PHAGE-T4 DNA [N-6-ADENINE] METHYLTRANSFERASE - KINETIC-STUDIES USING OLIGONUCLEOTIDES CONTAINING NATIVE OR MODIFIED RECOGNITION SITES

The DNA-[N-6-adenine] methyltransferase of T4 phage (T4 Dam MTase) cat alyzes methyl group transfer from S-adenosyl-L-methionine (AdoMet) to the N-6-position of adenine in the palindromic sequence, GATC, We have investigated the effect of eliminating different structural component s of the recognition site on the ability of a substrate to be bound an d methylated by T4 Dam, For this purpose, steady state binding (by gel shift assays) and kinetic parameters of methylation (using the methyl donor, [H-3-CH3]-AdoMet, at 25 degrees C) were studied using various synthetic duplex oligonucleotides containing some defect in the DNA-ta rget site; e.g., the absence of an internucleotide phosphate or a nucl eotide(s) within the recognition site, or a single stranded region. Th e salient results are summarized as follows: (1) Addition of T4 Dam to a complete reaction mixture (with a 20-mer duplex as substrate) resul ted in a 'burst' of H-3-methylated product, followed by a constant rat e of product formation that reflected establishment of steady-state co nditions, This suggests that the rate-limiting step is release of prod uct methylated DNA from the enzyme [and not the transfer of the methyl group]. (2) A number of the defects in duplex structure had only a we ak influence on the binding and K-m values, but strongly reduced the k (cat). At the same time, several poorly bound duplexes retained good s ubstrate characteristics, especially duplexes having uninterrupted GAT -sequences in both strands. Whereas having only one half of the recogn ition site element intact was sufficient for stable complex formation, the catalytic turnover process had a strict requirement for an uninte rrupted GAT-sequence on both strands, (3) There was no correlation bet ween K-m and binding capability; the apparent K-d for some duplexes wa s 5-70 times higher than K-m. This indicates that the T4 Dam methylati on reaction can not be explained by a simple Michaelian scheme.