Thrombosis Research 94 (1999) 307–316

REGULAR ARTICLE

l-Arginine Modulates Aggregationand Intracellular Cyclic 39,59-GuanosineMonophosphate Levels in Human Platelets:Direct Effect and Interplay with Antioxidative Thiol AgentGiovanni Anfossi, Isabella Russo, Paola Massucco,Luigi Mattiello, Paolo Perna, Francesco Tassone and Mariella TrovatiDiabetes Unit, Department of Clinical and Biological Sciences,University of Turin-Azienda Ospedaliera S. Luigi, Orbassano (TO), Italy.

(Received 24 July 1998 by G.F. Gensini; revised/accepted 11 December 1998)

enhanced l-arginine effect. l-arginine (at 120–500Abstractmmol/L) increased cyclic 39,59-guanosine mono-phosphate levels in resting platelets; the amino acidPlatelet nitric oxide is involved in the control ofalso determined an increase of cyclic 39,59-guano-aggregability via cyclic 39,59-guanosine monophos-sine monophosphate in platelets at the end of aden-phate synthesis. Since l-arginine provides a gua-osine 5-diphosphate-induced aggregation. Nitricnidino nitrogen group for nitric oxide synthesisoxide synthase inhibitor NG-monomethyl-l-argi-through nitric oxide synthase activity, we tried tonine prevented l-arginine effects on aggregationclarify whether an increased availability of thisand cyclic 39,59-guanosine monophosphate synthe-amino acid can directly modulate the response ofsis. Phosphodiesterase III inhibitor milrinone andhuman platelets. In our conditions, l-arginine (atantioxidative thiol N-acetyl-l-cysteine enhanced100–6000 mmol/L) was able to influence the re-the effect of l-arginine on cyclic 39,59-guanosinesponse of human platelets stimulated with adeno-monophosphate. In conclusion, l-arginine exertssine 5-diphosphate and collagen both in PRP andinhibitory effects on human platelet responsein whole blood. The anti-aggregating effect wasthrough a nitric oxide-dependent synthesis of cyclicnot present when d-arginine was used. Permeabil-39,59-guanosine monophosphate. A positive inter-ized platelets exhibited an increased sensitivity toplay on platelet response between l-arginine andl-arginine. Also, an increased availability of Ca21

milrinone or antioxidative thiol N-acetyl-l-cysteinewas evidenced. 1999 Elsevier Science Ltd. Allrights reserved.Part of this article was presented during the 30th Annual Scientific

Meeting of the European Society for Clinical Investigation (In-terlaken, April 24–27, 1996) and has been published as an abstract

Key Words: l-arginine; N-acetyl-l-cysteine; Nitric oxide;in the European Journal of Clinical Investigation (1996;26:A5).Nitric oxide synthaseAbbreviations: ADP, adenosine 5-diphosphate; cGMP, cyclic 39,59-

guanosine monophosphate; cNOS, constitutive nitric oxide syn-thase; l-NMMA, NG-monomethyl-l-arginine; NO, nitric oxide;

Nitric oxide (NO) acts as a potent inhibitorNOS, nitric oxide synthase; GTN, glyceryl trinitrate, nitroglycerin;GTP, guanosine triphosphate. of platelet responses to agonists via activa-Corresponding author: G. Anfossi, Diabetes Unit, Department of tion of soluble guanylate cyclase, which in-Clinical and Biological Sciences, University of Turin, Azienda

creases intracellular levels of cyclic 39,59 guanosineOspedaliera S. Luigi, I-10043 Orbassano (TO), Italy. Tel: 139(11) 902 6622, 139 (11) 902 6612; Fax: 139 (11) 903 8639. monophosphate (cGMP) [1–3].

0049-3848/99 $–see front matter 1999 Elsevier Science Ltd. All rights reserved.PII S0049-3848(99)00011-0

308 G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

Since a constitutive, calcium-dependent NO syn- and interferes with the platelet responses [24,25],we conducted all experiments in PRP and wholethase (cNOS) has been identified in human plate-

lets [4–6], different reports have provided evi- blood where blood proteins can reduce peroxy-nitrite formation through SH groups of proteins.dences that NO is synthesized [7,8] and is involved

in the control of platelet response and in overall Finally, we investigated the interplay with antioxi-dative thiol N-acetyl-l-cysteine, which is able toresponse to different agonists [1–4,9–11]. In partic-

ular, NO-dependent increase of intraplatelet cGMP break peroxynitrite to a S-nitrosothiol [26]. Inother studies, N-acetyl-l-cysteine has been shownis detectable at the end of aggregation induced by

different agonists, such as ADP, sodium arachido- to potentiate the inhibition of platelet aggregationby nitroglycerin [26–28] and to enhance the effectsnate, collagen, and thrombin [12–14], representingof NO donors and endogenous NO [29].a negative feedback control mechanism [14].

Since l-arginine provides a guanidino nitrogengroup for NO synthesis through NOS activity [2,3,

1. Research Design and Methods15,16], previous reports tried to clarify whether anincreased availability of this amino acid modulates

1.1. Chemicalsthe response of human platelets either in vivo orin vitro.

Adenosine 5-diphosphate (ADP) sodium salt, acidRecently, infusion or oral administration of l-argi-soluble collagen, l-arginine, d-arginine, NG-mono-nine in healthy subjects was shown to reduce plate-methyl-l-arginine (l-NMMA), calcium ionophorelet aggregability evaluated ex vivo [17,18]. TheseA23187, ionomycin, milrinone, saponin, and N-acetyl-studies, however, do not allow us to separate thel-cysteine were obtained from Sigma Chemical Co.effects on endothelium-derived NO from those on(St Louis, MO, USA). Glyceryl trinitrate was fromplatelet NO.Simes S.p.A. (Milan, Italy). Stock solutions of bothIn vitro, inhibition of platelet aggregating re-calcium ionophore A23187 and ionomycin weresponse to collagen and other agonists by preincuba-prepared in dimethylsulfoxide (DMSO); the sub-tion with l-arginine was shown in washed plateletsstances were further diluted in phosphate buffered[2,3], in PRP [19,20] and in whole blood [3]. Thesaline before use, and final concentration of DMSOdifferent studies indicated that l-arginine concen-did not exceed 20 mmol/L: this concentration, ontrations necessary to obtain antiaggregating effectthe basis of our previous studies, is unable to mod-are relatively high [19,20].ify platelet response.The mechanism involved in the effect of l-argi-

nine on platelet function has not been fully clarifiedsince availability of l-arginine is not considered a 1.2. Subjects and Platelet Preparationrate-limiting step for NO synthesis [21]. Further-more, l-arginine effect on cGMP has not been com- The studies were performed on 16 healthy male

volunteers (mean age: 34.762.0 years), who deniedpletely elucidated [3].In the present study, we aimed to further investi- taking any drugs for the previous 2 weeks. The

subjects gave informed consent, and the study wasgate whether an increased availability of l-argininethrough exogenous addition can influence the plate- approved by the Ethics Committee of our depart-

ment. The subjects were investigated after over-let function.In particular, we evaluated: (1) the direct effects night fasting. Venous blood samples were withdrawn

without stasis, anticoagulated with 3.8% trisodiumof l-arginine on platelet aggregation and cGMP;(2) the modulation of l-arginine effects by NOS citrate, pH 7.4 (blood/anticoagulant: 9/1), and used

for the investigations both in PRP and in wholeinhibitor NG-monomethyl-l-arginine [22] and phos-phodiesterase III inhibitor milrinone [23]; (3) the blood.

PRP was obtained by centrifugation at 1003grole of calcium ionophores in responsiveness tol-arginine. for 20 minutes. Platelet-poor plasma (PPP) was

obtained by further centrifugation at 8003g forSince some results obtained in washed plateletscan be influenced by the rapid reaction of NO with 10 minutes. Platelet counts were performed on a

model S-plus Coulter Counter (Coulter Electron-superoxide ion to form peroxynitrite, which is toxic

309G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

ics, Hertfordshire, UK); the platelet number ranged physiologic saline, and the tests were performed ina Chrono Log Whole Blood Aggregometer, Modelbetween 250,000 and 300,000/mL in PRP samples,

and it was not adjusted since the study design con- 500 at a constant stirring of 1000 rpm.Briefly, 1 mL of diluted whole blood was trans-sisted of the determination of platelet responses

in samples from the same PRP after addition of ferred to a plastic cuvette preheated to 378C; aTeflon-coated bar was added, and the electrodesbuffer solution or different drug concentrations for

each subject. were placed in the cuvette. The sample was brieflyallowed to reach equilibrium, and the instrumentwas calibrated. Aggregation was induced with 201.3. Protocolsmmol/L ADP or 8 mg/L collagen and was recordeduntil the maximal amplitude was reached. The maxi-1.3.1. Study Amum increase in resistance, expressed in ohms,Platelet aggregation was first evaluated on PRPwas calculated.according to Born’s method [30].

On the basis of results from experiments in PRP,Experiments were conducted on a model 500l-arginine (30–500 mmol/L) was added 20 minutesChrono Log aggregometer (Chrono Log Corp.before the addition of the aggregating agent. Ag-Havertown, PA, USA) at a constant stirring of 900gregation was prolonged for 10 minutes. We con-rpm . The percentage of light transmission was setducted the experiments also by incubation withat 10% with PRP and at 90% with PPP. Eachd-arginine for 20 minutes.aggregation rate was evaluated as an increase in

In some experiments, the samples were exposedlight transmission. Aggregation was induced byto 3-minute preincubation with 100 mmol/LADP (4 mmol/L) and collagen (8 mg/L). In orderl-NMMA followed by 20-minute incubation withto evaluate time- and dose-dependent effects, dif-l-arginine and stimulation with ADP (20 mmol/L)ferent concentrations of the amino acid were addedor collagen (8 mg/L).to PRP samples at 378C 3–20 minutes before the

addition of aggregating agent. We did not prolongincubation over 20 minutes to avoid a spontaneous 1.3.3. Study Creduction of platelet aggregability. To verify whether l-arginine influences intra-

To compare the different responses, maximal platelet concentrations of cGMP, the PRP samplesaggregation (MA) was quantified by Weiss’ for- (500 mL) were incubated without stirring for 20mula (ODo-ODm)3100/ODo, where ODo was the minutes with different concentrations of l-arginineinitial light absorption and ODm the minimum (30–500 mmol/L) or 100 mmol/L GTN (to verifylight absorption [31]. Data are expressed as a per- platelet responsivity).centage of MA induced by the aggregating agent In other experiments, PRP samples were prein-in the absence of any inhibition. cubated for 10 minutes with l-arginine (100 and

Some experiments were repeated using perme- 1000 mmol/L) and stimulated with ADP (4 mmol/L);abilized platelets prepared following previous indi- aggregation was performed for 8 minutes.cations [32,33]. Briefly, PRP samples were incu- Platelet reactions were stopped with 100 ml ofbated with saponin at the final concentration of 25 ice-cold 30% trichloroacetic acid. Precipitated pro-mg/mL for 10 minutes at 308C. Different concentra- teins were removed by centrifugation at 20003gtions of l-arginine (50–1000 mmol/L) were added for 10 minutes at 48C. After addition of 100 mLto PRP samples at 378C 3–5 minutes before the 1 mol/L HCl, the supernatant was submitted to 10addition of ADP (4 mmol/L). Aggregation was con- ether extractions and lyophylized. Samples wereducted as previously described. stored at 2708C until assay. Before determination,

the powder was dissolved in 300 mL of 50 mmol/Lsodium acetate buffer, pH 6.2. The determination1.3.2. Study Bwas performed by radioimmunoassay by using aTo further investigate the platelet response toKit from Advanced Magnetics Inc. (Cambridge,l-arginine, we conducted additional experimentsMA, USA). The sensitivity was less than 0.3 fmol/by whole blood aggregometry in accordance to the0.1 ml (acetylated method); the cross-reactivity ofmethod of Cardinal and Flower [34].

Citrated blood samples were diluted 1:1 with the antibody was less than 0.66% with cAMP and

310 G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

less than 0.001% with GMP, GDP, ATP, and GTP;the intra-assay coefficient of variation was 5.3%.

1.3.4. Study DTo evaluate the role of increased calcium availabil-ity, we utilized calcium ionophore A23187 and iono-mycin. In particular, we conducted the experimentby preincubation (1 minute) with A23187 (10 nmol/L)or ionomycin (1 mmol/L) followed by incubationof l-arginine (100 and 500 mmol/L) for 10 and20 minutes. Aggregation was induced by ADP asdescribed in study A.

1.3.5. Study EIn other experiments, PRP samples were preincu-bated for 3 minutes with milrinone (10 mmol/L),l-NMMA (100 mmol/L) or N-acetyl-l-cysteine (3and 6 mmol/L) and then incubated for 20 minuteswith l-arginine at 30–120 mmol/L. cGMP was mea-sured as in study C.

1.4. Statistical Analysis

Fig. 1. Effect l-arginine preincubation for 3, 10, and 20All values in this study are presented as means6minutes on platelet aggregation induced by 4 mmol/L ADPSEM. When experimental design consisted of re-in PRP samples. Data are expressed as mean6SEM, n58.peated measures of platelet responses to differentSignificance: 3-minute preincubation: ANOVA for re-drug concentrations in each subject, an analysis of peated measures p,0.05, Dunnett’s test: p,0.05 control

variance for repeated measures (ANOVA) and a vs. l-arginine at 2000 and 6000 mmol/L; 10-minute prein-Dunnett’s test were applied. Student’s t-test for cubation: ANOVA for repeated measures p,0.0001, Dun-

nett’s test: p,0.05 control vs. l-arginine at 100, 500, 2000,paired data was used when two measures had toand 6000 mmol/L; 20-minute preincubation: ANOVA forbe compared. A p-value less than 0.05 was consid-repeated measures p,0.0001, Dunnett’s test: p,0.05 con-ered significant.trol vs. l-arginine at 100, 500, 2000, and 6000 mmol/L.

2. Results6000 mmol/L (Dunnett’s test: p,0.05 vs. control);a 10-minute preincubation was required to evi-The employed l-arginine concentrations were un-

able to modify pH values in PRP samples, pH val- dence an effect of l-arginine at 100 and 500 mmol/L.ues being 7.39660.045 in the controls, 7.38860.024, Collagen-induced aggregation was decreased by7.37860.022, 7.40060.030, 7.37760.015, and 7.3676 PRP preincubation with l-arginine (10 minutes)0.020 with 30, 60, 150, 500, and 2000 mmol/L, re- in a dose-dependent way. Platelet response (per-spectively (not significant; n55). centage of control) in the presence of l-arginine

was: 90.263.1 with 100 mmol/L, 77.064.1 with 500mmol/L, and 69.365.9 with 1000 mmol/L (ANOVA2.1. Study Afor repeated measures: p,0.0001; Dunnett’s test:p,0.05 with l-arginine at 500 and 1000 mmol/L).The experiments with PRP showed both time- and

Permeabilized platelets showed a normal re-dose-related effects of l-arginine as shown in Fig-sponse to ADP (MA: 55.866.7 vs. 53.464.3 in non-ure 1. After a 3-minute preincubation, a significantpermeabilized PRP; p,0.757). In this condition,decrease of platelet response to ADP was evi-

denced with l-arginine concentrations of 2000 and we evidenced a significant antiaggregating effect

311G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

Fig. 2. Effect l-arginine preincubation for 3 and 5 minuteson platelet aggregation induced by 4 mmol/L ADP inpermeabilized PRP samples. Data are expressed asmean6SEM, n56. Significance for 3-minute and 5-minutepreincubation: ANOVA for repeated measures p,0.0001,Dunnett’s test: p,0.05 control vs. l-arginine at 50, 100,

Fig. 3. Effect of a 20-minute preincubation with l-arginine500, and 1000 mmol/L.on platelet aggregation induced by 20 mmol/L ADP (A)and 8 mg/L collagen (B) in whole blood samples. Maximalamplitude values (ohms) in control samples were: 15.861.9

of l-arginine (50–1000 mmol/L) after a 3-minute with ADP and 18.261.1 with collagen. Data are expressedas mean6SEM, n58. Significance for ADP-induced aggre-preincubation (Figure 2).gation: ANOVA for repeated measures p,0.0001, Dun-nett’s test: p,0.05 control vs. l-arginine at 30, 60, 150, and

2.2. Study B 500 mmol/L for collagen-induced aggregation: ANOVAfor repeated measures p,0.0001, Dunnett’s test: p,0.05

In whole blood, a 20-minute preincubation with control vs. l-arginine at 60, 150, and 500 mmol/L.l-arginine decreased in a dose-dependent fashionthe aggregating response to ADP and collagen(Figures 3A and 3B).

In the same conditions, the organic nitrate GTNl-NMMA blunted the inhibitory effects of l-argi-at 100 mmol/L markedly increased the cGMP con-nine on agonist-induced aggregation (Figures 4Acentrations (p,0.008) (Figure 5A).and 4B).

In the experiments with ADP-stimulated plate-d-arginine failed to modify the platelet responselets, irreversible aggregation was evidenced (MA:to collagen (8 mg/L) in whole blood being aggre-55.666.8%; n55). Aggregation was decreased bygation values (percentage of controls): 101.061.7,10-minute preincubation with l-arginine: platelet102.062.4, and 102.462.1 with 30, 120, and 500response (percentage of control) in presence ofmmol/L, respectively (n55; ANOVA for repeatedl-arginine was: 83.467.8 with 100 mmol/L andmeasures: not significant).60.367.8 with 1000 mmol/L (ANOVA for repeatedmeasures: p,0.0001; Dunnett’s test: p,0.05 with2.3. Study Cboth l-arginine concentrations; n55). IntraplateletcGMP levels at the end of ADP-induced aggrega-l-arginine increased cGMP levels in unstimulatedtion were increased in comparison with unstimu-PRP samples in the concentration range of 30–500lated platelets (control); the exposure to l-argininemmol/L (Figure 5A). The effect was statistically

significant for concentrations of 120–500 mmol/L. further enhanced cGMP (Figure 5B).

312 G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

Fig. 5. (A) Effect of 20-minute incubation with l-arginineFig. 4. Effect of 3-minute preincubation with NG-mono- on intraplatelet levels of cyclic 39,59-guanosine monophos-methyl-l-arginine (l-NMMA) at 100 mmol/L on antiag- phate (cGMP) Data are expressed as mean6SEM, n55.gregatory effect of 500 mmol/L l-arginine (20-minute incu- Significance: ANOVA for repeated measures p50.0001;bation) in whole blood stimulated with 20 mmol/L ADP Dunnett’s test: p,0.05 control vs. l-arginine at 120, 150,(A) or with 8 mg/L collagen (B). Data are expressed as and 500 mmol/L. 100 mmol/L GTN was used as positivemean6SEM, n58 for each experiment. control. (B) Effect of 10-minute preincubation with l-argi-

nine on intraplatelet levels of 39,59-cyclic guanosine mono-phosphate (cGMP) at the end of aggregation induced byADP (4 mmol/L). C indicates cGMP levels in unstimulated

2.4. Study D PRP. Data are expressed as mean6SEM, n55. Signifi-cance: ANOVA for repeated measures p50.004; Dunnett’stest: p,0.05 control vs. l-arginine at 100 and 1000 mmol/L.At the employed concentrations, both calcium ion-

ophore A23187 and ionomycin failed to modify plate-let response to ADP: after 20-minute incubation,

inhibitor l-NMMA (100 mmol/L) blunted the ef-MA (% of control) was 96.462.2 and 96.162.9,fect of l-arginine (Figure 8).respectively. These calcium ionophores signifi-

N-acetyl-l-cysteine, at concentrations of 3 andcantly increased the effect of l-arginine at 100 and6 mmol/L, markedly enhanced the effects of l-argi-500 mmol/L (Figure 6).nine on intraplatelet cGMP levels without differ-ences between the two doses employed (Figure 7B).

2.5. Study E

The addition of phosphodiesterase III inhibitor mil- 3. Discussionrinone clearly increased the effect of l-arginine; inthis condition, even the increase induced by the low- In the present study, we defined that l-arginine di-est l-arginine concentrations was significant (Fig- rectly influences the response of human plateletsure 7A). by inhibiting the aggregation induced by ADP and

collagen both in PRP and in whole blood. In theThe preincubation of PRP samples with NOS

313G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

Fig. 6. Effect of preincubation (1 minute) with 10 nmol/L Fig. 7. (A) Effect of 3-minute preincubation with milri-calcium ionophore A23187 (A) or 1 mmol/L ionomycin (B) none at 10 mmol/L on l-arginine-induced increase of in-on antiaggregating effects of l-arginine (100 and 500 traplatelet cGMP levels. Data are expressed as mean6mmol/L 10- and 20-minute incubation). Aggregation was SEM, n55. Significance was assayed by Student’s t-test forinduced by 4 mmol/L ADP. Data are expressed as paired data. (B) Effect of 3-minute preincubation withmean6SEM, n56. Significance was evaluated vs. controls N-acetyl-l-cysteine (NAC) at 3 and 6 mmol/L on l-argi-without preincubation with ionophores (Student’s t-test for nine-induced increase of intraplatelet cGMP levels. Datapaired data). *p,0.05; **p,0.01; ***p,0.001. are expressed as mean6SEM, n55. Significance was as-

sayed by Student’s t-test for paired data.

experiments in PRP samples, we evaluated time-This action is clearly dependent on activation ofdependence effect.

platelet NOS, since both cGMP increase and antiag-l-arginine concentrations of 100 to 500 mmol/Lgregating activity are prevented by pretreatmentwere sufficient to induce a significant reduction ofwith NOS inhibitor l-NMMA. This conclusion wasplatelet aggregability.reached already by Marietta et al. through theThese data complete previous reports from ourchanges of intraplatelet l-citrulline [20].group and from other investigators [3,19,20].

Data obtained with preincubation of plateletsFrom the studies on whole blood aggregation, weevidenced that the antiaggregating effect is specific with l-arginine and ionophores further support the

previous conclusion. Actually, it is known that plate-for l-amino acid, since d-arginine failed to modifyplatelet aggregation. let NOS is a calmodulin-activated enzyme and iono-

mycin is a potent stimulator of NOS activity throughIn further investigations, we defined a mechanisminvolved in the effect of l-arginine, evidencing that an increased Ca21 availability [35].

In summary, these data indicated that: (1) exoge-amino acid increases the intraplatelet levels of cyclicGMP in a dose-dependent way both in resting plate- nous l-arginine provides a suitable substrate for

augmented NO production from platelets, such aslets and at the end of ADP-induced aggregation.

314 G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

enhances the inibitory effect of Prostacyclin (PGI2)analog Iloprost, likely through cGMP-mediatedmodulation of intraplatelet cAMP levels [19]. In thisstudy, we showed that the inhibition by milrinone ofphosphodiesterase III, which controls intracellularcontent of both nucleotides 39,59-cyclic adenosinemonophosphate (cAMP) and cGMP, enhanced theeffects of l-arginine in particular when low aminoacid concentrations were added to PRP. This con-firms that an effect on nucleotide levels is presentfor a large spectrum of l-arginine concentrations.

N-acetyl-l-cysteine enhances the effects of l-argi-nine on cGMP production. This observation pro-vides further evidences that l-arginine effect on hu-man platelets is dependent on an increase of NO;actually, it is known from previuos reports that

Fig. 8. Effect of 3-minute preincubation with NG-mono- S-nitrosothiols derived by reactions between NOmethyl-l-arginine (l-NMMA) at 100 mmol/L on l-argi- and reduced thiols are biologically active intermedi-nine-induced increase of intraplatelet cGMP levels. Data

ates in the NO-induced guanylate cyclase activationare expressed as mean6SEM, n55. Significance for l-argi-[38–40]. Through this mechanism N-acetyl-l-cys-nine alone: ANOVA for repeated measures p50.0001;teine potentiates the effect of nitroglycerin and NODunnett’s test: p,0.05 control 150 and 500 mmol/L.

l-NMMA1l-arginine: ANOVA for repeated measures donors on platelet response and vascular tone [26–not significant. 29]. From our data, we suggest that the availability

of reduced thiols for S-nitrosothiol formation canincrease the l-arginine effects on human platelets.

demonstrated by other investigations in vascular en-The study has been supported by a financial grant from the Italiandothelial cells [36]; (2) the platelet activity is modu-Ministero dell’Universita’ e della Ricerca Scientifica.lated by l-arginine through cGMP; (3) increased

NO formation exerts inhibitory effects on plateletresponse to aggregating agents.

ReferencesOn this basis, our evidences can support thehypothesis that the antiaggregating effect of l-argi-nine infusion in human volunteers is mediated by 1. Furlong B, Henderson AH, Lewis MJ, Smith

JA. Endothelium derived relaxing factor inhib-an influence on platelets. Recent observation fromDiodati et al. evidencing that l-arginine inhibited its in vitro platelet aggregation. Br J Pharmacol

1987;90:687–92.platelet aggregation equally in vitro and in vivo [37]deeply supports this hypothesis. 2. Radomski MW, Palmer RMJ, Moncada S.

Characterization of the l-arginine:nitric oxideWith the second part of the study we investigatedsome mechanisms involved in the l-arginine effect. pathway in human platelets. Br J Pharmacol

1990;101:325–8.We confirmed that effective l-arginine concentra-tions are elevated in comparison to intracellular lev- 3. Radomski MW, Palmer RM, Moncada S. An

l-arginine/nitric oxide pathway present in hu-els (,1 mmol/L) and showed that prolonged incuba-tion is required to evidence antiaggregating effect. man platelets regulates aggregation. Proc Natl

Acad Sci USA 1990;87:5193–7.Platelet permeability to l-arginine is involved, sincepermeabilization of platelets with saponin increased 4. Moncada S, Higgs EA. Endogenous nitric

oxide: Physiology, pathology and clinical rele-sentivity to l-arginine by reducing both effectiveconcentration and lag time. vance. Eur J Clin Invest 1991;21:361–74.

5. Muragandam A, Mutus B. Isolation of nitricFinally, other experiments provided further infor-mation about modulation of l-arginine effect. synthase from human platelets. Biochim Bio-

phys Acta 1994;1200:1–6.In a previous report, we evidenced that l-arginine

315G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

6. Sase K Michel T. Expression of constitutive thelium-dependent dilatation in healthy youngmen. J Am Coll Cardiol 1995;26:1054–61.endothelial nitric oxide synthase in human

blood platelets. Life Sci 1995;57:2049–55. 18. Bode-Boger SM, Boger RH, Creutzig A, Tsi-kas D, Gutzki F-M. l-arginine infusion de-7. Malinski T, Radomski MW, Taha Z, Moncada

S. Direct electrochemical measurement of ni- creases peripheral arterial resistance and inhib-its platelet aggregation in healthy subjects. Clintric oxide released from human platelets. Bio-

chem Biophys Res Commun 1993;194:960–5. Sci 1994;87:303–10.19. Anfossi G, Massucco P, Mularoni E, Cavalot8. Zhou Q, Hellermann GR, Solomonson LP. Ni-

tric oxide release from resting human platelets. F, Mattiello L, Trovati M. Organic nitrates andcompounds that increase intraplatelet cyclicThromb Res 1995;77:87–96.

9. Radomski MW, Palmer RMJ, Moncada S. En- guanosine monophosphate (cGMP) levels en-hance the antiaggregating effects of the stabledogenous nitric oxide inhibits human platelet

adhesion to vascular endothelium. Lancet 1987; prostacyclin analogue Iloprost. ProstaglandinsLeukot Essent Fatty Acids 1993;49:839–45.ii:1057–8.

10. Alheid U, Frolich JC, Forstermann U. Endo- 20. Marietta M, Facchinetti F, Neri I, Piccinini F,Volpe A, Torelli G. l-arginine infusion de-thelium-derived relaxing factor from cultured

human endothelial cells inhibits aggregation of creases platelet aggregation through an in-traplatelet nitric oxide release. Thromb Reshuman platelets. Thromb Res 1987;47:561–9.

11. Freedman JE, Loscalzo J, Barnard MR, Alpert 1997;88:229–35.21. Arnal JF, Munzel T, Venema RC, James NL,C, Keaney J, Michelson AD. Nitric oxide re-

leased from activated platelets inhibits platelet Bai CL, Mitch WE, Harrison DG. Interactionsbetween l-arginine and l-glutamine changerecruitment. J Clin Invest 1997;100:350–6.

12. Davies T, Davidson ML, McClenaghan MD, NO production: An effect independent of NOsynthase substrate availability. J Clin InvestSay A, Haslam RJ. Factors affecting platelet

cyclic GMP levels during aggregation by colla- 1995;95:2565–72.22. Fukuto JM, Chaudhuri G. Inhibition of consti-gen and by arachidonic acid. Thromb Res

1976;9:387–405. tutive and inducible nitric oxide synthase: Po-tential selective inhibition. Annu Rev Pharma-13. Tokai Y, Kaibuchi K, Matsubara T, Nishizuka

Y. Inhibitory action of guanosine 39,59-mono- col Toxicol 1995;35:165–94.23. Conti M, Jin S-LC, Monaco L, Repaske DR,phosphate on thrombin-induced phosphatidyl-

inositol turn-over and protein phosphorylation Swinnen JV. Hormonal regulation of cyclic nu-cleotide phosphodiesterases. Endocr Rev 1991;in human platelets. Biochem Biophys Res

Commun 1981;101:61–7. 12:218–34.24. Anggard EE. Nitric oxide: Mediator, murderer14. Chirkov YY, Belushkina NN, Tyshchuk IA,

Severina IS, Horowitz JD. Increase in reactiv- and medicine. Lancet 1994;343:1199–206.25. Mayer B, Schrammel A, Klatt P, Koesling D,ity of human platelet guanylate cyclase during

aggregation potentiates the disaggregating ca- Schmidt K. Peroxynitrite-induced accumula-tion of cyclic GMP in endothelial cells andpacity of sodium nitroprusside. Clin Exp Phar-

macol Physiol 1991;18:517–24. stimulation of purified soluble guanylate cic-lase. Dependence on glutathione and possible15. Palmer RMJ, Ashton AS, Moncada S. Vascu-

lar endothelial cells synthesize nitric oxide role of S-nitrosation. J Biol Chem 1995;270:17355–60.from l-arginine. Nature 1988;333:664–6.

16. Schmidt HHHW, Nau H, Wittfoht W, Gerlach 26. Loscalzo J. N-acetyl-cysteine potentiates inhi-bition of platelet aggregation by nitroglycerin.J, Prescher K-E, Klein MM, Niroomand F,

Bohme E. Arginine is a physiological precursor J Clin Invest 1985;76:703–8.27. Stamler J, Cunningham M, Loscalzo J. Re-of endothelium-derived nitric oxide. Eur J

Pharmacol 1988;154:213–6. duced thiols and the effect of intravenous nitro-glycerin on platelet aggregation. Am J Cardiol17. Adams MA, Forsyth CJ, Jessup W, Robinson J,

Celermaier DS. Oral l-arginine inhibits plate- 1988;62:377–80.28. Stamler JS, Loscalzo J. The antiplatelet effectslet aggregation but does not enhance endo-

316 G. Anfossi et al./Thrombosis Research 94 (1999) 307–316

of organic nitrates and related nitroso com- tamagna C, Heller R, Garbarino G, Pescar-mona GP, Bosia A. Nitric oxide synthesis ispounds in vitro and in vivo and their relevance

to cardiovascular disorders. J Am Coll Cardiol impaired in glutathione-depleted human um-bilical endothelial cells. Am J Physiol 1993;264:1991;18:1529–36.

29. Stamler JS, Mendelsohn ME, Amarante P, C728–32.36. Bogle RG, Coade SB, Moncada S, Pearson JD,Smicl D, Andion N, Davies PF, Cooke JP, Los-

calzo J. N-acetylcysteine potentiates platelet Mann GE. Bradykinin and ATP stimulatedl-arginine uptake and nitric oxide release ininhibition by endothelium-derived relaxing

factor. Circ Res 1989;65:789–95. vascular endothelial cells. Biochem BiophysRes Comm 1991;180:926–32.30. Born GVR. Aggregation of blood platelets by

adenosine diphosphate and its reversal. Nature 37. Diodati JG, Dakak N, Gilligan DM, QuyyumiAA. Effect of atherosclerosis on endothelium-1962;194:927–9.

31. Weiss HJ Rogers J. Thrombocytopenia due dependent inhibition of platelet activation inhumans. Circulation 1998;98:17–24.to abnormalities in platelet release reaction.

Blood 1972;39:2–8. 38. Myers PR, Minor RL, Guerra R, Bates JN,Harrison DG. Vasorelaxant properties of the32. Brass LF, Joseph SK. A role for inositol-tri-

phosphate in intracellular Ca21 mobilization endothelium-derived relaxing factor moreclosely resemble S-nitrosocystein than nitricand granule secretion in platelets. J Biol Chem

1985;260:15172–9. oxide. Nature 1990;345:161–3.39. Stamler JS, Jaraki O, Osborne J, Simon DI,33. Bergh CH, Hjalmarson A, Holm G, Angwald

E, Jacobsson B. Studies on calcium exchange in Keaney J, Vita J, Singel D, Valeri CR, LoscalzoJ. Nitric oxide circulates in mammalian plasmaplatelets in human diabetes. Eur J Clin Invest

1988;18:92–7. primarily as an S-nitroso adduct of serum albu-min. Proc Natl Acad Sci USA 1992;89:7674–7.34. Cardinal DC, Flower RJ. The electronic ag-

gregometer: A novel device for assessing plate- 40. Loscalzo J. Antiplatelet and antithrombotic ef-fects of organic nitrates. Am J Cardiol 1992;let behaviour. J Pharmacol Meth 1980;3:135–58.

35. Ghigo D, Alessio P, Foco A, Bussolino F, Cos- 70(Suppl):18B–22B.