LncRNA LINC02381/miR-133b axis promotes tumorigenesis and predicts survival in pancreatic cancer.

Table  1 records the correlation between the expression of LINC02381 and the clinical features of included pancreatic cancer patients. LINC02381 expression in tissues of 112 patients with pancreatic cancer was detected and divided into low expression of LINC02381 group ( n  = 55) and high expression of LINC02381 group ( n  = 57) according to the mean value. As can be seen from the table, LINC02381 expression is not affected by the age ( P  = 0.855), gender ( P  = 0.697), venous involvement ( P  = 0.241), differentiation ( P  = 0.190), tumor size ( P  = 0.131), and location ( P  = 0.354) of pancreatic cancer patients, but is correlated with patients’ lymph node metastasis ( P  = 0.024), TNM stage ( P  = 0.034) and perineural invasion ( P  = 0.044). Table 1 Correlation between expression of LncRNA LINC02381 and clinical characteristics of pancreatic cancer patients Characteristics Cases ( n  = 112) LINC02381 expression P values Low ( n  = 55) High ( n  = 57) Age 0.855 ≤ 60 years 58 28 (48.3%) 30 (51.7%) > 60 years 54 27 (50.0%) 27 (50.0%) Gender 0.697 Male 59 30 (50.8%) 29 (49.2%) Female 53 25 (47.2%) 28 (52.8%) Lymph node metastasis No 83 46 (55.4%) 37 (44.6%) 0.024 Yes 29 9 (31.0%) 20 (69.0%) TNM stage 0.034 I-II 77 43 (55.8%) 34 (44.2%) III-IV 35 12 (34.3%) 23 (65.7%) Perineural invasion No 71 40 (56.3%) 31 (43.7%) 0.044 Yes 41 15 (36.6%) 26 (63.4%) Venous involvement 0.241 No 93 48 (51.6%) 45 (48.4%) Yes 19 7 (36.8%) 12 (63.2%) Differentiation 0.190 Well and Moderate 50 28 (56.0%) 22 (44.0%) Poor 62 27 (43.5%) 35 (56.5%) Tumor size 0.131 ≤ 2 cm 55 31 (56.4%) 24 (43.6%) > 2 cm 57 24 (42.1%) 33 (57.9%) Location 0.354 Head 64 29 (45.3%) 35 (54.7%) Body and Tail 48 26 (54.2%) 22 (45.8%) Correlation between expression of LncRNA LINC02381 and clinical characteristics of pancreatic cancer patients RT-qPCR detection was performed on 112 pancreatic cancer tissue samples and 75 normal tissue samples, and it was found that LINC02381 expression was highly enriched in tumor tissues, as shown in Fig.  1 A. Then, RT-qPCR detection in pancreatic cancer cells also showed that compared with normal pancreatic epithelial cells HPDE, LINC02381 expression was increased in Panc-1, BxPC-3, ASPC-1 and SUIT-2 cells (Fig.  1 B). The content of miR-133b was assessed by the same procedure. In both pancreatic cancer tissues and cells, miR-133b was expressed in a decreasing trend (Fig.  1 C and D). Fig. 1 Levels of LINC02381 and miR-133b in pancreatic cancer. A Elevated expression of LINC02381 in tissue samples was detected through RT-qPCR (*** P  < 0.001 vs. Normal Tissue). B LINC02381 expression was upregulated in pancreatic cancer cells by RT-qPCR (** P  < 0.01, *** P  < 0.001 vs. HPDE). C – D miR-133b was lowly expressed in pancreatic cancer tissues (*** P  < 0.001 vs. Normal Tissue) and cells (*** P  < 0.001 vs. HPDE) Levels of LINC02381 and miR-133b in pancreatic cancer. A Elevated expression of LINC02381 in tissue samples was detected through RT-qPCR (*** P  < 0.001 vs. Normal Tissue). B LINC02381 expression was upregulated in pancreatic cancer cells by RT-qPCR (** P  < 0.01, *** P  < 0.001 vs. HPDE). C – D miR-133b was lowly expressed in pancreatic cancer tissues (*** P  < 0.001 vs. Normal Tissue) and cells (*** P  < 0.001 vs. HPDE) The molecular mechanism of the regulation of LINC02381 on pancreatic cancer was studied by exploring the targeted factors of LINC02381. The ENCORI/Starbase website suggests that LINC02381 targets miR-133b and there are binding sites between LINC02381 and miR-133b (Fig.  2 A). Subsequently, the binding ability of LINC02381 and miR-133b was studied by luciferase experiment. Figure  2 B illustrates that the luciferase activity of MUT-LINC02381 did not change significantly, while the luciferase activity of WT-LINC02381 was downregulated by transfected miR-133b mimic and the luciferase activity of WT-LINC02381 was upregulated by transfected miR-133b inhibitor. Pearson correlation analysis confirmed that LINC02381 expression in pancreatic cancer tissues was negatively correlated with miR-133b ( r = -0.6967, P  < 0.0001), as shown in Fig.  2 C. Moreover, the results of RIP assay showed that Ago2 specifically captured LINC02381 and enriched miR-133b compared with the IgG antibody group, revealing their binding relationship (Fig.  2 D). Fig. 2 Correlation between LINC02381 and miR-133b. A The binding site of LINC02381 to miR-133b. B Luciferase assay confirmed that LINC02381 directly targeted miR-133b (*** P  < 0.001 vs. control). C Correlation between LINC02381 and miR-133b in pancreatic cancer tissues ( r = -0.6967, P  < 0.0001). D RIP assay illustrated that LINC02381 and miR-133b were enriched in Panc-1 cells (*** P  < 0.001 vs. IgG) Correlation between LINC02381 and miR-133b. A The binding site of LINC02381 to miR-133b. B Luciferase assay confirmed that LINC02381 directly targeted miR-133b (*** P  < 0.001 vs. control). C Correlation between LINC02381 and miR-133b in pancreatic cancer tissues ( r = -0.6967, P  < 0.0001). D RIP assay illustrated that LINC02381 and miR-133b were enriched in Panc-1 cells (*** P  < 0.001 vs. IgG) In Fig.  3 A, LINC02381 level decreased significantly after transfection of si-LINC02381 into Panc-1 cells, compared with the transfection control group. Additionally, miR-133b was elevated in cells after knockdown of LINC02381, while this up-regulation trend was curbed after transfection with miR-133b inhibitor (Fig.  3 B). In the cell proliferation assay, the OD value of Panc-1 cells was lower than that of the control group after transfection with si-LINC02381. However, when co-transfected with si-LINC02381 and miR-133b inhibitor, the OD value increased (Fig.  3 C). Similarly, Transwell migration and invasion experiments showed that the migration number of Panc-1 cells was obviously reduced after LINC02381 knockdown, as was the number of invasive cells in Fig.  3 D and E. After transfection with si + miR-133b inhibitor, the above inhibitory effects were counteracted. After silencing LINC02381 and miR-133b were introduced into BxPC-3 cells, the transfection efficiency and biological function of BxPC-3 cells also reflected the same regulatory results (Fig.  4 A and E). In conclusion, silenced LINC02381 negatively regulates the activity of pancreatic cancer cells and inhibits cell proliferation behavior, migratory ability, and invasive processes, whereas the involvement of miR-133b inhibitor alleviates this inhibitory effect. Fig. 3 Effect of silencing LINC02381 on Panc-1 cells. A – B The expression of LINC02381and miR-133b in Panc-1 cells after transfection was detected by RT-qPCR (* P  < 0.05, *** P  < 0.001 vs. control; # P  < 0.05 vs. si-LINC02381). C The proliferative activity of Panc-1 cells after transfection with si-NC and si-LINC02381 was detected by CCK-8 assay (*** P  < 0.001 vs. control; ### P  < 0.001 vs. si-LINC02381). ( D ) and ( E ) The migration situation and invasion activity were performed by Transwell assay (*** P  < 0.001 vs. control; # P  < 0.05 vs. si-LINC02381) Effect of silencing LINC02381 on Panc-1 cells. A – B The expression of LINC02381and miR-133b in Panc-1 cells after transfection was detected by RT-qPCR (* P  < 0.05, *** P  < 0.001 vs. control; # P  < 0.05 vs. si-LINC02381). C The proliferative activity of Panc-1 cells after transfection with si-NC and si-LINC02381 was detected by CCK-8 assay (*** P  < 0.001 vs. control; ### P  < 0.001 vs. si-LINC02381). ( D ) and ( E ) The migration situation and invasion activity were performed by Transwell assay (*** P  < 0.001 vs. control; # P  < 0.05 vs. si-LINC02381) Fig. 4 The biological behavior of BxPC-3 cells was regulated by knocking down LINC02381 and miR-133b. A – B LINC02381 and miR-133b expression in BxPC-3 cells after transfection assays (*** P  < 0.001 vs. control; ## P  < 0.01 vs. si-LINC02381). C - E The biological behavior of BxPC-3 cells is mediated by the co-transfection of dysregulated LINC02381 and miR-133b (*** P  < 0.001 vs. control; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs. si-LINC02381) The biological behavior of BxPC-3 cells was regulated by knocking down LINC02381 and miR-133b. A – B LINC02381 and miR-133b expression in BxPC-3 cells after transfection assays (*** P  < 0.001 vs. control; ## P  < 0.01 vs. si-LINC02381). C - E The biological behavior of BxPC-3 cells is mediated by the co-transfection of dysregulated LINC02381 and miR-133b (*** P  < 0.001 vs. control; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs. si-LINC02381) Survival data of the included pancreatic cancer patients were collected for a period of five years after surgery. The Kaplan-Meier curve in Fig.  5 A showed that compared with the group with high expression of LINC02381 ( n  = 57), patients in the group with low expression of LINC02381 ( n  = 55) had a higher survival rate (Log-rank P  = 0.015). Moreover, patients in the low miR-133b group showed worse survival outcomes (Log-rank P  = 0.037; Fig.  5 B). What’s more, Univariate and multivariate Cox analysis also demonstrated that LINC02381 expression could be an independent prognostic factor for survival in patients with pancreatic cancer in Table  2 . In summary, overexpression of LINC02381 implies poor prognosis for pancreatic cancer patients and can be regarded as a prognostic marker for pancreatic cancer. Fig. 5 The Kaplan-Meier curve evaluated the five-year survival of enrolled pancreatic cancer patients. A The survival rate of LINC02381 (Log-rank P  = 0.015). B The survival rate of miR-133b (Log-rank P  = 0.037) The Kaplan-Meier curve evaluated the five-year survival of enrolled pancreatic cancer patients. A The survival rate of LINC02381 (Log-rank P  = 0.015). B The survival rate of miR-133b (Log-rank P  = 0.037) Table 2 Cox analysis of clinical characteristics for pancreatic cancer patients Characteristics Univariate Cox analysis Multivariate Cox analysis HR 95% CI P value HR 95% CI P value LINC02381 0.386 0.173–0.860 0.020 0.288 0.120–0.694 0.006 Age 0.975 0.487–1.951 0.942 1.174 0.522–2.637 0.698 Gender 0.679 0.336–1.373 0.281 0.675 0.305–1.495 0.333 Lymph node metastasis 0.362 0.180–0.727 0.004 0.429 0.195–0.945 0.036 TNM stage 0.352 0.173–0.716 0.004 0.465 0.203–1.065 0.070 Perineural invasion 2.679 1.095–6.552 0.031 3.151 1.212–8.193 0.019 Venous involvement 0.895 0.344–2.332 0.820 1.554 0.555–4.348 0.401 Differentiation 1.175 0.587–2.352 0.649 1.465 0.712–3.017 0.300 Tumor size 0.933 0.466–1.868 0.844 1.773 0.773–4.066 0.176 Location 1.108 0.540–2.273 0.779 1.238 0.589–2.599 0.573 Cox analysis of clinical characteristics for pancreatic cancer patients

A total of 112 patients with pancreatic cancer who were first diagnosed and underwent surgical treatment in The Affiliated Hospital of Panzhihua University from January 2015 to December 2016 were identified as subjects of this study. Pancreatic cancer patients who had previously received antitumor therapy or had common diseases such as heart disease, high blood pressure and diabetes were excluded from the study. Pancreatic cancer tissues and corresponding normal tissues of the subjects were collected, and the tissue samples were stored in the refrigerator at -80°C immediately after surgery. Meanwhile, clinical characteristics date of pancreatic cancer patients was summarized and recorded. All the above subjects signed informed consent voluntarily and knowingly, which cooperated with follow-up observation within five years after surgery. ATCC provided the pancreatic cancer cells Panc-1, BxPC-3, ASPC-1, SUIT-2 and normal control cells HPDE required for this study. The experimental cells were cultured according to a uniform standard, that is, the cells were inoculated in Dulbecco’s Modified Eagle’s medium (DMEM, HyClone) containing 10% fetal bovine serum (FBS, Gibco) in a humidified incubator at 37℃ with 5% CO 2 . To achieve low expression of LINC02381, represented by Panc-1 and BxPC-3 cells from cultured pancreatic cancer cells. Control group, negative control group (si-NC), and silenced LINC02381 (si-LINC02381) were transfected into Pan-1 and BxPC-3 cells by Lipofectamine 2000 transfection reagent (Invitrogen, USA). After the transfected cells were cultured at 37℃ for 48 h, LINC02381 level was detected by RT-qPCR to reflect the transfection efficiency. TRIZOL reagent (ThermoFisher, USA) was added to tissue or cell samples to extract total RNA. Then the RNA was reverse transcribed into cDNA using ReverTra Ace qPCR RT Kit (Toyobo, Japan), and the obtained cDNA as template, SYBR GREEN qPCR Kit (Vazyme, Nanjing) was used to prepare the reaction system and was amplified in the PCR instrument ABI 7500 for RT-qPCR detection. GAPDH and U6 were the reference genes of LINC02381 and miR-133b. Finally, the relative expression levels of LINC02381 and miR-133b were calculated by 2 −ΔΔCt method. Primer sequences included: LINC02381, forward 5’-CTGATGGCCACTCACGCTAT-3’ and reverse 5’-GATCCGGAGGGAGAGAGCATTC-3’; miR-133b, forward 5’-GGACCCCAACAACCAGCAA-3 and reverse 5’-CTGGTTGTTGGGGTCCTTT-3’; GAPDH, forward 5’-GGTATCGTGGAAGGACTCA-3’ and reverse 5’-TCATCATATTTGGCAGGT-3’; U6, forward 5’-CTCGCTTCGGCAGCAC-3’ and reverse 5’-AACGCTTCACGAATTTGCG-3’. The transfected Panc-1 and BxPC-3 cells were inoculated into 96-well cell culture plates, and CCK-8 detection reagent (Solarbio, Beijing) was added into each well every 24 h. After incubation in 37°C incubator for 2 h, the absorbance values of the cells were measured by enzyme-labeled apparatus. The Transwell chamber was used to measure the level of pancreatic cancer cell migration and invasion. After transfection, the cells were resuspended in DMEM medium and transferred to the upper chamber of the Transwell chamber. Then, DMEM medium with 10% FBS was added to the lower chamber of the Transwell chamber. After 24 h of culture, the migrated cells were fixed and stained with 100% methanol and 0.5% crystal violet. The cells were finally counted under an inverted microscope to assess the ability of migration. The difference is that the detection of cell invasion level required pre-coating the Transwell chamber with Matrigel (BD Biosciences). The ENCORI/Starbase website ( https://rnasysu.com/encori/ ) is widely used for RNA-RNA interactions, through which we identified the downstream target genes of LINC02381. The binding site of LINC02381 to miR-133b was amplified to obtain wild-type and mutant LINC02381 fragments, which were then cloned into the pmirGLO vector (Promega, Beijing). The WT-LINC02381 and MUT-LINC02381 were co-transfected into Pan-1 cells with control, mimic NC, miR-133b mimic, inhibitor NC, and miR-133b inhibitor to confirm the relationship between LINC02381 and miR-133b. The luciferase activity was determined by dual-luciferase gene reporting Kit (Promega, Beijing). Enrichment of lncRNA bound to miRNA was possible by anti-Ago2 antibody, which in turn verified the lncRNA-miRNA binding relationship. RIP assays were performed according to the protocol of Magna RIP kit (Millipore, MA, USA). Pan-1 cells were collected and lysed in RIP lysis buffer, treated with Ago2 and IgG antibody (Abcam, UK) and incubated at low temperature overnight. Then the immunoprecipitated RNA was purified and assayed for expression. Each part of the experiment should be measured more than three times to minimize operating errors. The obtained data were processed by SPSS 23 or GraphPad 7 software and expressed as mean ± standard deviation. Student t -test and One-way ANOVA with Tukey’s multiple comparative analysis was used for comparison between groups. Pearson correlation analysis was used to evaluate the correlation between LINC02381 expression and clinical features of pancreatic cancer patients. Prognostic ability of LINC02381 in pancreatic cancer was predicted by Kaplan-Meier survival curve and Cox test. P  < 0.05 was considered statistically significant.

Pancreatic cancer is mostly ductal adenocarcinoma originating from glandular duct epithelium, accounting for more than 90% of all patients [ 16 ]. There was evidence that genetic mutations and familial factors contribute to the development of hereditary pancreatic cancer [ 17 ], and precancerous lesions characterized by pancreatic intraepithelial neoplasia (PanIN) are also predisposing factors for tumor induction [ 18 , 19 ]. While other studies have shown that poor diet, long-term smoking or excessive body fat may aggrandize the risk of pancreatic cancer, the specific causes are still not completely clear [ 20 ]. In addition, the onset of pancreatic cancer is insidious, and some patients have no discomfort, or only suffer from lumbago, dyspepsia, abdominal discomfort and other common symptoms of digestive system diseases, which is not beneficial to the diagnosis of the disease [ 21 ]. When persistent symptoms such as severe abdominal pain, jaundice, and extreme fatigue appear, the pancreatic cancer patient may be in the advanced stage of the disease and may rapidly deteriorate or die within a short period of time [ 22 , 23 ]. Therefore, taking a more promising approach within a limited time may better solve the dire prognosis of pancreatic cancer patients. After decades of efforts, the biological function of lncRNAs in diseases has long become a research hotspot and some achievements have been made already. LINC02381 is located at 12q13.13 and has a length of 1411nt [ 24 ]. In recent years, high levels of LINC02381 have been elucidated in glioblastoma, and the micropeptide encoded by LINC02381 may serve as a therapeutic target to regulate ferroptosis in glioblastoma [ 25 ]. Additively, in a 2022 report, LINC02381 was presented as one of the prognostic risk model indicators for colorectal cancer, with visibly higher expression in colorectal cancer cells than in normal cells [ 26 ]. In this study, LINC02381 was highly enriched in pancreatic cancer. In the available evidence, LINC02381 has been suggested to have a pro-oncogenic role in gliomas and to influence tumor therapy by mediating downstream targets [ 27 ]. Similarly, up-regulated LINC00941 [ 28 ], TPT1-AS1 [ 29 ] and FAM83H-AS1 [ 30 ] were found in pancreatic cancer tissues and cell samples, and correlated with poor prognosis of patients. Furthermore, the binding site of miR-133b and LINC02381 was identified by bioinformatics tools. Therefore, this study further confirmed that LINC02381 was the sponge of miR-133b in pancreatic cancer through luciferase activity and RIP assays. We found that LINC02381 is involved in tumor development via negatively regulating miR-133b. Wu et al. proposed that lncRNA MEM147-AS1 regulates the tumorigenesis of prostate cancer by directly targeting miR-133b, thus affecting the invasion of tumor cells [ 31 ]. He and colleagues also investigated the increased expression of lncRNA NCK1-AS1 and down-regulated expression of miR-133b in esophageal squamous cell carcinoma, and the inhibition of NCK1-AS1 to slow tumor progression through regulation of miR-133b/ENPEP axis [ 32 ]. Minutely, miR-133b has been reported to bind directly to lncRNA TTN-AS1 in the pancreas, and the targeting of low-expression miR-133b by TTN-AS1 has opened new directions for the control of pancreatic cancer [ 33 ]. All of the above conclusions indicate that miR-133b can be used as a cancer regulator, promoting or hindering tumor progression through binding with lncRNAs. The in vitro assays demonstrated that the proliferation behavior, migratory capacity, and invasive processes of pancreatic cancer cells were significantly diminished following the knockdown of LINC02381. This suggests that silencing LINC02381 could play a crucial role in halting the progression of pancreatic cancer. Reversion experiments further indicated that co-transfection of si-LINC02381 with miR-133b inhibitor counteracted the inhibitory effects of LINC02381 silencing on the viability of pancreatic cancer cells. This finding implies that LINC02381 may govern tumor development through its regulatory relationship with miR-133b. Additionally, prognostic analyses revealed that patients exhibiting lower levels of LINC02381 experienced improved survival outcomes, while low miR-133b levels suggested poor prognosis. Notably, previous research has identified that the interaction of miR-133b with the downstream target SQLE is instrumental in the advancement of pancreatic cancer [ 33 ]. However, the current study has yet to provide a comprehensive exploration of the regulatory network involving LINC02381 and miR-133b. We intend to further investigate the downstream targets impacted by the actions of miR-133b in subsequent analyses. In addition, the prospective collection of more complete clinical parameters while recruiting more pancreatic cancer patients will also help to enhance the reference of subsequent studies. In summary, lncRNA LINC02381 is prominently expressed in pancreatic cancer, and it affects cell growth activity by directly targeting miR-133b, thus regulating tumor progression. Meanwhile, overexpression of LINC02381 greatly impaired the survival of patients, opening a new perspective for the study of prognostic biomarkers of pancreatic cancer.

Pancreatic cancer primarily arises from the epithelial cells of the pancreatic duct or the acinar cells of the pancreas, earning it the moniker king of cancer due to its aggressive nature and poor prognosis [ 1 , 2 ]. Patients with pancreatic cancer often exhibit nonspecific symptoms in the early stage, making diagnosis particularly challenging, the disease typically progresses at a rapid pace, resulting in a notably short survival duration [ 3 ]. The majority of individuals diagnosed with pancreatic cancer experience metastatic spread, with overall 5-year survival rates typically falling below 10%, as highlighted in recent literature [ 4 ]. For now, doctors will refer to the basic clinical conditions of patients, and reasonably formulate specific treatment plans based on existing diagnosis and treatment means, in order to control tumors, reduce complications and improve the quality of life of patients as much as possible. Among the various treatment options, surgical resection remains the most effective intervention; however, it is accessible to only a subset of patients [ 5 , 6 ]. Based on this, expanding our understanding of the molecular mechanisms underlying pancreatic cancer and identifying potential prognostic factors will facilitate timely monitoring of patient conditions, ultimately leading to improved quality of life and a heightened likelihood of successful treatment outcomes. Long noncoding RNAs (lncRNAs), a popular research subject, have been proven to play a role in a number of different diseases, including the progression of pancreatic cancer. For example, Zhang et al. reported that blocking the expression of lncRNA UPK1A-AS1 increased the sensitivity of pancreatic cancer cells to oxaliplatin, revealing the potential of UPK1A-AS1 as a therapeutic target for pancreatic cancer [ 7 ]. Zhou et al. demonstrated that lncRNA ABHD11-AS1 is strongly expressed in pancreatic cancer tissues, and that pancreatic cancer cell-derived exosomes can accelerate lymphocyte reproduction and lymphatic vessel formation by controlling ABHD11-AS1 expression [ 8 ]. In addition, microRNAs (miRNAs) have also been found to exert biological effects as downstream targets of lncRNAs and participate in tumor regulation [ 9 , 10 ]. It has been reported that lncRNA LINC01857 is overexpressed in pancreatic ductal adenocarcinoma, which regulates SMOC2 levels by targeting miR-19a-3p, thereby affecting tumor progression [ 11 ]. There was also evidence that lncRNA HOXD-AS1 acts as a sponge for miR-664b-3p to upregulate PLAC8 and accelerate cell biological function in pancreatic cancer mechanisms, which may serve as a novel therapeutic factor [ 12 ]. It is noteworthy that lncRNA LINC02381 has also been reported to play a corresponding role in endometriosis [ 13 ], cervical cancer [ 14 ], and glioma [ 15 ]. Unfortunately, the mechanism of action and regulatory capacity of LINC02381 have not been described in the investigation of pancreatic cancer. Based on the existing literature, this study hypothesized that LINC02381 may regulate the progression of pancreatic cancer through adsorption of downstream genes, and confirmed by a series of characterization experiments, which would bring new insights into the prognosis and treatment of pancreatic cancer patients in the future.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1