Protective effect of Juzen-taiho-to on hepatocarcinogenesis is mediated through the inhibition of Kupffer cell-induced oxidative stress

Abstract

Traditional herbal formulations, such as Juzen-taiho-to (TJ-48), are used extensively in medical practice in Asia even though their mechanism of action remains elusive. This study tested a hypothesis that TJ-48 is protective against hepatocarcinogenesis by impeding Kupffer cell-induced oxidative stress. Forty-eight patients were randomly assigned to receive TJ-48 (n = 10), or no supplementation (n = 38) for up to 6 years after surgical treatment for hepatocellular carcinoma (HCC). In addition, to investigate the mechanism of protective action of TJ-48, diethylnitrosamine-containing water was administered for 22 weeks to male mice that were fed regular chow or TJ-48-containing diet. Liver tumor incidence, cell proliferation, number of 8-hydroxy-2'-deoxyguanosine- or F4/80-positive cells, and cytokine expression were evaluated. Although most of the patients experienced recurrence of HCC, a significantly longer intrahepatic recurrence-free survival was observed in the TJ-48 group. In mice, TJ-48 inhibited the development of liver tumors, reduced oxidative DNA damage, inflammatory cell infiltration and cytokine expression. Administration of TJ-48 improves intrahepatic recurrence-free survival after surgical treatment of hepatocellular carcinoma. On the basis of animal experiments, we reason that the protective mechanism of TJ-48 involves inhibition of Kupffer cells. This leads to lower levels of pro-inflammatory cytokines and oxidants in liver which may slow down the process of hepatocarcinogenesis and improves hepatic recurrence-free survival in patients with HCC.

Figure 1

Schematic representation of the treatments used in the mouse model of DEN-induced liver carcinogenesis. See Material and methods section for experimental details.

Figure 2

Administration of TJ-48 to patients who underwent surgical treatment of hepatocellular carcinoma (HCC) improves the length of intrahepatic recurrence-free survival. Kaplan–Meier method was used to plot hepatic recurrence-free survival in a total of 48 consecutive patients who were randomly assigned to receive TJ-48 (7.5 g daily, n = 10, solid line), or no treatment (n = 38, dashed line). Asterisk (*) denotes statistical significant (p <0.05, Log-rank test) difference between groups.

Figure 3

TJ-48 blunts infiltration of inflammatory cells in the mouse model of DEN-induced liver carcinogenesis. Liver sections were obtained from C57BL/6N (left panel) and C3H/HeN (right panel) mice administered normal chow and tap water (Control), normal chow and DEN (25 mg/l)-containing drinking water (DEN), or TJ-48 (1.6% w/w)-supplemented chow and DEN-containing drinking water (DEN + TJ-48) for 10 weeks. Immunohistochemical analysis for F4/80-positive cells (intense brown staining of single cells in the sinusoids) was performed as detailed in Material and methods section. Original magnification, ×100. Representative photomicrographs.

Figure 4

TJ-48 diminishes expression of proinflammatory cytokines in the mouse model of DEN-induced liver carcinogenesis. Liver samples were obtained from C57BL/6N (black bars) and C3H/HeN (open bars) mice administered normal chow and tap water (Control), normal chow and DEN (25 mg/l)-containing drinking water (DEN), or TJ-48 (1.6% w/w)-supplemented chow and DEN-containing drinking water (DEN + TJ-48) for 10 weeks. Expression of Tnfα (top) and IL-1β (bottom) mRNA was evaluated using quantitative RT-PCR as described in Materials and Methods. Fold change in gene expression (mean±S.E.M., n = 5 in each group) was derived from 2^ΔΔCt values and normalized to expression of the housekeeping gene Gus-b. Asterisks denote statistically significant (p <0.05, one-way ANOVA with Newman-Keul’s post-hoc test) difference as compared to control (*), or DEN (^#) groups within corresponding mouse strain.

Figure 5

TJ-48 blocks release of pro-inflammatory cytokines from DEN-activated Kupffer cells both in vivo and in vitro. For in vivo experiments (a–c), Kupffer cells were isolated from C57BL/6N mice administered normal chow and tap water (Control, open bars), normal chow and DEN (25 mg/l)-containing drinking water (DEN, black bars), or TJ-48 (1.6% w/w)-supplemented chow and DEN-containing drinking water (DEN + TJ-48, grey bars) for 5 weeks. Cells were cultured and protein levels of Tnfα (a), Il-1β (b) and IL-6 (c) were measured in the media as detailed in Material and methods section. Absolute control values of each measurement were as follows Tnfα: 35.0 ± 3.0 pg/ml, IL-1β: 105.3 ± 27.5 pg/ml, and IL-6: 82.2 ± 49.1 pg/ml. For in vitro experiments (d–f), Kupffer cells were isolated from C57BL/6N mice administered normal chow and either tap water (DEN-, open bars), or DEN-containing drinking water (DEN +, closed bars) for 5 weeks. Cells were cultured in presence of TJ-48 (0–1,000 μg/mL) and protein levels of Tnfα (d), IL-1β (e) and IL-6 (f) were measured in the media as detailed in Material and methods section. Data are shown as mean ± S.E.M. (n = 5 in each group). Absolute control values in this experiment were as follows Tnfα: 22.5 ± 6.2 pg/ml, IL-1β: 80.2 ± 21.5 pg/ml, and IL-6: 73.0 ± 45.0 pg/ml. Asterisks denote statistically significant (p <0.05, one-way ANOVA with Newman-Keul’s post-hoc test) difference as compared to control (*), or DEN (^#) groups.

Figure 6

TJ-48 blocks oxidant production by DEN-activated Kupffer cells both in vivo and in vitro. For in vivo experiments (a), C57BL/6N mice were fed normal chow (control) or TJ-48 (1.6% w/w)-supplemented diets for 5 weeks. Kupffer cells were isolated and cultured in normal media (control), or media containing DEN (2.5 mg/ml) and production of oxidants was assessed as detailed in Material and methods section. For in vitro experiments (b), Kupffer cells were isolated from naive C57BL/6N mice and cultured with TJ-48 (0 or 1,000 μg/ml) and in absence or presence of DEN (2.5 mg/ml) and production of oxidants was assessed using DCF-fluorescence as detailed in Material and methods section. Data are shown as mean ± S.E.M. (n = 3 in each group). Asterisks denote statistically significant (p <0.05, one-way ANOVA with Newman-Keul’s post-hoc test) difference as compared to control (*), or DEN (^#) groups.