Sporadic mutations in the hMeCP2 gene, coding for a protein that preferentially binds symmetrically methylated CpGs, result in the severe neurological disorder Rett syndrome (RTT). In the present work, employing a wide range of experimental approaches, we shed new light on the many levels of MeCP2 interaction with DNA and chromatin. We show that strong methylation-independent as well as methylation-dependent binding by MeCP2 is influenced by DNA length. Although MeCP2 is strictly monomeric in solution, its binding to DNA is cooperative, with dimeric binding strongly correlated with methylation density, and strengthened by nearby A/T repeats. Dimeric binding is abolished in the F155S and R294X severe RTT mutants. MeCP2 also binds chromatin in vitro, resulting in compaction-related changes in nucleosome architecture that resemble the classical zigzag motif induced by histone H1 and considered important for 30-nm-fiber formation. In vivo chromatin binding kinetics and in vitro steady-state nucleosome binding of both MeCP2 and H1 provide strong evidence for competition between MeCP2 and H1 for common binding sites. This suggests that chromatin binding by MeCP2 and H1 in vivo should be viewed in the context of competitive multifactorial regulation.
Bacterial spores have long been recognized as the sturdiest known life forms on earth, revealing extraordinary resistance to a broad range of environmental assaults. A family of highly conserved spore-specific DNA-binding proteins, termed alpha/beta-type small, acid-soluble spore proteins (SASP), plays a major role in mediating spore resistance. The mechanism by which these proteins exert their protective activity remains poorly understood, in part due to the lack of structural data on the DNA-SASP complex. By using cryoelectron microscopy, we have determined the structure of the helical complex formed between DNA and SspC, a characteristic member of the alpha/beta-type SASP family. The protein is found to fully coat the DNA, forming distinct protruding domains, and to modify DNA structure such that it adopts a 3.2-nm pitch. The protruding SspC motifs allow for interdigitation of adjacent DNA-SspC filaments into a tightly packed assembly of nucleoprotein helices. By effectively sequestering DNA molecules, this dense assembly of filaments is proposed to enhance and complement DNA protection obtained by DNA saturation with the alpha/beta-type SASP.
Mutations of the methylated DNA binding protein MeCP2, a multifunctional protein that is thought to transmit epigenetic information encoded as methylated CpG dinucleotides to the transcriptional machinery, give rise to the debilitating neurodevelopmental disease Rett syndrome (RTT).In this in vitro study, the methylation-dependent and -independent interactions of wild-type and mutant human MeCP2 with defined DNA and chromatin substrates were investigated.A combination of electrophoretic mobility shift assays and visualization by electron microscopy made it possible to understand the different conformational changes underlying the gel shifts.MeCP2 is shown to have, in addition to its well-established methylated DNA binding domain, a methylation-independent DNA binding site (or sites) in the first 294 residues, while the C-terminal portion of MeCP2 (residues 295 to 486) contains one or more essential chromatin interaction regions.All of the RTT-inducing mutants tested were quantitatively bound to chromatin under our conditions, but those that tend to be associated with the more severe RTT symptoms failed to induce the extensive compaction observed with wild-type MeCP2.Two modes of MeCP2-driven compaction were observed, one promoting nucleosome clustering and the other forming DNA-MeCP2-DNA complexes.MeCP2 binding to DNA and chromatin involves a number of different molecular interactions, some of which result in compaction and oligomerization.The multifunctional roles of MeCP2 may be reflected in these different interactions.
Most cases of Rett syndrome (RTT) are caused by mutations in the methylated DNA-binding protein, MeCP2. Here, we have shown that frequent RTT-causing missense mutations (R106W, R133C, F155S, T158M) located in the methylated DNA-binding domain (MBD) of MeCP2 have profound and diverse effects on its structure, stability, and DNA-binding properties. Fluorescence spectroscopy, which reports on the single tryptophan in the MBD, indicated that this residue is strongly protected from the aqueous environment in the wild type but is more exposed in the R133C and F155S mutations. In the mutant proteins R133C, F155S, and T158M, the thermal stability of the domain was strongly reduced. Thermal stability of the wild-type protein was increased in the presence of unmethylated DNA and was further enhanced by DNA methylation. DNA-induced thermal stability was also seen, but to a lesser extent, in each of the mutant proteins. Circular dichroism (CD) of the MBD revealed differences in the secondary structure of the four mutants. Upon binding to methylated DNA, the wild type showed a subtle but reproducible increase in alpha-helical structure, whereas the F155S and R106W did not acquire secondary structure with DNA. Each of the mutant proteins studied is unique in terms of the properties of the MBD and the structural changes induced by DNA binding. For each mutation, we examined the extent to which the magnitude of these differences correlated with the severity of RTT patient symptoms.
The SWI/SNF complex disrupts and mobilizes chromatin in an ATP-dependent manner. SWI/SNF interactions with nucleosomes were mapped by DNA footprinting and site-directed DNA and protein cross-linking when SWI/SNF was recruited by a transcription activator. SWI/SNF was found by DNA footprinting to contact tightly around one gyre of DNA spanning ∼50 bp from the nucleosomal entry site to near the dyad axis. The DNA footprint is consistent with nucleosomes binding to an asymmetric trough of SWI/SNF that was revealed by the improved imaging of free SWI/SNF. The DNA site-directed cross-linking revealed that the catalytic subunit Swi2/Snf2 is associated with nucleosomes two helical turns from the dyad axis and that the Snf6 subunit is proximal to the transcription factor recruiting SWI/SNF. The highly conserved Snf5 subunit associates with the histone octamer and not with nucleosomal DNA. The model of the binding trough of SWI/SNF illustrates how nucleosomal DNA can be mobilized while SWI/SNF remains bound.
Methylated DNA binding protein 2 (MeCP2) is a methyl CpG binding protein whose key role is the recognition of epigenetic information encoded in DNA methylation patterns. Mutation or misregulation of MeCP2 function leads to Rett syndrome as well as a variety of other autism spectrum disorders. Here, we have analyzed in detail the properties of six individually expressed human MeCP2 domains spanning the entire protein with emphasis on their interactions with each other, with DNA, and with nucleosomal arrays. Each domain contributes uniquely to the structure and function of the full-length protein. MeCP2 is ∼60% unstructured, with nine interspersed α-molecular recognition features (α-MoRFs), which are polypeptide segments predicted to acquire secondary structure upon forming complexes with binding partners. Large increases in secondary structure content are induced in some of the isolated MeCP2 domains and in the full-length protein by binding to DNA. Interactions between some MeCP2 domains in cis and trans seen in our assays likely contribute to the structure and function of the intact protein. We also show that MeCP2 has two functional halves. The N-terminal portion contains the methylated DNA binding domain (MBD) and two highly disordered flanking domains that modulate MBD-mediated DNA binding. One of these flanking domains is also capable of autonomous DNA binding. In contrast, the C-terminal portion of the protein that harbors at least two independent DNA binding domains and a chromatin-specific binding domain is largely responsible for mediating nucleosomal array compaction and oligomerization. These findings led to new mechanistic and biochemical insights regarding the conformational modulations of this intrinsically disordered protein, and its context-dependent in vivo roles.
MeCP2 is a transcriptional repressor that contains an N-terminal methylated DNA-binding domain, a central transcription regulation domain, and a C-terminal domain of unknown function. Whereas current models of MeCP2 function evoke localized recruitment of histone deacetylases to specific methylated regions of the genome, it is unclear whether MeCP2 requires DNA methylation to bind to chromatin or whether MeCP2 binding influences chromatin structure in the absence of other proteins. To address these issues, we have characterized the complexes formed between MeCP2 and biochemically defined nucleosomal arrays. At molar ratios near 1 MeCP2/nucleosome, unmethylated nucleosomal arrays formed both extensively condensed ellipsoidal particles and oligomeric suprastructures. Furthermore, MeCP2-mediated chromatin compaction occurred in the absence of monovalent or divalent cations, in distinct contrast to all other known chromatin-condensing proteins. Analysis of specific missense and nonsense MeCP2 mutants indicated that the ability to condense chromatin resides in region(s) of the protein other than the methylated DNA-binding domain. These data demonstrate that MeCP2 assembles novel secondary chromatin structures independent of DNA modification and suggest that the ability of MeCP2 to silence chromatin may be related in part to its effects on large-scale chromatin organization. MeCP2 is a transcriptional repressor that contains an N-terminal methylated DNA-binding domain, a central transcription regulation domain, and a C-terminal domain of unknown function. Whereas current models of MeCP2 function evoke localized recruitment of histone deacetylases to specific methylated regions of the genome, it is unclear whether MeCP2 requires DNA methylation to bind to chromatin or whether MeCP2 binding influences chromatin structure in the absence of other proteins. To address these issues, we have characterized the complexes formed between MeCP2 and biochemically defined nucleosomal arrays. At molar ratios near 1 MeCP2/nucleosome, unmethylated nucleosomal arrays formed both extensively condensed ellipsoidal particles and oligomeric suprastructures. Furthermore, MeCP2-mediated chromatin compaction occurred in the absence of monovalent or divalent cations, in distinct contrast to all other known chromatin-condensing proteins. Analysis of specific missense and nonsense MeCP2 mutants indicated that the ability to condense chromatin resides in region(s) of the protein other than the methylated DNA-binding domain. These data demonstrate that MeCP2 assembles novel secondary chromatin structures independent of DNA modification and suggest that the ability of MeCP2 to silence chromatin may be related in part to its effects on large-scale chromatin organization. MeCP2 belongs to a family of DNA-binding proteins that can selectively bind methylated CpG dinucleotides. These proteins have a highly conserved methylated CpG-binding domain (MBD) 1The abbreviations used are: MBD, methylated CpG-binding domain; EM, electron microscopy; ECM, electron cryomicroscopy. (1Wade P.A. Oncogene. 2001; 20: 3166-3173Crossref PubMed Scopus (179) Google Scholar). The overall domain structure of MeCP2 is considerably different from that of other MBD family members (1Wade P.A. Oncogene. 2001; 20: 3166-3173Crossref PubMed Scopus (179) Google Scholar, 2Bird A.P. Wolffe A.P. Cell. 1999; 99: 451-454Abstract Full Text Full Text PDF PubMed Scopus (1573) Google Scholar). In particular, most of the middle portion of MeCP2 consists of a transcriptional repression domain, while at the extreme C terminus are repetitive sequences similar to those found in forkhead proteins (1Wade P.A. Oncogene. 2001; 20: 3166-3173Crossref PubMed Scopus (179) Google Scholar, 3Vacca M. Filippini F. Budillon A. Rossi V. Mercadante G. Manzati E. Gualandi F. Bigoni S. Trabanelli C. Pini G. Calzolari E. Ferlini A. Meloni I. Hayek G. Zappella M. Renieri A. D'Urso M. D'Esposito M. MacDonald F. Kerr A. Dhanjal S. Hulten M. J. Mol. Med. 2001; 78: 648-655Crossref PubMed Scopus (54) Google Scholar). MeCP2 is also fundamentally different from other MBD family members in that numerous specific mutations scattered throughout the entire polypeptide chain are closely associated with the molecular pathogenesis of the neurological disorder Rett syndrome (4Lee S.S. Wan M. Francke U. Brain Dev. 2001; 23: S138-S143PubMed Google Scholar, 5Van den Veyver I.B. Zoghbi H.Y. Brain Dev. 2001; 23: S147-S151Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). DNA methylation is classically associated with imprinted genes, transposable elements and their relics, and the inactive X chromosome in female mammals (6Bestor T.H. Hum. Mol. Genet. 2000; 9: 2395-2402Crossref PubMed Scopus (1636) Google Scholar). Because regions of dense cytosine methylation are so strongly correlated with transcriptional repression in vivo (2Bird A.P. Wolffe A.P. Cell. 1999; 99: 451-454Abstract Full Text Full Text PDF PubMed Scopus (1573) Google Scholar, 6Bestor T.H. Hum. Mol. Genet. 2000; 9: 2395-2402Crossref PubMed Scopus (1636) Google Scholar), virtually all molecular investigations to date have focused on the possible role of MeCP2 as a specific DNA methylation-dependent transcriptional repressor (1Wade P.A. Oncogene. 2001; 20: 3166-3173Crossref PubMed Scopus (179) Google Scholar, 2Bird A.P. Wolffe A.P. Cell. 1999; 99: 451-454Abstract Full Text Full Text PDF PubMed Scopus (1573) Google Scholar, 7Wade P.A. Bioessays. 2001; 23: 1131-1137Crossref PubMed Scopus (292) Google Scholar). The current paradigm for MeCP2-mediated transcriptional repression involves specific, targeted deacetylation of promoter chromatin mediated by the MeCP2-Sin3-HDAC1 corepressor complex (1Wade P.A. Oncogene. 2001; 20: 3166-3173Crossref PubMed Scopus (179) Google Scholar, 8Jones P.L. Veenstra G.J. Wade P.A. Vermaak D. Kass S.U. Landsberger N. Strouboulis J. Wolffe A.P. Nat. Genet. 1998; 19: 187-191Crossref PubMed Scopus (2271) Google Scholar, 9Nan X. Ng H.H. Johnson C.A. Laherty C.D. Turner B.M. Eisenman R.N. Bird A. Nature. 1998; 393: 386-389Crossref PubMed Scopus (2823) Google Scholar). However, it is well documented that MeCP2 can repress transcription independently of histone acetylation (7Wade P.A. Bioessays. 2001; 23: 1131-1137Crossref PubMed Scopus (292) Google Scholar), implying that MeCP2-mediated transcriptional repression does not exclusively involve local recruitment of histone deacetylases. Furthermore, to date, there have been no examinations of whether MeCP2 directly affects either local or global chromatin structure. It has previously been shown that a glutathione S-transferase-MeCP2 hybrid protein can bind to methylated DNA wrapped within a reconstituted mononucleosome (10Chandler S.P. Guschin D. Landsberger N. Wolffe A.P. Biochemistry. 1999; 38: 7008-7018Crossref PubMed Scopus (153) Google Scholar). These studies with Xenopus laevis MeCP2 showed that both the MBD and C terminus are involved in MeCP2-nucleosome interactions, suggesting that MeCP2 binding to chromatin substrates may be more complex than to naked DNA. Here, we have determined the hydrodynamic and electrophoretic properties of the nucleoprotein complexes formed in solution between human recombinant MeCP2 and defined 12-mer nucleosomal arrays and visualized their structures by electron cryomicroscopy and electron microscope tomography. We found that at molar ratios near 1 protein/nucleosome, binding of MeCP2 led to salt-independent conversion of extended nucleosomal arrays into extensively compacted 60 S ellipsoidal structures. Furthermore, at MeCP2/nucleosome molar ratios of ≥1.0, the 60 S particles assembled into morphologically defined oligomeric suprastructures. Studies of specific Rett syndrome mutants indicated that the ability to mediate chromatin compaction lies in a region of the protein other than the MBD. Thus, in addition to functions related to binding methylated DNA, human MeCP2 is a potent chromatin-condensing protein that mediates assembly of novel secondary chromatin structures. The relevance of these results to current models of MeCP2 function is discussed. Protein Purification—Histone octamers were isolated from chicken erythrocytes as described by Hansen et al. (11Hansen J.C. Ausio J. Stanik V.H. van Holde K.E. Biochemistry. 1989; 28: 9129-9136Crossref PubMed Scopus (201) Google Scholar). Purified histone octamer fractions were stored at 4 °C in the presence of 20 μg/ml each aprotinin and leupeptin. Human recombinant full-length MeCP2 (486 amino acids, 52,408 Da), MeCP2(R133C), and MeCP2(R168X) were prepared from clones provided by T. Yusufzai (12Yusufzai T.M. Wolffe A.P. Nucleic Acids Res. 2000; 28: 4172-4179Crossref PubMed Scopus (117) Google Scholar). Following induction with isopropyl-β-d-thiogalactopyranoside, bacteria were collected, washed (25 mm Tris-HCl (pH 7.5) and 100 mm NaCl), resuspended in Buffer A (25 mm Tris-HCl (pH 7.5), 500 mm NaCl, 0.25% Triton X-100, 5% glycerol, 1 mm phenylmethylsulfonyl fluoride, 0.5 mm benzamidine, and 20 μg/ml l-1-tosylamido-2-phenylethyl chloromethyl ketone), and sonicated. Following centrifugation at 15,000 rpm for 30 min, the cleared lysate was applied to chitin-agarose (New England Biolabs Inc.) pre-equilibrated in Buffer A. The column was washed with 10 column volumes of Buffer A, followed by 2 column volumes of Buffer A containing 50 mm dithiothreitol. The column was then sealed and incubated at 4 °C for 48 h. Following incubation, the column was washed with Buffer A, and fractions containing the bulk of the eluted MeCP2 were pooled. The chitin-agarose-purified protein was diluted with water to a conductance equivalent to 250 mm NaCl and applied to a heparin-Sepharose column pre-equilibrated in 25 mm Tris-HCl, 5% glycerol, 0.25 m NaCl, 1 mm phenylmethylsulfonyl fluoride, 0.5 mm benzamidine, and 20 μg/ml l-1-tosylamido-2-phenylethyl chloromethyl ketone). MeCP2 was eluted with a linear gradient of 0.25–1.0 m NaCl. Nucleosomal Array Reconstitution and Formation of MeCP2-Nucleosomal Array Complexes—208-12 DNA templates were purified from the pPol-I-208-12 plasmid (13Georgel P. Demeler B. Terpening C. Paule M.R. van Holde K.E. J. Biol. Chem. 1993; 268: 1947-1954Abstract Full Text PDF PubMed Google Scholar) as described (11Hansen J.C. Ausio J. Stanik V.H. van Holde K.E. Biochemistry. 1989; 28: 9129-9136Crossref PubMed Scopus (201) Google Scholar). Preparations of 208-12 nucleosomal arrays containing 10–12 nucleosomes were reconstituted from purified core histone octamers and 208-12 DNA using salt dialysis as described (14Hansen J.C. Lohr D. J. Biol. Chem. 1993; 268: 5840-5848Abstract Full Text PDF PubMed Google Scholar). 1.1 mol of histone octamer/mol of 208-bp DNA was used for the reconstitutions. The final dialysis step for electrophoretic mobility shift assay and analytical ultracentrifugation experiments was performed against TEN buffer (10 mm Tris-HCl, 0.25 mm EDTA, and 2.5 mm NaCl (pH 7.8)). The nucleosomal arrays used for electron microscopy (EM) imaging were dialyzed against HEGN buffer (10 mm Hepes, 0.25 mm EDTA, 10% glycerol, and 2.5 mm NaCl). Nucleoprotein complexes were formed by incubating 208-12 nucleosomal arrays or 208-12 DNA (200 ng in 20 μl) with increasing molar ratios of MeCP2 to 208-bp repeat (r MeCP2 = 0.1–2.0) at room temperature for 30 min. The incubation buffer was either TEN or HEGN buffer. Prior to loading of the samples, glycerol was added to a final percentage of 10%. EcoRI Digestion—Nucleoprotein complexes formed in the presence or absence of various molar ratios of MeCP2 were digested with 10 units of EcoRI/μg of DNA at 37 °C for 90 min. The reaction buffer contained 10 mm Tris, 0.25 mm EDTA, 7.5 mm NaCl, and 1.75 mm MgCl2. The digestion was stopped by adding EDTA to a final concentration of 15 mm. Half of the sample was deproteinized by treatment with 10 μg of proteinase K at 50 °C for 60 min, phenol/chloroform-extracted, and ethanol-precipitated. The native chromatin and deproteinized DNA samples were electrophoresed on a 1% agarose gel buffered with 1× Tris borate/EDTA at 8 V/cm for 2 h. Sedimentation Velocity—All sedimentation velocity studies were performed in a Beckman XL-I analytical ultracentrifuge. Samples were sedimented at 18,000 rpm. The A 260 was ∼0.6, and the temperature was 20 ± 0.1 °C. Data were acquired with the absorbance optics. The boundaries were analyzed by the method of van Holde and Weischet (15van Holde K.E. Weischet W.O. Biopolymers. 1978; 17: 1387-1405Crossref Scopus (320) Google Scholar) using the Ultrascan data analysis program to yield the integral distribution of sedimentation coefficients, plotted as the boundary fraction versus s 20,w. Electron Microscopy—Samples for negative staining or shadowing were first cross-linked as follows. 208-12 DNA or nucleosomal arrays at a DNA concentration of 50 μg/ml were incubated with MeCP2 at a the desired molar ratio for 1 h at room temperature in HEGN buffer and then fixed by dialysis against HEGN buffer containing 0.1% glutaraldehyde for 4 h at 4 °C, followed by two changes of HEGN buffer for a total of 16 h. For negative staining, fixed samples were diluted 8-fold with 50 mm NaCl, adsorbed to a glow-discharged carbon film, washed with double-distilled water, negatively stained with 2% uranyl acetate, and air-dried. Some MeCP2/DNA samples were rinsed briefly with water after uranyl acetate treatment to provide positive staining. For shadowing, samples were brought to 75% glycerol and then sprayed onto freshly cleaved mica as described (16Tyler J.M. Branton D. J. Ultrastruct. Res. 1980; 71: 95-102Crossref PubMed Scopus (301) Google Scholar). The mica was dried under high vacuum in a turbo-pumped evaporator (BA080, Baltec Products Inc., Hudson, NH) and, while under vacuum, shadowed with platinum at an angle of 6° and then coated with a thin layer of carbon. To provide both an accurate value for the size of the particles and a view of the complete shape, the sample was kept stationary for the first one-fourth of the platinum evaporation procedure and then rotated for the remaining three-fourths. After removal from the vacuum, the carbon films were floated off the mica on distilled water and mounted on 400-mesh grids. Samples were examined in a Tecnai 12 transmission electron microscope (FEI Co., Hillsboro, OR) operating at 100 kV, and images were recorded on a TVIPS slow-scan CCD (TVIPS GmbH, Gauting, Germany). Most shadowed preparations and all positively stained grids were examined using tilted beam dark-field optics. For tomographic reconstruction of negatively stained samples, tilt series were recorded with a pixel size at the image of 0.45 nm. The tilt data consisted of images at 2.5° intervals between –70° and +65° with an average dose of 0.5 electrons/A2. Alignment of the tilt series by cross-correlation and reconstruction by sinc-weighted back-projection was performed using SUPRIM software (17Schroeter J.P. Bretaudiere J.P. J. Struct. Biol. 1996; 116: 131-137Crossref PubMed Scopus (93) Google Scholar) on an SGI Octane workstation (Silicon Graphics, Mountain View, CA). The computed three-dimensional reconstructions were analyzed using Amira software (TGS Inc., San Diego, CA). Electron cryomicroscopy (ECM) was carried out by applying 3-μl drops of an unfixed sample of r MeCP2 = 1.0 nucleosomal arrays to glow-discharged Quantifoil grids (Quantifoil Micro Tools GmbH, Jena, Germany) in a humidity-controlled chamber maintained at 85% relative humidity at room temperature. The sample droplet was blotted with Whatman No. 4 filter paper before plunging the grid into liquefied ethane. The frozen grids were transferred under liquid nitrogen to a Gatan 626 crytotransfer system (Gatan Inc., Pleasanton, CA) and observed at –180 °C in the Tecnai 12 transmission electron microscope operating at 120 kV. Stereo pair images with a 15° separation were digitally recorded as described above. Three-dimensional measurements from stereo pairs were calculated as previously described (18Bednar J. Horowitz R.A. Dubochet J. Woodcock C.L. J. Cell Biol. 1995; 131: 1365-1376Crossref PubMed Scopus (141) Google Scholar), and model reconstructions were created using Vertigo software (Vertigo, Inc., Vancouver, British Columbia, Canada). Human Recombinant MeCP2 Binds to Long Linear DNA and Forms Cross-linked Aggregates—As a control, we initially examined whether human recombinant MeCP2 binds to linear 208-12 DNA. This DNA template consists of 12 tandem 208-bp repeats of Lytechinus 5 S rDNA (19Simpson R.T. Thoma F. Brubaker J.M. Cell. 1985; 42: 799-808Abstract Full Text PDF PubMed Scopus (387) Google Scholar) and is extensively used for studies of chromatin structure/function relationships (20Hansen J.C. Annu. Rev. Biophys. Biomol. Struct. 2002; 31: 361-392Crossref PubMed Scopus (425) Google Scholar). Human MeCP2 was purified from recombinant bacteria using chitin and heparin affinity chromatography (see “Experimental Procedures”) and judged to be ≥95% pure by SDS-PAGE/Coomassie Brilliant Blue staining (Fig. 1A). Recombinant protein was added to the DNA at molar ratios of 0.25–2.0 MeCP2 monomers/208-bp DNA. After incubation in low ionic strength TEN buffer for 30 min at room temperature, the samples were electrophoresed on native 1.0% agarose gels. This same large particle gel shift assay has been used previously to study binding of linker histones (21Carruthers L.M. Bednar J. Woodcock C.L. Hansen J.C. Biochemistry. 1998; 37: 14776-14787Crossref PubMed Scopus (210) Google Scholar) and yeast Sir3p (22Georgel P.T. Palacios DeBeer M.A. Pietz G. Fox C.A. Hansen J.C. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 8584-8589Crossref PubMed Scopus (54) Google Scholar) to 208-12 DNA and nucleosomal arrays. We found that there was a large progressive reduction in DNA mobility upon addition of increasing MeCP2 molar ratios (Fig. 1B). The smearing seen at the lower r MeCP2 values is indicative of the presence of multiple nucleoprotein species. A relatively discrete, more slowly migrating band was observed at r MeCP2 = 1.0, whereas at r MeCP2 = 2.0, none of the DNA was able to migrate through the gel. The greatly reduced band mobilities seen at r MeCP2 ≥ 0.5 indicate that MeCP2 binds to long unmethylated DNA under low ionic strength conditions and assembles large nucleoprotein complexes. The structural features of the MeCP2-DNA complexes formed at r MeCP2 = 1.0 were visualized by EM (Fig. 1C). As expected, 208-12 DNA appeared as an individual, randomly curved, linear molecule (Fig. 1C, panel i). In contrast, DNA samples bound to MeCP2 contained structures with thickened regions formed by the close apposition of two DNA strands (Fig. 1C, panels ii–v). On either side of the apposed regions, the two DNA molecules diverged (Fig. 1C, arrows), presumably because of their mutual electrostatic repulsion under these low salt conditions. In some cases, several 208-12 DNA molecules existed together in intricate structures (Fig. 1C, panel v). We interpret the thickened regions as places in which two DNA strands are bridged by MeCP2 molecules. Similar complexes have been observed upon incubation of naked DNA with MENT (23Grigoryev S.A. Bednar J. Woodcock C.L. J. Biol. Chem. 1999; 274: 5626-5636Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar) and the globular domain of linker histones (24Thomas J.O. Rees C. Finch J.T. Nucleic Acids Res. 1992; 20: 187-194Crossref PubMed Scopus (82) Google Scholar), both of which are proteins involved in local and global chromatin condensation. Taken together, the results shown in Fig. 1 (B and C) demonstrate that MeCP2 is able to “cross-link” long DNA fragments into large complex structures and that MeCP2-mediated assembly of such structures occurs in the absence of DNA methylation. These results also suggest a potential “nucleation” mechanism whereby MeCP2 binding to methylated genomic sites may lead to spreading of MeCP2 into unmethylated adjoining regions. MeCP2 Mediates Compaction of Nucleosomal Arrays into Novel Secondary Chromatin Structures in the Absence of Salts—MeCP2 functions in a chromatin environment in vivo (1Wade P.A. Oncogene. 2001; 20: 3166-3173Crossref PubMed Scopus (179) Google Scholar, 2Bird A.P. Wolffe A.P. Cell. 1999; 99: 451-454Abstract Full Text Full Text PDF PubMed Scopus (1573) Google Scholar) and is not likely to encounter long stretches of naked DNA. Hence, we next examined the interactions of MeCP2 with nucleosomal arrays assembled from 208-12 DNA and chicken erythrocyte histone octamers. The intrinsic solution behavior of these biochemically defined 12-mer nucleosomal arrays is very well understood (20Hansen J.C. Annu. Rev. Biophys. Biomol. Struct. 2002; 31: 361-392Crossref PubMed Scopus (425) Google Scholar), making them ideal templates for investigating MeCP2 interactions with nucleosomes and chromatin in vitro. Human recombinant MeCP2 was mixed with the nucleosomal arrays at molar ratios of 0.25–2.0 MeCP2/nucleosome and incubated in low ionic strength TEN or HEGN buffer for 30 min at room temperature. The samples were then electrophoresed on native 1.0% agarose gels and also characterized by EM (Fig. 2). Large electrophoretic mobility shifts were observed with increasing protein molar ratios, indicative of formation of MeCP2-nucleosomal array complexes in TEN buffer (Fig. 2A). The same results were obtained after incubation in HEGN buffer (data not shown). The unique structural features of these nucleoprotein complexes were revealed using a variety of EM techniques. Under subsaturating conditions (r MeCP2 = 0.5), shadowed EM images obtained in HEGN buffer showed areas of local compaction within individual nucleosomal arrays (Fig. 2B, panels b–f), presumably reflecting regions where MeCP2 was bound to nucleosomes. A relatively broad range of compacted array morphologies were observed, consistent with smearing of the r MeCP2 = 0.5 samples during electrophoresis (Fig. 2A). More striking are the images obtained at r MeCP2 = 1.0 (Fig. 2B, panels g–i). Under these conditions, the dominant structures were highly condensed, ellipsoidal particles. In some cases, we also observed what appeared to be oligomers of the ellipsoidal particles (Fig. 2C, panel i, arrows), which we interpret as corresponding to the smear of material seen directly above the predominant band on agarose gels (Fig. 2A, fifth lane). Importantly, there was no intrinsic folding of the nucleosomal arrays under the low ionic strength conditions of the binding reaction (Fig. 2B, panel a) (20Hansen J.C. Annu. Rev. Biophys. Biomol. Struct. 2002; 31: 361-392Crossref PubMed Scopus (425) Google Scholar). Hence, the intra-array compaction observed in low salt can be attributed entirely to MeCP2. To determine whether MeCP2-mediated compaction also occurs under more physiological ionic conditions, the experiment in Fig. 2A was repeated in the presence of either 2 mm MgCl2 or 150 mm NaCl. The resulting gel patterns were essentially identical to those shown in Fig. 2A (data not shown), indicating that MeCP2-mediated chromatin fiber compaction is independent of ionic strength and is not an artifact of low salt conditions. The structures of the compacted nucleoprotein particles formed at r MeCP2 = 1.0 were further characterized by sedimentation velocity in the analytical ultracentrifuge, restriction enzyme probing of linker DNA accessibility, and three-dimensional image reconstructions of complexes in negative stain and in the frozen hydrated state. Sedimentation velocity experiments were performed to obtain the sedimentation coefficients of the MeCP2-compacted arrays and to determine the distribution of nucleoprotein species present in solution. Data were analyzed by the method of van Holde and Weischet (15van Holde K.E. Weischet W.O. Biopolymers. 1978; 17: 1387-1405Crossref Scopus (320) Google Scholar) to yield the integral distribution of sedimentation coefficients, plotted as boundary fraction against s 20,w (Fig. 3). Such a plot will be vertical if the sample is homogeneous (15van Holde K.E. Weischet W.O. Biopolymers. 1978; 17: 1387-1405Crossref Scopus (320) Google Scholar). The sedimentation experiments were performed in low salt buffer to prevent intrinsic salt-dependent array compaction (20Hansen J.C. Annu. Rev. Biophys. Biomol. Struct. 2002; 31: 361-392Crossref PubMed Scopus (425) Google Scholar). Under these conditions, the parent nucleosomal arrays sedimented as a nearly homogeneous 27–29 S population (data not shown), corresponding to the extended “beads-on-a-string” conformation (Fig. 2B, panel a). However, at r MeCP2 = 1.0, ∼60% of the sample sedimented as a 60 S species, whereas the remainder was spread between 60 and 110 S (Fig. 3). Extensively condensed 12-mer nucleosomal arrays sediment near 55 S (11Hansen J.C. Ausio J. Stanik V.H. van Holde K.E. Biochemistry. 1989; 28: 9129-9136Crossref PubMed Scopus (201) Google Scholar, 21Carruthers L.M. Bednar J. Woodcock C.L. Hansen J.C. Biochemistry. 1998; 37: 14776-14787Crossref PubMed Scopus (210) Google Scholar). Hence, we conclude that the 60 S species present in the r MeCP2 = 1.0 sample corresponds to the highly compacted ellipsoidal structures observed in the electron microscope (Fig. 2B, panels g–i). As discussed below, the heterogeneous population of rapidly sedimenting species seen in Fig. 3 are oligomers of the 60 S particles (e.g. Fig. 2B, panel i). Further insight into the structure of the 60 S complexes was obtained from EM image reconstruction. All three imaging methods (shadowing, negative staining, and ECM) revealed roughly ellipsoidal particles, although the detailed shape of each individual particle was slightly different (Figs. 2B and 4). To better understand their internal structure, tomographic three-dimensional reconstructions were carried out on 60 S particles embedded in negative stain (Fig. 4, panels a–g). From the reconstructed volumes, it was possible to recognize the location of each nucleosome as well as the orientation of the nucleosomal disks. The three-dimensional reconstructions also allowed estimation of the mean volume of the 60 S particles as 15 × 103 nm3 (S.E. = 0.6 ± 103 nm3), which is ∼2.2 times the volume of 12 nucleosomes. In comparison, a solenoidal arrangement with six nucleosomes/turn occupies ∼6 times the volume of 12 nucleosomes. Whereas negative staining provides a relatively stable specimen and high contrast imaging, the processes of chemical fixation and grid adhesion may potentially alter the hydrated three-dimensional solution structure. These problems are obviated in frozen hydrated specimens obtained by ECM (25Woodcock C.L. Horowitz R.A. Methods Cell Biol. 1998; 53: 167-186Crossref PubMed Google Scholar). An overview of a field of unfixed, unstained complexes imaged by ECM is shown in Fig. 4 (panels h–l). Stereo pair ECM images of monomeric particles allowed the three-dimensional location (but not the orientation) of nucleosomes to be identified. Fig. 4 (panels i–l) shows the result of placing 10-nm diameter spheres at nucleosome positions derived from the specific particles denoted by the arrows in panel h. In the absence of substrate adhesion, the 60 S complexes were not compressed in one plane, and the probable linear path of the 12 connected nucleosomes was often evident. The “envelopes” of the three-dimensional structures occupied approximately the same volume in the ECM and negatively stained reconstructions, indicating that conformational distortions due to substrate adhesion in the latter are minimal. Nuclease digestion was used to probe both the linker DNA accessibility and the nature of the macromolecular interactions that stabilize the 60 S complexes. Each tandem repeat in the 5 S rDNA template is linked by EcoRI restriction sites (19Simpson R.T. Thoma F. Brubaker J.M. Cell. 1985; 42: 799-808Abstract Full Text PDF PubMed Scopus (387) Google Scholar), and most of these sites are located in the linker DNA of the parent 208-12 nucleosomal arrays (Fig. 5A, lane 7) (11Hansen J.C. Ausio J. Stanik V.H. van Holde K.E. Biochemistry. 1989; 28: 9129-9136Crossref PubMed Scopus (201) Google Scholar, 26Meersseman G. Pennings S. Bradbury E.M. J. Mol. Biol. 1991; 220: 89-100Crossref PubMed Scopus (112) Google Scholar). We digested samples assembled at r MeCP2 = 0–2.0 with EcoRI and examined the electrophoretic band pattern before and after deproteinization of the digests. The deproteinized digests reveal information about linker DNA accessibility (22Georgel P.T. Palacios DeBeer M.A. Pietz G. Fox C.A. Hansen J.C. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 8584-8589Crossref PubMed Scopus (54) Google Scholar). Digestion of the parent nucleosomal arrays produced a ladder of DNA bands on 1.0% agarose gels as expected (Fig. 5A, lane 7), indicative of protection of a fraction of the EcoRI sites by nucleosomes (11Hansen J.C. Ausio J. Stanik V.H. van Holde K.E. Biochemistry. 1989; 28: 9129-9136Crossref PubMed Scopus (201) Google Scholar, 26Meersseman G. Pennings S. Bradbury E.M. J. Mol. Biol. 1991; 220: 89-100Crossref PubMed Scopus (112) Google Scholar). Somewhat to our surprise, digestion of the MeCP2-compacted 60 S complexes yielded a pattern of bands essentially identical to that of the nucleosomal array control (Fig. 5A, lanes 8–10). Thus, the linker DNA segments in condensed 60 S structures are as equally accessible as those in the parental nucleosomal arrays. Electrophoresis of the non-deproteinized samples indicated whether cleavage of the linker DNA caused the 60 S complexes to dissociate into mononucleosomes. The results from electrophoresis of the native digests are shown in Fig. 5B. There was little change in the electrophoretic mobility of the samples; and in particular, the r MeCP2 = 1.0 sample remained almost entirely intact (compare Figs. 2A and 5B). Thus, either direct MeCP2-MeCP2 or bivalent DNA-MeCP2-DNA bridging interactions appear to be the m
