0021-972X/07/$15.00/0 Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 92(6):2215–2222 Copyright © 2007 by The Endocrine Society doi: 10.1210/jc.2006-2329
Identification and Developmental Changes of Aromatase and Estrogen Receptor Expression in Prepubertal and Pubertal Human Adrenal Tissues Marı´a Sonia Baquedano, Nora Saraco, Esperanza Berensztein, Carolina Pepe, Michele Bianchini, Estrella Levy, Javier Gon˜i, Marco A. Rivarola, and Alicia Belgorosky Molecular Biology Laboratory (M.S.B., N.S., E.B., C.P., J.G., M.A.R., A.B.), Endocrinology Department, Garrahan Pediatric Hospital, Buenos Aires, Argentina C1245AAM; and Centro de Investigaciones Oncolo´gicas (M.B., E.L.), Centro de Investigaciones Oncolo´gicas, Fundacio´n para la Investigacio´n y Prevencio´n del Cancer, Buenos Aires, Argentina C1426ANZ Context: The mechanisms of postnatal adrenal zonation remain unclear. Objective: To provide a clue for a possible role of estrogens in adrenarche, we studied the expression of estrogen receptor (ER)␣, ER, G protein-coupled receptor (GPR)30, and cP450aromatase (cP450arom) in human adrenal tissue. Design: Human adrenal tissue was collected from three postnatal age groups (Grs): Gr 1, younger than 3 months (n ⫽ 12), fetal zone involution; Gr 2, 3 months to 6 yr (n ⫽ 17), pre-adrenarche; and Gr 3, older than 6 –20 yr (n ⫽ 12), post-adrenarche period. Results: ER mRNA in Grs 1 and 3 was higher than in Gr 2 (P ⬍ 0.05). By immunohistochemistry and laser capture microdissection followed by RT-PCR, ER was expressed in zona reticularis and fetal zone, GPR30 in zona glomerulosa (ZG) and adrenal medulla, while ER␣ mRNA and protein were undetectable. cP450arom mRNA in Gr
A
DRENARCHE IS THE maturational activation of dehydroepiandrosterone (DHEA) and DHEA sulfate adrenal androgen secretion. This event of postnatal sexual maturation occurs only in higher primates (1), typically between 6 and 8 yr of age in humans, when the innermost layer of adrenal cortex, the zona reticularis (ZR), develops. Differentiation of the ZR cells implies a decrease in 3-hydroxysteroid dehydrogenase (3HSD) type II expression, an increase in 17,20 lyase activity of 17␣-hydroxylase (P450c17) due to an increase in cytochrome b5 association, along with an increase in sulfotransferase expression, which results in an increase of adrenal androgen output (1– 4). However, the physiological mechanisms controlling the differentiation of the ZR and human adrenal androgen secretion remain unknown, suggesting that adrenarche is probably the result of the interplay of several factors. Adrenal cortex and adrenal medulla are related ontogenFirst Published Online April 3, 2007 Abbreviations: AU, Arbitrary unit; cP450arom, cP450aromatase; DHEA, dehydroepiandrosterone; ER, estrogen receptor; FeZ, fetal zone; GPR, G protein-coupled receptor; Gr, group; 3HD, 3-hydroxysteroid dehydrogenase; LCM, laser capture microdissection; ZF, zona fasciculata; ZG, zona glomerulosa; ZR, zona reticularis. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.
3 was higher than in Grs 1 and 2 (P ⬍ 0.05), and localized to ZG and adrenal medulla by laser capture microdissection. cP450arom Immunoreactivity was observed in adrenal medulla in the three Grs and in subcapsular ZG of Gr 3. Double-immunofluorescence studies revealed that cP450arom and chromogranin A only colocalize in adrenal medulla of subjects younger than 18 months. In these samples, exon 1.b-derived transcript was 3.5-fold higher, while exon 1.a-, 1.c-, and 1.d-derived transcripts were 3.3-, 1.9-, and 1.7-fold lower, respectively, than in subjects older than 6 yr. Conclusions: Our results suggest that estrogens produced locally in adrenal medulla would play a role in zona reticularis functional differentiation through ER. The cP450arom and GPR30 expression in subcapsular ZG, colocalizing with a high-cell proliferation index, previously reported, suggests a local GPR30-dependent estrogen action in proliferation and migration of progenitor adrenal cells. (J Clin Endocrinol Metab 92: 2215–2222, 2007)
ically, anatomically, and functionally (5, 6). These two adrenal systems influence their respective secretory activities in a paracrine fashion, suggesting that adrenomedullary function might be linked to the process of adrenarche (7). In vitro and in vivo studies showed that estrogens might increase adrenal androgen production, suggesting that estrogen may modulate the ⌬5 pathway of steroidogenesis (8 –11). Estrogen biosynthesis is catalyzed by the enzyme cP450aromatase (cP450arom), the product of the CYP19 gene. In humans, it is expressed not only in gonadal but also in extra-gonadal tissues (12), where it exerts significant local biological influence (13). The CYP19 gene is tissue-specifically regulated through the alternative use of multiple exons 1 (12). A unique promoter region containing basic and regulatory promoter or enhancer elements flanks each tissuespecific exon 1 (12). Previous reports of cP450arom expression in normal human adult adrenal gland were controversial (14 –17). However, the presence of cP450arom was recently shown in the human adrenocortical cell line H295R (18) and in feminizing human adrenal tumors (15, 17), as well as in the human fetal adrenal gland (19). The effect of estrogens on target tissues are mediated by nuclear estrogen receptors (ERs) ER␣ and ER, but they also act through nongenomic processes involving membrane proteins comparable to classical ER or nonclassical receptors,
2215
2216
J Clin Endocrinol Metab, June 2007, 92(6):2215–2222
such as G protein-coupled receptor (GPR)30 (20 –22). ER immunoreactivity has been detected in the neonatal and adult adrenal glands of the rhesus monkey (23), and ER mRNA was reported in whole human fetal adrenal gland (24). However, the developmental expression and cortical zone-specific localization of ER␣ and ER in the postnatal human adrenal gland has not been determined. To provide a clue to estrogen biological actions at adrenarche, we studied the ontogenesis of ER␣, ER, and cP450arom mRNA expression and immunolocalization in human adrenal tissues from early infancy to puberty, as well as the alternative use of different exons 1 of the cP450arom gene. Subjects and Methods Tissue samples For the present study, the human adrenal tissue samples, analyzed in a previous report (25), were used. Briefly, 41 nonpathological human adrenal glands were obtained from multiorgan transplantation donors with less than 24 h after the diagnosis of brain death or patients undergoing resection of the kidney plus adrenal because of renal neoplasm (n ⫽ 18), or from necropsies with less than 6 h of postmortem time (n ⫽ 24). Histological identification of the adrenal zones has already been published (26). As previously described (25), for analysis of the results, subjects were divided into three age groups (Grs): Gr 1, younger than 3 months (n ⫽ 12); Gr 2, between 3 months and 6 yr old (n ⫽ 17); and Gr 3, older than 6 up to 20 yr (n ⫽ 12). This division was based on the mean age of postnatal fetal zone (FeZ) involution (27) for the first cutoff and the mean age of adrenarche (1) for the second. In Gr 3, only those samples, in which a continuous ZR was present, were included. For immunolocalization and exon-derived transcript studies of cP450arom, definition of age Grs was different (see later) after the particular maturation pattern of the adrenal medulla (27). After removing adherent fatty tissue, segments of adrenals were fixed in 4% paraformaldehyde and embedded in paraffin for histological examination or promptly frozen and stored in liquid nitrogen for subsequent RNA analysis. Donors or donor families gave consent for the use of organs for research and transplantation. The Ethical Committee of the Garrahan Pediatric Hospital approved the study.
RNA isolation Intact adrenal tissues. Total RNA was extracted from each adrenal tissue sample by homogenization in TRIzol reagent (Invitrogen, Buenos Aires, Argentina) following the manufacturer’s instructions. Purity and integrity of each total RNA sample were evaluated spectroscopically and by gel electrophoresis before RT. Total RNA concentration was assessed by spectrophotometric absorbance at 260 nm. Cells isolated by laser capture microdissection (LCM). Sections (5 m) of frozen adrenal tissues (n ⫽ 4, from boys aged 14, 15, and 18 yr, and a woman aged 20 yr) were mounted on slides for LCM (steel frames with polyethylene naphthalate; Leica Microsystems, Bensheim, Germany). Slides were fixed in methacarn (60% vol/vol absolute methanol, 30% vol/vol chloroform, and 10% vol/vol glacial acetic acid), hematoxylin and eosin stained, and dehydrated in rinsing ethanol and xylene. Cells from the ZR, zona fasciculata (ZF), zona glomerulosa (ZG), and adrenal medulla were captured using the AS LMD microscope (Leica Microsystems). Total RNA was extracted from the microdissected samples using the RNAqueous-Micro kit (AM-1931; Ambion Inc., Austin, TX) according to the manufacturer’s instructions.
RNA analysis RT-PCR of ER␣ and ER. Total RNA of intact adrenal tissues was reverse transcribed using MMLV reverse transcriptase (Amersham Biosciences, Buenos Aires, Argentina) following the manufacturer’s instructions. The RT products were then pooled and amplified by PCR with specific primers for ER␣ and ER. Each primer pair was localized on different
Baquedano et al. • Estrogens and Adrenal Development
exons to discriminate the products from genomic DNA and cDNA. The PCR reaction was performed using 2 l cDNA pool as a template in 25 l reaction volume containing 1.5 mm MgCl2, 200 m dNTPs, 0.24 m of each forward and reverse primer, and 1 U Taq polymerase (Amersham Biosciences). Negative controls lacking cDNA were included in all PCR reactions. Human ER␣ and ER cDNA clones (plasmids kindly supplied by Dr. K. Korach, National Institute of Environmental Sciences, Research Triangle Park, NC) were used as positive controls. PCR products were analyzed on 2% agarose gels containing ethidium bromide and visualized using a UV transilluminator. The identities of RT-PCR products were verified by sequencing analysis.
Semiquantitative RT-PCR of ER and cP450arom The cDNA of each adrenal tissue sample was amplified by PCR. The sequences of cP450arom and -actin primers were previously published (28, 29). Human -actin gene, which is expressed at stable levels regardless of age (4), was used as a housekeeping gene to control PCR reactions, and normalize mRNA levels of ER and cP450arom. Optimal PCR cycles required for linear amplification of each primer set were determined. Total amplification in each reaction was kept below saturation levels. A total of 0.12 m sense primer, 0.08 m 5⬘-end 32P-labeled sense primer, and 0.2 m antisense primer was added at the start of PCR. The thermocycler was programmed for 10 cycles for ER and 11 cycles for cP450arom; -actin primers were then added, and cycling was continued for an additional 16 and 19 cycles, respectively. The amplified products were separated by electrophoresis on 5% acrylamide 7-m urea gels and visualized by autoradiography. The intensities of the bands were quantified by densitometric analysis and normalized to that of -actin (Molecular Image System, Bio-Rad GS-505; Bio-Rad Laboratories, Hercules, CA). Each sample was analyzed at least twice, in duplicates, at two different dilutions, each time. Negative controls lacking cDNA were included in all PCR reactions. The identity of RT-PCR products was verified by sequence analysis. Within- and between-assay coefficients of variation were 11.4% and 13.6%, and 10.9% and 12.8% for mRNA abundance of cP450arom and ER, respectively. No significant difference between ER and cP450arom mRNA abundance in samples of Gr 2 collected from necropsies (n ⫽ 9) or organ donors (n ⫽ 8) was found.
RT-PCR of cP450arom, ER, and GPR30 in samples from LCM A total of 10 l RNA extract from medullary and cortical lasercaptured cells was reverse transcribed using random hexamers (Biodynamics SRL, Buenos Aires, Argentina) and 200 U reverse transcriptase (Superscript II; Invitrogen) and then amplified by PCR. PCR was performed for 35 cycles using specific intron-spanning primer pairs for 3HSD, P450c17, and chromogranin A, and 40 cycles for P450arom, ER, and GPR30. PCR amplification of -actin (30 cycles) was performed to check the quality of cDNA generated from the laser-captured samples.
Analysis of alternative usage of exons 1 in cP450arom mRNA The use of alternative exons 1 of cP450arom gene in adrenal tissue was investigated by RT-PCR. As previously reported (27), chromaffin cell functional maturation is not complete before the age of 18 months. Therefore, cDNA from subjects younger than 18 months was pooled separately from that of subjects older than 6 yr, before amplification by PCR. P450arom PCR was performed using a sense primer specific for exons 1a (placenta-type exon 1), 1b (adipose tissue/skin䡠fetal liver䡠bonetype exon 1), 1c (ovary-type exon 1), 1d (ovary/testis䡠prostate-type exon 1) (30), or 1.6 (bone-type exon 1) (31) and a 32P-labeled antisense primer, specific for exon 3. Optimal PCR cycles required for linear amplification for each primer pair were determined. Positive (18, 30, 31) and negative controls lacking cDNA were included in all PCR reactions. As an internal control, RT-PCR reactions amplifying the -actin mRNA were included. The PCR products were visualized by autoradiography after electrophoresis on 5% acrylamide 7-m urea gels. The identity of RT-PCR products was verified by sequence analysis.
Baquedano et al. • Estrogens and Adrenal Development
J Clin Endocrinol Metab, June 2007, 92(6):2215–2222
2217
FIG. 1. mRNA expression analysis of ER␣, ER, cP450arom, and GPR30 in normal human adrenal tissues. A, RT-PCR analysis of ER␣ (lanes 1– 4) and ER (lanes 6 –9). PCR amplification with primers located at the N terminal A/B region for ER␣ (sense 5⬘-aggctgcggcgttcggc-3⬘, antisense 5⬘-agccatacttcccttgtcat-3⬘, 272 bp) and ER (sense 5⬘-ttcccagcaatgtcactaact-3⬘, antisense 5⬘-ctctttgaacctggaccagta-3⬘, 260 bp) revealed that ER but not ER␣ mRNA was detected in a pool of human prepubertal and pubertal adrenal glands (Ad). Human ER␣ and ER cDNA clones (lanes 1 and 7, respectively) were used as positive controls (⫹C). Each reaction consisted of 40 cycles (1 min at 94 C, 1 min at 56 C, and 1 min at 72 C); no product was amplified in reactions that did not contain cDNA (n). M, 100-bp marker. B and C, Semiquantitative RT-PCR analysis of ER and cP450arom, respectively, in normal human adrenal glands of the three age Grs. Gr 1, n ⫽ 12; Gr 2, n ⫽ 17; and Gr 3, n ⫽ 12. The bars show mean mRNA levels, quantified densitometrically in relation to -actin mRNA (AUs); the error bars represent SD in each Gr. Results are expressed as mean ⫾ SD; *, P ⬍ 0.05. D, Zonal expression of mRNA encoding cP450arom, ER, and GPR30 in ZR, ZF, ZG, and adrenal medulla cells of human adrenal tissue from Gr 3 recovered by LCM. Da and Db, A representative photomicrograph of a 14-yr-old human adrenal gland before and after LCM of the ZR (Da) and adrenal medulla (Db). Dc, Total RNA was extracted from cells of the respective zones and analyzed by RT-PCR. Each panel is labeled according to the primers used as follows: ER (sense 5⬘-ttcccagcaatgtcactaact-3⬘, antisense 5⬘-ctctttgaacctggaccagta-3⬘); cP450arom (sense 5⬘-gaatattggaaggatgcacagact-3⬘, antisense 5⬘-gggtaaagatcatttccagcatgt-3⬘); GPR30 (sense 5⬘-ggctttgtgggcaacatc-3⬘; antisense 5⬘-cggaaagactgcttgcagg-3⬘); 3HSD (sense 5⬘-gggctggagctgccttgtga-3⬘, antisense 5⬘-tcgtggcgttctggatgat-3⬘); P450c17 (sense 5⬘-tggagaccaccacctctgtg-3⬘, antisense 5⬘-ggaggagacggttacggtca-3⬘); chromogranin A (Chrom A) (sense 5⬘-catctccgacacactttccaag-3⬘, antisense 5⬘-tcctctcttttctccataacatcc-3⬘); and -actin (sense 5⬘-ggacctgactgactacctcatgaa-3⬘, antisense 5⬘-gatccacatctgctggaaggtgg-3⬘, 524 bp). ER mRNA was observed primarily in the ZR, while cP450arom and GPR30 in ZG and adrenal medulla. Negative controls (no input cDNA) were run with all reactions (n). M, 100-bp marker. Experiments were reproduced with cDNAs from four adrenal tissues (15, 16, 21, and 22 yr of age). AM, Adrenal medulla.
Immunohistochemistry of ER␣, ER, and cP450arom Immunohistochemistry staining was performed using the streptavidinbiotin and peroxidase method using the manufacturer’s protocol (Dako Catalyzed Signal Amplification System, HRP; DakoCytomation, Carpinteria, CA), as we have previously described (25). Briefly, after deparaffinization, sections (5 m) were subjected to antigen retrieval and proteinase K (20 g/ml) treatment (10 min). Endogenous biotin activity was blocked, and endogenous peroxidase activity was quenched. The sections were further blocked (Protein Block Serum-Free, Dako X0909; DakoCytomation) for 15 min. Sections were then incubated with the primary antibody
for ER␣ (3 g/ml, sc-8002; Santa Cruz Biotechnology, Inc.), ER (1 g/mL, sc-8974; Santa Cruz Biotechnology, Inc.), cP450arom (1/1000; HauptmanWoodward Medical Research Inst., Buffalo, NY), or GPR30 (4 g/ml, Lot 9096, kindly provided by Dr. E. Filardo, Brown University, Providence, RI) for 18 h, at 4 C. After washing, tissues were incubated for 15 min with biotinylated goat antirabbit (ER, cP450arom, and GPR30) or biotinylated rabbit antimouse (ER␣) Igs. Bound antibodies were visualized with 3,3⬘diaminobenzidine tetrahydrochloride. As negative controls, normal rabbit serum (for Er, cP450arom, and GPR30) or normal mouse serum (for ER␣) was used instead of primary antibodies. Experiments were repeated twice.
2218
J Clin Endocrinol Metab, June 2007, 92(6):2215–2222
Baquedano et al. • Estrogens and Adrenal Development
FIG. 2. Immunohistochemistry localization of ER␣ (Aa– c), ER (Ba– c), and cP450arom (Ca– c) in paraffin sections of normal human adrenal glands of Gr 1 (Aa, Ba, and Ca), Gr 2 (Ab, Bb, and Cb), and Gr 3 (Ca, Cb, and Cc). Aa– c, ER␣ protein was undetectable in the three age Grs, while a strong immunostaining was present in sections of human breast carcinoma used as a positive control (⫹C). Ba– c, Immunoreactivity for ER was confined to FeZ of Gr 1 (Ba) and ZR of Gr 3 (Bc). Ca– c, Immunoreactivity for P450arom was seen in adrenal medulla of the three age Grs and in subcapsular ZG of Gr 3. Immunopositive cells were visualized with the 3,3⬘diaminobenzidine tetrahydrochloride colorimetric reaction, which results in a brown-colored precipitate at the antigen site. Counterstaining was performed using hematoxylin, which stains cell nuclei blue. Scale bar, 100 m (50 m for ⫹C). AM, Adrenal medulla.
Double-label immunofluorescence for cP450arom with chromogranin A Double-labeling experiments were performed by a sequential approach on paraffin tissue sections processed as described previously. After blocking with 1.5% normal goat serum in PBS for 1 h, sections were incubated with a primary anti-chromogranin A antibody (mouse monoclonal diluted at 1/25, NCL-CHROM; Novocastra Laboratories Ltd., Newcastle upon Tyne, UK) overnight at 4 C, followed by the secondary Alexa Fluor 488 goat antimouse antibody (10 g/ml) (Molecular Probes, Eugene, OR) for 1 h at room temperature. This was followed by an overnight incubation at 4 C with the primary anti-cP450arom antibody (rabbit polyclonal, diluted at 1/100) and then with the secondary Alexa Fluor 555 goat antirabbit IgG (10 g/ml) (Molecular Probes) for 1 h at RT. After the last wash, slides were mounted and examined in an Olympus FV300 confocal laser-scanning microscope (Olympus, Hamburg, Germany). To control for nonspecific immunofluorescent staining, sections were also incubated in solutions in which the primary antibody was omitted.
Statistical analysis For statistical analysis, mRNA data, expressed in arbitrary units (AUs), were log⫹1 transformed so that values approximated a normal distribution. Results are expressed as the mean ⫾ sd. Statistical significance was determined by ANOVA followed by the Bonferroni post hoc test. Differences were considered statistically significant when P ⬍ 0.05.
Results Detection of ER␣ and ER mRNAs and proteins, and GPR30 protein
Amplification and abundance of ER␣ and ER mRNA in Grs 1–3 are shown in Fig. 1, A and B. Mean ⫾ sd (AUs) ER mRNA value in Grs 3 (1.54 ⫾ 0.34) and 1 (1.53 ⫾ 0.75) was significantly higher than in Gr 2 (0.79 ⫾ 0.28; P ⬍ 0.05). By immunohistochemistry, both cytoplasmic and nuclear
Baquedano et al. • Estrogens and Adrenal Development
ER staining was observed in FeZ of Gr 1 and ZR cells of Gr 3 (Fig. 2, Ba–Bc). The high degree of intermingling between adrenal medulla and ZR was also evident. In contrast, ER␣ mRNA (Fig. 1A) and protein (Fig. 2, Aa–Ac) were undetectable in human adrenal tissue from any age Gr. Furthermore, GPR30 protein was strongly expressed in ZG and in the adrenal medulla of Gr 3 (Fig. 3). Detection of cP450arom mRNA and protein
Mean ⫾ sd (AU) cP450arom mRNA levels in Grs 1–3 is shown in Fig. 1C. Mean ⫾ sd cP450arom mRNA (1.43 ⫾ 0.40) increased significantly (P ⬍ 0.05) at the age when adrenarche occurs (Gr 3), but no significant difference between Grs 1 (0.97 ⫾ 0.27) and 2 (1.03 ⫾ 0.43) was found. Intense immunoreactivity for cP450arom was observed in adrenal medulla, independent of age and sex, as well as in the subcapsular ZG of Gr 3. Representative results of cP450arom cell immunolocalization in Grs 1–3 are shown in Fig. 2, Ca–Cc. Double-immunofluorescence labeling experiments with an antibody against chromogranin A, a general marker for epinephrine and norepinephrine secreting cells, are shown in Fig. 4. In the adrenal medulla of subjects older than 6 yr, cP450arom-positive cells were chromogranin A negative, suggesting that other cells different from functional mature chromaffin cells are the source of local estrogen in males and females (Fig. 4, G–L). However, in chromaffin cells that had not reached full functional maturity yet, as occurred in the adrenal medulla of subjects younger than 18 months (27), a coexpression of cP450arom and chromogranin A was observed (merged confocal images, Fig. 4, C and F). Analysis of gene expression in medullary and cortical cells isolated by LCM
LCM was used to obtain cDNA exclusively from the ZR, ZF, ZG, and adrenal medulla cells (Fig. 1D, a and b). In four
FIG. 3. Immunohistochemistry localization of GPR30 in paraffin sections of normal human adrenal glands of Gr 3. A, Immunopositive cells appear brown as a result of 3,3⬘-diaminobenzidine tetrahydrochloride colorimetric reaction. A strong immunostaining for GPR30 was present in ZG and adrenal medulla. The negative control (B) exhibited no signal. Counterstaining was performed using hematoxylin, which stains cell nuclei blue. Scale bar, 100 m. AM, Adrenal medulla.
J Clin Endocrinol Metab, June 2007, 92(6):2215–2222
2219
experiments with different human adrenal tissue from Gr 3, ER mRNA was found expressed in ZR cells, while cP450arom and GPR30 mRNAs could be detected in both medullary and ZG cells. Representative results of mRNA zonal expression are shown in Fig. 1Dc. PCR reactions with specific primers for 3HSD (ZF and ZG cell marker), P450c17 (adrenocortical cell marker), and chromogranin A (adrenomedullary cell marker) were performed to analyze the purity of each dissected adrenal zone (Fig. 1Dc). Use of multiple exons 1 in cP450aro mRNA of human adrenal tissue
As shown in Fig. 5A, we found that exons 1.a, 1.b, 1.c, 1.d, and 1.6 were selected in normal human adrenal tissue. However, exon 1.b-containing transcripts in the fraction from subjects younger than 18 months were 3.5-fold higher than in subjects older than 6 yr, while in the latter exon 1.a-, 1.d-, and 1.c-containing transcripts were 3.3-, 1.9-, and 1.7-fold higher, respectively (Fig. 5B). These findings suggest an agerelated switch in promoter use. Moreover, the fraction from subjects older than 6 yr expressed low levels of exon 1.6derived transcripts compared with the youngest fraction. However, the overall expression of 1.6-containing transcripts was low because a 40-cycle PCR was required for visualization. The unspliced transcript of exon 1.6, which contains all genomic sequences between exons 1.6 and 1.d (30), was the only splicing variant of promoter 1.6-driven transcripts detected in both fractions. Discussion
As in previous studies of human fetal adrenals (24), we did not detect expression of ER␣ in any age Gr. By contrast, we found expression of ER. Similar ER abundance between the youngest Gr, in which the FeZ is still present in most adrenals, and the oldest Gr, in which ZR is fully developed, was found. On the other hand, in the intermediate age Gr, ER mRNA expression was lower than in the other two Grs. Consistent with mRNA expression, our immunohistochemistry study revealed that ER expression was almost restricted to the ZR of Gr 3 and FeZ of Gr 1. Moreover, LMC followed by RT-PCR confirmed ER ZR-specific expression. Expression of the ER in the ZR provides a mechanism for mediating the action of estrogen directly within this cortical zone, at adrenarche. In this regard, previous studies have shown that E2 modulates the response of human adrenocortical cells to ACTH by increasing DHEA sulfate production (8 –10, 32). Therefore, we suggest that the estrogen-dependent increase of ACTH-regulated DHEA secretion by human adrenal gland after adrenarche and during the fetal period may be mediated by ER. Estrogen is not solely an endocrine factor, but it is also produced in a number of extragonadal sites, and acts locally in a paracrine and intracrine fashion. The presence of cP450arom activity in a given tissue can generate, locally, high levels of estrogens without significantly affecting circulating levels (13). Therefore, we analyzed the mRNA levels and zonal distribution of cP450arom in human adrenal tissue, as a function of age. Although cP450arom mRNA levels
2220
J Clin Endocrinol Metab, June 2007, 92(6):2215–2222
Baquedano et al. • Estrogens and Adrenal Development
FIG. 4. Laser scanning confocal microscopy of normal human adrenal gland of subjects younger than 18 months (A–F) and subjects older than 6 yr (G–L) for cP450arom (red) and chromogranin A (green). cP450arom Protein was not expressed in adrenal chromaffin cells of the post-adrenarche Gr (subjects older than 6 yr) (I and L). Immunoreactivities for cP450arom and chromogranin A (a chromaffin cell marker) were colocalized in adrenal glands from subjects younger than 18 months, as shown in the C and F merged images. There were no differences in cP450arom expression between female and male adrenal glands. Scale bar, 200 m.
increased at the age when adrenarche occurs, marked cP450arom immunoreactivity was detected in the adrenal medulla of the three age Grs. The increase of cP450arom mRNA in the oldest Gr might be related to the intense cP450arom immunoreactivity also observed in the subcapsular ZG. To confirm the zonal cP450arom expression, LCM followed by RT-PCR was used. As expected, the zonal expression pattern of cP450arom mRNA in microdissected samples was similar to that reveled by immunohistochemical analysis. Considering the fact that various cell systems, including chromaffin cells, tissue macrophages, and cortical cells (5), were found to be intermingled in the human adrenal medulla, double-immunofluorescence studies were performed to evaluate whether cP450arom is expressed in chromaffin cells or not. Surprisingly, dramatic age-related changes in the localization of cP450arom within the adrenal medulla were found. In functionally mature chromaffin cells, cP450arom was not expressed, while in the adrenal medulla of subjects younger than 18 months, virtually all chromaffin cells, which are still functionally immature (27), were cP450arom positive. In accordance with previous reports (33–35), this finding suggests that estrogens, in an autocrine
and paracrine manner, might mediate events associated with adrenomedullary development and functional differentiation. Indeed, a morphological and functional relationship of adrenocortical cells, and the adrenal medulla and its secretory products has been described (5). Furthermore, because ER is clearly expressed in the ZR, but not in adrenal medulla, we propose that estrogens, produced locally in the adrenal medulla, could act as a paracrine factor in the ZR. However, we cannot discard a role of estrogens in the medulla itself not mediated by ER (34). A great number of cP450arom positive cells were also observed in the ZG of Gr 3, suggesting a potential for autocrine and/or paracrine activity of estrogens. However, because classical ER was clearly not expressed in the periphery of the adrenal cortex, this plausible estrogen action should be mediated by an ER-independent mechanism. Several recent studies have shown ER-independent effects of estrogens in neurons and breast cancer cells (20 –22, 36, 37). Because GPR30, an orphan GPR, was recently reported to be a novel transmembrane ER (22), we examined its possible expression in human adrenal tissue from Gr 3. Intriguingly, GPR30 was highly expressed in the periphery of the adrenal cortex, pro-
Baquedano et al. • Estrogens and Adrenal Development
J Clin Endocrinol Metab, June 2007, 92(6):2215–2222
2221
FIG. 5. RT-PCR analysis of the use of alternative exons 1 of the P450arom gene in prepubertal and pubertal normal human adrenal gland. The adrenal cDNA from subjects younger than 18 months and older than 6 yr was pooled in two separate fractions. A, Exons 1.a, 1.b, 1.c, 1.d, and 1.6 are selected in normal human adrenal tissues. P450arom cDNA was amplified using a sense primer specific for exon 1a (5⬘ctggagggctgaacacgtgg-3⬘), 1b (5⬘-gaccaactggagcctgacag-3⬘), 1c (5⬘-ccttgttttgacttgtaacca-3⬘), 1d (5⬘-aacaggagctatagatgaac-3⬘), or 1.6 (5⬘-cacagcagaaccagcacatca-3⬘) and an 32P-labeled antisense primer (5⬘-ccaattcccatgcagtagccagg-3⬘), specific for exon 3; (30 PCR cycles for exon 1.a, 1.b, 1.c, and 1.d-containing PCR products, and 40 cycles for exon 1.6-containing products). Lanes 3 and 4, adrenal cDNA from subjects younger than 18 months. Lanes 5 and 6, adrenal cDNA from subjects older than 6 yr. Human placenta mRNA (exon 1.a- and 1.b-containing PCR products), cAMP-stimulated H295R mRNA (exon 1.c and 1.d-containing PCR products), and testicular tumor mRNA (exon 1.6-containing PCR products) were used as positive controls (lane 2). The sizes of the detected PCR products for exons 1.a, 1.b, 1.c, 1.d, and 1.6 are 304, 262, 270, 273, and 1077 bp, respectively. No product was amplified in reactions that did not contain cDNA (lane 1). B, There were 1.b-containing transcripts in the younger adrenal fraction 3.5-fold higher than in the older one, while 1.a-, 1.d-, and 1.c-containing transcripts increased 3.3-, 1.9-, and 1.7-fold in subjects older than 6 yr, respectively. The intensities of the amplified products in A were quantified by densitometry and normalized to that of -actin (AU). Columns show mean expression levels of exon 1 transcripts in the fraction from subjects older than 6 yr, normalized to subjects younger than 18 months (fold increase). The dashed line indicates the levels of subjects younger than 18 months normalized to one.
viding a possible membrane-associated GPR30 mechanism for mediating local subcapsular estrogen action. In a previous study, we have recently shown that the cell proliferation index is high in the outer ZG of human adrenal tissues, from early infancy to late puberty (25), in agreement with the migration model (38), and we have suggested that the IGF system could be involved in the postnatal mechanisms of progenitor adrenal cell proliferation and migration. However, the molecular mechanisms involved in this process remain poorly understood, and it seems logical to propose that multiple factors are playing a role. It has been reported that estrogens induce cell proliferation of the human adrenocortical H295R cell line (39). Therefore, we can speculate that in the oldest Gr, coincidently with the period of enlargement of ZR, estrogen produced in the outer ZG cells, might stimulate, locally, adrenal progenitor cell proliferation and migration by a GPR30-dependent mechanism. Furthermore, direct interactions and cross talks between the IGF-I and estrogen signaling pathways exist in several tissues and disease processes, involving genomic and nongenomic effects (20, 40–42). Because of the age-related changes found in the cell type distribution of cP450arom expression, we have hypothesized that exon 1 of the cP450arom gene used in human adrenal gland may change in a region- and age-dependent fashion. To define the overall usage of exon 1 at different age periods, pooled cP450arom mRNA from adrenals of subjects younger
than 18 months and those older than 6 yr was analyzed. Different patterns of use of exons 1 were observed in the two pools, even though the same exons 1 were selected. In summary, to our knowledge, this is the first comprehensive report on the localization of cP450arom protein in the human adrenal gland during childhood and puberty. Our results suggest that estrogen produced locally in the adrenal medulla, acting as a paracrine factor through ER, would play a role in the regulation of ZR functional differentiation. The high cP450arom expression in the periphery of the gland, colocalizing with a high cell proliferation index, might also suggest a local ER-independent estrogen action in the proliferation and migration of progenitor adrenal cells in the process of postnatal zonation. However, the role of estrogens in the human adrenal gland is not completely defined, and further studies are required to elucidate the mechanisms of action of estrogens in terms of both genomic and nongenomic effects. Finally, the significance of the developmental changes of cP450arom expression in chromaffin cells remains to be elucidated. Acknowledgments We thank Diego Chirico for hormone measurements, and K. Korach and E. Filardo for kindly supplying ER␣ and ER clones and anti-GPR30 antibody, respectively.
2222
J Clin Endocrinol Metab, June 2007, 92(6):2215–2222
Received October 25, 2006. Accepted March 27, 2007. Address all correspondence and requests for reprints to: Alicia Belgorosky, Endocrinology Service, Hospital de Pediatria Garrahan, C. de los Pozos 1881, Buenos Aires, Argentina C1245AAM. E-mail:
[email protected]. This study was supported by grants from Consejo Nacional de Investigaciones Cientı´ficas y Te´cnicas (CONICET), Fondo para la Investigacio´n Cientı´fica y Tecnologica (FONCYT), Fundacio´n Florencio Fiorini, Argentina, and Pfizer Endocrine Care Internacional Fund for Research and Education. Disclosure: The authors have no conflicts to disclose.
References 1. Havelock JC, Auchus RJ, Rainey WE 2004 The rise in adrenal androgen biosynthesis: adrenarche. Semin Reprod Med 22:337–347 2. Zhang L, Rodriguez H, Ohno S, Miller WL 1995 Serine phosphorylation of human P450c17 increases 17,20-lyase activity: implications for adrenarche and for the polycystic ovary syndrome. Proc Natl Acad Sci USA 92:10619 –10623 3. Gell JS, Carr BR, Sasano H, Atkins B, Margraf L, Mason JI, Rainey WE 1998 Adrenarche results from development of a 3-hydroxysteroid dehydrogenasedeficient adrenal reticularis. J Clin Endocrinol Metab 83:3695–3701 4. Dardis A, Saraco N, Rivarola MA, Belgorosky A 1999 Decrease in the expression of the 3 -hydroxysteroid dehydrogenase gene in human adrenal tissue during prepuberty and early puberty. Implications to the mechanism of adrenarche. Pediatr Res 45:384 –388 5. Ehrhart-Bornstein M, Hinson JP, Bornstein SR, Scherbaum WA, Vinson GP 1998 Intraadrenal interactions in the regulation of adrenocortical steroidogenesis. Endocr Rev 19:101–143 6. Merke DP, Chrousos GP, Eisenhofer G, Weise M, Keil MF, Rogol AD, Van Wyk JJ, Bornstein SR 2000 Adrenomedullary dysplasia and hypofunction in patients with classic 21-hydroxylase deficiency. N Engl J Med 343:1362–1368 7. Weise M, Einsenhofer G, Merke DP 2002 Pubertal and gender-related changes in the sympathoadrenal system in healthy children. J Clin Endocrinol Metab 87:5038 –5043 8. Mesiano S, Jaffe RB 1993 Interaction of insulin-like growth factor-II and estradiol directs steroidogenesis in the human fetal adrenal towards dehydroepiandrosterone sulfate production. J Clin Endocrinol Metab 77:754 –758 9. Winter JSD, Smail PJ 1983 Effect of ACTH and estradiol on steroid production by cultured adrenal cells from anencephalic fetus and from normal adults. Steroids 42:677– 685 10. Gell JS, Oh J, Rainey E, Carr BR 1998 Effect of estradiol on DHEAS production in the human adrenocortical cell line, H295R. J Soc Gynecol Investig 5:144 –148 11. Carmina E, Gonzalez F, Vidali A, Stanczyk FZ, Ferin M, Lobo RA 1999 The contributions of estrogen and growth factors to increased adrenal androgen secretion in polycystic ovary syndrome. Hum Reprod 4:307–311 12. Simpson ER, Clyne C, Rubin G, Boon WC, Robertson K, Britt K, Speed C, Jones M 2002 Aromatase–a brief overview. Annu Rev Physiol 64:93–127 13. Simpson ER, Rubin G, Clyne C, Robertson K, O’Donnell L, Jones M, Davis S 2000 The role of local estrogen biosynthesis in males and females. Trends Endocrinol Metab 11:184 –188 14. Kimura M, Itho N, Tsukamoto T, Kumamoto Y, Yakagy Y, Mori Y 1995 Aromatase activity in an estrogen producing adrenocortical carcinoma in a young man. J Urol 153:1039 –1040 15. Phornphutkul C, Okubo T, Wu K, Harel Z, Tracy TF, Pinar H 2001 Aromatase P450 expression in a feminizing adrenal adenoma presenting as isosexual precocious puberty. J Clin Endocrinol Metab 86:649 – 652 16. Watanabe T, Yasuda T, Noda H, Wada K, Kasukawa I, Someya T, Minamitani K, Minagawa M, Wataki K, Matsunaga T, Ohnuma N, Kohno Y, Harada N 2000 Estrogen secreting adrenal adenocarcinoma in an 18-month-old boy: aromatase activity, protein expression, mRNA and utilization of gonadal type promoter. Endocr J 47:723–730 17. Bouraı¨na H, Lireux B, Mittre H, Benhaim A, Herrou M, Mahoudeau J, Guillon-Metz F, Kottler ML, Reznik Y 2003 Major hyperestrogenism in a feminizing adrenocortical adenoma despite a moderate overexpression of the aromatase enzyme. Eur J Endocrinol 48:457– 461 18. Watanabe M, Nakajin S 2004 Forskolin up-regulates aromatase (CYP19) activity and gene transcripts in the human adrenocortical carcinoma cell line H295R. J Endocrinol 180:125–133 19. 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
Baquedano et al. • Estrogens and Adrenal Development
20. Coleman KM, Smith CL 2001 Intracellular signaling pathways: nongenomic actions of estrogens and ligand-independent activation of estrogen receptors. Front Biosci 6:D1379 –D1391 21. Levin ER 2002 Cellular functions of plasma membrane estrogen receptor. Steroids 67:471– 475 22. Filardo E, Thomas P 2005 GPR30: a seven-transmembrane-spanning estrogen receptor that triggers EGF release. Trends Endocrinol Metab 16:362–367 23. Hirst JJ, West NB, Brenner RM, Novy MJ 1992 Steroid hormone receptors in the adrenal glands of fetal and adult rhesus monkeys. J Clin Endocrinol Metab 75:308 –314 24. Takeyama J, Susuki T, Inoue S, Kaneko C, Nagura H, Harada N, Sasano H 2001 Expression and cellular localization of estrogen receptors ␣ and  in the human fetus. J Clin Endocrinol Metab 86:2258 –2262 25. Baquedano MS, Berensztein E, Saraco N, Dorn GV, Davila MTG de, Rivarola MA, Belgorosky A 2005 Expression of IGF system in human adrenal tissues from early infancy to late puberty. Implications for the development of adrenarche. Pediatr Res 58:451– 458 26. McNicol AM 1992 The human adrenal gland: aspects of structure, function and pathology. In: James VH, ed. The adrenal gland. 2nd ed. New York: Raven Press; 1– 42 27. Mesiano S, Jaffe RB 1997 Developmental and functional biology of the primate fetal adrenal cortex. Endocr Rev 18:378 – 403 28. Koos RD, Banks PK, Inkster SE, Yue W, Brodie AMH 1993 Detection of aromatase and keratinocyte growth factor expression in breast tumor using reverse transcription-polymerase chain reaction. J Steroid Biochem Mol Biol 45:217–225 29. Peng C, Huang TH, Jeung EB, Donaldson CJ, Vale WW, Leung PC 1993 Expression of the type II activin receptor gene in the human placenta. Endocrinology 133:3046 –3049 30. Harada N, Utsumi T, Takagi Y 1993 Tissue-specific expression of the human aromatase cytochrome P-450 gene by alternative use of multiple exons 1 and promoters, and switching of tissue-specific exons 1 in carcinogenesis. Proc Natl Acad Sci USA 90:11312–11316 31. Shozu M, Zhao Y, Bulun S, Simpson E 1998 Multiple splicing events involved in regulation of human aromatase expression by a novel promoter, I.6. Endocrinology 139:1610 –1617 32. Mesiano S, Katz SL, Lee JY, Jaffe RB 1999 Phytoestrogens alter adrenocortical function: genistein and daidzein suppress glucocorticoid and stimulate androgen production by cultured adrenal cortical cells. J Clin Endocrinol Metab 84:2443–2448 33. Lloyd T, Weisz J 1978 Direct inhibition of tyrosine hydroxylase activity by catechol estrogens. J Biol Chem 253:4841– 4843 34. Yanagihara N, Liu M, Toyohira Y, Tsutsui M, Ueno S, Shinohara Y, Takahashi K, Tanaka K 2006 Stimulation of catecholamine synthesis through unique estrogen receptors in the bovine adrenomedullary plasma membrane by 17-estradiol. Biochem Biophys Res Commun 339:548 –553 35. Conley AJ, Corbin CJ, Hinshelwood MM, Liu Z, Simpson ER, Ford JJ, Harada N 1996 Functional aromatase expression in porcine adrenal gland and testis. Biol Reprod 54:497–505 36. Diaz-Perez MJ, Zannis-Hadjopoulos M, Price GB, Wainer IW 1998 Receptor independent effects on DNA replication by steroids. J Cell Biochem 70:323–329 37. Singh M, Setalo Jr G, Guan X, Frail DE, Toran-Allerand CD 2000 Estrogeninduced activation of the mitogen-activated protein kinase cascade in the cerebral cortex of estrogen receptor-␣ knock-out mice. J Neurosci 20:1694 –1700 38. Wolkersdorfer G, Bornstein SR 1998 Tissue remodeling in the adrenal gland. Biochem Pharmacol 56:163–171 39. Montanaro D, Maggiolini M, Recchia AG, Sirianni R, Aquila S, Barzon L, Fallo F, Ando S, Pezzi V 2005 Antiestrogens upregulate estrogen receptor  expression and inhibit adrenocortical H295R cell proliferation. J Mol Endocrinol 35:245–256 40. Cardona-Gomez GP, Mendez P, DonCarlos LL, Azcoitia I, Garcia-Segura LM 2002 Interactions of estrogen and insulin-like growth factor I in the brain: molecular mechanisms and functional implications. J Steroid Biochem Mol Biol 83:211–217 41. Hamelers IH, Steenbergh PH 2003 Interactions between estrogen and insulinlike growth factor signaling pathways in human breast tumor cells. Endocr Relat Cancer 10:331–345 42. Karl M, Potier M, Schulman IH, Rivera A, Werner H, Fornoni A, Elliot S 2005 Autocrine activation of the local insulin-like growth factor I system is upregulated by estrogen receptor (ER)-independent estrogen actions and accounts for decreased ER expression in type 2 diabetic mesangial cells. Endocrinology 146:889 –900
JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.
