Remodeled nucleosomes do not accumulate on closed circular nucleosomal arrays. (A) Nucleosomal arrays were reconstituted on negatively supercoiled closed circular plasmids bearing a unique SalI/HincII site on a single 5S nucleosome positioning sequence. (B) The closed circular arrays were subjected to HincII digestion for 30 min followed by a 30-min incubation with (lane 2) or without (lane 1) SWI-SNF. Alternatively, the arrays were incubated for 30 min with SWI-SNF, followed by removal of ATP with apyrase and digestion with HincII for 30 min (lane 3). (C) A supercoiling assay was performed on the closed circular nucleosomal arrays (lane 1) in order to detect remodeling as changes in plasmid linking number. Addition of calf thymus topoisomerase I resulted in the appearance of approximately 15 bands after deproteination and agarose gel electrophoresis (lane 2). Incubation of the arrays with SWI-SNF and ATP for 30 min had no effect on topology (lane 3). Addition of SWI-SNF, ATP, and topoisomerase I resulted in a redistribution of the topoisomers (lane 4). Removal of ATP by addition of apyrase after 30 min of incubation in the presence of SWI-SNF plus topoisomerase did not affect the distribution of topoisomers (lane 5). Incubation of the arrays with SWI-SNF and ATP for 30 min followed by apyrase and then topoisomerase resulted in a topoisomer distribution similar to that in lane 2 (lane 6). Quantification of the autoradiograph indicates that for lanes 2 and 6, the predominant band corresponds to a linking number of 8; for lanes 4 and 5, the predominant band corresponds to a linking number of 6. Note that apyrase treatment of the circular array results in an increase in the nicked plasmid form (B, lane 3; C, lanes 5 and 6). Apyrase-induced plasmid nicking does not affect HincII activity or the relative distribution of topoisomers upon topoisomerase I treatment of the array (data not shown).

SWI-SNF action leads to increased protection of DNA ends. (A) Schematic of the 216-bp mononucleosome substrate. (B) The ends of the DNA fragment are more protected after a 30-min incubation with SWI-SNF. The graph represents the percentage of cleaved nucleosomal DNA after restriction enzyme digestion without SWI-SNF (−) or after a 30-min incubation with SWI-SNF followed by removal of ATP with apyrase (+). Error bars represent the standard deviation from at least three experiments. (C) Difference in percent nucleosomal DNA cleaved by the restriction enzyme in the absence (−) and presence (+) of SWI-SNF (see panel B). (D) Time course of BamHI DNA cleavage. Mononucleosomes were preincubated for 30 min in the absence of SWI-SNF (□), in the presence of SWI-SNF without ATP (⧫), or in the presence of SWI-SNF and ATP (●, ▴). Reactions containing SWI-SNF and ATP were then incubated with (▴) or without (●) apyrase, BamHI was added to all reactions, and the amount of cleavage was determined throughout a 50-min time course.

SWI-SNF-dependent remodeling changes nucleosome positions within reconstituted arrays. (A) Schematic representation of the DNA template showing positions of restriction endonuclease sites unique to the central nucleosome (open bars). EcoRI sites (filled bars) are located in the linker region between nucleosomes in every 5S repeat. (B) Change in the accessibility of restriction endonuclease sites at the nucleosome dyad (HincII) and at the linker region (NcoI and BamHI) following SWI-SNF reaction. The bars represent percentages of uncut arrays after restriction endonuclease cleavage of reconstituted arrays in the presence of ATP and/or SWI-SNF as indicated. In apyrase (APY) experiments, the SWI-SNF reaction was stopped with apyrase and restriction enzyme was then added. Error bars represent the standard deviation from at least three experiments. (C) SWI-SNF reaction results in a decrease in accessibility of linker regions throughout nucleosomal arrays. The EcoRI digestion of reconstituted nucleosomal arrays was carried out as in panel B; DNA was isolated and analyzed in a 4% acrylamide gel. Gel was stained with Vistra Green (Amersham) and scanned with a Molecular Dynamics Storm scanner. The limit digestion product (208-bp 5S DNA) is marked with an arrow. The 232-bp product represents the HincII/SalI-marked nucleosome. Larger products represent partial digestion products. Lane M contains a 100-bp DNA ladder (New England Biolabs).

Remodeled nucleosomes do not accumulate in the context of linear nucleosomal arrays. (A) Schematic representation of the nucleosomal DNA templates used for the coupled restriction enzyme–SWI-SNF remodeling assay. Each template is composed of head-to-tail repeats of a 5S rDNA nucleosome positioning sequence from L. variegatus. The first, second, or sixth nucleosome is tagged by a unique SalI/HincII restriction site. (B) The nucleosomal arrays were incubated with HincII (□), SWI-SNF (○, ▵), or both (◊). After 1 h, HincII (○) or HincII and apyrase (▵) were added to the reaction to test for accumulation of remodeled template or for ATP dependence, respectively. Cleavage rates were quantified as described in Materials and Methods. Similar results were obtained in three separate experiments.

SWI-SNF remodeling of isolated mononucleosomes leads to persistent DNA accessibility. (A) Purified mononucleosomes, obtained following PstI digestion of a 6-mer nucleosomal array, were treated with SWI-SNF, SalI, and/or apyrase as indicated. (B) Mononucleosomes assembled on a 154-bp DNA fragment were used as substrates in the same assays as panel A. Similar results were obtained in two additional, independent experiments.

SWI-SNF action alters the translational positioning of a mononucleosome. Mononucleosomes reconstituted onto a 416-bp DNA fragment that contains two copies of a 5S nucleosome positioning sequence were electrophoresed on a 4% native polyacrylamide gel for 8 h. Arrows denote eight electrophoretically distinct particles that reflect different nucleosome translational positions. Mononucleosomes were incubated for 30 min at 37°C in the absence of SWI-SNF (lane 1), in the presence of SWI-SNF but in the absence of ATP (lane 2), and in the presence of both SWI-SNF and ATP (lane 3). Prior to loading of samples, all three reactions received a 100-fold molar excess of unlabeled chicken oligonucleosomes to compete for binding of SWI-SNF to the mononucleosome substrate. Note that SWI-SNF and ATP shifts the distribution of mononucleosome particles to the faster-migrating species without leading to an increase in the amount of free, naked DNA. Similar results were obtained in at least three additional experiments.

Persistent randomization of nucleosome positions within reconstituted arrays as a result of SWI-SNF-dependent remodeling. (A) MNase digestion of reconstituted 11-mer arrays. [α-³²P]dATP-labeled reconstituted nucleosomal arrays were digested with increasing amounts of MNase in the presence of SWI-SNF and either in the presence or absence of ATP before or after addition of apyrase (APY). DNA was isolated and analyzed on a 1% agarose gel. Lane M, ³²P-labeled 1-kb DNA ladder (Gibco BRL). Note that 5S array DNA also exhibits a repeating pattern of MNase cleavages that is the inverse pattern of nucleosomal arrays. (B) SWI-SNF-dependent remodeling does not change the sedimentation properties of nucleosomal arrays. Shown are 10 to 40% linear glycerol gradient profiles of naked DNA and reconstituted nucleosomal arrays either before or after SWI-SNF-dependent remodeling. ³²P-labeled reconstituted arrays were incubated in the presence of ATP and/or SWI-SNF for 30 min, loaded on top of the gradient, and centrifuged in an SW28 rotor (Beckman) at 33,000 × g for 16 h. Fractions of 0.4 ml were collected from the top of the gradient and counted by scintillation. (C) Nucleosomes within remodeled arrays protect ∼150 bp of DNA. Nucleosomal arrays were incubated in the presence of SWI-SNF with or without ATP, the reaction was stopped with apyrase, and remodeled arrays were digested with a high concentration of MNase. DNA was isolated, fractionated in a 4% acrylamide gel, and visualized by staining with Vistra Green. The position of the 146-bp nucleosomal DNA is marked with an arrow. Lane M, 100-bp DNA ladder.