Pre-steady state partitioning analysis of the HhaI DNA methyltransferase directly demonstrates the catalytic competence of the enzyme⅐DNA complex and the lack of catalytic competence of the enzyme⅐S-adenosyl-L-methionine (AdoMet) complex. The enzyme⅐AdoMet complex does form, albeit with a 50-fold decrease in affinity compared with the ternary enzyme⅐AdoMet⅐DNA complex. These findings reconcile the distinct binding orientations previously observed within the binary enzyme⅐AdoMet and ternary enzyme⅐S-adenosyl-Lhomocysteine⅐DNA crystal structures. The affinity of the enzyme for DNA is increased 900-fold in the presence of its cofactor, and the preference for hemimethylated DNA is increased to 12-fold over unmethylated DNA. We suggest that this preference is partially due to the energetic cost of retaining a cavity in place of the 5-methyl moiety in the ternary complex with the unmethylated DNA, as revealed by the corresponding crystal structures. The hemi-and unmethylated substrates alter the fates and lifetimes of discrete enzyme⅐substrate intermediates during the catalytic cycle. Hemimethylated substrates partition toward product formation versus dissociation significantly more than unmethylated substrates. The mammalian DNA cytosine-C-5 methyltransferase Dnmt1 shows an even more pronounced partitioning toward product formation.

We present the first description of KmDNA, KdDNA, Kcat, and Kmethylation for a mammalian DNA methyltransferase. Homogeneous, 190 000 MTDNA (cytosine-5-)-methyltransferase isolated from mouse erythroleukemia cells has turnover constants of 0.15-0.59 h-1 with single-stranded and unmethylated double-stranded oligonucleotides containing a single CpG dinucleotide. These substrates were designed to mimic DNA transcriptional cis elements previously reported to have cytosine C-5-methylated regulation. The rate-limiting step for these substrates is the methylation step itself. In contrast, hemimethylated double-stranded substrates show burst kinetics, consistent with a rapid methylation event (3 h-1) followed by a slower step which determines steady-state Kcat. Hemimethylated and unmethylated double-stranded DNA shows similar binding affinities; these results reveal the molecular basis for the enzyme's preference for hemimethylated DNA to be the methyl transfer step. Substrates with multiple recognition sites do not show burst kinetics and have turnover rate constants of 6 h-1. Catalytic turnover for the mammalian enzyme is thus approximately 10-fold slower than that for the related bacterial enzymes. Our combined results show quantitatively that one enzyme is certainly capable of both maintenance and de novo methylation and that maintenance of the genomic methylation pattern is preferred over the de novo establishment of new patterns. Direct comparison of the mammalian enzyme with the bacterial DNA cytosine-C5 methyltransferase, M.SssI, indicates dramatic differences in preferences for single-stranded, double-stranded, and hemimethylated double-stranded substrates. Moreover, the specificity hierarchy shown for the M.SssI is derived from very different changes in K(m) and catalysis than those observed for the mammalian DCMTase. These results demonstrate that the M.SssI, and perhaps other DNA cytosine methyltransferases from bacteria, is functionally dissimilar to the mammalian enzyme.

The major DNA cytosine methyltransferase isoform in mouse erythroleukemia cells, Dnmt1, exhibits potent dead-end inhibition with a single-stranded nucleic acid by binding to an allosteric site on the enzyme. The previously reported substrate inhibition with doublestranded substrates also involves binding to an allosteric site. Thus, both forms of inhibition involve ternary enzyme-DNA-DNA complexes. The inhibition potency of the single-stranded nucleic acid is determined by the sequence, length, and most appreciably the presence of a single 5-methylcytosine residue. A single-stranded phosphorothioate derivative inhibits DNA methylation activity in nuclear extracts. Mouse erythroleukemia cells treated with the phosphorothioate inhibitor show a significant decrease in global genomic methylation levels. Inhibitor treatment of human colon cancer cells causes demethylation of the p16 tumor suppressor gene and subsequent p16 re-expression. Allosteric inhibitors of mammalian DNA cytosine methyltransferases, representing a new class of molecules with potential therapeutic applications, may be used to elucidate novel epigenetic mechanisms that control development.

DNA (cytosine-5-)-methyltransferase is essential for viable mammalian development and has a central function in the determination and maintenance of epigenetic methylation patterns. Steady-state and substrate trapping studies were performed to better understand how the enzyme functions. The catalytic efficiency was dependent on substrate DNA length. A 14-fold increase in KmDNA was observed as the length decreased from 5000 to 100 base pairs and kcat decreased by a third. Steady-state analyses were used to identify the order of substrate addition onto the enzyme and the order of product release. Double-reciprocal patterns of velocity versus substrate concentration intersected far from the origin and were nearly parallel. The kinetic mechanism does not appear to change when the DNA substrate is either 6250 or 100 base pairs in length. Isotope trapping studies showed that the initial enzyme-AdoMet complex was not catalytically competent; however, the initial enzyme-poly(dI.dC-dI.dC) complex was observed to be competent for catalysis. Product inhibition studies also support a sequential ordered bi-bi kinetic mechanism in which DNA binds to the enzyme first, followed by S-adenosyl-L-methionine, and then the products S-adenosyl-L-homocysteine and methylated DNA are released. The proposed mechanism is similar to the mechanism proposed for M. HhaI, a bacterial DNA (cytosine-5-)-methyltransferase. Evidence for an enzyme-DNA-DNA ternary complex is also presented.

No abstract