The Role of Long Non-coding RNAs in Cancer Metabolism: A Concise Review

Dysregulation of metabolic pathways in cancer cells is regarded as a hallmark of cancer. Identification of these abnormalities in cancer cells dates back to more than six decades, far before discovery of oncogenes and tumor suppressor genes. Based on the importance of these pathways, several researchers have aimed at modulation of these functions to intervene with the pathogenic course of cancer. Numerous genes have been shown to participate in the regulation of metabolic pathways, thus aberrant expression of these genes can be involved in the pathogenesis of cancer. The recent decade has experienced a significant attention toward the role of long non-coding RNAs (lncRNAs) in the biological functions. These transcripts regulate expression of genes at several levels, therefore influencing the activity of cancer-related pathways. Among the most affected pathways are those modulating glucose homeostasis, as well as amino acid and lipid metabolism. Moreover, critical roles of lncRNAs in regulation of mitochondrial function potentiate these transcripts as novel targets for cancer treatment. In the current review, we summarize the most recent literature regarding the role of lncRNAs in the cancer metabolism and their significance in the design of therapeutic modalities.


INTRODUCTION
Altered metabolic pathways in cancer has been attracting researchers for more than six decades when Warburg hypothesized that the tumorigenesis process is initiated by a deficient cellular respiration due to the mitochondrial function impairment (1). This research area remarkably precedes the identification of the role of oncogenes and tumor suppressors in the carcinogenesis (2). While normal cells obtain energy principally via mitochondrial oxidative phosphorylation (3), cancer cells can fulfill the requirements of their fast and uncontrolled proliferation by excessive glycolysis and the subsequent lactic acid fermentation even in the existence of plentiful oxygen supply. This kind of aerobic glycolysis has been characterized as the Warburg effect (4). Carcinogenesis process is accompanied by the extensive synchronized activation of metabolic pathways that maintain this process by dysregulation of PI3K-AKT-mTOR signaling pathways, deficiency of tumor suppressors, and induction of oncogenes (2). The altered metabolic functions in the cancer cells have been shown to facilitate the attainment and preservation of malignant features. Since some of these characteristics have been detected rather commonly across many kinds FIGURE 1 | (A) UCA1 can enhance expression of AKT1 and activate it through phosphorylation. AKT1 activates mTOR, thus facilitating STAT3 phosphorylation and nuclear translocation. These events lead to over-expression of HK2. In addition, mTOR inhibits miR-143 which targets HK2. Therefore, mTOR has a prominent effect in enhancing HK2 expression (21). (B) HK2 is located in the mitochondrial membrane and facilitates conversion of glucose to glucose-6-phosphte. Glucose-6-phosphate enters the pentose phosphate pathway to produce pentose phosphate for nucleic acid synthesis which is needed for proliferation of cancer cells. In addition, glucose-6-phosphate can be converted to pyruvate to enter tri-carboxylic acid (TCA) cycle and produce citrate which is needed for the synthesis of membrane components (phospholipids and cholesterol), thus being necessary for tumor proliferation (22). of cancer cells, altered metabolic function is regarded as a hallmark of cancer (5). This aberrant metabolic function facilitates anabolic growth in the course of nutrient-depleted situations, catabolism to sustain cell survival for the period of nutrient insufficiency, and protection of redox homeostasis to neutralize the metabolic influences of oncogene activation, tumor suppressor deficiency or other cellular stresses (6). Such metabolic reprogramming involves several genes and molecular pathways among them are long non-coding RNAs (lncRNAs) (7). These transcripts comprise a large proportion of human transcriptome, have sizes larger than 200 nucleotides and share several features with mRNA coding genes; yet, they lack considerable open reading frames (8). Not only can they regulate cell proliferation, cell death, migration, invasion and stemness properties (9), but also they have critical roles in the regulation of cancer metabolism (7). The latter has been supported by a bunch of evidence which reports aberrant expression of metabolismrelated lncRNAs in cancer cells. Moreover, functional studies have verified their roles in the context of cancer in some cases. The current review has focused on the role of lncRNAs in cancer metabolism and provides key examples in this regard. Based on the ever growing literature on this topic, this review cannot provide the exhaustive list of all related researches.

MECHANISMS OF lncRNAs FUNCTIONS IN REGULATION OF GENE EXPRESSION
LncRNAs can exert their regulatory functions through different mechanisms such as modulation of chromatin structure and DNA methylation status and interacting with transcription factors and DNA motifs, thus regulating transcription of target genes. They also influence mRNA processing to affect gene expression at post-transcriptional level. Besides, their interactions with certain proteins enable them to regulate protein translation and post-translational alterations such as phosphorylation and ubiquitination (10). These transcripts can function as miRNA sponges to modulate expression of miRNA target genes or serve as precursors for miRNA or small interfering RNAs (10). LncRNAs can also modulate alternative splicing processes and consequently modulate spatial and temporal expression of genes (10).

ONCOGENIC lncRNAs THAT REGULATE CANCER METABOLISM
Oncogenic lncRNAs regulate different aspects of cancer metabolism such as glutaminolysis and lipid metabolism. For instance, UCA1 has an established role in the regulation of glutamine metabolism (11). Expression of this lncRNA in bladder cancer tissues and cell lines is significantly correlated with GLS2 expression. Moreover, up-regulation of UCA1 enhances GLS2 expression and increases glutaminolysis in these cells. This function has been mediated through sponging miR-16, a miRNA that targets GLS2 (11). Besides, the oncogenic lncRNA CCAT2 has been shown to alter glutamine metabolism in colon cancer (12). Functional studies revealed the interaction between CCAT2 and CFIm complex, a protein complex that modulates the alternative splicing and the poly(A) site choosing in GLS transcript, leading to the privileged expression of the more aggressive variant GAC (12). The lncRNA PCGEM1 has been shown to enhance glucose entry in the cells to increase aerobic glycolysis, coupling with the pentose phosphate shunt to enhance lipogenesis in prostate cancer cells (12). Several other up-regulated lncRNAs in the cancer cells have been shown to alter cancer metabolism. The metabolisminduced tumor activator 1 (MITA1) is an lncRNA which has been shown to be over-expressed in hepatocellular carcinoma (HCC) and participates in the metastatic potential of these cells. This lncRNA is remarkably activated by energy stress. This process is controlled by the LKB1-AMPK pathway and DNA methylation (13). Another experiment in the HCC cells has shown correlation between expressions of the lncRNA RAET1K and both HIF1A and miR-100-5p. LncRNA RAET1K has been shown to act as a molecular sponge for miR-100-5p, thus inhibiting its expression. On the other hand, HIF1A binds with the promoter region of lncRNA RAET1K to enhance its transcription. LncRNA RAET1K knock down has inhibited proliferation and invasion of HCC cells and also overturned hypoxia-induced upsurge in lactate levels and glucose uptake. Functional studies have verified the role of the HIF1A/lncRNA RAET1K/miR-100-5p axis in regulation of hypoxia-induced glycolysis in HCC cells (14). PVT1, as an up-regulated lncRNA in HCC tissues and cell lines, can directly interact with miR-150 to suppress its expression and subsequently up-regulating expression hypoxia-inducible protein 2 (HIG2) which is targeted by the miR-150. The PVT1/miR-150/HIG2 axis has been shown to regulate iron metabolism in HCC cells (15). In gastric cancer cells, LINC00152 has been shown to modulate aerobic glycolysis through modulation of miR-139-5p and PRKAA1 expressions (16). LINK-A, the upregulated lncRNA in the glioma cells, has been shown to regulate expression of lactate dehydrogenase A (LDH-A), thus enhancing glycolysis and proliferation in these cells (17). Expression of HOTAIR has been increased in hepatocellular carcinoma samples. Its expression has been enhanced under hypoxia condition. Its silencing suppressed glycolysis in these cells. Functional studies verified the role of HOTAIR as a molecular sponge for miR-130a-3p. This miRNA has been shown to inhibit expression of HIF1A. HOTAIR silencing inhibited glycolysis through modulating miR-130a-3p and HIF1A in HCC cells under hypoxic conditions (18). This lncRNA regulates cancer metabolism in pancreatic adenocarcinoma cells as well. Up-regulation of HOTAIR enhances lactate synthesis, glucose uptake and ATP synthesis. Besides, it increases HK2 expression, while HK2 upregulation had no remarkable influence on HOTAIR expression amounts. HOTAIR has been shown to enhance cancer cell energy metabolism in this kind of cancer through increasing HK2 expression (19). UCA1 has been shown to increase mitochondrial function in bladder cancer cells. This lncRNA acts as a molecular sponge for miR-195 to control mitochondrial function through enhancing expression of ARL2. The role of UCA1 through miR-195/ARL2 axis in promotion of bladder tumor growth has been verified in animal models (20). Figure 1 shows a summary of role of UCA1 in the regulation of cancer metabolism. Table 1 summarizes the role of oncogenic lncRNAs in the cancer metabolism in all kinds of human malignancies.

TUMOR SUPPRESSOR lncRNAs THAT REGULATE CANCER METABOLISM
A number of studies have assessed the association between tumor suppressor lncRNAs and metabolic pathways. NEF as a downregulated lncRNA in NSCLC tissues has been shown to regulate cell proliferation and glucose uptake in these cells through modulation of GLUT1 expression. Thus, this lncRNA can target glucose transportation to suppress lung tumorigenesis (76).  Poor prognosis (88) LINC01537 as another tumor suppressor lncRNA has been demonstrated to enhance cellular sensitivity to nilotinib. This lncRNA also targets phosphodiesterase 2A (PDE2A) and enhance it expression through RNA-RNA interaction. Based on the role of PDE2A in energy metabolism, Warburg effect and mitochondrial respiration, LINC01537 has been identified as a regulator of cancer metabolism (77). The lncRNA GASL1 has been shown to enhance Bcl-2 expression, while down-regulating GLUT-1 expression. Thus, the role of this lncRNA in suppression of proliferation of prostate cancer cells has been exerted through modulation of metabolism (78). LINC01554, the down-regulated lncRNA in hepatocellular carcinoma has been shown to be suppressed by miR-365a. This lncRNA enhances the ubiquitinmediated destruction of PKM2 and suppresses Akt/mTOR signaling pathway to stop aerobic glycolysis in hepatocellular cancer cells (79). FILNC1 has been identified as an energy stress-induced lncRNA. FILNC1 silencing in renal cancer cells lessens energy stress-induced apoptosis and considerably induces progression of this type of cancer. Notably, FILNC1 silencing increases glucose uptake and lactate synthesis via induction of c-Myc. In energy stress conditions, this lncRNA binds with AUF1, a c-Myc interacting protein. Thus, it prevents AUF1 from binding with c-Myc transcript, resulting in under-expression of c-Myc protein (80). Expression of the lncRNA HAND2-AS1 has been decreased in osteosarcoma tissues and serum samples of the affected patients compared with control samples. There was a significant association between serum levels of this lncRNA and tumor size. Notably, in vitro studies revealed that HAND2-AS1-silencing enhances osteosarcoma cell proliferation, upsurges glucose uptake and increases GLUT1 levels. Thus, HAND2-AS1 has a tumor suppressor role in osteosarcoma through modulating glucose metabolism (81). GATA6-AS is another tumor suppressor lncRNA that regulates expression of GLUT1. Up-regulation of this lncRNA has reduced glucose uptake and GLUT1 expression in the mantle cell lymphoma. Thus, the lncRNA GATA6-AS might suppress cancer cell proliferation through decreasing GLUT1 expression (82). The lncRNA MORT has a similar role in suppression of glucose uptake and GLUT1 expression in prostate cancer cell lines (83). In prostate cancer cells, up-regulation of GASL1 has enhanced Bcl-2 expression and decreased GLUT-1 levels (78). CASC8 is also involved in the regulation of the glycolysis in bladder cancer cells through modulating expression of the fibroblast growth factor receptor 1 (FGFR1). The interaction between this lncRNA and FGFR1 has been shown to suppress FGFR1-associated lactate dehydrogenase A phosphorylation, which decreases the lactate synthesis from pyruvate (84). Table 2 summarizes the role of tumor suppressor lncRNAs in the cancer metabolism.   (69). Several other lncRNAs that regulate cancer metabolism have been identified as diagnostic/prognostic markers in cancer. Table 3 summarizes the results of studies which reported diagnostic/prognostic significance of these lncRNAs.

DISCUSSION
The carcinogenesis process is associated with high glucose uptake, lactate over-production, aerobic glycolysis as well as glutamine and lipid metabolism (89). The above-mentioned data support the role of lncRNAs in these metabolic pathways in the context of cancer. Notably, HCC has been the most investigated cancer type regarding the role of lncRNAs in the metabolic pathways. Apart from the function of lncRNAs in this regard, metabolic changes have been previously recognized to evidently distinguish HCC tumors. Several clinical parameters that are presently utilized to evaluate liver functions reveal alterations in both enzyme activity and metabolites. Actually, alterations in glucose and acetate consumption are regarded as effective clinical means for classification of patients with HCC. Besides, elevated serum lactate can differentiate HCC from healthy individuals, and serum lactate dehydrogenase is applied as a determinant of prognosis of HCC patients under therapeutic regimens (90). Thus, it is not surprising that the role of lncRNAs has been vastly assessed in this context. The underlying mechanism of participation of lncRNAs in the regulation of metabolic pathways has been clarified in several cases. Glucose transporters (GLUTs) as important modulators of glucose utilization which are commonly dysregulated in cancer (91), have been shown to be targeted by several lncRNAs such as LINC01638, Ftx, XIST, YIYA (LINC00538), HISLA, AWPPH, and UCA1. Most notably, several lncRNA/ miRNA/mRNA comprising axes have been shown to modulate cancer metabolism. Therefore, the complex interactions between these trios should be considered in the design of any therapeutic option. Moreover, numerous lncRNAs have direct or indirect interactions with the well-known oncogene c-Myc. This oncogene is an important modulator of pathways that regulate metabolism of glucose, glutamine and lipid in cancer (92). Thus, all of these lncRNAs are putative regulators of different aspects of cancer metabolism. Several oncogenic lncRNAs mainly exert their effects through modulation of these pathways. Thus, modulation of expression of these lncRNAs through application of antisense oligonucleotides or CRISPR/Cas9-based modalities can be regarded as a therapeutic option in cancer. Yet, the main obstacles in this regard are their off-target effects or unstable efficiency resulting from the space-time related features of lncRNAs (93). Small interfering (si)RNA-mediated silencing of oncogenic lncRNAs has been hampered by unavailability of efficient delivering systems. Yet, this such hurdle has been rather solved by the advent of biocompatible nanoparticle delivery systems (86).
The relevance of metabolism-associated lncRNAs in the treatment of cancer has been highlighted by a number of studies. For instance, the lncRNA-UCA1 has been shown to modulate radioresistance in cervical cancer cell through the HK2/glycolytic pathway (53). The same lncRNA has been shown to be upregulated in AML patients after Adriamycin (ADR)-based chemotherapy. UCA1 silencing has enhanced the cytotoxic effect of this chemotherapeutic agent and suppressed the HIF-1α-associated glycolysis in ADR-resistant AML cells. Based on these results, UCA1 has been shown to exert a positive role in conquering the chemoresistance in pediatric AML patients (72).
Several lncRNAs such as NEF, HISLA, MEG3, PVT1, HOTTIP, SNHG3, LINC00689, H19, HOTAIR, GASL1, NRCP, LINRIS, and FILNC1 have been identified as predictive markers for OS or DFS of cancer patients. The potential of a number of lncRNAs including AWPPH, NEF, HAND2-AS1, lncRNA-RP11-555H23.1, and GASL1 as diagnostic markers in cancer patients has also been verified. These data suggest the importance of these lncRNAs in diverse aspects of cancer biology.
Taken together, regulation of cancer metabolism is a critical role of lncRNAs which has been shown by several in vitro investigations and a number of clinical studies. Thus, these transcripts are putative therapeutic targets in cancer. The importance of this function of lncRNAs is further highlighted by the eminent role of tumor microenvironment in the evolution of cancer and the ubiquitous dysregulation of metabolism in different cancer types. Thus, therapeutic targeting of these lncRNAs can be applied in diverse cancer types.