Vascular B1 Kinin Receptors in Patients With Congestive Heart Failure

Abstract

Animal models suggest a vasomotor role for the B1 kinin receptor in cardiovascular disease states. In patients with heart failure treated with angiotensin-converting enzyme inhibition (ACEi), or combined B1/B2 receptor antagonism, but not B2 receptor antagonism alone, causes vasoconstriction. However, B1 agonism has no effect on vasomotor or fibrinolytic function. Findings from transgenic animals lacking the B2 receptor suggest that these conflicting data may be explained by cross-talk between B1 and B2 receptors. We hypothesized that B1 stimulation causes vasodilatation and tissue plasminogen activator release in the human forearm when B2 receptor signaling is inhibited. Forearm blood flow was measured in 16 patients with heart failure receiving ACEi. In double-blinded crossover studies, intrabrachial Lys-[Leu8]-des-Arg9-bradykinin (B1 antagonist), lys-des-Arg9-bradykinin (B1 agonist), bradykinin (B2 agonist), and sodium nitroprusside (endothelium-independent vasodilator) were infused alone or with HOE-140 (B2 antagonist). HOE-140 did not affect basal vascular tone or t-PA release, but it abolished bradykinin-induced vasodilatation and t-PA release (P < 0.0001). Blood flow and t-PA release were unaffected by B1 agonism or antagonism in the presence and absence HOE-140. Our findings do not support a role for crosstalk between the B1 and B2 kinin receptors in the human peripheral circulation.

INTRODUCTION

Bradykinin is a potent endothelium-dependent vasodilatator and powerful stimulant for the endothelial release of the endogenous fibrinolytic factor, tissue plasminogen activator (t-PA).^1,2 Bradykinin is principally released at sites of inflammation and coagulation, and it is strongly implicated in the protection of the circulation from acute thrombotic events, particularly those associated with atheromatous disease.³

Bradykinin is rapidly degraded by angiotensin-converting enzyme (ACE),⁴ and its actions are therefore markedly potentiated by ACE inhibitor therapy.⁵ We and others have suggested that bradykinin contributes to the haemodynamic^6,7 and antiischemic^5,8 benefits afforded by these drugs. Whilst ACE inhibition increases the plasma concentration of biologically active bradykinin, it also favours its metabolism via alternative pathways that include the removal of its C-terminal arginine, resulting in the production of des-Arg⁹-bradykinin. Des-Arg⁹-bradykinin is the principal ligand for the B1 kinin receptor, whilst bradykinin exerts its vascular effects via the constitutively expressed B2 receptor. Therefore, ACE inhibition may potentiate both B1 and B2 receptor-mediated effects.⁴

There is compelling evidence that the B2 receptor mediates the vascular effects of bradykinin,^1,9 but the role of the vascular B1 receptor in humans is less clear. The B1 kinin receptor is normally expressed very weakly in the vasculature, but it is markedly upregulated in the presence of left ventricular dysfunction,¹⁰ cardiovascular disease,¹¹ inflammation,¹² and ACE inhibition.¹³ Atheromatous human coronary arteries display intense B1 receptor expression¹⁴ that evoke vasodilatation when stimulated in vitro,¹⁵ and B1 receptor agonism causes vasodilatation and a decrease in blood pressure in animal studies.^12,16-18 Conversely, smooth muscle B1 receptors may evoke arterial constriction. Indeed, intracoronary administration of a B1 agonist in cardiac transplant patients causes minor vasoconstriction that is independent of endothelial function.¹⁹

In our own previous studies, we have shown that the combined B1 and B2 antagonist, B9340, causes vasoconstriction in the forearm arterial circulation of patients with heart failure in the presence, but not absence, of ACE inhibition.²⁰ However, neither selective B1 antagonism²¹ nor selective B2 antagonism²⁰ has been shown to evoke a similar effect. Furthermore, isolated B1 receptor agonism has no effect on peripheral arterial vasomotion or t-PA release.²¹ This apparent discrepancy may reflect crosstalk between the B1 and B2 kinin receptors, such that B1-mediated effects are apparent only in the absence of B2-mediated signalling.²¹ Indeed, both receptors are coupled to similar G-protein subtypes and share the same intracellular signalling pathways.⁴ In transgenic murine models engineered to lack the B2 receptor, B1 receptor expression becomes upregulated, and these receptors assume vascular functions normally associated with the B2 receptor.²² Furthermore, we have previously demonstrated that vasomotor responses to a potent and specific B1 agonist [Lys-des-Arg⁹-bradykinin (LDABK)] are potentiated in the presence of a B2 antagonist in human tissue in vitro.²¹

The aims of this study were to examine in patients with heart failure treated with chronic ACE inhibition the vasomotor and endogenous fibrinolytic effects of B1 receptor agonism and antagonism in the presence and the absence of concurrent B2 receptor antagonism.

METHODS

This study was performed with the approval of the local research ethics committee in accordance with the Declaration of Helsinki and with the written informed consent of each patient.

Patients

Sixteen patients with symptomatic heart failure (New York Heart Association class II or III) and evidence of left ventricular systolic dysfunction (ejection fraction less than 40%, shortening fraction less than 20%, or left ventricular end diastolic dimension greater than 55 mm) were recruited. Patients were maintained on maximally tolerated ACE inhibitor therapy (10 mg of enalapril twice daily or equivalent) for at least 6 months before enrolment.

Patients with a history of asthma, diabetes mellitus, coagulopathy, hyperlipidaemia, renal or hepatic failure, valvular heart disease, or significant concurrent illness and women of childbearing potential were excluded from participation. Subjects abstained from alcohol for 24 hours before and from food, tobacco, and caffeine-containing drinks on the day of the study. Diuretics were withheld on the morning of the study for patient comfort.

Drugs

Pharmaceutical-grade bradykinin (Clinalfa AG, Läufelfingen, Switzerland), Lys-des-Arg⁹-bradykinin (Clinalfa), Lys-[Leu⁸]-des-Arg⁹-bradykinin (Clinalfa), HOE-140 (Clinalfa; Table 1), sodium nitroprusside (David Bull Laboratories, Warwick, UK), and norepinephrine (Abbott Laboratories Ltd, Maidenhead, UK) were dissolved in physiological saline on the day of study.

TABLE 1: Nanomolar Affinity Estimates for Agonists and Antagonists at Human Kinin Receptors

Intraarterial Drug Administration

Patients attended twice for vascular studies, separated by at least 7 days. All studies were performed with the patient lying supine in a quiet, temperature-controlled (22 to 25°C) room. Subjects underwent brachial artery cannulation with a 27 standard-wire-gauge steel needle. The intraarterial infusion rate was kept constant at 1 ml/min throughout all studies.

Antagonist Study

Six patients were recruited for this initial study. Using a double-blind randomized crossover study design, patients received 10 mg of enalapril bid or 50 mg of losartan bid each for 6 weeks each. In the final week of each treatment period, they attended for a vascular study. After a 30-minute infusion of intraarterial 0.9% saline, patients received Lys-[Leu⁸]-des-Arg⁹-bradykinin (LLDABK, B1 antagonist; 1, 3, and 10 nmol/min), HOE 140 (B2 antagonist; 1.5, 4.5, and 13.5 nmol/min), norepinephrine (control vasoconstrictor; 60, 180, and 540 pmol/min), and a combination of B1/B2 antagonists (LLDABK + HOE140; 1 + 1.5 nmol/min, 3 + 4.5 nmol/min, and 10 + 13.5 nmol/min, respectively) in random order. Drugs were infused for 10 minutes at each dose separated by 20-minute saline infusions between each drug.

Agonist Study

Ten patients were recruited and received 10 mg of enalapril twice daily in place of their usual ACE inhibitor for 4 weeks. During the final 2 weeks, patients attended twice for vascular studies, separated by at least 7 days. After a 30-minute intraarterial infusion of 0.9% saline, subjects were randomized to receive an intrabrachial infusion of HOE-140 (13.5 nmol/min) or saline placebo for the remainder of the study visit in a double-blind randomized crossover design.^20,21 HOE-140 or saline placebo was infused alone for 20 minutes before co-infusion with bradykinin (30 to 300 pmol/min), Lys-des-Arg⁹-bradykinin (10 to 100 nmol/min), and sodium nitroprusside (2 to 8 μg/min) for 10 minutes at each dose.^20,21 The order in which the drugs were infused was randomized, although the order was maintained constant on both visits for each individual patient. Administration of each drug was separated by a 20-minute infusion of 0.9% saline.

Forearm Blood Flow and Blood Pressure

Forearm blood flow was measured at 10-minute intervals in both forearms using venous occlusion strain gauge plethysmography as previously described.^23,24 Heart rate and blood pressure were recorded in the non-infused arm at intervals throughout the study using a semiautomated non-invasive oscillometric sphygmomanometer (Takeda UA 751; Takeda Medical Inc, Japan).

Venous Sampling and Assays

In the agonist study only, 17-gauge venous cannulae were inserted bilaterally into large antecubital veins. Ten millilitres of blood were withdrawn simultaneously from each arm at baseline and during the last minute of the infusion of each dose of bradykinin and Lys-des-Arg⁹-bradykinin. 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, UK; for plasminogen activator inhibitor type 1 (PAI-1) assays]. Samples were kept on ice before centrifugation at 2000 × g for 30 minutes 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) and PAI-1 antigen and activity (Elitest PAI-1 antigen and Zymutest PAI-1 Activity; Hyphen Biomed, Neuville-Sur-Oise, France) concentrations were determined by enzyme-linked immunosorbant assays. Urea, electrolytes, liver function tests, and fasting glucose and lipid profile were assessed at the beginning of the study by the host institution's Clinical Biochemistry reference laboratory. Full blood count and hematocrit were measured at baseline and the end of the study.

Data Analyses and Statistics

Forearm plethysmographic data were analyzed as described previously. Estimated net release of t-PA and PAI-1 was defined 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 non-infused arms.²⁵ Variables are reported as mean ± SEM and analyzed using analysis of variance (ANOVA) and two-tailed Student t test as appropriate. Serial changes (within group) were compared by ANOVA with post hoc Bonferroni corrections for repeated measurements, and intergroup comparisons were made by 2 way ANOVA. Statistical analyses were performed with GraphPad Prism (Graph Pad Software), and statistical significance was defined as P < 0.05.

The authors had full access to the data and take responsibility for its integrity. All authors have read and agree to the manuscript as written.

RESULTS

Patients were predominantly male with mild to moderate cardiac failure as a result of ischemic heart disease or dilated cardiomyopathy (Table 2). There were no changes in blood pressure, heart rate, or hematocrit (data not shown) during the study.

TABLE 2: Subject Characteristics (n = 16)

Forearm Blood Flow

Antagonist Study

Intraarterial infusion of Lys-[Leu⁸]-des-Arg⁹-bradykinin (B1 antagonist) and HOE140 (B2 antagonist) had no effect on forearm blood flow, either alone or in combination (P = nonsignificant for all; ANOVA), irrespective of prior treatment with enalapril or losartan (P = nonsignificant for all; ANOVA enalapril versus losartan). Norepinephrine caused dose-dependent vasoconstriction (P = 0.009; ANOVA) in all studies (Figure 1).

FIGURE 1: Forearm blood flow during (a) intraarterial Lys-[Leu⁸]-des-Arg⁹-bradykinin ([B1 antagonist] 1 to 10 nmol/min), (b) HOE140 ([B2 antagonist] 1.4 to 13.5 nmol/min), (c) co-infusion of HOE140 plus Lys-[Leu⁸]-des-Arg⁹-bradykinin (Dose 1, 1 + 1.4 nmol/min; Dose 2, 3 + 4.5 nmol/min; Dose 3, 10 + 13.5 nmol/min HOE-140 + Lys-[Leu⁸]-des-Arg⁹-bradykinin, respectively) and, (d) norepinephrine (60 to 540 pmol/min) in the presence of chronic pretreatment with losartan (open symbols) or enalapril (closed symbols). *P = 0.009 for dose response, ANOVA.

Agonist Study

Intraarterial HOE-140 had no effect on resting forearm blood flow (P = nonsignificant, ANOVA; HOE-140 versus placebo: data not shown). Bradykinin caused dose-dependent vasodilatation (P < 0.0001, ANOVA; Figure 2) that was abolished by the co-infusion of HOE-140 (P < 0.0001, ANOVA; HOE-140 versus placebo). Lys-des-Arg⁹-bradykinin had no effect on forearm blood flow (P = nonsignificant, ANOVA; Figure 2) irrespective of the presence or absence of HOE-140 (P = nonsignificant, ANOVA; HOE-140 versus placebo). Sodium nitroprusside caused dose-dependent vasodilatation (P < 0.0001, ANOVA) that was unaltered by HOE-140 (P = non-significant, ANOVA; HOE-140 versus placebo; data not shown).

FIGURE 2: Forearm blood flow in the presence of chronic ACE inhibition during intraarterial Lys-des-Arg⁹-bradykinin (triangles), bradykinin (squares), and sodium nitroprusside (circles) infusion in the presence (closed symbols) and absence (open symbols) of HOE-140 (13.5 nmol/min). *P < 0.0001 in the presence versus the absence of HOE-140 (ANOVA).

Plasma Fibrinolytic Factors

HOE-140 did not affect baseline measures of plasma t-PA and PAI-1 antigen and activity concentrations. Bradykinin caused a dose-dependent net release of t-PA antigen and activity (P = 0.01 and P < 0.0001, respectively, ANOVA) that was abolished by HOE-140 (P = 0.001 and P < 0.0001, respectively, ANOVA; HOE-140 versus placebo). Lys-des-Arg⁹-bradykinin did not affect the net release of t-PA antigen or activity, either in the presence or absence of HOE-140 (P = nonsignificant for all, ANOVA; HOE-140 versus placebo: Figure 3).

FIGURE 3: Net release of t-PA antigen (top panels) and activity (bottom panels) in the presence of chronic ACE inhibition during intraarterial Lys-des-Arg⁹-bradykinin (triangles) and bradykinin (squares) in the presence (closed symbols) and absence (open symbols) of HOE-140 (13.5 nmol/min). *P < 0.0001 in the presence versus the absence of HOE-140 (ANOVA). *P = 0.01, † P < 0.0001 in the presence versus the absence of HOE-140 (ANOVA).

Net release of PAI-1 antigen was unchanged by bradykinin both in the presence and absence of HOE-140 (P = nonsignificant for all). Consistent with the marked release of t-PA, net PAI-1 activity decreased during bradykinin infusion in the absence (P = 0.01, ANOVA; HOE-140 versus placebo) but not the presence of HOE-140 (P = nonsignificant, ANOVA; HOE-140 versus placebo). Lys-des-Arg⁹-bradykinin had no effect on net PAI-1 activity and antigen, irrespective of the presence of HOE-140 (P = nonsignificant, ANOVA; HOE-140 versus placebo: Table 3).

TABLE 3: Net Release of Plasminogen Activator Type 1 (PAI-1) Antigen and Activity in Response to Lys-des-Arg9-bradykinin (LDABK) and Bradykinin (BK) in the Presence Versus the Absence of HOE140

DISCUSSION

In the present study, we have addressed the hypothesis that crosstalk exists between the vascular B1 and B2 kinin receptors in vivo in humans. Using a randomized double-blind crossover design, we have demonstrated that selective B1 receptor agonism has no effect on local vasomotor or endogenous fibrinolytic function either in the presence or absence of concurrent B2 receptor antagonism. Furthermore, selective B1 antagonism does not alter vascular tone in the presence or absence of B2 blockade and in the presence or absence of ACE inhibition. This argues against a major role of the B1 receptor in the cardiovascular effects of chronic ACE inhibition in patients with heart failure.

We have confirmed our earlier findings^21,26 that the B1 receptor does not cause forearm arterial vasodilatation or t-PA release in patients with heart failure treated with long-term ACE inhibition. Previous data from a transgenic rodent model suggested that, in the absence of functional B2 receptor signalling, the B1 receptor is upregulated and assumes the vascular actions normally associated with the B2 kinin receptor.²² We have demonstrated robustly in vivo in humans that the presence of B2 receptor signalling does not exert an inhibitory effect on vascular B1-mediated activity; conversely, the B1 receptor does not assume the role of B2 receptor in its functional absence.

In reaching these conclusions, it should be noted that we have previously confirmed the biological activities of the peptidic kinin receptor agonists and antagonists used in this study in human tissue in vitro.²¹ Similarly, the absence of a B1 receptor response cannot be attributed to a failure to achieve adequate local tissue concentrations. The doses of LDABK used in this study were an order of magnitude greater than those employed in our previous in vivo study to examine LDABK-mediated actions in a similar population.²¹ This dose is at least 200-fold greater than required to produce 50% of the maximal hypotensive effect in primates (EC₅₀ ≈ 0.1 pmol/kg)¹² and rodents (EC₅₀ ≈ 0.3 pmol/kg)¹⁶ and corresponds with an estimated tissue concentration of up to 4.0 μM.

At first glance, our current findings would appear to be at odds with earlier work demonstrating that combined B1 and B2 antagonism with B9340, but not B2 receptor antagonism alone, causes basal forearm arterial vasoconstriction in patients with heart failure in the presence, but not the absence, of ACE inhibition.²⁰ Both HOE-140⁹ and B9340²⁷ inhibit the endothelium-dependent vasodilatation and t-PA release induced by supraphysiological doses of exogenously administered bradykinin. Similarly, HOE-140²⁸ and B9340⁶ both attenuate the haemodynamic effects of ACE inhibition in the human systemic circulation. Previous work examining the contribution of bradykinin to basal vascular tone in the forearm circulation of patients with heart failure treated with enalapril demonstrated an apparent small and nonsignificant vasoconstriction with intraarterial HOE-140,²⁹ the magnitude of vasoconstriction observed being similar to that seen with B9340.²⁰ Furthermore, administration of HOE-140 attenuates the release of t-PA associated with acute ACE inhibition in the human forearm circulation at rest.⁸ Taken together, we believe that these findings suggest that endogenous bradykinin contributes to basal vascular tone and fibrinolytic function in patients with heart failure maintained on chronic ACE inhibition via a B2 receptor-mediated effect, but that the magnitude of this contribution is modest and at the limit of that detectable using the forearm technique. This may be further compounded by subtle differences in the pharmacological profiles of HOE-140 and B9340.

It could be argued that our failure to demonstrate a vascular response to B1 receptor agonism in the presence of concurrent B2 antagonism simply reflects the fact that the B1 receptor is maximally stimulated in patients with heart failure treated with ACE inhibition. However, we have also demonstrated that isolated B1 receptor antagonism has no effect on basal forearm vascular tone in patients with heart failure treated with enalapril, thus confirming our earlier findings.²¹ Furthermore, we have now demonstrated that the simultaneous infusion of both B1 and B2 receptor antagonists has no effect on resting forearm vascular tone. These data lend further support to our conclusion that neither crosstalk nor redundancy exists between B1 and B2 kinin receptors in the human forearm vasculature.

Having discounted inactivity or an insufficient dose of LDABK as reasons for the lack of vascular responses to B1 agonism in vivo, it could be argued that the study population did not have heart failure of sufficient severity to evoke the upregulation of B1 receptor expression. The participants in this study all had mild to moderate heart failure [New York Heart Association (NYHA) Class II and III]. To address this, we examined forearm vasomotor responses to B1 kinin receptor agonism in the subgroup of patients with NYHA class III heart failure from this and our previous study (n = 10).²¹ Consistent with the current study, LDABK did not evoke a peripheral vasomotor response in the subgroup with NYHA Class III symptoms (basal forearm blood flow 2.27 ± 0.23 ml/min/100 mL versus 2.05 ± 0.40 ml/min/100 mL during 10 nmol/min Lys-des-Arg9-bradykinin; P = NS, paired t test). Whilst assessment of patients with NYHA class IV heart failure may be pertinent, the practicalities of undertaking studies of this nature in this population are extremely challenging.

CONCLUSION

We have demonstrated that the B1 kinin receptor does not contribute to the vasomotor or fibrinolytic effects of kinins in the forearm circulation of patients with heart failure treated with long-term ACE inhibition. Specifically, we have rejected the hypothesis that crosstalk or redundancy exists between the vascular B1 and B2 receptor. In patients receiving an effective, evidence-based³⁰ and standardized dose of ACE inhibitor, these findings remain consistent irrespective of symptom severity and at doses of B1 agonist higher than previously assessed. Our results do not support a significant role for the B1 receptor in mediating the vascular effects associated with ACE inhibition.

ACKNOWLEDGMENTS

We thank Pamela Dawson and the staff of the Wellcome Trust Clinical Research Facility for their assistance with the conduct of this study.