gms | German Medical Science

Introduction: Hepatic tissue damage can occur due to several diseases like cholestasis, infections or liver tumors. Kupffer cells (KC) play an important role in mediating tissue damage as well as regenerative processes. KC activation by lipopolysaccharide (LPS), antigens or endogenous proteins such as released under necrotic conditions can lead to an activation of the NF-κB signaling pathway transmitted by reactive oxygen intermediates (ROI). Once activated, KC release different cytokine patterns that support inflammation or induce tolerance reactions. Aim of the present study was the establishment of a human in vitro model for the simulation of hepatic tissue damage which enables the investigation of immune mediated signaling in hepatic inflammatory reactions.

Material and methods: Primary human hepatocytes (PHH) and KC were isolated from human liver resectates using a two-step collagenase perfusion technique. KC activation was investigated by measuring the intracellular formation of ROI using the DCF assay with and without LPS treatment. Mitochondrial activity as a marker for cell activity/viability was measured using the XTT assay. In order to simulate the activation of KC following hepatocyte damage, KC were incubated with supernatants of PHH that had previously been treated with hepatotoxic compounds. KC reaction was determined by measurement of pro- and anti-inflammatory cytokine concentrations using specific cytokine ELISA sets.

Results: The isolated KC yield was 1.25±0.89x10⁶ cells/g liver tissue with a purity of >80% measured by immunofluorescence staining of CD 68 and phagocytosis of latex beads. Initial ROI levels in KC showed high donor variability, depending on the condition of the liver tissue and patient anamnesis. The activation of KC by LPS caused a concentration-dependent increase in ROI formation. Incubation of KC with supernatants from drug-treated PHH increased mitochondrial activity and lead to a change in the formation of ROI compared to KC incubated with supernatants of untreated PHH. Additionally we detected a compound and donor dependent increase in proinflammatory cytokines TNF-α and IL-6 or in anti-inflammatory cytokines PGE₂ and IL-10, respectively.

Conclusion: Taken together, the isolated KC populations showed different basal KC activations, which are linked to the quality and the state of disease of the donor liver tissue. Nevertheless, LPS and supernatants of compound-treated PHH induced an additional KC activation in vitro. Interpretation of the cytokine profile revealed that the KC reactions depends strongly on the initial degree of inflammation in the liver tissue suggesting a connection to the initiation of inflammatory or regenerative processes. First attempts aiming the transfer of this method into a transwell coculture and the determination of the donor-dependent cytokine production are made for the development of a system biological approach.

This study was supported by the Virtual-Liver-Network, BMBF:0315741