An Efficient Approach for Lipase-Catalyzed Synthesis of Retinyl Laurate Nutraceutical by Combining Ultrasound Assistance and Artificial Neural Network Optimization
Although retinol is an important nutrient, retinol is highly sensitive to oxidation. At present, some ester forms of retinol are generally used in nutritional supplements because of its stability and bioavailability. However, such esters are commonly synthesized by chemical procedures which are harm...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2017-11-01
|
| Series: | Molecules |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1420-3049/22/11/1972 |
| _version_ | 1811259953125523456 |
|---|---|
| author | Shang-Ming Huang Hsin-Ju Li Yung-Chuan Liu Chia-Hung Kuo Chwen-Jen Shieh |
| author_facet | Shang-Ming Huang Hsin-Ju Li Yung-Chuan Liu Chia-Hung Kuo Chwen-Jen Shieh |
| author_sort | Shang-Ming Huang |
| collection | DOAJ |
| description | Although retinol is an important nutrient, retinol is highly sensitive to oxidation. At present, some ester forms of retinol are generally used in nutritional supplements because of its stability and bioavailability. However, such esters are commonly synthesized by chemical procedures which are harmful to the environment. Thus, this study utilized a green method using lipase as a catalyst with sonication assistance to produce a retinol derivative named retinyl laurate. Moreover, the process was optimized by an artificial neural network (ANN). First, a three-level-four-factor central composite design (CCD) was employed to design 27 experiments, which the highest relative conversion was 82.64%. Further, the optimal architecture of the CCD-employing ANN was developed, including the learning Levenberg-Marquardt algorithm, the transfer function (hyperbolic tangent), iterations (10,000), and the nodes of the hidden layer (6). The best performance of the ANN was evaluated by the root mean squared error (RMSE) and the coefficient of determination (R2) from predicting and observed data, which displayed a good data-fitting property. Finally, the process performed with optimal parameters actually obtained a relative conversion of 88.31% without long-term reactions, and the lipase showed great reusability for biosynthesis. Thus, this study utilizes green technology to efficiently produce retinyl laurate, and the bioprocess is well established by ANN-mediated modeling and optimization. |
| first_indexed | 2024-04-12T18:39:15Z |
| format | Article |
| id | doaj.art-061f7e587d87418593f0823a4018240f |
| institution | Directory Open Access Journal |
| issn | 1420-3049 |
| language | English |
| last_indexed | 2024-04-12T18:39:15Z |
| publishDate | 2017-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Molecules |
| spelling | doaj.art-061f7e587d87418593f0823a4018240f2022-12-22T03:20:50ZengMDPI AGMolecules1420-30492017-11-012211197210.3390/molecules22111972molecules22111972An Efficient Approach for Lipase-Catalyzed Synthesis of Retinyl Laurate Nutraceutical by Combining Ultrasound Assistance and Artificial Neural Network OptimizationShang-Ming Huang0Hsin-Ju Li1Yung-Chuan Liu2Chia-Hung Kuo3Chwen-Jen Shieh4Biotechnology Center, National Chung Hsing University, 250 Kuokuang Road, Taichung 40227, TaiwanDepartment of Chemical Engineering, National Chung Hsing University, 250 Kuo-Kuang Road, Taichung 40227, TaiwanDepartment of Chemical Engineering, National Chung Hsing University, 250 Kuo-Kuang Road, Taichung 40227, TaiwanDepartment of Seafood Science, National Kaohsiung Marine University, 142 Haijhuan Road, Nanzih District, Kaohsiung 81143, TaiwanBiotechnology Center, National Chung Hsing University, 250 Kuokuang Road, Taichung 40227, TaiwanAlthough retinol is an important nutrient, retinol is highly sensitive to oxidation. At present, some ester forms of retinol are generally used in nutritional supplements because of its stability and bioavailability. However, such esters are commonly synthesized by chemical procedures which are harmful to the environment. Thus, this study utilized a green method using lipase as a catalyst with sonication assistance to produce a retinol derivative named retinyl laurate. Moreover, the process was optimized by an artificial neural network (ANN). First, a three-level-four-factor central composite design (CCD) was employed to design 27 experiments, which the highest relative conversion was 82.64%. Further, the optimal architecture of the CCD-employing ANN was developed, including the learning Levenberg-Marquardt algorithm, the transfer function (hyperbolic tangent), iterations (10,000), and the nodes of the hidden layer (6). The best performance of the ANN was evaluated by the root mean squared error (RMSE) and the coefficient of determination (R2) from predicting and observed data, which displayed a good data-fitting property. Finally, the process performed with optimal parameters actually obtained a relative conversion of 88.31% without long-term reactions, and the lipase showed great reusability for biosynthesis. Thus, this study utilizes green technology to efficiently produce retinyl laurate, and the bioprocess is well established by ANN-mediated modeling and optimization.https://www.mdpi.com/1420-3049/22/11/1972retinollipaseretinyl lauratesonicationartificial neural network (ANN)central composite design |
| spellingShingle | Shang-Ming Huang Hsin-Ju Li Yung-Chuan Liu Chia-Hung Kuo Chwen-Jen Shieh An Efficient Approach for Lipase-Catalyzed Synthesis of Retinyl Laurate Nutraceutical by Combining Ultrasound Assistance and Artificial Neural Network Optimization Molecules retinol lipase retinyl laurate sonication artificial neural network (ANN) central composite design |
| title | An Efficient Approach for Lipase-Catalyzed Synthesis of Retinyl Laurate Nutraceutical by Combining Ultrasound Assistance and Artificial Neural Network Optimization |
| title_full | An Efficient Approach for Lipase-Catalyzed Synthesis of Retinyl Laurate Nutraceutical by Combining Ultrasound Assistance and Artificial Neural Network Optimization |
| title_fullStr | An Efficient Approach for Lipase-Catalyzed Synthesis of Retinyl Laurate Nutraceutical by Combining Ultrasound Assistance and Artificial Neural Network Optimization |
| title_full_unstemmed | An Efficient Approach for Lipase-Catalyzed Synthesis of Retinyl Laurate Nutraceutical by Combining Ultrasound Assistance and Artificial Neural Network Optimization |
| title_short | An Efficient Approach for Lipase-Catalyzed Synthesis of Retinyl Laurate Nutraceutical by Combining Ultrasound Assistance and Artificial Neural Network Optimization |
| title_sort | efficient approach for lipase catalyzed synthesis of retinyl laurate nutraceutical by combining ultrasound assistance and artificial neural network optimization |
| topic | retinol lipase retinyl laurate sonication artificial neural network (ANN) central composite design |
| url | https://www.mdpi.com/1420-3049/22/11/1972 |
| work_keys_str_mv | AT shangminghuang anefficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT hsinjuli anefficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT yungchuanliu anefficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT chiahungkuo anefficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT chwenjenshieh anefficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT shangminghuang efficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT hsinjuli efficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT yungchuanliu efficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT chiahungkuo efficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization AT chwenjenshieh efficientapproachforlipasecatalyzedsynthesisofretinyllauratenutraceuticalbycombiningultrasoundassistanceandartificialneuralnetworkoptimization |