ELSEVIER

Contents lists available at ScienceDirect

Journal of Steroid Biochemistry and Molecular Biology

journal homepage: www.elsevier.com/locate/jsbmb

-

The Journal of Steroid Biochemistry & Molecular Biology

Viapy

Visualization of calcium channel blockers in human adrenal tissues and their possible effects on steroidogenesis in the patients with primary aldosteronism (PA)

Check for updates

Naoki Motomura ª, Yuto Yamazaki ª,*, Xin Gao ª, Yuta Tezuka b,c, Kei Omata b,c, Yoshikiyo Ono b,c, Ryo Morimoto ”, Fumitoshi Satoh b,c, Yasuhiro Nakamura ”, Jaeyoon Shim , Man Ho Choie, Akihiro Ito, Hironobu Sasanoª

a Department of Pathology, Tohoku University Graduate School of Medicine, Sendai, Japan

b Division of Clinical Hypertension, Endocrinology and Metabolism, Tohoku University Graduate School of Medicine, Sendai, Japan

· Division of Nephrology, Endocrinology, and Vascular Medicine, Tohoku University Hospital, Sendai, Japan

d Division of Pathology, Faculty of Medicine, Tohoku Medical and Pharmaceutical University, Sendai, Japan

e Molecular Recognition Research Center, Korea Institute of Science and Technology, Seoul, Republic of Korea

‘ Department of Urology, Tohoku University School of Medicine, Sendai, Japan

ARTICLE INFO

Keywords:

Primary aldosteronism MALDI imaging Liquid chromatography-mass spectrometry Blood pressure Calcium channel blocker HSD3B

ABSTRACT

Voltage-gated L-type calcium channel (CaV) isoforms are well known to play pivotal tissue-specific roles not only in vasoconstriction but also in adrenocortical steroidogenesis including aldosterone biosynthesis. Alpha-1C subunit calcium channel (CC) (CaV1.2) is the specific target of anti-hypertensive CC blockers (CCBs) and its Alpha-1D subunit (CaV1.3) regulates depolarization of cell membrane in aldosterone-producing cells. Direct effects of CCBs on aldosterone biosynthesis were previously postulated but their intra-adrenal distribution and effects on steroid production in primary aldosteronism (PA) patients have remained virtually unknown. In this study, frozen tissue specimens constituting tumor, adjacent adrenal gland and peri-adrenal adipose tissues of nine aldosterone-producing adenoma (APA) cases were examined for visualization of amlodipine and aldoste- rone themselves using matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI). Liquid chromatography-mass spectrometry (LC-MS) analysis was also performed to quantify amlodipine and 17 adrenal steroids in those cases above and compared the findings with immunohistochemical analysis of ste- roidogenic enzymes and calcium channels (CaV1.2 and CaV1.3). Effects of amlodipine on mRNA level of aldo- sterone biosynthetic enzymes were also explored using human adrenocortical carcinoma cell line (H295R). Amlodipine-specific peak (m/z 407.1 > 318.1) was detected only in amlodipine treated cases. Accumulation of amlodipine was marked in adrenal cortex compared to peri-adrenal adipose tissues but not significantly different between APA tumors and adjacent adrenal glands, which was subsequently confirmed by LC-MS quantification. Intra-adrenal distribution of amlodipine was generally consistent with that of CCs. In addition, quantitative steroid profiles using LC-MS and in vitro study demonstrated the lower HSD3B activities in amlo- dipine treated cases. Immunoreactivity of CaV1.2 and HSD3B2 were also correlated. We report the first demonstration of specific visualization of amlodipine in human adrenal tissues by MALDI-MSI. Marked amlo- dipine accumulation in the adrenal glands suggested its direct effects on steroidogenesis in PA patients, possibly targeting on CaV1.2 and suppressing HSD3B activity.

1. Introduction

Primary aldosteronism (PA) is one of the major causes of secondary hypertension, accounting for approximately 5-10% of all hypertensive

patients [1-3]. PA is clinically classified into aldosterone-producing adenoma (APA) and idiopathic hyperaldosteronism (IHA) [1]. In addi- tion, PA was also reported to frequently develop medical treatment-resistant hypertension and a higher risk of cardiovascular

* Corresponding author at: Department of Pathology, Tohoku University School of Medicine, 2-1 Seiryo-machi, Aoba-ku, Sendai, 980- 8575, Japan.

E-mail address: y.yamazaki@patholo2.med.tohoku.ac.jp (Y. Yamazaki).

Table 1 Clinicopathological characteristics of APA cases.
No.AgeSexFinal Histological Diagnosis (After CYP11B2 IHC)PAC (ng/ dL)PRA (ng/ mL/h)ARRMacroscopic (or Histological) Diameter [mm]Availability of adipose tissueAmlodipine administrationDrug administration other than amlodipine
172MDouble APAs236.50.21183Noamlodipine OD 2.5 mg x1eplerenone 125 mg x2 nifedipine OD 40 mg 2 tablets x2 bunazosin 3 mg
270FAPA92.5192.524Yesamlodipine 2.5 mg x2spironolactone 25 mg x4
351MAPA33.80.216913Yesamlodipine 5 mg x2spironolactone 25 mg x4
472FAPA580.229013Noamlodipine OD 2.5 mg 2 tablets x2spironolactone 25 mg 4 tablets x2
5 625 51M FAPA APA61.6 15.20.2 0.2308 7617 11No Yes0 mg 0 mgCilnidipine10 mg 2 tablets x2 nifedipine CR 40 mg 2 tablets x2 doxazosin 1 mg 2 tablets x2 spironolactone 200 mg x2 nifedipine CR 20 mg 2 tablets x2
757MAPA32.40.216210Yes0 mgeplerenone 25 mg 3 tablets x2 nifedipine OD 40 mg eplerenone 25 mg 4 tablets x2
864MAPA41.90.2209.510Yes0 mgeplerenone 25 mg 6 tablets nifedipine CR 20 mg 4 tablets doxazosin 2 mg
943FAPA55.10.2275.513No0 mg *eplerenone 150 mg x2

Abbreviation; M (Male), F (Female), PAC (Plasma aldosterone concentration), PRA (Plasma renin activity), ARR (Aldosterone renin ratio), OD (Oral Disintegration), N/ A: Not available. Case 9 had been treated with amlodipine but was suspended in several months before the adrenalectomy.

disease compared to essential hypertensive patients even at the same severity of hypertension, especially in younger patients [4,5]. Among medical therapeutic agents of PA patients, calcium channel blockers (CCBs) have been widely used to decrease blood pressure by inhibiting the intracellular calcium influx and subsequently relaxing the tone of vascular smooth muscle. Amlodipine is a third generation dihydropyr- idine CCB, which specifically targets on voltage-gated L-type calcium channel subunit alpha 1C (CaV1.2) present in the vascular smooth muscle and has become one of the most widely used and long-acting CCBs [6-8]. Amlodipine is also characterized by its unique pharmaco- kinetics, i.e., high accumulation in the cell membrane, 10,000 times higher than adjacent aqueous environment, which accounts for its relatively longer half-life time (30-50 h) compared to other CCBs [9-11].

Calcium is also established as a key second messenger of the adre- nocortical steroid hormone biosynthesis including aldosterone [12-14]. In aldosterone biosynthesis, voltage-dependent L-type calcium channel subunit alpha 1D (CaV1.3), which is highly homogeneous to CaV1.2, has been also well-known to play pivotal roles in regulating the intracellular calcium concentration in aldosterone-producing cells [15]. In addition, various aldosterone driver-gene somatic mutations in KCNJ5, CAC- NA1D, ATP1A1 and ATP2B3 genes have been reported in both APAs and IHAs [16-19]. These genetic alterations were also reported to contribute to aldosterone overproduction via increasing intracellular calcium con- centration mainly through the calcium channels above. Several previ- ously reported in vitro studies demonstrated that CCBs directly inhibited the aldosterone production and decreased the expression levels of ste- roidogenic enzymes [15,20,21]. In addition, previous in vivo study also reported the relatively lower plasma aldosterone concentration levels in PA patients treated with CCBs [22]. These findings above all indicated that CCBs could directly influence the aldosterone production in the adrenal glands. However, the possible intra-adrenal distribution of CCBs and which steroidogenic pathways to be influenced by CCBs in PA ad- renals have remained unknown.

Therefore, in this study, we attempted to visualize amlodipine and aldosterone in PA adrenals by matrix-assisted laser desorption/ioniza- tion mass spectrometry imaging (MALDI-MSI) analysis. In addition, tissue concentrations of adrenal steroids and amlodipine in the same adrenal glands studied by MALDI-MSI were quantitatively and

comprehensively analyzed using liquid chromatography-mass spec- trometry (LC-MS). Furthermore, in order to clarify the specific direct amlodipine effects on steroidogenic enzymes expression levels, mRNA levels were examined in human adrenocortical carcinoma cell line (H295R) with amlodipine treatment in different doses. Steroidogenic enzyme immunoreactivity was also compared with that of CaV1.2 and CaV1.3, respectively.

2. Materials and methods

2.1. Human tissues

9 cases of APAs (Case 1-9) and 5 cases (Case 2, Case 3, Case 6, Case 7 and Case 8) of corresponding peri-adrenal adipose tissues, and non- pathological liver tissue operated at Tohoku University Hospital, Sen- dai, Japan were examined in this study. All APA cases were clinically diagnosed as PA according to the Japanese Endocrine Society guidelines [23]. The liver tissues were obtained from the autopsy cases as a nega- tive control of aldosterone biosynthesis. We simultaneously prepared optimal cutting temperature (OCT) compound embedded tissues and 10 % formalin-fixed paraffin-embedded (FFPE) tissues on site. Both speci- mens contained the whole tumor area harboring their greatest dimen- sion in order to reflect the intra-tumoral heterogeneity as much as possible. The clinical features of APA cases, including the history of amlodipine administration and other hypertensive agents in individual patients were summarized in Table 1. Four cases (Case 1-4) were treated with amlodipine before adrenalectomy and 5 cases (Case 5-9) not. All medications were interrupted on the day of adrenalectomy. However, in Case 9, amlodipine administration was suspended for several months before adrenalectomy and was tentatively regarded as a non-treated case, based on the amlodipine’s half-life time [9].

Serial tissue sections were prepared from the OCT-embedded frozen tissues of APA and peri-adrenal adipose and liver for the analysis of MALDI-MSI and hematoxylin and eosin (H&E) staining. Serial tissue sections of APA and peri-adrenal adipose tissues were also prepared for immunohistochemistry (IHC) and LC-MS analysis. Another set of serial tissue sections were prepared from FFPE tissues of APAs used for H&E staining and IHC of steroidogenic enzymes and calcium channels.

This research protocol was approved by Tohoku University School of

Medicine IRB (2019-1-467).

2.2. H&E staining and IHC of frozen and FFPE sections

In frozen tissue sections, H&E staining and IHC were performed in serial tissue sections of liver, APA and peri-adrenal adipose tissues. They were sectioned at 5 um thickness using cryotome (thermo fisher scien- tific, Waltham, USA) and fixed in ethanol (Fujifilm Wako Pure Chem- icals Co., Ltd., Osaka, Japan) and diethyl ether (Fujifilm Wako Pure Chemicals Co., Ltd.) for H&E staining, and in 10 % neutrally-buffered formalin (Fujifilm Wako Pure Chemicals Co., Ltd.) for IHC, respec- tively. IHC protocols of CYP11B2 [24], CaV1.2 and CaV1.3 in frozen tissue sections were summarized in Supplemental Table 1-A. CYP11B2 IHC was performed in APA and peri-adrenal adipose tissues, and CaV1.2 and CaV1.3 IHC were performed only in APA tissues. In addition, CYP11B2 positive areas were illustrated in a more detailed fashion in Supplemental Fig. 1.

In FFPE tissue sections, H&E staining, and IHC of steroidogenic en- zymes (HSD3B1 and HSD3B2) and calcium channels (CaV1.2 and CaV1.3) were performed in APA specimens including the tumors, adja- cent adrenal glands and peri-adrenal adipose tissues. Sections were sliced at 3 um thickness using microtome (Yamato Kohki Industrial co., ltd., Saitama, Japan). IHC protocols used for FFPE sections were sum- marized in Supplemental Table 1-B.

2.3. Evaluation of immunoreactivity

All FFPE stained tissue sections were digitally scanned by Image Scope AT2 (Leica, Wetzler, Germany), and evaluated using HALO TM CytoNuclear ver.1.5 (Indica Labs, Corrales, NM) program as previously reported [25,26]. Whole target regions of each type of tissues including tumors, adjacent adrenal glands and peri-adrenal adipose tissues were annotated for the analysis. Cytoplasmic immunoreactivity of CaV1.2, CaV1.3, HSD3B1 and HSD3B2 were quantitatively analyzed and compared the findings among the target regions above. Positive cells were tentatively classified into +1 (Weak), +2 (Moderate) and +3 (Strong) according to their immunointensity. H-Score was subsequently calculated according to the following formula: _ (The number of indi- vidual immunopositive cells X Immunointensity score +1, +2, +3) / Number of total cells) X100 [27,28]. Immunolocalization of individual cortical layers; ZG (zona glomerulosa), ZF (zona fasciculata), ZR (zona reticularis) in adjacent adrenal cortex were also analyzed.

Frozen tissue sections were also made for the analysis of target markers’ distribution and compared the findings with MALDI-MSI images.

2.4. MALDI-MSI analysis of amlodipine

OCT-embedded frozen tissues were sliced at 8 um thickness and mounted on Indium-Tin Oxide (ITO) coated glass slide (Sigma-Aldrich, St. Luis, USA). Matrix application using iMLayer (Shimadzu, Kyoto, Japan) was performed and a-Cyano-4-hydroxycinnamic acid («-CHCA) (Sigma-Aldrich) was evenly deposited at the thickness of 0.7 um. MALDI-MSI analysis using iMScope (Shimadzu) was performed after the matrix application. Amlodipine specific peak was obtained by MS/MS analysis in the positive ion mode. The transition for MS/MS analysis was m/z 407.1 > 318.1 [29]. The analytical protocol was summarized in Supplemental Fig. 2. Parameters of MALDI-MSI were optimized in each frozen tissue section to maximize the signal to noise ratio. The peak intensity distribution was merged with microscopic image, and subse- quently converted into the integrated ones by Imaging MS Solution (Shimadzu).

In addition, distribution of peak intensity was statistically analyzed using Imaging MS Solution in a quantitative manner. Both adrenal tis- sues (tumors and adjacent adrenal gland) and peri-adrenal adipose tis- sues were mounted on the same slides in order to standardize the

conditions and parameters.

2.5. MALDI-MSI analysis of aldosterone

Frozen tissue specimens of APAs were prepared in the same way as amlodipine analysis mentioned above. In addition, in aldosterone visualization, liver tissue sections were mounted on the same ITO coated glass slide, and aldosterone (Sigma-Aldrich, molecular weight: 360.4) standard solution were administered and analyzed in order to further confirm the specificity and sensitivity of each experiment. The deriva- tization by Girard’s reagent T (Gir-T) (Sigma-Aldrich) and two step matrix application of a-CHCA were performed to improve the ionization efficiency as previously reported [30-33]. The frozen tissue sections prepared were sprayed with 10 mg/mL Gir-T of 20 % acetic acid (Fujifilm Wako Pure Chemicals Co., Ltd.), and the derivatizing reaction was performed at room temperature for 90 min. After the derivatization, two-step matrix application was performed. «-CHCA was evenly deposited at the thickness of 0.7 um using iMLayer. «-CHCA reagent at the concentration of 10 mg/mL in 30 % acetonitrile (Fujifilm Wako Pure Chemicals Co., Ltd.), 10 % isopropanol (Fujifilm Wako Pure Chemicals Co., Ltd.) and 0.1 % trifluoroacetic acid (Fujifilm Wako Pure Chemicals Co., Ltd.) were subsequently sprayed. Specific MALDI-MSI analysis using iMScope was performed to detect aldosterone because Gir-T de- rivatives of aldosterone and cortisone had the same ion transition for MS/MS analysis. Aldosterone specific peak with m/z 474.3 > 415.2 > 397.2 was obtained by one additional tandem mass spectrometry (MS/MS/MS) analysis in the positive ion mode [32,33]. MALDI-MSI parameters were optimized for each sample to maximize the signal to noise ratio and the resolution. The analytical protocol was summarized in Supplemental Fig. 3. The merged images were obtained in the same way as mentioned above.

2.6. Quantitative LC-MS analysis of amlodipine, adrenal steroids and their metabolic ratios

Individual target tissue areas of the tumors, adjacent adrenal glands, and peri-adrenal adipose tissues were separately isolated by micro- dissection under the stereomicroscope with care. Weight of each sam- ple was measured in advance for the normalization. Cortical steroids and amlodipine quantification using LC-MS/MS were subsequently per- formed. In steroids quantification, 17 steroid hormones (progesterone, 11-deoxycorticosterone, 17a-hydroxyprogesterone, 11ß-hydrox- yprogesterone, 18-hydroxycorticosterone, aldosterone, corticosterone, 11-deoxycortisol, 20x-dihydrocortisol, 18-hydroxycortisol, 21-deoxy- cortisol, cortisol, cortisone, 6ß-hydroxycortisol, pregnenolone, 17- hydroxypregnenolone, dehydroepiandrosterone) were quantitatively analyzed using LC-MS as previously reported [34]. Amlodipine was simultaneously quantified by LC-MS/MS. d4-Amlodipine (Fujifilm Wako Pure Chemicals Co., Ltd.) was used as the internal control. The MS/MS transitions were described as follows: d4-amlodipine; m/z 412.9 (precursor ion) > 238.1, 298.2, 206.1, and m/z 238.1 was selected for the quantification. Amlodipine; m/z 409.1 (precursor ion) > 238.1, 294.2, 206.1, and m/z 238.1 was selected for the quantification. The retention time of d4-amlodipine and amlodipine was 8.87 and 8.88 min, respectively.

The enzymic activities of HSD3B, CYP21A2, CYP11B1, CYP11B2, CYP17A1, HSD11B1, HSD11B2, 66-hydroxylase and 20x-reductase in tumor areas were subsequently analyzed by the corresponding meta- bolic ratios [35]. The metabolic ratios were calculated by dividing the absolute amounts of steroid hormones by the corresponding precursors as previously reported [35], and were compared between amlodipine treated and non-treated cases.

A

Fig. 1. Amlodipine visualization in both treated and non-treated tissues. Regions annotated by red circles corresponded to the tumor areas defined by CYP11B2 IHC. Imaging results of amlodipine in each tissue and preoperative amlodipine dosage were summa- rized. A: Amlodipine was specifically visualized in treated cases but not necessarily consistent with CYP11B2 distribution. B: Amlodipine signal was not detected in non-treated cases.

Treated cases

Case 1 Dose 2.5 mg × 2 /day

Case 3 Dose 5 mg × 2 /day

H&E staining

CYP11B2 IHC

Microscopic image

Amlodipine distribution

Superimposed image

318.05

318.05

1000 g.m

100

318.05

318.05

1000 1 m

100

LO

50

50

0

0

Case 2 Dose 2.5 mg ×2 /day

Case 4 Dose 5.0mg × 2/day

318.05

318.05

318.05

318.05

100

5

3

S

50

50

0

0

B

Non-treated cases

Case 5 Dose 0 mg/day

Case 8 Dose 0 mg/day

H&E staining

CYP11B2 IHC

Microscopic image

Amlodipine distribution

Superimposed image

100

318.05

318.05

100

50

50

0

Case 6 Dose 0 mg/day

318.05

100

318.05

Case 9 Dose 0 mg/day

50

0

318.05

318.05

100

Case 7 Dose 0 mg/day

50

318.05

318.06

100

O

50

0

0

2.7. Cell culture and steroidogenic enzyme analysis in H295R following amlodipine treatment by qRT-PCR

H295R was commercially obtained from American Type Culture Collection (ATCC, Manassas, USA) and cultured in growth medium, Dulbecco’s modified Eagle’s medium (DMEM)/Ham’s nutrient mixture F-12, with 10 % fetal bovine Serum (FBS) +1% ITS-Premix (Corning, New York, USA) [36]. These cells were incubated at 37 ℃ with a hu- midified 5% CO2 atmosphere. After the cells reached their confluence in the flask, they were seeded into 12-well plates and incubated for 24 h in the growth medium. The growth medium was then replaced to that with or without 0.1 umol/L Angiotensin II (ANG II), and the dose series of 0 pmol/L, 0.001 µmol/L, 0.1 µmol/L, 10 umol/L of amlodipine dissolved in dimethyl sulfoxide (Fujifilm Wako Pure Chemicals Co., Ltd.) was added and incubated for 24 h [17]. Following the treatment above, total RNA was extracted from these cells using TRI reagent (Cosmo Bio Co., Ltd, Tokyo, Japan). cDNA was subsequently synthesized using Quanti- Tect Reverse Transcription Kit (QIAGEN). RT-PCR was performed using LightCycler FastStart DNA Master SYBR Green I kit (Roche, Basel, Switzerland) for CYP11B2 and RPL13A analysis [26,37-42]. HSD3B1 and HSD3B2 qPCR were also performed using TaqMan probes and primers as previously reported [38]. The primer sequences and TaqMan probes used in our present study were summarized in Supplemental Table 2-A, B. RPL13A was used as an endogenous control gene, and the

relative gene expression was calculated by each standard curve as pre- viously reported [40-43].

2.8. Statistical analysis

Results of IHC and LC-MS studies were analyzed by the software “JMP Pro 16.0” and results of MALDI-MSI were analyzed by Imaging MS Solution. The activities of steroidogenic enzymes between amlodipine treated and non-treated tumors, and the immunoreactivity of calcium channels among tumors, adjacent adrenal glands, and peri-adrenal ad- ipose tissues, and transitions of H295R cells’ mRNA level were analyzed using Mann-Whitney test. In addition, the correlation of immunoreac- tivity between steroidogenic enzymes and calcium channels were also analyzed by Spearman’s rank correlation coefficient test. Mann-Whit- ney’s U test was applied to examine the difference of amlodipine accu- mulation obtained by studying the distribution of amlodipine specific peak intensity between tumor and adjacent adrenal glands, and between adrenal gland and adipose tissue. In all statistical analyses, the signifi- cance was regarded as P-value < 0.05.

Fig. 2. Aldosterone visualization, and concordance of its distribution with the amlodipine localization and CYP11B2 immunolocalization. Regions annotated by red circles corresponded to the tumor areas defined by CYP11B2 IHC. A: Imaging results of aldosterone in each tissue and plasma aldosterone concentration (PAC) were summarized. Aldosterone was detected only in tumors. B: Amlodipine distribution was compared with aldosterone distribution and CYP11B2 immunolocalization. They were partly overlapped but amlodipine was much more widely detected throughout the adrenal glands, and CYP11B2 immunoreactivity and aldosterone signal were detected in tumor areas.

A

Case 1

PAC 236.5 ng/dL

Case 5

PAC 61.6 ng/dL

397.21

397.21

397.21

397.21

1000 gr m

100

100

50

0

50

Case 6

PAC 15.2 ng/dL

397.21

397.21

0

100

2000 gr m

PAC 92.5 ng/dL

50

Case 2

0

397.22

397.22

100

Case 7

PAC 41.9 ng/dL

50

397.21

397.21

700 gr m

100

0

Case 3

PAC 33.8 ng/dL

50

397.22

397.22

100

0

50

Case 8

PAC 32.4 ng/dL

0

397.21

397.21

100

2000 g/ m

Case 4

PAC 58.0 ng/dL

50

397.21

397.21

1000 um

100

0

Case 9

PAC 55.1 ng/dL

50

397.21

397.21

1000 p

100

1000 Ar

50

0

0

B

H&E stainig

CYP11B2 IHC

Aldosterone Distribution Amlodipine Distribution

397.21

318.05

100

100

Case 1 Amlodipine

2.5 mg × 2 /day

50

50

0

0

397.22

318.05

Case 2 Amlodipine

100

100

2.5 mg ×2/day

50

50

0

0

318.05

100

100

Case 3 Amlodipine

5 mg × 2 /day

50

50

0

0

397.21

318.05

100

100

Case 4 Amlodipine

2.5 mg 2tablets × 2 /day

50

50

0

0

P

3. Results

3.1. Amlodipine visualization in both treated and non-treated tissues of the APA patients

Amlodipine visualization using MALDI-MSI along with the protocol mentioned above was newly developed in this study and performed in 9 cases of resected APA tissues. The detection specificity was confirmed by analyzing amlodipine (Fujifilm Wako Pure Chemicals Co., Ltd.) stan- dards on stainless steel plates (Shimadzu) (data not shown). Amlodipine

was detected and visualized only in its preoperatively treated 4 cases (Case 1-4, Fig. 1-A) but not in non-treated 5 cases, including Case No. 9 (Case 5-9, Fig. 1-B). Of particular interest, CYP11B2 immunolocalization (regions circled in red) and amlodipine localization were not necessarily concordant. There were no significant differences of amlodipine peak intensity distribution between tumor (CYP11B2 positive region) and adjacent adrenal gland (CYP11B2 negative regions) by examining the overview image captured (Fig. 1-A). However, CaV1.2 and CaV1.3 were also ubiquitous (Supplemental Fig. 4) and tended to be overlapped with the distribution of amlodipine.

m/z 318.1TumorAdj.
Average64.565.6
standard deviation144.0148.9
Central value0.00.0
P value8.380e-001

A

1500

1000

500

Case 1

0

318.05

318.05

Tumor

Adj.

1000 4 m

100

50

CYP11B2 IHC

0

4000

2000

0

Tumor

Adj.

Case 3

318.05

318.05

680

340

CYP11B2 IHC

0

B

Case 2

811

405

adrenal

CYP11B2 IHC

0

m/z 318.1

Adipose

Adrenal gland

318.05

318.05

Average

102.1

165.1

1000

600 x m

811

standard deviation

130.5

165.8

500

405

Central value

58.0

135.0

1.812e-035

0

adipose

CYP11B2 IHC

P value

0

Adipose

Adrenal gland

Case 3

650

318.05

318.05

m/z 318.1

Adipose

Adrenal gland

1000

340

Average

42.2

65.7

adrenal

CYP11B2 IHC

0

standard deviation

140.2

105.9

500

318.06

318.06

680

Central value

0.0

0.0

340

0

P value

8.805e-019

adipose

CYP11B2 IHC

0

Adipose

Adrenal gland

m/z 318.1TumorAdj.
Average251.2268.6
standard deviation412.1450.5
Central value0.00.0
P value0.4.981

Fig. 3. Semi-quantitative analyses of distribution of signal intensity.

MALDI-MSI analysis of amlodipine was performed in adrenal and peri-adrenal adipose tissues simultaneously and the distribution of signal intensity was subse- quently analyzed. When comparing amlodipine distribution between tumor and peri-adrenal adipose tissues, Case 2 and Case 3 were examined because of their availability of the peri-adrenal adipose tissue. Results of H&E staining, CYP11B2 IHC, microscopic image, distribution of peak intensity, superimposition of peak intensity distribution with microscope image, box plot, and the table of statistical figures were illustrated. A: Comparison between the tumor and the adjacent adrenal gland. There was no significant difference of amlodipine accumulation between them. B: Comparison between steroidogenic (tumor and adjacent adrenal gland) and non-steroidogenic (peri-adrenal adipose) tissues. Amlodipine was more abundantly detected in adrenal tissues than peri-adrenal adipose tissues in both cases (P < 0.05).

3.2. Aldosterone visualization and comparison with the amlodipine localization

Aldosterone visualization was performed in 9 cases. At first, the specificity of this method was further confirmed by analyzing dilution series of a standard solution of aldosterone (100 umol/L, 10 umol/L, 1 umol/L, 0.1 µmol/L, 0.01 µmol/L) on the liver tissue (Supplemental Fig. 5-A). Aldosterone was specifically and markedly detected in the areas where 10 umol/L and 100 umol/L of aldosterone standard solution

were administered, and not detected in aldosterone non-administered area. In addition, the sensitivity and specificity of each experiment were also confirmed by mounting the liver tissue on the same ITO coated glass slide as APA and subsequently analyzing the 100 umol/L of aldo- sterone standard solution administered on it (Supplemental Fig. 5-B). Aldosterone specific peak was clearly detected at least 100 umol/L of its concentration in each case. Localization of aldosterone was generally overlapped with CYP11B2 immunolocalization (regions circled in red) (Fig. 2-A). Aldosterone and amlodipine distribution were partly

Table 2 Amlodipine and 17 adrenal steroids quantification by LC-MS analysis.
sampleused amount (ng/g)
weightcasetissueProg11- deoxyB17a-OH- Prog11b- OH- Prog18-OHBAldosteroneB11- deoxyF20a- DHF18-OHF21- deoxyFFE6b- OHFPreg17-OH- pregDHEAAmlodipine
0.5 mg1tumor2719.72856.65904.36163.196687.692996.916491.871790.2274.342118.11196.627551.58748.8629.747115.85423.09103.151617.74
0.3 mg1adj2328.41573.157963.1157.392339.7727.047426.734303.12497.643323.882815.6648770.74654.86248.6211923.912737.13470.883210.11
1.7 mg2tumor3474.441086.46461.1356.28908.18196.642798.411759.73109.092678.55119.1513723.95749.8249.8714979.01773.32109.758490.82
1.9 mg2adipose7.53ND2.17ND22.2ND43.5719.537.4630.14ND213.9223.212.6333.84ND28.73545.76
0.2 mg3tumor6783.49878.662390.02374.732970.93906.395780.562214.63163.784948.16454.45211511225.0675.9252930.325616.75698.5427405.26
2.0 mg3adj847.41170.751641.7124.97236.0911.5738.472448.2125.061529.33778.998551.35795.0767.6311073.713368.3512035225.82
2.4 mg3adipose8.73ND5.13ND17.21ND16.2810.256.0832.57ND78.679ND40.48NDND640.07
2.1 mg4tumor687.5172.16374.4820.57393.68111.55960.33606.5848.982241.48190.79349.07850.2353.8525428.643384.91263.245550.76
1.0 mg5tumor7800.791244.665798.6213.161134.47601.313591.593885.6581.423609.68601.3816597.09556.1652.7114109.291503.37131.56NS
0.2 mg5adj1172.7540.94926.32394.03472.0212.992765.842232.93133.851569.721135.7823299.611164.3750.7417615.047450.313911.56NS
2.3 mg6tumor5557.69992.552781.93155.961658.82458.528495.71228.43140.791535.56143.2921943.761623.0772.826844.94602.138.74NS
1.1 mg6adj676.4171.831476.8736.84692.5915.023031.151332.66286.752337.4313.9222630.333061.13168.629782.971873.01446.78NS
2.4 mg6adipose19.4313.6728.812.2347.58ND138.6423.0219.8556.45ND983.7184.2219.3115.1284.6119.49NS
0.2 mg7tumor1968.22658.52101.84150.893227.021822.112065.6743.1639.41769.19221.915395.7311.520.935077.45352.7199.12NS
0.3 mg7adj4999.61582.038733.16217.53829.639.324396.346344.7358.394185.295577.3433009.83005.93143.540161.774170.461000.31NS
1.9 mg7adipose8.26ND1.21NDNDND9.925.18NDNDND28.45NDND36.25NDNDNS
1.1 mg8tumor13554.051455.635387.13961.925950.92146.3213377.451579.73109.963302.131066.9617332.151398.4452.1111641.991507.38274.6NS
2.8 mg8adj816.24143.751685.9815.95187.9410.04888.511526.34126.15669.57374.357833.68767.4560.256843.271758.05827.64NS
3.5 mg8adipose16.13ND17.35ND14.8ND27.0514.18ND10.61ND101.188.721.1151.9725.7548.2NS
1.8 mg9tumor6574.282371.42871.71257.374013.972148.844113.951051.823.762220.37120.433588.09210.6110.4814119.72598.560.68NS
1.0 mg9adj891.18208.491132.4429.47309.24123.47712.481212.6934.28317.91116.873161.09276.8810.457808.031541.48744.34NS

Abbreviation; Prog (progesterone), 11-deoxyB (11-deoxycorticosterone), 17a-OH-Prog (17a-hydroxyprogesterone), 11b-OH-Prog (11}-hydroxyprogesterone), 18-OHB (18-hydroxycorticosterone), Aldo (aldosterone), B (corticosterone), 11-deoxyF (11-deoxycortisol), 20a-DHF (20x-dihydrocortisol), 18-OHF (18-hydroxycortisol), 21-deoxyF (21-deoxycortisol), F (cortisol), E (cortisone), 6b-OHF (66-hydroxycortisol), Preg (pregnenolone), 17-OH-Preg (17-hydroxypregnenolone), DHEA (dehydroepiandrosterone), ND (not detected), NS (no signal corresponding amlodipine).

Among 9 cases of APA examined in this study, adjacent adrenal glands were available for examination in 7 cases and peri-adrenal adipose tissues in 5 cases. Amlodipine signals were detected in all treated cases of both adrenal glands and peri-adrenal adipose tissues but not detected (no signal corresponding amlodipine) in all the non-treated cases.

Fig. 4. Quantitative comparative LC-MS analysis of steroid hormones. Seventeen adrenal steroids were measured and statistically analyzed. A: Comparison of levels of adrenal steroids between adipose and adrenal (tumor and adjacent adrenal gland) tissues. B: Comparison of levels of adrenal steroids between amlodipine treated and non-treated tumors.

A

Pregnenolone

Progesterone

11-deoxyB

B

18-OHB

Aldosterone

00000

15000

15000

6000

50000

18-hydroxycorticosterone

5000

2000

40000

Pregnenolone

Progesterone

10000

corticosberone

10000

4000

1500

1000

3000

Aldosterone

10000

5000

500

5000

2000

1000

30000

500

·

-

0

-

0

0

-

D

-30000

·

-

choose

adicione

adrenal

Adipose

adrenal

adipos

adrenal

-1000

adrenal

P=0.0011

P=0.0011

P=0.0011

P=0.0011

P=0.0011

P=0.0010

17 a -OH-Prog

17 a -OH-Preg

11-deoxyF

F

18-OHF

E

1000

6000

000

35000

5000

3500

1000

.

4000

3000

17a-OH-Progesterone

6000

4000

5000

25000

2500

3000

4000

20000

3000

2000

2000

2000

Cortisol

15000

Cortisone

1500

:

2000

2000

2000

10000

5000

1000

500

·

-

0

-

0

-

0

-

0

0

-

adipose

adrenal

-3000

adipose

adipose

adrenal

-5000

adrenal

adipose

adrenal

500

P=0.0011

P=0.0011

P=0.0011

P=0.0011

P=0.0011

P=0.0011

6 B OHF

20 a -DHF

21-deoxyF

11 ß-OH-Prog

DHEA

400

6000

1400

150

5000

*

1200

Adipose : 5 cases

õ

65-011-cortisol

200-Dihydroxycortisol

250

4000

1000

50

2000

400

400

Adrenal (tumor + adj.) Tumor : 9 cases

9

200

200

α

®

4

-

®

-

4

adipose

adrenal

50

adpost

adrenal

1000

adipose

adrenal

adipose

adrenal

200

adipose

adrenal

Adj. : 7 cases

P=0.0044

P=0.0011

P=0.0010

P=0.0011

P=0.0044

B

Pregnenolone

Progesterone

11-deoxyB

B

18-OHB

Aldosterone

15000

7000

7000

50000

#5000

3000

52500

2500

12500

6000

40000

10000

5000

Pregnencione

13000

2000

corticosterone

10000

5000

-

7500

1500

7500

4000

4000

3000

.

5000

*

5000

2000

2000

30000

2500

=

500

2500

1000

1000

0-

administrated tumor

not: administrabed tumor

Nminstrated Turner

0

administrabed tumor

not administrabed tumor

0

not administrated turner

0

ted tumor

not: administrabed tumor

0

Administrated turner

not administrabe

P=0.1113

P= 0.2703

P= 0.7133

P=0.5403

P=0.5403

P=0.5403

17 a -OH-Prog

17 a -OH-Preg

11-deoxyF

F

18-OHF

E

6000

6000

4000

5000

5000

5000

2500

20000

1500

4000

170-OH-Progesterone

1000

17-OH-Pregnenolon

4000

3000

15000

Cortison

1000

3000

2500

3000

2000

2500

2000

30000

1500

2000

11000

1500

500

5000

1000

5000

1000

0

administrated turner

not adminstrated tumor

0

Administrated tumor

not. administrated tumor

500

not administrated turner

administrated tumor

not: administrated tumor

500

böminstrated tumor

not adminstrated tumor

·

not administrated turner

L

P=0.1113

P=0.3913

P=0.7133

P=1.0000

P=0.5403

P=0.7133

6 B OHF

20 a -DHF

21-deoxyF

11 ß -OH-Prog

DHEA

00

1000

800

150

20g-Dihydrowycortisol

21-Deowycortisol

Administrated tumor : 4 cases

000

1

800

6

900

600

600

500

400

V

8

et

50

300

30

200

200

100

Not-administrated

adminstrated tumgr

not administrated tumor

0

4

adminstrated tumor

not administrated tumor

administrabed tumor

not administrabedi tumor

a

administrated tumor

not administrated tumor

4

administrated turner

not administrated tumor

tumor : 5 cases

P=0.5403

P=0.7133

P=0.5403

P=0.3913

P=0.5403

overlapped but not necessarily concordant (Fig. 2-B); aldosterone was detected only in tumors while amlodipine in both tumors and adjacent adrenal glands.

3.3. Comparative analysis of amlodipine accumulation in adrenal glands and adipose tissues

Amlodipine accumulation was not significantly different between tumor and adjacent adrenal gland in Case 1 (P = 0.8380) and in Case 3 (P = 0.4981) (Fig. 3-A). Peri-adrenal adipose tissues were collected from 2 patients (Case 2 and Case 3) treated with amlodipine before adrenal- ectomy in order to further explore the differences of amlodipine accu- mulation between steroidogenic (tumor and adrenal gland) tissues and non-steroidogenic (adipose) tissues (Fig. 3-B). Amlodipine specific peak was detected in both types of the tissues above. Significant dif- ferences of amlodipine accumulation were also detected between

steroidogenic and non-steroidogenic tissues in both cases (P < 0.05).

3.4. Quantification of steroids and amlodipine using LC-MS analysis

Amlodipine and 17 adrenocortical steroids were evaluated by LC-MS/MS (Supplemental Fig. 6) in the quantitative fashion (Table 2). Amlodipine accumulation in the adrenal and adipose tissues was confirmed and was higher in adrenal gland than peri-adrenal adipose tissue (Case 2: 8490.82 ng/g in tumor, and 545.76 ng/g in peri-adrenal adipose tissue, Case3: 27405.26 ng/g in tumor, 5225.82 ng/g in adja- cent adrenal gland, and 640.07 ng/g in peri-adrenal adipose tissue). In addition, there were no consistent differences of amlodipine concen- trations between tumor and adjacent adrenal gland. (Case 1: 1617.74 ng/g in tumor, and 3210.11 ng/g in adjacent adrenal gland, Case 3: 27405.26 ng/g in tumor, and 5225.82 ng/g in adjacent adrenal gland). Amlodipine was also not detected (background noise was only detected

Fig. 5. Comparison of activities of steroidogenic enzymes.

HSD3B

CYP21A2

Prog/Preg

17a-OH-Prog /17a-OH-Preg

11-deoxyB/Prog

11-deoxyF/17a-OH-Prog

F/21-deoxyF

=

E

ca

!

YES

NO

YES

NO

YES

NO

YES

NO

YES

NO

P=0.0200

P=0.0373

P=0.7133

P=0.0373

P=0.1799

CYP11B1

CYP11B2

B/11-deoxyB

F/11-deoxyF

11b-OH-Prog/Prog

18-OHB/B

Aldosterone/18-OHB

18-OHF/F

15

i

B

:

E

E

YES

NO

YES

NO

YES

NO

YES

NO

YES

NO

YES

NO

P=0.7133

P=0.9025

P=0.7133

P=0.9025

P=0.1779

P=0.2703

CYP17A1

17a-OH-Prog/Prog

17a-OH-Preg/Preg

21-deoxyF/11b-OH-Prog

DHEA/17-OH-Preg

=

=

YES

NO

YES

NO

YES

NO

YES

NO

P=0.3913

P=1.000

P=0.2703

P=0.9025

HSD11B1

HSD11B2

6ß-hydroxylase 6b-OHF/F

20a-reductase 20a-DHF/F

F/E

E/F

Administrated tumor

: 4 cases

Not-administrated tumor

YES

NO

YES

NO

YES

NO

YES

NO

: 5 cases

P=0.5403

P=0.5403

P=0.0200

P=0.1779

Abbreviation; YES (administered tumor), NO (non-administered tumor). Both of progesterone/pregnenolone and 17«-progesterone/17«-pregnenolone, which indicated HSD3B activity, were significantly lower in amlodipine-administrated cases. In addition, 11-deoxycortisol/progesterone and 66-hydroxycortisol/cortisol, which indicated the enzyme activity of CYP21A2 and 66-hydroxylase respectively, were increased.

without a specific signal corresponding to amlodipine) in non-treated cases.

In steroid hormones quantification, tissue concentrations of all adrenocortical steroid hormones examined in this study were higher in adrenal gland (Fig. 4-A). In addition, there were no significant differ- ences of adrenocortical steroids concentrations in the adrenals between amlodipine treated and non-treated cases (Fig. 4-B).

3.5. Activities of steroidogenic enzymes examined by metabolic ratios in APA patients

Activity of steroidogenic enzymes in tumors was tentatively analyzed by calculating the metabolic ratio as mentioned above (Fig. 5). Only HSD3B activity was significantly lower in those treated with amlodipine compared to non-treated cases. 11-deoxycortisol/progesterone, which partly corresponded to CYP21A2 activity, was increased but 11-deoxy- corticosterone/progesterone and cortisol/21-deoxycortisol, which also correspond to CYP21A2 activity, were not different between amlodipine treated and non-treated cases. In addition, 66-hydroxycortisol/cortisol, which corresponded to the enzyme activity of 66-hydroxylase, was higher in treated cases.

3.6. In vitro study of amlodipine’s effects on steroidogenic enzyme mRNA level in H295R cells

mRNA levels of CYP11B2, HSD3B1, and HSD3B2 in H295R cells treated with different doses of amlodipine (0 pmol/L, 0.001 µmol/L, 0.1 umol/L, 10 umol/L) were examined using qRT-PCR. In H295R cells without ANG II treatment, mRNA levels of CYP11B2 and HSD3B1 were both significantly decreased with 10 umol/L of amlodipine solution (P < 0.05), but not in HSD3B2 (Fig. 6-A). In contrast, mRNA levels of all the enzymes examined (CYP11B2, HSD3B1 and HSD3B2) in H295R cells treated with ANG II and 10 umol/L of amlodipine solution were

significantly decreased (P < 0.05) (Fig. 6-B). In addition, mRNA levels of CYP11B2 were significantly decreased even by the treatment with 0.1 umol/L amlodipine.

3.7. Immunoreactivity of steroidogenic enzymes (HSD3B1 and HSD3B2) and calcium channels (CaV1.2 and CaV1.3)

In this study, immunoreactivity of HSD3Bs (HSD3B1 and HSD3B2) and calcium channels (CaV1.2 and CaV1.3) were examined. The repre- sentative images of H&E staining and individual IHC sections in areas of the tumor (Fig. 7-A), the adjacent adrenal (Fig. 7-B), and the peri- adrenal adipose tissues (Fig. 7-C) from Case 3 (treated case) and Case 6 (non-treated case) were illustrated. In all the cases examined, CaV1.2, CaV1.3, HSD3B1 and HSD3B2 were ubiquitously immunolocalized in tumor areas regardless of amlodipine treatment before adrenalectomy. In the adjacent adrenal gland, CaV1.2 was immunolocalized predomi- nantly in ZG but also in ZF and ZR. CaV1.3 was predominantly immu- nolocalized in ZG and ZR but also in ZF. HSD3B1 was immunolocalized in ZG and ZR but also in ZF, although its immunoreactivity was gener- ally weak throughout the adjacent adrenal gland. HSD3B2 was immu- nolocalized predominantly in ZF but also in ZG and ZR. In peri-adrenal adipose tissue, the immunoreactivity of CaV1.2, HSD3B1, and HSD3B2 was extremely weak. On the other hand, CaV1.3 immunoreactivity was relatively marked in peri-adrenal adipose tissue.

In all of the 9 cases examined in this study, CaV1.2 and CaV1.3 were both significantly lower in peri-adrenal adipose tissue than both tumors and adjacent adrenal glands (P < 0.05), but no significant differences were detected between tumors and adjacent adrenal glands (Fig. 7-D).

Of particular note, there were no significant differences of calcium- channels and HSD3Bs between amlodipine treated and non-treated cases. However, CaV1.3 tended to be higher and HSD3B1 lower in amlodipine treated APA cases (Fig. 7-E, F) but the differences did not reach statistical significance.

A

Fig. 6. Effects of amlodipine treatment on the mRNA levels of steroidogenic enzymes in H295R cells.

CYP11B2

CYP11B2/RPL13A

0.02

*

0.015

0.01

0.005

0

DMSO

0.001 umol/L amlodipine

0.1 umol/L amlodipine

10 umol/L amlodipine

HSD3B1

HSD3B2

2.5

18

16

HSD3B1/RPL13A

2

HSD3B2/RPL13A

14

1.5

*

12

10

1

8

6

0.5

4

2

0

0

DMSO

0.001 umol/L

0.1 umol/L amlodipine

10 umol/L amlodipine

DMSO

0.001 umol/L amlodipine

0.1 umol/L

10 umol/L

amlodipine

amlodipine

amlodipine

* P < 0.05

The mRNA levels of CYP11B2, HSD3B1 and HSD3B2 after the treatment with different doses of amlodipine standard solution (0 pmol/ L, 0.001 µmol/L, 0.1 umol/L, 10 umol/L) were summarized. In order to determine whether amlodipine treatment could result in significant changes of mRNA levels, a statistical compari- son of two-group was performed with cells treated without amlodipine (0 umol/L treat- ment). A: The transition of mRNA levels of the enzymes under the absence of 0.1 µmol/L ANG II. B: The transition of mRNA levels under the presence of 0.1 umol/L ANG II.

B

CYP11B2

0.03

CYP11B2/RPL13A

0.025

*

0.02

*

0.015

0.01

0.005

0

DMSO+ANG II

0.001 umol/L

0.1 umol/L

10 umol/L

amlodipine + ANG Il amlodipine + ANG Il amlodipine + ANG II

HSD3B1

HSD3B2

2

18

HSD3B1/RPL13A

HSD3B2/RPL13A

16

1.5

*

14

12

*

1

10

8

6

0.5

4

2

0

0

DMSO+ANG II

0.001 µmol/L

0.1 µmol/L

10 umol/L

DMSO+ANG II

0.001 umol/L

0.1 µmol/L

10 umol/L

amlodipine + ANG II amlodipine + ANG Il amlodipine + ANG II

amlodipine + ANG II amlodipine + ANG II amlodipine + ANG II

* P < 0.05

The analysis of correlation between individual calcium channel subtypes and HSD3B isoforms also revealed the significantly positive correlation between CaV1.2 and HSD3B2 (p = 0.6833, P = 0.0424) (Fig. 7-G).

4. Discussion

This is the first study to specifically visualize amlodipine in human adrenal tissues of PA patients treated with amlodipine before surgery using MALDI-MSI. Amlodipine was predominantly accumulated in the adrenal tissues, which were also subsequently confirmed by quantitative LC-MS analysis. In addition, we also obtained steroidogenic enzyme activities by examining metabolic ratios and compared the findings between amlodipine treated and non-treated cases.

Amlodipine was ubiquitously detected throughout the adrenal glands and its localization was not necessarily completely overlapped with aldosterone nor CYP11B2 immunolocalization (Fig. 1-A, 2-B). However, these findings above were also consistent with results of the previous studies that amlodipine was markedly accumulated within cell membrane [9-11]. In addition, immunoreactivity of CaV1.2 and CaV1.3 on the frozen sections was consistent with the intra adrenal distribution of amlodipine (Supplemental Fig. 4). In contrast, in our present study,

amlodipine accumulation was significantly higher in adrenal tissues than peri-adrenal adipose tissues (Fig. 3-A, B), which was also subse- quently verified by LC-MS analysis in our present study (Table 2). From the examination on FFPE sections, both CaV1.2 and CaV1.3 were pre- dominantly detected in adrenal glands compared to peri-adrenal adipose tissues but there were no significant differences of CaV1.2 and CaV1.3 immunoreactivity between tumor and adjacent adrenal glands (Fig. 7-D), which was also consistent with the intra-adrenal distribution of amlodipine above. These findings above also suggested that amlodi- pine localization could be influenced by the expression status of L-type calcium channels and augmented the hypothesis that amlodipine could have direct effects to the steroidogenesis via the calcium channels in human adrenal glands.

Comprehensive analysis of the activities of steroidogenic enzymes did demonstrate the lower HSD3B activity in amlodipine treated pa- tients (Fig. 5). The decrease of both HSD3B1 and HSD3B2 mRNA levels under amlodipine treatments was also confirmed by our in vitro study (Fig. 6-A, B). In addition, HSD3B1 immunoreactivity tended to be lower in amlodipine treated cases, although not significantly different (Fig. 7- F). Calcium was reported as a second messenger of HSD3B transcription [44-46], and amlodipine was reported to suppress aldosterone pro- duction as previously demonstrated by in vitro and in vivo studies [21,

A

Case 3

H&E staining

CaV1.2 IHC

CaV1.3 IHC

HSD3B1 IHC

HSD3B2 IHC

Case 6

H&E staining

CaV1.2 IHC

CaV1.3 IHC

HSD3B1 IHC

HSD3B2 IHC

B

Case 3

ZR

3

ZR ZF ZG

ZG

H&E staining

CaV1.2 IHC

CaV1.3 IHC

HSD3B1 IHC

HSD3B2 IHC

Case 6

ZR

ZR

ZR

ZR

ZF

ZF

ZG

ZG

H&E staining

CaV1.2 IHC

CaV1.3 IHC

HSD3B1 IHC

HSD3B2 IHC

C

Case 3

H&E staining

CaV1.2 IHC

CaV1.3 IHC

HSD3B1 IHC

HSD3B2 IHC

Case 6

H&E staining

CaV1.2 IHC

CaV1.3 IHC

HSD3B1 IHC

HSD3B2 IHC

Fig. 7. IHC of HSD3Bs and calcium channels.

D

160

200

180

140

120

150

160

CaV1.2

100

CaV1.2

CaV1.2

140

80

100

120

60

100

40

50

80

20

Adipose

Tumor

Adipose

Adjacent Adrenal

Adjacent Adrenal

Tumor

tissue

tissue

tissue

p=0.0036

p=0.0059

p=0.5254

200

200

200

150

150

150

CaV1.3

CaV1.3

CaV1.3

100

100

100

50

50

50

0

0

Adipose

Tumor

Adipose

Adjacent Adrenal

0

Adjacent Adrenal

Tumor

tissue

tissue

tissue

p=0.0423

p=0.0081

p=0.2443

Representative images of Case 3 (amlodipine treated case) and Case 6 (amlodipine non-treated case) were illustrated. A: Representative images of H&E staining, and IHC in tumor. B: Representative images of H&E staining, and IHC in adjacent adrenal. C: Representative images of H&E staining, and IHC in peri-adrenal adipose. D: Statistical comparison of CaV1.2 and CaV1.3 between the different types of tissues. CaV1.2 and CaV1.3 were compared between adipose (9 cases), adjacent adrenal gland (7 cases) and tumor (9 cases). CaV1.2 and CaV1.3 immunoreactivity in peri-adrenal adipose tissues was signifi- cantly lower than adjacent adrenal gland and tumor (P < 0.05). E: Statistical comparison of CaV1.2 and CaV1.3 between amlodipine treated and non-treated tumors, adjacent adrenal glands, and peri-adrenal adipose tissues. There were no signif- icant differences between these tissues. F: Statistical compari- son of HSD3B1 and HSD3B2 between amlodipine treated and non-treated tumors and adjacent adrenal glands. There were no significant differences between these tissues. G: Statistical analysis of the correlation between the CaV1.2 or CaV1.3 and HSD3B1 or HSD3B2. There was a significant correlation be- tween the CaV1.2 and HSD3B2 in tumors (p = 0.6833, P < 0.05).

Fig. 7. (continued).

E

150

120

140

CaV1.2 in tumor

CaV1.2 in adjacent adrenal

180

130

160

CaV1.2 in adipose

100

120

110

140

80

100

120

60

90

80

100

40

70

No

Yes

80

No

Yes

No

Yes

amlodipine

amlodipine

amlodipine

p=0.9025

p=0.8597

p=0.3913

120

200

CaV1.3 in adjacent adrenal

200

100

CaV1.3 in tumor

CaV1.3 in adipose

150

150

80

100

60

100

40

50

50

20

0

No

Yes

No

Yes

0

No

Yes

amlodipine

amlodipine

amlodipine

p=0.1779

p=0.1116

p=0.0662

110

160

HSD3B1 in tumor

108

150

HSD3B1 in

adjacent adrenal

140

106

130

104

120

102

110

No

Yes

100

No

Yes

amlodipine

amlodipine

p=0.1779

p=0.5959

160

120

140

110

HSD3B2 in tumor

120

HSD3B2 in adjacent adrenal

100

100

90

80

80

70

60

60

40

50

20

No

Yes

40

No

Yes

amlodipine

amlodipine

p=0.9025

p=0.5959

F

G

Tumor

Adjacent adrenal

CaV1.2 H-Score

CaV1.3 H-Score

CaV1.2 H-Score

CaV1.3 H-Score

p =- 0.1167

p=0.1500 p=0.7001

p =- 0.287 p=0.5345

p =- 0.2857

150

108

p=0.7650

150

p=0.5345

108

106

106

100

100

HSD3B1 H-Score

HSD3B1

104

104

H-Score

50

50

102

102

0

0

70

90

110

130

0

50

100

150

200

80

100

120

140

160

180

50

100

150

200

p=0.6833 p=0.0424

p=0.5333 p=0.1392

80

p =- 0.2143 p=0.6445

p =0.5357

140

140

80

*p=0.2152 .

120

120

60

60

100

100

80

80

HSD3B2

40

40

HSD3B2

H-Score

H-Score

60

60

20

20

40

40

20

20

0

0

70

90

110

130

:

0

50

100

150

200

80

100

120

140

160

180

50

100

150

200

22]. Beer NA et al. also reported that serum cortisol level was decreased and dehydroepiandrosterone sulfate level increased in amlodipine treated patients [47]. All of these findings above were consistent with decreased HSD3B activities detected in the tumors of amlodipine treated PA patients in our present study (Supplemental Fig. 7). Meanwhile, CYP11B2 mRNA level in H295R cell was reported to be lowered by amlodipine treatment [17], which was also confirmed in our present in vitro study (Fig. 6-A, B). However, in vivo decrement of enzyme activity was not necessarily detected in amlodipine treated cases (Fig. 5). KCNJ5, CACNA1D, ATP1A1 and ATP2B3 gene mutations in APA have been reported to contribute to the autonomous overproduction of aldosterone through the sustained calcium influx into cells [16-19]. These mechanisms of aldosterone overproduction were generally considered as one of the APA-specific genetic alterations, which could be different from those in adrenocortical carcinoma cell lines. Because of these APA-specific gene mutations, CYP11B2 activity in the tumor tissue of individual patients may not harbor these significant changes. In addition, one of the metabolic ratios indicating CYP21A2 and 60-hy- droxylase activities were higher in amlodipine treated cases in LC-MS analysis (Fig. 5). Ikeda K et al. reported the increased CYP21A2 expression following CCB administration in their in vitro study, which was also consistent with our present in vivo results [48]. The above changes in steroid synthase expressions and enzyme activity could have influenced the steroid profiles in the adrenal gland. Meanwhile, the increased enzymatic activity of 66-hydroxylase was consistent with the decreased serum cortisol levels as reported in CCB treated patients [47], but it is also true that Ma B et al. reported that CYP3A (66-hydroxylase) activity was decreased in CCBs treated cells [49]. In addition, the studies which indicated the increase of 66-hydroxylase by the CCB adminis- tration have not been reported to the best of our knowledge. Therefore, effects of CCB on the 66-hydroxylase have still remained unknown. Further investigations are therefore warranted for clarification.

CaV1.3 was reported to regulate aldosterone biosynthesis [15]. However, CaV1.3 was not necessarily correlated with HSD3Bs in our present study (Fig. 7-G). HSD3B activity was decreased in adrenal glands of amlodipine treated PA patients and CaV1.2 and HSD3B2 were significantly positively correlated in the tumor (Fig. 7-G). In addition, the intra-adrenal immunolocalization of CaV1.2 and HSD3Bs were partly overlapped (Fig. 7-A, B, C) [12]. These results above indicated that amlodipine could regulate aldosterone biosynthesis by mainly inhibiting CaV1.2 as originally reported [7], even in the adrenocortical tissues. In addition, CaV1.2 deficiency was reported to result in Timothy syndrome, characterized by cardiac arrhythmias and adrenal gland dysfunction [50]. This study of Timothy syndrome did indicate that CaV1.2 was also important for not only regulating aldosterone biosyn- thesis but also maintaining the overall adrenocortical functions, but it awaits further investigations for clarification including the study of larger cohort of APA patients.

5. Conclusion

We firstly achieved the intra-adrenal visualization of amlodipine in APA patients treated with amlodipine before adrenalectomy using MALDI-MSI method in human adrenal tissue sections obtained and subsequently confirmed the results using LC-MS analysis in the same adrenal specimens. Despite the relatively small number of PA cases examined, comprehensive analysis of tissue contents of 17 adrenocor- tical steroids firstly revealed the effects of amlodipine on aldosterone production through inhibiting HSD3B activity. In addition, amlodipine distribution was more abundant in steroidogenic tissue (adrenal tissue) than non-steroidogenic tissue (peri-adrenal adipose tissue), consistent with L-type calcium channel distribution in human adrenal glands. However, further studies are warranted to clarify the underlying mechanisms of direct effects of amlodipine on aldosterone biosynthesis.

Authors statement

Naoki Motomura: Conceptualization, Methodology, Formal anal- ysis, Investigation, Data Curation, Writing - Original Draft, Visualiza- tion. Yuto Yamazaki: Conceptualization, Methodology, Formal analysis, Review & Editing, Visualization, Supervision, Project admin- istration, Funding acquisition. Xin Gao: Formal analysis, Investigation. Yuta Tezuka: Resources, Writing - Review & Editing. Kei Omata: Resources, Writing - Review & Editing. Yoshikiyo Ono: Resources, Writing - Review & Editing. Ryo Morimoto: Resources, Writing - Review & Editing. Fumitoshi Satoh: Conceptualization, Resources, Writing - Review & Editing. Yasuhiro Nakamura: Writing - Review & Editing. Jaeyoon Shim: Methodology & Resources. Man-Ho Choi: Conceptualization, Methodology, Investigation, Resources, Writing - Review & Editing. Akihiro Ito: Writing - Review & Editing. Hironobu Sasano: Conceptualization, Methodology, Writing - Review & Editing, Supervision, Project administration, Funding acquisition.

Funding

This study was supported by grants from the Ministry of Health, Labor, and Welfare, Japan (No. 20FC1020).

Declaration of Competing Interest

The authors report no declarations of interest.

Acknowledgments

This work was technically supported by Mr. Takushi Yamamoto, Dr. Eiichi Matsuo, Dr. Kana Waki in Global Application Development Cen- ter, Analytical & Measuring Instruments Division, Shimadzu Corporation.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jsbmb.2022.106062.

References

[1] J.W. Funder, Medicine. The genetics of primary aldosteronism, Science 331 (Feb (6018)) (2011) 685-686, https://doi.org/10.1126/science.1202887.

[2] G.P. Rossi, G. Bernini, C. Caliumi, G. Desideri, B. Fabris, C. Ferri, C. Ganzaroli, G. Giacchetti, C. Letizia, M. Maccario, F. Mallamaci, M. Mannelli, M.J. Mattarello, A. Moretti, G. Palumbo, G. Parenti, E. Porteri, A. Semplicini, D. Rizzoni, E. Rossi, M. Boscaro, A.C. Pessina, F. Mantero, PAPY Study Investigators, A prospective study of the prevalence of primary aldosteronism in 1,125 hypertensive patients, J. Am. Coll. Cardiol. 48 (Dec 5 (11)) (2006) 2293-2300, https://doi.org/10.1016/ j.jacc.2006.07.059. Epub 2006 Nov 13.

[3] J.S. Williams, G.H. Williams, A. Raji, X. Jeunemaitre, N.J. Brown, P.N. Hopkins, P. R. Conlin, Prevalence of primary hyperaldosteronism in mild to moderate hypertension without hypokalaemia, J. Hum. Hypertens. 20 (Feb (2)) (2006) 129-136, https://doi.org/10.1038/sj.jhh.1001948.

[4] P. Milliez, X. Girerd, P.F. Plouin, J. Blacher, M.E. Safar, J.J. Mourad, Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism, J. Am. Coll. Cardiol. 45 (Apr 19 (8)) (2005) 1243-1248, https://doi.org/10.1016/ j.jacc.2005.01.015.

[5] M. Stowasser, J. Sharman, R. Leano, R.D. Gordon, G. Ward, D. Cowley, T. H. Marwick, Evidence for abnormal left ventricular structure and function in normotensive individuals with familial hyperaldosteronism type I, J. Clin. Endocrinol. Metab. 90 (Sep (9)) (2005) 5070-5076, https://doi.org/10.1210/ jc.2005-0681. Epub 2005 Jun 7.

[6] M. Tiwaskar, A. Langote, R. Kashyap, A. Toppo, Amlodipine in the era of new generation calcium channel blockers, J. Assoc. Physicians India 66 (Mar (3)) (2018) 64-69.

[7] M.J. Sinnegger-Brauns, A. Hetzenauer, I.G. Huber, E. Renström, G. Wietzorrek, S. Berjukov, M. Cavalli, D. Walter, A. Koschak, R. Waldschütz, S. Hering, S. Bova, P. Rorsman, O. Pongs, N. Singewald, J. Striessnig, Isoform-specific regulation of mood behavior and pancreatic beta cell and cardiovascular function by L-type Ca 2 + channels, J. Clin. Invest. 113 (May (10)) (2004) 1430-1439, https://doi.org/ 10.1172/JCI20208.

[8] D. Lipscombe, T.D. Helton, W. Xu, L-type calcium channels: the low down, J. Neurophysiol. 92 (Nov (5)) (2004) 2633-2641, https://doi.org/10.1152/ jn.00486.2004.

[9] K.G. Bulsara, M. Cassagnol, Amlodipine. StatPearls [Internet], StatPearls Publishing, Treasure Island (FL), 2021. Apr 7, 2021 Jan -.

[10] R.P. Mason, P. Marche, T.H. Hintze, Novel vascular biology of third-generation L- type calcium channel antagonists: ancillary actions of amlodipine, Arterioscler. Thromb. Vasc. Biol. 23 (Dec (12)) (2003) 2155-2163, https://doi.org/10.1161/01. ATV.0000097770.66965.2A. Epub 2003 Sep 25.

[11] R.P. Mason, D.G. Rhodes, L.G. Herbette, Reevaluating equilibrium and kinetic binding parameters for lipophilic drugs based on a structural model for drug interaction with biological membranes, J. Med. Chem. 34 (Mar (3)) (1991) 869-877, https://doi.org/10.1021/jm00107a001.

[12] S.J. Felizola, T. Maekawa, Y. Nakamura, F. Satoh, Y. Ono, K. Kikuchi, S. Aritomi, K. Ikeda, M. Yoshimura, K. Tojo, H. Sasano, Voltage-gated calcium channels in the human adrenal and primary aldosteronism, J. Steroid Biochem. Mol. Biol. (2014), https://doi.org/10.1016/j.jsbmb.2014.08.012. Oct;144 Pt B:410-6. Epub 2014 Aug 23.

[13] N. Cherradi, M.F. Rossier, M.B. Vallotton, A.M. Capponi, Calcium stimulates intramitochondrial cholesterol transfer in bovine adrenal glomerulosa cells, J. Biol. Chem. 271 (Oct 18 (42)) (1996) 25971-25975, https://doi.org/10.1074/ jbc.271.42.25971.

[14] T. Yang, M. He, C. Hu, Regulation of aldosterone production by ion channels: from basal secretion to primary aldosteronism, Biochim Biophys Acta Mol Basis Dis. 1864 (Mar (3)) (2018) 871-881, https://doi.org/10.1016/j.bbadis.2017.12.034. Epub 2017 Dec 26.

[15] C.B. Xie, L.H. Shaikh, S. Garg, G. Tanriver, A.E. Teo, J. Zhou, C. Maniero, W. Zhao, S. Kang, R.B. Silverman, E.A. Azizan, M.J. Brown, Regulation of aldosterone secretion by Cav1.3, Sci. Rep. 6 (Apr 21) (2016) 24697, https://doi.org/10.1038/ srep24697.

[16] M. Choi, U.I. Scholl, P. Yue, P. Björklund, B. Zhao, C. Nelson-Williams, W. Ji, Y. Cho, A. Patel, C.J. Men, E. Lolis, M.V. Wisgerhof, D.S. Geller, S. Mane, P. Hellman, G. Westin, G. Åkerström, W. Wang, T. Carling, R.P. Lifton, K+ channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension, Science 331 (Feb 11 (6018)) (2011) 768-772, https://doi.org/ 10.1126/science.1198785.

[17] E.A. Azizan, H. Poulsen, P. Tuluc, J. Zhou, M.V. Clausen, A. Lieb, C. Maniero, S. Garg, E.G. Bochukova, W. Zhao, L.H. Shaikh, C.A. Brighton, A.E. Teo, A. P. Davenport, T. Dekkers, B. Tops, B. Küsters, J. Ceral, G.S. Yeo, S.G. Neogi, I. McFarlane, N. Rosenfeld, F. Marass, J. Hadfield, W. Margas, K. Chaggar, M. Solar, J. Deinum, A.C. Dolphin, I.S. Farooqi, J. Striessnig, P. Nissen, M.J. Brown, Somatic mutations in ATP1A1 and CACNA1D underlie a common subtype of adrenal hypertension, Nat. Genet. 45 (Sep (9)) (2013) 1055-1060, https://doi.org/ 10.1038/ng.2716. Epub 2013 Aug 4.

[18] F. Beuschlein, S. Boulkroun, A. Osswald, T. Wieland, H.N. Nielsen, U. D. Lichtenauer, D. Penton, V.R. Schack, L. Amar, E. Fischer, A. Walther, P. Tauber, T. Schwarzmayr, S. Diener, E. Graf, B. Allolio, B. Samson-Couterie, A. Benecke, M. Quinkler, F. Fallo, P.F. Plouin, F. Mantero, T. Meitinger, P. Mulatero, X. Jeunemaitre, R. Warth, B. Vilsen, M.C. Zennaro, T.M. Strom, M. Reincke, Somatic mutations in ATP1A1 and ATP2B3 lead to aldosterone-producing adenomas and secondary hypertension, Nat. Genet. 45 (Apr (4)) (2013), https:// doi.org/10.1038/ng.2550, 440-4, 444e1-2. Epub 2013 Feb 17.

[19] K. Omata, F. Satoh, R. Morimoto, S. Ito, Y. Yamazaki, Y. Nakamura, S.K. Anand, Z. Guo, M. Stowasser, H. Sasano, S.A. Tomlins, W.E. Rainey, Cellular and genetic causes of idiopathic hyperaldosteronism, Hypertension 72 (Oct (4)) (2018) 874-880, https://doi.org/10.1161/HYPERTENSIONAHA.118.11086.

[20] T. Isaka, K. Ikeda, Y. Takada, Y. Inada, K. Tojo, N. Tajima, Azelnidipine inhibits aldosterone synthesis and secretion in human adrenocortical cell line NCI-H295R, Eur. J. Pharmacol. 605 (Mar 1 (1-3)) (2009) 49-52, https://doi.org/10.1016/j. ejphar.2008.12.041. Epub 2009 Jan 10.

[21] T. Yang, M. He, H. Zhang, P.Q. Barrett, C. Hu, L- and T-type calcium channels control aldosterone production from human adrenals, J. Endocrinol. 244 (Jan (1)) (2020) 237-247, https://doi.org/10.1530/JOE-19-0259.

[22] P. Mulatero, F. Rabbia, A. Milan, C. Paglieri, F. Morello, L. Chiandussi, F. Veglio, Drug effects on aldosterone/plasma renin activity ratio in primary aldosteronism, Hypertension 40 (Dec (6)) (2002) 897-902, https://doi.org/10.1161/01. hyp.0000038478.59760.41.

[23] T. Nishikawa, M. Omura, F. Satoh, H. Shibata, K. Takahashi, N. Tamura, A. Tanabe, Task Force Committee on Primary Aldosteronism, The Japan Endocrine Society, Guidelines for the diagnosis and treatment of primary aldosteronism-the Japan Endocrine Society 2009, Endocr. J. 58 (9) (2011) 711-721, https://doi.org/ 10.1507/endocrj.ej11-0133. Epub 2011 Aug 9.

[24] C.E. Gomez-Sanchez, X. Qi, C. Velarde-Miranda, M.W. Plonczynski, C.R. Parker, W. Rainey, F. Satoh, T. Maekawa, Y. Nakamura, H. Sasano, E.P. Gomez-Sanchez, Development of monoclonal antibodies against human CYP11B1 and CYP11B2, Mol. Cell. Endocrinol. 383 (Mar 5 (1-2)) (2014) 111-117, https://doi.org/ 10.1016/j.mce.2013.11.022. Epub 2013 Dec 8.

[25] Y. Yamazaki, K. Omata, Y. Tezuka, Y. Ono, R. Morimoto, Y. Adachi, K. Ise, Y. Nakamura, C.E. Gomez-Sanchez, Y. Shibahara, T. Kitamoto, T. Nishikawa, S. Ito, F. Satoh, H. Sasano, Tumor cell subtypes based on the intracellular hormonal activity in KCNJ5-mutated aldosterone-producing adenoma, Hypertension 72 (Sep (3)) (2018) 632-640, https://doi.org/10.1161/HYPERTENSIONAHA.118.10907.

[26] Y. Yamazaki, Y. Nakamura, K. Omata, K. Ise, Y. Tezuka, Y. Ono, R. Morimoto, Y. Nozawa, C.E. Gomez-Sanchez, S.A. Tomlins, W.E. Rainey, S. Ito, F. Satoh, H. Sasano, Histopathological classification of cross-sectional image-negative

hyperaldosteronism, J. Clin. Endocrinol. Metab. 102 (Apr 1 (4)) (2017) 1182-1192, https://doi.org/10.1210/jc.2016-2986.

[27] S. Konosu-Fukaya, Y. Nakamura, F. Satoh, S.J. Felizola, T. Maekawa, Y. Ono, R. Morimoto, K. Ise, K. Takeda, K. Katsu, F. Fujishima, A. Kasajima, M. Watanabe, Y. Arai, E.P. Gomez-Sanchez, C.E. Gomez-Sanchez, M. Doi, H. Okamura, H. Sasano, 36-Hydroxysteroid dehydrogenase isoforms in human aldosterone-producing adenoma, Mol. Cell. Endocrinol. 408 (Jun 15) (2015) 205-212, https://doi.org/ 10.1016/j.mce.2014.10.008. Epub 2014 Oct 22.

28] K.S. McCarty Jr, L.S. Miller, E.B. Cox, J. Konrath, K.S. McCarty Sr, Estrogen receptor analyses. Correlation of biochemical and immunohistochemical methods using monoclonal antireceptor antibodies, Arch. Pathol. Lab. Med. 109 (Aug (8)) (1985) 716-721.

[29] A. Jakimska, M. Śliwka-Kaszyńska, P. Nagórski, J. Namieśnik, A. Kot-Wasik, Phototransformation of amlodipine: degradation kinetics and identification of its photoproducts, PLoS One 9 (Oct 3 (10)) (2014), e109206, https://doi.org/ 10.1371/journal.pone.0109206.

[30] S. Shimma, H.O. Kumada, H. Taniguchi, A. Konno, I. Yao, K. Furuta, T. Matsuda, S. Ito, Microscopic visualization of testosterone in mouse testis by use of imaging mass spectrometry, Anal. Bioanal. Chem. 408 (Nov (27)) (2016) 7607-7615, https://doi.org/10.1007/s00216-016-9594-9. Epub 2016 May 26. Erratum in: Anal Bioanal Chem. 2016 Nov;408(27):7881.

[31] S. Shimma, Y. Takashima, J. Hashimoto, K. Yonemori, K. Tamura, A. Hamada, Alternative two-step matrix application method for imaging mass spectrometry to avoid tissue shrinkage and improve ionization efficiency, J. Mass Spectrom. 48 (Dec (12)) (2013) 1285-1290, https://doi.org/10.1002/jms.3288.

[32] Y. Sugiura, E. Takeo, S. Shimma, M. Yokota, T. Higashi, T. Seki, Y. Mizuno, M. Oya, T. Kosaka, M. Omura, T. Nishikawa, M. Suematsu, K. Nishimoto, Aldosterone and 18-oxocortisol coaccumulation in aldosterone-producing lesions, Hypertension 72 (6) (2018), https://doi.org/10.1161/HYPERTENSIONAHA.118.11243.

[33] E. Takeo, Y. Sugiura, T. Uemura, K. Nishimoto, M. Yasuda, E. Sugiyama, S. Ohtsuki, T. Higashi, T. Nishikawa, M. Suematsu, E. Fukusaki, S. Shimma, Tandem mass spectrometry imaging reveals distinct accumulation patterns of steroid structural isomers in human adrenal glands, Anal. Chem. 91 (Jul 16 (14)) (2019) 8918-8925, https://doi.org/10.1021/acs.analchem.9b00619. Epub 2019 Jun 26.

[34] C. Lee, J.H. Kim, S.J. Moon, J. Shim, H.I. Kim, M.H. Choi, Selective LC-MRM/SIM- MS based profiling of adrenal steroids reveals metabolic signatures of 17a- hydroxylase deficiency, J. Steroid Biochem. Mol. Biol. 198 (Apr) (2020) 105615, https://doi.org/10.1016/j.jsbmb.2020.105615. Epub 2020 Jan 31.

[35] S.H. Kim, J.Y. Moon, H. Sasano, M.H. Choi, M.J. Park, Body fat mass is associated with ratio of steroid metabolites reflecting 17,20-lyase activity in prepubertal girls, J. Clin. Endocrinol. Metab. 101 (Dec (12)) (2016) 4653-4660, https://doi.org/ 10.1210/jc.2016-2515. Epub 2016 Sep 20.

[36] X. Gao, Y. Yamazaki, Y. Tezuka, Y. Onodera, H. Ogata, K. Omata, R. Morimoto, Y. Nakamura, F. Satoh, H. Sasano, The crosstalk between aldosterone and calcium metabolism in primary aldosteronism: a possible calcium metabolism-associated aberrant “neoplastic” steroidogenesis in adrenals, J. Steroid Biochem. Mol. Biol. 193 (Oct) (2019) 105434, https://doi.org/10.1016/j.jsbmb.2019.105434. Epub 2019 Jul 24.

[37] S.J.A. Felizola, Y. Nakamura, F. Satoh, R. Morimoto, K. Kikuchi, T. Nakamura, A. Hozawa, L. Wang, Y. Onodera, K. Ise, K.M. McNamara, S. Midorikawa, S. Suzuki, H. Sasano, Glutamate receptors and the regulation of steroidogenesis in the human adrenal gland: the metabotropic pathway, Mol. Cell. Endocrinol. 382 (Jan 25 (1)) (2014) 170-177, https://doi.org/10.1016/j.mce.2013.09.025. Epub 2013 Sep 27.

[38] M. Doi, Y. Takahashi, R. Komatsu, F. Yamazaki, H. Yamada, S. Haraguchi, N. Emoto, Y. Okuno, G. Tsujimoto, A. Kanematsu, O. Ogawa, T. Todo, K. Tsutsui, G. T. van der Horst, H. Okamura, Salt-sensitive hypertension in circadian clock- deficient Cry-null mice involves dysregulated adrenal Hsd3b6, Nat. Med. 16 (Jan (1)) (2010) 67-74, https://doi.org/10.1038/nm.2061. Epub 2009 Dec 13.

[39] A. Azmahani, Y. Nakamura, S.J. Felizola, Y. Ozawa, K. Ise, T. Inoue, K. M. McNamara, M. Doi, H. Okamura, C.C. Zouboulis, S. Aiba, H. Sasano, Steroidogenic enzymes, their related transcription factors and nuclear receptors in human sebaceous glands under normal and pathological conditions, J. Steroid Biochem. Mol. Biol. (Oct;144 Pt B) (2014) 268-279, https://doi.org/10.1016/j. jsbmb.2014.07.010. Epub 2014 Jul 30.

[40] Y. Shibahara, Y. Miki, Y. Onodera, S. Hata, M.S. Chan, C.C. Yiu, T.Y. Loo, Y. Nakamura, J. Akahira, T. Ishida, K. Abe, H. Hirakawa, L.W. Chow, T. Suzuki, N. Ouchi, H. Sasano, Aromatase inhibitor treatment of breast cancer cells increases the expression of let-7f, a microRNA targeting CYP19A1, J. Pathol. 227 (Jul (3)) (2012) 357-366, https://doi.org/10.1002/path.4019. Epub 2012 Apr 30.

[41] Y. Nakamura, S. Suzuki, T. Suzuki, K. Ono, I. Miura, F. Satoh, T. Moriya, H. Saito, S. Yamada, S. Ito, H. Sasano, MDM2: a novel mineralocorticoid-responsive gene involved in aldosterone-induced human vascular structural remodeling, Am. J. Pathol. 169 (Aug (2)) (2006) 362-371, https://doi.org/10.2353/ ajpath.2006.051351.

[42] S.J. Felizola, Y. Nakamura, X.G. Hui, F. Satoh, R. Morimoto, K. M McNamara, S. Midorikawa, S. Suzuki, W.E. Rainey, H. Sasano, Estrogen-related receptor « in normal adrenal cortex and adrenocortical tumors: involvement in development and oncogenesis, Mol. Cell. Endocrinol. 365 (Jan 30 (2)) (2013) 207-211, https://doi. org/10.1016/j.mce.2012.10.020. Epub 2012 Oct 30.

[43] M.W. Pfaffl, A new mathematical model for relative quantification in real-time RT- PCR, Nucleic Acids Res. 29 (May 1 (9)) (2001) e45, https://doi.org/10.1093/nar/ 29.9.e45.

[44] E.F. Nogueira, Y. Xing, C.A. Morris, W.E. Rainey, Role of angiotensin II-induced rapid response genes in the regulation of enzymes needed for aldosterone

synthesis, J. Mol. Endocrinol. 42 (Apr (4)) (2009) 319-330, https://doi.org/ 10.1677/JME-08-0112. Epub 2009 Jan 21.

[45] S. Udhane, P. Kempna, G. Hofer, P.E. Mullis, C.E. Flück, Differential regulation of human 36-hydroxysteroid dehydrogenase type 2 for steroid hormone biosynthesis by starvation and cyclic AMP stimulation: studies in the human adrenal NCI- H295R cell model, PLoS One 8 (Jul 9 (7)) (2013), e68691, https://doi.org/ 10.1371/journal.pone.0068691.

[46] H. Enslen, T.R. Soderling, Roles of calmodulin-dependent protein kinases and phosphatase in calcium-dependent transcription of immediate early genes, J. Biol. Chem. 269 (Aug 19 (33)) (1994) 20872-20877.

[47] N.A. Beer, D.J. Jakubowicz, R.M. Beer, J.E. Nestler, The calcium channel blocker amlodipine raises serum dehydroepiandrosterone sulfate and androstenedione, but lowers serum cortisol, in insulin-resistant obese and hypertensive men, J. Clin.

Endocrinol. Metab. 76 (Jun (6)) (1993) 1464-1469, https://doi.org/10.1210/ jcem.76.6.8501151.

[48] K. Ikeda, T. Saito, K. Tojo, Efonidipine, a Ca(2+)-channel blocker, enhances the production of dehydroepiandrosterone sulfate in NCI-H295R human adrenocortical carcinoma cells, Tohoku J. Exp. Med. 224 (Aug (4)) (2011) 263-271, https://doi.org/10.1620/tjem.224.263.

[49] B. Ma, T. Prueksaritanont, J.H. Lin, Drug interactions with calcium channel blockers: possible involvement of metabolite-intermediate complexation with CYP3A, Drug Metab. Dispos. 28 (Feb (2)) (2000) 125-130.

[50] A. Marcantoni, C. Calorio, E. Hidisoglu, G. Chiantia, E. Carbone, Cav1.2 channelopathies causing autism: new hallmarks on Timothy syndrome, Pflugers Arch. 472 (Jul (7)) (2020) 775-789, https://doi.org/10.1007/s00424-020-02430- 0. Epub 2020 Jul 3.