Human SWI/SNF generates abundant, structurally altered dinucleosomes on polynucleosomal templates

Abstract

Human SWI/SNF (hSWI/SNF) is an evolutionarily conserved ATP-dependent chromatin remodeling complex required for transcriptional regulation and cell cycle control. The regulatory functions of hSWI/SNF are correlated with its ability to create a stable, altered form of chromatin that constrains fewer negative supercoils than normal. Our current studies indicate that this change in supercoiling is due to the conversion of up to one-half of the nucleosomes on polynucleosomal arrays into asymmetric structures, termed "altosomes," each composed of two histone octamers and bearing an asymmetrically located region of nuclease-accessible DNA. Altosomes can be formed on chromatin containing the abundant mammalian linker histone H1 and have a unique micrococcal nuclease digestion footprint that allows their position and abundance on any DNA sequence to be measured. Over time, altosomes spontaneously revert to structurally normal but improperly positioned nucleosomes, suggesting a novel mechanism for transcriptional attenuation as well as transcriptional memory following hSWI/SNF action.

FIG. 1.

Altered micrococcal nuclease footprint of SWI/SNF remodeled polynucleosomes. (a) Polynucleosomal arrays containing 12- by 208-bp sea urchin 5S rRNA gene repeat sequences were remodeled by hSWI/SNF (lanes 4 to 6) for 1 h. Remodeling was stopped with 20-fold excess ADP, followed by digestion with either 2.5 units/ml (lanes 1 and 4), 10 units/ml (lanes 2 and 5), or 40 units/ml (lanes 3 and 6) MNase. For the control reaction (lanes 1 to 3), ADP was added before hSWI/SNF. Naked DNA templates were completely digested by even the lowest concentration of MNase (data not shown). After DNA purification and PAGE, a 196-bp 5S repeat sequence DNA was used as a probe for Southern hybridization. For other conditions see Materials and Methods. Cartoons on right: positions of normal mono-, di-, tri-, and tetranucleosomes. Brackets, positions of sub- and internucleosomal footprints indicative of hSWI/SNF products. Below, percentage altered nucleosomes in lanes 2 and 5 (see Materials and Methods for calculations). (b) Top, cartoon depicting nucleosomes on the ClaI-linearized 5SG5E4 plasmid (2.7 kb of vector sequences, and a central 430-bp transcription cassette flanked on both sides by five 208-bp sea urchin 5S rRNA gene sequences) and probes used for Southern blotting: 196-bp EcoRI 5S, 430-bp EcoRI promoter, or 166-bp DdeI vector fragments. Bottom, as per panel a, except that remodeling was inhibited by omission of hSWI/SNF (lane 1) or ATP (lane 2), or addition of apyrase before hSWI/SNF (1st, lanes 5, 7, and 9). To examine stable remodeled products, apyrase was added after 1 h of remodeling and before MNase (2nd, lanes 4, 6, and 8). Samples of each reaction were treated to PAGE and Southern blotting, followed by hybridization with indicated probes. For these reactions, remodeling was done in a buffer containing 5 mM MgCl₂, 40 mM NaCl, 16 mM KCl, and 2 mM ATP/MgCl₂, and MNase digestion was done at 30°C for 5 min after addition of 2.5 mM CaCl₂ and either 55 U/ml (lanes 1 to 3) or 270 U/ml (lanes 4 to 9, to compensate for the moderate inhibitory effect of apyrase on MNase) MNase. The percentage of altered nucleosomes in lanes 5, 7, and 9 was estimated to be ∼20% (although accurate measurement, in this case, was made difficult by the presence of underdigested MNase bands). (c) Random-sequence bulk MCF7 cell chromatin (lanes 1 to 2) or the 12 by 208 5S array (lanes 3 to 4) were remodeled by hSWI/SNF as described in panel a and stopped by addition of ADP and EDTA before digestion with 8 U/ml MNase. For control reactions, ADP was added before hSWI/SNF. Products were detected by ethidium bromide staining.

FIG. 2.

Altered footprints are not associated with histone loss. 12 by 208 5S polynucleosomal arrays were remodeled, and reactions were stopped by addition of ADP and EDTA, followed by incubation at 37°C in reaction buffer containing 150 mM KCl, in the presence or absence of 20-fold excess of competitor linearized plasmid DNA. The percentage reversion for lanes 4 and 8 were 78% and 73%, determined as per Fig. 3.

FIG. 3.

MNase footprint changes and supercoiling changes arise from the same products. Circular p5SG5E4 plasmid chromatin (5.7 kb) was treated with SWI/SNF in the presence of topoisomerase I for 1 h. After the reaction was stopped by 20× excess ADP and 5 mM EDTA, the buffer was adjusted to 150 mM KCl, and polynucleosomes were allowed to revert for 0, 1, 3, and 6 h at 37°C, before MNase footprint (a) and supercoiling assays (b). Control reactions (“− Remod.”) had ADP and EDTA added before incubation with SWI/SNF. (a) Time-dependent changes in the MNase footprints of control (lanes 1 to 4) and remodeled (lanes 5 to 8) plasmid chromatin. Because Topo I inhibits MNase activity, and later time points had more Topo I, additional MNase was needed to give equivalent levels of digestion (40 U/ml at 0 and 1 h, 80 U/ml at 3 h, and 160 U/ml at 6 h). This resulted in a slightly higher frequency of ∼120-bp mononucleosome overdigestion bands in lanes 2 and 6. For quantitation, these bands were included in the normal mononucleosome region. Below, percentage altered nucleosomes determined as described in Materials and Methods, setting lanes 1 to 4 as controls for lanes 5 to 8, respectively. (b) Topology of plasmid chromatin samples. Two-dimensional gels (1st, no chloroquine; 2nd, 2 μg/ml chloroquine) were used to separate the topoisomers in each of the reactions above. From top left to bottom right: “relaxed,” bare DNA plasmid treated with Topo I in a mock reaction (top right panel); “control,” unremodeled chromatin; “0hrs,” remodeled chromatin assayed immediately after stopping hSWI/SNF; and “1hr,” “3hrs,” or “6hrs,” remodeled chromatin at 1, 3, or 6 h thereafter. Measured linking numbers were as follows: control, −28.25 supercoils; remodeled 0 h, −20.74; 1-h reversion, −24.62; 3-h reversion, −26.07; and 6-h reversion, −25.98. The linking number for the control reaction after 6 h was −27.12 (data not shown). (c) Reversion kinetics. To compare reversion rates, the data above were normalized by setting control reactions to 0% and the 0 h after remodeling time point to 100% remodeled. For supercoiling, the 1- and 3-h reversion results were normalized using the 0-h control.

FIG. 4.

Fractionation of nucleosome products with altered MNase footprints. 5SG5E4 plasmid chromatin (2 μg) was remodeled with SWI/SNF and ATP as in Fig. 1a, stopped by addition of ADP and EDTA, and digested with 40 U/ml MNase. ADP and EDTA was added to the control reaction before hSWI/SNF. MNase digestion was stopped with 15 mM EDTA, followed by addition of a 10-fold excess of competitor DNA, and the reactions were separated by ultracentrifugation on linear glycerol gradients. Note that panels c and d are from a separate gradient fractionation using 5 μg of control or remodeled chromatin. (a) Analysis of gradient samples on 5% acrylamide nucleoprotein (EMSA) gel. Relative gradient mobilities and cartoons depicting the gradient tubes are shown at the top. Lanes 1 to 4, peak fractions from the unremodeled gradient; lanes 5 to 8, remodeled. DNA was detected by Southern blotting and hybridization with a 196-bp 5S sequence probe. (b) Presented as in panel a, except that DNA from the fractions was purified before PAGE. Lanes 9 and 10, input control and remodeled reactions before fractionation. Note that smaller subnucleosomal fragments ran off the bottom of this gel—for best resolution of subnucleosomal bands see panel c. (c) Analysis of subnucleosomal DNA. The peak subnucleosomal DNA fraction (relative mobility of 0.29) from the remodeled gradient was either loaded directly on an EMSA gel (lane 1) or the DNA was purified before PAGE (lane 2, as per panels a and b, above). (d) SDS-PAGE and colloidal Coomassie staining analysis of the histone content of nucleosomal fractions. Lanes 1 and 2, pure core histones from the same stock used to assemble polynucleosomes (328 ng and 210 ng); lane 3, dinucleosomes from the control gradient; and lane 4, dinucleosomes from the remodeled gradient. The relative gradient mobility of both dinucleosome fractions was 0.75. To calculate the relative ratio of core histones for control octamers versus remodeled dinucleosomes, we divided the lane 2 signal by the lane 4 signal for each histone band and normalized by setting the ratio of H3 bands to 1.

FIG. 5.

Redigestion of gradient fractions with MNase. Gradient-purified control (lanes 1 to 4) and remodeled (lanes 5 to 8) dinucleosomes were treated with 0, 2.5, 5, or 10 U/ml MNase (lanes 1 and 5, 2 and 6, 3 and 7, and 4 and 8, respectively), followed by addition of 15 mM EDTA. Half of each sample was analyzed by EMSA (a) and half for DNA length (b) as per Fig. 4. The percentages of total signal due to mononucleosomes (in panel a) or mononucleosome footprint bands (in panel b) are indicated below each lane. The approximately 120- and 100-bp mononucleosome overdigestion fragments are indicated by dotted circles on the right in panel b. Both dinucleosome fractions had a relative gradient mobility of 0.75. p5SG5E4 plasmid was used as probe. Similar results were seen in two separate experiments.

FIG. 6.

Altosomes revert without reestablishing normal positions. 12 by 208 5S polynucleosomal arrays were remodeled, and the reaction was stopped by addition of ADP, followed by incubation at 37°C for 0 or 48 h in reaction buffer containing 30 mM KCl, before digestion with MNase (a) or restriction enzymes (c). In the control reaction, ADP was added before hSWI/SNF. (a) MNase footprint analysis (as per Fig. 1), using 2.5 U/ml MNase. Below, percent altered nucleosomes, calculated as described in Materials and Methods, and percentage of mononucleosome-length DNA (relative to total signal between 50 to 600 bp; “%mono.”). (b) Diagram showing EcoRI and MspI restriction sites relative to the nucleosome positioning sequence on one repeat of the 12 by 208 5S array. (c) Restriction enzyme accessibility analysis. Digestion with 10 units of EcoRI (lanes 1 to 3) or 10 units of MspI (lanes 4 to 6). The regions used to quantitate the proportion of products 1, 2, 3, or more repeats long are indicated by brackets. Note that the weak, short bands in the MspI digest result from digestion of the two repeats at each end of the template. The weak EcoRI bands result from incomplete digestion of one of the two EcoRI sites in each repeat or at the ends of the template. Bare DNA in a mock reaction containing hSWI/SNF was digested to completion under these conditions. The percent cutting for control reactions incubated for 0 or 48 h were the same. (d) Percent reversion for percent altered, percent mononucleosomal, and percent EcoRI and MspI cutting was calculated as follows: % reversion = 100 × [1 − (% at 48 h − % control)/(% at 0 h − % control)]. Similar results were seen in two additional experiments: one with reversion in 30 mM KCl for 48 h, where reversion was 55% for altosomes, 33% for mononucleosomes, 11% for EcoRI, and 29% for MspI, and a second with reversion in 150 mM KCl for 5 h, where reversion was 72% for altosomes, 46% for mononucleosomes, 22% for EcoRI, and 29% for MspI.

FIG. 7.

Linker histones do not block SWI/SNF remodeling. (a) Kinetics of MNase footprint changes. H1-lacking or H1-containing 12 by 208 5S rRNA gene polynucleosomal arrays were remodeled with SWI/SNF, with aliquots treated with ADP and EDTA to terminate remodeling at the indicated times before MNase digestion and Southern blotting. The positions of mononucleosome core (146-bp) and chromatosome (166-bp) bands are indicated. Note that the remodeling conditions used here do not cause any aggregation of arrays with or without H1 (as determined by centrifugal pelleting assays [; also data not shown]). (b) Quantitation of the percentage of altered nucleosomes in panel a and two other independent experiments, as described in Materials and Methods, except that the “abnormal” band regions quantitated were adjusted to account for the extended footprint of H1-bound control nucleosomes (brackets on right). (c) Formation and reversion of MNase footprint changes. After remodeling as in panel a, reactions were stopped by addition of ADP and EDTA and incubated at 37°C in the presence of 150 mM KCl for the indicated times before MNase footprint size analysis. (d) Quantitation of percentage altered nucleosomes in panel c.

FIG. 8.

Hypothetical models for altosome structure. (a) hSWI/SNF-driven movement of the lower nucleosome might pull DNA off the surface of the upper nucleosome. (b) This might create an altered structure containing an intact nucleosome as well as a partially unwrapped nucleosome, constraining only approximately −1/2 supercoils. (c) A short DNA loop with a positive crossing might arise from hSWI/SNF translocation along the phosphate backbone of the DNA to both push DNA into the nucleosome and twist it. (d) An altosome might be created if resolution of this loop were blocked by interaction with an immediately adjacent nucleosome. The overall structure constrains one negative supercoil. (e) DNA entering at the pseudodyad, crossing over from entry to exit point and exiting again at the pseudodyad, results in a structure that constrains one positive supercoil. (f) hSWI/SNF action might release DNA from the surface of the leftmost nucleosome that is then replaced by linker DNA from the far side of the rightmost nucleosome. (g) This could result in an altosome structure where DNA exits and enters at the pseudodyad and a second nucleosome is connected to the normal entry and exit points by short (lower) or long (upper) DNA bridges. The overall structure constrains no supercoils.