Review Papers Claim by Claim

• Protonation states modeled: pre-protonation OOH vs post-protonation OHH (two relevant c-ring glutamate protonation configurations) OOH/OHH state modeling
• Free-energy landscape: metastable intermediates appear every ~36° (consistent with elementary rotational periodicity). ~36° metastable spacing in PMFs
• Directionality: after protonation (OOH→OHH), the landscape becomes overall downhill toward forward rotation; they propose the forward step is stabilized by a persistent cE111J–aR239 salt bridge. directionality via downhill OHH and cE111J:aR239
• Proton transfer inference: they do not simulate covalent proton transfer directly; instead they infer a likely protonation configuration at P2 using water-wire statistics and a “proton-transfer score,” arguing direct side-chain contact between donor and acceptor is rare/absent in accessible states. P2 water-wire protonation inference

Strong / better-supported claims

• Computed directionality via protonation-dependent PMFs: the central computational observation is that the OHH (post-protonation) profile exhibits a strong downhill trend toward forward rotation while the OOH profile requires overcoming barriers, consistent with the proposed role of protonation in biasing rotation. PMF directionality OOH vs OHH
• Salt-bridge stabilization with essential arginine: they argue that a conserved essential a-subunit arginine (aR239 in their mapping) forms/maintains a salt bridge with the c-ring glutamate on the (leading or trailing) subunit across rotation, and is particularly stabilizing in the forward state. aR239 salt-bridge role
• Compatibility with modern structural/functional framework: the paper explicitly situates its model within a cryo-EM-supported understanding of c-ring/a-subunit channels and the importance of polar residues and sealing to limit leakage. cryo-EM: dual channels + aArg239 seal

Inferred / model-dependent steps (major epistemic bottlenecks)

• Proton transfer is not covalently simulated in the main PMFs: the framework uses protonation-state labels (OOH vs OHH) and infers which rotational state is permissive for transfer by analyzing hydration-water-wire statistics, not by explicit bond-making/breaking. Proton transfer inferred, not covalently simulated
• Water-wire → efficient proton transfer is probabilistic, not deterministic: the paper argues that short water wires (length ≤8) appear frequently around P2, and that proton transfer would be rate-limiting by water orientation while the actual chemical event occurs on femtosecond timescales—however, the “proton-transfer score” is ultimately an operational heuristic whose calibration and mapping to real proton kinetics cannot be directly read off from the MD snapshots. P2 water-wire occupancy and transfer score
• Electrostatic “directionality” is computed under fixed modeling assumptions: they assume that the transmembrane potential does not substantially change the rotary PMFs because modeled charge movements are parallel to the membrane plane. This is plausible but still a critical assumption; any hidden component of charge displacement normal to the membrane, or dielectric/polarization mismatch in the force field, could alter barrier asymmetries. Assumption about ΔΨ effect on PMFs

• Free energy method: extended Adaptive Biasing Force (eABF) on a 2D space: rotation angle (θ) and a salt-bridge-related distance d1 between cE111J and aR239. This can accelerate sampling of rare rotations but depends on the choice/coverage of collective variables (CVs). eABF 2D PMF with d1 coordinate
• Sampling time reported: roughly ~70 μs of eABF sampling plus ~30 μs for restrained c-ring simulations used for protonated-donor conditional analysis. This is substantial but still may not fully resolve the kinetics behind barrier heights, especially for lipid reorganization and slow water-network rearrangements. reported simulation times
• Water-wire detection: they use a breadth-first algorithm on hydrogen-bond networks and Dijkstra to select shortest wires, then compute a proton-transfer score from wire occupancy and length. This is a mechanistically motivated pipeline, but hydrogen-bond criteria, donor/acceptor definitions, and wire topology thresholds can influence inferred “permissive” states. water-wire detection + scoring pipeline

• Barrier height vs kinetics mismatch: they acknowledge that the ~16 kcal/mol barrier might be overestimated because of imperfect sampling of conformational dynamics. If actual kinetics differ strongly, the supposed rate-limiting substep might change. Acknowledged possible overestimate of barriers
• Protonation-state simplification: they model only OOH and OHH rotational PMFs and assume rotary surfaces are largely ΔΨ-insensitive for their processes. A more complete thermodynamic cycle with coupled protonation fluctuations, explicit dielectric polarization, and different residue protonation microstates could alter relative basin depths. Protonation simplifications and potential PMF shifts
• Model system limits: the study focuses on the F0 region (rotor/stator interface) rather than full Fo–F1 coupling. Since mechanical coupling and subunit compliance can affect state occupancy, one should be cautious about claiming that the in vitro/isolated stepping mechanism fully predicts the full ATP synthesis cycle kinetics. Fo-focused model scope
• Species generalization: while the architecture of ATP synthases is conserved, this work is grounded in Polytomella sp. mitochondrial architecture. Directionality details (e.g., how water wires form, the role of metal ions) could be different in other isoforms. cautious generalization across isoforms

• Cryo-EM structural basis is consistent: the paper aligns with cryo-EM evidence for dual proton pathways in subunit a and the role of aArg239 sealing. aArg239 seal and dual channels
• Water-mediated proton transfer is in line with broader Grotthuss-type discussions: their mechanistic claim is coherent with established physical chemistry of water-wire proton transfer, though the paper’s novelty is in mapping permissive states using rotational CVs. Grotthuss-type inference in P2
• Torque scale is near known order-of-magnitude: the predicted 200 pN·nm is ~5× higher than the 40 pN·nm E. coli benchmark they cite; this could reflect model assumptions (force fields, barrier/PMF scaling) and/or differences in experimental conditions. torque scale comparison