Heterozygous Nes-GFP transgenic mice [16] on the C57BL/6N genetic background were generated as described previously [14]. All mouse colonies were maintained under a 12:12 h light/dark cycle and fed ad libitum. Mice were sacrificed by cervical dislocation and all animal experiments were approved according to the German Animal Welfare Act by Landesdirektion Sachsen, Germany.

Adrenals from 2–3 mice (age 2–3 months, both sexes) per experiment were resected and placed in Petri dishes with ice-cold phosphate buffered saline (PBS). Fat tissue surrounding the adrenals was carefully removed and the adrenal cortex was thoroughly isolated from the medulla as described previously [17]. All cortical and medullary tissues, respectively, were pooled, pelleted (270× g, 5 min), and digested for 20 min at 37℃ while shaking [1.8 mg/mL collagenase, 10 mg/mL bovine serum albumin (BSA), 0.18 mg/mL deoxyribonuclease (DNase) in PBS; all from Merck]. The digestion was stopped by washing twice in PBS and primary adrenomedullary and adrenocortical cells were suspended in 1 mL complete medium consisting of Dulbecco’s modified eagle medium (DMEM/F12; Gibco, Thermo Fisher Scientific) containing 10% fetal bovine serum (FBS; Merck), 1% antibiotic-antimycotic solution (Gibco, Thermo Fisher Scientific), 1% L-glutamine (Merck) and 20 ng/mL basic fibroblast growth factor (bFGF; Merck).

Different amounts of cells (500–10,000 cells/well) suspended in a complete cell culture medium containing 20% of a 1.2% methylcellulose solution [0.24% (w/v) in DMEM/F12 medium] were seeded in round-bottom 96-well plates for suspension culture (Greiner Bio-One). In a second setup, cells were suspended in a complete medium and seeded in round-bottom 96-well Nunclon™ Sphera™ plates with an ultra-low binding surface (Thermo Fisher Scientific). In both setups, the medium was replaced every 3–4 days by carefully removing 100 µL medium and adding 100 µL fresh medium. In a third setup, adrenocortical and adrenomedullary cells suspended in complete medium were seeded in 24-well ultra-low attachment Sphericalplate 5D, where each well contains 750 rounded microwells (Kugelmeiers).

For cultivation of nestin-positive cells alone, green fluorescent protein (GFP)-positive and GFP-negative cells from the adrenal medulla and cortex of Nes-GFP mice were flow sorted on a BD FACSAria™ II flow cytometer (BD Biosciences) using BD FACSDiva™ Software using 10 µg/mL propidium iodide (BioLegend) as a live/dead cell marker. Cells were sorted directly in a complete medium containing bFGF. Afterwards, cells were seeded in Sphericalplate 5D (Kugelmeiers).

To assess the in vitro differentiation of isolated adrenal cells, spheroids (after 28 days of proliferation), were transferred to Corning^® 96-well Polypropylene Microplates (Corning) coated with 1 mg/mL poly-D-lysine (Merck Millipore^®) and 3 µL/mL fibronectin (R&D Systems) and cultured in the absence of bFGF as described previously [17]. The medium was replaced by fresh medium every 2–3 days.

Spheroids cultured for 10 days were transferred to poly-D-lysine and fibronectin-coated 96-well plates as described above. They were treated with 10 mmol/L 5-ethynyl-2’-deoxyuridine (EdU) for 3 h, followed by fixation in 4% paraformaldehyde (PFA) in PBS for 15 min. After careful washing with PBS for 3 × 5 min, spheroids were stained for EdU incorporation using the Click-iT™ Plus EdU Alexa Fluor 647 Imaging Kit (Thermo Fisher Scientific) following the instructions of the manufacturer. Nuclei were stained with 4’-6-diamidino-2-phenylindole (DAPI; Thermo Fisher Scientific) and slides were mounted with a fluorescent mounting medium (Aqua-Poly/Mount; Polysciences).

Immunofluorescence staining of adrenal sections was performed as described previously [14]. Briefly, adrenal glands were fixed in 4% PFA in PBS for 4 h and cryoprotected overnight in 30% sucrose/PBS at 4℃. Afterward, they were embedded in Tissue-Tek^® Medium (O.C.T.; Sakura Finetek), and stored at –80℃. Cryosections were sliced to 7 µm thickness (Leica CM1900; Leica Biosystems) and mounted on Superfrost™ Plus Microscope Slides (Thermo Fisher Scientific). Slides were immunostained using specific antibodies against the endothelin receptors A and B (Alomone Labs). Nuclei were stained with DAPI (Thermo Fisher Scientific) and slides were mounted with a fluorescent mounting medium (Aqua-Poly/Mount; Polysciences).

Confocal imaging was performed with a Zeiss LSM 880 inverted confocal laser scanning microscope and ZEN 2010 software (Zeiss). Fluorescence microscopy was performed with a Zeiss Axiovert 200M fluorescence microscope and AxioVision software (Zeiss). Image processing and analysis were carried out using ImageJ software, version 1.52h.

Spheroid formation was followed using an inverse microscope Axiovert 200M (Software: AxioVision 4.8; Carl Zeiss Microscopy). Each spheroid’s area (A) was analyzed using the software package Fiji (ImageJ, 1.52h; National Institutes of Health). The diameter (D) was calculated under the acceptance of an approximately spherical form of the spheroids:

There are striking parallels between the development of carotid body glomus cells and adrenal chromaffin cells. Therefore, to analyze the effect of endothelin 1 on the formation of the adrenal spheroids, the cell suspension (2.500 cells/well) containing 20% of a 1.2% methylcellulose solution was treated with endothelin 1 (10 nmol/L or 100 nmol/L; Sigma-Aldrich) starting on day 4 of cultivation. The medium was changed every 3 days.

The CellTiter 96^® AQueous One Solution Cell Proliferation Assay (Promega) was used to investigate cell viability. In analogy to the manufacturer’s instructions, primary adrenocortical cells were seeded in 96-well plates and incubated for 24 h. After 3 h incubation at 37℃ with CellTiter 96^® AQueous One reagent the absorption of the whole plate was measured at 490 nm.

Statistical analysis was performed using the GraphPad Prism program version 9.0.0 (GraphPad Software Inc.), and statistical significance was determined using two-way analysis of variance (ANOVA) followed by a Bonferroni multiple comparison test correction where appropriate. A student’s t-test was performed when only two means were compared. The significance was defined as not significant (ns) P > 0.05; * P < 0.05; ** P < 0.01; *** P < 0.001.

For the development of 3D adrenal spheroid cultures, we utilized adrenals from Nes-GFP mice to specifically follow nestin-positive stem cells/progenitors. To assess the ability of adrenocortical and adrenomedullary primary cells to form spheroids, two different methods were compared. One method made use of low-adherence round-bottom 96 well plates. In the other method, normal 96-well round bottom plates, with 0.24% methylcellulose added to the culture medium to avoid adherence, were used [18].

First, we tested five different amounts of total primary adrenocortical or adrenomedullary cells to achieve the optimal seeding density (500–10,000 cells/well). As shown previously, under these culture conditions within the first days fully differentiated cells die, whereas mainly stem and progenitor cells survive and form spheroids [14]. On day 1 after seeding, cells aggregated at the bottom of the culture plates, but no spheroids were assembled. As observed previously [14], in the medullary cultures around 10% of the cells were positive for GFP, whereas in the cortical cultures, only around 1–2% of the cells were GFP-positive (Figure 1A). On days 3–4, spheroid formation took place and the expression of nestin-GFP increased in both cell populations. Spheroids were formed with all amounts of cells, but within the cortical cultures, the spheroids quickly dissociated when using 5,000 or 10,000 cells per well demonstrating that the biggest stable spheroid diameter is around 100 µm. This was independent of the method used (methylcellulose or low-adherence) (Figure 1B). In the medullary cultures, the spheroids showed the highest stability using methylcellulose (Figure 1B). Therefore, we decided for future experiments to use 2,500 cells per well and culture them in normal 96-well round bottom plates in medium containing 0.24% methylcellulose.

To test whether primary medullary and cortical cells differ in their proliferation capabilities, we added the thymidine analog EdU to the cultures on day 9. At this time point, medullary cells showed a significantly higher proliferation rate than cortical cells (Figure 2A). Therefore, we measured the viability of adrenocortical cells after 4 weeks to exclude cell death. No statistically significant differences were found between the different cell concentrations, but viability tended to be higher at 2,500 or more cells per well than at 500–1,000 cells per well (Figure 2B). No difference between low-adherence plates and methylcellulose could be observed. We cultured both medullary and cortical spheroids for 4 weeks and then we changed the culture conditions to adherence and removed bFGF from the medium. Thereby, cells adhered and started to differentiate (Figure 1B) showing the long-term potential of the cultures.

Neural crest-derived stem cells in the adult mammalian carotid body are highly sensitive to hypoxia and endothelin [19, 20]. As there are striking parallels between carotid body glomus cells and adrenal chromaffin cell development [21], we decided to evaluate the effect of endothelin on adrenomedullary stem cells and compare them to adrenocortical cells.

Previously, we performed bulk RNA sequencing of nestin-GFP cells from the adrenal cortex and medulla, respectively. This showed that nestin-GFP from the cortex and medulla are distinct. Using this dataset (GEO: GSE1402479), in medullary nestin-GFP cells we observed a 2.97 log2fold (padj = 0.0047) higher expression of the endothelin receptor gene Ednrb and a 9.06 log2fold (padj = 2.02 × 10^–25) higher expression of the Edn3 gene encoding endothelin 3 than in cortical nestin-GFP cells. At the protein level, we observed both endothelin receptors A and B in the adrenal medulla, but with a very low expression in nestin-GFP cells (Figure 3A). We anyway added endothelin to our spheroids as they might contain other cells that are reactive to endothelin. However, for both medullary and cortical spheroids, no effect was observed on the spheroid diameter (Figure 3B) verifying that the spheroids mainly consist of cells that are not reactive to endothelin.

For potential high-throughput applications, such as drug screening or personalized medicine, we upscaled the number of spheroids, by using Sphericalplates 5D with 750 microwells per well in a 24-well plate. In these wells, both cortical and medullary adrenal cells formed spheroids comparable to those created in 96-well plates, although slightly smaller. The percentages of nestin-GFP positive cells (medulla: ~10%, cortex: 1–2%) were as for the spheroids cultured in 96-well plates. For cortical cells, because of the lower proliferation, 500 cells per microwell were necessary, whereas, for medullary cells, only 100 cells per microwell were needed (Figure 4A). To further test if nestin-GFP cells alone were able to create spheroids, we flow-sorted these cells from cortex and medulla, respectively, and small amounts per well were sorted in Sphericalplates 5D (Figure 4B). These cells divided and survived up to at least 4 weeks but did not form spheroids indicating that higher amounts of cells or other supporting cells are necessary for spheroid formation.