The ribbon antisense to hTR was constructed using the method described earlier (21). Oligonucleotides containing target sequences were synthesized using an Expedite 8909 DNA synthesizer (Applied Biosystem, Foster city, CA). The antisense and control (scrambled and mismatched) oligodeoxynucleotides (ODN) were phosphorylated at the 5' end. Both antisense to hTR and control ODN were used to form stemloop structures. The intra-molecular stem was formed with complementary sequences at both the 5' and 3' ends of each ODN. Two identical ODN molecules were joined by single-stranded sequences that were complementary to each other at the 5' ends. ODN molecules were mixed and heated for 2 min at 95℃, followed by gradual cooling to room temperature. One unit of T4 DNA ligase was added and incubated for 24 h at 16℃ to generate a ribbon type molecule ligated covalently with diad-symmetry. The ribbon type ODN consists of two loops connected by one stem. The ODNs were purified with a Poros HQ anion exchange column (PerSeptive Biosystems, Framingham, MA). Anion exchange was performed by gradient elution with 1M NaCl solutions. Purified ODNs were dried in an evaporator and desalted by reverse phase chromatography. Finally, the desalted ODNs were precipitated by ethanol and resuspended in ddH₂O.

The cellular uptake of hTR-RiAS into cancer cells was enhanced by employing a tripartite DPL (DNA: peptide: liposomes) transfection system, which had previously been found to be effective in our laboratory. Antisense oligos were complexed with a short modified peptide of the protein transduction domain (PTD) derived from the Tat protein, and the complex was then added to cationic liposomes, Lipofectamine (Invitrogen, Carlsbad, CA), to form the DPL complex. Each component of the tripartite complex, DNA/peptide/liposomes, was mixed in the ratio of 1:3:5 (w:w:w) and added to the target cells to improve their cellular uptake of the antisense molecules.

The telomerase activity was determined using a Telo TAGGG Telomerase PCR ELISA^PLUS Kit (Roche Diagnostics, Germany), as recommended by the manufacturer. HeLa cells were seeded at 1×10⁶ cells/well in 6-well plates. The cells were transfected with hTR-RiAS and incubated for 48 h. To compare the effects of the antisense molecules on the telomerase activity, equal amounts of mismatched control or liposome alone were also added to the cells, and simultaneously assayed. The cells were then trypsinized and lysed in 500µl of ice-cold lysis buffer for 30 min. The lysates were centrifuged at 13,000 rpm and 4℃ for 30 min, the supernatants rapidly frozen and stored at -70℃. One µl of each extract, corresponding to 1×10³ cells, were assayed in a total reaction mixture in 50µl to detect the telomerase activity. Telomeric repeats were added to a biotin-labeled primer during the first telomerase-mediated extension reaction. The elongation products were amplified by PTC-100 thermal cycler (MJ Research). In this step, a TS8(+ ve) template set and heat-treated HeLa cells were used as positive and negative controls for detection of telomerase expression, respectively, and simultaneously amplified. An aliquot of the PCR product was denatured, hybridized to a DIG-labeled, telomeric repeat-specific probe and bound to a streptavidin-coated 96-well plate. Finally, the immobilized PCR product was detected with an anti-DIG-POD antibody, visualized by a color reaction, using the substrate TMB, and quantified photometrically at 450 nm.

The MTT assay was performed to study the effect of hTR-RiAS molecules on the proliferation of cancer cells. HeLa cells were plated at 5×10³ cells/well in a 96-well plate one day prior to transfection and incubated in an atmosphere of 5% CO₂ at 37℃. The next day, the cells were washed twice with TOM-Transfection Optimized Medium (JBI, Deagu, Korea) and transfected with hTR-RiAS in a 50 µl volume for 6 h. The cells were then added with 100 µl complete medium, containing 20% FBS, and incubated for a further 4 days. The media were replaced with 50 µl of fresh medium, with the addition of 20 µl MTT reagent (5 mg/ml) prepared in phosphate buffered saline, followed by incubation in a CO₂ incubator at 37℃ for 4 h. 150 µl of dimethyl sulfoxide was added to the cells and incubated for 1 h at room temperature with gentle mixing. The absorbance was measured at 570 nm with a SpectraMAX 190 (Molecular Devices, Sunnyvale, CA, USA) to quantify the surviving cells. The percentage of cell growth inhibition in each well treated with the antisense was calculated by comparing the optical density with that of a sham-treated control, using the following formula: 1 - (absorbance of an experimental well/absorbance of a control well)×100.

HeLa cells were seeded at 1×10⁶ cells in a 6 cm-dish one day prior to hTR-RiAS treatment. To compare the effects of the antisense molecules on apoptosis, control ODNs (scrambled or mismatched) or an equal amount of liposomes alone were also added to the cells and simultaneously assayed. Cells were lysed with cell lysis buffer (1% NP-40, 25 mM EDTA, 50 mM Tris-HCl, pH 6.6) 24 h post-transfection. The cell lysates were treated with RNase A (final conc. 20 µg/ml) and incubated for 3 h at 58℃. Proteinase K was then added, to a final concentration of 20 µg/ml, and incubated for 3 h at 37℃. The cell lysates were mixed with 2.5 vol. of ice-cold ethanol and 0.1 vol. of 3M ammonium acetate and stored overnight at -20℃. The precipitated DNA was resuspended in 25 µlof TE buffer (pH 8.0) and run on a 1.8% agarose gel.

It is considered critical that antisense oligonucleotides (AS-oligos) exhibit stability toward nucleases, while maintaining sequence specific antisense activity. Ribbon antisense (RiAS) molecules, with excellent stability and antisense activity, without side effects, which have been reported with chemically modified counterparts (21).

The selection of target sites may be performed by the exhaustive simulation of secondary structures of target RNA. The entire sequence of hTR RNA was examined in order to find target sites readily accessible to antisense molecules. Five target sites were chosen; named hTR13701 (50 bp), hTR13702 (52 bp), hTR13703 (52 bp), hTR13704 (43 bp) and hTR13705 (43 bp). Mismatched and scrambled control oligos were also designed and designated hTR13706 and hTR13707, respectively (50 bp). The hTR13706 oligo had 8 mismatched bases in the sequence of hTR13701 (Table 1). Antisense molecules to the five potential target sites were ligated to form RiAS molecules, having a stem and two identical antisense loops at both ends of the stem (Fig. 1).

When examined for hTR expression, all six different cancer cell lines tested were found to be positive for hTR expression (Fig. 2A). These were two lung cancer cell lines (A549 and NCIH1299), a liver cancer cell line (Hep3B), a colon cancer cell line (SW480), a cervical cancer cell line (HeLa) and a skin cancer cell line (A375SM). The five different RiAS molecules were examined for antisense activity in HeLa cells. Twenty-four hours post-transfection, the total RNA was purified and subjected to RT-PCR. Among the 5 molecules, AS13701 was able to eliminate hTR mRNA to completion (Fig. 2B). The ribbon antisense was termed hTR-RiAS, and adopted for further studies.

To achieve successful antisense activity, it is critical to have efficient uptake of antisense oligos. The tripartite DPL (DNA/peptide/liposomes) complex found effective for the cellular uptake of both plasmid DNA and AS oligos (manuscript in preparation), was employed. The DPL complex was formed with hTR-RiAS, for enhanced transfection, and tested in various cancer cell lines in an effort to reduce the cellular level of hTR RNA. When hTR-RiAS was used for antisense activity in HeLa cells, the antisense was found to be especially potent, almost completely ablating cellular hTR RNA (Fig. 3A). hTR-RiAS was able to reduce hTR RNA by about 60% at a concentration of 0.1 µg and by 80% at 0.5 µg. Almost complete ablation of hTR RNA was observed with 1.0 µg of hTR-RiAS. The RT-PCR product obtained was then examined for sequence authenticity by Southern blotting. The DNA band was probed with a 27 mer internal primer (5'-CTCGCTGACTTTCAGCGGGC GGAAAAG-3') labeled with FITC. The data confirmed that the amplified DNA band was derived from hTR RNA (data not shown). A similar level of antisense activity was also observed in another two cancer cell lines (SW840 and NCIH1299) tested, indicating a broad utility of hTR-RiAS to various cancer cells (Fig. 3B and 3C). In contrast, both scrambled and mismatched oligos exhibited only mild reductions of hTR RNA. Treatment of liposomes alone also showed no reduction in the level of hTR RNA. Transfection of hTR-RiAS did not affect theβ-actin expression. The effect on the telomerase activity in HeLa cells treated with the hTR-RiAS was also examined. The cells treated with hTR-RiAS showed more than a 60% reduction in their telomerase activity compared with the cells treated with liposomes alone, mismatched oligos, or with the sham treated cells (Fig. 4).

These results confirm that the hTR-RiAS has specific and potent antisense activity to hTR RNA when efficiently delivered as a DPL complex.

The potency of the antisense activity of hTR-RiAS was then tested for inhibition of HeLa cell growth. The cells treated with increasing amounts (0.05 µg, 0.1 µg and 0.2 µg) of hTR-RiAS showed growth inhibition of about 35, 50 and 70%, respectively, in proportion to the increasing amounts of the antisense transfection (Fig. 5). In contrast, control treatments showed no significant growth inhibition.

It has been reported that treatment of antisense to telomerase complexes, including hTR RNA, can cause apoptosis (5). The cell death caused by hTR-RiAS treatment was investigated to see if it reflected the induction of apoptosis. For this, HeLa cells treated with the antisense molecules were subjected to a DNA fragmentation assay. The hTR-RiAS was found to cause characteristic DNA ladder formation 24 h after transfection, indicating apoptotic progression (Fig. 6). No DNA ladder formation was detected in the control treated cells when using scrambled oligos, mismatched oligos or liposomes alone.