Low doses of gamma ionizing radiation increase hprt mutant frequencies of TK 6 cells without triggering the mutator phenotype pathway

The TK6 lymphoblastoid cell line is known to be mismatch repair (MMR) and p53 proficient. Deficiency in MMR results in a mutator phenotype characterized by microsatellite instability (MSI) and increased hprt mutant frequency (MF). Increased hprt MF is also a biomarker of effect for exposure to ionizing radiation. In order to test if a mutator phenotype could be induced by low doses of gamma ionizing radiation, an hprt cloning assay and a MSI investigation were performed after radiation exposure. The spontaneous MF was 1.6 x 10. The groups exposed to 0.2, 0.5 and 1.0 Gy had hprt MFs of 2.3, 3.3 and 2.2 x 10, respectively. The spontaneous MSI frequency per allele in non-selected cells was 5.4 x 10, as evidenced at the loci D11S35, nm23-H1, D8S135 and p53. MSI frequencies in the groups exposed to 0.2, 0.5 and 1.0 Gy were found to be < 4.7, < 7.7 and < 12 x 10, respectively. The frequencies of hprt mutants and MSI found in this study suggest that low doses of ionizing radiation increase hprt mutant frequency without triggering the mutator phenotype pathway.

In this study, we used the hprt cloning assay and microsatellite analysis to investigate if low doses of gamma ionizing radiation can induce a persistent genetic instability through the mutator phenotype pathway in TK6 hprt mutant clones.hprt cloning assay.Three groups of TK6 cells (1.6 x 10 6 cells/mL) were exposed in vitro to the doses of 0.2, 0.5, and 1.0 Gy of gamma ionizing radiation, respectively.Cells were irradiated at a dose-rate of 9.7 ± 0.74% Gy/min, using a 137 Cs g-ray source (Gammacell 1000, Nordion International Inc., Ontario, Canada), and then incubated for three days at 37 °C and 5% CO 2 before the hprt cloning assay.For hprt cloning assay, 6 cells/well in the unexposed group and 3 cells/well in the exposed groups were plated into 96-well microtitre plates in non 6-TG-containing medium for determining plating efficiency, and at 10 4 cells/well in 2.0 mg/mL 6-TG for hprt mutant selection.RPMI 1640 medium was supplemented with 6 mg/mL penicillin, 10 mg/mL streptomycin, 1% fungizone and 10% bovine serum.After 14 days, clones were scored and expanded individually to 6-well plates in 2 mL of growth medium/well.Wells with approximately 10 6 cells/mL were harvested for microsatellite analysis.

Statistical analysis.
Chi-square analysis was performed to conclude if there were a difference between two plating efficiencies or between two mutant frequencies.
The spontaneous hprt mutant frequency found in this study was 1.57 x 10 -6 , whereas the frequencies for the groups exposed to 0.2, 0.5 and 1.0 Gy were 2.31, 3.28, and 2.18 x 10 -6 , respectively.The plating efficiencies and mutant frequencies for all groups are shown in Figure 1.Our spontaneous hprt mutant frequency was similar to that described by other authors, who reported spontaneous mutation frequencies ranging from 1.3 to 3.8 x 10 -6 (Giver et al., 1993;Nelson et al., 1994;Sussman et al., 1999;Tomita-Mitchel et al, 2000).The hprt mutant frequencies for the groups exposed to 0.2 and 0.5 Gy were closer to the mutant frequencies of TK6 cells exposed to 0.3 and 0.5 Gy described elsewhere (Grosovsky and Little, 1985).The previously reported hprt mutant fraction for lymphoblastoid cells exposed in vitro to 1.0 Gy is 17.1 ± 7.8 x 10 -6 (Phillips et al., 1995).Such a high value may be due to the selection of mutants using p53 non-proficient TK6 cells (WTK1 cell line) and further to the lower concentration of 6-TG (0.5 mg/mL) medium.For higher doses, Nelson and colleagues (1994) found a mutant fraction of 10.1 ± 0.4 x 10 -6 for TK6 cells exposed in vitro to 2.0 Gy.
The unexposed group and the one not selected by 6-TG showed six events of MSI in 1,112 alleles investigated (5.4 x 10 -3 MSI/allele), as described in Table 1.All the microsatellite instabilities observed were found to present a decrease of the wild-type allele size.Loci D11S35 (not shown) and nm23-H1 (Figure 2.A) exhibited two MSI per locus in different clones.Loci D8S135 (Figure 2.B) and p53 (Figure 2.C) exhibited one MSI per locus.However, because no MSI was detected in 6-TG R clones, MSI frequencies were estimated to be lower than 8.6, 4.7, 7.7 and 12 x 10 -3 for the unexposed cells and for the cells exposed to 0.2, 0.5 and 1.0 Gy, respectively.It is worth noting that MMR deficiency is likely to be identified by a .4x 10 -3 < 8.6 x 10 -3 < 4.7 x 10 -3 < 7.7 x 10 -3 < 12 x 10 -3 1 Microsatellite analysis was randomly performed among the positive clones, where each clone was considered as having two alleles.110.96, 296.33, and 31.96 for the mutant frequencies of the groups exposed to 0.2, 0.5, and 1.0 Gy, respectively.p = 0.001.
microsatellite instability frequency higher than 30%, using panel of microsatellite markers (Ayres et al., 2004).Moreover, hprt mutant selection is a rare event measured per ~10 -6 assayed cells, in which it is assumed that only one mutant clone gives raise to each positive colony (Albertini et al., 2000).Thus, although the exact frequencies of microsatellite instability and mutator phenotype in hprt mutants induced by ionizing radiation remain to be determined, these results indicate the presence of an efficient MMR system in our mutant clones.
The spontaneous MSI frequency of 5.4 x 10 -3 found in this study is closer to the data reported for peripheral T lymphocytes, which vary from 2.9 to 5 x 10 -3 (Shibata et al., 1994;Hackman et al., 1995).However, our spontaneous MSI frequency was higher than 8 x 10 -4 , as reported by Li and colleagues (1994) for unexposed TK6 cells.Additionally, Davies and colleague (1999) found a spontaneous MSI frequency of 1 x 10 -4 in hprt non-selected clones and of 3 x 10 -3 in 6-TG R T-lymphocytes, results that are in disagreement with the data from this study.The last mentioned authors attributed the low rate of MSI for non-selected clones to the large number of alleles investigated.Giver and Grosovksy (2000) reported one event of MSI induced by ionizing radiation at 59 tk mutants (6 x 10 -4 MSI/allele).
In conclusion, considering together the hprt mutant frequency for all the groups and the absence of MSI in the 6-TG R clones, our data suggest that ionizing radiation increases hprt mutant frequency without triggering the mutator phenotype pathway.

Figure 1 -
Figure 1 -Dose response of TK6 cells to ionizing radiation for (A) plating efficiencies and (B) mutant frequencies, using Poisson distribution, as described by Albertini et al. (2000).Error bars represent the standard deviation of each mutant frequency.Positive clones per well plated are shown in brackets.c 2 = 22.92 and 48.94 for the plating efficiencies of the groups exposed to 0.5 and 1.0 Gy, respectively.c 2 =110.96,296.33, and 31.96  for the mutant frequencies of the groups exposed to 0.2, 0.5, and 1.0 Gy, respectively.p = 0.001.

Table 1 -
Frequency of MSI at 6 microsatellite loci in TK6 cell clones exposed and unexposed to ionizing radiation.