Poster
T4077
Modeling of Active Transport and Metabolism for Hepatocyte Assays
with Application of In Vitro to In Vivo Extrapolation (IVIVE)
James Mullin , Viera Lukacova , Walter S. Woltosz , Michael B. Bolger
1
Simulations Plus, Inc.
1
1
1
QR Code
Only
1
DEVELOPING SCIENCE. IMPACTING HEALTH.
CONTACT INFORMATION: [email protected], Simulations Plus, Inc., 42505 10th Street West, Lancaster, CA 93534
PURPOSE
METHODS (CONT.)
Sandwich and suspended hepatocyte cultures are routinely used to assess either active transport
and/or metabolism of drug molecules. In vitro assays that evaluate critical drug clearance and
metabolism pathways are important in the prediction of hepatobiliary transport, drug-drug
interactions, drug induced liver injury (DILI), and the relative importance between active transport
and metabolism in hepatocytes. Physiologically based pharmacokinetic (PBPK) models provide
quantitative in vivo simulation of drug disposition and drug-drug interactions if Km and Vmax
values for the relevant transporters or enzymes can be extracted accurately from in vitro data. To
that end, a fully mechanistic simulation of drug transport in both sandwich and suspended
hepatocytes was developed in MembranePlus (Simulations Plus, Inc.) that allows simultaneous
determination of Km and Vmax values for enzymes as well as both influx and efflux transporters.
Both models account for additional processes such as drug diffusion through the unstirred
boundary layer, sample volume loss, protein binding in both media and cytosol, lysosomal trapping,
and drug partitioning into lipid bilayers. Two case studies are presented to demonstrate the
applicability of these models. The active uptake and biliary secretion of sodium taurocholate is
simulated in the sandwich hepatocyte model. The CYP2D6 metabolism of propafenone is evaluated
in the suspended hepatocyte model and, in conjunction with a PBPK model, is used to predict the in
vivo exposure.
OBJECTIVES
1. Determine intracellular Km/Vmax values for metabolism in suspended hepatocyte cell culture.
2. To facilitate the generation of metabolism inputs for GastroPlus (Simulations Plus, Inc.).
3. Calculate the biliary transport and disposition of sodium taurocholate.
METHODS
The mechanistic models for transport and metabolism in sandwich and suspended hepatocytes
were implemented in MembranePlus version 2 and are shown in Figure 1.
SUSPENDED HEPATOCYTES
SANDWICH HEPATOCYTES
Figure 1. Physical model for drug transport in suspended hepatocytes (left) and sandwich
hepatocytes (right).
The partial differential equations that describe drug transport are solved using the method of lines in
rectangular or spherical 1D geometries. The sandwich hepatocyte model was utilized to determine
the Vmax for active uptake and efflux of sodium taurocholate as shown in Figure 2. Because
disposition data was only measured at one dose concentration, (1 mM) Km values for MRP4, OSTa/b,
and BSEP mediated transport of sodium taurocholate could not be fitted and were used as obtained
from literature4,5.
The Km values are shown in Table 3. ADMET
Predictor 8.0 (Simulations Plus, Inc.) was used to
determine physicochemical properties and passive
membrane transport parameters for each case study
as shown in Table 2. The suspended hepatocyte
model was used to predict passive diffusion and
CYP2D6 metabolism of propafenone across 4 dose
levels from 0.05 to 5 mM. Experimental data for
media, cell, bile concentrations, or plasma
concentrations, as well as, in vitro experimental
settings (Table 1) in these systems was obtained from
literature data.1,2,3
Table 1. Cell assay parameters for MembranePlus
simulations
Cell Assay Inputs
Na. Taur. Propafenone
Feed Solution Conc.
1
0.05, 0.2, 1, 5
mM
Protein in Basolateral
Fluid (BSA)
4
NA
%
Well size
24
6
well
Volume
0.3
2
mL
Cell Volume
6.46
NA
pL
Cell Thickness/Diameter
18.6
16.79
micron
Cell Density
0.4
1
Mcell/well
RESULTS
In sandwich hepatocyte cells, Vmax values of 0.0404,
0.0863, and 0.0737 mmol/s/L cytosol were obtained
for NTCP (influx), BSEP (bile efflux), and OSTa/b
(efflux) transport of sodium taurocholate. The fitted
concentration profiles and Vmax values are shown in
Figure 3 and Table 3, respectively. The resulting
prediction of media and cell concentration including
bile had a mean absolute error of 13.6%. The accuracy
figure encompasses both the uptake and efflux (or
wash) phase where the cells are exposed to drug-free
buffer. In suspended hepatocytes, Km and Vmax
values of 0.0146 mM and 0.0927 mmol/s/L cytosol
were obtained for the CYP2D6 metabolism of
propafenone and the fitted predictions are shown in
Figure 4. The resulting parameters were used in the
GastroPlus PBPK model to predict propafenone PK as
shown in Figure 5. The IVIVE was reasonably accurate
with a predicted and observed AUC of 1343 and 1032
ng-hr/mL (23% error). The result indicates that the
metabolic clearance predictions obtained from
MembranePlus analysis of in vitro data and the
Lukacova method for tissue partition coefficients are
an acceptable combination in this case and show the
utility of both software tools for IVIVE.
OSTa/b
Blood Flow
Hepatocyte
BSEP
Bile
Acids
Blood Flow
NTCP
RESULTS (CONT.)
Bile
Pocket
Bile
Acids
Figure 2. Schematic of sodium taurocholate uptake
Table 2. Properties of sodium taurocholate
and propafenone for simulations in
MembranePlus and GastroPlus
Property
Na. Taur. Propafenone
0.0184
0.49
S+Sw (mg/mL) @ pH @ pH 2.78 @ pH 10.28
S+pKa (Base)
1.1
9.47
S+SF
906
114
S+logP
0.82
3.03
S+Peff (x10-4 cm/s)
0.36
1.76
DiffCoef (x10-5 cm/s)
0.53
0.65
Rbp
0.59
0.81
Fu plasma
16.8
19.03
Figure 4. Km/Vmax fitting results for propafenone in
suspended hepatocyte culture (left)
Figure 5. GastroPlus IVIVE prediction of
70 mg IV bolus dose in human (right)
CONCLUSION
The sandwich and suspended hepatocyte models within MembranePlus are
new utilities for scientists to analyze their in vitro experiments and extract
relevant Km and Vmax parameters for active transport and metabolism. We
have shown that the models and parameters extracted from MembranePlus
can then be utilized in a PBPK model to predict in vivo exposure. In the
future, we will expand the capabilities of the new hepatocyte models to
determine relevant in vitro drug-drug interaction parameters for predictions
of in vivo DDI potential.
Efflux Phase
REFERENCE
Figure 3. Model results for extraction of Vmax
values. For biliary excretion of sodium taurocholate
Table 3. Fitted Vmax values for sodium
taurocholate disposition in sandwich hepatocytes
Km Exp
Vmax Fit
Transporter (mM) (mmol/s/L)
BSEP
25.8
6.83E-03
OSTa/b
6
9.63E-02
NTCP
5
3.88E-02
Km Literature Source
Swift-Mol-Pharm-2010-7(2)-491500
J-Exp-Biol-2001-204-1673-1686
J-Exp-Biol-2001-204-1673-1686
1.
2.
3.
4.
5.
Guo, Cen, et al. ISSX (2014).
Komura, Hiroshi, et al., Drug metabolism and disposition 33.6 (2005): 726-732.
Schlepper and Olsson, Cardiac Arrhythmias, Springer-Verlag (1983): 125-132.
Swift, et al., Molecular Pharmaceutics 7.2 (2009): 491-500.
St-Pierre, et al., Journal of Experimental Biology 204.10 (2001): 1673-1686.
