R-α-Lipoic Acid and 4-Phenylbutyric Acid Have Distinct Hypolipidemic Mechanisms in Hepatic Cells

The constitutive activation of the mechanistic target of rapamycin complex 1 (mTORC1) leads to the overproduction of apoB-containing triacylglycerol-rich lipoproteins in HepG2 cells. R-α-lipoic acid (LA) and 4-phenylbutyric acid (PBA) have hypolipidemic function but their mechanisms of action are not well understood. Here, we reported that LA and PBA regulate hepatocellular lipid metabolism via distinct mechanisms. The use of SQ22536, an inhibitor of adenylyl cyclase, revealed cAMP’s involvement in the upregulation of CPT1A expression by LA but not by PBA. LA decreased the secretion of proprotein convertase subtilisin/kexin type 9 (PCSK9) in the culture media of hepatic cells and increased the abundance of LDL receptor (LDLR) in cellular extracts in part through transcriptional upregulation. Although PBA induced LDLR gene expression, it did not translate into more LDLR proteins. PBA regulated cellular lipid homeostasis through the induction of CPT1A and INSIG2 expression via an epigenetic mechanism involving the acetylation of histone H3, histone H4, and CBP-p300 at the CPT1A and INSIG2 promoters.


Introduction
Although the association between LDL cholesterol (LDL-C) and coronary heart disease (CHD) risk is prominent in observational studies and clinical trials, the attention to elevated triacylglycerides (TAG) and low HDL cholesterol (HDL-C) has increased with respect to risk estimation and therapeutic targets for intervention [1]. Owing to the metabolic interconnection between TAG-rich lipoproteins, HDL and insulin resistance, hypertriglyceridemia is frequently observed in combination with low HDL-C and type 2 diabetes. In clinical trials, subjects with hypertriglyceridemia and an elevated LDL-C/HDL-C ratio have a disproportionately higher risk of CHD events [2,3]. Hypertriglyceridemia, which affects~30% of U.S. adults [4], is defined as fasting blood TAG levels above 150 mg/dL (~1.8 mM) and according to the Adult Treatment Panel III guidelines, 150-199 mg/dL is considered borderline high, 200-499 mg/dL is considered high, and >500 mg/dL is considered very high [5]. Hepatic overproduction of VLDL is believed to be the main contributor to hypertriglyceridemia [6]. The accumulation of tissue TAG leads to increased synthesis and secretion of VLDL from the liver and impedes chylomicron (CM) hydrolysis due to competition of CM and VLDL in binding to lipoprotein lipase. This results in the delayed clearance of CM remnants within the circulation and raises blood TAG. It is critically important to identify new therapeutic strategies and dietary molecules capable of modulating TAG, VLDL and LDL metabolism simultaneously, and elucidate their regulatory points and mechanisms of action.
R-α-Lipoic acid (LA) is a naturally occurring enzyme cofactor synthesized from octanoic acid in most prokaryotic and eukaryotic microorganisms as well as plant and animal mitochondria, encouraging potential. Feeding LA to Zucker Diabetic Fatty rats increased liver LDLR protein levels 2-fold, decreased plasma PCSK9 by 70%, and significantly lowered blood cholesterol [39].
Previous work from our group demonstrated that the hyperactivation of the mechanistic target of rapamycin complex 1 (mTORC1) leads to the overproduction of apoB-containing TAG-rich lipoproteins in HepG2 cells [40], a phenomenon ameliorated by LA and PBA. mTORC1, an evolutionarily conserved serine/threonine kinase, integrates signals coming from nutrients and growth factors to regulate protein, nucleotide and lipid syntheses. Growth factors regulate mTORC1 activity through the PI3K/Akt and/or Ras/MAPK signaling pathways, which converge on the tuberous sclerosis complex 1/2 (TSC1/2) to suppress its GTPase activity upon Rheb (Ras homolog enriched in brain) [41,42] ( Figure 1). The deletion of either TSC1 or TSC2 increases the fractional GTP load of Rheb and activates mTORC1, whereas the elimination of scaffold protein Raptor abolishes mTORC1 activity. mTORC1 is required for insulin-mediated synthesis and processing of SREBP1 mRNA in rat hepatocytes [43]. In addition, mTORC1 activates SREBP1 through ribosomal protein S6 kinase beta-1 (p70S6K1), which phosphorylates lipin 1, thereby preventing its translocation to the nucleus and its inhibition of SREBP1 transcriptional activity [44]. However, the negative feedback inhibition of mTORC1 on the IRS/Akt pathway contributes to repressing lipid synthesis through Akt/INSIG2/SREBP1 when mTORC1 is constitutively active. The negative feedback is apparent in liver-specific TSC1-knockout mice, which are protected from diet-induced hepatic steatosis; they display impaired SREBP1 activity and lipogenesis [45,46], suggesting INSIG2 is a therapeutic target.

Reagents
R-α-lipoic acid (LA) from MAK Wood (#RALA1100134, Grafton, WI, USA); 4-Phenylbutyric acid (PBA) was from Acros Organics (#130,380,250, Thermo Fisher Scientific, Waltham, MA, USA); DMSO was from ATCC (4-X ™ , Manassas, VA, USA); and SQ22536 from Cayman Chemical (#13339, Ann Arbor, MI, USA). LA was dissolved in DMSO to a stock concentration of 200 mM and diluted to a final concentration of 200 µM. PBA was dissolved in DMEM medium to a final concentration of 8 mM. SQ22536 was dissolved in DMSO to a stock concentration of 50 mM and diluted to a final concentration of 100 µM. The final DMSO concentration did not exceed 0.2%.

Trichloroacetic Acid (TCA) Precipitation of Proteins Secreted in Conditioned Media
Conditioned media were spun to remove cell debris. Equal volumes of spun media and 20% (w/v) TCA solution were mixed, vortexed, incubated on ice for 30 min and centrifuged (15,000× g, 10 min, 4 • C). The supernatant was carefully aspirated leaving the protein pellet to which 100 µL of cold acetone was added, vortexed and centrifuged (15,000× g, 10 min, 4 • C) to remove traces of TCA. After three acetone washes, the protein pellet was resuspended by sonication in Laemmli sample buffer and heat denatured (95 • C, 5 min) prior to Western blotting.

Quantitative Real-Time PCR (qRT-PCR)
Total RNA was isolated from cells using Bio-Rad Aurum TM Total RNA mini kit (Bio-Rad #7326820, Hercules, CA), which includes an on-column DNase I treatment. First-strand cDNA was synthesized in the presence of oligo (dT) and random primers with iScript TM Reverse Transcription Supermix (BioRad #1708841). qRT-PCR was performed on a BioRad CFX96 Real-Time PCR Detection System using Sso Advanced ™ Universal SYBR ® Green Supermix (BioRad #1725272). Amplicon authenticity was confirmed by melt curve analysis and agarose gel electrophoresis. PCR efficiencies were assessed with serial dilutions of the template (0.0064-100 ng cDNA/reaction) and 0.3 µM of each primer, and by plotting the quantification cycle (Cq) vs. log amount of template. PCR efficiencies between target genes and housekeeping gene cyclophilin A (PPIA) were comparable, thus unknown amounts of target in the samples were determined relative to PPIA. For each primer pair, PCR controls included reaction containing total RNA instead of cDNA as template, or no template at all. Primer sequences are provided in Supplementary Table S1.

Chromatin Immunoprecipitation (ChIP) Assay
ChIP assays were performed as previously described [26]. Huh7-shTSC2 cells were crosslinked with 37% formaldehyde and chromatin was extracted. Samples were sonicated using Brandson SFX 550 to produce 250-700 bp chromatin fragments. After pre-clearing the sheared chromatin, samples were subjected to immunoprecipitation using the following the antibodies: anti-acetyl-CBP [Lys1535]/p300 [Lys1499] (Cell Signaling Technology #4771S, Danvers, MA, USA), anti-acetyl-histone H3 (Millipore #06-599, Burlington, MA, USA), anti-histone H3 (Millipore #07-690). Rabbit IgG was used as negative control and input (10%) was used as positive control. DNA was eluted and used for qRT-PCR analysis using the primers designed for the first intron of CPT1A and promoter region of INSIG2 (Supplementary Table S2). Acetyl CBP/p300 occupancy in the first intron of CPT1A was normalized to input DNA, acetyl-histone H3 was normalized to histone H3 abundance and input DNA.

Statistical Analysis
Results are shown as means ± SEM. Statistical significance was determined by unpaired two-tailed Student's t-test or one-way ANOVA followed by Tukey's multiple comparisons test to detect differences between groups using GraphPad Prism software. All statistical tests were performed at the 5% significance level.

cAMP Mediates the Lipid-Lowering Action of LA
Generated from ATP by adenylyl cyclase within cell membranes, cAMP partakes in the regulation of lipid-related genes. Given LA and PBA modulate several lipid-related genes (CPT1A, FASN, DGAT1 and DGAT2), we asked whether cAMP signaling was part of LA and PBA mechanisms of action. SQ22536, an inhibitor of adenylyl cyclase, significantly diminished LA-mediated stimulation of CPT1A expression in HepG2-shTSC2 cells revealing that cAMP mediated LA's cellular signaling ( Figure 3A). In contrast, PBA-mediated stimulation of CPT1A expression was unchanged in the presence or absence of SQ22536 ( Figure 3B). Of note, SQ22536 alone repressed FASN and DGAT2 expression ( Figure 3). regulation of lipid-related genes. Given LA and PBA modulate several lipid-related genes (CPT1A, FASN, DGAT1 and DGAT2), we asked whether cAMP signaling was part of LA and PBA mechanisms of action. SQ22536, an inhibitor of adenylyl cyclase, significantly diminished LA-mediated stimulation of CPT1A expression in HepG2-shTSC2 cells revealing that cAMP mediated LA's cellular signaling ( Figure 3A). In contrast, PBA-mediated stimulation of CPT1A expression was unchanged in the presence or absence of SQ22536 ( Figure 3B). Of note, SQ22536 alone repressed FASN and DGAT2 expression (Figure 3).

PBA Acts as HDAC Inhibitor to Drive CPT1A and INSIG2 Expression
Since our results revealed that cAMP was not involved in the mechanism of PBA, we asked whether PBA acted as HDAC inhibitor to remodel chromatin and thus influence gene expression. To that end, we examined the gene expression of CPT1A and INSIG2, the acetylation of histones, and the abundance of acetylated CBP/p300 and acetylated histone H3 at the CPT1A and INSIG2 promoters in Huh7-shTSC2 cells treated +/-PBA (8 mM, 6 h). The ChIP antibody we used detects CBP acetylated at Lys1535 or p300 acetylated at Lys1499. Acetylation of p300 at Lys1499 has been demonstrated to

PBA Acts as HDAC Inhibitor to Drive CPT1A and INSIG2 Expression
Since our results revealed that cAMP was not involved in the mechanism of PBA, we asked whether PBA acted as HDAC inhibitor to remodel chromatin and thus influence gene expression. To that end, we examined the gene expression of CPT1A and INSIG2, the acetylation of histones, and the abundance of acetylated CBP/p300 and acetylated histone H3 at the CPT1A and INSIG2 promoters in Huh7-shTSC2 cells treated +/− PBA (8 mM, 6 h). The ChIP antibody we used detects CBP acetylated at Lys1535 or p300 acetylated at Lys1499. Acetylation of p300 at Lys1499 has been demonstrated to enhance p300 histone acetyltransferase (HAT) activity [49]. Results showed PBA significantly induced the expression of CPT1A and INSIG2, and the acetylation state of histones H3 and H4 ( Figure 4A,B). ChIP data revealed that PBA caused a statistically significant enrichment of acetylated CBP/p300 (Ac-p300: +67% and +55%) and acetylated histone H3 (Ac-H3: +70% and +42%) at the CPT1A and INSIG2 promoter regions indicating PBA modulated transcription through epigenetic modification ( Figure 4C). enhance p300 histone acetyltransferase (HAT) activity [49]. Results showed PBA significantly induced the expression of CPT1A and INSIG2, and the acetylation state of histones H3 and H4 ( Figure  4A,B). ChIP data revealed that PBA caused a statistically significant enrichment of acetylated CBP/p300 (Ac-p300: +67% and +55%) and acetylated histone H3 (Ac-H3: +70% and +42%) at the CPT1A and INSIG2 promoter regions indicating PBA modulated transcription through epigenetic modification ( Figure 4C).

LA, But Not PBA, Suppresses PCSK9 Secretion
Earlier studies demonstrated that LA feeding to rats significantly lowered plasma PCSK9 and enhanced hepatic LDLR [39]. These findings prompted us to examine whether LA and PBA might affect the LDLR signaling in an mTORC1-dependent manner. We found PCSK9 mRNA levels correlated positively with mTORC1 activity in HepG2 cells ( Figure 5A,D). LA (200 µM, 24 h) lowered PCSK9 mRNA in shScramble cells (-22%, p < 0.05, Figure 5A) but did not in shTSC2 or shRaptor cells. However, LA did not significantly change PCSK9 mRNA levels in shTSC2 cells, LA decreased PCSK9 abundance in media (-72%, p < 0.05, Figure 5C). PBA (8 mM, 6 h) did not affect the gene expression, media or cellular levels of PCSK9 ( Figure 5D

LA or PBA Induces LDLR Gene Expression, But Only LA Upregulates LDLR Protein
Since PCSK9 negatively regulates LDL uptake by initiating LDL receptor (LDLR) degradation when binding to LDLR, we sought to determine LDLR status in liver cells treated +/-LA or +/-PBA. LA and PBA upregulated LDLR gene expression, and the effect was more pronounced in HepG2-shTSC2 cells (Figure 6A,C). Moreover, LA significantly increased cellular LDLR abundance ( Figure  6B). Together, the data show that although LA and PBA upregulate LDLR gene expression, LA (but not PBA) upregulates LDLR protein abundance through suppression of PCSK9 secretion. indicates a statistically significant effect of mTORC1 between stable cell lines; Student's t-test, p < 0.05, n = 4.

LA or PBA Induces LDLR Gene Expression, But Only LA Upregulates LDLR Protein
Since PCSK9 negatively regulates LDL uptake by initiating LDL receptor (LDLR) degradation when binding to LDLR, we sought to determine LDLR status in liver cells treated +/− LA or +/− PBA. LA and PBA upregulated LDLR gene expression, and the effect was more pronounced in HepG2-shTSC2 cells ( Figure 6A,C). Moreover, LA significantly increased cellular LDLR abundance ( Figure 6B). Together, the data show that although LA and PBA upregulate LDLR gene expression, LA (but not PBA) upregulates LDLR protein abundance through suppression of PCSK9 secretion.

Discussion
The liver plays a critical role in lipid homeostasis and failure to maintain a balance may cause dyslipidemia in the form of hypertriglyceridemia, low high-density lipoprotein (HDL), or elevated small dense LDL particles [50]. The rise in TAG-rich lipoprotein particles seen in metabolic disorders (e.g., insulin resistance, type II diabetes) results predominantly from the overproduction of apoB100containing VLDL by the liver, competition of VLDL and CM for binding to lipoprotein lipase in peripheral tissues, and inefficient uptake of TAG-rich lipoprotein remnants by hepatic LDLR [51,52]. Thus, it is critically important to identify new therapeutic agents and strategies that would lessen hypertriglyceridemia and also be safe and relatively inexpensive to implement.
We, and others, reported that LA and PBA possess TAG-rich lipoprotein-lowering properties in animal and cell models [17,18,22,39]. We postulated that the molecular mechanisms by which LA and PBA impart their lipid-lowering properties differ [40]. In this study, we explored the molecular mechanisms exerted by LA and PBA to impede TAG-rich lipoprotein overproduction. We first examined the expression of genes involved in lipolysis, lipogenesis, TAG synthesis, and lipoprotein assembly in HepG2 cells. Both LA and PBA significantly decreased SREBP1 and FASN expression and increased CPT1A and INSIG2. To our knowledge, this is the first report showing that PBA upregulates genes that contribute to lowering cellular neutral lipids (CPT1A and INSIG2) while concomitantly repressing lipogenic genes (SREBP1 and FASN). With regards to TAG synthesis and

Discussion
The liver plays a critical role in lipid homeostasis and failure to maintain a balance may cause dyslipidemia in the form of hypertriglyceridemia, low high-density lipoprotein (HDL), or elevated small dense LDL particles [50]. The rise in TAG-rich lipoprotein particles seen in metabolic disorders (e.g., insulin resistance, type II diabetes) results predominantly from the overproduction of apoB100-containing VLDL by the liver, competition of VLDL and CM for binding to lipoprotein lipase in peripheral tissues, and inefficient uptake of TAG-rich lipoprotein remnants by hepatic LDLR [51,52]. Thus, it is critically important to identify new therapeutic agents and strategies that would lessen hypertriglyceridemia and also be safe and relatively inexpensive to implement.
We, and others, reported that LA and PBA possess TAG-rich lipoprotein-lowering properties in animal and cell models [17,18,22,39]. We postulated that the molecular mechanisms by which LA and PBA impart their lipid-lowering properties differ [40]. In this study, we explored the molecular mechanisms exerted by LA and PBA to impede TAG-rich lipoprotein overproduction. We first examined the expression of genes involved in lipolysis, lipogenesis, TAG synthesis, and lipoprotein assembly in HepG2 cells. Both LA and PBA significantly decreased SREBP1 and FASN expression and increased CPT1A and INSIG2. To our knowledge, this is the first report showing that PBA upregulates genes that contribute to lowering cellular neutral lipids (CPT1A and INSIG2) while concomitantly repressing lipogenic genes (SREBP1 and FASN). With regards to TAG synthesis and lipidation of VLDL, LA significantly decreased expression of DGAT1 and DGAT2, which catalyze TAG synthesis from diacylglycerol and FFAs, and MTP, which transfers TAG to nascent apoB to facilitate lipoprotein secretion. In contrast, PBA did not change the transcript levels of DGAT1, DGAT2 and MTP; an observation that is consistent with our earlier work showing PBA decreased secretion of apoB-containing TAG-rich lipoproteins from HepG2 cells without decreasing cellular TAG concentration [40].
As a second messenger, cAMP participates in numerous signaling pathways among which pathways relevant to glucose and lipid metabolism [53]. cAMP stimulates the transcription of lipolytic genes through activation of protein kinase A (PKA) and its substrate, the transcription factor cAMP-response element-binding protein (CREB) [54,55]. cAMP also inhibits lipid synthesis via stimulation of AMP-activated protein kinase (AMPK), which represses mTORC1 activity and SREBP1c-mediated transcription of lipogenic genes (ACC, FASN, and SCD) [56,57], and enhances CPT1α-mediated β-oxidation of long-chain fatty acids [58]. By showing that the inhibition of cAMP production suppressed CPT1A transcription in HepG2 cells, our results are consistent with earlier reports of cAMP inducing CPT1A mRNA in primary rat hepatocytes [59]. The use of SQ22536, an inhibitor of adenylyl cyclase, revealed that LA's lipid-lowering properties in liver cells are in part mediated by cAMP. However, the primary target of LA leading to cAMP remains to be identified, as one of the most potent redox couples, LA(S-S)/LA(SH) 2 is anticipated to interact with extracellular as well as intracellular proteins, primarily those exposed to the oxidized cytoplasmic regions where local concentrations of oxidants (e.g., hydrogen peroxide) build up. In contrast to LA, cAMP did not take part in the mechanism of action of PBA.
The use of SQ22536 also revealed another facet of cAMP-mediated regulation of lipid metabolism; cAMP via CREB contributes to lipogenesis and adipogenesis. Fox et al. showed that the constitutive activation of CREB was sufficient to drive adipogenesis in 3T3-L1 cells, whereas a dominant-negative CREB mutant inhibited this process [60]. Mechanistically, CREB promotes the expression of both PPARγ and C/EBPβ, which are required and sufficient inducers of adipogenesis [61,62]. Here, we provide evidence that the direct blockage of cAMP production leads to a reduction of FASN and DGAT2 transcription. Our results are fully supported by other findings: (i) Antony et al. showed that mRNA levels of Fasn were markedly decreased in the lungs of Creb1 −/− mice [63], (ii) Erion et al. reported a significant decrease of Dgat2 mRNAs in ZDF rat livers following knockdown of Creb1 [64]. Given the dual, seemingly opposite, role of cAMP in lipid metabolism, therapeutic strategies aimed at stimulating cAMP production sustainably may reveal ineffective to control blood lipids and adiposity. In this regard, LA has not been shown to stimulate lipogenesis nor adipogenesis; to the contrary, LA lowers neutral blood lipids and adiposity [16,65,66]. Therefore, it suggests that LA also acts independently of cAMP/PKA.
We previously showed PBA downregulated the secretion of apoB-containing lipoproteins independently of acting as chemical chaperone to relieve ER stress in high glucose-treated HepG2 cells [40]. In the present study, we report that PBA stimulates the acetylation of histones H3 and H4, and histone-associated CBP-p300 to drive CPT1A and INSIG2 expression. Histone acetylation/deacetylation events are essential processes of gene regulation. Acetylation of histone tails by histone acetyltransferases (HAT) relaxes chromatin into a transcription-competent state that facilitates the DNA binding of transcription factors, whereas histone hypoacetylation catalyzed by histone deacetylases (HDAC) is associated with gene silencing [67]. A critical link between lipogenesis, acetyl-CoA availability and histone acetylation was revealed in mammalian cells lacking the acetyl-CoA producing enzyme ATP-citrate synthase (Acly) [68]. In that study, Wellen et al. showed that Acly knockdown correlated with a decrease in acetylated histones H3 and H4, a decrease in transcription of genes involved in glucose uptake and metabolism, and a decrease in 3T3-L1 adipocyte neutral lipid content. A link between histone acetylation and lipogenesis was further established by Feng et al. [69], who showed that hepatic de novo lipogenesis was markedly increased in mice lacking Hdac3 suggesting histone hypoacetylation suppresses lipogenesis; conversely histone acetylation stimulates lipogenic gene expression. Unlike Hdac3 depletion, which caused upregulation of lipogenic genes, our study shows that PBA (an inhibitor of class I and II HDACs; HDAC3 belongs to class I HDACs) selectively upregulated lipid-lowering genes (CPT1A and INSIG2) while downregulating lipogenic genes (SREBP1 and FASN). PBA's selectivity may lie in the transcription factors and transcriptional co-activator proteins it regulates, including PPARα [70], PPARγ [71] and CBP/p300.
We found that constitutively active mTORC1 was associated with the upregulation of LDLR and PCSK9 in HepG2 cells. These changes in gene expression did not translate into significantly elevated levels of LDLR or PCSK9 protein, although PCSK9 trended up under constitutive mTORC1 activation (HepG2-shTSC2 cells). This is not to say that mTORC1 and LDLR are not linked. Liu et al. observed an association between inflammation-mediated activation of mTORC1 and the LDLR pathway in ApoE −/− mice [72]. They reported a positive correlation between LDLR mRNA and p-mTOR (r = 0.802, p < 0.001, which corroborates our data) but a negative correlation between PCSK9 protein and p-mTOR (r = −0.675, p < 0.05), suggesting mTORC1 activation induced LDLR expression but repressed that of PCSK9 in ApoE −/− mice. We did not find this in stably transduced hepatoma cells, suggesting that the mTORC1 signaling and LDLR pathway are connected but PCSK9 is downregulated by the mTORC1 signaling in an ApoE-dependent manner.
A study by Carrier et al. found that LA supplementation increased hepatic LDLR protein two-fold and suppressed plasma PCSK9 by 70% in obese Zucker rats fed a high-fat diet [39]. Consistent with their results, we found LA increased LDLR expression (both mRNA and protein) and lowered PCSK9 secretion in media without changing PCSK9 mRNA or protein expression. PBA markedly enhanced LDLR transcription in its own right, but did not change LDLR protein levels. PBA did not affect PCSK9 expression in stable HepG2 cells. Costet et al. showed that SREBP1 promotes PCSK9 gene transcription by tethering to a sterol response element (SRE) in the PCSK9 promoter [73]. Although LA and PBA repressed SREBP1 transcription in HepG2-shTSC2 cells, it did not lower PCSK9 mRNA levels. Evidence of PBA's therapeutic potential against familial hypercholesterolemia caused by deleterious mutations in LDLR was presented by Tveten et al. [74]. LDLR mutants are functional because mutations do not affect the active site but they fail to localize to the plasma membrane and thus accumulate in the ER. In Chinese hamster ovary (CHO) cells expressing the LDLR (G544V) mutant, one of the predominant mutations in familial hypercholesterolemia, PBA (5 mM, 24 h) increased protein expression of the LDLR mutant on the plasma membrane and restored some LDL endocytic activity.
We conclude that (i) both LA and PBA downregulated de novo lipogenesis in an INSIG2/SREBP1-dependent manner, (ii) both LA and PBA activated CPT1A, (iii) LA downregulated glycerolipid synthesis genes (DGAT1 and DGAT2) and microsomal triglyceride transfer protein (MTP) whereas PBA did not, (iv) LA upregulated LDLR receptor abundance through downregulation of PCSK9 secretion whereas PBA did not, (v) PBA induced CPT1A and INSIG2 expression through epigenetic acetylation of CBP-p300 at the CPT1A and INSIG2 promoters (Figure 7). Importantly, the lipid-lowering functions of LA and PBA operated under mTORC1 hyperactivation; that is, under conditions needing lipid rebalancing. These results warrant further animal and human studies.

Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1, Table S1: Oligonucleotide primer sequences used for qRT-PCR, Table S2: Oligonucleotide primer sequences used for ChIP.  Most animal studies have indicated that LA has hypotriglyceridemic effects, which manifest by a lowering of blood and liver TAG content, and abdominal fat mass. By comparison, research on the TAG-lowering properties of LA in human subjects remains limited. After reviewing the available evidence, it appears that LA is most efficacious (i.e., decreases blood TG by 20-50%) in situations where blood TG is elevated above 200 mg/dL (~2.4 mM) pre-treatment [75][76][77]. In future studies, it will be critically important to consider factors that optimize LA blood plasma levels including the use of bioavailable forms of LA (R-enantiomer, physiologically appropriate salts), the route and timing of administration, as well as the subject characteristics such as the type and severity of hyperlipidemia, other health pre-conditions, gender, age, eating pattern, and lifestyle habits.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.