Society for Endocrinology
RESEARCH
A novel 2D and 3D model for primary adrenocortical carcinoma of advanced and metastasized stage co-secreting cortisol, aldosterone, testosterone, 18-oxocortisol and 18-hydroxycortisol
Igor Shapiro1, Chiara Kräuchi1,2, Edlira Luca 01, Susanne Dettwiler3, Mirko Peitzsch4, Weihong Qi5,6, Astrid Reul1, Felix Beuschlein1,7,8, Umberto Maccio3, Svenja Nolting1,7,9, Huguette Debaix1 and Constanze Hantel®1
1Department of Endocrinology, Diabetology and Clinical Nutrition, University Hospital Zurich (USZ) and University of Zurich (UZH), Zürich, Switzerland
2Institute for Pharmaceutical Sciences, ETH Zurich, Zurich, Switzerland
3Department of Pathology and Molecular Pathology, University Hospital Zurich, Zurich, Switzerland
4Institute of Clinical Chemistry and Laboratory Medicine, University Hospital and Faculty of Medicine Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
5Functional Genomics Center Zurich, ETH Zurich and University of Zurich, Zurich, Switzerland 6Swiss Institute of Bioinformatics SIB, Geneva, Switzerland
7Department of Medicine IV, LMU University Hospital, LMU Munich, Munich, Germany
8The LOOP Zurich-Medical Research Center, Zurich, Switzerland
9ENETS Center of Excellence Zurich, University Hospital Zurich, Zurich, Switzerland
Correspondence should be addressed to C Hantel: Constanze.Hantel@usz.ch
Abstract
Adrenocortical carcinoma (ACC) is a highly aggressive malignancy with poor survival rates and few treatment options. Preclinical models are indispensable to further strengthen our understanding of disease progression and development of novel therapeutic treatments. Here, we report the establishment of a new cell line named ZUC-1 originating from the resection of an advanced primary ACC and its characterization at the genomic, cellular and molecular level. ZUC-1 cells were successfully propagated as monolayer cultures and three-dimensional spheroids. LC-MS/MS analysis revealed for ZUC-1 cells co-secretion of cortisol, aldosterone and testosterone, and the model represented in direct comparison with other current ACC pre-clinical models furthermore significantly elevated expression of SF-1, CYP11B1 and CYP11B2 genes. Whole genome sequencing identified various mutations in genes linked to DNA repair/stress response, stemness, and also steroidogenesis. Interestingly, ZUC-1 represents genotypic and phenotypic variations that might be of interest beyond ACC, including congenital adrenal hyperplasia (CAH) and polycystic ovary syndrome (PCOS). Moreover, 18-oxocortisol and 18-hydroxycortisol release was detected in ZUC-1, conditions which are often linked to hyperaldosteronism, but forskolin, potassium and, at higher concentration, angiotensin II modulability of CYP11B2 for this model is retained. ZUC-1 spheroids exhibited furthermore an intra-spheroidal heterogeneous mix of canonical and non-canonical Wnt pathway activation. We conclude that due to its origin and unique geno- and phenotypes, ZUC-1 represents an intriguing model to further gain a basic understanding of adrenal function, the pathogenesis of ACC, but it might be also of interest in the context of CAH and PCOS.
Keywords: adrenocortical carcinoma; tumor spheroids; cell line; NCI-H295; MUC-1; ZUC-1; aldosterone; angiotensin II; cortisol; ACTH; androgens; potassium; beta-catenin; Ki67; vimentin; MMP9; lamin B1; beta-actin; steroids
Introduction
The adrenal glands are triangular-shaped organs, which are located on top of both kidneys, and play a critical role in hormone production and homeostasis (1, 2). Therefore, the adrenal glands and their specific hormones help to regulate essential functions, such as metabolism, immune response and blood pressure (3). Each adrenal gland consists of two main parts: the cortex and the medulla. The cortex, which represents the outer layer of the adrenal glands, is furthermore subdivided into three histologically and functionally distinct zones: the zona glomerulosa, fasciculata and reticularis. Each layer synthesizes and secretes specific steroid hormones (1, 4), including mineralocorticoids (such as aldosterone), glucocorticoids (such as cortisol) and androgens (such as dehydroepiandrosterone (DHEA), estrogens and testosterone). Dysregulation of adrenal hormone synthesis can lead to various endocrine disorders and is also associated with the pathogenesis of adrenal tumors. In general, tumors of the adrenal cortex can be categorized into benign adenomas and the rare, but highly malignant, adrenocortical carcinomas (ACCs) (5, 6). The prognosis for ACCs is poor with a five-year survival rate depending on the disease stage (65-82% for stage I, 58-68% for stage II, 41-55% for stage III and 10-20% for stage IV) (5). Of note, patients of advanced stage frequently present liver, lung and bone metastases at the time point of diagnosis, indicating predominant routes of metastatic dissemination (7, 8).
Due to the rarity, heterogeneity and aggressive nature of ACC, human cell lines are indispensable for the understanding of the pathogenesis of this disease and for the development of novel therapeutic approaches. In this context, there has been a continuing progress in the field with six established human cell lines, which originated either from primary tumors of non- metastasized stages or from ACC metastases derived from lymphatic or hematogenous dissemination (9, 10). However, a model representing advanced primary ACC of already metastasized stage is currently not available and would be of potentially high translational value for appropriate preclinical investigations.
In this study, we describe the establishment, authentication and characterization of ZUC-1, a novel 2D and 3D model derived from a patient with advanced primary ACC who presented at the time point of diagnosis already bone, hepatic and pulmonary metastases. We show that the developed cell line is continuously passageable and demonstrates its own, unique STR-profile matching the original patient tumor. We characterize and compare ZUC-1 with two gold
standard models for primary non-metastatic ACC (NCI-H295R (11, 12)) and drug-resistant ACC metastasis (MUC-1 (12, 13, 14)) regarding various clinically relevant endpoints (H&E, Ki67, SF-1, p53, ß-catenin, mutation analysis, secretome, functional stimulation tests and others) and reveal a highly interesting distinct model- phenotype including cortisol, aldosterone and androgen co-secretion.
Methods
ZUC-1 cell line, 2D culture and 3D tumor spheroids
Generation of patient-derived 2D primary cultures from freshly surgery-derived human ACC and culturing, fixation, paraffin embedding and treatment of NCI-H295R, MUC-1 and ZUC-1 tumor spheroids was done as previously described (14, 15). NCI-H295 R and MUC-1 were overall handled with general split ratios of 1:6-1:8 and 1:4-1:5 per week, respectively. Tumor sample collection was approved by the local ethics committee (Kantonale Ethikkommission Zürich, BASEC 2017-00771), and written informed consent was obtained prior to tumor sampling. In brief, ZUC-1 cells were cultured under the following conditions: cells were grown in Advanced DMEM/F12 medium (containing 10% FBS, 2 mM L-glutamine and 1% penicillin/streptomycin, all from Thermo Fisher Scientific, USA) and split 1:2 every 14-18 days (mean doubling time of 17.2 days). After passage 6, human fibroblast removal step was performed using the Anti-Fibroblast MicroBeads and Positive Selection Columns (both from Miltenyi Biotec, Germany) following the manufacturer’s instructions.
Morphological analysis
To evaluate the morphology of spheroids, sections were stained with hematoxylin and eosin. First, sections were deparaffinized and rehydrated through progressive immersions in xylene for 10 min for 3 times, followed by 3 washes in 100% ethanol for 5 min each. Next, the slides were incubated in 80% ethanol for 5 min and 70% ethanol for 2 min and finally washed in distilled water for 2 min. Staining was initiated by incubating slides in filtered, undiluted Harris’s HTX Hematoxylin (HistoLab®, 01800- EX, Biosystems, Switzerland) for 4 min and thereafter rinsed thoroughly under tap water until the water was clear. The slides were dipped 10 times in a differentiation solution (2.5 mL 37% HCl in 1 L 70% ethanol) and afterward
| Target | Characteristic | Code, company | Dilution |
|---|---|---|---|
| Primary antibodies | |||
| ATRX | Mouse monoclonal AB | Sc-55584, Santa Cruz Biotechnology, INC, USA | 1/100 |
| ß - catenin | Rabbit monoclonal AB | 8480, Cell Signaling Technology®, USA | 1/100 |
| ß - catenin | Mouse monoclonal AB | 610153, BD Transduction Laboratories™M, USA | 1/250 |
| CD44 | Mouse monoclonal AB | 3570, Cell Signaling Technology®, USA | 1/100 |
| CYP11B1 | Rabbit polyclonal AB | CSB-PA591028, Cusabio®, China | 1/25 |
| CYP11B2 | Mouse monoclonal AB | MABS1251, Sigma-Aldrich®, USA | 1/150 |
| Ki67 | Rabbit monoclonal AB | KI68IC0I, DCS Innovative Diagnostik-Systeme, Germany | 1/200 |
| Lamin B1 | Mouse monoclonal AB | Sc-365214, Santa Cruz Biotechnology, INC, USA | 1/100 |
| MC2R | Rabbit polyclonal AB | CSB-PA003204, Cusabio®, China | 1/250 |
| MMP9 | Goat polyclonal AB | AF911, R&D Systems®, USA | 1/100 |
| Nanog | Rabbit monoclonal AB | 4903, Cell Signaling Technology®, USA | 1/200 |
| Nestin | Rabbit monoclonal AB | 73349, Cell Signaling Technology®, USA | 1/800 |
| NR5A1 | Rabbit polyclonal AB | KAL-KO611, Cosmo Bio Co., LTD, Japan | 1/100 |
| P53 | Mouse monoclonal AB | M7001, Agilent Dako, USA | 1/50 |
| SOX2 | Rabbit polyclonal AB | CSB-PA16539A0Rb, Cusabio®, China | 1/150 |
| SOX9 | Rabbit monoclonal AB | CSB-RA202969A0HU, Cusabio®, China | 1/200 |
| Vimentin | Goat polyclonal AB | Abx431705, Abbexa®, UK | 1/100 |
| Wnt5 | Rabbit monoclonal AB | MA5-41244, Invitrogen, USA | 1/50 |
| Secondary antibodies | |||
| Anti-rabbit | Goat IgG Alexa Fluor 488 Conjugated | A11034, Invitrogen, USA | 1/500 |
| Anti-mouse | Goat IgG Alexa Fluor 568 Conjugated | A11004, Invitrogen, USA | 1/500 |
| Anti-mouse | Goat IgG Alexa Fluor 488 Conjugated | A21121, Invitrogen, USA | 1/500 |
| Anti-goat | Donkey IgG Alexa Fluor 488 Conjugated | A11055, Invitrogen, USA | 1/500 |
| Anti-rabbit | Biotinylated goat | BA-1000, Vector laboratories, USA | 1/200 |
| Anti-mouse | Biotinylated goat | A24528, Invitrogen, USA | 1/500 |
rinsed again under running tap water to remove residual differentiation solution. In the next step, slides were incubated in TWS (2 g NaHCO3, 20 g MgSO4 *H2O, in 1L ddH2O) for 4 min followed by an incubation in 95% ethanol for 30 s and counterstaining with filtered, undiluted Eosin Y solution (alcoholic - HT110132-1L, Sigma-Aldrich, USA) for 1 min. Following the staining, the slides were dehydrated stepwise, first incubated in 70% ethanol, 80% ethanol, and 90% ethanol and then in 100% ethanol 3 times, each step lasted 2 min. Finally, the slides were incubated in xylene 3 times for 5 min. For the preservation of the slides, the mounting medium Pertex® (HistoLab®, 00811-EX, Biosystems, Switzerland) was applied to the slides and a cover slip was added. All steps were performed at room temperature, and precautions were taken to ensure consistent hydration throughout the staining process, preventing the slides from drying out. Images were acquired with a Zeiss Axio Imager M2 equipped with the AxioCam 512 color, using ZEN 3.3 software (Carl Zeiss, Germany).
Immunofluorescence and immunohistochemistry
Immunofluorescence
Sections were deparaffinized, boiled in sodium citrate buffer for antigen retrieval, blocked and incubated overnight at 4℃ with specific primary antibodies. Secondary antibodies were applied for 1 h at room
temperature. In the case of multiple stainings, primary antibodies were applied sequentially. Primary and secondary antibodies are listed in Table 1. DAPI (D9542, Sigma-Aldrich, USA) was used at a final concentration of 1 µg/mL. For staining, 100 uL of solution were applied to each section on the slide. For the preservation of the slides, few drops of the aqueous non-fluorescing mounting media Hydromount (HS-106, National Diagnostics, USA) were applied to the slides and a cover slip was added. Pictures were acquired with a Zeiss Axio Observer 7 equipped with the Axiocam 305 mono, using ZEN 10 software (Carl Zeiss, Germany).
Immunohistochemistry
Sections were deparaffinized, boiled in sodium citrate buffer for antigen retrieval, blocked and incubated overnight at 4℃ with primary antibodies. Primary antibodies are listed in Table 1. Next day, sections were incubated with secondary antibodies (Table 1) and ABC solution (Vectastain Elite ABC-HRP Kit, Vector Laboratories, USA). For chromogenic detection, the DAB solution (SK-4100, Vector Laboratories, USA) was used and the reaction was stopped by washing the sections twice in PBS for 5 min and a final 5 min wash in deionized water. Some sections were counterstained with hematoxylin, differentiated, dehydrated and mounted with Pertex as described above. Pictures were acquired with a Zeiss Axio Imager M2 equipped with the AxioCam 512 color, using ZEN 3.3 software (Carl Zeiss, Germany).
Quantitative real-time PCR
Cell lysates were homogenized using the QIAshredder (79656, QIAGEN, Netherlands) prior to RNA purification to ensure uniform lysis. Subsequently, RNA was extracted using the RNeasy®Mini Kit (QIAGEN, Netherlands) according to the instructions given by the manufacturer and treated with the DNase kit (TURBO DNA-free™ Kit, Thermo Fisher Scientific, USA) to remove residual genomic DNA. To analyze the quality and quantity of the RNA, a NanoDrop® spectrophotometer (Thermo Fisher Scientific, USA) was used. Absorbance was acquired at 260 nm, and the A260/A280 ratio was used to assess the purity.
The RNA was reverse transcribed using the RevertAid Reverse Transcription Kit (Thermo Fisher Scientific, USA), and 500 ng of total RNA were used per sample. First, 1 µL of random hexamer primers was added to the RNA and the mixture was incubated at 70℃ for 5 min to denature the secondary structures and allow annealing of the primers. Subsequently, 7.5 uL reverse transcription mastermix were added, containing 5x reverse transcription buffer, RNase inhibitor, 10 mM dNTPs mix and reverse transcriptase enzyme. The reaction was carried out in the Biometra TRIO (Analytik Jena, Germany) with 10 min at 25℃, 60 min at 42℃ and 10 min at 70℃. The cDNA was stored at -20℃ until further use.
The following primers were used for quantitative real-time PCR: human SF-1 (forward: 5’-CAGCCTG-GATTTGAAGTTC CT; reverse: 5’-CAGCATTTCGATGAGCAGGT), human CYP11B1 (forward: 5’-TCCCGAGGGCCTCTAGGA; reverse: 5’-GGGACAAGGTCAGCAAGATCTT) and human CYP11B2 (forward: 5’-AGCTGGGACATTGGTACAGGT; reverse: 5’-GCA TGCCAAAGCCAAAGGG). Gene expression levels were normalized to the housekeeping gene GAPDH (forward: 5’-AGCCTCCCGCTTCGCTCTCT; reverse: 5’-CCAGGCGCCCAA TACGACCA). The quantitative PCRs were carried out in a 96-well plate (4346907, Applied Biosystems™, USA) and performed in the QuantStudio™ 5 Real-Time PCR Instrument and the QuantStudio-TM Design & Analysis Software 1.5.1 (Thermo Fisher Scientific, USA).
STR-authentication
Cell line authentication was done by STR (short tandem repeat) analysis and performed by Microsynth AG (Balgach, Switzerland). Sixteen independent PCR loci (D3S1358, TH01, D21S11, D18S51, Penta_E, D5S818, D13S317, D7S820, D16S539, CSF1PO, Penta_D, AMEL, vWA, D8S1179, TPOX and FGA) were tested and compared to the Cellosaurus database.
Liquid chromatography-tandem mass spectrometry (LC-MS/MS) steroid measurements
Steroid analysis was done as previously described (16).
Whole genome sequencing and mutation analysis
Genomics DNA was quality checked using Qubit Fluorometric Quantification. The Illumina whole genome shotgun (WGS) library was prepared with a total of 100 ng DNA as input, using the Illumina Truseq DNA Nano kit with the Illumina IDT 384 10 nt V1b Barcode Kit. The library was double-sided size selected, and the quality and quantity of the library were validated using the Fragment Analyzer (Agilent, USA). Cluster generation and sequencing were performed on a NovaSeqX System with a run configuration of paired-end (PE) at 2 × 150 bp. In total, 655 million read pairs were collected.
Illumina PE reads underwent quality checks using FastQC (v0.12.1). Adapter sequences and low-quality read ends (identified with a sliding window of 4 bp and a base quality lower than Q20) were trimmed away using Fastp (v0.23.4). Trimmed reads with low average quality (<Q20) and short length (<18 nt) were filtered out using the same tool. Trimmed and filtered reads were mapped to the human reference genome (GRCh38.p14) using bwa-mem (v0.7.18). A list of genes of interest was generated and curated manually. Genomic coordinates of the genes were extracted from the GENCODE Human Release 48 gene annotation. Small variants within genes of interest were identified using Mutec2 in GATK (v4.6.1.0). Multiallelic variant sites were split into biallelic records using bcftools (v1.20). SnpSift (v5.2c) was used to keep only variants with variant allele depth 5 and above and variant allele frequency 5% and above. Variants passing filtering were functionally annotated using SnpEff (v5.2c). Result files in variant call format were converted to tab-delimited files using SnpSift (v5.2c) for downstream analysis and interpretation.
Bioinformatics and statistical analysis
All statistical analysis and graphical representation of the data were conducted using GraphPad Prism software (version 10, GraphPad Software, USA). For statistical comparison, a one-way or two-way ANOVA was performed, followed by Dunnett or Bonferroni’s post hoc test. Results are presented as mean ± SEM. Statistical significance was set at P < 0.05 and indicated by asterisks (*P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001) in all figures, if not stated otherwise. For the quantitative evaluation of the pictures, ImageJ (Version 2.14.0, USA) was used. In particular, the Polygon-Tool was used to measure the areas and the point tool was used to count the positive and negative cells for calculating the Ki67 index.
Results
Clinical characteristics and cell line development
In a first step, we established a primary culture from a surgically resected ACC. The tumor was diagnosed as left adrenal carcinoma with thrombosis of the left adrenal and renal vein in a 41-year-old female patient who furthermore presented bone, hepatic and pulmonary metastases. The patient did not present clinical signs of hormone excess. The primary tumor was resected en bloc, measured approximately 18.5 cm and had a Ki67 index of 20% and macro- and microscopic infiltration of the draining vein with tumor cone formation extending into the renal vein (Fig. 1A, B, C, D, E). The established chemo-naïve primary culture showed since then continuous tumor growth (Fig. 1F; currently passage 21), indicating the establishment of a new cell line.
Next, cell line authentication via genomic DNA of the primary culture in direct comparison with the original patient tumor was done by STR (short tandem repeat) analysis. According to the analysis, summarized in Table 2, the samples positively matched each other. Moreover, there was no detectable overlap with potentially other cell lines according to a comparison with the Cellosaurus database. Based on this successful authentication, the newly established model was named ZUC-1.
Mutation analysis
In a next step, whole genome analysis was performed for ZUC-1 and a list of genes known as general cancer and/or specific ACC drivers, DNA damage repair genes, steroidogenesis genes and a selection of highly relevant stemness markers has been investigated for potential mutagenesis. For completeness, the names of candidates for which mutations were detected are as follows: ADAM21, ANKRD36C, APC, ATM, ATR, ATRX, BLTP2, BRCA1, BRCA2, CD44, CDH23, CFAP251, CHEK1, CREB3L2, CSMD2, CYP19A1, CYP21A2, DLK1, DNAH6, DNHD1, EGFR, EVI2B, FAM170A, FAM20A, FGFR4, HELZ2, HMGB4, HSD17B4, HSD3B2, IGF2R, KCNJ5, KREMEN1, MEN1, MSH3, MUC5B, MUTYH, MYC, MYCL, NANOG, NES, OMG, PCDH12, PPL, SHANK1, SLFN11, SREBF2, TP53, TTN and VSIR. However, the potential functional impact of the detected mutations certainly strongly varies. Thus, a full list of detected mutations (general, potential high and moderate impact), protein effect, codon effect and allele frequency is provided in Supplementary Table 1 (see section on Supplementary materials given at the end of the article), from which selected candidates will be discussed in more detail below.
Morphological assessment of 3D-cultured tumor spheroids
Parallel to culturing in 2D, we applied 3D-culture conditions for ZUC-1 as depicted in Fig. 1G. In a next step, various histological and immunohistochemical stainings, including H&E, Ki67, SF-1 and MC2R, were performed from paraffin-embedded 3D-cultured ZUC-1 cells, which confirmed successful tumor spheroid formation, retainment of adrenocortical marker SF-1 and intense MC2R staining (Fig. 2A, B, C, D, H). Strong nuclear positivity was observed in both SF-1 stainings, showing co-localization of SF-1 and DAPI or SF-1 and hematoxylin, respectively, confirming high rates of nuclear localization. Moreover, ZUC-1 tumor spheroid Ki67 indices, spheroid sizes and diameters were assessed in comparison with spheroids of the already well-established preclinical models NCI-H295R (primary non-metastatic) and MUC-1 (derived from ACC distant metastasis, Fig. 2E, F, G). First, Ki67 indices were calculated for NCI-H295R, MUC-1 and ZUC-1 tumor spheroids (Fig. 2E and G). NCI-H295R exhibited a significantly higher proliferation rate (75.72 + 2.04%) compared to MUC-1 (32.78 ± 1.69%, P < 0.0001, VS NCI-H295R) and ZUC-1 (31.93 ± 1.75%, P < 0.0001, VS NCI-H295R). The Ki67 indices of MUC-1 and ZUC-1 spheroids were assessed as comparable with each other as no statistically significant differences were observed.
Next, our analysis revealed that NCI-H295R spheroids form spheroids of largest sizes (area = 10,514.70 ± 67.61 um2, n = 9; Fig. 2F). In contrast, MUC-1 spheroids were significantly smaller (area = 2,630.00 ± 31.41 um2, P < 0.0001, vs NCI-H295R, n = 32; Fig. 2F) but often appeared to have larger nuclei. The overall area of the newly established ZUC-1 spheroids was significantly smaller than NCI-H295R spheroids (area = 4,694.85 ± 58.01 µm2, P < 0.0001, vs NCI-H295R, n = 17; Fig. 2F) but significantly larger than MUC-1 spheroids (P < 0.05; Fig. 2F). Spheroid diameters are consistent with spheroidal areas: NCI-H295R d = 150.56 ± 5.20 um, MUC-1 d = 66.81 + 2.40 um, and ZUC-1 d = 96.37 ± 2.88 um. For non-spherical spheroids, the average of the shortest and longest diameter was taken. Among the three spheroid types, measurements revealed for NCI-H295R the largest diameters, followed by ZUC-1 and MUC-1, with all differences being statistically significant (Fig. 2G).
Furthermore, p53 and ATRX stainings were performed in tumor spheroids of all three cell lines. As depicted in Fig. 2I, NCI-H295R spheroids showed no clearly detectable p53 positivity. In contrast, both MUC-1 and ZUC-1 demonstrated nuclear staining for p53 (Fig. 2J and K). However, the staining intensities and frequencies of p53-positive nuclei were higher in MUC- 1 compared to ZUC-1. Additional immunofluorescence stainings of ZUC-1 spheroids confirmed nuclear positivity of p53 through co-localization with DAPI and,
A
B
T
F
PO
G
Day 3
500 um
250 um
500 um
P3
Day 6
500 um
250 um
500 um
P5
Day 3
500 um
250 um
P7
Day 6
500 um
250 um
| DNA marker | Chromosomal location | ZUC-1 cell line | Original patient tumor | Database alleles |
|---|---|---|---|---|
| D3S1358 | Chr03 | 17 | 16/17 | N/A |
| TH01* | Chr11 | 6 | 6 | N/A |
| D21S11 | Chr21 | 29/31 | 29/31 | N/A |
| D18S51 | Chr18 | 13/14 | 13/14 | N/A |
| Penta_E | Chr15 | 13/18 | 13/18 | N/A |
| D5S818* | Chr05 | 10/11 | 10/11 | N/A |
| D13S317* | Chr13 | 12 | 11/12 | N/A |
| D7S820* | Chr07 | 10/12 | 10/12 | N/A |
| D16S539* | Chr16 | 11/12 | 11/12 | N/A |
| CSF1PO* | Chr05 | 10/12 | 10/12 | N/A |
| Penta_D | Chr21 | 10/11 | 10/11 | N/A |
| AMEL* | X/Y | X | X | N/A |
| vWA* | Chr12 | 14/17 | 14/17 | N/A |
| D8S1179 | Chr08 | 10/13 | 10/13 | N/A |
| ΤΡΟΧ* | Chr02 | 8/9 | 8/9 | N/A |
| FGA | Chr04 | 20/24 | 20/24 | N/A |
moreover, revealed cytoplasmic staining (data not shown). While we detected furthermore for ZUC-1 spheroids strong nuclear signals of ATRX, our studies revealed comparably a loss of ATRX proteins for MUC-1 and NCI-H295R under the same conditions (Fig. 2L, M, N).
Next, double-stainings of ß-catenin and lamin B1 and B-catenin and WNT5A were performed, respectively. Immunofluorescence stainings revealed a mix of cytoplasmic distribution, plasma membranous localization and nuclear accumulation for ß-catenin, as illustrated in Fig. 3A. Lamin B1, a nuclear lamina marker, showed a precise pattern surrounding the DAPI-stained nucleus, consistent with its known localization at the nuclear membrane. WNT5A demonstrated high protein levels in ZUC-1 with some diffuse basal cytoplasmic staining but also often co-localized with ß-catenin and intense accumulation hotspots (Fig. 3B).
Further cytoskeletal and extracellular matrix-related markers were assessed next (Fig. 4). Since the matrix metalloproteinase 9 plays an important role in potential ECM degradation during metastasization, this marker was first investigated in ZUC-1 spheroids that were obtained from a primary ACC of already metastasized stage. The immunofluorescence staining for MMP9 generally revealed a weak signal throughout the spheroid with distinct regions of elevated intensities often in the spheroid periphery or spheroid core. Vimentin, a key intermediate filament protein, which is known to promote cancer metastasis by regulating cell motility and epithelial-mesenchymal transition (EMT), showed cytoplasmic staining again often with intense hotspot regions at the spheroid center and periphery. Nestin, an intermediate filament of type VI and also an
important stem cell marker, which is furthermore known to participate in various pathophysiological processes, including angiogenesis and EMT, was also distinctly detectable. To further follow up on stemness potential in tumor spheroids of ZUC-1, the occurrence and cellular localization of further key stem cell markers were investigated (Fig. 4). These markers are commonly associated with self-renewal capacity and stemness properties. Both SOX2 and SOX9 showed broad stainings throughout the spheroids. Immunofluorescence stainings revealed furthermore also the occurrence of the highly relevant stem cell marker Nanog. Nanog protein appeared to be localized to distinct regions within the spheroid, but no co-localization with DAPI was observed, indicating non-nuclear localization. For CD44, immunofluorescence analysis of ZUC-1 spheroids revealed a heterogeneous distribution, with hotspots in some distinct spheroidal regions likely at the plasma membrane. No co-localization of DAPI and CD44 is observed, indicating extra-nuclear localization.
Investigation of steroidogenic characteristics
Finally, the steroidogenic properties of the new cell line ZUC-1 were investigated. LC-MS/MS measurements of passages 1-7 revealed continuous endocrine functionality, including cortisol, aldosterone, 18-oxocortisol, 18-hydroxycortisol and androgen co-secretion (Fig. 5).
Next, gene expression, protein occurrence and localization of CYP11B1 and CYP11B2 were examined for ZUC-1 in direct comparison with NCI-H295R and MUC-1 (Fig. 6). As expected, both CYP11B1 and CYP11B2 proteins were detectable by immunofluorescence in NCI- H295R tumor spheroids and revealed a diffuse staining pattern with often increased cytoplasmic intensities (Fig. 6A and B). CYP11B1 localized more weakly also in MUC-1 in the cytoplasm, with some hotspots in the periphery of the spheroids. ZUC-1 showed a CYP11B1 pattern comparable to MUC-1 but with hotspots of higher intensity. NCI-H295R displayed for CYP11B2 a punctuated pattern in the cytoplasm, potentially associated with endomembranous compartments, comparable to that observed but weaker than in ZUC-1 spheroids. A punctuated pattern of CYP11B2 was also detected in MUC-1 spheroids, although with lower intensity compared to ZUC-1 and NCI-H295R (Fig. 6B).
In a next step SF-1, CYP11B1 and CYP11B2 gene expression was quantified for the cell lines NCI-H295R, MUC-1 and ZUC-1 and comparably high expression of CYP11B1 and CYP11B2 was observed in ZUC-1 relative to the current functional gold standard model NCI-H295R (Fig. 6C, D, E). As expected, lowest SF-1 expression and no gene expression of CYP11B1 and CYP11B2 were detectable under these conditions for MUC-1.
A
B
&
D
50 μm
5Q um
50 μm
50 um
50 μm
SQ, um
20 Jim
20 jam
£
31 μm
31 pm
31 pm
31 μm
31 um
20 Jim
20 pm
E
F
Spheroid size
G
Spheroid diameter
…
H
100
…
ns
15000
200
Ki-67 index (%)
80-
150
60-
area (um2)
10000-
diameter (um)
40
100
5000-
20-
50
25
ME
0
0
0
NCI-H295R
MUC-1
ZUC-1
NCI-H295R
MUC-1
ZUC-1
NCI-H295R
MUC-1
ZUC-1
J
K
NCI-H295R control
p53
MUC-1 control
Es
p53
ZUC-1 control
p53
?
50 μm
50 um
50 μm
50 μm
50 um
50 μm
31 um
31 um
31 um
31 um
31 μm
31 μm
INCI-H295R Merge
M
MUC-1 Merge
N
ZUC-1 Merge
20 um
20 pm
20 pm
20 um
20 pm
20 um
ATRX
ATRX
ATRX
20 pm
20 pm
20 pm
20 um
20 pm
20 um
DAPI
DAPI
DARI
20 pm
20 pm
20 pm
20 pm
20 pm
20 pm
| A Merge 200 um 100 um 50 um 20 um 20 um 20 um | Beta-Catenin 200 um 100 um | Lamin B1 200 um | DAPI 200 um 100 um |
|---|---|---|---|
| 100 um | |||
| 50 um | 50 um | 50 um 20 um 20 um | |
| 20 um | 20 um 20 um | ||
| 20 um | |||
| 20 um | 20 um | 20 um | |
| B Merge 20 pm | Wnt5A 20 um | Beta-Catenin 20 um | DAPI 20 um |
| 20 um | 20 um | 20 um | 20 um |
| 20 um | 20 um | 20 um | 20 um |
Figure 3 ß-catenin/lamin B1 (A) and Wnt5A/B-catenin (B) immunofluorescence co-stainings in different magnifications from low to high in ZUC-1 tumor spheroids. A full color version of this figure is available at https://doi.org/10.1530/ERC-25-0385.
Finally, stimulation experiments were performed for ZUC-1 applying potassium, ACTH, angiotensin II and forskolin as potential modulators of steroidogenesis (Fig. 6F, G, H, I, J, K, L, M). Subsequent investigation of CYP11B1 and CYP11B2 gene expression revealed highest
induction of CYP11B1 under forskolin (Fig. 6F) and of CYP11B2 under potassium (Fig. 6G) stimulation. Moreover, significant induction of CYP11B1 was also detectable upon potassium (Fig. 6H) and of CYP11B2 upon forskolin (Fig. 6I) and angiotensin II (Fig. 6J)
Merge
MMP9
DAPI
Merge
Vimentin
DAPI
20 pm
20 pm
20 pm
20 pm
20 um
20 um
20 um
20 um
20 um
20 um
20 pm
20 um
Nestins
SOX2
SOX9
Nanog
CD44
20 um
20 um
20 um
20 um
20 um
stimulation. In contrast, ACTH lacked in our experiments stimulatory potency on the expression of both enzymes in ZUC-1 and demonstrated rather some inhibitory effects on CYP11B1 in comparison with solvent controls. However, further experiments will be required to confirm and further characterize this observation.
Discussion
Adrenocortical carcinoma is a rare and, particularly at advanced and metastasized stages, highly aggressive disease with poor prognosis. One potential reason for the high failure rate of translationally developed systemic regimes in the past was presumably a lack of clinically relevant models that reflect ACC patients’ collective heterogeneity, including appropriate disease stages. Fortunately, since 2016, there is strong progress in the field, which led to the establishment of a panel of various human cell lines of primary and metastatic origins. Such cell lines often closely reflect the clinical characteristics of the original patient tumors they have been originally obtained from, including either a lack of spontaneous metastasization such as for NCI-H295R (obtained from a primary ACC with no clinical evidence of metastatic disease at the time point of resection) or regional (TVBF-7, local lymph node metastasis) and distant (MUC-1, neck metastasis) dissemination (9). Predominant metastatic routes from advanced ACC stages include liver, lung and bone metastases (7, 8). However, until today, only primary ACC-derived cell lines prior to potential dissemination
could be reported, which, thus, likely represent earlier and more benign ACC stages. Here, we report on the establishment of ZUC-1, a novel adrenocortical cell line derived from an advanced ACC after clinically confirmed dissemination to bone, liver and lung. In this study, we confirm that the cells can be continuously passaged that they demonstrate their own unique STR-profile not overlapping with already existing cells lines and that this profile furthermore matches the original patient tumor.
As 3D-organization of tumors and their respective preclinical models is furthermore an essential and highly emerging aspect, we have established in parallel ZUC-1 tumor spheroids and compared these with 3D- models of NCI-H295R and MUC-1. Morphological assessment showed that NCI-H295R tumor spheroids are significantly larger than those formed by MUC-1 and ZUC-1 and that MUC-1 and ZUC-1 show comparable Ki67 indices significantly lower than those detected for NCI-H295R spheroids. Notably, the spheroids from the newly established cell line ZUC-1 appear to be in size, diameter and shape more comparable to MUC-1 spheroids. Recently, Feely et al. demonstrated that NCI- H295R spheroids develop hypoxic cores during ongoing culturing, whereas this phenomenon was not demonstrated for MUC-1 spheroids under the same culture conditions (17). This observation is in agreement with our presented results and studies suggesting that spheroids with a diameter smaller than 200 um typically develop no necrosis and hypoxia in the core (18, 19). Similar effects were previously observed in primary solid tumors, where it could be demonstrated
Aldosterone
18-oxocortisol
18-hydroxycortisol
DHEA
Androstenedione
11-Deoxycortisol
0.15
1.5
80
1.0-
15
60
0.8
ng/ml
0.10
ng/ml
1.0
60
ng/ml
ng/ml
0.6
10
40
40
ng/ml
ng/ml
0.05
0.5
0.4
20
5
20
0.2
0.00
0.0
0
0.0
0
0
A
18oxoF
18OHF
DHEA
A4
S
Cortisol
11-Deoxycorticosterone
Corticosterone
Pregnenolone
Progesterone
17OH-Progesterone
2000
3
80
6
0.4
2.5
1500
60
0.3
2.0
ng/ml
2
4
1.5
1000
ng/ml
ng/ml
40
ng/ml
ng/ml
0.2
ng/ml
1
2
1.0
500
20
0.1
0.5
0
0
0
0
0.0
0.0
F
DOC
B
P5
P4
17OHP
21-Deoxycortisol
Cortisone
11Dehydrocorticosterone
DHEAS
Testosterone
25
150
0.8
250
0.15
20
0.6
200
15
100
ng/ml
ng/ml
ng/ml
ng/ml
150
0.4
ng/ml
0.10
10
50
100
0.05
5
0.2
50
0
0
0.0
0
0.00
21DF
E
11DHC
DHEAS
T
that larger primary tumor volume is associated with significantly higher hypoxic volume and higher hypoxic fractions compared to smaller tumors (20). Our studies indicate for primary, non-metastatic-derived NCI-H295R continuously increasing spheroid sizes, while both models of metastatic nature, MUC-1 and ZUC-1, rather appear to actively regulate spheroid sizes and form instead higher numbers of spheroids of highly distinct, but significantly smaller sizes. These differences might reflect specific cellular plasticities and metastatic nature, as smaller sizes and low proliferation rates might be advantageous during metastatic niche-formation to evade from the immune system (21).
TP53, a potential tumor suppressor gene, is often dysregulated in patients with cancer, including ACC. Most of the already established and characterized ACC cell lines carry TP53 mutations. In particular, a large deletion in the TP53 gene in NCI-H295R cells and a somatic deletion and frameshift mutation in MUC-1 were previously reported (9). Our immunohistochemistry stainings on 3D-cultured tumor spheroids revealed no specific immunopositivity for
NCI-H295R, likely due to the reported large deletion in the TP53 gene, resulting in a loss of gene expression. In contrast, strong nuclear positivity was detected for MUC-1. Both findings are in line with the results previously reported by us for NCI-H295R and MUC-1 xenografts (14). For the newly established ZUC-1 model, we identified in this study a TP53 missense variant (p.Pro72Arg, c.215C>G, allele frequency 0.984). Of note, this TP53 rs1042522 SNP is associated with susceptibility to several malignancies but shows overall contradictory results in terms of general cancer risk (22, 23). ZUC-1 tumor spheroids presented furthermore in immunohistochemical stainings p53 nuclear positivity along with diffuse cytoplasmic staining. Of note, wild- type p53 underlies dynamic shuttling between nuclei and is in unstressed cells rather weakly localized to both (24). DNA damage and other stress result in general p53 stabilization and nuclear accumulation, but of course also TP53 mutations can influence its subcellular localization. In this context, it should be mentioned that the DNA damage response pathway involves the recruitment of ATM (ataxia telangiectasia mutated) and
A
B
NCI-H295R
Merge
CYP11B1
DAPI
Merge
CYP11B2
DAPI
20 um
20 um
20 um
20 mm
20 um
20 pam
MUC-1
20 pm
20 um
20 pm
20 Mm
20 Pm
20 mm
ZUC-1
20 pm
20 am
20 um
20 Mm
30 Mm
20 Mm
C
D
E
4
150
0030
SF-1 / GAPDH [arbitrary unit]
CYP11B1 / GAPDH [arbitrary unit]
3
100
CYP11B2 / GAPDH [arbitrary unit]
0025
0020
50
2
0015
0.0010
0.4
0.0008
1
0.0006
0.2
0.0004
0.0002
0
0.0
0.0000
NCI-H295R
MUC-1
ZUC-1
NCI-H295R
MUC-1
ZUC-1
NCI-H295R
MUC-1
ZUC-1
F
CYP11B1
G
CYP11B2
CYP11B1 / GAPDH (% of SC)
25000
CYP11B2 / GAPDH (% of SC)
15000
5000
10000
1000
5000
500
400
200
0
0
SC
K+ 5mM
K+ 10mM
K+ 20mM
Fsk 5M
Fsk 10UM
Fsk 25uM
ACTH 5nM
ACTH 15nM
ACTH 25nM
Angll 20nM
Angli 100nM
Angli 250nM
SC
K+ 5mM
K+ 10mM
K+ 20mM
Fsk 5uM
Fsk 10uM
Fsk 25uM
ACTH 5nM
ACTH 15nM
ACTH 25nM
Angli 20nM
Angli 100nM
Angli 250nM
I
J
CYP11B1 / GAPDH (% of SC)
CYP11B1
CYP11B2 / GAPDH (% of SC)
CYP11B2
CYP11B2 / GAPDH (% of SC)
CYP11B2
1000
500
200-
800
400
150
600
300
100
400
200
200
100
50
0
0
0
SC
K+ 5mM
K+ 10mM
K+ 20mM
SC
Fsk 5uM
Fsk 10uM
Fsk 25uM
SC
Angli 20nM
Angli 100nM
Angli 250nM
that the key downstream target of ATM is p53. Interestingly, it has recently been demonstrated that even single synonymous mutations have the potential to alter the p53 mRNA secondary structure in a way that essential interactions between ATM and p53 are altered (25, 26). In parallel to the reported TP53 mutation, our analyses revealed also ATM mutations. Among them are c.1236-3dupT and Ser707Pro. Variant c.1236-3dupT is classified as a likely benign ATM variant. However, while multifactorially adjusted hazard ratios for ATM Ser707Pro comparing heterozygotes versus noncarriers revealed for cancer overall a value of 0.8 (95% CI: 0.6-1.2), the same study revealed a value of 10 (95% CI: 1.1-93) for thyroid and other endocrine tumors, indicating a rather potentially strong influence for endocrine tumors (27).
The known BRCA1 variant c.2612C>T (p.Pro871Leu) leads to an amino acid substitution in a region where recombinase RAD51, a multifunctional protein with a central role in DNA replication and homologous recombination repair, interacts with BRCA1. Published studies provide contradictory reports on the single role of this variation in different cancers (28, 29). However, in recent years, not only effects of high-penetrance mutations but also cumulative effects of single nucleotide variations have been assessed. Among them is a study on breast cancer risk involving 25 variants of previously defined low penetrance but of potential functionality in BRCA1, BRCA2, ATM, TP53 and CHEK2. Four of these 25 exemplary variants reported to represent independently minor, but cumulatively significant, increased risk of breast cancer were also among those identified in ZUC-1 (BRCA1: p.Lys1183Arg and p.Ser1613Gly, BRCA2: p.Asn372His, ATM: p.Ser707Pro and TP53: p.Pro72Arg (30)). Four mutations were furthermore identified in the MSH3 gene that encodes a protein critical for DNA mismatch repair and is thereby involved in maintaining genome stability.
For ATRX, we detected the rather likely benign SNP c.2785G>C plus ATRX positivity in immunofluorescence stainings of ZUC-1 spheroids. In contrast, for the two other models, we observed ATRX loss. An ATRX mutation has been previously reported for MUC-1 but not for NCI- H295R (31).
Furthermore, we identified a KCNJ5 mutation in ZUC-1. Human KCNJ5 is mainly expressed in the adrenal gland but is also detectable in other organs (https://gtexportal. org/home/gene/KCNJ5). Mutations of KCNJ5, often within the selectivity filter protein domain, are known to disrupt the ion channel’s normal function and are linked to primary hyperaldosteronism, particularly in aldosterone-producing adenomas (32). From the current point of view, in ZUC-1, the observed variant is rather considered to be benign. However, although this variant does not localize close to the selectivity filter, interestingly, the same genetic variation has been previously reported in the context of cardiovascular
diseases, such as sinoatrial node dysfunction and LongQT syndrome (33) (https://www.ncbi.nlm.nih.gov/ clinvar/RCV001730572/; https://www.ncbi.nlm.nih.gov/ clinvar/RCV000622178/). In this context, it should be also mentioned that we detected in parallel 18-oxocortisol and 18-hydroxycortisol in ZUC-1 cell culture supernatants. The secretion of both steroids is known to be very low in healthy subjects but linked to familial hyperaldosteronism types 1 (glucocorticoid-remediable aldosteronism) and 3 (caused by KCNJ5 germline mutations) (34). However, a stimulatory effect of ACTH on both, CYP11B1 and CYP11B2, would have been expected for conditions related to familial hyperaldosteronism type 1. Overall, further functional analysis, including electrophysiology, will be required to bring further insights.
ANKRD36C, a protein yet functionally uncharacterized in humans, has recently been demonstrated to be a novel genomic biomarker for the progression of chronic myeloid leukemia specifically in patients with advanced disease stages and associated with early disease progression (35, 36). Moreover, studies indicate a role in cancer, association with poor prognosis and ANKRD36 mutations being of ‘likely pathogenic’ status (36, 37, 38, 39). ANKRD36 may be related to inflammation, and ANKRD36C is, interestingly, predicted to enable ion channel inhibitory activity (40, 41, 42) (https://www.ncbi. nlm.nih.gov/gene/400986). Potential ion channel interaction partners are, however, yet unknown. Our studies reveal for ZUC-1 various mutations, including some of very high impact.
HSD3B2 is critical for the production of all major steroid hormones, and germline mutations in the HSD3B2 locus lead to a rare, but severe form of congenital adrenal hyperplasia (CAH) called classical 3B-HSD2 deficiency. Under these conditions, patients experience decreased production of cortisol and aldosterone and increased levels of DHEA, ACTH, androstenedione and testosterone. For ZUC-1, we detected a HSD3B2 c.281_282delAG mutation, a deletion frameshift variant with potentially high functional impact and a variation allele frequency of 0.221, indicating that only a small fraction of the cells is affected. Hypothesizing based on a partial 3B-HSD2 deficiency for ZUC-1 and that mutationally affected cells might produce consequently increasing amounts of ACTH could lead to continuous ACTH stimulation of the remaining other fraction. This could explain the observed overall extraordinary high levels of cortisol detected, strong MC2R-receptor immunostaining and lack of ACTH- inducibility of CYP11B1 gene expression, while potassium and forskolin stimulabilities for CYP11B1 are retained. The enzyme CYP21A2 is known in the context of the classical form of CAH, and the detected CYP21A2 stop gained mutation could, thus, further contribute to this effect. However, more detailed studies, which were out of the scope of a first cell line description, will be required to clarify the underlying mechanisms for ZUC-1.
For completeness, the detected common missense polymorphism of CYP19A1 (p.Arg264Cys, rs710059) was previously demonstrated to be associated with susceptibility to polycystic ovary syndrome and a potential increase in aromatase enzymatic activity (43). The detected SBREF-2 polymorphism was already previously associated with non-alcoholic fatty liver (44), atherosclerosis (45) and the development of hyperlipidemia (46).
Cancer stem cells are a subpopulation of cancer cells that exhibit properties of normal stem cells, including self-renewal and pluripotency. To investigate the stem- like properties of ZUC-1, the expression of relevant stem cell markers was assessed. High nuclear expression of KLF4 and strong signals for Nestin and Nanog were observed, SOX2 and SOX9 showed diffuse patterns with regions of higher intensity, and CD44 expression was weak with focal hotpots. Similarly, we have recently reported varying patterns of the same stemness markers for NCI-H295R, TVBF-7 and MUC-1 (47). Such cancer stem cell characteristics may contribute to variations in aggressive behavior in the same way as the investigated candidates Wnt5A, vimentin or MMP9, which were linked to cytoskeletal remodeling and invasive potential in 3D- models of ACC (Experimental and Molecular Medicine, 2025, in press). Although ZUC-1 originates from a primary ACC, the patient had already developed metastases, indicating that ZUC-1 cells may have acquired and retained the molecular machinery for metastatic behavior. The immunofluorescent stainings of the previously mentioned markers support this hypothesis based on our recent findings (Experimental and Molecular Medicine, 2025, in press), even though this is of course only a first step of investigation and further studies will be needed to reveal the full potential of the newly developed ZUC-1 model.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/ERC-25-0385.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.
Funding
This research was supported by Swiss 3RCC (CH).
Acknowledgments
Whole genome sequencing was performed at the Functional Genomics Center Zurich (FGCZ) of University of Zurich and ETH Zurich.
References
1 Chu Y, Setayesh J, Dumontet T, et al. Adrenocortical stem cells in health and disease. Nat Rev Endocrinol 2025 21 464-481. (https://doi.org/10.1038/s41574-025-01091-2)
2 Miller WL & White PC. History of adrenal research: from ancient anatomy to contemporary molecular biology. Endocr Rev 2023 44 70-116. (https://doi.org/10.1210/endrev/bnac019)
3 Kanczkowski W, Sue M & Bornstein SR. Adrenal gland microenvironment and its involvement in the regulation of stress- induced hormone secretion during sepsis. Front Endocrinol 2016 7 156. (https://doi.org/10.3389/fendo.2016.00156)
4 Lyraki R & Schedl A. Adrenal cortex renewal in health and disease. Nat Rev Endocrinol 2021 17 421-434. (https://doi.org/10.1038/s41574-021-00491-4)
5 Fassnacht M, Puglisi S, Kimpel O, et al. Adrenocortical carcinoma: a practical guide for clinicians. Lancet Diabetes Endocrinol 2025 13 438-452. (https://doi.org/10.1016/s2213-8587(24)00378-4)
6 Mete O, Erickson LA, Juhlin CC, et al. Overview of the 2022 WHO classification of adrenal cortical tumors. Endocr Pathol 2022 33 155-196. (https://doi.org/10.1007/s12022-022-09710-8)
7 Libe R, Borget I, Ronchi CL, et al. Prognostic factors in stage III-IV adrenocortical carcinomas (ACC): an European Network for the Study of Adrenal Tumor (ENSAT) study. Ann Onco/ 2015 26 2119-2125. (https://doi.org/10.1093/annonc/mdv329)
8 Owen DH, Patel S, Wei L, et al. Metastatic adrenocortical carcinoma: a single institutional experience. Horm Cancer 2019 10 161-167. (https://doi.org/10.1007/s12672-019-00367-0)
9 Luca E, Abate A, Wang K, et al. Human and murine cell lines for adrenocortical carcinoma and pheochromocytoma. Endocrines 2024 5 261-276. (https://doi.org/10.3390/endocrines5030019)
10 Sigala S, Rossini E, Abate A, et al. An update on adrenocortical cell lines of human origin. Endocrine 2022 77 432-437. (https://doi.org/10.1007/s12020-022-03112-w)
11 Gazdar AF, Oie HK, Shackleton CH, et al. Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res 1990 50 5488-5496.
12 Sigala S, Bothou C, Penton D, et al. A comprehensive investigation of steroidogenic signaling in classical and new experimental cell models of adrenocortical carcinoma. Cells 2022 11 1439. (https://doi.org/10.3390/cells11091439)
13 Bothou C, Sharma A, Oo A, et al. Novel insights into the molecular regulation of ribonucleotide reductase in adrenocortical carcinoma treatment. Cancers 2021 13 4200. (https://doi.org/10.3390/cancers13164200)
14 Hantel C, Shapiro I, Poli G, et al. Targeting heterogeneity of adrenocortical carcinoma: evaluation and extension of preclinical tumor models to improve clinical translation. Oncotarget 2016 7 79292-79304. (https://doi.org/10.18632/oncotarget.12685)
15 Bornstein S, Shapiro I, Malyukov M, et al. Innovative multidimensional models in a high-throughput-format for different cell types of endocrine origin. Cell Death Dis 2022 13 648. (https://doi.org/10.1038/s41419-022-05096-x)
16 Peitzsch M, Dekkers T, Haase M, et al. An LC-MS/MS method for steroid profiling during adrenal venous sampling for investigation of primary aldosteronism. J Steroid Biochem Mol Biol 2015 145 75-84. (https://doi.org/10.1016/j.jsbmb.2014.10.006)
17 Feely S, Mullen N, Donlon PT, et al. Development and characterization of 3-dimensional cell culture models of adrenocortical carcinoma. Endocrinology 2024 166 bqae159. (https://doi.org/10.1210/endocr/bqae159)
18 Daster S, Amatruda N, Calabrese D, et al. Induction of hypoxia and necrosis in multicellular tumor spheroids is associated with resistance
to chemotherapy treatment. Oncotarget 2017 8 1725-1736. (https://doi.org/10.18632/oncotarget.13857)
19 Sharifi F, Firoozabadi B & Firoozbakhsh K. Numerical investigations of hepatic spheroids metabolic reactions in a perfusion bioreactor. Front Bioeng Biotechno/ 2019 7 221. (https://doi.org/10.3389/fbioe.2019.00221)
20 Hildingsson S, Gebre-Medhin M, Zschaeck S, et al. Hypoxia in relationship to tumor volume using hypoxia PET-imaging in head & neck cancer - a scoping review. Clin Transl Radiat Oncol 2022 36 40-46. (https://doi.org/10.1016/j.ctro.2022.06.004)
21 Jehanno C, Vulin M, Richina V, et al. Phenotypic plasticity during metastatic colonization. Trends Cell Biol 2022 32 854-867. (https://doi.org/10.1016/j.tcb.2022.03.007)
22 Dong Z, Zheng L, Liu W, et al. Association of mRNA expression of TP53 and the TP53 codon 72 Arg/Pro gene polymorphism with colorectal cancer risk in Asian population: a bioinformatics analysis and meta- analysis. Cancer Manag Res 2018 10 1341-1349. (https://doi.org/10.2147/cmar.s164892)
23 Yasin M, Holland SJ, Jolly AM, et al. Polymers for biodegradable medical devices. VI. Hydroxybutyrate-hydroxyvalerate copolymers: accelerated degradation of blends with polysaccharides. Biomaterials 1989 10 400-412. (https://doi.org/10.1016/0142-9612(89)90132-4)
24 Maki CG. p53 Localization. In p53, pp 117-126. Boston, MA, USA: Springer, 2010. (https://link.springer.com/chapter/10.1007/978-1-4419-8231-5_8)
25 Karakostis K, Malbert-Colas L, Thermou A, et al. The DNA damage sensor ATM kinase interacts with the p53 mRNA and guides the DNA damage response pathway. Mol Cancer 2024 23 21. (https://doi.org/10.1186/s12943-024-01933-z)
26 Karakostis K, Vadivel Gnanasundram S, Lopez I, et al. A single synonymous mutation determines the phosphorylation and stability of the nascent protein. J Mol Cell Biol 2019 11 187-199. (https://doi.org/10.1093/jmcb/mjy049)
27 Dombernowsky SL, Weischer M, Allin KH, et al. Risk of cancer by ATM missense mutations in the general population. J Clin Oncol 2008 26 3057-3062. (https://doi.org/10.1200/jco.2007.14.6613)
28 Kour A, Sambyal V, Guleria K, et al. Screening of BRCA1 variants c.190T>C, 1307delT, g.5331G>A and c.2612C>T in breast cancer patients from North India. Genet Mol Biol 2020 43 e20190014. (https://doi.org/10.1590/1678-4685-gmb-2019-0014)
29 Miao L, Yu Y, Ji Y, et al. Association between BRCA1 P871L polymorphism and cancer risk: evidence from a meta-analysis. Oncotarget 2017 8 30587-30594. (https://doi.org/10.18632/oncotarget.15739)
30 Johnson N, Fletcher O, Palles C, et al. Counting potentially functional variants in BRCA1, BRCA2 and ATM predicts breast cancer susceptibility. Hum Mol Genet 2007 16 1051-1057. (https://doi.org/10.1093/hmg/ddm050)
31 Arakawa Y, Elloumi F, Varma S, et al. A database tool integrating genomic and pharmacologic data from adrenocortical carcinoma cell lines, PDX, and patient samples. Cancer Res Commun 2024 4 2384-2398. (https://doi.org/10.1158/2767-9764.crc-24-0100)
32 Choi M, Scholl UI, Yue P, et al. K+ channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension. Science 2011 331 768-772. (https://doi.org/10.1126/science.1198785)
33 Meyer KM, Malhotra N, Kwak JS, et al. Relevance of KCNJ5 in pathologies of heart disease. Int J Mol Sci 2023 24 10849. (https://doi.org/10.3390/ijms241310849)
34 Lenders JWM, Williams TA, Reincke M, et al. Diagnosis of endocrine disease: 18-oxocortisol and 18-hydroxycortisol: is there clinical utility of these steroids? Eur J Endocrinol 2018 178 R1-R9. (https://doi.org/10.1530/eje-17-0563)
35 Absar M, Alanazi N, Siyal A, et al. Clinical validation of Ankrd36 mutations as a novel biomarker for monitoring early progression and timely clinical interventions in blast crisis cml. J Popul Ther Clin Pharmacol 2022 29 311-320. (https://doi.org/10.53555/jptcp.v29i02.4161)
36 Iqbal Z, Absar M, Akhtar T, et al. Integrated genomic analysis identifies ANKRD36 gene as a novel and common biomarker of disease progression in chronic myeloid leukemia. Biology 2021 10 1182. (https://doi.org/10.3390/biology10111182)
37 Cong T, Liu GX, Cui JX, et al. [Exome sequencing of gastric cancers screened the differences of clinicopathological phenotypes between the mutant and the wide-type of frequently mutated genes]. Zhonghua Yi Xue Za Zhi 2018 98 2242-2245. (https://doi.org/10.3760/cma.j.issn.0376-2491.2018.28.006)
38 Perilli L, Tessarollo S, Albertoni L, et al. Silencing of miR-182 is associated with modulation of tumorigenesis through apoptosis induction in an experimental model of colorectal cancer. BMC Cancer 2019 19 821. (https://doi.org/10.1186/s12885-019-5982-9)
39 Yamada Y, Arai T, Kojima S, et al. Regulation of antitumor miR-144-5p targets oncogenes: direct regulation of syndecan-3 and its clinical significance. Cancer Sci 2018 109 2919-2936. (https://doi.org/10.1111/cas.13722)
40 Basu MK, Massicano F, Yu L, et al. Exome sequencing identifies abnormalities in glycosylation and ANKRD36C in patients with immune-mediated thrombotic thrombocytopeniaurpura. Thromb Haemost 2021 121 506-517. (https://doi.org/10.1055/s-0040-1719030)
41 Fang Y, Wang X, Li W, et al. Screening of circular RNAs and validation of circANKRD36 associated with inflammation in patients with type 2 diabetes mellitus. Int J Mol Med 2018 42 1865-1874. (https://doi.org/10.3892/ijmm.2018.3783)
42 Wang Q, Tao S, Zhu N, et al. Silencing circular RNA circANKRD36 remits lipopolysaccharide-induced inflammatory damage by regulating microRNA-15/MyD88. J Cell Biochem 2020 121 2704-2712. (https://doi.org/10.1002/jcb.29490)
43 Wang H, Li Q, Wang T, et al. A common polymorphism in the human aromatase gene alters the risk for polycystic ovary syndrome and modifies aromatase activity in vitro. Mol Hum Reprod 2011 17 386-391. (https://doi.org/10.1093/molehr/gar046)
44 Bhatt SP, Nigam P, Misra A, et al. SREBP-2 1784 G/C genotype is associated with non-alcoholic fatty liver disease in north Indians. Dis Markers 2011 31 371-377. (https://doi.org/10.1155/2011/950102)
45 Robinet P, Vedie B, Chironi G, et al. Characterization of polymorphic structure of SREBP-2 gene: role in atherosclerosis. Atherosclerosis 2003 168 381-387. (https://doi.org/10.1016/s0021-9150(03)00144-8)
46 6 Duan X, Zhu W, Li Y, et al. The effect of sterol regulatory element- binding protein 2 polymorphism on the serum lipid in northern Chinese subjects. J Lipid Res 2005 46 252-257. (https://doi.org/10.1194/jlr.m400166-jlr200)
47 Shapiro I, Debaix H, Kräuchi C, et al. (Non)canonical Wnt signaling, cytoarchitecture and stemness: new insights from primary nonmetastatic, primary metastatic, regional and distant metastatic models of adrenocortical carcinoma. Exp Mol Med 2025 57 2657-2670. (https://doi.org/10.1038/s12276-025-01507-z)