Effect of caffeine on DNA synthesis in mammalian cells.

Alkaline sucrose sedimentation studies of DNA from mouse lymphoma cells (L5178Y) treated with caffeine have demonstrated the following effects. Caffeine (at a concentration of 1.6 mM) does not introduce strand breaks into preformed DNA nor does it inhibit the rejoining of gamma-ray-induced strand breaks. Although it does not affect the over-all rate of DNA synthesis, pulse labeling experiments show that the DNA strands synthesized in its presence are smaller than those made in its absence. This could be the result of (a) DNA being made in shorter replicating units or (b) small gaps in the daughter DNA strands within normal-sized replicating units. These two alternative models were indirectly distinguished as follows. After a pulse label with thymidine-(3)H in the presence of caffeine, cells were incubated without caffeine in bromodeoxyuridine (BrdUrd). During this incubation, growing strands are elongated and hypothetical gaps (model b) filled in with bromuracil (BrUra)-substituted DNA. The BrUra-containing DNA segments will now be of different lengths on the two models. With smaller replicating units (a) the "elongation segments" will be somewhat smaller than but the same order of magnitude as those in untreated cells, whereas with small gaps (b) the "filled-in gap segments" would be expected to be at least an order of magnitude smaller. The BrUra-containing regions of DNA can be selectively broken open by exposing the cells to light at 313 nm. The exposure required to break open the BUra-substituted regions is inversely related to, and hence gives a measure of, the size of these regions. In caffeine-treated cells these regions were found to be somewhat smaller than but of comparable size with those in untreated cells; this is consistent with the DNA being synthesized in smaller units and argues against the presence of small gaps in the daughter strands.