Finding. The open conformation of mammalian complex I finally has a complete ubiquinone tunnel, because mycothiazole holds the catalytic mess still. Akisada, Murai, Yokoyama and colleagues put a mycothiazole-type inhibitor on bovine heart submitochondrial particles and do cryo-electron microscopy. A photoreactive analogue had already labeled the 49-kDa subunit and ND1, both walls of the tunnel. Bound mycothiazole reorganizes and stabilizes the ubiquinone catalytic region around iron-sulfur cluster N2 in both open and closed particles. That is enough to model the entire tunnel in the open state, which had been a disordered gap. The surprise is not only completeness. With the ligand bound, PSST loops that form the catalytic region look closed, while the transmembrane helices of ND1 and ND6 still look open. Open-versus-closed is not one switch. Catalysis-site loops and proton-pump helices can rearrange on their own.
Why this paper matters
Complex I is the first enzyme of the mitochondrial respiratory chain and a disease, aging, and inhibitor target. The field has been arguing whether the open state is catalytic, damaged, or a pumping intermediate, in part because the open Q-site would not sit still for a model. A complete open tunnel is a coordinate set you can dock, mutate, and simulate. A hybrid architecture is a mechanistic claim: the Q-site and the ND1/ND6 pump gate are not obligatorily locked.
This is bovine heart, in membranes, with a natural-product-class probe. It is not a detergent curiosity.
What they actually measured
Photoaffinity mapping, then cryo-EM with and without inhibitor, open and closed classes. The abstract is explicit that mycothiazole creates a common hybrid architecture in both conformations.
How to read the score
Mid 90s. Primary mammalian complex I structure, a missing tunnel, and an independence claim about catalysis versus pumping parts. Confidence is high for the structural description. The functional reading of the hybrid is an argument.
Caveats
Ligand-bound is not turning over. Bovine, not human. Do not retire the two-state language entirely; add a hybrid as a third allowed arrangement.
What to do with it
If you simulate or dock Q-site ligands, replace the disordered open tunnel with this model. If you mutate coupling residues, look at PSST loops versus ND1/ND6 helices as separately movable. Pull the 49-kDa/ND1 label sites. Do not call mycothiazole a drug from this brief.