PERGAMON
TIV Toxicology in Vitro
Effects of 3-MeSO2-DDE and some CYP inhibitors on glucocorticoid steroidogenesis in the H295R human adrenocortical carcinoma cell line
M.K. Johanssonª,*, J.T. Sandersonb, B-O. Lunda
a Department of Environmental Toxicology, Uppsala University, Norbyv. 18A, SE-752 36 Uppsala, Sweden
b Research Institute of Toxicology, PO Box 80.176, Yalelaan 2, NL-3508 TD Utrecht, The Netherlands
Accepted 28 October 2001
Abstract
The formation of steroids in the H295R human adrenocortical carcinoma cell line was analysed by HPLC or RIA, and based on these data the apparent catalytic activities of CYP11A, CYP17, CYP21 and CYP11B1 in this cell line were calculated. The envir- onmental pollutant 3-methylsulfonyl-DDE (3-MeSO2-DDE) and the cytochrome P450 (CYP) inhibitors ketoconazole, metyrapone and aminoglutethimide were studied for their effects on the steroid formation. Metyrapone (IC50 of 1 µM) and 3-MeSO2-DDE (10 µM: 66±10% of control) were found to inhibit the apparent CYP11B1 activity. Ketoconazole inhibited all enzymes examined with the greatest effects on CYP11B1 (IC50 of 2.5 UM). Aminoglutethimide was examined only for effects on CYP11A activity and was shown to inhibit pregnenolone formation (20 µM: 61±4% of control). The possibility of studying all CYP enzymes in the corti- costeroidogenesis makes this cell line a valuable test system to examine effects of chemicals, such as suspected endocrine disruptors, on the human glucocorticoid hormone synthesis. The inhibition of cortisol formation by 3-MeSO2-DDE supports an interaction with the active site of CYP11B1, as previously reported in mouse adrenocortical Y1 cells. In mice, this interaction led to metabolic activation and a high adrenotoxicity of 3-MeSO2-DDE. Therefore studies on the adrenotoxicity of 3-MeSO2-DDE in humans are needed. C) 2002 Elsevier Science Ltd. All rights reserved.
Keywords: H295R; Glucocortocoid steroidogenesis; MeSO2-DDE; DDT; Adrenal cortex
1. Introduction
Even though the use of DDT is prohibited in many countries, this insecticide is still used extensively for controlling insect-borne diseases, such as malaria (Smith, 1999). DDT and its persistent and lipophilic metabolites have been identified at various trophic levels of the ecosystem, including in mammals, birds and fish. DDT and its metabolites are also suspected to be endo- crine disruptors in wildlife and humans (Brandt et al.,
1998). The DDT metabolite 3-methylsulfonyl-2,2-bis(4- chlorophenyl)-1,1-dichloroethene (3-MeSO2-DDE) was first discovered in the blubber of Baltic grey seals (Jen- sen and Jansson, 1976). More recently, 3-MeSO2-DDE was found to be present in Swedish human breast milk, blood and adipose tissue (Norén et al., 1996; Weistrand and Norén, 1997).
In experimental studies in mice, 3-MeSO2-DDE was found to accumulate as an irreversibly bound residue in the zona fasciculata of the adrenal cortex (Lund et al., 1988). At a single dose as low as 3 mg/kg, 3-MeSO2- DDE caused disorganization and disappearance of the central cristae in the mitochondria (Jönsson et al., 1991; Lindhe et al., 2001). The adrenotoxicity of 3-MeSO2- DDE is believed to be caused by its metabolism by the adrenocortical, mitochondrial CYP11B1 to a reactive intermediate which binds irreversibly to cellular proteins (Lund and Lund, 1995). The toxic potency of 3-MeSO2- DDE in humans is not known, although the compound
Abbreviations: 3ßHSD, 3ß-hydroxysteroid dehydrogenase; 3- MeSO2-DDE, 3-methylsulfonyl-2,2-bis(4-chlorophenyl)-1,1-dichloro- ethene; CYP, cytochrome P450; DDT, dichlorodiphenyltri- chloroethane; DMSO, dimethyl sulfoxide; ER, endoplasmatic reticulum; HPA, hypothalamicary-adrenal; PCB, poly- chlorinated biphenyl; RIA, radioimmunoassay.
* Corresponding author. Tel .: +46-18-471-2618; fax: +46-18- 518843.
was reported to be activated and bound by a human adrenal preparation (Jönsson and Lund, 1994).
The adrenal cortex is frequently affected by sub- stances due to direct, or stress-related toxicity via the hypothalamicuitary-adrenal (HPA) axis (Harvey, 1996). The rich blood supply to the adrenal gland in relation to its small tissue mass makes this endocrine organ vulnerable to toxic chemicals. Because of the high lipid content of the adrenal cortex, lipophilic com- pounds readily accumulate in the cortical area (Szabo and Lippe, 1989). Yet, toxic effects on the adrenal gland are most often overlooked during risk assessment of environmental pollutants (US EPA, 1998). Recently, we reported that persistent aryl methyl sulfone metabolites of PCBs and DDT competitively inhibited the CYP11B1 activity in mouse adrenocortical Y1 cells (Johansson et al., 1998). Among these aryl methyl sul- fones, 3-MeSO2-DDE inhibited CYP11B1 activity with a potency close to that of the drug metyrapone, a potent inhibitor of CYP11B1. The question of whether aryl methyl sulfones are selective inhibitors of CYP11B1 and/or also affect other enzymes in the glucocorticoid synthesis pathway was not addressed, because the Y1 cell line is derived from a mouse adrenocortical carci- noma containing only few CYP enzymes (Es-souni et al., 1992). Many drugs, such as etomidate (Wagner et al., 1984), ketoconazole (Ideyama et al., 1999), 4-amino- pyridin (Fraser et al., 1986), spironolactone (Colby, 1994) benznidazole (de Castro et al., 1992) and procaine (Noguchi et al., 1990), have been reported to inhibit various enzymes in the glucocorticoid synthesis path- way. Thus, there is a need for a human adrenocortical cell model to study effects of xenobiotics on the human glucocorticoid synthesis pathway. The human adreno-
cortical carcinoma cell line NCI-H295 (Gazdar et al., 1990), and the subculture H295R (Rainey et al., 1993), appear to provide useful model systems for mechanistic studies. The H295R cell line expresses mRNA for all the CYP enzymes when stimulated with 8-Br-cAMP, for- skolin or K + (Rainey et al., 1993; Staels et al., 1993). So far, this cell line has been used successfully for studies of the regulation of mRNA levels of the steroidogenic enzymes (Rainey et al., 1994).
Glucocorticoids, for example cortisol, are produced in the zona fasciculata, and to some extent in the zona reticularis and zona glomerulosa, of the adrenal cortex (Fig. 1). The adrenal cortex, which is under the influence of the HPA axis, is stimulated by adrenocorticotropin (ACTH) to produce glucocorticoids. The first and rate- limiting step in steroidogenesis is the mobilisation of chol- esterol for subsequent conversion to pregnenolone by the mitochondrial CYP11A (P450scc or 20,22-desmo- lase). In the endoplasmatic reticulum (ER), pregneno- lone is converted to progesterone by 3ß-hydroxysteroid dehydrogenase (3ßHSD). The next steps in glucocorticoid synthesis are catalyzed by the microsomal enzymes CYP17 (with 17a-hydroxylase and 17/20-desmolase activity), which converts progesterone to 17a-hydroxyprogesterone and CYP21 (21-hydroxylase), which converts 17x-hydro- xyprogesterone to 11-deoxycortisol. The final step is the conversion of 11-deoxycortisol to the active glucocorti- coid cortisol by the mitochondrial CYP11B1 (11ß- hydroxylase). Cortisol exerts numerous metabolic, devel- opmental, immunosuppressive, anti-inflammatory and other functions in the body. The catabolic properties of cortisol promote breakdown of carbohydrates, proteins and lipids. Thus, a well-functioning glucocorticoid syn- thesis is essential (for review, see Buckingham et al., 1997).
Dehydroepiandrosterone
☒
CYP17
3฿HSD
17a-Hydroxypregnenolone
Androstenedione
CYP17 ☒
3ßHSD
CYP17
Cholesterol
CYP11A
Pregnenolone
3ßHSD
Progesterone
CYP17
17a-Hydroxyprogesterone
CYP21
CYP21
11-Deoxycorticosterone
11-Deoxycortisol
CYP11B1
CYP11B1
Corticosterone
Cortisol
The present study was undertaken to investigate whether the human adrenocortical cell line H295R can be used as a valid test system for studying the effects of suspected endocrine disruptors on catalytic activities of the CYP enzymes active in the glucocorticoid synthesis pathway. By determination of the concentration of vari- ous steroid hormones secreted by the H295R cells dur- ing 30 min or 24-h incubation in the presence or absence of test substance, the apparent catalytic activities of the CYP11A, CYP17, CYP21 and CYP11B1 were exam- ined. The well-documented CYP enzyme inhibitors aminoglutethimide (an inhibitor of CYP11A), metyr- apone (a potent inhibitor of CYP11B1) and ketocona- zole (a potent inhibitor of various CYP enzymes), were used as test substances. In addition, the effects of 3-MeSO2-DDE were studied as a first step in the evalua- tion of the human relevance of the adrenotoxicity of 3-MeSO2-DDE previously reported in rodents. The results show that the H295R cell line is a useful test system for studies of catalytic enzyme activities and that the four tested substances inhibit enzymes of the gluco- corticoid synthesis pathway to various extents.
2. Materials and methods
2.1. Reagents and supplies
3-MeSO2-DDE (>99% purity) was a kind gift from Dr. Åke Bergman (Dept of Environmental Chemistry, Stockholm University, Sweden) (Bergman and Wacht- meister, 1977). Ketoconazole was bought from ICN Biomedicals Inc. (Aurora, OH, USA). All steroids, for- skolin, metyrapone and dimethyl sulfoxide (DMSO) were obtained from Sigma (St Louis, MO, USA). Ace- tonitrile, chloroform and methanol for HPLC analysis were purchased from Tamro (Mölndal, Sweden). Preg- nenolone tritium kit for radioimmunoassay was bought from ICN (Costa Mesa, CA, USA). Trilostane (pro- duced by Sanofi-Winthrop) was a kind gift from Dr. George Margetts (Stegram Pharmaceuticals Ltd, Bill- inghurst, UK). SU-10603 was a kind gift from Dr. Honora Cooper Eckhardt (Novartis Pharmaceuticals Corporation, Summit, USA).
2.2. Cell culture
H295R cells were obtained from the American Type Culture Collection (ATCC # CRL-2128) and grown in 75-cm2 flasks (Life Technologies, Täby, Sweden) under culture conditions published previously (Rainey et al., 1993, 1994). Briefly, cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM-F12, Gibco BRL, UK) supplemented with 1% ITS Plus Premix (contain- ing insulin (6.25 µg/ml), transferrin (6.25 µg/ml), sele- nium (6.25 ng/ml), BSA (1.25 mg/ml), and linoleic acid
(5.35 µg/ml); Collaborative Biomedical Products via Labora, Sollentuna, Sweden), 50 units/ml penicillin and 50 µg/ml streptomycin (Gibco BRL, UK). For the determination of the pregnenolone formation, cells were cultured in the presence of 2% of the steroid-free serum replacement Ultroser SF (Soprachem, France), whereas for the determination of 17a-hydroxyprogesterone, androstenedione, 11-deoxycortisol and cortisol forma- tion, cells were cultured in the presence of 2% NuSerum (NuS; Collaborative Biomedical Products via Labora, Sollentuna, Sweden). The cells were grown at 37℃ in a chamber containing 5% CO2 and a humidity of 95%. The cell-doubling time of the H295R cells was approxi- mately 3 days in both laboratories. All test substances, i.e. 3-MeSO2-DDE, ketoconazole, metyrapone and aminoglutethimide (only examined for effects on CYP11A), as well as forskolin and enzyme substrates used in the studies, were dissolved in DMSO and added from 1000-fold stock solutions. Controls were treated with DMSO. The DMSO concentration in the culture medium never exceeded 0.1%.
2.3. Studies of effects of forskolin and test substances on the pregnenolone formation
Before each experiment, the H295R cells were seeded in 24-well plates. When the cells were almost confluent in each well (between 1 and 2×105 cells/well), the cells were exposed to test substances (20 µM) and forskolin (10 µM) in the reaction buffer [KREBS buffer, pH 7.4, containing 1 µM 22(R)-hydroxycholesterol, the 3ß- hydroxysteroid dehydrogenase inhibitor trilostane (10 AM) and the CYP17 inhibitor SU-10603 (0.2 µM)] and incubated for 30 min at 37℃. The inhibitors were used to ensure that no further metabolism of pregnenolone would occur in the cells (see Fig. 1). The reaction was stopped by placing the plates on ice. From each well, 400 ul of reaction buffer was removed to a plastic vial and mixed. A 5-ul aliquot (or more, depending on the amount of pregnenolone formed during the reaction) was taken from each vial and added to a glass tube suit- able for radioimmunoassay (RIA) studies. The RIA for pregnenolone was performed according to instructions from the manufacturer. The rate of pregnenolone for- mation is considered to represent the CYP11A activity.
2.4. Studies of effects of forskolin and test substances on the formation of 17a-hydroxyprogesterone, androstenedione, 11-deoxycortisol and cortisol
Before each experiment, the H295R cells were seeded in six-well tissue culture plates (2 ml medium, Corning Costar Corp., Cambridge, MA, USA) and cultured as described above. On day 1, at approximately 80% con- fluency, old medium was removed and 1 ml fresh serum- free medium containing 1% ITS Plus Premix was added
to the wells together with forskolin (10 µM) and/or test substances (0.01-10 µM) and incubated for 24 h. For- skolin was used to stimulate the enzyme activities via activation of the protein kinase A pathway and was used in controls as well. Progesterone, 17x-hydroxy- progesterone and 11-deoxycortisol were used as sub- strates for measurement of the apparent activities of CYP17, CYP21 and CYP11B1, respectively (Fig. 1). Pro- gesterone (20 µM) and 17x-hydroxyprogesterone (20 µM) were added (premixed in 40 ul medium) to the cells 3 h before terminating the experiment. To obtain measurable levels of cortisol, 11-deoxycortisol (20 µM) was added together with test substances and forskolin at the start of the experiment. In all experiments, the cells were exposed to forskolin and/or a test substance for 24 h before an ali- quot of the medium was collected and analyzed for steroid concentration by HPLC. The remaining medium was removed and the cells rinsed with phosphate buffered saline and dissolved in NaOH (0.1 M). Cellular toxicity was evaluated by visual inspection of the cell morphol- ogy, cell attachment and protein content of the wells. No effects were observed on any of these parameters at the test substance concentrations used (data not shown). The protein content of the cells in each well was deter- mined spectrophotometrically according to Lowry et al. (1951) using bovine serum albumin as standard.
2.5. HPLC analysis of steroid concentrations
A 500-750-ul aliquot of the medium was taken for extraction with three volumes of chloroform:methanol (2:1). After a second extraction of the medium with chloroform, the chloroform phases were pooled, evapor- ated and dissolved in acetonitrile:water (1:1, v/v) for analysis of steroids. The HPLC column (4.6 mm×250 mm) was packed with Lichrosorb RP18 (5 um diameter) and eluted at a flow rate of 1.0 ml/min with a mixture of 18% acetonitrile:tetrahydrofuran (9:1, v/v) and 82% methanol:water (2:3, v/v). After injection, the concentra- tion of acetonitrile:tetrahydrofuran was increased line- arly from 18 to 35% over a period of 25 min, then from 35 to 60% over 5 min and from 60 to 80% over 2 min, after which it was kept constant for 10 min. Steroids were detected by their UV absorbance at 241 nm. The steroids were quantified using a standard curve based on synthetic steroids. The retention times of the steroids pro- gesterone, 17a-OH-progesterone, deoxycorticosterone, androstenedione, 11-deoxycortisol and cortisol were 35, 26, 23, 22, 16 and 11 min, respectively. The retention time of metyrapone was 10 min. The sum of the synthesized 17a-hydroxyprogesterone, androstenedione and 11-deoxy- cortisol was assumed to represent the apparent CYP17 activity. The apparent activities of CYP21 and CYP11B1 were measured by the formation of 11-deoxycortisol and cortisol, respectively. The enzyme activities were expressed as nmol or pmol steroid(s)/mg protein/h.
2.6. Statistics
The studies of the apparent CYP17, CYP21 and CYP11B1 activities were carried out with controls (n=3-6 wells/treatment) and various concentrations of test substances (0.01-10 µM, n=1-3). At least four independent dose-response experiments were performed (at different occasions) for each test substance. Statis- tical analysis was conducted on the absolute values using the paired, one-tailed Student’s t-test, except for the study of the apparent CYP11A activity and the effect of metyrapone on CYP11B1 activity, where trip- licate samples were used of each concentration, and the unpaired, one-tailed Student’s t-test was used for statis- tical evaluation. All data were assumed to be normally distributed.
3. Results
3.1. Forskolin stimulation
Forskolin treatment increased the apparent enzyme activities about three-fold, with the increases of CYP17 and CYP11B1 activities slightly exceeding those of CYP21 and CYP11A. A comparison of the activ- ities of the enzymes revealed that the microsomal enzymes CYP17 and CYP21 had relatively high activities, whereas the activities of the mitochondrial enzymes CYP11A and CYP11B1 were low (Fig. 2). There was no difference between the cellular protein content of the forskolin treated (10 µM) (0.19±0.02 mg protein/well, n=5) and untreated cells (0.18±0.03 mg protein/well, n=5).
7.5
nmol steroid(s)/mg
protein/h
5.0
2.5
0.0
CYP11A
CYP17
CYP21
CYP11B1
3.2. Effects of test substances on CYP11A activity
The apparent CYP11A activity was determined by measuring the pregnenolone formation from 22(R)- hydroxycholesterol in the presence of the enzyme inhi- bitors trilostane and SU-10603. After 30 min of incu- bation, the pregnenolone formation in the control wells was 272±26 ng/mg protein/h (mean±S.D., n=3). 20 UM of the drugs aminoglutethimide, ketoconazole or metyrapone inhibited the CYP11A activity, but only ketoconazole had an IC50 value of less than 20 µM (Fig. 3). 3-MeSO2-DDE did not affect this enzyme.
3.3. Effects of test substances on CYP17 activity
The apparent CYP17 activity was determined by measuring the sum of the synthesized 17x-hydroxy- progesterone, androstenedione and 11-deoxycortisol using progesterone as a substrate. Data from one typical experiment are shown in Table 1 as an illustration of the production of the different steroid hormones. After 24 h of incubation, ketoconazole (IC50 approx. 8 UM) significantly inhibited the CYP17 activity, whereas 3-MeSO2-DDE did not affect the CYP17 activity at any of the concentrations (Fig. 4).
| Concentration of ketoconazole (UM) | 17-Hydroxyprogesterone | Androstenedione | 11-Deoxycortisol | Sum of steroids |
|---|---|---|---|---|
| 0 | 2.07±0.13a | 0.42±0.01a | 1.90±0.06a | 4.40±0.18a |
| 0.01 | 1.99 | 0.39 | 1.84 | 4.22 |
| 0.1 | 1.86 | 0.34 | 1.71 | 3.91 |
| 1 | 1.77 | 0.23 | 1.58 | 3.58 |
| 5 | 1.57 | 0.48 | 0.62 | 2.67 |
| 10 | 0.51 | 0.55 | 0.23 | 1.29 |
Progesterone (20 µM) was added as a substrate 3 h before terminating the experiment. Steroids were analyzed by HPLC and are expressed as nmol/mg protein/h.
a Mean±S.D., n=3.
CYP11A
400
CYP11A activity (ng pregnenolone/mg protein/h)
300
*
**
200
100
0
control
metyrapone
aminoglutethimide
ketoconazole
3-MeSO2-DDE
100
*
*
T
% of control
T
**
50
0
1
5
10
concentration of test substances (LM)
metyrapone
MeSO2-DDE
ketoconazole
100
% of control
*
*
50
**
0
1
5
10
concentration of test substances (LM)
☐ metyrapone
MeSO2-DDE
ketoconazole
Metyrapone had a weak effect at 5 µM, but it was con- sidered a chance finding due to lack of effect at 10 µM, and thus we conclude that no biologically relevant inhi- bition of the CYP17 activity occurred at these con- centrations of metyrapone.
3.4. Effects of test substances on CYP21 activity
The apparent CYP21 activity was determined by measuring 11-deoxycortisol formation using 17x- hydroxyprogesterone as a substrate. After 24 h of incu- bation, the mean 11-deoxycortisol formation in the control cells was ranging from 3.7 to 6.9 nmol/mg/h in the performed experiments, with a S.D. never exceeding 0.3 nmol/mg/h (three wells/experiment). Ketoconazole (IC50 approx. 8 µM) and 3-MeSO2-DDE (80% of con- trol activity at 10 µM) significantly inhibited the CYP21 activity (Fig. 5). Metyrapone did not affect this enzyme.
3.5. Effects of test substances on CYP11B1 activity
The apparent CYP11B1 activity was determined by measuring the formation of cortisol using 11-deoxy- cortisol as a substrate. After 24 h of incubation, the mean cortisol formation in the control cells was ranging from 45 to 79 pmol/mg/h in the performed experiments, with a SD never exceeding 14 pmol/mg/h (3-6 wells/ experiment). Metyrapone (one of four experiments is shown), 3-MeSO2-DDE and ketoconazole significantly inhibited the CYP11B1 activity (Fig. 6). The IC50 values of metyrapone and ketoconazole were 1 and 2.5 M,
150
% of control
100
H
#
$
**
**
I
**
50
**
**
0
-9
-8
-7
-6
-5
concentration of test substances (log M)
metyrapone ^ ketoconazole
· 3-MeSO2-DDE
respectively. 10 µM 3-MeSO2-DDE inhibited cortisol formation to 66% of the control value.
4. Discussion
Catalytic enzyme activities are difficult to study in living cells. In almost all cases, a substrate is not con- verted to a single product by a single enzyme. Rather, the substrate (precursor steroid) is usually metabolized by more than one enzyme, and the steroid product may be further metabolized. Thus, our results give an indi- cation of the catalytic CYP enzyme activity in the H295R cell line. In order to examine the response of the H295R cells, three well-known cytochrome P450 inhi- bitors ketoconazole, metyrapone and aminogluteth- imide were studied. Their effects on the apparent CYP enzyme activities were compared with previously pub- lished data.
The antimycotic agent ketoconazole is known to inhibit several CYP enzyme activities by binding to the heme of the enzyme, thus preventing activation of molecular oxygen (Feldman, 1986). In the H295R cells, ketoconazole inhibited all the studied CYP enzyme activities. Although this is the first report to show inhi- bition of human CYP11A activity by ketoconazole, this finding is not surprising in view of the relatively non- selective inhibitory properties of ketoconazole, and the fact that it has also been found to inhibit CYP11A activity in rat adrenal cells, rat testis and bovine mito- chondrial fractions (Loose et al., 1983; Kan et al., 1985;
Nagai et al., 1986). The overall inhibition profile of ketoconazole in our cell system resembles that found in most previously reported human adrenal in vitro studies (Albertson et al., 1988; Ayub and Levell, 1989; Ideyama et al., 1999). As a comparison, the inhibitory potency of ketoconazole on the steroidogenic CYP enzymes in this study seemed lower than its potency to inhibit the human hepatic drug-metabolizing CYP3A family (Ki in the nanomolar range) (Venkatakrishnan et al., 2000).
In H295R cells, the mitochondrial CYP enzyme activities CYP11A and CYP11B1 were inhibited by metyrapone. Similar results have been demonstrated in human and bovine adrenal mitochondrial fractions and in dispersed human adrenocortical cells (Carballeira et al., 1976; Tobes et al., 1985; Lamberts et al., 1987). Mety- rapone is a drug designed to be a selective CYP11B1 inhi- bitor. Nevertheless, the inhibition of CYP11A activity in our study shows that higher concentrations of metyrapone may lead to non-selective inhibition of other cytochrome P450 enzymes in the glucocorticoid synthesis pathway. Metyrapone and ketoconazole were almost equipotent inhibitors of CYP11B1 activity in H295R cells. In contrast, mouse adrenocortical Y1 cells were 20 times more sensitive to ketoconazole than to metyrapone in an identical experimental regimen (Johansson et al., 1998). There are thus species differ- ences in sensitivity to these test substances, which emphasizes the importance of having species-specific models and using human cells when appropriate.
Aminoglutethimide was originally developed as an anticonvulsant, but has also been used in breast cancer therapy as an aromatase inhibitor (Nicholls et al., 1986). In glucocorticoid steroidogenesis, aminoglutethimide inhibits the enzyme activities of CYP11A, CYP21 and CYP11B1 to various extents (Harvey, 1996). In NCI- H295 cells, aminoglutethimide has been demonstrated to decrease the CYP11B1 activity (Fassnacht et al., 1998). Our results show that aminoglutethimide inhib- ited the CYP11A activity in the H295R cells, although with a rather low potency.
The H295R cell line is responsive to forskolin (Rainey et al., 1993; Staels et al., 1993), indicating that activa- tion of the protein kinase A pathway is functional. In our cell system, the steroid secretion pattern resembled that of NCI-H295 cells, for instance with regard to cortisol being a minor product (Gazdar et al., 1990). However, Rainey and co-workers (1993) reported that forskolin-treated H295R cells primarily secreted corti- sol. In long-term culture of NCI-H295 cells, a selection of different clones which express different patterns of CYP enzymes have been observed (Zenkert et al., 2000). This observation could possibly explain the relatively low secretion of cortisol in our H295R cells. In addition, human adrenal carcinomas showed lower activities of the mitochondrial CYP11A and CYP11B1 compared to normal adrenal tissue, which was suggested to be due to
a defective NADPH generation in the adrenal carcino- matous mitochondria (Brown and Fishman, 2000). There is thus a difference in mitochondrial enzyme activities between at least some clones of H295R cells and normal adrenal cells, but the presence of all CYP enzymes in H295R cells still makes this cell line a suit- able tool for studying potential effects of endocrine disruptors.
The persistent metabolite of DDT, 3-MeSO2-DDE, was recently shown to be a competitive inhibitor of CYP11B1 in mouse Y1 cells, with a potency close to that of the drug metyrapone (Johansson et al., 1998). The metabolic activation of 3-MeSO2-DDE by a human adrenal mitochondrial preparation (Jönsson and Lund, 1994), implies that this substance may be a CYP11B1 substrate in humans as well. The present study shows that 3-MeSO2-DDE inhibits CYP11B1 activity in H295R cells, and thus may interfere with human gluco- corticoid steroidogenesis. In fact, the inhibitory effect of 3-MeSO2-DDE on human CYP11B1 activity is only somewhat lower than that found in mouse Y1 cells (Lund and Lund, 1995). In that study, it was demon- strated that mouse CYP11B1 catalyses the metabolism of 3-MeSO2-DDE to a toxic reactive intermediate that binds irreversibly to cellular proteins. This CYP11B1- mediated bioactivation of 3-MeSO2-DDE is proposed to be the cause of the adrenotoxicity seen in mice (Jönsson et al., 1991). Thus, there may also be a risk for CYP11B1-mediated bioactivation of 3-MeSO2-DDE in humans, resulting in adrenotoxicity. In human H295R cells, the drugs metyrapone and ketoconazole were almost equipotent in inhibiting the CYP11B1 activity (IC50 of 1 and 2.5 uM, respectively) and roughly one order of magnitude more potent than 3-MeSO2-DDE. When interpreting the inhibitory potency of the test substances, it should be kept in mind that relatively high substrate concentrations were used. Thus, the actual Ki values of the test substances are probably lower than indicated by the IC50 values.
In certain areas of the world DDT intake by breast- fed infants clearly exceeds the Acceptable Daily Intake (ADI) recommended by the World Health Organization (Smith, 1999). In Swedish human breast milk, the DDT metabolite 3-MeSO2-DDE is the most abundant aryl methyl sulfone, and is present at concentrations of 0.4-5 ng/g lipids (Norén et al., 1996). Taken together, the adrenotoxic potency of 3-MeSO2-DDE in fetal, suckling and adult mice (Lund et al., 1988; Jönsson et al., 1991, 1992), the previously demonstrated bioactivation of 3-MeSO2-DDE by human adrenocortical mitochondria (Jönsson and Lund, 1994), and the results of this study, warrant a detailed study of the adrenotoxicity of this compound in humans.
In conclusion, we have demonstrated that the cyto- chrome P450 inhibitors ketoconazole, metyrapone, and aminoglutethimide differentially inhibited the apparent
catalytic CYP enzyme activities according to their known in vitro potencies in adrenocortical cells. We therefore propose the H295R cell line as a useful model for adrenocortical toxicity studies and to screen sub- stances, such as suspected endocrine disruptors, for their potential to interfere with human glucocorticoid synthesis pathway. Furthermore, we have demonstrated that the persistent environmental pollutant 3-MeSO2- DDE inhibited the apparent activities of CYP11B1 and CYP21 in this cell system. Considering the presence of 3-MeSO2-DDE in human breast milk and the possible adrenotoxicity of this compound in humans, our results emphasize the need for further studies to support a human risk assessment of 3-MeSO2-DDE.
Acknowledgements
This study was supported financially by the Swedish National Board for Laboratory Animals, the Oscar and Lili Lamm’s Memorial Foundation, and the Swedish Environmental Protection Agency.
References
Albertson, B.D., Maronian, N.C., Frederick, K.L., DiMattina, M., Feuillan, P., Dunn, J.F., Loriaux, D.L., 1988. The effect of ketoco- nazole on steroidogenesis II. Adrenocortical enzyme activity in vitro. Research Communications in Chemical Pathology and Phar- macology 61, 27-34.
Ayub, M., Levell, M.J., 1989. Inhibition of human adrenal steroido- genic enzymes in vitro by imidazole drugs including ketoconazole. Journal of Steroid Biochemistry 32, 515-524.
Bergman, Å., Wachtmeister, C.A., 1977. Synthesis of methylsulphonyl deriatives of 2,2-bis(4-chlorophenyl)-1,1-dicloroethylene (p,p’-DDE) present in seal from the baltic. Acta Chemica Scandinavia B31, 90- 91.
Brandt, I., Berg, C., Halldin, K., Brunström, B., 1998. Developmental and reproductive toxicity of persistent environmental pollutants. Archives of Toxicology 20 (Suppl.), 111-119.
Brown, J.W., Fishman, L.M., 2000. Biosynthesis and metabolism of steroid hormones by human adrenal carcinomas. Brazilian Journal of Medical and Biological Research 33, 1235-1244.
Buckingham, J.C., Gillies, G.E., Cowell, A .- M. (Eds). 1997. Stress, Stress Hormones and the Immune System. J. Wiley, Chichester.
Carballeira, A., Fishman, L.M., Jacobi, J.D., 1976. Dual sites of inhi- bition by metyrapone of human adrenal steroidogenesis: correlation of in vivo and in vitro studies. Journal of Clinical Endocrinology and Metabolism 42, 687-695.
Colby, H.D., 1994. In vitro assessment of adrenocortical toxicity. Journal of Pharmacological and Toxicological Methods 32, 1-6.
de Castro, C.R., Diaz de Toranzo, E.G., Castro, J.A., 1992. Benzni- dazole-induced ultrastructural alterations in rat adrenal cortex. Mechanistic studies. Toxicology 74, 223-232.
Es-souni, M., Ramirez, L.C., Bournot, P., 1992. 18-hydroxylation in the Y-1 adrenal cell line: response to ACTH and to culture conditions. Journal of Steroid Biochemistry and Molecular Biology 43, 535-541.
Fassnacht, M., Beuschlein, F., Vay, S., Mora, P., Allolio, B., Reincke, M., 1998. Aminoglutethimide suppresses adrenocorticotropin receptor expression in the NCI-h295 adrenocortical tumor cell line. Journal of Endocrinology 159, 35-42.
Feldman, D., 1986. Ketoconazole and other imidazole derivatives as inhibitors of steroidogenesis. Endocrine Reviews 7, 409-420.
Fraser, R., Holloway, C.D., Kenyon, C.J., 1986. Inhibition of corti- costeroid 11beta hydroxylation in bovine zona fasciculata cells by the potassium entry blocker 4-aminopyridine. Journal of Steroid Biochemistry 24, 777-778.
Gazdar, A.F., Oie, H.K., Shackleton, C.H., Chen, T.R., Triche, T.J., Myers, C.E., Chrousos, G.P., Brennan, M.F., Stein, C.A., La Rocca, R.V., 1990. Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Research 50, 5488-5496.
Harvey, P.W., 1996. An overview of adrenal gland involvement in toxicology. In: Harvey, P.W. (Ed.), The Adrenal in Toxicology: Target Organ and Modulator of Toxicity. Taylor & Francis Ltd, London, pp. 3-19.
Ideyama, Y., Kudoh, M., Tanimoto, K., Susaki, Y., Nanya, T., Nakahara, T., Ishikawa, H., Fujikura, T., Akaza, H., Shikama, H., 1999. YM116, 2-(1H-imidazol-4-ylmethyl)-9H-carbazole, decreases adrenal androgen synthesis by inhibiting C17-20 lyase activity in NCI-H295 human adrenocortical carcinoma cells. Japanese Journal of Pharmacology 79, 213-220.
Jensen, S., Jansson, B., 1976. Antropogenic substances in seal from the Baltic: methyl sulfone metabolites. Ambio 5, 257-260.
Johansson, M., Larsson, C., Bergman, Å, Lund, B .- O., 1998. Struc- ture-activity relationship for inhibition of CYP11B1-dependent glucocorticoid synthesis in Y1 cells by aryl methyl sulfones. Pharmacology and Toxicology 83, 225-230.
Jönsson, C .- J., Lund, B .- O., 1994. In vitro bioactivation of the envir- onmental pollutant 3-methylsulphonyl-2, 2-bis(4-chlorophenyl)-1,1- dichloroethene in the human adrenal gland. Toxicology Letters 71, 169-175.
Jönsson, C .- J., Lund, B .- O., Bergman, Å, Brandt, I., 1992. Adreno- cortical toxicity of 3-methylsulphonyl-DDE; 3: studies in fetal and suckling mice. Reproductive Toxicology 6, 233-240.
Jönsson, C.J., Rodriguez-Martinez, H., Lund, B.O., Bergman, A., Brandt, I., 1991. Adrenocortical toxicity of 3-methylsulfonyl-DDE in mice II. Mitochondrial changes following ecologically relevant doses. Fundamental and Applied Toxicology 16, 365-374.
Kan, P.B., Hirst, M.A., Feldman, D., 1985. Inhibition of steroidogenic cytochrome P-450 enzymes in rat testis by ketoconazole and related imidazole anti-fungal drugs. Journal of Steroid Biochemistry 23, 1023-1029.
Lamberts, S.W., Bons, E.G., Bruining, H.A., de Jong, F.H., 1987. Differential effects of the imidazole derivatives etomidate, ketoco- nazole and miconazole and of metyrapone on the secretion of cor- tisol and its precursors by human adrenocortical cells. Journal of Pharmacology and Experimental Therapeutics 240, 259-264.
Lindhe, Ö., Lund, B .- O., Bergman, Å. Brandt, I., 2001. Irreversible binding and adrenocorticolytic activity of the DDT metabolite 3-methylsulfonyl-DDE examined in tissue-slice culture. Environ- mental Health Perspectives 109, 105-110.
Loose, D.S., Kan, P.B., Hirst, M.A., Marcus, R.A., Feldman, D., 1983. Ketoconazole blocks adrenal steroidogenesis by inhibiting cytochrome P450-dependent enzymes. Journal of Clinical Investi- gation 71, 1495-1499.
Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the folin reagent. Journal of Biological Chemistry 193, 265-275.
Lund, B.O., Bergman, A., Brandt, I., 1988. Metabolic activation and toxicity of a DDT-metabolite, 3-methylsulphonyl-DDE, in the adrenal zona fasciculata in mice. Chemico-Biological Interactions 65, 25-40.
Lund, B.O., Lund, J., 1995. Novel involvement of a mitochondrial steroid hydroxylase (P450c11) in xenobiotic metabolism. Journal of Biological Chemistry 270, 20895-20897.
Nagai, K., Miyamori, I., Ikeda, M., Koshida, H., Takeda, R., Suhara, K., Katagiri, M., 1986. Effect of ketoconazole (an imidazole anti- mycotic agent) and other inhibitors of steroidogenesis on cytochrome
P450-catalyzed reactions. Journal of Steroid Biochemistry 24, 321- 323.
Nicholls, P.J., Daly, M.J., Smith, H.J., 1986. Pharmacology of amino- glutethimide: structure/activity relationships and receptor interac- tions. Breast Cancer Research and Treatment 7, S55-S67.
Noguchi, A., Takamura, M., Yamada, K., Tou, S., Kawamura, M., 1990. Procaine inhibits cyclic AMP-induced steroidogenesis in iso- lated bovine adrenocortical cells. Japanese Journal of Pharmacology 52, 81-85.
Norén, K., Lundén, Å., Pettersson, E., Bergman, Å., 1996. Methyl- sulfonyl metabolites of PCBs and DDE in human milk in Sweden, 1972-1992. Environmental Health Perspectives 104, 766-772.
Rainey, W.E., Bird, I.M., Mason, J.I., 1994. The NCI-H295 cell line: a pluripotent model for human adrenocortical studies. Molecular and Cellular Endocrinology 100, 45-50.
Rainey, W.E., Bird, I.M., Sawetawan, C., Hanley, N.A., McCarthy, J.L., McGee, E.A., Wester, R., Mason, J.I., 1993. Regulation of human adrenal carcinoma cell (NCI-H295) production of C19 steroids. Jour- nal of Clinical Endocrinology and Metabolism 77, 731-737.
Smith, D., 1999. Worldwide trends in DDT levels in human breast milk. International Journal of Epidemiology 28, 179-188.
Staels, B., Hum, D.W., Miller, W.L., 1993. Regulation of ster- oidogenesis in NCI-H295 cells: a cellular model of the human fetal adrenal. Molecular Endocrinology 7, 423-433.
Szabo, S., Lippe, I.T., 1989. Adrenal gland: chemically induced struc- tural and functional changes in the cortex. Toxicologic Pathology 17, 317-329.
Tobes, M.C., Hays, S.J., Gildersleeve, D.L., Wieland, D.M., Beier- waltes, W.H., 1985. Adrenal cortical 11 beta-hydroxylase and side- chain cleavage enzymes (Requirement for the A- or B-pyridyl ring in metyrapone for inhibition). Journal of Steroid Biochemistry 22, 103-110.
US EPA, 1998. ORD Research Plan for endocrine Disruptors. EPA/ 600/R-98/087. US Environmental Protection Agency, Washington, DC, pp. 1-55.
Wagner, R.L., White, P.F., Kan, P.B., Rosenthal, M.H., Feldman, D., 1984. Inhibition of adrenal steroidogenesis by the anesthetic etomi- date. New England Journal of Medicine 310, 1415-1421.
Weistrand, C., Norén, K., 1997. Methylsulfonyl metabolites of PCBs and DDE in human tissue. Environmental Health Perspectives 105, 644-649.
Venkatakrishnan, K., von Moltke, L.L., Greenblatt, D.J., 2000. Effects of the antifungal agents on oxidative drug metabolism: clin- ical relevance. Clinical Pharmacokinetics 38, 111-180.
Zenkert, S., Schubert, B., Fassnacht, M., Beuschlein, F., Allolio, B., Reincke, M., 2000. Steroidogenic acute regulatory protein mRNA expression in adrenal tumours. European Journal of Endocrinology 142, 294-299.