Ets domain transcription factor PE1 suppresses human interstitial collagenase promoter activity by antagonizing protein-DNA interactions at a critical AP1 element.

In MC3T3E1 calvarial osteoblasts, fibroblast growth factor receptor (FGFR) signaling elicits multiple transcriptional responses, including upregulation of the interstitial collagenase/matrix metalloproteinase 1 (MMP1) promoter. FGF responsiveness maps to a bipartite Ets/AP1 element at base pairs -123 to -61 in the human MMP1 promoter. Under basal conditions, the MMP1 promoter is repressed in part via protein-DNA interactions at the Ets cognate, and minimally two mechanisms convey MMP1 promoter upregulation by FGF2: (a) transcriptional activation via Fra1/c-Jun containing DNA-protein interactions at the AP1 cognate and (b) derepression of promoter activity regulated by the Ets cognate. To identify osteoblast Ets repressors that potentially participate in gene expression in the osteoblast, we performed reverse transcription-polymerase chain reaction (RT-PCR) analysis of mRNA isolated from MC3T3E1 cells, using degenerative amplimers to the conserved Ets DNA binding domain to survey the Ets genes expressed by these cells. Six distinct Ets mRNAs were identified: Ets2, Fli1, GABPalpha, SAP1, Elk1, and PE1. Of these, only PE1 has extensive homology to the known Ras-regulated Ets transcriptional repressor, ERF. Therefore, we cloned and characterized PE1 cDNA from a mouse brain library and performed functional analysis of this particular Ets family member. A 2 kb transcript was isolated from brain that encodes a approximately 57 kDa protein; the predicted protein contains the known N-terminal Ets domain of PE1 and a novel C-terminal domain with signficant homology to murine ERF. The murine PE1 open reading frame (ORF) is much larger than the previously reported human PE1 ORF. Consistent with this, affinity-purified rabbit anti-mouse PE1 antibody specifically recognizes an approximately 66 kDa protein present only in the nuclear fraction of MC3T3E1 osteoblasts. Recombinant PE1 binds authentic AGGAWG Ets DNA cognates, and transient transfection studies demonstrate that PE1 represses MMP1 promoter activity. Surprisingly, although deletion of the MMP1 Ets cognate at nucleotides -88 to -83 abrogates FGF2 induction, it does not prevent suppression of the AP1-dependent MMP1 promoter by PE1. PE1 regulation maps to the MMP1 promoter region -75 to -61, suggesting that PE1 suppresses transcription via protein-protein interactions with AP1. Consistent with this, recombinant GST-PE1 specifically inhibits the formation of protein-DNA interactions on the MMP1 AP1 site (-72 to -66) when present in an admixture with MC3T3E1 crude nuclear extract. In toto, these data indicate that PE1 participates in the transcriptional regulation of the MMP1 promoter in osteoblasts. As observed with other transcriptional repressors of MMP1 gene expression, transcriptional suppression by PE1 occurs via inhibition of AP1-dependent promoter activity.

The RRM domain of MINT, a novel Msx2 binding protein, recognizes and regulates the rat osteocalcin promoter.

Msx2 is a homeodomain transcriptional repressor that exerts tissue-specific actions during craniofacial skeletal and neural development. To identify coregulatory molecules that participate in transcriptional repression by Msx2, we applied a Farwestern expression cloning strategy to identify transcripts encoding proteins that bind Msx2. A lambdagt11 expression library from mouse brain was screened with radiolabeled GST-Msx2 fusion protein encompassing the core suppressor domain of Msx2. A cDNA was isolated that encodes a novel protein fragment that binds radiolabeled Msx2. Homeoprotein binding activity was confirmed by Farwestern analysis of the T7-epitope-tagged recombinant protein fragment, and interactions in vitro require Msx2 residues necessary for transcriptional suppression in vivo. On the basis of biochemical analyses, this novel protein was named MINT, an acronym for Msx2-interacting nuclear target protein. The original clone is part of a 12.6 kb transcript expressed at high levels in testis and at lower levels in calvarial osteoblasts and brain. Multiple clones isolated from a mouse testis library were sequenced to construct a MINT cDNA contig of 11 kb. Starting from an initiator Met in good Kozak context, a large nascent polypeptide of 3576 amino acids is predicted, in contiguous open reading frame with the Msx2 interaction domain residues 2070-2394. Protein sequence analysis reveals that MINT has three N-terminal RNA recognition motifs (RRMs) and four nuclear localization signals. Western blot analysis of fractionated cell extracts reveals that mature approximately 110 kDa (N-terminal) and approximately 250 kDa (C-terminal) MINT protein fragments accumulate in chromatin and nuclear matrix fractions, cosegregating with Msx2 and topoisomerase II. In gel shift assays, the MINT RRM domain selectively binds T- and G-rich DNA sequences; this includes a large G/T-rich inverted repeat element present in the proximal rat osteocalcin (OC) promoter, overlapping three cognates that support OC expression in osteoblasts. MINT and OC mRNAs are reciprocally regulated during differentiation of MC3T3E1 calvarial osteoblasts. Consistent with its proposed role as a nuclear transcriptional factor, transient expression of MINT(1-812) suppresses the FGF/forskolin-activated OC promoter, does not significantly regulate CMV promoter activity, but markedly upregulates the HSV thymidine kinase promoter in MC3T3E1 cells. In toto, these data indicate that the novel nuclear protein MINT binds the homeoprotein Msx2 and coregulates OC during craniofacial development. Msx2 and MINT both target an information-dense, osteoblast-specific regulatory region of the OC proximal promoter, nucleotides -141 to -111. The N-terminal MINT RRM domain represents an authentic dsDNA binding module for this novel vertebrate nuclear matrix protein. Acting as a scaffold protein, MINT potentially exerts both positive and negative regulatory actions by organizing transcriptional complexes in the nuclear matrix.

Reciprocal regulation of osteocalcin transcription by the homeodomain proteins Msx2 and Dlx5.

Osteocalcin (OC) is a small calcium binding protein expressed in bones and teeth undergoing mineralization. OC expression in calvarial osteoblasts and odontoblasts is regulated in part via protein-protein interactions between the homeodomain repressor, Msx2, and components of the cell type-specific basal OC promoter. Recent work suggests that homeodomain proteins form heterodimers that confer transcriptional regulation. Since the homeodomain proteins Dlx5 and Msx2 are both expressed by primary rat calvarial osteoblasts, we examined whether Msx2 and Dlx5 functionally interact to regulate the OC promoter. In both primary rat calvarial and MC3T3E1 mouse calvarial osteoblasts, transient expression of Dlx5 only mildly augments basal OC promoter (luciferase reporter) activity, while Msx2 suppresses transcriptional activity by ca. 80%. However, Dlx5 completely reverses Msx2 repression of the OC promoter. Structure-function analyses using far-Western blot and transient cotransfection assays reveal that (i) Msx2 and Dlx5 can form dimers, (ii) Dlx5 residues 127-143 are necessary for dimerization and to reverse Msx2-dependent OC repression, and (iii) intrinsic DNA binding activity of Dlx5 is not required for OC regulation. Msx2 inhibits the DNA binding activity of a third complex, the OC fibroblast growth factor response element binding protein (OCFREB), that supports activity of the basal OC promoter. Dlx5 completely abrogates Msx2 suppression of OCFREB DNA binding activity, and residues required for Dlx5 transcriptional de-repression in vivo are also required for reversing inhibition of OCFREB binding in vitro. Finally, Dlx5 reverses Msx2 inhibition of OC promoter activation by FGF2/forskolin. Thus, Dlx5 regulates the expression of the OC promoter in calvarial osteoblasts in part by de-repression, antagonizing Msx2 repression of transcription factors that support basal OC promoter activity.

Stimulus-selective inhibition of rat osteocalcin promoter induction and protein-DNA interactions by the homeodomain repressor Msx2.

Osteocalcin (OC) is a matrix calcium-binding protein expressed in osteoblasts and odontoblasts undergoing mineralization. OC expression is up-regulated in part by signals initiated by basic fibroblast growth factor (FGF2), cyclic AMP or forskolin (FSK), and calcitriol via defined elements and DNA-protein interactions in the OC promoter. We identified the OC gene as a target for transcriptional suppression by Msx2, a homeodomain transcription factor that controls ossification in the developing skull. In this study, we examine the effects of Msx2 expression on OC promoter activation (luciferase reporter) by FGF2/FSK and calcitriol in MC3T3-E1 osteoblasts. Expression of Msx2 decreases basal activity of the 1-kilobase (-1050 to +32) rat OC promoter by 80%; however, the promoter is still inducible 3-fold by calcitriol. By contrast, OC promoter induction by FGF2/FSK is completely abrogated by Msx2. Because intrinsic Msx2 DNA binding activity is not required for the Msx2 suppressor function, we assessed whether Msx2 represses OC activation by regulating DNA-protein interactions at the FGF2 response element (OCFRE) and compared these interactions with those occurring at the calcitriol response element (VDRE). Treatment of MC3T3-E1 cells with FGF2/FSK or calcitriol up-regulates specific DNA-protein interactions at the OCFRE or VDRE, respectively, as detected by gel shift assay. Preincubation of crude nuclear extracts with recombinant glutathione S-transferase (GST)-Msx2 dose-dependently inhibits OCFRE DNA binding activity, whereas GST has no effect. Msx2 itself does not bind the OCFRE. Residues 132-148 required for Msx2 core suppressor function in transfection assays are also required to inhibit OCFRE DNA binding activity. By contrast, GST-Msx2 has no effect on calcitriol-regulated DNA-protein interactions at the VDRE. Using gel shift as an assay, the OCFRE DNA-binding protein OCFREB was purified to about 50% homogeneity from MG63 osteosarcoma cells. Recombinant Msx2 inhibits purified OCFREB DNA binding activity, whereas the Msx2 variant lacking residues 132-148 is inactive. Thus, Msx2 abrogates up-regulation of the OC promoter by FGF2/FSK in part by inhibiting OCFREB binding to the OCFRE.

Structure-function analysis of Msx2-mediated transcriptional suppression.

Osteocalcin (OC) is a calcium binding protein expressed in mature osteoblasts undergoing mineralization. The OC gene has been identified as a target for transcriptional suppression by Msx2, a homeodomain transcription factor that controls ossification in calvarial bone of the developing skull. We have initiated systematic structure-function analyses of Msx2, using OC promoter suppression (luciferase reporter) in MC3T3-E1 calvarial osteoblasts as an assay. Msx2 variants were epitope ("FLAG")-tagged for monitoring Msx2 protein expression by Western blot analysis. Functional analyses of N- and C-terminally truncated molecules identify Msx2 residues 97-208 as the core suppressor domain. Internal deletion analyses indicate that suppressor function is dependent upon structural features encoded by residues 132-148--upstream of the homeodomain and overlapping the homeodomain N-terminal extension--but not upon residues in the three homeodomain helices. Mutations that enhance DNA binding activity do not proportionally enhance Msx2 suppressor function; moreover, a Msx2 point mutant Msx2(T147A) that completely lacks DNA binding activity is indistinguishable from wild-type Msx2 in its ability to suppress the OC promoter, demonstrating that direct interaction with DNA is not required for Msx2 suppressor function. This suggests that Msx2 suppresses transcription via protein-protein interactions with components of the basal transcriptional machinery, either alone or in concert with co-regulators. Using interaction "Far Western" blotting assays, we systematically tested for protein-protein interactions between Msx2 and components of the basal transcriptional machinery known to mediate transcriptional activation (TBP, TFIIB, and TFIIF). Msx2 binds both components of TFIIF (RAP74, RAP30), but not TFIIB or TBP. Msx2(55-208) encompasses core suppressor domain residues and binds TFIIF; in this context, deletion of the seventeen amino acid residues 132-148 that are required for core suppressor function abrogates interactions with TFIIF components. Co-expression of RAP74 in MC3T3-E1 cells partially reverses (>50%) suppression of OC promoter activity by Msx2, while co-expression of TFIIB or RAP30 has no effect. Thus the core suppressor domain of Msx2 participates in functionally important interactions with RAP74 that regulate OC promoter activity in calvarial osteoblasts.

Fibroblast growth factor receptor signaling activates the human interstitial collagenase promoter via the bipartite Ets-AP1 element.

Interstitial collagenases participate in the remodeling of skeletal matrix and are regulated by fibroblast growth factor (FGF). A 0.2-kb fragment of the proximal human interstitial collagenase [matrix metalloproteinase (MMP1)] promoter conveys 4- to 8-fold induction of a luciferase reporter in response to FGF2 in MC3T3-E1 osteoblasts. By 5'-deletion, this response maps to nucleotides -100 to -50 relative to the transcription initiation site. The 63- bp MMP1 promoter fragment -123 to -61 confers this FGF2 response on the rous sarcoma virus minimal promoter. Intact Ets and AP1 cognates in this element are both required for responsiveness. The AP1 site supports basal and FGF-inducible promoter activity. The intact Ets cognate represses basal transcriptional activity in both heterologous and native promoter contexts and is also required for FGF activation. FGF2 up-regulates a DNA-binding activity that recognizes the MMP1 AP1 cognate and contains immunoreactive Fra1 and c-Jun. Both constitutive and FGF-inducible DNA-binding activities are present in MC3T3-E1 cells that recognize the MMP1 Ets cognate; prototypic Ets transcriptional activators are not present in these complexes. Inhibitors of protein kinase C, phosphatidyl inositol 3-OH kinase, and calmodulin-dependent protein kinase do not attenuate MMP1 promoter activation. FGF2 activates ERK1/ERK2 signaling in osteoblasts; however, 25 microM MAPK-ERK kinase (MEK) inhibitor PD98059 (inhibits by > 85% the phosphorylation of ERK1/ERK2) has no effect on MMP1 promoter activation by FGF2. Ligand-activated and constitutively active FGF receptors initiate MMP1 induction. Dominant negative Ras abrogates MMP1 induction by constitutively active FGFR2-ROS, but dominant negative Rho and Rac do not inhibit induction. The mitogen-activated protein kinase (MAPK) phosphatase MKP2 [inactivates extracellular regulated kinase (ERK) = Jun N-terminal kinase (JNK) > p38 MAPK] completely abrogates MMP1 activation, whereas PAC1 (inactivates ERK = p38 > JNK) attenuates but does not completely prevent induction. Thus, a Ras- and MKP2-regulated MAPK pathway, independent of ERK1/ERK2 MAPK activity, mediates FGF2 transcriptional activation of MMP1 in MC3T3-E1 osteoblasts, converging upon the bipartite Ets-AP1 element. The DNA-protein interactions and signal cascades mediating FGF induction of the MMP1 promoter are distinct from two other recently described FGF response elements: the MMP1 promoter (-123 to -61) represents a third FGF-activated transcriptional unit.

The rat osteocalcin fibroblast growth factor (FGF)-responsive element: an okadaic acid-sensitive, FGF-selective transcriptional response motif.

We recently identified a bipartite element in the rat osteocalcin (OC) promoter that confers synergistic induction by fibroblast growth factor receptor 2 (FGF2) and cAMP. A GCAGTCA motif (OCFRE) at -146 to -138 in the OC promoter is necessary for synergy and participates in a FGF2-regulated DNA-protein interaction. We have isolated the FGF-regulated component of this transcriptional response for detailed study. Two or three copies of the OC promoter fragment -154 to -113 with the intact OCFRE confer 10- or 30-fold FGF2-inductive responses, respectively, on the unresponsive basal promoter 92 OCLUC (luciferase reporter) in MC3T3-E1 cells; a single copy is insufficient. As in the native context, induction depends upon an intact OCFRE motif (mutant GCATTTA motifs unresponsive). FGF receptor 1 can mediate activation; expression of this receptor in L6 cells (no endogenous FGF receptors) permits FGF2 induction of (OCFRE)2 92 OCLUC. FGF2 induction of (OCFRE)2 92 OCLUC in MC3T3-E1 cells is not recapitulated by platelet-derived growth factor-BB, epidermal growth factor, insulin-like growth factor I, or transforming growth factor-beta (< 10% the activity of FGF2). OCFRE activation is not inhibited by kinase inhibitors H-89, wortmannin, staurosporine, KN-62, or H-7. However, the phosphoprotein phosphatase inhibitors okadaic acid (OKA), calyculin A, and vanadate decrease FGF induction of (OCFRE)2 92 OCLUC or (OCFRE)3 92 OCLUC, without inhibiting induction of the interstitial collagenase promoter. OKA and calyculin A do not decrease OCFRE DNA-protein interactions, suggesting that important protein-protein interactions are phosphatase regulated. These data provide evidence that; 1) FGF receptors elaborate transcriptional activation signals that functionally differ from those of other receptor tyrosine kinases; 2) an OKA-sensitive phosphatase participates in FGF receptor-dependent activation of the OCFRE; and 3) two transcriptional activation signals are initiated by FGF receptor activation in MC3T3-E1 cells, reflected in the divergent sensitivities of OCFRE and interstitial collagenase promoter induction to OKA and vanadate.

Cell-cycle-dependent modulation of EGF-receptor-mediated signaling.

In A431 cells synchronized by treatment with thymidine, the level of EGF-stimulated tyrosine protein kinase activity in cells in S and G2/M phases was reduced approximately 40% relative to that seen in cells in G1. This decrease in receptor tyrosine protein kinase activity did not correlate with a decrease in cell surface EGF receptor expression, indicating that the reduced activity could not be attributed to receptor loss. EGF-stimulated PI 3-kinase activity was also reduced by approximately 60% during S phase as compared to G1 phase. The change was not due to decreased PI 3-kinase expression since Western blot analyses indicated that cellular p85 levels remained constant throughout the cell cycle. These data suggest that the ability of EGF to stimulate biological responses varies during the cell cycle and implicate cell-cycle-dependent processes in the regulation of EGF-receptor-mediated signaling.

Mutation of proline-1003 to glycine in the epidermal growth factor (EGF) receptor enhances responsiveness to EGF.

We have shown previously that the epidermal growth factor (EGF) receptor is phosphorylated at Ser-1002 and that this phosphorylation is associated with desensitization of the EGF receptor. Ser-1002 is followed immediately by Pro-1003, a residue that may promote the adoption of a specific conformation at this site or severe as a recognition element for the interaction of the EGF receptor with other proteins. To examine these possibilities, we have mutated Pro-1003 of the EGF receptor to a Gly residue and have analyzed the effect of this mutation on EGF-stimulated signaling. Cells expressing the P1003G EGF receptors exhibited higher EGF-stimulated autophosphorylation and synthetic peptide phosphorylation compared to cells expressing wild-type EGF receptors. In addition, the ability of EGF to stimulate PI 3-kinase activity and mitogen-activated protein kinase activity was enhanced in cells expressing the P1003G EGF receptor. Cells expressing P1003G receptors also demonstrated an increased ability to form colonies in soft agar in response to EGF. These results indicate that mutation of Pro-1003 leads to a potentiation of the biological effects of EGF. The findings are consistent with the hypothesis that Pro-1003 plays a role in a form of regulation that normally suppresses EGF receptor function.