The potential role and mechanism of circRNA/miRNA axis in cholesterol synthesis

Cholesterol levels are an initiating risk factor for atherosclerosis. Many genes play a central role in cholesterol synthesis, including HMGCR, SQLE, HMGCS1, FDFT1, LSS, MVK, PMK, MVD, FDPS, CYP51, TM7SF2, LBR, MSMO1, NSDHL, HSD17B7, DHCR24, EBP, SC5D, DHCR7, IDI1/2. Especially, HMGCR, SQLE, FDFT1, LSS, FDPS, CYP51, and EBP are promising therapeutic targets for drug development due to many drugs have been approved and entered into clinical research by targeting these genes. However, new targets and drugs still need to be discovered. Interestingly, many small nucleic acid drugs and vaccines were approved for the market, including Inclisiran, Patisiran, Inotersen, Givosiran, Lumasiran, Nusinersen, Volanesorsen, Eteplirsen, Golodirsen, Viltolarsen, Casimersen, Elasomeran, Tozinameran. However, these agents are all linear RNA agents. Circular RNAs (circRNAs) may have longer half-lives, higher stability, lower immunogenicity, lower production costs, and higher delivery efficiency than these agents due to their covalently closed structures. CircRNA agents are developed by several companies, including Orna Therapeutics, Laronde, and CirCode, Therorna. Many studies have shown that circRNAs regulate cholesterol synthesis by regulating HMGCR, SQLE, HMGCS1, ACS, YWHAG, PTEN, DHCR24, SREBP-2, and PMK expression. MiRNAs are essential for circRNA-mediated cholesterol biosynthesis. Notable, the phase II trial for inhibiting miR-122 with nucleic acid drugs has been completed. Suppressing HMGCR, SQLE, and miR-122 with circRNA_ABCA1, circ-PRKCH, circEZH2, circRNA-SCAP, and circFOXO3 are the promising therapeutic target for drug development, specifically the circFOXO3. This review focuses on the role and mechanism of the circRNA/miRNA axis in cholesterol synthesis in the hope of providing knowledge to identify new targets.


Introduction
Cholesterol is an important component of vertebrate organisms' membrane and plasma lipoproteins and regulates membrane fluidity and permeability. Cholesterol is also a precursor of steroid hormones, bile acids, and vitamin D. However, plasma cholesterol levels have been firmly the initiating factor of atherosclerosis, cardiovascular disease (ASCVD), and cancer, which are the leading causes of disease and death worldwide. Therefore, controlling cholesterol levels is essential for preventing and treating atherosclerosis [1][2][3]. The human body gets 300-500 mg of cholesterol from the diet every day and produces about 700-900 mg of cholesterol from scratch [4]. Approximately 50% of endogenous Ivyspring International Publisher cholesterol is synthesized in the liver. HMG-CoA reductase (HMGCR) is the rate-limiting enzyme in cholesterol synthesis. Statins, which are the HMGCR inhibitors, have been widely used for the treatment of ASCVD. Statins also increase survival rates for cancer patients. However, the efficacy of statins was limited by compensatory increases in HMGCR protein.

CircRNA_ABCA1
CircRNA_ABCA1 (also named circRNA_36781) is located in the exonic of ABCA1. CircRNA_ABCA1 expression was increased in aortic vessels of HFDinduced apoE-/-mice and H2O2-induced mouse aortic endothelial cells (MAECs) injury model, suggesting that circRNA_ABCA1 is a potential diagnostic biomarker for atherosclerosis. CircRNA_ ABCA1 could serve as miR-140-3p sponge increasing vascular endothelial injury and atherosclerosis by regulating the miR-140-3p/MAP2K6 axis [119]. MiR-140-3p also suppressed cholesterol biosynthesis by binding and suppressing the 3'UTR of HMGCR and HMGCS1 [120], suggesting that circRNA_ABCA1 promoted cholesterol biosynthesis by regulating miR-140-3p/HMGCR and HMGCS1 axis. It is worth mentioning that ABCA1 promotes cholesterol efflux to apolipoprotein A-I (apoA-I) to suppress foam cell formation. Previous studies from our laboratory and others have shown that ABCA1 promoted cholesterol efflux to suppress foam cell formation and atherosclerosis development [121][122][123][124][125]. However, the role of circRNA_ABCA1 on ABCA1 expression and cholesterol efflux remains unclear.

CircRNA/miR-133b and miR-221-5p/SQLE axis
CircRNA/miR-133b/SQLE axis Zeste homolog 2 (EZH2) could encode circRNAs, including circEZH2 (also named hsa_circ_0006357) and hsa_circ_0008324. Many studies have shown that EZH2 plays a crucial role in cholesterol synthesis and atherosclerosis development. EZH2 siRNA and inhibitors promoted cholesterol synthesis by enhancing multiple genes expression, including HMGCS1, FDFT1, SQLE, LSS, CYP51A1, DHCR7, DHCR24, and HMGCR [135], suggesting that EZH2 suppressed cholesterol synthesis. However, EZH2 promoted atherosclerosis development in vivo. Specifically, myeloid EZH2 deficiency reduced atherosclerosis development by reducing neutrophil migration and macrophage foam cell inflammatory responses, such as nitric oxide (NO), IL-6, and IL-12 [136]. promoter region and then reduced cholesterol efflux to promote foam cell formation and atherosclerosis development [137,138]. EZH2 regulated miR-139-5p methylation and its target STAT1 expression through H3K27me3 and then promoted ox-LDL-induced HASMCs apoptosis, plaque formation, and inflammatory response in atherosclerosis mice [139]. EZH2 promoted the expression of MMP2 and MMP9 and their-mediated migration of aortic smooth muscle cells (MASMCs) and atherosclerosis development by promoting the methylation of TIMP2 [140]. As mentioned above, EZH2 is a parental gene of circEZH2 [141,142]. suggesting that circEZH2 may regulate cholesterol synthesis and atherosclerosis development by regulating EZH2 expression. However, more studies are needed.
CircRNAs/miR-221-5p/SQLE axis Sterol regulatory element binding protein (SREBP) cleavage activating protein (SCAP) could encode circRNAs, including circRNA-SCAP (also named circSCAP, hsa_circ_0001292), has_circRNA_ 103352, hsa_circ_0065214, hsa_circ_0007291. These circRNAs are located in the exonic of SCAP. SCAP also regulated cholesterol synthesis. SCAP could bind to SREBPs and form SCAP-SREBP complex. When cholesterol in the endoplasmic reticulum (ER) is too low (below 5 %), SCAP binds to the Coat Protein complex II (COPII) protein and escorts the SCAP-SREBP complex from the ER to the Golgi. After several conformational changes, SREBP2 separates from the SCAP-SREBP2 complex and enters the nucleus. SREBPs promoted cholesterol synthesis genes by binding to HMGCR and SQLE [153]. As mentioned above, circRNA-SCAP is located in SCAP, suggesting that circRNA-SCAP may regulate cholesterol synthesis by regulating SCAP expression and SCAP-SREBP2 complex.

Conclusions and Future Directions
Many genes play a central role in cholesterol synthesis, including CYP51, DHCR7, DHCR24, EBP, FDFT1, FDPS, HMGCR, HMGCS1, HSD17B7, IDI1/2, LBR, LSS, MSMO1, MVD, MVK, NSDHL, PMK, SC5D, SQLE, and TM7SF2. Many circRNAs regulate cholesterol synthesis by regulating ACSL1, DHCR24, HMGCR, HMGCS1, PTEN, SQLE, and YWHAG expression by sponging miRNAs. Some circRNAs were also involved in other atherosclerotic risk factors (Table 3). Notable, CYP51, EBP, FDFT1, FDPS, HMGCR, LSS, and SQLE, are promising therapeutic targets for drug development due to many specific inhibitors have been approved and entered into clinical research by targeting these genes. Many circRNAs regulated cholesterol biosynthesis by regulating HMGCR expression via sponging miR-122. Several drugs targeting miR-122 have completed the phase II trial for the treatment of HCV infections, including Miravirsen and RG-101. Thus, the circRNA/miR-122/HMGCR axis is a promising therapeutic axis for drug development. However, several interesting and critical tasks remain to be explored: (1) The naming of circRNA is not uniform and even a little confusing, such as HMGCS1 could encode five circRNA, including hsa_circ_0072391, hsa_circ_0072387, hsa_circ_0008621, hsa_circ_ 0072389, and hsa_circ_0072386. However, hsa_circ_ 0072391 is also named circ-HMGCS1 or circHMGCS1, while hsa_circ_0008621 is also named circHMGCS1-016. SCAP could encode circRNAs, including hsa_circ_0001292, has_circRNA_103352, hsa_circ_ 0065214, and hsa_circ_0007291. However, only hsa_circ_0001292 is also named circRNA-SCAP or circSCAP. (2) Several circRNAs not only promoted cholesterol biosynthesis but also promoted cholesterol efflux or suppressed proinflammatory cytokines, including circUGGT2, circ-PRMT5, circ-HMGCS1, circOgdh, circMED12L, circ_FURIN, circ_CSNK1E, circ0036602, circ-LRP1B, circHUWE1, circITGA7, circNFIX, circRNA-CIDN, circ_0009910, circ_0039569, hsa_circ_0018069, and circPTK2. The role of these circRNAs in atherosclerosis remains to be investigated in vivo. (3) The state of the disease may affect circRNAs studies, such as circHMGCS1-016. CircHMGCS1-016 may exhibit an anti-atherogenic effect in the early stages of the disease. However, as the disease progresses, circHMGCS1-016 may exhibit a pro-atherogenic effect. The development of drugs and diagnostic reagents must consider the state of disease progression. (4) CircRNAs regulate gene expression through various mechanisms, including sponge miRNA, protein scaffold and sponge, encoding protein, and regulation of splicing and transcription. However, so far, almost all circRNAs regulate cholesterol synthesis genes through sponge miRNA. Whether there are other mechanisms is not clear. (5) Until now, most circRNA's role in cholesterol synthesis has been studied in vitro. However, there are many factors influencing the development of the disease. The effect of circRNAs on the disease still needs to be studied in vivo. (6) Given that inhibits miR-122 completed the phase II trial, circFOXO3 is a promising target for drug research by sponging miR-122. However, more studies are needed. (7) Many drugs have been approved for market by targeting HMGCR and SQLE expression. Several circRNAs may be promising therapeutic targets for drug development by targeting HMGCR and SQLEM, such as circRNA_ABCA1, circ-PRKCH, circEZH2, and circRNA-SCAP. However, more studies are needed. (8) The development of new drugs usually requires preclinical studies in multiple animal models before clinical application to improve drug development's success rate. The development of circRNAs drugs also requires much research. (9) Current methods of circRNA synthesis are limited by low cyclization efficiency and the high cost of enzymes and other reagents. There is an urgent need to address these issues. (10) The current study has shown that circRNAs have many targets. However, it may be caused by different dosing doses. Whether there are multiple targets in vivo still needs much research.
In summary, drugs that target CYP51, EBP, FDFT1, FDPS, HMGCR, LSS, SQLE, and miR-122 have entered the stage of market or clinical trials. CircRNA_ABCA1, circ-PRKCH, circEZH2, circRNA-SCAP, and circFOXO3 are promising therapeutic targets for drug development, specifically circFOXO3. With the progress of science and technology, the deepening of research, and the cooperation of scientific research, we believe there will be the clinical application of circRNAs agents soon.  [196,198] miR-503-5p/smurf1, smurf2, Smad7 axis Suppressed proinflammatory cytokines and adhesion molecules level [179,197]

Author contributions
WC and JX participated in the writing-original draft. YW, MY, DW, and XH participated in supervision, and resources. BL, CS, LL, and WH participated in formal analysis and investigation. YS and DX participated in conceptualization, writing review & editing, project administration, and funding acquisition. All authors read and approved the final version of the manuscript.