Endocrinology
EARLY RELEASE:
ENDOCRINE SOCIETY
ATR-101, a Selective and Potent Inhibitor of Acyl-CoA Acyltransferase 1, Induces Apoptosis in H295R Adrenocortical Cells and in the Adrenal Cortex of Dogs
Christopher R. LaPensee1, Jacqueline E. Mann2, William E. Rainey3, Valentina Crudo3, Stephen W. Hunt III4, Gary D. Hammer1
1Department of Internal Medicine, 2Department of Pathology, 3Department of Molecular and Integrative Physiology, Univ of Michigan, Ann Arbor, MI; 4Atterocor, Inc. Ann Arbor, MI
ATR-101 is a novel, oral drug candidate currently in development for the treatment of Adreno- cortical Cancer (ACC). ATR-101 is a selective and potent inhibitor of acyl-coenzyme A:cholesterol O-acyltransferase 1 (ACAT1), an enzyme located in the endoplasmic reticulum (ER) membrane that catalyzes esterification of intracellular free cholesterol. We aimed to identify mechanisms by which ATR-101 induces adrenocortical cell death. In H295R human adrenocortical carcinoma cells, ATR- 101 decreases the formation of cholesteryl esters and increases free cholesterol levels, demon- strating potent inhibition of ACAT1 activity. Caspase-3 levels and TUNEL-positive cells are increased by ATR-101 treatment, indicating activation of apoptosis. Exogenous cholesterol markedly poten- tiates the activity of ATR-101, suggesting that excess free cholesterol that cannot be adequately esterified increases caspase-3 activation and subsequent cell death. Inhibition of calcium release from the ER or the subsequent uptake of calcium by mitochondria reverses apoptosis induced by ATR-101. ATR-101 also activates multiple components of the Unfolded Protein Response (UPR), an indicator of ER stress. Targeted knockdown of ACAT1 in an adrenocortical cell line mimicked the effects of ATR-101, suggesting that ACAT1 mediates the cytotoxic effects of ATR-101. Finally, in vivo treatment of dogs with ATR-101 decreased adrenocortical steroid production and induced cellular apoptosis that was restricted to the adrenal cortex. Together, these studies demonstrate that inhibition of ACAT1 by ATR-101 increases free cholesterol, resulting in dysregulation of ER calcium stores that result in ER stress, the UPR, and ultimately apoptosis.
A drenocortical carcinoma (ACC) is a rare cancer of the adrenal cortex, with an estimated worldwide annual incidence of 0.5 to 2 cases per million, accounting for 0.2% of cancer deaths annually. ACC is highly aggressive with many patients presenting with metastases upon di- agnosis. Therapeutic options for the treatment of ACC are limited. Mitotane, an adrenolytic agent derived from the insecticide dichlorodiphenyltrichloroethane (DDT), is the only FDA-approved agent for ACC. While mitotane and cytotoxic chemotherapy are often used to treat ACC, the response rates are low, and discontinuation of mitotane- based therapy is common due to significant toxicity. Thus,
there exists an unmet need for new drugs or drug combi- nations that target key regulatory pathways in adrenocor- tical carcinoma cells that decrease tumor growth and ul- timately increase survival of patients with ACC.
The adrenal cortex is subdivided into three distinct zones of hormone synthesis: the zona glomerulosa, the zona fasciculata, and the zona reticularis, which express distinct enzymes to produce various steroids from a com- mon and requisite precursor, cholesterol (1). In adreno- cortical cells, esterified cholesterol is stored in lipid drop- lets, and serves as a large, obligate reservoir for the production of steroid hormones (2). At the interface of the
Abbreviations:
Copyright @ 2016 by the Endocrine Society
Received December 14, 2015. Accepted March 10, 2016.
doi: 10.1210/en.2015-2052
lipid droplet and mitochondrial membrane, esterified cho- lesterol is de-esterified by hormone sensitive lipase prior to the transfer of the resultant free cholesterol (FC) to ste- roidogenic acute regulatory protein (StAR) - that carries out the rate-limiting transfer of the FC to the inner mito- chondrial membrane for conversion to pregnenolone. The endoplasmic reticulum (ER) resident enzyme Acyl-CoA: cholesterol acyltransferase (ACAT), also known as sterol O-acyltransferase (SOAT), catalyzes (3) the conversion of cholesterol to its storage form cholesteryl esters, by form- ing an ester linkage between the 3-ß OH moiety in cho- lesterol, and the carboxyl group of a long chain fatty acid donated from the long chain fatty acyl coenzyme A (4).
Inhibition of ACAT1 has been considered a potential strategy for the treatment of several pathological condi- tions including hypercholesterolemia and atherosclerosis. Inhibition of cholesterol esterification was hypothesized to prevent the formation of atherosclerotic lesions com- prised of “foam” cells (5). However, mice with macro- phages engineered to lack ACAT1 had increased athero- sclerotic lesions, suggesting that the inability to esterify cholesterol in macrophages leads to the accumulation of FC in the artery wall that promotes, rather than inhibits, lesion development (6). Subsequent studies using ACAT1 inhibitors have shown that excess FC leads to toxicity in cultured macrophages (7, 8), demonstrating the impor- tance of ACAT1 activity in protecting cells that contain high levels of cholesterol (9).
While blockage of ACAT1 activity was hypothesized to decrease hepatic and intestinal cholesterol ester forma- tion, this approach to hypercholesterolemia management was abandoned due to in vivo toxicity, curiously limited to the adrenal cortex of multiple animal species (10-12). The toxic effects of the ACAT1 inhibitor PD 132 301-2 that were restricted to the adrenal cortex, raised the possibility of targeted ACAT1 inhibition as a treatment for ACC. PD 132 301-2, also known as ATR-101, is currently in Phase I Clinical development for the treatment of ACC. In this study, we determined the mechanism by which ATR-101 induces adrenocortical cell death.
Materials and Methods
Cell culture
The H295R and HAC clone 15 (HAC15) human adrenocor- tical cancer cell lines (both generous gifts from Dr. W.E. Rainey) were maintained as previously described (13). Briefly, H295R cells were grown in DMEM:F12 (1:1) containing 10% Nu Serum (BD Biosciences, Franklin Lakes, NJ), 1% Insulin-Transferrin- Selenium-X (Life Technologies, Grand Island, NY), and Pen/ strep (Life Technologies, Grand Island, NY). Cells were passaged by trypsinization (Life Technologies, Grand Island, NY).
HAC15 cells were grown in DMEM/F12 medium supplemented with 10% Cosmic Calf Serum (HyClone, Logan, UT), antibiot- ics, and 1% insulin/transferrin/selenium (Gibco). Treatment me- dia was prepared from a stock of 30 mm ATR-101 in sterile DMSO. For inhibitor studies, cells were pretreated for 30 min- utes with 100 nm U18666A (Sigma, St. Louis, MO), 100 um 2-APB (Tocris, Minneapolis, MN), 0.5 pm Xestospongin C (Tocris, Minneapolis, MN), 15 um Ruthenium Red (Calbio- chem, Billerica, MA), or 10 um Cyclosporin A (Tocris, Minne- apolis, MN).
Cytotoxicity assay
Cells were plated at a density of 15 000 cells/100 pl/well in clear 96 well plates. The following day, cells were incubated with ATR-101 in the presence or absence of cholesterol (C4951 Sigma, St. Louis, MO). Cholesterol-containing media was pre- pared from a stock of 7.5 mg/ml water-soluble cholesterol-meth- yl-ß-cyclodextrin. Cytotoxicity was determined by the 3-(4,5- dimethylthiazol-2-yl)2,5-diphenyl tetrazolium bromide (MTT) method. 15 ul of MTT was added at a final concentration of 0.5 mg/ml for 2 hours. Following medium aspiration, the formazan dye was extracted with 50 ul DMSO and absorbance was read at 570 nm using a plate reader.
Crystal Violet Staining
Cells were plated at a density of 15 000 cells/100 pl/well in clear 96 well plates. The following day, cells were incubated with ATR-101 in the presence or absence of cholesterol. Cell viability was measured by crystal violet staining. Following treatment, cells were washed with 1x PBS and fixed for 15 minutes in 4% paraformaldehyde in PBS. Cells were washed 3x with water and stained for 10 minutes with 0.4% crystal violet in 10% ethanol. Cells were washed 3x with water, allowed to air dry, solubilized in 1% sodium dodecyl sulfate for 15 minutes, and absorbance was measured at 570nm.
Caspase assay
Caspase 3/7 levels were measured using a luminescence-based kit (G6320) from Promega (Madison, WI). Cells (15,000/100 pl/well) were plated in an opaque 96-well plate. The following day, media was removed and replaced with 100 ul of treatment media, and plates were incubated at 37℃ for 5 hours. 100 ul of caspase detection reagent was added to each well, and light out- put, which correlates with caspase-3/7 activation as an indicator of apoptosis, was measured with a luminometer.
TUNEL Staining
Cells were plated on 8-well glass chamber slides. The follow- ing day, cells were incubated with ATR-101 in the presence or absence of cholesterol for 16 hours. TUNEL (terminal deoxy- nucleotidyl transferase dUTP nick end labeled) staining was per- formed using the DeadEnd™M Fluorometric TUNEL System from Promega (Madison, WI). Cells were counterstained with DAPI and mounted using ProLong Gold antifade reagent (Life Tech- nologies, Grand Island, NY),
Mitochondrial Membrane Potential Measurement
Cells were plated on a black, clear bottom 96-well plate. The following day, media was removed and replaced with treatment media, plates were incubated at 37°℃ for 5 hours, and mito-
chondrial membrane potential was measured using the TMRE- MMP Assay Kit (Abcam, Cambridge, MA). Following treat- ment, 2x TMRE (2 pm) was added to each well, and cells were incubated at 37 C for 15 minutes. Media was aspirated and replaced with 1x PBS containing 0.2% BSA and live cell TMRE levels were measured on a fluorescence plate reader.
Determination of intracellular free and esterified cholesterol
H295R or HAC15 cells were plated at 200 000 cells/well in a 24 well plate. After overnight incubation at 37°℃, free choles- terol loading was achieved by removing growth media and re- placing with treatment media containing 0 or 45 µg/ml choles- terol. After incubation for 5 hours, media was removed and cells were washed twice with cold PBS. 200ul of Hexane/Isopropanol (3:2) was added to each well, the plate was rocked at room temperature for 1 hour, and the supernatent was transferred to glass vials and allowed to air dry. Residual lipids were resus- pended and measured using the Cholesterol Fluorometric Assay Kit (Cayman Chemical, Ann Arbor, MI). For calculation of es- terified cholesterol, the amount of free cholesterol was sub- tracted from total cholesterol level, as determined by addition of esterase in the assay.
Determination of ACAT activity
ATR-101 inhibitory activity and EC50 was measured using a fluorescent cell-based assay measuring esterification of 22-[N- (7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-23,24-bisnor-5- cholen-3-ol (NBD-cholesterol). AC29 cells lacking endogenous ACAT activity were transfected with constructs expressing hu- man ACAT1 or ACAT2, and esterification was determined as described (14).
RT-PCR
Cells (200,000/well) were plated in a 12 well plate, incubated overnight at 37℃, and treated for 5 hours. Total RNA was extracted using RNeasy Plus minikits (QIAGEN, Valencia, CA). 1 µg of total RNA was reverse transcribed using the high-ca- pacity cDNA archive kit (Applied Biosystems, Foster City, CA). Quantitative PCR for human CHOP gene expression was per- formed on cDNA using Power SYBR Green PCR Mastermix (Applied Biosystems, Foster City, CA) and intron spanning prim- ers: forward 5’-AGAGCCCTCACTCTCCAGATTCCA-3’, re- verse 5’-TCTGTTTCCGTTTCCTGGTTCTCC-3’. Gene ex- pression was normalized to GAPDH (forward 5’- CTTCCAGGAGCGAGATCCCTC-3’, reverse 5’- TCCACGATACCAAAGTTGTCAT-3’) using the 24AcT method and expressed as fold-change compared to the vehicle treated group, which is normalized to 1. Product purity was confirmed by DNA melting curve analysis. Detection of XBP1 splicing was performed as described previously (15). The for- ward and reverse primers used for XBP1 splicing were 5’ GAA CCA GGA GTT AAGAACACG 3’ and 5’ AGG CAA CAGTGT CAG AGT CC 3’, respectively.
Western blotting
For detection of phosphorylated PERK, cells were homoge- nized in lysis buffer containing 20 mm HEPES, 150 mm NaCl, 1% Triton X-100, 10% glycerol, 1 mm EDTA, 10 mm sodium pyrophosphate, 100 mm sodium fluoride and protease inhibi-
tors. Protein concentration was determined by the Pierce bicin- choninic acid assay. Lysates (40 µg) were separated by SDS- PAGE and transferred to a nitrocellulose membrane. Phosphorylated PERK was detected using an Anti-PERK anti- body (Abcam, Cambridge, MA) at a 1:1000 dilution. Proteins were visualized using the Odyssey scanning system (LI-COR Bio- sciences, Lincoln, NE). Molecular weight was estimated using MagicMark Western Standard from Invitrogen.
Knockdown of ACAT1
ACAT1 expression was silenced in HAC15 cells using GIPZ Lentiviral shRNA technology (Dharmacon, Lafayette, CO); ACAT1 (SOAT1 in humans) (V2LHS_153311) ATAAGAAG- TACAAATACTG. Lentiviral particles containing SOAT1 shR- NAs were transduced into HAC15 cells (MOI 3.0) with 2.4 ug/ml of polybrene (EMD Millipore Corporation, Billerica, MA). Transduced cells were then selected by 10 µg/ml of puro- mycin (Sigma-Aldrich, St. Louis, MO). The concentration of puromycin was determined according to a puromycin selectivity test with HAC15 cells. After the antibiotic selection, cells were plated at a density of 150 000 per well and target gene expression was analyzed after RNA extraction and qPCR analysis. qPCR reactions were performed in the ABI StepOnePlus Real-Time PCR systems (Applied Biosystems, Foster City, CA). Primer and probe mixtures for the amplification of the SOAT1 (Hs00162077_m1) and PPIA (Hs99999901_m1) target se- quences were purchased from Applied Biosystems. PPIA tran- script was used for normalization of sample loading. Relative gene expression was determined using the 24AcT method.
Serum cortisol and steroid and steroid intermediate/metabolite analysis
Studies were conducted at MPI Research, Mattawan, Mich- igan, and were approved by the Institutional Animal Care and Use Committee at MPI Research. Three naïve male beagle dogs (approximately 10 to 12 kg) were administered ATR-101 in 0.5% hydroxypropylmethylcellulose via oral gavage once a day at approximately the same time at 3 mg/kg/d for 7 days followed by treatment at 30 mg/kg/d for an additional 7 days. The dose volume was 5 mL/kg. Animals were observed twice daily for morbidity, mortality, injury, and availability of food and water. The dogs were fed Certified Canine Diet #5007, from PMI Nu- trition International, Inc. This diet is described as a palatable and complete life-cycle diet for reproduction, growth and mainte- nance of laboratory dogs, and contains cholesterol. The animals were fed ad libitum and were in a normal state of cholesterol intake and utilization. Clinical observations were conducted pre- test and approximately 1 hour postdose weekly. Body weights were recorded pretest on prior to dosing on days 1, 8, and 14. Food consumption was measured daily. Blood samples for anal- yses of serum cortisol and steroid and steroid intermediates/me- tabolites were collected from all animals at approximately the same time each morning predose and 1 hour after intravenous (IV) administration of Cortrosyn® [@ 1-24 corticopin synthetic subunit of ACTH; 5 µg/kg] on days -3, 1, 3, 7, 8, 10, and 14. Serum was isolated, and the samples were divided into two ap- proximately equal aliquots. The first aliquot was analyzed for serum cortisol levels by MPI Research (Mattawan, MI) using the Canine Cortisol Quantitative ELISA Kit, (Endocrine Technolo- gies, Newark CA). The second aliquot was sent to the University
of Michigan Metabolomics Biomedical Core Research Facilities, Ann Arbor, Michigan and analyzed for concentrations of ste- roids and steroid intermediates/metabolites.
Mass spectrometry
Samples for analysis of cortisol, corticosterone, testosterone, 11-deoxycortisol, 11-deoxycorticosterone, 17-hydroxyproges- terone, and androstenedione, were prepared by a combination of protein precipitation and liquid-liquid extraction and analyzed by liquid chromatography-tandem mass spectrometry (LC/MS- MS). 100 uL aliquots of serum were diluted 2:1 with DI H2O in a 1.5 mL centrifuge tubes. Protein was precipitated by sequential addition of 100 uL of methanol and 200 uL acetonitrile. After addition of 100 ul of deuterated internal standard [2H4] (C/D/N Isotopes, Pointe-Claire, Quebec), the suspension was vortex mixed and centrifuged for 5 minutes at 13 000 rpm. The super- natant was transferred to a 2 mL tube with 0.5 mL water, and steroids were extracted with 1 mL of methyl-t-butyl ether (MTBE). The organic phase was transferred to a clean 2 ml tube and dried under UHP nitrogen. The dried extract was reconsti- tuted with 100 uL of 50% aqueous methanol (v/v) and trans- ferred to a 300 uL vial insert. Samples were analyzed with an Agilent 1290 HPLC and 6490 triple quadrupole LC-MS/MS us- ing electrospray ionization in positive ionization mode. The fol- lowing steroids (cortisone, cortisol, corticosterone, testosterone, 11-deoxy cortisol, 11-deoxycorticosterone, progesterone, 17- hydroxyprogesterone) were analyzed in positive ionization mode and resolved on a Kinetex 50 × 2.1 mm, 2.6 um particle size C 8 column (Phenomenex, Torrence, CA) using gradient elution with 10 mM ammonium acetate (mobile phase A) and methanol (mobile phase B). Quantitation was by external stan- dardization with a nine-point calibration curve using multiple reaction monitoring. Samples were prepared for analysis of DHEAS by performing a second liquid-liquid extraction on the 100 ul aliquot after MTBE extraction. The remaining aqueous phase was pH adjusted by addition of 100 uL of 0.1% phos- phoric acid. The samples were extracted with 1.0 ml of a 50:50 (v/v) chloroform/2-butanol solution. The extract was dried un- der UHP nitrogen and reconstituted with 50:50 methanol/water. DHEAS was analyzed in negative ionization mode using gradient elution with 0.2 mM ammonium fluoride (mobile phase A) and methanol+0.2 mM ammonium fluoride (mobile phase B). Quantitation was by external standardization with a nine-point calibration curve using multiple reaction monitoring.
Canine necropsy
At study termination, animals were euthanized by an IV over- dose of sodium pentobarbital solution followed by exsanguina- tion via severing the femoral vessels, and necropsy examinations were performed. Examination of adrenal glands was performed under procedures approved by a veterinary pathologist on all animals. The adrenal glands of all ATR-101-treated animals were collected, as well as adrenals from one naïve stock dog for comparison. The left adrenal gland was sent to Wake Forest University for assessment of free cholesterol and cholesterol es- ters. The right adrenal gland was fixed in 10% neutral buffered formalin and processed to slides for light microscopy evaluation. Hematoxylin and eosin, and TUNEL-stained slides were exam- ined for histopathologic changes including apoptosis. The per-
centage of cells that stained positive for TUNEL was calculated using Image J software.
Adrenal esterified and free cholesterol analysis
Esterified and free cholesterol concentrations in the adrenal were determined at The Lipid, Lipoprotein, and Atherosclerosis Analysis Laboratory at Wake Forest Innovation Quarter, Wake Forest University, Winston-Salem, North Carolina as described (16). Briefly, lipid extracts were prepared by treatment of adrenal homogenates extracted with detergent, and the esterified and free cholesterol concentrations in the lipid extracts were deter- mined using an enzymatic assay kit.
ATR-101 tissue distribution analysis
Studies were conducted at ITR Laboratories, Quebec, Can- ada, and the blood and tissue sample concentration analyses were conducted at AIT Bioscience, Indianapolis, Indiana. 3 fe- male beagle dogs approximately 19 months old weighing 7 to 10 kg were given daily oral doses of ATR-101 in 0.5% hydroxy- propylmethylcellulose at 3 mg/kg/d for 7 days using a dose vol- ume of 5 mL/kg. On the last day of dosing, blood was collected from all animals 4 hours postdose for determination of ATR-101 concentrations in plasma and erythrocytes. Animals were then euthanized and exsanguinated, and adrenal and ovary weights were recorded. Adrenal glands, kidney, liver, skeletal muscle, subcutaneous fat, ovaries, and a sample of cerebrospinal fluid (CSF) were collected for determination of ATR-101 content. All samples were analyzed using AIT Bioscience Bioanalytical Method BAM.0095.01, which was developed and validated for the quantitation of ATR-101 in dog plasma. This method was validated over a range of 1.00 to 1000 ng/ml using ATR-101-13 C4 as an internal standard. Briefly, tissue samples were homog- enized, and aliquots of the homogenates, blood, plasma, and CSF were extracted and analyzed using LC-MS. The lower limit of quantitation was 1.0 ng/ml in plasma, erythrocytes, and CSF, and 2.5 ng/g in tissue samples. Results were calculated using peak area ratios of analytes to internal standard, and calibration curves were generated using a weighted (1/x2, where x = con- centration) linear least-squares regression. Statistics were calcu- lated from individual animal data, and tissue concentration as a percentage of plasma concentration in each species was determined.
Results
ATR-101 induces cytotoxicity in H295R adrenocortical carcinoma cells
We examined the sensitivity of H295R adrenocortical carcinoma cells to increasing concentrations of ATR-101. As shown in Figure 1A, 3 nm - 3 um ATR-101 exhibited no toxicity in the absence of exogenous cholesterol, as measured by MTT assay, while 30 um ATR-101 treat- ment in the absence of cholesterol reduced survival by ~40% within 24 hours. Coincubation of ATR-101 and cholesterol markedly increased toxicity in a dose-depen- dent manner, where 3 nm ATR-101 in the presence of 60 ug/ml cholesterol reduced survival by 60% after 24 hours.
All doses of ATR-101 (3 nm - 30 um) induced cytoxicity in the presence of cholesterol, whereas treatment with cho- lesterol in the absence of ATR-101 had no effect on cell viability. To demonstrate that the effects of ATR-101 seen by MTT assay were not due to changes in mitochondrial activity, cell viability was assessed by crystal violet stain- ing. Figure 1B shows that coincubation of 30nm or 300nm ATR-101 and cholesterol decreased cell viability, while ATR-101 alone had no effect on cells. Treatment with cholesterol-free methyl-ß-cyclodextrin alone had no effect on cell toxicity or viability (not shown).
Apoptosis is induced by ATR-101 in H295R adrenocortical carcinoma cells
To determine whether the toxic effects of ATR-101 involve apoptotic pathways, caspase 3/7 activity was mea- sured in cells treated with ATR-101 and cholesterol. Fig- ure 1C shows that 30 nm ATR-101 induced apoptosis in the presence, but not the absence of cholesterol (45 µg/ml), while cholesterol alone did not increase caspase 3/7 activ-
ity. We next tested whether 36-[2-(diethylamino)ethoxy]- androst-5-en-17-one hydrochloride (U18666A), which selectively impairs cholesterol trafficking to the ER (17, 18), could prevent the apoptosis induced by ATR-101. Also shown in Figure 1C, coincubation of cells with 30 nm ATR-101 and 45 µg/ml cholesterol in the presence of 100 nm U18666A protected the cells from apoptosis. These observations indicate that, localization of cholesterol at the ER mediates cell death induced by incubation with low nanomolar concentrations of ATR-101 in the presence of cholesterol. Apoptosis was also assessed by TUNEL stain- ing. Figure 1D shows that cells treated with cholesterol alone exhibited no TUNEL staining, whereas treatment with ATR-101 and cholesterol significantly increased the number of apoptotic cells. Cells coincubated with ATR- 101 and cholesterol also exhibited pyknotic nuclei char- acteristic of apoptotic cells.
ATR-101 selectively inhibits ACAT1 and increases FC levels
A
B
Viability (% of Control)
125
24hr
Cholesterol (µg/ml)
Viability (% of Control)
125
24hr
Cholesterol (µg/ml)
30
100
100
MTT assay
0
Crystal Violet
*
45
*
75-
*
*
**
75
**
**
* * .
15
50
**
#
30
50
25-
#
#
*#
#
#
25
#
60
0 0
0
0.003
0.03
0.3
0
3
30
0
0
0.03
0.3
ATR-101 (UM)
ATR-101 (UM)
C
D
300
Caspase 3/7 activity (% vs Control)
Chol
200
100
ATR-101
0
+
ATR-101 Chol
Chol
+
+
+
-
-
-
+
+
+
-
-
-
U18666A
+
+
-
-
-
-
To investigate the selectivity of ATR-101 toward ACAT1 and ACAT2, we used two cell lines in which either human ACAT1 and ACAT2 genes were transfected into the ACAT-deficient Chinese ham- ster ovary cell line AC29. ATR-101 potently inhibited ACAT1 com- pared to ACAT2, exhibiting EC50 values in ACAT1 and ACAT2 of 0.009 um vs 0.368 um, respectively (Figure 2A).
To determine if ATR-101 inhibits esterification of cholesterol in adre- nocortical carcinoma cells, intracel- lular free and esterified cholesterol levels were measured in H295R cells treated with ATR-101. As shown in Figure 2B, the FC:CE ratio in vehicle- treated and 9 nm ATR-101 treated cells in the absence of exogenous cholesterol is ~7:1, with no differ- ence in the amount of total intracel- lular cholesterol content. Treatment with 9 nm ATR-101 in presence of cholesterol results in a FC:CE ratio of ~5:1. Incubation of cells with 45 ug/ml cholesterol in the absence of ATR-101 for 5 hours decreases the FC:CE ratio to 1.4:1. FC levels were approximately 70% higher in ATR- 101 treated cells. The increased FC
levels and FC:CE ratio observed in ATR-101-treated com- pared to vehicle treated cells is consistent with inhibition of ACAT1 activity in the presence of cholesterol. The total amount cholesterol (FC and CE combined) was not dif- ferent between ATR-101 and vehicle treated cells.
ATR-101 activates the Unfolded Protein Response
ER membranes normally contain low levels of choles- terol. Because excess FC that accumulates in the ER mem- brane has been shown to decrease membrane fluidity and ultimately perturb ER functions, (9, 19) we hypothesized that excess FC caused by inhibition of ACAT1 disrupt the ER in ACC cells. Predicated on this hypothesis, we exam- ined whether ATR-101 induced ER stress in H295R cells. The Unfolded Protein Response (UPR) is initiated by ER- resident transmembrane protein kinases that become phosphorylated under conditions of ER stress. These ki- nases activate downstream transcription factors that in- duce genes involved in restoring normal ER function. When stress is prolonged or severe, this programmed re- sponse can induce apoptotic cell death.
IRE1 is an ER membrane endonuclease that oligomer- izes and trans-autophosphorylates in response to ER fold- ing stress. Active IRE1 facilitates splicing of the mRNA for X-box-binding protein-1 (XBP-1), leading to the synthe- sis of a highly active splice variant of the transcription factor called XBP-1 that is highly specific to the UPR. Coincubation of 30 nm ATR-101 and 45 µg/ml choles- terol induced XBP-1 splicing (XBP-1s) after 4 hours (Fig- ure 3A), while treatment of H295R cells with 30 nm ATR- 101 or 45 µg/ml cholesterol alone did not affect XBP-1 splicing (not shown).
Like IRE1, (Protein kinase-like ER Kinase) PERK is an ER membrane protein activated by trans-autophosphor- ylation during ER stress. PERK Phosporylation is detected
as an upward shift in protein mobility on immunoblots. Figure 3B shows that coincubation of 30 nm ATR-101 and 45 µg/ml cholesterol for 5 hours resulted in delayed mi- gration reflective of active phosphorylated PERK, while cholesterol alone did not activate PERK. Thapsigargin, a potent inhibitor of SERCA that induces ER stress and sub- sequent PERK phosphorylation served as a positive con- trol. Accumulation of the transcription factor C/EBP ho- mologous protein (CHOP) is a downstream event that is indicative of activation of the UPR and PERK phosphor- ylation. We observe that CHOP mRNA expression is in- duced ~2.5-fold in H295R cells treated with 30 nm ATR- 101 and 45 µg/ml cholesterol (Figure 3C). Induction of CHOP mRNA was not observed in the absence of cho- lesterol (not shown). These observations suggest that ER stress is induced by FC when ACAT1 activity is inhibited.
Calcium mediates ATR-101 induced apoptosis in adrenocortical cells
A critical mechanism by which prolonged or extreme ER stress enables apoptosis in some cell types is through release of ER calcium stores (20). To determine whether ER-derived calcium plays a role in ATR-101-induced ap- optosis in adrenocortical cells, we performed a series of studies in the presence of Xestospongin C (21) and 2-Ami- noethoxydiphenyl borate (2-APB) (22), potent inhibitors of calcium release from the ER via inositol 1,4,5-triphos- phate receptors (IP3R). Figure 4 shows that treatment with 30 nm ATR-101 in the presence of cholesterol for 5 hours induced apoptosis, while pretreatment with (A) Xe- stospongin C or (B) 2-APB protected cells from apoptosis induced by coincubation with ATR-101 and cholesterol. In macrophages, increased cytosolic calcium as a result of ER stress induces cell death in part through a process that involves mitochondrial uptake of calcium and subsequent release of cytochrome C, an activator of apo-
ptotic caspases (20, 23). To deter- ** mine whether ATR-101 induces ad- renocortical cell death through a similar pathway, we next utilized Ruthenium Red (RR), a pharmaco- logical inhibitor of the mitochon- T drial calcium uniporter (24). As shown in Figure 4C, coincubation of cells with 30 nm ATR-101 and 45 + + ug/ml cholesterol in the presence of - + RR protected the cells from caspase activation, indicating that ATR- 101-induced caspase activation re- quires mitochondrial uptake of cal- cium derived from the ER (Figure 4A
A
B
125
ACAT1
50
100
ACAT2
FC
% ACAT activity
UM Cholesterol
40
75
CE
30
50
1
25
20
0
10
-3
ATR-101 (log uM)
-2
-1
0
1
2
-25
0
ACAT1 ACAT2
ATR-101
-
-
EC50 (μM)
.009
.368
Chol
-
+
and 4B). Next, we examined whether Cyclosporin A, an inhibitor of mitochondrial membrane permeabilization (25) and subsequent cytochrome c release, could block the activation of caspase 3/7 by ATR-101. Figure 4C also shows that ATR-101 did not activate caspase 3/7 in the presence of Cyclosporin A. We then assessed whether ATR-101 induced mitochondrial membrane permeabili- zation, as judged by TMRE (tetramethyl rhodamine), a dye that readily accumulates in active mitochondria. De- polarized or inactive mitochondria have decreased mem- brane potential and fail to sequester TMRE. Figure 4D shows that ATR-101 decreased mitochondrial membrane potential, an effect that was reversed by Cyclosporin A. No effect of ATR-101 or cholesterol alone was observed. The mitochondrial uncoupler FCCP (Carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone) (10 um) de- creased TMRE by ~70% (not shown).
Deficiency of endogenous ACAT1 sensitizes adrenocortical carcinoma cells to cholesterol- induced death
To verify that the effects of ATR-101 are mediated through ACAT1, we used shRNA to generate stable ad- renocortical HAC15 cells with downregulated ACAT1 expression. As determined by PCR, ACAT1 expression was decreased by 80% in shACAT1 cells compared to cells transduced with nontargeting shRNA (not shown). As shown in Figure 5A, the FC:CE ratio in ACAT1-deficient and control cells in the absence of exogenous cholesterol is ~6:1, with no difference in the amount of total intra-
A
C
XBP-u
5
Chol
ATR-101 + Chol
XBP-s
4
*
Chop mRNA
3
B
2
P-PERK
EPERK
Chol
ATR-101 + Chol
Thaps
1
0
ATR-101
+
-
Chol
+
+
cellular cholesterol content. Cholesterol treatment of cells transduced with nontargeting shRNA for 5 hours de- creased the FC:CE ratio to ~1:1, whereas incubation of ACAT1-deficient cells in presence of cholesterol resulted in a FC:CE ratio of ~4:1. FC levels and FC:CE ratio ob- served in ACAT1-deficient cells were comparable to ATR- 101 treated cells (Figure 2), and are consistent with inhi- bition of ACAT1 activity in the presence of cholesterol. The total amount or cholesterol (FC and CE combined) was not different between ACAT1-deficient and shCon- trol cells.
The effects of cholesterol on cell toxicity were next ex- amined in ACAT1-deficient cells. Treatment with FC de- creased the viability of ACAT1-deficient cells after 24 hours (Figure 5A) and 72 hours (Figure 5B), and increased apoptosis after 4 hours (Figure 5C). These findings are consistent with ACAT1 inhibition by ATR-101 in the presence of cholesterol (Figure 2). Notably, cholesterol alone is nontoxic to cells transduced with a nontargeting shRNA. Figure 5D shows that cholesterol-induced apo- ptosis in ACAT1-deficient cells is blocked by U18666A, Ruthenium Red, and Cyclosporin A. These observations are consistent with blockade of ATR-101-induced cell death by the inhibitors above shown in Figures 2 and 4.
Effects of ATR-101 in dogs
ATR-101 was administered to three female dogs, orally, once daily for 7 days at 3 mg/kg/d followed by treatment at 30 mg/kg/d on for an additional 7 days. Plasma concentrations of ATR-101 doses (Figure S1) ap- proximated those previously seen in a toxicology study (not shown) at 4 hours postdose at 3 mg/kg/d for 14 days followed by 30 mg/kg/d for 14 days. Tissue distribution analysis of ATR-101 in treated dogs revealed preferential distribution of ATR-101 in adrenal glands after oral dos- ing at 3 mg/kg/d for 7 days (Figure S2). When ATR-101 concentrations in tissues were normalized to plasma, only the adrenal had concentrations approaching equivalence to plasma. No other tissues demonstrated ATR-101 con- centrations that were even 30% of the ATR-101 plasma concentration. ATR-101 was not measurable in any of the samples of CSF.
ATR-101 had no significant effects on body weight or any other clinical parameter in the male canines. A sepa- rate, 28 day toxicity study of ATR-101 in male dogs showed no effects on testes or macrophages (not shown). However, analysis of all three ATR-101-treated animals revealed histologic changes in the adrenal gland, including apoptosis, cortical atrophy and substantial cellular vacu- olation when compared to untreated animals at 14 days following treatment initition (Figure 6A). These changes primarily affected the zona fasciculata; however, mono-
nuclear cell infiltrates extended multifocally into the zona glomerulosa and reticularis. These findings were also as- sociated with minimal to substantial mixed inflammatory cell infiltration composed of lymphocytes, with fewer plasma cells and histiocytes. Intranuclear TUNEL stain- ing, indicative of apoptotic cell death was detected in ATR-101-treated dogs within the zona glomerulosa, zona fasciculata and zona reticularis, but not the medulla (Fig- ure 6B). Analysis using Image J software revealed that ~60% of cortical cells in each representative field of ad-
renal glands from ATR-101 treated dogs stained positive for TUNEL. There were no microscopic abnormalities and TUNEL staining detected within the adrenal gland of a naïve control animal. Adrenal glands of dogs treated with ATR-101 were analyzed for free and esterified cholesterol content. A marked decrease in CE (191 µg/mg protein) was detected in the adrenal glands or dogs treated with ATR-101, compared to 606 µg/mg protein in a nontreated dog, consistent with inhibition of cholesterol esterification by ATR-101. Adrenal FC levels were 37 µg/mg protein in ATR-101-treated dogs compared to 38 µg/mg protein in a nontreated dog (Figure S3).
A
B
250
250
T
Caspase 3/7 activity (% vs Control)
200
T
200
150
150
100
1
100
50
50
0
0
ATR-101
+
+
ATR-101
+
+
-
-
-
-
Xesto C
+
+
2-APB
+
+
-
-
-
-
Chol
+
+
+
+
Chol
+
+
+
+
C
D
250
150
Caspase 3/7 activity (% vs Control)
200
TMRE (% vs Control)
100
150
T
T
100-
50
50
0
ATR-101
+
+
+
ATR-101
+
+
+
+
-
-
-
RR
+
+
Cyc A
+
+
-
-
-
-
-
Cyc A
+
+
Chol
+
+
+ +
-
-
-
-
Chol
+ +
+ +
Adrenocortical function was as- sessed by measuring serum pre- ACTH and post-ACTH stimulated levels of several steroids and their precursors. Prior to ATR-101 treat- ment, ACTH stimulated an increase in cortisol levels from 0.8 to 29 µg/dl (Figure 7A). A marked decrease in ACTH-stimulated cortisol was ob- served over the course of ATR-101 treatment. After 7 days of ATR-101 treatment at 3 mg/kg/d, ACTH-stim- ulated cortisol levels were decreased by 62%. On day 14 (following an additional 7 days of ATR-101 treat- ment at 30 mg/kg/d), ACTH-stimu- lated cortisol was reduced by 71%. In addition to profound decreases in glucocorticoids, ACTH-stimulated levels of the mineralocorticoid corti- costerone (Figure 7B) were signifi- cantly lower in dogs treated with ATR-101 for 14 days. The upstream glucocorticoid precursors 17-OH- progesterone and 11-deoxycortisol (Figure 7A), as well as the mineralo- corticoid precursor 11-deoxycorti- costerone (Figure 7B) were similarly decreased. Pre-ACTH-stimulated levels of the adrenal androgens an- drostenedione and DHEA-S (Figure 7C) were significantly lower in dogs treated with ATR-101 for 14 days. Remarkably, ACTH stimulation did not result in any increase in adrenal androgens in this cohort (data not shown). Lastly, even the pre-ACTH levels of a panel of multiple steroid intermediates were significantly
lower following 14 days of ATR-101 treatment when compared to nontreated dogs (Table 1). These results, ob- served after treatment with ATR-101, are consistent with both the in vitro and in vivo studies documenting a de- crease in adrenal steroid production/secretion and ulti- mate adrenocortical cell death resulting from ACAT1 inhibition.
Discussion
Adrenocortical cells import and store large amounts of cholesterol to meet the need for production of steroid hor- mones. ACAT1 esterifies FC to maintain a large inert res- ervoir of cholesterol for steroid production. In non- steroidogenic cell types, such as macrophages that uptake large amounts of cholesterol (23), ACAT1 activity pre- vents the accumulation of toxic FC in various cell mem- branes. Based on these observations, we reasoned that adrenal glands are susceptible to perturbations in choles- terol homeostasis, and hypothesized that inhibition of ACAT1 by ATR-101 leads to toxic accumulation of FC in adrenal cells. Several observations in this study support this notion. First, ATR-101 in the absence of cholesterol
A
B
shC
shACAT1
24hr
72hr
**
UM Cholesterol
50
Viability (% of Control)
125
*
125
**
**
40.
100
T
T
T
100
1
FC
CE
T
30-
75
75.
20
50
50
10
T
25
25.
T
0
Chol
0
0
+
+
Chol
+
Chol
+
-
-
-
-
shControl
shACAT1
C
250
D
250
*
Caspase 3/7 activity (% vs Control)
200
shC
Caspase 3/7 activity (% vs Control)
200
shC
150
shACAT1
150
shACAT1
**
**
T
T
100
100
T
T **
T
Y
50
50
0
Chol
-
+
Chol
0
+
-
+
+
+
-
-
-
UA
+
+
-
-
-
-
-
-
RR
-
-
+
+
-
-
-
-
Cyc A
-
-
-
-
-
+
+
-
was nontoxic to H295R cells at concentrations ranging from 3 nm to 3 um. Included in this range were concen- trations demonstrated to potently and selectively inhibit ACAT1 activity, which first led us to conclude that inhi- bition of ACAT1 in the absence of its substrate cholesterol had no effect on adrenocortical cell viability. Addition of exogenous cholesterol alone was also nontoxic, but ex- posure of adrenocortical cells to ATR-101 in the presence of cholesterol dramatically increased cell death. Consis- tent with this observation, shRNA-mediated knockdown of ACAT1 was toxic in the presence, but not the absence of cholesterol.
Blocking the trafficking of cholesterol to the ER where ACAT1 resides (using U18666A) reverses the toxic effects of ACAT1 inhibition achieved by either ATR-101 or si- lencing of the ACAT1 gene. Based on these observations, we concluded that FC mediates the cytotoxic effects of ATR-101, and thus all subsequent experiments in this study were performed in the presence of exogenous cho- lesterol. Studies designed to elucidate mechanisms under- lying cholesterol-induced cell death in models of athero- sclerotic lesions have utilized a similar experimental paradigm, where macrophage death is studied in the pres- ence of exogenous FC.
The cholesterol content of ER membranes is low, and ER function is particularly sensitive to excessive FC (26). FC enrichment of the nor- mally fluid ER membrane causes stiffening that inhibits sarcoendo- plasmic reticulum ATPase (SERCA), an ER membrane Ca++ transporter that is crucial for maintaining high concentrations of ER calcium (19, 27). SERCA inhibition can induce ER stress by rapidly depleting the lu- minal Ca2+ required for proper ER chaperone function and protein fold- ing (9, 28), resulting in XBP splicing, phosphorylation of PERK, induc- tion of CHOP expression, caspase-3 activation, and ultimately apoptotic cell death (17, 19, 29-32). Cell sur- vival in the initial stages of ER stress gives repair processes mediated by transcription factors such as XBP and CHOP a chance to work. Among the genes upregulated during the UPR in yeast (33) is Are1 (Acyl- coenzyme A: cholesterol acyl trans- ferase-Related Enzyme), a homolog of ACAT1 that esterifies FC, raising
the idea that some cell types have developed mechanisms for reducing the burden of FC accumulation in the ER membrane (26).
Given that ER calcium depletion is a known inducer of the UPR, this event most probably contributes to FC-in- duced UPR activation in macrophages, as well as in ATR- 101-treated H295R cells and ACAT1-deficient HAC15 cells. Conversely, U18666A prevents FC-induced ER stress, activation of the UPR, and apoptosis in macro- phages (17, 30-32). These observations indicate that the ER is a key mediator of FC-induced UPR and cell death in macrophages, and that calcium plays a key role in this process (26). We find that ATR-101 also upregulates sev- eral indicators of dysregulated ER function in adrenocor- tical cells. Key indicators of the activated UPR, including splicing of XBP-1, phosphorylation of PERK, and CHOP
A
zG
B
zF
ZG
ZR
Med
zF
C
D
zG
ZG
zF
ZR
Med
zF
E
zG
F
ZG
zF
zR
zF
Med
G
H
zG
zF
zG
ZR
zF
Med
mRNA were increased by ATR-101 (34, 35). The fact that inhibition of calcium release from ER-IP3Rs by Xesto- spongin C or 2-APB reversed cell death in ATR-101 treated and ACAT1-deficient cells strongly indicate that calcium mediates cell death under these conditions, and is consistent with prior work indicating that pharmacolog- ical blockade of IP3R prevents apoptotic cell death in- duced by FC in macrophages or SERCA inhibitors in ß cells (29, 36). Apoptosis often involves mitochondrial dys- function triggered by the mitochondrial uptake of ER- released calcium through a highly selective ion channel in the setting of ER stress (20, 23, 37, 38). In the presence of excess FC, ER-derived calcium taken up by mitochondria leads to a loss of mitochondrial membrane potential, mi- tochondrial membrane permeabilization, and apoptosis (23). Our findings that Ruthenium Red, an inhibitor of mitochondrial calcium uptake, and Cyclosporin A, which prevents disruption of the mitochondrial transmembrane potential during apoptosis (39, 40), blocked apoptosis in- duced by ATR-101 or during ACAT1 deficiency suggest that ER-derived calcium leads to release of apoptotic ef- fectors from mitochondria. Other reports have shown that proapoptotic mitochondrial dysfunction can be triggered by the mitochondrial uptake of ER-released calcium in the setting of ER stress (41). While it is possible that nonmi- tochondrial targets of Cyclosporin A, such as Calcineurin, could contribute to the protective effects against ATR- 101, the loss of mitochondrial membrane potential in ATR-101-treated cells and reversal by Cyclosporin A is consistent with the hypothesis that mitochondrial dys- function is a downstream event mediating cell death fol- lowing inhibition of ER-residing ACAT1. Similar obser- vations have been reported in other cell types, where inhibition of mitochondrial calcium uptake can prevent apoptosis induced by FC (20) or calcium-induced (24) cell death. It is unlikely that FC loading of mitochondria con- tributes to cell death, because incubation of isolated mi- tochondria with excess cholesterol in vitro actually stabi- lizes mitochondrial function (42). Moreover, ATR-101 does not appear to directly disrupt mitochondrial func- tion, as determined by the lack of effect of ATR-101 on electron transport chain complexes I, II, and III in isolated mitochondria (Figure S4).
FC loading in macrophages triggers mitochondrial membrane permeabilization and release of cytochrome c from mitochondria, leading to activation of the effector caspase 3 and subsequently apoptosis (23). Given that cytochrome c is required for caspase-3 activation, and el- evated caspase-3 activity is observed in the presence of ATR-101 and in ACAT1 deficient cells, our data are con- sistent with the hypothesis that FC-induced apoptosis in- volves caspase activation by cytochrome c that is released
| Pathway | Steroid/intermediate | Pre-ACTH levels Day -3 to Day 14 (% decrease) | Post-ACTH levels Day -3 to Day 14 (% decrease) |
|---|---|---|---|
| Glucocorticoids | 17OH-Progesterone | 100 | 100 *** |
| 11-Deoxycortisol | 11 | 78 *** | |
| Cortisol | 32 | 71 *** | |
| Cortisone | 19 | 44* | |
| Mineralocorticoids | 11-Deoxycorticosterone | 67* | 86 *** |
| Corticosterone | 50 | 87 ** | |
| Adrenal Androgens | DHEAS | 100 *** | 100 *** |
| DHEA | 52 | 50 | |
| Androstenedione | 78* | 87 *** |
Significant differences from day -3 levels are designated by * (P < 0.05), ** (P < 0.005), and *** (P < 0.0005).
| Peptide/protein target | Antigen sequence (if known) | Name of Antibody | Manufacturer, catalog #, and/or name of individual providing the antibody | Species raised in; monoclonal or polyclonal | Dilution used |
|---|---|---|---|---|---|
| Protein kinase-like endoplasmic reticulum kinase (PERK) | Anti-PERK | Abcam ab65142 | rabbit polyclonal | 1:1000 |
from dysfunctional mitochondria (25, 43, 44). Because cytochrome c has been reported to bind and activate IP3Rs (45), an IP3R-calcium-mitochondria pathway of apopto- sis is part of a positive feedback amplification cycle that increases cell death. In this scenario, inhibition of either calcium release from the ER, calcium uptake by mitochon-
A
B
17-OH Progesterone
3
11-Deoxycorticosterone
800
2
T
pg/ml
600
-ACTH
ng/ml
400
*
+ACTH
1
*
200
-3
1
T
Day of treatment
0
Z
I
-3
1
3
Day of treatment
7
10
14
30
Corticosterone
20
T
ng/ml
T
25
11-Deoxycortisol
10
ng/ml
20
0
*
-3
1
15
T
**
Day of treatment
10
T
C
5
T
T
3
T
0
2
7
T
Androstenedione
-3
1
3
7
10
14
ng/ml
2
Day of treatment
**
1
200
Cortisol
0
-3
Day of treatment
1
ng/ml
150
T
T
100
T
*
200
T
DHEAS
ng/ml
50
T
100
0
0
-3
1
3
7
10
14
Day of treatment
-3
1
Day of treatment
8
14
dria, or cytochrome c release from mitochondria could break this cycle and prevent FC-induced apoptosis. Each of the above steps has been shown to mediate FC-induced macrophage death, and their inhibition can reverse apo- ptosis (20, 23, 36, 46).
Because accumulation of excess FC in the ER mem- brane can be cytotoxic, a major func- tion of ACATs is to protect against -ACTH the unnecessary buildup of FC. Mac- +ACTH
rophages are also protected from the
accumulation of excess FC by cho- lesterol efflux (5). Although the 8 FC:CE ratio changes in response to 14 ATR-101 treatment or in ACAT1- deficient cells, we observe no differ- * ** ence in total intracellular FC and CE T content compared to control cells. 8 This suggests that adrenocortical 14 cells do not compensate by exporting FC. While it may be possible that -ACTH ATR-101 alters other aspects of ste- roidogenesis, such as StAR activity,
we found that ATR-101 had no ef- 8 14 fect on the expression of STAR, SF-1, Cyp11B1, or Cyp11B2.
We found that at high exposures, ATR-101 induced significant atro- phy, vacuolation, and apoptosis that is completely restricted to the adre- nal cortex in dogs. In accord with similar toxicities having previously been observed following administra- tion of ACAT1 inhibitors, it is likely that the observed adrenocortical toxicity is related directly to pharma-
cological inactivation of ACAT. Moreover, it is probable that adrenocortical cells, which handle large amounts of cholesterol in order to meet steroidogenic needs, are ex- tremely sensitive to ACAT1 inhibition, and that cell death results from excess FC, as determined in our accompany- ing in vitro studies. Since ACAT1 is the main ACAT en- zyme in adrenal glands (47-51), the potent selectivity of ATR-101 for ACAT1 and selective distribution to adre- nals is the most probable cause of the observed effects.
Biochemical analysis of adrenal FC and cholesterol es- ter content demonstrated an ATR-101-treatment related reduction in cholesterol ester concentrations in the adre- nals of dogs, indicative of in vivo inhibition of cholesterol esterification. Although FC levels were not found to have changed after 14 days of ATR-101 treatment, it is likely that the measured levels of FC were impacted by the pres- ence of degenerative changes in the dog adrenals. None- theless, the marked reduction of cholesterol esters in ATR- 101 treated dogs was a clear biochemical indicator of ACAT1 inhibition.
In a tissue distribution study in dogs, we found that ATR-101 achieves highest levels in the adrenals compared to several other tissues (Figure S2), which may be part of its adrenal selectivity. Given that ATR-101 is a highly li- pophilic compound (52), its transport to the adrenal cor- tex may be facilitated by its concentration in the lipopro- tein fraction and mediate its adrenocortical toxicity. The high adrenocortical lipoprotein receptor density and cen- tripetal pattern of normal adrenal blood flow may also contribute to the tissue specific effects of ATR-101.
The experiments detailed in our studies use ACTH to demonstrate deficiency of cortisol stimulation in ATR- 101 treated animals. ATR-101 treatment in vivo also caused progressive reductions in pre- and post-ACTH- stimulation steroid and steroid intermediate concentra- tions indicative of reduced adrenal gland function. These findings are believed to result first from adrenocortical cells that lack the necessary reservoir of esterified choles- terol required to produce steroid hormones following low dose ATR-101 exposure (Day 1-7), and subsequently ad- renocortical cells that have undergone cell death and are no longer capable of producing hormones (Day 14).
The fact that adrenocortical cells in dogs, following 30 mg/kg ATR-101 exposure, exhibit markers of apoptosis supports the notion that the reduced function is a reflec- tion of adrenocortical cell death. The prominent adreno- cortical vacuolation and degeneration observed in ATR- 101 treated dogs indicate ATR-101 induced cellular toxicity. Given that steroid intermediates were found to be reduced in ATR-101 treated dogs, it is also unlikely that reduced steroidogenesis reflects reduced activity or ex-
pression of terminal steroidogenic enzymes such as Cyp11B1.
While a recent report additionally observed ACAT1- independent cell death in adrenocortical cell death in re- sponse to ATR-101(53), the experiments were only con- ducted in vitro with ATR-101 at concentrations greater than 1000-fold higher than its EC50 to inhibit ACAT1 (Figure 2) and in the absence of exogenous cholesterol. We have demonstrated that treatment with ATR-101 at 1000- fold lower concentrations (30 nm) is a potent inducer of cell death, only in the presence of low, nontoxic levels of cholesterol through inhibition of endoplasmic reticulum ACAT1. These observations are consistent with reports in macrophages, where treatment with either cholesterol or an ACAT inhibitor alone is not detrimental to cell viabil- ity, whereas coincubation is extremely cytotoxic (54). Based on our studies (Figure 1), we hypothesize that the abundance of cholesterol present is one determinant of the effective concentration of ATR-101 in vivo. It is unclear whether ACAT1/SOAT1 expression is limiting in the re- sponsiveness to ATR-101. Notably, SOAT1 expression correlates with responsiveness to the adrenolytic drug mi- totane, which can also inhibit ACAT1, albeit only at high micromolar concentrations (55). Translational efforts such as the ongoing phase 1 clinical trial testing ATR-101 efficacy in adrenocortical carcinoma may ultimately clar- ify the requirement for personalized treatment based on tumoral ACAT1 expression.
In conclusion, ATR-101 induces adrenocortical cell death through its potent and selective inhibition of cho- lesterol esterification by endoplasmic reticulum ACAT1. ATR-101-induced adrenocortical cell death occurs via ap- optosis in vitro and in vivo, suggesting a common mech- anism of cell death where excess FC results in dysregula- tion of ER-calcium homeostasis and subsequent activation of apoptosis. The cytotoxic effects of ATR-101 are restricted to the adrenal cortex, and provide a unique opportunity for targeted treatment of ACC.
Acknowledgments
We thank Dr. Eric Fearon (University of Michigan) and Dr. An- tonio Lerario (University of Michigan) for critical reading of the manuscript.
Address all correspondence and requests for reprints to: Gary D. Hammer M.D., Ph.D., University of Michigan School of Med- icine. 1528 Biomedical Science Research Building. 109 Zina Pitcher Place, Ann Arbor MI, 48 109. Phone 734-615-2421. Fax 734-647-9559. E-mail: ghammer@med.umich.edu
This work was supported by This study was funded by a sponsored research agreement between Atterocor Inc. and The University of Michigan ..
Disclosure Summary: GDH is cofounder, chairman of the scientific advisory board and member of the clinical advisory board at Atterocor, Inc. SWH is an employee of Atterocor, Inc. WER is a member of the scientific advisory board at Atterocor, Inc. CRL, JEM, and VC have no conflict of interest.
References
1. Lin D, Sugawara T, Strauss JF, 3rd, Clark BJ, Stocco DM, Saenger P, Rogol A, Miller WL. Role of steroidogenic acute regulatory pro- tein in adrenal and gonadal steroidogenesis. Science 1995;267;5205: 1828-31.
2. Kraemer FB, Shen WJ, Harada K, Patel S, Osuga J, Ishibashi S, Azhar S. Hormone-sensitive lipase is required for high-density lipoprotein cholesteryl ester-supported adrenal steroidogenesis. Mol Endocri- nol 2004;18;3:549-57.
3. Liscum L, Munn NJ. Intracellular cholesterol transport. Biochim Biophys Acta 1999;1438;1:19-37.
4. Chang TY, Chang CC, Cheng D. Acyl-coenzyme A:cholesterol acyl- transferase. Annu Rev Biochem 1997 66: 613-38.
5. Tabas I. Consequences of cellular cholesterol accumulation: basic concepts and physiological implications. J Clin Invest 2002 110;7: 905-11.
6. Fazio S, Major AS, Swift LL, Gleaves LA, Accad M, Linton MF, Farese RV, Jr. Increased atherosclerosis in LDL receptor-null mice lacking ACAT1 in macrophages. J Clin Invest 2001 107;2: 163-71.
7. Warner GJ, Stoudt G, Bamberger M, Johnson WJ, Rothblat GH. Cell toxicity induced by inhibition of acyl coenzyme A:cholesterol acyltransferase and accumulation of unesterified cholesterol. J Biol Chem 1995 270;11: 5772-8.
8. Maxfield FR, Tabas I. Role of cholesterol and lipid organization in disease. Nature 2005 438;7068: 612-21.
9. Feng B, Yao PM, Li Y, Devlin CM, Zhang D, Harding HP, Sweeney M, Rong JX, Kuriakose G, Fisher EA, Marks AR, Ron D, Tabas I. The endoplasmic reticulum is the site of cholesterol-induced cyto- toxicity in macrophages. Nat Cell Biol 2003 5;9: 781-92.
10. Dominick MA, Bobrowski WA, MacDonald JR, Gough AW. Mor- phogenesis of a zone-specific adrenocortical cytotoxicity in guinea pigs administered PD 132301-2, an inhibitor of acyl-CoA:choles- terol acyltransferase. Toxicol Pathol 1993 21;1: 54-62.
11. Dominick MA, McGuire EJ, Reindel JF, Bobrowski WF, Bocan TM, Gough AW. Subacute toxicity of a novel inhibitor of acyl-CoA: cholesterol acyltransferase in beagle dogs. Fundam Appl Toxicol 1993 20;2: 217-24.
12. Wolfgang GH, MacDonald JR, Vernetti LA, Pegg DG, Robertson DG. Biochemical alterations in guinea pig adrenal cortex following administration of PD 132301-2, an inhibitor of acyl-CoA:choles- terol acyltransferase. Life Sci 1995 56;13: 1089-93.
13. Xing Y, Cohen A, Rothblat G, Sankaranarayanan S, Weibel G, Royer L, Francone OL, Rainey WE. Aldosterone production in hu- man adrenocortical cells is stimulated by high-density lipoprotein 2 (HDL2) through increased expression of aldosterone synthase (CYP11B2). Endocrinology 2011 152;3: 751-63.
14. Lada AT, Davis M, Kent C, Chapman J, Tomoda H, Omura S, Rudel LL. Identification of ACAT1- and ACAT2-specific inhibitors using a novel, cell-based fluorescence assay: individual ACAT uniqueness. J Lipid Res 2004 45;2: 378-86.
15. Uemura A, Oku M, Mori K, Yoshida H. Unconventional splicing of XBP1 mRNA occurs in the cytoplasm during the mammalian un- folded protein response. J Cell Sci 2009 122;Pt 16: 2877-86.
16. Carr TP, Andresen CJ, Rudel LL. Enzymatic determination of tri- glyceride, free cholesterol, and total cholesterol in tissue lipid ex- tracts. Clin Biochem 1993 26;1: 39-42.
17. Kellner-Weibel G, Geng YJ, Rothblat GH. Cytotoxic cholesterol is generated by the hydrolysis of cytoplasmic cholesteryl ester and transported to the plasma membrane. Atherosclerosis 1999 146;2: 309-19.
18. Underwood KW, Andemariam B, McWilliams GL, Liscum L. Quantitative analysis of hydrophobic amine inhibition of intracel- lular cholesterol transport. J Lipid Res 1996 37;7: 1556-68.
19. Li Y, Ge M, Ciani L, Kuriakose G, Westover EJ, Dura M, Covey DF, Freed JH, Maxfield FR, Lytton J, Tabas I. Enrichment of endoplas- mic reticulum with cholesterol inhibits sarcoplasmic-endoplasmic reticulum calcium ATPase-2b activity in parallel with increased or- der of membrane lipids: implications for depletion of endoplasmic reticulum calcium stores and apoptosis in cholesterol-loaded mac- rophages. J Biol Chem 2004 279;35: 37030-9.
20. Timmins JM, Ozcan L, Seimon TA, Li G, Malagelada C, Backs J, Backs T, Bassel-Duby R, Olson EN, Anderson ME, Tabas I. Cal- cium/calmodulin-dependent protein kinase II links ER stress with Fas and mitochondrial apoptosis pathways. J Clin Invest 2009 119; 10: 2925-41.
21. Narasimhan K, Pessah IN, Linden DJ. Inositol-1,4,5-trisphosphate receptor-mediated Ca mobilization is not required for cerebellar long-term depression in reduced preparations. J Neurophysiol 1998 80;6: 2963-74.
22. Maruyama T, Kanaji T, Nakade S, Kanno T, Mikoshiba K. 2APB, 2-aminoethoxydiphenyl borate, a membrane-penetrable modulator of Ins(1,4,5)P3-induced Ca2+ release. J Biochem 1997 122;3: 498- 505.
23. Yao PM, Tabas I. Free cholesterol loading of macrophages is asso- ciated with widespread mitochondrial dysfunction and activation of the mitochondrial apoptosis pathway. J Biol Chem 2001 276;45: 42468-76.
24. Bae JH, Park JW, Kwon TK. Ruthenium red, inhibitor of mitochon- drial Ca2+ uniporter, inhibits curcumin-induced apoptosis via the prevention of intracellular Ca2+ depletion and cytochrome c re- lease. Biochem Biophys Res Commun 2003 303;4: 1073-9.
25. Walter DH, Haendeler J, Galle J, Zeiher AM, Dimmeler S. Cyclo- sporin A inhibits apoptosis of human endothelial cells by preventing release of cytochrome C from mitochondria. Circulation 1998 98; 12: 1153-7.
26. Travers KJ, Patil CK, Wodicka L, Lockhart DJ, Weissman JS, Walter P. Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degrada- tion. Cell 2000 101;3: 249-58.
27. East JM. Sarco(endo)plasmic reticulum calcium pumps: recent ad- vances in our understanding of structure/function and biology (re- view). Mol Membr Biol 2000 17;4: 189-200.
28. Michalak M, Robert Parker JM, Opas M. Ca2+ signaling and cal- cium binding chaperones of the endoplasmic reticulum. Cell Cal- cium 2002 32;5-6: 269-78.
29. Luciani DS, Gwiazda KS, Yang TL, Kalynyak TB, Bychkivska Y, Frey MH, Jeffrey KD, Sampaio AV, Underhill TM, Johnson JD. Roles of IP3R and RyR Ca2+ channels in endoplasmic reticulum stress and beta-cell death. Diabetes 2009 58;2: 422-32.
30. Liscum L, Faust JR. The intracellular transport of low density lipo- protein-derived cholesterol is inhibited in Chinese hamster ovary cells cultured with 3-beta-[2-(diethylamino)ethoxy]androst-5-en- 17-one. J Biol Chem 1989 264;20: 11796-806.
31. Harmala AS, Porn MI, Mattjus P, Slotte JP. Cholesterol transport from plasma membranes to intracellular membranes is inhibited by 3 beta-[2-(diethylamino)ethoxy]androst-5-en-17-one. Biochim Biophys Acta 1994 1211;3: 317-25.
32. Lange Y. Cholesterol movement from plasma membrane to rough endoplasmic reticulum. Inhibition by progesterone. J Biol Chem 1994 269;5: 3411-4.
33. Yang H, Bard M, Bruner DA, Gleeson A, Deckelbaum RJ, Aljinovic G, Pohl TM, Rothstein R, Sturley SL. Sterol esterification in yeast: a two-gene process. Science 1996 272;5266: 1353-6.
34. Yoshida H, Matsui T, Yamamoto A, Okada T, Mori K. XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 2001 107;7: 881-91.
35. Calfon M, Zeng H, Urano F, Till JH, Hubbard SR, Harding HP, Clark SG, Ron D. IRE1 couples endoplasmic reticulum load to se- cretory capacity by processing the XBP-1 mRNA. Nature 2002 415; 6867: 92-6.
36. Li G, Mongillo M, Chin KT, Harding H, Ron D, Marks AR, Tabas I. Role of ERO1-alpha-mediated stimulation of inositol 1,4,5- triphosphate receptor activity in endoplasmic reticulum stress-in- duced apoptosis. J Cell Biol 2009 186;6: 783-92.
37. Kirichok Y, Krapivinsky G, Clapham DE. The mitochondrial cal- cium uniporter is a highly selective ion channel. Nature 2004 427; 6972: 360-4.
38. Rizzuto R, Pinton P, Ferrari D, Chami M, Szabadkai G, Magalhaes PJ, Di Virgilio F, Pozzan T. Calcium and apoptosis: facts and hy- potheses. Oncogene 2003 22;53: 8619-27.
39. Pritchard DE, Singh J, Carlisle DL, Patierno SR. Cyclosporin A in- hibits chromium(VI)-induced apoptosis and mitochondrial cyto- chrome c release and restores clonogenic survival in CHO cells. Carcinogenesis 2000 21;11: 2027-33.
40. Friberg H, Ferrand-Drake M, Bengtsson F, Halestrap AP, Wieloch T. Cyclosporin A, but not FK 506, protects mitochondria and neu- rons against hypoglycemic damage and implicates the mitochon- drial permeability transition in cell death. J Neurosci 1998 18;14: 5151-9.
41. Deniaud A, Sharaf el dein O, Maillier E, Poncet D, Kroemer G, Lemaire C, Brenner C. Endoplasmic reticulum stress induces calci- um-dependent permeability transition, mitochondrial outer mem- brane permeabilization and apoptosis. Oncogene 2008 27;3: 285- 99.
42. Graham JM, Green C. The properties of mitochondria enriched in vitro with cholesterol. Eur J Biochem 1970 12;1: 58-66.
43. Lakhani SA, Masud A, Kuida K, Porter GA, Jr., Booth CJ, Mehal WZ, Inayat I, Flavell RA. Caspases 3 and 7: key mediators of mi- tochondrial events of apoptosis. Science 2006 311;5762: 847-51.
44. Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 1997 91;4: 479-89.
45. Boehning D, Patterson RL, Sedaghat L, Glebova NO, Kurosaki T, Snyder SH. Cytochrome c binds to inositol (1,4,5) trisphosphate receptors, amplifying calcium-dependent apoptosis. Nat Cell Biol 2003 5;12: 1051-61.
46. Lim WS, Timmins JM, Seimon TA, Sadler A, Kolodgie FD, Virmani R, Tabas I. Signal transducer and activator of transcription-1 is critical for apoptosis in macrophages subjected to endoplasmic re- ticulum stress in vitro and in advanced atherosclerotic lesions in vivo. Circulation 2008 117;7: 940-51.
47. Lee RG, Willingham MC, Davis MA, Skinner KA, Rudel LL. Dif- ferential expression of ACAT1 and ACAT2 among cells within liver, intestine, kidney, and adrenal of nonhuman primates. J Lipid Res 2000 41;12: 1991-2001.
48. Meiner V, Tam C, Gunn MD, Dong LM, Weisgraber KH, Novak S, Myers HM, Erickson SK, Farese RV, Jr. Tissue expression studies on the mouse acyl-CoA: cholesterol acyltransferase gene (Acact): find- ings supporting the existence of multiple cholesterol esterification enzymes in mice. J Lipid Res 1997 38;9: 1928-33.
49. Leon C, Hill JS, Wasan KM. Potential role of acyl-coenzyme A:cho- lesterol transferase (ACAT) Inhibitors as hypolipidemic and anti- atherosclerosis drugs. Pharm Res 2005 22;10: 1578-88.
50. Lee O, Chang CC, Lee W, Chang TY. Immunodepletion experi- ments suggest that acyl-coenzyme A:cholesterol acyltransferase-1 (ACAT-1) protein plays a major catalytic role in adult human liver, adrenal gland, macrophages, and kidney, but not in intestines. J Lipid Res 1998 39;8: 1722-7.
51. Cases S, Novak S, Zheng YW, Myers HM, Lear SR, Sande E, Welch CB, Lusis AJ, Spencer TA, Krause BR, Erickson SK, Farese RV, Jr. ACAT-2, a second mammalian acyl-CoA:cholesterol acyltrans- ferase. Its cloning, expression, and characterization. J Biol Chem 1998 273;41: 26755-64.
52. Sliskovic DR, Krause BR, Picard JA, Anderson M, Bousley RF, Hamelehle KL, Homan R, Julian TN, Rashidbaigi ZA, Stanfield RL. Inhibitors of acyl-CoA: cholesterol O-acyl transferase (ACAT) as hypocholesterolemic agents. 6. The first water-soluble ACAT in- hibitor with lipid-regulating activity. J Med Chem 1994 37;5: 560-2.
53. Cheng Y, Kerppola RE, Kerppola TK. ATR-101 disrupts mitochon- drial functions in adrenocortical carcinoma cells and in vivo. Endocr Relat Cancer 2016.
54. Tabas I, Marathe S, Keesler GA, Beatini N, Shiratori Y. Evidence that the initial up-regulation of phosphatidylcholine biosynthesis in free cholesterol-loaded macrophages is an adaptive response that prevents cholesterol-induced cellular necrosis. Proposed role of an eventual failure of this response in foam cell necrosis in advanced atherosclerosis. J Biol Chem 1996 271;37: 22773-81.
55. Sbiera S, Leich E, Liebisch G, Sbiera I, Schirbel A, Wiemer L, Mat- ysik S, Eckhardt C, Gardill F, Gehl A, Kendl S, Weigand I, Bala M, Ronchi CL, Deutschbein T, Schmitz G, Rosenwald A, Allolio B, Fassnacht M, Kroiss M. Mitotane Inhibits Sterol-O-Acyl Trans- ferase 1 Triggering Lipid-Mediated Endoplasmic Reticulum Stress and Apoptosis in Adrenocortical Carcinoma Cells. Endocrinology 2015 156;11: 3895-908.