Placenta 32 (2011) 222e229
Contents lists available at ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
Mitochondrial heat shock protein participates in placental steroidogenesis S. Olvera-Sanchez, M.T. Espinosa-Garcia, J. Monreal, O. Flores-Herrera, F. Martinez* Departamento de Bioquimica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70-159, 04510 Mexico City, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 15 December 2010
The human placenta, which does not express the StAR protein, synthesizes large amounts of progesterone. The rate-limiting step for steroidogenesis is the transport of cholesterol which is divided into two steps: 1) cholesterol flux from cytoplasm to outer membrane mitochondria, and 2) cholesterol transport from outer to inner mitochondrial membrane. The proteins mediating placental cholesterol influx have not been clearly identified. We investigated the proteins involved in the transport of cholesterol in syncytiotrophoblast mitochondria from human placenta. Two proteins, one of 30 kDa, and another of 60 kDa, were identified using anti-MLN64 antibodies. The 30 kDa protein corresponds to a fragment of MLN64, and the 60 kDa protein was identified as a heat shock protein. During steroidogenesis, mitochondria released MLN64 protein to supernatant. When this supernatant was added to fresh isolated mitochondria, progesterone synthesis increased; a similar result was obtained with the addition of the recombinant MLN64-START protein. In the presence of flurescein-5-maleimide or N-ethyl-maleimide, the mitochondrial synthesis of progesterone was inhibited in a dose-dependent fashion without changes in mitochondrial respiration. 2D-electrophoretic pattern showed that flurescein-5-maleimide- fluorescence was associated with HSP60. Both MLN64 and HSP60 were identified in mitochondrial contact sites. The results suggest that HSP60 is involved in the steroidogenic metabolism of human placenta. A tight association between MLN64 and HSP60 is suggested for cholesterol transport in the human placenta. Ó 2010 Elsevier Ltd. All rights reserved.
Keywords: Steroidogenesis Heat shock proteins Cholesterol transport MLN64 Human placenta mitochondria Maleimides
1. Introduction The mechanisms controlling steroidogenesis in human placental mitochondria are still unknown. However, there is enough evidence suggesting that cholesterol transport to mitochondria and between the mitochondrial membranes are rate-limiting steps. Similarly to other steroidogenic tissues, contact sites have been proposed as the gate for cholesterol to P450scc in placental mitochondria [1]. Observations indicate that all necessary elements and enzymes for steroidogenesis are present and functional in placental mitochondrial contact sites. This organization may facilitate the metabolism of cholesterol delivered to the outer mitochondrial membrane into steroid hormones by the inner mitochondrial membrane cholesterol side chain cleavage system [2]. In this sense, steroidogenic regulatory protein (StAR) interacts with voltage-dependent channel 1 (VDAC1) and with phosphate carrier protein (PCP); moreover, it has been suggested that this interaction is part of a larger outer mitochondrial membrane (OMM)
Abbreviations: FM, fluorescein-5-maleimide; NEM, N-ethyl-maleimide; P4M, progesterone synthesis medium; ANT, adenine nucleotide translocase. * Corresponding author. Tel.: þ52 (55) 5623 2168; fax: þ52 (55) 5616 2419. E-mail address:
[email protected] (F. Martinez). 0143-4004/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.placenta.2010.12.018
complex including StAR, VDAC1, PCP [3] and unidentified 60 kDa protein [4]. Also, the sterol carrier protein-2 (SCP-2) was found in the adrenal gland, testis and ovary, suggesting that SCP-2 could play a role in steroidogenesis [5]. Proteins localized in OMM and contact sites like PKAR1a, PAP7, and the interaction between StAR and translocator protein (TSPO), are necessary for importing cholesterol into the mitochondria [6,7], suggesting that multiple protein associations are required to transport cholesterol during mitochondrial steroidogenesis. The human placenta does not transcribe the StAR gene, but it synthesizes large amounts of progesterone [8,9]. Thus, in human placental steroidogenesis there is a StAR-independent mechanism to transport cholesterol into the mitochondria. MLN64 is an endosome protein of 54 kDa highly expressed in breast cancers, which has a START domain to bind cholesterol in its C-terminal [10]; both StAR and MLN64 have a START domain to bind cholesterol. Truncated forms of MLN64 have StAR activity and may have a direct role in fostering the flux of cholesterol from the mitochondrial membrane. Bose has proposed that MLN64 is processed in vivo to a product similar to N-218 MLN64, which then acts to promote steroidogenesis by the same mechanism as StAR [11,12]. Although in murine models the deletion of MLN64 has mild effects on the reproductive functions [13], and it has been demonstrated that MLN64 and StAR are differently located in cells,
S. Olvera-Sanchez et al. / Placenta 32 (2011) 222e229
223
the function of MLN64 remains unknown [14]. However, it has been found that MLN64 mRNA is expressed in the human placenta, and that recombinant MLN64 protein containing the START domain stimulates progesterone synthesis in isolated syncytiotrophoblast mitochondria, suggesting that MLN64 could function as protein carrier for cholesterol in the placenta [11,15]. Several proteins have been identified by Western blots against MLN64 in whole human placental homogenate. Among these, a 60 kDa protein was the most abundant, compared to 30 kDa MLN64 [13]. Mitochondrial contact sites allow cholesterol flow from outer to inner membrane [16]. Ultrastructural observations of adrenal gland mitochondria stimulated with ACTH have shown an increased number of contact sites [17], as well as an increase of cholesterol transport at semi-purified contact sites [18]. We have demonstrated that purified contact sites of human syncytiotrophoblast mitochondria exhibitied the 90, 72, 40 and 27 kDa heat shock proteins, cytochromes b, c1 and P450scc, MLN64, porin (VDAC), NADP-isocitrate dehydrogenase, and ATPdiphosphohydrolase [2]. This result suggests that multiple proteins are involved in the transport of cholesterol from and between mitochondrial membranes. Since knowledge of proteins involved in the mitochondrial cholesterol transport from human placenta is limited [1,2], the aim of this study was to determine the proteins associated with placental steroidogenesis in isolated syncytiotrophoblast mitochondria. The results showed a protein of 60 kDa corresponding to the heat shock protein HSP60, associated to MLN64, which participates in placental steroidogenesis. 2. Materials and methods 2.1. Materials The chemical reagents were purchased from Sigma Chemical Co. (St Louis), MO, USA. Complete (protease inhibitor cocktail) and Protein A-agarose, from Roche Applied Science (Mexico). Anti-rabbit anti-MLN64 (PA1-562), from ABR Inc. (Thermo Scientific, USA). Donkey anti-goat anti-MLN64 (sc-26062), and antibodies against VDAC1 (sc-8829), Rab5 (sc-309), HSP60 N-terminal (sc-13115) or HSP60 C-terminal (sc-13966), ANT (sc-11433), from Santa Cruz Biotechnology, Inc., USA. Anti-NPC1 (BC 400-126) antibody, from Novus Biologicals (USA); Super Signal West Pico (Chemiluminescence detection system), N-ethyl-maleimide (NEM), and fluorescein-5-maleimide (FM), from Pierce Biotechnology (IL, USA); culture cell media DMEM, and FBS, from Gibco (USA); Coat-a-CountÒ Progesterone Kit for progesterone determinations, from Siemens Healthcare Diagnostic, USA; Amicon Ultra Centrifugal Filter system (PL-10, 10000 Nominal Molecular Weight Limit), from Millipore Corporation (USA); Started-Kit, Ready Strips, and materials for electrophoresis, from Bio-Rad (USA). Other reagents were of analytical grade. 2.2. Isolation of syncytiotrophoblast mitochondria and steroidogenic contact sites Following approval by the government and an ethics committee, human placentas at term were collected from a hospital. Syncytiotrophoblast mitochondria from the human placenta (hereinafter referred to as mitochondria) were isolated as previously described [19], with modifications as observed in Fig. 1. Protein concentration was determined by the Bradford method using bovine serum albumin as standard [20]. Oxygen consumption was determined polarographically using a Clark type electrode [21]. Only syncytiotrophoblast mitochondria with respiratory control higher than three were used in order to assure integrity of mitochondrial membranes. Contact sites were isolated as reported by Uribe et al. [2]. 2.3. Succinate dehydrogenase activity The activity of succinate dehydrogenase was assessed as described elsewhere [22], by reducing cytochrome c. An extinction coefficient of 22.7 mmol1 cm1 was used to calculate the specific activity, which was reported as mmol of cytochrome c reduced/mg/min. 2.4. Progesterone synthesis Progesterone synthesis was started by incubating intact mitochondria (1 mg/ml) at 37 C during 20 min in a medium called P4M containing 10 mM MOPS, 0.5 mM EGTA, 120 mM KCl, pH 7.4, supplemented with the protease inhibitor Complete (Roche) and 5 mM isocitrate as an oxidizable substrate of mitochondrial isocitrateNADP dehydrogenase, which feeds the redox chain associated to cytochrome
Fig. 1. Identification of proteins recognized by anti-MLN64-START antibodies in the steps of syncytiotrophoblast mitochondrial isolation. A) The scheme represents the different steps accomplished in order to isolate syncytiotrophoblast mitochondria. Numbers below indicate the relative centrifuge force. B) Immunoblot with rabbit antiMLN64 antibody of different placental fractions (S1, S2, S3; P1, P2, P3, P4; SM). C) Succinate dehydrogenase activity (mmol of cytochrome c reduced/mg/min) in different fractions (S1, S2, S3; P1, P2, P3, P4). S1, S2 and S3 ¼ supernatant 1, 2 and 3. P1, P2, P4 and P4 are the different pellets containing mitochondria. Syncytiotrophoblast mitochondria (SM) were washed with KCl as described in the Materials and Methods Section; these mitochondria were used in this study. This is a representative result from three different placentas.
P450scc. When mitochondria were incubated in a medium without substrates, no synthesis of progesterone was observed. However, if an oxidizable substrate to produce NADPH such as isocitrate is added to the incubation media, the progesterone synthesis was observed, as previously was reported [23]. The reaction was stopped with methanol at indicated times and the mixture was placed in ice-cold bath. Some experiments were performed in the presence of the protease inhibitor cocktail (CI-DMSO) and 1 mM PMSF. The effect of 0.1 nMe3 mM N-ethyl-maleimide (NEM) or 5-Fluorescein-maleimide (FM) on steroidogenesis was assessed in mitochondria (50 mg/ml) or contact sites (50 mg/ml) incubated in its presence for 45 min at 4 C. Then, mitochondria were incubated in P4M medium while contact sites in the same medium were supplemented with 1 mM ADP and 1 mM Pi at 37 C [17]. 22-(R)-OH-cholesterol was used to determine the highest progesterone synthesis. In some experiments, mitochondria were recovered after incubation by centrifugation at 9200 g for 10 min, at 4 C. Supernatant was concentrated in an Amicon Ultra Centrifugal Filter system; then the succinate dehydrogenase activity was determined and the identity of some proteins were obtained by Western blot. The supernatant of various mitochondrial preparations were mixed in order to study its effect on steroidogenesis. Steroidogenesis in JEG-3 human choriocarcinoma cells was performed in cultures at 37 C under a 95% air, 5% CO2 humidified atmosphere in DMEM medium supplemented with 10% FBS, 2 mM glutamine, 100 IU/ml penicillin, 100 mg/ml streptomycin, 1 mM sodium pyruvate, 20 mM HEPES, pH 7.4. Cells were cultured in a 4-well plate at a density of 1.2 105/well. When cells reached 50e60% confluence, they were treated with 0.1e500 nM FM for 24, 48 or 72 h. 25 mM 22-(R)-OHcholesterol was added as control. After treatment, cells were incubated in lysis buffer containing 150 mM NaCl, 5 mM EDTA, 5 mM TriseHCl, pH 9, 1% Nonidet-P40 and a mixture of protease inhibitors and then analyzed by 2D-electrophoresis and Western blot using anti-MLN64 and anti-HSP60 antibodies. Progesterone was
224 determined by radioimmunoassay using manufacturer’s instructions.
S. Olvera-Sanchez et al. / Placenta 32 (2011) 222e229 125
I-progesterone Kit, following the
The samples were observed with a confocal microscope. Anti-MLN64 (CY2) was labeled in red and anti-HSP60 N-terminal (CY7) was labeled in green.
2.5. Western blot analysis
2.9. Statistical analysis
Equal amounts of samples (50mg/lane) were loaded into a 10% SDS-PAGE as described elsewhere [24] and Western blot was carried out as described [25]. The dilutions of antibodies were: anti-MLN64 (1:2000), anti-HSP60 N-terminal (1:10,000), anti-HSP60 C-terminal (1:10,000), anti-VDAC1 (1:2500), anti-Rab5 (1:500) and antiNPC1 (1:2000). After washing, blots were incubated with a corresponding second antibody. Protein-antibody complexes were visualized by an enhanced chemiluminescence detection system, according to the manufacturer’s instructions.
Data are expressed as mean SD of at least three independent experiments carried out with three different placental preparations. Statistical analysis was performed as indicated in each figure legend.
2.6. 2D-electrophoresis The isoelectrofocusing (IEF) of different samples was performed using Ready Strips from pH 5e8 or 3e10 in a ProteanÒ IEF Cell apparatus. After IEF, proteins were resolved by its molecular weight in a 10% SDS-PAGE, electrotransfered to PVDF membranes, and processed to Western blot analysis with anti-MLN64 or anti-HSP60 antibodies, as mentioned above. Specific spots of proteins from 2D-electrophoresis were sent to the Core Laboratory for Protein Microsequencing of the University of Massachusetts, USA, to determine their identity. 2.7. Immunoprecipitation Mitochondria (10 mg) were incubated for 20 min at 37 C in P4M medium, followed by lysis treatment, and incubated overnight with anti-MLN64, anti-HSP60 (C-terminal), or anti-HSP60 (N-terminal) antibodies. Then, mitochondria were incubated with Protein G-Agarose (Roche), according to the manufacturer’s instructions. Samples were processes by Western blot to identify VDAC, ANT, HSP60 (C-terminal), HSP60 (N-terminal), or MNL64 proteins. 2.8. Immunohistology Cotyledons from the human placenta were dissected and fixed in 30% glutaraldehyde, as described previously [26] with cuts of 10 mm. After overnight incubation with anti-MLN64 and anti-HSP60 antibodies (dilution 1:1000), the second fluorescent antibodies with different color markers were added at appropriate dilutions.
3. Results Fig. 1A shows the differential centrifugation steps for mitochondrial isolation from the human placenta. Western blots against MLN64 from the different fractions showed mainly one band of 30 kDa in the supernatants S1, S2 and S3, while 60 and 75 kDa proteins were associated with P1, P2, P4 and mitochondrial samples (Fig. 1B). Although other proteins of low and high molecular mass were observed, their presence was irregular and no further characterization done. In supernatants S1 and S2, the specific activity of succinate dehydrogenase, an enzyme exclusively found in mitochondria, was 6.8 and 7.7 mmol of cytochrome c reduced/mg/min, respectively (Fig. 1C); this activity was not observed in supernatant S3. Fractions P1, P2, P3, P4, and mitochondria showed high succinate dehydrogenase activity, while Western blots against MNL64 showed mainly a protein of 60 kDa (Fig. 1, B and C). Notably, mitochondria had the highest specific succinate dehydrogenase activity, with 111.3 mmol of cytochrome c reduced/mg/min. NPC1, an exclusive protein of endosomes, was not detected in the mitochondrial fraction by Western blot analysis (Supplemental Fig. 1). To determine the role of the 60 kDa protein in steroidogenesis, mitochondria were incubated in the P4M medium supplemented with isocitrate, and at the indicated times the samples were
Fig. 2. Placental mitochondria electrophoretic pattern, 2D-electrophoresis and 60 kDa proteins sequence. A) Placental mitochondria were incubated 20 min at 37 C in P4M and centrifuged. Mitochondrial pellet and supernatant were separated and processed for SDS-PAGE. Western blot analysis using anti-MLN64 antibody was carried out as described in the Materials and Methods Section. Mitochondrial pellet (line 1) and supernatant (line 2) supernatant was placed in an Amicon Ultra Centrifugal Filter System with 10 kDa molecular cut membrane and concentrated at 0.5 ml (line 4). The filtrate was also collected (line 3). The concentrated sample was treated with methanol, centrifuged and the supernatant (line 5) and the pellet (line 6) were collected. Progesterone determination was performed for each fraction. B) Isoelectrofocusing (pH 5e8) from placental mitochondria was carried out as described in the Materials and Methods Section. C) Western blot of isoelectrofocusing was performed using the anti-MLN64 antibody to detect proteins containing the START domain for 15 or 60 s of exposure. Spot 1 showed a MW 59.50 kDa with a pI 5.82 0.029. Spot 2 showed a MW 58.90 kDa with a pI 5.92 0.027. Spot 3 shows a MW 58.45 kDa with a pI 5.99 0.025. D) The amino terminal sequencing of spot 1 and 2 corresponded to heat shock proteins of 60 kDa. For progesterone, the values are the mean SD from three independent determinations carried out with three different placental preparations.
S. Olvera-Sanchez et al. / Placenta 32 (2011) 222e229
centrifuged at 10,000 g for 20 min, and the supernatant and mitochondria were recovered separately to identify the proteins of 30 and 60 kDa by Western blot against MLN64. The Western blot against MLN64 showed a band of 60 kDa in mitochondria (Fig. 2A, lane 1), while in the supernatant the most abundant was a protein of 30 kDa (Fig. 2A, lane 2). The ultra-concentration of supernatant proteins showed that 87.9% 7.7% of the progesterone synthesized remained associated to the protein fraction (Fig. 2A, lane 4), while in the filtrate only 12% of progesterone was detected (Fig. 2A, lane 3), suggesting an association between progesterone and the proteins from this fraction. When the ultra-concentrate of supernatant proteins was denatured with methanol and precipitated by centrifugation, 95.5% 5.6% of progesterone was recovered in the protein freesupernatant (Fig. 2A, lane 5) while proteins of 60 and 30 kDa were identified with antiMLN64 in the precipitate (Fig. 2A, lane 6). Additionally, after the incubation and centrifugation steps, the respiratory control of mitochondria was similar to non-incubated mitochondria (Supplemental Fig. 2), and the succinate dehydrogenase activity was not detected in supernatant, suggesting that mitochondria remained unbroken. However, several proteins were released in the supernatant, which showed a different electrophoretic pattern compared to proteins from mitochondria (Supplemental Fig. 3). Three proteins of 60 kDa were recognized by antibodies against MLN64 in a 2D-electrophoresis of mitochondria (Fig. 2B), although at a short time of exposure only one protein was recognized (spot 2 from Fig. 2B). Proteins 1 and 2 were sequenced and identified as heat shock proteins of 60 kDa (HSP60; NCBI ID: 31542947) (Fig. 2C and Supplemental Fig. 4). Protein 1 showed a MW of 59.5 kDa with
225
a pI of 5.8 0.03; protein-2 had a MW 58.9 kDa with a pI of 5.9 0.03, suggesting that these proteins could be isoforms of HSP60. Western blot of 2D-electrophoresis showed an identical pattern with both anti-MNL64 and anti-HSP60 antibodies (Fig. 3A), suggesting a cross reaction. However, mitochondrial progesterone synthesis in the presence of antibodies against anti-HSP60 or antiMNL64 (1:40 dilution each) was 97.0% 4.0% and 91.4% 7.6%, respectively, suggesting that proteins were not accessible to these antibodies. The addition of ultra-concentrates of supernatant proteins (60e100 mg per assay) to fresh mitochondria increased de novo progesterone synthesis 1.98 0.37 times compared to control. Similar results were obtained when 1 mM recombinant MLN64START was added to intact mitochondria, where de novo progesterone synthesis increased 2.59 0.44 times. After mitochondrial protein immunoprecipitation with anti-MLN64 antibody, both MLN64 and HSP60 proteins were recognized by Western blot at the same molecular weight (Fig. 3B). Similarly, the immunoprecipitation with anti-HSP60 antibody showed the presence of MLN64 protein (Fig. 3C). When mitochondria were pre-incubated in P4M medium for 20 min at 37 C, a protein of 30 kDa was identified with both antibodies (Fig. 3B, lane 1). The co-immunoprecipitation of MLN64 and HSP60 protein suggests an association between these two proteins. When mitochondria were incubated at 4 C (Fig. 3B, lane 2), a lower amount of the 30 kDa protein was detected with both antibodies, suggesting a decrease in the enzymatic transformation of HSP60. This was confirmed by immunolocation of HSP60 and MLN64 in placental tissue, which co-localize in the syncytiotrophoblast cells (Fig. 3D).
Fig. 3. Immunoprecipitation and immunolocalization of HSP60 and MLN64. A) Western blot of 2D-electrophoresis (pH 5e8) from placental mitochondria were carried out as described in the Materials and Methods Section, using anti-HSP60 and anti-MLN64 antibodies. B) Syncytiotrophoblast mitochondria (10 mg of protein) were immunoprecipitated with anti-MLN64 and anti-HSP60 antibodies and identified with anti-MLN64 and anti-HSP60 C-terminal antibodies, respectively. Line 1: Mitochondria incubated in P4M for 20 min at 37 C. Line 2: Mitochondria incubated in P4M for 20 min at 4 C. C) Mitochondria (10 mg of protein) were incubated at 37 C for 20 min and immunoprecipitated with anti-HSP60 N-terminal (line 1) or anti-HSP60 C-terminal (line 2) antibodies, then MLN64 protein was detected with anti-MLN64 antibody. D) Cotyledons from human placentas were dissected, fixed and then incubated with anti-HSP60 N-terminal (green) and anti-MLN64 (red) antibodies. Samples were observed with a confocal microscopy. The yellow staining indicates co-localization of both antibodies in the syncytiotrophoblast. This is a representative result from three different experiments. Positions of molecular weight markers (kDa) are shown on the left.
226
S. Olvera-Sanchez et al. / Placenta 32 (2011) 222e229
To determine the participation of HSP60 in steroidogenesis, the progesterone synthesis from mitochondria was determined in the presence of FM (Fig. 4A). The results showed inhibition of steroidogenesis between 50 and 75 nM FM; this effect seems to be selective, since oxygen consumption was unaltered at the same FM concentrations (Fig. 4B). Similar results were obtained with NEM (data not shown). In 2D-electrophoresis after incubation of mitochondria plus 500 nM FM, the 60 kDa protein was the most intensively labeled with the fluorescent probe; additionally, this protein was detected with both anti-HSP60 and anti-MLN64 antibodies (Fig. 4CeE, respectively). Similar results were observed in steroidogenic contact sites (Fig. 5A), where a 60 kDa protein was detected with both anti-HSP60 or anti-MLN64 antibodies (Fig. 5B). Progesterone synthesis was inhibited about 70% when steroidogenic contact sites were incubated in the presence of 10 nM FM (Fig. 5C), suggesting that the chemical labeling of HSP60 with FM affects steroidogenesis. To explore the role of HSP60 in placental steroidogenesis, JEG-3 cells were incubated with FM. The results showed that the viability of JEG-3 cells remained constant during three days with or without 500 nM FM; and the 2D-electrophoresis showed several proteins labeled by FM, among which the 60 kDa protein had the highest fluorescence (data not shown). Progesterone synthesis in JEG-3 cells was inhibited in the presence of 500 nM FM (Fig. 5D). However, the addition of 22-(R)-OH-cholesterol (25 mM) restored the maximal cell steroidogenesis (Fig. 5D), suggesting that cytochrome P450 remains active.
4. Discussion Steroidogenesis requires the transport of cholesterol between mitochondrial membranes, a process thought to be mediated by proteins located at contact sites [2]. MLN64 has been suggested to transport cholesterol at mitochondrial level in the human placenta [11,12]. Watari et al. [11] have concluded that MLN64 is directly involved in lysosomal cholesterol mobilization, and that it may have an additional role in steroidogenesis. This group has also suggested that unknown proteins may promote the function of MLN64 in the late endosome, as well as in cholesterol transport. Thus, proteineprotein interactions appear to be a requirement for cholesterol transport between mitochondrial membranes, as described in Leydig cell mitochondria, where TSPO, StAR, PAP7 and PKA have been associated with cholesterol transport [3,27]. Similarly, placental steroidogenesis requires several proteins [1]; where HSP60 and MLN64 appear to be necessary for this process. Although Alpy et al. [28] have suggested that the role of MLN64 in steroidogenesis is unlikely, because mice lacking the MLN64-START domain did not show defects in steroidogenesis [12], both studies only involved the analysis of cells expressing StAR and MLN64 proteins. Given that a StAR protein is not present in the human placenta, the role of MLN64 in the steroidogenesis of this organ may be more significant. Similarly, TSPO protein is necessary for steroidogenesis in Leydig cells [29], but not in BeWo cells or placental mitochondria [30]. This suggests that each steroidogenic tissue has a particular set of proteins involved in cholesterol transport.
Fig. 4. Effect of FM on progesterone synthesis and mitochondrial oxygen uptake. A) Syncytiotrophoblast mitochondria (1 mg/ml) were incubated at 4 C for 45 min in the presence of FM (0.1 nMe1 mM). Then, progesterone synthesis was performed in P4M medium at 37 C for 20 min. 22-OH-Chol (25 mM) was used as control of progesterone synthesis. Mitochondria were incubated in respiratory medium at 37 C for 2e3 min for state 4 (data not shown). B) Effect of FM (0.1e500 nM). State 3 oxygen consumption was obtained by adding of ADP (100e200 mM). C) Isoelectrofocusing (pH 5e8). Placental mitochondria protein labeled with FM (500 nM). Mitochondria were incubated for 45 min at 4 C in the presence of 500 nM FM. The samples were processed by 2D-SDS-PAGE (pH 5e8) and subjected to Western Blot analysis using D) anti-HSP60 or E) anti-MLN64 antibodies as described in the Materials and Methods Section. For progesterone synthesis, values are the mean SD of eight measurements carried out with different placental preparations. Statistical difference for progesterone synthesis was analyzed by Dunnet test (p < 0.05 p < 0.006). For oxygen consumption the statistical difference was analyzed by Tukey test compared “a” vs “b” (p > 0.4) and “a” vs “c” (p > 0.05).
S. Olvera-Sanchez et al. / Placenta 32 (2011) 222e229
227
Fig. 5. Isoelectrofocusing (pH 3e10) and progesterone synthesis in steroidogenic contact sites and JEG-3 cells. A) Steroidogenic contact sites were incubated for 45 min at 4 C in the presence of 500 nM FM. The samples were processed to isoelectrofocusing (pH 3e10). B) Western blot analysis of previous samples using anti-MLN64 and anti-HSP60 antibodies was performed as described earlier in the Materials and Methods Section. C) Progesterone synthesis by steroidogenic contact sites. The samples were incubated 20, 40, or 60 min at 37 C in P4M medium in the presence (gray bars) or absence (black bars) of FM and progesterone synthesis was assayed as described in the Materials and Methods Section. D) Progesterone synthesis in JEG-3 cells. JEG-3 cells were incubated for 24, 48 or 72 h with 500 nM FM and progesterone synthesis was determined using an aliquot of supernatant from incubation medium as described in the Materials and Methods Section. Data are expressed as mean SD of at least three independent experiments. Statistical difference for progesterone synthesis by steroidogenic contact sites was analyzed by GameseHowell test (p < 0.05 p < 0.001). For progesterone synthesis in JEG-3 cells the statistical difference was analyzed by Dunnet test compared test (p < 0.05 p < 0.01).
The evidence of MLN64 participation in placental steroidogenesis was evaluated by cotransfections of cells [10], in vitro incubation of isolated mitochondria with recombinant MLN64-START [15], or its identification by Western blot [2]. Western blot analysis of isolated syncytiotrophoblast mitochondria with an antibody against the Cterminal of MLN64 identified several proteins, in agreement with Watari et al. [11], among which the most abundant were the 60 and 30 kDa proteins. The 30 kDa protein is the active form of MNL64 identified previously [11,12,15]. The formal identification after the Western blot was made by sequencing of the 60 kDa protein. This protein corresponded to an HSP60, according to the N-terminal sequence. It was recognized with anti-MLN64 and anti-human-HSP60 antibodies, suggesting the presence of similar antigenic epitopes. Sequence analysis showed that 65 and 55 amino acids of HSP60 overlapped with StAR and MLN64, and an identity of 26.1 and 18.6% was estimated, respectively (Supplemental Fig. 5). We named the 60 kDa protein HSP60-like. Syncytiotrophoblast mitochondria were analyzed after incubation for progesterone synthesis and MLN64 was recognized by Western blot, suggesting that the HSP60-like protein could be a precursor involved in cholesterol movement between the mitochondrial membranes. This process implies the presence of proteases required for placental steroidogenesis, since recombinant MLN64START added to mitochondria was degraded. The proteolysis of StAR protein has been similarly described in mitochondria from the adrenal gland [31]. Proteolysis in syncytiotrophoblast mitochondria was not modified by protease inhibitors, but the appearance of the 30 kDa protein diminished significantly in supernatant when mitochondria were
incubated at 4 C (Supplemental Fig. 6), suggesting that proteolysis also occurs at the level of the mitochondrial membranes. Interestingly, it was not possible to observe the proteolysis of HSP60-like in sonicated mitochondria (data not shown), indicating that integrity of mitochondria is required. The common location of HSP60-like and MLN64 in syncytiotrophoblast suggests that both contribute to placental steroidogenesis. One role of the HSP60-like protein could be, besides the movement of cholesterol, the transport of progesterone from syncytiotrophoblast to maternal blood, since the mitochondrial matrix recovered by centrifugation after 20 min of incubation for progesterone synthesis, and incubated with fresh medium for a second steroidogenic cycle, retained an important percentage of the new progesterone (data not shown). It is likely that in vivo the exchange of these proteins keeps progesterone from being constantly released from the mitochondria, thus maintaining pregnancy. Even more, the HSP60-like protein and the 30 kDa protein from the supernatant retained more than 90% of progesterone, which was released after treatment with methanol, suggesting that progesterone is closely associated to these proteins. Interestingly, HSP60 plays several physiological roles in addition to functioning as a chaperone, since it has been associated with a number of autoimmune diseases, such as Alzheimer’s disease, heart disease, and diabetes [32e34]. Moreover, the association between antibodies against HSP60 and high serum cholesterol has been proposed as the beginning of the atherosclerosis in human beings [35,36]. Thus, the possible role of the HSP60-like protein in cholesterol distribution, as in placental steroidogenesis, becomes relevant.
228
S. Olvera-Sanchez et al. / Placenta 32 (2011) 222e229
In this sense, HSP60 has three cysteine residues, being Cys-442 and 447 related to its activity [37], and the blocking of Cys-442 inhibits its functions [38,39]. An alternative way to inhibit the activity of proteins containing cysteine is using maleimides [40]. Our results show that progesterone synthesis was inhibited at nanomolar concentrations of NEM or FM even in syncytiotrophoblast mitochondria, steroidogenic contact sites, or JEG-3 cells. The inhibition by maleimides is apparently specific for steroidogenesis; given that oxygen consumption and cellular viability remained unaltered, the transporters and the enzymes related to respiration or cell survival were most likely not affected by maleimides. Moreover, HSP60 was the main protein labeled with FM in all biological models used in this study. These data strongly support the assumption that the HSP60-like protein is related to placental steroidogenesis. Finally, our results support the assumption that multiprotein complexes are required to move cholesterol between mitochondrial membranes in the human placenta, as we [41,42] and other authors [43,44] have proposed. Acknowledgments This work was partially supported by Grants IN203006 and IN217609 from the Dirección General de Apoyo al Personal Académico de la Universidad Nacional Autónoma de México. Jessica Monreal is student of the Biomedical Ph.D. Program, UNAM, and has a Grant from CONACYT to support her doctoral studies. The authors thank Dr. Lilia Graue Olmos and Dr. José Luis Pérez García for checking the proper English usage in this manuscript. Appendix A. Supplementary material
[12]
[13]
[14] [15]
[16] [17] [18]
[19]
[20]
[21]
[22]
[23]
[24] [25]
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.placenta.2010.12.018.
[26]
References
[27]
[1] Espinosa-Garcia MT, Strauss 3rd JF, Martinez F. A trypsin-sensitive protein is required for utilization of exogenous cholesterol for pregnenolone synthesis by placental mitochondria. Placenta 2000;21:654e60. [2] Uribe A, Strauss 3rd JF, Martinez F. Contact sites from human placental mitochondria. Characterization and role in progesterone synthesis. Arch Biochem Biophys 2003;413:172e81. [3] Bose M, Whittal RM, Miller WL, Bose HS. Steroidogenic activity of StAR requires contact with mitochondrial VDAC1 and phosphate carrier protein. J Biol Chem 2008;283:8837e45. [4] McEnery MW, Snowman AM, Trifiletti RR, Snyder SH. Isolation of the mitochondrial benzodiazepine receptor: association with the voltage-dependent anion channel and the adenine nucleotide carrier. Proc Natl Acad Sci U S A 1992;89:3170e4. [5] Gallegos AM, Atshaves BP, Storey SM, Starodub O, Petrescu AD, Huang H. Gene structure, intracellular localization, and functional roles of sterol carrier protein-2. Prog Lipid Res 2001;40:498e563. [6] Li H, Degenhardt B, Tobin D, Yao ZX, Tasken K, Papadopoulos V. Identification, localization, and function in steroidogenesis of PAP7: a peripheral-type benzodiazepine receptor- and PKA (RIalpha)-associated protein. Mol Endocrinol; 2001:2211e28. [7] Hauet T, Yao ZX, Bose HS, Wall CT, Han Z, Li W. Peripheral-type benzodiazepine receptor-mediated action of steroidogenic acute regulatory protein on cholesterol entry into leydig cell mitochondria. Mol Endocrinol 2005;19:540e54. [8] Sugawara T, Holt JA, Driscoll D, Strauss 3rd JF, Lin D, Miller WL, et al. 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 U S A 1995;92:4778e82. [9] Pollack SE, Furth EE, Kallen CB, Arakane F, Kiriakidou M, Kozarsky KF, et al. Localization of the steroidogenic acute regulatory protein in human tissues. J Clin Endocrinol Metab 1997;82:4243e51. [10] Moog-Lutz C, Tomasetto C, Regnier CH, Wendling C, Lutz Y, Muller D, et al. MLN64 exhibits homology with the steroidogenic acute regulatory protein (STAR) and is over-expressed in human breast carcinomas. Int J Cancer 1997;71:183e91. [11] Watari H, Arakane F, Moog-Lutz C, Kallen CB, Tomasetto C, Gerton GL, et al. MLN64 contains a domain with homology to the steroidogenic acute
[28] [29]
[30]
[31]
[32]
[33]
[34]
[35] [36]
[37] [38]
[39]
regulatory protein (StAR) that stimulates steroidogenesis. Proc Natl Acad Sci U S A 1997;94:8462e7. Bose HS, Whittal RM, Huang MC, Baldwin MA, Miller WL. N-218 MLN64, a protein with StAR-like steroidogenic activity, is folded and cleaved similarly to StAR. Biochemistry 2000;39:11722e31. Kishida T, Kostetskii I, Zhang Z, Martinez F, Liu P, Walkley SU, et al. Targeted mutation of the MLN64 START domain causes only modest alterations in cellular sterol metabolism. J Biol Chem 2004;279:19276e85. Alpy F, Tomasetto C. Give lipids a START: the StAR-related lipid transfer (START) domain in mammals. J Cell Sci 2005;118:2791e801. Zhang M, Liu P, Dwyer NK, Christenson LK, Fujimoto T, Martinez F, et al. MLN64 mediates mobilization of lysosomal cholesterol to steroidogenic mitochondria. J Biol Chem 2002;277:33300e10. Stocco DM, Clark BJ. Regulation of the acute production of steroids in steroidogenic cells. Endocrinol Rev 1996;17:221e44. Jefcoate CR, McNamara BC, DiBartolomeis MJ. Control of steroid synthesis in adrenal fasciculata cells. Endocrinol Res 1986;12:315e50. Cherradi N, Rossier MF, Vallotton MB, Timberg R, Friedberg I, Orly J, et al. Submitochondrial distribution of three key steroidogenic proteins (steroidogenic acute regulatory protein and cytochrome p450scc and 3beta-hydroxysteroid dehydrogenase isomerase enzymes) upon stimulation by intracellular calcium in adrenal glomerulosa cells. J Biol Chem 1997;272:7899e907. Martinez F, Kiriakidou M, Strauss 3rd JF. Structural and functional changes in mitochondria associated with trophoblast differentiation: methods to isolate enriched preparations of syncytiotrophoblast mitochondria. Endocrinology 1997;138:2172e83. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248e54. Martinez F, Espinosa-Garcia MT, Flores-Herrera O, Pardo JP. Respiratory control induced by ATP in human term placental mitochondria. Placenta 1993;14:321e31. Scott DM, Storey BT, Lee CP. L-3-Glycerol phosphate oxidation with energy coupling in submitochondrial particles from skeletal muscle mitochondria. Biochem Biophys Res Commun 1978;83:641e8. Garcia-Perez C, Pardo JP, Martinez F. Ca2þ modulates respiratory and steroidogenic activities of human term placental mitochondria. Arch Biochem Biophys 2002;405:104e11. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680e5. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A 1979;76:4350e4. Liu J, Rone MB, Papadopoulos V. Protein-protein interactions mediate mitochondrial cholesterol transport and steroid biosynthesis. J Biol Chem 2006;281:38879e93. Rone MB, Fan J, Papadopoulos V. Cholesterol transport in steroid biosynthesis: role of proteineprotein nteractions and implications in disease states. Biochem Biophys Acta 2009;1791:646e58. Alpy F, Tomasetto C. MLN64 and MENTHO, two mediators of endosomal cholesterol transport. Biochem Soc Trans 2006;34:343e5. Papadopoulos V, Baraldi M, Guilarte TR, Knudsen TB, Lacapere JJ, Lindemann P, et al. Translocator protein (18kDa): new nomenclature for the peripheral-type benzodiazepine receptor based on its structure and molecular function. Trends Pharmacol Sci 2006;27:402e9. Maldonado-Mercado MG, Espinosa-Garcia MT, Gomez-Concha C, MonrealFlores J, Martinez F. Steroidogenesis in BeWo cells: role of protein kinase A and benzodiazepines. Int J Biochem Cell Biol 2008;40:901e8. Bose HS, Whittal RM, Debnath D, Bose M. Steroidogenic acute regulatory protein has a more open conformation than the independently folded smaller subdomains. Biochemistry 2009;48:11630e9. Lin L, Kim SC, Wang Y, Gupta S, Davis B, Simon SI, et al. HSP60 in heart failure: abnormal distribution and role in cardiac myocyte apoptosis. Am J Physiol Heart Circ Physiol 2007;293:H2238e47. Imatoh T, Sugie T, Miyasaki M, Tanihara S, Baba M, Momose Y, et al. Is heat shock protein 60 associated with type 2 diabetes mellitus? Diabetes Res Clin Pract 2009;85(2):208e12. Di Domenico F, Sultana R, Tiu GF, Scheff NN, Perluigi M, Cini C. Protein levels of heat shock proteins 27, 32, 60, 70, 90 and thioredoxin-1 in amnestic mild cognitive impairment: an investigation on the role of cellular stress response in the progression of Alzheimer disease. 2010;1333:72e81. Foteinos G, Xu Q. Immune-mediated mechanisms of endothelial damage in atherosclerosis. Autoimmunity 2009;42:627e33. ska M, Szprynger K, Zwolin ska D. The impact of dialysis Musia1 K, Szczepan modality on serum heat shock proteins in children and young adults with chronic kidney disease. Kidney Blood Press Res 2009;32:366e72. Xu Z, Horwich AL, Sigler PB. The crystal structure of the asymmetric GroELGroES-(ADP)7 chaperonin complex. Nature 1997;388:741e50. Nagumo Y, Kakeya H, Shoji M, Hayashi Y, Dohmae N, Osada H. Epolactaen bunds human Hsp60 Cys442 resulting in the inhibition of chaperone activity. Biochem J 2005;387(Pt 3):835e40. Nagumo Y, Kakeya H, Yamaguchi J, Uno T, Shoji M, Hayashi Y, et al. Structureactivity relationships of epolctaene derivates: structural requirements for inhibition of Hsp60 chaperone activity. Bioorg Med Chem Lett 2004;14 (17):4425e9.
S. Olvera-Sanchez et al. / Placenta 32 (2011) 222e229 [40] Smyth D, Blumenfeld O, Koningsberg W. Reactions of N-Ethylmaleimide with peptides and amino. J Acids Biochem 1964;91:589e95. [41] Navarrete J, Flores-Herrera O, Uribe A, Martinez F. Differences in cholesterol incorporation into mitochondria from hepatoma AS-30D and human term placenta. Placenta 1999;20:285e91. [42] Flores-Herrera O, Uribe A, Garcia-Perez C, Milan R, Martinez F. 50 -p-fluorosulfonylbenzoyl adenosine inhibits progesterone synthesis in human placental mitochondria. Biochim Biophys Acta 2002;1585:11e8.
229
[43] Kowluru R, Yamazaki T, McNamara BC, Jefcoate CR. Metabolism of exogenous cholesterol by rat adrenal mitochondria is stimulated equally by physiological levels of free Ca2þ and by GTP. Mol Cell Endocrinol 1995;107 (2):181e8. [44] Xu T, Bowman EP, Glass DV, Lambeth JD. Stimulation of adrenal mitochondrial cholesterol side-chain cleavage by GTP, steroidogenesis activator polypeptide (SAP), and sterol carrier protein2. GTP and SAP act synergistically. J Biol Chem 1991;266(11):6801e7.
