Background

Ixodes ricinus (Acari, Ixodidae) is an important vector of many pathogens of medical and veterinary importance with a wide host and distribution range across the entire West Palearctic region, from British Islands to Russian Ural and from North Africa to Scandinavia [1].

Despite the complexity of the genetic structure of I. ricinus populations, two patterns have been consistently shown, regardless of the approach used: the genetic distance of ticks on the British Islands compared with that of populations in North Africa [2,3,4] and the genetic dissimilarity between the North African tick population and that on continental Europe [5, 6]. In 2014, the latter pattern resulted in a description of the new species Ixodes inopinatus, based on the morphological characteristics and the partial sequence of the 16S rRNA gene of ticks from Spain (type locality), Portugal, and North Africa (Algeria, Tunisia, Morocco) [7].

Since then, ticks referred to as I. inopinatus have been reported in several European countries including Germany, Austria, Switzerland, Romania, and Turkey. These reports are based solely on morphology [8,9,10,11,12], morphology combined with sequencing of 16S ribosomal DNA (rDNA) [13,14,15,16,17,18], and sometimes using other genetic markers [2, 19,20,21]. However, with the rising number of available sequences, the number of reports of a failure to distinguish the two species has been growing, resulting in the use of the term I. ricinus/inopinatus complex [22,23,24,25]. In our surveillance of I. inopinatus in the Czech Republic, we also encountered difficulties in differentiating I. ricinus from I. inopinatus based on the morphology and sequencing of 16S rDNA. This resulted in our search for other genetic and easy-to-use markers that would differentiate these two sympatric and morphologically close to identical species.

So far, no consensus has been reached on the selection of a single molecular marker to differentiate ixodid tick species. The short fragment of mitochondrial 16S rRNA is often the first choice for tick identification together with the morphological description [26,27,28], followed by sequencing of the cytochrome c oxidase I subunit (COI) [29,30,31]. As for nuclear markers, the internal transcribed spacer 2 (ITS2) was used not only for species delineation [32,33,34] but also for the detection of natural hybrids between Ixodes persulcatus and I. ricinus as well as between I. ricinus and I. pavlovskyi [35, 36]. Use of other genes is scarce and in the context of I. ricinus/inopinatus, only the tick receptor for the OspA (TROSPA) and defensin genes showed discrimination power [5, 19,20,21].

In this study, we aimed for an easy and fast method for differentiation of the closely related and possibly sympatric species, I. ricinus and I. inopinatus, and searched for I. inopinatus in the Czech Republic. During the validation of a multiplex polymerase chain reaction (PCR) targeting the TROSPA gene, we were driven to a deeper study of the genetic diversity of these two species using mitochondrial and nuclear markers.

Methods

Adult ticks were collected by flagging in the Czech Republic (CZ) and Algeria (ALG) between 2015 and 2020. Ticks were collected in three different regions in CZ: Libava, northern Moravia (n = 114), Prostredni Porici, central Moravia (n = 110), and Podyji, southern Moravia (n = 103), and from one locality in ALG: El-Tarf province (n = 47). Based on the morphological characteristics according to Estrada-Peña et al. [7], all ticks were identified as I. ricinus or I. inopinatus. For phylogenetic purposes, another five Ixodes spp. were used: I. frontalis (Italy, 2021), I. gibbosus (Italy, 2021), I. hexagonus (Czech Republic, 2021), I. persulcatus (Russia, 2019), and I. ventalloi (Italy, 2013). These ticks were collected by our team as part of other ongoing projects with DNA of I. ventalloi obtained from colleagues from Italy [37]. Ticks were identified by BLASTn (Nucleotide Basic Local Alignment Search Tool) analysis of their 16S sequences. All samples were stored in 70% EtOH at −20 °C until further analysis.

Genomic DNA was isolated from a longitudinal half of each tick using the Exgene Cell SV mini 250p Kit (GeneAll, Seoul, Korea) according to the standard protocol for animal tissues with 100 µl of elution buffer added in the final step. The other half of the tick was stored for potential reanalysis.

For an easy and fast way to distinguish the two main variants of the TROSPA gene sequences reported as I. ricinus and I. inopinatus, we designed a multiplex PCR. A specific pair of primers for each variant was designed within the intron region based on the alignment of available sequences in GenBank (Table 1). The resulting amplicons differed by 126 base pairs (bp) for an easy on-gel identification. PCR was performed in a total volume of 25.0 µl using 2× PCRBIO Taq Mix Red (PCR Biosystems, UK), 0.4 µM of each of the four primers and 2.0 µl of template DNA. Reaction conditions followed manufacturer instructions with the annealing temperature of 52 °C and the elongation time of 15 s for 40 cycles.

Table 1 Primers used in this study
Full size table

For validation of the multiplex PCR and for subsequent sequence analysis, the 824 bp long fragment of the TROSPA gene including the entire intron was amplified and sequenced. To assess the genetic variability in more detail, fragments of two nuclear (ITS2 and calreticulin) and two mitochondrial genes (16S rRNA and cytochrome C oxidase subunit I—COI) were also amplified and sequenced. Primer sequences and PCR conditions are shown in Table 1.

Amplification of TROSPA, 16S rDNA, and ITS2 was performed in a total volume of 25 µl using 2× PCRBIO Taq Mix Red (PCR Biosystems, UK), 0.4 µM of each primer and 2 µl of template DNA. Reaction conditions followed manufacturer instructions for 40 cycles. COI and calreticulin genes were amplified in the total volume of 20 µl using Phusion Green Hot Start II High-Fidelity PCR Master Mix (Thermo Fisher Scientific, USA), 0.5 µM of each primer, and 2 µl of template DNA. Reaction conditions followed manufacturer instructions for 40 cycles. All PCR reactions were visualized on 1.5% agarose gel with the Midori Green Advance system (Nippon Genetics Europe, Germany). All products of expected length were cut from the gel, purified by the Gel/PCR DNA Fragments Extraction Kit (Geneaid Biotech Ltd., Taiwan), and sequenced by the Macrogen capillary sequencing services (Macrogen Europe, Netherlands) in both directions using the amplification primers.

Obtained sequences were assembled and visually inspected using Geneious R11.1.5 [38]. The identity of the amplicons was confirmed by BLASTn analysis (NCBI GenBank). Due to the appearance of double peaks in otherwise high-quality sequences (TROSPA and calreticulin genes), the detection of heterozygotes was performed using the Geneious plugin Find Heterozygotes followed by visual inspection and assigning ambiguous bases in positions with double peaks detected in both strands (with settings of peak similarity 30% and peak detection height 50%).

Representative samples with a high number of double peaks detected in the TROSPA gene as well as samples with a sudden loss of sequencing signal followed by an apparently mixed product chromatogram in the ITS2 region were cloned using pGEM®-T Easy Vector Systems (Promega Corporation, USA). Acquired plasmid DNA was purified from the bacterial culture using the GenElute™ Plasmid Miniprep Kit (Sigma-Aldrich, USA) and sequenced by the Macrogen capillary sequencing services (Macrogen Europe, The Netherlands) using universal T7/SP6 primers.

For phylogenetic analyses, sequences representing 16S rDNA, COI, TROSPA, ITS2, and calreticulin from various tick species within the genus Ixodes, preferably from different studies and geographical origins, were selected from the GenBank. Sequences originating from this study were limited to representative sequences in the case of 16S rRNA and COI, representative sequences and unique clones for ITS2, all sequences with 0–5 ambiguities and representative clones for TROSPA, and all sequences with a maximum of two ambiguous nucleotides for calreticulin (Additional file 2: Table S2).

Phylogenetic analyses were conducted by the ClustalW alignments built in Geneious R11.1.5 [38]. After manual editing of poorly aligned regions (especially the 16S rRNA gene), phylogenies were calculated by the maximum likelihood method using IQ-TREE multicore version 2.1.3 [39]. The best-fit evolution models were chosen based on the Bayesian information criterion (BIC) computed by ModelFinder [40]. Branch support was assessed by the ultrafast bootstrap (UFBoot) approximation [41] and by the SH-like approximate likelihood ratio test (SH-aLRT) [42]. Trees were visualized and edited in FigTree v1.4.4 and Inkscape 1.1.1.

Results

In total 374 adult ticks preselected by morphology as the I. ricinus/inopinatus complex were screened by multiplex PCR (Table 2) and three gel patterns were observed (Additional file 1: Figure S1). A single band of the size corresponding to the I. ricinus allele was observed in 321 ticks (317 from CZ and 4 from ALG) and a single band corresponding to the I. inopinatus allele was seen in 43 ticks from ALG and none from CZ. In 10 ticks from CZ, two bands were detected, each corresponding to one of the two species (Table 2, Additional file 1: Figure S1).

Table 2 Tick identification by multiplex PCR from four localities
Full size table

To validate the newly designed assay, the entire intron region of the TROSPA gene was amplified and sequenced from randomly selected representatives of both species from all localities and all 10 ambiguous samples. We were able to consistently amplify and sequence 670 bp out of the expected length of 824 bp, resulting in high-quality chromatograms from 112 I. ricinus and 19 I. inopinatus (based on the multiplex PCR) and all 10 ambiguous samples. Chromatograms commonly revealed double peaks; in fact, only 13 samples had no double peaks. In 117 samples, 1 to 15 clear double peaks in otherwise flawless chromatograms were observed, and in 10 samples (all assigned as ambiguous by multiplex PCR) 25 to 32 double peaks were detected in both strands (Fig. 1).

Fig. 1
figure 1

Schematic representation of the partial TROSPA gene sequences depicting the nucleotide positions with the double peaks (small yellow bars) (KF041821 is used as a reference sequence). The chromatograms depict the forward and reverse strands of sequencing for the uncloned PCR product. 441 cl.1 and 441 cl.5 are sequences after cloning resolving the double peaks of hybrid ticks

Full size image

By cloning and sequencing of PCR products of the TROSPA gene from two ambiguous ticks and their alignment to sequences without double peaks from our study as well as with sequences from GenBank, we identified 23 positions consistently different between I. ricinus and I. inopinatus alleles (Fig. 1, Additional file 2: Table S1). In all 10 samples yielding bands corresponding to both I. ricinus and I. inopinatus in the multiplex PCR, the positions of double peaks corresponded to 23 single-nucleotide polymorphisms (SNPs) differentiating the two species. All other positions with double peaks showed no regular pattern and were detected in random positions.

To support the above-described analyses and to put the TROSPA species differentiation power in a larger context, we amplified and sequenced the same fragment from other Ixodes spp. (minimum of three individuals per species). In the phylogenetic analyses, I. ricinus and I. inopinatus sequences from this study together with sequences from the GenBank database form two well-supported sister clades (Fig. 2). The cloned sequences of the ambiguous samples based on the multiplex PCR, the two variants of alleles representing the I. ricinus and I. inopinatus species, fell within the respective clades. All other Ixodes spp. form well-supported and distinguished monophyletic clades (Fig. 2, Additional file 1: Figure S2).

Fig. 2
figure 2

Phylogenetic tree of ticks based on the TROSPA gene samples from this study are indicated in bold font. CZ Czech Republic, ALG Algeria. 441 M CZ cl.1/cl.5 and 331F CZ cl.3/13 are sequences after cloning of hybrid ticks showing a clear split to I. ricinus and I. inopinatus branches

Full size image

Fragments of two mitochondrial genes, 16S rDNA and COI, the most commonly used molecular markers for ixodid ticks, were amplified and sequenced. The fragment of 16S rDNA was amplified and sequenced from 222 ticks from CZ and 43 ticks from ALG. Thirty-six unique haplotypes (26 CZ and 10 ALG) with sequence similarity 96.02–99.73% were identified. From these unique haplotypes, seven (three CZ and four ALG) had the “AG” haplotype assigned previously to I. inopinatus, 28 (23 CZ and five ALG) had the “CT” haplotype referring to I. ricinus [13, 16], and one new AT haplotype (1 ALG) was detected. In the phylogenetic analyses, all representative sequences from this study form a single, highly supported clade together with the I. ricinus and I. inopinatus sequences retrieved from the GenBank database. However, no structure based on the species or geography was detected within the clade(s) (Fig. 3). Other Ixodes spp. form well-distinguished and supported clades (with the exception of I. affinis and I. pararicinus forming a single clade, Additional file 1: Figure S3).

Fig. 3
figure 3

Phylogenetic tree of ticks based on the 16S rRNA gene showing the lack of power to distinguish I. ricinus and I. inopinatus. Samples from this study are indicated in bold font. CZ Czech Republic, ALG Algeria

Full size image

Amplification and sequencing of the COI gene were done on 285 ticks (245 CZ and 40 ALG). In the phylogenetic analysis of the COI gene, all representative sequences from this study form a single clade together with the I. ricinus and I. inopinatus sequences retrieved from the GenBank database (Fig. 4). Although the bootstrap support of the clade is low, its resolution from the most closely related I. laguri is clear. All I. inopinatus sequences (as assigned based on the TROSPA analyses or by their name in GenBank) form a separate, highly supported subclade, although with very short branch length. Other Ixodes spp. form well-distinguished and supported clades, some with the intraspecific subclade structure (e.g., I. affinis and I. persulcatus) (Additional file 1: Figure S4).

Fig. 4
figure 4

Phylogenetic tree of ticks based on the COI gene depicting a very close relationship between I. ricinus and I. inopinatus. Samples from this study are indicated in bold font. CZ Czech Republic, ALG Algeria

Full size image

In addition to the TROSPA gene, two other nuclear markers were also amplified and sequenced from the subset of our tick samples. After direct sequencing, the ITS2 region yielded high-quality chromatograms only from 24 ticks (13 CZ and 11 ALG). In other samples, a sudden loss of the sequencing signal followed by an apparently mixed product chromatogram was observed. PCR products from 15 samples (eight CZ including two ticks assigned as ambiguous by multiplex PCR, and seven ALG) were cloned and 71 individual clones were sequenced (4–7 clones per sample). Among these, 37 unique clones (18 CZ and 19 ALG) were observed. Phylogenetic analyses of the available sequences representing various Ixodes spp. showed a pattern similar to that of the COI gene. All I. ricinus and I. inopinatus sequences (directly sequenced and cloned in this study as well as from GenBank) form a single, highly supported clade. All sequences of I. inopinatus (as assigned by the TROSPA gene) assembled into the unsupported subclade (Fig. 5, Additional file 1: Figure S5). Clones originating from a single individual always fell into a single subclade (I. ricinus/inopinatus). All clones from the two ambiguous samples (80F and 42F) clustered within the I. ricinus subclade.

Fig. 5
figure 5

Phylogenetic tree of ticks based on the ITS2 gene Samples from this study are indicated in bold font. CZ Czech Republic, ALG Algeria. Matching colors indicate sequences after cloning to resolve sequence ambiguities

Full size image

A part of the calreticulin gene was amplified and sequenced from 34 ticks (19 CZ and 15 ALG). Sequences with 0–2 double peaks (ambiguous bases were assigned) from 26 ticks (15 CZ and 11 ALG) were used for the phylogenetic analyses. The fragment was not suitable for distinguishing the Ixodes species since no clades were formed in the phylogeny (Additional file 1: Figure S6).

Discussion

Accurate identification of ticks at the species level is critical from several perspectives including distribution mapping, life cycle, and host range/preference, and most importantly for vector capacity for pathogens. Typically, methods based on the morphology and sequencing of 16S rDNA have been used for tick identification and description of new species [26, 28, 29], and other genes have rarely been used [27, 30, 33]. However, morphological identification relies on acarological expertise and specimen quality, and is rather time-consuming on large datasets. Furthermore, very low morphological variability makes it impossible to use for the identification of closely related taxa [9]. Sequencing of 16S rDNA is considered a gold standard for the identification of ticks and many other organisms, including bacteria. However, I. ricinus could not be differentiated from I. inopinatus by the commonly used 16S rDNA fragment [13, 16] due to the high haplotype diversity of this gene [8, 10, 11, 16, 22,23,24,25]. The COI gene is another common marker used for species delineation within the animal kingdom (e.g., the BOLD [Barcode of Life Data System] database); however, similarly to 16S rDNA, its analyses do not have the power to detect potential hybridization between closely related species. Mitochondrial markers are typically inherited uniparentally and therefore do not reflect the genetic history of an organism [43, 44].

Nuclear genes can reveal potential hybridization [35, 36], but these often have several copies resulting in mixed chromatograms in Sanger sequencing and the consequent need for cloning. So far, the TROSPA gene was the only marker that consistently distinguished tick populations from North Africa and Europe [5]. Internal transcribed spacers 1 and 2 (ITS1 and ITS2) are useful for subtyping due to the high intraspecific diversity [5]; however, for many tick species, sequences for these loci are not available in GenBank [45]. Calreticulin was found to be completely inappropriate for distinguishing tick species, which is consistent with Babkin et al. [46].

Clearly, the TROSPA and ITS2 genes seem to be good candidates for differentiation of the North African lineage of ticks referred to as I. inopinatus from the European population of I. ricinus. However, relatively small differences in these two markers between I. ricinus and I. inopinatus (in comparison to differences among other Ixodes species) opened a question of the natural gene flow between tick populations in North Africa and Europe. Our TROSPA data indicate natural hybridization followed by gene introgression and that hybrids of I. ricinus and I. inopinatus survive and may backcross the European parental population potentially resulting in unidirectional introgression [47]. However, this needs to be investigated further with larger sets of ticks, especially from North Africa.

Distribution patterns of arthropods, amphibians, reptiles, and mammals demonstrate biogeographical affinities between Europe and North Africa at the species level [48]. The distribution of primarily Palaearctic species across the Mediterranean has attracted considerable attention, showing North Africa as a refugium and differentiation center for Western Palaearctic thermophilic species. However, this applies to non-flying organisms only. Ixodes ricinus is a tick species commonly reported on birds (especially nymphs and larvae) [45], including migratory species [21, 49, 50]. We hypothesize, that I. inopinatus is adapted to climatic conditions in North Africa, and possibly the southernmost areas of Europe. African ticks are likely regularly carried by migratory birds between North Africa and Europe, as documented in the case of Hyalomma spp. [51, 52] as well as I. ricinus and I. inopinatus [53].

Since we did not find any signs of hybridization in North Africa, we hypothesize, that I. ricinus ticks from higher latitudes and their hybrids with the African population do not survive well in North African climate and that only North African ticks carried to Europe successfully hybridize and backcross with I. ricinus in Europe. However, it is important to point out that our data set from Algeria is much smaller than that from the Czech Republic, and follow-up studies are needed. To address this, we are currently conducting a surveillance study of ticks across Italy and additional sampling in Algeria (in preparation). The surveillance of ticks on migratory birds in the Czech Republic is also underway.

Similarly to the study from Germany [54], our data put into question studies reporting I. inopinatus from Central Europe based on morphology and/or sequencing of 16S rDNA, and we suggest that these should be re-examined. Even when TROSPA and other nuclear genes were used, double peaks and signs of hybridization and introgression have not been reported previously. In conclusion, we offer a fast and reliable multiplex PCR method for the identification of I. ricinus and I. inopinatus. Morphological similarity to I. ricinus and phylogenetic analyses both suggest African I. inopinatus to be “a species in statu nascendi” evolving from I. ricinus. Additional studies on the genetic diversity and the full genome sequencing of Ixodes ricinus/inopinatus in North Africa and regions of likely sympatry with I. ricinus in Europe (Spain, Portugal, Italy) are needed. Questions including the potential differences in vector competence between I. ricinus and I. inopinatus and their hybrids remain to be answered.