Conserved Amblyomma americanum tick Serpin19, an inhibitor of blood clotting factors Xa and XIa, trypsin and plasmin, has anti-haemostatic functions

Abstract

Tick saliva serine protease inhibitors (serpins) facilitate tick blood meal feeding through inhibition of protease mediators of host defense pathways. We previously identified a highly conserved Amblyomma americanum serpin (AAS) 19 that is characterized by its reactive center loop being 100% conserved in ixodid ticks. In this study, biochemical characterization reveals that the ubiquitously transcribed AAS19 is an anti-coagulant protein, inhibiting the activity of five of the eight serine protease blood clotting factors. Pichia pastoris-expressed recombinant (r) AAS19 inhibits the enzyme activity of trypsin, plasmin and blood clotting factors (f) Xa and XIa, with stoichiometry of inhibition estimated at 5.1, 9.4, 23.8 and 28, respectively. Similar to typical inhibitory serpins, rAAS19 forms irreversible complexes with trypsin, fXa and fXIa. At a higher molar excess of rAAS19, fXIIa is inhibited by 82.5%, and thrombin (fIIa), fIXa, chymotrypsin and tryptase are inhibited moderately by 14 – 29%. In anti-hemostatic functional assays, rAAS19 inhibits thrombin but not ADP and cathepsin G activated platelet aggregation, delays clotting in recalcification and thrombin time assays by up to 250 s, and up to 40 s in the activated partial thromboplastin time assay. Given AAS19 high cross-tick species conservation, and specific reactivity of rAAS19 with antibodies to A. americanum tick saliva proteins, we conclude that rAAS19 is a potential candidate for development of a universal tick vaccine.

Graphical abstract

1. Introduction

Ticks and tick borne diseases (TBDs) have a global veterinary and public health impact that in livestock production amounts to large monetary losses worldwide (Jongejan and Uilenberg, 2004; Grisi et al., 2014). For many years, ticks and TBDs were mostly a veterinary problem. However, following the description of Borrelia burgdorferi, the causative agent of Lyme disease, in the 1980s (Burgdorfer et al., 1982), the impact of human TBDs has increased. Recently, 17 human TBD agents were listed in a study advocating for One Health solutions (Dantas-Torres et al., 2012). Long considered a nuisance tick (Childs and Paddock, 2003), A. americanum is the principal vector of Ehrlichia chaffeensis and Ehrlichia ewingii, the causative agents of human monocytic ehrlichiosis (Anderson et al., 1993; Wolf et al., 2000). It is also associated with epidemiology of Francisella tularensis (Taylor et al., 1991); a yet to be described causative agent of the southern rash illness (STARI) (James et al., 2001); and a newly identified heartland virus (Savage et al., 2013). In veterinary health, A. americanum transmits Theileria cervi, a parasite of white tailed deer (Laird et al., 1988), and heavy infestations caused mortality in white tailed deer fawns (Yabsley et al., 2005) as well as production losses in cattle (Barnard, 1985, 1990).

Ticks feed by disrupting host tissue and sucking up blood that pools in the feeding lesion, which provokes host defense responses including pain, hemostasis (to limit blood loss), inflammation, complement activation (to protect against invading microbial organisms) and tissue repair responses (to heal the feeding lesion) (Francischetti et al., 2009). To date, many proteins derived from tick saliva have been identified which have a direct role at the feeding site, allowing the tick to successfully acquire its blood meal (Steen et al., 2006; Maritz-Olivier et al., 2007). Following the blood meal acquisition, it is also necessary that blood remains fluid for subsequent digestion, in which different tick proteins have been described to act as anti-coagulant molecules in the tick midgut (Ricci et al., 2007; Anderson et al., 2008; Liao et al., 2009). Serine proteases mediate some host defense pathways to tick feeding and are controlled in some pathways by inhibitors belonging to the serine protease inhibitors (serpins) family (Gettins, 2002; Huntington, 2006; Rau et al., 2007). From this perspective, it is proposed that ticks inject serpins into the host to mediate evasion of host defenses and thus they could be suitable targets for tick vaccines (Mulenga et al., 2001). Several tick serpin-encoding cDNAs have been cloned and characterized, including serpins of A. americanum (Mulenga et al., 2007; Porter et al., 2015), Amblyomma maculatum (Karim et al., 2011), Ixodes scapularis (Ribeiro et al., 2006; Mulenga et al., 2009), Ixodes ricinus (Leboulle et al., 2002; Prevot et al., 2006; Chmelar et al., 2011), Rhipicephalus microplus (Rodriguez-Valle et al., 2012; Tirloni et al., 2014b), Rhipicephalus appendiculatus (Mulenga et al., 2003a), Rhipicephalus haemaphysaloides (Yu et al., 2013), and Haemaphysalis longicornis (Sugino et al., 2003; Imamura et al., 2005, 2006). As of January 2015, more than 200 tick serpin-encoding cDNAs were available in public databases (Porter et al., 2015).

The concept that ticks utilize serpins to evade host defense mechanisms assumes that ticks inject inhibitory serpins into the host during feeding. Indeed, the presence of serpins in tick saliva was well demonstrated though saliva proteomic studies in Dermacentor andersoni (Mudenda et al., 2014) and R. microplus (Tirloni et al., 2014a) as well as being inferred from transcriptional analysis of salivary glands from A. americanum (Mulenga et al., 2007; Porter et al., 2015), A. maculatum (Karim et al., 2011), Amblyomma triste, Amblyomma parvum, Amblyomma cajennense (Garcia et al., 2014), Amblyomma variegatum (Ribeiro et al., 2011), Hyalomma marginatum rufipes (Francischetti et al., 2011), I. scapularis (Valenzuela et al., 2002; Ribeiro et al., 2006; Mulenga et al., 2009), I. ricinus (Leboulle et al., 2002; Schwarz et al., 2013, 2014), R. appendiculatus (Mulenga et al., 2003a, b), R. microplus (Tirloni et al., 2014b), R. haemaphysaloides (Yu et al., 2013), H. longicornis (Sugino et al., 2003; Imamura et al., 2005), and Antricola delacruzi (Ribeiro et al., 2012). Accordingly, inhibitory tick serpins have been found and characterized in tick saliva, including A. americanum salivary serpin (AAS) 6 (Mulenga et al., 2007; Chalaire et al., 2011), a cross-class inhibitor of papain and trypsin-like proteases able to inhibit blood clotting and complement activation (Mulenga et al., 2013). A blood meal-induced salivary I. scapularis serpin has been shown to act upon thrombin and platelet aggregation (Ibelli et al., 2014). In related studies, an inhibitor of pro-inflammation proteases, elastase, cathepsin G and chymase was found in I. ricinus (Prevot et al., 2006, 2009; Chmelar et al., 2011). Similarly, I. ricinus serpin Iris2 inhibited inflammation by inhibiting cathepsin G and chymase (Chmelar et al., 2011). Rhipicephalus haemaphysaloides has two serpins which are able to inhibit chymotrypsin (Yu et al., 2013). In another study, Rodriguez-Valle et al. (2012) reported a characterization of R. microplus serpin-3, an inhibitor of trypsin and thrombin that is recognized by naturally tick-infested bovine serum, and antibodies against an epitope of this protein impairs tick fertility. Indeed, tick vaccine efficacy studies showed tick-feeding efficiency is reduced when H. longicornis (Sugino et al., 2003; Imamura et al., 2005), R. appendiculatus (Imamura et al., 2006, 2008), R. microplus (Jittapalapong et al., 2010), and I. ricinus (Prevot et al., 2007) serpins are used as antigen. The goal of the present study was to characterize the role(s) of A. americanum tick serpin-19 (AAS19) in tick feeding success. This study demonstrates that AAS19 is a potential target for development of a universal tick vaccine that is effective against more than one tick species.

2. Materials and methods

2.1. Ethics statement

All animal work was conducted and approved according to the Texas A&M University, USA, Institutional Animal Care and Use Committee (AUP 2011–0207).

2.2. Tick feeding, dissections, total RNA extractions and cDNA synthesis

Amblyomma americanum ticks were purchased from the tick laboratory at Oklahoma State University (Stillwater, OK, USA). Routinely, ticks were fed on rabbits according to animal use protocols approved by the Texas A & M University Institutional Animal Care and Use Committee. Feeding was performed as previously described (Mulenga et al., 2013; Kim et al., 2014a). Amblyomma americanum ticks were restricted to feed on the outer part of the ear of New Zealand rabbits with orthopedic stockinets glued with Kamar Adhesive (Kamar Products Inc., Zionsville, IN, USA). Six male ticks were pre-fed for 3 days prior to introducing 15 female ticks in each of the ear stockinets (total of 30 female ticks per rabbit).

Ticks were collected and dissected as previously described (Mulenga et al., 2013). Five ticks were manually detached every 24 h for 5 days (24 – 120 h). Within the first hour of detachment, tick mouthparts were inspected to remove remnant tissue and washed in RNase inhibitor diethylpyrocarbonate (DEPC)-treated water to prepare for dissection. Dissected tick organs, salivary glands (SG), midgut (MG), ovary (OV), synganglion (SYN), Malpighian tubules (MT) and carcass (CA, the remnants after removal of other organs) were placed in RNA Later (Life Technologies, Carlsbad, CA, USA) or 1 mL of Trizol total RNA extraction reagent (Life Technologies) and stored at −80°C until total RNA extraction.

Total RNA was extracted using the Trizol reagent according to the manufacturer’s instructions (Life Technologies) and re-suspended in DEPC-treated water. Total RNA was quantified using a UV-VIS Spectrophotometer DU-640B (Beckman Coulter, Brea, CA, USA) or the Infinite M200 Pro plate reader (Tecan, Männedorf, Switzerland). Up to 1 μg of total RNA was used to synthesize cDNA using the Verso cDNA Synthesis Kit following the manufacturer’s instructions (Thermo Scientific, Waltham, MA, USA).

2.3. Structural alignment, amino acid motif scanning and comparative modeling

To gain an insight into the relationship of AAS19 (GenBank accession number: GAYW01000076; Porter et al., 2015) protein from the A. americanum tertiary structure compared with its homologs in other tick species, amino acid sequences from UniProt in other tick species (Rhipicephalus pulchellus (L7LRY7), I. ricinus (V5IHU3), A. maculatum (G3ML49 and G3ML50), R. microplus (V9VP22), Ix. scapularis (B7QJF1), A. triste (A0A023GPF9) and A. cajenense (A0A023FM57)) were subjected to structure-based ClustalW alignment using the AAS19 tertiary structure as a template. The alignment sequences were subsequently viewed using GeneDoc software (http://www.nrbsc.org/gfx/genedoc/ebinet.htm). Additionally, the AAS19 amino acid sequence was manually inspected for annotated glycosaminoglycan (GAG) binding motifs as previously reviewed (Hileman et al., 1998). To determine potential N- or O-linked glycosylation sites, AAS19 amino acid was scanned using NetNGlyc 1.0 and NetOGlyc 4.0 servers (www.cbs.dtu.dk).

The three-dimensional (3-D) structure of AAS19 was predicted using a comparative modeling approach. The serpin protein C inhibitor structure (2OL2) (Li and Huntington, 2008) was retrieved from the Protein Data Bank (PDB) (http://www.rcsb.org) and used as a molecular template for AAS19 modeling based on 30% and 53% sequence identity and similarity, respectively. Sequence alignments were generated using the ClustalW algorithm (Larkin et al., 2007) and used as input in the Modeller 9v14 program (Webb and Sali, 2014). Models generated were evaluated using QMEAN4 and PROCHECK to estimate model reliability and predict quality (Morris et al., 1992; Benkert et al., 2008). The electrostatic potential of AAS19 and protein C inhibitor (2OL2, positive control template) was calculated using two approaches. First, the Adaptive Poisson-Boltzmann Solver (APBS) was used and protonation states were assigned using the parameters for solvation energy (PARSE) force field for each structure by PDB2PQR (Unni et al., 2011). Execution of APBS and visualization of resulting electrostatic potentials were performed by using the Visual Molecular Dynamics (VMD) program (Humphrey et al., 1996) at ±5·kT/e of positive and negative contour fields. In the second approach, electrostatic potential was computed using APBS and visualized in the SWISS PDB viewer (Guex and Peitsch, 1997) (http://www.expasy.org/spdbv/) set to default parameters.

2.4. Temporal and spatial quantitative reverse transcription (qRT)-PCR transcription analyses of AAS19

Transcription analysis was done using a two-step quantitative (q) reverse transcription (RT)-PCR using the Applied Biosystems 7300 Real Time PCR System (Life Technologies) as described (Kim et al., 2014a). AAS19-specific qRT-PCR forward (⁵′GACAAGACGCACGGCAAAA³′) and reverse (⁵′GAAGTCCGGCGGCTCAT³′) primers were used to determine transcript abundance in triplicate pools of cDNA of dissected SG, MG and remnant tissues as CA of unfed, 24 and 48 h fed (n = 15 ticks per pool), 72 and 96 h fed (n = 10 ticks per pool) and 120 h fed ticks (n = five ticks). Cycling conditions were as follows: stage one at 50°C for 2 min, stage two at 95°C for 10 min, and stage three contained two steps with 40 cycles of 95°C for 15 s and 60°C for 1 min. Reaction volumes, in triplicate, contained 10-fold diluted cDNAs that were originally synthesized from 1 μg of total RNA, 350 nM of forward and reverse AAS19 primers, and 2X SYBR Green Master Mix (Life Technologies). For an internal reference control, forward (⁵′GGCGCCGAGGTGAAGAA³′) and reverse (⁵′CCTTGCCGTCCACCTTGAT³′) primers targeting the 40S ribosomal protein S4 (RPS4; accession number GAGD01011247.1), which is stably expressed in I. scapularis during feeding (Koci et al., 2013), was used. Relative quantification (RQ) of AAS19 transcript was determined using the comparative C_T (2^−ΔΔCt) method as previously described (Livak and Schmittgen, 2001) and adapted in Kim et al. (2014a). The data was presented as the percent mean (M) transcript abundance ± S.E.M. per tissue type.

2.5. Expression and affinity purification of recombinant AAS19

Recombinant (r) AAS19 protein was expressed using the Pichia pastoris and pPICZα plasmid expression system (Life Technologies) as described previously (Mulenga et al., 2013; Kim et al., 2014b). A mature AAS19 protein open reading frame (Porter et al., 2015) was sub-cloned into pPICZαA EcoRI and NotI restriction sites using forward (⁵′GAATTCGCAGAGCCCGACGAAGATGGC³′) and reverse (⁵′GCGGCCGCGAGGGCGTTAATTTCGCCCAG³′) primers with added restriction enzyme sites in bold. The pPICZαA-AAS19 expression plasmid was linearized with PmeI and electroporated into the Pichia pastoris X-33 strain (Life Technologies) using an ECM600 electroporator (BTX Harvard Apparatus Inc., Holliston, MA, USA) with parameters set to 1.5 kV, 25 uF and 186 Ω. Transformed colonies were selected on Yeast Extract Peptone Dextrose Medium with Sorbitol (YPDS) agar plates with zeocin (100 μg/μl), and methanol utilization on Minimal Methanol (MM) agar plates, both incubated at 28°C. Positive transformants were inoculated in buffered glycerol-complex medium (BMGY) and grown overnight at 28°C with shaking (230–250 rpm). Subsequently, the cells were used to inoculate buffered methanol-complex medium (BMMY) to A₆₀₀ of 1 after which protein expression was induced by adding methanol to 0.5% final concentration every 24 h for 5 days. rAAS19 in spent culture media was precipitated by ammonium sulfate saturation (525 g/L of media) with stirring overnight at 4°C. The precipitate was pelleted at 11,200 g for 1 h at 4°C and re-suspended in, and dialyzed against, 20 mM Tris-HCl buffer pH 7.4. To verify expression of rAAS19, western blotting analysis was performed using the horseradish peroxidase (HRP)-labeled antibody to the C-terminal hexahistidine tag (Life Technologies) diluted to 1:5000 in 5% blocking buffer (5% skim milk powder in PBS with Tween-20). The positive signal was detected using a metal-enhanced DAB chromogenic substrate kit (Thermo Scientific). Subsequently, rAAS19 was affinity purified under native conditions using Hi-Trap Chelating HP Columns (GE Healthcare Bio-Sciences, Pittsburgh, PA, USA). Affinity purified putative rAAS19 was dialyzed against 20 mM Tris-HCl buffer pH 7.4 for downstream assays. To verify purity and background contamination, affinity purified rAAS19 was resolved on a 10% SDS-PAGE gel and silver stained. Samples with the least background were selected and concentrated by either ammonium sulfate precipitation or by centrifugation using MicroSep Centrifugal Concentration Devices (Pall Corporation, Port Washington, NY, USA).

2.6. N- and O- linked deglycosylation assay

Preliminary amino acid sequence analyses predicted N- and O-linked glycosylation sites in AAS19. To determine whether rAAS19 was N-glycosylated and/or O-glycosylated, 5 μg of affinity purified rAAS19 were treated with protein deglycosylation enzyme mix according to the manufacturer’s instructions (New England Biolabs, Ipswich, MA, USA). Deglycosylation was verified by western blotting analysis using the antibody to the C-terminal hexahistidine tag (Life Technologies) and the positive signal detected using HRP chromogenic substrate (Thermo Scientific).

2.7. Determining whether native AAS19 is injected into the host during tick feeding

To determine whether native AAS19 is injected into the host during feeding, glycosylated and deglycosylated affinity-purified rAAS19 was subjected to routine western blotting analyses using antibodies to replete-fed A. americanum tick saliva proteins. The antibodies to replete-fed tick saliva proteins were produced as previously described (Mulenga et al., 2013).

2.8. Protease inhibitor (PI) profiling

Inhibitory activity of rAAS19 was tested against a panel of 16 mammalian serine proteases related to host defense pathways against tick feeding. Mammalian proteases (per reaction) tested were: bovine thrombin (43 U), pancreatic porcine elastase (21.6 nM), pancreatic bovine trypsin (24.6 nM), pancreatic bovine α-chymotrypsin (96 nM), pancreatic porcine kallikrein (33 U), human chymase (10 U), human tryptase (10 U), human plasmin (10 nM) (Sigma-Aldrich, St. Louis, MO, USA), human neutrophil cathepsin G (166 nM) (Enzo Life Sciences Inc., Farmingdale, NY, USA), human factor (f) XIa (3.68 nM), bovine fIXa (306 nM), human fXIIa (7.6 nM), human t-PA (32 nM), human u-PA (47.2 nM) (Molecular Innovations, Inc., Novi, MI, USA), fXa (5.8 nM) (New England Biolabs) and proteinase-3 (68 U) (EMD Millipore, Billerica, MA, USA). Substrates were used at 0.20 mM final concentration and purchased from Sigma-Aldrich: N α-benzoyl-DL-Arg-pNA for tryptase, N-Succinyl-Ala-Ala-Pro-Phe-pNA for chymase, cathepsin G and chymotrypsin, N-Benzoyl-Phe-Val-Arg-pNA for thrombin and trypsin, N-Succinyl-Ala-Ala-Ala-p-nitroanilide for elastase. The following substrates were purchased from Diapharma Inc. (Philadelphia, PA, USA): Bz-Ile- Glu(γ-OR)-Gly-Arg-pNA for fXa, H-D-Val-Leu-Lys-pNA for plasmin, and H-D-Pro- Phe-Arg-pNA for kallikrein, fXIa and fXIIa. The substrate CH₃SO₂-D-CHG-Gly-Arg-pNA was purchased from Molecular Innovations and used for fIXa, u-PA and t-PA. The substrate N-Methoxysuccinyl-Ala-Ala-Pro-Val-pNA was purchased from Enzo Life Sciences and used for proteinase-3.

Reagents were mixed at room temperature in triplicate. One μM of rAAS19 was pre-incubated with amounts of the enzyme listed above for 15 min at 37°C in 20 mM Tris-HCl, 150 mM NaCl, BSA 0.1%, pH 7.4 buffer. The corresponding substrate for each enzyme was added in a 100 μL final reaction volume and substrate hydrolysis was measured at A_405nm every 11 s for 30 min at 30°C using the Infinite M200 Pro plate reader (Tecan). Acquired A_405nm data were subjected to one-phase decay analysis in Prism 6 software (GraphPad Software, La Jolla, CA, USA) to determine plateau values as proxies for initial velocity of substrate hydrolysis (V_max) or residual enzyme activity. The percent enzyme activity inhibition level was determined using the formula: $100-{\mathrm{V}}_{max}({\mathrm{V}}_{\mathrm{i}})/{\mathrm{V}}_{max}({\mathrm{V}}_0)\times 100$ where V_max (V_i) = activity in presence of, and V_max (V₀) = activity in absence of rAAS19. Data are presented as mean ± S.E.M. of triplicate readings.

2.9. Stoichiometry of inhibition (SI)

We determined stoichiometry of inhibition (SI) indices against five proteases (trypsin, plasmin, fXa, fXIa and fXIIa) that were inhibited by more than 80% in the PI profiling assay described in Section 2.8. Different molar ratios (serpin:protease) of rAAS19 were pre-incubated for 1 h with a constant concentration of trypsin (10 nM), fXa (5 nM), fXIa (5 nM) and plasmin (10 nM). The residual enzyme activity was measured using colorimetric substrates specific for each enzyme as described in Section 2.8. The data were plotted as the residual activity (V_i/V₀) versus the inhibitor to enzyme molar ratio. SI, or the molar ratio of rAAS19 to protease when enzyme activity is completely inhibited, was determined by fitting data onto the linear regression line in PRISM.

2.10. rAAS19 and protease complex formation

In varying molar ratios, affinity purified rAAS19 was incubated with trypsin, plasmin, fXa and fXIa, in Tris-HCl reaction buffer (20 mM Tris-HCl, 150 mM, NaCl, pH 7.4) for 1 h at 37°C. Denaturing sample buffer was added to the reaction mix and incubated at 99.9°C for 5 min in a thermocycler. Samples were subjected to SDS-PAGE electrophoresis on a 12.5% acrylamide gel and stained with Coomassie brilliant blue using routine protocols.

2.11. Anti-platelet aggregation function of rAAS19

Anti-platelet aggregation function(s) of rAAS19 was determined using platelet rich plasma (PRP) isolated from citrated (acid citrate dextrose) whole bovine blood (WBBL) as previously described (Horn et al., 2000; Berger et al., 2010). To prepare PRP, freshly citrated WBBL was centrifuged at 200 g for 20 min at 18°C. Subsequently, the PRP (top layer) was transferred into a new tube and centrifuged at 800 g for 20 min at 18°C. The pellet containing platelets was washed and diluted with Tyrode buffer, pH 7.4 (137 mM NaCl, 2.7 mM KCl, 12 mM NaHCO₃, 0.42 mM Na₂HPO₄, 1 mM MgCl₂, 0.1% glucose, 0.25% BSA), until A₆₅₀ = 0.15. To determine anti-platelet aggregation function, various amounts of rAAS19 (1, 0.5, 0.25 and 0 μM) were pre-incubated for 15 min at 37°C with agonist 10 NIH-U thrombin, 20 μM ADP or 0.7 μM cathepsin G in a 50 μL reaction. Adding 100 μL of pre-warmed PRP triggered platelet aggregation. Vice versa, PRP was pre-incubated with various amounts of rAAS19 prior to addition of the agonist. Platelet aggregation was monitored every 20 s for 30 min at A_650nm using the Infinite M200 Pro plate reader (Tecan). In this assay, a higher O.D. was observed in our blank control (platelet only), and increased platelet aggregation was correlated with a reduction in O.D. To determine the percentage of platelet aggregation inhibition, O.D. data was fitted to the one-phase decay equation in PRISM (GraphPad) to determine the plateau O.D. The percent reduction in platelet aggregation was calculated using this $\mathrm{Y}=100-(\mathrm{\Delta }\text{Pn})/\mathrm{\Delta }\text{PC})\ast 100$, where Y = % reduction in platelet aggregation, ΔPn and ΔPC = a respective rAAS19-treated and positive control plateau O.D. subtracted from the blank plateau O.D. Data are presented as mean ± S.E.M. of duplicate platelet aggregation assays.

2.12. Anti-blood clotting function of rAAS19

The effect of rAAS19 against blood clotting was assessed using modified recalcification time (RCT), activated partial thromboplastin time (aPTT), thrombin time (TT) and prothrombin time (PT) to, respectively, measure the effect on the entire blood cascade, the intrinsic, extrinsic and common blood clotting activation pathways, as described (Mulenga et al., 2013). All assays were done in duplicate, with clotting time monitored at A_650nm using the Infinite M200Pro plate reader (Tecan) set to 37°C. In these assays, clot formation was directly proportional to an increase in O.D., and results presented as mean ± S.E.M.

In the RCT assay, 50 μL of universal coagulation reference human plasma (UCRP) (Thermo Scientific) were pre-incubated with 10, 5, 1.25, 0.625, 0 μM rAAS19 in 90 μL of Tris-HCl reaction buffer (20 mM Tris-HCl, 150 mM NaCl, pH 7.4) for 15 min at 37°C. Adding 10 μL of pre-warmed 150 mM calcium chloride (CaCl₂) triggered plasma clotting. Plasma clotting was monitored at every 20 s for 30 min.

In the aPTT assay, various amounts of rAAS19 (indicated above) were pre-incubated for 15 min at 37°C with 50 μL of UCRP diluted to 100 μL with 20 mM Tris-HCl, 150 mM NaCl, pH 7.4. Subsequently, 50 μL of the aPTT reagent (Thermo Scientific) was added to plasma and incubated at 37°C for an additional 5 min to activate the reaction. Addition of 50 μL of 25 mM CaCl₂ to the reaction triggered clotting, and clot formation was monitored every 10 s for 5 min.

In the TT assay, various amounts of rAAS19 (indicated above) were pre-incubated at 37°C for 15 min with 25 μL of the TT reagent (Thermo Scientific) containing CaCl₂ in 50 μL of Tris-HCl reaction buffer with 20 mM Tris-HCl, 150 mM NaCl, pH 7.4. Adding 50 μL of pre-warmed UCRP started plasma clotting and clotting time was monitored every 10 s for 5 min.

In the PT assay various rAAS19 amounts as above were pre-incubated at 37°C for 15 min with 100 μL of PT reagent (Thermo Scientific) containing CaCl₂ diluted up to 150 μL with 20 mM Tris-HCl, 150 mM NaCl, pH 7.4 reaction buffer. Adding 50 μL of the pre-warmed UCRP (Thermo Scientific) triggered plasma clotting and clotting time was monitored every 10 s for 5 min.

2.13. Statistical analysis

Statistical software packages in PRISM version 6 (GraphPad Software Inc.) were used.

3. Results

3.1. AAS19 tertiary structure retains features of a typical serpin

AAS19 amino acid sequence is homologous to serpins from other tick species, being that its functional domain reactive center loop (RCL) is 100% conserved among the majority of ixodid tick species (Porter et al., 2015). In this study we conducted comparative amino acid motif and secondary structure analyses to determine the relationship of AAS19 to its homologues in other tick species (Fig. 1). Comparative modeling using protein C inhibitor tertiary structure (2OL2) as a template showed that AAS19 predicted tertiary structure retains the typical serpin fold formed by nine α-helices, three β sheets (five strands each in βsA and βsB, and three strands in βsC) (Figs. 1, 2) (Gettins, 2002). Amino acid motif scanning analyses showed conservation of the “RGD” motif (positions 40 – 42, Fig. 1), which is the binding site to integrin GPIIb-IIIa (Srinivasan et al., 2010; Nurden, 2014). Additionally, two N-glycosylation sites at positions 97 – 99 and 198 – 200 in AAS19 and in the majority of its homologues are conserved as noted with “#” symbol below the sequence (Fig. 1). Visual inspection of the alignment in Fig. 1 revealed seven clusters of basic amino acid residues that show similarity to annotated GAG binding sites (BS) (reviewed in Hileman et al., 1998). The seven clusters of basic residues (basic residues in bold, amino acid positions based on structural alignment in Fig. 1) in AAS19 are:

Fig. 1.

Structure-based sequence alignment of Amblyomma americanum serpin-19 (AAS19; GenBank accession number: GAYW01000076) protein sequence with its homologs in other tick species from the Uniprot database: Rhipicephalus pulchellus (L7LRY7), Ixodes ricinus (V5IHU3), Amblyomma maculatum (G3ML49 and G3ML50), Rhipicephalus microplus (V9VP22), Ixodes scapularis (B7QJF1), Amblyomma triste (A0A023GPF9) and Amblyomma cajenense (A0A023FM57). Secondary structures, assigned based on AAS19 comparative tertiary structure, are labeled “H” for α-helix and “E” for β-strand. Helices are labeled from “hA” to “hI” and β-strands that constitute β-sheets A – C are labeled “s1 – 6A”, “s1 – 5B” and “s1 – 4C”, respectively. The “RGD” motif is noted by the “&” symbol and the two putative N-glycosylation sites are noted by the “#” symbols below the sequence. The residues that correspond to the basic patches are indicated by an asterisk (*). The high and low conserved residues are highlighted in black and gray, respectively. The 100% conserved reactive center loop (RCL) region is denoted with black broken lines (---) above the sequence.

Fig. 2.

Structure and putative glycosaminoglycan-binding sites of Amblyomma americanum serpin-19 (AAS19). The AAS19 model was constructed using the coordinates generated with the Modeler 9v14 program and protein C inhibitor (PCI) as template (PDB 2OL2). Calculation of the electrostatic potential surface map was generated using the Adaptive Poisson-Boltzmann Solver (APBS) tool in the Visual Molecular Dynamics (VMD) program. (A) Electrostatic surface potential (indicated by a dark surface for positively, and light surface for negatively, charged regions) and ribbon diagram of the native PCI (template) showing the glycosaminoglycan (GAG)-binding along the helix-H. Residues responsible for GAG-binding activity are shown as sticks and labeled (Li and Huntington, 2008). (B) Electrostatic surface potential and ribbon diagram of the AAS19 model showing putative GAG-binding sites (GAGBS 1 – 4). Clusters of basic amino acid residues potentially involved in formation of GAGBS 1 – 4 [K¹³⁷-R¹³⁹-R¹⁴⁰-K¹⁶¹-K²⁶⁴-K²⁶⁸-K³⁰⁸], [K⁸-K²³⁶-R²⁴⁶-R²⁵⁰], [R¹⁷²-K¹⁷⁴-R¹⁷⁶-K¹⁸⁵-K¹⁹²- K¹⁹⁴-K³⁴³] and [K¹¹⁷-K¹¹⁸-R¹⁵¹-K²⁷²-K²⁹⁸] are shown as sticks and labeled.

• [⁵⁹KVLALFREQLDASR⁷²],

• [¹¹⁷KKGEEAVEKINNWVSDKTHGKIRR¹⁴⁰],

• [¹⁵¹RLILLNAVYYKGTWLYEFNKARTKPR¹⁷⁶],

• [¹⁸⁵KVLVPMMKMK¹⁹⁴],

• [²²⁹RNGLEHLKSVLTTQTLNRAISRMYPKDMKFRMPKLKLDTKYTLK²⁷²],

• [²⁸¹KKKIFSADADLSGISGAKNLYVSDVLHK³⁰¹] and

• [³⁴³KVYVDHPFIFLIR³⁵⁵].

To gain further insight regarding the possibility that clusters of basic amino acid residues in AAS19 form basic patches, we calculated surface electrostatic potentials on the AAS19 model (Fig. 2B). This analysis predicts that [K¹³⁷-R¹³⁹-R¹⁴⁰-K¹⁶¹-K²⁶⁴-K²⁶⁸-K³⁰⁸], [K⁸-K²³⁶-R²⁴⁶-R²⁵⁰], [R¹⁷²-K¹⁷⁴-R¹⁷⁶-K¹⁸⁵-K¹⁹²-K¹⁹⁴-K³⁴³], and [K¹¹⁷-K¹¹⁸-R¹⁵¹-K²⁷²-K²⁹⁸], respectively, interact to form putative GAG binding sites (GAGBS) 1 – 4 on the AAS19 surface (Fig. 2B). This methodology was able to recover the GAGBS in PCI, a protein whose GAGBS was previously described (Fig. 2A) (Li and Huntington, 2008).

3.2. AAS19 mRNA is expressed in several tick tissues during feeding

To determine the relationship of AAS19 transcription with the A. americanum tick-feeding phase, we determined its spatial and temporal transcription profile. Fig. 3 shows AAS19 mRNA is expressed during blood meal and in unfed ticks. AAS19 is transcribed in SG (Fig. 3A), MG (Fig. 3B) and other tissues (eg. CA) (Fig. 3C). AAS19 transcription in the SG does not change from unfed to 120 fed ticks. Contrarily in MG and CA,, AAS19 transcription increases (RPS4 was used for normalization) with feeding (Fig. 3B, C).

Fig. 3.

Temporal and spatial expression analysis of Amblyomma americanum serpin-19 (AAS19) mRNA through 120 h post-attachment. AAS19 mRNA relative expression analysis in unfed (UF), 24, 48, 72, 96, and 120 h fed tick dissected tissues: (A) salivary gland, (B) midgut and (C) carcass analyzed by quantitative reverse transcription-PCR. To determine the AAS19 mRNA relative expression (y-axis), data were analyzed using the 2^−ΔΔCt method comparing with ribosomal protein S4 (RPS4) in three biological replicates of different tissues in relation to time of feeding (x-axis). The lowest expressed time point was used as a calibrator for each tissue.

3.3. AAS19 protein is injected into the host during feeding

Fig. 4A and B summarize expression and affinity purification of rAAS19 in P. pastoris. To verify expression, daily samples of 1 mL of yeast spent media were concentrated by ammonium sulfate saturation and subjected to western blotting analysis using a specific antibody to the C-terminal histidine tag (as summarized in Fig. 4A). For large-scale expression, ammonium sulfate-precipitated rAAS19 was affinity purified under native conditions and purity checked by SDS-PAGE with silver staining (Fig. 4B). Fractions W2, E1, E2 and E4 in Fig. 4B, the fractions that showed the smallest amount of contaminating protein, were pooled, concentrated and dialyzed in appropriate buffer by spin columns and used for further assays. Fraction E3 was used in western blotting analysis assays summarized in Fig. 4C – E. As shown in Fig. 4C, a 1:50 dilution of rabbit pre-immune serum did not recognize rAAS19, while a 1:250 dilution of antibodies against saliva proteins of fully engorged A. americanum (Fig. 4B), and against the C-terminal histidine tag (Fig. 4E) recognize both glycosylated (lane 1) and de-glycosylated (lane 2) rAAS19. In Fig. 4D and E, there is smearing in lane 1 but not in lane 2. Additionally, there is a downward molecular weight shift in lane 2. This shift in molecular weight and disappearance of smears following de-glycosylation validates that rAAS19 is glycosylated.

Fig. 4.

Expression and affinity purification of recombinant (r) Amblyomma americanum serpin-19 (AAS19) in Pichia pastoris. Construction of expression plasmid, induction and validation of expression levels using the antibody to the C-terminal hexahistidine tag and affinity purification of rAAS19 were performed as described in Section 2.5. (A) Daily expression levels of rAAS19 through 5 days. (B) Affinity purified rAAS19 resolved on a 10% SDS-PAGE and silver staining (L, molecular weight ladder; T, total protein loaded onto column; R, column run through; B, binding buffer wash; W1 and W2, wash buffers; E1 – E4, eluted rAAS19 fractions). Purified rAAS19 fractions used for assays in Figs. 5 – 9 are noted with asterisks (*). Western blotting analysis with (C) 1:50 dilution pre-immune sera, (D) 1:250 dilution immune sera to replete-fed A. americanum tick saliva proteins, and (E) antibody to C-terminal hexahistidine tag. Lanes L, 1 and 2 in C, D and E show a molecular weight ladder, glycosylated and de-glycosylated rAAS19.

3.4. rAAS19 is a broad spectrum inhibitor of trypsin-like proteases

Protease inhibition profiling was done against 16 mammalian proteases that regulate host defenses against tick feeding such as hemostasis, wound healing/tissue remodeling and inflammatory response (Table 1). Molar excess of rAAS19 inhibited the activity of trypsin and trypsin-like proteases associated with hemostasis (Fig. 5). Pre-incubation of 1 μM rAAS19 inhibited, respectively, approximately 80, 82, 86, 95 and 98% the activity of plasmin (34.8 nM), fXIIa (7.62 nM), fXa (5.81 nM), fXIa (3.68 nM) and trypsin (24.6 nM) (Fig. 5A, B). Additionally, rAAS19 inhibits thrombin (43 U), tryptase (10 U), fIXa (30.6 nM) and chymotrypsin (96 nM) by approximately 13, 16, 20 and 28%, respectively.

Table 1.

Proteases used in recombinant Amblyomma americanum serpin-19 (rAAS19) protease inhibitor profiling.

Proteases	Biological function	References
Thrombin	Common, intrinsic and extrinsic coagulation pathways; fibrinolysis pathway; platelet activation and aggregation pathways via activation of protease-activated receptors (PARs); wound healing/tissue remodeling; inflammation.	Reimers et al., 1976; Gartner et al., 1978; Davie et al., 1991; Hoffman, 2003; Olszewska-Pazdrak et al., 2010
Trypsin	Digestion; wound healing/tissue remodeling; inflammation and pain via activation of PAR2.	Walsh et al., 1964; White et al., 2013; Cattaruzza et al., 2014
Cathepsin G	Platelet activation; wound healing/tissue remodeling; inflammation; antimicrobial properties.	Renesto and Chignard, 1993; Mak et al., 2003; Korkmaz et al., 2008
Kallikrein	Wound healing/tissue remodeling; inflammation; intrinsic coagulation pathway.	Li et al., 2007; Chao et al., 2010
Elastase	Wound healing/tissue remodeling; inflammation; digestion of elastin; activation of platelets and lymphocytes; antimicrobial properties.	Renesto and Chignard, 1993; Belaaouaj, 2002; Korkmaz et al., 2008; Kessenbrock et al., 2008, 2011
Plasmin	Fibrinolysis pathway; inflammation.	Collen, 1999; Li et al., 2012; Carmo et al., 2014; Mancek-Keber, 2014
Factor IXa	Intrinsic coagulation pathway; platelet amyloid precursor protein pathway	Davie et al., 1991; Hoffman, 2003
Factor Xa	Common, intrinsic and extrinsic coagulation pathways.	Davie et al., 1991; Hoffman, 2003
Factor XIa	Intrinsic coagulation pathway; platelet amyloid precursor protein pathway	Davie et al., 1991; Hoffman, 2003
Factor XIIa	Intrinsic coagulation pathway.	Davie et al., 1991; Hoffman, 2003
Tissue-type plasminogen activator (t-PA)	Fibrinolysis pathway; platelet amyloid precursor protein pathway.	Angles-Cano, 1994; Angles-Cano et al., 1994; Collen, 1999
Urokinase plasminogen activator (u-PA)	Fibrinolysis pathway; platelet amyloid precursor protein pathway.	Angles-Cano, 1994; Angles-Cano et al., 1994; Collen, 1999
α-chymotrypsin	Wound healing/tissue remodeling; inflammation; digestion; regulation of proteases: chymotrypsin, cathepsin G, mast cell chymase.	Algermissen et al., 1999a; Matsunaga et al., 1994, 2000; Korkmaz et al., 2008
Chymase	Wound healing/tissue remodeling; iInflammation.	Algermissen et al., 1999a, b; Caughey, 2007
Tryptase	Wound healing/tissue remodeling; inflammation.	Algermissen et al., 1999a; Hallgren and Pejler, 2006; Caughey, 2007
Proteinase-3	Wound healing/tissue remodeling; inflammation; platelet activation. Antimicrobial properties.	Renesto et al., 1994; Algermissen et al., 1999a; Kessenbrock et al., 2008; Korkmaz et al., 2008;

Fig. 5.

Protease inhibitor function profiling. Enzymes, at concentrations indicated in Section 2.8, were pre-incubated with recombinant Amblyomma americanum serpin-19 (rAAS19) (1 μM) at 37°C for 15 min. Subsequently, specific colorimetric substrates were added and hydrolysis monitored at A_450nm over 30min at 30°C. The percent enzyme activity inhibition level was determined using the formula: $100-{\mathrm{V}}_{max}({\mathrm{V}}_{\mathrm{i}})/{\mathrm{V}}_{max}({\mathrm{V}}_0)\times 100$ where V_max (V_i) = activity in presence of, and V_max (V₀) = activity in absence of, rAAS19. Data are presented as mean ± S.E.M. of triplicate readings.

To evaluate rAAS19 inhibitory efficiency, we determined the SI of proteases that were inhibited by more than 80% by rAAS19. Fig. 6A, C, E and G show the kinetics of the hydrolysis of specific substrates by these enzymes in the absence (a) and in the presence (b – g) of rAAS19 at various concentrations. These data were fitted to a linear regression, as shown in Fig. 6B, D, F and H. The rAAS19 SI index was estimated from the X-axis intercept (Y = 0) for trypsin, plasmin, fXa and fXIa at 5, 9, 23 and 28, respectively. The SI index for fXIIa was estimated at 103 (not shown). The rAAS19 mechanism of action as a typical inhibitory serpin was confirmed by the inability of heating and SDS (Huntington et al., 2000) to dissociate it from trypsin (Fig. 7A), fXa (Fig. 7B) and fXIa (Fig. 7C). Irreversible complex between rAAS19 and the target enzyme was observed only at inhibitor enzyme molar ratios (Fig. 7A lanes 1 – 3, Fig. 7B lanes 1 – 2 and Fig. 7C lanes 1 – 6) where enzyme activity inhibition was observed (Fig. 6A, E, G).

Fig. 6.

Stoichiometry inhibition (SI) assay for Amblyomma americanum serpin-19 (AAS19). Residual enzyme activity without (a) and with presence of increasing (b – g) amounts of rAAS19 were pre-incubated for 1 h at 37°C with a constant concentration of trypsin (A,B: 10 nM), plasmin (C,D: 10 nM), factor Xa (E,F: 5 nM), and factor XIa (G,H: 5 nM), resulting in rAAS19 inhibitor : enzyme protease (I:E) molar ratios varying from 0 – 10 (for plasmin and trypsin) or 0 – 20 (for factor Xa and factor XIa). Residual enzymatic activity was measured using specific colorimetric substrate for each enzyme noted in Section 2.9. The data were plotted as enzymatic residual activity (V_i/V₀) versus molar ratio (rAAS19 : enzyme protease). The SI was determined by fitting data onto a linear regression line.

Fig. 7.

SDS-stable complex formation assay. Increasing amounts of recombinant Amblyomma americanum serpin-19 (rAAS19) were pre-incubated for 1 h at 37°C with a constant concentration of trypsin (A), factor Xa (B) and factor XIa (C), resulting in molar ratios varying from 0.625:1 to 10:1 (rAAS19: enzyme protease). Samples were resolved on 12.5% SDS-PAGE and Coomassie blue-stained to identify SDS-stable complexes. The rAAS19 : enzyme protease complex formation is denoted with an asterisk.

3.5. rAAS19 inhibits thrombin-triggered platelet aggregation

PRP was used in platelet aggregation assays using three agonists: thrombin (0.03 U/μL), ADP (20 μM) and cathepsin G (0.7 μM). rAAS19 inhibited thrombin activated platelet aggregation (Fig. 8). In data shown in Fig. 8A, we did not observe platelet aggregation without addition of the agonist (a). However with platelets activated by thrombin that was pre-incubated with 1 (b), 0.5 (c), 0.25 (d) and 0 μM (e), rAAS19 showed variable levels of platelet aggregation. Pre-incubating 3U of thrombin with 1 μM and 0.5 μM rAAS19, platelet aggregation was reduced by 56% and 27%, respectively, while at 0.25 μM rAAS19 did not have any effect (Fig. 8B). The rAAS19 did not have any effects on ADP and cathepsin G activated platelet aggregation. On the other hand, pre-incubation of rAAS19 with platelets prior to addition of the agonist did not affect platelet aggregation.

Fig. 8.

Effect of recombinant Amblyomma americanum serpin-19 (rAAS19) on thrombin activated platelet aggregation. (A) The platelet aggregation function assay was done using the platelet rich plasma (PRP) approach described in Section 2.11. Tyrode buffer without thrombin (a), various amounts of rAAS19 with 1 μM (b), 0.5 μM (c), and 0.25 μM (d) rAAS19 and thrombin only (e) were pre-incubated with thrombin (3 U) in a 50 μL reaction for 15 min at 37°C. Platelet aggregation was initiated with the addition of 100 μL of pre-warmed PRP and monitored at intervals of 20 s over 30 min at A_650nm. (B) Percent reduction of thrombin-induced platelet aggregation inhibition by rAAS19. In the assay used here, platelet aggregation was directly proportional to reduced O.D. (A_650nm). Data are presented as mean ± S.E.M. of duplicate platelet aggregation assays.

3.6. rAAS19 delays plasma clotting

Plasma clotting time (CT), RCT, activated aPTT and TT were delayed by rAAS19 in a dose-response manner (Fig. 9). However, rAAS19 did not affect PT. In the RCT assay, to measure the effect of on the entire blood clotting activation system, 0.625, 1.25 and ⁵ μM (arrowhead R-CT2), and 10 μM (arrowhead R-CT3) rAAS19 dose responsively delayed RCT-CT by ~100 and 250 s compared with CT in the absence of rAAS19 (arrowhead R-CT1, 0 μM) (Fig. 9A). In the aPTT assay to measure the effect on the intrinsic blood clotting activation pathway (Fig. 9B), 0.625 and 1.25 μM rAAS19 did not have any effect (arrowhead A-CT1), while 5 and 10 μM rAAS19 delayed aPTT-CT by ~20 (A-CT2) and 40 (A-CT3) s, respectively. In the TT assay (Fig. 9C) 0.625 μM rAAS19 was not different from the control (arrowhead T-CT1), while 1.25 and 5 μM rAAS19 (arrowhead T-CT2), and 10 μM rAAS19 (arrowhead T-CT3) delayed clotting by ~180 and 250 s, respectively.

Fig. 9.

The effect of recombinant Amblyomma americanum serpin-19 (rAAS19) on plasma clotting time in the recalcification time (RCT), activated partial thromboplastin time (aPTT) and thrombin time (TT) assays. (A) Universal coagulation reference human plasma (UCRP) (50 μL) was incubated with 0 (arrowhead R-CT1), 0.625, 1.25 and 5 (arrowhead R-CT2), and 10 (arrowhead R-CT3) μM rAAS19 in 90 μL of Tris-HCl reaction buffer for 15 min at 37°C followed by the addition of 150 mM CaCl₂ (10 μL) thereafter clotting was measured every 20 s for 30 min. (B) UCRP (50 μL) was incubated with 0, 0.625, 1.25 (arrowhead A-CT1), 5 (arrowhead A-CT2), and 10 (arrowhead A-CT3) μM rAAS19 up to 100 μL of Tris-HCl reaction buffer for 15 min at 37°C, followed by the addition of aPTT reagent and incubation for 5 min at 37°C before the addition of 25 mM CaCl₂ (10 μL); clotting time was monitored every 10 s for 5 min. (C) rAAS19 with 0, 0.625 (arrowhead T-CT1), 1.25 and 5 (arrowhead T-CT2), and 10 (arrowhead T-CT3) μM up to 25 μL of Tris-HCl reaction buffer were incubated with 25 μL of the TT reagent containing CaCl₂ for 15 min at 37°C before addition of 50 μL of pre-warmed UCRP with clotting time monitored every 10 s for 5 min. All assays were done in duplicate, with clotting time monitored at A_650nm.

4. Discussion

This study was prompted by the characteristic high conservation of AAS19 in all ixodid tick species, for which data are available (Porter et al., 2015). We were particularly interested in AAS19 due to its functional domain RCL being 100% conserved in other ixodid ticks, suggesting this serpin has a role in regulating proteolytic pathways crucial to all ixodid tick species. The observation that AAS19 motifs and secondary structure are conserved in AAS19 homologues in several tick species further suggests this protein has the potential to regulate processes that are important to ixodid ticks.

In proposing that ticks use serpins to evade host defense mechanisms, the assumption is that ticks inject serpins into the host during feeding. Although several serpin cDNAs have been cloned, only a few reports confirm that saliva actually contains serpins (Mulenga et al., 2013; Ibelli et al., 2014; Mudenda et al., 2014; Tirloni et al., 2014a). Thus, observations in this study demonstrating native AAS19 is injected into the host during tick feeding and its activity inhibiting proteases participating in host defenses reinforces the idea that inhibitory serpins have a role in the tick-host relationship. Based on the rAAS19 inhibitory activity against trypsin and trypsin-like proteases including plasmin and blood clotting factors Xa and XIa, we conclude this serpin inhibits proteases involved in host defense mechanisms against tick feeding. Albeit at a low rate, it is notable that rAAS19 also inhibits other blood clotting enzymes including fXIIa, fIIa (thrombin) and fIXa, as well as chymotrypsin and tryptase, further confirming that AAS19 has a broad spectrum of inhibitory functions. Although AAS19 inhibits these enzymes, we are cautious about these enzymes being physiological targets for AAS19 because the SI indices estimated in this study were not near 1:1, as would be expected for an inhibition by serpins (Irving et al., 2000; Silverman et al., 2001). However, we note here that the observed high SI indices are not unique for rAAS19. For instance, a SI index of 10 was estimated for Drosophila melanogaster Spn1, which inhibits trypsin activity by 96% (Fullaondo et al., 2011). A high SI index could also be explained by the possibility that AAS19 requires a co-factor to enhance inhibition. Indeed, nearly all serpin acting upon the blood clotting activation cascade and fibrinolysis (antithrombin III, heparin-cofactor II, plasminogen activator inhibitor 1, protein C inhibitor and protease nexin inhibitor) need to bind GAGs in order to accelerate the inhibition rate, which can be up to 10,000-fold higher compared with the unbound native serpin (Rein et al., 2011). Accordingly, Z-protein-dependent inhibitor (ZPI) has been shown to accelerate the inhibitory effect of ZPI upon fXa by 1000-fold (Broze, Jr., 2001; Munoz and Linhardt, 2004). Our demonstration that AAS19 has four putative GAGBSs suggests its activity could be modulated by heparin or another GAG. Studies to investigate the effects of heparin on rAAS19 function are underway.

From the perspective of tick feeding physiology, the AAS19 inhibition activity upon trypsin, fXa and fXIa could be rationally explained. Trypsin, mostly known for its role as a digestive enzyme (Rawlings and Barrett, 1994), is also associated with inflammation in the skin (Cattaruzza et al., 2014; Meyer-Hoffert et al., 2004), while blood clotting factors Xa and XIa are critical proteases of the blood clotting activation cascade (Monroe and Hoffman, 2006; Roberts et al., 2006). On the other hand, AAS19 inhibition of plasmin can, at a glance, be viewed as contradictory since plasmin is mostly known for its role in digesting blood clots (fibrinolysis) (Angles-Cano, 1994; Angles-Cano et al., 1994), an activity that seems beneficial to tick feeding. Similarly, plasmin degrades and inactivates blood-clotting factors V, VIII, IX and X in vitro, suggesting plasmin has anti-coagulant functions (Hoover-Plow, 2010) and is thus beneficial to tick feeding. On the other hand, plasmin has been reported to participate in several processes such as pro-inflammatory cytokine release (Syrovets et al., 2001), inducing monocyte and dendritic cell chemotaxis (Li et al., 2010), modifying IL-8 and producing a potent attractant of neutrophils (Mortier et al., 2011), tissue remodeling and wound healing (Shen et al., 2012), all of which can negatively impact tick-feeding success. From this perspective, AAS19 inhibition of plasmin seems to contribute to feeding on blood.

Given that AAS19 is injected into the host during tick feeding and due to its inhibitory activity on blood clotting factors, we investigated its action on platelet aggregation and plasma clotting. Platelet aggregation is important to stop bleeding in injured small blood vessels, as observed in tick feeding (Francischetti et al., 2009). Platelet aggregation is activated by multiple agonists including thrombin, collagen, ADP and cathepsin G (Selak et al., 1988; Ohlmann et al., 2000; Brass, 2003). Since thrombin is considered the most efficient platelet aggregation agonist (Furman et al., 1998; Davi and Patrono, 2007), the observation that rAAS19 reduced thrombin-activated platelet aggregation by up to 56 ± 14% suggests that AAS19 may play a crucial role in tick modulation of platelet function. Thrombin activates platelet aggregation through activation of Protease Activated Receptors (PARs) 1, 3 and 4 (Kahn et al., 1999; Asokananthan et al., 2002), and thus it is plausible that native AAS19 may contribute to binding and inhibiting thrombin at the tick feeding site, preventing the formation of the platelet plug and thus impairing the integrity of the blood clot. The observation that pre-incubating rAAS19 with platelets for 15 min prior to addition of the agonist had minimal effect suggests that the observed rAAS19 anti-platelet aggregation effects were mediated through inhibition of thrombin activity. However, it is interesting to note that the “RGD” motif, which is known to interfere with platelet aggregation binding the integrin GPIIb-IIIa on activated platelets (Varon et al., 1993; Katada et al., 1997) is conserved in AAS19 and its homologues. If this motif was functional, pre-incubation with platelets should interfere with platelet aggregation function.

Platelet aggregation is followed by fibrin clot formation, which reinforces the platelet plug (Clemetson, 2012). Blood clotting can be activated via the extrinsic pathway initiated upon release of tissue factor from injured tissue or via the intrinsic pathway when factors XII, XI, IX and VIII are converted into their active forms. These pathways converge into the common pathway when fXa activates prothrombin to thrombin, which in turn catalyzes formation of the fibrin clot (Hoffman and Monroe, III, 2001; Hoffman, 2003; Monroe and Hoffman, 2006). The effects of anti-blood clotting agents are routinely addressed by using RCT assays, while effects on the extrinsic, intrinsic and common blood clotting activation pathways are examined using PT, aPTT, and TT assays (Davie et al., 1991), as were done in this study. Eight of the 11 blood clotting factors are serine proteases (Walsh and Ahmad, 2002). Thus, although empirical evidence is needed, it is conceivable that the observed rAAS19 anti-coagulant function could be attributed to its inhibitory activity against five (fIIa, fIXa, fXa, fXIa and fXIIa) of the eight serine protease mediators of the blood clotting activation cascade. Except for fVIII, the other four blood clotting factors (IXa, Xa, XIa, and XIIa) in the intrinsic pathway (aPTT assay) are serine proteases, all of which were inhibited by rAAS19, and this could explain the observation of up to 40 s plasma clotting time delay in the aPTT assay. Similarly, in the TT (common pathway) assay, the observed 250 s delay in plasma clotting time could be explained by rAAS19 inhibitory activity against thrombin. The discrepancy of 250 s delay in the TT assay and ~14% inhibition of thrombin activity in substrate hydrolysis is interesting and raises questions for further investigation. There is a possibility that the observed significant clotting time delay in the TT assay could be explained by a possibility of a yet unknown factor in host serum that potentiates AAS19 anti-thrombin activity. Given that AAS19 has four putative heparin binding sites, it is conceivable that GAGs in serum binding to AAS19 could potentiate its activity against thrombin. Experiments to resolve this question are underway.

We would like to note here that arthropods do have hemolymph-clotting cascades that are also controlled by serpins (Agarwala et al., 1996; Iwanaga et al., 1998; Kanost, 1999). Thus, there is a possibility that native AAS19 could be a well-conserved regulator of the tick’s hemolymph clotting in itself and the observed anti-coagulant function in this study could be an artifact. However, the observation in this study that native AAS19 was injected into the host during tick feeding strongly suggests the role of this protein at the tick-feeding site. Alternatively, AAS19 may function both in the tick and at the tick-feeding site. It is important to note here that based on transcript abundance, AAS19 is predominant in the tick MG, which could strongly indicate a role for this protein in the MG. Tick digestion of the blood meal nutrients is intracellular (Lara et al., 2005; Franta et al., 2010), and for this to happen, host blood must remain in the unclotted state. Could it be that AAS19 anti-coagulant function is part of the mechanism in the MG to prevent host blood from clotting before digestion commences?

In conclusion, data in this study contribute towards our understanding of the molecular basis of tick feeding physiology. Data here contribute to the growing list of tick saliva proteins that potentially regulate tick evasion of the mammalian host’s defense mechanisms to tick feeding as well as having role(s) in other tick physiological systems. On the basis of AAS19 being highly conserved, coupled with the fact that its RCL is 100% conserved, we conclude that tick physiological processes regulated by AAS19 are evolutionary conserved in all tick species. From the perspective of our long-term interest to find target tick vaccine antigens, AAS19 represents an interesting candidate. Human and animal tick borne diseases are transmitted by different species of ticks, and thus developing universal tick vaccines based on highly conserved proteins such as AAS19 has been advocated (Fragoso et al., 1998; Mulenga et al., 2001, 2013; de la Fuente and Kocan, 2006; Azhahianambi et al., 2009; Parizi et al., 2012). Work toward validating the tick vaccine efficacy of rAAS19 is warranted.

Highlights.

• AAS19 protein sequence is highly conserved with other ixodid tick species.

• rAAS19 is a broad spectrum inhibitor of trypsin and trypsin-like proteases.

• rAAS19 prevents platelet aggregation and delays plasma clotting.