For in vitro experiments, NCI-H295R (hereon referred to as H295R) human ACC cells (from passage 7 to 12) obtained from Gustave Roussy, Universite Paris Sud, Villejuif, France and used in our previous studies (2-4,11), were cultured as previously described (3). The H295R cells were cultured in DMEM/HAM’S F-12 (PAA, Les Mureaux, France) supplemented with 20 mM HEPES (Invitrogen, Life Technologies/Thermo Fisher Scientific, Waltham, MA, USA), antibiotics (penicillin 100 IU/ml and streptomycin 100 µg/ml) and 2 mM glutamine. The medium for H295R cell culture was enriched with 10% fetal bovine serum and a mixture of insulin/transferrin/selenium. The cells were cultured at 37°C in a humidified incubator with 5% CO₂. o,p′-DDD (Sigma-Aldrich, St. Louis, MO, USA), and rosuvastatin (Sigma-Aldrich) were solubilized in dimethyl sulfoxide (DMSO; Sigma-Aldrich) and used at the indicated concentrations ranging from 0 to 100 µM. In all experiments, the percentage of DMSO in the culture medium never exceeded 0.1% v/v. Given that o,p′-DDD induces hepatic CYP3A4 activity (12), we selected rosuvastatin for use in our experiments, a statin not metabolized by CYP3A4.

Cell viability assays were performed using WST1 assay (Roche, Basel, Switzerland) and apoptosis tests were performed using the Caspase-Glo 3/7 assay (Promega, Madison, WI, USA) according to the manufacturer’s recommendations. The cells were cultured in 96-well plates and treated with 0-100 µM mitotane alone or with rosuvastatin for various periods of time (0 to 72 h). The number of cells per well was 3 to 10×10³. Optical densities were measured 4 h after the addition of WST1 solution (10 µl per well) by spectrophotometry at 450 nm (Viktor multilabel plate reader; PerkinElmer, Waltham, MA, USA). The results were validated by cell counting with the cell counter method (TC20 automated cell counter; Bio-Rad Laboratories, Hercules, CA, USA). Luminescence was measured 1 h after the addition of Caspase-Glo 3/7 solution (equal volume) by luminometry (Viktor multilabel plate reader; PerkinElmer).

Total protein extracts were prepared and western blot analyses were performed as previously described (11). Total protein extracts were prepared from cells lysed in lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA, 30 mM Na pyrophosphate, 50 mM Na fluoride and 1% Triton X-100) and 1X protease inhibitor (Sigma-Aldrich), 40 µg proteins were loaded by lane. Following protein blotting on an Odyssey nitrocellulose membrane (LI-COR, Lincoln, NE, USA), the blots were incubated for 1 h at room temperature in a blocking buffer [5% fat-free milk in phosphate-buffered saline (PBS) with 0.1% Tween-20] before an overnight incubation at 4°C. Primary antibodies were a rabbit polyclonal anti-poly(ADP-ribose) polymerase (PARP) antibody (dilution, 1:200; #9542; Cell Signaling Technology, Danvers, MA, USA). The antibody detected total PARP (116 kDa) and cleaved PARP (89 kDa), the ratio of which reflects the pro-apoptotic status. The normalizing antibody was the anti-α-tubulin (dilution 1:1,000; AB_10013740; Sigma-Aldrich). Secondary antibodies were goat anti-mouse IgG (H+L) cross adsorbed secondary antibody (DyLight 680 conjugated, AB_614942) and goat anti-rabbit IgG (H+L) DyLight 800 conjugated (dilution 1:10,000, AB_614947) (both from Thermo Fisher Scientific).

Antibodies were diluted in PBS 0.1% Tween-20 buffer 5% non-fat milk and added to the membranes for 1 h at room temperature or overnight at 4°C, followed by incubation with the indicated secondary antibody for 1 h at room temperature. Target proteins were detected using Odyssey Fc, Dual-Mode Western Imaging (LI-COR) by fluorescence (680 nm wavelength for anti-mouse antibody and 800 nm for anti-rabbit antibody) and quantified using the Odyssey Fc Dual-mode Western Imaging apparatus from LI-COR as indicated.

The concentrations of free intracellular cholesterol were measured using the Cholesterol Quantification kit (Sigma-Aldrich) according to the manufacturer’s recommendations. The cells were cultured in F12 plates (4 wells per condition) treated with mitotane and/or rosuvastatin for 24 h. In these experiments, we used serum-free culture media to exclude exogenous cholesterol uptake. The absorbance was read at 570 nm 1 h after the addition of 2 µl of cholesterol probe and 2 µl of cholesterol enzyme mixture per well. Cholesterol concentrations were calculated according to the established standard curve and normalized to the initial cell count.

The measurement of HMGCR was carried out using HMGCoA Reductase Assay (Sigma-Aldrich), according to the manufacturer’s recommendations. Incubation medium included HMGCoA (enzyme substrate), NADPH (reduced nicotinamide adenine dinucleotide phosphate), buffer solution and HMG-CoA reductase (provided in the kit). Specific absorbance at 340 nm was compared in the presence or in the absence of pravastatin, a potent HMGCR activity inhibitor. HMGCR activity was determined by the difference in absorbance slope between these two conditions.

Total RNA was extracted from the H295R cells with the RNeasy kit (Qiagen) according to the manufacturer’s recommendations. A total of 1 µg total RNAwere subjected to DNase I treatment (Invitrogen/Thermo Fisher Scientific) and reverse-transcribed with 200 units of reverse transcriptase (Superscript II, Invitrogen/Thermo Fisher Scientific). PCR was performed with 100 ng cDNA in the presence of qPCR™ Mastermix Plus for Sybr™-Green I (Eurogentec, Seraing, Belgium) containing 300 nM of specific primers (Table SI). qPCR was carried on an ABI Step One Plus (Applied Biosystems, Foster City, CA, USA) whose parameters were as follows: A pre-cycle at 95°C for 20 sec then 40 cycles at 95°C for 1 sec followed by 40 cycles at 60°C for 20 sec. The amount of cytochrome c oxidase subunit II COX2 transcript in the samples was determined by comparison with the standard range and related to the amount of the 18S gene of nuclear origin. For standards preparation, amplicons were subcloned into pGEMT-easy plasmid (Promega) and sequenced to confirm the identity of each sequence. Standard curves were generated using serial dilutions of linearized standard plasmids. Samples were amplified in duplicate or triplicate. Ribosomal 18S was used as an internal control for data normalization. qPCR was performed using the Fast SYBR Green Master Mix (Life Technologies/Thermo Fisher Scientific) and carried out on a QuantStudio 6 Flex (Life Technologies/Thermo Fisher Scientific). The relative expression of each gene was expressed as the ratio of attomoles of specific gene to attomoles of 36B4 mRNA or femtomoles of 18S rRNA.

Steroid measurements (progesterone, 17OHP) were assayed in the cell supernatants under various conditions after 48-h treatment, by means of liquid chromatography (LC)-mass spectrometry (MS)/MS analysis. LC-MS/MS was performed using a Waters Xevo TQS triple-quadrupole mass spectrometer connected to a Waters Acquity UPLC H-class (Waters SAS, Saint Quentin Yvelines, France). Chromatographic separation was performed on a BEH C18 column (1.7 µm, 100×2.1) at a flow rate of 0.3 ml/min at 40°C. The mobile phase consisting of methanol and 5 mmol/l ammonium formate in water was delivered according to the following gradient: 35% methanol from 0 to 1.5 min, linear increase to 45% methanol (1.5-3 min), then to 63% methanol (3-7 min) followed by 100% methanol (7-9 min). Following column washing with 100% methanol (9-11.5 min), the gradient was reversed to reach initial conditions at 14 min. The injection volume was 10 µl and the sample manager was maintained at 10°C. Detection was performed on a Xevo TQS tandem mass spectrometer (Waters, Paris, France). Instrument optimization for the analytes was conducted by infusing standard solution (100 pg/ml) of the analytes by the built-in syringe pump at a flow rate of 10 µl/min. The following optimized operating conditions were used for the multiple reaction monitoring mode: Capillary voltage, 3.5 kV; cone voltage, 4-60 V; collision energy, 15-34 eV; dwell time, 0.03-0.1 sec, depending on the steroid. The mass spectrometer parameters were configured as follows: Desolvation temperature, 500°C; desolvation nitrogen flow, 790 l/h; source temperature, 150°C; cone nitrogen flow, 145 l/h. Argon was used as collision gas with a flow rate of 0.14 ml/min. Two mass transitions were monitored for each steroid. System control and data acquisition were achieved with the MassLynx 4.0 software (Waters). Cells were cultured in F6 plates (3 wells per condition) treated with 25 µM mitotane (M25) and/or 50 µM rosuvastatin (R50) for 48 h.

Results are expressed as the means ± SEM of n independent replicates performed in the same experiment or from separate experiments (n). The non-parametric Mann-Whitney U test was used when appropriate and differences between groups were analyzed using non-parametric Kruskall-Wallis multiple comparison tests followed by a post hoc Dunn’s test (Prism software, GraphPad, CA, USA). A P-value of 0.05 was considered to indicate a statistically significant difference.

First, we examined the effects of mitotane or rosuvastatin alone on cell viability at the concentration to 100 µM and at different time periods (up to 48 h). Mitotane (50 µM; M50) (Fig. 1A) reduced the absorbance in a time-dependent manner, confirming its anti-proliferative effect. Rosuvastatin alone also induced a time-dependent inhibition of cell viability, providing support for a specific effect of rosuvastatin. This effect, however, was only observed at high concentrations starting from 50 µM (R50) (Fig. S1). The combination of mitotane (50 µM; M50) and rosuvastatin (100 µM; R100) potentiated the inhibition of cell viability at 48 h (Fig. 1A), while rosuvastatin (100 µM) had no additive effect on the anti-proliferative effects of mitotane at 72 h (Fig. S2).

Using similar experimental conditions, we then analyzed the index of H295R cell apoptosis using caspase-3/7 activity (Fig. 1B) and cleaved PARP expression (Fig. 1C and D). As expected, mitotane (50 µM; M50) alone induced a significant increase in caspase-3/7 activity, while rosuvastatin (100 µM; R100) alone had no effect. However, the combination of mitotane and rosuvastatin led to an over-induction of caspase-3/7 activity. Moreover, similar potentiation effects between mitotane and rosuvastatin were observed when examining the expression of cleaved PARP (Fig. 1C and D), confirming the induction of H295R cell apoptosis when both molecules were used in combination (M50 and R50). As the effects of rosuvastatin were observed at 50 µM, this concentration was considered as the most relevant for use in our experiments.

Since statins and mitotane alter intracellular lipid metabolism (9,13), in this study, we examined the effects of mitotane, rosuvastatin and their association on the level of intracellular free cholesterol, the precursor of steroidogenesis. As shown in Fig. 2, we confirmed that mitotane (50 µM; M50) for 24 h significantly increased the concentration of intracellular free cholesterol, whereas rosuvastatin (50 µM; R50) alone had no effect. However, no potentiating effect on the free intracellular cholesterol concentration was observed when the cells were exposed to both mitotane and rosuvastatin.

Since mitotane increases the intracellular free cholesterol concentrations, we then sought to examine the hypothesis that mitotane may directly increase the activity of HMGCR and may thereby stimulate the mevalonate pathway.

HMGCR activity was measured at 1,815 U/mg protein under control conditions and at 1,876 U/mg following the addition of up to 100 µM mitotane, indicating that mitotane exerts no direct effect on the activity of HMGCR at least in vitro. The lack of a direct effect of mitotane in vitro associated with the difficulties to accurately evaluate HMGCR activity (data not shown) in the cells did not prompt us to examine the activity in H295R cells under various experimental settings.

We then examined the expression of several genes involved in cholesterol metabolism, including the HMGCR gene, encoding a key player in the intracellular free cholesterol balance, and the ABCA1 gene, encoding a protein that allows cholesterol efflux from the cell.

Rosuvastatin (50 µM; R50) alone did not exert any effect, but acted in combination with mitotane at 50 µM to significantly reduce HMGCR expression (Fig. 3A). Rosuvastatin (50 µM; R50) significantly reduced ABCA1 gene expression, alone or in combination with mitotane 25 and 50 µM (Fig. 3B). However, rosuvastatin (50 µM) did not alter LDLR or SrB1 gene expression, regardless of the duration of treatment, while mitotane inhibited the expression of these genes (data not shown).

We then examined the expression of genes involved in steroidogenesis, including StAR, encoding the cholesterol transporter facilitating transfer to the mitochondria, and CYP11A1, encoding the first limiting step of steroid synthesis catalyzing cholesterol to pregnenolone (Fig. 3C and D). The expression of StAR was significantly reduced by mitotane in a concentration- (25 and 50 µM; M25 and M50). Rosuvastatin (50 µM; R50) did not exert any significant effect when used alone, but slightly prevented the mitotane-induced reduction in StAR expression. With respect to CYP11A1 expression, mitotane alone significantly reduced its expression (25 and 50 µM; M25 and M50) (Fig. 3D). Rosuvastatin (50 µM; R50) alone significantly reduced the expression of CYP11A1.

We then evaluated the steroid-secreting capacities of the H295R cells (Fig. 4A). As previously demonstrated (2,11), mitotane (25 µM; M25) alone decreased the concentration of cortisol and corticosterone in the supernatants of H295R cells following 48 h of treatment (Fig. S3). Rosuvastatin (50 µM; R50) alone decreased the progesterone and 17OHP concentrations. The combination of mitotane 25 µM and rosu-vastatin 50 µM exhibited a significant potentiation effect in inhibiting steroidogenesis, with progesterone secretion reduced by 47% (Fig. 4A) and that of 17OHP reduced by 37% (Fig. 4B).

A visual summary of the mechanisms of action of mitotane and rosuvastatin in the H295R cells described in this study is presented in Fig. 5.