TGF-β induces HLA-G expression through inhibiting miR-152 in gastric cancer cells

Abstract

Background: Mounting evidences have showed the important role of transforming growth factor-β (TGF-β) in immunological surveillance of tumors. Some studies have also indicated human leukocyte antigen (HLA)-G-associated immune escape involving TGF-β management in gastric cancer (GC). However, the mechanism underlying it is unclear. This study aims to verify the correlations between HLA-G and TGF-β, involving the potential targeting of miR-152 on HLA-G.

Results: TGF-β and HLA-G levels were analyzed in blood samples from twenty GC patients with ELISA assays, while TGF-β showed directly proportional to HLA-G levels in GC patients, and TGF-β induced HLA-G up-regulation was also confirmed in GC cell lines. Furthermore, miR-152 expression could be inhibited by TGF-β, and the negative post-transcriptionally regulation of miR-152 on HLA-G was also demonstrated through gain- and loss-of-function studies. Besides, miR-152 overexpression repressed HLA-G up-regulation induced by TGF-β. And, miR-152 expression levels showed inversely proportional to both HLA-G and also TGF-β levels in GC patients.

Conclusion: TGF-β could induce HLA-G expression in GC by inhibiting miR-152, involving its negative regulation on HLA-G. Since TGF-β induced HLA-G up-regulation plays important role in immune escape, a potential application of miR-152 was suggested in GC treatment, or miR-152 might be one potential biomarker for GC.

Fig. 1

Correlation between TGF-β and HLA-G in GC. a. The concentration of TGF-β showed directly proportional to the HLA-G levels from the peripheral blood of twenty GC patients. R² = 0.5455; p < 0.001. b. After treated with TGF-β (2.5, 5 or 10 ng/ml) for 24 h, HLA-G mRNA level was significantly upregulated in a dosage dependent manner in GC cell lines like BGC823 and SGC7901. c. ELISA analysis showed increased HLA-G level in the same manner after TGF-β treatment for 48 h. *P < 0.05; **P < 0.01

Fig. 2

TGF-β inhibited miR-152 levels in GC cell line. The expression of miR-152 was downregulated under TGF-β treatment in dosage (2.5, 5 or 10 ng/ml for 12 h; a) and time (5 ng/ml for 6, 10 or 24 h; b) dependent manner. *P < 0.05; **P < 0.01

Fig. 3

MiR-152 regulated HLA-G expressions in BGC823 and SGC7901 cells. a. The binding sites of miR-152 on the wild-type or mutated 3’UTR sequences of HLA-G, wherein “CTG” and “ACT” were mutated to “AAA” and “GGA” respectively in Mut-HLA-G. b. The expression change of miR-152 in BGC823 and SGC7901 cells after transfected with miR-152 mimic or inhibitor for 12 h. c. HLA-G mRNA expression in miR-152 overexpressing or inhibiting cells after transfection for 24 h. d. Primers used for the construction of HLA-G luciferase reporter (PGL3-HLA-G) within HLA-G 3'-UTR sequences. Mutation sites within Mut-HLA-G sequences were indicated with rectangle boxes. e. HLA-G 3’-UTR activity change resulted from miR-152 dysregulation through luciferase assay. The results were normalized to those of Renilla luciferase activities, and statistically analyzed as compared to that from the cotransfection with negative miRNA. f. Detection for Mut-HLA-G 3’-UTR activity change. g. HLA-G level detection after miR-152 mimics or inhibitor transfection for 48 h. *P < 0.05; **P < 0.01

Fig. 4

TGF-beta induced HLA-G expression by inhibiting miR-152. BGC823 and SGC7901 cells were transfected with miR-152 mimics for 12 h, then incubated for 12 h (for qPCR analysis) or 24 h (for ELISA analysis) and finally treated with TGF-β (5 ng/ml). TGF-β induced increase of mRNA level (a) or concentration (b) of HLA-G was attenuated by miR-152 overexpression. *P < 0.05; **P < 0.01. c. The expression of miR-152 in GC tissues was inversely proportional to the HLA-G levels of GC patients. R² = 0.5267; p < 0.001. D. MiR-152 expression levels also showed inversely proportional to the concentration of TGF-β. R² = 0.5944; p < 0.001