Cd/Ca/dum
(WQO)ll,
0143-4160,90/0011-03@71$10.00
397-403
QLongmanGrcupUKLidlQ30
The effects of IWIg*+on rat liver microsomal Ca*+ sequestration G.H. ZHANG and N. KRAUS-FRIEDMANN
Depaftment of Physiology and Cell Biology, University of Texas Medical School, Houston, Texas, USA Abstract - The effects of Mg2+ on the hepatic microsomal Ca2+-sequestering system was tested. Ca2+-ATPase activity and Ca2+ uptake were both dependent on the concentration of free Mg2+, reaching maximum levels at 2 mM. The effects of Mg-ATP were also influenced by the concentration of free Mg2; being maximally effective at a ratio of 1:i. The results suggest that Mg2’ influences Ca ’ sequestration at various steps, namely in addition to forming the substrate of the Ca2+-ATPase reaction, Mg-ATP, Mg2+ stimulates the reaction at an additional step, as indicated by its stimulatory effect on the Ca2+-ATPase reaction and on Ca2+ uptake, even at optimal Mg-ATP levels. The stimulatory effect of Mg2+ was evident at various pH levels tested, and it was nucleotide specific. The stimulatory effect of Mg2+ might be exerted at the dephosphorylation step of the enzymatic reaction or at an other, yet undefined, site. The results demonstrate a plural effect of Mg2+ on the hepatic microsomal sequestration system. This indicates that, dependinn on its magnitude, changes in Mg2+ distribution might influence cytosolic Ca2’ levels. Calcium sequestration in the endoplasmic reticulum is a physiologically important process contributing to the regulation of cytosolic Ca2+ levels [l]. The sequestration of Ca2+ reflects the balance between Abbreviations : adenosine S-triphosphate; CIP, cytidine ATP, S-triphosphate; GTP, guanosine S&phosphate; ITP, inosine S-triphosphate; UTP, midine S-triphosphate; ethylene EGTA, glycol bis(oxyethylenenitrilo) N,N,N’,N’-tetraacetic acid; MES. 2-(N-morpholino) ethane erazine sulfonic acid, HEPES, 4-(2-hydroxyethyl)-l-pi K ethane sulfonic acid, p[Mg*‘]f, -log (free Mg wncentration); p[Mg-A’IP], -log (Mg-ATP concentration); NTPase, nucleotide 5’-triphosphatase. or adenosine, cytidine, guanosine, inosine or uridine 5’-triphosphatase; CHAPS, 3-[(3-cholamidopropyl) dimethylammonio]-lpropane sulfonate.
two processes: Ca2+ uptake, which is catalyzed by the Ca2’ pump, and the release of Ca2’ through Ca2+ channels. The Ca2+-ATPase activity in the endoplasmic reticulum-derived microsomal fraction has been characterized and found to be similar to the corresponding enzymes in cardiac and skeletal muscles. Thus, the hepatic enzyme also forms a phosphorylated intermediate as part of the reaction sequence which is hydroxylamine labile [2, 31. The molecular weight of the enzyme is around 100-107 kD and the enzyme cross reacts with antibodies raised against the purified Ca2+-ATPase of skeletal muscle SR [4, 51. The hepatic Ca2+-ATPase was recently purified to homogeneity using conventional methods [5]. 397
cJ5LLscALcIuM
398
201 A
Fig. 1 Effect of free Mg’+ on Ca”
uptake and Ca”-ATPase activity. Ca2’ uptake and Ca’+-ATPase activity were determined as described in Materials and Methods. The free Ca2’ and Mg2+ concentrations were 20 p.M and 10 pM - 10 mM, respectively. Mg-ATP concentration WBS1 mM in all determinations. In the Ca2+ uptake experiment, the reaction was started by addition of ATP, and the total Ca2’ uptake was deWed when the Ca2’ concentration reached a plateau by addition of 1 pM ionophore A23187 to release all of the Ca2’ tramported into the vesicles. The data presented are means f SEM of 3 separate experiments using different preparations of microsomes. Open symbols represent Ca2+ uptake, filled symbols Ca2+ATPase activity
In contrast to the relatively abundant information available on the Ca2+-pump, very little is known about the Ca2’ channels present in the endoplasmic reticulum. We recently described the activation of Ca2+-release from the endoplasmic reticulum by ryanodine and -SH reagents 16.71. Though second messengers are known to release Ca2+ from the endoplasmic reticulum in whole cell preparations, their addition to isolated microsomal fractions does not result in a reproducible effect as it does in the It is well demonstrated in cellular preparation. skeletal and cardiac muscles, that Mg2+ has an effect on both aspects of Ca2’ sequestration. It is
required for the Ca2+-pump and it affects the function of Ca2+ channels. Magnesium was shown to affect several steps ir+the ATPase sequence cycle [S-l 11. Howeve;; Mg is not counter transported durinEj+active Ca accumulation [12]. Th;+activity of Ca -channels also is influenced by M Thus, %+ in cardiac and skeletal muscles Mg is an important modifier of the Ca2+ sequestration process 113, 141. The possibility that Mg2’ also might have a role in modifying hepatic Ca2’ sequestration is raised by observations demonstrating that Mg2’ fluxes are altered in the liver after the administration of different agonists [15, 161. Though these observations did not resolve whether changes in Mg2+ ‘distribution are functionally linked to the changes in Ca2+ distribution or not, this possibility cannot be excluded. Therefore, it was important to examine the effects of Mg2+ on hepatic microsomal Ca2’ sequestration, a key factor in cellular Ca2’ responses.
Materialsand Methods Chemicals ATP, Cl?, GTP, ITP, UTP, EGTA, HBPES, MBS and DTT were purchased from Sigma (St Louis, MO, USA). 45CaC12 was purchased from Amersham (Arlington Heights, IL, USA). All other reagents were of the highest purity. Microsomalpreparation Livers from male fed Sprague-Dawley rats were homogenized in a medium composed of 250 mh4 sucrose, 20 mM HBPBS (pH 7.4), 1 mM EGTA and 1 mM D’IT. The microsomal fraction was prepared a8 previously described [2, 171. In short, the homogenate was centrifuged at 1100 g for 10 min in a SS34 rotor of a Sorvall RC2-B centrifuge. The supematant was centrifuged at 7700 g for 20 min in The microsomal fraction was the same rotor. sedimented from the 7700 g supematant by centrifugation at 110,000 g for 60 min using a Ti 70 rotor in a Beckman L5-50 uhracemrituge. The microsomal fraction was resuspended in 100 mM
Mg2+ EFFECTS ON RAT LIVER MICROSOMAL Ca2+ SEQUESTRATION
KCl, 20 mM HEPES, pH 6.8, to a final protein concentration of 20-30 w/ml. We determined the plasma membrane aad contamination of mitochoadria b the assay of their respective marker Y enzymes, Na -K+-ATPase [18] and succiaate dehydrogeaase [19]. The contaminations of the microsomal preparation with plasma membrane or mitochoadria was less than two percent of the protein. Measurement
ofCa2+ transport
Ca2’ uptake by the microsomes (0.5 rag/ml) was determined in a medium composed of 100 mM KCl, 20 mM HEPES, pH 6.8 with a Ca2’ electrode (Orion, Model 93-20, Cambridge, MA, USA) at 37°C. In some experiments, 20 mM MES was used instead of HEPES to buffer the solution to pH 6.0-6.7. The Ca2+ electrode was calibrated usi;$ Ca-EGTA buffers of known ionized Ca which concentrations, were prepared and standardized using a Ca2+ standard solution purchased from Orion Associates Inc. The free, Ca2+, Mg2’ and Mg-ATP concentrations were calculated by a computer program using the constants described by Fabiato et al. [20] as previously described [211. NTPase assay
Ca2+-NTPase activity was determined using the method previously described [22]. The reaction medium containing 100 mM KCl, 20 mM HEPES (pH 6.8) 1 mM ouabaia, 1 raM NaN3, 1 pM ioaophore A23187 and 0.5 mg/ml of microsomal protein. The microsomes were pieincubated at 37°C for 10 min. The reaction was started by addition of either ATP, CTP, GTP, ITP or UTP. After the reaction at 37°C for 10 mia, the ice-cold trichloroacetic acid (final concentration 7%) was added to stop the reaction. The protein-free supemataat was obtained by centrifugatioa at 3500 g for 5 min and assayed for Pi concentration [23]. Mannose-bphosphatase
assay
Maanose-6-phosphatase was determined as described by Vaastapel et al. [24]. The incubation
399
mixtare coataiaed 250 mM sucrose, 100 mM imidazole, pH 6.5. Reaction was started by addition of 1 mM rnaaaose-6-phosphate. Assays were performed for 5 mia at 20°C aad quenched by addition of ice-cold 8% trichloroacetic acid. The protein-free supemataat was assayed for Pi [23]. Protein assay
Protein concentration was measured as described by Smith et al. [25] using bovine serum albumin as standard.
Results
The effects of free Mg2’ on Ca2+ uptake and Ca2+-ATPase activity are shown in Figure 1. The uptake of Ca2+ increased . with the free Mg2+ concentration and plateaued at the free Mg2+ concentration of 1 mM (Fig. IA). This response is consistent with the physiological concentration of free Mg2+ of around 1 mM [26]. The Km of the Ca2’ uptake was 0.4 mM of free Mg2’. The Ca2’-ATPase activity showed a somewhat different response to the changes in free Mg2+ (Fig. 1B). First, the ATPase activity decreased when free Mg2’ concentration was higher than 2.5 mM, sad second, the peak of response was at a free Mg2+ concentration of 2.5 mM, whereas the peak of Ca2’ uptake was at a Mg2+ concentration of 1 mM. It has been su ested that Mg-ATP is the main g substrate of the Ca -ATPase. The effect of varying Mg-ATP concentrations on Ca2’ u take is shown in Figure 2A. The uptake of Ca2P increased with increasing Mg-ATP c;nceatratioas, and reached its peak where the [M~+IdlMg-ATP] ratio was 1. After this point, Ca uptake decreased despite further increases ia Mg-ATP concentration (Fig. 2A). Ca2+-ATPase activity showed a similar response to Mg-ATP at the higher free Mg2+ concentrations. At a low concentration of free Mg2+ (0.1 mM), the saturation of Ca2+-ATPase activity was not apparent. The biphasic response to Mg-ATP was correlated with the corresponding free ATP concentrations. The uptake of Ca2+ increased until the free ATP level reached 0.1 mM. At higher
CELL CALCIUM
400
l 0 [Mg2+], = 2.5mM = q [Mg2+], = 0.5mM A A [tvlg2+J,= O.lmM lO[Mg2+],= 0.02mM
mM free Mg2+, the Ca2+ uptake induced by ATP increased 16 times and the Ca2+ uptake induced by CTP, GTP, ITP and UTP increased about 2 times (Fig. 3,A). The changes in Ca2+-NTPase activity induced by Mg2+ showed some differences amon 28 the various nucleotides. At 0.02 mM free Mg Ca2+-NTPase activities were similar, but at 2.5 mh; free Mg2+, the ATPase activity increased 2.3 fold, UTPase increased 78%, while CIFase, GTPase and ITPase activity decreased 42%, 23% and 60% respectively (Fig. 3B). Because the Ca2’ uptake and Ca2+-ATPase activi are pH-dependent, we determined the effect Yt of Mg on the pH dependence of Ca2+ uptake and Ca2+-ATPase. As shown in Figure 4A at high concentration of free Mg2’ (2.5 mM), Ca2’ uptake reached the maximum at pH 6.8, and decreased thereafter. At a low concentration of free Mg2’ (0.038 mM), Ca2+ uptake increased continuously
251
OJ 5.0
4.5
4.0
3.5
3.0
2.5
ATP CTP GTP ITP UTP
ATP CTP GTP ITP UTP
B
,W’W, Fig . 2
A
Effect of Mg-ATP and A’IIQ on Ca2’ uptake and
Ca2+-ATPase activity.
Ca2+ uptake and Ca2+-ATPase activity
were determined as in Figure 1. The data presented NIXmeans f Open SEM of 3 experiments using different micmsomes. symbols represent Ca2+ uptake, filled symbols Ca2’ ATPase activity ATP CTP GTP ITP IJTP
concentrations of free ATP, Ca2+ uptake was inhibited. The inhibition by ATP was more pronounced at higher Mg2+ concentrations (Fig. 2C). The nucleotide specificity at low and high Mg2+ concentrations was also tested (Fig. 3). We found that at 0.02 mM free Mg2+, ATP, CTP, GTP, ITP and UTP did not stimulate Ca”’ uptake, but at 2.5
[Mg2+] t= 0.02mM
ATP CTP GTP
ITP UTP
[Mg2+] t= 2.5mM
Fig. 3
Effect of Mg2+ on the substrate specificity of the Ca2’ The Ca2’ uptake and Ca2+-NTPase activity were P-P. The free Ca2’ and Mg-NTP determined as in Figure 1.
concentration were 20 p.M and 1 mM. respectively.
The data
presented are means f SEM of 3 separate experiments using different microsomal preparations
Mg2+ EFFECTS
ON RAT LIVER MICROSOMAL
Ca2t SEQUESTRATION
o o [Mg2+], = 2.5mM n q [Mg2+lt = 0.038mM
6.0
6.5
7.0
7.5
8.0
401
effect of Mg2+ on membrane leakiness was tested (Table 1). As an indicator of membrane integrity of the intravesicular enzyme, the activity mannose-6-phosphatase was determined in intact and detergent treated vesicles, as described by Vanstapel et al. [24]. As shown in Table 1, Mg2’ had no effect on membrane leakiness as indicated by the lack of difference in mannose-6-phosphatase activity in the presence and absence of Mg2+.
8.5
Discussion
6.0
’ 7.0
6.5
’ 7.5
’ 6.0
815
PH Fig. 4 Effect of Mg” Ca*+-ATPase
activity.
were determined was buffered
Ca”
Ca”
as in Figure
uptake
of Ca2’ uptake and
and Ca*+-ATPase
at pH 6.8-8.2
by 20 mM HEPEZS. The free Ca2’ and Mg-ATP were 20 pM and 1 mM, respectively.
The data on
uptake are n~eans f SEM of 3 separate experiments
data on Ca2+-ATPase experiments
activity
1. The reaction mixture at pH 6.0
by 20 mM MES, and the solutions
were buffered concentration
on the pH dependence
activity
and the
are means f SEM of 9 separate
using diKerent microsomal
preparations
The2ysults presented in this study demonstrate that has multiple effects on hepatic microsomal Fjt sequestering: (a) free Mg2 is essential for Ca2+-ATPase activity and Ca2’ uptake; this effect of free Mg2+ is not due to Mg-ATP being the substrate for the Ca2+-ATPase, but additional to it: and (b) Mg2+ also seems to affect a Ca2’ efflux pathway. The results illustrated in the experiments seen in Figures l-3, clearly demonstrate that Mg-ATP is essential for Ca2+-ATPase activity as is expected from its role as a substrate for the Ca2+-ATPase reaction. However, at all the Mg-ATP concentrations employed, increasing [Mg2+] to the iological, millimolar range, stimulated both the . . PCh”-ATPase acuvity and the uptake of Ca2’ . A possible mechanism by which free Mg2’ stimulates the activity of the enzyme is the one described for the sarcoplasmic reticulum enzyme, namely at the dephosphorylation step of the phosphoenzyme
with pH (Fig. 4A). However, the absolute amount Table 1 Effectsof Ca2’ and Mg2+on microsomal of Ca2+ taken up in low Mg2+ is less than 15 membrane permeability percent of the amozu+nttaken up in the presence of Similar1 concentrations. phxsio1ogica.l Mg Additions M-6-Pase activity J-G Ca -ATPase activity was lower in low Mg (nmol/m&nin) concentration and the enzyme activity decreased even further at basic pH (Fig. 4B). Control 1.47 f 0 192.95 f 6.64 In all the above experiments when Ca2+ uptake CHAPS Ca2+ 1.96 f 1.04 was measured, the net uptake of Ca2+ reflected the Ca2+ Mg2+ 2.82 f 1.24 balance between Ca2+ uptake due to the activity of the Ca2+ pump and the efflux of Ca2’ through Mannose-6-phosphatase (M-6-Pase) activity was de&mined as specific channels or non-specific membrane described in Materials and Methods. Control contained the ma&ion but&r only. Total activity was measuredinthepresence leakiness. In order to examine whether Mg2+ of 0.2% CHAPS whi& permeabilized the vesiclea. C!a* and affects Cazt efflux directly through Ca2’ channels Mg* were added tc give a final concen~ticm of 5 n&l. The results were means 5 SEih4 of 3 experiments or indirectly by increasing membrane leakiness, the
402
intemediate 18, 27, 281. At o~+t$al levels of [Mg-ATP], the effect of free Mg IS consistently stimulatory. The effect of Mg-ATP seems to be complex (Fig. 2) because it is of a biphasic nature, stimulatory at the lower and inhibitory at the higher concentrations. This effect of Mg-ATP seems to be the reflection of the associated changes in the concentration of free ATP. An examination of Figure 2B reveals that above 0.1 mM, further increases in Eree ATP are associated with a decrease in Ca2+ uptake. This effect of free ATP might be due to its effect on membrane permeability as has been suggested for the sarcoplasmic reticulum vesicles. It was shown by Chiesi and Wen that when the Ca2’-ATPase is phosphorylated in the absence of Mg2+ it forms a stable, phosphorylated intermediate [29]. Addition of ATP under this condition results in rapid Ca2’ release. Shoshan-Barmatz [30] reported that preincubation of sarcoplasmic reticulum vesicles with ATP inhibited subsequent Ca2’ uptake, an effect which was specific to free ATP. Meissner and Henderson [31] also reported on ATP induced Ca2’ release. The inhibitory effect of free ATP on Ca2’ above the concentration of 0.1 mM might be due to any of the above mechanism(s) or to some different, yet undefined, cause. It was outside the scope of this study to establish the mechanism(s) by which hi h St of free ATP prevent Ca concentrations It seems clear, however, that the accumulation. changes in [ATP]r are the underlying mechanism responsible for the biphasic responses to the changes in [Mg-ATP] levels. Additional effects of free Mg2+ are illustrated in Fi ure 3 where the nucleotide specificity of the CaSt -ATFQse reaction and Ca2’ u{p are shown. concentration As shown, increase in free Mg renders to the enzyme the ability to pump Ca2’. This effect might be due to the stimulation of the dephosphorylation step or at an additional site. The inhibitory effect of vanadate on the enzyme is also Mg2+ dependent as shown previously by Dawson and Fulton 1321. The stimulatory effect of Mg2+ on Ca2’ uptake is evident in the pH range of 6.0 - 8.2 (Fig. 4). As described originally by Moore et al. [33], Ca2+ uptake is pH sensitive and it declines at pH higher
CELL CALCIUM
than 6.8. Because the Ca2+-ATPase activity peaks at pH 7 5 the sharp decline in Ca2’ uptake observed betweed ;H 6.8 - 7.5 is likely to be caused by stimulation of Ca2+ efflux. The pronounced effects that pH has on various cellular processes involving Ca2+ are well known, as is the Ca2+ releasing effect of alkalinization [34-361. A Mg2+ induced inhibition of the Ca2+-ATPase at high pH in the sarcoplasmic reticulum was recently described [37]. However, in the present experimental conditions at low Mg2+ concentrations the enzyme activity was already very low so that possible additional pH effects were not evident. In summary, we present evidence showing that Mg2+ exerts a profound effect on microsomal Ca2’ sequestration. The physiological significance of these findings mi ht depend on the magnitude of the 2! changes in Mg distribution following hormonal stimulation of the liver.
Acknowledgements Supported by PHS grant NDDK-DK36916. The valuable suggestions of V. Shoshan-Bannatz and the critical reading of the manuscript by N. Karin were greatly appreciated.
References 1. Camfoli E. (1987) Intracellular calcium homeostasis. Annu. Rev. Biochem., 56.395-433. 2. Fleschner CR. Kraus-Friedmann N. Wibert GJ. (1985) Phosphorylated intermediates of two hepatic microsomal ATPases. Biochem. J., 226, 839-845. 3. Heihnann C. Spamer C. Gerok W. (1985) The calcium pump in rat liver endoplasmic reticulum. Demonstration of the phosphorylated intermediate. J. Biol. Chem., 260, 788-794. 4. Damiani E. Spamer C. Heihnann C. Salvatori S. Magreth A. (1988) Endoplasmic reticulum of rat liver contains two proteins closely related to skeletal sarcoplasmic reticulum Ca-ATPase and calsequestrin. J. Biol. them., 263, 340-343. 5. Jong YJ. Sheldon A. Kraus-Friedmann N. (1990) Purification of the hepatic microsomal Ca2+-ATPase from rat liver. (Submitted). 6. Shoshan-Bannatz V. Kraus-Friedmann N. (1990) High affinity ryanodine binding site in rat liver endoplasmic reticulum. (Submitted). 7. Zhang GH. Yamaguchi M. Kimura S. Higham S. Kraus-Friedmann N. (1990) Effects of heavy metals on the microsomal Ca2’-ATPase and Ca2” sequestering. Relation to SH-groups. J. Biol. Chem., 265,2184-2189.
Mg2+ EFFECTS ON RAT LIVER MICROSOMAL
Ca2+ SEQUESTRATION.
8. Martonosi AN. (1984) Mechanisms of Ca” release from satcoplasmic reticulum of skeletal muscle. Physiol. Rev., 64, 1240-I 319. 9. Champeil P. Gingold MP. Guillain F. Inesi G. (1983) Effects of magnesium on the calcium-dependent transient kinetics of sarcoplasmic reticulum ATPase, studied by stopped flow fluorescence and phosphorylation. J. Biol. Chem., 258,4453-4458. 10. Orlowski S. Lund S. Moller J. Champeil P. (1988) Phosphoenzymes formed from Mg.ATP and Ca.ATP during pm-steady state kinetics of sarcoplasmic reticulum ATPase. J. Biol. Chem., 263, 17576-17583. 11. Asturias FJ. Blasie JK. (1989) Effect of Mgzf concentration on Ca2+ uptake kinetics and structure of the sarcoplasmic reticulum membrane. Biophys. J., 55, 739-753. 12. Salama G. Scarpa A. (1985) Magnesium permeability of samoplasmic reticulum. Mg2+ is not countertransported during ATP-dependent Ca2+ uptake by sarcoplasmic reticulum. J. Biol. Chem., 260, 11697-11705. 13. Agus ZS. Kelepouris E. Dukes I. Morad M. (1989) Cytosolic magnesium modulates calcium-channel activity in mammahan ventricular cells. Am. J. Physiol., 256, C452C455. 14. White RE. Hartzell HC. (1989) Magnesium ions in cardiac junction. Regulator of ion channels and second messengers. B&hem. Pharmacol., 38,859-867. 15. Bond M. Vadasx G. Somlyo AV. Somlyo AP. (1987) Subcellular calcium and magnesium mobilization in rat liver stimulated in vivo with vasopressin and glucagon. J. Biol. Chem., 262, 15630-15636. 16. Jacob A. Becker J. Schottli G. Fritzsch G. (1989) aladrenergic stimulation causes Mg2+ release from perfused rat liver. FEBS Lett., 246, 127-130. 17. Zhang GH. Kraus-Friedmann N. (1989) Demonstration of adenylate kinase activity in hepatic micmsomes. Relevance to Ca2’ uptake. B&hem. Int., 19, 333-343. 18. Shamchmidt BF. Keeffe EB. Blankenship NM. Ockemer RB. (1979) Validation of a recording spectrophotometric method for measurement of membrane-associated Mg- and Na/K-ATPase activity. J. Lab. Clin. Med., 93, 790-799. 19. Veeger C. Der Vartanran DV. Zeylemaker WP. (1968) Succinate dehydrogenase. Methods Enzymol., 13, 81%. 20. Fabiato A. Fabiato F. (1979) Calculator programs for computing the composition of the solutions containing multiple metals and ligands used for experiments in skinned muscle cells. J. Physiol. (Paris), 75, 465-505. 21. Fleschner CR. Kraus-Friedmann N. (1987) Inhibition of microsomal Ca2+-ATPase by fluorescein 5’ isothiocyanate. Arch. B&hem. Biophys., 254,448-453. 22. Fleschner CR. Kraus-Friedmann N. (1986) The effect of Mg2+ on hepatic microsomal Ca2’ and Sr+ transport. Eur. J. B&hem., 154,313-320. 23. Lanzetta PA. Alvarez L. Reinach PS. Candia DA. (1979) An improved assay for nanomole amounts of inorganic
403
phosphate. Anal. Biochem., 100,95-97. 24. Vanstapel F. Pua K. Blanckaert NB. (1986) Assay of mannose&phosphate in untreated and detergent-disrupted rat-liver microsomes for assessment of integrity of microsomal preparations. Eur. J. B&hem., 156,73-77. 25. Smith PK. Krohn RI. Hermanson GT. et al. (1985) Measurement of protein using biconchoninic acid. Anal. B&hem., 150,76-85. 26. Murphy E. Steenbergen C. Levy LA. Raju B. London RE. (1989) Cytosolic free magnesium levels in ischemic rat heart. J. Biol. Chem., 264, 5622-5627. 27. Inesi G. (1985) Mechanism of calcium transport. Annu. Rev. Physiol., 47,573~601. 28. Vilsen B. Andersen JP. (1987) Effects of phospholipid, detergent and protein-protein interaction on stability and phosphoenzyme isomerixation of soluble sarcoplasmic reticulum Ca-ATPase. Eur. J. B&hem., 170.421-429. 29. Chiesi M. Wen YS. (1983) A phosphotylated conformational state of the (Ca2+-Mg’+)-ATPase of fast skeletal muscle sarcoplasmic reticulum can mediate rapid Ca” release. J. Biol. Chem.. 258, 60786085. 30. Shoshban-Barmatz V. (1987) Stimulation of calcium efflux from sarcoplasmic reticulum by preincubation with ATP and inorganic phosphate. B&hem. J., 247, 497-504. 31. Meissner G. Henderson JS. (1987) Rapid Ca” release from cardiac sarcoplasmic reticulum vesicles is dependent on Ca2’ and is modulated by Mg”’ adenine nucleotide, and calmodulin. J. Biol. Chem., 262, 3065-3073. 32. Dawson AP. Fulton DV. (1983) Some properties of Ca2+-stimulated ATPase of rat liver microsomal fraction. B&hem. J., 210,405-410. 33. Moore L. Chen T. Knapp HR. Landon El. (1975) Energy-dependent calcium sequestration activity in rat liver microsomes. J. Biol. Chem., 250, 4562-4568. 34. Sorenson MM. De Meis L. (1977) Effects of an ion, pH and magnesium on calcium accumulation and release by samoplasmic reticulum vesicles. Biochim. Biophys. Acta, 465,210-223. 35. Connett RI. (1978) Association of increased pH with calcium ion release in skeletal muscle. Am. J. Physiol., 234, CllO-c114. 36. Busa WB. Nuccitelli R. (1984) Metabolic regulation via intracellular pH. Am. J. Physiol., 246, R409-R438. 37. Bishop JE. AlShawi MK. (1988) Inhibition of sarcoplasmic reticulum Ca2+ -ATPase by Mg2+ at high pH. J. Biol. Chem., 263, 1886-1892. Please send reprint requests to : Dr N. Kraus-Friedmann, Department of Physiology and Cell Biology, University of Texas Medical School, P.O. Box 20708, Houston, TX 77225, USA. Received : 14 December 1989 Revised : 9 February 1990 Accepted : 9 March 1990
(WQO)ll,
0143-4160,90/0011-03@71$10.00
397-403
QLongmanGrcupUKLidlQ30
The effects of IWIg*+on rat liver microsomal Ca*+ sequestration G.H. ZHANG and N. KRAUS-FRIEDMANN
Depaftment of Physiology and Cell Biology, University of Texas Medical School, Houston, Texas, USA Abstract - The effects of Mg2+ on the hepatic microsomal Ca2+-sequestering system was tested. Ca2+-ATPase activity and Ca2+ uptake were both dependent on the concentration of free Mg2+, reaching maximum levels at 2 mM. The effects of Mg-ATP were also influenced by the concentration of free Mg2; being maximally effective at a ratio of 1:i. The results suggest that Mg2’ influences Ca ’ sequestration at various steps, namely in addition to forming the substrate of the Ca2+-ATPase reaction, Mg-ATP, Mg2+ stimulates the reaction at an additional step, as indicated by its stimulatory effect on the Ca2+-ATPase reaction and on Ca2+ uptake, even at optimal Mg-ATP levels. The stimulatory effect of Mg2+ was evident at various pH levels tested, and it was nucleotide specific. The stimulatory effect of Mg2+ might be exerted at the dephosphorylation step of the enzymatic reaction or at an other, yet undefined, site. The results demonstrate a plural effect of Mg2+ on the hepatic microsomal sequestration system. This indicates that, dependinn on its magnitude, changes in Mg2+ distribution might influence cytosolic Ca2’ levels. Calcium sequestration in the endoplasmic reticulum is a physiologically important process contributing to the regulation of cytosolic Ca2+ levels [l]. The sequestration of Ca2+ reflects the balance between Abbreviations : adenosine S-triphosphate; CIP, cytidine ATP, S-triphosphate; GTP, guanosine S&phosphate; ITP, inosine S-triphosphate; UTP, midine S-triphosphate; ethylene EGTA, glycol bis(oxyethylenenitrilo) N,N,N’,N’-tetraacetic acid; MES. 2-(N-morpholino) ethane erazine sulfonic acid, HEPES, 4-(2-hydroxyethyl)-l-pi K ethane sulfonic acid, p[Mg*‘]f, -log (free Mg wncentration); p[Mg-A’IP], -log (Mg-ATP concentration); NTPase, nucleotide 5’-triphosphatase. or adenosine, cytidine, guanosine, inosine or uridine 5’-triphosphatase; CHAPS, 3-[(3-cholamidopropyl) dimethylammonio]-lpropane sulfonate.
two processes: Ca2+ uptake, which is catalyzed by the Ca2’ pump, and the release of Ca2’ through Ca2+ channels. The Ca2+-ATPase activity in the endoplasmic reticulum-derived microsomal fraction has been characterized and found to be similar to the corresponding enzymes in cardiac and skeletal muscles. Thus, the hepatic enzyme also forms a phosphorylated intermediate as part of the reaction sequence which is hydroxylamine labile [2, 31. The molecular weight of the enzyme is around 100-107 kD and the enzyme cross reacts with antibodies raised against the purified Ca2+-ATPase of skeletal muscle SR [4, 51. The hepatic Ca2+-ATPase was recently purified to homogeneity using conventional methods [5]. 397
cJ5LLscALcIuM
398
201 A
Fig. 1 Effect of free Mg’+ on Ca”
uptake and Ca”-ATPase activity. Ca2’ uptake and Ca’+-ATPase activity were determined as described in Materials and Methods. The free Ca2’ and Mg2+ concentrations were 20 p.M and 10 pM - 10 mM, respectively. Mg-ATP concentration WBS1 mM in all determinations. In the Ca2+ uptake experiment, the reaction was started by addition of ATP, and the total Ca2’ uptake was deWed when the Ca2’ concentration reached a plateau by addition of 1 pM ionophore A23187 to release all of the Ca2’ tramported into the vesicles. The data presented are means f SEM of 3 separate experiments using different preparations of microsomes. Open symbols represent Ca2+ uptake, filled symbols Ca2+ATPase activity
In contrast to the relatively abundant information available on the Ca2+-pump, very little is known about the Ca2’ channels present in the endoplasmic reticulum. We recently described the activation of Ca2+-release from the endoplasmic reticulum by ryanodine and -SH reagents 16.71. Though second messengers are known to release Ca2+ from the endoplasmic reticulum in whole cell preparations, their addition to isolated microsomal fractions does not result in a reproducible effect as it does in the It is well demonstrated in cellular preparation. skeletal and cardiac muscles, that Mg2+ has an effect on both aspects of Ca2’ sequestration. It is
required for the Ca2+-pump and it affects the function of Ca2+ channels. Magnesium was shown to affect several steps ir+the ATPase sequence cycle [S-l 11. Howeve;; Mg is not counter transported durinEj+active Ca accumulation [12]. Th;+activity of Ca -channels also is influenced by M Thus, %+ in cardiac and skeletal muscles Mg is an important modifier of the Ca2+ sequestration process 113, 141. The possibility that Mg2’ also might have a role in modifying hepatic Ca2’ sequestration is raised by observations demonstrating that Mg2’ fluxes are altered in the liver after the administration of different agonists [15, 161. Though these observations did not resolve whether changes in Mg2+ ‘distribution are functionally linked to the changes in Ca2+ distribution or not, this possibility cannot be excluded. Therefore, it was important to examine the effects of Mg2+ on hepatic microsomal Ca2’ sequestration, a key factor in cellular Ca2’ responses.
Materialsand Methods Chemicals ATP, Cl?, GTP, ITP, UTP, EGTA, HBPES, MBS and DTT were purchased from Sigma (St Louis, MO, USA). 45CaC12 was purchased from Amersham (Arlington Heights, IL, USA). All other reagents were of the highest purity. Microsomalpreparation Livers from male fed Sprague-Dawley rats were homogenized in a medium composed of 250 mh4 sucrose, 20 mM HBPBS (pH 7.4), 1 mM EGTA and 1 mM D’IT. The microsomal fraction was prepared a8 previously described [2, 171. In short, the homogenate was centrifuged at 1100 g for 10 min in a SS34 rotor of a Sorvall RC2-B centrifuge. The supematant was centrifuged at 7700 g for 20 min in The microsomal fraction was the same rotor. sedimented from the 7700 g supematant by centrifugation at 110,000 g for 60 min using a Ti 70 rotor in a Beckman L5-50 uhracemrituge. The microsomal fraction was resuspended in 100 mM
Mg2+ EFFECTS ON RAT LIVER MICROSOMAL Ca2+ SEQUESTRATION
KCl, 20 mM HEPES, pH 6.8, to a final protein concentration of 20-30 w/ml. We determined the plasma membrane aad contamination of mitochoadria b the assay of their respective marker Y enzymes, Na -K+-ATPase [18] and succiaate dehydrogeaase [19]. The contaminations of the microsomal preparation with plasma membrane or mitochoadria was less than two percent of the protein. Measurement
ofCa2+ transport
Ca2’ uptake by the microsomes (0.5 rag/ml) was determined in a medium composed of 100 mM KCl, 20 mM HEPES, pH 6.8 with a Ca2’ electrode (Orion, Model 93-20, Cambridge, MA, USA) at 37°C. In some experiments, 20 mM MES was used instead of HEPES to buffer the solution to pH 6.0-6.7. The Ca2+ electrode was calibrated usi;$ Ca-EGTA buffers of known ionized Ca which concentrations, were prepared and standardized using a Ca2+ standard solution purchased from Orion Associates Inc. The free, Ca2+, Mg2’ and Mg-ATP concentrations were calculated by a computer program using the constants described by Fabiato et al. [20] as previously described [211. NTPase assay
Ca2+-NTPase activity was determined using the method previously described [22]. The reaction medium containing 100 mM KCl, 20 mM HEPES (pH 6.8) 1 mM ouabaia, 1 raM NaN3, 1 pM ioaophore A23187 and 0.5 mg/ml of microsomal protein. The microsomes were pieincubated at 37°C for 10 min. The reaction was started by addition of either ATP, CTP, GTP, ITP or UTP. After the reaction at 37°C for 10 mia, the ice-cold trichloroacetic acid (final concentration 7%) was added to stop the reaction. The protein-free supemataat was obtained by centrifugatioa at 3500 g for 5 min and assayed for Pi concentration [23]. Mannose-bphosphatase
assay
Maanose-6-phosphatase was determined as described by Vaastapel et al. [24]. The incubation
399
mixtare coataiaed 250 mM sucrose, 100 mM imidazole, pH 6.5. Reaction was started by addition of 1 mM rnaaaose-6-phosphate. Assays were performed for 5 mia at 20°C aad quenched by addition of ice-cold 8% trichloroacetic acid. The protein-free supemataat was assayed for Pi [23]. Protein assay
Protein concentration was measured as described by Smith et al. [25] using bovine serum albumin as standard.
Results
The effects of free Mg2’ on Ca2+ uptake and Ca2+-ATPase activity are shown in Figure 1. The uptake of Ca2+ increased . with the free Mg2+ concentration and plateaued at the free Mg2+ concentration of 1 mM (Fig. IA). This response is consistent with the physiological concentration of free Mg2+ of around 1 mM [26]. The Km of the Ca2’ uptake was 0.4 mM of free Mg2’. The Ca2’-ATPase activity showed a somewhat different response to the changes in free Mg2+ (Fig. 1B). First, the ATPase activity decreased when free Mg2’ concentration was higher than 2.5 mM, sad second, the peak of response was at a free Mg2+ concentration of 2.5 mM, whereas the peak of Ca2’ uptake was at a Mg2+ concentration of 1 mM. It has been su ested that Mg-ATP is the main g substrate of the Ca -ATPase. The effect of varying Mg-ATP concentrations on Ca2’ u take is shown in Figure 2A. The uptake of Ca2P increased with increasing Mg-ATP c;nceatratioas, and reached its peak where the [M~+IdlMg-ATP] ratio was 1. After this point, Ca uptake decreased despite further increases ia Mg-ATP concentration (Fig. 2A). Ca2+-ATPase activity showed a similar response to Mg-ATP at the higher free Mg2+ concentrations. At a low concentration of free Mg2+ (0.1 mM), the saturation of Ca2+-ATPase activity was not apparent. The biphasic response to Mg-ATP was correlated with the corresponding free ATP concentrations. The uptake of Ca2+ increased until the free ATP level reached 0.1 mM. At higher
CELL CALCIUM
400
l 0 [Mg2+], = 2.5mM = q [Mg2+], = 0.5mM A A [tvlg2+J,= O.lmM lO[Mg2+],= 0.02mM
mM free Mg2+, the Ca2+ uptake induced by ATP increased 16 times and the Ca2+ uptake induced by CTP, GTP, ITP and UTP increased about 2 times (Fig. 3,A). The changes in Ca2+-NTPase activity induced by Mg2+ showed some differences amon 28 the various nucleotides. At 0.02 mM free Mg Ca2+-NTPase activities were similar, but at 2.5 mh; free Mg2+, the ATPase activity increased 2.3 fold, UTPase increased 78%, while CIFase, GTPase and ITPase activity decreased 42%, 23% and 60% respectively (Fig. 3B). Because the Ca2’ uptake and Ca2+-ATPase activi are pH-dependent, we determined the effect Yt of Mg on the pH dependence of Ca2+ uptake and Ca2+-ATPase. As shown in Figure 4A at high concentration of free Mg2’ (2.5 mM), Ca2’ uptake reached the maximum at pH 6.8, and decreased thereafter. At a low concentration of free Mg2’ (0.038 mM), Ca2+ uptake increased continuously
251
OJ 5.0
4.5
4.0
3.5
3.0
2.5
ATP CTP GTP ITP UTP
ATP CTP GTP ITP UTP
B
,W’W, Fig . 2
A
Effect of Mg-ATP and A’IIQ on Ca2’ uptake and
Ca2+-ATPase activity.
Ca2+ uptake and Ca2+-ATPase activity
were determined as in Figure 1. The data presented NIXmeans f Open SEM of 3 experiments using different micmsomes. symbols represent Ca2+ uptake, filled symbols Ca2’ ATPase activity ATP CTP GTP ITP IJTP
concentrations of free ATP, Ca2+ uptake was inhibited. The inhibition by ATP was more pronounced at higher Mg2+ concentrations (Fig. 2C). The nucleotide specificity at low and high Mg2+ concentrations was also tested (Fig. 3). We found that at 0.02 mM free Mg2+, ATP, CTP, GTP, ITP and UTP did not stimulate Ca”’ uptake, but at 2.5
[Mg2+] t= 0.02mM
ATP CTP GTP
ITP UTP
[Mg2+] t= 2.5mM
Fig. 3
Effect of Mg2+ on the substrate specificity of the Ca2’ The Ca2’ uptake and Ca2+-NTPase activity were P-P. The free Ca2’ and Mg-NTP determined as in Figure 1.
concentration were 20 p.M and 1 mM. respectively.
The data
presented are means f SEM of 3 separate experiments using different microsomal preparations
Mg2+ EFFECTS
ON RAT LIVER MICROSOMAL
Ca2t SEQUESTRATION
o o [Mg2+], = 2.5mM n q [Mg2+lt = 0.038mM
6.0
6.5
7.0
7.5
8.0
401
effect of Mg2+ on membrane leakiness was tested (Table 1). As an indicator of membrane integrity of the intravesicular enzyme, the activity mannose-6-phosphatase was determined in intact and detergent treated vesicles, as described by Vanstapel et al. [24]. As shown in Table 1, Mg2’ had no effect on membrane leakiness as indicated by the lack of difference in mannose-6-phosphatase activity in the presence and absence of Mg2+.
8.5
Discussion
6.0
’ 7.0
6.5
’ 7.5
’ 6.0
815
PH Fig. 4 Effect of Mg” Ca*+-ATPase
activity.
were determined was buffered
Ca”
Ca”
as in Figure
uptake
of Ca2’ uptake and
and Ca*+-ATPase
at pH 6.8-8.2
by 20 mM HEPEZS. The free Ca2’ and Mg-ATP were 20 pM and 1 mM, respectively.
The data on
uptake are n~eans f SEM of 3 separate experiments
data on Ca2+-ATPase experiments
activity
1. The reaction mixture at pH 6.0
by 20 mM MES, and the solutions
were buffered concentration
on the pH dependence
activity
and the
are means f SEM of 9 separate
using diKerent microsomal
preparations
The2ysults presented in this study demonstrate that has multiple effects on hepatic microsomal Fjt sequestering: (a) free Mg2 is essential for Ca2+-ATPase activity and Ca2’ uptake; this effect of free Mg2+ is not due to Mg-ATP being the substrate for the Ca2+-ATPase, but additional to it: and (b) Mg2+ also seems to affect a Ca2’ efflux pathway. The results illustrated in the experiments seen in Figures l-3, clearly demonstrate that Mg-ATP is essential for Ca2+-ATPase activity as is expected from its role as a substrate for the Ca2+-ATPase reaction. However, at all the Mg-ATP concentrations employed, increasing [Mg2+] to the iological, millimolar range, stimulated both the . . PCh”-ATPase acuvity and the uptake of Ca2’ . A possible mechanism by which free Mg2’ stimulates the activity of the enzyme is the one described for the sarcoplasmic reticulum enzyme, namely at the dephosphorylation step of the phosphoenzyme
with pH (Fig. 4A). However, the absolute amount Table 1 Effectsof Ca2’ and Mg2+on microsomal of Ca2+ taken up in low Mg2+ is less than 15 membrane permeability percent of the amozu+nttaken up in the presence of Similar1 concentrations. phxsio1ogica.l Mg Additions M-6-Pase activity J-G Ca -ATPase activity was lower in low Mg (nmol/m&nin) concentration and the enzyme activity decreased even further at basic pH (Fig. 4B). Control 1.47 f 0 192.95 f 6.64 In all the above experiments when Ca2+ uptake CHAPS Ca2+ 1.96 f 1.04 was measured, the net uptake of Ca2+ reflected the Ca2+ Mg2+ 2.82 f 1.24 balance between Ca2+ uptake due to the activity of the Ca2+ pump and the efflux of Ca2’ through Mannose-6-phosphatase (M-6-Pase) activity was de&mined as specific channels or non-specific membrane described in Materials and Methods. Control contained the ma&ion but&r only. Total activity was measuredinthepresence leakiness. In order to examine whether Mg2+ of 0.2% CHAPS whi& permeabilized the vesiclea. C!a* and affects Cazt efflux directly through Ca2’ channels Mg* were added tc give a final concen~ticm of 5 n&l. The results were means 5 SEih4 of 3 experiments or indirectly by increasing membrane leakiness, the
402
intemediate 18, 27, 281. At o~+t$al levels of [Mg-ATP], the effect of free Mg IS consistently stimulatory. The effect of Mg-ATP seems to be complex (Fig. 2) because it is of a biphasic nature, stimulatory at the lower and inhibitory at the higher concentrations. This effect of Mg-ATP seems to be the reflection of the associated changes in the concentration of free ATP. An examination of Figure 2B reveals that above 0.1 mM, further increases in Eree ATP are associated with a decrease in Ca2+ uptake. This effect of free ATP might be due to its effect on membrane permeability as has been suggested for the sarcoplasmic reticulum vesicles. It was shown by Chiesi and Wen that when the Ca2’-ATPase is phosphorylated in the absence of Mg2+ it forms a stable, phosphorylated intermediate [29]. Addition of ATP under this condition results in rapid Ca2’ release. Shoshan-Barmatz [30] reported that preincubation of sarcoplasmic reticulum vesicles with ATP inhibited subsequent Ca2’ uptake, an effect which was specific to free ATP. Meissner and Henderson [31] also reported on ATP induced Ca2’ release. The inhibitory effect of free ATP on Ca2’ above the concentration of 0.1 mM might be due to any of the above mechanism(s) or to some different, yet undefined, cause. It was outside the scope of this study to establish the mechanism(s) by which hi h St of free ATP prevent Ca concentrations It seems clear, however, that the accumulation. changes in [ATP]r are the underlying mechanism responsible for the biphasic responses to the changes in [Mg-ATP] levels. Additional effects of free Mg2+ are illustrated in Fi ure 3 where the nucleotide specificity of the CaSt -ATFQse reaction and Ca2’ u{p are shown. concentration As shown, increase in free Mg renders to the enzyme the ability to pump Ca2’. This effect might be due to the stimulation of the dephosphorylation step or at an additional site. The inhibitory effect of vanadate on the enzyme is also Mg2+ dependent as shown previously by Dawson and Fulton 1321. The stimulatory effect of Mg2+ on Ca2’ uptake is evident in the pH range of 6.0 - 8.2 (Fig. 4). As described originally by Moore et al. [33], Ca2+ uptake is pH sensitive and it declines at pH higher
CELL CALCIUM
than 6.8. Because the Ca2+-ATPase activity peaks at pH 7 5 the sharp decline in Ca2’ uptake observed betweed ;H 6.8 - 7.5 is likely to be caused by stimulation of Ca2+ efflux. The pronounced effects that pH has on various cellular processes involving Ca2+ are well known, as is the Ca2+ releasing effect of alkalinization [34-361. A Mg2+ induced inhibition of the Ca2+-ATPase at high pH in the sarcoplasmic reticulum was recently described [37]. However, in the present experimental conditions at low Mg2+ concentrations the enzyme activity was already very low so that possible additional pH effects were not evident. In summary, we present evidence showing that Mg2+ exerts a profound effect on microsomal Ca2’ sequestration. The physiological significance of these findings mi ht depend on the magnitude of the 2! changes in Mg distribution following hormonal stimulation of the liver.
Acknowledgements Supported by PHS grant NDDK-DK36916. The valuable suggestions of V. Shoshan-Bannatz and the critical reading of the manuscript by N. Karin were greatly appreciated.
References 1. Camfoli E. (1987) Intracellular calcium homeostasis. Annu. Rev. Biochem., 56.395-433. 2. Fleschner CR. Kraus-Friedmann N. Wibert GJ. (1985) Phosphorylated intermediates of two hepatic microsomal ATPases. Biochem. J., 226, 839-845. 3. Heihnann C. Spamer C. Gerok W. (1985) The calcium pump in rat liver endoplasmic reticulum. Demonstration of the phosphorylated intermediate. J. Biol. Chem., 260, 788-794. 4. Damiani E. Spamer C. Heihnann C. Salvatori S. Magreth A. (1988) Endoplasmic reticulum of rat liver contains two proteins closely related to skeletal sarcoplasmic reticulum Ca-ATPase and calsequestrin. J. Biol. them., 263, 340-343. 5. Jong YJ. Sheldon A. Kraus-Friedmann N. (1990) Purification of the hepatic microsomal Ca2+-ATPase from rat liver. (Submitted). 6. Shoshan-Bannatz V. Kraus-Friedmann N. (1990) High affinity ryanodine binding site in rat liver endoplasmic reticulum. (Submitted). 7. Zhang GH. Yamaguchi M. Kimura S. Higham S. Kraus-Friedmann N. (1990) Effects of heavy metals on the microsomal Ca2’-ATPase and Ca2” sequestering. Relation to SH-groups. J. Biol. Chem., 265,2184-2189.
Mg2+ EFFECTS ON RAT LIVER MICROSOMAL
Ca2+ SEQUESTRATION.
8. Martonosi AN. (1984) Mechanisms of Ca” release from satcoplasmic reticulum of skeletal muscle. Physiol. Rev., 64, 1240-I 319. 9. Champeil P. Gingold MP. Guillain F. Inesi G. (1983) Effects of magnesium on the calcium-dependent transient kinetics of sarcoplasmic reticulum ATPase, studied by stopped flow fluorescence and phosphorylation. J. Biol. Chem., 258,4453-4458. 10. Orlowski S. Lund S. Moller J. Champeil P. (1988) Phosphoenzymes formed from Mg.ATP and Ca.ATP during pm-steady state kinetics of sarcoplasmic reticulum ATPase. J. Biol. Chem., 263, 17576-17583. 11. Asturias FJ. Blasie JK. (1989) Effect of Mgzf concentration on Ca2+ uptake kinetics and structure of the sarcoplasmic reticulum membrane. Biophys. J., 55, 739-753. 12. Salama G. Scarpa A. (1985) Magnesium permeability of samoplasmic reticulum. Mg2+ is not countertransported during ATP-dependent Ca2+ uptake by sarcoplasmic reticulum. J. Biol. Chem., 260, 11697-11705. 13. Agus ZS. Kelepouris E. Dukes I. Morad M. (1989) Cytosolic magnesium modulates calcium-channel activity in mammahan ventricular cells. Am. J. Physiol., 256, C452C455. 14. White RE. Hartzell HC. (1989) Magnesium ions in cardiac junction. Regulator of ion channels and second messengers. B&hem. Pharmacol., 38,859-867. 15. Bond M. Vadasx G. Somlyo AV. Somlyo AP. (1987) Subcellular calcium and magnesium mobilization in rat liver stimulated in vivo with vasopressin and glucagon. J. Biol. Chem., 262, 15630-15636. 16. Jacob A. Becker J. Schottli G. Fritzsch G. (1989) aladrenergic stimulation causes Mg2+ release from perfused rat liver. FEBS Lett., 246, 127-130. 17. Zhang GH. Kraus-Friedmann N. (1989) Demonstration of adenylate kinase activity in hepatic micmsomes. Relevance to Ca2’ uptake. B&hem. Int., 19, 333-343. 18. Shamchmidt BF. Keeffe EB. Blankenship NM. Ockemer RB. (1979) Validation of a recording spectrophotometric method for measurement of membrane-associated Mg- and Na/K-ATPase activity. J. Lab. Clin. Med., 93, 790-799. 19. Veeger C. Der Vartanran DV. Zeylemaker WP. (1968) Succinate dehydrogenase. Methods Enzymol., 13, 81%. 20. Fabiato A. Fabiato F. (1979) Calculator programs for computing the composition of the solutions containing multiple metals and ligands used for experiments in skinned muscle cells. J. Physiol. (Paris), 75, 465-505. 21. Fleschner CR. Kraus-Friedmann N. (1987) Inhibition of microsomal Ca2+-ATPase by fluorescein 5’ isothiocyanate. Arch. B&hem. Biophys., 254,448-453. 22. Fleschner CR. Kraus-Friedmann N. (1986) The effect of Mg2+ on hepatic microsomal Ca2’ and Sr+ transport. Eur. J. B&hem., 154,313-320. 23. Lanzetta PA. Alvarez L. Reinach PS. Candia DA. (1979) An improved assay for nanomole amounts of inorganic
403
phosphate. Anal. Biochem., 100,95-97. 24. Vanstapel F. Pua K. Blanckaert NB. (1986) Assay of mannose&phosphate in untreated and detergent-disrupted rat-liver microsomes for assessment of integrity of microsomal preparations. Eur. J. B&hem., 156,73-77. 25. Smith PK. Krohn RI. Hermanson GT. et al. (1985) Measurement of protein using biconchoninic acid. Anal. B&hem., 150,76-85. 26. Murphy E. Steenbergen C. Levy LA. Raju B. London RE. (1989) Cytosolic free magnesium levels in ischemic rat heart. J. Biol. Chem., 264, 5622-5627. 27. Inesi G. (1985) Mechanism of calcium transport. Annu. Rev. Physiol., 47,573~601. 28. Vilsen B. Andersen JP. (1987) Effects of phospholipid, detergent and protein-protein interaction on stability and phosphoenzyme isomerixation of soluble sarcoplasmic reticulum Ca-ATPase. Eur. J. B&hem., 170.421-429. 29. Chiesi M. Wen YS. (1983) A phosphotylated conformational state of the (Ca2+-Mg’+)-ATPase of fast skeletal muscle sarcoplasmic reticulum can mediate rapid Ca” release. J. Biol. Chem.. 258, 60786085. 30. Shoshban-Barmatz V. (1987) Stimulation of calcium efflux from sarcoplasmic reticulum by preincubation with ATP and inorganic phosphate. B&hem. J., 247, 497-504. 31. Meissner G. Henderson JS. (1987) Rapid Ca” release from cardiac sarcoplasmic reticulum vesicles is dependent on Ca2’ and is modulated by Mg”’ adenine nucleotide, and calmodulin. J. Biol. Chem., 262, 3065-3073. 32. Dawson AP. Fulton DV. (1983) Some properties of Ca2+-stimulated ATPase of rat liver microsomal fraction. B&hem. J., 210,405-410. 33. Moore L. Chen T. Knapp HR. Landon El. (1975) Energy-dependent calcium sequestration activity in rat liver microsomes. J. Biol. Chem., 250, 4562-4568. 34. Sorenson MM. De Meis L. (1977) Effects of an ion, pH and magnesium on calcium accumulation and release by samoplasmic reticulum vesicles. Biochim. Biophys. Acta, 465,210-223. 35. Connett RI. (1978) Association of increased pH with calcium ion release in skeletal muscle. Am. J. Physiol., 234, CllO-c114. 36. Busa WB. Nuccitelli R. (1984) Metabolic regulation via intracellular pH. Am. J. Physiol., 246, R409-R438. 37. Bishop JE. AlShawi MK. (1988) Inhibition of sarcoplasmic reticulum Ca2+ -ATPase by Mg2+ at high pH. J. Biol. Chem., 263, 1886-1892. Please send reprint requests to : Dr N. Kraus-Friedmann, Department of Physiology and Cell Biology, University of Texas Medical School, P.O. Box 20708, Houston, TX 77225, USA. Received : 14 December 1989 Revised : 9 February 1990 Accepted : 9 March 1990
