ENDOCRINE SOCIETY
OXFORD
Adrenocortical Mitochondria-Associated Membranes: Isolation, Characterization, and Lipidoproteomic Response to Mitotane
Alexander F. Krüger, 10 Werner Schmitz,2D Stephanie Lamer,3 Alexandra Triebig, 40D Tanja Maier,4 Carmina T. Fuss, 1,2,5,6[D José Pedro Friedmann Angeli,30D Andreas Schlosser,3D
Christian Stigloher,70 Martin Fassnacht, 1,2,5,6[D Isabel Weigand, 4[D and Matthias Kroiss 1,4[D
1Department of Internal Medicine I, Division of Endocrinology and Diabetes, University Hospital Würzburg, 97080 Würzburg, Germany 2Department of Biochemistry and Molecular Biology, University of Würzburg, Biocenter/Theodor-Boveri-Institute, 97074 Würzburg, Germany
3Rudolf-Virchow-Center for Experimental Biomedicine, University Hospital Würzburg, 97080 Würzburg, Germany
4Department of Medicine IV, University Hospital Munich, Ludwig-Maximilians-Universität München, 80336 Munich, Germany 5Comprehensive Cancer Center Mainfranken, Würzburg Site, 97080 Würzburg, Germany
6Bavarian Cancer Research Center, Würzburg Site, 97080 Würzburg, Germany
7Imaging Core Facility, University of Würzburg Biocenter/Theodor-Boveri-Institute, 97074 Würzburg, Germany
Correspondence: Matthias Kroiss, MD, PHD, Department of Medicine IV, University Hospital Munich, Ludwig-Maximilians-Universität München, Ziemssenstr. 5, München 80336, Germany. Email: Matthias.kroiss@med.lmu.de.
Abstract
Mitotane is an inhibitor of sterol O-acyltransferase 1 (SOAT1) approved for the treatment of adrenocortical carcinoma (ACC). In cells, mitotane increases reactive oxygen species, lipid peroxidation, and ultimately cell death. This mechanism is similar but distinct from ferroptosis, a cell death mechanism adrenal cortex cells are endogenously predisposed to. Both Acyl-CoA-Synthetase 4 (ACSL4), essential for ferroptosis, and SOAT1 are localized in mitochondria-associated membranes (MAM), specialized contact sites between mitochondria and endoplasmic reticulum (ER). Here, we used protein and lipid mass spectrometry to explore the role of MAMs in adrenocortical cell death.
MAMs were isolated from NCI-H295S cells treated with mitotane, the ferroptosis inducer RSL3, or control. Western blotting of marker proteins was used for quality control prior to lipid and protein mass spectrometry.
MAM fractions showed strong enrichment of SOAT1 and FATE1 (fetal and adult testis expressed 1) marker proteins, contained ACSL4, and were depleted from mitochondrial MTCO2 independent of treatment condition. Protein mass spectrometry identified IRE1alpha/ERN1, and PERK/EIF2AK3 implicated in the response to mitotane. Proteins involved in ER- and mitochondria-related processes were functionally enriched. We discovered the guanosine nucleotide exchange factor GRIPAP1 in MAMs of mitotane but not RSL3- or control-treated samples. In NCI-H295S cells mitotane upregulated GRIPAP1 expression. Mitotane but not RSL3 pronouncedly reduced the quantity of ubiquinone (Q10) and heme B in MAMs. In conclusion, locally reduced Q10 in MAM may contribute to impaired respiratory chain activity and free radical excess induced by mitotane. Recruitment of GRIPAP1 protein to MAMs may transduce cell death.
Key Words: adrenal gland, steroidogenesis, cell death, ferroptosis, cancer treatment, mitochondria-associated membranes
Abbreviations: ACC, adrenocortical carcinoma; ACSL, long-chain acyl-coenzyme A synthetase; ANOVA, analysis of variance; CANX, calnexin; DMSO, dimethyl sulfoxide; EM, electron microscopy; ER, endoplasmic reticulum; FATE1, fetal and adult testis expressed 1; FDR, false discovery rate; GRIPAP1, GRIP1 associated protein 1; LC/MS, liquid chromatography-mass spectrometry; LD, lipid droplet; LFQ, label-free quantification; MAM, mitochondria-associated membrane; MS/ MS, tandem mass spectrometry; MTCO2, mitochondrially encoded cytochrome c oxidase II; ROS, reactive oxygen species; SOAT1, sterol O-acyltransferase 1; StAR, steroidogenic acute regulatory protein.
Adrenocortical carcinoma (ACC) is a rare malignancy with overall unfavorable prognosis, yet varying outcomes between patients [1]. Complete tumor resection is a key factor for long- term survival and adjuvant treatment with mitotane has been shown to be associated with a lower proportion of local recur- rences and improved survival [2-4] except in low-risk cases [5]. Mitotane remains the sole adrenal-specific medication ap- proved for treating ACC [6] and potently counteracts tumoral steroid hormone secretion that is present in ~60% of cases [7, 8]. Mitotane treatment is complicated by multiple and common adverse drug effects that include severe neurologic
symptoms [5, 9]. Close monitoring of plasma concentrations is required to maintain therapeutic concentrations and limit adverse events [10, 11].
While mitotane has been a crucial element of therapy for decades, either as monotherapy or in combination with first- line chemotherapy [12], the precise pharmacological mechan- ism underlying its toxic effect on the adrenal cortex remains incompletely understood [13]. At the cellular level, mitotane causes mitochondrial damage [14], with decreased activity of the mitochondrial respiratory chain elicited by defective cytochrome c oxidase [15]. One potential key mechanism of
mitotane has been elucidated by our group earlier. We have demonstrated that mitotane inhibits sterol O-acyltransferase 1 (SOAT1), leading to the accumulation of toxic lipids and subsequent lipid-induced endoplasmic reticulum (ER) stress that triggers death in ACC cells [16]. Even though SOAT1 could be identified as a potential target molecule, a phase 1 study with the specific SOAT1 inhibitor nevanimibe failed to result in objective therapeutic response in advanced ACC [17]. This suggests that SOAT1 inhibition alone may not be sufficient as a treatment strategy in patients with advanced ACC. In line with this notion, tissue SOAT1 protein expres- sion was not significantly correlated with clinical response to mitotane treatment in a large multicenter study [18]. Lipid droplets (LDs) may play a more direct role in the mech- anism of action of mitotane in ACC cells. Warde et al demon- strated that mitotane targets LDs to induce lipolysis [19]. Therefore, in addition to previously proposed mitochondrial mechanisms and SOAT1 inhibition, alternative or comple- mentary pathways involving LDs should be considered.
SOAT1 is a tetrameric enzyme that catalyzes the synthesis of cholesterol esters from free cholesterol and long-chain fatty acids [20, 21]. SOAT1 has been shown to be localized at mitochondria-associated membranes (MAMs) of the ER [22, 23]. MAMs are specialized contact sites located between mito- chondria and the ER [24]. Proteins located in MAMs play crit- ical roles in maintaining the connection between the ER and mitochondria and contribute to the initiation of reactive oxy- gen species (ROS)-mediated mitochondrial apoptosis [25].
While mitotane-induced cell death in adrenocortical cells has been a subject of research for many years, the predisposition of adrenocortical cells to ferroptosis has only recently been discov- ered. Ferroptosis is an iron-dependent mode of cell death that results from the peroxidation of membrane lipids. The oxidative stress caused by steroidogenesis in ACC cells has been shown to be the primary driver of adrenocortical cells’ susceptibility to ferroptosis [26]. Experimentally, ferroptosis can be elicited in ACC cells through inhibition of the selenoprotein glutathione peroxidase 4 (GPX4) [27] with the compound RSL3 at nano- molar concentrations. Yet, ferroptosis and mitotane-induced cell death are distinct in that the latter cannot be blunted with ROS scavengers. Acyl-CoA-synthetase 4 (ACSL4) catalyzes fatty acid activation for subsequent esterification [18, 28, 29] and is essential for ferroptosis sensitivity. Like SOAT1, ACSL4 is localized at MAMs together with other marker pro- teins such as FATE1 (fetal and adult testis expressed 1), which is required for cell division and apoptosis by maintaining the distance between the ER and mitochondria. In adrenocortical tumors, increased FATE1 expression results in reduced mito- chondrial uptake of calcium, rendering cancer cells more resist- ant to cell death triggered by mitotane [30]. The MAM protein calnexin (CANX) is a ubiquitous chaperone protein facilitating protein folding and ER quality control [31].
In view of the convergence of ferroptosis and mitotane- induced cell death in lipid peroxidation and the preeminent role of MAM proteins in their execution, we here aimed to disentangle both cell death mechanisms through an integrated mass spectrometry approach focusing on MAMs. We made use of the NCI-H295S ACC cell line growing in suspension culture and determined the impact of mitotane and RSL3 on MAM composition by lipid and protein mass spectrometry techniques. An improved mechanistic understanding of mito- tane may ultimately contribute to the development of more effective and less toxic drugs for ACC.
Materials and Methods
Cell Culture
NCI-H295S cells grown in suspension were obtained from ATCC and cultivated in RPMI 1640 medium supplemented with 10% fetal calf serum, 0.13% sodium selenite, 1% trans- ferrin, and 0.04% insulin at a concentration of 4 µg/mL. Cells were cultured in flasks at 37 ℃ and 5% CO2 under hu- mid conditions. 50% of preconditioned medium was utilized for cell passaging. For assessment of viability and Western blot- ting, NCI-H295S cells in suspension or adherent NCI-H295R cells were incubated with mitotane (Sigma-Aldrich) or RSL3 (Selleckchem) and dimethyl sulfoxide (DMSO) or ethanol con- trol for 8 hours at 37 ℃. Cell viability was assessed using CellTiter-Glo (Promega) according to the manufacturer’s in- structions (Fig. 1). We decided to use a final concentration of 100µM mitotane in the cell culture flasks, as 65% of the NCI-S cells remained viable after 8 hours of treatment. This al- lowed the establishment of a sub-totally lethal concentration of mitotane over this period. For RSL3, a final concentration of 0.25uM was selected in the cell culture flasks, as 76% of the cells remained viable after 8 hours at this concentration. All cell culture flasks were thoroughly inspected for contamination and cell viability. In total, 80 large 175 cm2 cell culture flasks, each containing approximately 4 to 6 x 107 NCI-H295S cells, were required for the isolation of MAM from ACC cells. Of these, 26 selected cell culture flasks were aimed at achieving a final concentration of 100uM mitotane. Another 26 flasks were treated with a final concentration of 0.25uM RSL3. For control, 500 µL of DMSO was pipetted into each of the re- maining 26 cell culture flasks.
Preparation of Mitochondria-Associated Membranes From NCI-H295S Cells
The isolation process followed the method described by Wieckowski et al [24]. The procedure was optimized for NCI-H295S suspension cells, and the following steps performed for 16 to 24 x 108 cells treated for 8 hours with DMSO, RSL3, and mitotane. Cells were resuspended and transferred into tubes of 50 mL each, rinsed with 10 mL of phosphate-buffered saline (PBS) (without Ca2+ and Mg2+) and centrifuged at 600g for 5 minutes at 4℃. The cell pellet was resuspended, washed in 30 mL PBS (containing Ca2+ and Mg2+), combined and centri- fuged again before all cell suspensions were sedimented in a sin- gle tube at 600g for 5 minutes. The cell pellet was resuspended in 20 mL of ice-cold IB-cell buffer (225 mM mannitol, 75 mM sucrose, 0.1 mM EGTA, and 30 mM Tris-HCl pH 7.4 at 4 °℃). For homogenization, cells were resuspended and trans- ferred to a pre-cooled 30 mL glass tube with a Teflon pestle. Cells were homogenized gently and slowly using the Teflon pes- tle at 2000 rpm. After 35 cycles, cells were examined using a Countess cell counter (Thermo) and by phase contrast micros- copy to assess the extent of cell lysis with the goal to attain a cell disruption rate of 80% to 90%.
The homogenate was transferred into a 50 ml tube and centrifuged at 600g for 5 minutes at 4 ℃. The pellet contain- ing intact cells and nuclei was discarded and the supernatant centrifuged at 600g for 5 minutes at 4℃. The supernatant was collected, the pellet was discarded, and the sample was centrifuged at 8800 rpm (50.2 Ti rotor) for 10 minutes at 4 °℃. All ultracentrifuge tubes were balanced to 3 decimal pla- ces at each step [24].
A
B
150
viability (% of control)
150
*
viability (% of control)
*
100
100
T
50
50
0
0
0 µM
15μM
25uM
50 μΜ
100 μΜ
0 µM
10nM
50nM
100nM
250nM
500nM
1000nM
mitotane concentration
RSL3 concentration
The resulting pellet, containing mitochondria and MAMs, was gently resuspended in 20 ml of ice-cold IBcells-2 buffer using a glass Elvehjem homogenizer. First, 10 mL of buffer was added to the pellet, the sample was homogenized, and then another 10 mL of buffer was added.
The mitochondrial suspension was centrifuged at 8800 rpm (50.2 Ti rotor) for 10 minutes at 4 ℃. The supernatant was discarded, and the pellet was resuspended again in 20 ml of ice-cold IBcells-2 buffer. It was then centrifuged at 10 500 rpm (50.2 Ti rotor) for 10 minutes at 4 °C.
The supernatant was discarded, and the crude mitochon- drial pellet was resuspended in 2 mL of ice-cold MRB buffer using a glass Elvehjem homogenizer.
An aliquot of 0.2 mL was frozen at -20℃ for later Western blot analysis. All subsequent steps were done accord- ing to Wieckowski et al [24] (Fig. 2A).
Subsequently, 8 mL of Percoll medium were carefully pipet- ted into a 14 mL Ultra-Clear™M ultracentrifuge tube (14x 95 mm; Beckman Coulter) suitable for the SW40 rotor. The crude mitochondrial suspension in MRB buffer was gently layered on top using a cut pipette tip, followed by the addition of approximately 1 to 2 mL MRB buffer. The tube was filled to approximately 4 to 5 mm below the rim to prevent break- age during ultracentrifugation.
Centrifugation was performed using an SW40Ti swinging- bucket rotor at 27 400 rpm for 30 minutes at 4 ℃. After cen- trifugation, mitochondria appeared as a dense white band just above the bottom of the tube, whereas the MAM fraction formed a diffuse white band in the middle of the gradient (Fig 2A). The MAM fraction was carefully collected using a pipette and transferred to a 26 mL ultracentrifuge tube. The mitochondrial fraction was collected separately. Both MAM and mitochondrial fraction were diluted 10-fold (1:10) with MRB buffer.
Both the MAM and mitochondrial suspensions were subse- quently centrifuged at 8300 rpm for 10 minutes at 4 ℃ using
a 50.2 Ti rotor. The MAM-containing tubes were kept on ice, while further processing continued with the mitochondrial fraction. The mitochondrial supernatant was discarded, and the resulting pellet was gently resuspended in 20 mL MRB buffer using a glass Elvehjem homogenizer. Initially, 10 mL MRB buffer were added to the pellet for resuspension, fol- lowed by an additional 10 mL. The resuspended sample was centrifuged again at 8300 rpm for 10 minutes at 4 ℃.
The supernatant was removed, and the mitochondrial pellet was resuspended in 300 µL MRB buffer, transferred into a 1.5 mL tube, and stored at -20 ℃.
For the final step of MAM isolation, the MAM-containing supernatant kept on ice was transferred into a 26-mL ultra- centrifuge tube. After discarding the pellet, the supernatant was centrifuged at 33 200 rpm for 1 hour at 4 ℃ using the 50.2 Ti rotor. The resulting supernatant was discarded, and the MAM pellet was resuspended in 100 µL MRB buffer, transferred to a 1.5 mL tube, and stored at -20 ℃ until fur- ther analysis. The resulting MAM and mitochondrial frac- tions were used for subsequent analyses [24].
SDS-PAGE and Western Blot
Samples comprising 4.5 µg of protein from organelle prepara- tions or 15 µg cell lysate were loaded onto a 4% to 20% de- naturing gradient gel (BioRad) and separated using SDS-PAGE and transferred to a nitrocellulose membrane by tank blot (BioRad). The membrane was blocked with 5% skimmed milk powder in TBS-T (20 mM Tris, 200 mM NaCl, 0.01% Tween-20) at room temperature for 1 hour. The membrane was incubated overnight at 4 ℃ with the fol- lowing primary antibodies: ACSL4: Santa Cruz sc271800 (RRID:AB_10715092, https://scicrunch.org/resolver/AB_107 15092) at a dilution of 1:1 000; FATE1: Abcam AB111486 (RRID:AB_10890082, https://scicrunch.org/resolver/AB_1089 0082) 1:200; SOAT1: Abcam AB39327 (RRID:AB_778001,
A
MAMs
MRB
pure MAM
MAM
centrifuge and homogenize cells and layer mitochondria suspension on Percoll medium
Centrifuged at 27400 rpm for 30 min at 4 °℃ in a SW40 rotor
Centrifugation (2x)
2
Percoll Medium
~5x10º NCI-H295 ACC cells
mitochondria
Centrifugation (2x)
pure mitochondria
B
ER EtOH
NCI-H295S Lysate
ER mitotane
CM EtOH
cM mitotane
PM EtOH
pM mitotane
MAM EtOH
MAM mitotane
kDa
CANX 90 kDa
- 130
endoplasmic Reticulum
- 100
- 70
ACSL4 75 kDa
- 100
SOAT1
ACSL4
CANX
FATE1
- 70
MTCO2 60 kDA
- 70
- 55
SOAT1 50 kDa
- 55
H MTCO2
Mitochondria
- 35
FATE1
21 kDa
- 25
https:/scicrunch.org/resolver/AB_778001) 1:1 000; Calnexin: NOVUS NB300-518 (RRID:AB_3187515, https://scicrunch. org/resolver/AB_3187515) 1:1 000; MTCO2: Abcam #AB3298 (RRID:AB_303683, https://scicrunch.org/resolver/ AB_303683) 1:1 000; and GRAP1/GRIPAP1: Santa Cruz #sc398198 (RRID:AB_3712998, https://scicrunch.org/ resolver/AB_3712998) at a dilution of 1:100. HRP-labeled sec- ondary antibodies (goat anti-rabbit: Jackson ImmunoResearch Laboratories, 111-035-144 and goat anti-mouse: Jackson ImmunoResearch Laboratories, 115-035-003) were diluted 1:10 000 and incubated at room temperature for 1 hour. The protein-antibody complex was visualized using Amersham ECL Prime reagent (GE Healthcare) and captured on x-ray film (Fuji) using enhanced chemiluminescence. For the follow- up experiments, we used NCI-H295R cells, which represent the standard model for routine functional experiments in our la- boratory (Fig. 2B).
Electron Microscopy
Electron microscopy (EM) sample preparation was conducted as described in detail in Schaller et al with the modifications that the fraction from centrifugation was fixated in 4% PFA only [32]. EM sections were analyzed at a JEOL JEM-1400Flash electron
microscope at 120 kV acceleration voltage and recorded with a Matataki camera.
Lipid Mass Spectrometry
For lipid extraction from isolated MAMs, the BuMe method, a fast and simple chloroform-free technique for extracting to- tal lipids from animal tissue, was applied [33].
A 20-µL sample was acidified with 280 uL of 1% acetic acid (in water) to extract the lipids. Next, 300 µL of n-butanol/ methanol (3:1, v/v) was added to the sample. After mixing, the sample was treated with ultrasound. Afterward, 300 uL of heptane/ethyl acetate (3:1, v/v) was added, followed by another round of mixing, ultrasound treatment, and centrifugation. The upper phase was transferred to a new 1.5-mL tube.
For further sample purification, 300 µL of heptane/ethyl acet- ate (3:1, v/v) was added to the lower phase, followed by mixing, ultrasound treatment, and centrifugation. The upper phases were combined and evaporated under nitrogen at 45 ℃.
Liquid chromatography-mass spectrometry (LC/MS) ana- lysis was performed on a Thermo Scientific Dionex Ultimate 3000 UHPLC system connected to a Q Exactive mass spec- trometer (QE-MS) equipped with a HESI probe (Thermo
Scientific, Bremen, Germany). The prepared sample was dis- solved in 50 µL of isopropanol for LC/MS analysis. Chromatographic separation was achieved by applying 3 µL of sample on a Acclaim C8 column (100x2.1 mm, 3-um, Thermo), protected by a Supelco ColumnSaver particle filter (Merck) and a gradient of mobile phase A (CH3CN/H2O/for- mic acid (10/89.9/0.1, v/v/v)), mobile phase B (isopropanol/ CH3CN/H2O/formic acid (45/45/9.9/0.1, v/v/v/v)) maintain- ing a flow rate of 200 µL/min and a column temperature of 40 °C.
The LC gradient program was 20% mobile phase B for 2 minutes, followed by a linear increase to 100% B within 5 minutes, maintaining 100% B for 33 minutes, then return- ing to 20% B in 1 minute, followed by 5 minutes 20% B for column equilibration before each injection.
The eluent was directed to the QE-MS from 2 minutes to 40 minutes after sample application [33]. LC/MS analysis was performed as described earlier [34]. Mass detection was conducted in alternating pos./neg. full scan mode (at 70 k reso- lution, scan range m/z 200-1650, automatic gain control [AGC] target 3E6 and 200 ms max. injection time). HESI parameters: Sheath gas: 30, aux gas: 10, spray voltage: 2.5 kV in pos.mode and 3.6 kV in neg.mode, capillary temperature: 300 ℃, S-lens RF level: 55, aux gas heater temperature: 120 ℃.
Manual curation and integration of chromatographic peaks were performed with TraceFinder 5.1 using a mass tolerance of 6 mMUs.
The lipidomic analysis of lipids in purified organelles was performed in 3 biological replicates.
Preparative Protein Gel Electrophoresis and In-Gel Digestion
Protein precipitation was performed overnight at -20 ℃ with 4-fold volume of acetone. Pellets were washed with acetone at -20 ℃. Precipitated proteins were dissolved in NuPAGE® LDS sample buffer (Life Technologies), reduced with 50 mM DTT at 70 ℃ for 10 minutes and alkylated with 120mM io- doacetamide at room temperature for 20 minutes. Separation was performed on NuPAGE® Novex® 4% to 12% Bis-Tris gels (Life Technologies) with MOPS buffer according to manu- facturer’s instructions. Gels were washed 3 times for 5 minutes with water and stained for 60 minutes with Simply Blue™M Safe Stain (Life Technologies). After washing with water for 1 hour, each gel lane was cut into 15 slices. The excised gel bands were destained with 30% acetonitrile in 0.1M NH4HCO3 (pH 8), shrunk with 100% acetonitrile, and dried in a vacuum concen- trator (Concentrator 5301, Eppendorf, Germany). Digests were performed with 0.1 µg trypsin per gel band overnight at 37°℃ in 0.1M NH4HCO3 (pH 8). After removing the super- natant, peptides were extracted from the gel slices with 5% formic acid, and extracted peptides were pooled with the supernatant.
Mass Spectrometric Proteome Analyses
NanoLC-MS/MS analyses were performed on an Orbitrap Fusion (Thermo Scientific) equipped with a PicoView Ion Source (New Objective) and coupled to an EASY-nLC 1000 (Thermo Scientific). Peptides were loaded on a trapping col- umn (2 cm × 150 um ID, PepSep) and separated on a capillary column (30 cm × 150 um ID, PepSep) both packed with 1.9 umC18 ReproSil and separated with a 30-minute linear gradi- ent from 3% to 30% acetonitrile and 0.1% formic acid and a
flow rate of 500 nL/min. Both MS and MS/MS scans were ac- quired in the Orbitrap analyzer with a resolution of 60 000 for MS scans and 30 000 for MS/MS scans. HCD fragmentation with 35% normalized collision energy was applied. A Top Speed data-dependent MS/MS method with a fixed cycle time of 3 seconds was used. Dynamic exclusion was applied with a repeat count of 1 and an exclusion duration of 30 sec- onds; singly charged precursors were excluded from selection. Minimum signal threshold for precursor selection was set to 50 000. Predictive AGC was used with AGC a target value of 4x105 for MS scans and 5x104 for MS/MS scans. EASY-IC was used for internal calibration.
A total of 6 samples were analyzed, consisting of 2 repli- cates of MAMs treated with mitotane, RSL3, and DMSO, respectively.
Raw MS data files were analyzed with MaxQuant version 1.6.2.2 [35]. Database search was performed with Andromeda, which is integrated in the utilized version of MaxQuant. The search was performed against the UniProt Human Reference Proteome database (Release 2022_02, UP000005640, 79 684 entries). Additionally, a database con- taining common contaminants was used. The search was per- formed with tryptic cleavage specificity with 3 allowed miscleavages. Protein identification was under control of the false discovery rate (FDR; < 1% FDR on protein and peptide spectrum match [PSM] level). In addition to MaxQuant de- fault settings, the search was performed against following variable modifications: Protein N-terminal acetylation, Gln to pyro-Glu formation (N-term. Gln) and oxidation (Met). Carbamidomethyl (Cys) was set as fixed modification. Further data analysis was performed using R scripts devel- oped in-house. Label-free quantification (LFQ) intensities were used for protein quantitation [34]. Proteins with fewer than 2 razor/unique peptides were removed. Missing LFQ in- tensities were imputed with values close to the baseline. Data imputation was performed with values from a standard nor- mal distribution with a mean of the 5% quantile of the com- bined log10-transformed LFQ intensities and a SD of 0.1. For the identification of significantly enriched proteins, me- dian log2 transformed protein ratios were calculated from the 2 replicate experiments and boxplot outliers were identi- fied in intensity bins of at least 300 proteins. Log2 trans- formed protein ratios of sample vs control with values outside a 1.5x (significance 1) or 3x (significance 2) interquar- tile range (IQR), respectively, were considered as significantly enriched in the individual replicates.
Statistical Analyses
Statistical analyses were performed using Prism 9 software (GraphPad). The results of cell survival data are presented as mean ± standard error of the mean (SEM) and groups were com- pared using the Kruskal-Wallis test with Dunn’s correction for multiple testing. Lipid mass spectrometry data are presented as mean ± (SEM). A two-way analysis of variance (ANOVA) was conducted to determine statistical significance. Differences were considered significant for P values less than .05.
Pathway analysis of protein mass spectrometry data was performed using webgestalt (https://www.webgestalt.org/). Proteome data from MAM preparations were compared to whole cell proteome in the RampDB Genomics database. All datasets are available online [36]. Image analysis of electron micrographs was performed using Fiji (Image]), an open- source image processing platform.
Results
MAM Marker Proteins in ACC Cells With and Without Mitotane
To validate the cell fractionation method for the isolation of MAMs and to assay the subcellular distribution of marker pro- teins in MAMs with mitotane treatment we conducted fraction- ation experiments and Western blot analysis of different purified subcellular fractions. In the MAM fraction, the expres- sion of the 2 key established MAM marker proteins SOAT1 [23] and FATE1 [30] was clearly enriched compared to purified mitochondria (pM) while ACSL4 [23] and CANX [37] were de- tectable also in mitochondrial fractions (Fig. 2B). Importantly, the abundance of CANX and mitochondrially encoded cyto- chrome c oxidase II (MTCO2) was strongly reduced in MAM fractions of both control- and mitotane- treated cells. Although MTCO2 was strongly disenriched in MAM fractions, depletion was not complete, which is in line with the close phys- ical contact of MAMs and mitochondria. ACSL4, which also plays a critical role in steroidogenesis [28], was also present but not enriched in the MAM fractions compared to ER.
After successful establishment of MAM isolation from treated and untreated NCI-H295S ACC cells, we used SOAT1, FATE1, and MTCO2 expression as quality control
for each MAM preparation used for mass spectrometry analyses.
Proteomic Characterization of Isolated MAMs in ACC Cells
We next conducted protein mass spectrometry to more compre- hensively understand the protein composition of MAMs in ACC cells. We first successfully confirmed the presence of MAM markers ACSL4, FATE1, SOAT1, CANX, IRE1alpha/ ERN1, and PERK/EIF2AK3, RANBP2, STAR, and CYP17A1 by proteomics (Fig. 3A, Table 1 and Supplementary File S1 [36]). These MAM markers were detected in all 3 MAM sam- ples, except for EIF2AK3, which was present in only 2 samples.
As an unsupervised confirmation of our approach, we found substantial mass spectrometry signals of peptides identifying IRE1a and PERK-factors involved in the ER stress response in adrenocortical cells-as well as steroidogenic acute regula- tory protein (StAR) and CYP17A1, 2 central components of steroidogenesis present in the MAM fraction. StAR catalyzes the rate limiting step of steroidogenesis by transporting choles- terol to the inner mitochondrial membrane [38], where it undergoes conversion into a spectrum of steroid hormones [39, 40]. StAR is synthesized upon hormonal regulation or
A
4000
1000
RANBP2
Mol.weight (kDa)
400
EIF2AK3
ERN1
100
CANX
ACSL4
STAR
40
CYP17A1
SOAT1
0
FATE1
10
4
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
8
8.5
9
9,5
Log10 (Median iBAQ Intensity)
B
FDR ≤0.05
FDR>0.05
mitochondrial gene expression
cytoplasmic translation
Golgi vesicle transport
protein folding
protein-RNA complex organization
endosomal transport
vesicle organization
ribonucleoprotein complex biogenesis
mitochondrion organization
establishment of protein localization to organelle
0.0
Enrichment ratio
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
| Protein | Gene | Razor + unique peptides Experiment 1 | Razor + unique peptides Experiment 2 | Razor + unique peptides Experiment 3 |
|---|---|---|---|---|
| Long-chain fatty acid-CoA ligase 4 | ACSL4 | 5 | 7 | 6 |
| Serine/threonine-protein kinase/endoribonuclease IRE1; Serine/threonine-protein kinase; Endoribonuclease | ERN1 | 10 | 19 | 32 |
| Steroid 17-alpha-hydroxylase/17,20 lyase | CYP17A1 | 30 | 30 | 37 |
| Fibroblast growth factor receptor 1; Fibroblast growth factor receptor | FGFR1 | 11 | 13 | 12 |
| Calnexin | CANX | 55 | 61 | 67 |
| Sterol O-acyltransferase 1 | SOAT1 | 16 | 16 | 25 |
| Steroidogenic acute regulatory protein, mitochondrial | STAR | 11 | 13 | 8 |
| RAN binding protein 2 | RANBP2 | 27 | 20 | 70 |
| Fetal and adult testis-expressed transcript protein | FATE1 | 14 | 14 | 14 |
| Eukaryotic translation initiation factor 2-alpha kinase 3 | PERK/ EIF2AK3 | 0 | 9 | 12 |
under stress and is localized in MAM before being loaded into outer mitochondrial membrane [41]. CYP17A1 catalyzes the conversion of pregnenolone to dehydroepiandrosterone (DHEA) and progesterone to 17-hydroxyprogesterone, there- by holding a pivotal position in the biosynthesis of various ster- oid hormones [42]. Functional enrichment analysis was performed using WebGestalt with the significance threshold set at FDR <0.05. The analysis identified the biological proc- esses mitochondrial gene expression, cytoplasmic translation, Golgi vesicle transport, protein folding, and endosomal trans- port. Additional processes, such as protein targeting, vesicle organization, mitochondrion organization, establishment of protein localization to organelle, and ribonucleoprotein com- plex biogenesis, were also highly enriched. These findings suggest that MAMs are strongly associated with pathway com- ponents related to protein synthesis, intracellular transport, and mitochondrial function. The enrichment ratios for these processes ranged from ~0.8 to 2.8, indicating a significant over-representation relative to the RampDB Genomics back- ground data set (Fig. 3B).
GRIPAP1 in Mitotane-Treated MAMs
We next studied quantitative changes of the proteome induced by RSL3 and mitotane at the level of MAMs using proteome analysis. Among 4611 proteins, 49 were significantly upregu- lated, while 4 were significantly downregulated upon mito- tane treatment. The full list of regulated proteins is provided in Supplementary File S2 [36]. Of these, we focused on GRIPAP1 that was detected in both control- and mitotane- treated samples and decreased significantly in at least 2 repli- cates after mitotane, suggesting a drug-specific effect (Fig. 4A and B).
GRIPAP1 can strongly activate the JNK (c-Jun N-terminal kinase) signaling pathway leading to apoptosis. In an ex- ploratory experiment, we treated NCI-H295R cells for 6 hours with mitotane and assessed GRIPAP1 expression in whole cell lysates. In Western blot experiments, GRIPAP1 expression increased at 12.5uM mitotane but returned to baseline at higher concentrations (Fig. 4C). Mitotane further induced a dose-dependent increase of caspase 3 cleavage (Fig. 4D).
In MAMs treated with RSL3, 27 proteins were upregulated while 1 protein was significantly downregulated (Fig. 5). Manual curation of these proteins did not reveal a biologically relevant context for these. The complete list of regulated pro- teins is available in Supplementary File S3 [36].
Lipidomic Characterization of Isolated MAMs in ACC Cells
The preponderance of lipid-related proteins and the relevance of lipids in the execution of ferroptosis and mitotane-induced cell death led us to study the lipid composition of the MAM fractions isolated from NCI-H295S cells by lipid mass spec- trometry. The lipid classes of lysophosphatidic acid, phospha- tidyl ethanolamines, phosphatidyl inositols, phosphatidyl serines, and isoprenoids were studied, while oxilipids were not accessible with our method (Fig. 6A and File S4 [36]).
Interestingly, we observed that the overall lipid composition of MAMs remained largely unchanged upon treatment with both mitotane and RSL3. Most lipids analyzed did not exhibit significant differences between the drug-treated and control groups (Fig. 6A).
The isoprenoids ubiquinone and free cholesterol were re- markable exceptions (Fig. 6A and 6B). Mitotane treatment led to a significant downregulation of the mean relative nor- malized peak area of heme b in MAMs from 100 ± 42 to 19 ±7 and ubiquinone from a mean of 100±30 to 47±4 (ANOVA P =. 02) relative normalized peak area (%). No sig- nificant changes were observed in RSL3-treated cells, with heme b levels remaining stable (100 ± 42 to 93 ± 74) and ubi- quinone levels showing only a slight increase (100 ±30 to 123 ±29).
Furthermore, we found a nonsignificant reduction in the levels of cholesterol in mitotane-treated MAMs 100 ±7 to 66 ± 16 (ANOVA P =. 16) (Fig. 6B). These findings suggest that mitotane specifically affects the levels of these lipid com- ponents in MAMs, potentially impairing mitochondrial func- tion and cellular processes associated with MAMs.
In contrast, RSL3 treatment did not induce significant changes in the levels of these lipid species in MAMs, indicating a differential effect between the 2 drugs on lipid metabolism within this membrane compartment.
A
Marker by:
KRT1
(Row Number)
11
11
Color by:
sig. (2)
10,5
10,5
0
1
10
10
2
GRIPAP1
DSP
CELF1
Size by:
RAD50
FUBP3
Razor … unique.pepti.
9,5
EZR
CHMP4B
NDRG2
JUP
9,5
≥200
PLEC
== 0
o
PTMA
CAD
EIF5B
TARDBP;TDP43
intensity
9
DSC1
CHMP1A
intensity
9
RAB3GAP1
NOV
ABCF3
DSG1
8,5
MAPRE2
GRIPAP1
· DST
SRSF6
XPO5
8,5
MY01D
PHKB
DVL2
HDGF
8
RBM14
UBXN1
HNRNPF
8
HNRNPDL
GFPT1
PEG10
WDR62
PUM1
7,5
IKBKG
GALK1
MCM7
ERC1
PBK
7,5
TRIP13
DAZAP1
MCM6
AGL
7
ANKHD1
MAGOHB
PLAA
NUCKS1
7
HNRNPH2
SBSN
6,5
GCC1
6,5
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
median norm log LFQ
median norm log LFQ
B
6
Marker by
(Row Number)
Color by:
5
sig combined
@0
4
1
C
D
·2
3
4
Size by:
Mitotane [M]
sum razor and uniqu ..
2
200
NCI-H295 R
6h Mitotane [uM]
0
1
5
12.5
25
.- 0
GRIPAP1
0
12.5
25
50
Vinculin
117 kDa
1
- 100
GRIPAP1
cleaved Casp3
17 kDa
a
- 70
12 kDa
-1
2
-3
-5
4
3
-2
-1
0
1
2
3
4
5
6
Transmission Electron Microscopy of MAMs
Transmission electron microscopy of the MAM-enriched frac- tion clearly revealed the presence of extended membranous structures closely resembling those described by Lu et al [43]. These appear as continuous, electron-dense strands with characteristic bilayer morphology. The membranes are readily identifiable by their typical EM contrast: 2 parallel electron-dense lines separated by an electron-lucent space, in- dicative of a lipid bilayer structure. This contrast pattern is a well-established criterion for identifying lipid bilayers, as the hydrophilic head groups of the phospholipids appear electron-dense, while the hydrophobic core appears less so. Examples of such bilayer structures are visible and marked in the Supplementary Fig. S1 [36]. The samples may also con- tain putative remnants of mitochondrial membranes, as well as flattened or tubular ER-like structures.
Numerous discrete, spherical particles with diameters in the range of 25 to 30 nm were frequently observed along these membranes fitting with the expected size of ribosomes. A de- tailed visualization of electron microscopy images is given in Supplementary Fig. S2 [36].
Discussion
This study, to our knowledge, is the first comprehensive multi-omics analysis of MAMs isolated from adrenocortical cells. We used a differential centrifugation method for MAM isolation from large quantities of cultivated NCI-H295S cells. MAM enrichment was verified by the detec- tion of marker proteins. By applying lipid mass spectrometry, we found co-enzyme Q10 and heme B to be significantly de- pleted from MAMs after mitotane treatment of cells. This is likely to contribute to the therapeutically used toxic effects of mitotane, as Q10 and heme B are constituents of the mito- chondrial respiratory chain. In addition, we describe the mitotane-induced localization of GRIPAP1 at MAMs that may play a role in the execution of mitotane-induced cell death. At variance, no significant differences in MAM com- position were observed when cells were treated with RSL3, an inducer of ferroptosis.
MAMs play an important role at the interface between the ER and mitochondria. At present, the composition and func- tion of MAMs in different tissues is incompletely understood. Biochemical analyses have been hampered by their tedious
11,5
11
10,5
10
CISD2
NIPSNAP3A
9,5
UQCRQ
BRI3BP
intensity
LAMP1
FAHD1
9
HLA-C
EIF5A;EIF5AL1;EIF5A2
MACROD1
8,5
MRPL18
BAK1
ATXN2
GCDH
COPZ1
ACSF2
SCAMP4
8
BCKDK
DHTKD1
7,5
TOP2B
MRPL23
MRPS14
7
LACTB
MAGOHB
THNSL1
POLB
TRIP13
6,5
FAM21C
PCM1
-3
-2
-1
0
1
2
3
4
5
6
median norm log LFQ RSL3/DMSO
and vulnerable isolation and the complexity of downstream methodology. We here applied a published method of MAM isolation which was adapted to large scale adrenocortical cell culture. To validate our MAM isolation method, we used the used markers SOAT1, FATE1, ACSL4, and CANX. SOAT1 has repeatedly been described as a metabolic- ally important component of MAMs [23] and catalyzes the es- terification of cholesterol with polyunsaturated fatty acids. FATE1 encodes a cancer-testis antigen [44] and regulates calcium- and drug-induced cell death [30] by uncoupling the ER and mitochondria. Both SOAT1 and FATE1 were en- riched in purified MAM fractions. Additionally, the low but detectable abundance of CANX, a well-established ER and MAM marker, strengthens the notion that MAMs represent regions of close interaction between the ER and mitochondria. Although we found CANX to be localized in the ER fraction, CANX was described to be localized at MAMs as well [45]. Similar to SOAT1, ACSL4 catalyzes the esterification of fatty acids to form membrane phospholipids [28]. Several studies have observed the localization of ACSL4 in MAMs and support its significant involvement in mitochondrial fusion during steroidogenesis. In addition, expression of ACSL4 is crucial for cells to die upon RSL3-induced ferroptosis [29]. Although we cannot completely exclude contamination of purified MAMs with membrane components from other or- ganelles, our Western blot analysis provides evidence for a sig- nificant enrichment of these organelles.
The identification of ACSL4, FATE1, SOAT1, CANX, IRE1alpha/ERN1, and PERK/EIFA2K3 by mass spectrometry in purified MAMs reflects our immunoblot findings and adds additional potential MAM marker proteins [46]. IRE1alpha/ ERN1 and PERK/EIFA2K3 have been associated with endo- plasmic reticulum stress especially in the adrenal cortex as a key mechanism of mitotane action [16]. StAR plays a pivotal role in cholesterol transport into mitochondria, enabling ster- oid hormone synthesis in the adrenal cortex [38]. Its presence in MAMs has potential implications for cholesterol trafficking and steroidogenesis within this distinct subcellular compart- ment. Unexpectedly, CYP17A1 was also localized in MAMs
of ACC cells. Given the known functions of MAMs in lipid me- tabolism and calcium signaling, the presence of CYP17A1 in these membranes suggests localized subdomains of cholesterol and steroid metabolism.
Electron microscopy of the isolated MAM fraction revealed the presence of abundant membrane structures and electron-dense particles consistent with ribosomes. Based on their size and localization, the darkly stained particles are highly likely to represent ribosomes (Fig. S2 [36]). Together, these ul- trastructural features strongly support the presence of intact membrane complexes in the isolated fraction and are in line with the expected morphology of MAMs, including rough ER-like membranes and ribosome-like particles, while the ab- sence of intact mitochondria supports the specificity of the MAM isolation procedure. The EM images provide additional evidence that the isolated fraction contains lipid bilayers, sup- porting the presence of MAM structures (Fig. S3 [36]).
We next studied the MAM lipidome that reflects the specific functions of this subdomain in interorganelle communication, lipid metabolism, and calcium signaling and hence differs from other regions of the ER. Among the quantitatively predominant lipid classes in MAMs are cardiolipins, phosphatidylserine (PS), phosphatidylethanolamine (PE), cholesterol, and sphingolipids [47]. In line with the literature, MAM fractions exhibited a high- er concentration of cholesterol compared to the rest of the ER [44]. Among the 626 lipid subclasses accessible to our method, only cholesterol, ubiquinone/Q10 and heme B were significantly different between control- and mitotane-treated cells. Although this does not exclude that more subtle changes in the lipid com- position may contribute to the mechanism of mitotane, the de- crease in cholesterol is seemingly contradictory to previous experiments where an accumulation of free cholesterol was ob- served [16]. However, these experiments were obtained in whole cell lysates. Our finding is likely explained by the repression of cholesterol synthesis resulting from SOAT1 inhibition by mito- tane leading to the accumulation of free cholesterol. Reduced Q10 levels are a potential causative of the known disruption in the function of the respiratory chain. Q10 acts as an electron car- rier between complex I, II, and III of the electron transport chain. Its decrease in mitotane-treated cells may compromise electron transfer and lead to subsequent reduction in ATP production. This could contribute to the decline in cellular energy levels and the overall dysfunction of mitochondria respiratory chain, especially complex IV, in mitotane-treated cells that has been observed [15]. Interestingly, Q10 also plays a relevant role in pro- tecting cells from the deleterious effects of ROS as demonstrated by its role as a ferroptosis suppressor [48]. Q10 was unchanged after RSL3 treatment but differences in oxilipids would have es- caped our detection.
The reduced levels of heme B in mitotane-treated cells may contribute to the significant accumulation of oxygen free rad- icals. Heme B, an iron-containing molecule, serves as a crucial component of several proteins involved in the electron trans- port chain. Specifically, heme B is an essential cofactor of suc- cinate dehydrogenase (Complex II) and cytochrome-bc1 complex (Complex III). Its reduced quantity is likely to disrupt electron transfer reactions, leading to the leakage of electrons and subsequent generation of ROS.
Heme B also is a crucial cofactor for catalase, an enzyme in- volved in detoxifying hydrogen peroxide (H2O2). In the nor- mal cellular context, catalase converts H2O2 into water and molecular oxygen, preventing the accumulation of potentially harmful ROS. The reduced levels of heme B observed in
A
isoprenoids + heme B
lipid class DMSO-treated MAMs
mitotane-treated MAMs
RSL3-treated MAMs
lipid classes
sphingomyelines
cerebrosides
DMSO mitotane RSL3
PGE2
Fold change
phosphatidyl serines
phosphatidyl inositoles
>100
99-20
phosphatidyl ethanolamines
lyso-phosphatidy serines
Q10{H+}
<20
heme B
Average Rel. Norm Peak Area {%}
lyso-phosphatidates
acyl carnitines
cholesterol
B
relative normalized peak area
200
DMSO
*
*
₸
mitotane
150
RSL3
T
100
T
50
T
T
0
T
Q10 [H+]
heme B
cholesterol
mitotane-treated MAMs may impair catalase activity and re- duce its ability to eliminate H2O2, leading to increased oxida- tive stress.
The synthesis of heme A, a component of cytochrome c oxi- dase, depends on both the availability of heme B and isopre- noids for its side chain modification. Although we could not measure heme A directly, the downregulation of both isopre- noids and heme B through mitotane may explain the reduced cytochrome c oxidase activity observed by Hescot et al [15].
The combined effects of reduced Q10 and heme B synthesis, impaired mitochondrial respiration, and increased ROS pro- duction are plausible contributors to observed cellular stress upon mitotane treatment. It was therefore interesting to
observe increased GRIPAP1 in MAMs when cells were treated with mitotane.
GRIPAP1 has been demonstrated to function as a scaffold- ing protein that facilitates the phosphorylation of c-Jun N-terminal kinases (JNK) in vitro [49]. Notably, GRIPAP1 is a substrate of caspase-3, and the cleavage of GRIPAP1 by caspase-3 is required for JNK activation. Furthermore, activa- tion of the JNK signaling pathway is associated with the in- duction of apoptosis, linking GRIPAP1 to apoptotic signaling through its role in JNK activation. Upon activation, JNK translocates to the cell nucleus, where it phosphorylates specific target proteins such as c-Jun, thereby regulating apop- tosis, among other cellular stress processes [50]. Importantly,
mitotane
cholesterol
Acetyl-CoA
mevalonate pathway
isoprenoids
Q10
heme B
intermembrane space
ROS
Q10
1
III
Cyt c
inner mitochondrial membrane
IV
ATP synthase
II
×
X
mitochondrial matrix
×
Q10
impairment
heme B
impairement
we also show that mitotane can induce caspase 3 cleavage, a hallmark of apoptosis.
Taken together, our findings underscore the dynamic nature of the MAM proteome and its modulation in response to spe- cific drug treatments. The reproducibly increased abundance of GRIPAP1 in mitotane-treated MAMs indicates its potential contribution to the apoptotic signaling pathways activated by mitotane in adrenal carcinoma cells.
Within the scope of the current project, we did not have the opportunity to study the functioning of GRIPAP1 in more de- tail. However, it is tempting to propose a model where, on the one hand, mitotane inhibits cholesterol esterification leading to the downregulation of the isoprenoid pathway as demon- strated by Sbiera et al, and on the other hand, mitotane re- duces the abundance of Q10, a molecule involved in the electron transport chain and resistance to oxidative stress as shown here (Fig. 7). The reduced availability of heme B likely contributes to a higher burden of oxidative stress. These alter- ations in cellular metabolism, along with the increased local abundance of GRIPAP1 may initiate apoptotic pathways in mitotane-treated cells. Yet, the exact mechanisms by which GRIPAP1 promotes apoptosis and its interaction with other signaling pathways require further investigation.
Overall, our results demonstrate that treatment with mito- tane leads to a specific downregulation of isoprenoids, includ- ing ubiquinone and heme B. A reduction in free cholesterol was observed, although it did not reach statistical significance. Further investigations are warranted to elucidate the mechan- istic basis of these lipid changes induced by mitotane and their functional consequences in MAMs. Understanding the
specific impact of mitotane on the lipid composition of MAMs could contribute to a better understanding of the drug’s mode of action and its effects on cellular processes in adrenal carcinoma cells. It is noteworthy that the local lipid environment profoundly affects the consequences of mitotane incorporation in membranes that itself contributes to mito- tane activity [51].
Our study has several limitations: first, we rely on known marker proteins to confirm the identity of our membrane preparation with MAMs. The specificity of these markers re- mains unclear, as not all of them have been shown to be exclu- sively localized in MAMs. Further, some contamination with mitochondrial material appears likely given the detectable ex- pression of MTC02 in MAM fractions. Second, we only used NCI H295S cells. This could not be circumvented, due to the necessity to cultivate extreme quantities of cells, which results in considerable handling times for the preparation of mem- brane fractions, rendering more than 3 experimental sets im- possible. On the other hand, this cell line is the only established model with adrenal steroidogenesis readily in place. However, our primary aim was to establish and validate a methodological workflow for the isolation and characteriza- tion of adrenocortical MAMs. Future studies should extend this approach to additional ACC cell lines, such as CU-ACC1 and CU-ACC2, to strengthen the generalizability of the findings [52, 53]. Third, the number of experiments available for each analysis is still limited. This is again related to the required quantities of biological material and their downstream processing. Fourth, Immunogold labeling of spe- cific markers in these preparations was unsuccessful, most
likely due to the harsh conditions required for EM sample pro- cessing. Antibodies that perform reliably in immunofluores- cence often fail to bind in EM samples, as the hydrophobic environment of the embedding resins interferes with antibody recognition. Nonetheless, transmission EM images still reveal the structural homogeneity of the samples and support isola- tion of MAMs. Finally, we were unable to directly prove the role of individual proteins like GRIPAP1 and lipids like Q10 in executing mitotane effects within this study. Given the multiplicity of mitotane effects in adrenocortical cells, we con- sider it unlikely that interference with one pathway may com- pletely abolish cell death induced by mitotane.
Understanding the implications of GRIPAP1 upregulation in response to mitotane treatment could provide insights into the molecular mechanisms underlying the therapeutic ef- fects of mitotane on ACC cells. Together, our findings provide novel insights into the lipid alterations associated with mito- tane treatment and their potential implications for mitochon- drial and MAM-associated functions in ACC cells. This study establishes and validates a methodological framework for the isolation and characterization of MAMs in ACC cells. The re- sulting lipidomic and proteomic dataset represents an essen- tial reference resource that will enable future mechanistic studies addressing the functional role of MAMs in adreno- cortical tumor biology and mitotane response.
Funding
This study was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under project # 314061271 and # 237292849. The Transmission Electron Microscope JEOL JEM-1400Flash is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)-under project # 426173797.
Author Contributions
A.F.K., I.W., and M.K. contributed to the conceptualization and design. A.F.K., I.W., and M.K. wrote the original draft. M.K., I.W., and M.F. supervised the project. New reagents/ analytic tools: A.S. Analysis and interpretation of data: A.F.K., W.S., S.L., C.T.F., J.P.F.A., I.W., C.S., and M.K. All authors reviewed and edited the manuscript.
Disclosures
All authors declare no conflicts of interest related to this manuscript.
Data Availability
Original data generated and analyzed during this study are in- cluded in this published article or in the data repositories listed in References.
References
1. Elhassan YS, Altieri B, Berhane S, et al. S-GRAS score for prognos- tic classification of adrenocortical carcinoma: an international, multicenter ENSAT study. Eur J Endocrinol. 2021;186(1):25-36.
2. Terzolo M, Baudin AE, Ardito A, et al. Mitotane levels predict the outcome of patients with adrenocortical carcinoma treated adjuv- antly following radical resection. Eur J Endocrinol. 2013;169(3): 263-270.
3. Terzolo M, Angeli A, Fassnacht M, et al. Adjuvant mitotane treat- ment for adrenocortical carcinoma. N Engl J Med. 2007;356(23): 2372-2380.
4. Fassnacht M, Dekkers OM, Else T, et al. European Society of Endocrinology clinical practice guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European network for the study of adrenal tumors. Eur J Endocrinol. 2018;179:G1-G46.
5. Terzolo M, Fassnacht M, Perotti P, et al. Adjuvant mitotane versus surveillance in low-grade, localised adrenocortical carcinoma (ADIUVO): an international, multicentre, open-label, randomised, phase 3 trial and observational study. Lancet Diabetes Endocrinol. 2023;11:720-730.
6. Else T, Kim AC, Sabolch A, et al. Adrenocortical carcinoma. Endocr Rev. 2014;35(2):282-326.
7. Megerle F, Herrmann W, Schloetelburg W, et al. Mitotane mono- therapy in patients with advanced adrenocortical carcinoma. J Clin Endocrinol Metab. 2018;103(4):1686-1695.
8. Paragliola RM, Torino F, Papi G, Locantore P, Pontecorvi A, Corsello SM. Role of mitotane in adrenocortical carcinoma- review and state of the art. Eur Endocrinol. 2018;14(2):62-66.
9. Daffara F, De Francia S, Reimondo G, et al. Prospective evaluation of mitotane toxicity in adrenocortical cancer patients treated adjuv- antly. Endocr Relat Cancer. 2008;15(4):1043-1053.
10. Hermsen IG, Fassnacht M, Terzolo M, et al. Plasma concentrations of o,p’DDD, o,p’DDA, and o,p’DDE as predictors of tumor re- sponse to mitotane in adrenocortical carcinoma: results of a retro- spective ENS@T multicenter study. J Clin Endocrinol Metab. 2011;96(6):1844-1851.
11. Arshad U, Taubert M, Kurlbaum M, et al. Enzyme autoinduction by mitotane supported by population pharmacokinetic modelling in a large cohort of adrenocortical carcinoma patients. Eur J Endocrinol. 2018;179(5):287-297.
12. Fassnacht M, Terzolo M, Allolio B, et al. Combination chemother- apy in advanced adrenocortical carcinoma. N Engl J Med. 2012;366(23):2189-2197.
13. Hahner S, Fassnacht M. Mitotane for adrenocortical carcinoma treatment. Curr Opin Investig Drugs. 2005;6(4):386-394.
14. Lo Iacono M, Puglisi S, Perotti P, et al. Molecular mechanisms of mitotane action in adrenocortical cancer based on in vitro studies. Cancers (Basel). 2021;13(21):5255.
15. Hescot S, Slama A, Lombès A, et al. Mitotane alters mitochondrial respiratory chain activity by inducing cytochrome c oxidase defect in human adrenocortical cells. Endocr Relat Cancer. 2013;20(3): 371-381.
16. Sbiera S, Leich E, Liebisch G, et al. Mitotane inhibits sterol-O-acyl transferase 1 triggering lipid-mediated endoplasmic reticulum stress and apoptosis in adrenocortical carcinoma cells. Endocrinology. 2015;156(11):3895-3908.
17. Smith DC, Kroiss M, Kebebew E, et al. A phase 1 study of nevanimibe HCl, a novel adrenal-specific sterol O-acyltransferase 1 (SOAT1) in- hibitor, in adrenocortical carcinoma. Invest New Drugs. 2020;38(5): 1421-1429.
18. Weigand I, Altieri B, Lacombe AMF, et al. Expression of SOAT1 in adrenocortical carcinoma and response to mitotane monotherapy: an ENSAT multicenter study. J Clin Endocrinol Metab. 2020; 105(8):dgaa293.
19. Warde KM, Lim YJ, Ribes Martinez E, et al. Mitotane targets lipid droplets to induce lipolysis in adrenocortical carcinoma. Endocrinology. 2022;163(9):bqac102.
20. Rogers MA, Liu J, Song B-L, Li B-L, Chang CC, Chang T-Y. Acyl-CoA:cholesterol acyltransferases (ACATs/SOATs): enzymes with multiple sterols as substrates and as activators. J Steroid Biochem Mol Biol. 2015;151:102-107.
21. Gao P, Yang W, Sun L. Mitochondria-associated endoplasmic re- ticulum membranes (MAMs) and their prospective roles in kidney disease. Oxid Med Cell Longev. 2020;2020:3120539.
22. Lak B, Li S, Belevich I, et al. Specific subdomain localization of ER resident proteins and membrane contact sites resolved by electron microscopy. Eur J Cell Biol. 2021;100(7-8):151180.
23. Harned TC, Stan RV, Cao Z, et al. Acute ACAT1/SOAT1 blockade increases MAM cholesterol and strengthens ER-mitochondria con- nectivity. Int J Mol Sci. 2023;24(6):5525.
24. Wieckowski MR, Giorgi C, Lebiedzinska M, Duszynski J, Pinton P. Isolation of mitochondria-associated membranes and mitochondria from animal tissues and cells. Nat Protoc. 2009;4(11):1582-1590.
25. Verfaillie T, Rubio N, Garg AD, et al. PERK is required at the ER-mitochondrial contact sites to convey apoptosis after ROS-based ER stress. Cell Death Differ. 2012;19(11):1880-1891.
26. Weigand I, Schreiner J, Röhrig F, et al. Active steroid hormone synthesis renders adrenocortical cells highly susceptible to type II ferroptosis induction. Cell Death Dis. 2020;11(3):192.
27. Ingold I, Berndt C, Schmitt S, et al. Selenium utilization by GPX4 is required to prevent hydroperoxide-induced ferroptosis. Cell. 2018;172(3):409-22.e21.
28. Smith ME, Saraceno GE, Capani F, Castilla R. Long-chain acyl-CoA synthetase 4 is regulated by phosphorylation. Biochem Biophys Res Commun. 2013;430(1):272-277.
29. Doll S, Proneth B, Tyurina YY, et al. ACSL4 dictates ferroptosis sensitivity by shaping cellular lipid composition. Nat Chem Biol. 2017;13(1):91-98.
30. Doghman-Bouguerra M, Granatiero V, Sbiera S, et al. FATE1 an- tagonizes calcium- and drug-induced apoptosis by uncoupling ER and mitochondria. EMBO Rep. 2016;17(9):1264-1280.
31. Lynes EM, Raturi A, Shenkman M, et al. Palmitoylation is the switch that assigns calnexin to quality control or ER Ca2+ signal- ing. J Cell Sci. 2013;126(Pt 17):3893-3903.
32. Schaller E, Lamer S, Schlosser A, Stigloher C, Maher P, Decker M. Affinity-based protein profiling reveals IDH2 as a mitochondrial target of cannabinol in receptor-independent neuroprotection. Chemistry. 2025;31(33):e202501143.
33. Löfgren L, Forsberg GB, Ståhlman M. The BUME method: a new rapid and simple chloroform-free method for total lipid extraction of animal tissue. Sci Rep. 2016;6(1):27688.
34. Cox J, Hein MY, Luber CA, Paron I, Nagaraj N, Mann M. Accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ. Mol Cell Proteomics. 2014;13(9):2513-2526.
35. Cox J, Mann M. MaxQuant enables high peptide identification rates, individualized p.p.b .- range mass accuracies and proteome- wide protein quantification. Nat Biotechnol. 2008;26(12): 1367-1372.
36. Krüger AF, Schmitz W, Lamer S, et al. Supplemental Data Adrenocortical Mitochondria-Associated Membranes: Isolation, Characterization, and Lipidoproteomic Response to Mitotane. 2025. https://osf.io/rb6pt/overview?view_only=7ab3a22cb41d4a10ba9c158 0614c5ff5
37. Lynes EM, Bui M, Yap MC, et al. Palmitoylated TMX and calnexin target to the mitochondria-associated membrane. EMBO J. 2012;31(2):457-470.
38. Martin LA, Kennedy BE, Karten B. Mitochondrial cholesterol: mechanisms of import and effects on mitochondrial function. J Bioenerg Biomembr. 2016;48(2):137-151.
39. Arakane F, Sugawara T, Nishino H, et al. Steroidogenic acute regu- latory protein (StAR) retains activity in the absence of its mitochon- drial import sequence: implications for the mechanism of StAR action. Proc Natl Acad Sci U S A. 1996;93(24):13731-13736.
40. Prasad M, Pawlak KJ, Burak WE, et al. Mitochondrial metabolic regulation by GRP78. Sci Adv. 2017;3(2):e1602038.
41. Prasad M, Walker AN, Kaur J, et al. Endoplasmic reticulum stress enhances mitochondrial metabolic activity in mammalian adrenals and gonads. Mol Cell Biol. 2016;36(24):3058-3074.
42. Guengerich FP, McCarty KD, Tateishi Y, Liu L. Steroid 17a-hydroxylase/17, 20-lyase (cytochrome P450 17A1). Methods Enzymol. 2023;689:39-63.
43. Lu X, Gong Y, Hu W, et al. Ultrastructural and proteomic profiling of mitochondria-associated endoplasmic reticulum membranes reveal aging signatures in striated muscle. Cell Death Dis. 2022;13(4):296.
44. Doghman-Bouguerra M, Lalli E. ER-mitochondria interactions: both strength and weakness within cancer cells. Biochim Biophys Acta Mol Cell Res. 2019;1866(4):650-662.
45. Paskevicius T, Farraj RA, Michalak M, Agellon LB. Calnexin, more than just a molecular chaperone. Cells. 2023;12(3):403.
46. Kumar V, Maity S. ER Stress-sensor proteins and ER-mitochondrial crosstalk-signaling beyond (ER) stress response. Biomolecules. 2021;11(2):173.
47. Giorgi C, Missiroli S, Patergnani S, Duszynski J, Wieckowski MR, Pinton P. Mitochondria-associated membranes: composition, mo- lecular mechanisms, and physiopathological implications. Antioxid Redox Signal. 2015;22(12):995-1019.
48. Doll S, Freitas FP, Shah R, et al. FSP1 is a glutathione-independent ferroptosis suppressor. Nature. 2019;575(7784):693-698.
49. Ye B, Liao D, Zhang X, Zhang P, Dong H, Huganir RL. GRASP-1: a neuronal RasGEF associated with the AMPA receptor/GRIP com- plex. Neuron. 2000;26(3):603-617.
50. Weston CR, Davis RJ. The JNK signal transduction pathway. Curr Opin Cell Biol. 2007;19(2):142-149.
51. Scheidt HA, Haralampiev I, Theisgen S, et al. The adrenal specific toxicant mitotane directly interacts with lipid membranes and alters membrane properties depending on lipid composition. Mol Cell Endocrinol. 2016;428:68-81.
52. Hantel C, Shapiro I, Poli G, et al. Targeting heterogeneity of adreno- cortical carcinoma: evaluation and extension of preclinical tumor models to improve clinical translation. Oncotarget. 2016;7(48): 79292-79304.
53. Kiseljak-Vassiliades K, Zhang Y, Bagby SM, et al. Development of new preclinical models to advance adrenocortical carcinoma re- search. Endocr Relat Cancer. 2018;25(4):437-451.