- Email: [email protected]
Magnesium regulates intracellular free ionized calcium concentration and cell geometry in vascular smooth muscle cells
Biochimica et Biophysica Acta, 1134(1992) 25-29 ©
25
1992ElsevierScience PublishersB.V. All rights reserved0167-4889/92/$05.00
BBAMCR 13108
Magnesium regulates intracellular free ionized calcium concentration and cell geometry in vascular smooth muscle cells A i m i n Z h a n g a, T o n i P . O . C h e n g b a n d B u r t o n M . A l t u r a a a Department of Physiology and b Department of Anatomy and Cell Biology, State Unit~tsity of New yod¢. Health Science. Center at Brooklyn, Brooklyn, N Y (USA)
(Received 14 August1991)
Keywords: Magnesium;Calcium;Vascularsmoothmusclecell;Fura-2 Regulatory effects of extraceUular magnesium ions ([Mg2+]o) on intraceUular free ionized calcium ~[Ca2+~.) were studied in cultured vascular smooth muscle cells (VSMCs) from rat aorta by use of the fluorescent indicator fura-2 and digital imaging microf,copy. With normal Mg2+ (1.2 mM)-containing incubation media, [Ca2÷]i in VSMCS was 93.6+7.93 nM with a heterogeneous cellular distribution. Lowering [Mg2+]o to 0 mM or 0.3 mM (the lowest phlmiological range) resulted in 5.8-fold (579.5+39.99 nM) and 3.5-fold (348.0+31.52 nM) increments of [Ca2*]i, respectively, without influencing the ceHalar distribution of [Ca2+]i. Surprisingly, [Mg2÷]o withdgawal induced changes of cell geomeWy in many VSMCs, i.e., the cells rounded up. However, elevation of [Mg2+]o up to 4.8 mM only induced slight decrements of [Ca2+]i- (mean - 72.0 + 4.55 riM). The large increment of [Ca2+]i induced by [Mg2+]o withdrawal was totally inhibited when |Ca2+] o was removed. The data suggest that: (1) [M82+]o regulates the level of [Ca2÷]i in rat aortic smooth muscle cells, and (2) [Mg2+] acts as an iml~'tant regulatory ion by modulating cell shapes in cultured VSMc and their metabolism to control vascular contractile activities.
Introduction The importance of magnesium ions (Mg 2+) as cofactors for numerous enzlnnatic reactions and as a ubiquitous chelator of intracellular A T P has long been recognized, but there is growing evidence that Mg 2+ may also act as a regulatory ion [1,2]. In vascular smooth muscle cells (VSMCs), antagonistic effects of calcium ions (Ca 2+) and Mg 2+ have been established for contraction, and the interaction of Ca z+ and Mg 2+ is known to control vascular reactivity. Mg 2+ has been demonstrated to modulate basal tone, myogenic tone and contractile responsiveness of VSMCs to various physiological and pharmacological stimuli by modulating calcium content, binding and transport [3-5]. Mg z+ is thought to affect Ca 2+ flux across vascular muscle membranes and its release from internal stores, while an acute elevation or decrease of extracelluiar Mg 2+ concentration ([Mg2*]o) is known to exert rapid inln'b~-
Abbreviation:VSMCe,vascularsmoothmu.w.lecells. Correspondence:B.M. Altara, Beet31, Sun.v-HealthScience Center at Brooklyn,450 ~ Avenue,Brooklyn,NY II21D, U.S.A.
tion or potentiation of vascular contractile tension, respectively. Given its rather unique qualities compared to any known Ca 2+ channel blockers, Mg 2+ is believed to be a naturally occurring Ca 2+ antagenist
[6].
Since a change in ojtosolic free Ca z+ concentration ([Ca2+]i) is a very important second messenger in facilitation of contraction or relaxation [7], it may be physiologically relevant that Mg 2+ regulate the activity of VSMCs by competing with Ca 2+ and modulating the cytesolie free [Ca2+]i [6,8,9]. We repart here a novel effect of [Mg2+]o on regulation of [Ca2+]i and cell geometry in cultured VSMCs.
Materials t e t l Methods The distribution of [Ca2+]i in cultured VSMCs from rat aortas was determined using digital imaging fluorescence microscopy. Aortic smooth muscle cells from adult Wistar rats (200-250 g) were isolated according to previous e s t a ~ methods [10] and cultured in Dulbecco's modified Eagle's medimn at 37"C in a humidified atmosphere ~ of 95% air-5~£ CO 2. Cells for image ana~skt experintents were plated onto glass cover s l i ~ at a density of 1.0-10 s celis/ml. All
26 images were analyzed using a temperature-controlled Nikon fluorescence microscope and digitized by a TN8500 FluorPlex Image Analyzer (Tracer Northern, Madison), and using fura-2 as a fluorescent Ca 2+ indicator, which has a high selectivity for Ca 2+ over Mg 2+ and other divalent cations [11]. The coverslips containing a monolayer of cells were loaded with fura-2 by incubating them with 2 ~tM fura-2/AM in the culture media for 60 min under the above conditions. The coverslips were washed for 5 min with Hepes buffer solution (pH 7.4) containing 0, 0.3, 1.2 or 4.8 mM MgSO4, respectively. In addition to MgSO4, the Hepes buffer solution contained (mM): NaCI 118, KCI 4.7, CaCI 2 2.5, KH2PO 4 1.2, Hepes 5, glucose 10. The pH was adjusted to 7.4 with NaOH. The coverslips containing the fura-2-1oaded cells were affixed to holders (glass slides) that were placed on the temperature-controlled stage of the fluorescence microscope. The cells were excited, alternatively, at 340 n m and 380 nm; and the emission intensity was recorded at 510 nm, using a silicon intensified target (SIT) camera. The intensities of the recorded images at 340 and 380 n m were corrected by subtracting them from background fluorescence recorded at the corresponding wavelengths. The resulting images were then used to calculate [Ca2+] i in VSMCs, using external standards containing 2.54 mM Ca 2+ and 0 mM Ca 2+ plus 10 m M EGTA, a K d of 224 nM [11] for the f u r a - 2 / C a 2+ complex, and maximum (Rmu) and minimum (Rmi n) fluorescence ratios of the 340 and 380 nm-images. The fluorescent intensity in a single VSMC could be recorded accurately with high resolution. Visual examinations of fura-2 loaded cells revealed that a uniform fluorescence appeared when excited at 380 nm without any obvious centers of concentration, indicating to us a good distribution of the fura-2 throughout the treated cells. Results and Discussion When incubated with Hepes buffer solution containing 1.2 mM Mg 2+, the basal level of [Ca2+]i in cultured VSMCs, estimated from the ratio (F340/F380), is 93.2 + 7.9 nM (48 cells in four coverslips). The valt,cs varied among cells over a range between 14.1 nM and 177.7 nM, which is a quite comparable with basal values reported by others [12,13]. The resting distribution of [Ca2+]i , rather than the simple difference in
fura-2 concentration [13], also appeared heterogeneous within the cells by ratio image analysis (Fig. 1). Most of the VSMCs in the primary culture appeared spindleshaped (of. Fig. 1). The prenuclear region, arbitrarily chosen as the central ellipsoid area in the widest portion of the cell [14], revealed the most brightness, reflecting a relative higher [Ca2+]i than the peripheral cytoplasmic area. This is probably due to the fact that the pranuclear region contains most of the cellular organelles, such as the sarcoplasmic reticnlum (SR) a n d / o r rough endoplasmic reticulum (RER), which can sequester Ca 2+ by M~2~'-depandant proce.~es [5,13,14]. Therefore, the [C,a2+]i, we measured here, represents not only Ca 2+ in the cytosol, but also Ca 2+ sequestered in SR a n d / o r other intracellular Ca 2+ storge sites in VSMCs, and is probably an overestimate of the true [Ca2+]i. The lower [Ca2+]l in the peripheral cytoplasmic region may be relatively close to the true [Ca2+]i. The observation of a [Ca 2+] gradient in the cell suggests that the plasma membrane of VSMCs and the discrete localized processes regulate Ca 2+ activity subcellularly. In order to probe the effects of Mg 2+ on [Ca2+] i in VSMCs, the extracellular Mg 2+ concentrations ([Mg2+] o) in the incubation media (Hepes buffer solution) were altered. Such treatment did not, however, influence loading of the dye into the cells. Somewhat to our surprise, removal of extracellnlar Mg 2+ ([Mg2+]o z 0 mM), after only 5 min of washing VSMCs with Mg2+-free incubation media, induced a dramatic increment of [Ca2+]t up to 579.5 + 39.3 nM (91 cells from four individual coverslips), or about a 5.8-fold elevation compared to the control experiments ([Mg2+]oZ 1.2 mM) (Fig. 2). Concomitant with the latter, we noted large variations in changes of [Ca2+] i brought about by removal of extracellular Mg 2+ exhibiting a range from 137.7 nM to 2149.4 nM. Such rises of [Ca2+] i were maintained consistently, at least over a 30 min period of observation. To examine the [~2+]i changes produced under physiological conditions, [Ms2+] o was lowered to 0.3 mM, the lowest ionized physiological level seen clinically. This maneuver raised [Ca2+] i about 3.5-fold or to 348.0 + 31.5 nM (116 cells from five individual coverslips). However, the heterogeneous distribution of [Ca2+] i was still evident when [Mg2+]o was lowered to 0.3 raM, even though [Ca2+]t in the peripheral cytoplasmic region was in-
Fig. I. Ratio of 340/380 imagesof fura-2fluorescencein culturedvascularsmoothmusclecells incubatedin HEPES buffer solutionwith normal 1.2 mM Mg2+. The ratioed imageshowsthat [Ca2+]i is apparentlyhigherin the prenuclearregion.The numbersaround the color scale indicate the approximatenM concentrationsof free, ionizedintraceliolarCaz+. Fig. 2, Ratio of 340/380 imasesof fura-2 fluorescencein culturedvascularsmooth musclecells incubatedin MgZ+-froeHepes buffer solution. Note that the removal of [Mg2+]o raises [Ca2+]j both in the nuclear and the peripheral cytoplasmicregions, while the hetemgennons distributionof [Ca2+]l still persists.Someof the cultured cells dearly round up. The numbersaround the color scale indicatethe approximate nM concentrationsof fTee,ionizedintracellularCa2+.
F~.I.
28 TABLE I Effects of egtracellular magnesium on the regulation of intracellaior flee calcium ion in cultured vascular smooth muscle from rats
Values- means+S.E.; n - the number of cells; * Significantlydifferent from 1.2 mM MS2÷ (P<0.01). ** Significantly different from 1.2 mM MS2+ (P < 0.02). [MS+++],, (mM) 1.2 0 0.3 4.8 0
[Ca2+lo (mM) 2.5 2.5 2.5 2.5 0
n 48 91 !16 126 21
[Ca++]+ (nM) 93.6+ 7.90 579.5+39.3* 348.0+31.5* 72.0:1:4.55** 3.6+ 0.51*
creased. It is noteworthy that, simultaneously to the alteration of [ca2+]i, cell shapes were changed by lowering [Mg2+]o. Visual examinations revealed that many VSMCs were rounded up when [Mg2]o was with~ drawn indicating that these single cells can clearly change shape and undergo strong contractions in Mg2+-defieient media (el. Fig. 2), and that the 5.8-fold elevation of [ca2+]i is responsible for the latter events. Although the changes of cell geometry induced by contraction could influence the accuracy of [Ca2+] i measurements, the ratioed image, herein, reflects an absolute increase of [Ca2+]i . In contrast to lowering [Mg2+]o, elevation of [Mg2+] o to 4.8 mM (four times over normal [Mg2+] o) was found to decrease [ca2+]i to about 72.0 + 4.5 nM (126 cells from six individual coverslips), showing relatively less of an influence on [Ca2+]i compared to Mg 2+ removal (Table I). It is interesting to note that the intracellular heterogeneity of [Ca2÷] i WaS clearly less visible, and the brightness of the nuclear region became much weaker when [Mg2+]o was raised to 4.8 mM. These observations support our previous findings on intact and isolated blood vessels which indicate that reduction in [Ms 2+ ]o results in spasms of blood vessels, elevation of basal tone and increments in spontaneous mechanical activity [3,5,9]. The [Ca2+]i changes brought about by varying [Mg2+]o clearly indicate the dependence of [Ca2+]i and basal [ca2+] t upon [Ms2+] o. As shown in Table I, although there is relatively less of a change in [Ca 2+]i following changes of [Ms 2+]o above the normal physiological level ( > 1.2 mM), a reverse relationship of [Ca 2+ ]l vs. [Mg 2+ ]o was virtually exerted when [Mg2+] o is below this value, indicating for the first time a [Mg2+]o.dependent mechanism for regulating intracellular [ca2+]i distribution. To obtain further insights into the existence and n a t u r e of these regulatory processes, VSMCs were exposed 'to a modified Hepes buffer solution in which both Ca 2+ and Mg 2+ were removed. It was anticipated that such a maneuver would reveal the steady-state distribution of [ca2+]l a~'oss the sateolemma and
intracellular organelle membranes. A strong dependence of the [Mg2+]o-induced increment of [Ca2+] i upon [Ca2+]o was observed as exemplified by the marked lowering of [Ca2+] i (to 3.6 nM) in such treated cells. This represents about a 200-fold decrement compared with cells treated by [Mg2+]o withdrawal alone. The [ca2+] i gradient completely disappears in the absence of [Ca2+]o. Although the removal of extracellular Ca 2+ and Mg 2+ may increase membrane permeability and consequently result in an ebbing away of fura-2 from VSMCs via membrane leakage, as indicated by the decrement of fluorescent intensity at 380 nm, the sharp drop in [Ca 2÷] observed here reveals that the entry of external Ca 2+ occurs primarily in resting VSMCs when [Mg2+]o is lowered, and that alterations of [Ca2+] i induced by Mg 2+ seem to be reversible. However, such dye leakage induced by the removal of [Ca2+]o and [Mg2+]o, on the other hand, suggests that both divalent cations are required for maintaining membrane permeability and integrity of the cell membrane. The experiments described above demonstrate that Mg 2+ exerts profound influences on [Ca2+], in cultured VSMCs. At present, the mechanism by which [Mg2+]o regulates [Ca2+]i is unclear. The alterations in [Ca2+] i in this cell type could be a consequence of fluctuations either in the entry of external Ca 2+ or in its release from internal stores. However, both of these fluctuations could be affected by [Mg2+] o 69,15]. Since the heterogeneous distribution of [Ca2+]l is hardly affected by [Mg2+]o, it could be suggested that the intracellular compartment of Ca 2+ may not play a major role in these events, in spite of data which indicates that Mg 2+ is known io regulate Ca 2+ release and sequestration in VSMCs. The dependence of [Ca 2+ ]i on [Ca 2+ ]o, however, suggests that the mechanism whereby Mg 2+ regulates [Ca2+] t in VSMCs is linked mainly to a modulation of membrane permeability to [Ca2+]o. Lowering [Mg2+]o is known to inaease 4sea influx and efflux [3], as well as increase exchangeable and membrane-bound Ca 616], whereas high [Mg 2+]0 exerts opposite effects. Therefore, alterations in [Ca2+]i could be a consequence of Ca2+ entry governed by Mg2+-controlled membrane permeability. Previous experimental findings indicate that, unlike most Caz+ channel antagonists, Mg z+ can act on voltage-dependent 62,5] and receptor-operated channels [5], as well as leak.operatea channels in the resting state 65,16]. With regard to cellular Mg contents which seem to be well-below electrochemical equilibrium, it has been suggested that there are active Mg transport systems in the cells [17,18], such as M g 2 / M g 2+ exchange, Mg2+/Ca 2+ exchange or N a + / M g 2+ exchange, by which Ca2+ may compete with Mg 2+ for transport sites. The present, new findings indicate that Mg 2+ may regulate membrane permeability to [Ca2+]o
29 by a novel mechanism r a t h e r than as a simple blocking particle on a n open channel or a simple m e m b r a n e binding ion. Alternatively, active M g transport systems may play a major role as a result of [Mg2+]o regulation of both influx and efflux of Ca 2+ [3]. W h e t h e r or not changes in [Mg2+] o can alter, rapidly, intracellular free M g 2+ concentration ([Mg2+]i) in VSMCs of the rat aorta is at present not known. F u r t h e r studies are thus n e e d e d to clarify these events and to elucidate which one of these parameters, i.e., [Mg2+]o, [Mg2+] i or [Mg2+]o/[Mg2+]i, is most important for controlling m e m b r a n e permeability to [Ca2+]o in V~MCs. The present finding of a fine regulation of [Ca2+]i by [Mg2+] o may have several important implications. Irrespective of the precise m e c h a n i s m s whereby [Mg2+]o controls Ca 2+ entry and cell geometry, changes of [Mg2+] o could induce substantial long-term potentiation or inhibition of vascular reactivity and basal tone. Clinically, M g deficiency is associated with a n u m b e r of cardiovascular disorders [3,19]. These include hypertension, stroke, cardiac arrbythmias, cardiac ischemia and myocardial infarction, as well as atherosclerosis [20], M a n y of these pathological processas are associated with vascular spasm [19], a consequence of h y p o m a g n e s e m i a which could result in increm e n t s of [Ca2+] i in VSMCs by the mechanism described herein.
~
t
This work was s u p p o r t e d in part by A D A M H A G r a n t AA-08674 t o B.M.A.
References I Grubbs, R.D. and ~ ME. (1987) Maanesiam 6,113--127. 2 White, R.E. and Hastzen, EI.C.. (1989) Biochem. Pharmacol. 38, 859-867. 3 Altura, B.M. and Altura, B.T. (1961) Fed. Pm~ 40, 2672-2679. 4 Aitura, B.M., Altura, B.T, Carella, A. and TedalmtY, P.D.M.V. (1982) Can. J. Physiol. Phanaacel. 60, 459-4~. 5 Altura, B.M., Altma, B.T., Camlht, A., GelMewoM, A., Murakawa, T. and Nishio, A. (1997) Can J. ~ PharmacoL 65, 729-745. 6 Altm'a, B.M. and Altura, B.T. (1981) In New Pegspectiv~ on Calcium Antagonists (Wei~ G.B., ed.), pp. 131-145, Am. Phr~ iol. Soc., Washington. 7 Kuriyama, H., Ito, Y, Suzuki, H, Kitamera, If. and ltolk IL (1982) Am. J. Physiol. 243, I-I641-He6Z 8 Altura, B.M. and Altura, B.T. (1971)Am. J. ~ 220, 938-944. 9 Altura, B.M. and Altura, B.T. (19/4) ~ Res. 7,145-155. 10 Yamamoto, H., ~ a a k k , H., Nalumm'a, M. (1~3) Be. J. Exp. Pathol. 64,156-165. II Grsnkiewicz, G., Poeaie, M. and Tslon, R.Y. (1~5) J. Biol. Chew- 260, 3440-3450. 12 Nabel, E.B,, Berk, B.C., Brock T.A, and Smith T.W. (1988) Circ. Res. 62, 486-493. 13 Goto, A., Yamada, If.,hhii, M, Ymhleb, M,, hhit~ro, T, Eguchi, C and Sulgimoto,T. (1989) ~ e m i o n 13, 916-921. 14 Papngeorsion, P. and Mecgan, K.G. (1989) Prec. Soc. ~ Med. 193, 331-334. 15 Sjfgmn, A. and Edvimao~ L (1~8) Pharmm:oL TmicoL 62, 17-21. 16 Turlapa~, P.D.M.V. and Altura, B.M. (1978) Ear. J. PhmmacoL 52, 421-423. 17 Flatman, P.W. (lC~4) J. Membr. Biol. 80,1-14. 18 Altura, B.M. and Altma B.T. (1981) ~ B u l k t i a It, 102-114. 19 Altura, B.M. and Altwa, B.T. (1965) 1 ~ 4, 226-244. 20 Altura, B.T., Brust, M., Badmm', R.L, Bloom, S~ Stempak, J.K. and Altura, B.M. (1990) Proc. Nat. Acnd. Sci. USA. 87, 18401844.
25
1992ElsevierScience PublishersB.V. All rights reserved0167-4889/92/$05.00
BBAMCR 13108
Magnesium regulates intracellular free ionized calcium concentration and cell geometry in vascular smooth muscle cells A i m i n Z h a n g a, T o n i P . O . C h e n g b a n d B u r t o n M . A l t u r a a a Department of Physiology and b Department of Anatomy and Cell Biology, State Unit~tsity of New yod¢. Health Science. Center at Brooklyn, Brooklyn, N Y (USA)
(Received 14 August1991)
Keywords: Magnesium;Calcium;Vascularsmoothmusclecell;Fura-2 Regulatory effects of extraceUular magnesium ions ([Mg2+]o) on intraceUular free ionized calcium ~[Ca2+~.) were studied in cultured vascular smooth muscle cells (VSMCs) from rat aorta by use of the fluorescent indicator fura-2 and digital imaging microf,copy. With normal Mg2+ (1.2 mM)-containing incubation media, [Ca2÷]i in VSMCS was 93.6+7.93 nM with a heterogeneous cellular distribution. Lowering [Mg2+]o to 0 mM or 0.3 mM (the lowest phlmiological range) resulted in 5.8-fold (579.5+39.99 nM) and 3.5-fold (348.0+31.52 nM) increments of [Ca2*]i, respectively, without influencing the ceHalar distribution of [Ca2+]i. Surprisingly, [Mg2÷]o withdgawal induced changes of cell geomeWy in many VSMCs, i.e., the cells rounded up. However, elevation of [Mg2+]o up to 4.8 mM only induced slight decrements of [Ca2+]i- (mean - 72.0 + 4.55 riM). The large increment of [Ca2+]i induced by [Mg2+]o withdrawal was totally inhibited when |Ca2+] o was removed. The data suggest that: (1) [M82+]o regulates the level of [Ca2÷]i in rat aortic smooth muscle cells, and (2) [Mg2+] acts as an iml~'tant regulatory ion by modulating cell shapes in cultured VSMc and their metabolism to control vascular contractile activities.
Introduction The importance of magnesium ions (Mg 2+) as cofactors for numerous enzlnnatic reactions and as a ubiquitous chelator of intracellular A T P has long been recognized, but there is growing evidence that Mg 2+ may also act as a regulatory ion [1,2]. In vascular smooth muscle cells (VSMCs), antagonistic effects of calcium ions (Ca 2+) and Mg 2+ have been established for contraction, and the interaction of Ca z+ and Mg 2+ is known to control vascular reactivity. Mg 2+ has been demonstrated to modulate basal tone, myogenic tone and contractile responsiveness of VSMCs to various physiological and pharmacological stimuli by modulating calcium content, binding and transport [3-5]. Mg z+ is thought to affect Ca 2+ flux across vascular muscle membranes and its release from internal stores, while an acute elevation or decrease of extracelluiar Mg 2+ concentration ([Mg2*]o) is known to exert rapid inln'b~-
Abbreviation:VSMCe,vascularsmoothmu.w.lecells. Correspondence:B.M. Altara, Beet31, Sun.v-HealthScience Center at Brooklyn,450 ~ Avenue,Brooklyn,NY II21D, U.S.A.
tion or potentiation of vascular contractile tension, respectively. Given its rather unique qualities compared to any known Ca 2+ channel blockers, Mg 2+ is believed to be a naturally occurring Ca 2+ antagenist
[6].
Since a change in ojtosolic free Ca z+ concentration ([Ca2+]i) is a very important second messenger in facilitation of contraction or relaxation [7], it may be physiologically relevant that Mg 2+ regulate the activity of VSMCs by competing with Ca 2+ and modulating the cytesolie free [Ca2+]i [6,8,9]. We repart here a novel effect of [Mg2+]o on regulation of [Ca2+]i and cell geometry in cultured VSMCs.
Materials t e t l Methods The distribution of [Ca2+]i in cultured VSMCs from rat aortas was determined using digital imaging fluorescence microscopy. Aortic smooth muscle cells from adult Wistar rats (200-250 g) were isolated according to previous e s t a ~ methods [10] and cultured in Dulbecco's modified Eagle's medimn at 37"C in a humidified atmosphere ~ of 95% air-5~£ CO 2. Cells for image ana~skt experintents were plated onto glass cover s l i ~ at a density of 1.0-10 s celis/ml. All
26 images were analyzed using a temperature-controlled Nikon fluorescence microscope and digitized by a TN8500 FluorPlex Image Analyzer (Tracer Northern, Madison), and using fura-2 as a fluorescent Ca 2+ indicator, which has a high selectivity for Ca 2+ over Mg 2+ and other divalent cations [11]. The coverslips containing a monolayer of cells were loaded with fura-2 by incubating them with 2 ~tM fura-2/AM in the culture media for 60 min under the above conditions. The coverslips were washed for 5 min with Hepes buffer solution (pH 7.4) containing 0, 0.3, 1.2 or 4.8 mM MgSO4, respectively. In addition to MgSO4, the Hepes buffer solution contained (mM): NaCI 118, KCI 4.7, CaCI 2 2.5, KH2PO 4 1.2, Hepes 5, glucose 10. The pH was adjusted to 7.4 with NaOH. The coverslips containing the fura-2-1oaded cells were affixed to holders (glass slides) that were placed on the temperature-controlled stage of the fluorescence microscope. The cells were excited, alternatively, at 340 n m and 380 nm; and the emission intensity was recorded at 510 nm, using a silicon intensified target (SIT) camera. The intensities of the recorded images at 340 and 380 n m were corrected by subtracting them from background fluorescence recorded at the corresponding wavelengths. The resulting images were then used to calculate [Ca2+] i in VSMCs, using external standards containing 2.54 mM Ca 2+ and 0 mM Ca 2+ plus 10 m M EGTA, a K d of 224 nM [11] for the f u r a - 2 / C a 2+ complex, and maximum (Rmu) and minimum (Rmi n) fluorescence ratios of the 340 and 380 nm-images. The fluorescent intensity in a single VSMC could be recorded accurately with high resolution. Visual examinations of fura-2 loaded cells revealed that a uniform fluorescence appeared when excited at 380 nm without any obvious centers of concentration, indicating to us a good distribution of the fura-2 throughout the treated cells. Results and Discussion When incubated with Hepes buffer solution containing 1.2 mM Mg 2+, the basal level of [Ca2+]i in cultured VSMCs, estimated from the ratio (F340/F380), is 93.2 + 7.9 nM (48 cells in four coverslips). The valt,cs varied among cells over a range between 14.1 nM and 177.7 nM, which is a quite comparable with basal values reported by others [12,13]. The resting distribution of [Ca2+]i , rather than the simple difference in
fura-2 concentration [13], also appeared heterogeneous within the cells by ratio image analysis (Fig. 1). Most of the VSMCs in the primary culture appeared spindleshaped (of. Fig. 1). The prenuclear region, arbitrarily chosen as the central ellipsoid area in the widest portion of the cell [14], revealed the most brightness, reflecting a relative higher [Ca2+]i than the peripheral cytoplasmic area. This is probably due to the fact that the pranuclear region contains most of the cellular organelles, such as the sarcoplasmic reticnlum (SR) a n d / o r rough endoplasmic reticulum (RER), which can sequester Ca 2+ by M~2~'-depandant proce.~es [5,13,14]. Therefore, the [C,a2+]i, we measured here, represents not only Ca 2+ in the cytosol, but also Ca 2+ sequestered in SR a n d / o r other intracellular Ca 2+ storge sites in VSMCs, and is probably an overestimate of the true [Ca2+]i. The lower [Ca2+]l in the peripheral cytoplasmic region may be relatively close to the true [Ca2+]i. The observation of a [Ca 2+] gradient in the cell suggests that the plasma membrane of VSMCs and the discrete localized processes regulate Ca 2+ activity subcellularly. In order to probe the effects of Mg 2+ on [Ca2+] i in VSMCs, the extracellular Mg 2+ concentrations ([Mg2+] o) in the incubation media (Hepes buffer solution) were altered. Such treatment did not, however, influence loading of the dye into the cells. Somewhat to our surprise, removal of extracellnlar Mg 2+ ([Mg2+]o z 0 mM), after only 5 min of washing VSMCs with Mg2+-free incubation media, induced a dramatic increment of [Ca2+]t up to 579.5 + 39.3 nM (91 cells from four individual coverslips), or about a 5.8-fold elevation compared to the control experiments ([Mg2+]oZ 1.2 mM) (Fig. 2). Concomitant with the latter, we noted large variations in changes of [Ca2+] i brought about by removal of extracellular Mg 2+ exhibiting a range from 137.7 nM to 2149.4 nM. Such rises of [Ca2+] i were maintained consistently, at least over a 30 min period of observation. To examine the [~2+]i changes produced under physiological conditions, [Ms2+] o was lowered to 0.3 mM, the lowest ionized physiological level seen clinically. This maneuver raised [Ca2+] i about 3.5-fold or to 348.0 + 31.5 nM (116 cells from five individual coverslips). However, the heterogeneous distribution of [Ca2+] i was still evident when [Mg2+]o was lowered to 0.3 raM, even though [Ca2+]t in the peripheral cytoplasmic region was in-
Fig. I. Ratio of 340/380 imagesof fura-2fluorescencein culturedvascularsmoothmusclecells incubatedin HEPES buffer solutionwith normal 1.2 mM Mg2+. The ratioed imageshowsthat [Ca2+]i is apparentlyhigherin the prenuclearregion.The numbersaround the color scale indicate the approximatenM concentrationsof free, ionizedintraceliolarCaz+. Fig. 2, Ratio of 340/380 imasesof fura-2 fluorescencein culturedvascularsmooth musclecells incubatedin MgZ+-froeHepes buffer solution. Note that the removal of [Mg2+]o raises [Ca2+]j both in the nuclear and the peripheral cytoplasmicregions, while the hetemgennons distributionof [Ca2+]l still persists.Someof the cultured cells dearly round up. The numbersaround the color scale indicatethe approximate nM concentrationsof fTee,ionizedintracellularCa2+.
F~.I.
28 TABLE I Effects of egtracellular magnesium on the regulation of intracellaior flee calcium ion in cultured vascular smooth muscle from rats
Values- means+S.E.; n - the number of cells; * Significantlydifferent from 1.2 mM MS2÷ (P<0.01). ** Significantly different from 1.2 mM MS2+ (P < 0.02). [MS+++],, (mM) 1.2 0 0.3 4.8 0
[Ca2+lo (mM) 2.5 2.5 2.5 2.5 0
n 48 91 !16 126 21
[Ca++]+ (nM) 93.6+ 7.90 579.5+39.3* 348.0+31.5* 72.0:1:4.55** 3.6+ 0.51*
creased. It is noteworthy that, simultaneously to the alteration of [ca2+]i, cell shapes were changed by lowering [Mg2+]o. Visual examinations revealed that many VSMCs were rounded up when [Mg2]o was with~ drawn indicating that these single cells can clearly change shape and undergo strong contractions in Mg2+-defieient media (el. Fig. 2), and that the 5.8-fold elevation of [ca2+]i is responsible for the latter events. Although the changes of cell geometry induced by contraction could influence the accuracy of [Ca2+] i measurements, the ratioed image, herein, reflects an absolute increase of [Ca2+]i . In contrast to lowering [Mg2+]o, elevation of [Mg2+] o to 4.8 mM (four times over normal [Mg2+] o) was found to decrease [ca2+]i to about 72.0 + 4.5 nM (126 cells from six individual coverslips), showing relatively less of an influence on [Ca2+]i compared to Mg 2+ removal (Table I). It is interesting to note that the intracellular heterogeneity of [Ca2÷] i WaS clearly less visible, and the brightness of the nuclear region became much weaker when [Mg2+]o was raised to 4.8 mM. These observations support our previous findings on intact and isolated blood vessels which indicate that reduction in [Ms 2+ ]o results in spasms of blood vessels, elevation of basal tone and increments in spontaneous mechanical activity [3,5,9]. The [Ca2+]i changes brought about by varying [Mg2+]o clearly indicate the dependence of [Ca2+]i and basal [ca2+] t upon [Ms2+] o. As shown in Table I, although there is relatively less of a change in [Ca 2+]i following changes of [Ms 2+]o above the normal physiological level ( > 1.2 mM), a reverse relationship of [Ca 2+ ]l vs. [Mg 2+ ]o was virtually exerted when [Mg2+] o is below this value, indicating for the first time a [Mg2+]o.dependent mechanism for regulating intracellular [ca2+]i distribution. To obtain further insights into the existence and n a t u r e of these regulatory processes, VSMCs were exposed 'to a modified Hepes buffer solution in which both Ca 2+ and Mg 2+ were removed. It was anticipated that such a maneuver would reveal the steady-state distribution of [ca2+]l a~'oss the sateolemma and
intracellular organelle membranes. A strong dependence of the [Mg2+]o-induced increment of [Ca2+] i upon [Ca2+]o was observed as exemplified by the marked lowering of [Ca2+] i (to 3.6 nM) in such treated cells. This represents about a 200-fold decrement compared with cells treated by [Mg2+]o withdrawal alone. The [ca2+] i gradient completely disappears in the absence of [Ca2+]o. Although the removal of extracellular Ca 2+ and Mg 2+ may increase membrane permeability and consequently result in an ebbing away of fura-2 from VSMCs via membrane leakage, as indicated by the decrement of fluorescent intensity at 380 nm, the sharp drop in [Ca 2÷] observed here reveals that the entry of external Ca 2+ occurs primarily in resting VSMCs when [Mg2+]o is lowered, and that alterations of [Ca2+] i induced by Mg 2+ seem to be reversible. However, such dye leakage induced by the removal of [Ca2+]o and [Mg2+]o, on the other hand, suggests that both divalent cations are required for maintaining membrane permeability and integrity of the cell membrane. The experiments described above demonstrate that Mg 2+ exerts profound influences on [Ca2+], in cultured VSMCs. At present, the mechanism by which [Mg2+]o regulates [Ca2+]i is unclear. The alterations in [Ca2+] i in this cell type could be a consequence of fluctuations either in the entry of external Ca 2+ or in its release from internal stores. However, both of these fluctuations could be affected by [Mg2+] o 69,15]. Since the heterogeneous distribution of [Ca2+]l is hardly affected by [Mg2+]o, it could be suggested that the intracellular compartment of Ca 2+ may not play a major role in these events, in spite of data which indicates that Mg 2+ is known io regulate Ca 2+ release and sequestration in VSMCs. The dependence of [Ca 2+ ]i on [Ca 2+ ]o, however, suggests that the mechanism whereby Mg 2+ regulates [Ca2+] t in VSMCs is linked mainly to a modulation of membrane permeability to [Ca2+]o. Lowering [Mg2+]o is known to inaease 4sea influx and efflux [3], as well as increase exchangeable and membrane-bound Ca 616], whereas high [Mg 2+]0 exerts opposite effects. Therefore, alterations in [Ca2+]i could be a consequence of Ca2+ entry governed by Mg2+-controlled membrane permeability. Previous experimental findings indicate that, unlike most Caz+ channel antagonists, Mg z+ can act on voltage-dependent 62,5] and receptor-operated channels [5], as well as leak.operatea channels in the resting state 65,16]. With regard to cellular Mg contents which seem to be well-below electrochemical equilibrium, it has been suggested that there are active Mg transport systems in the cells [17,18], such as M g 2 / M g 2+ exchange, Mg2+/Ca 2+ exchange or N a + / M g 2+ exchange, by which Ca2+ may compete with Mg 2+ for transport sites. The present, new findings indicate that Mg 2+ may regulate membrane permeability to [Ca2+]o
29 by a novel mechanism r a t h e r than as a simple blocking particle on a n open channel or a simple m e m b r a n e binding ion. Alternatively, active M g transport systems may play a major role as a result of [Mg2+]o regulation of both influx and efflux of Ca 2+ [3]. W h e t h e r or not changes in [Mg2+] o can alter, rapidly, intracellular free M g 2+ concentration ([Mg2+]i) in VSMCs of the rat aorta is at present not known. F u r t h e r studies are thus n e e d e d to clarify these events and to elucidate which one of these parameters, i.e., [Mg2+]o, [Mg2+] i or [Mg2+]o/[Mg2+]i, is most important for controlling m e m b r a n e permeability to [Ca2+]o in V~MCs. The present finding of a fine regulation of [Ca2+]i by [Mg2+] o may have several important implications. Irrespective of the precise m e c h a n i s m s whereby [Mg2+]o controls Ca 2+ entry and cell geometry, changes of [Mg2+] o could induce substantial long-term potentiation or inhibition of vascular reactivity and basal tone. Clinically, M g deficiency is associated with a n u m b e r of cardiovascular disorders [3,19]. These include hypertension, stroke, cardiac arrbythmias, cardiac ischemia and myocardial infarction, as well as atherosclerosis [20], M a n y of these pathological processas are associated with vascular spasm [19], a consequence of h y p o m a g n e s e m i a which could result in increm e n t s of [Ca2+] i in VSMCs by the mechanism described herein.
~
t
This work was s u p p o r t e d in part by A D A M H A G r a n t AA-08674 t o B.M.A.
References I Grubbs, R.D. and ~ ME. (1987) Maanesiam 6,113--127. 2 White, R.E. and Hastzen, EI.C.. (1989) Biochem. Pharmacol. 38, 859-867. 3 Altura, B.M. and Altura, B.T. (1961) Fed. Pm~ 40, 2672-2679. 4 Aitura, B.M., Altura, B.T, Carella, A. and TedalmtY, P.D.M.V. (1982) Can. J. Physiol. Phanaacel. 60, 459-4~. 5 Altura, B.M., Altma, B.T., Camlht, A., GelMewoM, A., Murakawa, T. and Nishio, A. (1997) Can J. ~ PharmacoL 65, 729-745. 6 Altm'a, B.M. and Altura, B.T. (1981) In New Pegspectiv~ on Calcium Antagonists (Wei~ G.B., ed.), pp. 131-145, Am. Phr~ iol. Soc., Washington. 7 Kuriyama, H., Ito, Y, Suzuki, H, Kitamera, If. and ltolk IL (1982) Am. J. Physiol. 243, I-I641-He6Z 8 Altura, B.M. and Altura, B.T. (1971)Am. J. ~ 220, 938-944. 9 Altura, B.M. and Altura, B.T. (19/4) ~ Res. 7,145-155. 10 Yamamoto, H., ~ a a k k , H., Nalumm'a, M. (1~3) Be. J. Exp. Pathol. 64,156-165. II Grsnkiewicz, G., Poeaie, M. and Tslon, R.Y. (1~5) J. Biol. Chew- 260, 3440-3450. 12 Nabel, E.B,, Berk, B.C., Brock T.A, and Smith T.W. (1988) Circ. Res. 62, 486-493. 13 Goto, A., Yamada, If.,hhii, M, Ymhleb, M,, hhit~ro, T, Eguchi, C and Sulgimoto,T. (1989) ~ e m i o n 13, 916-921. 14 Papngeorsion, P. and Mecgan, K.G. (1989) Prec. Soc. ~ Med. 193, 331-334. 15 Sjfgmn, A. and Edvimao~ L (1~8) Pharmm:oL TmicoL 62, 17-21. 16 Turlapa~, P.D.M.V. and Altura, B.M. (1978) Ear. J. PhmmacoL 52, 421-423. 17 Flatman, P.W. (lC~4) J. Membr. Biol. 80,1-14. 18 Altura, B.M. and Altma B.T. (1981) ~ B u l k t i a It, 102-114. 19 Altura, B.M. and Altwa, B.T. (1965) 1 ~ 4, 226-244. 20 Altura, B.T., Brust, M., Badmm', R.L, Bloom, S~ Stempak, J.K. and Altura, B.M. (1990) Proc. Nat. Acnd. Sci. USA. 87, 18401844.