Induction of endogenous γ-globin gene expression with decoy oligonucleotide targeting Oct-1 transcription factor consensus sequence

The K562 Human erythroleukemia cells were purchased from the American Type Culture Collection (ATCC, Rockville, MD) and maintained in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum.

All the oligonucleotides used in this study were synthesized either by the Keck Oligo Synthesis Laboratory at the Yale University School of Medicine or by Integrated DNA Technologies, Inc (Coralville, IA) and are listed in Table 1. A pair of phosphorothioate-modified complementary 24 mer oligonucleotides containing the Oct-1 consensus sequence was synthesized and annealed to form a 24 bp double-stranded DNA fragment, which was designated as the Oct-1 decoy oligonucleotide. A pair of phosphorothioate-modified complementary 24 mer oligonucleotides containing a scrambled sequence was also synthesized and annealed to form a 24 bp double-stranded DNA fragment, which was designated as the scrambled oligonucleotide.

A pair of complementary 24-mer oligonucleotides containing the Oct-1 consensus with a sequence of ^5'ACGTCAGTATGCAAATCGTATCAG^3' was synthesized. The complementary oligonucleotides were annealed to form a 24 bp Oct-1 consensus DNA fragment. The DNA fragment was radioactively labeled with [γ-³²P]-ATP by T4 polynucleotide kinase. The radioactively labeled Oct-1 consensus DNA fragment (1 × 10^-11 mole or ~20,000 cpm radioactively labeled Oct-1 DNA fragment) was incubated with a HeLa nuclear extract (5 μg) in a volume of 20 μl containing 1× Protein Binding buffer (20 mM HEPES, pH 7.9, 100 mM KCl, 0.2 mM EDTA, 0.5 mM DTT, and 20% glycerol) at room temperature for one hour. To reduce unspecific binding, denatured salmon sperm DNA (4 μ g) was also supplemented in each binding reaction. The unlabeled 24 bp scrambled or Oct-1 decoy oligonucleotide was added into some reactions for competitive binding of the Oct-1 transcription factor. The reactants were analyzed by polyacrylamide gel electrophoresis using a 6% gel and visualized by autoradiography.

The K562 cells were seeded at a density of 4 × 10⁴ cells/ml and treated with the Oct-1 decoy oligonucleotide at various concentrations by adding the 24 bp Oct-1 decoy oligonucleotide directly into the cell growth medium. As controls, some K562 cells were treated with either 10 μ M scrambled oligonucleotide or 75 μ M hemin, an effective γ-globin gene inducer, in parallel experiments. At various time points, the cells were harvested and total RNA was isolated using an RNeasy mini kit (Qiagen, Santa Clarita, CA). Total RNA was also isolated from the untreated K562 cells at the same time points in parallel experiments.

A two-step real time PCR assay was performed to measure the level of the γ-globin mRNA. The reverse transcription (RT) reaction was carried out in 2 μg of total RNA using the Taqman Reverse Transcription Master Mix (Applied Biosystems, Foster City, CA). A primer optimization step was then tested to determine the optimal primer concentrations. Once the optimal primer concentrations were determined, the cDNA sample (10 ng) was used for a quantitative PCR reaction to determine the value of cycle threshold (C _t ) of the γ-globin gene from each RNA sample using a Sybr Green Master Mix with ABI 7500 Fast Real Time PCR System (Applied Biosystems). The Ct value of the GADPH gene was also determined for each RNA sample. The real time PCR data was then analyzed to determine the levels of the γ-globin mRNA in each RNA sample. The level of the γ-globin mRNA in the untreated K562 cells of each time point was counted as 100% and the level of the γ-globin mRNA in the treated K562 cells of the same time point was then calculated as a percentage in comparison to that of the untreated K562 cells. The GAPDH gene was used as an internal control for normalization. Relative expression of the γ-globin mRNA was expressed as 2^-ΔΔCtwhere ΔC _t was calculated by subtracting the average normalization gene C _t (GAPDH) from the average target gene (γ-globin gene) C _t value in the same cell line and the ΔΔC _t was obtained by subtracting the ΔC _t of the untreated cells from the ΔC _t of the treated cells.

The K562 cells were seeded at a density of 6 × 10⁴ cells/ml. The cells were treated with either the 24 bp Oct-1 decoy oligonucleotide (2 μ M and 10 μ M) or the 24 bp scrambled oligonucleotide (10 μ M) for four days and then harvested. Approximately 6 × 10⁶ cells were harvested for each study in the experiment. As a control, the same number of cells was also harvested from the untreated K562 cells. The cells were fixed in 1% formaldehyde for 20 minutes at room temperature and washed in 1 × PBS three times. The cells were resuspended in the SDS cell lysis buffer at a density of 1 × 10⁶ cells/200 μ l and incubated on ice for 10 minutes. The cells were then sonicated (four cycles of 10 second sonications with a 30 second pulse) to lyse the cells. The cell lysates were centrifuged at 4°C for 10 minutes at 14,000 rpm to remove any insoluble cell debris and the supernatants were used in the IP study. The cell lysates (100 μ l) were diluted with 900 μ l of ChIP dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl, pH8.1, 167 mM NaCl) and then incubated with 20 μ l of Oct-1 antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at 4°C overnight. Protein A-conjugated agarose beads (40 μ l beads) (Sigma Inc., St. Louise, MO) were then added and the reactants were incubated at 4°C for two hours. The beads were collected from the reactants by centrifugation and washed three times with low salt washing buffer (0.1%SDS, 1.1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH8.1, 150 mM NaCl) and three times with high salt washing buffer (0.1% SDS, 1% Triton X-100, 0.5 mM EDTA, 20 mM Tris-HCl, pH8.1, 500 mM NaCl) to remove any unspecific binding proteins. Half of the beads were analyzed by western blot to determine the level of the Oct-1 protein precipitated with the beads in each reaction. The rest of the beads were resuspended in 100 μ l of 50 mM NaCl and incubated at 65°C for 5 hours to reverse the DNA-protein crosslinks. The proteins were removed by phenol chloroform extraction and the DNA was recovered by ethanol precipitation. The DNA was dissolved in 10 μ l of TE buffer and a quantitative PCR assay was performed to determine the amount of the γ-globin gene promoter region DNA contained in each reaction using a pair of primers that bind to the γ-globin gene at the positions of -350 to -330 and +50 to +30 respectively with the ABI 7500 Fast Real Time PCR System. The Ct value was determined for each reaction. The amount of γ-globin gene promoter DNA precipitated from the untreated K562 cells by the IP was calculated as 100% and the amount of γ-globin gene promoter DNA precipitated from the treated K562 cells by the IP was calculated as a percentage in comparison to that of the untreated K562 cells.

We first tested whether the decoy oligonucleotide designed for the Oct-1 transcription factor can effectively compete with the Oct-1 consensus probe in vitro. For this study, a 24 bp DNA fragment containing the Oct-1 consensus sequence was radioactively labeled with [γ-³²P]-ATP and used as a probe for the in vitro DNA-protein binding study. The DNA probe was incubated with HeLa nuclear extracts at room temperature for one hour to allow for the Oct-1 transcription factor to bind to the DNA probe. For a competitive binding, the HeLa nuclear extracts were incubated with both the radioactively labeled Oct-1 DNA probe and the unlabeled Oct-1 decoy oligonucleotide at room temperature for one hour. The reactants were then analyzed by polyacrylamide gel electrophoresis using a 6% gel (Fig. 1). The binding of the Oct-1 transcription factor caused a mobility shift of the Oct-1 DNA probe in the gel (Fig. 1, lane 2 vs lane 1). When the HeLa nuclear extracts were incubated with both the radioactively labeled Oct-1 DNA probe and the unlabeled Oct-1 decoy oligonucleotide, however, a dose-dependent reduction in the level of the shifted Oct-1 DNA probe was observed (Fig. 1, lanes 9–11). The Oct-1 transcription factor binding-caused gel mobility shift of the Oct-1 DNA probe was completely diminished when 1 × 10^-6 M of the Oct-1 decoy oligonucleotide was supplemented in the binding reaction (Fig. 1, lane 11 vs lane 2). As a control, no clear reduction in the level of the shifted Oct-1 DNA probe was observed when similar concentrations of the scrambled oligonucleotide were supplemented into the binding reactions (Fig. 1, lanes 6–8). This result suggests that the Oct-1 decoy oligonucleotide effectively competes with the Oct-1 consensus DNA probe for the binding of the Oct-1 transcription factor.

To determine if the Oct-1 decoy oligonucleotide treatment could result in reduced cell viability, we studied the cell viability of the K562 human erythroleukemia cells under the Oct-1 decoy oligonucleotide treatment. The K562 cells were seeded at a density of 4 × 10³ cells/ml and treated with the Oct-1 decoy oligonucleotide at various concentrations (0, 2, 5, and 10 μ M). As controls, some K562 cells were either untreated or treated with the scrambled oligonucleotide (10 μ M). The cell density was determined at various time points (1, 2, 3, 4, and 5 days) and the cell growth curve was determined for each treatment (Fig. 2). The cell growth was not significantly affected by the Oct-1 decoy oligonucleotide treatment even with the oligonucleotide at concentrations as high as 10 μ M (Fig. 2).

The results obtained from our in vitro protein binding assays revealed a strong competitive binding of the Oct-1 decoy oligonucleotide for the Oct-1 transcription factor. To determine if the Oct-1 decoy oligonucleotide can induce expression of endogenous γ-globin genes, the K562 cells were treated with the Oct-1 decoy oligonucleotide and the effect of the decoy oligonucleotide treatment on γ-globin expression was determined.

We first determined the Oct-1 decoy oligonucleotide treatment-induced transcription of the γ-globin gene. The K562 cells were either untreated or treated with the Oct-1 decoy oligonucleotide at various concentrations (0, 2, 5, and 10 μ M). As controls, some K562 cells were treated with the scrambled oligonucleotide (10 μ M) or hemin (75 μ M), an effective hemoglobin inducer [30‐32]. At various time points, the cells were harvested and total RNA was isolated. A real time PCR assay was then performed to determine the level of the γ-globin mRNA in each RNA sample (Fig. 3A). When treated with the scrambled oligonucleotide at 10 μ M, no significant increase in the level of γ-globin mRNA was observed in the K562 cells (Fig. 3A). When treated with the Oct-1 decoy oligonucleotide, however, significant increases in the level of the γ-globin mRNA were detected in the K562 cells even with the Oct-1 decoy oligonucleotide at concentrations as low as 2 μ M. For example, when the K562 cells were treated with 5 μ MOct-1 decoy oligonucleotide for two, three, four and five days, the level of the γ-globin mRNA increased 1.92, 2.93, 2.85, and 5.13 fold respectively. As a positive control, when the K562 cells were treated with 75 μ M hemin for two, three, four, and five days, the levels of the γ-globin mRNA increased 2.41, 4.47, 4.00, and 3.71 fold respectively (Fig. 3A). This result suggests that the Oct-1 decoy oligonucleotide treatment increases the transcription of the γ-globin genes in the K562 cells as effectively as that of hemin, a well-known γ-globin inducer.

Although the results obtained from our real time PCR study demonstrated increases in transcription of the γ-globin genes in the Oct-1 decoy oligonucleotide-treated K562 cells, whether these increases in transcription led to increases in the protein level of fetal hemoglobin (HbF, α2γ2) was unknown. Therefore, we further performed an immuno-blotting study to determine the level of HbF in the Oct-1 decoy oligonucleotide-treated K562 cells (Fig. 3B). Indeed, the results obtained from our immuno-blotting studies revealed that the treatment of the Oct-1 decoy oligonucleotide (2, 5, or 10 μ M) resulted in increases in the level of HbF in the K562 cells (Fig. 3B). As a negative control, the treatment of the K562 cells with 10 μ M scrambled oligonucleotide did not cause a significant increase in the level of HbF (Fig. 3B). As a positive control, the treatment with 75 μ M hemin also resulted in an increase in the level of HbF in the K562 cells (Fig. 3B).

To determine whether the increased γ-globin gene transcription in the Oct-1 decoy oligonucleotide-treated K562 cells was caused by the competitive binding of the Oct-1 decoy oligonucleotide to the Oct-1 transcription factor, we further performed an immunoprecipitation (IP) study [33, 34] to determine the binding status of the Oct-1 binding sites within the promoter region of the endogenous γ-globin gene under the Oct-1 decoy oligonucleotide treatment. Both untreated and the Oct-1 decoy oligonucleotide-treated K562 cells were fixed in 1% formaldehyde and sonicated to shear the chromosomal DNA into small fragments. The Oct-1 transcription factor was then immunoprecipitated from both the untreated and the decoy oligonucleotide-treated K562 cells by the IP protocol using an Oct-1 antibody and Protein A-conjugated agarose beads. As controls, the Oct-1 transcription factor was also immunoprecipitated from the scrambled oligonucleotide and hemin-treated K562 cells. Half of the beads were analyzed by western blot to determine the level of Oct-1 protein precipitated from each cell lysate (Fig. 4A) and rest of the beads were analyzed by a quantitative PCR assay to determine the level of the γ-globin gene promoter region DNA co-precipitated with the Oct-1 transcription factor from each cell lysate (Fig. 4B). The results of our western blots revealed that similar amounts of Oct-1 transcription factor were precipitated from individual cell lysates, suggesting a very successful IP of the study. The results of our quantitative PCR studies indicated that the Oct-1 decoy oligonucleotide treatment caused significant decreases in the level of the endogenous γ-globin gene promoter region DNA co-precipitated with the Oct-1 transcription factor by the IP protocol. In comparison to the amount of γ-globin gene promoter region DNA co-precipitated with the Oct-1 transcription factor in the untreated K562 cells, only 44% and 17% of the γ-globin gene promoter region DNA was co-precipitated with the Oct-1 transcription factor when the K562 cells were treated with 2 μ M and 10 μ M Oct-1 decoy oligonucleotide, respectively. When the K562 cells were treated with 10 μM scrambled oligonucleotide, however, more than 83% of the γ-globin gene promoter region DNA was still co-precipitated with the Oct-1 transcription factor. The treatment of K562 cells with 75 μ M hemin also resulted in a decrease in the level of the co-precipitated γ-globin gene promoter region DNA to 61% of the untreated K562 cells. This result revealed that the treatment of K562 cells with the Oct-1 decoy oligonucleotide caused a dose-dependent decrease in the binding of Oct-1 transcription factor to the endogenous γ-globin gene promoter region DNA sequence, suggesting that the Oct-1 decoy oligonucleotide effectively competed with the endogenous γ-globin gene promoter region Oct-1 consensus sequences for binding with the Oct-1 transcription factor, which resulted in activation of the endogenous γ-globin gene and increases in the level of γ-globin mRNA and HbF in the K562 cells.