Nucleosome chains fold and self-associate to form higher order structures whose internal organization is unknown. Here, cryo-electron tomography (cryo-ET) of native human chromatin reveals novel folding motifs such as 1) non-uniform nucleosome stacking, 2) intermittent parallel and perpendicular orientations of adjacent nucleosome planes, and 3) an inverse zigzag nucleosome chain path, which deviates from the direct zigzag topology seen in reconstituted nucleosomal arrays. By examining these self-associated structures, we observed prominent nucleosome stacking in-cis and anti-parallel nucleosome interactions in-trans, which are consistent with partial nucleosome interdigitation. Histone citrullination strongly inhibits nucleosome stacking and self-association with a modest effect on chromatin folding, while the reconstituted arrays showed a zigzag topology which undergoes a dramatic unfolding induced by histone citrullination. This study sheds light on the internal structure of compact chromatin nanoparticles and suggests a novel mechanism for how epigenetic changes in chromatin are retained across both open and condensed forms of chromatin.

Cryo-Electron microscopy and tomographic reconstruction

Chromatin samples incubated for 20 min. without or with appropriate concentrations of MgCl2 were mixed with a suspension of 10 nm fiduciary gold particles (Sigma Aldrich # 741957), which were coated in bovine serum albumin to prevent clustering. 3 ul chromatin samples with a concentration of about 0.2 mg/ml DNA were applied to Quantifoil R2/2 200 mesh copper grids (EMS Q250-CR2). Vitrification was conducted by plunging into liquid ethane using our FEI Vitrobot Mk IV Grid Plunging System at 100% humidity, 4o C, and setting the blotting strength at 5, and blotting time at 3.5 sec. Imaging of the vitrified samples was conducted on Titan Krios G3i 300 kV electron microscope, equipped with a K3 direct electron detector (Gatan, CA) at the Penn State Hershey cryo-EM core. We used Tomography-5.7.1. software (Thermo Fisher) for controlling data acquisition and collecting tilt-series. CryoEM tilt series (± 60o ) were collected at 5o intervals in dose symmetric mode, at either 2.2 angstroms/pixel (defocus -6 um) or 1.7 angstroms/pixel (defocus -5 um), with zero-loss peak energy filtration through a 20-eV slit. Images were collected in 1x counting mode. Each tilt series had a cumulative dose of 120 electrons/Å2. Tilt series were aligned using fiducials in the IMOD software suite (https://bio3d.colorado.edu/imod/), CTF corrected and SIRT reconstructed.

Centroid/axis/plane (CAP) modeling of nucleosome chain folding

The reconstructed un-binned tomograms were visualized and segmented into smaller subtomograms by IMOD/3dmod. Each volume was inverted using “newstack” to generate subtomograms with positive intensity corresponding to high density and filtered using IMOD command: “nad_eed_3d -n 30 -f -k 50” to reduce noise and enhance chromatin edges. The filtered subtomograms were exported into UCSF CHIMERA (RBVI, Univ. San Francisco, CA) for interactive visualization and analysis of nucleosome structures. In CHIMERA, the filtered volumes were fitted with nucleosome core X-ray crystal structure (pdb 2CV5) semi-automatically to correspondent electron densities in the volume using the ‘fitmap’ command. The fitting of each nucleosome followed the local maximal electron density and thus was independent from the observer’s bias. After fitting, the nucleosomes were overlaid with centroids selecting the centers of masses, center-to-center axes, and nucleosome planes crossing the nucleosome at the dyad axis using the structure analysis ‘Axels/Planes/Centroids’ tool. For stereological analysis, each nucleosome in an array was numbered, if possible, starting from one end of the array and numbering consecutive nucleosomes to the other end of the array. Where linkers were not visible (such as in the tertiary chromatin structures), the nucleosomes were labeled based on their spatial proximity. The following measurements were recorded from such Centroid-Axial-Plane (CAP) models for each nucleosome (n) in an array: a) center-to-center distance D to the next nucleosome (n+1) in the array, b) center-to-center distance N to the nearest nucleosome (nx) in the 3D space, angle a between the two axes connecting each nucleosome with the previous one (n-1) and the next one (n+1) in a chain, angle b between the planes of consecutive nucleosomes n and n+1, and an angle para between the plane of each nucleosome (n) and the plane of the nearest nucleosome (nx) in the 3D space (see Fig 3 B). All distance and angle measurements for individual nucleosome arrays are included in the Suppl. Table 2. The absolute nucleosome proximity values were obtained by subtracting the number of each nucleosome (n) from the number of its nearest nucleosome (nx) in the 3D space. The measured angle and plane values were analyzed statistically to determine the distribution profiles and average values that discriminate between the condensed and open nucleosomes arrays. Distance N and angle para were recorded for all nucleosomes. For some nucleosomes condensed by 0.75 mM Mg2+ (~21% of all nucleosomes) and for all nucleosomes condensed at > 1 mM Mg2+, the linkers were not resolved and corresponding distances D and angles alpha and beta were excluded from the statistical analysis. SDs were obtained from at least three tomograms and at least two independently cultured biological samples; p- values represent probability associated with a Student’s two-sample unequal variance t-test with a two-tailed distribution.

This data is separated into two different folders.

1) tilt_series_data - Contains all the original cryo-EM tilt series data used for analysis in the paper.

To view the data, you will need to download and install IMOD: https://bio3d.colorado.edu/imod/

To view a single tilt series use the “3dmod” . Like this,

3dmod tilt_series_name.mrc

2) CAP_models - Each subfolder contains folders with both the filtered .MRC volume and its corresponding .PY file.

To open this data you will need to download and install Chimera. https://www.cgl.ucsf.edu/chimera/

To view the data and the model, open Chimera and use it to open the .PY file within one of the directories. It should load the corresponding .MRC and display the placement of each centroid, axis and plane within it.