PARP1 covalently modifies all four core histone tails. (A) PARP1 trans-ADP-ribosylates histones isolated from calf thymus. An amount of 10 pmol recombinant PARP1 or PARP2 were incubated for 15 min (PARP1) or 30 min (PARP2) at 30°C with 5 pmol EcoRI-linker DNA, 100 nM radiolabeled NAD⁺ and 20 µg extracted histones from calf thymus. The reaction was stopped by addition of SDS-lysis buffer and the proteins were resolved on a 10–20% SDS–PAGE gradient gel. The gel was stained with Coomassie-R250 and incorporated ³²P was visualized by autoradiography. (B) Recombinant expressed and purified histones (3 µg) were ADP-ribosylated by PARP1 as in Figure 1A. (C) An amount of 1.5 µg of each purified GST-histone tail were used in PARP1 or PARP2 mediated trans-ADP-ribosylation reactions for 5 min at 30°C. (D) An amount of 1.5 µg of each purified GST-histone tail were used in PARP1 mediated trans-ADP-ribosylation reactions for 5 min at 30°C. Histone tails were either included during the reaction (pre) or added after the reaction had been stopped with a 100-fold excess of 3AB over radiolabeled NAD⁺ (post) to exclude non-covalent interaction of the histone tails with PAR. (E) Trans-ADP-ribosylation of the H2B tail is inhibited by the PARP-inhibitors PJ34 (0.01–100 µM) and DAM-TIQ-A (10 µM). (F) GST-histones were incubated with PARP1, EcoRI linker and increasing concentrations of NAD⁺ (0, 10, 100, 400 µM) for 10 min at 30°C. Poly(ADP-ribose) formation was assessed through western blotting with anti-PAR (LP-96-10) antibody. Unmodified GST-histone tails are marked by an empty asterisk, PARylated GST-histone tails are marked by a filled asterisk.

Mass spectrometric analysis of ADP-ribosylated histone peptides. (A) Extracted ion chromatogram of the biotin tagged H2A (aa 3–23) peptide incubated with 500 µM NAD⁺ and PARP1. The precursor masses of unmodified H2A peptide (2395.34 Da) and ADP-ribosylated H2A peptide (2936.40 Da) were plotted in a range of 10 ppm over time. ETD fragment spectrum of quintuply charged precursor ion of ADP-ribosylated H2A peptide (638.51 m/z) at K13, as indicated by the sequence plot. The spectrum shows next to the two major peaks of unfragmented charge reduced precursor ions (M+5H)ⁿ⁺ an almost complete series of N-terminal and C-terminal fragment ions (c-ions, z-ions, respectively). (B) Extracted ion chromatogram of the biotin tagged H2B (aa 18–37) peptide incubated with 500 µM NAD⁺ and PARP1 as in (A). ETD fragment spectrum of ADP-ribosylated H2B peptide (797.89 m/z) at K30, indicated by the sequence plot. (C) Extracted ion chromatogram of the biotin tagged H3 (aa 23–42) peptide incubated with PARP1 as in (A). ETD fragment spectrum of ADP-ribosylated H3 peptide (590.29 m/z) at K27, indicated by the sequence plot. (D) Extracted ion chromatogram of the biotin tagged H4 (aa 1–22) peptide incubated with 100 µM NAD⁺ and PARP1. ETD fragment spectrum of ADP-ribosylated H4 peptide (621.30 m/z) at K16, as indicated by the sequence plot. (E) Trans-ADP-ribosylation of the GST-H4 histone tail wild-type or K16A mutant by PARP1. (F) GST-pulldown of wild-type and mutated GST-histone H4 tails with recombinant PARP1. The GST-histone tails are marked by an asterisk.

ADP-ribosylation of the H4 peptide is impaired by H4K16 acetylation. (A) Elution profile of biotinylated H4 peptide (aa 1–22) and biotinylated H4K16ac peptide (aa 1–22) incubated with 100 μM NAD⁺ for 15 min without PARP1 (–PARP1) or in the presence of PARP1 (+PARP1) and subsequent ARH3 treatment. Acetone precipitated peptides were analyzed by LC–MS/MS using a C18 reversed-phase column and subsequent detection by mass spectrometry. (B) Histone H4 (aa 1–22) and acetylated H4K16ac (1–22) peptides were ADP-ribosylated with PARP1 for 15 min at 30°C with 100 nM ³²P-NAD⁺. The peptides were purified by microvolume-C18 reversed phase columns, eluted into scintillation liquid and counted for incorporated ³²P. Relative increase of counts per minutes was calculated over background (peptides added after termination of the reaction by 3AB). (C) Overview of the identified ADP-ribose acceptor sites within the amino-terminal core histone tails.

Confining the regions of putative ADP-ribose acceptor sites. (A) Schematic representation of the deletion strategy for GST-histone tails to identify the minimal ADP-ribosylated domain. (B–E) Trans-ADP-ribosylation of the indicated GST-histone deletion mutants by PARP1 with 100 nM ³²P-NAD⁺.

Histone H4 interacts with PARP1 by salt bridges with acidic residues in the catalytic domain. (A) Representative snapshot of the binding mode saved after 10 ns MD simulation of the H4 tetrapeptide AKRH. (B) Enlarged view of the catalytic cleft for the same snapshot as in (A). Amino acids in close proximity of the H4 peptide are highlighted. PARP1 and the H4 tetrapeptide are shown in surface render and sticks, respectively. The surface is colored according to atomic elements with carbon, oxygen and nitrogen in green, red, and blue, respectively. Carbon atoms of the H4 tetrapeptide are in cyan. (C) ADP-ribose was docked into the donor-site of PARP1 catalytic domain after 10 ns MD simulation of the H4 tetrapeptide. Amino acids in close proximity of the H4 peptide are highlighted. The orientation is rotated by about 180°C with respect to (A,B). (D) GST-pulldown of wild-type and mutated GST-histone H4 tail with recombinant PARP1 and subsequent western blot with anti-PARP1 antibody. GST-histone tails are indicated by an asterisk. (E and F) Trans-ADP-ribosylation of wild-type and mutated GST-H4 tails with PARP1 and radiolabeled NAD⁺. Coomassie stains of the input and autoradiographies are shown. (G) Trans-ADP-ribosylation of calf-thymus extracted histones by PARP1 wild-type or PARP1 D756K mutant as in Figure 1A.