Oxidative phosphorylation (OXPHOS)

The process by which mitochondria convert the reducing equivalents NADH and FADH₂ — generated by the tricarboxylic acid cycle and fatty-acid beta-oxidation — into ATP via a proton electrochemical gradient across the inner mitochondrial membrane (IMM). OXPHOS supplies ~95% of cellular ATP under aerobic conditions and is the dominant energetic engine of post-mitotic, high-demand tissues (heart, brain, skeletal muscle). Its progressive decline with age is a central driver of the mitochondrial-dysfunction hallmark, and the reactive oxygen species (ROS) it generates as by-products are the original substrate of the free-radical-theory-of-aging.

This page covers the molecular architecture and aging biology of the electron transport chain (ETC) + ATP synthase coupled system. Upstream metabolism (tca-cycle) and downstream mitochondrial quality control (mitophagy, mitochondrial-biogenesis) are treated on their own pages.

ETC architecture

Five multi-subunit membrane complexes embedded in the IMM execute OXPHOS ¹:

Complex	Common name	Subunit count (human)	Approx. mass	Prosthetic groups	Reaction
Complex I (CI)	NADH:ubiquinone oxidoreductase	~45	~1 MDa	FMN, 8 Fe-S clusters	NADH + ubiquinone → NAD⁺ + ubiquinol; pumps 4H⁺/2e⁻
Complex II (CII)	Succinate:ubiquinone oxidoreductase	4	~140 kDa	FAD, 3 Fe-S clusters, heme b	FADH₂ (from succinate) → ubiquinol; no proton pumping
Ubiquinone (CoQ₁₀)	— mobile carrier in IMM	—	—	Quinone head	Collects electrons from CI/CII → CIII
Complex III (CIII)	Cytochrome bc1 complex	11	~240 kDa	2 hemes b, heme c1, Rieske Fe-S	Ubiquinol → cytochrome c (reduced); pumps 4H⁺/2e⁻ via Q-cycle
cytochrome-c	— mobile carrier in IMS	1	~12 kDa	Heme c	Shuttles electrons from CIII → CIV
Complex IV (CIV)	Cytochrome c oxidase	13 (+ assembly factors)	~200 kDa	hemes a/a3, CuA/CuB	4 × cytochrome c(red) + O₂ → 2H₂O; pumps 2H⁺/e⁻ pair
Complex V (CV)	ATP synthase (F₀F₁)	~16 (2 subcomplexes)	~600 kDa	—	Proton flow through F₀ rotor drives ADP + Pi → ATP in F₁ head

Electron flow CI → CoQ → CIII → cytochrome-c → CIV generates the proton-motive force (PMF) — a combination of electrical potential (~−180 mV, negative inside) and pH gradient across the IMM. Complex V couples this PMF to ATP synthesis: ~2.7 ATP are synthesized per NADH oxidized at CI (theoretical P/O ratio ~2.5 for NADH, ~1.5 for FADH₂). In practice, uncoupling proteins (UCPs) and proton leak reduce measured P/O values in intact cells ².

Supercomplexes (respirasomes): CI, CIII, and CIV associate into higher-order supercomplexes (CI + CIII₂ + CIV, the “respirasome”) in the IMM. This organization is thought to enhance electron channeling efficiency and reduce CI-driven ROS production, though the extent of functional vs. structural coupling is debated. contradictory-evidence

Inhibitors as research tools

Standard pharmacological probes isolate individual complex activities:

Inhibitor	Target	Mechanism	Notes
Rotenone	CI	Blocks Q-binding site; prevents ubiquinol reduction	Model of Parkinson-like neurodegeneration; rotenone-exposed rodents recapitulate dopaminergic loss
TTFA (thenoyltrifluoroacetone)	CII	Blocks Q-binding site (proximal to SDHB)	Less commonly used; also inhibits succinate dehydrogenase
Antimycin A	CIII	Blocks Qi site; traps semiquinone radical → ↑↑ superoxide	Maximally amplifies mitochondrial ROS production
KCN / azide / CO	CIV	Compete with O₂ at heme a3/CuB center	Classical respiratory poisons
Oligomycin	CV (F₀)	Blocks proton channel of F₀ subunit c-ring	Dissociates ATPase activity from PMF; standard control for ATP production assays

Seahorse XF respirometry uses sequential oligomycin → FCCP (uncoupler) → rotenone/antimycin A injections to resolve basal respiration, ATP-linked respiration, maximal respiratory capacity, and proton leak in intact cells.

ROS production and the free-radical hypothesis

Under isolated mitochondrial conditions (mode 2: high Δp, reduced CoQ pool, no ATP synthesis), ~0.1–2% of electrons flowing through the ETC escape to molecular oxygen, generating superoxide (O₂•⁻) ². Murphy 2009 explicitly cautions that this value applies only to isolated mitochondria and cannot be extrapolated to the in vivo situation, where O₂•⁻ production is “far, far lower.” The two principal sites are:

Complex I — NADH-binding flavin (FMN site; reverse electron transport from ubiquinol under high PMF is especially productive) and the Q-binding module. CI is the dominant mitochondrial ROS source under physiological and high-CoQH₂/CoQ redox states.
Complex III — Outer Q-site (Qo); produces superoxide bidirectionally into the matrix and into the IMS. CIV and CII produce comparatively little.

Superoxide is rapidly dismutated by manganese SOD (SOD2) in the matrix and Cu/Zn SOD (SOD1) in the cytosol, generating H₂O₂. H₂O₂ can react with Fe²⁺ (Fenton chemistry) to generate the highly reactive hydroxyl radical (OH•). Collectively these are mitochondrial ROS (mtROS).

The free-radical-theory-of-aging (Harman, 1956; status: superseded as a strong causal claim) predicted that mtROS accumulation drives cumulative macromolecular damage. Key nuances from subsequent decades of work:

ROS as signaling molecules: Redox-sensitive cysteines on kinases, transcription factors, and ion channels mean mtROS serves physiological roles in nutrient sensing, immune activation, and preconditioning — not all mtROS is harmful. no-mechanism for the distinction between physiological vs. damaging ROS levels.
NMR paradox: heterocephalus-glaber (naked mole-rat) tolerates substantially elevated protein oxidative damage and lipid peroxidation compared to age-matched mice, yet lives ~8× longer per body mass — directly challenging a linear damage-aging relationship. needs-replication — NMR oxidative damage baseline needs independent validation across tissues.

Aging-specific decline

ETC function deteriorates progressively with age in high-demand tissues. Key mechanisms:

1. mtDNA mutations and heteroplasmy

The mitochondrial genome (16.6 kb, 37 genes encoding 13 OXPHOS subunits + 22 tRNAs + 2 rRNAs) is exposed to higher mutagenic stress than nuclear DNA — proximity to CI/CIII ROS sites, absence of histones, and limited mismatch repair. Point mutations and large deletions accumulate with age, especially in post-mitotic neurons, cardiomyocytes, and skeletal muscle fibers. Beyond a threshold (~60–80% heteroplasmy, tissue-dependent), respiratory chain function is impaired.

The PolG mutator mouse expresses a proofreading-deficient mitochondrial DNA polymerase gamma (D257A knock-in), leading to accelerated mtDNA mutation accumulation. These mice display a premature aging phenotype — hair loss, kyphosis, cardiomyopathy, anemia, osteoporosis, reduced fertility, reduced lifespan; median lifespan ~48 weeks (~11 months), with all homozygotes dying before 61 weeks ³. Whether the mechanism is primarily bioenergetic insufficiency or increased ROS (or apoptosis) remains unresolved; later work suggests ROS elevation is modest, implicating energy deficiency or apoptotic signaling. contradictory-evidence — causal weight of ROS vs. energy deficiency in PolG aging not settled.

Extrapolation (PolG → humans):

Dimension	Status
Pathway conserved in humans?	yes — mtDNA heteroplasmy accumulates in aged human tissues
Phenotype conserved in humans?	partial — human mtDNA deletion burden increases in muscle, neurons; cardiomyopathy associations known
Replicated in humans?	in-progress — rates and thresholds not fully characterized; no prospective human PolG equivalent

2. Complex-specific activity declines

Cross-sectional biochemical studies of aged human and rodent tissues show decreases in CI and CIV enzyme activity in skeletal muscle, heart, and brain; CII and CIII activities are more variable. Skeletal muscle data from humans are the most consistent: CI activity declines ~40% from young to very old in some studies. Liver appears relatively spared. Tissue specificity likely reflects differential mtDNA copy number, fusion/fission balance, and mitophagic capacity. needs-replication — most human data are cross-sectional; longitudinal data are limited.

3. IMM remodeling and cristae changes

OXPHOS complexes are organized in cristae — invaginations of the IMM that concentrate the ATP synthesis machinery. With aging, cristae become less tightly packed, OPA1-dependent cristae junctions widen, and supercomplex stability decreases. These structural changes precede or accompany bioenergetic decline. See cytochrome-c for the cristae-gating role of OPA1 in cytochrome-c mobilization during apoptosis.

Quality control — the OXPHOS-mitophagy axis

Damaged mitochondria with low membrane potential (ΔΨm) are selectively tagged for mitophagy via the PINK1/Parkin pathway. PINK1 is normally imported and cleaved at the IMM; when ΔΨm collapses, PINK1 accumulates on the outer membrane, phosphorylates ubiquitin and Parkin, leading to selective autophagic degradation of the compromised organelle. This closes a key feedback loop: OXPHOS dysfunction → PINK1 stabilization → mitophagy → removal of dysfunctional units, preventing ROS amplification. With aging, mitophagy flux itself declines (see autophagy), allowing dysfunctional mitochondria to persist and propagate mtDNA mutations.

mitochondrial-biogenesis, driven primarily by PGC-1α/TFAM/NRF1, replaces degraded mitochondria with new ones. The balance between biogenesis and mitophagy determines mitochondrial network quality. Both arms decline with sedentary aging; exercise activates both (see Therapeutic Angles below).

Therapeutic angles

Intervention	Mechanism	Aging/OXPHOS evidence	Human evidence
Exercise	Transient mitochondrial stress → PGC-1α → mitochondrial-biogenesis; AMPK-mediated mitophagy; complex activity increases	Robust in rodents + humans; skeletal muscle CI/CIV activity increases with training	Strong — exercise training improves peak VO₂ and OXPHOS markers in elderly dose-response-unclear for type/intensity/frequency in aged humans
urolithin-a	Induces mitophagy → turnover of dysfunctional mitochondria → improved OXPHOS competence	Ryu 2016 (C. elegans + aged mice); Andreux 2019 (human safety/mitophagy biomarkers)	ATLAS RCT: hamstring strength significant at both 500 mg and 1000 mg/day; aerobic endurance (VO₂ peak, 6MWT) significant at 1000 mg/day only; primary endpoint (peak power output) ns across all groups ⁴
MitoQ / SkQ1	Mitochondria-targeted antioxidants (plastoquinone conjugated to TPP⁺); accumulate ~500× in matrix vs. cytosol; scavenge CI/CIII superoxide	Lifespan extension in mice (MitoQ) contested; SkQ1 data largely from Skulachev group	Limited human data; no longevity trial completed needs-human-replication
NAD⁺ precursors (NR, NMN)	Boost NAD⁺ → sirtuins (SIRT1/3) → deacetylation of OXPHOS subunits and PGC-1α → complex activity, mitochondrial biogenesis	Strong preclinical; NMN mouse data positive	Human trials show NAD⁺ restoration in blood; muscle OXPHOS impact modest/mixed long-term-unknown

Limitations and gaps

Precise ROS flux rates in aged human tissues are technically inaccessible; most estimates derive from isolated mitochondria or permeabilized fibers, not intact aging tissue. needs-human-replication
Whether targeting ROS pharmacologically extends human lifespan is unresolved — antioxidant supplementation trials (vitamin E/C) have repeatedly failed or shown harm in some contexts. This challenges the free-radical theory in its strong form. See free-radical-theory-of-aging.
The relative contribution of CI vs. CIII to total aging-relevant ROS in specific human tissues is not quantified. dose-response-unclear
Supercomplex stability with age: limited cryo-EM structural data from aged human tissues. Most data are from detergent-solubilized preparations that may disrupt native architecture. needs-replication
mtDNA heteroplasmy thresholds for functional impairment vary substantially across tissue types and individual mitochondrial variants; no generalizable threshold established for human aging. needs-replication

Cross-references

mitochondrial-dysfunction — hallmark page; OXPHOS decline is a primary mechanistic contributor
mitophagy — selective removal of OXPHOS-impaired mitochondria; PINK1/Parkin axis
mitochondrial-biogenesis — replenishment of OXPHOS-competent mitochondria; PGC-1α axis
cytochrome-c — ETC electron carrier (CIII→CIV) + apoptosis initiator; dual role
free-radical-theory-of-aging — superseded strong-form hypothesis; mtROS central claim
heterocephalus-glaber — NMR oxidative damage paradox; evidence against strict ROS-aging proportionality
urolithin-a — mitophagy-inducing compound; improves OXPHOS competence in aging contexts
autophagy — OXPHOS decline contributes to mitophagy substrate accumulation; general flux context
ampk — OXPHOS deficiency → ↑AMP/ADP → AMPK activation → mitophagy and biogenesis
tca-cycle — upstream electron-donor cycle; NADH/FADH₂ supply to ETC (implicit stub)
pgc-1alpha — master transcriptional regulator of mitochondrial biogenesis (implicit stub)

Footnotes

doi:10.1038/191144a0 · Peter Mitchell 1961 · review/primary · model: theoretical + cell-free; Nobel Prize 1978 · chemiosmotic hypothesis; first formulation of proton-motive force coupling electron transfer to ATP synthesis · archive: not_oa (closed) ↩
doi:10.1042/BJ20081386 · Murphy MP 2009 · review · model: isolated mammalian mitochondria + submitochondrial particles · canonical quantitative description of 11 mitochondrial ROS-producing sites; CI and CIII identified as dominant; ~0.1–2% electron leak estimate · archive: OA bronze (pending download; citation count 7,798) ↩ ↩²
doi:10.1038/nature02517 · Trifunovic A et al. 2004 · in-vivo · n=PolG D257A knock-in mice (multiple litters; exact n per group variable across assays; survival curve: n=50 wt, n=38 mut/mut) · model: PolG proofreading-deficient knock-in mouse (C57BL/6 × 129Sv background) · premature aging phenotype; hair loss, kyphosis, cardiomyopathy, anemia, osteoporosis, reduced fertility; median lifespan ~48 weeks (~11 months, homozygotes); all homozygotes died before 61 weeks · archive: local PDF available ↩
singh-2022-atlas-rct-urolithin-a · doi:10.1016/j.xcrm.2022.100633 · n=88 randomized / 79 completed · rct (3-arm: placebo vs 500 mg vs 1000 mg UA/day × 4 months) · primary endpoint (peak power output) ns; hamstring strength significant at both doses; VO₂ peak + 6MWT significant at 1000 mg/day · model: untrained overweight middle-aged adults 40–64 yr · archive: local PDF available ↩

🧬 Aging Wiki

Explorer

oxphos