Interleukin (IL)-12 is thought to be a key factor for the induction of interferon gamma (IFN-gamma), a cytokine essential for the lethal effects of endotoxin. We report here on the release of the nonfunctional subunit of IL-12, p40, as well as biologically active heterodimeric IL-12, p70, after administration of a lethal (n = 5) or sublethal (n = 8) dose of live Escherichia coli to baboons. Remarkably, on lethal challenge, peak levels of p40 were observed at 3 hours that were about twofold lower than those elicited after sublethal challenge (2,813 +/- 515 pg/mL v 4,972 +/- 732 pg/mL, P < .05). This disparity was also observed, although to a lesser extent, for IL-12 p70 antigen, of which maximum levels of 91 +/- 47 pg/mL and 151 +/- 41 pg/mL were measured 6 hours after a lethal or sublethal dose of E coli, respectively. Circulating p70 antigen correlated with IL-12 biologic activity (r = 0.869; P < .001). When comparing lethal to sublethal conditions, lower peak levels of IL-12 on lethal E coli sharply contrasted with higher levels of other proinflammatory cytokines, such as tumor necrosis factor (TNF)-alpha, IL-1beta, IL-6, and IL-8 observed in these animals. Lower IL-12 concentrations in the lethal group may have resulted in part from the enhanced production of IL-10, a known inhibitor of IL-12 synthesis in vitro, as peak levels of this cytokine 3 hours postchallenge inversely correlated with peak levels of IL-12, in particular p40 (r = -0.802; P < .01). Contrary to what might be expected if IFN-gamma were solely induced by IL-12, lethally challenged baboons generated threefold more IFN-gamma at 6 hours than those receiving a sublethal dose (P < .05). Moreover, higher levels of IFN-gamma were associated with lower p40/p70 ratios, suggesting that, in agreement with observations in vitro, IFN-gamma may have preferentially upregulated the release of p70 over p40. These data show that IL-12 is released in experimental septic shock in nonhuman primates and suggest that IL-10 and IFN-gamma are involved in the regulation of this release. Furthermore, this study indicates that the systemic release of IL-12 might be essential, but is not likely sufficient, to promote lethal production of IFN-gamma in sepsis.