consistent throughout its range, although data from Spain [40], [53] and Italy [50], [51], [54] suggest that at least within these refuges there is a slightly higher parasite species diversity than further north (Table 2).

Evolutionary interpretations
A detailed comparison of the ecology of host both species is still missing. The existence of a relatively large zone of overlap can be viewed as a natural, large scale, common-garden experiment, which could be utilized for studying the ecological diversification of both lineages with a similar environmental background. The landscape genetic analysis from central Europe points to differences in the altitudinal distribution of EE and ER, indicating at least some ecological differentiation [10]. However, both species can occur in similar habitats and even syntopically, in rural, suburban and urban habitats. EE and ER seem to be essentially similar in their feeding ecology (e.g. [55][56][57][58][59][60][61]). The diet consists mainly of a variety of invertebrates, usually with a few main prey types such as beetles, caterpillars, earthworms, slugs and snails, which can act as intermediate hosts of various parasite species [13]. However animals from different regions show their own particular spectrum of prey items [13]. The fact that all parasite species with significantly different abundances in EE and ER are intestinal forms nevertheless indicates possible diversification of trophic niches between these sister hedgehog species.
Although definite quantitative differences were found in prevalences and intensities of infection by certain parasite species between the two hedgehog species, qualitative differences in terms of differences in species composition were not apparent. However, as species identification was carried out morphologically, this does not exclude the possibility of cryptic variation in studied species. In order to determine the degree of divergence, and potentially introgression after secondary contact, a molecular study of the parasites should be carried out.