Introduction

With an anticipated 2.26 million cases reported in 2020, breast cancer (BC) is the most often diagnosed disease in the world. BC is the main cause of cancer mortality in women, considered a serious threat to global health [1]. Genetic factors, hormones, lifestyle, and proliferative breast lesions with atypia are all risk factors for developing BC [2]. It is divided into three main subtypes, including triple-negative, ERBB2-positive, and hormone receptor–positive/ERBB2-negative, depending on the presence or absence of molecular markers for the human epidermal growth factor 2 (ERBB2; formerly HER2), progesterone receptor (PR), and estrogen receptor (ER) [3].

The benefit of transcriptomics has evolved recently thanks to the expansion of high-throughput sequencing methods. It has illustrated the function of non-coding RNAs (ncRNAs), especially long ncRNAs (lncRNAs), in human illness and cellular function [4]. LncRNAs are a class of recently discovered non-coding transcripts with a length of more than 200 nucleotides [5]. Recent studies have defined distinct roles for either particular lncRNAs or the process of lncRNA synthesis, despite the fact that at least some lncRNA production may represent transcriptional “noise” [6]. It has been suggested that lncRNAs play a role in tumorigenesis and tumor metastasis by controlling gene expression at the transcriptional and translational levels [7]. The tumor microenvironment (TME), autophagy, apoptosis, cell cycle, epithelial-mesenchymal transition (EMT), DNA repair, epigenetic alteration, and drug efflux are merely some of the recent studies that have demonstrated the essential and various functions of lncRNAs in BC chemoresistance [8]. Progress in analytical methods gradually made the early detection of BC possible through biomarker scanning in nipple aspirate fluid, sweat, tears, blood, breath, and urine [9]. The cellular damage produced by lncRNAs may be alleviated by targeting them [10]. RNA sequencing (RNA-seq) identified numerous up- and downregulated genes involved in many biological pathways, such as EMT and paths related to mammary gland development [11]. Although RNA-seq presents an unequaled platform for identifying appropriate targets for BC precision medicine, current molecular-based prognostic markers are not the best fit for RNA-seq data, necessitating comprehensive genomic testing [12]. Two major regulators of gene expression, lncRNAs and microRNAs (miRNAs), functionally intercede the mechanisms of pathogenesis. The lncRNA-miRNA axis might have oncogenic or tumor suppressor effects depending on the unique lncRNA/miRNA interaction, making it crucial to define the lncRNA-miRNA axis for evaluating targetability [13]. The identified pathways, genes, Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, and gene ontology (GO) terms may serve as a foundation for further research in BC and the potential creation of new therapies, diagnostics, and prognostic biomarkers [14].

In order to study potential markers associated with disease phenotype and prognosis, we can develop networks of interactions between genes and proteins by analyzing differentially expressed RNAs, such as lncRNAs, miRNAs, and mRNAs. In this study, following a thorough examination of MEG3 lncRNA and its interactions with RNAs, proteins, and medicines, we determined that MEG3 is a crucial gene involved in the invasion, prognosis, and other properties of BC.

Methods

Identifying lncRNA MEG3

The LncRNA and Disease Database version 2.0 (LncRNADisease v2.0) (https://www.cuilab.cn/lncrnadisease) was used to choose lncRNAs associated with BC. Based on patient information from OncoDB (https://oncodb.org/), differentially expressed genes (DEG) enrolling in tumor size, metastasis, and clinical stage of BC were separated. Common lncRNA between all four mentioned criteria was chosen by drawing a Venn diagram using InteractiVenn web-based tool (http://www.interactivenn.net/index.html).

Expression of lncRNA MEG3 in the Cancer Genome Atlas (TCGA) and GTEx Tissues

The GEPIA version 2.0 (http://gepia2.cancer-pku.cn) was used to compare the MEG3 lncRNA expression levels in BC and non-tumor tissues. A web-based tool, Kaplan–Meier plotter (https://kmplot.com/analysis/) and GEPIA were utilized to analyze the expression levels of MEG3 in relation to the prognosis of BC, which includes overall survival, disease-free survival, relapse-free survival, and progression-free survival.

Gene–Gene and Gene-Protein Network of the lncRNA MEG3

RAID database version 2.0 (https://www.rna-society.org/raid2/) was used to analyze MEG3-Gene and MEG3-Protein interactions, and the networks were created using Cytoscape version 3.9.1.

KEGG Pathway and GO Term Analysis

The COXPRESdb database version 8.1 (https://coxpresdb.jp/) was utilized to determine the genes associated with MEG3. Afterwards, Database for Annotation, Visualization and Integrated Discovery (DAVID) version 6.8 (https://david.ncifcrf.gov/) was used to carry out KEGG pathway and GO analyses of MEG3-associated genes.

Validation of MEG3 Expression Profiling in the GSE Dataset

Gene expression omnibus (GEO) dataset GSE2219 was utilized to confirm differential expression of the MEG3 in BC cell lines and mammary epithelial cells. Using the GEO2R web-tool (https://www.ncbi.nlm.nih.gov/geo/geo2r/), volcano plot of the DEGs included in the GSE1299 dataset was displayed.

Enrichment Analysis of MEG3

Correlation between MEG3 and cancer hallmarks and interactions with drugs were investigated using LncSEA version 2.0 (https://bio.liclab.net/LncSEA/index.php). Using OncoDB, the correlation between MEG3 and clinical parameters was investigated.

Results

Identifying lncRNA MEG3

Surveying the LncRNADisease v2.0, 200 distinct lncRNAs were discovered associated with BC. On the other hand, the OncoDB revealed 1026, 1673, and 386 DEGs in BC patients related to different tumor sizes, metastasis stages, and pathological stages, respectively. Based on the Venn diagram (Fig. 1a), MEG3 was chosen as the common lncRNA between these four criteria.

Fig. 1
figure 1

Selecting the common lncRNA related to BC clinicopathological traits and expression levels of MEG3 lncRNA. a MEG3 was identified as the common gene, using the InteractiVenn web-based tool. Each digit in the figure is related to the number of common genes between different sections. b Expression levels of MEG3 in BC (1085) and in normal breast (291) tissues indicate a downregulated level of its expression in BC. c Expression of MEG3 in different clinical stages of BC represents a significant difference. Red: BC tissues. Blue: normal breast tissues. Black dots: individual cases

Full size image

Expression and Survival Analysis of lncRNA MEG3

MEG3 expression in BC tissues was lower than in normal breast tissues, according to the results of gene expression in TCGA and GTEx samples (Fig. 1b). There was a significant difference between the clinical stages of BC and the levels of MEG3 expression (Fig. 1c; P = 0.03). Lower MEG3 expression levels were not shown to be substantially related to a worse prognosis for BC patients’ survival (Fig. 2a–f).

Fig. 2
figure 2

Kaplan–Meier survival curves based on MEG3 lncRNA expression level. a There is no significant relation between the overall survival rate of patients with BC and lncRNA MEG3 expression levels. b MEG3 expression level has no noticeable effect on the disease-free survival rate of patients with BC. c–f There is no significant relationship between MEG3 lncRNA expression level and c overall survival rate of patients with BC, d relapse-free survival rate of patients with BC, e disease-specific survival rate of patients with BC, and f progression-free survival rate of patients with BC

Full size image

Gene–Gene and Gene-Protein Network of the MEG3 lncRNA

Based on the RAID database, 73 miRNAs, 3 mRNAs, 2 lncRNAs, and 20 proteins had interactions with MEG3 (Fig. 3a and b). The most significant interactions of MEG3 were with TP53 mRNA, TUG1 and CRNDE lncRNAs, has-miR-140-5p miRNA, and TP53 protein, based on the combined score given to each interaction on the RAID database.

Fig. 3
figure 3

RAID analysis presented by Cytoscape and GO and KEGG analysis. a Using Cytoscape 3.9.1, a gene–gene interaction network and b a gene-protein interaction network for the MEG3 was created. Protein-coding genes are represented by blue rectangles, lncRNAs by orange ovals, and miRNAs by green diamonds. Symbolizing proteins are blue circles. Highest combined score is represented by edges with thicker and deeper hue (yellow to red). c GO term and d KEGG pathways enrichment plots of the MEG3 lncRNA and its associated genes enriched the cell adhesion, extracellular region, extracellular matrix structural constituent, and PI3K-Akt signaling pathway as biological process, cellular component, molecular function, and KEGG pathway, respectively. Deeper color (red to blue) bars represent higher log10 (P value)

Full size image

KEGG Pathway and GO Term Analysis

Using the COXPRESdb database, the top 200 genes associated with MEG3 were determined. Eight biological processes (BP), 6 cellular components (CC), and 8 molecular functions (MF) were significantly enriched, according to the GO enrichment results (Fig. 3c). Three KEGG pathways were considerably enriched, according to KEGG enrichment analyses (Fig. 3d). The most significant BP, CC, MF and KEGG pathway were cell adhesion (GO:0007155, P = 1.9E − 9), extracellular region (GO:0005576, P = 1.2E − 14), extracellular matrix structural constituent (GO:0005201, P = 7.6E − 22), and PI3K-Akt signaling pathway (hsa04151, P = 6.2E − 6), separately.

Enrichment Analysis of MEG3

In order to analyze drug interactions and cancer hallmarks, the LncSEA database was employed. The results revealed strong interactions between cisplatin and temozolomide with MEG3 as well as the major contribution of MEG3 with cancer hallmarks (Table 1). Based on OncoDB database, MEG3 is significantly downregulated in BC patients and significantly affected the tumor size, pathological stages, and metastasis stages (Fig. 4a).

Table 1 Enrichment analysis of MEG3 in BC patients using LncSEA database
Full size table
Fig. 4
figure 4

Correlation between the expression of lncRNA MEG3 and different metastasis, tumor size, and pathological stages of patients with BC and volcano plot of DEGs in GSE1299. a MEG3 expression decreased nearly 2.5-fold in BC samples than normals, and this reduction significantly affected the tumor sizes, pathological stages, and metastasis stages of BC. b DEGs between mammary epithelial cells and BC cell lines have been identified. There were 2396 DEGs in GSE1299. Among them, 530 and 600 genes were significantly (|log2FC|≥ 1) up- and downregulated. MEG3 is one of the significantly downregulated genes in this dataset and was represented with the yellow star. The red dots represent upregulated genes, and the blue ones represent downregulated genes. Black dots represent the insignificant data

Full size image

Validation of MEG3 Expression Profiling in the GSE Dataset

The volcano plot of the DEGs found in the GSE1299 dataset was displayed in Fig. 4b. Based on these findings, the expression of MEG3 in BC cell lines was lower compared to mammary epithelial cells (adj. P-value = 0.01, log2(fold change) =  − 0.955).

Discussion

BC is the most frequent cancer diagnosed globally, and its incidence has increased over the past few decades [15]. Developing precise diagnostic and prognostic biomarkers may increase the survival rates of BC patients. LncRNAs can act alone or in association with miRNAs and other molecules as part of multiple pathways, to stimulate or repress gene expression [16]. There is expanded evidence that lncRNAs play crucial roles in numerous biological processes, including the initiation and progress of cancer [17].

In this study, the LncRNADisease v2.0 was used to choose lncRNAs associated with BC. Using patients’ data from OncoDB, DEGs in terms of tumor size, metastasis, and pathological stage were isolated. Using the InteractiVenn web-based application, the MEG3 was selected as the common lncRNA between the four criteria.

MEG3 is a lncRNA located on chromosome 14q32.3 [18] and modifies the expression of target genes through transcription, translation, post-translational alterations, and epigenetic regulation [19]. Several primary human tumors exhibit downregulated MEG3 expression, which suggests that it could serve as a tumor suppressor [20]. According to the research by Wang et al., MEG3 is downregulated in colorectal cancer [21]. Overexpression of MEG3 lncRNA suppresses cell proliferation and invasion and induces apoptosis in ovarian cancer by sponging miR-205-5p [22] and its abnormal expression prognosticates low survival in ovarian cancer [23]. This lncRNA also prevents the development of non-small cell lung cancer by inhibiting telomere function, cell proliferation, telomerase activity, cell migration, and invasion through control of the expression of the DKC1 protein [24]. Moreover, the downregulation of MEG3 plays a significant role in the development of hepatocellular carcinoma [25], prostate [26], and ovarian cancer [27, 28] and findings point to the possibility of MEG3 expression as a prognostic marker and potential immunotherapeutic target for gliomas [29] bladder cancer [30] and cervical cancer [31]. Besides, multiple studies indicate the downregulation of MEG3 in BC which is consistent with our findings. For BC patients, the lower MEG3 levels is a diagnostic and unfavorable prognostic feature [32, 33].

Methylation of the MEG3 promoter by DNMT1 activity downregulates the MEG3 expression and promotes the malignant behavior of BC cells due to the freedom of miR-494-3p from the MEG3 trap [34]. Zhang et al. proposed that MEG3 inhibits the growth of BC and triggers apoptosis through the activation of the ER stress, NF-κB, and p53 pathways where NF-κB signaling is necessary for MEG3-induced p53 activation in BC cells [35]. As we proposed, the MEG3 level may affect the tumor sizes, metastasis, and pathological stages of BC. According to Zhang et al. [36], the expression rate of this lncRNA is associated with lymph node metastasis, various grades, and the TNM stage of BC. Although our results indicate no noticeable prognostic value for MEG3 in BC, it is potentially related to poor overall survival, progression-free survival, and 5-year survival rate, based on the research [36, 37]. MEG3 level may also differentiate between the triple-negative status of BC—negatively correlated with MEG3 expression level—and estrogen or progesterone receptor–positive status—which are positively related [37]. Though, our results show no significant differences between various grades of BC in the MEG3 rate.

In mining the RAID database, we achieved the most significant interactions of MEG3 with TP53 mRNA, TUG1 and CRNDE lncRNAs, has-miR-140-5p miRNA, and TP53 protein. TP53, an effective tumor suppressor gene, frequently targeted by tumorigenesis processes, and MEG3, like some other lncRNAs, interferes in regulating its expression in BC [38]. This lncRNA affects the transcriptional level of MDM2 and stabilizes TP53 to prevent the progression of BC [39]. As in BC [35], MEG3 induces apoptosis in hepatocellular carcinoma by promoting the NF‑κB-related ER stress and p53 pathway [40]. MEG3 is also correlated with TP53 by regulating the FOXP3 expression and upregulating the miR-149-3p [41]. Sequestration of miR-140-5p with MEG3 has a confirmed role in the osteogenic differentiation of human adipose tissue–derived stem cells (hASCs) [42] which authenticates the competing endogenous task of MEG3 about the miR-140-5p, as our study proposed. This correlation also protects the number of circulating endothelial progenitor cells (EPCs) in metabolic syndrome [43]. TUG1 expression indicates a coordinated rhythm with MEG3 expression in cisplatin-resistant tissues [23] and diabetic patients [44, 45]. It may present a close relationship between these two lncRNAs.

Gene Ontology analysis predicted that MEG3 participates in cell adhesion, and most of its target genes are constituents of the extracellular matrix. MEG3 is one of the genes that regulate epithelial-mesenchymal transition (EMT), so its downregulation induces EMT in C2C12 myoblasts and inhibits myogenesis [46]. This lncRNA represses the intracellular cell adhesion molecule-1 (ICAM-1) expression by sponging the miR-147 to reduce cell viability and migration [47]. miR-147 has a target site in the promoter of the ICAM-1 gene and positively regulates its expression [47].

Due to the importance of PI3K/Akt signaling in controlling the main biological processes like cell growth, apoptosis, and migration [48], this pathway is exploited by multiple genes, including MEG3, to interfere with tumorigenesis [48]. Zhu et al. indicated that MEG3 acts as a tumor inhibitor by reversing miR-21-mediated activation of PI3K/Akt pathway in BC cells [48]. The inhibitory interference of MEG3 in the PI3K/AKT signaling is demonstrated in hepatocellular carcinoma [49], inflammatory disorders [50], and diabetic retinopathy (DR) [51]. Consistent with them, we have also predicted the PI3K/AKT signaling as the crucial pathway exploited by MEG3 to prevent the BC progression based on the KEGG analysis.

LncSEA data guided our study toward a significant interaction between MEG3 and cisplatin. This interaction is not strange because, in triple-negative breast cancer (TNBC), cisplatin (DDP) utilizes MEG3 expression to exert its anti-tumor effect [52]. DDP-related upregulation of MEG3 promotes the NLRP3/caspase-1/GSDMD pyroptosis pathway and sensitizes TNBC to cisplatin [52]. The low expression level of MEG3 in cisplatin-resistant cells of non-small cell lung cancer (NSCLC) is another proof demonstrating the effect of MEG3 on the tumor cell response to cisplatin [53].

Some obstacles should be mentioned despite the scientific relevance of the findings. First, the results are obtained using organic bioinformatics techniques and are entirely reliant on the TCGA database. Even if certain conclusions had been confirmed by the GEPIA database, the discoveries needed to be evaluated. Second, a typical follow-up period should be used to examine the prognostic signature of MEG3 (three to 5 years).

Conclusion

According to the findings, MEG3 may have a significant role in critical pathways such as apoptosis and interactions with crucial genes and proteins, as well as in the prognosis, proliferation, migration, invasion, and metastasis of BC. More clinical data is necessary to determine the precise role of MEG3.