Zoonotic Bartonella species in Eurasian wolves and other free‐ranging wild mammals from Italy

Abstract Bartonellae are emerging vector‐borne pathogens infecting humans, domestic mammals and wildlife. Ninety‐seven red foxes (Vulpes vulpes), 8 European badgers (Meles meles), 6 Eurasian wolves (Canis lupus), 6 European hedgehogs (Erinaceus europaeus), 3 beech martens (Martes foina) and 2 roe deer (Capreolus capreolus) from Italian Nature Conservatory Parks were investigated for Bartonella infection. Several Bartonella species (9.84%; 95% CI: 4.55–15.12), including zoonotic ones, were molecularly detected among wolves (83.3%; 95% CI: 51–100.00), foxes (4.12%; 95% CI: 0.17–8.08), hedgehogs (33.33%; 95% CI: 0.00–71.05) and a roe deer. Bartonella rochalimae was the most common Bartonella species (i.e. in 4 foxes and 2 wolves) detected. Candidatus B. merieuxii and B. vinsonii subsp. berkhoffii were identified for the first time in wolves. Furthermore, Bartonella schoenbuchensis was identified in a roe deer and a new clone with phylogenetic proximity to B. clarridgeiae was detected in European hedgehogs. Zoonotic and other Bartonella species were significantly more frequent in Eurasian wolves (p < .0001), than in other free‐ranging wild mammals, representing a potential reservoir for infection in humans and domestic animals.


| INTRODUC TI ON
Many Bartonella species, which are facultative intracellular, fastidious, Gram-negative alpha-proteobacteria, have been described so far (Okaro et al., 2017). These bacteria are highly adapted to one or more mammalian hosts, infecting erythrocytes and endothelial cells, establishing a long-term silent infection in mammalian reservoirs, through escaping the immune response (Harms & Dehio, 2012). The spread of Bartonella spp. infection among mammalians occurs mainly via bloodsucking arthropod vectors, particularly fleas , disseminating the bacteria within specific reservoir communities and between different reservoirs. Currently, out of 40 Bartonella species/ subspecies, at least 17 are associated with an expanding spectrum of clinical signs in humans and animals (Breitschwerdt, 2014;Chomel & Kasten, 2010;Harms & Dehio, 2012;Okaro et al., 2017).
More than 60% of human pathogens are of animal origin with the majority coming from wildlife (Böhm et al., 2007). National or regional parks and protected areas offer habitats for such a diverse wild fauna, thus acting as potential reservoirs for many human and domestic animal pathogens including Bartonella spp. (Ambrogi et al., 2019;Millán et al., 2016). In Southern Italy, wild mammal populations in the National (the Cilento and Vallo di Diano) and Regional (Partenio and Monti Picentini) Parks have increased in recent decades. For instance, the Eurasian wolf (Canis lupus) population, close to extinction due to human activities until 1970s, has rebounded, with an estimated 1,500-1,800 individuals thanks to conservation policies and to an increase in its main prey species (Galaverni et al., 2015). In addition, red fox (Vulpes vulpes) is a significant free-ranging carnivore in these areas. Bartonella infection in foxes has been described worldwide (Bai et al., 2016;Chomel et al., 2012;Fleischman et al., 2015;Gerrikagoitia et al., 2012;Hodžić et al., 2018;Kosoy & Goodrich, 2019;López-Pérez et al., 2017;Marciano et al., 2016;Millán et al., 2016;Schaefer et al., 2011;Víchová et al., 2018), whereas it has rarely been reported in wolves from Spain (Gerrikagoitia et al., 2012). Since both wolf and red fox are sympatric with dogs especially feral and rural ones, they may play an important role in the Bartonella ecology (Bateman & Fleming, 2012;Gehrt et al., 2013). Finally, a number of badgers, hedgehogs, martens and roe deers are registered in the area, all of them shown to be susceptible to Bartonella spp. infection (Bitam et al., 2009(Bitam et al., , 2012Dehio et al., 2001;Gerrikagoitia et al., 2012;Harms et al., 2017;Sato et al., 2012). In Italy, B. bovis and B. chomelii are the only species reported in wildlife, from deer ticks . Notably, data on wild canid infections are lacking, although B. vinsonii subsp. berkhoffii and the uncultured Bartonella sp. strain HMD later shown to be Candidatus B. merieuxii have been detected in hunting and rural dogs in southern Italy Diniz et al., 2009). It is worth mentioning that B. vinsonii subsp. berkhoffii genotype III can be highly pathogenic for dogs (Shelnutt et al., 2017).
To gain insight for potential biologic threats for wildlife and for humans and their domestic animals within the One Health approach (Breitschwerdt, 2014), this study aimed at investigating the occurrence of Bartonella species in different free-ranging wildlife in South Italy.
The territory covers complex ecosystems and habitats allowing the presence of such diverse fauna (Guglietta et al., 2015).

| Molecular procedures
All the spleen samples were homogenized in Minimal Essential Medium (MEM, 50 mg/ml). DNA was extracted from 200 µl of homogenate using the DNeasy Tissue Kit (Qiagen, Milan, Italy) according to the manufacturer's instructions. DNA was eluted in 100 μl of AE buffer and stored at −20°C till testing. For quality assurance, a Bartonella-free spleen sample as an extraction negative control was used. DNA was carefully quantified using the fluorometric Qubit ® dsDNA HS (High Sensitivity) Assay kit, and the extracts were diluted at the final concentration of 2.5 ng/μ. Two microl were used for each qPCR and cPCR assays. Samples were screened using a Bartonella genus-specific quantitative real-time PCR (qPCR) assay targeting the transfer-mRNA ssrA (ssrA) gene as previously described (Diaz et al., 2012) (Table 2). qPCR amplification was conducted in Multiplate PCR plates (Bio_Rad ™ , Milan, Italy) using a CFX96 Touch Real-Time PCR Detection System (Bio_Rad ™ , Milan, Italy).
Maximum likelihood method with Tamura Nei 3-parameter substitution model, a proportion of invariable sites and a gamma distribution of rate variation across sites was applied supplying statistical support with subsampling over 1,000 replicates.

| Statistical analysis
An animal was considered Bartonella spp. infected if it was positive in the qPCR. Exact binomial 95% confidence intervals (CIs) were established for proportions. Proportion differences were tested for statistical significance using the Fisher's exact test/or Chi square, where appropriated. The statistical significance threshold for both tails test was set at p ≥ .05. All statistics were performed on WInepI software (http://winepi.net/; October 2020).

| RE SULTS
The concentration mean from the DNA spleen samples extracted  (Table 3). Bartonella DNA load ranging between 4.84 × 10 0 and 2.59 × 10 4 (2.27 × 10 3 mean, 5.15 × 10 1 median) DNA copies per μl. The frequencies and the spatial distribution of the detected Bartonella species are shown in Tables 1 and 3 and Figure 1.
merieuxii from Italian wolf was 99.22% similar to HMD clone detected in rural dogs in South Italy (Diniz et al., 2009) and to F040 clone  from Iraqi jackal. The ssrA sequence was 99.57% similar to the Iran-GT-3b clone from an Iranian dog and to clone Ca-1 detected in a jackal from Israel, respectively (Table 3) (Greco, Sazmand, et al., 2019;Marciano et al., 2016). No significant differences were recorded for locality (p = 1) or gender (p = .96).

| Nucleotide sequence accession numbers
The novel unique sequences obtained in the present study have  and beech martens, as previously reported in Spain (Márquez et al., 2009;Millán et al., 2016).

| D ISCUSS I ON
Bartonella infections in wild and domestic canids have been reported worldwide (Bai et al., 2016;Chomel et al., 2012;Fleischman et al., 2015;Gerrikagoitia et al., 2012;Greco, Sazmand, et al., 2019;Hodžić et al., 2018;Kosoy & Goodrich, 2019;López-Pérez et al., 2017;Marciano et al., 2016;Millán et al., 2016;Schaefer et al., 2011;Víchová et al., 2018). However, in Italy, F I G U R E 3 Phylogenetic trees displaying the diversity of Bartonella species detected in wild mammals from Italy. The phylogenetic trees for partial gltA (290 bp) (panel a), and ssrA (254 bp) (panel b) sequences from representative known Bartonella isolates were generated by using meGA-x v. 10.0.5 software. Maximum likelihood method with Tamura Nei 3-parameter substitution model, a proportion of invariable sites and a gamma distribution of rate variation across sites was applied supplying statistical support with subsampling over 1,000 replicates. GenBank accession numbers are provided for reference isolates with the sequence from Brucella abortus (AE017223) used as outgroup. The representative sequences generated in the present study are marked with arrows. Asterisks denote the nucleotide sequences identical to strains retrieved from red foxes. Scale bars indicate nucleotide substitutions per site only one seroprevalence survey in domestic dogs had reported a 6% prevalence for B. henselae infection in owned dogs in the Tuscany region (Ebani, Nardoni, et al., 2015). In addition, DNAs of B. vinsonii subsp. berkhoffii type III and of the uncultured Bartonella spp. strain HMD later shown to be Candidatus B. merieuxii were detected in rural and hunting dogs from the south and the centre of the country Diniz et al., 2009;Ebani, Nardoni, et al., 2015).
Up to date, no information was available on Bartonella infection in wild canids in Italy. DNA sequences of different Bartonella spp.
were detected in wild canids with occurrence being much higher in Eurasian wolves (>80%) than in red foxes (<5%), thus identifying this carnivore as significant reservoir for Bartonella spp. infection in southern Italy. Though the occurrence observed in the present study is probably biased because of the small wolf and fox sample sizes, it may be affected by the different behaviours of the two canids.
Indeed, the close physical contact between group members of social canids such as wolves, rather than the lonely foxes may have enhanced the likelihood of transmission of pathogens and vectors between wolves (Delamater et al., 2019 Hodžić et al., 2018;Millán et al., 2016). The discordant values of B. rochalimae occurrence observed in the two animal species overlapped previous studies in northern Spain, where prevalence of 33.3 % and 1.6% had been reported in wolves and foxes, respectively (Gerrikagoitia et al., 2012). The zoonotic B. rochalimae species has been recorded worldwide in coyotes, wolves, island foxes, grey foxes, red foxes, raccoons, skunk and domestic dogs (Fleischman et al., 2015;Gerrikagoitia et al., 2012;Greco, Sazmand, et al., 2019;López-Pérez et al., 2017;Marciano et al., 2016;Millán et al., 2016;Schaefer et al., 2011). The present study revealed the circulation of such a zoonotic agent within wild canids in regional and national parks in Southern Iran-GT-3b clones detected in rural dogs from South Italy and Iran (Diniz et al., 2009;Greco, Sazmand, et al., 2019), and with the F040 and Ca-1 clones from Iraqi and Israel jackals Marciano et al., 2016), respectively. The results of this study show that C. B. merieuxii is also infecting in Eurasian wolf, thus expanding the host range of this Bartonella species.
A female roe deer from the Regional Park of Monti Picentini in the Salerno province was infected with a B. schoenbuchensis strain matching with the R1 strain isolated from a roe deer blood sample from Germany (Dehio et al., 2001 (Bitam et al., 2009;Gehrt et al., 2013). An uncultured Bartonella spp. was detected in Southern white-breasted hedgehogs (33.3%) in Israel (Marciano et al., 2016).
In the present study, a new Bartonella clone was detected in the spleen of two (33.3%) European hedgehogs from peri-urban areas in Matera province (Basilicata region). By phylogenetic analyses of the ITS and ssrA sequences, the new detected Bartonella clone grouped in the same clade including B. clarridgeiae and B. rochalimae (García-Esteban et al., 2008;Harms et al., 2017;Sato et al., 2012). life style in south Italy (Greco, Brianti, et al., 2019;Otranto et al., 2017) further research should be conducted to investigate whether the Bartonella clone described in hedgehogs could be transmitted to cats through common vectors. These findings point to a cautious management of such animals hosting strains phylogenetically related to the zoonotic B. clarridgeiae species that is known to be a minor agent of CSD with cats and their fleas as reservoirs (Clarridge et al., 1995;Kim et al., 2009).
In conclusion, our findings stress the importance of wildlife disease surveillance, mainly for wildlife protection and as a useful and complementary component of human and domestic animal disease surveillance. Noteworthy, these findings underline the risk of exposure to Bartonella spp. infections in nature lovers, orienteering/ trekking competitors, hunters, wildlife rangers and local residents during outdoor activities along with their domestic animals, as they can easily encounter wildlife animals, arthropods and, eventually, be in contact with the pathogens they transmit.

ACK N OWLED G EM ENTS
The authors are grateful to Dr. Paolo Varuzza (University of Naples Federico II) for his technical assistance with the GIS software.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

E TH I C A L A PPROVA L
The study was conducted under the frame of a wildlife healthmonitoring plan authorized by the Regione Campania (Animal Ethics Committee, Decreto Dirigenziale 96 no. 210 -Piano B7 DPAR 2018) in order to assess the health status of domestic and wild free-ranging animals. Procedures according to the Italian Ministry of Health guidelines were strictly applied.