Objectives: Endotoxin (lipopolysaccharide) tolerance is characterized by a transient refractory state to a subsequent lipopolysaccharide challenge. Following human endotoxemia, ex vivo tolerance of circulating leukocytes to lipopolysaccharide resolves within 24 hrs. However, the duration of in vivo tolerance, assumed to be primarily mediated by tissue-resident macrophages, is unknown. DESIGN, SETTING, SUBJECTS, AND INTERVENTIONS: Clinical experimental study in 16 healthy male volunteers at an intensive care research unit. To compare ex vivo and in vivo tolerance kinetics, whole blood from healthy volunteers was stimulated with lipopolysaccharide before, 4 hrs after, and 1, 2, 3, and 4 wks following in vivo endotoxin (2 ng/kg; lipopolysaccharide derived from Escherichia coli O:113) administration. Furthermore, we compared the inflammatory response during two subsequent endotoxemia experiments in healthy volunteers with an interval of 2 wks. The cytokines tumor necrosis factor-α, interleukin-6, interleukin-10, interleukin-1 receptor antagonist, and transforming growth factor-β were measured.
Measurements and main results: Four hours after in vivo lipopolysaccharide administration, production of tumor necrosis factor-α, interleukin-6, and interleukin-10, but not interleukin-1 receptor antagonist in ex vivo lipopolysaccharide-stimulated whole blood was diminished. Ex vivo lipopolysaccharide tolerance completely resolved within 1 week. In contrast, in vivo lipopolysaccharide tolerance was still apparent after 2 wks. Compared to the first lipopolysaccharide administration, plasma peak levels of tumor necrosis factor-α, interleukin-6, interleukin-10, interleukin-1 receptor antagonist, and transforming growth factor-β were attenuated by 46%, 36%, 45%, 10%, and 14%, respectively (all p < .05).
Conclusions: While ex vivo lipopolysaccharide tolerance quickly resolves, in vivo lipopolysaccharide tolerance persists for at least 2 wks. These findings strengthen the notion that the in vivo response to lipopolysaccharide is mediated by tissue-resident macrophages and that ex vivo stimulation does not accurately reflect the in vivo innate immune response. Intervention studies utilizing the human endotoxemia model should be performed using parallel groups rather than a crossover design.