Search

Despite extensive investigations into oligonucleotide lipoplexes, virtually no work has addressed whether the physicochemical properties of these assemblies vary as a function of the constituent oligonucleotide (ODN) sequence and/or composition. The present study was aimed at answering this question. To this end, we complexed N-(1-(2,3-dioleoyloxy)propyl)-N,N,N-trimethylammonium chloride (DOTAP) liposomes, in dispersion, with either 18-mer phosphorothiote homo-oligonucleotides composed of either adenine, thymidine or cytosine; or one of three structurally related 18-mer phosphorothioate oligonucleotides (S-ODNs) (G3139, its reverse sequence and its two-base mismatch). After ODN addition to vesicles at different mole ratios, changes in pH and electrical surface potential at the lipid-water interface were analyzed by using the fluorophore heptadecyl-7-hydroxycoumarin while particle size distributions were analyzed by static-light scattering. The results indicate that each homo-oligonucleotide does indeed exhibit different complexation behavior. In particular, the maximal level of DOTAP neutralization by the polyadenine S-ODN is much lower than that for the two other homo-oligonucleotides and hence its lipoplex is much more positively charged. Much smaller electrostatic differences are also apparent between lipoplexes formed from each of the G3139-related ODNs. This paper identifies nucleotide base selection and sequence as a variable that can affect the physicochemical properties of oligonucleotide lipoplexes and hence probably their transfection competency.

Recent advances in the use of oligodeoxynucleotide and plasmid-derived RNA as antisense agents of special relevance to cancer gene therapy are summarized with emphasis on agents and systems which have lead to clinical trials and/or regression of established tumors in animal model systems. Transformed cell lines bearing plasmids and viruses designed for the transcription of antisense RNA have the advantage that they can be characterized thoroughly and the effects of antisense RNA on target gene expression and phenotype can be studied easily in vivo. Promising results make the considerable efforts of applying oligodeoxynucleotides in whole animals and in clinical trials more plausible. Conversely, oligodeoxynucleotide experiments which yield promising results in tissue culture can be generalized to the in vivo setting by development of clones of cells bearing plasmid-derived antisense RNA against the same target. Several examples of the concordant results for oligodeoxynucleotide and plasmid-derived antisense RNA against the same target are considered. The importance of examination of antisense effects in syngeneic and immunocompetent hosts is illustrated by studies of insulin-like growth factor and insulin-like growth factor receptor where tumor regression and protection against tumor formation have been observed for particular cell types in defined settings.

Oligonucleotides bearing a terminal lipophilic group attached through a biodegradable ester bond should be useful as antisense pro-drugs with improved cellular uptake. The synthesis of 5'-ester oligonucleotides is, however, problematic due to lability of the ester bond during aqueous ammonia treatment that is commonly used for the deprotection of synthetic oligonucleotides. The synthesis of 5'-palmitoyl oligodeoxynucleotides was accomplished in good yield by the use of a combination of base-labile tert-butylphenoxyacetyl amino protecting groups (t-BPA), the oxalyl-CPG anchor group, and ethanolamine (EA) as a deprotecting reagent.

Phosphorothioate analogs of oligodeoxynucleotides at a concentration of 2 microM protected Himalayan tahr cells from infection by caprine arthritis encephalitis virus (CAEV) and equine dermis cells from infection by equine infectious anemia virus (EIAV). The characteristics of this inhibition against these lentiviruses are similar to those previously described for the inhibition of HIV-1 in ATH8 cells [17]. Thus, the 28-mer homo-oligomer of cytidine [S-(dC)28] was at least as effective as three anti-sense sequences targeted to the LTR, gag, and env regions of CAEV. The effectiveness of homo-oligomers of equal length was in the order C >> A > T, and a random 28-copolymer with a composition of 2C:1G was as effective as S-(dC)28. Shorter oligonucleotides were less effective (28 > 14 > 5 mers) for all base compositions tested. While replication of a simian type D retrovirus was inhibited by S-(dC)28, this compound did not inhibit the cytopathogenicity of two type C retroviruses, amphotropic murine leukemia virus (MuLV), and baboon endogenous virus, when they were tested in the same cell lines used to support the replication of lentiviruses. Southern blot analysis of the high molecular weight DNA of drug-treated CAEV-infected cells showed that S-(dC)28 was acting at or before the reverse transcription step. Our present data and the earlier finding that S-(dC)28 is a potent in vitro inhibitor of the MuLV reverse transcriptase [15] suggest that S-(dC)28 is acting very early in the replication cycle of these lentiviruses. Since MuLV reverse transcriptase is inhibited in vitro, but its replication is not blocked in permissive cells, our data suggest that the phosphorothioate oligonucleotides are preventing virus attachment.

An all phosphorodithioate oligodeoxyribonucleotide (PS2; 17-mer) complementary to the coding region of the rabbit beta-globin mRNA was compared with the normal (PO2) and phosphorothioate (POS) oligonucleotide of the same size and sequence with respect to physicochemical properties and antisense activity in cell-free systems. The melting temperature (Tm) of the PS2-cDNA duplex was reduced by 17 degrees C relative to the PO2-cDNA duplex, compared to 11 degrees C for the POS-cDNA duplex, suggesting a decreased stability of the duplex with an increasing sulfur substitution. Like the POS-derivative, the PS2 oligonucleotide is quite stable against exonucleases, but these modified oligonucleotides showed different stability towards endonucleases and also towards different sub-cellular fractions of MCF-7 cells. During in vitro protein binding studies, the PS2 oligonucleotide showed similar binding (10-20%) to that of the PO2 oligonucleotide, while the POS oligonucleotide bound 60%. In cell-free translation, the PS2 oligonucleotide produced slightly higher specific translation inhibition of rabbit beta-globin mRNA compared to that of the PO2 oligonucleotide, and this was true only at concentration below 2 mM. The POS-derivative, except at 10 mM concentration, always showed higher translation arrest of the rabbit beta-globin mRNA compared to that of the other two oligonucleotides. The present study suggests that the PS2 oligonucleotide offers very little advantage over the POS oligonucleotide for use as an antisense analog.

To assess the usefulness of computer-assisted modeling of mRNA as an aid in design of antisense DNA, the efficiency of inhibition of translation of rabbit beta-globin mRNA by various antisense sequences was compared with calculated structures of the mRNA. The model obtained by consideration of 30 lowest-energy computer-simulated structures is consistent with the high accessibility of the AUG initiation codon region known from digestion with nucleases and with previous antisense inhibition studies reported in the literature. Additional antisense inhibition data were obtained with 20-mer phosphorothioate oligonucleotides, targeted to regions of beta-globin mRNA differing moderately in their degree of participation in intramolecular folding. The efficiency of translation arrest by the oligonucleotides in cell-free expression systems (wheat germ extract and rabbit reticulocyte lysate) was obtained by measuring incorporation of [35S]methionine into total protein, and corrected for sequence-nonspecific inhibition using brome mosaic virus mRNA. In the presence of RNase H (wheat germ system), the inhibitory activity of the oligonucleotides showed correlation with the calculated secondary structure of mRNA, in particular at low oligonucleotide-to-mRNA ratios (correlation coefficient, 0.95). No correlation was observed in the reticulocyte lysate system, in which the inhibition is mediated by translational arrest.

Efforts have been made to reduce the disadvantages associated with the natural oligonucleotides (all-PO) for antisense application by introducing phosphorothioate (PS) linkages into the molecule. A series of such oligodeoxynucleotide copolymers (17-mers) complementary to the coding region of the rabbit beta-globin mRNA, and containing different proportions and arrangements of PO and PS bonds, were synthesized and tested for their protein-binding properties, nuclease stability in vitro, hybridizing ability with the complementary DNA (cDNA), ability to form RNase H-sensitive substrates and antisense activity in cell-free systems. The melting temperatures (Tm) of the co-polymers were reduced by up to 6 degrees C relative to the all-PO oligo, compared to 11 degrees C for the all-PS compound, indicating intermediate hybridizing abilities of the co-polymers. The protein-binding studies with human serum albumin exhibited a linear correlation with the percentage of PS linkage present in the molecule. Nuclease susceptibilities of the co-polymers were also improved, but the number and position of the PS linkages played a significant role in such improvement. Translation inhibition by these oligonucleotides was only found in wheat germ agglutinin (WGA) extract, but not in rabbit reticulocyte lysate (RRL) cell-free system, suggesting the involvement of RNase H in their antisense activities. Provided they have > or = 50% PS linkages, the co-polymers produced almost the same increased inhibition in the WGA system as that of the all-PS oligo. The translation arrest in WGA extract is in good agreement with the in vitro cleavage found for rabbit globin mRNA in the oligo:mRNA duplex by RNase H alone. It is concluded that a copolymer of PO and PS might be preferable to either all-PO or all-PS for antisense applications.

As oligomers inhibited hepatoma growth in vitro. These cells have oligomer-specific binding sites. The oligomers were taken up by liver and have potential for hepatic growth modulation in intact animals.

Oligodeoxynucleotides can act as antisense complements to target sense sequences of natural mRNAs to selectively regulate gene expression by translation arrest. This is a form of interventional gene therapy. Chemically modified analogs that are nuclease-resistant enable this strategy to be utilized in practice. Of the chemically modified backbone analogs of oligodeoxynucleotides we have used the phosphorothioate (PS) analog, in which a non-bridging phosphate oxygen atom is substituted with a sulfur atom. We have shown that these oligodeoxynucleotide analogs inhibit beta-globin expression in cell free systems, and that they are taken up by cells. Specific sequences have been shown to selectively regulate viral and cellular gene expression, for example the bcl-2 oncogene that is found in ca. 90% of lymphomas. However, the PS analog has certain disadvantages, notably reduced hybridization and non-selective inhibition of translation. We have therefore synthesized a series of (PS-PO) co-polymers and characterized their properties. Other related approaches include catalytic ribozymes, and formation of triplexes by direct interaction of oligomers in the major groove of DNA. In general, a chemically modified oligodeoxynucleotide analog can be regarded as a novel form of informational drug.

A detailed comparison was made of the concentration dependence of translation inhibition by phosphorothioate and phosphodiester oligodeoxynucleotides of the same anti-beta-globin sequence in cell-free systems using beta-globin mRNA and unrelated mRNAs as controls. The results confirm that at low concentrations the phosphorothioate oligomer is more potent as an antisense compound, while at higher concentration (greater than 4 microM) it exhibits more nonspecific inhibition than the phosphodiester oligomer for RNase H-mediated translation inhibition.

Sequences from the gag, pol and rev regions of the RF strain of HIV-1 (HIV-1RF) were chosen as targets for antisense phosphorothioate oligodeoxynucleotides (S-oligos). These sequences were the p18/p24 junction in gag, the active site of HIV protease in pol; a sequence from the first exon of the rev gene and S-oligodeoxycytidylic acid controls. Compounds were tested against HIV-1 in both acutely and chronically infected cells. The results show that these phosphorothioate analogues tested in acutely infected cells were active in the 0.1-2 microM range, were dependent on chain length but had no sequence specificity. To study the mechanism of action, the time of addition of S-oligos to acutely infected cells was delayed for up to 48 h post-infection. It was found that antiviral activity was lost when compounds were added to the cultures later than 10 h post-infection. With chronically infected cells only the antisense rev sequence showed activity at 30 microM and neither of the gag or pol antisense sequences has a significant effect on HIV replication at 50 microM. These results are consistent with previous in vitro studies which demonstrate that antisense S-oligodeoxynucleotides have several modes of action.

Drugs that contain information encoded in them are intended to interfere with great specificity in the transfer of information within the cell, thus providing selective therapeutic advantage. It is proposed that such drugs can be classified as informational drugs.

Of the chemically modified backbone analogs of oligodeoxynucleotides that have been developed for antisense applications, the phosphorothioate (PS) analog has perhaps the best properties. Nevertheless, it also has certain disadvantages, notably reduced hybridization and increased non-selective inhibition of translation, compared to the natural phosphodiester (PO) compounds. We have therefore synthesized and characterized a series of co-polymers, with the same antisense beta-globin sequence, but with different repeated sequences of PO and PS and tested them for their comparative properties. The results indicate that a PO-PS co-polymer is the best backbone modification for an antisense compound.

The title compounds were chemically synthesized as their 5'-dimethoxytrityl derivatives by base-catalyzed reaction of 35S-enriched elemental sulfur with support-bound hydrogen phosphonate oligomer. This was derived from adamantane carbonyl chloride-activated coupling of nucleotide hydrogen phosphonate monomers, and similarly activated capping with isopropyl phosphite. A convenient, disposable, reversed-phase cartridge was utilized to purify and isolate the 5'-dimethoxytrityl derivative for subsequent in situ detritylation and elution of the final product. The specific activity obtained for the title compounds was ca. 10(7) cpm/mumols-eq P(O)S-. The procedure should be readily adaptable to appropriate syntheses of other P-S containing analogs of DNA and RNA.