The RNA22 tool¹ was employed to screen for lncRNAs that have a potential binding relationship with miR-210-3p. Data on the expression of DLEU2L in pancreatic tumor tissues and normal adjacent tissues were obtained from the GEPIA web site² using clinical samples in The Cancer Genome Atlas database.

The human pancreatic cell line PANC-1 was acquired from the cell bank of the Chinese Academy of Sciences (Shanghai, China) and cultured in Dulbecco’s modified Eagle medium (DMEM, SH30022.01B, Hyclone, Carlsbad, CA) containing 10% fetal bovine serum (FBS, 10270-106, Gibco, Waltham, MA) in an atmosphere containing 5% CO₂ at 37°C. For transfection, DLEU2L overexpression and miR-210-3p interference vectors were prepared using sequences synthesized by Wuhan Tianyi Huiyuan Bioscience and Technology Inc. (Wuhan, China). The overexpression and interference sequences are shown in Table 1. PANC-1 cells were transfected with either DLEU2L overexpression (ov-DLEU2L) or miR-210-3p interference (sh-miR-210-3p) fragments or the corresponding negative controls (ov-NC or sh-NC). After transfection, PANC-1 cells were treated with GEM (S1714, Selleck Chemicals, Houston, TX) dissolved in dimethyl sulfoxide at 0, 0.5, 1.5, or 15 μM for 48 h for western blot; 24, 48, or 72 for 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay; or 24 h for flow cytometry.

RNA22 (see text footnote 1) and starBase V2.0³ were employed to screen for potential target lncRNAs and genes of miR-210-3p, among which DLEU2L and breast cancer type 2 susceptibility protein (BRCA2) were selected, respectively. Binding sites between miR-210-3p and DLEU2L or the 3′ untranslated region (UTR) of BRCA2, were identified, and wild-type (WT) and mutant (MUT) 3′UTR segments of the binding sites were prepared based on the predicted sequences. Transfection was performed with pmirGLO plasmids (Addgene, Watertown, MA) using the LucPair^TM Duo-Luciferase Assay Kit (LF001, GeneCopoeia, Rockville, MD). After transfected cells were cultured with miR-210-3p mimics or negative control (NC), the signal of firefly luciferase and renilla luciferase was detected using a Glo-MAX 20/20 analyzer.

To evaluate the migration and invasion capability of PANC-1 cells after transfection, Transwell assays were carried out. Transwell chambers (Corning Inc., Corning, NY) and 24-well plates were first soaked with 1 × phosphate-buffered saline (PBS) for 5 min. For invasion assays, the top Transwell chambers were pre-coated with 80 μL of Matrigel (354230, BD Biosciences, Franklin Lakes, NJ) for 30 min at 37°C (chambers were non-coated for migration assay). Before the experiment, transfected cells were incubated in serum-free DMEM for 24 h. The cells were then trypsinized and suspended at 1 × 10⁵ cells/mL in DMEM containing 1% FBS. In the lower Transwell chambers, 750 μL of DMEM containing 10% FBS was added in each well of a 24-well plate, and 500 μL of cells were seeded on the top chamber. The plate was incubated at 37°C for 24 h, and 1 mL of 4% formaldehyde solution was added to each well for 10 min of fixation. After the solution was removed, the cells were washed once with 1 × PBS and 1 mL of 0.5% crystal violet solution (C1701, Bioswamp, Wuhan, China) was added to each well. The cells were stained for 30 min and washed three times with 1 × PBS. After drying, the non-migrated cells were removed using a cotton swab and the cells were visualized using an optical microscope for cell counting.

Total proteins were extracted from cell or tissue samples using radioimmunoprecipitation assay buffer containing protease and phosphatase inhibitors and quantified using a bicinchoninic acid assay. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed using 20 μg of denatured proteins in each lane, after which the proteins were transferred to polyvinylidene fluoride membranes. Blocking was performed by incubating the membranes in 5% skimmed milk for 2 h at room temperature. After blocking, the membranes were immersed in diluted primary antibodies (Table 2) overnight at 4°C and washed three times with PBS/Tween 20 (PBST) for 5 min each. Then, the membranes were incubated in horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature and rinse three times with PBST for 5 min each. An enhanced chemiluminescence reagent (WBKLS0010, Millipore, Burlington, MA) was added for color detection using a Tanon-5200 automatic analyzer (Tanon, Shanghai, China), and the protein bands were visualized and analyzed using Tanon GIS software.

Adenosine triphosphate (ATP) content (product code: A095) and lactic acid production (product code: A019-2) were evaluated using respective assay kits provided by the Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Glucose uptake was assessed using an assay kit (product code: KA4086) provided by Abnova (Taipei, Taiwan). All required reagents were provided in the kits and the assays were performed according to the manufacturer’s instructions.

Transfected PANC-1 cells were seeded in a 96-well plate at 5 × 10³ cells/well (180 μL/well). The plate was incubated overnight at 37°C in an atmosphere containing 5% CO₂ for the cells to adhere. The cells were then treated with GEM (0, 0.5, 1.5, or 15 μM) for 0, 24, 48, or 72 h. After GEM treatment, 20 μL of 5 mg/mL MTT solution (PAB180013, Bioswamp) was added to each well and the cells were further incubated for 4 h. Thereafter, the solution in the wells was aspirated and 150 μL of dimethyl sulfoxide was added to each well. The plate was gently shaken for 10 min and the absorbance of the wells was measured at 490 nm using a plate reader to evaluate relative cell proliferation.

Transfected and/or GEM-treated (24 h) PANC-1 cells were washed once with PBS and trypsinized. The cells were centrifuged at 1,000 × g for 5 min, the supernatant was removed, and the cells were resuspended in PBS. To assess apoptosis, 1 × 10⁶ cells were centrifuged at 1,000 × g at 4°C for 5 min. The supernatant was removed and 1 mL of cold PBS was added to resuspend the cells gently. This step was performed three times in total. After the final centrifugation, the cells were resuspended in 200 μL of binding buffer, along with 10 μL of annexin V-fluorescein isothiocyanate solution and 10 μL of propidium iodide, both provided in an apoptosis assay kit (556547, BD Biosciences). The mixture was gently mixed and incubated at 4°C in the dark for 30 min, and 300 μL of binding buffer was added. Flow cytometry was carried out and the data were analyzed using NovoExpress software (NovoCyte, ACEA Biosciences, Inc., San Diego, CA). To examine cell cycle progression, the suspended cells were fixed in 700 μL of anhydrous ethanol at −20°C for 24 h. The fixed sample was centrifuged at 3,000 × g for 30 s and the supernatant was removed. Then the precipitated cells washed twice with 1 mL of cold PBS. The precipitate was resuspended in 100 μL of 1 mg/mL RNAse A and the cells were incubated at 37°C to allow RNA to degrade, and 400 μL of 50 μg/mL propidium iodide was added. After 10 min of incubation in the absence of light, DNA content was determined using flow cytometry and the proportion of cells at each phase of the cell cycle was evaluated using NovoExpress software.

All animal experiments were performed in accordance with the Guidelines for Animal Care and Use of the Model Animal Research Institute and was approved by the ethics committee of Wuhan Myhalic Biotechnology Co., Ltd. (approval number HLK-20190225-01). The in vivo model of xenografted pancreatic cancer was established in thirty male BALB/c nude mice aged 4 weeks and weighing 18–20 g (obtained from Cavens Laboratories, Changzhou, China). The mice were housed in specific-pathogen-free conditions (22–26°C, 50–60% relative humidity) in a 12/12-h light/dark cycle and given free access to food and water. After a 7-day initial adaptive period, the mice were randomly divided into five groups (n = 6 per group): Control (non-transfected), ov-DLEU2L (DLEU2L overexpression), ov-NC (empty overexpression vector as negative control), sh-miR-210-3p (miR-210-3p interference), and sh-NC (empty interference vector as negative control). Based on the respective groups, the mice were subjected to subcutaneous injection of non-transfected or transfected (DLEU2L overexpression or miR-210-3p interference vectors, or their respective negative controls) PANC-1 cells at 1 × 10⁶ cells/mL at the right armpit. After 21 days of tumor growth, the mice were sacrificed with an overdose of sodium pentobarbital and the tumors were removed for subsequent characterization.

RNA extraction was carried out using TRIzol reagent (15596026, Ambion, Inc., Foster City, CA). The extracted RNA was reverse-transcribed into cDNA using the RevertAid First Strand cDNA Synthesis Kit (K1622, Thermo Scientific, Waltham, MA) and TaqMan kit (Applied Biosystems, Foster City, CA). qRT-PCR was performed using the SYBR Green PCR kit (KM4101, KAPA Biosystems, Wilmington, MA). The primer sequences are listed in Table 3 and the experimental conditions are described in Table 4. Data were acquired using the QuantStudio^TM 6 Flex Real-Time PCR System (Applied Biosystems) and analyzed by the 2^–ΔΔCt method.

The extracted tumors were fixed in 10% neutral formalin for 2 days and embedded in paraffin wax. Before sectioning, the paraffin-embedded tissue block was hardened at −20°C for several minutes. The tissue block was cut into 2-μm-thick sections and fixed onto microscope slides. For immunohistochemical staining, the sections were heated for 1 h at 65°C and deparaffinized twice in xylene for 15 min each. The deparaffinized sections were soaked in a graded concentration series of ethanol (100% twice, 95, 85, and 75%) for 5 min at each concentration, then washed with running water for 10 min. Antigen retrieval was performed by immersing the sections in 0.01 M sodium citrate buffer for 15 min. After three washes with PBS for 3 min each, endogenous peroxidase removal was performed by immersing the sections in 3% H₂O₂ for 10 min. The sections were washed three times with PBS for 3 min each and blocked in 0.5 bovine serum albumin for 30 min. Then, the sections were incubated with Ki67 MaxVision^TM primary antibodies (PAB30684, Bioswamp) at 1:50 for 2 h at 37°C and washed three times with PBS for 3 min each. Thereafter, the sections were incubated with horseradish peroxidase-polymer anti-rabbit secondary antibodies (KIT-5020, MaxVision) at room temperature for 30 min and washed three times with PBS for 3 min each. Diaminobenzidine was added to the sections and when color change was observed, the sections were washed with running water to remove the dye. Hematoxylin counterstaining was performed for 3 min and differentiation was performed using 1% hydrochloric alcohol. The degree of staining was controlled under a microscope and the reaction was terminated by washing the sections under running water for 10 min. Then the sections were dehydrated in ethanol (75, 85, 95, and 100% twice) for 5 min at each concentration and transparentized by soaking twice in xylene for 3 min each. The sections were sealed with neutral balsam gum and analyzed using a microscope.

All numerical data are shown as the mean ± standard deviation (SD) of three replicates (n = 3) for in vitro experiments and six replicates (n = 6) for in vivo experiments. The one-sample t-test was performed to compare the means between two groups and one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test was conducted to compare the means between more than two groups. p < 0.05 indicates statistical significance.

Using the RNA22 tool (Miranda et al., 2006), we screened for lncRNAs that exhibited binding sites with the oncogenic miR-210 (specifically miR-210-3p). This step was performed to identify potential miRNA-lncRNA targeting relationships. We revealed that miR-210-3p, which contains 22 nucleotides, binds to DLEU2L at position 1547 (Figure 1A). Through this prediction, we hypothesized that DLEU2L inhibits miR-210-3p expression by binding at the target site. To verify this hypothesis, we designed WT and MUT sequences of the DLEU2L binding site based on the prediction and performed a dual luciferase assay to determine whether binding occurs between miR-210-3p and DLEU2L. As anticipated, overexpression of miR-210-3p using mimics suppressed luciferase activity in WT-transfected cells (Figure 1B), successfully validating the binding between miR-210-3p and DLEU2L. We confirmed the relationship between DLEU2L and miR-210-3p by transfecting PANC-1 cells with DLEU2L overexpression (ov-DLEU2L) or interference (sh-DLEU2L) vectors (or their respective negative control/NC) and detecting the expression of miR-210-3p by qRT-PCR (Figure 1C). The results are consistent with those of luciferase activity assay, showing that ov-DLEU2L significantly downregulated the expression of miR-210-3p, whereas sh-DLEU2L significantly upregulated it. In other words, the targeting relationship was verified. Based on this result and our previous findings (Yang et al., 2019), we next assumed that DLEU2L plays a tumor-suppressing role, particularly in pancreatic cancer, as a target of (and in turn, by targeting) miR-210-3p. We used the GEPIA web server (Tang et al., 2017) to collect data on the median expression of DLEU2L in a variety of tumor types (Figures 1D,E). In clinical pancreatic adenocarcinoma (PAAD in Figure 1D) samples in The Cancer Genome Atlas database, DLEU2L was more highly expressed in adjacent normal tissues than in tumor tissues (the relative expression levels of DLEU2L and miR-210-3p in various tumor and normal tissues, detected in a pre-study, are shown in Supplementary Figure 1). Thus, we hypothesized that by overexpressing DLEU2L, miR-210-3p is conducive to over-binding with DLEU2L, thereby resulting in the suppression of its expression and the nullification or attenuation of its oncogenic function. The subsequent studies were designed to verify this hypothesis.

Based on the results of bioinformatics analysis, we investigated whether overexpression of DLEU2L or silencing of miR-210-3p affected the biological behavior of PANC-1 pancreatic cancer cells. We first looked at the migration and invasion capability of PANC-1 cells after transfection. Upon DLEU2L overexpression (ov-DLEU2L) or miR-210-3p silencing using shRNA (sh-miR-210-3p), the number of migrating (Figure 2A) and invading (Figure 2B) cells was decreased significantly, suggesting that the transfections affected cell motility. We then examined whether glucose metabolism (the Warburg effect in tumor behavior) is influenced by transfection. The expression of glucose transporter 1 (GLUT1), lactate dehydrogenase B (LDHB), hexokinase 2 (HK2), and pyruvate kinase isozyme M2 (PKM2), was remarkably downregulated by ov-DLEU2L or sh-miR-210-3p (Figure 2C), suggesting alterations in glucose metabolism. In addition, changes in energy metabolism were assessed by evaluating ATP levels, lactic acid production, and glucose uptake. All of these were suppressed in PANC-1 cells transfected with ov-DLEU2L or sh-miR-210-3p. In all cases, the corresponding empty vectors (negative control, NC) had no effect, and the effect of ov-DLEU2L was similar to that of sh-miR-210-3p (Figure 2D).

We next looked at the proliferation and apoptosis of transfected PANC-1 cells in the presence of GEM at various concentrations. We first performed MTT assays for up to 72 h to determine the effect of transfection and GEM administration on the growth and survival of PANC-1 cells. As anticipated, in both non-transfected and transfected (ov-DLEU2L or sh-miR-210-3p) cells, GEM inhibited cell proliferation in a concentration-dependent manner (Figure 3A). Looking at each GEM concentration separately, we note that transfection with ov-DLEU2L or miR-210-3p both suppressed PANC-1 cell proliferation compared to non-transfection (Figure 3B). These findings are consistent with the results of Transwell invasion assay. Flow cytometry of cell cycle progression (Supplementary Figure 2) revealed that in control (non-transfected) PANC-1 cells, the proportion of cells in the G0/G1 phase increased in a GEM concentration-dependent manner. Upon DLEU2L overexpression (ov-DLEU2L) or miR-210-3p silencing using shRNA (sh-miR-210-3p), the proportion of cells at the G0/G1 phase at each GEM concentration was decreased compared to that in control cells. Concurrently, the percentage of cells in the S phase was comparatively increased, which is indicative of S phase arrest and may signify the occurrence of DNA damage via double-strand breaks (Ye et al., 2003). In addition, flow cytometry was performed to evaluate the effect of transfection on PANC-1 cell apoptosis. While the percentage of apoptosis increased significantly in a GEM concentration-dependent manner, apoptosis at each concentration was further accentuated by ov-DLEU2L or sh-miR-210-3p (Figure 3C). This finding complements the results of cell cycle progression, implicating that the overexpression of DLEU2L and silencing of miR-210-3p effectively induced PANC-1 cell death, with more prominent effects at higher GEM concentrations.

The results of flow cytometry were verified by western blot detection of apoptosis-related proteins in transfected PANC-1 cells (ov-DLEU2L or sh-miR-210-3p) treated with GEM at various concentrations (Figures 3D,E). The expression of the pro-apoptosis proteins Bax, Cyt-c, and cleaved caspase-3 (cl-Casp-3) showed a generally increasing trend with increasing concentrations of GEM. Concurrently, the expression of the anti-apoptosis protein Bcl-2 showed a concentration-dependent downward trend. Moreover, Bax, Cyt-c, and cl-Casp-3 were significantly upregulated by transfection with ov-DLEU2L or sh-miR-210-3p at all GEM concentrations, whereas Bcl-2 was significantly downregulated. In addition to revealing the concentration-dependent effect of GEM on PANC-1 cell apoptosis, the western blot observations also demonstrated the accentuating effect of DLEU2L overexpression and miR-210-3p interference on apoptosis, in the presence of GEM.

Our previous research has revealed that exosomal miR-210 derived from GEM-resistant pancreatic cancer cells mediated the horizontal transfer of drug-resistant traits via mTOR signaling (Yang et al., 2019). We thus proposed that the tumor-suppressing properties of DLEU2L may be associated with the inactivation of the AKT/mTOR signaling pathway to impair GEM resistance. To confirm this, we examined the activation of AKT/mTOR by detecting their phosphorylation, as well as the phosphorylation of eukaryotic initiation factor 4E-binding protein (4EBP1) and ribosomal protein S6 kinase (S6K), which are downstream effectors of mTOR (Hay and Sonenberg, 2004). We observed that the levels of phosphorylated AKT, mTOR, 4EBP1, and S6K, relative to the total protein amount, showed an overall GEM concentration-dependent decrease (Figure 4). Compared to non-transfected PANC-1 cells, ov-DLEU2L or sh-miR-210-3p transfection further attenuated the levels of phosphorylated AKT, mTOR, 4EBP1, and S6K at each GEM concentration. The phosphorylation level also showed a decreasing trend in a GEM concentration-dependent manner after transfection. Considering the previously analyzed results of apoptosis, these observations signify that DLEU2L overexpression or miR-210-3p interference promoted pancreatic cancer cell death by inhibiting AKT/mTOR activation, and the effects are more prominent at higher GEM concentrations.

Target genes of miR-210-3p that exhibit a possible correlation with pancreatic cancer growth and progression were screened using RNA22, as described previously. Among the target genes, BRCA2 was identified as a tumor-suppressor gene involved in DNA damage repair (Naderi and Couch, 2002). Mutations in the BRCA2 are known to be associated with increased risk of several cancer types, including pancreatic cancer (Breast Cancer Linkage, 1999; Hahn et al., 2003). Thus, proper function of BRCA2 is crucial for cancer prevention, and miR-210-3p may target BRCA2 to impair its function and promote cancer progression (Figure 5A). Binding sites between miR-210-3p and the 3′UTR of the BRCA2 gene were determined using RNA22 (Figure 5B). Based on the predicted binding sites, WT and MUT sequences of the 3′UTR of BRCA2 were designed and a dual luciferase assay was performed to verify the binding relationship between miR-210-3p and BRCA2 (Figure 5C). We observed that overexpression of miR-210-3p using mimics reduced the luciferase activity in WT-transfected cells, confirming that binding existed between miR-210-3p and the BRCA2 gene. We then confirmed that ov-DLEU2L and sh-miR-210-3p (Figures 5D,E) drastically upregulated the protein expression of BRCA2 in PANC-1 cells at all concentrations of GEM. This result is consistent with that of the dual luciferase assay. In addition, the expression of BRCA2 was increased in a GEM concentration-dependent manner in both ov-DLEU2L- and sh-miR-210-3p-transfected cells.

Finally, we explored the anti-tumorigenic effect of DLEU2L overexpression and miR-210-3p interference in vivo. A xenografted pancreatic tumor model was established in mice using PANC-1 cells, and the effect of DLEU2L overexpression or miR-210-3p interference on tumor growth was investigated. Prior to injection, PANC-1 cells were transfected with either ov-DLEU2L, sh-miR-210-3p, or their respective negative controls. After 21 days of tumor growth, we observed that tumors formed from PANC-1 cells transfected with ov-DLEU2L or sh-miR-210-3p were evidently smaller than those formed by control PANC-1 cells as well as those transfected with negative controls (Figure 6A). In particular, sh-miR-210-3p had the most prominent effect on slowing down tumor growth (Figure 6B). Ki67 staining of tumor cell proliferation revealed large areas of brown deposition in the Control, ov-NC, and sh-NC groups, which were indicative of active tumor cell proliferation. However, ov-DLEU2L and sh-miR-210-3p induced a remarkable decrease in the distribution and intensity of brown staining, implicating that tumor cell proliferation was inhibited (Figures 6C,D). We measured the expression of DLEU2L and miR-210-3p in tumor tissues (Figures 6E,F). As anticipated, ov-DLEU2L and miR-210-3p significantly upregulated the expression of DLEU2L while downregulating that of miR-210-3p compared to the three control groups. Collectively, these findings indicate that DLEU2L overexpression or miR-210-3p interference effectively disrupted pancreatic tumor growth by suppressing tumor cell proliferation.