Accepted Manuscript
Title: Combined Steroidogenic Characters of Fetal Adrenal and Leydig Cells in Childhood Adrenocortical Carcinoma
Author: Yasuko Fujisawa Kimiyoshi Sakaguchi Hiroyuki Ono Rie Yamaguchi Fumiko Kato Masayo Kagami Maki Fukami Tsutomu Ogata
The Journal of Steroid Biochemistry & Molecular Biology
| PII: | S0960-0760(16)30045-0 |
| DOI: | http://dx.doi.org/doi:10.1016/j.jsbmb.2016.02.031 |
| Reference: | SBMB 4652 |
| To appear in: | Journal of Steroid Biochemistry & Molecular Biology |
| Received date: | 19-11-2015 |
| Revised date: | 29-1-2016 |
| Accepted date: | 27-2-2016 |
Please cite this article as: Yasuko Fujisawa, Kimiyoshi Sakaguchi, Hiroyuki Ono, Rie Yamaguchi, Fumiko Kato, Masayo Kagami, Maki Fukami, Tsutomu Ogata, Combined Steroidogenic Characters of Fetal Adrenal and Leydig Cells in Childhood Adrenocortical Carcinoma, Journal of Steroid Biochemistry and Molecular Biology http://dx.doi.org/10.1016/j.jsbmb.2016.02.031
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Combined Steroidogenic Characters of Fetal Adrenal and Leydig Cells in Childhood Adrenocortical Carcinoma
Abbreviated title: Steroidogenesis in childhood adrenocartinoma
Yasuko Fujisawaª1, Kimiyoshi Sakaguchia1, Hiroyuki Onoª, Rie Yamaguchiª, Fumiko Katoa,
Masayo Kagamib, Maki Fukamib, Tsutomu Ogataª* tomogata@hama-med.ac.jp
ªDepartment of Pediatrics, Hamamatsu University School of Medicine, Hamamatsu, Japan Department of Molecular Endocrinology, National Research Institute for Child Health and Development, Tokyo, Japan
Ccorresponding author.
1These authors contributed equally to this work.
ACCEPTED MANUSCRIPT
DHEA-S
Common ancestors
Carcinogenesis
Fetal adrenal
c-ACC
DHEA-S Testosterone
Testosterone
Testis
Highlights
· We report childhood adrenocortical carcinoma (c-ACC) with a germline TP53 mutation.
· Postzygotic tumorigenic factors are similar to those reported previously in c-ACCs.
· Testosterone and dehydroepiandrosterone are efficiently produced in this c-ACC.
· This c-ACC has combined steroidogenic property of fetal adrenal and Leydig cells.
Abstract
Although childhood adrenocortical carcinomas (c-ACCs) with a TP53 mutation are known to produce androgens, detailed steroidogenic characters have not been clarified. Here, we examined steroid metabolite profiles and expression patterns of steroidogenic genes in a c-ACC removed from the left adrenal position of a 2-year-old Brazilian boy with precocious puberty, using an atrophic left adrenal gland removed at the time of tumorectomy as a control. The c-ACC produced not only abundant dehydroepiandrosterone-sulfate but also a large amount of testosterone via the 45 pathway with 45-androstenediol rather than 44-androstenedione as the primary intermediate metabolite. Furthermore, the c-ACC was associated with elevated expressions of CYP11A1, CYP17A1, POR, HSD17B3, and SULT2A1, a low but similar expression of CYB5A, and reduced expressions of AKR1C3 (HSD17B5) and HSD3B2. Notably, a Leydig cell marker INSL3 was expressed at a low but detectable level in the c-ACC.
Furthermore, molecular studies revealed a maternally inherited heterozygous germline TP53 mutation, and several post-zygotic genetic aberrations in the c-ACC including loss of paternally derived chromosome 17 with a wildtype TP53 and loss of maternally inherited chromosome 11 and resultant marked hyperexpression of paternally expressed growth promoting gene IGF2 and drastic hypoexpression of maternally expressed growth suppressing gene CDKN1C. These results imply the presence of combined steroidogenic properties of fetal adrenal and Leydig cells in this patient’s c-ACC with a germline TP53 mutation and several postzygotic carcinogenic events.
Abbreviations
aCGH: array comparative genomic hybridization
CA: control adrenal
c-ACC: childhood adrenocortical carcinoma
45A-diol: androstenediol
44A-dione: androstenedione
DHEA: dehydroepiandrosterone
DHEA-S: DHEA-sulfate
ACCEPTED MANUSCRIPT
DHT: dihydrotestosterone
DMR: differentially methylated region
NR: normal range
P4: progesterone
P5: pregnenolone
17-OHP4: 17-OH progesterone
17-OHP5: 17-OH pregnenolone
qPCR: quantitative PCR
T: testosterone
TART: testicular adrenal rest tumor
Keywords: adrenocortical carcinoma; androgen production; steroidogenesis; tumorigenesis; TP53
1. Introduction
Childhood adrenocortical carcinoma (c-ACC) is highly prevalent in the south Brazilian area including the Paraná state, with an estimated prevalence being 10-15 times higher than that in the rest of the world (3.4-4.2 vs. 0.3 per million children under 15 years of age) [1, 2]. The high prevalence is primarily due to the presence of a founder p.R337H mutation at exon 10 of TP53 encoding tumor protein p53 in this area [2-5]. To date, extensive studies have been performed for the c-ACCs with the heterozygous germline TP53 founder mutation, revealing multiple underlying carcinogenic factors such as post-zygotic loss of chromosome 17 with the wildtype TP53 allele in virtually all c-ACCs [2, 5, 6], loss of the maternally inherited chromosome 11p15 and resultant hyperexpression of paternally expressed growth-promoting gene IGF2 and hypoexpression of maternally expressed growth-suppressing gene CDKN1C in most c-ACCs [5, 7], and other genetic aberrations including loss of chromosome 4q34 and gains of chromosome 9q33-q34 and chromosome 19p13-q13 in a substantial fraction of c-ACCs [6]. By contrast, other genetic aberrations that have frequently been found in ACCs, such as somatic gain-of-function mutations of CTNNB encoding ß-catenin, potentially disease-causing mutations in GNAS, NF2, and RB1, and copy number alterations of TERT, ZNRF3, and KREMEN1, are rarely identified in TP53 mutation positive c-ACCs [5-8]. Furthermore, the early onset of manifestations, the histological features, and the gene expression profiles including the weak HSD3B2 expression consistently imply that the TP53 mutation positive ACCs are of fetal adrenal origin [7, 9, 10]. In agreement with this, serum dehydroepiandrosterone (DHEA)-sulfate (DHEA-S) are markedly elevated in affected patients, and this would explain at least in part why such c-ACCs are primarily identified in boys with precocious puberty or in girls with virilization [11].
To our knowledge, however, detailed profiles of intratumoral and blood steroid metabolites and expression patterns of steroidogenic genes have not been examined. Thus, while elevated serum testosterone (T) has also been reported in several patients with c-ACCs [11], it remains unknown whether T is actually produced in c-ACCs or converted from DHEA in extratumoral tissue(s). Here, we report combined steroidogenic characters of fetal adrenal and Leydig cells in a TP53 germline mutation positive c-ACC, as well as detailed post-zygotic
carcinogenic factors.
2. Experimental procedures
2.1. Clinical history
A Brazilian boy was born in Japan to non-consanguineous parents at 38 weeks of gestation, with a birth weight of 2,75 kg (-0.6 SD for Japanese). The parents were of the Paraná state origin and, while the parents and the two siblings were clinically normal, three maternal relatives were allegedly afflicted with cancers (Figure 1A).
At 2 7/12 year of age, he was referred to us because of secondary sexual development and
accelerated growth that were noticed from 2 years old. His height was 96.2 cm (+2.0 SD), and
his weight 19.2 kg (+4.9 SD). Physical examination revealed several pubic hairs, large penis (
6 cm), and intrascrotal testes of prepubertal size (2 mL). There were no obvious Cushing-like
features. Standard radioimmunoassay showed increased serum DHEA-S (870 µg/dL =
23.7 µmol/L) (normal range [NR], < 40 µg/dL), T (4.3 ng/ml=15 nmol/L) (NR, <0.05 ng/ml),
and dihydrotestosterone (DHT) (0.5 ng/ml = 1.7 nmol/L) (NR, <0.01 ng/mL), and suppressed
serum luteinizing hormone (<0.2 IU/L) and follicle-stimulating hormone (<0.2 IU/L). Serum
cortisol at 8:00 am was 10.8 ug/dL (=298 nmol/L) (NR, 3-21 µg/dL), and was mildly
suppressed to 6.2 µg/dL (=171 nmol/L) after a low-dose dexamethasone intake (20 µg/kg p.o.
at 11 pm of the previous day). Abdominal computed tomography and magnetic resonance
imaging delineated a mass of~4×4×3 cm at the left adrenal position (Figure 1B). The mass of
26.4 g was completely removed, together with an atrophic left adrenal gland which was utilized
as the control adrenal (CA) in this study. Macroscopic examination at the time of surgery
showed no discernible metastatic region or invasion to the surrounding tissues. Histological
examinations showed high nuclear atypia, atypical mitotic figures, eosinophilic cytoplasm,
sinusoidal invasion, and diffuse architecture; immunohistochemical examinations revealed
positive NR5A1 (alias, SF-1) staining characteristic of steroidogenic cells [12], TP53 nuclear
staining suggestive of TP53 mutation [13], increased Ki67 index of 11% indicative of
malignancy (>5% is considered as malignant) [14], and faint HSD3B2 and strong CYP17A1
staining consistent with fetal adrenal [15] (Figure 1C). His hormone values were normalized
shortly after the surgery. Thus, the mass was diagnosed as the stage I c-ACC according to the Weiss criteria and the IPACTR staging [10, 16, 17]. At present, he is 3 5/12 years old, and is under close follow-up without further treatment, as well as his mother and younger brother who were found to have the TP53 p.R337H mutation.
2.2. Ethical approval
This study was approved by the Institutional Review Board Committee of Hamamatsu University School of Medicine, and performed after obtaining written informed consent.
2.3. Measurements of steroid metabolites
Multiple steroid metabolites were simultaneously measured by the liquid chromatography-tandem mass spectrometry [18], using ~100 mg of the c-ACC and the CA homogenates prepared at the time of the operation and ~0.3 ml of serum samples obtained before and one month after the operation. We measured steroid metabolites not only on the conventional frontdoor pathway that produces DHT via T but also on the alternative backdoor pathway that yields DHT without T intermediacy [15, 19]. The conversion factors from metric units to SI units for blood steroid metabolites are shown in Supplementary Table 1.
2.4. Molecular studies
mRNA was obtained from the c-ACC and the CA tissues. Genomic DNA was extracted from leukocytes, the c-ACC, and the CA of the patient, and from leukocytes of the siblings, the parents, and a control male. The primers utilized are shown in Supplementary Table 2.
Expression dosage analyses were carried out for genes involved in steroidogenesis and a Leydig marker INSL3 by quantitative PCR (qPCR) on StepOnePlus system (Life Technologies, Carlsbad, CA, USA), using ACTB, GAPDH, PPIA, and 18S as internal controls. We also performed DNA chip analysis (3D-Gene) (Toray, Tokyo, Japan) including CTNNB1, GNAS,
ZNRF3, and KREMEN1 that are often mutated in ACCs [20].
Direct sequencing and microsatellite genotyping were performed by the standard methods on the ABI 3130xl Genetic Analyzer (Life Technologies). Gene copy number analysis was carried out by qPCR on StepOnePlus system, using GAPDH as an internal control. Genomewide array comparative genomic hybridization (aCGH) and SNP array were performed using a catalog human array (SurePrint G3 Human CGH+SNP 4×180K format, ID G4890A) (Agilent Technologies, Santa Clara, CA, USA); aCGH was carried out using leukocyte DNA samples of the patient and a control male, and leukocyte and c-ACC genomic DNA samples of the patient. Methylation analysis was carried out for multiple differentially methylated regions (DMRs) (Supplementary Table 3) by pyrosequencing (PyroMark Q24) (Qiagen, Venlo, Netherlands), using bisulfite-treated genomic DNA samples.
2.3. Experiments using the NCI-H295R cells
The NCI-H295R cells (CRL-2128, American Type Culture Collection) originated from an adrenocortical carcinoma in an adult female [21] are known to have fetal adrenal-like character, as are c-ACCs [22]. Thus, we also examined steroid metabolite profiles and gene expression patterns of the H295R cells. In brief, the H295R cells were seeded into 6-well dishes (1×106 cells/well) and were cultured in DMEM/Ham’s F-12 medium (Life Technologies) supplemented with 2.5% Nu-Serum (BD Biosciences) and 1% insulin/transferrin/selenium premix (Corning) for 48 hours. Subsequently, ~2×107 of pooled cells (~80 mg of cell homogenates) and ~3 ml of pooled media were utilized to examine the steroidogenic character.
3. Results
3.1. Profiling of steroid metabolites
The results are shown in Figure 2A. As compared with the CA, pregnenolone (P5), 17-OH pregnenolone (17-OHP5), DHEA, and DHEA-S were obviously elevated in the c-ACC. Notably, androstenediol (45A-diol) and T were also markedly high in the c-ACC. The conversion from 17-OHP5 to DHEA was apparently compromised, as was that from 17-OH
progesterone (17-OHP4) to androstenedione (44A-dione). The conversion from P5 to progesterone (P4) and that from DHEA to 44A-dione were also apparently compromised, whereas the conversion from 17-OHP5 to 17-OHP4 was fairly preserved and that from 45A-diol to T was apparently smooth. DHT was barely produced in the c-ACC, and the backdoor pathway for androgen production was not operating in the c-ACC (Supplementary Figure 1). Pre-operation steroid metabolite profile was considerably different between the ACT and the serum, including the 17-OHP5/DHEA, 17-OHP4/44A-dione, 17-OHP5/17-OHP4, DHEA/44A-dione, and DHEA/45A-diol ratios. Serum steroid profile was normalized at one month after the operation.
3.2. Expression patterns of genes involved in steroidogenesis and a Leydig marker INSL3
The results are shown in Figure 2B. As compared with the CA, the c-ACC was associated with elevated expressions of CYP11A1, CYP17A1, POR, HSD17B3, and SULT2A1, a low but similar level of expression of CYB5A, and reduced expressions of AKR1C3 (HSD17B5) and HSD3B2. In addition, CYP2142 and CYP11B1 were expressed at low and slightly high levels, respectively, and SRD5A1, SRD542, and HSD3B1 were barely expressed. Notably, INSL3 was expressed at a low but detectable level in the c-ACC and the CA, with obviously stronger expression in the c-ACC than in the CA. The fold change (the c-ACC vs. the CA) was most remarkable for HSD17B3, followed by INSL3, SULT2A1, CYP11A1, and CYP17A1 (although the fold change was also remarkable for SRD5A1, SRD5A1 was poorly expressed in both tissues).
3.3. Experiments using H295R cells
The results are shown in Supplementary Figure 2. Steroid metabolite profiling revealed that the H295R cells contained markedly high level of P5, relatively high levels of 17-OHP5, 11-DOF, and DHEA, relatively low levels of 17-OHP4, DHEA-S, and 44A-dione, obviously low level of P4, and extremely low levels of cortisol, 45A-diol, T, and DHT. The media contained
markedly high level of 11-DOF, relatively high levels of DHEA-S and 44A-dione, relatively low levels of P5 and 17-OHP5, obviously low levels of cortisol and DHEA, and extremely low levels of P4, 17-OHP4, 45A-diol, T, and DHT. Gene expression analysis indicated that the H295R cells were associated with relatively high expressions of CYP11A1, CYP17A1, POR, CYB5A, AKR1C3 (HSD17B5), CYP21A2, and SULT2A1, relatively low expressions of HSD17B3, HSD3B1, HSD3B2, and CYP11B1, and severely low expressions of SRD5A1 and SRD5A2, as well as INSL3.
3.3. Germline TP53 mutation and post-zygotic genetic rearrangements in the c-ACC
A heterozygous p.R337H mutation of TP53 was identified in the leukocyte of this boy, the mother, and the younger brother, but not in the father and the elder sister (Figure 1A and Figure 3A). In the c-ACC, this mutation was present in an apparently hemizygous condition and, consistent with this, aCGH and SNP array analyses indicated loss of the whole chromosome 17 (Figure 3A). This was further confirmed by qPCR analysis for TP53 copy number, and microsatellite analysis for D17S831 and D17S949 revealed loss of the paternally inherited chromosome 17 with the wildtype TP53 from the c-ACC (Figure 3A). In this regard, the results obtained using the c-ACC genomic DNA, such as the minor TP53 “G” allele in direct sequencing and the minor peaks for paternally derived alleles in microsatellite analysis, were considered to derive from contaminated non-tumor tissue.
aCGH and SNP array analyses, qPCR for CDKN1C, and microsatellite analysis also delineated loss of the whole chromosome 11 of maternal origin in virtually all the c-ACC cells, with the trace of contaminated non-tumor tissue (Figure 3B). In agreement with this, pyrosequencing-based methylation analysis revealed the hypermethylated H19-DMR and the hypomethylated KvDMR1 in the c-ACC, and qPCR analysis showed marked hyperexpression of IGF2 and drastic hypoexpression of CDKN1C (Figure 3B).
aCGH, SNP array, and microsatellite analyses also delineated loss of the whole chromosome 4 of maternal origin and copy number gains of the middle part of chromosome 19p of paternal origin in actually all the c-ACC cells (Figure 3C). Furthermore, such analyses
indicated complex rearrangements including copy number gains and losses of the distal part of chromosome 19q, copy number gains of the whole chromosome 20 of maternal origin, copy number gains of most of the Xp, and copy number losses of most of the Xq and the whole Y in a substantial fraction of the c-ACC cells (Figure 3C). The copy number gains of maternally derived chromosome 20 were also supported by methylation analysis (Supplementary Table 3). In addition, variable degrees of copy number alterations were also implicated for chromosomes 1, 5, 7, 8, 14, and 15 in a certain fraction of the c-ACC cells by aCGH, SNP array, and methylation analyses.
3.4. Expression patterns of putative carcinogenic genes for ACTs
No significant alterations (fold change, >2.0 or <0.5) was identified for CTNNB1, GNAS, ZNRF3, and KREMEN1 by DNA chip analysis.
4. Discussion
We studied steroid metabolite profiles and gene expression patterns in the c-ACC of this Brazilian boy. The expression data imply that the c-ACC has steroidogenic properties of not only fetal adrenal but also Leydig cells, because the c-ACC was associated with (i) high CYP11A1 and CYP17A1 expressions common to both fetal adrenal and Leydig cells, (ii) severely reduced HSD3B2 expression and obviously high SULT2A1 expression indicative of fetal adrenal character, and (iii) markedly high HSD17B3 expression and a demonstrable level of INSL3 expression characteristic of Leydig cells [15, 23]. This notion would explain why the c-ACC was capable of producing not only abundant DHEA/DHEA-S but also a large amount of T primarily via the 45 pathway, with 45A-diol rather than 44A-dione being the primary intermediate metabolite.
Such combined steroidogenic characters of fetal adrenal and Leydig cells have also been reported in testicular adrenal rest tumors (TARTs) in males [24]. In this regard, steroidogenic cells in the fetal adrenal and the gonad are derived from the common ancestors, and human fetal
adrenal cells appear to express both AKRIC3 and HSD17B3 around 8-9 weeks post conception [25]. It would be possible, therefore, that ACTs and TARTs may have acquired pluripotential steroidogenic functions that have once been exhibited by the common ancestral cells. Furthermore, since serum T is also increased in virilizing girls with c-ACCs [11], c-ACCs in girls may also have acquired Leydig cell-like property.
Several findings should be pointed out with regard to the steroidogenic characters. First, the conversion from 17-OHP5 to DHEA and that from 17-OHP4 to 44A-dione were rather compromised in the c-ACC. This would primarily be due to the relatively low 17/20 lyase activity as compared with 17a-hydroxylase activity of CYP17A1 [26], and relatively low expression of CYB5A required for the 17/20 lyase function [15]. Furthermore, since DHEA would efficiently be converted into DHEA-S and 45A-diol, this would result in a drastic difference between 17-OHP5 and DHEA concentrations within the c-ACC. Second, the conversion from 17-OHP5 to 17-OHP4 and that from 45A-diol to T were apparently well preserved despite the weak HSD3B2 expression in the c-ACC. This may be explained by assuming that accumulation of a large amount of substrates maximally stimulated the residual HSD3B2 activity, especially for the conversion from 45A-diol to T in the c-ACC with a character of Leydig cells that produce T as the final bioactive product. Third, steroid metabolite profile was considerably different between the c-ACC and the serum. This would not be surprising, because serum steroid metabolites derive from not only the c-ACC but also extratumoral tissues. Furthermore, it is likely that intermediate metabolites such as 17-OHP5 are just leaked from the c-ACC to the blood flow, whereas biologically important metabolites such as DHEA/DHEA-S involved in the feto-placental unit [15] are positively secreted into the blood flow. Lastly, the backdoor pathway was not operating in the c-ACC.
We also attempted to reveal underlying carcinogenic factors of the c-ACC in this patient (Supplementary Figure 3). Notably, of the genetic aberrations identified in virtually all the c-ACC cells of this patient, loss of chromosomes 17 with a wildtype TP53, and loss of maternally inherited chromosome 11 and resultant marked hyperexpression of IGF2 and drastic hypoexpression of CDKN1C, have been detected in most, if not all, of TP53 mutation positive c-ACCs [7]. In this context, loss of the maternally derived chromosome 11 would have played a
major role in the carcinogenic process, by providing a drastic growth potential. Indeed, CDKN1C is strongly expressed in the adrenal, and gain-of-function mutations of CDKN1C lead to IMAGe syndrome with adrenal aplasia/hypoplasia [27], whereas loss-of-function mutations of CDKN1C result in Beckwith-Wiedemann syndrome characterized by the frequent occurrence of c-ACTs [28]. Furthermore, IGF2 expression was markedly increased, probably due to loss of functional CDKN1C in the c-ACTs [7, 29]. In addition, loss of chromosome 4 and copy number gain of the middle part of 19p have also frequently been identified in TP53 mutation positive c-ACCs [6]. By contrast, the c-ACC of this patient was free from altered expression dosages of CTNNB1, GNAS, ZNRF3, and KREMEN1, as in most TP53 mutation positive c-ACCs [6, 7]. Thus, fairly common carcinogenic factors appear to be operating in TP53 mutation positive c-ACCs.
The experiments using the H295R cells also provide useful implications. It is likely that the H295R cells basically have steroidogenic characters of fetal adrenal and the zona fasciculata and the zona reticularis of permanent adrenal, although they have barely contained CYP11B1 activity. Thus, the H295R cells, as well as the c-ACC, appear to have acquired pluripotential steroidogenic functions. In this regard, the difference in steroid metabolite profile including the T and cortisol values between the H295R cells and the c-ACC would be due to the difference in the original tissue and the underlying carcinogenic factors. While T production in the H295R cells was severely compromised despite the presence of a relatively high AKR1C3 expression, previous studies have shown that H295A cells are virtually incapable of producing T despite AKR1C3 expression level being similar to that of H295R cells [30]. Thus, adrenal T production would be subject to some factors other than the AKR1C3 expression level.
Furthermore, the discrepancy in steroid metabolite values between the H295R cells and the media would argue for the notion that intermediate metabolites (e.g., P5, 17-OHP4, and DHEA) are just leaked from the H295R cells into the media, whereas final products (probably 11-DOF, DHEA-S, and 45A-dione) are positively secreted from the H295R cells into the media.
5. Conclusions
The results imply the presence of combined steroidogenic characters of fetal adrenal and Leydig cells in this patient’s c-ACC with a germline TP53 mutation and several postzygotic carcinogenic events. However, the results have been obtained from a single c-ACC, and further studies are required to reveal steroidogenic and carcinogenic properties common to TP53 mutation positive c-ACCs and those specific to each c-ACC.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgments
We would like to thank Professor Hidenobu Sasano for the histopathological diagnosis. This work was supported by Grants-in-Aid for Scientific Research on Innovative Areas from the Ministry of Education, Culture, Sports, Science and Technology (22132004-A01), by Grants-in-Aid for Scientific Research (A) from the Japan Society for the Promotion of Science (25253023), and by Grants for Research on Intractable Diseases from the Ministry of Health, Labor and Welfare (H27-025).
References
[1] R. Sandrini, R.C. Ribeiro, L. DeLacerda, Childhood adrenocortical tumors, J. Clin. Endocrinol. Metab. 82 (1997) 2027-2031.
[2] R.C. Ribeiro, F. Sandrini, B. Figueiredo, G.P. Zambetti, E. Michalkiewicz, A.R. Lafferty, L. DeLacerda, M. Rabin, C. Cadwell, G. Sampaio, I. Cat, C.A. Stratakis, R. Sandrini, An inherited p53 mutation that contributes in a tissue-specific manner to pediatric adrenal cortical carcinoma, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 9330-9335.
[3] E.M. Pinto, A.E. Billerbeck, M.C. Villares, S. Domenice, B.B. Mendonca, A.C. Latronico, Founder effect for the highly prevalent R337H mutation of tumor suppressor p53 in Brazilian patients with adrenocortical tumors, Arq. Bras. Endocrinol. Metab. 48 (2004) 647-650.
[4] S. Garritano, F. Gemignani, E.I. Palmero, M. Olivier, G. Martel-Planche, F. Le Calvez-Kelm, L. Brugieres, F.R. Vargas, R.R. Brentani, P. Ashton-Prolla, S. Landi, S.V. Tavtigian, P. Hainaut, M.I. Achatz, Detailed haplotype analysis at the TP53 locus in p.R337H mutation carriers in the population of Southern Brazil: evidence for a founder effect, Hum. Mutat. 31 (2010) 143-150.
[5] T. Else, A.C. Kim, A. Sabolch, V.M. Raymond, A. Kandathil, E.M. Caoili, S. Jolly, B.S. Miller, T.J. Giordano, G.D. Hammer, Adrenocortical carcinoma, Endocr. Rev. 35 (2014) 282-326.
[6] E. Letouze, R. Rosati, H. Komechen, M. Doghman, L. Marisa, C. Fluck, R.R. de Krijger, M.M. van Noesel, J.C. Mas, M.A. Pianovski, G.P. Zambetti, B.C. Figueiredo, E. Lalli, SNP array profiling of childhood adrenocortical tumors reveals distinct pathways of tumorigenesis and highlights candidate driver genes, J. Clin. Endocrinol. Metab. 97 (2012) E1284-1293.
[7] A.N. West, G.A. Neale, S. Pounds, B.C. Figueredo, C. Rodriguez Galindo, M.A. Pianovski, A.G. Oliveira Filho, D. Malkin, E. Lalli, R. Ribeiro, G.P. Zambetti, Gene expression profiling of childhood adrenocortical tumors, Cancer Res. 67 (2007) 600-608.
[8] L.F. Leal, L.M. Mermejo, L.Z. Ramalho, C.E. Martinelli, Jr., J.A. Yunes, A.L. Seidinger, M.J. Mastellaro, I.A. Cardinalli, S.R. Brandalise, A.C. Moreira, L.G. Tone, C.A. Scrideli,
M. Castro, S.R. Antonini, Wnt/beta-catenin pathway deregulation in childhood adrenocortical tumors, J. Clin. Endocrinol. Metab. 96 (2011) 3106-3114.
[9] C.L. Coulter, Fetal adrenal development: insight gained from adrenal tumors, Trends Endocrinol. Metab. 16 (2005) 235-242.
[10] E. Michalkiewicz, R. Sandrini, B. Figueiredo, E.C. Miranda, E. Caran, A.G. Oliveira-Filho, R. Marques, M.A. Pianovski, L. Lacerda, L.M. Cristofani, J. Jenkins, C. Rodriguez-Galindo, R.C. Ribeiro, Clinical and outcome characteristics of children with adrenocortical tumors: a report from the International Pediatric Adrenocortical Tumor Registry, J. Clin. Oncol. 22 (2004) 838-845.
[11] W. Bonfig, I. Bittmann, S. Bechtold, B. Kammer, V. Noelle, S. Arleth, K. Raile, H.P. Schwarz, Virilising adrenocortical tumours in children, Eur. J. Pediatr. 162 (2003) 623-628.
[12] S. Sbiera, S. Schmull, G. Assie, H.U. Voelker, L. Kraus, M. Beyer, B. Ragazzon, F. Beuschlein, H.S. Willenberg, S. Hahner, W. Saeger, J. Bertherat, B. Allolio, M. Fassnacht, High diagnostic and prognostic value of steroidogenic factor-1 expression in adrenal tumors, J. Clin. Endocrinol. Metab. 95 (2010) E161-171.
[13] N. Slade, U.M. Moll, Mutational analysis of p53 in human tumors: immunocytochemistry, Methods. Mol. Biol. 234 (2003) 231-243.
[14] A. Stojadinovic, M.F. Brennan, A. Hoos, A. Omeroglu, D.H. Leung, M.E. Dudas, A. Nissan, C. Cordon-Cardo, R.A. Ghossein, Adrenocortical adenoma and carcinoma: histopathological and molecular comparative analysis, Mod. Pathol. 16 (2003) 742-751.
[15] W.L. Miller, R.J. Auchus, The molecular biology, biochemistry, and physiology of human steroidogenesis and its disorders, Endocr. Rev. 32 (2011) 81-151.
[16] L.M. Weiss, Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors, Am. J. Surg. Pathol. 8 (1984) 163-169.
[17] L.M. Weiss, L.J. Medeiros, A.L. Vickery, Jr., Pathologic features of prognostic significance in adrenocortical carcinoma, Am. J. Surg. Pathol. 13 (1989) 202-206.
[18] S. Arai, Y. Shibata, Y. Nakamura, B. Kashiwagi, T. Uei, Y. Tomaru, Y. Miyashiro, S. Honma, K. Hashimoto, Y. Sekine, K. Ito, H. Sasano, K. Suzuki, Development of prostate
cancer in a patient with primary hypogonadism: intratumoural steroidogenesis in prostate cancer tissues, Andrology 1 (2013) 169-74.
[19] M. Fukami, K. Homma, T. Hasegawa, T. Ogata, Backdoor pathway for dihydrotestosterone biosynthesis: implications for normal and abnormal human sex development, Dev. Dyn. 242 (2013) 320-329.
[20] C.C. Juhlin, G. Goh, J.M. Healy, A.L. Fonseca, U.I. Scholl, A. Stenman, J.W. Kunstman, T.C. Brown, J.D. Overton, S.M. Mane, C. Nelson-Williams, M. Backdahl, A.C. Suttorp, M. Haase, M. Choi, J. Schlessinger, D.L. Rimm, A. Hoog, M.L. Prasad, R. Korah, C. Larsson, R.P. Lifton, T. Carling, Whole-exome sequencing characterizes the landscape of somatic mutations and copy number alterations in adrenocortical carcinoma, J. Clin. Endocrinol. Metab. 100 (2015) E493-502.
[21] A.F. Gazdar, H.K. Oie, C.H. Shackleton, T.R. Chen, T.J. Triche, C.E. Myers, G.P. Chrousos, M.F. Brennan, C.A. Stein, R.V. La Rocca, Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis, Cancer Res. 50 (1990) 5488-5496.
[22] B. Staels, D.W. Hum, W.L. Miller, Regulation of steroidogenesis in NCI-H295 cells: a cellular model of the human fetal adrenal, Mol. Endocrinol. 7 (1993) 423-433.
[23] A. Ferlin, C. Foresta, Insulin-like factor 3: a novel circulating hormone of testicular origin in humans, Ann. N. Y. Acad. Sci. 1041 (2005) 497-505.
[24] E.E. Smeets, P.N. Span, A.E. van Herwaarden, R.A. Wevers, A.R. Hermus, F.C. Sweep,
H.L. Claahsen-van der Grinten, Molecular characterization of testicular adrenal rest tumors in congenital adrenal hyperplasia: lesions with both adrenocortical and Leydig cell features, J. Clin. Endocrinol. Metab. 100 (2015) E524-530.
[25] M. Goto, K. Piper Hanley, J. Marcos, P.J. Wood, S. Wright, A.D. Postle, I.T. Cameron, J.I. Mason, D.I. Wilson, N.A. Hanley, In humans, early cortisol biosynthesis provides a mechanism to safeguard female sexual development, J. Clin. Invest. 116 (2006) 953-960.
[26] M. Katagiri, N. Kagawa, M.R. Waterman, The role of cytochrome b5 in the biosynthesis of androgens by human P450c17, Arch. Biochem. Biophys. 317 (1995) 343-347.
[27] V.A. Arboleda, H. Lee, R. Parnaik, A. Fleming, A. Banerjee, B. Ferraz-de-Souza, E.C.
Delot, I.A. Rodriguez-Fernandez, D. Braslavsky, I. Bergada, E.C. Dell’Angelica, S.F. Nelson, J.A. Martinez-Agosto, J.C. Achermann, E. Vilain, Mutations in the PCNA-binding domain of CDKNIC cause IMAGe syndrome, Nat. Genet, 44 (2012) 788-792.
[28] R. Weksberg, C. Shuman, J.B. Beckwith, Beckwith-Wiedemann syndrome, Eur. J. Hum. Genet. 18 (2010) 8-14.
[29] F. Wilkin, N. Gagne, J. Paquette, L.L. Oligny, C. Deal, Pediatric adrenocortical tumors: molecular events leading to insulin-like growth factor II gene overexpression, J. Clin. Endocrinol. Metab. 85 (2000) 2048-2056.
[30] E. Samandari, P. Kempna, J.M. Nuoffer, G. Hofer, P.E. Mullis, C.E. Fluck, Human adrenal corticocarcinoma NCI-H295R cells produce more androgens than NCI-H295A cells and differ in 3beta-hydroxysteroid dehydrogenase type 2 and 17,20 lyase activities, J. Endocrinol. 195 (2007) 459-472.
Figure Captions
A
C
I
1
2
3
4
II
1
2
II
1
2
*
*
3
4
HE
NR5A1
N
*
1 *
2 *
3
7
B
TP53
Ki67
€
€
HSD3B2
CYP17A1
Figure 2. Steroidogenic characters of the c-ACC in this Brazilian boy. c-ACC: childhood adrenocortical carcinoma; and CA: control adrenal. A. Intratumoral and serum steroid metabolite values. All the metabolites were measured by the liquid chromatography-tandem mass spectrometry, except for serum DHEA-S value that was obtained by the standard radioimmunoassay (indicated by an asterisk). The metabolic pathway common to both fetal adrenal and Leydig cells is highlighted with green, that characteristic of the fetal adrenal cells with blue, and that characteristic of Leydig cells with yellow. For the convenience in the comparison, all the blood steroid metabolites are shown with the same metric unit (ng/ml). B. Expression patterns of genes involved in steroidogenesis and INSL3, using ACTB (ß-actin) as an internal control. Similar results have been obtained when GAPDH, PPIA, and 18S were utilized as internal controls. In the left part, minus values represent genes expressed more strongly than ACTB.
ACCEPTED MANUSCRIPT
A
DHEA-S
Adrenal 3,900 vs. N.M.
Cholesterol
Blood 8,700* - > N.M.
StAR CYP11A1
SULT2A1 PAPSS2
Fdx-FdR
P5
17-OHP5
DHEA
45A-diol
Adrenal
CYP17A1 (17a-OHlase) POR
Adrenal
CYP17A1 (17/20 lyase) POR
Adrenal
HSD17B3
Adrenal
5,910 vs. 270
10,960 vs. 110
428 vs. 42
392 vs. 3
Blood
Blood
Cyt b5
Blood
Blood
4.1→0.2
64.5→0.3
35.5→0.03
5.3 → <0.01
HSD3B2
HSD3B2
HSD3B2
HSD3B2
P4
17-OHP4
44A-dione
Testosterone
Adrenal
CYP17A1 (17a-OHlase) POR
Adrenal
CYP17A1 (17/20 lyase) POR
Adrenal
HSD17B3 AKR1C3
Adrenal
45 vs. 147
4,060 vs. 160
99 vs. 50
725 vs. 19
Blood
Blood
Cyt b5
Blood
Blood
0.1 → 0.03
2.8→0.2
1.6 →0.03
3.9 →0.01
CYP21A2 POR
SRD5A1/2
11-DOF
DHT
Adrenal 195 vs. 100
Adrenal
9 vs. < 1
Blood 1.5→0.7
Blood
0.5→<0.01
CYP11B1
Fdx-FdR
Cortisol
Steroid metabolite
Adrenal
Adrenal (ng/g wet tissue)
2,370 vs. 1,560
c-ACC vs. CA
Blood
Blood value (ng/ml)
69 →74
Pre → Post-operation
B
CYP11A1
CYP17A1
POR
CYB5A
HSD17B3
AKR1C3
SRD5A1
SRD5A2
HSD3B1 HSD3B2
CYP21A2
CYP11B1
SULT2A1
INSL3
5
0
+5
+10
+15
+20
+25
c-ACT
+30
CA
+35
| Ct | Fold Change | ||
|---|---|---|---|
| c-ACC | CA | c-ACC vs. CA | |
| CYP11A1 | 16.98 | 18.89 | 5.863 |
| CYP17A1 | 14.44 | 15.92 | 5.239 |
| POR | 13.78 | 14.51 | 3.123 |
| CYB5A | 23.95 | 22.99 | 0.807 |
| HSD17B3 | 22.35 | 31.87 | 1362.567 |
| AKR1C3 | 29.89 | 23.72 | 0.026 |
| SRD5A1 | 39.32 | 41.92 | 9.512 |
| SRD5A2 | 31.75 | 32.68 | 2.981 |
| HSD3B1 | 28.56 | 26.60 | 0.403 |
| HSD3B2 | 26.86 | 18.03 | 0.004 |
| CYP21A2 | 20.96 | 15.92 | 0.047 |
| CYP11B1 | 17.61 | 17.90 | 1.911 |
| SULT2A1 | 19.73 | 21.58 | 6.773 |
| INSL3 | 27.85 | 30.59 | 12.465 |
Figure 3. Representative molecular findings. F (L): father’s leukocyte genomic DNA; M (L): mother’s leukocyte genomic DNA; P (L): patient’s leukocyte genomic DNA; C (L): control male’s leukocyte genomic DNA; c-ACC: patient’s tumor genomic DNA, and CA: patient’s control adrenal genomic DNA. The red and blue star symbols on the microsatellite data indicate decreased and increased peaks in the c-ACCs, respectively. For the aCGH and SNP array findings, see below. A. Germline TP53 p.P337H (c.1010G>A) mutation of maternal origin (indicated by red asterisks) and loss of paternally inherited chromosome 17 with wildtype TP53 from the c-ACC. B. Loss of maternally derived chromosome 11 from the c-ACC. In the upper right figure indicating the chromosome 11p15.5 imprinted regions, paternally and maternally expressed genes are shown in blue and red, respectively; filled and open circles denote methylate and unmethylated CpG dinucleotides, respectively. Deleted alleles of maternal origin are indicated with gray boxes. For methylation analysis of the H19-DMR, a segment encompassing 21 CpG dinucleotides was PCR-amplified with F1 & R1 primers, and a sequence primer (S1) was hybridized to a single-stranded PCR products. Subsequently, the methylation indices (MIs, the ratio of methylated clones) were obtained for four CpG dinucleotides (CG1-CG4) (indicated with a green rectangle). The KvDMR1 was similarly examined using F2 & R2 primers and S2, and the MIs were obtained for CG5-CG10. C. Copy number alterations of chromosomes 4, 19, 20, X, and Y. For chromosome 19p, the relative ratio of the area under curves (AUCs) between the two peaks is: D19S1152, 1.0:0.83 for P (L) and 1.0:1.76 for c-ACC; and D19S256, 1.37:1.0 for P (L) and 2.45:1.0 for c-ACC. The results, together with those of aCGH and SNP array, indicate duplication of paternally derived alleles in virtually all the c-ACC cells. Similar analyses, including the comparison of the AUCs between P (L) and c-ACC, implicate copy number gains of maternally inherited chromosome 20, copy number gains of the Xp, copy number reductions of the Xq, and copy number reductions of the Y chromosome including the short arm pseudoautosomal region (PAR1), in a substantial fraction of the c-ACC cells. Note for the aCGH and SNP array findings In aCGH for autosomes, the black, the red, and the green dots denote signals indicative of the normal, the increased (log2 signal ratio >+0.4), and the decreased (log2 signal ratio ← 0.8) copy numbers, respectively; for sex chromosomes that appear in a heterogametic condition in a male, duplication leads to the
log2 signal ratios of +2.0, and deletion results in the log2 signal ratios of - o (thus, the log2 signal ratios for Xp, Xq, and Y chromosomes in this boy indicate the occurrence of copy number alterations in a substantial fraction of the c-ACC cells rather than in most of the c-ACC cells). In SNP array for autosomes, the dots for log2 signal ratios of “0 or 2” and “1” denote homozygous and heterozygous regions, respectively, and those for log2 signal ratios of “3” and “4” represent regions of increased copy numbers. Thus, the presence of the signals for “0 or 2” and the absence of those for “1” indicate hemizygosity (loss of heterozygosity) or full isodisomy. For the X chromosome in a male, the log2 signal ratio of “0” and “1” denote the hemizygosity for corresponding regions, and that of “3” represent the three copies of the .corresponding regions.
ACCEPTED MANUSCRIPT
A
TP53
Chromosome 17
1.2 1
TP53 (Gene copy)
T
F (L)
mamma
P (L)
1.0
+2.0
1 TP53
VS. ±0
0.8
P (L)
C (L) -2.0
0.6
CA
GTGAGCGCTTCGAG
0,4
c-ACC
M (L)
mmmmm
P (L)
+2.0
0.2
VS. ±0
0
GTGAGCGCTTCGAG
c-ACC-2.0
Exon 4 Exon 10
110 bp
220 bp
P(L)
mhmmm
P (L)
43210
F (L)
1
GTGAGCGCTTCGAG
A
M (L)
VANHO
1
c-ACC
c-ACC
P (L)
GTGAGCACTTCGAG (G)
c-ACC
*
c.1010G>A, p.R337H
D17S831 (17p13.3)
D17S949 (17q24.3)
B
Chromosome 11
CDKN1C (Gene copy)
Pat
IGF2
1.2
H
KCNQ1OT1
P (L) VS.
VCDKN1C
1.0
P (L)
H19
KCNQ1
CDKN1C
+2.0
0.8
CA
Mat<
O
H
±0
C (L) -2.0
0.6
c-ACC
H19-DMR
KVDMR1
0.4
Pat
CO
P (L) +2.0
0.2
0
F2→ S2D
Mat
R1→
AS1 <F1
+R2
VS.
±0
c-ACC-2.0
Exon 1
180 bp
120 bp
(CG1 → CG4)
(CG5 → CG10)
JUNHO
P (L)
F (L)
60
IGF2
CDKN1C
50
1.0
M (L)
11
N
40
30
JUNHO
P (L)
M
20
c-ACC
c-ACC
! *
10
0
D11S2071
D11S988
Exon 2-4
0
(11p15.5) (11p15.4)
Exon 1
Fold change
CA
c-ACC
| CG | 1 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| MI (patient) | |||||||||
| B | 37 40 | 39 | 42 58 58 | 52 | 54 | 58 | 60 | ||
| T | 64 72 | 73 | 75 | 22 | 22 | 20 | 20 | 24 | 22 |
| MI (50 controls) | |||||||||
| Mim | 36 36 | 39 | 37 49 | 52 | 41 | 42 | 55 | 55 | |
| Med | 45 46 | 50 | 48 | 58 | 61 | 48 | 48 | 67 | 64 |
| Max | 55 57 | 64 | 60 | 66 | 68 | 54 | 55 | 72 | 71 |
C
Chromosome 4
Chromosome 19
Chromosome 20
Chromosome X
Chromosome Y
P (L)
+2.0
VS. ±0 C (L) -2.0
P (L)
+2.0
VS.
-0
c-ACC -2.0
P (L)
JUNHO
JUNHO
c-ACC
170 bp
220 bp
240 bp 180 bp
230 bp
320 bp
190 bp
F (L)
11
4
1
1
Mads
M (L)
1
HA
1
P (L)
c-ACC
1
*
*
A
ali
1
*
X
D4S2936 (4p16.3)
D4S2633 (4p15.3)
D19S1152
D19S256
(19p13.12) (19p13.11)
D20S459 (20p12.1)
D20S93 (20p12.1)
DXYS228 (PAR1)