Cholesterol Biosynthesis and Trafficking in Cortisol-Producing Lesions of the Adrenal Cortex

Abstract

Context:

Cortisol-producing adenomas (CPAs), primary pigmented nodular adrenocortical disease (PPNAD), and primary macronodular adrenocortical hyperplasia (PMAH) cause ACTH-independent Cushing syndrome (CS). Investigation of their pathogenesis has demonstrated their integral link to the cAMP-dependent protein kinase signaling pathway.

Objective:

The aim of this study was to identify differences in cholesterol biosynthesis among different CS-causing adrenocortical tumors. Because of the concomitant associations of cAMP levels with cholesterol and with steroid biosynthesis, we hypothesized that benign cortisol-producing tumors would display aberration of these pathways.

Design and Setting:

Twenty-three patients with CPA, PPNAD, or PMAH who underwent adrenalectomy for CS were included in the study. Preoperative biochemical analyses were performed, and excised adrenal tissues were studied.

Main Outcome Measures:

Serum, urinary hormone levels, serum lipid profiles, and anthropometric data were obtained preoperatively. Adrenal tissues were analyzed for total protein, cholesterol, and neutral sterol content by mass spectrometry and expression of HMGCR, LDLR, ABCA1, DHCR24, and STAR genes.

Results:

There were differences in cholesterol content and markers of cholesterol biosynthesis and metabolism that distinguished CPAs from PMAH and PPNAD; cholesterol, lathosterol, and lathosterol/cholesterol ratio were significantly higher in CPAs. ABCA1 mRNA was lower among CPAs compared to tissues from bilateral adrenocortical hyperplasia (PMAH and PPNAD), and mRNA expression of LDL-R, DCHR24, and HMGCR tended to be higher in CPA tumor tissues.

Conclusion:

CPAs displayed characteristics of “cholesterol-starved” tissues when compared to PPNAD and PMAH and appeared to have increased intrinsic cholesterol production and uptake from the periphery, as well as decreased cholesterol efflux. This has implications for a potential new way of treating these tumors.

ACTH-independent Cushing syndrome (CS) is most commonly caused by cortisol-producing adenoma (CPA) (1). Bilateral adrenocortical hyperplasias (BAHs), including primary pigmented nodular adrenocortical hyperplasia disease (PPNAD) and primary macronodular adrenocortical hyperplasia (PMAH), are rare causes of ACTH-independent CS. PPNAD usually presents in childhood or early adulthood as multiple small (<1 cm diameter), pigmented, cortisol-producing adrenocortical nodules, usually surrounded by cortical atrophy (2, 3). Conversely, PMAH most frequently occurs among older patients (4–6). PMAH patients have grossly enlarged adrenal glands that bear multiple nodules (>1 cm diameter), usually in the absence of internodular atrophy (7).

The pathogenesis of these various lesions has been further delineated in recent years (8). PPNAD has been associated with Carney complex in almost 90% of cases; Carney complex is an autosomal dominant syndrome associated with multiple endocrine tumors and other lesions (9). About 60% of patients with PPNAD harbor an inactivating mutation of the PRKAR1A gene, which codes for the regulatory subunit 1α of cAMP-dependent protein kinase A (PKA), resulting in increased cAMP signaling (10, 11). In patients with PPNAD but without PRKAR1A mutations, other genes involved in cAMP signaling, including phosphodiesterase (PDE) 11A (PDE11A) and PDE8B, have been implicated (12, 13). We recently described PRKACA defects in both PPNAD and CPA (14, 15). PRKACA is the main catalytic subunit of PKA and is tightly regulated by PRKAR1A (16). In addition, we and others described ARMC5 mutations in PMAH (17, 18); the function of ARMC5, however, is largely unknown, and its role in increasing cAMP signaling remains to be elucidated. Although each of the various tumor types has distinct morphological and molecular features, a common thread among all identified defects is the resulting impact on intracellular cAMP signaling. Furthermore, the nodules in PMAH show altered expression of the PKA subunits and increased PKA signaling (19).

Ultimately, all of the aforementioned mechanisms lead to aberrant production of glucocorticoids and the presentation of CS. Additionally, gene expression of CYP11A, CYP17, HSD3B2, and CYP11B1, enzymes needed for corticosteroid production, was increased in cortisol-producing lesions compared to normal adrenals (20–23). Cortisol secretion was positively correlated with the expression of genes involved in cholesterol metabolism, glutathione S-transferase, and steroidogenic enzymes in a gene-profiling study of adrenocortical adenomas (24). Increased cAMP levels affect many of the steps involved in cholesterol and steroid biosynthesis (25). There is evidence that cAMP plays a role in regulating gene expression of steroidogenesis acute regulatory (StAR) protein and 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR), the rate-limiting steps of steroidogenesis and cholesterol biosynthesis, respectively (25, 26). Additionally, PKA has been implicated in the complex protein-protein interactions that regulate cholesterol transport in steroidogenic cells via StAR and then into the mitochondria via translocator protein (26, 27).

We hypothesized that cells originating from different cortisol-producing tumors would behave differently with respect to cholesterol trafficking and synthesis. Thus, we investigated the cholesterol content and expression of several genes critical in regulating cholesterol availability in various CS-causing adrenal tumors and the relationship between cholesterol content of tumors and serum hormone and lipid profiles.

Patients and Methods

Patients and clinical studies

All clinical studies were approved by the Eunice Kennedy Shriver National Institute of Child Health and Human Development Institutional Review Board. All included patients signed the appropriate informed consent documents upon enrollment in the study protocols. Twenty-three patients initially referred to the National Institutes of Health (NIH) and subsequently enrolled in NIH protocols (95CH059, 97CH0076, and 00CH160) for management of CS caused by various adrenocortical tumors were included in the study. Postoperative histological examinations determined the cause of CS to be CPAs (n = 9), PPNAD (n = 8), and PMAH (n = 6). Twenty of the patients were available for all or part of the preoperative serum and urinary studies (CPA, n = 8; PPNAD, n = 7; and PMAH, n = 6), and tumor tissues were obtained for 23 of the patients. Preoperative serum lipid data were analyzed for 16 additional CPA (n = 5), PPNAD (n = 5), and PMAH (n = 6) patients enrolled in the aforementioned NIH protocols to obtain enough data to perform statistical analysis.

Anthropometric data, along with baseline values of urinary 17-hydroxysteroids, urinary free cortisol (UFC), diurnal (8 am and midnight) cortisol levels, and creatinine, were obtained preoperatively (Table 1). Dexamethasone testing included the Liddle's test or, occasionally, an overnight 8-mg dexamethasone test. Most patients had a CRH stimulation test with measurements of both cortisol and ACTH levels. Preoperative lipid profiles, including total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and triglycerides were also obtained.

Table 1.

Demographics and Baseline Biochemical Values for the 23 Patients Enrolled in the Study

Sex	Age, y	Diagnosis	BMI, kg/m2 (BMI z-score)	Diurnal Cortisol		Dexamethasone Test		Mean UFC, μg/dL	Mean 17-OH, μg/dL	Chol, μg/dL	HDL, mg/dL	LDL, mg/dL	TG, mg/dL
				MN	8 am	Pre-Dex, μg/dL	Post-Dex, μg/dL
F	49	PMAH	31.3	15.1	17.7	20.4	15.4	237.6	12.1	152	NA	NA	NA
F	46	PMAH	33.5	24.4	22.8	NA	NA	140.7	13.1	215	NA	NA	87
M	55	PMAH	38.5	6.2	12.5	34.4	18.5	384.7	12.4	194	NA	NA	190
M	40	PMAH	26	51.3	63.9	67.8	26.9	1808.4	97.9	NA	NA	NA	NA
F	19	PPNAD	38.4	11.1	10.5	12.0	12.1	117.8	9.9	201	61	118	95
F	31	PPNAD	25.8	6.0	9.4	NA	NA	9.5	2.1	161	43	105	81
F	51	PPNAD	27.0	11.8	10.4	NA	NA	84.3	26.3	160	NA	NA	NA
F	36	PMAH	30.1	5.8	14.6	NA	NA	558.9	15.6	207	62	148	175
F	31	PMAH	38.5	24.1	21.9	NA	NA	506.4	17.1	351	NA	NA	NA
M	8	PPNAD	25.4 (2.31)	35.6	37.6	33.6	36.7	389.7	14.4	208	62	139	175
F	49	CPA	23.5	13.3	16.6	9.4	9.8	634.2	16.7	164	92	84	49
F	16	CPA	20.5 (0.05)	19.6	17.9	21.2	23.7	116.2	6.1	NA	NA	NA	NA
F	15	CPA	19.7 (−0.13)	47.3	33.6	NA	NA	280.2	9.4	169	NA	NA	NA
F	38	PPNAD	25.1	10.4	12.8	NA	NA	37.8	6.7	134	NA	NA	NA
F	24	PPNAD	37.6	26.3	30.8	33.4	36.2	551.9	9.3	NA	NA	NA	NA
F	37	PPNAD	36.2	19.8	15.2	44.6	30.8	8488.3	76.1	330	35	NA	NA
F	8	PPNAD	24.1 (2.06)	22.3	19.0	16.6	20.0	42.1	4.0	197	NA	NA	NA
F	27	CPA	36.9	14.4	5.0	8.5	3.2	7113.3	29.2	213.0	44	162	545
F	50	CPA	40.6	14.7	13.0	13.6	14.0	230.6	11.6	233.0	NA	NA	NA
F	30	CPA	32.2	11.6	13.6	24.6	25.0	654.7	2.5	220.0	NA	NA	NA
F	2	CPA	N/A	NA	NA	NA	NA	23.2	7.3	NA	NA	NA	NA
F	44	CPA	24.4	15.8	15.7	NA	NA	178.3	7.2	169	53	103	275
F	14	CPA	42.6 (2.58)	37.9	23	NA	NA	408.3	22.1	174	38	118	93

Abbreviations: F, female; M, male; MN, midnight; NA, not available; Pre-Dex, pre-dexamethasone; Post-Dex, post-dexamethasone; 17-OH, 17-hydroxycortisol; Chol, cholesterol; TG, triglyceride.

All samples have been previously sequenced for mutations in known genes, including PRKAR1A, PRKACA, PDE11A, PDE8B, and MEN1. Tumor tissues from CPAs were also screened for PRKACA by microdissecting from the excised adrenal; all sequencing was done as we have described elsewhere (12, 13).

Tissue and cholesterol studies

Twenty-one tumor tissues obtained at surgery were analyzed for cholesterol and lathosterol content by gas chromatography/mass spectrometry and corrected for both total protein and tissue mass.

Sample preparation was conducted as previously described by Kelley (28) with slight modifications. Briefly, the mouse tissue samples were weighed, then homogenized in 2 mL of 1X PBS; cell pellets were briefly homogenized in 2 mL of 1X PBS and subjected to the same conditions as the tissue samples. Twenty micrograms of coprostan-3-ol (Sigma) were added to each sample as a surrogate internal standard. The samples were saponified in 4% potassium hydroxide in 70% ethanol for 1 hour at 60°C. Samples were then extracted with an equal volume of ethyl acetate and centrifuged at 2000 × g for 5 minutes. The organic phase was removed and dried under a stream of nitrogen. The samples were derivatized with bis-trimethylsilyl trifluoroacetamide plus 1% trimethylchlorosilane (Pierce) for 1 hour at 60°C. Derivate samples were injected onto a gas chromatogram/mass spectrometer (Finnigan Trace DSQ, Thermo Electron GC/MS) utilizing a ZB-1701 30 m × 0.32 mm × 0.25 μm column (Phenomenex). On injection, the oven temperature was 170°C, ramped at 21°C per minute to 250°C, and then ramped at 3°C per minute to a final temperature of 290°C. Total amounts of cholesterol and 7- dehydrocholesterol (7DHC), 8DHC, desmosterol, and 7-dehydrodesmosterol were determined based on comparison to the surrogate internal standard coprostan-3-ol. Retention times were confirmed using standards (Sigma) of the available compounds. 8DHC and 7-dehydrodesmosterol were determined based on the total ion chromatogram in comparison to 7DHC and/or the National Institutes of Standards and Technology mass spectral library, data version 14.

Cholesterol biosynthesis molecular studies

RNA was isolated from adrenal samples using Trizol (Invitrogen) and was used to generate cDNA using an ABI High Capacity cDNA Archive Kit (Applied Biosystems). Quantitative real-time PCR was performed on an ABI 7000, using ABI probe and primer sets for the genes HMGCR (Hs00168352_m1), LDLR (Hs00181192_m1), ABCA1 (Hs00194045_m1), DHCR24 (Hs00207388_m1), and StAR (Hs00264912_m1). Reactions were run as a multiplex using human GADPH as an internal control. The ΔΔ cycle threshold method was determined using the CPA group mean as the comparison group to evaluate fold-change in expression (2^−ddCT).

Immunohistochemistry

Representative formalin-fixed paraffin-embedded tissues derived from the various CS-causing adrenocortical tumors were subjected to immunofluorescent staining using a commercially available anti-LDLR antibody (sc-18823; Santa Cruz) at a dilution of 1:500 and counterstained with DAPI nuclear stain. The secondary antibody contained Alexa Fluor 488 and enabled fluorescence microscopy to assess relative expression of LDLR.

Statistical analyses

Biochemical and molecular data were analyzed using ANOVA and Student's t test after first confirming normal distribution of data. Correlation analysis was performed for all data and for individual diagnostic groups. All statistical tests were performed using SAS version 9.3 (SAS Institute Inc). Data are all presented as the mean ± SEM. Data were initially analyzed keeping each of the three CS subgroups separate. When a pattern was identified among the BAHs (PMAH and PPNAD) that shared similar characteristics with respect to the data examined, we opted to group the BAHs as a single group, which enhanced the statistical power.

Results

Serum and urinary studies

UFC levels (corrected for creatinine) and serum diurnal (8 am and midnight) cortisol values did not differ significantly between the various CS diagnostic groups. Differences in cortisol levels after a high-dose dexamethasone test showed a tendency toward higher cortisol levels in PPNAD compared to CPA and PMAH patients (P = .07; Table 2), as we have described previously (29). There were no differences in the increment of cortisol during the CRH test (data not shown). Mean serum triglyceride levels tended to be higher in CPA patients compared to PMAH or PPNAD groups. When compared to the values of the BAH groups combined (PMAH and PPNAD), triglyceride levels in CPA patients were significantly higher (P = .04; Table 2). Total serum cholesterol, HDL, and LDL values did not differ among the diagnostic groups (Table 2). There were positive, although not statistically significant, correlations between tumor cholesterol content and 8 am cortisol (P = .18), and pre-dexamethasone (P = .09), and post-dexamethasone (P = .11) cortisol levels. Serum total cholesterol was positively correlated with average UFC and average 17-hydroxycorticosteroid values (P ≤ .05).

Table 2.

Preoperative Serum and Urinary Biochemical Measures of Patients Included in the Studies

	Diagnostic Group
	CPA	PMAH	PPNAD
Serum lipids
HDL, mg/dL	50.0 ± 5.0 (n = 11)	53.8 ± 3.1 (n = 7)	56.4 ± 5.3 (n = 8)
LDL, mg/dL	126.5 ± 7.5 (n = 11)	113.0 ± 13.1 (n = 7)	109.5 ± 12.8 (n = 8)
Total cholesterol, mg/dL	204.0 ± 7.8 (n = 13)	200.1 ± 19.6 (n = 10)	191.9 ± 16.9 (n = 12)
Triglycerides, mg/dL	192.8 ± 41.9 (n = 11)	107.9 ± 15.6 (n = 7)	114.6 ± 16.1 (n = 8)
Cortisol and metabolites
MN cortisol, μg/dL	21.8 ± 4.7 (n = 6)	21.2 ± 6.9 (n = 6)	17.9 ± 3.5 (n = 8)
8 am cortisol, μg/dL	17.3 ± 2.9 (n = 6)	25.6 ± 7.8 (n = 6)	18.2 ± 3.7 (n = 8)
Mean UFC, μg/dL	248.7 ± 119.1 (n = 6)	729.6 ± 362.9 (n = 6)	1931.8 ± 1684.7 (n = 8)
Mean 17-OH, μg/dL	11.7 ± 2.9 (n = 5)	25.8 ± 11.9 (n = 6)	14.2 ± 6.0 (n = 8)
Pre-Dex, μg/dL	3.2 ± 4.1 (n = 5)	14.1 ± 3.4 (n = 3)	6.0 ± 4.8 (n = 3)
Post-Dex, μg/dL	15.1 ± 4.1 (n = 5)	20.3 ± 3.4 (n = 3)	27.3 ± 4.8 (n = 5)

Abbreviations: MN, midnight; 17-OH, 17-hydroxycortisol; Pre-Dex, pre-dexamethasone; Post-Dex, post-dexamethasone. Values are expressed as mean ± SEM.

Tissue cholesterol analyses

Twenty-one tumor tissue samples were analyzed for their cholesterol content. A representative gas chromatogram is shown in Figure 1A. The following three parameters were assessed: cholesterol (μg/mg protein), lathosterol (μg/mg protein), and lathosterol/cholesterol ratio, an index of cholesterol biosynthesis. The initial analysis distinguished the CPA group from the BAHs with respect to measures of cholesterol content. Cholesterol content of CPAs was higher than that of BAH tissues (Figure 1B), as were the mean lathosterol concentration (Figure 1C) and the ratio of lathosterol to cholesterol (Figure 1D).

Figure 1.

Sterols were measured in adrenal tissue samples by gas chromatography/mass spectrometry. A, Representative total ion chromatogram (50 mg CPA tissue); peak 1 is coprostanol (internal surrogate standard), peak 2 is cholesterol, and peak 3 is lathosterol. Adrenal tumor tissue concentrations of cholesterol (B) and lathosterol (C), and the ratio of lathosterol:cholesterol (D) in patients with CPAs compared to those with BAHs (PMAH and PPNAD). Values are expressed as mean ± SEM. *, P < .05.

Genetic and molecular studies

mRNA expression of the ABCA1 was approximately 3-fold higher in BAH adrenal tissues compared to that in CPA tumors (Figure 2A; P = .005). Although there were tendencies toward increased expression of DHCR24 (Figure 2B), HMGCR (Figure 2C), and LDLR (Figure 2D) in CPA tumors compared to the levels of BAH samples, the differences did not achieve statistical significance; P values were 0.16, 0.12, and 0.11, respectively. mRNA expression of StAR did not differ among groups. Additionally, the presence or absence of a mutation of the PRKAR1A gene did not appear to impact expression of the genes examined, or tumor cholesterol content for that matter. There were no PRKACA mutations in the studied CPAs.

Figure 2.

Gene expression of cholesterol biosynthetic and transport enzymes in adrenocortical tumor tissues from CPA and BAH (PMAH and PPNAD) patients as determined by quantitative real-time PCR: ABCA1 (A), DHCR24 (B), HMGCR (C), LDLR (D), and StAR (E). Values are expressed as mean ± SEM. *, P < .05.

Immunohistochemical analysis

Figure 3 shows immunofluorescent staining for LDLR protein in adrenal tissues from patients with PMAH and PPNAD compared to tumor tissue from a patient with CPA. As indicated by the intensity and the density of the green fluorescence, the CPA appears to have substantially more LDL receptor at the protein level.

Figure 3.

Adrenal tissues from patients with PMAH and PPNAD (BAHs) compared to that from a CPA patient. Slides were prepared from formalin-fixed paraffin embedded tissues and subjected to immunofluorescent staining with anti-LDLR antibody (Alexa Fluor 488) and counterstained with DAPI. Magnification, 10×, top panels; 40×, lower panels.

Discussion

It has long been described that, under normal circumstances, the cholesterol needs for steroidogenesis in adrenal cells are met mainly via uptake of circulating cholesterol (30–32). Although adrenal cells can synthesize cholesterol de novo, only 20% of the total cholesterol required by adrenal cells is derived from biosynthesis within the adrenals (30). Human adrenal cells primarily rely on uptake of circulating cholesterol containing lipoproteins, which are taken up through either a nonselective pathway, via the LDL receptor, or by “selective” cholesterol uptake through a pathway that uses scavenger receptor class B type I to bind both LDL and HDL (31). However, the origin of cholesterol in pathological circumstances and the recruitment of diverse pathways by adrenal cells in different pathological entities are not well known. Several stimuli, such as ACTH, have been documented to stimulate the activity of 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA) reductase for de novo cholesterol synthesis and LDL receptors for uptake of LDL cholesterol (30, 32). Thus, in ACTH-independent diseases that lead to high steroid hormone levels such as CPA, PPNAD, and PMAH, it is unclear whether and to what degree adrenal cells depend on steroidogenesis for cholesterol uptake mechanisms from the periphery or recruitment of the intrinsic pathways for de novo synthesis of cholesterol. Gaining a clear understanding of how steroidogenesis may depend on cholesterol trafficking in adrenal cells could potentially be exploited pharmacologically.

It should be noted that the current study did not address the contribution of HDL cholesterol, an area that warrants further investigation. Haploinsufficiency of LDLR in patients with abetalipoproteinemia was not sufficient to impair adrenocortical response to ACTH, suggesting that HDL is important for adrenal steroidogenesis (33). Along the same line, subjects with low HDL with or without mutations in LCAT or ABCA1 attenuated basal but not ACTH stimulated corticosteroid metabolism (34). Additionally, mice lacking LCAT had impaired adrenal glucocorticoid output, despite the compensatory up-regulation of genes involved in cholesterol synthesis, including HMG-CoA reductase (35). Evidence from these and other studies supports the dynamic nature of adrenal cholesterol metabolism and implicates the need for both endogenous cholesterol production and effective substrate uptake by adrenal cells to enable cholesterol supplies to come from various sources.

We examined the lathosterol/cholesterol ratio as a surrogate index of cholesterol biosynthesis. Lathosterol is one of the precursors in the synthesis of cholesterol, and the ratio of lathosterol to cholesterol has been widely used as an indirect measure of cholesterol synthesis (36). The lathosterol/cholesterol ratio was nearly 10-fold higher in CPAs compared to that in tissues from BAHs (Figure 1D). Additionally, cholesterol and lathosterol levels of CPAs were significantly higher than those of BAH adrenal tissues (Figure 1, B and C). This observation emphasizes the enhanced utilization of the intrinsic pathway of cholesterol synthesis by CPA cells for steroid production. The dramatic differences in these direct and indirect measures of cholesterol content and biosynthesis, respectively, demonstrate differential patterns of cholesterol biosynthesis among CS subgroups, specifically in comparing CPA to BAH.

Unique patterns in cholesterol trafficking among subtypes of CS were also suggested by the observation of decreased ABCA1 expression, the gene codes for ATP-binding cassette transporter. ABCA1 is a major player in cholesterol efflux and cellular cholesterol homeostasis that responds to elevated intracellular cholesterol levels by exporting cholesterol and thus maintaining appropriate intracellular lipid levels (37). The elevated ABCA1 expression as well as the elevated cholesterol levels in CPA tumor tissue support the hypothesis that CPAs have increased needs for cholesterol and, in fact, retain more cholesterol than do cells from BAH tissues. The positive correlations we observed between tumor tissue cholesterol content and several measures of circulating glucocorticoid levels support the retention of cholesterol in adrenocortical cells as a source of cholesterol for steroidogenesis.

HMG-CoA reductase catalyzes the rate-limiting step in cholesterol synthesis, mediating the formation of mevalonate from acetyl coenzyme A (26). We found a tendency toward increased HMGCR expression in CPAs when compared to BAHs (PMAH and PPNAD). In conjunction with our other findings, it seems likely that these results may nonetheless be physiologically relevant because CPAs seemed to display increased intrinsic production of cholesterol. Expression of the LDLR gene that encodes for the LDL receptor, which imports cholesterol from the periphery, also tended to be higher in CPA tumor tissues compared to BAHs (Figure 2D). More strikingly, immunofluorescent staining for LDL receptor was clearly more intense in CPA than both PMAH and PPNAD adrenal tissues (Figure 3), thereby providing more evidence that increased LDL receptor action may aid in providing cholesterol to CPA cells via an elevated rate of cholesterol import from the periphery.

Interestingly, all three types of adrenocortical lesions we studied showed similar levels of StAR expression. StAR protein mediates the rate-limiting step for steroidogenesis by facilitating the transfer of cholesterol into the mitochondria for further processing (38). Because patients from the different diagnostic groups had similar serum and urinary hormonal levels, the finding of similar StAR expression across CPAs, PPNAD, and PMAH was not surprising and suggests that cholesterol flux does not change StAR activity when the latter is already up-regulated in these hypercortisolemic states.

In summary, this is the first study to investigate differences in cholesterol origin or absolute cholesterol content in CPA, PMAH, and PPNAD. Our investigation showed that sporadic CPAs behave distinctly regarding cholesterol trafficking and storage when compared to BAHs; CPAs act as “cholesterol-starved” tissues, as demonstrated by increased cholesterol biosynthesis and likely by enhanced cholesterol uptake with a concomitant decrease in cholesterol efflux. Compounds that can decrease cholesterol uptake or enhance efflux by adrenal tumor cells may have potential as therapeutics for CS caused by CPAs.