The primary goal of this study is to chemically refine a class of sequence- specific, DNA binding, triple helix-forming oligonucleotides (TFOs). the unique feature of these TFOs is that binding of the third strand to its duplex DNA target is pH independent and is stabilized by GGC and TAT triplets such that the preferred orientation places the bound TFO antiparallel with respect to more purine rich strand of the underlying duplex. Our initial studies with a 38 base long oligonucleotide (HIV38p), designed to bind to the promoter region in the long terminal repeat sequence of human immunodeficiency virus (HIV), indicated that HIV38p has a tendency to self-associate due to high G content. Subsequent studies with HIV38p have shown that substitution of G with 2'-deoxy-7- deazaguanosine residues reduces the self-association considerably. Thus, substitution of some or all guanine residues in HIV38p with selected guanine analogs may reduce or eliminate self-association and improve triplex formation. We now propose to prepare TFOs containing various levels of substitutions ranging from 10 to 50% with 2'-deoxy-7-deazaguanosine or 2'-deoxy-6- thioguanosine employing solid-phase phosphoramidite chemistry. The modified oligonucleotides thus obtained will be purified and adequately characterized. The ability of these novel TFOs to form stable triplexes will be evaluated using band shift analysis, DNase footprinting experiments, melting curves and CD measurements. It is hoped that such a substitution improves triple helix structure, reduces self-association, increases binding affinity and antiviral (HIV and HSV) activity. Detailed biological effects of such modified TFOs will be assessed in Phase II studies.