Prevalence and diversity of Rickettsia species in ectoparasites collected from small rodents in Lithuania

Background Rickettsiae are emerging pathogens causing public health problems in many countries around the world. Rickettsia spp. are found in association with a wide range of arthropods which feed on different species of animals. However, the distribution and natural cycle of Rickettsia species and their association with different arthropod vectors are not fully established. The aim of this study was to investigate the presence and prevalence of Rickettsia spp. in ticks, mites and fleas parasitizing different species of small mammals in Lithuania and to molecularly characterize the Rickettsia spp. obtained from different ectoparasites. Results A total of 1261 ectoparasites (596 Ixodes ricinus ticks, 550 mites of five species and 115 fleas of eight species) collected from 238 rodents in Lithuania during 2013–2014 were investigated for the presence of Rickettsia pathogens. Infection rates were calculated as the maximum likelihood estimation (MLE) with 95% confidence intervals (CI). The infection rate varied among ectoparasites and was found highest in fleas 43.5%, followed by I. ricinus ticks (MLE = 26.5%; 95% CI: 22.2–31.3%) and then mites (MLE = 9.3%; 95% CI: 7.0–12.2%). Sequence analysis of partial gltA and 17kDa genes revealed the presence of Rickettsia helvetica, R. felis, R. monacensis, Rickettsia sp. and rickettsial endosymbionts. Four Rickettsia spp. were identified in fleas, while three Rickettsia spp. were identified in Laelapidae mites and only one (R. helvetica) in I. ricinus ticks. Conclusions To our knowledge, this is the first report of the occurrence and molecular characterization of Rickettsia spp. in 11 species of ectoparasites of small rodents in Lithuania. The present data extend the knowledge on the distribution of Rickettsia spp. and their association with different arthropod vectors. Prior to our study, R. felis had never been identified in Lithuania. To our knowledge, this is also the first report of R. felis in L. agilis and H. microti mites and in Ct. agyrtes and H. talpae fleas, as well as the first detection of R. monacensis in Ct. agyrtes fleas. Electronic supplementary material The online version of this article (10.1186/s13071-018-2947-9) contains supplementary material, which is available to authorized users.


Background
Rickettsiae are obligate intracellular Gram-negative bacteria from the order Rickettsiales that are characterised by complex life-cycles and a diversity of hosts and transmission strategies. Hosts of Rickettsia spp. are found in freshwater, marine and terrestrial habitats, and include protozoans, arthropods, vertebrates, photosynthetic algae and plants [1][2][3].
Rickettsia spp. are best known as human pathogens vectored mainly by hematophagous arthropods and causing public health problems in many countries around the world. Some species can cause diseases in other mammals and birds. Rickettsia species are horizontally transmitted to vertebrates by a variety of arthropod vectors which feed on different species of animals. To date, ticks (Ixodidae), lice (Phtiraptera) and fleas (Siphonaptera) are known to be competent vectors of rickettsial agents. However, most Rickettsia spp. have been found exclusively in non-hematophagous arthropods with no known secondary host [3].
Currently, the genus Rickettsia comprises 31 recognized species and numerous uncharacterized strains causing diseases in both humans and domestic and wild animals [4][5][6]. Rickettsia species and associated human clinical diseases vary depending on the geographical location [1,4]. The genus Rickettsia was traditionally classified into three groups based on phenotypic characters: the spotted fever group (SFG), the typhus group (TG) and the scrub typhus group (STG). The development of molecular tools over recent decades has resulted in reorganizations in the rickettsiae taxonomy. Phylogenomic studies have shown that the genus could be divided into four different phylogenetic groups: (i) the SFG associated mainly with ticks and, less frequently, with fleas and mites, consisting of 23 validated species (using a whole genome approach, Rickettsia helvetica was characterized as separate group from SFG); (ii) the typhus group which includes the agents of epidemic typhus and murine typhus associated with lice and fleas; (iii) an ancestral group (consisting of R. bellii and R. canadensis); and (iv) a transitional group whose members are R. akari and R. felis [7].
In Europe, rickettsioses are caused mainly by tick-borne SFG Rickettsia [1,2,4]. Ticks play important roles as vectors and sometimes as reservoirs in the ecology of these pathogens. The tick Ixodes ricinus is the main vector of R. helvetica and R. monacensis, while R. massiliae, R. raoultii, R. vini and other rickettsial species that have yet to be fully characterized have also been detected in this tick species [1,[8][9][10][11].
Mites of the family Laelapidae (Acari: Mesostigmata) are distributed worldwide and are the most common ectoparasites of small rodents. Parasitic species from this family are frequently found on the bodies of rodents or in their nests [12][13][14][15]. The role of mesostigmatid mites in the circulation of some disease agents has been previously confirmed and recent studies have also demonstrated that mesostigmatid mites may be reservoirs as well as vectors of some pathogenic rickettsiae (reviewed by Merhej et al. [4], Miťková et al. [12]). Fleas from the order Siphonaptera are another group of blood-sucking ectoparasites found on rodents [16]. Fleas are important vectors for various pathogens including viruses, bacteria and tapeworms. Different Rickettsia species have been detected in more than 15 flea species [4]. In Europe, fleas infesting small rodents have been found to be infected with R. felis, R. helvetica and rickettsial endosymbionts [17,18].
Information on the circulation of rickettsiae in the territories of the Baltic countries is scarce. In Lithuania, the presence and prevalence of two SFG rickettsiae has been previously reported in questing ixodid ticks [19]: R. helvetica has been identified in I. ricinus (with a prevalence of 17%) and R. raoultii in Dermacentor reticulatus (4.9%). So far, no reported human clinical cases due to infection by Rickettsia species have been registered in Lithuania. Detection of R. helvetica, R. monacensis and "Candidatus R. tarasevichiae" have been reported in questing I. ricinus and I. persulcatus ticks in Estonia [20]. In Latvia, R. helvetica has been identified in I. ricinus ticks from birds [21]. There are currently no published reports of the presence of Rickettsia spp. in mites and fleas in Lithuania, nor in other Baltic states.
The distribution and natural cycle of Rickettsia spp. and their association with different arthropod vectors are not fully established. In order to extend knowledge on the relationships between arthropod vectors, hosts and Rickettsia pathogens, we aimed to investigate the presence and prevalence of Rickettsia spp. in ticks, mites and fleas parasitizing different species of small rodents in Lithuania and to molecularly characterize the Rickettsia spp. obtained from different ectoparasites.

Sample collection
Trapping of rodents was conducted on the Curonian Spit of Lithuania during 2013-2014. The Curonian Spit is a narrow sand peninsula (2 km wide and 98 km long, with half of the length in Lithuania) in the southeastern part of the Baltic Sea separating the Curonian Lagoon from the Baltic Sea [22]. The climate of the Curonian Spit is intermediate between marine and continental and is characterized by frequent and intense variability of weather, relatively mild winters and temperately warm summers. According to the phytogeographical classification, the territory of the Curonian Spit is considered to be within the zone of mixed coniferous-broad-leaved forest. The territory of the Curonian Spit is characterized by considerable flora species diversity [22]. Small rodents were captured at eight sampling sites ( Fig. 1) located in different habitats (coastal meadow, mixed forest and forest-meadow ecotone). The names and coordinates of the sampling sites are provided in Additional file 1: Table S1.
Ectoparasites were collected from hosts using tweezers and placed into 1.5 ml tubes with 70% ethanol. The collected ticks, mites and fleas were identified by the use of taxonomic keys [13,16,[23][24][25].

DNA extraction
DNA from ticks, mites and fleas was extracted by using 2.5% ammonium hydroxide solution [26]. DNA from engorged ticks was extracted using a Genomic DNA Purification Kit (Thermo Fisher Scientific, Vilnius, Lithuania) according to the protocol suggested by the manufacturer. Ectoparasites were processed individually or pooled (2-10 specimens) in groups from each host by species, life stage, sex and location.

PCR amplification and DNA sequencing
Rickettsia DNA in ectoparasites was detected using a nested PCR that targeted partial gltA gene (encoding rickettsial citrate synthase of Rickettsia spp.) and two primer sets RpCS.877p/RpCS.1258n (external primers) and RpCS.896p/RpCS.1233n (internal primers) as previously described [12]. The obtained specific products of 338 base pairs were considered as a positive result.
All gltA-positive samples for Rickettsia were further examined using a semi-nested PCR which amplifies the 450 bp fragment of the 17-kDa gene (encoding rickettsiae genus-specific 17-kDa outer membrane antigen) using the primers Rr17k.1p/Rr17k.539n and Rr17k.90p as described [27]. A negative control (double-distilled water) and a positive control (DNA of Rickettsia-infected ticks, confirmed by sequencing) were included in every PCR run. PCR products were subjected to electrophoresis on 1.5% agarose gel and analyzed by UV transilluminator. To prevent cross-contamination, separate rooms were used for DNA isolation, PCR-mix preparation, the PCR and nested-PCR reactions. Negative controls, which consisted of sterile, double-distilled water added to the first PCR mix rather than DNA, were included after every five experimental samples.
A selected number of Rickettsia-positive samples for both genes were purified using the GeneJET™ Gel Extraction Kit (Thermo Fisher Scientific) and sequenced (Macrogen, Amsterdam, the Netherlands). Obtained DNA sequences were edited using the MEGA 6.05 software package [28] and aligned to available data in GenBank with BLASTn. The gltA and 17kDa protein-coding gene based phylogenetic trees were constructed using the maximum-likelihood (ML) and neighbor-joining (NJ) methods. The most appropriate model of nucleotide substitution was determined according to the Bayesian information criterion (BIC) using the program jModelTest2 [29,30].
Partial 17kDa and gltA gene sequences for representative samples were submitted to the GenBank database under the accession numbers MF491767-MF491779, MH454244 and MF491780-MF491791, MH454245-MH454247 respectively.

Statistical analysis
Infection rates for pooled ticks and mites were calculated using maximum likelihood estimation (MLE) method with 95% confidence intervals (CI) for unequal pool sizes and expressed as the MLE of infection rate per 100 ticks/mites. MLE calculations are based on the number of pools, pool sizes (number of individuals per pool), and the number of positive pools [31]. We used the PooledInfRate estimation software (version 4.0) add-on with Microsoft Excel (http:// www.cdc.gov/westnile/resourcepages/mosqSurvSoft.html).
Rickettsia DNA was detected in eleven species of ectoparasites. Rickettsia-infected fleas were found parasitizing five rodent species, while infected I. ricinus ticks and mites were found only on two species of rodents (  Table 2).
PCR and sequence analysis of the partial 17kDa gene revealed the presence of three Rickettsia species: (i) R. helvetica: isolates derived from I. ricinus ticks, L. agilis mites and two species of fleas M. walkeri and M. turbidus were 99-100% identical (with one nucleotide    (Fig. 3).

Discussion
In this study, we report the occurrence and the molecular characterization of Rickettsia spp. in eleven species of ectoparasites parasitizing small rodents on the Curonian Spit, West Lithuania. The frequency of Rickettsia spp. infections in rodent-derived ectoparasites varied among species and was found highest in fleas, followed by I. ricinus ticks and then mites. Small rodents are important hosts for the immature stages of ixodid ticks and are considered carriers and reservoir hosts of tick-borne pathogens such as tick-borne encephalitis virus [32], Borrelia burdorferi (s.l.) [33], Borrelia miyamotoi [34], Babesia microti [35], Anaplasma phagocytophilum and "Candidatus Neoehrlichia mikurensis" [36]. The reservoir role of wild rodents in rickettsiae life-cycles is still unclear and there are a lack of studies focusing on the investigation of rickettsial pathogens in rodents and their ectoparasites across Europe [12,17,37,38]. Several molecular studies conducted in the Netherlands, Germany, Slovakia, Poland and Austria have showed the presence of rickettsial DNA in European rodents with a prevalence ranging from 2.7 to 29%. Infected rodents more frequently harbored R. helvetica than other rickettsiae species [12,18,37,39,40].
The presence of rickettsiae in ectoparasites may result from the acquisition of bacteria via blood meals from rickettsiemic rodents. Some SFG rickettsiae are thought to circulate in enzootic or epizootic cycles between wild vertebrates and arthropod vectors [41]. The high prevalence of R. helvetica previously obtained in small rodents suggests that they may play an important role as potential natural reservoir hosts for this pathogen [37,39,42]. In the present study, the most frequently captured rodent species in all habitats was A. flavicollis, followed by 3%, while other rodent species were found to be non-rickettsiemic (data not shown) [42]. All infected rodents harboured R. helvetica [42]. Our data showed that 66.6% of Rickettsia-infected I. ricinus ticks, 59% of Laelapidae mites and 55% of fleas were derived from rodents which had been previously proved to be non-rickettsiemic, while 33.4 % of Rickettsia-infected I. ricinus ticks, 41% of Laelapidae mites and 45% of fleas were collected from infected rodents [42]. This could have potentially influenced the detection rates estimated in our study for ectoparasites that harboured R. helvetica. It seems that PCR-positive ticks, mites and fleas feeding on non-infected rodents, as well as ectoparasites which harboured other Rickettsia species, did not acquire these pathogens through blood meals on rodents. Mites and fleas in these cases could be infected during previously feedings on another hosts (e.g. in common nests). Occurrence of PCR-positive I. ricinus larvae on non-infected rodents probably show transovarial transmission from engorged females.
To our knowledge, the present study is the first report on the presence of Rickettsia spp. in fleas and mites from the Baltic countries. Fleas and mites feeding on rodents harboured multiple Rickettsia species, including R. helvetica, R. felis, R. monacensis, rickettsial endosymbiont and Rickettsia sp. Four Rickettsia spp. were identified in fleas, while three Rickettsia spp. were identified in Laelapidae mites. In Rickettsia-positive I. ricinus samples subjected for sequence analysis, only R. helvetica species was detected.
Rickettsia helvetica is a human pathogenic Rickettsia species which was first discovered in I. ricinus from Switzerland in 1979 [43]. Human cases have been reported from Sweden, France, Switzerland, Italy, Denmark, Austria and Slovakia (reviewed by Parola et al. [1]). Ixodes ricinus is considered to be the main vector and a natural reservoir of R. helvetica [44] due to the ability of rickettsiae to survive perpetually in ticks and to be transmitted transstadially and transovarially [1]. To date, R. helvetica has been detected in I. ricinus in many European countries with a highly variable prevalence. This species has been previously detected in questing I. ricinus ticks in Lithuania with a prevalence ranging in different locations between 0-31.3% [19].
Rickettsia helvetica has also been isolated from other tick species such as D. reticulatus, I. hexagonus and I. arboricola, as well as from mites and fleas [1]. This species has been detected in Ct. agyrtes males collected from A. agrarius in Slovakia [17] and in two out of 24 fleas isolated from M. glareolus and A. sylvaticus in the Netherlands [39]. Miťková et al. [12] reported for the first time the detection of R. helvetica in L. agilis and Hg. nidi mites infesting small rodents in Slovakia.
In our study, R. helvetica was detected in I. ricinus ticks, L. agilis and M. gigas mites and Ct. agyrtes, M. turbidus and M. walkeri fleas from A. flavicollis; I. ricinus ticks and M. walkeri fleas from M. glareolus, and M. turbidus fleas from M. minutus. To our knowledge, this is the first detection of R. helvetica in M. gigas mites and M. turbidus and M. walkeri fleas. The presence of R. helvetica in the immature ticks may result from acquisition pathogens through a blood meal from rickettsiemic rodents or through a vertical (transovarial and transstadial) transmission. Mites and fleas infesting rodents could acquire R. helvetica infection while taking a blood meal on a R. helvetica-infected rodent or, possibly, by co-feeding with infected I. ricinus ticks. Recently, horizontal transmission through a shared blood meal was demonstrated for some rickettsial pathogens [45].
This study is the first report of R. felis infection in Ct. agyrtes and H. talpae fleas collected from A. flavicollis, and in H. microti and L. agilis mites collected from M. minutus in Lithuania. Rickettsia felis is the causative agent of flea-borne spotted fever in humans [46]. This species has worldwide distribution and infections of R. felis have been reported in over 25 countries spanning five continents [47]. The cat flea, Ctenocephalides felis, is a main vector and a reservoir of this pathogen [48]. However, more than 24 species of ectoparasites, such as fleas, ticks, mites, lice and mosquitoes, have been demonstrated as vectors for R. felis all over the world [4,13,17,27,49]. Recently, both intra-and interspecific transmission of R. felis between co-feeding arthropods on a vertebrate host has been demonstrated [44]. The finding of R. felis in naturally infected mites is not unexpected because it has been previously reported in other Mesostigmata mite species collected from rodents in Asia [50,51]. Sequence analyses of R. felis strains isolated from different arthropods revealed genomic heterogeneity and provides evidence for host-specific strain variation [52]. Additionally, increasing numbers of identified R. felis-like organisms (RFLOs) in different arthropods have been reported within the last decade.
To our knowledge, this study provides the first evidence of the presence of R. monacensis in Ct. agyrtes fleas collected from A. flavicollis (Fig. 2). Rickettsia monacensis is an etiological agent of human rickettsioses distributed throughout Europe. Ixodes ricinus is recognized as the main vector of this pathogen [4,13,53]. However, not all possible vector species are currently known [53]. Rickettsia monacensis has been detected in questing I. ricinus ticks in many European countries [1] and has also been detected in immature I. ricinus ticks collected from migratory birds in the Russian part of the Curonian Spit [54]. In Lithuania, R. monacensis has been previously detected in I. ricinus ticks removed from raccoon dogs (GenBank: KT033401). The first occurrence of mites infected with R. monacensis has recently been reported in Slovakia [12].
Based on sequence analysis of the gltA gene, the Rickettsia sp. detected in our study in the mite E. stabularis was not identified to the species level. The obtained 268 bp gltA sequence showed 100% identity to the corresponding sequences of R. vini, R. japonica, R. heilongjiangensis and R. raoultii deposited in the GenBank database. Unfortunately, the fragment of the 17kDa gene of this sample was not successfully amplified. R. vini was first detected in the ticks I. arboricola and I. ricinus collected from birds in Spain and named "Candidatus R. vini" [8,55]. This bacterium has since been detected in ticks feeding on birds in other European counties and Turkey [56]. The pathogenicity for humans and animals of this Rickettsia remains unknown [57]. Rickettsia vini has been confirmed in Lithuania in I. lividus ticks from sand martin (Riparia riparia) nests (GenBank: MH454248).
The present study indicates the presence of phylogenetically distinct rickettsiae in multiple ectoparasites feeding on different rodent hosts. Future studies should be performed to determine the specific roles of different species of mites and fleas parasitizing small rodents in the acquisition and transmission of different Rickettsia species.

Conclusions
To our knowledge, this is the first report of the occurrence and molecular characterization of Rickettsia spp. in 11 species of ectoparasites of small rodents in Lithuania. The data presented in this paper extend the knowledge on the distribution of Rickettsia species, their association with different arthropod vectors and rodent-parasites interactions. Prior to our study, R. felis had never been identified in Lithuania. To our knowledge, this is also the first report of R. felis in L. agilis and H. microti mites and in Ct. agyrtes and H. talpae fleas, as well as the first detection of R. monacensis in Ct. agyrtes fleas.