Improving the intracellular delivery and molecular efficacy of antisense oligonucleotides in chronic myeloid leukemia cells: a comparison of streptolysin-O permeabilization, electroporation, and lipophilic conjugation.

The hybrid gene BCR-ABL that typifies chronic myeloid leukemia (CML) represents an attractive target for therapy with antisense oligodeoxyribonucleotides (ODN). A central obstacle in the therapeutic application of ODN is their poor cellular uptake. Adding various lipophilic conjugates to the ODN backbone has been reported to improve uptake, and electroporation of target cells has also been shown to enhance intracellular ODN delivery. We have shown that (1) BCR-ABL-directed ODN will specifically decrease the level of BCR-ABL mRNA, provided that cells are first permeabilized with Streptolysin-O (SL-O), and (2) chimeric methylphosphonodiester:phosphodiester ODN directed against 9 bases either side of the BCR-ABL junction are more efficient ODN effectors than structures composed solely of phosphodiester or phosphorothioate linkages. In this study, we compared the efficacy of lipophilic conjugation, SL-O permeabilization and electroporation on the intracellular delivery and molecular effect of BCR-ABL-directed ODN. b2a2- and b3a2-directed chimeric ODN were synthesized either unmodified or with one of the following groups at the 5' end: cholesterol, vitamin E, polyethylene glycol of average molecular weight 2,000 or 5,000, N-octyl-oligo-oxyethylene, or dodecanol. ODN associated with Lipofectin was also studied. Comparison was made in untreated, electroporated, and SL-O permeabilized KYO1 cells. Uptake was examined by fluorescence microscopy and flow cytometry, using ODN structures that were 3' labeled with fluorescein. The effect on target BCR-ABL mRNA expression was analyzed by Northern blotting. Several conjugated structures associated avidly with the cell membrane without achieving significant intracellular uptake or molecular effect. Similarly, ODN:Lipofectin complexes moderately increased cell association, without enhancing intracellular levels of ODN or inducing detectable molecular effect. In SL-O permeabilized or electroporated cells, uptake was approximately 1 to 2 logs greater than in untreated cells, and rapid nuclear localization was seen, especially with unmodified chimeric ODN. In SL-O permeabilized cells treated with ODN directed to the b2a2 and b3a2 junctions respectively, b2a2 BCR-ABL mRNA levels at 4 hours were reduced to 2. 6% +/- 2.1% and 38.4% +/- 1.3% of control values. In cells permeabilized by electroporation, BCR-ABL mRNA levels were decreased to 4.0% +/- 1.4% of control levels by b2a2 directed ODN, although very little nontargeted suppression was seen with b3a2-targeted ODN (93.4% +/- 4.2% of control). Greater cell to cell variation in ODN uptake was seen for SL-O permeabilized cells when compared with electroporated cells, suggesting that, after SL-O permeabilization, relatively unpermeabilized and overpermeabilized populations may coexist. No structure had any effect on the level of irrelevant (p53, MYC, and GADPH) mRNA levels. We conclude that the conjugation of chimeric ODN with one of the above-mentioned lipophilic groups or the complexing of ODN with Liopfectin does not improve either intracellular delivery of ODN or the molecular effect. In contrast, both electroporation and SL-O permeabilization (1) considerably enhanced uptake of chimeric ODN (even for structures without a conjugate group) and (2) achieved significant suppression of target mRNA levels.