Current advancements and future perspectives of long noncoding RNAs in lipid metabolism and signaling

Graphical abstract

Highlights

• •Current advancements of lncRNAs in lipid metabolism were elaborated clearly.
• •The function and mechanism of lncRNAs in lipid signaling were reviewed.
• •Several promising lncRNA-based therapies were also reviewed.
• •The challenges about targeting lncRNAs were also summarized.
• •Future perspectives of lncRNAs in lipid metabolism and signaling were elaborated.

Abstract

Background

The investigation of lncRNAs has provided a novel perspective for elucidating mechanisms underlying diverse physiological and pathological processes. Compelling evidence has revealed an intrinsic link between lncRNAs and lipid metabolism, demonstrating that lncRNAs-induced disruption of lipid metabolism and signaling contribute to the development of multiple cancers and some other diseases, including obesity, fatty liver disease, and cardiovascular disease.

Aimof Review

The current review summarizes the recent advances in basic research about lipid metabolism and lipid signaling-related lncRNAs. Meanwhile, the potential and challenges of targeting lncRNA for the therapy of cancers and other lipid metabolism-related diseases are also discussed.

Key Scientific Concept of Review

Compared with the substantial number of lncRNA loci, we still know little about the role of lncRNAs in metabolism. A more comprehensive understanding of the function and mechanism of lncRNAs may provide a new standpoint for the study of lipid metabolism and signaling. Developing lncRNA-based therapeutic approaches is an effective strategy for lipid metabolism-related diseases.

Introduction

As one of the three major nutrients, lipid contributes greatly to cellular and organismal functions. Lipids not only functions as the structural element of plasma membranes and play a critical character in energy metabolism, but also act as signaling molecules, affecting membrane fluidity, or mediating post-translational modifications [1], [2]. Emerging studies have provided evidence that lipid metabolism and signaling are closely correlated with diseases such as cancer [3], obesity [4], cardiovascular disease [5], and aging [6]. Therefore, exploring the underlying regulatory mechanisms of lipid metabolism and signaling, and targeting the critical regulatory molecules will be beneficial for disease treatment.

With the continuous development of new research fields, researchers have realized that lipid metabolism is not just a self-regulation network. Technical developments in the human genome have revealed that noncoding RNAs are a new class of regulators in lipid metabolism [7], [8], [9]. Among all types of noncoding RNAs, lncRNAs have received great attention because of their broad biological functions. Depending on their genomic regions, lncRNAs can be classified into five categories: (1) intergenic, lncRNAs located between the protein-coding genes, which are modified by epigenetic marks similar to those of protein-coding genes; (2) intronic, lncRNAs located in the introns of protein-coding genes, most of which can regulate the transcription or alternative splicing of coding genes; (3) sense, lncRNAs that are transcribed from the sense strand of protein-coding genes’ exonic regions. They contain an ORF but cannot encode proteins; (4) antisense, lncRNAs that are transcribed from the antisense strand of a protein-coding gene; and (5) enhancer, lncRNAs can be transcribed and have H3K4me1 marks on the promoter regions (Fig. 1 A). Assorted lncRNAs participate in all aspects of life activity [10], [11], [12], [13], by serving as a molecular signal [14], [15], decoy [16], guide [17], [18], [19], scaffold [20] and enhancer [21] (Fig. 1 B).

Fig. 1.

Classical categories and general mechanisms of lncRNAs. A. According to their genomic region, lncRNAs can be divided into five class, comprising intergenic, intronic, sense, antisense, and enhancer lncRNAs. Intergenic lncRNAs are located between the protein-coding genes, and their transcription activation and methylation marks are similar to protein-coding genes. Sense lncRNAs are transcribed from the sense strand of protein-coding genes’ exonic regions. They contain an ORF but cannot encode protein. In contrast to sense lncRNAs, antisense lncRNAs are transcripts from the antisense strand of a protein-coding gene. Intronic lncRNAs are located in the introns of protein-coding genes, and most of them can regulate the transcription of coding genes or regulate alternative splicing of coding genes. In the protein-coding genes enhancer regions, enhancer lncRNAs can be transcribed and have H3K4me1 marks on the promoter regions. B. By serving as the molecular signal, decoys, guides, scaffolds, and enhancers, lncRNAs regulate the expression and function of multiple molecules. LncRNA p21 is transcribed as a signal molecular to trigger apoptosis. GAS5 binds with the glucocorticoid receptor as a molecular decoy. Some lncRNAs can bind with target proteins and guide them to the proper localization. ANRIL can bind with PRC1 and PRC2 directly. The transcriptional repression of the target INK4b locus would be impacted if disrupt either interaction. HOTTIP activates the HOXA distal genes through recruiting the WDR5 subunit of the MLL complex and reducing the H3K4me3 mark.

The function of lncRNAs is also determined by their subcellular localization. In the cytoplasm, lncRNAs can regulate the stability as well as the translation of target gene mRNAs. Moreover, as competing endogenous RNAs, lncRNAs in the cytoplasm can also influence the distribution of miRNAs on their targets, thereby regulating the expression of target genes. In the nucleus, lncRNAs can participate in transcriptional regulation in a variety of ways. They can recruit chromatin-modifying complexes to specific genomic loci thereby controlling the epigenetic status of target genes. Thus lncRNA MANTIS can interact with SWI/SNF chromatin remodel factor BRG1 and enable the binding of BRG1 to the promoter of angiogenic genes, like SOX18, to mediate vascular protection [22], or can regulate the expression of transcription factors or DNA binding [23]. Furthermore, lncRNAs in the nucleus can also act as co-activators/repressors or localize to different subdomains for transcriptional regulation. In addition to intracellular lncRNAs, extracellular vesicle-derived lncRNAs have been found to function extracellularly. Notably, lncRNAs from tumor-derived exosomes, such as lncRNA UAC1 and MAGEA3, have emerged as potential biomarkers for monitoring cancer progression [7].

With the development of sequencing technologies, more and more novel lncRNAs have been discovered and the relationships between lipid metabolism-related diseases and lncRNAs have been revealed. For example, to discover lncRNAs involved in fat cell lipid metabolism, the differential-expressed lncRNAs in white adipose tissue of female obesity patients, with or without undergoing Roux-en-Y gastric bypass surgery, were identified via microarray analysis. Changes in lncRNA expression were associated with the alteration in fat mass and fat cell size. Among these candidate lncRNAs, ADIPINT displayed the greatest fold change. Further research revealed that ADIPINT regulated lipid metabolism in human white adipocytes through interaction with pyruvate carboxylase in human white adipocytes [24]. Moreover, targeting lipogenesis-related lncRNA has been identified as an effective therapy for some cancers. LINC00958 can promote the expression of hepatoma-derived growth factors, facilitating HCC progression. Encapsulating LINC00958 siRNA into a poly lactic-co-glycolic acid-based nanoplatform has been identified as a therapeutic strategy with satisfactory antitumor efficacy [25]. However, the conservation of lncRNAs is a limiting factor for lncRNA study. The diversity of lncRNAs sequences, localization, and function in human and experimental animal limits the exploitation of related therapeutics.

Using a diverse interplay of mechanisms, lncRNAs regulate various biological processes. Current opinion based on previous studies has authenticated diverse lncRNAs as critical regulatory players during the lipid metabolism process and lipid signaling (Fig. 2). Here, we have highlighted the regulating effects and mechanisms of lncRNAs on lipid metabolism and signaling, which may help advance the comprehension of the detailed modulatory networks of lipid metabolism and afford a better theoretical foundation for clinical diagnosis and treatment of metabolic diseases.

Fig. 2.

LncRNAs participate in lipid metabolism and lipid signaling. In the major organs of lipid metabolism, lncRNAs regulate the synthesis and catabolism of multiple lipids by modulating the expression and function of key enzymes. Meanwhile, lncRNAs can also control the influx and efflux of lipids to maintain lipid homeostasis. In the process of lipid transport, lncRNAs regulate the expression of the apoproteins to help transmit lipid signaling. In addition, lncRNAs also modulate the expression and function of some lipid receptors to influence the activation of lipid signaling.

LncRNAs and the metabolism of diverse lipids

The types of lipids found in organisms are numerous. According to the Lipid MAPS consortium, there are eight categories of lipids including FAs, sterols, triglycerides, glycerophospholipids, saccharolipids, sphingolipids, prenols, and polyketides [26]. In addition to having abundant types, lipid metabolism is dynamic and complex, and the metabolic processes of different lipid species regulate corresponding physiopathological consequences. For example, the synthesis and catabolism of FAs and cholesterol, two essential lipids for multiple biological processes, are related to numerous diseases, including various types of malignancy, type 2 diabetes mellitus, and atherosclerosis [27], [28], [29], [30]. As a key regulator in lipid metabolism-related pathways, the role of lncRNAs in the metabolism of FAs, cholesterol, triglycerides, and sphingolipids has been gradually revealed (Fig. 3, Table 1). Lipid metabolism-related lncRNAs have also been regarded as novel therapeutic targets for multiple diseases, such as atherosclerosis, obesity, Alzheimer’s disease, and type 2 diabetes [31], [32].

Fig. 3.

LncRNAs modulate the metabolism of fatty acids, cholesterol, and triglyceride. During fatty acid synthesis, lncRNAs, like FLJ22763, HAGLR, HOTAIR, and PVT1, control the expression of FA synthesis enzymes, such as ACLY and FAS, to regulate the conversion from citrate to fatty acids. To enter bioactive pools, fatty acids need to be translated to FA-CoA. lncRNA HULC and XLOC_004244 promote the expression of ACSL to facilitate this process. FA-CoA generated from fatty acids then participates in FAO to produce NADPH. CPT1 is one of the rate-limiting enzymes of fatty acid transport into mitochondria for oxidation. LncRNA HCP5, H19, AGAP2-AS1, and MACC1-AS can promote the expression of CPT1 and NEAT1 can decrease CPT1 expression, leading to a modulation of fatty acid oxidation. Unused fatty acids will be stored, and most cells store fatty acids as TGs. DGAT is essential for the synthesis and accumulation of TGs, and its expression can be regulated by SPRY4-IT1. By regulating the transporter and key synthetases, such as LDLR, CD36, ABCA1, and HMGCR, lncRNA RP1-13D10.2, NEAT1,RP11-728F11.4, lnc-HC, CHROME, and AT102202 control the source of intracellular cholesterol. Meanwhile, lncRNA RP5-833A20.1 also modulates the reverse cholesterol transport process to regulate cholesterol homeostasis. The G3P pathway represents the de novo lipogenesis route in the synthesis of triglycerides. LncRNAs SPRY4-IT1 and lnc-KDM5D-4 can upregulate GPAT to promote triglyceride synthesis. In addition, lncRNA MSC-AS and HOXA11-AS1 influence triglyceride production by regulating PAPs and DGAT in triglyceride synthesis, respecivly. Meanwhile, lncRNA NEAT1 and SPRY4-IT1 also control the mutual transformation of triglyceride and fatty acid by modulating ATGL and DGAT respectively. In FA synthesis, SREBPs are the most critical modulator, lncRNA MALAT1, H19, HR1, and PU.1 AS can regulate the expression and stabilization of SREBPs, thus affecting the expression of FA synthesis related enzymes.

Table 1.

LncRNAs and diverse lipid metabolism.

Metabolic Process	LncRNA	Target	Function	Ref.
Fatty acid metabolism	FLJ22763	ACLY	Suppressing ACLY expression	[45]
	HAGLR	FAS	Activating FAS to increase fatty acid synthesis	[46]
	HOTAIR	FAS	Up-regulating FAS expression	[48]
	PVT1	FAS	Promoting FAS expression via negatively regulating miR-195	[51]
	HULC	ACSL1	Downregulating miR-9 to promote PPARα expression and then activate ACSL1	[52], [53]
	XLOC_004244	ACSL3	Suppressing ACSL3 expression	[54]
	MALAT1	SREBP-1c	Stabilizing nuclear SREBP-1c protein	[60]
	H19	SREBP-1c	Facilitating SREBP-1c mRNA binding to polypyrimidine tract-binding protein 1, and then increases the transcriptional activity of SREBP-1c	[62]
	HR1	SREBP-1c	Inhibiting SREBP‑1c levels through the phosphorylation of the PDK1/AKT/FoxO1 axis	[63]
	PU.1 AS	SREBP-1c	Interacting with EZH2 protein to suppress Sirt6 expression, leading to SREBP-1c decrease	[64]
	HCP5	CPT1	Sequestering miR-3619-5p to up-regulate PPAR coactivator‐1α, and then activate CPT1 transcription	[70]
	MACC1-AS	CPT1	Directly binding to miR-145-5q to promote the expression of CPT1	[71]
	AGAP2-AS1	CPT1	Forming AGAP2-AS1-HuR complex to improving CPT1 mRNA stability	[72]
	NEAT1	CPT1	Sponging miR-146a-5p to active Rho-kinase isoform 1, and lead to CPT1 expression decrease	[74]
	H19	CPT1	Interacting with miR-130b to activate PPARγ, which then promote the overexpression of CPT1	[75]
	LINC01028	ADH1C/CYP4A11/CYP4A22	Regulating expression of multiple FAO-related genes	[76]
Cholesterol and triglyceride metabolism	RP1-13D10.2	LDL receptor	Facilitating the expression of low-density lipoprotein receptors, thus increasing the uptake of plasma LDL-C	[81]
	NEAT1	CD36	NEAT1-induced paraspeckles inhibit CD36 expression	[83]
	RP11-728F11.4	CD36	Binding to RNA recognition domain of EWSR1, increasing FXYD6 transcription and promoting CD36 expression	[84]
	AT102202	HMGCR	Leading to a cis-acting effects in HMGCR transcription	[86]
	Lnc-HC	ABCA1	Forming a complex with hnRNPA2B1 and inhibiting ABCA1 expression	[90]
	CHROME	ABCA1	Inhibiting miR-27b, miR-33a/b, and miR-128 to promote ABCA1 expression	[91]
	RP5-833A20.1	NFIA	Decreasing NFIA expression by inducing hsa-miR-382-5p expression	[92]
	ANRIL	ADAM10	Inhibiting the transcription of ADAM10 via DNMT1-mediated DNA methylation and promoting cholesterol efflux	[93]
	Lnc-HC	PPARγ	Suppressing PPARγ expressing through increasing the level of miR-130b-3p, leading to the triglyceride decrease	[96]
	MSC-AS	GPAT	Interacting with miR-33b-5p to up-regulated GPAT and promote triglyceride synthesis	[98]
	SPRY4-IT1	Lipin 2	Binding lipin 2 directly to downregulate the production of triglyceride	[104]
	KDM5D-4	Lipin 2	Inhibiting the expression of lipin 2	[105]
	HOXA11-AS1	DGAT2	Promoting the expression of DGAT2	[106]
	NEAT1	ATGL	Emulously binding to miR-124-3p to improve the level of ATGL	[73]
	SRA	ATGL	Repressing ATGL expression through inhibiting the transcriptional activity of fork-head protein O1	[107]
Sphingolipids metabolism	SNHG1	Serine palmitoyltransferase	Acting as the ceRNA of miR-137 to increase serine palmitoyltransferase	[112]
	RP1-167A14.2–001	Serine palmitoyltransferase	Potential regulator of serine palmitoyltransferase	[113]
	TGFB2-OT1	CERS1	Sponging miR-146a-5p to facilitate CERS1 expression	[114]
	CERS6-AS1	CERS6	Binding with IGF2BP3 to maintain CERS6 mRNA stability	[115]
	RP24-131B6.1	Dihydroceramide desaturase 1	Transcriptomic analysis revealed the promoter regions of RP24-131B6.1 overlapped with dihydroceramide desaturase 1	[116]
	LOC100506036	Sphingomyelinase	Knockdown of LOC100506036 can suppress the expression of sphingomyelinase	[119]
	THAP9-AS1	Sphingomyelin synthase 2	Sponging miR-335-5p to promote sphingomyelin synthase 2 expression	[121]
	KCNQ1OT1	ACER3	Positively regulating ACER3 through interacting with miR-146a-5p	[125]
	MALAT1	SPHK1	Interacting with miR-124-3p as an endogenous sponge to promote SPHK1 expression	[129]
	MAFG-AS1	SPHK1	Relieving transcriptional repression of SPHK1 by binding with miR-125b-5p	[130]
	LINC00460	SPHK2	Sponging miR-613 to promote SPHK2 expression	[131]
	LINC00520	SPHK2	Sponging miR-577 to promote SPHK2 expression	[132]

Fatty acids

FA metabolism alterations in cancer and some metabolic diseases are becoming increasingly recognized [33], [34], [35]. In contrast to normal cells which prefer to obtain FAs from exogenous sources, cancer cell prefer to synthesize FA de novo [36]. Additionally, FAO is an essential NADPH source, which is critical to offset oxidative stress in cancer cells [37]. When FAO is inhibited, glioma cells exhibit markedly decreased NADPH levels and increased reactive oxygen species levels, resulting in cell death [38]. The function of lncRNAs in fatty acid metabolism and therapeutic applications for some diseases is attracting attention (Fig. 3).

In response to the high metabolic demand, FAs biosynthesis in cancer cells is extraordinarily active. Several enzymes, including ACLY [39], acetyl-CoA carboxylase [40], FAS [41], and ACSLs [42], are frequently overexpressed in tumors for converting carbons from citrate to bioactive fatty acids. In the FA decreasing availability model, cancer cell proliferation would be attenuated after inhibiting these enzymes, in which lncRNAs play a critical role (Fig. 3) [43].

By converting six-carbon citrate to the FA precursor, acetyl-CoA, ACLY regulates the initiation of FA synthesis [44]. In a recent study, researchers have identified lncRNA FLJ22763 as a suppressor gene for gastric cancer tumorigenesis. In BGC-823 human gastric cancer cells, there is a negative correlation between the level of FLJ22763 and the ACLY protein level. These results indicated that FLJ22763 may restrain gastric carcinoma tumor development by decreasing fatty acid synthesis by suppressing ACLY expression. However, the detailed mechanisms on how FLJ22763 regulates ACLY expression need further study [45].

As a multi-enzyme protein, FAS can catalyze acetyl-CoA and malonyl-CoA to synthesize palmitate [33]. In several cancers, upregulated FAS-induced FA synthesis increase has been observed. This increase in FAS is closely related to a poor prognosis in many instances [35]. Emerging evidence indicates that lncRNA HAGLR level is up-regulated in the majority of NSCLCs, and a high level of HAGLR promotes NSCLC progression by activating FAS to increase fatty acid synthesis in human NSCLC SPC-A1 and NCI-1703 cell lines [46]. LncRNA HOTAIR has been identified to play a critical role in tumorigenesis and metastasis in various carcinomas [47]. It has been demonstrated that by up-regulating FAS, HOTAIR promotes the progression and recurrence of NPC [48]. LncRNA PVT1 also participates in the proliferation and poor prognosis of various cancers [49], [50]. A study has shown that the level of PVT1 was more advanced in osteosarcoma tissues compared with corresponding noncancerous tissues. By negatively regulating miR-195, PVT1 promoted FAS expression and then aggravated the invasiveness of human osteosarcoma U2OS cells [51].

To enter bioactive pools, Fas must be translated to FA-CoA by ACSL, which has five isoforms and has different tissue specificities and preferred substrates [33]. As the first identified lncRNA that is specifically up-regulated in hepatocellular carcinoma, lncRNA HULC downregulates miR-9 and releases the inhibition of PPAR α by inducing methylation of CpG islands in the promotor in humans [52], [53]. PPARα then activates ACSL1 to initiate the metabolism of cellular long-chain fatty acid in human Huh7 and HepG2 hepatocytes cell lines. Meanwhile, in the liver, the formations of triglycerides and cholesterol also require ACSL1. Further, through a positive feedback loop, cholesterol can up-regulate HULC expression [53]. Thus, HULC-enhanced fatty acid synthesis accelerates liver cancer growth. Statins have been identified to improve endothelial function in atherosclerosis, and lncRNAs related to fatty acid synthesis may play a role in this process. A previous study using a human lncRNA microarray, profiled the variations in lncRNAs and mRNAs expression in human umbilical vein endothelial cells treated with pravastatin, and identified 181 lncRNAs with significant differences. LncRNA XLOC_004244 was one of these candidates. The upregulation of XLOC_004244 represses ACSL3 expression and contributes to a reduction in fatty acid synthesis [54].

Besides regulating FA synthesis enzymes, lncRNAs also regulate the expression of FA synthesis-related regulatory molecules. In FA synthesis, SREBPs are the most critical modulators [55], [56]. SREBP-1a and SREBP-1c are mainly promotors of fatty acid synthesis, while SREBP-2 is crucial for de novo cholesterol biosynthesis [57], [58]. As the most largely expressed lncRNAs in normal tissues, the role of lncRNA MALAT1 in the process of tumorigenesis is controversial [59]. Many studies described it as a tumorigenesis and metastasis promotor, while other reports indicated a tumor-suppressing activity of MALAT1 [59]. In hepatoma carcinoma, researchers have found that MALAT1 binds with SREBP-1c to maintain the stability of nuclear SREBP-1c protein via the ubiquitin–proteasome pathway, leading to lipid accumulation in human HepG2 hepatocytes cell [60]. Meanwhile, MALAT1 can also increase SREBP-2 mRNA stability, but the detailed mechanisms need further study for clarification [60], [61]. Similar to lncRNA MALAT1, lncRNA H19 also has a function in stabilizing SREBP-1c. In the cytoplasm of human Hepa-1 and Huh7 hepatocytes cell lines, H19 facilitates SREBP-1c mRNA binding to polypyrimidine tract-binding protein 1. Thus, H19 increases the transcriptional activity of SREBP-1c [62]. Contrary to the MALAT1 function, lncRNA HR1 is a transcriptional suppressor of SREBP-1c. LncRNA HR1 was first described as being regulated in human Huh7 hepatocytes cells infected with the hepatitis C virus. Studies have found that in the human Huh7 hepatocyte cell line, HR1 participates in the activation of the PDK1/AKT/FoxO1 axis, and then regulates the SREBP-1c level [63]. As an omnipresent metalloid toxicant, arsenic-induced lipid metabolic disequilibrium could contribute to the development of cardiovascular diseases. The study has indicated that lncRNA PU.1 AS was significantly increased in the liver of arsenic-exposed mice. Additionally, PU.1 AS could interact with EZH2 protein to suppress Sirt6 expression, therefore leading to decreased expression of SREBP-1c [64].

Overall, by regulating fatty acid synthesis and enhanced uptake, lncRNAs maintain fatty acid homeostasis. However, it is worth noting that some lncRNAs can act on the same enzyme or regulatory factors. Thus, HAGLR and PVT1 regulate FAS, or MALAT1 while HR1 and PU.1 AS can modulate the expression of SREBP-1c. Clarifying the relationship between these lncRNAs may provide a clearer roadmap for exploring mechanisms of maintaining cellular fatty acid homeostasis.

In most situations, the growth and survival of cancer cells are limited by cytosolic NADPH levels [65]. The cytosolic NADPH produced from FAO is critical for counteracting oxidative stress in cancer cells [66]. Within the mitochondria, Fas are rapidly translated to acetyl-CoAs and then feed into the Krebs cycle to produce reducing substances for oxidative phosphorylation. Increasing FAO to limit FA abundance could, in theory, be beneficial to cancer cells [67]. Studies have shown that lncRNAs regulate key enzymes in fatty acid oxidation [68], [69].

As the first rate-limiting enzyme of FAO, CPT1 can assist fatty acid transport into mitochondria [38]. A study has reported that MSC-induced lncRNA HCP5 sequestered miR-3619-5p to up-regulate PPAR coactivator-1α, activating transcription complex PPAR coactivator-1α and CPT1, which prompted FAO in gastric cancer cells. By driving FAO, HCP5 enhances stemness and chemo-resistance of human gastric cancer AGS and MKN45 cell lines [70]. Similar to HCP5, lncRNA MACC1-AS is also induced by MSCs in gastric cancer cells. miR-145-5p can depress the expression of CPT1, while lncRNA MACC1-AS directly binds to miR-145-5p to block its function, promoting the expression of CPT1, resulting in an FAO-dependent chemo-resistance and stemness of human AGS and MKN45 gastric cancer cells [71]. Meanwhile, in breast cancer, FAO has been shown to maintain the stemness and chemoresistance of cancer cells. It was shown that mesenchymal stem cells co-culture-induced lncRNA AGAP2-AS1 can regulate stemness and trastuzumab resistance in SKBR-3 and BT474 human breast cancer cell lines by activating FAO. Additionally, the AGAP2-AS1-HuR complex promoted CPT1 expression by enhancing the stability of CPT1 mRNA [72]. LncRNA NEAT1 is required for paraspeckles formation and lipolysis [73]. In the human HepG2 hepatocyte cell line, NEAT1 sponged miR-146a-5p to activate Rho-kinase isoform 1, which led to a decrease in CPT1 expression and facilitated lipid accumulation [74]. During the development of atherosclerosis, H19 can interact with miR-130b to prevent its negative regulation of PPARγ in human macrophage cells, which in turn promotes the overexpression of CPT1 and increases fatty acid uptake and foam cell formation[75]. Besides targeting a single key enzyme, some lncRNAs can regulate a series of genes involved in fatty acid oxidation. Thus, the functional human cardiometabolic trait-associated lncRNA, LINC01028, was identified by integration of GWAS and eQTL data and further bioinformatic and experimental analyses. Using a humanized mouse model, LINC01028 was reported to regulate the expression of multiple genes, such as ADH1C, CYP4A11, and CYP4A22, which were involved in fatty acid oxidation in the liver of humanized mice [76].

Cholesterol and triglyceride

Both cholesterol and triglyceride are major storage lipids and are critical in energy metabolism and signal transduction. These two lipids have been well described because of their unique role in the process of cardiovascular disease [77]. It has been demonstrated that the use of ASOs and siRNA technologies to target cholesterol and triglyceride synthesis pathways or degradation-related gene expression are therapeutic approaches to lower cholesterol and triglyceride for the prevention and therapy of cardiovascular diseases [78], [79]. Considering the regulating effect of lncRNAs in the synthesis and degradation of cholesterol and triglyceride (Fig. 3), targeting lncRNAs may provide novel strategies for reducing cardiovascular risk.

In blood, cholesterol is present in lipoproteins, and LDL in plasma is the leading carrier for transporting endogenous cholesterol. LDL particles in the blood are usually removed by low-density lipoprotein receptors [80]. LncRNA RP1-13D10.2 can facilitate the increase of low-density lipoprotein receptors, thus increasing the uptake of plasma LDL-C in human Huh7 and HepG2 hepatocytes cell lines. Moreover, in patients with a high LDL-C response upon statin treatment, an increased level of RP1-13D10.2 was observed. Conversely, low LDL-C response patients exhibited reduced RP1-13D10.2, indicating that RP1-13D10.2 may act as a novel biomarker and may play an important role in statin-mediated LDL-C decrease [81]. In the initial stage of the atherosclerotic process, circulating monocytes translocate to arterial intima where they differentiate into macrophages, which then bind and internalize oxidized LDL via CD36 to form foam cells. Internalized oxidized LDL further upregulates CD36 expression, which has been referred to as the “eat me signal” to promote uptake of more oxidized LDL[82]. The lncRNA NEAT1 regulates multiple biological processes, including tumorigenesis and adipogenesis. A study has found that oxidized low-density lipoprotein mediates NEAT1-induced paraspeckle formation in the human monocyte THP-1cell line. Paraspeckles then partially suppress cholesterol uptake by stabilizing CD36 mRNA, which inhibits CD36 expression [83]. Similarly, lncRNA RP11-728F11.4 is also related to the level of CD36 in human monocyte-derived macrophages. A study demonstrated that RP11-728F11.4 relieved the repressive activity of EWSR1 on FXYD6 by binding to its RNA recognition domain. Increased FXYD6 then promoted CD36 expression, resulting in intracellular cholesterol accumulation [84].

HMGCR is one of the rate-limiting enzymes in the process of cholesterol synthesis [85]. LncRNA AT102202 is located in the HMGCR gene locus and can lead to a cis-acting effect in HMGCR transcription [86]. In the human HepG2 hepatocytes cell line, silencing AT102202 can increase HMGCR expression. Studies have found that the natural compound polyphenol (-)-epigallocatechin gallate from green tea can enhance the expression of AT102202 and reduce HMGCR expression [87]. The relationship between natural products and lncRNAs related to cholesterol metabolism may provide a novel therapeutic method for lipid metabolism-related diseases.

Besides reducing cholesterol uptake, organisms remove and transport excess cholesterol to maintain cholesterol homeostasis via RCT [88]. During RCT, cholesterol is transported to apolipoproteins by ABCA1, which is effectively targeted by numerous lncRNAs [89]. The accumulation of cholesterol always induces the expression of lnc-HC. In the rat CBRH-7919 hepatocyte cell line and liver of hyperlipidemia rat model, lnc-HC can then form a complex with hnRNPA2B1 and inhibit ABCA1 expression, resulting in the inhibition of RCT [90]. CHROME is a primate-specific, cellular and systemic cholesterol homeostasis-related lncRNA. The expression of CHROME is frequently increased in the plasma and arterial plaques of cardiovascular disease patients. A previous study showed that CHROME could inhibit miR-27b, miR-33a/b, and miR-128 simultaneously. Knockdown of CHROME in human hepatocytes and macrophage reduces the expression of ABCA1, one of the genes repressed by these miRNAs, leading to a disruption of cholesterol efflux and high-density lipoprotein biogenesis [91]. In human THP-1 macrophage-derived foam cells, oxidized low-density lipoprotein or acetylated low-density lipoprotein enhances the expression of lncRNA RP5-833A20.1. Subsequently, it induces hsa-miR-382-5p and leads to the expression of nuclear factor IA, a major regulator involved in lipid homeostasis and adipocyte differentiation. The decrease of nuclear factor IA reduces RCT and affects cholesterol homeostasis [92]. LncRNA ANRIL has been reported to play critical roles in multiple diseases including atherosclerosis, diabetes, and cancer. For example, in atherosclerotic plaque tissue and THP-1 macrophage-derived foam cells, the expression of ANRIL is frequently downregulated. Overexpressing ANRIL recruits DNMT1 to the promotor region of ADAM10 to initiate its methylation. Thus, the inhibition of ADAM10 promotes cholesterol efflux and attenuates atherosclerosis [93].

The G3P and dihydroxyacetone phosphate pathways are two main pathways for triglyceride biosynthesis in the liver and adipose tissue [94], [95]. In these pathways, lncRNAs regulate the expression of several key enzymes to maintain the requirement of triglyceride. For example, aberrant hepatic lipid metabolism is identified to initiate NAFLD pathologies and lncRNAs contribute to this process. In vivo studies have identified that after reducing the expression of lnc-HC, miR-130b-3p expression decreases, inducing PPARγ expression, and increasing triglyceride, aggravating aberrant lipid accumulation in rat livers with hyperlipidemia [96]. The G3P pathway represents the de novo lipogenesis route in the synthesis of triglycerides. In the rate-controlling step of this pathway, GPAT promotes sn-glycerol-3-phosphate transformation to lysophosphatidic acid by catalyzing the acylation of sn-glycerol-3-phosphate with acyl-CoA [97]. LncRNA MSC-AS can interact with miR-33b-5p to up-regulated GPAT and promote triglyceride synthesis which then facilitates lung adenocarcinoma [98]. The lysophosphatidic acid is then transformed into phosphatidic acid, and PAPs, such as lipin 1 and lipin 2, can remove the phosphate group of phosphatidic acid to produce DAG, which is an essential intermediate in the biosynthesis of triglyceride [99], [100]. Increasing PAPs have been recently linked to several cancers, and lncRNAs have shown a close correlation with PAPs [101], [102], [103]. LncRNA SPRY4-IT1 is overexpressed in the A375 melanoma cell line. It can bind lipin 2 directly to downregulate the production of triglyceride, provoking cellular lipotoxicity [104]. In the human hepatocellular carcinoma HepG2 cell line, the underexpression of lncRNA lnc-KDM5D-4 causes overexpression of lipin 2, leading to increased lipid droplets formation [105]. In the last step of the glycerol-3-phosphate pathway, DAG can be esterified by DGAT to give rise to triglyceride. By regulating the expression of DGAT, lncRNA modulates the transformation between DAG and triglyceride. In human adipose-derived stem cells, lncRNA HOXA11-AS1 is highly expressed and can facilitate adipocyte differentiation. During the process of adipocyte differentiation, HOXA11-AS1 increased significantly and promoted the expression of DGAT2 [106].

When tissues require fatty acids for energy or other purposes, triglyceride is released through the action of ATGL, monoacylglycerol lipase, and hormone-sensitive lipase. As the rate-limiting enzyme of triglyceride catabolism, ATGL hydrolyses triglyceride to diacylglycerols. The modulation of ATGL is crucial for the initiation of triglyceride hydrolysis [106]. The role of lncRNAs in regulating ATGL has provided novel targets for regulating triglyceride metabolism in some physiological and pathological conditions. In HCC tissues, highly expressed ATGL is regarded as a biomarker that predicts a poor prognosis. LncRNA NEAT1 can improve the level of ATGL by emulously binding to miR-124-3p, and mediates HCC exacerbation through the miR-124-3p/ATGL/DAG axis. Taking into account the effects of both NEAT1 and ATGL may improve the prognostic accuracy for HCC [73]. Meanwhile, by inhibiting the transcriptional activity of fork-head box protein O1, lncRNA SRA has been found to repress ATGL and promote hepatic steatosis, resulting in the hepatic free fatty acid oxidation inhibition [107].

Sphingolipids

As important structural components of cell membranes, sphingolipids maintain the barrier function, and sphingolipid metabolites, such as sphingosine-1-phosphateand and ceramide, can serve as intermediaries in the signaling cascades related to stress responses, cell proliferation, and inflammation [108], [109]. Targeting sphingolipid metabolism-related enzymes for some disease treatments has been reported [108], [110], [111]. Studies have shown that some lncRNAs promote several diseases by regulating sphingolipid metabolism-related enzymes (Fig. 4). Targeting these lncRNAs may provide novel approaches for treating these diseases.

Fig. 4.

Function and mechanism of lncRNAs in sphingolipid metabolism. As the core molecule, the synthesis and transformation of ceramide are crucial for sphingolipid metabolism. During the synthesis of ceramide, lncRNAs participate in the regulation of several critical enzymes, like SPT, CRES, and DES. LncRNA THAP9-AS1 and LOC100506036 control the mutual transformation of ceramide and sphingomyelin. Meanwhile, in the process of producing sphingosine‑1‑phosphate, multiple lncRNAs, such as KCNQ1OT1, MALAT1 and LINC00460 are indispensable in regulating the expression or function of ceramidase and SPHKs.

As the central hub of sphingolipid metabolism, de novo synthesis of ceramide begins with the formation of 3-ketodihydrosphigosine by serine palmitoyltransferase. In a human cell model of Alzheimer’s disease, lncRNA SNHG1 acted as a ceRNA for miR-137. The decrease of miR-137 was found to be closely related to serine palmitoyltransferase increase [112]. In myelodysplastic syndrome patients and SKM-1 and THP-1 cell lines, lncRNA RP1-167A14.2–001 was shown to be significantly downregulated and bioinformatic analysis showed that serine palmitoyltransferase was the potential target gene of RP1-167A14.2–001 [113].

In the next step, 3-ketodihydrosphigosine is reduced to form dihydrosphingosine which is acylated by CERSs to form dihydroceramide. In this process, lncRNAs are critical for the expression of CERSs. In human vascular endothelial cells, inflammation induced by lipopolysaccharide and oxidized low-density lipoprotein can markedly increase the level of lncRNA TGFB2-OT1. TGFB2-OT1 acts as a competing endogenous RNA to sponge miR3960, facilitating the expression of CERS1, and promoting autophagy [114]. As the antisense lncRNA of CERS6, CERS6-AS1 was notably upregulated in human breast cancer tissue and cell lines (MDA-MB-436, MDA-MB-453, MCF-7, and MDA-MB-231) compared with normal tissue/cell line (MCF10A) and binds with IGF2BP3 to maintain CERS6 mRNA stability and facilitate the progression of breast cancer [115].

The final step in producing ceramide is catalysis by dihydroceramide desaturase. Transcriptomic analysis of high-fat diet fed mouse brain cortex revealed that the location of promotor regions of lncRNA RP24-131B6.1 overlapped with dihydroceramide desaturase 1, suggesting that RP24-131B6.1 may participate in the synthesis of ceramide by regulating dihydroceramide desaturase [116].

Hydrolysis of sphingomyelin is another way to generate ceramide. This process is catalyzed by the enzyme sphingomyelinase. The activation of sphingomyelinase is a major route for ceramide production in cells under stress states and is closely related to lncRNAs [117], [118]. In rheumatoid arthritis patients, lncRNA LOC100506036 is increased in activated Jurkat T lymphocyte cells. Knockdown of LOC100506036 in the human Jurkat T lymphocyte cell line can lead to the decreased expression of sphingomyelinase and nuclear factor in activated T cells [119].

Ceramide is subsequently transformed into various forms of sphingolipids to help modulate lifes activities. Ceramidases and SPHKs catalyze ceramide to generate sphingosine-1-phosphate. Alternatively, ceramide can generate complex sphingolipids like glycosphingolipids using glucosylceramide synthase. Additionally, sphingomyelin synthase or ceramide kinase can convert ceramide to sphingomyelin or ceramide-1-phosphate respectively [108], [109]. In these transformation pathways, the effects of lncRNAs on the key enzymes influence the generation of diverse sphingolipids, such as sphingosine-1-phosphate and sphingomyelin, and subsequent physiopathological processes.

As the main sphingolipid in the human body, sphingomyelin is involved in the formation of the plasma membrane and has a role in cell apoptosis [111], [120]. By modulating the expression of sphingomyelin synthase, lncRNAs modulate the synthesis of sphingomyelin. For example, in human esophageal squamous carcinoma KYSE-150 and TE-10 cell lines, lncRNA THAP9-AS1 can sponge miR-335-5p to promote sphingomyelin synthase 2 expression and facilitate the progression of esophageal squamous carcinoma [121]. In addition, sphingosine-1-phosphate is another important conversion product of ceramide, where lncRNAs are again also essential. ACER3 is one of the most ubiquitous ceramidases. It is strongly linked to multiple cancers, such as colitis-related cancers and myeloid leukemia, and targeting ACER3 has been proposed [122], [123], [124]. In HCC, lncRNA KCNQ1OT1 can positively regulate ACER3 through interacting with miR-146a-5p and decreasing apoptosis and radiosensitivity of human Huh7 and Hep3B hepatocytes cell lines [125]. In addition to ceramidases, SPHK1 and SPHK2 are also essential for the synthesis of sphingosine-1-phosphate. Antisense lncRNA KHPS1 can be tethered to the triplex-forming region of the SPHK1 enhancer via RNA-DNA triplex formation recruits p300 and E2F1, and then activate the expression of SPHK1 [126]. In the development of multiple cancers, some sphingosine kinase-related lncRNAs have been identified to play an important role [110], [127], [128]. In human osteosarcoma MG63, U2OS cells, lncRNA MALAT1 can interact with miR-124-3p as an endogenous sponge to promote SPHK1 expression, then modulate osteosarcoma progression [129]. Additionally, overexpressed SPHK1 promoted the proliferation of human bladder cancer HT01197 and HT-1376 cell lines, and lncRNA MAFG-AS1 can relieve transcriptional repression of SPHK1 by binding with miR-125b-5p, facilitating the progression of bladder cancer [130]. Similarly, SPHK2 also contributes to the development of some cancers and the regulation of SPHK2 expression by lncRNAs may provide a promising strategy for targeting SPHK2. It has been demonstrated that in human papillary thyroid carcinoma TPC-1, BCPAP, and IHH-4 cell lines, LINC00460 and LINC00520 can sponge miR-613 and miR-577 respectively, and then promote the expression of SPHK2, facilitating the progression of papillary thyroid carcinoma [131], [132] (Fig. 4).

As a critical regulatory factor of lipid homeostasis, the importance of lncRNAs has been gradually elucidated. Various lncRNAs seem to play a role in some diseases related to dysfunctional lipid metabolisms (e.g., atherogenesis and dyslipidemia). It is now necessary to ascertain whether lncRNAs participating in lipid metabolism are effective disease biomarkers and potential therapeutic targets. However, lncRNAs function through multiple and complex mechanisms, and they may be associated with many different signaling pathways. It is still necessary to confirm the specificity of lncRNAs as biomarkers for lipid metabolism disorder.

LncRNAs and lipid signaling

In addition to functioning as the fundamental structural elements of plasma membranes and key molecules in energy metabolism, numerous studies have identified that lipids are crucial signaling molecules that participate in the regulation of nuclear transcription and cellular communication [133]. In these lipid signaling pathways, signaling lipids, lipid chaperones, and lipid receptors cooperate to maintain normal pathway function. The disorder of lipid signaling pathways results in the progression of multiple diseases, including cardiovascular disease, neurodegenerative disease, and cancers [5], [134], [135]. Studies have indicated that the regulation of lipid signaling pathways also requires the participation of lncRNAs [136]. Besides regulating the signaling lipids production, lncRNAs also modulate the transport of signaling lipids and the activation of lipid signaling by regulating lipid chaperones and receptors. This section will focus on the relationships between the dysregulation of lipid chaperones and lipid receptors induced by lncRNAs and diseases (Fig. 5, Table 2).

Fig. 5.

LncRNAs affect lipid signaling by regulating lipid chaperones and receptors. By modulating the expression levels of FABPs and apolipoproteins, lncRNAs regulate the intracellular levels of signaling lipids, thus affecting downstream signaling pathways. Meanwhile, some lncRNAs also influence lipid receptors including GPCRs, K channels, TRP channels, and some lipid-sensing nuclear receptors to detect the signaling lipid and activate downstream signaling pathways and the transcription of critical molecules that control diverse physiopathological processes.

Table 2.

LncRNAs and lipid signaling.

LncRNA	Target	Mechanism	Ref.
LNMICC	FABP5	Recruiting nucleophosmin 1 to the loci of FABP5 to increase FABP5 expression	[142]
UCA1	FABP5	FABP5 transport into the nucleus and increase SP1, lead to a high level of UCA1	[143]
MIR31HG	FABP4	Enriching AcH3 and H3K4me3 modification in the FABP4 promoter to induce FABP4 expression	[144]
APOA1-AS	Apolipoprotein gene cluster	Recruiting SUZ12 to the APO gene cluster and promoting apolipoprotein expression	[145]
APOA4-AS	APOA4	Interacting with HuR and stabilize APOA4 mRNA	[146]
LSTR	APOC2	Suppressing APOC2 expression	[147]
NEAT1	APOG4	Sponging hsa-miR-372-3p to induce the overexpression of APOG4	[150]
KCNQ1OT1	S1PR1	Promoting S1PR1 expression via sponging miR-149	[159]
LISPR1	S1PR1	Facilitating RNA polymerase II binding to the S1PR1 promotor and inhibiting the binding of zinc finger protein 354C	[161]
MALAT1	CNR1	Suppressing the function of miR-30b to upregulate the CNR1	[166]
Xist	CB2R	Activated CB2R can increase the level of Xist	[167]
NONRATT023402.2	PTGER3	Slicing NONRATT023402 inhibited Ptger3	[168]
KCNQ1OT1	CPT1	Enhancing chromatin flexibility allowing the Kcnq1 promoter to constitute ectopic regulatory contacts, which leads to the overexpression of KCNQ1	[174]
GAS5	KCNQ3/ KCNK3	Sponging miR-135a-5p/ miR-23b-3p to induce the overexpression of KCNQ3/ KCNK3	[176], [177]
TCONS-00106987	KCNJ2	Inducing the transcription of KCNJ2 (Kir2.1) by sponging miR-26	[178]
POU6F2-AS1	KCNJ4	Sponging miR-34c-5p to facilitate KCNJ4 expression	[179]
BC1686887	TRPV1	Knockdown of lncRNA BC1686887 can reduce the expression level of TRPV1	[184]
MALAT1	TRPV4	Upregulating TRPV4 through sponging of miR-214	[186]
SNHG5/ DLX6-AS1	TRPC3	Sponging miR-26a to facilitate TRPC3 expression	[187], [188]
TUG1	TRPC6	Binding to miR-145-5p and promote TRPC6 expression	[189]
RP11-142A22.4	PPAR-γ	Suppressing miR-587 and then promote Wnt5β and PPAR-γ expression	[193]
Plnc1	PPAR-γ	Decreasing the methylation level of the PPAR-γ promoter and enhancing the transcriptional activity	[194]
LOC100996425	HNF4A	Reducing the expression of HNF4A by binding to HNF4A mRNA to form a silencing complex	[197]
LINC00858	HNF4A	Promoting HNF4A binding to the promoter area of WNK lysine deficient protein kinase 2	[198]
TUG1	FXR1	Negatively regulating miR-92R to induce the expression of FXR1	[199]
Blnc1	LXR	Endothelial differentiation-related factor 1 can interact with Blnc1, constituting a ribonucleoprotein transcriptional complex with LXR	[201]
LeXis	LXR	LXR activation increase the level of LeXis to reduce serum and hepatic cholesterol levels	[203]
MeXis	LXR	Interacting with and guiding the transcriptional coactivator DDX17 to bind to ABCA1 promoter in LXR/MeXis/ABCA1 pathway	[204]

LncRNAs in the regulation of lipid chaperones

To diffuse through the intracellular water environment, hydrophobic lipid molecules are required to bind to lipid chaperones. There are two key classes of lipid-binding proteins, FABPs, which facilitate the uptake of fatty acids and their derivatives, and apolipoproteins [6], [137], [138]. By regulating these lipid chaperones, lncRNAs modulate the flow of signaling lipids to affect lipid signaling pathways (Fig. 5).

It has been well recognized that FAs, which serve as the signaling lipids, are critical for the colonization of metastatic cancer cells in distant organs [139]. FABPs are not only an essential carrier of FA transport, but are also strongly linked to the invasiveness and metastasis of cancer cells. For instance, FABP5 can induce EMT in hepatocellular carcinoma cells and facilitate invasiveness and metastasis [36], [140], [141]. In human cervical cancer SiHa and MS751 cell lines, nucleophosmin 1 is recruited to FABP5 through lncRNA LNMICC, increases the expression of FABP5, then induces VEGF-C expression to promote lymphangiogenesis. miR-190 can bind with LNMICC to arrange an RNA-induced silencing complex and suppress the biological function of LNMICC [142]. Conversely, FABP5 is also a modulator of lncRNAs. As an example, in human gastric cancer HGC27 and MGC803 cell lines, palmitate acid can promote FABP5 transport into the nucleus and increase the level of nuclear protein SP1, which then induces a high level of lncRNA UCA1 to enhance the metastatic properties of gastric cancer [143]. FABP4 is another important FA transporter that is regulated by lncRNA. A study has shown that lncRNA MIR31HG can enrich AcH3 and H3K4me3 modifications in the FABP4 promoter to induce FABP4 expression in human adipose-derived stem cells, leading to the promotion of de novo formation of adipose tissue. Its overexpression can activate adipogenic factors, such as PPARγ and CCAAT enhancer binding protein alpha, which then promotes adipocyte differentiation [144].

Apolipoproteins are another category of lipid-binding proteins that serve as components of lipoprotein particles, lipid transport proteins, and ligands of some cell surface receptors. Their expression and function are also closely related to some lncRNAs. In plasma, APOA1 is one of the high-density lipoprotein components. It has been demonstrated that APOA1-AS can regulate the apolipoprotein gene cluster. Knockdown of APOA1-AS can disrupt the interaction between SUZ12 and the APO gene cluster and reduce the H3K27 trimethylation marks along the APOA1 promoter area, which then decreases the expression of these genes in human HepG2 hepatocytes cell line and African green monkey primary hepatocytes [145]. Similar to apolipoprotein A1, the expression of apolipoprotein A4 is also regulated by its antisense lncRNA APOA4-AS. APOA4-AS can directly interact with HuR and stabilize apolipoprotein A4 mRNA in mouse liver [146]. Considering the specificity that APOA4-AS shows in regulating apolipoprotein A4, it could be a potential therapeutic target for metabolic diseases. LncRNA LSTR was the first identified liver-rich lncRNA, and is involved in glucose and lipid metabolism in the liver. In the hyperlipidemic mouse model, liver-specific depletion of LSTR induced lowered plasma triglyceride and blood glucose levels by up-regulating APOC2, thereby promoting the catabolism of triglyceride mediated by lipoprotein lipase [147]. As an important immunosuppressive agent, rapamycin is crucial in improving the survival rate of liver transplant patients [148]. However, triglyceride accumulation in the liver and hypertriglyceridemia induced by rapamycin may cause non-liver-related death [149]. A study has demonstrated that rapamycin can promote the expression of lncRNA NEAT1, and then sponge hsa-miR-372-3p to induce the overexpression of APOG4 in human Huh7 and HepG2 hepatocytes cell lines and mouse liver, facilitating triglyceride accumulation and hypertriglyceridemia [150]. Targeting lncRNA NEAT1 may alleviate the side effects of rapamycin and improve the prognosis of liver transplantation patients.

LncRNAs and lipid receptors

For the lipid signaling pathway, lipid receptors are the trigger that recognizes the presence of signaling lipid and activate downstream signaling pathways. GPCRs comprise the largest superfamily with nearly 800 receptors in humans. Most GPCR signal transduction is influenced by lipid-protein interactions and dozens of drugs targeting lipid GPCRs are commercially available or under evaluation in clinical trials [151], [152]. In a similar manner to GPCRs, ion channels such as potassium (K) channels, TRP channels, and pentameric ligand-gated ion channels can also interact with phospholipids or cholesterol to transmit lipid signaling [153], [154], [155]. Lipid-sensing nuclear receptors, including PPARs, LXR, and FXR are another class of lipid receptors, whose transactivation activity is controlled by signaling lipids [156], [157]. These lipid receptors are critical to the function of lipid signaling pathways, while lncRNAs modulate these lipid receptors by acting as transcription factors or downstream effector molecules (Fig. 5).

As a class of lipid signaling molecules S1P targets GPCRs, S1PRs regulate fundamental biological processes, including endothelial integrity, vascular development, and lymphocyte trafficking [158]. In human Huh7 and SMMC-7721 hepatocytes cell lines, lncRNA KCNQ1OT1 can promote S1PR1 expression by sponging miR-149, which can downregulate S1PR1expression by targeting the 3′-UTR of S1PR1 mRNA, which then facilitate the progress of HCC [159]. In contrast to the high expression level of S1PR1 in some tumors, endothelial S1PR1 expression in human pulmonary diseases, like chronic obstructive pulmonary disease, is downregulated. The low levels of S1PR1 attenuate migration and spheroid outgrowth of endothelial cells induced by S1P, which accelerates the progress of pulmonary diseases [160]. It has been identified that in human umbilical vein endothelial cells, as well as human lung tissue, the lncRNA LISPR1 controls endothelial S1PR1 expression by facilitating RNA polymerase II binding to the S1PR1 promotor and inhibiting the binding of zinc finger protein 354C, the transcriptional repressor [161]. Targeting LISPR1 may afford a novel strategy for the therapy of multiple pulmonary diseases.

Cannabinoid receptors are another class of GPCRs, which are required in multiple physiological processes, such as appetite, memory, and pain sensation [162], [163], [164]. In some neurodegenerative diseases, cannabinoid receptors are critical for disease progression and lncRNAs participate in regulating the expression and activation of cannabinoid receptors [165]. In rat hippocampus and rat PC12 and C6 cell lines, lncRNA MALAT1 can suppress the function of miR-30b to upregulate CNR1 and promote neuronal recovery following Alzheimer’s disease [166]. Cannabinoid receptors can also perform their functions by regulating downstream lncRNAs. For example, the CB2R agonist AM1241 can activate CB2R, leading to increased levels of lncRNA Xist. Subsequently, Xist can sponge miR-133b-3p to promote Pitx3 expression and protect nerves in a Parkinson’s disease animal model [167]. In addition to S1PRs and cannabinoid receptors, lncRNAs can also regulate other lipid GPCRs. In levodopa-induced dyskinesia and Parkinson’s disease, lncRNA NONRATT023402.2 is downregulated. This downregulation can decrease the level of PTGER3 and reduce inflammation in neurons, leading to the progression of levodopa-induced dyskinesia and Parkinson’s disease [168].

Besides GPCRs, some lipids can also modulate the conformational changes of specific ion channels to trigger lipid signaling. Additionally, lncRNAs are involved in the regulation of lipid signaling by regulating the expression and functions of these ion channels. As the switch of K⁺ flux, K channels play key roles in both excitable and non-excitable cells. Some lipids, such as polyunsaturated fatty acids, phosphatidylinositol 4,5-bisphosphate, and cholesterol, modulate the activity of K channels to regulate intracellular lipid signaling pathways [169], [170], [171], [172], [173]. LncRNAs take part in these processes by regulating multiple K channels including voltage-gated potassium (Kv), inwardly rectifying potassium (Kir), and two-pore-domain potassium (K2P) channels. For example, lncRNA KCNQ1OT1 is located antisense of the Kcnq1 (Kv7.1) gene. In later cardiac development, the absence of KCNQ1OT1 can enhance chromatin flexibility allowing the Kcnq1 promoter to constitute ectopic regulatory contacts, which leads to the overexpression of KCNQ1 in the heart [174]. KCNQ3 is another Kv channel that is generally expressed in the nervous system [175]. In a rat epilepsy seizure model, lncRNA GAS5 acts as the competitive endogenous RNA of miR-135a-5p to induce KCNQ3 expression, then accelerate the progress of epilepsy [176]. In addition to KCNQ3, GAS5 can also sponge miR-23b-3p to induce the overexpression of KCNK3 (K2P3.1) in human pulmonary artery smooth muscle cells, facilitating the proliferation and migration of human pulmonary artery smooth muscle cells, aggravating pulmonary hypertension [177]. During the initiation and progression of atrial fibrillation, lncRNA TCONS-00106987 can induce the transcription of KCNJ2 (Kir2.1) by sponging miR-26, thereby adding inward-rectifier K⁺ current, promoting atrial electrical remodeling [178]. Another Kir2 channel, KCNJ4 (Kir2.3), is also modulated by lncRNA. In human lung adenocarcinoma Calu-3 and NCI-H460 cell lines, lncRNA POU6F2-AS1 can sponge miR-34c-5p and facilitate KCNJ4 expression, promoting the development of lung adenocarcinoma cells [179].

Similar to K channels, several TRP channel families, such as TRPV1 [180], TRPV4 [181], TRPC3 [182], and TRPC6 [183], are also modulated by phospholipids or cholesterol, and the sites of interaction may be druggable. These lipid signaling-related TRP channels are also regulated by specific lncRNAs. For instance, knockdown of lncRNA BC1686887 in a diabetic neuropathic pain rat model can reduce the expression level of TRPV1 and alleviate TRPV1-mediated diabetic neuropathic pain [184]. Although the detailed regulatory mechanism is still unclear, targeting BC1686887 may point to an original strategy for the prevention and treatment of diabetic neuropathic pain. TRPV4 is another member of the TRPV family, which can promote cell apoptosis in ischemia–reperfusion injury [185]. It has been demonstrated that in testicular ischemia–reperfusion injury, lncRNA MALAT1 can upregulate TRPV4 through sponging of miR-214 and then promote cell apoptosis in mouse testis and the mouse GC-1 spermatogenic cell line [186]. In laryngeal cancer HEp-2 and Tu-177 cell lines and melanoma A375 and A2508 cell lines, miR-26a can be sponged by lncRNA SNHG5 and DLX6-AS1 respectively, facilitating TRPC3 expression and leading to tumor growth [187], [188]. Another TRPCs member, TRPC6, is also critical in the development of cancer. In human colorectal cancer HCT116 and SW620 cell lines, lncRNA TUG1 can bind to miR-145-5p and promote TRPC6 expression, thereby accelerating tumor cell proliferation and migration [189] (Fig. 5).

Both ion channels and GPCRs are important lipid receptors in cell membranes. By regulating ion channels and GPCRs, lncRNAs play an essential role in the recognition of extracellular lipid signals. Targeting these lncRNAs may be a novel strategy for regulating lipid signaling pathways, including the treatment of some nervous system diseases and cancers. However, the exploration of ion channels and GPCRs-related lncRNAs remains limited and the role of lncRNAs in extracellular lipid signaling transmission needs further study.

By recognizing intracellular signaling lipid, lipid-sensing nuclear receptors evoke a transcriptional response and link lipid metabolism with multiple diseases such as nonalcoholic steatohepatitis, obesity, and cancers [190], [191], [192]. As the factor involved in gene expression, lncRNAs regulate diverse lipid-sensing nuclear receptors to impact on physiopathological processes. LncRNA RP11-142A22.4 can suppress miR-587 and then promote Wnt5β and PPAR-γ expression in preadipocytes isolated from human visceral adipose tissue to facilitate adipocyte differentiation [193]. In addition, lncRNA Plnc1 can also increase PPAR-γ transcription to control adipocyte differentiation by decreasing the methylation level of the PPAR-γ promoter and enhancing transcriptional activity in human adipose tissue and bone marrow stromal cells [194]. HNF4 receptors are another main mediator of FA-mediated signaling which have been discovered to be potential biomarkers in several cancers [195], [196]. In prostate cancer, high level lncRNA LOC100996425 can reduce the expression of HNF4A by binding to HNF4A mRNA to form a silencing complex, which then promotes human prostate cancer DU-145 cell line proliferation and migration [197]. In human colon cancer SW480 and HCT116 cell lines, lncRNA LINC00858 can promote HNF4A binding to the promoter area of WNK lysine deficient protein kinase 2, leading to WNK lysine deficient protein kinase 2 transcription inhibition, and subsequently promoting migration, invasion, and angiogenesis in colon cancer [198].

In atherosclerosis, lncRNA TUG1 can negatively regulate miR-92R to induce the expression of FXR1, leading to the downregulation of apolipoprotein M, which can modulate high-density lipoprotein metabolism to anti-atherosclerosis by facilitating S1P binding to the S1P receptor on endothelial cells [199]. LXR is the central regulator of the hepatic lipogenic gene program [200]. The activation of the LXR-SREBP-1c axis is closely related to aberrant hepatic lipogenesis in NAFLD. LncRNA Blnc1, which is involved in regulating thermogenic adipocyte differentiation, is required to express SREBP1c. Studies have demonstrated that endothelial differentiation-related factor 1 can interact with Blnc1, constituting a ribonucleoprotein transcriptional complex with LXR to activate lipogenic gene expression (e.g., SREBP1c) in the liver, resulting in insulin resistance and NAFLD [201]. Meanwhile, LXR also modulates cholesterol homeostasis-related genes and is involved in the pathogenesis of cardiovascular diseases [202]. In western diet or LXR pharmacologically activated mouse models, the expression of LXR-dependent hepatic lncRNA LeXis was significantly increased, resulting in lower serum and hepatic cholesterol levels. However, in SCAP (SREBP2 regulator) deleted mice, LeXis failed to lower cholesterol levels, implying a role for the SREBP2 pathway in the LeXis function [203]. Although LeXis is a mouse lncRNA, a putative orthologue lncRNA, TCONS_00016452, has been annotated in human hepatocyte cell lines. Assessing whether TCONS_00016452 is involved in cholesterol homeostasis in humans may provide new ideas for the treatment of cardiovascular diseases [32]. During the process whereby LXR promotes the expression of ABCA1, which is important for cholesterol efflux, lncRNA MeXis can serve as an amplifier by interacting with and guiding the transcriptional coactivator DDX17 to bind to the ABCA1 promoter in mouse liver [204]. Targeting the LXR/MeXis/ABCA1 pathway may provide a new strategy for treating or preventing atherosclerotic disease by modulating cellular cholesterol transport.

Lipid messengers, chaperones, and receptors are the critical components of lipid signaling. Besides regulating signaling lipids, lncRNAs also influence the expression or function of lipid chaperones and receptors, thus controlling the transfer of the signaling lipids and activation of lipid signaling cascades. Targeting these chaperones and receptors has been proposed as an effective way to modulate lipid signaling pathways and treat some lipid metabolism-related diseases. Mining the lncRNAs involved in lipid signaling may provide more novel strategies.

Targeting lncRNAs to diagnose and treat lipid metabolism-related diseases

As the most plentiful component of the mammalian transcriptome, the pivotal roles of lncRNAs have been shown in multiple essential metabolic processes. By regulating the key enzymes of lipid metabolism, modulating the transport of signaling lipids, and activation of lipid signaling pathways, lncRNAs not only control the synthesis, catabolism, and efflux of multiple lipids but also serve as regulators in lipids signaling pathways in various physiopathological conditions. Due to their essential and specific functions, lncRNAs are considered as novel therapeutic targets and biomarkers for clinic diagnosis and surveillance of lipid metabolism-related diseases, such as cancer, obesity, diabetes, and cardiovascular disease. A comprehensive understanding of lncRNAs in modulating body metabolism may provide new targets for the clinical diagnosis and treatment of metabolic diseases [205]. Considering the emerging function of lncRNAs in specific pathological processes, researchers and clinicians have sought to develop lncRNA-based diagnostic and prognostic biomarkers for lipid metabolic diseases. For example, in a clinical trial focused on the early diagnosis of vascular dementia, researchers sought to investigate circulating lncRNA expression and exosomal RNAs in stroke patients with/without cognitive dysfunction or dementia through next-generation sequencing technology. Meanwhile, they proposed to evaluate cerebral blood flow, metabolism, and oxygenation through the application of NIRS, a device that measures tissue oxygenated and deoxygenated hemoglobin concentration in tissues. A lncRNA-exosome RNA-NIRS-based scoring system for cognitive dysfunction was to be tested in a large independent validation cohort. This study would establish a set of lncRNA-based diagnostic and prognostic biomarkers to improve clinical prevention and therapeutic care for stroke-induced dementia (NCT03152630). Considering that lipid metabolic reprogramming is essential for tumor initiation and progression, targeting this process would be a potential anti-cancer strategy. At present, several drugs targeting lipid metabolic reprogramming have been developed. For example, the FAS inhibitor TVB-2640 is in phase II clinical trials (NCT02876796). The HMGCR inhibitors, also known as statins, have been widely used as cholesterol-metabolism-targeting agents in clinical studies of various cancers, such as CRC, prostate cancer, and multiple myeloma [206]. Notably, lncRNAs are also involved in the action of these antitumor drugs targeting lipid metabolism. Metformin has been evaluated as a potential antitumor agent with few side effects [207]. In breast cancer, metformin can inhibit autophagy by modulating lncRNA H19, thus inducing ferroptosis [208].

The unique functions and mechanisms of lncRNAs in specific diseases are continuously being revealed, which provides abundant potential drug target candidates for the novel lncRNA drug arena. Knock down of deleterious lncRNAs is a major strategy for targeting lncRNAs involved in disease. siRNAs can efficiently deplete cytoplasmic lncRNAs, which have been used effectively in multiple preclinical models to explore the therapeutic implications in metabolic diseases. For example, systematic administration of lncRNA DANCR siRNA nanoparticles in triple-negative breast cancer mouse models were shown to notably impede tumor progression [209]. The use of siRNAs to target cytoplasmic lipid metabolism-related lncRNAs, such as PVT1, HCP5, NEAT1, and H19, may be an effective means to improve dysregulated fatty acid metabolism. However, some limitations still remain, such as the immunogenicity of exogenous RNA. Clinical trials of siRNAs targeting VEGFR1 (NCT00499590 and NCT00395057) demonstrated that the anti-angiogenic effects of these siRNAs were more attributable to direct stimulation of TLR3 than to the expected silencing effect, thus these clinical trials were terminated [210]. Off-target effects are another potential problem. siRNAs targeting BCL2 have been found to exhibit a number of off-target effects, including the abnormal expression of stress-response and proliferation-associated genes [211]. In contrast to siRNAs, ASOs are chimeric RNA/DNA oligonucleotides that cleave complementary target nuclear lncRNAs [212]. In some benign solid tumors, ASOs have demonstrated excellent ability in administrating lncRNAs [213]. For instance, in humanized hepatocellular carcinoma models, interfering with the expression of lncRNA HAND2-AS1 shows significant antitumorigenic effects [214]. In bladder, cervical, and pancreatic cancer cells, lncRNA MALAT1 localizes in the nucleus. The conventional RNAi method cannot be employed to knock down it effectively, but MALAT1 ASOs have worked well in a preclinical cancer model [61], [215]. Other nuclear lipid metabolism-related lncRNAs, such as RP11-728F11.4, AT102202, and KCNQ1OT1, may also be effectively blocked by ASOs. However, the utilization of ASOs also needs to overcome the off-target or undesired on-target effects. For example, some patients receiving XIAP-targeted ASO have experienced chemotherapy-induced neuropathy due to downregulation of XIAP in oligodendrocytes, glial cells, or neuronal cells rather than cancer cells [216]. For this reason, researchers have addressed improving the safety, stability, and efficacy of oligonucleotides, for example through the use of locked nucleic acids [217]. For RNA-based therapeutic approaches, efficient delivery is one of the biggest challenges. Lipid nanoparticles, polymers, RNA conjugation, metal-based nanoparticles, and virus-based delivery methods have entered clinical trials, and several new delivery methods, such as exosome-mediated RNA delivery and bacteriophage and bacterial minicell delivery vehicles, are also being developed [218].

Meanwhile, small-molecule compounds that target lncRNAs can bind or block specific lncRNAs structures. LncRNA GAS5 is closely related to diabetes, and it is a high-profile drug target [219]. It has been shown that the small-molecule compound NP-C86 can disrupt the interaction between GAS5 and its transcriptional repressor, undifferentiated embryonic cell transcription factor 1, to elevate the level of GAS5, subsequently recovering the GAS5 level to that of normal levels and alleviating insulin resistance [220]. MALAT1 contains a triple helix structure at 3′-end, and diphenylfuran-based small molecules have been found to specifically target the triple-helix structure in MALAT1 [221]. However, some lncRNAs with dual functions such as MALAT1, which has both anti-oncogenic and oncogenic functions, should be only be used with caution. Compared with siRNAs and ASOs, small-molecule compounds are cheaper to produce and provide more convenient administration modes. However, small molecules work in a structure-specific manner and the large-scale development and utilization of small molecules still require improved techniques for analyzing the specific structure of lncRNA, and an understanding of the interaction between lncRNA and the targeted molecule is essential. Besides small synthetic molecules, plant-derived natural compounds have demonstrated a beneficial regulatory effect on lncRNAs that are related to metabolic diseases [222]. In traditional Chinese medicine, using phytochemicals present in a large number of herbal sources to treat diseases has a long history. Berberine, a coptis extract, has been identified to alter the expression of 538 lncRNAs in non-alcoholic fatty liver disease [223]. However, because of the lack of understanding of the specific targets and mechanisms, the utilization of many phytochemicals has been limited. LncRNAs may provide an innovative perspective to explain the bioactivity of these phytochemicals and possibly participate in clinical trials.

Future directions

In recent years, the rapid development of extracellular vesicle research has expanded the mechanisms underlying intercellular communication networks. It has been identified that lncRNAs can also be packaged into extracellular vesicles and act as messengers to mediate intercellular communication. However, the physiological and pathological role of lncRNAs from extracellular vesicles in lipid metabolism and signaling remains to be further investigated. Circulating lncRNAs in extracellular vesicles can be used as liquid biopsies and non-invasive biomarkers for early diagnosis, prognosis, and treatment of lipid metabolism-related diseases [224], [225], [226]. Moreover, extracellular vesicles have been identified as effective delivery vehicles for non-coding RNAs due to their good targeting property and biocompatibility [227], [228]. Extracellular vesicles extracted by chemically induced membrane blebbing, or from bovine milk, may be applied for the targeted delivery of lncRNAs or siRNAs/ASOs targeting specific lncRNAs for the treatment of lipid metabolism-related diseases [218].

Poor conservation is an “inconvenient truth” for lncRNA studies. The diversity of lncRNAs sequences, localization, and function in primate and non-primate animals limits the development of relevant therapeutic approaches. Nearly 80 % of lncRNAs are primate-specific, thus it is necessary to develop a humanized animal model that expresses the primate-specific lncRNA sequences and target genes [229]. In addition to an animal model, patient-derived organoids provide an in vitro model of human disease that more closely simulates the physiological state and can be studied in the laboratory to better understand the causes of disease and identify potential therapeutic strategies. In oncology research, diverse organoids have been used to elucidate the role of lncRNAs in cancer development [230], [231], [232].

To overcome the off-target effect of RNAi therapeutics, many improvements in RNAi design have already enhanced on-target specificity. Chemical modifications and locked nucleic acid modifications have improved the efficacy of antisense therapeutics. In recent years, circular siRNAs have attracted significant attention due to their higher stability and knockdown efficacy, as well as increased cellular uptake. Targeting lncRNAs using circular siRNAs may provide an alternative solution for lipid metabolism-related diseases.

We have highlighted the functions and mechanisms of lncRNAs in lipid metabolism and signaling using recent exemplars. Compared with the substantial number of lncRNA loci, our current knowledge of lncRNAs in metabolism is just the tip of the iceberg. The rest of the iceberg is a treasure chest waiting to be discovered.

Availability of data and materials

Not applicable.

Funding

This work was supported by National Key Research and Development Project of China (2020YFA0509400, 2020YFC2002705); Guangdong Basic and Applied Basic Research Foundation (2019B030302012); 1·3·5 project for disciplines of excellence, West China Hospital, Sichuan University (ZYGD22007, ZYJC21004); Chinese NSFC (82130082, 81821002, 81790251, 82103168).

CRediT authorship contribution statement

Jiufei Duan: Investigation, Writing – original draft, Writing – review & editing, Visualization. Zhao Huang: Investigation, Writing – original draft, Writing – review & editing, Visualization. Edouard C. Nice: Writing – review & editing, Visualization. Na Xie: Conceptualization, Supervision. Mingqing Chen: Conceptualization, Supervision. Canhua Huang: Conceptualization, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Biographies

Jiufei Duan Jiufei Duan is a PhD candidate in West China School of Basic Medical Sciences and Forensic Medicine, Sichuan University. His research interests focus on the role of reactive oxygen species and noncoding RNA in tumorigenesis.

Zhao Huang Dr. Zhao Huang received his PhD degree from State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University. Dr. Huang’s research interests focus on investigating the role of oxidative stress and autophagy in tumorigenesis.

Edouard C. Nice Professor Ed Nice is Head of Clinical Biomarker Discovery and Validation at Monash University and scientific advisor to the Monash Antibody Technologies Facility (MATF), for which he was director from 2009–2013. He is also Visiting Professor at Sichuan University/West China Hospital and an Adjunct Professor at Macquarie University. His research interests are in protein and peptide micropurification, biomarker discovery and validation, SPR analysis, high throughput monoclonal antibody production and validation, and clinical biomarker assay development, with a strong translational focus on colorectal cancer, especially the development of faecal proteomics for colorectal cancer detection and surveillance.

Na Xie Dr. Na Xie received his PhD degree from State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University. Dr. Xie’s research interests focus on investigating oxidative stress, metabolic regulation and immune response during virus-induced tumorigenesis and development.

Mingqing Chen Professor Mingqing Chen is the principal investigator in Hubei Key Laboratory of Genetic Regulation and Integrative Biology, School of Life Sciences, Central China Normal University. His research focuses on the health effects of indoor environmental pollutants.

Canhua Huang Professor Canhua Huang has been a principal investigator in State Key Laboratory of Biotherapy, West China Hospital, Sichuan University since 2005. Professor Huang is the chief scientist for Redox Proteomics, National Basic Research Program of China (973 Program). His research focuses on oxidative stress and redox regulation tumorigenesis and tumor progression using proteomics approaches. He is a member of Chinese Human Proteome Organization (CNHUPO), and also serves on the Editorial Board of Proteomics and Signal Transduct Target Ther.