Mechanisms Mediating the Regulation of Peroxisomal Fatty Acid Beta-Oxidation by PPARα

In mammalian cells, two cellular organelles, mitochondria and peroxisomes, share the ability to degrade fatty acid chains. Although each organelle harbors its own fatty acid β-oxidation pathway, a distinct mitochondrial system feeds the oxidative phosphorylation pathway for ATP synthesis. At the same time, the peroxisomal β-oxidation pathway participates in cellular thermogenesis. A scientific milestone in 1965 helped discover the hepatomegaly effect in rat liver by clofibrate, subsequently identified as a peroxisome proliferator in rodents and an activator of the peroxisomal fatty acid β-oxidation pathway. These peroxisome proliferators were later identified as activating ligands of Peroxisome Proliferator-Activated Receptor α (PPARα), cloned in 1990. The ligand-activated heterodimer PPARα/RXRα recognizes a DNA sequence, called PPRE (Peroxisome Proliferator Response Element), corresponding to two half-consensus hexanucleotide motifs, AGGTCA, separated by one nucleotide. Accordingly, the assembled complex containing PPRE/PPARα/RXRα/ligands/Coregulators controls the expression of the genes involved in liver peroxisomal fatty acid β-oxidation. This review mobilizes a considerable number of findings that discuss miscellaneous axes, covering the detailed expression pattern of PPARα in species and tissues, the lessons from several PPARα KO mouse models and the modulation of PPARα function by dietary micronutrients.


Introduction
As reported in the review by Latruffe and Vamecq [1], peroxisomes are ubiquitous, single membrane-bound organelles. They belong to the fundamental class of intracellular compartments named microbodies. According to the evolutionists, microbodies and eukaryotic cells appeared on Earth around 1.5 billion years ago. Based on their related cell origin, these organelles are defined as glycosomes, glyoxysomes, hydrogenosomes or peroxisomes. Peroxisomes are found in higher vertebrates; glycosomes exist only in trypanosomes; glyoxysomes are found in leaves and seeds; hydrogenosomes are found in anaerobic unicellular ciliates, flagellates, and fungi. The latter two microbody structures belong to lower eukaryotic species, and all these compartments metabolize hydrogen peroxide. According to the endosymbiotic theory, peroxisomes, mitochondria, and chloroplasts may have derived from free-living prokaryotic ancestors. Only mitochondria and chloroplasts are semi-autonomous organelles, containing a DNA genome, which encodes for just some of their proteins.
In the mammalian liver, very-long-chain-fatty acids (VLCFA) are exclusively shortened in peroxisome through a specific β-oxidation system. Then, shortened fatty acids are metabolized by mitochondrial β-oxidation. Peroxisomes also contain the first enzymatic steps of plasmalogen synthesis [1]. In addition, they are involved in maintaining a redox state through the NAD + /NADH balance, linked to the pyruvate/lactate level. In 1965, a milestone was reached by Hess et al. [2] who described, for the first time, hepatomegaly induced by clofibrate (ethyl 2-(4-chlorophenoxy)-2-methylpropanoate) in rats, subsequently established as a peroxisome proliferator in rodents and an activator of the fatty acid peroxisomal β-oxidation [3]. Later, Isseman and Green [4] identified peroxisome proliferators (PPs) as activator ligands of a special class of nuclear receptors termed Peroxisome Proliferator-Activated Receptor α (PPAR α). Afterward, several PPAR isoforms were characterized as members of the superfamily of the nuclear steroid receptors. It is recognized that the phylogenetic origin of PPARs dates back 200 million years to the fishmammalian divergence period [5]. PPARs evolved three times faster than other members of the hormone nuclear receptor superfamily, and are represented now in three isoforms (α, β/δ, and γ).
The second peroxisomal β-oxidation system (Figure 1), converting fatty carboxylates with a 2-methyl branch, such as pristanic acid and bile acid intermediates, includes the 2-methylacyl-CoA-specific oxidases (trihydroxycoprostanoyl-CoA oxidase and pristanoyl-CoA oxidase), the second multifunctional protein (named MFP-2) [11,14,15], and a 58 kDa sterol-carrier protein (SCP-2) containing thiolase activity [6,16] (Figure 1). Although both LBP/MFP-1 and DBP/MFP-2 provide hydratase and dehydrogenase activities, these proteins exhibit opposite stereospecificities. While LBP/MFP-1 hydrates 2-trans-enoyl-CoAs into L-3-hydroxyacyl-CoAs, and dehydrogenates the L-isomers [11,12], the DBP/MFP-2 transforms 2-trans-enoyl-CoAs into D-3-hydroxyacyl-CoA and dehydrogenates the Disomers [6,11,14,15]. Despite the fact that the MFP enzymes are structurally unrelated to each other, both MFPs can hydrate 2-methyl-enoyl-CoAs [14]. The 3-hydroxy isomers formed by MFP-2 have the same (3R, 2R) configuration, or (24R, 25R) configuration in bile acid intermediates, underlining the role of MFP-2 in both pristanic acid degradation and bile acid synthesis [15]. Recently, it was demonstrated that LBP/MFP-1 is indispensable for the β-oxidation of dicarboxylic acids and the production of their medium-chain derivatives [6,15,17,18].   [38]. (A) Sequence logo of the MA1148.1 Jaspar matrix, presenting the conservation of nucleotides from multiple alignments of 1000 PPARα/RXRα PPRE sequences. Adenosine (A), cytidine (C), guanosine (G), and thymidine (T) nucleotides are respectively green-, blue-, yellow-, and red-colored, and the relative size of the letters represents their frequency in the consensus. The total height of a logo position corresponds to the degree of conservation in the corresponding multiple sequence alignment. (B) A table representing the percentage of the four bases for each position of the consensus. The color gradient code highlights the percentage of conservation of bases from blue to red for the whole table.
Among the hundreds of genes known to be regulated by PPARα, eight are encoding enzymes that are commonly localized in the peroxisomal compartment (Figure 3), and belong to the three species: human, mouse, and rat. Table 1 presents validated PPREs sequences for functional genes, of which four (Acox1, Ehhadh, Acaa1b, and Scp2) are encoding very-long-chain fatty acid β-oxidation enzymes, and the remaining genes are Cat, encoding the catalase enzyme, and Mlycd (malonyl-CoA decarboxylase gene), expressing an enzyme with both cytoplasmic and peroxisomal localization. The latter form is believed to be involved in the peroxisomal degradation of malonyl-CoA produced by odd-chain-length dicarboxylic fatty acid β-oxidation [39], and finally, the Pex11a gene, which participates particularly in peroxisome biogenesis (Pex11α).

Figure 3.
Multiple alignments of PPRE sequences, identified in peroxisomal gene promoters, and experimentally proved to be regulated through PPARα binding. Adenosine (A), cytidine (C), guanosine (G) and thymidine (T) nucleotides are respectively green-, blue-, yellow-and red-colored.
The focus here will be on the PPARα isoform. PPARα is a type-II non-steroid ligandregulated nuclear hormone receptor [4,32,62] transcribed from the human PPARA gene, which spans~93.2 kb [53] and consists of eight exons [63]. It has been mapped to chromosome 15 in the mouse DNA and to chromosome 22 in humans [31]. The PPARα protein possesses five main functional domains (A-F) embodied in a modular canonical structure [64] (Figure 4). The N-amino terminal end harbors the activation function-1 (AF-1) or A/B domain, which operates autonomously in a ligand-independent manner. The 65 amino acid-long DNA-binding domain (DBD), or C domain, consists of 2 highly conserved zinc finger-like motifs that promote the receptor's binding to the PPRE sequence of the target genes. The D domain or hinge region that bridges the DBD to the ligand-binding domain (LBD) acts as a docking site for cofactors. In the C-terminal region, the LBD, or E/F domain, is responsible for ligand specificity and contains the activation function 2 (AF-2) [28,65]. This latter contains a tyrosine residue on the helix 12, which plays an ultimate role in interacting with the carboxyl group of the ligands [66]. When a ligand enters the LBD pocket of PPARα, the interface of AF-2 stabilizes and facilitates so that PPARα can recruit coactivators [67]. The LBD is still a center of interest in numerous pharmaceutical investigations. Recent publications to date on studies based on X-ray crystallography, referenced in the protein data bank website (PDB; http://www.pdb.org/, accessed on 1 July 2021), provide fascinating, detailed insight into the LBD domain structure, albeit limited to comparing it with other PPAR receptors. It describes a relatively large Y-shaped hydrophobic cavity in the PPARα-LBD pocket volume of 1400 Å 3 [68], which allows PPARα to interact with a broad range of structurally distinct natural and synthetic ligands [67,69]. Phosphorylation sites are labeled with yellow stars (6,12,21,179,230) amino acids, the corepressor site is marked with a red half-sphere, and the coactivator binding site is shown with a green ring. The panels on top show the number of amino acid residues. (B) Structural function of A/B, C, D, and E/F domains, respectively.

PPARα Ligands
Recently, PPARα-ligands have gained consistent interest in several complex metabolic disease investigations [60,67], such as lipid metabolism disorders. Due to their engagement in physiological and pathophysiological metabolic processes, and their role in activating transcriptional regulatory networks, these ligands are becoming intriguing bona fide treatment opportunities and present a way to unveil many relevant potential roles of PPARα, also known as promising versatile drug targets.
Evidence indicates that a wide variety of lipophilic molecules, the so-called ligands, can activate PPARα, encompassing natural saturated, unsaturated, and polyunsaturated fatty acids (PUFAs) [70,71], and synthetic ligands that are collectively referred to as PPARαactivators [72].
The abovementioned results were recently supported by Chen et al. [108], reporting that feeding animals a diet high in rapeseed oil (rich in erucic acid, a very-long-chain fatty acid) leads to PPARα activation with an adaptive elevation in peroxisomal β-oxidation capacity, which suggested that erucic acid might act as a potential ligand for PPARα. In line with prior communicated data, Maheshwari et al. [109] reported that treating rat Fao cells with a fungal lipid extract rich in monomethyl BCFAs (Conidiobolus heterosporous) increases mRNA levels of the PPARα target genes Acox1, Cyp4a1, Cpt1A, and Slc22A5, strongly suggesting that BCFAs are similarly potent PPARα activators [109]. Taken together, these relevant results from our laboratory and from others all affirm that the peroxisomal βoxidation substrates are potent PPARα ligands that modulate the expression of a battery of lipid-metabolizing enzymes to maintain lipid homeostasis and to alleviate the toxic effect of VLCFA and BCFA overload [57].
To date, various synthetic single, dual and pan agonists, respectively, are in clinical use as medications to treat dyslipidemia, hyperglycemia in patients with Type 2 diabetes mellitus, hypertriglyceridemia, and cardiovascular disease [28,72,91]. Indeed, potent synthetic ligands could elicit both desirable and undesirable side effects. Studies conducted by Preiss et al. [116] proved that the chronic administration of peroxisome proliferators to rodents causes hepatocellular carcinoma, and it may also increase the risk of gallstones and cause anemia and leukopenia [117]. Much of what we know about PPARα-ligands comes from a collective knowledge primarily derived from rodent studies, via the treatment of mice or rats with synthetic PPARα peroxisome proliferators or by using PPARα null mice [98]. It has been reported that human and mouse PPARα have different binding affinities and physiological effects [118] and are diversely activated by specific ligands, including phthalates and fibrates [119]. Nevertheless, these differences are negligible and do not call into question the tenet of the ultimate role that PPARα plays as a general lipid sensor in both species [98].
To date, tremendous efforts are in progress to develop new, highly PPARα-specific ligands with different activation and binding modes that could more selectively activate PPARα-RXRα transcriptional complex assembly, with tissue-selective and gene-selective activities, to reduce unwanted side effects and assure reasonable safety. In parallel, the "micronutrients" found in food that activate PPAR receptors are gaining increasing interest, as nutritional therapy becomes an unstoppable trend for treating lipid disorders [79].

PPARα and Coregulators
The identification of PPAR in the 1990s heralded a new era of biotic and xenobiotic sensing by the liver [4]. The PPAR subfamily of nuclear receptors functions as sensors for fatty acids and fatty acid derivatives and controls critical metabolic pathways involved in lipid and energy metabolism [120,121] and catabolism [122][123][124][125]. The transcriptional activation of genes is a complex process that involves the participation of many transcription factors [126]. While the nuclear receptors (NRs) mediated gene-regulation provide the backbone for the transcription factor-specific gene regulation, coregulators provide the much-needed tissue-, cell-, and species-specific differences in the peroxisome proliferatorinduced pleiotropic responses of PPARα [127,128]. However, we would like to focus this review section on the role of PPARα and its associated proteins in regulating peroxisomal beta-oxidation genes/pathways. Coregulators are proteins that bind to the nuclear receptor by a specific domain LXXLL, a hallmark for all coregulators [129]. Most coregulators have more than one LXXLL domain and are essential for protein-protein interactions between the nuclear receptor and the coregulator [130] (Figure 6). Moreover, each LXXLL could function in a specific nuclear interaction, suggesting that the coregulators are shared between different NRs.
Coregulators can be broadly classified into subgroups, such as essential vs. nonessential, repressors vs. activators, and DNA binding region (DBD) vs. ligand bindingregion (LBD) interacting coregulators [127]. Essential coregulators are proteins deemed critical for the survival of the offspring, and their absence results in embryonic lethality: cAMP-response element-binding protein (CBP); PPAR-interacting protein/activating signal cointegrator 2 (PRIP/ASC2); PPAR-binding protein/mediator complex subunit 1 (PBP/Med1); mediator complex subunit 25 (Med25) [131][132][133]. Non-essential coregulators are proteins with such critical functional responsibilities that they are usually represented by more than one isoform-steroid receptor coactivators (SRCs) [131,132,134], Asp-Glu-Ala-Asp (DEAD)-box helicases [135][136][137], sirtuins (SIRT) [96,97], PPARγ coactivators (PGCs) [138][139][140][141]-and the loss of one isoform is compensated by others. Repressors that bind to the nuclear receptor PPARα in the absence of/or independent of ligands prevent it from binding to the peroxisomal proliferator response elements (PPRE) of the target genes as nuclear corepressor (NCoR) and silencing mediator of retinoic acid and thyroid hormone (SMRT) [96] ( Figure 7A). This group of proteins usually bind to the AF-1 domain of the DNA binding region of the receptor. The ligand-independent coregulators (heat-shock protein-70) could also prevent the PPARα from proteolytic degradation in the cytosol before PPARα could translocate to the nucleus in the activated state [131] and, typically, these proteins bind to the hinge region of the nuclear receptor that interconnects the DNA binding region to the ligand-binding region of the receptor [127]. The activators, on the other hand, could help PPARα zero in onto the specific PPREs of the target genomic region, help attach it to the PPREs with the assistance of nucleosomal-specific functions such as histone methylases (SRC proteins) [142], histone acetyltransferases (CBP/p300) [143,144], DNA-helicases [145], PRIC285 [124], and PRIC320 [146]. The activators would also function by stabilizing the transcriptional complex (PRIP/ASC2 [147]) and potentiate the recruitment of RNA-polymerase complex proteins to the transitional complex (mediator complex, PBP [148,149] ( Figure 7B). Additionally, the activators would consist of proteins responsible for separating the transcribed mRNA from the genomic region (protein-L-isoaspartate (D-aspartate) O-methyltransferase (PIMT) [147]. These proteins activate the AF-2 domain of the nuclear receptor and enhance transcription by linking the liganded nuclear receptor to the basal transcription machinery. We have identified almost all these groups of coregulators using either a direct protein-protein interaction assay, such as a yeast two-hybrid assay [150], GST-pull downs [124], and ligand affinity chromatography [141] to identify the PPARα-interacting proteins and a functional transcriptional activation complex [131]. PPARs, like other nuclear receptors, interact with coactivators such as SRC-1 (steroid receptor coactivator-1) or corepressors such as NCoR and SMRT. PPARαinteracting coactivators and corepressors augment or repress, respectively, the PPARα transactivation activity. Since the cloning of SRC-1 twenty-five years ago, over 300 coactivators/coregulators have been identified, with new members still being added to this expanding spectrum. PPARα is known to interact with some of these coregulators [151]. These include CBP/p300-dependent binding complex [152], members of the SRC/p160 superfamily, members of PBP/MED1 complex (PBP/TRAP220/DRIP205/MED1 [133,149,153], members of PRIP/NCoA6 (ASC2/RAP250/TRBP/NRC), members of PRIC complex PRIC285 [124], PRIC295 [141], PRIC320 [146], PPAR gamma-binding proteins, PGC-1α [147,154], and PGC-1β [155,156], as well as coactivator-associated proteins PIMT [131] (NCoA6IP) and coactivator-associated arginine methyltransferase 1 (CARM-1) [157,158]. The PPARαinteracting coregulator (PRIC) complex isolated from rat liver nuclear extracts reveals many coregulators, presumably forming one mega-complex. An almost similar complex was isolated with ciprofibrate as the ligand in affinity chromatography. This diversity raises several issues about the evolutionary importance of the versatility and complexity of coregulatory molecules, their relative abundance in various cell types, and their affinity for a given nuclear receptor in orchestrating transcription in gene-, cell-, and developmental stage-specific transcription. In the absence of a specific ligand, PPARα interacts with the corepressors NCoR and SMRT, but the importance of PPARα action is not well documented, as endogenous ligands could potentially activate PPARα [159]. The homozygous deletion of NCoR or SMRT in mice is embryonically lethal, indicating that they cannot fully compensate for each other during development [160][161][162]. Furthermore, another corepressor, the receptor-interacting protein 140 (RIP140), which can interact with PPARα, is known to repress the activity of NRs by competing with coactivators and by recruiting downstream effectors such as histone deacetylases (HDACs) [163]. Interestingly, the phenotype of RIP140 knockout mice suggests a role for this corepressor in PPARα signaling, as these mice exhibit resistance to high-fat diet-induced obesity, resulting from the upregulation of genes involved in energy dissipation [163]. Interestingly, hepatic sirtuin 1 (SIRT1) regulates lipid homeostasis by positively regulating PPARα [164,165]. On the other hand, SIRT1 interacts with PPARγ and is regulated by PPARγ in a negative feedback mechanism [166]. SIRT6 binds NCOA2, a PPARα coactivator and part of the SRC family of coactivators; the binding results in the decrease of the acetylation of SRC2/NCOA2 K780 in the liver, thus, interaction with SIRT6 mediates the activation of PPARα and thus the inhibition of SREBP-dependent cholesterol and triglyceride synthesis [167]. The ligand binding to a nuclear receptor triggers a molecular switch that releases corepressors and begins the recruitment of coactivator complexes, such as members of the CBP/p300 family, which exhibit the histone acetyltransferase activity required to facilitate chromatin remodeling. The subsequent recruitment of other coregulators, either singly or as preassembled multisubunit protein complexes, including mediator complex and RNA polymerase machinery, is facilitated by the interaction of the general basal transcription machinery to enhance the transcription of a specific set of genes [97,168]. As discussed previously, coregulators contain an LXXLL motif that forms two turns of the α-helix and binds to a hydrophobic cleft on the surface of the nuclear receptor. The identification and characterization of coregulators have been derived mostly from in vitro experiments, but there is a paucity of information about individual coactivators in vivo cell-and gene-specific functional roles [131]. In the absence of the ligand, PPARα interacts with transcriptional corepressors (NCoR, SMRT, NRTR-P-1) by recognizing the AF2 region. The same process is at the RXR, where AF2 interacts with RIP140 as RXRα corepressor [169]. Due to the chromatin condensed state, the heterodimer cannot bind the PPRE properly. Part (B). In the presence of a PPARα ligand, either a long-chain or very-long-chain polyunsaturated fatty acid, leukotriene LTB4, fibrate or other chemicals [170], and a 9-cis retinoic acid as RXRα ligand, an exchange corepressor/coactivator is made by NCoEX, which suppresses the repressive of corepressor state by ubiquitinylation-inducing degradation by the proteasome system. The fixation of a ligand induces an allosteric LBD conformational change of AF2, allowing the recruitment of coactivators, either NCoR1, p300/CBP, or SRC1 for PPARα, and p120 and NCoPR for RXRα [171]. The CBP-dependent HAT activity induces the remodeling of chromatin and allows the PPARα/RXRα heterodimer to bind to PPRE correctly, then activates the Pol II transcription complex and triggers the transcription of lipid metabolism-encoding genes. Some post-translational modifications of PPARα regulate its activity [172,173]. For instance, phosphorylation stimulates PPARα transcriptional activity [174]. The HNF4α transcription factor recognizes a similar response element as the PPRE and interplay with PPARα [175]. A comparable mechanism has been reported with the Coup-TF transcription factor. While several works consider PGC-1α [176] as an important coregulator of PPARα, it seems to be more specific for PPAR γ. The 15(S)-HETE, 15-hydroxyicosatetraenoic acid, family of arachidonic acid metabolites; 9-cisRA, retinoic acid cis conformation in carbon 9; AF1, activating domain 1; AF2, activating domain 2; CBP, CREBP binding protein; CoPRs, COPR1 and COPR2 as corepressors of PPAR and RXR, respectively; COUP-TF, chicken ovalbumin upstream promoter transcription factor; CTBP-2, C-terminal binding protein-2; DBD, DNA binding domain; HAT, histone acetyl-transferase; HD, hinge domain; HNF-4α, hepatic nuclear factor 4 α; HDAC, histone de-acetyl transferase; LBD, ligand binding domain; LTB4, leukotrien B4; MBP, mono butyl phthalate; MEHP, mono ethyl hexyl phthalate; NCoA1, nuclear receptor coactivator 1; NCoEX, nuclear receptor corepressor Excit; NCoR1, nuclear receptor corepressor 1; NRTP-1, nuclear repressor transcription factor; p120, protein 120 kDa; p300, protein 300 kDa; Pol II, RNA polymerase class II; PGC-1αPPAR γ co-activator-1α; PPRE, peroxisome proliferator response element; PRIP/RAP250, PPAR interacting-protein methyl transferase; PUFA (LC &VLC), polyunsaturated fatty acids (longchain or very-long-chain); RIP140 receptor interacting protein corepressor; SMRT, silencing mediator of retinoid and thyroid receptors; SRC1, steroid receptor coactivator-1.

Metabolic Regulation of the Peroxisomal β-Oxidation Pathways
The regulation of the peroxisomal pathways is mainly associated with the cellular increase of the peroxisome population, which is highly promoted by several diverse natural and synthetic compounds nominated as peroxisome proliferators (PPs). Such compounds raise a peroxisome number quantitatively, mainly in hepatic parenchymal cells, and provoke delayed pleiotropic responses, including the development of hepatocarcinoma in rats and mice [8,131]. Based on several pieces of experimental evidence, earlier reports from Reddy's group proposed a receptor-mediated mechanism to explain the phenomenon of hepatic peroxisome proliferation. Accordingly, the induction of peroxisomal β-oxidation is a consequence of ligand hepatic overload, leading to lipid metabolism dysregulation, accompanied by an augmentation in the extrahepatic lipolysis and a substantial hepatic influx of free fatty acids [96]. Furthermore, the unique pleiotropic responses raised by structurally unrelated peroxisome proliferators in hepatocytes drive a synchronized transcriptional activation of the peroxisomal β-oxidation genes [13,96,131].
Lazarow and De Duve [7] demonstrated previously that clofibrate administration in rat liver strikingly enhances the peroxisomal β-oxidation activity. A similar observation was reported by Hashimoto and coworkers [177], showing that feeding a diet containing a phthalate ester plasticizer di-(2-ethylhexyl)phthalate, a PPARα activator, leads to a 20-fold increase in the expression of peroxisomal β-oxidation enzymes in rat liver. In addition, a previous study reported that synthetic ligands such as WY-14643 exhibited a high affinity to PPARα, compared to the natural endogenous ligand (16:0/18:1-GPC) in the induction of fatty acid β-oxidation [75]. Moreover, Rogue et al. [93] showed that Acox1 and Cpt1A genes in oleic-acid-overloaded HepaRG cells were significantly upregulated from 1 day, and remained at high levels after 14 days, upon treatment with the dual agonist muraglitazar, which stimulates the fatty acid β-oxidation pathway. These results are in close concordance with previous experiments conducted by Lee et al. [126], showing that after feeding hypolipidemic agents to mice lacking PPARα expression, the mutant animals accumulated lipid droplets in their tissues, which strongly supports the idea that PPARα activators promote the transcription of genes involved in the lipid catabolism process.
Structurally, PPs molecules may be chemically unrelated, including hypolipidemic drugs, such as clofibrate, ciprofibrate, gemfibrozil, and Wy-14,643, as well as some nutritional conditions, especially high-fat diet or vitamin E-deficient diet and leukotriene D4 receptor antagonists. In addition, several herbicides, such as 2,4-dicholophenoxyacetic acid or 4-chloro-2-methylphenoxyacetic acid [8,178] and certain phthalate ester plasticizers, induce a similar liver peroxisome proliferation as do prototypic fibrate derivatives. In addition, the administration to rodents of a C19-steroid, dehydroepiendrosterone, promotes peroxisomal fatty acid β-oxidation and peroxisome proliferation [179]. Though the response to PPs has been demonstrated in several tissues from PPs-treated rodents, the hepatic responsiveness is by far the most powerful, accounting for a 10-to 20-fold induction of peroxisomal fatty acid β-oxidation activities, accompanied by a proliferation of peroxisomes and strong hepatomegaly pathogenesis [8,131].
The description of PPARα-target genes shows that this nuclear hormone receptor largely governs those genes involved in hepatic and cardiac muscle transport, oxidation, and the degradation of lipids. Transcriptionally, PPARα activates several genes, including the lipoprotein lipase gene permitting the release of fatty acids from lipoprotein particles [180], genes encoding fatty acid translocase CD36, and fatty acid-binding proteinfacilitating fatty acids capture and transport them through the plasma membrane [8,180]. The acyl-CoA synthetase, activating fatty acids to acyl-CoAs, is another gene-target of PPARα [96,98]. Regarding the genes encoding peroxisomal β-oxidation enzymes, the induction of the peroxisomal fatty acyl-CoA ABC transporter D2 (ALDRP) by PPs was shown to be partially PPARα-dependent in mice hepatocytes [179]; however, the regulation of, e.g., ACOX1, L-PBE and ThB, are entirely reliant on PPARα [8,98,181]. Nevertheless, the regulation of genes implicated in the mitochondrial fatty acid βoxidation, including the carnitine palmitoyltransferase-1 and the medium chain-acyl-CoA dehydrogenase, is also coordinated by PPARα [98,182,183]. Thus, PPARα arises as a master regulator controlling the hepatic metabolism of free fatty acids. The development of PPARα null mice evidenced the crucial role played by PPARα in the concerted regulation of peroxisome proliferation and expression of its target genes involved in both βand ω-fatty acid oxidations [181]. By contrast to Ppara -/mice, which exhibit mild hepatic steatosis, Acox1 null mice develop strong hepatic steatosis, showing a hepatic peroxisome proliferation and the sustained activation of PPARα and expression of its target genes [147,184]. Thus, paradoxically, the defect in ACOX1 activity leads to the hepatic accumulation of ACOX1 substrates, of which some have been shown [147] as efficient endogenous PPARα ligands, mediating the sustained activation of PPARα. On the other hand, the strong PPARα activation of fatty acid β-oxidation genes increases hepatic dicarboxylic acid production and accumulation. Thus, in the absence of ACOX1 activity, these dicarboxylic acids are still unmetabolized and act as firm inhibitors of mitochondrial fatty acid β-oxidation [185]. Moreover, the Ppara -/-, Acox1 -/double-knockout mice exhibit a few periportal clusters of steatotic hepatocytes, and (re-)expression of human ACOX1 in mice liver results in a substantial reduction in both PPARα activation and hepatic steatosis [8,180]. Peroxisomal fatty acid β-oxidation is induced by starvation in a PPARα-dependent manner, as validated by its impairment in PPARα null mice [8,180]. Accordingly, the deacetylase sirtuin-1 is dispensable to PPARαinducing peroxisomal fatty acid β-oxidation and needs SIRT1-PPARα interaction, and the deletion of hepatic SIRT1 negatively impacts PPARα signaling [165]. The MAP kinase kinase kinase TGFβ-activated kinase 1 (TAK1) acts upstream to PPARα, and its deletion also impaired the PPARα-dependent induction of peroxisomal fatty acid β-oxidation [186]. PPARα signaling has also been shown to involve the AMPK-SIRT1-PGC-1α axis via the adiponectin receptors [187] (Figure 8). These results strongly highlight the detrimental role of the peroxisomal β-oxidation pathway in the sensing of PPARα activity.
Several peroxisomal β-oxidation substrates display a substantial role as PPARα modulators. It is believed that the activities of (inducible and non-inducible) peroxisomal fatty acid β-oxidation systems are modulated by PPARα [108]. Moreover, several findings provide significant evidence that VLCFA and BCFA, which are considered potentially toxic fatty acids, are potent inducers of PPARα that enhance the transcription of peroxisomal enzymes mediating fatty acid β-oxidation [57,188].
Interestingly, Oleoylethanolamide, a naturally occurring lipid regulating satiety and body weight, exhibited a high-affinity binding to PPARα and the activation of its lipidmetabolizing target genes [189]. Nonetheless, we should consider that most fatty acids are subject to elongation, desaturation, esterification, and β-oxidation, which could modify the availability of PPARα ligands. Accordingly, very-long-chain saturated and unsaturated fatty acids are exclusively metabolized by peroxisomal β-oxidation, which participates in their degradation, synthesis, or retro conversion. One defect in this pathway is associated with the accumulation of VLCFAs and a deficit in certain PUFAs' synthesis, such as DHA. Interestingly, a mouse deficiency of ACOX1, the rate-limiting enzyme in the peroxisomal β-oxidation, leads to the sustained activation of hepatic PPARα and the induction of its target genes [190]. The role of ACOX1 in PPARα lipid sensing was highlighted by Acox1 -/-; ob/ob double knockout mice. Thus, the sustained activation of PPARα when linked to the absence of ACOX1 activity attenuates the metabolic consequences of leptin deficiency, due to the ob/ob genotype, showing less obesity with the recovery of glucose homeostasis and alleviating insulin resistance [131,147]. Collectively, accumulated data underline the key role of peroxisomal β-oxidation in sensing PPARα-dependent lipid and energy metabolism.

PPARα Tissue Distribution
PPARα tissue expression is also ubiquitous, although on a different level. PPARα is mainly expressed in tissues with high rates of fatty acid catabolism, i.e., those involved in digestive function (liver, stomach, enterocytes) and muscular activity (heart, skeletal muscle, kidney at proximal tubules). In the nervous system, the expression is moderated (low in retinal, or lacking expression in the central nervous system). Low expression is found in the pancreas and adipose tissue [196], while in the brain, PPARα is found at the highest levels in neurons, followed by astrocytes, and is weakly expressed in microglia [62,197]-more likely, to upregulate the expression of several synaptic related genes coding proteins engaged in excitatory neurotransmission and the neuroprotective mechanism [198][199][200]. In the immune system, PPARα expression is detected in the spleen, monocytes/macrophages, and neutrophils [201]. In addition, expression is seen in reproductive organs and the epidermis. PPARα is also associated with tumorigenesis in colorectal carcinoma [202]. Concerning the expression in developmental tissue in rats, Ppara transcripts are detectable in mouse embryo at 13.5 gestation days, to reach the maximum level at birth [203].

Lessons from Pparα Knockout
This part of the manuscript provides recent findings from the last five years related to Ppara knockout animals, with the intent of disentangling the PPARα's various biological functions in health and disease and to evaluate its engagement in fatty acid catabolism and clearance in liver and heart tissues, where PPARα and FAO are both abundant. A growing body of evidence indicates that PPARα is a crucial regulator of systemic lipid metabolism. PPARα deficiency is considered to be a prime factor that either causes or exacerbates fatty acid metabolism impairment, which leads inevitably to the development of numerous metabolic diseases, to name but a few-obesity [204,205], type 2 diabetes mellitus, insulin resistance, dyslipidemia, myocardial infarction, hepatic steatosis without ethanol consumption, termed non-alcoholic fatty liver disease (NAFLD), which includes severe phenotypes such as non-alcoholic steatohepatitis (NASH), liver fibrosis, and hepatocellular carcinoma [206][207][208][209][210]. Therefore, many investigations were conducted using mainly PPARα knockout mouse models, because of the relative equivalent expression of Ppara mRNA between mice and humans in different tissues [98]. Knockout animal models are generated either with the global (Ppara -/-) or hepatocyte-specific abrogation of the Ppara gene, such as Ppara Hep-/-(reviewed by Wang et al. [181]). The goal was to identify the pathophysiological mechanisms underlying the abnormal phenotypes associated with PPARα dysfunction and to assess the distinct contribution of hepatic and extrahepatic PPARα to global energy and immune system homeostasis in vivo.

Lessons from Pparα-KO in the Liver
Hepatic PPARα activation occurs during suckling [211], with a high-fat diet, and during fasting [212][213][214], boosting fatty acid oxidation (FAO), which participates in the restoration of energy homeostasis and provides energy supply for the extrahepatic tissues. For that reason, most of the studies were focused on hepatic PPARα. Furthermore, hepatic PPARα can protect the liver against fasting/high-fat diet-induced steatosis, by transactivating the genes required for fatty acid catabolism and repressing several inflammatory genes. Thus, during the fasting process, metabolic substrates stored in white adipose tissue are released into the circulation and captured by the liver. Subsequently, this increases β-oxidation and ketogenesis to maintain the energy balance [212]. It was observed that fasted Ppara -/and Ppara Hep-/mice developed hypoketonemia, hypoglycemia, and hypothermia with decreased serum triglycerides. Additionally, the ectopic accumulation of medium-chain fatty acids and long-chain fatty acids in the liver manifests as an increase of hepatic fat mass, termed steatosis, with pronounced oxidative stress and lipid peroxidation compared to wild-type mouse liver. These effects result from the altered mitochondrial and peroxisomal fatty acid β-oxidation pathways in the liver [212,214,215]. Furthermore, mice in which Ppara was deleted uniquely in hepatocytes could not modulate bone marrow monocyte egress upon fasting [216], suggesting that PPARα contributes to the regulation of monocyte homeostasis during fasting.
Regarding high-fat diet (HFD)-induced obesity, mice with the hepatocyte-specific deletion of Ppara develop steatosis and inflammation [217]. These observations corroborate previous results communicated by Stec and al. [205], showing that Ppara Hep-/mice on HFD had worsened hepatic inflammation associated with steatosis, and exhibited high levels of LDL, which is considered an emerging risk factor for cardiovascular disease in NAFLD. PPARα could also protect against obesity. In ob/ob obese mice, the absence of PPARα resulted in increased obesity and led to severe hepatic steatosis [184]. Interestingly, mice lacking only hepatocyte-PPARα developed steatosis spontaneously but without obesity in aging [212,214]. Indeed, extrahepatic PPARα activity blunts and compensates when hepatic PPARα is disrupted, by elevating FAO and lipase activity in other tissues to increase and utilize excess lipid, thus maintaining lipid homeostasis [215]. Likewise, the transcriptome, lipidome, and metabolome results communicated by Régnier et al. and Batatinha et al. [217,218] demonstrate the significant contribution of extrahepatic PPARα activity to the metabolic homeostasis response to HFD consumption. By using double-knockout mice, Ppara -/-/Cyp2a5 -/-, Chen et al. [108,206] together indicate that PPARα interacts with CYP2a5 (cytochrome P450 2A5) an antioxidant enzyme to protect against steatosis. Fibroblast growth factor 21 (FGF21) acts as a downstream molecule of the PPARα signaling pathway to regulate the liver lipid metabolism and contribute to the CYP2a5 protective effects on alcoholic fatty liver disease [206]. In an experiment conducted by Brocker et al. [219], it was observed that treatment with WY-14643, a PPARα agonist, caused weight loss and severe hepatomegaly in WT and Ppara ∆Mac mice but not in Ppara Hep-/and Ppara -/mice, suggesting that cell proliferation is mediated exclusively by PPARα activation in hepatocytes in response to WY-14643 agonist treatment.
Pparab is one of the two Ppara paralogs, highly expressed in zebrafish tissues with high oxidative activity. Li and coworkers generated Pparab-knockout in the zebrafish model [194]. Pparab-null zebrafish demonstrated a lower expression of critical enzymes involved in FAO, and lower mitochondrial and peroxisomal FAO in the liver and muscle, associated with lipid accumulation in the liver. Furthermore, PPARab deficiency increases glucose oxidation, protein synthesis, and reduced amino acid breakdown, while in rodents, the loss of PPARα increases amino acid breakdown [194].

Lessons from Ppara-KO in the Heart
PPARα is a crucial regulator of substrate utilization in the heart. Fatty acids are a primary energy source for the heart, and fatty acid β-oxidation provides almost 70% of cardiac ATP; the remainder is obtained primarily from glycolysis and lactate oxidation [220]. Thus, Ppara KO mice, in response to chronic pressure overload, exhibit enhanced cardiac dysfunction. In contrast, mild PPARα activation in mice showed a positive effect on myocardial energetic functions, especially during progressive and pressure-overloaded heart failure, revealing the virtue of PPARα-associated FAO modulation as a promising therapeutic strategy for heart failure [221]. In addition, Ppara ablation exacerbated myocardial ischemiareperfusion injury in Ppara KO mice models subjected to cardiac ischemia-reperfusion, and interestingly, after the treatment with PEA microparticles (PEA-um ®® , 10 mg/Kg), an endogenous PPARα ligand, only Ppara WT mice showed the cardioprotective effect of PEA-um ®® , but not in Ppara KO mice. Although PEA-um ®® had a protective and beneficial effect on inflammatory disorders associated with ischemic myocardial failure, it also negatively regulates inflammation through PPARα activation by reducing the activation of the nuclear factor-kB (NF-kB) pathway and production of pro-inflammatory cytokines [222]. Thus, PPARα could augment heart function and cardiac fatty acid oxidation, whereas in the Ppara KO mouse model, a more severe sepsis phenotype is observed due to deteriorated cardiac performance and fatty acid oxidation, associated with both a hyperinflammatory cytokine storm as well as immune paralysis [223]. Furthermore, during sepsis, WT hearts showed a decrease in PPARα and other FAO genes' mRNA expression, and this reduction was more dramatic in Ppara-null mouse hearts [223]. Taken together, PPARα expression increased fatty acid oxidation and subsequently supported the hyperdynamic cardiac response early during sepsis or pressure-overloaded heart failure, which may prevent morbidity and mortality.

PPARα and Micronutrients
As reported above, PPARα is activated by different ligands of both natural and synthetic origins, involved in several signaling and metabolic pathways. Some natural ligands are issued from the lipid metabolism, such as PUFAs and their derivatives. Interestingly, micronutrients, such as minerals, vitamins, phytophenols, and phytosterols are non-energetic compounds with essential signaling activity. Of particular interest, polyphenols, oil prod-ucts, and some terpenoids and alkaloids impact cell functions through the modulation of PPARα activity.

Resveratrol
Resveratrol, or 3,4 ,5-trihydroxystilbene, is a natural polyphenol present in large amounts in Japanese knotweed (Polygonum cuspidatum) root. This phytoalexin is produced by a wide variety of plants, some of which are edible for humans, such as grapes, blackberries, blackcurrants, blueberries, and cranberries, to name but a few [224]. However, in the last two decades, the effect of resveratrol on animal models related to several disorders, such as autism spectrum disorder, mitochondrial myopathies, type 2 diabetic nephropathy, or renal lipotoxicity has been increasingly reported.
The effect of resveratrol in the presence of quercetin has been studied on PPARαmediating uncoupling protein regulation in visceral white adipose tissue from metabolic syndrome rats. Resveratrol treatment leads to a significantly increased expression of both Ppara mRNA and protein levels [225]. Remarkably, resveratrol prevents renal lipotoxicity in a high-fat diet-treated mouse model by regulating the PPARα pathway, enhancing the expression of lipolytic genes, and raising the renal PPARα protein level and AMPK phosphorylation level [226]. Due to known dyslipidemia in autism spectrum disorders, PPARs have been proposed as therapeutic targets of resveratrol. Furthermore, in autism, impaired mitochondrial fatty acid oxidation suggests the potential implications for regulating mitochondrial oxidation flux by PPAR activators, especially resveratrol [227].
Numerous natural ligands, including polyphenolic compounds, control the expression of PPAR receptors [228]. They have several health-promoting properties, including antioxidant, anti-inflammatory, and antineoplastic activities. Resveratrol is an active biological modulator of several signaling proteins, including PPARα. Resveratrol activates the AMPK-SIRT1-PGC-1α axis and PPARα via the adiponectin receptors in the renal cortex [187]. Adiponectin has multiple functions, including insulin sensitization and lipid metabolism regulation. Similarly, in mitochondrial myopathy, resveratrol has been shown to potentially target many mitochondrial metabolic pathways comprising fatty acid β-oxidation and oxidative phosphorylation, leading to the up-regulation of the energy supply via AMPkinase-SIRT1-PGC-1α signaling pathways [229].

EGCG (Epigallocatechin-3-Gallate)
Epigallocatechin-3-gallate (EGCG) is catechin conjugated with gallic acid. It belongs to the flavonol class and is found abundantly in green tea [232] and cocoa, which have the highest content of catechins, followed by prune juice, broad bean pods, and argan oil [224].
EGCG and green tea polyphenol extract display crosstalk with PPARα. Reported studies in cancer cells revealed that EGCG induced the expression level of PPARα protein in a dose-dependent manner. Clofibrate, a PPARα agonist, blocks heme oxygenase-1 (HO-1) induction and sensitizes cancer cells to EGCG-promoted cell death. Moreover, PPARα interacts with the PPRE of the HO-1 promoter. The activation of PPARα sensitizes cancer cells to epigallocatechin-3-gallate (EGCG) treatment by suppressing HO-1 expression [233].
In rats, green tea polyphenols reduce the renal oxidative stress induced by a high-fat diet through deacetylation of SIRT3 mediated by PPARα upregulation [234].

Anthocyanins
Among berries, blueberries contain higher amounts of anthocyanins. These polyphenols are known to exhibit hypolipidemic properties. Rimando et al. reported that both anthocyanins and catechins do not activate PPARα, while pterostilbene revealed the dosedependent activation of PPARα in H4IIEC3 hepatocytes [236]. In addition, pterostilbene showed a significant increase in Ppara gene expression, but at a lower extent than fenofibrate [236]. Although pterostilbene and resveratrol, as PPARα activators, are under the threshold for effective concentrations in blueberry extract, hepatic mRNA Ppara expression has increased in hamsters fed on a diet containing blueberry extract [236].

Coffee
Coffee consumption has been shown to upregulate mouse hepatic PPARα expression and its target-gene Acox1, consequently leading to the induction of liver peroxisomal fatty acid β-oxidation. Such FAO induction, with induced intestinal cholesterol efflux and reduced lipid digestion, prevents the high-fat diet-induced fatty liver through the lipid-sensing modulation of the gut-liver axis [237].

Edible Oil Products
The effect of polyphenols has been investigated in a rat model of bowel disease by 3 months diet supplementation with extra-virgin olive oil with a high or low phenolic content [238]. The presence of polyphenols in olive oil significantly attenuates the intestinal inflammation associated with hypocholesterolemia and the induction of PPAR-α gene expression in the liver [238]. In a model of insulin resistance of rats fed a high-fat diet, the administration of the major metabolite of oleuropein, hydroxytyrosol, increases the hepatic mRNA levels of Ppara and its target genes, i.e., fibroblast growth factor 21 and carnitine palmitoyltransferase 1a [239]. Similarly, mice receiving a high-fat diet develop hepatic steatosis and inflammation, which were attenuated by hydroxytyrosol supplementation through PPARα activation, Nrf2 (nuclear factor, erythroid 2 like 2) mediated-antioxidative pathway, and by the downregulation of NF-κB-associated inflammation [240]. Used as food supplementation, argan oil or olive oil was shown to restore the expression of genes involved in liver mitochondrial and peroxisomal fatty acid β-oxidation and gluconeogenesis in the mice sepsis model when injected with lipopolysaccharides. This preventive effect of argan oil likely involves the hepatic upregulation of PPARα, PGC-1α, and the estrogen-related receptor α [241].

Conclusions and Future Directions
In all these tested situations, irrespective of the tissue, animal, or pathological condition, micronutrients appear to have an advantageous effect on Ppara expression and activity. Furthermore, almost all these compounds are potent antioxidants and can activate signaling pathways via PGC1-α and AMP kinase. Numerous natural products might modulate PPARα, including terpenes, polyketides, phenylpropanoids, polyphenols, and alkaloids; for instance, the linalool effect is ten times less compared to fenofibrate [88], demonstrating the potential beneficial effects of dietary micro-components to modulate PPARα functions desirably in a population with an ever-increasing high-fat diet consumption. The question is the dietary relevance of these effects, since most of the data were obtained from in vitro studies, and secondly, these micronutrients are often present at very low doses in the diet, except for some polyphenols.
Despite tremendous signs of progress on the critical role of PPARα-dependent regulation in lipid metabolism, the characterization of peroxisomal enzymes and transporters, there are still gaps that need to be filled to fully define the exact role and regulation of PPARα and peroxisomal fatty acid β-oxidation in the cellular homeostasis of lipid metabolism. Particular attention needs to be focused on: 1.
The shuttling of substrates and cofactors from and into peroxisome.

2.
What is the exact role of peroxisomal β-oxidation in lipid metabolism and cell signaling? 3.
How can peroxisome be a mediator and responder of metabolic and environmental stresses? 4.
What are the molecular events that are required at the metabolic level?
(a) Does heterodimerization of PPAR/RXR control the regulation? Is it controlled by coregulators? (b) What is the nature of ligands? (c) What is the nature of micronutrients? Are they natural agonists or antagonists or their balance? (d) Is PPARα the only nuclear receptor governing peroxisomal β-oxidationrelated genes? (e) How do coregulators play in concert to fine-tune metabolically peroxisomal β-oxidation pathway?
All these as yet unanswered questions deserve our complete focus in the near future. There is an increasing demand from health institutions and pharmaceutical industries for efficient drugs. PPARα binding pocket-ligand interactions are being increasingly recognized as a source for therapeutic interventions. Bio structural analysis based on X-ray crystallography and ligand structure pharmacophore modeling approaches afford new biophysical and structural parameters that are important in designing and developing novel potent and highly PPARα-specific ligands to preserve human health and safety. However, the overall goal of increasing the peroxisomal fatty acid oxidation and β-oxidation safely, without increasing the lipid peroxidation and free radical-based risk of non-genotoxic carcinogenesis in the high-fat Western diet-fed population, is a challenge that is still unmet and requires continuous exploration of avenues to activate PPARα dependent pathways safely.