The impact of coronary spasm on the onset of acute ischemic coronary events remains controversial. Several previous studies argued that coronary spasm was directly related to sudden cardiac death [1-4]. However, others recently indicated that the role of vasospasm in myocardial infarction remains unclear [5,6]. Thus, the relationship between coronary spasm and acute ischemic coronary events as well as that coronary spasm and the degree of atherosclerotic lesions have not been elucidated in detail. Previous studies demonstrated that the degree of coronary plaques on angiography was a predictor of vasospasm [7] and intravascular ultrasound studies revealed that atherosclerosis was present at sites with coronary spasm even in the absence of angiographically significant coronary artery disease [8]. However, vasospasm was also shown to occur at the site of angiographically normal coronary segments or in the presence of a nonobstractive atherosclerotic plaque [9]. Since difficulties are associated with determining the relationship between coronary plaques and vasospasm as well as that between coronary spasm and angina pectoris in clinical studies because of ethical problems, suitable animal models needed to be developed to more clearly understand coronary spasm and angina pectoris.

Several animals have been used as models in the study of coronary spasm. In ex vivo examinations, aortas and coronary arteries exposed to endogenous hyperlipidemia appeared to be hyper reactive to ergonovine [10]. Constrictive responses induced by constrictors were significantly augmented in a swine model fed a cholesterol diet, as was denudation of endothelium [11-13]. Although these findings contributed to clarifying the mechanisms involved in the provocation of vasospasm, the denudation of arterial endothelial layers does not occur physiological condition. Previous studies [13-15] examined the relationship between coronary spasm and the onset of acute coronary syndromes, but they did not use clinical findings. Lanza et al [16] reported that the cause of and mechanism responsible for coronary spasm were still poorly defined 60 years after the first report of variant angina despite of recent advances.

We developed a myocardial infarction-prone strain of the Watanabe heritable hyperlipidemic (WHHLMI) rabbit [17,18] by selectively breeding the WHHL rabbit [18,19], which has hypercholesterolemia because of a genetic defect in low-density lipoprotein receptors. We previously provoked coronary spasm in WHHLMI rabbits and demonstrated the associations between of coronary spasm and plaque disruption as well as ischemic myocardial damage [20]. However, we did not sufficiently examine the effects of coronary plaques on the provocation of coronary spasm or development of angina pectoris like symptoms, e.g. ventricular dysfunction and ischemic injury of myocardium.

We here in reanalyzed the data of our study [20] with the addition of data from four extra rabbits and ex vivo coronary contraction experiments to determine whether the severity of coronary plaques affected provocation of coronary spasm and acute ischemic myocardial events in WHHLMI rabbits.

Animals

We used 34 WHHLMI rabbits aged 12-29 months in experiments on the provocation of coronary spasm, and 15 male WHHLMI rabbits aged 8-15 months in the coronary contraction test. WHHLMI rabbits were bred at the Kobe University Graduate School of Medicine. As a control, 7 male Japanese White (JW) rabbits (Kitayama Labes, Co. Ltd., Ina, Japan) aged 8 months were used in the coronary spasm provocation test and examination of left ventricular function, while 15 age-matched male JW rabbits were used in the coronary contraction test. Rabbits resided individually in metal cages (550 mm wide, 600 mm deep and 450 mm high) with a flat floor, and consumed standard rabbit chow (LRC4, Oriental Yeast Co., Ltd., Tokyo, Japan) at 120 g/day and water ad libitum. The animal rooms were maintained under a constant temperature (22 ± 2°C), relative humidity (50-60%), ventilation rate (15 cycles/hour), and lighting cycle (12 hours light/dark). This study was approved by the Kobe University Animal Care and Use Committee (approval numbers: P080606, P091101), and animal experiments were conducted in accordance with the Regulations for Animal Experimentation of Kobe University, and Japanese regulations, such as the Act on Welfare and Management of Animals (Law No. 105; 1973, revised 2006), Standards Relating to the Care and Management of Laboratory Animals and Relief of Pain (Notification No. 88, 2006), and Fundamental Guidelines for Proper Conduct of Animal Experiment and Related Activities in Academic Research Institutions (Notice No.71, 2006).

Anesthesia and euthanasia

Rabbits were anesthetized with an intravenous injection of ketamine hydrochloride (15 mg/kg, Daiichi-Sankyo Co. Ltd., Tokyo, Japan) plus midazolam (1 mg/kg, Dormicum, Astellas Pharma Inc., Tokyo, Japan), and anesthesia was continued by the infusion of ketamine hydrochloride at 60 mg/kg/h. During experiments, oxygen was supplied through a face mask (2.0 L/min for rabbits), and rabbits were warmed with a heating pad. Rabbits were euthanized with exsanguination under the intravenous administration of sodium pentobarbital (30 mg/kg).

Isometric tension measurement of coronary arterial strips

The Left Anterior Descending artery (LAD) and Left Circumflex Arteries (LCX) were immediately isolated after euthanasia, cleaned of surrounding tissue, and cut into 2-mm-wide and 15-mm-long helical strips. The endothelium was denuded by rubbing the vessels with wet filter paper. Isometric tension was measured in the coronary strips as previously described [10,21]. Briefly, coronary strips were suspended in organ baths containing Krebs buffer, and an initial preload of 0.5 g was applied. After 2 hours for equilibration, test contractions were induced by adding 20 mmol/L KCl. When the developed tension attained its peak value, strips were relaxed by rinsing with buffer. Constriction responses were evaluated as the percentage of maximum tension induced by 20 mmol/L KCl. Concentration-response relationships were determined by the cumulative addition of norepinephrine and ergonovine (all from Sigma-Aldrich, St. Louis, MO). Furthermore, coronary strips were stimulated with cumulative doses of ergonovine after the tone had been moderately increased by a threshold concentration (~EC₂₀) of norepinephrine (1 μM).

Provocation of coronary spasm

The provocation and evaluation of coronary spasm were performed as previously described [20], and the study design was shown in Figure 1.

Figure 1: Study design for the provocation of coronary spasm. Coronary spasm was provoked with an intravenous injection of ergonovine maleate (Tokyo Chemical Industry Co., Ltd. Tokyo, Japan) at a dose of 0.45 μmol/kg during the infusion of norepinephrine (Daiichi-Sankyo Co., Ltd., Tokyo, Japan) at a rate of 12 nmol/kg/min through a marginal ear vein. Nitroglycerin (Hikari Pharmaceutical Co., Ltd., Tokyo, Japan) was injected intravenously (10 μg/kg) 20–30 minutes after the onset of ischemic changes on electrocardiogram (ECG) to reverse coronary spasm. The occurrence of coronary spasm was monitored with coronary angiograms using an X-ray apparatus (OPESCOPE PLENO; Shimadzu Corporation, Kyoto, Japan), and ECG were measured using an amplifier (AB-621G; Nihon Kohden, Tokyo, Japan) and PowerLab/8SP (ADInstruments Pty Ltd, Bella Vista, Australia), or an electrocardiograph (CARDIOSUNY α6000 AX-D, Fukuda M-E Kogyo, Co., Ltd., Tokyo, Japan) . Blood pressure was also monitored invasively at the femoral artery.

Evaluation of ventricular contractile dysfunction and myocardial ischemia

Echocardiograms were performed using 13 WHHLMI rabbits and 7 JW rabbits with a Philips Envisor C echocardiograph (Philips Inc., Eindhoven and the Netherlands) [20]. Left ventricular function was evaluated by fractional shortening, which was calculated as 1 – Systolic Left Ventricular Diameter (LVDs)/Diastolic Left Ventricular Diameter (LVDd). Serum biomarkers for ischemic myocardial damage (heart-type fatty acid-binding protein [H-FABP], cardiac troponin-I [cTroponin-I], and myoglobin) were assayed with ELISA kits (Life Diagnostics Inc., West Chester, PA, USA) before the injection of vasoconstrictors, and 4 hours after ischemic changes occurred on ECG using 20 WHHLMI rabbits.

Preparation of coronary sections

Rabbits were euthanized after the examination. Hearts were excised and immersion-fixed with a 10% neutral buffered formalin solution, and embedded in paraffin. Coronary arterial segments were prepared as reported previously [22]. Sections were stained with elastic van Gieson stain. Coronary stenosis was evaluated as cross-sectional narrowing (%) calculated by dividing the lumen area by the area surrounded with an internal elastic lamina.

Assay of serum lipid levels

Serum total cholesterol and triglyceride levels were assayed enzymatically with kits at 12 months of age using sera obtained after 15 hours of fasting.

Statistical analysis

Data are presented as the mean ± the Standard Error of the Mean (SEM). Statistical analyses were performed for mean values with the signed Wilcoxon test, Mann-Whitney U-test, or Student’s t-test; and for frequency with the chi-square test. To compare of mean values among multiple groups, we performed the Bonferroni test. A value of P<0.05 was taken as significant.

Baseline data of WHHLMI rabbits

As shown in Table 1, all WHHLMI rabbits had hypercholesterolemia and each coronary artery had atherosclerotic plaques. Coronary stenosis (evaluated as cross-sectional narrowing) of more than 75% was observed in 23 rabbits (67.6%).

Table 1: Baseline data of WHHLMI rabbits provoked with coronary spasm

Contractile responses in coronary helical strips

Norepinephrine and ergonovine induced contractions in control JW and WHHLMI coronary strips in a dose-dependent manner (Figure 2). There is no significant difference between WHHLMI and JW rabbits in maximum tension induced by 20 mmol/L KCl (data not shown) as similar to a previous study [9]. No significant differences were observed in the contraction by norepinephrine (triangles) between WHHLMI and control strips. The concentration-response curves for ergonovine after contractions with norepinephrine in WHHLMI strips (closed circles) exhibited a left and upward shift from the control (open circles). The E_max for combined treatment was 174 ± 15% in WHHLMI coronary strip, and was 89 ± 8% in the normal coronary strips. These results indicated that atherosclerotic WHHLMI coronary strips were hypersensitive and hyper reactive to these agents relative to normal coronary strips.

Figure 2: Contraction-response of the endothelium-denuded helical coronary strips of rabbits treated with norepinephrine and ergonovine. The abscissa shows the final concentration of the agonists, and the ordinate shows the percentage of maximal tension induced by 20 mmol/L KCL. Coronary strips were treated with norepinephrine (Nor, ▲ or Δ), ergonovine (Erg, ■ or □), and 1.0 μM norepinephrine plus ergonovine (Nor + Erg, ● or ○). Each point represents the mean ± SEM (n=5). Statistical analyses were performed with the Student’s t-test. Closed symbols (▲,■,●) represent data from WHHLMI rabbits (W) and open symbols (Δ,□, ○) represent data from normal JW rabbits (J). Bars indicate the SEM.

Occurrence of coronary spasm

During the vasospasm provocation test, ECG showed ischemic changes in WHHLMI rabbits (Figure 3), e.g. ST depression/ elevation, T-wave elevation/inversion, poor R-wave progression, deep Q-wave, and ventricular premature complex. The frequency of these abnormalities was markedly higher in WHHLMI rabbits (27/34, 79.4%, P=0.007) than in normal JW rabbits (2/7, 28.6%). We confirmed development of coronary spasm with coronary angiography. Figure 4 shows the provocation of angiographical coronary spasm in WHHLMI rabbits in vivo. Contrast flow was markedly decreased after the injection of ergonovine (Figure 4A), and the area with decreased perfusion in LCX corresponded to segments with plaques (sections 7-16) (Figure 4B). Regarding the relationship between coronary plaques and the development of coronary spasm (Figure 4C), the frequency of coronary spasm was proportional to the degree of coronary stenosis (P=0.01) and the number of coronary segments with lesions (P=0.024). Since coronary stenosis with >75% cross-sectional narrowing is often used as the level of significant stenosis in clinical analysis, we examined the relationship between significant stenosis and vasospasm provocation. The frequency of coronary segments with more than 75% stenosis was significantly high in the vasospasm-positive WHHLMI rabbits (32.3 ± 5.9%, P=0.036) than in vasospasm-negative WHHLMI rabbits (8.2 ± 6.0%). These results demonstrated that atherosclerotic coronary arteries were highly susceptible to the provocation of vasospasm.

Figure 3: The electrocardiograms (ECG) of rabbits treated with ergonovine bolus injection under norepinephrine infusion. (A) Representative ECG changes in a WHHLMI rabbit during the experiments. (B) Frequency of ischemic patterns on ECG in normal rabbits (n=7, open bars) and WHHLMI rabbits (n=34, solid bars) treated with an ergonovine bolus injection under the infusion of norepinephrine. Statistical analyses were performed with the chi-square test

Figure 4: Relationship between coronary plaques and coronary spasm in WHHLMI rabbits. (A) Coronary angiograms at the baseline (left), after the ergonovine injection (center), and after a nitroglycerine injection (right). (B) Coronary stenosis evaluated as cross-sectional narrowing of a left circumflex artery as indicated on the angiogram. (C) Relationship between coronary plaques and the provocation of coronary spasm. Data are represented as the mean ± SEM. Statistical tests were performed with the chi-square test for the frequency and with the Student’s t-test for the mean value. LAD, left anterior descending artery, LCX, left circumflex artery.

Changes in the motility of the left ventricular wall

Systolic blood pressure was significantly increased with the spasmogen treatments, but was significantly decreased at the end of the experiment (96 ± 6.2 mmHg, P=0.018) than the baseline (131 ± 4.6 mmHg). Since these results may suggest cardiac dysfunction due to coronary spasm, we further analyzed cardiac functions by echocardiogram. Table 2 shows the results of echocardiograms during the vasospasm provocation test. Although fractional shortening was already significantly lower at the baseline in WHHLMI rabbits (32.5 ± 1.2, P=0.004) than in normal JW rabbits (41.5 ± 3.1), it was further decreased after the spasmogen treatments (23.0 ± 0.6, P<0.001). These results indicated that coronary spasm lead to cardiac dysfunction in WHHLMI rabbits.

Table 2: Left ventricular wall motility in WHHLMI rabbits with coronary spasm

Evaluation of ischemic damage to the left ventricular wall

Table 3 shows changes in serum markers for ischemic myocardial injury in WHHLMI rabbits. These serum markers were within human normal ranges at the baseline. However, they were markedly increased after the provocation of vasospasm, which confirmed that coronary spasm induced ischemic injury in the myocardium.

Table 3: Changes in serum markers for ischemic myocardial damage in WHHLMI rabbits with coronary spasm (n=20)

These results of the present study demonstrated that coronary spasm was induced frequently in WHHLMI rabbits with diffuse atherosclerotic plaques showing more than 75% coronary stenosis by pharmacological treatments, and coronary spasm caused cardiac dysfunction and ischemic injury to the myocardium. These results suggest that angina pectoris and/or non-fatal myocardial infarction developed due to the provocation of coronary spasm.

In the present study, the frequency of coronary spasm in WHHLMI rabbits with severe coronary stenosis and diffuse atherosclerotic plaques was significantly higher than that in WHHLMI rabbits with less stenosis and focal plaques in spite of in serum lipid levels being similar (data not shown). Previous studies reported a reduction in arterial endothelial function in atherosclerotic arteries [23,24]; therefore, the present results suggest that a dysfunction in the arterial endothelium of atherosclerotic plaques may affect the constriction of coronary arteries. In addition, ex vivo studies using endothelial-denudated coronary strips demonstrated the hypersensitivity and hyperreactivity of atherosclerotic strips. These results are consistent with previous findings [11,12,21]. Therefore, we speculated that the enhanced contractile response observed in atherosclerotic coronary strips may primarily be attributed to the hyperreactivity of medial smooth muscle cells. Maseri et al [25] demonstrated that the response to constrictor stimuli also extended to the adjacent segments and coronary spasm can be multi-focal and/or diffuse. Miyao et al [26] and Tsujita et al [27] also showed that coronary spasm was not a localized, but a generalized disorder that involved the entire coronary arteries and was associated with intimal thickening. These findings were consistent with the angiographical results in the present study; therefore, coronary spasm in WHHLMI rabbits is, from a pathophysiological aspect, relevant to that in humans.

After the provocation of coronary spasm, the left ventricle motility in rabbits with coronary spasm evaluated with echocardiograms was decreased, and serum markers for ischemic myocardial injury were markedly increased. These results suggested the onset of angina pectoris-like symptoms and/or non-fatal myocardial infarction, although the frequency of occlusive thrombi after the disruption of coronary plaques was very low [20]. Previous studies asserted that coronary spasm plays an important role in the pathogenesis of not only variant angina, but ischemic coronary disease, including other forms of angina, acute myocardial infarction, arrhythmias, and ischemic sudden death [28-31]. These findings were consistent with the results of the present study; therefore, we speculated that the WHHLMI rabbit is useful as an animal model of experimentally provoked coronary spasm and subsequent myocardial ischemia.

The present results suggest that atherosclerosis in coronary arteries contributes to the development of coronary spasm, and coronary spasm in severe atherosclerotic lesions results in acute ischemic myocardial damage, which is similar to angina pectoris in humans. The WHHLMI rabbit is a good animal model for coronary spasm and related cardiac dysfunctions.

Nothing to declare.

This work was supported, in part, by a grant for Research on Biological Resources and Animal Models for Drug Development from the Ministry of Health, Labor, and Welfare of Japan, by Grants-in- Aid for scientific research from the Ministry of Education, Culture, Sports, and Technology, Japan 23300157), and by a research grant from Daiichi-Sankyo Co. Ltd., Tokyo, Japan.

We would like to thank Dr. Norihisa Nitta, and Dr. Akinaga Sonoda, Shiga University of Medical Science, Japan, for coronary angiography on rabbits.