By Thomas G. Brock, Ph.D.

When peroxisomes were first studied, in the 1960's, as a subcellular organelle similar in structure to lysosomes, they were found to consume oxygen to initiate the metabolism of long chain fatty acids through β-oxidation. Additionally, peroxisomes use molecular oxygen to enzymatically produce hydrogen peroxide, which, with peroxidase, is used to oxidize a variety of substrates, including alcohols and toxic compounds. Excess hydrogen peroxide is decomposed by peroxisomal catalase. While peroxisomes may be found in many cells and tissues, they are particularly important in the liver and kidney.

Peroxisomes can multiply prior to cell division, doubling in number so that mother and daughter cells have a full complement.¹ More relevant to this article, peroxisomes can form de novo while a cell is in interphase. Specific peroxisomal membrane proteins are synthesized in the endoplasmic reticulum (ER), with pre-peroxisomes budding off from the ER to form immature organelles. These may fuse with each other or with mature peroxisomes, while enlarged mature peroxisomes may undergo fission to generate smaller peroxisomes. Lumenal (matrix) proteins and additional membrane proteins are imported from the cytoplasm directly into peroxisomes.¹

As peroxisomal proliferation can coincide with mitosis and excess cell division is a hallmark of cancer, there was early interest that compounds that promoted peroxisomes to multiply might be associated with carcinogenesis. Moreover, it was known that clofibrate, a compound with lipid-lowering properties, causes enlargement of the liver (hepatomegaly) in rats that is associated with a profound increase in the number of peroxisomes in liver cells. Another lipid-lowering drug that causes peroxisomal proliferation, nafenopin, was found to cause hepatocellular carcinomas in mice with acatalasemia (a genetic disorder leading to a deficiency of catalase in erythrocytes) but not in wild type mice.² In the 1980's, concern moved from therapeutics to environmental factors when various phthalate and adipate esters, used as industrial plasticizers, were discovered to be peroxisome proliferator carcinogens in mice and rats.3 None of these compounds were found to directly cause DNA damage, the prevalent modus operandi for known carcinogens at the time. By and large, it was thought that an overabundance of peroxisomes could lead to the generation of oxygen radicals, which then would produce the DNA damage necessary for carcinogenesis. Today, with a deeper understanding of how these compounds work, it is understood that normal proliferation of peroxisomes, like normal cell proliferation, does not cause cancer. Instead, it is only with dysfunctional signaling that pathologies occur. In fact, enhanced peroxisomal proliferator signaling is, for some diseases, therapeutic.

The First Peroxisome Proliferator-Activated Receptor

By the end of the 1980's, it was known that clofibrate, phthalate esters, and other compounds that caused dramatic proliferation of hepatic peroxisomes as well as liver hyperplasia also increased transcription of genes required for the peroxisomal β-oxidation of long chain fatty acids and genes of the cytochrome p450 IV family. Narendra Lalwani had discovered a clofibrate- and nafenopin-binding protein in rat liver, suggesting that peroxisome proliferators might act via soluble receptors, like steroid hormones.⁴ In 1990, Isabelle Issemann and Stephen Green reported the cloning of a novel member of the steroid hormone receptor superfamily that was activated by peroxisome proliferators.5 This protein, named a peroxisome proliferator activated receptor (PPAR), had a DNA-binding domain (DBD) with high sequence identity (46-64%) with known nuclear receptors (Figure 1). A distinct putative ligand-binding domain (LBD) shared highest identity (38%) with the human erb-A-related (ear) receptor hear1. Highest expression of the PPAR message was found in liver, kidney, and heart, with very weak expression in brain and testis. Expression was also reported to occur in brown adipose tissue. Chimeric reporters engineered to combine the putative ligand-binding domain of this new receptor with DNA-binding domains of either the estrogen receptor or the glucorticoid receptor, were used to show that hypolipidemic peroxisome proliferators, including nafenopin, clofibrate, and phthalate esters, activated gene expression as expected. Thus, the first PPAR, later renamed PPARα, was discovered.

Figure 1. PPARγ (cyan) associated with RXRα (red) bound to DNA 

Note the DBDs (helices) fill major grooves on opposite sides of the DNA double helix and the LBDs (displayed in space-filling mode) overlap. From RCSB 3DZY¹⁷

PPARα is activated by a diverse array of compounds (refer to Table). Importantly, some agonists are specific for this isoform, whereas some also activate PPARβ/δ and/or PPARγ. While receptor activation causes peroxisome proliferation and hepatomegaly in mice and rats, this does not happen in non-rodent (including human) species. Similarly, PPARα agonists, including the fibrates, do not increase the incidence of liver cancer in humans. In fact, lipid-lowering drugs that act through PPARα, including the fibrates gemfibrozil and fenofibrate, have been used for years and their side effects are well-known (and are unique for each drug). For example, gemfibrozil causes a slight but significant increase in gastrointestinal reactions, while fenofibrate may adversely impact liver function tests. In exchange for these potential complications, PPARα activators potently decrease fatty acid and triglyceride levels. They are commonly used in combination with statins, which lower cholesterol by interfering with the cholesterol biosynthetic pathway. While fibrates are best known to stimulate the β-oxidation of long chain and very long chain fatty acids by peroxisomes for the treatment of hyperlipidemia, they are also used to correct atherogenic dyslipidemia in the context of obesity, diabetes, and coronary heart disease.⁶ Also, clofibrate has recently been shown to prevent nicotine reward and relapse in rats and squirrel monkeys, suggesting that fibrate medications might promote smoking cessation.⁷ Remarkably, although fibrates activate PPARα, direct binding has not been demonstrated.⁶ Moreover, clofibrate activates peroxisomal proliferation in plants like Arabidopsis thaliana, although Arabidopsis lacks a PPARα homolog.⁸

Activity of various PPAR agonists in cell-based transactivation assays.¹⁶

na = not active; Mur = Murine Receptor EC₅₀ (µM) PPAR; Hu = Human Receptor EC₅₀ (µM) PPAR

Nuts and Bolts

As noted earlier, PPARs contain distinct domains for DNA and ligand binding. These domains are separated by a stretch of over 100 aa. The amino terminus, referred to as activation function-1 (AF-1), is thought to have a transactivation function, folding back above the DBD to stabilize heterocomplex formation between the LBD and associating proteins. The large (189 aa) LBD contains a leucine zipper region of some 130 aa that is required for heterodimerization of PPARα with retinoic acid receptor-α (RXRα). RXRα similarly contains a DBD and LBD separated by a hinge region (Figures 1,2). In the classical model of PPARα signaling, PPARα is heterodimerized with RXRα on a PPAR response element (PPRE) consisting of direct repeats of AGGTCA separated by a single intervening nucleotide; this direct repeat PPRE is called DR-1 and is one of several that bind RXRα heterodimers.

Figure 2. The classical model of PPAR-RXR signaling, showing the PPAR-RXR corepressor complex (above) and the PPAR-RXR coactivator complex after ligand binding (below)

In the absence of ligand, the PPARα-RXRα dimer associates with a multiprotein complex that blocks the initiation of transcription, including nuclear receptor corepressors (e.g.,, NCoR1, SMRT), histone deacetylases (HDACs, SIRTs), and G protein pathway suppressor 2. The addition of ligand leads to dissociation of the corepressor complex followed by the recruitment of coactivators, such as PPAR coactivator-1 and the histone acetyltransferases p300 and CREB binding protein. Formation of the PPAR activation complex leads to histone modification (e.g.,, through acetylation), chromatin relaxation, and altered gene expression. PPARα affects gene expression relevant to altered lipid metabolism, lowering triglycerides and raising high-density lipoprotein in dyslipidemia (see related article on page 4). Natural ligands for PPARα include certain fatty acids, including metabolites of arachidonic acid, (e.g.,, leukotriene B4).⁹

Evidence suggests that PPARs can act in other ways. PPARα can directly interact with numerous proteins other than RXRα and this can occur in either the cytoplasm or nucleus. For example, PPARα binds directly to c-Jun and p65,10 proteins which, like PPARα, heterodimerize with other proteins to form functional transcription factors. These interactions prevent each transcription factor from acting. For example, PPARα binding to p65 prevents NF-κB-mediated expression of such genes as iNOS, COX-2, and IL-6, thus diminishing pro-inflammatory signaling. PPARα also forms DNA-binding heterodimers with other nuclear receptors, such as thyroid hormone receptor (TR) and liver X receptor (LXR), to alter gene expression. Notably, RXRα can also partner with nuclear receptors, including TR and vitamin D3 receptors. This competitively prevents signaling through PPARα.

Additional information is available in recent reviews.^11-15

References

1. Hettema, E.H. and Motley, A.M. J. Cell Sci.122, 2331-2336 (2009).

2. Reddy, J.K., Rao, M.S., and Moody, D.E. Cancer Res.36, 1211-1217 (1976).

3. Kluwe, W.M. Environ. Health Perspect.65, 271-278 (1986).

4. Lalwani, N.D., Alvares, K., Reddy, M.K., et al.Proc. Natl. Acad. Sci. USA84, 5242-5246 (1987).

5. Issemann, I. and Green, S. Nature347, 645-650 (1990).

6. Pahan, K. Cell Mol. Life Sci.63(10), 1165-1178 (2006).

7. Panlilio,L.V., Justinova,Z., Mascia,P., et al.Neuropsychopharmacology (2012).

8. León, J. Plant Signal. Behav.3(9), 671-673 (2008).

9. Murakami, K., Ide, T., Suzuki, M., et al.Biochem. Biophys. Res. Commun.260, 609-613 (1999).

10. Delerive, P., De Bosscher, K., Besnard, S., et al.J. Biol. Chem.274(45), 32048-32054 (1999).

11. Ament,Z., Masoodi,M., and Griffin,J.L. Genome Med.4, (2012).

12. Choi, J.-M. and Bothwell, A.L.M. Mol. Cells33, 217-222 (2012).

13. Nagy,L., Szanto,A., Szatmari,I., et al.Physiol. Rev.92, 739-789 (2012).

14. Neher,M.D., Weckbach,S., Huber-Lang,M.S., et al.PPAR Res.2012, 1-13 (2012).

15. Thomas, M.C., Jandeleit-Dahm, K.A., and Tikellis, C. PPAR Res.2012, 1-10 (2012).

16. Willson, T.M., Brown, P.J., Sternbach, D.D. , et al.J. Med. Chem. 43(4), 528-550 (2000).

17. Chandra, V., Huang, P., Hamuro, Y., et al.Nature 456(7220), 350-356 (2008).