Gene Therapy and Molecular Biology Vol 4, page 177
Gene Ther Mol Biol Vol 4, 177-182. December 1999.
ets-1 mRNA as target for antisense radio- oligonucleotide therapy in melanoma cells Research Article
Kalevi J. A. Kairemo1,3 , Ketil Thorstensen1, Merete Mack1, Mikko Tenhunen2, and Antti P. Jekunen4
1Dept. of Clinical Chemistry, Norwegian University of Science and Technology, Trondheim, Norway.
2Dept. of Oncology, Helsinki University Central Hospital, Finland
3Dept. of Clinical Chemistry, Helsinki University Central Hospital, Finland
4 Dept. of Clinical Pharmacology, Helsinki University Central Hospital, Finland
Correspondence: Kalevi J A Kairemo, MD, PhD, MSc (Eng), Department of Clinical Chemistry, Helsinki University Central
Hospital, FIN-00029, Finland. Tel. +358-9-47172565; Fax. +358-9-47176678; E-mail:Kalevi.Kairemo@huch.fi
Key Words: Antisense, phosphorothioate, oligonucleotide therapy, radiation, phosphorothioates, angiogenesis, endothelial cell,
Abbreviations: AS-PODNs, antisense phosphorothioate oligodeoxynucleotides; VEGF, vascular endothelial growth factor
Received: 9 August 1999; accepted: 29 August 1999
Angiogenesis provides a novel target for anticancer therapy, in particular radiochemo-therapy as endothelial cells in the vascular wall are sensitive to radiation. Antisense phosphorothioate oligodeoxynucleotides (AS-PODNs) may serve as vehicles for carrying cytotoxic or radioactive agents into a particular intracellular location. Radiolabelled AS-PODNs have the potential of having both antisense and radiation effects. Recently, vascular endothelial growth factor (VEGF)-induced invasiveness was shown to be specifically inhibited by AS-PODN directed against ets-1. Previous studies have shown that radio-oligonucleotide therapy may be effective with AS-PODNs internally labelled with 32P, 33P or 35S. Theoretically, 35S gave the smallest variation in nuclear dose in the different cell dimensions studied (Kairemo et al., Cancer Gene Ther 1998; 5: 408-12). This means that cell nuclear targets should be treated with the short range -emitters 35S or 33P for optimal radio-oligonucleotide therapy. Here we explore this possibility using 33P labeled 17-mer AS-PODNs directed against ets-1 in human melanoma cells in vitro. Inhibition of cell growth was observed in the following order: labeled AS-PODN > nonlabeled AS-PODN > labeled sense PODN > nonlabeled sense PODN > transfection agent. Even with a single 33P at the 5Â´-end of the AS- PODN melanoma cell uptake of label was approximately 0.5 mBq/cell. The nuclear doses in this experiment varied from 1.9 to 3.7 cGy. Thus, in vitro and in vivo use of radio-oligonucleotide therapy utilizing 33P radionanotargeting, e.g. in angiogenesis through ets-1, are highly recommended.
The oncogene v-ets was originally discovered as a component of a chimeric genome, along with a truncated v-myb gene, present in the genome of E26, an avian leukosis virus (LePrince et al. 1983; Nunn et al. 1983). Since then, a family of transcription factors, known as the ets family, involved in a wide variety of biological processes including growth control and development, transformation, and T-cell activation, have been cloned and sequenced from a variety of species ranging from
human to Drosophila. The common feature of ets proteins is a well-conserved 85 amino acid domain that binds specifically to DNA containing a (G/C)(A/C)GGAAGT consensus sequence (Macleod et al. 1992; Wasylyk et al.
1993; Timms and Kola 1994).
The ets gene family includes, in addition to ets-1 and ets-2, also erg, elk-1, elk-2, pu-1, fli-1 and E74 (Fisher et al. 1992). ets-1 encodes a set of phosphoproteins ranging in size from 39 to 51 kD (Fisher et al. 1992). However, whereas the chicken ets protein, which contains both the ets-1 and ets-2 domains,
Kairemo et al: Antisense ets-1 therapy for melanoma
distributes equally between the cytoplasm and nucleus, in the human and other mammals, the ets-1 protein is cytoplasmic and the ets-2 protein nuclear. This, together with their noncoordinate expression, suggests that ets-1 and ets-2 have different biologic functions (Fujiwara et al.
Ets-1 is preferentially expressed at high levels in B and T cells and is regulated during both thymocyte development and T cell activation (Chen 1985; Bhat et al.
1989). Studies in mice have shown that Ets-1 is essential for normal maintainance, survival and activation of B- and T-lineage cells (Bories et al. 1995; Muthusamy et al.
1995). Bhat et al. (1990) found that, following T-cell activation, ets-2 mRNA and proteins are induced, while ets-1 gene expression decreases to very low levels.
Amplification and rearrangement of ets-1 has also been implicated in human leukemia (Goyns et al. 1987; Rovigatti et al. 1986). ets-1 is also expressed in endothelial cells during blood vessel development and in fibroblasts adjacent to tumor cells in various invasive human carcinomas (Wernert et al. 1992; Wernert et al.
1994), suggesting that the ets-1 gene could be involved in angiogenesis associated with tumor growth and normal development.
then escape degradation by intracellular nucleases to achieve adequate concentrations in the correct intracellular compartment. As cellular nucleases effectively degrade phosphodiester ODNs, several more nuclease resistant ODNs have been developed. Of these the phosphorothioate ODNs (P-ODNs) in which a non-bridging oxygen atom has been replaced by a sulphur atom, are the most common.
Antisense ODNs may also function as specific carriers of cytotoxic drugs into cells, provided that the target sequence is specifically expressed in that particular cell type. Similarly, by attaching an appropriate radionuclide to the ODN, the selective delivery of radiation to a particular cell may be achieved. The primary target for ionizing radiation is the nuclear DNA and if the radiation source is in close proximity to the DNA, molecular damage would ensue (Kairemo et al. 1999).
Theoretical studies evaluating Auger and gamma- emitting radionuclides as well as beta-emitters (Kairemo et al. 1998) have suggested that short range beta-emitters such as 33P and 35S may be best suited for delivery of radiation confined to the cell nucleus. ODNs are easily labeled with 33P or 35S at their 5Â´- or 3Â´-end. Furthermore, since antisense P-ODNs contains both sulphur and
Tumor growth, progression and metastasis are
phosphorus atoms that could be exchanged with
dependent on the formation of new capillary blood vessels from existing vessel, a process termed angiogenesis (Folkman and Shing 1992); rapidly growing tumors are often hypoxic due to insufficient vascularization. Angiogenesis is a cascade of processes involving both soluble angiogenic factors and insoluble extracellular matrix factors (Jekunen and Kairemo 1997). Multiple soluble molecules that stimulate angiogenesis are released by tumor cells as well as host cells such as endothelial, epithelial and mesothelial cells and leukocytes. In SK- MEL-2 human melanoma cells cultured under hypoxic conditions the synthesis of Vascular Endothelial Growth Factor (VEGF) is stimulated (Claffey et al. 1996). In endothelial cells VEGF induces the expression of the proto-oncogene ets-1 and stimulates endothelial cell migration. On the other hand, antisense P-ODNs directed against ets-1 mRNA inhibit the ability of endothelial cells to migrate (Chen et al. 1997). In fact, induction of ets-1 expression appears to be a common phenomenon in endothelial cells stimulated by angiogenic growth factors (Iwasaka et al. 1996). Thus, the ets-1 gene apparently plays a direct role in angiogenesis.
Antisense oligodeoxynucleotides (ODNs) are short (typically 15 bases) stretches of synthetic DNA that are complementary to specific regions of cellular mRNA or DNA. This complementarity allows them to hybridize to specific parts of cellular mRNA or DNA, forming mRNA- DNA or DNA-DNA duplexes. The duplex formation disrupts the function of that particular gene either at the translational or transcriptional level. Due to their specificity antisense ODNs have become attractive potential tools for specific therapeutic applications, e.g. as specific inhibitors of malignant cell growth. In order to be effective the antisense ODNs must first enter the cell and
35S as part of their structure, these radionuclides offer benefits over e.g. transition metal nuclides since they do not require any extra coupling techniques for incorporation of the nuclide into the P-ODN.
The aim of this study was to investigate the potential of radiolabeled antisense oligodeoxynucleotides to specifically inhibit the growth and to destroy melanoma cells utilizing ets-1 mRNA as target ( Figure 1).
During the first 24 h of incubation with 100 nM 33P- labelled antisense or sense oligonucleotides the cells accumulated 33P to an activity of approximately 0.15-0.25 mBq/cell. No further cellular accumulation of label was observed during the next 24 h. Incubation of cells with
200 nM 33P-labeled oligonucleotides increased cellular accumulation of label to approximately 0.4-0.5 mBq/cell at 24 h incubation, with no further accumulation of label during the next 24 h (Table 1). Thus, cellular accumulation of 33P-labeled oligonucleotides apparently increases linearly with extracellular oligonucleotide concentration. The cellular accumulation of 33P at 24 h with 200 nM oligonucleotides corresponds to a cellular oligonucleotide uptake of approximately 2.5 pmol DNA per million cells.
The cumulative activity per cell was calculated for the two different oligonucleotide concentrations at 48 h incubation. At 100 nM oligonucleotide the cumulative activity was somewhat higher for the antisense compared to the sense oligonucleotide. At 200 nM oligonucleotide concentration there was no difference in cumulative activity between the antisense and the sense
Gene Therapy and Molecular Biology Vol 4, page 179
oligonucleotide (Table 1). Based on the data shown in Table 1 and assuming a cell diameter of 14 m, radiation doses per cell were calculated. The radiation doses varied between 1.2 and 4.1 cGy.
In cells that were allowed to accumulate 33P -labelled antisense oligonucleotide directed against ets-1 mRNA for
24 h cell growth was inhibited approximately 25% (Table
2). With other oligonucleotides (i.e. labeled sense,
unlabeled antisense or sense) the effect on cell growth was less pronounced, irrespective of whether the oligo was labelled or not. The transfection reagent alone inhibited cell growth by approximately 15%. Incubation of cells with 100 or 200 nM oligonucleotide apparently had no effect on cell growth.
After 48 h of incubation no effects on cell growth were observed.
Figure 1. Effects of Ets-1 protein and antisense approach for cell killing.
Table 1: Accumulation of 33P during incubation of cells with labelled oligonucleotides. Activities, cumulative activities and doses per cell are shown.
Oligo concentration (nM)
Cumulated activity (Bqs/cell)
Nuclear dose (cGy)
Cumulated activity (Bqs/cell)
Nuclear dose (cGy)
Kairemo et al: Antisense ets-1 therapy for melanoma
Table 2. Effect of oligonucleotide treatment on cell growth.
In this first, to our knowledge, experiment designing internal labeling characteristics for ODNs we used single
5Â´-end labeling with 33P to obtain a sufficient specific activity for subcellular dosimetric and cell killing experiments. In order to achieve high specific activity, many 33P-atoms may be incorporated as part of the ODN backbone during synthesis. We have shown here that doubling the ODN concentration in the cell may increase cellular radiation doses more than three-fold (Table 1 ). Theoretically, it is possible to introduce 17 more 33P- atoms in this ets-1 P-ODN system.
Human melanoma G361 cells were utilized to establish the in vitro model system. These cells are well characterized, simple to maintain in culture, and they express ets-1. They are rather resistant to radiation. These cells are well suited for use in xenograft models in mice, because they grow subcutaneously and they may develop new vessels (angiogenesis).
In this preliminary experiment, melanoma cells were incubated with two concentrations of 33P-labeled ODNs for
24 and 48 hours. The accumulation and elimination of ODNs by the cells seemed to be low. Total cellular uptake was less than 1% of added ODNs in the cell culture media. Furthermore, very little events between 24 and 48 hours, this is because influx and efflux are in balance (Table 1). It is important to find the optimal time and concentration for antisense treatment. Actually, in our experiments the lower ODN concentrations gave clearly different doses both for antisense ODN and sense ODN, respectively. The significance of this, if any, is unknown. At higher concentration the doses both for antisense ODN and sense ODN were rather similar. This demonstrates that it is important to find the optimal mode of ODN delivery.
During these experiments we did not yet find the optimal amount of transfection reagent, because very small differences on cell growth in various conditions were observed (Table 2). However it was clear that the best effect was obtained using labeled antisense-ODN. This finding has not been shown earlier in the literature. In fact, this to our knowledge, the first report of cell killing utilizing endocytotic therapy with 33P-radionuclide.
We calculated internal radiation doses as described previously (Kairemo et al. 1998) as D (target target). This means that cell-to-cell-interactions as well as activities in the cell media were neglected (negligible?). This should be the case in optimal radionanotargeting. Following loading of the cells with radiolabeled ODNs a simple subcellular fractionation into nuclear, membrane and cytosolic fractions can be performed. This will yield data on the nuclear, cytoplasmic, and cell surface distribution of label (data not shown). Using this approach internal radiation doses were calculated as previously described (Kairemo et al. 1999). Using cell diameter of 14 m, a nuclear dose of 3 cGy was obtained which is in
accordance with the cell doses shown in Table 1.
100 nM oligo
200 nM oligo
Cell growth (%)
Cell growth (%)
The human melanoma G361 cell line produces adequate levels of Ets-1 protein and mRNA under standard culture conditions. Utilizing known inducers (e.g. VEGF) or inhibitors (e.g. tissue plasminogen activator) of ets-1 expression the sensitivity of the detection systems may be changed. Following validation of the detection systems the ability of antisense ODNs to down-regulate the cellular expression of ets-1 can be studied. Our preliminary results indicate that radiolabeled antisense ODNs have to be evaluated with respect to any effects on ets-1 expression in addition to or synergistically with the pure antisense effect. Further studies are needed to decipher the molecular mechanisms of cell killing by radioactive antisense oligonucleotides .
The authors (KJAK, KT) were supported by a reseach grant from Sintef Unimed Foundation, Trondheim, Norway.
IV. Materials and Methods
A. Cell culture
Human melanoma cells G631 were cultured in RPMI1640 medium containing 2 mM L-glutamine, 50 U penicillin/ml 50 g streptomycin/ml and 10% (v/v) FCS, in an atmosphere of 5% CO2/95% air in 60 mm dishes. The doubling time of the cells was approximately 24h.
B. Phosphorothioate oligonucleotides
Antisense and sense phosphorothioate oligonucleotides against ets-1 were obtained from Amersham Pharmacia Biotech. The oligonucleotide sequences were as described by Chen et al. (Chen et al.
1997) as follows: Antisense: 5â- TCGACGGCCGCCTTCAT-3â; Sense: 5â- ATGAAGGCGGCCGTCGA-3â. The oligonucleotides were purified by FPLC from the manufacturer and were reconstituted in sterile 1xTE buffer.
Gene Therapy and Molecular Biology Vol 4, page 181
C. Labeling of oligonucleotides
The oligonucleotides were 5â-end labeled with -33P- ATP according to the manufacturers instructions, utilizing a kit from Amersham Pharmacia Biotech. Unincorporated
-33P-ATP was removed by passage through an anion exchange column (Qiaquick oligonucleotide removal kit, Qiagen). Incorporation of 33P into the oligonucleotides was in the range of 60-80% and specific activity was approximately 200 Bq/pmol DNA.
D. Incubation of cells
Cells were seeded in 60 mm dishes at a density of
5x105 cells per dish in culture medium (se above). The next day cells were washed twice in DPBS before fresh culture medium containing labelled or unlabelled oligonucleotides (200 nM) in SuperFectâ¢ (Qiagen) were added. The cells were incubated in the presence of oligonucleotides for a maximum of 48 h.
E. Determination of cell growth
At the designated time points the cells were washed in DPBS, trypsinated and the resulting cell suspension counted in a Coulter Z1 (Coulter Electronics, Ltd.). The effect of treatment on cell growth is expressed relative to the control cells receiving no treatment.
F. Cell dosimetry
The nuclear dose of the internalized oligodeoxynucleotide was estimated using the principles of Medical Internal Radiation Dose (MIRD) schema (Loevinger and Berman 1976). The dose D was calculated as a product of cumulated activity Ã and specific absorption fraction, S, where all the radiation sources k are summed up together:
D= ÃkÂ·Sk k
The nuclear dose was calculated using the assumption of uniformly distributed activity and subcellular S-factors of Goddu et al. (Goddu et al. 1993) for the cellular and nuclear diameters of 7 and 4 m. We assumed that subcellular S-factors of phosphorus-33 are similar to those of sulphur-35, which gives the total S- factor (cellânucleus) to be 6.3 x 10-4 Gy/Bqs. The dose
from the adjacent cells or from the medium was not taken into account. The tracer concentration was assumed to increase from 0 to 24 hrs linearly and remain constant from 24 to 48 hours.
Bhat NK, Komschlies KL, Fujiwara S, et al. (1989) Expression of ets genes in mouse thymocyte subsets and T cells. J Immunol 142, 672-678.
Bhat NK, Thompson CB, Lindsten T, et al. (1990) Reciprocal expression of human ETS1 and ETS2 genes during T-cell activation. Proc Natl Acad Sci 87, 3723-3727.
Bories J-C, Willerford D, GrÃ©vin D, et al. (1995) Increased T- cell apoptosis and terminal B-cell differentiation induced by inactivation of the Ets-1 proto-oncogene. Nature 377,
Chen JH (1985) The proto-oncogene c-ets is preferentially expressed in lymphoid cells. Mol Cell Biol 5, 2993-3000. Chen Z-Q, Fischer RJ, Riggs CW, et al. (1997) Inhibition of
vascular endothelial growth factor-induced endothelial
cell migration by ETS1 antisense oligonucleotides.
Cancer Res 57, 2013-2019.
Claffey KP, Brown LF, del Aguila LF, et al. (1996) Expression of vascular permeability factor/ vascular endothelial growth factor by melanoma cells increases tumor growth, angiogenesis, and experimental metastasis. Cancer Res
Fisher RJ, Koizumi S, Kondoh A, et al. (1992) Human ETS1 oncoprotein. J Biol Chem 267, 17957-17965.
Folkman J, and Shing Y (1992) Angiogenesis. J Biol Chem
Fujiwara S, Fisher RJ, Seth A, et al. (1988) Characterization and localization of the products of the human homologs of the v-ets oncogene. Oncogene 2, 99-103.
Goddu SM, Howell RW, and Rao RV (1993) Cellular dosimetry: absorbed fractions for monoenergetic electron and alpha particle sources and S-values for radionuclides uniformly distributed in different cell compartments. J Nucl Med 35: 303-311.
Goyns MH, Hahn IM, Stewart J, et al. (1987) The c-ets-1 proto- oncogene is rearranged in some cases of acute lymphoblastic leukaemia. Br J Cancer 56, 611-613.
Iwasaka C, Tanaka K, Abe M, Sato Y (1996) Ets-1 regulates angiogenesis by inducing the expression of urokinase- type plasminogen activator and matrix metallo- proteinase-1 and migration of vascular endothelial cells. J Cell Physiol 169, 522-531.
Jekunen AP, and Kairemo KJA (1997) Inhibition of malignant angiogenesis. Cancer Treatm Rev 263-286.
Kairemo KJA, Tenhunen M, and Jekunen AP (1998) Gene therapy using antisense oligodeoxynucleotides labeled with Auger-emitting radionuclides. Cancer Gene Therapy 5, 408-412.
Kairemo KJA, Jekunen AP, Tenhunen M (1999) Dosimetry and optimization of in vivo targeting with radiolabeled antisense oligodeoxynucleotides (oligonucleotide radiotherapy). In Antisense Technology (Phillips MI, ed.), Meths Enzymol, (Abelson JN and Simon MI, eds.) Vol. 314, Academic Press, New York, pp506-524.
LePrince D, GÃ©gonne A, Coll J, et al. (1983) A putative second cell-derived oncogene of the avian leukaemia retrovirus E26. Nature 306, 395-397.
Loevinger R, and Berman MA (1976) A revised schema for calculating the absorbed dose from biologically distributed radionuclides. MIRD pamphlet no. 1, revised, p.3, Society of Nuclear Medicine, New York.
Macleod K, LePrince D, and Stehelin D (1992) The Ets gene family. Trends Biochem Sci 17, 252-256.
Muthusamy N, Barton K, and Leiden JM (1995) Defective activation and survival of T cells lacking the Ets-1 transcription factor. Nature 377, 639-642.
Kairemo et al: Antisense ets-1 therapy for melanoma
Nunn MF, Seeburg PM, Moscovici C, and Duesberg PH (1983) Tripartite structure of the avian erythroblastosis virus E26 transforming gene. Nature 306, 391-395.
Rovigatti U, Watson DK, and Yunis JJ (1986) Amplification and rearrangement of Hu-ets-1 in leukemia and lymphoma with involvement of 11q23. Science 232, 398-400.
Timms MJ, and Kola I (1994) Regulation of gene expression by transcription factors Ets-1 and Ets-2. Mol Reprod Dev
Wasylyk B, Hahn SL, and Giovane A. (1993) The ETS family of transcription factors. Eur J Biochem 211, 7-18.
Wernert N, Raes MB, Lasalle P, et al. (1992) c-ets1 proto- oncogene is a transcription factor expressed in endothelial cells during tumor vascularization and other forms of angiogenesis in humans. Am J Pathol 140, 119-
Wernert N, Gilles F, Fafeur V, et al. (1994) Stromal expression of c-Ets1transcription factor correlates with tumor invasion. Cancer Res 54, 5683-5688.