The brominated trihalomethanes (THMs) are mutagenic and carcinogenic disinfection by-products frequently found in chlorinated drinking water. They can be activated to mutagens by the product of the glutathione S-transferase-Theta (GSTT1++-1) gene in Salmonella RSJ100, which has been transfected with this gene. To evaluate this phenomenon in humans, we have examined the genotoxicity of a brominated THM, bromoform (BF), using the Comet assay in human whole blood cultures exposed in vitro. No differences were found in the comet tail length between cultures from GSTT1-1(+) versus GSTT1-1(-) individuals (1.67 +/- 0.40 and 0.74 +/- 0.54 microm/mM, respectively, P = 0.28). The high variability was due to the relatively weak induction of comets by BF. Combining the data from both genotypic groups, the genotoxic potency of BF was 1.20 +/- 0.34 microm/mM (P = 0.003). GSTT1-1 is expressed in red blood cells but not in the target cells (lymphocytes), and expression within the target cell (as in Salmonella RSJ100) may be necessary for enhanced mutagenesis in GSTT1-1(+) relative to GSTT1-1(-) cultures. To examine this, we exposed Salmonella RSJ100 and a control strain not expressing the gene (TPT100) to the most mutagenic brominated THM detected in Salmonella, dibromochloromethane (DBCM), either in the presence or absence of S9 or red blood cells from GSTT1-1(+) or GSTT1-1(-) individuals. S9 did not activate DBCM in the non-expressing strain TPT100, and it did not affect the ability of the expressing strain RSJ100 to activate DBCM. As with S9, red cells from either genotypic group were unable to activate DBCM in TPT100. However, red cells (whole or lysed) from both genotypic groups completely repressed the ability of the expressing strain RSJ100 to activate DBCM to a mutagen. Such results suggest a model in which exposure to brominated THMs may pose an excess genotoxic risk in GSTT1-1(+) individuals to those organs and tissues that both express this gene and come into direct contact with the brominated THM, such as the colon. In contrast, those organs to which brominated THMs would be transported via the blood might be protected by erythrocytes. Such a proposal is reasonably consistent with the organ specificity of drinking water-associated cancer in humans, which shows slightly elevated risks for cancer of the colon and bladder but not of the liver.