Experimental Cell Research 255, 56 – 66 (2000) doi:10.1006/excr.1999.4774, available online at http://www.idealibrary.com on
NPC1-Containing Compartment of Human Granulosa-Lutein Cells: A Role in the Intracellular Trafficking of Cholesterol Supporting Steroidogenesis 1 Hidemichi Watari,* E. Joan Blanchette-Mackie,† Nancy K. Dwyer,† Gwoshing Sun,‡ Jane M. Glick,‡ Shutish Patel,§ Edward B. Neufeld,† Peter G. Pentchev, 㛳 and Jerome F. Strauss III,* ,2 *Center for Research on Reproduction and Women’s Health, Department of Obstetrics and Gynecology and Cellular and ‡Molecular Engineering, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104; †Section of Lipid Cell Biology, National Institutes of Diabetes and Digestive and Kidney Diseases, and 㛳Cellular and Molecular Pathophysiology Section, Developmental and Metabolic Neurology Branch, National Institutes of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland 20892; and §Neurology Research Laboratory, Veterans Administration Connecticut Healthcare System, Newington, Connecticut 06111
recovered after U18666A was removed from the culture medium. U18666A treatment caused a 2-fold or more increase in NPC1 protein and mRNA levels, suggesting that disruption of NPC1’s function activates a compensatory mechanism resulting in increased NPC1 synthesis. We conclude that the NPC1 compartment plays an important role in the trafficking of LDLderived substrate in steroidogenic cells; that NPC1 expression is up-regulated when NPC1 action is blocked; and that the NPC1 compartment can be functionally separated from other intracellular pathways contributing substrate for steroidogenesis. © 2000
Steroidogenic cells represent unique systems for the exploration of intracellular cholesterol trafficking. We employed cytochemical and biochemical methods to explore the expression, regulation, and function of the Niemann-Pick C1 protein (NPC1) in human granulosalutein cells. NPC1 was localized in a subset of lysosome-associated membrane glycoprotein 2 (LAMP-2)positive vesicles. By analyzing the sensitivity of NPC1 N-linked oligosaccharide chains to glycosidases and neuraminidase, evidence was obtained for movement of nascent NPC1 from the endoplasmic reticulum through the medial and trans compartments of the Golgi apparatus prior to its appearance in cytoplasmic vesicles. NPC1 protein content and the morphology and cellular distribution of NPC1-containing vesicles were not affected by treatment of the granulosa-lutein cells with 8-Br-cAMP, which stimulates cholesterol metabolism into progesterone. In contrast, steroidogenic acute regulatory (StAR) protein levels were increased by 8-Br-cAMP. Incubation of granulosa-lutein cells with low-density lipoprotein (LDL) in the presence of the hydrophobic amine, U18666A, caused accumulation of free cholesterol in granules, identified by filipin staining, that contained LAMP-2 and NPC1. These granules also stained for neutral lipid with Nile red, reflecting accumulation of LDL-derived cholesterol esters. LDL-stimulated progesterone synthesis was completely blocked by U18666A, leaving steroid output at levels similar to those of cells incubated in the absence of LDL. The hydrophobic amine also blocked the LDL augmentation of 8-Br-cAMP-stimulated progesterone synthesis, reducing steroid production to levels seen in cells stimulated with 8-BrcAMP in the absence of LDL. Steroidogenesis
Academic Press
INTRODUCTION
Steroidogenic cells are valuable models for studying intracellular cholesterol trafficking because of their unique requirements for moving sterol to the mitochondria where the synthesis of hormones is initiated. Cholesterol derived from different sources is employed in the biogenesis of steroid hormones, including cholesterol synthesized de novo, cholesterol acquired from low-density lipoproteins (LDL) and high-density lipoproteins (HDL), and cholesterol located in the plasma membrane or deposited in cytoplasmic lipid droplets as sterol esters [6, 8, 10, 11, 31, 37]. Cholesterol in each of these potential substrate pools must move to the mitochondria where the first committed step in steroidogenesis takes place, the side-chain cleavage reaction [23]. The translocation of cholesterol from the outer mitochondrial membrane to the inner membrane where the cholesterol side-chain cleavage complex resides is governed by the steroidogenic acute regulatory protein (StAR) [36]. The itineraries followed by substrate cholesterol originating from the different pools, ultimately leading to the mitochondria, have not
1
Supported by NIH Grants HD06274 (J.F.S.) and NS34339 (S.P.) and a grant from the Ara Parseghian Medical Research Foundation. 2 To whom correspondence and reprint requests should be addressed at: 1354 BRB II/III, 421 Curie Boulevard, Philadelphia, PA 19038. Fax: (215) 573-5408. E-mail:
[email protected]. 0014-4827/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
56
NPC1 AND CHOLESTEROL TRAFFICKING IN STEROIDOGENIC CELLS
been fully mapped. Thus, it is not known if cholesterol substrate passes through central sorting stations prior to reaching the mitochondria, or whether sterol originating in different pools travels by pool-specific pathways. LDL-carried cholesterol is a major substrate for hormone synthesis in human steroidogenic cells [11], and the pathway of receptor-mediated uptake of LDL and its hydrolysis in lysosomes in these cells has been well documented [9, 35]. However, the manner in which LDL-derived cholesterol enters into the steroidogenic substrate pool and its relationship to cholesterol in other cellular compartments are still largely unknown. The recent discovery of the gene that is mutated in Niemann-Pick type C disease, named NPC1, has provided an important clue to the intracellular trafficking of cholesterol [4, 20]. A prominent feature of NiemannPick type C disease is delayed efflux of cholesterol from lysosomes [18]. The product of the NPC1 locus is a membrane glycoprotein with a sterol-sensing domain [4, 20, 41, 42]. Based on the cellular phenotype of Niemann-Pick type C disease, NPC1 is evidently critical for the movement of free cholesterol and other cargo out of late endosomes/lysosomes [26]. The role of NPC1 in the utilization of cholesterol in steroidogenic cells has received relatively little attention. However, deficient testicular testosterone production in mice homozygous for inactivating mutations in the homologous murine Npc1 gene has been documented, indicating an important role for NPC1 in trafficking of cholesterol entering the steroidogenic pool [33]. In the present study, we utilized cytochemical and biochemical methods and pharmacological probes to explore the expression, regulation, and function of NPC1-containing compartments in the process of steroidogenesis. MATERIALS AND METHODS Reagents. U18666A, (3-(2-(diethylamino)ethoxy)androst-5-en17-one) was obtained from the Upjohn Company (Kalamazoo, MI). Filipin was purchased from Polysciences (Warrington, PA). Progesterone was purchased from Sigma Chemical Co. (St. Louis, MO). Cell culture. Proliferating human granulosa-lutein cells were prepared from granulosa cells obtained from women undergoing in vitro fertilization/embryo transfer and cultivated as previously described [21]. Studies were carried out with cultures established from three different cell isolates. For an individual experiment, cells were cultured to 50 –90 % confluence. For 24 h before treatment, cells were incubated in media supplemented with 2% lipoprotein-deficient serum (LPDS) and then treated with various reagents in media supplemented with 2% LPDS for 24 h. The culture media were then collected and cells were analyzed for protein and sterol contents, Western blotting for NPC1 and StAR proteins, and Northern blotting for NPC1 and StAR mRNAs. CT60 Chinese hamster ovary cells, which display the NiemannPick type C phenotype [3, 5], were cultured and transfected with wild-type NPC1 or a NPC1 mutant lacking the four C-terminal amino acid residues containing the di-leucine lysosomal targeting motif as previously described [41].
57
Western blotting. Cells were scraped from the dishes into lysis buffer consisting of 100 mM Tris-HCl, pH 8.0, 1% Nonidet P-40, 150 mM NaCl, 0.02% sodium azide, 100 g/ml phenylmethylsulfonyl fluoride, and 1 g/ml aprotinin and centrifuged at 10,000g for 2 min at 4°C. Supernatants were subjected to SDS-PAGE and then Western blotting. Anti-NPC1 N-terminus polyclonal antiserum [42] and anti-recombinant human steroidogenic acute regulatory protein polyclonal antiserum [29] were employed in Western blot analyses. Glycosidase treatment. Cells were scraped from dishes, centrifuged at 1000g for 5 min at 4°C, resuspended in 1 ml of phosphatebuffered saline, and centrifuged as above. The cell pellet was resuspended in 0.4 ml of sucrose buffer (10 mM Hepes-KOH, pH 7.4, 10 mM KCl, 1.5 mM MgCl 2, 5 mM sodium EDTA, 250 mM sucrose), passed through a 22-gauge needle 20 times, and centrifuged at 15,000g for 10 min at 4°C. The resulting pellet was resuspended in 30 l of sucrose buffer containing 0.1 M NaCl. Aliquots were treated with 1 g of trypsin (Sigma) at 30°C for 30 min. The reaction was stopped by the addition of 400 units of soybean trypsin inhibitor (Sigma). Reaction mixtures were pelleted using speed vac at 4°C, and the resulting pellets were resuspended in distilled H 2O. Equal volumes of the aliquot were incubated with or without endogycosidase H or N-glycosidase F using the Endoglycosidase H Deglycosylation kit or the N-glycosidase F Deglycosylation kit, respectively (Roche Molecular Biochemicals, Indianapolis, IN), according to the manufacturer’s instructions. The mixtures were heated and subjected to SDS-PAGE and Western blotting as described above. Neuraminidase treatment. For treatment with Clostridium perfingens neuraminidase (New England Biolabs, Beverly, MA), membranes (60 l) were incubated with trypsin as described above and then subjected to sequential addition of 5 l of solution of 17X protease inhibitors (a concentration of 1X corresponding to 10 g/ml leupeptin, 5 g/ml pepstatin, and 2 g/ml aprotinin) and 8.5 l of 10% (vol/vol) Triton X-100 [27]. After rocking at 4°C for 1 h, equal aliquots were incubated with or without neuraminidase (50 units) in the presence of 50 mM sodium citrate, pH 4.5, overnight at 37°C, and the reaction was stopped by addition of SDS sample buffer. The mixtures were then heated at 100°C for 5 min and subjected to SDS-PAGE and Western blotting as described above. Northern blotting. Total RNA was isolated from cultures with Trizol reagent (Gibco-BRL, Grand Island, NY) according to the manufacturer’s instructions. Equal amounts of total RNA (40 – 60 g) were separated on 1% agarose-formaldehyde denaturing gels and transferred to nylon membranes. Northern blots were probed with a NPC1 cDNA (nucleotides 196 –921), a human StAR cDNA [38], and a cDNA encoding human 28 S rRNA. Hybridization signals were quantified using a phosphoimager. Lipid analysis. Cells were washed with phosphate-buffered saline and then lipids were extracted with isopropanol as previously described [22]. After overnight incubation, isopropanol extracts were transferred to glass tubes and dried down under nitrogen in an evaporator at 55°C. The monolayer protein was dissolved in 0.1 M NaOH containing 1% SDS and the protein content measured with the Micro BCA kit (Pierce). The lipid extract was redissolved and dried down four times, initially in 1 ml toluene, then in 1 ml chloroform, then in 0.5 ml Triton X-100 in chloroform, and finally in 0.5 ml chloroform. Water (0.25 ml) was added to the lipid/Triton X-100 mixture, which was briefly heated briefly to 50°C to ensure solubilization of the lipids. Free and total cholesterol was measured using kits from Wako Chemical Co. adapted for use in microplates. Esterified cholesterol was calculated by the difference. Measurement of progesterone. Progesterone in conditioned media was quantified using a radioimmunoassay kit (Diagnostic Products Corporation, Los Angeles, CA) according to the manufacturer’s instructions. Each value was normalized to the amount of total cellular protein.
58
WATARI ET AL.
Immunocytochemical analyses. Cells were grown on uncoated glass coverslips. At the end of the culture period, cells were washed with phosphate-buffered saline and fixed with 3% paraformaldehyde for 30 min. Immunostaining was performed with rabbit affinity purified anti-peptide NPC1-C antibody [26, 28, 41] and mouse monoclonal anti-lysosome-associated membrane glycoprotein, LAMP-2, in blocking solution containing filipin (0.05%) and goat IgG (2.5 mg/ml) as previously described [41]. In some cultures, Nile red was used to identify neutral lipid droplets. Stained cells were examined with a Zeiss LSM 410, confocal microscope equipped with a UV laser. Statistical analysis. Experiments were conducted in triplicate and replicated on at least three separate occasions, except where indicated. Significant differences among treatment groups were identified using ANOVA and the Tukey-Kramer test with P ⬍ 0.05 considered to be significant.
RESULTS
Localization of NPC1 in Human Granulosa-Lutein Cells Immunocytochemistry localized NPC1 in human granulosa-lutein cells incubated with LDL to a subset of vesicles approximately 1 m in diameter that stained for the lysosomal membrane glycoprotein, LAMP-2 (Figs. 1A,B,C,D). The NPC1-containing vesicles (Fig. 1A) were distributed in the cytoplasm in a pattern that was distinctly different from that of free cholesterol (Fig. 1E), identified by staining with the polyene antibiotic, filipin, which complexes with sterol 3-hydroxyl groups (Fig. 1G, merge of NPC1 immunostaining in red and filipin staining in blue). The free cholesterol-containing vesicles, which were occasionally prominent in scattered cells (Fig. 1E) were clearly separated from the cytoplasmic neutral lipid droplets, identified by Nile red-staining, present in all cells incubated with LDL (Fig. 1F). NPC1-containing vesicles were also separated from cytoplasmic neutral lipid droplets (Fig. 1H, merge of NPC1 immunostaining in red and neutral lipid staining in green). Although NPC1 immunostaining was not identified in structures consistent with the endoplasmic reticulum or Golgi apparatus, biochemical evidence for the transit of NPC1 through these organelles was obtained through the examination of the sensitivity of NPC1 N-linked oligosaccharide chains to glycosidases. N-linked oligosaccharide chains incorporated into
proteins in the endoplasmic reticulum are sensitive to cleavage by endoglycosidase H [14]. These oligosaccharide chains are later trimmed in the medial Golgi apparatus and in this process become resistant to endoglycosidase H cleavage. However, at all stages the oligosaccharides can be released by peptide N-glycosidase F. Some of the trypsin-generated fragments containing the N-glycosylated NPC1 N-terminus from human granulosa-lutein cells were sensitive to endoglycosidase H digestion, but most were endoglycosidase H resistant yet sensitive to peptide N-glycosidase F (Fig. 2). As a control for the enzymatic digestions, we examined glycosidase sensitivities of wildtype NPC1 and a NPC1 mutant lacking the C-terminal four amino acids (C-4) transfected into a NPC1-null cell line, CT60 Chinese hamster ovary cells. The C-4 mutant remains trapped in the endoplasmic reticulum, whereas the wild-type protein enters into small late endosome/lysosomal vesicles in these cells [41]. The trypsin-generated fragments from the C-4 mutant were completely endoglycosidase H sensitive reflecting the lack of Golgi-mediated processing, whereas a portion of the wild-type protein was endoglycosidase H resistant. These observations confirm that endoglycosidase H sensitivity identifies NPC1 species that have not passed through the Golgi apparatus. Sialic acid residues are added to glycoproteins in the trans Golgi apparatus. Treatment of trypsin-generated fragments with neuraminidase resulted in the loss of one of the immunoreactive N-terminal bands, indicating that NPC1 is sialylated (Fig. 3A). A similar pattern of neuraminidase sensitivity was observed for the wildtype protein prepared from transfected CT60 cells, but the neuraminidase-sensitive fragment was not detected in CT60 cells transfected with the C-4 mutant which does not pass through the Golgi (Fig. 3B). Collectively, these observations indicate that NPC1 passes through the medial and trans Golgi prior to entering the late endosome/lysosomal compartment. Effects of 8-Br-cAMP on NPC1 Expression Treatment of human-granulosa lutein cells with 1 mM 8-Br-cAMP for 24 h increased progesterone pro-
FIG. 1. Human granulosa-lutein cells were incubated with LDL (50 g/ml) and stained with filipin to localize free cholesterol and Nile red to localize neutral lipid and immunostained for NPC1 and LAMP-2 proteins as indicated. NPC1-containing vesicles (A) and LAMPcontaining vesicles (B) are randomly distributed in the cytoplasm. Note there are more LAMP-positive vesicles than NPC1-positive vesicles. Higher magnification shows that NPC1 protein (C) is present in LAMP-positive vesicles (D); arrows point out colocalization between NPC1 and LAMP. Thus NPC1 is located in a subset of LAMP-positive vesicles. The little free cholesterol that accumulates intracellularly (E) is present in the perinuclear Golgi region whereas most cellular free cholesterol is present in the cell membranes and illuminates cell outlines when stained with filipin (E). Nile red staining (F) shows neutral lipid cytoplasmic droplets not in the same compartment as intracellular free cholesterol (compare E and F). Merge of NPC1 protein immunostaining and filipin staining (G) shows that NPC1 protein (red) does not colocalize with accumulations of free cholesterol (blue). Merge of NPC1 protein immunostaining and Nile red staining (H) shows that NPC1 protein (red) does not colocalize with cytoplasmic neutral lipid droplets (green). A and B same magnification, bar ⫽ 5 m; C and D same magnification, bar ⫽ 5 m, E and F same magnification, bar ⫽ 25 m, G and H same magnification, bar ⫽ 5 m.
NPC1 AND CHOLESTEROL TRAFFICKING IN STEROIDOGENIC CELLS
59
60
WATARI ET AL.
FIG. 2. Posttranslational processing of NPC1 in human granulosa-lutein cells. (A) Susceptibility of NPC1 to endoglycosidase H and peptide N-glycosidase F digestion. Membrane fractions from human granulosa-lutein cells were treated with trypsin and the digests subsequently treated with endoglycosidase H (Endo H) or peptide N-glycosidase F (Endo F) followed by Western blotting as described in the text. Arrows indicate the Endo H and Endo F-generated products demonstrating that most of the NPC1 is Endo H resistant while all of it is Endo F sensitive. (B and C) Trypsin digests of NPC1 from membrane fractions of CT60 cells transfected with a wild-type NPC1 expression plasmid (WT) or a truncated NPC1 cDNA lacking the C-terminal di-leucine motif (C-4) which causes the protein to be trapped in the endoplasmic reticulum. Some of the wild-type NPC1 is Endo H sensitive while all of the C-4 mutant is Endo H sensitive (B), whereas in both cases the NPC1 fragments are sensitive to Endo F (C).
duction approximately 4-fold (vide infra) but did not alter NPC1 mRNA or protein levels (Fig. 4). Moreover, 8-Br-cAMP treatment did not affect the morphology or distribution of NPC1-containing vesicles in the cells, but it did substantially reduce the number of Nile red-staining cytoplasmic droplets (data not shown). In contrast, levels of StAR protein and mRNA were increased 4.5-fold and 5-fold, respectively (Fig. 4). U18666A Causes Accumulation of LDL-Derived Cholesterol in Lysosomes Containing NPC1 and Blocks LDL-Supported Steroidogenesis The hydrophobic amine, U18666A, is known to block various intracellular trafficking events including movement of cholesterol from the plasma membrane to the endoplasmic reticulum and movement of cholesterol from lysosomes to the endoplasmic reticulum and plasma membrane [17, 19, 40]. All of the granulosalutein cells treated with U18666A (2 g/ml) in the presence of LDL (50 g/ml) accumulated large perinuclear vesicles that stained intensely with filipin (Fig. 5A compared to images shown in Figs. 1E and 1G, which were taken of control cells for the U18666A experiment). The filipin-positive vesicles (Fig. 5C) stained with antibodies to LAMP-2 (data not shown) and NPC1 protein (Fig. 5D). The filipin-positive vesicles (Fig. 5A) also stained with Nile red (Fig. 5B), reflecting the accumulation of neutral lipid. The profound accumulation of large filipin-stained granules was not seen in granulosa-lutein cells incubated with LDL in the absence of U18666A (Figs. 1E and 1G). The free cholesterol content of granulosa-lutein cells incu-
bated in the presence of LDL and U18666A nearly doubled (Table 1). The observed increase in the mass of esterified cholesterol in the U18666A-treated cells compared to cells incubated with LDL alone reflects the accumulation of unhydrolyzed LDL sterol esters in late endosomes/lysosomes since the cytochemical studies described above showed colocalization of free cholesterol and neutral lipid in the LAMP-2, NPC1-positive granules. Thus, the profound arrest of free cholesterol efflux from lysosomes caused by U18666A evidently impairs acid lipase hydrolysis of LDL esterified lipids. Incubation of granulosa-lutein cells with LDL resulted in a 2.5-fold increase in progesterone secretion (Fig. 6). However, U18666A completely inhibited the LDL-supported steroidogenesis. Treatment of the granulosa-lutein cells with 8-Br-cAMP in the absence of LDL increased progestin production nearly 4-fold. The effects of 8-Br-cAMP were augmented in the presence of LDL resulting in a 6-fold increase in progesterone production. The supplemental effects of LDL on 8-Br-cAMP-stimulated progesterone synthesis were blocked by U18666A. The inhibitory effects of U18666A on LDL-sponsored progesterone synthesis were not due to interference with the cholesterol side-chain cleavage enzyme since 22(R)-OH-cholesterol (5 g/ml), a less hydrophobic steroidogenic precursor than cholesterol which readily enters into mitochondria [38], prevented the inhibition of progesterone production caused by U18666A (Control ⫹ 22(R)-OH-cholesterol ⫽ 100%; U18666A ⫹ 22(R)-OH-cholesterol ⫽ 90 ⫾ 14% of control progesterone production; mean ⫾ SD, N ⫽ 3;
NPC1 AND CHOLESTEROL TRAFFICKING IN STEROIDOGENIC CELLS
61
promotes accumulation of free cholesterol in lysosomes (Fig. 7). When granulosa-lutein cells incubated with LDL in the absence or presence of U18666A were subsequently placed in lipoprotein-deficient media in the absence of U18666A, progesterone production by cells never exposed to U18666A declined whereas progesterone production by cells previously incubated with U18666A and LDL increased, suggesting that free cholesterol accumulated in the filipin-positive granules had moved out of the lysosomal compartment and into the steroidogenic pool (Fig. 8). Filipin staining of the cells after a 48-h drug washout revealed the presence of free cholesterol laden cytoplasmic granules in most cells, although granule size and number were reduced compared to 0 time consistent with some mobilization of accumulated free cholesterol (data not shown). DISCUSSION
The present observations on human granulosa-lutein cells corroborate the localization of NPC1 in a subset of LAMP-2-positive vesicles, as has been recently reported in human fibroblasts [26]. The NPC1containing vesicles are thought to mediate the retrograde unloading of lysosomal cargo, a process requiring FIG. 3. NPC1 is a sialoglycoprotein. (A) Granulosa-lutein cell membranes were subjected to trypsin digestion and incubated without (⫺) or with neuraminidase (⫹) followed by Western blot analysis. Neuraminidase treatment resulted in the loss of one of the immunoreactive fragments (arrow). (B) Trypsin digests of membrane fractions isolated from CT60 cells transfected with wild-type NPC1 or the C-4 mutant were treated with neuraminidase as described above. Neuraminidase treatment removed the 55- to 60-kDa immunoreactive band (arrow) which was not present in trypsin digests of the C-4 mutant, which does not pass through the Golgi apparatus.
P ⬎ 0.05). U18666A is also known to block de novo cholesterol synthesis as well as cholesterol movement from the plasma membrane to internal menbranes. When granulosa-lutein cells were incubated in the absence of LDL as substrate, U18666A treatment reduced basal progesterone production by 17.5 ⫾ 7.4% (mean ⫾ SE, N ⫽ 4 separate experiments) and 8-BrcAMP-stimulated progesterone production by 20.8 ⫾ 6.5% (mean ⫾ SE, N ⫽ 3 separate experiments) compared to control cultures incubated in the absence of drug. Western blot analyses carried out on extracts of granulosa-lutein cells incubated with LDL in the absence or presence of U18666A revealed a 2-fold increase in the cellular content of NPC1, but no change in the levels of StAR (Fig. 7). Moreover, U18666A treatment caused a more than 2-fold increase in levels of NPC1 mRNA as did progesterone (30 M), which also
FIG. 4. Effects of 8-Br-cAMP treatment on NPC1 and StAR protein and mRNA contents of human granulosa-lutein cells. Protein and RNA were extracted from control cells or cells incubated with 1 mM 8-Br-cAMP for 24 h as described in the text and subjected to Western blot and Northern blot analysis for NPC1 and StAR. The bracket on the StAR Western blot indicates the multiple species of StAR including the 30-kDa mature form and the 37-kDa precursor.
62
WATARI ET AL.
FIG. 5. Effect of U18666A on free cholesterol, neutral lipid, and NPC1 distribution in human granulosa-lutein cells. Cells were cultured with LDL (50 g/ml) in the presence of U18666A (2 g/ml) for 24 h and then stained for free cholesterol deposits with filipin and neutral lipid with Nile red and immunostained for NPC1 protein. U18666A induces accumulation of free cholesterol (A) in the same compartment as Nile red-positive droplets (B) in granulosa cells. NPC1 protein (D) colocalizes with free cholesterol (C), pointed out at arrows. A and B, bar ⫽ 25 m; C and D, bar ⫽ 5 m.
only transient interaction of the NPC1 vesicles with lysosomes [26]. Thus, NPC1 vesicles do not colocalize with free cholesterol-containing structures, representing lysosomes containing LDL-derived free cholesterol, unless the cells are treated with U18666A or drugs with a similar action. U18666A appears to disrupt the
dynamic itinerary of the NPC1 vesicles, trapping NPC1 in lysosomes. The accumulation of immunoreacive-NPC1 in cholesterol-laden lysosomes was also observed when inactive NPC1 mutants were expressed in CT60 cells, which display the Niemann-Pick type C phenotype [41, 42]. Therefore, if NPC1 action is
NPC1 AND CHOLESTEROL TRAFFICKING IN STEROIDOGENIC CELLS
TABLE 1 Effect of U18666A on the Cholesterol Content of Human Granulosa-Lutein Cells Cholesterol (g/mg protein) Treatment
Free
Esterified
LDL LDL ⫹ U18666A
40 ⫾ 4.2 a 72.4 ⫾ 10.2 b
12 ⫾ 1.9 a 20.6 ⫾ 3.6 b
Note. Human granulosa-lutein cells were cultured for 24 h with LDL (50 g/ml) in the absence or presence of U18666A (2 g/ml) and cellular cholesterol contents were determined as descrtibed in the text. Values presented are means ⫾ SD from triplicate cultures in each treatment group. Means with different superscripts are significantly different (P ⬍ 0.05). Similar results were obtained in an independent replication of this experiment.
blocked by pharmacologic agents or the protein is incapable of carrying out its functions upon reaching the lysosomes, it is frozen in the lysosomal compartment. Our failure to detect wild-type NPC1 in either the endoplasmic reticulum or the Golgi apparatus at the light microscopic level in the face of biochemical evidence based on sensitivity to glycosidases indicating that the protein must move through these organelles could reflect the masking of epitopes in these locations or, more likely, a very transient residence time in the endoplasmic reticulum and Golgi. The cellular content of NPC1 is evidently not tightly linked to the steroidogenic program regulated by cAMP, unlike StAR gene expression [36]. However, because we only examined static images in our cytochemical studies, we cannot exclude the possibility that cAMP accelerates the dynamic movement of NPC1-containing vesicles and hence increases their ability to unload lysosomal cargo without a change in NPC1 content. Although 8-Br-cAMP did not affect NPC1 expression, treatment with U18666A increased both NPC1 protein and mRNA. These observations suggest that when the function of NPC1 is blocked a homeostatic mechanism is activated which results in increased synthesis of NPC1 and possibly other protein components of the late endosome/lysosomal system. This compensatory mechanism probably explains the increased intensity of NPC1 staining in U18666Atreated cells, reflecting increased NPC1 protein abudance. LDL-derived cholesterol used for steroidogenesis must pass through a compartment in human granulosa-lutein cells that is sensitive to blockade by the hydrophobic amine, U18666A. The exact mechanism by which hydrophobic amines like U18666A interfere with intracellular cholesterol trafficking is unknown [16, 17, 40]. These compounds may bind to lysobisphosphatidic acid, which is enriched in late endosomes/ lysosomes [13]. Recent studies using several different
63
approaches including antibodies against lysobisphosphatidic acid suggest that this phospholipid plays an important role in the movement of free cholesterol out of late endosome/lysosomal vesicles [13]. By analogy, U18666A may neutralize a critical action of lysobisphosphatidic acid. However, this mechanism does not explain how progesterone, which has no charge, produces a similar cellular phenotype. The role of lysosomes in the processing of LDL-derived cholesterol for steroidogenesis is not unanticipated since lysosomotropic agents have been known for some time to block the metabolism of LDL-carried radiolabeled cholesterol esters into steroid hormones [35]. U18666A evidently prevents the unloading of lysosomal free cholesterol but does not interfere with the movement and fusion of the late endosomes containing NPC1 with lysosomes. The present study confirms the role of the late endosome/lysosomal compartment in processing one pool of cholesterol supporting steroidogenesis. It also documents that this compartment can be functionally separated from other pools of cholesterol utilized for steroidogenesis since U18666A does not completely shut down progesterone synthesis. Cholesterol stored in neutral lipid droplets or cholesterol in the plasma membrane could represent the U18666Ainsensitive pool(s) employed when utilization of LDLderived substrate is blocked. The notion of multiple sterol trafficking pathways is consistent with the work of a number of investigators whose experiments demonstrate separable itineraries for cholesterol moving between lysosomes and plasma membrane and lysosomes and endoplasmic reticulum
FIG. 6. Effect of U18666A on progesterone secretion by human granulosa-lutein cells. (A) Cells were cultured with and without LDL (50 g/ml) in the absence or presence of U18666A (2 g/ml). (B) Effect of U18666A on 8-Br-cAMP-stimulated progesterone production. Cells were incubated with or without 8-Br-cAMP (1 mM), LDL (50 g/ml), and U18666A (2 g/ml). Progesterone in the media was quantitated by radioimmunoassay and cellular protein was determined. Values presented are means ⫾ SE from three separate experiments. Values with different superscripts are significantly different (P ⬍ 0.05).
64
WATARI ET AL.
FIG. 7. Effects of U18666A and progesterone treatment on NPC1 expression in human granulosa-lutein cells. Cells were cultured with and without LDL (50 g/ml) and U18666A (2 g/ml) or progesterone (30 M) for 24 h and extracted for Western blot analysis of NPC1 (A) and StAR (B) protein and Northern blot analysis (C) of NPC1 and 28S rRNA as described in the text. Similar results were obtained in two independent experiments.
[15, 25, 30, 39]. It should be noted that the pool of steroidogenic cholesterol affected by U18666A may be cell specific. Rodent testicular Leydig and adrenal tumor cells appear to utilize plasma membrane cholesterol as a primary substrate [8, 10, 30]. In these cells, U18666A also partially blocks utilization of cellular cholesterol in the absence of exogenous lipoproteins [12]. These results imply that the NPC1 compartment may, in certain cell types or certain circumstances, function in the trafficking of plasma membrane cholesterol. However, Porpaczy et al. [30] presented evidence that plasma membrane cholesterol is internalized through an endosomal compartment that can be distinguished from lysosomes. These data are again consistent with the existence of multiple sterol trafficking pathways. It should also be appreciated that cAMP can modify the trafficking pattern of cholesterol in steroidogenic cells, probably as a consequence of the enhanced conversion of sterol substrate into steroid hormones [7]. In rodent steroidogenic cells, but also in human ste-
roidogenic cells under certain conditions, HDL cholesterol esters can be selectively taken up and processed to free cholesterol for use in steroidogenesis by a nonlysosomal pathway [1, 31, 32]. The actual mechanism of the selective extraction of cholesterol esters from HDL particles without their internalization and the physical form in which they are incorporated into cells is still unknown. The process is not sensitive to drugs that interfere with lysosomal function and, therefore, may be independent of NPC1. Our human granulosa cell cultures do not respond with increased progesterone production or increase cholesterol storage in response to HDL (data not shown), so it was not possible for us to explore the impact of U18666A on HDLsupported steroidogenesis. Progesterone reversibly inhibits the movement of cholesterol out of lysosomes and causes accumulation of NPC1 in free cholesterol-engorged organelles [26]. Granulosa-lutein cells treated with 3 to 30 M progesterone accumulated filipin-positive, NPC1-positive granules in the cytoplasm, similar to cells treated with
NPC1 AND CHOLESTEROL TRAFFICKING IN STEROIDOGENIC CELLS
65
in up-regulating NPC1 expression at the time of luteinization of granulosa cells, thus endowing them with a greater capacity to process LDL-derived substrate for steroidogenesis. In conclusion, our findings implicate an organelle whose function is sensitive to hydrophobic amines, which contains NPC1 glycoprotein, in the utilization of LDL-derived cholesterol for steroidogenesis. This organelle is not the only sorting station for cholesterol entering into the steroidogenic pool since some basal and 8-Br-cAMP-stimulated progesterone production can continue in the presence of U18666A. The authors thank Ms. Judith Wood for help in preparation of this manuscript.
REFERENCES 1.
FIG. 8. Washout of U18666A restores progesterone production. Human granulosa-lutein cells were cultured with LDL (50 g/ml) in the absence or presence of U18666A (2 g/ml) as described in the text. The cells were then washed extensively and placed into a lipoprotein-deficient culture medium at 0 time. Progesterone released into the culture medium during the subsequent 48 h of incubation was determined. Values presented are means ⫾ SD from triplicate cultures in each treatment group. At 48 h of washout the U18666A cells were producing significantly more progesterone (P ⬍ 0.05) than control cells.
2.
U18666A (data not shown), and had increased levels of NPC1 mRNA. However, progesterone could not be used to perturb the NPC1 compartment in studies of steroidogenesis because exogenous progesterone could not be distinguished from hormone produced endogenously by the granulosa-lutein cells. Although 8-Br-cAMP stimulated progesterone production by human granulosa cells, the mass of steroid produced by these cells is relatively small (ng/mg protein) in contrast to the substantial quantities of progesterone produced by some primary luteinized granulosa cell cultures (g/mg protein) [35]. Thus, we would not have anticipated that the accumulated progesterone in the culture medium in the present studies would have resulted in a block to lysosomal free cholesterol efflux and a secondary impact on NPC1 expression. Preliminary reports of NPC1 expression in porcine ovarian cells indicate that luteinization of granulosa cells and treatment with gonadotropic hormones promotes an increase in NPC1 mRNA [24, 34]. In the porcine system, the gonadotropin-provoked increase in progesterone synthesis may have been sufficient to cause a block to lysosomal free cholesterol efflux and, therefore, indirectly resulted in an increase in NPC1 mRNA levels. It is possible that this action of progesterone has a physiological significance
5.
3.
4.
6.
7.
8.
9.
10.
11.
12.
Azhar, S., Tsai, L., Medicherla, S., Chandrasekher, Y., Guidice, L., and Reaven, E. (1998). Human granulosa cells use high density lipoprotein cholesterol for steroidogenesis. J. Clin. Endocrinol. Metab. 83, 983–991. Blanchette-Mackie, E. J., and Pentchev, P. G. (1998). Cholesterol distribution in Golgi, lysosomes and endoplasmic reticulum. In “Intracellular Cholesterol Transport” (D. Freeman and T. Y. Chang, Eds.), pp. 1–21. Kluwer Academic, Norwell, MA. Cadigan, K. M., Spillane, D. M., and Chang, T-Y. (1990). Isolation and characterization of chinese hamster ovary cell mutants defective in intracellular low-density lipoprotein cholesterol trafficking. J. Cell Biol. 110, 295–308. Carstea, E. D., Morris, J. A., Coleman, K. G., et al. (1997). Niemann-Pick C1 gene: Homology to mediators of cholesterol homeostasis. Science 277, 228 –231. Chang, T.-Y., Hasan, M. T., Chin, J., Chang, C. C., Spillane, D. M., and Chen, J. (1997). Chinese hamster ovary cell mutants affecting cholesterol metabolism. Curr. Opin. Lipid. 8, 65–71. Choi, Y.-S., and Freeman, D. A. (1999). The movement of plasma membrane cholesterol through the cell. In “Intracellular Cholesterol Transport” (D. Freeman and T. Y. Chang, Eds.), pp. 109 –121. Kluwer Academic, Norwell, MA. Freeman, D. A. (1987). Cyclic AMP mediated modification of cholesterol traffic in Leydig tumor cells. J. Biol. Chem. 262, 13061–13068. Freeman, D. A., and Ascoli, M. (1982). Studies on the source of cholesterol used for steroid biosynthesis in cultured Leydig tumor cells. J. Biol. Chem. 257, 14231–14238. Golos, T. G., Soto, E. A., Tureck, R. W., and Strauss III, J. F. (1985). Human chorionic gonadotropin and 8-bromo-adenosine 3⬘, 5⬘-monophosphate stimulate [125I]low density lipoprotein uptake and metabolism by luteinized human granulosa cells in culture. J. Clin. Endocrinol. Metab. 61, 633– 638. Gocze, P. M., and Freeman, D. A. (1993). Plasma membrane cholesterol is used as steroidogenic substrate in Y-1 mouse adrenal tumor cells and normal sheep adrenal cells. Exp. Cell Res. 209, 21–25. Gwynne, J. T., and Strauss III, J. F. (1982). The role of lipoproteins in steroidogenesis and cholesterol metabolism in steroidogenic glands. Endocr. Rev. 3, 295–329. Harmala, A. S., Porn, M. J., Mattjus, P., and Slotte, J. P. (1994). Cholesterol transport from plasma membrane to intracellular membranes is inhibited by 3-[2-(diethylamino)ethox]androst5-en-17-one. Biochim. Biophys. Acta 1211, 317–325.
66 13.
14. 15.
16.
17.
18. 19. 20.
21.
22.
23. 24. 25.
26.
27.
28.
WATARI ET AL. Kobayashi, T., Beuchat, M.-H., Lindsay, M., Frias, S., Palmiter, R. D., Sakuraba, H., Parton, R. G., and Gruenberg, J. (1999). Late endosomal membranes rich in lysobiphosphatidic acid regulate cholesterol transport. Nature Cell Biol. 1, 113–118. Kornfeld, R., and Kornfeld, S. (1985). Assembly of asparaginelinked oligosaccharides. Annu. Rev. Biochem. 50, 555–583. Lange, Y., Ye, J., and Steck, T. L. (1998). Circulation of cholesterol between lysosomes and the plasma membrane. J. Biol. Chem. 273, 18915–18922. Liscum, L., and Collins, G. J. (1991). Characterization of Chinese-hamster ovary cells that are resistant to 3-[2-(diethylamino)ethoxy]androst-5-en-17-one inhibition of low-density lipoprotein-derived cholesterol metabolism. J. Biol. Chem. 266, 16599 –16606. Liscum, L., and Faust, J. R. (1989). The intracellular transport of low-density lipoprotein-derived cholesterol is inhibited in Chinese-hamster ovary cells cultured with 3-[2-(diethylamino) ethoxy]androst-5-en-17-one. J. Biol. Chem. 264, 11796 –11806. Liscum, L., and Klansek, J. J. (1998). Niemann-Pick disease type C. Curr. Opin. Lipid. 9, 131–135. Liscum, L., and Munn, N. J. (1999). Intracellular cholesterol transport. Biochim. Biophys. Acta 1438, 19 –37. Loftus, S., Morris, J. A., and Carstea, E. D., et al. (1997). Murine model of Niemann-Pick C disease: Mutation in a cholesterol homeostasis gene. Science 277, 232–235. McAllister, J. M., Manson, I. J., and Byrd, W., et al. (1990). Proliferating human granulosa-lutein cells in long term monolayer culture: Expression of aromatase, cholesterol side-chain cleavage, and 3ß hydroxysteroid dehydrogenase. J. Clin. Endocrinol. Metab. 71, 26 –33. McCloskey, H. M., Rothblat, G. H., and Glick, J. M. (1987). Incubation of acetylated low density lipoprotein with cholesterol-rich dispersions enhances cholesterol uptake by macrophages. Biochim. Biophys. Acta 921, 320 –332. Miller, W. L. (1988). Molecular biology of steroid hormone synthesis. Endocr. Rev. 9, 295–318. Murphy, B. (1999). Luteinization: Models and markers. Biol. Reprod. 60, M16. Neufeld, E. B., Cooney, A. M., Pitha, J., Dawidowicz, E. A., Dwyer, N. K., Pentchev, P. G., and Blanchette-Mackie, E. J. (1996). Intracellular trafficking of cholesterol monitored with a cyclodextrin. J. Biol. Chem. 271, 21604 –21613. Neufeld, E. B., Wastney, M., Patel, S., Suresh, S., Cooney, A. M., Dwyer, N. K., Roff, C. F., Ohno, K., Morris, J. A., Carstea, E. D., Incardona, J. P., Strauss III, J. F., Vanier, M. T., Patterson, M. C., Brady, R. O., Pentchev, P. G., and BlanchetteMackie, E. J. (1999). The Niemann-Pick C1 protein resides in a vesicular compartment linked to retrograde transport of multiple lysosomal cargo. J. Biol. Chem. 274, 9627–9635. Nohturfft, A., DeBose-Boyd, R. A., Scheek, S., Goldstein, J. L., and Brown, M. S. (1999). Sterol regulate cycling of SREBP cleavage-activating protein (SCAP) between endoplasmic reticulum and golgi. Proc. Natl. Acad. Sci. USA 96, 11235–11240. Patel, S. C., Suresh, S., Kumar, U., Hu, C. Y., Cooney, A., Blanchette-Mackie, E. J., Neufeld, E. B., Patel, R. C., Brady, R. O., Patel, Y. C., Pentchev, P. G., and Ong, W.-Y. (1999). Localization of Niemann-Pick C1 protein in astrocytes: Impli-
Received October 13, 1999 Revised version received November 22, 1999
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
cations for neuronal degeneration in Niemann-Pick type C disease. Proc. Natl. Acad. Sci. USA 96, 1657–1662. Pollack, S. E., Furth, E. E., Kallen, C. B., Arakane, F., Kiriakidou, M., Kozarsky, K. F., and Strauss III, J. F. (1997). Localization of the steroidogenic acute regulatory protein in human tissues. J. Clin. Endocrinol. Metab. 82, 4243– 4251. Porpaczy, Z., Tomasek, J. J., and Freeman, D. A. (1997). Internalized plasma membrane cholesterol passes through an endosome compartment that is distinct from the acid vesicle-lysosome compartment. Exp. Cell Res. 234, 217–224. Reaven, E., Tsai, L., and Azhar, S. (1995). Cholesterol uptake by the selective pathway of ovarian granulosa cells. Early intracellular events. J. Lipid Res. 36, 1602–1617. Reaven, E., Tsai, L., and Azhar, S. (1996). Intracellular events in the “selective” transport of lipoprotein-derived cholesteryl esters. J. Biol. Chem. 271, 16208 –16217. Roff, C. F., Strauss III, J. F., Goldin, E., Jaffe, H., Patterson, M. C., Agritellis, G. C., Hibbs, A. M., Gardfield, M., Brady, R. O., and Pentchev, P. G. (1993). The murine Niemann-Pick type C lesion affects testosterone production. Endocrinology 133, 2924 –2934. Song, J. H., Dobias, M., Pescador, N., and Murphy, B. D. (1998). Luteinization and gonadotrophins up-regulate the NiemannPick gene in the porcine ovary. Biol. Reprod. 58, 188. Soto, E. A., Silavin, S. L., Tureck, R. W., Strauss III, J. F. (1984). Stimulation of progesterone synthesis in luteinized human granulosa cells by human chorionic gonadotropin and 8-bromo-adenosine 3⬘, 5⬘-monophosphate: The effect of low density lipoprotein. J. Clin. Endocrinol. Metab. 58, 831– 857. Stocco, D. M., and Clark, B. J. (1996). Regulation of the acute production of steroids in steroidogenic cells. Endocr. Rev. 17, 221–244. Strauss III, J. F., Schuler, L. A., Rosenblum, M. F., and Tanaka, T. (1981). Cholesterol metabolism by ovarian tissue. Adv. Lipid Res. 18, 99 –157. Sugawara, T., Holt, J. A., and Driscoll, D., et al. (1995). Human steroidogenic acute regulatory protein: Functional activity in COS-1 cells, tissue-specific expression, and mapping of the structural gene to 8p11.2 and a pseudogene to chromosome 13. Proc. Natl. Acad. Sci. USA 92, 4778 – 4782. Underwood, K. W., Jacobs, N. L., Howley, A., and Liscum, L. (1998). Evidence for a cholesterol transport pathway from lysosomes to endoplasmic reticulum that is independent of the plasma membrane. J. Biol. Chem. 273, 4266 – 4274. Underwood, K. W., Andemariam, B., McWilliams, G. L., and Liscum, L. (1996). Quantitative analysis of hydrophobic amine inhibition of intracellular cholesterol transport. J. Lipid Res. 37, 1556 –1568. Watari, H., Blanchette-Mackie, E. J., Dwyer, N. K., Glick, J. M., Patel, S., Neufeld, E. B., Brady, R. O., Pentchev, P. G., and Strauss III, J. F. (1999). Niemann-Pick C1 protein: Obligatory roles for N-terminal domains and lysosomal targeting in cholesterol mobilization. Proc. Natl. Acad. Sci. USA 96, 805– 810. Watari, H., Blanchette-Mackie, E. J., Dwyer, N. K., Watari, M., Neufeld, E. B., Patel, S., Pentchev, P. G., and Strauss III, J. F. (1999). Mutations in the leuciune zipper motif and sterol-sensing domain inactivate the Niemann-Pick C1 glycoprotein. J. Biol. Chem. 274, 21861–21866.
