システム生物学に基づいた薬剤作用機序推定法に関 する研究
TRANSCRIPT
九州大学学術情報リポジトリKyushu University Institutional Repository
システム生物学に基づいた薬剤作用機序推定法に関する研究
齊藤, 隆太
https://doi.org/10.15017/1807110
出版情報:Kyushu University, 2016, 博士(農学), 課程博士バージョン:権利関係:Fulltext available.
Studies on “Estimation of the Mechanism-of-Action of
Pharmaceutical Compounds Based on Systems Biology
Approach”
Ryuta Saito
Graduate School of Bioresource and Bioenvironmental Sciences Kyushu University
2 0 1 7
2
Table of Contents
1 Abstract
2 Introduction
3 Development of a computational model of adrenal steroidogenesis in human
adrenocortical carcinoma NCI-H295R cells
3.1 Materials and Methods
3.2 Results 3.3 Discussion
3.4 Tables and Figures
4 Estimating mechanism of adrenal action of endocrine-disrupting compounds
by using a hybrid optimization method of real-coded genetic algorithm and local search based on non-linear least squares method
4.1 Materials and Methods
4.2 Results
4.3 Discussion
4.4 Tables and Figures
5 Conclusions
6 Acknowledgements
7 References
3
1. Abstract
Recently systems biology is in an exponential development stage and has been widely used
in drug discovery and development to deeply understand molecular basis of disease and
pharmacological action. However, most of these conventional researches have been based on
the exploratory approach, and comprehensive technologies for quantitative estimation of
mechanism of drug action based on differentially omics analysis, such as metabolic profiling,
are not widely reported. The author has investigated about an integrative computational
methodology for elucidation of mechanism of drug action based on differentially metabolic
profiling as an application of systems biology. In this paper, I particularly focused on the
adrenal toxicity by endocrine-active compounds.
Adrenal toxicity is one of the major concerns in drug development. To quantitatively
understand the effect of endocrine-active compounds on adrenal steroidogenesis, and to
assess the human adrenal toxicity of novel pharmaceutical drugs, I developed a mathematical
model of steroidogenesis in human adrenocortical carcinoma NCI-H295R cells. The model
includes cellular proliferation, intracellular cholesterol translocation, diffusional transport of
steroids, and metabolic pathways of adrenal steroidogenesis, which serially involve
steroidogenic proteins and enzymes such as StAR, CYP11A1, CYP17A1, HSD3B2,
CYP21A2, CYP11B1, CYP11B2, HSD17B3, and CYP19A1. It was reconstructed in an
experimental dynamics of cholesterol and 14 steroids from an in vitro steroidogenesis assay
using NCI-H295R cells. Results of dynamic sensitivity analysis suggested that HSD3B2
plays the most important role in the metabolic balance of adrenal steroidogenesis. Based on
differential metabolic profiling of 12 steroid hormones and 11 adrenal toxic compounds, I
could estimate which steroidogenic enzymes were affected in this mathematical model. In
terms of adrenal steroidogenic inhibitors, the predicted action sites were approximately
matched to reported target enzymes. Thus, proposed computer-aided system based on a
4
systems biological approach may be useful to understand the mechanism of action of
endocrine-active compounds and to assess the human adrenal toxicity of novel
pharmaceutical drugs.
5
2. Introduction
Applications of systems biology for drug discovery
Systems biology provides a framework for constructing mathematical models of biological
and physiological systems from systematic measurements of multiple molecular levels.
Recently systems biology is in an exponential development stage and has been widely used in
drug discovery and development to better understand molecular basis of disease and
mechanism of drug action [1]. By considering the biological context of drug target, systems
biology, biological network analyses and dynamical modeling, provides new opportunities to
address disease mechanisms and approach drug discovery, which will facilitate the translation
of preclinical discoveries into clinical benefits such as novel biomarkers and therapies [2].
The application of computational and experimental systems biology methods to
pharmacology allows us to introduce the definition of “systems pharmacology” [3], which
describes a field of research that provides us with a comprehensive view of drug action
rooted in molecular interactions between drugs and their targets in a human cellular context.
Advances in systems pharmacology will, in the long term, assist in the development of new
drugs and more effective therapies for patient treatment management. There are several
important clinically motivated applications in drug discovery to which systems biological
approaches make significant contributions, such as drug-target networks [4 - 8], predictions
of drug-target interactions [9 - 12], investigations of the adverse effects of drugs [13 - 18],
drug repositioning [19 - 23], and predictions of drug combination [24 - 30]. However, most of
these conventional researches have been based on the exploratory approach, and
comprehensive technology for understanding of mechanism of drug action based on
differential analysis on systemic dynamics of molecules is not widely reported. And, an
unsolved problem is still remained in drug discovery and development, that it is unclear the
understanding of the biological systems and the mechanism of drug action, for example,
6
idiosyncratic drug-induced liver injury, proarrhythmia, and adrenal toxicity. Therefore, I
focused on the elucidation of mechanism of action of the endocrine-disrupting compounds for
the adrenal toxicity as an application of the systems biology to risk assessment of the clinical
adrenal toxicity.
Assessment of human adrenal toxicity on drug discovery
Because steroid hormones play an important role in a wide range of physiological processes,
the potential to disturb endocrine effects is a major concern in the development of novel
pharmaceutical drugs such as etomidate and aminoglutethimide [31]. The adrenal gland is the
most common target for toxicity in the endocrine system in vivo, because steroid hormones
are primarily synthesized through enzymatic reactions in the adrenal cortex [32 - 35]. Indeed,
in these studies based on chemically induced endocrine lesions observed in in vivo toxicity,
the most frequent site of reported effects was the adrenal gland. Therefore, the prediction of
human adrenal toxicity based on the mechanism of on- or off-target actions in the early stages
of drug discovery and development is important.
The NCI-H295R human adrenocortical carcinoma cell line has been used to elucidate
mechanisms of adrenal steroidogenic disrupting compounds [31, 36 - 40]. The NCI-H295R
cell line was established by Gazder and his collaborators in 1990 [41], which expressed all
key steroidogenic enzymes and steroidogenesis-related proteins [41 - 43]. H295R cells have
the physiological characteristics of zonally undifferentiated human fetal adrenal cells and the
ability to produce steroid hormones found in the adult adrenal cortex [31, 41, 43]. In vitro
bioassays using the NCI-H295R human cell line have been able to evaluate the effects of
chemicals on steroid hormone production [38, 44 - 48], steroidogenic enzyme activities [44,
49, 50], and the expression of steroidogenic genes [44, 51 - 55]. In transcriptome studies, the
mechanisms of action of many steroidogenic disrupting compounds have been qualitatively
elucidated in terms of adrenal toxicity. However, gene expression does not always reflect the
7
production of steroid hormones [53]. Furthermore, measuring a few specific steroid
hormones may not be a useful approach to study the mechanisms of steroidogenic disrupting
effects in complex pathways such as adrenal steroidogenesis. To systematically understand
how exogenous compounds affect adrenal steroidogenesis, simultaneous determination of all
detectable steroid hormones and integrative analysis of these complex data would be
important. As an exploratory approach to analysis of complex data, ToxClust developed by
Zhang and colleagues in 2009 is able to visualize concentration-dependent response
relationships in the characteristics of chemically induced toxicological effects [56]. However,
this exploratory approach is unable provide a quantitative understanding of the mechanism of
action of adrenal toxicants or reveal systematic information about the effect of each
enzymatic reaction, interactions, and feedback in the adrenal steroidogenesis pathway.
Adrenal steroidogenesis and systems biology
Systems biology based on computational models of biological processes and the
comprehensive measurement of biological molecules is the most powerful approach to
quantitatively understand the influence of each factor in complex biological pathways. In
recent studies by Breen and his colleagues, a computational model of adrenal steroidogenesis
has been developed in NCI-H295R cells, including the steroidogenic disrupting effects of
metyrapone to inhibit enzymatic reactions of CYP11B1 [57, 58]. These models reproduce the
dynamics of adrenal steroidogenesis in NCI-H295R cells and the influence of metyrapone. A
current computational model of adrenal steroidogenesis was incorporated with a reaction of
oxysterol synthesis as a bypass to consume cellular cholesterol [58]. In addition, all reactions
in this model were described by a kinetic equation of the first-order reaction [58]. It is
difficult to quantitatively evaluate the influence of each protein in the complicated system of
adrenal steroidogenesis using the reported models, because it is simple and any biochemical
and cellular biological information is not sufficient. For example, to clearly understand the
8
cause of the change from the differentially dynamic patterns of steroid hormones, it is
necessary to consider the substrate inhibition of steroidogenic enzyme because most of
steroidogenic enzymes recognize multiple steroids as the enzymatic substrate. However, the
substrate inhibition of steroidogenic enzyme cannot be described by the mathematical model
based on kinetic equations of first-order reaction that does not consider the Michaelis
constant Km expressing the affinity of the substrate. To quantitatively estimate the
mechanism of steroidogenic disrupting compounds from comprehensive experimental data of
adrenal steroidogenesis in NCI-H295R cells, the reported model should be improved
according to the following two points. First, the kinetic equation of enzymatic reactions
should be exchanged from the first-order equation to a steady-state kinetic equation based on
the mechanism of the enzymatic reaction. Because a mathematical model organized by
first-order equations operates in a simple structure-dependent manner, it does not show
complex behavior based on molecular interactions, feedback, or regulation. Second,
intracellular localization processes of cholesterol should be incorporated as a considerable
mechanism. Because intracellular cholesterol molecules are stored as cholesterol esters or
widely distributed as membrane components, only a few cholesterol molecules localized on
the mitochondrial inner membrane are available for the adrenal steroidogenesis pathway [59,
60]. Moreover, a cholesterol-trafficking process from the outer to inner mitochondrial
membranes, which is regulated by steroidogenic acute regulatory (StAR) protein, is one of
the rate-limiting steps in adrenal steroidogenesis [60]. By overcoming these limitations in the
reported steroidogenesis model, systems analysis of adrenal steroidogenesis in NCI-H295R
cells may be able to quantitatively estimate the mechanism of action of steroidogenic
disrupting compounds.
Purpose of the present study
In the present study, to quantitatively estimate the toxicological mechanism of
9
endocrine-active compounds in adrenal steroidogenesis and to predict human adrenal toxicity
of novel pharmaceutical drugs in the drug discovery phase, I developed a novel
computational model of steroidogenesis in NCI-H295R cells. It includes cholesterol transport
into intracellular regions from the extracellular space, the cholesterol translocation system in
intracellular regions, including oxysterol synthesis, the metabolic pathway of adrenal
steroidogenesis, and transport of steroid hormones. Global sensitivity analysis of this adrenal
steroidogenesis model is able to evaluate the influence of each steroidogenic enzyme and
related protein for each steroid hormone observed in an in vitro steroidogenesis assay of
NCI-H295R cells. Furthermore, the mechanisms of action of steroidogenic disrupting
compounds for steroidogenic enzymes can be estimated by the optimization method to solve
the reverse problem from the concentration changes of 12 steroid hormones measured by
liquid chromatography/mass spectrometry in the steroidogenesis assay of NCI-H295R cells in
vitro. Using this developed model of adrenal steroidogenesis and the analytical approach, the
in vitro steroidogenesis assay of NCI-H295R cells can assess the human adrenal toxicity of a
novel pharmaceutical drug based on quantitative understanding of its toxicological
mechanism in adrenal steroidogenesis. This proposed strategy was summarized in Fig. 2-1.
10
Fig. 2-1. Advanced systems biology approach for estimating of mechanism-of-
action of pharmaceutical compounds.
11
3. Development of a computational model of adrenal steroidogenesis in human adrenocortical carcinoma NCI-H295R cells
3.1. Materials and Methods
Cell culture
NCI-H295R human adrenocortical carcinoma cells were purchased from the American Type
Culture Collection (Cat# CRL-2128, Manassas,VA) and cultured at 37 °C in a humidified
atmosphere with 5% CO2. The cells were maintained in a 1:1 mixture of Dulbecco’s modified
Eagle’s medium (DMEM; GIBCO) and F-12 medium (ICN Biomedicals) supplemented with
15 mM HEPES (Dojindo), 0.00625 mg/mL insulin (SIGMA), 0.00625 mg/mL transferrin
(SIGMA), 30 nM sodium selenite (WAKO), 1.25 mg/mL bovine serum albumin (BSA;
SIGMA), 0.00535 mg/mL linoleic acid (SIGMA), 2.5% Nu Serum (Becton, Dickinson and
Company), 100 U/mL penicillin (Meiji), and 100 mg/L streptomycin (Meiji).
Steroidogenesis in human adrenal corticocarcinoma NCI-H295R
cells
NCI-H295R cells were stimulated with adrenocorticotrophic hormone (ACTH), forskolin,
and angiotensin II to initiate steroidogenesis. Changes in steroid concentrations over time
were measured after stimulation in both cells and culture medium to construct a simulation
model.
The cells were seeded at 6×105 cells/well in 6-well plates. After 3 days of culture, the
culture medium was changed to stimulation medium consisting of DMEM/F-12 (1:1)
medium supplemented with 0.00625 mg/mL insulin, 0.00625 mg/mL transferrin, 30 nM
sodium selenite, 1.25 mg/mL BSA, 0.00535 mg/mL linoleic acid, 10% fetal bovine serum
(GIBCO), 100 U/mL penicillin, 100 mg/L streptomycin, 50 nM ACTH (SIGMA), 20 μM
forskolin (SIGMA), and 100 nM angiotensin II (CALBIOCHEM). Culture media and cells
12
were collected at 0, 8, 24, 48, and 72 h after stimulation. The cells were collected in 100 μL
distilled water and sonicated to produce a cell lysate. The cultures were conducted in four
wells/time point (N=4).
The concentrations of 12 steroids, pregnenolone (PREG), 17α-hydroxypregnenolone
(HPREG), dehydroepiandrosterone (DHEA), progesterone (PROG),
17α-hydroxyprogesterone (HPROG), androstenedione (DIONE), testosterone (TESTO),
11-deoxycorticosterone (DCORTICO), 11-deoxycortisol (DCORT), corticosterone
(CORTICO), cortisol (CORT), and aldosterone (ALDO), in the medium and cell lysate were
measured by LC/MS. Concentrations of estrone (E1) and 17β-estradiol (E2) were measured
by enzyme-linked immunosorbent assays (WAKO). In addition, the concentration of
cholesterol was measured using a commercial kit (WAKO) based on the cholesterol oxidase
method.
Liquid chromatography
A Shimadzu LC-VP series (Kyoto, Japan) consisting of an SIL-HTc autosampler,
LC-10ADvp Pump, CTO-10ACvp column oven, and DGU-14AM degasser was used to set
the reverse-phase liquid chromatographic conditions. The column was a Cadenza CD-C18
column (100 × 2 mm i.d., 3 μm; Imtakt Corp., Kyoto, Japan) used at 45 °C. The mobile phase
included water/acetonitrile/formic acid 95/5/0.05 (v/v/v, Solvent A) and
water/acetonitrile/formic acid 35/65/0.05 (v/v/v, Solvent B). The gradient elution programs
were 0% B (0–1 min with an isocratic gradient), 0–40% B (1–2 min with a linear gradient),
40% B (2–7 min with an isocratic gradient), 40–100% B (7–12 min with a linear gradient),
100% B (12–14 min with an isocratic gradient), 100–0% B (14–15 min with a linear gradient),
and 0% B (15–16 min with an isocratic gradient) at a flow rate of 0.3 mL/min. The
autosampler tray was cooled to 45 °C and the injection volume was 5 μL. HPLC grade
acetonitrile and formic acid were purchased from WAKO.
13
Mass spectrometry
A triple quadrupole mass spectrometer API4000 (Applied Biosystems/MDS Sciex, Concord,
Canada) coupled with an electrospray ionization source was operated in the positive ion
mode. The optimized ion source conditions were as follows: collision gas, 6 psi; curtain gas,
40 psi; ion source gas 1, 50 psi; ion source gas 2, 80 psi; ion source voltage, 5500 V; ion
source temperature, 600 °C. Nitrogen was used as the collision gas in the multiple reaction
monitoring (MRM) mode. The conditions of declustering potential, collision energy, and
collision cell exit potential were optimized by every steroid. The transitions in MRM were as
follows: PREG m/z 317 → 299, HPREG m/z 315 → 297, DHEA m/z 289 → 271, PROG m/z
315 → 109, HPROG m/z 331 → 109, DIONE m/z 287 → 97, DCORT m/z 331 → 123,
DCORTICO m/z 347 → 161, CORTICO m/z 347 → 100, CORT m/z 363 → 309, ALDO m/z
361 → 343, and TESTO m/z 289 → 109. Mass spectroscopic data were acquired and
quantified using the Analyst 1.4.2 software package (Applied Biosystems/MDS Sciex).
Estimation of the cell volume
Cell volume was estimated from the number of cells in the well and the average diameter of
the cells. Cells were detached from the well using 0.025% trypsin (MP Biomedicals) in a
0.02% EDTA solution (Dojindo) at the start of pre-culture, start of stimulation, and at 24, 48
and 72 h after stimulation. The numbers and diameters of the cells were measured by a cell
counter (Vi-cell XR 2.01; Beckman Coulter) after trypan blue staining. Parameters of the cell
volume and number of cells were estimated to fit experimental time-course data using
exponential curves.
Mathematical modeling of adrenal steroidogenesis in NCI-H295R
cells
Steroid hormones secreted from human adrenal corticocarcinoma NCI-H295R cells are
synthesized from cholesterol through the C21-steroid hormone biosynthesis pathway. A
14
mathematical model of steroidogenesis in NCI-H295R cells was constructed with cholesterol
transport and the intracellular localization pathway, the oxysterol synthesis pathway as a
bypass of steroidogenesis, the C21-steroid hormone biosynthesis pathway as the main
steroidogenesis pathway, passive transport of steroid hormones, and cell proliferation (Fig.
3-1). In this model, two compartments, the intracellular space and culture medium, were
incorporated as the available region.
Equations to estimate the proliferation of NCI-H295R cells have been proposed by Breen
and his colleagues [57, 58]. In this model, the dynamics of the total number of cells (N )
and total cell volume (V ) were implemented by the following equations:
N N 72 ∙ exp ∙
V V ∙ N
where, N 72 is the initial number of NCI-H295R cells per well before incubation
(6×105 cells), is the cell proliferation rate of NCI-H295R cells (0.00878 1/h),
is the incubation time until stimulation (72 h), is the culture time after stimulation, and
V is the mean cell volume of individual NCI-H295R cells (1,499 μm3).
Cholesterol transport and the intracellular localization pathway were modified as a part of
the ACTH-stimulated cortisol secretion model described by Dempsher and coauthors [61].
Intracellular cholesterol was separated into five states based on localization in the adrenal
cells. There are stored cholesterol esters in the lipid droplets (CHOS), free cytosolic
cholesterol (CHOC), free mitochondrial cholesterol (CHOM), free mitochondrial cholesterol
close to CYP11A1 enzymes (CHON), and free mitochondrial cholesterol remote from
CYP11A1 enzymes (CHOR). Cholesterols (CHOL) almost always exist as a cholesterol ester
(CE) in extracellular culture medium. Therefore, imported CHOL from the culture medium is
first stored in lipid droplets (transition to CHOS from medium CHOL). CHOS is transformed
15
to free CHOL by cholesterol ester hydrolase (CEH) and distributed to the cytosolic space
(transition from CHOS to CHOC). CHOC is transported into mitochondria from the cytosol
by StAR protein (transition from CHOC to CHOM). CHOM is continuously translocated in
the vicinity of CYP11A1 enzymes by StAR (transition from CHOM to CHON), so that
CHON is available for the adrenal steroidogenesis pathway. On the other hand, CHOM also
passively recedes from CYP11A1 enzymes (transition from CHOM to CHOR). Moreover,
the oxysterol biosynthesis pathway was incorporated as a bypass of the steroidogenesis
pathway, which was proposed by Breen et al [58]. In fact, the total mass balance of
cholesterol and all steroids is not preserved in this in vitro system. However, the oxysterol
biosynthesis pathway and/or bypass pathway are expected to exist in NCI-H295R cells. In
this model, the oxysterol biosynthesis pathway was defined to branch from CHOC to
oxysterol (OXY). These variables belonging to cholesterol transport and the intracellular
transport system were described by the following ordinary differential equations (ODEs):
d CHOLdt
VV
∙
dCHOSdt
V ∙
dCHOCdt
V ∙
dCHOMdt
V ∙
dCHONdt
V ∙
dCHORdt
V ∙
dOXYdt
V ∙
16
CHOLCHOS CHOC CHOM CHON CHOR
V
where, V is the volume of culture medium (2 mL), V is the total cell volume,
is the cholesterol import rate into the cytosolic space from the
extracellular culture medium, is the enzymatic reaction rate of CEH, is the
mitochondrial transport rate by the StAR protein, is the reaction rate of
oxysterol biosynthesis, is the passive diffusion rate from CYP11A1 enzymes, is
the translocation rate close to CYP11A1 enzymes by StAR, and is the enzymatic
reaction rate of the first adrenal steroidogenic enzyme, CYP11A1.
The C21-steroid hormone biosynthesis pathway includes 14 steroid hormones, PREG,
HPREG, DHEA, PROG, HPROG, DIONE, TESTO, DCORTICO, DCORT, CORTICO,
CORT, ALDO, E1 and E2, and 17 enzymatic reactions catalyzed by nine steroidogenic
enzymes, cholesterol side-chain cleavage enzyme (CYP11A1), 17α-hydroxylase (CYP17H),
C17,20-lyase (CYP17L), 3β-hydroxysteroid dehydrogenase (HSD3B2), 21-hydroxylase
(CYP21A2), 11β-hydroxylase (CYP11B1), 18-hydroxylase (CYP11B2), 17β-hydroxysteroid
dehydrogenase (HSD17B3), and aromatase (CYP19A1). These variable parameters
belonging to the adrenal steroidogenesis system were described by the following ODEs:
V ∙d PREG
dtV
V V ∙∙ , ,
V ∙d HPREG
dtV
V V ∙∙ , , ,
V ∙d DHEA
dtV
V V ∙∙ , ,
V ∙d PROG
dtV
V V ∙∙ , , ,
17
V ∙d HPROG
dt
VV V ∙
∙ , , , ,
V ∙d DIONE
dt
VV V ∙
∙ , , , ,
V ∙d TESTO
dtV
V V ∙∙ , ,
V ∙d DCORTICO
dtV
V V ∙∙ , ,
V ∙d DCORT
dtV
V V ∙∙ , ,
V ∙d CORTICO
dtV
V V ∙∙ ,
V ∙d CORT
dtV
V V ∙∙ ,
V ∙d ALDO
dtV
V V ∙∙
V ∙d E1
dtV
V V ∙∙ , ,
V ∙d E2
dtV
V V ∙∙ , ,
where, (dimensionless) is the equilibrium constant of steroid hormone x between the
cytosol and extracellular culture medium and (μM/h) is the enzymatic reaction rate of
steroidogenic enzyme X.
Descriptions of all passive transport reactions of steroid hormones were based on the
quasi-equilibrium reaction. Therefore, steroid concentrations in the culture medium were
calculated from each cytosolic concentration and equilibrium constant.
18
PREG ∙ PREG
HPREG ∙ HPREG
DHEA ∙ DHEA
PROG ∙ PROG
HPROG ∙ HPROG
DIONE ∙ DIONE
TESTO ∙ TESTO
DCORTICO ∙ DCORTICO
DCORT ∙ DCORT
CORTICO ∙ CORTICO
CORT ∙ CORT
ALDO ∙ ALDO
E1 ∙ E1
E2 ∙ E2
In this mathematical model of adrenal steroidogenesis in NCI-H295R cells, the flux
velocities of molecular transportation and enzymatic reaction rates of steroidogenic enzymes
were described in terms of the first-order reaction and rapid-equilibrium enzyme kinetics,
respectively, by the following equations:
∙ CHOL ∙VV
∙ CHOS
19
∙ CHOC
∙ CHOC ∙ CHOM
∙ CHOM ∙ CHOR
∙ CHOM ∙ CHON
∙ CHON
CHON
,
∙PREG
1PREG PROG
,
∙PROG
1PREG PROG
,
∙HPREG
1HPREG HPROG
,
∙HPROG
1HPREG HPROG
,
∙PREG
1PREG HPREG DHEA
,
∙HPREG
1PREG HPREG DHEA
,
∙DHEA
1PREG HPREG DHEA
20
,
∙PROG
1PROG HPROG
,
∙HPROG
1PROG HPROG
,
∙DCORTICO
1DCORTICO DCORT
,
∙DCORT
1DCORTICO DCORT
∙ CORTICO
,
∙DIONE
1DIONE E1
,
∙E1
1DIONE E1
,
∙DIONE
1DIONE TESTO
,
∙TESTO
1DIONE TESTO
where, (1/h) is the rate constant of the first order reaction for metabolic flux X,
(μM/h) is the maximum velocity of enzyme X for substrate A, and is the equilibrium
dissociation constant of enzyme X for substrate A.
21
Protein expression dynamics of steroidogenic proteins regulated by signal transduction
cascades were not incorporated in this model. Because, the rate of signaling promoting
adrenal steroidogenesis is very fast, StAR protein is rapidly induced in response to
stimulation by tropic hormones such as ACTH [66 - 75]. And, the steroidogenic enzymes
such as CYP11A1, CYP17A1 and CYP21A2 are also quickly induced within 24 h by same
stimulus [43]. Therefore, I think that the kinetic parameters and the maximum velocities of
steroidogenic enzymes in this model are reflected as the final activities after steroidogenic
inducing signal.
All static parameters in this model are described in Table 3-1. Rate constants and the
maximum velocity were estimated by fitting to experimental time-course data of the
concentrations of cholesterol and all steroids. Michaelis constants of steroidogenic enzymes
were extracted from the comprehensive enzyme information system BRENDA
(http://www.brenda-enzyme.org). Equilibrium dissociation constants of each steroid transport
were the values of Breen and his colleagues [57, 58]. All initial values of variable parameters
in this model are described in Table 3-2. Initial values of cholesterol and the 14 steroid
concentrations were used in each experimentally measured value, and every steroid
concentration was assumed to rapidly reach the equilibrium state between the culture medium
and intracellular space.
Modeling environment and solution of differential equations
This computational model of adrenal steroidogenesis in NCI-H295R cells was developed on
the simBio platform which is a general environment of biological dynamic simulation and
computational model development [62]. ODEs were solved by the fourth order Runge-Kutta
method with a variable time step. The time step (dt) was adjusted to refer to the maximum
absolute value of flux velocities or enzymatic reaction rates at each time point, and the range
of the time step was from 1×10-5 to 10-2. To confirm whether the range of the time step was
22
suitable, the numerical error ratio was calculated by certain fixed time steps in the range of
the time step, which was under 1×10-8 in every time step. The duration time of computational
simulation of adrenal steroidogenesis in NCI-H295R cells was set at 72 h.
Parameter optimization
To reconstruct experimental time-course patterns of the concentrations of cholesterol and
the 14 steroids in the culture medium and intracellular space, I have optimized every rate
constant and maximum velocity of the steroidogenic enzymes. This parameter optimization
problem was solved by the Levenberg-Marquardt method which is one of the non-linear least
squares methods [63 - 65]. The objective function of optimization was used as the following
normalized least squares distance (NLSD):
NLSD , , , , ,
where is the compartment (culture medium or intracellular space), is the molecular
species (cholesterol and the 14 steroids), is the time point (0, 8, 24, 48, and 72 h), , , is
the experimentally measured concentration of molecule in compartment at time point ,
, , is the simulated concentration of molecule in compartment at time point , and
, is the maximum concentration of molecule in compartment over all time points.
Data points under the lower quantitation limit were excluded from the evaluation by the
objective function.
Effects of every static model parameter for parameter optimization were calculated from
differences of fitting the objective function using sensitivity analysis.
Global dynamic sensitivity analysis
The property of every kinetic parameter in this computational model of steroidogenesis in
NCI-H295R cells was evaluated by dynamic sensitivity analysis. The sensitivity (S , ) of a
dependent variable y with respect to a kinetic parameter x for was defined by the following
23
equation:
S ,Δy y⁄
Δx x⁄
where variable y was the concentration of a steroid hormone in the cytosolic space of
NCI-H295R cells. The perturbation for kinetic parameters was +10% (Δx x 0.1⁄ ). The
results of dynamic sensitivity analysis were visualized as a heat-map view. In visualization,
plotting points of dynamic sensitivity were until 72 h at 6 h intervals.
Statistical Analysis
Comparisons were performed by the two-sample Welch’s t-test with Bonferroni multiple
testing correction for each steroid hormone species. Statistically significant steroid hormones
were considered at adjusted p-values of less than 0.01. Statistical analysis was performed
using MATLAB software (MathWorks, Inc., Natick, MA).
3.2. Results
Steroidogenesis of NCI-H295R cells and the mass balance
All steroid hormones in the culture medium were significantly increased after 72 h of
stimulation with 50 nM ACTH, 20 μM forskolin, and 100 nM angiotensin II (Fig. 3-2A).
Mass balances in steroidogenesis of NCI-H295R cells under non-treatment and control
(stimulated) conditions are shown in Fig. 3-2B and 3-2C, respectively. Under stimulation, the
dynamics of net mass in these experiments were unchanged, and accumulated cholesterol was
converted to adrenal steroids.
Optimization of the mathematical model of adrenal
steroidogenesis in NCI-H295R cells
The mathematical model of adrenal steroidogenesis in NCI-H295R cells was optimized for
24
several kinetic parameters of cholesterol transport, intracellular localization, the oxysterol
pathway, and maximum velocity of steroidogenic enzymes to fit the experimental time-course
data. All optimized kinetic parameters are shown in Table 3-1. The optimized mathematical
model was reconstructed with the experimental dynamic patterns of cholesterol and the 14
steroid hormones in the intracellular space and culture medium. The fitness was 0.621761 of
NLSD values as the fitting objective function. The simulation results and experimental data
are shown in Fig. 3-3.
Sensitivities of optimized kinetic parameters for the fitting score are shown in Table 3-1. In
results of sensitivity analysis, nine kinetic parameters were extracted as high sensitive
parameters for the fitting score, , , , , ,
, , , and , which had over 3.0 of sensitivity for the
fitting score.
Dynamic sensitivity analysis of steroid concentrations secreted by
NCI-H295R cells
To comprehensively understand the dynamics of adrenal steroidogenesis, dynamic
sensitivities were calculated for steroid concentrations secreted by NCI-H295R cells using
my constructed mathematical model of steroidogenesis. The results of dynamic sensitivity
analysis are shown as a heat-map in Fig. 3-4.
For ranking the sensitive parameters affecting the overall of adrenal steroidogenesis pathway,
total area under the curve of dynamic sensitivities for cholesterol and 14 steroids in culture
medium were calculated. The top 10 parameters were , , , ,
, , , , , and in order from the top.
Cholesterol uptake ( ), StAR protein ( ), and CYP11A1 ( ),
determining factors of the capacity for steroidogenesis, promoted the production of
mineralocorticoids (DCORTICO, CORTICO, and ALDO) and restrained the synthesis of
25
glucocorticoids (DCORT and CORT) and sex steroids (DIONE, TESTO, and E1) because of
the accumulation of intermediate molecules in steroidogenesis (PREG, HPREG, DHEA,
PROG and HPROG) only by self-activation. The dynamic patterns of the intermediate
molecules in steroidogenesis are mainly dependent on the activity of CYP17H and HSD3B2
with PREG as the substrate of these enzymes, in which the dynamic sensitivities of
for HPREG and HPROG and for PROG, HPROG and DCORTICO are reversed
the direction at some time from 49 to 66 h. The dynamic sensitivities of the maximum
activities of HSD3B2 for PREG ( ) and CYP21A2 for PROG ( ) are related
to all steroids at 72 h. Almost all kinetic parameters have positive sensitivity for downstream
steroids in the adrenal steroidogenic pathway and negative sensitivity for direct-binding
steroids as substrates of the steroidogenic enzyme. The sensitivities of in all
steroidogenic enzymes are relatively higher than for the same steroid substrate.
Simulation of the metabolic balance of adrenal steroidogenesis
pathway
The results of dynamic sensitivity analysis suggested that enzymatic activities of CYP17H
and HSD3B2 are key regulators for metabolic balance in the adrenal steroidogenesis pathway.
To clearly show the property of the metabolic shift between mineralocorticoid and
glucocorticoid biosynthesis, I performed two-dimensional parameter scanning of the
enzymatic activities of CYP17H and HSD3B2 (Fig. 3-5). NCI-H295R cells lost the ability to
produce all steroid hormones when enzymatic activities of CYP17H and HSD3B2 were
changed by 60% and 30%, respectively. Activation of CYP17H and/or HSD3B2 induced the
metabolic shift that enhanced the glucocorticoid biosynthesis and deviated from the
mineralocorticoid biosynthesis. On the other hand, inhibition of CYP17H and/or HSD3B2
induced the metabolic shift that enhanced the mineralocorticoid biosynthesis and deviated
from the glucocorticoid biosynthesis. Moreover, the enzymatic activity of HSD3B2 regulated
26
the metabolic balance of sex steroids and the precursors on adrenal steroidogenesis of
NCI-H295R cells. NCI-H295R cells became more producing E1, TESTO and DIONE when
activated the enzymatic activity of HSD3B2. Conversely, NCI-H295R cells became more
producing E2 and DHEA when suppressed the enzymatic activity of HSD3B2. The
biosynthesis of downstream steroids in adrenal steroidogenesis pathway, such as
mineralocorticoids and glucocorticoids, was almost completely terminated when the
enzymatic activity of HSD3B2 was decreased by 80% and over.
3.3. Discussion
Importance of 3β-HSD activity in adrenal steroidogenesis
My systematic analysis using the mathematical model of adrenal steroidogenesis in
NCI-H295R cells revealed that the enzymatic activity of 3β-HSD, called as HSD3B2 in my
model, controls the dynamics of adrenal steroidogenesis. The activity of the StAR protein
controls the net capacity of steroidogenesis in steroidogenic cells, which is the transport of
cholesterol from the outer to inner mitochondrial membranes. Both the expression levels of
StAR protein and mRNA are rapidly elevated in response to stimulation by tropic hormones
such as ACTH [66 - 75]. Another important factor in adrenal steroidogenesisis is the
cholesterol side-chain cleavage enzyme CYP11A1, the first, rate-limiting, and hormonally
regulated step in the synthesis of all steroid hormones, which is conversion of cholesterol to
pregnenolone in mitochondria [76 - 83]. According to results of global sensitivity analysis
(Table 3-1 and Fig. 3-4D), in addition to CYP11A1 and StAR proteins, 3β-HSD was one of
the key regulators in adrenal steroidogenesis of NCI-H295R cells. And also, this result
suggests that a significant regulatory mechanism in steroidogenesis pathway is very
reasonable. StAR, CYP11A1, and 3β-HSD (isoform 1 or 2 in humans) proteins generally
27
respond to the same hormones that stimulate steroid production through common pathways
such as cAMP signaling in adrenal glands and testes [84, 85]. Moreover, my data also support
recent experimental evidence from clinical and in vivo studies, suggesting that the enzymatic
activity of 3β-HSD plays an important role in the regulation of mineralocorticoid synthesis in
adrenal steroidogenesis and contributes to hypertension caused by abnormal overproduction
of aldosterone [86 - 89]. Circadian clock-deficient Cry-null mice show salt-sensitive
hypertension due to abnormally high synthesis of aldosterone, which is caused by
constitutively high expression of HSD3B6 mRNA and protein in the adrenal cortex [86, 87].
Recent clinical observations have revealed predominant expression of HSD3B2 mRNA and
protein in tumor cells of aldosterone-producing adenoma (APA), and HSD3B1 mRNA
significantly correlated with CYP11B2 mRNA levels and plasma aldosterone concentrations
in APA patients [88, 89]. However, the relationship is unclear and disputed in a small-scale
clinical study indicating that genetic variation in HSD3B1 affects blood pressure and
hypertension in APA patients [90]. The results of my simulation study suggest that 3β-HSD
protein (human genes are HSD3B1 and HSD3B2) is one of the determination factors for the
dynamic property of adrenal steroidogenesis. My results support the clinical evidence of Doi
and colleagues [88], and I believe that the HSD3B1 enzyme has a promising potential as
novel drug target for endocrine hypertension.
Metabolic properties of adrenal steroidogenesis in NCI-H295R
cells
The metabolic properties of adrenal steroidogenesis in NCI-H295R cells were revealed by
dynamic sensitivity analysis using the mathematical model (Fig 3-4). Mineralocorticoids,
such as DCORTICO, CORTICO and ALDO, and intermediate steroids upstream of the
adrenal steroidogenesis pathway, such as PREG, HPREG, DHEA, PROG, HPROG, were
accelerated by reactions of cholesterol import ( ), StAR protein (
28
and ), and CYP11A1 ( ). On the other hand, glucocorticoids, such as DCORT
and CORT, and sex hormones, such as DIONE, TESTO and E1, were suppressed by these
model parameters. Therefore, enhancement of the net adrenal steroidogenesis capacity, which
supplies PREG precursor to the pathway, causes a production shift from glucocorticoids to
mineralocorticoids by substrate inhibitions of CYP17H, HSD3B2, and CYP21A2 caused by
accumulation of initial intermediate steroids such as PREG and PROG. Sensitivities of
CYP17H ( ) and HSD3B2 ( ) for the products were dynamically changed and
these parameters determined the metabolic balance of downstream steroids in the adrenal
steroidogenesis pathway. These results of dynamic sensitivity analysis of StAR, CYP11A1,
CYP17H, and HSD3B2 suggest that the enhancement of CYP17H and HSD3B2 activity
during ACTH stimulation were important to shift the steroidogenic output away from ALDO
biosynthesis towards CORT biosynthesis, as well as adrenal androgen production. This
suggestion partially supports a comparative animal study in which molecular and cellular
variations in CYP17H activity dramatically affect acute cortisol production, resulting in
distinct physiological and behavioral responses [91].
Results of two-dimensional parameter scanning of the enzymatic activities of CYP17H and
HSD3B2 quantitatively showed the detail of the metabolic relationship between
mineralocorticoid and glucocorticoid biosynthesis (Fig. 3-5). Particularly, the results showed
that the balance of these enzymatic activities was very important for the typical function of
NCI-H295R cells, namely the ability to produce all steroid hormones. NCI-H295R cells lost
this function when enzymatic activities of CYP17H and HSD3B2 were changed by 60% and
30%, respectively. In addition, they became mineralocorticoid (ALDO)-secreting cells when
the enzymatic activity of CYP17H or HSD3B2 was inhibited by 50% or glucocorticoid
(DCORT and CORT)-secreting cells when these enzymes were activated by 50%. In
particular, this analysis also showed that HSD3B2 is a key player in the adrenal
29
steroidogenesis of NCI-H295R cells, because HSD3B2 inhibition by 80% almost completely
inhibited the biosynthesis of downstream steroids. The ratio of CYP17A1 to HSD3B2 mRNA
expression level has been related to several endocrine diseases with a low level in APAs [92]
and high level in cortisol-producing adenomas [93]. Furthermore, the expression levels or
enzymatic activities of CYP17A1 and HSD3B1 have been related to androgen production in
polycystic ovary syndrome [94, 95]. These clinical studies support my simulation results
indicating that the balance of enzymatic activity of CYP17H and HSD3B2 determines the
shift in steroidogenic output to mineralocorticoids, glucocorticoids or androgens.
30
3.4. Tables and Figures
Table 3-1. Fixed parameters in the adrenal steroidgenesis model of NCI-H295R
cells
Parameter
number
Model parameter Optimized
value
Units Sensitivity
for the fitting
objective
function
Reference
1 0.0197596 1/h 4.6234 optimized
2 0.302850 1/h 0.0331 optimized
3 11.7203 1/h 1.4953 optimized
4 202.531 1/h 1.7737 optimized
5 5.17768 1/h 3.1793 optimized
6 0.0404461 1/h 1.8855 optimized
7 9.97234 1/h 8.0399 optimized
8 2339.88 1/h 5.8004 optimized
9 0.0654339 1/h 1.3421 optimized
10 0.5 μM 5.8000 [96]
11 675.701 μM/h 22.4750 optimized
12 0.25 μM 0.2342 [97]
13 0.45 μM 0.2570 [97]
14 108.063 μM/h 3.8388 optimized
15 1050.21 μM/h 0.2985 optimized
16 0.270 μM 0.1274 [98]
17 0.525 μM 0.1597 [99]
18 10.8710 μM/h 0.2107 optimized
19 69.7546 μM/h 0.3717 optimized
20 2.8 μM 0.3626 [100]
21 3.5 μM 0.4001 [100]
22 3.7 μM 0.0929 [100]
23 235.152 μM/h 30.9490 optimized
24 1901.65 μM/h 0.6361 optimized
25 525.917 μM/h 0.1046 optimized
26 1.5 μM 0.1652 [101]
27 1.6 μM 0.3853 [101]
28 213.911 μM/h 7.0023 optimized
31
29 544.496 μM/h 1.9225 optimized
30 2.50 μM 0.1095 [102]
31 0.882 μM 0.1042 [103]
32 214.848 μM/h 0.2427 optimized
33 36.9996 μM/h 0.1781 optimized
34 0.0698923 1/h 0.1634 optimized
35 0.7 μM 0.1022 [104]
36 3.3 μM 0.1111 [104]
37 0.81059 μM/h 0.1666 optimized
38 6.15932 μM/h 0.1769 optimized
39 0.215 μM 0.0035 [105]
40 0.370 μM 0.0003 [105]
41 10.0091 μM/h 0.7052 optimized
42 1.0e-6 μM/h 0.0000 optimized
43 0.0129 dimensionless --- [58]
44 0.0605 dimensionless --- [58]
45 0.0377 dimensionless --- [58]
46 0.0052 dimensionless --- [58]
47 0.0212 dimensionless --- [58]
48 0.0400 dimensionless --- [58]
49 0.0412 dimensionless --- [58]
50 0.0422 dimensionless --- [58]
51 0.0558 dimensionless --- [58]
52 0.0676 dimensionless --- [58]
53 0.0911 dimensionless --- [58]
54 0.0443 dimensionless --- [58]
55 0.0267 dimensionless --- [58]
56 0.0351 dimensionless --- [58]
Table 3-2. Initial values of variable parameters in the adrenal steroidogenesis
model of NCI-H295R cells.
Compartment Molecular parameter
name
Initial value Unit
Culture medium CHOL 81.1 M
Culture medium PREG 0.00085 M
32
Culture medium HPREG 0.06945 M
Culture medium DHEA 0.0 M
Culture medium PROG 0.00003 M
Culture medium HPROG 0.0 M
Culture medium ANDRO 0.00080 M
Culture medium DCORTICO 0.0 M
Culture medium DCORT 0.0 M
Culture medium CORTICO 0.00011 M
Culture medium CORT 0.00003 M
Culture medium ALDO 0.00091 M
Culture medium TESTO 0.00080 M
Culture medium E1 0.00011 M
Culture medium E2 0.00121 M
Intracellular space OXY 0.0 nmol
Intracellular space CHOS 16.08 nmol
Intracellular space CHOC 0.3503 nmol
Intracellular space CHOM 0.02943 nmol
Intracellular space CHON 0.0006431 nmol
Intracellular space CHOR 0.6350 nmol
Intracellular space PREG 0.008580 nmol
Intracellular space HPREG 0.0 nmol
Intracellular space DHEA 0.003165 nmol
Intracellular space PROG 0.00002892 nmol
Intracellular space HPROG 0.00009361 nmol
Intracellular space ANDRO 0.002115 nmol
Intracellular space DCORTICO 0.0007598 nmol
Intracellular space DCORT 0.06871 nmol
Intracellular space CORTICO 0.002065 nmol
Intracellular space CORT 0.003114 nmol
Intracellular space ALDO 0.0 nmol
Intracellular space TESTO 0.0 nmol
Intracellular space E1 0.001895 nmol
Intracellular space E2 0.0003870 nmol
33
Fig. 3-1. Schematic diagram of adrenal steroidgenesis in NCI-H295R cells.
Overview of the mathematical model of adrenal steroidogenesis in NCI-H295R cells,
including cholesterol transport and intracellular localization, oxysterol synthesis, the
C21-steroid hormone biosynthesis pathway, passive diffusional transport of steroid hormones,
and cell proliferation. CHOL: total cholesterol in medium culture, CHOS: stored cholesterol
esters in the endoplasmic reticulum, CHOC: intracellular free cholesterol, CHOM:
mitochondrial free cholesterol, CHON: mitochondrial free cholesterol close to CYP11A1
enzymes, CHOR: mitochondrial free cholesterol remote from CYP11A1 enzymes, PREG:
pregnenolone, HPREG: 17α-hydroxypregnenolone, DHEA: dehydroepiandrosterone, PROG:
progesterone, HPROG: 17α-hydroxyprogesterone, DIONE: androstendione, DCORTICO:
11-deoxycorticosterone, DCORT: 11-deoxycortisol, CORTICO: corticosterone, CORT:
cortisol, ALDO: aldosterone, TESTO: testosterone, E1: estrone, E2: 17β-estradiol, OXY:
oxysterol, CEH: cholesterol ester hydrolase, StAR: steroidogenic acute regulatory protein,
CYP11A1: P450 side chain cleavage enzyme, CYP17H: 17α-hydroxylase of CYP17,
CYP17L: C17-20 lyase of CYP17, HSD3B2: 3β-hydroxysteroid dehydrogenase, CYP21A2:
21-hydroxylase, CYP11B1: 11β-hydroxylase, CYP11B2: 18-hydroxylase, HSD17B3:
17β-hydroxysteroid dehydrogenase, CYP19A1: aromatase.
34
Fig. 3-2. Experimental data of metabolic profiling of adrenal steroid hormones
and the mass balance.
Concentrations of steroid hormones secreted from NCI-H295R cells in the culture medium
at 72 h were compared with untreated conditions and the stimulated condition by 50 nM
ACTH, 20 μM forskolin, and 100 nM angiotensin II (A). Net molecular amounts including
cholesterol and steroid hormones in the culture medium and intracellular space are plotted at
five time points (0, 8, 24, 48, and 72 h after stimulation) under the untreated condition (B)
and the stimulated condition (C). In the bar graphs, dark and light bars indicate the amount of
cholesterol and adrenal steroids, respectively. All data are shown as the mean ± SD (N = 4).
*P-values adjusted by Bonferroni correction < 0.01. PREG: pregnenolone, HPREG:
17α-hydroxypregnenolone, DHEA: dehydroepiandrosterone, PROG: progesterone, HPROG:
17α-hydroxyprogesterone, DIONE: androstendione, DCORTICO: 11-deoxycorticosterone,
DCORT: 11-deoxycortisol, CORTICO: corticosterone, CORT: cortisol, ALDO: aldosterone,
and TESTO: testosterone.
35
Fig. 3-3. Comparison of time-course profiles of cholesterol and adrenal
steroids produced by NCI-H295R cells between experimentally measured and
simulated data.
To intuitively confirm reconstruction of the measured experimental data in the developed
36
simulation model of NCI-H295R cells, dynamics of cholesterol and adrenal steroids produced
by NCI-H295R cells were plotted to overlay experimental data with the simulated results.
Graphs show the dynamics of medium concentrations of cholesterol (A) and adrenal steroids
(B–D), and intracellular concentrations of cholesterol (E) and adrenal steroids (F–H). Steroid
hormones were categorized into three groups by the concentration level. Major steroids were
PREG: pregnenolone, DCORTICO: 11-deoxycorticosterone, DCORT: 11-deoxycortisol,
CORTICO: corticosterone and CORT: cortisol (B and F). Moderate steroids were HPREG:
17α-hydroxypregnenolone, PROG: progesterone, HPROG: 17α-hydroxyprogesterone,
DIONE: androstendione and E1: estrone (C and G). Minor steroids were DHEA:
dehydroepiandrosterone, ALDO: aldosterone, TESTO: testosterone, and E2: 17β-estradiol (D
and H). Experimental data are shown as symbols with dotted lines. All data represent the
mean ± SD (N = 4). Simulation data are shown as solid lines.
37
Fig. 3-4. Heat-map of the dynamic sensitivity analysis of adrenal steroid
concentrations produced by NCI-H295R cells.
The global dynamic sensitivity analysis is a powerful tool to comprehensively understand
the dependencies of the model parameters in the mathematical model of the biological
complex system. Dynamic sensitivities of model parameters in the mathematical model of
adrenal steroidogenesis in NCI-H295R cells were calculated for all steroid concentrations in
the culture medium every 6 h until 72 h after stimulation. To clarify the heat-map of global
38
dynamic sensitivity analysis, imaginary data of the dynamics of steroid concentrations in the
original model (blue line) and perturbed model for sensitivity analysis (red line) were
prepared (A). Using this imaginary data, the calculated dynamic sensitivities (B) and the
visualized dynamic sensitivity as one block of the heat-map (C) were shown respectively.
By the same method that explained using imaginary data, the large-scale data of the global
dynamic sensitivity analysis on the mathematical model of adrenal steroidogenesis in
NCI-H295R cells was comprehensively visualized as a big graph of heat-map (D). Parameter
numbers in the horizontal axis are 1. , 2. , 3. , 4. , 5.
, 6. , 7. , 8. , 9. , 10. , 11. , 12.
, 13. , 14. , 15. , 16. , 17. , 18. ,
19. , 20. , 21. , 22. , 23. , 24. , 25.
, 26. , 27. , 28. , 29. , 30. , 31.
, 32. , 33. , 34. , 35. , 36. , 37.
, 38. , 39. , 40. , 41. , and 42. .
PREG: pregnenolone, HPREG: 17α-hydroxypregnenolone, DHEA: dehydroepiandrosterone,
PROG: progesterone, HPROG: 17α-hydroxyprogesterone, DIONE: androstendione,
DCORTICO: 11-deoxycorticosterone, DCORT: 11-deoxycortisol, CORTICO: corticosterone,
CORT: cortisol, ALDO: aldosterone, and TESTO: testosterone.
39
Fig. 3-5. Metabolic categories of steroidogenic cells determined by the
balance of HSD3B2 and CYP17H activities.
The two-dimensional parameter scanning analysis by the perturbation of two focused
parameters clarifies the interaction and the relationship between the two parameters in the
complex system. Functional cellular categories of steroidogenic cells were defined by the
levels of a mineralocorticoid (ALDO), glucocorticoids (DCORT and CORT), and androgens
(DHEA and DIONE) at 72 h after stimulation. Enzymatic activities of HSD3B2 and CYP17
were normalized by standard values of the simulation model in NCI-H295R cells. Scanning
ranges of HSD3B2 and CYP17 activities were 0–200%, each 10%. Green regions are all 14
steroids produced by NCI-H295R cells. Red, blue and purple regions are
mineralocorticoid-producing cells, glucocorticoids-producing cells, and both
corticoid-producing cells, respectively. Yellow regions are steroidogenesis of NCI-H295R
cells terminated upstream of the adrenal steroidogenesis pathway.
40
4. Estimating mechanism of adrenal action of endocrine-disrupting compounds by using a hybrid optimization method of real-coded genetic algorithm and local search based on non-linear least squares method
4.1. Materials and Methods
Cell culture
NCI-H295R human adrenocortical carcinoma cells were purchased from the American Type
Culture Collection (Cat# CRL-2128, Manassas,VA) and cultured at 37 °C in a humidified
atmosphere with 5% CO2. The cells were maintained in a 1:1 mixture of Dulbecco’s modified
Eagle’s medium (DMEM; GIBCO) and F-12 medium (ICN Biomedicals) supplemented with
15 mM HEPES (Dojindo), 0.00625 mg/mL insulin (SIGMA), 0.00625 mg/mL transferrin
(SIGMA), 30 nM sodium selenite (WAKO), 1.25 mg/mL bovine serum albumin (BSA;
SIGMA), 0.00535 mg/mL linoleic acid (SIGMA), 2.5% Nu Serum (Becton, Dickinson and
Company), 100 U/mL penicillin (Meiji), and 100 mg/L streptomycin (Meiji).
Steroidogenesis in human adrenal corticocarcinoma NCI-H295R
cells
NCI-H295R cells were stimulated with ACTH, forskolin, and angiotensin II to initiate
steroidogenesis. Changes in steroid concentrations over time were measured after stimulation
in both cells and culture medium to construct a simulation model.
The cells were seeded at 6×105 cells/well in 6-well plates. After 3 days of culture, the
culture medium was changed to stimulation medium consisting of DMEM/F-12 (1:1)
medium supplemented with 0.00625 mg/mL insulin, 0.00625 mg/mL transferrin, 30 nM
sodium selenite, 1.25 mg/mL BSA, 0.00535 mg/mL linoleic acid, 10% fetal bovine serum
(GIBCO), 100 U/mL penicillin, 100 mg/L streptomycin, 50 nM ACTH (SIGMA), 20 μM
41
forskolin (SIGMA), and 100 nM angiotensin II (CALBIOCHEM). Culture media and cells
were collected at 72 h after stimulation. The cells were collected in 100 μL distilled water and
sonicated to produce a cell lysate. The cultures were conducted in four wells/time point
(N=4).
The concentrations of 12 steroids, PREG, HPREG, DHEA, PROG, HPROG, DIONE,
TESTO, DCORTICO, DCORT, CORTICO, CORT, ALDO, in the medium and cell lysate
were measured by LC/MS.
Liquid chromatography
A Shimadzu LC-VP series (Kyoto, Japan) consisting of an SIL-HTc autosampler,
LC-10ADvp Pump, CTO-10ACvp column oven, and DGU-14AM degasser was used to set
the reverse-phase liquid chromatographic conditions. The column was a Cadenza CD-C18
column (100 × 2 mm i.d., 3 μm; Imtakt Corp., Kyoto, Japan) used at 45 °C. The mobile phase
included water/acetonitrile/formic acid 95/5/0.05 (v/v/v, Solvent A) and
water/acetonitrile/formic acid 35/65/0.05 (v/v/v, Solvent B). The gradient elution programs
were 0% B (0–1 min with an isocratic gradient), 0–40% B (1–2 min with a linear gradient),
40% B (2–7 min with an isocratic gradient), 40–100% B (7–12 min with a linear gradient),
100% B (12–14 min with an isocratic gradient), 100–0% B (14–15 min with a linear gradient),
and 0% B (15–16 min with an isocratic gradient) at a flow rate of 0.3 mL/min. The
autosampler tray was cooled to 45 °C and the injection volume was 5 μL. HPLC grade
acetonitrile and formic acid were purchased from WAKO.
Mass spectrometry
A triple quadrupole mass spectrometer API4000 (Applied Biosystems/MDS Sciex, Concord,
Canada) coupled with an electrospray ionization source was operated in the positive ion
mode. The optimized ion source conditions were as follows: collision gas, 6 psi; curtain gas,
40 psi; ion source gas 1, 50 psi; ion source gas 2, 80 psi; ion source voltage, 5500 V; ion
42
source temperature, 600 °C. Nitrogen was used as the collision gas in the multiple reaction
monitoring (MRM) mode. The conditions of declustering potential, collision energy, and
collision cell exit potential were optimized by every steroid. The transitions in MRM were as
follows: PREG m/z 317 → 299, HPREG m/z 315 → 297, DHEA m/z 289 → 271, PROG m/z
315 → 109, HPROG m/z 331 → 109, DIONE m/z 287 → 97, DCORT m/z 331 → 123,
DCORTICO m/z 347 → 161, CORTICO m/z 347 → 100, CORT m/z 363 → 309, ALDO m/z
361 → 343, and TESTO m/z 289 → 109. Mass spectroscopic data were acquired and
quantified using the Analyst 1.4.2 software package (Applied Biosystems/MDS Sciex).
Test compounds for mechanistic analysis of adrenal toxicity
NCI-H295R cells were exposed to seven well-characterized inhibitors of steroidogenesis,
and then the concentrations of the steroids in the culture medium were measured to estimate
the enzyme inhibition to evaluate the performance of the simulation model. The adrenal
steroidogenic inhibitors included aminoglutethimide (AGT; BACHEM), o,p’-DDD (DDD;
Aldrich Chem. Co.), spironolactone (SP; SIGMA), metyrapone (MP; Aldrich Chem. Co.),
ketoconazole (KC; WAKO), miconazole (MC; WAKO), and daidzein (DZ; SIMGA). The
cells were also exposed to four adrenal toxicants whose adrenal toxicity is not mediated
through steroidogenesis inhibition. The toxicants were acrylonitrile (AN; WAKO),
salinomycin (SM; SIGMA), thioguanine (TG; Tokyo Kasei), and fumaronitrile (FN; WAKO).
All chemicals were dissolved in DMSO (WAKO) and added to the culture medium at 1:1000
dilutions.
Metabolic steroid profiling of adrenal toxicants (in vitro)
NCI-H295R cells were cultured for 3 days in 6-well plates, and then stimulated with the
above-mentioned compounds. Upon the start of stimulation, various concentrations of test
chemicals were added to the cultures. After a further 3 days of culture with the chemicals, the
concentrations of 12 steroids (PREG, HPREG, DHEA, PROG, HPROG, DIONE,
43
DCORTICO, DCORT, CORTICO, CORT, ALDO, and TESTO) in the culture medium were
measured by LC/MS/MS. The test concentrations of the chemicals were determined by
dose-finding cytotoxicity assays. The cytotoxicity assay was conducted in 96-well plates
using ATP content in cells as an endpoint (CellTiter-GloTM Luminescent Cell Viability
Assay; Promega). Concentrations that caused more than 20% cytotoxicity were not used in
the steroidogenesis assay. The test concentrations of adrenal steroidogenesis inhibitors and
other compounds are shown in Table 4-1.
Quantitative estimation of the mechanism of action of adrenal
toxicants
Metabolic steroid profiling and differential patterns of the adrenal steroid hormones by
chemical perturbation were reconstructed to optimize the relative activities of the
steroidogenic enzymes. The relative activity is defined as scaling factor for of each
steroidogenic enzymes. The input data for the quantitative mechanistic analysis of adrenal
toxic compounds was a fold change (ratio) of the measured 12 steroid concentrations induced
by drug exposure for 72 h. The hybrid optimization method of the real-coded genetic
algorithm (RCGA) and local search was adopted as a global optimization method in the
quantitative mechanistic analysis of adrenal toxic compounds.
The operations of the crossover and generation alteration model in RCGA were used for the
adaptive real-coded ensemble crossover (AREX) and just generation gap (JGG) [106 - 109].
As the initial parameters of RCGA, maximum generation, population size, selection size of
parent individuals, population size of child individuals and termination criteria were 1000,
100, 6, 25 and under 0.1 of NLSD, respectively. The search space for the relative activities of
the steroidogenic enzymes was from 1/100 to 100. To evaluate the fitness of each individual,
the sum of squared residuals for fold changes of measured 12 steroid concentrations was used
as the objective function. The fitness score of each individual was described the following
44
equation:
Fitness Score
where is the molecular species (the 12 steroids), is the experimentally measured
fold change of medium concentration of molecule at 72 hours after drug-exposure, and
is the simulated fold change of medium concentration of molecule at 72 hours after
drug-exposure. Data points under the lower quantitation limit were excluded from the
evaluation by the objective function.
Non-linear least squares optimization by the Levenburg-Marquardt method was used as a
local search [63 - 65]. As the estimated mechanisms of actions of the adrenal toxic
compounds, the relative activities of eight steroidogenic enzymes (CYP11A1, CYP17H,
CYP17L, HSD3B2, CYP21A2, CYP11B1, CYP11B2, and HSD17B3) were optimized by the
above-mentioned hybrid optimization method. Every optimization calculation was duplicated
to check the numerical stability of the optimal parameters.
Statistical Analysis
Differential metabolic steroid profiles were visualized by principal component analysis and
classified by hierarchical cluster analysis. In the hierarchical cluster analysis, pairwise
distances were calculated as standardized Euclidean metric and linkage method was used
Ward’s method. Statistical analysis was performed using MATLAB software (MathWorks,
Inc., Natick, MA).
45
4.2. Results
Cytotoxicity of adrenal toxicants
Viabilities of cells treated with each compound were expressed as a relative value to the ATP
level of the control. Effects of AN, SM TG, FN, AGT, DDD, SP, MP, KC, MC, and DZ on
cell viability were determined to be valid under 80% of the relative ATP level at 7 days after
treatment. AN, SM, TG, and FN showed cytotoxicity at 100, 1, 10 and ≥10 μM, respectively.
AGT, MP, and DZ did not affect cell viability at up to 100 μM. DDD, SP, KC, and MC
induced less than 80% of cell viability at >50, 50, 10 and 50 μM, respectively.
Differentially steroid profiling of adrenal toxicants
After NCI-H295R cells were exposed to each test compound during three days, the
concentrations of 12 steroid hormones in the culture medium were simultaneously measured
by LC/MS/MS. The average concentrations of adrenal steroid hormones in the culture
medium are shown in Fig. 4-1. The effects of test compounds on adrenal steroidogenesis
were evaluated at the concentration without any overt cytotoxicity. The differential metabolic
steroid profiles of 11 adrenal toxic compounds were visualized by using principal component
analysis (Fig. 4-2) and classified by using hierarchical clustering analysis (Fig. 4-3).
Four adrenal toxicants without steroidogenic inhibition, AN, SM, TG, and FN, did not
change the medium concentrations of all steroid hormones by more 2-fold. Abovementioned
4 compounds at every condition and 7 adrenal steroidogenic inhibitors at the low-exposure
concentration were gathered into a big cluster as non-change group. The 7 steroidogenic
inhibitors at the high-exposure concentration were identified the characteristic differential
metabolic steroid pattern each other, but differential steroid profiles of 100 μM DZ and 10μM
SP were classified as a cluster.
AGT drastically decreased the medium concentrations of PREG, HPREG, DHEA, PROG,
DCORTICO, CORTICO, and ALDO at 100 μM. DDD dose-dependently decreased the
46
medium concentrations of PROG, DCORTICO, CORTICO, CORT, and ALDO at >10 μM
and decreased PREG, HPREG, DHEA, PROG, HPROG, DIONE, and DCORT at the
maximum exposure concentration of 25 μM. SP increased PREG, HPREG and DHEA, and
decreased PROG, DIONE, DCORTICO, DCORT, CORTICO, ALDO, and TESTO at 10 μM.
MP dose-dependently decreased CORTICO, CORT, and ALDO and decreased DHEA,
HPROG, DIONE, and TESTO at the maximum exposure concentration of 100 μM. KC
drastically decreased the medium concentrations of PREG, HPREG, DHEA, HPROG,
DIONE, DCORTICO, DCORT, CORTICO, CORTO, ALDO, and TESTO at 10 μM. MC
increased the medium concentrations of PROG and decreased DIONE, DCORT, CORT, and
TESTO at 10 μM. DZ increased PREG, HPREG, and DHEA, and decreased DIONE,
DCORTICO, DCORT, CORTICO, CORT, ALDO, and TESTO at 100 μM.
Quantitative mechanistic analysis of adrenal toxicants
Effects of adrenal toxic compounds on steroidogenic enzymes were quantitatively predicted
from the change in the ratio of the measured medium concentrations of the 12 steroid
hormones at 72 h after drug exposure using the mathematical model of adrenal
steroidogenesis in NCI-H295R cells. The reproducibility of the estimated results was
confirmed by performing the test twice. The estimated effects of 11 adrenal toxic compounds
on eight steroidogenic enzymes are shown in Fig. 4-4. The adrenal toxic compounds without
steroidogenic inhibition, such as vasculotoxic agents (AN, SM, TG and FN), were not
estimated for the target steroidogenic enzymes under non-cytotoxic conditions. Every fitness
score under 0.05 of NLSD values was used as the fitting objective function (Fig. 4-4A ~
4-4D). Other steroidogenic inhibitors (AGT, DDD, SP, MP, KC, MC, and DZ) are described
in detail below.
AGT
The mechanism of action of AGT in adrenal steroidogenesis was estimated by inhibition of
47
CYP11A1, HSD3B2, CYP21A2, and CYP11B1 at 100 μM (estimated inhibitions were 77.0%,
78.0%, 81.1%, and 59.8%, respectively) (Fig. 4-4E). Measured steroid profiling of AGT were
reconstructed 0.020501, 0.015160, 0.101938 and 0.305751 of NLSD values to fit to changing
ratio at 0.1, 1, 10 and 100 μM, respectively. AGT has been reported to inhibit CYP11A1,
CYP21A2, CYP11B1, and CYP11B2 [36, 110 - 113]. My results were mostly consistent with
the previous reports. In particular, CYP11A1 appeared to be inhibited strongly by AGT. In
my study, HSD3B2 inhibition of AGT was shown by mechanistic analysis based on systems
biology approaches as a novel mechanism of action of AGT. Inhibition of AGT by CYP11B2
was not estimated in my study. However, the concentration of ALDO in the culture medium
decreased to 3.8% of the normal stimulated condition. Inhibition of AGT by CYP11B2 has
been shown using sheep adrenal homogenates as well as a human adrenal homogenate from a
patient with Cushing’s syndrome [113]. The activity of 18-hydroxylase induced by CYP11B2
was determined as the conversion of corticosterone to 18-hydroxycorticosterone in the
previous study. The cause of the discrepancy regarding the effect of AGT on CYP11B2 was
suggested to be substrate inhibition, because the intracellular concentration of CORTICO was
increased by over 10 times of that in the culture medium to reach 50 μM. Another possibility
was poor quantitative reliability of the experimental data, because the ALDO concentration
was under the lower limit of quantification at 100 μM AGT. Hecker and his colleagues
reported that 3 μM AGT decreases PREG and PROG concentrations and increases the
TESTO concentration [38]. However, AGT did not increase the TESTO concentration in my
study. One possibility is that the concentration of TESTO was already enhanced by about
3.3-fold through stimulation with ACTH, forskolin, and angiotensin II.
DDD
The mechanism of action of DDD in adrenal steroidogenesis was estimated by
dose-dependent inhibition of CYP11A1, HSD3B2, CYP21A2, and CYP11B1 (estimated
48
inhibitions at 25 μM were 87.0%, 86.9%, 76.9%, and 84.9%, respectively) (Fig. 4-4F).
Measured steroid profiling of DDD were reconstructed 0.010620, 0.021612, 0.072898 and
0.268142 of NLSD values to fit to changing ratio at 0.1, 1, 10 and 25 μM, respectively. DDD
has been reported to inhibit CYP11A1, HSD3B2, CYP21A2, CYP11B1, and CYP11B2 [112,
114, 115]. Inhibition of CYP11B2 by DDD was not estimated in my study. However, the
concentration of ALDO in the culture medium decreased to 3% of that in the normal
stimulated condition. Inhibition of DDD by CYP11B2 has been shown using mitochondrial
and microsomal fractions prepared by standard centrifugation procedures from a bovine
adrenal cortex homogenate [115]. The cause of the discrepancy regarding the inhibition of
DDD by CYP11B2 could not be explained by same effect in the case of AGT.
SP
The mechanism of action of SP in adrenal steroidogenesis was estimated by inhibition of
HSD3B2, CYP21A2, and HSD17B3 (estimated inhibitions at 10 μM were 70.2%, 59.5%, and
59.3%, respectively) (Fig. 4-4G). Measured steroid profiling of SP were reconstructed
0.038331 and 0.079659 of NLSD values to fit to changing ratio at 1 and 10 μM, respectively.
SP has been reported to inhibit CYP17H, CYP17L, CYP11B1, and CYP11B2 [36, 116 - 118].
The inhibitory effect of SP on the HSD3B2 enzyme was a novel mechanism of action. The
main action of SP is as a mineralocorticoid receptor (MR) antagonist. SP has also been
reported to exert some off-target effects by binding to androgen, glucocorticoid and
progesterone receptors [119 - 121]. SP has been shown to inhibit the production of ALDO
and CORT from PREG induced by angiotensin II in H295R cells, but the specific MR
antagonist eplerenone did not show the inhibitory effects [122]. Therefore, HSD3B2
inhibition by SP is not mediated via MR, and the action might be direct inhibition of
HSD3B2 enzymes or a part of known off-target effects mediated through other nuclear
hormone receptors. Regarding CYP17H and CYP17L, my results were consistent with
49
previous reports [116, 117]. 7α-thiospironolactone, which is synthesized by deacetylation of
SP, inhibits CYP17H and CYP17L [117]. The fact that there were no inhibitions of CYP17H
or CYP17L in my study suggests that SP might not be deacetylated to 7α-thiospironolactone
in NCI-H295R cells. Regarding CYP11B1 and CYP11B2, my results were unclear compared
with a previous study. It has been shown that 30 μM SP inhibits CYP11B1 and CYP11B2 in
human and bovine adrenal mitochondria [118]. The cause of CYP11B1 and CYP11B2
inhibition by SP could not be determined in my study, which might be due to the lower
maximum exposure concentration of SP than that in the previous report. I could not examine
of SP concentrations over 10 μM because these concentrations were cytotoxic in NCI-H295R
cells.
MP
The mechanism of action of MP in adrenal steroidogenesis was estimated by
dose-dependent inhibition of CYP11B1 (estimated inhibitions at 1, 10 and 100 μM were
57.1%, 82.7%, and 98.2%, respectively) (Fig. 4-4H). Measured steroid profiling of MP were
reconstructed 0.036027, 0.031511, 0.042620 and 0.133944 of NLSD values to fit to changing
ratio at 0.1, 1, 10 and 100 μM, respectively. MP has been reported to inhibit CYP11B1 as its
major effect, as well as inhibit CYP11A1 and CYP11B2 as a weak effect [36, 112, 123 – 126].
The results were able to show that MP is a selective inhibitor of CYP11B1 in the previous
report. However, the estimated effect of MP at a high concentration, 100 μM as the maximum
exposure concentration, was unclear. According to the previous report, selectivity of MP for
CYP11B1/CYP11B2 is about five times [126]. In addition, 20 μM MP has a slight inhibition
effect on CYP11A1 in H295R cells [112].
KC
The mechanism of action of KC in adrenal steroidogenesis was estimated by inhibition of
CYP11A1, CYP17H, CYP17L, HSD3B2, CYP21A2, CYP11B1, and CYP11B2 (estimated
50
inhibitions at 10 μM were 92.6%, 94.3%, 51.8%, 83.0%, 88.2%, 97.4%, and 79.8%,
respectively) (Fig. 4-4I). Measured steroid profiling of KC were reconstructed 0.018088,
0.017081 and 0.468604 of NLSD values to fit to changing ratio at 0.1, 1 and 10 μM,
respectively. KC has been reported to inhibit CYP11A1, CYP17H, CYP17L, HSD3B2,
CYP21A2, and CYP11B1 [36, 112, 127 - 130]. My results were almost consistent with the
previous reports. KC inhibits CYP11A1, CYP17H, CYP21A2 and CYP11B1 in NCI-H295R
cells at 10 μM [72] and CYP17H, CYP17L, CYP21A2 and CYP11B1 in human adrenal
mitochondria and Leydig cell microsomes at 2–5 μM [131, 132]. However, KC has shown
only weak inhibition of HSD3B2 and HSD17B3 in Leydig cells at the millimolar level [131,
132]. Regarding CYP11B2 and HSD17B3, I considered that these estimated inhibitions of
KC did have sufficient reliability in terms of quantitative prediction precision, because
ALDO and TESTO concentrations were less than the lower limit of quantification at 10 μM
KC.
MC
The mechanism of action of MC in adrenal steroidogenesis was estimated by inhibition of
CYP17H, CYP17L, CYP11B1, and HSD17B3 (estimated inhibitions at 10 μM were 69.1%,
53.0%, 76.4%, and 57.1%, respectively) (Fig. 4-4J). MC has been reported to inhibit not only
CYP17H and CYP17L, but also CYP11A1, CYP21A2, and CYP11B1 [128, 133, 134].
Measured steroid profiling of MC were reconstructed 0.012021, 0.036330 and 0.191708 of
NLSD values to fit to changing ratio at 0.1, 1 and 10 μM, respectively. The results in the
previous reports were able to estimate that MC is a CYP17 inhibitor. However, CYP11A1
inhibition by MC, probably instead of a reduction in StAR expression, was not clearly
detected in my study using NCI-H295R cells, because there were no decreases in the
concentrations of PREG and PROG in the culture medium. Indirect inhibition of CYP11A1
via the peripheral-type benzodiazepine receptor has been reported in mouse adrenocortical
51
Y-1 cells treated with MC in the absence of stimuli by measuring PREG production [133].
On the other hand, reductions of StAR protein expression and/or transport activity without
affecting total steroid synthesis or CYP11A1 and HSD3B2 enzyme expression or activities
have been reported in (BU)2cAMP-stimulated MA-10 Leydig tumor cells treated with MC by
measuring PROG production [134]. Therefore, the effect of MC on the initial reaction in
adrenal steroidogenesis from cholesterol should be different according to the cell type and
stimulation condition. Inhibition of CYP21A2 and CYP11B1 by MC has been reported as
decreases in the consumption of PROG and DCORTICO, respectively [128]. Inhibition of
CYP11B1 was estimated by the action of MC in this study, but that of CYP21A2 was not
detected. In the previous experimental report, inhibitory sites by MC might have been
reflected by inhibition of CYP17H activity, because CYP21A2 activity was measured as a
decrease in labeled PROG.
DZ
The mechanism of action of DZ in adrenal steroidogenesis was estimated by inhibition of
CYP11A1, HSD3B2, CYP21A2, CYP11B1, and HSD17B3 (estimated inhibitions at 100 μM
were 58.6%, 94.1%, 96.5%, 87.2%, and 98.1%, respectively) (Fig. 4-4K). Measured steroid
profiling of DZ were reconstructed 0.043871, 0.029722, 0.061256 and 0.222954 of NLSD
values to fit to changing ratio at 0.1, 1, 10 and 100 μM, respectively. DZ has been reported to
inhibit HSD3B2 and CYP21A2 [135]. The results of HSD3B2 and CYP21A2 were consistent
with the previous report. However, inhibitions have not been reported for CYP11A1,
CYP11B1, and HSD17B3. These estimated effects of DZ on CYP11B1 and HSD17B3 were
unclear, because the concentrations of ALDO and TESTO were less than the lower limit of
quantification at 100 μM. In addition, these enzymes act downstream of the strong action
points of DZ, such as HSD3B2 and CYP21A2.
52
4.3. Discussion
Methodologies of quantitative mechanistic analysis for drug
discovery
According to my results obtained using the mathematical model of steroidogenesis in
NCI-H295R cells, such as sensitivity analysis, comprehensive analysis based on systems
biology is available to quantitatively estimate the mechanism of action of steroidogenic
disrupting compounds from differential profiling of adrenal steroid hormones. Because,
dynamic patterns of steroid hormones in adrenal steroidogenesis pathway are highly complex.
My proposed method of quantitative mechanistic analysis of steroidogenic inhibitors was
able to predict known action sites in the adrenal steroidogenesis pathway at only one time
point (72 h after drug exposure). Although it is possible to estimate the mechanism of action
even at one point, if you need more high quality results, I recommend that time points of
experimental data are added for improving accuracy. Moreover, according to the results of
sensitivity analysis (Fig. 3-4), the sensitivity of for all steroidogenic enzymes was
more sensitive than that of , because the intracellular concentrations of steroid hormones
were almost maintained at sufficiently high levels compared with the values of
steroidogenic enzymes. These results suggested that estimation of the mechanism of action of
drugs is more effective and detectable when using the influences of as the searching
parameters such as my proposed method. My data showed that the proposed method based on
a systems biology model is a very powerful tool for exploratory screening of steroidogenic
disrupting compounds.
RCGA as a solver of parameter estimation problems in systems biology has been applied to
biological network identification of gene regulatory networks and metabolic pathways, and
optimization of biological processes using experimentally observed time-course data [136 -
141]. In this study, RCGA was useful to estimate the mechanism of action of novel
53
pharmaceutical drug candidates for adrenal steroidogenesis as a new application of RCGA in
systems biology. I had two issues when applying RCGA to the quantitative mechanistic
analysis of drug actions. These issues were the vast calculation cost and multimodality of
quasi-optimum solutions in solving the optimization problem, because the mathematical
model in systems biology consists of many equations and parameters. A proposed
optimization strategy using RCGA based on AREX/JGG was a highly stable and efficient
calculation method for a better quasi-optimum solution than the unimodal normal distribution
crossover (UNDX)/minimum generation gap (MGG) method that is well applied in the
engineering field. In addition, I expanded the RCGA optimization program based on
AREX/JGG to a hybrid method of GA and then applied as a local search as recommended by
Harada and Kobayashi [108, 142]. A final optimal solution was obtained with a good
convergence property. Because these problems are general in systems biology studies, I
suggest that the proposed hybrid method based on AREX/JGG is a very useful tool for
quantitative mechanistic analysis of novel pharmaceutical drugs, not limited to steroidogenic
disrupting compounds.
Limitation
Proposed computational approach based on a comprehensive mathematical model of adrenal
steroidogenesis and a hybrid optimization method of RCGA and local search was able to
estimate mechanism-of-action of endocrine-disrupting compounds by using differential
metabolic profiling of 12 steroid hormones at 3 days after exposure. However, this proposed
approach depends on the accuracy and the structure of the mathematical model of adrenal
steroidogenesis. For example, although a computational analysis based on the present
mathematical model is possible to estimate the drug effects for steroidogenic enzymes as end
phenotypes, it is not able to correctly predict the true mechanism-of-action of chemicals
affecting to other proteins, such as adrenocorticotrophin receptor MC2R, renin-angiotensin-
54
aldosterone system and some ion channels. Because, the present mathematical model of
adrenal steroidogenesis does not have any components for signal transduction cascades of
ACTH and Ang II and electrophysiological effects of potassium and calcium ions. Therefore,
to expand estimating the mechanism-of-action for upstream pathways of adrenal
steroidogenesis, detail functions of signal transduction cascades, protein expression dynamics
and electrophysiological ion dynamics in NCI-H295R cells will need to incorporate into the
mathematical model of adrenal steroidogenesis. On the other hands, it is very hard problem
that complex effects of the chemical action for biological system are correctly predicted by
only one profiling data, such as metabolomics. Therefore, I would like to investigate about
not only incorporating additional functions for signal transduction cascades, protein
expression dynamics of steroidogenic enzymes and electrophysiology in NCI-H295R cells
into the present mathematical model, but also clearly understanding of mechanism of drug
action by the integration of multiple profiling data, such as transcriptomics, proteomics,
phosphoproteomics, metabolomics, metabolic flux analysis and others.
55
4.4. Tables and Figures
Table 4-1. Adrenal toxicities and actions of tested compounds.
Test chemical
Test
concentrations
(μM)
Pathological
features of
adrenal toxicity
Reported enzyme
inhibitions References
Acrylonitrile
(AN)
0.1, 1, 10, 100 Hemorrhagic
adrenal necrosis
Not reported [143]
Salinomycin
(SM)
0.00001,
0.001, 0.01, 1,
10
Damage to
adrenal medullae
Not reported [144]
Thioguanine
(TG)
0.01, 1, 10,
and 100
Hemorrhagic
adrenal necrosis
Not reported [143]
Fumaronitrile
(FN)
0.1, 1, 10, 100 Hemorrhagic
adrenal necrosis
Not reported [143]
Aminoglutethimide
(AGT)
0.1, 1, 10, 100 Hypertrophy,
Vascular
degeneration
CYP11A1, CYP21A2,
CYP11B1, CYP11B2
[36, 110 -
113]
o,p’-DDD
(DDD)
0.1, 1, 10, 25,
100
Atrophy CYP11A1, HSD3B2,
CYP21A2, CYP11B1,
CYP11B2
[112, 114,
115]
Spironolactone
(SP)
1, 10, 50, 100 Hypertrophy CYP17H, CYP17L,
CYP11B1, CYP11B2
[36, 116 -
118]
Metyrapone
(MP)
0.1, 1, 10, 100 Hypertrophy,
Vascular
degeneration
CYP11A1, CYP11B1,
CYP11B2
[36, 112,
123 - 126]
Ketoconazole
(KC)
0.1, 1, 10, 100 Hypertrophy CYP11A1, CYP17H,
CYP17L, HSD3B2,
CYP21A2, CYP11B1
[36, 112,
127 - 130]
Miconazole
(MC)
0.1, 1, 10, 25,
50, 100
Hypertrophy CYP11A1, CYP17H,
CYP17L, CYP21A2,
CYP11B1
[128, 133,
134]
Daidzein
(DZ)
0.1, 1, 10, 100 Unknown HSD3B2, CYP21A2 [135]
56
Fig. 4-1. Differential metabolic profiling of 12 steroid hormones by exposure to
adrenal toxicants.
Concentrations of 12 adrenal steroids secreted from NCI-H295R cells were changed by
57
exposure to adrenal toxicity compounds. The 12 adrenal steroid hormones were quantitatively
measured by LC-MS/MS simultaneously. Differential metabolic profiling of 12 steroids by
exposure to vasculotoxic agents acrylonitrile: AN (A), fumaronitrile: FN (B), salinomycin:
SM (C), and thioguanine: TG (D) are shown. These agents were used as negative control
compounds for steroidogenic inhibitors. In addition, differential steroid profiles induced by
steroidogenic inhibitors aminoglutethimide: AGT (E), o,p’-DDD: DDD (F), spironolactone:
SP (G), metyrapone: MP (H), ketoconazole: KC (I), miconazole: MN (J), and daidzein: DN
(K) are shown. The exposure concentration of each adrenal toxic compound was a
non-cytotoxic concentration. PREG: pregnenolone, HPREG: 17α-hydroxypregnenolone,
DHEA: dehydroepiandrosterone, PROG: progesterone, HPROG: 17α-hydroxyprogesterone,
DIONE: androstendione, DCORTICO: 11-deoxycorticosterone, DCORT: 11-deoxycortisol,
CORTICO: corticosterone, CORT: cortisol, ALDO: aldosterone, and TESTO: testosterone.
All data represent the mean (N = 4).
58
Fig. 4-2. Principal component analysis of differential metabolic profiles of 12
steroid hormones by exposure to adrenal toxicants.
The differential metabolic steroid profiling of adrenal toxicants were visualized by using the
principal component analysis. Concentrations of 12 adrenal steroids secreted from
NCI-H295R cells were changed by exposure to adrenal toxicity compounds. The 12 adrenal
steroid hormones were quantitatively measured by LC-MS/MS simultaneously. Differential
metabolic profiling of 12 steroids by exposure to vasculotoxic agents acrylonitrile: AN,
fumaronitrile: FN, salinomycin: SM, and thioguanine: TG are shown. These agents were used
as negative control compounds for steroidogenic inhibitors. In addition, differential steroid
profiles induced by steroidogenic inhibitors aminoglutethimide: AGT, o,p’-DDD: DDD,
spironolactone: SP, metyrapone: MP, ketoconazole: KC, miconazole: MC, and daidzein: DN
are shown. Exposure concentrations of adrenal toxic compound were described in brackets of
the sample name and the units were prepared as μM. A blue circle was classified as a group of
non-change samples. PREG: pregnenolone, HPREG: 17α-hydroxypregnenolone, DHEA:
dehydroepiandrosterone, PROG: progesterone, HPROG: 17α-hydroxyprogesterone, DIONE:
androstendione, DCORTICO: 11-deoxycorticosterone, DCORT: 11-deoxycortisol,
CORTICO: corticosterone, CORT: cortisol, ALDO: aldosterone, and TESTO: testosterone.
59
Fig. 4-3. Hierarchical cluster analysis of differential metabolic profiles of 12
steroid hormones by exposure to adrenal toxicants.
Adrenal toxicants were classified by using the differential metabolic profiling of 12 steroid
hormones. Concentrations of 12 adrenal steroids secreted from NCI-H295R cells were
changed by exposure to adrenal toxicity compounds. The 12 adrenal steroid hormones were
quantitatively measured by LC-MS/MS simultaneously. Differential metabolic profiling of 12
steroids by exposure to vasculotoxic agents acrylonitrile: AN, fumaronitrile: FN,
salinomycin: SM, and thioguanine: TG are shown. These agents were used as negative
control compounds for steroidogenic inhibitors. In addition, differential steroid profiles
induced by steroidogenic inhibitors aminoglutethimide: AGT, o,p’-DDD: DDD,
spironolactone: SP, metyrapone: MP, ketoconazole: KC, miconazole: MC, and daidzein: DN
are shown. Exposure concentrations of adrenal toxic compound were described in brackets of
the sample name and the units were prepared as μM. A blue cluster was classified as a group
of non-change samples. PREG: pregnenolone, HPREG: 17α-hydroxypregnenolone, DHEA:
dehydroepiandrosterone, PROG: progesterone, HPROG: 17α-hydroxyprogesterone, DIONE:
androstendione, DCORTICO: 11-deoxycorticosterone, DCORT: 11-deoxycortisol,
CORTICO: corticosterone, CORT: cortisol, ALDO: aldosterone, and TESTO: testosterone.
60
Fig. 4-4. Estimated mechanism of action of adrenal toxicants by using the
mathematical model of adrenal steroidogenesis in NCI-H295R cells.
Mechanisms of action of adrenal toxicity compounds were quantitatively estimated from
experimental results of differential steroid profiling using the mathematical model of adrenal
steroidogenesis in NCI-H295R cells. The drug action was defined as a scaling factor of
enzymatic activity in the simulation model. These scaling factors were optimized to fit
experimental data by a hybrid method of the RCGA and non-linear least squares. Estimated
drug actions by exposure of vasculotoxic agents acrylonitrile: AN (A), fumaronitrile: FN (B),
salinomycin: SM (C), and thioguanine: TG (D) and the steroidogenic inhibitors
aminoglutethimide: AGT (E), o,p’-DDD: DDD (F), spironolactone: SP (G), metyrapone: MP
(H), ketoconazole: KC (I), miconazole: MC (J), and daidzein: DN (K) are shown in a spider
radar chart. CYP11A1: P450 side chain cleavage enzyme, CYP17H: 17α-hydroxylase of
CYP17, CYP17L: C17-20 lyase of CYP17, HSD3B2: 3β-hydroxysteroid dehydrogenase,
CYP21A2: 21-hydroxylase, CYP11B1: 11β-hydroxylase, CYP11B2: 18-hydroxylase, and
HSD17B3: 17β-hydroxysteroid dehydrogenase.
61
5. Conclusions
The novel mathematical model of adrenal steroidogenesis constructed in this study,
including cholesterol transport and distribution, the C21-steroid hormone pathway, steroid
transport, and cell proliferation, could reproduce adrenal steroidogenesis in NCI-H295R cells.
According to the results of dynamic sensitivity analysis using the new model, HSD3B2 plays
the most important role in the metabolic balance of adrenal steroidogenesis in NCI-H295R
cells. Moreover, to quantitatively estimate mechanisms of action of adrenal toxic compounds,
I developed a computational system by using this new mathematical model and a hybrid
optimization method of RCGA and local search (non-linear least squares method). By
differential metabolic profiling of 12 steroid hormones at 3 days after exposure to 11 adrenal
toxic compounds and a developed computational system, it was able to estimate which sites
of steroidogenic enzymes were affected by these compounds. Vasculotoxic agents were
estimated to have no effect according to the results obtained by my computational system. In
terms of adrenal steroidogenic inhibitors, the predicted action sites were approximately
matched to the target enzymes as reported in the literature. Thus, my computational system
based on a systems biology approach may be useful to analyze the mechanism of action of
endocrine-disrupting compounds and to assess the human adrenal toxicity of novel
pharmaceutical drugs based on steroid hormone profiling.
62
6. Acknowledgements
I would like to thank Dr. Natsuko Terasaki and Mr. Makoto Yamazaki at Mitsubishi Tanabe
Pharma Corporation for performing all in vitro experiments and collected data in this study.
I would like to thank Dr. Michael S Breen and Dr. Miyuki Breen at EPA for helpful
discussion concerning their mathematical models of adrenal steroidogenesis.
This work was partially supported by the Grant-in-Aid for Scientific Research on Innovative
Areas, “Synthetic Biology” (No.23119001 (MO)) from Ministry of Education, Culture,
Sports, Science and Technology in Japan.
63
7. References
1. Chuang HY, Hofree M, Ideker T. A decade of systems biology. Annu Rev Cell
Development Biol. 2010; 26:721–44.
2. Zou J, Zheng MW, Li G, Su ZG. Advanced Systems Biology Methods in Drug
Discovery and Translational Biomedicine. BioMed Res Int. 2013; 742835.
3. Zhao S, Iyengar R. Systems pharmacology: network analysis to identify multiscale
mechanisms of drug action. Annu Rev Pharmacol Toxicol. 2012; 52:505–21.
4. Yildirim MA, Goh KI, Cusick ME, Barabasi AL, Vidal M. Drug-target network. Nat
Biotechnol. 2007; 25(10): 1119–26.
5. Paolini GV, Shapland RHB, Van Hoorn WP, Mason JS, Hopkins AL. Global mapping
of pharmacological space. Nat Biotechnol. 2006; 24(7): 805–15.
6. Hopkins AL. Network pharmacology: the next paradigm in drug discovery. Nat Chem
Biol. 2008; 4(11): 682–90.
7. Keiser MJ, Roth BL, Armbruster BN, Ernsberger P, Irwin JJ, Shoichet BK. Relating
protein pharmacology by ligand chemistry. Nat Biotechnol. 2007; 25(2): 197–206.
8. Li J, Zhu X, Chen JY. Building disease-specific drug-protein connectivity maps from
molecular interaction networks and PubMed abstracts. PLoS Comput Biol. 2009;
5(7): Article ID e1000450.
9. Keiser MJ, Setola V, Irwin JJ, Laggner C, Abbas AI, Hufeisen SJ, et al. Predicting
new molecular targets for known drugs. Nature. 2009; 462(7270): 175–81.
10. Campillos M, Kuhn M, Gavin AC, Jensen LJ, Bork P. Drug target identification using
side-effect similarity. Science. 2008; 321(5886): 263–6.
11. Takarabe M, Kotera M, Nishimura Y, Goto S, Yamanishi Y. Drug target prediction
using adverse event report systems: a pharmacogenomic approach. Bioinformatics.
2012; 28(18): i611–8.
12. Cheng F, Liu C, Jiang J, Lu W, Li W, Liu G, et al. Prediction of drug-target
interactions and drug repositioning via network-based inference. PLoS Comput Biol.
2012; 8(5): e1002503.
13. Lounkine E, Keiser MJ, Whitebread S, Mikhailov D, Hamon J, Jenkins JL, et al.
Large-scale prediction and testing of drug activity on side-effect targets. Nature. 2012;
486(7403): 361–7.
64
14. Kuhn M, Al Banchaabouchi M, Campillos M, Jensen LJ, Gross C, Gavin AC, Bork P.
Systematic identification of proteins that elicit drug side effects. Mol Syst Biol. 2013;
9: 663.
15. Yang L, Chen J, He L. Harvesting candidate genes responsible for serious adverse
drug reactions from a chemical-protein interactome. PLoS Comput Biol. 2009; 5(7):
e1000441.
16. Yang L, Wang K, Chen J, Jegga AG, Luo H, Shi L, et al. Exploring off-targets and
off-systems for adverse drug reactions via chemical-protein
interactome—clozapine-induced agranulocytosis as a case study. PLoS Comput Biol.
2011; 7(3): e1002016.
17. Chang RL, Xie L, Xie L, Bourne PE, Palsson BO. Drug off-target effects predicted
using structural analysis in the context of a metabolic network model. PLoS Comput
Biol. 2010; 6(9): e1000938.
18. Chen L, Lu J, Zhang J, Feng KR, Zheng MY, Cai YD. Predicting chemical toxicity
effects based on chemical-chemical interactions. PLoS ONE. 2013; 8(2): e56517.
19. Iorio F, Bosotti R, Scacheri E, Belcastro V, Mithbaokar P, Ferriero R, et al. Discovery
of drug mode of action and drug repositioning from transcriptional responses. Proc
Natl Acad Sci USA. 2010; 107(33): 14621–6.
20. Gottlieb A, Stein GY, Ruppin E, Sharan R. PREDICT: a method for inferring novel
drug indications with application to personalized medicine. Mol Syst Biol. 2011; 7:
496.
21. Sirota M, Dudley JT, Kim J, Chiang AP, Morgan AA, Sweet-Cordero A, et al.
Discovery and preclinical validation of drug indications using compendia of public
gene expression data. Sci Transl Med. 2011; 3(96): 96ra77.
22. Jin G, Fu C, Zhao H, Cui K, Chang J, Wong STC. A novel method of transcriptional
response analysis to facilitate drug repositioning for cancer therapy. Cancer Res.
2012; 72(1): 33–44.
23. Iskar M, Zeller G, Blattmann P, Campillos M, Kuhn M, Kaminska KH, et al.
Characterization of drug-induced transcriptional modules: towards drug repositioning
and functional understanding. Mol Syst Biol. 2013; 9: 662.
24. Jia J, Zhu F, Ma X, Cao ZW, Li YX, Chen YZ. Mechanisms of drug combinations:
interaction and network perspectives. Nat Rev Drug Discov. 2009; 8(2): 111–28.
25. Fitzgerald JB, Schoeberl B, Nielsen UB, Sorger PK. Systems biology and
65
combination therapy in the quest for clinical efficacy. Nat Chem Biol. 2006; 2(9):
458–66.
26. Iadevaia S, Lu Y, Morales FC, Mills GB, Ram PT. Identification of optimal drug
combinations targeting cellular networks: integrating phospho-proteomics and
computational network analysis. Cancer Res. 2010; 70(17): 6704–14.
27. Zou J, Luo SD, Wei YQ, Yang SY. Integrated computational model of cell cycle and
checkpoint reveals different essential roles of Aurora-A and Plk1 in mitotic entry. Mol
BioSystems. 2011; 7(1): 169–79.
28. Zou J, Ji P, Zhao YL, Li LL, Wei YQ, Chen YZ, Yang SY. Neighbor communities in
drug combination networks characterize synergistic effect. Mol BioSystems. 2012;
8(12): 3185–96.
29. Lehar J, Krueger AS, Avery W, Heilbut AM, Johansen LM, Price ER, et al. Synergistic
drug combinations tend to improve therapeutically relevant selectivity. Nat Biotechnol.
2009; 7: 659–66.
30. Kwong LN, Costello JC, Liu H, Jiang S, Helms TL, Langsdorf AE, et al. Oncogenic
NRAS signaling differentially regulates survival and proliferation in melanoma. Nat
Med. 2012; 18(10): 1503–10.
31. Harvey PW, Everett DJ, Springall CJ. Adrenal toxicology: a strategy for assessment of
functional toxicity to the adrenal cortex and steroidogenesis. J Appl Toxicol. 2007;
27(2):103-15.
32. Ribelin WE. The effects of drugs and chemicals upon the structure of the adrenal
gland. Fundam Appl Tox. 1984; 4(1):105-19.
33. Colby HD, Longhurst PA. Toxicology of the adrenal gland. In: Atterwill CK, Flack JD,
eds. Endocrine Toxicology. Cambridge University Press, Cambridge; 1992; 243-81.
34. Colby HD. In vitro assessment of adrenocortical toxicity. J Pharmacol Toxicol
Methods. 1994; 32(1):1-6.
35. Rosol TJ, Yarrington JT, Latendresse J, Capen CC. Adrenal gland: structure, function,
and mechanisms of toxicity. Txicol Pathol. 2001; 29(1):41-8.
36. Harvey PW, Everett DJ. The adrenal cortex and steroidogenesis as cellular and
molecular targets for toxicity: critical omissions from regulatory endocrine disrupter
screening strategies for human health? J Appl Toxicol. 2003; 23(2): 81-7.
37. Harvey PW, Everett DJ. Regulation of endocrine-disrupting chemicals: critical
66
overview and deficiencies in toxicology and risk assessment for human health. Best
Pract Res Clin Endocrinol Metab. 2006; 20(1):145-65.
38. Hecker M, Newsted JL., Murphy MB, et al. Human adrenocarcinoma (H295R) cells
for rapid in vitro determination of effects on steroidogenesis: hormone production.
Toxicol Appl Pharmacol. 2006; 217(1):114-24.
39. Hinson JP, Raven PW. Effects of endocrine-disrupting chemicals on adrenal function.
Best Pract Res Clin Endocrinol Metab. 2006; 20(1):111-20.
40. Sanderson JT. The steroid hormone biosynthesis pathway as a target for
endocrine-disrupting chemicals. Toxicol Sci. 2006; 94(1):3-21.
41. Gazdar AF, Oie HK, Shackleton CH, et al. Establishment and characterization of a
human adrenocortical carcinoma cell line that expresses multiple pathways of steroid
biosynthesis. Cancer Res. 1990; 50:5488–96.
42. Rainey WE, Bird IM, Sawetawan C, et al. Regulation of human adrenal carcinoma
cell (NCI-H295) production of C19 steroids. J Clin Endocrinol Metab. 1993;
77(3):731-7.
43. Staels B, Hum DW, Miller WL. Regulation of steroidogenesis in NCI-H295 cells: a
cellular model of the human fetal adrenal. Mol Endocrinol. 1993; 7(3):423-33.
44. Oskarsson A, Ulleras E, Plant KE, Hinson JP, Goldfarb PS. Steroidogenic gene
expression in H295R cells and the human adrenal gland: adrenotoxic effects of
lindane in vitro. J Appl Toxicol. 2006; 26(6):484-92.
45. Muller-Vieira U, Angotti M, Hartman RW. The adrenocortical tumor cell line
NCI-H295R as an in vitro screening system for the evaluation of CYP11B2
(aldosterone synthase) and CYP11B1 (steroid-11beta-hydroxylase) inhibitors. J
Steroid Biochem Mol Biol. 2005; 96(3-4):259-70.
46. Imagawa K, Okayama S, Takaoka M, et al. Inhibitory effect of efonidipine on
aldosterone synthesis and secretion in human adrenocarcinoma (H295R) cells. J
Cardiovasc Pharmacol. 2006; 47(1):133-8.
47. Voets M, Muller-Vieira U, Marchais-Oberwinkler S, Hartmann RW. Synthesis of
amidinohydrazones and evaluation of their inhibitory effect towards aldosterone
synthase (CYP11B2) and the formation of selected steroids. Arch Pharm. 2004;
337(7):411-6.
48. Nielsen FK, Hansen CH, Fey JA, et al. H295R cells as a model for steroidogenic
disruption: a broader perspective using simultaneous chemical analysis of 7 key
67
steroid hormones. 2012; 26:343-50.
49. Ohno S, Shinoda S, Toyoshima S, Nakazawa H, Makino T, Nakajin S. Effects of
flavonoid phytochemicals on cortisol production and on activities of steroidogenic
enzymes in human adrenocortical H295R cells. J Steroid Biochem Mol Biol. 2002;
80(3):355-63.
50. Canton RF, Sanderson JT, Nijmeijer S, Bergman A, Letcher RJ, van den Berg M. In
vitro effects of brominated flame retardants and metabolites on CYP17 catalytic
activity: a novel mechanism of action? Toxicol Appl Pharmacol. 2006; 216(2):274-81.
51. Hilscherova K, Jones PD, Gracia T, et al. Assessment of the effects of chemicals on
the expression of ten steroidogenic genes in the H295R cell line using real-time PCR.
Toxicol Sci. 2004; 81(1):78-89.
52. Zhang X, Yu RM, Jones PD, et al. Quantitative RT-PCR methods for evaluating
toxicant-induced effects on steroidogenesis using the H295R cell line. Environ Sci
Technol. 2005; 39(8):2777-85.
53. Gracia T, Hilscherova K, Jones PD, et al. The H295R system for evaluation of
endocrine-disrupting effects. Ecotoxicol Environ Saf. 2006; 65(3):293-305.
54. Blaha L, Hilscherova K, Mazurova E, et al. Alteration of steroidogenesis in H295R
cells by organic sediment contaminants and relationships to other endocrine disrupting
effects. Environ Int. 2006; 32(6):749-57.
55. Xu Y, Yu RM, Zhang X, et al. Effects of PCBs and MeSO2-PCBs on adrenocortical
steroidogenesis in H295R human adrenocortical carcinoma cells. Chemosphere. 2006;
63(5):772-84.
56. Zhang X, Newsted JL, Hecker M, Higley EB, Jones PD, Giesy JP. Classification of
chemicals based on concentration-dependent toxicological data using ToxClust.
Environ Sci Technol. 2009; 43(10):3926-32.
57. Breen MS, Breen M, Terasaki N, Yamazaki M, Conolly RB. Computational model of
steroidogenesis in human H295R cells to predict biochemical response to
endocrine-active chemicals: model development for metyrapone. Environ Health
Perspect. 2010; 118(2):265-72.
58. Breen M, Breen MS, Terasaki N, Yamazaki M, Lloyd AL, Conolly RB. Mechanistic
computational model of steroidogenesis in H295R cells: role of oxysterols and cell
proliferation to improve predictability of biochemical response to endocrine active
chemical-metyrapone. Toxicol Sci. 2011; 123(1):80-93.
68
59. Kraemer FB. Adrenal cholesterol utilization. Mol Cell Endocrinol. 2007;
265-266:42-5.
60. Miller WL. Steroidogenic acute regulatory protein (StAR), a novel mitochondrial
cholesterol transporter. Biochim Biophys Acta. 2007; 1771(6):663-76.
61. Dempsher DP, Gann DS, Phair RD. A mechanistic model of ACTH-stimulated cortisol
secretion. Am J Physiol. 1984; 246(4 Pt 2):R587-96.
62. Sarai N, Matsuoka S, Noma A. simBio: a Java package for the development of
detailed cell models. Prog Biophys Mol Biol. 2006; 90(1-3):360-77.
63. Levenberg K. A method for the solution of certain non-linear problems in least
squares. Q Appl Math. 1944; 2:164-8.
64. Marquardt D. An algorithm for least-squares estimation of nonlinear parameters.
SIAM J Appl Math. 1963; 11(2):431-41.
65. Gill PE, Murray W. Algorithms for the solution of the nonlinear least-squares problem.
(1978) SIAM J Numer Anal. 1978; 15(5):977-92.
66. Pon LA, Orme-Johnson NR. Acute stimulation of steroidogenesis in corpus luteum
and adrenal cortex by peptide hormones. Rapid induction of a similar protein in both
tissues. J Biol Chem. 1986; 261(14):6594-9.
67. Pon LA, Hartigan JA, Orme-Johnson NR. Acute ACTH regulation of adrenal
corticosteroid biosynthesis. Rapid accumulation of a phosphoprotein. J Biol Chem.
1986; 261(28):13309-16.
68. Epstein LF, Orme-Johnson NR. Regulation of steroid hormone biosynthesis.
Identification of precursors of a phosphoprotein targeted to the mitochondrion in
stimulated rat adrenal cortex cells. J Biol Chem. 1991; 266(29):19739-45.
69. Stocco DM, Sodeman TC. The 30-kDa mitochondrial proteins induced by hormone
stimulation in MA-10 mouse Leydig tumor cells are processed from larger precursors.
J Biol Chem. 1991; 266(29):19731-8.
70. Clark BJ, Wells J, King SR, Stocco DM. The purification, cloning, and expression of
a novel luteinizing hormone-induced mitochondrial protein in MA-10 mouse Leydig
tumor cells. Characterization of the steroidogenic acute regulatory protein (StAR). J
Biol Chem. 1994; 269(45):28314-22.
71. Stocco DM, Clark BJ. Regulation of the acute production of steroids in steroidogenic
cells. Endocr Rev. 1996; 17(3):221-44.
69
72. Kallen CB, Arakane F, Christenson LK, Watari H, Devoto L, Strauss JF 3rd.
Unveiling the mechanism of action and regulation of the steroidogenic acute
regulatory protein. Mol Cell Endocrinol. 1998; 145(1-2):39-45.
73. Christenson LK, Strauss JF 3rd. Steroidogenic acute regulatory protein (StAR) and
the intramitochondrial translocation of cholesterol. Biochim Biophys Acta. 2000;
1529(1-3):175-87.
74. Christenson LK, Strauss JF 3rd. Steroidogenic acute regulatory protein: an update on
its regulation and mechanism of action. Arch Med Res. 2001; 32(6):576-86.
75. Hagen IJ, Kusakabe M, Young G. Effects of ACTH and cAMP on steroidogenic acute
regulatory protein and P450 11beta-hydroxylase messenger RNAs in rainbow trout
interrenal cells: relationship with in vitro cortisol production. Gen Comp Endocrinol.
2006; 145(3):254-62.
76. Stone D, Hechter O. Studies on ACTH action in perfused bovine adrenals: aspects of
progesterone as an intermediary in corticosteroidogenesis. Arch Biochem Biophys.
1955; 54(1):121-8.
77. Halkerston ID, Eichhorn J, Hechter O. A requirement for reduced triphosphopyridine
nucleotide for cholesterol side-chain cleavage by mitochondrial fractions of bovine
adrenal cortex. J Biol Chem. 1961; 236:374-80.
78. Koritz SB, Kumar AM. On the mechanism of action of the adrenocorticotrophic
hormone. The stimulation of the activity of enzymes involved in pregnenolone
synthesis. J Biol Chem. 1970; 245(1):152-9.
79. Tuckey RC, Cameron KJ. Catalytic properties of cytochrome P-450scc purified from
the human placenta: comparison to bovine cytochrome P-450scc. Biochim Biophys
Acta. 1993; 1163(2):185-94.
80. Kuwada M, Kitajima R, Suzuki H, Horie S. Purification and properties of cytochrome
P-450 (SCC) from pig testis mitochondria. Biochem Biophys Res Commun. 1991;
176(3):1501-8.
81. Shikita M, Hall PF. Cytochrome P-450 from bovine adrenocortical mitochondria: an
enzyme for the side chain cleavage of cholesterol. I. Purification and properties. J Biol
Chem. 1973; 248(16):5598-604.
82. Shikita M, Hall PF. Cytochrome P-450 from bovine adrenocortical mitochondria: an
enzyme for the side chain cleavage of cholesterol. II. Subunit structure. J Biol Chem.
1973; 248(16):5605-9.
70
83. Simpson ER. Cholesterol side-chain cleavage, cytochrome P450, and the control of
steroidogenesis. Mol Cell Endocrinol. 1979; 13(3):213-27.
84. Lavoie HA, King SR. Transcriptional regulation of steroidogenic genes: STARD1,
CYP11A1 and HSD3B. Exp Biol Med (Maywood). 2009; 234(8):880-907.
85. Raucci F, D'Aniello A, Di Fiore MM. Stimulation of androgen production by
D-aspartate through the enhancement of StAR, P450scc and 3β-HSD mRNA levels in
vivo rat testis and in culture of immature rat Leydig cells. Steroids. 2014; 84:103-10.
86. Doi M, Takahashi Y, Komatsu R, et al. Salt-sensitive hypertension in circadian
clock-deficient Cry-null mice involves dysregulated adrenal Hsd3b6. Nat Med. 2010;
16(1):67-74.
87. Okamura H, Doi M, Yamaguchi Y, Fustin JM. Hypertension due to loss of clock:
novel insight from the molecular analysis of Cry1/Cry2-deleted mice. Curr Hypertens
Rep. 2011; 13(2):103-8.
88. Doi M, Satoh F, Maekawa T, et al. Isoform-specific monoclonal antibodies against
3β-hydroxysteroid dehydrogenase/isomerase family provide markers for
subclassification of human primary aldosteronism. J Clin Endocrinol Metab. 2014;
99(2):E257-62.
89. Konosu-Fukaya S, Nakamura Y, Satoh F, et al. 3β-hydroxysteroid dehydrogenase
isoforms in human aldosterone-producing adenoma. Mol Cell Endocrinol. 2015;
408:205-12.
90. Verwoert GC, Hofland J, Amin N, et al. Expression and gene variation studies deny
association of human HSD3B1 gene with aldosterone production or blood pressure.
Am J Hypertens. 2015; 28(1):113-20.
91. Hough D, Storbeck K, Cloete SW, Swart AC, Swart P. Relative contribution of
P450c17 towards the acute cortisol response: Lessons from sheep and goats. Mol Cell
Endocrinol. 2015; 408:107-13.
92. Sakuma I, Suematsu S, Matsuzawa Y, et al. Characterization of steroidogenic enzyme
expression in aldosterone-producing adenoma: a comparison with various human
adrenal tumors. Endocr J. 2013; 60(3):329-36.
93. Tong A, Jia A, Yan S, Zhang Y, Xie Y, Liu G. Ectopic cortisol-producing
adrenocortical adenoma in the renal hilum: histopathological features and
steroidogenic enzyme profile. Int J Clin Exp Pathol. 2014; 7(7):4415-21.
94. Hirsch A, Hahn D, Kempna P, et al. Metformin inhibits human androgen production
71
by regulating steroidogenic enzymes HSD3B2 and CYP17A1 and complex I activity
of the respiratory chain. Endocrinology. 2012; 153(9):4354-66.
95. Xu B, Gao L, Cui Y, et al. SET protein up-regulated testosterone production in the
cultured preantral follicles. Reprod. Biol Endocrinol. 2013; 11:9.
96. Warne PA, Greenfield NJ, Lieberman S. Modulation of the kinetics of cholesterol
side-chain cleavage by an activator and by an inhibitor isolated from the cytosol of the
cortex of bovine adrenals. Proc Natl Acad Sci U S A. 1983; 80(7):1877-81.
97. Beaudoin C, Lavallee B, Tremblay Y, et al. Modulation of
17alpha-hydroxylase/17,20-lyase activity of guinea pig cytochrome P450c17 by
site-directed mutagenesis. DNA Cell Biol. 1998; 17(8):707-15.
98. Rosa S, Steigert M, Lang-Muritano M, I'Allemand D, Schoenle EJ, Biason-Lauber A.
Clinical, genetic and functional characteristics of three novel CYP17A1 mutations
causing combined 17alpha-hydroxylase/17,20-lyase deficiency. Horm Res Paediatr.
2010; 73(3):198-204.
99. Dalla Valle L, Ramina A, Vianello S, Belvedere P, Colombo L. Kinetic analysis of
duodenal and testicular cytochrome P450c17 in the rat. J Steroid Biochem Med Biol.
1996; 58(5-6):577-84.
100. Thomas JL, Mason JI, Brandt S, Spencer BR Jr, Norris W. Structure/function
relationships responsible for the kinetic differences between human type 1 and type 2
3beta-hydroxysteroid dehydrogenase and for the catalysis of the type 1 activity. J Biol
Chem. 2002; 277(45): 42795-801.
101. Riepe FG, Hiort O, Sippell WG, Krone N, Holterhus PM. Functional and structural
consequences of a novel point mutation in the CYP21A2 gene causing congenital
adrenal hyperplasia: potential relevance of helix C for P450
oxidoreductase-21-hydroxylase interaction. J Clin Endocrinol Metab. 2008;
93(7):2891-5.
102. Fisher A, Fraser R, Mc Connell J, Davies E. Amino acid residue 147 of human
aldosterone synthase and 11beta-hydroxylase plays a key role in 11beta-hydroxylation.
J Clin Endocrinol Metab. 2000; 85(3):1261-6.
103. LaSala D, Shibanaka Y, Jeng AY. Coexpression of CYP11B2 or CYP11B1 with
adrenodoxin and adrenodoxin reductase for assessing the potency and selectivity of
aldosterone synthase inhibitors. Anal Biochem. 2009; 394(1):56-61.
104. Baker ME. Unusual evolution of 11beta- and 17beta-hydroxysteroid and retinol
72
dehydrogenases. BioEssays. 1996; 18(1):63-70.
105. Corbin CJ, Mapes SM, Lee YM, Conley AJ. Structural and functional differences
among purified recombinant mammalian aromatases: glycosylation, N-terminal
sequence and kinetic analysis of human, bovine and the porcine placental and gonadal
isozymes. Mol Cell Endocrinol. 2003; 206(1-2):147-57.
106. Akimoto Y, Hasada R, Sakuma J, Ono I, Kobayashi S. Generation Alternation Model
for Real-Coded GA Using Multi-parent: Proposal and Evaluation of Just Generation
Gap (JGG). SICE Symp Decentralized Auton Syst. 2007; 19:341-46.
107. Akimoto Y, Nagata Y, Sakuma J, Ono I, Kobayashi S. Proposal and Evaluation of
Adaptive Real-Coded Crossover AREX. Trans Jpn Soc Artif Intel. 2009;
24(6):446-58.
108. Kobayashi S. The Frontiers of Real-Coded Genetic Algorithms. Trans Jpn Soc Artif
Intel. 2009; 24(1):147-62.
109. Akimoto Y, Nagata Y, Sakuma J, Ono I, Kobayashi S. Analysis of the behavior of
MGG and JGG as a selection model for real-coded genetic algorithms. Trans Jpn Soc
Artif Intel. 2010; 25(2):281-9.
110. Santen RJ, Misbin RI. Aminoglutethimide: review of pharmacology and clinical use.
Pharmacotherapy. 1981; 1(2):95-120.
111. Carella MJ, Dimitrov NV, Gossain VV, Srivastava L, Rovner DR. Adrenal effects of
low-dose aminoglutethimide when used alone in postmenopausal women with
advanced breast cancer. Metabolism. 1994; 43(6):723-7.
112. Johansson MK, Sanderson JT, Lund BO. Effects of 3-MeSO2-DDE and some CYP
inhibitors on glucocorticoid steroidogenesis in the H295R human adrenocortical
carcinoma cell line. Toxicol In Vitro. 2006; 16(2):113-21.
113. Touitou Y, Bogdan A, Legrand JC, Desgrez P. Aminoglutethimide and glutethimide:
effects on 18-hydroxycorticosterone biosynthesis by human and sheep adrenals in
vitro. Acta Endocrinol (Copenh). 1975; 80(3):517-26.
114. Hart MM, Straw JA. Studies on the site of action of o,p'-DDD in the dog adrenal
cortex. 1. Inhibition of ACTH-mediated pregnenolone synthesis. Steroids. 1971;
17(5):559-74.
115. Ojima M, Saitoh M, Itoh N, Kusano Y, Fukuchi S, Naganuma H. The effects of
o,p'-DDD on adrenal steroidogenesis and hepatic steroid metabolism. Nippon
Naibunpi Gakkai Zasshi. 1985; 61(3):168-78.
73
116. Canosa LF, Ceballos NR. Effects of different steroid-biosynthesis inhibitors on the
testicular steroidogenesis of the toad Bufo arenarum. J Comp Physiol B. 2001;
171(6):519-26.
117. Kossor DC, Kominami S, Takemori S, Colby HD. Role of the steroid 17
alpha-hydroxylase in spironolactone-mediated destruction of adrenal cytochrome
P-450. Mol Pharmacol. 1991; 40(2):321-5.
118. Cheng SC, Suzuki K, Sadee W, Harding BW. Effects of spironolactone, canrenone
and canrenoate-K on cytochrome P450, and 11beta- and 18-hydroxylation in bovine
and human adrenal cortical mitochondria. Endocrinology. 1976; 99(4):1097-106.
119. Struthers A, Krum H, Williams GH. A comparison of the aldosterone-blocking agents
eplerenone and spironolactone. Clin Cardiol. 2008; 31(4):153-8.
120. Yamasaki K, Sawaki M, Noda S, et al. Comparison of the Hershberger assay and
androgen receptor binding assay of twelve chemicals. Toxicology. 2004;
195(2-3):177-86.
121. Eil C, Edelson SK. The use of human skin fibroblasts to obtain potency estimates of
drug binding to androgen receptors. J Clin Endocrinol Metab. 1984; 59: 51-5.
122. Ye P, Yamashita T, Pollock DM, Sasano H, Rainey WE. Contrasting effects of
eplerenone and spironolactone on adrenal cell steroidogenesis. Horm Metab Res.
2009; 41(1):35-9.
123. Yanagibashi K, Haniu M, Shively JE, Shen WH, Hall P. The synthesis of aldosterone
by the adrenal cortex. Two zones (fasciculata and glomerulosa) possess one enzyme
for 11 beta-, 18-hydroxylation, and aldehyde synthesis. J Biol Chem. 1986;
261(8):3556-62.
124. Morishita K, Okumura H, Ito N, Takahashi N. Primary culture system of
adrenocortical cells from dogs to evaluate direct effects of chemicals on
steroidogenesis. Toxicology. 2001; 165(2-3):171-8.
125. Lamberts SW, Bons EG, Bruining HA, de Jong FH. Differential effects of the
imidazole derivatives etomidate, ketoconazole and miconazole and of metyrapone on
the secretion of cortisol and its precursors by human adrenocortical cells. J Pharmacol
Exp Ther. 1987; 240(1):259-64.
126. Yin L, Lucas S, Maurer F, Kazmaier U, Hu Q, Hartmann RW. Novel
imidazol-1-ylmethyl substituted 1,2,5,6-tetrahydropyrrolo[3,2,1-ij]quinolin-4-ones as
potent and selective CYP11B1 inhibitors for the treatment of Cushing's syndrome. J
74
Med Chem. 2012; 55(14):6629-33.
127. DiMattina M, Maronian N, Ashby H, Loriaux DL, Albertson BD. Ketoconazole
inhibits multiple steroidogenic enzymes involved in androgen biosynthesis in the
human ovary. Fertil Steril. 1988; 49(1):62-5.
128. Ayub M, Levell MJ. Inhibition of human adrenal steroidogenic enzymes in vitro by
imidazole drugs including ketoconazole. J Steroid Biochem. 1989; 32(4):515-24.
129. Engelhardt D, Weber MM, Miksch T, Abedinpour F, Jaspers C. The influence of
ketoconazole on human adrenal steroidogenesis: incubation studies with tissue slices.
Clin Endocrinol (Oxf). 35(2):163-8.
130. Bhasin S, Sikka S, Fielder T, et al. Hormonal effects of ketoconazole in vivo in the
male rat: mechanism of action. (1986) Endocrinology. 1986; 118(3):1229-32.
131. Albertson BD, Frederick KL, Maronian NC, et al. The effect of ketoconazole on
steroidogenesis: I. Leydig cell enzyme activity in vitro. Res Commun Chem Pathol
Pharmacol. 1988; 61(1):17-26.
132. Albertson BD, Maronian NC, Frederick KL, et al. The effect of ketoconazole on
steroidogenesis. II. Adrenocortical enzyme activity in vitro. Res Commun Chem
Pathol Pharmacol. 1988; 61(1):27-34.
133. Zisterer DM, Williams DC. Calmidazolium and other imidazole compounds affect
steroidogenesis in Y1 cells: lack of involvement of the peripheral-type benzodiazepine
receptor. J Steroid Biochem Mol Biol. 1997; 60(3-4):189-95.
134. Walsh LP, Kuratko CN, Stocco DM. Econazole and miconazole inhibit
steroidogenesis and disrupt steroidogenic acute regulatory (StAR) protein expression
post-transcriptionally. J Steroid Biochem Mol Biol. 2000; 75(4-5):229-36.
135. Ohno S, Shinoda S, Toyoshima S, Nakazawa H, Makino T, Nakajin S. Effects of
flavonoid phytochemicals on cortisol production and on activities of steroidogenic
enzymes in human adrenocortical H295R cells. J Steroid Biochem Mol Biol. 2002;
80(3):355-63.
136. Kikuchi S, Tominaga D, Arita M, Takahashi M, Tomita M. Dynamic modeling of
genetic networks using genetic algorithm and S-system. Bioinformatics. 2003;
19(5):643-50.
137. Shikata N, Maki Y, Nakatsui M, et al. Determining important regulatory relations of
amino acids from dynamic network analysis of plasma amino acids. Amino Acids.
2009; 38(1):179-87.
75
138. Nakatsui M, Ueda T, Maki Y, Ono I, Okamoto M. Method for inferring and extracting
reliable genetic interactions from time-series profile of gene expression. Math Biosci.
2008; 215(1):105-14.
139. Komori A, Maki Y, Nakatsui M, Ono I, Okamoto M. Efficient Numerical
Optimization Algorithm Based on New Real-Coded Genetic Algorithm, AREX + JGG,
and Application to the Inverse Problem in Systems Biology. Appl Math. 2012;
3(10A):1463-70.
140. Roubos JA, van Straten G, van Boxtel AJB. An evolutionary strategy for fed-batch
bioreactor optimization; concepts and performance. J Biotechnol. 1999;
67(2-3):173-87.
141. Sarkar D, Modak JM. Genetic algorithms with filters for optimal control problems in
fed-batch bioreactors. Bioprocess Biosyst Eng. 2004; 26(5):295-306.
142. Harada K, Ikeda K, Sakuma J, Ono I, Kobayashi S. Hybridization of Genetic
Algorithm with Local Search in Multiobjective Function Optimization:
Recommendation of GA then LS. Trans Jpn Soc Artif Intel. 2006; 21:482-92.
143. Szabo S, Lippe IT. Adrenal gland: chemically induced structural and functional
changes in the cortex. Toxicol Pathol. 1989; 17(2):317-29.
144. Chen-Pan C, Pan IJ, Yamamoto Y, Sakogawa T, Yamada J, Hayashi Y. Prompt
recovery of damaged adrenal medullae induced by salinomycin. Toxicol Pathol. 1999;
27(5):563-72.