Nicotinamide-Adenine Dinucleotide-Related Enzymes in Tumour-Cell Nuclei

with DNA. In no case is the detailed conformation of a histone in its complex with DNA known but several lines of evidence (Simpson, 1970) indicate that the major groove of DNA is a probable binding site. A possible model for the binding of the N-terminal region of histone IV has been proposed (Sung & Dixon, 1970; Shih & Bonner, 1970) in which the first 18 residues from the N-terminus are in an a-helix occupying the major groove of DNA a d the positive changes of the four lysine residues 5, 8, 12 and 16 bind to a series of four phosphate groups on one strand of the DNA. However, Boublik et nf. (1970) estimated the probability of a-helix formation in various regions of histone IV in solution by arranging the known sequences according to the helical wheel con’formation of Schiffer & Edmundson (1967). Using Prothero’s (1966) rule, they came to the conclusion that the N-terminal basic region (residues 1-36) had a very low probability of a-helix formation due to charge repulsion between the positively charged lysine and arginine groups. Thus, the spontaneous formation of a significant fraction of a-helix in the N-terminal region of free, unmodified histone IV seems unlikely. However, theenzymic acetylation of lysine residues 5,8,12 and 16 would greatly increase the probability of a-helix formation i n this region, since the main obstacle to helix formation, namely, charge repulsion of the cationic lysine residues, would disappear. Thus the observations reported here that newly synthesized histone IV must pass through a series of acetylated forms is certainly consistent with the notion that these modifications are necessary to enable a particular conformation, in this case an a-helix, to form before its correct binding to the major groove of DNA. Sequential deacetylation could then ‘lock’ histone IV into place by regeneration of the positive charges of lysine residues 5,8,12 and 16 so as to allow tight ionic interactions with four DNA phosphates.