Platelet activation and erythrocyte lysis during brief exposure of blood to pathophysiological shear stress in vitro

Abstract

BACKGROUND:

Interaction of von Willebrand factor (VWF) with circulating platelets is the trigger for thrombosis in a region of arterial stenosis. These events are typically studied in vitro under conditions where platelets adhere to a VWF-coated surface. Our approach assesses platelet responses in the absence of adhesion.

OBJECTIVE:

To characterize extent of platelet activation and erythrocyte lysis in an artificial stenosis model.

METHODS:

Whole blood is perfused through a length of polyetheretherketone tubing that includes an artificial stenosis, comprising narrow-bore (89–381 μm) tubing. Secretion of [¹⁴C] serotonin and hemoglobin release was measured to evaluate platelet activation and hemolysis respectively at various perfusion rates and different stenosis dimensions.

RESULTS:

Platelet activation and erythrocyte lysis increased progressively with increasing perfusion rate and decreasing stenosis diameter; the length of the stenosis had negligible influence. Modest platelet activation (5–10% secretion of [¹⁴C] serotonin) occurred without significant erythrocyte lysis under a limited range of perfusion conditions (4–6 mL/min flow through a 127 μm stenosis).

CONCLUSIONS:

Our experimental approach mimics conditions in severe arterial stenosis or a mechanical heart valve. It could be a valuable aid in the development of novel drugs to treat arterial thrombosis and in the design of heart valves.

Keywords

Arterial stenosis erythrocyte platelet shear stress thrombosis

Abbreviations

[Ca2+]i

intracellular free calcium concentration

GPIb–IX–V

glycoprotein Ib–IX–V

GPIIb–IIIa

glycoprotein IIb–IIIa

PEEK

polyetheretherketone

VWF

von Willebrand factor

1 Introduction

Blood platelets play a predominant role in arterial thrombosis through interaction with the multimeric glycoprotein von Willebrand factor (VWF). Platelets flowing through a region of arterial stenosis encounter a high shear stress. This promotes rolling adhesion of platelets through interaction of the glycoprotein Ib–IX–V (GPIb–IX–V) receptor complex with VWF exposed on a damaged region of vascular subendothelium [1 –3]. Such interaction triggers a complex series of signaling events leading to an increase in intracellular calcium concentration [Ca²⁺]i in the platelet and functional activation of another receptor, glycoprotein IIb–IIIa (GPIIb–IIIa) [1 , 4–6]. Binding of fibrinogen or VWF to GPIIb–IIIa promotes platelet aggregation. The intracellular signaling events also lead to secretion from the platelet granules of substances, such as ADP, serotonin and VWF, that further enhance platelet activation and aggregation, ultimately leading to formation of a thrombus. Exposure to high shear stress without adhesion to the subendothelium appears to activate platelets through a similar VWF-dependent mechanism [7, 8]. Studies performed in vitro have shown little increase in [Ca²⁺]i when platelets are subjected to a high shear stress in a cone and plate viscometer in the absence of exogenous VWF. However, a pronounced [Ca²⁺]i rise is seen in the presence of VWF. A limitation of most such studies is that the platelets are typically exposed to a high shear stress for a much longer duration than occurs in vivo under physiological or pathophysiological conditions [9].

Blood flowing through a mechanical heart valve is also subjected briefly to a very high shear stress. Prosthetic valves are used to treat certain heart conditions, such as rheumatic heart disease. A problem with such valves is that they are prone to cause thrombosis at the site of the prosthetic valve or thromboembolism when a clot in the valve becomes dislodged [10]. The foreign surfaces that blood is exposed to within a prosthetic valve and the elevated shear stress both contribute to the increased risk of thrombosis [11]. Blood platelets flowing through an artificial heart valve encounter a complex hemodynamic flow pattern. The high velocity leakage flow generates regions of high shear stress that are the most likely sites for platelet activation; platelets also tend to become trapped in regions of stagnant flow or recirculation in the vicinity of the valve, creating an environment that is highly conducive to clot formation [12 –14]. Moreover, the extremely high shear forces near the edge of the leaflet in a typical prosthetic heart valve may directly cause the rupture of red blood cells or platelets [12 , 15]. Release of ADP and ATP from the red blood cells and platelet δ-granules would then lead to further platelet activation.

We have developed a simple artificial model of a stenosis to study platelet activation during very brief exposure of blood to a high shear stress. The aim of the current study was to characterize how the extent of platelet activation and erythrocyte lysis varied with the dimensions of this artificial stenosis and with the rate of blood flow. We also assessed the pressure gradient across the stenosis as an indicator of whether or not the pattern of blood flow complied with the predictions for laminar flow of a Newtonian fluid.

2 Materials and methods

2.1 Collection of human blood, loading of platelets with [¹⁴C] serotonin, and reconstitution of blood

Venous blood (50 mL) was drawn using a light tourniquet from apparently healthy adult volunteers who had not taken aspirin or a non-steroidal anti-inflammatory drug for at least 10 days prior to the blood draw. The protocol was approved by the local Institutional Review Board (protocol# 2007-64), and informed consent obtained from each donor. Blood was collected into 40 μmol/L (final concentration) Phe-Pro-Arg-chloromethylketone (Haematologic Technologies, Essex Junction, VT, USA) as anticoagulant. This agent is a direct inhibitor of thrombin and is an adequate anticoagulant for studies on whole blood as it maintains the normal free calcium concentration and pH in plasma [16]. Platelet-rich plasma was obtained by centrifugation of the blood at 100 g for 20 min at 23°C. After addition of 5 mmol/L phosphocreatine and 25 U/mL creatine phosphokinase (Sigma Chemicals, St Louis, MO), the platelets were isolated by centrifugation at 1,000 g for 15 min at 23°C. The platelets were resuspended in calcium-free, magnesium-free Tyrode’s solution (137 mmol/L NaCl, 2.7 mmol/L KCl, 12 mmol/L NaHCO3, 0.4 mmol/L NaH2PO4, 5.5 mmol/L D-glucose), supplemented with 10 mmol/L HEPES, pH 7.35, 0.35% (w/v) bovine serum albumin, 1 mmol/L EGTA, 5 mmol/L phosphocreatine and 25 U/mL creatine phosphokinase.

The platelet δ-granules were loaded with 1 μCi 5-hydroxy-[2-¹⁴C] tryptamine (serotonin) binoxalate (40–60 Ci/mol; PerkinElmer, Boston, MA) through 1 h incubation at 23°C. The platelets were then precipitated by centrifugation, resuspended in autologous platelet-poor plasma and mixed with the packed erythrocyte pellet to attain a platelet count of 300,000 cells/μL and a hematocrit of 0.4 (40%). The platelet count for each reconstituted blood sample was confirmed with a Cell-Dyn 900 Hematology Analyzer (Sequoia-Turner, Mountain View, CA, USA) and the hematocrit checked with a CritSpin microcentrifuge (StatSpin, Norwood, MA, USA).

2.2 Artificial stenosis model for assessment of platelet activation under high shear stress

An artificial stenosis was created by connecting a 30–80 mm length of precision narrow-bore (60–380 μm internal diameter) polyetheretherketone (PEEK) tubing (Upchurch Scientific, Oak Harbor, WA, USA) between two longer segments of wide-bore (1 mm diameter) PEEK tubing using zero dead-volume connectors (Fig. 1). One segment of the wide-bore tubing was connected through a T-junction to an 8 mL stainless steel syringe mounted on a high-pressure perfusion pump (model PHD 4400, Harvard Apparatus, Holliston, MA, USA). A further segment of wide-bore tubing attached to the side-arm of the T-junction provided a means to fill or refill the syringe with blood without it passing through the artificial stenosis. The stainless steel syringe and the segment of PEEK tubing with the artificial stenosis were maintained at 37°C by means of a water jacket and a dry-bath heater, respectively.

Fig.1

Experimental set-up for artificial stenosis model. An artificial stenosis comprising a 30–80 mm length of narrow-bore (60–380 μm internal diameter) PEEK tubing was assembled between two segments of wide-bore (1 mm diameter) tubing using zero dead-volume connectors. One end of the wide-bore tubing was connected through a T-junction to an 8 mL stainless steel syringe mounted on a perfusion pump. The other end of the wide-bore tubing was directed into a microcentrifuge vial for collection of perfusate samples. Another piece of wide-bore tubing attached to the side-arm of the T-junction provided a means to fill or refill the syringe with blood. The outlet tube was plugged while filling or refilling the syringe, and the filling tube plugged during perfusion through the stenosis. The stainless steel syringe and the artificial stenosis were maintained at 37°C.

Reconstituted whole blood (containing [¹⁴C] serotonin-loaded platelets) was supplemented with 1 μmol/L imipramine (Sigma Chemicals, St Louis, MO, USA) to preclude further reuptake of released serotonin, and equilibrated to 37°C for 5 min in the stainless steel syringe before perfusion at a defined rate (2–12 mL/min) through the tubing with the artificial stenosis. Duplicate 1.5 mL samples of blood were perfused under each condition. The perfusate was collected from the outflow tubing into a solution of 1 mmol/L aspirin and 5 mmol/L EDTA to preclude further platelet activation. It was then centrifuged at 14,000 g for 2 min at 23°C, and the supernatant fraction aspirated carefully to assess platelet activation and erythrocyte lysis.

The degree of platelet activation was determined by measuring [¹⁴C]serotonin released from the platelet δ-granules as a percentage of the total [¹⁴C]serotonin in the blood; the total content was assessed after cell lysis with 1% (vol/vol) Triton X-100. The general approach was based on the procedure of Holmsen and Dangelmaier [17]. However, hemoglobin released from whole blood causes severe color quenching and interferes with quantification of ¹⁴C. Such interference was eliminated by Zn²⁺ chelation of the released hemoglobin before scintillation counting. A portion (100–200 μL) of each supernatant fraction or total lysate was mixed with 3 ml distilled water before addition of ZnCl₂ to 5 mmol/L final concentration. Centrifugation at 14,000 g for 5 min at 23°C selectively precipitated the hemoglobin in a complex with Zn²⁺ [18]. This procedure was repeated a second time to ensure effective removal of the hemoglobin. Control studies verified that the procedure did not cause any significant loss of [¹⁴C] serotonin during removal of the hemoglobin. The ¹⁴C content of each processed sample was assessed in a Tri-Carb 2500 liquid scintillation counter (PerkinElmer) using a quench correction protocol.

2.3 Assessment of erythrocyte lysis

Erythrocyte lysis was also assessed for the samples of whole blood perfused through the artificial stenosis under various conditions. The hemoglobin content of the supernatant fraction from each perfusate was measured as an index of erythrocyte lysis. Samples (20 μL) were diluted appropriately (from 1 : 50 to 1 : 1,000 based on visual inspection of color) in 0.01% (w/v) Na2CO3 solution. Hemoglobin content was assessed from the difference in absorbance at 405 nm (Soret peak region for most forms of hemoglobin [19]) and 450 nm (correction for albumin and other constituents of plasma), measured with a THERMOmax microplate reader (Molecular Devices, Sunnyvale, CA, USA). Data are expressed as percentage erythrocyte lysis on the basis of a calibration curve for dilutions of the reconstituted whole blood lysed with 1% Triton X-100.

2.4 Measurement of pressure differential across the artificial stenosis

The tubing arrangement for our artificial stenosis model was reconfigured to include an additional T-junction on each side of the narrow-bore stenosis tubing to allow measurement of the pressure drop across the stenosis section. The side-arms of these T- junctions were connected to the + and –ports of a DP15TL pressure transducer (Validyne, Northridge, CA, USA) by further segments of wide-bore PEEK tubing filled with 40% (v/v) glycerol. The pressure transducer was equipped with a 3–64 diaphragm and was calibrated using a GaugeCalXP pressure comparator (Crystal Engineering Corp, San Luis Obispo, CA, USA). Whole blood (8 mL) was perfused at a defined rate (2–12 mL/min) at 37°C through the artificial stenosis and the pressure drop across the stenosis was recorded midway through the perfusion.

2.5 Statistical analysis

Data are presented as a mean value±standard error of the mean (SEM). One-way repeated-measures analysis of variance (ANOVA) was used in conjunction with Tukey’s post-hoc test to determine the statistical significance of differences between experimental groups.

3 Results

3.1 Blood flow through the artificial stenosis is consistent with predictions for laminar flow

A series of studies was conducted in 5 different volunteers to assess how the pressure drop across the stenosis varied with the rate of perfusion of the blood as well as the length and internal diameter of the stenosis. The pressure drop across a stenosis of fixed length (40 mm) and internal diameter (127 μm) increased proportionally as the perfusion rate was increased (Fig. 2A). Similarly, with a fixed perfusion rate (12 mL/min), the pressure drop increased in proportion with the length of the stenosis (Fig. 2B). Changes in the internal diameter of the stenosis had a more profound effect on the pressure drop, which varied in inverse proportion to the fourth power of the diameter (Fig. 2C).

Fig.2

Effect of perfusion rate, length and internal diameter of the stenosis on the pressure differential across the stenosis tube. Whole blood was perfused through an artificial stenosis comprising a section of narrow-bore PEEK tubing between two segments of wide-bore (1 mm) PEEK tubing. The pressure differential was assessed with a pressure transducer. (A) The perfusion rate (2–12 mL/min) was varied with a fixed length (40 mm) and internal diameter (127 μm) for the narrow-bore tubing. (B) The length (30–80 mm) of the narrow-bore tubing was varied with a fixed perfusion rate (12 mL/min) and stenosis diameter (127 μm). (C) The diameter (89–381 μm) of the narrow-bore tubing was varied with a fixed perfusion rate (12 mL/min) and stenosis length (40 mm). Data (mean±SEM) from 3–5 studies of each type are shown together with the predicted pattern (hatched line) for Poiseuille flow.

3.2 Platelet activation and erythrocyte lysis increase with increasing perfusion rate through the stenosis

Platelet activation and erythrocyte lysis were assessed in parallel for samples of reconstituted whole blood perfused at different rates (2–12 mL/min) through a stenosis of 40 mm length and 127 μm diameter in a total of 7 donors, two of whom gave blood twice. Control measurements were made in the absence of the stenosis in 4 separate donors. Platelet activation during perfusion through the stenosis increased progressively as the perfusion rate was increased from 2 to 12 mL/min (Fig. 3A). In contrast, there was negligible increase in platelet activation in the absence of the stenosis; [¹⁴C] serotonin in the supernatant fraction (6% typically) under the latter circumstances reflects contributions from serotonin that was not taken up into the platelets and serotonin released spontaneously before the perfusion. Release of [¹⁴C] serotonin during perfusion at 4 mL/min or faster through the stenosis was significantly higher than in the control conditions without a stenosis (P < 0.02, one-way repeated measures ANOVA with Tukey’s post-hoc test). Up to 14% serotonin release was noted at the highest perfusion rate.

Fig.3

Effect of perfusion rate on the extent of platelet activation and erythrocyte lysis in the stenosis tube. Whole blood was perfused at different rates (2–12 mL/min) through an artificial stenosis section comprising a 40 mm length of narrow-bore (127 μm diameter) PEEK tubing between segments of wide-bore (1 mm) tubing. (A) Platelet activation was monitored through the release of [¹⁴C]serotonin. (B) Erythrocyte lysis was assessed from the release of hemoglobin. Data (mean±SEM) from 9 studies with the stenosis (hatched bars) and 4 studies without a stenosis (open bars) are shown; there was negligible erythrocyte lysis without a stenosis. ^*Significantly different from platelet activation or erythrocyte lysis at 2 mL/min perfusion rate (P < 0.001). ^#Significantly different from platelet activation or erythrocyte lysis without a stenosis (P < 0.02). Shear stresses were estimated assuming laminar flow and an effective viscosity of 0.0035 Pa.s for blood under arterial flow conditions [12, 13].

Perfusion of the blood at a higher rate through the stenosis caused progressively greater lysis of the red blood cells (Fig. 3B). In contrast, negligible (<0.1%) erythrocyte lysis occurred in the absence of the stenosis. Erythrocyte lysis was not significantly greater in the presence of the stenosis for perfusion rates in the 2–6 mL/min range (P > 0.5). Perfusion through the stenosis at 8 mL/min or faster caused significantly greater lysis of erythrocytes than perfusion at the slowest rate of 2 mL/min (P < 0.01). About half of the erythrocytes were lysed at the highest perfusion rate of 12 mL/min when the stenosis was present.

3.3 Platelet activation and erythrocyte lysis increase with decreasing stenosis diameter

Platelet activation and erythrocyte lysis were evaluated in parallel for perfusion of blood at 12 mL/min through a series of artificial stenosis of different internal diameter (89–1016 μm). There was negligible activation of the platelets with stenosis of diameter 203–1016 μm (Fig. 4A), but platelet activation increased significantly with the stenosis of diameter 127 μm (P < 0.02) and showed a substantial increase with that of diameter 89 μm (P < 0.001).

Fig.4

Effect of stenosis diameter on the extent of platelet activation and erythrocyte lysis in the stenosis tube. Whole blood was perfused at 12 mL/min through an artificial stenosis section comprising a 40 mm length of narrow-bore (89–1016 μm diameter) PEEK tubing between two segments of wide-bore (1 mm) tubing. Platelet activation (A) and erythrocyte lysis (B) were monitored. Data (mean±SEM) from 5–7 studies are shown. ^*Significantly different from platelet activation or erythrocyte lysis with 1016 μm diameter stenosis (P < 0.02).

There was negligible erythrocyte lysis during perfusion of blood through the stenosis of diameter 203–1016 μm (Fig. 4B). Perfusion through the stenosis of diameter 127 μm caused substantial erythrocyte lysis (P < 0.001 compared with each of the wider stenosis). Essentially complete erythrocyte lysis occurred with the narrowest stenosis of diameter 89 μm (P < 0.001).

3.4 Platelet activation and erythrocyte lysis are unaffected by the duration of exposure to shear stress

Platelet activation and erythrocyte lysis were evaluated in parallel for perfusion of blood at 2 or 12 mL/min through a series of artificial stenosis of different length (30–80 mm) in a total of 9 donors, 2 of whom gave blood twice. The extent of platelet activation at a perfusion rate of 12 mL/min was essentially the same regardless of the length of the stenosis (Fig. 5A; P > 0.5). There was negligible platelet activation with a perfusion rate of 2 mL/min.

Fig.5

Effect of stenosis length on the extent of platelet activation and erythrocyte lysis in the stenosis tube. Whole blood was perfused at 2 mL/min (open bars) or 12 mL/min (hatched bars) through an artificial stenosis comprising various lengths (30–80 mm) of narrow-bore (127 μm) PEEK tubing between segments of wide-bore (1 mm) tubing. Platelet activation (A) and erythrocyte lysis (B) were monitored. Data (mean±SEM) from 3–10 studies are shown. ^*Significantly different from platelet activation or erythrocyte lysis at 2 mL/min perfusion rate (P < 0.001).

Erythrocyte lysis at a perfusion rate of 12 mL/min was not significantly affected by the length of the stenosis (Fig. 5B; P > 0.1). There was negligible erythrocyte lysis at the slower perfusion rate of 2 mL/min.

3.5 Correlation between platelet activation and erythrocyte lysis

The extent of platelet activation under the various perfusion conditions (different perfusion rates, stenosis diameters and lengths) was compared with the degree of erythrocyte lysis under the same conditions (Fig. 6). A close correlation was found between erythrocyte lysis and platelet activation (r = 0.894, P < 0.001).

Fig.6

Correlation between platelet activation and erythrocyte lysis in the stenosis tube under different perfusion conditions. Data on platelet activation are compared with those on erythrocyte lysis for the studies with different perfusion rates (open circles), different stenosis diameters (open triangles) and different stenosis lengths (open squares). Composite data (closed circle) derived from all three sets of studies are shown for the standard stenosis (40 mm length, 127 μm diameter) and perfusion conditions (12 mL/min perfusion rate). Basal serotonin secretion (6%) in the absence of a stenosis is depicted by the hatched horizontal line; this reflects contributions from serotonin that was not taken up into the platelets and serotonin released spontaneously before the perfusion.

4 Discussion

Blood is exposed very briefly to a high shear stress when it passes through a region of arterial stenosis in vivo. Rupture of atherosclerotic plaque in the stenosis may expose the subendothelial matrix, a highly thrombogenic surface rich in VWF and collagen. Platelets adhere to this matrix, initially through interaction of platelet GPIb–IX–V with VWF then subsequently through interaction of glycoprotein VI and integrin α2β1 (glycoprotein Ia–IIa) with collagen [1 , 20]. These interactions trigger an intricate network of intracellular signaling events that lead to functional activation of GPIIb–IIIa, secretion from the α-granules and δ-granules, and ultimately formation of a thrombus [5, 20]. Exposure to a high shear stress without adhesion to the subendothelial matrix also appears to activate platelets through a VWF-dependent mechanism [7, 8].

Assessment of platelet activation under shear stress in vitro can yield invaluable information about the signaling mechanism that cannot readily be ascertained from studies in vivo. Several approaches have been developed to assess shear-induced platelet activation in vitro¹. One approach is to subject whole blood, platelet-rich plasma or washed platelets to a defined shear stress in a cone and plate viscometer. An advantage of this method is that it allows real-time measurement of changes in platelet [Ca²⁺]i and aggregation when a platelet suspension plasma is used [7, 8]. A disadvantage is that prolonged exposure (1–2 min) to a high shear stress appears necessary if platelet responses are to occur. This exposure time is not representative of exposure to high shear stress under physiological or pathophysiological conditions in vivo [21]. Another approach is to perfuse blood or platelet-rich plasma through a parallel-plate perfusion chamber that has one surface coated with VWF or collagen [22 –25]. An advantage of this approach is that the duration of exposure of individual platelets to a high shear stress is relatively short. Another advantage is that the behavior of individual platelets can be observed by videomicroscopy [6, 25]. A disadvantage is that the rectangular geometry of the perfusion chamber differs from the circular cross-section of an artery, so the flow streamlines are not directly comparable. This disadvantage has been obviated in a few studies by replacing the parallel-plate perfusion chamber with a glass capillary tube coated with VWF or collagen [26, 27].

We also adopted a perfusion approach for our studies, but chose PEEK tubing for the shear region so as to examine platelet activation in the absence of adhesion. The inert polymer PEEK provides a non-thrombogenic surface to which there is minimal adhesion of proteins [28]. A particular advantage of PEEK tubing is that it is available commercially in a wide range of different internal diameters. This allowed us to construct a series of artificial stenoses differing in severity and length to define how the geometry of the stenosis governed the extent of platelet activation. Our findings indicate that the severity (internal diameter) of the stenosis is the most critical factor in determining platelet activation (Fig. 4A). There was negligible activation of the platelets during perfusion of blood through stenoses of diameter 203–1016 μm, whereas platelet activation was significant with a stenosis of diameter 127 μm and substantial with a 89 μm stenosis.

Fallon et al. have reached a similar conclusion about the critical importance of cross-sectional diameter in a high shear region to thrombus formation [11], albeit using a quite different experimental approach. Blood was collected into 3.2% sodium citrate anticoagulant and gradually recalcified during their study. Whole blood was exposed to a high shear stress during passage through a circular or rectangular orifice that mimics a mechanical heart valve. Exposure to high shear was repetitive, as blood was recirculated through the orifice using a centrifugal bypass pump. The smallest diameter (200 μm) circular orifice was invariably occluded completely by thrombus at the end of their studies, whereas there was minimal or no thrombosis with orifices of larger diameter [11]. This study together with our data suggests that there may be a critical threshold for the diameter of a stenosis below which platelet activation readily occurs. Moreover, the threshold diameter of 127 μm observed in our studies corresponds closely to the minimum hinge diameter in some mechanical heart valves [13, 29].

On the basis of measurements of platelet adhesion in stenosis models, Bluestein et al. have proposed that the extent of shear-induced platelet activation depends on the product of shear stress and exposure duration integrated over the total exposure period [30]. However, our findings with the different perfusion rates (Fig. 3A) and stenosis sections of different length (Fig. 5A) are at variance with this hypothesis. Assuming that laminar flow conditions prevail, an increase in perfusion rate with a stenosis of fixed length should result in a proportional increase in shear stress in the stenosis but a proportional decrease in the duration of exposure to shear stress. The product of shear stress and exposure duration should thus be unchanged. Nonetheless, our experimental data indicate that platelet activation increased progressively with increasing perfusion rate (Fig. 3A). An increase in the length of the stenosis with a fixed perfusion rate does not alter the magnitude of the shear stress in the stenosis but results in a proportional increase in its duration. Despite the increased duration of exposure to shear stress in a stenosis of 80 mm length, platelet activation was no greater than in a 30 mm long stenosis (Fig. 5A). The average duration of exposure to a high shear stress is estimated as 1.9 ms for the 30 mm long stenosis and 5.1 ms for the 80 mm stenosis. This short duration for exposure of the platelets to a high shear stress is more representative of flow of blood through a region of arterial stenosis or a mechanical heart valve than the typical durations for exposure in a cone and plate viscometer [9, 21].

Extensive erythrocyte lysis occurred under most conditions where platelet activation was seen in our studies. Moreover, the degree of platelet activation under various perfusion conditions correlated strongly with the extent of erythrocyte lysis under the same conditions (Fig. 6). These patterns suggest that release of ADP during lysis of the erythrocytes may play an important role in triggering platelet activation under the pathophysiological shear conditions of our artificial stenosis model. However, significant platelet activation also occurred in some situations where there was little or no lysis of the erythrocytes. With a stenosis of 127 μm diameter and 40 mm length, for example, significant activation of the platelets occurred at a perfusion rate of 4 mL/min (Fig. 3A), but no erythrocyte lysis was detected (Fig. 3B). This observation implies that platelet activation in our model, under at least some conditions, is mediated by a mechanism other than ADP release from lysed erythrocytes. Further studies are needed to assess whether a VWF-dependent mechanism is involved and to define the signaling pathway that underlies platelet activation in our stenosis model.

Our measurements of the pressure differential across the stenosis section under various conditions yielded data consistent with the predictions for laminar flow of a Newtonian fluid in accord with the Poisseuile equation. Blood cannot truly be described as a Newtonian fluid, as its viscosity varies with shear rate and its components tend to separate during rapid flow through a blood vessel or tube [31, 32]. Due to their larger size and deformability, the erythrocytes have a propensity to occupy the center of the vessel where the velocity of flow is greatest but shear stress is least. In contrast, the platelets are forced towards the periphery of the vessel where the velocity is least but the shear rate is greatest. The platelets thus tend to experience much higher shear forces than the erythrocytes. An approximate estimate can be made of the shear stress encountered by platelets adjacent to the wall of the tubing in our artificial stenosis, if laminar flow is assumed to prevail in this region. Shear stress is estimated to range from 70 to 400 Pa (700–4,000 dyn cm–2) in the 127 μm diameter stenosis as the perfusion rate varies from 2 to 12 mL/min. An even higher shear stress of 1,300 Pa would pertain in the 89 μm diameter stenosis with a perfusion rate of 12 mL/min. In contrast, shear stresses in the range 40–200 Pa have been estimated for an arterial stenosis in vivo [3 , 34] and in the range 100–1,000 Pa in the hinge region of a mechanical heart valve [11 , 14]. The shear stresses at which platelet activation was observed in our artificial stenosis model are thus at the upper limit or above those likely to be encountered in vivo. Nonetheless, it should be emphasized that we have assessed platelet activation after a single passage through the stenosis in vitro, whereas blood would recirculate through this region multiple times in vivo, presumably resulting in progressively greater activation of the platelets. Furthermore, our results might be more significant if we enrolled subjects with a history of atherosclerotic disease [35].

Our in vitro model of an arterial stenosis grossly simplifies the situation that pertains in vivo. One limitation of our artificial stenosis model is that it comprises rigid sections of PEEK tubing, lacking the compliancy intrinsic to a blood vessel. Flow of blood through this model stenosis is steady, as opposed to the pulsatile flow in a blood vessel. Thus, our model may not accurately depict the pathophysiological shear stresses in a real stenosis in the human body. Another limitation is that the contribution of the coagulation cascade to platelet activation has effectively been eliminated in our system through use of the thrombin inhibitor PPACK as the anticoagulant. Anti-thrombotic factors typically released from intact endothelial cells and pro-thrombotic factors in the subendothelial matrix are also absent from our model. Furthermore, physically damaged platelets may lead to microparticle shedding and platelet activation. Future studies measuring direct markers of platelet activation with CD62p expression might be considered [36 –39]. Nonetheless, the use of [¹⁴C] serotonin provides an excellent objective measure of platelet activation and the short duration of exposure to high shear stress correlates well with the exposure time in vivo in an arterial stenosis or an artificial heart valve.

5 Conclusion

This is the initial characterization of a new method to assess platelet activation and erythrocyte lysis during brief exposure to pathophysiological shear stress. Our findings indicate that the severity of stenosis is the most critical factor in determining platelet activation under certain conditions. Our experimental approach could prove to be a valuable aid in assessing thrombotic risk during the design of mechanical heart valves.

Sources of funding

This work was generously supported by Research Grant 0443903 (to G.W.B. and J.C.K.) from the National Science Foundation.

Declaration of interest

None of the authors have any conflict of interest to disclose.

Footnotes

Acknowledgments

We are greatly indebted to Joe Ed Smith for constructing a water jacket to maintain blood at 37°C during perfusion. We are also grateful to the volunteer donors who provided blood samples for these studies.

References

Furie

and Furie

B.C.

, In vivo thrombus formation, Journal of Thrombosis and Haemostasis: JTH 5(Suppl 1) (2007), 12–17. PubMed PMID: 17635703.

Jackson

S.P.

, The growing complexity of platelet aggregation, Blood 109(12) (2007), 5087–5095. PubMed PMID: 17311994.

Kroll

M.H.

, Hellums

J.D.

, McIntire

L.V.

, Schafer

A.I.

and Moake

J.L.

, Platelets and shear stress, Blood 88(5) (1996), 1525–1541. PubMed PMID: 8781407.

Ikeda

, Handa

, Kawano

, Kamata

, Murata

, Araki

, et al., The role of von Willebrand factor and fibrinogen in platelet aggregation under varying shear stress, The Journal of Clinical Investigation 87(4) (1991), 1234–1240. PubMed PMID: 2010539. Pubmed Central PMCID: 295144.

Jackson

S.P.

, Nesbitt

W.S.

and Kulkarni

, Signaling events underlying thrombus formation, Journal of Thrombosis and Haemostasis: JTH 1(7) (2003), 1602–1612. PubMed PMID: 12871297.

Mazzucato

, Pradella

, Cozzi

M.R.

, De Marco

and Ruggeri

Z.M.

, Sequential cytoplasmic calcium signals in a 2-stage platelet activation process induced by the glycoprotein Ibalpha mechanoreceptor, Blood 100(8) (2002), 2793–2800. PubMed PMID: 12351387.

Chow

T.W.

, Hellums

J.D.

, Moake

J.L.

and Kroll

M.H.

, Shear stress-induced von Willebrand factor binding to platelet glycoprotein Ib initiates calcium influx associated with aggregation, Blood 80(1) (1992), 113–120. PubMed PMID: 1611079.

Ikeda

, Handa

, Kamata

, Kawano

, Kawai

, Watanabe

, et al., Transmembrane calcium influx associated with von Willebrand factor binding to GP Ib in the initiation of shear-induced platelet aggregation, Thrombosis and Haemostasis 69(5) (1993), 496–502. PubMed PMID: 1611079.

Zhang

J.N.

, Bergeron

A.L.

, Yu

, Sun

, McBride

, Bray

P.F.

, et al., Duration of exposure to high fluid shear stress is critical in shear-induced platelet activation-aggregation, Thrombosis and haemostasis 90(4) (2003), 672–678. PubMed PMID: 14515188.

10.

Cannegieter

S.C.

, Rosendaal

F.R.

and Briet

, Thromboembolic and bleeding complications in patients with mechanical heart valve prostheses, Circulation 89(2) (1994), 635–641. PubMed PMID: 8313552.

11.

Fallon

A.M.

, Shah

, Marzec

U.M.

, Warnock

J.N.

, Yoganathan

A.P.

and Hanson

S.R.

, Flow and thrombosis at orifices simulating mechanical heart valve leakage regions, Journal of Biomechanical Engineering 128(1) (2006), 30–39. PubMed PMID: 16532615.

12.

Alemu

and Bluestein

, Flow-induced platelet activation and damage accumulation in a mechanical heart valve: Numerical studies, Artificial Organs 31(9) (2007), 677–688. PubMed PMID: 17725695.

13.

Krishnan

, Udaykumar

H.S.

, Marshall

J.S.

and Chandran

K.B.

, Two-dimensional dynamic simulation of platelet activation during mechanical heart valve closure, Annals of Biomedical Engineering 34(10) (2006), 1519–1534. PubMed PMID: 17013660.

14.

Leo

H.L.

, He

, Ellis

J.T.

and Yoganathan

A.P.

, Microflow fields in the hinge region of the CarboMedics bileaflet mechanical heart valve design, The Journal of Thoracic and Cardiovascular Surgery 124(3) (2002), 561–574. PubMed PMID: 12202873.

15.

Reinthaler

, Braune

, Lendlein

, Landmesser

and JungPlatelets

, and coronary artery disease: Interactions with the bloodvessel wall and cardiovascular devices, Biointerphases 11(2) (2016), 029702. PubMed PMID: 27277200.

16.

Lyon

M.E.

, Fine

J.S.

, Henderson

P.J.

and Lyon

A.W.

, D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone (PPACK): Alternative anticoagulant to heparin salts for blood gas and electrolyte specimens, Clinical Chemistry 41(7) (1995), 1038–1041. PubMed PMID: 7600685.

17.

Holmsen

and Dangelmaier

C.A.

, Measurement of secretion of serotonin, Methods in Enzymology 169 (1989), 205–210. PubMed PMID: 2654560.

18.

Frantzen

, Grimsrud

, Heggli

D.E.

and Sundrehagen

, Selective precipitation of human hemoglobin by organic solvents and metal cations, Hemoglobin 21(2) (1997), 155–172. PubMed PMID: 9101284.

19.

Merrick

M.F.

and Pardue

H.L.

, Evaluation of absorption and first-and second-derivative spectra for simultaneous quantification ofbilirubin and hemoglobin, Clinical Chemistry 32(4) (1986), 598–602. PubMed PMID: 3955808.

20.

Mendolicchio

G.L.

and Ruggeri

Z.M.

, New perspectives on von Willebrand factor functions in hemostasis and thrombosis, Seminars in Hematology 42(1) (2005), 5–14. PubMed PMID: 15662611.

21.

Zhang

J.N.

, Bergeron

A.L.

, Yu

, Sun

, McIntire

L.V.

, Lopez

J.A.

, et al., Platelet aggregation and activation under complex patterns of shear stress, Thrombosis and Haemostasis 88(5) (2002), 817–821. PubMed PMID: 12428100.

22.

Alevriadou

B.R.

, Moake

J.L.

, Turner

N.A.

, Ruggeri

Z.M.

, Folie

B.J.

, Phillips

M.D.

, et al., Real-time analysis of shear-dependent thrombus formation and its blockade by inhibitors of von Willebrand factor binding to platelets, Blood 81(5) (1993), 1263–1276. PubMed PMID: 8443388.

23.

Fressinaud

, Baruch

, Rothschild

, Baumgartner

H.R.

and Meyer

, Platelet von Willebrand factor: Evidence for its involvement in platelet adhesion to collagen, Blood 70(4) (1987), 1214–1217. PubMed PMID: 3498520.

24.

Sakariassen

K.S.

, Aarts

P.A.

, de Groot

P.G.

, Houdijk

W.P.

and Sixma

J.J.

, A perfusion chamber developed to investigate platelet interaction in flowing blood with human vessel wall cells, their extracellular matrix, and purified components, The Journal of Laboratory and Clinical Medicine 102(4) (1983), 522–535. PubMed PMID: 6619647.

25.

Savage

, Saldivar

and Ruggeri

Z.M.

, Initiation of platelet adhesion by arrest onto fibrinogen or translocation on von Willebrand factor, Cell 84(2) (1996), 289–297. PubMed PMID: 8565074.

26.

Dopheide

S.M.

, Maxwell

M.J.

and Jackson

S.P.

, Shear-dependent tether formation during platelet translocation on von Willebrand factor, Blood 99(1) (2002), 159–167. PubMed PMID: 11756166.

27.

Maxwell

M.J.

, Westein

, Nesbitt

W.S.

, Giuliano

, Dopheide

S.M.

and Jackson

S.P.

, Identification of a 2-stage platelet aggregation process mediating shear-dependent thrombus formation, Blood 109(2) (2007), 566–576. PubMed PMID: 16990596.

28.

Wang

, Xu

, Zhang

, Kwok

D.T.

, Jiang

, Wu

, et al., Mechanical and biological characteristics of diamond-like carbon coated poly aryl-ether-ether-ketone, Biomaterials 31(32) (2010), 8181–8187. PubMed PMID: 20692699.

29.

Ellis

J.T.

, Healy

T.M.

, Fontaine

A.A.

, Saxena

and Yoganathan

A.P.

, Velocity measurements and flow patterns within the hinge region of a Medtronic Parallel bileaflet mechanical valve with clear housing, The Journal of Heart Valve Disease 5(6) (1996), 591–599. PubMed PMID: 8953436.

30.

Bluestein

, Niu

, Schoephoerster

R.T.

and Dewanjee

M.K.

, Fluid mechanics of arterial stenosis: Relationship to the development of mural thrombus, Annals of Biomedical Engineering 25(2) (1997), 344–356. PubMed PMID: 9084839.

31.

Turitto

V.T.

and Hall

C.L.

, Mechanical factors affecting hemostasis and thrombosis, Thrombosis Research 92(6 Suppl 2) (1998), S25–S31. PubMed PMID: 9886907.

32.

Zwaginga

J.J.

, Nash

, King

M.R.

, Heemskerk

J.W.

, Frojmovic

, Hoylaerts

M.F.

, et al., Flow-based assays for global assessment of hemostasis. Part 1: Biorheologic considerations, Journal of Thrombosis and Haemostasis: JTH 4(11) (2006), 2486–2487. PubMed PMID: 16938127.

33.

Ruggeri

Z.M.

, Orje

J.N.

, Habermann

, Federici

A.B.

and Reininger

A.J.

, Activation-independent platelet adhesion and aggregation under elevated shear stress, Blood 108(6) (2006), 1903–1910. PubMed PMID: 16772609. Pubmed Central PMCID: 1895550.

34.

Slack

S.M.

, Cui

and Turitto

V.T.

, The effects of flow on blood coagulation and thrombosis, Thrombosis and Haemostasis 70(1) (1993), 129–134. PubMed PMID: 8236088.

35.

Braune

, Gross

, Walter

, Zhou

, Dietze

, Rutschow

, et al., Adhesion and activation of platelets from subjects with coronary artery disease and apparently healthy individuals on biomaterials, Journal of Biomedical Materials Research Part B, Applied Biomaterials 104(1) (2016), 210–217. PubMed PMID: 25631281.

36.

Klaus

, Korfer

, Mottaghy

, Reul

and Glasmacher

, In vitro blood damage by high shear flow: Human versus porcine blood, The International Journal of Artificial Organs 25(4) (2002), 306–312. PubMed PMID: 12027141.

37.

Leytin

, Allen

D.J.

, Gwozdz

, Garvey

and Freedman

, Role of platelet surface glycoprotein Ibalpha and P-selectin in the clearance of transfused platelet concentrates, Transfusion 44(10) (2004), 1487–1495. PubMed PMID: 15383023.

38.

O’Brien

J.R.

, Etherington

M.D.

, Brant

and Watkins

, Decreased platelet function in aortic valve stenosis: High shear platelet activation then inactivation, British Heart Journal 74(6) (1995), 641–644. PubMed PMID: 484121.

39.

Reininger

A.J.

, Platelet function under high shear conditions, Hamostaseologie 29(1) (2009), 21–22, 4. PubMed PMID: 19151841.

Platelet activation and erythrocyte lysis during brief exposure of blood to pathophysiological shear stress in vitro

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

Abbreviations

1 Introduction

2 Materials and methods

2.1 Collection of human blood, loading of platelets with [14C] serotonin, and reconstitution of blood

2.2 Artificial stenosis model for assessment of platelet activation under high shear stress

2.4 Measurement of pressure differential across the artificial stenosis

2.5 Statistical analysis

3 Results

3.1 Blood flow through the artificial stenosis is consistent with predictions for laminar flow

5 Conclusion

Sources of funding

Declaration of interest

Footnotes

Acknowledgments

References

2.1 Collection of human blood, loading of platelets with [¹⁴C] serotonin, and reconstitution of blood