During the last decade major efforts have been made to develop synthetic reagents that bind sequence specifically (or selectively) to double stranded DNA thereby in principle being mimics of DNArecognizing proteins such as gene repressors or transcription factors. Apart from the scientific importance of such reagents for understanding the molecular chemistry of DNA recognition, sequence specific DNA-recognizing ligands have obvious potentials for the development of gene-targeted drugs and molecular biology tools (2).

Most success in this area has been obtained using oligonucleotides that bind to double-stranded DNA by triple helix formation (Figure 1)(2-7), and although a general solution for the problem of recognizing mixed purine-pyrimidine sequences is yet to come, major progress has been made in recent years (8-23). We decided to try exploitingthe “triple helix principle” for constructing a fully synthetic DNA-recognizingligand which would rely on hydrogen bonding/recognition by heterocycles like the natural nucleobases. Especially, we wished to construct an oligomeric type of reagent in which the heterocycles were connected via a backbone not being composed of deoxy ribose phosphate esters. Thus, in essence we aimed at making a “DNA analogue" in which the backbone had been replaced by one which from a synthetic point of view (ease of synthesis and synthetic flexibility) would be more amicable. The new backbone was designed using computer model building combined with chemical “common sense”(24-25) to be structurally homomorphous to the deoxyribose phosphate backbone, synthetically accessible, and ame-nable to automated assembly synthesis. We decided to rely on the Merrifield solid-phase synthesis of peptides using the terf-butyloxycarbonyl (Boc) protection strategy. The resulting target molecule is composed of a backbone containing 2V-(2-aminoethyl) glycine units in which the nucleobase is attached tothe glycine nitrogen via a methylene carbonyl linker (Figure 2)(24-29). Since we consider these molecules as chimera between nucleic acids (the nucleobases) and (pseudo) peptides (the backbone) we termed them peptide nucleic acid (PNA). We ac-knowledge that this is not strictly a chemically correct name, since PNA molecules are neither acids nor natural peptides. However, with the name we wished to emphasize that peptide chemistry is usedfor the oligomerization and that PNA is a very close analogue of nucleic acids. Some colleagues have suggested the name“polyamide nucleic acid analogue” be more appropriate. However, we still