Sage Journals: Discover world-class research

Abstract

Through detailed interrogation of the molecular pathways that contribute to the development of pulmonary arterial hypertension (PAH), the separate but related processes of oxidative stress and cellular metabolic dysfunction have emerged as being critical pathogenic mechanisms that are as yet relatively untargeted therapeutically. In this review, we have attempted to summarize some of the important existing studies, to point out areas of overlap between oxidative stress and metabolic dysfunction, and to do so under the unifying heading of redox biology. We discuss the importance of precision in assessing oxidant signaling versus oxidant injury and why this distinction matters. We endeavor to advance the discussion of carbon-substrate metabolism beyond a focus on glucose and its fate in the cell to encompass other carbon substrates and some of the murkiness surrounding our understanding of how they are handled in different cell types. Finally, we try to bring these ideas together at the level of the mitochondrion and to point out some additional points of possible cognitive dissonance that warrant further experimental probing. The body of beautiful science regarding the molecular and cellular details of redox biology in PAH points to a future that includes clinically useful therapies that target these pathways. To fully realize the potential of these future interventions, we hope that some of the issues raised in this review can be addressed proactively.

Keywords

bone morphogenetic protein receptor 2 Warburg 3-nitrotyrosine nitrosylation

Considered in a positive light, the intense investigative efforts into the pathogenesis of pulmonary arterial hypertension (PAH) over the past several decades can be viewed as a major success. We continue to gain more-detailed insights into the molecular mechanisms that cause and sustain disease. New therapeutics are being successfully tested and approved for human use, with more in the developmental pipeline. Emerging data suggest not only that the thoughtful and aggressive use of existing therapies has improved the quality of life and functional status for patients with PAH but also that these therapies have actually had a measurable positive impact on survival.¹

However, most investigators in the field remain driven by a lingering dissatisfaction. Although existing therapies are certainly beneficial in terms of improving functional status and quality of life and likely have a measurable survival benefit, many patients still will progress to the end points of death or need for transplantation. In tissues collected from these individuals, either at autopsy or at explantation of the diseased lungs, it appears that targeted therapies have not changed the course of the pulmonary vascular pathology.² New drugs continue to be developed and approved for clinical use, but most of these target the same signaling pathways (e.g., prostacyclin, endothelin, and nitric oxide) as the first generation of PAH drugs.^3–5 Despite impressive progress, then, we still lack some fundamental insights into the processes that initiate and sustain PAH, and thus it remains an incurable, progressive, and ultimately fatal disease.

At the 2013 Grover Conference, we discussed data from our investigations into the downstream pathogenic mechanisms that begin with loss-of-function mutations in bone morphogenetic protein receptor 2 (BMPR2) and end with what we recognize as PAH. We discussed data regarding oxidant injury and lipid peroxidation, mitochondrial dysfunction, metabolic reprogramming, and how we think these processes contribute to disease. In this review, we further expand on the separate but intimately related topics of oxidative stress and molecular metabolism, attempting to unite them under the broader heading of redox biology. We hope to highlight some “gray areas” that are in need of more in-depth investigation and clarification, as we think that there is real therapeutic benefit to be realized from using redox biology as one source for the next generation of possible treatments for PAH. We focus our discussion as much as possible on World Health Organization (WHO) group I disease, recognizing that many experimental PAH systems straddle the line between group I and other classifications (e.g., group III) and that at least some pathways may be common to more than one form of pulmonary hypertension.

BRIEF REVIEW OF REACTIVE OXYGEN SPECIES IN PAH

To begin, a review of what is meant by “reactive oxygen species” (ROS) as pertains to PAH is in order. In general, this term refers to a group of small molecules formed from reactions that require the participation of at least one oxygen atom (typically from molecular oxygen or from water) and involve a configuration of electrons around the oxygen atom that renders it chemically unstable on its own. The relative reactivity of any individual ROS depends on the specific configuration of electrons around the oxygen atom, the possible alternative/resonance structures within the molecule containing the reactive oxygen atom, and local conditions outside of the molecule containing the ROS (e.g., solvent conditions). The reactivity of any given ROS is what determines how injurious it will be within a cell or tissue and also determines its ability to function in a signaling capacity.

In the pathogenesis of PAH, several ROSs have been particularly well studied. Perhaps the most relevant ROS in PAH is nitric oxide (NO). NO is formed by nitric oxide synthase from the combination of the guanidino-group nitrogen in arginine with molecular oxygen to yield NO and water, converting arginine to citrulline in the process with electrons provided by NADPH and transferred by the heme iron. By virtue of its electronic configuration (with an unpaired electron in an antibonding π orbital), NO is relatively unreactive toward most biomolecules, reacting primarily with other ROSs—most notably, reacting at near the diffusion limit with superoxide to form peroxynitrite, itself an ROS that results in cellular damage by both oxidation and nitration of molecular targets. Superoxide anion, the reaction partner of NO that forms peroxynitrite, is itself an ROS formed from the 1-electron reduction of molecular oxygen. Superoxide can be produced by the specific action of NADPH oxidase as part of the oxidative burst of neutrophils. In this way, inflammation can contribute directly to oxidative stress. More commonly, superoxide is produced by an incomplete reaction cycle of enzymatic systems that typically reduce molecular oxygen completely to water. In PAH, such enzymatic systems include uncoupled nitric oxide synthase, electron-transfer enzyme complexes in mitochondria (see below), cell-free hemoglobin, and any number of other proteins containing redox-active metals. Under normal conditions, superoxide is fairly rapidly converted to hydrogen peroxide (H₂O₂) through the action of superoxide dismutase. H₂O₂ itself is considered by many to be an ROS, although its reactivity is substantially less than that of most other biologically relevant ROSs (see below). These ROSs are summarized in Table 1.

Table 1.

Summary of major reactive oxygen species in pulmonary arterial hypertension

Reactive oxygen species	Common sources	Common reaction targets
Nitric oxide (NO^•)	Nitric oxide synthase	Heme iron, free sulfhydryl groups
Superoxide (O₂^•−)	NADPH oxidase, mitochondrial electron transport chain, uncoupled NOS, heme-containing proteins	Nucleotides (especially guanine), lipids (especially polyunsaturated fatty acids), proteins
Peroxynitrite (ONOO⁻)	Reaction of NO and superoxide	Tyrosine residues in proteins
Hydrogen peroxide (H₂O₂)	Superoxide dismutase	Metal atoms

OXIDATIVE STRESS, OXIDANT SIGNALING, AND OXIDANT INJURY

Mounting evidence for some time has implicated oxidative stress as an important pathogenic mechanism in PAH, both in patients with disease and in virtually every experimental model of PAH. Evidence for increased production of ROSs has been demonstrated in cells, tissues, blood, urine, and even breath condensate with a variety of analytical techniques. The often-unstated underlying assumption is that increased production of ROSs is uniformly detrimental, presumably through direct, deleterious chemical modification of proteins, lipids, and DNA.

The terminology as it is most often used makes two generalizations. First, the term “oxidative stress” assumes that any observed increase in an assay for “ROSs” above the level observed in the control or normal condition is indicative of a disease-associated process. The strict definition of oxidative stress is the production of oxidizing species of all types that, in total, exceed the capacity of enzymatic and chemical antioxidant defenses and that cause a deviation from homeostasis. The second generalization is that all ROSs are more or less functionally equivalent.

The problems with “oxidative stress”

The problems with these generalizations are fairly readily apparent. An increase in ROS production that is measurable but does not exceed the cell's antioxidant capacity and/or does not cause a significant deviation from homeostasis is not creating a true stress. Also, it is clear that not all ROSs are equivalent. For example, H₂O₂ is often thought of as an ROS, as many assays used to quantify ROS measure H₂O₂. However, it is not an especially reactive substance by itself (a dilute, 3% solution is sold by the bottle by most pharmacies in the United States). Often, the Fenton reaction (Fig. 1) is invoked to explain how H₂O₂ causes oxidative stress. In the presence of free transition metals (e.g., iron or copper), H₂O₂ can undergo a single-electron reduction to either a hydroxyl radical or a hydroperoxyl radical, both of which are extremely reactive and will oxidatively modify nearby macromolecules with a speed at or near the diffusion limit. However, the availability of free transition metals in most biological systems is extremely low, and there is some debate as to whether Fenton chemistry happens to any appreciable degree in vivo.¹² The Haber-Weiss cycle (Fig. 1) has been proposed as an alternative mechanism for how H₂O₂ and superoxide anion together can result in hydroxyl radical production. This may have more relevance to in vivo settings, although in the absence of free transition metals to catalyze this series of reactions, the Haber-Weiss cycle is fairly slow and would require relatively high concentrations of H₂O₂ and superoxide to proceed with any efficiency.¹³ Further, a strong case can be made for H₂O₂ as a bona fide signaling molecule, possessing characteristics of regulated synthesis and breakdown, specific molecular targets, localized cellular production, and measurable biologic effects that exist along a spectrum of intensity that coincides with that of a signal.^14–16 Superoxide anion, while more reactive than H₂O₂, is still a relatively stable ROS; the reactivity of the hydroxyl radical is between 5 and 10 orders of magnitude greater. NO, which is itself a free radical (and one could debate whether it is an ROS or a reactive nitrogen species), is so stable as to serve an almost exclusively signaling role and becomes damaging only after reacting with superoxide to form the peroxynitrite anion. Many colorimetric and fluorimetric assays cannot distinguish between the different ROSs, and the functional differences are substantial.^17,18

Figure 1.

Reactive oxygen species (ROSs) and products of redox reactions exist along a spectrum from oxidant injury to oxidant signaling. The ROSs that are most commonly studied under the umbrella of “oxidative stress” can be separated on the basis of how they behave as injurious stimuli versus how they behave as signaling species/second messengers. A partial listing is shown in this figure. For example, the hydroxyl radical covalently reacts essentially at the diffusion limit with any nearby macromolecule (lipid, protein, or nucleic acid), kinetics that are too rapid for regulated signaling and that functionally can only drive oxidant injury. By contrast, at the other end of the spectrum is nitric oxide, a free radical that is quite chemically stable and has a very well established role as a bona fide second messenger. Similarly, the products of free-radical reactions with macromolecules can be ordered along a similar spectrum. Products such as F₂-isoprostanes and protein carbonyls have predominantly been shown to be markers of oxidant injury (although F₂-isoprostanes can signal, there does not appear to be tightly regulated production and degradation that would qualify them as an oxidant signal). However, production of 8-oxoguanosine has been shown to exhibit regulated production to control gene expression. GSSG: glutathione disulfide; HNE: hydroxynonenal.

These are not merely semantic distinctions, as evidence exists in the literature to suggest that more-specific determinations about the oxidants produced in PAH and their role in disease pathogenesis could have a significant impact on novel therapeutic development. As suggested above, the differences between oxidant injury and oxidant signaling are separable, and these processes should thought of as related to but distinct from each other (Fig. 2). Oxidant injury more clearly refers to the end result of oxidative macromolecule modification in an essentially unregulated fashion. Rather than relying on techniques that provide some readout of the presence and production of ROSs, assessing oxidant injury involves quantification of the stable end products of macromolecular damage—lipid peroxidation products, oxidatively modified nucleotides, protein oxidation products, and carbohydrate oxidation products. By contrast, oxidant signaling involves the coordinated, regulated production of ROSs to achieve a specific, measurable biological end point. Oxidant signaling possesses other hallmarks of signal transduction cascades, including signal amplification, signal termination, and a relatively narrow spectrum of downstream responses. Both oxidant signaling and oxidant injury have been demonstrated in PAH.

Figure 2.

Summary of the Fenton reaction and the Haber-Weiss cycle. The Fenton reaction and the Haber-Weiss cycle are often invoked as important sources of damaging reactive oxygen species (particularly the hydroxyl radical) in vivo. There is some debate as to whether either of these processes proceeds with any appreciable efficiency in living cells and tissues.

Methods for assessing “oxidative stress”

Before discussing the particulars of oxidant injury and oxidant signaling, we present a brief review of the methods for measuring ROSs and the results of injurious chemical reactions driven by ROSs. Several methods exist for quantifying ROSs themselves.

Electron paramagnetic resonance (EPR)/electron spin resonance (ESR) spectroscopy can be used to quantify specific radical species, either through spectroscopic examination of the radicals themselves or through the use of spin traps that form more-stable radical species to permit study. EPR/ESR is quantitative and specific and can be used with biological samples, but the technique requires a great deal of technical expertise and specialized equipment. As in the use of spin traps, there are a variety of dyes that become fluorescent upon reaction with particular ROSs. These are notable for being easy and relatively inexpensive to use, but as discussed above, specificity and true quantitation are significant drawbacks to many of these dyes. Aside from quantifying ROSs themselves, many methods are available for quantifying the myriad products of ROS interactions with biological macromolecules. Lipid peroxidation products (e.g., isoprostanes, isofurans, hydroxynonenal, malondialdehyde, and total lipid hydroperoxides) can be quantified by mass spectrometry, the TBARS (thiobarbituric acid reactive substances) assay, enzyme-linked immunosorbent assay (ELISA), high-performance liquid chromatography, and various colorimetric assays. Similarly, nucleotide oxidation products (e.g., 8-oxoguanosine) can be quantified by many of these same methods. Protein oxidation can be quantified by measurement of “protein carbonyls,” a general measure of protein oxidation, or by more specific assays, such as quantification of 3-nitrotyrosine by ELISA, Western blot, mass spectrometry, or other methods. All of these methods vary in their performance characteristics (e.g., limits of detection, interfering substances and tolerance of sample quality, specificity of the species quantified), and while a complete review is well beyond the scope of this work (and the topic has been amply reviewed elsewhere^19–24), the specific technique employed in any given experiment should be tailored to the question being asked.

Oxidant signaling in PAH

Two recent studies from the pulmonary hypertension literature provide fairly clear, illustrative examples of oxidant signaling in PAH. Al-Mehdi et al.²⁵ published an elegant series of studies in 2012 showing that chemical oxidation of nucleotides (specifically, guanosine bases) in the promoter region of the vascular endothelial growth factor (VEGF) gene is required for efficient hypoxia-inducible factor (HIF)-mediated transcription of the gene under hypoxic conditions. The oxidation of the promoter region was a carefully orchestrated event that involved translocation of mitochondria to a perinuclear region, followed by a nuclear-specific increase in ROSs. Prevention of mitochondrial clustering prevented the oxidation of the VEGF promoter, decreased the binding efficiency of HIF-1α, and decreased production of VEGF messenger RNA without changing the amount of HIF-1α in the nucleus. In 2010, Archer and colleagues²⁶ demonstrated that in the fawn-hooded rat model of PAH, superoxide dismutase 2 (SOD2) is epigenetically silenced, leading to a hyperproliferative, apoptosis-resistant phenotype in pulmonary artery smooth muscle cells (PASMCs). Interestingly, in the absence of normal SOD2 expression, PASMCs showed less ROS production (with data support ing H₂O₂ as the relevant ROS). Restoration of SOD2 expression or replacement with a pharmacologic SOD2 mimetic (MnTBAP) increased ROS production, restored normal apoptosis, normalized PASMC proliferation, and ameliorated the PAH phenotype in vivo. The fact that increasing ROS production led to restoration of programmed cell death—a tightly regulated process required for normal cellular homeostasis—strongly supports an argument for oxidant signaling.

Oxidant injury in PAH

Oxidant injury, as defined above, has also been demonstrated to associate with and to contribute to PAH pathogenesis. Cracowski and colleagues^10,27 found that levels of F₂-isoprostanes—a robust biomarker for lipid peroxidation and oxidant injury in vivo—measured in the urine of PAH patients at the time of diagnostic right heart catheterization independently predicted increased mortality in both unadjusted and adjusted risk prediction models. Previously, Robbins et al.²⁸ had published that the major urinary metabolite of F₂-isoprostanes decreased in PAH patients after the initiation of epoprostenol therapy. Oxidized guanosine and 3-nitrotyrosine levels have also been demonstrated to be elevated in experimental models of PAH and in lung tissue from patients.^29–31 We and others have shown that disruption of extracellular superoxide dismutase (EC-SOD), either globally or in specific cell populations, results in exacerbation of experimental PAH, with accompanying increases in markers of oxidant injury, and that enhancing EC-SOD activity ameliorates disease.^32–34 To help solidify a causative link between oxidant injury and PAH, we have shown that driving mitochondrial oxidant injury (quantified by measurement of lipid peroxidation products) using brief daily exposure to hyperoxia worsens PAH in BMPR2-mutant mice and also drives dysregulation of glucose homeostasis, suggesting a link between oxidant injury and cellular metabolism.³⁵

Ambiguity in distinguishing oxidant injury from signaling

It can sometimes be difficult to distinguish between oxidant injury and oxidant signaling, and there is certainly a gray area in the middle. For example, 3-nitrotyrosine is formed from the non-enzymatic nitration of tyrosine residues by peroxynitrite, an aggressive pro-oxidant produced by the reaction of superoxide anion with NO. Thus, 3-nitrotyrosine is appropriately considered a product of oxidant injury. However, nitration of tyrosine residues does not result in widespread, nonspecific protein dysfunction. Indeed, in pulmonary microvascular endothelial cells from BMPR2-mutant mice (Fig. 3, left) and in skin fibroblasts from heritable PAH patients with BMPR2 mutations (Fig. 3, right), a Western blot for 3-nitrotyrosine shows multiple specific bands of higher-intensity staining than is seen in wild-type cells. Similarly, a group of investigators at the University of Illinois in Chicago^36–38 has very nicely demonstrated that knockout of caveolin-1 results in both increased superoxide and increased NO production, causing nitration of specific protein targets (e.g., protein kinase G, p190RhoGAP-A), and leads to the development of experimental PAH. In keeping with these findings, mutations in caveolin-1 in humans have been shown to cause heritable PAH.³⁹ Thus, only a subset of proteins within cells exhibit significant 3-nitrotyrosine formation, and certain tyrosines within those proteins are preferential targets for nitration. Moreover, the presence of 3-nitrotyrosine has specific effects on the function of individual proteins, with some exhibiting significant loss of function (e.g., SOD2, prostacyclin synthase) while others show notable gain of function (e.g., increased peroxidase activity of cytochrome c after tyrosine nitration).⁴⁰ With a narrower range of possible targets and target-specific functional effects that are observed with nitration of a relatively small percentage of the total possible targets, formation of 3-nitrotyrosine has several features of a signaling pathway, much like other posttranslational modifications, such as phosphorylation.

Figure 3.

Pulmonary arterial hypertension–causing bone morphogenetic protein receptor-2 (BMPR2) mutations drive tyrosine nitration of specific targets. In both murine pulmonary microvascular endothelial cells (PMVECs) and skin fibroblasts from patients, expression of any one of several mutant BMPR2 isoforms is sufficient to drive nitration of tyrosines in multiple protein targets, as shown in this Western blot from total protein lysates probed with an anti-nitrotyrosine antibody. Fold increase is indicated by the numbers and calculated as a ratio of the densitometry in the BMPR2 mutant sample to that in control/wild-type cells (average densitometry for the two, in the case of the two different BMPR2 mutations represented by the human fibroblasts). Of note, bands appear as a ladder in both wild-type and mutant cells, indicating that tyrosine nitration is not a stochastic event. Further, only certain bands (arrows) show increased intensity in the BMPR2 mutant cells, suggesting further specificity.

As implied in the above discussion, mitochondria are a key site of convergence for both oxidant injury and oxidant signaling. The mitochondria are one of the most important cellular sources of ROSs, which can be directed into oxidant signaling pathways or which can escape control and mediate oxidant injury. Mitochondria contain high concentrations of redox-active metals (e.g., heme-containing cytochromes, iron-sulfur clusters) and are the primary site for single-electron redox reactions in the cell. Mitochondria contain many of the enzymes for the key pathways involved in maintaining redox balance in the cell by way of synthesizing cellular reducing equivalents, such as NADH. The convergence of energy metabolism, regulation of apoptosis, and redox homeostasis allows mitochondrial function to influence and to be influenced by oxidant fluxes and the products of oxidant signaling and injury.

CARBON METABOLISM AND REDOX BIOLOGY IN PAH: BEYOND GLUCOSE OXIDATION

The past 10–15 years have seen a renaissance of interest in intermediary metabolism, and PAH research has definitely been part of this resurgence. The role of molecular metabolic program as a primary driver of disease in the modern era is commonly traced back to the work of Otto Warburg in the early twentieth century. Warburg^41,42 demonstrated that a shift to lactate-producing glycolysis as a primary means of carbon utilization, even in the presence of ample oxygen to allow complete glucose oxidation, not only was a feature of malignant cells but also was central to malignant transformation. The term “Warburg effect” is often used as shorthand for any cell or tissue that appears to undergo a change in metabolic program to a reliance on lactatogenic glycolysis under normoxic conditions.

Evidence for the Warburg effect in PAH came first from experimental models, such as the monocrotaline-treated rat and the fawn-hooded rat, and was followed by studies supporting increased lactatogenic glycolysis as a primary carbon-use program in cells and tissues from PAH patients.^43–46 Several underlying molecular mechanisms have been proposed, with the weight of data in support of normoxic activation of HIF and increased “oxidative stress” by a variety of mechanisms (e.g., epigenetic silencing of SOD, a direct increase in superoxide flux from mitochondria). This subject has been very thoroughly reviewed elsewhere, with the summary being that a shift from glucose oxidation to lactatogenic glycolysis is a key pathogenic mechanism in multiple models of experimental PAH.^47–53

Glucose oxidation versus fat oxidation: which to inhibit, which to activate?

In normally functioning cells, glucose oxidation exists in a reciprocal relationship with fatty acid oxidation by way of a feedback mechanism termed the glucose–fatty acid cycle, or Randle's cycle, first described by endocrinologist Philip Randle^54,55 in the 1960s. As fatty acid oxidation proceeds, rising concentrations of multiple intermediates provide feedback inhibition that limits carbon flux through the glycolytic pathway.⁵⁶ Specifically, rising acetyl coenzyme A (CoA) levels and NADH levels inhibit pyruvate dehydrogenase, and rising citrate levels inhibit phosphofructokinase and, indirectly, hexokinase. Reciprocal inhibition of fatty acid oxidation is provided by the conversion of glucose-derived citrate to malonyl-CoA through the sequential action of adenosine triphosphate (ATP) citrate lyase (which converts citrate to acetyl-CoA) and acetyl-CoA carboxylase. Malonyl-CoA provides feedback inhibition of carnitine palmitoyltransferase 1, which limits the availability of acyl chains for oxidation and shifts the balance toward fatty acid esterification. As malonyl-CoA accumulates, then, glucose oxidation is required to meet the energetic needs of the cell. These reciprocal relationships are summarized schematically in Figure 4.

Figure 4.

Schematic of Randle's cycle. A reciprocal relationship exists in most cells between glucose oxidation and fatty acid oxidation. As shown here and discussed in the text, acetyl coenzyme A (CoA) and citrate feed back to inhibit glucose oxidation, mainly at pyruvate dehydrogenase (PDH) and phosphofructokinase (PFK) and indirectly at hexokinase and glucose transporter 4. Malonyl-CoA will feed back to inhibit fatty acid oxidation through inhibition of carnitine palmitoyltransferase 1 (CPT1). Note that a number of enzymes (e.g., adenosine triphosphate citrate lyase) have been left out to allow focus on the key nodes of regulation between the two pathways that constitute Randle's cycle.

Much of the work that has been done thus far to leverage molecular metabolism in PAH has been focused on the restoration of glucose oxidation. This has been accomplished in experimental PAH in a variety of ways. Administration of inhibitors of pyruvate dehydrogenase kinase, such as dichloroacetate (DCA, most commonly used to treat inborn errors of metabolism), has long been known to activate pyruvate dehydrogenase and to facilitate entry of glucose-derived carbon into the tricarboxylic acid (TCA) cycle to permit its oxidative metabolism, effectively counteracting the Warburg effect.⁵⁷ DCA has shown efficacy in several models of experimental PAH, ameliorating pathologic changes in the pulmonary vasculature as well as maladaptive right ventricle (RV) remodeling.^{8,45,48,58–62} A small phase I, open-label trial of DCA in PAH has been conducted by the University of Alberta and Imperial College London (DCA for the treatment of PAH; ClinicalTrials.gov identifier: NCT01083524), and the forthcoming results are eagerly anticipated. Strategies to leverage Randle's cycle have also shown efficacy in experimental PAH.^61,63,64 By inhibiting fatty acid oxidation pharmacologically with compounds such as ranolazine or trimetazidine or by increasing malonyl-CoA through genetic deletion of malonyl-CoA decarboxylase (the major enzyme that breaks down malonyl-CoA), glucose oxidation can be facilitated and at least some of the PAH pathology improved in experimental models. Currently, there are 4 registered trials examining ranolazine's safety and initial efficacy in pulmonary hypertension and one registered trial to study trimetazidine in PAH.

While it is likely that facilitating pyruvate entry into the TCA cycle by using DCA should have some beneficial effect, the actual realized benefit of inhibition of fatty acid oxidation is less clear. Indeed, the data are unclear regarding what happens to fatty acid oxidation in the normal progression of the molecular metabolic pathology that characterizes PAH. With inhibition of glucose oxidation, other things being equal, one might expect a compensatory increase in oxidative fat metabolism, based on Randle's cycle. Indeed, Zhao and colleagues⁶⁵ reported an upregulation of the expression of genes involved in fatty acid oxidation as well as increases in metabolic intermediates in lung tissue from PAH patients, compared to controls. In this setting, then, inhibition of oxidative fat metabolism would make sense. However, metabolic remodeling in the heart in PAH is a different story. Normal myocardium generates up to 70% of its ATP needs from fat oxidation.⁶⁶ In several models of PAH, fatty acid oxidation has been shown to be decreased in the myocardium of the RV, compared to that in normal animals.^62,67,68 Supportive evidence for suppression of fat oxidation has also been found in RV myocardium from humans with PAH.⁶⁹ With this in mind, it is more difficult to imagine how further inhibition of fatty acid oxidation will ultimately be beneficial, particularly given that RV function is the major determinant of survival in patients with PAH.

Broader metabolic reprogramming: the TCA cycle and redox balance

More importantly, if the data regarding suppression of both glucose and fat oxidation in PAH (or at least in the RV in PAH) are taken at face value, then it appears that a more fundamental breakdown in metabolic regulation underlies PAH. If Randle's cycle were functioning as expected, suppression of glucose oxidation should result in upregulation of fat oxidation, and vice versa. The fact that this seems not to happen in PAH suggests that additional coexisting metabolic dysregulation is a key part of the molecular pathogenesis of PAH. One piece of the puzzle is likely a disruption of the normal mechanisms that regulate the inflow and outflow of carbon in the TCA cycle through pathways other than pyruvate/acetyl-CoA. The inflow reactions are collectively called anaplerosis, and the outflow reactions are collectively referred to as cataplerosis. We have published that in pulmonary endothelial cells expressing PAH-causing BMPR2 mutations, several key anaplerotic pathways are disrupted compared to wild-type endothelium.⁷⁰ This has also been demonstrated in the RV with respect to a specific anaplerotic pathway, glutaminolysis.⁷¹ The amino acid l-glutamine can be converted to glutamate and then to 2-oxoglutarate (also called α-ketoglutarate), which enters directly into the TCA cycle through a route that involves neither glucose nor fatty acids. Glutaminolysis has been shown to be upregulated in the RV in PAH, and this may indeed be part of the broader metabolic dysregulation that would allow for simultaneous suppression of glucose oxidation and fatty acid oxidation. Other anaplerotic pathways (e.g., formation of oxaloacetate directly either from pyruvate via pyruvate carboxylase or from aspartate via aspartate aminotransferase) have not been fully explored in PAH, although we have shown that the metabolism of multiple amino acids is altered in the PAH endothelium.⁷⁰

Metabolism in service to redox balance

The end result of most of the carbon metabolism pathways in the cell is the generation of reducing equivalents that participate in critical cellular redox reactions. The reducing equivalents produced include NADH, produced from NAD⁺ by glycolysis, fatty acid oxidation, and the TCA cycle; NADPH, produced from NADP⁺ primarily by the pentose phosphate pathway, with additional contributions from isoforms of malic enzyme and isocitrate dehydrogenase (IDH); and FADH2, primarily from flavoproteins involved in fatty acid oxidation (electron transport flavoprotein complex) and the TCA cycle (e.g., pyruvate dehydrogenase, 2-oxoglutarate dehydrogenase, and succinate dehydrogenase). Not only is the balance between the reduced and oxidized forms of these electron carriers a key determinant of the overall redox balance within the cell, but levels of these intermediates also regulate the metabolic pathways themselves. As mentioned above, NADH produced from fatty acid oxidation provides feedback inhibition to pyruvate dehydrogenase to limit the complete oxidation of glucose-derived carbon. This is one of the key molecular underpinnings of Randle's cycle.

In trying to understand the complex regulation of molecular metabolism in PAH, perhaps more focus should be placed on understanding redox balance at the level of electron carriers in specific cellular compartments. For example, one possible mechanism by which glucose oxidation and fatty acid oxidation could both be simultaneously suppressed is generation of NADH through an alternative pathway, such as upregulation of the activity of NAD⁺-dependent IDH isoforms. Indeed, we have shown that IDH1/2 activity is increased in the plasma of PAH patients.⁷⁰ As a second example, recent publications have suggested that decreased activity of one or more sirtuin isoforms may be an important contributor to PAH pathogenesis.⁷² The sirtuins are lysine deacetylases in the same superfamily as histone deacetylases. Several sirtuin isoforms have been shown to be localized to mitochondria (Sirt3, Sirt4, and Sirt5) and to be master regulators of metabolic program. Their metabolic regulatory function, particularly in the case of Sirt3, is thought to be directly linked to their use of NAD⁺ as a required cofactor, allowing their activity to be dynamically regulated by the redox state of the cell. Although decreased expression of sirtuin isoforms could certainly contribute to the development of PAH, it is at least as likely that sirtuin activity is regulated by cofactor availability. Indeed, the concept of cofactor availability as a regulatory mechanism for posttranslational modifications has been very nicely captured in other work from this same group.⁷³ The most dynamic regulation of sirtuin activity almost requires that a significant portion of sirtuin activity be regulated by factors other than expression levels, such as cofactor availability and subcellular localization. Viewed in this way, the availability of NAD⁺ as a cofactor would be a critical regulator of sirtuin function in PAH, and any change to steady state that decreased NAD⁺ concentrations (e.g., conversion of NAD⁺ to NADH by increased activity of IDH1/2) would decrease sirtuin activity. Although existing data certainly do not support the above IDH-dependent mechanism as even one of the critical molecular metabolic alterations underlying PAH, the above cases are nonetheless illustrative of the fact that we are only just beginning to understand the complex interplay between redox (im)balance and metabolic dysfunction in PAH.

MITOCHONDRIA IN PAH: THE CENTER OF CELLULAR REDOX BIOLOGY

As discussed in the above sections, the processes of oxidant injury, oxidant signaling, redox balance, and cellular metabolism are all intimately interconnected. The main point of integration for all of these processes is the mitochondrion.⁷⁴ Oxidative carbon-substrate metabolism and the resultant electron carriers converge on the mitochondrial electron transport system to generate the proton gradient that lies at the heart of chemiosmotically driven ATP synthesis. The proton gradient (or proton-motive force) is generated by the proton-pumping action of several mitochondrial complexes and is powered by the free energy liberated in sequential electron transfers. The ultimate electron acceptor in this process is molecular oxygen, which is converted to water when completely reduced. (Note that the overall stoichiometry of the combination of carbon substrate with oxygen to release carbon dioxide, water, and energy is equivalent to a combustion reaction.) However, no process happens with perfect efficiency, and there are multiple points within the electron transport system where electrons can be prematurely transferred to molecular oxygen (with cytochrome c oxidase—also known as complex IV—being the site where oxygen is normally reduced to water). A 1-electron reduction of molecular oxygen results in the formation of superoxide anion, and in this way there is a direct link between metabolism and oxidant signaling and injury.

A number of conclusions can be derived from the relationships depicted in Figure 5. For example, one would predict that different fuel sources would yield different efficiencies when expressed as the ratio of electrons transferred to superoxide produced. For each turn of the TCA cycle, electrons are introduced via NADH oxidation by complex I and by conversion of succinate to fumarate by complex II. However, only complex I serves as a major site of superoxide generation, and so free-radical leak is minimized. For the same number of electrons transferred through the electron-transfer flavoprotein complex (responsible for fatty acid oxidation), one would predict a much higher rate of superoxide formation, as the route of entry for electrons into the electron transport system is the same as the route of inappropriate egress that results in superoxide formation. Indeed, increased fatty acid oxidation has been shown to lead directly to increased generation of ROSs.⁷⁵

Figure 5.

Major sites for superoxide generation in the mitochondrial electron transport system. Shown are the major sites for free-radical leak, the overall direction of electron flow, and the overall direction of proton flow in the mitochondrion. ADP: adenosine diphosphate; ATP: adenosine triphosphate; CoQ: coenzyme Q/ubiquinol/ubiquinone; Cx: complex; cyt c: cytochrome c; ETF: electron transport flavoprotein system; FAO: fatty acid oxidation; Pi: phosphate ion.

Hypoxia and redox biology in PAH. The case of hypoxia deserves special mention with regard to redox biology in PAH. Chronic hypoxia was one of the earliest models of experimental pulmonary hypertension, and although the presence of hypoxia in WHO group I PAH is secondary to vessel dropout and falling cardiac output (as opposed to being primarily causative in WHO group III pulmonary hypertension), many publications on experimental PAH use hypoxia as part of the model system. The effects of hypoxia to activate HIF and cause metabolic remodeling, characterized by increased glycolysis, decreased carbon flow through pyruvate dehydrogenase, increased reliance on glutamine metabolism to support fatty acid synthesis, and isoform switching of cytochrome c oxidase, are all fairly well known.^76–80 However, there is a great deal of literature regarding increased oxidative stress under conditions of hypoxia. On the face of it, this is somewhat difficult to understand. With relatively less molecular oxygen available, it would seem that there should be an absolute reduction in superoxide production, at least from the mitochondria. The reverse has been demonstrated, in which increased molecular oxygen concentrations (hyperoxia) increase superoxide production from mitochondria, increase mitochondrial oxidant injury, and increase PAH.^35,81–84 For the case of hypoxia, Hoffman and Brookes,⁸⁵ in 2009, published a beautiful set of experiments in isolated mitochondria that quantitatively answers some of these questions. The summary is that, in hypoxic conditions with ample substrate, the proportion of molecular oxygen converted to superoxide can and does increase, but the absolute number of molecules of oxygen available for 1-electron reduction is far lower. The net result is that absolute superoxide generation actually decreases in hypoxic conditions. Moreover, these and other published studies suggest that the site(s) of maximum superoxide generation are a complex function of oxygen availability, relative substrate concentrations, relative expression of different components of mitochondrial metabolism, and perhaps even regulatory signaling outside the mitochondria.^75,86,87 Finally, and perhaps not surprisingly, there is mounting evidence that pulmonary hypertension driven by hypoxia, while sharing some mechanisms and even some of the features of the redox pathobiology associated with heritable and idiopathic PAH, is a fundamentally different disease. Indeed, some of the disrupted metabolic regulatory mechanisms that play a role in PAH pathogenesis have been shown not to affect pulmonary hypertension driven by chronic hypoxia.⁸⁸ Of course, these mechanisms are central to the pathogenesis of WHO group III pulmonary hypertension.

SUMMARY AND FUTURE DIRECTIONS

Investigations examining the facets of redox biology, including oxidant signaling and stress and molecular metabolic reprogramming, have taught us much about the pathogenesis of PAH. There are many promising new therapeutic targets that come from these mechanisms, some of which are already being explored in PAH while others are currently under development in cancer. To continue to advance our understanding and to maximize the likelihood of truly disease-modifying therapies being developed for PAH, we must redouble our efforts to elucidate fine mechanistic details in relevant model systems and then verify those mechanisms in humans. The devil, as ever, is in the details. We must be as specific as possible when studying “oxidative stress” and its effects in any given model system. We should seek to examine all of the relevant interrelationships between metabolic pathways and identify regulatory nodes that, when disrupted or altered, explain more of the variability in molecular metabolism in PAH than our current models. We must broaden our thinking about redox biology in PAH to move beyond old concepts and interventions.

Footnotes

Source of Support: Nil.

Conflict of Interest: None declared.

References

Zheng

Liu

Chen

Xiong

. Oral targeted therapies in the treatment of pulmonary arterial hypertension: a meta-analysis of clinical trials. Pulm Pharmacol Ther 2014;29(2):241–249.

Stacher

Graham

Hunt

Gandjeva

Groshong

McLaughlin

Jessup

. Modern age pathology of pulmonary arterial hypertension. Am J Respir Crit Care Med 2012;186(3):261–272.

Channick

Delcroix

Ghofrani

Hunsche

Jansa

Le Brun

Mehta

. Effect of macitentan on hospitalizations: results from the SERAPHIN trial. JACC Heart Fail 2015;3(1):1–8.

Ghofrani

Galiè

Grimminger

Grünig

Humbert

Jing

Keogh

. Riociguat for the treatment of pulmonary arterial hypertension. N Engl J Med 2013;369(4):330–340.

Simonneau

Torbicki

Hoeper

Delcroix

Karlócai

Galiè

Degano

. Selexipag: an oral, selective prostacyclin receptor agonist for the treatment of pulmonary arterial hypertension. Eur Respir J 2012;40(4):874–880.

Lane

Talati

Austin

Hemnes

Johnson

Fessel

Blackwell

. Oxidative injury is a common consequence of BMPR2 mutations. Pulm Circ 2011;1(1):72–83.

Grobe

Wells

Benavidez

Oishi

Azakie

Fineman

Black

. Increased oxidative stress in lambs with increased pulmonary blood flow and pulmonary hypertension: role of NADPH oxidase and endothelial NO synthase. Am J Physiol Lung Cell Mol Physiol 2006;290(5):L1069–L1077.

Archer

Gomberg-Maitland

Maitland

Rich

Garcia

Weir

. Mitochondrial metabolism, redox signaling, and fusion: a mitochondria-ROS-HIF-1α-Kv1.5 O₂-sensing pathway at the intersection of pulmonary hypertension and cancer. Am J Physiol Heart Circ Physiol 2008;294(2):H570–H578.

Tabima

Frizzell

Gladwin

. Reactive oxygen and nitrogen species in pulmonary hypertension. Free Radic Biol Med 2012;52(9):1970–1986.

10.

Cracowski

Degano

Chabot

Labarère

Schwedhelm

Monneret

Iuliano

. Independent association of urinary F₂-isoprostanes with survival in pulmonary arterial hypertension. Chest 2012;142(4):869–876.

11.

Mikkelsen

Wardman

. Biological chemistry of reactive oxygen and nitrogen and radiation-induced signal transduction mechanisms. Oncogene 2003;22(37):5734–5754.

12.

Valko

Morris

Cronin

. Metals, toxicity and oxidative stress. Curr Med Chem 2005;12(10):1161–1208.

13.

Benov

. How superoxide radical damages the cell. Protoplasma 2001;217(1–3):33–36.

14.

Quinlan

Perevoshchikova

Hey-Mogensen

Orr

Brand

. Sites of reactive oxygen species generation by mitochondria oxidizing different substrates. Redox Biol 2013;1:304–312.

15.

Bréton-Romero

Lamas

. Hydrogen peroxide signaling in vascular endothelial cells. Redox Biol 2014;2:529–534.

16.

Marinho

Real

Cyrne

Soares

Antunes

. Hydrogen peroxide sensing, signaling and regulation of transcription factors. Redox Biol 2014;2:535–562.

17.

Kalyanaraman

Darley-Usmar

Davies

Dennery

Forman

Grisham

Mann

Moore

Roberts

II Ischiropoulos

. Measuring reactive oxygen and nitrogen species with fluorescent probes: challenges and limitations. Free Radic Biol Med 2012;52(1):1–6.

18.

Zielonka

Kalyanaraman

. Hydroethidine- and MitoSOX-derived red fluorescence is not a reliable indicator of intracellular superoxide formation: another inconvenient truth. Free Radic Biol Med 2010;48(8):983–1001.

19.

Kadiiska

Basu

Brot

Cooper

Csallany

Davies

George

. Biomarkers of oxidative stress study. V. Ozone exposure of rats and its effect on lipids, proteins, and DNA in plasma and urine. Free Radic Biol Med 2013;61:408–415.

20.

Kadiiska

Gladen

Baird

Dikalova

Sohal

Hatch

Jones

Mason

Barrett

. Biomarkers of oxidative stress study: are plasma antioxidants markers of CCl₄ poisoning? Free Radic Biol Med 2000;28(6):838–845.

21.

Kadiiska

Gladen

Baird

Germolec

Graham

Parker

Nyska

. Biomarkers of oxidative stress study. II. Are oxidation products of lipids, proteins, and DNA markers of CCl₄ poisoning? Free Radic Biol Med 2005;38(6):698–710.

22.

Kadiiska

Gladen

Baird

Graham

Parker

Ames

Basu

. Biomarkers of oxidative stress study. III. Effects of the nonsteroidal anti-inflammatory agents indomethacin and meclofenamic acid on measurements of oxidative products of lipids in CCl₄ poisoning. Free Radic Biol Med 2005;38(6):711–718.

23.

Kadiiska

Hatch

Nyska

Jones

Hensley

Stocker

George

. Biomarkers of oxidative stress study. IV. Ozone exposure of rats and its effect on antioxidants in plasma and bronchoalveolar lavage fluid. Free Radic Biol Med 2011;51(9):1636–1642.

24.

Kadiiska

Peddada

Herbert

Basu

Hensley

Jones

Hatch

Mason

. Biomarkers of oxidative stress study. VI. Endogenous plasma antioxidants fail as useful biomarkers of endotoxin-induced oxidative stress. Free Radic Biol Med 2015;81:100–106.

25.

Al-Mehdi

Pastukh

Swiger

Reed

Patel

Bardwell

Pastukh

Alexeyev

Gillespie

. Perinuclear mitochondrial clustering creates an oxidant-rich nuclear domain required for hypoxia-induced transcription. Sci Signal 2012;5(231):ra47. doi:10.1126/scisignal.2002712.

26.

Archer

Marsboom

Kim

Zhang

Toth

Svensson

Dyck

. Epigenetic attenuation of mitochondrial superoxide dismutase 2 in pulmonary arterial hypertension: a basis for excessive cell proliferation and a new therapeutic target. Circulation 2010;121(24):2661–2671.

27.

Cracowski

Bessard

Pepin

Bessard

Schwebel

Stanke-Labesque

Pison

. Increased lipid peroxidation in patients with pulmonary hypertension. Am J Respir Crit Care Med 2001;164(6):1038–1042.

28.

Robbins

Morrow

Christman

. Oxidant stress but not thromboxane decreases with epoprostenol therapy. Free Radic Biol Med 2005;38(5):568–574.

29.

Bowers

Cool

Murphy

Tuder

Hopken

Flores

Voelkel

. Oxidative stress in severe pulmonary hypertension. Am J Respir Crit Care Med 2004;169(6):764–769.

30.

Voelkel

Bogaard

Al Husseini

Farkas

Gomez-Arroyo

Natarajan

. Antioxidants for the treatment of patients with severe angioproliferative pulmonary hypertension? Antioxid Redox Signal 2013;18(14):1810–1817.

31.

Masri

Comhair

Dostanic-Larson

Kaneko

Dweik

Arroliga

Erzurum

. Deficiency of lung antioxidants in idiopathic pulmonary arterial hypertension. Clin Transl Sci 2008;1(2):99–106.

32.

Kamezaki

Tasaki

Yamashita

Tsutsui

Koide

Nakata

Tanimoto

. Gene transfer of extracellular superoxide dismutase ameliorates pulmonary hypertension in rats. Am J Respir Crit Care Med 2008;177(2):219–226.

33.

Guo

Fassett

. Exacerbated pulmonary arterial hypertension and right ventricular hypertrophy in animals with loss of function of extracellular superoxide dismutase. Hypertension 2011;58(2):303–309.

34.

Chow

Fessel

Ihida-Stansbury

Schmidt

Gaskill

Alvarez

Graham

. Dysfunctional resident lung mesenchymal stem cells contribute to pulmonary microvascular remodeling. Pulm Circ 2013;3(1):31–49.

35.

Fessel

Flynn

Robinson

Penner

Gladson

Kang

Wasserman

Hemnes

West

. Hyperoxia synergizes with mutant bone morphogenic protein receptor 2 to cause metabolic stress, oxidant injury, and pulmonary hypertension. Am J Respir Cell Mol Biol 2013;49(5):778–787.

36.

Zhao

Mirza

Huang

Potula

Vogel

Brovkovych

Yuan

Wharton

Malik

. Persistent eNOS activation secondary to caveolin-1 deficiency induces pulmonary hypertension in mice and humans through PKG nitration. J Clin Invest 2009;119(7):2009–2018.

37.

Siddiqui

Komarova

Vogel

Gao

Bonini

Rajasingh

Zhao

Brovkovych

Malik

. Caveolin-1-eNOS signaling promotes p190RhoGAP-A nitration and endothelial permeability. J Cell Biol 2011;193(5):841–850.

38.

Maniatis

Shinin

Schraufnagel

Okada

Vogel

Malik

Minshall

. Increased pulmonary vascular resistance and defective pulmonary artery filling in caveolin-1^–/– mice. Am J Physiol Lung Cell Mol Physiol 2008;294(5):L865–L873.

39.

Austin

LeDuc

Rosenzweig

Berman E

Borczuk

Phillips

III Palomero

. Whole exome sequencing to identify a novel gene (caveolin-1) associated with human pulmonary arterial hypertension. Circ Cardiovasc Genet 2012;5(3):336–343.

40.

Radi

. Nitric oxide, oxidants, and protein tyrosine nitration. Proc Natl Acad Sci USA 2004;101(12):4003–4008.

41.

Weinhouse

Warburg

Burk

Schade

. On respiratory impairment in cancer cells. Science 1956;124(3215):267–272.

42.

Warburg

. On the origin of cancer cells. Science 1956;123(3191):309–314.

43.

Rehman

Archer

. A proposed mitochondrial-metabolic mechanism for initiation and maintenance of pulmonary arterial hypertension in fawn-hooded rats: the Warburg model of pulmonary arterial hypertension. Adv Exp Med Biol 2010;661:171–185.

44.

Kakinuma

Miyauchi

Suzuki

Yuki

Murakoshi

Goto

Yamaguchi

. Enhancement of glycolysis in cardiomyocytes elevates endothelin-1 expression through the transcriptional factor hypoxia-inducible factor-1α. Clin Sci 2002;103(suppl. 48):210S–214S.

45.

Marsboom

Wietholt

Haney

Toth

Ryan

Morrow

Thenappan

. Lung ¹⁸F-fluorodeoxyglucose positron emission tomography for diagnosis and monitoring of pulmonary arterial hypertension. Am J Respir Crit Care Med 2012;185(6):670–679.

46.

Koeck

Lara

Neumann

DiFilippo

Koo

Janocha

. Alterations of cellular bioenergetics in pulmonary artery endothelial cells. Proc Natl Acad Sci USA 2007;104(4):1342–1347.

47.

Allard

. Energy substrate metabolism in cardiac hypertrophy. Curr Hypertens Rep 2004;6(6):430–435.

48.

Piao

Marsboom

Archer

. Mitochondrial metabolic adaptation in right ventricular hypertrophy and failure. J Mol Med 2010;88(10):1011–1020.

49.

Erzurum

. Endothelial cell energy metabolism, proliferation, and apoptosis in pulmonary hypertension. Compr Physiol 2011;1(1):357–372.

50.

Rabinovitch

. Molecular pathogenesis of pulmonary arterial hypertension. J Clin Invest 2012;122(12):4306–4313.

51.

Rich

. Right ventricular adaptation and maladaptation in chronic pulmonary arterial hypertension. Cardiol Clin 2012;30(2):257–269.

52.

Cottrill

Chan

. Metabolic dysfunction in pulmonary hypertension: the expanding relevance of the Warburg effect. Eur J Clin Invest 2013;43(8):855–865.

53.

Tuder

Davis

Graham

. Targeting energetic metabolism: a new frontier in the pathogenesis and treatment of pulmonary hypertension. Am J Respir Crit Care Med 2012;185(3):260–266.

54.

Garland

Randle

Newsholme

. Citrate as an intermediary in the inhibition of phosphofructokinase in rat heart muscle by fatty acids, ketone bodies, pyruvate, diabetes, and starvation. Nature 1963;200(4902):169–170.

55.

Randle

Garland

Hales

Newsholme

. The glucose fatty-acid cycle: its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 1963;281(7285):785–789.

56.

Hue

Taegtmeyer

. The Randle cycle revisited: a new head for an old hat. Am J Physiol Endocrinol Metab 2009;297(3):E578–E591.

57.

Abdelmalak

Lew

Ramezani

Shroads

Coats

Langaee

Shankar

Neiberger

Subramony

Stacpoole

. Long-term safety of dichloroacetate in congenital lactic acidosis. Mol Genet Metab 2013;109(2):139–143.

58.

Bonnet

Michelakis

Porter

Andrade-Navarro

Thébaud

Bonnet

Haromy

. An abnormal mitochondrial–hypoxia inducible factor-1α–Kv channel pathway disrupts oxygen sensing and triggers pulmonary arterial hypertension in fawn hooded rats: similarities to human pulmonary arterial hypertension. Circulation 2006;113(22):2630–2641.

59.

Bonnet

Archer

Allalunis-Turner

Haromy

Beaulieu

Thompson

Lee

. A mitochondria-K⁺ channel axis is suppressed in cancer and its normalization promotes apoptosis and inhibits cancer growth. Cancer Cell 2007;11(1):37–51.

60.

Piao

Fang

Cadete

Wietholt

Urboniene

Toth

Marsboom

. The inhibition of pyruvate dehydrogenase kinase improves impaired cardiac function and electrical remodeling in two models of right ventricular hypertrophy: resuscitating the hibernating right ventricle. J Mol Med 2010;88(1):47–60.

61.

Archer

Fang

Ryan

Piao

. Metabolism and bioenergetics in the right ventricle and pulmonary vasculature in pulmonary hypertension. Pulm Circ 2013;3(1):144–152.

62.

Piao

Sidhu

Fang

Ryan

Parikh

Hong

Toth

. FOXO1-mediated upregulation of pyruvate dehydrogenase kinase-4 (PDK4) decreases glucose oxidation and impairs right ventricular function in pulmonary hypertension: therapeutic benefits of dichloroacetate. J Mol Med 2013;91(3):333–346.

63.

Fang

Piao

Hong

Toth

Marsboom

Bache-Wiig

Rehman

Archer

. Therapeutic inhibition of fatty acid oxidation in right ventricular hypertrophy: exploiting Randle's cycle. J Mol Med 2012;90(1):31–43.

64.

Sutendra

Bonnet

Rochefort

Haromy

Folmes

Lopaschuk

Dyck

Michelakis

. Fatty acid oxidation and malonyl-CoA decarboxylase in the vascular remodeling of pulmonary hypertension. Sci Transl Med 2010;2(44):44ra58. doi:10.1126/scitranslmed.3001327.

65.

Zhao

Peng

Hsin

Mura

Chu

. Metabolomic heterogeneity of pulmonary arterial hypertension. PLoS ONE 2014;9:e88727. doi:10.1371/journal.pone.0088727.

66.

Neely

Morgan

. Relationship between carbohydrate and lipid metabolism and the energy balance of heart muscle. Annu Rev Physiol 1974;36:413–459.

67.

Gomez-Arroyo

Santos-Martinez

Aranda

Pulido

Beltran

Muñoz-Castellanos

Dominguez-Cano

Sonnino

Voelkel

Sandoval

. Differences in right ventricular remodeling secondary to pressure overload in patients with pulmonary hypertension. Am J Respir Crit Care Med 2014;189(5):603–606.

68.

Buermans

Redout

Schiel

Musters

Zuidwijk

Eijk

van Hardeveld

. Microarray analysis reveals pivotal divergent mRNA expression profiles early in the development of either compensated ventricular hypertrophy or heart failure. Physiol Genomics 2005;21(3):314–323.

69.

Hemnes

Brittain

Trammell

Fessel

Austin

Penner

Maynard

. Evidence for right ventricular lipotoxicity in heritable pulmonary arterial hypertension. Am J Respir Crit Care Med 2014;189(3):325–334.

70.

Fessel

Hamid

Wittmann

Robinson

Blackwell

Tada

Tanabe

Tatsumi

Hemnes

West

. Metabolomic analysis of bone morphogenetic protein receptor type 2 mutations in human pulmonary endothelium reveals widespread metabolic reprogramming. Pulm Circ 2012;2(2):201–213.

71.

Piao

Fang

Parikh

Ryan

Toth

Archer

. Cardiac glutaminolysis: a maladaptive cancer metabolism pathway in the right ventricle in pulmonary hypertension. J Mol Med 2013;91(10):1185–1197.

72.

Paulin

Dromparis

Sutendra

Gurtu

Zervopoulos

Bowers

Haromy

. Sirtuin 3 deficiency is associated with inhibited mitochondrial function and pulmonary arterial hypertension in rodents and humans. Cell Metab 2014;20(5):827–839.

73.

Sutendra

Kinnaird

Dromparis

Paulin

Stenson

Haromy

Hashimoto

Zhang

Flaim

Michelakis

. A nuclear pyruvate dehydrogenase complex is important for the generation of acetyl-CoA and histone acetylation. Cell 2014;158(1):84–97.

74.

Schumacker

Gillespie

Nakahira

Choi

Crouser

Piantadosi

Bhattacharya

. Mitochondria in lung biology and pathology: more than just a powerhouse. Am J Physiol Lung Cell Mol Physiol 2014;306(11):L962–L974.

75.

St-Pierre

Buckingham

Roebuck

Brand

. Topology of superoxide production from different sites in the mitochondrial electron transport chain. J Biol Chem 2002;277(47):44784–44790.

76.

Wise

Ward

Shay

Cross

Gruber

Sachdeva

Platt

DeMatteo

Simon

Thompson

. Hypoxia promotes isocitrate dehydrogenase-dependent carboxylation of α-ketoglutarate to citrate to support cell growth and viability. Proc Natl Acad Sci USA 2011;108(49):19611–19616.

77.

Semenza

Roth

Fang

Wang

. Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem 1994;269(38):23757–23763.

78.

Rumsey

Abbott

Bertelsen

Mallamaci

Hagan

Nelson

Erecinska

. Adaptation to hypoxia alters energy metabolism in rat heart. Am J Physiol Heart Circ Physiol 1999;276(1):H71–H80.

79.

Semenza

. Oxygen-dependent regulation of mitochondrial respiration by hypoxia-inducible factor 1. Biochem J 2007;405():1–9.

80.

Fukuda

Zhang

Kim

Shimoda

Dang

Semenza

. HIF-1 regulates cytochrome oxidase subunits to optimize efficiency of respiration in hypoxic cells. Cell 2007;129(1):111–122.

81.

Freeman

Topolosky

Crapo

. Hyperoxia increases oxygen radical production in rat lung homogenates. Arch Biochem Biophys 1982;216(2):477–484.

82.

Turrens

Freeman

Crapo

. Hyperoxia increases H₂O₂ release by lung mitochondria and microsomes. Arch Biochem Biophys 1982;217(2):411–421.

83.

Turrens

Freeman

Levitt

Crapo

. The effect of hyperoxia on superoxide production by lung submitochondrial particles. Arch Biochem Biophys 1982;217(2):401–410.

84.

Fessel

Porter

Moore

Sheller

Roberts

II . Discovery of lipid peroxidation products formed in vivo with a substituted tetrahydrofuran ring (isofurans) that are favored by increased oxygen tension. Proc Natl Acad Sci USA 2002;99(26):16713–16718.

85.

Hoffman

Brookes

. Oxygen sensitivity of mitochondrial reactive oxygen species generation depends on metabolic conditions. J Biol Chem 2009;284(24):16236–16245.

86.

Guzy

Schumacker

. Oxygen sensing by mitochondria at complex III: the paradox of increased reactive oxygen species during hypoxia. Exp Physiol 2006;91(5):807–819.

87.

Hoffman

Salter

Brookes

. Response of mitochondrial reactive oxygen species generation to steady-state oxygen tension: implications for hypoxic cell signaling. Am J Physiol Heart Circ Physiol 2007;292(1):H101–H108.

88.

Waypa

Osborne

Marks

Berkelhamer

Kondapalli

Schumacker

. Sirtuin 3 deficiency does not augment hypoxia-induced pulmonary hypertension. Am J Respir Cell Mol Biol 2013;49(6):885–891.

Redox Biology in Pulmonary Arterial Hypertension (2013 Grover Conference Series)

Abstract

Keywords

BRIEF REVIEW OF REACTIVE OXYGEN SPECIES IN PAH

OXIDATIVE STRESS, OXIDANT SIGNALING, AND OXIDANT INJURY

The problems with “oxidative stress”

Methods for assessing “oxidative stress”

Oxidant signaling in PAH

Oxidant injury in PAH

Ambiguity in distinguishing oxidant injury from signaling

CARBON METABOLISM AND REDOX BIOLOGY IN PAH: BEYOND GLUCOSE OXIDATION

Glucose oxidation versus fat oxidation: which to inhibit, which to activate?

Broader metabolic reprogramming: the TCA cycle and redox balance

Metabolism in service to redox balance

MITOCHONDRIA IN PAH: THE CENTER OF CELLULAR REDOX BIOLOGY

SUMMARY AND FUTURE DIRECTIONS

Footnotes

References