Shrew's venom quickly causes circulation disorder, analgesia and hypokinesia

Abstract

Multiple representatives of eulipotyphlan mammals such as shrews have oral venom systems. Venom facilitates shrews to hunt and/or hoard preys. However, little is known about their venom composition, and especially the mechanism to hoard prey in comatose states for meeting their extremely high metabolic rates. A toxin (BQTX) was identified from venomous submaxillary glands of the shrew Blarinella quadraticauda. BQTX is specifically distributed and highly concentrated (~ 1% total protein) in the organs. BQTX shares structural and functional similarities to toxins from snakes, wasps and snails, suggesting an evolutional relevancy of venoms from mammalians and non-mammalians. By potentiating thrombin and factor-XIIa and inhibiting plasmin, BQTX induces acute hypertension, blood coagulation and hypokinesia. It also shows strong analgesic function by inhibiting elastase. Notably, the toxin keeps high plasma stability with a 16-h half-life in-vivo, which likely extends intoxication to paralyze or immobilize prey hoarded fresh for later consumption and maximize foraging profit.

Keywords: Blarinella quadraticauda; Thrombin. FXIIa; Toxin; Venom.

Fig. 1

Amino acid (aa) sequence and tissue distribution of BQTX. A cDNA sequence encoding BQTX precursor. Signal peptide is shown in italics. Amino acid sequences of N-terminal and interior fragments, determined by automated Edman degradation, are boxed. Two domains of BQTX are underlined. B Tissue distribution of BQTX was determined by western blot analysis. Lane 1, submaxillary gland; Lane 2, heart; Lane 3, liver; Lane 4, lung; Lane 5, skin; Lane 6, muscle. To assess equal loading of protein, the membranes were stripped and reprobed for β-actin. C Sequence alignment of domains 1 and 2 of BQTX with serine protease inhibitor toxins from snake, wasp, and snail venom. Identical residues are highlighted in blue rectangles. The P1 residues are shown in red. *cysteine

Fig. 2

Effects of BQTX on serine proteases and its stability in plasma. A–E Effects of BQTX on hydrolytic ability of trypsin, elastase, plasmin, thrombin, and FXIIa were determined using synthetic substrates. Testing enzyme was incubated with different concentrations of BQTX in 60 μl of Tris-buffer for 5 min at 37 °C, and then a certain concentration of chromogenic substrate was added. Absorbance was monitored at 405 nm immediately. Data represent mean ± SD of five independent experiments, *P < 0.05, **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test (D and E). F BQTX showed high stability in vivo. BQTX (2 mg) was injected into mouse tail vein. Blood was drawn from tail every 2 h, and plasma was collected and used for ELISA

Fig. 3

BQTX directly interacts with plasmin, thrombin, and FXIIa, and inhibits/potentiates hydrolytic abilities of proteases. A BQTX was immobilized on CM5 sensor chip by covalent bond. Plasmin (30 μl) diluted with HEPES-EB buffer (pH 7.4) was applied to surface of CM5 with a flow rate of 10 μl/min. B Representative SDS-PAGE analysis of fibrinogen hydrolysis by plasmin. Lane 1, 20 μg of fibrinogen; lane 2, 20 μg of fibrinogen was incubated with 0.1 NIH units of plasmin in Tris–HCl (50 mM, pH 7.4) for 30 min; lane 3–5: 20 μg of fibrinogen was incubated with 0.1 NIH units of plasmin and 1 000, 200, and 40 nM BQTX for 30 min in Tris–HCl (50 mM, pH 7.4), respectively. C Quantification of release of degradation products (DP) of ‘B’. D SPR analysis of interaction between BQTX and thrombin. E Representative RP-HPLC analysis of release of fibrinopeptide A and fibrinopeptide B (FbpA and FbpB). Panel 1, 40 μl of human thrombin (0.1 NIH units) in 25 mM Tris–HCl (pH 7.4) was added to 500 μl of 10 mg/ml fibrinogen in 50 mM Tris–HCl (pH 7.8) containing 0.15 M NaCl. After incubation for 30 min at 37 °C, 500 μl of 20% trichloroacetic acid (TCA) was added to stop reaction, with mixture then centrifuged 4 °C for 10 min at 12 000 rpm to precipitate insoluble protein. A 700-μl aliquot of supernatant was injected into RP-HPLC C₁₈ column. 1, 2, 3, and 4 indicate RP-HPLC analysis after treatment with 0, 40, 200, and 1 000 nM BQTX, respectively. F Quantification of release of FbpA and FbpB of ‘E’. G SPR analysis of interaction between BQTX and FXII. H Representative SDS-PAGE analysis of prekallikrein hydrolysis by FXIIa. Lane 1: 10 μg of prekallikrein; lane 2: 10 μg of prekallikrein was incubated with 0.01 NIH units of FXIIa in Tris–HCl buffer (50 mM, pH 7.4) for 30 min; lane 3, 4 and 5: 10 μg of prekallikrein was incubated with 0.01 NIH units of FXIIa and 1 000, 200, and 40 nM BQTX in Tris–HCl buffer (50 mM, pH 7.4) for 30 min, respectively. I Quantification of release of height chain of kallikrein (HC) of ‘H’. Data represent mean ± SD of five independent experiments, *P < 0.05, **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test (C, F, and I)

Fig. 4

Two Kazal domains of BQTX specifically target different serine proteases. A Schematic of mature BQTX, BQTX domain 1 (D1), BQTX domain 1 29th D mutated to A (D1 D29A), BQTX domain 1 24th E mutated to A (D1 E24A), BQTX domain 2 (D2), BQTX domain 2 112th E mutated to A (D2 E112A), and BQTX domain 2 107th D mutated to A (D2 D107A). B BQTX D1 showed inhibitory ability against plasmin and trypsin. (C) BQTX D2 inhibited elastase without affecting plasmin and trypsin. D, E BQTX D1 enhanced hydrolytic ability of thrombin and FXIIa. F, G BQTX D2 enhanced hydrolytic ability of thrombin and FXIIa. H, I Effects of BQTX D1 mutants on hydrolytic ability of thrombin and FXIIa were determined. J, K Effects of BQTX D2 mutants on hydrolytic ability of thrombin and FXIIa were determined. Data represent mean ± SD of six independent experiments, *P < 0.05, **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test (D–K)

Fig. 5

BQTX induces acute thrombosis and persistent hypertension. A BQTX enhanced platelet aggregation induced by thrombin. Washed human platelets were pretreated with BQTX (1 000, 200, and 40 nM) for 5 min and then stimulated with thrombin (40 nM) for 5 min. n = 8 per group. B Maximum platelet aggregation rate (MPAR) of ‘A’ was calculated. C BQTX dose-dependently reduced plasma recalcification time. Plasma (20 μl) was pretreated with BQTX (0, 40, 200, or 1 000 nM) in 60 μl of HEPES buffer for 10 min. Then, 60 μl of 25 mM CaCl₂ preheated at 37 °C for 10 min was added. n = 10 per group. D Clotting time of ‘C’ was calculated by measuring time to half maximal increase in absorbance. E BQTX dose-dependently reduced activated partial thromboplastin time (APTT) and prothrombin time (PT), n = 8 per group. F, G BQTX D1 and BQTX D2 dose-dependently reduced APTT and PT, n = 8 per group. Data represent mean ± SD of five independent experiments, *P < 0.05, **P < 0.01 by one-way ANOVA with Dunnett’s post hoc test (B and D). H BQTX decreased mouse (BALB/c) tail bleeding time. At 20 min after BQTX (0.4, 0.2, and 0.1 mg/kg), 6-aa (0.4 mg/kg), or saline administration, mouse tails were transected at approximately 2 mm from tail tip. Bleeding tails were immediately immersed in normal saline, and time from initiation to termination of bleeding was recorded as bleeding time. n = 6 per group. I, J BQTX aggravated mouse (BALB/c) tail and paw thrombosis induced by carrageen. Mice received (i.v.) BQTX (0.4, 0.2, and 0.1 mg/kg), 6-aa (0.2 mg/kg), or saline. At 30 min after injection, mice were administered (i.p.) 0.25 ml of 1% carrageenan, then maintained at 17 °C. Thrombosis region lengths in tails and grayscale change in paws were measured and photographed at 24 h after carrageenan injection. n = 6 per group. K, L Effects of BQTX on FeCl₃-induced carotid artery thrombus formation in C57BL/6J mice. Representative images of carotid artery blood flow (top) and quantification (bottom) are shown. Red: blood flow; Blue and black area: background; Color bar on right indicates perfusion unit scale (0–240). Left common carotid arteries were exposed by cervical incision and separated from adherent tissue and vagus nerve. Injury was induced by contact with a cross-section of a filter paper strip (2 × 2 mm) saturated with 10% FeCl₃ (W/V) for 90 s. n = 5 per group. Animal experiments were repeated three times independently. M Effects of BQTX on blood pressure in rhesus macaques (Macaca mulatta). After macaques were acclimated for 30 min, both systolic and diastolic blood pressure levels were continuously recorded for 20–30 min before and 100–120 min after intravenous administration of different concentrations of BQTX in right forearm, n = 3 per group. Data represent mean ± SD, *P < 0.05, **P < 0.01, by one-way ANOVA with Dunnett’s post hoc test (H, I, J, and L)

Fig. 6

BQTX shows analgesic effects. A, B Pretreatment with BQTX (5 and 1 mg/kg) selectively attenuated mouse (BALB/c) nociceptive behavior in phase II (10–30 min) following right hind paw injection of 20 μl of 5% formalin. Formalin was injected 30 min after BQTX (5, 1, and 0.2 mg/kg), morphine (0.2 mg/kg), or saline pretreatment. Phase II licking time following BQTX treatment decreased significantly compared with that in saline-pretreated mice. Phase I (0–10 min) licking time did not differ significantly from that observed in corresponding saline group. n = 5 per group. C BQTX reduced writhing response to acetic acid in BALB/c mice. Number of writhes was counted between 0 and 30 min after i.p. injection of 0.1 ml/10 g body weight of acetic acid (concentration of 0.6% (v/v)). Mice received an injection (i.v.) of BQTX (5, 1, and 0.2 mg/kg), morphine (0.2 mg/kg), or saline 30 min before acetic acid injection. n = 5 per group. D Effects of BQTX and morphine on reaction time to thermal stimulus in BALB/c mice. BQTX had no effect on thermal pain. Mice were pretreated (i.v.) with saline, morphine (0.2 mg/kg), or BQTX (5, 1, and 0.2 mg/kg), with reaction times measured 30 min after injection. n = 5 per group. Animal experiments were repeated three times independently. Data represent mean ± SD, *P < 0.05, **P < 0.01, by one-way ANOVA with Dunnett’s post hoc test (A–D)