10-Formyltetrahydrofolate dehydrogenase (EC 1.5.1.6) was previously identified as a folate-binding protein in rat liver cytosol (R.J. Cook and C. Wagner, Biochemistry 21, 4427-4434, 1982) by virtue of the tetrahydrofolate polyglutamate tightly bound to the partially purified enzyme. In this current study we provide evidence to show that when liver cytosol was rapidly processed to identify the protein bound folate, large amounts of both 10-formyl- and 5-formyltetrahydrofolate were present. After overnight storage of the cytosol at 5 degrees C before processing, almost no formylfolates were present and the major protein-bound form was tetrahydrofolate. This suggests that 10-formyltetrahydrofolate polyglutamates are tightly bound to the enzyme in vivo and are converted to tetrahydrofolate forms during isolation by the hydrolase activity associated with the enzyme. Covalent binding of the stable folate analogue, 5-formyltetrahydrofolate, to the purified enzyme resulted in 2 mol bound per mole of enzyme subunit. This is consistent with earlier reports suggesting the enzyme is capable of carrying out both oxidative and hydrolytic conversion of 10-formyltetrahydrofolate to tetrahydrofolate at the same time. Partial tryptic digestion of the purified enzyme selectively inhibited dehydrogenase activity of the enzyme but did not affect the hydrolase or aldehyde dehydrogenase activities.