Sonic Hedgehog Signaling is Active in Human Adrenal Cortex Development and Deregulated in Adrenocortical Tumors Débora C. Gomes1, Letícia F. Leal1, Livia M. Mermejo1, Carlos A. Scrideli1, Carlos E. Martinelli Jr.1, Maria C. B. V. Fragoso2, Ana C. Latronico2, Luis G. Tone1, Silvio Tucci1, Jose A. Yunes3, Izilda A. Cardinalli3, Maria J. Mastellaro3, Silvia R. Brandalise3, Fernando Ramalho1, Ayrton C. Moreira1, Leandra N. Ramalho1, Margaret de Castro1, Sonir R. R. Antonini1 School of Medicine of Ribeirao Preto -University of Sao Paulo, Brazil1, School of Medicine - University of Sao Paulo2, Boldrini Children’s Center3

Background: The sonic hedgehog (SHH) pathway plays a key role in rodent adrenal cortex development and is involved in tumorigenesis in several human tissues, but data in human adrenal glands are limited. Objectives: To analyze the involvement of the SHH pathway in human adrenal development and tumorigenesis and the effects of SHH inhibition on an adrenocortical tumor (ACT) cell line. Patients & Methods: Expression of SHH pathway components was evaluated by immunohistochemistry (IHC) in 51 normal adrenals (33 fetal) and 34 ACTs (23 pediatric), and by qPCR in 81 ACTs (61 pediatric) and 19 controls (10 pediatric). The effects of SHH pathway inhibition on gene expression and cell viability in the NCI-H295A adrenocortical tumor cell line after cyclopamine treatment were analyzed. Results: SHH pathway proteins were present in fetal and postnatal normal adrenals and showed distinct patterns of spatiotemporal expression throughout development. Adult ACCs presented with higher expression of PTCH1, SMO, GLI3 and SUFU compared with normal adult adrenal cortices. Conversely, pediatric ACTs showed lower mRNA expression of SHH, PTCH1, SMO, GLI1 and GLI3 compared with normal pediatric adrenal cortices. In vitro treatment with cyclopamine resulted in decreased GLI3, SFRP1, CTNNB1 mRNA expression and beta-catenin staining, as well as decreased cell viability. Conclusions: The SHH pathway is active in human fetal and postnatal adrenals, up regulated in adult ACCs and down regulated in pediatric ACTs. SHH pathway antagonism impaired cell viability. The SHH pathway is deregulated in ACTs and might provide a new target therapy to be explored.

enign adrenocortical tumors (ACTs) are relatively frequent in adults, whereas adrenocortical carcinomas (ACCs) are rare and exhibit very aggressive behavior (1). Increased incidence of ACTs in children has been observed in the South and Southeastern regions of Brazil, often in association with the germline p.R337H P53 mutation (2, 3). Recent data have shown that only 3% of children car-

B

rying this mutation develop ACTs (4). Therefore, it is likely that additional genetic events contribute in adrenocortical tumor development. Studies on the molecular basis of ACTs have demonstrated that genes and pathways involved in normal adrenal cortex development and maintenance are important in tumorigenesis (5). Outstanding among these factors are activation of the Wnt/␤-catenin

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received November 14, 2013. Accepted March 28, 2014.

Abbreviations:

doi: 10.1210/jc.2013-4098

J Clin Endocrinol Metab

jcem.endojournals.org

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

1

2

SHH in adrenal development and tumorigenesis

pathway and IGF2 overexpression (3, 6 – 8). However, the knowledge that has, until now, been accumulated is insufficient to account for the events required for tumor initiation and malignant transformation (9, 10). The hedgehog (Hh) pathway is important in cell growth and differentiation, embryonic development, adult tissue homeostasis and tumorigenesis, and plays a role in mammalian adrenal cortex development (11–14). In vertebrates, Hh activation is triggered by Hh ligand binding to a receptor complex formed by the transmembrane protein patched 1 (PTCH1) and a coreceptor (cell adhesion associated, oncogene regulated – CDO, BOC cell adhesion associated, oncogene regulated – BOC or growth arrestspecific1 - GAS1). In the presence of the ligand, the smoothened, frizzled class receptor (SMO) is released from PTCH1 inhibition and activates the transcription factors GLI1, GLI2 and GLI3. In the absence of the ligand, GLI2 and GLI3 are targeted to proteasomal processing into repressor forms, which act as transcriptional repressors (15). In mice and rat adrenals, Sonic hedgehog (SHH) is the main ligand of the Hh pathway (16). Shh is mainly expressed underneath the adrenal capsule and Gli1 is expressed in the capsule (12–14, 17). Conditional Shh inactivation in the adrenal cortex produces adrenal cortex and capsule hypoplasia (12, 13). Moreover, Gli1-positive cells migrate centripetally, supporting the role of this pathway in adrenocortical growth (13, 14). On the other hand, studies on the role of SHH signaling in human adrenal physiology are limited. Nevertheless, the discovery of adrenal hypoplasia in some patients with Pallister-Hall Syndrome, which is caused by GLI3 mutations, suggests its importance (18). Moreover, abnormalities in the SHH pathway have been linked to several cancers, such as basal cell carcinoma, medulloblastoma, glioblastoma, and rhabdomyosarcoma (19). Based on these data, we hypothesized that SHH pathway abnormalities may be important in adrenocortical tumorigenesis. In the present study, we show that key proteins of the SHH pathway present distinct patterns of expression depending on the developmental stage in the normal human adrenal cortex. Moreover, we observed differential expression of SHH pathway genes in pediatric and adult ACTs. In addition, we demonstrated that SHH signaling inhibition results in cell viability impairment in NCIH295A adrenocortical tumor cell line.

SUBEJCTS AND METHODS Normal fetal and postnatal adrenal panel A retrospective study was carried out in a group of 51 normal adrenal tissues obtained from spontaneously miscarried fetuses

J Clin Endocrinol Metab

or from children who underwent routine autopsy examination in the Department of Pathology, School of Medicine of Ribeirao Preto. Tissues specimens were analyzed according to gestational or postnatal age: 20 to 25 weeks (n ⫽ 14), 26 to 31 weeks (n ⫽ 10), 32 to 37 weeks (n ⫽ 9), and neonates to 10-year-old children (n ⫽ 18). Fetuses or children with multiple malformations or children with previous use of glucocorticoids were excluded. This section of the study was approved by the local Ethics Committee (protocol 2011/11614).

Immunohistochemical (IHC) analyses- Normal Adrenals Routine hematoxylin-eosin staining was performed to evaluate adrenal morphology. IHC studies were performed using the primary antibodies anti-SHH (sc-1194), anti-GLI1 (sc-20687), anti-GLI2 (Ab-7181), and anti-GLI3 (sc-20688), as has previously been validated (20, 21). Labeling was developed with DAB (Vector Laboratories Inc.) and the sections were counterstained with Harris hematoxylin.These samples were analyzed by D.C.G. and S.R.A and confirmed by an experienced pathologist (L.Z.R.). A descriptive analysis in each developmental stage is presented. Detailed methodology is shown in the Supplemental Data.

Adrenocortical tumors Patients Eighty-one patients with ACTs diagnosed between 1991 and 2012 at three reference centers in southeastern Brazil were enrolled in this study. Routine diagnostic tests and imaging evaluations were performed before surgical treatment and the patients were followed up after surgery, as previously reported (3). Disease stages were classified according to the classifications of Sandrini and Macfarlane, modified by Sullivan for children and adults, respectively (22, 23). In addition, ACTs with available tissues were classified according to the Weiss score (24). Adult ACTs were classified as adenoma (ACA) or carcinoma (ACC), according to Weiss ⬍ 3 or ⱖ 3, respectively. In the pediatric group, we did not categorize ACTs as ACA or ACT according to the Weiss score, because this criterion is not entirely applicable and does not accurately predict clinical outcomes (24). Nineteen normal adrenal cortices were analyzed as controls: 10 pediatric specimens were obtained after surgical treatment of Wilms’ tumor and 9 from adult subjects with sudden deaths who underwent routine autopsy. This study was approved by the Ethics Committees (protocols 2011/5721 and 0005.0.144.000 –10). Informed written consent was obtained from patients and controls or their parents.

Quantitative real time PCR (qPCR) The mRNA expression was analyzed with qPCR as previously described (3) using specific Taqman assays (Applied Biosystems): SHH (Hs00179843 m1), PTCH1 (Hs00181117 m1), SMO (Hs01090242 m1), GLI1 (Hs01110766 m1), GLI2 (Hs01119974 m1), GLI3 (Hs00609233 m1), and SUFU (suppressor of fused; Hs00171981 m1). GUSB (beta glucuronidase; 4326320E) was used as the endogenous control. Normal pediatric and adult adrenal cortices were used as controls for pediatric and adult ACTs, respectively. Relative mRNA expression values were determined by the 2-⌬⌬Ct method.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2013-4098

IHC analyses- ACTs IHC for SHH, GLI1 and GLI3 proteins was performed as described above in a subset of 34 tumors (23 pediatric, 11 adults) depending on the availability of paraffin-embedded samples. Each marker was evaluated randomly in at least 10 representative high-power fields (x400). According to staining, each sample was defined as negative (less than 1%), ⫹1 (weak, 1%–10%), ⫹2 (moderate, 11%–50%) and ⫹3 (strong, more than 50%). The normal adrenal cortex tissues present in the sample were considered as the control.

In Vitro Study NCI-H295A adrenocortical cell line NCI-H295A adrenocortical tumor cells, a lineage established from an adult adrenocortical carcinoma harboring a CTNNB1 mutation (25), were cultured in RPMI 1640 (GIBCO, Life Technologies, Foster City, CA), supplemented with 2% fetal bovine serum, 1% ITS Premix (BD Biosciences) and 1% penicillin/streptomycin (100 mg/ml; GIBCO), and maintained at 37°C with 5% CO2.

jcem.endojournals.org

3

Titer 96 AQueous one solution (MTS, Promega, Madison, WI) was added to each well. After 2 hours of incubation, absorbance at 490nm was measured using a Sunrise microplate reader (Tecan, Salzburg, Austria) in two independent experiments. The half maximal inhibitory concentration (IC50) values were calculated in dose-response curves using CalcuSyn software (Biosoft, Ferguson, MO).

Statistical analyses Statistical analyses were performed using SAS software, version 9.2 (SAS Institute Inc., Cary, NC) and GraphPad Prism 5.02 (Graphpad, San Diego, CA). Continuous variables were expressed as mean and range or standard deviation. Differences among variables were evaluated using linear regression (adding tumor weight and patient age for pediatric ACTs as control variables, when appropriate); data values were log transformed before tests. One-way ANOVA followed with Dunnett’s post-test were used in the in vitro study analyses. Overall survival was defined as the time from the date of diagnosis to death or last follow-up and analyses were carried out using Kaplan-Meier curves and the Log-rank test. The significance level was set at P ⬍ .05 for all analyses.

SHH pathway inhibition (Cyclopamine) Cells were plated in 24-well plates at a density of 200 ⫻ 103cells/well. After 24 hours, the medium was replaced and cells were treated with 5 or 20 ␮M cyclopamine (Sigma-Aldrich, St. Louis, MO), an SMO inhibitor. Dimethylsulfoxide (DMSO- Sigma-Aldrich, St. Louis, MO) 0.1% was used as a control. After 24 hours incubation, medium was removed and 250 ␮L TRIzol was added to each well for RNA extraction. Expression of GLI1, GLI3, SFRP1 (secreted frizzled-related protein 1; Hs00610060 m1),CTNNB1 (beta catenin Hs00170025 m1) and GUSB mRNA was evaluated with qPCR in three independent experiments. GUSB gene was used as the endogenous control.

Immunofluorescence Staining For immunofluorescence assays, cells were cultured over coverslips added to 24-well plates. After treatment with cyclopamine the medium was removed and cells were fixed with methanol for 3 minutes, washed 3x with 0.01M phosphate buffered saline (PBS), and incubated with 10% normal horse serum for 1 hour. Slides were then incubated with anti-␤-catenin antibody (#610154 BD Biosciences, 1:2000) overnight at room temperature, washed 3x with 0.01M PBS, incubated with AffiniPure donkey antimouse CyTM5-conjugated antibody (red color, 1:250, #715–175–150, Jackson ImmunoResearch, West Grove, PA) for 4 hours, washed with 0.01M PBS 3x and incubated with 4⬘,6-diamidino-2-phenylidone (DAPI, blue color, #4083 Cell Signaling Technology, Danvers, MA-1:25,000) for 2 minutes for nuclear counterstaining, then washed with PBS and mounted with Fluoromount (Sigma-Aldrich, St. Louis, MO). Staining was analyzed with a Leica TCS SP5 laser scanning confocal microscope (Leica Microsystems, Wetzlar, Germany) with a fixed time of exposure for all samples.

Results Expression of SHH pathway proteins during normal adrenal development (Figure 1) Hematoxylin-eosin staining confirmed the histomorphology of human adrenal tissues. In general, SHH staining was mainly found, although weakly, in the cytoplasm throughout the fetal and the definitive adrenal cortex. SHH staining tended to be more intense next to the capsule in later stages of fetal life and in the postnatal adrenal cortex. The GLI family of transcription factors showed different staining patterns. In early fetal stages, GLI1 labeled all regions of the cortex, particularly the fetal zone. As the adrenal cortex developed, GLI1 staining tended to be stronger in the capsule and adjacencies. Independently of the developmental stages, GLI1 staining was weaker than GLI2 and GLI3 staining. GLI2 staining was positive in all regions during fetal stages. In earlier stages, it was stronger in the definitive cortex. In the postnatal adrenal, GLI2 staining was scattered throughout the cortex. In fetal adrenals, GLI3 staining was weak in the capsule and strong in the definitive cortex, as well as in the fetal cortex. In postnatal adrenals, GLI3 staining remained intense throughout the cortex, but not in the capsule. Adrenocortical tumors

Cell viability assay

Clinical presentation of ACTs

NCI-H295A cells were plated in 96-well plates at a density of 20,000 cells/well. After growth for 48 hours, the medium was replaced with medium containing 5 to 30 ␮M cyclopamine or 0.1% DMSO (control). After 24 to 96 hours, 20 ␮L of the Cell-

Adults Of the twenty adults evaluated, 18 were females (90%). Mean age at diagnosis was 35 years (17– 66) and 90% of

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

4

SHH in adrenal development and tumorigenesis

20 - 25 weeks GA

26 - 31 weeks GA

J Clin Endocrinol Metab

32 - 37 weeks GA

Postnatal

SHH

20X 40X

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

GLI3

GLI2

GLI1

A

Figure 1. Immunohistochemical staining of SHH, GLI1, GLI2 and GLI3 proteins in the human adrenal cortex in different stages of development (20 and 40X). Protein staining was developed with DAB (brown) and counterstained with Harris hematoxylin (blue). A-D: SHH staining; weak, mainly in the cytoplasm throughout the fetal and the definitive adrenal cortices (A and B) and more intense next to the capsule in later stages of the fetal and postnatal adrenal cortices (C and D). E-H: GLI1 staining; positive in the fetal cortex in early fetal stages (E); abundant in the capsule and its adjacencies and weaker in the cortex in other stages (F-H). I-L: GLI2 staining: positive in all regions during fetal stages, most intense in the definitive cortex in earlier stages (I-K) and scattered throughout the cortex in postnatal adrenals (L). M-P: GLI3 staining: weak in the capsule and strong in the definitive cortex, as well as in the fetal cortex in fetal adrenals (M-O): intense throughout the cortex but not in the capsule in postnatal adrenals (P).

these patients presented with hormone secreting tumors (androgen: 2, cortisol: 6, androgen and cortisol: 8, and inhibin: 2). MacFarlane stages were I in 5, II in 6, III in 4, and IV in 5 cases. The Weiss scores ranged from 1 to 8; six tumors were classified as ACA, thirteen as ACC and one was unspecified. Mean follow-up was 70.3 months. CTNNB1 and the p.R337H P53 mutations were found in 35% and 20% of cases, respectively. Part of these clinical data had been previously reported (26) and are presented in the Supplemental Data (Table 1). Pediatric Of the 61 children evaluated, 48 were girls (79%) and mean age at diagnosis was 3.4 years (0.4 –15.5). Fifty-nine (97%) patients had hormone secreting tumors (androgen: 33, cortisol: 2, androgen and cortisol: 24). Sandrine stages were I in 37, II in 8, III in 8, and IV in 8 cases. Mean follow-up was 69.9 months. A favorable prognosis was associated with age ⬍ 48 months (P ⬍ .01), tumor weight ⬍ 200g (P ⬍ .01), tumor stage I (P ⬍ .01) and normal cortisol secretion (P ⫽ .04). Weiss score was determined in a subset of 50 tumors and ranged from 3 to 9. All patients presenting a Weiss score ⬍ 6 survived and this score was associated with a good prognosis (P ⬍ .01). CTNNB1 and the p.R337H P53 mutations were found in 6% and 85% of cases, respectively. Part of these clinical

data had been previously reported (3) and are presented in the Supplemental Data (Table 2). mRNA expression of SHH pathway genes Normal Adrenal Cortices Overall, the mRNA expression of SHH pathway genes was higher in pediatric than in adult normal adrenal cortex samples (Figure 2A): SHH (P ⫽ .01), PTCH1 (P ⫽ .01), SMO (P ⬍ .01), GLI1 (P ⬍ .01), GLI2 (P ⫽ .02) and GLI3 (P ⫽ .01). Adult ACTs Adult ACCs presented higher mRNA expression levels of PTCH1 (P ⫽ .04), SMO (P ⫽ .03), GLI3 (P ⬍ .01) and SUFU (P ⬍ .01) than normal adult adrenal cortices. These differences were not observed comparing ACA and controls (Figure 2B). Noncarriers of the p.R337H P53 mutation presented higher GLI3 mRNA expression levels than carriers (P ⫽ .04).

Pediatric ACTs Pediatric ACTs presented lower mRNA expression levels of SHH (P ⬍ .01), PTCH1 (P ⫽ .03), SMO (P ⬍ .01), GLI1 (P ⬍ .01) and GLI3 (P ⫽ .01) than normal adrenal cortices obtained from children (Figure 2C). Cortisol secreting pediatric ACTs showed higher mRNA expression of PTCH1 (P ⬍ .01) and SUFU (P ⫽ .04). No differential mRNA expression was observed in association with p.R337H P53 and CTNNB1 mutations. No association between mRNA expression of SHH pathway components and overall survival was observed in either the adult or pediatric ACT groups. IHC In agreement with the qPCR data, IHC showed that GLI1 and GLI3 staining was more frequent and intense in adult ACAs and ACCs, compared with pediatric ACTs. GLI1 and GLI3 labeling was positive in 30% and 87% of the pediatric ACTs and in 83% and 100%, and 100% and 100% of the adult ACAs and ACCs, respectively (Figure 3A and 3B).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2013-4098

jcem.endojournals.org

5

␤-catenin staining, both in the cytoplasm and the nucleus (Figures 4B). Cyclopamine effect on cell viability Cyclopamine treatment did not affect NCI-H295A cell viability in the first 24 hours (Figure 4C). However, a dose dependent reduction in cell viability was observed after 96 hours in cells treated with 20 ␮M (24 ⫾ 16%; P ⬍ .01) and 30 ␮M (30 ⫾ 7; P ⬍ .001) cyclopamine, compared with the control (Figure 4D). The IC50 was 61 ␮M (r ⫽ 0.95) after 96 hours of incubation.

Discussion The expression pattern of the SHH pathway genes in different stages of development of the human adrenal cortex or in pediatric ACTs had not been explored to date. The present study shows that the SHH pathway is active in fetal and postnatal human adrenal cortices. In human fetuses, the expression of SHH components in the adrenal cortex was detected from midgestation and continued Figure 2. Expression levels of mRNA SHH pathway genes in normal adrenal throughout postnatal life. Interestcortices (A) and in adrenocortical tumors (B and C). A: The normal adrenal cortices of children presented higher mRNA expression of SHH, PTCH1, SMO, GLI1, GLI2, GLI3 than adult ingly, the mRNA expression levels of normal adrenal cortices. B: PTCH1, SMO, GLI3 and SUFU were increased in adult ACCs when SHH components were higher in the compared with normal adult adrenal cortices. C: SHH, PTCH1, SMO, GLI1 and GLI3 were normal adrenal cortices from childecreased in pediatric ACTs when compared with normal pediatric adrenal cortices. (*P ⬍ .05; dren than in those from adults. In **P ⬍ .01) (mRNA expression presented as median, interquartile range and fifth /95th percentile). agreement with our findings, studies in animal models indicated that the SHH pathway is active and plays a Adrenocortical tumor cell line -NCI-H295A role in the development and maintenance of the adrenal Inhibition of the SHH pathway impairs Wnt/␤- cortex (12–14). Similarly to what had been detected in catenin signaling earlier stages of adrenal cortex development in rodents To evaluate a possible interaction between the SHH (12), we detected SHH protein expression in the human and Wnt/␤-catenin pathways, we treated adrenocortical fetal adrenal cortex during the earlier fetal stages. In conNCI-H295A cells with cyclopamine, a selective SMO intrast, SHH protein expression tended to be more intense hibitor. Twenty-four hour treatment with 20 ␮M cyclopamine caused a significant reduction of GLI3 mRNA next to the capsule in later stages of fetal development and expression (P ⫽ .01) and of components of the Wnt path- in the postnatal adrenal cortex. This finding is in agreeway, SFRP1 (P ⫽ .01) and CTNNB1 (P ⫽ .03) (Figure ment with a previous report showing the presence of SHH 4A). Under basal conditions, immunofluorescence analy- mRNA expression in the periphery of the adult human sis showed strong cytoplasmic and nuclear ␤-catenin adrenal cortex (27). GLI1, a transcription factor induced by SHH, is constaining. In agreement with the mRNA results, a 24-hour treatment with 20 ␮M cyclopamine resulted in decreased sidered to be a marker of SHH activation (11). In agree-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

6

SHH in adrenal development and tumorigenesis

ment with rodent models and a study in adult human adrenals (12, 13, 28), we found GLI1 protein expression mostly in the capsule and adjacencies in postnatal adrenals. The same pattern was also observed in later fetal stages. However, in earlier developmental stages, GLI1 staining was detected mostly in the fetal adrenal cortex. GLI1-positive cells have been considered a marker of progenitor cells (14). Therefore, our data suggest that in earlier stages of fetal life, stem/progenitor cells are present in the entire human adrenal cortex, but later become restricted to the capsule. These findings agree with the hypothesis that the adult cortex arises from both cells of the fetal cortex and cells of the capsule/subcapsular region (29 –31). Positive protein expression of GLI2 and GLI3 was found during all stages of development in the analyzed adrenal cortices. Interestingly, in postnatal adrenals, GLI3 expression remained strong throughout the cortex, but not in the capsule. In mouse models, GLI2 and GLI3 mRNA levels were detected in the subcapsular region of the cortex during embryonic stages and at birth, but not in the adult adrenal cortex (12). In fact, GLI3 seems to be important for human adrenal development, since some patients harboring GLI3 inactivating mutations present with adrenal hypoplasia (18). It is unclear whether GLI3 is important in adrenal zonation. However, in neural and

Figure 3. Immunohistochemical analysis in ACTs. A: Representative image of SHH, GLI1 and GLI3 immunostaining in ACTs (20x magnification). Protein staining was developed with DAB (brown) and counterstained with Harris hematoxylin (blue). Staining intensity classification: SHH (2⫹: moderate), GLI1 (3⫹: strong) and GLI3 (3⫹: strong); B: Percentage and intensity of positive SHH, GLI1 and GLI3 protein staining in pediatric and adult ACTs.

J Clin Endocrinol Metab

bone tissues, GLI3 modulates the temporal kinetics of cell differentiation by interacting with the WNT and FGF pathways (32, 33). Analyzing the expression of SHH pathway genes in ACCs of adult patients, we found positive GLI1 protein expression in all tumors, as well as increased PTCH1 mRNA expression and normal SHH expression in ACCs when compared with normal adrenal cortices. Conversely, in aldosterone producing adenomas, which were not evaluated in our study, SHH staining was found in all analyzed samples (27). The expression of PTCH1, SMO, GLI3 and SUFU mRNA was higher in adult ACCs than in normal adrenal cortices. This differential expression was not observed in ACAs when comparing with normal adrenals. However, due to the small number of ACAs analyzed, this result should be further tested and confirmed. Previously, microarray studies found higher GLI2 mRNA expression in ACCs compared with ACAs (9, 34, 35) and with normal adrenal cortices (34). In addition, Soon et al (2009) found lower expression of SHH and PTCH1 mRNA in ACCs than in ACAs (35). We did not find these differences. All but two adult patients evaluated in our series presented with secreting tumors. Of note, a recently published study analyzing a small number of samples found evidence of SHH pathway activity in nonsecreting adrenocortical adenomas (28). Differing from the adult group, pediatric ACTs were observed to have lower expression of most SHH pathway genes, as compared with normal pediatric adrenal cortices. Noteworthy, most of the pediatric ACTs in our cohort are associated with a specific, low penetrance TP53 mutation (p.R337H), which could characterize a specific subtype of childhood ACTs. However, our finding is in agreement with data from a microarray analysis that also found lower expression of PTCH1 mRNA in pediatric ACTs compared with normal adrenals (36). Our data reinforces that pediatric and adult ACTs differ in both origin and clinical behavior (37). The mechanism underlying the involvement of the SHH pathway in ACTs is unknown, but might be related to abnormal Wnt/␤catenin signaling. Previous studies suggested an interaction between SHH and WNT pathways (38 – 40). In order to evaluate a possible interaction of SHH and WNT pathways in adrenocortical tumor cells, we treated NCI-H295A cells with cyclopamine. Our data showed that SMO inhibition resulted in decreased CTNNB1 mRNA expression and cytoplasmic and nuclear ␤-catenin staining. Thus, SMO inhibition in adrenocortical tumor cells may impair WNT/ ␤-catenin signaling. In addition, cyclopamine treatment also resulted in a decrease of GLI3 mRNA expression. Of note, we found increased expression of GLI3 mRNA lev-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2013-4098

A

jcem.endojournals.org

24h

C

24h

7

portant in adrenocortical tumorigenesis and may be a potential target in adjuvant therapy for adrenocortical carcinomas.

Acknowledgments 24h Beta-catenin

D DAPI

Merge

D

96h

24h

Cyclopamine 20

DMSO

B

We thank Professor Claudimara Lotfi for providing the NCI-H295A cell line, Wendy Turatti and Rogerio Zuliani for their technical assistance and Silvia G. Ruginsk and Rodrigo C. Rorato for their kind help. We also thank Leandro M. Colli for his help with the microarray studies and data mining and the statistician Davi C. Aragon for help with the statistical analysis.

Figure 4. Responses to cyclopamine treatment in the adrenocortical tumor cell line NCI-H295A.A: After 24 hours of 20 ␮M cyclopamine treatment GLI3, SFRP1 and CTNNB1 mRNA expression was decreased when compared with 0.1% DMOS; B: After 24 hours, 20 ␮M cyclopamine treatment leads to lower cytoplasmic and nuclear ␤-catenin staining when compared with 0.1% DMSO (␤-catenin –red color; nucleus – blue color; 20x magnification); C: Viability assay after 24 hours: no changes in viability were observed; D: Viability assay after 96 hours: impairment of cell viability was observed with 20 and 30 ␮M cyclopamine treatment compared with 0.1%DMSO (*P ⬍ .05; **P ⬍ .01; ***P ⬍ .001).

els in adult ACCs and GLI3-positive staining in all evaluated adult ACTs. Conversely, GLI3 expression was decreased in pediatric ACTs. Interestingly, the prognosis of pediatric patients is usually better than in adults (37). Our original result showing GLI3 overexpression in adult ACTs may be important, since GLI3 overexpression has been associated with cell survival, epithelial mesenchymal transition and tumor chemical resistance in other tumors types (41, 42). Our data showing decreased NCI-H295A cell viability after cyclopamine treatment is in agreement with a recently published study by an independent group (28). We must consider that cyclopamine may induce nonspecific toxicity (43), but the concentrations used in our study were moderate and cell viability impairment occurred only after 96 hours of incubation, suggesting a direct SMO inhibition effect. These data should be further explored, particularly by in vivo animal studies. Of note, several Hedgehog pathway inhibitors are currently undergoing phase I and II clinical trials for treatment of different neoplasms (44). In conclusion, we demonstrated that the SHH pathway is active in human adrenals and presents distinct spatiotemporal expression patterns throughout adrenal development. In addition, we showed that the SHH pathway is deregulated in ACTs and in vitro SHH pathway inhibition impaired WNT/␤-catenin signaling and reduced tumor cell viability. Thus, the SHH pathway appears to be im-

Address all correspondence and requests for reprints to: Sonir R. Antonini, MD, PhD, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Avenida Bandeirantes, 3900 – Monte Alegre, CEP14049 –900, Ribeirão Preto – São Paulo, Brasil, email: [email protected]. Disclosure Summary: The authors have nothing to disclose. This work was supported by FINANCIAL SUPPORT: Sao Paulo State Research Council (FAPESP), grant #2011–13807– 4 and Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq), grant #2011– 474273.

References 1. Fassnacht M, Libe R, Kroiss M, Allolio B. Adrenocortical carcinoma: a clinician’s update. Nat Rev Endocrinol. 2011;7:323–335. 2. Latronico AC, Pinto EM, Domenice S, Fragoso MC, Martin RM, Zerbini MC, Lucon AM, Mendonca BB. An inherited mutation outside the highly conserved DNA-binding domain of the p53 tumor suppressor protein in children and adults with sporadic adrenocortical tumors. J Clin Endocrinol Metab. 2001;86:4970 – 4973. 3. Leal LF, Mermejo LM, Ramalho LZ, Martinelli CE, Jr., Yunes JA, Seidinger AL, Mastellaro MJ, Cardinalli IA, Brandalise SR, Moreira AC, Tone LG, Scrideli CA, Castro M, Antonini SR. Wnt/betacatenin pathway deregulation in childhood adrenocortical tumors. J Clin Endocrinol Metab. 2011;96:3106 –3114. 4. Custodio G, Parise GA, Kiesel Filho N, Komechen H, Sabbaga CC, Rosati R, Grisa L, Parise IZ, Pianovski MA, Fiori CM, Ledesma JA, Barbosa JR, Figueiredo FR, Sade ER, Ibanez H, Arram SB, Stinghen ST, Mengarelli LR, Figueiredo MM, Carvalho DC, Avilla SG, Woiski TD, Poncio LC, Lima GF, Pontarolo R, Lalli E, Zhou Y, Zambetti GP, Ribeiro RC, Figueiredo BC. Impact of Neonatal Screening and Surveillance for the TP53 R337H Mutation on Early Detection of Childhood Adrenocortical Tumors. J Clin Oncol. 2013;31:2619 –2626. 5. Simon DP, Hammer GD. Adrenocortical stem and progenitor cells: implications for adrenocortical carcinoma. Mol Cell Endocrinol. 2012;351:2–11.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

8

SHH in adrenal development and tumorigenesis

6. Berthon A, Sahut-Barnola I, Lambert-Langlais S, de Joussineau C, Damon-Soubeyrand C, Louiset E, Taketo MM, Tissier F, Bertherat J, Lefrancois-Martinez AM, Martinez A, Val P. Constitutive betacatenin activation induces adrenal hyperplasia and promotes adrenal cancer development. Hum Mol Genet. 2010;19:1561–1576. 7. Tissier F, Cavard C, Groussin L, Perlemoine K, Fumey G, Hagnere AM, Rene-Corail F, Jullian E, Gicquel C, Bertagna X, VacherLavenu MC, Perret C, Bertherat J. Mutations of beta-catenin in adrenocortical tumors: activation of the Wnt signaling pathway is a frequent event in both benign and malignant adrenocortical tumors. Cancer Res. 2005;65:7622–7627. 8. Almeida MQ, Fragoso MC, Lotfi CF, Santos MG, Nishi MY, Costa MH, Lerario AM, Maciel CC, Mattos GE, Jorge AA, Mendonca BB, Latronico AC. Expression of insulin-like growth factor-II and its receptor in pediatric and adult adrenocortical tumors. J Clin Endocrinol Metab. 2008;93:3524 –3531. 9. Heaton JH, Wood MA, Kim AC, Lima LO, Barlaskar FM, Almeida MQ, Fragoso MC, Kuick R, Lerario AM, Simon DP, Soares IC, Starnes E, Thomas DG, Latronico AC, Giordano TJ, Hammer GD. Progression to adrenocortical tumorigenesis in mice and humans through insulin-like growth factor 2 and beta-catenin. Am J Pathol. 2012;181:1017–1033. 10. Drelon C, Berthon A, Ragazzon B, Tissier F, Bandiera R, SahutBarnola I, de Joussineau C, Batisse-Lignier M, Lefrancois-Martinez AM, Bertherat J, Martinez A, Val P. Analysis of the role of Igf2 in adrenal tumour development in transgenic mouse models. PLoS One. 2012;7:e44171. 11. Ingham PW, McMahon AP. Hedgehog signaling in animal development: paradigms and principles. Genes Dev. 2001;15:3059 – 3087. 12. Ching S, Vilain E. Targeted disruption of Sonic Hedgehog in the mouse adrenal leads to adrenocortical hypoplasia. Genesis. 2009; 47:628 – 637. 13. Huang CC, Miyagawa S, Matsumaru D, Parker KL, Yao HH. Progenitor cell expansion and organ size of mouse adrenal is regulated by sonic hedgehog. Endocrinology. 2010;151:1119 –1128. 14. King P, Paul A, Laufer E. Shh signaling regulates adrenocortical development and identifies progenitors of steroidogenic lineages. Proc Natl Acad Sci U S A. 2009;106:21185–21190. 15. Robbins DJ, Fei DL, Riobo NA. The Hedgehog signal transduction network. Sci Signal. 2012;5:re6. 16. Bitgood MJ, McMahon AP. Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. Dev Biol. 1995;172:126 –138. 17. Guasti L, Paul A, Laufer E, King P. Localization of Sonic hedgehog secreting and receiving cells in the developing and adult rat adrenal cortex. Mol Cell Endocrinol. 2011;336:117–122. 18. Kang S, Graham JM, Jr., Olney AH, Biesecker LG. GLI3 frameshift mutations cause autosomal dominant Pallister-Hall syndrome. Nat Genet. 1997;15:266 –268. 19. Ng JM, Curran T. The Hedgehog’s tale: developing strategies for targeting cancer. Nat Rev Cancer. 2011;11:493–501. 20. Kim KH, Kim JM, Choi YL, Shin YK, Lee HC, Seong IO, Kim BK, Chae SW, Chung YS, Kim SH. Expression of sonic hedgehog signaling molecules in normal, hyperplastic and carcinomatous endometrium. Pathol Int. 2009;59:279 –287. 21. Souzaki R, Tajiri T, Souzaki M, Kinoshita Y, Tanaka S, Kohashi K, Oda Y, Katano M, Taguchi T. Hedgehog signaling pathway in neuroblastoma differentiation. J Pediatr Surg. 2010;45:2299 –2304. 22. Sandrini R, Ribeiro RC, DeLacerda L. Childhood adrenocortical tumors. J Clin Endocrinol Metab. 1997;82:2027–2031. 23. Sullivan M, Boileau M, Hodges CV. Adrenal cortical carcinoma. J Urol. 1978;120:660 – 665. 24. Lau SK, Weiss LM. The Weiss system for evaluating adrenocortical neoplasms: 25 years later. Hum Pathol. 2009;40:757–768. 25. Rodriguez H, Hum DW, Staels B, Miller WL. Transcription of the human genes for cytochrome P450scc and P450c17 is regulated

J Clin Endocrinol Metab

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

differently in human adrenal NCI-H295 cells than in mouse adrenal Y1 cells. J Clin Endocrinol Metab. 1997;82:365–371. Mermejo LM, Leal LF, Colli LM, Fragoso MCV, Latronico AC, Tone LG, Scrideli C, Tucci S, Martinelli CE, Yunes JA, Mastellaro MJ, Seidinger AL, Brandalise SR, Moreira AC, Ramalho LN, Antonini SR, Castro M. 2013 Noncanonical and canonical Wnt pathways abnormalities in pediatric and adult adrenocortical tumors. In: Society TE ed. The Endocrine Society’s Annual Meeting, Expo. San Francisco, CA: Endocr Rev; MON-17. Boulkroun S, Samson-Couterie B, Golib-Dzib JF, Amar L, Plouin PF, Sibony M, Lefebvre H, Louiset E, Jeunemaitre X, Meatchi T, Benecke A, Lalli E, Zennaro MC. Aldosterone-producing adenoma formation in the adrenal cortex involves expression of stem/progenitor cell markers. Endocrinology. 2011;152:4753– 4763. Werminghaus P, Haase M, Hornsby PE, Schinner S, Schott M, Malendowicz LK, Lammers BJ, Goretzki PE, Muller-Mattheis V, Markus G, Willenberg HS. Hedgehog-signaling is upregulated in non-producing human adrenal adenomas and antagonism of hedgehog-signaling inhibits proliferation of NCI-H295R cells and an immortalized primary human adrenal cell line. J Steroid Biochem Mol Biol. 2013 Kim AC, Barlaskar FM, Heaton JH, Else T, Kelly VR, Krill KT, Scheys JO, Simon DP, Trovato A, Yang WH, Hammer GD. In search of adrenocortical stem and progenitor cells. Endocr Rev. 2009;30: 241–263. Wood MA, Hammer GD. Adrenocortical stem and progenitor cells: unifying model of two proposed origins. Mol Cell Endocrinol. 2011; 336:206 –212. Zubair M, Parker KL, Morohashi K. Developmental links between the fetal and adult zones of the adrenal cortex revealed by lineage tracing. Mol Cell Biol. 2008;28:7030 –7040. Alvarez-Medina R, Cayuso J, Okubo T, Takada S, Marti E. Wnt canonical pathway restricts graded Shh/Gli patterning activity through the regulation of Gli3 expression. Development. 2008;135: 237–247. Probst S, Zeller R, Zuniga A. The hedgehog target Vlk genetically interacts with Gli3 to regulate chondrocyte differentiation during mouse long bone development. Differentiation. 2013;85:121–130. Giordano TJ, Thomas DG, Kuick R, Lizyness M, Misek DE, Smith AL, Sanders D, Aljundi RT, Gauger PG, Thompson NW, Taylor JM, Hanash SM. Distinct transcriptional profiles of adrenocortical tumors uncovered by DNA microarray analysis. Am J Pathol. 2003; 162:521–531. Soon PS, Gill AJ, Benn DE, Clarkson A, Robinson BG, McDonald KL, Sidhu SB. Microarray gene expression and immunohistochemistry analyses of adrenocortical tumors identify IGF2 and Ki-67 as useful in differentiating carcinomas from adenomas. Endocr Relat Cancer. 2009;16:573–583. West AN, Neale GA, Pounds S, Figueredo BC, Rodriguez Galindo C, Pianovski MA, Oliveira Filho AG, Malkin D, Lalli E, Ribeiro R, Zambetti GP. Gene expression profiling of childhood adrenocortical tumors. Cancer Res. 2007;67:600 – 608. Faria AM, Almeida MQ. Differences in the molecular mechanisms of adrenocortical tumorigenesis between children and adults. Mol Cell Endocrinol. 2012;351:52–57. Arimura S, Matsunaga A, Kitamura T, Aoki K, Aoki M, Taketo MM. Reduced level of smoothened suppresses intestinal tumorigenesis by down-regulation of Wnt signaling. Gastroenterology. 2009;137:629 – 638. He J, Sheng T, Stelter AA, Li C, Zhang X, Sinha M, Luxon BA, Xie J. Suppressing Wnt signaling by the hedgehog pathway through sFRP-1. J Biol Chem. 2006;281:35598 –35602. Yanai K, Nakamura M, Akiyoshi T, Nagai S, Wada J, Koga K, Noshiro H, Nagai E, Tsuneyoshi M, Tanaka M, Katano M. Crosstalk of hedgehog and Wnt pathways in gastric cancer. Cancer Lett. 2008;263:145–156. El Khatib M, Kalnytska A, Palagani V, Kossatz U, Manns MP, Malek NP, Wilkens L, Plentz RR. Inhibition of hedgehog signaling

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2013-4098

attenuates carcinogenesis in vitro and increases necrosis of cholangiocellular carcinoma. Hepatology. 2013;57:1035–1045. 42. Kang HN, OH SC, Kim JS, Yoo YA. Abrogation of Gli3 expression suppresses the growth of colon cancer cells via activation of p53. Exp Cell Res. 1997;318:539 –549.

jcem.endojournals.org

9

43. Ng JM, Curran T. The Hedgehog’s tale: developing strategies for targeting cancer. Nat Rev Cancer. 2013;11:493–501. 44. Sandhiya S, Melvin G, Kumar SS, Dkhar SA. The dawn of hedgehog inhibitors: Vismodegib. J Pharmacol Pharmacother. 2013;4:4 –7.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 05 July 2014. at 02:02 For personal use only. No other uses without permission. . All rights reserved.

Sonic hedgehog signaling is active in human adrenal cortex development and deregulated in adrenocortical tumors.

The sonic hedgehog (SHH) pathway plays a key role in rodent adrenal cortex development and is involved in tumorigenesis in several human tissues, but ...
2MB Sizes 0 Downloads 0 Views