Neuropharmacology 53 (2007) 318e329 www.elsevier.com/locate/neuropharm
Intracellular cholesterol changes induced by translocator protein (18 kDa) TSPO/PBR ligands Angela Maria Falchi a, Barbara Battetta b, Francesca Sanna b, Marco Piludu a, Valeria Sogos a, Mariangela Serra c, Marta Melis a, Martina Putzolu a, Giacomo Diaz a,* a
Department of Cytomorphology, Cittadella Universitaria, University of Cagliari, Monserrato, 09100 Cagliari, Italy Department of Biomedical Science & Biotechnology, Cittadella Universitaria, University of Cagliari, 09100 Cagliari, Italy c Department of Experimental Biology & Center of Excellence for Neurobiology of Drug Dependence, Cittadella Universitaria, University of Cagliari, 09100 Cagliari, Italy b
Received 10 April 2006; received in revised form 4 May 2007; accepted 20 May 2007
Abstract One of the main functions of the translocator protein (18 kDa) or TSPO, previously known as peripheral-type benzodiazepine receptor, is the regulation of cholesterol import into mitochondria for steroid biosynthesis. In this paper we show that TSPO ligands induce changes in the distribution of intracellular cholesterol in astrocytes and fibroblasts. NBD-cholesterol, a fluorescent analog of cholesterol, was rapidly removed from membranes and accumulated into lipid droplets. This change was followed by a block of cholesterol esterification, but not by modification of intracellular cholesterol synthesis. NBD-cholesterol droplets were in part released in the medium, and increased cholesterol efflux was observed in [3H]cholesterol-prelabeled cells. TSPO ligands also induced a prominent shrinkage and depolarization of mitochondria and depletion of acidic vesicles with cytoplasmic acidification. Consistent with NBD-cholesterol changes, MTT assay showed enhanced accumulation of formazan into lipid droplets and inhibition of formazan exocytosis after treatment with TSPO ligands. The effects of specific TSPO ligands PK 11195 and Ro5-4864 were reproduced by diazepam, which binds with high affinity both TSPO and central benzodiazepine receptors, but not by clonazepam, which binds exclusively to GABA receptor, and other amphiphilic substances such as DIDS and propranolol. All these effects and the parallel immunocytochemical detection of TSPO in potentially steroidogenic cells (astrocytes) and non-steroidogenic cells (fibroblasts) suggest that TSPO is involved in the regulation and trafficking of intracellular cholesterol by means of mechanisms not necessarily related to steroid biosynthesis. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: TSPO; Translocator protein (18 kDa); PBR; Cholesterol; Mitochondria; Lipid droplets; MTT; Acidic vesicles; Immunocytochemistry
1. Introduction One of the most extensively characterized functions of the translocator protein (18 kDa) or TSPO (Papadopoulos et al., 2006), previously known as peripheral-type benzodiazepine receptor (PBR), is the mitochondrial import of cholesterol (Hauet et al., 2005; Miller, 1988; Simpson and Waterman, 1983) delivered to mitochondria by cholesterol-transfer
* Corresponding author. Tel.: þ39 070 6754081; fax: þ39 070 675 4003. E-mail address:
[email protected] (G. Diaz). 0028-3908/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2007.05.016
proteins (StAR, MLN64 and, less specifically, SCP-2) (Chanderbhan et al., 1982; Clark et al., 1994; Watari et al., 1997). This represents the first and rate-determining step in steroid hormone biosynthesis. Accordingly, the localization, structure and pharmacology of TSPO and the effects induced by TSPO ligands have been widely investigated in steroidogenic cells (for review see Casellas et al., 2002; Gavish et al., 1999; Papadopoulos, 2004). On the other hand, effects of TSPO ligands have also been found in non-steroidogenic cells, affecting different physiological mechanisms such as respiration in heart, kidney and liver (Moreno-Sa`nchez et al., 1991; Veenman and Gavish, 2006), generation of reactive oxygen species in
A.M. Falchi et al. / Neuropharmacology 53 (2007) 318e329
neurons (Jayakumar et al., 2002) and HL60 human leukemia cells (Fennell et al., 2001), anion transport in kidney (Basile et al., 1988), mitochondrial permeability transition in cardiomyocytes (Chelli et al., 2001), inhibition of cell proliferation in human fibroblasts and fibrosarcoma cells (Kletsas et al., 2004) and apoptosis in various cell lines (Chelli et al., 2004; Decaudin et al., 2002). Increased tumorigenicity has been found in TSPO antisense knockdown cells (Levin et al., 2005; Weisinger et al., 2004). Many of these effects have been attributed to the intimate association of TSPO with a complex of proteins present in mitochondrial membranes, including the voltage-dependent anion channel (VDAC), the adenine nucleotide transporter (McEnery et al., 1992), and the mitochondrial transition pore complex (Kroemer et al., 1997). However, in some cases, effects of TSPO ligands have been observed also in TSPO-deficient (Hans et al., 2005) and TSPO-knockdown cells (Gonzalez-Polo et al., 2005; Kletsas et al., 2004), raising the issue of the possible presence of TSPO-independent mechanisms of actions. One of these mechanisms may rely on the inhibition of the mitochondrial inner membrane anionic channel (IMAC) exerted by several TSPO ligands, namely Ro5-4864 (a 40 -chloro derivative of diazepam) and PK 11195 (an isoquinoline carboxamide derivative) (Beavis, 1989, 1992). This hypothesis has recently been supported by data obtained by Aon et al. (2003) in cardiomyocytes where mitochondrial potential
319
oscillations were prevented by PK 11195, Ro5-4864 but also by DIDS, a stilbene-2,20 -disulfonate which inhibits IMAC. In the present study we investigated the acute effect of TSPO ligands on the intracellular distribution of cholesterol, using the fluorescent analog NBD-cholesterol, and the concomitant effects on mitochondria and acidic vesicles, using potential-sensitive (TMRM) and pH-sensitive (acridine orange) probes. We also evaluated cholesterol synthesis, esterification and efflux. Our findings indicate consistent changes in the distribution, esterification and efflux of intracellular cholesterol, accompanied by mitochondrial depolarization and depletion of acidic vesicles. These changes were observed in different cell types such as rat astrocytes, fibroblasts and 3T3 mouse fibroblasts, which represent potentially steroidogenic and non-steroidogenic cells. This suggests that cholesterol changes induced by TSPO ligands are not necessarily related to steroid biosynthesis. 2. Materials and methods 2.1. Cell cultures Mouse Swiss 3T3 fibroblasts (ATCC collection) and primary cultures of rat newborn astrocytes and rat newborn fibroblasts were used in this study. Experimental procedures meet the UK and EC ethical policy guidelines and regulations for the use of laboratory animals. Astrocytes were characterized by positivity to glial fibrillary acidic protein antibody. Rat fibroblasts were
Fig. 1. TSPO localization in mouse Swiss 3T3 fibroblasts. TSPO and mitochondria (anti-Complex V) immunostaining are shown in separate gray images and in a composite color image (TSPO red, mitochondria green, co-localization in yellow-orange). Strong TSPO immunoreactivity is found in the inner cytoplasm surrounding the nucleus, where the number of mitochondria is high. The plot shows the intensity of TSPO and mitochondria immunofluorescences sampled at the spatial resolution of 0.3 mm along a line across a single cell (white line in the color image). Major TSPO peaks fit well with mitochondrial profiles. However, defined TSPO peaks, not associated to mitochondria, are also present in the peripheral cytoplasm (asterisks) and in the plasma membrane (PM). A similar TSPO distribution was found in rat astrocytes.
320
A.M. Falchi et al. / Neuropharmacology 53 (2007) 318e329
characterized by positivity to vimentin and negativity to desmin and factor VIII antibodies. All cells were grown in Dulbecco’s modified Eagle’s medium with high glucose, supplemented with 10% fetal bovine serum, in a 5% CO2 incubator at 37 C. Experiments were performed when cells were subconfluent. Cells were plated in glass-bottomed dishes (MatTek, Ashland, MA) to allow the use of immersion objectives in fluorescence and contrast microscopy. Before treatments, serum-supplemented medium was substituted with serumfree medium. Experiments were replicated in different cell types using the same probe and drug concentrations, incubation times and data acquisition parameters.
2.2. Drugs Cells were treated in vivo with: 30e60 mM 7-chloro-5-(4-chlorophenyl)1,3-dihydro-1-methyl-2H-1,4-benzodiazepin-2-one (Ro5-4864), acutely added (a.a.); 30e60 mM 1-(2-chlorophenyl-N-methyl-1-methylpropyl)-3-isoquinoline-carboxamide (PK 11195), a.a.; 100e400 mM 2H-1,4-benzodiazepin2-one (diazepam), a.a.; 100e400 mM 5-(2-chlorophenyl)-7-nitro-3H-1,4-benzodiazepin-2(1H)-one (clonazepam), a.a.; 100e400 mM 4,40 -diisothiocyanato-stilbene-2,20 -disulfonate (DIDS), a.a.; 60e400 mM propranolol, a.a.; 20e40 mM carbonyl cyanide 3-chloro-phenylhydrazone (CCCP), a.a.; 2 mM
Fig. 2. Effect of TSPO ligands on NBD-cholesterol distribution. (A) Enhancement of NBD-cholesterol droplets after treatment with 60 mM PK 11195 (see also Fig. 6). Bar is 10 mm. (B) Fluorescence increase of a natural cluster of seven NBD-cholesterol droplets, shown in the strip picture, at baseline, 30 s and 3 min after treatment with PK 11195. (C) Per cent increase of the number of NBD-cholesterol droplets counted in 3 cells at baseline, 30 s and 3 min after treatment with PK 11195. (D) Fluorescence intensity of NBD-cholesterol droplets after treatment with PK 11195, Ro5-4864, diazepam and clonazepam, tested at 60 and 400 mM concentrations. Changes induced by 60 mM PK 11195, 60 mM Ro5-4864 and 400 mM diazepam, compared to baseline, were statistically significant. No changes were found with clonazepam. Rat newborn astrocytes.
A.M. Falchi et al. / Neuropharmacology 53 (2007) 318e329 cyclosporin A (CsA) for 30 min; 4 mM Sandoz-compound 58-035, an inhibitor of acylCoA/cholesterol acyltransferase 1 (ACAT1) for 30 min or 24 h; 1 mg/ml phalloidin for 30 min; 1 mM paclitaxel for 45 min; 100 nM colchicine for 30 min or 2 h. Vehicles were: Me2SO for PK 11195, Ro 5-4864, diazepam, CCCP, propranolol, DIDS, paclitaxel and Sandoz 58-035; ethanol for CsA; water for phalloidin and colchicine. Stock solutions were 1000-fold concentrated, to not exceed the 0.1% concentration of vehicle in the medium. The same concentration of vehicle was added to control cells. Ro5-4864, PK 11195, diazepam, propranolol, DIDS, CsA, CCCP, colchicine, phalloidin and paclitaxel were from Sigma (St. Louis, MO, USA). Sandoz 58-035 was kindly donated by Novartis (Dayton, NJ, USA).
2.3. Probes Cells were supravitally stained with the following probes (ex, em ¼ fluorescence excitation and emission): 5 mM 22-[N-nitrobenz-2-oxa-1,3-diazol-4-yl) amino]-23,24-bisnor-5-cholen-3-ol (NBD-cholesterol) for 1 h (ex 460 25, em 535 20); 300 nM Nile Red for 15 min (ex 460 25, em 535 20 for nonpolar lipids; ex 540 12, em 590 LP for all lipids); 100 nM acridine orange for 10 min (ex 540 12.5, em 590 LP); 100 nM tetramethylrhodamine methyl ester perchlorate (TMRM) for 30 min (ex 540 12.5, em 590 LP); 100 nM Mito Tracker Green FM (MitoTracker) for 20 min (ex 460 25, em 535 20); 20 mM dihydrofluorescein diacetate (DHF) for 30 min (ex 460 25, em 535 20); 1.25 mM MitoSOX for 10 min (ex 540 12.5, em 590 LP); 45 mM thiazolyl blue tetrazolium bromide (MTT) for 5e20 min; 10 mM calcein for 15 min (ex 480 20, em 520 20). NBD-cholesterol is a fluorescent analog of cholesterol where the terminal segment of the alkyl tail is replaced by the NBD fluorophore. NBD-cholesterol fluorescence is environment-sensitive, being weak in aqueous media and very bright in nonpolar media (Rukmini et al., 2001). NBD-cholesterol has been successfully used to study the mechanism of sterol transfer to mitochondrial membranes by the steroidogenic acute regulatory protein (StAR) and sterol carrier protein-2 (SCP-2) (Petrescu et al., 2001). Owing to the overlap of NBD-cholesterol and Nile Red emissions in the 535 20 band, for co-localization cells were first incubated with NBD-cholesterol, then fixed in 4% paraformaldehyde to block cytoplasm movements, photographed mapping the field position, overstained with Nile Red and re-photographed. MTT is a colorless compound readily taken up by living cells and reduced to formazan (FMZ), an intensely purple substance which can be observed in brightfield microscopy. FMZ is extruded by exocytosis forming characteristic needle-like crystals on the cell surface. Acridine orange is a pH-sensitive probe, which selectively stains acidic vesicles. TMRM is a potential-sensitive probe, which selectively accumulates in the mitochondrial matrix. MitoTracker is a potential-insensitive mitochondrial probe, which binds covalently the mitochondrial matrix, thus enabling structural changes to be detected even after mitochondrial depolarization. DHF is a non-fluorescent derivative of fluorescein, which reverts to fluorescent fluorescein upon oxidation by reactive oxygen species (ROS). MitoSOX is a fluorigenic dye, which selectively detects superoxide in mitochondria. However, MitoSOX fluorescence is not permanent as it leaves mitochondria after depolarization. Calcein is a membrane-impermeant fluorescent dye used to test plasma membrane integrity. Vehicles were: Me2SO for TMRM, DHF, MitoSOX, Nile Red, MitoTracker and calcein; chloroform for NBD-cholesterol; water for MTT and acridine orange. Stock solutions were 1000-fold concentrated, to not exceed the 0.1% concentration of vehicle in the medium. MTT was from Sigma (St. Louis, MO, USA); Nile Red and calcein from Fluka (Buchs, SG, Switzerland); acridine orange from Merck (Darmstadt, Germany). Other fluorescent probes were from Molecular Probes (Eugene, OR, USA).
2.4. TSPO immunocytochemistry Cells were fixed in 4% paraformaldehyde in PBS for 30 min, washed three times in PBS for 5 min and incubated overnight with a polyclonal anti-TSPO antibody (6361-PC-100, R&D Systems, Minneapolis, MN, USA). In some experiments, the anti-TSPO antibody was used in combination with a mitochondrial antibody (anti-Complex V, Molecular Probes) to verify the association of TSPO with mitochondria. Co-localization was semiquantitatively assessed by
321
linear profiles of BPR and Complex V immunofluorescences sampled at the spatial resolution of 0.3 mm (Fig. 1).
2.5. Imaging Observations were made using an Olympus IX 71 inverted microscope (Olympus, Tokyo, Japan) equipped with 20 and 60 planapochromatic objectives (Olympus UPlanSApo series) with efficient chromatic correction, which minimized the focus and planar drift between different fluorescence filters. This was a critical requisite for the co-localization of probes, as the size of small lipid droplets was near to the microscope resolution limit. The nominal resolutions of images taken with 20 and 60 objectives were 0.3 and 0.1 mm/pixel, respectively. The use of the inverted microscope excluded any risk of dye contamination, as there was no contact between the immersion objective and the medium. Images were taken with a 12-bit cooled CCD camera (Sensicam PCO, Kelheim, Germany), electronically coupled to a mechanical shutter interposed between the 100 W Hg lamp and the microscope, to limit illumination of cells for the time strictly required for the acquisition of images. Excitation light was attenuated with a 6% transmittance neutral density filter.
Fig. 3. Identification of NBD-cholesterol spots as lipid droplets. (A) Co-localization of NBD-cholesterol and Nile Red. Cells were incubated with 5 mM NBD-cholesterol for 60 min, fixed, photographed at 460 25/535 20, over-stained with Nile Red and photographed again at 540 12/590 LP to select Nile Red emission. The plot shows a close matching of NBD-cholesterol and Nile Red major peaks that correspond to lipid droplets. (B) Resistance of Nile Red spots to Triton X-100. After extraction with 0.1% Triton X-100 for 30 min at room temperature, only lipid droplets (LD) were still visible whereas perinuclear membranes (PNM) were lost. The plot shows the intensity profile of Nile Red fluorescence (540 12/590 LP) before and after Triton X-100 extraction. In the abscissa, 1 pixel corresponds to 0.1 mm. Mouse Swiss 3T3 fibroblasts.
322
A.M. Falchi et al. / Neuropharmacology 53 (2007) 318e329 Image analysis and measurements were performed using the ImagePro Plus package (Media Cybernetics, Silver Springs, MD).
2.6. Analysis of cholesterol synthesis To determine the rate of cholesterol synthesis, mouse Swiss 3T3 fibroblasts were incubated 6 h at the cell density of 105 cells/ml with 185 kBq/ml of sodium [14C]acetate. After incubation cells were separated by centrifugation and collected. Cell lipids were extracted with cold acetone and neutral lipids separated by TLC on Kiesegel plates using a solvent system containing heptane/ isopropylether/formic acid (60:40:2 by volume). Free cholesterol and cholesterol ester bands were identified by comparison with standards running simultaneously with samples and visualized using iodine vapor (Folch et al., 1957). For counting, the silica scraped from the bands was added directly to the counting vials containing 10 ml of Ultima Gold. All incubations were carried out in triplicate and the results for individual experiments are given as the mean values. Variation between triplicates was less than 10%. All data are expressed as the rate of [14C]acetate incorporation into cholesterol per mg protein.
2.7. Analysis of cholesterol esterification Cells were incubated for 6 h in a medium containing [14C]oleate bound to bovine serum albumin (BSA). To prepare the oleate-BSA complex, 3.7 MBq of [14C]oleic acid in ethanol (Dupont, NEN specific activity 2.035 GBq/ mmol) was mixed with 1.4 mg KOH and the ethanol evaporated. PBS (1.5 ml) without Ca2þ and Mg2þ containing 4.24 mg BSA (fatty acid-free, Sigma) was added and the mixture shaken vigorously. This solution was added to each well at a final concentration of 74 kBq/ml. After incubation cells were washed with ice-cold PBS and lipids extracted with acetone. Neutral lipids were separated by TLC as described above and incorporation of [14C]oleate into cholesterol esters was measured.
2.8. Analysis of cholesterol efflux Subconfluent cells were preincubated for 24 h with 0.185 MBq/ml of [3H]cholesterol. After this period radioactive medium was removed, the cells were washed extensively with PBS and incubated for 6 h at 37 C in conditioned medium 10% FCS. At the end of incubation cells and medium were collected separately. Lipids from media were extracted with chloroform/methanol (2:1 by volume) (Folch et al., 1957) and neutral lipids were separated as described above. Efflux of cholesterol was expressed as the percentage of total (cell plus medium) [3H]cholesterol recovered from the medium. [14C]Acetic acid (specific activity 2.0 GBq/mmol), [14C]oleic acid (specific activity 1.9 GBq/mmol) and [3H]cholesterol (specific activity 1.9 TBq/mmol) were from Dupont, New England Nuclear (Boston, MA, USA). Kiesegel plates were obtained from Merck (Darmstadt, Germany) and the Ultima Gold scintillation fluid was from Packard (Meridien, CT, USA).
Fig. 4. Effect of PK 11195 on cholesterol esterification, synthesis and efflux. (A) Cholesterol esterification. Cells received PK 11195 (20 and 40 mM) and [14C]oleate bound to bovine serum albumin (BSA) at the final concentration of 74 kBq/ml. After 6 h of incubation cells were washed with ice-cold PBS
and processed as described in Section 2. (B) Cholesterol synthesis. Cells received PK 11195 (20 and 40 mM) and 185 kBq/ml of sodium [14C]acetate. After 6 h incubation cells were separated and processed as described in Section 2. (C) Cholesterol efflux. Cells were incubated for 24 h with 0.185 MBq/ml of [3H]cholesterol. After this period the radioactive medium was removed, cells were washed extensively with PBS and incubated for 6 h at 37 C in conditioned medium 10% FCS and in the presence of PK 11195 (20 and 40 mM). At the end of incubation, cholesterol efflux was assessed as the percentage of the total (cell plus medium) [3H]cholesterol recovered from the medium. All experiments were carried on subconfluent mouse Swiss 3T3 fibroblasts. Each value represents the mean S.E.M. of triplicate experiments. *P < 0.001 vs. untreated cells by t-test.
A.M. Falchi et al. / Neuropharmacology 53 (2007) 318e329
3. Results 3.1. TSPO immunocytochemistry TSPO immunoreactivity was found in both astrocytes and 3T3 fibroblasts. The intensity of immunoreaction was much higher in astrocytes than in 3T3 fibroblasts. However, the intracellular distribution of TSPO immunostaining was similar in the two types of cells. Dual immunostaining of TSPO and mitochondria (Complex V) showed a strongly preferential localization of TSPO in mitochondria, mostly concentrated in the perinuclear cytoplasm. However, TSPO immunostaining was also detected in peripheral regions of cytoplasm, devoid of mitochondria, and in the plasma membrane (Fig. 1). 3.2. Enhancement of NBD-cholesterol droplets In untreated cells, the fluorescent cholesterol analog NBDcholesterol produced a moderate fluorescence of organelle membranes (endoplasmic reticulum, mitochondria, etc.) and few intensely fluorescent spots apparently similar to lipid droplets. Surprisingly, treatment with TSPO ligands resulted in an increase of fluorescence of NBD-cholesterol spots, both in astrocytes (Fig. 2A) and fibroblasts (Fig. 5). The change was very rapid, being apparent as soon as 30 s after treatment. The
323
strongest effect was observed with 60 mM PK 11195, resulting in an increase of both the fluorescence intensity and number of NBD-cholesterol spots (Fig. 2B, C). Diazepam (400 mM) and Ro5-4864 (60 mM) produced only an increase of fluorescence intensity (Fig. 2D). The increase of fluorescence could not be attributed to an extra NBD-cholesterol uptake or endocytosis, as cells were washed several times after incubation with the fluorescent probe. Moreover, the fluorescence increase of NBDcholesterol spots was accompanied by a decrease of the perinuclear fluorescence, suggesting a process of NBD-cholesterol transfer from cytoplasmic membranes to isolated spots. To verify the nature of NBD-cholesterol spots, cells were overstained with Nile Red, a specific stain of lipid droplets. Though Nile Red can be used in vivo, cells were fixed to block organelle movements and co-localize probes at micron resolution. All NBD-cholesterol spots were stained by Nile Red, thus revealing the nature of lipid droplets (Fig. 3A). In addition, NBD-cholesterol spots resisted extraction with 0.1% Triton X-100 for 30 min at room temperature, whereas fluorescent traces of cytoplasmic membranes were lost (Fig. 3B). These findings support the hypothesis that NBD-cholesterol is initially partitioned between cytoplasmic membranes and lipid droplets. Treatment with TSPO ligands induces a redistribution of NBD-cholesterol, which preferentially accumulates in lipid droplets.
Fig. 5. PK 11195-induced enhancement of NBD-cholesterol droplets is not inhibited by previous mitochondrial depolarization with the protonophore CCCP. The panel shows the changes of NBD-cholesterol and TMRM, a potential-sensitive mitochondrial probe, after treatment with 40 mM CCCP followed by 60 mM PK 11195. CCCP uncoupling resulted in a rapid and extensive mitochondrial depolarization, but did not prevent the enhancement of NBD-cholesterol droplets with PK 11195. Rat newborn fibroblasts. Bar is 10 mm.
324
A.M. Falchi et al. / Neuropharmacology 53 (2007) 318e329
Although NBD-cholesterol has been proven to be a suitable substrate for esterification by acylCoA/cholesterol acyltransferase (ACAT) (Lada et al., 2004), overnight treatment with the Sandoz-compound 58-035, a potent ACAT inhibitor, did not prevent the enhancement of NBD-cholesterol droplets induced by TSPO ligands. Experiments with radioactive markers, in the absence of NBD-cholesterol, showed that PK 11195 itself (Ro5-4864 not tested) is a strong inhibitor of cholesterol esterification (Fig. 4A). Owing to the relatively long (6 h) duration of PK 11195 treatments required to assess cholesterol esterification and cholesterol synthesis, PK 11195 concentrations were slightly lower than those used in acute administration
(20e40 mM rather than 30e60 mM). PK 11195 (40 mM) induced a complete block of cholesterol esterification, whereas the effect of 20 mM PK 11195 was remarkably lower. However, PK 11195 did not modify endogenous cholesterol synthesis (Fig. 4B) nor did it affect cell viability, as evaluated by trypan blue exclusion test. The relative agreement between the range of effective concentrations required to inhibit cholesterol esterification and to enhance NBD-cholesterol droplets suggests the existence of a common mechanism of action. The accumulation of unesterified NBD-cholesterol in lipid droplets does not contrast with the composition of lipid droplets which are known to contain mostly esterified cholesterol.
Fig. 6. (A) Release of NBD-cholesterol droplets in the medium. 10 min after PK 11195 treatment, a small but consistent number of NBD-cholesterol droplets was found in the medium. The droplets (encircled particles) are better visualized after contrast enhancement. The absence of necrotic cells excluded the release of fluorescent particles by a mechanism of cell fragmentation. Mouse Swiss 3T3 fibroblasts. (B) Release of FMZ granules in the medium. The pictures show the same microscopic field observed at baseline and 10 min after 60 mM PK 11195 treatment. Several FMZ granules are found in the medium. Rat newborn fibroblasts. Bar is 10 mm.
A.M. Falchi et al. / Neuropharmacology 53 (2007) 318e329
325
Free cholesterol has been found on the surface of lipid droplets of adipocytes (Prattes et al., 2000) and peritoneal macrophages (McGookey and Anderson, 1983), and is generally assumed that the lipid droplet surface represents a region of exchange between the pool of free cholesterol present in membranes and the pool of esterified cholesterol present in the inner droplet core. The enhancement of NBD-cholesterol droplets induced by TSPO ligands was not inhibited by previous depolarization of mitochondria with CCCP or rotenone (Fig. 5), indicating that the effect of TSPO ligands was independent of the energetic condition of mitochondria. Ten to fifteen minutes after PK treatment, a few free-floating NBD-cholesterol droplets were also found in the medium. The number of fluorescent particles present in the medium was only a minimal fraction (
