ADP-ribosylation of histone H2B. Identification of glutamic acid residue 2 as the modification site.

The site of in vitro ADP-ribosylation of histone H2B was determined. From chromatin of rat liver incubated with [ribose(NMN)-14C]NAD (210 pCi), the histone H2B fraction was extracted with acid and purified by column chromatographies on Bio-Gel P-60 and Sephadex G-150. The final preparation (96 mg of protein), which consisted of ADP-[*4C]ribosylated (1.8 pCi) and nonADP-ribosylated histone H2B’s at a ratio of 1:12, gave a single protein band on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The bond between ADP-ribose and histone H2B was labile in neutral NHzOH and mild alkali, suggesting an ester-like nature. Digestion of ADP-ribosyl histone H2B with trypsin and snake venom phosphodiesterase produced a single radioactive peptide material. The material was purified by sequential chromatographies on Sephadex 6-25, Dowex 1, and Bio-Gel P-2 in a 42% yield of “C. The amino acid sequence of the peptide was Pro-Glu-ProAla-Lys, which was identical with the reported sequence of NHz-terminal residues 1 to 5 of calf thymus and other histone H2B’s. Further digestion of this material with acid protease €3 and carboxypeptidase Y yielded a tripeptide, Pro-Glu-Pro, associated with a stoichiometric amount of ribose 5-phosphate. The properties of the linkage between ribose 5-phosphate and the tripeptide was also labile in neutral NH20H and mild alkali. The electrophoretic mobilities of the dephosphorylated derivatives (ribosyl pentapeptide and ribosyl tripeptide) at pH 6.5 indicated a blockage of the y-carboxyl group of glutamic acid residue 2. All of these results suggested that histone H28 was ADP-ribosylated at this residue through an ester bond.