Studies of [3H]diazepam binding to intact rat pineal cells were carried out in tissue culture preparations. The binding was saturable, reversible and proportional to the number of cells used. Scatchard analysis resulted in a linear plot [Kd = 23 nM, maximum binding sites (Bmax) = 1.56 pmol/mg of protein for cells in monolayer culture; Kd = 7 nM, Bmax = 1.3 pmol/mg of protein for cells in suspension culture]. Inhibition constants (Ki) for clonazepam (500 nM), flunitrazepam (38 nM) and Ro-5-4864 (5 nM) indicated that the binding sites were probably of the "peripheral" type. In addition, the effects of diazepam on norepinephrine-stimulated N-acetyltransferase (NAT) activity were studied in organ culture and dissociated cell culture. Diazepam (10-50 microM) both prolonged and increased the magnitude of the norepinephrine-induced increase in NAT activity but did not affect the initial rate of rise of enzyme activity. The effect was dose-dependent and was also seen with clonazepam, flunitrazepam and Ro-5-4864, but not with Ro-15-1788. Diazepam, by itself, at these concentrations, had no effect on NAT, but enzyme activity was increased by higher concentrations (0.1-1 mM). Although a relationship between the [3H]diazepam binding sites described here and the effect of benzodiazepines on NAT cannot be established from these studies, the data suggest that the benzodiazepines may alter melatonin levels through their action on NAT.