The properties of yeast trehalose-6-phosphate synthase were reinvestigated in relation with the recent claim made by Panek et al. [Panek, A. C., de Araujo, P. S., Moura-Neto, V. and Panek, A. D. (1987) Curr. Genet. II, 459-465] that the enzyme would be stimulated by ATP and partially inactivated by cAMP-dependent protein kinase. Trehalose-6-phosphate synthase activity was measured by the sum of [14C]trehalose 6-phosphate and [14C]trehalose formed from UDP-[14C]glucose and glucose 6-phosphate. The activity measured in an extract of Saccharomyces cerevisiae was not affected by any treatment of the cells, such as incubation in the presence of glucose or of dinitrophenol, which are known to greatly increase the intracellular concentration of cAMP, nor by preincubation of the extract in the presence of ATP-Mg, cAMP and bovine heart cAMP-dependent protein kinase. The activity was also not significantly different in several mutants affected in the cAMP system. The kinetic properties of the partially purified enzyme were investigated; no effect of ATP could be detected but Pi acted as a potent noncompetitive inhibitor (Ki = 2 mM). The activity of trehalose-6-phosphate phosphatase was measured by the amount of [14C]trehalose formed from [14C]trehalose 6-phosphate. The enzyme could be separated from other phosphatases and appeared to be highly specific for trehalose 6-phosphate. It was Mg dependent and its kinetics for trehalose 6-phosphate was hyperbolic. Studies performed with intact cells, crude extracts or the purified enzyme did not reveal any cAMP-dependent change in its activity. Remarkably, trehalose-6-phosphate synthase and trehalose-6-phosphate phosphatase copurified in the course of different chromatographic procedures, suggesting that they are part of a single bifunctional protein. A 50-fold purification of the two enzymes could be achieved with a yield of only 2% by chromatography on Mono S followed by gel filtration on Superose 6B.