The kinin B₁ receptor contributes to the cardioprotective effect of angiotensin-converting enzyme inhibitors and angiotensin receptor blockers in mice

Abstract

Recent studies have shown that inhibition of angiotensin-converting enzyme (ACE) or angiotensin II receptors causes upregulation of the B₁ receptor (B₁R). Here we tested the hypothesis that activation of the B₁R partly contributes to the cardiac beneficial effect of ACE inhibitor (ACEi) and angiotensin II receptor blockers (ARB). B₁R knockout mice (B₁R^−/−) and C57Bl/6J (wild-type control animals, WT) were subjected to myocardial infarction (MI) by ligating the left anterior descending coronary artery. Three weeks after MI, each strain of mice was treated with vehicle, ACEi (ramipril, 2.5 mg kg⁻¹ day⁻¹ in drinking water) or ARB (valsartan, 40 mg kg⁻¹ day⁻¹ in drinking water) for 5 weeks. We found that: (1) compared with WT mice, B₁R^−/− mice that underwent sham surgery had slightly but significantly increased left ventricular (LV) diastolic dimension, LV mass and myocyte size, whereas systolic blood pressure, cardiac function and collagen deposition did not differ between strains; (2) MI leads to LV hypertrophy, chamber dilatation and dysfunction similarly in both WT and B₁R^−/− mice; and (3) ACEi and ARB improved cardiac function and remodelling in both strains; however, these benefits were significantly diminished in B₁R^−/− mice. Our data suggest that kinins, acting via the B₁R, participate in the cardioprotective effects of ACEi and ARB.

Kinins are vasodilator peptides released from low- and high-molecular-weight kininogens by plasma and tissue kallikreins and hydrolysed by angiotensin (Ang)-converting enzyme (ACE), neutral endopeptidase-24.11 (NEP) and other peptidases (Erdös, 1979, 1990). Two types of kinin receptors, B₁ and B₂ (B₁R and B₂R), have been well defined; the B₁ receptor is preferentially activated by its endogenous ligands, des-Arg⁹-bradykinin or des-Arg¹⁰-kallidin [carboxypeptidase N/kininase I metabolites of bradykinin (BK) or kallidin (KD)], whereas the B₂ receptor is activated by BK or KD (Regoli & Barabe, 1980). Most biological effects of kinins involved in vasodilatation, vascular permeability, inflammation and tissue repair are known to be mediated via the constitutively expressed B₂R. The B₁R is weakly expressed in physiological conditions but is strongly induced in response to pathological stimuli such as inflammation or tissue injury (Ignjatovic et al. 2002a; Lagneux et al. 2002; Xu et al. 2005). Up-regulation of the B₁R is also demonstrated when the B₂R is absent, which may in part compensate for the loss of the B₂R (Duka et al. 2001; Ignjacev-Lazich et al. 2005). Kakoki et al. (2007) have shown that renal ischaemia–reperfusion injury is more severe in mice lacking both B₁R and B₂R compared with mice lacking B₂R alone, suggesting a cardio-renal protective role of the B₁R. Activation of the B₁R also reportedly contributes to the beneficial effects of ACE inhibitors (ACEi) and angiotensin II receptor blockers (ARB; Ignjatovic et al. 2002a,b; Tschöpe et al. 2004). For example, Ignjatovic and co-workers reported that in endothelial cells and Chinese hamster ovary (CHO) cells transfected with B₁R, ACEi activate the B₁R directly via activation of the Zn²⁺-binding consensus sequence, HEXXH, in the absence of ACE and B₁R agonist, which was associated with increased [Ca²⁺] and nitric oxide (NO) release (Ignjatovic et al. 2002a,b). In rats with myocardial infarction (MI), the ARB irbesartan improved cardiac function and increased B₁R expression in the heart (Tschöpe et al. 2004).

Based on these findings, we hypothesized that B₁R plays an important role in the cardioprotective effects of ACEi and ARB. For this study, we used B₁R-null mice (B₁R^−/−) and studied whether lack of B₁R: (1) worsens cardiac remodelling and dysfunction post-MI; and (2) diminishes the cardioprotective effect of ACEi and ARB.

Methods

Animal procedures

This study was approved by the Institutional Animal Care and Use Committee of Henry Ford Health System. Two breeding pairs of B₁ kinin receptor knockout mice (B₁R^−/−) in a C57/J6 genetic background were obtained from Dr Bader's laboratory (Pesquero et al. 2000) and bred in our Mutant Mouse Facilities. Since B₁R^−/− mice have been backcrossed with the C57BL/6 strain for 10 generations and are C57BL/6J congenicas, we used C57BL/6J as wild-type (WT) control animals. The C57BL/6J mice were purchased from Jackson Laboratories (Bar Harbor, ME, USA). Animals were housed in an air-conditioned room with a 12 h–12 h light–dark cycle and received standard mouse chow and tap water. For induction of myocardial infarction, female mice 10−12 weeks of age were anaesthetized with sodium pentobarbitone (50 mg kg⁻¹, I.P.), both sham MI and MI were surgically induced by ligating the left anterior descending coronary artery as described previously (Xu et al. 2004).

Experimental protocols

Mice of both WT and B₁R^−/− strains were subjected to sham MI or MI and divided into four groups, as follows: (1) sham MI; (2) MI–vehicle; (3) MI–ACEi (ramipril, 2.5 mg kg⁻¹ day⁻¹ in drinking water, Sigma, St Louis, MO, USA); and (4) MI–ARB (valsartan, 40 mg kg⁻¹ day in drinking water, provided by Novartis, Cambridge, MA, USA). Treatments were started 3 weeks after MI and continued for 5 weeks. We have previously shown that in mice the vasopressor effect of exogenous Ang I and II was significantly inhibited by ramipril at 2.5 mg kg⁻¹ day and valsartan at 40 mg kg⁻¹ day⁻¹, respectively (Xu et al. 2002, 2004).

Measurement of blood pressure and cardiac function

Systolic blood pressure (SBP) and heart rate (HR) were measured in conscious mice weekly using a non-invasive computerized tail-cuff system (BP-2000, Visitech Systems, Apex, NC, USA) as described previously (Krege et al. 1995). Left ventricular diastolic dimension (LVDd), mass, cardiac index (CI) and ejection fraction (EF) were measured monthly with a Doppler echocardiographic system equipped with a 15 MHz linear array transducer (Acuson c256, Mountain View, CA, USA) as described previously (Yang et al. 1999; Xu et al. 2002). All studies were performed on awake mice before MI and periodically thereafter.

Histopathological study

Eight weeks after induction of sham MI or MI, mice were anaesthetized with sodium pentobarbitone (50 mg kg⁻¹, I.P.) and the heart stopped at diastole by intraventricular injection of 15% KCl (50 μl). The heart, lungs and liver were weighed, corrected by body weight and expressed as a ratio of organ wet weight (in milligrams) to body weight (10 g). The left ventricle (LV) was sectioned transversely into three slices from apex to base and rapidly frozen in isopentane precooled in liquid nitrogen, then stored at −70°C. Infarct size was determined by Gomori trichrome staining and expressed as the ratio of the infarcted portion to total LV circumference. For myocyte cross-sectional area (MCSA) and interstitial collagen fraction (ICF), 6 μm sections from each slice were double-stained with: (1) fluorescein-labelled peanut agglutinin to delineate the MCSA and interstitial space; and (2) rhodamine-labelled Griffonia simplicifolia lectin I to show the capillaries (Liu et al. 1997). The MCSA was measured by computer-based planimetry (Jandel, Corte Madera, CA, USA). The ICF was calculated as the percentage of the total surface area occupied by the interstitial space minus the percentage of the total surface area occupied by the capillaries. Mice with less than 20% of infarct size were excluded from analysis.

Statistical analysis

Student's two-sample unpaired t test was used to compare the difference between groups, where groups were either between strains or between treatments within strains. When multiple comparisons were performed, Hochberg's step-up procedure was used to adjust the P values (Hochberg & Benjamini, 1990). The type I error rate is set at 0.05 level, and data are expressed as means ± S.E.M.

Results

Mortality

The surgical mortality of MI tended to be higher in the B₁R^−/− group than in WT control mice (12.5 and 3%, respectively), but the differences were not statistically significant. Early (within 1 week of MI) and late mortality (between 1 and 8 weeks of MI) was similar between strains with and without treatment. No animals died in the sham groups (both strains) either during or after the operation.

Systolic blood pressure, heart rate, tissue weight and infarct size

The SBP and HR were similar between strains with either sham MI or MI. The ACEi decreased SBP in both WT and B₁R^−/− mice over a 5 week treatment period, but no strain difference was found (Fig. 1, upper panel). The ARB had no significant influence on SBP. Heart rate was not altered by either ACEi or ARB treatment (Fig. 1, lower panel). Body weight was similar in all groups. Organ weights in sham MI groups were no different between strains. Myocardial infarction significantly increased LV and total heart weight to a similar extent in both WT and B₁R^−/− mice. The ACEi significantly reduced LV and total heart weight in WT mice, and this response was diminished in B₁R^−/− mice. The ARB reduced LV and total heart weight in both WT and B₁R^−/− mice; however, the magnitude of reduction was significantly smaller in B₁R^−/− compared with WT control animals. There was no significant change in lung or liver weight after MI. Infarct size was similar among groups (Table 1).

Figure 1.

Effect of angiotensin-converting enzyme inhibitor (ACEi) and angiotensin II receptor blocker (ARB) on systolic blood pressure (SBP) and heart rate (HR) in wild-type (WT) and kinin B₁R knockout (B₁R^−/−) mice with myocardial infarction (MI)

††P < 0.01 ACEi versus vehicle within strain.

Table 1.

Effect of ACE inhibitor or ARB on tissue weight and infarct size at 8 weeks after myocardial infarction in WT and B₁R^−/− mice

	Sham MI		MI + vehicle		MI + ACEi		MI + ARB
Parameter	WT	B1R−/−	WT	B1R−/−	WT	B1R−/−	WT	B1R−/−
n	10	10	11	10	10	12	10	12
BW (g)	22.8 ± 0.4	22.2 ± 1.1	22.4 ± 0.3	24.0 ± 0.6	22.0 ± 0.5	22.3 ± 0.5	22.6 ± 0.4	22.7 ± 0.6
LVW (mg (10 g)−1)	31.4 ± 0.7	32.7 ± 1.2	45.3 ± 1.3	44.6 ± 1.2	35.5 ± 0.8††	41.2 ± 1.9‡	37.7 ± 1.0††	40.5 ± 0.9†‡
THW (mg (10 g)−1)	41.8 ± 0.9	43.7 ± 1.4	60.1 ± 1.9**	60.0 ± 1.8***	50.4 ± 1.3††	56.2 ± 2.1	52.9 ± 1.4††	54.9 ± 1.1†
Lungs (mg (10 g)−1)	69.1 ± 2.3	71.1 ± 2.3	73.8 ± 5.4	74.1 ± 5.9	77.6 ± 4.3	75.6 ± 2.8	81.1 ± 5.4	75.8 ± 3.7
Liver (mg (10 g)−1)	448 ± 12	445 ± 12	461 ± 12	459 ± 19	433 ± 8	454 ± 14	438 ± 10	413 ± 11
Infarct size (%)	–	–	34.9 ± 2.5	32.1 ± 1.9	32.5 ± 2.9	33.0 ± 2.5	31.7 ± 1.7	32.9 ± 1.8

Abbreviations: Sham, sham operation; MI, myocardial infarction; ACEi, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; WT, wild-type control animals; B₁R^−/−, kinin B₁ receptor knockout animals; BW, body weight; LVW, left ventricular weight corrected by body weight; and THW, total heart weight corrected by body weight.

^**P < 0.01

^***P < 0.001, MI + vehicle versus sham MI within strain

^†P < 0.05

^††P < 0.01 ACEi or ARB versus vehicle within strain

^‡P < 0.05 between strains with same treatment.

Cardiac function and remodelling

As shown in Fig. 2, EF and CI were similar between WT and B₁R^−/− mice subjected to sham MI; however, LVDd was greater in B₁R^−/− than in WT mice in basal conditions. After MI, EF and CI decreased and LVDd increased to a similar extent in both strains receiving vehicle. Both ACEi and ARB (5 weeks treatment) increased EF and reduced LVDd in both WT and B₁R^−/− animals; however, the responses to ACEi and ARB were significantly diminished in B₁R^−/− compared with WT mice (Figs 3 and 4).

Figure 2.

Comparison of cardiac phenotypes between WT and B₁R^−/− mice with sham MI or MI

Abbreviations: EF, ejection fraction; CI, cardiac index; and LVDd, left ventricular diastolic dimension. †P < 0.05, ††P < 0.01 and †††P < 0.001 for sham versus MI within strain.

Figure 3.

Effect of ACEi and ARB on ejection fraction (EF) in WT and B₁R^−/− mice with MI

Bar graph shows change in EF after treatment (week 3 minus average of weeks 6 and 8) between strains. Abbreviation: Veh, vehicle. †P < 0.05 and ††P < 0.01 for ACEi and ARB versus vehicle within strain.

Figure 4.

Effect of ACEi and ARB on left ventricular diastolic dimension (LVDd) in WT and B₁R^−/− mice with MI

Bar graph shows change in LVDd after treatment (week 3 minus average of weeks 6 and 8) between strains. Abbreviation: Veh, vehicle. †P < 0.05 and ††P < 0.01 for ACEi and ARB versus vehicle within strain.

Myocyte size and interstitial fibrosis

In sham-operated mice, myocyte cross-sectional area was larger in B₁R^−/− than in WT animals (Figs 5 and 6), whereas interstitial collagen fraction (ICF) was similar between strains (Figs 5 and 7). After MI, MCSA and ICF increased in similar manner in vehicle-treated groups of both strains. Treatment with ACEi or ARB significantly decreased MCSA and ICF in both strains of mice, and these effects were diminished in B₁R^−/− animals (Figs 5–7).

Figure 5.

Representative micrographs showing myocyte cross-sectional area and interstitial collagen deposition (green staining) in WT and B₁R^−/− mice with or with out MI, as well as their response to ACEi and ARB

The scale bar represents 100 μm.

Figure 6.

Effect of ACEi and ARB on myocyte cross-sectional area (MCSA) in WT and B₁R^−/− mice with sham MI or MI

The right-hand bar graph shows the difference in MCSA between vehicle and ACEi or ARB. ***P < 0.001, vehicle versus sham within strain. †P < 0.05 and ††P < 0.01, ACEi and ARB versus vehicle within strain.

Figure 7.

Effect of ACEi and ARB on interstitial collagen fraction (ICF) in WT and B₁R^−/− mice with sham MI or MI

The right-hand bar graph shows the difference in ICF between vehicle and ACEi or ARB. ***P < 0.001, vehicle versus sham within strain. †P < 0.05 and ††P < 0.01, ACEi and ARB versus vehicle within strain.

Discussion

In the present study using B₁R^−/− mice, we found that in basal conditions, lack of the B₁ receptor had no effect on SBP and cardiac function; however, B₁R^−/− mice showed enlarged LV chamber dimension and myocyte size compared with the WT control animals. When subjected to MI, the development and severity of cardiac dysfunction and remodelling were similar in B₁R^−/− and WT mice. Inhibition of ACE or blockade of the AT₁ receptor improved cardiac function and remodelling in both strains, as evidenced by increased EF and reduced LV chamber dimension, heart weight, myocyte size and interstitial collagen deposition; however, these effects were diminished in B₁R^−/− mice. This study provides important evidence that although the B₁R does not seem to have a role in regulation of cardiac haemodynamics and function, it appears to be involved in maintaining myocardial structure integrity, since B₁R^−/− mice had enlarged LV chamber and myocyte size in basal conditions. More importantly, we believe ours is the first study to show that the B₁R participates in the effects of both ACEi and ARB, since lack of B₁R diminished the cardioprotective effects of ACEi and ARB in mice post-MI.

Most known cardiovascular actions of kinins are attributed to activation of constitutively expressed B₂R. In contrast, the B₁R is generally absent or expressed at low levels in most tissues, but is up-regulated by inflammation and tissue injury. The pathophysiological role of up-regulation of B₁R in the cardiovascular system in response to stimuli has recently drawn attention. Duka and co-workers showed that B₁R expression in the heart was significantly increased in B₁R^−/− mice, and that B₁R overexpression was further enhanced in response to Ang II infusion (Kintsurashvili et al. 2001; Duka et al. 2003). Kakoki et al. (2007) also demonstrated that B₁R expression in the kidney was markedly induced in the absence of the B₂R. These studies suggest that up-regulation of the B₁R may play a compensatory role when the B₂R is absent.

In the present study, we found that lack of the B₁R does not alter blood pressure or cardiac function either in normal conditions or after MI. This agrees with our previous finding that the B₁R does not seem to be involved in the regulation of cardiac haemodynamics and may not play a significant pathophysiological role in the development of LV remodelling and dysfunction (Xu et al. 2005). However, we cannot exclude the possibility that lack of B₁R may lead to up-regulation of B₂R expression, which compensates for loss of the B₁R, particularly in the presence of pathological stimuli such as MI. This assumption is supported by our previous findings that in B₁R^−/− mice subjected to MI, blockade of the B₂R worsened cardiac dysfunction and remodelling compared with B₁R^−/− animals treated with vehicle, indicating a compensatory role of the B₂R in mice lacking the B₁R. However, we did find that B₁R^−/− mice have enlarged LV chamber dimension and increased LV mass and myocyte size in basal conditions, indicating that existence of the B₁R may be necessary for maintaining cardiac morphological and structural integrity in physiological conditions.

There is evidence that kinins contribute to the cardioprotective effect of ACEi (Liu et al. 1997; Leung et al. 2000; Yang et al. 2001; Tschope et al. 2002). We previously showed that kininogen-deficient Brown Norway Katholiek rats (BNK) or mice lacking the B₂R had a diminished response to ACEi (Liu et al. 2000; Yang et al. 2001), whereas Koch et al. (2008) recently reported that ACE inhibition is sufficient to improve LV function despite kininogen deficiency. The difference between the data of Koch and co-workers and our own may be due to the fact that in Koch's study the ACEi was initiated 2 days after MI, whereas we started it 2 months after MI. Nevertheless, Koch and co-workers did show that impairment of LV function post-MI is more severe in rats with kinin deficiency (Koch et al. 2008). Recently, the role of the B₁R in the beneficial effects of ACEi and ARB has been explored (Emanueli et al. 2000; Ignjatovic et al. 2002a,b; Tschöpe et al. 2004). In rats, as well as in B₂R-null mice, chronic ACEi administration increased B₁R expression in the vasculature and blockade of the B₁R diminished the effect of ACEi, suggesting that the functional up-regulation of the B₁R contributes to the hypotensive effect of ACEi (Marin-Castaño et al. 2002). Duka and co-workers recently offered evidence that lack of the B₁R increases B₂R mRNA expression; however, upregulation of the B₂R is insufficient to accommodate the kinin-mediated effect of ACEi in B₁R-null mice (Duka et al. 2008), emphasizing the importance of the B₁R in the cardioprotective effect of ACEi. The B₁R may also contribute to the cardioprotective effect of ARB. Tschöpe et al. (2004) showed that in rats with MI, the ARB irbesartan improved LV contractile function and this effect was reversed by blockade of B₁R. In the present study using B₁R^−/− mice subjected to MI, we found that ACEi and ARB reduced cardiac hypertrophy, dilatation and collagen deposition and improved ejection fraction in both WT and B₁R^−/− animals; however, the degree of improvement was significantly diminished in B₁R^−/− compared to WT mice. We also found that ACEi and ARB decreased SBP in both strains to a similar extent. Our data further confirm that the B₁R participates in the cardioprotective effect of ACEi and ARB, independent of a blood-pressure-lowering effect.

The precise mechanism(s) by which the B₁R mediates the effect of ACEi and ARB are not fully understood. It has been shown that in rats chronic ACE inhibition was associated with functional up-regulation of the B₁R in kidney, heart and vasculature (Marin-Castaño et al. 2002). Ignjatovic et al. (2002a,b)) showed that in cultured endothelial cells (ECs), the ACEi enalaprilat activates the B₁R via activation of the Zn²⁺-binding consensus sequence, HEXXH, in the B₁R. It was also shown that B₁R activation by enalaprilat in creases the uptake of l-arginine and leads to prolonged NO release, which is blocked by a B₁R antagonist (Ignjatovic et al. 2002b, 2004). Agata and Chao and co-workers (Agata et al. 2000; Chao et al. 2007) demonstrated that the B₁R contributes to kinin-induced reduction of vascular and renal injury via NO–cGMP and mitogen-activated protein kinase signalling pathways. However, using isolated rabbit aorta or mouse stomach preparations, Fortin et al. (2003) had failed to demonstrate that ACEi acts as a B₁R agonist to stimulate tissue response. Concerning the role of the B₁R in the mechanism of action of the ARB, Tschöpe et al. (2004) showed that in transgenic rats with MI, overexpression of the angiotensin II type 1 receptor (AT₁R) causes down-regulation of the B₁R, whereas ARB treatment leads to upregulation of the B₁R, suggesting that the AT₁R may exertadirect inhibitory effect on the B₁R and this interaction is eliminated by the ARB. However, further evidence is needed to demonstrate the existence of a cross-talk between the AT₁R and the B₁R, and the intracellular mechanisms responsible for such interaction need to be explored as well.

In conclusion, the kinin B₁R does not seem to play an essential role in cardiac haemodynamics and function either in normal conditions or during development of heart failure; however, it may be involved in maintaining cardiac structure and morphological integrity, since mice with targeted deletion of the B₁R had increased LV mass and chamber dimension in basal conditions. Furthermore, it appears that the B₁R is necessary for the action of ACEi and ARB, since lack of B₁R diminishes their cardioprotective effects.