Machine learning guided prediction of liquid chromatography-mass spectrometry ionization efficiency for genotoxic impurities in pharmaceutical products.

The limitation and control of genotoxic impurities (GTIs) has continued to receive attention from pharmaceutical companies and authorities for several decades. Because GTIs have the ability to damage deoxyribonucleic acid (DNA) and the potential to cause cancer, low-level quantitation is required to protect patients. A quick and easy method of determining the liquid chromatography-mass spectrometry (LC/MS) conditions for high-sensitivity analysis of GTIs may prospectively accelerate pharmaceutical development. In this study, a quantitative structure-property relationship (QSPR) model was developed for predicting the ionization efficiency of compounds using liquid-chromatography-mass spectrometry (LC/MS) parameters and molecular descriptors. Before implementing the QSPR prediction model, linear regression analysis was performed to model the relationship between the ionization efficiency and the LC/MS parameters for each compound. Comparison of the predicted peak areas with the experimentally observed peak areas showed good agreement based on the coefficient of determination (R2 > 0.96). The machine learning-based QSPR approach begins with computation of the molecular descriptors expressing the physicochemical properties of a compound, followed by a genetic algorithm-based feature selection. Linear and nonlinear regression were performed, and support vector machine (SVM) was selected as the best machine learning algorithm for the prediction. The SVM algorithm was developed and optimized using 1031 experimental data points for nine compounds, including well-known GTIs. Validation of the model by comparison of the predicted and observed relative ionization efficiencies (RIE) showed a high coefficient of determination (R2 = 0.96) and low root mean squared error value (RMSE = 0.118). Finally, this established prediction model was applied to hydrophilic interaction liquid chromatography coupled with MS for a new compound in new mobile phase compositions and new MS parameter settings. The RMSE of the predicted versus observed RIE was 0.203. This prediction accuracy was sufficient to determine the starting point of the LC/MS method development. The methodology demonstrated in this study can be used to determine the LC/MS conditions for high sensitivity analysis of GTIs.

Mass spectrometry data from proteomic analysis of human skin keratins after exposure to UV radiation.

A mass spectrometry (MS)-based proteomic methodology was employed to monitor oxidative modifications in keratins, the main constituents of human skin ("Non-invasive proteomic analysis of human skin keratins: screening of methionine oxidation in keratins by mass spectrometry" [1], "UV irradiation-induced methionine oxidation in human skin keratins: mass spectrometry-based non-invasive proteomic analysis" [2]). Human skin proteins were obtained non-invasively by tape stripping and solubilized in sodium dodecyl sulfate (SDS) buffer, followed by purification and digestion using the filter-aided sample preparation method. The tryptic peptides were then analyzed by liquid chromatography (LC)/electrospray ionization (ESI)-MS, tandem MS (MS/MS), and LC/ESI-selected reaction monitoring (SRM)/MS. The MS/MS data were generated to confirm amino acid sequences and oxidation sites of tryptic peptides D(290)VDGAYMTK(298) (P1) and N(258)MQDMVEDYR(267) (P2), which contain the most susceptible oxidation sites (Met(259), Met(262), and Met(296) in K1 keratin) upon UVA irradiation [2]. Subsequently, quantitative determination of the relative oxidation levels of P1 and P1 [2] was achieved by LC/ESI-SRM/MS analyses of P1 and P2 together with their oxidized forms after exposure to UVA radiation or treatment with hydrogen peroxide (H2O2).

UV irradiation-induced methionine oxidation in human skin keratins: Mass spectrometry-based non-invasive proteomic analysis.

Ultraviolet (UV) radiation is the major environmental factor that causes oxidative skin damage. Keratins are the main constituents of human skin and have been identified as oxidative target proteins. We have recently developed a mass spectrometry (MS)-based non-invasive proteomic methodology to screen oxidative modifications in human skin keratins. Using this methodology, UV effects on methionine (Met) oxidation in human skin keratins were investigated. The initial screening revealed that Met(259), Met(262), and Met(296) in K1 keratin were the most susceptible oxidation sites upon UVA (or UVB) irradiation of human tape-stripped skin. Subsequent liquid chromatography/electrospray ionization-MS and tandem MS analyses confirmed amino acid sequences and oxidation sites of tryptic peptides D(290)VDGAYMTK(298) (P1) and N(258)MQDMVEDYR(267) (P2). The relative oxidation levels of P1 and P2 increased in a time-dependent manner upon UVA irradiation. Butylated hydroxytoluene was the most effective antioxidant for artifactual oxidation of Met residues. The relative oxidation levels of P1 and P2 after UVA irradiation for 48 h corresponded to treatment with 100mM hydrogen peroxide for 15 min. In addition, Met(259) was oxidized by only UVA irradiation. The Met sites identified in conjunction with the current proteomic methodology can be used to evaluate skin damage under various conditions of oxidative stress. BIOLOGICAL SIGNIFICANCE: We demonstrated that the relative Met oxidation levels in keratins directly reflected UV-induced damages to human tape-stripped skin. Human skin proteins isolated by tape stripping were analyzed by MS-based non-invasive proteomic methodology. Met(259), Met(262), and Met(296) in K1 keratin were the most susceptible oxidation sites upon UV irradiation. Met(259) was oxidized by only UVA irradiation. Quantitative LC/ESI-SRM/MS analyses confirmed a time-dependent increase in the relative oxidation of target peptides (P1 and P2) containing these Met residues, upon UVA irradiation of isolated human skin. The relative oxidation levels of P1 and P2 along with the current proteomic methodology could be applied to the assessment of oxidative stress levels in skin after exposure to sunlight.

Non-invasive proteomic analysis of human skin keratins: screening of methionine oxidation in keratins by mass spectrometry.

Keratins are the main constituent of human skin and have been identified as major oxidative target proteins. However, there has been a lack of studies aimed at identifying the oxidation sites of keratins because of the difficulties associated with their insolubility and handling. Here, we introduce a mass spectrometry (MS)-based proteomic methodology to screen oxidative modifications in human skin keratins. Human skin proteins were obtained non-invasively by tape stripping and solubilized in SDS buffer, followed by purification and digestion using the modified filter-aided sample preparation method. The tryptic peptides were then analyzed by MALDI-TOF/MS, LC-ESI/MS, and MS/MS. PMF analyses have identified keratins K1 and K10 as the major proteins of human skin. Met(259), Met(262), Met(296), and Met(469), located in the α-helical rod domain of K1, were the most susceptible sites to oxidation induced by hydrogen peroxide in vitro and in vivo. Our results indicate a potential use of the identified methionine residues as biomarkers of oxidative skin damage. The present methodology is the first MS-based approach to detecting oxidative modifications in keratins obtained directly from human skin and can be easily applied to the monitoring of other keratin modifications in various skin conditions.