Enzymatic Resolution: 97.4% Yield and >99% e.e. for (2S,3R)-p-MPSE vs. (2R,3S) Isomer via Engineered Lipase
A machine learning-engineered lipase variant (E115F/L67D/I421Q/P317I) achieved complete kinetic resolution of racemic threo-p-MPSE, yielding the target (2S,3R)-p-MPSE with enantiomeric excess >99% and an isolated yield of 97.4% (based on theoretical single-enantiomer amount) from 1 M substrate in 4 h. The unwanted (2R,3S) isomer was recycled via chemical racemization and re-esterification [1]. This represents a 15.4-fold improvement in catalytic efficiency (kcat/Km) over the wild-type lipase.
| Evidence Dimension | Enantioselectivity of lipase-catalyzed kinetic resolution |
|---|---|
| Target Compound Data | (2S,3R)-p-MPSE: >99% e.e., 97.4% yield, 1 M substrate, 4 h reaction time |
| Comparator Or Baseline | Wild-type BSL lipase: low enantioselectivity toward (2R,3S)-p-MPSE; baseline Novzyme 435: 36.83% conversion, 35.13% e.e. |
| Quantified Difference | 15.4-fold improvement in kcat/Km; conversion rate improved from 36.83% to complete resolution; e.e. from 35.13% to >99% |
| Conditions | Engineered Bacillus subtilis lipase, 1 M rac-threo-p-MPSE, coupled enzymatic resolution–chemical racemization system |
Why This Matters
This data demonstrates that only the (2S,3R) isomer is obtainable in near-quantitative enantiopurity via this advanced process, making it the sole viable procurement option for downstream florfenicol synthesis requiring >99% chiral purity.
- [1] Chemical Engineering Journal (2026). Machine Learning-Driven Lipase Engineering for Highly Enantioselective Synthesis of Chiral Florfenicol Intermediate via Enzymatic Resolution–Chemical Racemization Coupling System. View Source
