The mechanism of histone H1 cross-linking by poly(ADP-ribosylation). Reconstitution with peptide domains.

One of the major products of the poly(ADP-Rib) polymerase reaction is the H1 dimer, formed by cross-linking two H1 molecules with a 15-unit poly(ADP-Rib) chain. In the present study we have attempted to characterize those regions of H1 molecules which participate in cross-linking, and in particular, to determine whether the reaction involves two NH2-terminal domains, two COOH-terminal domains, or a combination of these two. When we used reconstituted chromatin, formed by the addition of various H1 peptides to H1-depleted chromatin, we observed poly(ADP-ribosylation) in those complexes containing fragments of the NG-H1 domain (i.e. NH2-terminal tail plus the globular (G) region). Furthermore, with this material, increasing NAD concentration and incubation time led to a progressive elongation of poly(ADP-Rib) on the peptide approaching a chain length of 11-12 units. Reconstituted chromatin that contained other H1-derived peptides, such as C-H1 and G-H1 (i.e. only carboxyl or only globular domains respectively), were not poly(ADP-ribosylated). CG-H1 was noted to be a substrate in the reconstituted system. However, in contrast to NG-H1, where extensive chain elongation was observed, only very short chain, probably mono(ADP-ribosylation) occurred on this H1 peptide. In a complementary approach, the H1 dimer complex or its precursors were cleaved with N-bromosuccinimide, and peptide fragments isolated. This analysis confirmed that both the NH2- and COOH-terminal domains of H1 are poly(ADP-ribosylated) and that elongation of the ADP-ribose chain occurred primarily at the NH2 terminus during H1 dimer synthesis.