Inhibition of NPC Intracellular Cholesterol Transporter 1 Dually Regulates Aldosterone Secretion Via the Steroidogenic Acute Regulatory‐Related Lipid Transfer Domain‐3‐Voltage‐Dependent Anion Channel 1 Axis and Inositol 1,4,5‐Trisphosphate Receptor Type 3‐Calcium Signaling

Abstract

Background

Aldosterone‐producing adenomas, a prevalent cause of endocrine hypertension, arise from uncontrolled aldosterone production. NPC1 (NPC intracellular cholesterol transporter 1) is a cholesterol transporter located on the lysosomal limiting membrane. Although cholesterol serves as the primary precursor for aldosterone synthesis, the mechanism governing its supply and metabolism within aldosterone‐producing adenomas remains unclear.

Methods

In this study, we used quantitative proteomics and observed that NPC1 was significantly downregulated in aldosterone‐producing adenoma tissues.

Results

Liquid chromatography/tandem mass spectrometry analysis found that inhibition of NPC1 increased aldosterone secretion in H295R cells. Mechanistically, NPC1 deficiency promoted aldosterone production through 2 pathways: (1) immunofluorescence and coimmunoprecipitation experiments confirmed that NPC1 deficiency enhanced lysosome‐mitochondria interaction via STARD3‐VDAC1 (steroidogenic acute regulatory‐related lipid transfer domain‐3‐voltage‐dependent anion channel 1), leading to mitochondrial cholesterol overload; and (2) Western Blot and calcium measurement showed that NPC1 deficiency activated of cytoplasmic calcium signaling through IP3R3 (inositol 1,4,5‐trisphosphate receptor type 3)‐mediated endoplasmic reticulum calcium release, resulting in upregulated expression of aldosterone synthase.

Conclusions

Our findings demonstrate that NPC1 downregulation represents a novel mechanism driving elevated aldosterone production, linking lysosomal‐mitochondria cholesterol transport to aldosterone high production. These results suggest that NPC1 may offer a new understanding for aldosterone overproduction mechanism of aldosterone‐producing adenomas.

Nonstandard Abbreviations and Acronyms

APA

aldosterone‐producing adenoma

ER

endoplasmic reticulum

NFAT

nonfunctioning adrenocortical adenoma

Research Perspective

What New Question Does This Study Raise?

• NPC1 (NPC intracellular cholesterol transporter 1) is downregulated in aldosterone‐producing adenomas. In H295R cells, NPC1 inhibition increases aldosterone secretion.

• NPC1 inhibition increases aldosterone secretion via enhanced lysosomes‐mitochondria interaction and mitochondrial cholesterol overload mediated by the STARD3‐VDAC1 (steroidogenic acute regulatory‐related lipid transfer domain‐3‐voltage‐dependent anion channel 1) complex; NPC1 inhibition increases aldosterone secretion via upregulated CYP11B2 (cytochrome P450 family 11 subfamily B member 2) expression induced by IP3R3 (inositol 1,4,5‐trisphosphate receptor type 3)‐mediated Ca²⁺ release from the endoplasmic reticulum.

What Question Should Be Addressed Next?

• NPC1 regulatory role in aldosterone production in vivo and the mechanisms underlying reduced NPC1 expression in aldosterone‐producing adenomasremain unclear and require further investigation.

Hypertension and associated cardiovascular/cerebrovascular disorders pose significant threats to global public health. As a common endocrine form of secondary hypertension, primary aldosteronism affects 5% to 10% of populations with hypertension. ¹ Aldosterone‐producing adenoma (APA) is one of the most common causes, accounting for ∼30% of cases of primary aldosteronism. ¹ This syndrome arises from adrenal cortical lesions causing dysregulated aldosterone synthesis, diagnostically characterized by elevated aldosterone levels, suppressed plasma renin activity, and sustained hypertension. Aldosterone, the major mineralocorticoid hormone, mediates systemic blood pressure regulation through renal absorption of sodium and water. Its biosynthesis in the adrenal zona glomerulosa is critically controlled by aldosterone synthase CYP11B2 (cytochrome P450 family 11 subfamily B member 2) expression. Notably, physiological agonists such as angiotensin II, potassium ions, or pathological mutations in genes like potassium inwardly rectifying channel subfamily J member 5 (KCNJ5) can activate intracellular Ca²⁺ signaling pathways that upregulate CYP11B2 transcription and aldosterone production. ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹

Aldosterone biosynthesis in adrenal glomerulosa cells uses cholesterol as the primary precursor, which is acquired through several sources: de novo biosynthesis, mobilization of cholesteryl esters from lipid droplets, low‐density lipoprotein receptor‐mediated endocytosis, and high‐density lipoprotein‐cholesterol uptake. ¹⁰ Angiotensin II not only induces upregulation of low‐density lipoprotein receptor expression but also activates Ca²⁺ signaling. Furthermore, angiotensin II activates the PKD (protein kinase D)‐CREB (cAMP responsive element binding protein) signaling pathway to phosphorylate mitochondrial StAR (steroidogenic acute regulatory protein). This posttranslational modification enhances mitochondrial cholesterol import efficiency, ultimately driving aldosterone production. ¹¹ , ¹² , ¹³ Thus, aldosterone production is physiologically regulated by both Ca²⁺‐CYP11B2 signaling and cholesterol metabolism, suggesting that excessive aldosterone secretion in APA may result from more than just dysregulated Ca²⁺ signaling, which dominates current APA research pathogenesis. ¹⁴ However, the mechanistic links between cholesterol metabolism and APA pathogenesis remain unclear.

NPC1 (NPC intracellular cholesterol transporter 1), a lysosomal transmembrane cholesterol transporter, regulates cholesterol trafficking from lysosomes to other cellular membranes. ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ In Niemann‐Pick disease type C, loss of function in NPC1 results in characteristic lysosomal cholesterol accumulation, disruption of cellular cholesterol homeostasis, ²⁰ , ²¹ , ²² , ²³ and dysfunctional Ca²⁺ signaling, ²⁴ , ²⁵ , ²⁶ the 2 key features supporting aldosterone biosynthesis. However, whether NPC1 regulates aldosterone biosynthesis or how NPC1 modulates this process in adrenal cells remains unknown.

Here, we report that NPC1 is downregulated in APA. In H295R cells, NPC1 inhibition increases aldosterone secretion through 2 mechanisms: (1) enhanced lysosome‐mitochondria interaction and mitochondrial cholesterol overload mediated by the STARD3‐VDAC1 (steroidogenic acute regulatory‐related lipid transfer domain‐3‐voltage‐dependent anion channel 1) complex; and (2) upregulated expression of CYP11B2 induced by IP3R3 (inositol 1,4,5‐trisphosphate receptor type 3)‐mediated Ca²⁺ release from the endoplasmic reticulum. Thus, understanding how NPC1 regulates aldosterone may provide insights into the pathogenesis of APA or other hyperaldosteronism diseases.

METHODS

The data underlying this article are available in the article and its online supplementary material.

Cell Culture

HEK 293T cells were cultured in DMEM (C11995500CP) supplemented with 10% FBS. NCI‐H295R cells were maintained in DMEM/F‐12 (11320033) supplemented with 10% FBS (10099–141) and 1% insulin‐transferrin‐sodium selenite (41400–045). These materials were obtained from Gibco (United States). Both cell lines were cultured in humidified incubators at 37 °C with 5% CO₂.

Plasmids

The coding regions of human STARD3, VDAC1 were polymerase chain reaction (PCR) amplified and cloned into pcDNA3.1 containing 3× Flag or 3× HA tag vector. All mutants were generated by site‐directed mutagenesis. mito‐RCaMP1h (105013, Addgene, USA), ER‐GCaMP6‐150 (86918, Addgene, USA). The target sequences were as follows: Ctrl‐shRNA:CCTAAGGTTAAGTCGCCCTCG, NPC1‐shRNA:CCTAAGGTTAAGTCGCCCTCG, STARD3‐shRNA:AACACAGGCATCCGTAAGAAC.

Reagents and Antibodies

Cholesterol (C3045) and filipin (F9765) were obtained from Sigma Aldrich (Germany). LysoTracker Red DND‐99 (A66439) and MitoTracker Green FM (A66441) were obtained from Thermo Fisher Scientifc (USA). U18666A (HY‐107433), 2‐APB (HY‐W009724), BAPTA (HY‐100168), and Methyl‐β‐cyclodextrin (HY‐101461) were obtained from Medchemexpress (USA). VBIT‐12 (S8936) were obtained from Selleck (USA). Lipid depleted FBS was obtained from VivaCell, China (C3840‐0100). Cal‐520 AM (21130), Mag‐Fluo‐4 AM (20401), and Rhod‐2 AM (21064) were obtained from AAT Bioquest (USA). The primary antibodies used in this study were as follows: mouse anti‐β‐actin (66009‐1‐Ig), rabbit anti‐HA‐tag (51064‐2‐AP), rabbit anti‐Flag‐tag (20543‐1‐AP), and rabbit anti‐ITPR1 (inositol 1,4,5‐trisphosphate receptor type 1; 19962‐1‐AP), which were obtained from Proteintech (China). Other antibodies and their sources were as follows: rabbit anti‐ITPR2 (PA1‐904, Invitrogen, USA); rabbit anti‐TOM20 (sc‐11 415), mouse anti‐LAMP1 (lysosomal‐associated membrane protein 1; sc‐20 011), and mouse anti‐StAR (sc‐166 821) (Santa Cruz Biotechnology, USA); rabbit anti‐STARD3 (ab3478), rabbit anti‐NPC1 (ab134113), and mouse anti‐VDAC1 (ab186321) (Abcam); mouse anti‐ITPR3 (610 312, BD Biosciences, USA), and mouse anti‐CYP11B2 (MABS1251, Millipore, Germany).

Western Blotting

Cells or adrenal tumor were lysed in RIPA buffer containing protease inhibitor cocktail. Lysates were centrifuged at 12000 g for 15 minutes at 4 °C to obtain supernatants. Protein extracts were separated by electrophoresis on 8% to 15% SDS‐PAGE gels and transferred to polyvinylidene fluoride membranes (Millipore, Germany). Membranes were blocked with 5% skim milk for 1 hour at room temperature, then incubated overnight at 4 °C with primary antibodies. Subsequently, membranes were incubated with secondary antibodies for 1 hour at room temperature. After 3 5‐minute washes with TBST, proteins were detected using ECL Western Blotting Substrate (advansta, K‐12045‐D50, USA).

Immunohistochemistry

Adrenal samples were fixed in 4% paraformaldehyde for 48 hours, embedded in paraffin, and sectioned at 5 μm thickness. For immunohistochemistry, sections were deparaffinized, rehydrated, and subjected to antigen retrieval in citrate buffer (pH 9.0). After blocking with 10% goat serum (AR0009, BOSTER, China), sections were incubated overnight at 4 °C with primary antibodies against CYP11B2 and NPC1. Following washes, sections were incubated with horseradish peroxidase‐conjugated secondary antibodies. Signal detection was performed using DAB substrate, and images were captured with an Olympus VS200 slide scanner (Japan).

Immunofluorescence

Cells were fixed with 4% paraformaldehyde for 15 minutes at room temperature, then washed 3 times with cold PBS (5 minutes per wash). Subsequently, cells were permeabilized with 0.1% Triton X‐100 for 10 minutes. After blocking with 1% BSA in PBS for 1 hour at room temperature, cells were incubated with primary antibodies overnight at 4 °C. Following 3 5‐minutes PBS washes, cells were incubated with fluorophore‐conjugated secondary antibodies (Thermo Fisher Scientific, USA) for 1 hour at room temperature. Images were acquired using an Olympus SpinSR confocal microscope (Japan). Images were analyzed on ImageJ (National Institutes of Health) using the colocalization finder plug for automatic extraction of colocalization values.

Immunoprecipitation

To investigate STARD3‐VDAC1 interactions, HEK 293T cells were cotransfected with the indicated plasmids. After 48 hours, cells were lysed on ice with 500 μL lysis buffer (1% Triton X‐100, 5 mM MgCl₂, 100 mM NaCl, 50 mM Tris–HCl) supplemented with protease inhibitor cocktail. Lysates were centrifuged at 12000 g for 15 minutes at 4 °C. Supernatants were incubated with anti‐FLAG M2 magnetic beads (Sigma‐Aldrich, Germany) overnight at 4 °C. Beads were washed 3 times with 1 mL lysis buffer, and immunoprecipitated proteins were eluted by boiling in 2× loading buffer at 95 °C for 10 minutes for subsequent Western blot analysis.

For endogenous interactions, H295R cells were lysed with 500 μL lysis buffer, and lysates were incubated with Protein A/G Magnetic Beads (HY‐K0202, MedChemExpress, USA) conjugated to either STARD3 antibody (20292‐1‐AP, Proteintech, China) or control IgG (2729S, Cell Signaling Technology, Germany) overnight at 4 °C. After 3 washes with lysis buffer, proteins were eluted by boiling in 2× loading buffer and analyzed by Western blotting.

Aldosterone Measurement

Concentration of aldosterone in H295R cell supernatant was measured by an UltriMate Thermo 3000 UPLC and TSQ Endura triple quadrupole mass spectrometer (Thermo Fisher Scientific, USA). Briefly, the supernatant of H295R cells (750 mL) was added with equal volume of premixed solution (0.1 mol/L zinc sulfate and methanol, v/v=1/1) in 2 mL Eppendorf tubes to precipitate the proteins. After vortex mixing and centrifugation, the supernatant was loaded onto a solid‐phase extraction column (Bond Elut C18 cartridge, No. 12102025, Agilent, USA) that had been preactivated and equilibrated with methanol and water, respectively. The column was then rinsed with an alkaline eluent containing ammonia and methanol, and the eluent was discarded. Subsequently, steroids were eluted with methanol and collected into new 2 mL Eppendorf tubes. The eluent was evaporated to dryness under a gentle nitrogen stream and reconstituted with 20% methanol. After another vortex mixing and centrifugation, the supernatant was transferred to a 96‐well plate for liquid chromatography/tandem mass spectrometry analysis.

RNA Isolation and Quantitative Real‐Time PCR

Total RNA was extracted from H295R cells or adrenal tumor using TRIzol (Invitrogen) following the manufacturer’s protocol. Subsequently, 2 μg RNA was reverse transcribed to cDNA using the PrimeScript RT reagent Kit (Takara). Quantitative PCR was performed on the Applied Biosystems 7500 RealTime PCR System with SYBR Green (Takara) according to the manufacturer protocol. Primers were designed using PrimerBank and are listed in the TableTable.

Table 1.

Primer Sequences Used for qPCR

Target	Forward primer	Reverse primer
CYP11A1	AGCCAGCATCAAGGAGACAC	TCTCGGCCCAGAGCATAGAT
CYP21A2	CCTTGCTCAATGCCACCATCG	CTTGGAGGTTCGGAATGATGACTG
HSD3B2	CCACCGTATTGGAGTTGAACA	CGCGGCTAATGTCTCCTGG
CYP11B1	CCGGGTCCCCAGGACAGT	GTACTTCCAGGTGCAGGTCCT
CYP11B2	CCCTCAACACTACACAGGCA	GTCATCAGCAAGGGAAACGC
NPC1	CACTTCTGCTAAAGGACTGGATGAG	CAGGACTGCGATGCTGAATGAC
ITPR1	TCTCAGACCAGAGTACGACTT	CAGACAGCACCCGAATACAG
ITPR2	CCAATCAGCTACTTCTGCTACT	CAAGCTGTTCCACTGTCCT
ITPR3	AGCTGAAGATCCTGGAAATCC	AACACCTCCACAAACTCCTTC
ACTB	GACCCAGATCATGTTTGAGACCTTC	CCAGAGGCGTACAGGGATAGC

CYP11A1 indicates cytochrome P450 family 11 subfamily A member 1; CYP11B1, cytochrome P450 family 11 subfamily B member 1; CYP11B2, cytochrome P450 family 11 subfamily B member 2; CYP21A2, cytochrome P450 family 21 subfamily A member 2; HSD3B2, hydroxy‐delta‐5‐steroid dehydrogenase, 3 beta‐ and steroid delta‐isomerase 2; ITPR1, inositol 1,4,5‐trisphosphate receptor type 1; ITPR2, inositol 1,4,5‐trisphosphate receptor type 2; ITPR3, inositol 1,4,5‐trisphosphate receptor type 3; NPC1, NPC intracellular cholesterol transporter 1; and qPCR, quantitative polymerase chain reaction.

Ca²⁺ Imaging

Cells were incubated with calcium indicators by incubating for 30 minutes at 37 °C in Ringer’s solution (160 mM NaCl, 2 mM CaCl2, 1 mM MgCl2, 2.5 mM KCl, 10 mM HEPES, 8 mM glucose) containing 5 μM Cal‐520 AM, 5 μM Rhod‐2 AM, or 5 μM Mag‐Fluo‐4 AM. Following loading, cells were washed and incubated in indicators‐free Ringer’s solution for 30 minutes to allow complete deesterification. Imaging was performed using an Olympus SpinSR confocal microscope with Cal‐520 AM‐ or Mag‐Fluo‐4 AM‐loaded cells excited at 488 nm, and Rhod‐2 AM‐loaded cells excited at 546 nm. Images were analyzed on ImageJ for automatic extraction of intensity values.

Mitochondrial and ER Ca²⁺ Imaging

To normalize for expression levels of the transfected constructs, control and U18666A‐treated H295R cells were transfected with same amount of either Mito‐RCamPh1 or ER‐GCaMP6‐150 plasmids. Before stimulation, cells were bathed in Ca²⁺‐free Ringer’s solution to ensure baseline conditions. Subsequently, cells were stimulated with Ringer’s solution containing 2 mM Ca²⁺ and imaging was performed using an Olympus SpinSR confocal microscope. Mitochondrial RCamPh1 was excited at 546 nm and ER‐GCaMP6‐150 was excited at 488 nm. Images were analyzed on ImageJ for automatic extraction of intensity values.

Electron Microscopy

Cells were fixed in a mixture of 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M cacodylate buffer for 2 hours at 4 °C. After postfixation in 1% osmium tetroxide followed by 3% uranyl acetate, samples were dehydrated and embedded in TAAB‐812 resin. Ultrathin sections were prepared using an ultramicrotome (Leica UC7, Germany) and imaged with a transmission electron microscope (HT8700, Hitachi, Japan).

For quantification of the percentage of lysosome‐mitochondria contacts, the number of lysosomes in contact with mitochondria was divided by the total number of lysosomes in the region of interest. For quantification of the length of membrane contacts, membrane contacts between lysosomes and mitochondria were defined as areas where the opposing membranes were within 30 nm of each other using ImageJ. ²⁷

Subjects and Sample Collection

All patients were recruited from the CONPASS (Chongqing Primary Aldosteronism Study), which aimed to improve primary aldosteronism management in the Chinese population. ²⁸ Adrenal tumor tissues were collected from patients who were diagnosed with APAs or nonfunctioning adrenocortical adenomas (NFATs) and underwent laparoscopic surgery, with the approval of the Ethics Committee of the first affiliated hospital of Chongqing Medical University (ClinicalTrials.gov number, NCT03224312). Written informed consent was obtained from each individual enrolled.

The adrenal gland where the APA is located and the tumor of NFAT was surgically resected by experienced urological surgeon. The APA was dissected from the whole adrenal gland. All the sections of APA and NFAT were used for pathological staining, somatic mutations sequencing, Western blot, and quantitative PCR. Pathological diagnosis of APA and NFAT was confirmed by CYP11B2 immunohistochemistry and Western blotting using established protocols. ²⁹

Image Analysis

Lysosome‐mitochondria colocalization images were analyzed on ImageJ using the Colocalization finder plug for automatic extraction of colocalization values. The proportion of prolonged lysosome‐mitochondria interactions was quantified as the percentage of lysosomes colocalized with mitochondria lasting >10 seconds in time‐lapse images divided by the total number of lysosomes in the region of interest. ³⁰

Mitochondria Isolation and Cholesterol Measurement

Mitochondria were isolated from H295R cells using the Cell Mitochondrial Isolation Kit (C3601, Beyotime, China) according to the manufacturer’s protocol. Cholesterol was measured using the Amplex Red Cholesterol Assay Kit (A12216, Invitrogen, USA) and normalized to protein concentration.

Statistical Analysis

The statistical data are reported as mean±SD. Relative expression and fluorescence values were log‐transformed using the log₂X data transformation before analysis. The statistical differences between groups was estimated using permutation versions of t test or 2‐way ANOVA. P<0.05 was considered statistically significant.

RESULTS

NPC1 Is Downregulated in APA

To identify key regulators of cholesterol supply for aldosterone biosynthesis, we analyzed existing proteomics data. ²⁹ This analysis revealed no significant changes in cholesterol biosynthesis pathways, ²⁹ but critically, the cholesterol trafficking regulator NPC1 was significantly reduced in APA versus NFATs (fold change=0.66, P=0.008) (Figure 1A and 1B). Western blot analysis confirmed reduced NPC1 protein levels in APAs (Figure 1C and 1D), but without corresponding mRNA changes (Figure 1E). Immunohistochemistry further validated NPC1 protein downregulation in APA tumor tissues. Intriguingly, comparative analysis of tumor versus adjacent peritumoral tissues revealed that NPC1 was decreased in APA tumors but remained unchanged in NFAT tumors (Figure 1F; Figure S1A). Moreover, NPC1 mRNA levels were unchanged in APA tumors or NFAT tumors (Figure S1B). Given that NPC1 deficiency is involved in both cholesterol homeostasis and Ca²⁺ signaling regulation, the changed expression of NPC1 suggests its potential role in aldosterone production.

Figure 1. NPC1 is downregulated in APA.

A, Heatmap depicting key regulators of cholesterol trafficking in NFATs (n=10) and APAs (n=15) based on proteomic analysis. B, LFQ intensity of NPC1 in NFAT vs APA. C–D, NPC1 was downregulated in APA. Representative Western blot and quantification of NPC1 protein expression in APA samples (n=6). E, NPC1 mRNA levels were unchanged in APA. mRNA expression levels of NPC1 in APA samples were analyzed by qPCR (n=5). F, NPC1 was downregulated in APA. Immunohistochemistry staining of NPC1 in NFAT and APA tissues (n=6). Scale bars, 100 μm (4X), 50 μm (10X). Data are mean±SD. ABCA1 indicates ATP binding cassette subfamily A member 1; ABCG1, ATP binding cassette subfamily G member 1; APA, aldosterone‐producing aderoma; CYP11B2, cytochrome P450 family 11 subfamily B member 2; GRAMD1B, GRAM domain containing 1B; LDLR, low density lipoprotein receptor; LFQ, label‐free quantification; LIPA, lipase A, lysosomal acid type; LIPE, lipase E, hormone sensitive type; NFAT, nonfunctional adrenocortical aderoma; NPC1, NPC intracellular cholesterol transporter 1; NPC2, NPC intracellular cholesterol transporter 2; qPCR, quantitative polymerase chain reaction; SCARB1, scavenger receptor class B member 1; STAR, steroidogenic acute regulatory protein; STARD3, StAR related lipid transfer domain containing 3; and VDAC1, voltage dependent anion channel 1.

NPC1 Deficiency Promotes Aldosterone Secretion

To determine the functional role of NPC1 inhibition on aldosterone secretion—a hallmark of APA, we treated H295R cells with the NPC1‐specific inhibitor U18666A. Treatment with U18666A significantly increased aldosterone secretion by 70% (Figure 2A). Western blotting confirmed concomitant upregulation of CYP11B2 protein by 2‐fold (Figure 2B). Notably, quantitative PCR revealed a 15‐fold increase in CYP11B2 mRNA levels following NPC1 inhibition (Figure 2C). Similarly, NPC1 knockdown elevated aldosterone secretion by 41% and CYP11B2 mRNA by 2.3‐fold (Figure 2D‐F). Collectively, these data demonstrate that NPC1 suppression enhances aldosterone production.

Figure 2. NPC1 deficiency promote aldosterone secretion.

A–C, H295R cells were treated with or without the NPC1 inhibitor U18666A (5 μg/mL) for 24 hours. A, Aldosterone secretion was increased in NPC1‐inhibited cells. Aldosterone secretion in the cell culture supernatant was quantified by LC/MS (n=5). B, CYP11B2 protein levels were increased in NPC1‐inhibited cells. Aldosterone synthase (CYP11B2) protein expression was validated by Western blot analysis (n=5). C, CYP11B2 mRNA levels were increased in NPC1‐inhibited cells. mRNA expression levels of aldosterone synthase‐related genes (CYP11A1, HSD3B2, CYP21A2, CYP11B1, and CYP11B2) were analyzed by qPCR (n=6). D–F, H295R cells were transduced with sh‐NPC1 or sh‐Ctrl lentivirus to knock down NPC1 expression. D, Aldosterone secretion was increased in sh‐NPC1 cells. Aldosterone secretion in the cell culture supernatant was quantified by LC/MS (n=6). E, NPC1 protein levels were decreased in sh‐NPC1 cells. NPC1 protein expression was validated by Western blot analysis (n=6). F, CYP11B2 mRNA levels were increased in sh‐NPC1 cells. mRNA expression levels of aldosterone synthase‐related genes were analyzed by qPCR (n=6). Data are mean±SD. CYP11A1 indicates cytochrome P450 family 11 subfamily A member 1; CYP21A2, cytochrome P450 family 21 subfamily A member 2; CYP11B1, cytochrome P450 family 11 subfamily B member 1; CYP11B2, cytochrome P450 family 11 subfamily B member 2; HSD3B2, hydroxy‐delta‐5‐steroid dehydrogenase, 3 beta‐ and steroid delta‐isomerase 2; LC/MS, liquid chromatography/mass spectrometry; NPC1, NPC intracellular cholesterol transporter 1; qPCR, quantitative polymerase chain reaction; sh‐CTRL, control shRNA; and sh‐NPC1, shRNA targeting NPC1.

Inhibition of NPC1 Leads to Cholesterol Accumulation and Increased Lysosome‐Mitochondria Interaction

Downregulation of NPC1 leads to cholesterol accumulation in lysosomes and mitochondria. ³¹ , ³² , ³³ , ³⁴ , ³⁵ Consistent with this, immunofluorescence staining using TOM20 (mitochondria), LAMP1 (lysosomes), and filipin (cholesterol) showed that NPC1 inhibition significantly increased filipin colocalization with TOM20 and LAMP1, indicating elevated cholesterol levels in both mitochondria and lysosomes of H295R cells (Figure 3A–3D). In addition, NPC1 deficiency reduced lysosome–endoplasmic reticulum (ER) contacts, thereby decreasing ER cholesterol while promoting mitochondrial cholesterol accumulation. ²⁷ , ³⁶ Therefore, we hypothesized that elevated mitochondrial cholesterol originates from lysosomes. To test this, live‐cell imaging revealed a marked increase in lysosome‐mitochondria colocalization following NPC1 inhibition (Figure 3E and 3F). Interestingly, the proportion of prolonged lysosome‐mitochondria interactions (defined as colocalization lasting >10 seconds) increased approximately 2‐fold (Figure 3G; Video S1). To directly validate membrane contact, transmission electron microscopy was performed, demonstrating a significantly increased percentage of lysosome‐mitochondria contacts (1.5‐fold) and contact length (2.8‐fold) in stimulated cells (Figure 3H–3J). Collectively, these data indicate that NPC1 inhibition transports lysosomal cholesterol to mitochondria via enhanced lysosome‐mitochondria interactions.

Figure 3. NPC1 inhibition drives mitochondrial cholesterol accumulation and enhances lysosome‐mitochondria contact.

Cells were treated with or without the NPC1 inhibitor U18666A (5 μg/mL) for 24 hours. A–D, Mitochondrial and lysosomal cholesterol was increased in NPC1‐inhibited cells. Mitochondrial and lysosomal cholesterol distribution was assessed by fluorescence co‐staining of filipin (cholesterol) with TOM20 (mitochondrial marker) or LAMP1 (lysosomal marker). Colocalization was quantified using Pearson’s correlation coefficient (n=15–16) (B, D). E–G, Lysosome‐mitochondria colocalization was increased in NPC1‐inhibited cells. Lysosome‐mitochondria colocalization was evaluated by live‐cell imaging with MitoTracker (mitochondria) and LysoTracker (lysosomes) (E). Colocalization and proportion of prolonged lysosome‐mitochondria interactions (defined as colocalization lasting >10 seconds) were quantified in (E) (F, n=8; G, n=10). H–J, Lysosome‐mitochondria contacts were increased in NPC1‐inhibited cells. Lysosome‐mitochondria membrane contacts were analyzed by transmission electron microscopy. Percentage of lysosome‐mitochondria contacts (I, n=4), and contacts length were quantified (J, n=7). Scale bar:10 μm (A, C); 10 μm (E); 250 nm (H). Data are mean±SD. LAMP1 indicates lysosomal associated membrane protein 1; NPC1, NPC intracellular cholesterol transporter 1; TEM, transmission electron microscopy; and TOM20, translocase of outer mitochondrial membrane 20.

Inhibition of NPC1 Regulates Lysosome‐Mitochondria Interaction Via STARD3‐VDAC1

Next, we investigated how NPC1 inhibition regulates lysosome‐mitochondria interaction. Because STARD3 mediates cholesterol transport from lysosomes to mitochondria, ³⁵ , ³⁷ , ³⁸ we investigated its involvement in NPC1‐dependent aldosterone dysregulation. STARD3 knockdown decreased by 76% the elevated aldosterone secretion induced by NPC1 inhibition (Figure 4A, B). STARD3 is a homologous protein of StAR, a protein with a sterol‐binding domain that interacts with VDAC1 to regulate mitochondria cholesterol transfer. ³⁹ , ⁴⁰ We therefore hypothesized that STARD3 similarly interacts with VDAC1 to regulate lysosome‐mitochondria cholesterol transfer. To test this, coimmunoprecipitation in HEK293T cells expressing Flag‐STARD3 and HA‐VDAC1, and coimmunoprecipation of endogenous STARD3 and VDAC1 in H295R cells confirmed their interaction (Figure 4C and 4D). We next assessed the role of VDAC1 in aldosterone regulation. Treatment of H295R cells with the VDAC1‐specific inhibitor VBIT‐12 suppressed NPC1 inhibition induced aldosterone secretion by ∼50% and reduced CYP11B2 mRNA levels by 20‐fold (Figure 4E and 4F). Importantly, filipin staining and biochemical measurement of cholesterol revealed that Vbit‐12 attenuated mitochondrial cholesterol accumulation triggered by NPC1 downregulation (Figure 4G and 4H). Moreover, immunofluorescence demonstrated that VBIT‐12 treatment suppressed lysosome‐mitochondria colocalization, and transmission electron microscopy showed reductions of 58% in the percentage and 50% in the length of lysosome‐mitochondria contact compared with the U18666A group (Figure 4I through 4M). However, STARD3 and VDAC1 protein expression remained unchanged between APAs and NFATs or in NPC1 inhibition H295R cells (Figure S2A, B). Together, these results establish that NPC1 deficiency enhances lysosome‐mitochondria interaction via STARD3‐VDAC1.

Figure 4. Inhibition of NPC1 regulates lysosome‐mitochondria contacts via STARD3‐VDAC1.

A, STARD3 knockdown in H295R cells. Cells were transduced with sh‐STARD3 or sh‐Ctrl. Western blot confirming knockdown efficiency (n=6). B, STARD3 knockdown decreased aldosterone in NPC1‐inhibited cells. Aldosterone secretion in sh‐Ctrl/sh‐STARD3 H295R cells±U18666A (5 μg/mL, 24 hours) was quantified by LC/MS (n=6). C, STARD3‐VDAC1 interaction in HEK293T cells. Coimmunoprecipitation (anti‐Flag) of lysates from cells cotransfected with Flag‐STARD3 and HA‐VDAC1 (n=3). D, Endogenous STARD3‐VDAC1 interaction in H295R cells. Coimmunoprecipitation with anti‐STARD3 antibody or IgG control (n=3). E–M, H295R cells treated with VDAC1 inhibitor VBIT‐12 (20 μM)±U18666A (5 μg/mL, 24 h). E, VBIT‐12 treatment decreased aldosterone in NPC1‐inhibited cells. Aldosterone secretion was quantified by LC/MS (n=6). F, CYP11B2 mRNA expression quantified by qPCR (n=4). G, VBIT‐12 treatment decreased mitochondrial cholesterol in NPC1‐inhibited cells. Mitochondrial cholesterol was evaluated by filipin (cholesterol) and TOM20 (mitochondria) costaining. Inset: Quantification of colocalization (Pearson’s correlation coefficient, n=15). H, VBIT‐12 treatment decreased purified mitochondrial cholesterol in NPC1‐inhibited cells. Mitochondrial cholesterol content was measured in purified mitochondria (n=5). I, VBIT‐12 treatment decreased lysosome‐mitochondria colocalization in NPC1‐inhibited cells. Lysosome‐mitochondria colocalization was analyzed by live‐cell imaging with MitoTracker (mitochondria) and LysoTracker (lysosomes). J, Quantification of colocalization in (I) (Pearson’s correlation coefficient; n=15). K, VBIT‐12 treatment decreased lysosome‐mitochondria contacts in NPC1‐inhibited cells. Representative TEM images of lysosome‐mitochondria contacts. L, M, Quantification of percentage of lysosome‐mitochondria contacts (n=4), and contacts length in (K) (n=7). Scale bars: 10 μm (G, I); 250 nm (K). Data: mean±SD. CYP11B2 indicates cytochrome P450 family 11 subfamily B member 2; IB, immunoblotting; IP, immunoprecipitation; LC/MS, liquid chromatography/mass spectrometry; NPC1, NPC intracellular cholesterol transporter 1; qPCR, quantitative polymerase chain reaction; sh‐CTRL, control shRNA; sh‐STARD3, shRNA targeting STARD3; STARD3, StAR related lipid transfer domain containing 3; TEM, transmission electron microscopy; TOM20, translocase of outer mitochondrial membrane 20; and VDAC1, voltage dependent anion channel 1.

The Interaction Between STARD3 and VDAC1 Regulated Cholesterol Transport

Next, to identify which domain of STARD3 interacts with VDAC1, we generated truncation mutants of STARD3. Coimmunoprecipitation assays demonstrated that both the MENTAL (MLN64 amino‐terminal shared with MENTHO) and START (STAR‐related transfer) domains of STARD3 bind VDAC1 (Figure 5A and 5B). Subsequently, we tested whether the VDAC1‐STARD3 complex regulates cholesterol transport. Following cholesterol depletion and refeeding treatment, the interaction between VDAC1 and full‐length STARD3 intensified (by 70%) or with the START domain (by 32%), whereas that with the MENTAL domain showed no enhancement (Figure 5C through 5E; Figure S3A–C). Importantly, these effects were abolished by mutations disrupting cholesterol‐binding sites in STARD3 (M307R, N311D) and VDAC1 (E73Q) (Figure 5F and 5G; Figure S3D,E). Furthermore, immunofluorescence staining confirmed augmented STARD3‐VDAC1 colocalization after cholesterol refeeding and U18666A treatment (Figure 5H and 5I). Collectively, these data establish that the STARD3‐VDAC1 interaction mediates cholesterol transport.

Figure 5. STARD3‐VDAC1 interaction regulated by cholesterol.

A, Schematic of STARD3 protein domains and variants. B, STARD3‐VDAC1 interaction in HEK293T cells. Cells were cotransfected with HA‐tagged STARD3 constructs and Flag‐VDAC1. Co‐IP with anti‐Flag beads (n=3). C–F, Cholesterol depletion and refeeding treatment intensified interaction between VDAC1 and full‐length STARD3 or the START domain. HEK293T cells transfected with HA‐tagged STARD3 variants and Flag‐VDAC1. After sterol depletion (MCD)±cholesterol refeeding, Co‐IP was performed with anti‐Flag beads (n=3–5). G, Cells transfected with Flag‐STARD3 and HA‐tagged VDAC1 E73Q variant. Co‐IP was performed after sterol depletion (MCD)±cholesterol refeeding (n=3). H, Cholesterol depletion and refeeding treatment increased VDAC1‐STARD3 colocalization. Immunofluorescence staining of VDAC1 and STARD3 in H295R cells following MCD treatment±cholesterol refeeding. Inset: Quantification (Pearson’s correlation coefficient; n=7). I, VDAC1‐STARD3 colocalization was increased in sh‐NPC1 cells. Immunofluorescence of VDAC1 and STARD3 in sh‐Ctrl vs sh‐NPC1 cells. Inset: Quantification (Pearson’s correlation; n=14). Scale bars: 5 μm (H, I). Data: mean±SD. Co‐IP indicates coimmunoprecipitation; MCD, methyl‐β‐cyclodextrin; MENTAL, MLN64 amino‐terminal shared with MENTHO; NPC1, NPC intracellular cholesterol transporter 1; sh‐Ctrl, control shRNA; sh‐NPC1, shRNA targeting NPC1; STARD3 indicates StAR related lipid transfer domain containing 3; START, STAR‐related transfer; and VDAC1, voltage dependent anion channel 1.

NPC1 Regulates Endoplasmic Reticulum Ca²⁺ Release Via IP3R3

Ca²⁺ dysregulation occurs in NPC1‐deficient fibroblasts. ⁴¹ , ⁴² To determine whether similar effects occur in adrenal cells, we measured Ca²⁺ levels in H295R cells following U18666A treatment. Fluorescence imaging revealed increased cytosolic (Cal‐520, 53%) and mitochondrial Ca²⁺ (Rhod‐2 am, 20%) but decreased ER Ca²⁺ (Mag‐4 am, 34%) (Figure 6A and 6B). For direct organellar Ca²⁺ quantification, we transfected H295R cells with the ER‐targeted indicator ER‐GCaMP6‐150 and the mitochondrial indicator Mito‐RCamP1h. Fluorescence intensity showed that NPC1 inhibition reduced ER Ca²⁺ (22%) and elevated mitochondrial Ca²⁺ (25%) (Figure 6C and 6D). These data demonstrate Ca²⁺ dysregulation occurs in NPC1 inhibition H295R cells.

Figure 6. NPC1 regulates endoplasmic reticulum calcium release via IP3R3.

A, Cytosolic Ca²⁺ and mitochondrial Ca²⁺ were increased but ER Ca²⁺ was decreased in NPC1‐inhibited cells. H295R cells treated with NPC1 inhibitor U18666A (5 μg/mL, 24 h) and calcium was detected using Cal‐520 AM (cytosolic Ca²⁺), Rhod‐2 AM (mitochondrial Ca²⁺), and Mag‐4 AM (ER Ca²⁺). B, Quantification of fluorescence intensity from (A) (n=4). C, Mitochondrial Ca²⁺ was increased but ER Ca²⁺ was in NPC1‐inhibited cells. Live‐cell calcium imaging in U18666A‐treated H295R cells expressing ER‐GCaMP6‐150 (ER Ca²⁺ reporter) and mito‐RCamPh1 (mitochondrial Ca²⁺ reporter). D, Quantification of reporter fluorescence intensity from (C) (n=10). E, F, IP3R3 protein levels were increased in NPC1‐inhibited cells. IP3R family protein expression in U18666A‐treated cells or NPC1 knockdown cells. Representative Western blot (n=3 or n=6). G, IP3R3 protein levels were increased in sh‐NPC1 cells. IP3R3 subcellular localization in U18666A‐treated cells. Immunofluorescence staining (n=10). H, IP3R3 mRNA levels were increased in sh‐NPC1 cells. IP3R family mRNA expression in sh‐NPC1 H295R cells (qPCR, n=9). Scale bars: 50 μm (A); 5 μm (C, G). Data: mean±SD. ER indicates endoplasmic reticulum; IP3R3, inositol 1,4,5‐trisphosphate receptor type 3; NPC1, NPC intracellular cholesterol transporter 1; qPCR, quantitative polymerase chain reaction; sh‐Ctrl, control shRNA; and sh‐NPC1, shRNA targeting NPC1.

Given that ER Ca²⁺ depletion and cytoplasmic Ca²⁺ elevation, combined with the established role of IP3 receptors (IP3Rs) in mediating ER calcium release. ⁴³ , ⁴⁴ We hypothesized that NPC1 inhibition disrupts Ca²⁺ homeostasis via IP3R. Western blotting showed a 2.5‐fold increase in IP3R3 protein levels, with no significant changes in IP3R1 or IP3R2, in U18666A‐treated cells compared with controls (Figure 6E). Consistent with inhibitor results, NPC1 knockdown increased IP3R3 protein level by 2.8‐fold (Figure 6F). Furthermore, immunofluorescence staining confirmed elevated IP3R3 fluorescence intensity after NPC1 inhibition (Figure 6G). Notably, whereas IP3R1 mRNA levels remained unchanged, IP3R2 and IP3R3 mRNA levels increased by 48% and 78% (Figure 6H). Collectively, these data demonstrate that NPC1 deficiency selectively upregulates IP3R3 expression in H295R cells.

To establish the role of IP3R3 in mediating Ca²⁺‐dependent aldosterone secretion after NPC1 inhibition, we treated cells with the IP3R inhibitor 2‐APB. 2‐APB reduced the NPC1 inhibition induced cytoplasmic Ca²⁺ increase (Figure S4A). Furthermore, analysis using ER‐ and mitochondria‐targeted Ca²⁺ indicators revealed that 2‐APB reversed both the ER Ca²⁺ decrease and mitochondrial Ca²⁺ increase induced by NPC1 inhibition (Figure S4B–D). We then tested whether 2‐APB suppressed aldosterone elevation in NPC1 inhibition cells. H295R cells were treated with U18666A, or U18666A and 2‐APB, followed by aldosterone measurement. Notably, 2‐APB suppressed NPC1 inhibition induced aldosterone secretion by 66% (Figure S4E). Consistently, CYP11B2 protein (48%) and mRNA (89%) upregulation were suppressed by 2‐APB treatment (Figure S4F–H). Taken together, these results demonstrate that NPC1 inhibition drives aldosterone secretion via IP3R3‐dependent Ca²⁺ release.

We propose a mechanistic model wherein NPC1 inhibition promotes aldosterone secretion (Figure 7). In H295R cells, NPC1 inhibition increases aldosterone secretion through 2 pathways: (1) enhanced lysosome‐mitochondria interactions and mitochondrial cholesterol accumulation via the STARD3‐VDAC1; and (2) upregulated CYP11B2 expression induced by IP3R3‐mediated Ca²⁺ release from the endoplasmic reticulum.

Figure 7. Schematic model of NPC1 regulate aldosterone secretion.

Under NPC1 deficiency, the STARD3‐VDAC1 complex facilitates cholesterol transport from lysosomes to mitochondria, enabling aldosterone synthesis. Concurrently, NPC1 inhibition upregulates IP3R3, triggering endoplasmic reticulum calcium release. This calcium signaling promotes CYP11B2 transcription, ultimately enhancing aldosterone production. CYP11B2 indicates cytochrome P450 family 11 subfamily B member 2; IP3R3, inositol 1,4,5‐trisphosphate receptor type 3; NPC1, NPC intracellular cholesterol transporter 1; STARD3, StAR related lipid transfer domain containing 3; and VDAC1, voltage dependent anion channel 1.

DISCUSSION

Here we identify that NPC1 deficiency is a key component of elevated aldosterone production in adrenocortical cells, operating both by enhancing mitochondrial cholesterol overload and upregulating CYP11B2 expression. Furthermore, our molecular and functional characterization elucidates 2 novel mechanisms: enhanced lysosome‐mitochondria interactions mediated by STARD3‐VDAC1, and activation of cytoplasmic calcium signaling via IP3R3‐mediated endoplasmic reticulum calcium release in NPC1‐deficient H295R cells (Figure 7). These in vitro mechanisms demonstrate that abnormal cholesterol transfer and Ca²⁺ signaling activation are accompanied by substantial increases in aldosterone synthesis, which may provide new mechanistic insights into excessive aldosterone production in APAs.

More than 90% of APA carry somatic mutations in ion transport regulators, including K⁺ channel Kir3.4 (KCNJ5), Ca²⁺ channel CaV1.3 (CACNA1D), plasma membrane Ca²⁺ transporting ATPase 3 (ATP2B3). ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ Most mutations dysregulate ion channels, causing calcium dyshomeostasis and aldosterone hypersecretion. Thus, research on primary aldosteronism pathogenesis and physiological aldosterone production has focused predominantly on calcium signaling over the past decade. However, cholesterol’s role in APAs or physiological aldosterone production remains poorly defined, despite being the essential precursor for aldosterone synthesis. We previously reported that dysregulated cholesterol metabolism occurred in APAs. ²⁹ Here, we found reduced NPC1 expression in APA tissues. Moreover, we found that NPC1 inhibition elevates aldosterone secretion and upregulates CYP11B2 protein and mRNA levels in H295R cells. Our results provide evidence for how NPC1 regulates aldosterone production via cholesterol, findings with important implications for metabolic and APA research.

NPC1 deficiency leads to increased mitochondrial cholesterol that is dependent on STARD3. ²⁷ Similarly, we found that knocking down STARD3 rescues the increases in mitochondrial cholesterol and aldosterone levels induced by NPC1 deficiency. However, how STARD3 regulates lysosome‐mitochondria interactions and cholesterol transport remains unclear. Recent evidence indicates that StAR (a STARD3 homolog) functions within a protein complex that regulates mitochondrial cholesterol delivery. ¹² This complex comprises outer mitochondrial membrane proteins including VDAC, TSPO (translocator protein), ACBD3 (acyl‐coenzyme A binding domain containing 3), PKA‐RIα (protein kinase A regulatory subunit 1α), and inner mitochondrial membrane protein ATAD3A (ATPase family AAA domain‐containing protein 3A). ⁴⁹ Upon hormonal stimulation, complex assembly promotes cholesterol pooling at the outer mitochondrial membrane, where VDAC anchors StAR and TSPO to initiate cholesterol transport into the inner mitochondrial membrane. ⁴⁰ , ⁴⁹ We found that STARD3 interacts with VDAC1 to regulate lysosome‐mitochondria interactions and mediate cholesterol transfer, regarding other components of this complex, whether STARD3 interacts with them and the functional significance of such interactions requires further study.

Moreover, we found that both the MENTAL and START domains of STARD3 bind VDAC1, whereas full‐length STARD3 and the START domain interaction intensified following cholesterol depletion refeeding treatment. This aligns with established evidence that the STARD3 START domain transfers cholesterol to mitochondria and potentiates steroidogenesis. ⁵⁰ The N‐terminal MENTAL domain of STARD3 exhibits high homology to STARD3NL (formerly MENTHO), a late endosomal transmembrane protein, and overexpression of either protein extends endosome–ER contacts, indicating their role in establishing these membrane junctions. ⁵¹ Thus, we assume that when STARD3 interacts with VDAC1, its MENTAL domain may anchor lysosomes to mitochondria, whereas its START domain likely transfers lysosomal cholesterol to mitochondria.

NPC1 deficiency had previously been shown to elevate cytosolic Ca²⁺ in neurons. ⁵² Similarly, we demonstrate that in H295R adrenal cells, NPC1 inhibition increases both cytosolic and mitochondrial Ca²⁺ while depleting ER Ca²⁺. In neurons, this dyshomeostasis is mediated by IP3R1. ⁵³ However, in H295R cells, we found that NPC1 deficiency selectively upregulates IP3R3 (but not IP3R1) at both protein and mRNA levels. This tissue‐specific divergence may reflect differential IP3R isoform expression and function across organs. A limitation of this study is that our findings are based on in vitro experiments; further validation in vivo is needed, for example, using adrenal‐specific NPC1‐knockout mice to confirm its regulatory role in aldosterone production. Additionally, the mechanisms underlying reduced NPC1 expression in APAs remain unclear and require further investigation.

Conclusions

Our findings demonstrate that NPC1 downregulation represents a novel mechanism driving elevated aldosterone production, linking lysosomal‐mitochondrial cholesterol transport to excessive aldosterone synthesis. These results suggest that NPC1 may offer new insights into aldosterone overproduction of APA pathogenesis that require further investigation.

Sources of Funding

This work was supported by the national key research and development plan of China, major project of prevention and treatment for common diseases (2022YFC2505300, subproject: 2022YFC2505301, 2022YFC2505302, 2022YFC2505306). Other sourcs of funding include the National Natural Science Foundation of China (U21A20355, 82 170 825, 82 270 878), The Chongqing Technology Innovation and Application Development Special Key Project (CSTB2024TIAD‐KPX0039), The Chongqing Science and Health Joint Medical Research Project (2025GGXM004), Chongqing Outstanding Youth Science Fund project (CSTB2023NSCQ‐JQX0028), Medical research project of the Chongqing Health Commission (2024WSJK109), Program for Youth Innovation in Future Medicine, Chongqing Medical University, Science and technology research project of Chongqing Education Commission (KJQN202200403), and Joint Medical Research Project of Chongqing Science and Technology Commission & Chongqing Health and Family Planning Commission (Major Project, 2022ZDXM003).

Disclosures

None.

Supporting information

Data S1: Figures S1‐S4, Videos S1‐S2, Unedited Gels.

Video S1

Video S2

Data S2