Molecular identification of zoonotic Rickettsia species in Ixodidae parasitizing wild lagomorphs from Mediterranean ecosystems

Received: 29 June 2021 Revised: 1 October 2021 Accepted: 30 October 2021 DOI: 10.1111/tbed.14379 ORIGINAL ARTICLE Molecular identification of zoonotic Rickettsia species in Ixodidae parasitizing wild lagomorphs from Mediterranean ecosystems Susana Remesar1Sabrina Castro-Scholten2David Cano-Terriza2 Pablo Díaz1Patrocinio Morrondo1Débora Jiménez-Martín2 Carlos Rouco3Ignacio García-Bocanegra2 1Investigación en Sanidad Animal: Galicia (Grupo INVESAGA), Facultade de Veterinaria, Universidade de Santiago de Compostela, Lugo, Spain 2Animal Health and Zoonosis Research Group (GISAZ), Department of Animal Health, University of Cordoba, Córdoba, Spain 3Departamento de Botánica, Ecología y Fisiología Vegetal, Universidad de Córdoba, Córdoba, Spain Correspondence PabloDíaz,InvestigaciónenSanidadAnimal: Galicia(GrupoINVESAGA),Facultadede Veterinaria, UniversidadedeSantiagode Compostela,Lugo, Spain. Email:[email protected] Funding information Ministeriode Cienciae Innovación(España), Grant/AwardNumber:PID2019-111080RB- C21;Universidadde Córdoba(España), Grant/AwardNumber:UCO-FEDER-1264967; Ministeriode Ciencia,Innovacióny Universidades(España), Grant/AwardNumber: FPU19/06026;European SocialFund;Ministeriode Ciencia,Innovacióny Universidades; UniversidaddeCórdoba; Ministeriode Ciencia eInnovación Abstract A survey study was carried out to identify tick species parasitizing wild lagomorphs in Mediterranean ecosystems in southern Spain and to determine the occurrence of Rickettsia species present in these ticks in this region. A total of 1304 European wild rabbits (Oryctolagus cuniculus) and 58 Iberian hares (Lepus granatensis) were individually examined for the presence of ticks. Ticks were found in 42.9% and 50% of the wild rabbits and hares sampled, respectively. A total of 1122 ticks were collected and five species,includingRhipicephalus pusillus,Hyalomma lusitanicum,Haemaphysalis hispanica, Ixodes ventalloi and Rhipicephalus sanguineus sensu lato (s.l.), were microscopically and molecularly identified at the 16S rRNA gene. This is the first study on Ixodidae parasitizing Iberian hares. The presence of Rickettsia DNA was assessed in 254 tick pools (according to hunting states, lagomorph species, tick species and tick development stage) using PCR assays targeting the rOmpA, rOmpB and gltA. Twenty-seven pools (10.6%) were positive to Rickettsia DNA. Five zoonotic Rickettsia species were identified, being Rickettsia massiliae the most frequent (4.7%), followed by Rickettsia sibirica subsp. mongolitimonae (2.8%), Rickettsia slovaca (2.0%), Rickettsia aeschlimannii (0.8%) and Rickettsia africae (0.4%). The results suggest that wild rabbits and Iberian hares are parasitized by a wide range of tick species and that these lagomorphs may play an important role in the sylvatic cycle of some zoonotic Rickettsia species in Mediterranean ecosystems. Our data represent the first report of R. massiliae, R. aeschlimannii, R. slovaca and Rafricaein ticks collected in wild lagomorphs in Europe, and the first report of not imported R. africae in this continent. Since R. slovaca and R. africae DNA was detected in tick species different to their main vectors, further studies are warranted to unravel the role of wild lagomorphs in the epidemiology of these vectorborne pathogens. KEYWORDS Iberian hare, Rickettsia africae, Rickettsia slovaca, Spain, wild rabbit, zoonotic This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2021 The Authors. Transboundary and Emerging Diseases published by Wiley-VCH GmbH. e992 wileyonlinelibrary.com/journal/tbed Transbound Emerg Dis. 2022;69:e992–e1004. REMESAR ET AL.e993 1INTRODUCTION The European wild rabbit (Oryctolagus cuniculus) and the Iberian hare (Lepus granatensis) are native mammals of the Iberian Peninsula (Amori et al., 2008). They play an important role in the ecology and diversity of the Spanish Mediterranean ecosystems since they are staple preys for more than 40 predators including the threatened Spanish imperial eagle (Aquila adalberti) and the Iberian lynx (Lynx pardinus) (Delibes-Mateos et al., 2008). In addition, these lagomorphs may also represent significant animal and public health concerns for several reasons. First, both species show a wide distribution in Spain since they are very adaptable to different ecological conditions, sharing habitat and resources with other sympatric domestic and wild species, as well as humans. Second, the wild rabbit and the Iberian hare are among the main small game species in Spain, being generally consumed without sanitary inspection (Gortázar et al., 2015). Third, wild lagomorphs have shown to be natural carriers or reservoirs of different zoonotic pathogens, including vector-borne pathogens (González-Barrio et al., 2015; Jiménez et al., 2014). Previous investigations reported that Rhipicephalus pusillus, Hyalomma lusitanicum, Haemaphysalis hispanica and Ixodes ventalloi are the major tick species detected in wild rabbits from Spain (Ciceroni et al., 1988; Gilot & Aubert, 1985; Gonźalez et al., 2016; Lópes de Carbalho et al., 2016; Pereira et al., 2018; Varcárcel et al., 2015). Up to now, available information on the presence of ectoparasites in Iberian hares is lacking. Two previous studies carried out on European hares (Lepus europaeus) from northern Spain revealed that Ixodes ricinus is the most common tick species detected in this hare species, followed by Rhipicephalus spp. and Haemaphysalis spp. (Alzaga et al., 2009; Astobiza et al., 2011). All these results have an important epidemiological significance since some of these ticks are considered competent vectors of zoonotic pathogens such as Anaplasma phagocytophilum, Crimean–Congo haemorrhagic fever virus and some Rickettsia species (Estrada-Peña et al., 2012; Márquez, 2008). Concerning the presence of Rickettsia infections in humans in Europe, Mediterranean spotted fever (MSF), which is endemic in the Mediterranean basin, is the most frequent rickettsiosis (Parola et al., 2005). Although Rickettsia conorii subspecies are the main etiological agents of MSF, other Rickettsia species such as Rickettsia monacensis, Rickettsia massiliae and Rickettsia aeschlimannii have also been reported in Spanish patients with similar clinical manifestations (Oteo & Portillo, 2012). In addition, other zoonotic Rickettsia species such as Rickettsia slovaca and Rickettsia sibirica subsp. mongolitimonae have been detected in humans in Spain (Aguirrebengoa et al., 2008; Oteo & Ibarra, 2002). However, data on the presence and identification of Rickettsia species in ticks from wild lagomorphs in Europe are very limited, and the role of these mammals on the sylvatic epidemiological cycles of these tickborne pathogens is unknown. In this respect, previous studies carried out in America suggested that wild lagomorphs may play an important role in the maintenance of some Rickettsia species due to their long-lasting asymptomatic rickettsiaemia (Rovery et al., 2008). Even though in this continent some Rickettsia strains detected in ticks collected from rabbits were identified as avirulent (Eremeeva et al., 2018; Roth et al., 2017), Rickettsia rickettsii strains identical to those previously detected in human clinical samples were also identified in Haemaphysalis leporispalustris collected from these lagomorphs in Costa Rica (Hun et al., 2008). The aims of the present study were (a) to identify tick species parasitizing wild rabbits and Iberian hares in Mediterranean ecosystems in southern Spain and (b) to detect and identify the Rickettsia species present in ticks parasitizing these wild lagomorph species. 2MATERIALS AND METHODS 2.1 Study area The present study was carried out in Andalusia (southern Spain) (36 ◦N-38◦60′N, 1◦75′W-7◦25′W) between 2018 and 2020. This region has a Mediterranean climate characterized by dry summers and mild, humid winters; the average annual temperature is 16◦Cand the average annual precipitation is 590 mm (CMAOT, 2009). However, the presence of noticeable climatic differences in this region allowed establishing six bioclimatic regions: Mediterranean oceanic (MedOc), sub-continental Mediterranean with warm winters (SubMedWa), subcontinental Mediterranean with cold winters (SubMedCo), Mediterranean sub-desert (MedDes), Mountainous (Mount) and Mediterranean sub-tropical (MedSubTro) (CMAOT, 2009). 2.2 Wild lagomorphs sampling A total of 1304 wild rabbits and 58 Iberian hares were individually examined in the field for the presence of ticks. Rabbits and hares were legally hunted in 60 and 29 hunting states (limited geographical areas where recreational hunting activities are allowed under the government legislation), respectively, located in four of the six bioclimatic regions: MedOc (seven and five hunting states for rabbits and hares, respectively), SubMedWa (10 and 10 hunting states), SubMedCo (34 and 13 hunting states) and MedDes (nine and one hunting states) (Figure 1a,d). 2.3 Tick collection and identification Atotalof 1122tickswere collectedusingtweezers, individuallybagged up and labelled. Samples were kept frozen at −20◦C until examination. Tick specimens were identified to species level using morphological keys (Pérez-Eid, 2007). In order to confirm the microscopic identifications, a subset of each tick species was further molecularly analyzed (Table 1). First, ticks were disrupted using a MagNaLyser Instrument (Roche Diagnostic, Manheim, Germany) at 6000 rpm for 60 s. DNA was then extracted using a commercial kit (High Pure PCR Template Preparation Kit, Roche Diagnostics GmbH, Mannheim, Germany) following the manufacturer’s instructions. A 460 bp fragment of the 16S rRNA gene of ticks was amplified using previously reported PCR protocols e994 REMESAR ET AL. FIGURE 1 Distribution of tick (a and c) and Rickettsia (b and d) species in wild rabbits (Oryctolagus cuniculus) (a and b) and Iberian hares (Lepus granatensis) (c and d) from southern Spain TABLE 1 Total number of ticks captured, and pools processed from wild rabbits (Oryctolagus cuniculus) and Iberian hares (Lepus granatensis) from southern Spain Larvae Nymph Female Male Total Host species Number of ticks Number of pools Number of ticks Number of pools Number of ticks Number of pools Number of ticks Number of pools Number of ticks Number of pools Wild rabbit Rhipicephalus pusillus 18 – 407 65 251 54 221 50 897 169 Rhipicephalus sanguineus s.l.0– 0 – 0– 32 32 Haemaphysalis hispanica 0 – 26 13 0 – 0 – 26 13 Hyalomma lusitanicum 0 – 93 23 0 – 0 – 93 23 Ixodes ventalloi 0– 0 – 74 32 106 Total 18 – 526 101 258 58 227 54 1029 213 Iberian hare Rhipicephalus pusillus 1 – 5 3 20 11 22 10 48 24 Rhipicephalus sanguineus s.l. 0 – 0 – 0 – 25 11 25 11 Haemaphysalis hispanica 0 – 0 – 0 – 0 – 0 0 Hyalomma lusitanicum 0 – 16 4 4 2 0 – 20 6 Ixodes ventalloi 0 – 0 – 0 – 0 – 0 0 Total 1 – 21 724 13 47 21 93 41 Total 19 – 547 108 282 71 274 75 1122 254 (Norris et al., 1996; Simon et al., 1994). PCR products were separated by electrophoresis on 1.5% agarose gels stained with RedSafe (iNtRON Biotechnology, South Korea) and visualized using a Fluor-S MultiImager (Bio-Rad Laboratories, California, USA). Selected fragments were purified and sequenced on an ABI 3730xl sequencer (Applied Biosystems, Foster City, CA, USA) at the Sequencing and Fragment Analysis Unit of the Santiago de Compostela University. Sequences were aligned and edited using ChromasPro (Technelysium, Brisbane, Australia), and consensus sequences were then scanned against the Gen- Bank database using BLAST. Unique 16S rRNA sequences identified in this study were deposited in GenBank under accession numbers MZ420711–MZ420717. REMESAR ET AL.e995 A phylogenetic analysis was carried out using MrBayes 3.2.7 software (Ronquist et al., 2012) by Bayesian approach with Markov Chain Monte Carlo sampling (10,000,000 generations sampling every 1000 steps). A general time reversible substitution model with gammadistributed rate variation across sites (GTR +G) was used for the analysis of tick sequences at the 16s rRNA gene. The model was selected based on Akaike information criterion (AIC) value using the free software jModelTest v.2.1.10 (Darriba et al., 2012). The tree was visualized and edited using FigTree 1.4.3 (http://tree.bio.ed.ac.uk/software/ figtree/). 2.4 Rickettsia detection in ticks For Rickettsia spp. detection, ticks from the same hunting state were pooled including those specimens of the same species and development stage. Ticks from rabbits and hares were independently pooled. In addition, if the number of ticks of the same hunting state, species and development stage was higher than 10, more than one pool was prepared. The number of ticks included in each pool is summarized in Table 1. DNA was extracted using the same commercial kit previously indicated (High Pure PCR Template Preparation Kit; Roche Diagnostics GmbH, Mannheim, Germany). The presence of Rickettsia DNA in tick samples was detected by using two nested PCR assays targeting the genes encoding for two major outer membrane proteins of Rickettsia spp. (rOmpA and rOmpB) using previously reported primers and protocols (Choi et al., 2005; Regnery et al., 1991; Roux et al., 1996). All samples were analyzed in parallel using both PCR protocols. In addition, in order to increase the diagnostic sensibility, positive samples to rOmpA and/or rOmpB were further analyzed using a third PCR assay targeting the gltA of Rickettsia spp. (Labruna et al., 2004). In each assay, DNA of Rickettsia amblyommatis, a species absent in Europe, and nuclease free water were included as positive and negative controls, respectively. PCR products were processed and sequenced as previously indicated. Partial Rickettsia spp. sequences identified in this study were deposited under accession numbers MZ420707–MZ420710 and OK205202– OK205222. Finally, the obtained sequences were scanned against the GenBank database. A phylogenetic analysis of the partial rOmpA was carried out as previously indicated. A Hasegawa–Kishino–Yano substitution model with gamma-distributed rate variation across sites (HKY+G) was used for the analysis of Rickettsia spp. sequences at the rOmpA.The model was selected based on AIC value using the software jModelTest v.2.1.10 (Darriba et al., 2012). 2.5 Statistical analysis The association between the total number of tick specimens when considering the development stage (larvae, nymph and adult) was assessed using an analysis of variance (ANOVA) test. The maximum likelihood estimation (MLE) was used to estimate the prevalence of Rickettsia spp. in pooled ticks (Williams & Moffitt, 2005). The possible influence of tick species and development stage, lagomorph species and bioclimatic area on the Rickettsia spp. infection was analyzed by multivariate analysis using a multiple logistic regression model as described by Hosmer and Lemeshow (1989). Factors were eliminated from the initial model using a method based in AIC value until the best model was built. All pairwise interactions were evaluated. Odds ratio were computed by raising ‘e’ to the power of the logistic coefficient over the first category of each factor (reference category). All statistical analyses were performed using the statistical softwareR(RCoreTeam,2020), and the level of significance was set at pvalues <.05. A quantum geographic information system software was used to depict spatial distribution of ticks and Rickettsia species in the different hunting states and bioclimatic regions (QGIS.org, 2021). 3RESULTS 3.1 Tick species parasitizing wild lagomorphs Ticks were found in 560 of the 1304 (42.9%) and 29 of the 58 (50.0%) wild rabbits and Iberian hares examined, respectively. A total of 1029 tick specimens were collected in wild rabbits; nymphs and adults were predominant, whereas larvae were only occasionally found (Table 1). Significant differences were only detected between the number of nymphs and larvae (F=8.225; p=.004). Ninety-three Ixodidae were recovered from hares, being the number of adults higher than that of nymphsand larvae(Table1),but thesedifferenceswerenot statistically significant (p>.05). Five different tick species were identified in the lagomorphs examined; all of them were found in rabbits and three in hares (Table 1, Figure 1a,c). The number of specimens collected according to the tick species and the development stage is summarized in Table 1.R. pusillus was the most abundant tick species in both wild rabbits and Iberian hares. Co-infestations with two types of tick species were detected in both lagomorph species. In rabbits, two co-infestations were detected R. pusillus/H. lusitanicum (6.6%; 86/1304) and R. pusillus/H. hispanica (2.0%; 26/1304). In hares, the most commonly co-infestation detected was R. pusillus/Rhipicephalus sanguineus s.l. (8.6%; 5/58) followed by the associations H. lusitanicum/R. pusillus (3.4%; 2/58) and H. lusitanicum s.l./R. sanguineus s.l. (3.4%; 2/58). Sequence analysis confirmed the morphological identification of all tick species except for H. hispanica because no sequences of this species are currently deposited in the GenBank database. The sequence obtained in the present study showed a homology of 91.8% and 91.3% with those of Haemaphysalis sulcata (MT799946.1) and Haemaphysalis longicornis (MT555306.1), respectively. In addition, the phylogenetic analysis showed that our sequence formed a clade separated from other Haemaphysalis species deposited in GenBank (Figure 2). R. pusillus was the only species found in different development stages in the same rabbit being the co-infestation nymph/adult (19.5%; 254/1304) the most commonly detected, followed by larvae/nymph (3.5%; 45/1304) and larvae/adult (3.5%; 45/1304). The e996 REMESAR ET AL. FIGURE 2 Phylogenetic tree clustering of the partial 16S rRNA gene of the ticks. The tree was obtained using a general time reversible substitution model with gamma-distributed rate variation across sites (GTR +G) with the software MrBayes 3.2.7 (Ronquist et al., 2012)by Bayesian approach with Markov Chain Monte Carlo sampling (10,000,000 generations sampling every 1000 steps). This analysis involved 48 nucleotide sequences. The nucleotide sequence of Argas persicus was used as an outgroup. Isolates identified in this study (*) percentage of rabbits parasitized by all the development stages of this species was low (1.2%; 15/1304). In hares, no co-infestation with different development stages of the same tick species were observed. Ticks were collected from wild rabbits and hares from all sampled hunting states, but differences in the geographical distribution of tick species were found (Figure 1a,c). Thus, I. ventalloi was the less spread tick since it was only detected in hunting states with MedOc and Sub- MedCo climate (Table 2). The remaining tick species were found in MedOc, SubMedCo and SubMedWa areas, but only R. pusillus and H. lusitanicum were collected in areas with MedDes climate (Figure 1a,c;). R. pusillus shown the widest distribution on wild lagomorphs from southern Spain, since it was found in the 95% and 80% of the hunting states where rabbits and hares were sampled, respectively (Figure 1a,c). 3.2 Rickettsia species in ticks from wild lagomorphs 3.2.1 Percentage of pools positive to Rickettsia spp Rickettsia DNA was detected in 27 out of 254 (10.6%) tick pools. Rickettsia-positive pools were found in 22 out of 79 (27.9%) hunting states (Figure 1b,d). The percentage of positive pools was higher in Iberian hares (11/41; 26.8%) than in wild rabbits (16/213; 7.5%), and the logistic regression results (Table 3) showed that these differences were statistically significant [OR =5.5; 95% confidence interval (CI): 1.5–22.1; p=.012]. Only four (14.8%) of the 27 Rickettsia-positive pools were positive to all the three studied genes. Most positive pools (92.6%; 25/27) were detected using the PCR protocol targeting the rOmpA gene, whereas amplification was only observed in 8 (29.6%) and 14 (51.9%) pools when analyzed at the rOmpB and gltA genes, respectively. The highest percentage of positivity was detected in pools composed of R. sanguineus s.l. (38.5%; 5/13) and H. lusitanicum (37.9%; 11/29), followed by H. hispanica (7.7%; 1/13) and R. pusillus (5.2%; 10/193). No I. ventalloi pools were positive to Rickettsia spp. (Table 2). The logistic regression showed that the number of H. hispanica positive pools was significantly lower than H. lusitanicum positive pools (OR =0.1; 95%CI: 0.03– 0.4; p<.001). Similarly, a significantly lower percentage of R. pusilluspositive pools was detected when compared to H. lusitanicum-positive pools (OR =0.09; 95%CI: 0.01–0.6; p=.040). The multivariate analysis (Table 3) also identified a significantly higher number of Rickettsiapositive pools in SubMedWa area than in SubMedCo (OR =5.4; 95%CI: 1.7–20.2; p=.007). 3.2.2 Rickettsia species identified Five different Rickettsia species were molecularly identified in the tick pools (Table 4,Figure1b,d). The most commonly detected Rickettsia species was R. massiliae (4.7%; 12/254), followed by R. sibirica subsp. mongolitimonae (2.8%; 7/254), R. slovaca (2.0%; 5/254), R. aeschlimannii (0.8%; 2/254) and Rickettsia africae (0.4%; 1/254). No coinfections were found in the tick pools analyzed. All the Rickettsia species identified were found in ticks collected from wild rabbits except R. africae, being R. sibirica subsp. REMESAR ET AL.e997 TABLE 2 Percentage of pools positive to Rickettsia spp. and maximum likelihood estimation (MLE) from wild rabbits (Oryctolagus cuniculus) and Iberian hares (Lepus granatensis) when considering the tick development stage and the climatic region Stage of development Sampling area Total Nymphs Female Male MedOc SubMedWa SubMedCo MedDes % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) % Wild rabbit Rhipicephalus pusillus 5/65 (7.7%) 1.3 (0.5–2.8) 1/54 (1.9%) 0.40 (0.0–1.8) 1/50 (2%) 0.5 (0.0–2.0) 0/13 (0%) –5/32 (15.6%) 2.53 (0.9–5.4) 2/87 (2.3 %) 0.52 (0.1–1.6) 0/37 (0%) – 7/169 (4.1%) Rhipicephalus sanguineus s.l. – – – – 0/2 (0%) –––0/2 (0%) –––––0/2 (0%) Haemaphysalis hispanica 1/13 (7.7%) 3.9 (0.2–16.2) –– –– 0/5 (0%) –1/5 (20%) 7.42 (0.4–28.9) 0/3 (0%) –––1/13 (7.7%) Hyalomma lusitanicum 8/23 (34.8%) 9.2 (4.3–16.5) –– –– 2/8 (25%) 4.37 (0.7–12.9) 2/2 (100%) 100 (25.2– 100) 2/9 (22.2%) 6.90 (1.2–19.8) 2/4 (50%) 21.18 (3.9–52.5) 8/23 (34.8%) Ixodes ventalloi –– 0/4 (0%) 0/2 (0%) 0/4 (0%) –– 0/2 (0%) –– 0/6 (0%) Iberian hares Rhipicephalus pusillus 1/3 (33.3%) 26.3 (1.7–78.0) 2/11 (18.2%) 10.90 (1.9–30.2) 0/10 (0%) –1/4 (25%) 14.50 (0.9–50.9) 2/6 (33.3%) 29.5 (5.5–68.0) 0/11 (0%) –0/3 (0%) –3/24 (12.5%) Rhipicephalus sanguineus s.l. – – – – 5/11 (45.5%) 20.5 (7.9–39.0) 2/2 (100%) 100 (38.3– 100) 2/6 (33.3%) 11.5 (2.0–31.4) 1/3 (33.3%) 20.00 (1.3–62.8) – – 5/11 (45.5%) Hyalomma lusitanicum 3/4 (75%) 52.4 (12.0– 96.3) 0/2 (0%) –––1/3 (33.3%) 20.63 (1.2–73.7) – – 2/3 (66.7%) 29.00 (4.9–73.1) – – 3/6 (50%) Abbreviations: CI, confidence interval; MedDes, Mediterranean sub-desert; MedOc, Mediterranean oceanic; SubMedCo, sub-continental Mediterranean with cold winters; SubMedWa, sub-continental Mediterranean with warm winters. e998 REMESAR ET AL. TABLE 3 Logistic regression model for the prevalence of Rickettsia spp. in ticks collected in both wild rabbits (Oryctolagus cuniculus) and Iberian hares (Lepus granatensis) from southern Spain Estimate Z-value p-Value OR CI 95% (Intercept) 1.881 1.968 .050 – – Lagomorph species Wild rabbit – – – – – Iberian hare 1.698 2.512 .012 5.46 1.49–22.11 Tick species H. lusitanicum ––––– R. pusillus −2.449 −2.055 .040 0.09 0.01–0.64 H. hispanica −2.161 −3.340 <.001 0.12 0.03–0.40 Bioclimatic area SubMedCo – – – – – SubMedWa 1.693 2.703 .007 5.44 1.67–20.15 Abbreviations: CI, confidence level; OR, odds ratio; SubMedCo, sub-continental Mediterranean with cold winters; SubMedWa, sub-continental Mediterranean with warm winters. mongolitimonae the most frequently detected in this lagomorph species (Table 4). In contrast, R. massiliae was the major Rickettsia species detected in ticks collected from Iberian hares; it is worth noting that R. sibirica subsp. mongolitimonae was not detected in this lagomorph (Table 4). All the identified Rickettsia species were found in H. lusitanicum and all but R. africae were detected in R. pusillus. In addition, only a single Rickettsia species was identified in R. sanguineus s.l. (R. massiliae) and H. hispanica (R. sibirica subsp. mongolitimonae). Rickettsia-DNA was not detected in I. ventalloi (Table 4). Through the study of rOmpA (Figure 3), R. massiliae and R. aeschlimannii sequences showed a percentage of identity higher than 99.6% with the sequences obtained in Greece (MG521363), China (MF098409, KU365969 and MF098409) and Kenya (KX227782) and a percentage of identity higher than 99.4% with the reference sequences U43799 and U43800 of R. massiliae and R. aeschlimannii, respectively (Fournier et al., 2003). The obtained sequence of R. africae was identical to that obtained in Kenya (KX227785) and showed a percentage of identity of 99.0% with the reference sequence U43790 (Fournier et al., 2003). The sequence of R. sibirica subsp. mongolitimonae was identical to the obtained in Turkey (MF379309) and to the reference sequence U43796 (Fournier et al., 2006). Finally, the sequences of R. slovaca were identical to those obtained in Turkey (MH548522), Italy (HM161798) and China (MG598413) and showed a percentage of identity higher than 99.6% with the reference sequence U43808 (Fournier et al., 2003). Sequence analysis at the rOmpB and gltA showed that all sequences were identical to those obtained in questing R. turanicus (KY233281, KY233273 and MF002497) and Hyalomma spp. (KX227786, HM050273, KY233273 and KU961540) from Lebanon, Senegal, Kenya and China; in I. ricinus (MK301603 and MK301607) from Spain and in Dermacentor nuttalli (MF002541) from China. In addition, rOmpB sequences showed a percentage of identity higher than 99.4% with the reference sequences AF123714, AF123705, AF123706, DQ097083 and AF123723 of R. massiliae,R. aeschlimannii, R. africae, R. sibirica subsp. mongolitimonae and R. slovaca (Fournier et al., 2003). 4DISCUSSION 4.1 Ixodidae fauna of wild lagomorphs The results obtained in the present study reveal that tick infestation is very common in the wild rabbit and Iberian hare populations from southern Spain since more than 42% of the examined animals were parasitized, and tick-positive lagomorphs were found in all sampled hunting states. Previous investigations carried out in northern Spain reported lower prevalences than those observed in our study since tickswere only recoveredfrom the17%of wildrabbitsand the14–31% of European hares (Alzaga et al., 2009; Astobiza et al., 2011). In contrast, a higher prevalence has been reported in Sicily (southern Italy), where ticks were found on the 75.2 % of the sampled wild rabbits (Napoli et al., 2021). It is worth noting that no significant differences were observed between the number of adults and immature ticks found in both wild lagomorph species, whereas most investigations performed in other European countries reported a predominance of tick immature stages (Dantas-Torres et al., 2011; González et al., 2016; Pereira et al., 2018; Talleklint & Jaenson, 1997). Nevertheless, it is important to point out that the tick species detected, their ecological preferences and behaviour, the time of the year when the animals were sampled (Napoli et al., 2021; Sobrino et al., 2012) as well as the difficulties in capturing tick immature stages due to its size in field conditions, are possible factors that may explain the discrepancies observed among studies. Five different tick species were found in wild rabbits and Iberian hares in the present study. Our results are consistent with the data REMESAR ET AL.e999 TABLE 4 Percentage of pools positive to Rickettsia species detected and maximum likelihood estimation (MLE) from wild rabbits (Oryctolagus cuniculus) and Iberian hares (Lepus granatensis)from southern Spain Rickettsia sibirica subsp. mongolitimonae Rickettsia massiliae Rickettsia slovaca Rickettsia africae Rickettsia aeschlimannii Total % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) % MLE (95%CI) Wild rabbit Rhipicephalus pusillus 5/169 (3.0%) 0.6 (0.2– 1.2) 1/169 (0.6%) 0.1 (0.0–0.5) 0/169 (0%) – 0/169 (0%) – 1/169 (0.6%) 0.1 (0.0–0.5) 7/169 (4.1%) 0.8 (0.3–1.5) Rhipicephalus sanguineus s.l. 0/2 (0%) –0/2 (0%) –0/2 (0%) –0/2 (0%) –0/2 (0%) –0/2 (0%) – Haemaphysalis hispanica 1/13 (7.7%) 3.9 (0.2–16.2) 0/13 (0%) –0/13 (0%) –0/13 (0%) –0/13 (0%) –1/13 (7.7%) 3.9 (0.2–16.2) Hyalomma lusitanicum 1/23 (4.4%) 1.1 (0.1–4.7) 3/23 (13.0%) 3.4 (0.9–8.5) 4/23 (17.4%) 4.4 (1.4–9.4) 0/23 (0%) –0/23 (0%) –8/23 (34.78%) 9.20 (4.3–16.5) Ixodes ventalloi 0/6 (0%) –0/6 (0%) –0/6 (0%) –0/6 (0%) –0/6 (0%) –0/6 (0%) – Iberian hares Rhipicephalus pusillus 0/24 (0%) –2/24 (8.3%) 4.4 (0.7–12.8) 1/24 (4.2%) 2.2 (0.1–9.4) 0/24 (0%) –0/24 (0%) –3/24 (12.5%) 6.68 (1.7–16.4) Rhipicephalus sanguineus s.l. 0/11 (0%) –5/11 (45.5%) 20.5 (7.9–39.0) 0/11 (0%) –0/11 (0%) –0/11 (0%) –5/11 (45.5%) 20.45 (7.9–39.0) Hyalomma lusitanicum 0/6 (0%) –1/6 (16.7%) 5.0 (0.3–20.2) 0/6 (0%) –1/6 (16.7%) 5.8 (0.3–23.4) 1/6 (16.7%) 6.19 (0.4–25.3) 3/6 (50%) 25.5 (6.5–59.5) e1000 REMESAR ET AL. FIGURE 3 Phylogenetic tree clustering of the partial rOmpA of Rickettsia spp. The tree was obtained using a Hasegawa–Kishino–Yano substitution model with gamma-distributed rate variation across sites (HKY +G) with the software MrBayes 3.2.7 (Ronquist et al., 2012)by Bayesian approach with Markov Chain Monte Carlo sampling (10,000,000 generations sampling every 1000 steps). This analysis involved 34 nucleotide sequences. The nucleotide sequence of Rickettsia felis was used as an outgroup. Isolates identified or identical to those identified in this study (*) available for wild rabbits and European hares from Europe (Ciceroni et al., 1988; Gilot & Aubert, 1985; Gonźalez et al., 2016; Lópes de Carbalho et al., 2016; Pereira et al., 2018; Varcárcel et al., 2015). Nevertheless, noticeable differences on the presence and abundance of tick species in wild lagomorphs have been reported in Europe, probably related to their environmental preferences. Thus, I. ricinus is the most frequent Ixodidae in wild lagomorphs from northern European countries (Buttler et al., 1994; Gilot & Aubert, 1985; Talleklint & Jaenson, 1997) since it is sensitive to desiccation and extreme heat and needs a moist microclimate to live (A. R. Walker et al., 2009). In contrast and agreeing with our findings, the most abundant tick species in wild lagomorphs from the Mediterranean basin were H. lusitanicum and R. pusillus (Astoriza et al., 2011; Ciceroni et al., 1988; Dantas-Torres et al., 2011; González et al., 2016; Lopes de Carballo et al., 2016;Orkun& Karaer, 2017; Pereira et al., 2018; Psaroulaki et al., 2014;Varcarcel et al., 2015), which are adapted to dry Mediterranean ecosystems (J. B. Walker et al., 2000). The most abundant and distributed tick species in wild rabbits and Iberian hares from the study area, R. pusillus and H. lusitanicum, feed mainly on lagomorphs at least in one stage of development (Estrada- Peña et al., 2017). However, while all development stages of R. pusillus primary feed on lagomorphs (Guglielmone & Nava, 2014), only immature stages of H. lusitanicum feed on these species (Varcárcel et al., 2020). This could explain the large number of nymphs of the latter species found in the present study. Although H. hispanica and I. ventalloi are also endophilic species, which mainly parasite lagomorphs, they were not found in MedDes regions since they are adapted to Iberian Mediterranean sclerophyllous and mixed ecoregion (Estrada- Peña et al., 2017), which are not present in MedDes bioclimatic regions of the study area. R. sanguineus s.l. is a nidicolous tick well adapted to a wide range of climatic conditions which feed mainly in dogs; however, it can also be found in other hosts (Dantas-Torres, 2010). The diversity of Ixodidae found in Iberian hares was lower than in wild rabbits since I. ventalloi and H. hispanica were not detected in this hare species. I. ventalloi has been previously found in European hares from Cyprus (Psaroulaki et al., 2014). Although this tick species is mainly a parasite of wild rabbits (Estrada-Peña et al., 2017), in the present study, a low number of specimens were found in this lagomorph. Therefore, its absence in Iberian hares may be related to the low number of examined individuals. It should be noted that the number of hares hunted in the studied area is 10-fold lower than the number of rabbits (CAPDS, 2019). To the best of our knowledge, H. hispanica has not been previously detected in hares. This tick species has an endophilic and monotropic behaviour and all stages feed predominantly on rabbits (Estrada-Peña et al., 2017). The identification of some of these tick species may be difficult, especially for those belonging to the genus Rhipicephalus spp., and misidentifications may exist (Estrada-Peña et al., 2017). In addition, no sequences of H. hispanica are currently deposited in GenBank. For these reasons, the identification of all tick species detected was confirmed through the study of the 16S rRNA gene and a phylogenetic analysis was performed. In the obtained phylogenetic tree (Figure 2), R. sanguineus s.l., R. pusillus and H. hispanica sequences were included in different and independent clades close to other Rhipicephalus spp. and Haemaphysalis spp. deposited sequences confirming the morphological identifications.