Quinuclidine N-oxygenation mediated by flavin-containing monooxygenase (FMO) 1 and FMO3 in kidney and liver microsomes from humans, monkeys, dogs, and pigs.

Flavin-containing monooxygenases (FMOs) are a family of enzymes that are involved in the oxygenation of heteroatom-containing molecules. In humans, FMO3 is the major hepatic form, whereas FMO1 is predominant in the kidneys. FMO1 and FMO3 were also identified in monkeys, dogs, and pigs. The predicted contribution of human FMO3 to drug candidate N-oxygenation could be estimated using the classic base dissociation constants of the N-containing moiety. A basic quinuclidine moiety was found in the natural quinine and medicinal products. Consequently, N-oxygenation of quinuclidine was evaluated using liver and kidney microsomes from humans, monkeys, dogs, and pigs as well as recombinant FMO1, FMO3, and FMO5 enzymes. Experiments using simple reversed-phase liquid chromatography with fluorescence monitoring revealed that recombinant FMO1 mediated quinuclidine N-oxygenation with a high capacity in humans. Moreover, recombinant FMO1, FMO3, and/or FMO5 in monkeys, dogs, and pigs exhibited relatively broad substrate specificity toward quinuclidine N-oxygenation. Kinetic analysis showed that human FMO1 efficiently, and pig FMO1 moderately, mediated quinuclidine N-oxygenation with high capacity, which is consistent with the reported findings for larger substrates readily accepted by pig FMO1 but excluded by human FMO1. In contrast, human FMO3-mediated quinuclidine N-oxygenation was slower than that of the typical FMO3 substrate trimethylamine. These results suggest that some species differences exist in terms of FMO-mediated quinuclidine N-oxygenation in humans and some animal models (monkeys, dogs, and minipigs); however, the potential for quinuclidine, which has a simple chemical structure, to be inhibited clinically by co-administered drugs should be relatively low, especially in human livers. Significance Statement The high capacity of human flavin-containing monooxygenase (FMO) 1 to mediate quinuclidine N-oxygenation, a basic moiety in natural products and medicines, was demonstrated by simple reversed-phase liquid chromatography using fluorescence monitoring. The substrate specificity of FMO1 and FMO3 toward quinuclidine N-oxygenation in monkeys, dogs, and pigs was suggested to be relatively broad. Human FMO3-mediated quinuclidine N-oxygenation was slower than trimethylamine N-oxygenation. The likelihood of quinuclidine, with its simple chemical structure, being clinically inhibited by co-administered drugs is relatively low.

In Vivo and In Vitro Induction of Cytochrome P450 3A4 by Thalidomide in Humanized-Liver Mice and Experimental Human Hepatocyte HepaSH cells.

Autoinduction of cytochrome P450 (P450) 3A4-mediated metabolism of thalidomide was investigated in humanized-liver mice and human hepatocyte-derived HepaSH cells. The mean plasma ratios of 5-hydroxythalidomide and glutathione adducts to thalidomide were significantly induced (3.5- and 6.0-fold, respectively) by thalidomide treatment daily at 1000 mg/kg for 3 days and measured at 2 h after the fourth administration (on day 4). 5-Hydroxythalidomide was metabolically activated by P450 3A4 in HepaSH cells pretreated with 300 and 1000 µM thalidomide, and 5,6-dihydroxythalidomide was detected. Significant induction of P450 3A4 mRNA expression (4.1-fold) in the livers of thalidomide-treated mice occurred. Thalidomide exerts a variety of actions through multiple mechanisms following bioactivation by induced human P450 3A enzymes.

Modeled Hepatic/Plasma Exposures of Omeprazole Prescribed Alone in Cytochrome P450 2C19 Poor Metabolizers Are Likely Associated with Hepatic Toxicity Reported in a Japanese Adverse Event Database.

Omeprazole, a gastric acid pump inhibitor, is repeatedly administered and is oxidatively metabolized mainly by polymorphic cytochrome P450 2C19. The prescribed dosage of omeprazole was discontinued or reduced in 47 of the 135 patients who received omeprazole alone in this survey, as recorded in the Japanese Adverse Drug Event Report database. The days to onset of omeprazole-related disorders were 3-4 d (median) and 16 d for intravenous 20-40 mg and oral 20 mg daily doses, respectively, in 34 patients for whom relevant data were available. The maximum plasma concentration of omeprazole was pharmacokinetically modeled after a single oral 40-mg dose in P450 2C19-defective poor metabolizers and was 2.4-fold higher than that in extensive metabolizers. The modeled area under the hepatic concentration curves of omeprazole in P450 2C19 poor metabolizers after virtual daily 40-mg doses for 7 d was 5.2-fold higher than that in the extensive metabolizers. Omeprazole-induced P450 2C19 (approx. 2-fold), resulting in increased hepatic intrinsic clearance in repeated doses, was considered after the second day. Virtual plasma/hepatic exposure estimated using pharmacokinetic modeling in subjects with P450 2C19 poor metabolizers indicated that these exposure levels virtually estimated could be one of causal factors for unexpected hepatic disorders induced by prescribed omeprazole, such as those resulting from drug interactions with repeatedly co-administered medicines.

Modeled Hepatic/Plasma Exposures of Fluvastatin Prescribed Alone in Subjects with Impaired Cytochrome P450 2C9*3 as One of Possible Determinant Factors Likely Associated with Hepatic Toxicity Reported in a Japanese Adverse Event Database.

Fluvastatin is a 3-hydroxy-3-methylglutaryl CoA reductase inhibitor that competitively inhibits human cytochrome P450 (P450) 2C9 in vitro. Drug interactions between a variety of P450 2C9 substrates/inhibitors and fluvastatin can increase the incidence of fluvastatin-related hepatic or skeletal muscle toxicity in vivo. In this survey, the prescribed dosage of fluvastatin was reduced or discontinued in 133 of 164 patients receiving fluvastatin alone, as recorded in the Japanese Adverse Drug Event Report database of spontaneously reported events. The median days to onset of fluvastatin-related disorders were in the range 30-35 d in the 87 patients. Therefore, we aimed to focus on fluvastatin and, using the pharmacokinetic modeling technique, estimated the virtual plasma and hepatic exposures in subjects harboring the impaired CYP2C9*3 allele. The plasma concentrations of fluvastatin modeled after a virtual oral 20-mg dose increased in homozygotes with CYP2C9*3; the area under the plasma concentration curve was 4.9-fold higher than that in Japanese homozygotes for wild-type CYP2C9*1. The modeled hepatic concentrations of fluvastatin in patients with CYP2C9*3/*3 after virtual daily 20-mg doses for 7 d were 31-fold higher than those in subjects with CYP2C9*1/*1. However, heterozygous Chinese patients with CYP2C9*1/*3 reportedly have a limited elevation (1.2-fold) in plasma maximum concentrations. Virtual hepatic/plasma exposures in subjects harboring the impaired CYP2C9*3 allele estimated using pharmacokinetic modeling indicate that such exposure could be a causal factor for hepatic disorders induced by fluvastatin prescribed alone in a manner similar to that for interactions with a variety of co-administered drugs.

Species Specificity and Selection of Models for Drug Oxidations Mediated by Polymorphic Human Enzymes.

Many drug oxygenations are mainly mediated by polymorphic cytochromes P450 (P450s) and also by flavin-containing monooxygenases (FMOs). More than 50 years of research on P450/FMO-mediated drug oxygenations have clarified their catalytic roles. The natural product coumarin causes hepatotoxicity in rats via the reactive coumarin 3,4-epoxide, a reaction catalyzed by P450 1A2; however, coumarin undergoes rapid 7-hydroxylation by polymorphic P450 2A6 in humans. The primary oxidation product of the teratogen thalidomide in rats is deactivated 5'-hydroxythalidomide plus sulfate and glucuronide conjugates; however, similar 5'-hydroxythalidomide and 5-hydroxythalidomide are formed in rabbits in vivo. Thalidomide causes human P450 3A enzyme induction in liver (and placenta) and is also activated in vitro and in vivo by P450 3A through the primary human metabolite 5-hydroxythalidomide (leading to conjugation with glutathione/nonspecific proteins). Species differences exist in terms of drug metabolism in rodents and humans, and such differences can be very important when determining the contributions of individual enzymes. The approaches used for investigating the roles of human P450 and FMO enzymes in understanding drug oxidations and clinical therapy have not yet reached maturity and still require further development. SIGNIFICANCE STATEMENT: Drug oxidations in animals and humans mediated by P450s and FMOs are important for understanding the pharmacological properties of drugs, such as the species-dependent teratogenicity of the reactive metabolites of thalidomide and the metabolism of food-derived odorous trimethylamine to non-odorous (but proatherogenic) trimethylamine N-oxide. Recognized differences exist in terms of drug metabolism between rodents, non-human primates, and humans, and such differences are important when determining individual liver enzyme contributions with substrates in in vitro and in vivo systems.

Variants of Flavin-Containing Monooxygenase 3 Found in Subjects in an Updated Database of Genome Resources.

Single-nucleotide substitutions of human flavin-containing monooxygenase 3 (FMO3) identified in the whole-genome sequences of the updated Japanese population reference panel (now containing 38,000 subjects) were investigated. In this study, two stop codon mutations, two frameshifts, and 43 amino-acid-substituted FMO3 variants were identified. Among these 47 variants, one stop codon mutation, one frameshift, and 24 substituted variants were already recorded in the National Center for Biotechnology Information database. Functionally impaired FMO3 variants are known to be associated with the metabolic disorder trimethylaminuria; consequently, the enzymatic activities of the 43 substituted FMO3 variants were investigated. Twenty-seven recombinant FMO3 variants expressed in bacterial membranes had similar activities toward trimethylamine N-oxygenation (∼75%-125%) to that of wild-type FMO3 (98 minutes-1). However, six recombinant FMO3 variants (Arg51Gly, Val283Ala, Asp286His, Val382Ala, Arg387His, and Phe451Leu) had moderately decreased (∼50%) activities toward trimethylamine N-oxygenation, and 10 recombinant FMO3 variants (Gly11Asp, Gly39Val, Met66Lys, Asn80Lys, Val151Glu, Gly193Arg, Arg387Cys, Thr453Pro, Leu457Trp, and Met497Arg) showed severely decreased FMO3 catalytic activity (<10%). Because of the known deleterious effects of FMO3 C-terminal stop codons, the four truncated FMO3 variants (Val187SerfsTer25, Arg238Ter, Lys416SerfsTer72, and Gln427Ter) were suspected to be inactive with respect to trimethylamine N-oxygenation. The FMO3 p.Gly11Asp and p.Gly193Arg variants were located within the conserved sequences of flavin adenine dinucleotide (positions 9-14) and NADPH (positions 191-196) binding sites, which are important for FMO3 catalytic function. Whole-genome sequence data and kinetic analyses revealed that 20 of the 47 nonsense or missense FMO3 variants had moderately or severely impaired activity toward N-oxygenation of trimethylaminuria. SIGNIFICANCE STATEMENT: The number of single-nucleotide substitutions in human flavin-containing monooxygenase 3 (FMO3) recorded in the expanded Japanese population reference panel database was updated. One stop mutation, FMO3 p.Gln427Ter; one frameshift (p.Lys416SerfsTer72); and 19 novel amino-acid-substituted FMO3 variants were identified, along with p.Arg238Ter, p.Val187SerfsTer25, and 24 amino-acid-substituted variants already recorded with reference SNP (rs) numbers. Recombinant FMO3 Gly11Asp, Gly39Val, Met66Lys, Asn80Lys, Val151Glu, Gly193Arg, Arg387Cys, Thr453Pro, Leu457Trp, and Met497Arg variants showed severely decreased FMO3 catalytic activity, possibly associated with the trimethylaminuria.

Pharmacokinetic Models Scaled-up from Humanized-liver Mouse Data Can Account for Drug Monitoring Results of Atomoxetine and Its 4-Hydroxylated and N-Demethylated Metabolites i n Pediatric Patients Genotyped for Cytochrome P450 2D6.

Atomoxetine is a cytochrome P450 (P450) 2D6 probe substrate and an approved medicine for attention-deficit/hyperactivity disorder. In this humanized-liver mouse study, interactions between atomoxetine and the P450 2D6 probe drug paroxetine were observed. Human physiologically based pharmacokinetic (PBPK) models were established by scaling up humanized-liver mouse data obtained in the absence or presence of paroxetine. These models could explain the drug monitoring results of atomoxetine and its primary 4-hydroxylated and N-demethylated metabolites in Japanese children aged 8-14 years and could be used to help establish the correct dosage and for the evaluation of clinical outcomes. The results of simple PBPK models (using input parameters that reflected the subjects' small body size and normal or reduced P450 2D6-dependent clearance) were in general agreement with one-point measured plasma concentrations of atomoxetine and its 4-hydroxylated and N-demethylated metabolites in 13 pediatric participants. Unexpectedly high hepatic exposure, possibly in intermediate-metabolizer patients harboring CYP2D6*10 or 2D6*36 alleles, might in part explain the adverse effects of atomoxetine prescribed alone recorded in a Japanese adverse-event database. The steady-state, one-point drug monitoring data from the participants indicated extensive biotransformation of atomoxetine to 4-hydroxyatomoxetine under individually prescribed doses of atomoxetine. These results also suggest that a relatively narrow range of 4-hydroxyatomoxetine and N-desmethylatomoxetine concentration ratios in spot urine and/or plasma samples from pediatric patients could be a simple semiquantitative determinant factor for P450 2D6 intermediate metabolizers, compared with the wide range of concentrations of the two primary metabolites and substrate in extensive metabolizers. Significance Statement Validated simple pharmacokinetic models are able to predict steady-state plasma concentrations of the approved medicine atomoxetine and its primary metabolites in the majority of pediatric patients. The package insert advises careful dose escalation, especially for poor metabolizers; however, no simple way exists to determine P450 2D6 phenotypes. A relatively narrow range ratio of 4-hydroxyatomoxetine and N-desmethylatomoxetine in spot urine/plasma samples could be a simple semi-quantitative determinant factor for P450 2D6 intermediate metabolizers to optimize or confirm the correct dosage.

Metabolic Syndrome and the Increased Risk of Medically Certified Long-term Sickness Absence: A Prospective Analysis Among Japanese Workers.

BACKGROUND: Metabolic syndrome (MetS) has been associated with various chronic diseases that may lead to long-term sickness absence (LTSA), but there is lacking information on the direct association between MetS and LTSA. The present study aimed to investigate the all-cause and cause-specific associations between MetS and the risk of medically certified LTSA among Japanese workers. METHODS: We recruited 67,403 workers (57,276 men and 10,127 women), aged 20-59 years from 13 companies in Japan during their health check-ups in 2011 (11 companies) and 2014 (2 companies), and we followed them for LTSA events (≥30 consecutive days) until March 31, 2020. MetS was defined according to the Joint Interim Statement. A Cox proportional hazards regression model was used to estimate hazard ratios (HRs) and its 95% confidence intervals (CIs) for LTSA associated with MetS and its components. RESULTS: During 408,324 person-years of follow-up, 2,915 workers experienced LTSA. The adjusted HR for all-cause LTSA was 1.54 (95% CI, 1.41-1.68) among those with MetS compared to those without MetS. In cause-specific analysis, HRs associated with MetS significantly increased for LTSA due to overall physical disorders (1.76); cardiovascular diseases (3.16); diseases of the musculoskeletal system and connective tissue (2.01); cancers (1.24); obesity-related cancers (1.35); mental, behavioral, and neurodevelopmental disorders (1.28); reaction to severe stress and adjustment disorders (1.46); and external causes (1.46). The number of MetS components were also significantly associated with increased LTSA risk. CONCLUSION: MetS was associated with an increase in the risk of LTSA due to various diseases among Japanese workers.

Modeled Rat Hepatic and Plasma Concentrations of Chemicals after Virtual Administrations Using Two Sets of in Silico Liver-to-Plasma Partition Coefficients.

The hepatic elimination of chemical substances in pharmacokinetic models requires hepatic intrinsic clearance (CLh,int) parameters for unbound drug in the liver, and these are regulated by the liver-to-plasma partition coefficients (Kp,h). Both Poulin and Theil and Rodgers and Rowland have proposed in silico expressions for Kp,h for a variety of chemicals. In this study, two sets of in silico Kp,h values for 14 model substances were assessed using experimentally reported in vivo steady-state Kp,h data and time-dependent virtual internal exposures in the liver and plasma modeled by forward dosimetry in rats. The Kp,h values for 14 chemicals independently calculated using the primary Poulin and Theil method in this study were significantly correlated with those obtained using the updated Rodgers and Rowland method and with reported in vivo steady-state Kp,h data in rats. When pharmacokinetic parameters were derived based on individual in vivo time-dependent data for diazepam, phenytoin, and nicotine in rats, the modeled liver and plasma concentrations after intravenous administration of the selected substrates in rats using two sets of in silico Kp,h values were mostly similar to the reported time-dependent in vivo internal exposures. Similar results for modeled liver and plasma concentrations were observed with input parameters estimated by machine-learning systems for hexobarbital, fingolimod, and pentazocine, with no reference to experimental pharmacokinetic data. These results suggest that the output values from rat pharmacokinetic models based on in silico Kp,h values derived from the primary Poulin and Theil model would be applicable for estimating toxicokinetics or internal exposure to substances.

Liver and Plasma Concentrations of Food Chemicals after Virtual Oral Doses Extrapolated Using in Silico Estimated Input Pharmacokinetic Parameters to Confirm Reported Liver Toxicity in Rats.

The estimation of health risks of chemical substances was historically investigated using animal studies; however, current research focuses on reducing the number of animal experiments. The toxicity of chemicals in fish screening systems is reportedly correlated with their hydrophobicity. The inverse relationship between absorption rates (intestinal cell permeability) and virtual hepatic/plasma pharmacokinetics of diverse chemicals has been previously evaluated by modeling oral administration in rats. In the current study, internal exposures, i.e., virtual maximum plasma concentrations (Cmax) and areas under the concentration-time curves (AUC), of 56 food chemicals with reported hepatic lowest-observed-effect levels (LOELs) of ≤1000 mg/kg/d in rats were pharmacokinetically modeled using in silico estimated input pharmacokinetic parameters. After a virtual single oral dose of 1.0 mg/kg of 56 food chemicals, the output Cmax and AUC values in rat plasma generated by modeling using the corresponding in silico estimated input parameters were not significantly correlated with the reported hepatic LOEL values. However, significant inverse relationships between hepatic/plasma concentrations of selected lipophilic food chemicals (i.e., octanol-water partition coefficient log P > 1) using forward dosimetry and reported LOEL values (≤300 mg/kg/d) were observed (n = 14, r=-0.52-0.66, p ≤ 0.05). This simple modeling approach, which uses no experimental pharmacokinetic data, has the potential to play a significant role in reducing the use of animals to estimate toxicokinetics or internal exposures of lipophilic food components after oral doses. Therefore, these methods are valuable for estimating hepatic toxicity by using forward dosimetry in animal toxicity experiments.

Plasma and Hepatic Exposures of Celecoxib and Diclofenac Prescribed Alone in Patients with Cytochrome P450 2C9*3 Modeled after Virtual Oral Administrations and Likely Associated with Adverse Drug Events Reported in a Japanese Database.

The impacts of polymorphic cytochrome P450 (P450 or CYP) 2C9 on drug interactions and the pharmacokinetics of cyclooxygenase inhibitors have attracted considerable attention. In this survey, the prescribed dosage was reduced or discontinued in 150 and 56 patients, respectively, receiving celecoxib and diclofenac prescribed alone, as recorded in a Japanese database of adverse drug events. Among the factors underlying adverse events, intrinsic drug clearance rates may be a contributing factor. The pharmacokinetically modeled plasma concentrations of celecoxib after an oral 200-mg dose increased in CYP2C9*3 homozygotes: the area under the plasma concentration curve was 4.7-fold higher than that in CYP2C9*1 homozygotes. In patients with CYP2C9*3/*3, the virtual hepatic concentrations of diclofenac after three daily 25-mg doses for a week were 11-fold higher than the plasma concentrations in subjects with CYP2C9*1/*1. The in vivo and in vitro fractions of the victim drug metabolized by a specific polymorphic P450 form is an important determining factor for estimating drug-drug interactions. Virtual hepatic and plasma exposures estimated by pharmacokinetic modeling in patients harboring the impaired CYP2C9*3 allele could represent a causal factor for adverse events induced by celecoxib or diclofenac in a manner similar to that for drug interactions.

A decreasing plasma concentration of a toxicologically active metabolite 9-carboxymethoxymethylguanine after dialysis - A potential new clinical biomarker for improving encephalopathy in patients treated with acyclovir.

Although acyclovir is a key drug for the treatment of herpes infections, a consciousness disorder known as "acyclovir encephalopathy" is among its side effects. We encountered a patient with encephalopathy and measured the plasma and cerebrospinal fluid concentrations of acyclovir and its toxicologically active metabolite 9-carboxymethoxymethylguanine (CMMG). Before dialysis, cerebrospinal fluid concentrations of acyclovir and CMMG in this patient with a consciousness disorder were approximately 10% and 1%, respectively, of their plasma concentrations. After 3 days of dialysis, plasma CMMG levels decreased to detectable but below quantitative levels (<0.1 µg/mL), resulting in normal consciousness. These results suggest that decreasing plasma CMMG concentration could be one of clinical biomarkers for improving consciousness in patients with encephalopathy associated with acyclovir.

Diagnosis-specific Cumulative Incidence of Return-to-work, Resignation, and Death Among Long-term Sick-listed Employees: Findings From the Japan Epidemiology Collaboration on Occupational Health Study.

BACKGROUND: While it is essential to understand how long is sufficient for return-to-work when designing paid sick-leave systems, little attempt has been done to collect cause-specific information on when and how many of sickness absentees returned to work, became unemployed, or passed away. METHODS: We studied the first sick-leave episode of ≥30 consecutive days in those ≤55 years of age during 2012-2013 among employees of 11 Japanese private companies (n = 1,209), which were followed until 2017. Overall and disease-specific cumulative incidences of return-to-work, resignations, and deaths were estimated using competing risk analysis. RESULTS: During the 3.5-year period (follow-up rate: 99.9%), 1,014 returned to work, 167 became unemployed, and 27 died. Overall, return-to-work occurred within 1 year in 74.9% of all absentees and in 89.3% of those who successfully returned to work. Resignation occurred within 1 year in 8.7% of all absentees and in 62.9% of all subjects who resigned. According to ICD-10 chapters, the cumulative incidence of return-to-work ranged from 82.1% for mental disorders (F00-F99) to 95.3% for circulatory diseases (I00-I99). The cumulative incidence of return-to-work due to mental disorders ranged from 66.7% in schizophrenia (F20) to 95.8% in bipolar affective disorders (F31). Death was rarely observed except for cases of neoplasms (C00-D48), of which the cumulative incidence of death reached 14.2% by 1.5 years. CONCLUSION: Return-to-work and resignations occurred commonly within 1 year of sick leave among long-term sickness absentees in the Japanese private companies. Our findings may assist occupational physicians and employers in developing effective social protection schemes.

Trivariate Linear Regression and Machine Learning Prediction of Possible Roles of Efflux Transporters in Estimated Intestinal Permeability Values of 301 Disparate Chemicals.

A system for predicting apparent bidirectional permeability (Papp) across Caco-2 cells of diverse chemicals has been reported. The present study aimed to investigate the relationship between in silico-generated Papp (from apical to basal side, Papp A to B) for 301 substances with diverse structures and a binary classification of the reported roles of efflux P-glycoprotein or breast cancer resistant protein. The in silico log(Papp A to B/Papp B to A) values of 70 substances with reported active efflux and 231 substances with no reported active efflux were significantly different (p < 0.01). The probabilities of active efflux transport estimated by trivariate analysis with log MW, log DpH 6.0, and log DpH 7.4 for the 70 active-efflux-positive compounds were higher than those of the other 231 substances (p < 0.01); the area under the corresponding receiver operating characteristic (ROC) curve was 0.81. Further probability values estimated using a machine learning algorithm with 30 chemical descriptors as inputs yielded an area under the ROC curve of 0.79. Using a secondary set of 52 efflux-positive and 48 efflux-negative medicines, the final trivariate-generated probabilities resulted in no significant differences between these binary groups (p = 0.09); however, the final machine learning model demonstrated a good area under the ROC curve of 0.79. Consequently, a combination of the previously established system for generating the permeability coefficients across intestinal monolayers (a continuous variable) and the currently proposed system for predicting the roles of additional active efflux (a binary classification) could prove useful; high accuracy was achieved by applying machine learning using in silico-generated chemical descriptors.

Updated in Silico Prediction Methods for Fractions Absorbed and Key Input Parameters of 355 Disparate Chemicals for Physiologically Based Pharmacokinetic Models for Time-Dependent Plasma Concentrations after Virtual Oral Doses in Humans.

Human metabolic profiles for substances such as toxic food-derived compounds are usually allometrically extrapolated from traditionally determined in vivo rat concentration profiles. To evaluate internal exposures in humans without any reference to experimental data, physiologically based pharmacokinetic (PBPK) modeling could be used if the model input parameters could be estimated in silico. This approach would simplify the use of PBPK models for forward dosimetry after oral doses. In this study, the in silico estimation of input parameters for PBPK models (i.e., fraction absorbed × intestinal availability, absorption rate constants, and volumes of the systemic circulation) was updated for an panel of 355 chemicals (212 previously analyzed and 143 additional substances) using a light gradient boosting machine learning algorithms (LightGBM) based on between 11 and 29 in silico-calculated chemical descriptors. Simplified human PBPK models were then used to calculate virtual maximum plasma concentrations (Cmax) and areas under the concentration-time curve (AUC) based on two sets of input parameters, i.e., traditionally derived values from in vivo data and those calculated in silico using the current updated systems. Both sets of Cmax and AUC data were well correlated (r = 0.87 and r = 0.73, respectively; p < 0.01, n = 355). Therefore, input parameters for human PBPK models for a diverse range of compounds could be successfully estimated using chemical descriptors and in silico tools. This approach to pharmacokinetic modeling has potential for application in computational toxicology and in the clinical setting for assessing the potential risk of general chemicals.

Machine Learning Prediction of the Three Main Input Parameters of a Simplified Physiologically Based Pharmacokinetic Model Subsequently Used to Generate Time-Dependent Plasma Concentration Data in Humans after Oral Doses of 212 Disparate Chemicals.

Physiologically based pharmacokinetic (PBPK) modeling has the potential to play significant roles in estimating internal chemical exposures. The three major PBPK model input parameters (i.e., absorption rate constants, volumes of the systemic circulation, and hepatic intrinsic clearances) were generated in silico for 212 chemicals using machine learning algorithms. These input parameters were calculated based on sets of between 17 and 65 chemical properties that were generated by in silico prediction tools before being processed by machine learning algorithms. The resulting simplified PBPK models were used to estimate plasma concentrations after virtual oral administrations in humans. The estimated absorption rate constants, volumes of the systemic circulation, and hepatic intrinsic clearance values for the 212 test compounds determined traditionally (i.e., based on fitting to measured concentration profiles) and newly estimated had correlation coefficients of 0.65, 0.68, and 0.77 (p < 0.01, n = 212), respectively. When human plasma concentrations were modeled using traditionally determined input parameters and again using in silico estimated input parameters, the two sets of maximum plasma concentrations (r = 0.85, p < 0.01, n = 212) and areas under the curve (r = 0.80, p < 0.01, n = 212) were correlated. Virtual chemical exposure levels in liver and kidney were also estimated using these simplified PBPK models along with human plasma levels. These results indicate that the PBPK model input parameters for humans of a diverse set of compounds can be reliability estimated using chemical descriptors calculated using in silico tools.

In Silico Prediction of Input Parameters for Simplified Physiologically Based Pharmacokinetic Models for Estimating Plasma, Liver, and Kidney Exposures in Rats after Oral Doses of 246 Disparate Chemicals.

Recently developed computational models can estimate plasma, hepatic, and renal concentrations of industrial chemicals in rats. Typically, the input parameter values (i.e., the absorption rate constant, volume of systemic circulation, and hepatic intrinsic clearance) for simplified physiologically based pharmacokinetic (PBPK) model systems are calculated to give the best fit to measured or reported in vivo blood substance concentration values in animals. The purpose of the present study was to estimate in silico these three input pharmacokinetic parameters using a machine learning algorithm applied to a broad range of chemical properties obtained from several cheminformatics software tools. These in silico estimated parameters were then incorporated into PBPK models for predicting internal exposures in rats. Following this approach, simplified PBPK models were set up for 246 drugs, food components, and industrial chemicals with a broad range of chemical structures. We had previously generated PBPK models for 158 of these substances, whereas 88 for which concentration series data were available in the literature were newly modeled. The values for the absorption rate constant, volume of systemic circulation, and hepatic intrinsic clearance could be generated in silico by equations containing between 14 and 26 physicochemical properties. After virtual oral dosing, the output concentration values of the 246 compounds in plasma, liver, and kidney from rat PBPK models using traditionally determined and in silico estimated input parameters were well correlated (r ≥ 0.83). In summary, by using PBPK models consisting of chemical receptor (gut), metabolizing (liver), excreting (kidney), and central (main) compartments with in silico-derived input parameters, the forward dosimetry of new chemicals could provide the plasma/tissue concentrations of drugs and chemicals after oral dosing, thereby facilitating estimates of hematotoxic, hepatotoxic, or nephrotoxic potential as a part of risk assessment.

An Updated In Silico Prediction Method for Volumes of Systemic Circulation of 323 Disparate Chemicals for Use in Physiologically Based Pharmacokinetic Models to Estimate Plasma and Tissue Concentrations after Oral Doses in Rats.

Updated algorithms for predicting the volumes of systemic circulation (V1), along with absorption rate constants and hepatic intrinsic clearances, as input parameters for physiologically based pharmacokinetic (PBPK) models were established to improve the accuracy of estimated plasma and tissue concentrations of 323 chemicals after virtual oral administrations in rats. Using ridge regression with an enlarged set of chemical descriptors (up to 99), the estimated input V1 values resulted in an improved correlation coefficient (from 246 compounds) with the traditionally determined values. The PBPK model input parameters for rats of diverse compounds can be precisely estimated by increasing the number of descriptors.

Metabolic Profiles of Tetrabromobisphenol A in Humans Extrapolated from Humanized-Liver Mouse Data Using a Simplified Physiologically Based Pharmacokinetic Model.

Tetrabromobisphenol A, a brominated flame retardant, is increasingly prevalent worldwide and presents a potential health risk. Adjusted animal biomonitoring equivalents of tetrabromobisphenol A after orally administered doses in humanized-liver mice were scaled up to humans using known species allometric scaling factors to set up simplified physiologically based pharmacokinetic (PBPK) models. Absorbed tetrabromobisphenol A was slightly, moderately, and extensively metabolized in vivo to its glucuronide in rats, control mice, and humanized-liver mice tested, respectively. In silico estimated hepatic exposures of tetrabromobisphenol A and its glucuronide generated using the rat PBPK model-generated plasma concentration profiles were consistent with the reported values. The extent of hepatic injury in humanized-liver mice caused by tetrabromobisphenol A was evaluated by detecting human albumin mRNA in mouse plasma after oral administration of a high dose of tetrabromobisphenol A (1000 mg/kg). Reverse dosimetry analyses were carried out using two human PBPK models (set up based on the humanized-liver-mouse model and by optimizing the input parameters for reported human plasma concentrations of tetrabromobisphenol A glucuronide) to estimate the tetrabromobisphenol A daily intake based on reported human serum concentrations of total tetrabromobisphenol A from biomonitoring data. Within the predictability of the forward and reverse dosimetry estimations, the calculated daily intake was found to be far below established health benchmark levels (i.e., the suggested daily reported reference dose) with a wide (4 orders of magnitude) safety margin. These results suggest that the simplified PBPK models can be successfully applied to forward and reverse dosimetry estimations of tissue and/or blood exposures of tetrabromobisphenol A in humans after oral doses.

Smoking Cessation, Weight Gain, and the Trajectory of Estimated Risk of Coronary Heart Disease: 8-Year Follow-up From a Prospective Cohort Study.

INTRODUCTION: The effect of weight gain following smoking cessation on cardiovascular risks is unclear. We aimed to prospectively investigate the association of weight gain following smoking cessation with the trajectory of estimated risks of coronary heart disease (CHD). METHODS: In a cohort of 18 562 Japanese male employees aged 30-64 years and initially free of cardiovascular diseases, participants were exclusively grouped into sustained smokers, quitters with weight gain (body weight increase ≥5%), quitters without weight gain (body weight increase <5% or weight loss), and never smokers. Global 10-year CHD risk was annually estimated by using a well-validated prediction model for the Japanese population. Linear mixed models and piecewise linear mixed models were used to compare changes in the estimated 10-year CHD risk by smoking status and weight change following smoking cessation. RESULTS: During a maximum of 8-year follow-up, both quitters with and without weight gain had a substantially decreased level of estimated 10-year CHD risk after quitting smoking, compared with sustained smokers (all ps for mean differences < .001). The estimated 10-year CHD risk within the first year after cessation decreased more rapidly in quitters without weight gain than in quitters with weight gain (change rate [95% confidence interval, CI] -0.90 [-1.04 to -0.75] vs. -0.40 [-0.60 to -0.19] % per year, p < .0001). Thereafter, the estimated 10-year CHD risk in both groups increased at similar rates (change rate [95% CI] -0.07 [-0.21 to 0.07] vs. 0.11 [-0.09 to 0.30] % per year, p = .16, from year 1 to year 2; and 0.10 [0.05 to 0.15] vs. 0.11 [0.04 to 0.18] % per year, p = .80, from year 2 to year 8). CONCLUSIONS: In this population of middle-aged, Japanese male workers, smoking cessation greatly reduces the estimated 10-year risk of CHD. However, weight gain weakens the beneficial effect of quitting smoking in a temporary and limited fashion. IMPLICATIONS: To the best of our knowledge, this study is the first to examine the effect of weight gain following smoking cessation on the trajectory of the absolute risk of CHD. Our data imply that the benefits of cessation for reducing the absolute risk of CHD outweigh the potential risk increase due to weight gain, and suggest that in order to maximize the beneficial effects of quitting smoking, interventions to control post-cessation weight gain might be warranted.