Long Non-coding RNA DLEU2L Targets miR-210-3p to Suppress Gemcitabine Resistance in Pancreatic Cancer Cells via BRCA2 Regulation

Abstract

Gemcitabine (GEM) resistance remains a challenging clinical issue to overcome in chemotherapy against pancreatic cancer. We previously demonstrated that miR-210 derived from pancreatic cancer stem cells enhanced the GEM-resistant properties of pancreatic cancer cells, thus identifying miR-210 as an oncogenic miRNA. Herein, we report the existence of an upstream effector that acts as a competing endogenous RNA (ceRNA) to miR-210. Bioinformatic screening was performed to identify lncRNAs with a binding relationship to miR-210. Overexpression and interference vectors were constructed to demonstrate the effect of ceRNA activity in pancreatic cell behavior, both in vitro and in vivo. DLEU2L (deleted in lymphocytic leukemia 2-like), which is expressed at low levels in pancreatic cancer tissues, was shown to exhibit a binding relationship with miR-210-3p. Overexpression of DLEU2L and silencing of miR-210-3p suppressed the proliferation, migration, and invasion of pancreatic cancer cells while promoting apoptosis. These effects occurred via the inhibition of the Warburg effect (aerobic glycolysis) and AKT/mTOR signaling. In addition, we showed that BRCA2 is a target gene of miR-210-3p, and the downregulation of miR-210-3p by DLEU2L effectively induced an upregulation of BRCA2 via the ceRNA mechanism. In vivo, DLEU2L overexpression and miR-210-3p interference suppressed pancreatic tumor progression, consistent with the results of in vitro studies. The findings of our study establish DLEU2L as a ceRNA to miR-210-3p and reveal the critical role of the DLEU2L/miR-210-3p crosstalk in targeting GEM resistance.

Keywords: Warburg effect; cancer metabolism; competing endogenous RNA; drug resistance; lncRNA; microRNA.

FIGURE 1

Targeting relationship between DLEU2L and miR-210-3p and expression of miR-210-3p in pancreatic cancer. (A) Binding sites between DLEU2L (from position 1,547) and miR-210-3p, predicted by RNA22. Dot-bracket notation illustrates binding at specific nucleotides. (B) Dual luciferase activity assay of binding relationship between DLEU2L and miR-210-3p, using miR-210-3p mimics or negative control (NC). Relative luciferase activity represents the ratio of firefly luciferase to renilla luciferase. (C) qRT-PCR of the expression of miR-210-3p in PANC-1 cells that were transfected with DLEU2L overexpression (ov-DLEU2L) or interference (sh-DLEU2L) vectors (or their respective negative control/NC). (D) Median expression of DLEU2L in various types of tumor tissues compared to that in adjacent normal tissues in clinical samples. The plot was generated using data from the GEPIA web site. (E) Relative expression of DLEU2L in PAAD tumor samples and adjacent normal tissue samples. The plot was generated using data from the GEPIA web site. PAAD, pancreatic adenocarcinoma. Numerical data in (B,C) are expressed as the mean ± SD (n = 3); ^∗p < 0.05. Statistical analysis was carried out using ANOVA.

FIGURE 2

Effect of DLEU2L overexpression and miR-210-3p silencing on PANC-1 cell migration/invasion and glucose metabolism. Transwell assay and quantification of the number of (A) migrating and (B) invading PANC-1 cells, transfected with DLEU2L overexpression (ov-DLEU2L) or miR-210-3p interference (sh-miR-210-3p) vectors (or the respective negative controls), after 24 h of culture. Scale bar, 100 μm. (C) Western blot and quantification of the expression of proteins associated with glucose metabolism (GLUT1, LDHB, HK2, and PKM2), in PANC-1 cells transfected with DLEU2L overexpression (ov-DLEU2L) or miR-210-3p interference (sh-miR-210-3p) vectors (or the respective negative controls). (D) ELISA of ATP production, lactic acid production, and glucose uptake in PANC-1 cells transfected with DLEU2L overexpression (ov-DLEU2L) or miR-210-3p interference (sh-miR-210-3p) vectors (or the respective negative controls). All numerical data are expressed as the mean ± SD (n = 3); ^∗p < 0.05. Statistical analysis was carried out using ANOVA.

FIGURE 3

Effect of DLEU2L overexpression and miR-210-3p silencing on PANC-1 cell proliferation and apoptosis in the presence of GEM. MTT assay of PANC-1 cell proliferation plotted based on (A) cell transfection and (B) GEM concentration. For MTT assay, PANC-1 cells were non-transfected (Control) or transfected with DLEU2L overexpression (ov-DLEU2L) or miR-210-3p interference (sh-miR-210-3p) vectors and cultured in the presence of GEM at 0, 0.5, 1.5, or 15 μM for 0, 24, 48, or 72 h. (C) Flow cytometry of apoptosis and quantification of the percentage of late apoptotic PANC-1 cells (upper right quadrant). Western blot and quantification of the expression of proteins associated with apoptosis (Bax, Bcl-2, Cyt-c, and cl-Casp-3), in PANC-1 cells transfected with (D) DLEU2L overexpression (ov-DLEU2L) or (E) miR-210-3p interference (sh-miR-210-3p) vectors. All numerical data are expressed as the mean ± SD (n = 3); ^∗p < 0.05; ^#p < 0.05 within the same group (comparing GEM concentrations); data with the same letters in (D) and (E) are not significantly different (p > 0.05) in each plot of western blot. Statistical analysis was carried out using ANOVA.

FIGURE 4

Effect of DLEU2L overexpression and miR-210-3p silencing on AKT/mTOR signaling in PANC-1 cells in the presence of GEM. Western blot and quantification of the expression of proteins associated with AKT/mTOR signaling (phosphorylation of AKT, mTOR, and S6K), in PANC-1 cells transfected with DLEU2L overexpression (ov-DLEU2L) or miR-210-3p interference (sh-miR-210-3p) vectors. Expression of phosphorylated protein was normalized to the respective total protein content. All numerical data are expressed as the mean ± SD (n = 3); ^∗p < 0.05; data with the same letters are not significantly different (p > 0.05) in each plot of western blot. Statistical analysis was carried out using ANOVA.

FIGURE 5

Involvement of BRCA in miR-210-3p sponging and effect on DNA damage repair. (A) Schematic of the proposed ceRNA sponging relationship between DLEU2L, miR-210-3p, and BRCA2 and effect on pancreatic tumor progression. (B) Binding sites between BRCA2 (from position 11532) and miR-210-3p, predicted by RNA22. Dot-bracket notation illustrates binding at specific nucleotides. (C) Dual luciferase activity assay of binding relationship between BRCA2 and miR-210-3p, using miR-210-3p mimics or negative control (NC). Relative luciferase activity represents the ratio of firefly luciferase to renilla luciferase. (D) Western blot of the expression of BRCA2 in PANC-1 cells transfected with DLEU2L overexpression (ov-DLEU2L) or miR-210-3p interference (sh-miR-210-3p) vectors. (E) Quantification of the band gray values in (D). All numerical data are expressed as the mean ± SD (n = 3); ^∗p < 0.05; data with the same letters are not significantly different (p > 0.05) in each plot of western blot. Statistical analysis was carried out using ANOVA.

FIGURE 6

In vivo model of pancreatic tumor incorporating DLEU2L overexpression or miR-210-3p interference. (A) Xenografted pancreatic tumors extracted from mice on day 28 of tumor growth. (B) Tumor volume during the 28-day tumor growth period. Tumor volume was calculated using the formula volume = long diameter × (short diameter)² × 0.5. (C) Ki-67 staining of tumor tissues. Brown staining indicates proliferating tumor cells. Scale bar, 50 μm. (D) Quantification of the intensity of Ki-67 staining. qRT-PCR of the expression of (E) DLEU2L and (F) miR-210-3p in tumor tissues. All numerical data are expressed as the mean ± SD (n = 3); ^∗p < 0.05. Statistical analysis was carried out using ANOVA.