A comparison of the radiation sensitivities of non-tumorigenic and tumorigenic human hybrid cell lines

The CGL1 (HeLa × Normal Skin Fibroblast) Human Hybrid Cell Line: A History of Ionizing Radiation Induced Effects on Neoplastic Transformation and Novel Future Directions in SNOLAB

Cellular transformation assays have been utilized for many years as powerful in vitro methods for examining neoplastic transformation potential/frequency and mechanisms of carcinogenesis for both chemical and radiological carcinogens. These mouse and human cell based assays are labor intensive but do provide quantitative information on the numbers of neoplastically transformed foci produced after carcinogenic exposure and potential molecular mechanisms involved. Several mouse and human cell systems have been generated to undertake these studies, and they vary in experimental length and endpoint assessment. The CGL1 human cell hybrid neoplastic model is a non-tumorigenic pre-neoplastic cell that was derived from the fusion of HeLa cervical cancer cells and a normal human skin fibroblast. It has been utilized for the several decades to study the carcinogenic/neoplastic transformation potential of a variety of ionizing radiation doses, dose rates and radiation types, including UV, X ray, gamma ray, neutrons, protons and alpha particles. It is unique in that the CGL1 assay has a relatively short assay time of 18-21 days, and rather than relying on morphological endpoints to detect neoplastic transformation utilizes a simple staining method that detects the tumorigenic marker alkaline phosphatase on the neoplastically transformed cells cell surface. In addition to being of human origin, the CGL1 assay is able to detect and quantify the carcinogenic potential of very low doses of ionizing radiation (in the mGy range), and utilizes a neoplastic endpoint (re-expression of alkaline phosphatase) that can be detected on both viable and paraformaldehyde fixed cells. In this article, we review the history of the CGL1 neoplastic transformation model system from its initial development through the wide variety of studies examining the effects of all types of ionizing radiation on neoplastic transformation. In addition, we discuss the potential of the CGL1 model system to investigate the effects of near zero background radiation levels available within the radiation biology lab we have established in SNOLAB.

Neoplastic transformation induced by carbon ions

PURPOSE

The objective of this experiment was to compare the oncogenic potential of carbon ion beams and conventional photon beams for use in radiotherapy.

METHODS AND MATERIALS

The HeLa X human skin fibroblast cell line CGL1 was irradiated with carbon ions of three different energies (270, 100, and 11.4 MeV/u). Inactivation and transformation data were compared with those for 15 MeV photons.

RESULTS

Inactivation and transformation frequencies for the 270 MeV/u carbon ions were similar to those for 15-MeV photons. The maximal relative biologic effectiveness (RBE(alpha)) values for 100MeV/u and 11.4 MeV/u carbon ions, respectively, were as follows: inactivation, 1.6 +/- 0.2 and 6.7 +/- 0.7; and transformation per surviving cell, 2.5 +/- 0.6 and 12 +/- 3. The curve for dose-transformation per cell at risk exhibited a maximum that was shifted toward lower doses at lower energies.

CONCLUSIONS

Transformation induction per cell at risk for carbon ions in the entrance channel was comparable to that for photons, whereas for the lower energies, 100 MeV/u and 11 MeV/u, which are representative of the energies delivered to the tumor margins and volume, respectively, the probability of transformation in a single cell was greater than it was for photons. In addition, at isoeffective doses with respect to cell killing, the 11.4-MeV/u beam was more oncogenic than were photons.

Lack of differential G2 chromosomal radiosensitivity between non-tumorigenic and tumorigenic human hybrid cells (HeLa x skin fibroblasts)

Previous studies have shown, that for log phase cultures, tumorigenic segregants of HeLa x skin fibroblast human hybrid cells are slightly more radiosensitive in terms of cell killing than their nontumorigenic parents (Redpath et al. 1985, Colman et al. 1988). Other studies have shown that these same tumorigenic segregants exhibit a markedly enhanced G2 chromosomal radiosensitivity (Sanford et al. 1986) thus offering a possible explanation for the cell killing data. The present study set out to examine the G2-phase radiosensitivity of these cells in terms of cell killing with the expectation that an enhanced sensitivity in the tumorigenic cells would be seen. No enhanced sensitivity was observed. The G2 chromosomal radiosensitivity was then examined and no differential was seen between the non-tumorigenic and tumorigenic cells. This lack of confirmation of previously reported studies may be due to some technical differences in the experimental protocols.

Studies of the UVC-sensitivity of non-tumorigenic and tumorigenic human cell hybrids (HeLa x skin fibroblasts)

The UVC-sensitivities of a non-tumorigenic and a tumorigenic human cell hybrid (HeLa x skin fibroblasts) are compared and contrasted. The tumorigenic cells differ from the non-tumorigenic cells in that they have lost one copy each of chromosomes 11 and 14. For exponentially growing cultures, the tumorigenic cells are considerably more resistant than the non-tumorigenic cells. For confluent cultures, the differential in photosensitivity is much less. Flow cytometric studies of cell cycle distributions of both exponentially growing and confluent cultures of these cells indicate that the differences in photosensitivity cannot be explained by differences in cell cycle distribution. Studies of the kinetics of potentially lethal damage repair (PLDR) in confluent cultures of both cell lines indicate little or no recovery over the first 6 h followed by an increase in survival over the next 12-24 h. These data are consistent with previously published observations in human skin fibroblasts where the kinetics of PLDR reflected the kinetics of thymine dimer loss. The data are not consistent with 6-4 photoproducts being a potentially lethal lesion since such damage is rapidly repaired in human cells.