White-Toothed Shrews (Genus Crocidura): Potential Reservoirs for Zoonotic Leptospira spp. and Arthropod-Borne Pathogens?

Three species of white-toothed shrews of the order Eulipotyphla are present in central Europe: the bicolored (Crocidura leucodon), greater (Crocidura russula) and lesser (Crocidura suaveolens) white-toothed shrews. Their precise distribution in Germany is ill-defined and little is known about them as reservoirs for zoonotic pathogens (Leptospira spp., Coxiella burnetii, Brucella spp., Anaplasma phagocytophilum, Babesia spp., Neoehrlichia mikurensis and Bartonella spp.). We investigated 372 Crocidura spp. from Germany (n = 341), Austria (n = 18), Luxembourg (n = 2) and Slovakia (n = 11). West European hedgehogs (Erinaceus europaeus) were added to compare the presence of pathogens in co-occurring insectivores. Crocidura russula were distributed mainly in western and C. suaveolens mainly in north-eastern Germany. Crocidura leucodon occurred in overlapping ranges with the other shrews. Leptospira spp. DNA was detected in 28/227 C. russula and 2/78 C. leucodon samples. Further characterization revealed that Leptospira kirschneri had a sequence type (ST) 100. Neoehrlichia mikurensis DNA was detected in spleen tissue from 2/213 C. russula samples. Hedgehogs carried DNA from L. kirschneri (ST 100), L. interrogans (ST 24), A. phagocytophilum and two Bartonella species. This study improves the knowledge of the current distribution of Crocidura shrews and identifies C. russula as carrier of Leptospira kirschneri. However, shrews seem to play little-to-no role in the circulation of the arthropod-borne pathogens investigated.

in shrews was described for Sorex spp. in Germany [37]. A newly described Bartonella strain, named Bartonella florenciae, was previously isolated from the spleen tissue of a C. russula from France [38,39].
The causative agent of "Q-fever", Coxiella burnetii, is a globally distributed Gramnegative bacterium that causes infertility and abortions, mainly in ruminants (cattle, goats and sheep), and is excreted in great numbers with birth materials and, to a lesser extent in milk, faeces and urine. Farmers, veterinarians and abattoir employees are high-risk groups for infection. Numbers on reported human infections have fluctuated between 55 and 416 cases per year in Germany since 2001 [40]. Ticks (in Germany, supposedly Dermacentor marginatus) can shed C. burnetii in their faeces and transmission could potentially occur through inhalation of faecal dust rather than by the tick bite [41]. There is only limited information about the role of small mammals in the infection cycle of C. burnetii. A seroprevalence of 19% was previously reported for rodents in the UK [42,43]. In the vicinity of Q-fever-positive farms, seroprevalences of up to 53% in wild rats have been observed [44]. Conversely, a study on small mammals from Slovakia reported a seroprevalence of only 2.2%, while investigated Sorex spp. had no antibodies against C. burnetii [45].
Brucella spp. are facultative intracellular bacteria that cause brucellosis, a severe disease in animals (reproductive failure and abortion) and humans (feverish multi-organ failure). Germany is considered to be free of bovine, ovine and caprine brucellosis. To maintain this status, its potential reintroduction by wildlife should be closely monitored. However, reported human cases are increasing [46]. Several years ago, a new Brucella species, Brucella microti, was isolated from common voles in central Europe [47] and has since been detected in other wildlife [48,49]. Previous studies identified that 8% of all investigated soricine shrews [50] were Brucella spp.-positive, but so far no data are available on the presence of this pathogen in Crocidura spp. from Germany.
As data on the current distribution of greater, lesser and bicolored white-toothed shrews in Germany are incomprehensive and knowledge on their role as carriers for pathogens with zoonotic potential is limited, the objectives for this study were to (i) contribute to the current knowledge on the distribution of white-toothed shrews in Germany, (ii) detect and characterise Leptospira spp. in white-toothed shrews and (iii) evaluate whitetoothed shrews as reservoirs for arthropod-borne pathogens and compare the findings to European hedgehogs.

Collection and Dissection of Shrews and Hedgehogs
Shrews from Germany, Luxembourg, Austria and Slovakia were collected between 1999 and 2021 ( Figure 1, Table S1). The majority of these originated from a citizen-sciencebased project, where the public was asked to send in shrews trapped by cats or found dead. Additionally, shrews were trapped as by-catch during various rodent monitoring studies and pest control measures in Germany [32,51]. European hedgehogs were collected at a rescue center in Offenbach, Germany. Information on collection date and site were recorded; the latter was defined by common postal code as it was the most precise information available for specimens from prey of cats. All animals were transported on dry ice to the laboratory and stored at −20 • C until further processing. Kidney and spleen tissues were taken during a standardised necropsy procedure [52] and stored at −20 • C. Morphological metadata on body weight and sex were taken during necropsy (Table S2).

Nucleic Acid Extraction
Nucleic acids were extracted from kidney and spleen tissue using a Nucleo Mag Vet Kit (Macherey & Nagel, Düren, Germany) and a KingFisher™ Flex Purification System (Thermo Fisher Scientific, Darmstadt, Germany) according to the manufacturer's instructions.

Molecular Species Identification
Species identification for each shrew was performed based on the molecular analysis of the almost-complete cytochrome b gene and sequence comparison to GenBank entries as previously described [53].

Polymerase-Chain-Reaction-Based Screening for Leptospira spp. DNA
Kidney-derived DNA was screened in pools of two for the presence of Leptospira spp. DNA with a real-time PCR (qPCR) targeting the lipl32 gene (expected amplicon size: 242 base pairs, bp), encoding for an outer membrane lipoprotein [54]. Positive pools were retested for each individual, and samples with a cycle threshold (Ct) value below 41 were considered as Leptospira-positive. As positive control, DNA of a laboratory strain of L. kirschneri serovar Grippotyphosa was used [55]. Three C. leucodon samples were investigated previously by conventional lipl32 gene PCR [32].
PCR products were prepared with DNA Gel Loading Dye (6x) (Thermo Fisher Scientific, Darmstadt, Germany) for gel electrophoresis in 2% agarose, and gels were stained with HDGreen Plus DNA Stain (Intas Science Imaging Instruments GmbH, Göttingen, Germany). Amplification products were visualised by UV light using the UVP GelSolo streamlined gel documentation (Analytik Jena AG, Jena, Germany). The samples were purified for sequencing using a NucleoSpin Gel and PCR clean-up kit (Macherey-Nagel GmbH & Co. KG, Düren, Germany) as recommended by the manufacturer. The sequences were trimmed using Bionumerics v.7.6.1. (Applied Maths Inc., Austin, TX, USA) and compared to available data in GenBank with the Basic Local Alignment Search Tool (BLAST) (https://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 7 August 2022). The obtained sequences were uploaded to GenBank (accession numbers: OQ865429-OQ865435). The presence of Bartonella spp. was evaluated in individual spleen DNA samples by conventional PCR targeting the nicotinamide adenine dinucleotide hydrogen dehydrogenase (NADH) subunit (nuoG) with an amplicon size of 346 bp [57]. DNA from a cultured B. henselae Marseille strain was used as positive control. Positive samples were further analysed by PCR targeting the gltA gene (amplicon size: 378 bp) [57,58]. Positive samples were purified and sequenced commercially (Interdisziplinäres Zentrum für Klinische Forschung, Leipzig, Germany). The obtained sequences were uploaded to GenBank (accession numbers: OQ865426-OQ865428). Spleen-derived DNA pools of two individuals were screened with qPCRs for the presence of Anaplasma phagocytophilum DNA targeting the msp2 (major surface protein 2) gene (amplicon size: 77 bp) [59] and Neoehrlichia mikurensis DNA targeting the groEL gene (amplicon size: 99 bp) as previously described [60]. As positive controls, we used DNA from an A. phagocytophilum culture and DNA from a N. mikurensis positive yellow-necked field mouse (Apodemus flavicollis) from Leipzig, Germany, that was trapped in 2016 [61], respectively. Positive pools were retested on an individual level. Spleen DNA samples in pools of three were used for the detection of Babesia spp. DNA by conventional PCR targeting a fragment (411-452 bp) of the 18S rRNA gene [62]. For the detection of Coxiella burnetii DNA and Brucella spp. DNA, all individual spleen-derived DNA samples were screened using a qPCR targeting the multicopy insertion element IS1111 [63] or the bcsp31 gene [64], respectively.

Statistical Analysis
All statistics were performed in the GraphPad Prism Software v. 4.0 (GraphPad Software Inc., San Diego, CA, USA). Mean prevalence and confidence intervals (95% CI) for Leptospira spp. were determined using the Clopper and Pearson method with an alpha value of 0.05. For the prevalence of Leptospira and the sex of different Crocidura species, Fisher's exact test was used to test independence. Tests were considered to be significant if p (probability) < 0.05.
The shrews originated from the southwest (n = 9), northwest (n = 103), northeast (n = 110) and southeast (n = 118) of Germany ( Figure 2). Crocidura russula was the most abundant species, especially in the western parts of Germany-northwest: 99% (n = 103) and southwest: 100% (n = 9). Only one C. leucodon (1%) was collected in the southeast of Lower Saxony close to the Harz mountain range (Table S1). In the eastern half of Germany, the situation was more diverse. All three species could be found in the northeast, with 70% C. russula (n = 77), 13.6% C. leucodon (n = 15) and 16.4% C. suaveolens (n = 18). Crocidura russula was still the predominant species in northeast Germany, but it was not collected in the state of Brandenburg (BB), which is far northeast, where mainly C. suaveolens was found (78.3% of all investigated C. suaveolens). In the southeast, especially in the south of Bavaria, C. leucodon was the most prominent (56.8%, n = 67), and C. russula (39%, n = 46) was mainly found in Franconia and further north. Of all the collected white-toothed shrews from the southeast 4.2% were C. suaveolens (n = 5) (Figure 2, Table S1). The species composition varied per site. The occurrence of C. russula and C. leucodon overlapped at five sites (Figure 2), and C. leucodon and C. suaveolens overlapped at four sites. Crocidura russula and C. suaveolens were only found together at one site in the northeast of Germany. We did not find all three species at the same site. A few white-toothed shrews from neighbouring countries in central Europe were included in our study: two C. russula from Luxembourg, two C. russula and one C. leucodon from Vorarlberg, Austria, three C. leucodon and twelve C. suaveolens from the eastern state of Steiermark, Austria, and five C. leucodon and six C. suaveolens from Slovakia ( Figure 1).

Detection and Sequence Type Identification of Leptospira spp.
Leptospira spp. DNA was detected in kidney samples from 28 out of 227 C. russula (12.3%, 95% CI: 8.6-17.3) and three out of 81 C. leucodon (3.7%, 95% CI: 0.8-10.7) samples from Germany (Table 1). Table 1. Results for the detection of Leptospira spp. with lipl32-qPCR in kidney tissue and Neoehrlichia mikurensis (groEL-qPCR), Anaplasma phagocytophilum (msp2-qPCR) and Coxiella burnetii (multicopy IS1111 element-qPCR), Brucella spp. (bcsp31-qPCR) and conventional PCR results for the detection of All of the C. russula and C. leucodon samples from Luxembourg and Austria and all of the 22 C. suaveolens tested negative for the presence of Leptospira spp. DNA (0%, 95% CI: 0-17.6). Thus, the prevalence was significantly lower in C. leucodon and C. suaveolens compared to C. russula (p = 0.003). Out of the 28 lipl32 qPCR-positive C. russula, six were identified as Leptospira kirschneri by sequencing the secY PCR product. MLST was successful for an additional six individuals (C. russula) and were determined to be the same sequence type: Leptospira kirschneri ST 100. The sequencing of the secY PCR product of the lipl32 qPCR-positive C. leucodon was not possible, which was most likely due to the poor sample DNA quality. There was no significant difference in the prevalence between female (10.3%, 95% CI: 5.8-17.2) and male C. russula (14.6%, 95% CI: 8.8-23.1) (p = 0.337).

Current Distribution of White-Toothed Shrews in Germany
The collection of 341 white-toothed shrews allowed, albeit with limitations due to the heterogenous sampling, an update on the current distribution of Crocidura spp. in Germany. The latest comprehensive survey on the distribution of white-toothed shrews in Germany covered only southeast Germany (Bavaria) [65] and was mainly based on the identification of skeletal remains in owl pellets. With our citizen science project, which exploited cats' aversion to consume shrews, we were able to collect fresh carcasses to accurately identify the species using molecular techniques and to perform an initial screening of their accompanying pathogens, which allowed us to determine health risks to cats and their owners.
Over the past decades, multiple studies [9,[66][67][68][69] have monitored the distribution boundaries of white-toothed shrews on local levels [14,[70][71][72], describing fluctuations in total white-toothed shrew numbers [17] and uncertain boundaries. The core distribution range of C. russula expands from the western European countries into central Germany and is slowly expanding further east [15,73,74]. The collection of C. russula in our study in western and southeastern Germany coincided with the easternmost expansion into Franconia, Bavaria [65]. In regions where C. russula occurred, C. russula predominated over the other two species, which may have led to the local extinction of C. suaveolens as they are considered parapatric species [15,18,74]. Whether this is solely due to the size difference between the larger C. russula and the smaller C. suaveolens or due to differences in adaptations to synanthropic habitats and climate conditions, as C. russula copes better with drier, hotter summers, and therefore, out-competition is still under debate [15,18]. The same applies to C. leucodon, as C. russula was primarily found in former typical C. leucodon habitats [18,[74][75][76]. The eastwards expansion of C. russula and the replacement of C. leucodon has also been observed in Switzerland [16] and Austria [8,77]. Although limited by number, we observed the same trend with C. russula, it being found in the northwest of Austria, while in the east of Austria so far only C. leucodon and C. suaveolens were collected. We primarily detected C. suaveolens in the northeastern part of Germany, supporting the westwards expansion trend described by Jentzsch and Trost [78]. Crocidura suaveolens were sporadically found in the southeast, but not at all in the western parts of Germany. Similarly, the absence of C. leucodon from the southwest was consistent with previous reports describing a decline in C. leucodon occurrence in the western half of Germany [68,75,76]. Information on the exact origin of an individual is needed to determine territory size and sym-and parapatry, which was not possible with our sample collection as it was greatly influenced by the cats' behaviour. We decided to use postal codes as the smallest common spatial factor. All three species were not found together, but the co-occurrence of C. leucodon and C. russula versus C. leucodon and C. suaveolens was almost equally frequent (n = 5 vs. n = 4); however, C. suaveolens and C. russula were only collected together at one site in northeastern Germany. Between 1995 and 2010, the co-occurrence of all three species was described for east Thuringia [74] and west Saxony [18]. There are multiple possible explanations for the ongoing fluctuation and expansion of the species' distribution ranges, including ongoing postglacial expansion [5], man-made factors due to alterations in land use and climate [79] or simply the translocation of individuals [16]. Anthropogenic movement has a great influence in the range expansion, as shrews might be transported via feed (e.g., haystacks) or soil. Once translocated, shrews easily establish new colonies [80][81][82], as seen in the introduction of the greater white-toothed shrew to Ireland, most likely due to human activity, in the early 21st century [12]. Since then, C. russula has expanded at a pace of 15 km/year, which is much faster than described for continental Europe.
The current knowledge on Leptospira in crocidurine shrews in central Europe is scarce. Leptospira spp. was detected in C. russula already in the 1970s [83]. In Germany, Leptospira kirschneri was found in Crocidura russula [84] and Crocidura leucodon [32], but no further sequence typing was performed. Here, we detected Leptospira kirschneri in 28 C. russula and two C. leucodon with a mean prevalence of 25% (5.6-40%) at 15 trapping sites. Leptospira spp. was irregularly distributed in Germany, as demonstrated by its absence in white-toothed shrews from Saxony (this study, [85]). The irregular distribution and broad variation in the prevalence per trapping site might be caused by a biased sample size per site and the geographic origin of the samples. Water and moist areas play an important role in the maintenance and spread of Leptospira spp. outside their animal hosts [29]; crocidurine shrews prefer more open, arid habitats, which might explain the lower Leptospira spp. prevalence compared to Sorex spp. and rodents. The observed difference in prevalence between C. russula and C. leucodon could be due to the differences in habitat use between the species. Crocidura russula is a range-expanding invader [86] and may therefore have a higher exposure to Leptospira. Unfortunately, a comparison of the exact habitat use between the shrew species was not possible due to our sampling method. Although Leptospira kirschneri has been described as the most abundant genomospecies in small mammals, Jeske et al. [32] detected Leptospira borgpetersenii in sympatric rodents from trapping sites, where L. kirschneri was found in C. leucodon. Interestingly, the investigated hedgehogs carried two Leptospira species, L. kirschneri ST 100 and Leptospira interrogans ST 24, with the latter one commonly found in forest-dwelling rodents such as yellow-necked field mice and wood mice (Apodemus sylvaticus) [30].
MLST allowed us to determine the ST of Leptospira spp., and it is widely used to evaluate the spread of a specific pathogen within a population to distinguish detection in maintenance hosts from spill-over and host-switch events. In small mammal populations, different sequence types are seen within the same species and the same ST in different animal species. Common shrews from various locations in Germany have been shown to carry Leptospira kirschneri of two different sequence types (ST 110, ST 136) as well as Leptospira borgpetersenii of ST 146 [30]. Leptospira kirschneri ST 110 is strongly associated with voles of the genus Microtus and is the most common source of leptospirosis outbreaks in strawberry pickers in Germany [30]. Interestingly, we found only a single Leptospira kirschneri ST (ST 100) in all the C. russula samples from the different trapping sites across Germany, suggesting a possible host species specificity and may identify C. russula as maintenance host rather than spill-over host. However, this ST was also found in a European hedgehog (this study) and was previously isolated from a Portuguese house mouse (Mus musculus) [87]. This ST has been associated to the serovar Mozdok, a serovar that is widely distributed in small mammals (mainly Apodemus agrarius) in central Europe [88], which causes canine leptospirosis [89] and is also associated with human infections [90]. Further investigations on sympatric small mammals from the same trapping sites are needed to determine how widespread ST 100 is within the small mammal community. Unfortunately, for the publicly available ST 100 isolate (Leptospira isolate 15-LE00367-0 [91]) from Germany, the host species and its precise origin in Lower Saxony, Germany, is not specified.

Identification of White-Toothed Shrews as Reservoirs for Arthropod-Borne Pathogens
A high prevalence of tick-borne pathogens has been described for common shrews [36,37], but little is known about the prevalence of these pathogens in white-toothed shrews. A comparable study from Spain found A. phagocytophilum in one of six C. russula samples [92], whereas a previous study from Germany did not detect A. phagocytophilum, Babesia spp. and N. mikurensis in any C. russula sample [60]. Even though our sample size (n = 372) was much larger than that of previous studies (n = 4), we still did not detect A. phagocytophilum in any white-toothed shrew. Anaplasma phagocytophilum is present in the small mammal community in Germany, as confirmed here by the prevalence of about 10% in European hedgehogs (this study, [27]) and in crowned shrews and bank voles (Clethrionomys glareolus) [37]. We detected N. mikurensis DNA in two C. russula samples at different urban sites in northwestern and southeastern Germany, a finding that seems to be in contradiction to the assumption of previous studies that insectivores do not play a role in the transmission and maintenance of N. mikurensis [93]. The detection and further characterization of Bartonella spp. from soricine shrews in Germany revealed host-specific Bartonella taylorii-associated strains [37,94]. So far, Bartonella spp. have only been detected in Crocidura spp. outside of Germany [95,96], e.g., the detection of the new species Bartonella refiksaydamii in the blood of a lesser white-toothed shrew from northwestern Turkey by Celebi et al. [97]. In this study, we did not detect Bartonella spp. DNA in any of the white-toothed shrews, but we identified the Bartonella clarridgeiae strain 73 and an "uncultured Bartonella spp." in the hedgehogs. Bartonella clarridgeiae is commonly present in cats [38], is transmitted by cat fleas (Ctenocephalides felis) and was once found in an asymptomatic blood donor in Brazil [98]. The role of small mammals and shrews in particular for the transmission of Babesia spp. and Coxiella burnetii is ill-defined. In our study, we did not detect Babesia spp. DNA in any of the crocidurine shrews or hedgehogs, even though Bown et al. [36] reported a Babesia microti prevalence of 30.3% in common shrews occupying the same habitat as field voles (30.4% B. microti-prevalence). Despite reports of a high seroprevalence for C. burnetii in rodents [42], all of the insectivores tested here were negative according to the PCR analysis. Assuming that small mammals are exposed to C. burnetii, shrews and hedgehogs do not seem to play a role as reservoirs. Fleas collected from C. suaveolens were tested for the presence of C. burnetii and rickettsiae, but they did not contain any respective DNA [99]. Previous detection of Brucella spp. in soricine shrews [50] could not be demonstrated for crocidurine shrews, as all of the insectivores tested here were negative.
Little is known about ectoparasites on shrews, but a white-toothed-shrew specific "ectoparasite milieu" [99,100], reducing the possible transmission of arthropod-borne pathogens from other (small mammal) species, might be an explanation for the observed low pathogen prevalence. Even though different life stages of Ixodes ricinus and Dermacentor reticulatus could be collected from C. leucodon and C. suaveolens trapped in Slovakia, the numbers were much lower than those from sympatric rodent species [101].

Conclusions
This study provides an update on the current distribution of white-toothed shrews in Germany. Altogether, white-toothed shrews seem to play a minor role in the transmission of Leptospira spp. and arthropod-borne pathogens. However, our study was limited by its sample size and sampling approach, heavily relying on the cooperation of the public. In the future, a more systematic and longitudinal study, ideally in a One Health setting, is needed to evaluate the potential infection risks of shrews and hedgehogs. The short life expectancy and high turnover rate of local shrew populations, including frequent extinction and fast recolonization events as described for C. russula [82], potentially influencing pathogen persistence in shrew communities, should be taken into account.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/pathogens12060781/s1, Table S1: Information on the origin of investigated shrews.; Table S2: Metadata on investigated white-toothed shrews and hedgehogs. conceptualization, project administration, funding acquisition, R.G.U. and M.P. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Federal Ministry of Education and Research within the Network "Zoonotic Infectious Diseases" (ZooKoInfekt, grant numbers 01KI1903A to M.P. and 01KI1903B to R.G.U.). The collection of small mammals was funded within the projects "Long-term population dynamics of rodent hosts: Interaction of climate change, land-use and biodiversity" to J.J., "Effects of climate change on rodents, associated parasites and pathogens" and "Effectiveness and optimization of risk mitigation measures for the use of biocidal anticoagulant rodenticides with high environmental risk" to J.J. These studies were commissioned and funded by the Federal Environment Agency ( ). The majority of the small mammals originated from a citizen-science-based project (cat preys, found dead); therefore, no further permits were required.

Informed Consent Statement: Not applicable.
Data Availability Statement: All data are presented within the manuscript and its Supplementary Materials. Sequence data were uploaded to GenBank (accession numbers: OQ865426-OQ865435).