Thermodynamic criteria for high hit rate antisense oligonucleotide design.

Antisense oligonucleotides are used for therapeutic applications and in functional genomic studies. In practice, however, many of the oligonucleotides complementary to an mRNA have little or no antisense activity. Theoretical strategies to improve the 'hit rate' in antisense screens will reduce the cost of discovery and may lead to identification of antisense oligonucleotides with increased potency. Statistical analysis performed on data collected from more than 1000 experiments with phosphorothioate-modified oligonucleotides revealed that the oligo-probes, which form stable duplexes with RNA (DeltaG(o)37 < or = -30 kcal/mol) and have small self-interaction potential, are more frequently efficient than molecules that form less stable oligonucleotide-RNA hybrids or more stable self-structures. To achieve optimal statistical preference, the values for self-interaction should be (DeltaG(o)37) > or = -8 kcal/mol for inter-oligonucleotide pairing and (DeltaG(o)37) > or = -1.1 kcal/mol for intra-molecular pairing. Selection of oligonucleotides with these thermodynamic values in the analyzed experiments would have increased the 'hit rate' by as much as 6-fold.

Fully modified 2' MOE oligonucleotides redirect polyadenylation.

Many genes have been described and characterized that have alternative polyadenylation signals at the 3'-end of their pre-mRNAs. Many of these same messages also contain destabilization motifs responsible for rapid degradation of the mRNA. Polyadenylation site selection can thus determine the stability of an mRNA. Fully modified 2'-O:-methoxy ethyl/phosphorothioate oligonucleotides that hybridize to the 3'-most polyadenylation site or signal of E-selectin were able to inhibit polyadenylation at this site and redirect it to one of two upstream cryptic sites. The shorter transcripts produced after antisense treatment have fewer destabilization sequences, increased mRNA stability and altered protein expression. This study demonstrates that antisense oligonucleotides can be successfully employed to redirect polyadenylation. This is the first demonstration of the use of oligonucleotides to increase, rather than decrease, abundance of a message.

Identification of sequence motifs in oligonucleotides whose presence is correlated with antisense activity.

Design of antisense oligonucleotides targeting any mRNA can be much more efficient when several activity-enhancing motifs are included and activity-decreasing motifs are avoided. This conclusion was made after statistical analysis of data collected from >1000 experiments with phosphorothioate-modified oligonucleotides. Highly significant positive correlation between the presence of motifs CCAC, TCCC, ACTC, GCCA and CTCT in the oligonucleotide and its antisense efficiency was demonstrated. In addition, negative correlation was revealed for the motifs GGGG, ACTG, AAA and TAA. It was found that the likelihood of activity of an oligonucleotide against a desired mRNA target is sequence motif content dependent.

Effects of RNA secondary structure on cellular antisense activity.

The secondary and tertiary structures of a mRNA are known to effect hybridization efficiency and potency of antisense oligonucleotides in vitro. Additional factors including oligonucleotide stability and cellular uptake are also thought to contribute to antisense potency in vivo. Each of these factors can be affected by the sequence of the oligonucleotide. Although mRNA structure is presumed to be a critical determinant of antisense activity in cells, to date little direct experimental evidence has addressed the significance of structure. In order to determine the importance of mRNA structure on antisense activity, oligonucleotide target sites were cloned into a luciferase reporter gene along with adjoining sequence to form known structures. This allowed us to study the effect of target secondary structure on oligonucleotide binding in the cellular environment without changing the sequence of the oligonucleotide. Our results show that structure does play a significant role in determining oligonucleotide efficacy in vivo. We also show that potency of oligonucleotides can be improved by altering chemistry to increase affinity for the mRNA target even in a region that is highly structured.

Predicting oligonucleotide affinity to nucleic acid targets.

A computer program, OligoWalk, is reported that predicts the equilibrium affinity of complementary DNA or RNA oligonucleotides to an RNA target. This program considers the predicted stability of the oligonucleotide-target helix and the competition with predicted secondary structure of both the target and the oligonucleotide. Both unimolecular and bimolecular oligonucleotide self structure are considered with a user-defined concentration. The application of OligoWalk is illustrated with three comparisons to experimental results drawn from the literature.

What affects the effect of 2'-alkoxy modifications? 1. Stabilization effect of 2'-methoxy substitutions in uniformly modified DNA oligonucleotides.

The thermostability of hybrid duplexes with uniformly 2'-methoxy modified DNA strands (D'R and RD'), their unmodified DNA:RNA counterparts (DR and RD), and corresponded RNA:RNA (RR) duplexes for six sequences with different GC and deoxypyrimidine (dPy) content was measured. The linear correlation between the total stabilization effect of 2'-methoxy modifications (Delta DeltaG(o)37(D'R-DR)) and the relative stability of corresponding unmodified hybrids compared to the RR counterparts (Delta DeltaG(o)37(RR-DR)) suggests that the initial conformational and the thermodynamic state of the "parent" unmodified hybrid governs the effect of 2'-methoxy (and may be other 2'-alkoxy) modifications whose mechanism of action includes an S --> N conformational shift resulting in an RNA-like A-form duplex. We also found a correlation between the "hydrophobic" part of the total effect (Delta DeltaG(o)37(D'R-RR)) and the dA fraction in the modified DNA strand, suggesting that the "hydrophobic" effect of the 2'-methoxy groups results mainly from intraresidue steric effects increasing rigidity of the modified sugar rings. The correlations observed enabled us to predict the stability of hybrids with 2'-methoxy modified DNA strands for any sequence except for sequences with (dU)10 and (dA)10 strings.

Evaluation of pyrimidine PNA binding to ssDNA targets from nonequilibrium melting experiments.

Slow kinetics of homopyrimidine PNA binding to single stranded DNA and RNA targets is manifested in significant hysteresis in thermal UV absorption experiments. We have compared temperatures of dissociation (Tdis) and reassociation (Tass) for triplexes formed by DNA and single or bis PNAs with K50 derived from gel mobility experiments. Results indicated there was no correlation between Tdis and K50 while reasonable correlation between Tass and K50 was found. This correlation enabled use of easy thermal UV absorption experiments for evaluation of PNA binding to DNA/RNA targets.

Strategies for rapid deconvolution of combinational libraries: comparative evaluation using a model system.

Synthesis and testing of complex mixtures maximize the number of compounds that can be prepared and tested in a combinatorial library. When mixtures of compounds are screened, however, the identity of the compound(s) selected may depend on the deconvolution procedure employed. Previously, we developed a model system for evaluation of deconvolution procedures and used it to compare pooling strategies for iterative and noniterative deconvolution [Freier et al. J. Med. Chem. 1995, 38, 344-352]. We have now extended the model studies to include simulations of procedures with overlapping subsets such as subtractive pooling [Carell et al. Angew, Chem., Int. Ed. Engl. 1994, 33, 2061-2064], bogus coin pooling [Blake and Litzi-Davis. Bioconjugate Chem. 1992, 3, 510-513], and orthogonal pooling [D'Prez et al. J. Am. Chem. Soc. 1995, 117, 5405-5406]. These strategies required synthesis and testing of fewer subsets than did the more traditional nonoverlapping iterative strategies. The compounds identified using simulations of these strategies, however, were not the most active compounds in the library and were substantially less active than those identified by simulations of more traditional strategies.

The ups and downs of nucleic acid duplex stability: structure-stability studies on chemically-modified DNA:RNA duplexes.

In an effort to discover novel oligonucleotide modifications for antisense therapeutics, we have prepared oligodeoxyribonucleotides containing more than 200 different modifications and measured their affinities for complementary RNA. These include modifications to the heterocyclic bases, the deoxy-ribose sugar and the phosphodiester linkage. From these results, we have been able to determine structure-activity relationships that correlate hybridization affinity with changes in oligonucleotide structure. Data for oligonucleotides containing modified pyrimidine nucleotides are presented. In general, modifications that resulted in the most stable duplexes contained a heteroatom at the 2'-position of the sugar. Other sugar modifications usually led to diminished hybrid stability. Most backbone modifications that led to improved hybridization restricted backbone mobility and resulted in an A-type sugar pucker for the residue 5'to the modified internucleotide linkage. Among the heterocycles, C-5-substituted pyrimidines stood out as substantially increasing duplex stability.

Sequence-specific antitumor activity of a phosphorothioate oligodeoxyribonucleotide targeted to human C-raf kinase supports an antisense mechanism of action in vivo.

To determine the mechanism of action responsible for the in vivo antitumor activity of a phosphorothioate antisense inhibitor targeted against human C-raf kinase (ISIS 5132, also known as CGP69846A), a series of mismatched phosphorothioate analogs of ISIS 5132 or CGP69846A were synthesized and characterized with respect to hybridization affinity, inhibitory effects on C-raf gene expression in vitro, and antitumor activity in vivo. Incorporation of a single mismatch into the sequence of ISIS 5132 or CGP69846A resulted in reduced hybridization affinity toward C-raf RNA sequences and reduced inhibitory activity against C-raf expression in vitro and tumor growth in vivo. Moreover, incorporation of additional mismatches resulted in further loss of in vitro and in vivo activity in a manner that correlated well with a hybridization-based (i.e., antisense) mechanism of action. These results provide important experimental evidence supporting an antisense mechanism of action underlying the in vivo antitumor activity displayed by ISIS 5132 or CGP69846A.

Deconvolution of combinatorial libraries for Drug discovery: theoretical comparison of pooling strategies.

Synthesis and testing of mixtures of compounds in a combinatorial library allow much greater throughput than synthesis and testing of individual compounds. When mixtures of compounds are screened, however, the possibility exists that the most active compound will not be identified. The specific strategies employed for pooling and deconvolution will affect the likelihood of success. We have used a nucleic acid hybridization example to develop a theoretical model of library deconvolution for a library of more than 250,000 compounds. This model was used to compare various strategies for pooling and deconvolution. Simulations were performed in the absence and presence of experimental error. We found iterative deconvolution to be most reliable when active molecules were assigned to the same subset in early rounds. Reliability was reduced only slightly when active molecules were assigned randomly to all subsets. Iterative deconvolution with as many as 65,536 compounds per subset did not drastically reduce the reliability compared to one-at-a-time testing. Pooling strategies compared using this theoretical model are compared experimentally in an accompanying paper.

Deconvolution of combinatorial libraries for drug discovery: experimental comparison of pooling strategies.

An experimental evaluation of several different pooling strategies for combinatorial libraries was conducted using a library of 810 compounds and an enzyme inhibition assay (phospholipase A2). The library contained compounds with varying degrees of activity as well as inactive compounds. The compounds were synthesized in groups of three and pooled together in various formats to realize different pooling strategies. With one exception, all iterative deconvolution strategies and position scanning resulted in identification of the same compound. The results are in good agreement with the predicted outcome from theoretical and computational methods. These data support the tenet that active compounds for pharmaceutically relevant targets can be successfully identified from combinatorial libraries organized in mixtures.

Kinetics of G-quartet-mediated tetramer formation.

The phosphorothioate and phosphodiester oligodeoxynucleotides d(TTGGGGTT) form parallel-stranded tetramer structures stabilized by guanosine quartets. The phosphorothioate tetramer has been shown to inhibit human immunodeficiency virus (HIV) in vitro. The kinetics of association and dissociation of both tetramers have been determined as a function of temperature using size exclusion chromatography to measure the ratio of single strand to tetramer. In phosphate buffered saline (pH 7.2) at 37 degrees C, the fourth-order association rate of the phosphorothioate tetramer was 6.1 (+/- 0.5) x 10(4) M-3 s-1; the dissociation rate was 8.2 (+/- 0.2) x 10(-6) min-1, resulting in a t(1/2) of about 60 days. The association rate of the phosphodiester was about one order of magnitude faster and the dissociation rate about one order of magnitude slower than that of the phosphorothioate tetramer. The association reaction had a negative energy of activation for both compounds. Despite thermodynamic instability of the tetramer at low concentrations, the extremely slow dissociation rate may allow use of the phosphorothioate tetramer for AIDS chemotherapy.

'Mutational SURF': a strategy for improving lead compounds identified from combinatorial libraries.

Synthesis and testing of mixtures of compounds in a combinatorial library offers the potential of much greater throughput than the 'one compound, one well' approach. When mixtures of compounds are screened, however, pooling and deconvolution strategies must be employed to identify the most active compound in the library. The possibility exists that the most active compound will not be identified. We have developed a theoretical model of library deconvolution using the well characterized properties of nucleic acid hybridization to calculate activities of individual molecules in libraries of more than 250,000 compounds. Calculations using this model have been employed to evaluate strategies for pooling and deconvolution. In the presence of errors in synthesis and testing, iterative deconvolution or position scanning sometimes identified a compound with sub-optimal activity. We describe a procedure called 'mutational SURF' in which 'mutants' of the selected compound are individually synthesized and tested. Simulations of mutational SURF using our model libraries suggest that mutational SURF provides an efficient method for improving the activity of lead compounds identified from combinatorial libraries.

Inhibition of NF-kappa B specific transcriptional activation by PNA strand invasion.

Peptide nucleic acid (PNA) strand invasion offers an attractive alternative to DNA oligonucleotide directed triplex formation as a potential tool for gene inhibition. Peptide nucleic acid has been shown to interact with duplex DNA in a process which involves strand invasion of the duplex and binding of one of the DNA strands with two PNA oligomers. By blocking the interaction of a transcription factor with 5' regulatory sequences, PNA might specifically down-regulate gene activity. Here we demonstrate that PNA is capable of specifically blocking interaction of the transcription factor NF-kappa B with the IL2-R alpha NF kappa-B binding site in vitro. We further demonstrate that this interaction is sufficient to prevent transcriptional transactivation both in vitro and when transfected into cells in culture.

Relative thermodynamic stability of DNA, RNA, and DNA:RNA hybrid duplexes: relationship with base composition and structure.

Fourteen oligonucleotides 8-21 nucleotides in length and their complements were synthesized as DNA and RNA. For each sequence, four kinds of duplexes, DNA:DNA, RNA:RNA, DNA:RNA, and RNA:DNA, were prepared. Twelve sequences had A.T/U content varying from 25 to 80% and dPy content in the DNA strands varying from 0 to 100%. Thermodynamic stabilities of four duplexes for each sequence were determined in solution containing 100 mM Na+, 10 mM phosphate, and 0.1 mM EDTA, pH 7.1. CD spectra and electrophoretic mobility on native polyacrylamide gel were measured for most duplexes. Quantitative correlations of hybrid stability both with deoxypyrimidine content and, at fixed dPy content, with the fraction of A.T/U in duplexes were found. We also demonstrated that hybrids with 70-80% deoxypyrimidine DNA strand and a high or moderate A.T/U fraction displayed the highest relative stability compared to their RNA counterparts. Relationships of relative intensities of CD bands at 210 nm and relative electrophoretic mobilities of hybrids with relative hybrid stability suggested that hybrid conformation varies continuously between A- and B-form and is the decisive factor in relative hybrid stability.

Characterization of fully 2'-modified oligoribonucleotide hetero- and homoduplex hybridization and nuclease sensitivity.

The nuclease stability and melting temperatures (Tm) were compared for fully modified oligoribonucleotide sequences containing 2'-fluoro, 2'-O-methyl, 2'-O-propyl and 2'-O-pentyl nucleotides. Duplexes formed between 2' modified oligoribonucleotides and RNA have typical A-form geometry as observed by circular dichroism spectroscopy. Modifications, with the exception of 2'-O-pentyl, were observed to increase the Tm of duplexes formed with complementary RNA. Modified homoduplexes showed significantly higher Tms, with the following Tm order: 2'-fluoro:2'fluoro > 2'-O-propyl:2'-O-propyl > 2'-O-methyl:2'-O- methyl > RNA:RNA > DNA:DNA. The nuclease stability of 2'-modified oligoribonucleotides was examined using snake venom phosphodiesterase (SVPD) and nuclease S1. The stability imparted by 2' modifications was observed to correlate with the size of the modification. An additional level of nuclease stability was present in oligoribonucleotides having the potential for forming secondary structure, but only for 2' modified oligoribonucleotides and not for 2'-deoxy oligoribonucleotides.

Deconvolution of combinatorial libraries for drug discovery: a model system.

Iterative synthesis and screening strategies have recently been used to identify unique active molecules from complex synthetic combinatorial libraries. These techniques have many advantages over traditional screening methods, including the potential to screen large numbers of compounds to identify an active molecule while avoiding analytical separations and structural determination of unknown compounds. It is not clear, however, whether these techniques identify the most active molecular species in the mixtures and, if so, how often. Two key factors which may affect success of the selection process are the presence of many active compounds in the library with a range of activities and the chosen order of unrandomization. The importance of these factors has not been previously studied. Moreover, the impact of experimental errors in determination of subset activities or in randomization during library synthesis is not known. We describe here a model system based on oligonucleotide hybridization that addresses these questions using computer simulations. The results suggested that, within achievable experimental and library synthesis error, iterative deconvolution methods generally find either the best molecule or one with activity very close to the best. The presence of many active compounds in a library influenced the profile of subset activities, but did not preclude selection of a molecule with near optimal activity.