Molecular Detection of Spotted-Fever Group Rickettsiae in Ticks Collected from Domestic and Wild Animals in Corsica, France

To obtain a better understanding of the current magnitude of tick-borne rickettsioses in Corsica, we used molecular methods to characterize the occurrence of Rickettsia spp. in ixodid ticks collected from domestic and wild animals. The presence of Rickettsia spp. was evaluated using real-time polymerase chain reaction targeting the gltA gene and by sequencing of gltA and ompA partial genes for species identification and phylogenetic analysis. Infection rates were calculated as the maximum-likelihood estimation (MLE) with 95% confidence intervals (CI). In total, 1117 ticks belonging to four genera (Rhipicephalus, Hyalomma, Ixodes, and Dermacentor) were collected from cattle, sheep, wild boars, and companion animals during July–August 2017 and July 2018–January 2019. Overall, Rickettsia DNA was detected in 208 of 349 pools of ticks (MLE = 25.6%, 95% CI: 22.6–28.8%). The molecular analysis revealed five different rickettsial species of the spotted-fever group (SFG). We highlighted the exclusive detection of Candidatus Ri. barbariae in R. bursa and of Ri. aeschlimanii in H. marginatum. Rickettsia slovaca was detected in D. marginatus collected from wild boars. This study provides the first evidence of the presence of Ri. monacensis in I. ricinus ticks isolated from a dog in Corsica. In conclusion, our data revealed wide dispersal of SFG Rickettsiae and their arthropod hosts in Corsica, highlighting the need for surveillance of the risk of infection for people living and/or working close to infected or infested animals.


Introduction
Rickettsiae are obligate intracellular Gram-negative bacteria that are transmitted to vertebrate hosts by arthropod vectors such as ticks, fleas, lice, and mites. The genus Rickettsia (family Rickettsiaceae; order Rickettsiales) comprises 31 species that cause diseases in vertebrate hosts, including humans and domestic and wild animals [1]. Ticks can transmit these bacteria to humans and animals by feeding. Members of the genus Rickettsia can be classified into four phylogenetic groups: (1) the spotted-fever group (SFG), (2) the typhus group (TG), which includes the agents of epidemic typhus and murine typhus, (3) the ancestral group (AG), consisting of Ri. bellii and Ri. canadensis, and (4) the transitional group, the Ri. akari group (TRG) [2]. The SFG and AG Rickettsiae are mainly associated with ticks, while TG and TRG Rickettsiae are associated with other arthropods such as lice, fleas, and mites [3].
Rickettsia spp. and their associated human clinical diseases vary depending on geographical location [4]. In Europe, rickettsioses are caused mainly by tick-borne SFG Rickettsiae [4]. In the Mediterranean area, Rickettsiae species have been detected in several tick species, but human rickettsioses are usually considered to be caused by Ri. conorii, the causative agent of Mediterranean spotted fever (MSF) [1,5]. Rickettsia aeschlimanii, Ri. massiliae, and Ri. monacensis can also cause a similar clinical presentation. Other species that have been associated with human disease and detected in Europe include Ri. slovaca and Ri. raoultii, both causes of tick-borne lymphadenopathy and Dermacentor-borne necrosis lymphadenopathy (TIBOLA/DEBONEL) [6,7].
Information about the circulation and distribution of Rickettsia-infected tick species in Corsica, a French Mediterranean island, is scarce and fragmented. Rickettsia aeschlimanii and Candidatus Ri. barbariae were identified in Hyalomma marginatum and Rhipicephalus bursa collected from domestic animals in 2016 [8]. Rickettsia slovaca has been detected in ticks collected from vegetation. Rickettsia massiliae was reported from an R. turanicus collected from a dog in Corsica in 2005, and Ri. africae was detected in one Amblyomma variegatum collected from bovine in August 2018.
To obtain a better understanding of the current magnitude of tick-borne rickettsioses in Corsica, we aimed to use molecular methods to describe the presence of tick-borne Rickettsiae in tick species collected from domestic and wild animals. From a public health perspective, a better understanding of local tick species and of Rickettsia species carried by them is essential to improve local surveillance systems.
Ten ticks were collected from 500 sheep in one farm (Biguglia, Haute-Corse) in June 2017. All these were R. bursa (Table 1).
The infestation rate of sheep was 2% and it was not drastically different from the 4% observed in cattle during the same period (May 2017) (p = 0.68).
The infestation rate observed in wild boars was 38.6% (46/119). Dermacentor marginatus was mainly collected from wild boars throughout the surveillance period (August 2018-January 2019), with a peak of activity in December 2018. September 2018 was characterized by the co-circulation of four species of ticks in wild boars (Figure 1b).
The infestation rate of sheep was 2% and it was not drastically different from the 4% observed in cattle during the same period (May 2017) (p = 0.68).
The infestation rate observed in wild boars was 38.6% (46/119). Dermacentor marginatus was mainly collected from wild boars throughout the surveillance period (August 2018-January 2019), with a peak of activity in December 2018. September 2018 was characterized by the co-circulation of four species of ticks in wild boars (Figure 1b).
In sheep, the 10 ticks were analyzed in two pools. The detection rate of Rickettsia spp. DNA for R. bursa was 100%.
In sheep, the 10 ticks were analyzed in two pools. The detection rate of Rickettsia spp. DNA for R. bursa was 100%.

Rickettsia Species Identification Based on Sequence Analysis
The molecular characterization of the Rickettsiae in ticks was based on gltA (n = 91 of 208 positive pools; 43.8%). A subset of positive samples (n = 33) was also characterized by sequencing ompA to confirm Rickettsia spp. (Table 4). The molecular analysis of gltA and ompA sequences revealed five different rickettsial species.
All Rickettsia detected in H. marginatum tick pools (n = 56) collected from cattle and wild boars were exclusively Ri. aeschlimanii (Table 4). Candidatus Ri. barbariae was the only bacterium detected in R. bursa tick pools collected from sheep (n = 2) and from cattle in 2017 (n = 5). The DNA of Ri. aeschlimanii and Candidatus Ri. barbariae were respectively detected in five and eleven R. bursa pools collected from cattle in 2018. The DNA of Ri. slovaca was detected in six D. marginatus tick pools, in two I. ricinus tick pools and in four R. bursa tick pools, all collected from wild boars. It is to be noted that Ri. slovaca DNA was also detected in one D. marginatus tick collected from a cat. Rickettsia massiliae and Ri. monacensis DNA have been detected in one pool of R. sanguineus s.l. ticks and in one pool of I. ricinus ticks collected from dogs.

Phylogenetic Analysis
The results of phylogenetic trees based on gltA and ompA sequences are illustrated in Figure 2A,B

Phylogenetic Analysis
The results of phylogenetic trees based on gltA and ompA sequences are illustrated in Figure 2  A

Discussion
To our knowledge, this is the first reported evidence of the circulation of five bacteria from the genus of Rickettsia in four genera of ticks (Rhipicephalus, Hyalomma, Dermacentor, and Ixodes), collected from wild and domestic animals mainly reared in Northern Corsica.
In the present study, five tick species were collected, namely, R. bursa, H. marginatum, I. ricinus, D. marginatus, and R. sanguineus s.l. Rhipicephalus ticks were the most abundant species collected in this study. The diversity and seasonal pattern of ticks observed in this study, though collected not over 12 months and on selected animals kept under various conditions, is consistent with that reported previously for Corsica [8,9] and other Mediterranean countries [10][11][12]. In this study, overall, the low infestation rate observed in sheep with respect to cattle could be due to the short collection period and to a low density of tick populations usually reported in pastures where livestock had grazed with respect to undisturbed fallow lands [13].
In the present study, R. bursa was the tick species most frequently collected in cattle and sheep, with a peak of activity observed in the summer. This typical Mediterranean species is well established in the whole of Corsica [8,9]. In this study overall, almost 50% of R. bursa tick pools, mainly collected from cattle and sheep, were positive for Rickettsia spp. Interestingly, Candidatus Ri. barbariae DNA was exclusively detected in tick pools of R. bursa removed from sheep (100% of sequenced tick pools) and mostly in cattle (76%). Our detection of Candidatus Ri. barbariae in R. bursa collected from domestic animals reinforces its presence in Corsica ( [8], pp. 606-613) and highlights the possible role played by R. bursa ticks in the natural maintenance of this bacterium. To understand better the role of R. bursa in the maintaining of Candidatus Ri. barbariae, ticks should be collected from nature. Candidatus Ri. barbariae is a presumed new species of the genus Rickettsia genetically characterized for the first time in R. turanicus from Sardinia [14], a Mediterranean island separated from Corsica by 10 km of sea. The presence of this bacterium has also been reported in several Rhipicephalus spp. in Portugal, Italy, France, Algeria, Cyprus, Israel, Cameroon, Lebanon [12,[15][16][17][18][19], and in Hyalomma ticks in the West Bank [20]. The pathogenic role of Candidatus R. barbariae remains unknown for humans and animals, although it has been identified in an R. bursa tick removed from a woman in Greece [21].
Hyalomma marginatum was the second main tick species collected from cattle in the northern region of Corsica. This tick species was almost exclusively collected from cattle (99%). One H. marginatum was collected from a wild boar. Rickettsia spp. DNA was detected in more than 80% of H. marginatum pools collected from cattle sampling in Northern Corsica. Hyalomma marginatum ticks were exclusively infected with the human pathogen Ri. aeschlimanii, regardless of the host species (cattle or wild boars). In Corsica, cattle are reared outside all year round and acaricide treatments are rarely performed, increasing the opportunity of a wide circulation of this pathogen in Corsica. Such a high infection rate is consistent with the previously reported rate of infection of Hyalomma by Ri. aeschlimanii (>50% and >70%) in Corsica [8,22], in Croatia (64%) [23], and in Germany (50%) [24]. Rickettsia aeschlimanii may be spread through migratory birds from Africa, and it was detected in Hyalomma ticks collected from barn swallows (Hirundo rustica) in Corsica [25]. The first human case of infection was described in a patient who developed symptoms after returning from Morocco [26].
Even if wild boars and cattle share the same environment in Corsica [9], dominant tick species collected in the present study from these animals differed. In the present study, we confirmed that D. marginatus was the main tick species collected from wild boars, confirming the host preference previously reported. This tick species was not observed in cattle investigated in this study.
In the present study, most specimens of D. marginatus were collected during winter because the adults are active within a temperature range of 4-16 • C [27]. More than half of the pools of D. marginatus collected were positive for Rickettsia spp. Sequence analyses revealed for the first time the presence of Ri. slovaca DNA in D. marginatus ticks collected from wild boars in Corsica. This bacterium was previously detected in D. marginatus collected by flagging in Corsica and in other regions of the northern Mediterranean area. Rickettsia slovaca was identified in D. marginatum ticks collected on swine and wild boars in Sardinia [28]. Overall, these results suggested that the role of D. marginatus is important for the maintenance of Ri. slovaca. In Corsica, recreational hunting of wild boars could provide an ideal environment for the transmission of Ri. slovaca between wild boars, humans, and companion animals (e.g., hunting dogs). At present, the D. marginatus tick is recognized as the main vector and reservoir for Ri. slovaca in Mediterranean areas, including southern Europe and North Africa [1]. This human pathogen has been detected in almost half of D. marginatus ticks collected from rodents in the northern Apennines (Italy) and by flagging in Kazakhstan [29,30]. Rickettsia slovaca is associated with a syndrome characterized by neck lymphadenopathy following tick bites [6]. Human infection with Ri. slovaca has been described in several European countries, including France [31]. We also reported for the first time the detection of Ri. slovaca DNA in one D. marginatus collected from a cat.
In this study, R. sanguineus s.l. ticks were exclusively collected from dogs. We reported the detection of Ri. massiliae DNA in one pool of R. sanguineus s.l. collected from a dog. Rhipicephalus sanguineus s.l. is the likely reservoir of Ri. massiliae, with transovarian passage rates up to 100%. Rickettsia massiliae was reported from a specimen of R. turanicus collected from a dog in Corsica [32] and has been putatively linked to mild to moderately severe illnesses in dogs in California [33]. Rickettsia massiliae is recognized as a pathogenic Rickettsia causing spotted fever in humans [1].
In the present study, I. ricinus, mainly collected during autumn, was sporadic and represented almost 2% of all ticks sampled. The DNA of Ri. slovaca was detected in pools of I. ricinus ticks collected from wild boars. We report for the first time the detection in Corsica of Ri. monacensis DNA in one pool of I. ricinus ticks collected from one dog. Rickettsia monacensis, an emerging human pathogen of the SFG Rickettsiae family, has been previously detected in I. ricinus collected from dogs in Spain [34] and in dog blood samples from the Maio Islands [35].
This study has some limitations. Finding the DNA of human and veterinary pathogens in feeding ticks is only a marker of the community of tick-transmitted pathogens circulating in the target territory [36]. Ticks were mostly analyzed in pooled samples and not individually. Working with pools can lead to underestimation of the infection rate, which must be interpreted with caution.
The ticks were mostly collected in a municipal slaughterhouse in Northern Corsica. We have no data about the southern part of the island. Moreover, ticks from cattle, wild boars, and dogs were collected from several locations while the sheep were only from one farm.

Study Area and Ticks
Ticks were manually collected at several sites in Corsica ( Figure 3) as follows: (i) in May 2017 from sheep (n = 500) bred in one farm located at the edge of the largest lagoon in Corsica, the Biguglia Nature Reserve; (ii) in July-August 2017 and July-December 2018 from cattle (n = 140 and n = 241, respectively) in the Ponte-Leccia slaughterhouse, and (iii) from wild boars (n = 119) during the 2018 hunting season (August 2018-January 2019). In July-August 2017, ticks were sampled from 13 dogs and one cat that presented at three veterinary clinics for a variety of reasons and not specifically for symptoms related to tick infestation.
Ticks were collected from sheep during milking mainly around the perineum and udder. The farmer declared he had never treated sheep against ectoparasites. The breeding system was extensive; thus, the sheep live in open pastures.
During each visit to the slaughterhouse, the whole skin of slaughtered animals was inspected, and ticks were collected manually. The national cattle identification system, which uses ear tags, allowed the origin of the animals to be tracked and the farm owners to be identified. One to 20 ticks were collected from each infested animal.

Morphologic Identification of Ticks at the Species Level
Ticks were identified at species level using a pictorial guide [38]. Specific attention was brought to Ixodes ticks to identify I. inopinatus [39].

Morphologic Identification of Ticks at the Species Level
Ticks were identified at species level using a pictorial guide [37]. Specific attention was brought to Ixodes ticks to identify I. inopinatus [38].

DNA Extraction
Ticks were washed once in 70% ethanol for 5 min and then twice in distilled water for 1 min each time. Following species identification, ticks were analyzed individually or using monospecific pools consisting of two to six ticks according to developmental stage (nymphs, not engorged females, and male adults) and host. Individual or pools of ticks were crushed in phosphate-buffered saline using TissueLyser II (Qiagen, Hilden, Germany) at 5500 rpm for 20 s. DNA extraction was performed using QIAcube HT (Qiagen) and a QIAamp cador Pathogen Mini Kit according to the manufacturer's instructions. DNA was eluted in 150 µL of buffer and stored at −20 • C. For each polymerase chain reaction (PCR) reaction, the template DNA was at a final amount of <200 ng.

Molecular Identification of Ticks at the Species Level
Morphological identification of ticks was confirmed using PCR amplification and sequencing of mitochondrial 16S rDNA [39] (Table 5).

PCR Detection of Rickettsia
The rPCR assays for Rickettsia spp. were performed using the primers and probes previously described for the amplification of the citrate synthase gene (gltA) ( Table 5) [42]. Reactions were performed on a 96-well Applied Biosystems™ QuantStudio™ 3 Real-Time PCR System using QuantiFast Pathogen + Internal Control Kits (Qiagen). Internal and negative controls were included in each run.
Positive samples detected using rPCR were then analyzed using conventional PCR using primers that amplified the 850-bp fragment of the gltA gene common to the Rickettsia genus [40,41] and primers to amplify the 532-bp fragment of the 190-kDa outer membrane protein (ompA) gene specific to the SFG Rickettsiae [43] (Table 5). The reactions were carried out using an Applied Biosystems GeneAmp PCR System 9700 (Applied Biosystems, Courtaboeuf, France). The PCR products were separated on 2% agarose gels in tris-acetate-ethylenediaminetetraacetic acid (TAE buffer) and were visualized under ultraviolet light after staining with ethidium bromide. A 100-bp DNA ladder was used as a standard marker.
Strong positive samples were purified and directly sequenced using an Applied Biosystems model 3730XL (Fisher Scientific S. A. S., Illkirch-Graffenstaden, France). The newly generated sequences were aligned using Mega X [44]. All sequences were assembled and compared with similar sequences retrieved from the GenBank nucleotide database using BLASTn [45]. Phylogenetic analyses were inferred using the maximum-likelihood method implemented in the Mega X program [44].

Statistical Analysis
Infection rates for pooled ticks were calculated using the MLE method with 95% CI for unequal pool sizes and expressed as the MLE of infection rate per 100 ticks. MLE calculations are based on the number of pools, pool sizes (number of individuals per pool), and the number of positive pools [46].
The rates of infection or of tick infestation were compared using Fisher's exact test (p < 0.05). The analysis was performed using the R statistical platform (version 3.1.2) [47].

Conclusions
In conclusion, our results contribute to a better knowledge about SFG Rickettsiae in Corsica and provide a useful contribution to understanding their epidemiology. This study reports the first evidence of the circulation of five species of the Rickettsia genus in Corsican ticks representing four genera, namely, Rhipicephalus, Hyalomma, Dermacentor, and Ixodes, that were collected from wild and domestic animals. We highlighted the exclusive detection of Candidatus Ri. barbariae in R. bursa collected from sheep and the wide circulation of the human pathogen Ri. aeschlimanii in H. marginatum collected from cattle. Moreover, the high percentage of D. marginatus with positive detection to Ri. slovaca in wild animals indicates a wide distribution of this human pathogen in Corsica. This study reports evidence for the first time that Ri. monacensis, a human pathogen, could circulate in ticks collected from dogs in Corsica. Implications for human health of the circulation of these two pathogens merits further investigation and should be taken into account by physicians seeing patients with unexplained febrile illness.

Ethics Approval and Consent to Participate
The inspected cattle were slaughtered for human consumption. Living sheep were examined with the assistance of their owner. Companion animals were examined by veterinarians with the agreement of the owners. The wild boars collected were legally hunted during the hunting season.

Abbreviations DEBONEL
Dermacantor-borne necrosis lymphadenopathy gltA citrate synthase MLE maximum likelihood estimation MSF Mediterranean spotted fever ompA outer membrane proteins A PCR polymerase chain reaction SFG spotted fever group TIBOLA tick-borne lymphadenopathy