ARHGAP25 Inhibits Pancreatic Adenocarcinoma Growth by Suppressing Glycolysis via AKT/mTOR Pathway

Abstract

Increasing evidence reveals that the Rho GTPase-activating protein is a crucial negative regulator of Rho family GTPase involved in tumorigenesis. The Rho GTPase-activating protein 25 (ARHGAP25) has been shown to specifically inactivate the Rho family GTPase Rac1, which plays an important role in pancreatic adenocarcinoma (PAAD) progression. Therefore, here we aimed to clarify the expression and functional role of ARHGAP25 in PAAD. The ARHGAP25 expression was lower in PAAD tissues than that in normal pancreatic tissues based on bioinformatics analysis and immunohistochemistry staining. Overexpression of ARHGAP25 inhibited cell growth of AsPC-1 human pancreatic cancer cells in vitro, while opposite results were observed in BxPC-3 human pancreatic cancer cells with ARHGAP25 knockdown. Consistently, in vivo tumorigenicity assays also confirmed that ARHGAP25 overexpression suppressed tumor growth. Mechanically, overexpression of ARHGAP25 inactivated AKT/mTOR signaling pathway by regulating Rac1/PAK1 signaling, which was in line with the results from the Gene set enrichment analysis on The Cancer Genome Atlas dataset. Furthermore, we found that ARHGAP25 reduced HIF-1α-mediated glycolysis in PAAD cells. Treatment with PF-04691502, a dual PI3K/mTOR inhibitor, hampered the increased cell growth and glycolysis due to ARHGAP25 knockdown in PAAD cells. Altogether, these results conclude that ARHGAP25 acts as a tumor suppressor by inhibiting the AKT/mTOR signaling pathway, which might provide a therapeutic target for PAAD.

Keywords: AKT/mTOR signaling; ARHGAP25; glycolysis; pancreatic adenocarcinoma; proliferation..

Figure 1

Rho GTPase activating protein 25 (ARHGAP25) expression was downregulated in pancreatic adenocarcinoma (PAAD). A, The gene expression of ARHGAP25 was reduced in PAAD tissues compared with that in normal tissues from the Cancer Genome Atlas (TCGA) database. B, ARHGAP25 gene expression was downregulated in PAAD tissues in GEO datasets (GSE15471 and GSE16515). C,D, mRNA expressions (n=3) (C) and protein levels (D) of ARHGAP25 were decreased in a panel of PAAD cell lines compared with that in Human pancreatic normal ductal epithelial cells (HPNE). E, The relative immunohistochemical (IHC) scores of ARHGAP25 was significantly lower in PAAD tissues than that of normal tissues. F, Representative IHC staining of ARHGAP25 in normal and tumor tissues. Scale bar: 200 μm. Data are presented as box plots with median and interquartile range (25% and 75%). *P < .05, **P < .01, ***P < .001.

Figure 2

Rho GTPase activating protein 25 (ARHGAP25) inhibited pancreatic adenocarcinoma (PAAD) proliferation and tumorigenicity. A,B, Verification of ARHGAP25 overexpression and knockdown efficiency in PAAD cells revealed that the mRNA expression of ARHGAP25 was significantly higher in ARHGAP25-OE AsPC-1 cells (A) and lower in sh-ARHGAP25 BxPC-3 cells (B) compared to that in control cells (n=3 per group). C, CCK-8 assays revealed that the proliferation of ARHGAP25-OE AsPC-1 cells was significantly reduced, while that of sh-ARHGAP25 BxPC-3 cells was significantly increased (n=3 per group). D, Tumor growth curves showed that xenografts generated from ARHGAP25-OE AsPC-1 cells grew slower than those from ARHGAP25-NC cells (n=6). Conversely, xenografts generated from sh-ARHGAP25 BxPC-3 cells had increased growth rates compared with those from sh-NC cells (n=6). E,F, The weights and volumes of tumors in the ARHGAP25-OE group were significantly lower than those in the ARHGAP25-NC group (n= 6). The weights and volumes of tumors in the sh-ARHGAP25 group were significantly higher than those in the sh-NC group (n=6). G, Representative images (left panel) and quantification (right panel) of immunofluorescence staining with PCNA in xenograft tumors stably transfected as indicated were shown, Scale bar: 50 μm. Data are shown as the mean ± SD (n ≥ 3). *P < .05, **P < .01, ***P < .001.

Figure 3

Rho GTPase activating protein 25 (ARHGAP25) attenuated cytotoxic effects of gemcitabine or 5-fluorouracil in pancreatic adenocarcinoma (PAAD) cells. A,B, ARHGAP25-overexpressing AsPC-1 cells were treated with gemcitabine (0.5µM) (A) and 5-fluorouracil (5 µM) (B) for 72 h. Upregulation of ARHGAP25 enhanced both gemcitabine and 5-fluorouracil cytotoxicity in AsPC-1 cells. C,D, ARHGAP25-knockdown BxPC-3 cells were treated with gemcitabine (5 µM) (C) and 5-fluorouracil (10 µM) (D) for 72 h. ARHGAP25 knockdown reduced the chemosensitizing effect of gemcitabine or 5-fluorouracil. The cell viability was measured by CCK-8 assay at 450nm. Data are shown as the mean ± SD. **P < .01, ***P < .001.

Figure 4

Rho GTPase activating protein 25 (ARHGAP25) inhibited AKT/mTOR signaling pathway. A, Gene set enrichment analysis (upper panel) showed the enrichment of AKT/mTOR signaling in PAAD tumors with downregulated ARHGAP25 expression. Differential expressions (lower panel) of the gene signature (CREIGHTON_AKT1_SIGNALING_VIA_MTOR_DN) are shown in the heatmap created by the GSEA software. B,C, Protein levels of ARHGAP25, total and phosphorylated AKT, and mTOR were examined by western blotting in PAAD cells. In sh-ARHGAP25 BxPC-3 cells, the expressions of p-AKT and p-mTOR were significantly increased (B). In ARHGAP25-overexpresisng AsPC-1 cells, the expressions of p-AKT and p-mTOR were significantly decreased (C). D,E, Western blotting was used to determine protein expression in xenografts. The expressions of p-AKT and p-mTOR were significantly decreased in the ARHGAP25-OE xenografts. (n=6) (D). Conversely, the expressions of p-AKT and p-mTOR were significantly increased in the xenografts with ARHGAP25 knockdown (n=6) (E). Data are shown as the mean ± SD. **P < .01, ***P < .001.

Figure 5

Rho GTPase activating protein 25 (ARHGAP25) suppressed glycolysis in PAAD in vitro. A, Protein levels of ARHGAP25, HIF-1α, PKM2, and LDHA were examined by western blotting in PAAD cells. In ARHGAP25-overexpresisng AsPC-1 cells, the expressions of HIF-1α, PKM2, and LDHA were significantly decreased. B, Glucose uptake was measured using the 2-NBDG uptake assay kit by flow cytometry. Overexpression of ARHGAP25 significantly decreased 2-NBDG uptake (n=3). C,D, Extracellular lactate levels and intracellular ATP levels were measured using the lactate assay kit and the ATP assay kit, respectively. Lactate production (C) and intracellular ATP levels (D) were significantly decreased in ARHGAP25-overexpresisng AsPC-1 cells (n=3 per group). E, In sh-ARHGAP25 BxPC-3 cells, the expressions of HIF-1α, PKM2, and LDHA were significantly increased. F,G, Lactate production (F) and intracellular ATP levels (G) were significantly increased in sh-ARHGAP25 BxPC-3 cells (n=3 per group). H, ARHGAP25 knockdown significantly increased 2-NBDG uptake (n=3). Data are shown as the mean ± SD. **P < .01, ***P < .001.

Figure 6

Rho GTPase activating protein 25 (ARHGAP25) suppressed glycolysis in PAAD in vivo. A,B, Western blotting was used to determine protein expression in xenografts. The expressions of HIF-1α, PKM2, and LDHA were significantly decreased in the ARHGAP25-OE xenografts (n=6). C,D, The expressions of HIF-1α, PKM2, and LDHA were significantly increased in the xenografts with ARHGAP25 knockdown (n=6). Data are shown as the mean ± SD. *P < .05, ***P < .001.

Figure 7

Silencing of Rho GTPase activating protein 25 (ARHGAP25) promoted glycolysis and proliferation by activating AKT/mTOR signaling in PAAD cells. A, Western blot analysis showed that PI3K and mTOR inhibitor PF-04691502 reduced activation of AKT/mTOR signaling induced by ARHGAP25 knockdown. B, Increased glucose uptake in sh-ARHGAP25 BxPC-3 cells was remarkably reduced upon PF-04691502 treatment (n=3). C,D, Upregulated lactate production (C) and intracellular ATP level (D) upon ARHGAP25 knockdown were abolished by PF-04691502 in BxPC-3 cells (n=3 per group). E, The effect of ARHGAP25 knockdown on promoting cell growth was remarkably abrogated by PF-04691502 (n=3). F, The schematic diagram showed the mechanism of ARHGAP25 regulating glycolysis and cell growth through AKT/mTOR signaling in PAAD.