Peptide nucleic acid (PNA) is capable of enhancing hammerhead ribozyme activity with long but not with short RNA substrates
Peptide nucleic acid (PNA) is capable of enhancing hammerhead ribozyme activity with long but not with short RNA substratesEckhard Jankowsky+, Günther Strunk1 and Bernd Schwenzer*
Institut für Biochemie, Technische Universität Dresden, Mommsenstraße 13, 01069 Dresden, Germany and 1Abteilung Biochemische Kinetik, Max-Planck-Institut für Biophysikalische Chemie, Am Faßberg, 37018 Göttingen, Germany
Received May 9, 1997;Revised and Accepted May 30, 1997
ABSTRACT
Long RNA substrates are inefficiently cleaved by hammerhead ribozymes in trans. Oligonucleotide facilitators capable of affecting the ribozyme activity by interacting with the substrates at the termini of the ribozyme provide a possibility to improve ribozyme mediated cleavage of long RNA substrates. We have examined the effect of PNA as facilitator in vitro in order to test if even artificial compounds have facilitating potential. Effects of 12mer PNA- (peptide nucleic acid), RNA- and DNA-facilitators of identical sequence were measured with three substrates containing either 942, 452 or 39 nucleotides. The PNA facilitator enhances the ribozyme activity with both, the 942mer and the 452mer substrate to a slightly smaller extent than RNA and DNA facilitators. This effect was observed up to PNA facilitator:substrate ratios of 200:1. The enhancement becomes smaller as the PNA facilitator:substrate ratio exceeds 200:1. With the 39mer substrate, the PNA facilitator decreases the ribozyme activity by more than 100-fold, even at PNA facilitator:substrate ratios of 1:1. Although with long substrates the effect of the PNA facilitator is slightly smaller than the effect of identical RNA or DNA facilitators, PNA may be a more practical choice for potential applications in vivo because PNA is much more resistant to degradation by cellular enzymes.
INTRODUCTION
Hammerhead ribozymes are catalytic RNA molecules capable of cleaving RNA substrates in trans (1 ,2 ). By choosing target complementary stems flanking the catalytic hammerhead motif, ribozyme activity can be directed against almost any RNA and may thereby be useful for therapeutical purposes (3 -5 ).
Usually the intention is to inactivate RNAs containing several hundred nucleotides (nt). However, compared with substrates of 20-60 nt in length, ribozyme activity decreases as the length of the substrates increases (6 ). For potential therapeutical applications it is essential that hammerhead ribozymes are able to cleave long RNAs as well. Therefore, it is important to explore possibilities of enhancing hammerhead activity to cleave long RNAs (7 ).
Besides the improvement of ribozyme activity through the addition of proteins (8 ,9 ), oligonucleotides capable of interacting with the substrate at the termini of the ribozyme, so called oligo- nucleotide facilitators (10 ,11 ), were found to enhance hammerhead activity with long RNAs (12 ).
RNA- as well as DNA-facilitators have the potential to pre-form potentially structured substrates for ribozyme attack (13 ,14 ). Since higher order structures are considered to be the main hindrance for efficient ribozyme-mediated cleavage of long RNAs (6 ,15 ), it is pertinent to examine further possibilities in order to force structural changes of the substrates such that hammerhead activity is enhanced.
For this purpose peptide nucleic acid (PNA) oligonucleotides are promising. PNA is capable of binding with nucleic acids by Watson-Crick base pairing (16 ,17 ). The stability of PNA-RNA hybrids is considerably less sensitive to cation concentration than the hybrid stability of natural nucleic acids and PNA-RNA hybrids are usually more stable then RNA-RNA or RNA-DNA hybrids, respectively (17 ). Moreover, homopyrimidine PNA is able to bind with double stranded nucleic acids by a strand displacement mechanism (18 ,19 ). With a view to a potential application in cellular systems, of primary interest is the resistance of PNA to both cellular nucleases and proteases (20 ).
Attracted by these potentially favorable properties of PNA, we examined invitro the effect of a PNA facilitator on the hammerhead ribozyme activity with substrates of different length.
MATERIALS AND METHODS
RNA/DNA synthesis and labelling
The ribozyme, short substrates (39mer), RNA- and DNA-facilitators were chemically synthesized and purified as described (14 ,21 ). Long substrates (942mer and 452mer) were synthesized by T7 in vitro transcription from PCR-generated templates (14 ). 39mer substrates were 5'-labelled with [[gamma]-32P]ATP, long substrates were internally labelled during transcription (14 ).
PNA synthesis
The 12mer PNA (H-TGA AGG GTT TGG-NH2) was synthesized by solid phase t-Boc chemistry as described by Nielsen et al. (16 ) using an automated PNA-synthesizer (PerSeptive Biosystems). The crude product of a 1 [mu]mol setup was purified by reversed phase HPLC on a Ultrasphere ODS C-18, 4.6 * 250 mm column (Beckman) using an acetonitrile/TFA gradient (from A to B linear in 30 min: A, water/0.1% TFA; B, acetonitrile/0.1% TFA).
Ribozyme reactions
Ribozyme reactions were performed at 37oC in 10-20 [mu]l 50 mM Tris-HCl (pH 7.5) with 10 mM MgCl2 as described (14 ). Aliquots were taken at appropriate times and quenched by addition of 8 [mu]l ice cold stop buffer (8 M urea, 50 mM EDTA, 7.5% glycerine, 0.05% bromophenol blue and 0.05% xylene cyano blue). Products were separated from uncleaved substrate by either 4% denaturing PAGE (long substrates) or 20% denaturing PAGE (39mer substrates). Gels were analyzed by radioanalytic scanning.
RESULTS
The hammerhead substrates (Fig. 1 A) represent domains of the processed human tissue factor mRNA (22 ). The ribozyme was constructed with 7 nt in every stem in order to ensure the combination of sufficient catalytic activity and high substrate specificity (23 ,24 ). 12mer oligonucleotide facilitators which complement the substrat adjacent to the 3'-end of the ribozyme were recently proved to be efficient for enhancing the hammerhead activity with the long substrates (14 ).
DISCUSSION
PNA facilitators affect the ribozyme activity with all substrates tested. While with long substrates an enhancement was observed in the same order of magnitude as the enhancement caused by RNA and DNA facilitators, the ribozyme activity was strongly inhibited with a short substrate as the PNA-substrate complex was formed to full extent.
With the long substrates, a 10-fold excess of PNA facilitator is necessary to achieve the maximal enhancement at the concentrations used. This is very similar to the excess of RNA and DNA facilitator required for their maximal enhancement. Since the dependence of the facilitator effect on the facilitator:substrate ratio is determined by the stability of the facilitator-substrate complex (14 ), it is evident that under the reaction conditions the PNA-long-substrate complexes are of similar stability as the complexes formed between DNA/RNA facilitators and these substrates. Thus, in our system, PNA is not more efficient than an RNA or a DNA facilitator in pre-forming long RNA substrates for ribozyme attack.
The inhibition of the hammerhead activity by higher PNA substrate ratios might be a result of multiple binding of PNA to the substrate.
The unexpected inhibition of the hammerhead activity with the 39mer substrates should be the result of an influence of the cleavage step, which is suggested by the inhibition of both reactions with and without ribozyme-substrate preannealing. The inhibiting effect is also independent of whether the PNA facilitator binds 3'-end or 5'-end to the ribozyme (Fig. 3 C). Non-specific binding of the PNA to the ribozyme could be ruled out, since the reactions were performed under single turnover conditions, i.e., with an excess of ribozyme over the substrate. Under these conditions, direct binding of the PNA to the ribozyme have resulted in a `trapping' of a fraction of ribozyme and the reduction of the activity would have been considerably smaller than actually observed.
Most likely, formation of multimeric PNAn-RNA aggregates causes the inhibition at the conditions used. Since no more than two bands on native PAGE appeared (Fig. 3 A), these interactions should be weak. Because the ribozyme's catalytic core has inherent flexibility (25 ,26 ), the ribozyme-substrate complex might be trapped in less active conformations forced by these aggregations.
Although, in our system, the effect of a PNA facilitator is not higher than the effect of a DNA or RNA facilitator, PNA is useful for enhancing the activity of a hammerhead ribozyme to cleave long RNA substrates. Due to the resistance of PNA against nucleases and proteases this result is of interest especially for a potential invivo application.
ACKNOWLEDGEMENTS
We thank Dr Anna Marie Pyle and Justin B. Green for helpful comments on the manuscript. We are grateful to the researchers at the Institute of Physiological Chemistry of the Dresden University for the opportunity to make use of their laboratory facilities. Our work was supported by a grant from the DFG (Schw638/1-1).
+Present address: Department of Biochemistry and Molecular Biophysics, College of Physicians and Surgeons, Columbia University, New York, NY 10032, USA
