Interleukin 2 (IL-2) mediates the regression of metastatic cancer, but its clinical use is limited by associated toxicities including hepatic dysfunction. To determine the mechanism for IL-2-induced hepatic dysfunction, we hypothesized that IL-2 activation of Kupffer cells cause leukocyte-endothelial adhesion and decreases hepatic sinusoidal blood flow. C57BL/6 mice were given injections of latex particles and prepared for intravital hepatic microscopy 2 h after i.p. IL-2 administration. Liver tissue was also prepared to quantitate hepatic tumor necrosis factor (TNF) mRNA and processed for light and electron microscopy. Phagocytosing Kupffer cells and leukocytes adherent to the endothelium were counted, and surface sinusoidal blood flow was quantitated. Kupffer cell activity was quantitated as the ratio of phagocytosing Kupffer cells to sinusoidal blood flow. IL-2 significantly increased Kupffer cell activity (0.56 ± 0.05 for controls versus 0.84 ± 0.05 for IL-2), significantly caused leukocyte-endothelial adhesion (26.7 ± 7.9 for controls versus 87.0 ± 27.6 for IL-2, WBC/mm2 endothelial surface), and significantly decreased the number of sinusoids containing blood flow per microscopic field (6.66 ± 0.15 for controls versus 5.79 ± 0.13 for IL-2) without causing changes in systemic hemodynamic parameters. In IL-2-treated livers, light and electron microscopy showed the constriction of sinusoids associated with swollen or ruptured mitochondria, which was consistent with hypoxic deterioration near central venules. Adherent platelets, neutrophils, and lymphocytes within sinusoids and central venules were also observed. PCR revealed that IL-2 significantly induced TNF mRNA expression in the liver. These data suggest that IL-2 activate Kupffer cells in association with the release of monokines including TNF, which causes activation of circulating leukocytes as well as hepatic sinusoidal endothelial cells. The resultant leukocyte and platelet adhesion to the endothelium may then physically impede the sinusoidal microcirculation, resulting in microscopic areas of hepatic ischemia and explaining the mechanism of IL-2-induced hepatic dysfunction.
This work was supported by NIH Grant CA-57527.