Author + information
- Received January 7, 2008
- Revision received March 25, 2008
- Accepted April 2, 2008
- Published online July 1, 2008.
- Ninian N. Lang, MRCP⁎,⁎ (, )
- Ingibjörg J. Guðmundsdóttir, MRCP⁎,
- Nicholas A. Boon, MD, FRCP†,
- Christopher A. Ludlam, PhD, FRCP⁎,
- Keith A. Fox, FRCP, FESC⁎ and
- David E. Newby, PhD, FRCP⁎
- ↵⁎Reprint requests and correspondence:
Dr. Ninian N. Lang, Centre for Cardiovascular Science, The University of Edinburgh, Chancellor's Building, Edinburgh, EH16 4SU, United Kingdom.
Objectives We sought to test the hypothesis that cigarette smoking adversely alters protease-activated receptor type 1 (PAR-1)-mediated vascular effects in vivo in humans.
Background Distinct from its role in the coagulation cascade, thrombin exerts its major cellular and cardiovascular actions via PAR-1. The activation of PAR-1 causes endothelium-dependent arterial vasodilation and the release of endogenous fibrinolytic factors.
Methods Forearm blood flow was measured with venous occlusion plethysmography in 12 cigarette smokers and 12 age- and gender-matched nonsmokers during intrabrachial infusions of PAR-1–activating-peptide (SFLLRN; 5 to 50 nmol/min), bradykinin (100 to 1,000 pmol/min), and sodium nitroprusside (2 to 8 μg/min). Plasma tissue plasminogen activator (t-PA) and plasminogen-activator inhibitor 1 antigen and activity concentrations were measured throughout the experiment.
Results All agonists caused dose-dependent increases in forearm blood flow (p < 0.0001 for all). Although bradykinin and sodium nitroprusside caused similar vasodilation, SFLLRN-induced vasodilation was attenuated in smokers (p = 0.04). Smokers had modest reductions in bradykinin-induced active t-PA release (reduced by 37%, p = 0.03) and had a marked impairment of SFLLRN-induced t-PA antigen (p = 0.02) and activity (p = 0.006) release, with a 96% reduction in overall net t-PA antigen release. The use of SFLLRN also caused similar (p = NS) increases in inactive plasminogen-activator inhibitor 1 in both smokers and nonsmokers (p ≤ 0.002 for both).
Conclusions Cigarette smoking causes marked impairment of PAR-1–mediated endothelial vasomotor and fibrinolytic function. Relative arterial stasis and near abolition of t-PA release will strongly promote clot propagation and vessel occlusion. These findings suggest a major contribution of impaired endothelial PAR-1 action to the increased atherothrombotic risk of cigarette smokers.
Smoking tobacco remains one of the most important and consistent modifiable risk factors for myocardial infarction and fatal coronary artery disease (1). The recent INTERHEART (A Study of Risk Factors for First Myocardial Infarction in 52 Countries and Over 27,000 Subjects) study revealed that smoking tobacco increases the risk of nonfatal myocardial infarction by as much as 7-fold (2). The pathophysiological mechanisms underlying this association are likely to be a combination of accelerated atherosclerosis (3) and a propensity to acute coronary thrombosis (1,4).
The endogenous fibrinolytic system is responsible for the dissolution of arterial thrombi that are frequently found on the surface of atherosclerotic plaques at areas of endothelial denudation (5,6). It is regulated by the profibrinolytic factor, tissue plasminogen activator (t-PA), and its endogenous inhibitor, plasminogen-activator inhibitor type 1 (PAI-1) (7–9). The rapid mobilization of t-PA from the endothelium is crucial, with thrombus dissolution being much more effective if t-PA is incorporated during, rather than after, thrombus formation (10). Indeed, acute stimulated t-PA release predicts the future risk of cardiovascular events (11).
Thrombin plays a central role in the coagulation cascade and thrombosis. It is one of the most powerful physiological agonists in the cardiovascular system, and its actions are fundamental to the processes of atherothrombosis. Distinct from its enzymatic role in the coagulation cascade, thrombin causes direct cellular activation through stimulation of a novel family of G-protein-coupled receptors, protease-activated receptors (PARs) (12). These receptors have a unique mechanism of activation whereby agonist-induced proteolytic cleavage of the extracellular domain reveals a short peptide sequence that remains tethered and causes autoactivation of the receptor. To date, 4 different types of PARs have been identified: PAR-1, -3, and -4 are all activated by thrombin whereas PAR-2 is mainly activated by trypsin (13).
PAR-1 is the principal receptor that mediates the cardiovascular actions of thrombin. The hexapeptide, SFLLRN, represents the short peptide sequence revealed during PAR-1 activation and can be used as a selective agonist of the human PAR-1 thrombin receptor without activation of the coagulation cascade. Using SFLLRN, we have recently described the in vivo effects of PAR-1 activation in platelets, endothelium, and vascular smooth muscle in humans. For the first time, we were able to show that thrombin has unique and contrasting effects in the human vasculature, including arterial dilation, venous constriction, platelet activation, and tissue-type plasminogen activator (t-PA) release (14).
We, and others, have previously reported that pharmacological stimulation of acute t-PA release in the peripheral (15,16) and coronary (17,18) arterial circulations is markedly attenuated in smokers. In this study, we hypothesized that smokers have impaired PAR-1–mediated vascular responses. We, therefore, examined PAR-1–mediated t-PA release and vasomotor responses in the forearm circulation of cigarette smokers and healthy nonsmoking control subjects.
Twelve healthy cigarette smokers (5 to 20 cigarettes/day) and 12 age- and gender-matched nonsmokers between ages 20 and 46 years participated in the study, which was undertaken with the approval of the local research ethics committee and in accordance with the Declaration of Helsinki. The written informed consent of each subject was obtained before entry into the study. Exclusion criteria included a history of asthma, hypertension, diabetes mellitus, coagulopathy, hyperlipidemia, or vascular disease. Control subjects were lifelong nonsmokers and were not exposed to regular environmental tobacco smoke. Smokers had a history of regular daily cigarette smoking of at least 5 years and maintained their normal smoking habits in the week before attendance.
None of the subjects received vasoactive or nonsteroidal anti-inflammatory drugs in the week before the study, and all abstained from alcohol for 24 h before and from food, tobacco, and caffeine-containing drinks on the day of the study. All studies were performed in a quiet, temperature-controlled room maintained at 22°C to 24°C.
Intra-arterial drug administration
All subjects underwent brachial artery cannulation with a 27 standard-wire-gauge steel needle. The intra-arterial infusion rate was kept constant at 1 ml/min throughout all studies. Forearm blood flow was measured in the infused and noninfused arms by venous occlusion plethysmography with mercury-in-silastic strain gauges as described previously (15,19). Supine heart rate and blood pressure were monitored at intervals throughout each study with the use of a semiautomated noninvasive oscillometric sphygmomanometer.
After a 20-min intra-arterial infusion of 0.9% saline, the glycoprotein IIb/IIIa antagonist, tirofiban (1.25 μg/min; Merck, Sharp and Dohme, Hoddesdon, United Kingdom), was infused and continued throughout the study to inhibit potential PAR-1-induced platelet aggregation. This dose of tirofiban does not affect forearm blood flow (14).
During tirofiban administration, subjects received intra-arterial infusions of the PAR-1–activating peptide, SFLLRN (5, 15, and 50 nmol/min; Clinalfa, Läufelfingen, Switzerland), bradykinin (an endothelium-dependent vasodilator that causes the release of t-PA; 100, 300, and 1,000 pmol/min; Clinalfa), and sodium nitroprusside (an endothelium-independent vasodilator that does not release t-PA; 2, 4, and 8 μg/min; David Bull Laboratories, Warwick, United Kingdom). Study drugs were infused in random order for 10 min at each dose and were separated by a 20-min infusion of 0.9% saline.
Venous cannulae (17-gauge) were inserted into large subcutaneous veins of the antecubital fossae of both arms. Blood samples were drawn simultaneously from each arm at the beginning of the study and during infusion of each dose of PAR-1–activating peptide (SFLLRN), bradykinin, and sodium nitroprusside. Venous blood was collected into acidified buffered citrate (Stabilyte, Trinity Biotech Plc, Co., Wicklow, Ireland; for t-PA assays) and into citrate (BD Vacutainer, BD UK Ltd., Oxford, United Kingdom; for PAI-1 and von Willebrand factor [vWF] assays). Samples were kept on ice before centrifugation at 2,000 g for 30 min at 4°C. Platelet-free plasma was decanted and stored at 80°C before assay. Plasma t-PA antigen and activity (t-PA Combi Actibind Elisa Kit, Technoclone, Vienna, Austria), PAI-1 antigen and activity (Elitest PAI-1 antigen and Zymutest PAI-1 Activity, Hyphen Biomed, Neuville-Sur-Oise, France), and vWF antigen (Dako A/S, Glostrup, Denmark) concentrations were determined with the use of enzyme-linked immunosorbent assays. Full blood count and hematocrit were measured at baseline and the end of the study.
Data analysis and statistics
Forearm plethysmographic data were analyzed as described previously (19). Estimated net release of plasma t-PA, PAI-1, and vWF has been defined previously as the product of the infused forearm plasma flow (based on the mean hematocrit and the infused forearm blood flow) and the concentration difference between the infused and noninfused arms (19). Variables are reported as mean ± SEM and analyzed with repeated measures analysis of variance (ANOVA) and a 2-tailed Student t test as appropriate. Statistical analysis was performed with GraphPad Prism (GraphPad Software, San Diego, California) and statistical significance taken at the 5% level. The authors had full access to the data and take responsibility for its integrity. All authors have read and agreed to the report as written.
There were no differences in baseline characteristics between cigarette smokers and nonsmokers (Table 1). There were no changes in blood pressure, heart rate, or hematocrit (data not shown) during the study. Smokers had a mean cigarette consumption of 15 ± 1 cigarettes per day over a mean period of 9 ± 2 years (7 ± 2 pack-years).
Forearm blood flow
The use of tirofiban did not affect forearm blood flow (data not shown). Intra-arterial sodium nitroprusside, bradykinin, and the PAR-1–activating peptide, SFLLRN, all caused dose-dependent vasodilation in the infused arm of smokers and nonsmokers (p < 0.0001 for all; ANOVA). There were no changes in blood flow in the noninfused arm (data not shown).
Although there was no difference with bradykinin (p = 0.64; ANOVA smokers vs. nonsmokers), vasodilatation to SFLLRN was attenuated in smokers (p = 0.044; ANOVA smokers vs. nonsmokers). Endothelium-independent vasodilation evoked by the use of sodium nitroprusside was similar in both groups (p = 0.74; ANOVA smokers vs. nonsmokers) (Fig. 1).
Plasma fibrinolytic and hemostatic factors
Baseline plasma t-PA antigen and activity (Table 2) and vWF antigen (Table 3) concentrations were similar in smokers and nonsmokers. There appeared to be a trend toward greater absolute plasma PAI-1 antigen and activity concentrations in smokers, but this difference did not reach statistical significance (smokers vs. nonsmokers: PAI-1 antigen, p = 0.07 and p = 0.10, and PAI-1 activity, p = 0.18 and p = 0.24; infused and noninfused arms, respectively) (Table 4).
The use of SFLLRN caused a dose-dependent net release of t-PA antigen in nonsmokers (p < 0.0005; ANOVA) but not smokers (p = 0.18; ANOVA) (Fig. 2). In comparison with nonsmokers, the release of t-PA antigen and activity by SFLLRN was markedly attenuated in smokers (p = 0.02 and p = 0.006, respectively; ANOVA). However, SFLLRN induced a dose-dependent net release of PAI-1 antigen release in both nonsmokers (p = 0.0002; ANOVA) and smokers (p = 0.001; ANOVA). The response was similar in both groups (p = 0.36; ANOVA) and was associated with no change in net PAI-1 activity (p = NS for all; ANOVA [data not shown]) or vWF antigen release (p = NS; ANOVA) (Table 3).
Bradykinin caused a dose-dependent net release of t-PA antigen and activity in both smokers and nonsmokers (p < 0.01 for all; ANOVA). Bradykinin also evoked a dose-dependent increase in absolute t-PA activity in the noninfused arm of both nonsmokers (p < 0.0001; ANOVA) and smokers (p = 0.008; ANOVA). Net release of t-PA activity induced by bradykinin was less in smokers than nonsmokers (p = 0.032, smokers versus nonsmokers; ANOVA) (Fig. 2). Bradykinin caused no change in net PAI-1 antigen or activity and did not affect vWF antigen in either group (p = 0.91, nonsmokers; p = 0.98 nonsmokers; ANOVA). As expected (20). sodium nitroprusside caused no change in absolute or net release of t-PA, PAI-1 or vWf (data not shown).
To our knowledge, we have shown for the first time that thrombin-mediated vascular responses are markedly impaired in cigarette smokers, with a substantial reduction observed in PAR-1–mediated endothelial t-PA release and forearm arterial vasodilation. This impaired vasomotor and fibrinolytic response may represent an important shift in the fine balance between intravascular thrombosis and fibrinolysis that could account for the increased incidence of atherothrombosis in cigarette smokers.
Smoking and PAR-1–induced arterial vasomotion
As reported by others (16), we observed no effect of smoking status on endothelium-dependent vasodilation to bradykinin or endothelium-independent vasodilation to sodium nitroprusside. One of the important novel observations from our study is that vasodilation evoked via PAR-1 is impaired in smokers, especially at the greater doses of SFLLRN. Because homeostatic mechanisms attempt to maintain vessel patency and minimize intravascular thrombus formation in healthy arteries, we have previously hypothesized the arterial vasodilation to PAR-1 activation represents a protective feedback mechanism. In the presence of a developing thrombus, PAR-1–mediated vasodilation will increase blood flow to limit arterial thrombosis by facilitating its rapid clearance and dissolution (14). Thus, this specific impairment of PAR-1–induced vasodilation may have major pathophysiological consequences during acute thrombotic events such as myocardial infarction.
Smoking and PAR-1–induced release of endothelium-derived factors
Over and above diminished vasomotion, the major finding of our study was the almost complete abolition of PAR-1–mediated t-PA antigen release in cigarette smokers. Furthermore, PAR-1 activation caused only a very modest increase in t-PA activity despite causing substantial t-PA antigen and activity release in nonsmokers.
The current findings confirm previous studies from our own and other groups reporting reduced t-PA release in cigarette smokers (15,16,18). Although not demonstrated with t-PA antigen, the present finding of reduced bradykinin induced active t-PA release is consistent with similar observations previously reported by Pretorius et al. (16) However, the magnitude of the reduction in t-PA release is substantially greater for PAR-1–evoked responses than it is for bradykinin or substance P (96% vs. 40% to 50%) (15,16). We would therefore argue that SFLLRN-evoked t-PA release has the potential to be a more sensitive and pathophysiologically relevant assessment of endothelial vasomotor and fibrinolytic function.
Of note, PAR-1 activation also caused the release of PAI-1 antigen but did not cause an appreciable increase in PAI-1 activity, and neither indices were altered by smoking status. This increase in PAI-1 antigen without a change in activity suggests that SFLLRN is releasing PAI-1 from platelets rather than the endothelium because platelet-derived PAI-1 is relatively inactive as a result of the absence of the stabilizing effects of vitronectin (16,21). Furthermore, our own recent work has demonstrated a concomitant increase in beta-thromboglobulin, suggesting degranulation of platelet alpha granules (22). Therefore, the contribution of the endogenous fibrinolytic system to the prothrombotic state found in cigarette smokers is likely to be driven by impaired endothelial t-PA release and not by alterations in PAI-1 release or activity.
PAR-1 activation as a pathophysiologically relevant marker of endothelial function
The authors of previous studies to assess the endothelial release of endogenous fibrinolytic factors have used diverse methods. Historical means of stimulating t-PA release have included the systemic intravenous infusion of desmopressin and bradykinin, but this method causes significant confounding effects by altering systemic hemodynamics, activation of the sympathetic nervous system, and concomitant release of other mediators (6). By assessing the regional release of t-PA and PAI-1 in response to locally acting agonists, we can avoid such confounding effects.
We have previously demonstrated that substance P-induced t-PA release in the coronary (17) and peripheral (15) arterial circulations is impaired in cigarette smokers, and allows one to predict future adverse cardiovascular events in patients with coronary heart disease (11). However, although substance P has been a useful pharmacologic tool, it is unclear whether substance P is likely to act as a major pathophysiological mediator in atherothrombosis. In contrast, bradykinin may have a more direct role because it is released during the contact phase of coagulation and there is enhanced activation of the kallikrein system and bradykinin release in patients with acute coronary syndromes (23). However, we would argue that, given its central role in thrombosis and inflammation, thrombin is the most powerful and pathophysiologically relevant mediator in this setting. Our present findings not only reinforce previous findings but give the clearest indication yet that impaired endothelial function is of critical and dynamic importance in the setting of coronary heart disease and acute coronary syndromes.
Smoking and endothelium-dependent mechanisms
We have previously demonstrated that PAR-1 mediates arterial vasodilation via 2 endothelium-dependent mechanisms, namely nitric oxide (NO) and endothelium-derived hyperpolarizing factor (EDHF) (22). The pathways via which PAR-1 activation causes the endothelial release of t-PA are less clear and, in fact, inhibition of NO synthesis causes augmented SFLLRN-induced t-PA release (22). This discrepancy has raised the question as to whether EDHF is responsible for t-PA release and, in the absence of NO, whether EDHF responses undergo a compensatory up-regulation. Although the bulk of evidence suggests that smoking predominantly affects endothelial function by increasing oxidative stress with consequent disruption of NO production (24,25), studies specifically examining the effect of smoking upon EDHF-mediated responses are lacking.
The forearm circulation has been an extremely reliable model for the assessment of vascular physiology and pathophysiology. We do accept that our findings in the forearm may not be accurately representative of the coronary circulation. However, we and others have previously demonstrated consistent findings of impaired endothelial t-PA release in both the forearm (15,16) and coronary (17,18) circulations of cigarette smokers. Although the forearm vascular bed is relatively protected from the development of atheroma, it therefore seems likely that changes in its fibrinolytic capacity are indicative of the coronary circulation.
Establishing the receptor-mediated effects of thrombin in the vasculature is of major physiological and therapeutic relevance. It could be argued that, in our studies, the safety requirement for the coadministration of tirofiban with SFLLRN detracts from these advantages. However, we used locally active doses of glycoprotein IIb/IIIa inhibitor that abolish SFLLRN-mediated platelet aggregation without affecting platelet-monocyte binding, a sensitive marker of platelet activation. Furthermore, it has no effect upon basal forearm blood flow or fibrinolytic responses to SFLLRN (14). We therefore believe that SFLLRN remains an important and relevant tool to assess these fundamental pathophysiological aspects of endothelial function.
We have demonstrated an important impairment of fibrinolytic capacity in smokers, but it remains unclear whether this reflects an impairment of synthesis, storage, and release of t-PA, or indeed acceleration of its degradation. Addressing these questions will be challenging and is likely to require specifically designed in vitro studies.
In healthy vessels, thrombin's powerful procoagulant and prothrombotic effects are offset by its ability to evoke the release of t-PA and induce arterial vasodilation. We have shown here that cigarette smoking causes a marked impairment in PAR-1–mediated endothelial vasomotor and fibrinolytic function. Relative arterial stasis and abolition of t-PA release will strongly enhance clot expansion and vessel occlusion. Taken together, these findings suggest a major contribution of impaired endothelial PAR-1 action to the increased atherothrombotic risk of smokers. These important and novel findings are of direct relevance to our understanding of the pathophysiology by which cigarette smoking causes an increased propensity to atherothrombotic disorders including acute myocardial infarction and stroke.
We are grateful to the staff of the Wellcome Trust Clinical Research Facility in Edinburgh and to Pamela Dawson for their help with these studies.
Support was provided by Bristol-Myers Squibb Cardiovascular Prize Fellowship, United Kingdom (to Dr. Lang), the British Heart Foundation Junior Research Fellowship, United Kingdom (FS/05/028; to Dr. Guðmundsdóttir), and the British Medical Association Lansdell and Lawson Research Grant, United Kingdom. Drs. Lang and Guðmundsdóttir contributed equally to this work.
- Abbreviations and Acronyms
- analysis of variance
- endothelium derived hyperpolarizing factor
- nitric oxide
- plasminogen activator inhibitor type 1
- protease-activated receptor
- tissue-type plasminogen activator
- von Willebrand factor
- Received January 7, 2008.
- Revision received March 25, 2008.
- Accepted April 2, 2008.
- American College of Cardiology Foundation
- Ambrose J.A.,
- Barua R.S.
- Davies M.J.,
- Woolf N.,
- Rowles P.M.,
- Pepper J.
- Oliver J.J.,
- Webb D.J.,
- Newby D.E.
- Jansson J.H.,
- Olofsson B.O.,
- Nilsson T.K.
- Fox K.,
- Robison A.,
- Knabb R.,
- Rosamond T.,
- Sobel B.,
- Bergmann S.
- Robinson S.D.,
- Ludlam C.A.,
- Boon N.A.,
- Newby D.E.
- Hollenberg M.D.,
- Compton S.J.
- Guðmundsdóttir I.J.,
- Megson I.L.,
- Kell J.S.,
- et al.
- Newby D.E.,
- Wright R.A.,
- Labinjoh C.,
- et al.
- Newby D.E.,
- McLeod A.L.,
- Uren N.G.,
- et al.
- Seiffert D.,
- Ciambrone G.,
- Wagner N.,
- Binder B.,
- Loskutoff D.
- Guðmundsdóttir I.J.,
- Lang N.L.,
- Boon N.A.,
- et al.
- Hoffmeister H.,
- Jur M.,
- Wendel H.,
- Heller W.,
- Seipel L.
- Kiowski W.,
- Linder L.,
- Stoschitzky K.,
- et al.