The ability of astroglia-rich primary cultures derived from the brains of neonatal rats to take up and metabolize various sulfur containing compounds to cysteine was investigated using the content of intracellular glutathione as an indicator. Astroglial cells were partially depleted of glutathione by starvation for 24 h. Subsequent feeding for 4 h with glucose, glycine, and glutamate resulted in a restoration of the glutathione level, if cysteine was present. Substitution of cysteine by cystine during resynthesis of glutathione led to a glutathione content which exceeded that of cysteine-refed cells by 41%. Half-maximal content of glutathione was found at a concentration of about 12 Β΅M cysteine and a maximal content at a concentration of at least 50 Β΅M cysteine. In contrast, no plateau in the glutathione level was reached with increasing concentrations of cystine. The cystine effect could not be due to a contamination, since it was abolished after reduction of cystine by dithiothreitol. Since the cystine effect was not affected by inhibiting β₯-glutamyl transpeptidase, a promotion of cystine uptake by formation of β₯-glutamylcystine can also be excluded. Of the potential cysteine precursors tested, N-acetylcysteine was able to replace cysteine half-maximally at a concentration of 1 mM and fully at 5 mM. Feeding 2-oxothiazolidine-4-carboxylic acid at a concentration of 5 mM resulted in 64% of the glutathione level found in the presence of cysteine. A half-maximal glutathione content was attained at 50 Β΅M 2-oxothiazolidine-4-carboxylic acid. While cystathionine could partially replace cysteine, methionine and homocysteine were not at all able to substitute for cysteine. These results demonstrate that astroglial cells prefer cystine from cysteine for glutathione synthesis and express uptake systems for N-acetylcysteine, 2-oxothiazolidine-4-carboxylic acid, and cystathionine, as well as the enzymes N-deacetylase, 5-oxoprolinase, and cystathionine β₯-lyase.