One form of third-generation inhibitors under study comprises recombination intermediates with binding sites for methylation activators that are linked to methyltransferase traps like those found in second-generation inhibitors. These inhibitors are to be guided to the nucleus of the cell by linked proteins with nuclear localization signals, as shown schematically in Fig. 6. The design parameters of such an inhibitor are given in Fig. 7. To construct this machine, the NLS-M*£coRII fusion was cloned and the fusion protein was purified as described in Subheading 2.4. Final assembly was monitored by mobility shift analysis using a Bioanalyzer 2100 and DNA analysis microfluidics chips. As can be seen in Fig. 2, each step in the final assembly process could be monitored by the appearance of the expected intermediate in the microfluidics analysis. The data clearly show that methyltransferase-directed nucleoprotein
Fig. 5. (previouspage) Nucleic acid-based inhibitor action. (A) Saturation plots for hemimethylated DNA and non-CG methylated DNA. Although each substrate carries methyl groups at cytosines on only one strand, the hemimethylated substrate (dCS1) targets hDnmtl to a single cytosine residue. The second substrate (NdCS2) mistargets the enzyme to dG, dA, and dT residues as well. Based on the two-substrate reaction scheme on the right, increasing the concentration of (NdCS2) at a constant concentration of (dCS1) is expected to inhibit the enzyme. (B) Inhibition of human DNA methyltransferase by singly and multiply mistargeted substrate analogs. The active substrate (dCS1) gives rapid incorporation of 3H-methyl groups from AdoMet into DNA (normalized to 100% activity). This ongoing reaction was inhibited by increasing the amounts of a second substrate, NdCS2.
assembly can be used to produce a machine for inhibition of human DNA methyltransferase based on a bionanotechnological design (see Fig. 2, Y-Junc-tion + M^EcoRII lane). The design is now being tested for its efficacy in the inhibition of hDnmt1 in vitro and in vivo.
More sophisticated designs based on arrayed protein signals or binding sites are possible. Designs of this type, however, would require four-arm or X-Junc-tions in order to accommodate the arrayed functions. A detailed discussion of these designs is beyond the scope of this chapter.
Was this article helpful?