We conducted this study in the Slovak Karst National Park in Slovakia (48°36′N, 20°52′E) during 2004–2011. We carried out lizard capture and sample collections with official permits (6103/2007-2.1 and 5498/2011-2.2) issued by the Ministry of Environment of the Slovak Republic. We captured 103 green lizards (Lacerta viridis) and collected blood from 40 (30 males and 10 females). We collected 235 I. ricinus ticks (118 larvae and 117 nymphs) from 63 green lizards and 271 questing I. ricinus ticks (132 nymphs, 76 males, and 63 females) from the same area and immediately stored them in 70% ethanol.

We isolated DNA from lizard blood using a DNeasy Blood & Tissue Kit (QIAGEN, Hilden Germany) and isolated DNA from ticks by alkaline hydrolysis. We performed PCR amplification in 25-μL (total) reaction mixtures using the MasterTaq DNA Polymerase Kit (Eppendorf AG, Hamburg, Germany). We amplified sequences using the primer combinations EHR747 plus EHR521 or fD1 plus rP2 (9), which spanned almost the entire 16S rRNA sequence (Table 1). We examined the ≈250-bp gene fragment of 16S rRNA by single-strand conformation polymorphism (SSCP) analysis to determine Anaplasmataceae species type (10). We performed SSCP analysis following the protocol of Derdakova et al. (11). We ran positive control samples A. phagocytophilum, A. ovis, Wolbachia sp., and Candidatus N. mikurensis with each reaction. We purified the PCR products obtained using the GenElute PCR Clean-Up Kit (Sigma-Aldrich, Buchs, Switzerland) and sequenced both strands. We edited variants obtained in this study (1,410 bp) using MEGA 4.0.2 (https://megasoftware.net/) and checked by eye. We made comparisons to sequences in GenBank with BLASTn 2.2.26 (https://pods.iplantcollaborative.org/wiki/display/DEapps/Blastn-2.2.26). For phylogenetic analysis of our variant (GenBank accession no. MG924904), we aligned 17 related sequences obtained from the GenBank database and constructed a phylogenetic tree using the Bayesian inference method (12).

We examined all blood and ticks collected for the presence of family Anaplasmataceae bacteria. Anaplasmataceae family members were present in 36 lizards (28 males and 8 females). Of the ticks removed from lizards, 87 (37%) were infected, and of questing ticks, 18 (6.6%) were infected (Table 2).

Denatured and electrophoresed PCR products from samples demonstrated several SSCP profiles, of which 1 was clearly distinguishable from the profiles of the Anaplasmataceae species used as controls (Figure 1). We detected this unique profile in all lizard blood samples, all ticks feeding on lizards, and some questing ticks. We sequenced representatives of this unidentified SSCP profile (≈247 bp; GenBank accession nos. KY031322–3) and compared them with DNA fragments in the GenBank database. The closest related (99% identity) 16S rRNA sequences were from uncultured Anaplasma sp. isolates from questing I. ricinus ticks from Morocco (GenBank accession no. AY672415), Tunisia (GenBank accession no. AY672420), and France (GenBank accession no. GU734325). Sequencing of a longer (1,410-bp) fragment of the 16S rRNA gene revealed 99.1% similarity with the Candidatus C. californiense isolate from I. pacificus ticks in California (Figure 2). The 16S rRNA sequence obtained in this study was found to share a maximum of 94% identity with A. phagocytophilum Norway variant 2 (GenBank accession no. CP015376). The phylogenetic tree we constructed using 16S rRNA gene sequences showed that the reptile-associated Candidatus Cryptoplasma sp. REP (reptile) clustered in a separate branch with Candidatus C. californiense, indicating the isolate represents a lineage distinct from other known Anaplasmataceae species (e.g., A. phagocytophilum, A. marginale, A. platys, Ehrlichia muris, E. chaffeensis, and E. ewingii).

The role of ectotherm animals, especially lizards, in the maintenance of vectorborne pathogens is not clear. The interaction between reptiles and Anaplasmataceae family members has only been investigated in a few studies. Our findings expand knowledge on this research topic. Only limited information about the reptile–Anaplasma relationship exists. Ekner et al. suggested that sand lizards could potentially serve as a reservoir host for species of the Anaplasmataceae family when she discovered that ticks collected from these lizards in Poland were infected with Anaplasma-like pathogens (8). Although A. phagocytophilum might be transmitted by reptiles to a limited extent (5), the Anaplasma-like species detected in reptiles could also be a novel species, as suggested by Rejmanek et al. (6).

Despite the fact that lizards are exposed to a number of family Anaplasmataceae bacteria through infected ticks, our findings suggest that, except for Canditatus Cryptoplasma sp. REP, green lizards do not acquire infections with these species. In short, we detected Canditatus Cryptoplasma sp. REP in 90% of examined lizards, 37% of ticks feeding on lizards, and 6.6% of questing ticks in localities with lizards.

On the basis of our results, we cautiously speculate that Canditatus Cryptoplasma sp. REP is selected for and other genospecies selected against in ticks feeding on lizards. The Canditatus Cryptoplasma sp. REP variant had a high homology (100%) with a sequence obtained from an Apodemus agrarius mouse from Slovakia (13), which indicates that rodents or other mammals might also become infected with this bacterium and contribute (to a lesser extent) to the circulation of these bacteria in nature.

In conclusion, we found a yet to be named species of Canditatus Cryptoplasma sp. (Canditatus Cryptoplasma sp. REP) in questing I. ricinus ticks, I. ricinus ticks collected from and feeding on green lizards, and the blood of green lizards in Slovakia. These results indicate that green lizards serve as an intermediate host for this bacterium and that lizards can influence the enzootic maintenance and circulation of bacteria in the environment. However, other hosts besides reptiles could be involved in the Canditatus Cryptoplasma sp. REP lifecycle as well, though probably to a lesser extent.