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

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),


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]














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),



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


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


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,


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,


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.

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

Documents

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