Liver X receptor inhibition potentiates mitotane-induced adrenotoxicity in ACC

Adrenocortical carcinoma (ACC) is a rare aggressive malignancy with a poor outcome largely due to limited treatment options. Here, we propose a novel therapeutic approach through modulating intracellular free cholesterol via the liver X receptor alpha (LXR α ) in combination with current first-line pharmacotherapy, mitotane. H295R and MUC-1 ACC cell lines were pretreated with LXR α inhibitors in combination with mitotane. In H295R, mitotane (20, 40 and 50 µM) induced dose-dependent cell death; however, in MUC-1, this only occurred at a supratherapeutic concentration (200 µM). LXR α inhibition potentiated mitotane-induced cytotoxicity in both cell lines. This was confirmed through use of the CompuSyn model which showed moderate pharmacological synergism and was indicative of apoptotic cell death via an increase in annexinV and cleaved-caspase 3 expression. Inhibition of LXR α was confirmed through downregulation of cholesterol efflux pumps ABCA1 and ABCG1 ; however, combination treatment with mitotane attenuated this effect. Intracellular free-cholesterol levels were associated with increased cytotoxicity in H295R ( r 2 = 0.5210) and MUC-1 ( r 2 = 0.9299) cells. While both cell lines exhibited similar levels of free cholesterol at baseline, H295R were cholesterol ester rich, whereas MUC-1 were cholesterol ester poor. We highlight the importance of LXR α mediated cholesterol metabolism in the management of ACC, drawing attention to its role in the therapeutics of mitotane sensitive tumours. We also demonstrate significant differences in cholesterol storage between mitotane sensitive and resistant disease. an LXR α mediated of enhanced mitotane-induced cytotoxicity. validated these results by stimulating LXR α activity using the ligand 27HC reduced mitotane-induced cytotoxicity


Introduction
Adrenocortical carcinoma (ACC) is a rare, aggressive malignancy with an incidence of 2-5 per million and which carries a poor prognosis due to local invasion or distant metastases at the time of diagnosis in the majority of cases (Fassnacht et al. 2013, Libe 2019. Complete surgical resection (R0) is the only curative therapy (de Reynies et al. 2009, Kerkhofs et al. 2013). Yet, for those on whom R0 resection is achieved, disease recurs in 50-85% (Pommier & Brennan 1992, Terzolo et al. 2007. Mitotane is the only currently licensed pharmacotherapy for adjuvant use or for recurrent disease and remains the most effective drug (alone or in combination with platinum based chemotherapy regimens) in preventing and treating recurrence (Haak et al. 1994, Terzolo et al. 2007, Fassnacht et al. 2018). Mitotane's use is limited by its narrow therapeutic window, whereby at serum concentrations below 14 mg/L, it demonstrates poor efficacy, while at concentrations >20 mg/L, it is associated with unacceptable toxic effects (Hermsen et al. 2011. There is significant clinical need to improve the therapeutic options in managing persistent or recurrent ACC, and in this context, the current study investigates therapeutic strategies to enhance the efficacy and tolerability of mitotane. Mitotane has been used for 50 years in the management of adrenocortical neoplasm (Montgomery & Welbourn 1965), yet its adrenotoxic mechanism remains incompletely understood (Hescot et al. 2013, Lehmann et al. 2013, Poli et al. 2013. Sbiera et al. demonstrated that mitotane-induced adrenotoxicity is associated with intracellular free-cholesterol (FC) accumulation causing endoplasmic reticulum (ER) stress (Sbiera et al. 2015). The authors also demonstrated that mitotane-induced FC accumulation is associated with inhibition of sterol-O-acyltransferase-1 (SOAT1) which usually converts FC to inert cholesteryl esters (CE) for intracellular storage. The adrenal-cytotoxic effects of SOAT1 inhibition have also been demonstrated by LaPensee et al. through the use of the selective inhibitor ATR-101 (LaPensee et al. 2016). Consequently, this lipotoxic mechanism mediated by intracellular FC accumulation is considered significant or even predominant in terms of mitotane-induced adrenotoxicity.
We theorised that the lipotoxic effects of mitotane on ACC cells could be enhanced by targeting complementary mechanisms to that of SOAT1 inhibition, which would further increase intracellular FC. Liver X receptor alpha (LXRα/NR1H3) inhibition represents one such mechanism. LXRα is a nuclear receptor, sensitive to intracellular FC and represents the predominant transcriptional mechanism mediating reverse cholesterol transportation to the liver (Costet et al. 2000, Kennedy et al. 2005). At rest, LXRα is bound to its nuclear response element and held in an inactive state by co-repressor molecules. Upon ligand activation, it recruits co-activator proteins to dimerise with the retinoid X receptor (RXR), and in turn transcribes regulatory proteins for cholesterol efflux, namely ABCA1 and ABCG1. These deliver excessive intracellular FC to circulating apolipoprotein (Apo) A1 or high density lipoprotein (HDL 1 ) (Janowski et al. 1996, Fu et al. 2001, Hu et al. 2003, Huuskonen et al. 2004, Phelan et al. 2008. LXRα is highly expressed in liver, innate immune cells and adrenal cortex (Cummins et al. 2006) and its sensitivity to intracellular cholesterol is effected through endogenous ligand molecules, oxysterols, generated from precursor cholesterol. The common oxysterol, 27-hydroxycholesterol (27HC, also known as (25R)26-HC), is catalysed by CYP27A1 and highly expressed in the adrenal cortex (Chen et al. 1998, Fedorova et al. 2015. Higher intracellular FC results in increased oxysterol production, activating LXRα dependent transcription of cholesterol efflux pumps. The cholesterol regulatory function of LXRα is conserved across cell types, mediating reverse cholesterol transport as well as providing cellular protection against toxic FC accumulation (Cignarella et al. 2005, Engel et al. 2007).
In the adrenal, oxysterols and LXRα also have a tissue-specific role, whereby they increase steroidogenesis by upregulating steroidogenic acute regulatory (StAR) protein (Cummins et al. 2006, Nilsson et al. 2007. In breast and prostate cancer, LXRα stimulation facilitates cancer cell survival which is partially explained by its role to enhance aerobic glycolysis, while also hypothesised to reduce FC-induced cytotoxicity (Nelson et al. 2013, Flaveny et al. 2015, Kim et al. 2018. We hypothesised that inhibition of LXRα-modulated cholesterol efflux pump expression could potentiate mitotane-induced lipotoxic adrenocortical cell death at usually subtherapeutic mitotane concentrations. We investigated the dose-dependent interactions of mitotane with pharmacological inhibition of LXRα and dominantnegative (DN) interference of LXRα activity in two human ACC cell lines (H295R and MUC-1). Using this strategy, we demonstrate an enhanced ability of mitotane in combination with LXRα inhibition to kill ACC cells at lower concentrations than for mitotane alone. We show that this effect is associated with higher intracellular FC accumulation.

Plasmids, transfection and luciferase assay
H295R were transfected using Lipofectamine 3000 (Thermo Fisher) using either pCMX-hLXRα dominant-negative (LXR-DN) or WT (LXR-WT) expressing plasmids (Willy et al. 1995). Transfection efficiency was evaluated using green fluorescent protein co-expression (pmaxGFP, Lonza) using microscopy and flow cytometry (Supplementary Fig. 1I and J, see section on supplementary materials given at the end of this article). H295R were transfected in antibiotic free media for 24 h prior to drug treatments. Luciferase assay was carried out using firefly (pGL3-TK-LXRRE-Luc) and renilla luciferase vectors (10:1). Activity were measured following 24-h stimulation with LXRα agonist T0901317 (1 µM) using the Dual-Glo Luciferase Assay System (Promega) according to the manufacturer's guidelines. Luminescence was recorded on the Synergy HT Multi-Detection Microplate Reader (BioTek), normalised for renilla luciferase activity and represented as relative light units (RLU).

Gene expression
Following drug treatments, cells were washed with cold phosphate buffered saline (PBS) and total RNA was extracted from adherent cells using TRI Reagent®. Following spectrophotometric quantification, 1 μg of total RNA was reverse transcribed into cDNA using the Qiagen RT 2 first strand kit. PCR assay was carried out using Go Taq® Green Master Mix (Promega), according to manufacturer's guidelines. Amplified products were separated using a 4% agarose gel, visualized using SYBR Safe (Thermo Fisher) stain and imaged using the ChemiDoc™ XRS+ Imaging system (Bio-Rad Laboratories). Real-time quantitative PCR expression was assayed on Applied Biosystems StepOne Plus (Applied Biosystems) and carried out using SYBR green mastermix (Promega). Primer sequences are listed in Supplementary Table 1. Fold change was calculated using the ΔΔCT method. Data is represented as fold change of mRNA relative to vehicle control.

Western blotting
Samples were harvested on ice and lysed using radioimmuno-precipitation (RIPA) buffer and 1X protease inhibitor cocktail (Calbiochem). Lysates were quantified by BCA assay, denatured at 95°C and separated by SDS-PAGE using a 10% Tris-glycine gel. Proteins were transferred to PVDF membrane, blocked using 5% milk and incubated with relevant antibodies. Membranes were incubated with primary antibodies overnight at 4°C followed by relevant HRP-linked secondary antibodies (Supplementary Table  2). Imaging was carried out using the ChemiDoc™ XRS+ Imaging system (Bio-Rad Laboratories) with SuperSignal™ HRP substrate (Thermo Fisher) and Image Lab™ Software (Bio-rad, V5.2.1). Semi-quantitative densitometry was carried out using ImageJ and normalised to the loading control (β Actin).

Metabolic activity assay
MTT assay was carried out in 96 well plates, cells were seeded at 1 × 10 3 cells per well and grown for 3, 7, 10 and 14 days. Drug treatments GSK2033 (5 μM) and SR9243 (1 μM) were given at the indicated time points (day 0, 3, 7 and 10) for 24 h followed by mitotane (50 μM) for 24 h. Following incubation, MTT reagent (0.5 mg/mL) was added. The assay was terminated by solubilising MTT using DMSO and absorbance was read at 570 nm as described previously. Experimental conditions were carried out in replicates of six and the experiment was repeated four times. Data is represented as % metabolic activity adjusted to matched vehicle control (100%) for each individual time point.

Cholesterol analysis
Filipin complex III (Sigma) from Streptomyces filipinensis was used to assess free cholesterol (Hassall & Graham 1995). Filipin is a fluorescent marker of FC that is excited by a 405-nm laser followed by a 450/50 bandpass filter. Cells were treated as indicated, harvested via trypsinisation and fixed using fixation buffer (BioLegend, San Diego, CA, USA) according to manufacturer's guidelines. Staining was carried out at 1 mg/mL for 1 h at 4°C in darkness and analysed by microscopy and flow cytometry. CholEsteryl BODIPY™ FL C12 (cholesteryl 4,4-difluoro-5,7dimethyl-4-bora-3a,4a-diaza-s-indacene-3-dodecanoate) (Invitrogen) was used to label cholesterol esters (Hsieh et al. 2012). Cells were grown to 60% confluence and media was supplemented with 5 µM CholEsteryl BODIPY™ FL C12 overnight. Cells were harvested via trypsinisation and analysed by flow cytometry on the BD FACSCantoII (BD Bioscience). Sytox blue (Invitrogen) (1 µM) was used as a viability dye. Fluorescence was recorded using a 488-nm excitation laser followed by a 530/30 bandpass filter (CholEsteryl BODIPY™ FL C12) and a 405-nm laser followed by 450/50 bandpass filter (Sytox Blue). Single, live cells were gated according to Supplementary Fig. 6 and the resolution metric was calculated as described (Brando & Sommaruga 1993). Additionally, cells were analysed via imaging flow cytometry using the Amnis® ImageStream® x Mk II (Luminex Corporation, Seattle, WA, USA). High sensitivity mode (60×) was used to record 3500 events per sample. Analysis was carried out using IDEAS® software (V6.2) and, briefly, focused cells were gated according to Gradient RMS_M01_Ch01 (<50). Single cells were gated using Area_M01 vs Aspect Ratio_M01 (size vs circularity). Live cells were gated by using Sytox Blue (Invitrogen) viability dye according to Intensity_MC_ Ch07. The population of live, single, focused cells was used for analysis of CholEsteryl BODIPY™ FL C12 according to Intensity_MC_Ch02. Image display properties were standardised as follows: X Range (37, 563), Midpoint (300, 127) and X-Axis Scale (32, 568).

Data analysis
Drug synergism was evaluated using CompuSyn (download from www.compusyn.com) according to the Chou-Talalay method (Chou 2008). Data were input as single agent and combined (non-constant ratio) agents as mean ±s.e.m. Pharmacological synergism was interpreted according to the combination index (CI) model. CI plots and data tables are available in Supplementary Fig. 1K, L and M. Linear regression and statistical analyses were performed using GraphPad® Prism V 8.0 (GraphPad Software). Data is represented as mean ± s.e.m. unless stated otherwise. Paired sample analyses were performed using a two-sided Student's t-test. Multiple-group comparisons were carried out using an ANOVA followed by Tukey's post-hoc test. Statistical significance for two-tailed analyses (P-value) was assigned for values <0.05.
We validated these findings of pharmacological LXRα inhibition by transfecting H295R cells with a LXRα dominant-negative construct (LXRα-DN, transfection efficiency: 40%) that blocks LXRα-WT functioning ( Supplementary Fig. 1G, H, I and J). LXRα-DN-expressing cells reached significantly higher rates of mitotaneinduced cytotoxicity for all mitotane concentrations, compared to LXRα-WT (11% (WT:mitotane 20 μM) vs Viability data represented as percentage dead cells relative to vehicle control. Metabolic activity data represented as percentage cells relative to vehicle control (adjusted to 100% at each respective time point). All data shown as mean ± s.e.m., n = 4. Statistical significance is denoted as *P < 0.05, **P < 0.01, ***P < 0.01 and ****P < 0.001 relative to vehicle control or as otherwise indicated. K M Warde et al.
LXRα, a novel therapeutic approach to ACC

LXRα inhibition reduces expression of cholesterol efflux pumps in H295R and MUC-1 ACC cells
In H295R and MUC-1, LXRα expression at baseline (Fig.  3A and B) was equivalent to expression in the positive control HepG2 hepatocellular carcinoma cells. Similarly, high expression levels of the oxysterol synthetic enzyme CYP27A1 were detected ( Fig. 3A and B). Transfection with a luciferase construct under the control of an LXRα-responsive element (TK-LXRRE-Luc) showed significant constitutive baseline activation of LXRα, which was increased by stimulation with the LXRα agonist T0901317 (1 μM) (Fig. 3C). These data indicate high activity of the LXRα/oxysterol pathway in adrenal cells, equivalent to that of hepatic tissue.
Expression of LXRα target genes, the FC efflux pumps, ABCA1 and ABCG1, were reduced by each LXRα inhibitor in H295R and MUC-1, respectively (Fig. 3D, E, F, G and Supplementary Fig. 4A, B, C, D). In line with these findings, expression of ABCA1 and ABCG1 increased in both cell lines when stimulated using the non-steroidal  = 4), statistical comparisons were performed using a two-tailed t-test or a two-way ANOVA followed by Tukey's post-hoc analysis. Statistical significance is denoted as *P < 0.05, **P < 0.01, ***P < 0.01 and ****P < 0.001 relative to vehicle control or as otherwise indicated. K M Warde et al.
LXRα, a novel therapeutic approach to ACC

Endocrine-Related Cancer
LXRα agonist T0901317 (Fig. 3D, E, F and G). Expression for ABCA1 and ABCG1 also decreased in the presence of the LXRα-DN construct (Fig. 3H), confirming direct involvement of LXRα in their expression. The responses of ABCA1 and ABCG1 in the presence of increasing concentrations of mitotane did not significantly change (Fig. 3I and J). However, the efficacy of GSK2033 or SR9243 to reduce ABCA1 and ABCG1 expression were attenuated with mitotane treatment in a dose-dependent manner, which is unlikely to reflect a direct effect of mitotane on cholesterol efflux pumps. However, this may be explained by increasing intracellular FC accumulation at higher mitotane doses. We next investigated intracellular cholesterol within this model.

LXRα inhibition enhances mitotane-induced toxic accumulation of free cholesterol in H295R ACC cells
Mitotane exposure resulted in a dose-dependent increase in intracellular FC in H295R, reaching significance at all mitotane concentrations (Fig. 4A) (15, 16). The inhibition of LXRα with GSK2033 and SR9243 significantly increased intracellular FC in H295R at baseline and in combination with mitotane, with higher levels for each dose of mitotane vs mitotane alone ( Fig. 4A and Supplementary  Fig. 5A). For MUC-1, only the cytotoxic 200-μM mitotane concentration resulted in a significant increase in intracellular FC, increased further with SR9243 (Fig. 4B). Cell death and intracellular cholesterol correlated for both cell lines: H295R (r 2 = 0.5210, P = 0.008) and MUC-1 (r 2 = 0.9299; P = 0.0001) (Supplementary Fig. 5B and C). While intracellular FC differed in response to mitotane and LXRα inhibition, baseline levels were similar for H295R and MUC-1 (Fig. 4C and D). Interestingly, intracellular lipid droplet storage of CE was undetectable in MUC-1, when compared to H295R (Fig. 4E, F, G and Supplementary Fig. 6A, B).
SOAT1 expression was significantly reduced by mitotane at doses <40 µM in H295R ( Supplementary Fig.  5D). However, SOAT1 expression was unaffected by LXRα inhibition alone in both cell lines (Supplementary Fig.  5E and F), suggesting that intracellular FC is increasing due to the alternative mechanism of reduced efflux pump expression (Fig. 3).

Discussion
Adrenocortical carcinoma (ACC) has demonstrated resistance to most cytotoxic chemotherapeutic approaches, with mitotane representing the single best pharmacotherapeutic for disease control; however, with a limited therapeutic window. Recent work demonstrated the putative adrenotoxic mechanism of mitotane through inhibition of the enzyme SOAT1, resulting in intracellular FC accumulation and suggests SOAT1-specific inhibitors as stand-alone therapy or adjuncts to mitotane in the management of ACC (Sbiera et al. 2015, LaPensee et al. 2016. In the study presented herein, we explored LXRα/cholesterol efflux pathway and proposed this as a complementary mechanism to enhance toxic intracellular FC accumulation in ACC. We investigated these effects on two validated in vitro models of ACC: H295R cells which are mitotane-sensitive and MUC-1, a metastatic, mitotane-insensitive cell line (Hantel et al. 2016).
Transcellular cholesterol flux in steroidogenic cells, such as ACC, is central to cellular function and cholesterol provides the principle substrate for steroidogenesis. In steroidogenic cells intracellular cholesterol (1) is stored within lipid droplets following conversion to cholesteryl esters by SOAT1 (Cignarella et al. 2005, Dove et al. 2006); (2) provides a substrate for steroidogenesis following transfer to the mitochondrion via the STAR protein (Stocco & Clark 1996) or (3) is removed from the cells by the cholesterol efflux pumps, ABCA1 or ABCG1 to circulating ApoA1 and HDL 1 , respectively (Kraemer 2007). Cholesterol efflux is tightly regulated by the oxysterol/LXRα system through conversion of surplus FC to oxysterols through the action of enzymes such as CYP27A1. In the adrenal, the role of LXRα and its ligand oxysterols are of key relevance in all of these processes. This was highlighted by Cummins et al. who demonstrated the contribution of oxysterol/LXRα to (1) regulate steroidogenesis, (2) control adrenocortical cholesterol metabolism and (3) act as a 'safety valve' through the ABCA1/ABCG1-mediated efflux pathway, to provide a basal protective mechanism in preventing free-cholesterol induced adrenal lipotoxicity (Cummins et al. 2006).
Our data show that pharmacological inhibition or functional blockade of LXRα significantly reduced cholesterol efflux pump expression (ABCA1 and ABCG1) and is accompanied by higher intracellular FC concentrations, ER stress, apoptosis and cell death markers. Using the combined therapeutic approach, the mitotanesensitive H295R underwent apoptosis and cell death at subtherapeutic mitotane concentrations. However, MUC-1, while combined mitotane/LXRα inhibition considerably enhanced the cytotoxic effect of mitotane alone, retained their resistance to pharmacologically acceptable mitotane concentrations below 200 µM. In both cell lines, higher intracellular FC levels were associated with increased cell death and when challenged with a cholesterolloaded extracellular environment, the cytotoxic effect of mitotane alone and in combination with LXRα inhibition was further enhanced. These findings support our overarching hypothesis and are also in line with the findings of other groups (Sbiera and La Pensee) who have investigated SOAT1 inhibition (Sbiera et al. 2015, LaPensee et al. 2016. While H295R were very sensitive to combined LXRα/mitotane treatment and MUC-1 demonstrated enhanced sensitivity to the combination at high mitotane doses, MUC-1 gave disappointing results in terms of a therapeutic strategy. However, when taken together, we believe our data convincingly support a role for LXRα and cholesterol storage/efflux in ACC cell adaptation and survival.
It is unclear why MUC-1 are insensitive to LXRα/ mitotane treatment. MUC-1 were derived from a metastatic neck lesion in a young, mitotane-resistant patient and K M Warde et al. LXRα, a novel therapeutic approach to ACC

27:6
Endocrine-Related Cancer for whom four cycles of combination chemotherapy were unsuccessful in limiting the extent disease spread. When we evaluated intracellular cholesterol in MUC-1, we demonstrated an almost complete absence of stored cholesteryl esters (CE) at baseline and even under cholesterol-loaded conditions (data not shown). In contrast, mitotane-sensitive H295R supported a rich CE store within their intracellular lipid droplets under baseline and cholesterol-loaded culture conditions. Yet, in spite of markedly different intracellular CE stores, intracellular FC was similar for both cell lines. The difference in intracellular cholesterol handling between MUC-1 and H295R is interesting, particularly in the context of the putative mechanism of mitotane activity through SOAT1. We did not perform a SOAT1 activity assay; however, the reduced storage of CE in MUC-1 in the presence of similar FC suggests low baseline SOAT1 activity, irrespective of the expression of this enzyme. This also suggests that, in contrast to H295R, MUC-1 favor alternative mechanism(s) of regulation and clearance of intracellular cholesterol, providing a possible escape mechanism from mitotane-induced lipotoxicity. Seidel et al. have described similar findings in relation to FC and CE handling in a strain of HAC15 with induced resistance to mitotane (Seidel et al. 2020). Detailed evaluation of lipid/cholesterol metabolism in ACC, using the full range of cell lines and banked tumour samples, is necessary to evaluate the significance for development of future therapeutic strategies.
In the presence of increasing FC accumulation, higher mitotane doses attenuated the inhibitory effect of LXRα blockade on cholesterol efflux pump expression in H295R. In our study, mitotane alone had no significant effect on either ABCA1 or ABCG1 expression. Sbeira et al. previously demonstrated a short-lived decrease in ABCG1 which quickly increased back to baseline with prolonged therapy (Sbiera et al. 2015). It is therefore unlikely that this is a direct pharmacodynamic effect of mitotane. It is a possible secondary effect of intracellular FC accumulation which is a substrate for catalysis to 27HC by CYP27A1 and will compete with the LXRα antagonists. In line with this observation, we have also demonstrated that stimulating LXRα with 27HC reduces the therapeutic efficacy of mitotane. Whether or not this represents a mechanism whereby ACC cells can overcome long-term susceptibility to mitotane needs more detailed investigation.
The role of LXRα in supporting the cancer microenvironment and promoting cancer cell survival is established in other human cancer models (Kim et al. 2018. Higher tumour grade and enhanced metastasis in breast cancer have been associated with higher LXRα expression and activation (Nelson et al. 2013). Work by Flaveny et al. also demonstrated that the LXRα inverse agonist SR9243 induced apoptosis across models of lung, colon and prostate cancer (Flaveny et al. 2015). Other work has demonstrated the potential for LXRα inhibition to induce lipotoxicity through FC accumulation (Schuster et al. 2002, Bradley et al. 2007. We propose that, based on our date, the use of LXRα inhibitors can be translated into a therapeutic strategy for mitotane-sensitive ACC by adopting a complementary approach to one of mitotane's known mechanisms of action: intracellular FC-induced lipotoxicity. We show that enhanced FC accumulation within H295R in response to LXRα blockade increased the cytotoxicity of mitotane at half the usual minimum therapeutic dose. The main limitations to mitotane's clinical use are (1) no significant improvements in progression or recurrencefree survival of disease below 14 mg/L , Berruti et al. 2017 and (2) poor patient tolerability, with significant impact on quality of life and inability to continue therapy above 20 mg/mL (Haak et al. 1994). Recent data have suggested that adjunctive mitotane should continue indefinitely even in individuals with R0 resection without tumour recurrence (Berruti et al. 2017). Therefore, the development of strategies to facilitate the long-term tolerability of this drug is an important clinical consideration. In relation to mitotaneresistant disease, we demonstrated increased cell death for MUC-1 in response to combined mitotane and LXRα inhibition, but show that mitotane-resistance was not overcome. Therefore, while adopting this combination strategy offers promise in mitotane-sensitive disease and to enhance adjuvant mitotane, it is unlikely to be of benefit in disease that is mitotane resistant. However, the differences in cholesterol handling between mitotanesensitive and mitotane-resistant ACC cells demonstrate the merits of better understanding the adaptation of ACC cells to facilitate metastasis and cell survival. Finally, we propose that the selective induction of high cholesterol uptake or selective delivery of cholesterol to ACC cells as a combination therapy represents an interesting challenge for future translational research.
In summary, two cell lines representing some of the best validated human cell culture models of ACC, the mitotane-resistant metastatic MUC-1 cell line and the mitotane-sensitive H295R, representing the standard for early mechanistic/therapeutic investigation, were K M Warde et al. LXRα, a novel therapeutic approach to ACC

27:6
Endocrine-Related Cancer used to investigate LXRα inhibitors and mitotane's effect to induce adrenal cytotoxicity (Rainey et al. 1994, Bird et al. 1995, Wang & Rainey 2012, Hantel et al. 2016. Our data suggest a complementary pathway to enhance mitotane-induced cytotoxicity by preventing FC efflux from ACC cells, mediated through blockade of the LXRα pathway and enhanced by cholesterol loading. Therefore, it is reasonable to conclude that the mitotane-LXRα inhibition combined therapeutic approach exploits separate but complementary and synergistic mechanisms to accumulate toxic levels of FC and to induce ACC cytotoxicity. The findings provide valuable mechanistic insights and an important backdrop to support further animal and pre-clinical studies of the effects of LXRα inhibition as a therapeutic target for ACC in mitotanesensitive disease. Of the two pharmacological inhibitors chosen for investigation, the inverse agonist SR9243 is well tolerated in animals and has a favorable toxicity profile (Flaveny et al. 2015, Wu et al. 2019). This drug therefore presents good translational feasibility in future in vivo studies. We also highlight the ongoing challenge of approaching effective strategies for drugresistant metastatic disease and the importance of better understanding cholesterol metabolism in ACC cells in this regard.