Mitochondrial gene heterogeneity of the bat soft tick Argas vespertilionis (Ixodida: Argasidae) in the Palaearctic

Background Recently, a high degree of mitochondrial gene heterogeneity was demonstrated between conspecific ixodid ticks of bats in Eurasia. Argas vespertilionis is a soft tick species of mainly vespertilionid bats, also with a wide distribution in the Old World. The aim of this study was to investigate the morphology, mitochondrial gene heterogeneity and host range of A. vespertilionis in the Old World. Results Altogether 318 soft tick larvae were collected from 17 bat species (belonging to six genera) in seven countries. Based on the general morphology (setal arrangement) of 314 A. vespertilionis larvae, and the detailed measurements of fifteen larvae, only minor morphological differences (in dorsal plate size and the type of serrate setae) were observed between specimens from Europe and Vietnam. On the other hand, cytochrome c oxidase subunit 1 (cox1) and 16S rRNA gene sequence analyses of 17 specimens showed that A. vespertilionis from Europe is genetically different (with up to 7.5% cox1 and 5.7% 16S rRNA gene sequence divergence) from specimens collected in Vietnam, and their phylogenetic separation is well supported. Conclusion In its evaluated geographical range, no larval phenotypic differences justify the existence of separate species under the name A. vespertilionis. However, phylogenetic analyses based on two mitochondrial markers suggest that it represents a complex of at least two putative cryptic species. The broad host range of A. vespertilionis might partly explain its lower degree of mitochondrial gene heterogeneity in comparison with ixodid bat tick species over the same geographical region of Eurasia.


Background
Soft ticks (Acari: Ixodida: Argasidae) include approximately 200 species that are obligate haematophagous ectoparasites of hosts from all classes of terrestrial vertebrates. Argas vespertilionis is a cosmopolitan soft tick species of mainly vespertilionid bats (especially Pipistrellus spp.), with a wide distribution in the Old World, from the Palaearctic to South Africa [1]. Apart from bats, A. vespertilionis was reported to feed on humans [1,2] and domestic animals [3]. This soft tick species is a potential vector of zoonotic viruses [3] and bacteria [4], and the piroplasm Babesia vesperuginis [5].
The taxonomy of the Argasidae is controversial, because the majority of soft tick species can be assigned to more than one genus [6]. Accordingly, the taxonomical status of A. vespertilionis also appears to be uncertain. It was originally the type species of the genus Carios [7], but in the most recent list of valid tick names it is mentioned as a member of the genus Argas [8]. Recent phylogenetic analyses do not support A. vespertilionis as a member of the genus Argas, as this species has been misplaced into the subfamily Argasinae [9], and based on its 12S rRNA gene it should belong to Ornithodorinae [6]. In addition, the homogeneity of A. vespertilionis on the species level has long been questioned [7].
Recently, high degree of mitochondrial gene heterogeneity was demonstrated between conspecific ixodid ticks of bats in Eurasia [10]. The aim of this study was to investigate A. vespertilionis in the same context, i.e. its morphology, mitochondrial gene heterogeneity and host range in the Old World.
To date, molecular analyses of soft ticks focused on the 16S rRNA gene, which has the potential for resolving phylogenetic relationships among closely related species in Argasidae [6]. In addition to this gene, the 5' region of the mitochondrial cytochrome c oxidase subunit 1 (cox1) gene was chosen for phylogenetic analysis of A. vespertilionis in the present study, because it is regarded as the standard marker for tick species identification by DNA barcoding [11], and is particularly suitable to track separation among soft tick species [12]. Since there are few soft tick cox1 reference sequences in GenBank, a South African isolate of A. transgariepinus and a neotropical isolate of an Ornithodoros sp. were also included in the cox1 phylogenetic analysis. The nomenclature used in the manuscript complies with the valid tick names listed by Guglielmone et al. [8], who follow Hoogstraal in his classification of the Argasidae, although that classification, including the re-assignment of Carios as a subgenus of Argas, has never been supported by any analysis using actual data.

Sample collection and morphological analyses
Soft ticks were collected from bats captured for ringing and monitoring purposes ( Table 1). All ticks were stored in 70% ethanol. Morphological identification was based on the description of A. vespertilionis and A. transgariepinus larvae by Hoogstraal [7,13], and of Ornithodoros larvae (on the genus level) according to Barros-Battesti et al. [14], and Jones & Clifford [15]. Structures of representative specimens from each country (A. vespertilionis: eight larvae from Vietnam, three larvae from Italy, four larvae from Romania; except A. vespertilionis from Kenya, which was damaged) were measured under a Jenaval light microscope (Carl Zeiss GmbH, Jena, Germany) after clearance with lactic acid. The means of these data sets were compared by using two-tailed Student's t-test, and were considered significantly different if P < 0.05.

DNA extraction, molecular and phylogenetic analyses
DNA was extracted from the larvae with the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instruction, including an overnight digestion in tissue lysis buffer with 6.6% Proteinase-K at 56°C. Twenty DNA extracts were used in this study for molecular analyses (eight from Hungary, two from Romania, three from Italy, four from Vietnam, one from Kenya, one from South Africa and one from Mexico). From these samples two mitochondrial markers were amplified: a 710 bp long fragment of the cytochrome c oxidase subunit 1 (cox1) gene, and an approx. 460 bp part of the 16S rRNA gene, as reported [16].
PCR products were visualized in 1.5% agarose gel. Purification and Sanger dideoxy sequencing (twice for each sample) were done by Biomi Inc. (Gödöllő, Hungary). Obtained sequences were manually edited, then aligned and compared to reference GenBank sequences by nucleotide BLASTN program (https://blast.ncbi.nlm.nih.gov). Representative sequences were submitted to GenBank (see accession numbers in Table 1). The MEGA model selection method was applied to choose the appropriate model for phylogenetic analyses. In the phylogenetic analyses reference sequences with high coverage (i.e. 98-100% of the region amplified here) were retrieved from GenBank, and trimmed to the same start and stop positions (cox1: 652 bp in length, 16S rRNA gene: 439-442 bp in length). This dataset was resampled 1,000 times to generate bootstrap values. Phylogenetic analyses were conducted with the Maximum Likelihood method and Tamura-Nei model by using MEGA version 6.0.

Results
Altogether 318 soft tick larvae were collected from 17 bat species (belonging to six genera) in seven countries. All, except four soft tick larvae, were morphologically identified as A. vespertilionis ( Table 1). The majority of A. vespertilionis larvae (59.1%: 188 out of 318, CI: 53.5-64.6%) were found on Pipistrellus spp. (Table 1). Myotis alcathoe is a new host for this soft tick species.
One specimen from South Africa was identified as A. transgariepinus (based on idiosomal setae, palpal articles, coxae and tarsus I). Three larvae from Mexico represented the genus Ornithodoros (based on the elongated piriform dorsal plate with non-parallel sides), but could not be identified on the species level because of the lack of hypostome.
Measurements of selected, diagnostically important structures of A. vespertilionis larvae revealed no significant differences between specimens from Europe and Vietnam (Table 2), except for the length and width of the dorsal plate (plate length of ticks from Italy/Romania vs Vietnam: t = 3.49, df = 13, P = 0.004; plate width of ticks from Italy/Romania vs Vietnam: t = 3.21, df = 13, P = 0.006). Dorsal plate shape index (length:width ratio), as well as hypostome shape and dentition (4/4 anteriorly, 2/2 behind apex) were not significantly different between these categories. Sternal and anal setae were consistently pointed (needle-like), whereas dorsal setae were serrate. The morphology of serrate setae showed minor difference between geographically distant specimens ( Fig. 1 Vietnam). Clustering of A. vespertilionis isolates with two members of Ornithodorinae received moderate (72%) support (Fig. 2).

Discussion
In this study 314 A. vespertilionis larvae were collected and morphologically investigated. Finding of only larvae of soft ticks on bats is in line with the life cycle of A. vespertilionis, i.e. larvae (unlike nymphs and adults) suck blood for several weeks on their bat hosts (14-31 days: [1]), therefore almost exclusively these can be collected from bats. In the present study A. vespertilionis was found on 15 bat species, most of which are already reported hosts (including four Pipistrellus spp., seven Myotis spp. and two Eptesicus, as well as two Nyctalus spp. [17,18]).
While A. vespertilionis is mentioned in the most recent list of valid tick names as a member of the genus Argas [8], morphological and phylogenetic analyses do not support this assumption [6,9]. Instead, based on its 12S rRNA gene, A. vespertilionis was demonstrated to belong to Ornithodorinae [6]. Phylogenetic analyses of the present study also reflected that haplotypes of A. vespertilionis clustered outside the Argasinae.
During the past few decades scientific debate tried to establish morphological features suitable to solve the taxonomical uncertainty among the Argasidae. In the  larval stage the number of setae according to anatomical location is an important feature to recognize genera [14], and the shape of dorsal plate, the morphology of hypostome and the length of setae may be used to distinguish closely related species [15,19]. The possibility that more than one species exist under the name A. vespertilionis has already been suggested by Hoogstraal [7], but this remained hitherto unevaluated. In this study the great majority of relevant parameters were not significantly different between A. vespertilionis larvae from Europe and Vietnam, although these larvae proved to be well separated based on two mitochondrial genetic markers. Similarly, in a previous study comparing neotropical bat soft ticks, selected measurements (length of certain setae) differed slightly between larvae from different countries, but these were considered to represent the same species [19]. Intraspecific variations in body outline of A. vespertilionis have also been reported [7]. Furthermore, despite the differences in the mean length and width of the dorsal plate between A. vespertilionis from Europe and Vietnam, as demonstrated here, these alone cannot serve to delineate species, because the shape of the dorsal plate (reflected here by similar length:width ratios) is regarded as more relevant in this context [15,19].
A minor difference was also observed between the serrate setae of A. vespertilionis larvae collected in distant regions of Eurasia. However, while the types of fringed setae were reported to be different between larvae of closely related Ornithodoros spp. [15], the latter were also shown to differ in the ranges of their setal lengths and hypostome (unlike A. vespertilionis larvae here). Therefore, in the absence of further distinguishing characteristics, the present data suggest that A. vespertilionis in Europe and Vietnam belong to the same species, and observed minor differences (i.e. dorsal plate size) should be interpreted as intraspecific variations between populations. The morphology of specimens analyzed in this study also suggests that they are conspecific with A. vespertilionis reported from Japan [20].
On the other hand, specimens from Europe and Vietnam had cox1 sequence divergence (7.1-7.5%) exceeding that proposed for closely related ixodid tick species (6%, see [11]). Accordingly, morphologically similar, but genetically distinct populations of A. vespertilionis exist in Europe and Southeast Asia, suggesting that this soft tick should be regarded as a complex (group) of at least two putative cryptic species. This seems to be justified from the morphology of the larval stage alone (because differences between argasid larvae served to describe new soft tick species, e.g. in [15]), but morphological investigation of adult specimens from both regions and molecular/phylogenetic analyses of nuclear markers (18S and 28S rRNA genes) should ultimately confirm this conclusion.
Compared in the same context, the sequence divergence between A. vespertilionis from Kenya and Europe was less pronounced than between samples from Europe and Vietnam, suggesting that genetic exchange has been more likely in this direction (although a larger sample size from sub-Saharan Africa is needed to draw final conclusion in this respect). In support of this possibility, some of the main hosts of A. vespertilionis in the present study, most notably Pipistrellus nathusii is known to migrate long distances (up to 1,900 km) in the northeastern to south-western direction [21]. Another important host, P. kuhlii is widespread in certain regions across Europe, the Middle-East, North Africa and Asia [22].
In Eurasia, high degree of mitochondrial gene heterogeneity (i.e. up to 16% cox1 sequence divergence) has recently been demonstrated between ixodid bat ticks that had been regarded as conspecific [10]. This was explained by the preference of each tick species for bat hosts from a single genus, as well as by the geographical separation of relevant bat host species [10]. In comparison with ixodid bat ticks, the less pronounced difference (in terms of both morphology and genetics) between geographically distant isolates of A. vespertilionis, as shown here, may root in the fact that this soft tick species has a broad host spectrum (involving vespertilionid bats from at least six genera, as also shown here), thus preventing complete allopatric separation of its populations.

Conclusions
In its evaluated geographical range, no larval phenotypic differences justify the existence of separate species under the name A. vespertilionis. However, phylogenetic analyses based on two mitochondrial markers suggest that it represents a complex of at least two putative cryptic species. The broad host range of A. vespertilionis might partly explain its lower degree of mitochondrial gene heterogeneity in comparison with ixodid bat tick species over the same geographical region of Eurasia.