The early cellular signals associated with contractile activity initiate the activation and induction of transcription factors that regulate changes in skeletal muscle phenotype. The transcription factors Egr-1, Sp1, and serum response factor (SRF) are potentially important mediators of mitochondrial biogenesis based on the prevalence of binding sites for them in the promoter regions of genes encoding mitochondrial proteins, including PGC-1α, the important regulator of mitochondrial biogenesis. Thus, to further define a role for transcription factors at the onset of contractile activity, we examined the time-dependent alterations in Egr-1, Sp1, and SRF mRNA and the levels in electrically stimulated mouse C2C12 skeletal muscle cells. Early transient increases in Egr-1 mRNA levels within 30 min (P < 0.05) of contractile activity led to threefold increases (P < 0.05) in Egr-1 protein by 60 min. The increase in Egr-1 mRNA was not because of increased stability, as Egr-1 mRNA half-life after 30 min of stimulation showed only a 58% decline. Stimulation of muscle cells had no effect on Sp1 mRNA but led to progressive increases (P < 0.05) in SRF mRNA by 30 and 60 min. This was not matched by increases in SRF protein but occurred coincident with increases (P < 0.05) in SRF-serum response element DNA binding at 30 and 60 min as a result of SRF phosphorylation on serine-103. To assess the importance of the recovery period, 12 h of continuous contractile activity was compared with four successive 3-h bouts, with an intervening 21-h recovery period after each bout. Continuous contractile activity led to a twofold increase (P < 0.05) in Egr-1 mRNA, no change in SRF mRNA, and a 43% decrease in Sp1 mRNA expression. The recovery period prevented the decline in Sp1 mRNA, produced a decrease in Egr-1 mRNA, and had no effect on SRF mRNA. Thus continuous and intermittent contractile activity evoked different specific transcription factor expression patterns, which may ultimately contribute to divergent qualitative, or temporal patterns of, phenotypic adaptation in muscle.
- phenotypic adaptation
- mitochondrial biogenesis
- cytochrome c
- transcription factors
chronic contractile activity of skeletal muscle, either in continuous fashion or that which is intermittent and followed by a recovery period, evokes the remodeling of muscle, stimulates mitochondrial biogenesis, and ultimately results in improved muscle function (17). The specific cellular events producing these mitochondrial adaptations likely include the combination of rapid ATP turnover, a mismatch between ATP supply and demand, and fluctuations in intracellular Ca2+ levels. These signals originate from propagating action potentials, causing sarcoplasmic reticulum Ca2+ release, subsequent activation of myosin ATPase and muscle contraction, and the simultaneous activation of signal transduction cascades, which can influence transcription factor function and expression.
Transcription factors play an essential role in regulating the expression of eukaryotic genes, including genes encoding mitochondrial proteins, by binding to specific nucleotide sequences within their promoter regions. Thus these regulatory proteins ultimately represent important intermediates in the regulation of mitochondrial biogenesis (17). Several studies have demonstrated changes in transcription factor gene expression that occur concurrently, or preceding, the contractile activity-induced regulation of mitochondrial biogenesis in skeletal muscle (7, 19, 24, 27, 31, 41). With the use of a variety of different experimental models, these studies have identified numerous transcription factors, including a family of immediate early genes, the early growth response gene-1 (Egr-1), c-fos, and c-jun, that respond to both acute and chronic exercise (2, 7, 24, 29, 36). In addition, our laboratory (7, 14) also recently demonstrated that specificity protein 1 (Sp1) can be induced by contractile activity and that both Egr-1 and Sp1 are involved in regulating the transcription of cytochrome c, a nuclear-encoded protein of the electron transport chain. Egr-1 and Sp1 are nuclear-encoded transcription factors that bind to GC-rich sequences within the promoter regions of target genes. They are hypothesized to play an important role in orchestrating the events involved in mitochondrial biogenesis based on the abundance of GC-rich sequences within the promoter regions of genes encoding mitochondrial proteins (17, 22, 28, 42).
Another transcription factor that may be important in regulating mitochondrial biogenesis is the serum response factor (SRF). It is known that SRF is involved in skeletal muscle remodeling as a result of hypertrophic stimuli such as mechanical overload or stretch in which SRF expression is increased (5, 11, 15). In contrast, mechanical unloading of muscle resulting in muscle atrophy occurs coincident with a decrease in SRF expression (15). Given the presence of SRF binding sites within the promoter regions of a number of nuclear genes regulating mitochondrial biogenesis, including peroxisome proliferator-activated receptor-γ (PPARγ) and PPARγ coactivator-1α (PGC-1α), it is suggested that SRF may also play a role in stimulating organelle synthesis in skeletal muscle if the appropriate stimulus (i.e., chronic exercise) is provided.
Mitochondrial biogenesis occurs as the result of a complex series of events that begins with the first bout of exercise and requires the accumulation of adaptive responses to multiple acute exercise bouts (17). The ultimate adaptation to contractile activity appears to be augmented by an intervening recovery period subsequent to the exercise bouts (24, 29–32, 36). Thus, in this study, we examined the effects of acute and repetitive bouts of contractile activity and the role of the recovery period on the expression of Egr-1, SRF, and Sp1 in a muscle cell culture system developed for electrical stimulation (7). We hypothesized that the acute changes in transcription factor gene expression would be enhanced after the introduction of a recovery period, an effect that may be mediated by changes in DNA binding. Using this cell culture stimulation model, we have recently shown time-dependent, contractile activity-induced changes in the expression of nuclear respiratory factor-1, PGC-1α, and mitochondrial transcription factor A, important regulators of mitochondrial biogenesis (19). These changes appear to coincide with the activation of a number of intracellular signaling pathways and may be linked to rapid changes in the expression of immediate early genes such as Egr-1. Furthermore, we also wanted to define some of the upstream events that mediate the rapid change in Egr-1 mRNA expression, since Egr-1 appears to be directly involved in the transcription of cytochrome c (14). Our findings uniquely demonstrate that contractile activity induces a differential pattern of transcription factor expression that is time dependent and that relies, in part, on the presence or absence of an intervening recovery period.
Actinomycin D (ActD) was purchased from Sigma (Oakville, Ontario). [α-32P]dCTP, [γ-32P]dATP, nitrocellulose, and nylon membranes (Hybond N) were obtained from Amersham Pharmacia Biotech (Baie D'Urfé, Québec). Fetal bovine serum and DMEM were purchased from Sigma (St. Louis, MO). Horse serum was purchased from Invitrogen (Burlington, Ontario). Trizol was purchased from Invitrogen. Polyclonal antibodies directed toward Egr-1 (588), Sp1 (PEP2), and SRF (G-20) and oligodeoxynucleotides used for the SRF electromobility shift assay were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).
Cell culture and electrical stimulation.
Murine C2C12 skeletal muscle cells were cultured as previously described (7). Myoblasts were proliferated in DMEM containing 10% fetal bovine serum and were differentiated into myotubes in DMEM containing 5% heat-inactivated horse serum. Treatments were routinely carried out when myotubes reached 80–90% confluence (∼6 days). Electrical stimulation (5 Hz, 55–65 V) of myotubes (7) was performed either acutely for 5, 15, 30, 60, or 180 min, chronically for 12 consecutive hours, or chronically with four successive 3-h bouts of contractile activity (12 h total; Refs. 7, 19) with an intervening 21-h recovery period between each 3-h bout.
Steady-state mRNA measurements and Northern blotting.
Total RNA was isolated from control and stimulated C2C12 cells using Trizol reagent (Invitrogen) following the manufacturer's recommendations. Total RNA was resuspended in diethyl pyrocarbonate-treated sterile H2O. The concentration and purity of the RNA were determined by ultraviolet spectrophotometry as the 260- to 280-nm readings. Thirty micrograms of total RNA were electrophoresed through denaturing formaldehyde-1% agarose gels, which were transferred and subsequently fixed to nylon membranes (Hybond-N, Amersham Pharmacia). After 2 h of prehybridization, membranes were probed and hybridized overnight at 42°C with random primer-labeled [32P]dCTP cDNAs specific for Egr-1, Sp1, and SRF. Equal loading was verified by inspection of the ethidium bromide-stained gel. The levels of Egr-1, Sp1, and SRF mRNA were quantified by electronic autoradiography (Instantimager, Packard).
mRNA stability measurements.
C2C12 myotubes were electrically stimulated for 30 min as previously described (7) or left untreated. Immediately afterward, a subset of myotubes was treated with either 10 μg/ml ActD or methanol as a vehicle-matched control for periods of 30, 60, 120, or 180 min. Total RNA was extracted, and 30 μg were electrophoresed through denaturing formaldehyde-1% agarose gels. Total RNA was transferred and subsequently fixed to nylon membranes (Hybond-N, Amersham) and subjected to Northern blot analyses. We calculated Egr-1 mRNA half-life using exponential decay analysis.
Electromobility shift assay.
DNA binding proteins were prepared from control and stimulated C2C12 myotubes as previously described (7) and were used in electromobility shift assays after protein concentration was determined. DNA binding proteins (50 μg) from control or stimulated myotubes were incubated with a [γ32P]dATP end-labeled oligodeoxynucleotide corresponding to the proximal serum response element (SRE) found within the mouse Egr-1 promoter (37). The sequence was 5′-GGA TGT CCA TAT TAG GAC ATCT-3′ (Santa Cruz Biotechnology). The labeled (40,000 counts/min) oligodeoxynucleotide was incubated with 0.5 μg/ml poly(dI-dC) and 50 μM pyrophosphate in binding buffer [20 mM Tris (pH 7.6), 50 mM EDTA (pH 8.0), 50 mM NaCl, 10% glycerol, 0.3 mg/ml BSA, and 1 mM DTT] at room temperature for 20 min. To determine the specificity of binding, competition and supershift assays were conducted by preincubating extracts with either a 100 molar excess of cold oligodeoxynucleotide or an antibody directed to SRF for 20 min before the addition of the labeled oligodeoxynucleotide. In addition, DNA binding proteins extracted from control and stimulated cells were also preincubated with a phosphospecific antibody that specifically detects SRF phosphorylation on serine-103 (Ser-103) (33). Samples were electrophoresed through 4% acrylamide gels (29:1 acrylamide-bisacrylamide) at 200 V for 2–3 h. Gels were fixed in acetic acid-methanol-H2O (10:30:60, vol/vol/vol), dried at 90°C for 1 h, and quantified using an Instantimager (Packard).
Total protein extracts were made from control and stimulated C2C12 myotubes using 1× lysis buffer (Promega, Madison, WI). Protease and phosphatase inhibitors [1 μl/ml leupeptin (0.3 μg/μl), aprotinin (0.3 μg/μl), pepstatin (0.3 μg/μl), 0.5 mM PMSF, 1.0 mM DTT, 5 μl/ml sodium pyrophosphate (250 mM), and 5 μl/ml sodium orthovanadate (250 mM)] were added before extraction. Protein (150 μg) was electrophoresed through 8% SDS-polyacrylamide gels, followed by transfer to nitrocellulose membranes. Total protein on the membrane was visualized by Ponceau red staining, and this was used to verify equal loading among the lanes. Blots were blocked (1 h) with 5% milk in 1× Tris-buffered saline-Tween 20 (TBST) and Tris·HCl (pH 7.4), followed by overnight incubation with antibodies directed toward Egr-1 (1:500), Sp1 (1:500), and SRF (1:500 diluted in 5% milk-TBST). After three 5-min washes at room temperature with TBST, blots were incubated at room temperature (1 h) with an anti-rabbit secondary antibody conjugated with horseradish peroxidase (1:1,000 diluted in 5% milk-TBST). Blots were again subjected to three 5-min washes at room temperature with TBST, visualized with an enhanced chemiluminescence kit (Amersham Pharmacia), and quantified using SigmaGel (Jandel).
Data from at least three separate experiments were pooled and presented as means ± SE and analyzed using a one-way ANOVA, followed by Tukey's post hoc test to determine individual differences. In experiments in which the effect of a 12-h bout of contractile activity on transcription factor mRNA expression was assessed, unpaired Student's t-tests were used for analysis. Differences were considered statistically significant at P < 0.05.
Egr-1, Sp1, and SRF mRNA and protein expression during acute stimulation.
We evaluated the acute responses of the transcription factors Egr-1, Sp1, and SRF to 5, 15, 30, and 60 min of stimulation. Egr-1 mRNA reached a maximum of approximately fourfold by 30 min (P < 0.001) and returned toward control values by 60 min of stimulation (Fig. 1, A and B). This is consistent with the rapid and transient response of Egr-1 mRNA to acute stimulation, which we and others have previously observed both in vitro (2, 6) and in vivo (22). The mRNA expression of Sp1 and SRF in response to contractile activity differed considerably from that of Egr-1. No change in Sp1 mRNA expression was observed at any stimulation time point. However, increases in SRF mRNA occurred progressively, reaching 2- and 2.5-fold above control cells by 30 and 60 min of stimulation (P < 0.0001; Fig. 1, A and B). We then assessed whether these responses were associated with changes in protein levels (Fig. 1C). As expected from the mRNA data, no change in Sp1 protein was observed over the 60-min contraction period. Furthermore, despite the progressive increase in SRF mRNA levels, there was no evidence that this was reflected at the protein level within the 60-min contraction period. In contrast, Egr-1 protein levels increased (P < 0.05) progressively, reaching a maximum of threefold by 60 min of stimulation (Fig. 1D).
DNA binding and mRNA stability during acute stimulation.
Binding sites for SRF exist within its own promoter, as well as in the Egr-1 promoter (3, 5). Thus, to evaluate whether the contractile activity-induced increases in SRF or Egr-1 mRNA could be transcriptionally regulated by SRF, we measured SRF-SRE DNA binding using an electromobility shift assay (Fig. 2A). Stimulation of cells led to a progressive increase (P < 0.05) in SRF-SRE DNA-binding, reaching a maximum of 1.7- and 1.8-fold above control levels after 30 and 60 min (Fig. 2B). SRF-SRE binding was inhibited by a 100-fold molar excess of cold oligodeoxynucleotide (Fig. 2A, lane 2) and supershifted with an antibody directed toward SRF (Fig. 2A, lane 3). When we used a phosphospecific antibody that detects SRF phosphorylated on Ser-103 (33), we also observed a supershift at 30 and 60 min (Fig. 2A, lanes 6 and 7). These data suggest the possibility that phosphorylated SRF is bound in greater amounts to the SRE after 30 and 60 min of stimulation, an effect that may be important for the contractile activity-induced expression of Egr-1 mRNA expression.
The contractile activity-induced change in Egr-1 mRNA could also be a result of increases in mRNA stability. To evaluate this, we measured Egr-1 mRNA decay in cells after ActD administration. The Egr-1 mRNA half-life was calculated to be 66 min in control, nonstimulated myotubes (Fig. 2C). In contrast, the half-life was reduced by 58% to 28 min in myotubes that had been stimulated for 30 min before ActD administration. This indicates that Egr-1 mRNA stabilization cannot account for the increase in Egr-1 mRNA observed as a result of contractile activity and that transcriptional activation, possibly via SRF, is the more likely mechanism involved.
Effects of time and recovery on Egr-1, Sp1, and SRF mRNA expression.
Our previous findings (6) indicated that chronic contractile activity (3 h/day followed by 21 h of recovery for 4 consecutive days) significantly induced elevations in Sp1 but not in Egr-1 protein levels, coincident with increases in DNA binding to the cytochrome c promoter (6). This led us to examine the effects of time and the recovery period in the contractile activity-mediated changes in Sp1 and Egr-1 mRNA expression. We also extended our analysis to include SRF, since it appears that SRF has an important role in regulating Egr-1 transcription (5), as well as genes involved in mitochondrial biogenesis (16, 26). Three hours of stimulation also resulted in divergent mRNA responses of Egr-1, Sp1, and SRF (Fig. 3), as reflected at the earlier time points (Fig. 1). Egr-1 mRNA was 1.7-fold higher compared with nonstimulated control cells after 3 h of stimulation. With the introduction of the recovery period 21 h after stimulation (i.e., day 1), Egr-1 mRNA levels were reduced to values that were only 35% of those found in nonstimulated cells. A similar reduction was observed at all days (2, 3, and 4) after each 21-h recovery period. SRF mRNA was 2.2-fold (P < 0.05) above control, nonstimulated values after 3 h of stimulation. However, in contrast to Egr-1, the recovery period did not produce a decline in SRF mRNA to levels below those found in control, nonstimulated cells. Sp1 mRNA expression was unchanged throughout the stimulation and recovery time courses.
We then compared the mRNA responses described above with those observed after a single, continuous 12-h bout of contractile activity to determine whether the changes in transcription factor gene expression would be altered by the presence of a recovery period. In response to 12 h of continuous stimulation, Egr-1 mRNA expression increased (P < 0.05) twofold, SRF mRNA expression remained unchanged, and Sp1 mRNA levels declined (P < 0.05) to 43% of control values (Fig. 3, A and B, right). Thus the recovery period plays a role in determining the adaptive response of Egr-1 and Sp1 mRNA, but it does not significantly modify the expression of SRF mRNA to contractile activity.
The contractile-activity mediated coordination of early molecular events plays an important role in initiating, and eventually sustaining, the biochemical adaptations observed during mitochondrial biogenesis in muscle (17, 18, 38). This coordination involves the timely activation of signaling molecules, followed by the covalent modification of transcription factors and transcriptional coactivators. We undertook the present study with the purpose of characterizing the time-dependent, contractile activity-induced changes in transcription factor gene expression, using an isolated cellular system in the absence of neural and humoral factors.
One of the earliest alterations in gene expression in response to contractile activity is the rapid and transient increase in Egr-1 mRNA expression (7, 24). Functionally, Egr-1 acts as a DNA-binding transcription factor, which has been shown to regulate the expression of many genes, including some nuclear genes encoding mitochondrial proteins (9, 22, 28, 42). Our data confirm that Egr-1 expression is regulated by a contractile activity stimulus, as previously shown in vivo (24). Indeed, the induction of Egr-1 mRNA expression can be elicited by a variety of intracellular events that occur during muscle contraction, including electrical (2, 7, 39) and mechanical stimuli (16, 26), changes in intracellular Ca2+ concentration (1, 14), and growth factor availability. In this study, we demonstrate that the early increase in Egr-1 mRNA expression is mediated via rapid increases in transcription, not through elevations in the stability of Egr-1 mRNA. In fact, our data indicate for the first time that the increase in Egr-1 mRNA was accomplished despite a marked contractile activity-induced reduction in mRNA stability, suggesting that the transcriptional activation of the Egr-1 gene was likely augmented to a greater extent than the mRNA levels would suggest (i.e., greater than 4-fold). Furthermore, our data indicate that rapidly induced mRNA transcripts are labile and subject to rapid decay in skeletal muscle. This is clearly a feature of immediate early gene (i.e., Egr-1) mRNA stability, since mRNAs encoding structural proteins in skeletal muscle are normally resistant to mRNA decay (8).
In other cell types, the induction of Egr-1 mRNA expression has been shown to occur via transcriptional activation of the Egr-1 promoter, primarily through a proximal SRE (6). Thus we investigated the possibility that the contractile activity-mediated induction of Egr-1 expression may involve SRF. The increase in SRF mRNA expression was slower in onset and increased progressively, reaching its maximum level by 30 and 60 min of stimulation. This delayed response is unlike other immediate early genes, such as Egr-1 and c-fos, which exhibit rapid and transient changes in response to contractile activity (7, 26). Although SRF protein expression did not change, SRF-SRE DNA binding increased by 30 and 60 min of stimulation. SRF is a nuclear phosphoprotein that is covalently modified by phosphorylation on a Ser residue at position 103, an effect that serves to enhance the rate and affinity with which SRF binds the SRE (33). This suggests that the increased SRF-DNA binding is mediated via a contractile activity-induced activation of an upstream kinase, a result confirmed by our electromobility shift analysis using a phosphospecific SRF antibody to supershift the SRF-SRE DNA complex. This increased DNA binding could also be influenced by the contractile activity-induced activation of other accessory factors binding at sites adjacent to the SRE. To date, the identity of the contractile activity-induced kinase that phosphorylates SRF is unknown, although a likely candidate is a Ca2+/calmodulin-dependent kinase (11–13).
In addition to its involvement in Egr-1 expression (4, 6), SRF also autoregulates its own transcription via SREs present within its own upstream regulatory region (3, 35). Functional SRE binding sites also exist within the β-F1 ATPase promoter (28) as well as within the 5′-regulatory region of the important transcriptional coactivator PGC-1α (10). PGC-1α is an important component of the transcriptional machinery involved in regulating the expression of numerous genes implicated in mitochondrial biogenesis (17, 40). Thus our results are consistent with the possibility that SRF could serve as an important upstream regulator of mitochondrial biogenesis.
The contractile activity-induced expression of multiple genes appears to be augmented by an intervening recovery period (27, 29–32). This response also applies to mitochondrial proteins (30, 36), suggesting that the recovery period represents an important consequence of chronic exercise. We evaluated whether this could be extended to the expression of transcription factors by stimulating C2C12 myotubes in the presence and absence of a recovery period. We compared the effect of a single, continuous 12-h bout of contractile activity to four successive 3-h bouts with an intervening 21-h recovery period after each bout. We hypothesized that the changes brought about by continuous contractile activity would be further enhanced after the introduction of recovery time. Our results indicate that the responses were not uniform among the transcription factors measured. The most dramatic effect of the influence of the recovery period occurred with Egr-1. Although Egr-1 mRNA expression increased after 12 h of continuous contractile activity, the introduction of a recovery period led to a marked reduction in Egr-1 mRNA expression. This may be due to the contractile activity-induced increase in mRNA decay (Fig. 2), a process that may extend for a prolonged period of time during the recovery period, leading to a reduced mRNA level. During continuous contractile activity, the higher transcription rate compared with the rate of degradation produces the increase in mRNA observed. It is also notable that the lability of Egr-1 mRNA levels during contractile activity and recovery apparently has a minimal effect on Egr-1 expression at the protein level, since we have previously observed that Egr-1 protein expression is unchanged relative to nonstimulated cells using the same contractile activity-recovery paradigm (7). This clearly illustrates the independence of mRNA and protein turnover control mechanisms within muscle cells.
In contrast to Egr-1, Sp1 expression displayed an opposite pattern. Twelve hours of continuous contractile activity resulted in a 43% decrease in Sp1 mRNA expression. The inclusion of a recovery period served to maintain Sp1 mRNA levels at values that were not different from those of nonstimulated control cells. This sustained mRNA level may help to drive Sp1 translation, a process that could be accelerated by contractile activity, leading to the higher Sp1 protein level that we have previously observed with this experimental model (7). The increase in Sp1 protein, along with possible posttranslational modifications to the Sp1 protein (23), may be important during mitochondrial biogenesis, since Sp1 binding sites are ubiquitously found within promoter regions of nuclear genes encoding mitochondrial proteins (22, 28, 42).
In comparison to Sp1 and Egr-1, SRF mRNA expression returned to control values after a single, continuous 12-h bout of stimulation, and this effect was not altered with the introduction of the recovery period. Preliminary observations indicate that SRF protein levels are increased after 4 days of intermittent contractile activity, as used in the present study (unpublished observations). We suspect that this is mediated by the synthesis of SRF protein, which is relatively stable (i.e., has a low turnover rate) within the first 3 h of contractile activity (Fig. 3) and which accumulates during the 4 days of stimulation.
In conclusion, the data presented in this study, as well in our laboratory's previous work (7, 14), suggest that Egr-1, SRF, and Sp1 may be involved in mediating some of the adaptations occurring in skeletal muscle in response to elevated contractile activity, including mitochondrial biogenesis. Our results demonstrate that the contractile activity-mediated induction of transcription factor mRNA expression is highly divergent, despite similar alterations in the cellular environment (e.g., changes in Ca2+ levels or energy state) that are brought about by contractile activity and restored by the recovery period. This suggests that the early, upstream signal transduction events (i.e., kinase activation) that lead to these changes in mRNA diverge sufficiently to produce very different, and specific, transcription factor mRNA responses. Furthermore, the contractile activity-induced responses differ depending on the presence or absence of intervening recovery periods. This variability in transcription factor mRNA induction and turnover is likely important for the time-dependent, gene-specific transcription events that are responsible for many muscle phenotypic adaptations to contractile activity at the level of contractile proteins, the Ca2+ handling system, and mitochondrial biogenesis. Future work combining the use of RNA interference and overexpression strategies, along with acute or repetitive contractile activity, can be used to directly address the role of these transcription factors in mediating the muscle phenotypic adaptations that are initiated at the onset of exercise.
This work was supported by grants from the Natural Science and Engineering Research Council of Canada (NSERC) and the Canadian Institutes of Health Research. I. Irrcher was the recipient of a NSERC postgraduate scholarship. D. A. Hood is the holder of a Canada Research Chair in Cell Physiology.
The authors thank Dr. M. E. Greenberg (Harvard Medical School, Boston, MA) for the provision of the phosphospecific SRF antibody, Drs. J. A. Carson (University of South Carolina, Columbia, SC) and R. J. Schwartz (Baylor College of Medicine, Houston, TX) for the SRF cDNA, and Dr. T. Haas (York University, Toronto, ON, Canada) for the Sp1 cDNA. We thank Sabena Lowe for help with the electromobility shift assays.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society