Design, synthesis and biological evaluation of novel analogues of distamycin

Analogues of distamycin which retain the parent ligand's ability to bind in the DNA minor groove with sequence specificity are widely recognised as potential candidates for drug development. Distamycin binds to DNA with significant, albeit not complete, AT specificity, spanning five base pairs. Its pyrrole rings can partially accommodate the bulky guanine C2-NH2 groups in the minor groove especially at the edges of the binding site, thus leading to some tolerance of GC base pairs. A key objective of this project was to modify a portion of the distamycin structure by substituting its N-terminal pyrrole(-4yl)formamido unit with novel biaryl motifs in an attempt to enhance the AT specificity of the parent ligand. In order to achieve this, solution and solid phase combinatorial approaches were adopted and parallel libraries synthesised using a variety of heterocycles as building blocks. Solid phase combinatorial chemistry, performed on Mimotope Lanterns™, was employed to accelerate the synthesis of polyamides. A protocol for the assembly of the polyamide chain was successfully developed, and Library 3.1 and 3.2 compounds (3.1a-f and 3.2a-f) were prepared rapidly and efficiently. FRET-based DNA melting experiments revealed that some of these molecules are capable of stabilizing a DNA-hairpin oligonucleotide. Library members synthesized in solution phase (2.3a-g) were screened against a 512-member hairpin-DNA oligonucleotide library, containing all possible (non-degenerate) sequences in order to identify recognition events. This screen was based on the ethidium bromide displacement assay reported by Boger, and revealed that enhanced AT sequence selectivity had been achieved with some molecules. In particular, a compound containing the pyrazole-thiazole motif (2.3{5.16}) was found to bind to AT-rich sequences more selectively than distamycin. This result was confirmed by quantitative DNA footprinting experiments using two different DNA fragments in which 2.3{5.16} was shown to bind to TATA sequences with greater selectivity than distamycin, which bound to additional AT- containing sites. Some studies have also been carried out on the synthesis of novel lexitropsins (4.1a-c) specifically designed to target GC sequences. In summary, the results obtained have demonstrated the usefulness of biaryl building blocks in modifying the sequence selectivity of minor-groove binding agents, and lay the foundation for third generation compounds spanning up to five base pairs and designed to have specific recognition of individual bases within their binding site span.