Development of QSAR models for microsomal stability: identification of good and bad structural features for rat,human and mouse microsomal stability

High throughput microsomal stability assays have been widely implemented in drug discovery and many companies have accumulated experimental measurements for thousands of compounds. Such datasets have been used to develop in silico models to predict metabolic stability and guide the selection of promising candidates for synthesis. This approach has proven most effective when selecting compounds from proposed virtual libraries prior to synthesis. However, these models are not easily interpretable at the structural level, and thus provide little insight to guide traditional synthetic efforts. We have developed global classification models of rat, mouse and human liver microsomal stability using in-house data. These models were built with FCFP_6 fingerprints using a Naïve Bayesian classifier within Pipeline Pilot. The test sets were correctly classified as stable or unstable with satisfying accuracies of 78, 77 and 75% for rat, human and mouse models, respectively. The prediction confidence was assigned using the Bayesian score to assess the applicability of the models. Using the resulting models, we developed a novel data mining strategy to identify structural features associated with good and bad microsomal stability. We also used this approach to identify structural features which are good for one species but bad for another. With these findings, the structure-metabolism relationships are likely to be understood faster and earlier in drug discovery.     

In silico models for the prediction of dose-dependent human hepatotoxicity

The liver is extremely vulnerable to the effects of xenobiotics due to its critical role in metabolism. Drug-induced hepatotoxicity may involve any number of different liver injuries, some of which lead to organ failure and, ultimately, patient death. Understandably, liver toxicity is one of the most important dose-limiting considerations in the drug development cycle, yet there remains a serious shortage of methods to predict hepatotoxicity from chemical structure. We discuss our latest findings in this area and present a new, fully general in silico model which is able to predict the occurrence of dose-dependent human hepatotoxicity with greater than 80% accuracy. Utilizing an ensemble recursive partitioning approach, the model classifies compounds as toxic or non-toxic and provides a confidence level to indicate which predictions are most likely to be correct. Only 2D structural information is required and predictions can be made quite rapidly, so this approach is entirely appropriate for data mining applications and for profiling large synthetic and/or virtual libraries.     

Comparison of in silico models for prediction of Daphnia magna acute toxicity

Eight in silico modelling packages were evaluated and compared for the prediction of Daphnia magna acute toxicity from the viewpoint of the European legislation on chemicals, REACH. We tested the following models: Discovery Studio (DS) TOPKAT, ACD/Tox Suite, ADMET Predictor?, ECOSAR (Ecological Structure Activity Relationships), TerraQSAR?, T.E.S.T. (Toxicity Estimation Software Tool) and two models implemented in VEGA on 480 industrial compounds for 48-h median lethal concentrations (LC₅₀) to D. magna, matching them with experimental values. The quality of the estimates was compared using a standard statistical review and an additional classification approach in which the hazard predictions were grouped using well-defined regulatory criteria. The regression parameters, correlation coefficient being the most influential, showed that four models (ADMET Predictor?, DS TOPKAT, TerraQSAR? and VEGA DEMETRA) had similar reliability. These performed better than the others, but the coefficient of determination was still low (r² around 0.6), considering that at least half the predicted compounds were inside the training sets. Additionally, we grouped the results in four defined toxicity classes. TerraQSAR? gave 60% of correct classifications, followed by DS TOPKAT, ADMET Predictor? and VEGA DEMETRA, with 56%, 54% and 48%, respectively. These results highlight the challenges associated with developing reliable and easily applied acceptability criteria for the regulatory use of QSAR models to D. magna acute toxicity.     

Element content and element correlations in Chinese human liver

The amounts of the 19 elements As, Br, Ca, Cd, Ce, Co, Cr, Cs, Fe, K, La, Mo, Na, Rb, Sb, Sc, Se, Sm, and Zn in 92 lyophilized autopsy human liver samples from normal subjects have been analyzed by instrumental neutron-activation analysis (INAA). For intercomparison and quality control ten samples were independently analyzed in two institutes, the Institute of High Energy Physics in China and the Joef Stefan Institute in Slovenia. Most of the element contents determined by the two institutes were in quite good agreement, even though different experimental conditions were applied, indicating the reliability of the analytical results. Analysis of the chemical species of mercury present in the ten liver samples was also performed in Slovenia. Possible differences between the element content of male and female liver samples were studied by means of Students t-test, but significant differences were found only for Ce, Co, Fe, La, Mo, and Zn. The results obtained were also compared with those reported from other areas of the world; no appreciable differences were observed. Correlation among the various elements in the human liver samples was studied using multivariate statistics. It was found that there was relatively close correlation between some elements, for example As–Fe, Ca–Fe, Cd–Co, Cd–Zn, Mo–Zn, Co–Se, Cs–Rb, Br–Rb, Sc–Sm, La–Sm, La–Ce, etc.; these correlations could be rationally explained by the similarity of the electronic structures of the elements and/or their physiological functions in the human body.     

Knowledge acquisition and development of accurate rules for predicting protein stability changes

Knowing the mechanisms by which protein stability change is one of the most important and valuable tasks in molecular biology. The conventional methods of predicting protein stability changes mainly focus on improving prediction accuracy. However, it is desirable to extract domain knowledge from large databases that is beneficial to accurate prediction of the protein stability change. This paper presents an interpretable prediction tree method (named iPTREE) that produces explanatory rules to explore hidden knowledge accompanied with high prediction accuracy and consequently analyzes the factors influencing the protein stability changes. To evaluate iPTREE and the knowledge upon protein stability changes, a thermodynamic dataset consisting of 1615 mutants led by single point mutation from ProTherm is adopted. Being as a predictor for protein stability changes, the rule-based approach can achieve a prediction accuracy of 87%, which is better than other methods based on artificial neural networks (ANN) and support vector machines (SVM). Besides, these methods lack the ability in biological knowledge discovery. The human-interpretable rules produced by iPTREE reveal that temperature is a factor of concern in predicting protein stability changes. For example, one of interpretable rules with high support is as follows: if the introduced residue type is Alanine and temperature is between 4 °C and 40 °C, then the stability change will be negative (destabilizing). The present study demonstrates that iPTREE can easily be used in the application of protein stability changes where one requires more understandable knowledge.     

Improvements in protein identification confidence and proteome coverage for human liver proteome study by coupling a parallel mass spectrometry/mass spectrometry analysis with multi-dimensional chromatography separation

Recently, matrix-assisted laser desorption ionization (MALDI) technique has been shown to be complementary to electrospray ionization (ESI) with respect to the population of peptides and proteins that can be detected. In this study, we tried to hyphenate MALDI-TOF-TOF-MS and ESI-QUADRUPOLE-TOF-MS with a single 2D liquid chromatography for complicated protein sample analysis. The effluents of RPLC were split into two parts for the parallel MS/MS detection. After optimizing the operation conditions in LC separation and MS identification, a total of 1149 proteins were identified from the global lysate of normal human liver (NHL) tissue. Compared to the single MS/MS detection, the combined analysis increased the number of proteins identified (more than 25%) and enhanced the protein identification confidence. Proteins identified were categorized and analyzed based upon their cellular location, biological process and molecular function. The identification results demonstrated the application potential of a parallel MS/MS analysis coupled with multi-dimensional LC separation for complicated protein sample identification, especially for proteome analysis, such as human tissues or cells extracts.     

Data-driven machine learning models for quick prediction of thermal stability properties of OLED materials

Organic light-emitting diode (OLED) materials have exhibited a wide range of applications. However, the further development and commercialization of OLEDs requires higher quality OLED materials, including materials with a high thermal stability. Thermal stability is associated with the glass transition temperature (T_g) and decomposition temperature (T_d), but experimental determinations of these two important properties generally involve a time-consuming and laborious process. Thus, the development of a quick and accurate prediction tool is highly desirable. Motivated by the challenge, we explored machine learning (ML) by constructing a new dataset with more than 1,000 samples collected from a wide range of literature, through which ensemble learning models were explored. Models trained with the LightGBM algorithm exhibited the best prediction performance, where the values of mean absolute error, root mean squared error, and R² were 17.15 K, 24.63 K, and 0.77 for T_g prediction and 24.91 K, 33.88 K, and 0.78 for T_d prediction. The prediction performance and the generalization of the ML models were further tested by two applications, which also exhibited satisfactory results. Experimental validation further demonstrated the reliability and the practical potential of the ML-based models. In order to extend the practical application of the ML-based models, an online prediction platform was constructed. This platform includes the optimal prediction models and all the thermal stability data under study, and it is freely available at http://www.oledtppxmpugroup.com. We expect that this platform will become a useful tool for experimental investigation of T_g and T_d, accelerating the design of OLED materials with desired properties.     

Proteome analysis of human liver tumor tissue by two-dimensional gel electrophoresis and matrix assisted laser desorption/ionization-mass spectrometry for identification of disease-related proteins

Hepatocellular carcinoma (HCC) is a common malignancy worldwide and is a leading cause of death. To contribute to the development and improvement of molecular markers for diagnostics and prognostics and of therapeutic targets for the disease, we have largely expanded the currently available human liver tissue maps and studied the differential expression of proteins in normal and cancer tissues. Reference two-dimensional electrophoresis (2-DE) maps of human liver tumor tissue include labeled 2-DE images for total homogenate and soluble fraction separated on pH 3-10 gels, and also images for soluble fraction separated on pH 4-7 and pH 6-9 gels for a more detailed map. Proteins were separated in the first dimension by isoelectric focusing on immobilized pH gradient (IPG) strips, and by 7.5-17.5% gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels in the second dimension. Protein identification was done by peptide mass fingerprinting with delayed extraction-matrix assisted laser desorption/ionization-time of flight-mass spectrometry (DE-MALDI-TOF-MS). In total, 212 protein spots (117 spots in pH 4-7 map and 95 spots in pH 6-9) corresponding to 127 different polypeptide chains were identified. In the next step, we analyzed the differential protein expression of liver tumor samples, to find out candidates for liver cancer-associated proteins. Matched pairs of tissues from 11 liver cancer patients were analyzed for their 2-DE profiles. Protein expression was comparatively analyzed by use of image analysis software. Proteins whose expression levels were different by more than three-fold in at least 30% (four) of the patients were further analyzed. Numbers of protein spots overexpressed or underexpressed in tumor tissues as compared with nontumorous regions were 9 and 28, respectively. Among these 37 spots, 1 overexpressed and 15 underexpressed spots, corresponding to 11 proteins, were identified. The physiological significance of the differential expressions is discussed.     

Simple high-performance liquid chromatographic assay for melphalan in perfusate, rat liver and tumour tissue

A simple, sensitive and selective reversed-phase liquid chromatographic assay has been developed and validated for the anti-cancer agent melphalan in perfusate, liver and tumour tissue originating from isolated rat liver perfusion studies. Melphalan was extracted from the matrix using ice-cold methanol. The drug and the internal standard, propylparaben, were detected using ultraviolet absorbance at 262 nm. The assay has been validated in the 0.05-25 microg/mL range for perfusate; the lower limit of quantification (LLQ) is 0.05 microg/mL in perfusate and 0.25 ng/mg in liver and tumour tissues. Accuracies ranged from 89 to 110% and the inter-assay precisions were all below 15% (20% at the LLQ). Melphalan in a biological matrix has to be processed between 0 and 4 degrees C and is stable under all relevant processing and storage conditions tested. The assay has been exhaustively used in isolated liver perfusion studies with the drug demonstrating its applicability.     

Assessment of in silico models for acute aquatic toxicity towards fish under REACH regulation

We evaluated the performance of eight QSAR in silico modelling packages (ACD/ToxSuite?, ADMET Predictor?, DEMETRA, ECOSAR, TerraQSAR?, Toxicity Estimation Software Tool, TOPKAT? and VEGA) for acute aquatic toxicity towards two species of fish: Fathead Minnow and Rainbow Trout. For the Fathead Minnow, we compared model predictions for 567 substances with the corresponding experimental values for 96-h median lethal concentrations (LC50). Some models gave good results, with r² up to 0.85. We also classified the predictions of all the models into four toxicity classes defined by CLP. This permitted us to assess other parameters, such as the percentage of correct predictions for each class. Then we used a set of 351 substances with toxicity data towards Rainbow Trout (96-h LC50). In this case the predictability was unacceptable for all the in silico models. The calculated r² gave poor correlations (≤0.53). Another analysis was performed according to chemical classes and for mode of action. In the first case, all the classes show a high percentage of correct predictions, in the second case only narcotics and polar narcotics were predicted with good confidence. The results indicate the possibility of using in silico methods to estimate aquatic toxicity within REACH regulation, after careful evaluation.     

Assessing and improving the stability of chemometric models in small sample size situations

Small sample sizes are very common in multivariate analysis. Sample sizes of 10–100 statistically independent objects (rejects from processes or loading dock analysis, or patients with a rare disease), each with hundreds of data points, cause unstable models with poor predictive quality. Model stability is assessed by comparing models that were built using slightly varying training data. Iterated k-fold cross-validation is used for this purpose. Aggregation stabilizes models. It is possible to assess the quality of the aggregated model without calculating further models. The validation and aggregation methods investigated in this study apply to regression as well as to classification. These techniques are useful for analyzing data with large numbers of variates, e.g., any spectral data like FT-IR, Raman, UV/VIS, fluorescence, AAS, and MS. FT-IR images of tumor tissue were used in this study. Some tissue types occur frequently, while some are very rare. They are classified using LDA. Initial models were severely unstable. Aggregation stabilizes the predictions. The hit rate increased from 67% to 82%.     

Polyacrylamide gel electrophoresis followed by sodium dodecyl sulfate gradient polyacrylamide gel electrophoresis for the study of the dimer to monomer transition of human transthyretin

Familial amyloidotic polyneuropathy (FAP) is caused by mutations which destabilize transthyretin (TTR) and facilitate the aggregation into extracellular amyloid fibrils preferentially in peripheral nerve and heart tissues. Therapeutic and preventive trials for FAP at the plasma TTR level require a careful study of the destabilization of TTR under variable conditions. We have developed a simple double one-dimensional (D1-D) electrophoretic procedure with polyacrylamide gel electrophoresis (PAGE) followed by sodium dodecylsulfate (SDS) gradient PAGE to study the dimer to monomer transition. TTR is first isolated by PAGE from other plasma proteins. The gel strip containing the TTR fraction is incubated in 2% SDS under varying conditions of temperature, buffer composition, pH, and additives like urea and/or a sulfhydryl-reactive agent, followed by SDS-gradient PAGE for the separation of TTR dimers and monomers. We demonstrate that an unidirectional dimer to monomer transition of normal TTR is achieved at 70-80 degrees C in neutral to mild alkaline buffers or at 37 degrees C and slightly acidic pH (6-7). Addition of urea favors the transition into monomers. Amyloidogenic mutations like amyloidogenic TTR (ATTR)-V30M or ATTR-I107V favor the transition into monomers in buffer systems close to the physiological pH of human plasma. We conclude that this finding has to be considered by any hypothesis on ATTR-derived amyloidogenesis.     

Using a new solvation model for thermodynamic study on the interaction of nickel with human growth hormone

Thermodynamics of the interaction between Ni²⁺ and human growth hormone (hGH) were determined at 27 °C in Nail solution by isothermal titration calorimetry. A new method to predict protein penetration and the effect of metal ions on the stability of proteins is introduced. The new solvation model was used to reproduce the enthalpies of Ni²⁺-hGH interaction over the whole range of Ni²⁺ concentrations. The solvation parameters recovered from the new equation, attributed to the structural change of hGH and its biological activity.     

On-line SFE-GC for determination of PCBs in human milk and blood serum

On-line coupled supercritical fluid extraction and gas chromatography (SFE-GC) has been utilized for the determination of PCBs and other organochlorine compounds in human milk and blood serum. The procedure involved preconcentration of the sample on C₁₈-silica sorbent in an extraction cell: after precipitation of the proteins up to 20 ml of human milk was concentrated on 0.5 g of sorbent. Serum (up to 5 ml) was applied to the C₁₈ material without pretreatment. Basic alumina was utilized as a selective adsorbent for lipids in the on-line SFE-GC system. The method was used to analyze milk and serum spiked with 0.5 and 10 ng of Aroclor 1260 and the results compared with those obtained by liquid–liquid extraction of serum.     

Development and validation method for determination of Paroxetine and its metabolites by nonaqueous capillary electrophoresis in human urine. Experimental design for evaluating the ruggedness of the method

A simple, rapid, and sensitive procedure using nonaqueous capillary electrophoresis (NACE) to measure Paroxetine (one of the mostly used antidepressants for mental diseases treatment) and three metabolites has been developed and validated. Optimum separation of paroxetine and metabolites was obtained on a 57 cm x 75 microm capillary using a nonaqueous buffer system of 9:1 methanol-acetonitrile containing 25 mM ammonium acetate and 1% acetic acid, with temperature and voltage of 25 degrees C and 15 kV, respectively, and hydrodynamic injection. Fluoxetine was used as an internal standard. Good results were obtained for different aspects including stability of the solutions, linearity, accuracy, and precision. Detection limits between 9.3 and 23.1 microg.L(-1) were obtained for paroxetine and its metabolites. A ruggedness test of the method was carried out using the Plackett-Burman fractional factorial model with a matrix of 15 experiments. This method has been used to determine paroxetine and its main metabolite B at clinically relevant levels in human urine. Prior to NACE determination, the samples were purified and enriched by means of an extraction-preconcentration step with a preconditioned C18 cartridge and eluting the compounds with methanol.     

Development of a micro-planar Ag/AgCl quasi-reference electrode with long-term stability for an amperometric glucose sensor

This paper reports a micro-planar Ag/AgCl quasi-reference electrode (QRE) with long-term stability which is characterized by both long-term potential stability and practical immunity to interference species, and which has been applied for use with an amperometric glucose sensor for plasma glucose. For fabrication, we coated a silver/silver chloride (Ag/AgCl) electrode first with γ-aminopropyltriethoxysilane (γ-APTES) and then with perfluorocarbon polymer (PFCP). Tests demonstrate the new electrode’s ability to remain stable over an 82-day period in 150 mM KCl, and also show its imperviousness to the effects of interference species (1 mM KI and 1 mM KBr), pH, and serum. Furthermore, in tests for glucose concentrations in plasma samples, a good correlation coefficient, 0.954 (n=30, Y=1.02X+0.20), was demonstrated between results obtained with a clinical analyzer and those obtained with an amperometric glucose sensor that used the developed Ag/AgCl QRE, showing that the Ag/AgCl QRE functions well as a reference electrode for plasma samples.     

客体分子所处晶穴结构不同对甲烷水合物稳定性的影响

采用分子动力学方法模拟了SⅠ型甲烷水合物受热分解微观过程,并对水合物分解过程中不同晶穴结构内客体分子对甲烷水合物稳定性的作用进行了研究.通过最终构象、均方位移和势能等性质的变化规律对分别缺失大晶穴和小晶穴中客体分子的2种水合物体系随模拟温度升高稳定性的变化进行了分析.模拟结果显示,随温度的上升,水合物稳定性逐渐下降直至彻底分解;而水合物分解速度与2种晶穴各自部分晶穴占有率相关,不能简单的通过整体晶穴占有率表示.对比相同注热过程中2种水合物体系分解状况,发现位于大晶穴内的客体分子对水合物稳定性影响更大,缺失大晶穴内客体分子的水合物更容易随温度升高而分解.     

Ultra-performance liquid chromatographic-electrospray mass spectrometric determination (UPLC-ESI-MS) of O-demethylated metabolite of paeonol in vitro: Assay development, human liver microsome activities and species differences

A simple and sensitive method for determination of the O-demethylation activity of rat, dog, minipig, and human liver micrsomes toward paeonol using ultra-performance liquid chromatography with mass detection (UPLC-MS) has been developed. The method uses chemically synthesized O-demethylated metabolite of paeonol (2,4-dihydroxyacetophenone, DHA) as a standard for method validation. Validation was done with respect to specificity, linearity, detection limit, recovery, stability, precision and accuracy. The chromatographic separation was achieved on a UPLC BEH C₁₈ column (50 mm × 2.1 mm i.d., 1.7 μm), with phase of acetonitrile-water (ratio 30:70). Selective ion reaction (SIR) monitor was specific for paeonol, DHA and I.S. The method was specific since there were no interference peaks from the reaction matrix. The calibration curve for DHA was linear from 0.5-100 μM with r² = 0.9999. The newly developed method has good precision and accuracy. The method was successfully used to determine the kinetics of DHA activities toward paeonol in liver microsomes from different species. Dog liver microsomes (DLMs) were the most active in paeonol O-demethylation (709.7 pmol/min/mg protein) followed by rat liver microsomes (RLMs) (579.6 pmol/min/mg protein), HLMs (569.3 pmol/min/mg protein), and then minipig liver microsomes (PLMs) (417.3 pmol/min/mg protein). The developed method was appropriated for rapid screening paeonol O-demethylation activity in liver microsomes from different species.     

测定人体尿液中苄基丙二酸的免疫分析新方法研究

将苄基丙二酸(BA)与牛血清白蛋白通过活化酯法偶联制备人工免疫原,然后免疫新西兰大白兔制得抗苄基丙二酸多克隆抗体;使用该抗体建立了测定人体尿液中的苄基丙二酸的间接竞争酶联免疫吸附方法(ic-ELISA),方法线性范围为1.0×10-2～1.0×103ng/mL,最低检测浓度为0.01 ng/mL,板内差异为5.5%,板间差异在3.8%～7.8%的范围内;用于实际尿样中苄基丙二酸的测定,回收率为87.8%～115.1%。