Báo cáo Y học: Conditionally immortalized adrenocortical cell lines at undifferentiated states exhibit inducible expression of glucocorticoid-synthesizing genes

Eur. J. Biochem. 269, 69–81 (2002) (cid:211) FEBS 2002

Conditionally immortalized adrenocortical cell lines at undifferentiated states exhibit inducible expression of glucocorticoid-synthesizing genes

Kuniaki Mukai1, Hideko Nagasawa1,*, Reiko Agake-Suzuki1, Fumiko Mitani1, Keiko Totani1, Nobuaki Yanai2, Masuo Obinata2, Makoto Suematsu1 and Yuzuru Ishimura1

1Department of Biochemistry and Integrative Medical Biology, School of Medicine, Keio University, Shinjuku-ku, Tokyo, Japan; 2Institute of Development, Aging and Cancer, Tohoku University, Sendai, Japan

Notably, these cell lines displayed distinct expression pat- terns of the steroid 11b-hydroxylase P45011b gene respon- sible for the zone-specific synthesis of corticosterone. AcA201 cells expressed the P45011b gene at 33 (cid:176)C, showing the property of the zona fasciculata cells, while AcE60 cells expressed it upon a shift to a nonpermissive temperature (39 (cid:176)C). On the other hand, AcA101 expressed the P45011b gene at 39 (cid:176)C synergistically with exposure to dibutyryl cAMP. None of these clones express the zona glomerulosa- specific aldosterone synthase P450aldo gene under the con- ditions we tested. These results show that AcE60 and AcA101 cells display a pattern of the steroidogenic gene expression similar to that of the undi(cid:128)erentiated cell-layer and are capable of di(cid:128)erentiating into the zona fasciculata- like cells in vitro.

steroid hormone;

immor-

Keywords: adrenal cortex; talization; simian virus 40 large T-antigen.

To facilitate studies on di(cid:128)erentiation of adrenocortical cells and regulation of steroidogenic genes, we established cell lines from adrenals of adult transgenic mice harboring a temperature-sensitive large T-antigen gene of simian virus 40. Adrenal glands of the mice exhibited normal cortical zonation including a functionally undi(cid:128)erentiated cell-layer between the aldosterone-synthesizing zona glomerulosa cells and the corticosterone-synthesizing zona fasciculata cells. At a permissive temperature (33 (cid:176)C), established cell lines AcA201, AcE60 and AcA101 expressed steroidogenic genes side-chain encoding steroidogenic factor-1, cholesterol cleavage P450scc, and steroidogenic acute regulatory pro- tein, which are expressed throughout adrenal cortices and gonads. Genes encoding 3b-hydroxysteroid dehydrogenase and steroid 21-hydroxylase P450c21, which catalyze the intermediate steps for syntheses of both aldosterone and corticosterone, were inducible in the three cell lines in tem- perature- and/or dibutyryl cAMP-dependent manners.

enzymes,

aldosterone

synthase

hydroxysteroid dehydrogenase (3bHSD), and 21-hydroxy- lase cytochrome P450c21 (P450c21, the Cyp21a gene product) [1,3]. These enzymes are present throughout the adrenal cortex [4–6]. On the other hand, two structurally related cytochrome P450aldo (P450aldo, or the CYP11b-2 gene product) and 11b-hydroxylase cytochrome P45011b (P45011b, or the Cyp11b-1 gene product), convert deoxycorticosterone into aldosterone in the zona glomerulosa and into corticoster- one in the zona fasciculata, respectively [7–9]. Thus, the zonal differences in the steroid products are attributable to localization of the two enzymes responsible for the last steps in the steroidogenesis [10].

Adrenal cortices in mammals are composed of morpho- logically and functionally differentiated cell zones [1,2]. The outer zone, the zona glomerulosa, synthesizes aldos- terone, the most potent mineralocorticoid. The middle zone, the zona fasciculata produces corticosterone in rodents and cortisol in humans and other animals. The inner zone, the zona reticularis secretes adrenal androgens in humans and in some other animals. In rodents, aldosterone and corticosterone are produced from a common substrate, deoxycorticosterone. Deoxycorticoster- one is synthesized from cholesterol by a successive action of cholesterol side-chain cleavage enzyme cytochrome P450scc (P450scc, or the Cyp11a gene product), 3b-

Correspondence to K. Mukai, Department of Biochemistry and Integrative Medical Biology, School of Medicine, Keio University, 35 Shinano- machi, Shinjuku-ku, Tokyo 160-8582, Japan. Fax: + 81 3 3358 8138, Tel.: + 81 3 5363 3752, E-mail: mukaik@sc.itc.keio.ac.jp Abbreviations: StAR, steroid acute regulatory protein; Bt2cAMP, dibutyryl cAMP; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; 3bHSD, 3b-hydroxyl steroid dehydrogenase and isomerase; P450, cytochrome P450; P450scc, cholesterol side-chain cleavage P450; P450c21, steroid 21-hydrogenase P450; P45011b, steroid 11b-hydroxylase P450; P450aldo, aldosterone synthase P450; PDL, population doubling levels; SF-1, steroidogenic factor-1; SV40, simian virus 40; ts, temperature-sensitive; HBSS, Hank’s balanced salt solution. Enzymes: glyceraldehyde 3-phosphate dehydrogenase (EC 1.2.1.9); 3b-hydroxyl steroid dehydroxylase and isomerase (EC 1.1.1.145); cytochrome P450 side-chain cleavage (EC 1.14.15.6); cytochrome P450 steroid 21-hydroxylase (EC 1.14.99.10); cytochrome P450 steroid 11b-hydroxylase (EC 1.14.15.4); cytochrome P450 aldosterone synthase (EC 1.14.15.4). *Present address: Faculty of Engineering, The University of Tokushima, Tokushima, Japan. (Received 23 July 2001, revised 22 October 2001, accepted 22 October 2001)

70 K. Mukai et al. (Eur. J. Biochem. 269)

(cid:211) FEBS 2002

Cell culture

I

Recent studies on regulation of adrenocortical steroi- dogenesis have focused on mechanisms of cell-specific transcription of genes encoding the steroid hydroxylases (reviewed in [11]) and steroid acute regulatory protein (StAR) [12]. Among transcription factors that have been found to regulate the steroidogenic genes, it has been demonstrated that steroidogenic factor-1 (SF-1, also referred to as Ad4BP) [13,14] is essential for development of steroidogenic organs such as adrenal cortex and gonads [15,16]. However, molecular mechanisms for development of the adrenocor- tical zonation and its maintenance have not been clarified. We previously showed the presence of a functionally undifferentiated cell layer between the aldosterone-produc- ing zona glomerulosa cells and the corticosterone-producing zona fasciculata cells in rats [17–20]. It was immunohisto- chemically recognized as the region devoid of both P450aldo and P45011b [17–21]. We also showed that the cell layer is composed of the inner half of the zona glomerulosa and the transitional zone (also referred to as zona intermedia) the latter of which has been described by previous investigators [22–29]. We have provided further evidence that the layer locates in the middle of the region containing proliferating cells, suggesting the presence of precursor or progenitor cells that could differentiate into the glomerulosa and/or fasciculata cells [17,30–33]. This view is consistent with recent observation that adrenocortical cells radially arranging from the zona glomerulosa to the zona reticularis share the same clonal origin [34]. However, such processes for development and differentiation of the adrenal cortices have not fully been investigated because precursor or progenitor cell lines remain ambiguous.

A conditionally immortalizing gene such as a temperature- sensitive (ts) large T-antigen gene of simian virus 40 (SV40) has been utilized for generation of cell lines [35,36]. We have previously generated transgenic mice [37] carrying a ts SV40 large T-antigen gene tsA58 [38] that is driven by its own promoter. These transgenic mice have been used to establish various cell lines from different tissues [39,40]. In this study, we have attempted to establish conditionally immortalized adrenocortical cell lines suitable for in vitro analyses of cell differentiation by using the transgenic mice [37].

Ten adrenal glands from 8-week-old male mice and eight adrenal glands from 10-week-old female mice were used in separate experiments. The adrenals were minced and treated with 1.5 mL of Hank’s balanced salt solution (HBSS) containing 2 mgÆmL)1 collagenase type V (Sigma, St Louis, MO, USA), 0.05 mgÆmL)1 DNase (Sigma), and 5 mgÆmL)1 bovine serum albumin (Sigma) at 37 (cid:176)C for 1 h with gentle shaking. After pipetting to disperse cells, they were collected by centrifugation, and were resuspended with HBSS. The suspension contained 1.0 · 106 and 7 · 105 cells from 10 and eight adrenals, respectively. The cells were collected by centrifugation and resuspended at 5 · 105 cells per mL with one of two cell culture media: medium A, a 1 : 1 mixture of Dulbecco’s modified Eagle’s medium and Ham’s F12 medium with 15% heat-inactivated horse serum (Life Technologies, Rockville, MD, USA), 2.5% heat- inactivated fetal bovine serum (Hyclone, Logan, UT, USA), 200 UÆmL)1 penicillin, and 200 lgÆmL)1 streptomycin (Life Technologies); medium E, RITC80-7 medium ([42]; Kyo- kuto Pharmaceutical Industrial, Tokyo, Japan) with the same additives included in medium A. The cell suspension (5 · 104 cells) was placed on the center of a 9.2-cm2 well. The dishes had been coated with bovine fibronectin (Life Technologies) by incubation of multiwell plates overnight at 37 (cid:176)C with serum-free medium containing 1 lgÆmL)1 fibronectin. After incubation of the cells at 37 (cid:176)C for 4 h, 2 mL of medium was added gently into each well. Gas- phase used was humidified atmosphere containing 5% CO2. The next day, the temperature was shifted to 33 (cid:176)C and the medium was changed at 3- to 4-day intervals. The cells were transferred to fibronectin-coated plates every week using 0.05% trypsin-0.53 mM EDTA (Life Technologies). At a third transfer, 25–100 cells were plated in fibronectin-coated 60-mm dishes in the same media. Visual inspection of the plates verified the absence of pairs or groups of cells. After 4 weeks, colonies (3–4 mm in diameter) were isolated using cloning rings and trypsin-EDTA. Each clone was grown successively in 1.8-cm2 wells, 9.2-cm2 wells, and then larger dishes by subculturing. Some cells were used for RNA extraction and others were frozen for subsequent experi- ments.

M A T E R I A L S A N D M E T H O D S

Mice and adrenal glands

Adrenal glands used in this study were excised from the transgenic mice [37] carrying a ts mutant of SV40 large T-antigen gene tsA58 [38]. They were maintained on a standard diet containing 0.3% (w/w) Na and with water ad libitum in accordance with the institutional animal care guidelines of Keio University School of Medicine. The adrenal glands apparently developed no tumor and had normal histology based on examination with hematoxylin/ eosin staining until at least 10 weeks old.

Immunohistochemistry

Immunohistochemical localization of P45011b and P450aldo was performed on 6-lm sections of fresh-frozen adrenal glands from the transgenic mice as previously described [17,41]. The antibodies used were raised in rabbits against rat P45011b and P450aldo [21].

The cell lines obtained at the permissive temperature for the T-antigen were examined for expression of mRNAs for SF-1, P450scc, P45011b, and P450aldo by RT-PCR anal- ysis as described below. SF-1 and P450scc mRNAs were used as adrenocortical cell markers. The reason that we adopted SF-1 and P450scc as criteria for adrenocortical cells was that they were detected in the adrenogenital primordi- um and throughout the cortex in adults [43–46]. P450aldo and P45011b mRNAs were used as the zonal differentiation markers that were responsible for production of aldosterone and corticosterone, respectively [21]. We chose three cell lines AcA101, AcA201 (obtained with medium A), and AcE60 (obtained with medium E). They showed different expression patterns of SF-1, P450scc, and P45011b genes (see Results). Their expression patterns of the steroidogenic genes, morphological appearance, and growth rates were unchanged over population doubling levels (PDL) of 200. To further characterize the cell-lines, cells were treated with porcine corticotropin (23 mUÆmL)1), human angio- tensin II (50 nM), dibutyryl cAMP (Bt2cAMP) (1 mM), KCl

(cid:211) FEBS 2002

Immortalized adrenocortical cell lines (Eur. J. Biochem. 269) 71

(medium + 15 mM), BAY K6844 (1 lM), A23187 (1 lM), ionomycin (1 lM), or 12-O-tetredecanoylphorbol 13 acetate (TPA; 160 nM) for 24 h or 4 days under the standard culture media described above. These reagents were products of Sigma. During the treatments, the cells were cultured at 33 or 39 (cid:176)C. Total RNA was extracted and analysed as described below.

annealing temperatures for each primer pair were: 56 (cid:176)C for P450scc, P45011b, P450aldo, and 3bHSD; 69 (cid:176)C for P450c21; 50 (cid:176)C for StAR; 54 (cid:176)C for glyceraldehyde 3-phos- phate dehydrogenase (GAPDH). PCR products (5 lL) were analyzed by agarose gel electorphoresis followed by visualization with ethidium bromide. Nucleotide sequences of primer pairs used for PCR were as follows (numbers in parenthesis were nucleotide positions of the cDNA sequences):

Northern blot analysis

SCC-F, 5¢-GCACACAACTTGAAGGTACAGGAG-3¢ (1018–1041); SCC-R, 5¢-CAGCCAAAGCCCAAGTACC GGAAG-3¢ (1348–1361) [50].

m11b-F, 5¢-AAGAAAACTTAGAGTCCTGGGATT-3¢ (761–784); m11b-R, 5¢-GTGTCAGTGCTTCCAGCAAT GAGT-3¢ (927–950) [8].

mAldo-F, 5¢-AAGAACATTTCGATGCCTGGG ATG-3¢ (761–784); mAldo-R, 5¢-GTGTCAACGCTCCC AGCGGTGAGC-3¢ (930–953) [8].

mStAR-F,

5¢-AAGAGCTCAACTGGAGAGCAC-3¢ (170–190); mStAR-R, 5¢-TACTTAGCACTTCGTCCC CGT-3¢ (380–400) [51].

3bHSD-F,

5¢-GCAGACCATCCTAGATGTCAAT CTG-3¢ (412–436); 3bHSD-R, 5¢-CAAGTGGCTCATAG CCCAGATCTC-3¢ (1160–1137) [50].

m21-F, 5¢-CTTCACGACTGTGTCCAGGACTTG-3¢ (553–576); m21-R, 5¢-CAGCAGAGTGAAGGCCTGCA GCAG-3¢ (1309–1332) [52].

GAPDH-F, 5¢-TGAAGGTCGGTGTGAACGGATT TGGC-3¢ (51–76); GAPDH-R, 5¢-CATGTAGGCCATGA GGTCCACCAC-3¢ (1010–1033) [53].

corresponding to the

rat

[8,47]; P450aldo,

761–950

The forward and reverse primers reside in different exons of the genes. The PCR products were digested with appropriate restriction enzymes to ensure the specificity of the PCR reactions by comparing of sizes of digests with those expected from published DNA sequences. Total RNA from Y-1 cells was used as a positive control for detection of the mRNAs except that adrenal total RNA from C57BL/6 mice was used as a positive control for detection of P450c21 mRNA.

Total RNA was extracted with a modified single-step isolation method using Trizol reagent (Life Technologies). Northern blot analysis was performed as described previ- ously [47] except that probes were 32P-labeled DNA. Before transfer to positively charged nylon membranes (Roche Diagnostics, Mannheim, Germany) rRNAs were visualized by ethidium bromide. Densitometric analysis of 28S rRNA bands verified that amounts of RNA loaded were similar (< (cid:139) 10%) and that degradation of the RNA prepara- tions was undetectable under the experimental conditions. DNA fragments were labeled with [a-32P]dCTP (3000 CiÆmmol)1, Amersham-Pharmacia Biotech, Piscataway, NJ, USA) and High Prime (Roche Diagnostics) according to the manufacturer’s instructions. Hybridization signals were detected with a Kodak BioMax film with an intensi- fying screen. DNAs used for labeling each contained a fragment as follows: SV40 large T-antigen gene, 1.7-kb PvuII–EcoRI fragment of pMT-1ODtsA [48]; SF-1, AccI–EcoRI fragment corresponding to the 3¢ untransla- ted region of a mouse cDNA [13]; P450scc, a mouse cDNA fragment cDNA nucleotides 1018–1361 [49]; P45011b, a mouse cDNA nucleotides a mouse cDNA nucleotides 761–953 [8]. The plasmids carrying SV40 large T-antigen gene and mouse SF-1 were generous gifts from H. Ariga (Hokkaido University) and K. L. Parker (University of Texas South-Western Medical Center, TX, USA), respectively. cDNAs clones encoding P450scc, P45011b, and P450aldo were obtained by PCR with the primer pairs described below. Mouse adrenocortical Y-1 and fibroblast NIH3T3 cells were cultured with Dulbecco’s modified penicillin Eagle’s medium containing (100 IUÆmL)1), streptomycin sulfate (100 lgÆmL)1) and 10% heat-inactivated fetal bovine serum at 37 (cid:176)C under a humidified atmosphere containing 5% CO2.

RT-PCR

To estimate relative amounts of mRNA among the cells cultured under different conditions (33 or 39 (cid:176)C in the presence or absence of Bt2cAMP), intensities of PCR products stained with ethidium bromide (see below) were determined by densitometric analysis. All experiments for the determination were performed within the exponential phase of the amplification reactions to obtain the linear response concerning the initial RNA amounts. Each experiment was performed at least twice to assure the reproducibility. The intensities were normalized with GAPDH cDNA, and the relative amounts of mRNA were expressed in Table 1 as the values of the mRNA level in Y-1 cells or mouse adrenal glands were taken as 1.0.

Analysis of steroids

Cells (1–2 · 106 cells per 21-cm2 dish) were cultured at 33 or 39 (cid:176)C in the presence or absence of 1 mM Bt2cAMP for (20 lM; Sigma) was 4 days. Water-soluble cholesterol added at 24 h before removal of the medium. Steroids in the medium (2 mL) were extracted with 8 mL of dichlo- romethane. The extracts were treated with 2 mL of 0.1 M NaOH and then washed with 2 mL of water. The resulting extracts were evaporated to dryness and redissolved with

Expression of mRNA was analyzed with RT-PCR. cDNA was synthesized from total RNA (2 lg) with an oligo dT18 primer and Moloney murine leukemia virus reverse transcriptase using a first-strand cDNA synthesis kit (Amersham-Pharmacia Biotech) according to the manufac- turer’s instructions. Aliquots (1 lL) of the reaction solution were used as template for PCR. PCR mixture contained 10 mM Tris/HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, 0.2 mM each deoxynucleotide triphosphate, 0.5 lM deoxy- oligonucleotide primers, and Taq DNA polymerase (1.25 U, Takara Shuzo, Shiga, Japan) in a total volume of 25 lL. Amplification conditions were 45 s at 94 (cid:176)C, 45 s at the annealing temperature for each primer pair as described below, and 2 min at 72 (cid:176)C for 35 cycles or appropriate cycle numbers as indicated followed by 7 min at 72 (cid:176)C. The

72 K. Mukai et al. (Eur. J. Biochem. 269)

(cid:211) FEBS 2002

Table 1. Expression of genes involved in adrenocortical steroidogenesis in cell lines AcA101, AcA201, and AcE60. Based on the results from RT-PCR, relative levels of mRNA were normalized using the results of GAPDH mRNA as described in Materials and methods, and are expressed as the mRNA levels in Y-1 cells or mouse adrenal glands were taken as 1.0. Value 0 indicates that the level was < 0.01 of that of Y-1 cells or mouse adrenal glands. ND, not determined. Bt2cAMP presence or absence is indicated by + and –, respectively.

AcA101 AcA201 AcE60

33 (cid:176)C 33 (cid:176)C 39 (cid:176)C 33 (cid:176)C 39 (cid:176)C 39 (cid:176)C

– + – + – + – + – + – + Y-1 Adrenal glands mRNA

P450scc StAR 3bHSD P450c21 P45011b P450aldo 0.04 0.11 0 0 0 0 0.12 0.55 0 0.18 0 0 0.04 0.35 0 0 0 0 0.30 1.7 1.4 0.34 1.4 0 0.25 0.16 0 0 0.10 0 0.42 1.25 0 0.06 0.15 0 0.71 0.85 0 0 0.62 0 1.1 1.5 0.31 0.12 1.85 0 0.01 0.33 0.02 0 0 0 0.06 0.65 0.13 0.14 0 0 0.02 0.31 0.01 0 0.16 0 0.09 1.9 0.07 0 0.19 0 1.0 1.0 1.0 ND 1.0 1.0 ND ND ND 1.0 ND ND

R E S U L T S

Histology of adrenal glands of transgenic mice carrying a temperature-sensitive oncogene

Adrenal glands of the transgenic mice harboring SV40 large T-antigen tsA58 gene appeared quite normal in size and shape as compared with those of nontransgenic animals, suggesting that the ts oncogene developed no tumor in the adrenal glands at body temperature. As judged by the haematoxylin/eosin staining, zonation of their cortices including the zonae glomerulosa, fasciculata, and reticularis were indistinguishable from those of the normal animals (data not shown). The medullary tissues also appeared to be normal.

We then examined imminohistochemically expression of steroidogenic enzymes that occur in a zone-specific manner, namely, P45011b and P450aldo. As shown in Fig. 1,

50 lL of methanol, and then 50 lL of water was added. An aliquot (25 lL) of each sample was subjected to HPLC using a C18 column (4.6 mm · 150 mm; Cosmosil 5C18- AR, Nacalai Tesque, Kyoto, Japan). Steroids were sepa- rated by isocratic elution with 65% methanol in water at 0.8 mLÆmin)1 and detected at 254 nm. For detection of corticosterone, 55% methanol was used as the eluent. Authentic steroid standards were used for identification of steroid products by comparing elution times. To convert pregnenolone, which is hardly detectable at 254 nm, into progesterone, the steroid products were treated with 0.53 U of cholesterol oxidase (26.8 UÆmg)1; Toyo Jozo Co., Ltd, Shizuoka, Japan) [54] in a reaction mixture of 100 lL consisting of 20 mM potassium phosphate buffer, pH 7.4, and 0.3% (v/v) Tween 20. The reaction mixture was incubated at 37 (cid:176)C for 20 min, and extracted with dichlo- methane. The extracts were analyzed by HPLC under the same conditions.

Fig. 1. Adrenocortical zonation of transgenic mice harboring a temperature-sensitive SV40 large T-antigen gene. Fresh-frozen sections (6 lm) from adrenal glands of the transgenic mice harboring a temperature-sensitive (ts) SV40 large T-antigen gene were analyzed immunohistochemically with an antibody specific to corticosterone-synthesizing 11b-hydroxylase cytochrome P450 (P45011b) (A) and with an antibody to aldosterone synthase cytochrome P450 (P450aldo) (B) as described in Materials and methods. Localization of P45011b is shown with a brown color and that of P450aldo is shown with a blue color. These immunohistochemical results obtained with the transgenic mice were indistinguishable from those with nontransgenic normal mice. Sizes, shapes and cytology of the adrenal glands (including medulla) of the transgenic mice also seemed to be normal. Note that the thickness (marked with a) where P45011b is absent is larger than the thickness (marked with b) where signals of P450aldo are present, indicating that there is a cell-layer which neither has P45011b nor P450aldo. Bar (cid:136) 50 lm.

(cid:211) FEBS 2002

Immortalized adrenocortical cell lines (Eur. J. Biochem. 269) 73

P45011b was detected in the entire regions of the zonae fasculata-reticularis (Fig. 1A), while P450aldo was detected in the outermost cells of the zona glomerulosa (Fig. 1B). Such distribution was indistinguishable from that observed with nontransgenic animals (data not shown). Thus, the distribution of the two enzymes was not affected by introduction of SV40 large T-antigen gene tsA58. These observations suggest that the transgenic manipulation does not interfere with development of the adrenal zonation in the mice.

As indicated in Fig. 1, the region of P45011b-negative cells (parenthesis in Fig. 1A) was thicker than the region of P450aldo-positive cells (parenthesis in Fig. 1B). Evidently, there was a cell-layer where neither P45011b nor P450aldo were detectable, suggesting that the cells in this layer were unable to produce either corticosterone or aldosterone. It was also noted that P450aldo was only detectable in the outermost area of the zona glomerulosa under dietary conditions with normal Na contents. Together with our previous results [17], these results indicate that mice exhibit a functionally undifferentiated cell layer analogous to that observed in rats.

Establishment of immortalized adrenocortical cell lines

A number of cell lines were derived from primary cells prepared from whole adrenal glands of the transgenic mice. The permissive temperature (33 (cid:176)C) for the T-antigen mutant was used to establish cell lines in which the oncoprotein was kept active. To select cell lines exhibiting properties of adrenocortical cells, RNAs from the cells were examined by RT-PCR analysis to detect SF-1 and P450scc mRNAs. The cell lines were further examined for detection of P45011b and P450aldo mRNAs, the functional markers for the zone-specific differentiation of the cells. The results of RT-PCR analyses indicated that the cell lines were categorized into three different groups. The first group constituted cell lines expressing SF-1, P450scc, and P45011b mRNAs but not P450aldo mRNA. These cell lines had the property of the zona fasciculata cells. The second group was composed of a small number of the cell lines that expressed SF-1 and P450scc mRNAs but not P45011b or P450aldo mRNAs, showing the gene expression pattern observed in the undifferentiated cell layer. The last cell lines were those that expressed none of the SF-1, P450scc, P45011b, and P450aldo mRNAs and were characterized by their fibro- blast-like appearance. There were no cell lines that expressed P450aldo mRNA regardless of expression of SF-1, P450scc, or P45011b mRNAs.

cell-lines. Phase

flatter appearance and were less retracted than AcA101 and AcA201 cells. The doubling time of these cells was 24–30 h at 33 (cid:176)C.

Among these cell lines, AcA101, AcA201, and AcE60 were chosen for further detailed characterization. When cultured at 33 (cid:176)C, AcA201 was one of cell lines that displayed mRNAs for SF-1, P450scc, and P45011b but not P450aldo mRNA. On the other hand, AcA101 and AcE60 were two different cell lines that displayed expression of SF-1 and P450scc mRNAs but not P45011b and P450aldo mRNAs at 33 (cid:176)C; the latter two exhibited distinct expres- sion patterns of steroidogenic genes that were different to each other (see below). Their phenotypes and growth rates of the three cell lines were unchanged over a PDL of 200 at 33 (cid:176)C. Morphologies of these cells cultured at 33 (cid:176)C are shown in Fig. 2. AcA101 and AcA201 cells displayed retracted appearances. AcE60 cells showed a larger and

Fig. 2. Morphology of adrenocortical contrast Photomicrographs depict morphologies of adrenocortical cell-lines AcA101 (A), AcA201 (B), and AcE60 (C) which were cultured under the permissive temperature (33 (cid:176)C) for the ts SV40 large T-antigen. The cells were cultured at subconfluent stages under the conditions as described in Materials and methods. Bar (cid:136) 50 lm.

74 K. Mukai et al. (Eur. J. Biochem. 269)

(cid:211) FEBS 2002

Growth of AcA101, AcA201, and AcE60 cells under a nonpermissive temperature for the T-antigen (39 (cid:176)C) was examined upon shifting the temperature from 33 to 39 (cid:176)C. Their rates of growth were reduced within 2 days after the start of the temperature shift. At one week, AcA101 and AcA201 cells were mostly detached from the culture dish surface, and completely lost adhesivity until 4 weeks. In contrast, AcE60 cells were able to attach onto the dishes even after the disappearance of cell division (data not shown).

The three cell lines cultured at 33 (cid:176)C were characterized by Northern blot analysis. As shown in Fig. 3A, Northern

blotting for SV40 T-antigen mRNA indicated that AcA101 (lane 1), AcA201 (lane 2), and AcE60 (lane 3) cells expressed the transgene mRNA with different signal intensities. As expected, Y-1 adrenocortical cells (lane 4) and fibroblast NIH3T3 cells (lane 5) gave no hybridization signal. SF-1 mRNA (Fig. 3B) was detectable in the established cell lines (lanes 1–3). The electrophoretic mobility of the hybridiza- tion signal was the same as that of Y-1 cells (lane 4). Signal intensities of P450scc mRNA varied markedly among the three cell lines (Fig. 3C). The P450scc mRNA level in AcA201 cells was evident, while only a faint signal was detectable in AcA101 cells. P450scc mRNA in AcE60 cells was undetectable under the current experimental conditions. As shown in Fig. 3B,C, Y-1 cells (lanes 4) produced hybridization signals of SF-1 and P450scc mRNA, whereas NIH3T3 cells (lanes 5) did not, indicating that the hybrid- ization was specific. As seen in Fig. 3D, electrophoretic patterns of ribosomal RNAs indicated that amounts of total RNA subjected to the Northern analyses were equivalent and its degradation was undetectable. Although attempting to detect P45011b and P450aldo mRNAs using the same RNA blot, we could not detect these mRNAs under the current experimental conditions (data not shown).

Figure 4 shows the results from RT-PCR analysis with greater sensitivity for the detection of P450scc, P45011b, and P450aldo mRNAs in the three cell lines cultured at 33 (cid:176)C. In addition to AcA101 (Fig. 4A, lane 1) and AcA201 (lane 2), AcE60 cells (lane 3) exhibited a detectable level of P450scc mRNA. Differences in intensities of the amplified DNA fragments among the three were consistent with the results from Northern blotting (Fig. 3C). On the other hand, P45011b mRNA (Fig. 4B) was detectable in AcA201 cells (lane 2), but not in AcA101 (lane 1) or in AcE60 (lane 3) cells. P450aldo mRNA (Fig. 4C) was not detectable in the three cell lines (lanes 1–3), although it was detected in Y- 1 cells (lane 4). Digestion of the amplified DNA fragments with restriction enzymes verified specificity of PCR (right panels of Fig. 4A–C). As judged from the results for GAPDH mRNA, the efficiency of RT-PCR was compara- ble among the cell lines (Fig. 4D). These results suggest that, under conditions where the T-antigen is active, AcA201 cells have the property of the zona fasciculata cells, while AcA101 and AcE60 cells do not display the zone-specific markers of steroidogenesis.

Cyclic AMP-dependent alterations in steroidogenic gene expression upon inactivation of the T-antigen

To examine effects of inactivation of the T-antigen on expression of the genes for steroidogenesis, we cultured AcA101, AcA201, and AcE60 cells at 39 (cid:176)C for up to 4 days and analyzed levels of mRNAs for P450scc, StAR, 3bHSD, P450c21, P45011b, and P450aldo by RT-PCR. At the same time, the cells were cultured in the presence of regulators of the steroidogenic gene expression such as Bt2cAMP, ACTH, or angiotensin II, in combination with the temper- ature shift. As described below, a simple shift of temperature for 4 days affected the mRNA levels of some of these steroidogenic genes. Treatments with either corticotropin or angiotgensin II did not alter the mRNA levels significantly at both 33 and 39 (cid:176)C under the current experimental conditions (data not shown). On the other hand, treatment with Bt2cAMP for 4 days turned out to alter the mRNA

Fig. 3. Northern analysis for expression of SV40 large T-antigen, steroidogenic factor-1 (SF-1, or Ad4BP), cholesterol side-chain cleavage cytochrome P450 (P450scc, or Cyp11a) in cell lines AcA101, AcA201, and AcE60 at a permissive temperature for the ts T-antigen. Total RNA was prepared from AcA101, AcA201, and AcE60 cells which were cultured at 33 (cid:176)C with standard media as described in Materials and methods. RNA from mouse adrenocortical Y-1 and fibroblast NIH3T3 cells, neither of which express the T-antigen gene, were used as positive and negative controls, respectively, for detection of adrenocortical cell marker mRNAs. Total RNA (10 lg per lane) was analyzed by Northern blotting with 32P-labeled cDNA probes enco- ding (A) SV40 T-antigen (B) SF-1 and (C) P450scc genes as described in Materials and methods. Ribosomal RNAs (D) visualized by ethi- dium bromide show that amounts of total RNAs were comparable to each other and that degradation of RNA were undetectable under the experimental conditions.

(cid:211) FEBS 2002

Immortalized adrenocortical cell lines (Eur. J. Biochem. 269) 75

levels at both temperatures as described below. These results were summarized in Table 1.

P450sccand StARgenes

increased significantly. Responsiveness of AcA101 (lanes 2 and 4) and AcE60 (lanes 12 and 14) cells to Bt2cAMP was evident at both 33 and 39 (cid:176)C, while in AcA201 cells (lanes 7 and 9) responsiveness was much less. Synergistic effects of a temperature shift and Bt2cAMP were also shown in these cell lines (lanes 4, 9, and 14).

Figure 5A shows effects of a temperature shift in the presence or absence of Bt2cAMP on P450scc mRNA levels in AcA101, AcA201, and AcE60 cells. P450scc mRNA levels upon a simple shift to 39 (cid:176)C were almost unchanged in AcA101 (lanes 1 and 3) and AcE60 (lanes 11 and 13) cells, while in AcA201 cells (lanes 6 and 8), mRNA levels

We next examined the ability of AcA101, AcA201 and AcE60 cells to express StAR mRNA, a key factor for the acute induction of adrenocortical steroidogenesis (Fig. 5B). In the absence of the cAMP analog at 33 (cid:176)C, the mRNA levels were detectable but with only faint bands in the three

Fig. 4. RT-PCR analysis for expression of P450scc, P45011b, and P450aldo genes in the cell lines AcA101, AcA201, and AcE60 cells cultured at 33 (cid:176)C. cDNA was synthesized with oligo dT primer using total RNA (2 lg) from AcA101 (lanes 1 and 5), AcA201 (lanes 2 and 6), AcE60 (lanes 3 and 7), or Y-1 (lanes 4 and 8) cells, and the resulting cDNAs were amplified by PCR using specific primer pairs for (A) P450scc, (B) P45011b, (C) P450aldo and (D) glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as described in Materials and methods. Cycle numbers in PCR were 35 in (A), (B), and (C), and 25 in (D). Left, PCR products (5 lL) were analyzed through 1% agarose gels. Right, PCR products (5 lL except for lane 7 of panel A where 10 lL was used) were digested with BstXI (A,B) and with SacI (C). The digests were separated on 8% polyacrylamide gels. DNA fragments were visualized by ethidium bromide. Size marker (lanes M) is HaeIII-digested /X174 DNA. Numbers with arrowheads indicate sizes of PCR products or their digests. ns (C) indicates a nonspecific band. Amplifying conditions except for lanes 2 and 4 of (A) were within the exponential phase of the reactions to obtain the linear dose–response concerning the initial RNA amounts.

76 K. Mukai et al. (Eur. J. Biochem. 269)

(cid:211) FEBS 2002

although the level after the treatment was lower (lanes 6–9). Thus, synergies of the T-antigen and inactivation of treatment with the cAMP analog were evident in the induction of 3bHSD mRNA in AcA101 and AcA201 cells. The mRNA level in AcE60 cells in the absence of Bt2cAMP at 33 (cid:176)C was hardly detected (lane 11), while treatment with the cAMP analog at 33 (cid:176)C induced the mRNA level (lane 12). A temperature shift to 39 (cid:176)C did not enhance the 3bHSD mRNA level (lane 13), and the induction by the cAMP analog at 39 (cid:176)C was smaller than that observed at 33 (cid:176)C (lane 14). It should be noted that, although being expressed in adrenal cortex and gonads (as are the P450scc and StAR genes), the 3bHSD gene of these cell lines is expressed in a manner (Fig. 6A) distinct from those of P450scc and StAR gene expression (Fig. 5A,B).

cell lines (lanes 1, 6, and 11). When AcA101 and AcA201 cells were cultured either in the presence of Bt2cAMP (lanes 2 and 7) or at 39 (cid:176)C (lanes 3 and 8), the mRNA levels were increased markedly. In contrast, AcE60 cells displayed only a small induction of the mRNA level in the presence of Bt2cAMP (lane 12), while showing no notable change upon a temperature shift (lane 13). Interestingly, however, AcE60 cells exhibited the greatest synergy between the temperature shift and Bt2cAMP among the three cell lines tested (lanes 4, 9, and 14). On the other hand, the results obtained with GAPDH primers showed similar intensities of the PCR products among the groups, indicating that efficiencies of reverse transcription and PCR were comparable to one another (Fig. 5C). These results indicated that P450scc and StAR genes responsible for the initial steps for synthesis of both adrenocortical and sex steroid hormones were expressed constitutively and inducible through a cAMP- dependent pathway in the established cell lines.

3bHSDand P450c21genes

Figure 6B illustrates differences in P450c21 mRNA levels among the cell lines. The mRNA level in AcA101 cells in the absence of Bt2cAMP was undetectable at 33 (cid:176)C (lane 1), whereas treatment with the cAMP analog induced a weak signal at 33 (cid:176)C (lane 2). A simple shift to 39 (cid:176)C did not induce the mRNA (lane 3), but the same procedure in the presence of Bt2cAMP increased the level significantly (lane 4). The results obtained with AcA201 cells (lanes 6–9) were similar to those obtained with AcA101 cells except that the induction in AcA201 was weak. In AcE60 cells, P450c21 mRNA was detectable upon treatment with Bt2cAMP at 33 (cid:176)C (lanes 11 and 12). At 39 (cid:176)C, P450c21 mRNA became undetectable irrespective of the presence of Bt2cAMP (lanes 13 and 14). Thus, AcA101, AcA201 and AcE60 cells were

3bHSD and P450c21 catalyze the reactions in the middle of the synthetic pathways for both corticosterone and aldos- terone. Figure 6A shows mRNA levels of 3bHSD in the same sets of RNA preparations used for analysis of P450scc and StAR mRNA levels in Fig. 5. 3bHSD mRNA in AcA101 cells was detected only in the presence of Bt2cAMP at 39 (cid:176)C (lanes 1–4). AcA201 cells also expressed 3bHSD mRNA in a similar manner to that seen in AcA101 cells,

Fig. 5. Expression of P450scc and steroidogenic acute regulatory protein (StAR) genes in AcA101, AcA201, and AcE60 cells upon a temperature shift and/or treatment with dibutyryl cAMP (Bt2cAMP). AcA101, AcA201, and AcE60 cells were plated at 33 (cid:176)C, and allowed to attach to dishes for 24 h, and were further cultured at 33 or 39 (cid:176)C for 4 days in the absence or presence of 1 mM Bt2cAMP. Total RNA was prepared from the cells and was subjected to RT-PCR analysis using the primer pairs for (A) P450scc, (B) StAR and (C) GAPDH as described in Materials and methods. RNA preparation from Y-1 cells were used as positive controls for P450scc and StAR mRNA. PCR products (5 lL) were analyzed through 2% agarose gels followed by visualization with ethidium bromide. Cycle numbers in PCR for P450scc (A) were 32 in AcA101, 28 in AcA201, and 35 in AcE60 cells, and those in PCR for StAR (B) and GAPDH (C) were 35 and 25, respectively, for the three cell lines.

(cid:211) FEBS 2002

Immortalized adrenocortical cell lines (Eur. J. Biochem. 269) 77

able to express the P450c21 gene, the marker expressed exclusively in the entire regions of adrenal cortices but not in gonads, indicating that the cell lines retain a feature of adrenocortical cells.

P45011b and P450aldogenes

Fig. 4B, lane 2), and was not increased by the Bt2cAMP treatment at 33 (cid:176)C (lane 7 of Fig. 7). A simple shift to 39 (cid:176)C enhanced the mRNA level (lane 8), and the additive effects of the cAMP analog appeared to be small, if any (lane 9). AcE60 cells did not exhibit a detectable level of P45011b mRNA either in the absence (lane 11) or presence (lane 12) of Bt2cAMP at 33 (cid:176)C. Interestingly, AcE60 cells showed the ability to induce the mRNA level upon the temperature shift to 39 (cid:176)C (lane 13), although the level after the Bt2cAMP treatment remained almost unchanged at 39 (cid:176)C (lane 14). These results indicated that the three cell lines are able to express P45011b gene, which is a determinant for synthesis of corticosterone in the zona fasciculata cells. Moreover, these three cell lines responded to distinct stimulatory conditions to express the P45011b gene.

Figure 7 shows the levels of P45011b mRNA in the absence or presence of Bt2cAMP at 33 and 39 (cid:176)C. AcA101 cells did not have a detectable level of P45011b mRNA in the absence of the cAMP analog at 33 (cid:176)C (Fig. 7, lane 1; Fig. 4B, lane 1). Either treatment with the cAMP analog (Fig. 7, lane 2) or a temperature shift (lane 3) did not induce the mRNA. Upon the combined treatment with the cAMP analog and a temperature shift, however, the cells expressed the P45011b gene (lane 4). A similar synergistic effect in AcA101 cells was observed on the levels of 3bHSD mRNA (Fig. 6A). The in AcA201 cells at 33 (cid:176)C was P45011b mRNA level lane 6; detectable in the absence of Bt2cAMP (Fig. 7,

As mentioned earlier in this article, P450aldo mRNA was undetectable in AcA101, AcA201, and AcE60 cells at 33 (cid:176)C cultured in the absence of Bt2cAMP at 33 (cid:176)C. In addition, a temperature shift and/or treatment with Bt2cAMP failed to

Fig. 6. Expression of 3b-hydroxysteroid dehydrogenase (3bHSD) and 21-hydroxylase P450 (P450c21) upon a temperature shift and/or treatment with Bt2cAMP. PCR was performed with primer pairs for (A) 3bHSD and (B) P450c21 using the cDNAs synthesized in the experiments in Fig. 5. Y-1 cells and mouse adrenal glands were used as positive controls for detection of 3bHSD and P450c21 mRNA, respectively. PCR products (5 lL) were analyzed through 1% agarose gels followed by visualization with ethidium bromide. Cycle numbers in PCR for 3bHSD and P450c21 were 35 for the three cell lines.

Fig. 7. Expression of P45011b gene upon a temperature shift and/or treatment with Bt2cAMP. PCR was performed with the primer pair for P45011b using the cDNAs synthesized in the experiments in Fig. 5. Y-1 cells were used as a positive control for detection of P45011b mRNA. PCR products (5 lL) were analyzed through 2% gels followed by visualization with ethidium bromide. Cycle numbers in PCR were 35 for the three cell lines.

78 K. Mukai et al. (Eur. J. Biochem. 269)

(cid:211) FEBS 2002

induce P450aldo mRNA (data not shown). Furthermore, to induce expression of the P450aldo gene, these cells were cultured in the presence of either of angiotensin II, KCl, or reagents that could change intracellular concentration of calcium ion at 33 or 39 (cid:176)C (see Materials and methods). These stimuli, however, could not induce detectable levels of P450aldo mRNA in the cell lines under current experimen- tal conditions (data not shown).

Steroidogenesis of the cell lines

genesis to activation of protein kinase A. Although pregn- enolone is a possible intermediate of the steroidogenesis, it is hardly undetectable in the eluate of HPLC by monitoring at 254 nm. To examine amounts of pregnenolone in the culture medium of the three cell lines, the dichloromethane extracts were treated with cholesterol oxidase for conversion into progesterone. After treatment, a remarkable peak of progesterone was detected in the culture medium of AcA201 (Fig. 8F), suggesting that AcA201 cells secreted large amounts of pregnenolone. Similarly, AcA101 and AcE60 cells synthesized significant amounts of pregnenolone when they were cultured in the presence of Bt2cAMP at 39 (cid:176)C (data not shown). These results were consistent with the results from RT-PCR analysis of mRNA levels of the steroidogenic genes (Figs 5–7).

D I S C U S S I O N

The ability of the cell lines to produce steroids was examined by using reversed phase HPLC. As shown in Fig. 8A–D, when AcA101 cells were cultured in the presence of Bt2cAMP for 4 days at 39 (cid:176)C, progesterone and a small amount of deoxycorticosterone were detected, while corti- costerone was undetectable. On the other hand, AcA201 cells synthesized only a detectable amount of progesterone when the cells were stimulated with Bt2cAMP at 39 (cid:176)C (Fig. 8E). AcE60 cells secreted progesterone in the presence of Bt2cAMP at both 33 and 39 (cid:176)C (data not shown). Thus, the three cell lines showed responsiveness of their steroido-

Previous studies on regulatory mechanisms of adrenocor- tical steroidogenesis have often utilized mouse Y-1 [55] and human NCI-H295 [56] cells. Y-1 cells have been known to display expression patterns of steroidogenic genes analo- gous to those of the zona fasciculata cells except that the endogenous P450c21 gene is not expressed, and that the P450aldo gene is constitutively expressed though its mRNA level is one-tenth of that of P45011b mRNA [8]. On the other hand, NCI-H295 cells have been reported to exhibit the phenotypes of both zona glomerulosa and fasciculata cells simultaneously. Although these cell lines have been in vitro cell culture systems, the shown to be useful phenotypes of Y1 and NCI-H295 cells do not precisely correspond to either of the glomerulosa or fasciculata cells. Other adrenocortical cell lines were also established by using the wild-type SV40 T-antigen gene [57,58]. When compared with the cell lines established previously, our cell lines described in the present study have several distinct features suitable for studies on differentiation of adreno- cortical cells and regulation of the steroidogenic genes. First, immortalization is conditional so that activity of the oncogene can be removed. Secondly, multiple cell lines with the identical genetic background exhibit different pheno- types in steroidogenic gene expression from one another. Finally, and most importantly, established AcA101 and lines have the ability to convert from an AcE60 cell undifferentiated stage into the differentiated one analogous to zona fasciculata-like cells.

Because the ts T-antigen transgene did not affect cytogenesis and zonal differentiation of the adrenocortical cells of the mice, the adrenocortical cells in vivo were likely to be in a normal pathway of their differentiation. It is unknown whether the ts T-antigen is active in the adrenals in vivo. However, we have previously noted that amounts of the ts T-antigen protein in other cells, which were obtained by transfection, were decreased markedly at 37 (cid:176)C when compared with amounts of wild-type T-antigen at 37 (cid:176)C. It is presumable that levels of the ts T-antigen protein are very low at body temperature. At 33 (cid:176)C, on the other hand, the capability of the established cell lines to grow over PDL 200, appeared to be a result of T-antigen activation. Being consistent with the view, inactivation of the T-antigen by culturing at the nonpermissive temperature reduced their growth rates. Thus, the cell lines established under the permissive conditions for the T-antigen could be returned to

Fig. 8. Analysis of steroid production. Cells were cultured at 33 or 39 (cid:176)C in the absence or presence of 1 mM Bt2cAMP for 4 days. A water-soluble form of cholesterol (20 lM) was added at 24 h before removal of the medium. Dichloromethane extracts of the incubated medium were prepared and were analyzed by reversed phase HPLC as described in Materials and methods. AcA101 cells were cultured at 33 (cid:176)C in the absence (A) or presence (B) of Bt2cAMP and at 39 (cid:176)C in the absence (C) or presence (D) of Bt2cAMP. Dichloromethane extract of the medium of AcA201 cells, which were incubated at 39 (cid:176)C for 4 days in the presence of 1 mM Bt2cAMP (E), was treated with cho- lesterol oxidase to convert pregnenolone into progesterone (F), and was analyzed by HPLC. Peaks other than those corresponding to deoxycorticosterone and progesterone are unidentified compounds which are also extracted from medium without incubation of the cells. D, deoxycorticosterone. P, progesterone.

(cid:211) FEBS 2002

Immortalized adrenocortical cell lines (Eur. J. Biochem. 269) 79

expressing cells in vivo receive specific signals from the capsule.

the normal pathway of their differentiation after a temper- ature shift.

The present results showing that the cell lines responded to Bt2cAMP but not to ACTH raise a question whether the cell lines are able to express the MC2-receptor. Although we have examined expression of MC2-receptor mRNA by RT- PCR analysis using two different PCR primer sets [63,64], no specific signal was obtained from RNA of the three cell lines under our experimental conditions, while Y-1 cells gave specific signals (not shown). These observations suggested that the cells were not expressing MC2-receptor mRNA. Absence of the receptor might be a possible explanation for the response to cAMP analog but not to ACTH.

The three cell lines were able to synthesize large amounts of pregnenolone, and detectable levels of progesterone. AcA101 cells were likely to produce small amounts of deoxycorticosterone. The ability to produce the steroids together with enhancement of the synthesis by treatment with a cAMP analog were consistent with the expression levels of mRNA molecules of the steroidogenic genes, which were regulated through the protein kinase A pathway. Accumulation of pregnenolone and low levels of progester- one, deoxycorticosterone, and corticosterone suggested that P450scc, including the electron donor system, was active and that the following synthesizing steps such as that catalyzed by 3bHSD were not fully functional.

The three cell lines characterized in the present study retained a common phenotype of steroidogenic cells, i.e. expression of SF-1, P450scc, and StAR genes. These genes are known to be expressed throughout adrenal cortices as well as in the steroidogenic cells in gonads [43,46,59]. Although 3bHSD [6] and P450c21 [5] genes are also expressed throughout rodent adrenal cortices including the undifferentiated cell layer, their mRNAs in the three cell lines were undetectable or at very low levels at 33 (cid:176)C under the standard culture conditions. This suggests that the cell lines were dedifferentiated through a generation of cell lines. Previous studies, however, showed that the 3bHSD gene was expressed in nonsteroidogenic tissues such as liver and kidney [60] in addition to the classical steroidogenic tissues [6], suggesting that regulation of the 3bHSD gene is different from that of the former three genes. On the other hand, expression of P450c21 gene was undetectable in Y-1 and other mouse adrenocortical cell line [58], implying that the gene tends to be repressed by unknown mechanisms in vitro. Nevertheless, the present results that our cell lines showed inducible expression of 3bHSD and P450c21 genes indicate that the cell lines have the property of adrenocortical cells. Especially, expression of the endogenous P450c21 gene in lines contrasts strikingly with the the established cell characteristics of Y-1 and other mouse adrenocortical cell lines [58].

Based on the results from analysis for expression of the zonal markers in the established cell lines, AcA201 cells displayed the feature of the zona fasciculata cells at 33 (cid:176)C, while AcE60 and AcA101 cells were at undifferentiated stages that have a similar pattern of gene expression to the undifferentiated cell layer. Furthermore, differences in induction of P45011b gene between AcE60 and AcA101 were demonstrated and were more evident than the differences in the other steroidogenic genes; the P45011b gene in AcE60 cells is expressed after inactivation of the T- antigen, while the gene in AcA101 cells requires additional stimulation by cAMP for expression. It is noted that the same procedure for generating cell lines allowed us to establish the three different cell lines. As whole adrenal glands were used as the source for the cell culture in the present study, it was unknown whether the cell lines AcE60 and AcA101 were derived from cells in the functionally undifferentiated cell layer. Nevertheless, several lines of experimental evidence conceivably support a notion that AcE60 and AcA101 cells were derived from the undiffer- entiated cell layer.

Adrenal cortices in rats have been known to regenerate in vivo from the capsular portion (capsule and adherent cells) of the glands after enucleation surgery [1,2]. The capsular and decapsular portions were found to be separated by enucleation at the boundary between the zona glomerulosa and zona fasciculata (F. Mitani, unpublished observation). In the early days after enucleation, it was demonstrated that the cells of the capsular portion dedif- ferentiate into those devoid of both P450aldo and P45011b, proliferate, and then differentiate into morphologically and functionally distinct cells, i.e. the glomerulosa cells and fasciculata cells [62,65]. Thus, the dedifferentiated cells seem to be a bipotential stem cell population [32,33]. Considering that the dedifferentiated cells in the regenerating adrenal cortex express SF-1 and P450scc, but not P45011b nor P450aldo [65,66], such a steroidogenic phenotype was the same as those of AcA101 and AcE60 cells cultured at 33 (cid:176)C. Based on the present results that AcA101 and AcE60 cells were able to differentiate into the fasciculata-like cells, they may be referred to as precursor cell lines for the zona fasciculata cells. The available evidence does not support the hypothesis that these cell lines represent bipotential progen- itor cell lines or stem cell-like lines. Further studies are obviously necessary to examine whether AcA101 and AcE60 cells are able to differentiate into cells with the property of the zona glomerulosa cells.

A C K N O W L E D G E M E N T S

Although a number of cell lines were obtained in the present study, there was no cell line expressing the P450aldo gene. During initial passages, however, we found that the cells from the transgenic mice were able to express the P450aldo gene (data not shown). Being consistent with these observations, previous studies [61] have also reported that the zona glomerulosa cells prepared from rat adrenal glands gradually cease to produce aldosterone in culture. These phenomena may imply that additional factors are required for expression of the P450aldo gene. In the zona glomerul- osa cells in vivo, the cells expressing P450aldo were adjacent to the capsule, as reported previously by other studies [62] and ours [17,20,30]. It is possible that the P450aldo-

We thank Drs K. L. Parker and H. Ariga for generous gifts of plasmids and Drs T. Ogishima and P. J. Hornsby for helpful suggestions for analysis of steroids. This work was supported by Grants-in-aid for General Scientific Research from the Ministry of Education, Science and Culture of Japan, by National Grants-in-Aid for the Establishment of High-Tech Research Center in a Private University, and by grants from Uehara Memorial Foundation, the Ichiro Kanehara Foundation, and from Keio University School of Medicine.

80 K. Mukai et al. (Eur. J. Biochem. 269)

(cid:211) FEBS 2002

R E F E R E N C E S

20. Mitani, F., Mukai, K., Ogawa, T., Miyamoto, H. & Ishimura, Y. (1997) Expression of cytochromes P450aldo and P45011b in rat adrenal gland during late gestational and neonatal stages. Steroids 62, 57–61.

1. Orth, D.N., Kovacs, W.J. & DeBold, C.R. (1992) The adrenal cortex. In Williams Textbook of Endocrinology (Wilson, J.D. & Foster, D.W., eds), pp. 489–619. Sanders, Philadelphia, PA. 2. Hornsby, P.J. (1985) The regulation of adrenocortical function by control of growth and sturucture. In The Adrenal Cortex (Anderson, D.C. & Winter, J.S.D., eds), pp. 1–31. Butterworths Publishers Ltd, London, UK. 21. Ogishima, T., Suzuki, H., Hata, J., Mitani, F. & Ishimura, Y. (1992) Zone-specific expression of aldosterone synthase cyto- chrome P-450 and cytochrome P-45011b in rat adrenal cortex: histochemical basis for the functional zonation. Endocrinology 130, 2971–2977. 3. Miller, W.L. (1988) Molecular biology of steroid hormone syn- thesis. Endocrinol. Rev. 9, 295–318. 22. Cain, A.J. & Harrison, R.G. (1950) Cytological and histochemical variations in the adrenal cortex of the albino rat. J. Anat. 84, 196– 226.

23. Cater, D.B. &Lever, J.D.(1954) The zona intermedia of the adrenal cortex. A correlation of possible functional significance with development, morphology and histochemistry. J. Anat. 88, 437–454. 24. Deane, H.W. & Greep, R.O. (1946) A morphological and histo- chemical study of the rat’s adrenal cortex after hypophysectomy, with comments on the liver. Am. J. Anat. 79, 117–145. 4. Rogler, L.E. & Pintar, J.E. (1993) Expression of the P450 side- chain cleavage and adrenodoxin genes begins during early stages of adrenal cortex development. Mol. Endocrinol. 7, 453–461. 5. Raschella, G., Smets, G., Claeys, A., Verdood, P., Romeo, A. & Hooghe-Peters, E.L. (1989) Transcriptional pattern of 21- hydroxylase gene (P-450C21) during embryonic development, before, and after birth in mice as determined by in situ hybrid- ization. J. Histochem. Cytochem. 37, 751–756.

25. Hall, K. & Korenchevsky, V. (1937) Histological changes pro- duced by castration and by sex hormones in the adrenals of nor- mal and of castrated male rats. Nature 140, 318.

6. Engeland, W.C., Levay-Young, B.K., Rogers, L.M. & Fitzgerald, D. (1997) Di(cid:128)erential gene expression of cytochrome P45011b- hydroxylase in rat adrenal cortex after in vivo activation. Endo- crinology 138, 2338–2346.

26. Hartroft, P.M. & Eisenstein, A.B. (1957) Alterations in the adrenal cortex of the rat induced by sodium deficiency: correlation of histologic changes with steroid hormone secretion. Endocrinology 60, 641–651.

7. Ogishima, T., Mitani, F. & Ishimura, Y. (1989) Isolation of aldosterone synthase cytochrome P-450 from zona glomerulosa mitochondria of rat adrenal cortex. J. Biol. Chem. 264, 10935– 10938. 27. Mitchell, R.M. (1948) Histological changes and mitotic activity in the rat adrenal during postnatal development. Anat. Rec. 101, 161–185.

8. Domalik, L.J., Chaplin, D.D., Kirkman, M.S., Wu, R.C., Liu, W.W., Howard, T.A., Seldin, M.F. & Parker, K.L. (1991) Dif- ferent isozymes of mouse 11b-hydroxylase produce mineralocor- ticoids and glucocorticoids. Mol. Endocrinol. 5, 1853–1861. 28. Nicander, L. (1952) Histological and histochemical studies on the adrenal cortex of domestic and laboratory animals. Acta Anat. 14 (Supplement), 16.

29. Reiss, M., Balint, J., Oestreicher, F. & Aronson, V. (1936–7) Zur morphogenetischen Wirkung und biologishen Eichung der kortikotropen Wirksto(cid:128)es. Endokrinologie 18, 1–10. 9. Mukai, K., Imai, M., Shimada, H. & Ishimura, Y. (1993) Isolation and characterization of rat CYP11B genes involved in late steps of mineralo- and glucocorticoid syntheses. J. Biol. Chem. 268, 9130– 9137.

30. Mitani, F., Mukai, K., Miyamoto, H. & Ishimura, Y. (1998) Localization of replicating cells in rat adrenal cortex during the late gestational and early postnatal stages. Endocrinol. Res. 24, 983–986.

10. Rainey, W.E. (1999) Adrenal zonation: clues from 11b-hydroxylase and aldosterone synthase. Mol. Cell. Endocrinol. 151, 151–160. 11. Waterman, M.R. (1994) Biochemical diversity of cAMP-depen- dent transcription of steroid hydroxylase genes in the adrenal cortex. J. Biol. Chem. 269, 27783–27786. 31. Mitani, F., Mukai, K., Miyamoto, H., Suematsu, M. & Ishimura, Y. (1999) Development of functional zonation in the rat adrenal cortex. Endocrinology 140, 3342–3353.

12. Reinhart, A.J., Williams, S.C. & Stocco, D.M. (1999) Tran- scriptional regulation of the StAR gene. Mol. Cell. Endocrinol. 151, 161–169.

32. Miyamoto, H., Mitani, F., Mukai, K., Suematsu, M. & Ishimura, Y. (1999) Studies on cytogenesis in adult rat adrenal cortex: cir- cadian and zonal variations and their modulation by adrenocor- ticotropic hormone. J. Biochem. 126, 1175–1183.

13. Lala, D.S., Rice, D.A. & Parker, K.L. (1992) Steroidogenic factor I, a key regulator of steroidogenic enzyme expression, is the mouse homolog of fushi tarazu-factor I. Mol. Endocrinol. 6, 1249–1258. 14. Morohashi, K., Honda, S., Inomata, Y., Handa, H. & Omura, T. (1992) A common trans-acting factor, Ad4-binding protein, to the promoters of steroidogenic P-450s. J. Biol. Chem. 267, 17913– 17919.

33. Miyamoto, H., Mitani, F., Mukai, K., Suematsu, M. & Ishimura, Y. (2000) Daily regeneration of rat adrenocortical cells: circadian and zonal variations in cytogenesis. Endocrinol. Res. 26, 899–904. 34. Morley, S.D., Viard, I., Chung, B.C., Ikeda, Y., Parker, K.L. & Mullins, J.J. (1996) Variegated expression of a mouse steroid 21-hydroxylase/b-galactosidase transgene suggests centripetal migration of adrenocortical cells. Mol. Endocrinol. 10, 585–598. 15. Luo, X., Ikeda, Y. & Parker, K.L. (1994) A cell-specific nuclear receptor is essential for adrenal and gonadal development and sexual di(cid:128)erentiation. Cell 77, 481–490.

35. Frederiksen, K., Jat, P.S., Valtz, N., Levy, D. & McKay, R. (1988) Immortalization of precursor cells from the mammalian CNS. Neuron 1, 439–448. 16. Parker, K.L. & Schimmer, B.P. (1997) Steroidogenic factor 1: a key determinant of endocrine development and function. Endo- crinol. Rev. 18, 361–377.

36. Renfranz, P.J., Cunningham, M.G. & McKay, R.D. (1991) Re- gion-specific di(cid:128)erentiation of the hippocampal stem cell line HiB5 upon implantation into the developing mammalian brain. Cell 66, 713–729.

37. Yanai, N., Suzuki, M. & Obinata, M. (1991) Hepatocyte cell lines established from transgenic mice harboring temperature-sensitive simian virus 40 large T-antigen gene. Exp. Cell Res. 197, 50–56. 17. Mitani, F., Suzuki, H., Hata, J., Ogishima, T., Shimada, H. & Ishimura, Y. (1994) A novel cell layer without corticosteroid- synthesizing enzymes in rat adrenal cortex: histochemical detec- tion and possible physiological role. Endocrinology 135, 431–438. 18. Mitani, F., Ogishima, T., Miyamoto, H. & Ishimura, Y. (1995) Localization of P450aldo and P45011b in normal and regenerating rat adrenal cortex. Endocrinol. Res. 21, 413–423.

38. Jat, P.S. & Sharp, P.A. (1989) Cell lines established by a temper- ature-sensitive simian virus 40 large-T-antigen gene are growth restricted at the nonpermissive temperature. Mol. Cell. Biol. 9, 1672–1681. 19. Mitani, F., Miyamoto, H., Mukai, K. & Ishimura, Y. (1996) E(cid:128)ects of long term stimulation of ACTH and angiotensin II- secretions on the rat adrenal cortex. Endocrinol. Res. 22, 421–431.

(cid:211) FEBS 2002

Immortalized adrenocortical cell lines (Eur. J. Biochem. 269) 81

52. Stromstedt, M. & W.Aterman, M.R. (1995) Messenger RNAs encoding steroidogenic enzymes are expressed in rodent brain. Mol. Brain Res. 34, 75–88. 39. Obinata, M. (1997) Conditionally immortalized cell lines with di(cid:128)erentiated functions established from temperature-sensitive T-antigen transgenic mice. Genes Cells 2, 235–244.

53. Sabath, D.E., Broome, H.E. & Prystowsky, M.B. (1990) Glycer- aldehyde-3-phosphate dehydrogenase mRNA is a major inter- leukin 2-induced transcript in a cloned T-helper lymphocyte. Gene 91, 185–191. 40. Noble, M., Groves, A.K., Ataliotis, P., Ikram, Z. & Jat, P.S. (1995) The H-2KbtsA58 transgenic mouse: a new tool for the rapid generation of novel cell lines. Transgenic Res. 4, 215– 225.

54. Sugano, S., Morishima, N., Ikeda, H. & Horie, S. (1989) Sensitive assay of cytochrome P450scc activity by high-performance liquid chromatography. Anal. Biochem. 182, 327–333. 55. Schimmer, B.P. (1979) Adrenocortical Y1 cells. Methods Enzymol. 41. Mukai, K., Mitani, F., Shimada, H. & Ishimura, Y. (1995) Involvement of an AP-1 complex in zone-specific expression of the CYP11B1 gene in the rat adrenal cortex. Mol. Cell. Biol. 15, 6003–6012. 58, 570–574.

42. Yamane, I., Kan, M., Hoshi, H. & Minamoto, Y. (1981) Primary culture of human diploid cells and its long-term transfer in a serum-free medium. Exp. Cell. Res. 134, 470–474. 56. Rainey, W.E., Bird, I.M. & Mason, J.I. (1994) The NCI-H295 cell line: a pluripotent model for human adrenocortical studies. Mol. Cell. Endocrinol. 100, 45–50.

43. Ikeda, Y., Lala, D.S., Luo, X., Kim, E., Moisan, M.P. & Parker, K.L. (1993) Characterization of the mouse FTZ-F1 gene, which encodes a key regulator of steroid hydroxylase gene expression. Mol. Endocrinol. 7, 852–860.

44. Hatano, O., Takayama, K., Imai, T., Waterman, M.R., Taka- kusu, A., Omura, T. & Morohashi, K. (1994) Sex-dependent expression of a transcription factor, Ad4BP, regulating steroido- genic P-450 genes in the gonads during prenatal and postnatal rat development. Development 120, 2787–2797. 57. Cheng, C.Y., Ryan, R., Vo, T.P. & Hornsby, P.J. (1989) Cellular senescence involves stochastic processes causing loss of expression of di(cid:128)erentiated function genes: transfection with SV40 as a means for dissociating e(cid:128)ects of senescence on growth and on di(cid:128)eren- tiated function gene expression. Exp. Cell Res. 180, 49–62. 58. Mellon, S.H., Miller, W.L., Bair, S.R., Moore, C.C., Vigne, J.L. & Weiner, R.I. (1994) Steroidogenic adrenocortical cell lines pro- duced by genetically targeted tumorigenesis in transgenic mice. Mol. Endocrinol. 8, 97–108.

45. Ikeda, Y., Shen, W.H., Ingraham, H.A. & Parker, K.L. (1994) Developmental expression of mouse steroidogenic factor-1, an essential regulator of the steroid hydroxylases. Mol. Endocrinol. 8, 654–662.

46. Clark, B.J., Soo, S.C., Caron, K.M., Ikeda, Y., Parker, K.L. & Stocco, D.M. (1995) Hormonal and developmental regulation of the steroidogenic acute regulatory protein. Mol. Endocrinol. 9, 1346–1355.

59. Morohashi, K., Iida, H., Nomura, M., Hatano, O., Honda, S., Tsukiyama, T., Niwa, O., Hara, T., Takakusu, A., Shibata, Y. et al. (1994) Functional di(cid:128)erence between Ad4 BP and ELP, and their distributions in steroidogenic tissues. Mol. Endocrinol. 8, 643–653. 60. Mason, J.I., Keeney, D.S., Bird, I.M., Rainey, W.E., Morohashi, K., Leers-Sucheta, S. & Melner, M.H. (1997) The regulation of 3b- hydroxysteroid dehydrogenase expression. Steroids 62, 164–168. 61. Hornsby, P.J., O’Hare, M.J. & Neville, A.M. (1974) Functional and morphological observations on rat adrenal zona glomerulosa cells in monolayer culture. Endocrinology 95, 1240–1251. 47. Mukai, K., Mitani, F., Agake, R. & Ishimura, Y. (1998) Adrenocorticotropic hormone stimulates CYP11B1 gene tran- scription through a mechanism involving AP-1 factors. Eur. J. Biochem. 256, 190–200.

62. Engeland, W.C., Levay-Young, B.K., Paul, J.A. & Fitzgerald, D.A. (1995) Expression of cytochrome P450 aldosterone synthase and 11b-hydroxylase mRNA during adrenal regeneration. Endo- crinol. Res. 21, 449–454. 48. Tsuyama, N., Miura, M., Kitahira, M., Ishibashi, S. & Ide, T. (1991) SV40 T-antigen is required for maintenance of immortal growth in SV40-transformed human fibroblasts. Cell Struct. Funct. 16, 55–62.

63. Cammas, F.M., Pullinger, G.D., Barker, S. & Clark, A.J. (1997) The mouse adrenocorticotropin receptor gene: cloning and char- acterization of its promoter and evidence for a role for the orphan nuclear receptor steroidogenic factor 1. Mol. Endocrinol. 11, 867–876.

49. Oonk, R.B., Krasnow, J.S., Beattie, W.G. & Richards, J.S. (1989) Cyclic AMP-dependent and -independent regulation of cholesterol side chain cleavage cytochrome P-450 (P-450scc) in rat ovarian granulosa cells and corpora lutea. cDNA and deduced amino acid sequence of rat P-450scc. J. Biol. Chem. 264, 21934– 21942.

64. Qiu, R., Frigeri, C. & Schimmer, B.P. (1998) A role for guanyl nucleotide-binding regulatory protein b- and c-subunits in the expression of the adrenocorticotropin receptor. Mol. Endocrinol. 12, 1879–1887.

50. Keeney, D.S., Ikeda, Y., Waterman, M.R. & Parker, K.L. (1995) Cholesterol side-chain cleavage cytochrome P450 gene expression in the primitive gut of the mouse embryo does not require steroidogenic factor 1. Mol. Endocrinol. 9, 1091–1098.

65. Halder, S.K., Takemori, H., Hatano, O., Nonaka, Y., Wada, A. & Okamoto, M. (1998) Cloning of a membrane-spanning protein with epidermal growth factor-like repeat motifs from adrenal glomerulosa cells. Endocrinology 139, 3316–3328.

66. Engeland, W.C. & Levay-Young, B.K. (1999) Changes in the glomerulosa cell phenotype during adrenal regeneration in rats. Am. J. Physiol. 276, R1374–R1382. 51. Hasegawa, T., Zhao, L., Caron, K.M., Majdic, G., Suzuki, T., Shizawa, S., Sasano, H. & Parker, K.L. (2000) Developmental roles of the steroidogenic acute regulatory protein (StAR) as revealed by StAR knockout mice. Mol. Endocrinol. 14, 1462– 1471.