Differential effects of arsenic trioxide on chemosensitization in human hepatic tumor and stellate cell lines

Abstract

Background: Crosstalk between malignant hepatocytes and the surrounding peritumoral stroma is a key modulator of hepatocarcinogenesis and therapeutic resistance. To examine the chemotherapy resistance of these two cellular compartments in vitro, we evaluated a well-established hepatic tumor cell line, HepG2, and an adult hepatic stellate cell line, LX2. The aim was to compare the chemosensitization potential of arsenic trioxide (ATO) in combination with sorafenib or fluorouracil (5-FU), in both hepatic tumor cells and stromal cells.

Methods: Cytotoxicity of ATO, 5-FU, and sorafenib, alone and in combination against HepG2 cells and LX2 cells was measured by an automated high throughput cell-based proliferation assay. Changes in survival and apoptotic signaling pathways were analyzed by flow cytometry and western blot. Gene expression of the 5-FU metabolic enzyme, thymidylate synthase, was analyzed by real time PCR.

Results: Both HepG2 and LX2 cell lines were susceptible to single agent sorafenib and ATO at 24 hr (ATO IC(50): 5.3 μM in LX2; 32.7 μM in HepG2; Sorafenib IC(50): 11.8 μM in LX2; 9.9 μM in HepG2). In contrast, 5-FU cytotoxicity required higher concentrations and prolonged (48-72 hr) drug exposure. Concurrent ATO and 5-FU treatment of HepG2 cells was synergistic, leading to increased cytotoxicity due in part to modulation of thymidylate synthase levels by ATO. Concurrent ATO and sorafenib treatment showed a trend towards increased HepG2 cytotoxicity, possibly due to a significant decrease in MAPK activation in comparison to treatment with ATO alone.

Conclusions: ATO differentially sensitizes hepatic tumor cells and adult hepatic stellate cells to 5-FU and sorafenib. Given the importance of both of these cell types in hepatocarcinogenesis, these data have implications for the rational development of anti-cancer therapy combinations for the treatment of hepatocellular carcinoma (HCC).

Figure 1

ATO inhibits HepG2 and LX2 Cell Proliferation. HepG2 (A) and LX2 (B) cells were treated with ATO at the indicated concentrations for 24 hrs, 48 hrs, 72 hrs and 96 hrs (HepG2 only) and cell proliferation assessed by MTT assay. For each concentration, percent inhibition values were calculated and data normalized to vehicle control. Relative IC₅₀ values shown in the corresponding tables were determined by non-linear regression in GraphPad Prism (n = 3 replicates).

Figure 2

ATO induces apoptosis of LX2 cells. HepG2 and LX2 cells were untreated or incubated in the presence of 5 μM or 25 μM ATO for 24 hrs. Apoptotic cells were determined by propidium iodide staining and FACS analysis. The red arrow indicates the sub –G1 peak which represents an early apoptotic cell population. A representative example of three independent experiments is shown.

Figure 3

5-FU inhibits HepG2 and LX2 Cell Proliferation at 72 hrs. HepG2 (A) and LX2 (B) cells were treated with 5-FU at the indicated concentrations for 24 and 72 hrs and cell proliferation was assessed by MTT assay. For each concentration, percent inhibition values were calculated and data normalized to vehicle control. IC₅₀ values shown in the corresponding tables were determined by non-linear regression in GraphPad Prism (n = 3 replicates).

Figure 4

Combination of ATO with 5-FU in HepG2 and LX2 cell lines. HepG2 cells (A) were treated with increasing concentrations of 5-FU in the absence or presence of ATO (5 μM) for 24 hrs (left panel). The bar graph (right panel) demonstrates relative proliferation values for HepG2 cells when treated with 5 mg/ml 5-FU alone, 5 μM ATO alone and the combination of 5 mg/ml 5-FU and 5 μM ATO (p value < 0.0001). (B) LX2 cells were treated with increasing concentrations of 5-FU in the absence or presence of ATO (10 μM) for 24 hrs (left panel). The bar graph (right panel) demonstrates relative proliferation values for LX2 cells when treated with 5 mg/ml 5-FU alone, 5 μM, 10 μM ATO alone and the combination of 5 mg/ml 5-FU + 5 μM or 10 μM ATO. For each concentration, percent inhibition values were calculated and data normalized to vehicle control and represented as n = 3 replicates+/− SEM. (C) Dose effect curve for ATO and 5-FU combination in HepG2 cells was generated using Calcusyn software. (D) HepG2 treated cell lysates (vehicle, 20 μM, 50 μM or 75 μM ATO, 5 mg/ml 5-FU or 5 μM ATO + 5 mg/ml 5-FU) and LX2 treated cell lysates (vehicle, 10 μM, 20 μM or 50 μM ATO, 5 mg/ml 5-FU or 10 μM ATO + 5 mg/ml 5-FU) assessed for procaspase-9 and XIAP expression by western blot analysis. GAPDH was utilized as a loading control.

Figure 5

ATO inhibits thymidylate synthase expression in a dose dependent fashion. HepG2 cells were treated for 24 hrs with vehicle, 0.5 μM, 5 μM or 25 μM ATO. Thymidylate synthase mRNA expression was assessed by qRT-PCR and normalized to actin levels. Each sample was run in triplicate. A representative example of four independent experiments is shown. Bar graph demonstrates fold change relative to vehicle control +/− SEM.

Figure 6

ATO treatment of HepG2 cells results in sustained activation of MAPK and JNK. HepG2 and LX2 cells were treated with vehicle or 5–50 μM ATO as indicated for 24 hrs. Levels of phosphorylated and total MAPK (A) and JNK (B) were assessed by western blot analysis. GAPDH was utilized as a loading control.

Figure 7

Treatment of HepG2 cell lines with ATO and Sorafenib results in enhanced cytotoxicity and inhibits MAPK activation. HepG2 (A) and LX2 (B) cells were treated with Sorafenib at the indicated concentrations for 24 hr, 48 hr, 72 hr and 96 hr and cell proliferation assessed by MTT assay. For each concentration, percent inhibition values were calculated and data normalized to vehicle control. IC₅₀ values shown in the corresponding tables were determined by non-linear regression in GraphPad Prism. HepG2 (C) and LX2 (D) cells were treated with increasing concentrations of Sorafenib in the absence or presence of ATO (2.5 μM) for 48 h. Inhibition values for ATO alone (2.5 μM) are shown (◊). (E). HepG2 cells were untreated or treated with vehicle, 2.5 μM ATO or 0.1–20 μM sorafenib +/− 2.5 μM ATO as indicated for 24 hrs. Phosphorylated and total MAPK levels were assessed by western blot analysis. GAPDH was utilized as a loading control.