Gaining Target Access for Deoxyribozymes

https://doi.org/10.1016/j.jmb.2004.03.064Get rights and content

Abstract

Antisense oligonucleotides and ribozymes have been used widely to regulate gene expression by targeting mRNAs in a sequence-specific manner. Long RNAs, however, are highly structured molecules. Thus, up to 90% of putative cleavage sites have been shown to be inaccessible to classical RNA based ribozymes or DNAzymes. Here, we report the use of modified nucleotides to overcome barriers raised by internal structures of the target RNA. In our attempt to cleave a broad range of picornavirus RNAs, we generated a DNAzyme against a highly conserved sequence in the 5′ untranslated region (5′ UTR). While this DNAzyme was highly efficient against the 5′ UTR of the human rhinovirus 14, it failed to cleave the identical target sequence within the RNA of the related coxsackievirus A21 (CAV-21). After introduction of 2′-O-methyl RNA or locked nucleic acid (LNA) monomers into the substrate recognition arms, the DNAzyme degraded the previously inaccessible virus RNA at a high catalytic rate even to completion, indicating that nucleotides with high target affinity were able to compete successfully with internal structures. We then adopted this strategy to two DNAzymes that we had found to be inactive in our earlier experiments. The modified DNAzymes proved to be highly effective against their respective target structures. Our approach may be useful for other ribozyme strategies struggling with accessibility problems, especially when being restricted to unique target sites.

Introduction

Antisense oligonucleotides and nucleic acids enzymes are valuable tools to regulate the expression of target genes in a sequence-specific manner.1., 2., 3. These approaches employ oligonucleotides that bind to complementary mRNA molecules by Watson–Crick-type base-pairing. In theory, antisense oligonucleotides, ribozymes and deoxyribozymes can be designed to interact site-specifically with any RNA species. Despite this seemingly simple concept, one usually encounters numerous obstacles on the way to an efficient regulation of target gene expression. One of the major hurdles is the identification of accessible sites that allow the efficient interaction of ribozymes or antisense oligonucleotides with their specific target.

Numerous efforts have focused on solving the accessibility problem. Computer-based structural models of the target RNAs were employed to design efficient antisense oligonucleotides or ribozymes. Structure prediction, however, has severe limitations when focussing on complex targets like mRNAs. Therefore, a variety of experimental strategies have been developed to define accessible target sites. These approaches include the use of randomized or sequence-specific oligonucleotide libraries, ribozyme expression cassettes or, even more demanding, DNA arrays. For details the interested reader is referred to the reviews by Sohail & Southern4 and Gautherot & Sodoyer.5

To identify accessible target sites on a complex and highly structured RNA by any of these methods is a time and labour consuming process that does not always meet with success. In general, only one out of eight antisense oligonucleotides is thought to be efficient for knockdown of target genes.6 Likewise, up to 90% of the putative target sites on long RNAs (>700 bases) have been reported to be resistant to cleavage by deoxyribozymes.7., 8. In cases where the window of potential target sites within the sequence is narrow, this resistance may be prohibitive for antisense or ribozyme strategies.

In our present study, we report a strategy to design catalytically active oligonucleotides to cleave a seemingly uncleavable site. In the course of our attempt to design an effective antiviral strategy against picornaviruses, we employed the “10–23” DNAzyme, which has been obtained by in vitro selection.9., 10., 11., 12. As a target site we chose a region that is highly conserved between a large number of rhino- and coxsackieviruses.13 While this DNAzyme proved to be effective against the RNA of the human rhinoviruses (HRV) 2, 4, 14 and 16, it showed virtually no detectable degradation of coxsackievirus mRNA. By introducing 2′-O-methyl RNA or locked nucleic acids (LNA) monomers into the binding arms of the DNAzyme, we succeeded in cleaving the apparently unsuitable target site. These findings were confirmed with the mRNA of the vanilloid receptor subtype 1 (VR1, TRPV1) as a second target system, suggesting that our strategy may be generally applicable.

Section snippets

Results

Recently, we reported the design of an optimized 10–23 DNAzyme with enhanced catalytic activity and stability against the 5′ non-coding region of human rhinovirus 14 (see Figure 1).13 The deoxyribozyme, which we designated DH5, was directed towards a 19 nucleotide long target site that is highly conserved among a great number of rhino- and enteroviruses (Table 1) and has been shown to cleave the 5′ UTR of HRV14 (Figure 2(A) and (C)).13 Although the 5′ UTR of CAV-21 contains the identical

Discussion

The design of efficient antisense oligonucleotides or ribozymes is usually guided by the accessibility of the target site. For many applications, it may be an easy and obvious strategy to comply with the requirements of the target structure and to exclude those sites that are found inactive of further considerations. In some cases, however, one may be committed to a particular target sequence.

As it was our aim to develop a single oligonucleotide-based antiviral agent to target numerous

Oligonucleotides

Unmodified oligodeoxynucleotides and phosphorothioates were obtained from MWG-Biotech AG, Ebersberg, Germany. 2′-O-methyl-containing oligonucleotides and RNA oligonucleotides were purchased from IBA GmbH (Göttingen, Germany). Oligonucleotides containing LNA were obtained from Proligo, Boulder, CO, USA. The sequences of the unmodified oligonucleotides and DNAzymes used in this study are:

H5CCG GGG AAA CAG AAG TGC T
DH5CCG GGG AAA GGC TAG CTA CAA CGA AGA AGT GCT
Target sequence DH5AGC ACU UCU GUU

Acknowledgements

The authors are grateful for financial support by the Deutsche Forschungsgemeinschaft (KU 1436/1-1) and the Fonds der Chemischen Industrie. The partial financial support to H.Z. by the Bundesministerium für Bildung und Forschung and the RiNA GmbH (grant no. 0311957A/4.14) and the Gemeinnützige Hertie-Stiftung (grant no. GHS 191/00/02) is gratefully acknowledged. We furthermore thank Proligo, Boulder, USA for supplying LNAs.

References (34)

  • A.R Feldman et al.

    A new and efficient DNA enzyme for the sequence-specific cleavage of RNA

    J. Mol. Biol.

    (2001)
  • L.J Scherer et al.

    Approaches for the sequence-specific knockdown of mRNA

    Nature Biotechnol.

    (2003)
  • J.A Doudna et al.

    The chemical repertoire of natural ribozymes

    Nature

    (2002)
  • J Kurreck

    Antisense technologies. Improvement through novel chemical modifications

    Eur. J. Biochem.

    (2003)
  • I Gautherot et al.

    A multi-model approach to nucleic acid-based drug development

    Biodrugs

    (2004)
  • C.A Stein

    The experimental use of antisense oligonucleotides: a guide for the perplexed

    J. Clin. Invest.

    (2001)
  • M.J Cairns et al.

    Target site selection for an RNA-cleaving catalytic DNA

    Nature Biotechnol.

    (1999)
  • Cited by (54)

    • Development and characterization of DNAzyme candidates demonstrating significant efficiency against human rhinoviruses

      2019, Journal of Allergy and Clinical Immunology
      Citation Excerpt :

      Some are still under clinical evaluation, such as the capsid binders pleconaril or vapendavir (BTA798),68 the major drawback of which is the rapid emergence of resistance.65,68 Finally, several antisense technology–based drug development studies attempting to create candidate therapeutics directly targeting RV infection were reported 10 or more years ago70-73; however, none of those molecules has ever entered clinical trials.17,74 Early or even preventive management of RV infections can reduce the incidence of common colds, asthma/COPD exacerbations, and possibly childhood asthma development.

    • Site-specific cleavage of mutant ABL mRNA by DNAzyme is facilitated by peptide nucleic acid binding to RNA substrate

      2012, FEBS Letters
      Citation Excerpt :

      Previous attempts to improve ribozyme or DNAzyme activity have used facilitator oligonucleotides that bind RNA substrate at the termini of the ribozyme or DNAzyme [8–12]. Alternatively, biding arm of DNAzyme has been modified by incorporation of 2′-O-methyl RNA or locked nucleic acid (LNA) nucleoside analogues to enhance target affinity in vitro [13–16]. One of the facilitator oligonucleotides, peptide nucleic acid (PNA, Fig. 1B), is a DNA/RNA mimic in which the phosphodiester linkage is replaced with a peptide bond [17].

    • Reduction in intracellular HCV RNA and virus protein expression in human hepatoma cells following treatment with 2′-O-methyl-modified anti-core deoxyribozyme

      2008, Virology
      Citation Excerpt :

      Until recently many laboratories tested various oligonucleotide-based platforms for in vitro- or culture-based efficacy using subgenomic RNA fragments (Oketani et al., 1999; Smith et al., 2002; Trepanier et al., 2006b), however one must acknowledge that recognition of subgenomic targets by a therapeutic oligonucleotide does not always faithfully translate into an equivalent or sufficient level of efficacy when challenged with endogenous, full-length RNA (Smith et al., 2007; Swiatkowska et al., 2007). Long RNAs, such as the HCV genome, are highly structured molecules wherein up to 90% of putative cleavage sites were found to be inaccessible upon testing (Schubert et al., 2004). Even with the advent of the HCV 2b JFH-1strain/Huh-7 hepatoma replicative cell culture model, investigators are still restricted in their drug testing (Wakita et al., 2005).

    View all citing articles on Scopus
    View full text