Publicação
QSAR modeling studies of a library of Human Tyrosinase inhibitors
| Resumo: | Melanogenesis is the chemical process responsible for synthesizing melanin, which occurs in melanocytes, in subcellular lysosome-like organelles called melanosomes. Melanin plays a vital role in protecting the skin from damage caused by ultraviolet rays. However, excess melanin production or abnormal distribution can cause various pigmentation disorders, such as over-tanning, age spots, and melasma. Skin disorders like these, have prompted the development of skin-whitening compounds to reduce melanin content. Furthermore, inhibition of melanin synthesis is considered a valid therapeutic strategy for treating advanced melanotic melanomas Human tyrosinase (hsTYR) is the most important enzyme involved in the melanogenesis process, as it catalyzes, at least, its first two steps. Tyrosinase from the white button mushroom Agaricus bisporus (abTYR) has been widely available at low cost from commercial sources for several decades, whereas hsTYR is still expensive and difficult to produce. The importance of discovering more and better hsTYR inhibitors has been widely discussed, as when tested against hsTYR, several abTYR inhibitors provide disappointing results, including some of the most extensively used depigmenting compounds now used in dermocosmetics. An in silico methodology that can be used to predict compound bioactivities is QSAR (quantitative structure-activity relationship) modelling. A QSAR model tries to find correlations between a biological activity of interest and molecular descriptors calculated from the compound structure. In this work, a QSAR model was developed to predict hsTYR inhibition activity using the PYTHON computer language and its PyQSAR package. To develop a QSAR model, a library of 196 known hsTYR inhibitors was gathered, and compounds were divided into 6 groups according to their scaffold structure. A total of 33 QSAR models were prepared using different combinations of the defined groups and different pools of molecular descriptors. QSAR model 32 was selected for further use as it presented good statistical robustness and had the highest number of compounds, 41 in total. Of the 28,933 molecular descriptors calculated by the OCHEM platform for the 41 compounds used, PyQSAR selected 4 to be used in the model: C-026; DISSM2C; MaxdssC; WHALES90_Rem. The statistical data obtained after the validation of the QSAR model by cross-validation was excellent, namely the determination coefficient (R2CV=0.9147), the value of the square root of the mean error (RMSE CV=0.1878) and the mean value of the score of the multiple linear regression method (Q2CV=0.8922). This QSAR model originates a mathematical equation that allows the prediction of hsTYR inhibition activity by new compounds with similar structures. A library of natural compounds, with a structure similar to those used to develop QSAR model 32, was created using the COCONUT database of natural compounds. A total of 1,628 natural compounds were gathered, their molecular descriptors were calculated, and the QSAR model 32 equation was applied. The results are displayed on a website and can be viewed by accessing the URL http://esa.ipb.pt/qsar/. The ZINC15 database was used to determine which of the compounds in the developed natural compound library would be available for purchase after predicting the hsTYR inhibitory activity of each compound in the library. A total of 18 different compounds were bought from different companies. To evaluate these compounds experimental ability to inhibit hsTYR and thus validate QSAR model 32, the compounds will be tested against this enzyme. If those compounds activity is confirmed, they may be used in cosmeceutical applications. |
|---|---|
| Autores principais: | Mateus, Cristiano |
| Assunto: | QSAR PYTHON PyQSAR Molecular descriptor Melanin hsTYR abTYR OCHEM COCONUT ZINC15 |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Instituto Politécnico de Bragança |
| Idioma: | inglês |
| Origem: | Biblioteca Digital do IPB |
| Resumo: | Melanogenesis is the chemical process responsible for synthesizing melanin, which occurs in melanocytes, in subcellular lysosome-like organelles called melanosomes. Melanin plays a vital role in protecting the skin from damage caused by ultraviolet rays. However, excess melanin production or abnormal distribution can cause various pigmentation disorders, such as over-tanning, age spots, and melasma. Skin disorders like these, have prompted the development of skin-whitening compounds to reduce melanin content. Furthermore, inhibition of melanin synthesis is considered a valid therapeutic strategy for treating advanced melanotic melanomas Human tyrosinase (hsTYR) is the most important enzyme involved in the melanogenesis process, as it catalyzes, at least, its first two steps. Tyrosinase from the white button mushroom Agaricus bisporus (abTYR) has been widely available at low cost from commercial sources for several decades, whereas hsTYR is still expensive and difficult to produce. The importance of discovering more and better hsTYR inhibitors has been widely discussed, as when tested against hsTYR, several abTYR inhibitors provide disappointing results, including some of the most extensively used depigmenting compounds now used in dermocosmetics. An in silico methodology that can be used to predict compound bioactivities is QSAR (quantitative structure-activity relationship) modelling. A QSAR model tries to find correlations between a biological activity of interest and molecular descriptors calculated from the compound structure. In this work, a QSAR model was developed to predict hsTYR inhibition activity using the PYTHON computer language and its PyQSAR package. To develop a QSAR model, a library of 196 known hsTYR inhibitors was gathered, and compounds were divided into 6 groups according to their scaffold structure. A total of 33 QSAR models were prepared using different combinations of the defined groups and different pools of molecular descriptors. QSAR model 32 was selected for further use as it presented good statistical robustness and had the highest number of compounds, 41 in total. Of the 28,933 molecular descriptors calculated by the OCHEM platform for the 41 compounds used, PyQSAR selected 4 to be used in the model: C-026; DISSM2C; MaxdssC; WHALES90_Rem. The statistical data obtained after the validation of the QSAR model by cross-validation was excellent, namely the determination coefficient (R2CV=0.9147), the value of the square root of the mean error (RMSE CV=0.1878) and the mean value of the score of the multiple linear regression method (Q2CV=0.8922). This QSAR model originates a mathematical equation that allows the prediction of hsTYR inhibition activity by new compounds with similar structures. A library of natural compounds, with a structure similar to those used to develop QSAR model 32, was created using the COCONUT database of natural compounds. A total of 1,628 natural compounds were gathered, their molecular descriptors were calculated, and the QSAR model 32 equation was applied. The results are displayed on a website and can be viewed by accessing the URL http://esa.ipb.pt/qsar/. The ZINC15 database was used to determine which of the compounds in the developed natural compound library would be available for purchase after predicting the hsTYR inhibitory activity of each compound in the library. A total of 18 different compounds were bought from different companies. To evaluate these compounds experimental ability to inhibit hsTYR and thus validate QSAR model 32, the compounds will be tested against this enzyme. If those compounds activity is confirmed, they may be used in cosmeceutical applications. |
|---|