Targeting miR-148b-5p Inhibits Immunity Microenvironment and Gastric Cancer Progression

Abstract

Background: MicroRNAs (miRNAs) have been discovered to dictate the development of various tumors. However, studies on the roles of miRNAs in the progression of gastric cancer (GC) are still lacking.

Methods: Herein, by analyzing GC cell lines and patients samples, we observed that miR-148b-5p was significantly downregulated in GC. We also confirmed that miR-148b-5p overexpression significantly inhibited GC cell proliferation and invasion in vitro and in vivo.

Results: Overexpression of miR-148b-5p not only reprogrammed the metabolic properties of GC but also regulated the immune microenvironment by shifting lymphocyte and myeloid populations. Mechanistically, ATPIF1, an important glycolysis-associated gene, was identified as a direct target of miR-148b-5p and mediated the effect of miR-148b-5p. Notably, the low level of miR-148b-5p was significantly related with poor prognosis of GC patients (P < 0.001). Importantly, the levels of miR-148b-5p significantly changed the sensitivity of GC cells to several anti-cancer drugs (Doxorubicin, P < 0.05, Paclitaxel, P < 0.01, Docetaxel, P < 0.05).

Conclusions: Targeting miR-148b-5p inhibits immunity microenvironment and gastric cancer progression.

Keywords: ATPIF1; gastric cancer; immune microenvironment; metabolic reprogramming; miR-148b-5p.

Figure 1

The level of miR-148-5p is significantly reduced in GC and tightly associated with poor prognosis. (A) The qRT-PCR analysis to measure the expression levels of miR-148b-5p in normal gastric epithelial cells and four GC cell lines. The result was normalized to the U6 expression. (B) The expression levels of miR-148b-5p in 12 pairs of GC patients. (C) The expression levels of miR-148b-5p in non-metastatic (N = 20) and metastatic tissues of GCs (N = 20) were analyzed by qRT-PCR. (D) Assays of Kaplan–Meier survival of 46 GC patients indicated that the lower expression of miR-148b-5p was associated with poor patient prognosis. *P < 0.05, ***P < 0.001.

Figure 2

MiR-148b-5p reprograms metabolic pathways and inhibits GC development. (A) CellTiter -Glo methods and reagents were used to examine the effects of miR-148b-5p on GC cell growth. (B) The effect of miR-148b-5p on 3D spheroid growth at day 10. (C) The effect of miR-148b-5p on GC cell migration at 24 h. (D) The effect of miR-148b-5p on GC by injecting the indicated cells into the fat pads of nude mice (n = 10/group). (E) The representative images of H&E staining and Ki67 IHC from tumors. (F) H&E staining indicates liver and lung metastatic nodules from GC. (G) OCR assay was performed as indicated. (H) ECAR assay was performed as indicated. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 3

ATPIF1 is identified as a downstream target of miR-148b-5p in GC cells. (A) Target gene of miR-148b-5p by predicted using three online databases. (B) One predicted binding motif of miR-148b-5p on the 3′-UTR of ATPIF1 was presented. (C) Western blot presented ATPIF1 protein expression levels in five GC cell lines (AGS, SGC-7901, MKN74, MKN45, and SNU-16) and four normal human gastric cells. (D) Western blot data indicated the effect of transient reconstitution of miR-148b-5p using a mimic (0–40 pmol) on the expression of ATPIF1 in SCG-7901 cells. The effects of reconstitution of miR-148b-5p on the expression of ATPIF1 in AGS cells (E), MKN74 cells (F), MKN45 cells (G), and mice GC organoids (H). The effects of of miR-148b-5p on WT or mutant ATPIF1-3′-UTR luciferase reporter in AGS (I) or MKN45 cells (J–L) The effects of miR-148b-5p on ATPIF1 mRNA gene expression level using qRT-PCR analysis in AGS and MKN45 cells. ***P < 0.001; n. s., no statistical difference.

Figure 4

Restoration of ATPIF1 significantly rescues the effects of miR-148b-5p on GC progression. (A, B) The levels of ATPIF1 mRNA and protein after siRNA treatment. Three days after transfection, GC cell proliferation (C), 3D tumor growth (D) and migration assay (E) were examined respectively. (F) Western blots of ATIF1 levels were performed. And the proliferation (G) and migration assays (H) were conducted. (I) The effects of Paclitaxel, (100 nmol/L), Doxorubicin (0.3 mmol/L), and Docetaxel (10 nmol/L) on the cell viability of GC cells were examined after hours. The relative growth rate or viability after drug treatment was shown. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.001; n. s., no statistical difference.

Figure 5

MiR-148b-5p overexpression reprograms GC immune microenvironment. (A) Representative spleen pictures were presented and weight was measured from BALB/c nude mice xenograft bearing indicated GC cells. (B) Spleens tissues were stained by using CD11b antibody and its percentage was shown. (C) The levels of TNFa, IL-6, and CSF1 mRNA were analyzed using qRT-PCR in GC tissues of mice bearing mir-148b-5p overexpressing-MKN45 cells or not. (D) Secreted TNFα, IL-6, and CSF1 in GC cells were measured. (E) The levels of TNFα, IL-6, and CSF1 mRNA were analyzed using qRT-PCR in mice bearing mir-148b-5p overexpressing-AGS cells or not. (F) IHC staining were performed with TNFα and CD11b antibodies in GC tissues of mice bearing miR-148b-5p overexpressing-MKN45 cells or not. (G) Profiling of immune cell populations in in GC tissues of mice bearing miR-148b-5p overexpressing-MKN45 cells or not were performed using FACS analysis. *P < 0.05; **P < 0.01; ***P < 0.001, ****P < 0.001.