Machine-Learning-Guided Engineering of an NADH-Dependent 7ß-Hydroxysteroid Dehydrogenase for Economic Synthesis of Ursodeoxycholic Acid.

Enzymatic synthesis of ursodeoxycholic acid (UDCA) catalyzed by an NADH-dependent 7ß-hydroxysteroid dehydrogenase (7ß-HSDH) is more economic compared with an NADPH-dependent 7ß-HSDH when considering the much higher cost of NADP+/NADPH than that of NAD+/NADH. However, the poor catalytic performance of NADH-dependent 7ß-HSDH significantly limits its practical applications. Herein, machine-learning-guided protein engineering was performed on an NADH-dependent Rt7ß-HSDHM0 from Ruminococcus torques. We combined random forest, Gaussian Naïve Bayes classifier, and Gaussian process regression with limited experimental data, resulting in the best variant Rt7ß-HSDHM3 (R40I/R41K/F94Y/S196A/Y253F) with improvements in specific activity and half-life (40 °C) by 4.1-fold and 8.3-fold, respectively. The preparative biotransformation using a "two stage in one pot" sequential process coupled with Rt7ß-HSDHM3 exhibited a space-time yield (STY) of 192 g L-1 d-1, which is so far the highest productivity for the biosynthesis of UDCA from chenodeoxycholic acid (CDCA) with NAD+ as a cofactor. More importantly, the cost of raw materials for the enzymatic production of UDCA employing Rt7ß-HSDHM3 decreased by 22% in contrast to that of Rt7ß-HSDHM0, indicating the tremendous potential of the variant Rt7ß-HSDHM3 for more efficient and economic production of UDCA.

Discovery and Engineering of a Novel Baeyer-Villiger Monooxygenase with High Normal Regioselectivity.

Baeyer-Villiger monooxygenases (BVMOs) are remarkable biocatalysts for the Baeyer-Villiger oxidation of ketones to generate esters or lactones. The regioselectivity of BVMOs is essential for determining the ratio of the two regioisomeric products ("normal" and "abnormal") when catalyzing asymmetric ketone substrates. Starting from a known normal-preferring BVMO sequence from Pseudomonas putida KT2440 (PpBVMO), a novel BVMO from Gordonia sihwensis (GsBVMO) with higher normal regioselectivity (up to 97/3) was identified. Furthermore, protein engineering increased the specificity constant (kcat /KM ) 8.9-fold to 484 s-1 mM-1 for 10-ketostearic acid derived from oleic acid. Consequently, by using the variant GsBVMOC308L as an efficient biocatalyst, 10-ketostearic acid was efficiently transformed into 9-(nonanoyloxy)nonanoic acid, with a space-time yield of 60.5 g L-1 d-1 . This study showed that the mutant with higher regioselectivity and catalytic efficiency could be applied to prepare medium-chain ω-hydroxy fatty acids through biotransformation of long-chain aliphatic keto acids derived from renewable plant oils.

An Energy Optimization Strategy Based on the Perfect Conformation of Prolyl Endopeptidase for Improving Catalytic Efficiency.

Prolyl endopeptidases (PEPs) hydrolyze proteins to yield bioactive peptides and are effective in the treatment of celiac disease. However, the catalytic efficiency of PEPs still has the potential to be improved, which could further strengthen their industrial and therapeutic applications. Herein, a novel rational design strategy based on a "near-attack conformation" of the catalytic state of PEP was adopted. Constrained dynamic simulations were applied, followed by the virtual screening of potentially favorable mutants according to their binding free energy. We redesigned Sphaerobacter thermophiles PEP with high-temperature activity/stability, a wide range of pH stabilities, and high proline specificity. As a result, the kcat value of two PEP mutants (I462W and Q560Y) increased by 208.2 and 150.1%, respectively, and the kcat/KM increased by 32.7 and 6.3%, respectively. These data revealed that the PEP mutants had improved catalytic efficiency and that our strategy can be applied for enzyme engineering.