Finding
APOE4 mouse astrocytes loaded with cholesterol wreck the inner membrane in a specific way. Cristae become sparse and short, crista junctions pile up, Complexes I and IV drop, and respiration falls. Membrane potential does not fall with them. It rises. Leak is low, and ATP synthase runs backwards, hydrolyzing ATP. Those cells die when oligomycin blocks ATP synthase. Pull cholesterol out with methyl-β-cyclodextrin and the oligomycin toxicity goes away. Cut cholesterol by nutrient restriction and the cristae come back, with a partial return of the respiratory complexes. Epicatechin, used here as a block on the reverse mode, reduces death from the cholesterol stress and from respiratory-chain inhibitors.
Why it matters
APOE4 and atherosclerosis are both cholesterol diseases with a mitochondrial rumor attached. This preprint makes the rumor structural. Cholesterol amount changes crista shape, respiratory-complex abundance, and the direction ATP synthase runs. The cyclodextrin result is the causality test: sequester cholesterol, and the lethal dependence on ATP synthase lifts. That is a stronger claim than a correlation between APOE genotype and a fuzzy bioenergetic deficit.
Metal sites and redox species
Two respiratory complexes are lost, and each has a defined metal inventory that this paper does not measure.
Complex I carries FMN and a chain of iron-sulfur clusters, [2Fe-2S]2+/+ and ferredoxin-type [4Fe-4S]2+/+, with terminal cluster N2 donating to ubiquinone. Those are electron-carrier sites. The abstract says the complex is lost and respiration is lower. It does not say which cluster left or whether remaining clusters are stuck in the oxidized or reduced core state.
Complex IV reduces oxygen at heme a3/CuB. Electrons arrive via mixed-valence CuA and heme a. CuB is the catalytic oxygen site, not a simple carrier. Losing Complex IV protein would take those centers with it, or the centers could be empty inside a remaining shell. The paper does not distinguish those possibilities. No copper or heme measurement is reported.
The enzyme they do catch in the act is ATP synthase. In the forward direction, protons return through the Fo ring and F1 makes ATP. The catalytic site uses Mg2+ to coordinate nucleotides and phosphate. In reverse, the same site hydrolyzes ATP and Fo can pump protons, supporting membrane potential when the respiratory chain is weak. The authors report that reverse activity, together with low proton leak, accounts for a higher membrane potential despite less respiration. Oligomycin blocks proton passage through Fo. Hypersensitivity to it means survival had come to depend on that path. Epicatechin is described as stopping the reverse reaction and lowering cell death, which fits a picture in which ATP hydrolysis is both a way to hold the potential and a drain that injures the cell. Magnesium itself is not a measured analyte. It is the mechanistic ion of the F1 site they are functionally interrogating.
What was measured
Cristae architecture, Complexes I and IV, respiration, proton leak, reverse ATP synthase, membrane potential, oligomycin toxicity, cholesterol sequestration with methyl-β-cyclodextrin, epicatechin, and nutrient restriction, in APOE4 murine astrocytes.
Caveats
Mouse astrocytes. Cofactor occupancy of Complexes I and IV is unknown. MICOS is not named, so the extra junctions should not be assigned to a specific junction protein. Epicatechin's physical target is taken from the authors' functional description. Preprint.
What to do with it
Use this when asking how cholesterol touches oxidative phosphorylation. The chain to test is cholesterol, crista geometry, loss of Complexes I and IV, reverse ATP synthase, and oligomycin sensitivity. Keep copper, heme, and iron-sulfur clusters as the unmeasured cofactors of the complexes that disappeared.