Biomolecular Investigation of Bartonella spp. in Wild Rodents of Two Swiss Regions

Rodents represent a natural reservoir of several Bartonella species, including zoonotic ones. In this study, small wild rodents, collected from two sites in rural areas of Switzerland, were screened for Bartonella spp. using molecular detection methods. In brief, 346 rodents were trapped in two rural sites in the Gantrisch Nature Park of Switzerland (Plasselb, canton of Fribourg, and Riggisberg, canton of Bern). Pools of DNA originating from three animals were tested through a qPCR screening and an end-point PCR, amplifying the 16S-23S rRNA gene intergenic transcribed spacer region and citrate synthase (gltA) loci, respectively. Subsequently, DNA was extracted from spleen samples belonging to single animals of gltA positive pools, and gltA and RNA polymerase subunit beta (rpoB) were detected by end-point PCR. Based on PCR results and sequencing, the prevalence of infection with Bartonella spp. in captured rodents, was 21.10% (73/346): 31.78% in Apodemus sp. (41/129), 10.47% in Arvicola scherman (9/86), 17.05% in Myodes glareolus (22/129), and 50% in Microtus agrestis (1/2). A significant association was observed between Bartonella spp. infection and rodent species (p < 0.01) and between trapping regions and positivity to Bartonella spp. infection (p < 0.001). Similarly, prevalence of Bartonella DNA was higher (p < 0.001) in rodents trapped in woodland areas (66/257, 25.68%) compared to those captured in open fields (9/89, 10.11%). Sequencing and phylogenetic analysis demonstrated that the extracted Bartonella DNA belonged mainly to B. taylorii and also to Candidatus “Bartonella rudakovii”, B. grahamii, B. doshiae, and B. birtlesii. In conclusion, the present study could rise public health issues regarding Bartonella infection in rodents in Switzerland.


Introduction
Bartonella species are Gram-negative, facultative intracellular and emerging zoonotic bacteria infecting both domestic and wild mammals [1]. Rodents are probably the most common wildlife host of Bartonella, and some rodent-associated Bartonella spp. may induce infections in humans [2].
Bartonella spp. are widespread worldwide, but the prevalence is higher in areas where the climatic conditions favor spreading of the arthropod vectors [3]. Although ectoparasites (ticks, fleas, and mites) are the principal vectors allowing Bartonella spp. transmission among animal hosts, the ecology of these bacteria is more complex and still not well understood [4].
Human Bartonellosis can manifest with various clinical signs that are often correlated with the immune status of the subject and, obviously, with the species and bacterial load of Bartonella that infect the host [5].
Some Bartonella species lead well-known human diseases, such as B. henselae, responsible for cat scratch disease, and B. quintana, causative agent of trench fever. Others are associated with different clinical conditions such as weight loss, muscle fatigue, and neurological manifestations [4] as well as emerging diseases, including endocarditis [6], chronic lymphadenopathy, bacillary angiomatosis and peliosis, uveitis, and vasculitis [7]. Bartonella infection often leads to febrile illnesses and the clinical condition may be similar to those triggered by other pathogens (e.g., Borrelia spp.) [8]. This suggests that the diseases associated to Bartonella could be under-estimated. Recently, the number of newly detected Bartonella species increased significantly and, to date, 45 different species have been isolated [9]. These were identified in humans, domestic [10][11][12] and wild animals, including bats [13], deer [14], marine mammals [15], rodents [16], and sheep [17]. Molecular evidence of Bartonella spp. was reported also in some migratory bird species and sea turtles [18,19].
Rodents represent a natural reservoir of several Bartonella species, and different Bartonella spp. could infect numerous rodent species with various prevalence worldwide [20]. B. tribocorum and B. elizabethae often associated to human bartonellosis [21,22] were identified in rats. Also B. henselae has been identified in wild rodents, such as Rattus rattus from New Zealand [23], Apodemus spp. in Denmark [24], and in the Pianosa Island, Italy [25].
Bartonella spp. are slow-growing microorganisms. They need complex media as Bartonella-Alphaproteobacteria growth medium based on an insect growth medium and culture conditions such as 5% CO 2 , water-saturated atmosphere [26][27][28]. Moreover they are often weak reactors to many biochemical tests [29]. These characteristics hinder their isolation and identification at species level, therefore several molecular detection methods based on specific loci have been designed for the identification of Bartonella [28,30].
In this study, small wild rodents collected from two sites in rural areas of Switzerland were screened for Bartonella spp. using molecular detection methods.

Results
A total of 84/116 DNA pools (72.4%) yielded Cq values lower than 35 in the qPCR analysis for Bartonella spp. 16S-23S rRNA intergenic transcribed spacer (ITS). R 2 , slope, primer efficiency, and Cq mean values of qPCR were 0.998, −3.386, 97.4%, and 25.16, respectively. The subsequent conventional PCR (cPCR) identified 43 out of these 84 pools (51.2%) as also positive for the Bartonella spp. citrate synthase (gltA) locus. DNA from spleen of single animals belonging to gltA positive pools were then extracted, for a total of 129/346 animals. Seventy-three (56.6%) and 64 (49.6%) out of these 129 samples showed amplicons consistent with gltA and RNA polymerase subunit beta (rpoB) loci, respectively (Table S1). Based on cPCR results ( Figure S1) and following sequencing, the prevalence of infection with Bartonella spp. in captured rodents, was 73/346 (21.10%) and a significant association (p < 0.01) was observed between Bartonella spp. infection and rodent species. Prevalence recorded for Bartonella spp. in Apodemus sp. was 31.78% (41/129): 39/129 animals tested were positive for both gltA and rpoB loci by cPCR and 2/129 positive only for rpoB locus. Sequencing confirmed the results. Bartonella DNA was detected in 9/86 (10.47%) samples of Arvicola scherman, and in two specimens only gltA was amplified. gltA and rpoB loci were detected by cPCR in 24 out of 129 Myodes glareolus, but two of them were not confirmed by sequencing: therefore, the prevalence of Bartonella spp. in this rodent species was 17.05% (22/129); in details, 3/22 animals were positive only to gltA. Moreover, gltA locus amplification was observed in 1/2 (50%) samples of Microtus agrestis.
Prevalence of Bartonella DNA identified in the rodents captured in the two municipalities was 30.71% (43/140) in Riggisberg (BE) and 15.05% (31/206) in Plasselb (FR), and a significant association (p < 0.001) between trapping region and positivity to Bartonella spp. was observed. Similarly, prevalence of Bartonella DNA was significantly higher (p < 0.001) in rodents trapped in woodland areas (66/257, 25.68%) compared to those captured in open fields (9/89, 10.11%). No statistically significant association between Bartonella DNA presence and gender or age of captured rodents was observed.

Discussion
In this study the prevalence and molecular diversity of Bartonella in small rodent populations from Switzerland were firstly described.
Wild rodents could be potential reservoirs causing Bartonella infections and more than 20 Bartonella species are associated with these small mammals [31]. They include some zoonotic species, such as B. elizabethae, B. grahamii, and B. vinsonii subsp. arupensis [20,32].
In central Europe, prevalences of Bartonella spp. ranging from 3.3 to 65.8% [33] have been observed in wild rodents. The prevalence of about 21% reported in our study was similar to the ones observed in Lithuania (24%) in 2013-2014 period [34] and Poland (11-48%) [20]. In particular, B. taylori-like DNA was the most common detected species in this study, followed by B. grahamii and B. birtlesii, three of the four most widespread species in European rodents [33].
Phylogenetic analysis based on gltA and rpoB loci demonstrated that in wild rodents of Switzerland multiple Bartonella species were identified and four genogroups were recognized, in particular B. grahamii, B. taylorii, B. doshiae and B. birtlesii. [20]. According to the lineages previously specified by Engel et al. [35], Bartonella species detected in the present study belong to lineage three and four.
B. taylorii was isolated for the first time by Birtles et al. [36]. DNA of B. taylorii strain Far East II was previously identified in rodents including Apodemus agrarius from Russian Far East in 2005 [37] and also in the present study. In Europe, B. taylorii was principally detected in Myodes sp. and Microtus sp. [38].
In this study, Bartonella DNA relative to B. grahamii was identified in a Myodes glareolus, and it is known that this bacteria may cause neuroretinitis in humans [39]. In one Apodemus sp. a probable co-infection of B. grahamii with B. taylorii was observed, confirming the wide range of hosts and the worldwide distribution of B. grahamii-like organisms, as described in Szewczyk et al., [33]. Buffet et al. [38] reported the presence of B. grahamii in Microtus spp. and Apodemus spp. in France.
Bartonella sequences identified in four Myodes glareolus trapped in the municipality of Riggisberg were 100% identical to Candidatus "Bartonella rudakovii" identified in 2007 in small wild mammals in Western Siberia (unpublished, GenBank: EF682090.1).
RpoB sequences of three Apodemus sp. captured in the present project were very close to B. birtlesii, isolated for the first time in small rodents in Germany and France [40].
From three animals, DNA similar to gltA of B. doshiae was identified (94.82-100% of identity). This species was reported in mice and voles in Europe [41] and in Sigmodon hispidus in the United States [42]. Vayssier-Taussat et al. [43] highlighted their novel potential zoonotic properties.
In this study, Apodemus sp. was the rodent genus more frequently affected by Bartonella sp. Paziewska et al. [44] obtained similar results, showing that A. flavicollis was the species in which Bartonella sp. was more present in Poland. Moreover, as in Poland, the present findings showed that B. taylorii was the most common Bartonella species, with a higher prevalence in Apodemus than in Myodes. Similarly, in 2019, a new study on the presence of Bartonella spp. in rodents was conducted in the Baltic region. The prevalence of Bartonella spp. was 54.8% and, in particular, A. flavicollis and M. agrestis were the most infected rodent species [45].
In this study, rodents trapped in woodland areas were more often infected with Bartonella spp. compared to those captured in open fields. A possible explanation for this phenomenon is the presence of vectors, such as fleas, which prefer wet conditions of woodland, allowing a better survival of larval stages [46]. In fact, ectoparasites are influenced by host characteristics, host environment, and season [46][47][48]. However, temperature and humidity conditions can affect the various vector species differently, therefore further investigations on parasites present in the considered geographical area would be useful. The distribution of reservoirs or arthropod vectors are likely the reasons why the prevalence of Bartonella infection varied among different rodent species or locations.

Materials and Methods
Wild rodents studied were trapped between April and November 2017 in the Gantrisch Nature Park (Switzerland) as described in Peterhans et al. [49], in the municipalities of Plasselb (canton of Fribourg) and Riggisberg (canton of Bern).
Topcat traps (Andermatt Biocontrol, Switzerland) were used in open fields, whereas live traps (Longworth, Penlon Ltd., Abingdon, UK) were placed in the woodland.
Trapped mice were visually examined and then euthanized by exposure to carbon dioxide on site. Mice were then transported refrigerated to the laboratory, where postmortem examination and sampling were performed as previous described in Peterhans et al. [49] ( Table 1). The age of Arvicola was determined by measuring the weight of dry crystalline lenses [50,51]. For the remaining species, the development of the sexual organs was used as in Beerli et al. [52]. This project was performed in accordance with the Swiss Animal Welfare Act (SR 455) and the regulations of the Cantons of Bern and Fribourg (permit number BE145/16).
Lungs, spleen, liver, mandibular, and mesenteric lymph nodes were collected from 346 animals, and tissue samples of three animals were pooled, resulting in a total of 116 pools. These were homogenized and genomic DNA was obtained using DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany). A real-time PCR was performed on the pools to detect Bartonella spp. DNA as described in Divari et al. [25]. ITS region of Bartonella spp. was amplified (about 200 bp) using published primers 321s and H493as [53], [26] and amplification was performed using the CFX ConnectTM Real-Time PCR Detection System (BioRad, Hercules, CA, USA). In brief, iTaq Universal SYBER ® Green Supermix (BioRad, Hercules, CA, USA) was used in the reaction mix and the protocol consisted of a 4-min step at 94 • C, followed by 45 cycles at 95 • C for 5 s and 60 • C for 20 s. A melting curve (from 65 • C to 95 • C) was obtained at the end of each run, to detect the PCR products dissociation. Examples of Tm values were described in Supplementary Materials (Table S2). The efficiency of qPCR was calculated on the slope of the standard curve constructed for ITS region amplification using scalar dilution of DNA from the positive control (Bartonella sp. FG4-1).
Pools showing a quantification cycle (Cq) less than 35 were further analyzed by cPCR, detecting a 340 bp segment of the gltA specific for Bartonella spp. cPCR was performed using previously described primers 443f and 781r [25,54] and a master mix (HotStarTaq; Qiagen, Hilden, Germany). The gltA gene fragment was amplified by a protocol consisting of a first step at 95 • C for 15 min, followed by 45 cycles at 95 • C for 5 s and 60 • C for 45 s.
PCR products were electrophoresed in 1.5% and 2% agarose gels, stained by GelRed Nucleic Acid Gel Stain (Biotium Inc., Fremont, CA, USA) and visualized under UV light. Bartonella spp. FG4-1 DNA (NCBI: txid545598) was used as a positive control and nucleasefree water included as a negative control in each PCR run. Pools showing amplicons consistent in size with the amplified locus were considered positive to Bartonella spp. infection. Therefore, DNA from spleen of single animals belonging to positive pools was extracted using DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany) and retested for gltA (as described above) and rpoB loci. For this latest gene, an 800 bp segment was amplified using previously described primers 1400f and 2300r [23,55] and amplification was conducted under the following conditions: 95 • C for 15 min, followed by 35 cycles of denaturation at 95 • C for 30 s, annealing at 53 • C, and extension at 72 • C for 1 min. At the end of the reaction, an additional extension step at 72 • C for 2 min was applied. PCR products were visualized by electrophoresis, purified through MinElute PCR purification kit (Qiagen) and sequenced in both directions using Sanger method by a commercial sequencing provider (BMR Genomics, Padova, Italy).
The raw sequences were edited using Geneious Prime version 2021.2.2 [56] and compared to sequences deposited in NCBI using BLAST (https://blast.ncbi.nlm.nih.gov/ Blast.cgi (accessed on 2 August 2021)). All sequences were deposited in the GenBank under accession numbers described in Supplementary Materials (Table S3).
GltA and rpoB sequences from this study and from GenBank were aligned (MUSCLE alignment algorithm) and phylogenetic relations were estimated through a Bayesian inference method using a GTR substitution model with the MrBayes [57] plugin in Geneious Prime version 2021.2.2 [56], with 1,100,000 chain length, 100,000 burn-in length.
Differences of Bartonella spp. prevalence among small rodent species, sampling locations, gender and age were assessed by Fisher's exact and Chi-square test and 95% confidence intervals were set. Statistical analysis was conducted using GraphPad Prism 6 version 6.07 and a p value < 0.05 was considered significant.

Conclusions
Wild rodents infected with zoonotic Bartonella species were detected in two rural areas of Switzerland. However, these regions are not distant from urban areas and, thus, contacts between humans and infected rodents are possible. Therefore, the Bartonella-infected wild rodents might represent a potential pathogen reservoir in Switzerland and should be considered of public health importance.
Supplementary Materials: The following tables are available online at https://www.mdpi.com/ article/10.3390/pathogens10101331/s1, Table S1: Sequencing results of Bartonella spp. positive rodents. Table S2: examples of Tm values calculated for DNA of samples analysed by qPCR. Table  S3: Accession number of sequences deposited in GenBank. Figure S1: examples of cPCR products, relative to gltA and rpoB loci amplification.

Informed Consent Statement: Not applicable.
Data Availability Statement: The sequences obtained in the current study were submitted to Gen-Bank under the accession numbers MZ680369 to MZ680501.