2-5A-ANTISENSE Our laboratory has developed the use of an antisense oligonucleotide to address a specific RNA sequence and subsequent localized activation of the 2-5A-dependent RNase (RNase L) to effect selective RNA degradation is a new approach to the control of gene expression called 2-5A-antisense. The previously reported biological activity of the 2-5A:AS chimeric oligonucleotide [p5'(A2'p)3A-antiPKR1], directed against nucleotides 55-73 of the coding sequence of the PKR mRNA, have been used as a point of reference to examine the effect of introducing mismatches into the chimeric oligonucleotide, altering the chain length of the antisense domain of the chimeras, removal of the 5'-monophosphate moiety, shortening the 2',5'-oligoadenylate domain, and substitution of 3',5'-linked 2'-deoxyadenosine nucleotides for the 2-5A domain. When the biological activity of these new chimeric oligonucleotides were compared to the parent chimera, 2-5A-aPKR, for their ability to effect target PKR RNA cleavage in a cell-free and in an intact cell assay, it was determined that there was a close correlation between the activity of 2-5A-antisense chimeras and their affinity (Tm) for a targeted nucleic acid. In addition, there was also a close correlation between activity of the 2-5A-antisense chimeras and their ability to activate the 2-5A-dependent RNase L. In order to stabilize 2-5A-antisense chimeras to exonucleases, we have synthesized chimeric oligonucleotides in which the last phosphodiester bond at the 3'-terminus of the antisense domain was inverted from the usual 3',5'-linkage to a 3',3'-linkage. The structures of such terminally inverted linkage chimeras of the general formula pA4-[pBu]2-(pdNn3'-3'dN) were corroborated by a combination of snake venom phosphodiesterase digestion in the presence or in the absence of bacterial alkaline phosphatase. Most characteristically, the presence of the 3'-terminal inverted phosphodiester linkage produced an unnatural dinucleotide of general composition, dN3'p3'dM. 2-5A-antisense chimeras of this structural class, pA4-[pBu]2-(pdNn3'-3'dN), were 5-6-fold more stable than their unmodified congeners, pA4-[pBu]2-(pdN)n, to degradation by a representative phosphodiesterase from snake venom. In 10% human serum, the new 2-5A-antisense chimeras, pA4-[pBu]2-(pdNn3'-3'dN), possessed a half-life that was 28-fold longer than the unmodified chimeras. These results provide entry to a second generation of 2-5A-antisense chimeras.
