Preferential breakage of chromosomes at specific sites (so-called "fragile sites") has been observed to occur spontaneously, and has been induced by some
metal salts and chemicals. Furthermore, a heterochromatic region of the long arm of the Chinese hamster ovary (CHO) X-chromosome is known to be susceptible to a disproportionately high frequency of spontaneous breakage; unless there is physical displacement of
chromatin the resulting achromatic lesions are not scored as structural aberrations. We have encountered such anomalous breakage associated with C-band positive regions of the chromosomes of a CHO-K1 cell line following exposure of the cells to toxic doses of U-68,553B and in this report present evidence that the apparent breaks are due to undercondensed
heterochromatin (UH) and evidence that the phenomenon appears to occur at higher frequency in a particular cell line of Chinese hamster. This finding has important implications on the assessment of potential risk due to exposure to the
drug. Such apparent breaks at sites of UH in chromosome 1 was not observed in an alternate CHO cell line (CHO-WBL) which supports the notion that the UH associated achromatic lesions in the CHO-K1 line may be a cell line specific phenomenon. Furthermore, careful electron microscopy of the chromosomes revealed
chromatin fibers connecting the apparently broken chromosomes. The UH was not observed in the presence of added metabolic activation (S9), and thus the significance of the phenomenon in risk assessment is further reduced. The data presented here provide evidence that sites of UH occur preferentially at locations of C-band positive constitutive
heterochromatin in CHO cells; we believe that this is the first report of induced fragile sites in rodent cells in vitro documented in this way. In addition, evidence is presented that U-68,553B lacks the ability to induce breakage in vivo in rodents and lacks the ability to induce
chromosome breakage in human peripheral lymphocytes in vitro. Therefore, it is concluded that the positive results with CHO-K1 cells treated with U-68,553B are unlikely to be predictive of a genotoxic hazard. This is a specific example of the importance of careful followup to an in vitro result in risk assessment.