Computational Studies of Asymmetric Conformational Dynamics of AAA+ Machines Involved in Protein Disaggregation

Caseinolytic (Clp) ATPases are AAA+ (ATPases associated with diverse cellular Activities) biological nanomachines that assist essential quality control processes of protein degradation or disaggregation. ClpB is a double-ring hexamer that resolubilizes protein aggregates in repetitive cycles of non-concerted conformational transitions coupled with mechanical forces applied by pore loops onto substrate proteins. To study the allosteric regulation and asymmetric conformational dynamics of this machine, we performed all-atom molecular dynamics simulations of two different conformational states (“ring” and “spiral”) of ClpB in nucleotide and/or substrate-bound or apo configurations. Simulation results highlight distinct mechanisms of dynamic stabilization of rings formed by two nucleotide binding domains (NBDs) of ClpB. Whereas the network of salt bridges stabilizing the NBD1 ring involves the canonical loops, in accord with experimental data, that of NBD2 involves non-canonical loops. Using machine learning approaches, we find that strong intra-ring collaboration involves correlated motions of NBD domains. Analysis of relaxation dynamics of canonical loops indicates coupling of local and collective motions on time scales consistent with experimental results.