MicroRNA-145 exerts tumor-suppressive and chemo-resistance lowering effects by targeting CD44 in gastric cancer

Abstract

Aim: To determine the potential roles of CD4 and microRNA (miR)-145 in gastric cancer.

Methods: The levels of CD44 and miR-145 were determined in gastric cancer cells. Quantitative real-time polymerase chain reaction was used to measure to the level of CD44 mRNA. A luciferase reporter assay and western blotting were performed to examine the effect of miR-145 on CD44 expression. Tumor sphere and MTT assays were carried out to evaluate the self-renewal and chemo-resistance properties of gastric cancer cells.

Results: The expression of CD44 was greatly increased and miR-145 was decreased in gastric cancer cells that were highly enriched in cancer stem cells (CSCs). The results demonstrated that miR-145 regulated CD44 by targeting directly the CD44 3'-untranslated region (3'-UTR). In gastric cancer cells, overexpression of miR-145 repressed the activity of the CD44 3'-UTR, and disruption of miR-145/CD44 3'-UTR interactions abrogated the silencing effects. In addition, miR-145 inhibition stimulated CD44 3'-UTR activity and disruption of miR-145/CD44 3'-UTR interactions abrogated this stimulatory effect. Enforced CD44 expression greatly increased tumor sphere formation and chemo-resistance in gastric cancer cells. Furthermore, the inhibition of CSCs and the chemo-sensitivity of gastric cancer cells treated with miR-145 were significantly abrogated by overexpression of CD44.

Conclusion: miR-145 targeting of CD44 plays critical roles in the regulation of tumor growth and chemo-resistance in gastric cancer.

Keywords: CD44; Chemo-resistance; Gastric cancer stem cells; miR-145.

Figure 1

miR-145 and CD44 expression in gastric cancer cells with self-renewal properties. ^aP < 0.05, ^bP < 0.01, ^eP < 0.001 vs the monolayer cells; data are the mean ± SEM of at least three independent experiments. A: Representative image of a tumor sphere of MGC-803 cells. MGC-803 cells were cultured in stem cell medium as described in the Materials and Methods section; B: miR-145 expression in tumor spheres and monolayer cells. miR-145 expression was determined by quantitative real-time polymerase chain reaction (qPCR); C: CD44 mRNA expression in tumor spheres and monolayer cells. CD44 mRNA expression was determined by qPCR; D: CD44 protein expression in tumor spheres and monolayer cells. Cells were harvested for western blotting analysis; E: The expression of several gastric cancer stem cell markers in tumor spheres and monolayer cells. Sox2, Oct4, and Nanog mRNA expression was determined by real-time PCR.

Figure 2

miR-145 targets the CD44 3’UTR directly in gastric cancer cells. ^aP < 0.05, ^bP < 0.01, ^eP < 0.001 vs the control; data are the mean ± SEM of at least three independent experiments. A: A putative miRNA-recognition element (MRE) for miR-145 on the 3’UTR of CD44; B: miR-145 regulated CD44 3’UTR activity negatively; C: MRE site-mutation abolished the effects of miR-145 on CD44 3’UTR activity. MGC-803 cells were co-transfected with pMT-CD44-3’ UTR-luc or pWT-CD44-3’UTR-luc with or without miR-145 mimics or an miR-145 inhibitor, respectively, and the transfected cells were harvested 36 h later for luciferase reporter assays as described; D: miR-145 mimics inhibited CD44 protein expression; E: The miR-145 inhibitor increased CD44 protein expression. MGC-803 cells were transfected with or without miR-145 mimics or a miR-145 inhibitor, respectively, and the transfected cells were harvested 48 h later for western blotting analysis. RLU: Relative luciferase activity.

Figure 3

Overexpression of CD44 abolished the inhibitory effect of miR-145 on tumor sphere formation in gastric cancer cells. ^eP < 0.001 vs cells transfected with the control; ^bP < 0.01 vs cells transfected with miR-145 mimics; data are the mean ± SEM of at least three independent experiments. A: Transfection with plasmid pLV-CD44 increased CD44 protein expression significantly; B: CD44 overexpression abolished the inhibitory effect of miR-145 on tumor sphere formation. MGC-803 cells were co-transfected with pLV-CD44 with or without miR-145 mimics, respectively, and the transfected cells were collected 12 h later for tumor sphere formation assays as described.

Figure 4

Overexpression of CD44 abolishes the chemo-resistance lowering effect of miR-145 in gastric cancer cells. ^aP < 0.05, ^bP < 0.01 vs cells transfected with the control; ^eP < 0.001 vs cells transfected with miR-145 mimics; data are the mean ± SEM of at least three independent experiments. A: CD44 overexpression reduced MGC-803 cell chemo-resistance to cisplatin; B: CD44 overexpression reduced MGC-803 cell chemo-resistance to 5-FU. MGC-803 cells were co-transfected with pLV-CD44 with or without miR-145 mimics, respectively, the transfected cells were collected 12 h later, and then various concentrations of cisplatin or 5-FU were used to treat the cells. Cell viability was determined as described; C: ABCG2 expression following transfection of miR-145 or pLV-CD44. MGC-803 cells were co-transfected with pLV-CD44 with or without miR-145 mimics, respectively, the transfected cells were collected 36 h later, and ABCG2 mRNA expression was determined by quantitative real-time polymerase chain reaction.