Abstract
Pheochromocytoma (PCC) and abdominal paraganglioma (aPGL) (together abbreviated PPGL) frequently present with an underlying genetic event in a PPGL driver gene, and additional susceptibility genes are anticipated. Here, we re-analyzed whole-exome sequencing data for PCC patients and identified two patients with rare missense variants in the calcium voltage-gated channel subunit 1H gene (CACNA1H). CACNA1H variants were also found in the clinical setting in PCC patients using targeted sequencing and from analysis of The Cancer Genome Atlas database. In total, CACNA1H variants were found in six PCC cases. Three of these were constitutional, and two are known to have functional consequences on hormone production and gene expression in primary aldosteronism and aldosterone-producing adrenocortical adenoma. In general, PPGL exhibited reduced CACNA1H mRNA expression as compared to normal adrenal. Immunohistochemistry showed strong CACNA1H (CaV3.2) staining in adrenal medulla while PPGL typically had weak or negative staining. Reduced CACNA1H gene expression was especially pronounced in PCC compared to aPGL and in PPGL with cluster 2 kinase signaling phenotype. Furthermore, CACNA1H levels correlated with HIF1A and HIF2A. Moreover, TCGA data revealed a correlation between CACNA1H methylation density and gene expression. Expression of rCacna1h in PC12 cells induced differential protein expression profiles, determined by mass spectrometry, as well as a shift in the membrane potential where maximum calcium currents were observed, as determined by electrophysiology. The findings suggest the involvement of CACNA1H/CaV3.2 in pheochromocytoma development and establish a potential link between the etiology of adrenomedullary and adrenocortical tumor development.
Introduction
Pheochromocytoma (PCC) is a neuroendocrine tumor that arises from chromaffin cells in the adrenal medulla, and abdominal paraganglioma (aPGL) exhibits a highly related cellular origin (Mete et al. 2022). Based on their similarities, PCC and aPGL are frequently grouped together as PPGL. Affected patients commonly exhibit symptoms due to increased catecholamine production. PPGLs have a prominent genetic background, as up to 40% of cases carry a constitutional mutation in one of the known susceptibility genes (Mete et al. 2022, Buffet et al. 2020, Juhlin et al. 2015, Hadrava Vanova et al. 2022), which partly overlap the somatic mutation spectra in PPGL (Liu et al. 2014, Welander et al. 2018, Mete et al. 2022).
Based on gene expression profiling, three molecular subgroups of PPGL have been identified (Fishbein et al. 2017, Juhlin 2021). Cluster 1 tumors demonstrate a pseudohypoxia expression profile, exhibit a hypermethylation phenotype, are more prone to metastasize, and are more likely of extra-adrenal origin (Fishbein et al. 2017, Jochmanova & Pacak 2018). Cluster 2 tumors (Fishbein et al. 2017, Jochmanova & Pacak 2018) display activation of kinase signaling pathways, are typically of adrenal origin, and usually show a non-metastatic course with good patient outcomes, with the exception of tumors displaying ATRX mutations (Jochmanova & Pacak 2018). Furthermore, a Wnt-altered subtype has been recognized (Fishbein et al. 2017).
The alpha 1 subunit of the family of T-type low voltage-operated calcium channels (VOCCs) includes three isoforms encoded by the genes CACNA1G (CaV3.1), CACNA1H (CaV3.2), and CACNA1I (CaV3.3) (Weiss & Zamponi 2020). The alpha 1 subunit forms four complexes each with six segments (Fig. 1), of which segments 5 and 6 will form the pore needed for calcium influx (Scholl et al. 2015). These channels are involved in neural excitability and play a role in hormone release in neuroendocrine organs (Weiss & Zamponi 2020). VOCCs have important functions in physiological processes, and several channelopathies have been linked to mutations in the encoding genes (Weiss & Zamponi 2020). Constitutional CACNA1H mutations are an underlying cause of familial forms of primary hyperaldosteronism, characterized by hypertension at a young age due to increased calcium influx in adrenocortical cells of the zona glomerulosa (Scholl et al. 2015). Subsequently, somatic CACNA1H mutations have been found to underlie the development of subsets of sporadic aldosterone-producing adrenocortical adenomas (Nanba et al. 2020). In PC12 rat pheochromocytoma cells, hypoxia increases the expression of CACNA1H / CaV3.2 and its function, with an increase in CaV3.2-mediated calcium influx (Sellak et al. 2014). Furthermore, hypoxia-inducible factor 1 alpha (HIF1A) binds to a region of the CACNA1H promoter (Sellak et al. 2014), which is of particular interest for PPGL. In addition, cluster 1 tumors have a molecular background resulting in HIF activation, and HIF2A (EPAS1) is one of the known tumorigenic drivers in PPGL.
To further clarify the molecular genetic background of PPGL, we analyzed previously generated unpublished constitutional data from a next-generation sequencing study (Juhlin et al. 2015). Based on the identification of genetic variants in the VOCC CACNA1H, investigations were performed in a larger case series concerning gene and protein expression and their relation to clinical presentation and molecular phenotypes, as well as effects on protein expression profiles and calcium currents.
Material and methods
Discovery PCC cohort
Fifteen PCC cases from the Karolinska University Hospital were previously analyzed by whole-exome sequencing (WES). In a previous report, somatic mutations were found by comparing PCC tumors with matched normal tissue, and the data were mapped to the human reference genome (Juhlin et al. 2015).
The Karolinska PPGL cohort
Tumor tissue samples from 97 primary PPGL cases (85 PCC and 12 aPGL) operated on at Karolinska University Hospital were used in the study. The full clinical and genetic information is detailed in Supplementary Table 1, see section on supplementary materials given at the end of this article. Detailed data for the pheochromocytoma of the adrenal gland scaled score (PASS) have been published for the majority of cases (Stenman et al. 2019). All tumors were previously investigated for mutations in PPGL-related genes (Supplementary Table 1), and a subset of the tumors were subjected to WES analysis (Welander et al. 2014, Juhlin et al. 2015, Stenman et al. 2016a). Affymetrix-based mRNA expression profiles were previously generated for the majority of tumors (Stenman et al. 2016b, 2019). Based on the mutation status and/or mRNA expression profile, tumors were assigned as cluster 1 (SDHx, VHL, EPAS1, and EGLN1) or cluster 2 (RET, NF1, TMEM127, MAX, and HRAS).
Fresh frozen tumor tissue samples were collected from the local Biobank for all 97 PPGL. For 29 of the cases, formalin-fixed paraffin-embedded (FFPE) PPGL slides were also obtained for immunohistochemistry. Fresh frozen non-tumorous adrenal tissue samples were used as references for qRT-PCR. FFPE tissue sections from de-identified normal testis, normal pancreas, pancreatic neuroendocrine tumor (Pan-NET), and normal adrenal gland (from patients operated on for other metastatic lesions to the adrenal) were included as immunohistochemical controls.
The TCGA cohort
For the TCGA cohort of PPGL (Fishbein et al. 2017), the Firehose Legacy datasets used were downloaded from https://www.cbioportal.org/ in August 2020. Only primary tumors with a histological diagnosis of PCC or aPGL were selected. Head and neck PGL were excluded, and only one sample per patient was enrolled. The resulting cohort consisted of 177 samples (148 PCC and 29 aPGL) from 100 females and 77 males, with a median age of 46 years at diagnosis (range 19–83 years). Fifteen patients had recurrent or metastatic disease. Methylation density values for CACNA1H were downloaded from UCSC Xena Functional Genomics Explorer https://xenabrowser.net/ in September 2020. Data for mRNA expression clusters 1 or 2, and DNA methylator cluster (hyper-methylated M1 or intermediate/low-methylated non-M1) were obtained from Fishbein et al. (2017).
The following data were obtained and analyzed: CACNA1H variants and mutation status of known PPGL susceptibility genes for the three cases with a CACNA1H variant; mRNA expression profiles; CACNA1H mRNA expression; CACNA1H DNA methylation; CACNA1H copy number status; HIF1A mRNA expression; EPAS1 (HIF2A) mRNA expression; mRNA cluster M1/non-M1 phenotype; and clinical data for patient sex, tumor type, tumor weight, metastasis, and absolute outcome.
Analysis of next-generation sequencing data
WES data were used from the previously published discovery cohort (Juhlin et al. 2015). In this study, candidate PPGL genes were searched for, this time focusing on variants present in both tumor and matched constitutional tissue but absent in the human reference genome and absent or rare in gnomAD. An additional case (E2) was detected by clinical routine sequencing via the Clinical Genetics Department at the Karolinska University Hospital. For case E2, a tissue sample from normal myocardium was obtained at autopsy with consent from the nearest of kin; however, the tumor was not available. WES analysis was performed with analysis of an established gene panel for endocrine tumors (Supplementary Table 3) as described in Supplementary Document (Lindstrand et al. 2019, Muth et al. 2019). Detected variants were assessed for predictions of deleterious variants and allele frequencies from gnomAD, as detailed in Supplementary Document.
Sanger sequencing
Ninety-five PPGL were screened for variants in “hotspot” areas of exons 16, 17, 22, 25, and 34 of the CACNA1H gene using primers detailed in Supplementary Table 4 and experimental procedures described in Supplementary Document.
Quantitative real-time polymerase chain reaction (qRT-PCR)
CACNA1H mRNA expression was successfully analyzed by qRT-PCR in 83 PPGL and eight normal adrenal tissue samples using assays for CACNA1H (Hs00234934, Thermo Fisher) and the housekeeping gene B2M (Hs99999907, Thermo Fisher), following experimental procedures described in Supplementary Document.
Immunohistochemistry
CACNA1H protein expression was investigated by immunohistochemistry in 29 PPGL cases (20 PCC and nine PGL; Supplementary Table 5) as described in detail in Supplemental Document. When available, normal adrenal medulla and/or cortex on the same slide was used for comparison. Normal testis, normal pancreas, a Pan-NET, and normal adrenal were analyzed in parallel as controls. Slides were incubated with primary antibody against the T-type Ca++ CP α1H (G-10, sc-377510, Santa Cruz Biotechnology) at a dilution 1:400. The results were evaluated by a pathologist (CCJ) blinded to clinical and genetic data, and cytoplasmatic immunoreactivity was graded as: minimal or no staining (0); mixed staining pattern with negatively and positively stained cells (±); weak staining intensity (+); or strong staining intensity (++).
Cell culture, plasmids, and transfection
The established rat adrenal pheochromocytoma cell line PC12 was used for overexpression of rat rCacna1h and analyzed as detailed in Supplementary Document.
Mass spectrometry, data analyses, and protein ontology analyses
Mass spectrometry analysis of PC12 cells transfected with wild-type rCacna1h plasmid and control vector plasmid was performed at the Proteomics Biomedicum core facility at Karolinska Institutet, applying previously reported procedures (Shafi et al. 2023) with modifications as detailed in Supplementary Document.
Electrophysiology
For electrophysiological experiments, the analyses largely followed previously published experimental procedures (Lu et al. 2010) with modifications as described in Supplementary Document. PC12 cells transfected with wild-type rCacna1h plasmid and control vector plasmid were incubated in petri dishes at 37˚C and 5% CO2 overnight, and VOCCs activity was recorded using the patch-clamp technique (Hamill et al. 1981).
Statistical analyses and illustrations
IBM SPSS Statistics Version 25 and 26 were used for statistical analyses and graphic work, as described in Supplementary Document.
Results
Identification of CACNA1H variants
Previously generated and unpublished WES data of constitutional DNA for the 15 PCC cases in the discovery cohort were investigated and compared to the public reference genome (Juhlin et al. 2015). This identified two cases (6 and 19) that carried CACNA1H missense variants, W1530S and G1064R, respectively (Table 1, Supplementary Table 2). Since constitutional CACNA1H variants have previously been associated with primary hyperaldosteronism (Scholl et al. 2015), CACNA1H was selected for further analysis. One additional constitutional CACNA1H variant was subsequently found in a PCC patient from WES with analysis of an endocrine tumor gene panel (Supplementary Table 3) in the clinical routine setting during the study period. An indel frameshift A1989Rfs*33 was detected in constitutional tissue from this case E2 (Fig. 1, Table 1). Investigation of the TCGA dataset of 177 PPGL tumors identified three additional CACNA1H variants, two of which were missense variants (I1430T and L1447M) and one frameshift deletion (E1259Gfs*85) (Table 1).
Details of CACNA1H variants detected in human PCC cases in the Karolinska cohort and the TCGA database.
Case number | Type of variant | Exon no. | Genetic alteration | Protein effect | Detected in | Location in CACNA1Ha | Reported in/functional consequencesb | ||
---|---|---|---|---|---|---|---|---|---|
Normal | Tumor | ||||||||
Karolinska cohort | |||||||||
6 | Missense | 25 | c.4589G>C | W1530S | Blood | Prim tumor | Membrane | Repeat III-S6 (pore-forming) | – |
19 | Missense | 16 | c.3190G>A | G1064R | Blood | Prim tumor | Cytoplasm | Between repeat II and III | Primary hyperaldosteronism/aldosterone production |
E2 | FS del | 34 | c.5965_5987del | A1989Rfs*33 | Myocardium | – | Cytoplasm | After repeat IV | – |
TCGA database | |||||||||
WB-A81D | Missense | 22 | c.4289T>C | I1430T | Not reported | Prim tumor | Membrane | Repeat III-S5 (pore-forming) | Aldosterone-producing adrenocortical adenoma (S)/aldosterone production |
QR-A70U | Missense | 22 | c.4339T>A | L1447M | Not reported | Prim tumor | Membrane | Repeat III-S5 (pore-forming) | – |
S7-A7WP | FS del | 18 | c.3776del | E1259Gfs*85 | Not reported | Prim tumor | Cytoplasm | Between repeat II and III | – |
TCGA database and Fishbein et al. 2017 by comparing tumor and normal tissue. Reference sequence NM_021098.3. aAccording to Uniprot.org; b Roomp et al. 2016, Nanba et al. 2020.
FS del, frameshift deletion; Prim tumor, primary tumor; S, somatics.
The Karolinska cohort of 95 PPGL tumors was successfully screened for CACNA1H variants by Sanger sequencing of five focal areas of exons 16, 17, 22, 25, and 34 based on the localization of the CACNA1H variants (Fig. 1) found previously. The investigation verified the presence of the tumor DNA of the CACNA1H variants initially found by WES in cases 6 and 19 but did not reveal additional deleterious variants.
In total, six potentially damaging CACNA1H variants were found in the Karolinska cohort and in the TCGA database (Table 1). Two variants (W1530S and G1064R) have been reported in gnomAD at frequencies less than 0.0001, while the other four variants (A1989Rfs*33, I1430T, L1447M, and E1259Gfs*85) are not known in gnomAD, suggesting they are very uncommon in the general population (Table 2). In silico predictions and reported frequencies are summarized in Table 2, which revealed that the missense variants were predicted as pathogenic by two or more prediction tools. Based on WES data from tumor and/or constitutional DNA, mutations in known PPGL genes were not revealed in any of the six cases (Table 3).
In silico predictions and reported frequencies of detected CACNA1H variants.
W1530S | G1064R | A1989Rfs*33 | I1430T | L1447M | E1259Gfs*85 | |
---|---|---|---|---|---|---|
Cohort | Karolinska | Karolinska | Karolinska | TCGA | TCGA | TCGA |
Case no. | 6 | 19 | E2 | WB-A81D | QR-A70U | S7-A7WP |
DNA alteration | c.4589G>C | c.3190G>A | c.5965_5987del | c.4289T>C | c.4339T>A | c.3776del |
Conserved amino acid | Highly | Highly | – | Highly | Highly | – |
In silico predictions | ||||||
Mutation Taster (v2021) | Deleterious | Deleterious | Deleterious | Deleterious | Benign | Deleterious |
Tree vote (del/benign) | 75/25 | 69/31 | – | 86/14 | 27/73 | – |
PolyPhen2 | Probably damaging | Probably damaging | – | Probably damaging | Probably damaging | – |
Score | 0.998 | 1.000 | – | 0.998 | 0.999 | – |
CADD (v1.6) phred score | 29.4 | 28.9 | – | 25.1 | 21.8 | – |
SIFT (v6.2.0) | Deleterious | Deleterious | – | Deleterious | Deleterious | – |
Score | 0.00 | 0.00 | – | 0.00 | 0.00 | – |
Align GVGD (v2007) | Benign | Likely benign | – | Damaging | Benign | – |
Class | C0 | C15 | – | C65 | C0 | – |
Grantham distance | Large | Moderate | – | Moderate | Small | – |
gnomAD | ||||||
Variant ID | 16-1261968-G-C | 16-1258048-G-A | Not reported | Not reported | Not reported | Not reported |
Frequency all | 0.0016% | 0.0051% | None | None | None | None |
EUR allele frequency | 0.0035% | 0.0075% | None | None | None | None |
EUR carrier frequency | 1 in 14,500 | 1 in 7000 | None | None | None | None |
Reference sequence NM_021098.3.
FS del, frameshift deletion.
Clinical details of human PCC cases with detected CACNA1H variants.
Panel number | Age (years) | Sex (M/F) | Diagnosis | Size (mm) | Malignant behavior | Follow-up | Other tumors | Other diseases and features | Wild-type PPGL genes | ||
---|---|---|---|---|---|---|---|---|---|---|---|
Time | Outcome | Constitutional | Tumor | ||||||||
Karolinska cohort | |||||||||||
6 | 44 | F | PCC | 20 | No | 16.7 y | Alive | Breast cancer | Anosmia, toxic MNG | By WESa | By WESa |
19 | 37 | M | PCC | 100 | No | 27 y | Alive | Seminoma neurofibromas | Vertigo | By WESa | By WESa |
E2 | 51 | M | PCC | 45 | No | 0 y | Dead | – | –– | By WES-panelb | – |
TCGA database | |||||||||||
TCGA-WB-A81D | 82 | F | PCC | 75 | No | 4 y | Alive | – |
– |
By WESc | By WESc |
TCGA-QR-A70U | 46 | F | PCC | 57 | No | 4 y | Alive | – |
– |
By WESc | By WESc |
TCGA-S7-A7WP | 25 | F | PCC | 50 | No | 2 y | Alive | – |
– |
By WESc | By WESc |
Age, age at diagnosis; F, female; M, male; MNG, multinodular goiter; −, not reported; PCC, pheochromocytoma;WES, whole-exome sequencing; y, years.
aJuhlin et al. 2015; bGene panel detailed in the Supplementary Table 3; cFishbein et al. 2017.
All CACNA1H variants are summarized in Tables 1, 2, and3 and schematically illustrated in Fig. 1. All four missense variants are located in evolutionarily conserved amino acids (Supplementary Fig. 1). The missense variants W1530S, I1430T, and L1447M are located in the membrane, affecting the pore-forming segments 5 or 6 in repeat III, while G1064R is located in the cytoplasm (Fig. 1, Table 1). The two frameshift variants A1989Rfs*33 and E1259Gfs*85 occur at amino acids located in the cytoplasm, resulting in truncation/loss of amino acids after repeat IV and between repeats III/IV, respectively (Fig. 1). Two variants, G1064R and I1430T, have also been reported constitutionally in a family with primary hyperaldosteronism and as a recurrent somatic event in aldosterone-producing aldosteronoma, respectively (Roomp et al. 2016, Nanba et al. 2020).
Clinical details are summarized in Table 3 and described for cases 6, 19, and E2 in Supplementary Document. None of the three cases from Karolinska with a CACNA1H variant showed signs of primary hyperaldosteronism at the time of diagnosis, neither from clinical chemistry analyses nor at subsequent histopathology work-up.
Expression of CACNA1H / CaV3.2 is frequently suppressed in PPGL as compared to adrenal medulla
Quantification of CACNA1H mRNA expression in the Karolinska cohort showed that the tumor samples had lower expression compared to the adrenal references (P < 0.001, Mann–Whitney U-test, Fig. 2A). This finding also confirmed using microarray-based expressional data from a largely overlapping cohort, showing lower CACNA1H expression in PPGL compared to adrenal samples (P = 0.001, Mann–Whitney U-test, Fig. 2A).
CACNA1H/CaV3.2 protein expression was subsequently examined by immunohistochemistry in 29 PPGL cases and reference tissues. Examples of control tissues are shown for pan-NET (positive) and pancreas (negative) in Fig. 2E. In a subset of PPGL, the presence of adrenal medulla (12 samples) and adrenal cortex (18 samples) on the same slides permitted direct comparison of the staining pattern with tumor tissue (Fig. 2; Supplementary Table 5). Strong cytoplasmic staining (++) was found in the adrenal medulla for all 12 PPGL samples, while the adrenal cortex typically showed weak staining (+) (Fig. 2B and C). A similar staining pattern was observed in normal adrenal tissue from non-PPGL disease, with strong staining of the medulla together with weak staining of the cortex, including both zona glomerulosa and zona reticularis (Fig. 2D). Among the 29 PPGL, 15 samples had weak staining (+), seven showed varied staining with a mixture of negatively and positively stained cells (±), including the two cases with missense CACNA1H variants (Supplementary Fig. 2), and seven cases showed strong staining (++) in the tumor tissue.
CACNA1H methylation levels correlate with CACNA1H expression
A possible correlation between CACNA1H methylation density and gene expression levels was investigated using the TCGA dataset (Fig. 3). The mean methylation density across the CACNA1H gene locus was moderately correlated with CACNA1H gene expression (R = 0.453; P < 0.001, Fig. 3A). With regard to the individual CG sites, highly variable methylation density was observed, determined as a mean for each of the 177 PPGLs included (range 0.02–0.99; Fig. 3C; Supplementary Table 6), with the lowest methylation density observed within the promoter region. For 113 out of 187 CG sites, there was a significant correlation between CACNA1H mRNA expression and methylation density (P < 0.01). In total, 43 of these CG sites showed a positive correlation with R-values above 0.4, and three sites showed an inverse correlation with R-values below −0.4 (Supplementary Table 6).
We also compared CACNA1H expression between PPGL in different global methylation clusters using data obtained from Fishbein et al. (2017). This revealed that the M1 hypermethylated tumor group (n = 29) was associated with higher CACNA1H mRNA levels compared to the non-M1 tumors (n = 141) (P = 0.001; Mann–Whitney U-test; Fig. 3B).
Low CACNA1H mRNA levels are associated with PCC and cluster 2 phenotype
Comparison of CACNA1H gene expression levels in PCC and aPGL revealed lower levels in PCC as compared to aPGL in the TCGA cohort (P = 0.012, Mann–Whitney U-test), and a trend toward lower CACNA1H expression in PCC vs aPGL in the Karolinska cohort (Fig. 4A). Furthermore, in the Karolinska cohort, cases with normal norepinephrine levels had lower CACNA1H gene expression than cases with elevated norepinephrine levels (P = 0.009, Mann–Whitney U-test; Fig. 4B). No other correlations or associations were found in the Karolinska cohort between CACNA1H expression and the clinical parameters.
In the Karolinska cohort, cluster 2 tumors were found to have lower CACNA1H expression than cluster 1 tumors (P = 0.001, Mann–Whitney U-test, Fig. 4C). Similarly, in the TCGA cohort, cluster 2 tumors had lower CACNA1H expression than cluster 1 tumors (P < 0.001, Mann–Whitney U-test). Given that HIF is known to bind to a CACNA1H promoter region (Sellak et al. 2014), a possible correlation between CACNA1H and HIF1A expression was investigated using TCGA data, revealing a weak correlation (P = 0.008, R = 0.199, Spearman’s rank order correlation; Fig. 4D). Furthermore, a similar analysis of HIF2A (EPAS1, related to cluster 1) revealed a correlation between CACNA1H and HIF2A (EPAS1) mRNA expression levels (P < 0.001, R = 0.537, n = 177) (Fig. 4E).
Comparison of PPGL with CACNA1H variant or copy number alteration
To investigate possible functional effects of CACNA1H variants, we first used the TCGA dataset to compare RNA-Seq profiles between tumor groups. Altogether, 136 genes out of all 17,338 genes were significantly over-expressed, and 146 were under-expressed in the three cases with CACNA1H variants compared to the 174 CACNA1H wild-type cases. Enrichment analyses showed affected pathways related to neuronal excitability (Supplementary Fig. 3).
Additionally, copy number loss of CACNA1H was observed in 12 cases, while 141 were diploid and seven showed a gain. No significant difference in CACNA1H mRNA levels was observed between groups. In two tumors with CACNA1H variants, the CACNA1H locus was diploid and did not have any copy number loss, while the third sample´s copy number status is not known.
Differentially expressed proteins
Subsequently, we ectopically expressed wild-type rat rCacna1h and vector control in PC12 rat pheochromocytoma cells and compared the effects on global protein expression patterns and calcium currents by electrophysiology. Immunoblotting showed a very low/undetectable intrinsic level of rCacna1h in PC12 cells (Fig. 5A). Recombinant Cacna1h was transfected into PC12 cells, resulting in increased CACNA1H protein compared to the transfection control (Fig. 5A).
Using LC-MS/MS, a total of 3553 proteins were identified and quantified (Supplementary Table 7). Principal component analysis revealed the separation between samples in the two groups (Fig. 5B). Differentially expressed proteins (DEPs) were identified by statistical comparisons between rCacna1h-transfected and vector-control cells. Altogether, 335 DEPs with a P-value of 0.1 or below were found by t-test (186 up-regulated and 149 down-regulated, Supplementary Table 8). Fold-change analysis identified 59 DEPs (37 up-regulated and 22 down-regulated proteins) with a fold-change of 1.2 or more (Supplementary Table 9). Volcano plot analysis identified 37 DEPs as the most regulated proteins (26 up-regulated and 11 down-regulated, Supplementary Table 10). The clustering of DEPs is illustrated in a heatmap for the most regulated proteins (30 up-regulated and 20 down-regulated) in Fig. 5D (Supplementary Table 11). As illustrated in Fig. 5C and D, one of the top most up-regulated proteins was rCacna1h (accession number Q9EQ60), thus supporting the transfection efficiency. Among other DEPs, several have known roles in cancer (e.g. Dad1 and Brd9), adrenocortical tumors (Gnas), and cytochrome c oxidase subunits (e.g. Cox6c2 and Cox7a2l).
DEPs were assayed using the Reactome Pathway Database and String analysis. While no statistically significant associations between DEPs and annotated pathways were found when adjusting for the false discovery rate using Reactome, String analysis of up-regulated DEPs revealed enrichment of pathways related to rCacna1h and the potassium channel Kcnh3 (Fig. 5E).
Electrophysiology
Possible effects of CACNA1H overexpression were assessed by electrophysiological measurements of VOCC activity in PC12 cells using the patch-clamp technique (Fig. 6). rCacna1h-transfected PC12 cells and vector-transfected control cells were cultured in parallel in petri dishes 2–4 days after transfection. Cells were randomly selected in both groups’ petri dishes and clamped according to the voltage protocol presented in Fig. 6A. In total, 13 cells were analyzed in both groups, and all cells exhibited voltage-activated Ca2+ currents (Fig. 6C). Whole-cell currents were normalized to whole-cell capacitance (6.1 ± 0.7 pF in rCacna1h-transfected, 6.0 ± 0.5 pF in control cells, n.s.) with no detectable difference in peak current (188 ± 22 pA/pF vs 186 ± 5 pA/pF) (summary in Fig. 6D). However, rCacna1h-transfected PC12 cells displayed a slightly right-shifted I–V curve, with a peak current at a more positive potential. To further analyze the right shift, each I–V curve was fitted to a multi-variable regression model, and plotting the variable corresponding to the peak voltage for each group is shown in Fig. 6B.
Discussion
In this study, the genetic background of PPGL was further explored with the identification of rare CACNA1H gene variants in PCC patients, including two truncating variants. Moreover, supporting the role of CACNA1H in PPGL development, CACNA1H expression was shown to be suppressed in PPGL compared with normal adrenal medulla.
For the six PCC with potentially damaging CACNA1H variants, mutations in other known PPGL susceptibility genes were not identified, with the exception of one case (no. 6) with a constitutional variant G2735S of unknown significance in KMT2D (Table 3). In WES analyses that have been performed on tumor DNA and normal DNA from cases 6 and 19 (Juhlin et al. 2015) and the three TCGA cases (cBioPortal) (Fishbein et al. 2017), additional somatic or constitutional mutations in known PPGL genes were not observed. However, we cannot exclude additional somatic PPGL gene variants in case E2, and additional genes may be added to the growing list of PPGL genes.
Six potentially damaging CACNA1H variants were found by combined analyses of the Karolinska cohort and the TCGA database. In the TCGA cohort, the observed frequency of CACNA1H variants was 1.7% in PPGL and 2% in PCC. Taken together, the TCGA and Karolinska cohorts suggest a frequency of CACNA1H variants of 1.8% in PPGL and 2.2% in PCC.
In addition, Flynn et al. reported a somatic missense variant R2322Q/c.6965G>A predicted as benign in a PCC case with an NF1 mutation (Flynn et al. 2015).
Several observations support the role of CACNA1H variants in PPGL. Four of the detected CACNA1H variants were not present in gnomAD, and two were present at very low frequencies. Three of the four missense variants were predicted as damaging by multiple programs (Adzhubei et al. 2010), and the two frameshift deletions are also expected to be damaging. The missense variant L1447M demonstrated damaging predictions from two computational prediction tools, contrasting with benign predictions from alternative computational analyses. The variant G1064R is annotated in ClinVar with uncertain pathogenicity. Hence, elucidation of their pathogenicity awaits scrutiny through functional assays. Furthermore, all missense variants affect amino acids that are evolutionarily preserved in orthologs (Supplementary Fig. 1). Two of the missense variants have been reported in conditions affecting the adrenal cortex. Constitutional G1064R has been reported in a family with primary hyperaldosteronism, and the somatically occurring variant I1430T was observed in three aldosterone-producing adrenocortical adenomas. Both variants have been investigated in vitro, showing functional consequences of the investigated variant with effects on aldosterone production and gene expression (Roomp 2016, Nanba et al. 2020). Additional CACNA1H variants have been reported in primary hyperaldosteronism and aldosterone-producing adrenocortical adenoma (Fig. 1) (Wu et al. 2021, Tseng et al. 2022, Liu et al. 2021, Daniil et al. 2016). Of these, M1549V and M1549I are located close to W1530S observed in the present study, which both affect the pore-forming segment 6 in repeat III (Fig. 1). Similarly, L1447M and I1430T affect segment 5 of repeat III, which is also involved in the pore for calcium-flux. Also, V1937M and M1549V have been shown to have functional consequences in vitro and in mouse models, respectively (Gürtler et al. 2020, Tseng et al. 2022, Seidel et al. 2021). Hence, there is reason to predict the functional effects of the CACNA1H variants observed in PCC, especially for those assessed in primary hyperaldosteronism and aldosterone-producing adenoma, respectively. At the same time, none of the three cases with demonstrated constitutional CACNA1H variants presented signs of primary aldosteronism. This could possibly be related to reduced penetrance or the influence of other factors influencing the development of PCC or primary aldosteronism in carriers of CACNA1H variants. Additional investigations into the functional consequences of the detected variants could possibly clarify the role of CACNA1H in PPGL development.
CACNA1H gene expression was demonstrated by multiple methodologies to be generally suppressed in PPGL tumor tissue in comparison to the non-tumorous adrenal tissue consisting of both the adrenal medulla and cortex. Immunohistochemistry allows precise tissue observations, and in our cohort, the tumor tissue was generally less intense than the normal medulla, demonstrating that the protein is downregulated in the tumor.
In PC12 rat cells, CACNA1H is upregulated following cellular stress such as chronic hypoxia (Mahapatra et al. 2012), and hypoxia has been shown to increase the expression of CaV3.2 and its function due to the binding of HIF1A to the CACNA1H promoter region (Sellak et al. 2014). In our study, HIF1A and CACNA1H expressions were weakly correlated, and cluster 1 tumors, characterized by pseudohypoxia, were found to have higher expression of CACNA1H. In agreement, the expression of HIF2A (EPAS1) on which the pseudohypoxia phenotype is based, was correlated with CACNA1H mRNA levels. Based on mRNA expression profiles, five of the CACNA1H variant cases were cluster 2, while one was cluster 1, although without typical aggressive features or high methylation phenotype.
Additional mechanisms of CACNA1H suppression may therefore be operative in PPGL. The observation of correlations between DNA methylation density across the CACNA1H gene and its mRNA expression could suggest epigenetic suppression as an additional mechanism involved in the tumor-specific suppression. Of interest, CACNA1H has been reported as a differentially methylated gene distinguishing adrenocortical tumor subgroups (Clay et al. 2019).
The hypermethylator phenotype in PPGL recognizes the M1 subgroup with hypermethylation of multiple promoter CpGs as compared to non-M1 tumors, including CACNA1H (Letouzé et al. 2013). The methylation of three sites in the promoter region reported in Letouze et al. was not associated with decreased CACNA1H expression among M1 tumors (Letouzé et al. 2013). While the corresponding sites 45−47 are located in a region of generally low methylation when looking at all 177 PPGL, increased methylation for multiple sites of the gene is associated with increased CACNA1H expression (Fig. 3). In addition, CACNA1H levels were higher in PPGL with an M1 hypermethylator phenotype.
Gene enrichment analyses comparing tumors with CACNA1H variants and wild-type CACNA1H revealed a set of affected pathways (Supplementary Fig. 3). Interestingly, since CACNA1H is a regulator of neuronal excitability, this could potentially indicate a functional difference between the two groups.
In functional studies in HEK293 cells of the recurrent CACNA1H variant M1549V, reported in primary hyperaldosteronism, loss of normal channel inactivation and activation at different potentials were observed (Scholl et al. 2015). Depolarization through increased intracellular calcium ion concentrations in PC12 cells has been linked to changes in overall morphology, suggesting that an altered influx of calcium in PPGLs could influence tumoral behavior (Starikova et al. 2000). We, therefore, performed functional experiments overexpressing rat CACNA1H (rCacna1h) specifically to address if this particular calcium channel influences the electrophysiologic state of PC12 rat pheochromocytoma cells, as well as the overall proteome of these cells. Some of the observed DEPs are of particular interest for the adrenal gland. Among these can be mentioned GNAS Complex locus (GNAS) which is known to be mutated in a subset of adrenocortical tumors. Furthermore, we observed upregulation of cytochrome c oxidase subunits (e.g. Cox6c2 and Cox7a2l), which are part of the mitochondrial electron transport chain of central importance in PPGL. The patch-clamp experiments showed that the membrane potential for the peak current of Ca2+ currents was right shifted in PC12 cells overexpressing rCacna1h compared to control cells. However, no difference was seen in peak current or integrated current. Possible explanations for these unexpected findings are that transfection with rCacna1h affects the expression of other regulatory factors or Ca2+ channels. Another possible explanation is that in a population with a less than 100% transfection ratio, some of the randomly patched cells may be non-transfected, thereby underestimating the effect of the rCacna1h-transfected group. Indeed, our findings suggest that rCacna1h over-expressing PC12 cells exhibited a shift in the membrane potential for Ca2+ currents as well as changes in the proteome.
T-type low VOCCs lower the threshold of the action potential, thereby facilitating the cells to respond to sympathetic stimulation at lower voltage levels but also creating a continuous release of catecholamines at resting potential (Marcantoni et al. 2008). Scholl et al. compared CACNA1H missense mutations observed in primary aldosteronism to wild-type CACNA1H in a patch clamp experiment and observed that the mutated cells left-shifted, thus obtaining a lower threshold for activation and an almost ten-fold slower inactivation than wild-type channels, thereby increasing the secretion of aldosterone (Scholl et al. 2015). Conversely, when we overexpressed CACNA1H in PC12 cells, we observed a rightward shift in our patch clamp experiment. This shift suggests that the cells were less likely to respond to normal triggers because it takes a stronger electrical signal to reach the threshold required for activation. Hence, the reduced CACNA1H levels in tumors compared to normal and the truncating variants in our study could potentially confer a similar response to membrane potential as observed in primary aldosteronism, although it is important to emphasize that comparing site-directed mutagenesis studies and enforced overexpression experiments can be challenging.
The expression of CACNA1H in the adrenal coincides with tyrosine hydroxylase, a marker indicating chromaffin cells with catecholamine secretion (Levitsky & López-Barneo 2009). However, in normal adrenal medullary tissue, several types of calcium channels (L, N, P/Q, R, and T) are involved in chromaffin cell excitability and exocytosis of catecholamines (García-Palomero et al. 2000). In our study, we observed associations between CACNA1H expression and norepinephrine, as well as cluster 1 vs 2 phenotypes, which could theoretically reflect the difference in norepinephrine levels between cluster 1 and 2 cases and/or a causative relation. Cluster 1 PPGL exhibits an immature secretion apparatus with higher secretion of catecholamines (Eisenhofer et al. 2011). At the same time, we showed that in vitro expression of CACNA1H led to a higher threshold for activation. This discrepancy could partly be explained by the diversity in channels involved in the secretion (such as L, N, P/Q, R, and T). To fully explain the relationships between the different calcium channels in chromaffin cells, specifically CACNA1H, further analyses at the individual cell level would have to be made.
In conclusion, the findings suggest a role for CACNA1H/CaV3.2 in the development of PPGL. Genetic CACNA1H variants are reported, and CACNA1H/CaV3.2 expression was shown to be generally suppressed in pheochromocytoma/abdominal paragangliomas compared to non-tumorous adrenal medulla, with possible relation to epigenetic inactivation and influence of hypoxia. The observations propose a possible overlap in the etiology of adrenocortical tumors and provide the rationale for further studies into the role of CACNA1H in the development of PPGL.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/ERC-23-0061.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
The study was financially supported by the Swedish Research Council, the Swedish Cancer Society, the Gustav V Jubilee Foundation, Stockholm county council (ALF), and Karolinska Institutet.
Ethics approval and consent to participate
All tissue samples were collected with informed consent, and the study of the tissue material was approved by the Swedish Ethical Review Authority.
Data availability
The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository (Perez-Riverol et al. 2022) with the dataset identifier PXD040446.
References
Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P, Kondrashov AS & & Sunyaev SR 2010 A method and server for predicting damaging missense mutations. Nature Methods 7 248–249. (https://doi.org/10.1038/nmeth0410-248)
Buffet A, Burnichon N, Favier J & & Gimenez-Roqueplo AP 2020 An overview of 20 years of genetic studies in pheochromocytoma and paraganglioma. Best Practice and Research 34 101416. (https://doi.org/10.1016/j.beem.2020.101416)
Clay MR, Pinto EM, Cline C, Tran QT, Lin T, Dyer MA, Shi L, Wu H, Pounds SB, Zambetti GP, et al.2019 DNA methylation profiling reveals prognostically significant groups in pediatric adrenocortical tumors: a report from the international pediatric adrenocortical tumor registry. JCO Precision Oncology 3. (https://doi.org/10.1200/PO.19.00163)
Daniil G, Fernandes-Rosa FL, Chemin J, Blesneac I, Beltrand J, Polak M, Jeunemaitre X, Boulkroun S, Amar L, Strom TM, et al.2016 CACNA1H mutations are associated with different forms of primary aldosteronism. EBiomedicine 13 225–236. (https://doi.org/10.1016/j.ebiom.2016.10.002)
Eisenhofer G, Pacak K, Huynh TT, Qin N, Bratslavsky G, Linehan WM, Mannelli M, Friberg P, Grebe SK, Timmers HJ, et al.2011 Catecholamine metabolomic and secretory phenotypes in pheochromocytoma. Endocrine-Related Cancer 18 97–111. (https://doi.org/10.1677/ERC-10-0211)
Fishbein L, Leshchiner I, Walter V, Danilova L, Robertson AG, Johnson AR, Lichtenberg TM, Murray BA, Ghayee HK, Else T, et al.2017 Comprehensive molecular characterization of pheochromocytoma and paraganglioma. Cancer Cell 31 181–193. (https://doi.org/10.1016/j.ccell.2017.01.001)
Flynn A, Benn D, Clifton-Bligh R, Robinson B, Trainer AH, James P, Hogg A, Waldeck K, George J, Li J, et al.2015 The genomic landscape of phaeochromocytoma. Journal of Pathology 236 78–89. (https://doi.org/10.1002/path.4503)
García-Palomero E, Cuchillo-Ibáñez I, García AG, Renart J, Albillos A & & Montiel C 2000 Greater diversity than previously thought of chromaffin cell Ca2+ channels, derived from mRNA identification studies. FEBS Letters 481 235–239. (https://doi.org/10.1016/s0014-5793(0001984-0)
Gürtler F, Jordan K, Tegtmeier I, Herold J, Stindl J, Warth R & & Bandulik S 2020 Cellular pathophysiology of mutant voltage-dependent Ca2+ channel CACNA1H in primary aldosteronism. Endocrinology 161 bqaa135. (https://doi.org/10.1210/endocr/bqaa135)
Hadrava Vanova K, Pang Y, Krobova L, Kraus M, Nahacka Z, Boukalova S, Pack SD, Zobalova R, Zhu J, Huynh TT, et al.2022 Germline SUCLG2 variants in patients with pheochromocytoma and paraganglioma. Journal of the National Cancer Institute 114 130–138. (https://doi.org/10.1093/jnci/djab158)
Hamill OP, Marty A, Neher E, Sakmann B & & Sigworth FJ 1981 Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflugers Archiv 391 85–100. (https://doi.org/10.1007/BF00656997)
Jochmanova I & & Pacak K 2018 Genomic landscape of pheochromocytoma and paraganglioma. Trends in Cancer 4 6–9. (https://doi.org/10.1016/j.trecan.2017.11.001)
Juhlin CC 2021 Challenges in paragangliomas and pheochromocytomas: from histology to molecular immunohistochemistry. Endocrine Pathology 32 228–244. (https://doi.org/10.1007/s12022-021-09675-0)
Juhlin CC, Stenman A, Haglund F, Clark VE, Brown TC, Baranoski J, Bilguvar K, Goh G, Welander J, Svahn F, et al.2015 Whole-exome sequencing defines the mutational landscape of pheochromocytoma and identifies KMT2D as a recurrently mutated gene. Genes, Chromosomes and Cancer 54 542–554. (https://doi.org/10.1002/gcc.22267)
Letouzé E, Martinelli C, Loriot C, Burnichon N, Abermil N, Ottolenghi C, Janin M, Menara M, Nguyen AT, Benit P, et al.2013 SDH mutations establish a hypermethylator phenotype in paraganglioma. Cancer Cell 23 739–752. (https://doi.org/10.1016/j.ccr.2013.04.018)
Levitsky KL & & López-Barneo J 2009 Developmental change of T-type Ca2+ channel expression and its role in rat chromaffin cell responsiveness to acute hypoxia. Journal of Physiology 587 1917–1929. (https://doi.org/10.1113/jphysiol.2009.168989)
Lindstrand A, Eisfeldt J, Pettersson M, Carvalho CMB, Kvarnung M, Grigelioniene G, Anderlid BM, Bjerin O, Gustavsson P, Hammarsjö A, et al.2019 From cytogenetics to cytogenomics: whole-genome sequencing as a first-line test comprehensively captures the diverse spectrum of disease-causing genetic variation underlying intellectual disability. Genome Medicine 11 68. (https://doi.org/10.1186/s13073-019-0675-1)
Liu T, Brown TC, Juhlin CC, Andreasson A, Wang N, Bäckdahl M, Healy JM, Prasad ML, Korah R, Carling T, et al.2014 The activating TERT promoter mutation C228T is recurrent in subsets of adrenal tumors. Endocrine-Related Cancer 21 427–434. (https://doi.org/10.1530/ERC-14-0016)
Liu X, Jin L, Zhang H, Ma W, Song L, Zhou X & & Cai J 2021 A Chinese pedigree with glucocorticoid remediable aldosteronism. Hypertension Research 44 1428–1433. (https://doi.org/10.1038/s41440-021-00685-3)
Lu M, Bränström R, Berglund E, Höög A, Björklund P, Westin G, Larsson C, Farnebo LO & & Forsberg L 2010 Expression and association of TRPC subtypes with Orai1 and STIM1 in human parathyroid. Journal of Molecular Endocrinology 44 285–294. (https://doi.org/10.1677/JME-09-0138)
Mahapatra S, Calorio C, Vandael DH, Marcantoni A, Carabelli V & & Carbone E 2012 Calcium channel types contributing to chromaffin cell excitability, exocytosis and endocytosis. Cell Calcium 51 321–330. (https://doi.org/10.1016/j.ceca.2012.01.005)
Marcantoni A, Carabelli V, Comunanza V, Hoddah H & & Carbone E 2008 Calcium channels in chromaffin cells: focus on L and T types. Acta Physiologica (Oxford, England) 192 233–246. (https://doi.org/10.1111/j.1748-1716.2007.01815.x)
Mete O, Asa SL, Gill AJ, Kimura N, de Krijger RR & & Tischler A 2022 Overview of the 2022 WHO classification of paragangliomas and pheochromocytomas. Endocrine Pathology 33 90–114. (https://doi.org/10.1007/s12022-022-09704-6)
Muth A, Crona J, Gimm O, Elmgren A, Filipsson K, Stenmark Askmalm M, Sandstedt J, Tengvar M & & Tham E 2019 Genetic testing and surveillance guidelines in hereditary pheochromocytoma and paraganglioma. Journal of Internal Medicine 285 187–204. (https://doi.org/10.1111/joim.12869)
Nanba K, Blinder AR, Rege J, Hattangady NG, Else T, Liu CJ, Tomlins SA, Vats P, Kumar-Sinha C, Giordano TJ, et al.2020 Somatic CACNA1H mutation as a cause of aldosterone-producing adenoma. Hypertension 75 645–649. (https://doi.org/10.1161/HYPERTENSIONAHA.119.14349)
Perez-Riverol Y, Bai J, Bandla C, García-Seisdedos D, Hewapathirana S, Kamatchinathan S, Kundu DJ, Prakash A, Frericks-Zipper A, Eisenacher M, et al.2022 The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences. Nucleic Acids Research 50 D543–D552. (https://doi.org/10.1093/nar/gkab1038)
Roomp K, Grzyb K, Wolf C, Rhayem Y, Oliver N, Wardas B, Beck A, Pico A, Beuschlein F, Skupin A, et al.2016 New germline mutation in CACNA1H calcium channel causing primary aldosteronism. Endocrine Abstracts 41 GP7. (https://doi.org/10.1530/endoabs.41.GP7)
Scholl UI, Stölting G, Nelson-Williams C, Vichot AA, Choi M, Loring E, Prasad ML, Goh G, Carling T, Juhlin CC, et al.2015 Recurrent gain of function mutation in calcium channel CACNA1H causes early-onset hypertension with primary aldosteronism. eLife 4 e06315. (https://doi.org/10.7554/eLife.06315)
Seidel E, Schewe J, Zhang J, Dinh HA, Forslund SK, Markó L, Hellmig N, Peters J, Muller DN, Lifton RP, et al.2021 Enhanced Ca(2+) signaling, mild primary aldosteronism, and hypertension in a familial hyperaldosteronism mouse model (Cacna1h(M1560V/+). PNAS 118 e2014876118. (https://doi.org/10.1073/pnas.2014876118)
Sellak H, Zhou C, Liu B, Chen H, Lincoln TM & & Wu S 2014 Transcriptional regulation of alpha1H T-type calcium channel under hypoxia. American Journal of Physiology. Cell Physiology 307 C648–C656. (https://doi.org/10.1152/ajpcell.00210.2014)
Shafi AM, Végvári Á, Zubarev RA & & Penha-Gonçalves C 2023 Brain endothelial cells exposure to malaria parasites links type I interferon signalling to antigen presentation, immunoproteasome activation, endothelium disruption and cellular metabolism. Frontiers in Immunology 14 1149107. (https://doi.org/10.3389/fimmu.2023.1149107)
Starikova AM, Pogorelaya NC & & Kostyuk PG 2000 Long-term depolarization changes morphological parameters of PC12 cells. Neuroscience 95 923–926. (https://doi.org/10.1016/s0306-4522(9900485-6)
Stenman A, Juhlin CC, Haglund F, Brown TC, Clark VE, Svahn F, Bilguvar K, Goh G, Korah R, Lifton RP, et al.2016a Absence of KMT2D/MLL2 mutations in abdominal paraganglioma. Clinical Endocrinology 84 632–634. (https://doi.org/10.1111/cen.12884)
Stenman A, Welander J, Gustavsson I, Brunaud L, Bäckdahl M, Söderkvist P, Gimm O, Juhlin CC & & Larsson C 2016b HRAS mutation prevalence and associated expression patterns in pheochromocytoma. Genes, Chromosomes and Cancer 55 452–459. (https://doi.org/10.1002/gcc.22347)
Stenman A, Svahn F, Hojjat-Farsangi M, Zedenius J, Söderkvist P, Gimm O, Larsson C & & Juhlin CC 2019 Molecular profiling of pheochromocytoma and abdominal paraganglioma stratified by the PASS algorithm reveals chromogranin B as associated with histologic prediction of malignant behavior. American Journal of Surgical Pathology 43 409–421. (https://doi.org/10.1097/PAS.0000000000001190)
Tseng CS, Peng KY, Wang SM, Tsai YC, Huang KH, Lin WC, Hu YH, Wu VC & & Chueh JS 2022 A novel somatic mutation of CACNA1H p.V1937M in unilateral primary hyperaldosteronism. Frontiers in Endocrinology 13 816476. (https://doi.org/10.3389/fendo.2022.816476)
Weiss N & & Zamponi GW 2020 Genetic T-type calcium channelopathies. Journal of Medical Genetics 57 1–10. (https://doi.org/10.1136/jmedgenet-2019-106163)
Welander J, Andreasson A, Juhlin CC, Wiseman RW, Bäckdahl M, Höög A, Larsson C, Gimm O & & Söderkvist P 2014 Rare germline mutations identified by targeted next-generation sequencing of susceptibility genes in pheochromocytoma and paraganglioma. Journal of Clinical Endocrinology and Metabolism 99 E1352–E1360. (https://doi.org/10.1210/jc.2013-4375)
Welander J, Łysiak M, Brauckhoff M, Brunaud L, Söderkvist P & & Gimm O 2018 Activating FGFR1 mutations in sporadic pheochromocytomas. World Journal of Surgery 42 482–489. (https://doi.org/10.1007/s00268-017-4320-0)
Wu CH, Peng KY, Hwang DY, Lin YH, Wu VC & & Chueh JS 2021 Novel mutations detection with next-generation sequencing and its association with clinical outcome in unilateral Primary aldosteronism. Biomedicines 9 1167. (https://doi.org/10.3390/biomedicines9091167)