This report describes the regulation of thymidine kinase (TK) enzyme level in mouse 3T6 fibroblasts experiencing a serum-induced transition between the resting (G0) and growing states. The specific activity of TK is about 200-500 pmoles TMP formed × min-1 × mg protein-1 in exponentially growing cells but is scarcely detectable (usually less than 5 pmoles TMP formed × min-1 × mg protein-1) in cells resting at confluence in medium containing 0.5 % serum. When the resting cells are serum-stimulated to re-enter the cell cycle, TK activity remains at the resting level until about 12 h following stimulation, but then increases dramatically as the cells enter S phase. Experiments with inhibitors of protein or RNA synthesis indicate that the increase in enzyme level is the result of de novo synthesis of the enzyme, and that the increase depends on transcription. The increase in TK level is not blocked when cells are stimulated in the presence of inhibitors of DNA synthesis, indicating that there is no tight coupling between these two processes. The increase is not blocked by the presence of thymidine ranging in concentration from 10-7 to 10-3 M, suggesting that fluctuations in the intracellular concentrations of thymidine or thymidylic acid do not affect the increase in TK level. When cells are treated with cycloheximide, the level of TK decreases with a 4-5 h half-life, indicating that the enzyme is relatively unstable. When the serum stimulus is withdrawn from cells in S phase (20 h following stimulation), the TK level decreases with the same 4 h half-life beginning about 6 h following serum withdrawal. The ability of the cell to control the rate of production of TK over a very wide range, along with the relative instability of the enzyme appear to be responsible for the rapid and wide variations in TK level during the cell cycle. The increase in the level of TK appears to be regulated in a manner similar to that of dihydrofolate reductase and thymidylate synthetase in serum-stimulated 3T6 cells, suggesting that the expression of these genes (and perhaps those for other S phase enzymes) may be coordinated by a common control signal.