Calcium-dependent Regulation of Cytochrome c Gene Expression in Skeletal Muscle Cells IDENTIFICATION OF A PROTEIN KINASE C-DEPENDENT PATHWAY*

Mitochondrial biogenesis can occur rapidly in mammalian skeletal muscle subjected to a variety of physiological conditions. However, the intracellular signal(s) involved in regulating this process remain unknown. Using nuclearly encoded cytochrome c, we show that its expression in muscle cells is increased by changes in cytosolic Ca using the ionophore A23187. Treatment of myotubes with A23187 increased cytochrome c mRNA expression up to 1.7-fold. Transfection experiments using promoter-chloramphenicol acetyltransferase constructs revealed that this increase could be transcriptionally mediated since A23187 increased chloramphenicol acetyltransferase activity by 2.5-fold. This increase wasnotchangedbyKN62,aninhibitorofCa/calmodulindependent kinases II and IV, and it was not modified by overexpression of protein kinase A and cAMP response element-binding protein, demonstrating that the A23187 effect was not mediated through Ca/calmodulin-dependent kinaseor protein kinase A-dependent pathways. However, treatment of myotubes with staurosporine or 12-O-tetradecanoylphorbol-13-acetate reduced the effect of A23187 on cytochrome c transactivation by 40–50%. Coexpression of the Ca-sensitive protein kinase C isoforms a and bII, but not the Ca -insensitive d isoform, exaggerated the A23187-mediated response. The short-term effect of A23187 was mediated in part by mitogen-activated protein kinase (extracellular signalregulated kinases 1 and 2) since its activation peaked 2 h after A23187 treatment, and cytochrome c transactivation was reduced by PD98089, a mitogen-activated protein kinase/extracellular signal-regulated kinase kinase inhibitor. These results demonstrate the existence of a Ca-sensitive, protein kinase C-dependent pathway involved in cytochrome c expression and implicate Ca as a signal in the up-regulation of nuclear genes encoding mitochondrial proteins.

Mitochondrial biogenesis can occur rapidly in mammalian skeletal muscle subjected to a variety of physiological conditions. However, the intracellular signal(s) involved in regulating this process remain unknown. Using nuclearly encoded cytochrome c, we show that its expression in muscle cells is increased by changes in cytosolic Ca 2؉ using the ionophore A23187. Treatment of myotubes with A23187 increased cytochrome c mRNA expression up to 1.7-fold. Transfection experiments using promoter-chloramphenicol acetyltransferase constructs revealed that this increase could be transcriptionally mediated since A23187 increased chloramphenicol acetyltransferase activity by 2.5-fold. This increase wasnotchangedbyKN62,aninhibitorofCa 2؉ /calmodulindependent kinases II and IV, and it was not modified by overexpression of protein kinase A and cAMP response element-binding protein, demonstrating that the A23187 effect was not mediated through Ca 2؉ /calmodulin-dependent kinase-or protein kinase A-dependent pathways. However, treatment of myotubes with staurosporine or 12-O-tetradecanoylphorbol-13-acetate reduced the effect of A23187 on cytochrome c transactivation by 40 -50%. Coexpression of the Ca 2؉ -sensitive protein kinase C isoforms ␣ and ␤ II , but not the Ca 2؉ -insensitive ␦ isoform, exaggerated the A23187-mediated response. The short-term effect of A23187 was mediated in part by mitogen-activated protein kinase (extracellular signalregulated kinases 1 and 2) since its activation peaked 2 h after A23187 treatment, and cytochrome c transactivation was reduced by PD98089, a mitogen-activated protein kinase/extracellular signal-regulated kinase kinase inhibitor. These results demonstrate the existence of a Ca 2؉ -sensitive, protein kinase C-dependent pathway involved in cytochrome c expression and implicate Ca 2؉ as a signal in the up-regulation of nuclear genes encoding mitochondrial proteins.
Ionized calcium (Ca 2ϩ ) acts as a second messenger in a variety of biological processes, including gene expression (1,2), cell cycle regulation, (3) and cell death (4). Since Ca 2ϩ cannot be metabolized like other second messengers, cells tightly reg-ulate intracellular Ca 2ϩ concentration through a set of specialized molecules by sequestering Ca 2ϩ into intracellular compartments. In skeletal muscle, depolarization causes the release of Ca 2ϩ from the sarcoplasmic reticulum and a subsequent increase in cytosolic Ca 2ϩ concentration. This rise triggers muscle contraction (5) and promotes mitochondrial metabolism (6,7). However, it is not known to what extent Ca 2ϩ is involved in the phenotypic adaptations observed in skeletal muscle in response to an altered functional demand, such as chronic contractile activity.
Mitochondrial biogenesis is one of the striking responses observed in skeletal muscle cells in response to a variety of physiological conditions (see Refs. 8 -10 for reviews). The synthesis of mitochondrial proteins involves the expression of genes originating from two distinct genetic compartments. Thirteen proteins are derived from the mitochondrial genome, whereas the remaining gene products are transcribed from the nuclear genome, translated, and subsequently imported into the mitochondria. In skeletal muscle, chronic contractile activity is known to markedly increase the mRNA levels of several nuclear and mitochondrial genes encoding mitochondrial proteins (11)(12)(13)(14). Since contractile activity necessarily involves the mobilization of Ca 2ϩ from intracellular stores (5) and since Ca 2ϩ is an important signaling ion in muscle and other tissues (15), we hypothesized that Ca 2ϩ could play a role as an intracellular signal leading to mitochondrial biogenesis in skeletal muscle. Several arguments support this possibility. First, changes in internal Ca 2ϩ concentration in skeletal muscle cells using the Ca 2ϩ ionophore A23187 are known to induce alterations in gene expression (16,17). Second, a Ca 2ϩ -regulated, calcineurin-dependent pathway has been implicated in determining skeletal muscle phenotype (18). Third, the treatment of myotubes in culture with the Ca 2ϩ ionophore A23187 has been shown to increase the mitochondrial enzyme activities of citrate synthase, malate dehydrogenase, and fumarate hydratase (16,19). Fourth, myotubes cultured in low Ca 2ϩ medium possess very low levels of mitochondrial citrate synthase (20). Fifth, DNA sequencing and footprinting analyses within the rat cytochrome c promoter have demonstrated the existence of multiple binding sites, including two cAMP response elements (CRE) 1 (21,22). Phosphorylation of the cAMP response element-binding protein (CREB) on Ser 133 by calcium/calmodulindependent protein kinase (CaM kinase) IV and its binding to the CRE are known to enhance the transcriptional activation of specific genes (23,24). Finally, electrical stimulation of skeletal muscle cells has been shown to induce the translocation of the Ca 2ϩ -dependent protein kinase C (PKC) from the cytosolic to the plasma membrane fraction (25) and to cause an increase in the activity of PKC in the nucleus (26).
Thus, to define more precisely the potential role of Ca 2ϩ in skeletal muscle mitochondrial biogenesis, we began with a systematic study of the effect of the Ca 2ϩ ionophore A23187 on cytochrome c gene expression in skeletal muscle. We report here, using the L6E9 muscle cell line, that the treatment of myotubes with A23187 increases cytochrome c gene expression through a PKC-dependent pathway that depends, in part, on the activation of MAPK.

EXPERIMENTAL PROCEDURES
Materials-A23187, KN62, TPA, and staurosporine were purchased from Sigma. They were prepared as stock solutions in Me 2 SO at concentrations of 0.25 mM, 2 mM, 0.25 mM, and 1 M, respectively. PD98059 was obtained from Calbiochem and prepared as a 25 mM solution in Me 2 SO. [ 14 C]Chloramphenicol, [␣-32 P]dCTP, nylon membrane (Hybond N), and the ECL immunoblot detection reagents were obtained from Amersham Pharmacia Biotech (Mississauga, Ontario, Canada). Thinlayer chromatography plates were purchased from Fisher Scientific (Unionville, Ontario). The CREB antibody was purchased from Santa Cruz Biotechnology, whereas the MAPK antibody was from New England Biolabs Inc. (Beverly, MA). Other cell culture and molecular biology reagents were purchased from Sigma or Life Technologies, Inc. (Burlington, Ontario) and were of the highest grade available.
Dose-response and Time Course Experiments-All experiments were done with differentiated cells (myotubes) and were matched with vehicle-treated controls. The dose-dependent effects of A23187 on cytochrome c mRNA were evaluated at final concentrations ranging between 0.25 and 1 M. Total RNA was extracted 72 h later as described below. In time course experiments, myotubes were incubated with 1 M A23187 for 24, 48, or 72 h prior to the RNA extraction.
Total RNA Isolation, Northern Blotting, and Radiolabeled Probes-To isolate total RNA, cells were harvested and centrifuged at 4°C for 5 min. The pellet was resuspended with 1 ml of lysis solution consisting of 1 volume of solution containing 4 M guanidinium thiocyanate, 25 mM sodium citrate, pH 7.0, 0.5% Sarkosyl, and 0.1 M ␤-mercaptoethanol; 1 volume of phenol; and 0.1 volume of 2 M sodium acetate, pH 4.0. The cell lysate was then mixed with 0.2 ml of chloroform/isoamyl alcohol (24:1, v/v) and centrifuged at 4°C for 15 min. The upper phase was removed and added to 0.5 ml of isopropyl alcohol to precipitate the RNA. Total RNA was centrifuged for 10 min at 4°C, washed with 1 ml of 75% ethanol, dried in a vacuum desiccator, and finally resuspended in 30 -50 l of diethyl pyrocarbonate-treated water. Total RNA (10 g) was separated on a denaturing formaldehyde-containing 1% agarose gel and transferred overnight to a nylon membrane. The quality of the total RNA isolated in this fashion was high as indicated by 1) an ϳ2-fold greater intensity of the 28 S rRNA band, compared with 18 S rRNA, on ethidium bromide-stained gels; 2) high A 260 /A 280 ratios; and 3) a lack of visible RNA degradation or DNA contamination. Membranes were hybridized with random primer-labeled cDNA probes encoding cytochrome c and 18 S rRNA, as done previously (27,28). Final stringent washes of the blots were done in 0.1ϫ SSC and 0.1% SDS for 15 min at 55°C followed by 15 min at 60°C. Signals were quantified by electronic autoradiography (Packard Instrument Co.), and specific cytochrome c mRNA levels were corrected for differences in loading using the 18 S rRNA signal.
Plasmids-Plasmid constructs used for cytochrome c mRNA expression (pRC4) and promoter analysis (pRC4CAT) were generously provided by Dr. Richard Scarpulla (Department of Cell and Molecular Biology, Northwestern University, Chicago, IL) (20,28). The pRC4CAT/ Ϫ726 construct contains sequences of the rat cytochrome c gene extending from position Ϫ726 upstream of the start site of transcription fused to the chloramphenicol acetyltransferase (CAT) reporter gene. Truncated cytochrome c promoter constructs (pRC4CAT/Ϫ326 and pRC4CAT/Ϫ66) were also used. pRSV/␤-gal was used to direct the synthesis of ␤-galac- respectively. Expression vectors encoding PKC isoforms ␣, ␤ II , and ␦ were originally described by Ono et al. (32) and were obtained from Dr. J. McDermott (Department of Kinesiology, York University, Toronto, Canada).
DNA Transfections and Expression Assays-L6E9 myoblasts at 60% confluence were transfected with the appropriate cytochrome c promoter-reporter construct (10 g/100-mm dish). pRSV/␤-gal (2 g/dish) was used as an internal control to assess and correct for variations in transfection efficiency. However, no differences in ␤-galactosidase activity were found in A23187-and vehicle-treated cells. Where indicated, 5 g of pRSV/CREB, pC␣EV (PKA), PKC␣, PKC␤ II , or PKC␦ were cotransfected. Total DNA concentration was maintained constant in transfection experiments by adding pGEM-4Blue (Promega) or the specific empty vector as a carrier plasmid. Transfections were done using the poly-L-ornithine method followed by a Me 2 SO shock (33,34). Cells were then switched to low serum medium to allow differentiation. Where indicated, myotubes were preincubated for 24 h with KN62, staurosporine, or TPA prior to A23187 treatment. Some cells were treated with PD98059 (12.5 M) for 0.5 h or with EGTA (1 mM) for 5 h prior to A23187 treatment.
To prepare cell extracts for CAT and ␤-galactosidase expression assays, myotubes were scraped and centrifuged at 4°C for 5 min. The cell pellet was resuspended in 100 l of 0.25 M Tris, pH 7.9, and then subjected to three freeze-thaw cycles in an ethanol/dry ice slurry. Cell debris was centrifuged at 4°C for 5 min. The resulting supernatant was removed and used directly in the assays. CAT activity was determined using 10 -20 l of cell extract in the presence of 1 mM acetyl-CoA and 0.23 Ci of [ 14 C]chloramphenicol in a total volume of 40 l. The mixture was incubated at 37°C for 1.5 h, extracted with ethyl acetate, and dried in a vacuum desiccator. Acetylated and non-acetylated forms of [ 14 C]chloramphenicol were resuspended in 30 l of ethyl acetate and separated by thin-layer chromatography for 30 min at room temperature with chloroform/methanol (95:5, v/v) as the mobile phase. The percent conversion of [ 14 C]chloramphenicol to its acetylated products was quantified by electronic autoradiography. Results were corrected for transfection efficiency using ␤-galactosidase activity of the cotransfected plasmid. ␤-Galactosidase activity was measured using 10 -30 l of cell extract in a reaction mixture consisting of 60 mM Na 2 HPO 4 , 40 mM NaH 2 PO 4 , 1 mM MgCl 2 , 50 mM ␤-mercaptoethanol, and 0.66 mM o-nitrophenyl-␤-D-galactopyranoside. After incubation for 2.5 h at 37°C, the reaction was stopped by the addition of 1 M NaCO 3 , and the reaction product absorbance was read at 420 nm.
Immunoblotting-To prepare whole cell extracts for MAPK immunoblotting, myotubes were rinsed with cold phosphate-buffered saline and scraped in 200 l of modified Laemmli buffer (20% glycerol, 2% SDS, and 5% ␤-mercaptoethanol in 62.5 mM Tris-HCl, pH 6.8) containing protease inhibitors. Cells were sonicated (3 ϫ 10-s pulses) on ice, heat-denatured at 95°C for 5 min, and spun briefly to pellet insoluble material. Extracts (100 g) were electrophoresed on SDS-10% polyacrylamide gels and transferred to nylon membranes by electrotransfer (35). The extracts used to detect CREB were the same as those used in the CAT assay, with the addition of 1 volume of lysis buffer (10% glycerol, 2.3% SDS, 5% ␤-mercaptoethanol, and 62.5 mM Tris-HCl, pH 6.8). The extracts were heat-denatured for 5 min, and 30 g of protein/ lane were subjected to electrophoresis on SDS-10% polyacrylamide gels followed by transfer to nylon membranes. Primary antibody incubations using dilutions of 1:1000 (CREB) and 1:250 (MAPK) were carried out at 4°C overnight. Signals were detected by the use of an anti-rabbit antibody coupled to horseradish peroxidase and enhanced chemiluminescence according to the manufacturer's instructions (Amersham Pharmacia Biotech).
Statistics-The data are presented as means Ϯ S.E. One-way analyses of variance were used to evaluate the effect of A23187 incubation time and concentration on cytochrome c mRNA levels. Individual mean differences were assessed with Scheffé's post-hoc test. Two-way analyses of variance were employed to examine the effects of A23187 incubation time and cotransfection of plasmids on cytochrome c transactivation. Unpaired t tests were used for evaluating the effects of PD98059, staurosporine, and TPA on CAT activity compared with A23187 alone and to evaluate the effect of PKC isoforms on CAT activity. An ␣ level of 0.05 was used to indicate statistical significance.

Calcium-dependent Regulation of Cytochrome c Gene
Expression-Northern blot analyses were first used to determine the effect of A23187 treatment on cytochrome c mRNA levels. Concentrations of A23187 ranging from 0.25 to 1.0 M elicited a progressive increase in cytochrome c mRNA levels by up to 1.7-fold (p Ͻ 0.05) (Fig. 1A). Higher concentrations did not further enhance the response (data not shown). The effect of A23187 on cytochrome c mRNA levels was also time-dependent (Fig. 1B). By 48 and 72 h of exposure to the ionophore, mRNA levels were increased by 1.6-and 1.7-fold (p Ͻ 0.05), respectively. This effect was abolished when myotubes were preincubated with the extracellular Ca 2ϩ -chelating agent EGTA (Fig.  1A). Taken together, these data show a calcium-dependent increase in cytochrome c mRNA levels and indicate that A23187 exerts its effect by increasing the intracellular calcium concentration from the extracellular Ca 2ϩ pool.
Preliminary work suggested that the A23187 effect on cytochrome c mRNA was not mediated by a change in mRNA stability. 2  levels through transcriptional activation. In addition, as previously shown with cytochrome c mRNA, preincubation of the myotubes with 1 mM EGTA abolished the A23187-dependent increase in transcriptional activation, indicating a requirement for extracellular Ca 2ϩ (Fig. 2B).
Ca 2ϩ -mediated Cytochrome c Transcriptional Activation Is Not Mediated via the CRE-Next, we investigated the potential role of cis-elements present in the cytochrome c promoter in response to A23187, particularly the role of the two Ca 2ϩ /cAMP response elements. One CRE (CRE-1) is located Ϫ281 base pairs upstream of the transcription start site, whereas another CRE (CRE-2) is found at position Ϫ119 (22). Analysis of variance indicated that significant main effects of A23187 and promoter construct existed, but that the effect of A23187 depended on the construct used. A23187 increased cytochrome c transactivation mediated by the Ϫ726 promoter construct by 2.4-fold (p Ͻ 0.05). Elimination of 400 base pairs downstream of position Ϫ726 using pRC4CAT/Ϫ326 did not modify the magnitude of the transcriptional activation induced by A23187 (Fig. 2B), as expected on the basis of the locations of the CREs. Surprisingly, myotubes transfected with pRC4CAT/Ϫ66 showed a greater relative response (3.9-fold) to A23187 treatment compared with cells transfected with pRC4CAT/Ϫ726 or pRC4CAT/Ϫ326, despite drastic reductions in absolute CAT activity. These experiments verify that the DNA sequence encompassed between positions Ϫ326 and Ϫ66 is necessary for a high constitutive level of cytochrome c gene expression in L6E9 cells and that A23187-induced transcriptional activation does not seem to involve the activation of the CRE sites.
To specifically investigate the role of CREB in cytochrome c gene expression, cotransfection experiments were carried out in the presence and absence of CREB and PKA expression plasmids. A23187 alone produced a 3-fold increase (p Ͻ 0.05) in CAT activity. Transfection of myotubes with pRSV/CREB elicited 7.2 Ϯ 1.3-fold (n ϭ 5) increases in CREB protein levels (Fig. 3A, inset). Cotransfection with CREB or PKA alone or in combination did not further enhance the typical response elicited by A23187 (Fig. 3A). These results show that cytochrome c gene expression in skeletal muscle cells does not appear to involve a PKA-mediated phosphorylation of CREB.
Ca 2ϩ -mediated Cytochrome c Transcriptional Activation Is Not Mediated via a CaM Kinase-dependent Pathway-To examine the possibility that the Ca 2ϩ -mediated effect on cytochrome c gene expression occurs through a CaM kinase-dependent pathway, KN62, a potent inhibitor of both CaM kinases II and IV, was used. Inhibition of CaM kinases was ineffective in abolishing the A23187-induced increase (p Ͻ 0.05) in cytochrome c transcriptional activation, suggesting that neither CaM kinase II nor CaM kinase IV mediates the intracellular pathway triggered by A23187 (Fig. 3B).
Ca 2ϩ -mediated Cytochrome c Transcriptional Activation Is Mediated via a PKC-dependent Pathway-Next, we examined the possibility that the induction of cytochrome c gene expression by A23187 was mediated through a PKC-dependent pathway. In these experiments, A23187 induced a 2.6-fold increase in cytochrome c transactivation. In the presence of staurospo- rine, a nonspecific inhibitor of PKC, cytochrome c transactivation by A23187 was still evident (p Ͻ 0.05), but was markedly reduced to a 1.4-fold effect, significantly less than that observed in the absence of staurosporine (Fig. 3B). To more specifically inhibit PKC, myotubes were preincubated for 24 h with the phorbol ester TPA. Long-term treatment with TPA has been shown to down-regulate PKC expression in skeletal muscle (26) and in L6 cells (36), thus providing a useful tool to test the involvement of PKC in the signaling pathway. TPA treatment reduced the effect of A23187 on cytochrome c transactivation from 2.6-to 1.6-fold (40%) (Fig. 3C), indicating that the effect of A23187 is partly mediated by a PKC-dependent pathway. Therefore, we also evaluated the role of specific PKC isoforms in determining this effect. Cotransfection experiments were performed using pRC4CAT/Ϫ326 along with either the Ca 2ϩ -sensitive PKC␣ or PKC␤ II isoform or the Ca 2ϩ -insensitive PKC␦ isoform. Transcriptional activation by A23187 was significantly (p Ͻ 0.05) enhanced in the presence of the ␣ and ␤ II isoforms compared with the empty vector pTB, but not in the presence of the Ca 2ϩ -insensitive PKC␦ isoform (Fig. 4A). These experiments confirmed the involvement of Ca 2ϩ -sensitive PKC isoforms in the response triggered by A23187.
Ca 2ϩ -mediated Cytochrome c Transcriptional Activation Involves the Activation of MEK and MAPK-Since various lines of evidence (37) indicate that PKC isoforms activate the ERK1 and ERK2 MAPKs, we examined the possible activation of a MAPK cascade in response to A23187 treatment. Myotubes that were preincubated for 30 min with the MEK inhibitor PD98059 (38) demonstrated a 38% decrease (p Ͻ 0.05) in Ca 2ϩinduced cytochrome c gene transactivation (Fig. 4B). Immunoblot analyses using a phospho-specific MAPK antibody showed that the effect of A23187 was a relatively early event since a transient 4.5 Ϯ 0.7-fold (n ϭ 4) activation of ERK1 (p42 MAPK) and ERK2 (p44 MAPK) occurred with a peak at 2 h (Fig. 4C). This activation was completely prevented by preincubation with the MEK inhibitor PD98059. The stimulatory effect of A23187 was still present at 24 h, but was undetectable at 48 h after treatment with A23187 (data not shown). This transient pattern observed could also not be explained by differences in gel loading since both Ponceau staining and immunostaining of the same blot with a p44 antibody to illustrate total p44 levels revealed equal loading in all lanes (data not shown).

DISCUSSION
It is well known that oscillations in cytosolic Ca 2ϩ concentration play a fundamental role in the contraction and relaxation of striated muscles. It is also established, mainly from work in other tissue types, that Ca 2ϩ is active in signaling alterations in gene expression (1,2,15). Recent investigations have documented that this occurs in skeletal myofibers (18) and that artificially induced increases in Ca 2ϩ concentration lead to muscle phenotypic adaptations (16). These investigations are physiologically relevant since sustained elevations in internal Ca 2ϩ occur during contractile activity (39), a situation in which marked alterations in muscle phenotype have been documented (40). In view of earlier work that demonstrated that mitochondrial enzyme activities could be modified by the presence or absence of Ca 2ϩ (19,20), we hypothesized that modifications in Ca 2ϩ concentration could represent a putative signal leading to the enhanced expression of nuclear genes encoding mitochondrial proteins, thereby enhancing organelle biogenesis. To begin addressing this question, we have specifically studied the behavior of cytochrome c, a nuclearly encoded mitochondrial protein of the respiratory chain, in response to a model of augmented cellular Ca 2ϩ concentration using the Ca 2ϩ ionophore A23187. Increases in intracellular Ca 2ϩ concentration of 3-10-fold are typically obtained 10 min after the addition of the ionophore, apparently lasting up to 16 days following addition of the drug (16). This treatment has previously been used to document changes in acetylcholine receptor (17), acetylcholinesterase (41), and myosin isoform (16) gene expression. Here, we show for the first time that an increase in intracellular Ca 2ϩ concentration brought about by A23187 can increase the transcriptional activation of a nuclear gene encoding a mitochondrial protein (i.e. cytochrome c). This effect may be applicable to other nuclearly encoded genes since we have observed a similar finding using the cytochrome c oxidase subunit Vb promoter. 3 We were surprised to find that the augmentation of cytochrome c transcriptional activation was independent of the involvement of the two CRE sites present in the cytochrome c promoter as well as the activation of the CaM kinase-depend-3 P. Kumar and D. A. Hood, unpublished observations.

FIG. 4. Role of PKC isoforms and MEK inhibition in cytochrome c transcriptional activation and effect of A23187 on ERK1 and ERK2 activities.
A, effect of PKC isoform overexpression on cytochrome c transcriptional activation. L6E9 myoblasts were cotransfected with 10 g of pRC4CAT/Ϫ326 and 5 g of PKC expression vectors or empty vector (pTB). Following transfection, the myoblasts were induced to differentiate by switching to low serum conditions until the myotubes reached 85-90% confluence. The myotubes were then treated with A23187 (0.75 M) or vehicle (dimethyl sulfoxide (DMSO)) for 48 h. CAT assays were normalized with ␤-galactosidase activity (*, p Ͻ 0.05 versus the empty vector pTB). B, role of MEK in A23187mediated cytochrome c transactivation. L6E9 myoblasts were transfected with 10 g of pRC4CAT/Ϫ326. Following differentiation, the myotubes were treated with PD98059 (PD; 12.5 M) or vehicle 30 min prior to the addition of A23187 (A; 0.75 M) or vehicle (dimethyl sulfoxide (D)) for 48 h. CAT assays were normalized with ␤-galactosidase activity (*, p Ͻ 0.05 versus without PD98059). C, time course of ERK1 (p42) and ERK2 (p44) MAPK activation in response to A23187. L6E9 myotubes were treated with A23187 (A; 0.75 M) or vehicle (dimethyl sulfoxide (D)) for the indicated time points or were not treated (NT). Some cells were also treated with PD98059 (PD; 12.5 M) for 30 min prior to the addition of the ionophore. Whole cell lysates were prepared for immunoblot analyses as described under "Experimental Procedures" using a phospho-specific ERK antibody. ent pathway since a CREB/CRE pathway of cytochrome c transcriptional activation was previously found in Balb/c/3T3 cells (22). However, our observations are consistent with those of others using cardiac myocytes, in which the CRE appears to interact with c-Jun rather than CREB (42). The difference between these results and those previously reported in Balb/c/ 3T3 cells (22) suggests that cytochrome c gene expression is regulated via different transduction pathways in a variety of cell types. This appears to be true even within striated muscles since we have previously shown differences in the regulation of cytochrome c gene expression between heart and skeletal muscle during development and in response to variations in thyroid status (43).
In agreement with previous reports using other cell types (21,29), we show that the DNA sequence of the cytochrome c promoter encompassed between positions Ϫ326 and Ϫ66 is necessary for a high constitutive level of cytochrome c gene expression in L6E9 muscle cells. Despite this, the minimal cytochrome c promoter (pRC4CAT/Ϫ66) still possessed a high degree of Ca 2ϩ responsiveness. The Ca 2ϩ -responsive element downstream of position Ϫ66 is currently under investigation. Although there are no apparent AP-1 or CRE sites in this region, the response may be related to the presence of CCAAT box-like sequences at positions ϩ23 and ϩ48 of the first intron. This sequence has previously been shown to confer A23187mediated transcriptional activation of the grp78 promoter (44). It does not appear that Sp1 by itself is involved since preliminary investigations have shown that Sp1 overexpression actually reduces transcriptional activation of the full-length cytochrome c promoter and that restoration of transcriptional activation by A23187 is independent of Sp1. 2 PKC exists as a family of 11 homologous serine/threonine kinases, of which four isoforms (␣, ␤ I , ␤ II , and ␥) are Ca 2ϩ -dependent (45). Here, we show that the effect of the ionophore on cytochrome c transactivation is dramatically reduced by PKC inhibitors, indicating that a Ca 2ϩ -sensitive, PKC-dependent pathway is involved in L6E9 cells. This conclusion is further fortified by the fact that cells cotransfected with the Ca 2ϩsensitive PKC␣ and PKC␤ II isoforms illustrate a greater response to the Ca 2ϩ ionophore than cells cotransfected with the Ca 2ϩ -insensitive PKC␦ isoform. Since it was reported that the only detectable Ca 2ϩ -sensitive isoform in L6 cells under nontransfected conditions is PKC␣ (36) and that this isoform is known to be down-regulated by prolonged treatment with TPA (36), these data suggest that the Ca 2ϩ -mediated effect on cytochrome c gene expression occurs via PKC␣. However, it appears that other regulatory influences exist as well since ϳ25% of the response triggered by A23187 was not abolished by TPA treatment, but was eliminated by the nonspecific PKC inhibitor staurosporine. The broader spectrum of action of staurosporine likely inhibited other kinases involved in parallel signal transduction pathway(s) that could have been activated by the ionophore. Such a candidate may be c-Jun N-terminal kinase since recent work using contractile activity to increase cytosolic Ca 2ϩ levels in cardiac myocytes (46) and skeletal muscle (47) have demonstrated its activation.
In vitro, PKC␣ has been shown to phosphorylate c-Raf (48), and c-Raf is known to phosphorylate and activate MEK directly (49), thus leading to the activation of the ERK1 and ERK2 MAPKs. The facts that the A23187-induced increase in cytochrome c transactivation was enhanced by PKC␣, that it was reduced by treatment of the cells with the MEK inhibitor PD98059, and that A23187 induced an increase in ERK1 and ERK2 activities with a time course consistent with their involvement in downstream cytochrome c transactivation strongly suggest that this pathway is active in Ca 2ϩ -mediated cytochrome c expression. Thus, the following sequence of events is supported by our data: 1) A23187 mobilizes Ca 2ϩ from the extracellular pool to increase its intracellular concentration, which 2) triggers the activation of a PKC/MEK/MAPK-dependent pathway leading to 3) the transactivation of cytochrome c. This is the likely cause of the subsequent increase in cytochrome c mRNA observed. Further work will define the requirement for de novo protein synthesis as well as the Ca 2ϩresponsive element involved and whether an increase in intracellular Ca 2ϩ levels can act as a general signal mediating an increase in the expression of a variety of nuclear genes encoding mitochondrial proteins.