Journal of Human Hypertension (2001) 15, 203–207  2001 Nature Publishing Group All rights reserved 0950-9240/01 $15.00 www.nature.com/jhh ORIGINAL ARTICLE ␣ Doxazosin, an 1-adrenoceptor antagonist, inhibits serotonin-induced shape change in human platelets

IA Jagroop and DP Mikhailidis Department of Molecular Pathology and Clinical Biochemistry, Royal Free and University College Medical School (University of London), Royal Free campus, London, UK

Patients with peripheral vascular disease or diabetes and aliquots were removed at specific time points for mellitus tend to have elevated circulating levels of nat- median platelet volume (MPV) measurement (using a urally occurring platelet agonists like serotonin (5 hyd- high-resolution channelyser). The MPV was used as an roxytryptamine; 5-HT). This bioamine can induce plate- indicator of PSC. PRP was also incubated with doxazo- let shape change (PSC) an early phase of platelet sin (final concentration: 0.33 ␮M, a concentration similar activation, which is essentially aspirin resistant. In to therapeutic plasma levels) prior to the addition of addition, 5-HT exerts other harmful effects (eg stimulat- each of the above-mentioned agonists. Doxazosin sig- at 30 sec ,0.008 ؍ and P 0.007 ؍ ing vascular smooth muscle proliferation and inducing nificantly inhibited (P vasoconstriction in atheromatous coronary vessels). and 60 sec, respectively) the 5-HT-induced increase in The aim of this study was to determine whether doxazo- MPV. Doxazosin did not significantly inhibit ADP- or sin inhibits 5-HT-induced PSC. Doxazosin is a long act- U46619-induced PSC. The inhibitory effect of doxazosin ␣ ing 1-adrenoceptor antagonist, used in the treatment of seems to be specific to platelet 5-HT2 receptors, since essential hypertension and/or benign prostatic hyper- there was no effect on ADP- or U46619-induced PSC. plasia (BPH). Platelet rich plasma (PRP) was prepared This inhibition of platelet activation may be an five males and three additional, clinically relevant, advantage. Future in vivo ;8 ؍ from healthy volunteers (n females with a median age of 32 years, range: 26–57). studies should consider assessing the effect of doxazo- Agonists (5-HT, 0.06–0.5; ADP, 0.1–0.2 ␮mol/l or U46619, sin on 5-HT-induced platelet activation. ␮ a TXA2 analogue, 0.025–0.05 mol/l) were added to PRP Journal of Human Hypertension (2001) 15, 203–207

Keywords: doxazosin; platelet shape change; human platelets

Introduction humans.5–7 This is a relevant topic for further research ␣ because patients with peripheral vascular disease Doxazosin is a long acting 1-adrenoceptor antagonist, (PVD) or diabetes mellitus have elevated circulating used in the treatment of essential hypertension and/or levels of 5-HT.8–11 These conditions are associated 1–3 benign prostatic hyperplasia (BPH). These effects with a high prevalence of hypertension and vascular are complemented, by favourable changes in serum 12 1–3 events. PVD is associated with platelet hyperactiv- lipids, haemostatic factors and insulin sensitivity. ity.13 Elevated plasma levels of 5-HT were also asso- Hypertensive patients are at a risk of thrombotic 4 ciated with an increased incidence of graft reoc- events. It follows that it would be an advantage for clusion after infrainguinal by-pass surgery.9 These antihypertensive drugs to possess anti-thrombotic associations may be attributed to the harmful proper- properties.4 In this context, there is evidence that 5–7 ties of 5-HT that include stimulating vascular muscle doxazosin inhibits platelet activation. However, proliferation and platelet aggregation as well as doxazosin did not inhibit serotonin (5-hydroxy- inducing vasoconstriction in atheromatous coronary tryptamine; 5-HT)-induced platelet aggregation in vessels.9–11 Platelets are the main (95%) store of circulating 5-HT.9–11 This bioamine is released from activated Correspondence: Dr Dimitri P Mikhailidis MD FRCPath, Reader platelets8 and can induce platelet shape change and Honorary Consultant, Department of Molecular Pathology (PSC), an early phase of platelet activation, which is and Clinical Biochemistry, Royal Free and University College essentially aspirin resistant.14–27 PSC can be induced Medical School (University of London), Royal Free campus, Pond Street, London NW3 2QG, UK. Fax: +44 (0) 171 830 2235 by considerably lower concentrations of 5-HT than 15 Received 26 May 2000; revised 25 July 2000; accepted 18 aggregation (0.06–0.5 vs 10.0 ␮mol/l, respectively). August 2000 Similarly, PSC can be inhibited by lower concen- Doxazosin and shape change in human platelets IA Jagroop and DP Mikhailidis 204 trations of antagonists when compared to aggre- Table 1 Effect of doxazosin (0.33 ␮M) on a combination of ϩ = gation.15 We therefore used this technique to reas- 5HT ADP-induced PSC (n 1). PSC is expressed in terms of median platelet volume (MPV) in fl sess the effect of doxazosin on 5-HT-induced platelet activation in human platelets. This investi- ␣ Incubation Incubation gation is based on the evidence that -adrenoceptor Time (30 sec) Time (60 sec) antagonists also block 5-HT receptors.28,29

H2O (control) 5.42 5.36 ADP (only)a 5.61 5.61 Patients and methods 5HT (only)a 5.61 5.58 ADP+5HT 5.84 5.81 Venous blood was collected in 3.8% citrate (1 part ADP+5HT 5.62 5.58 citrate to 9 parts blood) from healthy volunteers + doxazosin (n = 8; five males, three females with a median age of 32 years, range: 26–57). Volunteers denied taking aThe dose of ADP alone (0.04 ␮M) and 5-HT alone (0.06 ␮M) was any drugs for at least 2 weeks prior to sampling. The selected to produce a middle range increase in MPV. This volunteers were fasting overnight and were all non- response could then be inhibited or enhanced. smokers. The citrated blood was centrifuged at 160 g for 15 min at room temperature. The platelet rich plasma (PRP) was then carefully collected with a Pasteur pipette. Aliquots of PRP (450 ␮l), in sili- inhibition was specific for any of the agonists. conised glass cuvettes, were placed in a Chronology Because a harmful synergistic effect of various agon- dual channel optical aggregometer (Coulter Elec- ists may occur in vivo, we examined the effect of tronics, Luton, Beds, UK) and kept at 37°C for doxazosin on a combination of two such agonists at 15 min prior to the start of the experiment.15–20,22–26 sub-maximal concentrations (ADP, 0.04 ␮mol/l + 5- Agonists (5-HT, 0.06–0.5; ADP, 0.1–0.2 ␮mol/l or HT, 0.06 ␮mol/l—see Table 1). ␮ U46619, a TXA2 analogue, 0.025–0.05 mol/l) were Doxazosin mesylate (mol. wt. 547.6) was prepared added (as 10–50 ␮l volumes) to PRP and stirred at by dissolving the compound in distilled water. 1000 rpm. These agonists concentrations and sam- Saline solutions, or other media-containing chloride pling times were determined in previous studies; ions, were avoided as this result in the precipitation they generate the most reproducible results.15–20,22–26 of the less soluble chloride salt of doxazosin.5 We adjusted the concentration of agonists to obtain a middle range response that could be either inhibited or enhanced.15–20,22–26 The volumes of distilled Statistical analysis and presentation of the results water (control) added to samples were matched to The results are presented as median and interquar- those of the agonist (as 10–50 ␮l volumes). ␮ tile range. Comparisons are by non-parametric PRP (100 l aliquots) was removed and mixed analysis (paired Wilcoxon tests, two-tailed). with 400 ␮l of fixative, (4% aqueous glutaraldehyde) at 30 sec, 60 sec (see ‘Results’ for details) after adding the agonists. This sample was further diluted Results 400-fold in Isoton II (Coulter Electronics). These samples were then processed by a Coulter counter Effect of doxazosin on 5-HT-induced PSC ␮ ZM (electrical impedance method using a 70- m The 5-HT-induced increase in MPV was signifi- diameter sampling tube orifice) coupled to a C-256 cantly inhibited (P = 0.007 and P = 0.008, at 30 sec channelyser (Coulter Electronics). The analyser sam- and 60 sec, respectively) by doxazosin (Table 2). 15– ple ‘window’ was set between 2.67 and 19.12 fl. Doxazosin alone did not affect the MPV (incubation 20,22–26 The counter was calibrated using latex-par- time: up to 4 min). ticles of 9 fl (Coulter Electronics). For particle analy- sis, data were accumulated to a maximum of 500 platelets in each of the 256 channels. The median Effect of doxazosin on ADP-and U46619-induced platelet volume (MPV) was the volume of the chan- PSC nel on each side of which 50% of the platelet popu- lation is distributed.15–20,22–26 The increase in MPV The increase in MPV induced by ADP and U46619 was taken as representing the PSC.15–20,22–26 was not inhibited by doxazosin (Table 3). The platelet count was monitored throughout the experiment to exclude PRP samples where signifi- + Ͼ Effect of doxazosin on a combination of ADP 5- cant ( 5%) aggregation occurred. HT-induced PSC PRP was also incubated for 4 min with doxazosin (final concentration 0.33 ␮mol/l, a concentration Doxazosin inhibited PSC induced by a combination similar to therapeutic plasma levels)1 prior to the of ADP + 5-HT (Table 1). In a representative experi- addition of each of the above-mentioned agonists. ment, the combined effect of ADP + 5-HT became This was carried out to determine whether doxazo- equivalent to the MPV response seen with ADP or sin could inhibit PSC. We also established if this 5-HT alone (at 30 and 60 sec).

Journal of Human Hypertension Doxazosin and shape change in human platelets IA Jagroop and DP Mikhailidis 205 Table 2 Effect of doxazosin (0.33 ␮M) on 5HT-induced PSC (n = 8). PSC is expressed in terms of median platelet volume (MPV) in fl. The results are presented as median and interquartile range

a a + + H2O 30 sec H2O 60 sec 5HT 30 sec 5HT 60 sec Doxazosin 5HT Doxazosin 5HT (control) (control) 30 sec 60 sec

5.36 5.31 5.61 5.56 5.32 5.35 (5.26–5.56) (5.07–5.48) (5.32–5.96) (5.37–5.85) (5.04–5.58) (5.00–5.64)

Statistical analysis: 5HT 30 sec vs doxazosin+5HT 30 sec: P = 0.007; 5HT 60 sec vs doxazosin+5HT 60 sec: P = 0.008. aThe dose of 5HT (0.06–0.5 ␮M) was selected to produce a middle range increase in MPV. This response could then be inhibited or enhanced.

Table 3 Effect of doxazosin (0.33 ␮M) on PSC induced by ADP bladder detrusor.29 Doxazosin also influences plate- = = (n 7) and U46619 (n 6). PSC is expressed in terms of median let calcium mobilisation.5,33 This mechanism may platelet volume (MPV) in fl. The results are presented as median and interquartile range account for the inhibition, by high concentrations (30 ␮M) of doxazosin, of adrenaline-induced aggre- 5 Sampling times gation potentiated by collagen, ADP or serotonin and possibly for the PSC effect we observed. How- 30 sec 60 sec ever, the considerably lower concentration of dox- azosin we used (0.33 ␮M) and the agonist specificity H2O (control) 5.53 5.49 of this effect, the binding of doxazosin to 5-HT3 (5.36–5.70) (5.39–5.77) receptors29 and the 5-HT antagonist activity of other ADPa 5.67 5.64 ␣-blockers28,29 favours a receptor-mediated mode of (5.64–6.24) (5.61–6.19) action. Furthermore, we previously showed that ket- + ␣ ADP doxazosin 5.74 5.67 anserin, a 5-HT2 antagonist which also has some - (5.71–6.28) (5.61–6.40) blocker activity, inhibits 5-HT-induced PSC.15 U46619a 5.96 6.15 Others demonstrated that doxazosin inhibited (5.70–6.27) (5.93–6.37) ADP-induced platelet aggregation.6,7 This finding is U46619 + doxazosin 5.96 6.01 not in agreement with our results or those of others.5 (5.64–6.28) (5.69–6.37) This discrepancy is probably the result of inappro- priate methodology5 used in the earlier studies.6,7 Statistical analysis: Doxazosin did not significantly inhibit ADP- The sensitivity of the PSC technique was demon- or U46619-induced PSC. 5 aThe dose (ADP, 0.1–0.2 ␮M or U46619, 0.025–0.05 ␮M) was strated by the fact that others, did not detect an selected to produce a middle range increase in MPV. This inhibitory effect of doxazosin on 5-HT-induced response could then be inhibited or enhanced. aggregation despite adding considerably higher con- centrations of this drug than we did (30 vs 0.33 ␮mol/l). Discussion Our results should be interpreted after consider- ing that, in humans, the circulating levels of 5-HT Doxazosin inhibited 5-HT-induced platelet acti- 10 vation (represented by PSC) in samples collected are in the 4–15 nmol/l range. Nevertheless, near from healthy volunteers. These findings are of clini- therapeutic levels of doxazosin blocked PSC cal interest because PSC, an early phase of platelet induced by considerably higher 5-HT concen- ␮ activation, is essentially aspirin resistant.15–20,22– trations (up to 0.50 mol/l). 26,30–32 Furthermore, the other harmful effects of 5- Our findings are limited because they were carried HT in the vasculature (see ‘Introduction’) and else- out in vitro and in samples obtained from healthy where (eg, bladder or prostate)29 may be inhibited by subjects. Furthermore, we rely on a single index of doxasosin. These possibilities deserve investigation. platelet function. Because in vivo several agonists may act syner- Platelets contribute to thrombosis and atheroscler- gistically,5 we evaluated PSC after adding 5- osis and inhibiting platelet activity is of clinical HT + ADP. The increase in MPV caused by this com- importance.34 Several antihypertensive drugs alter bination became equivalent to that seen with ADP haemostatic factors.19,35–39 This additional beneficial or 5-HT alone. We conclude that the contribution action may influence prescribing.4 The antiplatelet of 5-HT to the synergistic effect was inhibited since action of antihypertensives may vary in specificity. doxazosin did not influence ADP-induced PSC. For example, doxazosin inhibits 5-HT-mediated pla- 19 Doxazosin may only inhibit the 5-HT2 receptor- telet activation whereas losartan inhibits the acti- dependent actions of 5-HT because these receptors vation induced by angiotensin II and thromboxane 15 mediate platelet activation and doxazosin did not A2. These differences may gain importance because alter ADP- or U46619-induced PSC. However, dox- of the need to use of multiple drugs to treat hyper- azosin significantly reduced [3H]-5-HT binding and tension.12 The variability in the inhibitory speci- 5-HT3-mediated contractions in the rabbit urinary ficity of different antihypertensives may be relevant

Journal of Human Hypertension Doxazosin and shape change in human platelets IA Jagroop and DP Mikhailidis 206 if several agonists mediate platelet activation in ecular weight heparin, nadroparin (Fraxiparine), vivo. These drugs may then exert a synergistic inhibits thrombin induced platelet shape change and inhibitory action on platelets. Future studies should does not enhance spontaneous platelet aggregation. Br assess this possibility as well as the effect of doxazo- J Clin Pharmacol 1996; 41: 163–165. sin on 5-HT-induced responses. 18 Barradas MA et al. Effect of milrinone on human plate- let shape change, aggregation and thromboxane A2 synthesis: an in vitro study. Thromb Res 1993; 71: References 227–236. 19 Jagroop IA, Mikhailidis DP. Angiotensin II can induce 1 Bainbridge AD, Elliott HL. The pharmacology of dox- and potentiate shape change in human platelets: effect azosin. Rev Contempt Pharmacother 1992; 3: 1–7. of losartan. J Hum Hypertens 2000; 14: 581–585. 2 Fulton B, Wagstaff AJ, Sorkin EM. Doxazosin: an 20 Jagroop IA, Clathworthy I, Lewin J, Mikhailidis DP. update of its clinical pharmacology and therapeutic Shape change in human platelet-measurements with a applications in hypertension and benign prostatic channelyzer and visualisation by electron microscopy. hyperplasia. Drugs 1995; 49: 295–320. Platelets 2000; 11: 28–32. ␣ 3 Sever PS. 1-blockers in hypertension. Curr Med Res 21 O’Brien JR, Heywood JB. Effects of aggregating agents Opin 1999; 15: 95–103. and their inhibitors on the mean platelet shape. J Clin 4 Papadakis JA et al. Effect of hypertension and its treat- Pathol 1966; 19: 148–153. ment on lipid, lipoprotein (a), fibrinogen, and bilirubin 22 Nystrom ML, Barradas MA, Mikhailidis DP. N-formyl- levels in patients referred for dyslipidemia. Am J Hy- methionine-leucine-phenylalanine [fMLP], a bacterial pertens 1999; 12: 673–681. chemotactic peptide, stimulates platelet shape change 5 Alarayyed NA, Graham BR, Prichard BNC, Smith CCT. in human whole blood. Platelets 1993; 4: 156–158. The potentiation of adrenaline-induced in vitro plate- 23 Whitworth NH, Barradas MA, Mikhailidis DP, Dan- let aggregation by ADP, colagen and serotonin and its dona P. An investigation into the effects of bacterial inhibition by naftopidil and doxazosin in normal lipopolysaccharide on human platelets. Eur J Haema- human subjects. Br J Clin Pharm 1995; 39: 369–374. tol 1989; 43: 112–119. 6 Hernandez HR et al. The effect of doxazosin on platelet 24 Nystrom ML, Barradas MA, Jeremy JY, Mikhailidis DP. aggregation in normotensive subjects and patients with Platelet shape change in whole blood: differential hypertension: an in vitro study. Am Heart J 1991; 121: effects of endotoxin. Thromb Haemost 1994; 71: 389–394. 646–650. 7 Hernandez HR et al. Evidence of an antiplatelet aggre- 25 Jagroop IA, Mikhailidis DP. An investigation of the gation action of doxazosin in patients with hyperten- serotonergic effects of fenfluramine, dexfenfluramine sion: an ex vivo study. Am Heart J 1991; 121: 395–401. and dexnorfenfluramine using platelets as neuronal 8 Barradas MA, Jagroop IA, Mikhailidis DP. Naftidrofu- models. Platelets 2000; 11: 161–165. ryl inhibits the release of 5-hydroxytryptamine and 26 Mikhailidis DP, Barradas MA, O’Donoghue S, Dan- platelet-derived growth factor from human platelets. dona P. Evidence for in vivo platelet activation follow- Clin Chim Acta 1994; 230: 157–167. ing the injection of conventional unfractionated hep- 9 Cheshire NJW et al. Smoking and plasma fibrinogen, arin. Platelets 1990; 1: 189–192. lipoprotein (a) and serotonin are markers for post-oper- 27 Frojmovic MM, Milton JG. Human platelet size, shape ative infrainguinal graft stenosis. Eur J Vas Endovasc and related functions in health and disease. Physiol Surgery 1996; 11: 479–486. Rev 1982; 62: 186–261. 10 Fawcett A et al. Platelet activation and serotonin 28 Frenken M, Kaumann AJ. Interconversion into a low release during colorectal surgery. Platelets 1998; 9: active state protects vascular 5-HT2-receptors against 109–113. irreversible antagonism by phenoxybenzamine. Naunyn- 11 Barradas MA et al. Intraplatelet serotonin in patients Schmiedebergs Arch Pharmacol 1987; 335: 481–490. with diabetes mellitus and peripheral vascular dis- 29 Khan MA et al. Doxazosin modifies serotonin- ease. Eur J Clin Invest 1988; 18: 399–404. mediated rabbit urinary bladder contraction. Potential 12 Hansson L et al, for the HOT Study Group. Effects of clinical relevance. Urol Res 2000; 28: 116–121. intensive blood pressure lowering and low-dose 30 Mikhailidis DP, Jagroop IA. Is clopidogrel markedly aspirin in patients with hypertension: principal results superior to aspirin in patients with peripheral vascular of the patients with hypertension optimal treatment disease? Platelets 1998; 9: 273–278. (HOT) randomised trial. Lancet 1998; 351: 1755–1762. 31 Earnest JP, Santos GF, Zuerbig S, Fox JE. Dystrophin- 13 Mikhailidis DP et al. Heparin-induced platelet aggre- related protein in the platelet membrane skeleton. gation in anorexia nervosa and in severe peripheral Integrin-induced change in detergent-insolubility and vascular disease. Eur J Clin Invest 1985; 15: 313–319. cleavage by calpain in aggregating platelets. J Bio 14 Jagroop IA, Mikhailidis DP. Effect of endothelin-1 on Chem 1995; 270: 27259–27265. human platelet shape change: reversal of activation by 32 Bearer EL. Cytoskeletal domains in the activated plate- naftidrofuryl. Platelets 2000; 11: 227–282. let. Cell Motility & the Cytos 1995; 30: 50–66. 15 Barradas MA, O’Donoghue S, Mikhailidis DP. 33 Alarayyed NA et al. In vitro adrenaline and collagen- Measurement of platelet volume using a channelyzer: induced mobilisation of platelet calcium and its inhi- Assessment of the effect of agonists and antagonists. bition by naftopidil, doxazosin and nifedipine. B J Clin In vivo 1992; 6: 629–634. Pharm 1997; 43: 415–420. ␣ 16 Jagroop IA, Mikhailidis DP. Doxazosin a selective 1- 34 Anti Platelet Trialists’ Collaboration. Collaborative adrenoceptor antagonist, also inhibits serotonin- overview randomised trials of antiplatelet treatment. I: induced shape change in human platelets. Br J Pharm Prevention of death, myocardial infarction and stroke 1999; 127: 91. by prolonged antiplatelet therapy in various categories 17 Jagroop IA, Barradas MA, Mikhailidis DP. A low mol- of patients. BMJ 1994; 308: 81–106.

Journal of Human Hypertension Doxazosin and shape change in human platelets IA Jagroop and DP Mikhailidis 207 35 Islim IF, Beevers DG, Bareford D. The effect of antihy- nimodipine on platelet function and vascular prosta- pertensive drugs on in vivo platelet activity in essen- cyclin synthesis. Drugs Under Exptl Clin Res 1985; 11: tial hypertension. J Hypertens 1992; 10: 379–383. 645–651. 36 Steiner JA, Mikhailidis DP, Hughes RJD, Dandona P. 38 James IM et al. Treatment of hypertension with capto- The effect of nadolol on platelet aggregation. In: Gross pril: preservation of regional blood flow and reduced F (ed). International Experience with Nadolol, a Long- platelet aggregation. J Hum Hypertens 1988; 2: 21–25. acting Beta-blocking agent. Royal Soc Med Int Cong 39 Lip GY, Blann AD, Beevers DG. Prothrombotic factors, Sympos Series no. 37. Academic Press & Royal Soc endothelial function and left ventricular hypertrophy Med: London, 1981, pp 65–67. in isolated systolic hypertension compared with sys- 37 Jeremy JY, Barradas MA, Mikhailidis DP, Dandona P. tolic-diastolic hypertension. J Hypertens 1999; 17: An investigation into the effects of nifedipine and 1203–1207.

Journal of Human Hypertension