Mitochondrial Supercomplexes Do Not Enhance Catalysis by Quinone Channeling

Abstract

Mitochondrial respiratory supercomplexes, comprising complexes I, III, and IV, are the minimal functional units of the electron transport chain. Assembling the individual complexes into supercomplexes may stabilize them, provide greater spatiotemporal control of respiration, or, controversially, confer kinetic advantages through the sequestration of local quinone and cytochrome c pools (substrate channeling). Here, we have incorporated an alternative quinol oxidase (AOX) into mammalian heart mitochondrial membranes to introduce a competing pathway for quinol oxidation and test for channeling. AOX substantially increases the rate of NADH oxidation by O₂ without affecting the membrane integrity, the supercomplexes, or NADH-linked oxidative phosphorylation. Therefore, the quinol generated in supercomplexes by complex I is reoxidized more rapidly outside the supercomplex by AOX than inside the supercomplex by complex III. Our results demonstrate that quinone and quinol diffuse freely in and out of supercomplexes: substrate channeling does not occur and is not required to support respiration.

Keywords: alternative oxidase; channeling; mitochondria; oxidative phosphorylation; respirasome; supercomplex; ubiquinone.

Graphical abstract

Figure 1

Pictorial Representation of Quinone/Quinol Cycling during NADH Oxidation by a Respiratory-Chain Supercomplex in the Presence of AOX (A) The CI₁:CIII₂:CIV₁ supercomplex might confine channeled quinone/quinol (Q_C, blue pathway) to shuttle between CI and CIII within the supercomplex assembly, or depend on quinone/quinol exchange with the quinone pool (Q_P, red pathway). The competing quinol oxidase (AOX) can only react with quinol from the pool. Image created from the structures of the ovine respirasome (Letts et al., 2016) and T. brucei AOX (Shiba et al., 2013). (B) Flow diagram showing flux through the CIII/CIV and AOX pathways if quinone is channeled (blue pathway) or behaves as a shared pool (red pathways).

Figure 2

Addition of AOX Stimulates Catalysis by SMPs from Bovine Heart Mitochondria (A) Kinetic assay trace showing how catalysis responds to addition of 0.1 mg AOX per mg SMP. Rates of catalysis are marked in μmol NADH min⁻¹ mg⁻¹. A total of 200 μM NADH, 10 μg mL⁻¹ SMPs, 400 μM NaCN, 0.1 mg AOX mg⁻¹, and 500 nM ascofuranone were added sequentially as indicated. (B) Rates of NADH oxidation in SMPs, uncoupled using 1 μg mL⁻¹ gramicidin and treated with 0.1 mg AOX mg⁻¹ and/or 300 nM ascofuranone. DMSO (the vehicle for ascofuranone) was added to ascofuranone-free experiments at 0.1%. All values are mean ± SD (n = 6). (C and D) The effects of supplementing SMPs with increasing concentrations of AOX on the rates of succinate:O₂ (C) and NADH:O₂ (D) oxidoreduction. Conditions: 10 μg mL⁻¹ SMPs, 200 μM NADH, 400 μM NaCN, 1 μg mL⁻¹ gramicidin. Control: 0.2% DMSO (the vehicle for gramicidin). All values are mean ± SD (n = 6).

Figure 3

Supplementation with 0.1 mg AOX/mg SMPs Does Not Affect the Respiratory-Chain Supercomplexes, the Membrane Integrity, or Δp (A) Blue native PAGE analysis of SMPs with and without 0.1 mg AOX mg⁻¹ (10 μg SMPs per lane) stained with Coomassie R250 (left) or using an NADH oxidase activity stain (right). SC denotes the supercomplex bands. (B) Catalysis of the RET reaction (NAD⁺ reduction) by SMPs as a function of AOX (blue) and DDM (red) concentrations. Each point is a mean ± SD (n = 6). Blue squares show the uncoupling effect of adding 5 μg mL⁻¹ gramicidin in the presence of different AOX concentrations (n = 3). (C) Dependence of the rate of NADH oxidation (or NAD⁺ reduction) on ΔE, modulated by varying the fumarate concentration against fixed NADH, succinate, and NAD⁺ concentrations (STAR Methods). SMPs (50 μg mL⁻¹) were treated with 0.1 or 1.0 mg AOX mg⁻¹ or 5 μg mL⁻¹ gramicidin, as indicated. The data are presented as mean ± SD (SMPs, n = 11; SMPs + gramicidin, n = 3; AOX-SMPs [0.1 mg mg⁻¹], n = 10; AOX-SMPs [1 mg mg⁻¹], n = 3).

Figure 4

Addition of AOX to SMPs to Determine the H⁺/2e^– Stoichiometry of Complex I The rate of ATP synthesis driven by NADH oxidation through CI/CIII/CIV or CI/AOX is shown as a function of the rate of NADH oxidation. See STAR Methods for experimental details. The linear regression fits through the origin have slopes (±SE of the fit) of 0.485 ± 0.013 (r² = 0.8663) and 0.193 ± 0.0046 (r² = 0.8032) ATP NADH⁻¹, respectively. The ratio of slopes equates to the CI stoichiometry (n^CI) shown.