GATA-6 can act as a positive or negative regulator of smooth muscle-specific gene expression.

The GATA-4/5/6 family of transcription factors is important for the development of the cardiovascular system and the visceral endoderm. GATA-6 is the only family member expressed in vascular smooth muscle cells and has been shown to be important for controlling the phenotype of these cells following vascular injury. To clarify further the role of GATA-6 in regulating vascular smooth muscle differentiation, we directly examined its ability to regulate the promoters of smooth muscle-specific genes. This analysis revealed that GATA-6 strongly repressed telokin promoter activity. In contrast, GATA-6 activated the smooth muscle myosin heavy chain and smooth muscle alpha-actin promoters and had no significant effect on the SM22alpha promoter. Gel mobility shift assays demonstrate that GATA-6 binds to a consensus site adjacent to the CArG box in the telokin promoter. GATA-6 did not interfere with the serum-response factor-stimulated promoter activity but blocked myocardin-induced activation of the telokin promoter. In contrast, GATA-6 and myocardin resulted in synergistic activation of the smooth muscle myosin heavy chain promoter. Consistent with these findings, overexpression of GATA-6 in smooth muscle cells selectively inhibited expression of endogenous telokin, while simultaneously increasing expression of other smooth muscle proteins. These data suggest that GATA-6 selectively inhibits telokin expression by triggering the displacement of myocardin from the serum-response factor. As GATA-6 is expressed at high levels in vascular smooth muscle, this finding may explain the relatively low levels of telokin expression in the vascular system. These data also reveal a novel transcription regulatory mechanism by which GATA-6 can modulate the activity of the myocardin-serum-response factor complexes.

Smooth muscle cells arise from diverse populations of precursor cells during embryonic development. All differentiated smooth muscle is characterized by the presence of unique isoforms of contractile proteins such as smooth muscle ␣and ␥-actin, smooth muscle myosin heavy chain (Sm-MHC), 1 SM22␣, telokin, and calponin (1)(2)(3)(4)(5). Although the molecular mechanisms regulating gene expression during smooth muscle development are poorly understood, a number of transcription factors have been identified that regulate expression of these smooth muscle contractile proteins (6). Among these factors, the serum-response factor (SRF), which binds to CArG elements, plays a key role in the regulation of most smooth muscle restricted genes (reviewed in Ref. 7). SRF is a member of the MADS (MCM1, Agamous, and Deficiens, SRF) box family of transcription factors that are important regulators of many genes associated with cell growth and differentiation. SRF is enriched in cardiac, skeletal, and smooth muscle progenitor cells during embryogenesis, as well as in terminally differentiated adult muscle cells (8,9). Targeted disruption of the mammalian SRF gene leads to malformation of the mesoderm, indicating a critical role for SRF in mesoderm development (10,11).
In addition to forming complexes with SRF, several SRFassociated proteins, including the GATA proteins, also act as independent transcription factors upon binding to their own cognate cis-acting elements. GATA proteins are a family of zinc finger transcription factors that play essential roles in development through their interaction with DNA regulatory elements that contain a consensus WGATAR motif. In vertebrates, six GATA family members have been identified that can be divided into two subgroups based on sequence homology and tissue distribution (27). The GATA-1/2/3 subfamily is expressed in the hematopoietic cell lineage and plays a critical role in the development of this lineage (28). On the other hand, the GATA-4/5/6 subfamily is thought to be involved with cardiac, gut, and blood vessel formation (29). During early murine embryonic development, the patterns of GATA-6 and -4 gene expression are similar, with expression being detected in the precardiac mesoderm, the embryonic heart tube, and the primitive gut. However, during development GATA-6 becomes the only member of the family expressed in vascular smooth muscle cells (VSMC) (30,31). GATA-6 expression in VSMCs is rapidly down-regulated upon mitogen stimulation or vascular injury (32). Adenovirus-mediated GATA-6 gene transfer to the vessel wall after balloon injury partially inhibited lesion formation and reversed the down-regulation of Sm-MHC, smooth muscle ␣-actin (SM ␣-actin), calponin, vinculin, and metavinculin expression that is normally associated with injury-induced VSMC phenotypic modulation (32). These data suggest that GATA-6 plays a critical role in the maintenance of the differentiated phenotype in VSMCs.
In the current study, we further investigated the roles of GATA-6 in regulating smooth muscle-specific gene transcription. We found that GATA-6 strongly repressed the smooth muscle-specific telokin promoter activity while activating the Sm-MHC and SM ␣-actin promoters. GATA-6-mediated repression of the telokin promoter occurred through direct competition between GATA-6 and myocardin for binding to SRF. The GATA-6-binding site in the telokin promoter located adjacent to the CArG box was critical for this inhibitory activity. Together our data reveal a novel transcription regulatory mechanism by which GATA-6 can modulate the activity of myocardin-SRF complexes.

EXPERIMENTAL PROCEDURES
Plasmid Constructs and Promoter-Reporter Gene Assays-Mouse GATA-6 wild type cDNA and adenovirus were both kindly provided by Dr. Jeffery D. Molkentin (Department of Pediatrics, University of Cincinnati, Children's Hospital Medical Center). Mouse myocardin wildtype cDNA in pcDNA3.1-myc/his vector was kindly provided by Dr. Eric N. Olson (University of Texas, Southwestern Medical Center, Dallas). An expression plasmid containing the human SRF cDNA was the generous gift from Dr. Ron Prywes (Columbia University). All promoterreporter genes were constructed by cloning fragments of promoters into the pGL 2 B luciferase vector (Promega). The rabbit telokin promoter (T400)-luciferase reporter gene used includes nucleotides Ϫ256 to ϩ147 of the telokin gene as described previously (33). The GATA site (Ϫ60 to Ϫ57) deletion mutant of the rabbit telokin promoter, T400 (⌬GATA), was generated by PCR-mediated mutagenesis using the QuickChange site-directed mutagenesis kit, according to the manufacturer's directions (Stratagene). The SM22␣-luciferase reporter gene includes nucleotides Ϫ475 to ϩ61 of the mouse SM22␣ gene (34). The SM ␣-actin promoter fragment extended from nucleotide Ϫ2,555 to ϩ2,813 (35) and the Sm-MHC promoter from Ϫ4,200 to ϩ11,600 (36); these plasmids were kindly provided by Dr. Gary Owens (University of Virginia).
Plasmids were transfected into rat A10 smooth muscle cells using FuGENE 6 (Roche Applied Science). A10 smooth muscle cells were grown in high glucose Dulbecco's modified Eagle's medium containing 50 units/ml penicillin, 50 mg/ml streptomycin, and 20% fetal bovine serum. A10 cells to be transfected were seeded at 3 ϫ 10 4 cells/well in 24-well plates. 16 -18 h post-seeding, each dish was washed once with phosphate-buffered saline (pH 7.4), replaced with 0.5 ml of complete medium, and incubated with a total of 1 g of plasmid DNA and 2 l of FuGENE in 50 l of Dulbecco's modified Eagle's medium. 24 h later, 10 l of cleared extracts (100 l/well) were prepared for dual luciferase assays by using a dual luciferase reporter assay system according to the manufacturer's directions (Promega). Reporter gene luciferase activities were normalized to the luciferase activity of the internal control.
Gel Mobility Shift Assays-Nuclear extracts were prepared from COS cells transfected with GATA-6 or SRF expression plasmids. Gel mobility shift assays were performed in a final volume of 15 l. Binding mixtures contained 0.2 ng (1.5 ϫ 10 4 cpm) of end-labeled double-stranded DNA probe, 200 ng of poly(dI-dC), 4.5 g of bovine serum albumin, and various amounts of nuclear extract protein. For cold competition experiments, a 50-fold excess of unlabeled double-stranded oligonucleotide competitors was included in some reactions as indicated in the figure legends. Annealed oligonucleotides were labeled using [ 32 P]dCTP and Klenow DNA polymerase (Promega). Unincorporated [ 32 P]dCTP was removed by agarose gel electrophoresis. All binding reactions were incubated for 20 min at room temperature, and the DNA-protein complexes were then resolved by electrophoresis through 4% polyacrylamide gels containing 6.75 mM Tris (pH 7.9), 3.3 mM sodium acetate (pH 7.9), 1 mM EDTA, and 2.5% glycerol. The gel was dried and autoradiographed with intensifying screens at Ϫ80°C overnight.
Adenovirus Transduction and Western Blotting-For adenoviral transduction, A10 cells were seeded in 6-well plates at a density of 2 ϫ 10 5 cells/well and grown overnight to near-confluence. These cells were washed with phosphate-buffered saline to remove serum and infected with adenovirus encoding LacZ or GATA-6 in phosphate-buffered saline at a multiplicity of infection of 100 for 4 h at 37°C. These conditions resulted in close to 100% infection of cells. 72 h following infection, cell protein extracts were prepared using RIPA buffer, and protein concentrations were determined by using the BCA protein assay kit (Pierce). Western blotting analysis was carried out essentially as described previously (37). Thirty micrograms of protein were fractionated on 7.5 or 15% SDS-polyacrylamide gels. The protein sample was electrophoretically transferred to a polyvinylidene difluoride membrane and verified by Ponceau S staining. The membrane was then probed with a series of antibodies. The secondary antibody, anti-mouse or anti-rabbit IgG (1: 10,000 dilution), conjugated with horseradish peroxidase was visualized using Supersignal West Pico chemiluminescent substrate (Pierce) according to the manufacturer's instructions. Chemiluminescence was detected and quantitated by using a CCD camera system (Fujifilm, Stamford, CT). Antibodies used in this study were as follows: polyclonal antibodies against telokin (1:6,000), SM22␣ (1:6000; a gift from Dr. Len Adam), nonmuscle myosin heavy chain IIb (NMHCIIb, 1:5000) (1), GATA-6 (1:10,000; Santa Cruz Biotechnology), and monoclonal antibodies against calponin (1:10,000; Sigma, clone hCP), SM ␣-actin (1:10,000; Sigma, clone 1A4), and MLCK (1:10,000; Sigma, clone K36).

Primary Mouse Colon Smooth Muscle Cell
Preparation-Mouse colonic smooth muscle cells were prepared as follows. Briefly, the colon was dissected out of 4-week-old mice. The epithelial layer was removed by blunt dissection, and the remaining muscle layer was minced in 5 ml of tissue digestion buffer (1 mg/ml collagenase type IV (Sigma), 0.125 mg/ml elastase, 0.25 mg/ml trypsin inhibitor, 2 mg/ml bovine serum albumin, 0.2 mg/ml DNase in phosphate-buffered saline containing 1 mM CaCl 2 ). The minced tissue was digested for 1 h at 37°C with shaking. Cells were passed through cell sieve, washed three times with Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, and then plated at a density of ϳ6 ϫ 10 4 cells per well in a 24-well plate. The media were changed on the cells daily, and 5 days after plating they were infected with adenovirus as described above.
Statistical Analysis-Data are expressed as the mean Ϯ S.E. The statistical significance of the differences between the means of the groups was determined by one-way analysis of variance or unpaired two-tailed Student's t tests. A value of p Ͻ 0.05 was considered significant.

GATA-6 Selectively Represses the Activity of the Telokin Promoter, While
Activating the SM ␣-Actin and Sm-MHC Promoters-To examine the effect of GATA-6 on smooth muscle-specific promoter activity, GATA-6 expression plasmids were cotransfected into A10 cells together with various reporter plasmids consisting of a firefly luciferase gene driven by SM ␣-actin, Sm-MHC, SM22␣, or telokin promoters. Each of these promoters has been shown previously (34 -36) to direct smooth muscle-restricted transgene expression in vivo. Results showed that GATA-6 significantly increased the activity of SM ␣-actin and Sm-MHC promoters, but it did not change SM22␣ promoter activity, and GATA-6 overexpression strongly repressed telokin promoter activity by 47 Ϯ 2% (Fig.  1A). The effect of GATA-6 on telokin promoter activity was also examined in an A10 cell line stably expressing an integrated mouse telokin promoter (Ϫ190 to ϩ171) firefly luciferase reporter gene. Overexpression of GATA-6 by adenoviral transduction significantly repressed telokin promoter activity relative to cells transduced with a control ␤-galactosidase adenovirus (Fig. 1B).
GATA-6 Binds to a Consensus Binding Sequence in an AT-rich Region Adjacent to the CArG Box of the Telokin Promoter-To determine directly the binding site for GATA-6 within the core of the telokin promoter, gel mobility shift assays were per-formed. A gel mobility shift assay using a probe that encompasses Ϫ81 to Ϫ42 of the rabbit telokin promoter, including an E-box, AT-rich region, and CArG box ( Fig. 2A), demonstrated that SRF and GATA-6 nuclear extracts both bind specifically to this fragment. These mobility-shifted complexes could be supershifted by antibodies to the respective proteins but not by the nonspecific antibodies (Fig. 2B). The GATA-6 mobilityshifted complex could be competed away by unlabeled fragments encompassing the full-length cold wild type probe and with the AT-rich region alone, but not by a fragment that included only the E box or CArG box alone (Fig. 2C). This result suggests that GATA-6 binds to the AT-rich region between Ϫ71 and Ϫ59 of the telokin promoter. At the 3Ј end of the AT-rich region, adjacent to the CArG box, there is a consensus binding sequence (TATC) for GATA factors (Fig. 3A). To determine whether this consensus binding sequence is required for GATA-6 binding, a series of gel mobility shift probes were generated that contained single nucleotide mutations or a 4-bp deletion within the AT-rich region. Results from these assays demonstrated that deletion of the GATA consensus sequence (⌬GATA) abolished GATA-6 binding, and mutation of the ad-jacent cytosine to a guanine decreased binding (Fig. 3B, 12C-G). In contrast, single base mutations within the remainder of the AT-rich region did not prevent GATA-6 binding. As a control we demonstrated that all probes were able to bind to SRF with similar efficacy (Fig. 3B, lower panel). These results demonstrate that GATA-6 binds to a consensus binding sequence (TATC) in the AT-rich region, adjacent to the CArG box, in the core of the telokin promoter. To determine whether GATA-6 binding to this site is required for GATA-6 to inhibit telokin promoter activity, the consensus binding sequence (TATC) was deleted within the context of the Ϫ256 to ϩ147 telokin reporter gene. Deletion of the GATA-binding site from the telokin promoter abolished the ability of GATA-6 to repress the promoter (Fig. 3C), demonstrating that direct binding of GATA-6 to the telokin promoter is required for its ability to inhibit promoter activity.
Effect of GATA-6 on SRF-induced Activation of the Telokin Promoter-Given that both SRF and GATA proteins interact FIG. 1. GATA-6 selectively inhibits the activity of the smooth muscle-specific telokin promoter. A, A10 vascular smooth muscle cells were transiently transfected with the indicated promoter-luciferase reporter genes and either a GATA-6 expression vector or an empty vector together with an internal control Renilla luciferase plasmid. 24 h following transfection, cells were lysed and assayed for luciferase activity. Promoter activity was normalized to the internal control (Renilla luciferase). Promoter activity relative to vector control transfections is presented as the mean Ϯ S.E. of six samples. B, a clonal A10 cell line stably expressing a mouse telokin promoter (Ϫ190 to ϩ171) firefly luciferase reporter gene was infected with LacZ or GATA-6 adenovirus in phosphate-buffered saline at a multiplicity of infection of 100 for 4 h at 37°C. 24 h later, cells were lysed and assayed for luciferase activity. Promoter activity was normalized to total protein in each extract. Promoter activity is presented as mean Ϯ S.E. of six samples. *, p Ͻ 0.01 versus control group.
FIG. 2. GATA-6 binds to an AT-rich region adjacent to the CArG box of the telokin promoter. A, probes used for gel mobility shift assays in B and C. B, 32 P-labeled double-stranded oligonucleotide encompassing the E box, AT-rich region, and CArG box from the telokin promoter (Ϫ81 to Ϫ42, Core probe in A) was incubated with SRF, GATA-6, or nontransfected COS cell nuclear extracts, as indicated. Following incubation for 20 min at room temperature, anti-SRF, anti-GATA-6, or anti-Sp1 antibodies were added and incubated for a further 1 h on ice. Samples were run on a 4% polyacrylamide gel, and mobilityshifted complexes were visualized by autoradiography. The positions of the specific mobility-shifted complexes are indicated by the arrows, and complexes resulting from antibody supershifts are indicated by the asterisks. C, to further refine the GATA-binding site within the telokin promoter core, the probe used in B was incubated with 1 g of GATA-6 nuclear extract in the presence of a 50-fold excess of each of the cold competitor oligonucleotides as indicated.

GATA-6-Myocardin Competition
with adjacent binding sites in the telokin promoter, and previous data showing the importance of SRF for telokin promoter activity (38), we tested the possibility that the repressive effect of GATA-6 may result from GATA-6 preventing SRF from interacting with the promoter. Luciferase reporter gene assays in A10 smooth muscle cells showed that overexpression of SRF produced a 3-fold increase in promoter activity. Overexpression of GATA-6 and SRF produced a small synergistic response, indicating that GATA-6 does not block the ability of SRF to activate the promoter (Fig. 4A).
Effect of GATA-6 on Myocardin-induced Activation of the Telokin and Sm-MHC Promoters-Myocardin is a smooth and cardiac muscle-restricted transcriptional coactivator of SRF that is required for smooth muscle differentiation (20 -22, 39). Myocardin has been shown to increase telokin expression in rat aortic smooth muscle cells (24), and we have recently shown that myocardin strongly activated the telokin promoter to levels similar to that reported for other smooth muscle-specific promoters (Fig. 4). 2 As both GATA-6 and myocardin interact with SRF, we examined the effect of GATA-6 on myocardininduced promoter activity. Results show that myocardin upregulates telokin promoter activity about 10-fold in A10 cells.  However, increasing concentrations of GATA-6 resulted in a dose-dependent decrease in the ability of myocardin to activate the telokin promoter (Fig. 4B). In contrast, GATA-6 augmented the activation of the Sm-MHC promoter by myocardin in A10 cells (Fig. 4C). Thus, GATA-6 selectively inhibits the ability of myocardin to activate the telokin promoter.
GATA-6 Competes with Myocardin for Binding to SRF-To confirm that GATA-6 and myocardin compete for binding to SRF, a mammalian two-hybrid assay was utilized. As shown in Fig. 5A, increased reporter activity, indicating a protein-protein interaction, was observed in cells expressing pBIND-SRF and pACT-myocardin or pBIND-SRF and pACT-GATA-6. However, coexpression of pBIND-GATA-6 and pACT-myocardin have no effect on the reporter activity in our systems, indicating that these two proteins do not interact directly or interact with each other in an unstable and transient manner. To dem-onstrate a direct competition between GATA-6 and myocardin for SRF binding, A10 cells were transfected with pG5Luc reporter plasmid and expression plasmids for pBIND-SRF and pACT-myocardin in the presence or absence of pcDNA3.1-GATA-6. Results show that increased expression of GATA-6 repressed reporter activity, indicating decreased myocardin-SRF interactions (Fig. 5B). These data indicate that GATA-6 can compete directly with myocardin for binding to SRF.
Effect of GATA-6 on Endogenous Smooth Muscle Protein Expression-To examine further the role of GATA-6 in regulating smooth muscle protein expression in vivo, the ability of adenovirally expressed GATA-6 to regulate endogenous smooth muscle protein expression in A10 smooth muscle cells was evaluated. A10 smooth muscle cells were transduced with adenovirus encoding LacZ or GATA-6. At 72 h after infection, cells were harvested, and endogenous protein expression was analyzed by Western blotting. This analysis revealed that GATA-6 down-regulated endogenous telokin expression in A10 cells, whereas expression of the 220-kDa MLCK, SM22␣, and calponin was significantly increased (Fig. 6). Expression of SM ␣-actin was not significantly altered by increased GATA-6 expression. Overexpression of GATA-6 in 10T1/2 cells was not able to FIG. 5. GATA-6 interferes with the binding of SRF and myocardin. Mammalian two-hybrid assays were performed to evaluate the binding interactions between myocardin, SRF, and GATA-6. A, A10 cells were transfected with a pG5Luc reporter plasmid and expression plasmids for fusion proteins of GAL4 DNA binding domain-SRF (pBIND-SRF), VP16 activation domain-myocardin (pACT-myocardin), GAL4 DNA binding domain-GATA-6 (pBIND-GATA-6), and VP16 activation domain-GATA-6 (pACT-GATA-6). Empty vectors, pBIND and pACT, were used as controls, as indicated. Increased luciferase activity, indicative of protein-protein interactions, was seen when pBIND-SRF and pACT-GATA-6 were cotransfected and when pBIND-SRF and pACT-myocardin were cotransfected. No changes in luciferase activity were observed when pBIND-GATA-6 and pACT-myocardin were cotransfected. B, A10 cells were transfected with pG5Luc reporter plasmid together with expression plasmids for fusion proteins of GAL4-SRF (pBIND-SRF) and VP16-myocardin (pACT-myocardin) in the presence or absence of pcDNA3.1-GATA-6 (0.1-0.3 g). Cells were maintained in Dulbecco's modified Eagle's medium with 20% serum for 24 h after transfection and then harvested for luciferase assays. Promoter activity is expressed as luciferase activity normalized by Renilla luciferase control (mean Ϯ S.E., n ϭ 6). *, p Ͻ 0.05; **, p Ͻ 0.01 versus pBIND and pACT-Myocardin cotransfection group alone. induce expression of endogenous smooth muscle genes, suggesting that GATA-6 alone is not sufficient to drive fibroblast cells to differentiate into smooth muscle cells (data not shown). The effect of GATA-6 on telokin and Sm-MHC expression was also examined in primary mouse colonic smooth muscle cells. Immunoblot analysis revealed that transduction with GATA-6 virus for 72 h down-regulated the expression of telokin, whereas the expression of Sm-MHC was markedly increased (Fig. 7). DISCUSSION In this study, we found that GATA-6 selectively repressed telokin promoter activity while activating the Sm-MHC promoter. GATA-6 binds to a consensus sequence in the telokin promoter adjacent to a CArG box that binds SRF, whereas the GATA-binding site in the Sm-MHC promoter is distal to important CArG boxes (40). This specific arrangement of the GATA-6 and SRF-binding sites may provide the structural basis for specific inhibitory effects of GATA-6 on the telokin promoter.
Previous studies have shown that GATA-6 binds a conserved GATA-like motif within the Sm-MHC promoter and that p300 and GATA-6 cooperate in activating the Sm-MHC promoter (40). In the current study, we confirm these findings, and we also demonstrate that overexpression of GATA-6 in primary smooth muscle cells increased endogenous Sm-MHC expression (Fig. 7). In contrast, we found that GATA-6, binding to a consensus binding site in the telokin promoter, inhibited promoter activity and repressed expression of endogenous telokin in smooth muscle cells (Figs. 1, 6, and 7). As GATA-6 is expressed at high levels in vascular smooth muscle cells as compared with intestinal smooth muscle cells and GATA-6 inhibits telokin expression, this may explain why telokin is expressed at higher levels in intestinal smooth muscle as compared with vascular smooth muscle cells (1). Overexpression of GATA-6 in A10 smooth muscle cells also increased expression of the 220-kDa MLCK and SM22␣ (Fig. 6). The observed induction of 220-kDa MLCK expression is interesting as the 220-kDa MLCK and telokin are independent products of the same MLYK gene. The data suggest that GATA-6 can independently regulate at least two of the distinct promoters within the MLYK gene. Although SM22␣ expression was increased following GATA-6 overexpression in A10 cells, we observed no direct effects of GATA-6 on the SM22␣ promoter in these cells (Fig. 1). This discrepancy could be explained if GATA-6 induced endogenous SM22␣ expression through an indirect pathway or if the GATA-6 response element is located outside of the promoter fragment used in our reporter gene assays. Conversely, although we observed GATA-6 activation of the SM ␣-actin promoter, we were not able to detect any changes in endogenous SM ␣-actin expression following overexpression of GATA-6 in A10 cells. This may reflect the already high levels of endogenous SM ␣-actin in these cells, as previous studies (32) have shown that overexpression of GATA-6 prevents the down-regulation of SM ␣-actin expression that occurs following vascular injury. Together these data would suggest that direct binding of GATA-6 to the SM ␣-actin promoter is required to maintain expression of SM ␣-actin in the normal vasculature.
The GATA-6-induced inhibition of telokin promoter activity and endogenous telokin expression was unexpected as GATA-6 is The region of myocardin that interacts with SRF (green box) is distinct from the region that interacts with GATA-6 (dark blue box) (20,26,41,43). B, model depicting GATA-6, SRF, and myocardin interactions on the smooth muscle-restricted telokin promoter. In this model GATA-6 bound to the GATA site (thick purple line), adjacent to a CArG box (thick yellow line), interacts with SRF bound to the CArG box and prevents myocardin from binding to SRF, thereby inhibiting promoter activity. Within each molecule the protein interaction domains are indicated as described in A. C, in promoters such as the Sm-MHC promoter, the location of a GATA site more distal to important CArG boxes may not allow bound GATA-6 to affect SRF-myocardin interactions. GATA-6 bound to the more distal elements can recruit coactivators such as p300 and further enhance promoter activity (40). GATA-6 may also directly interact with SRF-bound myocardin and further modulate transcriptional activity (41). usually a potent transcriptional activator. However, recent studies (41) on GATA-4 have also demonstrated that this GATA family member can either augment or repress activation of promoters for myocardin in a promoter-specific manner. These studies showed that GATA-4 can repress the myocardin-induced activation of the SM22␣ and atrial natriuretic factor promoters while augmenting the myocardin-induced activation of the Nkx2.5 promoter. In contrast, we observed no effect of GATA-6 on the basal activity of the SM22␣ promoter in A10 cells and an increase in expression of endogenous SM22␣ following GATA-6 infection of A10 cells. This may suggest that GATA-4 and GATA-6 have distinct roles in regulating SM22␣ promoter activity or the apparently disparate results may suggest that either myocardin plays little role in regulating SM22␣ promoter activity in A10 cells or that other proteins are modulating the activity of the GATA factors in a cell type-specific manner.
As both GATA-6 and myocardin have been shown previously to interact with the MADS domain of SRF, we examined the possibility that GATA-6 and myocardin may compete with each other for binding to SRF. As expected, our results show that both GATA-6 and myocardin bind to SRF, and the binding of SRF and myocardin can be disrupted by increasing concentrations of GATA-6 (Figs. 5B and 8A). Indeed, our results show that on the telokin promoter GATA-6 could serve as a competitor for myocardin. Experiments using in vitro glutathione Stransferase pull-down assays have suggested a direct binding of GATA-4 and myocardin (41). In contrast, although we were able to detect interactions between GATA-6 and SRF and between myocardin and SRF using a mammalian two-hybrid assay, we were not able to detect an interaction between GATA-6 and myocardin (Fig. 5). This discrepancy is not due to the different properties of GATA-6 and GATA-4 as we were also not able to detect GATA-4-myocardin interactions using the mammalian two-hybrid assay (data not shown). This more than likely suggests that the binding of GATA-6 and myocardin is of a lower affinity than the binding of GATA-6 and SRF or myocardin and SRF and is below the threshold of detection of the two-hybrid assay. Whether GATA-6 interacts with myocardin in vivo remains to be confirmed; however, it is well accepted that both GATA-6 and myocardin can interact with SRF in vivo, and our current data strongly suggest that GATA-6 can prevent the interaction of SRF and myocardin. As the same domain of SRF binds to GATA and myocardin (20,26), it is likely that a single SRF molecule is able to interact with either GATA or myocardin but not both molecules (Fig. 8A). Based upon our data and other recent studies (41), we propose that on the telokin promoter GATA-6 binds to a GATA element adjacent to the CArG box, interacts with SRF, and thereby prevents myocardin from interacting with SRF (Fig.  8B). It is possible that binding of GATA-6 to this site may increase the local effective concentration of GATA-6, allowing it to more efficiently compete with soluble myocardin for binding to SRF (Fig. 8B). In the absence of GATA-6 or on promoters in which there is no GATA site adjacent to the CArG box, myocardin is able to interact with SRF and stimulate promoter activity (Fig. 8C). In the Sm-MHC promoter, the GATA site is located distal to the CArG box, and this may account for the inability of GATA-6 binding to this site to interfere with myocardin binding to SR, thus allowing GATA-6 to act as an independent activator (Fig. 8C). In addition, similar to GATA-4 (41), GATA-6 may be able to interact directly with SRF-associated myocardin to further enhance or modulate promoter activity.
Finally, recent studies (42) have also shown that myocardin and Elk-1 compete for binding to a common docking site on SRF and that Elk-1 acts as a signal-responsive repressor of smooth muscle gene expression by displacing myocardin from SRF within the context of native chromatin. Myocardin, Elk-1, and GATA-6 all bind to the MADS domain of SRF, thus the MADS domain provides a cofactor switching platform that responds to different growth and developmental signals. This would suggest that the relative nuclear concentrations of SRF-associated proteins will have dramatic effects on the SRF complexes that form in a cell at a given time. As smooth muscle-restricted genes may be differentially responsive to activation by distinct SRF-protein complexes, the expression levels of these complexes may then play a significant role in determining the phenotype of a given smooth muscle cell.