Combinatorial Synthesis of Pilicides

Sammanfattning: This thesis describes a combinatorial approach to design and synthesis of three libraries of compounds, referred to as pilicides, designed to target periplasmic chaperones present in uropathogenic Escherichia coli bacteria that are the main cause of urinary tract infections. Periplasmic chaperones are required for assembly of adhesive organelles, i.e. pili, which recognize and bind to specific receptors present on the cell surface of host tissue and thereby initialize bacterial colonization and infection. Prevention or inhibition of pilus assembly renders the bacterium less pathogenic, and pilicides therefore represents a potential class of new antibiotics. A generic pilicide structure, consisting of N-alkylated and N-acylated amino acids, was designed using information from the structure of the complex between periplasmic chaperone PapD and a C-terminal peptide from the pilus adhesin PapG. The selection of building blocks and compounds for combinatorial synthesis of pilicide libraries involved molecular design and statistical molecular design. The affinity of the resulting pilicides for chaperones PapD and FimC was screened in a direct binding assay based on surface plasmon resonance using a Biacore 3000 instrument. This assay allowed ranking of pilicides in classes of binders with strong, medium and low affinity. In summary, compounds based on hydrophobic amino acids, such as phenylalanine and tyrosine substituted with a 2-(3-indolyl)ethyl group in combination with a 2-naphthyl group on the amino group, bound strong to the chaperones. Omportant, these pilicides bound only weak to non-target proteins (Protein A, Streptavidin, and an anti-myoglobin mAb), suggesting them to be specific for their chaperone targets. Binding to PapD for the strongest binder, i.e. N-[2-(3-indolyl)ethyl]-N-(naphthalene-2-carbonyl)-tyrosine, was also investigated by 1H NMR spectroscopy. Spectra recorded from a mixture of N-[2-(3-indolyl)ethyl]-N-(naphthalene-2-carbonyl)-tyrosine, 1-naphthylacetic acid and PapD (1:1:1), revealed that the signals from the pilicide was reduced when the spectra were recorded with a delay of 200 ms, which indicated binding of the pilicide to PapD. N-[2-(3-Indolyl)ethyl]-N-(naphthalene-2-carbonyl)-tyrosine was also able to dissociate the FimC-FimH complex in vitro, suggesting that this pilicide bind to the active site of chaperones. This thesis also presents the concept of use of fluorinated linkers for gel-phase 19F NMR spectroscopy monitoring of reactions in solid-phase synthesis. Three fluorinated linkers were synthesized and one of these linkers (3-fluoro-4-(hydroxymethylphenoxy)acetic acid) alone, or in combination with fluorinated building blocks, enabled quantification and optimization of reaction conditions for solid-phase synthesis of pilicides. Optimized conditions for attachment of the first building block the solid support and cleavage of target pilicides after synthesis could be found by monitoring the change of the 19F chemical shift originating from the linker 3-fluoro-4-(hydroxymethylphenoxy)acetic acid. Use of 3-fluoro-4-(hydroxymethylphenoxy)acetic acid as an internal standard in combination with fluorinated building blocks allowed optimization of reaction conditions for reductive N-alkylation of amino acids with aromatic and aliphatic aldehydes, a key step in solid-phase synthesis of pilicides selected from a statistical molecular design. In summary, on-bead 19F NMR spectroscopy using the fluorinated linker proved to be an invaluable tool in troubleshooting difficult steps and also for optimization of reaction conditions in solid-phase synthesis.