A series of hexadeoxyribonucleotides (6-mers), d(TGGGAG), substituted with a variety of aromatic groups at the 5'-end were synthesized and tested for anti-human immunodeficiency virus type 1 (HIV-1) activity. While unmodified d(TGGGAG) (31) had no anti-HIV-1 activity, compound 23 with a 3,4-di(benzyloxy)benzyl (DBB) group at the 5'-end potently inhibited the HIV-1IIIB-induced cytopathicity of MT-4 cells in vitro (IC50 = 0.37 microM) without cytotoxicity up to 40 microM. A thermal denaturation study on the 5'-end-substituted 6-mers by means of the circular dichroism (CD) spectra demonstrated that the aromatic substituent attached at the 5'-end of the 6-mer strongly enhanced the formation of a parallel helical structure consisting of four strands (quadruplex). On the contrary, compound 36, in which one of the guanosines of 23 was replaced by a thymidine, did not form a quadruplex, thus exhibiting no anti-HIV-1 activity. Moreover, both compound 15, with a tert-butyldiphenylsilyl group solely at its 3'-end, and compound 21, with a relatively small substituent, a benzyl group, at the 5'-end, formed quadruplexes but had no anti-HIV-1 activity. These findings led us to the conclusion that both the quadruplex structure and the aromatic substituent with adequate size at the 5'-end are crucial for the interaction of the 5'-end-substituted 6-mers with the V3 loop as well as the CD4 binding site on viral gp120, resulting in anti-HIV-1 activity.

These results suggest that capuramycin analogues exhibit strong antimycobacterial potential and should be considered for further evaluation in the treatment of M. tuberculosis and M. avium-M. intracellulare complex infections in humans.

The recognition mode of adenophostin A at the D-myo-inositol 1,4, 5-trisphosphate [Ins(1,4,5)P(3)] receptor was investigated. Comparison of conformations of Ins(1,4,5)P(3) and adenophostin A by using the combination of NMR and molecular mechanics (MM) calculations demonstrated that adenophostin A adopted a moderately extended conformation regarding the distance between the 2'-phosphoryl group and the 3' ',4' '-bisphosphate motif, as suggested previously [Wilcox, R. A. et al. (1995) Mol. Pharmacol. 47, 1204-1211]. Based on the nuclear Overhauser effect (NOE) observed between 3'-H and 1' '-H and on MM calculations, the molecular shape of adenophostin A proved to be an extended form at least in solution, in contrast to Wilcox's compactly folded, preliminary hairpin model. GlcdR(2,3',4')P(3), an adenophostin analogue without adenine moiety, was found to be less potent than adenophostin A and almost equipotent to Ins(1,4,5)P(3). We propose the possibility that (i) the optimal spatial arrangement of the three phosphoryl groups and/or (ii) the interaction of the adenine moiety of adenophostin A with the putative binding site, if it exists in the vicinity of the Ins(1,4,5)P(3)-binding site, might account for the exceptional potency of adenophostin A.

No abstract

No abstract

No abstract

No abstract

No abstract