Chloroquine potentiates the anti-cancer effect of 5-fluorouracil on colon cancer cells

Abstract

Background: Chloroquine (CQ), the worldwide used anti-malarial drug, has recently being focused as a potential anti-cancer agent as well as a chemosensitizer when used in combination with anti-cancer drugs. It has been shown to inhibit cell growth and/or to induce cell death in various types of cancer. 5-Fluorouracil (5-FU) is the chemotherapeutic agent of first choice in colorectal cancer, but in most cases, resistance to 5-FU develops through various mechanisms. Here, we focused on the combination of CQ as a mechanism to potentiate the inhibitory effect of 5-FU on human colon cancer cells.

Methods: HT-29 cells were treated with CQ and/or 5-FU, and their proliferative ability, apoptosis and autophagy induction effects, and the affection of the cell cycle were evaluated. The proliferative ability of HT-29 was analyzed by the MTS assay. Apoptosis was quantified by flow-cytometry after double-staining of the cells with AnnexinV/PI. The cell cycle was evaluated by flow-cytometry after staining of cells with PI. Autophagy was quantified by flow-cytometry and Western blot analysis. Finally, to evaluate the fate of the cells treated with CQ and/or 5-FU, the colony formation assay was performed.

Results: 5-FU inhibited the proliferative activity of HT-29 cells, which was mostly dependent on the arrest of the cells to the G0/G1-phase but also partially on apoptosis induction, and the effect was potentiated by CQ pre-treatment. The potentiation of the inhibitory effect of 5-FU by CQ was dependent on the increase of p21Cip1 and p27Kip1 and the decrease of CDK2. Since CQ is reported to inhibit autophagy, the catabolic process necessary for cell survival under conditions of cell starvation or stress, which is induced by cancer cells as a protective mechanism against chemotherapeutic agents, we also analyzed the induction of autophagy in HT-29. HT-29 induced autophagy in response to 5-FU, and CQ inhibited this induction, a possible mechanism of the potentiation of the anti-cancer effect of 5-FU.

Conclusion: Our findings suggest that the combination therapy with CQ should be a novel therapeutic modality to improve efficacy of 5-FU-based chemotherapy, possibly by inhibiting autophagy-dependent resistance to chemotherapy.

Figure 1

Effect of 5-FU and CQ on the proliferative activity of HT-29 cells. (A) The proliferative activity of 5-FU-treated HT-29 cells for 24, 48, 72 hours assessed by the MTS assay. The y-axis represents the proliferation rate, calculated as the ratio to control untreated cells. The 5-FU-induced inhibitory effect of HT-29 cells was time and dose-dependently increased. (B) The proliferative activity of CQ-treated HT-29 cells for 12 and 24 hours assessed by the MTS assay. The y-axis represents the proliferation rate, calculated as the ratio to control untreated cells. CQ treatment at 1000 μM for 12 h or 100 μM and higher doses for 24 h resulted in significant inhibition of the proliferative activity of HT-29 cells, but not at lower doses. Values were given as mean.

Figure 2

Effect of CQ in 5-FU-induced cell proliferation and growth inhibition. The effect of CQ-pretreatment on the inhibitory effect of 5-FU on the proliferative activity/cell growth of HT-29 cells was investigated by MTS assay/trypan blue exclusion staining. The y-axis represents the proliferation/growth rate, calculated as the ratio to control untreated cells. Values were given as mean ± SD. (A) 5-FU at 5 μM is reported to cause minimal adverse health effects to human beings. Pre-treatment of cells with CQ at 80 μM for 12 hours prior to exposure to 5-FU at 5 μM for 48 hours, resulted in potentiation of the inhibitory effect (33% vs. 67% inhibition for 5-FU alone and CQ + 5-FU, respectively). (*, p < 0.05 vs. control) (B) When cells were treated with CQ at 80 μM for 12 hours prior to exposure to 5-FU at 5 μM for 48 hours, the growth rate reduction was potentiated (60% vs 23% live cells for 5-FU alone and CQ + 5FU, respectively). (*, p < 0.05 vs. control)

Figure 3

CQ increased the formation of Acidic vesicular organelles (AVOs). (A) The formation of AVOs was quantified by flow-cytometry after acridine orange staining in HT-29 cells treated with 5-FU for 48 h after 12-h pretreatment with CQ. Treatment of HT-29 cells with 5-FU alone resulted in increased AVOs formation, but the effect was potentiated by CQ-pretreatment. The results are expressed as the mean fluorescein intensity of acridine orange stained cells. (B) The similar results were obtained with the fluorescence microscopic examination, confirming the induction of autophagy. HT-29 cells treated with CQ and 5-FU displayed great number of red fluorescence vesicles in the cytoplasm, which represent AVOs. Because CQ inhibits the late step of autophagy, the accumulation, and not inhibition, of AVOs was induced by the treatment with CQ.

Figure 4

CQ potentiates the accumulation of LC3-II induced by 5-FU. (A) The expression of LC3-II was quantified by Western blot in HT-29 cells treated with 5-FU alone (left) for 48 h or after 12-h pretreatment with CQ (right). Treatment of HT-29 cells with 5-FU alone resulted in increased accumulation of LC3-II time-dependently, and the effect was potentiated by CQ-pretreatment. The expression levels of LC3-II are expressed as the density measured by the Image J software, standardized by the density of β-actin. (B) The intracellular localization of LC3 was analyzed by immunofluorescence microscopy, by staining the cells with fluorescent antibodies to LC3. Control (untreated) and CQ-treatment cells exhibited a weak and diffuse cytoplasmic staining with LC3-associated green fluorescence, whereas those treated with 5-FU and CQ pre-treatment + 5-FU exhibited an evident punctuate green fluorescence pattern of LC3, which is a typical feature of LC3 distribution within autophagosomes (LC3-II).

Figure 5

CQ potentiates apoptosis of HT-29 cells induced by 5-FU. The population of annexin V⁺ apoptotic cells were evaluated by FCM using annexin V-FITC/PI staining in HT-29 cells after CQ-pretreatment at 80 μM for 12 hours, followed by 5-FU at 5 μM for 48 hours. The percentage of annexin V⁺ apoptotic cells increased by pretreatment with CQ followed by 5-FU (16.5% vs 19.2% for 5-FU alone and CQ+5-FU, respectively). Values were given as mean ± SD.

Figure 6

CQ potentiates the G0/G1 arrest induced by 5-FU. (A) The analysis of changes in cell cycle was quantified by flow-cytometry after PI staining in HT-29 cells treated without or with 5-FU for 24, 48 h after 12-h pretreatment without or with CQ. Values were given as mean ± SD. Treatment of HT-29 cells with 5-FU alone resulted in increased intra-S arrest, but the G1 arrest was potentiated by CQ-pretreatment. Furthermore, G2/M progression of HT-29 cells was blocked by the treatment with 5-FU alone, and it was potentiated by the 12-h pretreatment of CQ (*, p < 0.05 vs. control). (B) The colony formation rate was quantified in HT-29 cells treated without or with 5-FU for 48 h after 12-h pretreatment without or with CQ. Values were given as mean ± SD. Treatment of cells with 5-FU alone resulted in a significant delay in the colony-forming ability, but at day 11 of culture, approximately 90% of the cells have formed colonies, suggestive that these cells were in a dormant state. Pre-treatment of cells with CQ prior to exposure to 5-FU resulted in potentiation of the inhibitory effect on the colony forming ability, which was reduced to approximately 35% of control untreated cells at day 11. CQ alone partially inhibited the colony forming ability of HT-29, but it was almost completely recovered by day 11.

Figure 7

Changes in the expression of cyclins, CDKs, CDKs inhibitors. Changes in the expression of cyclins, CDKs, CDKs inhibitors were quantified by Western blot in HT-29 cells treated with 5-FU for 48 h after 12-h pretreatment with CQ. Treatment of HT-29 cells with 5-FU alone resulted in decreased expression of cyclin D1, p21^Cip1 and p27^Kip1, time-dependently. In addition, 5-FU treatment resulted in accumulation of CDK2. Pre-treatment of cells with CQ resulted in inhibition of 5-FU-induced down-regulation of p21^Cip1 and p27^Kip1 expressions and, on the other hand, decreased the expression of CDK2. The expressions of cyclin E, CDK4 and CDK6 were not affected by 5-FU and/or CQ treatments. The expression levels of cyclins, CDKs, and CDKs inhibitors are expressed as the density measured by the Image J software, standardized by the density of β-actin.