The C-terminal region of HMG-17 hinders the acetylation of nucleosomal H3. A gel assay was used to monitor the acetylation of nucleosomal H3 as a function of added HMG-17 (●), HMG-14 (○), or HMG17ΔC37 (▵, a truncation mutant of HMG-17 lacking 37 residues from the C terminus). HMG-14 and HMG-17 inhibited H3 acetylation; in contrast, the C-terminal truncation mutant of HMG-17 was a poor inhibitor of nucleosomal H3 acetylation.

Specificity of the acetylation reaction with HMG-17 truncation mutants. Panel A shows the protein gel; panel B is the corresponding phosphorimage showing the incorporation of [¹⁴C]acetate. Substrates are as indicated (HMG-17ΔC37, an HMG-17 truncation mutant lacking 37 residues from the C terminus; HMG-17ΔN16, an HMG-17 truncation mutant lacking 16 residues from the N terminus; HMG-17ΔC22, an HMG-17 truncation mutant lacking 22 residues from the C terminus; HMG-14ΔC26, an HMG-14 truncation mutant lacking 26 residues from the C terminus; and HMG-17ΔN4, an HMG-17 truncation mutant lacking the first 4 residues from the N terminus).

Specific acetylation of HMG-17 by PCAF. The specificity of PCAF acetylation was tested by using several substrates. Panel A shows the protein gel, and panel B is the corresponding phosphorimage showing the incorporation of [¹⁴C]acetate. Substrates are as indicated (Hist.H3, histone H3; Nucl., chicken nucleosome core particles; Hist.H1, calf thymus histone H1; and Cyt. C, cytochrome c). The arrow to the right of each panel designates the position of PCAF.

HMG-17 inhibits the acetylation of nucleosomal H3. Nucleosomes reconstituted with various amounts of HMG-17 were acetylated by PCAF, and the proteins were resolved on SDS-polyacrylamide gels. The incorporation of [¹⁴C]acetate into each protein was determined by phosphorimaging. The figure depicts the decrease in specific activity of nucleosomal H3 as a function of added HMG-17. The inset shows the protein gel (left panel) and the corresponding phosphorimage (right panel) showing ¹⁴C incorporation either into the nucleosomes alone (−) or into nucleosomes containing stoichiometric amounts of HMG-17 (+).

Mass spectrum identification of the acetylated lysine in HMG-17. (A) Mass analysis of the full-length acetylated HMG-17. The peak designated HMG-17 is the full-length unmodified protein (m/z = 9,261), and the peak designated AcHMG-17 corresponds to HMG-17 with a single modification (one acetyl group, total m/z = 9,308). (B) Mass analysis of the Glu-C peptide (generated from the full-length acetylated species) containing the modification and identification of the modified lysine (position 2). (C) Region of the mass spectra containing the modified lysine. The mass analysis from panel B was performed from the spectra depicted in panel C.

Determinants for the specific acetylation of HMG-17. (A) The acetylation of HMG-17 is inhibited by a deletion mutant lacking the first 4 amino acids but is not inhibited by a deletion mutant lacking the first 16 amino acids. In these experiments the acetylation of HMG-17 was performed in the presence of 0.2 μg of competitor per μl. (B) Schematic diagram summarizing the studies indicating that residues 5 to 16 and a peptide longer than 52 residues of HMG-17 are required for the specific acetylation of HMG-17. Acetylation and competition studies were done as described in Materials and Methods. The designations for the truncation mutants are as indicated in Fig. 2. Three peptides were used: p14N11, a peptide containing the first 11 residues of HMG-14 (contains the conserved Lys-2); p17N16, a peptide containing the first 16 residues of HMG-17 (including the acetylated Lys-2); and p1752-67, a peptide containing residues 52 to 67 of HMG-17 (residues included in the acetylated HMG-17ΔC22 mutant but absent in the HMG-17ΔC37 mutant that is not a substrate). The HMG-17ΔC22 mutant was not tested (ND) as a competitor since it can function as a substrate. The HMG-17ΔC37 mutant was not tested as a competitor since it contains a fully intact N terminus, including the conserved acetylated Lys-2, yet it cannot function as a substrate. It is also indicated whether the substrate could (+) or could not (−) function as a substrate for acetylation or as a competitor for the acetylation of HMG-17. The boxes in the diagram indicate the evolutionarily conserved domains in HMG-14/-17. NA, not applicable.

Acetylation of HMG-17 reduces its binding affinity to nucleosome cores. Equilibrium dialysis (described in Materials and Methods) was used to determine the relative binding affinities of acetylated versus unacetylated HMG-17 to nucleosomes. (A) A mixture of acetylated and unacetylated HMG-17 (set 1, control, 80 ng/ml; set 2, 80 ng/μl; set 3, 40 ng/μl; and set 4, 20 ng/μl) were placed in one chamber, and the adjoining chambers were filled with either buffer (2× TBE [set 1, control]) or chicken nucleosome cores (60 ng/μl [sets 2 to 4]). The inset shows the quantity of HMG-17 in the control chambers of set 1 (with no nucleosome added). The specific activity of the HMG-17 in these control chambers is shown as set 1, columns a and b. Note that the specific activity in both chambers was identical. The specific activity of HMG-17 in each of the two chambers in the remaining sets is given as either bound (B) or free (F). Note that the specific activity of the free form was always greater than that of the bound form, indicating that the unmodified HMG bound with greater affinity. The ratio, F/B, is an indication of the increase in specific activity of the unbound HMG-17. (B) Nucleoprotein gel assay showing that heat-treated (95°C, 5 min) HMG-17 bound to core particles (Cp) in a manner that is indistinguishable from the untreated HMG-17. The samples were mixed on ice in the acetylation buffer and contained 0.43 μg of the nucleosome core particles per ml and the indicated amounts of HMG-17. Labels: Cp, core particle with no HMG-17 added; Cp+1HMG, core particle with one HMG-17 bound; and Cp+2HMG, saturated complex of core particle with two bound HMG-17s.