Engineering a highly active thermophilic F1-ATPase by homolog-guided exploration and machine-learning-assisted prioritization

The rotary motor F1-ATPase, widely studied as a molecular machine model, poses challenges in rational engineering of its catalytic activity due to ATP hydrolysis regulation by long-range intersubunit allostery and large conformational transitions. Researchers devised a homolog-guided engineering approach to boost the thermophilic Bacillus PS3 F1-ATPase (TF1)'s maximum rotation rate.

By comparing TF1 with homologous enzymes boasting higher maximum rotation rates—bovine mitochondrial F1 (bMF1) and Paracoccus denitrificans F1 (PdF1)—candidate mutation sites were pinpointed. Subsequent systematic examination of these sites revealed four activity-enhancing hotspots, followed by targeted hotspot investigation and machine-learning-assisted prioritization of combinatorial mutants.

The most promising mutant, TF1({beta}Y313L/{beta}E332S), demonstrated a 1.8-fold higher maximum rotation rate than the wild-type (WT) TF1 while maintaining its functional thermostability. Interestingly, activity-enhancing substitutions didn't solely rely on conserved residues in both bMF1 and PdF1, suggesting that bMF1-PdF1 consensus substitutions did not uniquely define the optimal amino acid.

Machine-learning-assisted exploration effectively prioritized highly active mutants, albeit with limited predictive performance due to a small training dataset and epistatic interactions among mutations. Kinetic and structural comparisons offered mechanistic insights into the enhanced catalytic activity of the engineered mutant.

These findings present a practical strategy for engineering complex molecular motors by merging homolog-guided hotspot identification with focused hotspot exploration.