Prostacyclin Receptor Desensitization Is a Reversible Phenomenon in Human Platelets
Background Long-term exposure of platelets to endogenous or exogenous prostacyclin or its analogues might result in desensitization of the platelet prostacyclin receptor in vitro and in vivo accompanied by a loss in receptor density on the platelet surface and a reduced sensitivity toward the inhibitory effects of prostacyclins. However, the reversibility of this process in platelets has not yet been investigated.
Methods and Results Human platelets desensitized by the chemically stable prostacyclin analogue iloprost showed a significant reduction in [3H]-iloprost binding sites that was reversed by saponin permeabilization. This indicates functionally active internalized prostacyclin receptors. To assess whether the internalized prostacyclin receptors recycle to the cell surface after withdrawal of the agonist, iloprost sensitivity and prostacyclin receptor binding properties of iloprost (30 nmol/L, 2 hours) desensitized platelets incubated in iloprost-free autologous plasma were investigated. While desensitized platelets showed a significant increase in IC50 for iloprost inhibition of thrombin-induced platelet aggregation, serotonin release, and p-selectin expression and a reduced iloprost-stimulated cAMP formation, platelet iloprost sensitivity was restored 3 hours after iloprost withdrawal. In addition, the significant reduction in Bmax and the increase in KD of prostacyclin receptors in desensitized platelets as revealed by [3H]-iloprost binding studies also returned to the initial values.
Conclusions These results indicate that prostacyclin receptors internalized during short-term desensitization are not degraded but can be recycled rapidly to the platelet surface in a functionally active form after withdrawal of the agonist.
Prostacyclin and its chemically stable analogue iloprost are potent inhibitors of platelet function. Both compounds inhibit platelet activation on several levels, including platelet aggregation and release of α-granula content. In addition, these compounds are active vasodilators and exert cytoprotective effects that are not fully understood.1 2 In pathophysiological situations such as unstable angina, urinary values of 2,3-dinor-6-keto–PGF1α, a prostacyclin degradation product, are increased, suggesting elevated levels of circulating endogenous prostacyclin.3 There is evidence suggesting that elevated endogenous prostacyclin might result in platelet prostacyclin receptor desensitization in vivo in patients with acute ischemic heart disease.4 Additionally, when administered intravenously, iloprost shows beneficial effects in patients with peripheral vascular disease.5 However, after prolonged exposure, a decrease in platelet responsiveness to the agonist has been observed in vivo.6 7 8 As a consequence, intrainfusion hyperreactivity of platelets was observed by several investigators. This hyperreactivity was concluded from the elevated ex vivo platelet aggregability observed7 9 10 and an increased plasma level of platelet granula contents7 or TxB2.9
There is evidence that this decreased prostacyclin sensitivity is due to a desensitization of the platelet prostacyclin receptor.8 11 This desensitization that had been observed in vitro12 13 is accompanied by a loss in high-affinity prostacyclin receptors on the platelet surface. However, no significant alterations in prostacyclin receptor affinity were observed by these authors in radioligand binding studies in human platelets. In contrast to some evidence for a restoration of prostacyclin responsiveness after withdrawal of the agonist in vivo,8 the reversibility of the prostacyclin receptor desensitization in human platelets has not yet been demonstrated in vitro. Therefore, the objective of the present study was to investigate whether desensitized platelets are able to regain their initial prostacyclin responsiveness after removal of the agonist in vitro.
In Vitro Desensitization/Resensitization of Human Platelets
Blood was drawn from the antecubital vein of healthy male volunteers and collected in citric acid dextrose (Biostabil, Biotest). Platelet-rich plasma (PRP) was obtained by centrifugation at 180g for 10 minutes. PRP was incubated for 2 hours at 37°C in the presence or in the absence of iloprost. To remove iloprost, an equal volume of TRIS-maleat buffer (200 mmol/L; pH 5.3) was added and platelets were centrifuged at 400g for 10 minutes. The pellet underwent an additional washing step with citrate-buffered saline (pH 6.5) containing 100 mmol/L NaCl, 5 mmol/L KCl, 2 mmol/L CaCl2, 1 mmol/L MgCl2, 30 mmol/L sodium citrate, 5 mmol/L dextrose, and 5% (wt/vol) bovine serum albumin (Sigma). With the use of [3H] -labeled iloprost (Amersham-Buchler), complete removal of the agonist by this washing procedure was confirmed. Platelets were resuspended in autologous platelet-poor plasma obtained by centrifugation of PRP for 10 minutes at 4000g and reincubated for 3 hours. Washed platelets for the different assays were obtained by an identical washing procedure. Platelets were finally resuspended in a modified Tyrode’s medium containing 131.7 mmol/L NaCl, 2.7 mmol/L KCl, 12 mmol/L NaHCO3, 0.415 mmol/L Na2PO4, 1 mmol/L MgCl2, and 10 mmol/L HEPES (pH 7.4) and adjusted to 3×108 platelets/mL.
Platelet Aggregation Assay
Aggregation of washed platelets was measured by a turbidometric method in a Born aggregometer (Apact Fibrintimer) in a volume of 300 μL. CaCl2 was added at a final concentration of 200 μmol/L, and the cell suspension was incubated at 37°C and stirred at 1000 rpm. After 30 seconds of incubation and registration of light transmission, iloprost was added in the concentrations indicated. After 1 minute of stirring, aggregation was induced by addition of 0.1 U/mL thrombin (Sigma). Aggregation was calculated as percent change in light transmission of washed platelet suspension in relation to the light transmission of buffer.
Scatchard Plot Analysis of [3H]-Iloprost Binding Studies
The assay was performed in triplicate experiments with the use of 96-well multititer plates (NUNC) in a final volume of 100 μL. Washed platelets were incubated with [3H]-iloprost in concentrations ranging from 10 to 300 nmol/L for 2 hours at 4°C. Nonspecific binding was determined by the presence of a 1000-fold excess of unlabeled iloprost. Free iloprost was removed by transferring platelet suspensions to glass fiber mats (Filtermat A, Wallace Oy) and washing with Tyrode’s buffer for 10 seconds. After drying in a microwave oven, scintillation cocktail (Rotiszint Eco plus, Roth) was added and the glass fiber mats were counted in a β-counter (1014 LKB Wallac). Specific binding was calculated by subtracting nonspecific binding from total binding of the respective determination. Binding curves and Scatchard plot analysis were performed by using the computer program PRIMER (McGraw-Hill, Inc, Version 1.0).
Washed platelets (2×106 cells/cap) were incubated in modified Tyrode’s buffer for 2 minutes at 37°C. Iloprost was added at final concentrations ranging from 30 pmol/L to 10 nmol/L. After 2 minutes of incubation, thrombin was added at a final concentration of 0.1 U/mL and incubation continued for another 2 minutes. Platelets were fixed with 0.5% paraformaldehyde solution for 30 minutes at room temperature. After washing with PBS, platelets were incubated with a p-selectin (CD62P) reactive monoclonal antibody (mAb) (clone CLB/thr.6, CellSystems). As a positive control, staining with a mAb directed against the constitutively expressed platelet glycoprotein CD61 (clone SZ21, Dianova) was included. To determine unspecific binding, the mAb OX-6 not cross-reacting with human platelets was included. After washing, mAb binding was detected using FITC-labeled goat anti-mouse IgG F(ab′)2 (Medac). After washing, platelets were suspended in 250 μL PBS and analyzed using a FACScan flow cytometer (Becton Dickinson). All incubation steps with the respective antibody were carried out for 10 minutes at room temperature in a volume of 50 μL of PBS supplemented with 5% human serum. All washing steps were carried out with PBS in a volume of 1 mL.
Platelets were incubated in PRP for 10 minutes at 37°C in the presence of 20 μmol/L [3H]-serotonin (Amersham-Buchler; specific activity, 88 Ci/mmol). Noninternalized serotonin was removed by the washing procedure described above. Washed platelets were incubated for 2 minutes in the presence or absence of iloprost in concentrations ranging from 30 pmol/L to 10 nmol/L in a final volume of 100 μL. After addition of thrombin at a final concentration of 0.1 U/mL, incubation was continued for another 2 minutes. Platelets were pelleted by centrifugation at 4000g for 10 minutes at 4°C. Eighty microliters of the supernatant was transferred to counting vials, mixed with 3 mL scintillation cocktail, and counted in a β-counter. Blank values, that is, radioactivity released under identical conditions in the absence of iloprost and thrombin, were subtracted from the individual values. IC50 values for iloprost were calculated as percent radioactivity released in relation to the radioactivity released by thrombin alone. Total serotonin uptake was measured from supernatants of platelet suspensions permeabilized by 1% sodium dodecyl sulfate. Thrombin (0.1 U/mL) released 52.3±4.2% of the [3H]-serotonin internalized.
Washed platelets were incubated in a volume of 400 μL with 100 nmol/L iloprost for 2 minutes at 37°C. The reaction was stopped by the addition of trichloroacetic acid at a final concentration of 5% (vol/vol). After centrifugation for 10 minutes at 12 000g, the supernatant was extracted three times with the quintuple volume of water-saturated diethylether. The aqueous phase was lyophylized and dissolved in assay buffer. cAMP was determined by a commercially available radioimmunoassay kit (Amersham-Buchler).
Results are shown as mean±1 SEM. Statistical comparisons between means were performed by Wilcoxon signed rank tests; a level of P=.05 was regarded as statistically significant.
Effect of Platelet Permeabilization on the Binding of [3H]-Iloprost to Control and Desensitized Platelets
Human platelets incubated for 3 hours in the presence of 100 nmol/L of the prostacyclin analogue iloprost showed a marked reduction in [3H]-iloprost binding sites to 41.4% (P<.01) compared with control platelets, as shown in Fig 1⇓. Receptor binding studies in the presence of 0.01% saponin, a nondenaturating detergent permeabilizing cell membranes, had no effect on prostacyclin receptor density in control platelets (92.0%), making the existence of an internal pool of functionally active prostacyclin receptors in nondesensitized platelets highly unlikely. However, saponin permeabilization elevated [3H]-iloprost binding sites of desensitized platelets to values not significantly different from permeabilized control platelets (91.2%; P=.28). In contrast, while KD values were slightly but significantly increased from 109.6±4.0 to 128.6±17.9 nmol/L upon desensitization, saponin permeabilization had no effect on receptor affinity. Taken together, these observations point to the presence of internalized prostacyclin receptors in desensitized platelets that are still functionally active with respect to their agonist binding properties. Therefore, we decided to investigate the possibility that these receptors could be recycled to the platelet surface after iloprost removal, that is, if platelet prostacyclin receptor desensitization is a reversible phenomenon.
Reversibility of Platelet Prostacyclin Receptor Desensitization at the Functional Level
The kinetics and dose dependency of alterations in platelet iloprost sensitivity during desensitization with three different iloprost concentrations (10 to 100 nmol/L) and after withdrawal of the prostacyclin analogue with respect to inhibition of thrombin-induced p-selectin expression are shown in Fig 2⇓. Iloprost incubation dose-dependently induced iloprost desensitization of platelets evident as increases in IC50 values from 0 to D. While platelet desensitization induced by 10 and 30 nmol/L iloprost was reversible 3 hours after iloprost removal (R3), platelets desensitized by incubation with 100 nmol/L iloprost showed no significant resensitization during the time interval investigated. Therefore, incubation of platelets with 30 nmol/L for 2 hours and incubation of desensitized platelets for 3 hours in iloprost-free plasma was chosen for further experiments.
The IC50 for iloprost inhibition of thrombin-induced platelet aggregation was also significantly increased from 408±26 to 842±68 pmol/L if the platelets were desensitized by iloprost (30 nmol/L) for 2 hours, as depicted in Fig 3⇓. After removal of the agonist and incubation in autologous plasma, platelet iloprost sensitivity was again restored in this assay system and the IC50 value returned to 388±28 pmol/L. Platelets incubated in the absence of iloprost showed no significant alteration in iloprost sensitivity throughout the 5-hour duration of the experiment (Fig 3⇓).
In addition, platelet sensitivity toward iloprost inhibition of thrombin-induced serotonin release decreased significantly after incubation with the prostacyclin analogue. In accordance with the p-selectin and aggregation data, platelet iloprost responsiveness returned to the initial value 3 hours after incubation in iloprost-free plasma, as shown in Fig 4⇓. Again, platelets incubated under identical conditions but in the absence of iloprost were unaltered in their responsiveness toward the antagonist.
Reversibility of Diminished Prostacyclin-Induced cAMP Formation by Desensitized Platelets
As depicted in Fig 5⇓, iloprost-induced platelet cAMP formation was also significantly reduced in desensitized platelets but was no longer decreased 3 hours after iloprost removal when compared with platelets not desensitized but incubated for the same period of time. This is true despite the fact that the overall capacity of platelets to generate cAMP continuously decreased in relation to the incubation time.
Platelet Prostacyclin Receptor Desensitization Is Reversible at the Receptor Level
Receptor binding studies using [3H]-iloprost revealed a significant loss in the density of specific binding sites by 27.9% of control on the platelet cell surface after incubation of platelets with iloprost (30 nmol/L). In addition, a significant decrease in receptor affinity by 38.8% was detected, as shown in the Table⇓. In accordance with the functional data, prostacyclin receptor density and affinity were almost completely restored 3 hours after washout of iloprost (Table⇓).
The desensitization of the platelet prostacyclin receptor after incubation of platelets with exogenous prostacyclin or its analogues in vitro is a well-known phenomenon. A similar desensitization of the platelet prostacyclin receptor has been observed in vivo in patients with acute coronary syndromes such as acute myocardial infarction or unstable angina.4 14 The reversibility of this process has not yet been studied. Because of a mean platelet survival time of ≈7.5 days, an irreversible loss of prostacyclin sensitivity of desensitized platelets would result in platelets being less sensitive to endogenous prostacyclin for a considerable part of their individual life span. This might result in severe thrombotic complications for a period of up to 1 week after treatment with this agonist, comparable to the irreversible acetylation of platelet cyclooxygenase by aspirin. To our knowledge, there is only one report that hints toward a reversibility of prostacyclin receptor desensitization after infusion of iloprost.8 However, the reappearance of prostacyclin sensitivity observed ex vivo by these authors does not necessarily represent an actual resensitization but might be explained by a selective removal of desensitized platelets caused by aggregation or adhesion to the injured vessel wall of the patients. This is probably caused by a loss of sensitivity toward endogenous prostacyclin in these platelets. Furthermore, Sinzinger et al11 were unable to confirm the restoration of iloprost sensitivity ex vivo. Therefore, the objective of the present study was to investigate the reversibility of the prostacyclin receptor desensitization in human platelets in vitro.
Human platelets incubated in plasma in the presence of 30 nmol/L iloprost showed a marked decrease in iloprost sensitivity. The IC50 values for inhibition of thrombin-induced platelet aggregation, serotonin release, and p-selectin expression were significantly increased. In addition, a significant reduction in high-affinity [3H]-iloprost binding sites accompanied by a significant decrease in the capacity of desensitized platelets to form the second-messenger cAMP in response to iloprost stimulation was detected. This complies with previous reports concerning human platelets.12 13 Additionally, a slight but significant increase in KD was observed that is in contrast to the results of Jaschonek et al13 and Alt et al,12 who were unable to detect a significant change in prostacyclin receptor affinity. The reason for this discrepancy, however, is not entirely clear. It might be due to the different desensitization conditions used. While platelet prostacyclin receptor desensitization in our system was achieved by incubation of PRP with 30 nmol/L iloprost for 2 hours, Jaschonek et al and Alt et al used higher iloprost concentrations (1 μmol/L or 100 nmol/L, respectively) and longer incubation periods (24 hours or 12 hours, respectively) at room temperature instead of the physiological temperature of 37° C used in our system. In fact, in our experiments platelets desensitized with a higher iloprost concentration (100 nmol/L) showed no tendency to recover from desensitization during the time interval observed (Fig 2⇑). However, it cannot be excluded that resensitization of these platelets takes more than 4 hours, a phenomenon that could not be studied in vitro because of a marked decrease in platelet viability after the 6-hour in vitro incubation at 37°C. On the other hand, it should be taken into consideration that an iloprost concentration as high as 100 nmol/L will not be achieved in patients because of the side effects of the drug caused by vasodilatation observed after administration of doses higher than 2 ng/kg×minutes.
Our in vitro data demonstrate that the prostacyclin receptor desensitization in human platelets after prolonged contact with moderate concentrations of iloprost is a reversible phenomenon. The sensitivity for iloprost was restored 3 hours after removal of the agonist in all three functional assays as well as in the cAMP formation experiments. In addition, receptor density and affinity was also restored. These results, which also correspond to ex vivo data observed in our laboratory after an iloprost infusion for 4 hours (data not shown), confirm the results of Modesti et al,8 who described a restoration of platelet iloprost responsiveness 6 hours after an infusion of iloprost for 6 hours.
The reversibility of the prostacyclin receptor desensitization has also been shown in cultured NCB-20 and NG 108-15 cells.15 16 In contrast to human platelets, Krane et al16 were able to demonstrate that desensitization of NG 108-15 cells could be partially blocked by the lysosomotrophic drug chloroquin, which impairs lysosomal breakdown. This finding points to an involvement of lysosomal degradation of the internalized prostacyclin receptor in desensitized NG 108-15 cells. In addition, the authors were able to show that the resensitization of these cells is abolished in the presence of the protein synthesis inhibitor cycloheximide. This is in accordance with Leigh and MacDermot,15 showing that the resensitization in NCB-20 cells is also blocked by cycloheximide and by actinomycin D, an inhibitor of transcription. Together these data strongly suggest that de novo protein synthesis is required for resensitization in these cell lines. Because human platelets are cellular fragments devoid of a nucleus, they are incapable of de novo protein synthesis. The fact that platelets fully recovered from prostacyclin receptor desensitization induced by moderate iloprost concentrations makes it likely that the internalized receptor is not degraded after short-term incubation of platelets with iloprost. The experiments performed during this study do not definitely exclude the possibility that a partial loss in internalized receptors occurs after prolonged exposure to the agonist. This could possibly explain the failure of other investigators to show a resensitization after prolonged incubation with iloprost in vivo for 7 days11 or in vitro for 12 hours.13
Our data clearly show that prostacyclin receptor desensitization in human platelets after short time exposure to moderate iloprost concentrations is a reversible phenomenon. Therefore, the loss in prostacyclin sensitivity seems to be a transient effect, persisting <3 hours after termination of exposure to this platelet inhibitor. Therefore, only during this short period an increased risk of platelet aggregate formation and arterial thrombosis might be taken into consideration after prostacyclin or prostacyclin analogue administration or pathophysiological situations resulting in increased levels of endogenous prostacyclin.
The authors wish to thank Anne-Kristin Gröning for expert technical assistance in performing the experiments. The study was supported by the Deutsche Forschungsgemeinschaft, grant No. Da 168/2-1.
- Received September 16, 1996.
- Revision received February 5, 1997.
- Accepted February 7, 1997.
- Copyright © 1997 by American Heart Association
Darius H, Osborne JA, Reibel DK, Lefer AM. Protective actions of a stable prostacyclin analog in ischemia induced membrane damage in rat myocardium. J Mol Cell Cardiol. 1987;19:243-250.
Schrör K, Ohlendorf R, Darius H. Beneficial effects of a new carbacyclin derivative, ZK 36 374, in acute myocardial ischemia. J Pharmacol Exp Ther. 1981;219:243-249.
FitzGerald GA, Smith B, Pedersen AK, Brash AR. Increased prostacyclin biosynthesis in patients with severe atherosclerosis and platelet activation. N Engl J Med. 1984;310:1065-1068.
Kahn NN, Mueller HS, Sinha AK. Impaired prostaglandin E1/I2 receptor activity of human blood platelets in acute ischemic heart disease. Circ Res. 1990;66:932-940.
Grant SM, Goa KL. Iloprost: a review of its pharmacodynamic and pharmacokinetic properties, and therapeutic potential in peripheral vascular disease, myocardial ischaemia and extracorporal circulation procedures. Drugs. 1992;43:889-924.
Sinzinger H, Silberbauer K, Horsch AK, Gall A. Decreased sensitivity of human platelets to PGI2 during long-term intraarterial prostacyclin infusion in patients with peripheral vascular disease: a rebound phenomenon? Prostaglandins. 1981;18:56-58.
Sinzinger H, Horsch AK, Silberbauer K. The behaviour of various platelet function tests during long-term prostacyclin infusion in patients with peripheral vascular disease. Thromb Haemost. 1983;50:885-887.
Modesti PA, Fortini A, Poggesi L, Boddi M, Abbate R, Gensini GF. Acute reversible reduction of PGI2 platelet receptors after iloprost infusion in man. Thromb Res. 1987;48:663-669.
Yardumian DA, Mackie IJ, Bull H, Goldstone AH, Machin SJ. Platelet hyperaggregability occuring during prolonged continuous intravenous infusion of the prostacyclin analogue ZK 36374. Br J Haematol. 1985;60:109-116.
Darius H, Hossmann V, Schrör K. Antiplatelet effects of intravenous iloprost in patients with peripheral arterial obliterative disease. Klin Wochenschr. 1986;64:545-551.
Sinzinger H, Steurer G, Fitscha P, Kraupp O. The intrainfusion rebound platelet activation during continuous prostaglandin I2 infusion occurs at the receptor level. In: Singinger H, Schror K, eds. Prostaglandins in Clinical Research. New York, NY: Alan R. Liss; 1987:19-23.
Alt U, Leigh PJ, Wilkins AJ, Morris PK, MacDermot J. Desensitization of iloprost responsiveness in human platelets follows prolonged exposure to iloprost in vitro. Br J Clin Pharmacol. 1986;22:118-119.
Jaschonek K, Faul C, Schmidt H, Renn W. Desensitization of platelets to iloprost: loss of specific binding sites and heterologous desensitization of adenylate cyclase. Eur J Pharmacol. 1988;147:187-196.
Kahn NN, Mueller HS, Sinha AK. Restoration by insulin of impaired prostaglandin E1/I2 receptor activity of platelets in acute ischemic heart disease. Circ Res. 1991;68:245-254.
Leigh PJ, MacDermot J. Desensitization of prostacyclin responsiveness in a neuronal hybrid cell line: selective loss of high affinity receptors. Mol Pharmacol. 1985;85:237-247.
Krane A, MacDermot J, Keen M. Desensitization of adenylate cyclase responses following exposure to IP prostanoid receptor agonists: homologous and heterologous desensitization exhibit the same time course. Biochem Pharmacol. 1994;47:953-959.