Renal Epithelial Mitochondria: Implications for Hypertensive Kidney Disease

Abstract

According to the Centers for Disease Control and Prevention, 1 in 2 U.S. adults have hypertension, and more than 1 in 7 chronic kidney disease. In fact, hypertension is the second leading cause of kidney failure in the United States; it is a complex disease characterized by, leading to, and caused by renal dysfunction. It is well-established that hypertensive renal damage is accompanied by mitochondrial damage and oxidative stress, which are differentially regulated and manifested along the nephron due to the diverse structure and functions of renal cells. This article provides a summary of the relevant knowledge of mitochondrial bioenergetics and metabolism, focuses on renal mitochondrial function, and discusses the evidence that has been accumulated regarding the role of epithelial mitochondrial bioenergetics in the development of renal tissue dysfunction in hypertension.

Introduction

Recent years have seen the “renaissance” of mitochondrial research in the context of kidney diseases. Mitochondrial dysfunction, however, is often loosely defined and could mean various things in terms of bioenergetics, redox balance, redox signaling, or mitochondrial quality control. There are many open questions remaining regarding the role of mitochondriarelated processes in renal pathophysiology. A large fraction of the literature investigating renal mitochondrial function and dysfunction focuses on either acute kidney injury (AKI) or diabetic kidney diseases (DKD) as it is increasingly clear that mitochondrial injury plays a central role in the pathogenesis of these disorders (20, 21, 42, 64, 66, 141, 147, 153, 176, 177, 188). Much less is known about changes in mitochondrial function and bioenergetics in hypertensive renal damage, despite a large body of information accumulated about “oxidative stress” in general. Hypertension is a global burden and a risk factor for the development of cardiovascular and renal complications (76). Despite the availability of multiple antihypertensive drugs, up to 18% of hypertensive patients exhibit resistant hypertension that is not controlled by three or more medications (54), which highlights the need for the development of new treatments.

The kidney is a key organ for the development of hypertension (41, 44). Energy production and substrate availability in the kidney contribute to renal blood pressure control (157, 193). This article is focused on hypertensive renal damage and provides an overview of the current state of the field, starting from the basics of mitochondrial biochemistry and biophysics to fundamental research in hypertensive animal models. We provide a general introduction to the bioenergetics of the mitochondrial electron transport chain (ETC) and oxidative phosphorylation. Then—given the very heterogeneous landscape of the kidney—we discuss differences in mitochondrial density and function in various renal cells, their redox regulation, and dynamics. Lastly, we discuss the historical and most recent advances in renal mitochondrial research focusing on hypertensive renal damage. Our goal is to discuss the established and emerging aspects of the current research, highlighting open questions and future opportunities in this promising area.

Brief Overview of the Mitochondrial Electron Transport Chain Function

The chemiosmotic theory developed by Peter Mitchell in the 1960s fundamentally changed how we think about the bioenergetics of cells and mitochondria. His experiments laid the foundation for our perception of cell metabolism, energy needs, and adenosine triphosphate (ATP) production, and while it initially infuriated the scientific community, it gained Dr. Mitchell the Nobel Prize in chemistry in 1978. He hypothesized that electron transport and ATP synthesis are embedded in the same membrane, these components are in contact with one another and the other side of the membrane, and the membrane is impermeable to protons (134). Thus, there should be an exchange of protons across the mitochondrial membrane, and the electron transport versus ATP synthesis are separate processes. Throughout the past four decades, as the available tools and approaches evolved, we gained a much deeper understanding of the mitochondrial ETC, its components, and its functions.

To briefly summarize, in the ETC electrons are transferred through a series of redox reactions with the help of protein complexes that drive oxidative phosphorylation (Figure 1). This transfer is coupled with protons (H⁺) being pumped across the inner membrane, creating an electrochemical gradient. This gradient then drives energy production in the form of ATP generation. The transfer of electrons occurs in the inner mitochondrial membrane, with many of the ETC complexes embedded in the bilayer. Through a series of reduction and oxidation reactions, free energy is released, which is used to pump protons across the membrane and into intermembrane space. This proton gradient is then used for the synthesis of ATP by the ATP synthase complex (7). Specifically, equivalents such as nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FADH₂) are used to transfer the electrons to the ETC, with the help of four multi-subunit protein complexes and other mobile electron carriers. Electrons are passed through the ETC in a manner that electron donors always pass electrons to an acceptor with higher redox potential. This acceptor then becomes a donor and donates an electron to yet another acceptor, with the process continuing until reaching the final acceptor, oxygen, which is then finally reduced to water (112). Each step releases free energy because reaction products always have lower energy than the preceding donor. This energy (electrochemical gradient) is used to phosphorylate adenosine diphosphate (ADP) to create ATP in oxidative phosphorylation (135). The energy from these redox reactions also establishes a proton gradient across the inner membrane. The proton gradient is largely responsible for maintaining the mitochondrial membrane potential (ΔΨm) (224). Because of this membrane potential, it is possible for the ATP synthase (Complex V) to utilize H⁺ and direct them back to the matrix for ATP production; furthermore, ΔΨm is being used for transporting metabolites or other ions in and out of the mitochondria. The redox reactions of the ETC use oxygen, products of the TCA cycle as well as fatty acid and amino acid metabolism, making mitochondria the central part of cellular metabolism and bioenergetics.

Figure 1. Summary and general schematics of mitochondrial electron transport chain (ETC) and oxidative phosphorylation.

Electrons are transferred through Complexes I to IV from a higher energy donor to a lower energy electron acceptor, with the help of cofactors. As a net result, protons are pumped at Complexes I, III, and IV into the intermembrane space. This proton gradient is then used to produce ATP in oxidative phosphorylation. TCA: tricarboxylic acid cycle, NAD⁺/NADH: nicotinamide adenine dinucleotide/reduced form, FAD/FADH₂: Flavin adenine dinucleotide/reduced form, CoQ: coenzyme Q, ADP: adenosine diphosphate, ATP: adenosine triphosphate.

Each of the four major protein complexes of the ETC (commonly known as Complex I, II, III, and IV) is highly specialized and contains multiple subunits. Complex I is an NADH ubiquinone oxidoreductase with 45 known subunits that work together to oxidize NADH in a process that yields four protons, by removing electrons from NADH and transferring them to ubiquinone (coenzyme Q or coQ) (146). More specifically, first NADH is converted to NAD⁺ by the reduction of flavin mononucleotide to FMNH₂ in a two-electron step. Then, FMNH₂ is oxidized in two one-electron steps, which are realized through an intermediate semiquinone. The electrons are transferred to an iron-sulfur cluster in the complex and then to ubiquinone. During this process, first a semiquinone free radical forms from coQ, and then the transfer of the other electron converts the semiquinone to ubiquinol (QH₂). As a result, a total of four H⁺ are translocated to the intermembrane space through Complex I. The mitochondrial substrates that support reactions through Complex I are pyruvate (carbohydrate-based), glutamate (amino acid-based), and their combinations with malate.

Complex II is made up of four subunits, and the official enzyme nomenclature for this complex is succinate dehydrogenase or succinate CoQ-reductase (180). Again, the name reveals the function: delivering additional electrons from succinate to the coQ pool. Unlike Complex I, Complex II is coupled with FAD, not NAD⁺, and thus electrons are transferred via FAD to coQ. The four subunits of Complex II are succinate dehydrogenase (SDHA), mitochondrial succinate dehydrogenase iron-sulfur subunit (SDHB), succinate dehydrogenase complex subunit C (SDHC), and subunit D (SDHD). At Complex II, no protons are transported to the intermembrane space, thus Complex II contributes less energy to the whole ETC process. As Complex II is also part of the TCA cycle (via SDHA/B), the substrate succinate feeds through this complex to energize reactions.

Complex III or cytochrome bc1 complex (sometimes simplified as cytochrome c reductase) is the next proton pump in ETC that contains 11 subunits (217). Complex III catalyzes the reduction of cytochrome c by oxidation of coQ. Cytochrome c is a small water-soluble heme protein located in the intermembrane space. During the reaction, known as the ubiquinone or Q cycle (113), two ubiquinols (QH2) are oxidized to ubiquinone and one ubiquinone is reduced back to ubiquinol. The two electrons are transferred from QH2 to ubiquinone through two cyt c intermediates. As a result, four protons are released to the intermembrane space but only two protons are taken up from the matrix side, creating a proton gradient across the membrane.

Finally, Complex IV, otherwise known as cytochrome c oxidase, is a 13-subunit enzyme (29). Complex IV is a copper-containing protein which also has several heme subunits (197). Its role is to facilitate the removal of the four electrons from cytochrome c, and pass them to the final electron acceptor, oxygen (O₂). In parallel, eight protons are removed from the matrix. Four protons produce two H₂O, while the other four protons are translocated across the membrane, contributing to the overall proton gradient.

At the last step, F₀F₁ ATP synthase (Complex V, a 16-subunit enzyme) couples the electron transfer process to oxidative phosphorylation (OXPHOS) and takes us back to Mitchell’s chemiosmotic theory. ETC and OXPHOS are coupled by a H⁺ gradient since Complex V utilizes this gradient to make ATP. The F₀ unit acts as an ion channel, through which the protons enter back into the mitochondrial matrix. F₀ consists of three subunits (a, b, and c) and is driven by the protonmotive force. The F₀ motor rotates the core and induces conformational changes in the F₁ unit, resulting in ATP synthesis (102). When the protons re-enter the matrix through F₀, free energy is released which is then used for ATP synthesis driven by the F₁ unit of Complex V. If necessary, uncoupling of the ATP synthesis and electron transfer can occur, where instead of ATP generation, free energy is dissipated as heat. This naturally occurs in brown adipose tissue, which is a specialized type of fat to produce heat in response to cold (34, 45, 100). In the kidney, in some cases, it can be useful to diminish ROS production at the expense of ATP synthesis (66).

Both electron and proton transfers are not entirely perfect processes. Electrons will always “leak”, escaping the transfer to the next acceptor; as a result, these leaked electrons can convert available O₂ to superoxide anion. As such, mitochondria in their natural state will always produce reactive oxygen species (ROS) to a certain degree. H⁺ leak also occurs, as the inner mitochondrial membrane is partially permeable to protons. Is it possible to have a reverse electron flow in ETC? Naturally, a reverse electron flow would require a significant amount of energy. Theoretically, the oxidized form of electron donors can be reduced by reverse transfer. Several studies have been conducted regarding the role of reverse electron flow and its role in increased mitochondrial ROS production. Situations where it could occur are “traffic jam” situations such as blockage of ATP synthase (leading to proton buildup), or a chronically overfed state with substrate excess. Its existence in vivo is debated, and the role of reverse electron flow in renal pathologies is currently also under investigation (107, 165). Figure 1 illustrates the ETC, proton gradient(s), and ATP synthesis.

Mitochondrial Dynamics

Initially, mitochondria were thought to be relatively steady organelles. Later discoveries revealed that mitochondria are dynamic, with the ability to form constantly changing networks in cells. They undergo either fusion, where contents of different mitochondria are mixed, or fission, where a mitochondrion splits into two daughter mitochondria (Figure 2). These processes are important for mitochondrial quality control (195, 198) and beyond, as fast adaptations may be necessary for a variety of cellular functions like cell cycle, cytoskeletal dynamics, or apoptosis. Fusion can contribute to the maintenance of healthy mitochondria and lowering cellular stress by integrating damaged mitochondria into another intact one. As a consequence, genetic complementation allows for the genome of two mitochondria (with different defects) to redundantly encode everything necessary for a functional organelle. Fission is a method to remove injured or otherwise dysfunctional mitochondria. A healthy level of fission is important for quality control and physiological function of the cell. The transitional steps in fusion and fission rely on GTPases from the dynamin family (179, 182). In fission, dynamin-related protein 1 (Drp1) gets recruited and undergoes oligomerization. This is followed by constriction of the mitochondria orchestrated by Drp1 in a synchronized fashion with another molecule called dynamin 2 (121). The result is mitochondrial fragmentation (31). Fusion is largely mediated by mitofusins 1 and 2 as well as the mitochondrial dynamin-like GTPase Opa1 (88, 98). Separate processes (which we will not detail herein) take care of the constriction or fusion of the inner mitochondrial membrane (Figure 2).

Figure 2. Mitochondrial fusion, fission, and mitophagy.

A simple schematic of mitochondria “life cycle” is shown. Mitochondria are dynamic organelles and constantly undergo fission and fusion. Damaged and dysfunctional mitochondria can be removed by fission (red arrow), which is a process assisted by the GTPase proteins Drp1, Drp2, and the outer membrane receptor Mff. Mitochondria can also undergo fusion, to “mix” and refresh genetic material to counter impairments (blue arrows). A healthy mitochondrion will fuse with an injured one with the help of Mfn ¹/2 and OPA1. The result is an elongated mito-chondrion. Mitophagy also contributes to mitochondria life cycle. In a healthy mitochondrion, PINK1 is recruited but degraded. In a damaged mitochondrion, PINK1 is not degraded, thus it recruits Parkin which then ubiquitinylates outer membrane proteins triggering mitophagy. Drp1/2: dynamin-related protein 1/2, Mff:mitochondrial fission factor, Mfn 1/2: mitofusin 1/2, OPA1: optic atrophy 1 protein. PINK1: PTEN-induced kinase 1, Parkin: E3 ubiquitin ligase Parkin.

In addition, mitophagy—a specific form of autophagy removing damaged mitochondria—plays a significant role in maintaining the health of the cells (110, 219). As mitochondria can accumulate damage with age (which is a factor in many renal diseases), injured mitochondria need to be removed from the cell. If fission (and/or fusion) is not sufficient anymore, mitophagy is used to eliminate damaged mitochondria and components. This mechanism becomes very important in mitochondria-rich cells in the kidney in pathological states. Accumulation of damaged mitochondria is undesired and can lead to detrimental consequences, from impairments of cellular metabolism to cell death. Mitophagy is induced by various pathways. Most widely known and studied are PINK1 and Parkin (101). PINK1 stands for PTEN-induced kinase 1, a molecule that has been implicated in mitochondrial quality monitoring. If the mitochondrial membrane is healthy and intact, PINK1 (which contains a mitochondrial targeting sequence) is imported through the outer membrane and in part, through the inner membrane. Through sequential cleavage, it is degraded. In a damaged mitochondrion, where the membrane is depolarized, PINK1 is not properly cleaved, and its concentration increases. PINK1 then is able to recruit Parkin (which is a cytosolic ubiquitin ligase) so that Parkin is able to ubiquitinylate proteins in the outer membrane, initiating mitophagy. Recently, it has been discovered that oxidation of cardiolipin, a phospholipid specific to the inner mitochondrial membrane, can also serve as a specific apoptotic signal for the removal of damaged mitochondrial and components (103). One of the next sections will discuss these studies briefly.

Mitochondrion as a Redox Organelle in the Kidney

Mitochondria participate in regulating multiple cellular functions, including but not limited to Ca²⁺ signaling, biomolecule synthesis, autophagy, redox balance, and homeostasis. The view of “mitochondrial ROS” was initially very simplistic, but as the available technology and experimental methods improved through decades, we now have a much better understanding of ROS-producing sites and redox reactions of the mitochondria.

Superoxide radical anion

The most prominent form of ROS from the mitochondrial ETC is the superoxide anion radical—O₂^•− (18, 35, 72, 116). Superoxide is formed because electrons escaping the ETC transfer to oxygen instead of the next electron acceptor in the ETC cascade. When the sites of superoxide formation were initially investigated, isolated mitochondria were often used, placing the organelle outside of the physiological milieu, or chemical blockers and inhibitors of one complex or the other were introduced, often artificially creating ROS or overestimating the amount of ROS emission (14). An elegant series of studies by Brand et al. as well as Van Remmen revealed that mitochondria in reality have a complex network of ROS production sites and modes (12, 13, 138, 150, 211). These include various subunits of complexes I and III, CoQ, and also proteins that are not part of the ETC such as the enzyme pyruvate dehydrogenase (PDH) or 2-oxoglutarate dehydrogenase. Complex I can produce superoxide during both forward (NADH-oxidizing) or reverse (NAD⁺-reducing) electron flow (151), presumably from two distinct sites (flavin moiety and Q-binding site). Respiratory complex II is also capable of significant superoxide production, oxidizing succinate to fumarate, when complexes I and III are inhibited and succinate concentrations are low (149), and thus may also be an important contributor of ROS in renal physiology and pathology.

The topology of the inner membrane and ROS production are also important to understand the potential reactions of superoxide. In principle, Complex I produces superoxide toward the matrix side, while complex III can produce equal rates of superoxide to the matrix and the intermembrane space (138) under certain conditions. A further issue is that ROS production will also depend on what substrate is being used for respiration, which renal cell is in question, what the oxygen tension is in the tissue, and so on. During forward electron transfer, with NAD-linked substrates (such as pyruvate), mitochondria in general produce high rates of ROS. If other physiologically relevant substrates are used, for example, fatty acids for renal proximal tubules (see more in the sections below), superoxide may be produced from sites that are less active during pyruvate oxidation. During reverse electron transfer, using succinate as substrate, complex I can produce superoxide again at high rates (158).

Finally, detection methods and in vitro approaches used in cultured renal cells can also complicate interpretation. First, the popular use of fluorescent probes comes with artifacts (105)—some of them like dichlorofluorescein diaceate (DCFDA/H2DCFDA) can produce ROS themselves (213, 214). Dihydroethidium (DHE) can be autoxidized (which is ~80% of the reaction) yielding only 20% of the measured fluorescence attributable to cellular ROS (106). Second, the use of inhibitors—just like in isolated mitochondria—such as rotenone, poses problems. Which specific sites of mitochondrial ROS production are most active in renal cells under physiological or disease conditions, in the absence of chemical inhibitors is unknown. The topic will require continued investigation to fully understand the extent and significance of mitochondrial ROS in renal health and disease.

The heterogeneity of renal cells—both in terms of function and mitochondrial density—then brings several interesting questions: will renal cells rich in mitochondria make more ROS when stressed? As different renal cells use different forms of metabolism, their ROS production may vary. In addition, various renal cells may have different defense mechanisms against the free radical attack and oxidative stress. Finally, what redox species can be produced in the kidney, and what redox signaling mechanisms are prominent? The primary form of ROS produced by mitochondrial respiration is the superoxide radical anion. With regards to the kidney, the overproduction of mitochondrial superoxide as a major form of oxidative stress in a plethora of kidney diseases has been postulated (20, 21, 42, 66, 176, 177). Certainly, mitochondrial defects due to pathologies can cause an increase in superoxide production which can further damage the ETC locally, cause aberrant electron transport, or affect mitochondrial fission, leading to leakage of mitochondrial DNA into the cytosol and activation of inflammatory pathways (37, 75, 206).

It is important to note that many cellular modulators of injury response or disease development are in the cytosol or in the nucleus. As a charged radical anion, superoxide is restricted in its capacity to have impacts on most biological structures outside of mitochondria. This occurs for the following three major reasons, serving as “barriers” (67-69). Self-dismutation of superoxide is a spontaneous process that occurs fast at k ~ 10⁶ M⁻¹ s⁻¹ (178). Even if produced in large amounts, superoxide is rapidly removed by mitochondrial SOD (superoxide dismutase) at a nearly diffusion-controlled rate (at a rate constant of k = ~10⁹ M⁻¹ s⁻¹) (127). This reaction is several orders of magnitude faster than either the production or self-dismutation of superoxide, meaning that mitochondrial superoxide is very short-lived, being converted to hydrogen peroxide (H₂O₂) very quickly. The third reason is its negative charge. Superoxide cannot cross membranes due to this charge. While it has been shown that a channel localized to the outer mitochondrial membrane, the voltage-dependent anion channel (VDAC), allows superoxide to exit from intermembrane space (127), evidence is lacking on how matrix-generated superoxide, for example, would get to the intermembrane space (133). In addition, it is worth mentioning that superoxide is a reductant and is not very efficient at oxidizing targets (22). Thus, the question should always be asked: is it reasonable to expect a large and sustained increase in mitochondrial superoxide levels in renal physiology and diseases? It is important to understand though that superoxide is a significant and primary initiator of further redox reactions. Without superoxide, there is no H₂O₂ emission, no hydroxyl radical formation through the Fenton reaction, and no initiation of further signaling or damage. Our proposition is to carefully consider each case in various kidney cell types and evaluate whether superoxide is indeed a major player in the signaling cascade, or a downstream reaction is important to lead to changes in cellular redox balance.

Hydrogen peroxide

Superoxide is dismutated in the mitochondria, and the resulting H₂O₂ is a better and stronger oxidant (95). It is able to diffuse farther from its originating source and acts as a secondary redox messenger molecule (6, 173). However, there are also multiple lines of defense against excess H₂O₂. It is removed in the cytosol by catalase—an enzyme that is nearly omnipresent in all tissues; the reaction yields water and oxygen. Mitochondria, however, do not contain catalase; defenses against H₂O₂ are peroxiredoxins, thioredoxin, and glutathione peroxidase 1 (43, 139) (Figure 3). The peroxiredoxin family (Prdx) consists of six isoforms (Prdx 1 to 6), also relatively abundant in many cells (155). The mitochondrial forms are Prdx 3 and 5. Unlike catalase, the active site of peroxiredoxin enzymes contains a redox-sensitive cysteine residue. When reacting with H₂O₂, the cysteine undergoes oxidation to sulfenic acid. The oxidized peroxiredoxin then is converted back to an active (reduced) form by thioredoxins (212).

Figure 3. Mitochondrial defenses against oxidative stress.

Mitochondrial superoxide dismutase (SOD2) is a major defense against mitochondrial superoxide, as it catalyzes the dismutation of superoxide to hydrogen peroxide (H₂O₂) at an extremely fast rate. Mitochondria do not contain catalase, thus hydrogen peroxide is eliminated by peroxiredoxins (Prdx1), thioredoxins (Trx), and glutathione peroxidases (Gpx 1). Hydrogen peroxide can also participate in the Fenton reaction with iron available from damaged iron-sulfur complexes, resulting in hydroxyl radical (OH) production. Mitochondria have an ideal membrane-rich environment conducive for Fenton-reaction-induced lipid peroxidation. Phospholipid peroxides (LOOH) are membrane diffusible and can only be eliminated by glutathione peroxidase 4 (GPx4). The brackets explain the glutathione peroxidase 1 and peroxiredoxin/thioredoxin system and their regeneration in detail. Other abbreviations are GSH: glutathione, reduced form, GSSG: glutathione, the oxidized form, Cat: catalase, NADP+/NADPH: nicotinamide adenine dinucleotide phosphate/reduced form.

Glutathione peroxidase 1 is a member of the glutathione peroxidase family (eight isoforms), which are selenoproteins, present in the cytosol as well as in the mitochondrion (16). Besides H₂O₂ removal, Gpx1 detoxifies organic hydroperoxides as well, playing an important role in mitochondrial ROS defense. The activity of these enzymes is tightly connected to cellular metabolism via cofactors: NAD(P)H and glutathione (GSH). If the metabolism of the cell changes (in hypertensive insult, diabetes, metabolic disease), the availability of these cofactors can change. The result would be insufficient removal of H₂O₂.

Lipid peroxides

Excess H₂O₂ readily reacts with transition metals like iron (Fe²⁺) in the mitochondria (99, 204). This reaction (known as a Fenton reaction) produces the hydroxyl radical, which is one of the best initiators of lipid peroxidation (82, 109) The close proximity of phospholipid-rich structures of the inner mitochondrial membrane adjacent to redox-active transition metal centers (such as Fe-S clusters) in the ETC complex proteins is a natural and ideal place for lipid peroxide formation. Lipid peroxide generation is a chain reaction, where propagation gives the necessary amplification and distance for these reactive species to reach targets far from the originating mitochondria, being ideal examples of redox messengers in renal cells.

What other advantages do lipid peroxides have compared to ROS such as superoxide or H₂O₂? In contrast to superoxide or H₂O₂, some of the specific oxidized (phospho)lipids also have another, distinct biochemical advantage: they have an affinity to react with amino acid residues (such as for example, cysteine) (86, 122). Thus, their target proteins can be rather specific. Some of the oxidized phospholipids, for example, cardiolipin, which is unique to the mitochondrial inner membrane, can be externalized by the cell, thus serving as a “signal” for damaged membranes to become autophagy/mitophagy targets (103). The basic tenets of lipid peroxidation reactions are well-known in biology (83). Much less is known however about specific and unique oxidized phospholipid molecules as redox signaling molecules in the normal and hypertensive kidney (or for that matter in chronic kidney diseases in general). How these unique molecules modulate biological responses in renal cells is an area for potential future discoveries and novel interventions. The only enzyme capable of protecting against and removing phospholipid peroxides in membranes (besides organic peroxides and H₂O₂) is glutathione peroxidase 4 (GPx4) (Figure 3). It is the only isoform of the GPx superfamily that is indispensable for life (154, 218). GPx4 resides in the cytosol, nucleus, and mitochondria. Still, importantly, the cytosolic form is the one that enters the mitochondria of somatic cells (mitochondrial GPx4 only confers specific functions in spermatogenesis) and is the form that is crucial to protect cells against apoptosis and ferroptosis under oxidative stress (124, 169). The cytosolic form of GPx4 was shown to be highly enriched in the intermembrane space, implying that this form is extremely important to protect the mitochondrial membrane from peroxidation (128).

Nitric oxide

Finally, we do have to mention nitric oxide (NO), a molecule that is important in blood pressure regulation, and thus for hypertensive kidney damage as well. NO is not of mitochondrial origin, but the molecule and its derivatives can impact mitochondria in a number of ways. NO is a gasotransmitter free radical; the discovery that NO is the endothelium-derived relaxing factor (EDRF) gave Robert Furchgott, Louis Ignarro, and Ferid Murad the prestigious Nobel prize in physiology and medicine in 1998 (23, 63, 74). In the cytosol, NO is produced from the substrate l-arginine by nitric oxide synthases (NOS). The enzymes convert l-arginine to citrulline, while catalyzing the formation of NO, which then is diffusible and activates soluble guanylate cyclase (sGC), resulting in cyclic GMP formation from GTP and vasorelaxation. The three major isoforms of the NOS family are: endothelial (eNOS or NOS3), neuronal (nNOS or NOS1), and inducible (iNOS or NOS2) (185). While the first two are constitutive enzymes, the third one is induced by various stressors (typically inflammatory reactions). Posttranslational factors, such as cofactor tetrahydrobiopterin (BH₄) availability and phosphorylation level also play a role in the concerted regulation of eNOS (70). Impaired NO production has been shown to play a role in the pathogenesis of hypertension, preeclampsia, obesity, diabetes, and diabetic complications such as retinopathy or nephropathy, migraine, and erectile dysfunction (5, 26, 79, 85, 115, 145, 174, 175, 225). Importantly, while nNOS initially has been discovered in neurons and brain, more recent studies have shown that that it also plays a key role in vascular physiology and diseases (104, 184).

Inducible NOS can produce large amounts of NO upon stress, such as acute or chronic inflammation, NFkB activation, or similar conditions (38). If the NOS enzyme is uncoupled, for example, due to BH₄ deficiency (202), it can also be a significant source of superoxide besides NO, opening the possibility to a fast, diffusion-controlled reaction of peroxynitrite (ONOO⁻) formation. ONOO⁻ is a potent molecule that can damage protein residues or change their function by nitration (97). Still, it cannot get too far from its production site due to its reactivity and (depending on the pH) fast protonation. Large amounts of NO or ONOO⁻ can also enhance inflammatory tissue damage, making NO a molecule which can bring benefits or cause further harm depending on the modes and sites of action (96, 144). Both NO and its derived species have been shown to affect mitochondria. High levels of NO inhibit mitochondrial respiration through binding to cytochrome c oxidase (19). NO can also react with oxygen and form ONOO⁻, or react with ubiquinol (170). Importantly, however, smaller amounts of NO can stimulate biogenesis in various cell types, likely through impacting cGMP and transcription factors (140).

Mitochondria in Renal Cells

The kidney is one of the most heterogeneous organs in the body, both in terms of the cell types and their function. The first single-cell sequencing studies described at least a dozen different cell types in mouse or human kidneys (36, 148, 210). Different nephron segments have distinct functions such as reabsorption, filtration, electrolyte balance, or maintaining the glomerular filtration barrier. Every one of these functions has a different energy need, from very high to low. To match and meet this energy need, each type of cell prefers a form of metabolism, with more or less reliance on mitochondria.

The kidneys receive about 20% of the cardiac output at rest. Oxygenation in kidney tissues is determined by arterial blood oxygen concentrations, how much oxygen is consumed by each cell, and arterial-venous shunting (181). Most of the oxygen consumption is used for sodium transport, and the remaining amount is utilized for metabolism and similar cell function activities. However, oxygenation and perfusion across the kidney are very heterogeneous. The medullary region is relatively hypoxic compared to the cortex—starting with a partial pressure of approximately 50mmHg and gradually decreasing to approximately 10 to 15mmHg in the inner medulla (193). Furthermore, the outer medullary region has high metabolic demands, in spite of relatively low oxygen concentrations. Thus, it is important to discuss each nephron segment, their metabolism and metabolic demand, oxygenation, and mitochondrial density (Figure 4).

Figure 4. Oxygenation and partial pressure in the kidney.

While the kidney cortex has a partial oxygen pressure (pO₂) of around 60 to 30 mmHg, this pressure drops significantly toward the outer and inner medullary regions, placing some of the highly metabolic, mitochondria-rich segments of the nephron in hypoxic conditions.

Glomerulus and glomerular filtration barrier

Located at the beginning of a nephron, glomeruli and their surrounding Bowman’s capsule constitute a basic filtration unit of the kidney. The glomeruli are a network of small capillaries (tuft) receiving blood supply from the afferent arteriole. The structure of a glomerulus is very unique, with fenestrated endothelium, glomerular basement membrane (GBM), specialized cells called podocytes and their foot processes, and the mesangium. The filtrate passes through the GBM and between podocyte foot processes to enter the capsule as ultrafiltrate and leaving toward the proximal tubule. Podocytes, which are essential and highly specialized cells of the glomerular filtration barrier, are not particularly mitochondria-dense. Not only are these cells more glycolytic, but they can also utilize anaerobic glycolysis to maintain the energy needs of the filtration barrier (17). Anaerobic glycolysis is a form of metabolism where small amounts of ATP are generated very rapidly, to match the energy demand. The idea of podocytes switching to this form of metabolism has been supported through evidence obtained in vitro and in vivo in mouse models (17, 123). Of course, just because they have fewer mitochondria than other renal cells, it does not necessarily mean that dysfunction and damage to mitochondria do not impact podocytes. As we mentioned before, mitochondria have a complex role and many functions in cellular homeostasis, which are not limited to energy production only, but extend to signaling, precursor biosynthesis, and much more. Indeed, there is a growing body of literature discussing the effects of altered mitochondrial function in podocytes. The metabolic need of these cells may also change depending on the injury or glomerular disease, diabetic or hypertensive renal damage present. Several studies discuss the tight regulation of Pgc1a (mitochondrial biogenesis), Kruppel-like factor 6 (Klf6), Drp1, and Rock1 (mitochondrial fission) in podocytes and their importance for maintenance of podocyte health (81, 89, 206). Changes in NO bioavailability have been shown to impact podocytes, partially through formation of ONOO⁻ (175), which then can lead to a decline in mitochondrial function, oxidative and nitrosative stress, ultimately causing proteinuria, and chronic kidney disease.

In general, less is known about glomerular endothelial cells (GEC) compared to podocytes. The molecular mechanisms of interaction between glomerular cell types during pathogenesis of renal damage and mitochondrial function in glomerular endothelial cells remain poorly understood, but the field is developing rapidly. It has been shown that in humans with either type 1 or type 2 diabetes, glomerular endothelial dysfunction represents an initiating step in albuminuria (166). Consistent with these studies, mitochondrial dysfunction in glomerular endothelial cells has also been shown to be essential to trigger podocyte loss in a diabetes-susceptible mouse model (28, 46).

Proximal tubules

Proximal tubular epithelial cells are highly energy demanding in the kidney (9, 60). These cells have more active transport-related mechanisms compared to other renal cells, as they reabsorb the largest amount of filtrate. Thus, PTCs contain copious amounts of mitochondria, almost similar to the levels found in white muscle or active cardiomyocytes (142). It was none other than Hans Adolf Krebs—who we know because of his discoveries about tricarboxylic acid metabolism named after him as Krebs cycle—who was among the first ones to study and describe the metabolism of PTC isolated from rat kidneys in 1969 (208). Mitochondrial fatty acid oxidation is the primary form of metabolism of PTC, which provides the large amounts of ATP necessary for the high energy needs of these cells. In addition, PTCs also have a high capacity for gluconeogenesis, compared to other nephron segments where gluconeogenesis is almost completely absent (77). It is, therefore, logical to suggest that derailments in mitochondrial function and impairments in fatty acid oxidation will take a toll on PTC in metabolic diseases or hypertension (108, 114, 130, 132, 167). Stimulating mitochondrial biogenesis has also been shown to be beneficial in protecting the PTC from injury, especially in—but not limited to—acute injury (27, 39, 40).

Loop of Henle

The loop of Henle is a heterogeneous nephron segment with a “hairpin” configuration. The primary site of sodium reabsorption in the LH is the thick ascending limb (TAL). The medullary portion in the last part of the LH has one of the highest content of mitochondria of all mammalian cells, as their need for oxidative metabolism and to sustain high ATP production for active transport is high (25). However, this high metabolic need is coupled with relatively low oxygenation (Figure 4), making this nephron segment susceptible to the consequences of mitochondrial dysfunction, as it has been shown for example in sepsis-related AKI (147). With relation to blood pressure, salt-sensitive rats for example show elevated reabsorption in the TAL, which may contribute to their impaired natriuresis (162). Renal medullary blood flow is also decreased in these rats, just within a few days of starting the high salt diet (136). Upon feeding a high salt diet to salt-sensitive rats, mitochondria seem to shorten in the TAL region, which is consistent with compromised mitochondria function. Interestingly, cells in the medullary region in general also have more efficient mitochondria when compared to cortex (168). Mitochondria in the TAL seem to have adopted a strategy to have sustained ATP production under nonideal conditions (decreasing levels of blood flow, lower tissue oxygen, and partial pressure), by having high oxygen affinity and a high degree of coupling.

Collecting duct

The collecting duct system in the kidney consists of connecting tubules and collecting ducts. The connecting tubules join the distal convoluted tubules (DCT) to cortical collecting ducts. The outer segment of the medullary collecting duct (OMCD) follows the cortical collecting duct and then descends deeper into the medulla forming the inner medullary collecting duct (IMCD). The collecting ducts have two main cell types: principal and intercalated cells. While principal cells are numerous, and sodium reabsorption occurs through these cells, they are sparse in organelles including mitochondria. Intercalated cells on the other hand have numerous mitochondria (78, 163, 172). These cells participate in acid-based homeostasis. Metabolism in the DCT is also heavily dependent on mitochondrial oxidation of fatty acids but glycolysis is present to some extent as well, while the DCT not having much potential for gluconeogenesis (80).

In summary, mitochondrial density varies in kidney cells and seems to be directly correlated with the ability to utilize oxidative metabolism for energy production. Some of the highly specialized cells in the kidney can use other forms of metabolism (glycolysis) as well, but that of course does not mean that mitochondrial injury and dysfunction do not impact these cells at all.

Renal Blood Pressure Control and Mitochondrial Bioenergetics and Metabolism

Chronic control of blood pressure relies on a variety of physiological factors and processes which work together to orchestrate the short-term and long-term maintenance of the blood pressure setpoint (15). To a great extent, our blood pressure depends on the ability to maintain sodium balance, which is controlled by the kidney; patients with chronic kidney disease (CKD) are vulnerable to even moderate elevations of blood pressure (11). The signaling pathways implicated in the development of hypertensive renal damage are far from being comprehensively characterized. In the recent decades, thanks to the development of tools that allow for precise characterization of the gene, protein, and metabolite networks, as well as related functional pathways in single cells and tissues, we were able to expand our understanding of the new dimension of hypertension, which relies on metabolic and bioenergetic abnormalities. Hypertension has been proposed to be a mitochondrial and metabolic disease (57, 193); such perspective on hypertension has opened up new avenues of research and potential for therapeutic developments such as repurposing of existing drugs. As summarized in the introductory sections above, renal tissue has a lot of heterogeneity in terms of its energy demands, substrate reliance, mitochondrial density, and metabolic processes.

Renal mitochondrial structure, dynamics and quality control in hypertension

There are multiple reports on the contribution of mitochondrial dynamics and biogenesis in acute kidney injury and diabetic kidney disease (8, 200). Unfortunately, the information regarding such changes in hypertensive renal damage is scarce, although some instances of changes in mitochondrial structure and dynamics have been documented in hypertensive cardiac or vascular damage (56, 159, 171). Indirect evidence of renal mitochondrial structural damage in human hypertension can be devised from an elegant study by Lerman group, which demonstrated a correlation between serum creatinine levels and estimated glomerular filtration rate with the increase in urinary mitochondrial DNA, implicating mitochondrial injury in kidney damage in human hypertension (59). The same group demonstrated that mitochondrial biogenesis is impaired in the kidney of hypertensive pigs (evaluated by renal expression of PGC-1α, nuclear respiratory factor-1, GA-binding protein, peroxisome proliferator-activated receptor (PPAR)-α, PPAR-δ, heme oxygenase-1, and sirtuin-1) (58). In the Goldblatt’s 2 kidney 1 clip rat model of hypertension, dysfunctional renal mitochondrial biogenesis, and quality control were reflected by decreased Parkin and PGC1-α levels (61). In recent studies, it was shown that mitophagy/biogenesis may serve as a mechanism counteracting the oxidative stress in high-salt-induced renal and vascular injury in stroke-prone spontaneously hypertensive rats in a cascade involving mitochondrial uncoupling protein 2 (UCP2) (51, 71, 164). A thorough study by He et al. was the first one to address the ultrastructure of mitochondria in different nephron segments in salt-sensitive hypertension (84). The authors convincingly showed that there are ultrastructural abnormalities in the medullary thick ascending limbs and proximal convoluted tubules of Dahl Salt-Sensitive (SS) rats prior to the development of histological injury, which allows to posit that mitochondrial damage is a significant contributor to initiation and development of hypertensive renal damage in this case. Overall, changes in mitochondrial biogenesis, a shift from fusion to fission, or upregulation of mitophagy are important factors that may compromise energy production and overall renal cell function in hypertension. The evidence that has been accumulated supports the role of these mitochondrial processes in hypertensive renal damage and prompts further careful investigation in different models of hypertension, with an emphasis on different nephron segments.

Renal mitochondrial metabolism in hypertension

In 1941, Rodbard and Katz, in a manuscript devoted to the role of renal metabolism in hypertension, postulated that “although the identity of the chemical mediator of renal hypertension has not as yet been established, the development of [hypertensive animal models] has stimulated studies which may lead to its identification and isolation” (160). After many decades of search for the forever elusive “chemical mediator of hypertension,” it has become clear that it is naïve to assume that there is just one. Complex metabolic disturbances are strong determinators of kidney injury in essential hypertension, forming “syndrome X” or “metabolic syndrome”; carbohydrate, lipid, and purine metabolism disturbances enhance kidney injury (199). Kincaid-Smith formulated a hypothesis in 2004 that hypertensive renal damage is linked more closely to metabolic alterations such as insulin resistance and obesity than to blood pressure (111). It is becoming increasingly clear that metabolic abnormalities are essential contributors to the development and progression of kidney damage in hypertension (Figure 5). Yet, we are just starting to understand the complex interconnected signaling and metabolic networks involved in this process. Liang has recently emphasized that we have made several key advances in our understanding of the role of energy and substrate metabolism in hypertension (193), however, the intricacies of these processes are yet to be fully defined.

Figure 5. A brief overview of the mitochondrial impact in the kidney during hypertension.

Hypertension can impact mitochondria (and vice versa) in renal cells in various ways, including, but not limited to, imbalance in redox homeostasis and oxidative stress, impaired mitochondrial biogenesis, defective mitochondrial fusion/fission, derailed substrate metabolism, impairments in Ca²⁺ homeostasis and mitophagy/autophagy.

A brief introduction to mitochondrial metabolism

While it is beyond the scope of this article to delve into all the details of mitochondrial metabolism (there are other excellent reviews in the literature about each nephron segment (77, 193)), we must briefly introduce how the mitochondria get their energy sources before we dive into the intricacies of renal mitochondrial metabolism. All three major forms of metabolism—carbohydrate- and fatty acid oxidation as well as amino acid metabolism—can be used by mitochondria (Figure 6). All three major metabolic pathways culminate in acetyl-CoA (AcCoA) formation, providing the basic “building block” for biosynthesis and energy production. As AcCoA is not membrane permeable, excess can be shuttled out of the mitochondria through linkage to carnitine. Carnitine acetyl-transferase (CrAT) is catalyzing this reaction, with preference to AcCoA and short-chain acyl-CoAs.

Figure 6. Brief summary of mitochondrial energy metabolism.

Mitochondria can oxidize substrates from all three major forms of metabolism: carbohydrate-, fatty acid-, or amino acid metabolism. Ultimately, the electrons harnessed from these processes are used for energy production, through passing onto cofactors NADH and FADH₂. All of these pathways are relevant in mitochondrial redox homeostasis as well. ATP: adenosine triphosphate, αKG: alpha-ketoglutarate, Asp: aspartate, Ala: alanine, Cpt1: carnitine palmitoyl-transferase 1, CrAT: carnitine acetyl-transferase, carn: carnitine, ETC: electron transport chain, FAO: fatty acid β-oxidation, Gls: glutamine synthetase, Got2: mitochondrial aspartate aminotransferase, Gpt2: mitochondrial alanine aminotransferase, GSH: glutathione, LDH: lactate dehydrogenase, PDH: pyruvate dehydrogenase, Sdhd: succinate dehydrogenase (subunit d), Slc1a5: neutral amino acid transporter, TCA: tricarboxylic acid cycle.

As the end-product of carbohydrate metabolism, pyruvate feeds through the pyruvate dehydrogenase complex into the mitochondrion where it is oxidized, fueling the TCA cycle as well as oxidative phosphorylation. Fatty acids enter the mitochondrial β-oxidation. The enzyme catalyzing mitochondrial entry of long-chain FA is carnitine-palmitoyl transferase 1 (Cpt1). Beta-oxidation is a complex process involving several enzymes, where each step of two-carbon removal provides one AcCoA molecule and a shorter fatty acyl CoA. The process continues to convert all the carbons to AcCoA. Beta oxidation is highly energetic and can produce large amounts of ATP (depending on the carbon chain length of the starter fatty acid substrate) through NADH, FADH₂, and a full turn of the TCA cycle.

Mitochondrial amino acid metabolism is an equally complex system where 17 amino acids have mitochondria-associated metabolic pathways (183). From the redox homeostasis viewpoint, we only show the example of glutamate/glutamine in Figure 6. Glutamine is transported into the mitochondria through a neutral amino acid carrier protein (Slc1a5). Glutamate—which can be produced from the TCA cycle—can also be converted to glutamine with the help of glutamine synthetase. Glutamate is essential for glutathione synthesis (GSH) and as glutathione is part of the major cytosolic redox defense system, derailments in glutamine/glutamate metabolism can take a toll on cellular redox balance.

Renal TCA cycle-related metabolism in hypertension

Mitochondrial dysfunction and altered renal mitochondrial metabolism have been reported in different models of hypertension, with the majority of the studies done in the Dahl SS and the SHR (spontaneously hypertensive) rats. For example, in the Dahl SS rat (a model that closely mimics human SS hypertension) Tian et al. discovered a novel role for renal fumarate metabolism (192, 194). Global metabolic profiling of plasma of these rats compared to the SS.13BN controls revealed higher levels of TCA cycle metabolites (fumarate, cis-aconitate, isocitrate, citrate, and succinate) as well as pyruvate (205). Wang et al. demonstrated that renal mitochondria of SS rats exhibit complicated metabolic alterations compared to control strain even when the animals were maintained a low salt diet, while high salt feeding further aggravates the observed dysfunctions (207). Zheng et al. reported that high salt diet contributes to hypertension by weakening the medullary TCA cycle and overall antioxidant system in these rats (222). Interestingly, these changes were also related to fumarase (the enzyme which catalyzes the interconversion of fumarate and l-malate). Lower activity levels of fumarase, isocitrate dehydrogenase, and succinyl-CoA synthetase were observed in renal mitochondria of SS rats compared with SS.13BN controls on low salt diet, while supplementation with fumarate increased blood pressure and H₂O₂ production in the SS rats, likely due to the imbalance of glutathione metabolism (221). On the other hand, supplementation of aspartate or malate increased renal levels of l-arginine and NO and attenuated hypertension in SS rats (90); in l-NAME-treated rats, malate reduced blood pressure and reduced GFR (55). Furthermore, it was shown that fumarase overexpression was able to abolish hypertension in Dahl SS rats, which was attributed to the ability of this enzyme to restore NO production (201, 216). In a recent study, when BAIBA (β-aminoisobutyric acid) was administered to SS rats, it exhibited an antihypertensive effect, accompanied with prevention of fumarase’s activity reduction caused by high salt, and an increase in NO content (223). Overall, this series of studies shed light on the TCA cycle dysfunction in SS hypertension; clearly, there are interconnected multi-step TCA cycle-related metabolic pathways which might be pharmacologically targeted in future to tip the scales of hypertension, when more details are revealed.

Renal mitochondrial amino acid metabolism in hypertension

Mitochondria are capable of modulating amino acid homeostasis according to the particular requirements and energy needs of a cell, and renal mitochondrial amino acid metabolism has garnered a lot of attention. Observational studies reported associations of circulating amino acids with cardiovascular diseases, including hypertension (65). Lin et al. provided first evidence for causal association of circulating amino acids with blood pressure in a Mendelian randomization study (125). The authors demonstrated that the increased levels of BCAAs are causally associated with hypertension, while glycine levels predicted a lower risk of high blood pressure. Findings from Teymoori et al. also revealed that the dietary amino acid pattern rich in branched chain, aromatic, and alcoholic amino acids, as well as proline could increase the risk of hypertension (191). In cross-sectional hypertension-metabolomics analysis that integrated data from nine intercontinental cohorts from the COnsortium of METabolomics Studies, Louca et al. confirmed in pathway analysis that amino-acids, serine/glycine, and bile acids are implicated in hypertension regulation (126). Øvrehus et al. demonstrated that in the biopsies from patients with early hypertensive nephrosclerosis (compared to healthy controls), there are distribution differences in 11 amino acids (143). In mechanistic studies on animal models, Mattson’s group has convincingly shown in a series of manuscripts that dietary protein source affects blood pressure development and renal damage in Dahl SS rats (1-3, 47, 49, 129). Rinschen et al. have recently offered a concept of “metabolic rewiring” of the hypertensive kidney in their multi-layer omics study aimed at deciphering the metabolic aspects of hypertension-induced glomerulosclerosis (157). Using the Dahl SS rat, they demonstrated that at early time points of hypertension in glomeruli there is activation of anaplerotic pathways to regenerate energy (likely used to counteract metabolic stress), indicated by depletion of branched-chain amino acids (BCAA, such as leucine, isoleucine, and valine) and proline. Taken together, these emerging data are opening new and plausibly diet-based interventions for prevention of hypertension.

Perhaps, the most well-known amino acid for its implications in mediation of hypertensive renal damage is l-arginine, due to its role as a precursor of NO formation (87). In the hypertensive Dahl SS rats compared to SS.13BN control strain, renal tissue levels of l-arginine and NO were reported to be lower (90). The activity of different NOS types was shown to be lower in the SS rats fed a high salt diet compared with normotensive control strains (94, 190). In SHR rats, perinatal administration of l-arginine combined with antioxidants was able to alleviate adult blood pressure (50, 152). l-arginine supplementation was also beneficial for blood pressure reduction, renal function, and NO bioavailability in Dahl SS rats (32, 33, 73, 136, 137, 187). Importantly, despite potential advantages, l-arginine supplementation may augment age-related renal functional decline and mortality, particularly in females (93), suggesting that long-term dietary l-arginine supplementation should potentially be avoided in the elderly population.

There are reports regarding the importance of amino acids in hypertensive renal damage. Tryptophan metabolism has been widely suggested as a target to promote longevity (30); tryptophan is used in the kynurenine metabolic pathway, which is the sole de novo NAD⁺ biosynthetic pathway that generates this molecule from ingested tryptophan. Tryptophan-related pathways were suggested to be used in reprogramming strategies to prevent hypertensive renal damage (92). Hsu et al. reported that maternal tryptophan supplementation protects offspring against hypertension, which was linked to NO bioavailability (91). In the early 90s, it was shown that chronic dietary tryptophan administration prevented salt-induced hypertension in the Dahl SS rat, however, since urinary sodium was not affected by this treatment, the authors concluded that the protective effect was not at the level of the kidney (117, 118). Essential amino acid lysine has also been recently implicated in renoprotective mechanisms; in the Dahl SS rat, lysine administration diminished development of hypertension and kidney injury and induced diuresis (156). The authors also demonstrated that lysine administration effectively inhibited tubular albumin uptake and reduced TCA cycle metabolites in patients at risk for hypertension and kidney disease versus kidney-healthy volunteers.

Therefore, selective amino acids supplementation might prove a novel and interesting potential treatment strategy for hypertension. In summary, there is significant interest and evidence regarding the mechanistic relationship between the amino acid metabolism and hypertension. The studies focusing on the interplay among dietary protein, renal mitochondrial function, and hypertension may provide insight into how the potential dietary interventions (via effects on mitochondrial amino acid metabolism) could serve as new therapeutic strategies for hypertension.

Renal mitochondrial bioenergetics and mitochondria-derived oxidative stress in hypertension

Although we are mainly focusing on mitochondrial ROS sources here, it should be mentioned there are other key nonmitochondrial sources of ROS in the kidneys, such as NADPH oxidases (NOX) (196); the elevated expression and activity of Nox are clearly shown in hypertension (119). Notably, there is more and more evidence regarding crosstalk between mitochondria and NADPH oxidases in the pathophysiology of hypertension; mitochondrial ROS may stimulate NADPH oxidases, feeding a vicious cycle of ROS production and cell damage, which can be pharmacologically targeted through inhibiting ROS production by mitochondria and reducing NADPH oxidase activity (52). Overall, the relative contribution of the mitochondrial and nonmitochondrial (NADPH oxidases, xanthine oxidase, endoplasmic reticular oxidases, as well as uncoupled NOS) renal ROS sources to the development of hypertension is not well-established and requires further careful studies.

When renal oxygen utilization is considered in hypertension, it has been established that oxygen utilization efficiency is reduced in the SHR rat versus WKY control, indicative of less efficient O₂ usage for Na⁺ transport in the hypertensive strain (209). Furthermore, renal NO is decreased in SHR rat, resulting in increased oxygen consumption (4); this can result from potential activation of mitochondrial enzymes, including ETC complexes and cytochrome oxidase, which can occur when NO levels are low (215). Lee et al. further studied the contribution of mitochondrial oxidative phosphorylation to the hypertensive phenotype in the SHR rat versus WKY; they demonstrated that it is higher in renal proximal tubules isolated from SHR rat, and the PHD complex is a determinant of increased mitochondrial metabolism and a key contributor to hypertensive phenotype in this strain (120). Interestingly, in SS hypertension oxygen consumption rate was lower than the IMCD isolated from Dahl SS rat fed a high salt diet versus SS.BN13 controls (220). We have similarly recently showed decreased oxygen consumption and higher oxidative stress in the glomeruli isolated from Dahl SS rats fed a high salt diet compared to their normal salt diet fed counterparts (53). The renal oxidative stress that accompanies hypertension development in the Dahl SS rats may be resulting from lower levels of antioxidant enzymes such as SOD and catalase, which may contribute to accelerated renal cell senescence (62, 161, 189). Furthermore, there is a possibility that uncoupling of the oxidative phosphorylation is involved; De Miguel et al. have recently revealed that overexpression of an uncoupling protein UCP2 in the kidney is able to increase blood pressure in mice (48). In SHR rats, it was suggested that UCP2 has a protective role, which is impaired in hypertension (51). UCP2 gene silencing in renal mesangial cells leads to increased ROS generation, inflammation, apoptosis, and necrosis; high-salt diet downregulated the antioxidant UCP2-dependent mechanism in kidneys of stroke-prone, but not stroke-resistant SHR (51). Overall, there is little known regarding the role of uncoupling proteins in hypertension in the renal epithelial cells, and studying the metabolic and adaptive ramifications of targeted mitochondrial uncoupling on renal function in hypertension is an exciting and novel avenue of research.

Inhibition of ROS production or genetic deletion of the components of the ROS cascades often attenuates hypertension and improves kidney function. However, the actual mitochondrial ROS-producing sites and redox reactions are not well-studied in hypertensive renal damage. One of the key questions in the mitochondrial redox-focused studies in hypertension is the maintenance of the delicate balance of pro- and antioxidant systems in the development of kidney disease. Novel approaches are required to target the renal mitochondrial oxidative stress pathways in a way that gently tips this balance but does not completely derail the endogenous antioxidant defenses and depart from the previously singular concept of ROS versus antioxidants balance.

Conclusions and Perspectives

In 1960 Bartels and Phelps stated that “the most encouraging progress … recently has been in the elucidation of certain metabolic changes in the kidney and their role as specific causes of hypertension” (10). This statement curiously remains true to this day; thanks to the significant advances that have been made in recent years with the use of the “omics technologies,” we are starting to unravel the intricate metabolic pathways that are affected in hypertension. It has become clear that metabolic disturbances are important for the development of renal damage. Some studies have provided evidence that mitochondrial structural and metabolic abnormalities occur prior to tissue damage or renal dysfunction, which inspires cautious optimism regarding targeting renal mitochondria for therapeutic applications. Despite these advances, there is a significant need for detailed temporal resolution of the mitochondrial metabolic processes during the initiation and development of hypertension rather than just obtaining their “snapshots” at certain time points. Notably, when renal metabolism is considered in time, it should be aligned with circadian rhythms of blood pressure. Furthermore, among the least studied topics related to mitochondrial function in hypertension is mitochondrial biogenesis and dynamics. Since mitochondria are very quickly adapting organelles, these processes should be studied in time similarly to metabolic changes. Emerging approaches to in vivo imaging of mitochondria in renal tissues (24) would be of high importance for such studies.

In order to tailor therapies to particular types and locations of damage (proximal tubular, glomerular, distal nephron, or roughly, cortical vs medullary), we require detailed knowledge about mitochondrial function in each segment. Emerging technology, such as Desorption Electrospray Ionization mass spectrometry, MALDI imaging, and spatial transcriptomics can provide region-specific resolution for in-depth insight regarding the location and molecular entity of certain metabolites (131). Development and application of high-resolution (single-cell) transcriptomic, proteomic, and metabolomic approaches combined with high-resolution imaging and targeted delivery of compounds to cell type of interest will undoubtedly move the field forward.

Although some research has been devoted to mitochondria as central targets for sex differences in pathology (203), very limited knowledge has been accumulated regarding the implications of sex differences in renal mitochondrial metabolism and function in hypertension. Given the crucial role of sex hormones in the regulation of mitochondrial bioenergetics (186), it is important that renal mitochondrial metabolism is studied in the context of renal sex differences, and disease susceptibility in males and females, and during gender-affirming hormonal treatments.

Lastly, it is imperative to go a step beyond the umbrella terms and better define the meaning of “ROS” and “oxidative stress” in hypertensive kidney damage. Reactive species are such diverse molecules in terms of reactivity, kinetics, and behavior that they cannot simply be covered by a singular concept. Identifying their molecular entity and their source may take us closer to both understanding reaction mechanisms and discovering specific, targeted interventions. Last but not least, the prevalent oxidative-stress-focused theories of the pathogenesis of hypertension should be expanded to make sure that the mitochondrion is properly incorporated in the understanding of this complex disease. Research into mitochondrial structure, dynamics, and overall bioenergetic and metabolic processes will reveal new druggable targets, which will allow us to expand current therapeutic options, and hopefully tip the scales of the treatment-resistant hypertension.

Didactic Synopsis.

Major teaching points

• Abnormalities in mitochondrial function and bioenergetics affect the development of chronic kidney disease and hypertension-associated kidney injury.

• Mitochondrial redox biology and energy production are complex biochemical processes.

• When considering mitochondrial function, it is imperative to define “reactive oxygen species” (ROS) and “oxidative stress.” ROS are such diverse molecules in terms of reactivity, kinetics, and behavior that they cannot simply be covered by a singular concept.

• Focusing on the exact production sites and the molecular entities that form reactive species (superoxide anion, hydrogen peroxide, and others) is important to understand the role they play in hypertensive kidney disease.

• All three major forms of metabolism—carbohydrate, fatty acid oxidation, and amino acid based—can be used by mitochondria in the kidney tissue.

• The kidney is a heterogeneous organ in terms of its energy demands, substrate reliance, mitochondrial density, and metabolism.

• The use of the metabolic pathways depends on a specific nephron segment. In order to tailor therapies to particular types and locations of damage (proximal tubular, glomerular, distal nephron, or, roughly, cortical vs medullary), detailed knowledge about mitochondrial function in each segment is required. Focusing on how mitochondrial metabolism evolved along the nephron is crucial for the development of targeted therapies and comprehensive understanding of the underlying metabolic processes.

• Reactive oxygen species production, TCA cycle, and amino acid metabolism are key mitochondrial pathways that have been identified to be affected in hypertensive kidney damage.