Role of the Peroxisome Proliferator Activated Receptors in Hypertension

Abstract

Nuclear receptors represent a large family of ligand-activated transcription factors which sense the physiological environment and make long-term adaptations by mediating changes in gene expression. In this review, we will first discuss the fundamental mechanisms by which nuclear receptors mediate their transcriptional responses. We will focus on the PPAR family of adopted orphan receptors paying special attention to PPARγ, the isoform with the most compelling evidence as an important regulator of arterial blood pressure. We will review genetic data showing that rare mutations in PPARγ cause severe hypertension and clinical trial data which show that PPARγ activators have beneficial effects on blood pressure. We will detail the tissue- and cell-specific molecular mechanisms by which PPARs in the brain, kidney, vasculature and immune system modulates blood pressure and related phenotypes such as endothelial function. Finally, we will discuss the role of placental PPARs in preeclampsia, a life-threatening form of hypertension during pregnancy. We will close with a viewpoint on future research directions and implications for developing novel therapies.

Introduction to Nuclear Receptors and PPARs

The human body is continuously working to sustain homeostasis. For the purpose of this review, homeostasis is defined as the preservation of blood pressure (BP) within a narrow range to ensure effective perfusion of body tissues. Disruption of these homeostatic systems can cause hypertension, and during pregnancy, can cause preeclampsia. The sensors designed to preserve homeostasis come in many flavors including receptors that sense and respond to mechanical or chemical signals, among others. This review will focus on a category of sensors called nuclear receptor transcription factors which detect signals produced by the body in the form of ligands and exert effects through changes in gene expression. They represent a translationally important group of proteins, because like G protein coupled receptors, they are targets of a wide range of pharmacological agents which treat disease. We will focus on peroxisome proliferator activated receptors (PPAR), spotlighting PPARγ as one member of the nuclear receptor superfamily for which there is extensive evidence for a major participation in BP regulation, hypertension, and preeclampsia.

In humans, the nuclear receptor superfamily includes 48 evolutionarily conserved and structurally homologous members which control the expression of a repertoire of target genes. Each carries a DNA binding domain which provides specificity to define its targets, a ligand binding domain (LBD) which provides specificity to the response, and transactivation domains which mediate interactions with other members of the transcriptional machinery. Although there are formally 6 classes of nuclear receptors, it is expedient to divide them into 4 groups: 1) steroid hormone receptors (e.g., glucocorticoid receptor), 2) non-steroid hormone receptors (e.g., thyroid hormone receptor), 3) orphan receptors whose ligands remain unidentified, and 4) adopted orphan receptors, previous orphan receptors which have since been associated with at least one ligand. The PPARs are members of this latter group.

There are 3 PPAR isoforms, α, βδ and γ, which differ in their ligand selectivity, tissue-specific expression, and repertoire of target genes. PPARα (NR1C1) is expressed in liver, brown adipose tissue, heart, and kidney, and uses arachidonic acid metabolites and polyunsaturated fatty acids as natural ligands, and fibrates as synthetic ligands. Among its myriad of effects, PPARα has been implicated to regulate arterial BP and modulate endothelial function through its effects as a regulator of anti-oxidant and anti-inflammatory gene expression.¹ PPARβδ (NR1C2) is expressed in adipose tissue, skin and brain, and is associated with development and oxidative capability in muscle.² PPARβδ is the least recognized PPAR family member evidenced to regulate BP.³ PPARγ (NR1C3) is ubiquitously expressed and is best known for its essential role in adipogenesis.⁴ Fatty acids and eicosanoids can act as PPARγ ligands, but the primary physiologically relevant ligand(s) for this receptor remain unclear.⁵ PPARγ’s physiological and medical significance was highlighted by the recognition that the thiazolidinediones (TZD) drugs which improve insulin sensitivity are high affinity ligands of PPARγ.⁶

To understand how the PPARs function, it is instructive to contrast their mechanism of action with other members of the superfamily (Figure 1A). Steroid hormone receptors, such as glucocorticoid receptor, are localized in the cytoplasm and are coupled to cytoplasmic chaperones. When they bind ligand, they homodimerize, translocate to the nucleus, bind DNA as homodimers, and recruit a co-activator protein complex to transactivate transcription of their target genes. Alternatively, PPARγ resides in the nucleus already bound to its cognate DNA binding sites as obligate heterodimers with retinoid X receptor (RXR). In the absence of ligand, they interact with co-repressors and histone deacetylases to form a co-repressor complex which promotes gene repression. In response to ligand binding, co-repressors are dismissed, and co-activators and histone acetyltransferases are recruited to form a co-activator complex which promotes transcription. Genome wide chromatin immunoprecipitation studies revealed that PPARγ/RXR can bind to over 5000 sites in the genome.⁷ However, transcriptome analysis revealed that the number of genes activated are substantially smaller.⁸ Thus, whereas the repertoire of potential target genes is large, the actual alterations in gene expression are smaller and cell-specific.

Figure 1: Mechanisms Distinguishing Steroid Hormone Receptors from Adopted Orphan Receptors.

Figure 1: A) Steroid hormone receptors are localized in the cytoplasm coupled to cytoplasmic chaperones under baseline conditions. In response to endogenous or exogenous ligand, the chaperone is dismissed, the receptor/ligand dimerizes and translocates into the nucleus. The dimerized nuclear receptor finds its cognate binding sites (Nuclear Receptor Response Element, NRE) in the genome and recruits a series of co-activator (CoA) proteins which facilitates the opening of chromatin and activation of transcription of its repertoire of genes in that cell type. Adopted orphan receptors such as PPARγ are localized in the nucleus complexed to its cognate binding site as a heterodimer with RXR. Co-repressor (CoR) and histone deacetylases maintain a repressive condition on transcription. In response to ligand binding, corepressors are dismissed, transcriptional coactivators are recruited which facilitates transactivation. B-D) Alternative PPARγ’s mechanism of actions includes: a trans-repression mechanism requiring SUMOylation of a lysine reside in the ligand binding domain, ubiquitination of the NF-κB p65 subunit, and shuttling the NF-κB complex off the nucleus. ER: estrogen receptor, GR: glucocorticoid receptor, PPRE: PPAR response element, RNAP: RNA polymerase. (Illustration credit: Ben Smith).

In additional to the classical mechanism of transactivation, PPARγ represses inflammatory gene expression via a transrepression mechanism (Figure 1B).⁹ This mechanism involves SUMOylation of a lysine reside in the LBD which targets SUMOylated-PPARγ to a co-repressor complex on the promoter of inflammatory genes such as inducible nitric oxide synthase. The formation of this complex, which interestingly, occurs independently of PPARγ ligand and PPARγ binding to a PPAR response element (PPRE), prevents ubiquitination and proteasomal activity of the co-repressor complex and prevents the activation of transcription.

PPARγ has also been reported to contain a “really interesting new gene” (RING) domain (Figure 1C).¹⁰ These domains are present in E3 RING ubiquitin ligases which function to regulate the stability of proteins by targeting them for ubiquitination and proteasomal degradation. PPARγ has been reported to function as an E3 RING ligase for the p65 subunit of NK-кB.¹⁰ In our studies in vascular smooth muscle cells, PPARγ limited NK-кB activity, not by functioning as an E3 ligase, but by binding to p65 to shuttle it out of the nucleus (Figure 1D).¹¹ Interestingly, a mutant form of PPARγ which causes human hypertension prevents its binding to p65 and enhances inflammatory gene expression.¹¹

Genetic Evidence for PPARγ as a Regulator of BP

Of the three PPAR isoforms, the strongest genetic evidence implicating a member of the family as a major regulator of blood pressure is for PPARγ. This is largely based on the seminal finding that mutations in PPARγ (V290M and P467L) not only caused diabetes, and insulin resistance, but also severe hypertension.¹² The mutations caused a loss of basal transcriptional activity and acted in a dominant negative fashion, meaning they inhibited the activity of the wildtype PPARγ (Figure 2). A “knock in” mouse model in which the mouse PPARγ gene was mutated to emulate the human P467L mutation (mouse P465L) exhibited modest hypertension, but incompletely phenocopied the dyslipidemia phenotypes of the human mutation.¹³ Two additional rare mutations in PPARγ, R165T and L339X (X = termination codon) were identified in familial partial lipodystrophy (FPLD3) associated with early-onset severe hypertension.¹⁴ Increased activity of the renin-angiotensin system genes, exaggerated responses to Ang-II stimulation, and increased NF-кB activity all of which was dependent on angiotensin-II (Ang-II) type-1 receptor (AT₁R) signaling were identified in cultured cells derived from patients carrying these mutations.

Figure 2: Mechanism of Dominant Negative PPARγ Mutations.

A) As illustrated in Figure 1, PPARγ forms a heterodimer with RXR and binds to a PPAR response element (PPRE) in chromatin in the regulatory region of a PPARγ target gene even in the unliganded state. This is not a static process, because the complex cycles through DNA bound and unbound states. Consequently, the rate of transcription under these circumstances reflects the occupancy of the site by the PPARγ-RXR-corepressor complex and the binding of other stimulatory transcription factors and RNA polymerase to other sequences in the regulatory region of the gene. Ligand binding causes dismissal of the corepressors (CoR) and a recruitment of coactivators (CoA). The coactivator complex facilitates a conformation of chromatin favorable for transcription. B) Replacement of wild-type PPARγ with P467L PPARγ causes increased occupancy (reduced cycling) of the PPRE by the PPARγ-RXR-corepressor complex, which further reduces baseline transcription of actively repressed PPARγ target genes. Figure was previously reported in ⁴⁶.

In additional to rare variants which have profound physiological effects, more frequent variants of PPARγ have been examined for association with hypertension. A meta-analysis examining 21 studies looking at the association between the P12A polymorphism, the most extensively studied polymorphism, and BP suggested the variant could influence hypertension risk in Asians.¹⁵ More recently, PPARγ was identified as a locus which interacts at the level of chromatin and thus is a distal associated gene associated with BP traits in a study of 1 million people.¹⁶

Clinical Evidence for PPARs as Regulators of BP

PPARα agonists are primary used to treat dyslipidemia. Consequently, there are only a few clinical studies and trials of PPARα agonists where blood pressure was an outcome or measure. Fenofibrate, a PPARα agonist, reduced blood pressure in salt-sensitive, but not salt-resistant volunteers in a small study of 37 hypertensive subjects fed high salt.¹⁷ The beneficial effects of fenofibrate on lowering BP were amplified by the AT₁R blocker candesartan in hypertensive patients with hypertriglyceridemia suggesting a potential link between PPARα and the renin-angiotensin system.¹⁸ The Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) study evaluated the effect of fenofibrate in 9,795 type-2 diabetes patients. There was no significant benefit from major coronary events, although there was significant reduction in total cardiovascular events when compared with placebo.¹⁹ Blood pressure was lowered and indices of renal function were improved in the Helsinki sub-study of FIELD where intensive 5-year fenofibrate treatment was studied.²⁰

There are many more clinical studies and trials of PPARγ agonists that included BP as an outcome. Early studies in humans revealed that the TZDs, in addition to improving glycemic control and improving insulin sensitivity, also decreased blood pressure. Troglitazone decreased BP in nondiabetic obese subjects and outpatients with essential hypertension and mild diabetes.^21,22 Similar modest BP lowering effects were observed with rosiglitazone and pioglitazone, findings replicated in many small clinical trials.²³ In the Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes (RECORD) trial, addition of rosiglitazone to metformin or a sulfonylurea reduced ambulatory BP to a greater level than when metformin and a sulfonylurea were combined without TZD.²⁴ Similarly, pioglitazone reduced diastolic BP in 522 patients with type 2 diabetes mellitus (T2DM) with a high risk for stroke in the Primary Prevention of High Risk Type 2 Diabetes in Japan study (PROFIT-J).²⁵ Finally, pioglitazone reduced the composite of all-cause mortality, non-fatal myocardial infarction, and stroke in 5238 patients with T2DM in the Prospective Pioglitazone Clinical Trial In Macrovascular Events (PROactive) trial.²⁶ BP was modestly but significantly lowered in the pioglitazone group.

It is important to point out that this discussion is not designed to suggest that TZDs are effective antihypertensives or have major cardiovascular benefit. In fact, the use of some TZDs, particularly rosiglitazone, has markedly declined in recent years due to adverse effects, some severe, although some of these adverse effects may be drug specific.²⁷ Instead, our goal was to highlight some of the evidence for the physiological and translational relevance of the PPARγ pathway. As illustrated below, there are tissue-specific benefits to PPARγ that merit further attention in the development of therapies that preserve its beneficial effects while minimizing detrimental effects.

Tissue-Specific Mechanisms of PPAR Action that Regulate Arterial BP

Early evidence for the importance of PPARs as regulators of BP was obtained from studies of PPAR agonists in animals. Activation of PPARα with clofibrate or fenofibrate was shown to lower blood pressure in a number of hypertensive models including the spontaneously hypertensive rat (SHR), and in hypertension models induced by DOCA-salt, Ang-II, and endothelin, among others.^28–33 PPARγ activation by TZD administration lowered BP in a variety of rat models including Dahl salt-sensitive (Dahl-S), one-kidney, one-clip Sprague-Dawley rats, obese Zucker rats, diet-induced obese rats, and in hypertensive transgenic mice, hyperlipidemic rabbits, and spontaneously obese, insulin-resistant rhesus monkeys.^34–40 Similar to the other isoforms, activation of PPARβδ reduced blood pressure in a number of animals models of hypertension (reviewed in ³). These observations have led many investigators to examine the tissue-specific mechanisms by which this occurs. Here we review evidence for the beneficial tissue-specific effects of the PPARs (with a focus on PPARγ) in the vasculature, brain, kidney, and immune system on blood pressure.

A. Vasculature

That PPARγ activators consistently lower BP in the face of volume expansion (see Kidney section) suggests that many of the protective effects of PPARγ are likely to be extra-renal. Indeed, there has been a long-standing hypothesis that many of the cardioprotective effects are mediated by PPARγ activity in vascular endothelium and smooth muscle. This has led to studies defining these protective mechanisms with the goal of identifying clinically therapeutic pathways that preserve the beneficial aspects of PPARγ activation while limiting adverse effects.

Early studies revealed that rosiglitazone prevented the development of hypertension and partially protected endothelium-dependent vasodilation in the mesenteric arteries in insulin-resistant Zucker fatty rats.⁴⁰ Similar effects were observed in models of hypertension that occur independent of hyperlipidemia and hyperglycemia. For example, both rosiglitazone and pioglitazone attenuated hypertension induced by Ang-II in male Sprague Dawley rats.⁴¹ Rosiglitazone blunted hypertension and improved vascular remodeling and endothelium-dependent vasodilation in deoxycorticosterone acetate (DOCA)-salt treated rats and had similar effects in hypertensive transgenic mice expressing the human renin and angiotensinogen genes.^33,37 Of course, the beneficial effects on the vasculature could be mediated by decreases in arterial pressure. Thus, more direct early evidence for a role of vascular PPARγ came from studies showing it was expressed in both endothelium and vascular smooth muscle and regulated the expression of vasoactive agents such as endothelin and Ang-II and pro- and antioxidant genes such as copper-zinc superoxide dismutase and subunits of the nicotinamide adenine dinucleotide phosphate oxidase enzyme.^42–45

It is now generally accepted that PPARγ in the endothelium is required to maintain nitric oxide (NO) bioavailability and restrain oxidative stress (reviewed in ⁴⁶). This is based on studies of endothelial-specific PPARγ-deficient mice and mice selectively expressing hypertension-causing dominant negative PPARγ mutation (V290M) in the endothelium (so called E-V290M mice). In endothelial-specific PPARγ-deficient mice, NO production was significantly decreased in aorta while serum reactive oxygen metabolites were significantly elevated.⁴⁷ Concomitantly, these mice exhibited an impaired endothelium-dependent vasodilation response and an augmented pressor response to Ang-II. Endothelial-specific PPARγ-deficient mice were also more sensitive to high-fat diet induced hypertension and exhibited accelerated vascular aging.^48,49 E-V290M mice displayed a modest increase in BP at baseline, an augmented pressor response to Ang-II, and impaired endothelial-dependent vasodilation response to acetylcholine and Ang-(1–7).^50,51 The endothelial dysfunction was reversed by antioxidants. Oxidative stress-mediated endothelial dysfunction was also observed in the aorta of E-V290M mice treated with IL-1β, in the cerebral artery of E-V290M mice challenged with a low-salt diet to induce the endogenous renin-angiotensin system, and in E-V290M mice treated with either a sub-pressor dose of Ang-II or a high-fat diet.^50,52–54 Consistent with this, gene expression profiling revealed altered expression of both pro-oxidant and anti-oxidant genes in response to dominant negative PPARγ.⁸

We identified the retinol binding protein 7 (RBP7, previously known as CRBP-III) gene as an endothelial-specific PPARγ target gene.^8,55 RBP7-deficient mice exhibited impaired vasodilation in cerebral vessels that was normalized by antioxidant treatment suggesting its protective role against oxidative stress.⁵⁶ In fact, RBP7-deficient mice exhibited the same impaired endothelial function phenotype as E-V290M mice suggesting that RBP7 may mediate the anti-oxidant effects of PPARγ. Further studies revealed that another PPARγ target gene, adiponectin, was required to mediate the antioxidant effects of RBP7. Recombinant adiponectin reduced oxidative stress and restored normal endothelial function in RBP7-deficient mice treated with either high-fat diet or Ang-II.

It is notable that RBP7 belongs to the family of fatty acid-binding proteins.⁵⁷ Other members of this family act as co-factors for other nuclear receptors and have been shown to shuttle ligands through the cytoplasm or nucleus to nuclear receptor transcription factors.⁵⁸ Based on its similarity with other members of this family, we hypothesized a mechanism where PPARγ stimulates expression of RBP7, which then acts as a vehicle to deliver a naturally occurring endogenous ligand to PPARγ in the nucleus where it further promotes expression of RBP7 and adiponectin (reviewed in ⁵⁹). We termed this the PPARγ:RBP7 regulatory hub (Figure 3). Indeed, RBP7-deficiency prevented the induction of several PPARγ target genes by rosiglitazone.⁵⁶

Figure 3: Model Illustrating the PPARγ/RBP7 Transcriptional Hub.

PPARγ induces expression of RBP7, an endothelium-specific PPARγ target gene. RBP7 facilitates transcriptional activation of RBP7-dependent PPARγ target genes including adiponectin. Expression of these genes is impaired in the absence of RBP7. Expression of adiponectin mediates an antioxidant response when the mice are challenged with either high-fat diet or Ang-II. Impairment of this protective response under conditions where either RBP7 or PPARγ is impaired causes oxidative stress and endothelial dysfunction in the presence of cardiovascular stressors. Adapted from ⁵⁶.

Mice with vascular smooth muscle specific PPARγ-deficiency exhibited elevated vasoconstriction, vascular remodeling, oxidative stress, and inflammation upon Ang-II infusion, whereas the vasodilation response to acetylcholine was significantly reduced.⁶⁰ Interestingly, smooth muscle PPARγ was reported to protect vessels from Ang-II-induced abdominal aortic aneurysms.⁶¹ It is notable that unlike vascular smooth muscle specific PPARγ-deficiency, the phenotypes of smooth muscle specific expression of a different dominant negative PPARγ mutation (P467L, in S-P467L mice) appeared independent of oxidative stress.

S-P467L mice exhibited elevated BP at baseline, accompanied by arterial stiffness, enhanced vasoconstriction, and severely impaired vasodilation, and develop augmented hypertension when fed a high salt diet.^62,63 They also exhibited increased formation and rupture of cerebral aneurysms when challenged with Ang-II.⁶⁴ Importantly, interference with smooth muscle PPARγ not only impaired endothelium-dependent vasodilation to agonists such as acetylcholine and Ang-(1–7), but also endothelium-independent agonists such as sodium nitroprusside, suggesting a state of NO-resistance.^65,66 Concomitant with a loss of vasodilatory tone, was a gain in vasoconstrictor tone. Vessels from S-P467L mice exhibited marked elevated contraction to endothelin, Ang-II, and serotonin, among others.^62,66 Similarly, myogenic tone in both mesenteric and cerebral vessels was markedly augmented.^67,68 Mechanistic studies revealed the augmented myogenic tone was mediated by a loss of Regulator of G Protein Signaling 5 (RGS5), a PPARγ (and PPARβδ) target gene. RGS5-deficiency exacerbated hypertension and vasoconstriction in mice subjected to Ang-II infusion.⁶⁹ Smooth muscle PPARγ also restrains vascular remodeling by regulating expression of Tissue Inhibitor of Metalloproteinases 4 (TIMP4).⁷⁰ This is consistent with studies reporting that PPARγ agonists prevent cell-cycle progression and telomerase expression.^71,72

RhoA and Rho Kinase are well-known to mediate many vasoconstrictor signals. The ROCK inhibitor Y27632 blocked the hypercontractile phenotype of vessels from S-P467L.⁶⁶ Notably, the level of RhoA protein but not mRNA was increased in the aorta from S-P467L leading us to hypothesize there was a defect in the RhoA turnover. Indeed, Cullin-3 E3 RING ubiquitin ligase mediates the ubiquitination and degradation of RhoA, and the levels of Cullin-3 protein was significantly decreased in aorta from S-P467L.^66,73 Further studies revealed that Cullin-3 plays a critical role as a downstream mediator of the protective effects of PPARγ in vascular smooth muscle.^74–78 Mutations in Cullin-3 cause a Mendelian form of hypertension.⁷⁹

In terms of PPARγ-dependent mechanisms regulating vasodilation, we identified Rho Related BTB Domain Containing 1 (RhoBTB1) as a novel PPARγ target gene.⁶⁶ Genetic complementation (e.g. restoration) of RhoBTB1 in smooth muscle cells of S-P467L mice normalized blood pressure, arterial stiffness, and impaired vasodilation response to both acetylcholine and sodium nitroprusside (Figure 4).⁸⁰ Surprisingly, restoration of RhoBTB1 did not reduce the enhanced vasoconstriction, indicating that its protective effect is mainly mediated by promoting the vasodilation pathway. That the improvement in arterial stiffness occurred in the face of preserved vasoconstriction is a remarkable finding that requires further study. Experiments revealed that RhoBTB1 controlled the level of cGMP in smooth muscle cells via phosphodiesterase 5 (PDE5). RhoBTB1 and PDE5 directly interact and RhoBTB1 functions as a substrate adaptor for PDE5 to the Cullin-3-RING dependent E3 ubiquitin ligase to restrain PDE5 activity. Thus, PPARγ resides upstream of two Cullin-3 pathways which control vasoconstriction through RhoA and vasodilation through PDE5 (Figure 5A).

Figure 4. RhoBTB1 Regulates BP and Arterial Stiffness.

RhoBTB1 is a PPARγ target gene in vascular smooth muscle. Its expression is decreased in S-P467L mice due to the action of the dominant negative mutation in PPARγ. A) S-P467L mice exhibited hypertension, vascular dysfunction and arterial stiffness which was not improved by tamoxifen. B) We generated a transgene which inducibly re-activates RhoBTB1 in response to Cre-recombinase. In the absence of Tamoxifen (Top), RhoBTB1 remains silent and the mice exhibit the same phenotype as S-P467L. In response to Tamoxifen, the transgene is induced and RhoBTB1 expression is restored to normal levels. This resulted in decreased BP, improved vascular function and a regression of arterial stiffness back to normal within 2 weeks of treatment. Data were previously reported in ⁸⁰.

Figure 5. PPARγ-dependent Pathways in the Vasculature.

A) Mutations in either Cullin-3 or PPARγ cause hypertension. Cullin-3 regulates both vasoconstriction and vasodilation pathways: Cullin-3 suppresses the RhoA/ROCK pathway via adaptor protein BACURD; Cullin-3 targets PDE5, a negative regulator of the NO/cGMP pathway, using a PPARγ target gene, RhoBTB1, as the adaptor. RhoBTB1 regulates the activity of PDE5 by ensuring that excess PDE5 is targeted for Cullin-3 dependent ubiquitination and proteasomal degradation. Mutations in PPARγ or treatment with Ang-II causes RhoBTB1-deficiency. Experimental data suggest that arterial stiffness correlates better with the state of vasodilation rather than contraction. Previously reported in ⁸⁰. B) Precise BP regulation requires coordination between vasodilator and vasoconstrictor signals in the endothelium and smooth muscle (SMC). Endothelium-derived nitric oxide (NO) is among the key signals which instruct the SMC to dilate or contract. The NO pathway is coordinately regulated through transcriptional and post-translational pathways initiated by PPARγ. Our data support the concepts that PPARγ: 1) acts as a sensor in the endothelium to regulate redox state, and through this, bioavailability of NO, and 2) regulates the responsiveness of SMC to NO by independently controlling a RhoA/Rho kinase (ROCK) activity that promotes constriction, and the production and stability of cyclic GMP (cGMP), a critical mediator of vasodilation.

Appropriate regulation of arterial BP requires coordination between vasodilator and vasoconstrictor signals in the endothelium and smooth muscle (Figure 5B). Evidence suggests that the NO pathway is coordinately regulated through transcriptional and post-translational pathways initiated by PPARγ. PPARγ acts as a sensor in the endothelium to regulate redox state, and through this, bioavailability of NO. It also regulates the responsiveness of smooth muscle to NO by independently controlling a RhoA/Rho kinase activity that promotes constriction, and production and stability of cyclic GMP (cGMP), a critical mediator of vasodilation. The range of PPARγ-dependent molecular mechanisms in both cell types is surprisingly complex; requiring novel transcriptional co-factors (e.g. RBP7) which form a transcriptional regulatory hub with PPARγ, and post-translational regulation of critical SMC mediators (RhoA and PDE5) by Cullin-3 E3 ubiquitin ligase-mediated protein turnover (Figure 5B). Importantly, this PPARγ initiated “final common pathway” has profound effects on vasomotor function, BP and vascular stiffness, and the studies proposed below have potential implications for the treatment of these disorders.

PPARα and PPARβδ are also expressed in endothelial and vascular smooth muscle cells. PPARα agonists induce expression of eNOS in endothelium which increases NO bioavailability and promotes vasodilation.⁸¹ PPARα can also suppress agonist-induced ET-1 secretion from endothelial cells which could blunt vasoconstriction.⁸²

The PPARβδ agonist GW501516 normalized the impairment in vasodilation caused by palmitic acid by activating the antioxidant gene dihydrofolate reductase.⁸³ Similarly, PPARβδ agonist improves the vasodilation response in high glucose fed mice and db/db mice.⁸⁴ PPARβδ activation abolishes NADPH oxidase-dependent vascular oxidative stress, normalizes endothelial dependent vasodilation, and lowers blood pressure in NZBWF1 lupus mice.⁸⁵ The beneficial anti-oxidant and pro-vasodilatory effects of PPARβδ activation were also observed in spontaneous hypertensive rats.⁸⁶ Moreover, the protective effects of PPARβδ in endothelial cells were reported to be mediated by Nrf2 and endoplasmic reticulum stress.^87,88 Other beneficial effects such as exercise also improve the vasodilation response in a PPARβδ-dependent manner.⁸⁹

B. Brain

The autonomic nervous system plays a fundamental role in the maintenance of cardiovascular function. Its importance in both the short-term and long-term regulation of BP via the baroreflex is accepted.^90,91 Increased activity of the sympathetic nervous system is associated with elevated BP in animal models of hypertension and hypertensive obese patients.^92–94 Moreover, dysfunctional autonomic control is particularly important in the genesis of obesity-induced hypertension and salt-sensitive hypertension.^93,95 In the brain, PPARγ is expressed in both neuronal and non-neuronal cells, is distributed in several anatomical structures regulating BP, and is involved in brain development.^96,97

Studies on brain PPARγ were initially designed to understand the mechanisms causing some of the undesirable effects of TZDs such as weight gain.⁹⁸ Studies in neuron-specific PPARγ knockout mice revealed that the metabolic effects of TZDs, including insulin-sensitizing actions and weight gain, are attributable to their actions in the central nervous system. Many phenotypic characteristics of neuron-specific PPARγ knockout mice were recapitulated in transgenic mice that conditionally expressed the hypertension-causing dominant-negative PPARγ mutant (P467L) in neurons.⁹⁷ A summary of PPARγ effects in the brain is illustrated in Figure 6.

Figure 6: Involvement of PPARγ in Neural Mechanisms of Metabolic and Cardiovascular Control.

Activation of PPARγ by TZD decreases BP and sympathetic outflow in part by attenuating oxidative stress, brain renin-angiotensin system (RAS) activation, and suppressing neuroinflammation in specific cardiovascular centers of the brain. The metabolic effects of TZDs (insulin-sensitization and obesogenic side-effects) were attributed to a direct effect of PPARγ ligands in the pro-opiomelanocortin neurons located in the arcuate nucleus of the hypothalamus. PPARγ activation can blunt the thermogenic and anorexigenic effects of leptin in the central nervous system. However, whether PPARγ can also attenuate leptin-induced sympathoexcitation and pressor responses, a pathophysiological mechanism that has been linked to obesity-associated hypertension, is still unknown.

Leptin acts on receptors located in the arcuate nucleus of the hypothalamus where it influences both metabolic and cardiovascular function.⁹⁹ Intracerebroventricular infusion of PPARγ antagonist, GW9662, or genetic ablation of neuronal PPARγ enhances leptin elicited appetite suppression and body weight loss in diet-induced obesity.^98,100 These data implicate a mechanistic link between PPARγ and leptin and suggest that central PPARγ may play a role in leptin resistance. Leptin resistance may play an important role in some forms of hypertension.¹⁰¹ Activation of brain PPARγ diminishes the firing rate of anorexigenic pro-opiomelanocortin (POMC) neurons while promoting the activity of obesogenic agouti-related peptide (AgRP) neurons.¹⁰² Ablation of PPARγ in POMC neurons augments the anorexic effects of leptin when mice are exposed to a high-fat diet, an effect that was surprisingly not observed in mice conditionally expressing dominant-negative PPARγ (P467L) in POMC neurons.^103,104

In addition to its metabolic effects, leptin can also induce renal sympathetic nerve activation and BP elevation.¹⁰⁵ In healthy individuals, leptin exerts protective effects by suppressing appetite and promoting thermogenesis. In contrast, in obese subjects, despite increased circulating leptin, the beneficial metabolic effects of leptin are impaired, whereas leptin-induced sympathoexcitation and increased BP is preserved.¹⁰⁶ Given the importance of the selective resistance to leptin in the genesis of obesity-induced hypertension, it will be important to assess whether PPARγ in POMC neurons also regulates the cardiovascular actions of leptin.¹⁰¹

Studies in prediabetic rats, SHR, and patients with acute myocardial infarction and T2DM revealed that the activation of PPARγ can mitigate autonomic dysfunction in certain pathological conditions.^107–109 It has been hypothesized that the anti-hypertensive effects TZDs might be attributable to their effects on the central nervous system because TZDs can cross the brain-blood barrier, and models of hypertension exhibit lower levels of PPARγ in the brain.^110,111 Supporting this hypothesis, intracerebroventricular infusion of pioglitazone prevents Ang-II-induced hypertension and attenuates neuroinflammation in the subfornical organ and the paraventricular nucleus.¹¹² Other studies support a mechanistic link between PPARγ and the renin-angiotensin system in the brain. Intracerebroventricular infusion of pioglitazone attenuated sympathoexcitation and a concomitant decrease in hypothalamic expression of AT₁R in rats subjected to coronary artery ligation.¹¹³ Another interaction between AT₁R signaling and PPARγ has been identified in the brainstem. Superoxide production elicited by the activation of the renin-angiotensin system in the rostral ventrolateral medulla, a key brainstem nucleus controlling vasomotor sympathetic nerve activity, induces sympathoexcitation and BP elevation, an effect abrogated by coadministration of rosiglitazone.^114–116 The upregulation of mitochondrial uncoupling protein 2 (UCP2) represents a protective mechanism against oxidative stress in the rostral ventrolateral medulla (RVLM). Ang-II induces UCP2 via p38 MAPK-mediated phosphorylation of PPARγ coactivator 1α (PGC-1α) and promotes the formation of PCG-1α/PPARγ complexes.¹¹⁶ Notably, oral rosiglitazone is sufficient to induce UCP2 in the RVLM, and the anti-hypertensive/antioxidant effects of oral rosiglitazone can be partially abrogated when UCP2 is ablated in the RVLM.¹⁰⁸ This PPARγ-mediated UCP2-dependent antioxidant mechanism is not unique to neurons but is also active in cerebral vasculature.¹¹⁷ Together, these studies support a protective role of PPARγ against central Ang-II signaling, oxidative stress, and sympathoexcitation in experimental hypertension.

Nuclear factor erythroid 2-related factor 2 (Nrf2) is a key regulator of redox hemostasis in the RVLM and the targeted ablation of Nrf2 in the RVLM leads to hypertension.¹¹⁸ PPARγ can induce expression of Nrf2 at the transcriptional level. However, cellular protein levels of Nrf2 are tightly regulated by the Kelch-like-ECH-associated-protein 1 (Keap)/Cullin-3 ubiquitin E3 ligase complex.¹¹⁹ Since mice expressing dominant negative PPARγ exhibited impaired Cullin-3 activity we hypothesize PPARγ might restrain Nrf2 activity in the RVLM by promoting its ubiquitination and proteasomal degradation.⁶⁶ Thus, whether Nrf2 plays a role as an alternative mechanism mediating the antioxidant effects of PPARγ agonists in the brainstem requires additional studies.

Evidence has been reported that some of the beneficial BP effects of PPARα activation may be mediated by improving baroreflex afferent function via PPARα-mediated transactivation of uncoupling protein 2 and reduction of oxidative stress in several regions of the brain regulating BP.¹²⁰ In Ang-II hypertension, GW0742, a PPARβδ agonist decreased BP and restored sympathetic tone, through a mechanism involving decreased Ang-II AT₁R signaling as a result of up-regulation of the PPARβδ target gene RGS5.¹²¹

C. Kidney

There are several beneficial effects elicited by PPARγ activation within the kidney. For example, PPARγ activation improves glomerular endothelial NO bioavailability through increased eNOS expression and activity and Nrf2-dependent antioxidant responses.¹²² In human glomerular endothelial cells, knockdown of PPARγ reduces Nrf2 expression and Nrf2-dependent upregulation of eNOS expression, phosphorylation and activity.¹²² Interestingly, PPARγ itself is a Nrf2 target gene. Activated Nrf2 binds to the antioxidant response element in the PPARγ promotor region to induce its transcription, an effect abolished by Nrf2 knockdown. Thus, PPARγ-Nrf2 reciprocal interaction supports an antioxidant defense required to maintain renal NO bioavailability in response to pro-oxidant stimuli such as excess salt intake. Consistent with this, mice selectively expressing the human hypertension-causing mutation P467L in vascular smooth muscle (S-P467L) exhibit severely impaired vasodilation in renal vessels, renal NO deficiency, and augmented hypertension responses to high salt diet.⁶³ The renal NO deficiency in HSD-fed S-P467L mice is associated with blunted renal blood flow, elevated Na⁺-K⁺−2Cl⁻ cotransporter (NKCC2) protein, and increased sodium retention compared to littermate controls. The impaired natriuresis/diuresis and salt-induced increase in BP are corrected by pharmacological inhibition of NKCC2. These observations suggest that the antioxidant effects of PPARγ in the renal vasculature may contribute to the overall cardiovascular protection mediated by global PPARγ activation.

One of the adverse effects of global PPARγ activation by TZDs is systemic fluid retention, edema, and weight gain. Because of this and other potential adverse effects, TZDs were withdrawn as frontline medications for T2DM, especially in those complicated with congestive heart failure.¹²³ TZD-induced fluid retention could be recapitulated in rodents, and could be reversed by treatment with the collecting duct-targeted diuretic amiloride implicating a role of the epithelial sodium channel (ENaC).^124,125 PPARγ is most abundantly expressed in the collecting duct among nephron segments; and in the collecting duct, Scnn1g which encodes ENaCγ is thought to be a PPARγ-target gene (Figure 7).¹²⁴ Consistent with this, rosiglitazone-induced increases in ENaCγ mRNA expression, sodium reabsorption and plasma volume were abolished by collecting duct-specific ablation of PPARγ in mice.^124,125 These studies provided evidence supporting a role of ENaC in the distal nephron in mediating PPARγ-dependent fluid retention.

Figure 7: Mechanism of PPARγ-mediated Salt/Fluid Reabsorption in Kidney.

In the renal collecting duct, TZDs induce PPARγ-dependent expression of Scnn1g encoding the γ subunit of the epithelial sodium channel (ENaCγ). This promotes sodium reabsorption in the distal nephron, which is inhibitable by ENaC-targeted inhibitors such as Amiloride. This was initially thought to be a primary mechanism for the adverse fluid retention, edema and weight gain associated with TZDs. However, further studies have found conflicting results including the expression of a non-selective cation channel (23ps-C) in response to TZDs which provided an alternative explanation for the sodium and fluid retention after functional ENaC knockdown. The molecular identity of this non-selective cation channel remains unknown. The diagram also illustrates key molecules in the collecting duct, including Na⁺-K⁺-ATPase on the basolateral membrane and the renal outer medullary potassium channel (ROMK) on the apical membrane. (Illustration credit: Ben Smith).

Another study revealed that conditional deletion of Scnn1a which encodes ENaCα failed to prevent rosiglitazone-induced weight gain, fluid retention, and plasma volume expansion despite functional knockdown of ENaC-mediated Li⁺/Na⁺ uptake in the collecting duct.¹²⁶ In the littermate controls, rosiglitazone induced weight gain and fluid retention in the absence of any change in ENaC α, β, or γ protein. Rosiglitazone and pioglitazone failed to alter the density of membrane bound ENaC channels or its Na⁺ conductance in patch clamp studies of primary cortical collecting duct. Instead, a 23-pS nonselective cation channel, which is biophysically distinct from the highly Na⁺-selective ENaC (approximately 6–8 pS), was identified in inner medullary collecting duct, providing a possible explanation for TZD-induced sodium and fluid retention after functional ENaC knockdown (Figure 7). Moreover, increased expression of NKCC2, sodium hydrogen exchanger (NHE3), and aquaporin 2/3 have been reported in the whole kidney of rats treated with rosiglitazone, questioning a primary role of ENaC in TZD-induced sodium and fluid retention.¹²⁷ Further, TZDs have been shown to suppress ENaC mRNA and protein in kidney cortex or collecting duct cell cultures.¹²⁸ Thus, although renal sodium retention and fluid expansion have been well documented as adverse effects associated with TZD use, the underlying molecular mechanisms remain controversial.

PPARα activation by clofibrate lowers BP and sodium retention through renal cytochrome P4504A (CYP4A) / 20-Hydroxyeicosatetraenoic acid (20-HETE) pathway in DOCA-salt hypertension, an effect abolished by PPARα deficiency.^29,129 Similarly, the BP lowering effect of clofibrate requires renal induction of CYP4A/2C and 20-HETE-mediated sodium excretion in salt-induced hypertensive Sprague-Dawley rats.¹³⁰ Clofibrate decreased BP, glomerular filtration rate and cumulative sodium balance which was correlated with increased CYP4A expression and decreased NHE3 in the proximal tubule in high fat diet fed rats.¹³¹ PPARα agonists prevent, but do not reverse salt-sensitive hypertension in Dahl SS rats.¹³²

PPARβδ is ubiquitously expressed in all nephron segments. PPARβδ agonists protect the kidney from acute ischemic/reperfusion injury and systemic lupus erythematosus (SLE) associated albuminuria and chronic renal damage.^85,133 In renal microvessels, PPARβδ mediates prostacyclin-induced vasodilation which is essential for the maintenance of renal blood flow.¹³⁴

D. Immune System

The involvement of the immune system in the development of hypertension was initially proposed in 1967 when it was observed that transferring lymphoid cells from hypertensive rats to normotensive recipients increased blood pressure.¹³⁵ Later it was demonstrated that the lack of lymphocytes in mice caused an attenuated response to Ang-II, but their responses to Ang-II were restored when T cells were adoptively transferred to these mice.¹³⁶ Similarly, depletion of monocytes from the innate arm of the immune system dampens the responses to Ang-II.¹³⁷

PPARγ is expressed in almost every immune cell type and plays multiples roles in the regulation of the immune function (reviewed in ¹³⁸). Generally, PPARγ is considered anti-inflammatory since PPARγ activators have been shown to suppress the recruitment of immune cells in the vasculature, kidney, and heart.^139,140 As discussed above, the anti-inflammatory actions of PPARγ are reported to be mediated by trans-repression, and termination of NF-кB signaling by facilitating either p65 turnover or nuclear export.^9–11 Some natural and synthetic ligands of PPARγ can suppress cytokine release even in macrophages lacking PPARγ indicating that part of their anti-inflammatory actions may be PPARγ-independent.¹⁴¹ A schematic showing the effects of PPARγ in the immune system is shown in Figure 8.

Figure 8: PPARγ Modulation of the Innate and Adaptive Immunity.

PPARγ suppresses the innate immune response by preventing the activation of the master inflammatory transcription factor NF-κB via a trans-repression mechanism resulting in a downregulation of proinflammatory cytokines interleukin (IL)-1β and IL-6, inducible nitric oxide synthase (iNOS), and chemokine receptor (CCR) 2. PPARγ is also involved in the differentiation of macrophages towards the anti-inflammatory M2 macrophages expressing IL-10 and arginase 1. Formation of neoantigens (Ag) such as isoketals are captured, processed and presented by the immunogenic dendritic cells (DC) to T cells resulting in the differentiation of naïve T cells to effector T cells. Among the wide variety of T helper cell subsets interferon (IFN)-γ expressing T_H1 and IL-17 expressing T_H17 are key players in the development of autoimmune disease and hypertension. PPARγ suppresses the adaptive immune response by preventing TH17 and B cells, but also by inducing regulatory T cells (Tregs), presumably via induction of tolerogenic DC. (Illustration credit: Ben Smith).

Studies using mice lacking PPARγ in macrophages and the myeloid lineage confirmed the immunoregulatory role of PPARγ.¹³⁹ Macrophage-specific PPARγ knockout mice exhibited larger vascular lesions, higher macrophage uptake of oxidized LDL and elevated immune cell infiltration in models of atherosclerosis. Myeloid-specific ablation of PPARγ also results in aggravated cardiac injury in myocardial infarction induced by coronary artery ligation.¹⁴² Although PPARγ appears to elicit protection from vascular, cardiac and renal injury in several animal models, studies attempting to elucidate whether PPARγ in immune cells provides protection in models of hypertension are still scarce. PPARγ modulates the innate immune response by suppressing the expression of pro-inflammatory cytokines (interleukin (IL)-1β and IL-6), chemokine receptor 2, inducible nitric oxide synthase among other mediators.^142–145 Notably, the blockade of IL-1β with neutralizing antibodies blunts the inflammatory response in myeloid-PPARγ-deficient mice subjected to myocardial infarction.¹⁴² In addition, PPARγ is involved in the polarization of macrophages to the alternative M2 phenotype, which are considered anti-inflammatory.¹⁴⁶ We observed that endothelium-specific overexpression of wildtype PPARγ or treatment with rosiglitazone ameliorated endothelial dysfunction in response to IL-1β, whereas endothelium-specific overexpression of dominant negative PPARγ (E-V290M) aggravated it.⁵² This indicates that PPARγ might modulate the immune response both by suppressing NF-кB in immune cells and by reducing oxidative stress.

Synthetic PPARγ agonists have shown promising results in several immune diseases such as SLE, psoriasis, arthritis, and autoimmune encephalomyelitis among others.^147–150 TZDs can attenuate the production of autoreactive antibodies and ameliorate vascular and kidney injury in a mouse model resembling human SLE. Some of these protective effects were attributed to adiponectin, a PPARγ target gene, and the induction of tolerogenic dendritic cells.^151,152 In female NZBWF1 mice, a model of SLE hypertension, rosiglitazone reduced albuminuria, renal inflammatory macrophage infiltration and glomerulosclerosis, and also blunted high blood pressure.¹⁵³ Similarly, rosiglitazone ameliorated salt sensitivity of BP in ovariectomized female Dahl salt-sensitive rats and this was associated with reduced renal resident macrophages and inflammation.¹⁵⁴

In addition to its role in innate immunity, PPARγ also protects from aberrant adaptive immune responses. PPARγ suppresses effector T cell activation and stimulates the differentiation and migration of regulatory T cells, which are known to protect from hypertension and vascular injury elicited by Ang-II.^155–157 In addition, activation of PPARγ in T cells suppresses autoimmunity and prevents the differentiation of naïve T cells to the interleukin 17-producing T helper 17 cells by controlling retinoic acid receptor-related orphan receptor γδ.¹⁵⁸ Together, these studies suggest that suppression of both innate and adaptive immune responses may contribute to the overall anti-hypertensive effects of global PPARγ activation. However, whether the selective modulation of PPARγ in different subset of T cells is sufficient to attenuate hypertension and end organ damage remains unclear.

PPARα is the dominant PPAR isoform in T and B lymphocytes.¹⁵⁹ In aged animals, PPARα-deficiency leads to a proinflammatory phenotype in immune cells evidenced by elevated NF-kB activity and inflammatory cytokine production.¹⁶⁰ Treatment of T cells with PPARα agonist induces repression of NF-kB resulting in attenuated secretion of pro-atherogenic cytokines including IFN-γ, TNF-α, and IL-2, and activation of other transcription factors, although some of the actions of fibrates might be PPARα independent.^159,161,162

PPARβδ agonist GW0742 decreases plasma levels of pro-inflammatory cytokines and autoantibodies against double-stranded DNA, alleviates splenomegaly and reduces lupus disease activity in a mouse model of SLE.⁸⁵ These anti-inflammatory actions may contribute to the anti-hypertensive effects of PPARβδ activation.

Hypertension in Women: Evidence for a Role for PPAR in Preeclampsia

Preeclampsia is a pregnancy-specific disorder of hypertension and end-organ dysfunction that poses serious clinical problems for women and their offspring that extend beyond the peripartum period. Hypertensive disorders of pregnancy, including preeclampsia, are one of the leading causes of maternal mortality and complicates up to 8% of all pregnancies.¹⁶³ The effects of preeclampsia appear to be long lasting for both the mother and offspring. Women with a history of preeclampsia have an increased risk of cardiovascular disease, cerebrovascular events and cardiovascular mortality.¹⁶³ Infants born to mothers with preeclampsia also have lasting complications, not only those related to placental dysfunction, which include preterm birth, prematurity, and growth restriction, but also lifelong consequences unrelated to these antepartum and intrapartum events.

A. Placental Development

Abnormal placentation is at the core of preeclampsia. Both PPARγ and PPARβδ deficiency in mice are lethal providing evidence that they play an important regulatory role during development.^4,164 PPARγ-deficient mice exhibit numerous placental defects including malformed labyrinth zone, lack of infiltration of fetal blood vessels, and rupture of maternal blood sinuses, while PPARβδ-deficient placentas exhibited defects in the placental-decidua interface.¹⁶⁴

All three PPAR isoforms are expressed in the amnion, decidua, and villous placenta, but PPARγ remains the most robustly studied isoform in placenta.¹⁶⁵ In early placentation, PPARγ inhibits extravillous cytotrophoblast invasion, promotes trophoblast differentiation and may control expression of human chorionic gonadotrophin.^166–168 Trophoblast differentiation is important as the cell characteristics change. For one, the syncytiotrophoblasts are a multinucleated layer that lines the placental villi. Interestingly, syncytiotrophoblasts are more resistant than cytotrophoblasts to hypoxic injury, so PPARγ-mediated differentiation may make the cells more resistant to hypoxia-induced apoptosis.¹⁶⁹ When human cytotrophoblasts differentiate into syncytiotrophoblasts, they accumulate lipid droplets. PPARγ plays a role in this as PPARγ-deficient placentas lack lipid droplets.^4,168 Studies in a cell line carrying PPAR response element-luciferase reporter demonstrated that sera from pregnant women are capable of activating PPARγ.¹⁷⁰ This suggests that PPARγ activating factors (presumable ligands) are present throughout pregnancy. It was reported that serum from preeclamptic women had reduced levels of PPARα and PPARγ ligands even prior to the onset of clinical features of the disease suggesting that downregulation of PPAR signaling might participate in the pathogenesis of preeclampsia.^171,172 This is consistent with plasma from preeclamptic patients being capable of stimulating an inflammatory response in monocyte culture resulting in decreased levels of PPARγ and increased IL-1a, IL-6, and TNF-α.¹⁷³

B. Genetics

Perhaps one of the most convincing links of PPARγ and preeclampsia comes from the description of human patients with known dominant negative PPARγ mutations. One of the female probands had two pregnancies which were both affected by preeclampsia.¹² On the contrary, there was no association between PPARγ polymorphism and the occurrence of preeclampsia in a Finnish cohort.¹⁷⁴ Several other reports have been published with some conflicting results. There was no difference in PPARγ protein expression between normal and preeclamptic pregnancies and those with growth restricted fetuses.¹⁷⁵ But, in a subsequent study, preeclampsia with co-existing growth restriction of the fetus demonstrated increased PPARγ expression and DNA binding activity.¹⁷⁶ Decreased PPARγ expression, independent of hypoxia-inducible factor and histone deacetylases, was found in hypoxia, a common finding in preeclamptic placenta.¹⁷⁷

C. Animal Models

In a rat model of preeclampsia caused by daily treatment with an inhibitor of nitric oxide synthase, vascular endothelial growth factor (VEGF) gene expression was lower in the preeclampsia cohort, whereas omega 3 fatty acids, potent activators of PPARγ, improved levels of VEGF as well as PPARγ expression in late-onset preeclampsia.¹⁷⁸ Similarly, rats treated with PPARγ inhibitors in the second half of pregnancy exhibited a decrease in fetal weights and impaired vasodilation of uterine arteries in response to agonists.¹⁷⁹ Chronic administration of a PPARγ antagonist to pregnant rats resulted in elevated blood pressure, proteinuria, endothelial dysfunction, reduced pup weight, increased platelet aggregation and reduced plasma VEGF.¹⁸⁰ Defect in endothelial PPARγ in offspring from vasopressin-exposed mice exhibiting a preeclampsia-like pregnancy infers an increased risk of endothelial dysfunction when exposed to a cardiovascular stressor in that offspring’s adulthood.¹⁸¹ As indicated above RGS5 is a PPARγ and PPARβδ target gene. Studies in pregnant mice revealed that PPARβδ and PPARγ agonist treatments blunted the hypertensive response to Ang-II infusion. That the protective effects of these agonists were absent in homozygous RGS5-deficient mice suggests the RGS5 is required to mediate these responses.¹⁸²

Conclusion and Perspectives

Herein, we provided substantial evidence from genetic, clinical and mechanistic studies that PPARs, particularly PPARγ are critical to achieving and maintaining BP homeostasis. Impaired PPARγ function has serious implications for BP control, renal regulation, vascular function, immunity, and importantly, a healthy pregnancy. Despite this, PPARγ activators, particularly the TZD class of drugs, remain on the decline due to adverse effects. The drugs are particularly contraindicated for heart failure patients where edema could worsen the condition and be life threatening. Investigators have used a two-pronged approach to deal with this. First, new non-agonist PPARγ activators continue to be explored as drugs which could modulate PPARγ activity without fully activating it in all cell types. As an analogy, consider a stereo system, as the volume increases, the music eventually becomes noise. TZDs fully activate PPARγ and cause massive changes in gene expression throughout the body. Drugs which could modulate the PPARγ “rheostat” appropriately, might provide beneficial effects without the adverse effects.¹⁸³ By way of example, SR1664 is a non-agonist PPARγ ligand which serves to prevent the phosphorylation of PPARγ (which inactivates PPARγ) and thus preserves its basal activity.^184,185 SR1664 was reported to have potent antidiabetic actions but did not cause fluid retention or inhibit bone formation. It is unknown if it has the same beneficial effect on blood pressure as the TZDs. Moreover, the selective PPARγ modulators such as MRL24, INT-131, amorfrutins, and telmisartan (also an angiotensin receptor blocker), which act as partial PPARγ agonists with lower affinity, might also offer fewer adverse effects.¹⁸⁶ Whether another PPARγ targeted drug will ever be approved for treatment remains unclear.

The second line of research is to identify PPARγ target genes with the goal of identifying new druggable targets. From our research alone, we identified RGS5, TIMP4, RBP7, RhoBTB1 as PPARγ targets, and through RhoBTB1, PDE5, each of which provides protection from hypertension. RhoBTB1 has remarkable effects on arterial stiffness. Future studies will need to be performed to assess if these can be targets of drugs to improve cardiovascular control and to reduce the risk of cardiovascular disease.

PPARα and PPARβδ have also emerged as potential regulators of BP, but the mechanisms governing their action are not as well defined. Whether some of the molecular mechanisms we defined for PPARγ are shared by other members of the family and whether the benefits of activating multiple PPAR isoform is additive would need to be investigated in the future.

Non-standard Abbreviations and Acronyms

AgRP

agouti-related peptide

Ang

angiotensin

AT1R

Ang-II type-1 receptor

BP

blood pressure

Dahl-S

Dahl salt-sensitive

DOCA

deoxycorticosterone acetate

ENaC

epithelial sodium channel

FIELD

Fenofibrate Intervention and Event Lowering in Diabetes

FPLD3

familial partial lipodystrophy 3

Keap

Kelch-like-ECH-associated-protein 1

IFN

interferon

IL

interleukin

LBD

ligand binding domain

NHE3

sodium hydrogen exchanger

NKCC2

Na⁺-K⁺−2Cl⁻ co-transporter

NRE

Nuclear Receptor, Response Element

NO

nitric oxide

Nrf2

Nuclear factor erythroid 2-related factor 2

PDE5

phosphodiesterase 5

PPAR

Peroxisome proliferator activated receptor

PPRE

PPAR response element

POMC

pro-opiomelanocortin

PROactive

Prospective Pioglitazone Clinical Trial In Macrovascular Events

PROFIT-J

Primary prevention of high-risk Type 2 diabetes in Japan

RECORD

Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes

RBP7

retinol binding protein 7

RING

really interesting new gene

RhoBTB1

Rho Related BTB Domain Containing 1

RGS5

Regulator of G Protein Signaling 5

RVLM

rostral ventral lateral medulla

RXR

Retinoid X Receptor

SHR

Spontaneously Hypertensive Rats

SLE

systemic lupus erythematosus

T2DM

Type 2 Diabetes Mellitus

TIMP4

Tissue Inhibitor of Metalloproteinases-4

TNF-α

Tumor necrosis factor-α

TZD

thiazolidinediones

UCP2

uncoupling protein 2

20-HETE

20-Hydroxyeicosatetraenoic acid