Oligodeoxyribonucleotides covalently linked to an intercalating agent via a polymethylene linker were synthesized. Oligothymidylates attached to an acridine dye (Acr) through the 3'-phosphate group [(Tp),(CH2)mAcr] specifically interact with the complementary sequence. The interaction is strongly stabilized by the intercalating agent. By using absorption and fluorescence spectroscopies, it is shown that complex formation between (Tp),(CH2)mAcr and poly(rA) involves the formation of n A-T base pairs, where n is the number of thy. mines in the oligonucleotide. The acridme ring intercalates between ANT base pairs. Fluorescence excitation spectra reveal the existence of two environments for the acridine ring, whose relative contributions depend on the linker length (m). The binding of (Tp)4(CH2)mAcr to poly(rA) is analyzed in terms of site binding and cooperative interactions between oligonucleotides along the polynucleotide lattice. Thermodynamic parameters show that the covalent attachment of the acridine ring strongly stabilizes the binding of the oligonucleotide to its complementary sequence. The stabilization depends on the linker length; the compound with m = 5 gives a more stable complex than that with m = 3. These results open the way to the synthesis of a family of molecules exhibiting both high-affinity and high-specificity for a nucleic acid base sequence.Molecules with high affinity and base-sequence specificity are required to control gene expression at different levels. Several possibilities may be contemplated depending on whether double-stranded or single-stranded nucleic acids are chosen as targets. For example, the regulation of transcription involves proteins that recognize specific duplex DNA sequences in both prokaryotes and eukaryotes (1, 2). However, there is no general rule yet established that could allow us to build an oligopeptide sequence aimed at recognizing a given base or base pair sequence (3). Moreover, the amino acid side chains that are involved in contacts with base or base pairs are not contiguous in the polypeptide chain.The most obvious candidate to allow for the specific recognition of a nucleic acid fragment is an oligonucleotide with the complementary sequence, provided the nucleic acid bases in the target sequence have their hydrogen bonding sites available. This is obviously so if the target nucleic acid region exists as a single-stranded structure. This might also be true in a duplex structure if a sufficient supplementary energy of interaction were provided either by modifying the oligonucleotide (see below) or by imposing constraints on the duplex structure (e.g., in superhelical DNA or upon binding of melting proteins). Also it must be kept in mind that important regulatory regions in transcription and replication must be transiently opened as observed-e.g., in the complex formed by bacterial RNA polymerase with a promoter (the so-called "open complex") (4) or at the replication fork (5). Actively transcribed genes in eukaryotes also possess regions upstream f...