Forkhead box protein A2 (FOXA2) protein stability and activity are regulated by sumoylation.

The forkhead box protein A2 (FOXA2) is an important regulator of glucose and lipid metabolism and organismal energy balance. Little is known about how FOXA2 protein expression and activity are regulated by post-translational modifications. We have identified that FOXA2 is post-translationally modified by covalent attachment of a small ubiquitin related modifier-1 (SUMO-1) and mapped the sumoylation site to the amino acid lysine 6 (K6). Preventing sumoylation by mutating the SUMO acceptor K6 to arginine resulted in downregulation of FOXA2 protein but not RNA expression in INS-1E insulinoma cells. K6R mutation also downregulated FOXA2 protein levels in HepG2 hepatocellular carcinoma cells, HCT116 colon cancer cells and LNCaP and DU145 prostate cancer cells. Further, interfering with FOXA2 sumoylation through siRNA mediated knockdown of UBC9, an essential SUMO E2 conjugase, resulted in downregulation of FOXA2 protein levels. Stability of sumoylation deficient FOXA2K6R mutant protein was restored when SUMO-1 was fused in-frame. FOXA2 sumoylation and FOXA2 protein levels were increased by PIAS1 SUMO ligase but not a SUMO ligase activity deficient PIAS1 mutant. Although expressed at lower levels, sumoylation deficient FOXA2K6R mutant protein was detectable in the nucleus indicating that FOXA2 nuclear localization is independent of sumoylation. Sumoylation increased the transcriptional activity of FOXA2 on Pdx-1 area I enhancer. Together, our results show that sumoylation regulates FOXA2 protein expression and activity.

PIAS1 is a GATA4 SUMO ligase that regulates GATA4-dependent intestinal promoters independent of SUMO ligase activity and GATA4 sumoylation.

GATA4 confers cell type-specific gene expression on genes expressed in cardiovascular, gastro-intestinal, endocrine and neuronal tissues by interacting with various ubiquitous and cell-type-restricted transcriptional regulators. By using yeast two-hybrid screening approach, we have identified PIAS1 as an intestine-expressed GATA4 interacting protein. The physical interaction between GATA4 and PIAS1 was confirmed in mammalian cells by coimmunoprecipitation and two-hybrid analysis. The interacting domains were mapped to the second zinc finger and the adjacent C-terminal basic region of GATA4 and the RING finger and the adjoining C-terminal 60 amino acids of PIAS1. PIAS1 and GATA4 synergistically activated IFABP and SI promoters but not LPH promoters suggesting that PIAS1 differentially activates GATA4 targeted promoters. In primary murine enterocytes PIAS1 was recruited to the GATA4-regulated IFABP promoter. PIAS1 promoted SUMO-1 modification of GATA4 on lysine 366. However, sumoylation was not required for the nuclear localization and stability of GATA4. Further, neither GATA4 sumoylation nor the SUMO ligase activity of PIAS1 was required for coactivation of IFABP promoter by GATA4 and PIAS1. Together, our results demonstrate that PIAS1 is a SUMO ligase for GATA4 that differentially regulates GATA4 transcriptional activity independent of SUMO ligase activity and GATA4 sumoylation.

Serum response factor expression is enriched in pancreatic ß cells and regulates insulin gene expression.

Serum response factor (SRF) is an essential regulator of myogenic and neurogenic genes and the ubiquitously expressed immediate-early genes. The purpose of this study is to determine SRF expression pattern in murine pancreas and examine the role of SRF in pancreatic gene expression. Immunohistochemical analysis of wild-type pancreas and LacZ staining of pancreas from SRF LacZ knock-in animals showed that SRF expression is restricted to ß cells. SRF bound to the rat insulin promoter II (RIP II) serum response element, an element conserved in both rat I and murine I and II insulin promoters. SRF activated RIP II, and SRF binding to RIP II and the exon 5-encoded 64-aa subdomain of SRF was required for this activation. Transient or stable knockdown of SRF leads to down-regulation of insulin gene expression, suggesting that SRF is required for insulin gene expression. Further, SRF physically interacted with the pancreas and duodenum homeobox-1 (Pdx-1) and synergistically activated RIP II. Elevated glucose concentration down-regulated SRF binding to RIP II SRE, and this down-regulation was associated with decreased RIP II activity and increased SRF phosphorylation on serine 103. Together, our results demonstrate that SRF is a glucose concentration-sensitive regulator of insulin gene expression.

GATA factors in gastrointestinal malignancy.

GATA factors are unique transcription factors with conserved DNA-binding domains. They serve diverse roles in embryogenesis, cell differentiation, regulation of tissue-specific genes, and carcinogenesis. The subfamily GATA-4, -5, and -6 are highly expressed in endoderm-derived organs and regulate multiple gut-specific genes. Multiple studies have analyzed the role of GATA factors in gastrointestinal (GI) malignancy, such as those of the stomach, pancreas, and colon, and premalignant lesions such as Barrett's esophagus. The GATA factors appear to have distinct roles in regulating key genes involved in GI malignancy. Understanding the precise role of GATA factors in malignancy may lead to the development of effective molecular targets for cancer therapy.

GATA6 promotes colon cancer cell invasion by regulating urokinase plasminogen activator gene expression.

GATA6 is a zinc finger transcription factor expressed in the colorectal epithelium. We have examined the expression of GATA6 in colon cancers and investigated the mechanisms by which GATA6 regulates colon cancer cell invasion. GATA6 was overexpressed in colorectal polyps and primary and metastatic tumors. GATA6 was strongly expressed in both the nuclear and cytoplasmic compartments of the colon cancer cells. GATA6 expression was upregulated in invasive HT29 and KM12L4 cells compared with the parental HT29 and KM12 cells and positively correlated with urokinase-type plasminogen activator (uPA) gene expression. Small interfering RNA (siRNA) knockdown of GATA6 resulted in reduced uPA gene expression and cell invasion. GATA6 bound to the uPA gene regulatory sequences in vivo and activated uPA promoter activity in vitro. uPA promoter deletion analysis indicated that the promoter proximal Sp1 sites were required for GATA6 activation of the uPA promoter. Accordingly, GATA6 physically associated with Sp1 and siRNA knockdown of Sp1 decreased GATA6 activation of the uPA promoter activity suggesting that Sp1 recruits GATA6 to the uPA promoter and mediates GATA6 induced activation of the uPA promoter activity. On the basis of our results, we conclude that GATA6 is an important regulator of uPA gene expression, and the dysregulated expression of GATA6 contributes to colorectal tumorigenesis and tumor invasion.

Histone deacetylase 7 and FoxA1 in estrogen-mediated repression of RPRM.

Activation of estrogen receptor alpha (ERalpha) results in both induction and repression of gene transcription; while mechanistic details of estrogen induction are well described, details of repression remain largely unknown. We characterized several ERalpha-repressed targets and examined in detail the mechanism for estrogen repression of Reprimo (RPRM), a cell cycle inhibitor. Estrogen repression of RPRM is rapid and robust and requires a tripartite interaction between ERalpha, histone deacetylase 7 (HDAC7), and FoxA1. HDAC7 is the critical HDAC needed for repression of RPRM; it can bind to ERalpha and represses ERalpha's transcriptional activity--this repression does not require HDAC7's deacetylase activity. We further show that the chromatin pioneer factor FoxA1, well known for its role in estrogen induction of genes, is recruited to the RPRM promoter, is necessary for repression of RPRM, and interacts with HDAC7. Like other FoxA1 recruitment sites, the RPRM promoter is characterized by H3K4me1/me2. Estrogen treatment causes decreases in H3K4me1/me2 and release of RNA polymerase II (Pol II) from the RPRM proximal promoter. Overall, these data implicate a novel role for HDAC7 and FoxA1 in estrogen repression of RPRM, a mechanism which could potentially be generalized to many more estrogen-repressed genes and hence be important in both normal physiology and pathological processes.

PDX-1 acts as a potential molecular target for treatment of human pancreatic cancer.

OBJECTIVES: The purpose of this study was to investigate whether pancreatic and duodenal homeobox factor 1 (PDX-1) could serve as a potential molecular target for the treatment of pancreatic cancer. METHODS: Cell proliferation, invasion capacity, and protein levels of cell cycle mediators were determined in human pancreatic cancer cells transfected with mouse PDX-1 (mPDX-1) alone or with mPDX-1 short hairpin RNA (shRNA) and/or human PDX-1 shRNA (huPDX-1 shRNA). Tumor cell growth and apoptosis were also evaluated in vivo in PANC-1 tumor-bearing severe combined immunodeficient mice receiving multiple treatments of intravenous liposomal huPDX-1 shRNA. RESULTS: mPDX-1 overexpression resulted in the significant increase of cell proliferation and invasion in MIA PaCa2, but not PANC-1 cells. This effect was blocked by knocking down mPDX-1 expression with mPDX-1 shRNA. Silencing of huPDX-1 expression in PANC-1 cells inhibited cell proliferation in vitro and suppressed tumor growth in vivo which was associated with increased tumor cell apoptosis. PDX-1 overexpression resulted in dysregulation of the cell cycle with up-regulation of cyclin D, cyclin E, and Cdk2 and down-regulation of p27. CONCLUSIONS: PDX-1 regulates cell proliferation and invasion in human pancreatic cancer cells. Down-regulation of PDX-1 expression inhibits pancreatic cancer cell growth in vitro and in vivo, implying its use as a potential therapeutic target for the treatment of pancreatic cancer.

PDX-1 expression is associated with islet proliferation in vitro and in vivo.

BACKGROUND: Transcription factor pancreatic duodenal homeobox-1 (PDX-1) is critical for beta-cell differentiation and insulin gene expression. In this study, we investigated the role of PDX-1 in ductal-to-islet cell transdifferentiation, islet cell apoptosis, and proliferation in addition to other regulators associated with these processes in two developing beta-cell models. MATERIALS AND METHODS: CAPAN-1 cells were cultured with the GLP-1 analogue Exendin-4 (Ex-4) to induce transdifferentiation to an insulin-producing phenotype. Expression patterns of PDX-1, somatostatin receptors (SSTR) 1, 2, and 5, p27, and p38 were analyzed. To model pancreatic regeneration in vivo, subtotal pancreatectomies were performed in rats and remnant pancreata were compared to sham laparotomy controls to determine islet size, morphology, apoptosis, and PDX-1 expression. RESULTS: In Ex-4-treated cells, PDX-1 expression increased 67% above basal levels within 24 h and was followed by a 10-fold decline in expression by the end of the study. Expression of cell-cycle inhibitor p27 was down-regulated by 81% at 24 h, while levels of the pro-apoptotic modulator p38 significantly increased 4-fold. When compared to controls, SSTR1 expression declined, while SSTR2 and SSTR5 expression were significantly up-regulated in treated cells. Immunofluorescence of pancreatic remnants following subtotal pancreatectomy revealed increased PDX-1 staining at 24 h followed by a significant decline at 72 h post-pancreatectomy. CONCLUSION: GLP-1 analogue Ex-4 resulted in up-regulation of PDX-1 in CAPAN-1 cells and PDX-1 was up-regulated in proliferating islets following subtotal pancreatectomy in rats. The increase was seen in the first 24 h. These findings suggest a possible relationship between PDX-1 and the state of islet proliferation, islet-to-ductal transdifferentiation, apoptosis, and the expression of SSTRs.

Cooperation between GATA4 and TGF-beta signaling regulates intestinal epithelial gene expression.

Members of the transforming growth factor-beta (TGF-beta) family have been shown to play an important role in the regulation of gut epithelial gene expression. We have used the intestinal alkaline phosphatase (IAP) and intestinal fatty acid binding protein (IFABP) promoters to dissect the mechanisms by which TGF-beta1 signaling regulates gut epithelial gene expression. TGF-beta signaling alone was not sufficient for activation of IAP and IFABP promoters. However, TGF-beta signaling cooperated with the gut epithelial transcription factor GATA4 to synergistically activate IAP and IFABP promoters. Coexpression of GATA4 along with the TGF-beta1 signal transducing downstream effectors such as Smad2, 3, and 4 resulted in synergistic activation of both IAP and IFABP promoters. This synergistic activation was reduced by simultaneous expression of dominant-negative Smad4. -40 and -89 GATA binding sites in the IFABP promoter were required for the synergistic activation by Smad2, 3, and 4 and GATA4. GATA4 and Smad2, 3, and 4 physically associated with each other and this interaction was mediated through the MH2 domain of Smad2, 3, and 4 and the second zinc finger and the COOH-terminal basic domain of GATA4. The COOH-terminal activation domain and the Smad-interacting second zinc finger domain of GATA4 were required for the synergistic activation of the IFABP promoter. Naturally occurring oncogenic mutations within the GATA4-interacting MH2 domain of Smad2 reduced the coactivation of IFABP promoter by Smad2 and GATA4. Our results suggest that the TGF-beta signaling regulates gut epithelial gene expression by targeting GATA4.

LIM-only protein, CRP2, switched on smooth muscle gene activity in adult cardiac myocytes.

Smooth muscle alpha-actin gene activity appears in promyocardial cells well before cardiac myocyte differentiation and is down-regulated during the onset of rhythmic contractility and cardiac morphogenesis. The levels of LIM-only CRP2 correlated well with smooth muscle gene activity. Cardiomyocyte-specific expression of CRP2 in transgenic mice showed robust expression of smooth muscle cell-specific transcripts and protein filaments in the adult heart. Protein transduction of a recombinant CRP2 protein, fused to the protein transduction domain of HIV, into neonatal heart cells induced de novo synthesis of smooth muscle cell-specific transcripts and proteins. The LIM zinc fingers in CRP2 were found to collaborate with Brg1 of the SNF/SWI complexes, recruited serum response factor, and remodeled smooth muscle target gene chromatin through histone acetylation. CRP2 may have a cytoskeletal role, but as a nuclear protein, CRP2 acted as a potent transcription coadaptor that remodeled silent cardiac myocyte chromatin and directed serum response factor-dependent smooth muscle gene activity.

Conditional mutagenesis of the murine serum response factor gene blocks cardiogenesis and the transcription of downstream gene targets.

Serum response factor (SRF) homozygous-null embryos from our backcross of SRF(LacZ/)(+) "knock-in" mice failed to gastrulate and form mesoderm, similar to the findings of an earlier study (Arsenian, S., Weinhold, B., Oelgeschlager, M., Ruther, U., and Nordheim, A. (1998) EMBO J. 17, 6289-6299). Our use of embryonic stem cells provided a model system that could be used to investigate the specification of multiple embryonic lineages, including cardiac myocytes. We observed the absence of myogenic alpha-actins, SM22alpha, and myocardin expression and the failure to form beating cardiac myocytes in aggregated SRF null embryonic stem cells, whereas the appearance of transcription factors Nkx2-5 and GATA4 were unaffected. To study the role of SRF during heart organogenesis, we then performed cardiac-specific ablation of SRF by crossing the transgenic alpha-myosin heavy chain Cre recombinase line with SRF LoxP-engineered mice. Cardiac-specific ablation of SRF resulted in embryonic lethality due to cardiac insufficiency during chamber maturation. Conditional ablation of SRF also reduced cell survival concomitant with increased apoptosis and reduced cellularity. Significant reductions in SRF (> or =95%), atrial naturetic factor (> or =80%), and cardiac (> or =60%), skeletal (> or =90%), and smooth muscle (> or =75%) alpha-actin transcripts were also observed in the cardiac-conditional knock-out heart. This was consistent with the idea that SRF directs de novo cardiac and smooth muscle gene activities. Finally, quantitation of the knock-in LacZ reporter gene transcripts in the hearts of cardiac-conditional knock-out embryos revealed an approximately 30% reduction in gene activity, indicating SRF gene autoregulation during cardiogenesis.

Serum response factor, an enriched cardiac mesoderm obligatory factor, is a downstream gene target for Tbx genes.

We tested the idea that T-box factors direct serum response factor (SRF) gene activity early in development. Analysis of SRF-LacZ "knock-in" mice showed highly restricted expression in early embryonic cardiac and skeletal muscle mesoderm and neuroectoderm. Examination of the SRF gene for regulatory regions by linking the promoter and 5'-flanking sequences, up to 5.5 kb, failed to target LacZ transgene activity to the heart and the tail pre-somitic mesenchyme. However, linkage of a minimal SRF promoter with the SRF 3'-untranslated region (UTR), inundated with multimeric T-box binding sites (TBEs), restored robust reporter gene activity to embryonic heart and tail. Finer dissection of the 3'-UTR to a small cluster of TBEs also stimulated transgene activity in the cardiac forming region and the tail, however, when the TBEs contained within these DNA sequences were mutated, preventing Tbx binding, transgene activity was lost. Tbx2, Tbx5, and the cardiac-enriched MYST family histone acetyltransferase TIP60, were observed to be mutual interactive cofactors through the TIP60 zinc finger and the T-box of the Tbx factors. In SRF-null ES cells, TIP60, Tbx2, and Tbx5 were sufficient to stimulate co-transfected SRF reporter activity, however this activity required the presence of the SRF 3'-UTR. SRF gene transactivation was blocked by two distinct TIP60 mutants, in which either the histone acetyltransferase domain was inactivated or the Zn finger-protein binding domain was excised. Our study supports the idea that SRF embryonic cardiac gene expression is dependent upon the SRF 3'-UTR enhancer, Tbx2, Tbx5, and TIP60 histone acetyltransferase activity.

Recruitment of the androgen receptor via serum response factor facilitates expression of a myogenic gene.

Androgen receptor (AR) induced precocious myogenesis in culture and myogenic specified gene activity. Increased levels of AR expression in replicating C2C12 myoblasts stimulated fusion into post-differentiated multinucleated myotubes and the appearance of skeletal alpha-actin transcripts, even in the absence of ligand. Furthermore, AR activated the skeletal alpha-actin promoter, which lacks GRE sites, in co-transfected C2C12 cells. AR co-activation of the skeletal alpha-actin promoter required co-expressed full-length serum response factor (SRF). In vitro, AR associated with SRF and was recruited by SRF to a alpha-actin promoter SRF binding site. Our data suggest that AR is capable of activating myogenic genes devoid of consensus AR binding sites via its recruitment by the myogenic enriched transcription factor, SRF.

Serum response factor is alternatively spliced in human colon cancer.

BACKGROUND: Serum response factor (SRF) is a transcription factor that plays an important role in cellular differentiation and cell cycle regulation. SRF function is regulated in part by alternative splicing. Little is known about the expression or role of these alternatively spliced isoforms during tumorigenesis. We hypothesized that there is a change in the splice variants during intestinal tumorigenesis and that this change promotes the tumor phenotype. MATERIALS AND METHODS: SRF expression was determined by Western blotting of benign intestinal cells and human colon cancer cell lines. To determine the effect of alternative splicing of SRF on intestinal growth and proliferation, the predominant alternatively spliced isoform of SRF that we identified in colon cancer cells, SRFDelta5, was transfected into IEC-6 cells. IEC-6 and IEC-6SRFDelta5 cells were plated and cell numbers were determined at four time points. RESULTS: Western blotting demonstrates that full-length SRF is the predominant form of SRF in rat IEC-6 cells, normal human colonic mucosa, and HT-29 cells, derived from a well-differentiated human colonic adenocarcinoma. In the colon cancer cell lines derived from poorly differentiated tumors (WiDr, HCT 116, LoVo, and SW480), SRFDelta5 is the predominant isoform expressed. There was a significant increase in cell survival in IEC-6 cells transfected with SRFDelta5 compared to parental cells. CONCLUSION: Our data demonstrate that an alternatively spliced isoform of SRF, SRFDelta5, is expressed in human colon cancer cell lines. Additionally, these data demonstrate that expression of SRFDelta5 may contribute to the tumor phenotype by affecting cell survival. This is the first study to document a change in expression of the alternatively spliced isoform of SRF in human malignancy.

Cysteine-rich LIM-only proteins CRP1 and CRP2 are potent smooth muscle differentiation cofactors.

Cysteine-rich LIM-only proteins, CRP1 and CRP2, expressed during cardiovascular development act as bridging molecules that associate with serum response factor and GATA proteins. SRF-CRP-GATA complexes strongly activated smooth muscle gene targets. CRP2 was found in the nucleus during early stages of coronary smooth muscle differentiation from proepicardial cells. A dominant-negative CRP2 mutant blocked proepicardial cells from differentiating into smooth muscle cells. Together with SRF and GATA proteins, CRP1 and CRP2 converted pluripotent 10T1/2 fibroblasts into smooth muscle cells, while muscle LIM protein CRP3 inhibited the conversion. Thus, LIM-only proteins of the CRP family play important roles in organizing multiprotein complexes, both in the cytoplasm, where they participate in cytoskeletal remodeling, and in the nucleus, where they strongly facilitate smooth muscle differentiation.

Developmental expression of serum response factor in the rat central nervous system.

Serum response factor (SRF), a transcription factor known to be essential for early embryonic development as well as post-natal regulation of both cellular proliferation and myogenic differentiation, is expressed broadly in neurons within the adult mammalian central nervous system (CNS). The function of SRF within the developing CNS is not well established, but it is likely to play an important role in neuraxial development and neuronal function, since many of its known target genes (e.g., c-fos) and transcriptional partners (e.g., Elk-1) are also highly expressed in neurons. Immunohistochemical survey of the post-natal developing rat brain revealed a progressive increase in SRF immunoreactivity in neurons of the cerebral and cerebellar cortices, and in selective subcortical regions from birth (P0) through post-natal day 28 (P28). SRF immunoreactivity stabilized from P28 into adulthood. A few loci, such as the nucleus of cranial nerve VII, showed the reverse expression pattern (strong immunoreactivity at P0-P7, declining by P28). The developmental expression pattern of SRF overlaps significantly with that of myotonic dystrophy protein kinase, a potential upstream regulator, and of the LIM-only genes Lmo1, Lmo2 and Lmo3, whose products belong to a family of proteins known to be strong positive regulators of SRF's transcriptional activity. These data suggest that SRF has a significant function in the early post-natal development of the CNS.

Transforming growth factor-beta induction of smooth muscle cell phenotpye requires transcriptional and post-transcriptional control of serum response factor.

Transforming growth factor-beta induces a smooth muscle cell phenotype in undifferentiated mesenchymal cells. To elucidate the mechanism(s) of this phenotypic induction, we focused on the molecular regulation of smooth muscle-gamma-actin, whose expression is induced at late stages of smooth muscle differentiation and developmentally restricted to this lineage. Transforming growth factor-beta induced smooth muscle-gamma-actin protein, cytoskeletal localization, and mRNA expression in mesenchymal cells. Smooth muscle-gamma-actin promoter-luciferase reporter activity was enhanced by transforming growth factor-beta, and deletion analysis revealed that CArG box 2 in the promoter was necessary for this transcriptional activation. CArG motifs bind transcriptional activator serum response factor; gel shift analyses revealed increased binding of serum response factor-containing complexes to this site in response to transforming growth factor-beta, paralleled by increased serum response factor protein expression. Serum response factor expression was found to be up-regulated by transforming growth factor-beta via transcriptional activation of the gene and post-transcriptional regulation. Using mesenchymal cells stably transfected with wild type or dominant-negative serum response factor, we demonstrated that its expression is sufficient for induction of a smooth muscle phenotype in mesenchymal cells and is necessary for transforming growth factor-beta-mediated smooth muscle induction.