Methylation-mediated LINC00261 suppresses pancreatic cancer progression by epigenetically inhibiting c-Myc transcription.

Background: Due to the limitations of strategies for its early diagnosis and treatment, pancreatic cancer (PC) remains a substantial human health threat. We previously discovered a methylation-mediated lncRNA, LINC00261, which is downregulated in PC tissues. However, the underlying role of LINC00261 in PC remains largely unknown.
Methods: Quantitative real-time PCR and in situ hybridization were performed to evaluate the expression levels of LINC00261 in PC, adjacent nontumor and normal pancreas tissues. The clinical significance of LINC00261 was assessed in multicenter PC samples. The functions of LINC00261 in PC were investigated by gain- and loss-of-function assays in vitro and in vivo. Potential downstream pathways and mechanisms were explored via RNA sequencing and bioinformatic analyses. RNA immunoprecipitation and chromatin immunoprecipitation assays were used to validate the underlying mechanisms. Pyrosequencing and targeted demethylation of the LINC00261 promoter were performed to explore the upstream epigenetic mechanisms and therapeutic potential.
Results: LINC00261 was significantly downregulated in PC tissues, and its expression was positively associated with the prognosis of PC patients. Phenotypic studies indicated that LINC00261 overexpression significantly suppressed PC cell proliferation, migration and metastasis in vitro and in vivo. c-Myc was identified as a downstream target of LINC00261. LINC00261 repressed c-Myc transcription by physically interacting and binding with the bromo domain of p300/CBP, preventing the recruitment of p300/CBP to the promoter region of c-Myc and decreasing the H3K27Ac level. Moreover, the methylation level of the LINC00261 promoter was high in PC tissues and was correlated with poor prognosis. Targeted demethylation of the LINC00261 promoter inhibited PC progression both in vitro and in vivo. Conclusions: Our findings indicate that methylation-mediated LINC00261 suppresses PC progression by epigenetically repressing c-Myc expression. These findings expand the therapeutic potential of LINC00261, possibly providing evidence to support the development of epigenetic drugs or therapeutic strategies. This research adds further insights into the etiology of PC and indicates that LINC00261 may be a prognostic and therapeutic target in PC. © The author(s).

Introduction

According to the latest global cancer statistics, pancreatic cancer (PC) is the seventh leading cause of cancer-related death 1. Due to the limitations of strategies for its early diagnosis and treatment, the five-year survival rate for PC patients is less than 7% 2. Moreover, the incidence and mortality rates of PC continue to increase 3; thus, novel therapeutic targets and derived treatment strategies are urgently needed for physicians and patients.

Recent studies have shown that lncRNA dysregulation is involved in various malignancies, such as lung cancer 4, esophageal cancer 5, gastric cancer 6, liver cancer 7 and PC 8, 9, by affecting diverse tumor biological behaviors, such as proliferation, invasion, migration and metastasis, both in vitro and in vivo. Increasing numbers of lncRNA studies in PC have been reported 9-12. However, detailed evidence elucidating the underlying functions of lncRNAs, especially their regulatory roles in driving the key proto-oncogenes and oncogenes in PC—such as Kras 11, p53 12 and c-Myc 13—remains scarce.

Epigenetic regulation of lncRNAs has been suggested to be an important mechanism contributing to cancer progression 14. Although lncRNAs can act as tumor suppressors in various cancers 15, some lncRNAs with tumor suppressor capability might be inactivated in cancers due to methylation of their promoter regions 16. Thus, the application of targeted demethylation techniques 17 to these tumor suppressor lncRNAs might contribute to the discovery of novel treatment methods such as epigenetic drugs.

We applied data mining techniques to public Gene Expression Omnibus (GEO) datasets of lncRNA microarray data to identify differentially expressed lncRNAs between PC and normal pancreas (NP) tissues. Among the numerous dysregulated lncRNAs, the DNA-methylated LINC00261 exhibited markedly lower expression levels in PC tissue than in NP tissue. Previous studies showed that LINC00261 might be a biomarker 18 and metastasis inhibitor in PC 19, 20. However, these studies were performed in vitro, possibly limiting the power of their results, and the prognostic value of LINC00261 has not been directly validated in multicenter PC cohorts. More importantly, the upstream and downstream mechanisms underlying the epigenetic role and potential therapeutic utility of LINC00261 remain unclear. To address these issues, we aimed to comprehensively investigate the molecular mechanisms related to LINC00261 and the potential therapeutic and prognostic value of this lncRNA in PC.

Materials and Methods

Cohorts and tissue samples

All fresh-frozen tissue samples, including the PC tissues with adjacent noncancerous tissues used for qPCR (30 NP tissues and 150 PC tissues, 40 pairs of PC and adjacent nontumor tissues), were obtained from the Institute of Hepatopancreatobiliary Surgery, Southwest Hospital, Army Medical University. Of the 298 formalin-fixed, paraffin-embedded tissues contained in the two independent tissue microarrays, 55 PC (Cohort 1) and 13 NP tissues were obtained from the archival collections of Southwest Hospital, 50 PC tissues were obtained from Soochow University (Cohort 2) and 100 PC and 80 NP tissues were purchased from Outdo Biotech (Shanghai, China) (Cohort 3). For the 850k microarray, 28 PC tissues and 18 adjacent noncancerous tissues were obtained from the Southwest Hospital. NP samples were obtained from organ donors. The pathological type of all PC tissues in the present study was pancreatic ductal adenocarcinoma, and diagnoses were made based on surgical pathology. No patient received either chemotherapy or radiotherapy before surgery. The clinical stage was evaluated based on the guidelines in the American Joint Committee on Cancer (AJCC) 7th edition. Resected specimens were cut into blocks of a proper size, immediately submerged in the RNA preserving reagent RNAlater (Thermo, USA) and either frozen in liquid nitrogen for further RNA extraction and qPCR analysis or formaldehyde fixed and paraffin embedded for further histological analysis. Each sample was used for only one specific type of assay, for example, in situ hybridization (ISH) or qPCR. Follow-up was performed every three months after surgery to obtain the survival status. The study protocol was approved by the Ethics Committee of Southwest Hospital (No. KY201875), Army Medical University. All patients provided written informed consent upon admission for the use of human specimens. All procedures, including the use of human tissue specimens and analysis of clinical data, were carefully handled to meet the guidelines of the Declaration of Helsinki.

In vivo animal experiments

Four- to six-week-old female nude mice were purchased from Southwest Hospital (Chongqing, China). All animal experimental procedures were carried out under aseptic conditions. To establish the subcutaneously implanted tumor model, 2×106 cells (in a total volume of 0.1 ml of PBS) were injected into the dorsal region of each mouse at the sixth week. The body weight and tumor growth of each mouse were measured weekly. All efforts were made to minimize suffering, and all mice were sacrificed for measurement of tumor weights 5 weeks after establishment of the model. To establish the metastasis model in nude mice, a midline incision was made in the anterior abdominal wall, and 2×106 cells (in a total volume of 0.1 ml of PBS) were directly injected into the distal pancreatic parenchyma at the sixth week. Mice were anesthetized with isoflurane inhalation or pentobarbital sodium. After 6 weeks, the liver metastatic ability of the PC cells was observed by harvesting of liver and pancreas tissues. The animal experiments were approved by the Institutional Animal Care and Use Committee of Southwest Hospital, Chongqing, China.

Supplemental materials and methods

The Supplemental Materials are provided in . And the Supplemental Methods, such as the microarray analysis, RNA sequencing analysis, DNA methylation analysis, RNA ISH, fluorescence in situ hybridization (FISH), RNA immunoprecipitation (RIP), Chromatin immunoprecipitation (ChIP) and immunoprecipitation (IP), are provided in . The workflow of the clinical samples used in the present study is provided in .

Results

LINC00261 is downregulated in PC tissue and is associated with survival

To investigate the lncRNA expression profiles in PC, we applied data mining techniques to GEO datasets (GSE15471 and GSE16515) to identify the differentially expressed lncRNAs between PC and NP tissues. Through integrative gene microarray analysis, we screened 6 upregulated lncRNAs and 15 downregulated lncRNAs in PC tissues (). In our previous work, we studied the upregulated lncRNAs 21. Thus, we focused on LINC00261 among the 15 downregulated lncRNAs because it was the only annotated lncRNA associated with both the overall survival and disease-free survival of PC patients according to GEPIA database analysis 22 (Figure ). We further validated LINC00261 expression by qRT-PCR in PC tissues, paired adjacent nontumor tissues, unpaired adjacent nontumor tissues and NP tissues. LINC00261 expression was markedly downregulated in PC tissues compared with nontumor and NP tissues (Figure -C), consistent with the microarray analysis results. Receiver operating characteristic (ROC) curve analysis performed in parallel revealed the potential diagnostic value of this lncRNA in distinguishing the paired tumors from the adjacent nontumor tissues and distinguishing the tumors from NP tissues (Figure -E). Overall survival and disease-free survival were analyzed in TCGA data 22 (Figure -G). We further detected LINC00261 expression in tissue microarrays via RNA ISH. Samples were stratified into high and low expression groups according to the ISH staining score standard (). Survival analysis based on the ISH results revealed that LINC00261 was positively associated with overall survival. Patients in the high LINC00261 expression group had longer median survival times (21.7 months) than those in the low LINC00261 expression group (10 months, Figure ). LINC00261 expression was markedly downregulated in PC tissues, but was expressed at high levels in NP tissues (Figure -J). These results demonstrate that LINC00261 is downregulated in PC tissue and is positively associated with survival.

LINC00261 expression is correlated with progression and poor prognosis in PC

We then analyzed the association between LINC00261 expression and clinical features using the PC samples from the three cohorts underwent ISH. LINC00261 expression was negatively correlated with the pathological differentiation grade, clinical stage and lymph node metastasis in PC patients (all P < 0.05, ). Univariate and subsequent multivariate Cox regression analysis revealed that LINC00261 was an independent protective factor for survival (). To expand the clinical significance of these results, we developed a nomogram based on the four independent covariates (age, clinical stage, differentiation grade and LINC00261 expression level) to predict the survival probability for PC patients (Figure ). The nomogram showed good performance in identifying outcome events (Figure , bootstrap corrected AUC=0.766, 95%CI=0.698-0.834). In addition, the Hosmer-Lemeshow test indicated that the calibration of the nomogram was good (Figure P=0.784). The decision curve showed that if the threshold probability of survival is >20%, the nomogram added more net benefit than the treat-all-patients scheme or the treat-none scheme 23. In addition, the net benefit of the nomogram with LINC00261 included was higher than that of the nomogram without LINC00261 included at threshold probabilities ranging from 55% to 95% (Figure ). These data demonstrate that LINC00261 is significantly associated with favorable clinical characteristics and is an independent protective factor for PC survival.

LINC00261 induces cell cycle arrest and inhibits tumor cell proliferation, migration and invasion in vitro

To investigate the effect of LINC00261 on the malignant phenotypes of PC cells, we first assessed LINC00261 expression in four PC cell lines and performed gain- and loss-of-function studies in PC cells by constructing a LINC00261 overexpression vector and LINC00261-siRNA (). LINC00261 overexpression induced cell cycle arrest and reduced cell proliferation, viability, migration and invasion relative to the corresponding properties in control cells, as shown by the results of the EdU assay, flow cytometric analysis, CCK8 assay, Transwell assay (migration) and Matrigel assay (invasion) (Figure ). Furthermore, the Western blot (WB) results indicated that cell proliferation and the expression levels of the cell cycle-related markers CCND1, CDK4 and CDK6 were significantly decreased in LINC00261-overexpressing cells compared to control cells (Figure ). Simultaneously, the expression level of the epithelial-mesenchymal transition (EMT)-related epithelial marker E-cadherin was significantly increased, while those of N-cadherin, Vimentin and Slug were decreased (Figure ). In contrast, downregulation of LINC00261 enhanced cell cycle progression and increased cell proliferation, viability, migration and invasion relative to the corresponding properties in control cells (Figure ). WB results showed significantly increased expression levels of the cell cycle-related markers CDK4, CDK6 and CCND1 (Figure ) and a markedly decreased expression level of the EMT-related epithelial marker E-cadherin but increased expression levels of N-cadherin, Vimentin and Slug (Figure -H). These results demonstrate that LINC00261 plays an important role in inhibiting the proliferation and EMT-related invasion and migration of PC cells in vitro.

LINC00261 suppresses tumor progression in vivo

We then evaluated the inhibitory effect of LINC00261 on PC progression in vivo. First, we established cell lines with stable LINC00261 upregulation and downregulation via lentiviral transfection. Successful overexpression and knockdown of LINC00261 in CFPAC-1 and PANC-1 cells were validated by qRT-PCR (Figure ). To ascertain the effect of LINC00261 on tumorigenicity in vivo, we subcutaneously transplanted the transfected cells into the dorsal region of nude mice. Overexpression and knockdown of LINC00261 affected tumor growth in vivo (Figure ). Then, stably transfected PC cells were injected into the distal pancreatic tissues of nude mice to assess whether LINC00261 influences liver metastasis of PC cells in vivo. Obviously, one mice (17%) in the LINC00261 overexpression group exhibited liver metastasis, while four mice (67%) exhibited intrahepatic metastasis in the LINC00261 knockdown group (Figure ). Taken together, our data demonstrate that LINC00261 acts as a suppressor of pancreatic tumor progression in vivo.

c-Myc is a key downstream target of LINC00261

Since the molecular function of lncRNA largely depends on its subcellular location 24, we performed FISH and found that LINC00261 was expressed in both the nucleus and cytoplasm, indicating that it may regulate a variety of biological processes (Figure ). To investigate the downstream pathway of LINC00261 in PC, we performed RNA-seq on two lines of PANC-1 cells: cells with LINC00261 downregulation and control cells (each cell line was biologically repeated in triplicate). Then, we performed Gene Set Enrichment Analysis (GSEA, Broad Institute) on our RNA-seq dataset (by dividing the transfected PC cells into the si-LINC00261 group and the si-NC group) and analyzed the top 10 enriched pathways (Figure ). We next performed the same analysis on the TCGA dataset (by dividing the patients into the LINC00261-high group and the LINC00261-low group) for independent validation (Figure ). The combined results of both analyses revealed that the differential expression of LINC00261 was statistically significantly related to Myc, E2F and G2M hallmarks (Figure ). We selected the Myc and E2Fs because they are important transcription factors in PC progression 25 and they both could regulate the cell cycle, migration and invasion from KEGG database (Pathways in cancer). The G2M was excluded as the change in cell cycle is not obvious in Figure . Subsequent cluster analysis showed that LINC00261 knockdown led to alterations in gene expression, including in the Myc and E2F families. Importantly, the gene expression analysis indicated that Myc was differentially expressed with a higher fold change than E2F related genes; thus, we focus on the gene Myc (Figure ). The correlation analyses of TCGA data revealed that LINC00261 was negatively correlated with Myc. Combined with KEGG analysis, other known downstream targets of Myc, such as CDK4, CCND1, MMPs and VEGFA 26-30, were also negatively correlated with LINC00261 ().

Importantly, Kaplan-Meier survival analysis of patient subgroups showed that the group with higher LINC00261 expression and lower Myc expression had the best prognosis (). These results suggest that Myc is a target of LINC00261. To further confirm this finding, we performed qPCR and WB analyses on PC cells with overexpression and downregulation of LINC00261. c-Myc expression was negatively regulated by LINC00261 (Figure -G & ). Similarly, the expression levels of CDK4, CCND1, MMP2, MMP9 and VEGFA were decreased when LINC00261 was overexpressed (Figure ). In contrast, the expression levels of these genes were increased when LINC00261 was downregulated (Figure ). Then, the expression levels of LINC00261 and c-Myc were evaluated in microarrays containing tissues from the same group of patients. LINC00261 was also negatively associated with c-Myc in these PC tissues by the Spearman's test and the Fisher's test (Figure -K). Furthermore, subgroup Kaplan-Meier analysis showed that patients with high LINC00261 expression and low c-Myc expression had the best prognosis (Figure ), consistent with our prior observations in the TCGA dataset. Taken together, these results demonstrate that c-Myc is negatively regulated by LINC00261 and is likely associated with the clinical prognosis of PC patients via interaction with LINC00261.

Regulation of cell proliferation, migration and invasion by LINC00261 is dependent on c-Myc

Next, we used EdU, Transwell and Matrigel assays to investigate whether the LINC00261-induced inhibition of cell proliferation, migration and invasion is dependent on c-Myc. Compared with control cells, PC cells transfected with LINC00261-specific siRNA exhibited enhanced malignant behaviors, namely, proliferation, migration and invasion. However, these increases were significantly reversed in cells cotransfected with LINC00261-specific siRNA, c-Myc-specific siRNA or a c-Myc inhibitor (Figure , C & ). We inversely validated these results by overexpressing LINC00261 alone or with c-Myc. As expected, overexpression of LINC00261 inhibited the proliferation, migration and invasion of PC cells, while these inhibitory effects were reversed when c-Myc was overexpressed (Figure ). These functional assay results are summarized in the histograms (Figure ). The mRNA and protein expression levels of c-Myc downstream targets, including MMP9, MMP2, CCND1, and CDK4, were increased when LINC00261 was downregulated, and this effect was reversed when LINC00261 and c-Myc were simultaneously downregulated (Figure ). In addition, the expression levels of these targets were decreased when LINC00261 was overexpressed, and this effect was reversed when LINC00261 and c-Myc were simultaneously overexpressed (Figure ). Taken together, these results indicate that the inhibitory effects of LINC00261 on the proliferation, migration and invasion of PC cells are dependent on c-Myc.

LINC00261 physically interacts with the p300/CBP complex to epigenetically repress c-Myc transcription

A previous study suggested that lncRNAs epigenetically mediate gene transcription by modifying histone acetylation or methylation levels via interaction with acetylases 31 or methylases 32. Considering that c-Myc transcription is regulated by histone acetylation as a classical mechanism 33, we sought to determine whether this mechanism was involved in LINC00261-mediated regulation. First, we focused on H3K27 acetylation (H3K27Ac) in the c-Myc gene promoter and found that the c-Myc promoter region can undergo H3K27Ac modification, according to analysis of the UCSC genome database (http://genome.ucsc.edu/, Figure ). We constructed the WT c-Myc promoter vector and performed reporter gene assays. LINC00261 overexpression significantly decreased the luciferase activity, and the opposite pattern was observed in the LINC00261 knockdown group (). We performed ChIP-PCR on PC cells, and the results validated high enrichment of H3K27Ac at the c-Myc promoter region (Figure ). We further investigated whether LINC00261 modifies the H3K27Ac level at the c-Myc gene promoter. Consistent with our prediction, the H3K27Ac level at the c-Myc promoter was decreased after LINC00261 overexpression. In contrast, the H3K27Ac level was increased after LINC00261 downregulation (Figure ).

To further investigate whether LINC00261 modifies the H3K27Ac level in the c-Myc promoter by interacting with acetylase or deacetylases, we first used an online tool to predict possible LINC00261-protein interactions (http://pridb.gdcb.iastate.edu/RPISeq/index.html, ). The results of a subsequent RNA immunoprecipitation (RIP) assay verified that LINC00261 could bind to the acetylase p300/CBP but not to HDAC1 or HDAC2 (Figure ). p300/CBP is an important acetylase complex that can regulate the H3K27Ac level 34. We speculated that LINC00261 modifies H3K27Ac through the p300/CBP complex. The ChIP-PCR results showed that overexpression of LINC00261 decreased the H3K27Ac level at the c-Myc promoter, while this decrease was significantly reversed in cells cotransfected with the LINC00261 overexpression vector and the p300/CBP overexpression vector. In contrast, downregulation of LINC00261 in PANC-1 cells increased the H3K27Ac enrichment at the c-Myc promoter, and this increase was significantly reversed in cells cotransfected with LINC00261-specific siRNA and p300/CBP-specific siRNA (Figure ). c-Myc mRNA expression was validated by qPCR, and similar results were observed (Figure ). Furthermore, LINC00261 overexpression decreased but LINC00261 downregulation increased the binding of p300/CBP to the c-Myc promoter (Figure ).

These results indicated that LINC00261 regulates H3K27Ac by interacting with p300/CBP and obstructing its binding to the acetylation site in the c-Myc promoter. Besides, we used EdU, Transwell and Matrigel assays to investigate whether the inhibitory effects of LINC00261 on cell proliferation, migration and invasion are dependent on p300/CBP. The inhibition of these malignant behaviors in CFPAC-1 cells induced by LINC00261 overexpression was reversed by cotransfection of the p300/CBP-specific overexpression vector. In contrast, the enhancement of these malignant behaviors in PANC-1 cells induced by downregulation of LINC00261 was reversed after cotransfection with p300/CBP-specific siRNA (Figure ).

LINC00261 physically binds to the bromo domain of p300/CBP

To further clarify the interacting mechanism of LINC00261 and p300/CBP, we first investigated the potential binding region using the online algorithm CatRAPID (http://service.tartaglialab.com/page/catrapid_group), which rapidly predicts RNA-protein interactions and domains to evaluate the interaction tendency based on secondary structures, hydrogen bonding, and molecular interatomic forces. CatRAPID analysis showed that the 1048 aa-1158 aa region of p300 and the 1084 aa-1194 aa region of CBP are the most likely regions bound by LINC00261 (Figure -L). Importantly, these regions contain critical protein domains, i.e., bromo domains (Human Protein Reference Database). The bromo domain has been reported to be an important functional structure for p300/CBP acetylation 35-37; thus, we speculated that LINC00261 inhibits the effect of p300/CBP by interacting with this domain. We constructed mutant p300/CBP expression plasmids () and evaluated the interaction of mutant p300/CBP with LINC00261 in PC cells and observed lower enrichment in cells transfected with either the mutant p300 vector (del-1048-1158aa) or the mutant CBP vector (del-1084-1194aa) group compared with cells transfected with the WT constructs (Figure ). Furthermore, we found that c-Myc expression was weakened in PC cells transfected with the mutant p300/CBP vector group compared with control (Figure ). In addition, we constructed a mutant LINC00261 vector according to the predicted region to identify the p300/CBP binding sites of LINC00261 region (del-4400bp-4700bp). The RIP assays showed minimal enrichment in the mutant LINC00261 group compared with the WT group (Figure ), consistent with our prior CatRAPID prediction. These findings demonstrate that LINC00261 directly interacts with the bromo domain of p300/CBP and participated in the regulation of c-Myc.

Taken together, these results demonstrate that LINC00261 physically interacts with the bromo domain of p300/CBP and prevents p300/CBP from binding to the acetylation site, which in turn decreases the H3K27Ac level and influenced the c-Myc expression (Figure ).

High methylation levels of the LINC00261 promoter were observed in PC patients and correlated with poor prognosis

High levels of promoter methylation might cause downregulation of lncRNAs in cancer, as previously described 38. To explore whether similar upstream mechanisms are involved in the regulation of LINC00261, we used Methylation EPIC BeadChip data (850K microarray) to evaluate the methylation level of the LINC00261 promoter. The methylation levels of LINC00261 in the gene body and promoter region were evaluated in PC and NP tissues, and we found that the methylation level of LINC00261 in PC tissues was higher than that in NP tissues (Figure ). Furthermore, the only methylation locus cg1279011 in the LINC00261 promoter was highly methylated in the three cohorts (Figure ). Next, we evaluated the expression level of LINC00261 and the methylation level of cg1279011 and observed that LINC00261 was downregulated in PC tissues compared to ANT tissues. Pyrosequencing analysis revealed that the methylation level of cg1279011 was higher in PC tissues than in ANT tissues (Figure ). Moreover, the LINC00261 expression and cg1279011 methylation levels were negatively correlated (Figure ). Patients were stratified into the high- and low-methylation-level groups, and patients in the high-methylation-level group had significantly shorter overall survival times (19.9 months) than those in the low-methylation-level group (37.7 months, Figure ). Next, we treated PC cells with azacitidine (AZA) and decitabine (DAC) and found that the expression level of LINC00261 in the BXPC-3, CFPAC-1, PANC-1 and SW1990 cell lines increased significantly (Figure ). Taken together, these results demonstrate the methylation level of LINC00261 is high in PC cells and that this high methylation level is associated with the prognosis of PC patients. In addition, this high methylation level might cause LINC00261 downregulation in PC.

Targeted demethylation of the LINC00261 promoter inhibits the progression of PC

To validate the relationship between LINC00261 promoter methylation and PC phenotypes, a CRISPR/dCas9 system 17 was used to enhance LINC00261 expression by demethylating its promoter (Figure ). We designed three sgRNAs targeting the cg1279011 site (). After transfection of PC cells with EF1a-Dcas9-Tet1CD-CMV-EGFP/sgLINC00261, we used pyrosequencing analysis to validate the methylation level of cg1279011 and found that it was decreased in CFPAC-1 and BXPC-3 cells (Figure ). In addition, the expression levels of LINC00261 in BXPC-3 and CFPAC-1 cells were increased (Figure ). Off-target effects are a major concern for the practical application of any Cas9-based technique; thus, evaluation of the off-target effects of the dCas9-based demethylation system was necessary. To this end, the top 15 potential off-target loci identified according to the CRISPOR web tool (http://crispor.tefor.net/) were selected for examination of whether their mRNA transcription levels were altered as a result of interference or random demethylation by CRISPR/dCas9. No significant differences in the mRNA expression levels of these potential off-target genes were found between cells transfected with EF1a-Dcas9-Tet1CD-CMV-EGFP/sgLINC00261 and control cells. Collectively, these results confirmed the specific effect of CRISPR/dCas9 editing on LINC00261 promoter demethylation (-C). We then performed EdU, Transwell and Matrigel assays and found that targeted demethylation of the LINC00261 promoter inhibited PC cell proliferation, viability, invasion and migration in vitro (Figure ). In addition, this targeted demethylation suppressed tumor growth (Figure ) and liver metastasis in vivo (Figure ). WB analysis was further performed to evaluate the expression levels of protein markers of EMT and the cell cycle in PC cells. The expression levels of c-Myc, MMP9, MMP2, N-cadherin, vimentin, CCND1, CDK6 and CKD4 were decreased, while that of E-cadherin was increased (Figure ). Taken together, these data reveal that targeted demethylation of the LINC00261 promoter upregulates LINC00261 expression and inhibits PC progression.

Discussion

In the present study, we demonstrated that LINC00261 is downregulated and has potential diagnostic value in PC. Furthermore, we found that LINC00261 represses c-Myc expression by interacting with and blocking the histone acetylase complex p300/CBP to decrease H3K27Ac enrichment at the c-Myc promoter. Downregulation of c-Myc had an important impact on its downstream target genes, eventually leading to the suppression of PC progression. In addition, the high methylation level of LINC00261 contributes to its low expression in PC, and targeted demethylation of the LINC00261 promoter restored the expression and function of LINC00261. These results might provide evidence helping us to further comprehend the molecular function of LINC00261 and inspiring novel diagnostic and therapeutic strategies for PC.

LINC00261 is an intergenic non-protein-coding RNA located on chromosome 20p11.21. In PC, LINC00261 has been suggested to be a prognostic marker 18, but the potential molecular mechanism remains largely unknown 19, 20. We validated the expression of LINC00261 from lncRNA microarray data by comparing data from the public cancer databases GEO and TCGA and showed that LINC00261 is downregulated in PC, consistent with the results of previous studies 18-20. More importantly, we found by analysis of tissue microarrays established from multicenter cohorts that LINC00261 is an independent protective factor for the survival of PC patients. To our knowledge, this study is the first to validate the prognostic value of LINC00261 in PC patients using multivariate analysis in multicenter cohorts with this sample size. Furthermore, the nomogram we developed using the cohort data has potential to guide clinical decisions. Although independent validation in external cohorts is needed to further validate its efficacy.

Competing endogenous RNA activity, such as miRNA sponging, is a well-known mechanism by which LINC00261 affects other cancers, as previously indicated 39. However, LINC00261 is 4924nt long, and we identified multiple protein-binding sites in its sequence, suggesting that LINC00261 might also act as a protein sponge by binding key regulatory molecules and further influencing the functions of their target genes such as CDK4 (). The results suggest a possible mechanism that LINC00261 competitively binds to the c-Myc protein to block its transcriptional regulation on downstream targets. Although this mechanism requires further research, it suggests that the mechanism by which LINC00261 regulates c-Myc function might be complicated.

Considering the extreme importance of c-Myc in PC, many novel therapeutic approaches have been developed or are currently being developed to target c-Myc in order to suppress its function 40. However, studies reporting that c-Myc expression is epigenetically repressed by lncRNAs are scarce. A recent study has shown that the lncRNA PVT1 promotes liver cancer progression by disturbing EZH2-regulated histone methylation of the c-Myc promoter 32. Interestingly, our study revealed that LINC00261 physically interacts with the acetylase p300/CBP and blocks the binding of p300/CBP to the promoter region of c-Myc, leading to downregulation of c-Myc transcription. Importantly, through previous bioinformatic predictions and RIP experiments, we found that LINC00261 mainly bind to the bromo domain of p300/CBP, which mainly affects the biological function of acetylation. And the other transcription factors recruitment and binding domain is not in this region. Thus, we speculated that LINC00261 will not affect the binding of transcription factors to p300/CBP, but more experiments are needed to further verify this problem in the future (). These results indicate the complexity of lncRNA functions in cancer.

In lung cancer, LINC00261 is epigenetically regulated, with high methylation levels at the cg15058464 and cg07003030 loci, resulting in activation of the DNA damage response 16. In contrast, we demonstrated in our study that LINC00261 exhibits a high methylation level at the cg12179011 locus in PC and that this high methylation level is associated with poorer survival. These results suggest that methylation of LINC00261 can vary across tissues and organs. Because the benefits of chemotherapy are limited, patients with advanced PC have few treatment options. However, demethylation therapy 17, 41, 42, especially via the dCas9 system, which can target key tumor suppressor genes, may provide insight for curative therapies for PC. A novel technology reported in 2016 43, targeted demethylation therapy has already shown promise in treating other cancers. After the dCas9 system is constructed and transfected into cancer cells, it can precisely and specifically demethylate regions of the target gene, resulting in substantial activation of tumor suppressor gene expression. Our findings validated this therapy in vitro and in vivo, suggesting that targeted epigenetic therapies using dCas9-sgLINC00261 as a promising epigenetic drug target may be an alternate strategy for the treatment of PC, because this approach can restore the expression of LINC00261 and produce tumor inhibitory effects via c-Myc, ultimately improving the clinical prognosis of patients. Finally, regarding the methylation site cg12179011 that was our focus in the present study, our database prediction results indicated that the LINC00261 promoter contains multiple binding sites for several key transcription factors (). The methylation status at these sites might influence the functions of these transcription factors. Further studies are anticipated to reveal the related mechanisms. These collective results may help to circumvent the long-standing obstacles to PC treatment.

Conclusion

These results indicate that LINC00261 acts as a novel potential tumor suppressor gene in PC through epigenetic inhibition of c-Myc transcription and subsequent suppression of c-Myc-driven PC cell proliferation, viability, migration and EMT-mediated invasion. Furthermore, downregulation of LINC00261 might occur due to the high methylation level of the LINC00261 promoter. Targeted demethylation of LINC00261 restores its expression and tumor inhibitory function. We anticipate that these findings will facilitate improvements in the current status of PC diagnosis and treatment.

Supplementary figures, tables, and methods.

Click here for additional data file.