Ectoparasites of the European wildcat (Felis silvestris) in Germany

Understanding the impact of parasites on wildlife populations is an important aspect of conservation management. However, research on ectoparasites in wildlife can be difficult, as examinations of live animals which are not habituated to human handling are often impossible. The European wildcat (Felis silvestris) is a strictly protected wildlife species whose population has been recovering in Germany in recent decades. Several studies from different European countries have investigated the parasitological status of European wildcat populations. However, most of these studies assessed endoparasite infections, whereas ectoparasite infestations have often been neglected. To fill this knowledge gap for wildcats in Germany, 131 dead found specimens were examined for ectoparasites by macroscopic and microscopic examination of the fur and the ear canals. Infestation with ectoparasites was present in 84.0% (110/131) of the wildcats. Ticks showed the highest prevalence with 72.5% (95/131) of wildcats infested, with 49.6% (65/131) infested with Ixodes ricinus and 36.6% (48/131) with Ixodes hexagonus/canisuga. A total of 27.5% (36/131) of the wildcats were positive for at least one flea species. Of the nine different flea species identified by morphology and/or molecular analyses, Ceratophyllidae were most common (16.8% [22/131]), with Ceratophyllus sciurorum confirmed on 12.2% (16/131) and Nosopsyllus fasciatus on 1.5% (2/131) animals, followed by Pulex irritans (5.3% [7/131]), Spilopsyllus cuniculi (3.8% [5/131]), Chaetopsylla spp. (3.1% [4/131]) (2/131 Chaetopsylla trichosa and 1/131 Chaetopsylla globiceps), Ctenocephalides felis (1.5% [2/131]), Archaeopsylla erinacei (1.5% [2/131]) and Ctenophthalmus baeticus (0.8% [1/131]). Further, 23.7% (31/131) of the wildcats harboured mites, identified as Trombicula autumnalis in 12.2% (16/131) and Otodectes cynotis in 4.8% (6/124) of cases. The only louse species detected was Felicola hercynianus with a prevalence of 2.3% (3/131). Infestation intensities ranged from 1 to 86 ticks, 1–49 fleas, 1–1896 mites, and 1–92 F. hercynianus per wildcat. This study demonstrates that a variety of ectoparasites infests wildcats in Germany, but they do not seem to have a serious impact on the general health of wildcats, as judged by the hosts' mostly good or very good nutritional condition. In addition, the potential risk to domestic cats (Felis catus) and humans posed by the wildcats’ ectoparasites, appears to be low but present.


Introduction
Parasitism in wildlife is common, whereby the parasite benefits most when the host is weakened but not immobilised or killed, so that effective reproduction and spread of the parasite can take place (Ewald, 1983).However, if such an optimal virulence level is not maintained, excessive ectoparasite infestation may have detrimental effects on a population's health, either directly by causing e.g.dermatitis, inflammatory skin reactions, secondary bacterial infections, and anaemia, or indirectly via transmission of vector-borne pathogens (Deplazes et al., 2020;Wall and Shearer, 2008).
The European wildcat (Felis silvestris; hereafter wildcat) population was close to being eradicated mainly from hunting and trapping in the early 20th century and habitat fragmentation in the late 20th century, after it was widespread throughout Germany in former times (Piechocki, 1990).Today, wildcats are considered as Threatened according to the German Red List (Meinig et al., 2020) and are still strictly protected (Bundesnaturschutzgesetz, 2009;EEC, 1992).Due to various species conservation projects, e.g.those of the German Federation for the Environment and Nature Conservation, Friends of the Earth Germany (Bund fuer Umwelt und Naturschutz Deutschland, BUND), the population's recovery in central and south western parts of Germany was promoted, with an estimated population size of 5000 to 7000 individuals in 2017 (Balzer et al., 2018).
In addition to potential harm from ectoparasites, wild animals, such as wildcats, may serve as natural pathogen reservoirs and can be a source of spillover events to domestic animals and humans (Thompson et al., 2010).Therefore, parasite surveillance is required to evaluate the necessity for further conservation measures and to identify possible health risks for domestic animals in terms of the One Health approach.So far, little is known about the relevance of ectoparasite infestation of the wildcat.
The aim of this study was to provide an overview on ectoparasite species, their prevalence and infestation intensity in wildcats in Germany, to aid in elucidating whether wildcats could serve as an ectoparasite reservoir for domestic cats with potential risks of spillover events.

Origin of wildcat samples
The hides of 131 wildcats found dead from 2018 to 2020 in the German federal state Rhineland-Palatinate (Fig. 1) were available from the project "Monitoring of dead wildcats in Rhineland-Palatinate (Totfundmonitoring Wildkatze in Rheinland-Pfalz)" of the German Federation for the Environment and Nature Conservation, Friends of the Earth Germany, state association of Rhineland-Palatinate (Bund fuer Umwelt und Naturschutz Deutschland (BUND), Landesverband Rheinland-Pfalz).Specimens were dissected by the BUND and the cooperating institutions Will and Liselott Masgeik Foundation, OEKOLOG field research and the Clinic for Birds, Reptiles, Amphibians and Fish, Justus-Liebig-University Giessen (Leonhardt et al., 2021).Parameters such as hybrid status, sex, age, nutritional condition, and state of decomposition were determined during dissections as previously described by Leonhardt et al. (2021).The classification of the nutritional condition was carried out on the basis of subcutaneous, visceral, coronary and kidney fat deposits as described in Bisterfeld et al. (2022), resulting in the categories "very good", "good", "moderate", "bad" and "very bad/cachectic".The classification of the state of decomposition ("fresh", "moderate fresh", "moderate rotten", "proceeded rotten") was carried out as part of the BUND project (Leonhardt et al., 2021) as described by Eskens et al. (2016).Skin with fur was collected from the wildcats and stored at − 20 • C. According to the recommendations of OIE (International Office of Epizootics, 2008), hides were frozen at − 80 • C for at least 48 h before further processing to prevent potential transmission of zoonotic agents, especially Echinococcus multilocularis.

Examination of fur and ear canals and morphological identification of ectoparasites
The condition of the skin and fur of each specimen was evaluated regarding missing, wet, bloody, sticky or rotten parts.Subsequently, hides were macroscopically examined for ectoparasites using a flea comb, after we determined that the undercoat was too dense by parting the coat with the thumb or fingers.Cerumen was scraped out of the ear canals and microscopically examined at 100x magnification (Primostar 1; Carl Zeiss Microscopy Deutschland GmbH).If at least one mite was present, additional cerumen was collected from both ear canals and heated in 10% KOH (Carl Roth GmbH + Co. KG, Karlsruhe, Germany) for approximately 1 min until boiling for a few seconds using a microwave.Parasites were counted and microscopically identified to family, genus, or species level by morphological keys (Beck and Prosl, 2010;Biebel, 2007;Boch et al., 2006;Deplazes et al., 2020;Emerson and Price, 1981;Estrada-Peña et al., 2018;Kéler, 1957;Leone et al., 2013; Peus, 1938; Pratt and Wiseman, 1962;Prosl et al., 2004;Smit, 1966;Wall and Shearer, 2008).Due to difficulties in morphologically differentiating of I. hexagonus from I. canisuga larvae and nymphs, they were summarised as I. hexagonus/canisuga.

Molecular identification of ectoparasite species
Morphological species identification was confirmed by molecular analysis of at least one specimen of each detected species, except for T. autumnalis due to a lack of reference sequences available in National Center for Biotechnology Information (NCBI) GenBank.There was no reference sequence for F. hercynianus either, but sequencing was nevertheless carried out to generate the first sequences, and to compare them with available sequences of F. subrostratus.Due to distinct morphologies, adult ticks were not molecularly identified.

Statistical analyses
Statistical analyses were performed using R version 4.1.0(R Core Team, 2021).The prevalence of the different stages (adults, nymphs, and larvae) of I. ricinus and I. hexagonus/canisuga was compared within and between the two tick species by χ 2 tests, followed by Bonferroni-Holm correction of P-values to account for multiple comparisons.To test for negative or positive associations between ectoparasite orders, mathematically expected coinfestation frequencies were calculated by multiplication of the respective prevalence values, which were compared to the observed coinfestation frequencies by χ 2 tests, or Fisher's Exact tests in case of values ≤ 5, respectively.
Moreover, the influence of the predictor variables 'season of finding' and 'state of decomposition' on the prevalence of I. ricinus and I. hexagonus/canisuga was calculated using generalised linear models (GLMs) with binominal error structure.Multiple comparisons were performed for the factor 'season of finding' using Tukey contrasts with single-step P-value adjustment.For both models, a likelihood ratio test was carried out for comparison with a null model that contained only the intercept.Due to missing data for three wildcats, a subset of 128 animals was included in the GLM calculations.
Because calculation of meaningful multivariate models was not possible for the remaining parasites due to low prevalence, seasonal comparisons for total ectoparasites, ticks, fleas, T. autumnalis, O. cynotis, and F. hercynianus were conducted using Fisher's Exact tests in case of values up to five, or χ 2 tests in case of values above five, respectively, and subsequent Bonferroni-Holm correction of P-values.

Key data of the wildcats
Of 131 wildcats, 48.1% came from Palatinate, 21.4% from Hunsrueck, 12.2% from Westerwald, 10.7% from Eifel, and 3.1% from Taunus (Fig. 1).Two of the animals were suspected to be hybrids between wildcat and domestic cat, the others were morphologically or genetically confirmed as wildcats as previously reported (Bisterfeld et al., 2022;von Thaden et al., 2020).General data of the examined wildcats, e.g.sex, age, nutritional condition, and state of composition are shown in Table 2. Data on sex and/or nutritional status were not available for all cats which were in an advanced state of decomposition or when parts of the carcasses were missing (Leonhardt et al., 2021).There were slightly more males (51.9%) than females (39.7%).Due to advanced decomposition or destruction of the carcasses, sex determination was not possible for the remaining 8.4% of wildcats.The largest sample of wildcats was adults (45.0%).Most animals were in good or very good nutritional condition (57.3%), were mainly found in the autumn months of September, October, and November (43.5%) and almost half of all carcasses were in a moderate fresh state (48.9%).The majority of animals (93.9%) died as a result of trauma, mostly caused by road traffic.
A total of nine different flea species were detected.Two mite species, T. autumnalis, and O. cynotis, were identified on the wildcats.Trombicula autumnalis was found on 12.2% [16/131]) of the animals.They were located in the ear canals (75.0 % [12/16]) as well as on the outer body surface (31.3% [5/16]), namely in the Henry's pocket (25.0%[4/16]) and the distal outer surface of the auricles (6.3% [1/ 16]).Interestingly, T. autumnalis was only located on the outer body surface on animals found in September.Otodectes cynotis was detected on 4.8% (6/124) of the wildcats and was located exclusively in the ear canals.For 11 wildcats, one mite each could not be identified or rediscovered after KOH processing.
Felicola hercynianus was the only louse species detected and was present on 2.3% (3/131) of the animals.

Discussion
Although collection of data on ectoparasite infestation from dead animals has proven to be difficult, 15 different ectoparasite species were detected in this study on German wildcats.Studies on ectoparasite infestation of dead found animals are rarely performed due to the fact that ectoparasites can leave their dead host (Leple, 2001;Lledo et al., 2015) and that the sample quality can be poor due to decay, destroyed, wet/bloody carcasses, or missing body parts.While there is comparatively many data available on endoparasites of wildcats from different European countries, e.g.Greece, Croatia, Germany, Romania, and Italy (Deak et al., 2022;Diakou et al., 2021;Krone et al., 2008;Martinković et al., 2017;Napoli et al., 2016;Steeb, 2015), fewer and less recent studies on ectoparasites, mostly limited to Slovenia, Spain, and France, have been published (Brglez and Železnik, 1976;Dominguez, 2004;Leple, 2001).To the best of our knowledge, this is the first study to determine the prevalence of ectoparasites on a large sample size of more than 130 wildcats in Germany.
Most ectoparasite species of wildcats could be identified successfully by morphological characteristics or genetic analyses.Although two genes (ITS-1 and COX-2) of the discovered C. baeticus specimen were amplified and numerous reference sequences of Ctenophthalmus spp.are available in GenBank, the subspecies could not be conclusively determined.As also reported by others, the ITS-1 region is not suitable for Ctenophthalmus species identification (Zurita and Cutillas, 2021) and the COX-2 sequence matched both C. baeticus arvernus and C. baeticus boisseauorum.Hence, the specimen is therefore most likely C. baeticus, but the subspecies could not be determined.
Unfortunately, there were no reference sequences for F. hercynianus available in NCBI GenBank.Nevertheless, the lice could be identified with certainty, as F. hercynianus can be easily delineated from other louse species that occur on small felids in Europe by its well described morphological characteristics (Kéler, 1957;Martín Mateo, 2009;Mey, 1988).Nevertheless, its COX-1 gene was sequenced so that the database could be supplemented.
In this study, ticks were found on 72.5% of the wildcats with I. ricinus being more prevalent (49.6%) than I. hexagonus/canisuga (36.6%).In France, the overall prevalence of ticks was lower with eight of 39 (20.5%)Ixodes spp.-infested wildcats (Leple, 2001).The distribution pattern of I. ricinus stages showed a significantly higher frequency of adult than immature ticks.This is in accordance with a study from Thuringia, where 1286 fox carcasses were examined for ticks.The majority of adult ticks (82.2% [3711/4513]) was identified as I. ricinus while adult I. canisuga (10.8% [486/4513]) and I. hexagonus (6.7% [303/4513]) were less common (Meyer-Kayser et al., 2012).Immature stages of I. hexagonus/canisuga, which were summarised due to difficulties in the morphological differentiation of larvae and nymphs, were significantly more common on wildcats than adult ticks, although a stage-dependent difference in host preference is rarely described for these species (Arthur, 1953;Estrada-Peña et al., 2018;Hillyard, 1996).In the fox study, 71.7% (2001/2790) of the I. canisuga and I. hexagonus were nymphs and only 28.3% (789/2790) were adults (Meyer-Kayser et al., 2012).An explanation might be that I. hexagonus and I. canisuga are nest-dwelling ticks with all stages parasitising the same host (Estrada-Peña et al., 2018).Since the population density of immature stages is higher than that of adult ticks, a higher infestation rate with immature stages is likely.
Fleas were detected less frequently (27.5%) than ticks, presumably because they are more mobile and can therefore quickly leave a dead host (Hsu and Wu, 2001).In Spain, the flea species C. sciurorum, S. cuniculi, C. felis, Ctenophthalmus spp., and P. s. spectabilis, were detected on four of six (66.7%) wildcats (Dominguez, 2004).With nine different flea species, the diversity of fleas in the present study was considerably higher.Many of these fleas are associated with prey species of wildcats or can be present in burrows of these prey animals.Ceratophyllus sciurorum mainly parasitises red squirrels (Sciurus vulgaris), but also small rodents such as dormice.Besides humans, P. irritans prefers animals that live in large burrows such as badgers and foxes.Rabbits are the main hosts of S. cuniculi, but they can also be found on domestic cats.
Chaetopsylla globiceps infest mainly foxes, while C. trichosa is found on both foxes and badgers.The main hosts for the other flea species that were found, namely N. fasciatus, C. felis, A. erinacei, and C. baeticus, are rats (Rattus spp.), domestic cats, hedgehogs, and rodents, respectively (Smit, 1966).Wildcats use e.g.hollow tree trunks and brushwood piles, but also fox and badger burrows as nursery sites (Piechocki, 1990).In Germany they prey mainly on small rodents and less frequently on hares and shrews (Soricidae) (Lang, 2016).Accordingly, this lifestyle creates interfaces of wildcats with many of the fleas' main hosts or their nests, making transmission to the wildcat likely.Piechocki (1990) mentioned the 'cat flea' C. felis as a common parasite of wildcats.However, in the present study, only two specimens were detected on one wildcat each.This is probably caused by the fact that C. felis is highly temperature-sensitive with most rapid development in warm indoor environments (Deplazes et al., 2020).
Trombicula autumnalis was detected on more than one-tenth (12.2%) of the wildcats.Since the ear canals were absent in seven animals, this prevalence might be slightly underestimated.Trombicula autumnalis is the most common chigger mite in Europe, occurring on mammals, including humans, birds, and reptiles (Bowman et al., 2002;Giannoulopoulos et al., 2012;Scholer et al., 2006).In the wildcats, T. autumnalis was predominantly located in the ear canals, but the Henry's pocket was also frequently infested.In domestic cats, T. autumnalis is also often located in the Henry's pocket, but also infests other areas of the head such as the chin, temples, lips, eyelids, and the body, e.g.interdigital areas, abdomen as well as the perianal region (Bowman et al., 2002;Leone et al., 2013).However, T. autumnalis was only found on the head  of the wildcats, possibly due to the long and dense fur on the rest of the body that hampered examination.Otodectes cynotis was detected less frequently than T. autumnalis, namely on 4.8% of the wildcats.The ear mite frequently infests pets, thus being the most common cause of otitis externa in domestic animals (Bowman et al., 2002;Fanelli et al., 2020).
Ixodes ricinus occurred more frequently in spring and winter than in autumn.In contrast, on foxes from Thuringia, I. ricinus nymphs and adults were detected significantly more frequently in the warm (April to September) than in the cold (October to March) months (Meyer-Kayser et al., 2012).Interestingly, the highest infestation frequencies of I. ricinus and of ticks in general were noted on wildcats collected in winter (66.7% and 83.3%, respectively) and spring (65.5% and 82.8%, respectively), whereas activity of ticks is reported to be rather low in winter (Gethmann et al., 2020;Hagedorn, 2013;Schulz et al., 2014).Fleas were also detected most frequently in winter (54.2%) and significantly less frequently in autumn (21.1%).Possibly, ticks and fleas had lower movement activity due to cold temperatures during the winter months and were therefore not able to leave their dead host.Moreover, the carcasses were better preserved in winter with only 8.3% (2/24) being in a moderately rotten state.Therefore, the ticks and fleas might have been easier to find.
Trombicula autumnalis mites are only parasitic during their larval stage.They occur seasonally with high abundance in late summer and autumn (Wall and Shearer, 2008).Accordingly, 62.5% of the 16   wildcats infested with T. autumnalis were found in autumn in the present study but seasonal differences were not statistically significant.Infestations of domestic cats with these mites occur mainly in autumn, but cases during the rest of the year are also described (Leone et al., 2013).
For O. cynotis and F. hercynianus, no seasonal patterns in infestation frequency were detected.Besides prevalence, the infestation intensity is an important parameter in the assessment of ectoparasite infestation because high intensities can have detrimental effects on wildlife population health.In general, most of the wildcats were only mildly infested with few specimens of each parasite species, although high infestation intensities were noted in single individuals.Moreover, the majority of the animals were in good or very good nutritional condition, indicating that the wildcat population appears to be generally healthy.
Low infestation intensities in this and also other studies of wildlife populations are likely related to the fact that dead found animals were examined.Ectoparasites can leave their dead host, so intensities on living wildcats are likely higher.Hence, our intensities should be treated with caution.Because comparative data are not available and no further examinations could be performed, relating infestation intensities or coinfestations to the health status of the wildcats is difficult.Nevertheless, in the present study, infestation intensities on some of the wildcats were elevated compared to those of other specimens.For instance, one wildcat was infested with 86 ticks.On another wildcat, 48 S. cuniculi were detected, whereas the intensity of other flea infestations was limited to few (1-7) specimens.The observed S. cuniculi infestation intensities, combined with the frequent observation of S. cuniculi on wildcats (Dominguez, 2004;Piechocki, 1990;Steeb, 2015), suggests that wildcats are suitable hosts for the rabbit flea, whereas other less common fleas such as N. fasciatus and C. baeticus probably represent incidental infestations via prey.Both ticks and fleas may cause anaemia in young or weakened animals in heavy infestations (Anderson, 2000;Deplazes et al., 2020).
Unfortunately, comparative data on infestation intensities of miteand lice-infested wildcats do not exist.The infestation intensity with O. cynotis was highly variable with 2-1896 mites found per wildcat.Regarding domestic cats, the intensity of infestation is similarly variable with one case reporting 8500 mites in a single ear (Bowman et al., 2002;Preisler, 1985).Otodectes cynotis are distributed worldwide, infest various carnivores such as cats, lynx, foxes, and martens and can cause ear mange, otitis externa, itching, and head shaking (Deplazes et al., 2020;Sotiraki et al., 2001;Wall and Shearer, 2008).In the present study, it was noticeable that increased and coffee ground-like dark cerumen only appeared in the five wildcats infested with 288 or more specimens of O. cynotis but was absent in the wildcat with only two detected O. cynotis mites or those carrying T. autumnalis.However, the number of only six cases does not provide representative results.Furthermore, at least in domestic cats, the amount of discharge does not determine the clinical signs (Sotiraki et al., 2001), so no clinical consequences of O. cynotis infestation can be inferred from the data collected in this study.
Larval stages of T. autumnalis are not host specific and infest e.g.small mammals and pets such as cats and dogs.Typical symptoms caused by infestation with T. autumnalis are itching, pruritus and, as detected in the present study, orange crumbs on the skin, but in many cases, infestations are also asymptomatic (Deplazes et al., 2020;Leone et al., 2013;Wall and Shearer, 2008).
Concerning lice, pediculosis in cats normally does not affect healthy animals.In debilitated or older animals with diminished grooming behaviour, the number of chewing lice may increase, resulting in possible clinical signs such as itching, dull, scaly fur, and alopecia (Bowman et al., 2002;Deplazes et al., 2020;Wall and Shearer, 2008).The at least 48-month-old wildcat infested with 92 F. hercynianus in the present study was in a cachectic nutritional condition, suggesting that the lice were able to multiply on the weakened cat.
Because ectoparasites on wildcats can also be harmful to domestic animals, the question arises as to how high the probability may be for transmission from wild to domestic animals and humans.The risk of infestation of domestic cats with F. hercynianus is negligible, because F. hercynianus seems to solely infest wildcats, whereas domestic cats are usually parasitised by F. subrostratus (Bowman et al., 2002;Deplazes et al., 2020;Kéler, 1957;Wall and Shearer, 2008).Flea species that commonly infest wildcats, e.g. C. sciurorum, P. irritans, and S. cuniculi are rare on domestic cats and vice versa (Deplazes et al., 2020;Visser et al., 2001).Transmission of O. cynotis requires direct contact (Wall and Shearer, 2008) but low rates of hybridisation (3-5%) between wild and domestic cats in Germany (Tiesmeyer et al., 2020) indicate that direct contact between the two species is rather rare, wherefore transmission of O. cynotis to domestic cats seems to be negligible.

Conclusions
A broad spectrum of 15 different ectoparasite species, namely three tick, nine flea, two mite, and one chewing louse species, were present on more than 80% of the wildcats.While more than two-thirds of the wildcats harboured ticks and about one-fourth showed infestations with fleas and mites, chewing lice were only occasionally detected.Although the influence of parasites on the health of dead found hosts is difficult to assess, the studied wildcat population seems to be healthy based on their mostly good or very good nutritional condition, whereby ectoparasite infestations appear to play only a minor role in the overall health of the German wildcat population.Nevertheless, individual animals, especially those that are young, old, or weakened, may be harmed by direct or indirect mechanisms, such as blood loss, skin irritation, and transmission of vector-borne diseases.

Fig. 1 .
Fig. 1.Map of the German state Rhineland-Palatinate showing the sampling locations of dead found wildcats (orange dots).The scale bar represents 50 km and the arrow indicates a northern orientation.The map was modified from Bisterfeld et al. (2022) using QGIS version 3.22.2(QGIS Development Team, 2021).The map of Germany was obtained from GeoBasis-DE/BKG (Geo-Basis-DE / Bundesamt für Kartographie und Geodäsie, 2022).(For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5 .
Fig. 5.Total ectoparasite prevalence as well as prevalence of ticks, fleas, Trombicula autumnalis, Otodectes cynotis, and Felicola hercynianus on Felis silvestris from Germany for each season.Values for O. cynotis are printed in bold.The asterisk indicates a significant difference after Bonferroni-Holm correction of P-values (*P ≤ 0.05).

Table 3
Prevalence and infection intensities of ectoparasites in European wildcats from Germany (n = 131).

Table 4
Single and coinfestations of ectoparasites in European wildcats from Germany.

Table 5
Results of GLMs testing the influence of different predictor variables on the prevalence of Ixodes ricinus and Ixodes hexagonus/canisuga.