Direct Comparison of the Prediction of the Unbound Brain-to-Plasma Partitioning Utilizing Machine Learning Approach and Mechanistic Neuropharmacokinetic Model

The mechanistic neuropharmacokinetic (neuroPK) model was established to predict unbound brain-to-plasma partitioning (K_p,uu,brain) by considering in vitro efflux activities of multiple drug resistance 1 (MDR1) and breast cancer resistance protein (BCRP). Herein, we directly compare this model to a computational machine learning approach utilizing physicochemical descriptors and efflux ratios of MDR1 and BCRP-expressing cells for predicting K_p,uu,brain in rats. Two different types of machine learning techniques, Gaussian processes (GP) and random forest regression (RF), were assessed by the time and cluster-split validation methods using 640 internal compounds. The predictivity of machine learning models based on only molecular descriptors in the time-split dataset performed worse than the cluster-split dataset, whereas the models incorporating MDR1 and BCRP efflux ratios showed similar predictivity between time and cluster-split datasets. The GP incorporating MDR1 and BCRP in the time-split dataset achieved the highest correlation (R² = 0.602). These results suggested that incorporation of MDR1 and BCRP in machine learning is beneficial for robust and accurate prediction. K_p,uu,brain prediction utilizing the neuroPK model was significantly worse compared to machine learning approaches for the same dataset. We also investigated the predictivity of K_p,uu,brain using an external independent test set of 34 marketed drugs. Compared to machine learning models, the neuroPK model showed better predictive performance with R² of 0.577. This work demonstrates that the machine learning model for K_p,uu,brain achieves maximum predictive performance within the chemical applicability domain, whereas the neuroPK model is applicable more widely beyond the chemical space covered in the training dataset.

Prediction of Oral Pharmacokinetics Using a Combination of In Silico Descriptors and In Vitro ADME Properties

Accurate prediction of oral pharmacokinetics remains challenging. This study investigated quantitative approaches for the prediction of the area under the plasma concentration-time curve after oral administration (AUC_p,oral) to rats using the in vitro-in vivo extrapolation (IVIVE), in silico model using machine learning approaches and the combination of the in silico model and in vitro data. A set of 595 structurally diverse compounds with determined AUC_p,oral at 1 mg/kg, in vitro intrinsic clearance (CL_int), an unbound fraction in plasma (f_u,p) in rats, and kinetic solubility at pH 6.8 was used for this assessment. Prediction models developed by two different types of machine learning techniques (i.e., random forest regression and Gaussian processes) were evaluated using three validation methods implementing the time and cluster-split training and test set and fivefold cross-validation. The developed machine learning models have a square of correlation coefficient (R²) in the range of 0.381-0.685 with 33-45% of the compounds being predicted within 2-fold of the observed AUC_p,oral value. The predictivity was improved by incorporating CL_int, f_u,p, and solubility as explanatory variables with R² = 0.554-0.743. In cases where extraction by the liver is the main elimination pathway and intestinal extraction is negligible, AUC_p,oral can be expressed by dose, CL_int, and f_u,p based on a well-stirred model. By using this conventional IVIVE approach, only 1.7-5.0% of compounds were predicted within the 2-fold error with R² = 0.354-0.487. Two empirical scaling factors (ESFs) determined by linear regression analysis and machine learning approaches improved the predictivity of AUC_p,oral with 33-44% predicted within twofold variability. The IVIVE using ESF predicted by random forest regression showed better predictivity of AUC_p,oral with R² = 0.471-0.618, while it still showed lower predictivity than machine learning approaches applied directly to AUC_p,oral prediction. This study demonstrated that the combination of in silico and in vitro parameters is useful to improve the predictivity of the machine learning model for rat AUC_p,oral and supports consideration for predicting AUC_p,oral for human and other non-clinical species in a similar manner.

In Vitro Metabolism of Slowly Cleared G Protein-Coupled Receptor 139 Agonist TAK-041 Using Rat, Dog, Monkey, and Human Hepatocyte Models (HepatoPac): Correlation with In Vivo Metabolism

Hepatic metabolism of low-clearance compound TAK-041 was studied in two different in vitro model systems using rat, dog, monkey, and human suspended cryopreserved hepatocytes and HepatoPac micropatterned coculture model primary hepatocytes. The aim of this work was to investigate the most appropriate system to assess the biotransformation of TAK-041, determine any notable species difference in the rate and in the extent of its metabolic pathways, and establish correlation with in vivo metabolism. TAK-041 exhibited very low turnover in suspended cryopreserved hepatocyte suspensions for all species, with no metabolites observed in human hepatocytes. However, incubations conducted for up to 14 days in the HepatoPac model resulted in more robust metabolic turnover. The major biotransformation pathways of TAK-041 proceed via hydroxylation on the benzene ring fused to the oxotriazine moiety and subsequent sulfate, glucuronide, and glutathione conjugation reactions. The glutathione conjugate of TAK-041 undergoes further downstream metabolism to produce the cysteine S-conjugate, which then undergoes N-acetylation to mercapturic acid and/or conversion to β-lyase-derived thiol metabolites. The minor biotransformation pathways include novel ring closure and hydrolysis, hydroxylation, oxidative N-dealkylation, and subsequent reduction. The HepatoPac model shows a notable species difference in the rate and in the extent of metabolic pathways of TAK-041, with dogs having the fastest metabolic clearance and humans the slowest. Furthermore, the model shows its suitability for establishing correlation with in vivo metabolism of low-turnover and extensively metabolized compounds such as TAK-041, displaying an extensive and unusual downstream sequential β-lyase-derived thiol metabolism in preclinical species and human. SIGNIFICANCE STATEMENT: This study investigated the most appropriate in vitro system to assess the biotransformation of the low-turnover and extensively metabolized compound TAK-041, determine any notable species difference in the rate and in the extent of its metabolic pathways, and establish correlation with in vivo metabolism. The HepatoPac model was identified and showed its suitability for species comparison and establishing correlation, with in vivo metabolism displaying an extensive and unusual downstream sequential β-lyase-derived thiol metabolism in preclinical species and human.

Direct Comparison of Total Clearance Prediction: Computational Machine Learning Model versus Bottom-Up Approach Using In Vitro Assay

The in vitro-in vivo extrapolation (IVIVE) approach for predicting total plasma clearance (CL_tot) has been widely used to rank order compounds early in discovery. More recently, a computational machine learning approach utilizing physicochemical descriptors and fingerprints calculated from chemical structure information has emerged, enabling virtual predictions even earlier in discovery. Previously, this approach focused more on in vitro intrinsic clearance (CL_int) prediction. Herein, we directly compare these two approaches for predicting CL_tot in rats. A structurally diverse set of 1114 compounds with known in vivo CL_tot, in vitro CL_int, and plasma protein binding was used as the basis for this evaluation. The machine learning models were assessed by validation approaches using the time- and cluster-split training and test sets, and five-fold cross validation. Assessed by five-fold validation, the random forest regression (RF) and radial basis function (RBF) models demonstrated better prediction performance in eight attempted machine learning models. The CL_tot values predicted by the RF and RBF models were within two-fold of the observed values for 67.7 and 71.9% of cluster-split test set compounds, respectively, while the predictivity was worse in the time-split dataset. The predictivity of both models tended to be improved by incorporating in vitro parameters, unbound fraction in plasma (f_u,p), and CL_int. CL_tot prediction utilizing in vitro CL_int and the well-stirred model, correcting for the fraction unbound in blood, was substantially worse compared to machine learning approaches for the same cluster-split test set. The reason that CL_tot is underestimated by IVIVE is not fully explained by considering the calculated microsomal unbound fraction (cf_u,mic), extended clearance classification system (ECCS), and omitting high clearance compounds in excess of hepatic blood flow. The analysis suggests that in silico machine learning models may have the power to reduce reliance on or replace in vitro and in vivo studies for chemical structure optimization in early drug discovery.

In vitro and in vivo effects of 2,4 diaminoquinazoline inhibitors of the decapping scavenger enzyme DcpS: Context-specific modulation of SMN transcript levels

C5-substituted 2,4-diaminoquinazoline inhibitors of the decapping scavenger enzyme DcpS (DAQ-DcpSi) have been developed for the treatment of spinal muscular atrophy (SMA), which is caused by genetic deficiency in the Survival Motor Neuron (SMN) protein. These compounds are claimed to act as SMN2 transcriptional activators but data underlying that claim are equivocal. In addition it is unclear whether the claimed effects on SMN2 are a direct consequence of DcpS inhibitor or might be a consequence of lysosomotropism, which is known to be neuroprotective. DAQ-DcpSi effects were characterized in cells in vitro utilizing DcpS knockdown and 7-methyl analogues as probes for DcpS vs non-DcpS-mediated effects. We also performed analysis of Smn transcript levels, RNA-Seq analysis of the transcriptome and SMN protein in order to identify affected pathways underlying the therapeutic effect, and studied lysosomotropic and non-lysosomotropic DAQ-DCpSi effects in 2B/- SMA mice. Treatment of cells caused modest and transient SMN2 mRNA increases with either no change or a decrease in SMNΔ7 and no change in SMN1 transcripts or SMN protein. RNA-Seq analysis of DAQ-DcpSi-treated N2a cells revealed significant changes in expression (both up and down) of approximately 2,000 genes across a broad range of pathways. Treatment of 2B/- SMA mice with both lysomotropic and non-lysosomotropic DAQ-DcpSi compounds had similar effects on disease phenotype indicating that the therapeutic mechanism of action is not a consequence of lysosomotropism. In striking contrast to the findings in vitro, Smn transcripts were robustly changed in tissues but there was no increase in SMN protein levels in spinal cord. We conclude that DAQ-DcpSi have reproducible benefit in SMA mice and a broad spectrum of biological effects in vitro and in vivo, but these are complex, context specific, and not the result of simple SMN2 transcriptional activation.

Potent and cellularly active 4-aminoimidazole inhibitors of cyclin-dependent kinase 5/p25 for the treatment of Alzheimer's disease

Utilizing structure-based drug design, a 4-aminoimidazole heterocyclic core was synthesized as a replacement for a 2-aminothiazole due to potential metabolically mediated toxicity. The synthetic route utilized allowed for ready synthesis of 1-substituted-4-aminoimidazoles. SAR exploration resulted in the identification of a novel cis-substituted cyclobutyl group that gave improved enzyme and cellular potency against cdk5/p25 with up to 30-fold selectivity over cdk2/cyclin E.

Mechanism of oxime reactivation of acetylcholinesterase analyzed by chirality and mutagenesis

Organophosphates inactivate acetylcholinesterase by reacting covalently with the active center serine. We have examined the reactivation of a series of resolved enantiomeric methylphosphonate conjugates of acetylcholinesterase by two oximes, 2-pralidoxime (2-PAM) and 1-(2'-hydroxyiminomethyl-1'-pyridinium)-3-(4'-carbamoyl-1-pyridinium) (HI-6). The S(p) enantiomers of the methylphosphonate esters are far more reactive in forming the conjugate with the enzyme, and we find that rates of oxime reactivation also show an S(p) versus R(p) preference, suggesting that a similar orientation of the phosphonyl oxygen toward the oxyanion hole is required for both efficient inactivation and reactivation. A comparison of reactivation rates of (S(p))- and (R(p))-cycloheptyl, 3,3-dimethylbutyl, and isopropyl methylphosphonyl conjugates shows that steric hindrance by the alkoxy group precludes facile access of the oxime to the tetrahedral phosphorus. To facilitate access, we substituted smaller side chains in the acyl pocket of the active center and find that the Phe295Leu substitution enhances the HI-6-elicited reactivation rates of the S(p) conjugates up to 14-fold, whereas the Phe297Ile substitution preferentially enhances 2-PAM reactivation by as much as 125-fold. The fractional enhancement of reactivation achieved by these mutations of the acyl pocket is greatest for the conjugated phosphonates of the largest steric bulk. By contrast, little enhancement of the reactivation rate is seen with these mutants for the R(p) conjugates, where limitations on oxime access to the phosphonate and suboptimal positioning of the phosphonyl oxygen in the oxyanion hole may both slow reactivation. These findings suggest that impaction of the conjugated organophosphate within the constraints of the active center gorge is a major factor in influencing oxime access and reactivation rates. Moreover, the individual oximes differ in attacking orientation, leading to the presumed pentavalent transition state. Hence, their efficacies as reactivating agents depend on the steric bulk of the intervening groups surrounding the tetrahedral phosphorus.