Protonation State of Active-Site Histidines, Reaction Mechanism and Stereoselectivity in β-Alanine Synthase: A Computational Study

β-Alanine synthase (βAS), which is a dizinc metalloenzyme, catalyzes the irreversible hydrolysis of <i>N</i>-carbamyl-β-alanine (NCβA) to β-alanine. This enzyme has potential applications for β-amino acid production. Understanding the reaction mechanism and selectivity of βAS at atomic details can help design and engineer the enzyme for cascade biocatalysis. Here, the protonation states of two conserved active-site histidine residues (His262 and His397 in <i>Saccharomyces kluyveri</i>) of βAS were investigated by means of combined quantum mechanical and molecular mechanical (QM/MM) molecular dynamics (MD) simulation, as well as the ONIOM QM/QM' approach. The calculations predicted that both His262 and His397 should be neutral for efficient catalysis. Furthermore, the βAS reaction mechanism and its stereospecificity toward a series of NCβA substrates containing different β² and β³-β-alanine substitutions were studied, which suggested factors governing the origin of stereoselectivity of this enzyme. The mechanism for the conversion of NCβA into β-alanine, carbon dioxide, and ammonia by βAS involved four reaction steps: nucleophilic attack by a hydroxide ion, substrate protonation and formation of a zwitterionic intermediate, and C-N bond cleavage to produce β-alanine and carbamate, which is finally decomposed into carbon dioxide and ammonia. The rate-limiting step is the protonation of the amide nitrogen of the substrate by Glu159, with the overall reaction barrier (16.5 kcal/mol) consistent with the experimental data. <i>In silico</i> alanine scanning analysis of the reaction mechanism for four variants (His262Ala, His397Ala, Asn309Ala, and Arg322Ala) is performed, showing increased activation energies compared to the wild-type enzyme, which confirms the roles of these residues in catalysis. The results explain the enzyme's preference for linear <i>N</i>-carbamyl substrates, as large and branched substrates cannot fit in the active site, restricted by the residue of the loop/region of the enzyme. Overall, we have demonstrated that a combined use of QM/MM MD and ONIOM models can be a promising strategy to elucidate possible protonation states of the ionizable residues in the enzyme active site prior to catalysis.