2.1 Curcumin inhibited the proliferation of HCC cells
Fig. 1a shows the chemical structure of curcumin, a lipophilic polyphenol [33]. To assess the effect of curcumin on the proliferation of HCC cells, we treated HepG2 and SK-Hep-1 cells with curcumin at 10 µM, 20 µM, 40 µM, and 60 µM for 24, 48 and 72 h and then carried out CCK-8 analysis. The median inhibitory concentrations in HepG2 cells at 24, 48 and 72 h were 43.73 µM, 22.97 µM, and 21.17 µM, respectively. The median inhibitory concentrations of SK-Hep-1 cells at 24, 48 and 72 h were 40.11 µM, 27.76 µM, and 21.24 µM, respectively. HepG2 cells treated with curcumin at 10 µM, 20 µM, 40 µM, and 60 µM for 24 h had growth rates of 97.45%, 79.83%, 65.27%, and 30.26%, respectively. Under the same treatment conditions, SK-Hep-1 cells had growth rates of 99.91%, 94.02%, 71.73% and 40.21%, respectively (Fig. 1b). Cell cycle analysis using flow cytometry in HepG2 cells treated with vehicle and curcumin at 20 µM, 40 µM, and 60 µM, or 5-Fluorouracil (5FU, positive control) at 10 µM revealed the proportions of cells in G0/G1 phase to be 46.46%, 53.30%, 56.60%, 62.63%, and 71.03%, the proportions of cells in S phase to be 24.59%, 25.21%, 22.18%, 22.80% and 16.77%, and the proportions of cells in G2/M phase to be 28.95%, 21.83%, 21.21%, 14.57% and 12.20%. Under the same treatment conditions, the proportions of SK-Hep-1 cells in G0/G1 phase were 49.60%, 52.58%, 61.28%, 65.26%, and 74.72%, the proportions of cells in S phase were 20.96%, 21.21%, 16.59%, 16.78%, and 12.96%, and the proportions of cells in G2/M phase were 29.44%, 26.21%, 22.13%, 17.97%, and 12.32%. DNA content in the G0/G1 phase increased significantly, while that in G2/M phase and S phase decreased relatively (Fig. 1c). These results indicated that cells were blocked in G0/G1 phase and could not enter S phase, resulting in a decrease in cell number in G2/M phase and decreased proliferation index.
2.2 Curcumin promoted apoptosis in HCC cells
Next, we treated HepG2 and SK-Hep-1 cells for 24 h using curcumin at 20 µM, 40 µM, and 60 µM, or 5FU at 10 µM and examined their effect on apoptosis. Apoptosis is characterized by chromatin shrinkage, which appears as dense structures upon Hoechst 33258 staining. Hoechst 33258 staining revealed that curcumin increased the number of apoptotic cells dose-dependently (Fig. 2a). Annexin V / PI double staining revealed the apoptosis rate to be 5.00%, 13.92%, 16.16%, 26.87% and 39.59%, respectively, in HepG2 cells and 9.30%, 18.24%, 26.43%, 30.58%, and 32.14%, respectively, in SK-Hep-1 cells. Compared with the control group, apoptosis was significantly elevated in cells treated with curcumin at 40 µM or with 5FU (Fig. 2b). Western blot analysis revealed that when compared with the control group, treatment with curcumin increased the levels of the apoptosis-related proteins, Cleaved caspase-3 and Cleaved caspase-9, dose-dependently (Fig. 2c) and also enhanced the expression of the pro-apoptotic factors, Bax and Cyt-C, while suppressing the levels of the anti-apoptotic factor, Bcl-2 (Fig. 2d). To determine if curcumin induced mitochondrial apoptosis pathways, we evaluated changes in mitochondrial membrane potential using a fluorescence probe and observed increased green fluorescence and decreased red fluorescence in curcumin-treated HCC cells, indicating reduced mitochondrial membrane potential (Fig. 2e). These results suggest that curcumin promotes apoptosis in HCC cells via the mitochondrial apoptosis pathway.
2.3 Curcumin inhibits PI3K/AKT/GSK-3β signaling
To evaluate if curcumin triggers apoptosis in HCC cells through the PI3K/AKT/GSK-3β pathway, we treated HepG2 and SK-Hep-1 cells for 24 h with curcumin at 20 µM, 40 µM, and 60 µM, or 5FU at 10 µM. Subsequent western blot analysis revealed that curcumin significantly reduced the levels of PI3K, p-PI3K, AKT, p-AKT, GSK-3β, and p-GSK-3β (Fig. 3a). To further test the effect of curcumin on the PI3K/AKT/GSK-3β pathway, we pretreated cells with LY294002, a PI3K inhibitor, and observed that treatment with curcumin (40 µM) enhanced the inhibition of the PI3K/AKT/GSK-3β pathway, upregulated Bax levels, and downregulated Bcl-2 levels (Fig. 3b).
2.4 BCLAF1 knockout enhanced curcumin-induced apoptosis in HCC cells
Western blot analysis revealed that treatment with curcumin decreased BCLAF1 levels in a concentration-dependent manner (Fig. 4a). To test if the effects of curcumin on the PI3K/AKT/GSK-3β pathway induced apoptosis in the HCC cell lines by modulating BCLAF1, we knocked out BCLAF1 in HepG2 and SK-Hep-1 cells using CRISPR/Cas9 technology and treated them with curcumin at 40 µM for 24 h. Cell cycle analysis showed that compared with the control group and curcumin group, the amount of DNA in the G0/G1 phase was markedly higher in the curcumin+sgRNA group, while the proportion in the S and G2/M phase was significantly lower (Fig. 4b). Annexin V / PI double staining showed that compared with the control group and curcumin-treated group, apoptosis was significantly higher in the curcumin+sgRNA group (Fig. 4c). Western blot analysis revealed that compared with the curcumin-treated group, the levels of Bax and Cyt-c were increased in the BCLAF1 knock out cells while the level of Bcl-2 decreased (Fig. 4d). Moreover, curcumin further inhibited the expression of PI3K/AKT/GSK-3β pathway-related proteins. The protein expression levels of p-PI3K, p-AKT, and p-GSK-3β in the curcumin+sgRNA group were dramatically decreased when compared with the curcumin alone group (Fig. 4e). Analysis of mitochondrial membrane potential showed that compared with the control group and the curcumin-treated group, the curcumin+sgRNA group had higher levels of green fluorescence and lower levels of red fluorescence, indicating decreased mitochondrial membrane potential (Fig. 4f).
2.5 Curcumin inhibits tumor growth in a nude mouse xenograft model of HCC and regulates mitochondrial apoptosis in the PI3K/AKT/GSK-3β pathway mediated by BCLAF1
Next, given that in vitro analyses suggested that curcumin inhibits the proliferation of HCC cell lines, we sought to evaluate the antitumor effects of curcumin in vivo. To this end, we established a nude mouse xenograft model of HCC. We treated the mice with curcumin at 20 µM, 40 µM, and 60 µM, or with 5FU at 10 µM and observed that curcumin inhibited tumor growth in a dose-dependent manner, as indicated by the tumor volumes of 1832.49±1001.37, 1285.93±547.63, 1117.17±752.54, 713.36±262.69 and 271.69±154.69 mm3, respectively (Fig. 5a). Analysis of tumor volume and growth inhibition curves revealed that when compared with the negative control group, the volume of the xenograft tumors decreased with increasing concentration of curcumin (Fig. 5b–c). Next, we performed pathological analysis of the tumors using H&E staining. This analysis found that when compared with the negative control group, curcumin altered the cell structure of the xenograft tumor cells, reduced their cell volume and nucleo-plasma ratio, and caused apoptosis (Fig. 5d). Immunohistochemical analysis showed that Bcl-2 and Bax localized to the cytoplasm, that curcumin enhanced Bax expression and inhibited BCLAF1 expression, and that BCLAF1 was expressed in the cytoplasm and nucleus (Fig. 5e). Western blot analysis of apoptosis-related proteins showed that expression of the PI3K/AKT/GSK-3β pathway-related proteins, PI3K, p-PI3K, AKT, p-AKT, GSK-3β and p-GSK-3β was decreased, with a more apparent decrease in the levels of the phosphorylated proteins (Fig. 5f).