Introduction

Haemaphysalis flava is a predominant tick species in China, infesting cattle, pets, wildlife and humans (Chen et al. 2010). Haemaphysalis flava harbors viruses (Tarumizu tick virus, Kabuto Mountain virus, tick-borne encephalitis virus, and others; Yun et al. 2016, Fujita et al. 2017, Ejiri et al. 2018), bacteria (Rickettsiae spp., Coxiella spp., Citrobacter spp., Hepatozoon spp., Cercopithifilaria spp., and others; Duan and Cheng 2017) and protozoa (Toxoplasma gondii, Kim et al. 2020). Bites by those ticks may cause Tularemia (Ozawa et al. 1982), Encephalitis (Ko et al. 2010) and Lyme borreliosis (Moon et al. 2013). Though the ability of H. flava to transmit ‘severe fever with thrombocytopenia syndrome’ (SFTS) has not been extensively confirmed yet, H. flava has been shown to carry SFTS viruses (SFTSV) (Jo et al. 2019). At this point, the significance of H. flava for public health should not be neglected (He and Cheng 2017). Immune prophylaxis using vaccines seems to be a safe and environmental friendly alternative compared to applying pesticides (Šmit and Postma 2015).

AV422 was initially discovered in an investigation of the molecular basis of tick attachment phase in Amblyomma americanum, and it was predicted to contain a signal peptide, suggesting it was secreted extracellularly (Mulenga et al. 2007). Its expression was upregulated after partially feeding for 7 days compared with 1 and 3 days, and was higher in salivary glands compared with cuticles, midguts and ovaries, indicating a possible role in blood-feeding (Mulenga et al. 2007). There was no sequence or amino acid motif analogue of this protein in arthropods and vertebrates. It was injected into hosts as a constituent of saliva during blood-feeding, and elicited a humoral response in hosts (Mulenga et al. 2013). As tick bites frequently got unnoticed, detection of antibodies from host sera using recombinant AV422 could be used as confirmation of tick bites (Mihaljica et al. 2017a, b). Recently, AV422 homologues were found in the sialome of A. americanum (Kim et al. 2016; Lewis et al. 2015), Ixodes scapularis (Radulović et al. 2014), Rhipicephalus microplus (Tirloni et al. 2014) and H. longicornis (Tirloni et al. 2015), suggesting it is a widespread conserved protein among ticks, involved in blood-feeding.

In order to screen for possible antigen candidates for the development of anti-tick vaccines, we have been done multiple omics analysis in H. flava (Xu et al. 2015, 2016; Liu et al. 2018a, b). In proteomic analysis of feces from H. flava, we found a protein showing a high degree of identity to AV422 from A. americanum (Aa-AV422) (GenBank acc. no. KC222016), and retrieved an encoding gene fragment (Contig2959) in transcriptomic library of salivary glands in the same species (Xu et al. 2015). The Contig2959 was identical to the gene encoding AV422 (AV422), but was lacking 3′ and 5′ end as revealed by the sequence alignment. In this study we aimed to investigate conserved features and expression profiles of Hf-AV422 depending on the engorgement and to explore inhibitory roles of recombinant protein (rHf-AV422) in coagulation, to lay a foundation for selection of antigen candidates for the vaccine design against ticks.

Materials and methods

Ticks

Experimental proposals involving animal use were approved by the Institutional Animal Care and Use Committee at Hunan Agricultural University. Live female ticks were collected with the flagging method in Xinyang City, Henan Province, China. Both morphological and molecular identifications were carried out to verify that they were H. flava ticks according to Yan and Cheng (2015). Genes used for the species identification were COX1 and ITS1.

Haemaphysalis flava female adults were fed on rabbits. Their body weights were recorded to designate 1/4, 1/2, 3/4 and fully engorged ticks (naturally detached from hosts; Liu et al. 2020).

Full-length cDNA clone for Hf-AV422

Total RNA of engorged females was extracted by following the manual of EasyPure RNA Kit (TransGen Biotech, Beijing, China). cDNA was produced using EasyScript One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech). The system of RNA reverse transcription reaction was comprised of total RNA (2 µL, 750 ng/µL), Anchored Oligo(dT)18 Primer (1 µL, 0.5 µg/µL), 2 × ES Reaction Mix (10 µL), EasyScript RT/RI Enzyme Mix (1 µL), gDNA Remover (1 µL) and RNase Free dH2O (5 µL). Total RNA, Anchored Oligo(dT)18 Primer and RNase Free H2O mixtures were initially water-bathed at 65 °C for 5 min, then ice-bathed for 2 min. All other constituents in the system were mixed, incubated at 42 °C for 30 min and then at 85 °C for 5 s to synthesize the cDNA.

The 5′ end of Contig2959 was amplified by the 5′-Rapid Amplification of cDNA Ends (5′-RACE) with a SMARTer™ RACE cDNA Amplification Kit (Clontech, Dalian, China). The primer information was listed in Table S1. Reaction programs for the first amplification were an initial 94 °C for 3 min; followed by 20 cycles of 94 °C for 30 s, 68 °C for 30 s, 72 °C for 2 min; and a final 72 °C for 10 min. The second amplification was accomplished by 94 °C for 3 min; 18 cycles of 94 °C for 30 s, 61 °C for 30 s, 72 °C for 24 s; and a final 72 °C for 10 min.

The 3′-RACE was done as well to obtain the 3′ end of Contig2959 using the same kit. The reaction programs for the 1st amplification were similar to those of 5′ RACE except annealing at 59 °C and extension for 42 s. The conditions for the second amplification were comparable to those of 5′-RACE except that the cycle number was 30.

The amplification products were subject to 1.5% agarose gel electrophoresis. Target bands were purified from gels with DNA Purification Kit (Tiangen Biotech, Beijing, China), and cloned into pGM-T vectors (Tiangen Biotech). They were further used for transformation of DH5α competent cells (Tiangen Biotech), and positive clones (resistant to ampicillin) were selected and sequenced. An EasyPure Plasmid MiniPrep Kit (TransGen Biotech) was used for plasmid isolation and purification.

The full sequence of Hf-AV422 cDNA was acquired by aligning the sequences of Contig2959 and the RACE products with DNAMAN v.6.0 (Lynnon Biosoft, San Ramon, CA, USA). The open reading frame (ORF) and its encoding protein sequence was determined by ORF finder (https://www.ncbi.nlm.nih.gov/orffinder/). The primary structure of the encoding protein is predicted by DiANNA v.1.1 (http://clavius.bc.edu/~clotelab/DiANNA/). Signal peptides were detected by SignalP v.4.1 (http://www.cbs.dtu.dk/services/SignalP/). Comparation of motifs between Hf-AV422 and its homologs in other ticks was accomplished by MEME v.5.1.0 (http://meme-suite.org/tools/meme).

Expression of Hf-AV422 in whole ticks and organs of different engorging levels

Ticks were sterilized with 70% ethyl alcohol, and salivary glands, ovaries and cuticles (tick remnants after removal of salivary glands, ovaries and midguts) were harvested. In each qPCR assay, 18 ticks were used (n = 18). Total RNA extraction was completed from whole ticks and organs from with the TRIzol reagent (Invitrogen, Waltham, MA, USA). The cDNA synthesis was carried out by following a kit user manual (AT341, TransGen Biotech). Primers used for amplification of Hf-AV422 were 5′-TCGTACGGCGAGACCAAC-3′ and 5′-CAGGTAGGCGGAGAAGAGC-3′. Primers used for amplification of β-actin were 5′-CAAGGAGAAGCTCTGCTACGTC-3′ and 5′-CCGATGGTGATGACCTGAC-3′. Those primers were designed by DNAMAN v.6.0 (Lynnon Biosoft, Quebec, Canada) based on the Hf-AV422 gene (GenBank acc. no. KR269741) and β-actin gene (AY519367.1). The reaction of real-time fluorescence quantitative PCR (qPCR) was done in a 20 μL system including 2 μL cDNA templates, 10 μL TB Green Premix Ex Taq II (2 ×), 0.4 μL each of forward and reverse primers (10 μM), 0.4 μL ROX Reference Dye and 6.8 μL of ddH2O. The qPCR was completed in an ABI StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) with a same program as in our previous study (Liu et al. 2020). Dissociation curves and standard curves were made at the same time for purposes of method validation. The relative expression level of Hf-AV422 to β-actin was calculated by the 2−ΔΔCt method (Livak and Schmittgen 2001). Comparison of Hf-AV422 expression in the same organ between partially and fully engorged ticks was analyzed by Student’s t test (GraphPad Prism v.8, San Diego, CA, USA). Comparison of Hf-AV422 expression in whole ticks but different engorging status (1/4, 1/2, 3/4 and fully engorged) was analyzed by one-way ANOVA (GraphPad Prism v.8) In all tests, α = 0.05.

Recombinant protein expression of AV422

The ORF of Hf-AV422 was amplified by primer pairs 5′-TTAGGATCCTCGTGCAACTGCCACCTGCGC-3′ and 5′-TTGAAGCTTGTTGCCCAGGTAGGCGGAGAA-3′ in the conditions of 94 °C for 5 min, 33 cycles of 94 °C for 30 s, 68.5 °C for 30 s, 72 °C for 45 s, and finally 72 °C for 10 min. The template was the 1st-cDNA reversely transcribed from total RNA of whole ticks. PCR products were confirmed by electrophoresis, and inserted into pGM-T vectors (Tiangen Biotech). The positive recombinant plasmid (pGM-T-HfAV422), and pET28a (Dingguo Changsheng Bitotech, Beijing, China), were submitted to restriction enzyme digestion with BamH I and Hind III (New England Biolabs, Ipswich, MA, USA). The purified ORF of AV422 was inserted into pET28a, and later used for transformation of Escherichia coli strain DH5α (Tiangen Biotech). Plasmids from positive clones were identified, sequenced, and named as pET28a-HfAV422. pET28a-HfAV422 was used for transformation of E. coli strain BL21 (DE3) (Tiangen Biotech), and cultured in LB mediums containing kanamycin (Thermo Fisher, Waltham, MA, USA) at 37 °C for 16 h. Positive clones were picked and further cultured until OD600 reaching 0.6–0.8. Then isopropyl-β-d-thiogalactoside (IPTG) was added to a final concertation of 1 mM. After incubation with IPTG for 2, 4 and 6 h, the culture medium and bacteria were collected. An aliquot of whole cell lysates (the mixture of whole bacterial culture after bacteriolysis treatment), cell pellets and supernatants were subjected to 12% SDS-PAGE analysis. The supernatants containing the solubilized proteins were purified by affinity chromatography with a Ni–NTA His Bind Resin column (7sea Biotech, Shanghai, China). After rinsing extensively with column buffer (10 mM imidazole), the recombinant protein of interest was eluted using a step gradient of imidazole in the column buffer (30, 50, 100, 250 and 500 mM imidazole, 5 mL each time). The protein of interest was further confirmed by SDS-PAGE (5% stacking gel and 12% separating gel).

Anticoagulant activity of rHf-AV422 in vitro

We followed a protocol published previously (Liu et al. 2020). Three female New Zealand rabbits, body weights 3.0 ± 0.2 kg, were bought from Hunan Laboratory Animal Center (Changsha, China). Five mL of blood was drawn from auricular veins into a tube containing 0.5 mL 3.8% Na citrate (wt/vol). Plasma was collected by centrifuging at 1000×g for 10 min. Purified rHf-AV422 solution was diluted with the plasma at a ratio of 0:1, 1:1 and 3:1 (rHf-AV422: plasma, vol:vol), i.e., the final concentration of rHf-AV422 in the mixed solution was 0, 1 and 1.5 µM, respectively. Plasma diluted with normal saline instead of rHf-AV422 was used as the blank control. Plasma diluted with bovine serum albumin (BSA; Solarbio, Beijing, China) was used as the negative control. Serine protease inhibitor (Serpin) was known with anti-coagulation activities (Imamura et al. 2005). rHf-Serpin was prepared with the same protocol for rHf-AV422 in the present study, and rHf-Serpin diluted in plasma at the same concentrations as rHf-AV422, was used as positive control. Assays of prothrombin time (PT), activated partial thromboplastin time (APTT) and thrombin time (TT) were conducted by following the user’s guides of kits in a MC-1000 coagulometer (TECO Medical Instruments, Niederbayern, Germany). Data were analyzed by Excel 2016 (Redmond, WA, USA), and presented as mean ± SD value (three biological replicates were used, n = 3). Comparisons among groups were evaluated Tukey’s test following one-way ANOVA (α = 0.05).

Results

Full length cloning and bioinformatic analysis

The 5′-RACE product was about 250 bp in length, and the deletion sequence in the 5′ end of Hf-AV422 was determined to be 204 bp (Fig. 1). We gained a PCR product at approximately 400 bp by 3′-RACE, and found that the deletion sequence in the 3′ end of Hf-AV422 was 353 bp (Fig. 2). On the basis of Contig2959, we assembled the 5′ end and 3′ end sequences to get the full-length of Hf-AV422. The Hf-AV422 full length was 1152 bp after aligning.

Fig. 1
figure 1

Amplification of the 5′ end of Contig2959 by 5′RACE. A Agarose electrophoresis for 5′RACE products. M, Marker; Lane 1, blank control; Lane 2, amplification product of 5′ RACE. B The 5′ end sequence of Hf-AV422. The overlapping sequence is underlined with the designed primer sequences in both bold and italic

Full size image
Fig. 2
figure 2

Amplification of the 3′ end of Contig2959 by 3′RACE. A Agarose electrophoresis for 3′RACE products. M, Marker; Lane 1, blank control; Lane 2, the amplification product of 3′ RACE. B The 3′ end sequence of Hf-AV422. The overlapping sequence is underlined with the designed primer sequences in both bold and italic

Full size image

The ORF Hf-AV422 was 696 bp, encoding a protein of 231 aa (Fig. 3). There was a signal peptide at 1–20 aa. The molecular weight of the mature protein was estimated to be 23.05 kDa. The sequence and motifs of AV422 from different ticks revealed that it was highly conserved among genera (Fig. 4).

Fig. 3
figure 3

The full-length of Hf-AV422 and the primary structure of the encoding protein. Putative signal peptides are underlined, and cysteine residuals are highlighted in gray

Full size image
Fig. 4
figure 4

Multiple alignments of AV422 sequence in ticks. The sequence of Hf-AV422 was compared with 10 homologs in other tick species deposited in UniProt database (https://www.uniprot.org/) by MEME v.5.1.0 (http://meme-suite.org/tools/meme). The database accession numbers of homologs used for comparison are to the left of the species names. Each colored box represents a motif in the protein, with the motif name indicated in the box at the bottom. The p-value is defined as the probability that a random sequence (with the same length and conforming to the background) would have position p-values such that the product is smaller or equal to the value calculated for the sequence under test. (Color figure online)

Full size image

Expression level variations of Hf-AV422 in whole ticks and organs of different engorging levels

The 1/4, 1/2 and 3/4 engorged ticks were summarized and analyzed together as one group compared to fully engorged ticks (Fig. 5A). The expression level of Hf-AV422 in the salivary gland and the cuticle was higher than that in the ovary regardless of engorging levels (p < 0.001) (Fig. 5A). Its expression in salivary glands and ovaries was not statistically different between engorged females and partially engorged females. However, its expression in cuticles from partially-engorged females was higher than that from fully-engorged females (p < 0.001) (Fig. 5A). The relative expression of Hf-AV422 in whole ticks decreased during engorging (p ≤ 0.007) except that there was no statistical difference between the 3/4 engorged and the fully-engorged (Fig. 5B).

Fig. 5
figure 5

Mean (+ SD; n = 3) relative expression levels of Hf-AV422 in organs and whole ticks of different engorging levels. A Hf-AV422 expressions in salivary glands (SG), ovaries (O) and cuticle (C) of partially engorged (PE) and fully engorged (FE) ticks. B Hf-AV422 expressions in whole ticks of different engorging levels. Tick body weights were recorded to designate the 1/4 engorged (1/4 E), 1/2 engorged (1/2 E), 3/4 engorged (3/4 E) and FE (detached from hosts). Means within a panel capped with different letters are significantly different (Tukey’s test: p < 0.01)

Full size image

Induced expression of Hf-AV422 in Escherichia coli

In comparison with the blank control, a more intense band could be seen at ca. 30 kDa after 2 h inducible expression (Fig. 6A). Inducible expression for 4 and 6 h repeated the observation (Fig. 6A). This demonstrated that a protein with a molecular weight at ca. 30 kDa was obviously expressed in E. coli. Theoretically, the size of the putative protein encoded by Hf-AV422 was 23.05 kDa, besides an extra approx. 5 kDa tag from His-tag would also be added. The size of the actual expression products was therefore in line with the putative size. The electrophoresis photo showed a single clear band near 30 kDa for proteins after affinity chromatography (lane 5, Fig. 6B), indicating an effective purification process.

Fig. 6
figure 6

Expression of Hf-AV422 in a prokaryote expression system. A pET28a-HfAV422 was used for transformation of Escherichia coli strain BL21 (DE3). IPTG was added to the medium, and the induced expression lasted for 2–6 h. Proteins were extracted from E. coli cells, and were subject to SDS-PAGE (5% stacking gel and 12% separating gel). M, Protein marker (10–180 kDa); 2 h N, non-induced expression (no IPTG) for 2 h; 2 h I, induced expression for 2 h; 4 h N/I, non-induced/induced expression for 4 h; 6 h N/I, non-induced/induced expression for 6 h. B Purification of rHf-AV422. The extracted proteins were further purified by affinity chromatography with Ni-NTA His Bind Resin columns, and then were analyzed by SDS-PAGE. M, protein marker; 1, whole cell lysates (mixture of whole bacterial culture after bacteriolysis treatment); 2, cell pellets (precipitates of whole cell lysates after centrifuging); 3, supernatants (of whole cell lysates after centrifuging); 4, eluents; 5, purified products

Full size image

Anticoagulation of rHf-AV422 in vitro

Results of PT assays were shown in Fig. 7A. Prothrombin time of undiluted rabbit plasma was not statistically different between groups. As expected, PT of normal saline group (blank controls) and BSA group (negative controls) were lower than that of rHf-Serpin group (positive controls) at a dilution ratio of 1:1 and 3:1 (vol:vol) (p ≤ 0.046), indicating the feasibility of the PT assays. Compared with blank controls and negative controls, PT in rHf-AV422 group was significantly longer (p ≤ 0.012), especially at a dilution ratio of 3:1. Those data indicated that rHf-AV422 had the activity of PT extension.

Fig. 7
figure 7

Effects of rHf-AV422 on mean (+ SD; n = 3) A PT, B APTT and C TT when incubated with rabbit plasma. rHf-AV422 at different concentrations was mixed with rabbit plasma to conduct dose-effect assays on PT, APTT and TT. Plasma diluted by normal saline was used as the blank control; plasma diluted by BSA was used as the negative control; plasma diluted by rHf-Serpin was used as the positive control. Asterisks indicate significant differences among treatments (Tukey’s test: *0.01 < p < 0.05; **p < 0.01). A clotting time > 300 s was beyond the measuring range of the coagulometer, and it was regarded to be not coagulated

Full size image

Figure 7B displays the results of APTT assays. Activated partial thromboplastin time of undiluted plasma was not statistically different between groups. APTT of blank controls and negative controls were shorter than that of rHf-Serpin group (p < 0.001). APTTs in rHf-AV422 group were > 300 s, exceeding the full-scale value of the coagulometer. Those data indicated that rHf-AV422 incubation obviously prolonged the APTT. The assays of TT showed similar results with those of PT. Incubation with rHf-AV422 dose-dependently prolonged TT (Fig. 7C).

Discussion

We have previously done a proteomic analysis of feces from H. flava, and four unique peptides, i.e., APCLGQTLPDQK, CLTDIQAGLEK, HAPCLGQTLPDQKK and ELIGFVAEGSQELFK, were detected (Liu et al. 2018b). By searching our salivary gland transcriptomic library in the same species, we retrieved Contig2959 encoding the polypeptide containing those four unique peptides (Xu et al. 2015). The Contig2959, 1102 bp long, was homologous to the cDNA of AV422 from A. americanum, but was lacking 5′ end and 3′ end. Hence, here we cloned the cDNA of Hf-AV422 on the basis of Congtig2959.

The ORF of Hf-AV422 had a high similarity to that of Aa-AV422 with an identity of 90.9%. However, there were some differences between the downstream and upstream sequences of their ORF. The 5′ end of Hf-AV422 was longer, but its 3′ end was shorter compared with Aa-AV422. Many protein sequences that were homologous to Hf-AV422 could be retrieved by searching Uniprot database, but their annotations or names were different. MEME analysis indicated that the primary structure of those proteins was highly conserved cross-tick species, consistent with a previous study (Mulenga et al. 2007). We thus inferred that the function of Hf-AV422 should be similar to that of Am-AV422, which included high immunogenicity and antithrombogenicity.

AV422 was initially described as a salivary protein, and was expressed in salivary glands, cuticles, midguts and ovaries (Mulenga et al. 2007). The relative expression levels of Hf-AV422 were higher in salivary glands and cuticles than in ovaries. The observation was consistent with that in A. americanum ticks (Mulenga et al. 2007). Its high expression in salivary glands supported that it might play a significant part in the blood-feeding. Mulenga et al. (2007) reported an upregulation of AV422 in whole A. americanum ticks fed for 1–7 days. However, we observed a decreased expression profile in whole ticks during engorgement. The reason for the inconsistence is unknown.

Blood clotting takes place after complex biochemical cascades. Three events, namely, the activation of FX, the synthesis of thrombin, and the formation of fibrin, occur in succession during the whole process (Davie et al. 1991). It is widely acknowledged that the activation of FX is triggered through two pathways, the intrinsic pathway (or the contact activation pathway) and the extrinsic pathway (or the tissue factor pathway) (Davie et al. 1991). The two pathways will converge into the final common pathway to generate a burst of thrombin. In vitro blood clotting tests, PT is recorded to estimate the effectiveness of the extrinsic pathway, whereas APTT is commonly served an indicator for the intrinsic pathway (Triplett 2000). TT quantifies the ability of transforming fibrinogen into fibrin, and thus represents the effects on the common blood clotting activation pathway (Triplett 2000). We demonstrated that rHf-AV422 significantly extended PT, APTT and TT in a dose dependent manner. This meant that rHf-AV422 affected both the FX activation pathways and the common pathway. The extrinsic pathway becomes activated upon vascular injuries, which always happens in hosts during insertion of tick mouth parts. Injury in vessels leads to formation of FVIIa-tissue factor, and it will further activate FX. Activation of FX will initially generate a small amount of thrombin. However, the produced thrombin provides a positive feedback by activating of FV and FVIII, whose activation will finally result in production of a large amounts of thrombin. It is likely that AV422 intervenes in one or more of those steps. A previous study showed that rAamAV422 delayed plasma clotting and inhibited platelet aggregation, but no effect was observed in PT, APPT and TT assays (Mulenga et al. 2013). Although both studies indicated an anticoagulation property of AV422, the inconsistence of the PT, APTT and TT tests remains to be explained. Perhaps the difference of expression systems for recombinant protein production and the different amino acid residue(s) in their primary structures account for this.

The anti-coagulation property of AV422 may lead to applications in the tick control. Ticks are obligate hematophagous ectoparasites, and they require nutrients from blood-meal to grow, develop and reproduce. During blood-feeding, ticks will secrete many proteins in saliva with anti-clotting activities to maintain the blood-meal in fluid state (He et al. 2018). Moreover, blood-meal will be stored at digestive tracts in a fluid form for a long time before taken in by digestive cells. Blood clotting will potentially block both the blood-meal acquisition and the digestion of blood-meal (Narasimhan et al. 2013). Hence, anticoagulation is vital for the survival of ticks (Araman 1979). It seems to be a promising approach to control ticks by abolishing the anticoagulation property of those proteins by using their antibodies against them. These bio-active proteins are possible antigen candidates for the vaccine design.

It should be noted that here we did not provide a detailed anti-clotting mechanism of AV422. Further studies are demanded to explore the specific effects of AV422 on the blood coagulation pathways. For example, whether AV422 directly inhibits FVII, FIX, FXI, FX, etc., or impacts cofactors like calcium, vitamin K, etc., or affects regulators like protein C, antithrombin, tissue factor pathway inhibitor, etc.

In conclusion, AV422 from H. flava is a conserved tick protein and exerts an anti-clotting effect by inhibition of fibrinogen-fibrin conversion via both the extrinsic pathway and the intrinsic pathway. Additional investigations are necessary to test the immunogenic and protective properties to ensure its availability as an antigen candidate for the development of broad-spectrum vaccines against ticks.