Molecular Identification and Phylogenetic Analysis of Laelapidae Mites (Acari: Mesostigmata)

Simple Summary Mites from the family Laelapidae are frequently associated with small mammals, mainly rodents, and can be found on their body surface or in their nests. Classification of the Laelapidae is complicated because of high levels of their morphological and ecological variability. This study aimed to undertake molecular characterization and to assess the phylogenetic relationship among eight Laelapidae mite species collected from different rodent hosts in Lithuania, Norway, Slovakia, and the Czech Republic using the nuclear and mitochondrial molecular markers. Our study provides new molecular data on Laelaps agilis, Laelaps hilaris, Laelaps jettmari, Haemogamasus nidi, Eulaelaps stabularis, Hyperlaelaps microti, Myonyssus gigas, and Hirstionyssus sp. mites collected from seven different rodent hosts and three geographical regions in Europe. This study, for the first time, registered sequences of four mite species: H. microti, Hirstionyssus sp., M. gigas, and E. stabularis. Abstract The family Laelapidae (Dermanyssoidea) is morphologically and ecologically the most diverse group of Mesostigmata mites. Although molecular genetic data are widely used in taxonomic identification and phylogenetic analysis, most classifications in Mesostigmata mites are based solely on morphological characteristics. In the present study, eight species of mites from the Laelapidae (Dermanyssoidea) family collected from different species of small rodents in Lithuania, Norway, Slovakia, and the Czech Republic were molecularly characterized using the nuclear (28S ribosomal RNA) and mitochondrial (cytochrome oxidase subunit I gene) markers. Obtained molecular data from 113 specimens of mites were used to discriminate between species and investigate the phylogenetic relationships and genetic diversity among Laelapidae mites from six genera. This study provides new molecular data on Laelaps agilis, Laelaps hilaris, Laelaps jettmari, Haemogamasus nidi, Eulaelaps stabularis, Hyperlaelaps microti, Myonyssus gigas, and Hirstionyssus sp. mites collected from different rodent hosts and geographical regions in Europe.


Introduction
Mesostigmata mites represent the most taxon-rich group of Parasitiformes and comprise approximately 11,000 described species [1]. Numerous species of mesostigmatic mites can occasionally infest humans and cause dermatitis and severe allergic reactions. These mites can be potential vectors of the human pathogenic tick-borne encephalitis virus (TBEV) [2] and various rickettsial agents [3][4][5][6]. The superfamily Dermanyssoidea is the largest subdivision of mesostigmatid mites. It consists of 15 families [7], including Laelapidae, which is morphologically and ecologically the most diverse group of
All trapped rodents were marked and identified by species level and sex. Ectoparasites were collected using soft tweezers, placed into 1.5 mL tubes with 70% ethanol solution, and then stored at 4 • C until processed. The collected mites were determined using morphological identification keys by Mašán, Fend'a [15], Bregetova [30], Baker [31], and Kaminskienė et al. [32].

DNA Extraction
Ammonium hydroxide solution (2.5%) was used for DNA extraction from mites. The laelapid mites were taken from the ethanol solution, dried (3-5 min) on the paper towel at room temperature, and then put in a 0.5 mL microcentrifuge tube. A quantity of 40 µL of 2.5% NH 4 OH solution was added for each adult mite. In the solution the mites were crushed with a sterile plastic pestle and stored at room temperature for 30 min until incubated at 100 • C for 30 min, allowing for maximal DNA recovery. Subsequently, the tubes were centrifuged at 13,000/min for 1 min to collect condensate from the cap and sides of the tube. All opened tubes with the solution were placed back in the heating block and incubated at 100 • C for 20 min to evaporate the ammonia. After incubation, the tubes were closed and placed on the ice for 2-3 min. Then tubes were centrifuged at 13,000/min for 30 s. Extracted DNA was stored at −20 • C until further usage.

PCR Amplification and Sequencing
Domains 1-3 from the 28S nuclear ribosomal RNA gene region and the COI gene of mitochondrial DNA were used for molecular characterization and phylogenetic reconstruction within the family Laelapidae [33].
PCR products were subjected to electrophoresis on 1.5% agarose gel and analyzed by UV transilluminator. The DNA fragment was excised from agarose gel and purified using a GenJET PCR purification kit (Thermo Fisher Scientific Baltics, Vilnius, Lithuania) according to the manufacturer's protocol. All purified PCR products were sent for DNA sequencing to a sequencing service (Macrogen, Amsterdam, The Netherlands).

Sequence Analysis
The sequences obtained in this study were analyzed using the BLAST program to confirm the morphological identification of mite species and were aligned with the corresponding sequences of other laelapid mites available in GenBank using ClustalW [34] multiple alignments implemented in MegaX [35]. The partial 28S rRNA and COI gene sequences were aligned in two independent datasets. The intraspecific and interspecific pairwise genetic distances, variable sites, conserved sites, and parsimony-informative sites were computed by Mega X. The non-synonymous mutation rate (Ka) and synonymous mutation rate (Ks), haplotype diversity (Hd), nucleotide diversity (Π), and polymorphic sites (S) were calculated using DnaSP v5.10.01 [36]. The representative sequences of 28S rRNA and COI gene were deposited to GenBank.

Phylogenetic Analysis
Phylogenetic trees were constructed using maximum likelihood (ML) and Bayesian inference (BI) methods. The best-fitting nucleotide substitution model (GTR + I + G) was determined by the Bayesian Information Criterion (BIC) yielded using jModelTest v2.1.10 [37]. The ML trees were generated using the Tamura-Nei parameter model in MEGA X, with each node supported by 1000 bootstraps. Bayesian inference (BI) analyses were run with MrBayes v.3.2.7 [38]. The Markov chain was run with 40,000,000 generations, and trees were sampled every 1000th generation. The first 25% of samples were discarded as burn-in, and the remaining saved samples were used to estimate the posterior probabilities (PP) of each bipartition. The phylogenetic tree was visualized using FigTree v1.4.4 [39].
To estimate the phylogenetic relationships among the COI gene haplotypes of L. agilis derived from different rodent hosts and geographical regions, median-joining (MJ) networks were constructed using Network 10.2.0.0 [40].

Samples
The phylogenetic tree of 28S rRNA gene sequences constructed using the ML method is divided into two main clusters: one cluster groups sequences of twelve Laelaps genus species and H. microti, while the other cluster consists of six species of Hirstionyssus, Haemogamasus, Myonyssus, Brevisterna, and Eulaelaps genera. The members of each species form individual subclusters on the phylogenetic tree ( Figure 1).  The 28S rRNA gene sequences of L. jettmari and L. hilaris obtained in the present study were 100% identical to corresponding sequences derived from GenBank: GU440635 and GU440637, respectively ( Figure 1). Sequences of Hg. nidi (MZ061928, MZ061929, MZ061931, MZ061930) collected in Lithuania shared 98.95-99.08% similarity to Hg. reidi (synonym Hg. nidi) sequences from GenBank: GU440583.

COI Gene
The partial sequences of the COI gene were successfully obtained from six species of Laelapidae mites (L. agilis, L. jettmari, L. hilaris, Hg. nidi, H. microti, and M. gigas) collected from six species of small rodents (A. flavicollis, A. agrarius, A. sylvaticus, C. glareolus, M. arvalis and M. oeconomus). A total of 60 good-quality COI sequences were analyzed (among them 47 sequences of L. agilis, four sequences of L. jettmari, three sequences of L. hilaris, two sequences of Hg. nidi, two sequences of M. gigas, and two sequences of H. microti). COI sequences of Laelapidae mites ranged from 582 to 699 bp in length and from 64.9 to 74.6% in AT content (Table 4); there were 253 variable sites, 330 conserved sites, and 245 parsimony-informative sites. A total of 23 nucleotide variable sites were detected among L. agilis species ( Table 5). The mean value of Ka/Ks of COI gene sequences obtained in this study was 2.31.   Nine COI haplotypes (h = 9) between 23 L. agilis sequences were detected with estimated haplotype diversity of Hd = 0.870, nucleotide diversity Π = 0.00720, and a total number of polymorphic sites S = 23. In total, 559 conserved sites, one singleton site, and 19 parsimonyinformative sites were detected. Haplotype H_1 of L. agilis was the most frequent. It was found in three out of four different locations (Lithuania, Slovakia, and the Czech Republic) (Table 4, Figure 2). Haplotypes H_2 and H_3 (the Czech Republic), H_4-H_8 (Lithuania), and H_9 (Norway) of L. agilis were specific for their respective sampling locations. Nine COI haplotypes (h = 9) between 23 L. agilis sequences were detected with estimated haplotype diversity of Hd = 0.870, nucleotide diversity Π = 0.00720, and a total number of polymorphic sites S = 23. In total, 559 conserved sites, one singleton site, and 19 parsimony-informative sites were detected. Haplotype H_1 of L. agilis was the most frequent. It was found in three out of four different locations (Lithuania, Slovakia, and the Czech Republic) (Table 4, Figure 2). Haplotypes H_2 and H_3 (the Czech Republic), H_4-H_8 (Lithuania), and H_9 (Norway) of L. agilis were specific for their respective sampling locations.  In this study, six haplotypes of L. agilis were detected in Lithuania. From these sequences, four haplotypes of L. agilis detected in Lithuania (H_4, H_5, H_7, and H_8) were unique and differed from the most similar sequences in GenBank (Figure 2A). The distribution of L. agilis haplotypes in different areas of Lithuania showed that the highest haplotype diversity was detected in the Lithuanian coastal area-the Curonian Spit where five of six haplotypes (H_1, H_5-H_8; n = 21) were found. In the continental part of the country (northern and south-eastern parts), three haplotypes were detected (H_1, H_4, H_6; n = 13) (Figure 3). Distribution of different L. agilis haplotypes did not reveal specificity to host species. Five haplotypes were detected in A. flavicollis, four haplotypes in C. glareolus, and A. agrarius, M. oeconomus, and M. minutus each harbored one haplotype H_4, H_6, and H_7, respectively. This study detected three COI haplotypes of L. jettmari (n = 4) and two COI haplotypes of H. microti (n = 2). In contrast, only one haplotype was found among L. hilaris, Hg. Nidi, and M. gigas sequences (Table 4). In this study, six haplotypes of L. agilis were detected in Lithuania. From these sequences, four haplotypes of L. agilis detected in Lithuania (H_4, H_5, H_7, and H_8) were unique and differed from the most similar sequences in GenBank (Figure 2A). The distribution of L. agilis haplotypes in different areas of Lithuania showed that the highest haplotype diversity was detected in the Lithuanian coastal area-the Curonian Spit where five of six haplotypes (H_1, H_5-H_8; n = 21) were found. In the continental part of the country (northern and south-eastern parts), three haplotypes were detected (H_1, H_4, H_6; n = 13) (Figure 3). Distribution of different L. agilis haplotypes did not reveal specificity to host species. Five haplotypes were detected in A. flavicollis, four haplotypes in C. glareolus, and A. agrarius, M. oeconomus, and M. minutus each harbored one haplotype H_4, H_6, and H_7, respectively. This study detected three COI haplotypes of L. jettmari (n = 4) and two COI haplotypes of H. microti (n = 2). In contrast, only one haplotype was found among L. hilaris, Hg. Nidi, and M. gigas sequences (Table 4). The overall mean genetic distance between laelapid mites' COI gene sequences obtained in this study was 0.1215. The inter-and intraspecific genetic distances based on the COI gene are shown in Table 3  The overall mean genetic distance between laelapid mites' COI gene sequences obtained in this study was 0.1215. The inter-and intraspecific genetic distances based on the COI gene are shown in Table 3. The highest interspecific distances were detected between M. gigas and the other Laelapidae mite species. The intraspecific genetic distance among L. agilis sequences was 0.0074.
The phylogenetic analysis based on the COI gene included sequences of other dermanysoid mite species available in GenBank: Laelaps muricola (KU166735; KU166676; KU166784; KU166789), Laelaps giganteus (KU166660; KU166413; KU166425), L. kochi (MF914881; MG413303), Haemogamasus ambulans (KM831963), Gaeolaelaps debilis (MW367907), E. stabularis (OP960202), and Dermanyssus hirundinis (MN355089). The phylogenetic tree of COI gene sequences constructed using the ML method showed a clear separation of different species of Laelapidae mites into different clusters. L. agilis sequences were heterogenic and, together with L. jettmari and L. hilaris, formed a separate cluster on the phylogenetic tree ( Figure 4). (MW367907), E. stabularis (OP960202), and Dermanyssus hirundinis (MN355089). The phylogenetic tree of COI gene sequences constructed using the ML method showed a clear separation of different species of Laelapidae mites into different clusters. L. agilis sequences were heterogenic and, together with L. jettmari and L. hilaris, formed a separate cluster on the phylogenetic tree ( Figure 4). Another phylogenetic tree of Laelapidae mites was constructed using the BI method ( Figure 3). ML and BI phylogenetic trees differed slightly in topology and branching structures (Figures 4 and 5). The Bayesian tree ( Figure 5  Another phylogenetic tree of Laelapidae mites was constructed using the BI method ( Figure 3). ML and BI phylogenetic trees differed slightly in topology and branching structures (Figures 4 and 5). The Bayesian tree ( Figure 5  NO-Norway. The geographical structure of L. agilis was supported by the median-joining network, which showed at least three major Lineages A, B, and C (Figure 2A,B). In the median-joining (MJ) network were included the COI gene haplotypes of L. agilis (derived from different rodent hosts and geographical regions) detected in our study and available in GenBank. Haplotypes assigned to Lineage A were found in Norway, Finland, the United Kingdom, France, and the Czech Republic. Most of the haplotypes were not shared between different geographic areas, except for one haplotype identified in Germany and the Czech Republic ( Figure 2A). This lineage (A) showed haplotype sharing between three host species: A. flavicollis, A. sylvaticus, C. glareolus ( Figure 2B). A single haplotype was assigned to Lineage B, shared between the L. agilis from Italy, Greece, and the Czech Republic, and three unique to the individuals from Italy, Bulgaria, and Serbia. This lineage (B) only included mites collected from A. flavicollis. All sequences of L. agilis from Lithuania belonged to Lineage C. They clustered together with samples from Slovakia, Bulgaria, Austria, the Czech Republic, Hungary, and Finland. Additionally, the network showed that four haplotypes have so far been found only in Lithuania (Figure 2A). This lineage (C) was composed of L. agilis found in different host species (A. flavicollis, A. sylvaticus, A. agrarius, C. glareolus, M. minutus, and M. oeconomus) (Tables 2B and 3).

Discussion
In the present study, eight species of Laelapidae mites collected from different rodent hosts and geographical regions in Europe were molecularly characterized based on both nuclear 28S rRNA and mitochondrial COI gene regions. Our findings confirm that these molecular markers could be successfully used for molecular identification of Laelapidae mite species and inference of their phylogenetic relationships [7,[27][28][29]. On the other hand, mitochondrial DNA evolves much faster and is more evolutionarily variable than the ribosomal DNA of the nuclear genome [41]. Thus, the COI gene sequences are more appropriate for analyzing intraspecific phylogenetic relationships [26,42]. In this study, our results based on the COI gene indicated a high intraspecific variation (9 haplotypes out of 23 obtained sequences) in L. agilis species. Intraspecific variations on the COI gene were also detected in L. jettmari (three haplotypes identified among four obtained sequences) and H. microti (two haplotypes among two obtained sequences).
Our findings provide new data on the intra-and interspecific phylogenetic relationships of Laelapidae mites belonging to six genera. This study, for the first time, registered sequences of four mite species: H. microti, Hirstionyssus sp., M. gigas, and E. stabularis.
Phylogenetic relationships based on 28S rRNA exhibited polyphyly of the different species from the family Laelapidae. The previous study also determined a polytomy structure in the phylogenetic relationships [7]. In contrast, Li et al. [43] and Yang et al. [44] showed that based on mitochondrial barcoding region, the family Laelapidae is a monophyletic group.
The results of the phylogenetic analysis based on 28S rRNA revealed the separation of Laelapidae mites into two different groups. The first group consists of sequences belonging to obligate parasitic mites from two genera, Laelaps and Hyperlaelaps. The second group contains two clusters-one cluster consists of sequences belonging to facultative parasitic mites Eulaelaps, Haemogamasus, and Myonyssus, whereas sequences of obligate parasitic Hirstionyssus sp. formed a separate cluster (Figure 1).
It should be noticed that phylogenetic analysis based on both genes (28S rRNA and COI) indicated the clustering of H. microti with the species of the genus Laelaps and did not show separation into distinct clades. The differences between the molecular and morphological taxonomy of this species were also observed in recent studies [29,44].
In line with a previous study [28], our results of the phylogenetic analysis based on mt DNA also corroborated three lineages (Lineages A, B, and C) within L. agilis ( Figure 2). The results did not indicate clear specificity according to geographical locations. Lineages A and C comprised specimens from diverse geographical regions of Europe (North, Central-Eastern, and West) (Figure 2A), which was also revealed in a recent study [28]. However, our results supplemented Lineage A with one specimen from Norway and Lineage C with sequences from Lithuania ( Figure 2A). Moreover, our findings showed no clear host species specificity and confirmed the results previously obtained by Nazarizadeh et al. [28]. However, the number of host species in these lineages (A and C) was supplemented by three additional species (A. agrarius, M. minutus, and M. oeconomus) in this study. Only one L. agilis lineage (B) showed clear specificity according to host species (A. flavicollis) ( Figure 2B), and it is consistent with the results of the Nazarizadeh et al. study [28].
Considering several species of rodents as important hosts of the parasitic mites analyzed in this study, it should be mentioned that populations of rodents of the genera Apodemus and Clethrionomys in Europe are genetically heterogeneous. During the glaciation in the Quaternary, they survived in various refugia in southern Europe [45,46] and had complex recolonization routes in Europe. A specific species in this regard is Apodemus agrarius, which only relatively recently colonized Europe from Asia [47].
Based on published data, at least 21 parasitic mite species belonging to the Laelapidae family have been morphologically identified in Lithuania [32,[48][49][50][51]. This study provides the first molecular characterization of eight species of laelapid mites collected from different rodent hosts in Lithuania. Therefore, the more comprehensive phylogenetic analysis of Laelapidae mites in Lithuania must be further investigated.

Conclusions
Our study provides new molecular data on Laelaps agilis, Laelaps hilaris, Laelaps jettmari, Haemogamasus nidi, Eulaelaps stabularis, Hyperlaelaps microti, Myonyssus gigas, and Hirstionyssus sp. mites collected from seven different rodent hosts and three geographical regions in Europe. This study is the first molecular characterization of eight Laelapidae mite species in Baltic countries. Specifically, 28S rRNA and COI sequences of four mite species were, for the first time, registered in the NCBI database (2021-2022).

Data Availability Statement:
The data presented in this study are available within the article.