Introduction

Rickettsiae are gram-negative, obligate intracellular bacteria that are associated with arthropods and are able to grow only within the cytoplasm or, occasionally in the nucleus, of a variety of eukaryotic host cells. Based on the differences in etiology, serology, epidemiology, and intracellular growth characteristics, the genus Rickettsia has traditionally been divided into three different groups and clusters based on their molecular and antigenic similarities (Raoult and Roux 1997), namely, the typhus group (TG), the scrub typhus group (that includes only one species, Orientia tsutsugamushi), and the spotted fever group (SFG). Recently, a revision of the classification of Rickettsia was proposed by erecting the transitional group (TRG) as a distinct lineage that shares immediate ancestry with the members of the spotted fever group (SFG) rickettsiae, coupled with the TG and ancestral group (AG) rickettsiae (Gillespie et al. 2007, 2008). The TG comprises only two species, R. prowazekii and R. typhi. The SFG, the largest rickettsial group, contains a multitude of pathogenic and nonpathogenic antigenically related species (Sekeyova et al. 2001), whereas TRG includes R. felis and R. akari. The ancestral group also includes marginal species, such as R. belli, the most divergent species within the rickettsiae (Stothard et al. 1994), and R. canadensis. The transovarial and transtadial passage of SFG rickettsiae within tick vectors in nature ensures rickettsial survival with distribution limited to that of their tick vectors (Azad and Beard 1998; Raoult and Roux 1997; Walker and Fishbein 1991).

In the Mediterranean area, some SFG rickettsiae have been implicated in human disease and are therefore defined as the pathogenic species (Brouqui et al. 2007; Parola et al. 2005). These rickettsiae include R. conorii, formerly the causal agent of Mediterranean spotted fever, comprising a variety of genospecies of R. conorii (Zhu et al. 2005), R. slovaca (Sekeyova et al. 1998), implicated in development of TIBOLA-DEBONEL in humans (Lakos and Raoult 1999; Oteo et al. 2004; Raoult et al. 1997). In addition, several other species of SFG have been found initially in ticks and nowadays it is thought that they have low pathogenicity toward humans, as R. massiliae (Beati and Raoult 1993; Vitale et al. 2006), R. aeschlimanni (Beati et al. 1997; Raoult et al. 2002), R. helvetica (Beati et al. 1994; Fournier et al. 2000), R. monacensis (Jado et al. 2007; Simser et al. 2002) or R. akari (Radulovic et al. 1996).

The purpose of this study was to investigate, identify and characterize SFG rickettsiae in ticks (Acarina, Ixodidae) collected from environmental samples and wild and domestic host in Jaen province. Prevalence studies of infection in the vectors can be used as an indicator of a change in the intensity of Rickettsia transmission. However, these studies are difficult to carry out as prevalence in the vector is usually low and its estimation requires a large number of ticks to be dissected.

Material and methods

Sampling

The fieldwork was conducted in two ecologically different areas of medium altitude mountainous system surrounded by olive tree cultures from Jaen Province (Andalusia, Spain) between January 2000 to December 2007. First, Sierra of Andujar and Despeñaperros natural parks (38°16′ N, 4°6′ W and 38°23′ N, 3°32′ W), a siliceous inland formed by sandstone. Secondly, Sierra of Cazorla and Sierra Magina natural parks (38°5′ N, 2°45′ W and 37°44′ N, 3°28′ W), a calcareous inland area constituted by calcite and dolomite rocks. Both areas have a Mediterranean climate with average air temperature between 4.0–6.6°C in January and 23.3–27.7°C in July, and an altitude between 500 and 2,167 m. Rainfall varies with altitude, between 505 mm to 793 mm per year.

Ticks were collected on vegetation by dragging or removed from domestic (dogs, sheeps and goats) and wild mammals (greater white toothed shrew Crocidura russula, west European hedgehog Erinaceus europaeus, wood mouse Apodemus sylvaticus, wild rabbit Oryctolagus cuniculus, red deer Cervus elaphus, fallow deer Dama dama, mouflon Ovis aries musimon, wild boar Sus scrofa and stone marten Martes foina). After collection, the ticks were immediately placed in vials with 70% ethanol, properly labeled, and were later identified in the laboratory by species, gender and stage using existing taxonomic keys (Gil-Collado et al. 1979; Iori et al. 2005; Manilla 1998; Márquez et al. 1992; Walker et al. 2000).

Molecular methods

Ticks were rinsed with distilled water, dried on sterile filter paper and then crushed in sterile Eppendorf tubes. DNA was extracted individually using the Macherey-Nagel DNA tissue Kit (Düren, Germany) according to the manufacturer’s instructions. The efficiency of DNA extraction was verified in all samples by PCR assay, which amplifies 12S rRNA of tick origin as well in the cases that doubt persisted in discrimination of closely related species (i.e. between R. sanguineus and R. turanicus) using the oligonucleotides T1B and T2A (Beati and Keirans 2001) and amplification conditions described elsewhere (Bernasconi et al. 2002). Negative controls consisted of distilled water extracted in the same laboratory. Specific rickettsial sequences were detected by using PCR primers that amplify a portion of gltA, ompA and ompB genes, respectively (Márquez et al. 1998). Subsequent direct sequencing of amplified products was performed on selected samples in order to provide an objective and precise identification.

Positive PCR products were sequenced using PCR primers and the GenomeLab DTCS—Quick Start kit (Beckman Coulter) and a CEQ 2000XL capillary DNA sequencer (Beckman Coulter) according to the manufacturer’s instructions. The resulting sequences were manually aligned and analyzed with Bioedit vers. 7.0.1. sequence analysis software (Hall 1999) to obtain consensus sequences and to align and compare other rickettsiae sequences found on GenBank, including previously sequenced and identified species of Rickettsia, with homologous sequences obtained directly from ticks. Sequences were identified using the BLAST feature of GenBank (http://ncbi.nlm.nih.gov/blastn) (Altschul et al. 1990). Phylogenetic relationships among sequences from tick associated rickettsiae, representatives of the established Rickettsia taxa, and other partially characterized Rickettsia isolates were analyzed using the resulting alignment into the PAUP (Phylogenetic Analysis Using Parsimony) vers. 4.0 beta 10 win software package for parsimony analysis (Center for Biodiversity, IL Natural History Survey, Champaign, IL).

Results

A total of 1,246 ixodid ticks (624 males, 511 females and 111 nymphs), representing 13 species (Table 1) were sampled in both natural and nearby urbanized areas. Hyalomma lusitanicum was the predominant tick species in siliceous inland, while R. bursa was the main species in Sierra of Cazorla and Magina. Rh. sanguineus was the unique species collected in urban and suburban areas.

Table 1 Distribution by gender, stage and origin of ticks included in this study
Full size table

Overall, rickettsial DNA was detected in 243 (19.5%) of the examined ticks by PCR amplifying specific fragments of glta, as well as ompA and ompB genes (Table 2). The presence of rickettsiae was demonstrated in 29.73% of I. ricinus studied (33/111), and in 12.5% of I. ventalloi (2/16). Furthermore, 84.57% of D. marginatus (137/162), 13.4% of Ha. punctata (13/97) and 16.13% of Ha. sulcata (15/93) ticks examined were positive. In Rhipicephalus sanguineus group (Rh. sanguineus, Rh. turanicus and Rh. pusillus) ticks 15.25% were positive (43/282). None of I. hexagonus, Ha. hispanica, Hy. lusitanicum, Hyalomma sp, or Rh. bursa studied ticks contained rickettsiae. PCR water controls and DNA extraction control were negative.

Table 2 Distribution among ticks (by gender and stage) of SFG rickettsiae identified
Full size table

Sequence analysis based on the portion of the gltA, ompA and ompB genes revealed that D. marginatus ticks are infected with two rickettsial species. In 40 D. marginatus (29.2% of rickettsiae positive Dermacentor) a rickettsia indistinguishable from R. slovaca was found. Moreover, in 97 D. marginatus (70.8% of rickettsiae positive Dermacentor) and in 9 Ha. punctata a rickettsiae similar to R. raoultii (Mediannikov et al. 2008), including RpA4, candidatus R. rioja and others, was found. In the case of species of the genus Ixodes, 2 I. ventalloi and 19 I. ricinus were infected with IRS3 strain. Moreover, in 11 I. ricinus and one Ha. punctata, the sequences from a portion of the gltA and ompA genes were closely related to that of R. monancensis. In only three I. ricinus we detected R. helvetica (gltA and ompB). In Jaen province both Haemaphysalis (Herpetobia) sulcata and Ha. punctata appeared infected with an exclusive SFG rickettsia. The gltA sequence found in all positive Ha. sulcata and 43% of positive Ha. punctata (GenBank accession code EU863190) agrees (98% of similarity, 376/381 bp) with the endosymbiont of Ha. sulcata described in southern Croatia (Duh et al. 2006) (GenBank accession code DQ081187) and with candidatus R. hoogstraalii (Mattila et al. 2007) co-isolated along with cellular line CCE2 (98% similarity, 358/363 bp) (GenBank accession code EF629539). We detected gltaA, ompA and ompB sequences of R. massiliae (Beati and Raoult 1993) in 43 of 282 Rhipicephalus sanguineus group examined ticks (32/228 Rh. sanguineus, 6/29 Rh. turanicus and 5/25 Rh. pusillus). No double infection was detected in any case.

Discussion

The Iberian Peninsula (Bartolomé et al. 2005; Cardeñosa et al. 2003, 2006; Guerrero et al. 2006; Herrero-Herrero et al. 1989; Jado et al. 2007; Lledó et al. 2006; Oteo et al. 2006) and particularly Andalusia (Bernabeu-Wittel et al. 2005, 2006) are known to be endemic regions of Mediterranean spotted fever (MSF) for humans. The epidemiology of rickettsiae and rickettsial diseases in Andalusia (south of Iberian Peninsula) is not well known, and a limited number of previous studies affecting Rickettsia sp. prevalence in ticks from domestic or feral animals have been made (Márquez et al. 1998, 2006, 2008).

In our area, both I. ricinus and I. ventalloi were infected mainly with R. monacensis sensu lato (formerly “Cadiz agent”, IRS3, IRS4, etc.), and in the case of I. ricinus a lower proportion with R. helvetica. In previous studies in Castilla-Leon (northern Spain) I. ricinus was infected also with rickettsiae IRS3 (2.44%) and R. helvetica (0.61%) and lesser percentages for rickettsiae RpA4 (Fernández-Soto et al. 2004). No rickettsia was detected previously in I. hexagonus. In northern Portugal, a 5.9% of I. ventalloi parasitizing the short-eared owl Asio flammeus was infected with R. helvetica (Santos-Silva et al. 2006).

The prevalence of R. slovaca infection in D. marginatus ticks (29.19%) in this study is similar to that obtained by other investigators: 30.3–45.4% in Switzerland (Beati et al. 1994), 36.8% in Croatia (Punda-Polić et al. 2002). However, that value was higher than 21.1% observed in Hungary (Lakos and Raoult 1999), 5.5% in Portugal (Bacellar 1999) or 1.8–3.3% obtained in Russia (Shpynov et al. 2001). In Castilla-Leon (northern Spain) Fernández-Soto et al. (2003, 2004, 2006a, b) detected by PCR the presence in D. marginatus and D. reticulatus of R. slovaca (11.43% and 4.93% respectively), R. raoultii (Marne strain formerly RpA4, and Khabarovsk strain initially referred as DnS14/DnS28) (4.93% and 10.0%). Moreover, the frequency values for other rickettsiae genospecies like DnS1 and R. monacensis (formerly “Cadiz agent”, IRS3 and IRS4) were very low (Fernández-Soto et al. 2006a). Immature D. marginatus ticks usually feed on small mammals and birds, whereas adult ticks mainly feed on large mammals but frequently also on humans (Rehácek and Tarasevich 1991).

In Rh. sanguineus group only R. massiliae were detected. Rickettsia massiliae has been detected previously in Rh. sanguineus and/or Rh. turanicus from France (Beati and Raoult 1993), Portugal (Bacellar 1999), Spain (Beati et al. 1996; Merino et al. 2005; Márquez et al. 2008), Switzerland (Bernasconi et al. 2002), Greece (Psaroulaki et al. 2006) and Algeria (Bitam et al. 2006), and recently was signalled in United States (Eremeeva et al. 2006). Our result agrees with that obtained by Fernández-Soto et al. (2006a, b) in Castilla-León (northwestern Spain). In Portugal, Bacellar (1999) found a prevalence of SFG rickettsiae of 1.8% (38 positives over 2207 Rh. sanguineus tested for the presence of rickettsiae, collected from vegetation, dogs and other wild mammals). Over 25 Rickettsia isolates 22 were R. massiliae and 3 correspond to R. conorii. In addition, Rh. pusillus parasitizing Egyptian mongoose (Herpestes ichneumon) was signaled as reservoir of R. sibirica mongolotimonae in south Portugal (de Sousa et al. 2006a, b)

In north Spain, Ha. punctata has been noted infected with IRS3 and R. aechlimannii (Fernández-Soto et al. 2003). In this investigation, Ha. punctata was infected with R. monacensis, candidatus R. raoultii and Rickettsia sp. R. aechlimannii appeared highly associated with Hy. marginatum (Matsumoto et al. 2004), showing higher prevalences (5.86%) for this ticks species than for other co-feeding species such as Rh. sanguineus, Rh. bursa, Rh. turanicus, Ha. punctata and I. ricinus, with observed prevalences values lower than 2% (Fernández-Soto et al. 2003). Rickettsia aeschlimannii was originally detected in Morocco in Hy. marginatum, which is the only tick species found to be associated with this rickettsia (Beati et al. 1997). The first human case of R. aeschlimannii infection was documented in a man who traveled to Morocco (Raoult et al. 2002).

The presence of an undescribed rickettsiae, closer to the endosymbiont described in Croatia and to candidatus R. hoogtraalii, is documented for first time in Iberian Peninsula. The adults of Ha. sulcata were found feeding habitually on wild and domestic mountain ungulates (Iberian ibex, mouflon, domestic sheep and goat) in high Mediterranean mountain situations. Larvae and nymphs of Ha. sulcata were found in several species of reptiles during feeding, mainly in the lizard Psammodromus algirus attached to skin pocket (Salvador et al. 1999; personal observations), sometimes co-feeding with larvae and nymphs of Ha. punctata and I. ricinus. This exposure may explain the presence of this rickettsial organism in both Haemaphysalis species that share some of the hosts and the middle-high calcareous mountains range.

Based on molecular methods, we have demonstrated the presence of several SFG rickettsiae in the southern part of Iberian Peninsula with still uncertain pathogenicity for humans. When a microorganism is found in an arthropod capable of biting humans, it could be considered a potential human pathogen (Gilot et al. 1990; Raoult et al. 1997; Walker and Fishbein 1991). The data presented in this article extend the knowledge of the geographic distribution of SFG rickettsiae in the southwestern Mediterranean part of Europe. This area will be severely affected by climate change and by the emergence and potential increase of tick species that may play an important role as vector of microorganisms pathogenic for both humans and animals (de Sousa et al. 2006a, b). In a near future, with unpredictable climate conditions, ticks might be considered as epidemiological markers for a number of infectious agents transmitted by them.