Molecular and Cellular Endocrinology, 84 (1992) 15-22
15
0 1992 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207/92/$05.00
MOLCEL 02678
Regulation of Ca2+ uptake in skeletal muscle by 1,25dihydroxyvitamin role of phosphorylation and calmodulin
D,:
Virginia Massheimer, Luis M. Fernandez, Ricardo Boland and Ana R. de Boland Departamento de Biologia, Universidad National de1 Sur, 8000 Bahia Blanca, Argentina
(Received 9 May 1991; accepted 1 October 1991)
Key worak 1,25-Dihydroxyvitamin D,; Skeletal muscle; Microsomal membrane; Cyclic AMP; Calmodulin binding; Calcium uptake
Summary
Experiments were carried out to obtain information about the mechanism underlaying the fast action of 1,25dihydroxyvitamin D, (1,25(OH),D,) in skeletal muscle. N-2’-o-dibutyryladenosine-3’,5’-cyclic monophosphate (dbcAMP), similarly as 1,25(OH),D, (5 X lo-” M), rapidly increased 45Ca uptake by soleus muscle from vitamin D-deficient chicks ( + 25% and + 98% at 3 min and 10 min, respectively) in a dose-dependent manner. The effects of the CAMP analog (10 PM) and 1,25(OH),D, could be abolished by the Ca*+-channel blocker nifedipine and the calmodulin antagonist flufenazine. Calmodulin binding by two muscle microsomal proteins of 28 kDa and 30 kDa was stimulated within 1 min of exposure of the tissue to 1,25(OH),D,. Direct effects of the sterol on membrane calmodulin binding were shown with isolated microsomes. The 1,25(OH), D,-mediated rise of [ ‘251]calmodulin binding to microsomal membranes was dependent on the presence of medium ATP. Forskolin (10 PM) and CAMP (10 PM) also increased [ 12511calmodulin binding ( + 75% and + 64%, respectively, with respect to controls). Pretreatment of microsomal membranes with CAMP-dependent protein kinase inhibitor (1 pg/ml) or addition of alkaline phosphates (1 U/ml) after hormonal treatment caused complete inhibition of 1,25(OH),D,-induced [1251]calmodulin binding to microsomal membrane proteins. These results imply modifications of membrane protein phosphorylation through the CAMP signal pathway and in turn of calmodulin binding in the mechanism by which 1,25(OH),D, rapidly stimulates skeletal muscle Ca*+ uptake.
Introduction
The hormonally active metabolite of vitamin D,: 1,25_dihydroxyvitamin D, (1,25(OH),D,), has been involved in the regulation of muscle intracellular calcium levels (Curry et al., 1974; Matthews et al., 1977; Boland et al., 1983; Giu-
Correspondence to: Ana R. de Boland, Departamento de Biologia, Universidad National del Sur, 8000 Bahia Blanca, Argentina.
liani and Boland, 1984). The hormone is believed to act through receptor-mediated genomic events (Boland and Boland, 1985a; Boland et al., 1985; Simpson et al., 1985). However, several studies have demonstrated a 1,25(0H),D,-mediated rapid uptake of Ca*+, independent of gene transcription, in both classical and non-classical target tissues (Lieberherr, 1987; Nemere and Norman, 1987; Baran and Kelly, 1988; Boland et al., 1989). Exposure of intact chick skeletal muscle to the hormone causes an acute stimulation of muscle Ca*+ uptake which can be suppressed by
16
Ca2+-channel blockers (Boland and Boland, 1987) and calmodulin (CAM) antagonists (Boland et al., 1988). This change is paralleled by an increase in CAM content of muscle membranes at the expense of a decrease in cytosol CAM concentration (Bofand et al., 1988>, More recently, it has been shown that 15 mm treatment of vitamin D-deficient skeletal muscle with 1,25(OH),D, affects the binding properties of muscle membrane CAM-binding proteins and that forskolin, an activator of adenylate cyclase, mimics the effects of the hormone on muscle Cazf uptake, CAM redistribution and binding ability (Fernandez et al., 1990; Massheimer et al., 1990). Lack of additional key evidence precluded to firmly conclude on causal relationships between the fast modifications induced by the hormone. The objectives of the present study were to investigate whether 1,25fOH),D, affects CAM binding through phospho~lation of membrane proteins and to evaluate the role of these changes in the fast stimulation of muscle Ca2+ uptake by the sterol. Materials and methods
~-2’-o-dibut~rylade~osine-3’,5’-cyclic AMP (dbcAMP), nifedipine, verapamil, fluphenazine, compound 48/80, CAMP-dependent protein kinase inhibitor from rabbit muscle, forskolin and calmodulin were provided by Sigma Chemical Co. (St. Louis, MO, USA). 1,25_Dihydroxyvitamin D, was a gift of Dr. Milan Uskokovic (Hoffman-~a Roche, Rathway, NJ, USA). Animals Chicks were raised from 1 day of age on a vitamin D-deficient diet with 1.6% calcium and 1.0% phosphorus (Wasserman and Taylor, 1973) for 5 weeks in an environment deprived of light. In vitro treatment of muscle preparations Soleus muscle dissected from both legs of vitamin D-deficient chicks were preincubated in Krebs-Henseleit-glucose solution (Paul, 1975) for 30 min at 37 o C under O,/CO, (95%/S%) with
constant shaking. 1,25(OH),D, dissolved in ethanol (S x 10-r” M) or dbcAMP (1 PM-1 mM) were then added for l-15 min. Ethanol ( < 0.1%) alone was added to control samples. When nifedipine, verapamil, fluphenazine or compound 48,&I were used, they were added 5 min before addition of dbcAMP or 1,25(OH),D,. The concentrations of Ca2+-channel blockers were selected on the basis of dose-response studies (Boland and Boland, 1987).
Determination of 45Ca uptake were carried out using a modification of a procedure previously described (Giuliani and Boland, 1984; Boland and Boland, 1985b). Muscle samples prepared and treated as described above were incubated in Krebs-Henseleit-glucose containing 4sCaCI, 12 #Zi/ ml) at 37 o C for 5 min. The tissue was then quickly washed with cold unlabelled medium, blotted on filter papers and dissolved in hot 1 N NaOH. Aliquots were taken for determination of protein content (Lowry et al., 1951) and radioactivity using Aquasol as scintillation fluid.
Control and 1,25(0H),D,-treated soleus muscles were homogenized with an Ultraturrax homogenizer for 30 s using 50 mM Tris-HCl pH 7.0, 10 mM MgCl,, 0.5 nM phenylmethylsulfonyl fluoride, 5 mM EDTA, 0.2 mM EGTA, 0.25 M sucrose (10 ml,/g tissue). The homogenate was centrifuged for 10 min at 1200 x g and a Sorvall refrigerated centrifuge. The supernatant was fittered through a triple layer of cheese cloth and centrifuged 30 min at 11,300 X g. Microsomal membranes were sedimented from this supernatant in a Beckman W-50 B ultracentrifuge at 120,000 X g for 1 h. The microsomal fraction was then resuspended in homogenization buffer. The quality of membrane preparations was assessed by meassuring ouabain-sensitive Naf, K+-ATPase (plasma membrane), Ca*+,Mg+-ATPase (sarcoplasmic reticulum) and azide and oligomycin-sensitive CaZf-ATPase (mitochondria) (Boland et al., 1988).
17
Identification of calmodulin binding proteins
CAM-binding proteins of muscle membranes, isolated as indicated above, were detected using the CAM gel overlay technique of Glenney and Weber (1980) as modified by Nelson et al. (1983) as previously described (Massheimer et al., 2990). Briefly, equal aliquots of microsomal membranes (100 kg) were subjected to sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) according to Laemmli (1970) using a polya~~lamide gradient f5- 15%). After electrophoresis, the gels were fixed (40% methanol: 10% acetic acid), washed (10% ethanol) to remove SDS and incubated in buffer containing 5 PCi (1 mCi/mmol) [““I]CAM/gel. After 48 h of incubation, the gels were washed, stained with Coomassie brilliant blue, dried and exposed to Agfa-Gevaert film (Eastman-Kodak Co., Rochester, NJ, USA) for 3-4 weeks at - 20 0 C.
Quantitation of [‘“5i]calmodulin microsomal proteins
binding to total
Microsomes were isolated from vitamin D-deficient chick soleus muscle as indicated before, except that EDTA and EGTA were omitted from the homogenization buffer. Membrane samples (100 pg protein) were pre.incubated 1 min at 37’ C in 100 ,ul medium containing 50 mM TrisHCI, pH 7.0, 10 mM MgC12, 1 mM CaCl,, 1% bovine serum albumin (BSA) in the absence or presence of 1 mM ATP. The reaction was started by addition of 1,25(OH),D, (5 X lo-‘” M), forskolin (10 FM) or CAMP (10 yM). After l-10 min of incubation, [‘251]CAM (0.05 ,&I’) was added and the samples were further incubated for 10 min. The reaction was stopped by Millipore filtration (pore size, 45 pm). The filters were washed twice with 50 mM Tris-HCI, pH 7.0, 10 mM MgCl,, 1 mM CaCl,, dried at 24°C and the radioactivity quantitated in a gamma counter. When cAMP protein kinase inhibitor (PKI) (Walsh et al., 1971) was employed, it was added to the incubation medium 5 min before hormonal treatment. To test the effects of alkaline phosphatase (Mieskes et al., 1984; Tung et al., 1984), immediately after hormonal treatment and prior to the addition of the enzyme (1 U/ml, 5 min), the medium was adjusted to pH 10.4 with 1 N
NaOH. [1251]CAM was then added and the reaction allowed to proceed as described above. Results and discussion
Treatment of soleus muscle from vitamin Ddeficient chicks with l-10 PM N-2’-o-dibutyryladenosine-3’,5’-cyclic monophosphate (dbcAMP) for 5 min increases 45Ca uptake in a dose-dependent fashion (Table 1). Maximum responses were obtained with 10 PM dbcAMP (75% increase with respect to control levels). Higher doses of the cyclic AMP analog (0.1-l mM) had a less pronounced effect (28-27% increase, with respect to controls). It has been reported that some of the actions of dbcAMP may be mediated by butyrate (Yusta et al., 1988), formed upon hydroiysis of dbcAMP in the cytosol. It is possible then that at the higher concentrations of dbCAMP used (0.1-l mM) butyrate effects may superimpose CAMP-dependent results. Similarly as physiological concentrations of 1,25(OH),D,, dbcAMP (10 PM) stimulation of 45Ca uptake was time dependent (Fig. 1). A 28% (p < 0.05) increase above controls was observed after 3 min of exposure to the analog. Within 10 min 45Ca uptake increased 98% (p < 0.001) from controls. At 15 min, the increment was less pronounced, reaching 53% (p < 0.025) of control
TABLE 1 DOSE-RESPONSE EFFECTS OF dbcAMP ON 45Ca UPTAKE BY SOLEUS MUSCLE OF VITAMIN D-DEFICIENT CHICKS The tissue was incubated in the presence of the indicated concentrations of dbcAMP for 5 min. 4sCaC1, was then added and the uptake was measured as described in Materials and methods. Each set of data represents the mean of five independent-experiments. dbcAMP GnM)
45Ca uptake increase above control (9%)
0.001
0.005 0.01 0.1 1.0
18+_ 6 75+ 5** 28*10 * 27+ 8*
* p < 0.005; * * p < 0.001 with respect to controls.
IX
180
160
140 120 -3
1
1
0
3
5
10
15
TIME (MINUTES)
Fig. I. Rapid stimulation of skeletal muscle calcium uptake by dbcAMP and l,Z(OH),D,. Soleus muscles from vitamin Ddeficient chicks were incubated in Krebs-Henseleit-glucose solution in the absence and presence of dbcAMP (10 PM) at 37 ‘C for 3-15 min. The uptake of “CaCI, was then measured as indicated in Materials and methods. The effects of 1,25(OH),D, (5 x lo-“’ M) were compared to those of dbCAMP under similar treatment conditions. Results are the mean of four separate experiments+ SD. dbcAMP (0); 1,25(OH&D, (0). * p < 0.05; * * p < 0.025; * * * p < 0.001.
values. Prior addition of the calcium-channel blocker and calmodulin antagonist, nifedipine (30 PM) and fluphenazine (100 PM), respectively, abolished the increase in muscle 4”Ca uptake produced by dbcAMP and 1,25(0H),D, after 5 min (Fig. 2). Similar observations were made with the calcium-channel blocker verapamil (30 PM) and compound 48/80, a calmodulin antagonist (data not given). These results involve the CAMP signal pathway in the rapid, non-genomic effects of 1,2S(OH),D, in skeletal muscle. In agreement with this interpretation, inhibitors of CAMP-dependent protein kinase from rabbit muscle (10 U/ml) and porcine heart (50 U/ml) inhibited 100% and 84%, respectively, the fast increase in Ca2’ uptake induced by 1,25(OH),D, in cultured muscle cells (Boland et al., unpublished). Changes in calmodulin (CAM) binding by specific skeletal muscle membrane proteins upon 15 min treatment of chick skeletal muscle with 1,2S(OH),D, have been recently shown (Massheimer et al., 1990). The major CAM target was a Ca2+-independent microsomal protein of 28 kDa.
To study the time dependence of 1,25(OH),D, effects on CAM binding, soleus muscle of vitamin D-deficient chicks were treated in vitro (l-15 min) with physiological levels of 1,25(OH),D, and microsomes isolated thereafter. Microsomal proteins were then separated by SDS-PAGE (5 15% acrylamide) followed by incubation with [‘25IICAM and autoradiography to detect individual calmodulin binding proteins. Fig. 3 shows a representative autoradiogram of such experiments. By using polyacrylamide-gradient electrophoresis we were now able to resoIve microsoma1 CAM targets affected by the hormone into two proteins of 28 and 30 kDa. Stimulation in CAM binding by these proteins was already detected within only 1 min of exposure to 1,25(OH),D, and was evident up to 15 min of treatment with the hormone. Higher CAM levels of intestine brush border membranes in response to treatment with 1,25(OH),D, in vivo have been related to an increased CAM binding ability of a 102-103 kDa, Ca*+-independent membrane protein (Bikle et al,, 1984; Bikle and Munson, 1985). No information was provided on the specific mechanism by which the hormone affects the binding of CAM to this protein. We previously reported (Fernandez et al., 1990) that direct treatment of skeletal muscle microsomal mem1 7
0
CONTROL
fJ
1,25~OHl~D3
IlIt dbc
AMP
6 !5 $4 c
3
*FLUPHENAZINE
+NIFEOIPINE
Fig. 2. Suppression by calmodulin antagonists and Ca’+-channel blockers of the rapid stimulation of muscle calcium uptake caused by dbcAMP and 1,25(OH),D,. Vitamin D-deficient chick soleus muscles were treated for 5 min with dbcAMP (10 FM) or 1,25(OH),D, in the absence or presence of fluphenazine (100 FM) or nifedipine (30 PM). 45Ca uptake was then measured for 5 mm as indicated in Materials and methods. Each bar represents the mean value& SD of four experiments. * p < 0.01; * * p < 0.025.
19
TIME
(min)
Fig. 3. Calmodulin binding proteins of chick skeletal muscle microsomal membranes. The figure shows the autoradiogram of microsomes (75 pg protein) isolated from vitamin D-deficient chick soleus muscles which had been incubated (l-15 min) in the absence (control) and the presence of 1,25(OH),D, (5~ lo-” Ml. The proteins were separated by slab-gel electrophoresis and then incubated with [‘251]CAM followed by autoradiography. The positions of molecular weight markers are shown on the left.
branes with physiological amounts of 1,25(OH),D, stimulates the phosphorylation of several microsomal proteins including the 28 kDa protein whose CAM binding ability is also potentiated. The hormone also increases adenylate cyclase activity in isolated microsomes. Experiments were performed in the present study to ascertain whether phosphorylation mediates the action of 1,25(OH),D, on calmodulin binding by muscle membrane proteins. We evaluated whether direct treatment of muscle membranes with the hormone affects their CAM binding ability and whether its effects are mediated by ATP. To that end, microsomal membranes were preincubated (3 min) with 1,25(OH),D, (5 X lo-” M), [r*‘I]CAM was then added and incubated for 10 min (interval selected from the linear part of a curve of CAM binding vs. time, not shown), followed by Millipore filtration and quantitation of filter radioactivity. Fig. 4 shows that in the absence of ATP, treatment with 1,25(OH),D, failed to increase [1251]CAM binding to microsomal membranes. When 1 mM ATP was added to the incubation medium, the hormone increased 30% the total binding of [‘251]CAM to membranes (200% of ATP-dependent binding). The 1,25(OH),D,-dependent binding of [ 1251]CAM was specific since it could be displaced by an excess of
non-radioactive CAM (8.87 k 0.83, 11.20 & 0.75 and 5.6 k 0.23 cpm X lo-“/mg membrane protein for control, 1,25(OH),D, and 1,25(OHl,D, + lOOO-fold excess cold CAM, respectively). The stimulation of CAM binding to microsomal membranes in the presence of medium ATP was dependent on the time of exposure to the hormone (Fig. 5). After 1 min ATP-dependent binding increased by 55% (p < 0.05), maximum responses were obtained at 10 min (5-fold increase with respect to control). These changes to a great extent paralleled the modifications in Ca2+ uptake induced by 1,25(OH),D, (Fig. 1). To test the possibility that phosphorylation by CAMP-dependent protein kinase would be involved in changes in calmodulin binding to muscle membranes, we studied the direct effects of forskolin and CAMP on [‘2’I]CAM binding to chick muscle microsomes (Fig. 6). Similarly as 1,25(OH),D,, 3 min treatment of microsomal membranes with 10 PM forskolin or 10 PM CAMP increased CAM binding (75% and 64% (p < 0.051, respectively, with respect to control). These 1
-ATP
+ATP
Fig. 4. Stimulation by 1,25(OH),D, of calmodulin binding to skeletal muscle microsomal membranes: dependence of medium ATP. Microsomal membranes (100 Kg) isolated from vitamin D-deficient chick soleus muscle were incubated in 50 mM Tris-HCI pH 7.0, 10 mM MgCI,, 1% BSAk 1 mM ATP in the absence (open columns) and presence (filled columns) of 1,25(OH),D, (5 x lo- lo M) for 3 min at 37 o C. [“‘I]CAM was then added for 10 min and the binding reaction stopped by Millipore filtration. The filters were washed twice with incubation buffer, dried and the radioactivity determined in a gamma counter. Results shown are the average of three independent experiments performed in duplicate k SD. * p < 0.05.
TABLE 2 EFFECTS OF ALKALINE PHOSPHATASE AND CAMPDEPENDENT PROTEIN KlNASE INIiI~IT~R (PKI) ON &?XOH),D,-DEPENDENT STIMULATION OF [“‘IfCALMODULIN BINDING TO MICROSQMAL MEMBRANES Microsomal membranes (100 _ug)were incubated in a medium containing 1 mM ATP with 1,25tOH),D, (5 X 10- VJM) in the presence and absence of alkaline phosphatase (1 U/ml) or CAMP-dependent protein kinase inhibitor (1 mg/ml) as described in Materials and methods. [‘ZSIJCalmodulin was then added as described in the legend of Fig. 3. Results are the mean&SD of three independent experiments performed in duplicate.
Fig. 5, Time-course of 125(OH),D,
Treatment
~I~~~alrnoduIin binding icpmjmg protein)X IO- ’
Control +-Alkaline phosphatase t- PKI l,Z%OH),D, 1,25(OH),D, + alkaline phosphatase 1,2S(OH),D, + PKI
44.10* 2.55 42.62k2.13 37.82+ 1.51 56.45 + 1.46 *
effects un ATP-depen-
dent calmodulin binding to microsomal membranes. Microsomes were incubated with 1,25(OH),D1 (5X lO_“’ M) for I-10 min in the presence of 1 mM ATP as described in the legend of Fig. 4. ATP-dependent 1”%‘AM binding was calculated as the difference between binding in the presence and absence of 1 mM ATP. Results are the average of three independent experiments performed in duplicate _t SD. * p < 0.05.
38.17*4.60 ** 35.28k1.74 ***
* p < 0.0.5with respect to control; * * p < 0.025 and * * * p < 0.001 with respect to 1,2S(OH),D, treatment.
results are in accordance with previous observations which indicate that ~~P-dependent phosphorylation modulates cahnodulin affinity of several CAM binding protein (Conti and Adelstein,
Fig. 6. Effects of CAMP and forskolin on [i251]calmodulin binding to microsomal membranes. Microsomes (100 1+g)were incubated in the absence and presence of CAMP (10 PM), forskolin 110 FM) or i,2S(OH),D, (5 x 10- ‘* M). Binding of fi2’I]CAM was performed as described in the legend of Fig, 4 in the presence of 1 mM ATP. Results are the average of four independent experiments f SD. * p < a,05
1981; Mafanick and Anderson, 2982; Vi‘tar Palasi et al., 1983). The ability of CAMP-dependent protein kinase inhibitor (PKI) to block 1,25(OH),D, stimulation of CAM binding to microsomal membranes was also investigated. As shown in Table 2, PKI (1 pg,/ml) caused complete inhibition of 1125(OH)2Q-dependent fi2SIJCAM binding to microsomal membranes. In addition, alkaline phosphatase (1 U/ml; Table 2) was also effective in suppressing the stimuIatory effects of the hormone on CAM binding. Muscle microsomal membranes contain both plasma membranes and sarcoplasmic reticulum (SRI. However, in SR the most important phosphorylation system is Ca*‘/ calmodulin dependent whereas in sarcolemma CAMP-regulated protein phosphorylation predominates (Walaas et al,, 1988). This suggests that the changes observed would be localized in the plasma mFmbrane. Interestingly, a sarcolemmal protein of 29 kDa is a substrate for cup-dependent protein kinase (Andrew et al,, 1975; Walaas et al., 1988).
21
These results indicate that 1,25(OH),D, stimulates CAM binding to skeletal muscle microsoma1 proteins through phosphorylation via CAMPdependent protein kinase. Several lines of evidence suggest that modifications of membrane protein phospho~Iation and in turn of calmodulin binding are part of the mechanism by which 1,25(OH),D, elicits a rapid stimulation of muscle calcium uptake. The fact that both dbcAMP and 1,25(OH),D, induced fast time-dependent increases in 45Ca uptake which were suppressed by calcium-channel blockers supports this contention (Figs. 1 and 2). Moreover, the early increase of soleus muscle Ca2+ uptake caused by 1,25(OH),D, is accompanied by a significant elevation of tissue CAMP and forskolin also stimulates muscle Ca ‘+ fluxes through a dihydropyridine-sensitive pathway (Fernandez, 19901. In addition, forskolin and CAMP, like the hormone, promote similar rapid changes in calmodulin binding by microsomal proteins (Figs. 3-61. Furthermore, CAM antagonists inhibit Cazf fluxes regulated by 1,25(OH),Ds and activators of the CAMP pathway. The data are in general agreement with evidence involving protein phosphorylation and calmodulin in the modulation of muscle membrane Cazf channels (Sperelakis, 1984; Braily and Sperelakis, 1986). However, the possibility that the 1,25(0H),D,-induced enhancement of CAM binding by microsomal proteins in vitro and membrane calcium channel activation are unrelated events cannot be totally excluded and additional studies are required to substantiate this interpretation, Membrane protein phosphorylation has also been involved in the 1,25(OH),D, rapid stimulation of cardiac muscle Ca*+ uptake. Hormone-induced stimulation of the phosphorylation of 43 and 55 kDa microsomal proteins was observed (Selles and Boland, 199Ia, b). It is not known whether these heart proteins bind CAM. Differences in subunit composition of the Ca2+ channels from skeletal and cardiac muscle have been reported (Flockerzi et al., 1986; De Jongh et al., 1989). Finally, the possibility that 1,25(OHI,D, may affect other signalling systems, in addition to the CAMP-dependent pathway, should not be excluded. Recent work performed in our laboratory
indicates that in skeleta1 muscle the hormone affects protein kinase C activity (Massheimer and Boland, submitted for publication). Complex interactions between kinase systems have been involved in agonist signal transduction (Kikkawa et al., 1986).
Acknowledgements This work was supported by grants from the National Research Council of Argentina (CONICET) and Nestle Nutrition (Vevey, Switzerland).
References Andrew, C.G., Almon, R.R. and Appel, S.H. (1975) J. Biol. Chem. 250, 3972-3980. Baran, D.T. and Kelly, A.M. (1988) Endocrinology 122, 930934. Bikle, D.D. and Munson, S. (1985) J. Clin. invest. 76, 23122316. Bikle, D.D., Munson, S. and Chafouleas, J.G. (1984) FEBS Lett. 174, 30-33. Boland, A.R. de and Boland, R. (1985aJ Biochim. Biophys. Acta 845, 237-241. Boland, A.R. de and Boland, R. (1985b) Z. Naturforsch. 4Oc, 102-108. Boland, A.R. de and Boland, R. (1987) Endocrinolo~ 120, 1858-1864. Boland, R., Norman, A.W., Ritz, E. and Hasselbach, W. (1985) Biochem. Biophys. Res. Commun. 128, 305-311. Boland, A.R. de, Gallego, S. and Boland, R. (1983) Biochim. Biophys. Acta 733, 264-270. Boland, A.R. de, Massheimer, V. and Fernandez, L.M. (1988) Calcif. Tissue Int. 43, 370-375. Boland, A.R. de, Nemere, I. and Norman, A.W. (1989) Biochem. Biophys. Res. Commun. 166, 217-222. Braily, G. and Sperelakis, N. (1986) J. Cyclic Nucleotide Protein Phosphorylation Res. 11, 25-34. Conti, M.A. and Adelstein, R.S. (1981) J. Biol. Chem. 256, 3178-3181. Curry, O.B, Basten, J.F., Francis, M.J. and Smith, R. (1974) Nature 249, 83-87. De Jongh, K.S., Merrick, D.K. and Catterall, W.A (1989) Proc. Natl. Acad. Sci. USA 86, 8585-8589. Fernandez, L.M., Massheimer, V. and Boland, A.R. de (1990) Calcif. Tissue Int. 47, 314-319. Flockerzi, V., Ruth, P., Deken, J., Jung, B. and Hofmann, F. (1986) Horm. Cell Regul. 139, 235-248. Giuliani, D. and Boland, R. (1984) Calcif. Tissue Int. 36, 200-20s. Glenney, J.R. and Weber, J.C. (1980) J. Biol. Chem. 255, 10551-10554.
22 Kikkawa, U., Kitano, T., Saito, N., Kishimoto, A., Taniyama, K., Tanaka, C. and Nishizuka, Y. (19861 in Calcium and the Cell, p. 197, Willey, Chichester. Malenick, D.A. and Anderson, S.R. (19821 Biochemistry 21, 3480-3486. Massheimer, V., Fernandez, L.M. and Boland, A.R. de (1990) Z. Naturforsch, 45c, 663-670. Matthews, C., Heinberg, K.W., Ritz, E., Agostini, B., Fritsche, J. and Hasselbach, W. (1977) Kidney Int. 11, 227-235. Mieskes, G., Brand, I.A. and Soling, H.D. (1984) Eur. J. Biochem. 140, 375-383. Nemere, I. and Norman, A.W. (1987) J. Bone Miner. Res. 2, 99-107. Nelson, T.Y., Overweller, J.M., Chafouleas, J.G. and Boyd, A.E. (1983) Diabetes 32, 1126-1133. Laemmli, U.K. (1970) Nature 277, 680-685. Lieberherr, M. (1987) J. Biol. Chem. 262, 13168-13173. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randal, P.J. (1951) J. Biol. Chem. 193, 265-275. Paul, J. (1975) in Cell and Tissue Culture, p. 90, Churchill Livingstone, Edinburgh.
Selles, J. and Boland, R. (1991a) Mol. Cell. Endocrinol. 77, 67-73. Selles, J. and Boland, R. (1991b) Mol. Cell. Endocrinol. 82, 229-235. Simpson, R.U., Thomas, G.A. and Arnold, A.J. (1985) J. Biol. Chem. 260, 8882-8891. Sperelakis, N. (1984) Am. Heart J. 197, 347-357. Tung, H.Y.L., Resink, T.J., Hemmings, B.A., Shenolikan, S. and Cohen, P. (1984) Eur. J. Biochem. 138, 635-641. Vilar Palasi, C., Oshiro, D.L. and Kretsinger, R.H. (1983) Biochim. Biophys. Acta 757, 40-46. Walaas, IS., Horn, R.S., Nairn, A.C., Walaas, 0. and Adler, A. (1988) Arch. Biochem. Biophys. 262, 245-258. Walsh, D.A., Ashby, C., Gonzalez, C., Calkins, D., Fisher, E. and Krebs, E. (1971) J. Biol. Chem. 246, 1977-1985. Wasserman, R.H. and Taylor, A.N. (19731 J. Nutr. 103, 586599. Yusta, B., Ortiz-Caro, J., Pascual, A. and Aranda, A. (1988) Neurochemistry 51, 1801-1813.
15
0 1992 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207/92/$05.00
MOLCEL 02678
Regulation of Ca2+ uptake in skeletal muscle by 1,25dihydroxyvitamin role of phosphorylation and calmodulin
D,:
Virginia Massheimer, Luis M. Fernandez, Ricardo Boland and Ana R. de Boland Departamento de Biologia, Universidad National de1 Sur, 8000 Bahia Blanca, Argentina
(Received 9 May 1991; accepted 1 October 1991)
Key worak 1,25-Dihydroxyvitamin D,; Skeletal muscle; Microsomal membrane; Cyclic AMP; Calmodulin binding; Calcium uptake
Summary
Experiments were carried out to obtain information about the mechanism underlaying the fast action of 1,25dihydroxyvitamin D, (1,25(OH),D,) in skeletal muscle. N-2’-o-dibutyryladenosine-3’,5’-cyclic monophosphate (dbcAMP), similarly as 1,25(OH),D, (5 X lo-” M), rapidly increased 45Ca uptake by soleus muscle from vitamin D-deficient chicks ( + 25% and + 98% at 3 min and 10 min, respectively) in a dose-dependent manner. The effects of the CAMP analog (10 PM) and 1,25(OH),D, could be abolished by the Ca*+-channel blocker nifedipine and the calmodulin antagonist flufenazine. Calmodulin binding by two muscle microsomal proteins of 28 kDa and 30 kDa was stimulated within 1 min of exposure of the tissue to 1,25(OH),D,. Direct effects of the sterol on membrane calmodulin binding were shown with isolated microsomes. The 1,25(OH), D,-mediated rise of [ ‘251]calmodulin binding to microsomal membranes was dependent on the presence of medium ATP. Forskolin (10 PM) and CAMP (10 PM) also increased [ 12511calmodulin binding ( + 75% and + 64%, respectively, with respect to controls). Pretreatment of microsomal membranes with CAMP-dependent protein kinase inhibitor (1 pg/ml) or addition of alkaline phosphates (1 U/ml) after hormonal treatment caused complete inhibition of 1,25(OH),D,-induced [1251]calmodulin binding to microsomal membrane proteins. These results imply modifications of membrane protein phosphorylation through the CAMP signal pathway and in turn of calmodulin binding in the mechanism by which 1,25(OH),D, rapidly stimulates skeletal muscle Ca*+ uptake.
Introduction
The hormonally active metabolite of vitamin D,: 1,25_dihydroxyvitamin D, (1,25(OH),D,), has been involved in the regulation of muscle intracellular calcium levels (Curry et al., 1974; Matthews et al., 1977; Boland et al., 1983; Giu-
Correspondence to: Ana R. de Boland, Departamento de Biologia, Universidad National del Sur, 8000 Bahia Blanca, Argentina.
liani and Boland, 1984). The hormone is believed to act through receptor-mediated genomic events (Boland and Boland, 1985a; Boland et al., 1985; Simpson et al., 1985). However, several studies have demonstrated a 1,25(0H),D,-mediated rapid uptake of Ca*+, independent of gene transcription, in both classical and non-classical target tissues (Lieberherr, 1987; Nemere and Norman, 1987; Baran and Kelly, 1988; Boland et al., 1989). Exposure of intact chick skeletal muscle to the hormone causes an acute stimulation of muscle Ca*+ uptake which can be suppressed by
16
Ca2+-channel blockers (Boland and Boland, 1987) and calmodulin (CAM) antagonists (Boland et al., 1988). This change is paralleled by an increase in CAM content of muscle membranes at the expense of a decrease in cytosol CAM concentration (Bofand et al., 1988>, More recently, it has been shown that 15 mm treatment of vitamin D-deficient skeletal muscle with 1,25(OH),D, affects the binding properties of muscle membrane CAM-binding proteins and that forskolin, an activator of adenylate cyclase, mimics the effects of the hormone on muscle Cazf uptake, CAM redistribution and binding ability (Fernandez et al., 1990; Massheimer et al., 1990). Lack of additional key evidence precluded to firmly conclude on causal relationships between the fast modifications induced by the hormone. The objectives of the present study were to investigate whether 1,25fOH),D, affects CAM binding through phospho~lation of membrane proteins and to evaluate the role of these changes in the fast stimulation of muscle Ca2+ uptake by the sterol. Materials and methods
~-2’-o-dibut~rylade~osine-3’,5’-cyclic AMP (dbcAMP), nifedipine, verapamil, fluphenazine, compound 48/80, CAMP-dependent protein kinase inhibitor from rabbit muscle, forskolin and calmodulin were provided by Sigma Chemical Co. (St. Louis, MO, USA). 1,25_Dihydroxyvitamin D, was a gift of Dr. Milan Uskokovic (Hoffman-~a Roche, Rathway, NJ, USA). Animals Chicks were raised from 1 day of age on a vitamin D-deficient diet with 1.6% calcium and 1.0% phosphorus (Wasserman and Taylor, 1973) for 5 weeks in an environment deprived of light. In vitro treatment of muscle preparations Soleus muscle dissected from both legs of vitamin D-deficient chicks were preincubated in Krebs-Henseleit-glucose solution (Paul, 1975) for 30 min at 37 o C under O,/CO, (95%/S%) with
constant shaking. 1,25(OH),D, dissolved in ethanol (S x 10-r” M) or dbcAMP (1 PM-1 mM) were then added for l-15 min. Ethanol ( < 0.1%) alone was added to control samples. When nifedipine, verapamil, fluphenazine or compound 48,&I were used, they were added 5 min before addition of dbcAMP or 1,25(OH),D,. The concentrations of Ca2+-channel blockers were selected on the basis of dose-response studies (Boland and Boland, 1987).
Determination of 45Ca uptake were carried out using a modification of a procedure previously described (Giuliani and Boland, 1984; Boland and Boland, 1985b). Muscle samples prepared and treated as described above were incubated in Krebs-Henseleit-glucose containing 4sCaCI, 12 #Zi/ ml) at 37 o C for 5 min. The tissue was then quickly washed with cold unlabelled medium, blotted on filter papers and dissolved in hot 1 N NaOH. Aliquots were taken for determination of protein content (Lowry et al., 1951) and radioactivity using Aquasol as scintillation fluid.
Control and 1,25(0H),D,-treated soleus muscles were homogenized with an Ultraturrax homogenizer for 30 s using 50 mM Tris-HCl pH 7.0, 10 mM MgCl,, 0.5 nM phenylmethylsulfonyl fluoride, 5 mM EDTA, 0.2 mM EGTA, 0.25 M sucrose (10 ml,/g tissue). The homogenate was centrifuged for 10 min at 1200 x g and a Sorvall refrigerated centrifuge. The supernatant was fittered through a triple layer of cheese cloth and centrifuged 30 min at 11,300 X g. Microsomal membranes were sedimented from this supernatant in a Beckman W-50 B ultracentrifuge at 120,000 X g for 1 h. The microsomal fraction was then resuspended in homogenization buffer. The quality of membrane preparations was assessed by meassuring ouabain-sensitive Naf, K+-ATPase (plasma membrane), Ca*+,Mg+-ATPase (sarcoplasmic reticulum) and azide and oligomycin-sensitive CaZf-ATPase (mitochondria) (Boland et al., 1988).
17
Identification of calmodulin binding proteins
CAM-binding proteins of muscle membranes, isolated as indicated above, were detected using the CAM gel overlay technique of Glenney and Weber (1980) as modified by Nelson et al. (1983) as previously described (Massheimer et al., 2990). Briefly, equal aliquots of microsomal membranes (100 kg) were subjected to sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) according to Laemmli (1970) using a polya~~lamide gradient f5- 15%). After electrophoresis, the gels were fixed (40% methanol: 10% acetic acid), washed (10% ethanol) to remove SDS and incubated in buffer containing 5 PCi (1 mCi/mmol) [““I]CAM/gel. After 48 h of incubation, the gels were washed, stained with Coomassie brilliant blue, dried and exposed to Agfa-Gevaert film (Eastman-Kodak Co., Rochester, NJ, USA) for 3-4 weeks at - 20 0 C.
Quantitation of [‘“5i]calmodulin microsomal proteins
binding to total
Microsomes were isolated from vitamin D-deficient chick soleus muscle as indicated before, except that EDTA and EGTA were omitted from the homogenization buffer. Membrane samples (100 pg protein) were pre.incubated 1 min at 37’ C in 100 ,ul medium containing 50 mM TrisHCI, pH 7.0, 10 mM MgC12, 1 mM CaCl,, 1% bovine serum albumin (BSA) in the absence or presence of 1 mM ATP. The reaction was started by addition of 1,25(OH),D, (5 X lo-‘” M), forskolin (10 FM) or CAMP (10 yM). After l-10 min of incubation, [‘251]CAM (0.05 ,&I’) was added and the samples were further incubated for 10 min. The reaction was stopped by Millipore filtration (pore size, 45 pm). The filters were washed twice with 50 mM Tris-HCI, pH 7.0, 10 mM MgCl,, 1 mM CaCl,, dried at 24°C and the radioactivity quantitated in a gamma counter. When cAMP protein kinase inhibitor (PKI) (Walsh et al., 1971) was employed, it was added to the incubation medium 5 min before hormonal treatment. To test the effects of alkaline phosphatase (Mieskes et al., 1984; Tung et al., 1984), immediately after hormonal treatment and prior to the addition of the enzyme (1 U/ml, 5 min), the medium was adjusted to pH 10.4 with 1 N
NaOH. [1251]CAM was then added and the reaction allowed to proceed as described above. Results and discussion
Treatment of soleus muscle from vitamin Ddeficient chicks with l-10 PM N-2’-o-dibutyryladenosine-3’,5’-cyclic monophosphate (dbcAMP) for 5 min increases 45Ca uptake in a dose-dependent fashion (Table 1). Maximum responses were obtained with 10 PM dbcAMP (75% increase with respect to control levels). Higher doses of the cyclic AMP analog (0.1-l mM) had a less pronounced effect (28-27% increase, with respect to controls). It has been reported that some of the actions of dbcAMP may be mediated by butyrate (Yusta et al., 1988), formed upon hydroiysis of dbcAMP in the cytosol. It is possible then that at the higher concentrations of dbCAMP used (0.1-l mM) butyrate effects may superimpose CAMP-dependent results. Similarly as physiological concentrations of 1,25(OH),D,, dbcAMP (10 PM) stimulation of 45Ca uptake was time dependent (Fig. 1). A 28% (p < 0.05) increase above controls was observed after 3 min of exposure to the analog. Within 10 min 45Ca uptake increased 98% (p < 0.001) from controls. At 15 min, the increment was less pronounced, reaching 53% (p < 0.025) of control
TABLE 1 DOSE-RESPONSE EFFECTS OF dbcAMP ON 45Ca UPTAKE BY SOLEUS MUSCLE OF VITAMIN D-DEFICIENT CHICKS The tissue was incubated in the presence of the indicated concentrations of dbcAMP for 5 min. 4sCaC1, was then added and the uptake was measured as described in Materials and methods. Each set of data represents the mean of five independent-experiments. dbcAMP GnM)
45Ca uptake increase above control (9%)
0.001
0.005 0.01 0.1 1.0
18+_ 6 75+ 5** 28*10 * 27+ 8*
* p < 0.005; * * p < 0.001 with respect to controls.
IX
180
160
140 120 -3
1
1
0
3
5
10
15
TIME (MINUTES)
Fig. I. Rapid stimulation of skeletal muscle calcium uptake by dbcAMP and l,Z(OH),D,. Soleus muscles from vitamin Ddeficient chicks were incubated in Krebs-Henseleit-glucose solution in the absence and presence of dbcAMP (10 PM) at 37 ‘C for 3-15 min. The uptake of “CaCI, was then measured as indicated in Materials and methods. The effects of 1,25(OH),D, (5 x lo-“’ M) were compared to those of dbCAMP under similar treatment conditions. Results are the mean of four separate experiments+ SD. dbcAMP (0); 1,25(OH&D, (0). * p < 0.05; * * p < 0.025; * * * p < 0.001.
values. Prior addition of the calcium-channel blocker and calmodulin antagonist, nifedipine (30 PM) and fluphenazine (100 PM), respectively, abolished the increase in muscle 4”Ca uptake produced by dbcAMP and 1,25(0H),D, after 5 min (Fig. 2). Similar observations were made with the calcium-channel blocker verapamil (30 PM) and compound 48/80, a calmodulin antagonist (data not given). These results involve the CAMP signal pathway in the rapid, non-genomic effects of 1,2S(OH),D, in skeletal muscle. In agreement with this interpretation, inhibitors of CAMP-dependent protein kinase from rabbit muscle (10 U/ml) and porcine heart (50 U/ml) inhibited 100% and 84%, respectively, the fast increase in Ca2’ uptake induced by 1,25(OH),D, in cultured muscle cells (Boland et al., unpublished). Changes in calmodulin (CAM) binding by specific skeletal muscle membrane proteins upon 15 min treatment of chick skeletal muscle with 1,2S(OH),D, have been recently shown (Massheimer et al., 1990). The major CAM target was a Ca2+-independent microsomal protein of 28 kDa.
To study the time dependence of 1,25(OH),D, effects on CAM binding, soleus muscle of vitamin D-deficient chicks were treated in vitro (l-15 min) with physiological levels of 1,25(OH),D, and microsomes isolated thereafter. Microsomal proteins were then separated by SDS-PAGE (5 15% acrylamide) followed by incubation with [‘25IICAM and autoradiography to detect individual calmodulin binding proteins. Fig. 3 shows a representative autoradiogram of such experiments. By using polyacrylamide-gradient electrophoresis we were now able to resoIve microsoma1 CAM targets affected by the hormone into two proteins of 28 and 30 kDa. Stimulation in CAM binding by these proteins was already detected within only 1 min of exposure to 1,25(OH),D, and was evident up to 15 min of treatment with the hormone. Higher CAM levels of intestine brush border membranes in response to treatment with 1,25(OH),D, in vivo have been related to an increased CAM binding ability of a 102-103 kDa, Ca*+-independent membrane protein (Bikle et al,, 1984; Bikle and Munson, 1985). No information was provided on the specific mechanism by which the hormone affects the binding of CAM to this protein. We previously reported (Fernandez et al., 1990) that direct treatment of skeletal muscle microsomal mem1 7
0
CONTROL
fJ
1,25~OHl~D3
IlIt dbc
AMP
6 !5 $4 c
3
*FLUPHENAZINE
+NIFEOIPINE
Fig. 2. Suppression by calmodulin antagonists and Ca’+-channel blockers of the rapid stimulation of muscle calcium uptake caused by dbcAMP and 1,25(OH),D,. Vitamin D-deficient chick soleus muscles were treated for 5 min with dbcAMP (10 FM) or 1,25(OH),D, in the absence or presence of fluphenazine (100 FM) or nifedipine (30 PM). 45Ca uptake was then measured for 5 mm as indicated in Materials and methods. Each bar represents the mean value& SD of four experiments. * p < 0.01; * * p < 0.025.
19
TIME
(min)
Fig. 3. Calmodulin binding proteins of chick skeletal muscle microsomal membranes. The figure shows the autoradiogram of microsomes (75 pg protein) isolated from vitamin D-deficient chick soleus muscles which had been incubated (l-15 min) in the absence (control) and the presence of 1,25(OH),D, (5~ lo-” Ml. The proteins were separated by slab-gel electrophoresis and then incubated with [‘251]CAM followed by autoradiography. The positions of molecular weight markers are shown on the left.
branes with physiological amounts of 1,25(OH),D, stimulates the phosphorylation of several microsomal proteins including the 28 kDa protein whose CAM binding ability is also potentiated. The hormone also increases adenylate cyclase activity in isolated microsomes. Experiments were performed in the present study to ascertain whether phosphorylation mediates the action of 1,25(OH),D, on calmodulin binding by muscle membrane proteins. We evaluated whether direct treatment of muscle membranes with the hormone affects their CAM binding ability and whether its effects are mediated by ATP. To that end, microsomal membranes were preincubated (3 min) with 1,25(OH),D, (5 X lo-” M), [r*‘I]CAM was then added and incubated for 10 min (interval selected from the linear part of a curve of CAM binding vs. time, not shown), followed by Millipore filtration and quantitation of filter radioactivity. Fig. 4 shows that in the absence of ATP, treatment with 1,25(OH),D, failed to increase [1251]CAM binding to microsomal membranes. When 1 mM ATP was added to the incubation medium, the hormone increased 30% the total binding of [‘251]CAM to membranes (200% of ATP-dependent binding). The 1,25(OH),D,-dependent binding of [ 1251]CAM was specific since it could be displaced by an excess of
non-radioactive CAM (8.87 k 0.83, 11.20 & 0.75 and 5.6 k 0.23 cpm X lo-“/mg membrane protein for control, 1,25(OH),D, and 1,25(OHl,D, + lOOO-fold excess cold CAM, respectively). The stimulation of CAM binding to microsomal membranes in the presence of medium ATP was dependent on the time of exposure to the hormone (Fig. 5). After 1 min ATP-dependent binding increased by 55% (p < 0.05), maximum responses were obtained at 10 min (5-fold increase with respect to control). These changes to a great extent paralleled the modifications in Ca2+ uptake induced by 1,25(OH),D, (Fig. 1). To test the possibility that phosphorylation by CAMP-dependent protein kinase would be involved in changes in calmodulin binding to muscle membranes, we studied the direct effects of forskolin and CAMP on [‘2’I]CAM binding to chick muscle microsomes (Fig. 6). Similarly as 1,25(OH),D,, 3 min treatment of microsomal membranes with 10 PM forskolin or 10 PM CAMP increased CAM binding (75% and 64% (p < 0.051, respectively, with respect to control). These 1
-ATP
+ATP
Fig. 4. Stimulation by 1,25(OH),D, of calmodulin binding to skeletal muscle microsomal membranes: dependence of medium ATP. Microsomal membranes (100 Kg) isolated from vitamin D-deficient chick soleus muscle were incubated in 50 mM Tris-HCI pH 7.0, 10 mM MgCI,, 1% BSAk 1 mM ATP in the absence (open columns) and presence (filled columns) of 1,25(OH),D, (5 x lo- lo M) for 3 min at 37 o C. [“‘I]CAM was then added for 10 min and the binding reaction stopped by Millipore filtration. The filters were washed twice with incubation buffer, dried and the radioactivity determined in a gamma counter. Results shown are the average of three independent experiments performed in duplicate k SD. * p < 0.05.
TABLE 2 EFFECTS OF ALKALINE PHOSPHATASE AND CAMPDEPENDENT PROTEIN KlNASE INIiI~IT~R (PKI) ON &?XOH),D,-DEPENDENT STIMULATION OF [“‘IfCALMODULIN BINDING TO MICROSQMAL MEMBRANES Microsomal membranes (100 _ug)were incubated in a medium containing 1 mM ATP with 1,25tOH),D, (5 X 10- VJM) in the presence and absence of alkaline phosphatase (1 U/ml) or CAMP-dependent protein kinase inhibitor (1 mg/ml) as described in Materials and methods. [‘ZSIJCalmodulin was then added as described in the legend of Fig. 3. Results are the mean&SD of three independent experiments performed in duplicate.
Fig. 5, Time-course of 125(OH),D,
Treatment
~I~~~alrnoduIin binding icpmjmg protein)X IO- ’
Control +-Alkaline phosphatase t- PKI l,Z%OH),D, 1,25(OH),D, + alkaline phosphatase 1,2S(OH),D, + PKI
44.10* 2.55 42.62k2.13 37.82+ 1.51 56.45 + 1.46 *
effects un ATP-depen-
dent calmodulin binding to microsomal membranes. Microsomes were incubated with 1,25(OH),D1 (5X lO_“’ M) for I-10 min in the presence of 1 mM ATP as described in the legend of Fig. 4. ATP-dependent 1”%‘AM binding was calculated as the difference between binding in the presence and absence of 1 mM ATP. Results are the average of three independent experiments performed in duplicate _t SD. * p < 0.05.
38.17*4.60 ** 35.28k1.74 ***
* p < 0.0.5with respect to control; * * p < 0.025 and * * * p < 0.001 with respect to 1,2S(OH),D, treatment.
results are in accordance with previous observations which indicate that ~~P-dependent phosphorylation modulates cahnodulin affinity of several CAM binding protein (Conti and Adelstein,
Fig. 6. Effects of CAMP and forskolin on [i251]calmodulin binding to microsomal membranes. Microsomes (100 1+g)were incubated in the absence and presence of CAMP (10 PM), forskolin 110 FM) or i,2S(OH),D, (5 x 10- ‘* M). Binding of fi2’I]CAM was performed as described in the legend of Fig, 4 in the presence of 1 mM ATP. Results are the average of four independent experiments f SD. * p < a,05
1981; Mafanick and Anderson, 2982; Vi‘tar Palasi et al., 1983). The ability of CAMP-dependent protein kinase inhibitor (PKI) to block 1,25(OH),D, stimulation of CAM binding to microsomal membranes was also investigated. As shown in Table 2, PKI (1 pg,/ml) caused complete inhibition of 1125(OH)2Q-dependent fi2SIJCAM binding to microsomal membranes. In addition, alkaline phosphatase (1 U/ml; Table 2) was also effective in suppressing the stimuIatory effects of the hormone on CAM binding. Muscle microsomal membranes contain both plasma membranes and sarcoplasmic reticulum (SRI. However, in SR the most important phosphorylation system is Ca*‘/ calmodulin dependent whereas in sarcolemma CAMP-regulated protein phosphorylation predominates (Walaas et al,, 1988). This suggests that the changes observed would be localized in the plasma mFmbrane. Interestingly, a sarcolemmal protein of 29 kDa is a substrate for cup-dependent protein kinase (Andrew et al,, 1975; Walaas et al., 1988).
21
These results indicate that 1,25(OH),D, stimulates CAM binding to skeletal muscle microsoma1 proteins through phosphorylation via CAMPdependent protein kinase. Several lines of evidence suggest that modifications of membrane protein phospho~Iation and in turn of calmodulin binding are part of the mechanism by which 1,25(OH),D, elicits a rapid stimulation of muscle calcium uptake. The fact that both dbcAMP and 1,25(OH),D, induced fast time-dependent increases in 45Ca uptake which were suppressed by calcium-channel blockers supports this contention (Figs. 1 and 2). Moreover, the early increase of soleus muscle Ca2+ uptake caused by 1,25(OH),D, is accompanied by a significant elevation of tissue CAMP and forskolin also stimulates muscle Ca ‘+ fluxes through a dihydropyridine-sensitive pathway (Fernandez, 19901. In addition, forskolin and CAMP, like the hormone, promote similar rapid changes in calmodulin binding by microsomal proteins (Figs. 3-61. Furthermore, CAM antagonists inhibit Cazf fluxes regulated by 1,25(OH),Ds and activators of the CAMP pathway. The data are in general agreement with evidence involving protein phosphorylation and calmodulin in the modulation of muscle membrane Cazf channels (Sperelakis, 1984; Braily and Sperelakis, 1986). However, the possibility that the 1,25(0H),D,-induced enhancement of CAM binding by microsomal proteins in vitro and membrane calcium channel activation are unrelated events cannot be totally excluded and additional studies are required to substantiate this interpretation, Membrane protein phosphorylation has also been involved in the 1,25(OH),D, rapid stimulation of cardiac muscle Ca*+ uptake. Hormone-induced stimulation of the phosphorylation of 43 and 55 kDa microsomal proteins was observed (Selles and Boland, 199Ia, b). It is not known whether these heart proteins bind CAM. Differences in subunit composition of the Ca2+ channels from skeletal and cardiac muscle have been reported (Flockerzi et al., 1986; De Jongh et al., 1989). Finally, the possibility that 1,25(OHI,D, may affect other signalling systems, in addition to the CAMP-dependent pathway, should not be excluded. Recent work performed in our laboratory
indicates that in skeleta1 muscle the hormone affects protein kinase C activity (Massheimer and Boland, submitted for publication). Complex interactions between kinase systems have been involved in agonist signal transduction (Kikkawa et al., 1986).
Acknowledgements This work was supported by grants from the National Research Council of Argentina (CONICET) and Nestle Nutrition (Vevey, Switzerland).
References Andrew, C.G., Almon, R.R. and Appel, S.H. (1975) J. Biol. Chem. 250, 3972-3980. Baran, D.T. and Kelly, A.M. (1988) Endocrinology 122, 930934. Bikle, D.D. and Munson, S. (1985) J. Clin. invest. 76, 23122316. Bikle, D.D., Munson, S. and Chafouleas, J.G. (1984) FEBS Lett. 174, 30-33. Boland, A.R. de and Boland, R. (1985aJ Biochim. Biophys. Acta 845, 237-241. Boland, A.R. de and Boland, R. (1985b) Z. Naturforsch. 4Oc, 102-108. Boland, A.R. de and Boland, R. (1987) Endocrinolo~ 120, 1858-1864. Boland, R., Norman, A.W., Ritz, E. and Hasselbach, W. (1985) Biochem. Biophys. Res. Commun. 128, 305-311. Boland, A.R. de, Gallego, S. and Boland, R. (1983) Biochim. Biophys. Acta 733, 264-270. Boland, A.R. de, Massheimer, V. and Fernandez, L.M. (1988) Calcif. Tissue Int. 43, 370-375. Boland, A.R. de, Nemere, I. and Norman, A.W. (1989) Biochem. Biophys. Res. Commun. 166, 217-222. Braily, G. and Sperelakis, N. (1986) J. Cyclic Nucleotide Protein Phosphorylation Res. 11, 25-34. Conti, M.A. and Adelstein, R.S. (1981) J. Biol. Chem. 256, 3178-3181. Curry, O.B, Basten, J.F., Francis, M.J. and Smith, R. (1974) Nature 249, 83-87. De Jongh, K.S., Merrick, D.K. and Catterall, W.A (1989) Proc. Natl. Acad. Sci. USA 86, 8585-8589. Fernandez, L.M., Massheimer, V. and Boland, A.R. de (1990) Calcif. Tissue Int. 47, 314-319. Flockerzi, V., Ruth, P., Deken, J., Jung, B. and Hofmann, F. (1986) Horm. Cell Regul. 139, 235-248. Giuliani, D. and Boland, R. (1984) Calcif. Tissue Int. 36, 200-20s. Glenney, J.R. and Weber, J.C. (1980) J. Biol. Chem. 255, 10551-10554.
22 Kikkawa, U., Kitano, T., Saito, N., Kishimoto, A., Taniyama, K., Tanaka, C. and Nishizuka, Y. (19861 in Calcium and the Cell, p. 197, Willey, Chichester. Malenick, D.A. and Anderson, S.R. (19821 Biochemistry 21, 3480-3486. Massheimer, V., Fernandez, L.M. and Boland, A.R. de (1990) Z. Naturforsch, 45c, 663-670. Matthews, C., Heinberg, K.W., Ritz, E., Agostini, B., Fritsche, J. and Hasselbach, W. (1977) Kidney Int. 11, 227-235. Mieskes, G., Brand, I.A. and Soling, H.D. (1984) Eur. J. Biochem. 140, 375-383. Nemere, I. and Norman, A.W. (1987) J. Bone Miner. Res. 2, 99-107. Nelson, T.Y., Overweller, J.M., Chafouleas, J.G. and Boyd, A.E. (1983) Diabetes 32, 1126-1133. Laemmli, U.K. (1970) Nature 277, 680-685. Lieberherr, M. (1987) J. Biol. Chem. 262, 13168-13173. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randal, P.J. (1951) J. Biol. Chem. 193, 265-275. Paul, J. (1975) in Cell and Tissue Culture, p. 90, Churchill Livingstone, Edinburgh.
Selles, J. and Boland, R. (1991a) Mol. Cell. Endocrinol. 77, 67-73. Selles, J. and Boland, R. (1991b) Mol. Cell. Endocrinol. 82, 229-235. Simpson, R.U., Thomas, G.A. and Arnold, A.J. (1985) J. Biol. Chem. 260, 8882-8891. Sperelakis, N. (1984) Am. Heart J. 197, 347-357. Tung, H.Y.L., Resink, T.J., Hemmings, B.A., Shenolikan, S. and Cohen, P. (1984) Eur. J. Biochem. 138, 635-641. Vilar Palasi, C., Oshiro, D.L. and Kretsinger, R.H. (1983) Biochim. Biophys. Acta 757, 40-46. Walaas, IS., Horn, R.S., Nairn, A.C., Walaas, 0. and Adler, A. (1988) Arch. Biochem. Biophys. 262, 245-258. Walsh, D.A., Ashby, C., Gonzalez, C., Calkins, D., Fisher, E. and Krebs, E. (1971) J. Biol. Chem. 246, 1977-1985. Wasserman, R.H. and Taylor, A.N. (19731 J. Nutr. 103, 586599. Yusta, B., Ortiz-Caro, J., Pascual, A. and Aranda, A. (1988) Neurochemistry 51, 1801-1813.
