Zoonotic Rickettsia Species in Small Ruminant Ticks From Tunisia

Tick-borne rickettsioses present a significant public health threat among emerging tick-borne diseases. In Tunisia, little is known about tick-borne Rickettsia pathogens. Therefore, the aim of this study was to investigate the presence of Rickettsia species in small ruminant ticks from Tunisia. Adult ticks (n = 694) were collected from goats and sheep in northern Tunisia. Obtained ticks were identified as Rhipicephalus turanicus (n = 434) and Rhipicephalus sanguineus sensu lato (n = 260). Selected ticks (n = 666) were screened for the presence of Rickettsia spp. by PCR targeting a partial sequence of the ompB gene followed by sequence analysis. Rickettsial DNA was detected in 122 (18.3%) tested tick samples. The infection rates in Rh. turanicus and Rh. sanguineus s.l. ticks were 23.4 and 9.5%, respectively. The overall prevalence of rickettsial DNA was markedly higher in ticks collected from goats (23.2%) compared to those infesting sheep (7.9%). The detection of rickettsial DNA was significantly higher in ticks from the governorate of Beja (39.0%) than those from the governorate of Bizerte (13.9%). Two additional genes, the outer membrane protein A gene (ompA) and the citrate synthase gene (gltA), were also targeted for further characterization of the detected Rickettsia species. Genotyping and phylogenetic analysis based on partial sequences (n = 106) of the three different genes revealed that positive ticks are infected with different isolates of two Spotted Fever Group (SFG) Rickettsia, namely, Rickettsia massiliae and Rickettsia monacensis, closely related to those infecting camels and associated ticks from Tunisia, and humans and small ruminant ticks from neighboring countries like Italy, France, and Spain.


INTRODUCTION
Rickettsia species (family Rickettsiaceae; order Rickettsiales) are included into four groups: the spotted fever group (SFG) rickettsiae, the typhus group, the Rickettsia bellii group, and the Rickettsia canadensis group (1). These pathogens infected several domesticated and wild vertebrate hosts through hematophagous arthropod vectors bites (mainly ticks, fleas, and mites). Besides, tick-borne rickettsioses are considered as one of the most virulent zoonotic diseases affecting humans especially in African countries (2). Spotted fever group rickettsioses (SFG) are actually considered as emerging and reemerging diseases affecting animals worldwide. They are caused by the pathogenic and zoonotic spotted fever Rickettsia bacteria mainly transmitted by ticks. Humans may be accidently infected especially in tropical areas (1,2).
In Tunisia, several SFG Rickettsia species have been previously reported, as Rickettsia conorii, that was described for the first time in humans since 1910 (3), and, recently, by Znazen et al. (4) and Khrouf et al. (5). In addition, R. conorii subsp. israelensis was identified in one human and tick specimens of Rhipicephalus sanguineus s.l. complex collected from dogs (4,6). Furthermore, R. aeschlimannii, R. helvetica, and R. africae were reported from camels' blood samples and infesting Hyalomma tick tissues in southern and central Tunisia (7,8). DNA of R. helvetica was also identified in questing Ixodes ricinus ticks (9).
Rickettsia monacensis was earlier detected in I. ricinus ticks from several European countries like Italy, Spain, Romania, Bulgaria, Hungary, and Serbia (1,12). In our country, the first identification of R. monacensis was also reported in I. ricinus ticks by Sfar et al. (9). Additionally, this human-pathogenic species was recently detected not only in Tunisian camels but also in associated H. impeltatum ticks removed from uninfected animals (8). This bacterium causes from moderate to severe infections in humans including fever, rash on palms and soles, and inoculation eschar (19,20). To better understand the epidemiology of Rickettsia species in Tunisia, we investigated, in the present molecular survey the occurrence of rickettsial bacteria in small ruminant ticks according to potential risk factors. Molecular characterization and phylogenetic analysis of revealed Rickettsia spp. isolates were also performed by using three different gene fragments.

Study Area Description
A cross-sectional study was carried out in five localities of Northern Tunisia (Figure 1)
All partially engorged ticks were collected by using a clamp from different preferred sites of small ruminant body (ears, neck, udder, and external genitalia) and separately categorized according to the examined animal host. Obtained specimens were morphologically identified using the taxonomic key of Walker et al. (21) and then classified according to tick species, life stage, and gender. Each tick specimen was individually conserved in a tube containing 70% ethanol and stored at −20 • C.

Total DNA Extraction and Tick DNA Amplification
Each identified tick was washed with sterile water, dried, and crushed individually using an automated TissueLyser LT system (Qiagen, Hilden, Germany). Genomic DNA extraction was performed from each tick sample using the DNeasy tissue kit (Qiagen, Hilden, Germany). Obtained DNA extracts were stored at −20 • C. DNA extraction efficiency was validated by PCR amplification step targeting the ribosomal RNA subunit (16S rRNA) gene using the tick-specific primers TQ16S+1F and TQ16S-2R as described by Black and Piesman (22) (Table 1).

Molecular Detection of Rickettsia spp.
In order to identify all species of the Rickettsia genus, tick DNA samples were subjected to nested PCR targeting a fragment (425 bp) of the rickettsial outer membrane protein B (ompB) gene (23) ( Table 1). For further characterization, the outer membrane protein A (ompA) and the citrate synthase protein (gltA) gene fragments (532 and 381 bp, respectively) were amplified by using nested and endpoint PCR, respectively ( Table 1). PCR reactions were performed in an automated DNA thermal cycler. Thermal cycling profiles were as described by Oteo et al. (24), and Regnery et al. (25), respectively.
The PCR reactions were carried out in a final volume of 50 µl composed of 0.125 U/µL of Taq DNA polymerase (Biobasic Inc., Markham, Canada), 1× PCR buffer, 1.5 mM MgCl2, 0.2 mM of dNTP, 3 µL of genomic DNA (50-150 ng) in the first PCR and 1 µL in the second PCR (for nested PCR), 0.5 µM of the primers, and autoclaved water. PCR products were visualized using electrophoresis in 1.5% agarose gels stained with ethidium bromide and observed under UV transillumination.

Statistical Analysis
Exact confidence intervals (CI) at the 95% level were estimated for prevalence rates according to different considered factors.  A comparison of the prevalence of Rickettsia species in ticks according to abiotic factors (geographic location and bioclimatic conditions) and factors related to ticks (gender, age, and host origin) was carried out using the Epi Info 6 software 01 (CDC, Atlanta, USA) and the χ 2 -test. A difference is considered statistically significant when the degree of significance p is ≤0.05.
In order to assess possible confusion between the risk factors, a Mantel-Haenszel χ 2 -test was performed.

DNA Sequencing and Obtaining Final Sequences
A total of 106 positive PCR products obtained after ompB, ompA, and gltA PCRs were randomly selected and purified using the GF-1 Ambi Clean kit (Vivantis, USA), according to the manufacturer's instructions. Purified DNA amplicons were sequenced in both directions, using the same primers as for the single gltA PCR and the second PCR of each nested PCR amplification targeting ompA and ompB genes. The Big Dye Terminator cycle sequencing ready reaction kit (Applied Biosystems, Foster City, USA) and an ABI3730XL automated DNA sequencer (Macrogen Europe, Amsterdam, The Netherlands) were employed. The chromatograms were evaluated with Chromas Lite v 2.01 (http://www.technelysium.com.au/chromas_lite.html). To obtain maximal data accuracy, sequences were determined on both forward and reverse strands. Indeed, the complementary strands of each sequenced product were manually assembled by using the DNAMAN software (Version 5.2.2; Lynnon Biosoft, Que., Canada). The primer region sequences were automatically removed and the overlapping parts were selected.

Sequence Alignment and Phylogenetic Study
Multiple-sequence alignments and sequence similarities were calculated using the CLUSTAL W method (26). BLAST analysis was performed to assess the level of similarity with previously reported sequences (http://blast.ncbi.nlm.nih.gov/). By using the DNAMAN software, genetic distances among the operational taxonomic units were computed by the maximum composite likelihood method (27) and were used to construct neighborjoining trees (28). Statistical support for internal branches of trees was evaluated by bootstrapping with 1,000 iterations (29).

Efficiency of DNA Isolation
DNA extracts were tested and validated in 666 samples (96%). No amplification products were obtained for 28 samples, reflecting a probable failure of the DNA extraction, and were thus excluded from the analysis. Thereby, a total of 666 ticks were selected from goats (452/666, 67.9%) and sheep (214/666, 32.1%) from the higher semiarid area (357/666, 53.6%) and the low humid area (309/666, 46.4%). Almost all analyzed ticks were collected from small ruminants located in the governorate of Bizerte (82.3%) while ticks collected from animals in El Alia are the most numerous (43.4%) compared to those in other localities (Figure 1 and Table 2). The sex ratio of tested ticks (M/F) was 1.15. After the validation of DNA extracts, a total of 423 Rh. turanicus (63.5%) and 243 Rh. sanguineus s.l. (36.5%) were subjected to Rickettsia spp. screening ( Table 2).

Molecular Characterization and Phylogenetic Analysis
Out of 122 Rickettsia-positive samples, 94 gave a clear band in the correct nucleotide size of the partial genes (ompA, ompB, and gltA) in at least one of the three genotyping PCRs. Partial sequences (n = 106) of the three analyzed genes were deposited under GenBank accession numbers presented in Table 4. Based on all revealed sequences of the three analyzed genes, we precisely selected Rickettsia spp. genotypes according to infecting tick species, and they differ from each other by at least one mutation in the nucleotidic sequence.
A phylogenetic analysis based on the alignment of Tunisian genotypes with 31 Rickettsia spp. ompB sequences obtained from GenBank shows the assignment of revealed genotypes to R. massiliae and R. monacensis clusters. The R. massiliae cluster is formed by three subclusters supported by robustness node rates ≥ to 81% (Figure 2). Tunisian strains were assigned to the first and third subclusters. Genotypes ompBRmasRs2 and ompBRmasRt2 were assigned to the first subcluster and clustered with strains isolated from H. impeltatum infesting camels in Tunisia and from Rh. sanguineus s.l. ticks located in Mediterranean countries such as Italy and Spain (Figure 2). Genotypes ompBRmasRs1 and ompBRmasRt1 were assigned to the third subcluster and clustered with strains isolated from Rh. sanguineus s.l. and Rh. turanicus ticks originated from North-Mediterranean countries (Figure 2). The R. monacensis cluster is also formed by three subclusters supported by robustness rates of nodes ≥ to 81% (Figure 2). Genotypes ompBRmonRs1  and ompBRmonRs2 were assigned, respectively, to the first and second subclusters. Genotype ompBRmonRs1 was closely related to isolates found in Tunisian camels and their infesting H. impeltatum ticks, and strains infecting human and ticks from different countries (Figure 2).
For this gene, a phylogenetic tree based on the alignment of ompA partial sequences of Rickettsia spp. found in GenBank showed the presence of our sequences in the three subclusters that formed the R. massiliae cluster and supported by robustness node rates ≥ to 84% (Figure 3). Genotype ompARmasRt7 formed separately subcluster 1, and genotypes ompARmasRt2 and ompARmasRs2 were assigned to the last subcluster and clustered with strains isolated from Rh. sanguineus s.l. located in different worldwide countries such as Italy, Austria, Argentina, and the USA. The remaining genotypes were clustered together in the second subcluster with several isolates infecting ticks from China and European countries (Figure 3).

Rickettsia spp. gltA Genotypes
Sequencing of 341 bp of the gltA partial sequence obtained from 25 specimens of Rh. turanicus-positive to Rickettsia spp. confirmed the infection with only one genotype (gltARmasRt1, GenBank accession number KJ663740) of R. massiliae (Tables 3, 4). This revealed that the genotype was 100% identical to strain 60B infecting Rh. sanguineus s.l. tick collected from Italian human (GenBank Accession Number KJ663740) ( Table 4).
Phylogenetic tree based on the gltA gene revealed that the gltARmasRt1 genotype clustered in the R. massiliae cluster especially in the first subcluster 1 with strains infecting Rh. sanguineus s.l. ticks from Italy and Argentina, Hyalomma asiaticum ticks from China, and R. turanicus tick specimens collected from birds in Portugal (Figure 4). FIGURE 2 | Neighbor-joining tree based on the alignment of partial ompB sequences (382 bp) of Rickettsia spp. obtained in this study with selected sequences representative of the Rickettsia genus. Numbers over the branches indicate the percentage of replicated trees in which the associated taxa clustered together in the bootstrap test (1,000 replicates, only percentages >50% were represented). The six partial ompB sequences representative of different Rickettsia spp. genotypes obtained in this study are indicated in bold. The host or vector, the genotype, strain or isolate name, the country of origin, and the GenBank accession number are indicated. One R. prowazekii ompB partial sequence was added as an outgroup.

DISCUSSION
Data about the occurrence and the genetic diversity of Rickettsia species in ticks is limited in North African countries (30,31), especially in Tunisia (6,8,9). In this report, adult ticks infesting small ruminants in northern Tunisia were examined and two species of Rhipicephalus genus (R. turanicus and R. sanguineus s.l.) were identified. This result is in agreement with other surveys which considered these two tick species as major ectoparasites of small ruminants in Tunisia (32,33).
To our knowledge, we report here for the first time the detection of SFG Rickettsia DNA in ticks collected from small ruminants raised in the north of Tunisia. Although this study does not conclude on the competence of these potential vectors, given that these results do not suggest that the tick species mentioned in this report can serve as a competent vector for detected bacteria, this study made a contribution to the knowledge of the presence of SFG rickettsiae in Tunisia. In addition, present data showed the need to search these bacteria in animal hosts and to increase the investigated areas, the potentially incriminated risk factors, and the number of analyzed tick samples, including questing ticks and different life stages. All these information may facilitate future prevention against SFG Rickettsial diseases in the country.
Specifically, the detection of Rickettsia spp. DNA in Rh. turanicus (23.4%) and Rh. sanguineus s.l. (9.5%) provides evidence that these tick species may be among the main vectors of Rickettsia species in northern Tunisia. These results are consistent with those reported by Khrouf et al. (6) who suggested a possible incrimination of Rhipicephalus ticks infesting dogs and sheep in the transmission of Rickettsia species in central Tunisia. Furthermore, according to Psaroulaki et al. (34), Rickettsia spp. were detected in Rhipicephalus ticks collected from domestic animals in Greece. Additionally, Germanakis et al. (35) reported that Rh. turanicus has been implicated as a potential vector transmitting to humans several pathogens including Rickettsia species. In the Northwest of China, Wei et al. (36) suggested that R. massiliae, R. aeschlimannii, and R. sibirica variants cocirculate in R. turanicus ticks. This data was confirmed by another study conducted by Song et al. (37) in the same country that indicates the occurrence of several SFG rickettsiae in Rh. turanicus collected from several ruminants. Rickettsia massiliae DNA was previously found in the salivary glands, and saliva of Rh. turanicus and its specific antibodies were also detected in patient sera. This may suggest, firmly, that Rh. turanicus act as a potential vector and reservoir for this bacterium (38).
Furthermore, analysis of potential risk factors demonstrated three interesting facts related to geographic regions, potential tick vector species, and infested hosts. Firstly, the positive rates of SFG Rickettsia in ticks were significantly higher in Beja (39%) than in Bizerte (13.9%) governorate. This discrepancy in prevalence rates according to geographic regions could be mainly explained by the diversity and heterogeneity of livestock population especially in El Alia locality and differences in husbandry practices, farm organization, wildlife reservoir hosts, and/or abiotic factors like the air temperature and the relative humidity that significantly affect the distribution of potential tick vectors. In addition, the higher rate of Rickettsia spp. observed in the governorate of Beja exclusively represented by the locality of Amdoun may be partly explained by the abundant presence in this region of I. ricinus considered to be one of the most important vectors of rickettsiae around the world (9). The infection of Rhipicephalus ticks with Rickettsia species may therefore come from infected small ruminants earlier infested with Rickettsia-positive I. ricinus ticks during wet seasons (9). Secondary, the positive rate in Rh. turanicus ticks (23.4%) was significantly higher compared to Rh. sanguineus s.l. (9.5%). This result is in line with those presented by Ghafar et al. (39) indicating a higher prevalence of R. massiliae and R. slovaca infections in Rh. turanicus ticks from Pakistan compared to other tick species. Furthermore, risk factor analysis showed that ticks collected from goats (23.2%) were more infected with Rickettsia spp. than those infesting sheep (7.9%) which is consistent with the same result of Ghafar et al. (39) in Pakistan.
In this study, R. massiliae was detected in Rh. turanicus and Rh. sanguineus s.l., thus confirming its occurrence especially in the north of Tunisia. In our country, previous studies have reported the presence of R. massiliae in Rh. sanguineus s.l. ticks collected from sheep situated in the center (6) and more recently in camels located in the center and the south (8). Similarly, R. massiliae has been also identified in Rh. turanicus and Rh. sanguineus s.l. from Algeria, Italy, Cyprus, and Greece (15,34,40), in Rh. sanguineus s.l. ticks from Morocco (41,42), Spain, and Italy (12,43), and in Rh. turanicus ticks from China (36) and Pakistan (39).
In the present study, R. monacensis DNA was detected in Rh. sanguineus s.l. tick specimens removed from goats. These results consolidate previous data describing the presence of this bacterium in questing I. ricinus ticks (9), and in camels and their infesting H. impeltatum ticks (8). Besides, wide geographical distribution of this pathogen was noted particularly in the Mediterranean region (Italy and Spain) and from other countries like Costa Rica and Nicaragua (44)(45)(46). Interestingly, this species was identified as a zoonotic pathogen able to cause from moderate to severe illness in humans (19). The detection in Tunisia of R. monacensis DNA in Rh. FIGURE 4 | Phylogenetical relationships based on nucleotide multiple alignments of partial Rickettsia spp. gltA sequences (341 bp). Numbers over the branches indicate the percentage of replicated trees in which the associated taxa clustered together in the bootstrap test (1,000 replicates, only percentages >50% were represented). The only R. massiliae gltA genotype revealed in this study from 25 positive samples is represented in bold. The host or vector, the genotype, sequence type, strain or isolate name, the country of origin, and the GenBank accession number are indicated. One R. prowazekii gltA partial sequence was added as an outgroup. sanguineus s.l. ticks collected from goats suggests that, even if the circulation in the environment is essentially maintained by I. ricinus ticks, there may be other species incriminated in the transmission of this bacterial species as suggested in other reports from several countries (19,47). Our findings highlight the need of extensive studies in the Rh. sanguineus s.l. tick complex collected from small ruminants and other domestic animals principally dogs to assess and predict the potential risks for humans.
However, given the growing occurrence of novel Rickettsia species with unidentified pathogenicity, it will be essential to carry out supplementary genetic characterization of the revealed Rickettsia spp. by using a combination of genetic markers such as ompA, and gltA, in addition to the ompB gene. In the present study, phylogenetic trees based on the three gene fragments showed higher genetic diversity among the revealed R. massiliae isolates by using ompA and ompB genes compared to the gltA gene. This result is in line with those presented by Ereqat et al. (11) and Chisu et al. (48) investigating Palestinian and Sardinian ticks, respectively.
By analyzing ompB partial sequences, two genotypes (ompBRmasRs1 and ompBRmasRt1) infecting Rh. turanicus and Rh. sanguineus s.l. tick specimens were found similar to that isolated from R. massiliae strain MTU5 (CP000683) recovered from Rh. turanicus ticks collected on horses in Camargues, France (49), suggesting its potential spread in several Mediterranean countries. The remaining genotypes (ompBRmasRs2 and ompBRmasRt2) also infecting both tick species were found identical to R. massiliae Bar29 (AF123710) earlier identified in Rh. sanguineus s.l. ticks from Spain based on the same gene (50) and from Tunisia based on the 23S-5S intergenic spacer (6). Additionally, on the basis of the ompA phylogenetic tree, we found that R. massiliae isolated from Rhipicephalus ticks showed genetic divergence with novel genotypes, which indicates that these isolates infecting different tick species may come from various origins, hosts, and reservoirs. Thus, this finding needs to be further investigated.
Based on ompB phylogeny, low genetic diversity was observed among R. monacensis genotypes identified in this study. Indeed, one genotype (ompBRmonRs1) was found to be 100% similar to the corresponding sequence of R. monacensis strain CN45Kr (EU883092) infecting a patient from South Korea (51), revealing its widespread distributions and potential risk for human. Thus, for a more accurate classification of our revealed R. monacensis isolates, further testing and phylogenetic analysis with additional genes are needed since no sequences of the two other genes isolated from this Rickettsia species were obtained in this study.
Therefore, the observation of these two zoonotic Rickettsia species, R. massiliae and R. monacensis, in investigated regions indicates a possible threat to resident humans. Indeed, infected tick species can also infest various domesticated animals and therefore constitute a possible risk for transmission of SFG rickettsiae to humans (3). However, the pathogenicity of this bacterium to humans is not well-understood (48). Consequently, supplementary trials are needed to investigate the pathogenicity of the revealed Rickettsia species and whether found tick species can transmit these pathogens in humans.

CONCLUSIONS
The present study confirms the occurrence of human-pathogenic Rickettsia species in Rh. sanguineus s.l. and Rh. turanicus ticks collected from small ruminants in Tunisia. Our findings expand knowledge on ticks collected from domestic animals and highlight the range of infectious agents that may be transmitted by ticks to humans and animals.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/supplementary material.

ETHICS STATEMENT
The animal study was reviewed and approved by The Ethics Committee of the National School of Veterinary Medicine of Sidi Thabet, University of Manouba. Written informed consent was obtained from the owners for the participation of their animals in this study.

AUTHOR CONTRIBUTIONS
HB, LM, and MB conceived the idea. HB and MD-J carried out the fieldwork. HB, RS, and SZ performed the experiments.
HB and MB performed risk factor analysis, genotyping, and phylogenetic study. HB and MB wrote the manuscript and HB, RS, LM, and MB finalized it. All authors read and approved the final version.