Inhibitory effect of high-density lipoprotein on platelet function is mediated by increase in nitric oxide synthase activity in platelets

Life Sciences, Vol. 55, No. 23, pp. 1815-1821, 1994 Copyright © 1994 Elsevier Science Ltd Printed in the USA. All rights reserved 0024-3205/94 $6.00 + .00

Pergamon 0024-3205(94)00354-8

I N H I B I T O R Y EFFECT OF HIGH-DENSITY L I P O P R O T E I N ON P L A T E L E T FUNCTION IS MEDIATED BY INCREASE IN NITRIC OXIDE SYNTHASE ACTIVITY IN PLATELETS L.Y. Chen and J. L. Mehta From the Department of Medicine, College of Medicine, University of Florida and the VA Medical Center, Gainesville, Florida (Received in final form September 29, 1994) Summary Although high-density lipoprotein (HDL) has been found to decrease platelet function per se, little is known regarding the mechanism of its platelet inhibitory effect. In this study, we confirmed the inhibitory effect of HDL on platelet aggregation and 14C-serotonin release in thrombin-activated washed human platelets. The inhibition of platelet function was associated with an increase in nitric oxide synthase activity, measured as the conversion of 3H-k-arginine to 3H-L-citrulline as well as nitlite release in the platelet supemates. The inhibition of platelet function by HDL was reversed by preincubation of washed platelets with an inhibitor of nitric oxide synthase, NG-nitro-L-arginine methyl ester (CNAME), and potentiated by co-incubation with the precursor of nitric oxide, Carginine. These observations suggest that HDL decreases platelet function by increasing nitric oxide synthase activity in human platelets.

Key Words: high density lipoproteins, platelcts, nitric oxide Platelet hyper-reactivity and impairment of vasorelaxation are characteristics of atherosclerosis. Platelet activation contributes to atherogenesis by facilitating formation of thrombus and release of growth factors (1). Hypercholesterolemia, a hallmark of atherosclerosis, per se enhances platelet activity (2) and reduces endothelium-derived nitric oxide (NO) mediated vasorelaxation(3). On the other hand, high density lipoprotein (HDL) antagonizes the inhibitory effects of oxidized low density lipoprotein (LDL) on soluble guanylyl cyclase in the endothelium and NO-mediated impaired relaxation in rabbit aorta (4, 5). NO has now been identified as a potent inhibitor of platelet function. The inhibitory effect of NO on platelet function is associated with elevation of intracellular cyclic GMP (cGMP) levels (6). There is convincing evidence that platelets themselves generate NO (7). This study was designed to examine the relationship between platelet function, NO synthase activity and HDL in washed human platelets. Materials and Methods

Reagents: 2,3,4,5-3H-L-arginine (69 Ci/mmol; 1 Ci=37 GBq; 1.0 BCi/I.tL) was obtained from Amersham, Arlington Heights, IL. Other chemicals were purchased from Sigma\. Preparation of HDL: HDL (d-1.063-1.21 g/mL) was isolated from plasma obtained from normolipemic lasting volunteers by stepwise ultracentrifugation as described previously (8). Reprint requests to: J.L. Mehta, M.D., Ph.D., University of Florida College of Medicine, P.O. Box 100277 JHMHC, Gainesville, FL 32610-0277

1816

HDL and Nitric Oxide Synthase Activity

Vol. 55, No. 23, 1994

Platelet Aggregation and ~4C-Serotonin in Washed Platelets : Washed platelets were prepared as described previously (9) and were incubated with t4C-serotonin (1 ~tCi/mL) for ! hr at room temperature, and washed twice. Washed platelet aliquots were incubated with buffer, HDL (10, 25, 50 and 100 ~tg protein/mL), t.-arginine (1 mM), L-NAME (1 mM) and HDL (100 lag protein/mL) plus L-NAME (1 raM) for 1 hr at 37°C. In parallel experiments, HDL (50 lag protein/mL) plus L-arginine (1 mM) with or without L-NAME (1 mM) were incubated with washed platelets for 1 hr. After incubation, platelet aggregation was induced by thrombin (threshold concentration 0.5-1 U/mL) in a dual-channel aggregometer as described previously (10). The threshold concentration of thrombin was kept constant and repeatedly checked during the experiment to ensure that it did not change in each experiment. EDTA (13.4 raM) was added to washed platelets at 5 min after the onset of aggregation, and the sample was centrifuged at 800 g for 15 rain. Supernatant (175 gL) was removed for scintillation counting. An aliquot of washed platelet suspension with 14C-serotonin was saved for total count. Static release count was expressed as described earlier (9). Determination of NO Synthase Activity: In control experiments, time course of L-arginine uptake was measured by incubating washed platelets with 3H-L-arginine at 37°C from 5 to 60 rain. The uptake of 3H-L-arginine had a linear increase up to 45 min and it stabilized thereafter. NO Synthase activity was determined as conversion of 3H-L-arginine to 3H-L-citrulline (11). Briefly, washed platelets (107/mL) were incubated with 3H-L-arginine (7.25 nM: average count 1,000,000 dpm) with or without HDL (10, 25, 50 100, 150 and 200 ~tg protein/mL) for 60 rain in 1 mL NO buffer (composition in mM: Hepes 25, NaC1 140, KC1 5.4, CaC12 1, MgC12 1, pH 7.4) at 37°C. In parallel experiments, platelets were preincubated with thrombin (1U/ml) for 5 rain before added 3H-L-arginine and HDL. The reaction was stopped with 1 mL cold buffer (composition in mM: Hepes 25, NaC1 118, KC1 4.7, KH2PO4 1.18, NaHCO3 24.8, EDTA 4, N~°-nitro-L-arginine 5, pH 5.5) and centrifuged at 800 g for 20 min at 4°C. Supernate was discarded and the pellet was disrupted by adding 1 mL of 0.3 M HC104 and neutralized with 65 ~L of 3 M K2CO3. An aliquot was applied to a 2 mL column of Dowex AG 50W-X8 (Na+ form) which was eluted with 6 mL water. 3H-L-citrulline in the eluent was quantified by liquid scintillation spectroscopy. The 3H-L-citrulline in the eluent has been confirmed previously by thin layer chromatography (11). The conversion of 3H-L-arginine to 3H-Lcitrulline was calculated as: 3 H - L - c i t r u l l i n e counts in the eluent x 100 3 H - L - a r g i n i n e counts in disrupted platelet suspension Determination of Nitrite in Platelets : Nitrite in platelets was measured by the Griess reaction (12). In brief, washed platelets (108 cells/mL) were suspended in NO buffer (composition in mM: Hepes 25, NaCI 140, KC1 5.4, CaC12 1, MgC12 l, pH 7.4) containing 1.44 mM NADPH and incubated with buffer, HDL (10, 25, 5(1 and 100 ~g protein/mL), HDL (100 gg protein/mL) plus L-NAME (1 raM), arid HDL (50 lag protein/mL) plus L-arginine (1 raM) with or without L-NAME (1 raM) for 1 hr at 37°C. The reaction was stopped by fi'eeze-thawing the sample, which was then sonicated. Each sample was incubated for I hr at 37°C after addition of nitrate reductase (20 mU), which reduces nitrate to nitrite. After centrifugation at 10,000 rpm for 15 rain, the supernate was allowed to react with the Griess reagent (1% sulfanilamide/0.1% naphthylenediamine dihydrochloride/2.5% H3PO4). The chromophore absorption was read at 543 nm. Nitrite concentration was determined with sodium nitrite in water as standard. Statistics: All data are expressed in mean_+SE. Statistical analyses were performed using ANOVA or Student's t-test (paired or unpaired data), as appropriate, and correlation coefficient.

Vol. 55, No. 23, 1994

HDL and Nitric Oxide Synthase Activity

1817

R%ul~ Effects of HDL on Platelet Aggregation and Serotonin Release: Incubation of HDL with washed platelets resulted in inhibition of aggregation. This effect of HDL was dependent on the concentration of HDL, potentiated by L-arginine and inhibited by L-NAME. A representative experiment is shown in Figure 1 and data from multiple experiments are summarized in the Table.

fControl

L-NAME 1mM+ HDL 10?~g/mL

HDL 50 IJg/mL

L-arginine 1 mM

HDL 100 IJg/mL

L-arginine + HDL 50 IJg/mL

[3b

<

A

A

A

i

Thrombin Fig. 1. Representative experiments showing inhibition of platelet aggregation by HDL, inhibition of platelet aggregation inhibitol2¢ effect of HDL by L-NAME, and its potentiation by L-arginine. Release of 14C-serotonin from thrombin-stimulated washed platelets was also reduced by HDL in a concentration-dependent manner. L-NAME antagonized the inhibitory effect of HDL on 14C-serotonin release and L-arginine enhanced this effect of HDL. Co-incubation with L-arginine overcame the effect of L-NAME (Table). In control experiments, L-arginine alone also decreased platelet aggregation and 14C-serotonin release, whereas L-NAME increased platelet aggregation and 14C-serotonin release. Thge effects of HDL plus L-arginine were abolished by coincubation with L-NAME.

Effect of HDL on NO Synthase Activity: HDL increased the conversion of 3H-L-arginine to 3H-L-citrulline both in thrombin stimulated and unstimulated platelets, indicating increase in NO synthase activity. The increase in NO synthase activity was dependent on the concentration of HDL. Whereas in unstimulated platelets, high concentrations of HDL was required to identify an increase in NO synthase, HDL at concentration of 25 ~g protein/ml significantly increased the conversion of 3H-L-arginine to 3H-L-citrulline in thrombin-stimulated platelets (Fig. 2, 3). HDL, however, had no effect on the uptake of 3H-L-arginine in platelets (data not shown). The increase in NO synthase activity was reflected in nitrite formation in platelet supemates. As shown in the Table, the release of nitrite also almost doubled in response to 100 gg proteirdmL of HDL. The nitrite concentrations decreased in the presence of L-NAME and Larginine negated the effect of L-NAME. The increase in nitrite levels COlTelated inversely with the decrease in platelet aggregation (r-0.88) as well as with the decrease in 14C-serotonin release (r-0.78) (Fig. 4).

1818

HDL and Nitric Oxide Synthase Activity

Vol. 55, No. 23, 1994

TABLE Effect of HDL, L-NAME and L-arginine on Platelet Function and Nitrite Formation Platelet Aggregation (%) 45_+1

Control HDL 10 gg protein/mL

14C-Serotonin Release (%) 20.8_+3.5

Nitrite (nmoles/mg protein) 3.43_+0.27

42_+1

15.3_+0.9*

3.93+0.10

25 ~tg protein/mL

34_+3*

12.8+3.4"

3.87_0.27

50 gg proteirdmL

24_+5*

7.8+0.7 *

5.04+0.07

100 ~tg protein/mL

2(~_4"

5.5_+0.9

6.07_+-+0.19"

L-arginine (1 mM)

32_+5*

11.5_+1.6"

ND

L-NAME (1 mM)

54_+1"

33.6_+3.2*

ND

HDL (100 lag protein/mL) + L-NAME (1 mM)

39_+3t

14.3-+2.7~-

HDL (50/.tg/mL) +L-arginine (1 mM)

15-+59

6.4+1.6

4.85_+0.99*

HDL (50 ~tg proteirl/mL) +

33_+4§

13.6+3.9

3.7(~0.32§

3.59_+0.49t

L-NAME (1 mM) + L-arginine (1 mM) Data from 3-5 experiments in mean_+SE. *P<0.05 vs control, ?P<0.05 vs HDL alone, §P<0.05 vs HDL+L-arginine; Abbreviations: L-NAME - NG-nitro-L-arginine methyl ester 45 o~ 4 0 ¢ 35 0 u) 3 0 I,.

> 25 ¢0 o 20 r-15 t,,'5~ 1 0 I,.

?

,_1

s o ........

- ......

HDL

(/.tg/mL)

........

Fig. 2 Effect of HDL on conversion of 3H-L-arginine to 3H-L-citrulline in unstimulated platelets. Data from 4 experiments in mean+SE. Concentration _<50 ~tg protein/mL had no effect.

Vol. 55, No. 23, 1994

HDL and Nitric Oxide SynthaseActivity

1819

o~ *P
c 0

o l

> c 0 rr-

o ~

.,I 0

Control

10

25

.......... HDL(txg/ml

50

100

)..........

Fig. 3 Effect of HDL on conversion of 3H-L-arginine to 3H-L-citrulline in thrombin stimulated platelets. Data in mean~+SE. Discus~i0n Epidemiologic studies have shown high levels of oxidized LDL and low levels of HDL to be associated with atherosclerosis. HDL exerts antiatherosclerotic effects by removal of LDL. Experimental studies have indicated that HDL reverses the decrease of NO-mediated vasorelaxation found in hyperlipidemia (4). Platelet hyperactivity, a characteristic of hyperlipidemia, is also reversed by HDL (13, 14). We confirmed the inhibitory effects of HDL on platelet function, measured as thrombin-induced aggregation and t4C-serotonin release in this study. In addition, we show that the inhibitory effect of HDL on platelet function is mediated by increase in endogenous NO synthase activity. Radomski et al (7) first showed that platelets possess L-arginine-NO pathway, which regulates platelet activity. The platelet aggregation inhibition by L-arginine, precursor of NO, is mediated by increase in guanylyl cyclase activity (15). The presence of L-arginine-NO pathway as a mechanism of platelet inhibitory effects of HDL in our study became evident as NO synthase inhibitor L-NAME attenuated the effects of HDL and L-arginine enhanced the effects of HDL. Notably, co-incubation with L-arginine overcame the effect of L-NAME. The effects of HDL on platelet NO synthase activity were quantitated by two different methods-conversion of 3H-L-arginine to 3H-L-citrulline and measurement of nitrite levels in the supernates of sonicated platelets. Using both techniques we found HDL concentrationdependent increase in NO synthase activity. Furthermore, the increase in NO synthesis correlated with reductions in platelet aggregation and 14C-serotonin release. The inhibitory effects of low concentrations of HDL (10 and 25 Bg protein/mL) on platelet aggregation, serotonin release and NO synthase activity in thrombin stimulated platelets were quite significant. However, only >50 protein I.tg/ml concentration of HDL increased NO synthase activity in unstimulated platelets. These observations suggest that stimulated platelets exhibit greater NO synthase activity than the unstimulated platelets, probably due to the release of stored calcium which can activate constitutive NO synthase. Indeed, a recent preliminary study (16) demonstrated marked increase in NO synthase activity in stimulated platelets.

1820

HDL and Nitric Oxide Synthase Activity

Vol. 55, No. 23, 1994

Moncada et al (15) also suggested greater NO synthase activity in stimulated than in unstimulated platelets. 50 45

•

•

°Q ° 0

40

35 im

30 25, 20 15

3

3;s

4~s

;

Nitrite ( n m o l e s / m 9 protein)

s;s

~

G;5

27.5

25

g

22.5 20

•

•

vv

3 IS

4

g, 17.s '~

12.5

7.5 5 2.5

I

4 ;s

I

;

I

S.Is

"

;

"

~.l s

Nitrite ( n m o l e s / m g protein)

Fig. 4 Correlation between nitrite levels in platelet supernates and platelet aggregation and 14C-serotonin release under the influence of HDL. Maximal concentrations of HDL employed in this study are lower than the physiological concentrations of HDL (400-800 btg cholesterol/ml). HDL even at 10-200 btg protein/ml concentrations was found to be effective in ]educing platelet activation and stimulating NO synthase activity in this study. It is likely that higher concentraions of HDL have a more pronounced effect than reported here. Although NO synthase activity correlated with the decrease in platelet function in the presence of HDL, other effects of HDL, such as its effect on cholesterol transport (17) and inhibition of LDL oxidation (18), may also relate to its overall effects on platelet function. HDL also stimulates release and stabilization of PGI2 (19), which inhibits platelet function. However, platelets do not synthesize PGI2 and this mechanism may not be operative in in vitro conditions employed by us. In summary, this study provides evidence for increase in NO synthasc activity as a mechanistic basis of HDL-induced reduction in human platelet aggregation and serotonin release.

Vol. 55, No. 23, 1994

HDL and Nitric Oxide Synthase Activity

Reference~ 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

R. ROSS N Engl J Med ~14488-500 (1986). N. CHETI'Y and N.H. NARANG Clin Chim Acta 2131-13 (1992). T.J. VERBEUREN, F.H. JORDAENS, L.L. ZONNEDEYN, C.E.VANHOVE, M.C. CORNE and A.G. HERMAN Circ Res 58, 552-564 (1986). K. SCHMIDT, P. KLATT, W.F. GRANIER, G.M. KOSTNER and W.R. KUKOVETZ Biochem Biophys Res Commun 18_._22302-308 (1992). F. PLANE, K.R. BRUCKDORFER, P. KERR, A. STEUER and M. JACOBS Brit J Pharmacol 10.___55216-222 (1992) B.T. MELLION, L.J. IGNARRO, E.H. OHLSTEIN, E.G. PONTECORVO, A.L. HYMAN and P.J. KADOWITZ Blood 57946-955 (1981). M.W. RADOMSKI, R.M.J. PALMER and S. MONCADA Br J Pharmaco1101325328 (1990). G.M. KOSTNER and P. ALANPOVIC Biochemistry 1_[13419-3428 (1972). L.Y. CHEN and J.L. MEHTA Thromb Res 74555-563 (1994). F.A. NICOLINI, A.C. WILSON, P. MEHTA and J.L. MEHTA J Lab Clin Med 116 147-152 (1990). J.B. BRYANT, P. MEHTA, A. VON DER PORTEN and J.L. MEHTA Biochem Biophys Res Commun 189558-564, (1992). L.C. GREEN, D.A. WAGNER, J. G L O G O W S K I , P.L. SKIPPER, J.S. WISHNORK and S.R. TANNENBAUM Ann Biochem 126131-138 (9182). M. HIGASHIHARA, M. K. NOSHITA, S. KUME, T. TERAMOTO and K. KUROKAWA Biochem Biophys Res Commun 1811331-1336 (1991). T.A. PEREIRA, G.S. SHAH and N.P. DAS Biochem Med Met Biol 4___55326-332 (1991). S. MONCADA, R.M.J. PALMER and A.J. HIGGS Pharmacol Rev 43109-142, (1991). Q. ZHOU, G.R. HELLERMAN and L.P. SOLOMONSON Proc Sci Con Mol Cell Biol Vasc Wall AHA (abst) p. 67 (1993). G. SCHMITZ, T. BRUNING, E. WILLIAMSON and G. NOWICKA Eur Heart J ll(Sut~ol E) 197-211 (1990). S. PAi~-THASARATHY, J. BARNETI? and L.G. FONG Biochim Biophys Acta 1044 275-283 (1990). A.M. LEFER Circulation 812013-2015 (1990).

1821