Platelet Adhesion and Aggregation on Human Type VI Collagen

Platelet Adhesion and Aggregation on Human Type VI Collagen Surfaces Under Physiological Flow Conditions

By Julia M. Ross, Larry V. Mclntire, Joel L. Moake, and Jacob H. Rand

Type VI collagen is a subendothelial constituent that binds von Willebrand factor (vWF) and platelets. The interaction of platelets with type VI collagen and the roles of platelet glycoprotein (GP) receptors and vWF were studied under flow conditions using epi-fluorescent videomicroscopy coupled with digital image processing. We found that surface coverage was less than 6% on collagen VI at a relatively high-wall shear rate (1,000 S-’) and was approximately 60%

most effective in eliminating adhesion (surface coverage, 0.8961, followed by anti-GPlb (4.3%). and ATA (12.6%). Experi- ments with von Willebrand disease blood indicate that vWF is involved in platelet adhesion to collagen VI at 100 S-’. In the absence of vWF, there may be direct binding of platelet GPllb-llla complexes to collagen VI. Adhesion and aggregation on collagen VI are different in shear rate dependence from collagen 1. Our results suggest a possible role for colla-

at a low-wall shear rate (l00 S“). The molecular mechanisms gen VI and vWF in platelet adhesion and aggregation in vas- involved in low-shear platelet binding were studied using cular regions with low shear rates. monoclonal antibodies to platelet GPlb and GPllb-llla, and 0 1995 by The American society of /f8mi3tO/OgY. polymeric aurin tricarboxylic acid. Anti-GPllb-llla was the

T HE ADHESION OF BLOOD platelets to the exposed vascular subendothelial matrix proteins at the site of

injury is a crucial initiating step in the hemostatic and throm- botic processes. The glycoprotein von Willebrand factor (vWF) is present in the subendothelium,l,2 where it mediates platelet adhesion to the subendothelium3.‘ under conditions of high shear stress.’.K Platelets contain two membrane-binding sites for vWF. Initial adhesive (platelet-vessel wall) interactions at high-wall shear rates primarily involve glycoprotein Ib (GPIb), whereas subsequent platelet-platelet interactions are mediated by the GPIIb-IIIa complex.’.’” The vWF-binding site in the subendothelium is currently not known but has been shown not to be fibrillar collagen.”,’*

One vWF-binding protein extracted from human vascular subendothelium is collagen VI.” The collagen VI molecule has been found in the vascular subendothelium, where it colocalizes with vWF,I4 and is anatomically positioned to mediate platelet adhesion after endothelial injury. Recently, it has been shown that type VI collagen stimulates platelet adhesion, and that this interaction requires the GPIa-IIa (VLA- 1) platelet integrin re~eptor.’~

The structure of collagen VI is unique within the collagen family and consists of a short 105 nm triple helix with globular domains of unequal size on either end that give the protein a dumbbell shape.“.” The globular portions of collagen VI have repetitive domains that are homologous to the A domains of vWF monomers.’8.” In vWF, these domains contain

From the Cox Laboratory for Biomedical Engineering, Rice Uni- versity, and the Division of Hematology, Baylor College of Medicine, Houston, 7X: and the Hematology Division, Mt. Sinai School of Medicine, New York, NY.

Submitted May 13, 1994; accepted October 25, 1994. Supported by National Institutes of Health Grants No. HL-18672,

HL-32200, NS-23327; the Robert A. Welch Foundation Grant No. C-938; and a grant from the Butcher Fund.

Address reprint request., to Larry V. Mclntire, PhD, Cox Labora- t o n ,for Biomedical Engineering, Rice University, P.O. Box 1892, Houston. 7X 77251.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with I8 U.S.C. section 1734 solely to indicate this fact. 0 1995 by The American Society of Hematology. 000ci-4971/y5/8507-0026$3.00/0

the GPIb, heparin, and fibrillar collagen types I and 111 binding sites.20-23 The triple helical region of the collagen VI molecule contains several RGD sequences.24 The platelet GPIIb-IIIa receptor has been shown to bind to the RGD amino acid sequence25 in vWF,’~ fibrinogen:’ fibronectin,2K and vitronectin.”

The purpose of this study was to examine and quantitate the kinetic interactions of platelets with collagen VI surfaces in real-time under various physiologic wall shear stress conditions. Also, details of the possible role of vWF in the low shear environment and the molecular mechanisms involved in mural adhesion and aggregation were examined.

MATERIALS AND METHODS

Blood collection and preparation. Blood was collected from healthy, nonsmoking, aspirin-free donors into polypropylene syrin- ges containing anticoagulant and was used within 3 hours. The final concentration of anticoagulant was either I O U/mL porcine heparin (heparin sodium; Elkins-Sinn Inc, Cherry Hill, NJ) or 0.38% wt/ v01 sodium citrate. Mepacrine (quinacrine dihydrochloride; Sigma Chemical CO, St Louis, MO), a fluorescent dye that concentrates in the dense granules of platelets and leukocytes, was added at a final concentration of 10 pmol/L, a level that has been shown to have no effect on normal platelet f~nction.~” Erythrocyte fluorescence is quenched by hemoglobin, and leukocytes can be distinguished from platelets visually using epi-fluorescent videomicroscopy. It is proba- ble that the platelets do secrete their granular contents after activation by collagen. After secretion, rapid reuptake of the mepacrine may take place (as with serotonin). This may explain why the net mepacrine fluorescence intensity does not change during the experiments. When a single platelet is monitored throughout the course of an experiment (2 or 10 minutes), no decrease in mean intensity is measured by the digital imaging system.

Collagen VI isolation and purijcation. Type VI collagen was prepared following the method of Kuo et al,” which was modified to use umbilical cords as the starting material. We had previously shown the presence of type VI collagen in umbilical veins’4 and had identified type VI collagen as a binding site for vWF in that tissue.”

Preparation of collagen suspensions and coating of coverslips. A suspension of type I acid-insoluble bovine achilles tendon collagen fibrils (Sigma) in 0.5 mom acetic acid was preparedg2 at a concentration of 980 pg/mL found by hydroxyproline a~say.”.’~ Glass coverslips (no. 1, 24 mm X 50 mm; Corning Inc, Corning, NY) were coated with 200 pL of the collagen I suspension on all but the first 10 mm of the slide width and were placed in a humid environment for at least 45 minutes, The glass slides were then rinsed with 1 0 mL of isotonic saline before assembly into the flow chamber. The

1826 Blood, Vol 85, No 7 (April 1). 1995: pp 1826-1835

For personal use only.on December 11, 2018. by guest www.bloodjournal.orgFrom

http://www.bloodjournal.org/

http://www.bloodjournal.org/site/subscriptions/ToS.xhtml

MURAL THROMBOSIS ON TYPE VI COLLAGEN

collagen concentration on the glass slides was measured by the difference in weight of 20 clean slides versus the same 20 slides coated with collagen in the above manner and subsequently dried. The average collagen concentration on the glass slides was 2.5 pgl cm'.

Suspensions of human type VI collagen were prepared as above with all materials scaled down by a factor of 200. The concentration of this suspension was found to be 76.2 pglmL by amino acid analysis. Slides were coated with 20 pL of collagen VI suspension over an area of 60 mm' and were humidified. The slides were rinsed with 2 mL of isotonic saline, and the collagen concentrations were measured by amino acid analysis as previously described.'s The collagen VI concentration on the glass slides was calculated to be 2.2 pglcm'.

Monoclonal antibodies (MoAbs). The MoAbs U-CP8 and U- Ibl were kindly provided by Dr Zaverio M. Ruggeri (Scripps Re- search Institute, La Jolla, CA). U-CP8 is a murine monoclonal IgGl that reacts with GPllb-llla and blocks vWF, fibronectin, fibrinogen, and vitronectin binding.'" U-lbl is a murine monoclonal IgGl that completely inhibits vWF-GPlb interaction." Both MoAbs were puri- fied on protein-A sepharose (Sigma) and dialyzed against HEPES buffer.zx Antibody concentrations in whole blood were 100 pglmL in all experiments.

Polymeric aurin rricarboxylic acid (ATA). ATA interacts with vWF and inhibits the binding of large vWF multimers to platelet GPIb.29 The trisodium salt of ATA (Aldrich Chemical, Milwaukee, W!) was dissolved in phosphate-buffered saline (10 mmol/L NalHPO,, 140 mmol/L NaCI, pH 7.4). Higher molecular weight polymers were separated from low molecular weight polymers by dialysis at 4°C with a 50-kD cutoff membrane (SpectrdPor; Spec- trum Medical Industries, Inc, Los Angeles, CA). A filter-sterilized stock of 10 mglmL ATA with a molecular mass of more than 2.5 kD was prepared and stored at room temperature. The concentration of polymeric ATA in all runs was 240 pg/mL in whole blood.

von Willebrand diseuse (vWD) blood. Blood from a patient with severe vWD was anticoagulated with 0.38% wt/vol sodium citrate and tested on collagen VI surfaces at a shear rate of 1 0 0 S - ' . Plasma vWF antigen levels in this patient have been less than 1.5 UldL on 12 different occasions by immunoradiometric assay and enzyme- linked immunosorbent assay, and vWF multimeric patterns have been either undetectable or barely visible by sodium dodecyl sulfate- electrophoresislautoradiography using patient plasma or platelets!"

Epi-fluorescent videomicroscopy. Epi-fluorescent videomicroscopy coupled with digital image processing was used to visualize and quantitate platelet adhesion and subsequent aggregation in whole

Before use, blood was incubated at 37°C for 30 minutes. For the antibody studies, either antibody or phosphate-buffered saline was added at 25 minutes for a 5-minute incubation. A parallel plate flow chamber was used to model a damaged blood vessel and has been described elsewhere!2 The flow chamber was assembled using a coverslip coated with either type I or type VI collagen and was filled with isotonic saline. The blood was infused through the flow chamber and across the collagen-coated coverslip at a constant flow rate for 2 to IO minutes via a syringe pump (model 935; Harvard Apparatus, South Natick, MA). Wall shear rates of 1 0 0 S" (a venous shear rate) and 1.OOO S - ' (an arterial shear rate)" were attained by varying the flow rate on the pump and are a function of the flow chamber geometry." These shear rates correspond to wall shear stresses of approximately 4 and 40 dyne/cm2, respectively, for whole blood at a normal hematocrit in laminar flow. The flow chamber was mounted on an inverted-stage microscope (DIAPHOT-TMD; Nikon, Garden City, NY) equipped with an epi-fluorescent illumination attachment (TMD-EF Nikon), a 60x FLUOR objective, a 5 X projection lens (Nikon), and a silicon-intensified target video camera (model CIOOO; Hamamatsu, Waltham, MA) suitable for very low

blood.".'"

A W Collagen VI

c3 Collagen I (n-5)

2 min 10 min 2 min

Shear Rate (8")

B 16[ 14

(n=5)

l

100 2 min

100. loop 10 mm 2 mm

Shear Rate (8") .

Fig 1. Effect of shear rate on platelet adhesion (A) and aggregation (B1 on bovine collagen type I and human collagen type VI surfaces in whole blood anticoagulated with porcine heparin is shown.

light levels. The microscope stage, syringe pump, and blood were maintained at 37°C with a thermostatic air bath (model 279; Labora- tory Products, Boston, MA) and an air curtain incubator. Excitation light was filtered with a 400-nm interference filter before passing through a standard Nikon BV filter cassette. Experiments were re- corded in real time on a 0.5-in video cassette recorder (JVC model BR-3 IOOU) at 2 mm downstream from the edge of the flow chamber. Run times were 2 minutes for the 1,OOO S" wall shear rate experiments and I O minutes for the 1 0 0 S" shear rate experiments. Initially, the low shear experiments were run for 20 minutes so that the amount of blood that passed over the surface was the same as for the high shear experiments. No significant difference was observed in platelet accumulation between 10 and 20 minutes of flow; therefore, the run time was cut to IO minutes to minimize the amount of blood required. Early experiments at the lower flow rate indicated higher platelet accumulation on the area of the surface exposed to extended UV illumination than on the other parts of the surface. This phenomenon began only after approximately 4 minutes into experimental runs.




1828 ROSS ET AL

Table 1. Platelet Adhesion and Aggregation on Collagen I: Platelet Thrombus Size Distributions

% of Thrombi Containing the Following No. of Platelets

Time 0 to 5 6 to 20

Shear rate = 1,000 S ' 20 S 85 (47) 13 (33) 40 S 70 (11) 17 (17) 60 S 72 (3) 11 (3) 80 S 64 (2) 24 (3) 100 61 (1) 20 (1) 120 S 65 (1 1 10 (1)

Shear rate = 100 S ' ' 2 min 49 (2) 18 (6) 4 min 32 (1) 22 (21 6 min 46 (1) 16 (2) 8 min 54 (1) 17 (2) 10 min 54 (1 1 19 (2)

Values in parentheses indicate the percentage of total accumulated

To obviate this potential problem, during the IO-minute low shear experiments, the port on the microscope to the UV lamp was opened only every 2 minutes for approximately I O seconds to collect images for analysis.

Digirul imugeproceusing. Images from the videotapes were digitized at 15-second intervals for the 2-minute high shear runs and at 2-minute intervals for the IO-minute low shear runs (IC-300 Modular Image Processing Workstation; lnovision Corporation, Durham, NC).4' In addition, a background image and a 2-second image were digitized immediately after the initiation of flow for each experimental run. The images were processed by subtracting a background image for each experimental run, by eroding all thrombi found on the surface by I pixel to give individual thrombi more distinct bound- aries, and by extracting data from each thrombus found on the surface. These data included the total number of pixels covered by each thrombus, the mean intensity level of each thrombus, and the integral of the intensity over the entire area of the thrombus (total intensity). The 2-second image consisted of only single platelets and was used to determine the total intensity of each platelet. The intensity level of each platelet was assigned a mean gray level between 0 (black) and 255 (white) by the imaging software. This value was multiplied by the total number of pixels covered by a single platelet. The resulting number was averaged for all of the single platelets to determine the average total intensity for a platelet in each experiment.

Fluorescent intensity has been shown to be directly proportional to thrombus height for platelet densities up to 1,OOO platelets11,000 pm2."s All thrombi in this study were well within this range. The total intensity of a thrombus is a "volume" measurement (height X pixel area). By knowing the mean total intensity or "volume" of a single platelet, it is then possible to divide the total intensity of a thrombus by the mean total intensity of a single platelet to determine the number of platelets in the individual thrombus.

Platelet adhesion is expressed by the percentage of surface coverage. Platelet aggregation is expressed by the total number of platelets on the surface, which is calculated as described above from the total intensity multiplied by the conversion factor representing the inverse of the total intensity for a single platelet. After smoothing the images. three dimensional representations of the platelet thrombi found at various times during the experimental runs were constructed by plotting the intensity as a height on the z-axis. These images are consistent with three-dimensional images obtained using confocal microscopy.4"~17

Sruristicul unu1y.si.q. Statistical significance of differences of

-

21 to 100 101 to 500 501 to 1,000 -- 1.000

2 (20) 12 (45) 1 (27) 12 (17) 3 (17) 1 (16) 1 (44) 9 (6) 2 (61 0 1 (83)

17 (4) 0 0 2 (94) 21 (3) 2 (1) 0 2 (94)

22 (20) 9 (44) 2 (28) 33 (16) 7 (30) 4 (26) 2 (25) 23 (11) 10 (201 2 (11) 3 (55) 17 (7) 7 (161 0 4 (74) 17 (7) 6 ( 1 1 ) 0 4 (791

platelets in thrombi of various sizes in the area of observation.

means between control and test runs was calculated using a two- tailed paired Student's r-test in all the antibody studies. To test for statistical significance within the 2 X 2 factorial design experiment involving type I collagen and type VI collagen at both high and low shear rates, a two-way analysis of variance for unbalanced data was applied.4x,4' In all cases, a probability value of .05 was used to test for significance. The mean t- SEM was calculated for experiments based on data from at least 3 donors with duplication.

RESULTS

Flow studies of type I and type VI collugen-coated sur- ,fuces ut shear rates of 100 S" and 1,000 .S". Figure 1 shows summary data of the adhesion and aggregation results for heparinized whole blood on collagen I and collagen VI surfaces at both high- (1,000 S l ) and low-wall shear rates (100 S l ) . Table 1 gives platelet thrombus distributions for a typical donor for an experiment on a collagen I surface at various time points. This includes analysis of the percentage of thrombi of various sizes on the surface as well as the percentage of total accumulated platelets in thrombi of different sizes. At a shear rate of 1,000 S - . ' , extensive adhesion and aggregation occurred on collagen I surfaces. Surface coverage was 38.0% 2 6.770, with 8,900 t 2,300 platelets deposited per surface area (3.1 X 1 O4 pm') after I20 seconds. Also at this time, very large thrombi containing greater than 1,000 platelets accounted for 94% of all the platelets on the surface (Fig 2). At a shear rate of 100 S - ' on collagen I, surface coverage was 12.6% 5 2.4%, with 2,705 2 408 total platelets deposited per surface area after 2 minutes. The largest thrombi that form after 2 minutes of flow contain between 501 and 1,000 platelets. After 10 minutes of blood flow, surface coverage was 34.4% 2 2.3%, with 9,700 t 600 total platelets deposited per surface area. In this case, very large thrombi (> 1000 platelets) accounted for only 79% of all the platelets on the surface (Fig 2).

The results on type VI collagen surfaces are very different and are shown in Fig 1, with detailed dynamic information presented in Table 2 for a typical donor. Adhesion and subsequent aggregation on collagen VI surfaces at a wall shear rate of 1,000 S - ' produced only 5.7% ? 1.3% surface cover-




MURAL THROMBOSIS ON TYPE VI COLLAGEN 1829

I,

1. ,'..:..

I 2 min 2 min 10 min

I ' I

I

" 1 2 min L mm IU mm

1000 /S 100 /S

F , I - I

Fig 2. Three-dimensional representations of platelet adhesion and subsequent aggregation onto bovine collagen type I and human collagen type VI in heparinized whole blood are shown. The area of each repre- sentation in this and subsequent figures is 3.1 x 10' pm2.

Intensity m

Highest - iowest

"

I 4 min 6 min

8 min 7 10 min

1 .

I , 1

. P

P I OT-

l

I 2 min

Fig 3. Three-dimensional representations are shown of platelet adhesion and subsequent aggregation onto human collagen type VI during a 10-minute infusion of heparinized whole blood a t a wall shear rate of 100 S".

L




1830 ROSS ET AL

Table 2. Platelet Adhesion and Aggregetion on Collagen VI: Platelet Thrombus Size Distributions


Time 0 to 5 6 to 20 21 to 100 101 to 500 501 to 1,000 -- 1,000

age, with 700 i: 200 platelets deposited per surface area after 120 seconds. The thrombi formed under these high shear conditions were relatively small, containing less than 100 platelets per thrombus. At the end of the experiments, 33% of the total platelets on the surface were in small thrombi containing between 0 and 5 platelets (Fig 2). At the lower wall shear rate of 100 S" platelet adhesion and aggregation produced 1 1.6% +- 2.8% surface coverage with 2,080 t 643 total platelets deposited per surface area after 2 minutes. However, after I O minutes of shear, adhesion and aggregation produced 59.7% t 5.0% surface coverage, with 13,400 t 2,400 total platelets deposited per surface area. Large thrombi containing more than 1,000 platelets accounted for 98% of all the platelets on the surface (Figs 2 and 3).

There was a statistically significant ( P = .05) difference between type I and type VI collagens at both a high (1,000 S - ') wall shear rate after 120 seconds and at low (100 S")

wall shear rate after 10 minutes in surface coverage and total number of platelets deposited per surface area. In addition, the differences in surface coverage and total platelets deposited per surface area on type VI collagen surfaces at 1,000 S - ' after 120 seconds, 100 S" after 2 minutes, and 100 S"

after 10 minutes were also statistically significant. However, there was no statistically significant difference in surface coverage and total number of platelets deposited per surface area between type I and type VI collagen at the low (100 S - ' ) wall shear rate after only 2 minutes.

Efject of anticoagulant on platelet adhesion and uggrega- tion to collagen VI surfaces ut low shear. Comparisons were made of control experiments on collagen VI surfaces at a shear rate of 100 S" with blood anticoagulated with either heparin or sodium citrate. Sodium citrate had a moder- ate effect on platelet adhesion to the collagen VI surface and a somewhat greater effect on aggregation compared to heparin. When heparin was used as the anticoagulant, the surface coverage was 59.7% t 5.0%, with 13,400 ? 2,400 total plate!ets deposited per surface area after I O minutes, as reported above. When sodium citrate was used as the anticoagulant, the surface coverage was 46.0% ? 4.2%, with

7,300 t 1,200 total platelets deposited per surface area after 10 minutes. Because these differences were statistically significant, all experiments with sodium citrate as anticoagulant included control runs that also used sodium citrate.

Eflect of severe vWD blood on platelet adhesion to collagen VI sLu$uces at low shear. Whole blood from a severe vWD patient was anticoagulated with citrate and was used in low shear rate (100 S - ' ) studies on type VI collagen over a period of 2 days (platelet counts, >200,0oO/pL). Before treatment (day l ) , the level of vWF antigen in patient platelet-poor plasma was less than 1.5 U/dL, and the hematocrit was 23%. As shown in Table 3, the surface coverage under these conditions was I l%, with 1,700 total platelets deposited per surface area after 10 minutes of flow. Because the low hematocrit of 23% would contribute to decreased platelet transport to the collagen VI surface," normal blood was adjusted to an equivalent hematocrit for the control experiments. The quantity of citrate relative to the volume of red blood cells (RBCs) for this control was the same as in the vWD blood. The percentage of surface coverage was 34.3% for a control donor with a hematocrit of 34% and an antigen level of l32 UfdL. The percentage of surface coverage for the same donor with a hematocrit of 23% was 25.0%, which was significantly higher than that of the vWD blood (1 l .O%). The patient was then treated with human cryoprecipitate that contained all vWF multimeric forms in normal plasma and with 4 U of packed RBCs. On the second day (one day after

Table 3. Effect of vWD Blood on Platelet Adhesion on Collagen VI at Shear Rate of 100 S"

% Surface Total Platelets Coverage Deposited*

Day 1, pretransfusion of vWF 11.0 1.7 Day 2, posttransfusion of vWF 49.8 10.2

The anticoagulant used was citrate. On day 1, the antigen level was 0 U/dL and the hematocrit was 23%. On day 2, the antigen level was 80 U/dL and the hematocrit was 32%.

* Values are the total platelets deposited x lo3 per surface area. Surface area is 3.1 x lo4 pm2.





Fig 4. Three-dimensional representations are shown of platelet adhesion and subsequent aggregation during a 10-minute infusion with citrated blood from a normal donor and from a patient with severe vWD (plasma vWF, <1.5 UldL, or 1.5%) at a wall shear rate of 100 S" on human collagen VI.

?

Control donor Antigen level = 132U/dl Hematocrit = 36%

Control donor Antigen level =132U/dl Hematocrit = 23%

vWD pre treatment Antigen level = OU/dl Hematocrit = 23%

W vWD post treatmeni Antigen level = 80U/dl Hematocrit = 32%

Control + Anti-GPIIb-IIIa

+Anti-GPIb + ATA I

Fig 5. Three-dimensional representations are shown of platelet adhesion and subsequent aggregation onto human collagen type VI from normal citrated blood in the presence of 100 pg/mL of anti- GPllb-Ma, 100 ag/mL of anti-GPlb, or 240 pg/mL of ATA. Blood was incubated for 5 minutes with the MoAb or ATA at 37°C and was infused at a wall shear rate of 100 S" for 10 minutes.




1832 ROSS ET AL

Table 4. Effect of MoAbs and Polymeric ATA on Platelet Adhesion on Collagen VI at Shear Rate of 100 S"

Total Platelets % Surface Coverage Deposited (% of

Treatment Groups (% of Control) Control)*

Control (n = 9) 46.0 2 4.2 7.3 2 1.2 Anti-GPllb-llla (n = 3) 0.8 2 0.3 (1.7)t 0.06 2 0.02 (0.8)t Anti-GPlb (n = 3) 4.3 f 1.2 (9.3)t 0.4 2 0.1 (5.5)t ATA (n = 3) 12.6 2 2.9 (27.4)t 1.3 _f 0.3 (17.8)t

The anticoagulant used was citrate. Values are expressed as mean

* Values are the total platelets deposited x IO3 per surface area. 2 SEM.

Surface area is 3.1 x lo4 pm'.

two-tail Student's t-test. t Statistically significant at a = .05 relative to respective controls;

treatment) the plasma vWF antigen concentration was 80 U/ dL, and the hematocrit was 32%. Surface coverage under these conditions rose to 49.8%, with 10,200 total platelets deposited per surface area. These results fall into the 95% confidence interval for normal donors. Three-dimensional representations of the thrombi formed using normal donor blood with hematocrits of 34% and 23% as well as thrombi formed at the end of the experiments using the severe vWD blood on each day are shown in Fig 4.

Effect of a MoAb to GPIIb-IIIa on platelet adhesion and aggregation to collagen VI surfaces ut low shear. Whole blood anticoagulated with citrate was incubated with LJ-CP8 for 5 minutes and infused over collagen VI surfaces at a shear rate of 100 S-' to study the effect of blocking platelet GPIIb-IIIa under these conditions. Three donors were used for the experiments with duplicates of all runs and results are shown in Table 4. When LJ-CP8 was added to the blood, the percentage of surface coverage decreased to only 1.7% of the control value, and the total number of platelets deposited per surface area decreased to 0.8% of control. These differences are statistically significant at P = .05. In addition, after 8 minutes of blood flow in the presence of LJ-CP8, the largest thrombi formed on the surface contained S platelets or less. After 10 minutes of blood flow in the presence of LJ-CP8, the largest thrombi formed on the surface had between 21 and 100 platelets, and 98% of the thrombi on the surface were comprised of no more than S platelets. Three- dimensional representations of thrombi that formed on the collagen VI surface in the presence or absence of LJ-CP8 are shown in Fig S . The addition of anti-GPIIb-IIIa greatly reduced aggregation subsequent to platelet adhesion on the collagen VI surface.

Effect ojpolymeric ATA and a MoAb to GPIb on platelet adhesion and aggregation to collagen VI surfaces at low shear. Whole blood anticoagulated with citrate was incubated with LJ-Ibl for S minutes and infused over collagen VI surfaces at a shear rate of 100 S" to study the effect of blocking vWF-GPIb interactions. Three donors were used for the experiments with duplicates of all runs. Summary results from these experiments are shown in Table 4. In the presence of LJ-Ibl, the percentage of surface coverage decreased to 9.3% of the control value, and the total number of platelets deposited per surface area decreased to 5.5% of

Table 5. Effect of Anti-GPlb on Platelet Adhesion and Aggregation on Collagen VI at Shear Rate of 100 S":

Platelet Thrombus Size Distributions


Time Oto 5 6 to 20 21 to 100 101 to 500 501 to 1,000 ,1,000

2 min 84 (26) 9 (17) 7 (57) 4 min 85 (28) 10 (20) 5 (52) 6 min 82 (36) 12 (23) 6 (41) 8 min 74 (15) 13 (17) 13 (68) 10 min 68(9) 15 (12) 15(51) 2 (28)

Values in parentheses indicate the percentage of total accumulated platelets in thrombi of various sizes in the area of observation

control. These values are statistically significant at P = .OS. In the presence of LJ-Ibl antibody, the largest thrombi formed contained 101 to SO0 platelets, with 68% of the thrombi on the surface being comprised of no more than 5 platelets after 10 minutes (Table S ) . Three-dimensional representations of thrombi that formed on the collagen VI surface in the presence or absence of LJ-Ibl are shown in Fig 5.

Similar experiments were run using polymeric ATA, which binds to vWF and prevents its binding to platelet GP- Ib.39 Results of the ATA studies are also summarized in Table 4. Incubation of normal whole blood with polymeric ATA before infusion caused the percentage of surface coverage by platelets to decrease to 27.4% of the control value, and the total number of platelets deposited per surface area to decrease to 17.8% of control. These differences are statistically significant at P = .05. The thrombus size distributions are shown in Table 6. As with anti-GPIb, the largest thrombi contained 101 to SO0 platelets, and 66% of the thrombi were comprised of no more than 5 platelets after 10 minutes. Three-dimensional representations of thrombi formed on the collagen VI surfaces in the presence or absence of ATA are shown in Fig 5.

DISCUSSION

The effect of shear stress on platelet adhesion and aggregation onto collagen VI surfaces is dramatic. Confirming the findings of Saelman et aI,l5 the surface appears most reactive to platelets at a low shear rate (100 S - ' ) , when platelet trans-

Table 6. Effect of ATA on Platelet Adhesion and Aggregation on Collagen VI at Shear Rate of 100 S":

Platelet Thrombus Size Distributions


Time Oto 5 6 to20 21 to 100 101 to500 501 to 1,000 >1,000

Values in parentheses indicate the percentage of total accumulated platelets in thrombi of various sizes in the area of observation.





port to the surface is relatively slow. Under high shear rate conditions (1,OOO S-') at which platelet transport to the surface is high, the collagen VI surface is not capable of sup- porting mural platelet adhesion and subsequent aggregation. Collagen VI is considerably more reactive with platelets than is collagen I under low shear conditions, as judged by surface coverage and the total number of platelets per surface area deposited after 10 minutes of blood flow. Our results suggest a possibly important role for collagen VI in hemostasis and thrombosis in vascular regions associated with low shear rates, such as in the venous circulation and downstream of bifurcations in the arterial circulation. The latter have been shown to be predisposed to atherosclerotic lesion develop- ment in h~rnans.~'

To investigate the molecular mechanisms of platelet binding to type VI collagen, we differentiated between platelet adhesion (measured by the percentage of the surface covered by platelets) and platelet aggregation (measured as the total number of platelets that were deposited per unit surface area of 3.1 X lo4 pm2). Differential levels of aggregation can also be qualitatively observed from the three-dimensional representations of thrombi that formed on the surfaces.

Under low shear rate conditions, anti-GPIb and anti- GPIIb-IIIa both significantly decreased platelet adhesion to collagen VI. The anti-GPIIb-IIIa was more effective in reducing adhesion and also nearly eliminated platelet aggregation. This indicates that the platelet GPIIb-IIIa receptor is not only absolutely required for aggregation, but is also involved in platelet adhesion onto collagen VI. In addition, a role for the GPIb receptor in platelet adhesion is indicated by our results. These results differ from those found in our laboratory using collagen I surfaces at a high shear rate of 1,500 S-'. Under those conditions, GPIb was found to be responsible for platelet adhesion to collagen I, and GPIIb- IIIa was found to be important only for subsequent platelet aggregation." It has recently been reported that the GPIa- IIa complex is also important in platelet adhesion to type VI collagen. l5

Results from the ATA experiments confirm our results found with the anti-GPIb. Although anti-GPIb was a better inhibitor of adhesion than ATA, the addition of either sub- stance to blood resulted in similar thrombus size distributions. Because ATA binds to vWF and inhibits vWF binding to GPIb, our ATA data are consistent with vWF mediation of the platelet GPIb-collagen VI adhesion interaction.

Results from experiments using blood from a severe vWD patient also indicate that platelet adhesion to collagen VI under low shear conditions is mediated by vWF. The initial experiment with vWD blood before treatment with cryoprecipitate and packed RBCs showed a reduction in platelet adhesion measured by the percentage of surface coverage as compared with a normal donor with the same hematocrit. Experiments with the same vWD patient's blood 1 day post- treatment with cryoprecipitate and packed RBCs indicated that the percentage of surface coverage and the total number of platelets deposited per surface area were within the 95% confidence range for normal donors. Blocking the GPIb platelet receptor was more effective in reducing adhesion than using blood from the vWD patient. This may be caused

by trace amounts of vWF (between 0 and 1.5 U/dL) present in the vWD blood that could mediate a small amount of platelet adhesion.

Because GPIIb-IIIa is also known to bind to vWF, this receptor may be involved in platelet adhesion to collagen VI through a vWF-mediated mechanism. GPIIb-IIIa receptors may also interact directly with the numerous RGD sequences in the triple helical portion of collagen VI. This could also explain the partial adhesion of severe vWD platelets to the collagen VI surface in the absence of vWF. Our finding that treatment with either anti-GPIb or anti-GPIIb- IIIa caused considerable inhibition of platelet adhesion to collagen VI at low shear may indicate that blocking one of these receptors inhibits binding at the other receptor sites. An alternative explanation would be that both GPIb and GPIIb-IIIa must normally be bound to vWF for optimal adhesion to occur under flow and that, by blocking either receptor, the mechanism of adhesion becomes much less effective.

Regardless of the precise mechanism involved, it is clear from our results that platelet adhesion to collagen VI at low shear occurs both via the GPIb receptor and the GPIIb-IIIa receptor and is mediated, at least in part, by vWF. Saelman et all5 recently reported that anti-GPIa MoAb also inhibits platelet adhesion to collagen VI. Preliminary results in our laboratory confirm this finding (data not shown). These data together suggest the possibility of a complex mechanism requiring GPIa-IIa, GPIb, and GPIIb-IIIa receptors for stable platelet adherence to collagen VI surfaces. However, the mechanism of aggregation subsequent to platelet adhesion onto collagen VI at low shear is clearly, from our results, mediated by the platelet GPIIb-IIIa receptor.

The ability of collagen VI to bind platelets at low-wall shear rates when platelet transport to the surface is low, suggests that this surface has a higher binding affinity for platelets than does collagen I under the same flow conditions. This study shows that the mechanisms that are responsible for platelet binding at a low-wall shear rate to collagen VI are different from those that occur at high-wall shear rates on collagen I, but that vWF plays a role in both situations. Because we have previously shown that collagen VI is lo- cated directly beneath the endothelial cell layer in human umbilical veins, that it binds vWF, and that it is in a physiological position to bind platelets after vascular injury,14 these findings also indicate the possible importance of collagen VI in the processes of hemostasis and thrombosis in low shear regions of the arterial and venous circulation.

ACKNOWLEDGMENT

The authors would like to thank Dorian Cross, RN, for assistance in obtaining umbilical cords, Drs Barry J. Potter and Robert Glanville for helpful advice, Nayana Pate1 and Nancy Turner for technical assistance, and Dr Elaine Schwartz for amino acid analyses.

REFERENCES

1. Rand JH, Sussman 11, Gordon RE, Chu SV, Solomon V: Local- ization of factor VIII-related antigen in human vascular subendothelium. Blood 55:752, 1980

2. Rand JH, Gordon RE, Sussman 11, Chu SV, Solomon V: Elec- tron microscopic localization of factor-VIII-related antigen in adult human blood vessels. Blood 60:627, 1982




1834 ROSS ET AL

3. Tschopp TB, Weiss HJ, Baumgartner HR: Decreased adhesion of platelets to subendothelium on von Willebrand’s disease. J Lab Clin Med 83:296, 1974

4. Sakariassen KS, Bolhuis PA, Sixma JJ: Human blood platelet adhesion to artery subendothelium is mediated by factor VIII-von Willebrand factor bound to the subendothelium. Nature 279:636, 1979

5. Stel HV, Sakariassen KS, deGroot PG, vanMourik JA, Sixma JJ: Von Willebrand factor in the vessel wall mediates platelet adherence. Blood 6535, 1985

6. Turitto VT, Weiss HJ, Zimmerman TS, Sussman 11: Factor VIII/von Willebrand factor in subendothelium mediates platelet adhesion. Blood 65:823, 1985

7. Weiss HJ: Effect of shear rate on platelet interaction with subendothelium in citrated and native blood. I. Shear rate-dependent decrease of adhesion in von Willebrand’s disease and the Bernard- Soulier syndrome. J Lab Clin Med 92:750, 1978

8. Turitto VT, Weiss HJ, Baumgartner HR: Platelet interaction with rabbit subendothelium in von Willebrand’s disease: Altered thrombus formation distinct from defective platelet adhesion. J Clin Invest 74:1730, 1984

9. Sakariassen KS: The role of platelet membrane glycoproteins Ib and IIb-IIIa in platelet adherence to human artery subendothelium. Br J Haematol 63:681, 1986

10. Alevriadou BR, Moake JL, Turner NA, Ruggeri ZM, Folie BJ, Phillips MD, Schreiber AB, Hrinda ME, McIntire LV: Real-time analysis of shear-dependent thrombus formation and its blockade by inhibitors of von Willebrand factor binding to platelets. Blood 81:1263, 1993

11. Wagner DD, Urban-Pickering M, Marder VJ: von Willebrand protein binds to extracellular matrices independently of collagen. Proc Natl Acad Sci USA 81:471, 1984

12. deGroot PG, Ottenhof-Rovers M, vanMourik JA, Sixma JJ: Evidence that the primary binding site of von Willebrand factor that mediates platelet adhesion on subendothelium is not collagen. J Clin Invest 82:65, 1988

13. Rand JH, Patel ND, Schwartz E, Zhou S-L, Potter BJ: 150- kD von Willebrand factor binding protein extracted from human vascular subendothelium is type VI collagen. J Clin Invest 88:253, 1991

14. Rand JH, Wu X-X, Uson RR, Potter BJ, Gordon RE: Co- localization of von Willebrand factor and type VI collagen in human vascular subendothelium. Am J Pathol 1425343, 1993

15. Saelman EUM, Nieuwenhuis HK, Hese KM, deGroot PG, Heijnen HFG, Sage EH, Williams S, McKeown L, Gralnick HR, Sixma JJ: Platelet adhesion to collagen types I through VI11 under conditions of stasis and flow is mediated by GPIa/IIa (a2pl-Inte- grin). Blood 85:1244, 1994

16. vonDerMark H, Aumailley M, Wick G, Fleiscbmajer R, Timpl R: Immunochemistry, genuine size and tissue localization of collagen VI. Eur J Biochem 142:493, 1984

17. Furthmayr H, Wiedemann H, Timpl R, Odermatt E, Engel J: Electron-microscopical approach to a structural model of intima collagen. Biochem J 211:303, 1983

18. Titani K, Kumar S , Takio K, Ericsson LH, Wade RD, Ashida K, Walsh KA, Chopek MW, Sadler JE, Fujikawa K Amino acid sequence of human von Willebrand factor. Biochem 25:3171, 1986

19. Bonaldo P, Russo V, Bucciotti F, Bressan GM, Colombatti A: a I Chain of chick type VI collagen. J Biol Chem 264:5575, 1989

20. Pareti FI, Fujimura Y, Dent JA, Holland LZ, Zimmerman TS, Ruggeri ZM: Isolation and characterization of a collagen binding domain in human von Willebrand factor. J Biol Chem 261:15310, 1986

21. Fujimura Y, Titani K, Holland LZ, Russel SR. Roberts JR, Elder JH, Ruggeri ZM, Zimmerman TS: von Willebrand factor: A

reduced and alkylated 52/48-kDa fragment beginning at amino acid residue 449 contains the domain interacting with platelet glycoprotein Ib. J Biol Chem 261:381, 1986

22. Fujimura Y, Titani K, Holland LZ, Roberts JR, Kostel P, Ruggeri ZM, Zimmerman TS: A heparin-binding domain of human von Willebrand factor. Characterization and localization to a tryptic fragment extending from amino acid residue Val-449 to Lys-728. J Biol Chem 262:1734, 1987

23. Titani K, Walsh KA: Human von Willebrand factor: The molecular glue of platelet plugs. Trends Biochem Sci 13:94, 1988

24. Chu M-L, Conway D, Pan T-C, Baldwin C, Mann K, Deutz- mann R, Timpl R: Amino acid sequence of the triple-helical domain of human collagen type VI. J Biol Chem 263:18601, 1988

25. Pytela R, Pierschbacher MD, Ginsberg MH, Plow EF, Ruos- lahti E: Platelet membrane glycoprotein IIb/IIIa: Member of a family of Arg-Gly-Asp-specific adhesion receptors. Science 23 1: 1559, 1986

26. Ruggeri ZM, DeMarco L, Gatti L, Bader R, Montgomery RR: Platelets have more than one binding site for von Willebrand factor. J Clin Invest 72:1, 1983

27. Nachman RL, Leung LLK: Complex formation of platelet membrane glycoproteins IIb and IIIa and fibrinogen. J Clin Invest 69:263, 1982

28. Gardner JM, Hynes RO: Interaction of fibronectin with its receptor on platelets. Cell 42:439, 1985

29. Thiagarajan P, Kelly K: Arg-Gly-Asp-dependent interaction of vitronectin with activated platelets. Blood 70:363, 1987

30. Dise CA, Bursh JW, Goodman DBD: Direct interaction of mepacrine with erythrocyte and platelet membrane phospholipid. J Biol Chem 257:4701, 1982

31. Kuo H-J, Keene D, Glanville RW: Orientation of type VI collagen monomers in molecular aggregates. Biochemistry 28:3757, 1989

32. Folie BJ, McIntire LV, Lasslo A: Effects of a novel antiplate- let agent in mural thrombogenesis on collagen-coated glass. Blood 72:1393, 1988

33. Prockop DJ, Udenfriend S : A specific method for the analysis of hydroxyproline in tissue and urine. Anal Biochem 1:228, 1960

34. Bergman I, Loxley R: Two improved and simplified methods for the spectrophotometric determination of hydroxyproline. Anal Chem 35:1961, 1963

35. Einarsson S , Josefsson B, Lagerkvist S : Determination of amino acids with 9-fluorenylmethyl chloroformate and reversed- phase high-performance liquid chromatography. J Chromatogr 282:609, 1983

36. Niiya K, Hodson E, Bader R, Byers-Ward V, Koziol JA, Plow EF, Ruggeri ZM: Increased surface expression of the membrane glycoprotein IIblIIIa complex induced by platelet activation. Rela- tionship to the binding of fibrinogen and platelet aggregation. Blood 70475, 1987

37. Handa M, Titani K, Holland LZ, Roberts JR, Ruggeri ZM: The von Willebrand factor-binding domain of platelet membrane glycoprotein Ib. J Biol Chem 261:12579, 1986

38. Ey PL, Prowse SJ, Jenkin CR: Isolation of pure IgGI, IgG2a and IgG2b immunoglobulins from mouse serum using protein A- Sepharose. Immunochemistry 15:429, 1978

39. Phillips MD, Moake JL, Nolasco L, Turner N: Aurin tricarboxylic acid: A novel inhibitor of the association of von Willebrand factor and platelets. Blood 72:1898, 1988

40. Moake JL, Turner NA, Stathopoulos NA, Nolasco LH, Hel- lums 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 78:1456, 1986





41. Hubbell JA, McIntire LV: Technique for visualization and analysis of mural thrombogenesis. Rev Sci Instrum 57:892, 1986

42. Abbassi 0, Kishimoto TK, McIntire LV, Anderson DC, Smith CW: E-Selectin supports rolling in vitro under conditions of flow. J Clin Invest 92:2719, 1993

43. Goldsmith HL, Turitto VT: Rheological aspects of thrombosis and haemostasis: Basic principles and applications. Thromb Haemost 55:415, 1986

44. Frangos JA, McIntire LV, Eskin SG: Shear stress induced stimulation of mammalian cell metabolism. Biotechnol Bioeng 32:1053, 1988

45. Wagner W R , Hubbell JA: Local thrombin sythesis and fibrin formation in an in vitro thrombosis model result in platelet recruit- ment and thrombus stabilization on collagen in heparinized blood. J Lab Clin Med 116:636, 1990

46. Alevriadou BR, Ware J, Ruggeri ZM: Use of laser scanning confocal microscopy to evaluate platelet thrombus formation in flowing blood. Annual Meeting of the American Institute of Chemi- cal Engineers, St Louis, MO, Nov 7-12, 1993 (abstr)

47. Alevriadou BR: Personal communication. 1994 48. Woolson RF: Statistical methods for the analysis of biomedi-

cal data. New York, NY, Wiley, 1987, p 513 49. Daniel WW: Biostatistics: A foundation for analysis in the

health sciences (ed 5). New York, NY, Wiley, 1991 , p 740 50. Wang N-HL, Keller KH: Augmented transport of extracellu-

lar solutes in concentrated erythrocyte suspensions in couette flow. J Col1 Interface Sci 103:210, 1985

5 1. Glagov S, Zarins C, Giddens D, Ku DN: Hemodynamics and atherosclerosis: Insights and perspectives gained from studies of human arteries. Arch Pathol Lab Med 112:1018, 1988




1995 85: 1826-1835

JM Ross, LV McIntire, JL Moake and JH Rand surfaces under physiological flow conditionsPlatelet adhesion and aggregation on human type VI collagen

http://www.bloodjournal.org/content/85/7/1826.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American


http://www.bloodjournal.org/content/85/7/1826.full.html

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

http://www.bloodjournal.org/site/subscriptions/index.xhtml





Documents

Platelet Adhesion and Aggregation on Human Type VI Collagen