Understanding coenzyme Q

Abstract

Coenzyme Q (CoQ), also known as ubiquinone, comprises a benzoquinone head group and a long isoprenoid side chain. It is thus extremely hydrophobic and resides in membranes. It is best known for its complex function as an electron transporter in the mitochondrial electron transport chain (ETC) but is also required for several other crucial cellular processes. In fact, CoQ appears to be central to the entire redox balance of the cell. Remarkably, its structure and therefore its properties have not changed from bacteria to vertebrates. In metazoans, it is synthesized in all cells and is found in most, and maybe all, biological membranes. CoQ is also known as a nutritional supplement, mostly because of its involvement with antioxidant defenses. However, whether there is any health benefit from oral consumption of CoQ is not well established. Here we review the function of CoQ as a redox-active molecule in the ETC and other enzymatic systems, its role as a prooxidant in reactive oxygen species generation, and its separate involvement in antioxidant mechanisms. We also review CoQ biosynthesis, which is particularly complex because of its extreme hydrophobicity, as well as the biological consequences of primary and secondary CoQ deficiency, including in human patients. Primary CoQ deficiency is a rare inborn condition due to mutation in CoQ biosynthetic genes. Secondary CoQ deficiency is much more common, as it accompanies a variety of pathological conditions, including mitochondrial disorders as well as aging. In this context, we discuss the importance, but also the great difficulty, of alleviating CoQ deficiency by CoQ supplementation.

Keywords: CoQ; CoQ deficiency; coenzyme Q; mitochondrial disease; ubiquinone.

Graphical abstract

FIGURE 1.

Structure and redox states of coenzyme Q (CoQ). CoQ exists in 3 redox states: the fully oxidized form (CoQ) accepts 2 electrons to form CoQH₂ or accepts 1 electron to form the ubisemiquinone intermediate, followed by acceptance of an additional electron to form CoQH₂. The number of isoprene units in the tail varies between species from 6 to 10.

FIGURE 2.

Functions of CoQ in the mitochondrial respiratory chain. CoQ is a pivotal component of the mitochondrial electron transport chain, acting as a mobile electron carrier shuttling electrons from CI and CII to CIII. During this process, CoQ cycles between reduced and oxidized states. In addition to moving randomly and colliding with CI and CII, CoQ is also present in CI- and CIII-containing respiratory supercomplexes (SCs), formed by the dynamic association of ETC complexes. In SCs, CIII is normally observed as a dimer (CIII₂). All CoQ in the IMM likely behaves as a single functional pool, that is, CoQH2 can diffuse out of the CI and CIII assembled in SCs and become oxidized by CIII found outside of SCs. Conversely, CoQH₂ generated independently of SCs can diffuse in, and be oxidized by, CIII attached to CI assembled in SCs. See glossary for other abbreviations.

FIGURE 3.

CoQ is a cofactor for mitochondrial dihydroorotate dehydrogenase (DHODH). DHODH catalyzes the oxidation of dihydroorotate to orotate during the fourth step of the de novo biosynthesis of pyrimidine. The reaction is coupled to the reduction/oxidation of flavin mononucleotide (FMN) and CoQ. Orotate diffuses back to the cytosol to be converted by uridine monophosphate synthase (UPMS) to uridine 5-monophosphate (UMP), the precursor of all pyrimidine nucleotides. Preexisting uridine can be phosphorylated to UMP by uridine kinase (UCK), thereby bypassing the need for the DHODH-catalyzed step. See glossary for other abbreviations.

FIGURE 4.

CoQ levels modulate sulfide-quinone oxidoreductase (SQOR) activity. SQOR catalyzes the initial oxidation of hydrogen sulfide (H₂S) and utilizes CoQ as the electron acceptor. Sulfur is primarily transferred to glutathione (GSH) under physiological conditions, forming glutathione persulfide (GSSH), which is then oxidized by sulfur dioxygenase (SDO), to produce sulfite ($\mathrm{S}{\mathrm{O}}_3^{2\mathrm{-}}$) and regenerate GSH. SDO is also known as ethylmalonic encephalopathy protein1 (ETHE1), as its mutations are associated with ethylmalonic encephalopathy, an infantile metabolic disorder. GSSH is also a substrate for thiosulfate sulfurtransferase (TST), which converts $\mathrm{S}{\mathrm{O}}_3^{2\mathrm{-}}$ to thiosulfate (${\mathrm{S}}_2{\mathrm{O}}_3^{2\mathrm{-}}$). Alternatively, $\mathrm{S}{\mathrm{O}}_3^{2\mathrm{-}}$ is converted to sulfate ($\mathrm{S}{\mathrm{O}}_4^{2\mathrm{-}}$) by sulfite oxidase (SUOX) residing in the intermembrane space. A further effect of H₂S accumulation is inhibition of the enzymatic activity of short-chain acyl-CoA dehydrogenase (SCAD) that catalyzes the first reaction in the β-oxidation of short-chain fatty acids. Elevated blood butyrylcarnitine (C4) is the hallmark biomarker of SCAD deficiency. In addition to acting as a cofactor of SQOR, CoQ levels regulate SQOR transcriptionally by an unknown mechanism. See glossary for other abbreviations.

FIGURE 5.

Diagram of a submitochondrial particle (SMP). A SMP is an inside-out vesicle of the inner mitochondrial membrane (IMM). It retains all the respiratory chain (RC) components, and the inversion of the IMM exposes CI and CII to the medium, allowing unrestricted access to oxidation substrates, including NADH, which could not pass through the IMM. See glossary for other abbreviations.

FIGURE 6.

ROS production sites in CI. A: with forward electron transport, NADH is oxidized at the flavin mononucleotide (FMN, the I_F site). Electrons are then passed via several iron-sulfur (Fe-S) clusters to the CoQ binding site (I_Q), where CoQ is reduced before it dissociates from CI. B: reverse electron transport occurs when electrons from an overreduced CoQ pool flow back to CI and reduce NAD⁺. The I_F site has been considered to be the main site of ROS production from CI under the oxidation of NADH-linked substrates. ROS production during reverse electron transport mainly originates from electron leakage from reduced CoQ formed at the I_Q site, but the I_F site has also been shown to contribute. Rotenone (ROT) blocks the flow of electrons by inhibiting the binding of CoQ to I_Q, whereas S1QELs suppress electron leak from CoQ^•− to oxygen at the I_Q site specifically. They do this without interfering with normal electron flow, and therefore this is expected to affect ROS generation during reverse electron transport. The O₂^•− produced by CI is released into the matrix, where superoxide dismutase 2 (SOD2) converts it to H₂O₂. See glossary for other abbreviations.

FIGURE 7.

Schematics of the mechanism of the Q cycle. The Q cycle mechanism defines 2 reaction sites in CIII: CoQH₂ oxidation (Q_o) and CoQ reduction (Q_i). The Q_o site is located between the Rieske iron-sulfur protein (RISP) and heme b_L, toward the intermembrane space, whereas the Q_i site is close to the matrix side. It takes 2 CoQH₂ oxidation cycles to complete the Q cycle. At first, a CoQH₂ moves into the Q_o site and undergoes oxidation, with 1 electron being transferred to RISP and then to cyt c via cyt c₁. The other electron passes through 2 b-type hemes (b_L and b_H) across the membrane to the Q_i site, where a bound CoQ is reduced to CoQ^•− and finally to CoQH₂. CoQ and CoQH₂ are recycled back to the CoQ pool from the Q_o and Q_i sites, respectively, after being fully oxidized or reduced. Oxidation of each CoQH₂ molecule releases 2 protons into the intermembrane space, and in the second half of the cycle 2 protons from the matrix are used to reduce CoQ^•−. Given that it takes 2 electrons to fully reduce a CoQ molecule, a CoQ^•− intermediate is expected to be formed at the 2 distinct CoQ binding sites. Stigmatellin and myxothiazol are Q_o site inhibitors, whereas antimycin A blocks electron transfer from b_H to the CoQ molecule at the Q_i site. See glossary for other abbreviations.

FIGURE 8.

Role of CoQ in the plasma membrane redox system (PMRS). The PMRS consists of multiple component operations that result in electron transfer from cytosolic reducing equivalents to extracellular electron acceptors. NADH-cytochrome b₅ reductase (CYB5R), NAD(P)H:quinone oxidoreductase 1 (NQO1), and ferroptosis suppressor protein 1 (FSP1) are CoQ reductases that oxidize NADH or NADPH to reduce CoQ. The cell surface protein ENOX is the terminal oxidase by catalyzing electron transport from CoQH₂ to extracellular electron acceptors, including oxygen (O₂) and ascorbyl (monodehydroascorbate) free radical (AFR). Besides oxidizing CoQH₂, ENOX also possesses an alternative activity, which is catalyzing protein disulfide-thiol interchange. Other enzymes of the PMRS include the NADPH/NADH oxidase (NOX) that directly catalyzes the 1-electron transfer from cytosolic NADPH to molecular oxygen, the voltage-dependent anion-selective channel (VDAC) that reduces extracellular ferricyanide using NADH as electron donor, and the duodenal cytochrome b (DCYTB) that utilizes ascorbate (Asc) in the cytosol as an electron donor to reduce either extracellular ferric iron (Fe³⁺), cupric copper (Cu²⁺), or AFR. See glossary for other abbreviations.

FIGURE 9.

Redox metabolism of ascorbate. Ascorbate (Asc) can undergo 2 consecutive 1-electron oxidations that generate the ascorbyl free radical (AFR) as an intermediate and the complete oxidation product dehydroascorbate (DHA). Free radical-mediated oxidative stress results in the oxidation of Asc, yielding AFR. The CoQ-dependent plasma membrane redox system (PMRS) transfers reducing equivalents from intracellular electron donors to AFR outside of the cell, converting AFR back to reduced Asc. Two molecules of AFR can react with each other to form 1 DHA and 1 Asc. DHA made extracellularly can be transported through glucose transporters (GLUT) into the cell, where it can be recycled back to Asc using glutathione (GSH) as a reductant, yielding glutathione disulfide (GSSG). Extracellular AFR can also be reduced by duodenal cytochrome b (DCYTB) using intracellular Asc as an electron donor in some species and tissues. See glossary for other abbreviations.

FIGURE 10.

The lipid peroxidation process. It is initiated by the radical-mediated abstraction of a hydrogen atom from a bis-allylic methylene group in a lipid (LH). This leads to the formation of a carbon-centered lipid radical (L^•), which undergoes molecular rearrangement and then reacts with oxygen to form a peroxyl radical (LOO^•). In turn, LOO^• propagates a chain reaction through the formation of a new L^• by hydrogen abstraction from other lipids, while it itself is converted to a lipid hydroperoxide (LOOH). Decomposition of LOOH yields lipid alkoxyl (LO^•) and the formation of aldehydes such as malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) and other secondary and end products of lipid peroxidation.

FIGURE 11.

Antioxidant action of vitamin E (VE) against lipid peroxidation. VE scavenges lipid peroxyl radical (LOO^•) before it attacks other lipids (LH) to form lipid hydroperoxide (LOOH) and a new lipid radical (L^•), by which it terminates lipid peroxidation chain reactions. This leaves behind the vitamin E radical (VE^•). VE^• can be converted back to the reduced antioxidant form by CoQ or vitamin C (VC). VE^• can also react with another LOO^•, forming poorly reactive nonradical adducts, decay by reaction with another VE^• molecule to give inactive dimers, or be completely oxidized to vitamin E quinone (VEQ). VC^•, vitamin C radical. See glossary for other abbreviations.

FIGURE 12.

Relationship between the levels of CoQ and ROS in human skin fibroblasts. Severe (<30% of residual CoQ₁₀) and moderate (>50% of residual CoQ₁₀) defects are not associated with induction of oxidative stress. However, an intermediate defect (30–50% of residual CoQ₁₀) results in an increase of cellular ROS levels. WT, wild type. See glossary for other abbreviations.

FIGURE 13.

Role of CoQ in ferroptosis regulation. Ferroptosis suppressor protein 1 (FSP1) acts as an oxidoreductase mainly localized on the plasma membrane and reduces CoQ to CoQH₂ using electrons from reduced nicotinamide adenine dinucleotide (NADH) or reduced nicotinamide adenine dinucleotide phosphate (NADPH). By directly scavenging lipid peroxy radicals (LOO^•) generated from lipid peroxidation, the fully reduced form of CoQ, CoQH₂, prevents excessive lipid peroxidation and thus inhibits ferroptosis. In the cytosol and mitochondria, glutathione peroxidase 4 (GPX4) converts glutathione (GSH) to oxidized glutathione (GSSG) while reducing lipid hydroperoxides (LOOH) to lipid alcohols (LOH), which is the main mechanism to regulate ferroptosis. Mitochondrial ROS production likely contributes to ferroptosis. On the other hand, CoQH₂ generation from respiration and other dehydrogenases such as dihydroorotate dehydrogenase (DHODH) and glycerol-3-phosphate dehydrogenase (GPDH) likely enhance defense against ferroptosis by inhibiting lipid peroxidation. See glossary for other abbreviations.

FIGURE 14.

A tentative model of involvement of CoQ in UCP regulation. UCP-mediated uncoupling can be activated by free fatty acids (FFA), an effect that is sensitive to purine nucleotides (PNs). The CoQ redox state has no effect on basal and FFA-induced UCP-catalyzed H⁺ conductance in the absence of PNs, but it modulates the sensitivity of UCP to inhibition by PNs. At a given fatty acid concentration, increased CoQ reduction by the respiratory chain decreases the binding affinity of PN to UCP, possibly by directly interfering with PN binding to UCPs due to structural similarities, thereby promoting UCP activation. Conversely, at lower CoQH₂ levels, no negative regulation occurs and UCP activity is inhibited by PNs. Additionally, it has been proposed that altered levels of CoQ or its redox state potentially could affect ROS-mediated UCP activation through its antioxidant activity against lipid peroxidation. See glossary for other abbreviations.

FIGURE 15.

The CoQ biosynthesis pathway in the yeast S. cerevisiae and humans. The proteins are in blue (S. cerevisiae) or green (humans). Dotted arrows designate multiple-step reactions, and the steps that are specific to yeast are shown in blue. R indicates the poly-isoprenoid tail. The synthesis of the isoprenoid depends on the mevalonate pathway, which produces the biosynthetic precursors of isoprenoids. Coq1 in yeast and a heterotetrameric protein formed by PDSS1 and PDSS2 in humans determine the number of isoprene units in the polyisoprene tail. The main ring precursor used by both yeast and humans is 4-hydroxybenzoic acid (4-HB), synthesized from tyrosine in the cytosol. The first and last steps of this pathway have been defined in the yeast, where 5 aminotransferases, Aro8, Aro9, Bat2, Bna3, and Aat2, redundantly convert tyrosine to 4-hydroxyphenylpyruvate (4-HPP) and the last step is catalyzed by Hfd1. After its transport into mitochondria, Coq2/COQ2 attaches the isoprenoid tail to 4-HB. Subsequent to this step, the CoQ ring undergoes several sequential modifications before yielding CoQ. The intermediates detected in yeast are shown. A recent study using a human embryonic kidney cell line (HEK293) showed that the decarboxylation and hydroxylation of position C1 occur in a single oxidative decarboxylation step and it is catalyzed by COQ4. The reaction catalyzed by COQ4 preferentially occurs before the C5 hydroxylation by COQ6; however, the alternative sequence of reactions, that is COQ6 and COQ3 may act before COQ4, is also possible (488). Yeast can also utilize para-aminobenzoic acid (pABA) for CoQ synthesis, which is made from chorismate in 2 steps catalyzed by Abz1 and Abz2. The nitrogen-containing intermediates generated from its utilization are also depicted. Coq6 and Coq9 are able to deaminate the ring C4 position on the intermediate derived from pABA. Whether and where Coq4 is involved before the deamination step remains to be demonstrated. pABA is also an intermediate in the synthesis of folate in the yeast. Two additional compounds that can also serve as ring precursors both in S. cerevisiae and mammals are shown at top: para-coumaric acid and resveratrol. The absence of Coq6 activity leads to the accumulation of 3-hexaprenyl-4-aminophenol (4-AP) and 3-hexaprenyl-4-hydroxyphenol (4-HP) when yeast cells are grown in pABA and 4-HB, respectively. The Coq/COQ proteins with unclear molecular functions are listed in a box. CoQ and intermediates are shown in their reduced forms. DDMQH₂, 3-hexaprenyl-5-methoxy-1,4-benzenediol; DMeQH₂, 2-methyl-3-hexaprenyl-5-methoxy-1,4,6-benzenetriol; DMQH₂, 2-methyl-3-hexaprenyl-5-methoxy-1,4-benzenediol; FPP, farnesyl diphosphate; GPP, geranyl pyrophosphate; HAB, 3-hexaprenyl-4-aminobenzoic acid; 4-HBz, 4-hydroxbenzaldehyde; HHAB, 4-amino-3-hexaprenyl-5-hydroxybenzoic acid; HHB, 3-hexaprenyl-4-HB; IPP, isopentenyl pyrophosphate. See glossary for other abbreviations.

Figure 16.

The chemical structure of 2,4-dihydroxybenzoic acid (2,4-DHB), 3,4-dihydroxybenzoic acid (3,4-DHB), and vanillic acid (VA) and their use as alternative benzoquinone ring precursors for CoQ biosynthesis. Of note, as mentioned in FIGURE 15. the order of Coq3-, Coq4-, and Coq6-catalyzed reactions has not been clearly established. A representative scheme of their use in the CoQ biosynthetic pathway is shown. However, it is possible that different sequences of reactions coexist or that the sequence of reactions changes depending on the availability of different precursors. See glossary for other abbreviations.

Figure 17.

Gene products and pathways whose defects are shown to cause secondary CoQ deficiency. See text for details. See glossary for abbreviations.