Integrated Genomic Analysis of Nodular Tissue in Macronodular Adrenocortical Hyperplasia: Progression of Tumorigenesis in a Disorder Associated with Multiple Benign Lesions
Madson Q. Almeida, Michelle Harran, Eirini I. Bimpaki, Hui-Pin Hsiao, Anelia Horvath, Chris Cheadle, Tonya Watkins, Maria Nesterova, and Constantine A. Stratakis
Section on Endocrinology and Genetics (M.Q.A., M.H., E.I.B., H .- P.H., A.H., M.N., C.A.S.), Program on Developmental Endocrinology and Genetics and Pediatric Endocrinology Inter-institute Training Program (C.A.S.), Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892; and Genomics Core (C.C., T.W.), Division of Allergy and Clinical Immunology, School of Medicine, Johns Hopkins University, Baltimore, Maryland 21224
Context: Massive macronodular adrenocortical disease or ACTH-independent macronodular ad- renal hyperplasia (AIMAH) is a clinically and genetically heterogeneous disorder.
Objective and Design: Whole-genome expression profiling and oligonucleotide array comparative genomic hybridization changes were analyzed in samples of different nodules from the same patients with AIMAH. Quantitative RT-PCR and staining were employed to validate the mRNA array data.
Results: Chromosomal gains were more frequent in larger nodules when compared with smaller nodules from the same patients. Among the 50 most overexpressed genes, 50% had a chromosomal locus that was amplified in the comparative genomic hybridization data. Although the list of most over- and underexpressed genes was similar between the nodules of different size, the gene set enrichment analysis identified different pathways associated with AIMAH that corresponded to the size; the smaller nodules were mainly enriched for metabolic pathways, whereas p53 signaling and cancer genes were enriched in larger nodules. Confirmatory studies demonstrated that BCL2, E2F1, EGF, c-KIT, MYB, PRKCA, and CTNNB1 were overexpressed in the larger nodules at messenger and/or protein levels. Chromosomal enrichment analysis showed that chromosomes 20q13 and 14q23 might be involved in progression of AIMAH from smaller to larger tumors.
Conclusion: Integrated transcriptomic and genomic data for AIMAH provides supporting evidence to the hypothesis that larger adrenal lesions, in the context of this chronic, polyclonal hyperplasia, accumulate an increased number of genomic and, subsequently, transcript abnormalities. The latter shows that the disease appears to start with mainly tissue metabolic derangements, as suggested by the study of the smaller nodules, but larger lesions showed aberrant expression of oncogenic pathways. (J Clin Endocrinol Metab 96: E728-E738, 2011)
A CTH-independent macronodular adrenal hyperpla- sia (AIMAH), also known as massive macronodular adrenal disease, is known as a rare cause of Cushing’s syndrome (CS), accounting for less than 1% of all cases of
ACTH-independent CS (1-3). However, the diagnosis of AIMAH associated with subclinical hypercortisolism has lately increased. This is not surprising, because approxi- mately 10-15% of the presumably clinically silent adrenal
Abbreviations: AIMAH, ACTH-independent macronodular adrenal hyperplasia; CGH, com- parative genomic hybridization; CS, Cushing’s syndrome; DEX, dexamethasone; ECM, extracellular matrix; GSEA, gene set enrichment analysis; oligo-aCGH, oligonucleotide array CGH; PKA, protein kinase A.
masses that are detected after various imaging studies are indeed bilateral (4).
AIMAH is a sporadic disease in the majority of cases, but several families with AIMAH, in whom the disease was inherited as an autosomal dominant trait, have been reported (5-7). Cortisol production in AIMAH can be regulated by the aberrant expression of G protein-coupled receptors other than ACTH, such as those for glucose- dependent insulinotropic peptide (GIPR), ß-adrenergic re- ceptors (ß-AR), vasopressin (V2-V3-vasopressin recep- tor), serotonin (5-HT7 receptor), angiotensin II (AT1R), glucagon (GCGR), and LH/human chorionic gonadotro- pin (LH/hCGR) (8, 9).
A number of genetic abnormalities have been detected in adrenocortical adenomas and carcinomas (10-16). Bourdeau et al. (17) showed somatic losses of the 17q22-24 region and protein kinase A (PKA) subunit and enzymatic activity changes in AIMAH, demonstrating that cAMP/PKA signaling is altered in this disorder, not unlike the case in primary pigmented nodular adrenocor- tical disease caused by PRKAR1A-inactivating mutations or sporadic adrenal tumors that harbor 17q (the PRKAR1A locus) losses (16). Whole-genome expression studies in AIMAH and primary pigmented nodular adre- nocortical disease, the main forms of bilateral adrenal hyperplasia, confirmed the involvement of the cAMP sig- naling pathway (18, 19) but also pointed to the overex- pression of genes that regulate or are part of the Wnt sig- naling pathway such as WISP2, GSK3B, and CTNNB1 (20, 21).
Previous cytogenetic analysis of adrenal tumors iden- tified no gains or losses in adenomas less than 5 cm by comparative genomic hybridization (CGH) (10). Adeno- mas more than 5 cm size had one chromosomal gain or loss or both, whereas adrenal carcinomas (range, 7-20 cm) frequently had losses involving chromosomes 2, 11q, and 17p and gains at chromosomes 4 and 5 (10). Sidhu et al. (11) also found a significantly high frequency of CGH changes in adrenal carcinomas when compared with ad- enomas. Although a strong correlation between the size of adrenal tumor and the number of cytogenetic changes has been demonstrated by several reports (10-14), the hy- pothesis that genetic changes in adrenal hyperplasias cor- relates with the nodule size in the same patients (22) re- mains to be explored.
In the present study, we tested this hypothesis after microdissection of different nodules from each of two un- related patients with AIMAH and analysis of the whole genome at the DNA and RNA levels. The data point to the significance of size of adrenocortical lesions as a major determining factor of their tumor potential. It is not only the number of changes that increased with size; the data
also show a qualitative difference in the pathways altered in smaller vs. larger nodules within AIMAH, suggesting a progression from a relatively simple metabolic derange- ment at the beginning of the process to the gradual in- volvement of oncogenic pathways in larger nodules.
Subjects and Methods
Subjects
Two patients with AIMAH were admitted to the National Institutes of Health (NIH) Warren Magnuson Clinical Center from 2000-2006 for the work-up and treatment of adrenocor- tical tumors under protocol 00-CH-160. The Eunice Kennedy Shriver National Institute of Child Health and Human Devel- opment Institutional Review Board approved this study, and informed consents were obtained from the two subjects.
Patient 1
A 42-yr-old male patient with a history of mild hypertension and weight gain (60 lbs over 10 yr), plethoric face, and violaceous striae was admitted to the NIH Clinical Center to investigate CS. The hormonal evaluation revealed absence of plasma cortisol suppression after 1 mg overnight dexamethasone (DEX) test and, after the DEX-ovine corticotropin-releasing hormone test, high 24 h urinary free cortisol levels (159.1 µg; normal, <90 µg/24 h), abnormal midnight serum cortisol levels (11.6 µg/dl), and suppressed ACTH levels (<5 pg/ml). Bilateral enlargement of the adrenal glands with macronodules was seen on the com- puter tomography scan. The patient was diagnosed with ACTH- independent CS and underwent bilateral adrenalectomy (Fig. 1A).
Patient 2
A 42-yr-old female patient presented with a 4-yr history of secondary amenorrhea, easy bruising, hirsutism, weight gain (25 lbs), muscle weakness, and behavioral changes. Biochemical evaluation showed absence of plasma cortisol suppression after 1 mg overnight DEX test and, after the DEX-ovine corticotropin- releasing hormone test, high 24-h urinary free cortisol levels (270 µg; normal, <90 µg/24 h) and suppressed ACTH levels (<5 pg/ml). Computer tomography scan revealed bilateral adrenal macronodules. She was diagnosed with ACTH-independent CS and underwent laparoscopic bilateral adrenalectomy (Fig. 2A).
Aberrant G protein-coupled receptor expression (GIPR, AGTR1,ADRB1,-2,and-3,AVPR1,-2,and-3,HTR7, GCGR, and LHCGR) was not detected in tissue samples from both pa- tients; however, these patients were not tested in vivo for their illegitimate receptor responses.
Hormone assays
Plasma cortisol and ACTH levels were measured as described elsewhere (23). Urinary free cortisol excretion was measured by direct RIA (24). The intraassay coefficient of variation was 5%, and the interassay coefficient of variation was 10%.
Comparative genome hybridization
Adrenal tissue was collected from both patients, and the nod- ules were macroscopically dissected from the surrounding tissue and snap frozen in liquid nitrogen. DNA extraction was per-
| Size | ||||
|---|---|---|---|---|
| Aberrations | Nodule 1 | Nodule 2 | Nodule 3 | Nodule 4 |
| Amplification | 8 | 18 | 28 | 30 |
| Deletion | 7 | 15 | 24 | 19 |
| Total | 15 | 33 | 52 | 49 |
A
B
C
Nodule 1
Nodule 4
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
×
Y
13
14
15
16
17
18
19
20
21
22
x
Y
..
formed in the four nodules (microdissected tissue) from patient 1 (sizes were 1, 3, 5, and 7 cm, respectively) and in two available from patient 2 (sizes were 1 and 3.5 cm, respectively) using the QIAGEN DNeasy blood and tissue kit (QIAGEN, Valencia, CA). Oligonucleotide array CGH (oligo-aCGH) by GeneDX was used to identify regions of gains or losses throughout the chromosomes. A set of 105,000 oligonucleotide probes was used to cover 37 kb of the human genome sequence (22 autosomes and X and Y chromosomes) per probe (GenomeDx version 2.0). Agi- lent CGH Analytics version 3.5 software was used to analyze the oligo-aCGH data.
Microarray analysis
Total RNA extraction was performed from seven different- sized nodules using the Trizol reagent method (Invitrogen, Carls- bad, CA). The RNA samples were further purified using the RNeasy columns (QIAGEN), and the quality of RNA was as- sessed using the Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA). Three commercially available pools of human ad- renal total RNA (Clontech, Mountain View, CA; BioChain, Hayward, CA; and Ambion, Austin, TX) were employed as ref- erence samples. Preparation of cRNA from total RNA, hybrid- ization in Sentrix HumanRef-8 Expression BeadChips, scanning, and image analysis was done as previously described (25). Anal- ysis of Illumina data was performed using Illumina BeadStudio software (Illumina, San Diego, CA), which returns the trimmed mean average intensity for each single gene probe type (nonnor- malized). Any gene consistently with a P detection value above 0.01 for all samples was eliminated from further analysis. Z- transformation for normalization was performed for each Illu- mina sample/array (25, 26). First, the raw intensity data for each
sample was log10-transformed, and Z scores were calculated by subtracting the overall average gene intensity (within a single experiment) from the raw intensity data for each gene, then, dividing that result by the SD of all of the measured intensities. Changes in gene expression (ratio) between different Z-trans- formed datasets (nodules compared with the average of the nor- mal adrenal pools) were calculated as differences between the corresponding Z scores and then divided by the SD of each Z difference dataset (25). A 2-fold change was employed as the cutoff value to identify over- and underexpressed genes in adre- nal nodules in relation to normal adrenal pools. Microarray data are in compliance with the minimal information about a mi- croarray experiment (MIAME) format. The raw and normalized microarray data reported in this paper have been deposited in the Gene Expression Omnibus (GEO) database (accession no. GSE25031).
Analysis of the mRNA profiling
The functional analysis of the whole-genome transcriptome profiling was performed using the Database for Annotation, Vi- sualization and Integrated Discovery (DAVID) Bioinformatic Resources 2008 (National Institute of Allergy and Infectious Diseases, NIH, http://david.abcc.ncifcrf.gov/home.jsp) (27, 28). The lists of genes (induced or repressed) were submitted to the DAVID database (http://david. abcc.ncifcrf.gov); genes are there clustered according to a series of common keywords. The pro- portion of each keyword in the list is compared with the one in the whole genome, making it possible to compute P values and enrichment scores (geometric mean of the inverse log of each P value). The detailed information of gene alterations was system- atically reported on KEGG pathways.
| Aberrations | Nodule 1 (small) | Nodule 2 (large) |
|---|---|---|
| Amplification | 24 | 59 |
| Deletion | 15 | 13 |
| Total | 39 | 72 |
A
B
30-05-72-7
C
Nodule 1
Nodule 2
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
x
Y
13
14
15
16
17
18
19
20
21
22
x
Y
Gene set enrichment analysis (GSEA) was performed by GSEA Software version 2.0 (www.broad.mit.edu) in pairwise comparisons (29). Gene expression results derived from mi- croarray experiments were correlated with chromosome gene sets. GSEA was performed using gene set permutation type as default, and the number of permutations was set to 1000. Statistical significance levels were defined as nominal P value <0.05.
Real-time quantitative RT-PCR
Quantitative real-time PCR was performed in the ABI Prism 7700 sequence detector using using the Oncogene and Tumor Suppressor Genes PCR Array Plate (Supplemental Table 1, pub- lished on The Endocrine Society’s Journals Online web site at http:/jcem.endojournals.org; SABiosciences, Frederick, MD). An average of three commercially available pools of human ad- renal total RNA (Clontech, BioChain, and Ambion) was used as control. Relative quantification was performed using the 2-AACT method (30).
Immunohistochemistry
All immunohistochemistry was performed in collaboration with Histoserv, Inc. (Germantown, MD) using standard proce- dures. The following primary antibodies were used: CTNNB1 (ab6302; Abcam), BCL-2 (C-2, sc-7382; Santa Cruz Biotechnol- ogy, Santa Cruz, CA), and c-KIT (E1, sc-17806), Ca PKA sub- unit (C-20, sc-903), CB PKA subunit (C-20, sc-904), and PRKX (gift from Dr. Robert M. Kotin, National Heart, Lung, and Blood Institute, NIH, Bethesda, MD). The following grading system was used for staining evaluation: negative (absence of expres-
sion), weak staining (from 1-25% of immunoreactive cells), and strong staining (>25% of immunoreactive cells). Immunoreac- tivity for BCL-2, c-KIT, and PKA subunits was evaluated in the cytoplasm, whereas CTNNB1 staining was assessed in the nucleus.
PKA activity
PKA enzymatic activity was measured in tissue extracts as previously described (31).
Statistical analysis
All statistical analyses were performed with the SPSS version 16.0 (SPSS Inc., Chicago, IL). Continuous data are expressed as mean ± SD. All the experiments were performed in triplicate. A two-sample t test was used for paired samples. A P value <0.05 was considered significant.
Results
Oligo-aCGH analysis
Oligo-aCGH was used to map DNA copy number ab- errations that occur in nodules of different sizes. A sum- mary of all the changes is presented for patients 1 and 2 (Figs. 1 and 2, respectively). Chromosomal gains were more frequent in larger nodules when compared with smaller nodules. In patient 1, 30 amplifications were de- tected in nodule 4 (larger), whereas only eight amplifica-
A
NI adrenal pool 1
NI adrenal pool 2
NI adrenal pool 3
Size
Nodules
1234567
CRELD1-cvsteine-rich with EGF-like domains 1 (CRELD1) , transcript variant 1. mRNA.
NEKBIA-nuclear factor of kamma light polypeptide dene enhancer in B-cells inhibitor, alpha (NEKBIA) , MRNA. MERTK-c-mer proto-oncogene tyrosine kinase (MERTK) . MRNA.
LEF1-1vmohoid enhancer-binding factor 1 (LEF1) . MRNA.
MYLIP-mvosin regulatory light chain interacting protein (MYLIP) . MRNA.
ZFP36-zinc finger protein 36. C3H type. homolog (mouse) (ZFP36) . MRNA.
FOS-V-fos FBJ murine osteosarcoma viral oncogene homolog (FOS) , MRNA. EGR1-earlv growth response 1 (EGR1) . mRNA.
RASD1-RAS, dexamethasone-induced 1 (RASD1) , MRNA.
CASP9-caspase 9. anontosis-related cvsteine peptidase (CASP9) , transcript variant alpha, mRNA.
ZNF331-zinc finger protein 331 (ZNF331) . MRNA.
WDR63-W repeat domain 63 (WDR63) , MRNA.
UGCG-UDP-alucose ceramide alucosyltransferase (UGCG) , MRNA. NSUN5C-NOL1/NOP2/Sun domain family. member 5C (NSUN5C) , transcript variant 4. mRNA.
TTLL3-tubulin tvrosine ligase-like family. member 3 (TTLL3) , transcript variant 2, mRNA.
SLIT2-slit homoloa 2 (Drosophila) (SLIT2) . MRNA.
DMD-dvstrophin (muscular dystrophy. Duchenne and Becker tvnes) (DMD) . transcript variant Dp40, mRNA.
FGF9-fibroblast growth factor 9 (alia-activating factor) (FGF9) . MRNA.
NR4A2-nuclear receptor subfamily 4. arow A. member 2 (NR4A2) , transcript variant 1, mRNA.
WNT4-wingless-tve MMTV integration site family, member 4 (WNT4) , MRNA.
TOB1-transducer of ERBB2. 1 (TOB1) . RNA.
GCN5L2-GCN5 general control of amino-acid synthesis 5-like 2 (veast) (GCN5L2) . MRNA.
SLC25A37-solute carrier family 25. member 37 (SLC25A37) , transcript variant 1. mRNA.
ERN1-endoplasmic reticulum to nucleus signalling 1 (ERN1) . transcript variant 2. mRNA. SUV420H2-suppressor of variegation 4-20 homolog 2 (Drosophila) (SUV420H2) , MRNA. PCP4-Purkinie cell protein 4 (PCP4) . MRNA.
FOXL2-forkhead box L2 (FOXL2) . MRNA.
NR4A3-nuclear receptor subfamily 4. arow A. member 3 (NR4A3) , transcript variant 4, mRNA.
NR4A3-nuclear receptor subfamily 4. arow A. member 3 (NR4A3) , transcript variant 3, mRNA.
SRRM2-serine/arginine repetitive matrix 2 (SRRM2) , MRNA.
RGS16-regulator of G-protein signalling 16 (RGS16) . MRNA.
FOSB-FBJ murine osteosarcoma viral oncogene homolog B (FOSB) , MRNA.
MUC20-mucin 20 (MUC20) . MRNA.
CDH12-cadherin 12. tvwe 2 (N-cadherin 2) (CDH12) . MRNA.
DOM3Z-dom-3 homoloa Z (C. elegans) (DOM3Z) . MRNA.
Pathway Analysis
B
| Pathways | Genes (n) | Fold enrich. | p o a |
|---|---|---|---|
| Nodule 1 (smaller) vs. Normal adrenal | |||
| Circadian rhythm | 6 | 8.4 | 0.0001 |
| Starch and sucrose metabolism | 9 | 3.7 | 0.002 |
| Aminosugars metabolism | 6 | 3.6 | 0.02 |
| MAPK signaling pathway | 23 | 1.6 | 0.03 |
| ECM-receptor interaction | 10 | 2.2 | 0.03 |
| Nicotinamide metabolism | 5 | 3.8 | 0.04 |
C
| Pathways | Genes (n) | Fold enrich. | p a |
|---|---|---|---|
| Nodule 7 (larger) vs. Normal adrenal | |||
| Ribosome | 30 | 7.1 | 0.0001 |
| p53 signaling pathway | 11 | 3.3 | 0.001 |
| Circadian rhythm | 4 | 6.3 | 0.02 |
| Pathways in cancer | 24 | 1.5 | 0.04 |
| ECM-receptor interaction | 9 | 2.2 | 0.05 |
| Colorectal cancer | 9 | 2.2 | 0.05 |
| MAPK signaling pathway | 20 | 1.5 | 0.06 |
FIG. 3. A, Heatmap visualization of gene expression data displaying differentially expressed genes in normal adrenals and AIMAH nodules from patient 2; B and C, functional analysis of whole-genome transcriptome profiling of nodule 1 (small) and nodule 7 (large) compared with normal adrenal tissue. The array functional analysis was performed using DAVID Bioinformatics Resources 2008, National Institute of Allergy and Infectious Diseases, NIH (http://david.abcc.ncifcrf.gov/home.jsp).
tions were found in nodule 1 (smaller) (Fig. 1, B and C). Similarly, a positive correlation between nodule size and number of gains of genetic material was observed for pa- tient 2 (Fig. 2, B and C). Copy number increases (ampli- fications) were found for 59 chromosomal regions in the large nodule from patient 2 but only for 24 chromosomal regions in the smaller nodule. Chromosomal losses were not as frequent as gains in both cases (35.8 vs. 64.2%). The number of chromosomal deletions was also associated with the nodule size in patient 1.
Whole transcriptome profiling
Microarray analysis was performed in three normal adrenal pools and in seven adrenal nodules from the same subject with AIMAH (patient 2) using the Illumina Bea- darrays system. All genes were displayed in the heat maps constructed by processing the data using unsupervised hi-
erarchical clustering (Fig. 3A). The most differentially ex- pressed genes in AIMAH were also compared with normal adrenal cortex (Fig. 3A and Supplemental Table 2). Among the 50 most overexpressed genes, 50% of them were found to be amplified in the CGH data. However, only nine of the 50 most underexpressed genes (18%) were located in regions of chromosomal losses.
The overexpressed genes associated with chromosomal gains included TFPI2 [7q22; ratio (log10) 6.45], TNFRSF12A (16p13.3; ratio 6.14), CDH12 [5p14.3; ratio 4.9], CASP9 (1p36.21; ratio 4.6), FOS (14q24.3; ratio 4.4), and FOSB (19q13.32; ratio 4.5), among others (Supplemental Table 2). On the other hand, some of the top overexpressed genes not associated with gene amplification were TSPAN8 (12q14.1-q21.1; ratio 9.0), NR4A2 (2q22-q23; ratio 7.17), IL7R (5p13; ratio 6.6), RGS1(1q31; ratio 4.8), and FGF9 (3q11-q12, ratio 4.3).
Wnt signaling activation in AIMAH was demonstrated in a previous gene array study from our group (21). In the current analysis, WNT4 was overexpressed similarly in all AIMAH nodules [nodule 1, ratio (log10) 4.0; nodule 2 = 4.5; nodule 3 = 3.8; nodule 4 = 3.5; nodule 5 = 3.8; nodule 6 = 3.3; and nodule 7 = 4.2]. Additionally, WISP1 was also overexpressed in AIMAH, regardless of the nod- ule size (nodule 1 = 2.2; nodule 2 = 1.7; nodule 3 = 1.8; nodule 4 = 1.8; nodule 5 = 2.5; nodule 6 = 1.5; and nodule 7 = 1.7).
Although the list of most over- and underexpressed genes was similar between nodules, the GSEA identified different pathways associated with AIMAH depending on the size of the examined nodules (Fig. 3B and Supplemen- tal Table 3). The functional analysis of the whole-genome transcriptome profiling was performed using the DAVID Bioinformatic Resources 2008 (National Institute of Al- lergy and Infectious Diseases, NIH) (28). Circadian rhythm and metabolic pathways (sucrose and aminosug- ars metabolism) were the most significantly enriched pathways in nodule 1 (smaller) in relation to normal adrenal (P < 0.05) (Fig. 3B). MAPK signaling and ex- tracellular matrix (ECM)-interaction pathways were also overexpressed in nodule 1 (P = 0.03). Interestingly, the functional analysis of nodule 7 (the largest one) revealed different pathways associated with AIMAH- tumor signature. An enrichment for p53 signaling (P = 0.001), other cancer pathways (P = 0.04), and, espe- cially, genes involved in colorectal cancer (P = 0.05) were found in nodule 7 (Fig. 3B).
Correlation of chromosomal enrichment by gene set enrichment analysis (GSEA) and cytogenetic data
GSEA was performed and correlated with CGH data. GSEA revealed chromosome 20q13 and 14q23 enrich- ment in nodule 7 (large) when compared with nodule 1 (small) gene signature (P = 0.01) (Table 1). Both chro- mosome 20q13 and 14q23 regions had a higher amplifi- cation in nodule 7 in relation to nodule 1. The enriched gene set in chromosome 20q13 included LAMA5, BIRC7, PKIG, PRIC285, and RGS19 (Table 1). In chromosome 14q23, RHOJ, TMEM30B, and PRKCH mainly ac- counted for the chromosomal enrichment of this amplified region.
Expression of oncogenes and PKA catalytic subunits in AIMAH nodules of different size
A quantitative RT-PCR array carrying 84 genes of on- cogenic pathways was performed to confirm the microar- ray data and evaluate which genes might be involved in tumor progression in AIMAH (Supplemental Table 1).
| Enrichment (%) | |
|---|---|
| Enriched genes at chromosomal 20q13 | |
| PFDN4: prefoldin subunit 4 | 47.3 |
| ZNF334, zinc finger protein 334 (ZNF334) | 73.8 |
| CSE1 liter, CSE1 chromosome segregation | 10.6 |
| 1-like (yeast) | |
| SS18L1, synovial sarcoma translocation gene on chromosome 18-like 1 | 10.8 |
| SLC2A4RG, SLC2A4 regulator | 14.5 |
| TCEA2, transcription elongation factor A (SII) | 92.9 |
| GTPBP5, GTP binding protein 5 | 16.9 |
| PPP1R3D, protein phosphatase 1, regulatory subunit 3D | 15.5 |
| PRIC285, peroxisomal proliferator-activated receptor A interacting complex 285 | 24 |
| RGS19, regulator of G-protein signaling 19 | 18 |
| LAMA5, laminin, & 5 | 100 |
| PKIG, protein kinase (cAMP-dependent, | 40 |
| catalytic) inhibitor y | |
| UBE2V1, ubiquitin-conjugating enzyme E2 variant 1 | 16.6 |
| BIRC7, baculoviral IAP repeat-containing 7 | 69.1 |
| (livin) | |
| TFAP2C, transcription factor AP-2 y | 13.5 |
| Enriched genes at chromosomal 14q23 | |
| TMEM30B, transmembrane protein 30B | 44.6 |
| RHOJ, ras homolog gene family, member J | 45.7 |
| NMNAT1, nicotinamide nucleotide | 10 |
| adenylyltransferase 1 | |
| VTI1B, vesicle transport through interaction | 26.2 |
| with t-SNAREs homolog 1B (yeast) | |
| PRKCH, protein kinase C, n | 11 |
Gene expression was compared between two groups, the smaller (nodules 1 and 2) and the larger (nodules 6 and 7) tumors (Fig. 4A). Overexpressed oncogenes in larger AIMAH nodules (and, thus, possibly associated with tu- mor progression) were BCL2, E2F1, EGF, c-KIT, MYB, PRKCA, and SERPINB5, among others. At protein level, BCL2 and c-KIT expression was strong in large nodules and weak in small nodules (Fig. 4B). Similarly, nuclear CTNNB1 staining was strong in a large nodule and weak in a small nodule (Fig. 4C).
PKA activity declined in larger nodules
PKA activity was measured in four AIMAH nodules (Fig. 5A). After cAMP exposure, smaller nodules had a tendency to have higher total kinase activity compared with larger ones (39.453 ± 4.682 vs. 32.637 + 1261 cpm/mg protein). Expression of PKA catalytic subunits (Ca, CB, and PRKX) decreased in macronodules when compared with micronodules (Fig. 5, B-D); this decrease in expression of the main catalytic subunits, may account
A
38
Large/Small Nodules
p= 0.001
p= 0.0004
*
36
*
34
32
mRNA level (fold change)
p= 0.001 *
30
p= 0.003 *
28
26
p= 0.005 *
24
p= 0.001 *
22
p= 0.03 *
p= 0.001 *
p= 0.01
T
20
*
18
16
14
p= 0.03
p= 0.03 *
12
*
p= 0.003 *
10
p= 0.02 *
8
p= 0.003 *
6
p= 0.004
p= 0.01
T
4
p= 0.03 *
p= 0.03 *
*
p= 0.01
p= 0.03
*
p= 0.02 *
p= 0.04
p= 0.004 *
2
*
T
T
*
+
*
2
T
0
BCL2
CDKN2A
CDKN2B
CDKN3
E2F1
EGF
ERBB2
ETS1
HGF
HIC1
IGF2R
KIT
KRAS
MET
MOS
MYB
MYCN
NRAS
PIK3C2A
PRKCA
RET
ROS1
SERPINB5
B
H&E
BCL2
C-KIT
Small nodule
Large nodule
C
-
CTNNB1
Small nodule
Large nodule
A
50000
45000
PKA activity (cpm/mg protein)
40000
35000
30000
25000
Basal
20000
CAMP
15000
10000
5000
0
Small nodules
Large nodules
Small nodule
Large nodule
B
Ca PKA subunit
C
CB PKA subunit
D
PRKX
for the tendency for lower total PKA activity in larger nodules.
Discussion
AIMAH is a clinically and genetically heterogeneous dis- order that can be associated with aberrant hormone re- ceptors (8, 9, 18). It is frequently associated with subclin- ical hypercortisolism or atypical CS (1, 2). Given the
genetic heterogeneity of AIMAH, genomic integrated approaches can re- veal novel aspects involved in tumor progression and offer the possibility of more informed clinical decision mak- ing and may yield novel therapeutic targets (15, 32-34). In this study, our data set represents, to our knowledge, the first fully integrated analysis of AIMAH with whole-genome profiling and oligo-aCGH in nodules of different size from the same patient.
Previous cytogenetic studies in adre- nal tumors showed that the number of chromosomal aberrations is strongly correlated with the size of the tumor (10, 11, 13). Here, we first demonstrated a positive association between chromo- somal gains and the size of the nodules within the same adrenal gland from AIMAH patients. With the direct com- parison of gene profiling and CGH re- sults performed on the same samples, an overlap between significantly overex- pressed genes and chromosomal ampli- fication was established in about half of the genes and respective genetic loci.
The tetraspanin 8 (TSPAN8) gene was the most overexpressed transcript in all AIMAH nodules. Interestingly, TSPAN8 overexpression was not asso- ciated with gene amplification in the CGH data. Tetraspanin 8 is a cell surface glycoprotein that is known to complex with integrins and mediates signal trans- duction events that play a role in the reg- ulation of cell development, activation, growth, and motility (35, 36). Tetraspa- nin 8 is an important angiogenesis in- ducer within tumors and also in tumor- free tissues (37).
Although the most over- and under- expressed genes were similar in all nodules, the functional analysis revealed different gene set enrichment that cor- responded to the nodule size. The smallest nodules were mostly enriched for metabolic pathways, ECM interac- tion, and MAPK signaling pathways, whereas a high sta- tistical association with p53 signaling and cancer path- ways was found in the largest nodules.
It should be noted that, although we have identified activation of oncogenic pathways in AIMAH, there is no clinical evidence, so far, supporting a high risk of malig-
nancy among patients with AIMAH. Confirmatory stud- ies demonstrated that BCL2, E2F1, EGF, c-KIT, MYB, PRKCA, and CTNNB1 were overexpressed in larger nod- ules at the messenger and/or protein levels but, AIMAH is a polyclonal and genetically heterogeneous disorder (18, 38). Chromosomal gains and enrichment of oncogenic pathways apparently provide functional advantage in longstanding AIMAH but may not necessarily predispose to cancer.
Chromosomal enrichment analysis by GSEA showed that chromosomes 20q13 and 14q23 were overexpressed in larger nodules when compared with smaller ones. Du- plication of 14q23 has been previously associated with congenital abnormalities, and 14q is frequently amplified in various sarcomas (39, 40). On the other hand, chro- mosomal 20q13 amplification has been associated with tumor progression and poor prognosis in ovarian and breast cancer (41-43). In our study, an enriched gene set in chromosome 20q13 revealed important genes that can be associated with AIMAH progression, such as LAMA5, BIRC7, PKIG, PRIC285, and RGS19. Laminin @5 protein encoded by LAMA5 gene belongs to the a-subfamily of laminin chains and is a major com- ponent of basement membranes. Recently, Paquet-Fi- field et al. (44) showed that laminin «5 promotes epi- thelial cell proliferation and skin regeneration by modifying the ECM microenvironment.
Wnt signaling pathway is a master regulator of tumor- igenesis driven by PKA dysregulation (21, 26, 45-47). We demonstrated here that WISP1 and WNT4 were signifi- cantly overexpressed in AIMAH nodules, as previously reported (21). Interestingly, nuclear CTNNB1 staining was found to be higher in macronodules when compared with micronodules. Furthermore, PKA catalytic subunits were strongly expressed in both micro- and macronodules, but their expression was lower in macronodules at the protein level. PKA signaling involvement in AIMAH was initially shown by our group (17); AIMAH displayed so- matic 17q22-24 allelic losses and an important increase in cAMP responsiveness.
In conclusion, our data provide a unique public re- source for the molecular features of AIMAH; in this study, we showed for the first time that larger nodules in this polyclonal disorder represented a progression to a more tumor-like profile with an increased number of chromo- somal aberrations and an expression signature that was enriched with cancer-involved pathways and oncogenic transformation. This is supportive of the original hypoth- esis (22) that proposed that adrenal tissue was not differ- ent from other tissues in the process of carcinogenesis; just like polyps in colonic tissues, smaller polyclonal nodules in adrenal cortex, when they grow, have the capacity of ac-
tivating oncogenic pathways and, thus, develop the po- tential, at least in theory, of monoclonal neoplastic transformation.
Acknowledgments
Address all correspondence and requests for reprints to: Constantine A. Stratakis, M.D., D(med).Sci., Section on En- docrinology and Genetics, Program on Developmental Endo- crinology and Genetics, Eunice Kennedy Shriver National In- stitute of Child Health and Human Development, National Institutes of Health, Building 10, CRC, Room 1-3330, 10 Center Drive, MSC1103, Bethesda, Maryland 20892. E-mail: stratakc@mail.nih.gov.
This work was supported by U.S. National Institutes of Health, Eunice Kennedy Shriver National Institute of Child Health and Human Development intramural project Z01-HD- 000642-04 (to C.A.S.).
Current address for H .- P.H .: Department of Pediatrics, Ka- ohsiung Municipal HsiaoKang Hospital, Kaohsiung Medical University, Taiwan.
Disclosure Summary: The authors have nothing to disclose.
References
1. Lacroix A 2009 ACTH-independent macronodular adrenal hyper- plasia. Best Pract Res Clin Endocrinol Metab 23:245-259
2. Lacroix A, Bourdeau I, Lampron A, Mazzuco TL, Tremblay J, Hamet P 2010 Aberrant G-protein coupled receptor expression in relation to adrenocortical overfunction. Clin Endocrinol (Oxf) 73: 1-15
3. Stratakis CA 2009 New genes and/or molecular pathways associ- ated with adrenal hyperplasias and related adrenocortical tumors. Mol Cell Endocrinol 300:152-157
4. Barzon L, Sonino N, Fallo F, Palu G, Boscaro M 2003 Prevalence and natural history of adrenal incidentalomas. Eur J Endocrinol 149:273-285
5. Findlay JC, Sheeler LR, Engeland WC, Aron DC 1993 Familial ad- renocorticotropin-independent Cushing’s syndrome with bilateral macronodular adrenal hyperplasia. J Clin Endocrinol Metab 76: 189-191
6. Minami S, Sugihara H, Sato J, Tatsukuchi A, Sugisaki Y, Sasano H, Wakabayashi I 1996 ACTH independent Cushing’s syndrome oc- curring in siblings. Clin Endocrinol (Oxf) 44:483-488
7. Gagliardi L, Hotu C, Casey G, Braund WJ, Ling KH, Dodd T, Manavis J, Devitt PG, Cutfield R, Rudzki Z, Scott HS, Torpy DJ 2009 Familial vasopressin-sensitive ACTH-independent ma- cronodular adrenal hyperplasia (VPs-AIMAH): clinical studies of three kindreds. Clin Endocrinol (Oxf) 70:883-891
8. Lacroix A, Baldacchino V, Bourdeau I, Hamet P, Tremblay J 2004 Cushing’s syndrome variants secondary to aberrant hormone recep- tors. Trends Endocrinol Metab 15:375-382
9. Lacroix A, Ndiaye N, Tremblay J, Hamet P 2001 Ectopic and ab- normal hormone receptors in adrenal Cushing’s syndrome. Endocr Rev 22:75-110
10. Kjellman M, Kallioniemi OP, Karhu R, Höög A, Farnebo LO, Auer G, Larsson C, Bäckdahl M 1996 Genetic aberrations in adrenocor- tical tumors detected using comparative genomic hybridization cor- relate with tumor size and malignancy. Cancer Res 56:4219-4223
11. Sidhu S, Marsh DJ, Theodosopoulos G, Philips J, Bambach CP, Campbell P, Magarey CJ, Russell CF, Schulte KM, Röher HD, Del-
bridge L, Robinson BG 2002 Comparative genomic hybridization analysis of adrenocortical tumors. J Clin Endocrinol Metab 87: 3467-3474
12. Figueiredo BC, Stratakis CA, Sandrini R, DeLacerda L, Pianovsky MA, Giatzakis C, Young HM, Haddad BR 1999 Comparative genomic hybridization analysis of adrenocortical tumors of child- hood. J Clin Endocrinol Metab 84:1116-1121
13. Zhao J, Speel EJ, Muletta-Feurer S, Rütimann K, Saremaslani P, Roth J, Heitz PU, Komminoth P 1999 Analysis of genomic altera- tions in sporadic adrenocortical lesions. Gain of chromosome 17 is an early event in adrenocortical tumorigenesis. Am J Pathol 155: 1039-1045
14. Dohna M, Reincke M, Mincheva A, Allolio B, Solinas-Toldo S, Lichter P 2000 Adrenocortical carcinoma is characterized by a high frequency of chromosomal gains and high-level amplifications. Genes Chromosomes Cancer 28:145-152
15. Szabó PM, Tamási V, Molnár V, Andrásfalvy M, Tömböl Z, Farkas R, Kövesdi K, Patócs A, Tóth M, Szalai C, Falus A, Rácz K, Igaz P 2010 Meta-analysis of adrenocortical tumour genomics data: novel pathogenic pathways revealed. Oncogene 29:3163-3172
16. Bertherat J, Groussin L, Sandrini F, Matyakhina L, Bei T, Stergio- poulos S, Papageorgiou T, Bourdeau I, Kirschner LS, Vincent-De- jean C, Perlemoine K, Gicquel C, Bertagna X, Stratakis CA 2003 Molecular and functional analysis of PRKAR1A and its locus (17q22-24) in sporadic adrenocortical tumors: 17q losses, somatic mutations, and protein kinase A expression and activity. Cancer Res 63:5308-5319
17. Bourdeau I, Matyakhina L, Stergiopoulos SG, Sandrini F, Boikos S, Stratakis CA 2006 17q22-24 chromosomal losses and alterations of protein kinase a subunit expression and activity in adrenocortico- tropin-independent macronodular adrenal hyperplasia. J Clin En- docrinol Metab 91:3626-3632
18. Hsiao HP, Kirschner LS, Bourdeau I, Keil MF, Boikos SA, Verma S, Robinson-White AJ, Nesterova M, Lacroix A, Stratakis CA 2009 Clinical and genetic heterogeneity, overlap with other tumor syn- dromes, and atypical glucocorticoid hormone secretion in adreno- corticotropin-independent macronodular adrenal hyperplasia com- pared with other adrenocortical tumors. J Clin Endocrinol Metab 94:2930-2937
19. Kirschner LS, Carney JA, Pack SD, Taymans SE, Giatzakis C, Cho YS, Cho-Chung YS, Stratakis CA 2000 Mutations of the gene en- coding the protein kinase A type I-a regulatory subunit in patients with the Carney complex. Nat Genet 26:89-92
20. Horvath A, Mathyakina L, Vong Q, Baxendale V, Pang AL, Chan WY, Stratakis CA 2006 Serial analysis of gene expression in adre- nocortical hyperplasia caused by a germline PRKAR1A mutation. J Clin Endocrinol Metab 91:584-596
21. Bourdeau I, Antonini SR, Lacroix A, Kirschner LS, Matyakhina L, Lorang D, Libutti SK, Stratakis CA 2004 Gene array analysis of macronodular adrenal hyperplasia confirms clinical heterogeneity and identifies several candidate genes as molecular mediators. On- cogene 23:1575-1585
22. Stratakis CA 2003 Genetics of adrenocortical tumors: gatekeepers, landscapers and conductors in symphony. Trends Endocrinol Metab 14:404-410
23. Papanicolaou DA, Yanovski JA, Cutler Jr GB, Chrousos GP, Nie- man LK 1998 A single midnight serum cortisol measurement dis- tinguishes Cushing’s syndrome from pseudo-Cushing states. J Clin Endocrinol Metab 83:1163-1167
24. Gomez MT, Malozowski S, Winterer J, Vamvakopoulos NC, Chrousos GP 1991 Urinary free cortisol values in normal children and adolescents. J Pediatr 118:256-258
25. Cheadle C, Vawter MP, Freed WJ, Becker KG 2003 Analysis of microarray data using Z score transformation. J Mol Diagn 5:73-81
26. Almeida MQ, Muchow M, Boikos S, Bauer AJ, Griffin KJ, Tsang KM, Cheadle C, Watkins T, Wen F, Starost MF, Bossis I, Nesterova M, Stratakis CA 2010 Mouse Prkar1a haploinsufficiency leads to an increase in tumors in the Trp53+/- or Rb1+/- backgrounds and
chemically induced skin papillomas by dysregulation of the cell cycle and Wnt signaling. Hum Mol Genet 19:1387-1398
27. Huang da W, Sherman BT, Tan Q, Kir J, Liu D, Bryant D, Guo Y, Stephens R, Baseler MW, Lane HC, Lempicki RA 2007 DAVID bioinformatics resources: expanded annotation database and novel algorithms to better extract biology from large gene lists. Nucleic Acids Res 35(Web Server issue): W169-W75
28. Huang da W, Sherman BT, Lempicki RA 2009 Systematic and in- tegrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4:44-57
29. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Me- sirov JP 2005 Gene set enrichment analysis: a knowledge-based ap- proach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 102:15545-15550
30. Livak KJ, Schmittgen TD 2001 Analysis of relative gene expression data using real-time quantitative PCR and the 2-44CT method. Methods 25:402-408
31. Rohlff C, Clair T, Cho-Chung YS 1993 8-Cl-cAMP induces trun- cation and down-regulation of the RI a-subunit and up-regulation of the RII ß-subunit of cAMP-dependent protein kinase leading to type II holoenzyme-dependent growth inhibition and differentiation of HL-60 leukemia cells. J Biol Chem 268:5774-5782
32. Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver BS, Arora VK, Kaushik P, Cerami E, Reva B, Antipin Y, Mitsiades N, Landers T, Dolgalev I, Major JE, Wilson M, Socci ND, Lash AE, Heguy A, Eastham JA, Scher HI, Reuter VE, Scardino PT, Sander C, Sawyers CL, Gerald WL 2010 Integrative genomic profiling of hu- man prostate cancer. Cancer Cell 18:11-22
33. Chitale D, Gong Y, Taylor BS, Broderick S, Brennan C, Somwar R, Golas B, Wang L, Motoi N, Szoke J, Reinersman JM, Major J, Sander C, Seshan VE, Zakowski MF, Rusch V, Pao W, Gerald W, Ladanyi M 2009 An integrated genomic analysis of lung cancer reveals loss of DUSP4 in EGFR-mutant tumors. Oncogene 28:2773-2783
34. Ding L, Getz G, Wheeler DA, Mardis ER, Mclellan MD, Cibulskis K, Sougnez C, Greulich H, Muzny DM, Morgan MB, Fulton L, Fulton RS, Zhang Q, Wendl MC, Lawrence MS, Larson DE, Chen K, Dooling DJ, Sabo A, Hawes AC, Shen H, Jhangiani SN, Lewis LR, Hall O, Zhu Y, Mathew T, Ren Y, Yao J, Scherer SE, Clerc K, Metcalf GA, Ng B, Milosavljevic A, Gonzalez-Garay ML, Osborne JR, Meyer R, Shi X, Tang Y, Koboldt DC, Lin L, Abbott R, Miner TL, Pohl C, Fewell G, Haipek C, Schmidt H, Dunford-Shore BH, Kraja A, Crosby SD, Sawyer CS, Vickery T, Sander S, Robinson J, Winckler W, Baldwin J, Chirieac LR, Dutt A, Fennell T, Hanna M, Johnson BE, Onofrio RC, Thomas RK, Tonon G, Weir BA, Zhao X, Ziaugra L, Zody MC, Giordano T, Orringer MB, Roth JA, Spitz MR, Wistuba II, Ozenberger B, Good PJ, Chang AC, Beer DG, Watson MA, Ladanyi M, Broderick S, Yoshizawa A, Travis WD, Pao W, Province MA, Weinstock GM, Varmus HE, Gabriel SB, Lander ES, Gibbs RA, Meyerson M, Wilson RK 2008 Somatic mu- tations affect key pathways in lung adenocarcinoma. Nature 455: 1069-1075
35. Roskoski Jr R 2007 Vascular endothelial growth factor (VEGF) sig- naling in tumor progression. Crit Rev Oncol Hematol 62:179-213
36. Gesierich S, Berezovskiy I, Ryschich E, Zöller M 2006 Systemic induction of the angiogenesis switch by the tetraspanin D6.1A/CO- 029. Cancer Res 66:7083-7094
37. Nazarenko I, Rana S, Baumann A, McAlear J, Hellwig A, Tren- delenburg M, Lochnit G, Preissner KT, Zöller M 2010 Cell surface tetraspanin Tspan8 contributes to molecular pathways of exosome- induced endothelial cell activation. Cancer Res 70:1668-1678
38. Beuschlein F, Reincke M, Karl M, Travis WD, Jaursch-Hancke C, Abdelhamid S, Chrousos GP, Allolio B 1994 Clonal composition of human adrenocortical neoplasms. Cancer Res 54:4927-4932
39. Cohen MM, Charrow J, Balkin NE, Harris CJ 1983 Partial trisomy 14 (q23 leads to qter) via segregation of a 14/X translocation. Am J Hum Genet 35:635-644
40. Ou Z, Martin DM, Bedoyan JK, Cooper ML, Chinault AC,
Stankiewicz P, Cheung SW 2008 Branchiootorenal syndrome and oculoauriculovertebral spectrum features associated with duplication of SIX1, SIX6, and OTX2 resulting from a complex chromosomal rearrangement. Am J Med Genet A 146A:2480- 2489
41. Tominaga E, Tsuda H, Arao T, Nishimura S, Takano M, Kataoka F, Nomura H, Hirasawa A, Aoki D, Nishio K 2010 Amplification of GNAS may be an independent, qualitative, and reproducible bio- marker to predict progression-free survival in epithelial ovarian can- cer. Gynecol Oncol 118:160-166
42. Hines WC, Bazarov AV, Mukhopadhyay R, Yaswen P 2010 BORIS (CTCFL) is not expressed in most human breast cell lines and high grade breast carcinomas. PLoS One 5:e9738
43. Plevova P, Cerna D, Balcar A, Foretova L, Zapletalova J, Silhanova E, Curik R, Dvorackova J 2010 CCND1 and ZNF217 gene ampli- fication is equally frequent in BRCA1 and BRCA2 associated and non-BRCA breast cancer. Neoplasma 57:325-332
44. Paquet-Fifield S, Schlüter H, Li A, Aitken T, Gangatirkar P, Blashki
D, Koelmeyer R, Pouliot N, Palatsides M, Ellis S, Brouard N, Zan- nettino A, Saunders N, Thompson N, Li J, Kaur P 2009 A role for pericytes as microenvironmental regulators of human skin tissue regeneration. J Clin Invest 119:2795-2806
45. Gaujoux S, Tissier F, Groussin L, Libé R, Ragazzon B, Launay P, Audebourg A, Dousset B, Bertagna X, Bertherat J 2008 Wnt/ß- catenin and 3’,5’-cyclic adenosine 5’-monophosphate/protein ki- nase A signaling pathways alterations and somatic ß-catenin gene mutations in the progression of adrenocortical tumors. J Clin En- docrinol Metab 93:4135-4140
46. Tadjine M, Lampron A, Ouadi L, Horvath A, Stratakis CA, Bour- deau I 2008 Detection of somatic ß-catenin mutations in primary pigmented nodular adrenocortical disease (PPNAD). Clin Endocri- nol (Oxf) 69:367-373
47. Iliopoulos D, Bimpaki EI, Nesterova M, Stratakis CA 2009 Mi- croRNA signature of primary pigmented nodular adrenocortical disease: clinical correlations and regulation of Wnt signaling. Can- cer Res 69:3278-3282
THE ENDOCRINE SOCIETY®
Sign up for eTOC alerts today to get the latest articles as soon as they are online. http://mend.endojournals.org/subscriptions/etoc.shtml