The reaction mechanism of the *Ideonella sakaiensis* PETase enzyme

Polyethylene terephthalate (PET) is a prevalent polyester plastic and a significant environmental pollutant. The discovery of *Ideonella sakaiensis*, a bacterium capable of degrading PET via a two-enzyme system (PETase and MHETase), opened new avenues for PET biorecycling. PETase directly acts on solid PET, liberating bis-hydroxyethyl terephthalate (BHET) and mono-hydroxyethyl terephthalate (MHET), while MHETase further cleaves MHET into ethylene glycol and terephthalic acid. Understanding PETase's catalytic mechanism is crucial for engineering improved enzymes for large-scale PET biodegradation. Previous studies suggested a two-step serine hydrolase mechanism involving a catalytic triad (Ser-His-Asp), but details remained unclear, particularly regarding the role of Trp185 and the proton transfer mechanism within the catalytic triad. Conflicting hypotheses existed about the interaction between Trp185 and the substrate, and the proton transfer mechanism (flipping vs. moving histidine, or charge relay) was debated. Computational studies had been undertaken, but often relied on pre-assumed reaction coordinates, potentially missing crucial mechanistic details. This study aimed to address these gaps using unbiased computational approaches.

The literature review summarizes existing knowledge on PETase and serine hydrolase mechanisms. It highlights the discovery of *I. sakaiensis* and its PET-degrading enzymes, along with previous engineering efforts to improve PETase activity. The active site of PETase, characterized as a serine hydrolase with a catalytic triad, suggests a two-step acylation-deacylation mechanism. However, the literature reveals conflicting hypotheses regarding Trp185's interaction with the substrate (edge-to-face vs. parallel-displaced π-π interaction) and the specific proton transfer mechanism within the catalytic triad (flipping, moving histidine, or charge relay). Prior computational studies, while insightful, often relied on pre-assumed reaction coordinates, leading to varying conclusions regarding the rate-limiting step and the nature of tetrahedral intermediates.

This study employed unbiased quantum mechanical/molecular mechanical (QM/MM) molecular dynamics simulations using transition path sampling (TPS) and inertial likelihood maximization (iLMax). First, a Michaelis complex of a hydroxyethyl-capped PET dimer bound to PETase was created. For both acylation and deacylation steps, QM/MM equilibration was performed using DFTB3. Aimless shooting (AS) was utilized to generate an ensemble of unbiased trajectories to find putative transition states. iLMax was then used to extract optimal reaction coordinates (RCs) from the AS ensemble, expressed as linear combinations of collective variables (CVs). The quality of the RCs was validated using a committor probability (*P_B*) histogram test. Umbrella sampling along the discovered RCs allowed the computation of free energy profiles, from which rate constants were estimated using the Eyring equation. Additionally, umbrella sampling simulations were performed along the Trp185 χ₁ dihedral to investigate its conformational changes during the reaction. Analysis included monitoring interactions within the catalytic triad, Trp185 conformation, oxyanion hole interactions, and charge buildup.

The key findings are: 1) Deacylation is the rate-limiting step in the PETase reaction, with a rate constant of 0.82 ± 0.10 s⁻¹. 2) The reaction proceeds via a moving histidine mechanism, without proton transfer between Asp206 and His237. 3) The oxyanion hole residues (Met161 and Tyr87) provide charge stabilization via hydrogen bonding to the PET carbonyl oxygen. 4) Trp185 undergoes conformational changes between reaction steps, facilitating catalysis; mutations restricting Trp185 motion reduce PETase activity. 5) Aromatic interactions between Trp185, Tyr87, and PET change throughout the reaction, supporting a hypothesis of an edge-to-face interaction during acylation shifting to a parallel-displaced interaction during deacylation. 6) Tetrahedral configurations correspond to transition states, not metastable intermediates, for both acylation and deacylation steps. The findings of this research were validated by rigorous statistical analysis such as committor probability analysis and high transmission coefficients. The results of the rate constants were also compared and found to be consistent with previous experimental work.

The findings of this study provide a detailed and unbiased picture of PETase's catalytic mechanism, resolving previous ambiguities regarding the role of Trp185 and the proton transfer mechanism. The identification of deacylation as the rate-limiting step provides a target for future enzyme engineering efforts. The importance of Trp185 flexibility suggests that mutations designed to increase this flexibility could significantly enhance catalytic efficiency. The elucidation of the moving histidine mechanism clarifies the role of His237 and Asp206. The study’s methodology, using unbiased TPS and iLMax, represents a significant advancement in the computational study of enzyme mechanisms. The mechanistic insights are likely transferable to other PET-degrading enzymes and cutinases, highlighting potential avenues for rational enzyme design.

This study provides a comprehensive, unbiased description of the *I. sakaiensis* PETase reaction mechanism using advanced QM/MM simulations. Key mechanistic insights, such as the identification of the rate-limiting step and the role of Trp185, offer valuable targets for future protein engineering endeavors aimed at enhancing PET degradation. The methodology employed also showcases the power of unbiased computational methods in elucidating complex enzymatic mechanisms. Future research could focus on investigating PETase's activity on longer PET chains, exploring interactions at the solid PET interface, and determining the impact of various mutations on enzyme flexibility and catalytic efficiency.

The study used a PET dimer as a substrate analog; the mechanism with longer PET chains or at the solid PET interface might differ. Simulations were conducted in solution, neglecting potential influences of a solid PET surface on enzyme conformation and dynamics. Furthermore, the impact of temperature on the reaction mechanism is not addressed.