The replacement of hydrogen of hydroxyl group with fluorine is an important and effective strategy in the design of analogues of biologically important molecules. The small size of fluorine and high electronegativty are factors that contribute to the value of many of these analogues as pharmacological tools and medicinal agents. Included in our research in this area has been the preparation and biological evaluation of ring fluorinated imidazoles, indoles, biogenic amines and amino acids and related compounds. We have recently extended this work to include chain fluorinated analogues of these important biological molecules. Side-chain fluorinated bioimidazoles and bioindoles: In our earlier work, we were unable to oxidize beta,beta-difluorohistidinol derivatives to the corresponding amino acids. Various alternative approaches to side-chain fluorinated L-histidines have been explored. For example, condensation of the Schollkopf chiral glycine equivalent with a protected imidazole-4-carboxyaldehyde followed by deoxyfluorinated produced a protected analogue of beta-fluoro-L-histidine. However, hydrolysis of the side chain to reveal the amino acid functionality led, in all cases, to loss of fluorine. Similarly, routes through the Evans chiral glycine equivalent were unsuccessful. Routes to tryptophan derivatives have also proven problematic. Reactivity of the benzylic fluoride is clearly a synthetic complication an becomes an issue in considering biological evaluations. Dr. Phil Cole, Johns Hopkins University, has prepared potent inhibitors of aromatic amine N-acetyltransferase by linking tryptamine to acetyl CoA. A similar bisubstrate analogue of beta,beta-difluortryptamine and acetyl CoA was prepared, but loss of fluoride to produce the corresponding ketone occurred under enzyme assay conditions. Further examination of a series of these beta,beta-difluorotryptamines (e.g. serotonin and melatonin) revealed rapid loss of fluoride, even on storage at low temperature. The reactivities of side-chain fluorinated analogues of imidazoles were examined in more detail. The half-life for hydrolytic loss of fluorine at pH 7.4 was determined: beta,beta-difluorohistidinol, 10.6 h; 4-difluoromethylimidazole, 2.45 h; 2-difluoromethylimidazole, 2.7 h. At pH values approaching the pKa of these analogues (ca 10.5), the loss of fluorine was too rapid to follow. The stabilizing effect of an additional fluorine substituent can be seen by comparison of previously determined rates of hydrolysis of 2-trifluoromethyl imidazole, having a limiting half life at high pH of 5.8 h, with 2-difluormethyl imidazole that has a half life of less than 9 sec at pH 10.5. The high reactivity of these derivatives presents problems as well as opportunities. For example, we are exploring systems wherein the reactive diazafulvene intermediate implied in fluoride loss could potentially be used as an affinity label. A New Fluorination Procedure. We developed the photochemical Schiemann reaction thirtyfive years ago. This procedure that consists of the in situ photochemical decomposition of diazonium fluoroborates in aqueous fluoroboric acid remains the only reliable method to prepare ring-fluorinated imidazoles. The mechanism presumably involves attack of the intermediate carbocation generated by nitrogen extrusion by nucleophilic fluoride. In aqueous fluoroboric acid water can compete in this process, and with benzenoid substrates, phenols can be isolated. In an approach to blocking this side reaction, we have investigated using anhydrous ionic liquids as solvent. Ethyl 4-aminoimidazole-5-carboxylate was dissolved in the commercially available ionic liquid, N-methyl-N-butylimidazolium tetrafluoroborate. Diazotization with nitrosonium tetrafluorborate followed by photolysis indeed produced the desired ethyl 4-fluoroimidazole-5-carboxylate in yields approaching 60%. We are optimizing conditions, including temperature and use of co-solvents, in an effort to improve the yields. Biochemical Incorporation of Fluorohistidine into Proteins. The successful completion of biosynthetic incorporation of fluorohistidine would provide a novel method, via 19F NMR, to monitor the structure and function of proteins that utilize histidine for catalysis, metal-ion binding, or stabilization. Additionally, because the pKa of both 2-fluoro- and 4-fluorohistidine is decreased from approximately 6.0-6.5 to approximately 1 and 3, respectively, incorporation of fluorohistidine would help verify the role of native histidine in pH-dependent processes. In collaboration with Dr. Jim Bann, Wichita State University, we have demonstrated the biosynthetic incorporation of both 4-F-His and 2-F-His into a mutant form of the chaperone protein PapD. Site-directed mutagenesis was used to introduce a single histidine residue at arginine-200 in PapD. The plasmid DNA of this mutant was transformed into the histidine auxotrophic bacterial strain, UTH780. Standard labeling and expression procedures were then used to incorporate both 2-F-His and 4-F-His into PapD(Arg200-His). NMR and mass spectral analysis confirmed the incorporation. Although we had previously shown facile biosynthetic incorporation of 2-fluorohisitidine, this work represents the first successful incorporation of both 2-fluoro- and 4-fluoro-histidine into a protein using biosynthetic methods. The ability to incorporate both regioisomers will provide added advantages for the structural and functional characterization of proteins that utilize histidine for physiological processes via the use of 19F NMR. Fluorinated cyclopropyl amines as inhibitors of amine oxidases: 1-Phenylcyclopropylamine is an irreversible inhibitor MOA with selectivity for MAO B over MAO A. We have shown that E-2-fluoro-1-phenycyclopropylamine is a potent irreversible MAO A selective inhibitor. We now have shown that the Z-isomer has modestly greater activity. Additional examples of 1-aryl-2-fluorocyclopropylamnes were prepared and evaluated (E- and Z-isomers, para-substituted with F, Cl, Me, MeO) and all show potent inhibition of MAO A. The most potent compound, Z-2-fluoro-1-(para-methylphenyl)cyclopropylamine, had an IC50 of 0.3 micromolar, or about 2400 times more potent than the parent 1-phenylcyclopropylamine. The inhibitory activity has been shown to be relatively insensitive to the stereochemistry of the cyclopropane ring, with all four stereoisomers of 2-fluoro-1-phenylcyclopropylamine having comparable IC50 values. The change in MAO A/MAO B selectivity by introduction of fluorine is approximately 1000. In order to gain more information on this dramatic effect of fluorine, more detailed kinetic analyses using purified enzyme are now being done by Professor Dale Edmondson, Emory University, Fluorophosphonate analogues of UDP-Glc-NAc as potential inhibitors of OTG transferase. The difluoromethylene group is an isosteric and isopolar replacement of oxygen in phosphate esters. Accordingly, difluorophoshonate analogues of biologically important phosphate esters have been prepared and studied extensively. We are using this strategy to prepare potential inhibitors of OGT transferase, the enzyme that catalyzes the transfer of GlcNAc to serine and threonine residues in proteins. We have prepared the phosphonate analogue of GlcNAc from a key C-allyl glycoside of GlcNAc. Improvements in key oxidation steps have increased availability of key intermediates. To date, deoxyfluorinations of hydroxylated phosphonates have been complicated by participation of the neighboring N-acetyl group. Synthetic strategies to work around this problem are on-going. Coupling of the non-fluorinated phosphonate has provided the UDP-GlcNAc analogue for inhibition studies.
Showing the most recent 10 out of 33 publications
