The inclusion complexes of 3-formyl-(1), 3-acetyl-(2), and 3-carboxyphenoxathiin (3) were investigated by steady-state fluorescence spectroscopy. In the presence of β-cyclodextrin (β-CD), the fluorescence spectra of 1 and 2 point out an enhancement of the emission and a hypsochromic shift of the maxima. In the case of the carboxy derivative the measurements were conducted in buffer solutions, pH ) 2.11 and pH ) 9.18, to ensure the presence of either the neutral (3a) or dissociated carboxylate (3b) form, as a single predominant species. A different behavior was observed for these two species: en enhancement of the emission for the neutral form and a quenching for 3b. For all the compounds the best fit to the experimental data was obtained for 1:1 stoichiometry and points out association constants in the range 5000-7000 M -1 , for 1, 2, and 3b and 1770 M -1 for 3a. Molecular modeling of the complexes was performed by both molecular mechanics and quantum semiempirical methods. The ligands were introduced in the host cavity, either with the unsubstituted phenyl ring (model A) or with the substituted moiety (model B). For the neutral compounds, about 70% of the binding energy, was due to the van der Waals contribution whereas for 3b the electrostatic interaction was the dominant term. The distance between the center of the cavity and the mass center of the four carbon atoms in the median heteroring of the ligand, points out a deep penetration of the ligands. The flexibility of the sulfur-containing ring could be a positive factor for a better fit of the ligand into the cavity accounting for the large values of the association constants. The calculated heat of formation of the complexes are in the range usually found for the inclusion complexes excepting 3b, for which unexpected large values were predicted. The calculations did not allow us to distinguish between the two possible ways the ligands approach the cavity, the results for model A and model B being similar.

No abstract

A general strategy is reported for the preparation of C-5-methylamino- or methyldiamino-d4T analogues of "different sizes". Reactions of the 2',3'-didehydro-2',3'-dideoxy-C-5 hydroxymethyl precursor (7) with either polymethylene diamines (n = 6, 8, 10 and 12) or propargylamine proceed regioselectively via substitution reactions at the C-5 position of uracil. The compounds were evaluated for antiviral activity and cytotoxicity. No significant activity was observed for compounds 9, 11, and 13, but 10 and 12 exhibited a weak activity against HIV-1.

A series of eleven heterodimers containing both a nucleoside analogue (d4U, d4T) and a non-nucleoside type inhibitor (Trovirdine analogue) were synthesized and evaluated for their ability to inhibit HIV replication. Unfortunately, the (N-3)d4U-Trovirdine conjugates (9a-e) and (N-3)d4T-Trovirdine conjugates (10a-f) were found to be inactive suggesting that the two individual inhibitor compounds do not bind simultaneously in their respective sites.

The target compounds 5-[N-(6-amino-hexyl)-acrylamide]-2',3'-didehydro-2',3'-dideoxy-uridine (12) and 5-[N-[5-(methoxycarbonyl)-pentyl]-acrylamide]-2',3'-didehydro-2',3'- dideoxy-uridine (15) were prepared by the palladium acetate-triphenylphosphine-catalyzed reaction of the 5'-O-acetyl-5-iodo-d4T analogue (3). These compounds 12 and 15 can be used to prepare nucleotide probes carrying fluorescent labels and were nevertheless screened for their anti-HIV activity. The biological data demonstrated that none of them were active against HIV-1.