Structural characterisation of chromatin remodelling intermediates supports linker DNA-dependent product inhibition as a mechanism for nucleosome spacing
- Molecular Cell and Developmental Biology, School of Life Sciences, University of Dundee, United Kingdom
Figures
Figure 1
DNA gaps at SHL +2, but not at SHL -2 prevent repositioning by Chd1.
(A) 2 bp gaps were introduced as illustrated on a linear representation of a DNA fragment, including the 601 nucleosome positioning sequence and an asymmetric 49 bp linker and (B) on a 3D representation of the same DNA bound by Chd1 (based on PDB 6 FTX). Positions are numbers relative to the dyad with the positive numbers proximal to extranucleosomal DNA; C or G designates the strand at which the dyad base is a C or G, respectively. (C) Representative native polyacrylamide gels showing how the repositioning of nucleosomes is affected on templates with DNA gaps introduced at the locations indicated. Repositioning reactions included 100 nM labelled nucleosomes, 5 nM Chd1, and 1 mM ATP; 10 μL aliquots were stopped with addition of 0.1 μg/ μL competitor DNA, 0.2 M sodium chloride, and 1.6% sucrose on ice immediately (t=0) and after sliding for 2, 4, 8, 16, and 32 min at 30°C. (D) Quantitation of 3 repeats for each DNA template as compared to intact nucleosomal DNA are fitted to a hyperbolic curve. Error bars indicate the Standard Error. (E) The extent of nucleosome repositioning is assessed at high resolution using site-directed hydroxyl radical cleavage directed via attachment of FeEDTA at S47C. Lane 1 contains a molecular weight marker, Lanes 2–19 include Chd1, with or without ATP as indicated and were subject to directed hydroxyl radical mapping as indicated. Nucleosomal DNA was either intact (lanes 2–4) or included 2 bp gaps as the locations indicated (lanes 5–19). Cleavage sites are detected following denaturing PAGE. Cleavages corresponding to the initial, intermediate and product nucleosome locations are indicated in the schematic to the left.
-
Figure 1—source data 1
Raw data files for gel scans in .tif and .adf formats displayed in Figure 1.
- https://cdn.elifesciences.org/articles/52513/elife-52513-fig1-data1-v2.zip
-
Figure 1—source data 2
Data images in Figure 1 as .pdf files with relevant areas labelled.
- https://cdn.elifesciences.org/articles/52513/elife-52513-fig1-data2-v2.zip
Figure 2
Access to SHL +2, but not SHL -2, is required for Chd1-directed nucleosome repositioning.
DNA fragments in which the 601 sequence is flanked by a 49 bp linker were generated with a biotin-modified base introduced at –17 (A) or +17 (B). Nucleosomes were assembled onto these fragments and found to be repositioned to more central locations following incubation of 100 nM nucleosome with 5 nM Chd1 and 1 mM ATP. In the presence of 400 nM streptavidin, both nucleosome and DNA fragments are supershifted (Lanes 6–10). A supershift corresponding to repositioning of streptavidin-bound nucleosomes was only observed for nucleosomes bound by streptavidin at SHL -2. (C) Nucleosome repositioning was measured at high resolution using site-directed hydroxyl radical cleavage. A change in cleavage consistent with repositioning by 20 bp is observed when streptavidin is bound at SHL-2 (lanes 1–5).
-
Figure 2—source data 1
Raw data files for gel scans in .tif and .adf formats displayed in Figure 2.
- https://cdn.elifesciences.org/articles/52513/elife-52513-fig2-data1-v2.zip
-
Figure 2—source data 2
Images displayed in Figure 2 as .pdf files with relevant areas labelled.
- https://cdn.elifesciences.org/articles/52513/elife-52513-fig2-data2-v2.zip
Figure 3 with 5 supplements
Chd1-nucleosome complexes detected in the presence of ATP.
Complexes between Chd1 and nucleosomes with 29 bp linker DNA on one side were incubated with 100 μM ATP for 10 min at 25°C prior to cross-linking and cryo-grid preparation. Following data processing three major Chd1-bound complexes are detected. (A) Complex I with DNA fully wrapped and in an identical state to that observed in unbound nucleosomes. (B) Complex II with poorly defined DNA ends suggesting an ensemble of intermediates detected as a result of ongoing remodelling. (C) Complex III in which Chd1 is bound in a configuration similar to that observed in previous structures except that DNA ends are not well defined (Partial density indicated in black) suggesting some repositioning has occurred. Density is coloured as follows: DNA strands - orange and purple, histone octamer - grey, Chd1 chromodomains - yellow, Chd1 ATPase Lobe1 and Lobe2 - light and dark blue, respectively.
Figure 3—figure supplement 1
Image processing and 3D reconstruction of Chd1-nucleosome complexes.
Flowchart showing different stages in the reconstruction of Chd1-nucleosome complexes. The number of particles and methods used in the different stages are shown.
Figure 3—figure supplement 2
Local resolution and FSC resolution estimation.
Local resolution maps computed for the free nucleosome and three Chd1-bound nucleosome complexes. The resolution range are span out between 2.2 Å-6.2 Å, which are colour coded. Blue represents the highest resolution and red the lowest. FSC curves were calculated between two independently refined half-maps, before (blue) and after (green, red, and purple) masking, and reported at 0.143 FSC cutoff.
Figure 3—figure supplement 3
Distinct states of DNA wrapping in Chd1 complexes.
Comparison of DNA path between three different Chd1-nucleosome complexes. DNA coloured in red represents the configuration observed in complex III (product-inhibited Chd1 complex), DNA coloured black represents the configuration in complex I (Chd1 bound adjacent to short linker side with short linker fully wrapped) and DNA coloured grey represents the configuration observed in complex II (Chd1 intermediate unwrapping). Chd1 domains are shown in surface representation with the Chromo, ATPase Lobe1, ATPase Lobe2 and DNABD coloured yellow, light blue, dark blue, and navy, respectively. The superhelical location (SHL) of DNA within the nucleosome are numbered as indicated.
Figure 3—figure supplement 4
Table illustrating key parameters for cryo-electron microscopy (cryo-EM) structures.
Figure 3—figure supplement 5
Illustrations of fit to density map.
The fit of structural models to electron density is shown for each of the three complexes determined in the study. The fits for the histone octamer, DNA, and ATPase lobes are shown separately. The bottom panel shows the fit of the ChEx domain observed in complex II and complex III in comparison to the previously published density for this region obtained from PDB 7TN2 (Nodelman et al., 2022).
Figure 4 with 1 supplement
ATP-dependent changes to the binding of Chd1 ATPase domains.
Hydroxyl radical mapping, directed by Chd1 S524C (300 nM), on nucleosomes (100 nM) with a 13 bp linker reports on the location of the ATPase lobes in apo, ADP-BeF3-, and ATP-bound states. (A) shows cleavage sites detected on the top strand whereas (B) shows cleavage detected on the bottom strand. In the apo and ADP-BeF3 states, cleavage occurs on the –2 side of the nucleosome, whereas in the presence of ATP, contact with both the +2 and –2 sides of the nucleosome is observed. Nucleosome mapping via hydroxyl radical cleavage shows that nucleosomes are not repositioned under the conditions of these reactions (C). The coloured asterisks indicate the locations of the fluorescent label used to detect DNA.
-
Figure 4—source data 1
Raw data files for gel scans in .tif and .adf formats displayed in Figure 4.
- https://cdn.elifesciences.org/articles/52513/elife-52513-fig4-data1-v2.zip
-
Figure 4—source data 2
Images displayed in Figure 4 as .pdf files with relevant areas labelled.
- https://cdn.elifesciences.org/articles/52513/elife-52513-fig4-data2-v2.zip
Figure 4—figure supplement 1
Chd1 S524C retains nucleosome repositioning activity.
(A) Repositioning assays performed with 300 nM Chd1 S524C and 100 nM nucleosomes with ATP concentration and times as indicated. (B) Native gel showing integrity of nucleosomes used in Figure 4A and B.
-
Figure 4—figure supplement 1—source data 1
Raw data files for gel scans in .tif and .adf formats displayed in Figure 4—figure supplement 1.
- https://cdn.elifesciences.org/articles/52513/elife-52513-fig4-figsupp1-data1-v2.zip
-
Figure 4—figure supplement 1—source data 2
Images displayed in Figure 4—figure supplement 1 as .pdf files with relevant areas labelled.
- https://cdn.elifesciences.org/articles/52513/elife-52513-fig4-figsupp1-data2-v2.zip
Figure 5
A mechanism for Chd1-mediated nucleosome spacing.
Chd1 is likely to initially bind nucleosomes in the conformation previously observed in the absence of nucleotides RSC PDB 7TN2 (Nodelman et al., 2022). In the presence of ATP, this is converted to Complex I, in which the organisation of nucleosomal DNA has the same arrangement as nucleosomes prior to remodelling, but Chd1 is bound on the opposite side. The lack of DNA dynamics suggests Complex I is inactive for repositioning, and in the absence of linker DNA on the entry side, the location of the DNA binding domain is dynamic and so not resolved. Transition to Complex II involves association of the ChEx domain. In Complex II, the DNA ends are poorly defined, consistent with ongoing ATP-dependent translocation, which is likely to occur in single base steps. Extension of exit DNA to a length of approximately 15 bp, sufficient to bind the DNA binding domain, allows formation of Complex III. Association of the DNA binding domain is likely to constrain further DNA translocation, meaning this Complex III can be considered product inhibited. Via a process similar to the initial change in ATPase lobe binding, Complex III could be converted to a conformation related to Complex I, but with the ATPase lobes bound on the other side of the nucleosome. These symmetry-related complexes are labelled with brackets (Complex I). Activation of this complex would result in a second round of repositioning but with opposing directionality. Multiple cycles are anticipated to drive bidirectional repositioning tuned to favour a nucleosomal repeat length close to 15 bp. Colouring of Chd1 and nucleosomes is as in Figure 3.
Tables
Table 1
Primers used for generating nucleosomal DNA.
| Primer | Sequence | Purpose |
|---|---|---|
| AFD 1698 | CTGCAGAAGCTTGGTCCCGGG | Unlabelled Nuc |
| AFD 1443 | Cy5-CTGCAGAAGCTTGGTCCC | Cy5 labelled Nuc |
| AFD 1629 | Cy3-CTGCAGAAGCTTGGTCCC | Cy3 labelled Nuc |
| AFD 874 | Cy5-TCAAGCTTGCATGCCTGCAGG | Cy5 labelled 49 bp (+) side extension |
| AFD 875 | Cy3-TCAAGCTTGCATGCCTGCAGG | Cy3 labelled 49 bp (+) side extension |
| AFD 1199 | Cy3-CCCCTACATGCACAGGATG | Cy3 labelled 13 bp (+) side extension |
| AFD 1203Cy5 | Cy5-CCCCTACATGCACAGGATG | Cy5 labelled 13 bp (+) side extension |
| 601C-73[49]* | CTGCAGAAGCTTGGTCCCGGGGCCGCTCAATTGGTCGTAGCAAGCTCTA | top strand with –24 base (+) side reduction |
| 601G-73[49]* | TAGAGCTTGCTACGACCAATTGAGCGGCCCCGGGACCAAGCTTCTGCAG | bottom strand with –24 base (+) side reduction |
| 601C-22 | TCCGCTTAATCGAACGTACGCG | PCR with AFD874 to isolate top, right strand for –24/22 gap Nuc |
| 601C-73[59]* | CTGCAGAAGCTTGGTCCCGGGGCCGCTCAATTGGTCGTAGCAAGCTCTAGATCCGCTTA | top strand of Nuc with –14 base reduction on (+) side |
| 601C-12 | CGAACGTACGCGCTGTCCCCCG | PCR with AFD874 to isolate top, right strand for –14/12 gap |
| 601C-73[69]* | CTGCAGAAGCTTGGTCCCGGGGCCGCTCAATTGGTCGTAGCAAGCTCTAGATCCGCTTAATCGAACGTA | top strand of Nuc with –4 base reduction on (+) side |
| 601C-2 | CGCTGTCCCCCGCGTTTTAACC | PCR with AFD874 to isolate top, right strand for –4/2 gap |
| 601G+5 Cy5 | Cy5-GGA CAG CGC GTA CGT TCG ATT AAG | PCR with AFD1698 to isolate top, left strand for +5/7 gap and SHL-2 biotin |
| 601C+7 | CGCGTTTTAACCGCCAAGGGGA | PCR with AFD874 to isolate top, right strand +5/7 gap |
| 601G+15 Cy5 | Cy5 labelled, PCR with 1698 to isolate top, right strand for +15/17 gap | |
| 601C+17 | CCGCCAAGGGGATTACTCCCTAG | PCR with AFD874 to isolate top, left strand for +15/17 gap |
| 601C-73[99]* | CTGCAGAAGCTTGGTCCCGGGGCCGCTCAATTGGTCGTAGCAAGCTCTAGATCCGCTTAATCGAACGTACGCGCTGTCCCCCGCGTTTTAACCGCCAAG | top strand with +25 base extension on (+) side |
| 601C+27[93]* | GATTACTCCCTAGTCTCCAGGCACGTGTCAGATATATACATCCTGTGCATGTAGGGGATTCTCTAGAGTCGACCTGCAGGCATGCAAGCTTGA | top strand with +27 base extension on (-) side |
| 601G+35 Cy5 | /Cy5/GAG TAA TCC CCT TGG CGG TTA AAA CG | Cy5 labelled, PCR with AFD1698 to isolate top, left strand for +35/37 gap |
| 601C+37 | TAG TCT CCA GGC ACG TGT CAG AT | PCR with AFD874 to isolate top, right strand for +35/37 gap |
| 601CBio-16 (TA-rich) | CTGCAGAAGCTTGGTCCCGGGGCCGCTCAATTGGTCGTAGCAAGCTCTAGATCCGC/BiodT/TAATCGAACGTACGCG | biotin @SHL-2, PCR with AFD874 |
-
*
HPLC purified oligonucleotides from Eurofins.
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Structural characterisation of chromatin remodelling intermediates supports linker DNA-dependent product inhibition as a mechanism for nucleosome spacing
eLife 14:e52513.
https://doi.org/10.7554/eLife.52513
