Antisense properties of 2'-O-dimethylaminooxyethyl (2'-O-DMAOE) oligonucleotides.

Antisense oligonucleotides with 2'-O-(2-[N,N-dimethyl)aminooxy]ethyl) or (2'-O-DMAOE) modification were synthesized and evaluated for nuclease resistance and pharmacology both in vitro and in vivo. This modification exhibits very high nuclease resistance and efficacy in various biological (ICAM-1, C-raf and PKC-alpha) targets.

2'-O-[2-[N,N-(dialkyl)aminooxy]ethyl]-modified antisense oligonucleotides.

[structure] Oligonucleotides with two novel modifications, 2'-O-¿2-[N, N-(dimethyl)aminooxy]ethyl¿ (2'-O-DMAOE) and 2'-O-¿2-[N, N-(diethyl)aminooxy]ethyl¿ (2'-O-DEAOE), have been synthesized. These modifications exhibit high binding affinity to target RNA (and not to DNA) and enhance the nuclease stability of oligonucleotides considerably with t(1/2) > 24 h as a phosphodiester.

Aminooxy functionalized oligonucleotides: preparation, on-support derivatization, and postsynthetic attachment to polymer support.

Three novel phosphoramidites, each bearing a phthaloyl-protected aminooxy tail, were prepared and applied in automated oligonucleotide synthesis. After chain assembly, the phthaloyl protection was removed with hydrazinium acetate. Normal succinyl linker turned to be stable under these conditions, and hence the support-bound oligonucleotide could be converted to a pyrene oxime conjugates by reacting with pyrene carbaldehyde or cis-retinal. Standard ammonolytic deprotection then released the deprotected conjugate in solution. Alternatively, the crude aminooxy-tethered oligonucleotide was immobilized to microscopic polymer particles by reacting the aminooxy function with the particle-bound aldehyde or epoxide groups. These immobilized oligonuceotides were shown to serve properly as probes in a mixed phase hybridization assay.

Syntheses, opioid binding affinities, and potencies of dynorphin A analogues substituted in positions, 1, 6, 7, 8 and 10.

Structural, stereochemical, stereoelectronic and conformational requirements for biological activity of dynorphin A1-11-NH2 analogues at opioid receptors were explored by substitution of Tyr1, Arg6, Arg7, Ile8 and Pro10 with other amino acid residues. Interestingly, substitution of Tyr1 with N alpha-Ac-Tyr1, D-Tyr1, Phe1 or p-BrPhe1 led to analogues that were quite potent at kappa opioid receptors, and additional substitution of Ile8 with D-Ala8 and/or Pro10 with D-Pro10 retained high potency in brain binding assay: [N alpha-Ac-Tyr1]- (1), [D-Tyr1]-(2) [Phe1]- (3), [Phe1,D-Ala8]- (5), [-BrPhe1, D-Ala8]- (6), [Phe1, D-Pro10]- (7) and [Phe1,D-Ala8, D-Pro10]- Dyn A1-11-NH2 (8) had IC50 (nM) binding affinities of 13.2, 18.6, 1.64, 1.26, 1.84, 2.44 and 1.62 nM, respectively. The D-Phe1 analogue 4, however, was only weakly active (610 nM). All of the analogues except 4 were modestly selective for kappa vs. mu guinea pig brain opioid receptor (11- to 88-fold) and quite selective for kappa vs. delta receptors (65-576). However, all of the analogues appeared to have very low or essentially no activity in the guinea pig ileum and mouse vas deference functional bioassays, and one analogue, 5, appeared to have weak antagonist activities. On the other hand, if constrained amino acids such as beta-methylphenylalanine or 1,2,3,4-tetrahydroisoquinoline carboxylic acid, and hydroxyproline were placed in the 1 position, inactive analogues or analogues with greatly reduced potency and biological activity were obtained (compounds 12-14). It had previously been suggested that the Arg6 and Arg7 residues were critical for biological activity.(ABSTRACT TRUNCATED AT 250 WORDS)

Oligodeoxynucleotides containing 2'-O-modified adenosine: synthesis and effects on stability of DNA:RNA duplexes.

Hybridization thermodynamics were compared for oligonucleotide sequences containing 2'-fluoro dA, 2'-O-methyl A, 2'-O-ethyl A, 2'-O-propyl A, 2'-O-butyl A, 2'-O-pentyl A, 2'-O-nonyl A, 2'-O-allyl A, and 2'-O-benzyl A in place of deoxyadenosine. Although the effect of 2'-modified adenosine on duplex stability is sequence dependent, a clear trend is apparent. For six sequences containing a few 2'-modified adenosines in a background of unmodified deoxynucleotides, the average delta TM per substitution ranged from +1.3 degrees C for 2'-fluoro dA to -2.0 degrees C for 2'-O-nonyl A. For the 2'-O-alkyl series, the average delta TM per substitution correlates well with size of the substituent; the order of stability is 2'-O-methyl A > 2'-O-ethyl A > 2'-O-propyl A > 2'-O-butyl A > 2'-O-pentyl A > 2'-O-nonyl A. This correlation also extends to 2'-fluoro dA, 2'-O-allyl A, and 2'-O-benzyl A if chain length is measured by number of carbon atoms. When examined in the background of 2'-O-methyl ribonucleotides, all 2'-modified adenosines with a substituent no larger than 2'-O-pentyl stabilized the duplex nearly 2 degrees C per substitution compared to unmodified dA. These thermodynamic results and CD spectra of modified and unmodified hybrids support a model of DNA:RNA hybrids in which the geometry is between that of B-form and A-form.

Evaluation of 2'-modified oligonucleotides containing 2'-deoxy gaps as antisense inhibitors of gene expression.

We have used a previously described 17-mer phosphorothioate (Monia, B.P., Johnston, J.F., Ecker, D. J., Zounes, M.A., Lima, W.F., and Freier, S.M. (1992) J. Biol. Chem. 267, 19954-19962) for structure-function analysis of 2'-sugar modifications including 2'-O-methyl, 2'-O-propyl, 2'-O-pentyl, and 2'-fluoro. These modifications were analyzed for hybridization affinity to complementary RNA and for antisense activity against the Ha-ras oncogene in cells using a highly sensitive transactivation reporter gene system. Hybridization analysis demonstrated that all of the 2'-modified oligonucleotides hybridized with greater affinity to RNA than an unmodified 2'-deoxy oligonucleotide with the rank order of affinity being 2'-fluoro > 2'-O-methyl > 2'-O-propyl > 2'-O-pentyl > 2'-deoxy. Evaluation of antisense activities of uniformly 2'-modified oligonucleotides revealed that these compounds were completely ineffective in inhibiting Ha-ras gene expression. Activity was restored if the compound contained a stretch of at least five 2'-deoxy residues. This minimum deoxy length correlated perfectly with the minimum length required for efficient RNase H activation in vitro using partially purified mammalian RNase H enzyme. These chimeric 2'-modified/deoxy phosphorothioates displayed greater antisense potencies in inhibiting Ha-ras gene expression, compared with the unmodified uniform deoxy phosphorothioate. Furthermore, antisense potency correlated directly with affinity of a given 2' modification for it's complementary RNA. These results demonstrate the importance of target affinity in the action of antisense oligonucleotides and of RNase H as a mechanism by which these compounds exert their effects.

Uniformly modified 2'-deoxy-2'-fluoro phosphorothioate oligonucleotides as nuclease-resistant antisense compounds with high affinity and specificity for RNA targets.

"Uniformly" modified phosphodiester or phosphorothioate oligonucleotides incorporating 2'-deoxy-2'-fluoroadenosine, -guanosine, -uridine, and -cytidine, reported herein for the first time, when hybridized with RNA afforded consistent additive enhancement of duplex stability without compromising base-pair specificity. CD spectra of the 2'-deoxy-2'-fluoro-modified oligonucleotides hybridized with RNA indicated that the duplex adopts a fully A-form conformation. The 2'-deoxy-2'-fluoro-modified oligonucleotides in phosphodiester form were not resistant to nucleases; however, the modified phosphorothioate oligonucleotides were highly nuclease resistant and retained exceptional binding affinity to the RNA targets. The stabilizing effects of the 2'-deoxy-2'-fluoro modifications on RNA-DNA duplexes were shown to be superior to those of the 2'-O-methylribo substitutions. RNA hybrid duplexes with uniformly 2'-deoxy-2'-fluoro-modified oligonucleotides did not support HeLa RNase H activity; however, incorporation of the modifications into "chimeric" oligonucleotides has been shown to activate mammalian RNase H. "Uniformly" modified 2'-deoxy-2'-fluoro phosphorothioate oligonucleotides afforded antisense molecules with (1) high binding affinity and selectivity for the RNA target and (2) stability toward nucleases.

Design and synthesis of highly potent and selective cyclic dynorphin A analogs. 2. New analogs.

We have designed and synthesized several cyclic disulfide-containing peptide analogs of dynorphin A (Dyn A) which are conformationally constrained in the putative "address" segment of the opioid ligand. Several of these Dyn A analogs exhibit unexpected apparent selectivities for the kappa and mu opioid receptors(s) of the central vs peripheral nervous systems. Thus, incorporation of conformational constraint in the putative "address" segment of Dyn A analogs has resulted in the kappa/mu opioid receptor ligands [L-Pen5,Cys11]Dyn A1-11-NH2 (4), [Cys5,Cys10]Dyn A1-11-NH2 (5), [Cys5,Cys9]DynA1-11-NH2 (6), and [Cys4,Cys9,Arg10]DynA1-11-NH2(7). All of these analogs possess high kappa and mu opioid receptor affinities for the central receptor (guinea pig brain), but effect only weak potency at peripheral kappa and mu opioid receptors (GPI). In fact cyclic dynorphin A analog 4 shows > 19,000-fold differences between central kappa opioid affinity and potency in the guinea pig ileum (GPI). Additionally analog 4 is not an antagonist in the GPI, suggesting possible receptor differences between these sites. Substitution of Tyr1 by Phe1 in the cyclic 1-11 series gave the analog [Phe1,Cys5,Cys11]Dyn A1-11-NH2 (1) that was surprisingly potent in the guinea pig brain binding assay (IC50 = 15.1 nM) at the kappa receptor, but was inactive in the GPI and mouse vas deferens bioassays. D-Ala2 and Tic4 analogs of 1 had lower affinity at brain kappa receptors and had very weak potencies in the GPI and MVD bioassays. On the other hand, [Cys6,Cys10]DynA1-11-NH2 (8), [Cys8,D-Cys13]DynA1-13-NH2 (9), [D-Cys8,D-Cys12]DynA1-13-NH2 (10), and [D-Pro10,Cys5,Cys13]-Dyn A1-13-NH2 (11) were surprisingly potent in the GPI bioassay, though considerable apparent selectivity for central receptors is still retained. The apparent lack of correlation between the pharmacological profiles observed in smooth muscle and in the brain binding assays, particularly with 1 and 4, may suggest the existence of different subtypes of the kappa and mu opioid receptors in the brain and peripheral systems.

Synthesis and cytotoxicity studies of 8-amino-6-methyl-2-beta-D-ribofuranosyl-1,2,3,5,6,7-hexaazaacenaphthyle ne (7-aza-TCN) and the corresponding 2'-deoxy- and arabinonucleoside analogues.

The tricyclic nucleoside 8-amino-6-N-methyl-2-beta-D-ribofuranosyl-1,2,3,5,6, 7-hexaazaacenaphthylene was synthesized from 8-amino-6-N-methyl-4-(methylthio)-2-beta-D-ribofuranosyl-1,2,3,5,6, 7-hexaazaacenaphthylene. The 2'-deoxy analogue of 5, 8-amino-6-N-methyl-2-(2-deoxy-beta-D-ribofuranosyl)-1,2,3,5,6, 7-hexaazaacenaphthylene (11), and the arabino analogue of (5), 8-amino-6-N-methyl-2-beta-D-arabinofuranosyl-1,2,3,5,6, 7-hexaazaacenaphthylene (14) were synthesized from 5. Nucleosides 2,3,4,5,11, and 14 were evaluated for potential anticancer activity by measuring their ability to inhibit the growth of L1210 and H. Ep. 2 tumor cells in vitro.

Design and synthesis of highly potent and selective cyclic dynorphin A analogues.

We have designed and synthesized several cyclic disulfide-containing peptide analogues of dynorphin A (Dyn A) which are conformationally constrained in the putative "address" segment of the opioid ligand. Several of these Dyn A analogues exhibit unexpected selectivities for the kappa and mu opioid receptors(s) of the central vs peripheral nervous systems. Thus, incorporation of conformational constraint in the putative "address" segment of Dyn A analogues has resulted in the kappa/mu opioid receptor ligands [Cys5,Cys11]Dyn A1-11-NH2 (1) and [Cys5,Cys11,D-Ala8]Dyn A1-11-NH2 (2), which possess high kappa and mu opioid receptor affinities centrally (guinea pig brain, GPB), but only weak activity at peripheral kappa and mu opioid receptors (guinea pig ileum, GPI). On the other hand, [Cys8,Cys13]Dyn A1-13-NH2 and [D-Cys8,D-Cys13]Dyn A1-13-NH2 (5) display high kappa potencies and selectivities at the peripheral (GPI) but not at the central (GPB) kappa opioid receptor. The lack of correlation between the pharmacological profiles observed in smooth muscle and in the brain binding assays suggests the existence of different subtypes of the kappa and mu opioid receptors in the brain and peripheral nervous systems.

The synthesis and biological activity of certain pentaazaacenaphthylenes, hexaazaacenaphthylenes and their corresponding nucleosides.

Synthesis of the tricyclic nucleoside 8-amino-6-N-methyl-2-(beta-D-ribo-furanosyl)-1,2,3,5,6,7-hexaazaacena phthylene (5) has been accomplished by the ring closure of an appropriately substituted pyrazolo[3,4-d]-pyrimidine nucleoside followed by the requisite chemical conversions. The formation, isolation and structural elucidation of two unexpected nucleosides formed by a reductive ring cleavage of the hexaazaacenaphthylene ring system is discussed. A comparison of the antitumor and biological activity of 5 with the structurally related tricyclic pentaazaacenaphthylene nucleoside which is currently in phase II clinical trials at the 5'-phosphate pro-drug is also presented.