Id Proteins Negatively Regulate Basic Helix-Loop-Helix Transcription Factor Function by Disrupting Subnuclear Compartmentalization
2003; Elsevier BV; Volume: 278; Issue: 46 Linguagem: Inglês
10.1074/jbc.m306056200
ISSN1083-351X
AutoresPeter O’Toole, Toshiaki Inoue, Lindsay J. Emerson, Ian Morrison, Alan R. Mackie, Richard J. Cherry, John D. Norton,
Tópico(s)Ubiquitin and proteasome pathways
ResumoId helix-loop-helix (HLH) proteins act as global regulators of metazoan cell fate, cell growth, and differentiation. They heterodimerize with and inhibit the DNA-binding function of members of the basic helix-loop-helix (bHLH) family of transcription factors. Using real time fluorescence microscopy techniques in single living cells, we show here that nuclear pools of chromatin-associated bHLH transcription factor are freely exchangeable and in constant flux. The existence of a dynamic equilibrium between DNA-bound and free bHLH protein is also directly demonstrable in vitro. By contrast, Id protein is not associated with any subcellular, macromolecular structures and displays a more highly mobile, diffuse nuclear-cytoplasmic distribution. When co-expressed with antagonist Id protein, the chromatin-associated sublocalization of bHLH protein is abolished, and there is an accompanying 100-fold increase in its nuclear mobility to a level expected for freely diffusible Id-bHLH heterodimer. These results suggest that nuclear Id protein acts by sequestering pools of transiently diffusing bHLH protein to prevent reassociation with chromatin domains. Such a mechanism would explain how Id proteins are able to overcome the large DNA-binding free energy of bHLH proteins that is necessary to accomplish their inhibitory effect. Id helix-loop-helix (HLH) proteins act as global regulators of metazoan cell fate, cell growth, and differentiation. They heterodimerize with and inhibit the DNA-binding function of members of the basic helix-loop-helix (bHLH) family of transcription factors. Using real time fluorescence microscopy techniques in single living cells, we show here that nuclear pools of chromatin-associated bHLH transcription factor are freely exchangeable and in constant flux. The existence of a dynamic equilibrium between DNA-bound and free bHLH protein is also directly demonstrable in vitro. By contrast, Id protein is not associated with any subcellular, macromolecular structures and displays a more highly mobile, diffuse nuclear-cytoplasmic distribution. When co-expressed with antagonist Id protein, the chromatin-associated sublocalization of bHLH protein is abolished, and there is an accompanying 100-fold increase in its nuclear mobility to a level expected for freely diffusible Id-bHLH heterodimer. These results suggest that nuclear Id protein acts by sequestering pools of transiently diffusing bHLH protein to prevent reassociation with chromatin domains. Such a mechanism would explain how Id proteins are able to overcome the large DNA-binding free energy of bHLH proteins that is necessary to accomplish their inhibitory effect. Id proteins function as global regulators of cell fate determination. They play a pivotal role in the coordinate regulation of gene expression during cell growth/cell cycle control, differentiation, and tumorigenesis (reviewed in Refs. 1Norton J.D. J. Cell Sci. 2000; 113: 3897-3905Crossref PubMed Google Scholar and 2Benezra R. Oncogene. 2001; 20: 8288-8289Crossref Google Scholar). Recent studies have also highlighted their role in cellular senescence (3Alani R.M. Young A.Z. Shifflet C.B. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 7812-7816Crossref PubMed Scopus (184) Google Scholar, 4Ohtani N. Zebedee Z. Huot T.J.G. Stinson J.A. Sugimoto M. Ohashi Y. Sharrocks A.D. Peters G. Hara E. Nature. 2001; 409: 1067-1070Crossref PubMed Scopus (524) Google Scholar) and in cell fate decisions in cells of specialized lineages such as lymphocytes (5Sugai M. Gonda H. Kusunoki T. Katakai T. Yokota Y. Shimizu A. Nat. Immunol. 2003; 4: 25-30Crossref PubMed Scopus (108) Google Scholar) (reviewed in Ref. 6Engel I. Murre C. Nat. Rev. Immunol. 2001; 1: 193-199Crossref PubMed Scopus (160) Google Scholar), vascular endothelial cells (7Lyden D. Young Z.A. Zagzag D. Yan W. Gerald W. O'Reilly R. Bader B.L. Hynes R.O. Zhaung Y. Manova K. Benezra R. Nature. 1999; 401: 670-677Crossref PubMed Scopus (769) Google Scholar, 8Volpert O.V. Pili R. Sikder H.A. Nelius T. Zaichuk T. Morris C. Shiflett C.B. Devlin M.K. Conant K. Alani R.M. Cancer Cell. 2002; 2: 473-483Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar), and neuronal cells (7Lyden D. Young Z.A. Zagzag D. Yan W. Gerald W. O'Reilly R. Bader B.L. Hynes R.O. Zhaung Y. Manova K. Benezra R. Nature. 1999; 401: 670-677Crossref PubMed Scopus (769) Google Scholar, 9Martinsen B.J. Bonner-Fraser M. Science. 1998; 281: 988-991Crossref PubMed Scopus (129) Google Scholar). The four members of the Id protein family (Id1-Id4) function by directly associating with and modulating the activity of several families of transcriptional regulators (1Norton J.D. J. Cell Sci. 2000; 113: 3897-3905Crossref PubMed Google Scholar, 10Lasorella A. Noseda M. Beyna M. Yokota Y. Iavarone A. Nature. 2000; 407: 592-598Crossref PubMed Scopus (430) Google Scholar, 11Yates P.R. Atherton G.T. Deed R.W. Norton J.D. Sharrocks A.D. EMBO J. 1999; 18: 968-976Crossref PubMed Scopus (128) Google Scholar, 12Roberts C.E. Deed R.W. Inoue T. Norton J.D. Sharrocks A.D. Mol. Cell Biol. 2001; 21: 524-533Crossref PubMed Scopus (118) Google Scholar). However, compelling biochemical and genetic data implicate members of the ubiquitously expressed Class A, basic helix-loop-helix (bHLH) 1The abbreviations used are: bHLH, basic helix-loop-helix; HLH, helix-loop-helix; FRAP, fluorescence recovery after photobleaching; FLIP, fluorescence loss in photobleaching; FRET, fluorescence resonance energy transfer; YFP, yellow fluorescent protein; EYFP, enhanced yellow fluorescent protein; ECFP, enhanced cyan fluorescent protein; FP, fluorescent protein; ROI, region of interest; Pipes, 1,4-piperazinediethanesulfonic acid.1The abbreviations used are: bHLH, basic helix-loop-helix; HLH, helix-loop-helix; FRAP, fluorescence recovery after photobleaching; FLIP, fluorescence loss in photobleaching; FRET, fluorescence resonance energy transfer; YFP, yellow fluorescent protein; EYFP, enhanced yellow fluorescent protein; ECFP, enhanced cyan fluorescent protein; FP, fluorescent protein; ROI, region of interest; Pipes, 1,4-piperazinediethanesulfonic acid. family of "E proteins" as the most important heterodimerization targets for Id proteins in the coordinate regulation of gene expression during cell fate determination (1Norton J.D. J. Cell Sci. 2000; 113: 3897-3905Crossref PubMed Google Scholar, 2Benezra R. Oncogene. 2001; 20: 8288-8289Crossref Google Scholar, 13Benezra R. Davis R. Lockshon D. Turner D. Weintraub H. Cell. 1990; 61: 49-59Abstract Full Text PDF PubMed Scopus (1785) Google Scholar, 14Massari M. Murre C. Mol. Cell Biol. 2000; 20: 429-440Crossref PubMed Scopus (1352) Google Scholar). In mammals, there are three E protein family genes, E2A (encoding three alternatively spliced variants, E12, E47, and E2-5), E2-2 (ITF2), and HEB (14Massari M. Murre C. Mol. Cell Biol. 2000; 20: 429-440Crossref PubMed Scopus (1352) Google Scholar). The E proteins bind to a consensus "E-box" recognition sequence, present in the transcriptional control regions of numerous cellular genes, either as a homodimer or, more commonly, as a heterodimeric partner with a member of the much larger family of tissue-specific, Class B bHLH proteins. Id proteins lack a basic, DNA-binding domain, and they heterodimerize avidly (via their HLH domain) with bHLH E proteins (13Benezra R. Davis R. Lockshon D. Turner D. Weintraub H. Cell. 1990; 61: 49-59Abstract Full Text PDF PubMed Scopus (1785) Google Scholar) to prevent the latter from binding to DNA (1Norton J.D. J. Cell Sci. 2000; 113: 3897-3905Crossref PubMed Google Scholar, 14Massari M. Murre C. Mol. Cell Biol. 2000; 20: 429-440Crossref PubMed Scopus (1352) Google Scholar). Since E proteins are obligate heterodimerization partners for tissue-specific bHLH proteins, this provides a common mechanism of negative regulation of transcription underlying cell fate decisions in multiple lineages (2Benezra R. Oncogene. 2001; 20: 8288-8289Crossref Google Scholar). Id proteins readily associate with and inhibit the DNA binding function of bHLH proteins in vitro and will inhibit their E-box-dependent gene regulatory and biological functions in vivo (1Norton J.D. J. Cell Sci. 2000; 113: 3897-3905Crossref PubMed Google Scholar, 2Benezra R. Oncogene. 2001; 20: 8288-8289Crossref Google Scholar). However, essentially nothing is known about how the opposing functions of Id proteins and their heterodimerization target bHLH E proteins are integrated at the level of the subcellular architecture. In particular, it is unclear precisely how, in a physiological context, Id proteins are able to overcome the large DNA-binding free energy of bHLH-bHLH homo-/heterodimers in order to sequester them as Id-bHLH heterodimers. To address these issues, we have used fluorescence microscopy techniques (fluorescence recovery after photo-bleaching (FRAP), fluorescence loss in photobleaching (FLIP), and fluorescence resonance energy transfer (FRET)) (15Lippincott-Schwartz J. Snapp E. Kenworthy A. Nat. Rev. Mol. Cell Biol. 2001; 2: 444-456Crossref PubMed Scopus (962) Google Scholar) with constructs expressing proteins tagged with either enhanced yellow fluorescent protein (EYFP) or enhanced cyan fluorescent protein (ECFP) to study "real time" diffusion dynamics and heterodimeric interactions of Id and bHLH protein in single living cells. Our results show that bHLH protein is localized within distinct chromatin-associated foci, where it exists in dynamic equilibrium between the DNA-bound and non-DNA-bound state. Id proteins disrupt this chromatin compartmentalization of bHLH proteins. This suggests that Id proteins act by sequestering pools of transiently diffusing nucleoplasmic bHLH protein and preventing reassociation with chromatin domains. This would provide a mechanism through which Id proteins could overcome the large DNA-binding free energy of bHLH proteins without stabilization of Id-bHLH heterodimers by higher order oligomeric structure and thereby function as potent transcription factor antagonists. DNA Constructs—cDNA constructs encoding full-length Id3 and E47 proteins (inserted in the vector, pcDNA3; Invitrogen) have been described previously (16Wilson J.W. Deed R.W. Inoue T. Balzi M. Becciolini A. Faraoni P. Potten C.S. Norton J.D. Cancer Res. 2001; 61: 8803-8810PubMed Google Scholar, 17Deed R.W. Armitage S. Norton J.D. J. Biol. Chem. 1996; 271: 23603-23606Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar). Constructs encoding fusions of Id3 and E47 with the enhanced green fluorescence protein variants, EYFP and ECFP, were generated by joining the coding regions at the N and C termini of the fluorescent protein moiety, in frame, onto the C- or N-terminal coding region of the Id3 or E47 genes. A mutant version of E47 that is defective in DNA binding (E47-EYFP-bm) was generated by mutating six amino acid residues in the basic, DNA-binding domain of full-length E47 as described previously (16Wilson J.W. Deed R.W. Inoue T. Balzi M. Becciolini A. Faraoni P. Potten C.S. Norton J.D. Cancer Res. 2001; 61: 8803-8810PubMed Google Scholar). The ΔN-E47 construct was generating by deleting the coding sequence from residue 1 to 475, leaving the DNA-binding and HLH regions intact. To construct the E-box-Luc reporter plasmid (pT81-4xE), synthetic double-stranded oligonucleotides containing four tandem repeats of the E-box sequence (upper strand, 5′-CAACACCTGCTGCCTCCCAACACCTGCTGCCTCCCAACACCTGCTGCCTCCCAACACCTGCC-3′) were ligated to the BamHI/XhoI sites upstream of the thymidine kinase minimal promoter of the pT81 plasmid (12Roberts C.E. Deed R.W. Inoue T. Norton J.D. Sharrocks A.D. Mol. Cell Biol. 2001; 21: 524-533Crossref PubMed Scopus (118) Google Scholar). DNA construction and PCR mutagenesis was performed by standard recombinant DNA procedures. Cell Culture and DNA Transfection—Human embryonic kidney epithelial cells (293) were cultured in Dulbecco's modified Eagle's medium, supplemented with 10% fetal calf serum either on glass cover slips, in plastic flasks, or in 12-well plates. DNA-mediated transfection was performed by the standard calcium phosphate method as described previously (16Wilson J.W. Deed R.W. Inoue T. Balzi M. Becciolini A. Faraoni P. Potten C.S. Norton J.D. Cancer Res. 2001; 61: 8803-8810PubMed Google Scholar). For reporter gene assays, cells were transfected in 12-well plates with a total of 6.0 μg of DNA. DNA inputs were normalized with the addition of empty vector, pcDNA3. Cells were harvested 48 h after transfection, and preparation of extracts and the luciferase assay were performed using the dual luciferase reporter assay (Promega), according to the manufacturer's instructions. For normalization of transfection efficiencies, Renilla luciferase expression plasmid (pRL-TK; Promega) was included in the transfections. In some experiments, cells were lysed in situ on glass cover slips (18Mancini M.G. Liu B. Sharp Z.D. Mancini M.A. J. Cell. Biochem. 1999; 72: 322-338Crossref PubMed Scopus (46) Google Scholar) using CSK buffer (10 mm Pipes, 300 mm sucrose, 100 mm NaCl, 3 mm MgCl2, 0.5% Triton X-100, Roche Applied Science complete protease inhibitor mixture, pH 6.8). For immunofluorescence detection of E47, cells were fixed in 4% paraformaldehyde and immunostained using anti-E47 antibody (sc 349; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) followed by detection with fluorescein isothiocyanate-labeled secondary antibody (F0205; DAKO) (17Deed R.W. Armitage S. Norton J.D. J. Biol. Chem. 1996; 271: 23603-23606Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar). Fluorescence Microscopy—Confocal microscopy was then performed using a Bio-Rad Radiance 2000 attached to an Olympus IX70 microscope fitted with a PlanApo ×60 oil immersion lens (numerical aperture = 1.4). EYFP was excited with the 514-nm argon line, and emission was detected via a 545/20 filter. ECFP was excited with the 457-nm argon line, and emission was detected via a 488/10 filter. Images were acquired sequentially to prevent bleed-through of the two probes. EYFP-tagged proteins were used for FRAP measurements. A single image z section was taken before a short bleach pulse (typically 0.5-1 s) in a region of interest (ROI), and then typically 60 images were acquired at either 1- or 2-s time intervals. Bleaching was performed using the 514-nm laser line, set at 100% laser power, and subsequent images were collected using no more than 4% laser power. Nontransfected cells and fixed transfected cells were used for control experiments. Pixels in the ROI were analyzed to determine mean and S.E. of the fluorescence signal at each time point; these data were then fitted by a single exponential to obtain the time constant (τ). Data was analyzed using in-house software. Routinely, data from 20-30 cells were collected for each experiment. FRAP was also performed by conventional fluorescence microscopy using a Physitemp model TS-4 microscope, as described previously but with the mechanical beam splitter replaced by an electro-optic modulator (19Ladha S. Mackie A.R. Harvey L.J. Clark D.C. Lea E.J.A. Brullemans M. Duclohier H. Biophys. J. 1996; 71: 1364-1373Abstract Full Text PDF PubMed Scopus (109) Google Scholar). The 514-nm argon laser line was used to excite EYFP, and emission was detected via a 550LP dichroic filter. FRAP measurements were recorded and analyzed as previously described (19Ladha S. Mackie A.R. Harvey L.J. Clark D.C. Lea E.J.A. Brullemans M. Duclohier H. Biophys. J. 1996; 71: 1364-1373Abstract Full Text PDF PubMed Scopus (109) Google Scholar), and fast diffusion data were also fitted by a single exponential in the same way as for confocal FRAP data. FRET was measured by a variant of the confocal FRAP method (20Day R.N. Periasamy A. Schaufele F. Methods. 2001; 25: 4-18Crossref PubMed Scopus (183) Google Scholar, 21Miyawaki A. Tsien R.Y. Methods Enzymol. 2000; 327: 472-500Crossref PubMed Scopus (376) Google Scholar). Cells expressing Id3-EYFP and E47-ECFP were initially imaged sequentially by 457- and 514-nm excitation, and YFP fluorescence in a ROI was then bleached by several scans at maximum 514-nm laser power. The ECFP fluorescence postbleach image was then obtained. Pixels in areas showing ECFP enhancement were analyzed to obtain the mean signals pre- and postbleach (excluding areas of high and low signal from "hot spots" and nucleoli). The value postbleach/prebleach of these signals provides the FRET ratio. Control areas in the same images quantify the reduction of the ECFP signal caused by the 457-nm laser. Electrophoretic Mobility Band Shift Assay—Electrophoretic mobility shift assays were performed using lysates from transfected cells essentially as described previously (22Inoue T. Kamiyama J. Sakai T. J. Biol. Chem. 1999; 274: 32309-32317Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar). Briefly, 2 μg of cell lysate was mixed with labeled probe (∼0.5 ng; 10,000 cpm) in a 25-μl reaction volume containing 20 mm Tris-HCl (pH 8.0), 100 mm NaCl, 10% glycerol, 1 μg of poly(dI-dC) (Amersham Biosciences). After a 20-min incubation, competitor lysate (2 μg) was added and the binding reaction terminated at time intervals by loading of the mix onto a 4% nondenaturing polyacrylamide gel with immediate electrophoresis. The μE5 E-box double-stranded oligonucleotide probe (GGCCAGAACACCTGCAGACG-3′) (23Jacobs Y. Xin X.Q. Dorshkind K. Nelson C. Mol. Cell Biol. 1994; 14: 4087-4096Crossref PubMed Scopus (19) Google Scholar) was end-labeled using the MEGALABEL DNA 5′-end labeling kit (Takara, Tokyo). Quantitation of band intensities was performed using a BAS2000 image analysis system (Fuji Film). Fluorescence-tagged E47 and Id3 Proteins Retain Their Gene-regulatory Functions—The bHLH E protein, E47, and its potent Id antagonist, Id3, have been widely studied at a molecular and cellular level as a model for Id-bHLH interactions (1Norton J.D. J. Cell Sci. 2000; 113: 3897-3905Crossref PubMed Google Scholar). Id3 will also inhibit gene regulatory programs and biological responses of E proteins functioning in heterodimeric mode with Class B bHLH proteins, exemplified by MyoD (24Atherton G. Travers H. Deed R. Norton J.D. Cell Growth Differ. 1996; 7: 1059-1066PubMed Google Scholar). Fig. 1A depicts the structures of fluorescent proteins generated by fusing the coding regions of the EYFP and ECFP genes onto the N- and C-terminal residue of E47 and onto the C-terminal residue of Id3. Since the size of the fluorescent protein (FP) tags was large, compared with the size of native proteins, particularly for Id3 (119 amino acid residues for Id3 versus 240 residues for EYFP/ECFP; see Fig. 1A), we tested fusion constructs for functional activity. In in vitro band shift assays, both N- and C-terminal fusions of native E47 bound to an E-box probe, whereas, as expected (16Wilson J.W. Deed R.W. Inoue T. Balzi M. Becciolini A. Faraoni P. Potten C.S. Norton J.D. Cancer Res. 2001; 61: 8803-8810PubMed Google Scholar), the E47 mutants in which the basic DNA-binding domain was disrupted (FP-E47bm and E47bm-FP in Fig. 1) did not. Fluorescently tagged Id3 also inhibited E-box probe binding of both wild type and FP fusions of E47, with efficiency comparable with that of wild type Id3 (data not shown). In transfected cells, expression of fluorescently tagged Id3 elicited a dose-dependent inhibition of E47-mediated transactivation of an E-box reporter gene construct (Fig. 1B). As shown for E47-EYFP in Fig. 1C, FP-tagged E47 was only slightly impaired in its ability to transactivate the E-box reporter (Fig. 1, compare B and C), and this activity could be competed using either wild type or FP-Id3 with comparable efficiency (Fig. 1C). In the following experiments, we used fusions of E47 in which the FP tag was located at the C terminus of the protein. Subnuclear Localization of E47—By virtue of the nuclear localization signal sequence (NLS in Fig. 1A), bHLH proteins, exemplified by E47, are localized to the cell nucleus. E47 is known to be associated with a variety of transcriptional coactivators within macromolecular complexes in chromatin (see Ref. 25Bradney C. Hjelmeland M. Komatsu Y. Yoshida M. Yao T-P. Zhuang Y. J. Biol. Chem. 2003; 278: 2370-2376Abstract Full Text Full Text PDF PubMed Scopus (60) Google Scholar and references therein). We initially examined the subnuclear distribution of endogenous E47 protein by immunofluorescence staining in fixed cells using confocal microscopy. As shown in Fig. 2 (middle panel), endogenous E47 was found to be localized to numerous discrete foci, superimposed on a diffuse background throughout the nucleoplasm, but largely excluded from the nucleolar regions. This nuclear "speckled" distribution pattern is characteristic of that described for several other transcription factors (18Mancini M.G. Liu B. Sharp Z.D. Mancini M.A. J. Cell. Biochem. 1999; 72: 322-338Crossref PubMed Scopus (46) Google Scholar, 26Grande M.A. van der Kraan I. de Jong L. van Driel R. J. Cell Sci. 1997; 110: 1781-1791Crossref PubMed Google Scholar, 27McNally J.G. Muller W.G. Walker D. Wolford R. Hager G.L. Science. 2000; 287: 1262-1265Crossref PubMed Scopus (637) Google Scholar, 28Kazansky A.V. Katotyanski E.B. Wyszomierski S.L. Mancini M.A. Rosen J.M. J. Biol. Chem. 1999; 274: 22484-22492Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 29Choi J.Y. Pratap J. Javed A. Zaidi S.K. Xing L.P. Balint E. Dalamangas S. Boyce B. van Wijnene A.J. Lian J.B. Stein J.L. J ones S.N. Stein G.S. Proc. Natl. Acad. Sci, U. S. A. 2001; 98: 8650-8655Crossref PubMed Scopus (240) Google Scholar). An identical subnuclear distribution was seen in live cells expressing E47 protein, tagged with a fluorescent EYFP (E47-EYFP) or ECFP (E47-ECFP) moiety either as a homodimer (Fig. 2, third panel) or as a heterodimer with co-expressed Class B bHLH protein, MyoD (data not shown). Nuclear E47-ECFP did not co-localize with EGFP-tagged splicing factor, ASF/SF2 (30Phair R.D. Misteli T. Nature. 2000; 404: 604-609Crossref PubMed Scopus (944) Google Scholar) or with chromatin-associated BRG1 (31Khavari P.A. Peterson C.L. Tamkun J.W. Mendel D.B. Crabtree G.R. Nature. 1993; 366: 170-174Crossref PubMed Scopus (531) Google Scholar), but, as with other transcription factors (26Grande M.A. van der Kraan I. de Jong L. van Driel R. J. Cell Sci. 1997; 110: 1781-1791Crossref PubMed Google Scholar), it displayed a minor co-localization (2-5%) with nascent transcriptional complexes, as detected by immunostaining of incorporated bromouridine triphosphate (data not shown). Rapid Exchange between Nuclear Pools of E47—To investigate whether nuclear pools of E47 localized in discrete subnuclear foci were exchangeable, the diffusion dynamics of E47-EYFP was studied in live cells using FRAP and FLIP techniques. In these and subsequent fluorescence dynamics experiments, transfected cells expressing a range of fluorescence intensities from dim (nearly physiological) to bright were analyzed. In all cases, the diffusion dynamics were not affected by expression level of exogenous protein. Fig. 3A shows the results of a typical FRAP experiment. Following photobleaching of a small ROI in the nucleus, there was a rapid fluorescence recovery (t = 10 ± 4 s) commensurate with the kinetics of fluorescence loss (t = 9 ± 3 s) from unbleached areas (Fig. 3, A and B). However, the localization of the E47-associated subnuclear structures themselves remained unchanged and exhibited only small displacements when observed over several minutes (Fig. 3A). Control cells within the same image frame showed no changes in fluorescence during the experiment (data not shown), demonstrating that the decrease was not due to bleaching during image acquisition. Repetitive bleaching of the ROI resulted in a complete loss of fluorescence (Fig. 3A , last panel), indicating the absence of any significant immobile fraction of E47-EYFP. Thus, FRAP and FLIP analysis show that essentially all of the nuclear pool of bHLH protein is freely exchangeable and in constant flux over a time scale of tens of seconds. This implies the existence of a dynamic equilibrium between the DNA-bound and -unbound nuclear pools of E47 protein. Since assembly of a functional bHLH homo-/heterodimer complex with DNA requires a functional DNA-binding domain, we also examined the subnuclear distribution and diffusion dynamics of the E47-EYFP-bm mutant (Fig. 1A) in which the DNA-binding domain is rendered nonfunctional. As shown by the example in Fig. 3C, the E47-EYFP-bm mutant displayed a distinctive pattern of subnuclear compartmentalization composed of much larger spherical structures than seen with wild type protein. Moreover, FRAP analysis showed that, in contrast to wild type protein, the DNA-binding mutant was completely immobile over a period of at least 5 min (Fig. 3C). Thus, although DNA binding function is not an absolute requirement for association with chromatin, it is required for correct subnuclear compartmentalization of the bHLH protein. Dynamic Equilibrium between DNA-bound and Free bHLH Protein Demonstrable in Vitro—Since the above diffusion dynamics data in intact live cells implied the existence of a dynamic equilibrium between the DNA-bound and -unbound nuclear pools of E47 protein, we sought to determine whether this was an inherent characteristic of the interaction between bHLH E protein and its cognate E-box-containing DNA that could be recapitulated in a cell-free system. As shown in Fig. 4, we employed a "competition" band shift assay under conditions of limited μE5 E-box probe input between two bHLH E proteins (wild type E47 and ΔN-E47), whose complexes with E-box DNA probe could be distinguished by electrophoretic mobility. The truncated, ΔN-E47 lacks most of the N-terminal region but still retains a DNA-binding domain. The addition of ΔN-E47 to preformed wild type E47-DNA complex led to the gradual appearance of ΔN-E47 complex with a corresponding loss of wild type E47-DNA complex (Fig. 4A). The time constant for the rate-limiting dissociation (off rate) of wild type E47-DNA complex in this experiment was (t = 40 s). Wild type E47 was similarly able to exchange with preformed ΔN-E47-DNA complex, with comparable kinetics (Fig. 4B). Thus, in an in vitro cell-free system, in the absence of any nuclear compartmentalization, a dynamic equilibrium between DNA-bound and free bHLH protein is clearly demonstrable. Id Protein Disrupts the Subnuclear Architecture of bHLH Protein and Dramatically Affects Its Nuclear Dynamics Properties—Previous immunofluorescence studies using fixed cells have shown that Id proteins, in contrast to bHLH proteins such as E47 that are localized to the nucleus, are distributed throughout the nucleus and cytoplasm; co-expression of these proteins leads to nuclear sequestration of Id protein, presumably as an Id-bHLH heterodimer (17Deed R.W. Armitage S. Norton J.D. J. Biol. Chem. 1996; 271: 23603-23606Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar). These findings were confirmed in live cells in which fluorescence-tagged E47 (E47-EYFP) and Id (Id3-ECFP) proteins were co-expressed (Fig. 5A). Significantly, in the presence of co-expressed Id protein, the subnuclear structural compartmentalization of E47 was completely abolished; the E47 acquired a diffuse distribution pattern throughout the nucleus, indistinguishable from that of nuclear Id protein alone (Fig. 5A). Co-expression of wild type Id3 protein also disrupted the subnuclear compartmentalization of E47-EYFP, whereas a mutant of Id3 that does not interact with bHLH protein because of a proline insertion in the H1 region of the HLH domain (17Deed R.W. Armitage S. Norton J.D. J. Biol. Chem. 1996; 271: 23603-23606Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar) had no effect (data not shown). We used FRET analysis to show that this disruption of nuclear compartmentalization of E47 in live cells was associated with direct physical interaction between the fluorescence-tagged E47 and Id protein (Fig. 5, B and C). As with other nuclear proteins that are physically associated with chromatin/nuclear matrix structures (18Mancini M.G. Liu B. Sharp Z.D. Mancini M.A. J. Cell. Biochem. 1999; 72: 322-338Crossref PubMed Scopus (46) Google Scholar, 27McNally J.G. Muller W.G. Walker D. Wolford R. Hager G.L. Science. 2000; 287: 1262-1265Crossref PubMed Scopus (637) Google Scholar), E47 (E47-EYFP) was retained in cells for several minutes following disruption with mild nonionic detergent, Triton X-100 (Fig. 6A , top panel). The addition of DNase I led to a rapid loss (within 2 min) of E47-EYFP protein from cells, consistent with a chromatin association of E47. However, in contrast to E47-EYFP alone, E47-EYFP/Id3-ECFP heterodimer was rapidly lost following detergent lysis, with kinetics indistinguishable from Id3-EYFP (or Id3-ECFP; data not shown) (see Fig. 6A). Confocal FRAP measurements revealed that the recovery time of E47-EYFP as a heterodimer with Id3-ECFP occurred too rapidly for accurate measurement. By using conventional FRAP, we determined that the recovery of the heterodimer is 100-fold faster than E47-EYFP protein alone (Fig. 6B). Similar results were obtained in experiments using the E47-EYFP-bm mutant (data not shown). The t for Id3-EYFP was 0.10 ± 0.04 s, and t for Id3-ECFP in the presence of E47-EYFP was 0.13 ± 0.04 s. A similar value was obtained for E47-EYFP in the presence of Id3-ECFP (t = 0.14 ± 0.03 s). The theoretical ratio of t values for an Id3-ECFP/E47-EYFP heterodimer to an Id3-ECFP monomer is calculated to be 1.50, assuming free diffusion and spherical structures. The difference in diffusion between Id3-EYFP and Id3-EYFP/E47-ECFP (Fig. 6B) is therefore consistent with the formation of a heterodimer, with no evidence for the existence of a significant contribution from more slowly diffusing species that would correspond to higher order oligomers. Accumulating data highlight the importance of the dynamic behavior of individual components of the transcriptional apparatus within a highly organized subnuclear architecture in the regulation of transcription (30Phair R.D. Misteli T. Nature. 2000; 404: 604-609Crossref PubMed Scopus (944) Google Scholar, 32Misteli T. Science. 2001; 291: 843-847Crossref PubMed Scopus (524) Google Scholar, 33Carmo-Fonseca M. Cell. 2002; 108: 513-521Abstract Full Text Full Text PDF PubMed Scopus (162) Google Scholar, 34Chubb J.R. Bickmore W.A. Cell. 2003; 112: 403-406Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar). The functional importance of subnuclear localization of transcription factors to discrete foci in chromatin has been widely reported (18Mancini M.G. Liu B. Sharp Z.D. Mancini M.A. J. Cell. Biochem. 1999; 72: 322-338Crossref PubMed Scopus (46) Google Scholar, 26Grande M.A. van der Kraan I. de Jong L. van Driel R. J. Cell Sci. 1997; 110: 1781-1791Crossref PubMed Google Scholar, 28Kazansky A.V. Katotyanski E.B. Wyszomierski S.L. Mancini M.A. Rosen J.M. J. Biol. Chem. 1999; 274: 22484-22492Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 29Choi J.Y. Pratap J. Javed A. Zaidi S.K. Xing L.P. Balint E. Dalamangas S. Boyce B. van Wijnene A.J. Lian J.B. Stein J.L. J ones S.N. Stein G.S. Proc. Natl. Acad. Sci, U. S. A. 2001; 98: 8650-8655Crossref PubMed Scopus (240) Google Scholar), although studies on subnuclear diffusion dynamics have largely focused on the family of ligand-dependent nuclear hormone receptors. For example, the glucocorticoid receptor (27McNally J.G. Muller W.G. Walker D. Wolford R. Hager G.L. Science. 2000; 287: 1262-1265Crossref PubMed Scopus (637) Google Scholar, 35Becker M. Baumann C. John S. Walker D.A. Vigneron M. McNally J.G. Hager G.L. EMBO Rep. 2002; 3: 118-1194Crossref Scopus (173) Google Scholar, 36Schaaf M.J.M. Cidlowski J.A. Mol. Cell Biol. 2003; 23: 1922-1934Crossref PubMed Scopus (145) Google Scholar), thyroid hormone receptor (37Baumann T. Maruvada P. Hager G.L. Yen P.M. J. Biol. Chem. 2001; 276: 11237-11245Abstract Full Text Full Text PDF PubMed Scopus (120) Google Scholar), dioxin receptor (38Elbi C. Misteli T. Hager G.L. Mol. Biol. Cell. 2002; 13: 2001-2015Crossref PubMed Scopus (64) Google Scholar), and proliferation-activated receptor (39Akiyama T.E. Baumann C.T. Sakai S. Hager G.L. Gonzalez F.J. Mol. Endocrinol. 2002; 16: 707-721Crossref PubMed Scopus (50) Google Scholar) reportedly undergo ligand-dependent intranuclear redistribution into discrete foci representing transcriptional complexes. The pools of nuclear receptor in these foci undergo rapid (within seconds) exchange with a more highly mobile fraction in the nucleoplasm. In the present study, we found that the nuclear pools of the prototype bHLH E protein, E47, display a comparable pattern of spatiotemporal dynamics, with no significant immobile fraction. As the third largest family of transcription factors in eukaryotes (40Tupler R. Perini G. Green M.R. Nature. 2001; 409: 832-833Crossref PubMed Scopus (299) Google Scholar), with over 200 distinct members (14Massari M. Murre C. Mol. Cell Biol. 2000; 20: 429-440Crossref PubMed Scopus (1352) Google Scholar), bHLH proteins display considerable molecular "promiscuity" in their heterodimeric interactions, and this affords a high level of combinatorial diversity for fine tuning of gene expression programs in cell fate determination and differentiation (14Massari M. Murre C. Mol. Cell Biol. 2000; 20: 429-440Crossref PubMed Scopus (1352) Google Scholar). The rapid exchange of E47 between chromatin domains is therefore particularly noteworthy, since this would facilitate a commensurately rapid change in bHLH dimer composition on multiple gene regulatory sequences in a coordinated fashion. Indeed, consistent with in vitro DNA-binding kinetic studies (41Spinner D.S. Liu S.H. Wang S.W. Schmidt J. J. Mol. Biol. 2002; 317: 431-445Crossref PubMed Scopus (19) Google Scholar), we observed a dynamic equilibrium between distinct bHLH E proteins and the formation of protein-E-box DNA complexes in a cell-free system. However, because of macromolecular crowding effects (42Ellis R.J. Trends Biochem. Sci. 2001; 26: 597-604Abstract Full Text Full Text PDF PubMed Scopus (1669) Google Scholar) within the confines of the chromatin architecture, the kinetics of bHLH protein-E-box DNA interaction in living cells is likely to be quite different from that observed in vitro. Under conditions in which heterodimeric interaction between Id and bHLH protein was directly demonstrable in live cells by FRET analysis, our studies showed that the subnuclear architecture and diffusion dynamics of bHLH protein were dramatically altered by association with Id protein. When co-expressed with antagonist Id protein, the chromatin-associated sublocalization of bHLH protein was abolished, and there was an accompanying 100-fold increase in its nuclear mobility demonstrable by FRAP as the bHLH protein acquired the diffuse nuclear distribution pattern of Id protein. bHLH-bHLH dimer interactions are strongly stabilized as a ternary complex by DNA binding (43Ellenberger T. Fass D. Arnaud M. Harrison S.C. Genes Dev. 1994; 8: 970-980Crossref PubMed Scopus (349) Google Scholar, 44Wendt H. Thomas R.M. Ellenberger T. J. Biol. Chem. 1998; 273: 5735-5743Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar, 45Maleki S.J. Royer C.A. Hurlburt B.K. Biochemistry. 2002; 41: 10888-10894Crossref PubMed Scopus (13) Google Scholar). In order to efficiently sequester bHLH proteins as Id-bHLH heterodimers, Id proteins must be able to overcome the large DNA-binding free energy of bHLH proteins. Based on kinetic reassociation studies employing synthetic peptides in vitro (which show that Id-bHLH complexes are stabilized by adopting a higher order oligomeric state), it has been proposed that Id-bHLH heterodimer formation is favored through adoption of higher order structures (46Fairman R. Beran-Steed R.K. Anthony-Cahill S.J. Lear J.D. Stafford W.F. DeGrado W.F Benfield P.A. Brenner S.L. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 10429-10433Crossref PubMed Scopus (93) Google Scholar). It is therefore significant that we did not observe any evidence for a higher order oligomeric state of Id-bHLH heterodimers in live cells that would explain the ability of Id proteins to overcome the large DNA-binding free energy of bHLH proteins, as observed in vitro. Because Id proteins lack a basic, DNA-binding domain, they are widely accepted to function as dominant negative antagonists of bHLH proteins through a mechanism involving inhibition of DNA-binding activity (through the formation of Id-bHLH heterodimers) (1Norton J.D. J. Cell Sci. 2000; 113: 3897-3905Crossref PubMed Google Scholar, 2Benezra R. Oncogene. 2001; 20: 8288-8289Crossref Google Scholar, 13Benezra R. Davis R. Lockshon D. Turner D. Weintraub H. Cell. 1990; 61: 49-59Abstract Full Text PDF PubMed Scopus (1785) Google Scholar, 14Massari M. Murre C. Mol. Cell Biol. 2000; 20: 429-440Crossref PubMed Scopus (1352) Google Scholar). However, our studies of Id/bHLH proteins in live cells show that inhibition of DNA binding per se does not fully account for the functional attributes of Id proteins. In common with wild type E47, a DNA binding-defective E47 mutant (that still localized to chromatin, albeit as an immobile pool in larger subnuclear structures) could also be rendered freely diffusible by co-expression with Id protein. The observation that Id protein disrupts the chromatin-associated, subnuclear compartmentalization of bHLH proteins represents a novel facet of Id protein function that might also explain the ability of these proteins to overcome the high DNA-binding free energy of bHLH proteins without invoking stabilization of Id-bHLH heterodimers by higher order structure. By sequestering transiently diffusing nucleoplasmic bHLH protein into Id-bHLH heterodimers that are evidently excluded from chromatin structures, Id proteins would be able to prevent DNA binding and thereby accomplish a rapid (within seconds) inhibition of bHLH transcription factor activity. It would be of particular interest in future studies to investigate the kinetics of this process, directly in vivo and/or by using a reconstituted in vitro system. Id proteins represent the prototype of transcriptional modulators that function by dominant negative inhibition of DNA binding. Other examples include some members of the bHLH (47Narumi O. Mori S. Boku S. Tsuji Y. Hashimoto N. Yokota Y. J. Biol. Chem. 2000; 275: 3510-3521Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar, 48Funato N. Ohyama K. Koroda T. Nakamura M. J. Biol. Chem. 2003; 278: 7486-7493Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar), leucine zipper (49Ron D. Habener J.F. Genes Dev. 1992; 6: 439-453Crossref PubMed Scopus (969) Google Scholar), and homeodomain (50Treacy M.N. He X. Rosenfeld M.G. Nature. 1991; 350: 577-584Crossref PubMed Scopus (145) Google Scholar) protein families. The way in which the function of Id proteins is integrated at the level of the subnuclear architecture as shown in the present study may also be relevant to these other transcription factor antagonists, in providing a general mechanism through which they can overcome the high DNA-binding free energy of their target proteins. We thank M. Mancini and M. T. Wilson for helpful discussions, T. Misteli for the EGFP-tagged ASF/SF2 construct, and C. Lupton for technical assistance.
Referência(s)