Antisense & nucleic acid drug development最新文献

Antisense oligonucleotides (ODNs) are powerful tools with which to determine the consequences of the reduced expression of a selected target gene, and they may have important therapeutic applications. Methods for predicting optimum antisense sites are not always effective because various factors, such as RNA-binding proteins, influence the secondary and tertiary structures of RNAs in vivo. To overcome this obstacle, we have attempted to engineer an antisense system that can unravel secondary and tertiary RNA structures. To create such an antisense system, we connected the constitutive transport element (CTE), an RNA motif that has the ability to interact with intracellular RNA helicases, to an antisense sequence so that helicase-binding hybrid antisense ODN would be produced in cells. We postulated that this modification would enhance antisense activity in vivo, with more frequent hybridization of the antisense ODN with its targeting site. Western blotting analysis demonstrated that a hybrid antisense ODN targeted to the bcl-2 gene suppressed the expression of this gene more effectively than did the antisense ODN alone. Our results suggest that the effects of antisense ODNs can be enhanced when their actions are combined with those of RNA helicases.

Oligonucleotide (ODN) therapy is a powerful tool for modulation of gene expression in vivo. With advances in ODN chemistry and progress in formulation development, ODNs are becoming widely acceptable drugs. This review summarizes the current status and future trend of the in vivo application of ODN therapeutics, especially antisense ODNs. Here, we review the current understanding of the tissue/organ distribution and cellular uptake of ODN drugs administered parenterally or nonparenterally to intact animals. The problems and advantages inherent in the use of different delivery methods for the treatment of particular diseases are discussed in detail. Emphasis is placed on the most widely studied ODN analogs, the phosphorothioates (PS). Lessons learned from antisense PS studies have broad implications for ODN therapeutics in general.

Short interfering RNA (siRNA)-mediated knockdown of deleterious endogenous transcripts has potential applications for the treatment of hereditary diseases. In situations where the mutant and wildtype transcripts cannot be discriminated from one another by siRNAs, it may be necessary to simultaneously carry out gene replacement with a modified form of the target RNA that is resistant to siRNA activity. To test this possibility, we have taken advantage of a potent siRNA that knocks down EGFP mRNA. In this system, wild-type EGFP expression is suppressed by the siRNA, whereas an EGFP construct with codon modifications in the target region that is otherwise fully functional is not downregulated. When expression of the wild-type message is inhibited, EGFP expression can be simultaneously restored by transfecting these cells with the codon-modified version of EGFP. These studies provide a detailed methodology and system for testing this strategy with RNA interference (RNAi).

A recently developed antisense splicing assay was used to determine the relative activities of 2'-O-methoxyethoxy (2'-MOE) phosphorothioate oligonucleotides containing base modifications. In the assay, RNase H-inactive oligonucleotides are used to block aberrant splicing and restore correct splicing of an Enhanced Green Fluorescence Protein (EGFP) reporter pre-mRNA stably expressed in HeLa cells. Thus, the extent of EGFP upregulation is proportional to the antisense activity of the tested molecule. The base modifications included C-5 propynyl analogs of uridine and cytidine and phenoxazine and G-clamp analogs of cytosine. Base-modified 2'-MOE oligonucleotides were delivered to the HeLa EGFP-654 test cells by cationic lipid transfection or scrape-loading or without any delivery method (free uptake). When delivered with a cationic lipid, the G-clamp and phenoxazine oligomers showed increases in activity over the unmodified 2'-MOE parent compound. However, when delivered by scrape-loading or without a delivery method, the unmodified oligomer performed best. The results suggest that base modifications do not enhance the free uptake activity of RNase H inactive 2'-MOE oligomers.

Inclusion of C-5 propynyl pyrimidines in phosphorothioate antisense oligonucleotides (ASOs) has been shown to significantly increase their potency for inhibiting gene expression in vitro. This increased potency is believed to be the result of enhanced binding affinity to target RNA. Our results show that C-5 propynyl pyrimidine-modified oligonucleotides caused an increase in the melting temperature (T(m)) of both oligodeoxynucleotides (ODNs) and 2'-O-(2-methoxy)ethyl (2'-MOE)-modified oligonucleotides. The in vitro data show a moderate increase in potency for an antisense oligodeoxynucleotide containing C-5 propynyl pyrimidines targeting the murine PTEN (MMAC1) transcript. Second-generation 2'-MOE chimeric ASOs containing C-5 propynyl pyrimidines showed no improvement in potency in PTEN target reduction in vitro or in vivo compared to their nonpropyne-modified parent. These results suggest that increasing affinity for target RNA beyond that achieved with the 2'-MOE modification does not further increase potency in cell-based assays. To evaluate whether this observation held true for in vivo applications, we evaluated both compounds in mice. We were unable to establish a dose-response relationship with C-5 propynyl pyrimidine-modified ODNs because of severe toxicity. The toxicity was characterized by mortality in animals receiving 50 mg/kg and an increase in infiltrating cells and apoptotic cells in livers of mice receiving 20 mg/kg. C-5 propynyl pyrimidine-modified chimeric oligonucleotides exhibited decreased hepatotoxicity compared with C-5 propynyl-modified ODNs but did not exhibit an increase in potency compared with unmodified chimeric oligonucleotides. The hepatotoxicity could be further limited if incorporation of propynyl pyrimidines was restricted to 2'-MOE nucleosides.

The proliferative response of primary B cells to CpG oligonucleotides (ODN) involves induction of nuclear activation promoting-1 (AP-1) transcription factor. AP-1 subunits c-Fos, Fos-B, Jun-B, and Jun-D, but not Fra-1 or Fra-2, were all induced by CpG ODNs in B cells within 30 minutes of stimulation, followed by c-Jun at 1-2 hours. c-Jun reached maximum at 6 hours. By 40 hours, Jun-B and Jun-D became dominant. Synthetic ODNs containing a single guanosine triplet/tetrad appropriately distanced from the 5' pyrimidine-rich unit, which inhibit CpG-driven cell cycle entry and apoptosis protection, blocked AP-1 induction by stimulatory ODNs when they were added simultaneously. After 30 minutes of stimulation, adding inhibitor no longer affected AP-1 at 6 hours. No AP-1 subunits escaped ODN inhibition. In a cell line transfected with an AP-1-beta-galactosidase reporter construct, CpG ODN-induced AP-1 transcriptional activity was prevented by inhibitory ODN, but lipopolysaccharide (LPS)-induced AP-1 activity was not. These data suggest that inhibitory ODNs block the CpG ODN-driven signaling pathway at a site proximal to AP-1 induction.

To decrease CD4 expression on T helper (Th) lymphocyte surface, antisense oligonucleotides (AS-ODNs), delivered by the cationic liposome DOTAP, were assayed in vitro on rat spleen lymphocytes. Four 21-mer ODNs (AS-CD4-1, AS-CD4-2, AS-CD4-3, and AS-CD4-4) directed against the translation start region of the cd4 gene were designed. AS-CD4-1 was phosphorothioate (PS)-modified in each base, and the other three were PS-modified at both ends and in the internal pyrimidine residues. Four ODN controls (fully PS-modified ODN-A and partially modified ODN-B, ODN-C, and ODN-D) were also assayed. CD4 resynthesis was stimulated by treatment with phorbol 12-myristate 13-acetate (PMA) at the same time as the incubations with the ODN. After 24 hours of treatment, CD4 expression was measured by immunofluorescence staining and flow cytometry. CD4 reexpression in rat PMA-treated lymphocytes was counteracted by 40% by means of AS-CD4-2 and AS-CD4-4 treatments. On the other hand, AS-CD4-3 produced only 20% inhibition, similar to that produced by ODN-B, and AS-CD4-1 did not have any significant effect compared with control ODNs. Both AS-CD4-2 and AS-CD4-4 decreased CD4 mRNA, as determined by RT-PCR, and in addition, they did not affect the expression of other surface lymphocyte molecules. Inhibition of surface CD4 expression remained at least 72 hours. The addition of both AS-ODNs did not further increase the effect obtained separately by each AS-ODN. Treatment of rat PMA-lymphocytes with two concentrations of AS-CD4-2 and AS-CD4-4 added 24 hours apart did not further decrease CD4 expression. In summary, AS-CD4-2 and AS-CD4-4 could constitute a good strategy to inhibit CD4 expression on Th lymphocytes and modulate their function.

Watson-Crick pairing between complementary oligomers is proving to be an effective means for rapidly directing radioisotopes specifically to the exterior surface of cancer cells in vivo. In such pretargeting applications, it is highly desirable that the excess of isotopically labeled oligomers, which do not bind to the cancer cells, be rapidly cleared from the body. In this context, understanding the influence of chain length and base sequence of the radiolabeled oligomers is critical. We had earlier determined that the kidneys are the principal targets of short-chain radiolabeled morpholino oligomers (MORFs). To explain these observations, MORFs consisting of uniform cytosines (Cs), uniform thymines (Ts), uniform adenines (As), and uniform AAG repeat were labeled with technetium-99m (99mTc) and studied in normal mice. In a limited investigation of the influence of oligomer backbone, a 20-mer MORF (MORF20) with a base sequence rich in Cs was compared with a phosphoromonothioate DNA (S-DNA20) of the same sequence. The in vivo behavior of the labeled MORFs was nearly identical in all organs, with the exception of kidneys. The kidney accumulations were about 25- to 80-fold higher for the uniform Cs relative to the other three uniform MORFs at 3 hours. The S-DNA20 rich in Cs showed only modest kidney accumulations compared with the equivalent MORF20, presumably because of preferential clearance of the S-DNA20 through the liver. Urine analysis showed no evidence of intact labeled S-DNA20 in contrast to fully intact labeled MORF20. We conclude that the high kidney levels observed by us previously for MORFs are most likely due largely to the C residues in the base sequence. In the case of S-DNAs, this phenomenon is partly disguised by the increased hepatic excretion and degradation. These results show that the base sequences of MORFs, and probably other oligomers as well, are an important determinant of biodistribution.

Human keratinocytes are highly responsive to mitogenic and antiapoptotic signaling by the insulin-like growth factor-I receptor (IGF-IR). IGF-IR hyperstimulation is a feature of hyperplastic skin conditions, making the IGF-IR an appealing target for antisense therapeutic intervention. In this study, we used a C5-propynyl-dU,dC-phosphorothioate oligo-2'-deoxyribonucleotide antisense 15-mer to the human IGF-IR mRNA, along with liposome transfection, to inhibit IGF-IR activity in a human keratinocyte cell line and demonstrated potent inhibition of cell growth despite the presence of serum. To investigate the sequence specificity of these effects and to establish the concentration range over which a purely antisense effect could be demonstrated, we introduced 1, 2, 4, 8, and 15 base mismatches into the oligonucleotide and analyzed changes in inhibitory efficacy. In the 10-30 nM concentration range, the introduction of 1 and 2 mismatches into the middle of the 15-mer only modestly affected inhibitory efficacy, whereas >4 mismatches profoundly reduced mRNA, protein, and growth-inhibitory effects. From these results, we conclude that (1) sequence-specific antisense inhibition of IGF-IR activity in keratinocytes is achievable, (2) potent anti-IGF-IR antisense inhibition can be achieved in vitro at concentrations as low as 10 nM, and (3) a sequence-dependent mechanism is likely to underpin the observed in vivo therapeutic effects (Wraight et al. Nat. Biotechnol. 2000;18:521) of these antisense oligonucleotides (AS-ODN) in cutaneous hyperplastic disorders, such as psoriasis.