Expression of aromatase and estrogen receptors in human adrenocortical tumors
Luisa Barzon . Giulia Masi . Monia Pacenti . Marta Trevisan . Francesco Fallo . Andrea Remo . Guido Martignoni . Daniela Montanaro · Vincenzo Pezzi · Giorgio Palù
Received: 30 July 2007 /Revised: 8 October 2007 / Accepted: 30 October 2007 / Published online: 21 December 2007 C Springer-Verlag 2007
Abstract We recently demonstrated that adrenocortical carcinoma cells express aromatase and estrogen receptors (ERs) and that 17ß-estradiol enhances adrenocortical cell proliferation. To provide a clue to the role of estrogens in adrenal tumorigenesis, we investigated the expression profile of genes involved in sex steroid hormone produc- tion and activity in a large series of normal and neoplastic human adrenocortical tissues. Quantitative reverse tran- scriptase-polymerase chain reaction, Western blotting, and immunohistochemistry showed that ER« and ERß, an- drogen receptor (AR), and aromatase were expressed in the adrenal cortex and in adrenocortical tumors. ERß was the predominant ER subtype and was mainly expressed in the zona glomerulosa and fasciculata. Western blot analysis revealed the presence of a truncated form of AR in adrenocortical tissues. With respect to the normal
adrenal cortex and adrenocortical adenomas, carcinomas were characterized by significantly lower ERß levels, ER« upregulation, and aromatase overexpression. ER expression correlated with expression of nuclear hormone receptors, suggesting they could be involved in ER modulation. In agreement with our in vitro findings, the results of this study suggest that estrogens, locally produced by aromatase, could enhance adrenocortical cell proliferation though autocrine/paracrine mechanisms. This study opens new perspectives on the potential use of antiestrogens and aromatase inhibitors as therapeutic agents against ACC.
Keywords Estrogen receptors · Androgen receptor . Adrenocortical tumors · Aromatase
L. Barzon · G. Masi · M. Pacenti · M. Trevisan · G. Palù Department of Histology, Microbiology and Medical Biotechnologies, University of Padova, Via A. Gabelli 63, 35121 Padova, Italy e-mail: luisa.barzon@unipd.it
F. Fallo Department of Medical and Surgical Sciences, University of Padova, Padova, Italy
A. Remo · G. Martignoni Department of Pathology, University of Verona, Verona, Italy
D. Montanaro · V. Pezzi Department of Pharmaco-biology, University of Calabria, Arcavacata di Rende, Cosenza, Italy
Introduction
Estrogens and androgens have been demonstrated to play a critical role in the etiology and progression of several endocrine-related malignancies, such as breast, ovarian, and prostate cancers. Some epidemiological and experimental studies suggest that sex steroid hormones could also be involved in adrenocortical tumorigenesis. In fact, adreno- cortical tumors are more frequent in women than in men, especially in those exposed to estro-progestins [5, 12]. Moreover, we recently demonstrated that 17ß-estradiol enhances proliferation of the human adrenocortical carci- noma (ACC) cell line NCI-H295R, whereas antiestrogens inhibit ACC cell growth [19]. At variance, androgens have been shown to inhibit NCI-H295R cell proliferation, an effect that is reverted by antiandrogens [27, 36].
To date, only a few studies have investigated expres- sion of estrogen receptors (ERs), androgen receptor
(AR), and aromatase-the enzyme responsible for estro- gen biosynthesis-in the human adrenal cortex and in adrenocortical tumors.
The effects of estrogens are mediated by two ER subtypes, ER& and ERß. While estrogens induce cancer cell proliferation through ER«, ERß seems to exert a protective effect [4]. ER expression has been studied in the human fetal adrenal cortex, where ERß is expressed at markedly higher levels than ER& [6, 34], and in pubertal human adrenal tissues, where ERß is detectable in the zona reticularis, whereas ER« is undetectable [3]. Only anec- dotal reports of ER expression in adrenal tumors are available in the literature [9, 22, 28]. Likewise, data on AR expression in the adrenal gland are limited to a few reports, which demonstrated the presence of AR messenger ribonucleic acid (mRNA) in normal and neoplastic human adrenocortical tissues [27]. The presence of aromatase has been recently demonstrated in the fetal adrenal gland [24], in the zona glomerulosa and medulla of the normal adrenal gland [3], in feminizing human adrenocortical tumors [25], and in NCI-H295R cells [19].
Expression and activity of ERs and AR are inhibited by the orphan nuclear hormone receptors DAX-1 (dosage- sensitive sex reversal, adrenal hypoplasia critical region, on chromosome X, gene 1) and SHP (small heterodimer partner) [11, 13, 37], which are also negative regulators of steroidogenic factor 1 (SF-1) and liver receptor homolog 1 (LRH-1) [33]. At variance, SF-1 and LRH-1 stimulate the expression of several steroidogeneic enzymes and nuclear hormone receptors. In breast adipose stromal cells, LRH-1 enhances aromatase expression, thus contributing to the growth of estrogen-positive breast tumors [7, 38].
Following our experimental studies on the effects of estrogens and antiestrogens on ACC cells and on the basis of the above evidence from the literature, in the present study, we investigated expression of genes involved in sex steroid hormone production and activity in normal and neoplastic human adrenocortical tissues to provide a clue to the role of estrogens in adrenal tumorigenesis.
Materials and methods
Adrenocortical samples
We investigated archived fresh-frozen or paraffin-embedded samples of adrenocortical tumors removed at surgery, normal adrenal cortex macroscopically dissected from adrenal glands of kidney donors, and fetal adrenal glands obtained at autopsy (Table 1). Tissue samples were obtained with the approval of local ethics committees and consent from patients, in accordance with the Declaration of Helsinki guidelines.
| Characteristics | Value |
|---|---|
| Normal adrenal cortex | N=14 |
| Female/male | 9/5 |
| Median age (range) | 45 (24-62) |
| Fetal adrenal gland | N=5 |
| Gestation weeks | 13, 16, 19, 20, 25 |
| Adrenocortical adenomas | N=33 |
| Female/male | 24/9 |
| Median age (range) | 42 (19-70) |
| Cortisol-producing | 4 |
| Aldosterone-producing | 13 |
| Androgen-producing | 1 |
| Nonfunctioning | 15 |
| Adrenocortical carcinomas | N=16 |
| Female/male | 11/5 |
| Median age (range) | 46 (27-69) |
| Cortisol-producing | 4 |
| Aldosterone-producing | 1 |
| Cortisol and androgen-producing | 5 |
| Nonfunctioning | 6 |
Diagnosis of malignancy was performed according to the histopathologic criteria proposed by Weiss et al. [35], with modification proposed by Aubert et al. [2]. Briefly, an adrenocortical neoplasm was defined as malignant in the presence of three or more of the following criteria: (1) high nuclear grade, according to the Fuhrman criteria [10], (2) greater than five mitoses per 50 high-power fields, (3) atypical mitotic figures, (4) less than 25% of tumor cells are clear cells, (5) diffuse architecture (>33% of tumor), (6) necrosis, (7) venous invasion, (8) sinusoidal invasion, and (9) capsular invasion.
Immunohistochemistry
Immunohistochemical evaluation of ER«, ERß, and AR protein expression was performed in a group of archive adrenocortical samples, including 16 ACCs, 5 normal adult adrenal glands, and 5 fetal normal adrenal glands. Specimens were fixed in 10% neutral buffered formalin. Immunohistochemical analysis was performed on depar- affinized sections pretreated for 30 min in 10 mM citrate buffer in steam. Mouse anti-human monoclonal anti- bodies to ER« (clone SP1, dilution 1:400, Lab-Vision, Neomarkers, Fremont, CA), ERß1 (clone PPG5/10, dilution 1:50, Serotec, Oxford, UK), and AR (clone AR441, dilution 1:20, Dako Italia S.p.A, Milan, Italy) were tested with a polymeric-labeling two-step method (Super sensitiveTM IHC detection system, Biogenex, San Ramon) in an automated staining system (GenoMx i6000, Biogenex). For ERß, an overnight incubation was done. Three carcinomas of the breast and three
adenocarcinomas of the prostate were used as external controls. Neoplastic cells were considered positive when they showed nuclear immunoreactivity, which was scored as follows: 0, undectable or equivocal individual single cells positivity; 1+, weak intensity; 2+, strong intensity. For each case, the percentage (%) of positive neoplastic cells was reported. Cytoplasmatic imunoreactivity, which was present in all cases of human adrenal glands, adrenal carcinomas, and in four of five fetal adrenal glands, was ignored.
Quantitative RT-PCR
Fifty-six adrenal tumor tissues, including 15 nonfunction- ing adrenocortical adenomas (NFAs), 4 cortisol-producing adenomas (CPAs), 13 aldosterone-producing adenomas (APAs), 1 androgen-producing adenoma, and 9 ACCs (six functioning, three nonfunctioning), and 14 normal adrenal cortex samples were studied. Reference RNA from human adrenal cortex (Human Adrenal Total RNA, Ambion, Austin, TX) was also employed as control. Total RNA was isolated using the RNeasy kit (Qiagen S.p.A. Milan, Italy), reverse transcribed, and subjected to quantitative real-time reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of ERa, ER3, AR, PGR (i.e., the progesterone receptor gene), CYP19 (i.e., the aromatase gene), SF-1, DAX-1, LRH-1, and SHP mRNA levels by using the oligonucleotide primer sequences reported in Table 2 and SYBR green reagents (Applied Biosystems, Foster City, CA). Absolute quantification was performed against a standard curve obtained by amplification of correspondent complementary deoxyribonucleic acids (DNAs) subcloned into the pCR2.1 vector (Invitrogen Life Technologies S.p.A, San Giuliano Milanese, Italy), as reported [8]. mRNA values are reported as copies per microgram total RNA, after statandardization against the
housekeeping gene RPLP0, which was also quantified by real-time RT-PCR.
Western blot analysis
Western blot analysis was performed as previously de- scribed [9, 19] in a subset of the above adrenocortical tissue samples, including two normal adrenal tissues, two CPAs, one APA, three NFAs, and four ACCs. Briefly, total protein extracts were separated on sodium dodecyl sulfate- polyacrylamide 11% gel and then electroblotted onto a nitrocellulose membrane. Blots were incubated overnight at 4℃ with anti-ERß antibody against the C-terminal region of ERß (1:500; Serotec), anti-ER« (F-10) against the C- terminal region of ER« (1:1,000; Santa Cruz Biotech, Santa Cruz, CA), mouse anti-human aromatase (1:1,000; Sero- tec); anti-SF-1 (1:2,000; Prof. Ken-ichirou Morohashi, National Institute for Basic Biology, Okazaki, Japan), anti-DAX-1 (2F4; 1:1,000) raised against amino acids 135-166 of human DAX-1 (Prof. Paolo Sassone-Corsi, School of Medicine University of California, Irvine, CA), and anti-AR (441; 1:1,000; Santa Cruz Biotech). Membranes were incubated with horseradish peroxidase-coniugated antibodies (Santa Cruz Biotech), and immunoreactive bands were visualized with ECL Plus Western blotting detection system (Amersham Pharmacia Biotech, UK) and quantified with Scion Immage software (Frederich, MD). As control loading, membranes were stripped and reprobed with actin antiserum.
Statistical analysis
Data are presented as mean±SD or median and range. Comparisons were performed by Mann-Whitney U test and Spearman Rank R correlation, setting the significance level to p<0.05 (Statistica 7.1, StatSoft Italia srl, Italy).
| Gene | Forward primer | Reverse primer | Amplicon length (bp) | Annealing temperature (°℃) |
|---|---|---|---|---|
| ER-a | CCACCAAACCAGTGCACCATT | GGTCTTTTCGTATCCCACCTTC | 108 | 64 |
| ER-ß | AGAGTCCCTGGTGTGAAGCAAG | GACAGCGCAGAAGTGAGCATC | 143 | 68 |
| AR | CCTGGCTTCCGCAACTTACAC | GGACTTGTGCATGCGGTACTCA | 168 | 64 |
| PGR | CGCGCTCTACCCTGCACTC | TGAATCCGGCCTCAGGTAGTT | 121 | 66 |
| CYP19 | TCACTGGCCTTTTTCTCTTGGT | GGGTCCAATTCCCATGCA | 83 | 60 |
| SF-1 | GGAGTTTGTCTGCCTCAAGTTCA | CGTCTTTCACCAGGATGTGGTT | 80 | 60 |
| DAX-1 | CCAAGGAGTACGCCTACCTCAA | ACTGGAGTCCCTGAATGTACTTCC | 90 | 60 |
| LRH-1 | TACCGACAAGTGGTACATGGAA | CGGCTTGTGATGCTATTATGGA | 89 | 60 |
| SHP | CCTCAATGCTGTCTGGAGTCCTT | CTGCAGGTGCCCAATGTG | 132 | 60 |
| RPLP0 | GGCGACCTGGAAGTCCAACT | CCATCAGCACCACAGCCTTC | 149 | 68 |
Results
Expression of sex steroid hormone receptors
Normal fetal and adult adrenal gland Immunohistochemical analysis of adult adrenal glands demonstrated heterogeneous expression of ERß (Fig. la-c) with nuclear positivity in the zona glomerulosa and fasciculata (score 1) but no expression in the zona reticularis and medulla. Fetal adrenal glands were negative for ERß expression, with the exception of a case with nuclear immunoreactivity (score 2) in all adrenal zones (Fig. 1d-f). Both adult and fetal normal adrenal glands were negative for AR and ER& immunostaining. The adenocarci- nomas of the prostate used as controls were positive for AR and the carcinomas of the breast immunoreacted for both ER& and ERß1 (data not shown).
In agreement with immunohistochemical results, Western blot analysis of two normal adrenal cortex samples demonstrated strong ERß expression and relatively low ER& levels (Fig. 2). As observed in MCF-7 control cells, ER& appeared as an immunopositive band corresponding to the full-length 66-kDa protein and a faint smaller band of 46 kDa, which could possibly represent the N-terminally deleted isoform of ER« with antagonist activity on ER& 66-kDa function [23]. In addition, ERß appeared as two bands, i.e., a faint band corresponding to a 55 kDa-protein variant and a more intense band corresponding to the expected full-length 59-kDa isoform of ERß1 [21]. At variance with immunohistochemistry, Western blot analysis of normal adrenocortical tissues allowed to detect the AR protein, which appeared as a double band with the highest of 75 kDa, whereas the full-length AR form of 110 kDa was absent. The apparent discrepancy between immunohis- tochemical and Western blot results could be explained by the relatively low levels of ER& and AR expression in the adrenal gland, which could be detected by Western blot but not by immunohistochemistry.
Quantitative RT-PCR analysis of sex steroid hormone receptors was performed in 14 normal adrenal cortex samples. It demonstrated variable levels of ERs, PGR, and AR mRNA (Fig. 3). In particular, mean ER& mRNA was 25,000±17,650 copies per µg total RNA; ERß mRNA levels were about twofold higher than ER& mRNA levels, in agreement with protein expression data, and there was a positive association between the amount of ERß mRNA and ER& mRNA levels (Spearman Rank R correlation, p< 0.05). AR mRNA was 17,700±8,300 copies per µg total RNA; PGR mRNA was 300,000±290,000 copies per µg total RNA.
Adrenocortical tumors Immunohistochemical analysis of sex steroid hormone receptors expression in ACCs demon-
strated that all ACCs were negative for ERx, with the exception of one case of ACC, which showed rare positive cells (Table 3). Western blot and RT-PCR analyses demonstrated higher levels of ER& protein and transcripts in ACCs than in the normal adrenal cortex (Figs. 2 and 3). At variance, quantitative real-time RT-PCR did not dem- onstrated any significant difference in ER& mRNA levels between normal adrenal cortex and adrenocortical adeno- mas, either functioning or nonfunctioning. No significant variation in the proportion of ER« protein isoforms was observed in the different tumor types at Western blot analysis.
Immunohistochemical staining demonstrated that 12 out of 16 (65%) ACCs were immunoreactive for ERß, including six ACCs composed by large, eosinophilic and pleomorphic score 2 cells and six ACCs composed of smaller cells that were frequently scored 1 (Table 3, Fig. 4a-d). A single tumor showing both types of cells showed both scores of intensity. The results of immuno- histochemistry were concordant with Western blot and RT- PCR analyses, which showed variable ERß expression among adrenocortical tumors. In fact, ERß levels were relatively low in several cases of ACCs and NFAs as compared to control normal adrenal tissues but increased in other cases. Statistical analysis demonstrated that, overall, ERß mRNA levels were significantly lower in ACCs than in the normal adrenal cortex (mean, 11,200±2,550 vs 44,580±8,050 copies per µg total RNA, p<0.01). More- over, in ACCs, the ER&/ERß ratio was significantly higher in ACCs than in the normal adrenal cortex (ERx/ERß ratio, 21.5±35.4 vs 0.51±0.29,p<0.05; Fig. 3), and ERß mRNA expression tended to be inversely related to ER« expres- sion, although the correlation was not statistically signifi- cant. PGR mRNA levels in ACCs were about fourfold higher (1,250,00±1,600,000 copies per µg total RNA) than in control normal adrenocortical tissues and correlated with that of ERa.
Six out of the 16 ACCs (38%) were positive for AR at immunohistochemical analysis and also displayed higher AR protein and mRNA levels than AR-negative ACCs at Western blot and quantitative RT-PCR analyses (Table 3). Among AR-positive ACCs, two displayed rare positive cells, whereas the other four cases were scored 2 with a percentage of positive cells ranging from 5 to 100%. (Table 3; Fig. 4e and f). AR expression was unrelated to any morphological features (e.g., mitotic rate, necrosis, vascular invasion, capsular invasion) present in ACCs. In benign adrenocortical tumor samples, Western blot and real-time RT-PCR analyses showed a wide variability of AR expression, with undetectable AR levels in four cases of NFAs and in the androgen-producing adenoma and high levels in four APAs (Figs. 2 and 3). Statistical analysis
Springer
a
b
C
d
e
f
ERa
+66 kDa
MCF7
46 kDa
ERB
+59 kDa
-MCF7
55 kDa
AR
+ 75 kDa
LnCap
CYP19
Plac
55 kDa
+55
SF-1
Plac
+50 kDa
DAX-1
+55 kDa
ß-actin
MCF7
+ 45 kDa
N1
N2
CPA1
CPA2
APA
NFA1
NFA2
NFA3
ACC1 ACC
C2 ACC3 C3
A
ACC4 M
C4 MCF7
demonstrated that, overall, AR mRNA levels were signif- icantly lower in NFAs (6,700±9,010 copies per ug total RNA) than in the normal adrenal cortex (p<0.01).
Expression of aromatase
Aromatase protein was expressed in normal adrenocortical tissues, although at lower levels than in human placenta, and was visualized at Western blot as a double band of 55 and 60 kDa. The small variation in molecular mass could be due to the level of glycosylation; however, glycosylation does not seem to have an impact on the enzymatic activity [20, 30]. Increased expression of aromatase protein and CYP19 mRNA was observed in ACCs, especially in cortisol- and androgen-secreting ACCs, than in the normal adrenal gland (16,090±9,140 vs 2,500±1,280 copies per µg total RNA; Figs. 2 and 3). CYP19 mRNA levels showed a positive association with ER& mRNA expression (Spearman Rank R correlation, p<0.05) but not with ERß expression.
Expression of orphan nuclear hormone receptors
In the normal adrenal cortex, SF-1 and DAX-1 were highly expressed, whereas LRH-1 and SHP mRNA levels were about tenfold lower (Fig. 3). Expression of SF-1, DAX-1, and LRH-1 was variable, with no significant differences between normal and neoplastic adrenal tissues, whereas SHP mRNA expression in APAs and NFAs was lower than in the normal adrenal cortex (Figs. 2 and 3). In the normal and neoplastic adrenal cortex, SF-1 and DAX-1 mRNA levels showed a positive correlation with LRH-1 mRNA but a negative correlation with SHP mRNA (p<0.05). As for sex steroid hormone receptors, ERß and ER& mRNAs
showed a positive association with LRH-1 expression but a negative correlation with SHP.
Discussion
This study represents the first investigation on expression of ERs and aromatase in the normal adrenal cortex and tumors derived thereof. It demonstrates that (1) ERs, AR, and aromatase are expressed in the human adrenal cortex and in adrenocortical tumors, (2) ERß is the predominant ER subtype in the adrenal cortex, (3) ACCs are character- ized by low ERß levels and/or increased amount of ER& and abnormal AR expression, (4) aromatase is expressed in the adrenal cortex and adrenocortical tumors and it is overexpressed in ACCs, and (5) expression of ERs correlates with nuclear hormone receptor expression.
By immunohistochemistry, Western blot, and real-time RT-PCR analyses, we demonstrated that both ERs are expressed in the normal adrenal cortex but that the amount of ERß is about twofold that of ER&. In adult normal adrenal cortex, we could localize ERß expression in the zona glomerulosa and fasciculata, at variance with findings in the prepubertal adrenal gland, where ERß was detectable in the zona reticularis [3]. ERß expression seems to be variable also during fetal adrenal development, as ERß has been detected in the definitive zone but not in the fetal zone of 20- and 38-gestational week fetuses [6], whereas, in neonates, ERß is expressed in the fetal zone but not in other zones [3]. In our experience, ERß was generally undetectable in fetal adrenal glands. Thus, the expression pattern of ERs in the adrenal cortex seems to change during adrenal development, suggesting a role for estrogens in this process.
1000000
art1oooo
100000
100000
*
10
*
9
8
10000
8
10000
X
ERa mRNA
ERB mRNA
1000
1000
100
100
10
10
1
1
0
0
Normal CPA APA NFA ACC
Normal
CPA
APA
NFA
ACC
100
100000
20
000
8
ERaJERß mRNA
18
9
10000
8
*
16
CYP19 mRNA
1000
p
14
*
12
a
10
100
8
10
6
4
1
2
8
0
£
8
8
0
Normal
CPA
APA
NFA
ACC
Normal
CPA
APA
NFA
ACC
100000
&
autoco
10000000
10000
8
1000000
adouo
ooooo
8
8
100000
&
AR mRNA
1000
8
PGR mRNA
10000
8
100
1000
10
100
10
1
1
0
0
Normal
CPA
APA
NFA
ACC
Normal
CPA
APA
NFA
ACC
1000000
1000000
100000
&
100000
oct.Do
оф
9
8
8
SF-1 mRNA
10000
DAX-1 mRNA
10000
8
1000
1000
100
100
10
10
1
1
Normal
CPA
APA
NFA
ACC
Normal CPA APA NFA ACO
10000
100000
ofco
8
10000
LRH-1 mRNA
1000
ano
P
8
8
8
SHP mRNA
8
1000
100
e
o
*
8
100
8
10
10
1
Normal CPA APA NFA ACC
1
Normal CPA APA NFA
ACC
| Immunohistochemical analysisª | Real-time RT-PCR analysisb | |||||
|---|---|---|---|---|---|---|
| ACC no. | ERa score | ERß1 score | AR score | ERa mRNA | ERß mRNA | AR mRNA |
| 1 | 0 | 2 (70%) | 0 | 7,640 | 27,900 | 22,280 |
| 2 | 0 | 1 (90%) | 2 (80%) | 16,940 | 13,280 | 9,370 |
| 3 | 0 | 1 (40%) | 0 | 34,240 | 9,920 | 4,900 |
| 4 | 0 | 0 | 0 | 25,770 | 7,190 | 17,430 |
| 5 | 0 | 2 (100%) | 0 | 12,300 | 12,800 | 7,410 |
| 6 | 0 | 1 (70%) | 2 (5%) | 39,320 | 4,020 | 42,830 |
| 7 | 0 | 2 (100%) | 0 | 16,570 | 24,150 | 10,690 |
| 8 | 0 | 2 (100%) | 2 (10%) | 18,420 | 10,760 | 61,080 |
| 9 | 0 | 2 (100%) | 0 (rare cells) | 86,560 | 4,780 | 14,460 |
| 10 | 0 (rare positive cells) | 0 | 0 (rare cells) | 91,800 | 1,200 | 28,000 |
| 11 | 0 | 0 | 2 (5%) | 51,860 | 508 | 50,440 |
| 12 | 0 | 1 (90%) | 2 (100%) | 84,500 | 27,890 | 98,000 |
| 13 | 0 | 1 (70%) | 0 | 10,100 | 3,120 | 3,760 |
| 14 | 0 | 2 (10%) | 0 | 15,750 | 8,620 | 6,890 |
| 15 | 0 | 0 | 2 (10%) | 6,460 | 1,750 | 5,340 |
| 16 | 0 | 1 (70%) | 0 | 91,150 | 997 | 15,070 |
a Neoplastic cells were considered positive when they showed nuclear immunoreactivity, which was scored as follows: 0 undectable or equivocal individual single cells positivity, 1+ weak intensity, and 2+ strong intensity. The percentage (%) of reactive neoplastic cells is reported within parenthesis. Cytoplasmatic imunoreactivity was ignored.
mRNA values are expressed as copies per ug total RNA.
Adrenocortical tumors were characterized by variable levels of ERs. Remarkably, several cases of ACCs showed lower ERß expression and/or higher ER& levels, resulting in increased ER&/ERß ratio, than the normal adrenal cortex. These results are in agreement with findings in other estrogen- dependent cancers, such as breast, ovary, colon, and prostate cancer, which show ER& overexpression and/or reduced ERß levels [4]. It is interesting to note that aromatase was also expressed at relatively high levels in the normal adrenal cortex and its expression markedly increased in ACCs. Based of these observations, we hypothesize that estrogens could contribute to adrenal tumorigenesis through an autocrine/paracrine loop. Estrogens could be produced locally in ACC cells by aromatase and exert their prolifer- ative effect through ER«, while the counteracting antiproli- ferative activity of ERß would be lost in cancer cells.
In this regard, we have recently demonstrated that H295R ACC cells express high levels of aromatase and are able to convert androgens to estrogens, which, through a short autocrine loop mediated by ER«, enhance H295R cell proliferation [19]. Treatment with the antiestrogens ICI 182,780 and 4-hydroxytamoxifen upregulated ERß expres- sion and inhibited basal and 17ß-estradiol-induced H295R cell proliferation. In particular, ICI 182,780 determined cell growth arrest, whereas treatment with 4-hydroxytamoxifen activated the Fas/FasL pathway, which in turn induced apoptosis. This effect could be due to 4-hydroxytamoxifen interaction with ERß and binding to the AP-1 site on the FasL promoter in H295R cells. A dose-dependent inhibi-
tion of H295R cell proliferation was also observed after treatment with the aromatase inhibitor Letrozole, thus further confirming the role of estrogens in H295R cell proliferation [19].
Androgens have been shown to inhibit H295R cell proliferation through their receptor [27]. This study demonstrates that AR was overexpressed in a subset of ACCs, whereas it was undetectable in some cases of NFA. It is interesting to note that at Western blot analysis, the full-length AR form of 110 kDa was absent, and AR appeared as a 75-kDa protein. The 75-kDa form has been also identified in prostate carcinoma and represents a truncated form of AR, generated by proteolytic cleavage [18]. This truncated AR has constitutive transcriptional activity, as it lacks a ligand binding domain, but it can bind androgen-responsive elements through its DNA-binding domain. We hypothesize that in adrenocortical cells, this truncated AR with androgen-independent activity could compete with the androgen-dependent negative regulation of adrenocortical cell growth. An imbalance toward the androgen-independent pathway could contribute to unre- strained cell proliferation in ACCs. This hypothesis is being investigated with experiments that are ongoing.
Finally, we investigated expression of orphan nuclear receptors to find clues on the regulation of sex steroid hormone receptors and aromatase expression in adrenocor- tical cells. We demonstrated variable expression of SF-1, DAX-1, LRH-1, and SHP in the adrenal cortex and its tumors, with increased DAX-1 and SF-1 levels in ACCs
a
b
C
d
*
e
f
and decreased SHP levels in APAs and NFAs. Variable SF- 1 and DAX-1 expression in adrenocortical tumors has been reported also in the literature [17, 26, 29, 31, 32], whereas no data are available on LRH-1 and SHP expression. The positive correlation between LRH-1 expression and ERs mRNA levels, as well as the negative association between SHP and ERs, suggest that LRH-1 and SHP might be transcriptional targets of ERs, as shown in other cell types
[1, 16]. On the other hand, increased LRH-1 expression and decreased SHP expression in ER-positive adrenal tumors could enhance estrogen-mediated effects on cell prolifera- tion, as LRH-1 has been shown to induce aromatase expression [7, 38], whereas SHP has been demonstrated to repress ER& and ERß transcriptional activity [14, 15].
In conclusion, our results demonstrate that ACCs are characterized by an imbalance between ER&x and ERß
levels and increased aromatase expression. Abnormal expression of AR and orphan nuclear hormone receptors could also contribute to the effects of estrogens in adrenocortical tumorigenesis. Because estrogens have a proliferative effect on ACC cells, the results of this study suggest that estrogens could contribute to adrenocortical tumorigenesis though autocrine/paracrine effects. This study opens new perspectives on the potential use of antiestrogens and aromatase inhibitors as therapeutic agents against ACC.
Acknowledgments Giulia Masi is a recipient of a fellowship from IRCCS-IOV (Instituto Oncologico Veneto). This work was supported by AIRC funds to Giorgio Palù.
References
1. Annicotte JS, Chavey C, Servant N, Teyssier J, Bardin A, Licznar A, Badia E, Pujol P, Vignon F, Maudelonde T, Lazennec G, Cavailles V, Fajas L (2005) The nuclear receptor liver receptor homolog-1 is an estrogen receptor target gene. Oncogene 24:8167-8175
2. Aubert S, Wacrenier A, Leroy X, Devos P, Carnaille B, Proye C, Wemeau JL, Lecomte-Houcke M, Leteurtre E (2002) Weiss system revisited: a clinicopathologic and immunohistochemical study of 49 adrenocortical tumors. Am J Surg Pathol 26:1612- 1619
3. Baquedano MS, Saraco N, Berensztein E, Pepe C, Bianchini M, Levy E, Goni J, Rivarola MA, Belgorosky A (2007) Identification and developmental changes of aromatase and estrogen receptor expression in prepubertal and pubertal human adrenal tissues. J Clin Endocrinol Metab 92:2215-2222
4. Bardin A, Boulle N, Lazennec G, Vignon F, Pujol P (2004) Loss of ERß expression as a common step in estrogen-dependent tumor progression. Endocr-Relat Cancer 11:537-551
5. Barzon L, Sonino N, Fallo F, Palù G, Boscaro M (2003) Prevalence and natural history of adrenal incidentalomas. Eur J Endocrinol 149:273-285
6. Brandenberger AW, Tee MK, Lee JY, Chao V, Jaffe RB (1997) Tissue distribution of estrogen receptors « (ER-x) and ß (ER-ß) mRNA in the midgestational human fetus. J Clin Endocrinol Metab 82:3509-3512
7. Clyne CD, Speed CJ, Zhou J, Simpson ER (2002) Liver receptor homologue-1 (LRH-1) regulates expression of aromatase in preadipocytes. J Biol Chem 277:20591-20597
8. Fallo F, Pezzi V, Barzon L, Mulatero P, Veglio F, Sonino N, Mathis JM (2002) Quantitative assessment of CYP11B1 and CYP11B2 expression in aldosterone-producing adenomas. Eur J Endocrinol 147:795-802
9. Fallo F, Pezzi V, Sonino N, Altavilla G, Barzon L (2005) Adrenal incidentaloma in pregnancy. A case report. J Endocrinol Investig 28:459-463
10. Greene LF, Page DL, Fritz A, Batch M, Haller DG, Morrow M (2002) American Joint Committee on Cancer (AJCC) Cancer staging manual, 6th edn. Springer, New York
11. Holter E, Kotaja N, Makela S, Strauss L, Kietz S, Janne OA, Gustafsson JA, Palvimo JJ, Treuter E (2002) Inhibition of androgen receptor (AR) function by the reproductive orphan nuclear receptor DAX-1. Mol Endocrinol 16:515-528
12. Hsing AW, Nam JM, Co Chien HT, Mclaughlin JK, Fraumeni JF Jr (1996) Risk factors for adrenal cancer: an exploratory study. Int J Cancer 65:432-436
13. Iyer AK, Zhang YH, McCabe ER (2006) Dosage-sensitive sex reversal adrenal hypoplasia congenita critical region on the X chromosome, gene 1 (DAX1) (NR0B1) and small heterodimer partner (SHP) (NR0B2) form homodimers individually, as well as DAX1-SHP heterodimers. Mol Endocrinol 20:2326-2342
14. Johansson L, Thomsen JS, Damdimopoulos AE, Spyrou G, Gustafsson JA, Treuter E (1999) The orphan nuclear receptor SHP inhibits agonist-dependent transcriptional activity of estrogen receptors ER& and ERB. J Biol Chem 274:345-353
15. Klinge CM, Jernigan SC, Risinger KE (2002) The agonist activity of tamoxifen is inhibited by the short heterodimer partner orphan nuclear receptor in human endometrial cancer cells. Endocrinology 143:853-867
16. Lai K, Harnish DC, Evans MJ (2003) Estrogen receptor alpha regulates expression of the orphan receptor small heterodimer partner. J Biol Chem 278:36418-36429
17. Lefrancois-Martinez AM, Bertherat J, Val P, Tournaire C, Gallo- Payet N, Hyndman D, Veyssiere G, Bertagna X, Jean C, Martinez A (2004) Decreased expression of cyclic adenosine monophosphate- regulated aldose reductase (AKR1B1) is associated with malignancy in human sporadic adrenocortical tumors. J Clin Endocrinol Metab 89:3010-3019
18. Libertini SJ, Tepper CG, Rodriguez V, Asmuth DM, Kung HJ, Mudryj M (2007) Evidence for calpain-mediated androgen receptor cleavage as a mechanism for androgen independence. Cancer Res 67:9001-9005
19. Montanaro D, Maggiolini M, Recchia AG, Barzon L, Fallo F, Andò S, Pezzi V (2005) Antiestrogens upregulate ERß expression and inhibit human adrenocortical cell proliferation. J Mol Endocrinol 35:245-256
20. Moslemi S, Vibet A, Papadopoulos V, Camoin L, Silberzahn P, Gaillard JL (1997) Purification and characterization of equine testicular cytochrome P-450 aromatase: comparison with the human enzyme. Comp Biochem Physiol B Biochem Mol Biol 118:217-227
21. Murphy LC, Watson PH (2006) Is oestrogen receptor-ß a predictor of endocrine therapy responsiveness in human breast cancer. Endocr Relat Cancer 13:327-334
22. Nakao M, Sato B, Koga M, Noma K, Kishimoto S, Matsuda M, Sonoda T, Matsumoto K (1986) Biochemical and immunological characterization of estrogen binding components in human neoplastic adrenocortical tissues. J Steroid Biochem 25:1-9
23. Penot G, Le Peron C, Merot Y, Grimaud-Fanouillere E, Ferriere F, Boujrad N, Kah O, Saligaut C, Ducouret B, Metivier R, Flouriot G (2005) The human estrogen receptor-« isoform hER«46 antago- nizes the proliferative influence of hER&x66 in MCF7 breast cancer cells. Endocrinology 146:5474-5484
24. Pezzi V, Mathis JM, Rainey WE, Carr BR (2003) Profiling transcript levels for steroidogenic enzymes in fetal tissues. J Steroid Biochem Mol Biol 87:181-189
25. Phornphutkul C, Okubo T, Wu K, Harel Z, Tracy TF, Pinar H, Chen S, Gruppuso PA, Goodwin G (2001) Aromatase P450 expression in feminizing adrenal adenoma presenting as isosexual precocious puberty. J Clin Endocrinol Metab 86:649-652
26. Reincke M, Beuschlein F, Lalli E, Arlt W, Vay S, Sassone-Corsi P, Allolio B (1998) DAX-1 expression in human adrenocortical neoplasms: implications for steroidogenesis. J Clin Endocrinol Metab 83:2597-2600
27. Rossi R, Zatelli MC, Valentini A, Cavazzini P, Fallo F, del Senno L, degli Uberti EC (1998) Evidence for androgen receptor gene expression and growth inhibitory effect of dihydrotestosterone on human adrenocortical cells. J Endocrinol 159:373-380
28. Sanfilippo JS, Wittliff JL (1984) Steroid hormone receptors in adrenal cortical carcinoma. Am J Obstet Gynecol 150:326-327
29. Sasano H, Shizawa S, Suzuki T, Takayama K, Fukaya T, Morohashi K, Nagura H (1995) Ad4BP in the human adrenal cortex and its disorders. J Clin Endocrinol Metab 80:2378-2380
30. Sethumadhavan K, Bellino FL, Thotakura NR (1991) Estrogen synthetase (aromatase). The cytochrome P-450 component of the human placental enzyme is a glycoprotein. Mol Cell Endocrinol 78:25-32
31. Shibata H, Ando T, Suzuki T, Kurihara I, Hayashi K, Hayashi M, Saito I, Murai M, Saruta T (1998) COUP-TFI expression in human adrenocortical adenomas: possible role in steroidogenesis. J Clin Endocrinol Metab 83:4520-4523
32. Shibata H, Ikeda Y, Mukai T, Morohashi K, Kurihara I, Ando T, Suzuki T, Kobayashi S, Murai M, Saito I, Saruta T (2001) Expression profiles of COUP-TF, DAX-1, and SF-1 in the human adrenal gland and adrenocortical tumors: possible implications in steroidogenesis. Mol Genet Metab 74:206-216
33. Sirianni R, Seely JB, Attia G, Stocco DM, Carr BR, Pezzi V, Rainey WE (2002) Liver receptor homologue-1 is expressed
in human steroidogenic tissues and activates transcription of genes encoding steroidogenic enzymes. J Endocrinol 174:R13- R17
34. Takeyama J, Suzuki T, Inoue S, Kaneko C, Nagura H, Harada N, Sasano H (2001) Expression and cellular localization of estrogen receptors & and ß in the human fetus. J Clin Endocrinol Metab 86:2258-2262
35. Weiss LM, Medeiros LJ, Vickery AL Jr (1989) Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 13:202-206
36. Zatelli MC, Rossi R, degli Uberti EC (2000) Androgen influences transforming growth factor-ß1 gene expression in human adreno- cortical cells. J Clin Endocrinol Metab 85:847-852
37. Zhang H, Thomsen JS, Johansson L, Gustafsson JA, Treuter E (2000) DAX-1 functions as an LXXLL-containing corepressor for activated estrogen receptors. J Biol Chem 275:39855-39859
38. Zhou J, Suzuki T, Kovacic A, Saito R, Miki Y, Ishida T, Moriya T, Simpson ER, Sasano H, Clyne CD (2005) Interactions between prostaglandin E2, liver receptor homologue-1, and aromatase in breast cancer. Cancer Res 65:657-663