The mechanism of augmentation of the primary antibody response in vitro by 2-mercaptoethanol (2-ME) was investigated. By using cystine-free RPMI 1640 medium, it was demonstrated that cyst(e)ine was absolutely required for eliciting the following murine lymphocyte reactions: antibody response to sheep erythrocytes, proliferative response to concanavalin A or lipopolysaccharide (LPS), and polyclonal antibody response induced by LPS. The maximal antibody response was attained with 2.5-5 mM cysteine or half-cystine. The serial feeding of fresh cysteine markedly amplified its capacity to support antibody response particularly when cysteine concentration was suboptimal. Such an effect was not observed in the serial addition of cystine. On the other hand, the dose-response curve of cystine was dramatically shifted to lower concentrations by the addition of 2-ME (1 x 10(-5) M), which alone could not elicit the antibody response in the absence of cystine, nor could it augment furthermore the maximal response induced by 2.5 mM half-cystine. Commercially available RPMI 1640 medium contains 0.41 mM half-cystine, which proved to be a suboptimal concentration for eliciting the maximal response. 35S-cystine was incorporated into murine lymphocytes five to six times more slowly than 35S-cysteine. The rate of cystine uptake, however, was accelerated by 2.5-fold in the presence of 1 x 10(-5) M 2-ME. A close correlation was observed between dose-response profiles of 2-ME in augmenting the antibody response and the stimulation of cystine uptake. These results strongly suggest that one of the roles of 2-ME in augmenting the antibody response in vitro is to facilitate the use of cystine contained in RPMI 1640 medium only at a suboptimal concentration.