LINC00960 regulates cell proliferation and glycolysis in pancreatic cancer through the miR‐326‐3p/TUFT1/AKT–mTOR axis

Abstract

Pancreatic cancer (PC) is a common malignant cancer characterized by high mortality and poor prognosis. LINC00690 was involved in the occurrence and progression of PC, but the underlying mechanisms require further investigation. The goal of this study was to figure out how LINC00960 mediates glycolysis in PC. LINC00960, miR‐326‐3p, and Tuftelin 1 (TUFT1) expression levels were detected in PC cell lines. LINC00960 and TUFT1 expression levels were increased in PC cells when compared with normal pancreatic cells, whereas miR‐326‐3p expression levels were decreased. The expression levels of LINC00690 affected glycolysis in PC, and inhibition of LINC00960 inhibited tumor growth in vivo. LINC00690 targeted and suppressed the expression of miR‐326‐3p. MiR‐326‐3p bound to TUFT1, and miR‐326‐3p inhibited AKT–mTOR pathway activation via TUFT1. In conclusion, the depletion of LINC00960 repressed cell proliferation and glycolysis in PC by mediating the miR‐326‐3p/TUFT1/AKT–mTOR axis. Thus, we present a novel mechanism underlying the progression of PC that suggests LINC00960 is a potential therapeutic target for this cancer.

INTRODUCTION

Pancreatic cancer (PC) is a common malignancy that ranks seventh in terms of deaths from cancer.¹ Surgery for PC remains a relatively effective therapeutic strategy.² However, >80% of patients are diagnosed with PC at an advanced stage, making them unsuitable for surgical treatment.³ Some molecules have become potential biomarkers for the diagnosis and prognosis of PC.^4–6 For example, the concentrations of serum biomolecules CEA, CA 19-9, and CA 242 have reliable sensitivity and specificity as biomarkers for the early diagnosis and prognosis of PC.⁴ It was reported that various long noncoding RNAs (lncRNAs) were aberrantly expressed in PC and played crucial roles in this disease.⁵ However, the underlying mechanisms need further enrichment.

LncRNAs are noncoding RNAs with >200 nucleotides; these have been extensively studied in the field of medicine, especially in cancer research.⁷ Indeed, they have been associated with the diagnosis and treatment of PC.^8,9 For instance, the lncRNA ZNFTR was found to act as an anticancer factor inhibiting the progression of PC.¹⁰ Another study revealed that LINC00960, a newly discovered lncRNA, was independently linked with the overall survival of patients with pancreatic ductal adenocarcinoma (PDAC).¹¹ Additionally, LINC00960 was upregulated in PDAC, mediating the growth of PDAC via the miR-146a-5p/IRAK1 axis.¹² LINC00960 could interact with various microRNAs (miRNAs) based on its sequence; thus, we questioned whether targeted miRNAs of LINC00960 are involved in the pathogenesis and progression of PC. Therefore, to further explore the molecular mechanism underlying the role of LINC00960 in PC, we sought to investigate the downstream genes and functions of LINC00960. Based on bioinformatics analysis, a binding site was found between miR-326-3p and LINC00960. In addition, low expression of miR-326-3p was detected in PC tissues and cells.¹³ However, the role of miR-326-3p in PC and whether it is regulated by LINC00960 are yet to be clarified, which is worth further explored.

The oncogene Tuftelin 1 (TUFT1) was reportedly overexpressed in PC and stimulated epithelial–mesenchymal transition.¹⁴ A research has revealed that TUFT1 was related to the mTOR signaling pathway, which was closely associated with glycolysis and involved in cancer development.¹⁵ As previously reported, the AKT pathway was also involved in glycolysis in breast cancer.¹⁶ Additionally, miRNAs can silence oncogenes and tumor suppressor genes at the posttranscription level and participate in cancer processes.¹⁷ Using bioinformatics analysis, the potential binding site between miR-326-3p and TUFT1 was predicted, and we speculated that miR-326-3p might participate in glycolysis in PC by regulating TUFT1, p-mTOR, and p-AKT expression.

Overall, the goal of this study was to determine the role of LINC00960 in PC and investigate the underlying mechanism by which LINC00960 is involved in the progression of this disease. This would improve the theoretical foundations of PC research and help in identifying biomarkers for this disease.

MATERIALS AND METHODS Cell culture

Normal pancreatic cells (HPNE and HPDE6c7) and PC cells (MiaPaCa-2, PANC-1, AsPC-1, CFPAC-1, Capan-1, and BxPC-3) were used in our experiments. AsPC-1, BxPC-3, and Capan-1 were purchased from ATCC (Manassas, Virginia). HPDE6c7, MiaPaCa-2, PANC-1, and CFPAC-1 were obtained from Cell Bank of Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). HPNE cells were cultured in F-12 K medium. MiaPaCa-2, PANC-1, CFPAC-1, and Capan-1 cells were maintained in Dulbecco's modified eagle medium (DMEM). AsPC-1 and BxPC-3 cells were maintained in RPMI-1640 medium. The aforementioned cell media (Thermo Fisher Scientific, Waltham, Massachusetts) were supplemented with 10% fetal bovine serum (Gibco, Carlsbad, California) and 1% penicillin and streptomycin (Beyotime Biotechnology, Shanghai, China). The cells were cultured at 37°C under 5% CO₂ condition. All cell lines were authenticated using short tandem repeat DNA profiling and routinely tested to confirm the absence of mycoplasma using the e-MycoTM VALiD Mycoplasma PCR Detection Kit (iNtRON, Burlington, Massachusetts).

RNA isolation, reverse transcription, and real-time quantitative polymerase chain reaction

TRIzol reagent (Invitrogen, California) was used to extract RNA from normal pancreatic cells, PC cells, and tumor tissues. Prime Script Reverse Transcription Reagent Kit (TaKaRa, Shiga, Japan) was used to synthesize cDNA, and SYBR Premix Ex Taq II Kit (Takara, Shiga, Japan) was used to perform real-time quantitative polymerase chain reaction (RT-qPCR). The qPCR reaction conditions were as follows: 95°C for 30 s and 40 cycles of 95°C for 10 s, 60°C for 20 s, and 72°C for 15 s. The primer sequences used herein were presented in Table 1. All data were analyzed using the 2^−ΔΔCt formula. GAPDH and U6 were used as reference genes.

TABLE 1 Primer sequences

Gene	Primer sequences
LINC00960 (F)	5′-CCTCTAAGCCTAAGCACCGCC-3′
LINC00960 (R)	5′-GGAAGCCTGGGCAAGGAATGG-3′
TUFT1 (F)	5′-AAAGGACGCCACCATCCAG-3′
TUFT1 (R)	5′-GTGCTGAAGTTGCCATGACTG-3′
GLUT1 (F)	5′-ACACTGGAGTCATCAATGCCC-3′
GLUT1 (R)	5′-CAGGATGCTCTCCCCATAGC-3′
LDHA (F)	5′-TACAGTTGTTGGGGTTGGTG-3′
LDHA (R)	5′-CACGTTACGCTGGACCAAAT-3′
miR-326-3p (F)	5′-CATCTGTCTGTTGGGCTGGA-3′
miR-326-3p (R)	5′-AGGAAGGGCCCAGAGGCG-3′
GAPDH (F)	5′-AACTTTGGCATTGTGGAAGG-3′
GAPDH (R)	5′-GGATGCAGGGATGATGTTCT-3′
U6 (F)	5′-CTCGCTTCGGCAGCACA -3′
U6 (R)	5′-AACGCTTCACGAATTTGCGT-3′

Western blotting

Radio immunoprecipitation assay (RIPA) lysis buffer (Beyotime Biotechnology) was used to extract proteins from normal pancreatic cells, PC cells, and tumor tissues. BCA Assay Kit (Beyotime Biotechnology) was used to quantify protein concentrations. The same amount of protein (30 μg/lane) was separated using 10% sodium dodecyl sulfate(SDS) polyacrylamide gel electrophoresis and transferred onto PVDF membranes. Next, skimmed milk (5%) was used to block the membranes for 1 h at 37°C. Subsequently, the polyvinylidene fluoride (PVDF) membranes were incubated with primary antibodies overnight at 4°C. Tris buffered saline with tween 20 (TBST) was used to wash the PVDF membranes five times, following which the membranes were soaked with horseradish peroxidase (HRP)-conjugated secondary antibodies (Beyotime Biotechnology) for 1 h. ECL Kit (Beyotime Biotechnology) was then used to visualize the protein bands. Densitometry analysis was conducted using Image J software (National Institutes of Health Bethesda, Maryland). The primary antibodies used, which were purchased from Thermo Fisher Scientific, were as follows: TUFT1 (no. PA5-55769, 1:5000; no. PA5-70773, 1:1000), AKT (no. 44-609G, 1:1000), p-AKT (no. MA5-38243, 1:1000), mTOR (no. PA5-34663, 1:2000), p-mTOR (no. 44-1125G, 1:1000), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (MA5-15738-D680, 1:1000).

Cell transfection

Sh-LINC00960, sh-KRAS, miR-326-3p mimics/inhibitor, overexpression of TUFT1, and their negative control groups (sh-NC, mimics NC, inhibitor NC, and pcDNA3.1) were obtained from GenePharma (Shanghai, China). PC cells were seeded onto 6-well plates, and when they reached 70% confluence, the corresponding sequences and plasmids were transfected into the cells using Lipofectamine 3000 (Invitrogen) following the manufacturer's protocol. After 48 h, the transfected cells were used in the experiments.

Cell counting kit-8 assay

Cell counting kit-8 (CCK-8) assay (Dojindo, Kumamoto, Japan) was used for detecting cell proliferation. The corresponding transfected cells were added to 96-well plates (4000 cells/well). At 1–5 days posttransfection, 10 μl of CCK-8 solution was added to the plates, followed by incubation for 2 h. The absorbance was then detected using a microplate reader at 450 nm (Thermo Fisher Scientific).

Clone formation assay

Cell proliferation was evaluated using clone formation assay. Briefly, PC cells with corresponding transfection were seeded onto 6-well plates at a low density (600 cells/well) and incubated at 37°C under 5% CO₂ condition. The medium was replaced every 48 h. After 2 weeks, 100% methanol was applied for 15 min to achieve cell fixation. Subsequently, the fixed cells were stained with 0.1% crystal violet (Sigma-Aldrich, St. Louis, Missouri). Finally, the cloned cells were assessed using Image J.

Lactate detection

Lactate was detected using Lactate Assay Kit II (Biovision, California). Briefly, the supernatant was collected for the detection of lactate production. According to manufacturer's protocol, the prepared reaction mixture was incubated for 30 min in darkness. A microplate reader was then used to measure the lactate levels at 450 nm.

Glucose uptake measurement

Glucose uptake was evaluated using Glucose Uptake Colorimetric Assay Kit (Biovision, California). The cells were seeded onto 96-well plates and incubated for 4 h. Subsequently, the phosphate buffered saline (PBS)-washed cells were incubated overnight in serum-free medium. Next, 100 μl of Krebs-Ringer-phosphate-4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer containing 2% bovine serum albumin (BSA) was used to preincubate the cells for 40 min, after which 10 μl of 10 mM 2-DG was used to incubate the cells for 20 min. The cells were then lysed with 90 μl of extraction buffer, frozen/thawed once, and then heated at 85°C for 40 min. The cell lysate was neutralized by adding 10 μl of neutralization buffer, and the cell supernatant was collected to examine glucose uptake. Finally, glucose uptake was determined using a microplate reader at 412 nm.

ATP measurement

ATP Colorimetric Assay Kit (Biovision) was used to detect ATP production. Briefly, cells treated with corresponding transfections were extracted using 100 μl of ATP Assay Buffer, and the supernatant was analyzed to determine ATP production at 570 nm using a microplate reader.

Dual-luciferase reporter assay

The sequences of LINC00960 or TUFT1 combined with miR-326-3p were cloned into psiChECK2 vectors (Ke Lei Biological Technology Co., Ltd, China) to construct the reporter vectors LINC00960 (WT/MUT) or TUFT1 (WT/MUT). MiR-326-3p mimics were transfected into PC cells coupled with LINC00960 (WT/MUT) or TUFT1 (WT/MUT) using Lipofectamine 3000. Luciferase activity was assessed using a dual-luciferase reporter assay system (Promega, Madison, Wisconsin).

RNA-binding protein immunoprecipitation

The interactions among LINC00960, miR-326-3p, and TUFT1 were assessed using RNA-binding protein immunoprecipitation (RIP) Assay Kit (Millipore, Bedford, Massachusetts). According to the manufacturer's instructions, ~2 × 10⁷ cells were collected and lysed using RIPA buffer. The cells were subjected to centrifugation, following which the supernatant was collected and incubated with magnetic beads conjugated with anti-Ago2 antibody (Abcam, no. ab32381) or antirat immunoglobin G (IgG) antibody (Abcam, no. ab109489) for immunoprecipitation. Finally, after the immunoprecipitated RNA was purified, the enrichment of the immunoprecipitated RNAs, including LINC00960, miR-326-3p, and TUFT1, was measured using RT-qPCR.

Tumor formation in nude mice

Animal experiments were approved by the ethics committee of the Affiliated Hospital of Qingdao University. Nude mice were randomly divided into two groups (five mice/group). After transfection with sh-NC or sh-LINC00960, 2 × 10⁶/ml PANC-1 cells were inoculated into the subcutaneous axilla of the left forelimb of nude mice. The tumor volume was then measured once per week for 4 consecutive weeks, following which the mice were euthanized. The two groups of tumor tissues were completely removed and used for immunohistochemistry (IHC), RT-qPCR, and western blotting.

Immunohistochemistry

IHC was performed to determine the protein expression of proliferating cell nuclear antigen (PCNA) and TUFT1 in tumor tissues. Briefly, the tumor tissues were embedded using 4% paraformaldehyde and paraffin. After the antigen repairing, the sections were blocked with 1% BSA for 20 min and incubated overnight at 4°C with antibodies against PCNA (Thermo Fisher Scientific; no. PA5-16797) and TUFT1 (Santa Cruz; no. sc-365632). The sections were then incubated with HRP-labeled antibodies for 30 min at room temperature. The slides were counterstained with hematoxylin, and images were acquired using a Nikon digital camera system combined with an Olympus microscope.

Statistical analysis

GraphPad Prism 6 was used to analyze data, which is presented as means ± SD. One-way analysis of variance was used to compare more than two groups, whereas Student's t-test was used to compare two groups. A p-value of <0.05 was considered statistically significant.

RESULTS LINC00960 and TUFT1 expression levels were increased and miR-326-3p expression levels were decreased in PC cells

First, we detected LINC00960, TUFT1, and miR-326-3p expression levels in six PC cell lines (MiaPaCa-2, PANC-1, AsPC-1, CFPAC-1, Capan-1, and BxPC-3). In comparison with normal pancreatic cells (HPNE and HPDE6c7), the RT-qPCR results revealed that LINC00960 and TUFT1 expression was increased in PC cells, whereas miR-326-3p expression was markedly decreased (Figures 1A–C and S4A–C), and these were not related to KRAS mutation (Figure S1). TUFT1 protein expression corresponded to the variation in TUFT1 mRNA expression (Figure 1D). Moreover, the proliferation of PC cells, detected using CCK-8 assay, was increased compared with that of HPNE cells (Figure 1E). Taken together, these findings suggested that the abnormal expression of LINC00960, TUFT1, and miR-326-3p in PC cells may be involved in PC progression.

FIGURE 1. LINC00960 and Tuftelin 1 (TUFT1) expression was increased but miR-326-3p expression was decreased in pancreatic cancer (PC) cells. (A–C) LINC00960, miR-326-3p, and TUFT1 expression were evaluated using real-time quantitative polymerase chain reaction in PC cells (MiaPaCa-2, PANC-1, AsPC-1, CFPAC-1, Capan-1, and BxPC-3) compared with normal pancreatic cells (HPNE). (D) TUFT1 protein levels were detected using western blotting in normal pancreatic cells and PC cells. (E) Cell proliferation was evaluated using cell counting kit-8 assay in normal pancreatic cells and PC cells. All of the data were obtained from at least three replicated experiments. The data were expressed as means ± SD. *p [less than] 0.05, **p [less than] 0.01, and ***p [less than] 0.001

Knocking down LINC00960 suppressed cell proliferation and glycolysis in PC

Based on the aforementioned results, the changes in the expression of LINC00960, TUFT1, and miR-326-3p were most significant in PANC-1 and AsPC-1 cells, which were used in subsequent experiments. To determine the effects of LINC00960 in PC, LINC00960 expression was knocked down using sh-LINC00960 (shRNA transfection had no cytotoxic effect; Figure S2). As shown in Figure 2A, RT-qPCR confirmed the successful knockdown of LINC00960 in PC cells, and cell proliferation was distinctly suppressed due to LINC00960 silencing in these cells (Figures 2B,C and S3A,B). The mRNA expression of the glycolysis-associated genes GLUT1 and LDHA was significantly reduced in PANC-1 and AsPC-1 cells transfected with sh-LINC00960 (Figure 2D,E). Moreover, the knock-down of LINC00960 reduced glucose uptake as well as lactate and ATP production in PC cells (Figures 2F–H and S3C,D). Thus, LINC00960 could regulate proliferation and glycolysis in PC cells.

FIGURE 2. Knocking down LINC00960 suppressed cell proliferation and glycolysis in PC. PANC-1 and AsPC-1 cells were transfected with sh-NC or sh-LINC00960. (A) Transfection efficiency was assessed via real-time quantitative polymerase chain reaction (RT-qPCR). (B,C) Cell proliferation was evaluated using cell counting kit-8 and clone formation assays. (D,E) GLUT1 and LDHA expression were evaluated using RT-qPCR. (F–H) Glucose uptake and lactate and ATP production were evaluated using commercial kits. All of the data were obtained from at least three replicated experiments. The data were expressed as means ± SD. *p [less than] 0.05, **p [less than] 0.01, and ***p [less than] 0.001

LINC00960 targeted miR-326-3p to regulate PC progression

To investigate the connection between LINC00960 and miR-326-3p in PC cells, we performed the following experiments. First, we tested whether the knock-down of LINC00960 affects miR-326-3p expression. We found that LINC00960 knockdown led to increased miR-326-3p expression in PC cells (Figure 3A). Second, the LncBase v2 and LncBook database were used to predict that putative binding sites existed between LINC00960 and miR-326-3p (Figure 3B and Supplementary Material S2). Furthermore, miR-326-3p mimics suppressed luciferase activity in the LINC00960-WT group, and both LINC00960 and miR-326-3p were enriched by Ago2 antibody treatment (Figure 3C,D), confirming the interaction between LINC00960 and miR-326. To assess the influence of miR-326-3p on the effects of LINC00960 in PC cells, rescue experiments were conducted. The expression of miR-326-3p was upregulated by LINC00960 depletion; however, this effect was abolished by the miR-326-3p inhibitor (Figure 3E). Additionally, the miR-326-3p inhibitor reversed the suppressive effect of sh-LINC00960 on cell proliferation (Figure 3F,G). Furthermore, the inhibitory effects of LINC00960 knockdown on processes associated with glycolysis, such as GLUT1 and LDHA expression, glucose uptake and lactate and ATP production, were counteracted by miR-326-3p inhibitor (Figure 3H–L). Therefore, LINC00960 interacted with miR-326, thereby participating in proliferation and glycolysis in PC cells.

FIGURE 3. LINC00960 targeted miR-326-3p to regulate pancreatic cancer (PC) progression. (A) miR-326-3p expression was evaluated using real-time quantitative polymerase chain reaction (RT-qPCR) in PC cells transfected with sh-NC or sh-LINC00960. (B) The binding sequences of LINC00960 on miR-326-3p were predicted using the LncBook database (https://ngdc.cncb.ac.cn/lncbook/index). (C,D) The interaction between LINC00960 and miR-326-3p was verified using dual-luciferase reporter and RNA-binding protein immunoprecipitation assays. PC cells were transfected with sh-LINC00960 alone or together with miR-326-3p inhibitor. (E) miR-326-3p expression was evaluated via RT-qPCR. (F,G) Cell proliferation was evaluated using cell counting kit-8 and clone formation assays. (H,I) GLUT1 and LDHA expression were evaluated using RT-qPCR. (J–L) Glucose uptake and lactate and ATP production were measured using commercial kits. All of the data were obtained from at least three replicated experiments. The data were expressed as means ± SD. *p [less than] 0.05, **p [less than] 0.01, and ***p [less than] 0.001.

miR-326-3p bound to TUFT1 and downregulated TUFT1, p-mTOR, and p-AKT expression

We also investigated the downstream genes of miR-326. RT-qPCR revealed that miR-326-3p mimics reduced TUFT1 expression (Figure 4A), and western blotting showed that miR-326-3p abundance inhibited TUFT1 expression as well as AKT and mTOR phosphorylation (Figure 4B). Further, StarBase predicted that miR-326-3p sequences could bind to TUFT1 (Figure 4C). Luciferase activity was reduced in both the miR-326-3p mimics and TUFT1-WT groups but was not affected in the TUFT1-MUT group (Figure 4D). RIP assay indicated that compared with the IgG group, miR-326-3p and TUFT1 expression levels were notably increased in the Ago2 antibody group (Figure 4E). These results demonstrated that miR-326-3p could directly interact with TUFT1 and negatively regulate TUFT1, p-mTOR, and p-AKT expression.

FIGURE 4. miR-326-3p bound to Tuftelin 1 (TUFT1) and downregulated TUFT1, p-mTOR, and p-AKT expression. Pancreatic cancer (PC) cells were transfected with miR-326-3p mimics. (A) miR-326-3p and TUFT1 expression were evaluated using real-time quantitative polymerase chain reaction. (B) TUFT1, AKT, p-AKT, mTOR, and p-mTOR protein expression were measured using western blotting. (C) The binding site between TUFT1 and miR-326-3p was predicted using StarBase (https://starbase.sysu.edu.cn/starbase2/). The interaction between TUFT1 and miR-326-3p was verified using dual-luciferase reporter assay (D) and RNA-binding protein immunoprecipitation assay (E). All of the data were obtained from at least three replicated experiments. The data were expressed as means ± SD. *p [less than] 0.05, **p [less than] 0.01, and ***p [less than] 0.001

miR-326-3p suppressed cell proliferation and glycolysis in PC by reducing TUFT1 expression

To determine whether TUFT1 participates in miR-326-mediated functions in PC, PC cells were subjected to miR-326-3p mimic transfection alone or in conjunction with overexpression of TUFT1. The miR-326-3p mimic reduced TUFT1 expression, but this phenomenon was reversed by TUFT1 overexpression (Figure 5A,B). In terms of cell proliferation, the suppressive effects of miR-326-3p mimics were neutralized by TUFT1 overexpression (Figure 5C,D). In addition, TUFT1 overexpression abolished the decrease in p-AKT and p-mTOR expression induced by miR-326-3p overexpression (Figure 5E). As expected, miR-326-3p mimics inhibited GLUT1 and LDHA expression, glucose uptake and lactate and ATP production, but these suppressive effects were impaired by TUFT1 overexpression (Figure 5F–J). Thus, miR-326-3p apparently targeted TUFT1 to repress cell proliferation and glycolysis in PC.

FIGURE 5. miR-326-3p suppressed cell proliferation and glycolysis in pancreatic cancer (PC) by reducing Tuftelin 1 (TUFT1) expression. PC cells were transfected with miR-326-3p mimics alone or together with overexpression of TUFT1. (A,B) TUFT1 expression was evaluated via real-time quantitative polymerase chain reaction (RT-qPCR) and western blotting. (C,D) Cell proliferation was evaluated using cell counting kit-8 and clone formation assays. (E) AKT, p-AKT, mTOR, and p-mTOR protein expression were measured using western blotting. (F,G) GLUT1 and LDHA expression were evaluated using RT-qPCR. (H-J) Glucose uptake and lactate and ATP production were measured using commercial kits. All of the data were obtained from at least three replicated experiments. The data were expressed as means ± SD. *p [less than] 0.05, **p [less than] 0.01, and ***p [less than] 0.001

Knocking down LINC00960 limited the proliferation of PC cells by reducing TUFT1 expression

To elucidate the biological function of TUFT1 in PC, PC cells were transfected with overexpression of TUFT1. As shown in Figure 6A, TUFT1 expression was notably increased by OE-TUFT1 transfection. Additionally, TUFT1 overexpression contributed to increasing TUFT1, p-AKT, and p-mTOR expression (Figure 6B). TUFT1 overexpression also increased GLUT1 and LDHA expression levels, glucose uptake and lactate and ATP production (Figure 6C–G). Subsequently, we explored the effects of TUFT1 expression on LINC00960-mediated biological functions in PC cells. Sh-LINC00960 alone or in combination with OE-TUFT1 was transfected into PC cells. LINC00960 depletion suppressed mRNA and protein expression levels of TUFT1, whereas TUFT1 overexpression attenuated the effects of LINC00960 inhibition (Figure 6H,I). Knocking down LINC00960 also inhibited the proliferation of PC cells, but this effect was reversed by TUFT1 overexpression (Figure 6J,K). As a downstream target gene of LINC00960, TUFT1 expression promoted glycolysis in PC cells, and TUFT1 overexpression abolished the sh-LINC00960-mediated proliferation of these cells.

FIGURE 6. LINC00960 knockdown impeded pancreatic cancer (PC) cell proliferation by reducing Tuftelin 1 (TUFT1) expression PC cells were transfected with overexpression of TUFT1. (A) TUFT1 expression was evaluated using real-time quantitative polymerase chain reaction (RT-qPCR). (B) TUFT1, AKT, p-AKT, mTOR, and p-mTOR protein expression were measured using western blotting. (C, D) GLUT1 and LDHA expression were evaluated using RT-qPCR. (E–G) Glucose uptake and lactate and ATP production were measured using commercial kits. PC cells were transfected with sh-LINC00960 alone or coupled with overexpression of TUFT1. (H,I) TUFT1 expression was evaluated via RT-qPCR and western blotting. (J,K) Cell proliferation was evaluated using cell counting kit-8 and clone formation assays. All of the data were obtained from at least three replicated experiments. The data were expressed as means ± SD. *p [less than] 0.05, **p [less than] 0.01, and ***p [less than] 0.001

LINC00960 knockdown inhibited tumor growth

We established a xenograft model bearing PANC-1 cells transfected with sh-LINC00960 in nude mice. When compared with the sh-NC group, the tumor volume and weight were decreased in the sh-LINC00960 group (Figure 7A–C). IHC images showed that PCNA and TUFT1 expression levels were downregulated in the sh-LINC00960 group (Figure 7D), in which the expression levels of LINC00960 and TUFT1 were low, whereas those of miR-326-3p were relatively high (Figure 7E). Furthermore, the expression of TUFT1, p-AKT and p-mTOR was inhibited in the tumor tissues of sh-LINC00960 group mice (Figure 7F). These data implied that knocking down LINC00960 inhibited tumor growth in vivo.

FIGURE 7. LINC00960 knockdown inhibited tumor growth. Nude mice were injected subcutaneously with PANC-1 cells transfected with sh-NC or sh-LINC00960 (five mice/group). (A) Images of xenograft tumors. (B) Tumor volume. (C) Tumor weight. (D) Proliferating cell nuclear antigen (PCNA) and Tuftelin 1 (TUFT1) expression were detected using immunohistochemistry. (E) LINC00960, miR-326-3p, and TUFT1 expression in tumor tissues were evaluated using real-time quantitative polymerase chain reaction. (F) TUFT1, AKT, p-AKT, mTOR, and p-mTOR protein expression were measured via western blotting. The data were expressed as means ± SD. *p [less than] 0.05, **p [less than] 0.01, and ***p [less than] 0.001

DISCUSSION

PC is a disease with a high fatality rate and a 5-year survival rate of ~10%.¹⁸ Patients with early-stage PC exhibit very mild clinical symptoms, and early detection markers are currently lacking, which contributes to the high mortality rate of PC.¹⁹ In this study, LINC00960 was found to facilitate cell proliferation and glycolysis in PC by inhibiting the miR-326-3p/TUFT1/AKT–mTOR pathway.

LncRNAs serve as carcinogenic or anticancer RNAs and participate in the occurrence and development of PC.^20,21 Herein, we studied the expression of LINC00960 and the molecular mechanism underlying its effects in PC. Corroborating the findings of Wu et al.¹¹ and Huang et al.,¹² LINC00960 expression increased in PC in this study, indicating that LINC00960 is an oncogene associated with PC. Another study showed that LINC00960 accelerated cell growth and metastasis in PDAC by inhibiting the miR-146a-5p/IRAK1 axis.¹² In lung adenocarcinoma (LADC), LINC00960 mediated the miR-124a/sphingosine kinase 1 axis to induce LADC development by promoting cell proliferation, migration, and invasion.²² We obtained similar outcomes, that is, LINC00960 silencing impeded PC cell proliferation. Although the proliferation rate in PC cells was not accurate due to the insufficient density,^23,24 we still confirmed that LINC00960 inhibited PC cell proliferation. Notably, an association between lncRNAs and glycolysis has been reported, and lncRNAs are known to regulate the progression of multiple cancers. For instance, lncRNA LINRIS bound to insulin-like growth factor 2 mRNA-binding protein 2 to accelerate aerobic glycolysis, expediting the progression of colorectal cancer.²⁵ LncMSC-AS1-regulated PFKFB3 expression and promoted glycolysis to stimulate cell growth in gastric cancer.²⁶ To the best of our knowledge, we were the first to report that LINC00960 increased the expression of glycolysis-related genes (GLUT1 and LDHA), elevated the capacity of glucose uptake, and enhanced the production of lactate and ATP, suggesting that LINC00960 was positively associated with glycolysis in PC cells.

MiR-326-3p has been studied in multiple cancer types, including breast cancer,²⁷ gastric cancer,²⁸ and LADC.²⁹ In these cancer types, miR-326-3p expression was low and it functioned as an anticancer factor.^27–30 Further, miR-326-3p expression was reported to be reduced in PC.^13,31 However, the role of miR-326-3p in PC and the underlying mechanism of its effects have rarely been investigated. It is well-accepted that lncRNAs, which act as competitive endogenous RNAs, interact with miRNAs and are involved in the occurrence and development of numerous tumors.³² In this study, miR-326-3p expression was reduced in PC cells, which was consistent with the findings of a previous study.³¹ Combining bioinformatics analysis, luciferase assay, and RIP assay, we verified that LINC00960 interacted with miR-326. However, Ago2 is involved in the process of miRNA maturation, and it can also combine with mature miRNA to mediate its function. The molecular sponge effect of lncRNA binding miRNA is based on the mechanism of binding mature miRNAs. LINC00960/miRNA-326 was found in the Ago2 complex pull-down products in our study. Furthermore, LINC00960 decreased miR-326-3p expression to mediate proliferation and glycolysis in PC.

TUFT1 is a well-known carcinogen in multiple cancers, including hepatocellular carcinoma,³³ renal cell carcinoma (RCC),³⁴ and PC.¹⁴ In this study, TUFT1 was highly expressed in PC. Lin et al.³⁴ reported that TUFT1 facilitated the growth of RCC cells by up-regulating the phosphorylation of AKT. In thyroid carcinoma, TUFT1 mediated the AKT–mTOR/GSK3β signaling pathway to accelerate cell invasion and proliferation.³⁵ In bladder cancer, HBXIP regulated the AKT/mTOR pathway to induce glycolysis.³⁶ Previous research has indicated that TUFT1 is involved in glycolysis in PC as it mediates the AKT/mTOR pathway. Our study revealed that TUFT1 overexpression increased p-AKT and p-mTOR levels and promoted glycolysis. miRNA sequences can partially or completely pair with the 3′-untranslated regions (UTR) region of the target gene, thereby resulting in the posttranscriptional silencing of the target gene.³⁷ In our study, miR-326-3p was combined with TUFT1, thereby down-regulating its expression. Additionally, miR-326-3p mimics reduced p-AKT and p-mTOR levels and inhibited cell proliferation and glycolysis in PC; however, these effects were reversed by TUFT1 overexpression. Based on these data, miR-326-3p inhibited cell proliferation and glycolysis via the TUFT1/AKT–mTOR pathway and impeded PC development.

Importantly, we also established xenograft models of mice inoculated with PANC-1 cells to verify our in vitro results in vivo. We confirmed that LINC00960 silencing reduced TUFT1, PCNA, p-AKT, and p-mTOR expression levels; increased miR-326-3p expression levels and inhibited tumor growth in mice. Taken together, our findings revealed that LINC00960 knockdown effectively inhibited cell proliferation and glycolysis in PC. Moreover, the LINC00960-mediated mechanism underlying PC progression was involved the adjustment of the miR-326/TUFT1 axis, which further affected the AKT–mTOR pathway to expedite PC development. Due to time constraints and the lack of sufficient clinical samples, we were unable to evaluate the relationships among LINC00960, miR-326-3p, and TUFT1 in PC, but we will continue to explore these relationships in subsequent clinical studies. Nevertheless, our results suggested that the LINC00960/miR-326-3p/TUFT1/AKT–mTOR axis contained markers or therapeutic targets for PC.

ACKNOWLEDGMENT

We would like to give our sincere gratitude to the reviewers for their constructive comments.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

ETHICS STATEMENT

Animal experiments were approved by the ethics committee of the Affiliated Hospital of Qingdao University.