Evaluation of platelet function under high shear condition in the small-sized collagen bead column

ORIGINAL ARTICLES Evaluation of platelet function under high shear condition in the small-sized collagen bead column MAKOTO KANEKO, TOSHIRO TAKAFUTA, OLGA CUYUN-LIRA, KANEO SATOH, MORIO ARAI, YUTAKA YATOMI, and YUKIO OZAKI YAMANASHI, KOBE, and TOKYO, JAPAN

We previously reported that platelet retention rates as measured with collagen-coated bead columns (the conventional column) reflect the processes of platelet adhesion and aggregation under low shear stress, and that this system could serve as an easy-to-use platelet aggregometry. With this column, platelet glycoprotein (GP) VI and GPIIb/IIIa, but not the GPIb–von Willebrand factor (VWF) interaction, play major roles in platelet activation. To develop a system that can better reflect the GPIb-VWF interaction under high shear stress, we designed a column containing small-sized beads (125–212 ␮m) coated with porcine collagen type I. As expected, the GPIb-VWF interaction played a crucial role in platelet retention rates at higher flow rates. Adenosine 5=-diphosphate, but not thromboxane A2, appears to support platelet activation in this system. The platelet retention rates among healthy individuals with the new columns are in the range wider than the conventional columns, and this diversity could be attributed to the broad range of the VWF antigen and/or its activity. It is suggested that this new column can serve as an easy-to-use method for evaluating the VWF antigen levels and its activity and for monitoring patients with thrombotic or bleeding disorders related to the VWF-GPIb interaction. (J Lab Clin Med 2005;146:64 –75) Abbreviations: A3P5P ⫽ adenosine 3= 5=-diphosphate; ADP ⫽ adenosine 5=-diphosphate; ARC69931MX ⫽ N6-(2-methylthioethyl)-2-(33,3-trifluoropropylthio)-␤,␥-dichloromethylene adenosine triphosphate; ASA ⫽ acetylsalicylic acid; CP ⫽ creatine phosphate; CPK ⫽ creatine phosphokinase; EDTA ⫽ ethylenediamine tetraacetic acid; GP ⫽ glycoprotein; MoAb ⫽ monoclonal antibody; PPP ⫽ platelet-poor plasma; PRP ⫽ platelet-rich plasma; TTP ⫽ thrombotic thrombocytopenic purpura; TXA2 ⫽ thromboxane A2; VWD ⫽ von Willebrand disease; VWF ⫽ von Willebrand factor; VWF:Ag ⫽ VWF:Antigen; VWF:RCo ⫽ VWF:Ristocetin Cofactor activity

B

lood flow through arteries often results in high shear rates. Under such hemodynamic conditions, platelets initially adhere to the subendothelium at sites of vascular injury through the platelet

From the Department of Clinical and Laboratory Medicine, Faculty of Medicine, University of Yamanashi, Yamanashi, Japan; Department of Hematology and Clinical Immunology, Nishi-Kobe Medical Center, Kobe, Japan; Department of Laboratory Medicine, Tokyo Medical University, Tokyo, Japan; and Department of Clinical Laboratory Medicine, Graduate School of Medicine, University of Tokyo, Tokyo, Japan. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan, by Health and Labour Sciences Research Grants H16-Iyaku-073, and by Charitable Trust Clinical Pathology Research Foundation of Japan and Kurozumi Medical Foundation.

64

membrane receptor glycoprotein Ib (GPIb)-von Willebrand factor (VWF) interaction. This reversible interaction leads to the conformational activation of integrin ␣IIb␤3 (GPIIb/IIIa) followed by platelet activation, which finally culminates in aggregate formation. Thus, the interaction between VWF and platelets is an imporReceived for publication December 2, 2004; accepted for publication April 4, 2005. Reprint requests: Yukio Ozaki MD, PhD, Department of Clinical and Laboratory Medicine, Yamanashi Medical University, 1110 Shimokato, Tamaho, Nakakoma, Yamanashi 409-3898, Japan; e-mail: [email protected] 0022-2143/$ – see front matter © 2005 Mosby, Inc. All rights reserved. doi:10.1016/j.lab.2005.04.007

J Lab Clin Med Volume 146, Number 2

tant step in primary hemostasis, and moreover, it can also contribute to the generation of pathologic thrombi at sites of atherosclerotic plaque rupture under high shear stress. The importance of VWF-platelet interaction in thrombus formation implies that the level of plasma VWF or VWF activity enhances platelet aggregate formation. A number of epidemiologic studies have shown a positive correlation between plasma VWF levels and the incidence of acute coronary syndromes.1–5 High levels of VWF are also related with cerebral vascular accidents.6 – 8 The pathogenesis of thrombotic thrombocytopenic purpura (TTP), which remained obscure until recently, is now suggested to be a dysfunction of ADAMTS13,9,10 a VWF cleaving metalloprotease with resultant increase in large VWF multimers. It is suggested that excessive adhesive activity of large VWF multimers results in platelet aggregate formation and thrombus formation in this disorder. The detection of large VWF multimer or enhanced platelet-VWF interaction at the early stage will provide useful information for the prevention of overt and/or full-blown symptoms. Thus, for the purpose of diagnosing TTP at the early stage or for monitoring patients with thrombotic disorders, it is important to develop appropriate methods for evaluating platelet functions related to VWF under various shear conditions. Among many methods for the evaluation of platelet functions, platelet adhesion and aggregation tests have been most frequently used.11,12 They have met with limited approval in clinical settings, because they have several shortcomings including relative insensitivity to detect weak platelet activation, lack of reproducibility or clinical relevance, and tedious procedures for measurements. To overcome these hurdles, a number of platelet function tests with improved technology and novel approaches to analysis have been developed recently. Some can only measure platelet function in the absence of red blood cells or white blood cells, which may also contribute to thrombus formation in vivo. Others only measure platelet activation under shear stress of a low degree. More recently, new instruments have been developed that can evaluate platelet function using whole-blood samples under high shear stress.13–18 However, specialized and expensive instruments are required for these methods, and thus these methods may not be suitable for screening tests in clinical settings in ordinary facilities. The glass bead column system was developed by Salzman, Hellem, and others, and has been modified to assess platelet adherent function.19 –21 However, because the interaction between platelets and the glass surface may not accurately represent physiologic platelet function in vivo and not only platelet adhesion but also platelet aggrega-

Kaneko et al

65

tion occurs in the glass bead column, this method has not seen wide clinical application. Recently, copolymer plastic sphere beads coated with porcine type I collagen have been developed to replace glass beads, and this collagencoated bead column (conventional column) is currently used for the measurement of platelet retention.22 We investigated the factors responsible for platelet retention and characterized this as an easy-to-use platelet function test. Our findings in conventional columns suggest that GPVI and GPIIb/IIIa are mainly involved in platelet retention, and that the results mainly reflect platelet aggregate formation at relatively lower shear rates. We have suggested that this can be a simple and reliable method for evaluating platelet aggregation using whole blood as samples. To develop a new aggregometer that can evaluate platelet activation related to VWF under high shear stress, we designed a new column that contains smallsized collagen beads. In the present study, we investigated the factors responsible for platelet retention in a small-sized collagen bead column. Our findings suggest that the level of platelet retention in this new type of column reflects the GPIb-VWF interaction and that it positively correlates with the plasma VWF activity. We suggest that this column, using whole blood as samples, can be a simple and reliable method for the evaluation of platelet function under high shear conditions. METHODS Materials. AJvW-2,2 an anti-VWF monoclonal antibody (MoAb), was a gift from Pharmaceutical Research Laboratories (Ajinomoto Co, Kawasaki, Japan). Dr. Jan J. Sixma (Thrombosis and Hemostasis Laboratory, Department of Hematology, University Medical Centre Utrecht, Netherlands) provided us with RU5,23 an anti-VWF MoAb that blocks the interaction between VWF and collagen. Confact F, a human factor VIII/VWF concentrate, was provided by Kaketsuken (Kumamoto, Japan). The following materials were obtained from the indicated suppliers: abciximab (ReoProTM) (Eli Lilly & Co. Ltd., Indianapolis, Ind); Asserachrom VWF (commercial kit for assay of factor VIII/VWF antigen) (Roche, Tokyo, Japan); peroxidase-conjugated rabbit polyclonal anti-human VWF Ab (Dako Cytomation Co. Ltd., Glostrup, Denmark); ristocetin (Sigma, St. Louis, Mo); and SeaKem HGT (P) Agarose (Takara Bio Inc., Otsu, Japan). Blood collection. This study was carried out according to the principles of the Declaration of Helsinki, and informed consent was obtained from all blood donors. Furthermore, the study was approved by the review board of the University of Yamanashi. Healthy drug-free volunteers and a patient with type 3 von Willebrand disease (VWD) served as donors, and blood withdrawn by venipuncture was mixed with 3.13% citrate (9:1, v/v) to make whole-blood samples as described previously.22 Platelet retention tests. Small-sized collagen bead column (PLA BEADS COLUMN-PAG, diameter 0.1– 0.2 mm) and conventional column (PLA BEADS COLUMN, diameter

66

Kaneko et al

J Lab Clin Med August 2005

Fig 1. Effects of monoclonal antibodies (MoAbs) (affecting platelet adhesion or aggregation) on platelet retention. Inhibitory effects of glycoprotein (GP)Ib-von Willebrand factor (VWF)-collagen interaction (AJvW-2 and RU5) and GPIIb/IIIa (abciximab) were evaluated on the platelet retention rates in the small-sized collagen bead column (A) and the conventional column (B), and in the small-sized collagen bead column at various flow rates (C). (A, B) The flow rates in the small-sized collagen bead column and conventional column were 2.25 mL/min and 0.75 mL/min, respectively. Whole-blood samples from healthy donors were incubated with 10 ␮g/mL of AJvW-2 (anti-VWF antibody), 10 ␮g/mL of RU5 (anti-VWF antibody), or 5 ␮g/mL of abciximab (anti-GPIIb/IIIa antibody) for 5 minutes, and platelet retention was determined as described in the Methods section. The results are expressed as percentages to the control value without pretreatment. Mean ⫾ standard deviation (SD) (columns and error bars); numbers of experiments performed (parentheses). (C) Wholeblood samples from healthy donors were preincubated with or without 10 ␮g/mL of AJvW-2 (anti-VWF antibody, closed boxes), 10 ␮g/mL of RU5 (anti-VWF antibody, open boxes), or 5 ␮g/mL of abciximab (anti-GPIIb/IIIa antibody, open circles) for 5 minutes, and platelet retention rates in the small-sized collagen bead column were determined. The results are expressed as percent inhibition compared with the control sample without antibody pretreatment. The data are compiled from 6 to 10 experiments for each antibody, and expressed as the mean ⫾ SD. *Significant differences between the data (t test, P ⬍ .01). (D) Platelet retention rates were examined in a patient with type 3 von Willebrand disease (VWD) by using the small-sized collagen bead column. Each experiment was performed in duplicate at each flow rate. The average of platelet retention rates (open columns).

0.4 – 0.6 mm) comprise spherical copolymer-plastic beads coated with porcine type I collagen, packed into polyvinyl tubing (with an internal diameter of 2 mm). Both types of columns are commercially available from ISK Co. Ltd. (Tokyo, Japan). The flow rate and direction of blood samples were controlled by a syringe pump (“Infusion/Withdrawal” Model 777; ISK Co. Ltd.). After a 5-minute incubation at 37°C, the whole-blood sample in the plastic tube was mixed by swirling and 1.5 mL of the sample was drawn into a 2.5-mL plastic syringe (Terumo, Tokyo, Japan). The syringe and column were set in the holders of the syringe pump, and samples (in the syringe) were passed through the small-sized collagen bead columns at a flow rate of 2.25 mL/min for 40 seconds; this flow rate was determined by the results shown in Fig 1C. Under this flow rate, the level of platelet retention almost equally represents platelet aggregate formation because of the GPIb-VWF interaction and subsequent activation of GPIIb/IIIa. The measurement procedure for the conventional column was described previously.22 All of the blood samples before and after passage through columns were collected into plastic tubes containing EDTA (Terumo, Tokyo, Japan), and platelet counts were measured by an NE-8000 automatic blood cell counter (Sysmex, Kobe, Japan). Platelet retention rates (%) were calculated as 100 ⫻ [(platelet count before the passage) ⫺ (platelet count after the passage)]/ (platelet count before the passage). von Willebrand factor:Ristocetin cofactor activity. The VWF:RCo was determined with a platelet aggregometer us-

J Lab Clin Med Volume 146, Number 2

ing washed human platelets according to the method of Weiss et al24 with slight modifications. Preparation of washed platelets. Platelet-rich plasma was obtained after centrifugation of whole blood at 150g for 12 minutes. Platelet-rich plasma was incubated with 1 mM acetylsalicylic acid (ASA) for 30 minutes to exclude the secondary effects of thromboxane A2 (TXA2). The platelets were washed twice as previously described25 and resuspended in a HEPES buffer at the final concentration of 2.5 ⫻ 108 cells/mL. Ristocetin-induced agglutination assay. Platelet-poor plasma (PPP) was obtained from whole-blood samples prepared by centrifugation at 1500g for 10 minutes, and PPP was stored at ⫺80°C until measurement. PPP was diluted with two volumes of HEPES buffer, and 100 ␮L of diluted plasma was added to an aggregometer cuvette that contained 200 ␮L of washed platelet suspension or HEPES buffer (as a control). Then ristocetin at a final concentration of 1.2 mg/mL was added, and the magnitude of platelet agglutination was measured at 1 minute after stimulation with a platelet aggregometer (PA-100, Kowa, Tokyo, Japan) with constant stirring. PPP from more than five healthy individuals was mixed, and the mixture was diluted with HEPES buffer at various ratios (1/2, 1/4, 1/8, 1/16, 1/32; v/v). The dilutants were also added to an aggregometry cuvette containing platelets, and the magnitude of platelet agglutination was measured as described to obtain the standard curve. The VWF activity of blood samples, that is, the activity of plasma to induce platelet agglutination, was compared with the standard curve of normal plasma and expressed as a percentage of the activity to that of normal plasma. von Willebrand factor:Antigen– enzyme-linked immunosorbent assay. VWF:Antigen (VWF:Ag) was assayed

with an Asserachrom VWF kit. von Willebrand factor multimeric analysis. VWF multimers were analyzed by horizontal electrophoresis in sodium dodecyl sulfate (SDS) and 1.4% agarose gels according to the method of Ruggeri and Zimmerman.26 After electrophoresis, VWF multimers were electroblotted, probed with peroxidaseconjugated rabbit polyclonal anti-human VWF antibody, and detected with a chemiluminescence reaction reagent (ECL, Amersham Biosciences Corp., Piscataway, NJ) and Konica X-ray film (Konica, Tokyo, Japan). Purification of von Willebrand factor. VWF was purified and fractionated by gel filtration from Confact F, human factor VIII/VWF concentrates, on a 5 ⫻ 100-cm column of Sepharose CL-4B (Amersham Biosciences Corp.) as described previously.27 The column was packed at room temperature with Sepharose CL-4B and was equilibrated with 20 mM imidazole buffer (pH 6.5), containing 20 mM ⑀-aminocaproic acid, 0.15 M NaCl, 10 mM Na3-citrate, and 0.05% NaN3. Then, 1500 units of Confact F was dissolved in 15 mL of the imidazole buffer and layered onto the column. The column was eluted at room temperature with the imidazole buffer at a flow rate of 1 mL/min using a peristaltic pump. The absorbance of the eluates at 280 nm was record continuously, and a 10-mL aliquot of the eluted VWF was fractionated into glass tubes. Each fraction was concentrated by

Kaneko et al

67

ultrafiltration with an Aquacide II (Calbiochem, San Diego, Calif) and then dialyzed in phosphate-buffered saline. Statistical analyses. When indicated, statistical analysis was performed by the paired t test. Multivariate analysis to define factors correlated with platelet retention rates was performed by the stepwise regression method. Briefly, we picked some possible explanatory variables, and multiple correlation coefficients were calculated using a stepwise regression process. The process would stop when the elimination of a variable would make no significant improvement in the amount of variation explained. RESULTS Platelet retention in the small-sized collagen bead column reflects platelet activation under high shear stress. We

reported in a previous article that platelet retention in the conventional column mainly involves GPVI and fibrinogen-GPIIb/IIIa interactions and does not reflect platelet activation related to the GPIb-VWF interaction.22 We then sought to develop a system in which the GPIb-VWF interaction plays a more important role than in the conventional column. Because small fluidflow space can produce higher shear stress under the same flow velocity in laminar flow, we designed a new column containing small-sized collagen beads.28 We first sought to determine whether higher shear stress could be generated within this column, judged by the inhibitory effects of several MoAbs. One antibody we used is AJvW-2, which recognizes the A1 domain of VWF. It effectively blocks high shear-induced platelet aggregation, but not aggregation induced under low shear conditions.2 Another is RU5, a MoAb that blocks the interaction of VWF with collagen by binding to the A3 domain of VWF.23 We also used abciximab, which blocks the interaction of GPIIb/IIIa and fibrinogen. Mouse immunoglobulin-G used as the control had no effects on platelet retention rates (data not shown). AJvW-2 and RU5 almost completely inhibited platelet retention rates in the small-sized collagen bead column (Fig 1A), whereas it had virtually no inhibitory effects in the conventional column (Fig 1B). Because the collagen-VWF interaction with subsequent binding of GPIb to VWF also plays an essential role in platelet activation under high shear stress but not under low shear stress, this finding suggests that shear stress at a level high enough to induce the GPIb-VWF interaction is generated within the small-sized collagen bead column. On the basis of a previous report using AJvW-2,2 it is estimated that shear stress equivalent to 108 dyn/ cm2 can be generated in this system. The inhibitory effects of abciximab also suggest that the interaction of GPIIb/IIIa with VWF or fibrinogen also takes place in the small-sized collagen bead column. We then evaluated the effects of these MoAbs on the

68

Kaneko et al

J Lab Clin Med August 2005

Fig 2. Effects of antiplatelet agents on platelet retention in the small-sized collagen bead column (A) and the conventional column (B), and effects of adenosine 5=-diphosphate (ADP) receptor antagonists on platelet retention in small-sized collagen bead columns (C). Whole-blood samples from healthy donors were incubated with 3 mM creatine phosphate (CP) plus 40 U/mL creatine phosphokinase (CPK) for 5 minutes, and/or 100 ␮M acetylsalicylate (ASA) for 30 minutes (A, B), or incubated with 1 mM A3P5P (a P2Y1 antagonist), 1 ␮M AR-C69931MX (a P2Y12 antagonist), or A3P5P plus ARC69931MX for 5 minutes (C), after which platelet retention in the column was determined. The results are expressed as percentages of the control value without pretreatment. Mean ⫾ SD (columns and error bars); numbers of experiments performed (parentheses). *Statistically significant differences (t test, P ⬍ .01).

platelet retention rates in the small-sized collagen bead columns at various flow rates (Fig 1C). With higher flow rates, the inhibitory effects of both AJvW-2 and RU5 were intensified. In contrast, that of abciximab decreased with higher flow rates. These findings suggest that, at higher flow rates with a proportional in-

crease in shear stress, platelet activation mediated by the GPIb/VWF interaction plays a role greater than that of GPIIb/IIIa in inducing platelet retention in the smallsized collagen bead column. On the other hand, GPIIb/ IIIa-mediated platelet activation appears to play a greater role in platelet retention under lower shear stress. We then evaluated platelet retention rates of blood samples from a patient with type 3 VWD who completely lacks plasma VWF. At lower flow rates, there was a considerable degree of platelet retention, whereas at higher flow rates there was virtually no platelet retention (Fig 1D). These findings confirm the notion of collagen-induced platelet activation with GPIIb/IIIarelated aggregate formation under low shear stress, and that the GPIb/VWF interaction is a prerequisite in platelet activation under high shear stress. These findings taken together suggest that the GPIb-VWF interaction plays a critical role in inducing platelet retention in this new column, and that shear stress at a level high enough to elicit the GPIb/VWF interaction is generated within this column. On the basis of the results illustrated in Fig 1C, the inhibitory effects of anti-GPIb MoAbs and anti-GPIIb/ IIIa MoAb are in the range of 70% to 90% at approximate flow rates of 2.25 mL/min, and the platelet retention rates under this condition well represent the process of platelet activation mediated by GPIb with subsequent activation of GPIIb/IIIa. We thus set the flow rate within the small-sized collagen bead column to 2.25 mL/min in experiments thereafter. Effects of antiplatelet agents on platelet retention. Shearinduced platelet aggregation, as measured by a coneplate aggregometer, is known to require adenosine 5=diphosphate (ADP) in addition to the involvement of the GPIb-VWF interaction, but is relatively independent of TXA2.29 To evaluate the role of ADP and TXA2, whole blood was preincubated with the ADP scavenger creatine phosphate/creatine phosphokinase,

J Lab Clin Med Volume 146, Number 2

Kaneko et al

69

or ASA, which inhibits TXA2 production by inactivating cyclooxygenase, and the platelet retention rate in the small-sized collagen bead column (Fig 2A) or the conventional column (Fig 2B) was measured. Creatine phosphate/creatine phosphokinase blocked the platelet retention by approximately 63%, suggesting that ADP seems to play vital roles in platelet activation (retention) in the small-sized collagen bead column (Fig 2A). In contrast with ADP, ASA had virtually no inhibitory effect. ONO 3708, a TXA2 receptor antagonist,30 also plays a negligible role similar to ASA, confirming the minor role of TXA2 in this system (data not shown). Platelet activation induced by ADP is mainly mediated by two G protein-coupled purinergic receptors (ie, P2Y1 and P2Y12).31 Involvement of these receptors was examined using specific antagonists. Adenosine 3=, 5=diphosphate (A3P5P), a selective P2Y1 antagonist, inhibited the platelet retention by 49%, and ARC69931MX, a selective P2Y12 antagonist, inhibited the platelet retention by 43% (Fig 2C). The combination of these two agents further increased the inhibition rate to 69% (Fig 2C), indicating a synergistic effect of P2Y1 and P2Y12 receptors on the involvement of ADP in enhancing platelet retention in the small-sized collagen bead columns. Comparison of platelet retention rates of healthy individuals between the conventional column and the small-

Platelet retention rates of 73 and 100 healthy donors were measured by the conventional column and the small-sized collagen bead column, respectively (Fig 3A). The mean values ⫾ standard deviation (SD) of platelet retention rates were 64% ⫾ 13% and 42% ⫾ 20%, respectively. In contrast with the large variation of platelet retention rates among healthy individuals with the small-sized collagen bead column, the reproducibility was satisfactory for the same individuals (Fig 3B), suggesting that the large variation represents individual differences of some nature. Because the GPIb-VWF interaction plays a critical role in the small-sized collagen bead column and the normal range of the VWF antigen is broad,32 we hypothesized that platelet retention rates assessed with this column are influenced by the levels of the VWF antigen and/or its activity in plasma. sized collagen bead column.

Effects of purified von Willebrand factor on platelet re-

We then examined the effect of purified VWF on platelet retention rates in this system. Whole-blood samples from healthy donors and a patient with type 3 VWD were incubated with purified VWF or with the same volume of vehicle (saline), and platelet retention rates were measured. Exogenous VWF increased the platelet retention rates, and the potentiating effects were more

tention in the small-sized collagen bead column.

Fig 3. (A) Comparison of platelet retention rates between the conventional column and the small-sized collagen bead column. Platelet retention rates of 73 and 100 healthy donors were measured by the conventional column and the small-sized collagen bead column, respectively. The average of duplicate measurements with each individual (dots); mean ⫾ SD of all the measurements (open circles and error bars). (B) Mean ⫾ SD of platelet retention rates of individuals whose whole-blood samples were assessed with the small-sized collagen bead column more than three times.

marked with individuals with lower retention rates, including the patient with type 3 VWD (Fig 4A). Effects of VWF multimers with different molecular sizes on platelet retention in the small-sized collagen bead column. The overall VWF function/activity in terms of

the ability to interact with platelet GPIb is determined by its multimer size and the VWF antigen level of plasma. Larger multimers have a more potent biologic activity33,34 and bind more efficiently to GPIb than small multimers. Thus, we sought to evaluate the effects of VWF multimer sizes in this system. VWF was purified and fractionated, and each fraction was analyzed by SDS–agarose gel electrophoresis as described in the Methods section (Fig 4B). Fractions 1 and 2 contained multimers larger than fractions 3 or 4. An aliquot of each fraction was added at the final concen-

70

Kaneko et al

J Lab Clin Med August 2005

Fig 4. Effects of purified VWF or VWF multimers with different molecular sizes on platelet retention in the small-sized collagen bead column. (A) Whole-blood samples from healthy donors (open columns, mean ⫾ SD of three experiments) and a patient with type 3 VWD (closed circles, average of two experiments) were incubated with or without purified VWF (final concentration of 4 or 20 ␮g/mL) for 5 minutes, after which platelet retention in the small-sized collagen bead column was determined. *Statistically significant differences (t test, P ⬍ .01) when comparison was made between whole-blood samples with and without VWF. (B) VWF was purified and fractionated by gel filtration from human factor VIII/VWF concentrates, and each fraction was analyzed by sodium dodecyl sulfate (SDS) and 1.4% agarose gel electrophoresis. (C) An aliquot of each fraction was added to blood samples from two healthy donors (open and shaded columns, average of duplicate experiments) or a patient with type 3 VWD (closed columns, average of two experiments) at the final concentration of 3 ␮g/mL of purified VWF. The original retention rates of two healthy donors were 31% and 29%, respectively. Five minutes after the addition of VWF, platelet retention in the small-sized collagen bead column was determined. The results are expressed as percentages of control samples from the same donor. All the results were significantly different (t test, P ⬍ .05) compared with control samples except for the closed column of lane 4. (*Differences with P ⬍ .01.)

tration of 3 ␮g/mL of VWF to blood samples from two healthy donors, whose retention rates were approximately 30% or a patient with type 3 VWD, and platelet retention rates in the small-sized collagen bead column

were determined (Fig 4C). Similar to the results of Fig 4A, exogenous VWF, especially large multimers, increased the platelet retention rate, and this increase was observed with whole-blood samples from healthy indi-

J Lab Clin Med Volume 146, Number 2

viduals and a patient with type 3 VWD. These findings strongly indicate that VWF multimer sizes also play a critical role in determining platelet retention rates within small-sized collagen bead column. Correlation between platelet retention rates and von Willebrand factor antigen/activity. Our findings thus far

suggest that the VWF antigen level and the VWF multimer size significantly affect platelet retention rates as assessed with the small-sized collagen bead column. We then sought to determine whether the plasma VWF:Ag or VWF:RCo of healthy donors correlates with platelet retention rates. VWF:RCo and VWF:Ag of 49 healthy donors were measured, and there was a good correlation between VWF:Ag and VWF:RCo (Fig 5A). Although the correlation coefficient values of VWF:Ag or VWF:RCo with the platelet retention rate were only 0.54 (P ⬍ 1 ⫻ 10⫺4) and 0.59 (P ⬍ 1 ⫻ 10⫺4), respectively (Fig 5B and C), the majority of the data seem to align with the regression lines. The ratio of the platelet retention rate to VWF:RCo (Fig 5D) clearly showed that there are six samples beyond the mean ⫹1.5 SD value. We assume that the platelet retention rates of these individuals were affected by some factors other than the VWF activity, although the factors responsible for the enhanced platelet retention with these individuals remain elusive. If these six data are excluded, the correlation coefficient value of VWF:RCo and VWF:Ag with the platelet retention rate is 0.81 (P ⬍ 1 ⫻ 10⫺4) and 0.72 (P ⬍ 1 ⫻ 10⫺4), which would suggest a good correlation between VWF function/ activity and the platelet retention rate in the small-sized collagen bead column. It is likely that various factors other than VWF also affect platelet retention rates in small-sized collagen bead columns; thrombocythemia35 often results in thrombotic accidents. It is suggested that high levels of hematocrit/hemoglobin predispose to stroke.35–37 Age is certainly one of the discrete risk factors for thrombotic disorders, which can be attributed to platelet hyperfunction.38 We used multivariate analysis to further identify the factors that determine the platelet retention rates of 49 healthy individuals assessed with small-sized collagen bead column (Table I). The factors analyzed included VWF:RCo, age, blood group, sex, neutrophil and platelet counts, and hemoglobin levels. Only VWF:RCo showed a statistically significant positive correlation with the level of platelet retention rates (P ⬍ 1 ⫻ 10⫺3). Males have a tendency to have higher levels of platelet retention rates (P ⬍ 1 ⫻ 10⫺3). Males have a tendency to have higher levels of platelet retention rates than females (P ⫽.069). The hemoglobin level gives the P value of .062, and this may explain the higher platelet retention rates among males, who ordinarily have higher levels of hemoglobin/hematocrit.

Kaneko et al

71

This may be attributed to ADP released from red blood cells, which enhances platelet retention rates in smallsized collagen bead columns. The VWF antigen level in plasma is known to be significantly lower in persons with blood type O.39,40 However, there was no correlation between platelet retention rates and blood groups (P ⫽ .53) DISCUSSION

At sites of vessel injury where high shear stress is generated, plasma VWF interacts with subendothelial matrices, the major component of which is collagen. The conformational change of VWF molecules induced not only by high shear stress but also collagen binding then leads to the interaction of platelet GPIb and VWF, resulting in platelet adhesion and aggregation.41,42 Thus, collagen-coated surfaces provide an ideal site of reaction for evaluating platelet activation under high shear stress. Because we reported in a previous article that the conventional collagen-coated bead column could serve as a new platelet aggregometer under blood-flow conditions of low shear stress, we sought to design a new column that can assess platelet activation under high shear stress. We assumed that higher shear stress could be generated by injecting blood at higher flow rates into columns packed with collagen beads of smaller size. The mode of flow within columns packed with beads is expected to be far from laminar flow, and thus the level of shear stress cannot be calculated from ordinary formulae. To confirm the generation of high shear stress, we used several MoAbs and evaluated their inhibitory effects on platelet retention rates in small-sized collagen bead columns (Fig 1A, B, and C). Under high shear stress, VWF interacts with platelet GPIb through two distinct processes. One involves the initial interaction of plasma VWF with subendothelial collagen exposed at sites of vascular injury, leading to subsequent VWF conformational changes. The other process is directly related to high shear stress, which presumably deforms plasma VWF molecules with subsequent exposure of its A1 domain. In either process, platelet GPIb then binds to the A1 domain of VWF and elicits downstream activation signals that finally lead to the formation of platelet thrombi.43 Under low shear stress, these processes play a minor role in platelet activation, and the interaction between GPIIb/IIIa and fibrinogen plays a decisive role.44,45 AJvW-2, which blocks the interaction between GPIb and VWF under high shear stress,2 and RU5, which interferes with the collagen-VWF interaction, almost completely inhibited platelet retention in the small-sized collagen bead column (Fig 1A), whereas they had virtually no effects in the conventional column (Fig 1B). Thus, it is suggested that the collagen-VWF interaction with subsequent in-

72

Kaneko et al

J Lab Clin Med August 2005

Fig 5. Correlation between platelet retention rates and VWF:RCo/ VWF:Ag. Correlation between platelet retention rates and VWF parameters such as VWF:RCo and VWF:Ag (A, B, C) were evaluated from the data of 49 healthy individuals. (A) VWF:Co versus VWF:Ag: Values of correlation coefficient were 0.879 (P ⬍ 1 ⫻ 10⫺4). (B) VWF:Ag versus retention rates: Correlation value was 0.535 (P ⬍ 1 ⫻ 1010⫺4). (C) VWF:RCo versus platelet retention: 0.591 (P ⬍ 1 ⫻ 10⫺4), respectively. Regression lines calculated from all the data (dotted lines) or recalculated from the data excluding closed circles (solid lines). The recomputed correlation value was 0.717 (P ⬍ 1 ⫻ 10⫺4) (VWF:Ag vs retention rates) and 0.809 (P ⬍ 1 ⫻ 10⫺4) (VWF:RCo vs platelet retention). (C) Individuals whose platelet retention rates were less than 20% and whose VWF:RCo levels were less than 50% (speckled circles). (D) The ratio of platelet retention rates to VWF:RCo. Data that were beyond the range of the mean ⫹1.5 SD (closed circles). Mean ⫾ SD, 1.5 SD, and 2 SD (horizontal bars on right).

teraction between GPIb and VWF plays a major role in determining the platelet retention rates in the smallsized collagen bead column, and that platelet retention in the conventional column is independent of the GPIbVWF interaction. Furthermore, blood samples from a patient with type 3 VWD showed remarkably low platelet retention rates with the small-sized collagen bead column (Fig 1D) under high shear stress. Platelet activation of type 3 VWD is independent of the GPIbVWF interaction, and these findings taken together suggest that the GPIb-VWF interaction plays a major role in platelet retention assessed by the small-sized collagen bead column, and that shear stress high enough to involve GPIb and VWF is generated within the columns. On the other hand, the platelet retention rate in the conventional column with blood samples from type 3 VWD was within the normal range,22 and abciximab, anti-GPIIb/IIIa MoAb, completely inhibited platelet retention (data not shown). These findings suggest that the GPIIb/IIIa-fibrinogen interaction, but not that of GPIb-VWF, plays the major role in platelet activation in the conventional column, and that only low levels of shear stress are generated in this type of collagen bead columns. The involvement of secondary mediators in platelet activation was also evaluated in this system. Factors that may possibly contribute to platelet activation in this system include ADP and TXA2. Our findings are in good accord with the notion that ADP, not TXA2, is an important secondary mediator of platelet activation under high shear stress conditions.29 Compared with the small-sized collagen bead column, ADP and TXA2 appear to play vital roles in the conventional column that corresponds to platelet activation under low shear stress condition (Fig 2B). Recent studies have revealed two major ADP recep-

J Lab Clin Med Volume 146, Number 2

tors on the platelet membranes, namely, P2Y1 and P2Y12.31 Our findings suggest that both contribute almost equally to platelet retention in the small-sized collagen bead column (Fig 2C). P2Y12 has been shown to be the target of thienopyridine drugs (including ticlopidine and clopidogrel), and these agents have proved useful in decreasing the occurrence rates of ischemic heart diseases and cerebrovascular disorders. Our findings imply that P2Y1 inhibitors also serve as good therapeutic agents and that the combined use of P2Y1 and P2Y12 blockers may have further therapeutic effects on vascular disorders related to high shear stress (Fig 2C). The platelet retention rates of the same individuals assessed by the small-sized collagen bead column gave good within-day reproducibility (data not shown). However, the variation of platelet retention rates was larger with a small-sized collagen bead column than that of the conventional column (Fig 3). Because the GPIb-VWF interaction is critical with the small-sized collagen bead column and the normal range of the plasma VWF antigen level is relatively broad, we hypothesized that the large variation of platelet retention rates assessed with the small-sized collagen bead column might be attributed to the VWF antigen level. As shown in Fig 4A, exogenous VWF increased platelet retention rates in individuals with lower retention rates, whereas the increase was not remarkable in individuals with high platelet retention rates. Because the activity of VWF in terms of inducing the GPIb-VWF interaction involves the VWF antigen level and the presence of VWF multimers, we then fractionated VWF preparations and evaluated the ability of each fraction to increase platelet retention rates (Fig 4B and C). In agreement with our expectation, large VWF multimers increased platelet retention rates more than smaller VWF molecules. These findings suggest that the variation of platelet retention rates as assessed by the smallsized collagen bead column can be at least partly explained on the basis of individual differences in the level of VWF antigen or size of VWF multimers. To further confirm our hypothesis, we investigated the correlation of platelet retention rates with the plasma VWF antigen and VWF:RCo in healthy individuals; VWF:RCo is a functional assay for VWF that measures the VWF-supported agglutination of human platelet by ristocetin, and it is known to represent the activity of VWF molecules to elicit the VWF-GPIb interaction.24,46 As expected, the VWF antigen and VWF:RCo are well correlated (Fig 5A). The VWF antigen and platelet retention rates, and the VWF activity assessed by VWF:RCo and platelet retention rates, are also correlated but with coefficient values lower than that of VWF antigen and VWF:RCo (Fig 5B

Kaneko et al

73

and C). These findings suggest that the plasma VWF antigen or VWF:RCo cannot totally explain the levels of platelet retention rates. To better understand the cause for this discrepancy, we calculated the ratio of platelet retention rates to VWF:RCo and visualized the distribution of the data (Fig 5D). It seems that several points are distinct from the majority of data in that platelet retention rates are relatively higher. With these data excluded, the correlation coefficient values are 0.79 for VWF:RCo versus platelet retention rates (Fig 5C). We assume that some as yet unidentified factors other than the VWF activity positively affected platelet retention rates in the small-sized collagen bead column with these individuals. Nonetheless, it can be concluded that low platelet retention rates correlate well with low VWF antigen levels and VWF:RCo. Two patients with type 1 VWD (data not shown) and a patient with type 3 VWD had platelet retention rates lower than 15%. Four individuals whose data are included in Fig 5C had platelet retention rates less than 20% and VWF:RCo levels less than 50%, which are in the diagnostic range of type 1 VWD,47,48 although they had no bleeding tendency. Taken together, we suggest that the platelet retention rates as assessed with this small-sized collagen bead column is useful for screening patients with low or abnormal VWF activities and monitoring patients whose platelet function is potentiated out of proportion to their VWF:RCo or VWF antigen levels. To explore the factors other than VWF that may affect platelet retention within the small-sized collagen bead column, we performed multivariate analysis on several factors that might correlate with platelet retention rates. Among the factors analyzed, hemoglobin and sex in addition to VWF:RCo showed a positive correlation, although the coefficient values did not reach statistically significant levels. Hematocrit levels correlated with the incidence of thrombosis. Red blood cells are known to promote platelet activation and to induce recruitment of additional platelets from microenvironments to forming thrombi. These effects have been attributed either to the mass effects of red blood cells, which may increase shear stress, or ADP released from red blood cells. Hematocrit levels of men are usually higher than those of women, and this may explain the reason for the sex difference. However, six samples that were beyond the mean ⫹ 1.5 SD range in Fig 5D included two men and four women, and their hematocrit/hemoglobin levels were not significantly higher than the others. Thus, factors other than the VWF antigen/activity that affect platelet retention in the small-sized collagen bead column still remain to be elucidated. In this study, we found that the small-sized collagen bead column represents platelet activation related to

74

Kaneko et al

VWF. It is widely accepted that the VWF antigen levels are elevated in patients with endothelial damage such as acute coronary syndrome, stroke, diabetes mellitus, and nephropathy.2 Furthermore, it has recently been reported that TTP is caused by deficiency in ADAMTS13 with a resultant increase in large VWF multimers. Thus, the measurement of the VWF antigen/activity gives valuable information in various clinical settings. This system may not only serve as a new and easy-touse tool for screening quantitative or qualitative abnormalities of VWF as can be observed with VWD and TTP, but also detect platelet hyperfunction independent of VWF. In conclusion, this study demonstrated that the platelet retention rate as determined with the use of the small-sized collagen bead column mainly involves the GPIb-VWF interaction. This system is an easy-to-use method that requires instruments of reasonable cost. It reflects the levels of the VWF antigen and/or its activity in plasma, and the effect of ADP seems to play a vital role in platelet activation (retention) in this system. With consideration of its high reproducibility and simple procedures for measurement, this method, which is capable of analyzing platelet function in whole blood, may be useful for monitoring patients with bleeding tendency or thrombotic disorders. We thank Dr. J. J. Sixma, ISK Co., Ltd., Ajinomoto Co., Inc., AstraZeneca R&D Charnwood, and Kaketsuken for generously providing us with antibodies and reagents. REFERENCES

1. Wiman B, Andersson T, Hallqvist J, Reuterwall C, Ahlbom A, deFaire U. Plasma levels of tissue plasminogen activator/plasminogen activator inhibitor-1 complex and von Willebrand factor are significant risk markers for recurrent myocardial infarction in the Stockholm Heart Epidemiology Program (SHEEP) Study. Arterioscler Thromb Vasc Biol 2000;20:2019 –23. 2. Eto K, Isshiki T, Yamamoto H, et al. AJvW-2, an anti-vWF monoclonal antibody, inhibits enhanced platelet aggregation induced by high shear stress in platelet-rich plasma from patients with acute coronary syndromes. Arterioscler Thromb Vasc Biol 1999;19:877– 82. 3. Folsom AR, Wu KK, Rosamond WD, Sharrett AR, Chambless LE. Prospective study of hemostatic factors and incidence of coronary heart disease: The Atherosclerosis Risk in Communities (ARIC) Study. Circulation 1997;96:1102– 8. 4. Goto S, Sakai H, Goto M, et al. Enhanced shear-induced platelet aggregation in acute myocardial infarction. Circulation 1999;99: 608 –13. 5. Jansson J, Nilsson T, Johnson O. von Willebrand factor in plasma: a novel risk factor for recurrent myocardial infarction and death. Br Heart J 1991;66:351–5. 6. Qizilbash N, Duffy S, Prentice C, Boothby M, Warlow C. Von Willebrand factor and risk of ischemic stroke. Neurology 1997; 49:1552– 6. 7. Uchiyama S, Yamazaki M, Maruyama S, et al. Shear-induced platelet aggregation in cerebral ischemia. Stroke 1994;25:1547– 51.

J Lab Clin Med August 2005

8. Folsom AR, Rosamond WD, Shahar E, et al. Prospective study of markers of hemostatic function with risk of ischemic stroke. Circulation 1999;100:736 – 42. 9. Moake JL. von Willebrand factor, ADAMTS-13, and thrombotic thrombocytopenic purpura. Semin Hematol 2004;41:4 –14. 10. Sadler JE. A new name in thrombosis, ADAMTS13. Proc Natl Acad Sci U S A 2002;99:11552– 4. 11. Born GVR. Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature 1962;194:927–9. 12. Carr ME Jr. In vitro assessment of platelet function. Transfus Med Rev 1997;11:106 –15. 13. Ikeda Y, Handa M, Kawano K, et al. The role of von Willebrand factor and fibrinogen in platelet aggregation under varying shear stress. J Clin Invest 1991;87:1234 – 40. 14. Harrison P. Progress in the assessment of platelet function. Br J Haematol 2000;111:733– 44. 15. Favaloro EJ. Appropriate laboratory assessment as a critical facet in the proper diagnosis and classification of von Willebrand disorder. Best Pract Res Clin Haematol 2001;14:299 –319. 16. Jilma B. Platelet function analyzer (PFA-100): a tool to quantify congenital or acquired platelet dysfunction. J Lab Clin Med 2001;138:152– 63. 17. Moake JL, Turner NA, Stathopoulos NA, Nolasco LH, Hellums JD. Involvement of large plasma von Willebrand factor (vWF) multimers and unusually large vWF forms derived from endothelial cells in shear stress-induced platelet aggregation. J Clin Invest 1986;78:1456 – 61. 18. Coller BS. Monitoring platelet GP IIb/IIIa antagonist therapy. Circulation 1998;97:5–9. 19. Salzman EW. Measurement of platelet adhesiveness. A simple in vitro technique demonstrating an abnormality in von Willebrand’s disease. J Lab Clin Med 1963;62:724 –35. 20. Hellem AJ. Platelet adhesiveness in von Willebrand’s disease. A study with a new modification of the glass bead filter method. Scand J Haematol 1970;7:374 – 82. 21. Rossi EC, Green D. A study of platelet retention by glass bead columns (‘platelet adhesiveness’ in normal subjects). Br J Haematol 1972;23:47–57. 22. Kaneko M, Cuyun-Lira O, Takafuta T, et al. Mechanisms of platelet retention in the collagen-coated-bead column. J Lab Clin Med 2003;142:258 – 67. 23. Romijn RA, Bouma B, Wuyster W, et al. Identification of the collagen-binding site of the von Willebrand factor A3-domain. J Biol Chem 2001;276:9985–91. 24. Weiss HJ, Hoyer LW, Rickles FR, Varma A, Rogers J. Quantitative assay of a plasma factor deficient in von Willebrand’s disease that is necessary for platelet aggregation. Relationship to factor VIII procoagulant activity and antigen content. J Clin Invest 1973;52:2708 –16. 25. Ohmori T, Yatomi Y, Asazuma N, Satoh K, Ozaki Y. Involvement of proline-rich tyrosine kinase 2 in platelet activation: tyrosine phosphorylation mostly dependent on alphaIIbbeta3 integrin and protein kinase C, translocation to the cytoskeleton and association with Shc through Grb2. Biochem J 2000;347:561–9. 26. Ruggeri ZM, Zimmerman TS. Variant von Willebrand’s disease: characterization of two subtypes by analysis of multimeric composition of factor VIII/von Willebrand factor in plasma and platelets. J Clin Invest 1980;65:1318 –25. 27. van Mourik JA, Mochtar IA. Purification of human antihemophilic factor (factor VIII) by gel chromatography. Biochim Biophys Acta 1970;221:677–9. 28. Ruggeri ZM. Structure of von Willebrand factor and its function in platelet adhesion and thrombus formation. Best Pract Res Clin Haematol 2001;14:257–79.

J Lab Clin Med Volume 146, Number 2

29. Moake JL, Turner NA, Stathopoulos NA, Nolasco L, Hellums JD. Shear-induced platelet aggregation can be mediated by vWF released from platelets, as well as by exogenous large or unusually large vWF multimers, requires adenosine diphosphate, and is resistant to aspirin. Blood 1988;71:1366 –74. 30. Kondo K, Seo R, Naka M, et al. Effects of ONO-3708, an antagonist of the thromboxane A2/prostaglandin endoperoxide receptor, on platelet aggregation and thrombosis. Eur J Pharmacol 1989;163:253– 61. 31. Gachet C. ADP receptors of platelets and their inhibition. Thromb Haemost 2001;86:222–32. 32. Gastineau DA, Moore SB. How important are ABO-related variations in coagulation factor levels? Transfusion 2001;41: 4 –5. 33. Sixma J, Sakariassen K, Beeser-Visser N, Ottenhof-Rovers M, Bolhuis P. Adhesion of platelets to human artery subendothelium: effect of factor VIII-von Willebrand factor of various multimeric composition. Blood 1984;63:128 –39. 34. Sporn L, Marder V, Wagner D. von Willebrand factor released from Weibel-Palade bodies binds more avidly to extracellular matrix than that secreted constitutively. Blood 1987;69:1531– 4. 35. Spivak JL, Barosi G, Tognoni G, et al. Chronic myeloproliferative disorders. Hematology 2003;2003:200 –24. 36. Kwaan HC, Wang J. Hyperviscosity in polycythemia vera and other red cell abnormalities. Semin Thromb Hemost 2003;29: 451– 8. 37. Knofler R, Weissbach G, Kuhlisch E. Release of adenosine triphosphate by adenosine diphosphate in whole blood and in erythrocyte suspensions. Am J Hematol 1997;56:259 – 65. 38. Emery JD, Leifer DW, Moura GL, Southern P, Morrissey JH, Lawrence JB. Whole-blood platelet aggregation predicts in vitro

Kaneko et al

39.

40.

41. 42. 43.

44.

45.

46.

47. 48.

75

and in vivo primary hemostatic function in the elderly. Arterioscler Thromb Vasc Biol 1995;15:748 –53. Matsui T, Shimoyama T, Matsumoto M, et al. ABO blood group antigens on human plasma von Willebrand factor after ABOmismatched bone marrow transplantation. Blood 1999;94:2895– 900. Gill J, Endres-Brooks J, Bauer P, Marks WJ, Montgomery R. The effect of ABO blood group on the diagnosis of von Willebrand disease. Blood 1987;69:1691–5. Ruggeri ZM. Von Willebrand factor, platelets and endothelial cell interactions. J Thromb Haemost 2003;1:1335– 42. Jackson SP, Nesbitt WS, Kulkarni S. Signaling events underlying thrombus formation. J Thromb Haemost 2003;1:1602–12. Satoh K, Asazuma N, Yatomi Y, et al. Activation of proteintyrosine kinase pathways in human platelets stimulated with the A1 domain of von Willebrand factor. Platelets 2000;11:171– 6. Savage B, Saldivar E, Ruggeri ZM. Initiation of platelet adhesion by arrest onto fibrinogen or translocation on von Willebrand factor. Cell 1996;84:289 –97. Frojmovic MM. Platelet aggregation in flow: differential roles for adhesive receptors and ligands. Am Heart J 1998;135:S119 – S31. Sarji KE, Stratton RD, Wagner RH, Brinkhous KM. Nature of von Willebrand factor: a new assay and a specific inhibitor. Proc Natl Acad Sci U S A 1974;71:2937– 41. Federici AB. Clinical diagnosis of von Willebrand disease. Haemophilia 2004;10:169 –76. Favaloro EJ, Lillicrap D, Lazzari MA, et al. von Willebrand disease: laboratory aspects of diagnosis and treatment. Haemophilia 2004;10:164 – 8.