Article pubs.acs.org/crt
Cite This: Chem. Res. Toxicol. 2019, 32, 211−218
Plasma and Hepatic Concentrations of Chemicals after Virtual Oral Administrations Extrapolated Using Rat Plasma Data and Simple Physiologically Based Pharmacokinetic Models Yusuke Kamiya,†,§ Shohei Otsuka,†,§ Tomonori Miura,† Hiroka Takaku,† Rio Yamada,† Mayuko Nakazato,† Hitomi Nakamura,† Sawa Mizuno,† Fumiaki Shono,‡ Kimito Funatsu,‡ and Hiroshi Yamazaki*,† Downloaded via UNIV OF WINNIPEG on January 25, 2019 at 12:24:05 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
†
Laboratory of Drug Metabolism and Pharmacokinetics, Showa Pharmaceutical University, 3-3165 Higashi-tamagawa Gakuen, Machida, Tokyo 194-8543, Japan ‡ Department of Chemical System Engineering, School of Engineering, The University of Tokyo, Bunkyo-ku, Tokyo 113-8656, Japan ABSTRACT: Only a small fraction of chemicals possesses adequate in vivo toxicokinetic data for assessing potential hazards. The aim of the present study was to model the plasma and hepatic pharmacokinetics of more than 50 disparate types of chemicals and drugs after virtual oral administrations in rats. The models were based on reported pharmacokinetics determined after oral administration to rats. An inverse relationship was observed between no-observed-effect levels after oral administration and chemical absorbance rates evaluated for cell permeability (r = −0.98, p < 0.001, n = 17). For a varied selection of more than 30 chemicals, the plasma concentration curves and the maximum concentrations obtained using a simple one-compartment model (recently recommended as a high-throughput toxicokinetic model) and a simple physiologically based pharmacokinetic (PBPK) model (consisting of chemical receptor, metabolizing, and central compartments) were highly consistent. The hepatic and plasma concentrations and the hepatic and plasma areas under the concentration−time curves of more than 50 chemicals were roughly correlated; however, differences were evident between the PBPK-modeled values in livers and empirically obtained values in plasma. Of the compounds selected for analysis, only seven had the lowest observed effect level (LOEL) values for hepatoxicity listed in the Hazard Evaluation Support System Integrated Platform in Japan. For these seven compounds, the LOEL values and the areas under the hepatic concentration−time curves estimated using PBPK modeling were inversely correlated (r = −0.78, p < 0.05, n = 7). This study provides important information to help simulate the high hepatic levels of potent hepatotoxic compounds. Using suitable PBPK parameters, the present models could estimate the plasma/hepatic concentrations of chemicals and drugs after oral doses using both PBPK forward and reverse dosimetry, thereby indicating the potential value of this modeling approach in predicting hepatic toxicity as a part of risk assessments of chemicals absorbed in the human body.
■
INTRODUCTION Current methods for estimating the health risks of general chemicals require animal testing studies that adhere to toxicity testing guidelines. Although big databases in chemical toxicity research have been set up, only a small fraction of chemicals possesses adequate in vivo data for assessing potential hazards.1 There are thousands of human-made chemicals in the environment that currently have limited toxicokinetic data. It is widely appreciated that drugs and their metabolites can cause adverse effects. The severity of such adverse effects is subject to intra- and interspecies differences mediated by the metabolic activities of enzymes in livers. These facts highlight the urgent need to develop more efficient and informative tools to determine hepatotoxicity. It is generally accepted that in vitro high-throughput toxicity screening assays combined with computational models may be able to provide a suitable alternative to traditional animal testing studies.1 Scientists from U.S. regulatory authorities have recommended the incorporation of dosimetry and exposure data © 2018 American Chemical Society
supported by full physiologically based pharmacokinetic (PBPK) modeling to the tools available for interpreting in vitro toxicity screening data.2,3 It is important that these considerations inform the determination of chemical testing priorities. Furthermore, in vitro to in vivo extrapolation of highthroughput toxicokinetic screening data to predict toxicokinetics from rapid in vitro measurements and chemical structure-based properties has been recently recommended.4 Against this background, we developed a simplified PBPK modeling system that uses a combination of algorithms along with empirical in vitro and in vivo data and literature resources.5,6 Because of its simplicity and utility, this model could also be employed by industry researchers and regulatory authorities in risk assessment to replace complex multicompartment models that are only rarely used. Received: October 20, 2018 Published: December 4, 2018 211
DOI: 10.1021/acs.chemrestox.8b00307 Chem. Res. Toxicol. 2019, 32, 211−218
Article
Chemical Research in Toxicology
Table 1. Selected Compounds with Reported NOEL or Hepatic LOEL Values for Evaluation of Apparent Permeably Using caco-2 Cell System compound
CAS no.
NOEL (mg/kg/day)
LOEL (mg/kg/day)
2-mercaptobenzimidazole 2,4-dimethylaniline N-methylaniline 3-cyanopyridine N-ethylaniline isophthalonitrile 3,5-dimethylaniline 2,3-dimethylaniline 3-nitroaniline terephthalonitrile 3,4-dimethylaniline m-cresol 2-aminobiphenyl bisphenol A 3-aminophenol 5-amino-2-chlorotoluene-4-sulfonic acid 3-aminobenzenesulfonic acid 4-nonylphenol 1,2,4-trimethylbenzene 1,2,3-trimethylbenzene paraacetaldehyde
583-39-1 95-68-1 100-61-8 100-54-9 103-69-5 626-17-5 108-69-0 87-59-2 99-09-2 623-26-7 95-64-7 108-39-4 90-41-5 80-05-7 591-27-5 88-53-9 121-47-1 84852-15-3 95-63-6 526-73-8 123-63-7
2 2 5 5 5 8 10 12 15 20 50 100 100 200 240 1000 1000
10 10 25 30 25 40 60 60 50 80 250 300 300 600 720
250 300 300 300
Papp (nm/s)
not not not not
673 661 463 569 660 805 674 624 520 573 541 851 576 321 513 20 21 determined determined determined determined
a
Values of NOEL and LOEL for hepatotoxicity for selected compounds were obtained from the Hazard Evaluation Support System Integrated Platform (HESS).10 current study represent a broad diversity of structures and are plotted in the two-dimensional plane shown in Figure 2.
The aim of the present study was to model the plasma and hepatic pharmacokinetics of more than 50 disparate types of chemicals and drugs after virtual oral administrations in rats. The models were based on reported rat plasma values and experimental pharmacokinetics determined after oral administration to rats. The current study employed two different models: one was a simple one-compartment model recently recommended by U.S. authorities as a high-throughput toxicokinetic model, and the other was a simple PBPK model consisting of a chemical receptor compartment, a metabolizing compartment, and a central compartment. Although the number of compounds was limited in the present study, the lowest observed effect levels for hepatoxicity from the Hazard Evaluation Support System Integrated Platform (HESS) in Japan and the areas under the hepatic concentration−time curves estimated using our PBPK model were inversely correlated. Overlapping of chemicals with adequate in vivo data for assessing potential hazards1 and with the lowest observed effect levels for hepatoxicity was limited at this moment. We report herein that, with a view to predicting hepatic toxicity as a part of chemical risk assessment, the present models could estimate the plasma/hepatic concentrations of chemicals and drugs after oral doses using both forward and reverse dosimetry.
■
Figure 1. One-compartment (A) and PBPK (B) models used in this study to simulate virtual oral administrations of chemicals to rats. F, bioavailability; k1 or ka, absorption constant; V, V1 or Vh, volume of distribution; kel, elimination constant; and CLh,int or CLr, clearance.
Permeability Study. The general procedures employed to prepare in vitro human intestinal caco-2 monolayers were described previously.8 Briefly, caco-2 cells obtained from (American Type Culture collection, Manassas, VA) were cultured on 12-well plates (1.0 × 105 cells/cm2) in Dulbecco’s Modified Eagle Medium supplemented with 10% (v/v) fetal calf serum, nonessential amino acids solution, and penicillin−streptomycin−amphotericin B suspension (Fujifilm Wako Pure Chemical) for 3 weeks. Inserts were placed on the cell cultures and cocultured in supplemented Dulbecco’s Modified Eagle Medium. The cells formed a confluent monolayer that exhibits the same characteristics as the intestinal barrier. The permeability coefficients (Papp, nm/sec) were calculated for timedependent in vitro absorption from the apical to basal sides of the caco-2 monolayer:
EXPERIMENTAL PROCEDURES
Chemical Space. The sources of chemicals shown in Table 1 were Sigma-Aldrich (St. Louis, MO) or Fujifilm Wako Pure Chemical (Osaka, Japan). To ensure the diversity of chemical structures in the original chemical space, the chemical structures described by 196 chemical descriptors were calculated using the open source chemoinformatics tool RDKit for approximately 50 000 randomly obtained molecules. The resulting chemical space was then projected onto a two-dimensional plane for visualization using generative topographic mapping methods.7 The compounds selected for analysis in the 212
DOI: 10.1021/acs.chemrestox.8b00307 Chem. Res. Toxicol. 2019, 32, 211−218
Article
Chemical Research in Toxicology
substrate concentration, respectively. With regard to evaluation of the predictive ability of the current PBPK models, within 3-fold errors of outputs of Cmax and AUC by the PBPK models to the matched values calculated by empirical compartment models were used as criteria after virtual oral administrations of 1.0 mg doses/kg. Statistical Analysis. Linear regression analyses were performed with the program Prism (GraphPad Software, San Diego, CA, U.S.A.) to investigate the relationship between drug concentrations/ parameters and toxicological properties [the no-observed-effect level (NOEL) or the lowest observed effect level (LOEL) for hepatoxicity] obtained from the Hazard Evaluation Support System Integrated Platform (HESS) in Japan10 and listed in Table 1.
■
Figure 2. Coordinate values of individual chemicals in a twodimensional plane illustrating variety in the chemical space. (A) The 17 chemicals selected for evaluation of caco-2 permeability are shown as triangles. (B) Indicated are the varied selection of 34 chemicals (closed circles), the plasma concentrations of which were evaluated using both one-compartment models and PBPK models, among the 53 chemicals (open and closed circles) that underwent PBPK modeling to calculate plasma and hepatic concentrations. Papp =
dX p dt
·
RESULTS Seventeen chemicals selected because they had NOEL values were evaluated for caco-2 permeability (Table 1) to understand the relationship between NOEL dose levels and chemical absorbance rates after oral administration (Figure 3). Reported
1 A · C0
where Xp is the amount of compound permeated across caco-2 cell monolayer (nmol) determined by high-performance liquid chromatography, A is the surface area of the insert membrane (cm2), and C0 is the initial concentration of each compound (nmol/mL) PBPK Study. The current study employed both a simple onecompartment model (Figure 1A), recently recommended by US authorities as a high-throughput toxicokinetic model,4 and a simple physiologically based pharmacokinetic (PBPK) model (Figure 1B) consisting of a chemical receptor compartment, a metabolizing compartment, and a central compartment.5,6,9 Pharmacokinetic parameters were derived from the respective plasma concentrations versus time curves using noncompartmental analysis in Phoenix WinNonlin 6.4 (Pharsight, Mountain View, CA). The area under the concentration−time curve from the time of virtual oral dosing to infinity with extrapolation of the terminal phase (AUC), the bioavailability (F), absorption constant (k1), the steady state volume of distribution (V/F), and the clearance (CL/F) were calculated using WinNonlin. For use in our simplified PBPK models, the values for the plasma unbound fraction (f u,p), the octanol−water partition coefficient, the liver-to-plasma concentration ratio (Kp,h), the blood-to-plasma concentration ratio (Rb) of test chemicals were estimated using in silico tools.5,6,9 The value of 0.853 L/h was used for the hepatic blood flow rate in rats (Qh). Values of the absorption rate constant (ka), the volume of the systemic circulation (V1), and the hepatic intrinsic clearance (CLh,int) were calculated to fit reported plasma substrate concentrations by simplex and modified Marquardt methods following previously established methods.5,6,9 Initiating values of hepatic clearance (CLh) and renal clearance (CLr) for PBPK modeling were derived the elimination constants in compartment modeling. Finally, in a similar way to that previously carried out,5,6,9 sets of differential equations were established for each test compound and solved to determine the concentrations in each compartment shown in Figure 1B: dXg (t ) dt
Figure 3. (A) Reported NOEL values were inversely correlated to the absorption rates of compounds (arbitrary units). (B) The absorption constant (ka) values for the PBPK modeling are compared with the log of the measured permeability for 2-mercaptobenzimidazole, mcresol, and bisphenol A. The 17 selected chemicals (listed in Table 1) evaluated for caco-2 permeability are shown as circles.
NOEL values were significantly inversely correlated with absorption rate (arbitrary units) of the 17 chemicals (Table 1) evaluated for caco-2 permeability (r = −0.98, p < 0.001, n = 17; Figure 3A). If both compounds implying moderate absorption in Figure 3A would be omitted from the analysis, a correlation coefficient was apparently decreased to be 0.48, with a p value of 0.067. The absorption constant values for 2mercaptobenzimidazole, m-cresol, and bisphenol A calculated for PBPK modeling were apparently correlated (Figure 3B) with the permeability, although this correlation was not significant. In the present study, the plasma and hepatic pharmacokinetics after virtual oral administrations in rats of 53 chemicals and drugs of disparate types were estimated using PBPK modeling. The parameters (shown in Tables 2 and 3) that define the 53 PBPK models were based on reported rat plasma values and experimental pharmacokinetics determined after oral administration to rats. Based on pharmacokinetic data from the literature, the parameter values for the onecompartment models (k1, V/F, and CL/F) and the PBPK models (ka, V1, and CLh,int) were determined by curve fitting, as described in the Experimental Procedures. The parameters for the one-compartment models for a varied selection of 34 chemicals are given in Table 2.
= − ka·Xg (t ), when t = 0, Xg (0) = dose
Vh
Q ·C h·R b dC h C = Q h·C b − h + ka·Xg − CL h,int · h ·fu,p dt K p,h K p,h
V1
Q ·C h·R b dC b = − Q h·C b + h − CLr · C b dt K p,h
where Xg, Vh, Ch, and Cb are the amount of compound in the gut, the volume of liver, the hepatic substrate concentration, and the blood 213
DOI: 10.1021/acs.chemrestox.8b00307 Chem. Res. Toxicol. 2019, 32, 211−218
a
214
103-90-2 28981-97-7 62-53-3 503612-47-3 41372-20-7 83015-26-3 83905-01-5 642-72-8 58-08-2 211914-51-1 125-71-3 3737-09-5 480449-70-5 61718-82-9 122892-67-3 191732-72-6 114798-26-4 108-39-4 2919-66-6 51384-51-1 583-39-1 59467-70-8 4376-20-9 84852-15-3 196618-13-0 106-44-5 848259-27-8 19171-19-8 366789-02-8 100-42-5 79-94-7 50-35-1 152-11-4 81-81-2
acetaminophen12 alprazolam13 aniline14 apixaban15 apomorphine16 atomoxetine17 azithromycin18 benzydamine19 caffeine20 dabigatran21 dextromethorphan20 disopyramide22 edoxaban15 fluvoxamine23 itopride24 lenalidomide25 losartan20 m-cresol26 melengestrol acetate27 metoprolol28 2-mercaptobenzimidazole29 midazolam20 mono(2-ethylhexyl) phthalate30 4-nonylphenol31 oseltamivir32 p-cresol26 pemafibratea pomalidomide33 rivaroxaban15 styrene34 tetrabromobisphenol A35 thalidomide36 verapamil37 warfarin38 14.4 4.6 17.8 0.4 29.0 1.4 0.4 8.2 7.2 4.1 15.4 1.9 1.0 1.9 11.0 2.7 5.3 3.5 1.8 1.4 3.3 2.7 3.7 1.3 9.7 4.2 19.0 0.7 0.5 65.1 5.8 0.2 70.7 1.6
k1 (1/h) 0.3 10.5 0.3 4.1 423 61.8 35.0 3.7 0.6 13.2 286 3.6 1.6 1.6 2.0 1.0 3.1 0.8 39.7 7.0 2.5 27.0 1.3 1.9 6.0 0.9 11.9 3.4 2.4 0.2 2.1 0.4 11.3 0.1
V/F (L) 0.3 7.8 0.3 1.5 89.7 16.2 1.6 1.5 0.3 4.0 116 0.9 0.2 0.2 0.6 1.0 0.7 1.2 551 9.8 0.1 18.9 0.1 0.4 3.1 1.5 1.7 0.4 1.1 0.3 0.2 0.1 0.6 0.04
CL/F (L/h) 14.3 32.4 8.4 0.4 4.0 0.9 0.5 6.3 5.7 2.5 14.4 2.1 1.1 1.9 9.6 1.0 6.7 2.9 2.2 1.6 1.9 5.6 0.7 1.4 5.5 3.1 5.8 0.8 0.4 24.0 0.2 0.2 22.7 1.5
0.2 6.3 0.2 1.2 0.4 1.4 6.4 1.2 0.3 0.7 4.1 1.9 1.0 0.5 0.9 0.2 2.1 0.1 2.2 0.3 1.6 0.1 1.0 1.2 1.5 0.1 0.8 0.5 1.3 0.1 0.1 0.4 1.5 0.0
V1 (L) 0.6 11.8 0.6 11.8 0.2 5.2 2.1 0.5 0.2 0.1 28.1 0.6 0.5 0.2 13.1 0.1 39.7 2.0 284 0.5 0.1 73.8 14.2 48.4 1.5 1.9 131 0.02 2.3 14.5 113 0.1 3.4 0.1
CLh,int, (L/h)
PBPK parameters ka (1/h)
Pharmacokinetics information for pemafibrate was taken from its package insert.
CAS no.
substrate, pharmacokinetic data reference
one-compartment parameters
473 16 393 22 1 3 5 54 291 15 1 50 111 134 105 138 69 143 4 13 95 6 168 90 31 116 19 50 38 17 113 238 21 3080
Cmax (ng/mL) 632 28 505 161 3 15 97 152 651 60 2 270 966 1440 382 232 348 161 19 24 1560 12 1690 569 71 122 134 543 225 21 1070 3150 288 37000
AUC (ng h/mL)
one-compartment modeling results
895 12 682 42 1 3 12 66 392 24 1 61 115 155 183 149 77 459 5 15 172 8 200 203 46 510 42 74 48 568 182 249 37 3020
Cmax (ng/mL)
Table 2. Plasma and Hepatic Concentrations of 34 Selected Compounds Obtained Using One-Compartment and PBPK Models
829 101 674 280 3 16 177 176 894 164 4 356 809 1300 477 281 417 421 20 42 2500 4 2190 1350 131 431 260 787 323 128 1620 3140 410 41600
AUC (ng h/mL) 877 288 571 79 5 53 233 3340 463 35 158 1390 109 1390 3690 133 5890 913 222 29 2550 81 1580 2440 1050 1010 2600 58 219 2250 1240 171 2470 13300
liver Cmax (ng/mL)
PBPK modeling results
444 117 399 353 13 118 484 1180 579 115 22 1810 451 9610 1501 230 2630 668 117 51 11600 20 14800 6570 377 682 1830 598 1080 297 10800 2160 2700 181000
liver AUC (ng h/mL)
Chemical Research in Toxicology Article
DOI: 10.1021/acs.chemrestox.8b00307 Chem. Res. Toxicol. 2019, 32, 211−218
Article
Chemical Research in Toxicology Table 3. Plasma and Hepatic Concentrations of 19 Additional Compounds that Underwent PBPK Modeling PBPK parameters substrate, pharmacokinetic data reference acrylonitrile6 bisphenol A39 chlorpyrifos40 dichlorodiphenyltrichloroethane 41 dichloromethane42 1,4-dioxane43 molinate44 nicotine45 omeprazole20 paraacetaldehyde46 PF-0493731947 2,3,5,6-tetrafluorobenzylalcohol48 tetramethylammonium49 tolbutamide12 toluene42 trichloroethylene42 trimethylamine50 1,2,3-trimethylbenzene51 1,2,4-trimethylbenzene52
PBPK modeling results
CAS no.
Ka (1/h)
V1 (L)
CLh,int (L/h)
Cmax (ng/mL)
AUC (ng h/mL)
liver Cmax (ng/mL)
liver AUC (ng h/mL)
107-13-1 80-05-7 2921-88-2 50-29-3 75-09-2 123-91-1 2212-67-1 54-11-5 73590-58-6 123-63-7 1245603-92-2 4084-38-2 75-59-2 64-77-7 108-88-3 79-01-6 75-50-3 526-73-8 95-63-6
3.3 3.5 3.7 0.3 121 0.3 5.0 1.1 2.9 2.9 8.5 0.03 1.7 1.8 131 12.1 1.2 1.4 1.5
1.9 2.6 8.1 0.9 0.2 0.1 5.1 0.7 0.6 0.2 1.6 1.2 0.1 0.04 0.7 0.1 0.4 1.6 1.7
76.2 62.4 67.4 1.1 11.4 0.02 63.9 5.4 339 0.1 0.2 2.3 0.2 0.2 33.5 23.0 0.2 7.3 7.6
1.9 14.6 9.7 217 111 1150 3.9 22.9 11.7 1070 6.2 2.8 65.5 6220 43.5 103 405 30.3 29.2
5.9 66.2 115 4520 33.3 12400 28.1 59.0 17.1 7930 50.8 54.9 82.8 6640 39.3 37.5 1790 180 175
9.8 1070 1870 1460 9.7 800 362 42.6 108 840 64.8 4.2 62.7 2640 18.1 259 336 630 683
3.5 373 797 31700 3.1 8520 101 49.8 44.7 5960 59.1 83.8 70.9 27800 16.3 41.2 1320 1130 1110
Figure 4. (A) Correlation between Cmax and (B) AUC values in plasma of 34 selected compounds after virtual oral administration (1.0 mg/kg) to rats estimated using one-compartment models and PBPK models. (C) Correlation between plasma and liver AUC values of 53 compounds obtained using PBPK models. The 53 compounds are made up of the 34 and 19 selected compounds shown in Tables 2 and 3, respectively.
PBPK-modeled liver AUC values (r = −0.78, p < 0.05, n = 7, Figure 5).
The varied selection of 34 chemicals that underwent onecompartment modeling also underwent PBPK modeling (these 34 chemicals are included in the 53 chemicals mentioned above). The plasma concentrations after virtual oral administration of 1.0 mg/kg doses of the 34 chemicals were estimated using both one-compartment and PBPK models. The maximum plasma concentrations (Cmax, Figure 4A) and the areas under the concentration−time curves (AUC, Figure 4B) of the 34 disparate chemicals after virtual oral administration (1.0 mg/kg) to rats obtained using highthroughput one-compartment toxicokinetic models (Figure 1A) and those obtained using our simple PBPK models (Figure 1B) were highly consistent (r = 0.98 and 0.99, respectively). In contrast, for the combined 53 compounds shown in Tables 2 and 3, the maximum hepatic and plasma concentrations (data not shown in Figure 4) and AUC values (r = 0.62, Figure 4C) were roughly correlated. However, the PBPK-modeled and empirically obtained values in the liver and plasma differed considerably. Finally, an inverse correlation was observed under the present conditions between the seven compounds with reported LOEL values for hepatotoxicity and
■
DISCUSSION The current study employed the simple one-compartment model recently recommended by US authorities as a highthroughput toxicokinetic model4 and our simple (PBPK)
Figure 5. There was an inverse correlation between reported LOEL values and the liver AUC values of seven compounds. The seven chemicals with reported hepatic LOEL values (Table 1) were evaluated. 215
DOI: 10.1021/acs.chemrestox.8b00307 Chem. Res. Toxicol. 2019, 32, 211−218
Article
Chemical Research in Toxicology model5,6,9 consisting of a chemical receptor compartment, a metabolizing compartment, and a central compartment (Figure 1). For a varied selection of 34 chemicals, the plasma AUC and maximum concentrations obtained using high-throughput toxicokinetic models and our simple PBPK models were quite consistent (Figure 4A,B). However, the maximum hepatic and plasma concentrations and the hepatic and plasma AUC values of 53 disparate chemicals obtained using the PBPK model were different (Figure 4C). Although the number of compounds (n = 7) for which represented a broad diversity of structures in the present chemical space and LOEL values for hepatoxicity were available [from the Hazard Evaluation Support System Integrated Platform (HESS)10 was limited, the LOEL values and the AUC values estimated using PBPK modeling were inversely correlated (Figure 5). On the basis of the present results, parameter estimation for PBPK modeling of new chemicals should be undertaken in future research projects. To establish a PBPK modeling-based hepatotoxicity prediction system, it would be worthwhile to set up the prediction equations for key parameters used in this simple PBPK modeling approach (Figure 1B). For example, using physiological parameters, we succeeded in estimating suitable albumin concentrations for enhanced drug oxidation activities mediated by liver microsomal cytochrome P450 enzymes.11 When more precise absorption constants of targeted compounds for one-compartment or PBPK models should be needed, it would be worthwhile to measure permeability values in vitro, which could be converted to absorption parameters in vivo. To establish the validity of estimated parameters required for PBPK modeling, the variety of informative compounds in the chemical space should be as wide as possible, as shown in Figure 2. In conclusion, we found an inverse relationship between NOEL levels and chemical absorbance rates after oral administration and an inverse relationship between LOEL levels and PBPK-modeled hepatic AUC values. Furthermore, the maximum plasma concentrations and plasma AUCs estimated using the one-compartment model and our PBPK model were highly consistent. This study provides important information to help simulate, in addition to plasma levels, the high hepatic levels of potent hepatotoxic compounds. Although metabolic activation of protoxicants by livers could not be ruled out for hepatoxicity, estimated accumulation of intake compounds orally in livers by PBPK modeling should be one of the determinant factors for potential of chemicals for hepatotoxicity. The present PBPK models with suitable parameters could estimate the relationships between plasma and hepatic concentrations of chemicals and drugs after virtual oral doses using both forward and reverse dosimetry, thereby indicating their potential value in predicting hepatic toxicity as a part of risk assessments of chemicals absorbed in the human body.
■
Funding
This work was supported in part by the METI Artificial Intelligence-based Substance Hazard Integrated Prediction System Project, Japan. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We are grateful to Makiko Shimizu, Norie Murayama, Miyu Iwasaki, Yui Kobayashi, Ayane Nakano, Ushio Ohnishi, Tatsurou Sasaki, and Manae Yoshizawa for their assistance. We also thank David Smallbones for his advice on English language usage.
■
REFERENCES
(1) Bell, S. M., Chang, X., Wambaugh, J. F., Allen, D. G., Bartels, M., Brouwer, K. L. R., Casey, W. M., Choksi, N., Ferguson, S. S., Fraczkiewicz, G., Jarabek, A. M., Ke, A., Lumen, A., Lynn, S. G., Paini, A., Price, P. S., Ring, C., Simon, T. W., Sipes, N. S., Sprankle, C. S., Strickland, J., Troutman, J., Wetmore, B. A., and Kleinstreuer, N. C. (2018) In vitro to in vivo extrapolation for high throughput prioritization and decision making. Toxicol. In Vitro 47, 213−227. (2) Rotroff, D. M., Wetmore, B. A., Dix, D. J., Ferguson, S. S., Clewell, H. J., Houck, K. A., LeCluyse, E. L., Andersen, M. E., Judson, R. S., Smith, C. M., Sochaski, M. A., Kavlock, R. J., Boellmann, F., Martin, M. T., Reif, D. M., Wambaugh, J. F., and Thomas, R. S. (2010) Incorporating human dosimetry and exposure into highthroughput in vitro toxicity screening. Toxicol. Sci. 117, 348−358. (3) Wetmore, B. A., Wambaugh, J. F., Ferguson, S. S., Sochaski, M. A., Rotroff, D. M., Freeman, K., Clewell, H. J., III, Dix, D. J., Andersen, M. E., Houck, K. A., Allen, B., Judson, R. S., Singh, R., Kavlock, R. J., Richard, A. M., and Thomas, R. S. (2012) Integration of dosimetry, exposure, and high-throughput screening data in chemical toxicity assessment. Toxicol. Sci. 125, 157−174. (4) Wambaugh, J. F., Hughes, M. F., Ring, C. L., MacMillan, D. K., Ford, J., Fennell, T. R., Black, S. R., Snyder, R. W., Sipes, N. S., Wetmore, B., Westerhout, J., Setzer, R. W., Pearce, R., Simmons, J. E., and Thomas, R. S. (2018) Evaluating in vitro−in vivo extrapolation of toxicokinetics. Toxicol. Sci. 163, 152. (5) Takano, R., Murayama, N., Horiuchi, K., Kitajima, M., Kumamoto, M., Shono, F., and Yamazaki, H. (2010) Blood concentrations of acrylonitrile in humans after oral administration extrapolated from in vivo rat pharmacokinetics, in vitro human metabolism, and physiologically based pharmacokinetic modeling. Regul. Toxicol. Pharmacol. 58, 252−258. (6) Yamazaki, H., Suemizu, H., Mitsui, M., Shimizu, M., and Guengerich, F. P. (2016) Combining chimeric mice with humanized liver, mass spectrometry, and physiologically-based pharmacokinetic modeling in toxicology. Chem. Res. Toxicol. 29, 1903−1911. (7) Bishop, C. M., Svensén, M., and Williams, C. K. I. (1998) GTM: the generative topographicmapping. Neural Comput 10, 215−234. (8) Cheng, K. C., Li, C., and Uss, A. S. (2008) Prediction of oral drug absorption in humans−from cultured cell lines and experimental animals. Expert Opin. Drug Metab. Toxicol. 4, 581−590. (9) Miura, T., Suemizu, H., Goto, M., Sakai, N., Iwata, H., Shimizu, M., and Yamazaki, H. Human urinary concentrations of monoisononyl phthalate estimated using physiologically based pharmacokinetic modeling and experimental pharmacokinetics in humanizedliver mice orally administered with diisononyl phthalate. Xenobiotica 2018, in press, DOI: 10.1080/00498254.2018.1471753 (10) Sakuratani, Y., Zhang, H. Q., Nishikawa, S., Yamazaki, K., Yamada, T., Yamada, J., Gerova, K., Chankov, G., Mekenyan, O., and Hayashi, M. (2013) Hazard Evaluation Support System (HESS) for predicting repeated dose toxicity using toxicological categories. SAR QSAR Environ. Res. 24, 351−363. (11) Shimura, K., Murayama, N., Tanaka, S., Onozeki, S., and Yamazaki, H. Suitable albumin concentrations for enhanced drug
AUTHOR INFORMATION
Corresponding Author
*Phone: +81-42-721-1406. Fax: +81-42-721-1406. E-mail:
[email protected]. ORCID
Kimito Funatsu: 0000-0002-9368-0302 Hiroshi Yamazaki: 0000-0002-1068-4261 Author Contributions §
These authors contributed equally. 216
DOI: 10.1021/acs.chemrestox.8b00307 Chem. Res. Toxicol. 2019, 32, 211−218
Article
Chemical Research in Toxicology oxidation activities mediated by human liver microsomal cytochrome P450 2C9 and other forms predicted with unbound fractions and partition/distribution coefficients of model substrates. Xenobiotica 2018, in press, DOI: 10.1080/00498254.2018.1482576. (12) Yamazaki, H., Kunikane, E., Nishiyama, S., Murayama, N., Shimizu, M., Sugiyama, Y., Chiba, K., and Ikeda, T. (2015) Human plasma concentrations of tolbutamide and acetaminophen extrapolated from in vivo animal pharmacokinetics using in vitro human hepatic clearances and simple physiologically based pharmacokinetic modeling for radio-labeled microdose clinical studies. Radioisotopes 64, 509−519. (13) Wang, Y., Roy, A., Sun, L., and Lau, C. E. (1999) A doublepeak phenomenon in the pharmacokinetics of alprazolam after oral administration. Drug Metab. Dispos. 27, 855−859. (14) Nguyen, H. Q., Stamatis, S. D., and Kirsch, L. E. (2015) A novel method for assessing drug degradation product safety using physiologically-based pharmacokinetic models and stochastic risk assessment. J. Pharm. Sci. 104, 3101−3119. (15) Zhang, W. L., Lou, D., Zhang, D. T., Zhang, Y., and Huang, H. J. (2016) Determination of rivaroxaban, apixaban and edoxaban in rat plasma by UPLC-MS/MS method. J. Thromb. Thrombolysis 42, 205− 211. (16) Borkar, N., Holm, R., Yang, M., Mullertz, A., and Mu, H. (2016) In vivo evaluation of lipid-based formulations for oral delivery of apomorphine and its diester prodrugs. Int. J. Pharm. 513, 211−217. (17) Mattiuz, E. L., Ponsler, G. D., Barbuch, R. J., Wood, P. G., Mullen, J. H., Shugert, R. L., Li, Q., Wheeler, W. J., Kuo, F., Conrad, P. C., and Sauer, J. M. (2003) Disposition and metabolic fate of atomoxetine hydrochloride: pharmacokinetics, metabolism, and excretion in the Fischer 344 rat and beagle dog. Drug Metab. Dispos. 31, 88−97. (18) Shepard, R. M., and Falkner, F. C. (1990) Pharmacokinetics of azithromycin in rats and dogs. J. Antimicrob. Chemother. 25 (Suppl A), 49−60. (19) Catanese, B., Grasso, A., and Silverstrini, B. (1966) Studies on the absorption and elimination of benzydamine in the mouse, rat, dog, and man. Arzneimittelforschung 16, 1354−1357. (20) Sakai, C., Iwano, S., Yamazaki, Y., Ando, Y., Nakane, F., Kouno, M., Yamazaki, H., and Miyamoto, Y. (2014) Species differences in the pharmacokinetic parameters of cytochrome P450 probe substrates between experimental animals, such as mice, rats, dogs, monkeys, and microminipigs, and humans. J. Drug Metab. Toxicol. 5, 173. (21) Li, J., Fang, J., Zhong, F., Li, W., Tang, Y., Xu, Y., Mao, S., and Fan, G. (2014) Development and validation of a liquid chromatography/tandem mass spectrometry assay for the simultaneous determination of dabigatran etexilate, intermediate metabolite and dabigatran in 50μL rat plasma and its application to pharmacokinetic study. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 973, 110−119. (22) Miyazaki, M., Maekawa, C., Iwanaga, K., Morimoto, K., and Kakemi, M. (2000) Bioavailability assessment of disopyramide using pharmacokinetic-pharmacodynamic (PK-PD) modeling in the rat. Biol. Pharm. Bull. 23, 1363−1369. (23) Liu, S., Shinkai, N., Kakubari, I., Saitoh, H., Noguchi, K., Saitoh, T., and Yamauchi, H. (2008) Automated analysis of fluvoxamine in rat plasma using a column-switching system and ion-pair highperformance liquid chromatography. Biomed. Chromatogr. 22, 1442− 1449. (24) Mushiroda, T., Douya, R., Takahara, E., and Nagata, O. (2000) The involvement of flavin-containing monooxygenase but not CYP3A4 in metabolism of itopride hydrochloride, a gastroprokinetic agent: comparison with cisapride and mosapride citrate. Drug Metab. Dispos. 28, 1231−1237. (25) Veeraraghavan, S., Thappali, S., Viswanadha, S., Nalla, S., Chennupati, S., Golla, M., Vakkalanka, S., and Rangasamy, M. (2014) Simultaneous quantification of idelalisib, fludarabine and lenalidomide in rat plasma by using high-performance liquid chromatography coupled with heated electrospray ionization tandem mass spectrometry. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 949−950, 63− 69.
(26) Morinaga, Y., Fuke, C., Arao, T., and Miyazaki, T. (2004) Quantitative analysis of cresol and its metabolites in biological materials and distribution in rats after oral administration. Leg. Med. 6, 32−40. (27) Tsukada, A., Suemizu, H., Murayama, N., Takano, R., Shimizu, M., Nakamura, M., and Yamazaki, H. (2013) Plasma concentrations of melengestrol acetate in humans extrapolated from the pharmacokinetics established in in vivo experiments with rats and chimeric mice with humanized liver and physiologically based pharmacokinetic modeling. Regul. Toxicol. Pharmacol. 65, 316−324. (28) Ma, Y. R., Rao, Z., Shi, A. X., Wang, Y. F., Huang, J., Han, M., Wang, X. D., Jin, Y. W., Zhang, G. Q., Zhou, Y., Zhang, F., Qin, H. Y., and Wu, X. A. (2016) Simultaneous determination of metformin, metoprolol and its metabolites in rat plasma by LC-MS-MS: Application to pharmacokinetic interaction study. J. Chromatogr. Sci. 54, 1−9. (29) Sakemi, K., Ito, R., Umemura, T., Ohno, Y., and Tsuda, M. (2002) Comparative toxicokinetic/toxicodynamic study of rubber antioxidants, 2-mercaptobenzimidazole and its methyl substituted derivatives, by repeated oral administration in rats. Arch. Toxicol. 76, 682−691. (30) Teirlynck, O. A., and Belpaire, F. (1985) Disposition of orally administered di-(2-ethylhexyl) phthalate and mono-(2-ethylhexyl) phthalate in the rat. Arch. Toxicol. 57, 226−230. (31) Doerge, D. R., Twaddle, N. C., Churchwell, M. I., Chang, H. C., Newbold, R. R., and Delclos, K. B. (2002) Mass spectrometric determination of p-nonylphenol metabolism and disposition following oral administration to Sprague-Dawley rats. Reprod. Toxicol. 16, 45− 56. (32) Ogihara, T., Kano, T., Wagatsuma, T., Wada, S., Yabuuchi, H., Enomoto, S., Morimoto, K., Shirasaka, Y., Kobayashi, S., and Tamai, I. (2009) Oseltamivir (tamiflu) is a substrate of peptide transporter 1. Drug Metab. Dispos. 37, 1676−1681. (33) Iqbal, M., Ezzeldin, E., Al-Rashood, K. A., and Shakeel, F. (2015) A validated UPLC-MS/MS assay using negative ionization mode for high-throughput determination of pomalidomide in rat plasma. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 983−984, 76−82. (34) Withey, J. R. (1976) Quantitative analysis of styrene monomer in polystyrene and foods including some preliminary studies of the uptake and pharmacodynamics of the monomer in rats. Environ. Health Perspect. 17, 125−133. (35) Schauer, U. M., Volkel, W., and Dekant, W. (2006) Toxicokinetics of tetrabromobisphenol A in humans and rats after oral administration. Toxicol. Sci. 91, 49−58. (36) Eriksson, T., Bjorkman, S., Fyge, A., and Ekberg, H. (1992) Determination of thalidomide in plasma and blood by highperformance liquid chromatography: avoiding hydrolytic degradation. J. Chromatogr., Biomed. Appl. 582, 211−216. (37) Lau-Cam, C. A., and Piemontese, D. (1998) Simplified reversed-phase HPLC method with spectrophotometric detection for the assay of verapamil in rat plasma. J. Pharm. Biomed. Anal. 16, 1029−1035. (38) Zhu, M., Chan, K. W., Ng, L. S., Chang, Q., Chang, S., and Li, R. C. (1999) Possible influences of ginseng on the pharmacokinetics and pharmacodynamics of warfarin in rats. J. Pharm. Pharmacol. 51, 175−180. (39) Teeguarden, J. G., Waechter, J. M., Jr., Clewell, H. J., III, Covington, T. R., and Barton, H. A. (2005) Evaluation of oral and intravenous route pharmacokinetics, plasma protein binding, and uterine tissue dose metrics of bisphenol A: a physiologically based pharmacokinetic approach. Toxicol. Sci. 85, 823−838. (40) Timchalk, C., Nolan, R. J., Mendrala, A. L., Dittenber, D. A., Brzak, K. A., and Mattsson, J. L. (2002) A physiologically based pharmacokinetic and pharmacodynamic (PBPK/PD) model for the organophosphate insecticide chlorpyrifos in rats and humans. Toxicol. Sci. 66, 34−53. (41) Yamazaki, H., Takano, R., Horiuchi, K., Shimizu, M., Murayama, N., Kitajima, M., and Shono, F. (2010) Human blood 217
DOI: 10.1021/acs.chemrestox.8b00307 Chem. Res. Toxicol. 2019, 32, 211−218
Article
Chemical Research in Toxicology concentrations of dichlorodiphenyltrichloroethane (DDT) extrapolated from metabolism in rats and humans and physiologically based pharmacokinetic modeling. J. Health Sci. 56, 566−575. (42) Beliveau, M., Tardif, R., and Krishnan, K. (2003) Quantitative structure−property relationships for physiologically based pharmacokinetic modeling of volatile organic chemicals in rats. Toxicol. Appl. Pharmacol. 189, 221−232. (43) Takano, R., Murayama, N., Horiuchi, K., Kitajima, M., Shono, F., and Yamazaki, H. (2010) Blood concentrations of 1,4-dioxane in humans after oral administration extrapolated from in vivo rat pharmacokinetics, in vitro human metabolism, and physiologically based pharmacokinetic modeling. J. Health Sci. 56, 557−565. (44) Yamashita, M., Suemizu, H., Murayama, N., Nishiyama, S., Shimizu, M., and Yamazaki, H. (2014) Human plasma concentrations of herbicidal carbamate molinate extrapolated from the pharmacokinetics established in in vivo experiments with chimeric mice with humanized liver and physiologically based pharmacokinetic modeling. Regul. Toxicol. Pharmacol. 70, 214−221. (45) Yamazaki, H., Horiuchi, K., Takano, R., Nagano, T., Shimizu, M., Kitajima, M., Murayama, N., and Shono, F. (2010) Human blood concentrations of cotinine, a biomonitoring marker for tobacco smoke, extrapolated from nicotine metabolism in rats and humans and physiologically based pharmacokinetic modeling. Int. J. Environ. Res. Public Health 7, 3406−3421. (46) Breen, K. J., Shaw, J., Alvin, J., Henderson, G. I., Hoyumpa, A. M., Jr., and Schenker, S. (1973) Effect of experimental hepatic injury on the clearance of phenobarbital and paraldehyde. Gastroenterology 64, 992−1004. (47) Sharma, R., Litchfield, J., Atkinson, K., Eng, H., Amin, N. B., Denney, W. S., Pettersen, J. C., Goosen, T. C., Di, L., Lee, E., Pfefferkorn, J. A., Dalvie, D. K., and Kalgutkar, A. S. (2014) Metabolites in safety testing assessment in early clinical development: a case study with a glucokinase activator. Drug Metab. Dispos. 42, 1926−1939. (48) Hooshfar, S., Mortuza, T. B., Rogers, C. A., Linzey, M. R., Gullick, D. R., Bruckner, J. V., White, C. A., and Bartlett, M. G. (2017) Gas chromatography/negative chemical ionization mass spectrometry of transfluthrin in rat plasma and brain. Rapid Commun. Mass Spectrom. 31, 1573−1581. (49) Lee, C. H., Wang, C. L., Lin, H. F., Chai, C. Y., Hong, M. Y., and Ho, C. K. (2011) Toxicity of tetramethylammonium hydroxide: review of two fatal cases of dermal exposure and development of an animal model. Toxicol. Ind. Health 27, 497−503. (50) Nnane, I. P., and Damani, L. A. (2001) Pharmacokinetics of trimethylamine in rats, including the effects of a synthetic diet. Xenobiotica 31, 749−755. (51) Swiercz, R., Majcherek, W., and Wasowicz, W. (2016) Hemimellitene (1,2,3-trimethylbenzene) in the liver, lung, kidney, and blood, and dimethylbenzoic acid isomers in the liver, lung, kidney and urine of rats after single and repeated inhalation exposure to hemimellitene. Int. J. Occup. Med. Environ. Health 29, 113−128. (52) Swiercz, R., Rydzynski, K., Wasowicz, W., Majcherek, W., and Wesolowski, W. (2002) Toxicokinetics and metabolism of pseudocumene (1,2,4-trimethylbenzene) after inhalation exposure in rats. Int. J. Occup. Med. Environ. Health 15, 37−42.
218
DOI: 10.1021/acs.chemrestox.8b00307 Chem. Res. Toxicol. 2019, 32, 211−218
