© 2003 Schattauer GmbH, Stuttgart
Platelets and Blood Cells
Affinity and kinetics of P-selectin binding to heparin
Jian-Guo Wang, Jian-Guo Geng
Laboratory of Molecular Cell Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences,
Chinese Academy of Sciences, Shanghai, China
Summary
P-selectin (CD62P), expressed on stimulated endothelial cells
and activated platelets, reacts with P-selectin glycoprotein
ligand-1 (PSGL-1, CD162) for leukocyte rolling. It also binds to
heparin and heparan sulfate proteoglycans (HSPGs), which
attenuates P-selectin mediated adhesions of leukocytes and
cancer cells. Here we report that P-selectin mediated adhesion, but not rolling, of the HSPGs bearing human malignant
melanoma A375 cells under shear stress. To understand its
underlying molecular mechanism, we measured the biophysical
properties of this interaction. Heparin inhibited the adhesion
of A375 cells to immobilized P-selectin under flow (IC50 = 3 µM
heparin) and neutralized the binding of P-selectin to A375 cells
(IC50 = 4 µM heparin). Using surface plasmon resonance technique, we found that P-selectin bound to heparin with a dissociation constant (Kd) of 115 ± 6 nM. The measured off rate (koff)
was 3.15 ± 0.34 × 10-3 s-1 and the calculated on rate (kon) was
2.75 × 104 M-1 s-1. Taken together, our data suggest that the
very slow koff and the reduced kon, but apparently not the Kd, are
responsible for adhesion, but not rolling of A375 cells, to Pselectin under flow.
Keywords
P-selectin, heparin, heparan sulfate proteoglycans, affinity,
kinetics
Thromb Haemost 2003; 90: 309–16
Introduction
In inflammatory responses, leukocytes are recruited to the site
of infection or tissue injury, where they function as “a doubleedged sword”, either fighting against invading pathogens or
causing own tissue damage. The recruitment of leukocytes
involves a cascade of cellular events, including rolling, weak
and firm adhesion, diapedesis, transendothelial migration and
chemotaxis. The binding of selectins (CD62) to their cognate
glycoprotein ligands mediates rolling and weak adhesion of leukocytes. Consequently, the interaction of integrins with the
immunoglobulin superfamily of cell adhesion molecules mediates firm adhesion and signal transduction, which eventually
triggers diapedesis and transendothelial migration of leukocytes. The emigrated leukocytes are then guided by several large
Correspondence to:
Dr. Jian-Guo Geng
Laboratory of Molecular Cell Biology
Institute of Biochemistry and Cell Biology
Shanghai Institutes for Biological Sciences
Chinese Academy of Sciences, Cell Biology Building Room 505
320 Yue-Yang Road
Shanghai 200031, China
Tel.: 086-21-54921351, Fax: 086-21-54921435
E-mail: [email protected]
families of soluble chemoattractants to migrate to their destinations where the insult occurs. It is generally believed that the
selectin family of cell adhesion molecules mediates the first step
of cell-cell interactions during the recruitment of leukocytes
(1-5).
P-selectin is a member of the selectin family of cell
adhesion molecules. It is a pre-synthesized protein stored in the
Weibel-Palade bodies of endothelial cells and the α–granules of
platelets. Upon inflammatory and thrombogenic challenges,
P-selectin rapidly translocates from these intracellular granules
to the cell surfaces of endothelial cells and platelets by exocytosis (in seconds). Further, P-selectin can be up-regulated by de
novo synthesis in the cytokine stimulated endothelial cells (in
hours; 1-5). L-selectin (CD62L) and β-2 leukocyte integrin,
Mac-1 (αMβ2, CD11b/CD18, CR3), are constitutively expressed
Received January 22, 2003
Accepted after revision May 13, 2003
Financial support:
This study was funded by grants from Chinese Academy of Sciences (KSCX2-2-02),
National Science Foundation of China (39925015, 30130090 and 39970794), Special
Funds for Major State Basic Research of China (G1999053907) and 973 Program
(2002CB513000) from Chinese Ministry of Science and Technology.
DOI: 10.1160/TH03-01-0045
309
Wang, Geng: Characteristics of P-selectin/heparin interaction
receptors on most leukocytes. After leukocyte activation,
L-selectin is shed from the cell surface while Mac-1 transforms
from the resting state (low affinity) to the activated state
(high affinity) for its ligands, such as ICAM-1 (Intercellular
Cell Adhesion Molecule-1, CD54) and fibrinogen.
Heparin and its analog, HSPGs, have been shown to bind to
cell adhesion molecules, such as P-selectin, L-selectin and
Mac-1. These bindings can inhibit adhesion of leukocytes
mediated specifically by these cell adhesion molecules in vitro
and in vivo (6-12). Recently, we have shown that the cell surface
HSPGs could directly mediate adhesion of human malignant
melanoma A375 cells and human tongue squamous cancer
Tca-8113 cells to P-selectin under flow (13). However, we
observed that although human promyeloid HL-60 cells rolled on
and adhered to P-selectin (a P-selectin/PSGL-1 pair), A375
cells and Tca-8113 cells only adhered to, but not rolled on,
P-selectin (a P-selectin/heparin pair) under flow. Similar manifestations were also reported previously (14, 15). To understand
the mechanisms governing for these distinctive cellular behaviors on the same substrate of P-selectin, we sought to
measure the binding affinity and kinetics of the interaction of
P-selectin with heparin, which might provide a biophysical
explanation for these divergent cellular phenotypes.
Materials and methods
Materials
The same lot of heparin (sodium salt grade I-A from porcine
intestinal mucosa, Cat. No. H3393, Lot No.98H0713, Sigma)
was used throughout the entire experimental study. Its molecular mass was 19.5 kDa and its sulfate contents (the main constituent for its charges) were 11.75% absolute content of total sulfate and 3.25% absolute content of N-sulfate (each heparin
chain was calculated to have 19.8 N-SO42- and 51.8 O-SO42anions), as determined before (16).
Proteins and antibodies
P-selectin receptor-globulin (P-selectin Rg; constructed by
fusing the lectin domain, the epidermal growth factor-like
domain and the first two complement protein-like repeats of
P-selectin with the Fc portion of human IgG1) was prepared as
before (17). G1 (a leukocyte adhesion blocking IgG1 mAb
against P-selectin) and PS1 (a leukocyte adhesion non-blocking
IgG1 mAb against P-selectin) were characterized as previously
described (13, 18). F(ab’)2 fragments of G1 and PS1 mAbs were
prepared using ImmunoPure F(ab’)2 Preparation Kit (Pierce,
Rockford, IL).
Cell culture
Human cell lines of promyeloid HL-60 cells (CCL 240) and
malignant melanoma A375 cells (CRL 1619) were purchased
from American Tissue Culture Collection (Rockville, MD). They
were cultured in RPMI 1640 medium (GIBCO BRL, Shanghai,
China) supplemented with 10% heat inactivated newborn bovine
calf serum (BCS), 4 mM L-glutamine, 100 units/ml penicillin
and 100 µg/ml streptomycin at 37oC in the presence of 5% CO2.
SDS-PAGE and gel filtration chromatography
The purified P-selectin Rg was resuspended in the SDS sample
buffer with or without 2% β-mercaptoethanol and boiled for
5 min. Samples were loaded on 7% SDS-polyacrylamide gel
(10 µg/lane) for electrophoresis. They were stained with
Coomassie brilliant blue. To determine the apparent molecular
mass of P-selectin Rg in a non-denatured condition, P-selectin
Rg (50 µg) was resuspended in 10 µl PBS (phosphate buffered
saline, pH 7.4) and injected to a Superdex 200 column (Smart
System; Amersham Pharmacia Biotech, Uppsala, Sweden) that
had been pre-equilibrated with PBS. Samples were run at
50 µl/min at 22°C. The gel-filtration chromatography was
calibrated by blue dextran (~2,000 kDa), horse spleen apoferritin (~443 kDa), sweet potato β-amylase (~200 kDa), yeast alcohol dehydrogenase (~150 kDa), and phenol red (0.354 kDa) (all
protein markers were purchased from Sigma, St. Louis, MO).
Laminar flow assay
Polystyrene slides were coated with 2 ml human IgG (Sigma)
or P-selectin Rg (both at 0.1 µg/ml) in 20 mM Tris-HCl, pH 9.5,
140 mM NaCl and 0.02% NaN3 at 4°C overnight and blocked
with 3% bovine serum albumin (BSA) in PBS at 22°C for 2 h.
Slides were mounted on the stage of an inverted phase contrast
Olympus microscope (Olympus Optical Co. Ltd., Tokyo,
Japan). It was connected to a time-lapse videocassette recorder
STLV-24P (Samsung Electronics, Suwon, Korea) using a
Panasonic color CCTV camera wv-GP410/G (Matsushita
Communication Industrial Co. Ltd., Okasa, Japan). HL-60 cells
and A375 cells were washed and resuspended at 0.5 × 106/ml in
PBS supplemented with 10 mM HEPES, pH 7.4, 2 mM CaCl2
and 2 mM MgCl2. They were injected through the flow chamber at 0.5 dyne/cm2 using a syringe pump at 22oC. The potential
interactions between the cellular Fc receptors and the Fc domain
of P-selectin Rg were eliminated by preincubation of the cells
with 10 µg/ml of human IgG at 22oC for 20 min. The adhesive
cells were quantified from videotape recordings of 10-20 fields
of view obtained (3 min after flowing cells through the chamber) while scanning the lower plate of the flow chamber using a
10× objective lens. For analysis of rolling velocity, the critical
velocity was defined as the velocity of a non-interacting cell in
a shear flow near the wall of the flow chamber, cells with a
translational velocity lower than the critical velocity were
defined as rolling (19). For antibody inhibition experiments, the
immobilized P-selectin Rg was preincubated with 0.3 µg/ml
G1 F(ab’)2 or PS1 F(ab’)2 at 22°C for 20 min (13). For EDTA
inhibition experiment, 2 mM EDTA, instead of 2 mM CaCl2
and 2 mM MgCl2, was used in the running buffer.
310
Wang, Geng: Characteristics of P-selectin/heparin interaction
Site density determinations
The density of the immobilized P-selectin Rg was determined
using 125I-1abeled PS1 (20). Briefly, 96-well culture plates were
coated with P-selectin Rg (40 µl/well, 0.1 µg/ml) and blocked
as described above. The 125I-1abeled PS1 (40 µl/well, 2 µg/ml)
in PBS containing 0.1% BSA was incubated for 1 h at 22°C.
The wells were washed five times and then counted in a γ
counter. Specific bindings were obtained after subtraction of
the bindings to the BSA-coated wells. Site densities were
calculated assuming the monovalent binding of the antibody at
saturation (20). All assays were performed in quadruplicate in
three separate experiments.
Flow cytometric assay
The adherent A375 cells were detached by PBS containing
0.02% EDTA (Versene; GIBCO BRL), washed once with
PBS, and resuspended in PBS/BSA (PBS supplemented with 2
mM CaCl2, 2 mM MgCl2 and 1% BSA; 1 ×106 cells/ml). Each
aliquot (0.5 ml) of cells was incubated with 1 µg human IgG or
P-selectin Rg followed by 1 µg of an FITC-conjugated Ab
against human IgG (Pierce) at 22oC for 1 h with end-to-end
rotation. Cells were centrifuged at 1,500 rpm for 5 min and
supernatants were discarded. For antibody inhibition experiments, 1 µg P-selectin Rg was preincubated with 3 µg of G1
F(ab’)2 or PS1 F(ab’)2 in 50 µl PBS/BSA at 22oC for 30 min.
For heparin inhibition experiments, 1 µg P-selectin Rg was preincubated with the indicated amounts of heparin in 50 µl
PBS/BSA at 22oC for 30 min. Each aliquot was then resuspended in 0.5 ml PBS/BSA for immediate flow cytometric analysis
(FACScan; Becton Dickinson & Co., Mountain View, CA).
Heparin conjugation
Unfractionated heparin was end-conjugated to BSA by a
method described for the synthesis of cellobiose conjugates
(21, 22). Briefly, 9.12 mg heparin and 34 mg BSA were
dissolved in 5 ml of 200 mM potassium phosphate buffer, pH
8.0. Sodium cyanoborohydride (25 mg) was then added and
the mixture was incubated at 37°C for 2 days. The mixture was
dialyzed thoroughly against H2O, lyophilized to the dryness
and reconstituted in 250 mM bicarbonate buffer, pH 8.0. The
resulting high molecular weight complex was separated from
unconjugated BSA and free heparin, by gel-filtration chromatography, on a Sepharose 4B column (1.6 × 100 cm; Amersham
Pharmacia Biotech). As a mock control, BSA (without heparin)
was treated with sodium cyanoborohydride under identical
conditions, as described above.
Surface plasmon resonance measurements
The mock BSA and the heparin-conjugated BSA were biotinylated with N-hydroxysuccinimidobiotin (Sigma), as previously described (23). Surface plasmon resonance experiments
were conducted at 22°C with a BIAcoreTM biosensor
Figure 1: Characterization of recombinant human P-selectin Rg.
A). Samples of the purified P-selectin Rg (10 µg/lane) were
boiled for 5 min in the presence (R) or absence (N) of 2% βmercaptoethanol (v/v).They were analyzed by 7% SDS-PAGE followed by Coomassie brilliant blue staining. B). P-selectin Rg was
loaded on a Superdex 200 column (Smart System;
Amersham Pharmacia Biotech) for gel- filtration chromatography.Vo and Vi were determined by separate injections of blue
dextran (~2,000 kDa) and phenol red (0.354 kDa), respectively.
Further, apoferritin (~443 kDa), β-amylase (~200 kDa) and
alcohol dehydrogenase (~150 kDa) were used for determination
of their respective elution profiles.
(Pharmacia Biosensor, Uppsala, Sweden) that was upgraded to
model 1000 specifications. The running buffer was 10 mM
HEPES, pH 7.4, 150 mM NaCl, 2 mM CaCl2, 2 mM MgCl2,
0.005% detergent-P20, which had been passed through 0.2 µm
filters and degassed. The biotinylated mock BSA and the biotinylated heparin-conjugated BSA were captured on SA5 sensor
chips containing immobilized streptavidin on their surfaces at a
flow rate of 10 µl/min for 1 min. A typical immobilization
resulted in the coupling of the heparin-conjugated BSA at a
311
Wang, Geng: Characteristics of P-selectin/heparin interaction
Figure 2: Adhesion, but not rolling, of A375 cells to P-selectin.The washed HL-60 cells and A375 cells were resuspended in PBS containing 2 mM CaCl2 and 2 mM MgCl2.They were perfused through the slides immobilized with P-selectin Rg in a parallel plate laminar
flow chamber. For antibody inhibition experiments, P-selectin Rg was preincubated with G1 F(ab’)2 (a leukocyte adhesion blocking
mAb to P-selectin) or PS1 F(ab’)2 (a leukocyte adhesion non-blocking mAb to P-selectin). For chelation experiments, EDTA was used
in all the incubation and washing buffers. All results of cell adhesion (A and B) and rolling (C and D) were expressed as the mean ±
S.D. values, determined in ten to twenty measurements of four to five separate experiments.
density of 0.4 ng/mm2 (400 RU). For measurements of binding
affinity and kinetics, P-selectin Rg, at various concentrations,
were injected at a flow rate of 20 µl/min for 30 sec, followed by
a 1-min dissociation phase, when only running buffer was
passed over the surface. Identical results were obtained when
different amounts of P-selectin Rg were injected at 10 µl/min or
when different lots of P-selectin Rg preparations were used
(data not shown). The bound P-selectin Rg was removed by
injection of 5 µl of 10 mM EDTA when less than 50 nM
P-selectin Rg was used or 5 µl of 20 mM SDS when more than
50 nM P-selectin Rg was used. For antibody inhibition experiments, 3 µg /ml P-selectin Rg was preincubated with 9 µg/ml
G1 F(ab’)2 or PS1 F(ab’)2 at 22°C for 20 min. For EDTA inhibition experiment, 2 mM EDTA, instead of 2 mM CaCl2 and
2 mM MgCl2, was used in the running buffer. Data on equilibrium binding were analyzed by the nonlinear curve fitting of
Langmuir binding isothern to the primary data using
BIAevaluation software (Pharmacia Biosensor) and by the
linear regression analysis of Scatchard plots. Recombinant
P-selectin Rg used here was calculated as a dimer (24, 25),
according to the determination data of gel-filtration chromatography (Fig. 1B).
Results
Characterization of recombinant P-selectin Rg
In this study, we investigated the biophysical properties for
the interaction of P-selectin with heparin. Figure 1A shows
the Coomassie brilliant blue staining of recombinant human
P-selectin Rg following electrophoresis on a polyacrylamide
gel. It had apparent molecular masses of ~85 kDa under reducing conditions and ~170 kDa under non-reducing conditions.
Figure 1B illustrates a single peak of the gel-filtrated P-selectin
Rg on a Superdex 200 column (Smart System). This
peak was larger than β–amylase (~200 kDa), but smaller than
apoferritin (~443 kDa), indicating that P-selectin Rg was a
dimeric molecule. No earlier elution peaks were observed,
suggesting that there had no detectable oligomers of P-selectin
Rg existed. To further make sure that P-selectin Rg used in
this study was not oligomeric, all binding experiments were
312
Wang, Geng: Characteristics of P-selectin/heparin interaction
Figure 3: Heparin inhibits adhesion of A375 cells to P-selectin
under flow.The immobilized P-selectin Rg was preincubated with
various amounts of heparin as indicated. HL-60 cells (A) and
A375 cells (B) in the presence of the same amounts of heparin
were perfused through the slides pretreated with the same
amount of heparin, as described above. All results were
expressed as the mean ± S.D. values determined in ten to twenty measurements of four to five separate experiments.
performed using the peak fractions of P-selectin Rg within 48 h
of the gel filtration chromatography.
Adhesion, but not rolling, of A375 cells
to P-selectin
We then compared the cellular behaviors of human promyeloid
HL-60 cells and human malignant melanoma A375 cells on
immobilized P-selectin Rg (the calculated density was 24.7 ±
0.9 molecules/µm2) under flow conditions. As shown in Figures
2A and B, HL-60 cells and A375 cells adhered avidly to immobilized P-selectin Rg under shear stress. These adhesions of HL-
Figure 4: Heparin inhibits the binding of P-selectin to A375
cells. A). A375 cells were incubated with human IgG or P-selectin Rg (designated as P-Rg) and a FITC-conjugated Ab against
human IgG. For antibody inhibition experiments, P-selectin Rg
was preincubated with G1 F(ab’)2 (a leukocyte adhesion blocking
mAb to P-selectin) or PS1 F(ab’)2 (a leukocyte adhesion nonblocking mAb to P-selectin).The binding events were analyzed
by flow cytometry (FACScan; Becton Dickinson & Co., Mountain
View, CA). B). P-selectin Rg was preincubated with heparin at
various concentrations, prior to addition of the washed A375
cells and a FITC-conjugated Ab against human IgG. Results were
presented as histograms of the log fluorescence intensities from
104 cells from the representative of three to five independent
experiments.
60 cells and A375 cells were abrogated by G1 F(ab’)2 (a leukocyte adhesion blocking mAb to P-selectin) and EDTA (a chelator for divalent cations), but not by PS1 F(ab’)2 (a leukocyte
adhesion non-blocking mAb to P-selectin). Interestingly, only
HL-60 cells, but not A375 cells, could roll avidly on immobilized P-selectin Rg under flow (Figs. 2C and D). Again, the rolling of HL-60 cells on immobilized P-selectin Rg was abolished
by G1 F(ab’)2 and EDTA, but not by PS1 F(ab’)2. Figures 3A
and B show that, in this laminar flow assay, heparin inhibited
adhesions of HL-60 cells and A375 cells to P-selectin Rg in
dose-dependent manners (IC50 = 10 M heparin for HL-60 cells
313
Wang, Geng: Characteristics of P-selectin/heparin interaction
Figure 5: Affinity and kinetics of
P-selectin binding to heparin.
A). Heparin-conjugated BSA was
biotinylated and captured on a sensor
chip SA5 containing immobilized
streptavidin on its surface. P-selectin
Rg (designated as P-Rg) was injected
over the sensor surfaces and the
sensographs were recorded. For antibody inhibition experiments, P-selectin
Rg was preincubated with G1 F(ab’)2
(a leukocyte adhesion blocking mAb
to P-selectin) or PS1 F(ab’)2 (a leukocyte adhesion non-blocking mAb to
P-selectin). For chelation experiments,
2 mM EDTA was used in the running
buffer. B). P-selectin Rg, at different
concentrations, was injected through
the sensor chips captured with the
mock BSA and the heparin-conjugated
BSA in parallel.The equilibrium affinity
of P-selectin binding to heparin was
plotted with the subtracted equilibrium responses (Req) versus the
concentrations of P-selectin Rg. C).
The dissociation constant derived
from nonlinear curve fitting. D).
Scatchard plot analysis.
and IC50 = 3 M heparin for A375 cells). These results confirmed the earlier reports from others (14, 15) that unlike HL-60
cells, A375 cells could adhere to, but could not roll on, P-selectin.
They further suggested a reasonably high affinity for the interaction of P-selectin with heparin. It should be pointed out that
although G1 F(ab’)2 and EDTA inhibited the adhesions of HL-60
cells (Fig. 2A) and A375 cells (Fig. 2B) to the base lines, heparin,
at as high as 1 mM concentration, apparently did not completely abrogate their adhesions to immobilized P-selectin (Fig. 3A
and B), indicating that there had the multifaceted and/or nonspecific effects of heparin in this laminar flow assay system.
Heparin inhibits P-selectin binding to A375
cells
To support the above findings, we employed a cell-surface
binding assay to measure the inhibitory effects of heparin on the
binding of P-selectin Rg to A375 cells. In this assay, a FITCconjugated Ab to human IgG was used to report the binding
events by flow cytometry. Figure 4A shows that compared to
human IgG, P-selectin Rg bound robustly to A375 cells.
Preincubation of P-selectin Rg with G1 F(ab’)2 (a leukocyte
adhesion blocking mAb to P-selectin), but not with PS1 F(ab’)2
(a leukocyte adhesion non-blocking mAb to P-selectin), attenuated this binding, demonstrating the specificity for this binding
assay. Using this assay, we carried out the dose course of heparin inhibition experiments. Fig. 4B shows that heparin potently
interfered with the binding of P-selectin Rg to A375 cells
(IC50 = 4 M heparin).
Kinetics of P-selectin binding
to heparin-conjugated BSA
Using surface plasmon resonance technique, we further measured, in real-time, the affinity and kinetics of P-selectin binding
to heparin-conjugated BSA. In our preliminary experiments, we
found that it was difficult to covalently couple heparinconjugated BSA in a sufficient quantity to a CM5 sensor chip
through its primary amino groups (data not shown). Thus,
we had to biotinylate the heparin-conjugated BSA and then
to capture it on a SA5 sensor chip carrying immobilized
streptavidin on its surface.
Injection of P-selectin Rg over the heparin-conjugated BSA
surface, but not over the unconjugated BSA surface, caused a
314
Wang, Geng: Characteristics of P-selectin/heparin interaction
binding response (Fig. 5A). In contrast, injections of human IgG
over the heparin-conjugated BSA surface did not induce a
significant binding response (data not shown). Preincubation of
P-selectin Rg with G1 F(ab’)2 (a leukocyte adhesion blocking
mAb to P-selectin), but not PS1 F(ab’)2 (a leukocyte adhesion
non-blocking mAb to P-selectin), blocked the binding. Further,
EDTA, a chelator for divalent cations, also abrogated this
binding.
As the binding detectable to BIAcore was presented as
a change in refractive index, the rapid kinetics of P-selectin
binding to heparin-conjugated BSA might be difficult to distinguish them from bulk refractive changes, especially when high
protein concentrations were used. We therefore conducted
parallel injections of P-selectin Rg on both surfaces of the heparin-conjugated BSA and the unconjugated BSA (as the
experimental control). Each of them was individually captured
on a SA5 sensor chip following the biotinylation. In this
manner, the response units (RU) attributed to bulk refractive
index on the unconjugated BSA surface could be subtracted
from the RU values measured on the heparin-conjugated BSA
surface for the specific bindings. It should be mentioned that
the RU values measured on the unconjugated BSA were
always much lower than those on the heparin-conjugated BSA
(Fig. 5B).
The equilibrium affinity of P-selectin binding to heparin
was plotted with the subtracted equilibrium responses (Req)
versus the concentrations of P-selectin Rg. The dissociation
constant (Kd = 115 ± 6 nM) was derived from nonlinear curve
fitting (Fig. 5C). Alternatively, Scatchard plot analysis (Kd =
117 ± 5 nM) was used (Fig. 5D). Both methods yielded very
similar results. The measured koff was 3.15 ± 0.34 × 10-3 s-1 and
the calculated kon was 2.75 ± 0.42 × 104 M-1 s-1.
Discussion
Heparin and HSPGs have been shown to bind to P-selectin and,
consequently, to inhibit the adhesions of leukocytes and cancer
cells to P-selectin. We previously reported that the cell surface
HSPGs directly mediated adhesion of human malignant melanoma A375 cells and human tongue squamous cancer Tca-8113
cells to P-selectin (13). Here we showed that P-selectin mediated adhesion, but not rolling, of A375 cells under shear stress.
Although others (14, 15) also reported the similar findings, its
molecular mechanism remains to be explored. In this context, it
has been shown that sialyl Lewisx mediates the rolling on
P-selectin while tyrosine sulfation, although not required,
enhances it (26, 27). Since both A375 cells and Tca-8113 cells
have no apparent expression of sialyl Lewisx on their cell
surfaces (13), our results appear to support this finding.
In the present study, we determined the potencies of heparin for inhibition of A375 cell adhesion to P-selectin under flow
(IC50 = 3 M heparin) and for attenuation of P-selectin binding
to the cell surface of A375 cells (IC50 = 4 M heparin). Using
the technique of surface plasmon resonance, we further examined the affinity and kinetics of P-selectin binding to heparin
(Kd = 115 ± 6 nM, koff = 3.15 ± 0.34 × 10-3 s-1 and kon =
2.75 × 104 M-1 s-1). Apparently, these very divergent approaches yielded the relatively convergent results for the
binding affinity of P-selectin to heparin.
It was worth noting that the Kd value of 115 ± 6 nM
measured by the technique of surface plasma resonance was
~20-fold lower than those determined by other methods, such as
the laminar flow assay (3 µM heparin; Fig. 3B) and the flow
cytometric assay (4 µM heparin; Fig. 4B). Although we currently did not know the exact explanation(s), we speculate that this
difference could result from a number of potentially variable
factors. These included the differences in the determination
methods, the amounts of P-selectin Rg used, and the densities
of heparin-conjugated BSA, the stoichiometry of P-selectin
Rg binding to heparin and the topology of HSPGs on the cell
surface of A375 cells.
In literature, P-selectin was reported to bind to sulfatides
secreted by granulocytes and tumor cells (28-30). Further, using
site-directed mutagenesis, the amino acids on the lectin domain
of P-selectin for the bindings of myeloid cells and sulfatides
were shown to be overlapping (31). Likewise, we found that G1,
a leukocyte adhesion blocking mAb to the lectin domain of
P-selectin, blocked the binding of P-selectin to heparin-conjugated BSA (Fig. 5A). These observations suggest that sialyl
Lewis x, sulfatides and heparin may bind to the same or very
similar motif(s) in the lectin domain of P-selectin.
The moderate binding affinity, the fast on rate (kon) and the
fast off rate (koff) are known to be the three biophysical elements
critically required for leukocyte rolling on selectin molecules
under shear stress (23, 24, 32, 33). Compared with the parameters for the interaction of P-selectin with PSGL-1 (Kd = 320 ±
20 nM, koff = 1.4 ± 0.1 s-1 and kon = 4.4 106 M-1 s-1) (32), the
interaction of P-selectin with heparin has a characteristically
slow koff (a ~1,000-fold increase) and a reduced kon (a ~100fold reduction). Given the similar binding affinities for the
P-selectin/PSGL-1 and P-selectin/heparin interactions, these
data suggest that the koff and kon values are apparently the key
determinators for rolling (fast koff and kon) of leukocytes versus
adhesion (slow koff and kon) of cancer cells on P-selectin under
flow. By extrapolation, as carbohydrate ligands can dramatically alter the parameters of the binding kinetics to P-selectin,
our finding also indicate that the carbohydrate ligands can
determine the biophysical properties of these selectin-mediated
cell-cell interactions.
315
Abbreviations
BCS, bovine calf serum; BSA, bovine serum albumin; HSPGs, heparan sulfate proteoglycans; PBS, phosphate buffered saline, pH 7.4; PSGL-1, P-selectin glycoprotein ligand-1; Rg, receptor-globulin; RU, response unit.
Wang, Geng: Characteristics of P-selectin/heparin interaction
References
1. Springer TA. Traffic signals for lymphocyte
recirculation and leukocyte emigration: The
multistep paradigm. Cell 1994; 76: 301-14.
2. Kansas GS. Selectins and their ligands: current concepts and controversies. Blood 1996;
88: 3259-87.
3. Vestweber D, Blanks JE. Mechanisms that
regulate the function of the selectins and their
ligands. Physiol Rev 1999; 79: 181-213.
4. Geng J.-G. Directional migration of leukocytes: their pathological roles in inflammation
and strategies for development of anti-inflammatory therapies. Cell Res 2001; 11: 85-8.
5. McEver RP. P-selectin and PSGL-1: exploiting connections between inflammation and
venous thrombosis. Thromb Haemost 2002;
87: 364-5.
6. Skinner MP, Lucas CM, Burns GF,
Chesterman CN, Berndt MC. GMP-140 binding to neutrophils is inhibited by sulfated glycans. J Biol Chem1991; 266: 5371-4.
7. Nelson RM, Cecconi O, Roberts WG, Aruffo
A, Linhardt RJ, Bevilacqua MP. Heparin
oligosaccharides bind L- and P-selectin and
inhibit acute inflammation. Blood 1993; 82:
3253-8.
8. Norgard-Sumnicht KE, Varki NM, Varki A.
Calcium-dependent heparin-like ligands for Lselectin in nonlymphoid endothelial cells.
Science 1993; 261: 480-3.
9. Giuffrè L, Cordey AS, Monai N, Tardy Y,
Schapira M, Spertini O. Monocyte adhesion to
activated aortic endothelium: Role of L-selectin and heparan sulfate proteoglycans. J Cell
Biol 1997; 136: 945-56.
10. Koenig A, Norgard-Sumnicht K, Linhardt R,
Varki A. Differential interactions of heparin
and heparan sulfate glycosaminoglycans with
the selectins. Implications for the use of
unfractionated and low molecular weight heparins as therapeutic agents. J Clin Invest 1998;
101: 877-89.
11. Xie X, Rivier AS, Zakrzewicz A, et al..
Inhibition of selectin-mediated cell adhesion
and prevention of acute inflammation by nonanticoagulant sulfated saccharides. Studies
with carboxyl-reduced and sulfated heparin
and with trestatin a sulfate. J Biol Chem 2000;
275: 34818-25.
12. Borsig L, Wong R, Feramisco J, Nadeau DR,
Varki NM, Varki A. Heparin and cancer revisited: mechanistic connections involving platelets, P-selectin, carcinoma mucins, and tumor
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
metastasis. Proc Natl Acad Sci U S A 2001;
98: 3352-7.
Ma Y.-Q., Geng J.-G. Heparan sulfate-like
proteoglycans mediate adhesion of human
malignant melanoma A375 cells to P-selectin
under flow. J Immunol 2000; 165: 558-65.
Giavazzi R, Foppolo M, Dossi R, Remuzzi A.
Rolling and adhesion of human tumor cells on
vascular endothelium under physiological
conditions. J Clin Invest 1993; 92: 3038-44.
Dardik R, Savion N, Kaufmann Y, Varon D.
Thrombin promotes platelet-mediated melanoma cell adhesion to endothelial cells under
flow conditions: role of platelet glycoproteins
P-selectin and GPIIb-IIIA. Br J Cancer 1998;
77: 2069-75.
Wang, J.-G., Mu, J.-S., Zhu, H.-S., Geng, J.-G.
N-desulfated non-anticoagulant heparin inhibits leukocyte adhesion and transmigration in
vitro and attenuates acute peritonitis and
ischemia and reperfusion injury in vivo.
Inflammation Res. 2002; 51: 435-43.
Asa D, Raycroft L, Ma L, Aeed P, Kaytes PS,
Elhammer ÅP, Geng J-G. The P-selectin glycoprotein ligand functions as a common
human leukocyte ligand for P- and E-selectin.
J Biol Chem 1995; 270: 11662-70.
Geng JG, Bevilacqua MP, Moore KL,
McIntyre TM, Prescott SM, Kim JM, Bliss
GA, Zimmerman GA, McEver RP. Rapid neutrophil adhesion to activated endothelium
mediated by GMP140. Nature 1990; 343: 75760.
Aigner S, Ramos CL, Hafezi-Moghadam A,
Lawrence MB, Friederichs J, Altevogt P, Ley
K. CD24 mediates rolling of breast carcinoma
cells on P-selectin. FASEB J 1998; 12:
1241–51.
Kevin LM, Steven FE, David EL, Henri SL,
Richard DC, McEver RP. The P-selectin glycoprotein ligand from human neutrophils displays sialylated, fucosylated, O-linked polyN-acetyllactosamine. J Biol Chem1994; 273:
23318-27.
Gray GR. Antibodies to carbohydrates: preparation of antigen by coupling carbohydrates
to proteins by reductive amination with cyanoborohydride. In: Methods in Enzymology.
Ginsburg, V. ed. Academic Press, FL, USA
1978; 155-60.
Saloua N, Roslyn VG, Myrtle YG, Christopher
CR. Characterization of human recombinant
interleukin 2 binding to heparin and heparin
316
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
sulfate using an ELISA approach. Cytokine
1997; 12: 1013-22.
Shattil SJ, Cunningham M, Hoxie JA.
Detection of activated platelets in whole blood
using activation-dependent monoclonal antibodies and flow cytometry. Blood 1987; 70:
307-15.
Nicholson MW, Barclay AN, Singer MS,
Rosen SD, van der Merwe PA. Affinity and
kinetic analysis of L-selectin (CD62L) binding to glycosylation-dependent cell-adhesion
molecule-1. J Biol Chem 1998; 273: 763-70.
Fritz J, Katopodis AG, Kolbinger F,
Anselmetti D. Force-mediated kinetics of single P-selectin/ligand complexes observed by
atomic force microscopy. Proc Natl Acad Sci
U S A. 1998; 95: 12283-88.
Rodgers SD, Camphausen RT, Hammer DA.
Sialyl lewis(x)-mediated, PSGL-1-independent rolling adhesion on P-selectin.
Biophysical J 2000; 79: 694–706.
Rodgers SD, Camphausen RT, Hammer DA.
Tyrosine sulfation enhances but is not required
for PSGL-1 rolling adhesion on P-Selectin.
Biophysical J 2001; 79: 2001–9.
Aruffo A, Kolanus W, Walz G, Fredman P,
Seed B. CD62/P-selection recognition of myeloid and tumor cell sulfatides. Cell 1991; 67:
35-44.
Todderud G, Alford J, Millsap KA, Aruffo A,
Tramposch KM. PMN binding to P-selectin is
inhibited by sulfatide. J Leukoc Biol 1992; 52:
85-8.
Needham LK, Schnaar RL. The HNK-1 reactive sulfoglucuronyl glycolipids are ligands
for L-selectin and P-selectin but not E-selectin. Proc Nat. Acad Sci U S A 1993; 90: 135963.
Bajorath J, Hollenbaugh D, King G, Harte
WJr, Eustice DC, Darveau RP, Aruffo A.
CD62/P-selectin binding sites for myeloid
cells and sulfatides are overlapping.
Biochemistry 1994; 33: 1332-39.
Mehta P, Cummings RD, McEver RP. Affinity
and kinetic analysis of P-selectin binding to Pselectin glycoprotein ligand-1. J Biol Chem
1998; 273: 32506-13.
Wild MK, Huang MC, Schulze-Horsel U, van
der Merwe PA, Vestweber D. Affinity, kinetics, and thermodynamics of E-selectin binding
to E-selectin ligand-1. J Biol Chem 2001; 276:
31602–12.