Unless specified otherwise, all reagents were purchased from Merck Life Science (Czech Republic).
Patients and samples
In total, 22 PPGL samples were included in this study. These samples included 8 primary PGLs (paragangliomas), 11 primary PHEOs (pheochromocytoma), 1 recurrent PHEO, 1 recurrent and metastatic PGL, and 1 metastatic PGL. PPGLs included in this study were surgically resected from enrolled patients and evaluated under a protocol approved by the Eunice Kennedy Shriver National Institute of Child Health and Human Development, NIH, USA Institutional Review Board (ClinicalTrials.gov Identifier: NCT00004847). Informed consent was obtained from all patients for clinical, genetic, biochemical, and imaging studies performed as part of the investigation. This study was conducted in accordance with the principles of the Declaration of Helsinki. Tumors were resected between August, 2006 and December, 2017, and all were histopathologically confirmed as PHEOs or PGLs by the NCI Laboratory of Pathology (Bethesda, MD, USA). Fresh tumor samples were stored and frozen within 1 h of surgical resection at −80 °C.
Adrenal glands were obtained from patients with primary aldosteronism who underwent curative adrenalectomies or from those with renal tumors who underwent radical nephrectomy, including adrenal glands. All adrenal samples were inspected by a skilled pathologist to exclude samples with potential pathological changes. Portions of adrenals were immediately stored and frozen at −80 °C. The adrenal medullas were manually dissected under stereomicroscope and stored at −80 °C. Pooled samples of chromaffin tissue were generated by combining isolated adrenal medulla dissected from 10 different individuals each, controlled for sex and surgical indication (nephrectomy vs. adrenalectomy). All participants provided informed consent for inclusion in this study. The study was conducted in accordance with the Declaration of Helsinki, and the protocols were approved by the Ethics Committee of the Faculty Hospital and Medical Faculty of Palacky University in Olomouc (approval no. 13/14) and the Ethics Committee of General University Hospital and First Faculty of Medicine, Prague (approval no. 120/14).
Mutation screening
Genetic testing for PPGL susceptibility genes was performed for each patient as part of their NIH clinical research evaluation or at their referral institution. Germline and somatic mutations in PPGL susceptibility genes are summarized in Table 1. Eighteen of the 22 patients had germline mutations in PPGL susceptibility genes. In nine patients, SDHB mutations were found, five had VHL mutations, and four had RET mutations. One patient (no. 15) carried a somatic mutation in the EPAS1/HIF2A gene and was negative for 19 germline mutations (National Heart, Lung, and Blood Institute [NHLBI] testing panel: RET, MAX, VHL, SDHA/B/C/D/AF2, TMEM127, NF1, KIF1Bbeta, EGLN1, EGLN2, K-RAS, IDH1, IDH2, FH, MDH2, and HIF2A). Three patients (Nos. 20, 21, and 22) were negative for mutations in well-known PPGL susceptibility genes. Tumors from two of these patients (Nos. 20 and 22) were concluded to be sporadic based on a negative family history for PPGL and negative genetic testing; the first included eight genes (Genetic PPGL Mayo Testing: RET, MAX, VHL, SDHB/C/D/AF2, and TMEM127) and the second included 19 genes (NHLBI testing: RET, MAX, VHL, SDHA/B/C/D/AF2, TMEM127, NF1, KIF1Bbeta, EGLN1, EGLN2, K-RAS, IDH1, IDH2, FH, MDH2, and HIF2A). The tumor from the third patient (No. 21) was also considered sporadic after negative genetic testing for RET (Genetic PPGL Mayo Testing) due to clinical suspicion of multiple endocrine neoplasia (MEN), despite having a somatic CTNBB1 somatic mutation (currently not implicated in hereditary PPGL) somatic mutation. For the remaining three patients (Nos. 10, 13, and 16), genetic testing reports were not available, but their clinical presentation and gene mutations listed and described in the clinical notes were conclusive for mutations attributed to PPGLs. Only two tumor samples (Nos. 15 and 21) were tested for somatic mutations.
Proteomic analyses
High pH–trypsin–cyanogen bromide (hpTC) method
hPTC method was applied in the same way as described in [8]. Briefly, approximately 100 mg of tissue pulverized in liquid nitrogen was kept in hypotonic buffer (10 mM NaCl, 2 mM MgCl2, 10 mM HEPES, pH 7.4) for 15 min and passed through a 20 G hypodermic needle repeatedly. The homogenate was centrifuged at 500 × g 5 min, and the supernatant was treated with 120 Kunitz units of bovine deoxyribonuclease I with 25 mM MgCl2 and 5 mM CaCl2 for 30 min at 37 °C, then centrifuged at 20,000 × g, 4 °C. The resuspended pellet was shaken in 100 mM Na2CO3 with 1 mM EDTA for 30 min on ice, and this step was repeated once. The pellets were resuspended again in 50 mM NH₄HCO₃ with 20 µg trypsin (Promega) and incubated at 37°C overnight. The suspension was pelleted and washed with Na2CO3 twice and three times, snap-frozen on dry ice, and thawed between the washes. The pellet was solubilized in 70% trifluoroacetic acid (TFA) with CNBr (20 mg/mL) and incubated in the dark at room temperature overnight. The evaporated material was solubilized in 80% acetonitrile, 10% isopropanol, and 5% formic acid (FA), and diluted 1:10 with 0.5% FA. The sample was desalted delipidated on an OptiTrap column (Optimize Technologies): flushed with 0.5 mL 0.5% FA, then 2.5 mL dichloromethane with 0.5% FA, then 0.5 ml 0.5% FA, and eluted with 80% acetonitrile, 10% isopropanol, 0.5% FA.
Solid phase enrichment of N-linked glycopeptides (SPEG)
The method was used as in [9]. Briefly, approximately 100 mg of tissue pulverized in liquid nitrogen was resuspended in a high-salt buffer (2 M NaCl, 1 mM EDTA, 10 mM HEPES-NaOH, pH = 7.4), sonicated, and centrifuged at 20,000 × g at 4 °C for 30 min. The pellet was resuspended and washed twice with 100 mM Na2CO3 and 1 mM EDTA (as in hpTC). This membrane fraction was digested according to [10]: the pellet was solubilized in 5% sodium deoxycholate (SDC) in 100 mM NH4HCO3, kept for 15 min at room temperature, resuspended, and sonicated. The insoluble debris was removed by centrifugation at 10,000 × g for 5 min. The sample was reduced and alkylated (20 mM dithiothreitol, 45 mM iodoacetamide), and the sample was diluted to final 1% SDC and 50 mM NH4HCO3. Trypsin (Promega) was added at a 1:50 trypsin:protein ratio and the sample was kept at 37 °C overnight. After digestion, TFA was added to pH < 3, and SDC was removed by phase transfer (ethyl acetate added at 1:1 vol, 1 min shaking, and centrifugation at 15,000 × g for 2 min) which was 4 times repeated. The sample was desalted using an OptiTrap column (Optimize Technologies), according to the manufacturer’s instructions.
The desalted peptides were solubilized in 500 µL of Affi-Gel Hz Coupling Buffer (Bio-Rad), oxidized with 10 mM NaIO4 for 1 h in the dark, and quenched for 10 min by the addition of 20 mM Na2S2O3. Affi-Gel Hz Hydrazide Gel beads (Bio-Rad) were washed ddH2O and twice in Coupling Buffer, the sample was added, and the beads were rotated overnight. The beads were then washed thoroughly (1.5 M NaCl, 80% acetonitrile, and 50 mM NH4HCO3, 1 ml each) and 3 units of peptide N-glycosidase F (PNGase F, Roche) were added to 25 µL of 50 mM NH4HCO3. The beads were rotated at 37 °C overnight. The supernatant and additional washes (2×50 µL of 50 mM NH4HCO3, 40 µL of 0.5 M NaCl and 40 µL of 80% acetonitrile) were collected, and the sample was desalted using OptiTrap column again.
N-glycopeptide enrichment with filter-aided sample preparation (N-glyco-FASP)
The initial protein digest of the membrane fraction was prepared in the same manner as for SPEG. The desalted peptides were solubilized in binding buffer (0.5 M NaCl, 1 mM MnCl2, 1 mM CaCl2, 20 mM tris-HCl, pH = 7.6) and 100 μg of wheat germ agglutinin, 100 μg of concanavalin A and 80 of μg Ricinus communis agglutinin I (RCA120) were added. The mixture was kept for 1 h on a rocker on a 10 kDa ultrafilter spin column (Microcon Ultracel-10 Membrane, 10 kDa), washed four times with 200 μL of binding buffer, twice with 50 mM NH4HCO3 and kept in 40 μl of 50 mM NH4HCO3 with 2 units of PNGase F at 37 °C overnight. The filter flow-through was collected by centrifugation together with additional washes (2×50 μL of 50 mM NH4HCO3 and 40 μL of 0.5 M NaCl). The samples were desalted using an OptiTrap column.
SDC-trypsin
Approximately 10 mg of tissue was pulverized in liquid nitrogen and digested according to [10]. The material was resuspended, sonicated in 5% SDC in 100 mM NH4HCO3 and digested the same way as described above (see SPEG).
nLC-MS/MS Analysis
The peptide samples were separated on a reversed-phase nano column (EASY-Spray column, 50 cm × 75 µm ID, PepMap C18, 2 µm particles, 100 Å pore size). Mobile phase buffer A was composed of water, 2% acetonitrile, and 0.1% formic acid. Mobile phase B was composed of 80% acetonitrile, 0.1% formic acid. Samples were loaded onto a trap column (Acclaim PepMap300, C18, 5 µm, 300 Å Wide Pore, 300 µm × 5 mm, 5 Cartridges) for 4 min at 15 μl/min, with a loading solution composed of water, 2% acetonitrile, and 0.1% TFA. After 4 min, the valve was switched, and mobile phase B was increased from 4% to 35% B in 120 min at 300 nl/min, followed by a wash with 75% B 5 min at 400 nl/min, and then 4% B for 5 min until the end of the run (for all samples except hpTC). For the hpTC samples, the mobile phase B was increased from 4% to 50% B in 120 min at 300 nl/min, followed by a wash with 75% B 5 min at 400 nl/min, and then 4% B for 5 min until the end of the run. Eluting peptide cations were converted to gas-phase ions by electrospray ionization and analyzed using a Thermo Orbitrap Fusion (Q-OT-qIT, Thermo). Survey scans of peptide precursors from 350 to 1400 m/z were performed at 120 K resolution (200 m/z) with a 5 × 105 ion count target. Tandem MS was performed by isolation at 1.5 Th with quadrupole, HCD fragmentation with a normalized collision energy of 30, and rapid-scan MS analysis in the ion trap. The MS2 ion count target was set to 104, and the maximum injection time was 35 ms. Only precursors with charge states 2–6 were sampled for MS2. The dynamic exclusion duration was set to 45 s with a 10 ppm tolerance around the selected precursor and its isotopes. The monoisotopic precursor selection was performed. The instrument was run in the top-speed mode with 2 s cycles [11].
Data processing
Raw mass spectrometry data files were analyzed and quantified using MaxQuant v1.6.0.7. The data were searched against the human subset of the Swiss-Prot database (20,395 sequences, May 2021) using different settings for each method. Trypsin/P with 2 max missed cleavages was present in all methods; in hpTC, CNBr (cleavage C-terminal to methionine) with 2 max missed cleavage sites was set. For all methods, N-terminal protein acetylation and methionine oxidation were used as variable modifications. In glyco-capture (SPEG and N-glyco-FASP, searched together as two fractions for each sample), asparagine deamidation was set, and in hpTC, methionine (C-terminal to any peptide) substitutions to homoserine lactone (Met –48.003) and homoserine (Met –29.993) were set. Carbamidomethylation of cysteine was set as a fixed modification in all methods except hpTC. The false discovery rate (FDR) was set to 1% for both, proteins and PSM.
Using these parameters, label-free quantification (LFQ) searches were performed, hpTC, glyco-capture, and trypsin separately, for all sets of tumor samples (cluster 1, cluster 2, and unassigned) with the control samples. In the resulting datasets, the LFQ intensity values were further normalized such that the sum of all intensities was equivalent to the average sum of the original intensities in the dataset. The datasets were further processed using Perseus v1.6.2.3. Protein groups identified based on at least two unique peptides present in at least 80% of either tumor samples or control tissue were kept. 4-fold upregulated protein levels (p < 0.05) were considered significant.
To identify integral membrane proteins, deepTMHMM agorithm [12] was used to predict the transmembrane segments. The cell surface proteins were identified using the GO annotation GO:0005886 “Plasma membrane” via the function “Retrieve/ID mapping“ on UniProt (https://www.uniprot.org/).
Unsupervised hierarchical clustering was performed using Perseus v1.6.2.3, based on an LFQ search containing all the “SDC-trypsin” samples. The data were processed in the same way as above, missing values were imputed from normal distribution, data were subjected to ANOVA, significant values were normalized by Z-score, and the hierarchical clustering function was executed with its default parameters. The mixOmics 6.16.3 R package was used for supervised discrimination analysis (sPLS-DA).
SDS-PAGE and western blotting
The samples for SDS-PAGE and western blotting were prepared the same way as samples for SDC-trypsin prior to reduction, alkylation, and digestion, in 5% SDC and 100 mM NH4HCO3. The protein concentration was determined using a bicinchoninic assay. The samples were mixed with sample buffer (final 2% SDS, 10% glycerol, and 5% 2-mercaptoethanol). The samples were separated on 10% gels with 5% stacking front, containing 29:1 acrylamide:bis solution (Bio-Rad) in Mini-Protean TetraCell module (Bio-Rad) in Tris/Glycine/SDS buffer (Bio-Rad). The electrophoreses were transferred to Trans-Blot Turbo Mini 0.2 µm PVDF membranes in the Trans-Blot Turbo Transfer System (Bio-Rad), washed in PBS with 0.1% Tween 20 (PBST), blocked in 5% (w/v) skimmed milk in PBST for 30 min, washed 3 times/5 min in PBST, sealed in small volume with primary antibody, and incubated for 15 min at room temperature and at 4 °C overnight. The membranes were washed 3 times/10 minutes in PBST, incubated with horseradish peroxidase-conjugated secondary antibody for 30 min, washed 3 times/10 minutes in PBST, and once in PBS. The membranes were incubated with KPL LumiGLO Chemiluminescent Substrate (SeraCare), and chemiluminescence was detected using the ChemiDoc MP System (Bio-Rad). The detected bands were quantified using ImageLab 6.0.1 (Bio-Rad) and normalized against total lane optical densities of Coomassie Brilliant Blue stained gel run in parallel with the gel used for western blotting.
The following antibodies were used at corresponding dilutions: anti-ACE, 1:100 (sc-23908; Santa Cruz); anti-ANO1, 1:1000 (MA5-16358, Invitrogen); anti-ARG2, 1:500 (ab137069, Abcam); anti-ATP8A1 1:1000 (21565-I-AP, Proteintech), anti-ATX 1:100 (sc-374222, Santa Cruz), anti-CD39, 1:1250 (ab108248, Abcam); anti-CD146, 1:200 (ab24577, Abcam); anti-CTR1, 1:500 (NBP1-91447, Novus Biologicals); anti-L1CAM, 1:100 (sc-53386, Santa Cruz); and anti-SHMT2, 1:1000 (HPA020543, Merck). Secondary antibodies were as follows: anti-mouse-HRP, 1:10 000 (115-035-003, Jackson) and anti-rabbit-HRP, 1:10 000 (711-035-152, Jackson).