(A) Silver stain gel of TAP-purified RSC2, Rpd3S and FACT complexes.
(B) The effect of Rpd3S on RSC-dependent nucleosome remodeling. 2 nM RSC was incubated with 0.3nM ³²P-labeled mononucleosome and 0, 13, 39 or 78 nM Rpd3S. The remodeling products were fractionated by native PAGE. A phosphorimage of the gel is shown. See also for the effect of H3K36me3 on Rpd3S in binding and RSC-mediated nucleosome remodeling reactions.
(C) The effect of Rpd3S on RSC dependent nucleosome eviction. Left panel, 6 nM RSC was incubated with 0.3nM mononucleosome and 0, 13, 39 or 78 nM Rpd3S in the presence of 10 ng of pGEM3Z601R acceptor DNA. Bar graph on the right represents quantitation by ImageQuant TL (GE) of the 3 independent experiments. The relative amounts of free DNA generated by eviction were plotted as a bar graph normalized to that generated by 6 nM RSC alone, which was assigned a value of 100. The error bars show +/− standard deviation (SD). The P value is calculated using student’s t-test. See also for the effect of H3K36me3 on Rpd3S in RSC-mediated nucleosome eviction.
(D) Rpd3S-mediated chromatin assembly assay. Left, the reaction contained 18 nM FACT, or 6, 12, 18 nM of Rpd3S respectively, with recombinant Xenopus octamers and a ³²P-labeled 601 DNA fragment. A phosphorimage of a native gel is shown. Graph on the right represents quantitation of the amounts of assembled nucleosomes by Rpd3S relative to no Rpd3S control (i.e., octamers alone). The free DNA and assembled nucleosome are indicated. The error bars show +/− standard deviation (SD). The P value is calculated using student’s t-test. For chaperone assays see also for the effect of H3K36me3 on Rpd3S, for the effect of Rpd3S mutants and for the effect of trichostatin.

(A) Silver stain gel of TAP-purified Rpd3L.
(B) The effect of Rpd3L on RSC-dependent nucleosome eviction. 6 nM RSC was incubated with 0, 15, 45, 90 nM Rpd3L, respectively, and analyzed as described in legend.
(C) Nucleosome assembly with Xenopus octamers and 18 nM Rpd3L or 3-subunit core complex, respectively, as in legend.
(D) Silver stain gel of recombinant 3-subunit core complex. The asterisk (*) indicates an unknown protein that co-purified with the 3-subunit core complex.
(E) The effect of 3-subunit core complex on RSC-dependent nucleosome eviction. 6 nM RSC was incubated with 601 nucleosome and 0, 15, 45, 90 nM of recombinant core complex, respectively, and analyzed as in legend. See also for silver stained gels of the individual subunits, for their effect on RSC-mediated nucleosome eviction and for chaperone assays.

(A) Schematic of the C-tail template. The template contains the 601 positioning sequence and a 20-nucleotide single-stranded C-tail with an intervening polylinker from pGEM3Z601R. Pol II initiates from the C-tail.
(B) The template was assembled into a mononucleosome with naïve recombinant Xenopus octamers and then preincubated with 3 nM RSC in the presence or absence of 30 or 60 nM Rpd3S, Rpd3L, 3-subunit core complex or FACT for 1h at 30°C. Pol II, α³²P-CTP, NTPs and RNase H were then added for 15 min at 30°C. The ³²P-labeled RNA products were fractionated on a 10% polyacrylamide/urea gel. A phosphorimage of the gel is shown.
(C) Same assay as performed in (B), except that naked C-tail DNA template was used instead of nucleosomal template.
(D) In vitro transcription was performed on naive or H3K36me3 nucleosomes. 3 nM RSC was incubated with 0, 5, 15, 45 nM Rpd3S or recombinant 3-subunit core complex, respectively.

(A) H3 levels were measured by ChIP in wild-type (WT), rpd3 H150A (H150A), and rpd3Δ (Vector) cells. ChIP DNA of Histone H3 and inputs were amplified, labeled, and hybridized to Agilent Tiling arrays. The average binding of 6,572 annotated genes and their upstream 500-bp regions are shown. Enrichment of H3 ChIP DNA is shown as the log2 ratios of ChIP versus input DNA. P values for the promoters were calculated using the Mann-Whitney test. See also for Rpd3 native vs. H150A protein levels, for global acetylation levels of H3K18 and H4K5 in Rpd3 native, H150A and null backgrounds, for Rpd3 association with Pol II and H3.
(B) Box and whisker plot for two subsets of targets that have high or low Rsc2 enrichment. Each subset has 495 and 506 targets, respectively. The P values are calculated using student’s t-test.
(C) H3 levels of the low Rsc2 targets were measured in wild-type, rpd3 H150A, and rpd3Δ cells. P values for the promoters were calculated using the Mann-Whitney test.
(D) H3 levels of the high Rsc2 targets were measured in wild-type, rpd3 H150A, and rpd3Δ cells. P values for the promoters were obtained from the Mann-Whitney test.