The authors' approach to the design of DNA polymerase-specific inhibitors, an approach based on the mechanism of action of 6-(p-hydroxyphenylazo)uracil, has been to disguise nucleic acid bases to mimic the purine substrates dGTP and dATP. Specifically, the strategy has been to synthesize bases with substituents that endow them with the capacity: to seek and react with unique features of the active site of a polymerase; and to form H bonds with complementary template pyrimidines apposing the active site. This strategy has yielded a series of novel, enzyme-specific dATP and dGTP analogues which are non-polymerizable and which inhibit their target polymerase by sequestering it to a complementary pyrimidine residue in primer:template. The work has involved primarily two replication-specific polymerases, B. subtilis DNA polymerase III (pol III) and mammalian DNA polymerase alpha (pol alpha). The initial design exploited the pyrimidine nucleus and produced inhibitors with Ki values in the micromolar range. Principles established with the pyrimidine derivatives have led to the development of bona fide purine nucleotide analogues which act as DNA polymerase inhibitors of high selectivity and unprecedented potency. For example, BuPdGTP, the 2'-deoxyribonucleoside 5'-triphosphate of N2-(p-n-butylphenyl)guanine (BuPG), lacks discernible activity against mammalian polymerases beta and gamma, whereas it inhibits mammalian pol alpha with a Ki of less than 10 nanomolar. Currently, the authors are exploiting BuPdGTP, BuPdGDP, and similar butylanilino derivatives of dATP to probe the active site of pol alpha and to develop other N2-substituted analogues which can bind selectively to the substrate sites of other important polymerases and nucleotide binding proteins.