Introduction

Ticks, as important ectoparasites, infect a broad range of animals and humans. They may also act as vectors of a range of pathogens (viruses, bacteria, fungi and protozoa) causing human and animal disease [3, 34, 36]. There are currently 896 species of ticks in the 3 families: the Ixodidae (hard ticks), the Argasidae (soft ticks), and the Nuttalliellidae. The Ixodidae is the most important family of veterinary and medical significance consisting of 702 species in 14 genera [15].

Haemaphysalis longicornis belongs to the Ixodidae and is distributed in Australia, New Zealand and eastern Asia [17]. This hard tick infests a variety of hosts (such as pigs, sheet, goat, cattle and dog), and causes lesions, dermatitis, weight loss, blood loss and even death [22]. More importantly, H. longicornis is a vector of many pathogens, such as Anaplasma spp., Babesia ovata, Ehrlichia canis and Rickettsia conorii [23, 30, 37, 38]. In China, H. longicornis is distributed in almost all the provinces or regions and is the dominant tick species in wild and domestic animals, causing major economic losses of livestock [31].

Mitochondrial DNA (mtDNA genes) sequences are preferred and reliable molecular markers for studying genetic diversity and phylogenetics in hard ticks [4, 18, 24, 25], given the advantages of mtDNA as a marker in the molecular research of species [14]. For example, mt cox1 (cytochrome c subunit 1) is a useful genetic marker for the identification and differentiation of ticks within the genus Rhipicephalus [26]. A recent study showed that mt genes can be used as standard genetic markers in discerning the genetic assemblages of R. microplus [27]. Furthermore, mt gene sequences are useful markers for studying genetic variation and phylogenetics of H. flava and R. sanguineus sensu lato tick populations [5, 25]. However, limited information is available about mt gene sequences of hard tick H. longicornis [6, 7]. Therefore, the objectives of the present study were to examine genetic variation and population genetics in three mt genes [cox1, rrnL (large subunit ribosomal RNA) and nad5 and nad5 (NADH dehydrogenase 5)] among H. longicornis isolates in China and to infer the phylogenetic relationships of H. longicornis with other Haemaphysalis ticks.

Materials and Methods

Parasites

All adult ticks of H. longicornis (n = 49) were obtained from different hosts (goat: Capra hircus; hedgehog: Erinaceus europaeus and cattle: Bos taurus) and four provinces in China (Table 1). The goat, hedgehog and cattle were fed and tick specimens were collected by the veterinarians. The ticks were repeatedly washed in physiological saline, fixed in 70% (v/v) ethanol and stored at − 20 °C until use. These tick samples were preliminarily identified based on host preference and morphological characters [9].

Table 1 The GenBank accession numbers for the three mtDNA sequences of 49 isolates Haemaphysalis longicornis from different hosts and geographical origins in China
Full size table

DNA Extraction, Genotyping and DNA Sequencing

Total genomic DNA was isolated from individual ticks using sodium dodecyl sulfate/proteinase K treatment, followed by spin column purification (Wizard® SV Genomic DNA Purification System, Promega, Madison, Wisconsin, USA). The primer sets (Table 2) were designed based on well-conserved mt sequences of H. longicornis (KY172954) and H. flava (NC_005292). PCR reactions (25 μL) were performed in 3.0 μL of MgCl2 (25 mM), 0.25 μL of each primer (50 pmol/μL), 2.5 μL 10 × rTaq buffer (100 mM Tris–HCl and 500 mM KCl), 2 μL of dNTP mixture (2.5 mM each), 0.25 μL of rTaq (5 U/μL) DNA polymerase (TaKaRa Biotechnology, Dalian, China) and 2 µL of DNA sample in a thermocycler (Biometra, Göttingen, Germany). The cycling conditions were: 95 °C for 5 min (initial denaturation), followed by 40 cycles of 95 °C for 30 s (denaturation), 55 °C for 1 min (annealing), 72 °C for 2 min (extension) and then 72 °C for 5 min (final extension). Negative control (without DNA template) was included in each amplification run. PCR products were visualized by electrophoresis in 1% (w/v) agarose gel to validate amplification efficiency. PCR products were sent to Sangon Company (Shanghai, China) for sequencing from both directions.

Table 2 Sequences of primers used to amplify a portion of the cox1, rrnL and nad5 of Haemaphysalis longicornis in the present study
Full size table

Sequence Analysis and Reconstruction of Phylogenetic Relationships

The sequences were aligned with the available mt gene sequences (such as H. longicornis and H. flava) using the software MAFFT 7.122 [20]. Genetic diversity values, including polymorphic sites, A + T contents, haplotype number, haplotype diversity, average number of nucleotide differences, Tajima’s D, Fu’s Fs tests, mismatch distribution and nucleotide diversity of each gene were calculated using DnaSP v5.0 [28]. Genetic differentiation within and among four different populations was estimated using analysis of molecular variance (AMOVA) implemented in Arlequin v3.5 [10]. Gene flow (Nm) among four populations was calculated as follows: (1-Fst)/2Fst using DnaSP v5.0 for 1000 permutations.

Phylogenetic trees inferred from the combined three mt gene datasets were constructed using the maximum likelihood (ML) method. In addition, H. formosensis (NC_020334), H. parva (NC_020335) and H. flava (NC_005292) were also included in the present study, with Ixodes pavlovskyi (NC_023831) as the outgroup. Phylogenetic analyses were performed using PhyML 3.0 [16]. The best fitting model with its parameter (GTR + I+G) of these sequence datasets was determined using JModeltest [32] based on the Akaike information criterion (AIC). ML analyses were checked on the basis of 100 bootstrap replicates (Br). Phylograms were drawn using FigTree v.1.31 (http://tree.bio.ed.ac.uk/software/figtree/).

Results and Discussion

The mt cox1, rrnL and nad5 gene regions were amplified and sequenced individually from 49 H. longicornis samples that obtained sequences which were deposited in GenBank database (Table 1). The lengths of the mt sequences of cox1, rrnL and nad5 were 776 bp, 409 bp and 510 bp, respectively. The 49 mt sequences were closely related to H. longicornis mt sequences, and three mt gene regions (cox1, rrnL and nad5) had more than 98.0% identity to previously published mt sequences for H. longicornis from Australia and China (GenBank accession nos. AF132820, KU986720 and FN394350, respectively).

The A + T content of these sequences was 67.0–67.7% for cox1, 77.8–78.0% for rrnL and 73.2–74.6% for nad5, respectively. The intra-specific sequence variation within H. longicornis was 0–2.8% for cox1, 0–2.9% for rrnL and 0–6.7% for nad5, however, the inter-specific sequence differences among other members of the genus Haemaphysalis were 13.8–15.3% for cox1, 14.7–15.7% for rrnL and 13.3–18.1% for nad5. Similarly, sequence diversity has also been detected in H. flava [25], H. qinghaiensis [29] and H. punctate [6] by analysis of mt gene sequences. These studies have clearly indicated that mt gene sequences provide reliable genetic markers for identification and differentiation of Haemaphysalis species.

Many studies have indicated that mt sequences are unique genetic markers to indicate geographical movements and population genetic structure of parasites [11,12,13]. In the present study, 22 polymorphic sites, 11 haplotypes, Hd = 0.696 and Pi = 0.00917 were determined in all sequences of pcox1. 11 polymorphic sites, 3 haplotypes, Hd = 0.041 and Pi = 0.0011 were determined in all sequences of prrnL. 34 polymorphic sites, 17 haplotypes, Hd = 0.849 and Pi = 0.01296 were determined in all sequences of pnad5 (Table 3). A moderate level of haplotype diversity (except for Yunnan) was maintained in the H. longicornis populations, but their nucleotide diversity was relatively low due to the richness of single-nucleotide substitutions. Similar results also were reported for I. ricinus in Baltic countries [33]. The low level of nucleotide diversity was found across all four H. longicornis populations, revealing a relative lack of genetic variation across the H. longicornis, regardless of geographical origin and hosts. The combined three mt gene sequences of H. longicornis gave a negative Tajima’s D value of − 0.2 (P > 0.05) and Fu’s Fs tests value of − 2.105 (P > 0.05). The positive values from Tajima’s D test signify that H. longicornis might not have experienced population expansion in the past. The H. longicornis sequences had a negative Tajima’s D value of − 0.2 and Fu’s Fs tests value of − 2.105, but the result was not statistically significant (P > 0.05). The genetic differentiation (66.2%) was mainly observed among populations, while the remaining 33.8% was observed between individuals within populations. These results indicated that there was higher genetic differentiation among populations across the four H. longicornis populations examined here. The AMOVA analysis has also confirmed that there was significant genetic differentiation across the four H. longicornis populations in the China mismatch distribution analysis of the combined three gene datasets which revealed the presence of a multi-peak not shown and a low rate of gene flow value (Nm = 0.18). The low levels of gene flow indicated less gene flow among the H. longicornis populations from the four provinces over time. Some studies have observed that R. appendiculatus went through a demographic expansion in Kenya [2, 21, 35]. However, our finding suggests that there was no rapid expansion event in the demographic history of all four H. longicornis populations.

Table 3 Genetic diversity indices of cox1, rrnL and nad5 genes in Haemaphysalis longicornis from four populations in China
Full size table

mtDNA genes are useful molecular markers for phylogenetic studies of many ectoparasites, including ticks [1, 8, 19, 21]. In the present study, all the H. longicornis isolates were grouped together, indicating that all studied isolates represent the species H. longicornis (Fig. 1). The H. longicornis forms a monophyletic group with high statistical support (Br = 100), and all the H. longicornis isolates were segregated into two major clades (Fig. 1). Isolates from Shandong and Yunnan provinces clustered together in one clade with high statistical support (Br = 97) (Fig. 1). However, isolates from Hunan and Henan provinces clustered together in another clade without reflecting geographical origin, with weak statistical support (Br = 31) (Fig. 1). Our results suggest that H. longicornis may exist in multiple genotypes or distinct lineages. A previous study also supports the division of I. scapularis into several distinct lineages based on mt cox1 and 16S genes, and nuclear genes (serpin2, ixoderin B and lysozyme) [35].

Fig. 1
figure 1

Phylogenetic relationship among Haemaphysalis longicornis isolates in China with other Haemaphysalis species inferred by maximum likelihood analyses using the combined dataset (cox1 +  rrnL + nad5), with Ixodes pavlovskyi (NC_023831) as outgroup. Bootstrapping frequency (Br) values were indicated at nodes

Full size image

In conclusion, our study made the first attempt to characterize genetic variation of H. longicornis isolated from different hosts and four provinces in China, by comparing and analyzing mt cox1, rrnL and nad5 genes. These datasets of H. longicornis provide a potential novel genetic marker for molecular epidemiology of H. longicornis in animals.