Presentation Information
[P01-033]A High-Throughput Screening Method for Lipase Regioselectivity
○Elvis Teng Chua1, Ping Han1, Elaine Chiang1, Chloe Ueda Teo1, Vani Octavia Sugiharto1, John Koh1, Yan Ping Lim1, Bo Xue1, Wen Shan Yew1 (1. National University of Singapore (Singapore))
Keywords:
lipase,regioselectivity,high-throughput,screening,mass spectrometry,ether lipid
Lipases are industrially relevant enzymes, and improving their properties remains a major research focus. Regioselectivity is particularly important for producing high-value structured lipids such as OPO and SOS. Current methods employ sn-1,3 lipases that hydrolyze triglycerides at the sn-1 and sn-3 positions, enabling esterification with desired fatty acids (e.g., oleic acid for OPO and stearic acid for SOS). However, these processes generate surplus by-products (palmitic acid for OPO and oleic acid for SOS), which could be further utilized by an sn-2 lipase to improve overall yield.To date, no strictly sn-2 lipase has been identified. Some lipases exhibit partial sn-2 selectivity but produce mixed products, requiring additional downstream separation and limiting industrial use.
Enzyme engineering strategies such as orthologue screening and site-directed mutagenesis can enhance performance, but they often generate large libraries, posing significant analytical challenges—especially when substrates and products are not spectrophotometrically detectable. Current methods rely on mass spectrometry to quantify substrates and products; however, each sample requires approximately 15 minutes to achieve highly resolved product isomer peaks.
Here, we present customized lipid substrates to assess lipase regioselectivity.An enzyme library was expressed in Pichia pastoris GS115, and recombinant proteins were secreted into the culture medium. Supernatants were used directly for screening. Purified CalA (more sn-2 selective) and ROL (strictly sn-1,3 selective) served as controls. Two ether lipid substrates (O-OA-O for sn-2 and O-O-OA for sn-1,3) were used, and hydrolysis products (O-OH-O and O-O-OH) were detected using an Agilent RapidFire system coupled to a triple quadrupole mass spectrometer, enabling analysis in under 10 seconds per sample.Validation confirmed the expected selectivity of ROL and CalA. Using this method, 58 lipases were screened, of which 16 showed activity with regioselectivity similar to CalA. Future work will expand orthologue screening and apply site-saturation mutagenesis to generate larger libraries. These results will improve understanding of lipase regioselectivity and support machine learning models. Workflow improvements are also underway to further reduce processing time.
Enzyme engineering strategies such as orthologue screening and site-directed mutagenesis can enhance performance, but they often generate large libraries, posing significant analytical challenges—especially when substrates and products are not spectrophotometrically detectable. Current methods rely on mass spectrometry to quantify substrates and products; however, each sample requires approximately 15 minutes to achieve highly resolved product isomer peaks.
Here, we present customized lipid substrates to assess lipase regioselectivity.An enzyme library was expressed in Pichia pastoris GS115, and recombinant proteins were secreted into the culture medium. Supernatants were used directly for screening. Purified CalA (more sn-2 selective) and ROL (strictly sn-1,3 selective) served as controls. Two ether lipid substrates (O-OA-O for sn-2 and O-O-OA for sn-1,3) were used, and hydrolysis products (O-OH-O and O-O-OH) were detected using an Agilent RapidFire system coupled to a triple quadrupole mass spectrometer, enabling analysis in under 10 seconds per sample.Validation confirmed the expected selectivity of ROL and CalA. Using this method, 58 lipases were screened, of which 16 showed activity with regioselectivity similar to CalA. Future work will expand orthologue screening and apply site-saturation mutagenesis to generate larger libraries. These results will improve understanding of lipase regioselectivity and support machine learning models. Workflow improvements are also underway to further reduce processing time.
Comment
To browse or post comments, you must log in.Log in