Nodelman IM et al. (2016), The Chd1 chromatin remodeler can sense both ent...

XB-ART-53745

Nucleic Acids Res 2016 Sep 19;4416:7580-91. doi: 10.1093/nar/gkw406.

Show Gene links Show Anatomy links

The Chd1 chromatin remodeler can sense both entry and exit sides of the nucleosome.

Nodelman IM , Horvath KC , Levendosky RF , Winger J , Ren R , Patel A , Li M , Wang MD , Roberts E , Bowman GD .

???displayArticle.abstract???
Chromatin remodelers are essential for establishing and maintaining the placement of nucleosomes along genomic DNA. Yet how chromatin remodelers recognize and respond to distinct chromatin environments surrounding nucleosomes is poorly understood. Here, we use Lac repressor as a tool to probe how a DNA-bound factor influences action of the Chd1 remodeler. We show that Chd1 preferentially shifts nucleosomes away from Lac repressor, demonstrating that a DNA-bound factor defines a barrier for nucleosome positioning. Rather than an absolute block in sliding, the barrier effect was achieved by altered rates of nucleosome sliding that biased redistribution of nucleosomes away from the bound Lac repressor site. Remarkably, in addition to slower sliding toward the LacO site, the presence of Lac repressor also stimulated sliding in the opposite direction. These experiments therefore demonstrate that Chd1 responds to the presence of a bound protein on both entry and exit sides of the nucleosome. This sensitivity to both sides of the nucleosome allows for a faster and sharper response than would be possible by responding to only the entry side, and we speculate that dual entry/exit sensitivity is also important for regularly spaced nucleosome arrays generated by Chd1 and the related ISWI remodelers.

???displayArticle.pubmedLink??? 27174939
???displayArticle.pmcLink??? PMC5027475
???displayArticle.link??? Nucleic Acids Res
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: chd1 pnp smarca5

???attribute.lit??? ???displayArticles.show???

	Figure 1. The presence of Lac repressor limits the extent and rate of nucleosome sliding by Chd1. Repositioning 0N80 nucleosomes by Chd1, monitored by histone mapping at H2B(S53C). Shown are representative examples of sliding reactions carried out in the absence of LacI using 0N80[LacO+1] nucleosomes (A), and reactions carried out in the presence of LacI for 0N80 nucleosomes with LacO binding sites located +1 bp (B), −11 bp (C) or −6 bp (D) from the right edge of the nucleosome. Top cartoons illustrate the relative positions of LacO and LacI on the nucleosome, whereas bottom cartoons summarize the major nucleosome positions resulting from each experiment. (E) Extended reach of the Chd1 DNA-binding domain does not alter the inhibitory effects of LacI. Nucleosome sliding was carried out using a variant of Chd1 containing a 121-residue insertion between the ATPase motor and DNA-binding domain, called Chd1RAM. Shown is a representative of three experiments. Time points for reactions in (A–E) were 0, 0.25 min (for LacO(−6) and LacO(−11) only), 0.5, 1, 2, 4, 8, 16, 32, 64 min. (F) Quantification of the nucleosome sliding experiments shown in (A–D), revealing a more rapid disappearance of the starting material when remodeled in the absence (open circles) compared with the presence (filled circles) of LacI. Each point represents the average values from two or more experiments, with error bars showing the range of values from two measurements, or the standard deviations for three or more measurements. Rate constants were calculated from single exponential fits to the data, with the error given as the standard deviation from averaging calculated rate constants. (G) Sliding in the presence of LacI does not permanently disrupt nucleosomes. Remodeling of 0N80[LacO-11R] nucleosomes was monitored by native PAGE. After 30 min of sliding, IPTG was added to release LacI, and the change in nucleosome mobility shows that this addition was sufficient for removing the LacI barrier.
	Figure 2. Exit side DNA defines the extent that Chd1 slides nucleosomes in the presence of Lac repressor. (A) Nucleosomes are dynamically shifted back-and-forth by Chd1 in the presence of LacI. End-positioned 0N70 nucleosomes containing a LacO(−11) site on the 70 bp side and an EcoRI cut site just inside the 0 bp side were digested by EcoRI and monitored by SDS-PAGE. The slower rate of digestion in the absence of ATP (gray) demonstrates that the EcoRI site is initially buried. In the presence of Chd1 and ATP, the same fraction of nucleosomal DNA becomes cleaved in the presence (filled circle) or absence (open circle) of LacI, demonstrating dynamic repositioning by Chd1 in the presence of LacI. Based on single exponential fits to the data, the sliding rate in the presence of LacI was calculated to be 7-fold slower than in the absence of LacI. Error bars indicate the standard deviations from five or more independent experiments. (B) Comparison of nucleosome sliding reactions carried out in the absence and presence of LacI, using (−10)N80[LacO-11R] substrates. Time points for these experiments were 0, 1, 4, 16, 64 min. (C) Comparison of the preferred distributions of nucleosome positions for 0N80 and (−10)N80 nucleosomes when Chd1 sliding was carried out in the presence of LacI. White peaks show zero time points and gray peaks are the nucleosome positions at 64 min time points.
	Figure 3. Chd1 can bind to nucleosomes at SHL2 in the presence of Lac repressor. Interactions between Chd1 and Cy3-labeled H4(A15C) nucleosomes were measured in the presence of the ATP analog AMP-PNP by monitoring changes in Cy3 fluorescence with increasing Chd1 concentrations. Error bars indicate the standard deviations from three or more fits to independent experiments.
	Figure 4. The ability of Chd1 to sense extranucleosomal DNA is not required to respond to a LacI barrier. (A) LacI diminished Chd1 sliding even in the absence of the Chd1 DNA-binding domain. Sliding experiments were performed with 0N80[LacO-11] nucleosomes using a Chd1 construct that lacked the DNA-binding domain (Chd1[ΔDBD]), carried out in the presence or absence of LacI. The left side shows a representative experiment, and the right side shows intensity profiles for each time point. (B) The bridge element of Chd1 is required for properly responding to extranucleosomal DNA. Nucleosome sliding experiments used pre-centered 40N40 nucleosomes. (C) LacI limits nucleosome sliding by the Chd1[GGS/bridge] variant. Shown are nucleosome sliding experiments using end-positioned 0N80[LacO-11] nucleosomes, carried out in the presence or absence of LacI. Time points for gels in (B) and (C) were 0, 0.25, 0.5, 1, 2, 4, 8, 16, 32 and 64 min. All experiments were analyzed by native acrylamide gels and carried out three or more times.
	Figure 5. LacI does not sequester Chd1 on the nucleosome. Shown are Chd1 sliding experiments carried out with one or two types of nucleosome, in the presence (filled symbols, solid lines) or absence (open circles and ‘X’, dotted lines) of LacI. Nucleosomes lacking the LacO sequence were Cy5 labeled (blue in the graph), whereas nucleosomes containing the LacO sequence were Cy3 labeled (pink). Data were quantified from native gel sliding experiments, with each point showing the mean from six to eight independent experiments and error bars representing the standard deviations.
	Figure 6. Chd1 shifts nucleosomes more rapidly away from Lac repressor. Centered 40N40 nucleosomes (150 nM) with a LacO site on the left (A) or right (B) were repositioned by Chd1 (50 nM) in the presence and absence of LacI. Nucleosome positions are shown by histone mapping at position H2B(S53C), with DNA fragments separated on DNA sequencing gels. Time points were 0, 0.25, 1, 4, 16, 64 min. Fits to a five state kinetic model for the quantified 40N40 sliding experiments are shown in the absence (C) and presence (D) of LacI. (E) Faster nucleosome sliding away from the LacO sites in the presence of LacI is required for fitting the kinetic model to the data. Fits for the plus-LacI data were carried out with the rate away from LacO (C→D for LacO(−11L) and C→B for LacO(−11R)) fixed to the slower values obtained in the absence of LacI. The dotted lines show the best fits from >800 000 trials (see Supplementary Figure S3 for the best 1000 trials), all of which fail to match the data. The colouring between the curves emphasizes the differences in fits with and without this rate constraint. (F) Summary of the kinetic models. For 40N40 nucleosomes with LacO on the left or the right, the relative rates of repositioning away from the LacO site was increased whereas shifting toward LacO was decreased compared with the rates obtained for the absence of LacI. States that are disfavoured in the presence of LacI due to asymmetry in forward/reverse rates are partially hidden by dotted boxes. Error bars in (C–E) indicate standard deviations from three independent mapping experiments for each condition. See also Supplementary Figure S1 and S2.
	Figure 7. Sensing both sides of the nucleosome allows for greater asymmetry in nucleosome sliding rates. In the absence of DNA-bound factors, nucleosomes are dynamically shifted in either direction by chromatin remodelers such as Chd1. The association of a DNA-binding protein such as a transcription factor at the edge of a nucleosome decreases the rate of sliding toward and increases sliding away from the bound factor, which promotes rapid redistribution of nucleosomes away from the bound factor. By enforcing asymmetry in the bidirectional sliding of the nucleosome, DNA-binding proteins effectively provide barriers that direct nucleosome positioning.

References [+] :

Bai, Gene regulation by nucleosome positioning. 2010, Pubmed