ELSEVIER
MCE
Molecular and Cellular Endocrinology
Transcription factors GATA-6, SF-1, and cell proliferation in human adrenocortical tumors
Sanne Kiiveri a, b, Jiangi Liuc, Johanna Arola c, d, Päivi Heikkilä c, d, Tiina Kuulasmaa e, Eero Lehtonen c, d, Raimo Voutilainen c, e, Markku Heikinheimo a, b, *
a Children’s Hospital, Biomedicum Helsinki, 00014 University of Helsinki, Finland
b Program for Developmental and Reproductive Biology, Biomedicum Helsinki, P.O. Box 63 (Haartmaninkatu 8), 00014 University of Helsinki, Finland
” Department of Pathology, 00014 University of Helsinki, Finland
d Department of Pathology, Helsinki University Central Hospital, 00290 Helsinki, Finland
e Department of Pediatrics, Kuopio University Hospital, 70210 Kuopio, Finland
Received 6 October 2004; received in revised form 13 January 2005; accepted 18 January 2005
Abstract
Transcription factor GATA-6 is expressed in fetal and adult human adrenal cortex and has been suggested to have a role in adrenal androgen synthesis. In other tissues GATA-6 has been linked to the cell cycle regulation and the dedifferentiation of carcinoma cells. GATA-6 has been shown to be downregulated in mouse adrenocortical tumors, but has not been studied in human adrenocortical tumors in detail. We have now analyzed GATA-6 expression in 20 human adrenocortical adenomas and 16 carcinomas using Northern blot analysis and immunohistochemistry. GATA-6 mRNA and protein expression was remarkably diminished in adrenocortical carcinomas as compared to normal adrenal cortex and adenomas (p<0.05). In opposite to other tumor types GATA-6 expression was, however, high in virilizing carcinomas. Steroidogenic factor 1 (SF-1) has been functionally linked to GATA-6, and the expression of these two factors correlated in the adrenal tumors. Furthermore, GATA-6 immunoreactivity was linked to P450c17 expression. In contrast to GATA-6, we found upregulated cyclin-dependent kinase inhibitor p21 and proliferation marker Ki67 in adrenocortical carcinomas indicating that GATA-6 is not linked to cell proliferation in human adrenal tumors. Taken together, the present and earlier results link GATA-6 to adrenocortical steroidogenesis and to the benign adrenocortical phenotype.
@ 2005 Elsevier Ireland Ltd. All rights reserved.
Keywords: Adrenal gland; Adrenocortical tumor; GATA-6; SF-1; Transcription factor
1. Introduction
Incidentally detected benign and hormonally inactive neoplasms are frequent abnormalities of the adrenal cor- tex (Kloos et al., 1995), whereas adrenocortical carcinoma (ACC) is a rare disease with an incidence of 0.5-2 cases per million per year (Bernstein and Gurney, 1999; Libertino and Libertino, 2003). In adults, adrenal tumors are usually discovered due to Cushing’s syndrome or a palpable abdom- inal mass, while in children they most often cause symp-
toms of virilization (Libertino and Libertino, 2003). In an analysis of 602, mainly adult, adrenal tumors 62% were considered functional (Libertino and Libertino, 2003). In children, 83-100% of ACCs have been reported to be hor- monally active (Sabbaga et al., 1993; Wajchenberg et al., 2000).
During the last decade several studies have elucidated genetic changes that accompany adrenal tumorigenesis. Most adrenal adenomas and carcinomas are monoclonal (Beuschlein et al., 1994). Rearrangements in the human chro- mosome 11p15 region and/or insulin-like growth factor II (IGF-II) overexpression are frequent features in adrenocorti- cal malignancies (reviewed in Reincke, 1998). In a smaller
* Corresponding author. Tel .: +358 9 47171975; fax: +358 9 47171947. E-mail address: markku.heikinheimo@helsinki.fi (M. Heikinheimo).
proportion of ACCs mutations in the p53 tumor-suppressor gene and deletions in ACTH receptor gene have been found (Reincke, 1998). However, it is most likely that there are ad- ditional, still unknown factors, which play a role in adreno- cortical tumorigenesis.
GATA transcription factors have been associated with the development and function of several endocrine organs. The GATA family consists of six conserved zinc finger proteins expressed with distinct developmental and tissue specific profiles (Orkin, 1992; Evans, 1997). GATA-4 and GATA-6 deficient mice die at an early embryonic stage demonstrat- ing essential roles for these factors during normal mouse development (Kuo et al., 1997; Molkentin et al., 1997; Morrisey et al., 1998). GATA-4 first appears in visceral and parietal endoderm during embryogenesis. Later in develop- ment, it is expressed in several tissues including heart and gut (Arceci et al., 1993; Laverriere et al., 1994). The transcripts for GATA-6 are prominent in visceral endoderm, develop- ing and adult heart, lung and liver (Morrisey et al., 1996; Narita et al., 1996; Suzuki et al., 1996). GATA-4 and GATA- 6 are also expressed in gonads, where they are under hor- monal regulation and are thought to have crucial roles in gonadal function and sex determination (Heikinheimo et al., 1997; Viger et al., 1998; Ketola et al., 1999; Anttonen et al., 2003).
Our previous studies have shown that GATA-6 mRNA is expressed in fetal and adult mouse as well as human adrenal cortex (Kiiveri et al., 2002). In contrast, marked GATA-4 expression can be detected in mouse and human adrenals only during fetal period. Inhibin «/Simian virus T-antigen transgenic mice develop adrenocortical carcinomas after go- nadectomy (Kananen et al., 1996), and these tumors express GATA-4 mRNA and protein while GATA-6 is absent (Kiiveri et al., 1999). Analogously to this mouse model, GATA-4 pro- tein was recently implicated in human ACCs (Barbosa et al., 2004). GATA-6 has not been studied in conjunction with human adrenal tumors in more detail. Given the expression of GATA-6 in normal adrenal cortex, the downregulation of GATA-6 in other types of carcinomas (Capo-chichi et al., 2003) and that it has been linked to cell cycle regulation (Nagata et al., 2000), it was of interest to examine possible roles for GATA-6 in adrenocortical tumors. Since GATA- 6 has been recently linked to adrenal androgen production (Jimenez et al., 2003), we also hoped to gain insight to this relationship by studying tumors with various hormonal ac- tivities.
In the present work, GATA-6 expression was compared to that of steroidogenic factor 1 (SF-1), steroidogenic enzyme P450c17 (cytochrome P450 17a-hydroxylase/17,20-lyase), cyclin-dependent kinase inhibitor p21(WAF-1) and Ki67 pro- liferation marker. We found diminished GATA-6 expression in carcinomas as compared to normal adrenal cortex and ade- nomas. There was also a tendency for low GATA-6 expression in the tumors with a high Weiss score, an indicator for the prognosis of adrenocortical tumors (Weiss, 1984; Weiss et al., 1989; Lucon et al., 2002).
2. Materials and methods
2.1. Patients and tissue samples
The tumor material has been described in detail previ- ously, and tumors were judged as nonfunctional or functional tumors according to the leading clinical features and the pre- dominant hormone production in a given patient (Arola et al., 2000). A total of 22 of the tumors included in the study were originally classified as adenomas and 17 as carcinomas. Tumors are called here as follows: nonfunctional, Conn’s (al- dosterone producing), Cushing’s (cortisol producing) and vir- ilizing (androgen producing) adenomas or carcinomas. The number of analyzed samples according to their hormonal sta- tus is shown in Tables 1 and 2. Paraffin-embedded tissue samples (34 formalin fixed and 7 Bouin fixed) were used for immunohistochemistry and frozen tissue (n=33) for RNA analysis. From three tumor samples (two adenomas, one car- cinoma) only RNA was available and they are not listed in Table 1. Of all tumor patients four were pediatric and 35 were adults, and 24 were females and 15 males. For com- parison, adrenal tissue adjacent to the tumor (i.e., the non- affected part of the tumor containing adrenal) was obtained from 10 patients. Normal adrenal tissue was also obtained from eight adult patients (four females, four males) who un- derwent surgery for kidney tumors. The aggressiveness of the tumors was assessed according to the criteria of Weiss (Weiss, 1984; Weiss et al., 1989) and a score from 0 (benign) to 9 (highly malignant) was given to each tumor (a consensus by two pathologists J.A. and P.H.).
The studies were approved by the Ethics Committees of the Hospital for Children and Adolescents and the Depart- ment of Internal Medicine, Helsinki University Central Hos- pital.
2.2. Immunohistochemistry
Paraffin-embedded human tissue samples were deparaf- finized. The tissue sections were then subjected to im- munochemistry using commercial polyclonal rabbit anti- GATA-6 IgG (1:50 and 1:100) (Santa Cruz Biotechnol- ogy Inc., cat# sc-9055, Santa Cruz, CA), polyclonal rab- bit anti-SF-1 IgG (1:200) (Upstate Biotechnology, cat# 06- 431, Lake Placid, NY), monoclonal mouse anti-p21(WAF1) IgG (1:100) (Oncogene Research Products, cat# OP64, San Diego, CA) and monoclonal mouse anti-Ki-67 (1:75) (DAKO A/S, clone MIB-1, Glostrup, Denmark). The specificity of the staining was assessed using the preimmune serum or PBS instead of the primary antibody during the staining protocol. The avidin-biotin immunoperoxidase system was used to visualize bound antibody according to the manufac- turer’s instructions (Vectastain Elite ABC Kit, Vector labora- tories, Inc., Burlingame, CA). 3,3’-Diaminobenzedine (DAP) (Sigma) or 3-amino-9-ethylcarbazole was used as the chro- mogen and the development reaction occurred in the presence of 0.03% H2O2. The tissues were counterstained with hema-
| Tissue sample | Patient | Immunohistochemistry | Weiss score | |||||
|---|---|---|---|---|---|---|---|---|
| Number | Sex | Age | GATA-6 | SF-1 | p21 | Ki67 | ||
| Adenoma | ||||||||
| Nonfunctional | 1 | F | 67 | 60 | 50 | 10 | 1 | 1 |
| 2 | M | 66 | 60 | 40 | 10 | 2 | 0 | |
| 3 | F | 53 | 90 | 10 | 30 | 1 | 1 | |
| 4 | F | 67 | 80 | 0 | 2 | 1 | 1 | |
| 5 | F | 71 | 10 | 0 | 1 | 1 | 0 | |
| Conn's | 6 | F | 39 | 50 | 50 | 4 | 2 | 1 |
| 7 | F | 47 | 50 | 30 | 2 | 1 | 0 | |
| 8 | F | 54 | 15 | 5 | 0 | NA | 0 | |
| 9 | M | 69 | 20 | 0 | 1 | NA | 0 | |
| 10 | M | 40 | 75 | 70 | 1 | 2 | 1 | |
| 11 | M | 65 | 80 | 50 | 20 | 1 | 1 | |
| 12 | F | 26 | 90 | 70 | 5 | 1 | 1 | |
| Cushing's | 13 | F | 54 | 80 | 10 | 50 | 1 | 1 |
| 14 | F | 65 | 70 | 60 | 0 | 1 | 0 | |
| 15 | F | 39 | 20 | 30 | 0 | 1 | 0 | |
| 16 | F | 69 | 50 | 60 | 5 | 1 | 0 | |
| 17 | M | 47 | 0 | 0 | 0 | 1 | 2 | |
| Virilizing | 18 | F | 37 | 95 | 100 | 75 | 2 | 1 |
| 19 | F | 51 | 50 | 50 | 1 | 2 | 3 | |
| 20 | F | 7 | 5 | 50 | 0 | 1 | 3 | |
| Carcinoma | ||||||||
| Nonfunctional | 21 | F | 55 | 0 | 5 | 3 | 30 | 9 |
| 22 | M | 21 | 40 | 70 | 2 | 5 | 5 | |
| 23 | M | 65 | 20 | 40 | 50 | 30 | 6 | |
| 24 | M | 69 | 10 | 10 | 50 | 50 | 8 | |
| 25 | M | 76 | 0 | 0 | 20 | NA | 7 | |
| 26 | M | 68 | 2 | 0 | 10 | NA | 5 | |
| Conn's | 27 | F | 76 | 0 | 0 | 5 | NA | 6 |
| 28 | F | 71 | 80 | 0 | 20 | 20 | 6 | |
| 29 | M | 50 | 0 | 0 | 10 | NA | 5 | |
| Cushing's | 30 | M | 57 | 0 | 10 | 20 | NA | 8 |
| 31 | F | 1 | 30 | 20 | 5 | 10 | 4 | |
| 32 | F | 39 | 50 | 3 | 75 | 20 | 6 | |
| 33 | M | 45 | 10 | 0 | 15 | NA | 4 | |
| Virilizing | 34 | F | 74 | 90 | 10 | 90 | 30 | 6 |
| 35 | F | 11 | 60 | 60 | 5 | 10 | 3 | |
| 36 | M | 1 | 80 | 95 | 30 | 30 | 8 | |
Results on immunohistochemical staining are based on the percentage of positive tumor cells for the given protein; NA: not available.
| Adenoma | Carcinoma | |||||
|---|---|---|---|---|---|---|
| n | Mean | Range | n | Mean | Range | |
| Nonfunctional | 2 | 1.6 | 1.5-1.7 | 2 | 0.6 | 0.5-0.7 |
| Conn's | 8 | 0.9 | 0.3-1.5 | 1 | 0.7 | 0.7 |
| Cushing's | 6 | 0.9 | 0.5-1.5 | 4 | 0.7 | 0.1-1.5 |
| Virilizing | 3 | 0.7 | 0.1-1.3 | 3 | 1.0 | 0.7-1.3 |
The intensities were compared to the expression in normal adrenal tissue (arbitrary unit for normal adrenal=1).
toxylin. Two researchers independently scored the slides by estimating the percentage of tumor cells showing character- istic immunoreactivity for each marker (from undetectable levels, 0%, to homogenous staining, 100%), and the result was a consensus between reviewing investigators.
2.3. RNA isolation and Northern blot analysis
Total RNA was isolated from the frozen tissues by ultra- centrifugation through a cesium chloride gradient (Chirgwin et al., 1979). Denatured RNA (10-20 µg) was subjected to electrophoresis in a 1% denaturing agarose gel and then transferred onto nylon membranes. The membranes were hy- bridized with two 32P-labelled (>6000 Ci/mmol, Amersham) synthetic oligonucleotide probes for human GATA-6. The oligonucleotides were prepared at the Institute of Biotechnol- ogy, University of Helsinki (Helsinki, Finland), and the se- quences were 5’-CGT CTG GAT GGA GCC GCA GTT CAC GCA CTC-3’ and 5’-AAG CCG CCG TGA TGA AGG CAC GCG CTT CTG-3’ corresponding to nucleotides 1074-1103 and 1200-1229 of human GATA-6 cDNA (GenBank ac- cession no. NM005257). 28S ribosomal RNA cDNA probe (Arnheim, 1979) was a loading control. The hybridizations with the oligonucleotide probes were performed at 60 ℃ over night (HB-1 D Hybridization oven, Techne, Cambridge, UK) and the filters were washed three times 20 min at 60 ℃ with 1 × SSC/0.1% SDS. Hybridization signals were detected by autoradiography using Agfa Curix Ortho ST-L film (Agfa- Gavaert N.V., Belgium). The size of the GATA-6 transcript was 3.8 kb. The intensities of the lanes were analyzed with Scion Image for Windows (Scion Corporation, Frederick, MD). All samples were normalized with the 28S ribosomal RNA values and the intensities were compared to the expres- sion of normal adrenal in the same membrane (arbitrary unit for normal adrenal = 1). When the same sample was analyzed in several membranes the mean value of the intensities was used.
2.4. cDNA synthesis
cDNA was synthesized using high-capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA, USA) according to manufacturer’s protocol. Reactions were made in total volume of 20 ul containing 1 µg total RNA, one-fold reaction buffer, dNTP mixture and random primers and 50 U MultiScribe reverse transcriptase. Reaction mixtures were in- cubated in +25 ℃ for 10 min following incubation in +37 ℃ for 2 h.
2.5. Quantitative real-time PCR
Quantitative real-time PCR was carried out in the Applied Biosystems 7500 real-time PCR System using TaqMan® Gene Expression Assays for P450c17 (CYP17; assay ID Hs00164375_m1) and glyceraldehyde-3-phosphate dehydro- genase (GAPD; assay ID Hs99999905_m1) with standard
B
curve method. A standard series of five dilutions contain- ing 48, 12, 6, 1.5 and 0.5 ng template cDNA was prepared from pooled sample cDNAs. Sample dilutions comprised of 6 ng template cDNA. All standards and samples were run in total volume of 20 ul in triplicate.
2.6. Statistical analysis
The percentage of positive cells (0-100%) in immunohis- tochemical staining and the level of mRNA intensity between the different groups of tumors (adenomas versus carcinomas, Weiss score 1-3 versus Weiss score 4-9) were compared with a non-parametric Mann-Whitney U-test as the values were not normally distributed. Correlations between the lev- els of different markers were analyzed with a bivariate Spear- man’s test. These analyzes were performed using SPSS 10.0 software (SPSS Inc., Chicago, IL). Regression analysis was performed to examine the relationship between GATA-6 im- munoreactivity and P450c17 mRNA levels (Statview 5.0.1 software for Macintosh®). For the analysis, the highest value of these factors was omitted to exclude the artificial effect of the extreme measures. A p-value of <0.05 was considered significant.
3. Results
3.1. GATA-6 expression is often diminished in ACCs
First, we analyzed GATA-6 protein expression in the nor- mal adrenals and tumors by immunohistochemistry. Most of the normal adrenocortical cells (97 and 98%) expressed GATA-6 (Figs. 1 and 3A). GATA-6 expression profile in each individual tumor is shown in Table 1, and represen- tative immunohistochemical stainings in various tumor types
are demonstrated in Fig. 2. Remarkably fewer GATA-6 posi- tive cells were seen in carcinomas as compared to adenomas (Fig. 3A). Despite some overlap in the pattern of GATA- 6 immunoreactivity between the groups, the difference be- tween adenomas and carcinomas was statistically significant (p<0.05). The difference between adenomas and carcino- mas was more prominent in nonfunctional than functional tumors (Fig. 3B). In opposite to other tumor types, GATA-6 protein expression in virilizing tumors remained high even in carcinomas (Fig. 3B).
We next analyzed GATA-6 mRNA expression levels in the adrenal tumors. RNA samples were available from 29 tumors, and their GATA-6 expression was examined by Northern blot analysis (Fig. 4). The mean GATA-6 mRNA level was 0.8 ar- bitrary units in carcinomas, 0.9 in adenomas and 1.0 in normal adrenal tissue (Table 2 and data not shown). The difference between adenomas (1.6) and carcinomas (0.6) was highest in nonfunctional tumors. A tendency of lower GATA-6 mRNA expression in carcinomas was seen supporting the results on GATA-6 immunoreactivity.
3.2. Low GATA-6 expression associates with a high Weiss score
Given the differences in GATA-6 expression between ade- nomas and carcinomas, we wanted to see whether the down- regulation of GATA-6 in carcinomas relates to the malignant behavior of these tumors. The results on the Weiss’s histo- logic criteria (scores 0-9) are shown in Table 1. GATA-6 tended to be lower with a higher Weiss score, but the cor- relation between these two parameters did not reach sta- tistical significance (p=0.06). However, reflecting the dif- ference between adenomas and carcinomas, GATA-6 lev- els were markedly higher in the tumors having a Weiss score 1-3 than in those with a Weiss score 4-9 (p<0.05) (Fig. 3C).
3.3. GATA-6 protein expression correlates with SF-1
SF-1 has been shown to be constantly expressed in both normal and pathological human adrenal tissues (Sasano et al., 1995a,b). Since GATA factors have been functionally linked to SF-1 (Tremblay and Viger, 1999, 2001), we wanted to see whether the expression pattern of SF-1 would correlate with that of GATA-6. The immunoreactivity (the percentage of positive tumor cells) of SF-1 is presented in Table 1. Repre- sentative immunohistochemical stainings of SF-1 in the var- ious tumor groups (same samples as for GATA-6 immuno- histochemistry in Fig. 1) are shown in Fig. 5.
SF-1 protein levels were somewhat higher in virilizing adenomas and carcinomas than in other tumors, and sur- prisingly no SF-1 protein was found in Conn’s carcinomas (Table 1). The expression levels of GATA-6 and SF-1 cor- related significantly (p<0.05), and this was most obvious in Conn’s and virilizing adenomas (Table 1). Single tumors with opposite expression patterns were seldom seen (Table 1,
Adenoma
Carcinoma
Nonf.
A
B
Conn’s
D
E
Cushing’s
K
G
Virilizing
H
GATA-6 positive nuclei (%)
100
GATA-6 positive nuclei (%)
90
GATA-6 positive nuclei (%)
*
100
100
80
90
90
*
70
*
80
80
*
60
70
70
50
60
60
40
50
50
40
40
30
20
30
30
20
20
10
10
10
0
0
0
(A)
Norm
Aden
Ca
(B)
Nonf.
Conn’s
Cushing’s
Viril.
(C) Weiss
1-3
4-6
7-9
Norm. adr.
Conn’s aden
Cushing’s aden
Virilizing aden
Nonf. aden
Cushing’s ca
Virilizing ca
Nonf. ca
Co
Co
Co
Co
GATA-6
3.8 kb
28S
cases #4 and #28). Similarly to earlier reports, the difference in SF-1 expression between adenomas and carcinomas was not significant. Neither did SF-1 correlate with the Weiss score.
3.4. GATA-6 correlates with P450c17 in the adrenal tumors
GATA-6 has been shown to regulate P450c17 in human adrenocortical cells (Flück and Miller, 2004), and therefore we examined the possible link between these two proteins in our tumor series. Frozen RNA was available of 20 tumor sam- ples and their P450c17 expression was analyzed using quan- titative RT-PCR. The mean P450c17 expression, normalized against GADP, was 1.9 (range 0-5.5). There was a positive correlation between GATA-6 and P450c17 (p<0.05), thus supporting a mechanistic link between these two proteins.
3.5. Proliferation markers p21 and Ki67 in adrenocortical tumors
A cyclin-dependent kinase inhibitor p21, which is re- lated to the regulation of the cell cycle, has been connected
Adenoma
Carcinoma
Nonf.
A
B
Conn’s
D
E
Cushing’s
F
G
Virilizing
H
I
GATA-6
p21
Ki67
A
B
C
Adenoma
D
E
F
Carcinoma
to GATA-6 (Perlman et al., 1998; Nagata et al., 2000) In addition, p21 has been shown to be upregulated in ACCs (Stojadinovic et al., 2002). We therefore analyzed the expres- sion of p21 together with that of Ki67, a widely used marker for cell proliferation, in the adrenal tumors. Both p21 and Ki67 were significantly upregulated in ACCs as compared to adenomas (Table 1, p<0.05) and their expression profiles thus diverged from that of GATA-6. Examples of low p21 expression in one adenoma (Table 1, case #10) and high ex- pression in one carcinoma (Table 1, case #32) are shown in Fig. 6.
4. Discussion
The GATA transcription factors play specific roles in cel- lular maturation and differentiation and their importance dur- ing normal development has been established in a variety of tissues (Arceci et al., 1993; Laverriere et al., 1994; Soudais et al., 1995; Morrisey et al., 1998). Several target genes for GATA proteins in the endocrine organs have also been identified. These include gonadotropin-releasing hormone (GnRH), Müllerian inhibiting substance (MIS), inhibin &, inhibin/activin BB-subunit and steroidogenic acute regula- tory protein (StAR) (Lawson et al., 1996; Feng et al., 1998, 2000; Viger et al., 1998; Ketola et al., 1999; Tremblay and Viger, 1999; Tremblay et al., 2002). Recent studies have im- plicated GATA factors also during malignant transformation in some endocrine tumors and other malignancies (Siltanen et al., 1999; Laitinen et al., 2000; Lin et al., 2000; Lassus et al., 2001; Wechsler et al., 2002; Siltanen et al., 2003). Inhibin &/Simian virus T-antigen transgenic mice and inbred
DBA mice develop adrenocortical tumors after gonadectomy (Feteke et al., 1941; Kananen et al., 1996). These tumors ex- press GATA-4 and a reciprocal disappearance of GATA-6 in the same tumor area is noted (Kiiveri et al., 1999; Bielinska et al., 2003). Studies on human adrenal tumors have linked the expression of GATA-4 to malignancy analogously to animal models (Kiiveri et al., 1999; Barbosa et al., 2004; Peterson et al., 2004). Alterations in the expression of GATA-6, as well as other transcription factors, in human adrenocortical tumors as compared to normal adrenal tissue have recently been de- scribed supporting our present findings (Barbosa et al. 2004; Suzuki et al., 2000; Shibata et al., 2001). Collectively, these data implicate important roles for GATA proteins not only in the endocrine development and function, but also in the cancer of these organs.
Our previous studies showed that GATA-6 is expressed during normal adrenal development in mice and humans from fetal period onwards, while GATA-4 expression is re- stricted to the fetal period and malignant adrenocortical tis- sue (Kiiveri et al., 1999; Kiiveri et al., 2002). In the adult human adrenal GATA-6 is expressed, besides other zones, in the zona reticularis, which is the major source for adrenal androgen production. In an adrenocortical cell line GATA-6 has been shown to act in synergy with SF-1 to maximally increase the expression of proteins needed for adrenal andro- gen production (e.g. StAR) (Jimenez et al., 2003). The co- operation of SF-1 and GATA factors, including GATA-6, has also been shown in the transactivation studies of the MIS pro- moter (Tremblay and Viger, 1999). Here, we demonstrate a correlation between GATA-6 and SF-1 expression in adreno- cortical tumors strengthening the concept that these factors have shared functions in the adrenal development and func-
tion. GATA-6 has also been shown to modulate the tissue- specific transcription of the steroidogenic enzyme P450c17 in a human adrenocortical cell line (Flück and Miller, 2004). In the present work, GATA-6 expression was maintained in virilizing carcinomas when compared to the other types of ACCs, and linked to P450c17 expression, suggesting roles for GATA-6 in the adrenal androgen production.
According to a recent study, GATA-6 function was lost in 82% of ovarian epithelial tumors along with epithelial markers (Capo-chichi et al., 2003). We find that GATA-6 is remarkably lower in adrenocortical carcinomas than adeno- mas, and the lost or disturbed GATA-6 function might well be linked to the dedifferentiation also in the adrenal neoplasms (Marx et al., 1996). Other studies have provided additional evidence for the role of GATA factors in enhancing changes towards more benign phenotypes. Accordingly, mouse ery- throleukemia cells were triggered to differentiate and lose their tumorigenicity by addition of an active form of GATA-1 (Choe et al., 2003). In another study, restored GATA-6 protein expression in vascular smooth muscle cells (VSMCs) inhib- ited injury induced hyperplasia (Mano et al., 1999). Opposite functions for GATA-6 have been suggested in the intestine, where its highest expression is detected in the proliferating progenitor cells (Gao et al., 1998). In the present study, the low GATA-6 immunoreactivity associated with a high Weiss score, measuring e.g. mitotic activity and invasion (Weiss, 1984; Weiss et al., 1989). The finding on diminished GATA- 6 expression in malignant tumors is therefore consistent with the possibility that GATA-6 has functions in maintaining the normal quiescent differentiated adrenal cell type.
GATA-6 has been linked to the regulation of the cell cy- cle and proliferation in VSMCs and glomerular mesangial cells (GMCs) (Suzuki et al., 1996; Nagata et al., 2000). First GATA-6 was shown to be downregulated in VSMCs in re- sponse to mitogen activation, and similar results were later obtained in GMCs. In addition, an overexpression of GATA-6 induced cell cycle arrest in GMCs associating with an upreg- ulation of the general cyclin-dependent kinase inhibitor p21 (Perlman et al., 1998; Nagata et al., 2000). Based on these studies we examined the p21 protein expression in our tumor series. As previously reported (Stojadinovic et al., 2002), p21 expression was upregulated in ACCs, and there was thus no correlation between p21 and GATA-6 protein levels in neo- plastic adrenocortical cells. We also wanted to study the pos- sible association of GATA-6 and cell proliferation in these tu- mors, but found no relation between the expression of GATA- 6 and Ki67 proliferation marker.
ACCs are generally monoclonal (Beuschlein et al., 1994; Gicquel et al., 1994), and tumor genotyping has been done in search for oncogenes and tumor suppressor genes. Most frequently loss of heterozygosity (LOH) has been described on chromosomes 11p15, 17p13, 2p16 and 18p21.1, involving e.g. IGF II and ACTH receptor genes (reviewed in Reincke, 1998; Kjellman et al., 2001). Mutations in the p53 tumor sup- pressor gene have also been reported in adrenal tumors (Lin et al., 1994). The disruption of GATA-6 expression and its
inactivation may happen at multiple levels. Human GATA-6 gene is located on chromosome 18q (Morrisey et al., 1996; Suzuki et al., 1996) and GATA-4 on chromosome 8p (White et al., 1995). Some reports of losses in chromosomes 8 and 18 have been published in ACCs and interestingly also gains in 8p (Figueiredo et al., 1999; Kjellman et al., 1999). It is therefore possible that genes for GATA-4 and GATA-6 are af- fected in these chromosomal aberrations, warranting further molecular genetic studies on GATA-4 and GATA-6 genes in these tumors. In colorectal and gastric cancer GATA-4 and GATA-5 have been shown to be silenced via promoter hy- permethylation (Akiyama et al., 2003). In the same tumors GATA-6 gene promoter was neither hypermethylated nor si- lenced, but hypermethylation remains a possible mechanism for diminished GATA-6 expression in ACCs.
All in all, we find GATA-6 downregulation in adreno- cortical carcinomas, and that this downregulation correlates with the higher Weiss score. GATA-6 expression also corre- lated with that of SF-1, which has been functionally linked to GATA transcription factors already before. As a widely expressed transcription factor GATA-6 has probably several functions in the adrenal cortex, some related to the develop- ment and differentiation of the adrenal, and others to the reg- ulation of steroidogenesis. While the regulation of androgen synthesis by GATA-6 has been recently elucidated (Jimenez et al., 2003), the suggested other functions of GATA-6 in the adrenal still remain to be established. In tumor cells, the fail- ure to express GATA-6 may be an important stage for the escape of the tumor cells from normal control mechanisms.
Acknowledgments
This work was supported by Finnish Pediatric Foundation (S.K., M.H.), Emil Aaltonen Foundation (S.K.), Maud Kuis- tila Foundation (S.K.), Finnish Medical Society Duodecim (S.K.), Jalmari and Rauha Ahokas Foundation (M.H., J.L., R.V.), Sigrid Juselius Foundation (R.V.), Finnish Cancer Foundation (J.A., P.H.), Academy of Finland (E.L., R.V.) and Kuopio University Hospital (R.V.). Dr. David B Wilson is thanked for constructive suggestions. Drs. Markku Kallio, Ilkka Ketola, Susanna Siltanen and Mia Westerholm-Ormio are thanked for their advice and Mrs Merja Haukka, Ritva Löfman and Taru Jokinen for the expert technical assistance.
References
Akiyama, Y., Watkins, N., Suzuki, H., Jair, K.W., van Engeland, M., Esteller, M., Sakai, H., Ren, C.Y., Yuasa, Y., Herman, J.G., Baylin, S.B., 2003. GATA-4 and GATA-5 transcription factor genes and po- tential downstream antitumor target genes are epigenetically silenced in colorectal and gastric cancer. Mol. Cell Biol. 23, 8429-8439.
Anttonen, M., Ketola, I., Parviainen, H., Pusa, A.K., Heikinheimo, M., 2003. FOG-2 and GATA-4 are coexpressed in the mouse ovary and can modulate Müllerian-inhibiting substance expression. Biol. Reprod. 68, 1333-1340.
Arceci, R.J., King, A.A., Simon, M.C., Orkin, S.H., Wilson, D.B., 1993. Mouse GATA-4: a retinoic acid-inducible GATA-binding transcription factor expressed in endodermally derived tissues and heart. Mol. Cell. Biol. 13, 2235-2246.
Arnheim, N., 1979. Characterization of mouse ribosomal gene fragments purified by molecular cloning. Gene 7, 83-96.
Arola, J., Liu, J., Heikkilä, P., Ilvesmäki, V., Salmenkivi, K., Voutilainen, R., Kahri, A.I., 2000. Expression of inhibin alpha in adrenocortical tumours reflects the hormonal status of the neoplasm. J. Endocrinol. 165, 223-229.
Barbosa, A.S., Giacaglia, L.R., Martin, R.M., Mendonca, B.B., Lin, C.J., 2004. Assessment of the role of transcript for GATA-4 as a marker of unfavorable outcome in human adrenocortical neoplasms. BMC Endocr. Disord., 4.
Bernstein, L., Gurney, J.G., 1999. Cancer Incidence and Survival among Children and Adolescents. In: National Cancer Institute monograph U.S.S.P., 1975-1995 ed. National Cancer Institute, USA, pp. 139- 147.
Beuschlein, F., Reincke, M., Karl, M., Travis, W.D., Jaursch-Hancke, C., Abdelhamid, S., Chrousos, G.P., Allolio, B., 1994. Clonal composition of human adrenocortical neoplasms. Cancer Res. 54, 4927-4932.
Bielinska, M., Parviainen, H., Porter-Tinge, S.B., Kiiveri, S., Genova,
E., Rahman, N., Huhtaniemi, I.T., Muglia, L.J., Heikinheimo, M., Wilson, D.B., 2003. Mouse strain susceptibility to gonadectomy- induced adrenocortical tumor formation correlates with the expres- sion of GATA-4 and luteinizing hormone receptor. Endocrinology 144, 4123-4133.
Capo-chichi, C.D., Roland, I.H., Vanderveer, L., Bao, R., Yamagata, T., Hirai, H., Cohen, C., Hamilton, T.C., Godwin, A.K., Xu, X.X., 2003. Anomalous expression of epithelial differentiation-determining GATA factors in ovarian tumorigenesis. Cancer Res. 63, 4967-4977.
Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J., Rutter, W.J., 1979. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry, 5294-5299.
Choe, K.S., Radparvar, F., Matushansky, I., Rekhtman, N., Han, X., Sk- oultchi, A.I., 2003. Reversal of tumorigenicity and the block to dif- ferentiation in erythroleukemia cells by GATA-1. Cancer Res. 63, 6363-6369.
Evans, T., 1997. Regulation of cardiac gene expression by GATA-4/5/6. Trends Cardiovasc. Med. 7, 75-83.
Feng, Z.M., Wu, A.Z., Chen, C.L., 1998. Testicular GATA-1 factor up- regulates the promoter activity of rat inhibin alpha-subunit gene in MA-10 Leydig tumor cells. Mol. Endocrinol. 12, 378-390.
Feng, Z.M., Wu, A.Z., Zhang, Z., Chen, C.L., 2000. GATA-1 and GATA-4 transactivate inhibin/activin beta-B-subunit gene transcription in tes- ticular cells. Mol. Endocrinol. 14, 1820-1835.
Feteke, E., Woolley, G., Little, C.C., 1941. Histological changes following ovariectomy in mice 1. dba high tumor strain. J. Exp. Med. 74, 1-8. Figueiredo, B.C., Stratakis, C.A., Sandrini, R., DeLacerda, L., Pianovsky, M.A., Giatzakis, C., Young, H.M., Haddad, B.R., 1999. Comparative genomic hybridization analysis of adrenocortical tumors of childhood. J. Clin. Endocrinol. Metab. 84, 1116-1121.
Flück, C.E., Miller, W.L., 2004. GATA-4 and GATA-6 modulate tissue- specific transcription of the human gene for P450c17 by direct inter- action with Sp1. Mol. Endocrinol. 26, 26.
Gao, X., Sedgwick, T., Shi, Y.B., Evans, T., 1998. Distinct functions are implicated for the GATA-4, -5, and -6 transcription factors in the regulation of intestine epithelial cell differentiation. Mol. Cell Biol. 18, 2901-2911.
Gicquel, C., Leblond-Francillard, M., Bertagna, X., Louvel, A., Chapuis, Y., Luton, J.P., Girard, F., Le Bouc, Y., 1994. Clonal analysis of human adrenocortical carcinomas and secreting adenomas. Clin. Endocrinol. (Oxf.) 40, 465-477.
Heikinheimo, M., Ermolaeva, M., Bielinska, M., Rahman, N.A., Narita, N., Huhtaniemi, I.T., Tapanainen, J.S., Wilson, D.B., 1997. Expression and hormonal regulation of transcription factors GATA-4 and GATA-6 in the mouse ovary. Endocrinology 138, 3505-3514.
Jimenez, P., Saner, K., Mayhew, B., Rainey, W.E., 2003. GATA-6 is expressed in the human adrenal and regulates transcription of genes required for adrenal androgen biosynthesis. Endocrinology 144, 4285-4288.
Kananen, K., Markkula, M., Mikola, M., Rainio, E.M., McNeilly, A., Huhtaniemi, I., 1996. Gonadectomy permits adrenocortical tumori- genesis in mice transgenic for the mouse inhibin alpha-subunit pro- moter/simian virus 40 T-antigen fusion gene: evidence for negative autoregulation of the inhibin alpha-subunit gene. Mol. Endocrinol. 10, 1667-1677.
Ketola, I., Rahman, N., Toppari, J., Bielinska, M., Porter-Tinge, S.B., Tapanainen, J.S., Huhtaniemi, I.T., Wilson, D.B., Heikinheimo, M., 1999. Expression and regulation of transcription factors GATA-4 and GATA-6 in developing mouse testis. Endocrinology 140, 1470-1480. Kiiveri, S., Siltanen, S., Rahman, N., Bielinska, M., Lehto, V.P., Huh- taniemi, I.T., Muglia, L.J., Wilson, D.B., Heikinheimo, M., 1999. Re- ciprocal changes in the expression of transcription factors GATA-4 and GATA-6 accompany adrenocortical tumorigenesis in mice and humans. Mol. Med. 5, 490-501.
Kiiveri, S., Liu, J., Westerholm-Ormio, M., Narita, N., Wilson, D.B., Voutilainen, R., Heikinheimo, M., 2002. Differential expression of GATA-4 and GATA-6 in fetal and adult mouse and human adrenal tissue. Endocrinology 143, 3136-3143.
Kjellman, M., Roshani, L., Teh, B.T., Kallioniemi, O.P., Hoog, A., Gray, S., Farnebo, L.O., Holst, M., Backdahl, M., Larsson, C., 1999. Geno- typing of adrenocortical tumors: very frequent deletions of the MEN1 locus in 11q13 and of a 1-centimorgan region in 2p16. J. Clin. En- docrinol. Metab. 84, 730-735.
Kjellman, M., Larsson, C., Backdahl, M., 2001. Genetic background of adrenocortical tumor development. World J. Surg. 25, 948-956.
Kloos, R.T., Gross, M.D., Francis, I.R., Korobkin, M., Shapiro, B., 1995. Incidentally discovered adrenal masses. Endocr. Rev. 16, 460-484.
Kuo, C.T., Morrisey, E.E., Anandappa, R., Sigrist, K., Lu, M.M., Par- macek, M.S., Soudais, C., Leiden, J.M., 1997. GATA-4 transcription factor is required for ventral morphogenesis and heart tube formation. Genes Dev. 11, 1048-1060.
Laitinen, M.P., Anttonen, M., Ketola, I., Wilson, D.B., Ritvos, O., Bützow, R., Heikinheimo, M., 2000. Transcription factors GATA-4 and GATA- 6 and a GATA family cofactor, FOG-2, are expressed in human ovary and sex cord-derived ovarian tumors. J. Clin. Endocrinol. Metab. 85, 3476-3483.
Lassus, H., Laitinen, M.P., Anttonen, M., Heikinheimo, M., Aaltonen, L.A., Ritvos, O., Bützow, R., 2001. Comparison of serous and mu- cinous ovarian carcinomas: distinct pattern of allelic loss at distal 8p and expression of transcription factor GATA-4. Lab. Invest. 81, 517-526.
Laverriere, A.C., MacNeill, C., Mueller, C., Poelmann, R.E., Burch, J.B., Evans, T., 1994. GATA-4/5/6, a subfamily of three transcription fac- tors transcribed in developing heart and gut. J. Biol. Chem. 269, 23177-23184.
Lawson, M.A., Whyte, D.B., Mellon, P.L., 1996. GATA factors are essen- tial for activity of the neuron-specific enhancer of the gonadotropin- releasing hormone gene. Mol. Cell Biol. 16, 3596-3605.
Libertino, N.G., Libertino, J.M., 2003. Adrenocortical carcinoma: diag- nosis, evaluation and treatment. J. Urol. 169, 5-11.
Lin, L., Aggarwal, S., Glover, T.W., Orringer, M.B., Hanash, S., Beer, D.G., 2000. A minimal critical region of the 8p22-23 amplicon in esophageal adenocarcinomas defined using sequence tagged site- amplification mapping and quantitative polymerase chain reaction in- cludes the GATA-4 gene. Cancer Res. 60, 1341-1347.
Lin, S.R., Lee, Y.J., Tsai, J.H., 1994. Mutations of the p53 gene in hu- man functional adrenal neoplasms. J. Clin. Endocrinol. Metab. 78, 483-491.
Lucon, A.M., Pereira, M.A., Mendonca, B.B., Zerbini, M.C., Saldanha, L.B., Arap, S., 2002. Adrenocortical tumors: results of treatment and study of Weiss’s score as a prognostic factor. Rev. Hosp. Clin. Fac. Med. Sao Paulo 57, 251-256.
Mano, T., Luo, Z., Malendowicz, S.L., Evans, T., Walsh, K., 1999. Re- versal of GATA-6 downregulation promotes smooth muscle differen- tiation and inhibits intimal hyperplasia in balloon-injured rat carotid artery. Circ. Res. 84, 647-654.
Marx, C., Wolkersdorfer, G.W., Brown, J.W., Scherbaum, W.A., Born- stein, S.R., 1996. MHC class II expression-a new tool to assess dignity in adrenocortical tumours. J. Clin. Endocrinol. Metab. 81, 4488-4491.
Molkentin, J.D., Lin, Q., Duncan, S.A., Olson, E.N., 1997. Requirement of the transcription factor GATA-4 for heart tube formation and ventral morphogenesis. Genes Dev. 11, 1061-1072.
Morrisey, E.E., Ip, H.S., Lu, M.M., Parmacek, M.S., 1996. GATA-6: a zinc finger transcription factor that is expressed in multiple cell lineages derived from lateral mesoderm. Dev. Biol. 177, 309-322.
Morrisey, E.E., Tang, Z., Sigrist, K., Lu, M.M., Jiang, F., Ip, H.S., Par- macek, M.S., 1998. GATA-6 regulates HNF4 and is required for dif- ferentiation of visceral endoderm in the mouse embryo. Genes Dev. 12, 3579-3590.
Nagata, D., Suzuki, E., Nishimatsu, H., Yoshizumi, M., Mano, T., Walsh, K., Sata, M., Kakoki, M., Goto, A., Omata, M., Hirata, Y., 2000. Cyclin A downregulation and p21(cip1) upregulation correlate with GATA-6-induced growth arrest in glomerular mesangial cells. Circ. Res. 87, 699-704.
Narita, N., Heikinheimo, M., Bielinska, M., White, R.A., Wilson, D.B., 1996. The gene for transcription factor GATA-6 resides on mouse chromosome 18 and is expressed in myocardium and vascular smooth muscle. Genomics 36, 345-348.
Orkin, S.H., 1992. GATA-binding transcription factors in hematopoietic cells. Blood 80, 575-581.
Perlman, H., Suzuki, E., Simonson, M., Smith, R.C., Walsh, K., 1998. GATA-6 induces p21(Cip1) expression and G1 cell cycle arrest. J. Biol. Chem. 273, 13713-13718.
Peterson 2nd, R.A., Kiupel, M., Bielinska, M., Kiiveri, S., Heikinheimo, M., Capen, C.C., Wilson, D.B., 2004. Transcription factor GATA-4 is a marker of anaplasia in adrenocortical neoplasms of the domestic ferret (mustela putorius furo). Vet. Pathol. 41, 446-449.
Reincke, M., 1998. Mutations in adrenocortical tumors. Horm. Metab. Res. 30, 447-455.
Sabbaga, C.C., Avilla, S.G., Schulz, C., Garbers, J.C., Blucher, D., 1993. Adrenocortical carcinoma in children: clinical aspects and prognosis. J. Pediatr. Surg. 28, 841-843.
Sasano, H., Shizawa, S., Suzuki, T., Takayama, K., Fukaya, T., Morohashi, K., Nagura, H., 1995a. Ad4BP in the human adrenal cortex and its disorders. J. Clin. Endocrinol. Metab. 80, 2378-2380.
Sasano, H., Shizawa, S., Suzuki, T., Takayama, K., Fukaya, T., Moro- hashi, K., Nagura, H., 1995b. Transcription factor adrenal 4 binding protein as a marker of adrenocortical malignancy. Hum. Pathol. 26, 1154-1156.
Shibata, H., Ikeda, Y., Mukai, T., Morohashi, K., Kurihara, I., Ando, T., Suzuki, T., Kobayashi, S., Murai, M., Saito, I., Saruta, T., 2001. Ex- pression profiles of COUP-TF, DAX-1, and SF-1 in the human adrenal gland and adrenocortical tumors: possible implications in steroidoge- nesis. Mol. Genet. Metab. 74, 206-216.
Siltanen, S., Anttonen, M., Heikkilä, P., Narita, N., Laitinen, M., Ritvos, O., Wilson, D.B., Heikinheimo, M., 1999. Transcription factor GATA- 4 is expressed in pediatric yolk sac tumors. Am. J. Pathol. 155, 1823-1829.
Siltanen, S., Heikkilä, P., Bielinska, M., Wilson, D.B., Heikinheimo, M., 2003. Transcription factor GATA-6 is expressed in malignant endo- derm of pediatric yolk sac tumors and in teratomas. Pediatr. Res. 54, 542-546.
Soudais, C., Bielinska, M., Heikinheimo, M., MacArthur, C.A., Narita, N., Saffitz, J.E., Simon, M.C., Leiden, J.M., Wilson, D.B., 1995. Targeted mutagenesis of the transcription factor GATA-4 gene in mouse em- bryonic stem cells disrupts visceral endoderm differentiation in vitro. Development 121, 3877-3888.
Stojadinovic, A., Ghossein, R.A., Hoos, A., Nissan, A., Marshall, D., Dudas, M., Cordon-Cardo, C., Jaques, D.P., Brennan, M.F., 2002. Adrenocortical carcinoma: clinical, morphologic, and molecular char- acterization. J. Clin. Oncol. 20, 941-950.
Suzuki, E., Evans, T., Lowry, J., Truong, L., Bell, D.W., Testa, J.R., Walsh, K., 1996. The human GATA-6 gene: structure, chromosomal location, and regulation of expression by tissue-specific and mitogen- responsive signals. Genomics 38, 283-290.
Suzuki, T., Takahashi, K., Darnel, A.D., Moriya, T., Murakami, O., Narasaka, T., Takeyama, J., Sasano, H., 2000. Chicken ovalbu- min upstream promoter transcription factor II in the human adrenal cortex and its disorders. J. Clin. Endocrinol. Metab. 85, 2752- 2757.
Tremblay, J.J., Viger, R.S., 1999. Transcription factor GATA-4 enhances Müllerian inhibiting substance gene transcription through a direct interaction with the nuclear receptor SF-1. Mol. Endocrinol. 13, 1388-1401.
Tremblay, J.J., Viger, R.S., 2001. GATA factors differentially activate multiple gonadal promoters through conserved GATA regulatory ele- ments. Endocrinology 142, 977-986.
Tremblay, J.J., Hamel, F., Viger, R.S., 2002. Protein kinase A-dependent cooperation between GATA and CCAAT/enhancer-binding protein transcription factors regulates steroidogenic acute regulatory protein promoter activity. Endocrinology 143, 3935-3945.
Wajchenberg, B.L., Albergaria Pereira, M.A., Medonca, B.B., Latronico, A.C., Campos Carneiro, P., Alves, V.A., Zerbini, M.C., Liberman, B., Carlos Gomes, G., Kirschner, M.A., 2000. Adrenocortical carcinoma: clinical and laboratory observations. Cancer 88, 711-736.
Wechsler, J., Greene, M., McDevitt, M.A., Anastasi, J., Karp, J.E., Le Beau, M.M., Crispino, J.D., 2002. Acquired mutations in GATA-1 in the megakaryoblastic leukemia of Down syndrome. Nat. Genet. 32, 148-152.
Weiss, L.M., 1984. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am. J. Surg. Pathol. 8, 163-169.
Weiss, L.M., Medeiros, L.J., Vickery Jr., A.L., 1989. Pathologic features of prognostic significance in adrenocortical carcinoma. Am. J. Surg. Pathol. 13, 202-206.
White, R.A., Dowler, L.L., Pasztor, L.M., Gatson, L.L., Adkison, L.R., Angeloni, S.V., Wilson, D.B., 1995. Assignment of the transcription factor GATA-4 gene to human chromosome 8 and mouse chromosome 14: GATA-4 is a candidate gene for Ds (disorganization). Genomics 27, 20-26.
Viger, R.S., Mertineit, C., Trasler, J.M., Nemer, M., 1998. Transcrip- tion factor GATA-4 is expressed in a sexually dimorphic pattern during mouse gonadal development and is a potent activator of the Müllerian inhibiting substance promoter. Development 125, 2665- 2675.