Differential histone tail citrullination by PAD Enzymes observed via NMR spectroscopy

Figures

• 

  Figure 1.. Utilizing real-time 1H-15N NMR spectroscopy to identify sites of citrullination by each PAD isozyme. (A) Structure of the nucleosome core particle (NCP) consisting of a histone octamer wrapped in 147 base-pairs of Widom 601 DNA (88). (B) Partial sequences of histones H2A, H2B, H3, and H4 with arginines in bold and the intrinsically disordered tail regions underlined. (C) Schematic depicting PAD-catalyzed citrullination of protein arginine residues. (D) Overview of the NMR-based citrullination assay. An initial spectrum is acquired of the 15N-labeled unmodified substrate, specifically H3(1-44) with R40 and R42 shown in this example. PAD isozyme is then added to the substrate, and spectra are acquired periodically as the reaction progresses. If an arginine is modified, the peak decays as a citrulline peak emerges.

• 

  Figure 2.. Spectra of arginines and citrullines in 15N-labeled histone tail peptides and reconstituted NCPs by PAD1. Top, left to right: 1H-15N SOFAST-HMQC spectra of 15N-labeled histone tail peptides of H3 (blue), H4 (green), H2A (yellow), and H2B (red) with their PAD1-citrullinated end-states in dark grey. Bottom, left to right: spectra of NCPs reconstituted with 15N-labeled histone H3 (blue), H4 (green), H2A (yellow), and H2B (red) with their PAD1-citrullinated end-states in dark grey. Spectra were acquired on either 100 μM tail peptide or 50 μM NCP (equivalent to 100 μM 15N-labeled histone) on an 800 MHz Bruker NMR spectrometer at 10°C (tail peptide) or 25°C (NCP) in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. Full spectra are shown in Figure S2.

• 

  Figure 3.. Progress curves of H3 peptide arginines and citrullines demonstrate site-specific differences in PADs 1, 2, and 4 versus PAD3. Top: Cartoon of unmodified H3 peptide (left) and 15N-H3 peptide arginine progress curves for PADs 1, 2, 3, 4 (right). Arginine progress curves are denoted by the shapes found above the H3(1-44) peptide sequence (middle). Black and blue residues represent unobserved and observed residues, respectively. Bottom: Cartoon of citrullinated H3 tail peptide (left) and graphs of 15N-H3 peptide citrulline progress curves for PADs 1, 2, 3, 4 (right). Citrulline progress curves are denoted by the shapes found immediately below the H3 peptide sequence. Data were collected with 100 μM 15N-H3(1-44) on an 800 MHz Bruker NMR Spectrometer at 10°C in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. PADs were added to a final concentration of 200 nM.

• 

  Figure 4.. Increasing arginine intensities after PAD addition suggest citrullination affects nucleosomal H3 tail dynamics. Top: Cartoon of unmodified H3-NCP (left) and graphs of H3-NCP arginine progress curves for PADs 1, 2, 3, 4 (right). Arginine progress curves are denoted by the shapes found above the H3 sequence (middle). Black and blue residues represent unobserved and observed residues, respectively. Bottom: Cartoon of citrullinated 15N-H3-NCP (left) and 15N-H3-NCP citrulline progress curves for PADs 1, 2, 3, 4 (right). Citrulline progress curves are denoted by the shapes found immediately below the H3 sequence. Please note differences in arginine vs. citrulline y-axis scaling. Data were collected with 50 μM 15N-H3-NCP (equivalent to 100 μM 15N-labeled histone) on an 800 MHz Bruker NMR spectrometer at 25°C in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. PADs were added to a final concentration of 200 nM.

• 

  Figure 5.. Progress curves of H4 tail peptide citrullination further highlight differences in histone modification sites by PADs 1, 2, and 4 versus PAD3. Top: Cartoon of unmodified H4 peptide (left) and graphs of 15N-H4 peptide arginine progress curves for PADs 1, 2, 3, 4 (right). Note that R17 intensity is influenced by the nearby growth of Cit23 (see Figure 2, top, for H4 tail peptide arginine and citrulline chemical shifts). Arginine progress curves are denoted by the shapes found above the H4(1-33) T30Y peptide sequence (middle). Black and green residues represent unobserved and observed residues, respectively. Bottom: Cartoon of citrullinated H4 peptide (left) and graphs of H4 peptide citrulline progress curves for PADs 1, 2, 3, 4 (right). Citrulline progress curves are denoted by the shapes found immediately below the H4 peptide sequence. Data were collected with 100 μM 15N-H4(1-33) T30Y on an 800 MHz Bruker NMR spectrometer at 10°C in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. PADs were added to a final concentration of 200 nM.

• 

  Figure 6.. Citrullination resolves the basic patch region of the nucleosomal H4 tail. Top: Cartoon of unmodified H4-NCP (left) and progress curves of the sole observable arginine in 15N-H4-NCP for PADs 1, 2, 3, 4 (right). Arginine progress curves are denoted by the shapes found above the H4 sequence (middle). Black and green residues represent unobserved and observed residues, respectively. Citrullination of the NCP resolves the H4 tail from A15 to K20 (see Figure S2). Bottom: Cartoon of citrullinated H4-NCP (left) and graphs of 15N-H4-NCP citrulline progress curves for PADs 1, 2, 3, 4 (right). Citrulline progress curves are denoted by the shapes found immediately below the H4 sequence. Please note differences in arginine vs. citrulline y-axis scaling. Data were collected with 50 μM 15N-H4-NCP (equivalent to 100 μM 15N-labeled histone) on an 800 MHz Bruker NMR spectrometer at 25°C in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. PADs were added to a final concentration of 200 nM.

• 

  Figure 7.. Broad citrullination of the H2A tail peptide is observed for all PADs. Top: Cartoon of unmodified H2A peptide (left) and progress curves of 15N-H2A peptide arginines with PADs 1, 2, 3, 4 (right). Arginine progress curves are denoted by the shapes found above the H2A(1-28) F25Y peptide sequence (middle). Black and yellow residues represent unobserved and observed residues, respectively. Bottom: Cartoon of citrullinated H2A peptide (left) and citrulline progress curve graphs of 15N-H2A peptide for PADs 1, 2, 3, 4 (right). Citrulline progress curves are denoted by the shapes found immediately below the H2A peptide sequence. Data were collected with 100 μM 15N-H2A(1-28) F25Y on an 800 MHz Bruker NMR spectrometer at 10°C in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. PADs were added to a final concentration of 200 nM.

• 

  Figure 8.. The sole visible arginine in H2A-NCP is citrullinated by PADs 1-4. Top: Cartoon of unmodified H2A-NCP (left) and progress curve graphs of the sole observable arginine in 15N-H2A-NCP for PADs 1, 2, 3, 4 (right). Arginine progress curves are denoted by the shapes found above the H2A sequence (middle). Black and yellow residues represent unobserved and observed residues, respectively. Citrullination of the NCP resolves the H2A tail through A10 (for PADs 2-4) or Cit11 (for PAD1) (see Figure S2). Bottom: Cartoon of H2A-NCP (left) and citrulline progress curves of 15N-H2A-NCP for PADs 1, 2, 3, 4 (right). Citrulline curves are denoted by the shapes found immediately below the H2A sequence. Please note differences in arginine vs. citrulline y-axis scaling. Data were collected with 50 μM 15N-H2A-NCP (equivalent to 100 μM 15N-labeled histone) on an 800 MHz Bruker NMR spectrometer at 25°C in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. PADs were added to a final concentration of 200 nM.

• 

  Figure 9.. PADs 1, 2, and 4 citrullinate H2B peptide arginines, but minimal citrullination is observed by PAD3. Top: Cartoon of unmodified H2B peptide (left) and 15N-H2B peptide arginine progress curves for PADs 1, 2, 3, 4 (right). Progress curves are denoted by the shapes found above the H2B(1-39) peptide sequence (middle). Black and red residues represent unobserved and observed residues, respectively. Bottom: Cartoon of citrullinated H2B peptide (left) and citrulline progress curve graphs of 15N-H2B peptide for PADs 1, 2, 3, 4 (right). Citrulline progress curves are denoted by the shapes found immediately below the H2B peptide sequence. Data were collected with 100 μM 15N-H2B(1-39) on an 800 MHz Bruker NMR spectrometer at 10°C in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. PADs were added to a final concentration of 200 nM.

• 

  Figure 10.. Citrullination of adjacent histones affects H2B lysine acetylation sites in the nucleosome. Top: H2B sequence. Black and red residues represent unobserved and observed residues, respectively. Left: cartoon of the H2B-NCP. Right, middle: Graphs of raw peak intensities of K5, K11, K16, K20, K27 as a function of time following PAD addition. Right, bottom: Graphs of normalized intensities (with respect to t=0) of K5, K11, K16, K20, K27 for PADs 1, 2, 3, and 4. The greatest relative intensity change is sensed closest to the core, at K27, suggesting the H2B tail becomes more dynamic in the NCP when adjacent histones are citrullinated. Data were collected with 50 μM 15N-H2B-NCP (equivalent to 100 μM 15N-labeled histone) on an 800 MHz Bruker NMR spectrometer at 25°C in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. PADs were added to a final concentration of 200 nM.

• 

  Figure 11.. Assessing PAD site-specificity by determining time to 50% completion in the context of histone peptides and NCP. Time to 50% completion (t50%) graphs for PADs 1, 2, 3, and 4 (top to bottom) with each of the histone peptide (A) or NCP (B) residues. Arginines (left bar) and citrullines (right bar) are in colors and greyscale, respectively, as found in the progress curve graphs in Figures 3–9. “X” indicates that a t50% was not determined for the following reasons: †For H3 peptide with PAD3, the Cit17 t50% was calculated to be a negative time value due to a minor decrease in intensity. ‡H4 tail peptide R17 could not be fit for any of the PADs due to overlap with Cit23. §H3 R8/17/26 in NCP could not be fit reliably due to non-monotonic progress curves. ‖Residues R17/19 in 15N-H4-NCP and R11 in 15N- H2A-NCP were unobserved in the initial spectra and thus unable to be fit for calculations. All data were collected on an 800 MHz Bruker NMR spectrometer in 20 mM MOPS pH 7.0, 100 mM KCl, 2 mM CaCl2, 0.5 mM TCEP, 0.1 mM EDTA, and 5% D2O. Experiments with histone peptide substrate were run at 10°C with 100 μM 15N-labeled peptide. Experiments with NCP substrate were run at 25°C with 50 μM NCP…

• 

  Figure 12.. Summary of PAD histone citrullination sites in nucleosomes and proposed models for effects of nucleosome citrullination. A. PADs 1, 2, and 4 largely citrullinate all observable and accessible arginines within the N-terminal intrinsically disordered H3, H4, and H2A histone tails. In contrast, PAD3 only citrullinates H4 R3 and H2A R3. B. The histone tails dynamically interact with nucleosomal DNA, predominantly existing in inaccessible tail ensembles (top, left). PAD citrullination shifts the histone tail equilibria toward accessible ensembles (top, right). (Note that the cartoons are simplified depictions and are not intended to capture all conformations.) Citrullination of nucleosomal histone tails by the PADs may affect the dynamics of a given tail i. directly, by reducing the charge of the tail through successive citrullination events (bottom, left), or ii. indirectly, by citrullinating adjacent histones (bottom, right).