|
|
 Previous Article | Table of Contents | Next Article 
Blood, Vol. 94 No. 3 (August 1), 1999:
pp. 976-983
Cyclophilin B Binding to Platelets Supports Calcium-Dependent
Adhesion to Collagen
By
Fabrice Allain,
Sandrine Durieux,
Agnès Denys,
Mathieu Carpentier, and
Geneviève Spik
From the Laboratoire de Chimie Biologique, Unité Mixte de
Recherche no. 8576 du CNRS, Université des Sciences et
Technologies de Lille, Villeneuve d'Ascq, France.
 |
ABSTRACT |
We have recently reported that cyclophilin B (CyPB), a secreted
cyclosporine-binding protein, could bind to T lymphocytes through
interactions with two types of binding sites. The first ones, referred
to as type I, involve interactions with the conserved domain of CyPB
and promote the endocytosis of surface-bound ligand, while the second
type of binding sites, termed type II, are represented by
glycosaminoglycans (GAG). Here, we further investigated the interactions of CyPB with blood cell populations. In addition to
lymphocytes, CyPB was found to interact mainly with platelets. The
binding is specific, with a dissociation constant (kd) of 9 ± 3 nmol/L and the number of sites estimated at 960 ± 60 per cell. Platelet glycosaminoglycans are not required for the
interactions, but the binding is dramatically reduced by active
cyclosporine derivatives. We then analyzed the biologic effects of CyPB
and found a significant increase in platelet adhesion to collagen. Concurrently, CyPB initiates a transmembranous influx of
Ca2+ and induces the phosphorylation of the P-20 light
chains of myosin. Taken together, the present results demonstrate for
the first time that extracellular CyPB specifically interacts with
platelets through a functional receptor related to the lymphocyte type
I binding sites and might act by regulating the activity of a
receptor-operated membrane Ca2+ channel.
© 1999 by The American Society of Hematology.
| |
INTRODUCTION |
CYCLOPHILINS are known to be the main
binding proteins for the immunosuppressive drug cyclosporine A
(CsA).1,2 The first characterized isoform was cyclophilin A
(CyPA), an abundant cytosolic protein that is considered to be the
major target for CsA into the cell.3,4 Cyclophilin B
(CyPB)4-6 and cyclophilin C (CyPC)7 are two
other isoforms structurally related to CyPA, but their mRNA encodes a
signal sequence thought to mediate translocation into the endoplasmic
reticulum. Both the cyclophilins and the structurally unrelated
FK506-binding proteins (FKBP) exhibit peptidyl-prolyl cis-trans
isomerase activity (PPIase)8-10 and inhibit the phosphatase activity of calcineurin in the presence of their respective
ligand.11,12 The latter property is thought to be relevant
to the immunosuppressive activity of both drugs. Indeed, the inhibition
of calcineurin activity has been shown to be a crucial step that
effectively blocks the early T-cell activation cascade and constitutes
the basis of the prevention of graft rejection.13
The presence of a released form of CyPB in human milk and
plasma6,14 has led us to investigate the properties of this protein. We first characterized specific surface binding sites on T
lymphocytes,15 mainly associated with the helper/inducer T-cell subset.16 Most recently, we provided evidence that
interactions of CyPB with sulfated glycosaminoglycans (GAG) may occur
on the T-cell surface. In addition, we identified a second type of CyPB binding sites, referred as type I sites versus type II for GAG interactions.17 The binding of CyPB to type I and type II
sites requires interactions with two distinct areas of the protein, the
catalytic/CsA-binding domain and the N-terminal extremity of CyPB,
respectively. Moreover, we demonstrated that the type I binding sites
are involved in an endocytosis process of the protein, which supports
the hypothesis that they may correspond to a functional receptor of
CyPB.17 The presence of surface binding sites on T
lymphocytes was consistent with the hypothesis that secreted CyPB may
interact with specific membrane receptors. However, the presence of
CyPB binding sites on the other human blood cells has not yet been
investigated. In the present report, we analyzed the distribution of
CyPB binding sites in blood cell populations. In addition to the
interactions with T lymphocytes, we found a significant binding of the
protein to platelets. CyPB interacts with platelets in a specific
manner, with a dissociation constant (kd) value similar to that of T
cells. However, CyPB binding to platelets did not involve interactions
with GAG, while it was strongly reduced in the presence of active
cyclosporine derivatives. Platelets are largely represented in human
blood and their activation is critically important in blood coagulation and inflammatory events.18 Here, we demonstrated that
incubation of platelets together with CyPB enhances adhesion to
collagen, but is ineffective in terms of degranulation or aggregation.
In addition, CyPB binding initiates an influx of extracellular
Ca2+ and some kinase activation, which demonstrates that
the platelet receptor is coupled to a transduction pathway. The present
results suggest that CyPB interacts with platelets through a functional receptor related to the lymphocyte type I binding sites and is probably
involved in the regulation of a receptor-operated membrane channel.
| |
MATERIALS AND METHODS |
Materials.
Human citrated venous blood samples from healthy donors were obtained
from the local blood transfusion center (Etablissement de Transfusion
Sanguine, Lille, France). Recombinant human CyPA and CyPB were produced
and purified as previously described.3,6 Recombinant human
CyPC7 and cyclosporine derivatives (CsA, CsG,
CsH)19 were a generous gift from Novartis (Basel,
Switzerland). Peptides that corresponded to the N- and C-terminal
extensions of CyPB were synthesized as described14 and were
provided with the tetrapeptide RGDS by Professor A. Tartar (Institut
Pasteur de Lille, France). Collagen mixture (essentially of type I and II) was purified from rat tail.20 [125I]CyPB
was prepared as described.15 The specific radioactivity ranged from 4 to 6 × 106 cpm/µg.
Preparation of platelets.
Platelet-rich plasma (PRP) was obtained by centrifugation of whole
blood at 200g for 30 minutes. Washed platelets were prepared by
filtration on a Sepharose 4-B column (Pharmacia, Uppsala, Sweden) equilibrated in Tyrode's buffer (NaCl, 137 mmol/L; KCl, 2.7 mmol/L; CaCl2, 2 mmol/L; MgCl2, 1 mmol/L;
Na2HPO4, 0.2 mmol/L; NaHCO3, 12 mmol/L; HEPES, 5 mmol/L; 0.1% glucose), pH 7.4, and resuspended in
Tyrode's buffer containing 0.5% bovine serum albumin (BSA). Concentrated platelet suspension was obtained by centrifugation of PRP
(1,600g, 10 minutes) and resuspension of the resulting pellet
with PRP to a final concentration of 2 × 109/mL.
Surface binding experiments with [125I]CyPB.
Surface binding of [125I]CyPB to blood cell populations
was performed by incubating blood samples (1 mL) in the presence of
radiolabeled ligand for 1 hour at room temperature. After washing off
the plasma, the surface-bound CyPB was counted in the blood cell
pellet, to obtain the total binding capacity, and in each isolated cell
population, after separation of blood cells on Ficoll separation medium
(Nycomed, Oslo, Norway). For platelet binding experiments, PRP was
previously incubated in the presence of 1 µmol/L prostaglandin
E1 for 30 minutes at 37°C. Platelets (2 × 108 per sample) were then allowed to bind
[125I]CyPB at various concentrations. After 1 hour at
22°C, the platelet suspension was filtered over a Whatman
glass fiber filter (Maidstone, UK) under mild vacuum and
washed twice. Filter-associated (bound ligand) and incubation medium
(free ligand) radioactivities were then measured. To analyze
interactions with GAG, washed platelets were treated with either 1 U/mL
heparinase I, 2 U/mL chondroitinase ABC, or both (Sigma Chemical, St
Louis, MO) for 2 hours at room temperature and directly used for
binding experiments as described.17 Control untreated
platelets were prepared under the same protocol. For all binding
experiments with [125I]CyPB, nonspecific interactions
were determined in the presence of a 200-fold molar excess of unlabeled
ligand and radioactivity was measured using a model 1282 Compugamma
LKB-Wallac counter (Gaithersburg, MD).
Platelet function analysis.
Platelet aggregation was typically performed at 37°C for 3 minutes
using a turbidimetric method. Aggregation was induced by the addition
of fibrinogen (1 mg/mL) or autologous plasma to washed platelet mixture
(2 × 108/mL) followed by the addition, 30 seconds
later, of CyPB or agonist. To measure platelet degranulation, PRP was
incubated with 5-[14C]hydroxytryptamine (5-HT) (0.05 mCi/mL) (ICN Biochemicals, Costa Mesa, CA) for 30 minutes
at 37°C. After gel filtration, platelets were incubated with CyPB
or agonist and processed as described.21 Released
5-[14C]HT was analyzed using a model LS 6000-TA Beckman
counter (Allendale, NJ). For platelet adhesion assays,
96-well microtiter plates were coated with 1 µg/well of collagen in a
sodium carbonate buffer, 10 mmol/L, pH 9.6, overnight at 4°C.
Nonspecific binding sites were blocked by addition of 2% BSA.
Platelets (1 × 107 per well) were incubated in the
presence of various concentrations of CyPB and added to collagen-coated
wells for a 30-minute incubation at 37°C. After washing, the
adherent platelets were quantified using the BCA reagent kit for
protein assay (Pierce Chemicals, Rockford, IL).
Calcium measurements.
Platelets were loaded with 3 µmol/L of Fluo 3-acetoxymethylester
(Fluo 3-AM) (Molecular Probes, Leiden, Netherlands)22 for 30 minutes at 37°C. After gel filtration to remove extracellular Fluo 3, the final platelet concentration was adjusted to 1 × 106/mL in Tyrode's buffer. Stimulation was induced by the
addition of either various concentrations of CyPB or thrombin at
37°C. Changes in fluorescence were recorded by flow cytofluorimetry using a Becton Dickinson FACScan cytofluorimeter (Mountain
View, CA), with excitation and emission wavelengths set on 488 and 515 nm, respectively. This method allowed the analysis of 2,000 fluorescent particles every 30 seconds. The Ca2+-Fluo 3 fluorescence
was calibrated with a maximum response induced by the addition of
ionomycin to the suspension. The levels of cytosolic Ca2+
were calculated for each fluorescence mean value, with a kd of 400 nmol/L for Fluo 3.22
Measurement of inositol phosphate formation.
Concentrated platelet suspensions were labeled with
myo-[3H]inositol (50 µCi/ml) (ICN Biochemicals)
for 3 hours at 37°C. Following washing and resuspension in a
myo-inositol-free buffer, platelets were stimulated with
either CyPB or thrombin and processed as described.23
Protein phosphorylation analysis.
Platelets (2 × 108 per sample) were incubated in the
presence of either CyPB or thrombin at 37°C. At various times,
platelets were transferred into Tyrode's buffer containing 10 mmol/L
EDTA, 1 mmol/L o-vanadate and 100 mmol/L NaF, to inhibit
protein phosphatases and rapidly washed by centrifugation
(2,500g, 30 seconds). Proteins from platelet lysates were
separated on a 12% sodium dodecyl sulfate-polyacrylamide gel
electropheresis (SDS/PAGE) and transblotted onto nitrocellulose paper.24 After blocking in Tris-buffered saline (TBS), pH
8.2, which contained 3% gelatin, blots were incubated with mouse
monoclonal antibodies to phosphoserine or phosphotyrosine residues
(Sigma) in TBS-gelatin 0.5% for 2 hours and exposed to horseradish
peroxidase-labeled antimouse IgG antibodies (1/2,500) (BioSys,
Compiègne, France) for another 2-hour incubation. Development was
performed with o-phenylene diamine kit (Sigma). Because
platelet controls from different individuals showed varying degrees of
protein phosphorylation, the effects of CyPB were evaluated in terms of
variations of phosphorylation intensity using Quantiscan
software (Biosoft, Cambridge, UK).
Statistical analysis.
Results are expressed as mean values ± SEM for at least three
independently performed experiments conducted with separate donors.
Statistical significance between the different values was analyzed by
Student's t-test for unpaired data with a threshold of
P < .05.
| |
RESULTS |
Characterization of CyPB binding sites on human blood cells.
On the assumption that only T lymphocytes were involved in the
interactions with CyPB, the total binding capacity of
[125I]CyPB was estimated at 50 fmol/mL in whole blood.
Indeed, we previously reported that approximately 40% of the
peripheral blood lymphocyte population showed significant binding of
CyPB, with a capacity ranging from 30,000 to 120,000 sites per
cell.15-17 Surprisingly, the surface-bound ligand was found
to be 415 ± 80 fmol/mL, much higher than the calculated value. The
distribution of [125I]CyPB was then analyzed after
separation of the blood cell populations (Table
1). Ten percent to 15% of the total
binding capacity of [125I]CyPB was found associated with
the lymphocyte fraction, which corresponds to the expected value we had
calculated. No significant amounts of radioactivity were measurable in
the monocyte population. A weak but significantly measurable proportion
of CyPB was associated with granulocytes. This value might reflect
either a poor expression of CyPB binding sites on the membrane of this
whole cell population, or a restricted CyPB binding on a specific
subpopulation of polymorphonuclear cells. More than 80% of the bound
protein was found associated with the platelet fraction. Owing to the
large amount of platelets in human blood, this value implies that the
number of surface-bound CyPB would range from 500 to 1,200 on
platelets. Finally, the radioactivity found associated with the
erythrocyte population is probably due to [125I]CyPB
binding to contaminating platelets or lymphocytes; otherwise it would
correspond to less than one binding site per erythrocyte, according to
the number of red blood cells.
Surface binding of CyPB to platelets.
The binding parameters of CyPB were determined by incubating platelets
with increasing concentrations of [125I]CyPB. Figure
1 illustrates one representative experiment
performed with washed platelets. The binding was specific since a
200-fold molar excess of unlabeled ligand inhibited
[125I]CyPB binding by 60% to 80%. After subtraction of
nonspecific interactions, the binding was found to be
concentration-dependent and saturable (Fig 1A). Scatchard analysis
resulted in a linear plot compatible with a single affinity binding
site. The apparent kd was 9 ± 3 nmol/L and the number of binding
sites was estimated at 960 ± 60 per platelet (Fig 1B). Similar data
were obtained when Ca2+ and Mg2+ were omitted
from the incubation medium, which indicates that the presence of the
divalent cations is not required for CyPB binding to platelets. In
additional experiments, platelets were incubated in the presence of 50 nmol/L [125I]CyPB for 1 hour at 22°C, then
extensively washed to remove unbound ligand and resuspended in the same
buffer containing a 10-fold molar excess of unlabeled CyPB. A rapid
removal of surface bound radiolabeled ligand was observed, which
indicates that the binding of CyPB to platelets is reversible (data not
shown). To ensure that CyPB binding may occur under physiologic
conditions, PRP was adjusted to 2 × 108 platelets/mL
with citrated buffer and directly used for binding experiments (n = 3).
In this case, the kd value and number of sites were estimated at 10 ± 3 nmol/L and 900 ± 120 per platelet, which indicates that the
presence of plasma does not significantly modify the binding
parameters. In addition, endogenous plasma CyPB levels were measured by
enzyme-linked immunosorbent assay (ELISA)14 and found to be
less than 5 nmol/L, which is too low to account for a large part in the
binding site occupancy.
View larger version (17K):
[in this window]
[in a new window]
| Fig 1.
Surface binding of [125I]CyPB to platelets.
Dose-dependence and saturation of CyPB binding were studied by
incubating platelets with the indicated concentrations of
[125I]CyPB for 1 hour at 22°C. The specific binding
( ) was obtained after subtraction of nonspecific ( ) from total
counts ( ). Points represent the mean values of triplicates from a
representative experiment (A). (B) Scatchard plot of the binding
data.
|
|
Specificity of CyPB binding to platelets.
The specificity of CyPB binding to platelets was first analyzed by
competitive experiments. For these studies, platelets were incubated
with 50 nmol/L of [125I]CyPB in the presence of various
concentrations of unlabeled CyPA, CyPB or CyPC (Fig
2A). As expected, [125I]CyPB
binding was inhibited by greater than 80% from a 200-fold molar excess
of unlabeled CyPB. The concentration of CyPB required for half-maximal
inhibition (IC50) of the total [125I]CyPB
binding was estimated at 250 nmol/L. CyPA was unable to displace the
radio-iodinated ligand from the platelet membrane, which shows that
this isoform has no affinity for CyPB binding sites. In contrast,
increasing concentrations of CyPC inhibited [125I]CyPB
binding to platelets, but to a lesser extent than CyPB. The
IC50 was estimated at 2,950 nmol/L, which shows that CyPC has a lower affinity for platelet binding sites. We then examined the
involvement of nonconserved regions of CyPB in the ligand binding by
using synthetic peptides that copy the most divergent parts of the
protein. As previously reported,17 increasing
concentrations of the C-terminal peptide were unable to
reduce [125I]CyPB binding. Surprisingly, the N-terminal
peptide was also inefficient at competing with the ligand for binding
to the platelet receptor, although it was previously shown to strongly
reduce interactions of CyPB with the lymphocyte type II
sites.17 Contrary to CyPA and CyPC, only CyPB possesses a
specific RGD motif.4-7 Nevertheless, the involvement of
this specific tripeptide in the interactions of CyPB with platelets is
unlikely, since the addition of increasing concentrations of the
tetrapeptide RGDS, which binds to glycoprotein (GP)IIb/IIIa and
inhibits its interaction with RGD-containing ligands such as
fibrinogen,25 was also ineffective at reducing
[125I]CyPB binding (Fig 2A). We then investigated the
role of the CsA-binding/catalytic domain of CyPB in the interactions
with the platelet receptor, by using cyclosporine derivatives as
competitive inhibitors (Fig 2B). Both CsA and the less active CsG
reduced [125I]CyPB binding to the platelet membrane. A
significant decrease was only obtained from a 10-fold molar excess of
both drugs, which indicates that the inhibition requires that CyPB was
maintained in a complexed form. By contrast, CsH, which is unable to
interact with CyPB, failed to prevent the ligand binding, which
confirms that occupancy of CsA-binding/catalytic domain by active
cyclosporine derivatives accounts for the loss of binding activity. On
the other hand, treatment of platelets with GAG-degrading enzymes did
not significantly modify the binding of [125I]CyPB to
platelets. Moreover, incubation of CyPB together with protamine, a
polypeptide that inhibits interactions with heparin-like molecules, had
no more effect on CyPB binding to platelets, which confirms that CyPB
does not interact with platelet GAG (Fig 2B).
View larger version (26K):
[in this window]
[in a new window]
| Fig 2.
Specificity of [125I]CyPB binding to
platelets. (A) Competitive binding assays with cyclophilin isoforms and
synthetic peptides copying specific sequences of CyPB. Platelets were
incubated in the presence of 50 nmol/L [125I]CyPB and
increasing concentrations of unlabeled CyPA (), CyPB (), CyPC
( ), N-terminal peptide of CyPB ( ), C-terminal peptide of CyPB
( ), or RGDS peptide ( ). After washing, the amounts of remaining
surface-bound [125I]CyPB, expressed as a percentage of
the ligand bound in the absence of competitors, are plotted against the
molar ratios of [125I]CyPB to the competitors. Data are
expressed as mean values from 3 separate experiments conducted with
platelets from different donors. (B) Sensitivity of CyPB binding to
cyclosporine derivatives, protamine, and GAG-degrading enzymes.
Platelets were incubated with of 50 nmol/L [125I]CyPB in
the absence (1) or presence of CsA 500 nmol/L (2); CsA 5 µmol/L (3);
CsG 500 nmol/L (4); CsG 5 µmol/L (5), CsH 500 nmol/L (6); CsH 5 µmol/L (7); protamine 500 nmol/L (8); or protamine 5 µmol/L (9). In
the last cases, platelets were first pretreated with heparinase type I
(10), chondroitinase ABC (11), or both (12), and directly used for
binding experiments. After washing, the amounts of remaining
surface-bound [125I]CyPB were expressed as a percentage
of the ligand bound in the absence of any treatment. Data are mean
values ± SEM from 3 separate experiments conducted with platelets
from different donors.
|
|
Analysis of CyPB activity on platelet functions.
In an attempt to understand the biological relevance of CyPB binding,
we analyzed its effects on platelet functions. The addition of CyPB at
different concentrations did not induce any significant aggregate
formation and only a poor release of 5-[14C]HT from
loaded platelets by comparison with thrombin (Table 2). We then examined whether CyPB may
potentiate platelet aggregation or secretion after low doses of
thrombin. Thrombin at 0.05 U/mL was used because it caused only
moderate but significant platelet activation at this concentration.
However, the addition of CyPB was found to have no significant
enhancing effect on aggregation or 5-HT release (Table 2). Similar
conclusions were observed when platelets were challenged in the
presence of plasma or activated with adenosine diphosphate (ADP) in
place of thrombin (data not shown), which confirms that CyPB is
ineffective at inducing any aggregation or degranulation processes and
does not act synergistically with low doses of agonist to activate
platelets.
For adhesion analysis, platelets were incubated in the presence of
various concentrations of CyPB, and the samples were added to
collagen-coated plates. Control values, obtained in the absence of
CyPB, were estimated at 20% to 32% of initially added platelets (27% ± 4%). These values were obtained in separate experiments conducted with the platelets from different donors (n = 10) and variations are likely to be due to individual differences in the response to collagen. In the presence of CyPB, a significant increase in platelet adhesion was observed with 5 nmol/L of the protein and the
optimal effect was measured at approximately 50 nmol/L. At this
concentration, adherent platelets ranged from 43% to 50% (47% ± 2.5%), which reflects an almost twofold increase in platelet adhesion
in comparison to control (Fig 3). To
determine whether this increasing effect is related to CyPB binding to
the platelet receptor, similar experiments were performed in the
presence of CsA. In the absence of CyPB, the immunosuppressant had no
marked effect on platelet adhesion to collagen. In contrast, the
addition of the drug at low and optimal concentrations of CyPB
decreased the enhancing effect of the protein. Nevertheless, high
concentrations of CsA were necessary to significantly reduce the
platelet adhesion to collagen. The basal level in platelet adhesion was
only restored in the presence of a 100-fold molar excess of the drug.
In contrast, CsA was ineffective in the presence of high concentrations
of CyPB, which may be explained by the presence of uncomplexed CyPB at
a concentration probably sufficient to obtain an optimal increase in
platelet adhesion (Fig 3). Two distinct mechanisms have been distinguished in the promotion of platelet adhesion, one dependent on
the presence of divalent cations, and the other cation-independent and
reflecting the participation of plasma proteins such as fibronectin or
von Willebrand factor. In the last mechanism, adhesive proteins are
thought to serve as a bridge between the platelet membrane and the
collagen fiber surface.26 To check this hypothesis, collagen-coated plates were pretreated with CyPB and extensively washed
before the addition of platelets. In these conditions, a CyPB-mediated
increase in platelet adhesion was not observed, which demonstrates that
CyPB does not act by forming a link between collagen and the platelet
receptor. In contrast, when platelets were pretreated with CyPB and
extensively washed to remove unbound ligand, the enhancing effect of
the protein on platelet adhesion was preserved, which suggests it may
be related to the binding of CyPB to a platelet signalling receptor
(Table 3). We then analyzed the role of
divalent cations, by incubating platelets in a
Ca2+/Mg2+-depleted medium supplemented with 2 mmol/L EGTA. In this case, platelet adhesion was significantly reduced,
from 30% to 12%. Nevertheless, no significant increase occurred in
the presence of CyPB, which implies that the presence of divalent
cations is required for the protein to enhance platelet adhesion. When
citrated plasma was used in place of buffer that contained EGTA,
platelet adhesion was approximately 20%, which indicates the
participation of plasma factors in promoting cation-independent
adhesion. Nevertheless, this mechanism was unmodified when CyPB was
added to citrated plasma. Finally, adhesion experiments were reproduced
with recalcified plasma. To prevent spontaneous aggregation, platelets
were pretreated with aspirin (100 µmol/L) and the drug was conserved
all along the experiment. In this case, the enhancing effect of CyPB on platelet adhesion to collagen was partially restored, with an almost
1.5-fold increase in adhesion by comparison to control (Table 3). Taken
together, these results demonstrate that the action of CyPB is only
dependent on the presence of divalent cations and may occur under
physiologic conditions.
View larger version (17K):
[in this window]
[in a new window]
| Fig 3.
Enhancing effect of CyPB on platelet adhesion. Washed
platelets were incubated in the presence of increasing concentrations
of CyPB in the absence () or presence of 500 nmol/L () or 5 µmol/L () of CsA and added to 96-well plates coated with collagen
(1 µg/ well) for a 30-minute incubation at 37°C. After washing,
adherent platelets were quantified using a BCA protein assay and
expressed as percentages of initially added platelets (1 × 107 platelets per well). Results are expressed as mean
values ± SEM from quadruplicates and are representative from at least
3 separate experiments conducted with platelets from different
donors.
|
|
View this table:
[in this window]
[in a new window]
|
Table 3.
Effects of the Modification of Either Incubation
Medium or Treatment Procedure on CyPB-Mediated Adhesion of
Platelets to Collagen
|
|
Effects of CyPB on Ca2+ movements and protein
phosphorylation.
In the following experiments, we investigated whether the enhancing
effect of CyPB on platelet adhesion may be related to the transduction
of intracellular signals. In this way, we analyzed possible pathways of
CyPB-induced platelet response by measuring the dose- and
time-responses of intraplatelet Ca2+ signal generation and
protein kinase activation following CyPB addition.
To analyze Ca2+ responses, a series of spectrofluorimetric
experiments was performed on platelets loaded with the Ca2+
fluorophore Fluo-3. Addition of CyPB (10 to 500 nmol/L) induced an
increase in cytosolic free Ca2+ and the concentration of
ligand required for a maximum response was estimated at 100 nmol/L.
These values are consistent with the CyPB concentrations required for
surface binding site occupancy. Stimulation with CyPB resulted in a low
and durable Ca2+ flux, with an increase from 90 ± 35 nmol/L to 225 ± 45 nmol/L in the first minute (Fig
4A). This elevation in cytosolic free Ca2+ concentration was nevertheless not comparable to that
induced by thrombin, which was estimated at 1,220 ± 205 nmol/L. To
inhibit extracellular Ca2+ entry, platelets were diluted 1 minute before analysis in a CaCl2-depleted buffer
containing 2 mmol/L EGTA. In this case, the effect of CyPB was similar
to that observed in the absence of any activator (Fig 4B), which
suggests that the elevation in cytosolic free Ca2+
initiated by CyPB is likely to be generated by a transmembranous influx
of extracellular Ca2+ through a membrane channel. We then
examined the influence of CyPB binding on the formation of inositol
phosphate (InsPs) derivatives, after platelet labeling with
myo-[3H]inositol. The exposure of platelets to
CyPB for times varying from 1 to 10 minutes did not increase the levels
of these second messengers (1,260 ± 320 cpm within 5 minutes) in
comparison to basal levels (930 ± 240 cpm), while thrombin induced
a large and rapid increase in InsPs concentration (14,400 ± 1,080 cpm within 5 minutes). These results further demonstrate that
CyPB-induced Ca2+ flux is quite different from that
initiated by thrombin and not dependent on the activation of PLC.
View larger version (14K):
[in this window]
[in a new window]
| Fig 4.
Effects of CyPB on Ca2+ flux and
phosphorylation of myosin light chains (P-20) and pleckstrin (P-47) in
platelets. Ca2+ mobilization, measured in Fluo-3-loaded
platelets, was analyzed after the addition of CyPB (100 nmol/L normal
(A) or Ca2+-depleted buffer (B). Changes in fluorescence,
reflecting changes in cytosolic Ca2+ concentration, were
monitored by flow cytofluorimetry. Tracings are representative of 3 distinct experiments conducted with platelets from separate
individuals. The arrows indicate addition of the ligand. Platelet
response to CyPB (100 nmol/L) was also analyzed in terms of variations
of P-20 and P-47 protein phosphorylation (C). Reactions were stopped at
the indicated times and the variations in the intensity of serine
phosphorylation of P-20 () and P-47 () were analyzed. Zero time
results were obtained without addition of the ligand. Data are
calculated as the percentage of intensity at indicated times relative
to that at time zero, and expressed as mean values ± SEM from 3 separate experiments conducted with platelets from different donors.
|
|
To analyze the effect of CyPB on protein kinase activation, platelets
were incubated in the presence of various concentrations of CyPB and
the patterns of phosphorylation were compared with those obtained in
the absence of agonist, or in the presence of thrombin (0.5 U/mL) taken
as a positive control. CyPB did not induce any significant changes in
tyrosine phosphorylation (data not shown). We then compared the
profiles of serine phosphorylation of P-47 pleckstrin, substrate for
protein kinase C (PKC)27 and P-20 myosin light chains,
substrate for myosin light chain kinase (MLCK).28,29 As
expected, the addition of thrombin to platelets resulted in the rapid
phosphorylation of P-47 and P-20. After CyPB stimulation, an increase
in serine phosphorylation was essentially observed for the 20-kD
proteins, while P-47 was not significantly modified (Fig 4C).
Phosphorylation of P-20 rapidly increased and was maximum within 5 to
10 minutes, which shows that it occurred after the generation of
Ca2+ influx. Moreover, the optimal concentration of CyPB
was estimated at 100 nmol/L, with a threefold increase in intensity, at
a level similar to that obtained with platelets challenged with
thrombin (290% ± 10% and 250% ± 15% for CyPB and thrombin,
respectively). In contrast, CyPB did not induce any increase in P-20
phosphorylation when similar experiments were reproduced in the absence
of extracellular Ca2+. Taken together, these results
indicate that the dose- and time-dependent responses to CyPB for
increasing P-20 phosphorylation paralleled those observed for elevation
of cytosolic free Ca2+ concentration, which suggests that
both events are related.
| |
DISCUSSION |
Our previous work showed that CyPB specifically binds to the surface of
human T cells.15-17 Here, we present new data demonstrating that specific binding sites are also expressed at the surface of human
platelets and exhibit similar affinity for CyPB. No binding was
observed to erythrocytes and monocytes, and most probably the methods
used were not sensitive enough to conclude to the presence or absence
of surface CyPB binding sites on granulocytes. We reported that CyPB
interacts with two types of binding sites present on the membrane of T
lymphocytes.17 The first ones, termed type I binding sites,
involve interactions with the CsA-binding domain of CyPB, while the
type II sites are mainly represented by GAG present on the T-cell
membrane and involve interactions with the N-terminal extension of the
protein. The present data demonstrate that the interactions of CyPB
with the platelet receptor and the lymphocyte type II binding sites are
quite different, which excludes a role for platelet GAG in CyPB
binding. Actually, the lymphocyte type II sites were found to mainly
correspond to molecules of the heparin/heparan sulfate
family.17 Conversely, platelet GAG are almost exclusively
represented by chondroitin-4-sulfate,30 which might explain
the absence of type II binding sites on platelets. The involvement of
the RGD tripeptide in the interactions of a large family of ligands,
such as fibrinogen and von Willebrand factor, with platelet membrane
receptors has been largely documented.25 Nevertheless, the
RGDS peptide was unable to compete with CyPB binding, which rules out
the role of integrins of the GPIIb/IIIa family in the binding of CyPB.
We then demonstrated that both CsA and CsG, but not CsH, reduced CyPB
binding to the platelet membrane. Actually, active drugs overlay the
binding domain of CyPB when complexed to the protein and probably lead
to a loss of accessibility for the platelet receptor. In addition,
CyPC, but not CyPA, was found here to compete with CyPB for binding to
platelet receptor. However, the area of the cyclophilin isoforms that
interacts with cyclosporine derivatives is strongly
conserved,4,6,7 which disagrees with the role of this
conserved catalytic domain in receptor recognition. Most probably,
divergent regions differentially influence the spatial conformation of
these proteins and therefore explain variations in the interactions
with specific receptors. Such properties were also observed for the
binding of CyPB to lymphocyte type I sites,17 which
strongly suggests that both CyPB receptors are related.
The events initiated by CyPB binding to the platelet receptor were
unexpected and appeared to differ in important ways to those induced by
agonists like thrombin.18 CyPB was found to increase
platelet adhesion to collagen, but was unable to promote degranulation
or aggregation. The mechanism by which CyPB increases platelet adhesion
to collagen is dependent on the presence of extracellular
Ca2+ and is accompanied by the elevation of cytosolic free
Ca2+ concentration and P-20 phosphorylation. However, it
does not require the generation of InsPs, which suggests that
stimulation of CyPB membrane sites mediates activation of effectors
other than PLC. Moreover, the absence of the PKC-dependent
phosphorylation of P-47 demonstrates that there is no generation of
diacylglycerol and further confirms the absence of any activation of
PLC. In contrast, the effect of CyPB on phosphorylation of P-20
indicates that the increase in cytosolic free Ca2+
concentration probably led to the activation of MLCK. Indeed, phosphorylation of the P-light chains of myosin may be induced by
direct activation of this kinase by the low elevation of intracellular Ca2+ concentration.18 Similar events are
observed when platelets are exposed to cold temperatures. Chilling
platelets was reported to promote a Ca2+ entry through the
inhibition of the membrane Ca2+/adenosine triphosphatase
(ATPase) channel by low temperature and to induce serine/threonine
phosphorylation.31,32 These events would be essentially
mediated by MLCK in response to the increase in cytosolic
Ca2+.33 This is consistent with our hypothesis
that CyPB-mediated Ca2+ influx is related to the
phosphorylation of P-20 in promoting the activation of MLCK. On the
other hand, a CyPB-associated protein, termed
calcium-signal-modulating cyclophilin ligand, was already reported
to participate in the transmission of Ca2+ influx signals
in T cells.34 This protein was postulated to regulate
intracellular Ca2+ release or generate a signal responsible
for opening plasma membrane channels.34 Such a CyPB-binding
protein might be expressed at the surface of platelets and be involved
in the control of Ca2+ influx in association with
extracellular CyPB. In this way, CyPB might interact with a
receptor-operated channel and initiate a transmembraneous influx of
Ca2+, leading to the activation of the
Ca2+-dependent MLCK. The phosphorylation of P-light chains
of myosin and increase in cytosolic free Ca2+ concentration
are thought to play a key role in the contractile events associated
with platelet shape changes and adhesion.21,28,29,35 It is
therefore conceivable to postulate that CyPB-mediated activation of
MLCK and Ca2+ entry is related to the enhancing effect of
the protein on platelet adhesion.
The addition of CsA significantly reduced the enhanced platelet
adhesion. These results indicate that the surface binding and activity
of CyPB are related and can be abolished by occupancy of the
CsA-binding domain of the protein. This inhibitory effect is likely to
be dependent on the concentration of the drug. Most probably, CsA
divided between CyPB and other binding sites, eg, platelet proteins and
lipids, and large molar excesses of the drug are necessary to form a
stable and inactive complexed form of CyPB. Elevated concentrations of
CsA are currently measured in blood from transplant recipients, which
suggests that the drug could interfere with the biologic activity of
CyPB and hemostatic parameters. Many in vivo observations report that
CsA increases the risk of thromboembolism in transplant patients, which
is mainly related to endothelium damage and abnormalities in platelets
and the coagulolytic system. However, data concerning the effect of CsA
on hemostasis and platelet functions are often confusing. A
prothrombotic effect of the drug was generally reported and attributed
to an increased activation of PKC after stimulation by
agonists.36,37 In contrast, CsA therapy was reported to increase the levels of antithrombin and protein C, two proteins known to protect against venous thromboembolism.38 Most
recently, CsA was demonstrated to have both proaggregatory but
anticoagulant effects, which are largely related to platelet reactivity
and drug concentration.39 Further studies will therefore be
necessary to determine the mechanisms by which CyPB in combination with hemostatic agents could modulate platelet reactivity and coagulation in
the presence of CsA.
We reported in a previous work that CyPB levels in plasma from healthy
donors were in the range of 5 nmol/L.14 However, we
demonstrated here that CyPB-mediated Ca2+ flux generation
and increase in platelet adhesion to collagen required higher
concentrations. Such concentrations have been measured in the plasma
from patients with sepsis, which indicates that CyPB may be secreted as
an inflammatory response and exert a cytokine-like
activity.40 On the other hand, CyPB was demonstrated to
escort and stabilize procollagen chains all along the secretory pathway.41 Thus, CyPB could be secreted together with
collagen and accumulate in the subendothelial matrix. As a response to blood vessel offense, platelets adhere to the site of injury and become
activated. These events are mainly mediated by the contact of collagen
with platelets.18 Therefore, the occurrence of collagen might be associated with the liberation of CyPB from the subendothelial matrix at the site of the injured vessel, which would allow the protein
to exert its enhancing effect on platelet adhesion.
In nonexcitable cells such as platelets, extracellular Ca2+
entry is thought to be controlled in part by agonists that act directly on plasma membrane Ca2+ channel.42 Studies with
permeabilized platelets demonstrated that guanine nucleotide regulatory
proteins were involved in the adhesive process to
collagen.43 In this way, the platelet CyPB receptor might
be associated with such regulatory proteins and control the activity of
a membrane channel. Our objective is now to characterize the platelet
receptor as a possible Ca2+ channel-associated protein and
to ascertain whether it is related to the lymphocyte type I binding
sites. This should allow further understanding on the biologic
functions of released CyPB.
| |
ACKNOWLEDGMENT |
We are grateful to Dr J.J. Huart, Director of the
Etablissement de Transfusion Sanguine, Lille, for providing us with
blood samples, and to Prof A. Tartar for the synthesis of the peptides used in this work. We also thank Drs J.F. Borel and M. Zurini for
generous gifts of cyclosporine derivatives and human recombinant CyPC.
| |
FOOTNOTES |
Submitted October 14, 1998; accepted April 5, 1999.
Supported by the Université des Sciences et Technologies de
Lille, CNRS (Unité Mixte de Recherche no. 111; Director:
Professor A. Verbert) and by a grant from the Conseil Régional du
Nord/Pas-de-Calais (contract: "Maladies
neurodégénératives et Vieillissement").
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 18 U.S.C. section
1734 solely to indicate this fact.
Address reprint requests to Prof Geneviève Spik, Laboratoire de
Chimie Biologique, Unité Mixte du CNRS no. 111, Université
des Sciences et Technologies de Lille, 8576 Villeneuve d'Ascq Cedex,
France; e-mail: genevieve.spik{at}univ-lille1.fr.
| |
REFERENCES |
1.
Handschumacher RE, Harding MW, Rice J, Drugge RJ, Speicher DW:
Cyclophilin: A specific cytosolic binding protein for cyclosporin A.
Science
226:544, 1984[Abstract/Free Full Text]
2.
Schreiber SL:
Chemistry and biology of the immunophilins and their immunosuppressive ligands.
Science
251:283, 1991[Abstract/Free Full Text]
3.
Haendler B, Hofer-Warbinek R, Hofer E:
Complementary DNA for human T-cell cyclophilin.
EMBO J
6:947, 1987[Medline]
[Order article via Infotrieve]
4.
Bergsma DJ, Eder C, Gross M, Kersten H, Sylvester D, Appelbaum E, Cusimano D, Livi GP, McLaughlin MM, Kasyan K, Porter TG, Silverman C, Dunnington D, Hand A, Prichett WP, Bossard MJ, Brandt M, Levy MA:
The cyclophilin multigene family of peptidyl-prolyl isomerases.
J Biol Chem
266:23204, 1991[Abstract/Free Full Text]
5.
Price ER, Zydowsky LD, Jin M, Baker CH, McKeon FD, Walsh CT:
Human cyclophilin B: a second cyclophilin gene encodes a peptidyl-prolyl isomerase with a signal sequence.
Proc Natl Acad Sci USA
88:1903, 1991[Abstract/Free Full Text]
6.
Spik G, Haendler B, Delmas O, Mariller C, Chamoux M, Maes P, Tartar A, Montreuil J, Stedman K, Kocher H, Keller R, Hiestand PC, Movva NR:
A novel secreted cyclophilin-like protein (SCYLP).
J Biol Chem
266:10735, 1991[Abstract/Free Full Text]
7.
Schneider H, Charara N, Schmitz R, Wehrli S, Mikol V, Zurini MGM, Quesniaux VFJ, Movva NR:
Human cyclophilin C: Primary structure, tissue distribution, and determination of binding specificity for cyclosporins.
Biochemistry
33:8218, 1994[Medline]
[Order article via Infotrieve]
8.
Fischer G, Wittmann-Liebold B, Lang K, Kiefhaber T, Schmid FX:
Cyclophilin and peptidyl prolyl cis-trans isomerase are probably identical proteins.
Nature
337:476, 1989[Medline]
[Order article via Infotrieve]
9.
Takahashi N, Hayano T, Suzuki M:
Peptidyl-prolyl cis-trans isomerase is the cyclosporin A-binding protein cyclophilin.
Nature
337:473, 1989[Medline]
[Order article via Infotrieve]
10.
Siekierka JJ, Hung SHJ, Poe M, Lin CS, Sigal NH:
A cytosolic binding protein for the immunosuppressant FK506 has peptidyl-prolyl isomerase activity but is distinct from cyclophilin.
Nature
341:755, 1989[Medline]
[Order article via Infotrieve]
11.
Liu J, Farmer JD, Lane WS, Friedman J, Weissman I, Schreiber SL:
Calcineurin is a common target of cyclophilin-cyclosporin A and FKBP-FK506 complexes.
Cell
66:807, 1991[Medline]
[Order article via Infotrieve]
12.
Liu J, Albers MW, Wandless TJ, Luan S, Alberg DG, Belshaw PJ, Cohen P, McKintosh C, Klee CB, Schreiber SL:
Inhibition of T cell signaling by immunophilin-ligand complexes correlates with loss of calcineurin phosphatase activity.
Biochemistry
31:3896, 1992[Medline]
[Order article via Infotrieve]
13.
Schreiber SL, Crabtree GR:
The mechanism of action of cyclosporin A and FK506.
Immunol Today
13:136, 1992[Medline]
[Order article via Infotrieve]
14.
Allain F, Boutillon C, Mariller C, Spik G:
Selective assay for CyPA and CyPB in human blood using highly specific anti-peptide antibodies.
J Immunol Meth
178:113, 1995[Medline]
[Order article via Infotrieve]
15.
Allain F, Denys A, Spik G:
Characterization of surface binding sites for cyclophilin B on a human tumor T-cell line.
J Biol Chem
269:16537, 1994[Abstract/Free Full Text]
16.
Denys A, Allain F, Foxwell B, Spik G:
Distribution of cyclophilin B binding sites in the subsets of human peripheral blood lymphocytes.
Immunology
91:609, 1997[Medline]
[Order article via Infotrieve]
17.
Denys A, Allain F, Carpentier M, Spik G:
Involvement of two classes of binding sites in the interactions of cyclophilin B with peripheral blood T-lymphocytes.
Biochem J
336:689, 1998
18.
Crawford N, Scrutton MC:
Biochemistry of the blood platelets, in
Bloom AL,
Forbes CD,
Thomas RP
(eds):
Haemostasis and Thrombosis. Singapore, Longman Singapore, 1994, p 89.
19.
Von Wartburg A, Traber R:
Chemistry of the natural cyclosporin metabolites.
Prog Allergy
38:28, 1986[Medline]
[Order article via Infotrieve]
20.
Borstein BM:
Reconstituted rat tail collagen used as substrate for time tissue cultures on coverslips in Maximow slides and roller tubes.
Lab Invest
7:134, 1958[Medline]
[Order article via Infotrieve]
21.
Smith JB, Selak MA, Dangelmaier C, Daniel JL:
Cytosolic calcium as a second messenger for collagen-induced platelet responses.
Biochem J
288:925, 1992
22.
Minta A, Kao JP, Tsien RY:
Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores.
J Biol Chem
264:8171, 1989[Abstract/Free Full Text]
23.
Faili A, Randon J, Francischetti IM, Vargaftig BB, Hatmi M:
Convulxin-induced platelet aggregation is accompanied by a powerful activation of the phospholipase C pathway.
Biochem J
298:87, 1994
24.
Towbin H, Staehelin T, Gordon J:
Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose.
Proc Natl Acad Sci USA
76:4350, 1979[Abstract/Free Full Text]
25.
Plow EF, Pierschbacher MD, Ruoslahti E, Marguerie GA, Ginsberg MH:
The effect of Arg-Gly-Asp-containing peptide on fibrinogen and von Willebrand factor binding to platelets.
Proc Natl Acad Sci USA
82:8057, 1985[Abstract/Free Full Text]
26.
Zijenah LS, Morton LF, Barnes MJ:
Platelet adhesion to collagen.
Biochem J
268:481, 1990[Medline]
[Order article via Infotrieve]
27.
Siess W, Lapetina EG:
Ca2+ mobilization primes protein kinase C in human platelets. Ca2+ and phorbol esters stimulate platelet aggregation and secretion synergistically through protein kinase C.
Biochem J
255:309, 1988[Medline]
[Order article via Infotrieve]
28.
Hathaway DR, Adelstein RS:
Human platelet myosin light chain kinase requires the calcium-binding protein calmodulin for activity.
Proc Natl Acad Sci USA
76:1653, 1979[Abstract/Free Full Text]
29.
Daniel JL, Molish IR, Holmsen H:
Myosin phosphorylation in intact platelets.
J Biol Chem
256:7510, 1981[Abstract/Free Full Text]
30.
Okayama M, Oguri K, Fujiwara Y, Nakanishi H, Yonekura H, Kondo T, Ui N:
Purification and characterization of human platelet proteoglycan.
Biochem J
233:73, 1986[Medline]
[Order article via Infotrieve]
31.
Shuttleworth TJ, Thompson JL:
Effect of temperature on receptor-activated changes in [Ca2+]i and their determination using fluorescent probes.
J Biol Chem
266:1410, 1991[Abstract/Free Full Text]
32.
Nakamura J:
pH and temperature resolve the kinetics of two pools of calcium bound to the sarcoplasmic reticulum Ca2+-ATPase.
J Biol Chem
264:17029, 1989[Abstract/Free Full Text]
33.
Nachmias VT, Kavaler J, Jacubowitz S:
Reversible association of myosin with the platelet cytoskeleton.
Nature
313:70, 1985[Medline]
[Order article via Infotrieve]
34.
Bram RJ, Crabtree GR:
Calcium signalling in T-cells stimulated by a cyclophilin B-binding protein.
Nature
371:355, 1994[Medline]
[Order article via Infotrieve]
35.
Daniel JL, Molish IR, Rigmaiden M, Stewart G:
Evidence for a role of myosin phosphorylation in the initiation of the platelet shape change response.
J Biol Chem
259:9826, 1984[Abstract/Free Full Text]
36.
Fernandes JB, Naik UP, Markell MS, Kornecki E:
Comparative investigation of the effects of the immunosuppressants cyclosporine A, cyclosporine G, and FK506 on platelet activation.
Cell Mol Biol Res
39:265, 1993[Medline]
[Order article via Infotrieve]
37.
Naik UP, Markell M, Ehrlich YH, Kornecki E:
Cyclosporine A enhances agonist-induced aggregation of human platelets by stimulating protein phosphorylation.
Cell Mol Biol Res
39:257, 1993[Medline]
[Order article via Infotrieve]
38.
Ueda D, Suzuki K, Malyszko J, Pietraszek MH, Takada Y, Takada A, Kawabe K:
Fibrinolysis and serotonin under cyclosporine A treatment in renal transplant recipients.
Thromb Res
76:97, 1994[Medline]
[Order article via Infotrieve]
39.
Muraki T, Taka T, Noguchi T, Ishii H, Seki J, Yamamoto J:
Effects of cyclosporine and FK506 on in vitro high shear-induced platelet reactivity in rat and human non-anticoagulated blood.
Transplantation
65:1132, 1998[Medline]
[Order article via Infotrieve]
40.
Tegeder I, Schumacher A, John S, Geiger H, Geisslinger G, Bang H, Brune K:
Elevated serum cyclophilin levels in patients with severe sepsis.
J Clin Immunol
17:380, 1997[Medline]
[Order article via Infotrieve]
41.
Smith T, Ferreira LR, Hebert C, Norris K, Sauk JJ:
Hsp47 and cyclophilin B traverse the endoplasmic reticulum with procollagen into pre-Golgi intermediate vesicles.
J Biol Chem
270:18323, 1995[Abstract/Free Full Text]
42.
Rink TJ, Sage SO:
Calcium signaling in human platelets.
Annu Rev Physiol
52:431, 1990[Medline]
[Order article via Infotrieve]
43.
Daniel JL, Dangelmaier C, Smith JB:
Evidence that adhesion of electrically permeabilized platelets to collagen is mediated by guanine nucleotide regulatory proteins.
Biochem J
286:701, 1992
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
B. Richards, J. Karpilow, C. Dunn, I. Peterson, A. Maxfield, L. Zharkikh, M. Abedi, A. Hurlburt, J. Hardman, F. Hsu, et al.
Genetic Selection for Modulators of a Retinoic-Acid-Responsive Reporter in Human Cells
Genetics,
March 1, 2003;
163(3):
1047 - 1060.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. Allain, C. Vanpouille, M. Carpentier, M.-C. Slomianny, S. Durieux, and G. Spik
Interaction with glycosaminoglycans is required for cyclophilin B to trigger integrin-mediated adhesion of peripheral blood T lymphocytes to extracellular matrix
PNAS,
February 20, 2002;
(2002)
52284899.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. M. Joussen, S. Huang, V. Poulaki, K. Camphausen, W.-D. Beecken, B. Kirchhof, and A. P. Adamis
In Vivo Retinal Gene Expression in Early Diabetes
Invest. Ophthalmol. Vis. Sci.,
November 1, 2001;
42(12):
3047 - 3057.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. Allain, C. Vanpouille, M. Carpentier, M.-C. Slomianny, S. Durieux, and G. Spik
Interaction with glycosaminoglycans is required for cyclophilin B to trigger integrin-mediated adhesion of peripheral blood T lymphocytes to extracellular matrix
PNAS,
March 5, 2002;
99(5):
2714 - 2719.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
TOP
ABSTRACT
INTRODUCTION