Introduction

The importance of ticks (Acari: Ixodida) has long been recognized due to their ability to feed on a large range of host species and to transmit microorganisms capable of infecting both human and animal hosts (de la Fuente et al. 2017; Brites-Neto et al. 2015). Identification and characterization of these circulating agents are crucial for the development of preventive measures in response to the gradually increasing exposure of humans to tick vectors. In Italy, the incidence of tick-borne rickettsioses (TBD) has increased over the last decade with 3170 clinical cases and 18 deaths documented by the Health Ministry from 2009 to 2013. The average incidence of Rickettsiosis for this period was 21.17 to 176.88 cases per million persons (http://old.iss.it/binary/publ/cont/16_1_web.pdf). Rickettsia species are the causative agents of human or animal diseases, including spotted fevers and murine or epidemic typhus which can cause a range of mild to fatal diseases, mostly through arthropod bites (El Karkouri et al. 2016).

This genus encompasses at least 27 Rickettsia species with validated and published names, and a number of putative novel Rickettsia species that have not been fully characterized, but were continually isolated from or detected in ticks (Parola et al. 2013; Merhej et al. 2014). The members of Rickettsia have been classified into four different groups, including the well-defined spotted fever group (SFG) and typhus group (TG), the Rickettsia bellii group, and the Rickettsia canadensis group (Merhej et al. 2014). Currently, several studies have revealed the extensive diversity of SFG rickettsiae in different tick species and geographic locations (Merhej et al. 2014). In Sardinia, the second largest island in the Mediterranean Sea, cases of notifiable tick-borne diseases are increasing (Madeddu et al. 2016) and Mediterranean Spotted Fever (MSF) rickettsiosis continues to be endemic with an incidence of 10/10,000 inhabitants per year (http://www.epicentro.iss.it/problemi/zecche/rickettsiosi.asp.). Members of the spotted fever group, capable of causing disease, have been detected in Sardinian ticks, such as R. aeschlimannii, R. massiliae, R. conorii israeliensis, R. slovaca, R. helvetica, and R. monacensis. In addition, two Rickettsiae of unknown pathogenicity namely Rickettsia hoogstralii and Candidatus Rickettsia barbariae have been previously identified (Chisu et al. 2018; Madeddu et al. 2016). Even if some of these spotted fever group rickettsia seem to have low pathogenicity, namely R. massiliae, R. monacensis, and R. raoultii, they were found to be involved in human disease (Parola et al. 2009; Jado et al. 2007; Vitale et al. 2006). The objective of this study was to determine the presence of Rickettsia species in ticks collected from several sites and to update the knowledge on tick-borne rickettsiae in Sardinia.

Material and methods

Tick sampling

Between March and November 2017, ticks were collected from humans, domestic mammals, and wildlife in Sardinia, Italy. Three ticks were also collected from vegetation in the areas grazed by vertebrate hosts. Domestic mammals (21 dogs, 4 goats, 7 cattle, and 1 cat) were reared in small farms, and owners provided the ticks in 15-ml tubes. Wild vertebrates sampled during the study included 1 fox (Cynotherium sardous), 3 mouflons (Ovis orientalis musimon), 7 wild boars (Sus scrofa meridionalis), 1 marten (Martes martes), 3 deer (Cervus elaphus corsicanus), and 3 birds (Corvus corone). They were brought dead to our laboratories for necropsy analyses and tick collection. All of the ticks removed from animals were adult and partially or completely engorged.

Ticks were carefully removed from their hosts by using fine-tipped tweezers and placed in vials with 70% ethanol at room temperature. Ticks were then identified to species level, developmental stage, and sex under a dissecting microscope using conventional taxonomic keys (Manilla 1998).

DNA extraction and amplification

Ticks were rinsed twice in distilled sterile water for 10 min and dried on sterile filter paper. Each sample was then longitudinally incised using an individual scalpel into two parts, where only one piece was crushed using unique scalpel in sterile tubes (Eppendorf; Hamburg, Germany). The remaining portion of each tick was kept at − 80 C for further control. All experiments and handling of blood and ticks were conducted in a laminar flow biosafety hood. A total of 1000 μL of homogenized ticks was transferred to a specific tubes prior to their extraction using QIAgen columns (QIAamp tissue kit, Qiagen, Hilden, Germany), according to the manufacturer’s instructions to obtain the final elution volume of 100 μL. A negative control was included every 10 DNA extractions to monitor the occurrence of false-positives. All biological materials including ticks DNA were stored at − 20 C until further use.

To investigate the presence of Rickettsia species in ticks, the samples were initially tested by PCR oligonucleotide primers Rp CS. 409p and Rp CS.1258n (Eurogentec, Seraing, Belgium), which amplify a 750-bp fragment of the citrate synthase gene (gltA) of Rickettsia, as previously reported (Roux et al. 1997). The conventional PCR assay amplification reactions were performed with a DNA thermal cycler under the following conditions: an initial denaturation step at 95 C for 15 min, followed by 40 cycles consisting of 1 min denaturation at 94 C, 1 min annealing at 60 C, and a 1 min extension at 72 C. A final extension cycle at 72 °C for 5 min was performed, and the reactions were cooled at 15 C. Next, the positive samples were amplified by using primers Rr 190.70, Rr 190.180, and Rr 190.701, which amplify a 629–632-bp fragment of the gene for outer membrane protein A (ompA) (Fournier et al. 1998). Finally, one more PCR reaction using primers 120-M59 and 120–807 that amplify a 865 bp of the outer membrane protein B (ompB) gene fragment was used (Roux and Raoult 2000). For primers targeting the ompA and ompB genes, the amplification conditions were as described above. A negative control of DNA extracted from non-infected laboratory ticks and a positive control of R. rickettsii DNA (one negative control for every 10 tested ticks) were included in each test. All laboratory procedures used in this study, including the amplification and sequencing of sample, have contributed to achieve the higher levels of Rickettsia species diversity, as evidenced in this study.

DNA purification, sequencing, and phylogenetic analyses

The gltA, ompA, and ompB PCR products were then purified using QIAquick PCR purification kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. Purified products were then directly sequenced by using an ABI Prism BigDye terminator cycle sequencing ready reaction kit (Life Technologies, Italy) on an ABI 377 DNA sequencer, according to the manufacturer’s recommendations. The sequencing reactions were performed with the forward and reverse primers used for the PCR amplifications. Sequences generated with gltA, ompA, and ompB primers were edited with Chromas 2.2 (Technelysium, Helensvale, Australia), aligned with CLUSTALX (Larkin et al. 2007) in order to assign them to unique sequence types, and then checked against the GenBank database with nucleotide blast BLASTN (Altschul et al. 1990). Pairwise/multiple sequence alignments and sequence similarities were calculated using the CLUSTALW (Thompson et al. 1994) and the identity matrix options of Bioedit (Hall 1999), respectively.

The evolutionary history was inferred by using the maximum likelihood method based on the protein-coding gltA, ompA, and ompB genes, in the MEGA6 (Tamura et al. 2013) program (http://megasoftware.net/). Statistical support for internal branches of the trees was evaluated by bootstrapping with 1000 iterations (Felsenstein 1985). Initial trees for the heuristic search were obtained by applying the neighbor-joining method to a matrix of pairwise distances estimated using the maximum composite likelihood (MCL) approach. All positions containing gaps and missing data were eliminated.

The sequence types generated with gltA primers were aligned with a set of 27 sequences representing gltA variability of the different species belonging to the genus Rickettsia. The reference sequences were as follows: R. conorii (U59728), R. bellii (CP000849), R. prowazekii (CP003394), R. typhi (CP003398), R. hoogstralii (KF791209), R. felis (CP000053), R. canadensis (CP000409), R. australis (CP003338), R. monacensis (LN794217), R. helvetica (KP866150), R. montanensis (U74756), R. massiliae (KU498299), R. rhipicephali (U59721), R. aeschlimannii (HQ335150), R. raoultii (KT899090), R. amblyommatis (KY273595), R. japonica (AP017602), R. heilongjiangensis (AB473812), R. slovaca (AY129301), R. parkeri (JN126320), R. rickettsii (HM446474), R. philipii (CP003308), R. africae (JN043505), R. israeli tick typhus (U59727), R. sibirica (KM288711), R. mongolotimonae (DQ097081), R. akari (CP000847).

The ompA sequence types generated during this study were also aligned with a set of 22 ompA sequences available in the GenBank database and representative of the Rickettsia genus. The ompA reference sequences were as follows: R. conorii (KT368818), R. felis (KY172882), R. canadensis (CP000409), R. australis (AF149108), R. monacensis (EU665233), R. montanensis (CP003340), R. massiliae (KR401145), R. rhipicephali (CP003342), R. aeschlimannii (HQ335157), R. raoultii (MF511255), R. amblyommatis (CP015012), R. japonica (KY484160), R. heilongjiangensis (AF179362), R. slovaca (KX506735), R. parkeri (EU715288), R. rickettsii (CP018914), R. philipii (CP003308), R. africae (KT633262), R. israeliensis (KF245449), R. sibirica (MF098408), and R. akari (CP000847).

Finally, sequence types obtained with primer targeting the the ompB genes were similarly processed and used to reconstruct evolutionary history. The dataset included 18 sequences: R. slovaca (KX506741), R. massiliae (MF098412), R. raoultii (KU961542), R. sibirica (AF123722), R. rhipicephali (KX018051), R. mongolotimonae (AF123715), R. africae (KF660535; AF123706), R. aeschlimannii (KU961544), R. rickettsii (GU723475), R. parkeri (KX018050), R. israeliensis (AF123712), R. conorii subsp. caspia (AY643093), R. honei (AF123711), R. amblyommatis (MG674591), R. montanensis (AF123716), R. japonica (KY364904), and R. hulinensis (AY260452).

Sequence accession numbers

Sequences of the rickettsial gltA, ompA, and ompB genes were deposited in the GenBank using the National Center for Biotechnology Information (NCBI; Bethesda, MD) BankIt v3.0 submission tool (http://www3.ncbi.nlm.nih.gov/BankIt/). Accession numbers are: MH064440-MH064462 (Rickettsia gltA gene), MH532235-MH532257 (Rickettsia ompA gene), MH532258-MH532280 (Rickettsia ompB gene).

Results

A total of 185 adult ticks from wild and domestic animals, human and vegetation, were collected from 23 collection sites and identified morphologically as: Rhipicephalus sanguineus s.l., R. bursa, R. annulatus, Hyalomma marginatum, Hy. lusitanicum, Haemaphysalis punctata, R. pusillus, and Dermacentor marginatus (Table 1). One exemplar of the soft tick Ornithodoros maritimus was also detected.

Table 1 Tick species analyzed in this study including the collection sites, host, number, stage and sex, and the total of gltA, ompA, and ompB positive PCR ticks
Full size table

By using gltA, ompA, and ompB PCR, the DNA of Rickettsia spp. was detected in 29 out of 185 (16%) of ticks removed from different wild and domestic hosts (dog, cattle, deer, wildboar) and from the vegetation (Table 1). The sequence analyses of 23 PCR products obtained by gltA, ompA, and ompB PCR primers returned clear sequencing signals, but the quality of sequencing of six samples was of low quality (presumably due to the presence of more than one bacterial species in the sample material) and they were excluded from all subsequent analysis.

Sequence analyses of gltA-positive amplicons showed that three R. sanguineus s.l. ticks shared 100% sequence identity with gltA sequences of R. massiliae; one Hy. lusitanicum and two Hy. marginatum positive ticks shared the 100% nucleotide identity with the gltA gene of R. aeschlimannii strains. Eight out of 29 D. marginatus ticks (28%) contained DNA of Rickettsia, which showed 100% sequence identity with the 750-bp fragment of the Ri. slovaca gltA gene. Three R. sanguineus s.l. and six D. marginatus ticks were positive for the presence of rickettsial DNA that shared 99–100% identity with that of R. raoultii strains isolated from ticks worldwide, which were identified as the closest match by nBLAST.

Ticks positive for rickettsial gltA were also positive when tested with the primers targeting the outer surface protein rOmpA (ompA) and the outer-membrane protein rOmpB (ompB) genes. Upon sequencing and ClustalX alignment, the comparative analysis of Rickettsia sequences obtained from the ompA gene was 100% identical with R. massiliae (sequence type Rm-ompA1 detected from three R. sanguineus s. l.), R. aeschlimannii (sequence type Ra-ompA2 from one Hy. lusitanicum and two Hy. marginatum ticks), R. raoultii (sequence type Rr-ompA4 from six D. marginatus and three R. sanguineus s. l. ticks), and R. slovaca (sequence type Rs-ompA3) strains. The ompB sequences obtained from three R. sanguineus s. l. showed that the closest sequences available in GenBank were those for R. massiliae (sequence type Rm-ompB1). Two sequences from one Hy. lusitanicum and two Hy. marginatum ticks showed 100% identity with R. aeschlimannii (sequence type Ra-ompB2); six D. marginatus and three R. sanguineus s. l. ticks were 100% identical with R. raoultii (sequence type Rr-ompB3). Finally, eight D. marginatus ticks shared a Rickettsia sequence that was 100% similar to R. slovaca (Rs-ompB4) as detailed in Table 2.

Table 2 Detection and BLASTN maximum identities of Rickettsia spp. identified from different tick species by gltA, ompA, and ompB genes
Full size table

Phylogenetic analysis based on the alignment of the five gltA sequence types with the rickettsial reference sequences allowed to identify two main groups: the spotted fever group including 26 sequences and the typhus group comprising R. prowazekii and R. typhi (Fig. 1). The phylogenetic tree indicated that Rm-gltA1 (obtained from 3 R. sanguineus s.l.) and Ra-gltA2 (from two Hy. marginatum and one Hy. lusitanicum) sequence types grouped with reference strains representative of R. aeschlimannii and R. massiliae, respectively. The two sequence types named Rr-gltA3 (derived from three R. sanguineus s.l.) and Rr-gltA4 (derived from six D. marginatum) were closely related to the R. raoultii sequence reference strain. Finally, the Rs-gltA5 (obtained from eight D. marginatus) sequence type formed a single clade with R. slovaca reference strain. The five sequence types detected in this study can be classified within the SFG rickettsiae. The main sequence clusters were statistically supported by bootstrap analyses.

Fig. 1
figure 1

Phylogenetic tree inferred with partial sequences of the gltA gene of Rickettsia species generated in this study and other sequences representative of the different species of the genus Rickettsia. The evolutionary history was inferred using the maximum likelihood method. Numbers next the branches indicate bootstrap values based on 1000 replicates. The Rickettsia sequences obtained in this study are represented in bold and marked with a circle

Full size image

Phylogenetic analyses conducted on the alignment of ompA and ompB sequence types obtained in this study with selected Rickettsia sequences found in GenBank are shown in Figs. 2 and 3.

Fig. 2
figure 2

Phylogenetic tree inferred with partial sequences of the ompA gene of Rickettsia species generated in this study and other sequences representative of the different species of the genus Rickettsia. The evolutionary history was inferred using the maximum likelihood method. Numbers next the branches indicate bootstrap values based on 1000 replicates. The Rickettsia sequences obtained in this study are represented in bold and marked with a circle

Full size image
Fig. 3
figure 3

Phylogenetic tree inferred with partial sequences of the ompB gene of Rickettsia species generated in this study and other sequences representative of the different species of the genus Rickettsia. The evolutionary history was inferred using the mximum likelihood method. Numbers next the branches indicate bootstrap values based on 1000 replicates. The Rickettsia sequences obtained in this study are represented in bold and marked with a circle

Full size image

Discussion

Sardinian ticks have been reported to infest wild and domestic hosts and to be a rich reservoir of well-known families of bacterial pathogens (Chisu et al. 2018; Masala et al. 2012a; Satta et al. 2011). The results of this study confirm that Sardinian ticks carry a wide range of bacteria from the Rickettsia genus and provide for the first time molecular evidence on the occurrence of R. raoultii in R. sanguineus s.l. and D. marginatus ticks collected from domestic (dogs and cattle) and wild vertebrates (wild boar and deer). Rhipicephalus ticks were the most abundant species detected in this study. The phylogenetic analyses showed that the Rickettsia species obtained in this study were closely related to representative SFG rickettsiae with known pathogenicity, which can cause human diseases (Fig. 1). Since the ticks have been collected directly from the hosts, no conclusion can be drawn regarding the circulation of the pathogens within the tick population, as every detection can be the result of ingesting infected blood.

The identification of R. massiliae in R. sanguineus s.l. ticks is consistent with previous reports where the potential role for Rhipicephalus ticks in the transmission of these pathogens has been postulated (Chisu et al. 2018, 2017; Parola et al. 2013). Rickettsia massiliae is recognized as a pathogenic species causing spotted fever in human (Parola et al. 2013) and where it has been putatively linked to mild to moderately severe illnesses in dogs in California (Beeler et al. 2011). Human cases of R. massiliae have been documented in patients from Sicily, Italy (Cascio et al. 2013; Vitale et al. 2006).

Detection of R. slovaca in D. marginatus confirms the role played by these tick species in R. slovaca transmission, as previously reported in other studies (Masala et al. 2012b; Chisu et al. 2017, 2018).

Rickettsia slovaca is associated with a syndrome characterized by scalp eschars and neck lymphadenopathy following tick bites. This syndrome was named TIBOLA (tick-borne lymphadenopathy) or DEBONEL (Dermacentor-borne necrotic erythema and lymphadenopathy). The term “SENLAT” (scalp eschar and neck lymphadenopathy after a tick bite) has also been recently proposed (Parola et al. 2013; Angelakis et al. 2010).

In this study, the human pathogen R. aeschlimannii was identified from Hy. lusitanicum and Hy. marginatum ticks collected from the vegetation and cattle, respectively. The presence of R. aeschlimannii in Hy. marginatum ticks was consistent with previous studies reported in Sardinia and in other countries (Chisu et al. 2017; Santos-Silva et al. 2006; Wallménius et al. 2014). Recently, D. marginatus and Hy. lusitanicum have been indicated as potentially new tick vectors of R. aeschlimannii (Parola et al. 2013). The pathogenicity of this bacterium to humans is not well understood, although a clinical picture similar to MSF-like lesions was reported. Infections in humans have been previously confirmed in Europe, South Africa, Algeria, and Tunisia (Blanda et al. 2017; Portillo et al. 2015; Germanakis et al. 2013; Demoncheaux et al. 2012; Pretorius and Birtles 2002; Raoult et al. 2002).

In this study, gltA, ompA, ompB-based molecular diagnosis allowed to identify the presence of R. raoultii sequences in D. marginatus and R. sanguineus s.l. ticks from vertebrate hosts. R. raoultii is a member of the spotted fever group rickettsiae and has been implicated in cases of DEBONEL/TIBOLA/SENLAT (Mediannikov et al. 2008). Rickettsia raoultii has been reported in many European and Asian countries in Dermacentor ticks. Other hard ticks, such as Haemaphysalis, Rhipicephalus, Hyalomma, and Amblyomma ticks, were also involved (Blanda et al. 2017).

These data implicate the broad and ubiquitous distribution of Rickettsia species in Sardinian ticks. The high percentage of R. sanguineus s.l. and D. marginatus ticks infected with R. slovaca and R. raoultii strongly indicates an increase in the prospect of medical intervention for persons in localities where these ticks occur. Clinicians should be aware that patients with tick-borne lymphadenopathy may be on the island.