ELSEVIER

Contents lists available at ScienceDirect

Molecular and Cellular Endocrinology

journal homepage: www.elsevier.com/locate/mce

Molecular and Celular Endocrinology

The adrenal specific toxicant mitotane directly interacts with lipid membranes and alters membrane properties depending on lipid composition

Holger A. Scheidt ª, Ivan Haralampiev b, Stephan Theisgen ª, Andreas Schirbel ”, Silviu Sbiera ª, Daniel Huster ª, Matthias Kroiss ª, Peter Müller b, *

a University of Leipzig, Institute of Medical Physics and Biophysics, Härtelstr. 16-18, 04107 Leipzig, Germany

b Humboldt University Berlin, Department of Biology, Invalidenstr. 42, 10115 Berlin, Germany

· University Hospital Würzburg, Department of Nuclear Medicine, Oberdürrbacher Straße 6, 97080 Würzburg, Germany

d University Hospital Würzburg, Department of Internal Medicine I, Endocrinology and Diabetes Unit, Oberdürrbacher Straße 6, 97080 Würzburg, Germany

ARTICLE INFO

Article history:

Received 8 December 2015 Received in revised form 26 February 2016 Accepted 16 March 2016

Available online xxx

Keywords: Mitotane Lipid vesicles Membrane structure Spin-labeled lipid Fluorescent lipid Order parameter Lipid-drug interaction

ABSTRACT

Mitotane (o,p’ .- DDD) is an orphan drug approved for the treatment of adrenocortical carcinoma. The mechanisms, which are responsible for this activity of the drug, are not completely understood. It can be hypothesized that an impact of mitotane is mediated by the interaction with cellular membranes. However, an interaction of mitotane with (lipid) membranes has not yet been investigated in detail. Here, we characterized the interaction of mitotane and its main metabolite o,p’-dichlorodiphenyldichloroacetic acid (o,p’-DDA) with lipid membranes by applying a variety of biophysical approaches of nuclear mag- netic resonance, electron spin resonance, and fluorescence spectroscopy. We found that mitotane and o,p’-DDA bind to lipid membranes by inserting into the lipid-water interface of the bilayer. Mitotane but not o,p’-DDA directly causes a disturbance of bilayer structure leading to an increased permeability of the membrane for polar molecules. Mitotane induced alterations of the membrane integrity required the presence of phosphatidylethanolamine and/or cholesterol. Collectively, our data for the first time char- acterize the impact of mitotane on the lipid membrane structure and dynamics, which may contribute to a better understanding of specific mitotane effects and side effects.

1. Introduction

Mitotane (o,p’-dichlorodiphenyldichloroethane, o,p’-DDD, Lysodren®) is the only drug approved for treatment of adrenocor- tical carcinoma (ACC) (Hahner and Fassnacht, 2005; Igaz et al., 2008). ACC is an orphan disease with an annual incidence of 0.5-2/million inhabitants (Golden et al., 2009). Complete tumor removal is still the only potentially curative option and is the initial treatment of choice in localized disease (Jurowich et al., 2013; Schteingart et al., 2005). Since local recurrence is frequent, mito- tane is recommended by most centers as an adjuvant treatment after complete resection (Terzolo et al., 2007, 2013; De Francia et al., 2012).

In advanced disease, mitotane is a cornerstone of the treatment

as well (Fassnacht et al., 2011, 2013; Else et al., 2014), but an objective response to monotherapy is only observed in 20% of the patients (Hahner and Fassnacht, 2005). Importantly, mitotane treatment is complicated by side-effects and drug interactions (Kroiss et al., 2011).

Despite its clinical use for more than five decades and efforts from several groups using various experimental approaches (Moore et al., 1980; Kruger et al., 1984; Jensen et al., 1987; Stigliano et al., 2008; Hescot et al., 2013; Poli et al., 2013) the molecular mecha- nisms underlying mitotane efficacy in ACC are only poorly under- stood. Only recently, we provided evidence that mitotane induces endoplasmic reticulum stress specifically in adrenocortical cells by accumulation of toxic lipids. We demonstrated that mitotane directly inhibits the Sterol-O-Acyl Transferase 1 (SOAT1), an enzyme bound to mitochondria-associated endoplasmic reticulum membranes (MAM) of the endoplasmic reticulum, thus explaining a perturbation of lipid homeostasis through mitotane (Sbiera et al., 2015). We also demonstrated that the majority of the cellular

* Corresponding author.

E-mail address: peter.mueller.3@rz.hu-berlin.de (P. Müller).

http://dx.doi.org/10.1016/j.mce.2016.03.022

Please cite this article in press as: Scheidt, H.A., et al., The adrenal specific toxicant mitotane directly interacts with lipid membranes and alters membrane properties depending on lipid composition, Molecular and Cellular Endocrinology (2016), http://dx.doi.org/10.1016/ j.mce.2016.03.022

Abbreviations

ACC adrenocortical carcinoma

Chol cholesterol

o,p’-DDA o,p’-dichlorodiphenyldichloroacetic acid

HBS HEPES buffered saline

HBS50

HBS containing 50 mM HEPES

DOPC 1,2-dioeloyl-sn-glycero-3-phosphocholine

DOPE 1,2-dioeloyl-sn-glycero-3-phosphoethanolamine

DOPS

1,2-dioeloyl-sn-glycero-3-phosphoserine

LUVs

large unilamellar vesicles

Mitotane o,p’-dichlorodiphenyldichloroethane

NBD-PC 1-palmitoyl-2-(12-[N-(7-nitrobenz-2-oxa-1,3-diazol- 4-yl)amino]dodecanoyl]-sn-glycero-3- phosphocholine

NBD-PE 1-palmitoyl-2-(12-[N-(7-nitrobenz-2-oxa-1,3-diazol- 4-yl)amino]dodecanoyl]-sn-glycero-3- phosphoethanolamine

NBD-PS 1-palmitoyl-2-(12-[N-(7-nitrobenz-2-oxa-1,3-diazol-

4-yl)amino]dodecanoyl]-sn-glycero-3-phosphoserine

POPC 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine

POPC-d31 perdeuterated POPC

POPE 1-palmitoyl-2-oleoyl-sn-glycero-3- phosphoethanolamine

POPE-d31 perdeuterated POPE

PS phosphatidylserine

SL-Chol 25-doxyl-cholesterol

SL-PC 1-palmitoyl-2-(4-doxylpentanoyl)-sn-glycero-3- phosphocholine

SL-PE 1-palmitoyl-2-(4-doxylpentanoyl)-sn-glycero-3- phosphoethanolamine

SL-PS 1-palmitoyl-2-(4-doxylpentanoyl)-sn-glycero-3- phosphoserine.

uptake of a radioactive mitotane analog occurred in association with lipids (Sbiera et al., 2015). Along these lines, recently the majority of mitotane circulating in the blood was found to be bound to lipoproteins (Kroiss et al., 2016; Hescot et al., 2015). Notably, a main metabolite of mitotane, o,p’-dichlorodiphenyldichloroacetic acid (o,p’-DDA), is inactive in vitro regarding induction of the hor- monal and cytotoxic effects observed with mitotane (Hescot et al., 2014) although in a clinical setting, determination of o,p’-DDA in serum may provide additional information about tumor response (Hermsen et al., 2011).

From a physicochemical point of view, mitotane and o,p’-DDA (Fig. 1) are very likely to interact with lipid membranes given the lipophilicity of their aromatic ring structures. For similar molecules like local anesthetics and other aromatic compounds a localization in the lipid-water interface of the membrane has been found by various biophysical methods (Yau et al., 1998; Huster et al., 2001; Scheidt et al., 2004; Weizenmann et al., 2012). In these cases, membrane interaction was shown to be caused by the hydrophobic effect, favorable dipole-dipole interactions, and hydrogen bonding

to more polar lipid groups (Yau et al., 1998; Huster et al., 2001).

To the best of our knowledge, only a single publication has dealt with the impact of mitotane on biological membranes (Jacobi et al., 2014). By using erythrocyte cell membranes after mitotane expo- sure, intracellular Ca2+ was found to be increased, which caused the appearance of phosphatidylserine (PS), a lipid which is normally localized on the inner membrane leaflet (Zachowski, 1993), on the outer membrane leaflet. Intriguingly, enhanced transbilayer phos- pholipid movement leading to the accumulation of PS in the outer membrane leaflet is a feature of apoptosis (Bevers et al., 1999; Bevers and Williamson, 2010), a process triggered by mitotane in ACC tumor cells (Lehmann et al., 2013; Sbiera et al., 2015). The mechanism(s), by which mitotane affected the erythrocyte mem- brane, were not explored further.

The current study aims at characterizing the interaction of mitotane and o,p’-DDA with lipid bilayers focusing on the two most important constituents of biological membranes, phospholipids and cholesterol. In particular, the interaction of mitotane and o,p’- DDA with lipid vesicles of defined composition was investigated,

Fig. 1. Chemical structures and isosurfaces of the electrical potential of mitotane (A) and o,p'-DDA (B). The molecular dipole moment is indicated by the arrows in the isosurface plots.

CĮ

H G

OHH

៛

which allows an adjustment of the membrane lipid mixture in or- der to determine the influence of specific lipid species. We applied different biophysical approaches (NMR, ESR, and fluorescence spectroscopy) in conjunction with deuterated, fluorescence- and, spin-labeled lipids.

2. Materials and methods

2.1. Materials

All lipids, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE), 1,2-dioeloyl-sn-glycero-3-phosphocholine (DOPC), 1,2- dioeloyl-sn-glycero-3-phosphoserine (DOPS), 1,2-dioeloyl-sn-glyc- ero-3-phosphoethanolamine (DOPE), cholesterol (chol), 1- palmitoyl-2-(12-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino] dodecanoyl]-sn-glycero-3-phosphocholine (NBD-PC), 1-palmitoyl- 2-(12-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]dodecanoyl]- sn-glycero-3-phosphoethanolamine (NBD-PE), 1-palmitoyl-2-(12- [N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]dodecanoyl]-sn- glycero-3-phosphoserine (NBD-PS) and the sn-1 chain perdeu- terated analogs of POPC and POPE (POPC-d31, POPE-d31) were pur- chased from Avanti Polar Lipids, Inc. (Alabaster, AL, USA). Spin- labeled (SL) phospholipids 1-palmitoyl-2-(4-doxylpentanoyl)-sn- glycero-3-phosphocholine (SL-PC), 1-palmitoyl-2-(4- doxylpentanoyl)-sn-glycero-3-phosphoethanolamine (SL-PE), and 1-palmitoyl-2-(4-doxylpentanoyl)-sn-glycero-3-phosphoserine (SL-PS) were prepared as described previously (Fellmann et al., 1994). 25-doxyl-cholesterol (SL-Chol) was synthesized according to the protocol of Maurin et al (Maurin et al., 1987). Mitotane was purchased from Sigma (Taufkirchen, Germany). The synthesis of o,p’-DDA has been described (Kroiss et al., 2016) and is provided in Supplementary Material (Fig. S1). All other chemicals were pur- chased from Sigma-Aldrich (Taufkirchen, Germany) and used without further purification.

2.2. Preparation of large unilamellar vesicles (LUVs)

LUVs were prepared by extrusion (Mayer et al., 1985). Aliquots of lipids were dissolved in chloroform and dried in a rotating round- bottom flask under vacuum until a lipid film was formed. Lipids were resuspended at first in a small volume of ethanol (final ethanol concentration was below 1% (v/v)). Subsequently, HBS (HEPES buffered saline, 145 mM NaCl and 10 mM HEPES, pH 7.4) was added to reach a final lipid concentration of 1 or 5 mM and the mixture was vortexed. To prepare LUVs, this suspension was sub- jected to five freeze-thaw cycles followed by extrusion of the lipid suspension 10 times through two 0.1 um polycarbonate filters at 40 ℃ (mini-extruder from Avanti Polar Lipids; filters from Costar, Nucleopore, Tübingen, Germany).

2.3. Sample preparation for NMR experiments

For the NMR measurements, mixtures of mitotane or o,p’-DDA and phospholipids were dissolved in chloroform at the respective molar ratios. After evaporating the solvent, the samples were re- dissolved in cyclohexane and lyophilized overnight at high vac- uum to obtain a fluffy powder. After hydration with 50wt% D2O for 1H MAS NMR or deuterium-depleted H2O for 2H NMR experiments, the samples were equilibrated by freeze-thaw cycles and gentle centrifugation and finally transferred into 4 mm HR MAS rotors with spherical Kel-F inserts.

2.4. 2H NMR measurements

2H NMR measurements were conducted on a Bruker DRX300 NMR spectrometer (Bruker Biospin GmbH, Rheinstetten, Germany) operating at a resonance frequency of 46.1 MHz for 2H using a double channel solids probe equipped with a 5 mm solenoid coil. The 2H NMR spectra were obtained using a phase-cycled quad- rupolar echo sequence (Davis et al., 1976) and a relaxation delay of 1 s. The two 7/2 pulses of ca. 3 us were separated by a 50 us delay. All spectra were measured at a temperature of 30 ℃. After depaking the spectra (Sternin et al., 1983) the smoothed order parameter profiles were calculated according to Lafleur et al. (1989).

2.5. 1H MAS NMR spectroscopy

1H MAS NMR measurements were carried out on a Bruker Avance III 600 MHz spectrometer using a 4 mm HR MAS probe at a MAS frequency of 8 kHz. A 2H lock was used for field stability. The Tt/2 pulse length was 4 us. In all 1H NMR spectra the chemical shift of the terminal methyl group of the lipid chains was calibrated at 0.885 ppm, which represents a referencing relative to TMS. All measurements were conducted at a temperature of 30 ℃.

Two-dimensional 1H MAS NOESY spectra (Jeener et al., 1979) were acquired at five mixing times (between 0.1 ms and 500 ms). At least 512 data points were acquired in the indirect dimension at a relaxation delay of 3.4 s.

The volume of the respective diagonal and cross peaks was in- tegrated using the Bruker Topspin 2.1 software package. NOE build- up curves were fitted in Origin (OriginLab Cooperation, North- ampton, MA) to the spin pair model yielding cross-relaxation rates (ôij) according to Scheidt and Huster (2008):

Ajj(0) Aij(tm) =^.(1-exp(-20jtm))exp(-tm/Tij) (1)

where Aij (tm) represents the cross peak volume at mixing time tm and Ajj (0) the diagonal peak volume at mixing time zero. The value 1/Tij defines a rate of magnetization leakage towards the lattice.

2.6. Reduction of spin-labeled phospholipids by ascorbate

For measuring the permeability of membranes, the reduction of the spin-labeled lipids by ascorbate was measured as described in Greube et al. (2001). 5 mM LUVs symmetrically labeled with 0.2 mM SL-PL were mixed with mitotane, o,p’-DDA or ethanol (with the same volume used for addition of the drug). Subsequently, sodium ascorbate in HBS was added (final concentration 20 mM) from a 100 mM stock solution adjusted to pH 7.4. ESR spectra were recorded at room temperature at different times using an EMX spectrometer (Bruker, Karlsruhe, Germany). Measuring parameters were as follows: modulation amplitude 1 G, power 20 mW, scan width 80 G, 1x accumulated. The decrease of ESR signal intensity was estimated by relating the intensities of the low field signal to those in the absence of ascorbate.

2.7. Reduction of fluorescent phospholipids by dithionite

Alternatively, for determining membrane permeability, an assay was used, which measures the permeation of the anion dithionite across membranes (Mcintyre and Sleight, 1991; Pomorski et al., 1994). LUVs of different lipid and NBD-labeled lipid composition were prepared. Vesicles were mixed in a cuvette with mitotane, o,p’-DDA or ethanol (same volume as added with the drugs). Sub- sequently, HBS50 (HEPES buffered saline, 145 mM NaCl and 50 mM HEPES, pH 7.4) was added and fluorescence of the NBD group was

monitored continuously at 540 nm (ex = 470 nm, slit width for excitation and emission, each 4 nm) at 37 ℃. After 30 s, sodium dithionite was added from a freshly prepared 1 M stock solution in 100 mM Tris (pH 10) to give a final concentration of 50 mM. Finally, after 300 s, Triton X-100 was added to a final concentration of 0.5% (w/v) enabling complete reaction of dithionite with the respective NBD-labeled lipid resulting in a loss of fluorescence. The curves were normalized to the fluorescence intensities before addition of dithionite and after addition of Triton X-100. Significances were tested using the open software R 2.15. Rate constants in the pres- ence of mitotane and o,p’-DDA were compared using a Mann- Whitney-U-test.

3. Results

3.1. Interaction of mitotane with membranes measured by 1H MAS NMR spectroscopy

In order to investigate whether (i) mitotane and o,p’-DDA attach to lipid membranes and, in case the molecules bind, (ii) where the molecules are located in the membrane, 1H MAS NMR measure- ments were performed. The respective 1H MAS NMR spectra of 20 mol% mitotane and o,p’-DDA in POPC membranes are shown in Fig. 2. For sensitivity reasons, such high drug concentrations were necessary in order to obtain a satisfactory signal to noise ratio in the NMR measurements (especially in the subsequent 2D NOESY measurements). In addition to the well resolved lipid signals, the 1H NMR signals of the aromatic ring protons of both mitotane or o,p’- DDA are observed in the region of 7-8 ppm. Due to the lack of resolution and the overlap of several signals, it was not possible to assign these resonances to a specific proton with full confidence. The literature data on the signal assignment of mitotane are con- tradictory (Alfonsi et al., 2013) (http://riodb01.ibase.aist.go.jp/sdbs/ cgi-bin/direct_frame_top.cgi). One has to assume that protons of both aromatic rings contribute to the signals used for the NOESY analysis. In addition, the proton HG (see Fig. 1) between the aro- matic rings is detected at 5.1 ppm for mitotane and 5.4 ppm for o,p’- DDA.

Quantitative information about the membrane localization of either molecule was obtained from the induced 1H NMR chemical shift data. Due to the presence of the aromatic molecules in the vicinity of the individual POPC segments, the POPC signals expe- rience a small chemical shift change, which is plotted in Fig. 3. These induced shift effects are caused by the ring current effect of the T-electrons of the aromatic rings and depend on the proximity to the respective molecular groups of POPC (Stamm and Jaeckel, 1989; Scheidt et al., 2004). Thus, a plot of the induced chemical shifts against the individual phospholipid segments can be inter- preted as an approximate distribution function of the aromatic rings in the lipid membrane. For both molecules, the largest induced chemical shifts are observed in the upper chain region of the phospholipids in the membrane (C-3, C-2, G-1). Especially in the presence of mitotane, a small bias in the distribution function towards the membrane-water interface for the mixed POPC/POPE membrane is observed (Fig. 3B). Interestingly, for o,p’-DDA a large influence on the head group of POPC can be seen as well (Fig. 3C, D).

Further insights into the membrane location of small molecules can be achieved in NOESY experiments and the analysis of the cross-relaxation rates between the molecular groups of the phos- pholipids and small membrane bound molecules (Huster et al., 1999; Scheidt et al., 2003, 2004; Scheidt and Huster, 2008). Due to the strong distance dependence of the cross-relaxation rate gij, a plot of the cross-relaxation rate for each individual segment of POPC provides an estimate for the distribution function of the respective membrane bound molecule within the lipid membrane.

Fig. 2. 1H MAS NMR spectra of POPC membranes in the presence of 20 mol% mitotane (A) and o,p'-DDA (B). The spectra were measured at a MAS frequency of 8000 Hz and a temperature of 30 ℃. The signals of POPC are assigned; the signals of mitotane or o,p'- DDA used for NOESY analysis are marked by arrows.

(CH2)n

HDO

CH3

CH=CH

CH2-C=

C-3

G-2

G-1

G-3

C-2

G-1

1H Chemical Shift /ppm

Fig. 4 shows the cross-relaxation rate between the aromatic pro- tons and the single proton HG of mitotane or o,p’-DDA and indi- vidual lipid segments of POPC. For all mitotane or o,p’-DDA signals investigated, a relatively broad distribution function in the lipid membrane was obtained, which is a consequence of the high mo- lecular disorder and mobility of liquid-crystalline lipid membranes (Huster et al., 1999; Scheidt and Huster, 2008). It is interesting, that the plots of the induced chemical shifts show a higher propensity of o,p’-DDA to interact with the lipid head group, which was not observed in the cross-relaxation rates.

The maximum of these distribution functions provides the mean position of the molecule in the lipid membrane. For both molecules a maximum for the cross-relaxation rates is found that corresponds - in confirmation to the results from the induced chemical shift measurements - to the upper chain region (C-3, C-2, G-1, G-3) of the phospholipid membrane, the lipid-water interface. Because the distribution functions for different protons in the

Fig. 3. Induced chemical shifts of individual phospholipid segments in POPC (A, C) or POPC/POPE (B, D) membranes by 20 mol% mitotane (A, B) or o,p'-DDA (C, D). The structure at the top shows the assignment of the individual lipid segments of POPC. Note the different scales of the ordinate.

CH2-C=

CH=CH

CH2-C=

C-3

C-2

CH3

O-U=O

₿

CH3

(CH2);

C-3

C-2

G-1

G-2

G-3

0.07

0.08

0.06

0.07

0.05

0.06

0.04

0.05

0.04

Induced Chemical Shift / ppm

0.03

0.02

0.01

0.00

0.10

0.14

0.12

0.08

0.10

0.08

0.06

0.04

0.02

0.00

CH3

(CH2)

CH2-C=

CH=CH

C-3

C-2

G-1

G-2

G-3

0.00

฿

CH3

(CH2)

CH2-C=

CH=CH

C-3

C-2

G-1

G-2

G-3

molecules are quite similar, one has to assume a fast reorientation of mitotane and o,p’-DDA in the membrane. The differences be- tween mitotane and o,p’-DDA are rather small. Only for the proton HG of mitotane an increased interaction to the acyl chains of POPC is observed, which may result in a different orientation or mobility of mitotane in the membrane compared to o,p’-DDA. A direct quan- titative comparison of values of the cross-relaxations rates between mitotane and o,p’-DDA is difficult, because these rates are also dependent on the correlation time of motion of these molecules, however, all measured cross-relaxation rates are in the same order of magnitude.

Comparing the cross-relaxations rates for mitotane and o,p’- DDA in POPC membranes, with those obtained for POPC/POPE membranes (Fig. 5), we find no qualitative difference in the dis- tribution functions of the molecules in the two lipid environments.

3.2. Impact of mitotane and o,p’-DDA on membrane properties

To investigate the influence of both molecules on the packing of the acyl chains in their host membranes, 20 mol% of the respective molecule were incorporated into either POPC bilayers or into mixed POPC/POPE membrane in a molar ratio of 2:1 (mol/mol). The 2H NMR spectra (not shown) of POPC-d31 as well as POPE-d31 in these samples exhibited the typical 2H NMR powder patterns, consisting of a superposition of Pake doublets known for lamellar liquid- crystalline lipid membranes (Davis, 1983). This indicates that the bilayer structure and phase state of the lipid vesicles is not

dramatically affected by the addition of 20 mol% of the respective molecule. The smoothed chain order parameter profiles for POPC- d31 in the presence of 20 mol% mitotane or o,p’-DDA are shown in Fig. 6A with a pure POPC-d31 membrane serving as control. The changes in the lipid chain order parameters induced by the drug molecules are moderate. While in the presence of mitotane an in- crease for the upper chain is observed, the order parameters decrease in the middle/lower chain region in the presence of o,p’- DDA. In mixed POPC/POPE membranes, the order parameters of POPC-d31 and POPE-d31 exhibit a different behavior in the presence of mitotane compared to the reference samples of pure lipids (Fig. 6B): while the order parameters of POPC-d31 are again slightly increased, the order parameters of POPE-d31 are decreased. Even though the differences are quite small, this might be evidence for a different interaction of mitotane with POPC and POPE, respectively. o,p’-DDA in general disorders lipid membranes. As observed for POPC membranes, the order parameters are decreased in POPC/ POPE membranes in the presence of o,p’-DDA (Fig. 6C), especially in the middle to lower chain region. In contrast to mitotane, there is no difference in these changes between POPC and POPE.

The 31P NMR spectra of the samples (not shown) exhibit the typical line shape of a lamellar liquid-crystalline phase state with a small isotropic component (2-7%) indicating that the overall bilayer structure of the vesicles is still intact also in the presence of mitotane and o,p’-DDA. There was no systematic correlation of this isotropic signal in the presence or absence of either mitotane or o,p’-DDA.

ARTICLE IN PRESS

Fig. 4. NOESY cross-relaxation rates (s-1) between the proton G (i.e. between the aromatic rings) (A) and between the protons of the aromatic rings (B, C) of mitotane (left) and o,p'- DDA (right) to the different molecular segments of POPC (see Fig. 3). The cartoon above illustrates the probable membrane location of the respective molecule.

Mitotane

DDA

エ

HHO

0.022-

0.10

0.020

0.018-

0.08

0.016-

0.014-

0.06

0.012-

0.010-

0.04

0.008

0.006-

0.004-

0.02

0.002

0.000-

0.00

Cross relaxation rate / s-1

0.014

Cross relaxation rate / s-1

0.045

0.012

0.040

0.035

0.010

0.030

0.008

0.025

0.006

0.020

0.015

0.004

0.010

0.002

0.005

0.000

0.014

0.12

0.012

0.10

0.010

0.08

0.008

0.006

0.06

0.004

0.04

0.002

0.02

0.000

CH3

(CH2)n

CH2-C=

C-2

G-1

G-3

0.00

C-3

₿

CH2

(CH2)n

CH2-C=

C-3

C-2

G-1

G-3

Fig. 5. NOESY cross-relaxation rates (s-1) between the proton G (i.e. between the aromatic rings) (A) and between the protons of the aromatic rings (B,C) of 20 mol% mitotane (left) or 20 mol% o,p'-DDA (right) to the different molecular segments of POPC (see Fig. 3) in a POPC/POPE (2:1 mol/mol) membrane. The cartoon above shows the probable membrane location of the respective molecule.

Mitotane

DDA

工

HHO

0.06

0.040

0.05

0.035

0.030

0.04

0.025

0.03

0.020

0.02

0.015

0.010

0.01

0.005

0.00

0.000

Cross relaxation rate / s-1

0.10

0.012

0.09

0.010

0.08

0.07

0.008

0.06

0.006

0.05

0.04

0.004

0.03

0.002

0.02

0.01

0.000

0.00

0.03

0.012

0.010

0.008

0.02

0.006

0.004

0.01

0.002

0.000

0.00

CH3

(CH2)n

CH2-C=

C-3

C-2

G-1

G-3

฿

(CH2),

CH2-C=

C-3

C-2

G-1

G-3

₿

3.3. Influence of mitotane and o,p’-DDA on membrane structure of LUV membranes measured by ESR spectroscopy

To investigate whether the binding of mitotane and o,p’-DDA to lipid membranes affects membrane permeability, two approaches were applied. In one assay, the ascorbate-mediated reduction of ESR signal intensity of spin-labeled phospholipids incorporated into the membranes was measured (Greube et al., 2001) and in the other the dithionite-mediated reduction of membrane embedded fluorescent phospholipids was determined (Langner and Hui,

1993).

For the former assay, LUVs comprising various phospholipids and the respective SL-PL were prepared and the kinetics of sodium ascorbate-mediated ESR signal reduction was measured (Fig. 7). In the absence of mitotane, addition of ascorbate to all lipid compo- sitions led to a rapid decrease of signal intensity to about 50% (Fig. 7, open circles). Subsequently, the curves leveled off and signal in- tensities remained almost constant. This observation indicates that solely spin-labeled analogues in the outer membrane leaflet are accessible to ascorbate and become rapidly reduced, whereas those

Fig. 6. 2H NMR chain order parameter of the sn-1 chain of POPC-d31 (squares) or POPE- d31 (triangle) in membranes in the presence of 20 mol% mitotane (blue symbols) or o,p'-DDA (red symbols). Order parameters were determined in POPC-d31 membranes (A) or mixed POPC/POPE (2:1 mol/mol) membranes (B, C). For comparison, in all plots the order parameters of pure lipid membranes are shown in black open symbols. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

0.25

0.20

0.15

0.10

0.05

0.00

0.25

Order parameter S

0.20

0.15

0.10

0.05

0.00

0.25

0.20

0.15

0.10

0.05

0.00

Carbon number n

localized in the inner layer are shielded from the reducing substance.

When the reduction assay was performed in the presence of mitotane (molar ratio PL/mitotane of about 4), the reduction ki- netics were different depending on the LUV species. The most significant effects were observed for DOPC/DOPE/Chol-LUVs. In

these vesicles, a rapid and complete disappearance (within 100 s) of the ESR signal of SL-PE was observed. This effect of mitotane was concentration dependent, since at lower concentration (molar ratio PL/mitotane of about 8), the ESR signal also completely vanished but with a slower rate. Mitotane also caused an increased signal reduction in DOPC/Chol- and DOPC/DOPE-LUVs, however, the impact of the drug on membrane was lower compared to DOPC/ DOPE/Chol-LUVs. For pure DOPC membranes (measured with SL- PC) and for DOPC/DOPS membranes (measured with SL-PS), no effect of mitotane on ascorbate permeation was observed compared to control vesicles.

Additionally, the impact of o,p’-DDA on LUVs was investigated. The reduction curves in the presence of o,p’-DDA were very similar to the control curves showing that the rate constants and the plateau values of reduction were similar at both conditions. Note, that for DOPC/DOPE-LUVs, the velocity of the first reduction component was somewhat slower in the presence of o,p’-DDA compared to control vesicles indicating some lower accessibility of the spin label moiety towards ascorbate in the presence of the drug.

These data indicate that mitotane disturbs the packing of the acyl chains in membranes depending on the lipid composition enabling ascorbate to permeate across the membrane and to reduce the analogues in the inner leaflet. In contrast, o,p’-DDA has no impact on membranes.

3.4. Influence of mitotane and o,p’-DDA on the membrane integrity of LUV membranes measured by fluorescence

Using a complementary assay, the influence of mitotane and o,p’-DDA was investigated by following the dithionite-mediated reduction of fluorescent phospholipids (Mcintyre and Sleight, 1991; Pomorski et al., 1994).

Dithionite rapidly quenches the fluorescence of the NBD-labeled lipid in the outer membrane leaflet by chemical reaction, which is reflected by a rapid initial decrease of fluorescence intensity (see Fig. 8). Subsequently, fluorescence intensity decreases slowly caused by a slow permeation of dithionite across the membrane thereby also reacting with the fluorescent lipids in the inner membrane leaflet. An increase of this slow dithionite permeation indicates a deterioration of membrane structure (Tannert et al., 2007). When the dithionite assay was performed e.g. on DOPC/ Chol/NBD-PC-LUVs in the presence of mitotane, NBD fluorescence also decreased rapidly to about 50% upon addition of dithionite (Fig. 8). In contrast, the fluorescence decrease of the second component was faster compared to the control curve indicating a higher membrane permeation of dithionite in the presence of mitotane. In contrast, the kinetics in the presence of o,p’-DDA was similar to that of the control.

To quantify these differences, curves were fitted to a bi- exponential equation giving the rate constants for the rapid fluo- rescence decrease (i.e. reduction of NBD-PC in the outer membrane leaflet) and those for the slow decrease (i.e. permeation of dithionite across the membrane). The latter rate constants were determined from kinetics measured on LUVs of different lipid and NBD-labeled lipid composition in the absence and presence of mitotane or o,p’-DDA (Fig. 9). The values in the presence of the drugs were normalized to the respective values in the absence of drugs since dithionite permeation also depends on the lipid composition of vesicles. E.g., in vesicles containing cholesterol (DOPC/Chol/NBD-PC-LUVs) dithionite permeation was slower compared to DOPC/NBD-PC-LUVs (data not shown) (Scheidt et al., 2013).

Fig. 9 shows that mitotane caused an increased dithionite permeation in vesicles containing PE and/or cholesterol, i.e. in DOPC/Chol/NBD-PC-, DOPC/DOPE/NBD-PE-, and DOPC/DOPE/Chol/

ARTICLE IN PRESS

Fig. 7. Influence of mitotane and o,p'-DDA on membrane integrity measured by the ascorbate assay. LUVs consisting of different lipids and spin-labeled phospholipids were prepared. The vesicles were mixed with sodium ascorbate to give final concentrations of 3.5 mM (phospholipids), 0.15 mM (SL-PL) and 20 mM (ascorbate) at time zero and transferred to an ESR capillary. ESR spectra were recorded at the given time points at 25 ℃ in the absence (O) or in the presence of 0.9 mM 0.5 mM mitotane, and 0.9 mM o,p'-DDA (A). ESR signal intensities (SI) of the low field peak were determined and normalized to those in the absence of ascorbate.

1.0

DOPC/SL-PC

1.0

DOPC/Chol/SL-PC

0.8

rel. SI

0.6

0.4

0.2

0.0

200

400

600

800

1000

1200

200

400

600

800

1000

1200

1.0

DOPC/DOPE/SL-PE

1.0

DOPC/DOPE/Chol/SL-PE

0.8

rel. SI

0.6

0.4

0.2

0.0

200

400

600

800

1000

1200

200

400

600

800

1000

1200

1.0

DOPC/DOPS/SL-PS

time [s]

0.8

rel. SI

0.6

0.4

0.2

0.0

200

400

600

800

1000

1200

time [s]

Fig. 8. Influence of mitotane and o,p'-DDA on the dithionite-mediated reduction of NBD-PC in DOPC/Chol-LUVs. Vesicles (lipid concentration 1 mM) were mixed in a cuvette with 0.1 mM mitotane (dashed line), 0.1 mM o,p'-DDA (dotted line) or ethanol (solid line, same volume as added with the drugs). Subsequently, 1.45 ml HBS50 was added and fluorescence was monitored with time at 540 nm (hex = 470 nm) at 37 ℃. At time 0, sodium dithionite was added (final concentration 50 mM). The curves were normalized to the fluorescence intensities before addition of dithionite and after addition of Triton X-100.

1.0

0.8

norm. FI

0.6

0.4

0.2

0.0

100

150

200

250

time [s]

NBD-PE-LUVs. In contrast, o,p’-DDA had no impact in these LUVs. Mitotane and o,p’-DDA had no influence on dithionite permeation in pure PC vesicles (DOPC/NBD-PC-LUVs) and in PS-containing vesicles (DOPC/DOPS/NBD-PS-LUVs). The results of the dithionite assay support the outcome of the ascorbate reduction experiments in that mitotane disturbs membrane structure depending on the presence of PE or cholesterol.

3.5. Influence of mitotane and o,p’-DDA on the mobility of spin- labeled phospholipids in LUV membranes

Recording the ESR spectra of spin-labeled lipids in membranes allows to detect changes of the analog mobility in the presence of membrane active molecules (Marsh and Horvath, 1998; Greube et al., 2001). We measured whether the incorporation of mitotane or o,p’-DDA into membranes influences the mobility of SL-PL by recording their ESR spectra. ESR spectra of SL-PL in LUVs of different lipid compositions were recorded immediately (i.e. after about 1 min) and 10 min after mixing mitotane/o,p’-DDA and LUVs (Fig. 10).

The ESR spectra of DOPC/DOPE/Chol/SL-PE-LUVs and DOPC/ Chol/SL-Chol-LUVs show in the presence of mitotane (molar ratio lipid/mitotane = 10) a signal broadening in comparison to control spectra, i.e. vesicles in buffer. This reflects a decreased rotational mobility of the respective analogs. The effect was similar whether the spectra were recorded immediately or 10 min after addition of mitotane. For DOPC/DOPE/SL-PE-LUVs, after 1 min solely a small

Fig. 9. Influence of mitotane and o,p'-DDA on membrane integrity measured by the dithionite assay. The reduction kinetics as shown in Fig. 8 were fitted to a bi-exponential equation. From the fits the rate constant k of the slower component was determined as a measure for dithionite permeation across membranes. All single values of the rate constant in the presence of mitotane (circles) or o,p'-DDA (squares) are shown which were normalized to those determined in the absence of the drugs. Note the different scales of the ordinate for the figures on the right side. Significant differences between the values in the presence of mitotane and o,p'-DDA are indicated (#P = 0.006, ##P = 0.041, ###P = 0.014).

2.0

3.0

2.5

1.5

norm. k

2.0

1.0

1.5

1.0

0.5

0.0

DOPC/NBD-PC

DOPC/Chol/NBD-PC

2.0

1.5

norm. k

1.0

0.5

0.0

DOPC/DOPE/NBD-PE

DOPC/DOPE/Chol/NBD-PE

2.0

1.5

norm. k

1.0

0.5

0.0

DOPC/DOPS/NBD-PS

immobilizing impact of mitotane was observed. The ESR spectrum recorded after 10 min shows a significant broadening of signals compared to the control spectrum. The spectra of DOPC/SL-PC- LUVs (see Fig. 10) and DOPC/DOPS/SL-PS-LUVs, DOPC/Chol/SL-PC- LUVs (see Supplementary Material, Fig. S2) indicate only a very small or no impact of mitotane. Notably, the spectra of all LUV species recorded in the presence of o,p’-DDA are identical to the control spectra. These measurements show that mitotane has an immobilizing impact on membrane lipids which depends on the lipid composition. In contrast, o,p’-DDA does not influence the mobility of lipids.

4. Discussion

We carried out a comprehensive biophysical study to charac- terize the interaction of mitotane and o,p’-DDA with lipid mem- branes and the impact of both drugs on membrane structure and dynamics. The results clearly show that both molecules insert into membranes. The membrane embedding of mitotane causes a slight increase in membrane packing density and leads to a small

immobilization of the membrane lipids depending on the compo- sition of the membrane. However, the interaction of mitotane with membranes mediates a disturbance of bilayer structure in that the permeation of polar molecules is increased. This impact of mitotane is observed in the presence of PE and/or cholesterol in the mem- brane. In contrast, o,p’-DDA shows no impact on membranes of either composition.

Most studies on the mechanism of action of mitotane have focused on cellular events like changes of gene expression or apoptosis (Hescot et al., 2013; Sbiera et al., 2015). So far, only a single paper dealt with the interaction of mitotane with mem- branes (Jacobi et al., 2014). We hypothesize that - due to the high serum concentrations clinically achieved - direct interaction with cellular membranes is a prerequisite for cellular mitotane action.

Chemically, mitotane and o,p’-DDA, are characterized by an ar- omatic ring structure, which additionally features some polar character. The isosurfaces of the electrical potential for the two molecules are shown in Fig. 1. The dipolar moments amount to 3.46 D (mitotane) and 4.85 D (o,p’-DDA) and point for both molecules towards the polar group between the aromatic rings. In order to

Fig. 10. Influence of mitotane on the mobility of spin-labeled lipids in LUV membranes. LUVs were prepared consisting of different lipids and spin-labeled lipids (see the labels on the spectra). Spectra of spin-labeled lipids were recorded immediately (i.e. after about 1 min) (left panel) or 10 min (right panel) after mixing vesicles (final lipid and label concentration: 4.56 mM and 0.19 mM, respectively) with HBS (black lines) or mitotane (final concentration: 0.5 mM) (blue lines) at 25 ℃. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

10 G

DOPC / SL-PC

DOPC / DOPE / SL-PE

DOPC / DOPE / Chol / SL-PE

DOPC / Chol / SL-Chol

quantify the hydrophobicity, octanol-water partition coefficients (logP) were calculated using the Molinspiration program (http:// www.molinspiration.com/). These values (6.89 for mitotane; 4.50 for o,p’-DDA) indicate for both molecules a comparatively high hydrophobicity. Due to these structural characteristics, an incor- poration of mitotane and o,p’-DDA into membranes can be assumed.

In line with this, by using NMR spectroscopy we found that both molecules bind to lipid membranes and incorporate very similarly into the upper part of the membrane, where the fatty acyl chains are linked to the glycerol backbone. This region is usually referred to as the lipid/water interface, which represents a rough and broad interface that provides numerous molecular groups for polar and apolar interactions (White et al., 2001). Notably, with regard to the different impact of mitotane and o,p’-DDA on membrane structure (see below), a similar localization of both molecules was also observed in POPC/POPE membranes, with a small bias of the dis- tribution function in direction towards the lipid head groups for mitotane. The line shapes of NMR signals indicate for both mole- cules a high reorientation rate within the membrane.

The membrane binding of mitotane and o,p’-DDA influences membrane structure and dynamics differently. The lipid chain or- der deduced from 2H NMR reflects slight changes in the presence of both molecules. While in pure POPC membranes, mitotane caused some increase of the order parameter in the upper chain region, o,p’-DDA induced a decrease in the middle and lower chain region. A similar disordering effect of o,p’-DDA on lipid chains was also observed in POPC/POPE membranes with similar disordering of both POPC-d31 and POPE-d31. In contrast, mitotane ordered the POPC component while it exerted a disordering effect on the POPE component. Although both molecules showed a similar distribution

function in POPC membranes, the order parameters may suggest that the more polar o,p’-DDA can preferentially interact with the upper chain region, which leads to some larger motional freedom for the lower chain segments. Furthermore, o,p’-DDA may also hydrogen bond to the PE head group, which constraints the molecule to the lipid water interface providing room for larger amplitude motions in the lower chain region. We acknowledge that these experiments were performed at rather high drug concen- trations, which were required to obtain sufficiently large NMR signals.

The ESR measurements also revealed an influence on membrane dynamics in particular for mitotane. The impact of mitotane on the rotational mobility of lipids depended on the membrane lipid composition and the used spin-labeled analog. A broadening of the ESR signals was especially observed in membranes containing PE or PE/Chol and using SL-PE as reporter revealing that mitotane im- mobilizes the lipids at these conditions. Similarly, there is an obvious role of cholesterol for mediating the influence of mitotane on membrane dynamics as seen from the ESR spectra in DOPC/Chol membranes. Whereas the ESR spectrum of SL-PC reflected no in- fluence of mitotane, that of SL-Chol showed a signal broadening. In contrast to mitotane, at all conditions investigated, o,p’-DDA had no influence on lipid mobility since the ESR spectra recorded in the presence of o,p’-DDA were nearly identical to the control spectra. We note that the different results obtained with ESR and NMR methods do not represent a contradiction since both methods are sensitive to different timescales with NMR typically probing somewhat slower motions.

The presence of PE and/or cholesterol in the membrane triggers a disturbance of bilayer morphology upon mitotane addition as we demonstrated by two approaches. Mitotane caused an increasing

reduction of fluorescent and spin-labeled lipid analogs localized in the inner membrane leaflet, especially in membranes containing PE and or cholesterol indicating an increased permeation of dithionite or ascorbate, respectively. We note, that the measured curves can also be explained by an increase of lipid flip-flop in the presence of mitotane. However, in any case, they reflect a drug-mediated membrane disturbance in that the transbilayer permeation of a polar molecule (dithionite, ascorbate) and/or part of a molecule (phospholipid head group) is increased. Notably, o,p’-DDA had no influence on the reduction kinetics at any condition, which agrees with the fact that this molecule had no effect on lipid mobility. It is noteworthy that the impact of mitotane but not o,p’-DDA on membrane integrity correlates with their respective biological ef- ficacy. O,p’-DDA has been demonstrated not to be active as mito- tane (Hescot et al., 2014). In preliminary experiments we have investigated another metabolite of mitotane, 1,1-(o,p’-dichlor- odiphenyl)-2,2 dichloroethene (o,p’-DDE), which also has no bio- logical activity like mitotane. We found, that o,p’-DDE, like o,p’- DDA, did not influence the reduction kinetics of dithionite in lipid membranes (data not shown) underlining the physiological rele- vance of the membrane perturbing impact observed for mitotane.

ESR experiments using lipid analogs dissolved in buffer exclude a direct interaction between mitotane and SL-PE (see Supplemen- tary Material, Fig. S3), indicating that a bilayer structure is required for the impact of mitotane. To explain the perturbing effect of mitotane, several mechanisms can be considered, which have also been proposed for describing the action of membrane per- meabilizing peptides (Shai, 1999; Zhang et al., 2001). First, it could be hypothesized that mitotane mainly inserts into the outer membrane layer, by which the outer leaflet area would be increased. Such an imbalance in the surface area may cause a perturbation of membrane structure leading e.g. to a leakage of the content of liposomes (Castano et al., 2000). However, small mole- cules are also subject to fast flip flop motions across the bilayer such that a large imbalance in the drug concentration between the two membrane leaflets appears unlikely. Second, the interaction of mitotane with specific lipids may disturb their structure and dy- namics and form local defects in the bilayer (Cheng et al., 2009). Third, mitotane could form local pores or channel-like structures within the membrane similar to peptide-mediated ‘barrel-stave’ mechanism (Matsuzaki, 1998). Since the NMR measurements indicate a preferential interaction of mitotane with the upper fatty acyl chain region, we suggest that this mechanism is less likely. In that case, i.e. a formation of transmembrane pores by mitotane, an interaction of the molecule also with the hydrophobic part of the fatty acyl chains should be observed. At the present stage of investigation, the process(es) triggered by mitotane in the mem- brane and the reasons for the behavior of mitotane and o,p’-DDA regarding their different influence on lipid mobility and membrane integrity despite similar binding properties, cannot be answered unequivocally. More experiments have to be performed on a mo- lecular level to answer these questions.

What is the relevance of our data for understanding the physi- ological effects of mitotane? We have demonstrated previously that mitotane is rapidly taken up into ACC cells (Sbiera et al., 2015). One study has investigated the interaction of mitotane with red blood cells finding cell hemolysis at a drug concentration of about 35 uM (Jacobi et al., 2014). However, the authors provided no information on the number of cells in their experiments in order to calculate the effective molar ratio of mitotane to membrane lipids. The concen- tration of 35 µM (11.5 mg/l) is well within the range of clinically used drug concentrations and hence the effective concentration may have been higher. Since PE is mainly localized in the inner membrane leaflet of plasma membranes (Zachowski, 1993), it is questionable whether interaction of mitotane with PE containing

membranes contributes to the effects in erythrocytes. In contrast, cholesterol is assumed to be symmetrically distributed between both membrane leaflets (Müller et al., 2011) and could mediate an impact of mitotane on red blood cells and putatively nucleated cells. Of course, our results on lipid membranes represent a first step in deciphering the interaction of mitotane with biological membranes, which will be probably more complex since biological membranes are much more heterogeneous in composition compared with the model membranes used here. So far, no data have been published investigating the interaction of mitotane with plasma membrane or organelle membranes of other (nucleated) cells. A propensity of mitotane to bind to biological membranes can be assumed from experiments finding a radioactive analog of the drug, after cellular uptake by ACC cells on the time scale of seconds, retrieved in organelle membranes whereas only a small amount retrieved in the cytosol (Sbiera et al., 2015), but previous reports suggest also binding to other membrane macromolecules (Gagliano et al., 2014).

Recently we discovered that mitotane directly inhibits SOAT1, which is localized in MAM (Sbiera et al., 2015). The endoplasmic reticulum is relatively rich in PE and contains only a small amount of cholesterol (van Meer et al., 2008). PE seems to be symmetrically distributed in the endoplasmic reticulum due to its rapid trans- bilayer movement (Herrmann et al., 1990; Marx et al., 2000). Therefore, deduced from the results of our study, any impact of mitotane on these organelle membranes should be realized via an interaction with PE. Importantly, this opens two additional mech- anistic possibilities: First, mitotane bound to PE-rich MAMs may modify activities of target enzymes such as SOAT1 by influencing the physico-chemical properties of the membrane as observed for many membrane anchored enzymes (Peters and Bywater, 2001; Rujiviphat et al., 2009; Nasr et al., 2013; Qin et al., 2013; Zhou et al., 2014; Pignataro et al., 2015). Such an influence may happen e.g. by changes of membrane fluidity, transbilayer lipid distribution, or lateral lipid arrangement. For reconstituted Sterol-O-Acyl Transferase, it has been shown that the activity of inhibitors de- pends on the lipid content of the surrounding membrane (Harte et al., 1995). It is conceivable, that mitotane interacts directly or indirectly with other MAM-associated proteins such as FATE1 (Doghman et al., 2014) or influences mitochondrial cholesterol import (Prasad et al., 2015). Second, we hypothesize that mem- brane disturbance per se may contribute to triggering of down- stream pathways such as ER-stress which might be caused for instance by increasing permeability of the ER-membrane for cal- cium. The lipid composition of adrenocortical cells to our knowl- edge has not been studied in detail. At the level of cell lines, NCI- H295 cells are characterized by a relatively high abundance of sphingomyelin compared to non-steroidogenic cells (Sbiera et al., 2015). Obviously, any purification of organelles or even sub- structures such as MAMs requires extensive validation by marker proteins and/or electron microscopy. Preliminary data from our lab using microsomal preparations from NCI-H295 cells indicate a relatively higher abundance of several lipid species (including PE and cholesterol) in the ER compared to whole cells (see Supple- mentary Material, Fig. S4). This, however, has to be clarified in detail in the future.

Future studies are required to investigate (i) the interaction of mitotane with plasma membranes, (ii) the physiological influence of mitotane on membranes of isolated organelles and in the cellular context, and (iii) the differences of mitotane and o,p’-DDA impact on membranes. In conclusion, our study shows that mitotane binds to lipid membranes and this binding modifies the physico-chemical properties of the membrane. The effects depend on the membrane lipid composition underlining a particular role of PE and choles- terol. The data should contribute to a better understanding of

H.A. Scheidt et al. / Molecular and Cellular Endocrinology xxx (2016) 1-14

mitotane effectiveness in ACC in that also changes in membrane structure and dynamics have to be considered.

Acknowledgments

This study was supported by grants of the Deutsche For- schungsgemeinschaft (KR 4371/1-1 to M.K. and Grant FA 466/4-1 to Martin Fassnacht). We thank Sabine Schiller (Humboldt University Berlin) for technical assistance.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.mce.2016.03.022.

References

Alfonsi, R., Attivi, D., Astier, A., Socha, M., Morice, S., Gibaud, S., 2013. Character- ization of mitotane (o,p’-DDD) - cyclodextrin inclusion complexes: phase- solubility method and NMR. Ann. Pharm. Fr. 71, 186-192.

Bevers, E.M., Comfurius, P., Dekkers, D.W.C., Zwaal, R.F.A., 1999. Lipid translocation across the plasma membrane of mammalian cells. Biochim. Biophys. Acta 1439, 317-330.

Bevers, E.M., Williamson, P.L., 2010. Phospholipid scramblase: an update. FEBS Lett. 584, 2724-2730.

Castano, S., Desbat, B., Dufourcq, J., 2000. Ideally amphipathic ß-sheeted peptides at interfaces: structure, orientation, affinities for lipids and hemolytic activity of (KL)mK peptides. Biochim. Biophys. Acta 1463, 65-80.

Cheng, J.T.J., Hale, J.D., Elliot, M., Hancock, R.E.W., Straus, S.K., 2009. Effect of membrane composition on antimicrobial peptides aurein 2.2 and 2.3 from australian southern bell frogs. Biophys. J. 96, 552-565.

Davis, J.H., 1983. The description of membrane lipid conformation, order and dy- namics by 2H-NMR. Biochim. Biophys. Acta 737, 117-171.

Davis, J.H., Jeffrey, K.R., Bloom, M., Valic, M.I., Higgs, T.P., 1976. Quadrupolar echo deuteron magnetic resonance spectroscopy in ordered hydrocarbon chains. Chem. Phys. Lett. 42, 390-394.

De Francia, S., Ardito, A., Daffara, F., Zaggia, B., Germano, A., Berruti, A., Di Carlo, F., 2012. Mitotane treatment for adrenocortical carcinoma: an overview. Minerva Endocrinol. 37, 9-23.

Doghman, M., Sbiera, S., Granatiero, V., Rizzuto, R., Fassnacht, M., Lalli, E., 2014. Role of FATE1, a Novel Steroidogenic Factor-1 Target, in the Regulation of Mito- chondrial Morphology and Function in Human Adrenocortical Cells, in: Adrenal Tumors: ACC & Adrenal Incidentaloma. Endocrine Society pp SAT-0786.

Else, T., Kim, A.C., Sabolch, A., Raymond, V.M., Kandathil, A., Caoili, E.M., Jolly, S., Miller, B.S., Giordano, T.J., Hammer, G.D., 2014. Adrenocortical carcinoma. Endocr. Rev. 35, 282-326.

Fassnacht, M., Kroiss, M., Allolio, B., 2013. Update in adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 98, 4551-4564.

Fassnacht, M., Libe, R., Kroiss, M., Allolio, B., 2011. Adrenocortical carcinoma: a cli- nician’s update. Nat. Rev. Endocrinol. 7, 323-335.

Fellmann, P., Zachowski, A., Devaux, P.F., 1994. Synthesis and use of spin-labeled lipids for studies of the transmembrane movement of phospholipids. In: Graham, J.M., Higgins, J.A. (Eds.), Biomembrane Protocols: II. Architecture and Function. Humana Press Inc., Totowa, NJ, pp. 161-175.

Gagliano, T., Gentilin, E., Benfini, K., Di Pasquale, C., Tassinari, M., Falletta, S., Feo, C., Tagliati, F., Uberti, E.D., Zatelli, M.C., 2014. Mitotane enhances doxorubicin cytotoxic activity by inhibiting P-gp in human adrenocortical carcinoma cells. Endocrine 47, 943-951.

Golden, S.H., Robinson, K.A., Saldanha, I., Anton, B., Ladenson, P.W., 2009. Clinical review: prevalence and incidence of endocrine and metabolic disorders in the United States: a comprehensive review. J. Clin. Endocrinol. Metab. 94, 1853-1878.

Greube, A., Müller, K., Töpfer-Petersen, E., Herrmann, A., Müller, P., 2001. Influence of the bovine seminal plasma protein PDC-109 on the physical state of mem- branes. Biochemistry 40, 8326-8334.

Hahner, S., Fassnacht, M., 2005. Mitotane for adrenocortical carcinoma treatment. Curr. Opin. Investig. Drugs 6, 386-394.

Harte, R.A., Yeaman, S.J., Jackson, B., Suckling, K.E., 1995. Effect of membrane environment on inhibition of acyl-CoA:cholesterol acyltransferase by a range of synthetic inhibitors. Biochim. Biophys. Acta 1258, 241-250.

Hermsen, I.G., Fassnacht, M., Terzolo, M., Houterman, S., den Hartigh, J., Leboulleux, S., Daffara, F., Berruti, A., Chadarevian, R., Schlumberger, M., Allolio, B., Haak, H.R., Baudin, E., 2011. Plasma concentrations of o,p’DDD, o,p’DDA, and o,p’DDE as predictors of tumor response to mitotane in adreno- cortical carcinoma: results of a retrospective ENS&T multicenter study. J. Clin. Endocrinol. Metab. 96, 1844-1851.

Herrmann, A., Zachowski, A., Devaux, P.F., 1990. Protein-mediated phospholipid translocation in the endoplasmic reticulum with a low lipid specificity. Biochemistry 29, 2023-2027.

Hescot, S., Paci, A., Seck, A., Slama, A., Viengchareun, S., Trabado, S., Brailly-

Tabard, S., Al Ghuzlan, A., Young, J., Baudin, E., Lombes, M., 2014. The lack of antitumor effects of o,p’DDA excludes its role as an active metabolite of mito- tane for adrenocortical carcinoma treatment. Horm. Cancer 5, 312-323.

Hescot, S., Seck, A., Guerin, M., Cockenpot, F., Huby, T., Broutin, S., Young, J., Paci, A., Baudin, E., Lombes, M., 2015. Lipoprotein-free mitotane exerts high cytotoxic activity in adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 100, 2890-2898.

Hescot, S., Slama, A., Lombes, A., Paci, A., Remy, H., Leboulleux, S., Chadarevian, R., Trabado, S., Amazit, L., Young, J., Baudin, E., Lombes, M., 2013. Mitotane alters mitochondrial respiratory chain activity by inducing cytochrome c oxidase defect in human adrenocortical cells. Endocr. Relat. Cancer 20, 371-381.

Huster, D., Arnold, K., Gawrisch, K., 1999. Investigation of lipid organization in biological membranes by two-dimensional nuclear overhauser enhancement spectroscopy. J. Phys. Chem. B 103, 243-251.

Huster, D., Muller, P., Arnold, K., Herrmann, A., 2001. Dynamics of membrane penetration of the fluorescent 7-nitrobenz-2-oxa-1,3-diazol-4-yl (NBD) group attached to an acyl chain of phosphatidylcholine. Biophys. J. 80, 822-831.

Igaz, P., Tombol, Z., Szabo, P.M., Liko, I., Racz, K., 2008. Steroid biosynthesis inhibitors in the therapy of hypercortisolism: theory and practice. Curr. Med. Chem. 15, 2734-2747.

Jacobi, J., Lang, E., Bissinger, R., Frauenfeld, L., Modicano, P., Faggio, C., Abed, M., Lang, F., 2014. Stimulation of erythrocyte cell membrane scrambling by mito- tane. Cell. Physiol. Biochem. 33, 1516-1526.

Jeener, J., Meier, B.H., Bachmann, P., Ernst, R.R., 1979. Investigation of exchange processes by two-dimensional NMR spectroscopy. J. Chem. Phys. 71, 4546-4553.

Jensen, B.L., Caldwell, M.W., French, L.G., Briggs, D.G., 1987. Toxicity, ultrastructural effects, and metabolic studies with 1-(o-chlorophenyl)-1-(p-chlorophenyl)-2,2- dichloroethane(o,p’-DDD) and its methyl analog in the guinea pig and rat. Toxicol. Appl. Pharmacol. 87, 1-9.

Jurowich, C., Fassnacht, M., Kroiss, M., Deutschbein, T., Germer, C.T., Reibetanz, J., 2013. Is there a role for laparoscopic adrenalectomy in patients with suspected adrenocortical carcinoma? A critical appraisal of the literature. Horm. Metab. Res. 45, 130-136.

Kroiss, M., Quinkler, M., Lutz, W.K., Allolio, B., Fassnacht, M., 2011. Drug interactions with mitotane by induction of CYP3A4 metabolism in the clinical management of adrenocortical carcinoma. Clin. Endocrinol. 75, 585-591.

Kroiss, M., Plonné, D., Kendl, S., Schirmer, D., Ronchi, C.L., Schirbel, A., Zink, M., Lapa, C., Klinker, H., Fassnacht, M., Heinz, W., Sbiera, S., 2016. Association of mitotane with chylomicrons and serum lipoproteins: practical implications for treatment of adrenocortical carcinoma. Eur. J. Endocrinol. 174, 343-353.

Kruger, R., Uhl, U.J., Voigt, K.H., 1984. Fine structural effects of a DDT derivative on a rat adrenocortical carcinoma. Histochem. J. 16, 406-408.

Lafleur, M., Fine, B., Sternin, E., Cullis, P.R., Bloom, M., 1989. Smoothed orientational order profile of lipid bilayers by 2H-nuclear magnetic resonance. Biophys. J. 56, 1037-1041.

Langner, M., Hui, S.W., 1993. Dithionite penetration through phospholipid bilayers as a measure of defects in lipid molecular packing. Chem. Phys. Lipids 65, 23-30.

Lehmann, T.P., Wrzesinski, T., Jagodzinski, P.P., 2013. The effect of mitotane on viability, steroidogenesis and gene expression in NCIH295R adrenocortical cells. Mol. Med. Rep. 7, 893-900.

Marsh, D., Horvath, L.I., 1998. Structure, dynamics and composition of the lipid- protein interface. Perspectives from spin-labelling. Biochim. Biophys. Acta 1376, 267-296.

Marx, U., Lassmann, G., Holzhütter, H.G., Wüstner, D., Müller, P., Höhlig, A., Kubelt, J., Herrmann, A., 2000. Rapid flip-flop of phospholipids in endoplasmic reticulum membranes studied by a stopped-flow approach. Biophys. J. 78, 2628-2640.

Matsuzaki, K., 1998. Magainins as paradigm for the mode of action of pore forming polypeptides. Biochim. Biophys. Acta 1376, 391-400.

Maurin, L., Morin, P., Bienvenue, A., 1987. A new paramagnetic analogue of cholesterol as a tool for studying molecular interactions of genuine cholesterol. Biochim. Biophys. Acta 900, 239-248.

Mayer, L.D., Hope, M.J., Cullis, R.P., Janoff, A.S., 1985. Solute distributions and trap- ping efficiencies observed in freeze-thawed multilamellar vesicles. Biochim. Biophys. Acta 817, 193-196.

Mcintyre, J.C., Sleight, R.G., 1991. Fluorescence assay for phospholipid membrane asymmetry. Biochemistry 30, 11819-11827.

Moore, R.N., Penney, D.P., Averill, K.T., 1980. Fine structural and biochemical effects of aminoglutethimide and o,p’-DDD on rat adrenocortical carcinoma 494 and adrenals. Anat. Rec. 198, 113-124.

Müller, P., Plazzo, A.P., Herrmann, A., 2011. Transbilayer movement and distribution of cholesterol. In: Devaux, P.F., Herrmann, A. (Eds.), Transbilayer Dynamics of Lipids. John Wiley & Sons, Inc, Hoboken, New Jersey, pp. 75-96.

Nasr, M.L., Shi, X., Bowman, A.L., Johnson, M., Zvonok, N., Janero, D.R., Vemuri, V.K., Wales, T.E., Engen, J.R., Makriyannis, A., 2013. Membrane phospholipid bilayer as a determinant of monoacylglycerol lipase kinetic profile and conformational repertoire. Protein Sci. A Publ. Protein Soc. 22, 774-787.

Peters, G.H., Bywater, R.P., 2001. Influence of a lipid interface on protein dynamics in a fungal lipase. Biophys. J. 81, 3052-3065.

Pignataro, M.F., Dodes-Traian, M.M., González-Flecha, F.L., Sica, M., Mangialavori, I.C., Rossi, J.P.F.C., 2015. Modulation of plasma membrane Ca2+- ATPase by neutral phospholipids: effect of the micelle-vesicle transition and the bilayer thickness. J. Biol. Chem. 290, 6179-6190.

Poli, G., Guasti, D., Rapizzi, E., Fucci, R., Canu, L., Bandini, A., Cini, N., Bani, D.,

H.A. Scheidt et al. / Molecular and Cellular Endocrinology xxx (2016) 1-14

Mannelli, M., Luconi, M., 2013. Morphofunctional effects of mitotane on mito- chondria in human adrenocortical cancer cells. Endocr. Relat. Cancer 20, 537-550.

Pomorski, T., Herrmann, A., Zachowski, A., Devaux, P.F., Müller, P., 1994. Rapid determination of the transbilayer distribution of NBD-phospholipids in eryth- rocyte membranes with dithionite. Mol. Membr. Biol. 11, 39-44.

Prasad, M., Kaur, J., Pawlak, K.J., Bose, M., Whittal, R.M., Bose, H.S., 2015. Mito- chondria-associated endoplasmic reticulum membrane (MAM) regulates ste- roidogenic activity via steroidogenic acute regulatory protein (StAR)-voltage- dependent anion channel 2 (VDAC2) interaction. J. Biol. Chem. 290, 2604-2616. Qin, S .- S., Yu, Y .- X., Li, Q .- K., Yu, Z .- W., 2013. Interaction of human synovial phos- pholipase A2 with mixed lipid bilayers: a coarse-grain and all-atom molecular dynamics simulation study. Biochemistry 52, 1477-1489.

Rujiviphat, J., Meglei, G., Rubinstein, J.L., McQuibban, G.A., 2009. Phospholipid as- sociation is essential for dynamin-related protein Mgm1 to function in mito- chondrial membrane fusion. J. Biol. Chem. 284, 28682-28686.

Sbiera, S., Leich, E., Liebisch, G., Sbiera, I., Schirbel, A., Wiemer, L., Matysik, S., Eckhardt, C., Gardill, F., Gehl, A., Kendl, S., Weigand, I., Bala, M., Ronchi, C.L., Deutschbein, T., Schmitz, G., Rosenwald, A., Allolio, B., Fassnacht, M., Kroiss, M., 2015. Mitotane Inhibits sterol-O-acyl transferase 1 triggering lipid-mediated endoplasmic reticulum stress and apoptosis in adrenocortical carcinoma cells. Endocrinology 156, 3895-3908.

Scheidt, H.A., Huster, D., 2008. The Interaction of small molecules with phospho- lipid membranes studied by 1H NOESY NMR under Magic-Angle Spinning. Acta Pharmacol. Sin. 29, 35-49.

Scheidt, H.A., Meyer, T., Nikolaus, J., Baek, D.J., Haralampiev, I., Thomas, L., Bittman, R., Müller, P., Herrmann, A., Huster, D., 2013. Cholesterol’s aliphatic side chain modulates membrane properties. Angew. Chem. Int. Ed. 52, 12848-12851.

Scheidt, H.A., Muller, P., Herrmann, A., Huster, D., 2003. The potential of fluorescent and spin-labeled steroid analogs to mimic natural cholesterol. J. Biol. Chem. 278, 45563-45569.

Scheidt, H.A., Pampel, A., Nissler, L., Gebhardt, R., Huster, D., 2004. Investigation of the membrane localization and distribution of flavonoids by high-resolution magic angle spinning NMR spectroscopy. Biochim. Biophys. Acta 1663, 97-107.

Schteingart, D.E., Doherty, G.M., Gauger, P.G., Giordano, T.J., Hammer, G.D., Korobkin, M., Worden, F.P., 2005. Management of patients with adrenal cancer: recommendations of an international consensus conference. Endocr. Relat. Cancer 12, 667-680.

Shai, Y., 1999. Mechanism of the binding, insertion and destabilization of

phospholipid bilayer membranes by o .- helical antimicrobial and cell non- selective membrane-lytic peptides. Biochim. Biophys. Acta 1462, 55-70.

Stamm, H., Jaeckel, H., 1989. Relative ring current effects based on a new model for aromatic-solvent-induced shift. J. Am. Chem. Soc. 111, 6544-6550.

Sternin, E., Bloom, M., Mackay, A.L., 1983. De-pake-ing of NMR spectra. J. Magn. Reson. 55, 274-282.

Stigliano, A., Cerquetti, L., Borro, M., Gentile, G., Bucci, B., Misiti, S., Piergrossi, P., Brunetti, E., Simmaco, M., Toscano, V., 2008. Modulation of proteomic profile in H295R adrenocortical cell line induced by mitotane. Endocr. Relat. Cancer 15, 1-10.

Tannert, A., Topfer-Petersen, E., Herrmann, A., Müller, K., Müller, P., 2007. The lipid composition modulates the influence of the bovine seminal plasma protein PDC-109 on membrane stability. Biochemistry 46, 11621-11629.

Terzolo, M., Angeli, A., Fassnacht, M., Daffara, F., Tauchmanova, L., Conton, P.A., Rossetto, R., Buci, L., Sperone, P., Grossrubatscher, E., Reimondo, G., Bollito, E., Papotti, M., Saeger, W., Hahner, S., Koschker, A.C., Arvat, E., Ambrosi, B., Loli, P., Lombardi, G., Mannelli, M., Bruzzi, P., Mantero, F., Allolio, B., Dogliotti, L., Berruti, A., 2007. Adjuvant mitotane treatment for adrenocortical carcinoma. N. Engl. J. Med. 356, 2372-2380.

Terzolo, M., Baudin, A.E., Ardito, A., Kroiss, M., Leboulleux, S., Daffara, F., Perotti, P., Feelders, R.A., deVries, J.H., Zaggia, B., De Francia, S., Volante, M., Haak, H.R., Allolio, B., Al Ghuzlan, A., Fassnacht, M., Berruti, A., 2013. Mitotane levels predict the outcome of patients with adrenocortical carcinoma treated adjuvantly following radical resection. Eur. J. Endocrinol. 169, 263-270.

van Meer, G., Voelker, D.R., Feigenson, G.W., 2008. Membrane lipids: where they are and how they behave. Nat. Rev. Mol. Cell Biol. 9, 112-124.

Weizenmann, N., Huster, D., Scheidt, H.A., 2012. Interaction of local anesthetics with lipid bilayers investigated by 1H MAS NMR spectroscopy. Biochim. Biophys. Acta 1818, 3010-3018.

White, S.H., Ladokhin, A.S., Jayasinghe, S., Hristova, K., 2001. How membranes shape protein structure. J. Biol. Chem. 276, 32395-32398.

Yau, W .- M., Wimley, W.C., Gawrisch, K., White, S.H., 1998. The preference of tryp- tophan for membrane interfaces. Biochemistry 37, 14713-14718.

Zachowski, A., 1993. Phospholipids in animal eukaryotic membranes: transverse asymmetry and movement. Biochem. J. 294, 1-14.

Zhang, L., Rozek, A., Hancock, R.E., 2001. Interaction of cationic antimicrobial pep- tides with model membranes. J. Biol. Chem. 276, 35714-35722.

Zhou, Q., Li, J., Yu, H., Zhai, Y., Gao, Z., Liu, Y., Pang, X., Zhang, L., Schulten, K., Sun, F., Chen, C., 2014. Molecular insights into the membrane-associated phosphati- dylinositol 4-kinase lio. Nat. Commun. 5, 3552-3561.

Quartz 4

Explorer

27002491

Molecular and Cellular Endocrinology

The adrenal specific toxicant mitotane directly interacts with lipid membranes and alters membrane properties depending on lipid composition

ABSTRACT

1. Introduction

Abbreviations

2. Materials and methods

2.1. Materials

2.2. Preparation of large unilamellar vesicles (LUVs)

2.3. Sample preparation for NMR experiments

2.4. 2H NMR measurements

2.5. 1H MAS NMR spectroscopy

2.6. Reduction of spin-labeled phospholipids by ascorbate

2.7. Reduction of fluorescent phospholipids by dithionite

3. Results

3.1. Interaction of mitotane with membranes measured by 1H MAS NMR spectroscopy

3.2. Impact of mitotane and o,p’-DDA on membrane properties

ARTICLE IN PRESS

3.3. Influence of mitotane and o,p’-DDA on membrane structure of LUV membranes measured by ESR spectroscopy

3.4. Influence of mitotane and o,p’-DDA on the membrane integrity of LUV membranes measured by fluorescence

ARTICLE IN PRESS

3.5. Influence of mitotane and o,p’-DDA on the mobility of spin- labeled phospholipids in LUV membranes

4. Discussion

H.A. Scheidt et al. / Molecular and Cellular Endocrinology xxx (2016) 1-14

Acknowledgments

Appendix A. Supplementary data

References

H.A. Scheidt et al. / Molecular and Cellular Endocrinology xxx (2016) 1-14

Graph View

Table of Contents

Backlinks