Office of Biological and Environmental Research

DOE Low Dose Radiation Program Workshop IV

Abstract

Title: Neoplastic Transformation in vitro Following Exposure to Low Doses of Low LET Radiation

Authors: J.L. Redpath, X-Y Lao, M. Pant, S. Ko, C. Christie, T. Taylor and E. Elmore

Aims

1. To establish dose-response curves including doses <10 cGy.
2. To investigate mechanisms of suppression of transformation at low doses.

Methods

The Hela x human skin fibroblast assay for neoplastic transformation in vitro was used for all studies. The experimental protocol involved both immediate post-irradiation plating and 24-hour delayed plating of cells. Data are presented as foci per surviving cell calculated using the null hypothesis.

Results

The main goal of the initial grant (DE-FG07-99ER62876) was to establish a dose-response curve, including doses <10 cGy of Cs-137 gamma radiation. This was accomplished and the data indicated evidence for suppression of transformation frequency below that seen spontaneously at doses of 10 cGy or less, particularly following delayed plating. This data was published in Radiation Research 156:700-707, 2001.

In an attempt to see how this suppression may vary with radiation quality we performed similar studies with 60 kVp x-rays. In this study, only the delayed plating protocol was used. Suppression of transformation frequencies to levels below that seen spontaneously was again seen. Indeed, there was evidence that the degree of suppression may be greater than that seen with Cs-137 gamma radiation, however, parallel prospective studies would be required to confirm or deny this suggestion. This data was published in the International Journal Radiation Biology (79:235-240, 2003).

The main goal of the second grant (DE-FG03-02ER63309) was to investigate possible mechanisms underlying the observed suppression of transformation. Several avenues have been explored in this regard. The first was based on the hypothesis that there exists a transformation-prone subpopulation of cells that also is hypersensitive to radiation-induced cell-killing at low doses. Thus, an effect of low doses would be to eliminate these cells from the population with a resulting reduction in transformation frequency. Earlier studies in our laboratory had shown G2/M cells to be 5- fold more prone to spontaneous transformation than G1/S-phase cells. Based on this evidence, we collaborated with Peter Johnston and Susan Short at the Gray Cancer Institute to examine the survival response of synchronized HeLa x skin fibroblast human hybrid cells at low radiation doses. The data indicated that G2/M cells were also the only cells that were killed at a dose of 5 cGy. Calculations were able to show that this killing, coupled with the sensitivity of this population to spontaneous transformation, was able to account for the suppression of transformation, at least at a dose of 5 cGy. These data were published in Radiation Research (159:433-436, 2003).

Since low dose-induced suppression of transformation had been seen in noncycling cells by other investigators (Azzam et al., Radiation Research 146:369-373, 1996), it was clear that hyper-radiosensistivity of a transformation prone subpopulation was unlikely to be the sole mechanism responsible for the observed suppression of transformation. We have therefore examined two other possibilities, and are in the process of examining a third. We have tested whether induction of reduced glutathione, a known radioprotective substance, could account for the observed suppression of transformation. We did not see any evidence for GSH-induction at low doses, nor did BSO-induced suppression of GSH modulate the transformation frequencies seen at low doses of 60 kVp x-rays. We conclude from this that modulation of GSH levels by low doses does not account for the suppression seen in our experimental system. The next possibility we tested for was upregulation of DNA repair by low doses. This was done by examining the effect of the PARP inhibitor, 3-aminobenzamide, on transformation frequencies at low doses. No significant effect was seen in controls or at a dose of 0.5 cGy. However, at a dose of 5 cGy there was a significant elevation of the transformation frequency in the presence of 3-AB. We conclude that the role of repair in modulating response is more important at 5, than at 0.5, cGy. This is, at least qualitatively, in agreement with recent conclusions of Rothkamm and Lobrich (Proceeding of the National Academy of the United States 100:5057-5062, 2003) based on the kinetics of removal of γ-H2AX nuclear foci from irradiated human cells. (Pant et al., Carcinogenesis 100:5057-5062, 2003). Since we have seen suppression of transformation at doses as low as 0.1 cGy, where not all cells will experience an ionizing event, we have to consider a possible role for bystander effects. We are currently testing for this using two approaches. The first is to ascertain any role for a gap-junction intercellular communication (GJIC) mediated bystander effect by examining the effect of inhibition of GJIC by lindane on transformation frequencies at low doses. Preliminary results indicate no effect. The second is to look at bystander effects mediated through factors secreted into the extracellular media. This is being done by medium transfer experiments, and using the endpoint of cell survival we see no killing effect of transferring medium from irradiated cells to unirradiated cells.

Conclusions

We have strong evidence for a suppression of transformation frequency at low doses of low-LET radiation. Mechanisms involved may well be dose-dependent over this low dose range. Evidence to date strongly suggests a role for HRS at the lowest doses and induced repair (IRR) at the higher doses. Future plans include extending these studies to high energy protons and heavy ions (DOE-FG02- 03ER63648).

Relevance to Risk of Radiation-Induced Cancer

A priori extrapolating from in-vitro transformation experiments to radiation risk in exposed humans would appear to be unacceptable. Nonetheless, relative risk estimates from our in-vitro data show remarkable agreement with those from epidemiologic studies in the dose range above 20 cGy (see Radiation Research 156:700-707, 2001). This is particularly true for leukemia and breast cancer, two cancers where there is some suggestion of relative risks <1, at doses < 5 cGy.

» Return to top