EXPERIMENTAL STUDY
Analysis by cDNA microarrays of gene expression patterns of human adrenocortical tumors
E P Slater, S M Diehl, P Langer, B Samans1, A Ramaswamy2, A Zielke and D K Bartsch
Department of Surgery, 1Institute of Medical Biometry and Epidemiology and 2 Department of Pathology, Philipps-University Marburg, Baldingerstrasse, 35033 Marburg, Germany
(Correspondence should be addressed to E P Slater; Email: slater@med.uni-marburg.de)
Abstract
Objectives: Adrenocortical carcinoma (ACC) is a rare malignant neoplasm with extremely poor prog- nosis. The molecular mechanisms of adrenocortical tumorigenesis are still not well understood. The comparative analysis by cDNA microarrays of gene-expression patterns of benign and malignant adrenocortical tumors allows us to identify new tumor-suppressor genes and proto-oncogenes under- lying adrenocortical tumorigenesis.
Design and methods: Total RNA from fresh-frozen tissue of 10 ACC and 10 benign adrenocortical ade- nomas was isolated after histologic confirmation of neoplastic cellularity of at least 85%. The refer- ence consisted of pooled RNA of 10 normal adrenal cortex samples. Amplified RNA of tumor and reference was used to synthesize Cy3- and Cy5-fluorescently labeled cDNA in a flip-color technique. D-chips containing 11 540 DNA spots were hybridized and scanned and the images were analyzed by ImaGene 3.0 software.
Results: The comparative analysis of gene expression revealed many genes with more than fourfold expression difference between ACC and normal tissue (42 genes), cortical adenoma and normal tissue (11 genes), and ACC and cortical adenoma (21 genes) respectively. As confirmed by real- time PCR, the IGF2 gene was significantly upregulated in ACCs versus cortical adenomas and normal cortical tissue. Genes that were downregulated in adrenocortical tumors included chromogra- nin B and early growth response factor 1.
Conclusions: Comprehensive expression profiling of adrenocortical tumors by the cDNA microarray technique is a very powerful tool to elucidate the molecular steps associated with the tumorigenesis of these ill-defined neoplasms. To evaluate the role of identified genes, further detailed analyses, including correlation with clinical data, are required.
European Journal of Endocrinology 154 587-598
Introduction
Adrenal masses are a common disorder, affecting 3-7% of the population. Most turn out to be benign adrenocortical adenomas, which may be functional or nonfunctional. Much more rarely, these masses represent primary adrenal carcinoma (1). Adrenocorti- cal carcinoma (ACC) is a highly malignant tumor with an incidence of ~ 1 per 1.7 million inhabitants per year in the West. Although ACC is rare, its highly aggressive behavior and 5-year survival rate of only 10-20% urgently require the decoding of its molecular basis to develop new strategies for diagnosis and treatment (2-6). The genetic background of adrenocortical tumorigenesis is poorly characterized. In other endo- crine tissues, such as thyroid, there is conclusive evi- dence that hyperplasia and adenomas can precede
cancer. In the adrenal, there are clinical cases of either hyperplasia or adenoma associated with later development of cancer. However, only a few studies have attempted to characterize this process on a mol- ecular basis (7). Although it is unclear whether there is an adenoma-carcinoma sequence, common patterns seen in adenomas and carcinomas, and the accumu- lation of chromosomal imbalances with tumor pro- gression support the existence of an adenoma- carcinoma sequence (8). X-chromosome inactivation analysis has shown that ACCs are of monoclonal origin, whereas benign adenomas may be monoclonal or polyclonal (9-11). The evidence gathered so far shows that the transition from adrenal adenoma to car- cinoma involves a monoclonal proliferation of cells that, among other yet to be characterized defects, have undergone chromosomal duplication at the
11p15.5 locus, leading to overexpression of the insulin- like growth factor (IGF)2 gene and abrogation of expression of the CDKN1C and H19 genes (12, 13). TP53 has been shown to be involved in progression to carcinoma in a subset of patients, and it has been suggested that the frequency of adrenocorticotropic hormone (ACTH) receptor deletion might also be involved (1). Other key oncogenes and tumor suppres- sor genes remain to be identified. However, a recent study has reported that chromosomal loci 1p, 2p16, 11q13 and 17p may harbor potential tumor-suppres- sor genes, and chromosomes 4, 5 and 12 potential oncogenes associated with adrenal tumorigenesis (1). Therefore, detailed analysis of the genes involved is highly desirable. The development of the cDNA micro- array technique offers the opportunity to analyze a large number of genes, and allows comparative analysis of gene-expression profiling in benign and malignant adrenocortical tumors and the identification of tumor- suppressor genes and proto-oncogenes associated with the initiation and progression of adrenocortical tumors. Findings from these analyses might clarify adrenocortical tumorigenesis and lead to the establish- ment of new diagnostic and prognostic markers as well as the characterization of novel strategies for treatment. Therefore, we have performed a comprehensive and representative analysis of neoplastic and nonneoplastic adrenocortical tissue samples. We were particularly interested in evaluating the differential profile of 11 500 genes with established importance for develop- ment and progression of malignant diseases (1), to assess whether combinations of genes can predict malignant tumors (2) and to validate the results by real time RT-PCR of selected candidate genes (3).
Materials and methods
Normal and tumor samples
The adrenocortical tissues analyzed in this study were obtained from the collection of fresh frozen adrenal tissue of the Department of Surgery, Philipps-University Marburg, Germany, collected between 1996 and 2003. For the purpose of this study, 10 ACCs and 10 adrenocor- tical adenomas (four from Conn’s syndrome, four from Cushing’s disease and two nonfunctional adenomas), as well as 10 nonneoplastic adrenal cortical tissue samples, were evaluated. The classification was deter- mined by both conventional histologic methods (14) and Weiss score (15) where adenomas met fewer than four and carcinomas at least four of the criteria. The ethics committee of the university approved this study, and all patients participating in the study consented to sampling. Frozen tumor samples were formalin-fixed and embedded in paraffin, and sections were evaluated by hematoxylin and eosin (HE) staining regarding diag- nosis and neoplastic cellularity. Only tumor samples with neoplasticity of greater than 85% were included
in the analysis. Control samples consisted of 10 histologically confirmed normal adrenal cortices.
RNA isolation was performed as follows. Frozen con- trol and tumor tissue samples were dissected by the pathologist (A M) and homogenized in the presence of TRIzol Reagent (Invitrogen, Karlsruhe, Germany) by the manufacturer’s protocol. Total RNA was then further purified by digestion with DNase I and recovery of RNA with the RNeasy kit (Qiagen, Hilden, Germany) by the supplier’s protocol. To determine the integrity of RNA, standard RT-PCR for the amplification of 17x- hydroxylase and ß-actin was performed on 25 ng RNA with Qiagen’s OneStep RT-PCR kit by the manu- facturer’s protocol. The primer sequences were as fol- lows:
· Cyp17_forward TCTCTTGCTGCTTACCCTAG; Cyp17_reverse TCAAGGAGATGACATTGGTT (GenBank accession no. NM_000102).
· ß-actin_forward GATGATGATATCGCCGCGCTCG- TCGTC; ß-actin_reverse GTGCCTCAGGGCAGCG- GAACCGCTCA (GenBank accession no. M10277).
The reaction mixture was incubated at 50 ℃ for 30 min and 95 ℃ for 15 min, followed by 30 cycles of standard PCR (1-min denaturation at 95℃, 1-min annealing at 58 °℃ and 1-min extension at 72°℃). PCR products were visualized by ethidium bromide staining of PAGE. Only samples showing expression of Cyp17 as well as ß-actin were included in analysis. RNA (2 µg) from 10 ACC and 10 adenomas (eight functional and two non- functional adenomas) was amplified with the Messa- geAmp aRNA Kit (Ambion, Huntingdon, UK). For the reference, the 10 control samples were amplified and then pooled. Amino allyl-cDNA was synthesized with 2 µg aRNA and then labeled and purified with the CyScribe Post-Labeling Kit (Amersham Biosciences, Frei- burg, Germany). Samples were fluorescently labeled with Cy3 and Cy5 by the flip-color technique.
For gene-expression profiling, the reference and tumor samples were mixed, denatured and then hybridized to microarrays for 16 h at 56 ℃ and washed at a stringency of 0.1% SSC and 0.1% SDS. The microarray contains 11 540 DNA spots; detailed protocols and data descrip- tion of the chip are available from the website: www.im- t.uni-marburg.de. Each experiment was performed as a sandwich hybridization with two arrays. Spot intensities were extracted from a scanned image with ImaGene 3.0 Software (BioDiscovery, Los Angeles, CA, USA). For each spot, median signal and background intensities for both channels were obtained. To account for spot differences, the background-corrected ratios of the two channels were calculated and log2-transformed. To balance the fluorescence intensities for the two dyes as well as to allow comparison of expression levels across exper- iments, the raw data were standardized. We used a spatial and intensity-dependent standardization (like Yang et al. (16)) to correct for inherent bias on
each chip (the lowest scatter-plot). As each gene was measured twice in the sandwich hybridization, mean log-ratios M were calculated from replicates. If gene repli- cates differed more than the maximum of threefold and 75% of the calculated average log-ratio, or the back- ground intensity was higher than signal intensity, the spot was excluded on that array. Differentially expressed genes were selected by a fold-change difference of at least 2 and an absolute value of the t-statistic of 1.96. Prior to the cluster analysis, the expression profile of each gene was centered by subtracting the mean observed value. Average linkage hierarchic clustering was then performed for genes as well as for chips with the Euclidean distance metric as implemented in the pro- gram Genesis (17).
The microarray results were validated for three candi- date genes known to be expressed in adrenocortical tissue, including chromogranin B (CgB), early growth response gene 1 (Egr-1), and IGF2 by real-time RT- PCR analysis. For validation, 10 µg total RNA were reverse transcribed with Superscript II reverse transcrip- tase (Invitrogen) and an oligo dT15 primer, according to the manufacturer’s instructions and previously pub- lished methods (18). Real-time PCR with the LightCycler System (Roche, Mannheim, Germany) was performed in a reaction mixture of 20 ul using the QuantiTect SYBR Green PCR Kit according to the manufacturer (Qiagen). Primers designed for analysis were as follows:
· Egr1 (GenBank accession no. NM001964) for- ward CGAGCAGCCCTACGAGCACCTGAC and reverse TGCGCAGCTCAGGGGTGGGCTCTG.
· IGF2 (GenBank accession no. BC000531) for- ward CCGTGCTTCCGGACAACT and reverse GGACTGCTTCCAGGTGTCATATT.
· CgB (GenBank accession no. AL035461) forward TGCCAGTGGATAACAGGAAC and reverse TCTT- CAGGACTTGGCGGCA.
· GAPDH forward CGTCTTCACCACCATGGAGA and reverse CGGCCATCACGCCACAGTTT.
The cycle threshold values for each gene were analyzed relative to those for GAPDH.
Results
For this study, 10 adrenocortical cancers, as established by histology, and 10 adenomas, of which eight were functional, including four aldosterone-producing ade- nomas, four cortisol-producing adenomas, and two nonfunctional adenomas (incidentalomas), were selected from the tissue bank of the Department of Sur- gery, Philipps-University Marburg. The clinicopatholo- gic data of the adrenocortical tumors analyzed are summarized in Table 1. Normal adrenal cortex, adreno- cortical adenoma and ACC were analyzed histologically
| Designation | Tissue type | Age | Gender | Stage* | Syndrome | Metastatic sites ** | Follow-up |
|---|---|---|---|---|---|---|---|
| A1 | ACC | 50 | F | II | None | None | DOD |
| A2 | ACC | 46 | F | I | None | None | NED |
| A3 | ACC | 59 | F | III | None | None | NED |
| A4 | ACC | 43 | F | IV | Cushing | Lung | DOD |
| A5 | ACC | 15 | M | II | Cushing | None | DOD |
| A6 | ACC | 74 | F | III | None | None | DURC |
| A7 | ACC | 52 | M | III | None | None | DOD |
| A8 | ACC | 51 | F | III | Cushing, AGS | None | DOD |
| A9 | ACC | 37 | F | II | Conn | None | NED |
| A10 | ACC | 37 | F | III | Cushing | None | AWD |
| A11 | ACC | 84 | F | II | None | None | NED |
| A12 | ACC | 43 | M | II | None | None | DOD |
| A15 | ACC | 17 | F | III | Cushing, AGS | None | NED |
| T1 | Adenoma | 46 | M | N.A. | Conn | NED | |
| T2 | Adenoma | 45 | F | N.A. | Conn | NED | |
| T3 | Adenoma | 47 | F | N.A. | Conn | NED | |
| T4 | Adenoma | 54 | M | N.A. | Conn | NED | |
| T5 | Adenoma | 65 | M | N.A. | Cush | NED | |
| T6 | Adenoma | 27 | F | N.A. | Cush | NED | |
| T7 | Adenoma | 47 | F | N.A. | Cush | NED | |
| T8 | Adenoma | 45 | F | N.A. | Cush | NED | |
| T9 | Adenoma | 65 | F | N.A. | Inc | NED | |
| T10 | Adenoma | 52 | F | N.A. | Inc | NED | |
| T11 | Adenoma | 64 | M | N.A. | Inc | NED | |
| T12 | Adenoma | 73 | F | N.A. | Inc | NED |
*Tumors graded according to UICC.
** At time of operation.
Cush: Cushing’s disease; Inc: Incidentaloma; N.A .: not applicable; Conn: Conn’s syndrome; AGS: adrenogenital syndrome; DOD: death of disease; NED: no evidence of disease; AWD: alive with disease; DURC: death of unrelated cause.
to confirm tissue origin, lack of necrosis and, for tumors, neoplasticity of greater than 85%, before pro- ceeding with nucleic acid purification. The normal tissue displayed regularly shaped nuclei and fat con- tent. Fig. 1 shows representative examples of HE stain- ing of tissue samples used for the RNA purification. The adrenal origin of the samples was additionally ensured by the RNA expression of 17a-hydroxylase (Cyp17), in addition to ß-actin, by RT-PCR. All samples chosen for further analysis were positive for both Cyp17 and
A
B
C
ß-actin (data not shown). Microarray analysis demon- strated that ACCs were more dissimilar to normal adrenal than adenomas. The adenomas were more closely related to each other and to normal adrenal; this is not an entirely unexpected result given the histo- logic similarity of the tissues. In particular, 40 differen- tially expressed genes were found in adenomas in comparison to normal adrenal (Table 2), and 144 dif- ferentially expressed genes were detected in ACCs (Table 3). As shown in Fig. 2 and Table 4, more than 60 genes were found with at least a threefold change in mRNA levels. Both up- and downregulated genes were identified (Table 4). These genes, with at least threefold differences in mRNA expression, were then subjected to cluster analysis. Transcriptional profiles, which distinguish between benign and malignant adrenocortical tumors, identified several differentially expressed transcripts as demonstrated by cluster analysis (Fig. 2). The sample dendrogram revealed the similarities among the adenomas (T) and the carci- nomas (A) respectively, and clearly distinguished the two groups. Two major clusters of genes were differen- tially expressed in the carcinomas compared with the adenomas: genes that were expressed at a higher level in the carcinomas and those that were expressed at a lower level in the carcinomas. The former include the gene for IGF-2 and potential oncogenes. The latter rep- resent potential tumor-suppressor genes.
Some of the differentially expressed genes were chosen for further analysis to ensure the validity of the microarray results. As expected from the literature (1, 12, 19), the IGF-2 gene expression was 3-7-fold greater in ACCs than in adenomas. Real-time RT-PCR confirmed this result where the cycle threshold crossing point (CT) for the ACC samples was 3-5 values less, representing a maximal increase in expression of 8-32-fold in comparison to the control or adenoma samples (Fig. 3).
As one of our main interests is the identification of potential tumor-suppressor genes whose investigation may contribute to the understanding of the pathome- chanism of ACC, we decided to concentrate in our initial analysis on clearly downregulated genes. The tissue-specifically expressed gene, chromogranin B (CgB), was found to be downregulated in both adenomas (28-fold) (Table 2) and ACCs (13-fold) (Table 3). The loss of a tissue-specifically expressed gene could reflect dedifferentiation of tumor tissue. This decrease in expression was confirmed by real- time RT-PCR, which demonstrated an increase in the CT value from 4 to 5, representing at most a 16-32- fold decrease in expression, for the adenomas analyzed, and from 3 to 4, or at most a 8-16-fold decrease for the ACCs.
The Egr-1 gene was downregulated in ACC in com- parison to normal adrenal by eightfold (Fig. 3). This finding was also confirmed by real-time RT-PCR, where the CT values for the ACCs were increased by
| Gene name | Accession no. | Fold regulation | Up/down |
|---|---|---|---|
| Protein tyrosine phosphatase, receptor type, f polypeptide (PTPRF), interacting protein (liprin), alpha 2 | H08850 | 4.42 | up |
| Chemokine (C-C motif) ligand 2 | AA425102 | 3.94 | up |
| Activity-regulated cytoskeleton-associated protein | H86117 | 3.84 | up |
| Hypothetical protein FLJ10052 | R54822 | 3.76 | up |
| Purkinje cell protein 4 | AA452826 | 3.36 | up |
| Mucolipin 3 | AA171718 | 3.29 | up |
| Neuronal pentraxin II | AA683041 | 2.40 | up |
| Cytoplasmic FMR1 interacting protein 2 | H12044 | 2.38 | up |
| B-cell CLL/lymphoma 11B (zinc finger protein) | H09748 | 2.27 | up |
| Chromogranin B (secretogranin 1) | W37769 | 28.46 | down |
| Myxovirus (influenza virus) resistance 1, interferon-inducible protein p78 (mouse) | AA457042 | 12.08 | down |
| EST | H11453 | 9.86 | down |
| yz80b09.s1 Soares_multiple_sclerosis_2NbHMSP Homo sapiens cDNA clone IMAGE:289337 3' | N92646 | 7.60 | down |
| Chromosome 1 open reading frame 29 | AA410188 | 7.54 | down |
| Insulin-like growth factor binding protein 6 | AA479428 | 7.35 | down |
| Brain expressed, X-linked 1 | W60582 | 4.70 | down |
| Contactin 1 | H19315 | 4.46 | down |
| Dickkopf homolog 3 (Xenopus laevis) | AA425947 | 4.33 | down |
| Sjogren syndrome antigen B (autoantigen La) | H29485 | 4.08 | down |
| Chemokine (C-C motif) ligand 2 | AA938563 | 3.99 | down |
| Cathepsin H | AA487231 | 3.93 | down |
| Sodium channel, voltage-gated, type III, beta | R53930 | 3.82 | down |
| Insulin-like growth factor binding protein 6 | AA478724 | 3.71 | down |
| Heat shock 105 kDa/110 kDa protein 1 | AA485151 | 3.71 | down |
| Homo sapiens cDNA FLJ38885 fis, clone MESAN2017417 | T89094 | 3.63 | down |
| Sodium channel, voltage-gated, type III, beta | AA134824 | 3.55 | down |
| Solute carrier family 40 (iron-regulated transporter), member 1 | T52564 | T57235 | 3.43 | down |
| Ubiquitin carboxyl-terminal esterase L1 (ubiquitin thiolesterase) | AA670438 | 3.35 | down |
| KIAA1576 protein | AA609348 | 3.34 | down |
| Protein tyrosine phosphatase, receptor type, N polypeptide 2 | AA464590 | 3.30 | down |
| Sal-like 2 (Drosophila) | H23254 | 3.29 | down |
| Connective tissue growth factor | AA598794 | 3.16 | down |
| Complexin 2 | H09966 | 3.10 | down |
| Integral membrane protein 2C | AA034213 | 3.08 | down |
| Hypothetical protein FLJ39155 | R08141 | 3.07 | down |
| Calmodulin 1 (phosphorylase kinase, delta) | R76554 | R76277 | 3.02 | down |
| Homo sapiens cDNA FLJ39226 fis, clone OCBBF2007232. | H09078 | 2.99 | down |
| Aldo-keto reductase family 1, member C1 (dihydrodiol dehydrogenase 1; 20-alpha (3-alpha)-hydroxysteroid dehydrogenase) | R93124 | 2.95 | down |
| gr04f12.s1 soares fetal liver spleen 1NFLS Homo sapiens cDNA clone IMAGE: 204335 | H59916 | 2.92 | down |
| Nucleolar protein 4 | AA430033 | 2.87 | down |
4, representing a maximal decrease in expression of up to 16-fold, Thus, for each of these candidate genes, the differential gene expression was confirmed by real-time RT-PCR, and the findings were comparable to those of the microarray analysis (Fig. 3).
Discussion
DNA microarray technology allows comprehensive examination of the transcriptional profile of tumors. It is rapidly being applied to various problems in pathology and oncology, such as tumor classification, and it is a useful gene discovery tool to complement other similar technologies. In this study, we used DNA microarrays
to generate transcriptional profiles of benign and malig- nant adrenocortical tumors with various hormonal secretion profiles (Table 1). We demonstrated for the first time that these profiles can distinguish normal and benign tissues from malignant tumors and identify differentially expressed genes that may help to explain the pathogenesis of the disease and have diagnostic and therapeutic implications. Surprisingly, only 11 genes were found to be differentially expressed more than fourfold in comparing adenomas to normal tissue (Table 2). A comparison of the ACCs to normal tissue resulted in a list of 42 genes that were differentially expressed more than fourfold (Table 3), and the com- parison ACCs to adenomas resulted in 21 genes that were more than fourfold differentially expressed.
| Gene name | Accession no. | Fold regulation | Up/down |
|---|---|---|---|
| Dopachrome tautomerase (dopachrome delta-isomerase. tyrosine-related protein 2) | AA478553 | 6.14 | up |
| Human neuropeptide Y receptor Y1 (NPYY1) | R43817 | 6.11 | up |
| Sodium channel, voltage-gated, type III, beta | R53930 | 5.78 | up |
| Fibronectin 1 | R62612 | 4.43 | up |
| Chemokine (C-C motif) ligand 2 | AA938563 | 4.30 | up |
| Sjogren syndrome antigen B (autoantigen La) | H29485 | 4.29 | up |
| Fibronectin 1 | R62612 | 4.28 | up |
| Insulin-like growth factor 2 (somatomedin A) | N54596 | 4.16 | up |
| GA binding protein transcription factor. alpha subunit 60 kDa | H96241 | 4.12 | up |
| Hypothetical protein MGC5306 | H47257 | 4.07 | up |
| EST | N59772 | 3.93 | up |
| Hydroxy-delta-5-steroid dehydrogenase. 3 betaand steroid delta-isomerase 1 | R68803 | 3.84 | up |
| Dickkopf homolog 3 (Xenopus laevis) | AA425947 | 3.77 | up |
| Aldehyde dehydrogenase 1 family. member A3 | AA455235 | 3.69 | up |
| Clone IMAGE:120162 mRNA sequence | T95274 | 3.67 | up |
| CDC28 protein kinase regulatory subunit 2. | AA397813 | 3.51 | up |
| Chemokine (C-C motif) ligand 2 | AA425102 | 3.27 | up |
| EST | H16389 | 3.24 | up |
| Insulin-like growth factor binding protein 3 | AA598601 | 3.22 | up |
| Immediate early response 3 | AA457705 | 3.21 | up |
| Dentatorubral-pallidoluysian atrophy (atrophin-1) | H08643 | 3.20 | up |
| Ectodermal-neural cortex (with BTB-like domain) | H72122 | 3.18 | up |
| Clone DKFZp434E235 | H22563 | 3.18 | up |
| yq51h12.s1 Soares fetal liver spleen 1NFLS cDNA clone IMAGE:199367 3' .. | R95691 | 3.11 | up |
| Dopa decarboxylase (aromatic L-amino acid decarboxylase) | AA702640 | 3.11 | up |
| Glutaminyl-peptide cyclotransferase (glutaminyl cyclase) | AA282134 | 3.03 | up |
| Fibulin 2 | AA452840 | 3.02 | up |
| c-src tyrosine kinase | AA079775 | 2.98 | up |
| Flotillin 1 | AA488175 | 2.92 | up |
| Fibroblast growth factor 14 | AA400047 | 2.89 | up |
| RWD domain containing 1 | AA487499 | 2.89 | up |
| Hypothetical protein FLJ37078 | R44292 | 2.84 | up |
| Sodium channel. nonvoltage-gated 1 alpha | AA458982 | 2.84 | up |
| Troponin I, skeletal, fast | AA181334 | 2.82 | up |
| Membrane-associated tyrosine- and threonine-specific cdc2-inhibitory kinase | AA478066 | 2.73 | up |
| Insulin-like growth factor 2 (somatomedin A) | N54596 | 2.63 | up |
| Parathymosin | R11526 | 2.53 | up |
| Cadherin 13. H-cadherin (heart) | R41787 | 2.50 | up |
| Adrenergic. alpha-2A -. receptor | T48692 | 2.49 | up |
| Glucocorticoid receptor DNA binding factor 1 | N72276 | 2.42 | up |
| CDC28 protein kinase regulatory subunit 2 | AA010065 | 2.41 | up |
| Signal recognition particle receptor, B subunit | R95693 | 2.37 | up |
| Cytochrome c oxidase subunit VIa polypeptide 1 | AA482243 | 2.35 | up |
| Ubiquitin-like 1 (sentrin) | AA488626 | 2.33 | up |
| Collagen, type V, alpha 1 | R75635 | 2.30 | up |
| Fibroblast growth factor receptor 1 (fms-related tyrosine kinase 2. Pfeiffer syndrome) | R54846 | 2.27 | up |
| Splicing factor. arginine/serine-rich 2. interacting protein | R91171 | 2.25 | up |
| EST | T97592 | 2.23 | up |
| Lymphocyte-specific protein 1 | T90632 | 2.18 | up |
| Likely ortholog of rat vacuole membrane protein 1 | AA485373 | 2.14 | up |
| Dishevelled. dsh homolog 2 (Drosophila) | R38325 | 2.13 | up |
| Kruppel-like zinc finger protein GLIS2 | R43826 | 2.12 | up |
| Dehydrogenase/reductase (SDR family) member 7 | H87144 | 2.12 | up |
| Insulin-like growth factor 2 (somatomedin A) | N74623 | 2.11 | up |
| Hypothetical protein BC013949 | W69741 | 2.11 | up |
| Dipeptidylpeptidase 4 (CD26. adenosine deaminase complexing protein 2) | W70234 | 2.11 | up |
| Nucleoredoxin | T64216 | 2.10 | up |
| Zinc finger protein 161 | AA232647 | 2.10 | up |
| Hypothetical protein from clone 643 | T53404 | 2.09 | up |
| BTG family. member 3 | N52496 | 2.07 | up |
| Enoyl Coenzyme A hydratase domain containing 1 | AA173573 | 2.06 | up |
| Glyceraldehyde-3-phosphate dehydrogenase | h52950 | 2.03 | up |
| EST | H48467 | 2.03 | up |
| Prefoldin 5 | AA446453 | 2.03 | up |
| Gene name | Accession no. | Fold regulation | Up/down |
|---|---|---|---|
| Integrin, alpha 7 | T60926 | 2.02 | up |
| Thymopoietin | T63980 | 2.02 | up |
| Myxovirus (influenza virus) resistance 1, interferon-inducible protein p78 (mouse) | AA457042 | 13.60 | down |
| Chromogranin B (secretogranin 1) | W37769 | 12.72 | down |
| Actin binding LIM protein 1 | AA406601 | 11.43 | down |
| EST | H11453 | 9.23 | down |
| Ferredoxin 1 | AA187349 | 8.96 | down |
| FBJ murine osteosarcoma viral oncogene homolog B | T61948 | 8.32 | down |
| yz80b09.s1 Soares_multiple_sclerosis_2NbHMSP Homo sapiens cDNA clone | N92646 | 8.17 | down |
| IMAGE:289337 3'. | |||
| Early growth response 1 | AA486533 | 7.54 | down |
| EST | N68993 | 7.54 | down |
| Fatty acid binding protein 5 (psoriasis-associated) | T60075 | 7.49 | down |
| Aldehyde dehydrogenase 1 family. member A1 | AA664101 | 7.35 | down |
| zb50h07.s1 Soares_fetal_lung_NbHL19W Homo sapiens cDNA clone | N93686 | 7.08 | down |
| IMAGE:307069 3' | |||
| Chromosome 1 open reading frame 29 | AA410188 | 6.96 | down |
| Contactin 1 | H19315 | 6.79 | down |
| Microsomal glutathione S-transferase 1 | AA495936 | 6.58 | down |
| Glucose phosphate isomerase | AA401111 | 6.54 | down |
| Regucalcin (senescence marker protein-30) | H05140 | 6.28 | down |
| Interferon-induced protein with tetratricopeptide repeats 1 | AA489640 | 5.04 | down |
| Chemokine (C-C motif) ligand 15 | R96626 | 5.03 | down |
| Solute carrier family 16 (monocarboxylic acid transporters) member 9 | W16424 | 4.87 | down |
| DKFZP586B1621 protein | H20543 | 4.85 | down |
| High mobility group AT-hook 1 | AI042404 | 4.80 | down |
| Serine (or cysteine) proteinase inhibitor. clade G (C1 inhibitor). member 1. | AA481438 | 4.64 | down |
| (angioedema. hereditary) | |||
| Mitogen-activated protein kinase kinase kinase 5 | AA150828 | 4.52 | down |
| Protein tyrosine phosphatase, receptor type, f polypeptide (PTPRF), interacting protein (liprin), alpha 2 | H08850 | 4.48 | down |
| Vascular cell adhesion molecule 1 | H07072 | 4.47 | down |
| Ubiquitin carboxyl-terminal esterase L1 (ubiquitin thiolesterase) | AA670438 | 4.31 | down |
| v-jun sarcoma virus 17 oncogene homolog (avian) | W96155 | 4.30 | down |
| Activity-regulated cytoskeleton-associated protein | H86117 | 4.26 | down |
| Heat shock 105 kDa/110 kDa protein 1 | AA485151 | 4.21 | down |
| Glutathione S-transferase theta 1 | H99813 | 4.21 | down |
| Insulin-like growth factor binding protein 6 | AA479428 | 4.08 | down |
| Atp-binding cassette. sub-family B (MDR/TAP). member 1 | AA459824 | 3.95 | down |
| Insulin-like growth factor binding protein 6 | AA478724 | 3.89 | down |
| Zinc finger protein 145 (Kruppel-like, expressed in promyelocytic leukemia) | AA101632 | 3.88 | down |
| Fibulin 1 | AA134871 | 3.85 | down |
| Chromosome 9 open reading frame 97 | AA235388 | 3.84 | down |
| KIAA1268 protein | T64956 | 3.82 | down |
| Solute carrier family 26 (sulfate transporter), member 2 | W15263 | 3.80 | down |
| Major histocompatibility complex, class II, DP beta 1 | AA486532 | 3.80 | down |
| CD163 antigen | AA401693 | 3.79 | down |
| Glutathione peroxidase 3 (plasma) | AA664180 | 3.79 | down |
| MAX protein | T89496 | 3.76 | down |
| Insulin-like growth factor 1 receptor | H13300 | 3.75 | down |
| Arachidonate 5-lipoxygenase-activating protein | T49652 | 3.74 | down |
| Interferon-induced transmembrane protein 1 (9-27) | AA058323 | 3.66 | down |
| Chromosome 10 open reading frame 10 | N55269 | 3.65 | down |
| Scavenger receptor class B, member 1 | AA443899 | 3.61 | down |
| Connective tissue growth factor | AA598794 | 3.59 | down |
| EST | T91100 | 3.57 | down |
| Discoidin domain receptor family, member 2 | AA243749 | 3.56 | down |
| Homo sapiens cDNA FLJ40165 fis. clone TESTI2015962. | H22854 | 3.50 | down |
| Solute carrier family 40 (iron-regulated transporter), member 1 | T52564 | T57235 | 3.44 | down |
| Hypothetical protein FLJ39155 | R08141 | 3.41 | down |
| Homo sapiens cDNA FLJ38885 fis. clone MESAN2017417. | T89094 | 3.40 | down |
| Plasminogen-like | T67549 | 3.38 | down |
| Periostin, osteoblast specific factor | AA598653 | 3.37 | down |
| ATP-binding cassette, subfamily C (CFTR/MRP), member 3 | AA429895 | 3.37 | down |
| Chemokine-like factor superfamily 3 | AA486561 | 3.36 | down |
| ALEX3 protein | N54456 | 3.33 | down |
| Gene name | Accession no. | Fold regulation | Up/down |
|---|---|---|---|
| Interferon, alpha-inducible protein (clone IFI-6-16) | AA448478 | 3.29 | down |
| Complement component 7 | AA598478 | 3.27 | down |
| KIAA1576 protein | AA609348 | 3.23 | down |
| Sodium channel, voltage-gated, type III, beta | AA134824 | 3.11 | down |
| Retinoic acid receptor responder (tazarotene induced) 2 | AA482067 | 3.07 | down |
| Complexin 2 | H09966 | 3.05 | down |
| Glutaredoxin (thioltransferase) | AA291163 | 3.03 | down |
| ATP-binding cassette. subfamily B (MDR/TAP), member 1 | AA135958 | 3.03 | down |
| AE binding protein 1 | AA490462 | 2.99 | down |
| Hypothetical protein FLJ20637 | AA487462 | 2.98 | down |
| Fibronectin type III domain containing 5 | N54901 | 2.92 | down |
| Glutathione S-transferase A4 | AA152347 | 2.91 | down |
| G protein-coupled receptor 116 | T63971 | 2.90 | down |
| Calmodulin 1 (phosphorylase kinase, delta) | R76554 | R76277 | 2.88 | down |
| Major histocompatibility complex; class Il; DR beta 5 | AA485739 | 2.86 | down |
| Scm-like with four mbt domains 1 | AA400512 | 2.26 | down |
| Notch homolog 3 (Drosophila) | T63511 | 2.22 | down |
| Peroxisomal lon protease | T67138 | 2.04 | down |
The gene for chromogranin B (CgB) was found to be downregulated in both adenomas (28-fold) (Table 2) and ACCs (13-fold) (Table 3). This finding was confirmed by real-time PCR (Fig. 3). Chromogranins are represen- tative proteins contained in endocrine cells of various organs, including some ductal cells of the breast. Hom- ology between the BRCA1 protein (1214-1223) and the chromogranins has been detected, suggesting that chromogranin may play the role of a tumor suppressor, like BRCA1 (20). Our results suggest that loss of CgB may be an early event in adrenocortical tumorigenesis. In patients with lymph-node-negative primary invasive ductal breast carcinoma, CgB-negative tumors demon- strated a significantly poorer prognosis than in patients with CgB-positive tumors. In univariate analysis, a sig- nificantly increased risk of disease progression and death was present in patients with CgA-poor and CgB- poor tumors respectively (21). It was concluded that the CgB immunostaining pattern of the primary tumor can distinguish patients with increased risk of death in patients with sporadic medullary thyroid carcinoma. The problem of whether CgB represents a prognostic factor in ACCs requires a large-scale study.
Another interesting potential tumor suppressor is the Egr-1 gene. As validated by real-time PCR, it was down- regulated in ACC in comparison to normal adrenal cortex tissue and adrenal adenoma by eightfold and threefold respectively (Fig. 3). Egr-1 is a transcription factor that has previously been suggested to be a master switch regulating inflammatory parameters (22). Egr-1 is also known as nerve growth factor induced-A (NGFI-A), Krox-24, ZIF268, ETR103 and TIS8. It is a phosphorylated zinc-finger-containing transcription factor often associated with cell growth stimulation (23). The gene for Egr-1 is located on the q31.1 ‘cytokine cluster’ region of chromosome 5 in man, and is rapidly and transiently induced by a
large number of stressful stimuli as well as growth fac- tors and cytokines. In fact, Egr-1 has been proposed as a master switch of gene expression underlying coordi- nated responses to various types of injury (23).
Apoptosis-regulating genes, such as TP53 and ANp73a, are known to increase the expression of Egr-1 (24, 25). Egr-1 has previously been implicated in the development and maintenance of prostate cancer (26, 27). Thus, further detailed, large-scale studies are warranted to clarify the role of Egr-1 in ACC.
One of the most interesting overexpressed genes is IGF2 on chromosome 11p15. Like other groups, we found a 3-7-fold overexpression of IGF2 in ACCs com- pared with adenomas and normal adrenal cortex (1, 13, 19). IGF-I and IGF-II are polypeptides involved in metabolism, growth and cell differentiation. They are synthesized in various tissues and have endocrine and auto/paracrine mechanisms of action depending on the tissue origin (2, 3). Both peptides are normally produced in adrenocotical cells (4, 19, 28). Through its action on steroidogenesis enzymes, IGF-I maintains the differentiated functions of the cell (19, 29). The precise role of IGF-II in mature adrenocortical gland is less clear. The IGF2 gene induction has previously been described as one of the most significant differences between ACCs and adrenal adenomas (1). Genetic altera- tions involving the 11p15 locus are very common in malignant tumors, but are found only in rare adrenal adenomas (13). The fact that we found this as well is further verification of our results. However, in adreno- cortical tumors associated with Beckwith-Wiedemann syndrome, allelic losses at the 11p15 locus have been found in both adenomas and carcinomas as well as in familial carcinomas. Current data suggest that abnorm- alities in structure and/or expression of the IGF-II gene are a late event in the multistep tumorigenesis of sporadic adrenocortical neoplasms.
3.0
1:1
-3.0
2
AA487231
AA17 17 18
R54848
RD3124
R026 12
R62612
H96241
T95274
R68803
R95001
T97592
AA485373
AA598053
R43910
R41787
N54596
AA488626
N54690
T53404
R11520
H47257
H12044
AA452820
AA282134
NO3680
H14208
AA484106
aa459824
T98298
R56774
AA+13899
R16838
R96626
AA482067
H00813
TO 1948
T60075
AA486533
AA437705
T49652
AA480001
AA478553
H15533
N59772
AI042404
N30000
H23137
R43817
T89490
AA401111
AA405030
AA40 1693
H16389
AA291103
AA484691
AA064180
AA455235
AA134871
AA406001
AA604101
AA152347
N59799
AA429895
H20543
AA187349
TO1100
₩15203
R41972
H28681
AA150828
H05140
H22854
H19203
AA484526
Several other genes were over- or underexpressed in ACCs compared with adrenocortical adenomas and normal adrenal cortex tissues, but we have not yet validated the expression pattern by real-time PCR. The presence of a gene on the list in Table 4 does not indicate that the gene product is either necessary or sufficient for causing ACC, but only that it is expressed as part of the complex pattern of gene expression that occurs during the initiation of the disease.
During the course of our studies, three other groups performed similar analyses of adrenocortical tissue (19, 30, 31). In the search for reliable markers for the
clinical management of adrenal tumors, de Fraipont et al. (31) designed an adrenal-specific microarray. They identified two clusters of genes, the IGF2 cluster and the steroidogenesis cluster, which, in combination, provide a good predictor of malignancy. As in our studies, they also confirmed the finding from the earlier analysis by Giordano et al. (19) demonstrating upregu- lation of IGF2 expression (30). Adrenal hyperplasia was analyzed in a similar fashion by Bourdeau et al. (30).
In summary, our findings indicate that microarray analysis can distinguish between ACC and adenomas by the differential expression of a set of the genes
| Gene name | Accession no. | Fold regulation | Up/down |
|---|---|---|---|
| Cathepsin H | AA487231 | 3.81 | up |
| Mucolipin 3 | AA171718 | 4.34 | up |
| Fibroblast growth factor receptor 1 (fms-related tyrosine kinase 2, | R54846 | 4.41 | up |
| Pfeiffer syndrome) | |||
| Aldo-keto reductase family 1, member C1 (dihydrodiol dehydrogenase 1; 20-alpha (3-alpha)-hydroxysteroid dehydrogenase) | R93124 | 3.45 | up |
| Fibronectin 1 | R62612 | 3.73 | up |
| Fibronectin 1 | R62612 | 3.51 | up |
| GAbinding protein transcription factor,alpha subunit 60 kDa | H96241 | 3.12 | up |
| Homo sapiens clone IMAGE:120162 mRNA sequence | T95274 | 3.24 | up |
| Hydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid deltaisomerase 1 | R68803 | 5.67 | down |
| yq51h12.s1 soares fetal liver spleen 1NFLS Homo sapiens cDNA clone | R95691 | 4.02 | up |
| IMAGE: 199367 3', mRNA sequence | |||
| EST | T97592 | 3.26 | up |
| Likely ortholog of rat vacuole membrane protein 1 | AA485373 | 2.89 | up |
| Periostin, osteoblast specific factor | AA598653 | 3.38 | up |
| yg22a04.s1 Soares infant brain 1NIB Homo sapiens cDNA clone | R43910 | 3.09 | up |
| IMAGE:32 962 3', mRNA sequence | |||
| Cadherin 13; H cadherin (heart) | R41787 | 3.57 | up |
| Insulin-like growth factor 2 (somatomedin A) | N54596 | 6.47 | up |
| Ubiquitin-like 1 (sentrin) | AA488626 | 2.96 | up |
| Insulin-like growth factor 2 (somatomedin A) | N54596 | 2.91 | up |
| Hypothetical protein from clone 643 | T53404 | 2.98 | up |
| Parathymosin | R11526 | 2.89 | up |
| Hypothetical protein MGC5306 | H47257 | 2.93 | down |
| Cytoplasmic FMR1 interacting protein 2 | H12044 | 6.03 | down |
| Purkinje cell protein 4 | AA452826 | 3.10 | down |
| Glutaminyl-peptide cyclotransferase (glutaminyl cyclase) | AA282134 | 3.08 | down |
| zb50h07.s1 Soares_fetal_lung_NbHL19W Homo sapiens cDNA clone | N93686 | 9.65 | down |
| IMAGE:307069 3', mRNA sequence | |||
| Paralemmin | H14208 | 3.55 | down |
| Hyaluronoglucosaminidase 1 | AA464196 | 3.41 | down |
| ATP-binding cassette, sub-family B (MDR/TAP), member 1 | AA459824 | 4.57 | down |
| Putative membrane protein | T98298 | 3.22 | down |
| Bone morphogenetic protein 1 | R56774 | 3.18 | down |
| Scavenger receptor class B, member 1 | AA443899 | 6.23 | down |
| Cytochrome p450, family 17, subfamily A, polypeptide 1 | R16838 | 2.87 | down |
| Chemokine (C-C motif) ligand 15 | R96626 | 4.73 | down |
| Retinoic acid receptor responder(tazarotene induced)2 | AA482067 | 3.39 | down |
| Glutathione S-transferase theta 1 | H99813 | 2.95 | down |
| FBJ murine osteosarcoma viral oncogene homolog B | T61948 | 6.82 | down |
| Fatty acid binding protein 5 (psoriasis-associated) | T60075 | 3.38 | down |
| Early growth response 1 | AA486533 | 3.26 | down |
| Immediate early response 3 | AA457705 | 3.04 | down |
| Arachidonate 5-lipoxygenase-activating protein | T49652 | 3.54 | down |
| Chemokine-like factor super family 3 | AA486561 | 3.19 | down |
| Dopachrome tautomerase(dopachrome delta-isomerase, tyrosinerelated protein 2) | AA478553 | 2.88 | down |
| Fibroblast growth factor 11 | H15533 | 3.62 | down |
| EST | N59772 | 6.40 | down |
| High mobility group AT-hook 1 | A1042404 | 3.73 | down |
| Glutathione S-transferase A3 | N30096 | 3.48 | down |
| Likely ortholog of mouse cancer related gene- liver 2 | H23137 | 3.13 | down |
| Human neuropeptide Y receptor Y1 (NPYY1) | R43817 | 3.60 | down |
| MAX protein | T89496 | 3.10 | down |
| Glucose phosphate isomerase | AA401111 | 8.53 | down |
| Microsomal glutathione S-transferase 1 | AA495936 | 6.87 | down |
| CD 163 antigen | AA401693 | 4.24 | down |
| Transcribed sequence with moderate similarity to protein pir:A36563 | H16389 | 3.28 | down |
| (H.sapiens) A36563 mannose receptor precursor - human | |||
| Glutaredoxin (thioltransferase) | AA291163 | 3.09 | down |
| Adlican | AA464691 | 3.50 | down |
| Glutathione peroxidase 3 (plasma) | AA664180 | 3.08 | down |
| Aldehyde dehydrogenase 1 family, member A3 | AA455235 | 3.64 | down |
| Fibulin 1 | AA134871 | 2.88 | down |
| Actin binding LIM protein 1 | AA406601 | 5.91 | down |
| Aldehyde dehydrogenase 1 family, member A1 | AA664101 | 5.25 | down |
| Gene name | Accession no. | Fold regulation | Up/down |
|---|---|---|---|
| Glutathione S-transferase A4 | AA152347 | 3.49 | down |
| Hypothetical protein MGC35366 | N59799 | 3.07 | down |
| ATP-binding cassette, subfamily C (CFTR/MRP), member 3 | AA429895 | 5.69 | down |
| DKFZP586B1621 protein | H20543 | 3.14 | down |
| Ferredoxin 1 | AA187349 | 9.33 | down |
| EST | T91100 | 2.94 | down |
| Solute carrier family 26 (sulfate transporter), member 2 | W15263 | 5.84 | down |
| DKFZp586P1124 | R41972 | 3.10 | down |
| Delta-notch-like EGF repeat-containing transmembrane | H28681 | 2.95 | down |
| Mitogen-activated protein kinase kinase kinase 5 | AA150828 | 2.93 | down |
| Regucalcin (senescence marker protein-30) | H05140 | 6.18 | down |
| Homo sapiens cDNA FLJ40165fis, clone TESTI2015962 | H22854 | 3.87 | down |
| Peroxiredoxin 3 | H19203 | 3.28 | down |
| Interleukin 1 receptor, type 1 | AA464526 | 2.84 | down |
ªListed in order of appearance in hierarchic cluster, Fig. 2.
IGF2
CgB
egr1
10
10
log2 expression (relative to reference)
5
5
cycles minus reference
0
0
-5
-5
-10
10
T6
T8
T10 T12
A4
A7
A9
A15
T6
T8
T10
T12
A4
A7
A9
A15
T6
T8
T10
T12
A4
A7
A9
A15
Adenomas
ACCs
Adenomas
ACCs
Adenomas
ACCs
analyzed. Further analysis of the IGF2 gene validated the experimental design. Underexpression of the CgB gene was found in both types of neoplasms, and the Egr-1 gene was downregulated in the ACCs. Further detailed analyses are warranted to elucidate the role of these genes in adrenal tumorigenesis.
Acknowledgements
We thank Michael Krause and Martin Eilers for assist- ance with the microarrays, Serdar Sel for help with the real-time PCR, and Brunhilde Chaloupka for excel- lent technical assistance.
References
1 Sidhu S, Gicquel C, Bambach CP, Campbell P, Magarey C, Robinson BG & Delbridge LW. Clinical and molecular aspects of adrenocortical tumourigenesis. Australian and New Zealand Journal of Surgery 2003 73 727-738.
2 Wooten MD & King DK. Adrenal cortical carcinoma. Cancer 1993 72 3145-3155.
3 Luton JP, Cerdas S, Billaud L, Thomas G, Guilhaume B, Bertagna X, Laudat MH, Louvel A, Chapus Y, Blondeau P, Bonnin A & Bricaire H. Clinical features of adrenocortical carci- noma, prognostic factors, and the effect of mitotane therapy. New England Journal of Medicine 1990 322 1195-1201.
4 Soreide JA, Braband K & Thoresen SO. Adrenal cortical carcinoma in Norway, 1970-1984. World Journal of Surgery 1992 16 663-668.
5 Copeland PM. The incidentally discovered adrenal mass. Annals of Internal Medicine 1983 98 940-945.
6 Lipsett MB, Hertz R & Ross GT. Clinical and pathophysiologic aspects of adrenocortical carcinoma. American Journal of Medicine 1963 35 374-383.
7 Stratakis CA. Genetics of adrenocortical tumors: gatekeepers, landscapers and conductors in symphony. Trends in Endocrinology and Metabolism 2003 14 404-410.
8 Russell AJ, Sibbald J, Haak H, Keith WN & McNicol AM. Increas- ing genome instability in adrenocortical carcinoma progression with involvement of chromosomes 3, 9 and X at the adenoma stage. British Journal of Cancer 1999 81 684-689.
9 Beuschlein F, Reincke M, Karl M, Travis WD, Jaursch-Hancke C, Abdelhamid S, Chrousos GP & Allolio B. Clonal composition of
human adrenocortical neoplasms. Cancer Research 1994 54 4927-4932.
10 Gicquel C, Leblond-Francillard M, Bertagna X, Louvel A, Chapula Y, Luton J-P, Girard F & Le Bouc Y. Clonal analysis of human adrenocortical carcinomas and secreting adenomas. Clini- cal Endocrinology 1994 40 465-477.
11 Kjellman M, Larsson C & Bäckdahl M. Genetic background of adrenocortical tumor development. World Journal of Surgery 2001 25 948-956.
12 Gicquel C, Raffin-Sanson M-L, Gaston V, Bertagna X, Plouin P-F, Schlumberger M, Louvel A, Luton J-P & Le Bouc Y. Structural and functional abnormalities at 11p15 are associated with the malignant phenotype in sporadic adrenocortical tumors: study on a series of 82 tumors. Journal of Clinical Endocrinology and Metabolism 1997 82 2559-2565.
13 Gicquel C, Bertagna X, Schneid H, Grancillard-Leblond M, Luton J-P, Girard F & Le Bouc Y. Rearrangements at the 11p15 locus and overexpression of insulin-like growth factor-II gene in sporadic adre- nocortical tumors. Journal of Clinical Endocrinology and Metabolism 1994 78 1444-1453.
14 Lack E. Tumors of the adrenal gland and extra-adrenal paraganglia. In Atlas of Tumor Pathology, 3rd Series, Fascicle 19, pp 125-151. Ed. J Rosai. Washington, DC: Armed Forces Institute of Pathology, 1997.
15 Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. American Journal of Sur- gery and Pathology 1984 8 163-169.
16 Yang YH, Dudoit S, Luu P, Lin DM, Peng V, Ngai J & Speed TP. Nor- malization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Research 2002 30 e15.
17 Sturn A, Quackenbush J & Trajanoski Z. Genesis: cluster analysis of microarray data. Bioinformatics 2002 18 207-208.
18 Higuchi R, Fockler C, Dollinger G & Watson R. Kinetic PCR ana- lysis: real-time monitoring of DNA amplification reactions. Biotechnology 1993 11 1026-1030.
19 Giordano TJ, Thomas DG, Kuick R, Lizyness M, Misek DE, Smith AL, Sanders D, Aljundi RT, Gauger PG, Thompson NW, Taylor JMG & Hanash SM. Distinct transcriptional profiles of adrenocortical tumors uncovered by DNA microarray analysis. American Journal of Pathology 2003 162 521-531.
20 Yoshida R, Ohuchi N & Kimura N. Clinicopathological study of chromogranin A, B and BRCA1 expression in node-negative breast carcinoma. Oncology Reports 2002 9 1363-1367.
21 Scopsi L, Sampietro G, Boracchi P & Collini P. Argyrophilia and chromogranin A and B immunostaining in patients with sporadic medullary thyroid carcinoma. A critical appraisal of their prog- nostic utility. Journal of Pathology 1998 184 414-419.
22 Yan SF, Fujita T, Lu J, Okada K, Shan ZY, Mackman N, Pinsky DJ & Stern DM. Egr-1, a master switch coordinating upregulation of divergent gene families underlying ischemic stress. Nature Medi- cine 2000 6 1355-1361.
23 Gashler A & Sukhatme VP. Early growth response protein 1 (Egr-1): prototype of a zinc-finger family of transcription factors. Progress in Nucleic Acid Research Molecular Biology 1995 50 191-224.
24 Kartasheva NN, Lenz-Bauer C, Hartmann O, Schäfer H, Eilers M & Dobbelstein M. ANp73 can modulate the expression of various genes in a p53-independent fashion. Oncogene 2003 22 8246-8254.
25 Madden SL, Galella EA, Zhu J, Bertelsen AH & Beaudry GA. SAGE transcript profiles for p53-dependent growth regulation. Oncogene 1997 15 1079-1085.
26 Abdulkadir SA, Qu Z, Garabedian E, Song SK, Peters TJ, Svaren J, Carbone JM, Naughton CK, Catalona WJ, Ackerman JJ, Gordon JI, Humphrey PA & Milbrandt J. Impaired prostate tumorigenesis in Egr1-deficient mice. Nature Medicine 2001 7 101-107.
27 Adamson ED & Mercola D. Egr1 transcription factor: multiple roles in prostate tumor cell growth and survival. Tumour Biology 2002 23 93-102.
28 Ji B, Bi Y, Simeone D, Mortensen RM & Logsdon CD. Human pan- creatic acinar cells lack functional responses to cholecystokinin and gastrin. Gastroenterology 2001 121 1380-1390.
29 Ji B, Chen X-Q, Misek DE, Kuick R, Hanash S, Ernst S, Najarian R & Logsdon CD. Pancreatic gene expression during the initiation of acute pancreatitis: identification of EGR-1 as a key regulator. Physiological Genomics 2003 14 59-72.
30 Bourdeau I, Antonini SR, Lacroix A, Kirschner LS, Matyakhina L, Lorang D, Libutti SK & Stratakis CA. Gene array analysis of macronodular adrenal hyperplasia confirms clinical heterogen- eity and identifies several candidate genes as molecular mediators. Oncogene 2004 23 1575-1585.
31 de Fraipont F, El Atifi M, Cherradi N, Le Moigne G, Defaye G, Houlgatte R, Bertherat J, Bertagna X, Plouin P-F, Baudin E, Berger F, Gicquel C, Chabre O & Feige J-J. Gene expression profiling of human adrenocortical tumors using complementary deoxyri- bonucleic acid microarrays identifies several candidate genes as markers of malignancy. Journal of Clinical Endocrinology and Metab- olism 2005 90 1819-1829.