INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

CORONARY VASOSPASM IS believed to play an important role in angina pectoris, as well as in acute myocardial infarction and sudden cardiac death (15, 16, 33). Although the pathogenesis of coronary vasospasm is not well understood, the primary mechanism appears to involve an altered reactivity of the coronary arteries. This altered reactivity results from enhancement of vasoconstricting responses of vascular smooth muscle to vasoactive stimuli (2, 15) and impairment of endothelium-dependent vasodilation (12, 31).

Platelets can cause blood vessels to contract by releasing serotonin and thromboxane A₂. Platelet function plays an important role in various ischemic heart diseases, including vasospastic angina. The platelets are activated during vasospastic anginal attacks (25), and platelet aggregation induces hyperconstriction at the site of vasospasm (28). Although intracellular Ca²⁺ is a major determinant of cellular function, the handling of Ca²⁺ in the platelets of patients with vasospastic angina is not well understood.

Magnesium has also been reported to be important in the pathogenesis of vasospastic angina. Magnesium exerts a profound effect on Ca²⁺ channel activity, cellular metabolism, and bioenergetics (10). Goto et al. (5) reported that an Mg deficiency, as estimated by the Mg infusion test, is present in patients with vasospastic angina. The intravenous infusion of Mg has been found useful in treating of coronary vasospasm (17). However, the cytosolic free Mg²⁺ concentration ([Mg²⁺]_i) in patients with vasospastic angina has not been examined adequately. Because a large percentage of the Mg in the body is in the intracellular compartment and ionized Mg mainly plays a physiological role in the intracellular space (10), determination of [Mg²⁺]_i is very important for achieving an understanding of Mg balance.

The present study investigated the [Mg²⁺]_i, Ca²⁺ handling, and aggregation response in platelets obtained from patients with vasospastic angina.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects. Thirty-six Japanese male patients who had experienced spontaneous angina attacks at rest associated with ST-segment elevation on the electrocardiogram underwent diagnostic cardiac catheterization. None of the patients had previously suffered myocardial infarction, heart failure, hypertension, or diabetes mellitus. The administration of antiplatelet agents was halted at least 4 wk before the study. The administration of other antianginal drugs was stopped at least 3 days before the study, except for nitroglycerin, which was stopped 3 h before the study. In 24 of these patients, the intracoronary administration of acetylcholine provoked 90% stenosis or 50% reduction of the diameter of the artery during coronary angiography (13, 35). Eight patients who showed 50% atherosclerotic stenosis on the coronary angiogram and four patients who could not demonstrate the vasospasm by the administration of acetylcholine were excluded. Therefore, 24 males with vasospastic angina (age range, 40-72 yr; mean age, 54 ± 10 yr) were included in the platelet study protocol. Twenty-four healthy Japanese males (age range, 32-74 yr; mean age, 51 ± 14 yr) served as control subjects. These control subjects were healthy male volunteers, none of whom had ever had typical chest pain. Their electrocardiographic findings at rest and during hyperventilation and exercise tests were all normal. Because of a reported gender difference in [Mg²⁺]_i (37), only males were included in the present study. Approval for this study was obtained from the local Ethics Committee, and written informed consent was obtained from each subject.

Catheterization procedure. During coronary catheterization, arterial pressure and a 12-lead electrocardiogram were monitored continuously. Coronary angiography was performed using a femoral approach. The angiogram was recorded with a cineangiography system (Siemens, Munich, Germany). After an appropriate view that allowed clear visualization of the coronary artery had been selected, control coronary angiography was performed. Distances between the patient and the X-ray focus and the image intensifier were kept constant during the study. Acetylcholine chloride (Daiichi Seiyaku, Tokyo, Japan) dissolved in 2 ml of warmed 0.9% saline, in incremental doses of 6 and 60 µg, was injected over 2 min into the coronary artery that was presumed to be responsible for the attack. Coronary angiography was performed at 1-min intervals after each dose of acetylcholine. When anginal pain or ST-segment change occurred, coronary angiography was performed immediately. When coronary vasospasm was documented, 0.2 mg nitroglycerin (Nihon Kayaku, Tokyo, Japan) was injected into a coronary artery. After the administration of nitroglycerin, coronary angiograms were again recorded in multiple views to assess the atherosclerotic stenosis of the coronary arteries.

Platelet preparation. During a nonanginal period, between 48 and 72 h after coronary catheterization, after drawing 1 ml of blood into a 2.5-ml syringe from an antecubital vein, a second syringe containing 2 ml of 3.8% trisodium citrate was connected to the needle, and 18 ml of blood were collected slowly and steadily (two-syringe technique) (19). The administration of antianginal drugs was stopped before the blood sampling. In the preliminary study, we compared the levels of resting cytosolic free Ca²⁺ concentration ([Ca²⁺]_i) before and 48-72 h after the coronary catheterization procedure in subjects who had not received any drugs for 4 wk. We confirmed that there were no differences between pre- and postcatheterization levels of resting [Ca²⁺]_i [25 ± 3 vs. 25 ± 3 nM (n = 10)]. Therefore, the coronary catheterization procedure should not affect our results. Platelet-rich plasma was prepared by centrifuging the blood samples at 800 g for 5 min at ~18°C (room temperature). Platelets were separated from the plasma by gel filtration using a Sepharose 2B-CL column (Pharmacia Biotechnology, Uppsala, Sweden) and then diluted with N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) buffer containing (in mM) 145 NaCl, 5 KCl, 1 MgSO₄, 10 HEPES, and 5 glucose, at pH 7.4, to a concentration of 10⁸ cells/ml. To minimize any time-dependent effect on platelet responsiveness as well as leakage of dyes, measurements of platelet [Mg²⁺]_i, [Ca²⁺]_i, and aggregation were performed separately by three independent investigators (Yoshimura, Hiraga, and Nakano, respectively) in the same blood specimens within 2 h after blood collection.

Measurement of [Mg²⁺]_i. The platelets were incubated with 2 µM mag-fura 2-acetoxymethyl ester (fura 2-AM) (Molecular Probes, Eugene, OR) and 0.02% Pluronic F-127 (Molecular Probes) for 30 min at 37°C. The cells were gel-filtrated again to remove the extracellular dye, and platelets were diluted with HEPES buffer to a concentration of 10⁷ cells/ml: CaCl₂ was added to the cell suspension at a final concentration of 1 mM. For fluorescence measurements, 2.5-ml aliquots of cell suspension were stirred continuously by a magnetic stir bar in a quartz cuvette at 37°C, and fluorescence was recorded with a spectrofluorophotometer (SPEX Fluorolog; SPEX Industries, Edison, NJ) using excitation wavelengths of 340 and 380 nm and an emission wavelength of 510 nm. Corrections were applied for extracellular leakage of mag-fura 2 by adding 3 mM EDTA (Dojindo Laboratories, Kumamoto, Japan) as described previously (37, 38) and for autofluorescence by subtracting the fluorescence of the unloaded platelets and test agents. A rapid initial drop in the fluorescence signal at 340 nm after the addition of EDTA was considered to reflect the contribution made by extracellular dye as extracellular Mg²⁺ and Ca²⁺ were chelated. [Mg²⁺]_i was calculated using a general formula (23) [Mg²⁺]_i = K_d × (R  R_min)/(R_max  R) × (S_{f, 380}/S_{b, 380}), where K_d is 1.5 mM for mag-fura 2, R is the ratio of fluorescence at excitation wavelengths 340 and 380 nm in a suspension of intact cells, and R_max was the 340-to-380-nm fluorescence ratio when the fluorescence dye was saturated with Ca²⁺ (because similar R_max values are obtained on saturation of the dye with Mg²⁺ and Ca²⁺) (23). R_min was the ratio at zero Mg²⁺ and Ca²⁺. S_{f, 380} and S_{b, 380} are the fluorescence intensities at 380 nm for mag-fura 2 with concentrations of zero and excess Mg²⁺ and Ca²⁺, respectively. The intracellular concentration of mag-fura 2 was determined by use of an in vitro mag-fura 2 calibration curve created by measuring the fluorescence of known concentrations of mag-fura 2 (not mag-fura 2-AM).

Measurement of [Ca²⁺]_i. [Ca²⁺]_i was measured as described previously (19), with a slight modification for humans. To measure [Ca²⁺]_i, platelets incubated with 1 µM fura 2-AM (Molecular Probes) were treated as in the determination of [Mg²⁺]_i. Fluorescence at excitation wavelengths of 340 and 380 nm and an emission wavelength of 510 nm was recorded with a spectrofluorophotometer (RF-5000; Shimadzu, Kyoto, Japan). After the recording of fluorescence in the basal state, the intracellular Ca²⁺ response to thrombin (Sigma Chemical) and ADP (Sigma Chemical) was evaluated, both in the presence of 1 mM Ca²⁺ and in Ca²⁺-free HEPES buffer prepared by the addition of 10 mM ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA, Dojindo Laboratories). We also evaluated the intracellular Ca²⁺ response to 5 µM ionomycin (Sigma Chemical) in Ca²⁺-free HEPES buffer, the concentration at which a maximal response was elicited, as an index of the intracellular Ca²⁺ discharge capacity (20). [Ca²⁺]_i was calculated using a general formula (6), [Ca²⁺]_i = K_d × (R  R_min)/(R_max  R) × (S_{f, 380}/S_{b, 380}), where K_d is 224 nM for fura 2, R is the ratio of fluorescence at excitation wavelengths 340 and 380 nm in the intact cell suspension, R_max and R_min are the 340-to-380-nm fluorescence ratios obtained at concentrations of saturating and zero Ca²⁺, respectively, and S_{f, 380} and S_{b, 380} are the fluorescence intensities at 380 nm for fura 2 with concentrations of zero and excess Ca²⁺, respectively. Corrections were applied for extracellular leakage of fura 2 by adding EGTA (19) and for autofluorescence by subtracting the fluorescence of the unloaded platelets and test agents by the method described above. The cytosolic fura 2 concentration was estimated from an in vitro fura 2 calibration curve by the method described above. Under both basal and stimulated conditions, the intra-assay coefficient of variation for [Ca²⁺]_i was <5.0% and the day-to-day variation in any one subject was <5.0% (n = 5).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Patient characteristics. The patients with vasospastic angina and the control subjects showed no significant difference in age, mean blood pressure, body mass index, fasting blood sugar, chemical lipid values, or serum electrolyte concentrations (Table 1). No differences between the two groups were detected in intracellular mag-fura 2 (Table 2) and fura 2 (Table 3) metabolism, indicating that platelets were loaded with the dyes to a similar extent.

Basal [Mg²⁺]_i and [Ca²⁺]_i. [Mg²⁺]_i in platelets of patients with vasospastic angina was significantly higher than that of control subjects (580 ± 160 vs. 460 ± 140 µM; P < 0.01, Fig. 1). Thrombin (0.03-1.0 U/ml) had no significant effect on [Mg²⁺]_i in Ca²⁺-containing medium (data not shown). Resting [Ca²⁺]_i was significantly higher in platelets from patients with vasospastic angina than in platelets from control subjects (26 ± 4 vs. 19 ± 5 nM; P < 0.0001), as shown in Fig. 2.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Resting cytosolic free Mg2+ concentration ([Mg2+]i) in platelets of control subjects (n = 24) and of patients with vasospastic angina (VSA) (n = 24). Bars represent means ± SD. Resting [Mg2+]i was significantly higher in platelets of VSA than in those of control subjects.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Resting cytosolic free Ca2+ concentration ([Ca2+]i) in platelets of control subjects (n = 24) and of VSA (n = 24). Bars represent means ± SD. Resting [Ca2+]i was significantly higher in platelets of VSA than in those of control subjects.

[Ca²⁺]_i responses to thrombin and ADP. The [Ca²⁺]_i responses to thrombin (0.03-1.0 U/ml) with extracellular Ca²⁺ were larger in patients with vasospastic angina than in control subjects, and the group differences in the responses to thrombin at concentrations of 0.03, 0.1, 0.3, and 1.0 U/ml were statistically significant (Fig. 3). The thrombin-induced increase in [Ca²⁺]_i consists of a Ca²⁺ influx across the plasma membrane and a Ca²⁺ release from intracellular stores. To assess the latter, we measured increases in [Ca²⁺]_i in response to thrombin (0.03-1.0 U/ml) in the absence of extracellular Ca²⁺. There was no difference between the two groups in [Ca²⁺]_i increase in response to various concentrations of thrombin in Ca²⁺-free medium (Fig. 4). There was no difference between the two groups in [Ca²⁺]_i increase in response to two concentrations of ADP in Ca²⁺-containing and in Ca²⁺-free medium (Table 4). The ionomycin-induced increase in [Ca²⁺]_i without external Ca²⁺, an index of the size of internal Ca²⁺ stores (19, 20), was similar in patients with vasospastic angina and in control subjects (680 ± 250 vs. 660 ± 170 nM).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Thrombin-stimulated increase in [Ca2+]i in platelets of control subjects (n = 24; ) and of VSA (n = 24; ). Data are expressed as means ± SD. Thrombin-induced increase in [Ca2+]i in the presence of extracellular Ca2+ was significantly greater in platelets of VSA than in those of control subjects. * P < 0.05, ** P < 0.01 vs. corresponding control value.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Thrombin-stimulated increase in [Ca2+]i in platelets of control subjects (n = 24; ) and of VSA (n = 24; ). Data are expressed as means ± SD. Thrombin-induced increase in [Ca2+]i in the absence of extracellular Ca2+ was similar in both groups.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Thrombin-induced maximum aggregation in platelets of control subjects (n = 24; ) and of VSA (n = 24; ). Data are expressed as means ± SD. The maximum aggregation to thrombin at concentrations of 0.1, 0.3, and 1.0 U/ml was significantly higher in platelets of VSA than in those of control subjects. # P < 0.005 vs. corresponding control value.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

A negative balance in the systemic level of Mg, as estimated by an intravenous Mg loading test, has been demonstrated in patients with vasospastic angina (5). Intravenous administration of Mg suppresses anginal attacks in patients with vasospastic angina (17). It has, therefore, been postulated that Mg deficiency is involved, at least in part, in the pathogenesis of coronary spasms. In the present study, however, neither serum nor [Mg²⁺]_i was decreased in patients with vasospastic angina. Thus our finding cannot support the Mg deficiency hypothesis for coronary spasms.

Less than 10% of cytosolic Mg²⁺ is thought to exist in the free form (4), and most of the remainder is thought to be bound to ATP (4). [Mg²⁺]_i is reportedly largely insensitive to changes in [Ca²⁺]_i or pH in an extended physiological range (29) and extracellular Mg²⁺ concentration (38). ATP is known to be an intracellular factor affecting [Mg²⁺]_i; ATP depletion causes a rapid increase of [Mg²⁺]_i (7, 29). Therefore, an elevated platelet [Mg²⁺]_i in patients with vasospastic angina might be attributable to decreased ATP content or to a varied ratio of the free and bound forms of cytosolic Mg²⁺. The resting value for [Mg²⁺]_i is much lower than would be expected from a passive distribution of Mg²⁺ across the plasmalemmal membrane (4, 37). One of the most important mechanisms maintaining this low [Mg²⁺]_i in human platelets is an Na⁺-dependent Mg²⁺ efflux through the plasmalemmal membrane (4, 38). Therefore, decreased platelet activity of the Mg²⁺/Na⁺ exchanger might also contribute to an elevated platelet [Mg²⁺]_i in patients with vasospastic angina.

Cellular Mg²⁺ status, as well as basal [Ca²⁺]_i, in platelets might be modified by hypertension (1, 24), non-insulin-dependent diabetes mellitus (24), and gender difference (37). However, our subjects were all male and none of them had hypertension, diabetes mellitus, or hyperlipidemia. In addition, there was no difference between the experimental and control groups in blood pressure, plasma glucose, or lipid metabolism. Therefore, the alteration in cellular metabolism of Ca²⁺ and Mg²⁺ in platelets of patients with vasospastic angina may not result from the change in clinical backgrounds.

Methodological issues are important in the assessment of [Ca²⁺]_i and [Mg²⁺]_i in fluorescent dye-loaded cells (18, 19). We minimized methodological problems that might have masked real alterations in the metabolism of these cations in vasospastic angina. First, variations in intracellular concentration and dye ester hydrolysis of fura 2 and mag-fura 2 may affect the calculation of [Ca²⁺]_i and [Mg²⁺]_i. We demonstrated that cellular metabolism of fluorescent dyes was similar in patients with vasospastic angina and controls. Second, the leakage of fluorescent dye from cells will result in the erroneous estimation of levels of [Ca²⁺]_i and [Mg²⁺]_i. When no correction was made for dye leakage, [Ca²⁺]_i and [Mg²⁺]_i were overestimated in standard buffer supplemented with Ca²⁺ and Mg²⁺, and variations in the measured [Ca²⁺]_i and [Mg²⁺]_i increased. Thus we can safely compare the data between patients with vasospastic angina and controls. Furthermore, as the difference in estimation method may produce different values in cytosolic free concentration of cation, the basal level of [Ca²⁺]_i reported in this study (~20 nM) was lower than those in some earlier reports (27, 30). Recently, careful investigators have reported levels of basal [Ca²⁺]_i similar to that in the present study (11, 18).

In response to agonists such as thrombin, inositol (1,4,5)-trisphosphate, which is formed by the hydrolysis of an inositol lipid precursor stored in the plasma membrane, mobilizes Ca²⁺ from intracellular stores. There is also an influx of extracellular Ca²⁺ through Ca²⁺ channels. Because the thrombin-induced [Ca²⁺]_i response in the absence of extracellular Ca²⁺ was similar in both groups, the accelerated [Ca²⁺]_i response to thrombin in patients with vasospastic angina may result from enhanced Ca²⁺ influx across the plasma membrane rather than enhanced Ca²⁺ discharge from intracellular stores. Because some aspects of platelet function (including the properties of glycoprotein IIb/IIIa) are affected by levels of extracellular Ca²⁺, there is a possibility that these aspects are different. Enhanced [Ca²⁺]_i response in thrombin-stimulated platelets cannot be explained by a change in stores regulating Ca²⁺ entry (capacitative model) (21), because Ca²⁺ discharge and intracellular Ca²⁺ discharge capacity were similar in the two groups. It is possible that the key mechanisms for abnormal Ca²⁺ handling by platelets in vasospastic angina may exist in signal transduction pathways that regulate Ca²⁺ channels, such as adenosine 3',5'-cyclic monophosphate, guanosine 3',5'-cyclic monophosphate, GTP-binding protein, protein kinase C, and myosin light-chain kinase (8, 22, 39).

Ca²⁺ release and influx induced by ADP are markedly different from those evoked by other agonists such as thrombin, thromboxane A₂, vasopressin, and platelet-activating factor. ADP induces Ca²⁺ release from intracellular Ca²⁺ stores independent of inositol (1,4,5)-trisphosphate formation (3) and evokes influx using a different transduction system that is more closely coupled to the Ca²⁺ entry system than that used by other agonists (26). Therefore, in this study, it is suggested in the platelets of patients with vasospastic angina that these specific Ca²⁺ transduction systems for ADP may not be impaired or the abnormality in Ca²⁺ handling by ADP-stimulated cells may be obscured by modification in other agonist-specific factors.

Although the inhibitory effect of extracellular Mg²⁺ on Ca²⁺ influx has been established (9), little is known about the effect of intracellular Mg²⁺ on Ca²⁺ influx. Because Mg²⁺ is extruded from the cytoplasm against its electrochemical gradient and the permeability of the membrane to Mg²⁺ is low (38), the effects of external and internal Mg²⁺ on cellular Ca²⁺ handling must be addressed separately. Although increases in [Mg²⁺]_i in the millimolar range inhibit Ca²⁺ influx in cardiac myocytes (32, 34), modulation of Ca²⁺ influx by changes in [Mg²⁺]_i in the physiological (submillimolar) range has not been characterized in platelets. We have previously reported that an increase in [Mg²⁺]_i in the physiological range accelerated Ca²⁺ influx in vascular smooth muscle cells (36). Therefore, elevated [Mg²⁺]_i in the physiological range may facilitate Ca²⁺ influx in platelets as well as in vascular smooth muscle cells. Further research is needed to present direct evidence linking an elevation of [Mg²⁺]_i at the micromolar level to accelerated Ca²⁺ influx in platelets of patients with vasospastic angina. Free Mg²⁺ is critical in DNA transcription, protein synthesis, and a number of enzymatic reactions (4, 10). It is, therefore, conceivable that a compensatory increase in [Mg²⁺]_i in vasospastic angina may have adverse effects on accelerated cell function.

In conclusion, platelet [Mg²⁺]_i in patients with vasospastic angina was significantly higher than that in control subjects. Whereas thrombin-induced [Ca²⁺]_i response in the Ca²⁺-free medium and intracellular Ca²⁺ discharge capacity were similar in both groups, the basal [Ca²⁺]_i and thrombin-induced Ca²⁺ influx across the plasma membrane may be greater in the patients with vasospastic angina than in controls. The increased platelet [Mg²⁺]_i and these alterations in Ca²⁺ handling may be associated with coronary vasospasm.

Perspectives