DEVELOPMENTAL
BIOLOGY
156,223-230 (1992)
Ca*+/Calmodulin-Dependent Phosphorylation of the 100-kDa Protein in Chick Embryonic Muscle Cells in Culture HYE SUN KIM, IHL HEE LEE, CHIN HA CHUNG,~ MAN-SIK Department of Molecular
Biology and SRC for Cell LX&rentiation,
KANG, AND Doo BONG HA
College of Natural Sciences, Seoul National
University, Seoul 151-742, Korea
Accepted December II, 1991 The pattern of protein phosphorylation was found to change in differentiating chick embryonic myoblasts in culture. The extent of phosphorylation of 42-, 50-, and lOO-kDa proteins increased while that of a 63-kDa protein declined in extracts of myoblasts that had been cultured for increasing periods. Of these, the increase in phosphorylation of the lOO-kDa protein occurred most dramatically in extracts of myoblasts in an early stage of differentiation and was specifically inhibited by trifluoperazine (TFP) and other calmodulin (CaM) antagonists including chlorpromazine and N-(6-aminohexyl)-5-chloro-l-naphthal~ne-sulfonamide (W-7). Treatment of increasing concentrations of TFP to culture medium also decreased the phosphorylation state of the lOO-kDa protein and the degree of myoblast fusion in parallel. In addition, levels of both the kinase activity and the lOO-kDa protein but not of CaM appeared to rise in the cells cultured for longer periods. These. results suggest that (1) a Caz+/CaM-dependent protein kinase is responsible for phosphorylation of the lOO-kDa protein, (2) the TFP-mediated myoblast fusion block may be associated with the inhibitory effect of the drug against the kinase activity, and (3) the increase in phosphorylation state of the lOO-kDa protein during myogenic differentiation is due to the rise in levels of the kinase and its substrate. o issz Academic FWSS, me.
CaM in vitro (Bar-Sagi and Prives, 1983). However, the mechanism of Ca2+ and CaM involvement in myoblast A prominent event in differentiation of skeletal musfusion is still unclear. cle cells is the fusion of mononucleated myoblasts into In many instances, the actions of Ca2+ are mediated multinucleated myotubes (Bischoff and Holtzer, 1969; by CaM (Cheung, 1980), which activates a number of O’Neill and Stockdale, 1972). Concurrent with the morenzymes including protein kinases that in turn phosphological changes, a large number of muscle-specific phorylate many crucial cell proteins (Schulman and proteins such as a-actin, myosin, creatine kinase, and Greengard, 1978). Several distinct Ca2+/CaM-dependent acetylcholine receptor are synthesized (Nadal-Ginard, protein kinases have been identified, and they are myo1978; Endo and Nadal-Ginard, 1987). Studies using cul- sin light chain kinase (Dabrowska et ml, 1978; Yagi et al, tured myoblasts have shown that Ca2’ plays an essen- 1978), phosphorylase kinase (Woodgett et al, 1982), CaM tial role in mediating myoblast fusion (David et cd, kinase I which phosphorylates synapsin I (Llinas et cd, 1981). Reduction in extracellular Ca2+ concentration 1985), CaM kinase II that phosphorylates a number of blocks myoblast fusion but the fusion is rapidly resumed protein substrates including glycogen synthase and miupon readdition of Ca2+ (Shainberg et a& 1969; Wakelam crotubule-associated protein 2 (Ahamd et al., 1982; Kenand Pette, 1982). An intracellular Ca2+-binding protein, nedy and Greengard, 1981), and CaM kinase III that uses calmodulin (CaM),2 has also been suggested to particia lOO-kDa protein as its exclusive substrate (Palfrey, pate in the regulation of myoblast fusion (Bar-Sagi and 1983; Nairn et aZ., 1985). In addition, it has recently been Prives, 1983). The phenothiazine trifluoperazine (TFP), found that the lOO-kDa protein exists in a variety of which binds to the Caz+-activated form of CaM (Levin mammalian cells and is elongation factor-2 (Nairn et al., and Weiss, 1978), inhibits myoblast fusion at concentra1985; Nairn and Palfrey, 1987; Ryazanov, 1987). tions that correspond closely to its antagonistic effect of As an attempt to elucidate the role of Ca2+ and CaM in myoblast fusion, we investigated the effect of TFP on phosphorylation of intracellular proteins in cultured 1 To whom all correspondence should be addressed. myoblasts. In the present study, we demonstrate the z Abbreviations: CaM, calmodulin; TFP, trifluoperazine; SDS, so- phosphorylation of a lOO-kDa protein increases during dium dodecyl sulfate; W-7, N-(6-aminohexyl)-5-chloro-l-naphthathe early stage of myoblast differentiation and is specifilene-sulfonamide; NEM, N-ethylmaleimide; EGTA, ethyleneglycolcally inhibited upon treatment of TFP. bis-( @aminoethyl ether)-N,N’-tetraacetic acid. INTRODUCTION
223
0012-1606/92 $3.00 Copyright All rights
0 1992 by Academic Press, Inc. of reproduction in any form reserved.
DEVELOPMENTALBIOLOGY VOLUME~~~,1992
224 MATERIALS
AND
METHODS
Cell Culture Myoblasts from breast muscle of 12-day-old chick embryos were prepared as described previously (Ha et al., 1979; Kwak et al, 1989). The cells were plated on collagen-coated culture dishes at a concentration of 5 X lo5 cells/ml in Eagle’s essential medium containing 10% horse serum, 10% chick embryo extract, and 1% antibiotic/antimycotic solution (GIBCO). One day after the cell seeding, the culture medium was changed with the same medium but containing 2% embryo extract. Degree of myoblast fusion was determined as described by O’Neill and Stockdale (1972).
been equilibrated with 50 mM Tris-HCl buffer (PH 7.5) containing 5 mlMMgCl,, 0.1 MNaCl, 1 mJfDTT, 0.5 mM EDTA, and 10% (v/v) glycerol. After collecting 4 ml of the flow-through fraction, the column was washed extensively with the same buffer. The protein bound to the column was then eluted with 4 ml of the buffer containing 0.2 M NaCl. Each fraction was concentrated to its original loading volume using Centricon (Amicon). RESULTS
Phosphorylation of Proteins in Cultured Myoblusts To determine if the pattern of protein phosphorylation in cultured myoblasts changes during differentiation, extracts were prepared from the cells that had been cultured for various periods and incubated for 2 Preparation of Cell Extract min in the presence of [T-~‘P]ATP. As shown in Fig. lA, Myoblasts cultured for appropriate periods were phosphorylation states of several proteins markedly washed three times with ice-cold phosphate-buffered changed in the extracts of differentiating myoblasts. In saline (PBS), harvested by centrifugation, and kept froparticular, phosphorylation of 42-, 50-, and lOO-kDa zen at -70°C until use. The cells were disrupted by soni- proteins increased during the early period of differencation for 30 set in 50 mM Tris-HCI buffer (pH 7.5) tiation while that of a 63-kDa protein declined gradually containing 5 mM MgCl,, 1 mlM dithiothreitol (DTT), 0.5 with time of culture. Since the increase in phosphorylam2M EDTA, and 10% (v/v) glycerol. After centrifugation was most evident in the lOO-kDa band, particularly tion at 15,OOOgfor 30 min, both supernatants and precipiduring the periods prior to myoblast fusion, we focused tates were collected and are referred to as cell extracts our further studies on this protein. and particulate fractions, respectively. The precipitates To examine the incubation time-dependency of the were washed twice and resuspended in the same buffer. lOO-kDa protein phosphorylation, extracts from myoblasts cultured for 24 and 48 hr were incubated for inPhosphorylation Assay creasing periods. After electrophoresis of the extracts The assays were performed by incubating reaction in the presence of SDS, the bands corresponding to the mixtures in final volumes of 50 ~1 containing 1.5 &i of lOO-kDa protein were cut off from the gels and counted [-r-32P]ATP (6000 Ci/mmole, New England Nuclear), 1 for their radioactivity. The extents of phosphorylation mM ATP, and 30 pg of cell extracts or particulate fracwere found to increase linearly with time until reaching tions obtained from cultured myoblasts. Effects of CaM, maximal levels by 5 min of incubation (data not shown). EGTA, TFP, chlorpromazine, W-7, and/of Walsh inhibiWe then quantified the changes in the phosphorylation tor (Sigma) were examined by treating them with the state of the lOO-kDa protein, which were revealed by the assay mixtures 30 min prior to initiation of the reaction incubation of the extracts obtained from cells cultured by adding ATP. After incubating the mixtures at 20°C for various periods (see Fig. 1A). Figure 1B again shows for proper periods, the reaction was terminated by addithat the extent of the lOO-kDa protein phosphorylation tion of 50 ~1 of 4% sodium dodecyl sulfate (SDS). dramatically rises in the cells cultured for 24-48 hr and The samples were then electrophoresed in 7-14% polythat the extent of the protein phosphorylation at 48 hr is acrylamide gradient gels in the presence of SDS at least sevenfold greater than that at 24 hr. In addition, (Laemmli, 1970). The gels were stained with Coomassie it is noteworthy that the phosphorylation state of the R250, dried under vacuum, and exposed to X-ray films lOO-kDa protein in cells cultured for 60 hr was repeat(Fuji). After the autoradiography, the gel bands that edly observed to be 5-20% lower than that in cells culcorrespond to lOO-kDa size were cut out, incubated in 1 tured for 48 or 72 hr; however, the reasons are presently ml each of H,Oz at 80°C for 24-48 hr, and counted for unknown. Under the same culture conditions, myoblasts their radioactivity using a scintillation counter (Packbegan to engage in active fusion at about 48 hr. These ard). Proteins were assayed as described by Bradford results suggest that the increase in phosphorylation of (1976). the lOO-kDa protein is differentiation stage specific. Separation of Kinase Fraction from Its Substrate in Cell Phosphwrylaticm of the 100-kDa Protein Is Extracts Ca’+/CaM-Dependent Cell extracts were prepared as above, and loaded on a To determine the type of protein kinase(s) that is reDEAE-cellulose column (2 ml bed volume) that had sponsible for phosphorylation of the lOO-kDa protein,
KIM ET AL.
A
24
36
40
60
225
CaM-kinase in Cultured Myoblasts
B
72
kDa 11697-
.
66-
0.0
I 24
I 40 CULTURE
TIME
I 72
0
(hr)
FIG. 1. Changes in the pattern of protein phosphorylation in differentiating myoblasts. Chick embryonic myoblasts were cultured and their extracts were prepared as described under Materials and Methods. Aliquots of the samples (30 pg each) were incubated with [-y-“PjATP at 20°C for 2 min, electrophoresed in the presence of SDS, dried, and autoradiographed (A). The arrowhead and arrows indicate the proteins showing significant changes in their phosphorylation state. After the autoradiography, the bands that correspond to the lOO-kDa protein were cut off from the dried gel and their radioactivity (0) was estimated as described in the text (B). Phosphorylation state of the lOO-kDa protein was maximal in the extract of cells cultured for 72 hr and was expressed as 1.0. Under the same culture conditions, the degree of myoblast fusion (0) was also determined.
extracts of myoblasts cultured for 72 hr were incubated in the presence and absence of Walsh inhibitor, an inhibitor of the CAMP-dependent protein kinase (Cheung et aZ., 1986) or CaM antagonists, such as TFP, chlorpromazine, and W-7 (Levin and Weiss, 1978). As shown in Fig. 2 (lane b), Walsh inhibitor showed little effect on the lOO-kDa protein phosphorylation. Addition of CAMP to the reaction mixture also showed no effect (data not a
b
c
d
e
kDa ZOS-
1169766-
4530-
shown). Thus, it appears unlikely that phosphorylation of the lOO-kDa protein is mediated by the CAMP-dependent system. Treatment of TFP, however, strongly inhibited phosphorylation of the lOO-kDa protein but not of the other proteins (Fig. 2, lane c). Half-maximal inhibition was observed at a concentration of about 25 &f TFP. Other CaM antagonists, such as chlorpromazine and W-7, also similarly inhibited the protein phosphorylation (lanes d and e, respectively). We then tested if treatment of exogenous CaM can reverse the inhibitory effect of TFP. As shown in Fig. 3, addition of CaM to the TFP-treated extract recovers phosphorylation of the lOO-kDa protein but to a level higher than that seen without the addition. Treatment of CaM alone also increased the extent of phosphorylation to a similar level. However, elimination of Ca2+ from the reaction mixture by treating with EGTA blocked phosphorylation of the lOO-kDa protein whether or not CaM was added. These results indicate that Ca2+/CaM-dependent protein kinase(s) is responsible for phosphorylation of the lOO-kDa protein. TFP-mediated Inhibition of Myoblast Fusion and Phosphcnylation of the NO-kDa Protein
FIG. 2. Effects of Walsh inhibitor and CaM antagonists on phosphorylation of the lOO-kDa protein. Extracts were prepared from myoblasts cultured for 72 hr and assayed for protein phosphorylation in the absence (lane a) and presence of 1 rg/ml of Walsh inhibitor(b), 50 &fTFP (c), 200 PMchlorpromazine (d), or 200 pM W-7 (e). The arrowhead indicates the position where the NO-kDa protein migrated.
TFP has been reported to inhibit myoblast fusion without affecting the synthesis of muscle-specific proteins (Bar-Sagi and Prives, 1983). Therefore, we tested the possibility that the TFP-mediated fusion block is associated with the inhibition of the lOO-kDa protein phosphorylation by the drug. Myoblasts were treated with increasing concentrations of TFP at 24 hr of the
226
DEVELOPMIENTAL BIOLOGY
EGTA
-
+
-
-
-
+
TFP
-
-
+
-
+
-
cat-A
-
-
-
-I.
f
+
FIG. 3. Inhibition of the NO-kDa protein phosphorylation by EGTA and TFP and its reversal by CaM. Extracts were prepared from myoblasts cultured for 72 hr and assayed for protein phosphorylation in the absence (-) and presence (+) of 10 mMEGTA, 50 j&fTFP, and/or 10 pg/ml of CaM. The arrowhead indicates the lOO-kDa protein.
culture. After culturing them for the next 48 hr, extracts were prepared and assayed for their capacity to phosphorylate the lOO-kDa protein. Also, the cells cultured under the same conditions were fixed and examined under a microscope to determine the degree of fusion. As shown in Fig. 4A, treatment with higher concentrations of TFP decreased more pronouncedly the phosphorylation state of the lOO-kDa protein but without affecting that of other proteins. In addition, TFP inhibited the myoblast fusion in a dose-dependent manner (Fig. 4B) in accord with the earlier report by Bar-Sagi and Prives (1983). Moreover, the fall in the extent of phosphorylation occurred in parallel with the decrease in the degree of myoblast fusion. Thus, it appears that the TFP-mediated inhibition of the lOO-kDa protein phosphorylation in cultured myoblasts is closely correlated with the block of myoblast fusion induced by the drug. Factors Regulating 100-kDa Protein
the Phosphmylatim
State of the
As was shown in Fig. 3, addition of exogenous CaM to the TFP-treated extract resulted in an increased phos-
VOLUME 150,199Z
phorylation of the lOO-kDa protein to a level higher than that seen without the treatment. This result indicates that CaM is limiting in the cell extracts. This observation also suggests that myoblasts cultured for 24 hr may contain lower amount of CaM than the fused cells, and therefore their extracts might have shown the reduced phosphorylation of the lOO-kDa protein (see Fig. 1). To test this possibility, CaM was added to each extract of the cells cultured for 24 and 72 hr and its effects were compared (Table 1). Addition of CaM significantly increased the abilities of both extracts in phosphorylation of the lOO-kDa protein. However, the effect of CaM was rather proportional and the difference in the extent of phosphorylation in the extracts remained unchanged. Thus, it is unlikely that CaM acts as a limiting factor in regulating the phosphorylation state of the lOO-kDa protein in cultured myoblasts. At least two other factors that can directly affect phosphorylation of the lOO-kDa protein are Ca’+/CaMdependent protein kinase and/or its 100-kDa substrate, whose levels may change in differentiating myoblasts. To test this possibility, we first fractionated the extracts obtained from the cells that had been cultured for 24 and 72 hr using a DEAE-cellulose column. As shown in Table 2, either fraction that eluted from or bound to the column did not reveal any phosphorylation of the lOO-kDa protein. However, the phosphorylation became evident when the two fractions from the extract of 72hr-cultured cell were added together. This result clearly indicates that the ion-exchange chromatography separates the kinase fraction from its lOO-kDa protein substrate. The NEM-sensitive, Ca’+/CaM-dependent protein kinase was then found to be present in the DEAEbound fraction upon further purification of the enzyme using Sephacryl S-200 and CaM-Sepharose affinity columns (unpublished observations).5 Thus, the lOO-kDa protein should be present in the DEAE-flow-through fraction. We then examined by the reconstitution experiments described below if the kinase and its lOO-kDa substrate levels change during myogenesis. When the DEAEflow-through substrate fraction from 24-hr-cultured cell extracts was incubated with the DEAE-bound kinase fraction from ‘72-hr cells, the extent of the protein phosphorylation was significantly increased compared to that from incubation of the kinase and its substrate fractions, both of which were from 24-hr cells (Table 2). However, the extent of phosphorylation did not reach that shown with the kinase and its substrate fractions,
* Jeon, Y. J., Kim, H. S., Chung, C. H., and Ha, D. B. (1991). Unpublished observations.
KIM ET AL. A
a
b
c
CaM-kinme
227
in Cultured Myoblwts
d
0.0
-
5
0 TFP
10
(uM)
FIG. 4. Effect of TFP-treatment to cultured myoblasts on phosphorylation of the lOO-kDa protein. Myoblasts that had been cultured for 24 hr were treated with 0 (lane a), 1 (b), 5 (c), and 10 p&ITFP (d) and further incubated for the next 48 hr. (A) The cells were washed three times with ice-cold PBS to eliminate residual TFP in the culture medium. Extracts were prepared from the cells, and their ability of protein phosphorylation was determined as in Fig. 1. The arrowhead indicates the 100~kDa protein. (B) Radioactivity in the lOO-kDa protein bands (0) was estimated, and that seen in the absence of TFP was expressed as 1.0. Myoblasts cultured under the same conditions were fixed, and their degree of fusion (0) was determined as described under Materials and Methods.
both of which were from 72-hr cells. These results suggest that the level of the lOO-kDa protein in 24-hr cells is lower than that in ‘72-hr cells and increases as myoblast cell differentiation proceeds. On the other hand, incubation of the kinase fraction from 24-hr cells with the substrate fraction from 72-hr cells led to little or no increase in the extent of the NO-kDa protein phosphorylation compared to that from both fractions from 24-hr cells. These results clearly suggest that the activity level of kinase in 24-hr cells is much less than that in 72-hr cells. Thus, it appears likely that the levels of both the lOO-kDa protein and the kinase activity are subjected to change during the early period of myogenic differentiation although the extent of level of change seems to differ.
Properties of Ca’+/CaM-Dependmt Protein Kinase The above studies used only the soluble extract of cultured myoblasts in determining the activity of Ca’+/ CaM-kinase. Therefore, we tested if the kinase activity also exists in particulate fractions. When the particulate fractions from the cells cultured for 72 hr were incubated, the extent of the lOO-kDa protein phosphorylation was much less significant than that of the soluble extracts alone (Fig. 5). Furthermore, little or no change in the extent of the protein phosphorylation was observed when the particulate fractions were incubated
TABLE
TABLE 1 EFFECT
DEAE-fractions
OF THE TREATMENT OF CaM ON PHOSPHORYLA~ON OF THE lOO-kDa PROTEIN
Radioactivity in lOO-kDa protein (cpm) CaM h/ml)
24 hr
72 hr
Ratio 72 hr/24 hr
0 10 20
70 132 145
301 645 698
4.3 4.9 4.8
Note. Extracts were prepared from myoblasts that had been cultured for 24 and 72 hr and assayed for their ability in phosphorylation of the 100~kDa protein in the presence of increasing concentrations of CaM. Radioactivity in lOO-kDa protein bands were then determined as described in Fig. 1.
2
SEPARATION OF Ca*+/CaM-DEPaJDEWr KINASE FROM ITS lOO-kDa PROTEIN SUEWRATE
72 hr 72 hr ‘72 hr 24 hr 24 hr 24 hr 24 hr 24 hr
flow through bound flow through flow through bound flow through flow through bound + 72
(30 pg each)
+ 72 hr bound + 24 hr bound + 72 hr bound hr flow through
ACWITY
Radioactivity in lOO-kDa protein (cpm) 9 4 550 0 5 30 220 44
Note. Extracts prepared from myoblasts cultured for 24 and 72 hr were fractionated using a DEAE-cellulose column (1 X 2.5 cm) as described under Materials and Methods. The flow through and bound fractions were assayed for their ability in phosphorylation of the lOOkDa protein in the presence of CaM (10 pg/ml). Radioactivity in lOOkDa protein bands was then determined as described in Fig. 1.
228
DEVELOPMENTAL BIOLOGY a
b
c
d
e
f
kDa
11697-
VOLUME 150.1992
kinase activity is indeed sensitive to inhibition by the salt, phosphorylation assays were performed in the presence of increasing concentrations of NaCI. As shown in Fig. 7, NaCl inhibited the phosphorylation activity against the lOO-kDa protein in a dose-dependent manner but without affecting that against other proteins. KC1 also blocked specifically the lOO-kDa protein phosphorylation (data not shown). Thus, the Ca2+/CaM-kinase activity found in the present study appears to be sensitive to inhibition by monovalent cations.
45-
DISCUSSION 30-
FIG. 5. Phosphorylation of membranous proteins from cultured myoblasts. Soluble and particulate fractions were obtained from myoblasts cultured for ‘72 hr as described under Materials and Methods. The kinase and its lOO-kDa substrate fractions in the soluble extracts were separated using a DEAE-column as in Table 2. Incubations were performed at 20°C for 2 min with [y-q]ATP and 30 pg each of the following: substrate fraction alone (a), kinase fraction alone (b), substrate and kinase fractions (c), particulate fraction alone (d), substrate and particulate fractions (e), kinase and particulate fractions (f). After the incubation, the samples were electrophoresed in the presence of SDS, dried, and autoradiographed.
with either the DEAE-separated kinase or its substrate fraction from the soluble extract of cells cultured for 72 hr. Similar data were obtained with the particulate fractions from 24- or 48-hr cells. These results indicate that Ca2+/CaM-kinase as well as its lOO-kDa protein substrate are primarily localized to the soluble fraction of the cells. As a preliminary attempt to elucidate the nature of the amino acid(s) that is phosphorylated by Ca2+/CaMkinase, an extract from myoblasts cultured for ‘72 hr was subjected to SDS-polyacrylamide gel electrophoresis in duplicate after the incubation with [T-~P]ATP. One of the gels was dried and autoradiographed. The other gel was also treated as above but after the incubation with 1 N NaOH. Figure 6 shows that the alkali-treatment dephosphorylates nearly all of the phosphoproteins except the lOO-kDa protein. It has been reported that phosphoThr and phospho-Tyr but not phospho-Ser are resistant to dephosphorylation by alkali-treatment (Cheung and Chen, 1981). Thus, it appears likely that Ca2+/CaM-kinase preferentially phosphorylates the Thr and/or Tyr residues in its lOO-kDa protein substrate. During separation of the kinase fraction from its lOOkDa protein substrate using DEAE-cellulose chromatography (see Table 2), the kinase activity eluted with 0.21K NaCl was found to increase several-fold after dialysis against the column buffer without the salt. To test if the
The present studies have demonstrated that phosphorylation of the lOO-kDa protein occurs in soluble extracts of cultured myoblasts and is Ca2+/CaM-dependent. This kinase activity appears to be distinct from that of CaM-kinase I or II by a number of factors: (1) It uses the lOO-kDa protein as its major, if not exclusive, substrate while both CaM-kinase I and II do not seem to phosphorylate proteins with similar size (Ahamd et uZ., 1982; Kennedy and Greengard, 1981; Llinas et al, 1985). (2) It appears to preferentially phosphorylate the Thr and/or Tyr residues while the phosphorylation sites by CaM-kinases I and II include Ser (Huttner et al, 1981; Kennedy and Greengard, 1981). (3) Its size (about llO140 kDa) under nondenaturing conditions (unpublished observations)3 is significantly different from that of CaM-kinase I or II (MeGuinness et al, 1985; Huttner et al, 1981). On the other hand, the Ca2+/CaM-dependent kinase from cultured myoblasts appears to share a number of similar properties with CaM-kinase III: (1) Both a
b
FIG. 6. Effect of alkali-treatment on dephosphorylation of the lOOkDa protein. Protein phosphorylation was carried out using the extracts prepared from myoblasts cultured for 72 hr. After the electrophoresis of the reaction sample, the gels were incubated at 37“C for 1 hr in deionized water (a) or 1 NNaOH (b). The gels were then autoradiographed. The arrowhead indicates the 100~kDa protein.
KIM ET AL. a
b
CaM-kinme in Cultured Myoblasts
c
FIG. 7. Effects of monovalent cations on phosphorylation of the 100kDa protein. Extracts were prepared from myoblasts cultured for 72 hr and assayed for protein phosphorylation in the presence of 0 (lane a), 50 (b), and 100 mM NaCl (c).
use same size, lOO-kDa protein as their substrates. (2) Both are sensitive to inhibition by Na+ ion (Togari and Guroff, 1985). (3) Both of their phosphorylated substrates are resistant to alkali-treatment (Togari and Guroff, 1985). (4) Both have a similar size under nondenaturing condition (Ryazanov et aZ., 1988; Togari and Guroff, 1985; Nairn et al, 1985). However, more studies are necessary for clarification of the identities of both the enzymes and their lOO-kDa substrates. TFP and other CaM antagonists have been reported to block myoblast fusion, and their effects have been attributed to their antagonistic action against CaM, thereby preventing the pivotal role of Ca” in the fusion process (Bar-Sagi and Prives, 1983). In the present study, we showed that TFP and the other CaM antagonists including W-7 and chlorpromazine selectively block the phosphorylation of the lOO-kDa protein (Fig. 2). In addition, extent of the lOO-kDa protein phosphorylation was found to sharply increase in extracts of myoblasts in rapid proliferation and alignment and reach a maximal level in myoblasts of the cells in active fusion (Fig. 1). Furthermore, treatment of the cultured cells with increasing concentrations of TFP resulted in a parallel inhibition of both myoblast fusion and phosphorylation of the lOO-kDa protein, but without affecting the phosphorylation state of other cell proteins (Fig. 4). Thus, it is tempting to speculate that TFP-mediated inhibition of the lOO-kDa protein phosphorylation may be responsible for the block of myoblast fusion, although the role(s) of the lOO-kDa protein in myoblast fusion remains totally unknown. It is noteworthy, however, that, for obtaining halfmaximal inhibition of the lOO-kDa protein phosphory-
229
lation, approximately fivefold higher concentration of TFP (25 PM) was required compared to that (5 pLM) for cultured cells (Fig. 4). The TFP concentration reported to inhibit CaM activity in vitro (ID, = 10 &V; Weiss and Levin, 1978) is also at least twofold greater than that required for the half-maximal inhibition of myoblast fusion (Fig. 4; Bar-Sagi and Prives, 1983). A possible explanation for these phenomena is that TFP may accumulate in the cells when added to culture medium. Although the TFP-induced fusion block can be reversed by drug removal, the recovery period appears to be much longer than that revealed by treatment-and-removal of other fusion blockers, such as EGTA (Bar-Sagi and Prives, 1983; Paterson and Strohman, 1972). Moreover, TFP is known to bind tightly to purified Ca2+-activated form of CaM with a binding constant of 1 &f(Levin and Weiss, 1977). Therefore, it seems that the TFP remained tightly bound to CaM was able to prevent phosphorylation of the lOO-kDa protein even after the removal of the drug from culture medium. It has been reported that Ca2+ influx in cultured myoblasts precedes membrane fusion (David et cd, 1981). Therefore, it is possible that the increase in intracellular Ca2+ concentration may induce activation of the Ca2+/CaM-dependent protein kinase(s) and consequently increase the extent of the 100~kDa protein phosphorylation. However, it is unlikely that the increased phosphorylation is mediated by alterations in the cells’ content of CaM, that may occur during myogenesis, since exogenous addition of CaM to extracts of proliferating myoblasts could not elevate the phosphorylation to the level of differentiated cells (Table 1). On the other hand, using the DEAE-separated kinase and its substrate preparations, we were able to demonstrate that the levels of both the kinase activity and its substrate increase during the early period of myogenic differentiation. Therefore, it appears that increase in the phosphorylation state of the lOO-kDa protein is at least in part due to differentiation-dependent changes in the levels of Ca2+/CaM-dependent protein kinase and its lOOkDa protein substrate. However, because these level changes were determined by the phosphorylation assay, we cannot exclude the possibility that other factors, such as endogenous activator or inhibitor of the kinase, may influence the phosphorylation state of the lOO-kDa protein in a differentiation-stage-specific manner. Thus, purification of the kinase and its substrate and preparation of antibodies against each protein are necessary to clarify the differentiation-specific regulation of the phosphorylation state of the,lOO-kDa protein in embryonic myoblasts in culture, and are in progress. We are grateful to Dr. Y. S. Kang (Kyung-Sang University) for his helpful discussion and to Ms. Y. J. Jeon for her help in some experi-
230
DEVELOPMENTALBIOLOGY
ments. This work was supported by grants from Korea Science and Engineering Foundation and the Ministry of Education. REFERENCES Ahamd, Z., Depaoli-Roach, A. A., and Roach, P. J. (1982). Purification and characterization of a rabbit liver calmodulin-dependent protein kinase able to phosphorylate glycogen syntbase. J. Bid &em, 257, 8348-8355. Bar-Sagi, D., and Prives, J. (1983). Trifluoperazine, a calmodulin antagonist, inhibits muscle cell fusion. J. Cell BioL 97,1375-1380. Bischoff, R., and Holtzer, H. (1969). Mitosis and process of differentiation of myogenic cells in vitro. J. Cell Biol 41,188-200. Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. And Biochem. 72,248-254. Cheung, H.-C., Kemp, B. E., Persons, R. B., Smith, A. J., Misconi, L., Pattern, S. M. V., and Walsh, D. A. (1986). A potent synthetic peptide inhibitor of the CAMP-dependent protein kinase. J. BioL Chew. 261,989-992. Cheung, W. Y. (1980). Calmodulin plays a pivotal role in cellular regulation. Science 207,19-27. Cheung, Y. E., and Chen, L. B. (1981). Detection of phosphotyrosinecontaining 34,000-dalton protein in the framework of cells transformed with Rous sarcoma virus. Proc NatL Acad. Sci. USA 78, 2388-2392. Dabrowska, R., Sherry, J. M. F., Aromatorio, D. K., and Hartshorne, D. J. (1978). Modulator protein as a component of the myosin light chain kinase from chicken gizzard. Biochemistry 17.253-258. David, J. D., See, W. M., and Higginbotham, C. A. (1981). Fusion of chick embryo skeletal myoblasts: Role of calcium influx preceding membrane union. Dev. Biol. 82,297-307. Endo, T., and Nadal-Ginard, B. (1987). Three types of muscle-specific gene expression in fusion-blocked rat skeletal muscle cells: Translational control in EGTA-treated cells. Cell 49,515-526. Ha, D. B., Boland, R., and Martonosi, A. (1979). Synthesis of the calcium transport ATPase of sarcoplasmic reticulum and other muscle cells in viwo and in vitro. Biochim Biophys. Ada 585,165-187. Huttner, W. B., DeGennaro, L. J., and Greengard, P. (1981). Differential phosphorylation of multiple sites in purified protein I by cyclic AMP-dependent and calcium-dependent protein kinases. J. BioL Chem. 256,1482-1488. Kennedy, M. B., and Greengard, P. (1981). Two calcium/calmodulindependent protein kinases, which are highly concentrated in brain, phosphorylate protein I at distinct site. Proc. NatL Acad Sci USA 78,1293-1297. Kwak, K. B., Lee, Y. S., Suh, S. W., Chung, C. S., Ha, D. B., and Chung, C. H. (1989). Purothionin from wheat endosperm reversibly blocks myogenic differentiation of chick embryonic muscle cells in culture. Eqn. Cell Res. 183,501-507. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680-685. Levin, R. M., and Weiss, B. (1977). Binding of trifluoperazine to the calcium-dependent activator of cyclic nucleotide phosphodiesterase. Mol PhammoL 13,690-697. Levin, R. M., and Weiss, B. (1978). Specificity of binding of trifluoper-
VOLUME150,1992 azine to the calcium-dependent activator of phosphodiesterase and to a series of other calcium-binding proteins. B&him Biqphys Acta 540,197-204. Llinas, R., McGuinness, T. L., Leonard, C. S., Sugimori, M., and Greengard, P. (1985). Intraterminal injection of synapsin I or calcium/ calmodulin-dependent protein kinase II alters neurotransmitter release at the squid giant synapse. Proc. NatL Acad Sci. USA 82,30353039. McGuinness, T. L., Lai, Y., and Greengard, P. (1985). Ca*+/calmodulindependent protein kinase II. J. Bid, Chew. 260,1696-1704. Nadal-Ginard, B. (1978). Commitment, fusion and biochemical differentiation of a myogenic cell line in the absence of DNA synthesis. Cell 15,855-864. Nairn, A. C., Bhagat, B., and Palfrey, C. (1985). Identification of calmodulin-dependent protein kinase III and its major Mr 100,000 substrate in mammalian tissues. Proc. NatL Acm! Sci USA 82,79397943. Nairn, A. C., and Palfrey, H. C. (1987). Identification of the major Mr 100,000 substrate for calmodulin-dependent protein kinase III in mammalian cells as elongation factor-2. J. BioL Chem 262, 1729917303. O’Neill, M. C., and Stockdale, F. E. (1972). A kinetic analysis of myogenesis in vitro. J. Cell BioL 52,52-65. Palfrey, H. C. (1983). Presence in many mammalian tissues of an identical major cytosolic substrate (Mr 100,000) for calmodulin-dependent protein kinase. FEBS I&. 157,183-190. Paterson, B., and Strohman, R. C. (1972). Myosin synthesis in cultures of differentiating chicken embryo skeletal muscle. Deu. BioL 29, 113-138. Ryazanov, A. G. (1987). Ca*+/calmodulin-dependent phosphorylation of elongation factor 2. FEBS L&t. 214,331-334. Ryazanov, A. G., Shestakova, E. A., and Natapov, P. G. (1988). Phosphorylation of elongation factor 2 by EF-2 kinase affects rate of translation. Nature 334,170-173. Schulman, H., and Greengard, P. (1978). Ca2+-dependent protein phosphorylation system in membranes from various tissues, and its activation by “calcium-dependent regulator”. Proc NatL Acad Ski. USA 75,5432-5436. Shainberg, A., Yagil, G., and Yaffe, D. (1969). Control of myogenesis in vitro by Ca2+ concentration in nutritional medium. Exp. CeU Res. 58, 163-16’7. Togari, A., and Guroff, G. (1985). Partial purification and characterization of a nerve growth factor-sensitive kinase and its substrate from PC12 cells. J. Bid Chem. 260,3804-3811. Wakelam, M. J., and Pette, D. (1982). The breakdown of phosphatidylinositol in myoblasts stimulated to fuse by the addition of Ca’+. B&hem. J. 202,723-729. Weiss, B., and Levin, R. M. (1978). Mechanism for selectivity inhibiting the activation of cyclic nucleotide phosphodiesterase and adenylate cyclase by antipsychotic agents. A&v. Cyclic Nwleotide Res. 9, 285-304. Woodgett, J. R., Tonks, N. K., and Cohen, P. (1982). Identification of a calmodulin-dependent glycogen synthase kinase in rabbit skeletal muscle, distinct from phosphorylase kinase. FEBS II&t. 148,5-11. Yagi, K., Yazawa, M., Kakiuchi, S., Ohshima, M., and Uenishi, K. (1978). Identification of an activator protein for myosin light chain kinase as the Ca2+-dependent modulator protein. J. Bid Chem 253, 1338-1340.
BIOLOGY
156,223-230 (1992)
Ca*+/Calmodulin-Dependent Phosphorylation of the 100-kDa Protein in Chick Embryonic Muscle Cells in Culture HYE SUN KIM, IHL HEE LEE, CHIN HA CHUNG,~ MAN-SIK Department of Molecular
Biology and SRC for Cell LX&rentiation,
KANG, AND Doo BONG HA
College of Natural Sciences, Seoul National
University, Seoul 151-742, Korea
Accepted December II, 1991 The pattern of protein phosphorylation was found to change in differentiating chick embryonic myoblasts in culture. The extent of phosphorylation of 42-, 50-, and lOO-kDa proteins increased while that of a 63-kDa protein declined in extracts of myoblasts that had been cultured for increasing periods. Of these, the increase in phosphorylation of the lOO-kDa protein occurred most dramatically in extracts of myoblasts in an early stage of differentiation and was specifically inhibited by trifluoperazine (TFP) and other calmodulin (CaM) antagonists including chlorpromazine and N-(6-aminohexyl)-5-chloro-l-naphthal~ne-sulfonamide (W-7). Treatment of increasing concentrations of TFP to culture medium also decreased the phosphorylation state of the lOO-kDa protein and the degree of myoblast fusion in parallel. In addition, levels of both the kinase activity and the lOO-kDa protein but not of CaM appeared to rise in the cells cultured for longer periods. These. results suggest that (1) a Caz+/CaM-dependent protein kinase is responsible for phosphorylation of the lOO-kDa protein, (2) the TFP-mediated myoblast fusion block may be associated with the inhibitory effect of the drug against the kinase activity, and (3) the increase in phosphorylation state of the lOO-kDa protein during myogenic differentiation is due to the rise in levels of the kinase and its substrate. o issz Academic FWSS, me.
CaM in vitro (Bar-Sagi and Prives, 1983). However, the mechanism of Ca2+ and CaM involvement in myoblast A prominent event in differentiation of skeletal musfusion is still unclear. cle cells is the fusion of mononucleated myoblasts into In many instances, the actions of Ca2+ are mediated multinucleated myotubes (Bischoff and Holtzer, 1969; by CaM (Cheung, 1980), which activates a number of O’Neill and Stockdale, 1972). Concurrent with the morenzymes including protein kinases that in turn phosphological changes, a large number of muscle-specific phorylate many crucial cell proteins (Schulman and proteins such as a-actin, myosin, creatine kinase, and Greengard, 1978). Several distinct Ca2+/CaM-dependent acetylcholine receptor are synthesized (Nadal-Ginard, protein kinases have been identified, and they are myo1978; Endo and Nadal-Ginard, 1987). Studies using cul- sin light chain kinase (Dabrowska et ml, 1978; Yagi et al, tured myoblasts have shown that Ca2’ plays an essen- 1978), phosphorylase kinase (Woodgett et al, 1982), CaM tial role in mediating myoblast fusion (David et cd, kinase I which phosphorylates synapsin I (Llinas et cd, 1981). Reduction in extracellular Ca2+ concentration 1985), CaM kinase II that phosphorylates a number of blocks myoblast fusion but the fusion is rapidly resumed protein substrates including glycogen synthase and miupon readdition of Ca2+ (Shainberg et a& 1969; Wakelam crotubule-associated protein 2 (Ahamd et al., 1982; Kenand Pette, 1982). An intracellular Ca2+-binding protein, nedy and Greengard, 1981), and CaM kinase III that uses calmodulin (CaM),2 has also been suggested to particia lOO-kDa protein as its exclusive substrate (Palfrey, pate in the regulation of myoblast fusion (Bar-Sagi and 1983; Nairn et aZ., 1985). In addition, it has recently been Prives, 1983). The phenothiazine trifluoperazine (TFP), found that the lOO-kDa protein exists in a variety of which binds to the Caz+-activated form of CaM (Levin mammalian cells and is elongation factor-2 (Nairn et al., and Weiss, 1978), inhibits myoblast fusion at concentra1985; Nairn and Palfrey, 1987; Ryazanov, 1987). tions that correspond closely to its antagonistic effect of As an attempt to elucidate the role of Ca2+ and CaM in myoblast fusion, we investigated the effect of TFP on phosphorylation of intracellular proteins in cultured 1 To whom all correspondence should be addressed. myoblasts. In the present study, we demonstrate the z Abbreviations: CaM, calmodulin; TFP, trifluoperazine; SDS, so- phosphorylation of a lOO-kDa protein increases during dium dodecyl sulfate; W-7, N-(6-aminohexyl)-5-chloro-l-naphthathe early stage of myoblast differentiation and is specifilene-sulfonamide; NEM, N-ethylmaleimide; EGTA, ethyleneglycolcally inhibited upon treatment of TFP. bis-( @aminoethyl ether)-N,N’-tetraacetic acid. INTRODUCTION
223
0012-1606/92 $3.00 Copyright All rights
0 1992 by Academic Press, Inc. of reproduction in any form reserved.
DEVELOPMENTALBIOLOGY VOLUME~~~,1992
224 MATERIALS
AND
METHODS
Cell Culture Myoblasts from breast muscle of 12-day-old chick embryos were prepared as described previously (Ha et al., 1979; Kwak et al, 1989). The cells were plated on collagen-coated culture dishes at a concentration of 5 X lo5 cells/ml in Eagle’s essential medium containing 10% horse serum, 10% chick embryo extract, and 1% antibiotic/antimycotic solution (GIBCO). One day after the cell seeding, the culture medium was changed with the same medium but containing 2% embryo extract. Degree of myoblast fusion was determined as described by O’Neill and Stockdale (1972).
been equilibrated with 50 mM Tris-HCl buffer (PH 7.5) containing 5 mlMMgCl,, 0.1 MNaCl, 1 mJfDTT, 0.5 mM EDTA, and 10% (v/v) glycerol. After collecting 4 ml of the flow-through fraction, the column was washed extensively with the same buffer. The protein bound to the column was then eluted with 4 ml of the buffer containing 0.2 M NaCl. Each fraction was concentrated to its original loading volume using Centricon (Amicon). RESULTS
Phosphorylation of Proteins in Cultured Myoblusts To determine if the pattern of protein phosphorylation in cultured myoblasts changes during differentiation, extracts were prepared from the cells that had been cultured for various periods and incubated for 2 Preparation of Cell Extract min in the presence of [T-~‘P]ATP. As shown in Fig. lA, Myoblasts cultured for appropriate periods were phosphorylation states of several proteins markedly washed three times with ice-cold phosphate-buffered changed in the extracts of differentiating myoblasts. In saline (PBS), harvested by centrifugation, and kept froparticular, phosphorylation of 42-, 50-, and lOO-kDa zen at -70°C until use. The cells were disrupted by soni- proteins increased during the early period of differencation for 30 set in 50 mM Tris-HCI buffer (pH 7.5) tiation while that of a 63-kDa protein declined gradually containing 5 mM MgCl,, 1 mlM dithiothreitol (DTT), 0.5 with time of culture. Since the increase in phosphorylam2M EDTA, and 10% (v/v) glycerol. After centrifugation was most evident in the lOO-kDa band, particularly tion at 15,OOOgfor 30 min, both supernatants and precipiduring the periods prior to myoblast fusion, we focused tates were collected and are referred to as cell extracts our further studies on this protein. and particulate fractions, respectively. The precipitates To examine the incubation time-dependency of the were washed twice and resuspended in the same buffer. lOO-kDa protein phosphorylation, extracts from myoblasts cultured for 24 and 48 hr were incubated for inPhosphorylation Assay creasing periods. After electrophoresis of the extracts The assays were performed by incubating reaction in the presence of SDS, the bands corresponding to the mixtures in final volumes of 50 ~1 containing 1.5 &i of lOO-kDa protein were cut off from the gels and counted [-r-32P]ATP (6000 Ci/mmole, New England Nuclear), 1 for their radioactivity. The extents of phosphorylation mM ATP, and 30 pg of cell extracts or particulate fracwere found to increase linearly with time until reaching tions obtained from cultured myoblasts. Effects of CaM, maximal levels by 5 min of incubation (data not shown). EGTA, TFP, chlorpromazine, W-7, and/of Walsh inhibiWe then quantified the changes in the phosphorylation tor (Sigma) were examined by treating them with the state of the lOO-kDa protein, which were revealed by the assay mixtures 30 min prior to initiation of the reaction incubation of the extracts obtained from cells cultured by adding ATP. After incubating the mixtures at 20°C for various periods (see Fig. 1A). Figure 1B again shows for proper periods, the reaction was terminated by addithat the extent of the lOO-kDa protein phosphorylation tion of 50 ~1 of 4% sodium dodecyl sulfate (SDS). dramatically rises in the cells cultured for 24-48 hr and The samples were then electrophoresed in 7-14% polythat the extent of the protein phosphorylation at 48 hr is acrylamide gradient gels in the presence of SDS at least sevenfold greater than that at 24 hr. In addition, (Laemmli, 1970). The gels were stained with Coomassie it is noteworthy that the phosphorylation state of the R250, dried under vacuum, and exposed to X-ray films lOO-kDa protein in cells cultured for 60 hr was repeat(Fuji). After the autoradiography, the gel bands that edly observed to be 5-20% lower than that in cells culcorrespond to lOO-kDa size were cut out, incubated in 1 tured for 48 or 72 hr; however, the reasons are presently ml each of H,Oz at 80°C for 24-48 hr, and counted for unknown. Under the same culture conditions, myoblasts their radioactivity using a scintillation counter (Packbegan to engage in active fusion at about 48 hr. These ard). Proteins were assayed as described by Bradford results suggest that the increase in phosphorylation of (1976). the lOO-kDa protein is differentiation stage specific. Separation of Kinase Fraction from Its Substrate in Cell Phosphwrylaticm of the 100-kDa Protein Is Extracts Ca’+/CaM-Dependent Cell extracts were prepared as above, and loaded on a To determine the type of protein kinase(s) that is reDEAE-cellulose column (2 ml bed volume) that had sponsible for phosphorylation of the lOO-kDa protein,
KIM ET AL.
A
24
36
40
60
225
CaM-kinase in Cultured Myoblasts
B
72
kDa 11697-
.
66-
0.0
I 24
I 40 CULTURE
TIME
I 72
0
(hr)
FIG. 1. Changes in the pattern of protein phosphorylation in differentiating myoblasts. Chick embryonic myoblasts were cultured and their extracts were prepared as described under Materials and Methods. Aliquots of the samples (30 pg each) were incubated with [-y-“PjATP at 20°C for 2 min, electrophoresed in the presence of SDS, dried, and autoradiographed (A). The arrowhead and arrows indicate the proteins showing significant changes in their phosphorylation state. After the autoradiography, the bands that correspond to the lOO-kDa protein were cut off from the dried gel and their radioactivity (0) was estimated as described in the text (B). Phosphorylation state of the lOO-kDa protein was maximal in the extract of cells cultured for 72 hr and was expressed as 1.0. Under the same culture conditions, the degree of myoblast fusion (0) was also determined.
extracts of myoblasts cultured for 72 hr were incubated in the presence and absence of Walsh inhibitor, an inhibitor of the CAMP-dependent protein kinase (Cheung et aZ., 1986) or CaM antagonists, such as TFP, chlorpromazine, and W-7 (Levin and Weiss, 1978). As shown in Fig. 2 (lane b), Walsh inhibitor showed little effect on the lOO-kDa protein phosphorylation. Addition of CAMP to the reaction mixture also showed no effect (data not a
b
c
d
e
kDa ZOS-
1169766-
4530-
shown). Thus, it appears unlikely that phosphorylation of the lOO-kDa protein is mediated by the CAMP-dependent system. Treatment of TFP, however, strongly inhibited phosphorylation of the lOO-kDa protein but not of the other proteins (Fig. 2, lane c). Half-maximal inhibition was observed at a concentration of about 25 &f TFP. Other CaM antagonists, such as chlorpromazine and W-7, also similarly inhibited the protein phosphorylation (lanes d and e, respectively). We then tested if treatment of exogenous CaM can reverse the inhibitory effect of TFP. As shown in Fig. 3, addition of CaM to the TFP-treated extract recovers phosphorylation of the lOO-kDa protein but to a level higher than that seen without the addition. Treatment of CaM alone also increased the extent of phosphorylation to a similar level. However, elimination of Ca2+ from the reaction mixture by treating with EGTA blocked phosphorylation of the lOO-kDa protein whether or not CaM was added. These results indicate that Ca2+/CaM-dependent protein kinase(s) is responsible for phosphorylation of the lOO-kDa protein. TFP-mediated Inhibition of Myoblast Fusion and Phosphcnylation of the NO-kDa Protein
FIG. 2. Effects of Walsh inhibitor and CaM antagonists on phosphorylation of the lOO-kDa protein. Extracts were prepared from myoblasts cultured for 72 hr and assayed for protein phosphorylation in the absence (lane a) and presence of 1 rg/ml of Walsh inhibitor(b), 50 &fTFP (c), 200 PMchlorpromazine (d), or 200 pM W-7 (e). The arrowhead indicates the position where the NO-kDa protein migrated.
TFP has been reported to inhibit myoblast fusion without affecting the synthesis of muscle-specific proteins (Bar-Sagi and Prives, 1983). Therefore, we tested the possibility that the TFP-mediated fusion block is associated with the inhibition of the lOO-kDa protein phosphorylation by the drug. Myoblasts were treated with increasing concentrations of TFP at 24 hr of the
226
DEVELOPMIENTAL BIOLOGY
EGTA
-
+
-
-
-
+
TFP
-
-
+
-
+
-
cat-A
-
-
-
-I.
f
+
FIG. 3. Inhibition of the NO-kDa protein phosphorylation by EGTA and TFP and its reversal by CaM. Extracts were prepared from myoblasts cultured for 72 hr and assayed for protein phosphorylation in the absence (-) and presence (+) of 10 mMEGTA, 50 j&fTFP, and/or 10 pg/ml of CaM. The arrowhead indicates the lOO-kDa protein.
culture. After culturing them for the next 48 hr, extracts were prepared and assayed for their capacity to phosphorylate the lOO-kDa protein. Also, the cells cultured under the same conditions were fixed and examined under a microscope to determine the degree of fusion. As shown in Fig. 4A, treatment with higher concentrations of TFP decreased more pronouncedly the phosphorylation state of the lOO-kDa protein but without affecting that of other proteins. In addition, TFP inhibited the myoblast fusion in a dose-dependent manner (Fig. 4B) in accord with the earlier report by Bar-Sagi and Prives (1983). Moreover, the fall in the extent of phosphorylation occurred in parallel with the decrease in the degree of myoblast fusion. Thus, it appears that the TFP-mediated inhibition of the lOO-kDa protein phosphorylation in cultured myoblasts is closely correlated with the block of myoblast fusion induced by the drug. Factors Regulating 100-kDa Protein
the Phosphmylatim
State of the
As was shown in Fig. 3, addition of exogenous CaM to the TFP-treated extract resulted in an increased phos-
VOLUME 150,199Z
phorylation of the lOO-kDa protein to a level higher than that seen without the treatment. This result indicates that CaM is limiting in the cell extracts. This observation also suggests that myoblasts cultured for 24 hr may contain lower amount of CaM than the fused cells, and therefore their extracts might have shown the reduced phosphorylation of the lOO-kDa protein (see Fig. 1). To test this possibility, CaM was added to each extract of the cells cultured for 24 and 72 hr and its effects were compared (Table 1). Addition of CaM significantly increased the abilities of both extracts in phosphorylation of the lOO-kDa protein. However, the effect of CaM was rather proportional and the difference in the extent of phosphorylation in the extracts remained unchanged. Thus, it is unlikely that CaM acts as a limiting factor in regulating the phosphorylation state of the lOO-kDa protein in cultured myoblasts. At least two other factors that can directly affect phosphorylation of the lOO-kDa protein are Ca’+/CaMdependent protein kinase and/or its 100-kDa substrate, whose levels may change in differentiating myoblasts. To test this possibility, we first fractionated the extracts obtained from the cells that had been cultured for 24 and 72 hr using a DEAE-cellulose column. As shown in Table 2, either fraction that eluted from or bound to the column did not reveal any phosphorylation of the lOO-kDa protein. However, the phosphorylation became evident when the two fractions from the extract of 72hr-cultured cell were added together. This result clearly indicates that the ion-exchange chromatography separates the kinase fraction from its lOO-kDa protein substrate. The NEM-sensitive, Ca’+/CaM-dependent protein kinase was then found to be present in the DEAEbound fraction upon further purification of the enzyme using Sephacryl S-200 and CaM-Sepharose affinity columns (unpublished observations).5 Thus, the lOO-kDa protein should be present in the DEAE-flow-through fraction. We then examined by the reconstitution experiments described below if the kinase and its lOO-kDa substrate levels change during myogenesis. When the DEAEflow-through substrate fraction from 24-hr-cultured cell extracts was incubated with the DEAE-bound kinase fraction from ‘72-hr cells, the extent of the protein phosphorylation was significantly increased compared to that from incubation of the kinase and its substrate fractions, both of which were from 24-hr cells (Table 2). However, the extent of phosphorylation did not reach that shown with the kinase and its substrate fractions,
* Jeon, Y. J., Kim, H. S., Chung, C. H., and Ha, D. B. (1991). Unpublished observations.
KIM ET AL. A
a
b
c
CaM-kinme
227
in Cultured Myoblwts
d
0.0
-
5
0 TFP
10
(uM)
FIG. 4. Effect of TFP-treatment to cultured myoblasts on phosphorylation of the lOO-kDa protein. Myoblasts that had been cultured for 24 hr were treated with 0 (lane a), 1 (b), 5 (c), and 10 p&ITFP (d) and further incubated for the next 48 hr. (A) The cells were washed three times with ice-cold PBS to eliminate residual TFP in the culture medium. Extracts were prepared from the cells, and their ability of protein phosphorylation was determined as in Fig. 1. The arrowhead indicates the 100~kDa protein. (B) Radioactivity in the lOO-kDa protein bands (0) was estimated, and that seen in the absence of TFP was expressed as 1.0. Myoblasts cultured under the same conditions were fixed, and their degree of fusion (0) was determined as described under Materials and Methods.
both of which were from 72-hr cells. These results suggest that the level of the lOO-kDa protein in 24-hr cells is lower than that in ‘72-hr cells and increases as myoblast cell differentiation proceeds. On the other hand, incubation of the kinase fraction from 24-hr cells with the substrate fraction from 72-hr cells led to little or no increase in the extent of the NO-kDa protein phosphorylation compared to that from both fractions from 24-hr cells. These results clearly suggest that the activity level of kinase in 24-hr cells is much less than that in 72-hr cells. Thus, it appears likely that the levels of both the lOO-kDa protein and the kinase activity are subjected to change during the early period of myogenic differentiation although the extent of level of change seems to differ.
Properties of Ca’+/CaM-Dependmt Protein Kinase The above studies used only the soluble extract of cultured myoblasts in determining the activity of Ca’+/ CaM-kinase. Therefore, we tested if the kinase activity also exists in particulate fractions. When the particulate fractions from the cells cultured for 72 hr were incubated, the extent of the lOO-kDa protein phosphorylation was much less significant than that of the soluble extracts alone (Fig. 5). Furthermore, little or no change in the extent of the protein phosphorylation was observed when the particulate fractions were incubated
TABLE
TABLE 1 EFFECT
DEAE-fractions
OF THE TREATMENT OF CaM ON PHOSPHORYLA~ON OF THE lOO-kDa PROTEIN
Radioactivity in lOO-kDa protein (cpm) CaM h/ml)
24 hr
72 hr
Ratio 72 hr/24 hr
0 10 20
70 132 145
301 645 698
4.3 4.9 4.8
Note. Extracts were prepared from myoblasts that had been cultured for 24 and 72 hr and assayed for their ability in phosphorylation of the 100~kDa protein in the presence of increasing concentrations of CaM. Radioactivity in lOO-kDa protein bands were then determined as described in Fig. 1.
2
SEPARATION OF Ca*+/CaM-DEPaJDEWr KINASE FROM ITS lOO-kDa PROTEIN SUEWRATE
72 hr 72 hr ‘72 hr 24 hr 24 hr 24 hr 24 hr 24 hr
flow through bound flow through flow through bound flow through flow through bound + 72
(30 pg each)
+ 72 hr bound + 24 hr bound + 72 hr bound hr flow through
ACWITY
Radioactivity in lOO-kDa protein (cpm) 9 4 550 0 5 30 220 44
Note. Extracts prepared from myoblasts cultured for 24 and 72 hr were fractionated using a DEAE-cellulose column (1 X 2.5 cm) as described under Materials and Methods. The flow through and bound fractions were assayed for their ability in phosphorylation of the lOOkDa protein in the presence of CaM (10 pg/ml). Radioactivity in lOOkDa protein bands was then determined as described in Fig. 1.
228
DEVELOPMENTAL BIOLOGY a
b
c
d
e
f
kDa
11697-
VOLUME 150.1992
kinase activity is indeed sensitive to inhibition by the salt, phosphorylation assays were performed in the presence of increasing concentrations of NaCI. As shown in Fig. 7, NaCl inhibited the phosphorylation activity against the lOO-kDa protein in a dose-dependent manner but without affecting that against other proteins. KC1 also blocked specifically the lOO-kDa protein phosphorylation (data not shown). Thus, the Ca2+/CaM-kinase activity found in the present study appears to be sensitive to inhibition by monovalent cations.
45-
DISCUSSION 30-
FIG. 5. Phosphorylation of membranous proteins from cultured myoblasts. Soluble and particulate fractions were obtained from myoblasts cultured for ‘72 hr as described under Materials and Methods. The kinase and its lOO-kDa substrate fractions in the soluble extracts were separated using a DEAE-column as in Table 2. Incubations were performed at 20°C for 2 min with [y-q]ATP and 30 pg each of the following: substrate fraction alone (a), kinase fraction alone (b), substrate and kinase fractions (c), particulate fraction alone (d), substrate and particulate fractions (e), kinase and particulate fractions (f). After the incubation, the samples were electrophoresed in the presence of SDS, dried, and autoradiographed.
with either the DEAE-separated kinase or its substrate fraction from the soluble extract of cells cultured for 72 hr. Similar data were obtained with the particulate fractions from 24- or 48-hr cells. These results indicate that Ca2+/CaM-kinase as well as its lOO-kDa protein substrate are primarily localized to the soluble fraction of the cells. As a preliminary attempt to elucidate the nature of the amino acid(s) that is phosphorylated by Ca2+/CaMkinase, an extract from myoblasts cultured for ‘72 hr was subjected to SDS-polyacrylamide gel electrophoresis in duplicate after the incubation with [T-~P]ATP. One of the gels was dried and autoradiographed. The other gel was also treated as above but after the incubation with 1 N NaOH. Figure 6 shows that the alkali-treatment dephosphorylates nearly all of the phosphoproteins except the lOO-kDa protein. It has been reported that phosphoThr and phospho-Tyr but not phospho-Ser are resistant to dephosphorylation by alkali-treatment (Cheung and Chen, 1981). Thus, it appears likely that Ca2+/CaM-kinase preferentially phosphorylates the Thr and/or Tyr residues in its lOO-kDa protein substrate. During separation of the kinase fraction from its lOOkDa protein substrate using DEAE-cellulose chromatography (see Table 2), the kinase activity eluted with 0.21K NaCl was found to increase several-fold after dialysis against the column buffer without the salt. To test if the
The present studies have demonstrated that phosphorylation of the lOO-kDa protein occurs in soluble extracts of cultured myoblasts and is Ca2+/CaM-dependent. This kinase activity appears to be distinct from that of CaM-kinase I or II by a number of factors: (1) It uses the lOO-kDa protein as its major, if not exclusive, substrate while both CaM-kinase I and II do not seem to phosphorylate proteins with similar size (Ahamd et uZ., 1982; Kennedy and Greengard, 1981; Llinas et al, 1985). (2) It appears to preferentially phosphorylate the Thr and/or Tyr residues while the phosphorylation sites by CaM-kinases I and II include Ser (Huttner et al, 1981; Kennedy and Greengard, 1981). (3) Its size (about llO140 kDa) under nondenaturing conditions (unpublished observations)3 is significantly different from that of CaM-kinase I or II (MeGuinness et al, 1985; Huttner et al, 1981). On the other hand, the Ca2+/CaM-dependent kinase from cultured myoblasts appears to share a number of similar properties with CaM-kinase III: (1) Both a
b
FIG. 6. Effect of alkali-treatment on dephosphorylation of the lOOkDa protein. Protein phosphorylation was carried out using the extracts prepared from myoblasts cultured for 72 hr. After the electrophoresis of the reaction sample, the gels were incubated at 37“C for 1 hr in deionized water (a) or 1 NNaOH (b). The gels were then autoradiographed. The arrowhead indicates the 100~kDa protein.
KIM ET AL. a
b
CaM-kinme in Cultured Myoblasts
c
FIG. 7. Effects of monovalent cations on phosphorylation of the 100kDa protein. Extracts were prepared from myoblasts cultured for 72 hr and assayed for protein phosphorylation in the presence of 0 (lane a), 50 (b), and 100 mM NaCl (c).
use same size, lOO-kDa protein as their substrates. (2) Both are sensitive to inhibition by Na+ ion (Togari and Guroff, 1985). (3) Both of their phosphorylated substrates are resistant to alkali-treatment (Togari and Guroff, 1985). (4) Both have a similar size under nondenaturing condition (Ryazanov et aZ., 1988; Togari and Guroff, 1985; Nairn et al, 1985). However, more studies are necessary for clarification of the identities of both the enzymes and their lOO-kDa substrates. TFP and other CaM antagonists have been reported to block myoblast fusion, and their effects have been attributed to their antagonistic action against CaM, thereby preventing the pivotal role of Ca” in the fusion process (Bar-Sagi and Prives, 1983). In the present study, we showed that TFP and the other CaM antagonists including W-7 and chlorpromazine selectively block the phosphorylation of the lOO-kDa protein (Fig. 2). In addition, extent of the lOO-kDa protein phosphorylation was found to sharply increase in extracts of myoblasts in rapid proliferation and alignment and reach a maximal level in myoblasts of the cells in active fusion (Fig. 1). Furthermore, treatment of the cultured cells with increasing concentrations of TFP resulted in a parallel inhibition of both myoblast fusion and phosphorylation of the lOO-kDa protein, but without affecting the phosphorylation state of other cell proteins (Fig. 4). Thus, it is tempting to speculate that TFP-mediated inhibition of the lOO-kDa protein phosphorylation may be responsible for the block of myoblast fusion, although the role(s) of the lOO-kDa protein in myoblast fusion remains totally unknown. It is noteworthy, however, that, for obtaining halfmaximal inhibition of the lOO-kDa protein phosphory-
229
lation, approximately fivefold higher concentration of TFP (25 PM) was required compared to that (5 pLM) for cultured cells (Fig. 4). The TFP concentration reported to inhibit CaM activity in vitro (ID, = 10 &V; Weiss and Levin, 1978) is also at least twofold greater than that required for the half-maximal inhibition of myoblast fusion (Fig. 4; Bar-Sagi and Prives, 1983). A possible explanation for these phenomena is that TFP may accumulate in the cells when added to culture medium. Although the TFP-induced fusion block can be reversed by drug removal, the recovery period appears to be much longer than that revealed by treatment-and-removal of other fusion blockers, such as EGTA (Bar-Sagi and Prives, 1983; Paterson and Strohman, 1972). Moreover, TFP is known to bind tightly to purified Ca2+-activated form of CaM with a binding constant of 1 &f(Levin and Weiss, 1977). Therefore, it seems that the TFP remained tightly bound to CaM was able to prevent phosphorylation of the lOO-kDa protein even after the removal of the drug from culture medium. It has been reported that Ca2+ influx in cultured myoblasts precedes membrane fusion (David et cd, 1981). Therefore, it is possible that the increase in intracellular Ca2+ concentration may induce activation of the Ca2+/CaM-dependent protein kinase(s) and consequently increase the extent of the 100~kDa protein phosphorylation. However, it is unlikely that the increased phosphorylation is mediated by alterations in the cells’ content of CaM, that may occur during myogenesis, since exogenous addition of CaM to extracts of proliferating myoblasts could not elevate the phosphorylation to the level of differentiated cells (Table 1). On the other hand, using the DEAE-separated kinase and its substrate preparations, we were able to demonstrate that the levels of both the kinase activity and its substrate increase during the early period of myogenic differentiation. Therefore, it appears that increase in the phosphorylation state of the lOO-kDa protein is at least in part due to differentiation-dependent changes in the levels of Ca2+/CaM-dependent protein kinase and its lOOkDa protein substrate. However, because these level changes were determined by the phosphorylation assay, we cannot exclude the possibility that other factors, such as endogenous activator or inhibitor of the kinase, may influence the phosphorylation state of the lOO-kDa protein in a differentiation-stage-specific manner. Thus, purification of the kinase and its substrate and preparation of antibodies against each protein are necessary to clarify the differentiation-specific regulation of the phosphorylation state of the,lOO-kDa protein in embryonic myoblasts in culture, and are in progress. We are grateful to Dr. Y. S. Kang (Kyung-Sang University) for his helpful discussion and to Ms. Y. J. Jeon for her help in some experi-
230
DEVELOPMENTALBIOLOGY
ments. This work was supported by grants from Korea Science and Engineering Foundation and the Ministry of Education. REFERENCES Ahamd, Z., Depaoli-Roach, A. A., and Roach, P. J. (1982). Purification and characterization of a rabbit liver calmodulin-dependent protein kinase able to phosphorylate glycogen syntbase. J. Bid &em, 257, 8348-8355. Bar-Sagi, D., and Prives, J. (1983). Trifluoperazine, a calmodulin antagonist, inhibits muscle cell fusion. J. Cell BioL 97,1375-1380. Bischoff, R., and Holtzer, H. (1969). Mitosis and process of differentiation of myogenic cells in vitro. J. Cell Biol 41,188-200. Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. And Biochem. 72,248-254. Cheung, H.-C., Kemp, B. E., Persons, R. B., Smith, A. J., Misconi, L., Pattern, S. M. V., and Walsh, D. A. (1986). A potent synthetic peptide inhibitor of the CAMP-dependent protein kinase. J. BioL Chew. 261,989-992. Cheung, W. Y. (1980). Calmodulin plays a pivotal role in cellular regulation. Science 207,19-27. Cheung, Y. E., and Chen, L. B. (1981). Detection of phosphotyrosinecontaining 34,000-dalton protein in the framework of cells transformed with Rous sarcoma virus. Proc NatL Acad. Sci. USA 78, 2388-2392. Dabrowska, R., Sherry, J. M. F., Aromatorio, D. K., and Hartshorne, D. J. (1978). Modulator protein as a component of the myosin light chain kinase from chicken gizzard. Biochemistry 17.253-258. David, J. D., See, W. M., and Higginbotham, C. A. (1981). Fusion of chick embryo skeletal myoblasts: Role of calcium influx preceding membrane union. Dev. Biol. 82,297-307. Endo, T., and Nadal-Ginard, B. (1987). Three types of muscle-specific gene expression in fusion-blocked rat skeletal muscle cells: Translational control in EGTA-treated cells. Cell 49,515-526. Ha, D. B., Boland, R., and Martonosi, A. (1979). Synthesis of the calcium transport ATPase of sarcoplasmic reticulum and other muscle cells in viwo and in vitro. Biochim Biophys. Ada 585,165-187. Huttner, W. B., DeGennaro, L. J., and Greengard, P. (1981). Differential phosphorylation of multiple sites in purified protein I by cyclic AMP-dependent and calcium-dependent protein kinases. J. BioL Chem. 256,1482-1488. Kennedy, M. B., and Greengard, P. (1981). Two calcium/calmodulindependent protein kinases, which are highly concentrated in brain, phosphorylate protein I at distinct site. Proc. NatL Acad Sci USA 78,1293-1297. Kwak, K. B., Lee, Y. S., Suh, S. W., Chung, C. S., Ha, D. B., and Chung, C. H. (1989). Purothionin from wheat endosperm reversibly blocks myogenic differentiation of chick embryonic muscle cells in culture. Eqn. Cell Res. 183,501-507. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227,680-685. Levin, R. M., and Weiss, B. (1977). Binding of trifluoperazine to the calcium-dependent activator of cyclic nucleotide phosphodiesterase. Mol PhammoL 13,690-697. Levin, R. M., and Weiss, B. (1978). Specificity of binding of trifluoper-
VOLUME150,1992 azine to the calcium-dependent activator of phosphodiesterase and to a series of other calcium-binding proteins. B&him Biqphys Acta 540,197-204. Llinas, R., McGuinness, T. L., Leonard, C. S., Sugimori, M., and Greengard, P. (1985). Intraterminal injection of synapsin I or calcium/ calmodulin-dependent protein kinase II alters neurotransmitter release at the squid giant synapse. Proc. NatL Acad Sci. USA 82,30353039. McGuinness, T. L., Lai, Y., and Greengard, P. (1985). Ca*+/calmodulindependent protein kinase II. J. Bid, Chew. 260,1696-1704. Nadal-Ginard, B. (1978). Commitment, fusion and biochemical differentiation of a myogenic cell line in the absence of DNA synthesis. Cell 15,855-864. Nairn, A. C., Bhagat, B., and Palfrey, C. (1985). Identification of calmodulin-dependent protein kinase III and its major Mr 100,000 substrate in mammalian tissues. Proc. NatL Acm! Sci USA 82,79397943. Nairn, A. C., and Palfrey, H. C. (1987). Identification of the major Mr 100,000 substrate for calmodulin-dependent protein kinase III in mammalian cells as elongation factor-2. J. BioL Chem 262, 1729917303. O’Neill, M. C., and Stockdale, F. E. (1972). A kinetic analysis of myogenesis in vitro. J. Cell BioL 52,52-65. Palfrey, H. C. (1983). Presence in many mammalian tissues of an identical major cytosolic substrate (Mr 100,000) for calmodulin-dependent protein kinase. FEBS I&. 157,183-190. Paterson, B., and Strohman, R. C. (1972). Myosin synthesis in cultures of differentiating chicken embryo skeletal muscle. Deu. BioL 29, 113-138. Ryazanov, A. G. (1987). Ca*+/calmodulin-dependent phosphorylation of elongation factor 2. FEBS L&t. 214,331-334. Ryazanov, A. G., Shestakova, E. A., and Natapov, P. G. (1988). Phosphorylation of elongation factor 2 by EF-2 kinase affects rate of translation. Nature 334,170-173. Schulman, H., and Greengard, P. (1978). Ca2+-dependent protein phosphorylation system in membranes from various tissues, and its activation by “calcium-dependent regulator”. Proc NatL Acad Ski. USA 75,5432-5436. Shainberg, A., Yagil, G., and Yaffe, D. (1969). Control of myogenesis in vitro by Ca2+ concentration in nutritional medium. Exp. CeU Res. 58, 163-16’7. Togari, A., and Guroff, G. (1985). Partial purification and characterization of a nerve growth factor-sensitive kinase and its substrate from PC12 cells. J. Bid Chem. 260,3804-3811. Wakelam, M. J., and Pette, D. (1982). The breakdown of phosphatidylinositol in myoblasts stimulated to fuse by the addition of Ca’+. B&hem. J. 202,723-729. Weiss, B., and Levin, R. M. (1978). Mechanism for selectivity inhibiting the activation of cyclic nucleotide phosphodiesterase and adenylate cyclase by antipsychotic agents. A&v. Cyclic Nwleotide Res. 9, 285-304. Woodgett, J. R., Tonks, N. K., and Cohen, P. (1982). Identification of a calmodulin-dependent glycogen synthase kinase in rabbit skeletal muscle, distinct from phosphorylase kinase. FEBS II&t. 148,5-11. Yagi, K., Yazawa, M., Kakiuchi, S., Ohshima, M., and Uenishi, K. (1978). Identification of an activator protein for myosin light chain kinase as the Ca2+-dependent modulator protein. J. Bid Chem 253, 1338-1340.
