Author + information
- Received April 30, 2001
- Revision received April 29, 2002
- Accepted May 24, 2002
- Published online September 4, 2002.
- Hisakazu Ogita, MD*,
- Koichi Node, MD*,* (, )
- Hiroshi Asanuma, MD*,
- Shoji Sanada, MD*,
- Yulin Liao, MD*,
- Seiji Takashima, MD*,
- Masanori Asakura, MD*,
- Hidezo Mori, MD§,
- Yoshiro Shinozaki, MD†,
- Masatsugu Hori, MD, FACC* and
- Masafumi Kitakaze, MD, FACC‡
- ↵*Reprint requests and correspondence:
Dr. Koichi Node, Department of Internal Medicine and Therapeutics, Osaka University Graduate School of Medicine, 2-2 Yamada-oka, Suita, 565-0871, Osaka, Japan.
Objectives We sought to investigate whether raloxifene reduces ischemia-reperfusion injury and what mechanisms are involved in the cardioprotective effects.
Background Estradiol-17-beta reduces myocardial infarct size in ischemia-reperfusion injury. Raloxifene, a selective estrogen receptor modulator, demonstrates immediate coronary artery vasorelaxing effects.
Methods The myocardial ischemia-reperfusion model included anesthetized open-chest dogs after 90-min occlusion of the left anterior descending coronary artery (LAD) and subsequent 6-h reperfusion. Raloxifene and/or other drugs were infused into the LAD from 10 min before coronary occlusion to 1 h after reperfusion without an occlusion period.
Results Infarct size was reduced in the raloxifene (5 μg/kg per min) group compared with the control group (7.2 ± 2.5% vs. 40.9 ± 3.9% of the area at risk, p < 0.01). Either NG-nitro-l-arginine methyl ester (l-NAME), the inhibitor of nitric oxide (NO) synthase, or charybdotoxin, the blocker of Ca2+-activated K+ (KCa) channels, partially attenuated the infarct size–limiting effect, and both of them completely abolished the effect. The incidence of ventricular fibrillation was also less in the raloxifene group than in the control group (11% vs. 44%, p < 0.05). Activity of p38 mitogen-activated protein (MAP) kinase increased with 15-min ischemia, and raloxifene pretreatment inhibited the activity. Myeloperoxidase activity of the 6-h reperfused myocardium was also attenuated by raloxifene.
Conclusions These data demonstrate that raloxifene reduces myocardial ischemia-reperfusion injury by mechanisms dependent on NO and the opening of KCa channels in canine hearts. Deactivation of p38 MAP kinase and myeloperoxidase by raloxifene may be involved in the cellular mechanisms of cardioprotection.
Estrogen replacement therapy may reduce cardiovascular events in postmenopausal women (1). We have previously demonstrated that estradiol-17-beta reduced the myocardial infarct (MI) size and the occurrence of ischemia- and reperfusion-induced ventricular arrhythmias (2) and that it increased coronary blood flow in canine ischemic hearts (3). These effects are mediated by both estrogen-induced enhancement of nitric oxide (NO) production and the probability of open Ca2+-activated K+ (KCa) channels. Despite these beneficial effects, estrogen replacement therapy appears to increase the risk of endometrial or breast cancer and may be associated with an increased risk of venous thromboembolism (4,5). Raloxifene, one of the selective estrogen receptor modulators, exerts estrogenic agonistic or antagonistic actions on different tissues and has been recently introduced as a drug for new hormone replacement therapy because of the reduced adverse effects of estrogen. It prevents bone loss and has lipid-lowering effects in humans (6–10). Although it also exhibits coronary-relaxing effects on the coronary artery in vitro (11), it remains unclear whether raloxifene has beneficial effects during ischemia and reperfusion in vivo. If raloxifene does not mediate cardioprotection, as estradiol-17-beta does, the rationale for the use of raloxifene as hormone replacement therapy may be weakened.
Therefore, we investigated the effects of raloxifene on MI size and the occurrence of ischemia- and reperfusion-induced ventricular arrhythmias. We infused raloxifene into the coronary artery of open-chest dogs and also examined whether the effects are involved in NO or the opening of KCa channels. Furthermore, we also examined the activities of three major mitogen-activated protein (MAP) kinases—p38 MAP kinase, c-Jun NH2-terminal protein kinase (JNK) and extracellular signal–regulated protein kinase (ERK)—during ischemia, because these MAP kinases are known to be the mechanism of cellular signaling for cardiac injury and protection (12–16).
Seventy-four beagle dogs weighing 8 to 12 kg were anesthetized with intravenous sodium pentobarbital (30 mg/kg), intubated with a cuffed endotracheal tube and ventilated with room air mixed with oxygen (1.5 l/min) with the use of a respirator. Intravenous infusion of sodium pentobarbital (approximately 30 mg/h) was continued during the experimental protocol to maintain hemodynamic stabilization. The method of preparation for the experiments of ischemia-reperfusion injury in the canine heart was previously described (2). Briefly, the left anterior descending coronary artery (LAD) was cannulated and perfused with blood through an extracorporeal tube from the left carotid artery. The femoral artery and left atrium were cannulated to obtain a reference blood flow sample for calculation of the regional myocardial blood flow and to inject microspheres, respectively. Heparin (500 U/kg) was administered intravenously every 3 h throughout the protocol. All studies conformed with the position of the National Institutes of Health (NIH), in its Guide for the Care and Use of Laboratory Animals, adopted in November 1984.
Protocol 1: effects of raloxifene on infarct size and ischemia- and reperfusion-induced ventricular arrhythmias
After hemodynamic stabilization, infusion of raloxifene (0.5 or 5 μg/kg per min; Eli Lilly, Indianapolis, Indiana), vehicle (0.1% [vol/vol] dimethyl sulfoxide) or saline (control group) was initiated into the bypass tube 10 min before coronary occlusion and continued until 1 h after reperfusion without the occlusion period (control group: n = 9; vehicle group: n = 9; raloxifene groups: 0.5 μg/kg per min, n = 6 and 5 μg/kg per min, n = 9). The intracoronary infusion of 5 μg/kg per min of raloxifene theoretically corresponded to 1 μmol/l in the coronary artery blood, which was shown in a previous study to maximally relax the coronary arteries (11), and 0.5 μg/kg per min of raloxifene was also infused to examine the dose-dependent effects of raloxifene.
In 10-min infusion, the coronary artery was occluded for 90 min and then reperfused for 6 h. Hemodynamic variables were measured before the administration of drugs and sustained ischemia at 10 and 90 min after the onset of ischemia and at 1, 3, and 6 h after the onset of reperfusion.
Protocol 2: role of NO and Kca channels in raloxifene-induced attenuation of infarct size and incidence of ischemia- and reperfusion-induced ventricular arrhythmias
The effects of raloxifene were examined in dogs pretreated with NG-nitro-l-arginine methyl ester (l-NAME, an inhibitor of NO synthase) or charybdotoxin (CTX, a blocker of KCa channels). Infusion of either l-NAME (10 μg/kg per min; Sigma) or CTX (1 μg/kg per min; Sigma, St. Louis, Missouri) into the bypass tube was initiated 10 min before infusion of raloxifene (5 μg/kg per min) and continued until 1 h of reperfusion after the 90-min occlusion period (raloxifene + l-NAME group, n = 7; raloxifene + CTX group, n = 7; raloxifene + l-NAME + CTX group, n = 7). To examine the effects of inhibitors of ischemia-reperfusion injury, we also infused only l-NAME and CTX (n = 8) into the bypass tube during the same period as described earlier. The dose of l-NAME completely abolished the release of NO (17), and the dose of CTX also maximally attenuated bradykinin-induced coronary vasodilation (18). Hemodynamic variables were measured at the same time as in protocol 1.
Protocol 3: effects of raloxifene on MAP kinase activity during ischemia
We used 12 dogs in this protocol and classified them into four groups: no ischemia, ischemia, 10-min pretreatment with raloxifene before ischemia and 10-min pretreatment with raloxifene, l-NAME and CTX before ischemia. In the group with no ischemia, the LAD of three dogs was cannulated and perfused with blood through a bypass tube for 25 min. In the group with ischemia, the tube was occluded for 15 min after 10 min of hemodynamic stabilization, and in the group with raloxifene pretreatment or raloxifene plus inhibitors pretreatment before ischemia, 10-min infusion of raloxifene with or without l-NAME and CTX into the LAD was performed before 15 min of complete coronary artery occlusion. After these procedures were done, we sacrificed the dogs, excised the hearts and quickly sampled myocardial tissue supplied by the LAD into liquid nitrogen and stored −80°C.
Criteria for exclusion
To ensure that all of the animals included in the analysis of infarct size data were healthy and exposed to similar degrees of ischemia, we adopted the following criteria for the exclusion of unsatisfactory dogs: 1) subendocardial collateral flow >15 ml/100 g per min; 2) heart rate >170 beats/min; or 3) more than two consecutive attempts required to correct ventricular fibrillation (VF) with low-energy, direct-current pulses applied directly to the heart. We calculated the survival percentage as the number of dogs that survived/number of assigned dogs × 100 .
Measurement of infarct size and regional myocardial blood flow
We measured infarct size as previously described (2). Briefly, after 6 h of reperfusion, we re-occluded the LAD and injected Evans’ blue dye into a systemic vein. Then, the sliced LAD was incubated in 1% 2,3,5-Triphenyltetrazolium chloride (Sigma) solution to detect the infarct zone. Infarct size was expressed as the percentage of the infarct zone that is contiguous with area at risk. Regional myocardial blood flow was determined by the microsphere technique (19).
Measurement of activity of MAP kinases
For the measurement of activity of MAP kinases, 0.5 g myocardial tissue was homogenized, and an in vitro kinase assay was carried out. Extracts from tissue homogenization were subjected to immunoprecipitation with anti-p38 MAP kinase antibody bound to protein A/agarose (Amersham Pharmacia Biotech, Piscataway, New Jersey), c-Jun fusion protein beads (Cell Signaling Technology, Beverly, Massachusetts), anti-ERK 1/2 antibody bound to protein A/agarose (Upstate Biotechnology, Waltham, Massachusetts). The immunoprecipitates were collected and washed extensively. Immunoprecipitates of anti-p38 MAP kinase and anti-ERK 1/2 antibody were mixed with 2 mg/ml dephosphorylated MAP kinase-2 activated protein and myelin basic protein (MBP), respectively, as a substrate, and 200 mmol/l adenosine triphosphate (ATP) for kinase assay. Pellets precipitated with glutathion-S-transferase/c-Jun/sepharose were mixed with 100 mmol/l ATP. Then, the reaction mixtures were further incubated for 30 min at 30°C. The kinase reaction was terminated by boiling in an appropriate volume of sodium dodecyl sulfate (SDS) sample buffer. The reaction components were separated by SDS/polyacrylamide gel electrophoresis and transferred to a nitrocellulose membrane. The phosphorylated MAPKAP kinase-2, c-Jun and MBP were detected by anti-phospho-MAPKAP kinase-2 (Upstate Biotechnology), antiphosphorylated c-Jun (Cell Signaling Technology), and antiphosphorylated MBP (Upstate Biotechnology) antibodies, respectively. The total content of p38 MAP kinase, c-Jun, or ERK in the samples was also determined by immunoblot analysis using anti-p38 MAP kinase, c-Jun, and ERK 1/2 antibodies. The band density was analyzed by NIH image software.
Measurement of myocardial myeloperoxidase (MPO) activity
Several myocardial tissue samples were taken from ischemic areas or nonischemic areas in each dog, frozen in liquid nitrogen and stored at −45°C until assay. The technical procedure was previously described (20). One unit of MPO activity was defined as that which degraded 1 mol hydrogen peroxide per minute at 25°C.
Data are expressed as the mean value ± SE. Statistical significance was performed with analysis of variance, and if differences were revealed among groups, they were assessed using the Fisher-Irwin test. A value of p < 0.05 was considered statistically significant.
Assignments and exclusions
The number of dogs assigned and excluded in each group is shown in Table 1.
Effects of raloxifene on infarct size
Heart rate and mean arterial blood pressure remained stable among the experimental groups throughout the study (Fig. 1). Both coronary blood flow and coronary perfusion pressure in all experimental groups did not significantly change during raloxifene infusion, although coronary blood flow in the raloxifene (5 μg/kg per min) group tended to increase at the end of raloxifene treatment, but it was not statistically significant, and the area at risk and collateral blood flow were also similar among all groups (Table 2).
The infarct size in the control group was 40.9 ± 3.9% of the area at risk, and vehicle did not have an infarct size–limiting effect (infarct size 37.4 ± 4.8%). Raloxifene (5 μg/kg per min) significantly reduced the infarct size compared with that of the control group (infarct size 7.2 ± 2.5%, p < 0.01), whereas a low dose of raloxifene (0.5 μg/kg per min) did not mediate the effect (infarct size 38.3 ± 2.0%). Figure 2 revealed that the infarct size–limiting effect was attenuated by either l-NAME (infarct size 21.4 ± 4.4%, p < 0.05 vs. control or raloxifene [5 μg/kg per min]) or CTX (infarct size 20.5 ± 5.0%, p < 0.05 vs. control or raloxifene [5 μg/kg per min]) and completely abolished by l-NAME + CTX (infarct size 37.7 ± 5.8%, p < 0.05 vs. l-NAME or CTX and p < 0.01 vs. raloxifene [5 μg/kg per min]). Only administration of the inhibitors l-NAME and CTX did not exaggerate the infarct size (40.5 ± 4.1%) compared with that of the control group.
Effects of raloxifene on VF during the reperfusion period
Pretreatment of dogs with raloxifene (5 μg/kg per min) significantly reduced the overall incidence of VF during the 6-h reperfusion period. This effect was abolished by l-NAME and CTX administered alone or together with raloxifene. At a dose of 0.5 μg/kg per min, raloxifene did not decrease the incidence of VF (Table 3).
Effects of raloxifene on MAP kinase activity during ischemia
Figure 3A shows the three major MAP kinase activities in each group, and Figure 3B demonstrates the representative results. Activities of p38 MAP kinase, JNK, and ERK in the group with ischemia were all significantly increased by 8.1 ± 1.2%, 10.4 ± 3.8% and 11.7 ± 4.9%, respectively, compared with the group with no ischemia. Activities of JNK and ERK were not different between the groups with ischemia and raloxifene pretreatment before ischemia, but only p38 activity was significantly reduced in the group with raloxifene pretreatment before ischemia, compared with the group with ischemia. Attenuation of p38 MAP kinase activity by pretreatment with raloxifene in the ischemic state was abolished by co-administration of l-NAME and CTX with raloxifene.
Myocardial MPO activity
Data on MPO activity for the ischemic and nonischemic regions of all experimental groups are shown in Table 4. These results showed that MPO activity significantly decreased only in the raloxifene (5 μg/kg per min) group, compared with the control group, suggesting that raloxifene reduced myocardial ischemia-reperfusion injury by attenuation of MPO activity.
We have shown that raloxifene reduces the MI size and the incidence of ischemia- and reperfusion-induced VF in dogs and that these cardioprotective effects are blocked by either inhibition of NO synthase or antagonism of KCa channels. Moreover, only treatment with l-NAME and CTX does not exacerbate the infarct size and incidence of VF, compared with the control group. These results suggest that augmentation of endogenous NO release and opening of KCa channels induced by raloxifene synergistically contribute to the alleviation of irreversible ischemia-reperfusion injury. Furthermore, p38 MAP kinase activity in myocardial ischemia was reduced by raloxifene, suggesting that inhibition of p38 MAP kinase–signaling pathways is cardioprotective, as previously described (13,14), and this may partly explain the effects of raloxifene on ischemia-reperfusion injury.
Mechanisms of cardioprotective effects against ischemia- and reperfusion-induced injury mediated by NO and opening of Kca channels
The present results demonstrate that raloxifene attenuated infarct size and incidence of VF during reperfusion by cardioprotective mechanisms mediated by both NO and the opening of KCa channels. Potentiation of NO release may be an effective pharmacologic intervention to limit MI size, given that the administration of an NO donor markedly attenuates ischemia-reperfusion injury (21,22). Several possible mechanisms may underlie the beneficial effects of NO on infarct size. First, because NO regulates the membrane Ca2+ current of cardiac cells (23), it may reduce the severity of ischemia by inhibiting the cytosolic accumulation of Ca2+. Second, NO may also reduce oxygen-derived free radical generation by decreasing lipolysis, thereby limiting the generation of radicals through lipid peroxidation (24). It is not clear how raloxifene increases the release of NO in the reperfused heart, but a previous report demonstrated that raloxifene caused an immediate and marked production and release of NO in human endothelial cells (25). Increases in the cardiac NO level can also provoke protective effects against VF during reperfusion.
Opening of KCa channels may hyperpolarize the cellular membrane and reduce Ca2+ overload during ischemia and reperfusion, similar to the protective effect of ATP-sensitive K+ channels (26), in addition to its coronary vasodilatory effect (17). The precise mechanism by which the opening of KCa channels reduces VF during reperfusion is unknown. Although opening of KCa channels may reduce the duration of action potentials during coronary occlusion, resulting in an arrhythmogenic effect, this opening reduces Ca2+ overload in cardiomyocytes as a result of a hyperpolarized membrane potential.
Role of p38 MAP kinases in infarct size–limiting and anti-arrhythmic effects
We have shown that three major MAP kinases, including p38 MAP kinase, JNK, and ERK, were activated by ischemia and that raloxifene reduced only p38 MAP kinase activity, but not JNK and ERK, in canine hearts. Activation of p38 MAP kinase, JNK, and ERK in cardiomyocytes subjected to ischemia was previously reported (15,16), but the present study is the first to demonstrate an increase of MAP kinase activity in ischemic canine hearts in vivo. Activation of ERK may be part of a “survival pathway,” whereas activation of p38 MAP kinase and JNK represents a “death pathway” in the heart (27). In our study, increased activity of ERK was equal to that of JNK and less than that of p38 MAP kinase (Fig. 3). It is possible that the beneficial effects of ERK against ischemia might be abrogated by activation of both p38 MAP kinase and JNK in the ischemic heart. Some reports demonstrated that inhibition of p38 MAP kinase was effective for cardioprotection (13,14,28). In our study, p38 MAP kinase activation was attenuated by raloxifene during ischemia, and it may be reasonable that raloxifene has MI size–limiting and anti-arrhythmic effects by inhibition of the p38 MAP kinase–signaling pathway.
p38 MAP kinase regulates transcription factors, which respond to transcriptional changes in nuclei. In this study, there was a rapid and pronounced effect of raloxifene against ischemia-reperfusion injury, suggesting that raloxifene develops its effect by non-genomic means. p38 MAP kinase mediates myocardial ischemia-reperfusion–induced cytokines and free radical production (29,30), neutrophil activation (31), and platelet aggregation (32). It is possible to explain that the infarct size–limiting and anti-arrhythmic effects induced by raloxifene in our early-phase experiments are partly due to a reduction of ischemia-induced neutrophil activation accompanied by p38 MAP kinase inhibition, because the increased myocardial MPO activity of the ischemic area was actually diminished to the level of the nonischemic area by raloxifene infusion (5 μg/kg per min) (Table 4). Although the relationship between p38 MAP kinase, KCa channels, and NO is unclear, co-administration of l-NAME and CTX with raloxifene abolished the attenuation of p38 MAP kinase activity by treatment with raloxifene in our present study (Fig. 3), suggesting that p38 MAP kinase activity may be inhibited by NO and opening of KCa channels under ischemic stress. In contrast, we could not deny the possible mechanism that inhibition of p38 MAP kinase activity by raloxifene in ischemia may result in both an increased NO level and opening of KCa channels, because inhibition of p38 MAP kinase activity reduces the production of free radicals during ischemia and reperfusion, subsequently increasing the impaired NO synthase activity (29). Nitric oxide, opening of KCa channels, and inhibition of p38 MAP kinase possibly contribute to the mechanisms of cardioprotection against ischemia and reperfusion.
To the best of our knowledge, this is the first report to demonstrate the MI size–limiting and anti-arrhythmic effects of raloxifene. In another report (33), idoxifene, one of the selective estrogen modulators, like raloxifene, had a beneficial effect on vascular remodeling in the balloon denudation rat model. Together with this action, raloxifene could be a good therapeutic drug for ischemic heart diseases. Moreover, raloxifene would be expected to be more advantageous and selective for the cardiovascular system, compared with estrogen, and it is possible to use raloxifene not only in postmenopausal women but also in premenopausal women with coronary heart diseases.
We thank Hideki Koyama, Akiko Ogai, Tomi Fukushima, and Junko Yamada for their technical assistance and advice.
☆ This study was partially supported by Comprehensive Research on Aging and Health, Ministry of Health and Welfare, Grants-in-Aid for Scientific Research (nos. 12470153 and 12877107) for the Ministry of Education, Culture, Sports, Science and Technology and by the Smoking Research Foundation in Japan.
- extracellular signal–regulated protein kinase
- c-Jun NH2-terminal protein kinase
- Ca2+-activated K+
- left anterior descending coronary artery
- NG-nitro-l-arginine methyl ester
- mitogen-activated protein
- myelin basic protein
- myocardial infarct
- nitric oxide
- Received April 30, 2001.
- Revision received April 29, 2002.
- Accepted May 24, 2002.
- American College of Cardiology Foundation
- Node K.,
- Kitakaze M.,
- Kosaka H.,
- Minamino T.,
- Funaya H.,
- Hori M.
- Node K.,
- Kitakaze M.,
- Kosaka H.,
- et al.
- Douketis J.D.,
- Gordon M.,
- Johnston M.,
- Julian J.A.,
- Adachi J.R.,
- Ginsberg J.S.
- Heaney R.P.,
- Draper M.W.
- the Multiple Outcomes of Raloxifene Evaluation (MORE) Investigators,
- Ettinger B.,
- Black D.M.,
- Mitlak B.H.,
- et al.
- Figtree G.A.,
- Lu Y.,
- Webb C.M.,
- Collins P.
- Ma X.L.,
- Kumar S.,
- Gao F.,
- et al.
- Mackay K.,
- Mochly-Rosen D.
- Yue T.L.,
- Wang C.,
- Gu J.L.,
- et al.
- Node K.,
- Kitakaze M.,
- Kosaka H.,
- Minamino T.,
- Hori M.
- Mori H.,
- Haruyama S.,
- Shinozaki Y.,
- et al.
- Ma X.L.,
- Weyrich A.S.,
- Lefer D.J.,
- Lefer A.M.
- Lefer D.J.,
- Nakanishi K.,
- Johnston W.E.,
- Vinten-Johansen J.
- Mery P.F.,
- Pavoine C.Belhassen,
- Pecker F.,
- Fishmeister R.
- Gaudiot N.,
- Ribiere C.,
- Jaubert A.M.,
- Giudicelli Y.
- Gross G.J.,
- Auchampach J.A.
- Abe J.,
- Baines C.P.,
- Berk B.C.
- Sanada S.,
- Kitakaze M.,
- Papst P.J.,
- et al.
- Saklatvala J.,
- Rawlinson L.,
- Waller R.J.,
- et al.
- Yue T.L.,
- Vickery-Clark L.,
- Louden C.S.,
- et al.