INTRAVENOUS ENDOTOXIN SUPPRESSES THE CYTOKINE RESPONSE OF PERIPHERAL-BLOOD MONONUCLEAR-CELLS OF HEALTHY HUMANS

When administered parenterally, endotoxin stimulates the synthesis of IL-1, TNF-alpha, and IL-6. However, this initial injection induces tol erance; a second injection of endotoxin results in lower levels of cir culating cytokines. In our study, five healthy male volunteers between the ages of 18 and 30 were injected with Escherichia coli endotoxin. Four subjects received only saline. Immediately before the injection a nd 3, 6, and 24 h afterward, PBMC were isolated and stimulated in vitr o with endotoxin, IL-1, or toxic shock syndrome toxin-1. Inasmuch as C D14+ monocytes are the primary source of the cytokines induced by thes e stimuli, results are expressed as cytokine production per 10(6) CD14 + cells. Six h after endotoxin injection, endotoxin-stimulated CD14+ c ells synthesized 66% less IL-1beta (p < 0.01), 47% less TNF-alpha (p < 0.001), 56% less IL-6 (p < 0.01), and 49% less IL-8 (p < 0.01) than c ells obtained before the injection. This suppression was not specific for endotoxin; IL-1beta-induced IL-1alpha and TNF-alpha were reduced b y 84% (p = 0.01) and 68% (p < 0.001), respectively. A decrease in cyto kine synthesis was also observed using toxic shock syndrome toxin-1 as a stimulus: 57% for IL-1beta (p = 0.06), 70% for TNF-alpha (p < 0.01) , 56% for IL-6 (p < 0.05), and 71 % for IL-8 (p = 0.001). When data we re expressed as cytokine production per 10(6) PBMC, cells isolated 3 h after endotoxin injection synthesized significantly less stimulus-ind uced IL-1, TNF-alpha, IL-6, and IL-8 than did PBMC from saline-injecte d controls. We conclude that endotoxin tolerance is due, in part, to c hanges in the stimulus-induced cytokine response of circulating CD14cells.