Suppression of cholangiocarcinoma cell growth by human umbilical cord mesenchymal stem cells: a possible role of Wnt and Akt signaling

Abstract

Emerging evidence indicates that human mesenchymal stem cells (hMSCs) can be recruited to tumor sites, and affect the growth of human malignancies. However, little is known about the underlying molecular mechanisms. Here, we observed the effects of hMSCs on the human cholangiocarcinoma cell line, HCCC-9810, using an animal transplantation model, and conditioned media from human umbilical cord-derived mesenchymal stem cells (hUC-MSCs). Animal studies showed that hUC-MSCs can inhibit the growth of cholangiocarcinoma xenograft tumors. In cell culture, conditioned media from hUC-MSCs inhibited proliferation and induced apoptosis of tumor cells in a dose- and time-dependent manner. The proliferation inhibition rate increased from 6.21% to 49.86%, whereas the apoptosis rate increased from 9.3% to 48.1% when HCCC-9810 cells were cultured with 50% hUC-MSC conditioned media for 24 h. Immunoblot analysis showed that the expression of phosphor-PDK1 (Ser241), phosphor-Akt (Ser 437 and Thr308), phosphorylated glycogen synthase kinase 3β (phospho-GSK-3β(Ser9)), β-catenin, cyclin-D1, and c-myc were down-regulated. We further demonstrated that CHIR99021, a GSK-3β inhibitor reversed the suppressive effects of hUC-MSCs on HCCC-9810 cells and increased the expression of β-catenin. The GSK-3β activator, sodium nitroprusside dehydrate (SNP), augmented the anti-tumor effects of hUC-MSCs and decreased the expression of β-catenin. IGF-1 acted as an Akt activator, and also reversed the suppressive effects of hUC-MSCs on HCCC-9810 cells. All these results suggest that hUC-MSCs could inhibit the malignant phenotype of HCCC-9810 human cholangiocarcinoma cell line. The cross-talk role of Wnt/β-catenin and PI3K/Akt signaling pathway, with GSK-3β as the key enzyme bridging these pathways, may contribute to the inhibition of cholangiocarcinoma cells by hUC-MSCs.

Figure 1. Characterization and differentiation of hUC-MSCs.

A–G, Flow-cytometric analysis of cell surface antigens of hUC-MSC; H, Alizarin red S staining of osteogenic differentiated hUC-MSCs (400×); I, Oil red O staining of adipogenic differentiated hUC-MSCs (400×).

Figure 2. The inhibitory effect of hUC-MSCs on intrahepatic cholangiocarcinoma.

HCCC-9810 and HUVEC controls cells were co-cultured with hMSCs for 48 h, after which cells were harvested for counting. The number of cells is represented as the mean±SD of three independent experiments. Results are shown as the percentage of cell number compared with the number of HUVEC control cells. Compared with that of the control, the cell proliferation rate decreased 38.58% (**P<0.05; A). Colony-forming assays showed that there were significantly fewer colony-forming units of HCCC-9810 cells treated with conditioned media from hUC-MSCs than that of control cells. The mean numbers of colony-forming units in the hUC-MSC media treatment group, and the HUVEC media control group were 14.7 and 35.7, respectively (P<0.01, B). HCCC-9810 cells were treated with various concentrations of conditioned media from hUC-MSCs (10%,25%,50% and 75%) for 12, 24 and 48 h. MTT assays showed that the inhibitory effects of test groups were significantly higher than that of the HUVEC control (C). The inhibition rate increased to 52.84% in the test group, while it only reached 6.87% in the controls after treatment for 48 h. This growth inhibition was not due to a lack of required nutrients due to consumption during preparation of the conditioned media since cell proliferation did not change when treated with 50% hUC-MSC conditioned media together with increasing concentrations of fetal calf serum (10%, 25%, and 50%) (D). The inhibitory effect was not limited to HCCC-9810 cells (E). There was a similar effect with Eca-109 cells, but not with MCF-7 or L-02 cells. When HCCC-9810 cells were co-injected with hUC-MSCs, smaller tumors formed (arrow) compared to those of HUVEC controls, and the formed tumors shrank in size after injection of conditioned media from hUC-MSC (F).

Figure 3. Induction of apoptosis by hUC-MSCs conditioned media in HCCC-9810 cells.

Cells were treated with 50% conditioned media from hUC-MSCs or HUVEC or without conditioned media (MOCK) for 48 h. The apoptotic tumor cells stained with Hoechst 33258 are designated by the arrow (A, 400×). The histogram shows the rates of Hoechst 33258-positive cells (P<0.05, B). Cells were incubated with 50% hUC-MSC conditioned media for 48 h. DNA fragmentation was detected by 1.5% agarose gel containing ethidium bromide. A representative blot was shown from three independent experiments (C).

Figure 4. Effect of conditioned media from hUC-MSCs on the Akt signaling pathway.

Immunoblot analysis showed that treatment of HCCC-9810 cells with 50% conditioned media from hUC-MSCs cultures led to the down-regulation of p-PI3K, p-PDK1^Ser241, p-Akt^Thr308, p-Akt^sSer473 and p- GSK-3β^Ser9 expression. However, this did not occur in HCCC-9810 cells with 50% conditioned media from HUVEC cultures. There were substantial differences in the phospho- to total Akt and Gsk-3β ratios between the treated group and control group (A). After treatment for 48 h with 50% conditioned media from hUC-MSCs cultures, the expression of phospho-Akt (B, 200×) and phospho-GSK-3β (C, 200×) in HCCC-9810 cells assessed by immunofluorescence. Results shown are representative of three independent experiments.

Figure 5. Effect of conditioned media from hUC-MSCs on the Wnt signaling pathway of HCCC-9810 cells.

Immunoblot analysis showed that the treatment of HCCC-9810 cells with 50% conditioned media from hUC-MSCs cultures led to the down-regulation of β-catenin, c-Myc, and cyclin-D1 expression, while the levels of these proteins in control groups was unchanged (A). After treatment for 48 h, 50% conditioned media from hUC-MSCs cultures prevented nuclear translocation of β-catenin in HCCC-9810 cells (B). Results shown are representative of three independent experiments.

Figure 6. Effect of the GSK-3β inhibitor/activator, and IGF-1 on hUC-MSCs-mediated intrahepatic cholangiocarcinoma cell toxicity.

HCCC-9810 cells were incubated with the GSK-3β activator SNP, and the GSK-3β inhibitor CHIR99021 as a single treatment or in combination with hUC-MSC conditioned media for 48 h, and cell survival was measured using the MTT assay (A, B). A similar MTT assay was carried out using IGF-1 (C). The expression of β-catenin and phospho-Akt (Ser 437 and Thr308) was measured by immunoblot analysis.

Figure 7. Thermolability of components of conditioned media.

The conditioned media were heated to various temperatures and cooled to room temperature prior to preparing the 50% conditioned media. When samples were heated to 75°C or above for 5 mins or longer, the growth inhibition decreased from 49.8% to 11.5%.

Figure 8. Proposed model by which hUC-MSC cultures mediates cross-talk between Wnt and Akt signaling in HCCC-9810 cells.

Inhibitory molecules in hUC-MSC conditioned media mediate Akt phosphorylation in HCCC-9810 cells, thereby decreasing phosphorylation of GSK-3βand increasing its activity. Activation of GSK-3βleads to decreased cellularβ-catenin levels. A decrease in β-catenin translocation to the nucleus to bind TCF4, and decreases the transcription of its specific target genes resulting in apoptosis.