Object. The presence of hypoxic cells in human brain tumors contributes to the resistance of these tumors to radiation therapy. However, because normal tissues are not hypoxic, the presence of hypoxic cells also provides the potential for designing cancer-specific gene therapy. Suicide genes can be expressed specifically in hypoxic conditions by hypoxia-responsive elements (HREs), which are activated through the transcriptional complex hypoxia-inducible factor-1 (HIF-1).
Methods. The authors have transfected the murine BAX-green fluorescent protein (GFP) fusion gene under the regulation of three copies of HRE into U-87 MG and U-251 MG cells and selected stably transfected clones. Even though BAX was expressed under both oxic and anoxic conditions in these clones, cell survival assays demonstrated increased cell killing under anoxic as compared with oxic conditions. Cells obtained from most of these clones did not grow in vivo, or the tumors exhibited highly variable growth rates. However, cells obtained from the U-251 MG clone A produced tumors that grew as well as tumors derived from parental cells, and examination of the tumor sections under fluorescent microscopy revealed GFP expression in localized regions. Western blot analyses confirmed an increased BAX expression in these tumors. Analysis of the results suggests that HRE-regulated BAX can be a promising tool to target hypoxic brain tumor cells. However, there are measurable levels of BAX-GFP expression in this three-copy HRE-mediated expression system under oxia, suggesting promoter leakage. In addition, most clones did not show significant induction of BAX-GFP under anoxia. Therefore, the parameters of this HRE-mediated expression system, including HRE copy number and the basal promoter, need to be optimized to produce preferential and predictable gene expression in hypoxic cells.
Therapy for malignant gliomas currently relies on a multimodality approach involving surgery, radiation therapy, and chemotherapy. However, the median survival time of patients with the most malignant glioma, GBM, is increased by only a few months when current treatments are applied. After surgery, radiation therapy is the most effective treatment for patients with malignant gliomas, and the efficacy of the treatment increases with increasing radiation dose. The dose-limiting tissue is the normal brain that surrounds the tumor, and the pivotal issue in the radiation therapy of gliomas is clearly a matter of therapeutic ratio -- doses of radiation that are tolerable to the patient are insufficient to control the tumor. Clearly, there is a need to devise strategies that markedly improve the therapeutic ratio, and this will require that we exploit differences between normal and malignant tissues.
One important abnormal characteristic found in most solid tumors, including human brain tumors, is the presence of regions with reduced oxygen concentrations. From a radiobiological perspective, these hypoxic cells are the most important subpopulation because they are much more resistant than oxic cells to radiation-induced damage. Historically, the radiobiologically important hypoxic cells have been thought to be those that arise in tumors when tumor growth produces cells that are further than 150µm from a blood vessel. Such chronic hypoxia is produced by an insufficient blood supply that occurs partly because tumor cells grow faster than the endothelial cells that make up the blood vessels and partly because the newly formed vascular supply is disorganized. Acute hypoxia can also develop due to intermittent blood flow, a phenomenon that is known to occur in a variety of disease states including tumors. Hypoxia affects the proliferation of cancer cells and their responsiveness to radiotherapy, because anoxic cells (PO2 ≤0.5 mm Hg) require approximately three times the radiation dose needed to kill oxic cells (PO2 ≥20 mm Hg). Cells at intermediate oxygen levels (0.5 mm Hg < PO2 < 20 mm Hg) display intermediate sensitivities to radiation and are also important contributors to the overall efficacy of radiation therapy.
The investigators of several studies have shown that the regulation of gene expression by oxygen is an important feature of many biological processes, and hypoxia is a powerful modulator of gene expression. Therefore, it should be possible to target hypoxic cells by developing a gene therapy strategy that uses plasmids containing suicide genes that are selectively expressed under hypoxic conditions. One example of hypoxia-regulated gene expression is provided by Epo, a hormone that regulates erythropoiesis in accordance with the oxygen-carrying capacity of the blood. The Epo gene contains an HRE consensus sequence in the 3' enhancer region. Hypoxia-inducible factor-1 is a transcription factor that accumulates in hypoxic cells, binds to HREs, and "turns on" the promoter regulating Epo. Previously, we transfected plasmids containing three copies of HRE derived from the human Epo gene into human brain tumor cells and tested their ability to induce LacZ gene expression under oxic and anoxic conditions. Gene expression under anoxic conditions increased approximately 12-fold for U-87 MG cells and approximately fourfold for U-251 MG cells, indicating HRE is functional in both brain tumor cell lines.
In the present study, we used HREs to regulate expression of the BAX suicide gene in response to hypoxia. A member of the Bcl-2 family of proteins, BAX can initiate apoptotic cell death in cell culture and in animals. We demonstrated previously that cells transiently transfected with BAX under HRE regulation are preferentially killed through apoptosis in anoxic conditions. In this present study, we transfected a vector containing a BAX-GFP fusion gene under the regulation of three HREs into two human GBM cell lines, and we isolated clones that were stably transfected with this plasmid. The GFP marker protein allows us to determine quickly the level and location of BAX expression in cells and in animal xenografts. The BAX-GFP fusion protein has been successfully used by other investigators, who found that the BAX-GFP fusion protein can produce cell killing as effectively as the BAX protein. Stably transfected clones that were made from these cell lines exhibit increased cell-killing ability under anoxic conditions. We implanted cells from BAX-containing clones into the flanks of athymic mice, and one of these clones produced tumors that grew as well as parental wild-type tumors. We detected gene expression in tissues removed from these tumors by observing GFP via fluorescence microscopy and by measuring BAX expression using Western blot analysis. The results of our studies indicate that BAX can be stably transfected into GBM cells. Such cells will be important for conducting future in vitro and in vivo bystander effect studies.