.49). During the same period, neither overall cell growth nor the concentrations of malate dehydrogenase (EC 1.1.1.37), acid phosphatase (EC 3.1.3.2), or alanine aminotransferase (EC 2.6.1.2) were significantly decreased by the base analog. Addition of thymidine to the growth medium rapidly counteracts the inhibition of tyrosine aminotransferase synthesis but restores the normal concentrations of lactate-, alcohol-, and glucose-6-phosphate dehydrogenases much more slowly. Growth of the cells for only one generation in the presence of bromodeoxyuridine, followed by the addition of thymidine, produces transient decreases in the concentrations of the three "late-responding" dehydrogenases, beginning 2-3 generations after exposure to the analog.It is concluded that the selective inhibitory effects of the analog could result from a mechanism in which bromodeoxyuridine is uniformly incorporated into cellular DNA, but inhibits the transcription of only certain genes into messenger RNA. A mathematical model is derived to account for the observed differences in the kinetics of the inhibition of synthesis of the gene products that are sensitive to the analog.The appearance of certain differentiated functions in embryonic tissues is suppressed in vitro by 5-bromodeoxyuridine, BrdU (1-4), as is the tumorigenicity of mouse melanoma cells (5). We reported earlier that the synthesis of the inducible enzyme tyrosine aminotransferase (TAT) is dramatically ieduced when an established line (HTC) of rat hepatoma cells is grown in culture with BrdU, and we have used this system to study the mechanisms by which BrdU affects specific cellular functions (6). These studies showed that the thymidine analog must be incorporated into DNA to decrease TAT synthesis, and that the effect of BrdU in HTC cells is rather selective, since the rates of cell growth and of general protein and RNA synthesis are relatively unaffected (6).
MATERIALS AND METHODSThe procedures described previously (6) were used for growing HTC cells, preparing cell-free extracts by freeze-thawing, and assaying for TAT activity. The same cell lysates were assayed spectrophotometrically for their contents of alcohol (7), lactate (8), glucose-6-phosphate (9), and malate (10) dehydrogenases as described in the references cited. Alanine amiinotransferase was assayed in cell extracts by a method described by Larsson and Tomkins'(to be published). Acid l)hosphatase was determined in uncentrifuged cell lysates prepared by thawing frozen cell pellets in 0.02 M Tris.HCI buffer, pH 7.6. The assa-was performed in 1 ml of 80 mMI sodium acetate buffer, pH's, containing 0.9 mMI p-nitrophenyl phosphate, 2 mMI EDlTA, and 0.4 mg/ml bovine serum albumin. The reacting was terminated with 0.1 ml of 10 N KOH, and p-nitrobphenol was measured spectrophotometrically at 400 nim. A unit of enzyme activity is defined as the production of 1 4mol of product per min at the temperature of the assay (370C for TAT and acid phosphatase, and room teml)erature for the others).
RESULTSAs reported previously ...