Radiation risk estimation in human populations: lessons from the radiological accident in Brazil

The development of radiological and nuclear technologies and the deployment of nuclear weapons have made ionizing radiation one of the most studied human mutagens. Exposure to ionizing radiation produces DNA damage which can result in mutation and cancer, making the risk associated with human exposure a critical issue. In this paper we estimate the risk associated with radiation exposure for individuals exposed to 137Cs during the 1987 Goiânia radiological accident. Using combined regression slopes from both the in vivo hprt mutant frequency and micronucleus frequency data we estimated a doubling dose of 173 (+/-47) cGy for these two endpoints. This is in close agreement with the published estimates for low dose rate and chronic exposure to low-LET radiation. We obtained risk estimates of about 24-fold increase in dominant disorders in the post-exposure generation of the directly exposed population. No detectable increase was found in the population at large. The risk of carcinogenesis in the directly exposed population was found to be increased by a factor in the range of 1.4 to 1.5. The small sample size in this study requires a large element of caution with respect to risk estimates interpretation. Moreover, the doubling dose estimates prepared here are derived from lymphocytes. This somatic data may require additional considerations for both cancer and certainly germ-line events. Nevertheless, the risk of carcinogenesis and genetic harm for this population are good indicators of the potential genetic damage imposed by ionizing radiation to the Goiânia population.

Increased hprt mutant frequencies in Brazilian children accidentally exposed to ionizing radiation

We have examined the effects of ionizing radiation on somatic mutations in vivo, using the hprt clonal assay. The study was performed on blood samples obtained from children exposed during a radiological accident that happened in 1987, in Goiânia, Brazil. The group of children exposed to ionizing radiation includes six males and four females ranging in age from 6 to 14 years at the time of exposure. The radiation doses ranged from 15 to 70 cGy. A Brazilian control group, not exposed to ionizing radiation, was also analyzed under similar conditions. the mean hprt mutant frequency for the exposed group was 4.6 times higher than the control group, although the cloning efficiency from the exposed group was significantly reduced. Linear regression analysis of the mutant frequency and ionizing radiation dose did not show a significant relationship between these two parameters. However, a reliable inverse relationship was demonstrated when the regression analysis was performed with nonselective cloning efficiency and ionizing radiation dose. It was demonstrated that nonselective cloning efficiency diminishes as ionizing radiation dose increases. To correct mutant frequencies for clonal events, the clonal relationship between the hprt mutant clones was examined by T-cell receptor analysis. The majority of the mutants analyzed represented individual clones, thus validating the observed mutant frequencies.

Determination of hprt mutant and mutation frequencies and the molecular characterization of human derived in vivo T-lymphocyte mutants

Using a T-lymphocyte clonal assay, 73 6-thioguanine resistant T-lymphocytes were isolated from two blood samples obtained 4 months apart from a 50-year-old male subject. Sixty-six of these mutants were characterized at the DNA sequence level using cDNA. One particular single base substitution was recovered a total of 23 times. The majority of T-cell receptors (TCR) of these mutants all share a common gamma-TCR rearrangement, and thus likely represent a single mutational event that underwent clonal expansion in vivo. Siblings of this clone were recovered in both collections. Three other single base substitutions were also recovered more than once. In two of the three cases, the mutants were also found to be clonally related, while in one case they were not. A number of identical exon loss events were also recovered, yet none of these were clonally related. This probably reflects the multiple pathways by which these mutations can arise. The TCR data was used to correct the observed mutant frequency to produce an estimate of the actual mutation frequency. The two mutant frequencies, 18 x 10(-6) and 19 x 10(-6), obtained from the first and second sampling periods, respectively, can thus be corrected to yield true mutation frequency's of 12 x 10(-6) each.