Mapping of RNA accessible sites for antisense experiments with oligonucleotide libraries

Target site selection for an RNA-cleaving catalytic DNA

A small catalytic DNA, known as the 10-23 DNA enzyme or deoxyribozyme, has been shown to efficiently hydrolyze RNA at purine-pyrimidine (R-Y) junctions in vitro. Although these potentially cleavable junctions are ubiquitous, they are often protected from deoxyribozyme activity by RNA secondary structure. We have developed a multiplex cleavage assay for screening the entire length of a target RNA molecule for deoxyribozyme cleavage sites that are efficient, both in terms of kinetics and accessibility. This strategy allowed us to simultaneously compare the RNA cleaving activity of 80 deoxyribozymes for a model target gene (HPV16 E6), and an additional 60 deoxyribozymes against the rat c-myc target. The human papilloma virus (HPV) target was used primarily to characterize the multiplex system and determine its validity. The c-myc target, coupled with a smooth muscle cell proliferation assay, allowed us to assess the relationship between in vitro cleavage efficiency and c-myc gene suppression in cell culture. The multiplex reaction approach streamlines the process of revealing effective deoxyribozymes in a functional assay and provides accessibility data that may also be applicable to site selection for other hybridization-based agents.

Antisense pharmacodynamics: critical issues in the transport and delivery of antisense oligonucleotides

This review critically examines current understanding of the kinetics and biodistribution of antisense oligonucleotides, both at the cellular level and at the level of the intact organism. The pharmacodynamic relationships between biodistribution and the ultimate biological effects of antisense agents are considered. The problems and advantages inherent in the use of delivery systems are discussed in the light of further enhancing in vivo pharmacological actions of oligonucleotides.

Regulation of the angiotensin type-1 receptor by antisense oligonucleotides occurs through an RNase H-type mechanism

Multiple, diverse sites in the coding region of the angiotensin type-1 receptor mRNA were targeted with 2'-deoxyribonucleotide antisense oligonucleotides (ONs). The uptake of 1 microM concentration of these ONs into Chinese hamster ovary cells was facilitated by the use of cationic liposomes. The antisense sequences reduced binding of 125I-angiotensin II by 57-73%, while mismatch ONs and reverse sequence ONs produced little reduction in receptor binding. These reductions in AT1 receptor binding were accompanied by comparable decreases in AT1 receptor mRNA levels. Furthermore, mRNA cleavage fragments corresponding in size to 3'-cleavage fragments were observed with two of the antisense ONs, consistent with the involvement of an RNase H-type enzyme. When 2'-methoxyribonucleotide analogs of these same sequences were tested, AT1 receptor mRNA levels were unchanged even though small reductions in AngII binding were observed. Antisense effects seen with these 2'-methoxyribonucleotide sequences may have arisen through a translational arrest mechanism. Direct comparisons between 2'-deoxyribonucleotide analogs and their 2'-methoxyribonucleotide counterparts show that antisense effects are significantly larger when they are mediated through an RNase H-type mechanism. 2'-methoxyribonucleotide sequences were most effective when they were directed against the translation initiation codon.

Recent developments in the hammerhead ribozyme field

Developments in the hammerhead ribozyme field during the last two years are reviewed here. New results on the specificity of this ribozyme, the mechanism of its action and on the question of metal ion involvement in the cleavage reaction are discussed. To demonstrate the potential of ribozyme technology examples of the application of this ribozyme for the inhibition of gene expression in cell culture, in animals, as well as in plant models are presented. Particular emphasis is given to critical steps in the approach, including RNA site selection, delivery, vector development and cassette construction.

Rapid determination and quantitation of the accessibility to native RNAs by antisense oligodeoxynucleotides in murine cell extracts

A major concern for antisense experiments is the prediction of effective oligonucleotide binding sites. We have developed a system to carry out oligodeoxyribonucleotide-RNA and ribozyme-RNA binding experiments in cell extracts to create a protein environment known to directly influence the structure of the mRNA. In these experiments the native, endogenous mRNA is probed using oligodeoxyribonucleotides (ODNs) to identify RNase H-accessible sites. The resulting RNase H-mediated cleavages in the cell extracts were quantified using RT-PCR with fluorescein and rhodaminetagged primers to generate fluorescent products that are analyzed and quantified on an automated DNA sequencer. As a model substrate for testing this system, we have targeted the murine DNA methyltransferase (MTase) mRNA. An ODN binding site in native MTase mRNA was identified that was cleaved by endogenous RNase H with an efficiency of 85% in the extracts. The ODN that was most effective in the cell extracts was also found to provide the best activity in vivo , resulting in a 75-85% reduction of the MTase mRNA. These data support the use of cell extracts and native transcripts to identify antisense and perhaps ribozyme target sites.

Modification of phosphorothioate oligonucleotides yields potent analogs with minimal toxicity for antisense experiments in the CNS

There is increasing evidence that phosphorothioate oligonucleotides infused into the brain can cause a host of undesired side effects which compromise the antisense experiment. In studies on the corticotropin releasing factor type-2 receptor, several phosphorothioate oligonucleotides administered intraventricularly produced significant weight loss in rats. Four different phosphodiester and phosphorothioate oligonucleotide analogs were examined to identify molecules which could eliminate these side effects while maintaining good potency for antisense inhibition. Of these, chimeric oligonucleotides consisting of a mixed phosphodiester-phosphorothioate backbone, and having 2'-methoxyribonucleotide modifications in 60% of the oligonucleotide were the most optimal. Rats treated with these chimeric oligonucleotides gained weight at rates identical to that of saline-treated controls. In addition, the antisense oligonucleotide but not the mismatch control sequence reduced corticotropin releasing factor type-2 receptor binding of 125iodo-sauvagine in the lateral septum by 40-60% after 5 daily injections. Increasing the dosing period to 9 days reduced receptor binding by 78%. Reductions in protein binding were accompanied by comparable reductions in the in situ hybridization signal of the corticotropin releasing factor type-2 receptor mRNA. However, when an oligonucleotide analog incapable of supporting ribonuclease H activity was used, neither protein nor RNA binding levels were changed compared to saline-treated controls. These results suggest that ribonuclease H or enzymes with similar activity are critical to the antisense inhibition observed in the lateral septum.

Tetranucleotide GGGA motif in primary RNA transcripts. Novel target site for antisense design

Selecting effective antisense target sites on a given mRNA molecule constitutes a major problem in antisense therapeutics. By trial-and-error, only 1 in 18 (6%) of antisense oligonucleotides designed to target the primary RNA transcript of tumor necrosis factor-alpha (TNF-alpha) strongly inhibited TNF-alpha synthesis. Subsequent studies showed that the area in RNA targeted by antisense oligonucleotides could be moved effectively 10-15 bases in either direction from the original area. We observed that only molecules that incorporated a tetranucleotide motif TCCC (complementary to GGGA on RNA) yielded potent antisense oligonucleotides against TNF-alpha. A comprehensive literature survey showed that this motif is unwittingly present in 48% of the most potent antisense oligonucleotides reported in the literature. This finding was prospectively used to predict the sequences of additional antisense oligonucleotides for the rat TNF-alpha primary RNA transcript. Over 50% of antisense constructs (13 of 22) containing the TCCC motif were found to effectively inhibit TNF-alpha synthesis. Marked reductions in mRNA were also observed. This motif was found to be most effective when targeting introns in the primary RNA transcript, suggesting a nuclear localization for the antisense action. Predicting target sites based on the presence of this motif in primary RNA transcripts should be of value in the development on new antisense pharmacotherapy.

Antisense drug discovery: can cell-free screens speed the process?

Many conditions must be satisfied for an antisense drug to function. It must colocalize with its target RNA at a sufficient concentration for a bimolecular reaction to occur, and it must have a structure that favors association with its target. In addition, if the antisense compound is to form Watson-Crick bonds with the target RNA, it must be complementary to sites that are amenable to binding. Unfortunately, the peculiarities that cause certain sites to be especially vulnerable to antisense compounds are undefined, as discussed previously (Branch, 1998). Because vulnerable target sites have no common properties allowing them to be identified by sequence analysis, most target sites and their antisense counterparts are found through a trial and error process in which oligomers--each complementary to a different site in the target RNA--are tested individually to find the one with the greatest specificity and lowest inhibitory concentration (IC50). However, testing antisense molecules one at a time can be a taxing process, and there is great interest in developing cell-free screening methods that can reduce the number of compounds that must be tested in cells and in whole animals. These cell-free screens are designed to generate short lists of target sites that include the ideal site--the site most vulnerable to antisense ablation in vivo. They are based on the unproven assumption that ideal sites have distinctive properties, such as susceptibility to RNase H-mediated cleavage, that allow them to be detected in cell-free assays. This is a review of data emerging from studies using RNase H-based screens and a summary of the challenges confronting these and any similar methods that use naked RNAs as surrogates for intracellular RNAs. It is not yet clear if cell-free screening methods will be effective.

Extending the cleavage rules for the hammerhead ribozyme: mutating adenosine15.1 to inosine15.1 changes the cleavage site specificity from N16.2U16.1H17 to N16.2C16.1H17

In this paper, we show that an adenosine to inosine mutation at position 15.1 changes the substrate specificity of the hammerhead ribozyme from N16.2U16.1H17to N16.2C16.1H17(H represents A, C or U). This result extends the hammerhead cleavage triplet definition from N16.2U16.1H17to the more general N16.2Y16.1H17. Comparison of cleavage rates using I15.1ribozymes for NCH triplets and standard A15.1 ribozymes for NUH triplets under single turnover conditions shows similar or slightly enhanced levels of reactivity for the I15. 1-containing structures. The effect of I15.1 substitution was also tested in nuclease-resistant 2'- O -alkyl substituted derivatives (oligozymes), showing a similar level of activity for the NUH and NCH cleaving structures. The availability of NCH triplets that can be targeted without loss of efficiency increases the flexibility of ribozyme targeting strategies. This was demonstrated by an efficient cleavage of an HCV transcript at a previously inaccessible GCA site in codon 2.