Activation‐induced changes in platelet surface receptor expression and the contribution of the large‐platelet subpopulation to activation

Abstract Objective Platelet surface receptors are also present subcellularly in organelle membranes and can be expressed on the surface upon platelet activation. However, some receptors were reported to be decreased after activation. We analyzed the mechanism of activation‐dependent expression for different receptors. Methods Flow cytometry using platelet‐rich plasma or washed platelets was used to analyze receptor‐expression changes after platelet activation by glycoprotein (GP) VI–specific agonists, crosslinked collagen‐related peptide (CRP‐XL) and convulxin (Cvx), and thrombin. Platelets prelabeled with fluorescent antibody specific for a receptor were allowed to adhere on immobilized collagen or fibrinogen and post‐stained with antibody against the same receptor labeled with another fluorophore, allowing us to differentiate preexisting receptors from newly expressed receptors. Results Surface expression of αIIbβ3 increased in CRP‐XL–, Cvx‐, or thrombin‐stimulated platelets, but GPIb decreased due to shedding and internalization. Both total and dimeric GPVI increased in thrombin‐induced platelets, but decreased in platelets stimulated by Cvx, as a result of internalization. The larger platelets showed a greater increase in surface receptor (α2β1, αIIbβ3, GPVI, GPIb) expression upon activation compared to the smaller ones. Pre‐ and postlabeling with antibody specific for the same receptor, but conjugated with different fluorophores, allowed us to differentiate the receptors expressed on the surface of resting platelets from receptors newly exposed to the surface upon platelet activation. Conclusions Increased receptor expressions after activation are mainly manifested in the larger platelets. On platelets adhered on fibrinogen, the newly expressed receptors, especially GPVI, are localized in the lamellipodia of the spread platelets.


Essentials
• Activation-dependent platelet surface expression of different receptors was analyzed.
• Changes in surface expression depended on both the receptor and the platelet agonist.
• Newly expressed receptors localize on lamellipodia of platelets spread on fibrinogen.
• Increased receptor expressions upon activation are mainly manifested in larger platelets.

| INTRODUCTION
Platelets are anucleate small blood cells, but they have several intracellular organelles and membrane systems whose localization and morphology are changed upon platelet activation by various stimulants. Activation transforms the smooth disclike shape of resting platelets to a disturbed spherical shape with numerous filopodial extrusions and lamellipodia, accompanied by marked changes in subcellular organelle localization. Secretory dense granules and α-granules extrude their contents to the extracellular medium or through the inside space of the open canalicular system (OCS), and granule membranes fuse with the plasma or OCS membrane. [1][2][3][4] Major receptor proteins contained in the OCS and α-granule membranes, including glycoprotein (GP) Ib and αIIbβ3, 1,2 become exposed to the surface when their membranes fuse with the platelet plasma membrane. This may explain increased surface αIIbβ3 expression in activated platelets, 1,2 but a decrease in surface GPIb 5,6 upon activation suggests that other mechanisms may be involved.
Platelets are essential for primary hemostasis since they adhere to subendothelial collagen exposed by vessel injury, become activated, aggregate, and form a thrombus to arrest bleeding.
Hyperactive platelets, however, lead to formation of unwanted thrombi, which can detach and travel to distal areas, causing ischemic stroke or cardiovascular disease (CVD). Larger platelet size, measured as increased mean platelet volume (MPV), is a risk factor for cardiovascular disease CVD. 7,8 MPV increases with age in mice, which might explain the increasing CVD risk in the elderly. 9 Circumstantial evidence suggests that large platelets are more active, but there is yet no direct evidence for this and why this may be so.
The aim of the present study is to compare larger platelets with the whole platelet population in terms of their surface expressions of receptors involved in thrombus formation in response to platelet activation using a clinically available method, flow cytometry.
In resting platelets, surface expressions of GPIb, αIIbβ3, α2β1, and GPVI were higher in the larger platelets, commensurate with their larger surface area. Expressions of αIIbβ3 and α2β1 were increased in activated platelets, but GPIb and GPVI decreased due to shedding, internalization, or both. Increased exposure of intracellular receptors upon activation was most prominent in the larger platelets.
These results suggest that platelets are a heterogeneous population, not only with respect to size but importantly with respect to activity and that the large platelets are the main determinants of platelet activation and function. and crosslinked collagen-related peptide (CRP-XL) 13 were previously reported.
In each experiment, the whole platelet population (P1) of a donor's platelets was divided into 2 subpopulations according to size by gating in the forward scatter/side scatter (FSC/SSC) plot: The larger half was defined as P2 (larger platelets) and the smaller half was defined as P3 (smaller platelets; Figure 1A shows an example).
Numbers of platelets in each subpopulation are calculated as percentage of total events.
Formaldehyde/glutaraldehyde solution was used, as it has less effect on cell membrane phospholipid distribution 14 and better maintained platelet antibody binding.

| Statistical analysis
Paired Student's t-test (Prism v8.9, GraphPad Software, La Jolla, CA, USA) was applied to compare differences between activated platelets and their corresponding resting platelets.

| Large platelets have higher levels of surface receptor expression
From the SSC/FSC scatter plot, resting platelets of each donor were gated into the larger platelets (P2) of the total population (P1; Figure 1A, resting platelets). Because each donor's platelets showed a different size distribution, gating was tailored to each donor's FSC/ SSC plot. P2 of resting platelets showed 1.5-fold higher binding of anti-GPVI dimer (204-11 Fab) than P1. CRP-XL activation slightly changed the scatter plot ( Figure 1A, activated) and P2, defined by the same gating as used for resting platelets, exhibited 1.9-fold higher 204-11 binding than P1.
P2/P1 of control resting platelets was about 130%, showing that P2 bound more control IgG than P1, indicating that P2 platelets have a larger cell surface area. For all receptors, except for total GPVI, P2/P1 ratios were similar to that of the control.

| Receptor expression after platelet activation
Surface receptor expressions in platelets activated by CRP-XL, Cvx, and thrombin were determined and compared to expression of each receptor in resting platelets ( Figure 2). For the total platelet population (P1) of PRP (Figure 2A), both total GPVI and GPVI dimer very markedly decreased after stimulation by Cvx (GPVI dimer: P = .03, n = 5; total GPVI: P = .000, n = 4), but no significant changes in expression of either in response to CRP-XL or thrombin.   and thrombin (P = .08, n = 5).
If we compare the differences in receptor expression in the P2 population of PRP and washed platelets ( Figure 2B,D), which has higher expression of each receptor than the total population (P1), the changes in surface receptor expression upon activation are even more dramatic.
To confirm these results, we determined receptor expression in stimulated washed platelets that were subsequently fixed with formaldehyde/glutaraldehyde; 204-11 Fab was not used in these experiments, as it does not react with fixed platelets. These activated and fixed platelets showed similar receptor changes; Cvx, especially, strongly reduced GPVI expression ( Figure 3A).

| Time course and agonist-concentration dependency of receptor expression in activated platelets
Flow cytometry of activated platelets ( Figure 3B) showed that total Thrombin-induced changes in expressions of GPVI dimer, P-selectin, and activated αIIbβ3 in P1-P3 were measured using 204-11 Fab, anti-CD62P, and PAC-1 antibodies, respectively.
All 3 receptors increased dose dependently with thrombin, but increases in P2 were much greater than those in P1 and P3 ( Figure 4A). Increase of GPVI dimer expression in P3 platelets was much less, and expressions of CD62P and active αIIbβ3 in P2 were more than 2-fold those in P3. Since anti-CD62P and PAC-1 bind only to activated platelets, these results indicate that the F I G U R E 3 Analyses of receptor expressions using fixed platelets. A, Receptor expressions in fixed platelets. After platelets were activated, as described in Figure 2, they were fixed, stained using specific antibodies for the indicated receptors, and receptor expressions determined by flow cytometry. The increase or decrease of specific receptors in fixed platelets were similar to those obtained with native platelets. GPVI expression is decreased by convulxin (Cvx; P = .003, n = 8), but increased by crosslinked collagen-related peptide (CRP-XL) and thrombin (P = .003 and P = .003, respectively, n = 8). Integrin α2β1 tended to increase, although not reaching statistical significance, in platelets activated by CRP-XL (P = .03, n = 4), Cvx (P = .13, n = 3), or thrombin (P = .06, n = 3). αIIbβ3 increased after the activation by CRP-XL (P = .002, n = 4) or Cvx (P = .03, n = 3) and tended to increase in platelets activated by thrombin (P = .12, n = 3). The expression of GPIb decreased significantly in platelets activated by Cvx (P = .02, n = 3) and tended decrease in platelets activated by CRP-XL (P = .19, n = 4) or thrombin (P = .07, n = 3). The coulumms show the means of the measurements and the error bars shows their SD. *P ≤ .05, **P ≤ .005, ***P ≤ .0005, ****P < .0001. B, Time course of the change in GPVI expression after activation. Washed platelets were activated by thrombin (0.5 units/mL) or Cvx (1 μg/mL) and incubated at 37°C. After the indicated time, aliquots of platelets were removed, fixed, and then analyzed by flow cytometry for total GPVI using 1G5 antibody, as described in the Methods section. Expression of total GPVI is increased by thrombin and decreased by Cvx in a rapid time course

| Analysis of GPIb and GPVI shedding
Platelet activation by CRP-XL, Cvx, or thrombin decreased surface GPIb

| Internalization of GPIb and GPVI in activated platelets
Platelets r e s t i n g r e s t i n g r e s t i n g

| Exposure of intracellular receptors on platelet spreading
Our flow cytometry data suggest that additional receptors are newly exposed on the platelet surface after platelet activation. Preexisting surface receptors were differentiated from newly exposed ones

| DISCUSSION
This study demonstrates that platelet activation changes the levels  In contrast to GPVI, many agonists have been reported to decrease GPIb expression, 21,24 suggesting platelet activation in general would cause GPIb internalization. Electron micrographs suggested cytoskeletal involvement in GPIb internalization. [19][20][21] However, αIIbβ3 binds to the cytoskeleton as well and is clustered after activation but does not internalize, so the precise mechanism of internalization remains inconclusive.
Using prestaining and post-staining with receptor-specific antibodies labeled with fluorophores that emit fluorescence of different wavelengths, we differentiated preexisting surface receptors in resting platelets from intracellular receptors newly exposed upon platelet activation. Prestained GPVI on fibrinogen-adhered platelets localizes to the cell body in the center of the spread cell, while post-stained GPVI localizes at the edges of the spread cell ( Figure 8A). However, platelets adhered on collagen show colocalization of prestained and post-stained GPVI clustered along the collagen fibers, as previously reported. 15 These results indicate that the flat, spread membrane of fibrinogen-adhered platelets would come from an intracellular resource, with newly exposed GPVI dispersed freely over the lamellipodia, but originally existing surface GPVI would remain at the same place. However, because of GPVI's high affinity for collagen, prestained (green) and post-stained GPVI (red) would move to the region of the membrane in contact with collagen fibers and become colocalized as membrane clusters on the fibers (yellow; Figure 8B). Distribution of prestained and post-stained GPIb on platelets adhered to immobilized fibrinogen and collagen are essentially like that of GPVI on platelets adhered on fibrinogen, but since these platelets are not fully spread like the GPVI-stained platelets, many of the spread cells show only a small separation of post-and prestaining. These results suggest that intracellular receptors become exposed in the spread membrane (lamellipodia) of adhered platelets and localize separately from the originally expressed surface receptors when they do not interact with substrate.
The P2/P1 ratio is increased upon activation, consistent with the larger platelets expressing more receptors on their surface compared to the smaller platelets. These results suggest that larger platelets are more activated than smaller platelets, as reported before. [25][26][27][28] Upon thrombin activation, P-selectin and PAC-1 expressions in the larger platelets are increased much more than in the total and smaller platelets ( Figure 4A). At higher thrombin concentrations, P-selectin and PAC-1 expressions in P2 became about 2-fold and 3-fold higher than that of P1 and P3, respectively. Since gating by flow cytometry cannot totally separate the larger platelets from the smaller platelets, these results suggest that activation involving integrin activation and secretion would mainly occur in the larger platelet population.
The large-platelet fraction isolated by differential centrifugation or flow cytometry was reported to be more active, [25][26][27] and a relationship between activity and platelet size was reported. 28 Although young platelets were thought to be large and more active, platelet size and age were indicated to independently affect platelet function. 29,30 Heterogeneity of platelet size was suggested to come from the heterogeneity of megakaryocytes 31 and activation of the mechanistic target of rapamycin complex1 in megakaryocytes may contribute to elevated platelet volume, 9 suggesting that megakaryocytes would produce different sizes of platelets and that the larger platelet subpopulation would have higher activities. Our data indicate that the larger subpopulation of activated platelets showed a higher level of agonist-induced receptor expression compared to the total platelet population, supporting the previous reports describing that larger platelets have higher activities. In this context, it is very interesting that a larger MPV has been indicated to be a risk factor for cardiovascular disease in both men and women 7,8,32 or just in men. 33 Our results importantly suggest that the large-platelet subpopulation would be the main contributor to higher platelet activity and our method, analyzing the receptor expression of resting and activated platelets by flow cytometry, would be a good method to analyze the risk factor of having larger platelets.