Drp1 Phosphorylation Is Indispensable for Steroidogenesis in Leydig Cells

Abstract

The initial steps of steroidogenesis occur in the mitochondria. Dynamic changes in the mitochondria are associated with their fission and fusion. Therefore, understanding the cellular and molecular relationships between steroidogenesis and mitochondrial dynamics is important. The hypothesis of the current study is that mitochondrial fission and fusion are closely associated with steroid hormone synthesis in testicular Leydig cells. Steroid hormone production, induced by dibutyryl cAMP (dbcAMP) in Leydig cells, was accompanied by increased mitochondrial mass. Mitochondrial elongation increased during the dbcAMP-induced steroid production, whereas mitochondrial fragmentation was reduced. Among the mitochondrial-shaping proteins, the level of dynamin-associated protein 1 (Drp1) was altered in response to dbcAMP stimulation. The increase in Drp1 Ser 637 phosphorylation correlated with steroid hormone production in the MA-10 Leydig cells as well as in the primary adult rat Leydig cells. Drp1 was differentially expressed in the Leydig cells during testicular development. Finally, gonadotropin administration altered the status of Drp1 phosphorylation in the Leydig cells of immature rat testes. Overall, mitochondrial dynamics is directly linked to steroidogenesis, and Drp1 plays an important regulatory role during steroidogenesis. This study shows that Drp1 level is regulated by cAMP and that its phosphorylation via protein kinase A (PKA) activation plays a decisive role in mitochondrial shaping by offering an optimal environment for steroid hormone biosynthesis in Leydig cells. Therefore, it is suggested that PKA-mediated Drp1 Ser 637 phosphorylation is indispensable for steroidogenesis in the Leydig cells, and this phosphorylation results in mitochondrial elongation via the relative attenuation of mitochondrial fission during steroidogenesis.

Steroid hormones are necessary for the maintenance of homeostasis and the regulation of diverse processes in the body. In mammals, steroid hormone biosynthesis occurs in the adrenal gland, ovary, testis, placenta, and brain. These organs contain several important steroidogenic proteins, including steroidogenic acute regulatory protein (StAR) (1), the cholesterol side chain cleavage enzyme [P450scc; cytochrome P450 11A1 (CYP11A1)] (2), translocator protein (TSPO) (3), and 3β-hydroxysteroid dehydrogenase (3β-HSD) (4). Whereas 3β-HSD is found in the mitochondria and endoplasmic reticulum (ER) as well as in the cytoplasm (5, 6), StAR, CYP11A1, and TSPO are predominantly present in the mitochondria. Therefore, the mitochondria and ER are the major cellular organelles responsible for steroid hormone production. Mitochondria, in particular, are central sites for the initial steps of steroidogenesis (7, 8).

Cholesterol is a critical material for steroid hormone production, and its delivery and entry to the mitochondria is a pivotal step for preparation of the steroidogenic process. Cholesterol access to the mitochondria is controlled by StAR (1). StAR, acting as the acute regulator of steroidogenesis, works with a multiprotein complex called the “transduceosome,” which resides on the outer mitochondrial membrane. The transduceosome complex includes the TSPO, the voltage-dependent anion channel 1 (VDAC1), acyl-coenzyme A binding domain containing 3 (ACBD3), and protein kinase cAMP-dependent type I regulatory subunit α (PRKAR1A) (7, 9). Once cholesterol reaches the inner mitochondrial membrane (IMM), it is subsequently converted to pregnenolone by CYP11A1 to initiate steroidogenesis. Mitochondrial CYP11A1 cleaves the 20,22 bond of insoluble cholesterol to produce soluble pregnenolone. This process is known to be the hormonally regulated, rate-limiting step in steroidogenesis. CYP11A1 acts with its two electron-transfer partners, ferredoxin reductase and ferredoxin, in the IMM (10). By applying this enzyme system, the conversion of cholesterol to pregnenolone is ultimately achieved, and thus net steroidogenic capacity is determined. Therefore, the mitochondrial CYP11A1 enzyme system is regarded as the chronic regulator of steroidogenesis (8).

Mitochondria are dynamic cellular organelles that contribute to energy conversion, metabolism, signaling, aging, and apoptosis. In the cytoplasm, mitochondria often organize a network, a reticulum of interconnected organelles shaped by the fusion and fission processes (11). They also closely interact with the ER to generate an essential interface in cell physiology and death (12). The morphology (shape) of mitochondria is determined by the balance between the opposing processes of fusion and fission (13–15). Unbalanced fissions result in mitochondrial fragmentation, and more extensive fusions lead to mitochondrial elongation (16). Additionally, mitochondrial fusion and fission are also indispensable for the bioenergetic function of mitochondria (17). Accordingly, specific research on mitochondrial fusion and fission has been collectively termed mitochondrial dynamics (18, 19). The mammalian mitochondrial fission machinery consists of the outer membrane protein mitochondrial fission 1 protein (Fis1) and dynamin-related protein 1 (Drp1) (20, 21). Drp1, a cytoplasmic dynamin GTPase, participates in the fragmentation of mitochondria and the ER (20). Fis1 leads to mitochondrial fragmentation that is Drp1-dependent (21). The phosphorylation of Drp1 at Ser 637 through protein kinase A (PKA) activation causes the inhibition of fission via suppression of Drp1 translocation to the mitochondria (22, 23). In contrast, Drp1 Ser 616 phosphorylation by cyclin-dependent kinase 1 (Cdk1) promotes mitochondrial fission via increased Drp1 translocation from the cytosol to the mitochondria (24, 25). The mitofusins Mfn1 and Mfn2 are key components of the mammalian mitochondrial fusion machinery (17, 26). Both Mfn1 and Mfn2 are localized to the outer membrane and mediate tethering of mitochondria (27). Similar to mitofusins, optic atrophy type 1 (OPA1) is required for mitochondrial fusion, but it is anchored to the IMM (28).

Leydig cells, located in the interstitial compartments of mammalian testes, produce testosterone (T) in response to pituitary LH (29). Binding of LH to its receptors (LHRs) on the plasma membrane triggers LHR coupling to G proteins, which causes the activation of adenylate cyclase and subsequently increases intracellular cAMP formation (30). Then, cAMP stimulates PKA activation, which phosphorylates serine and threonine residues on specific protein substrates that are required for steroid hormone production (31). As previously mentioned, cAMP-activated StAR on the outer mitochondrial membrane participates in cholesterol delivery into the mitochondria, and CYP11A1 converts cholesterol to pregnenolone in IMM. Pregnenolone then moves to the smooth ER, where it is converted to progesterone (P4) by 3β-HSD. P4 is then metabolized by cytochrome P450 aromatase to androstenedione, which is converted to T by 17β-hydroxysteroid dehydrogenase (29).

Based on the context of this potential relationship between steroidogenesis and the mitochondria, it was hypothesized in this study that mitochondrial dynamics (fission and fusion) is closely associated with steroid hormone synthesis in Leydig cells. To elucidate this, mitochondrial dynamics-involved proteins, Drp1, Fis1, Mfn1, and Mfn2, and mitochondrial morphology were first monitored in dibutyryl cAMP (dbcAMP)–treated Leydig cells. The responding protein was then chosen and its potential engagement with steroidogenesis in Leydig cells was determined by biochemical and hormonal analyses.

Materials and Methods

Reagents and antibodies

dbcAMP, human chorionic gonadotropin (hCG), H89, RO3306, dimethyl sulfoxide, poly-l-lysine solution, Bouin solution, hematoxylin, DMEM/Ham nutrient mixture F-12, medium 199, Percoll, gentamicin, and puromycin were purchased from Sigma-Aldrich (St. Louis, MO). Complete protease inhibitor cocktail tablets were from Roche Applied Science (Mannheim, Germany). Horse serum was from Gibco (Thermo Fisher Scientific, Waltham, MA). 10-N-nonyl-acridine orange (NAO) and MitoTracker Green FM (MTG) fluorescent dyes were from Molecular Probes (Thermo Fisher Scientific). Lipofectamine^® RNAiMAX reagent and Opti-MEM I were obtained from Invitrogen (Thermo Fisher Scientific). P4 and T ELISA kits were purchased from IBL (Hamburg, Germany). Western enhanced chemiluminescence detection reagent was purchased from Bio-Rad Laboratories (Hercules, CA). Anti-StAR (1:5000 dilution; RRID: AB_2115937) (32), Mfn1 (1:1000 dilution; RRID: AB_2250540) (33), Mfn2 (1:1000 dilution; RRID: AB_2142754) (34), 60-kDa heat shock protein (Hsp60; 1:3000 dilution; RRID: AB_627758) (35), and β-actin (1:3000 dilution; RRID: AB_2714189) (36) antibodies, as well as mouse Drp1 small interfering RNA (siRNA), control siRNA-A, and Polybrene, were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Anti–phosphorylated Drp1 (pDrp1; Ser 616) (1:2000 [western blot (WB)], 1:100 [immunocytochemistry (IC)] dilution; RRID: AB_2085352) (37), and pDrp1 Ser 637 [1:2000 (WB), 1:100 (IC) dilution; RRID: AB_10971640] (38) antibodies, as well as rabbit and mouse IgG conjugated with horseradish peroxidase, were purchased from Cell Signaling Technology (Danvers, MA). Anti-Fis1 (1:2000 dilution; RRID: AB_2102535) (39) antibody was from Alexis Biochemicals (Lausen, Switzerland). Anti-CYP11A1 (1:10,000 dilution; RRID: AB_90574) (40) antibody was from Chemicon (Millipore, Billerica, MA). Anti-Drp1 [1:2000 (WB), 1:100 (IC), 1:200 (immunohistochemistry) dilution; RRID: AB_398424] (41) antibody was obtained from BD Biosciences (Lexington, KY). Anti-Cox4 [1:3000 (WB), 1:200 (IC) dilution; RRID: AB_443304] (42) antibody was from Abcam (Cambridge, UK). Vectastain^® Elite ABC kit, Vectashield^® mounting medium, fluorescein anti-rabbit IgG (1:100 dilution; RRID: AB_2336197) (43), fluorescein anti-mouse IgG (1:100 dilution; RRID: AB_2336176) (44), and Texas Red anti-rabbit IgG (1:100 dilution; RRID: AB_2336199) (45) antibodies were from Vector Laboratories (Burlingame, CA).

Animals and sampling

For the hCG injection experiment, immature male Sprague-Dawley rats (21 days of age) were purchased from SamTako Bio-Korea (Osan, Korea). The rats were housed in a climate-controlled (21 ± 2°C) animal room at a constant 12-hour light/12-hour dark cycle with free access to food and water ad libitum. All procedures were performed in accordance with protocols approved by the Dong-A University Animal Care and Use Committee. The animals were treated humanely and with regard for alleviation of suffering. The rats were given an IP injection of diluent (PBS) or hCG (30 IU). Forty-eight hours after injection, the rats were euthanized by carbon dioxide asphyxiation and whole blood was collected by cardiac puncture. The testes were dissected, the right testes were used to collect interstitial fluid and to isolate Leydig cells, and the left testes were fixed in Bouin solution for histological studies. Testicular interstitial fluid (TIF) was collected according to a previously described method (46), and Leydig cells were isolated as described previously (47). Blood was allowed to clot at room temperature (RT) and centrifuged at 3000g for 10 minutes at 4°C. All collecting samples were stored at −80°C before the analyses. The testes of adult male Sprague-Dawley rats (12 weeks of age) were used for isolation of primary Leydig cells and employed for their culture experiments. The testes from different ages of male Sprague-Dawley rats (4, 7, 9, and 12 weeks of age) were used for immunohistochemical study.

MA-10 cell culture

The MA-10 cell line from mouse Leydig cell tumor (provided by Dr. M. Ascoli from the University of Iowa, Ames, IA) shows steroidogenic responses to PKA activation by dbcAMP, LH, and hCG stimulation (48). MA-10 cells were cultured with DMEM/Ham nutrient mixture F-12 modified to contain 1.2 g/L NaHCO₃, 20 mM HEPES, 50 μg/mL gentamicin, and 15% heat-inactivated horse serum (pH 7.4) at 37°C in a humidified atmosphere with 5% CO₂. Cells were maintained on 0.1% gelatin-coated culture dishes. The cells (5 × 10⁵ cells) were seeded in gelatin-coated six-well plates, and dbcAMP was treated at confluence with ∼50% to ∼60%.

Leydig cell isolation and culture

Primary Leydig cells were isolated from the testes as described previously (47), with slight modifications. In brief, testes were perfused with type III collagenase (1 mg/mL) in dissociation buffer [M-199 medium (pH 7.4) with 2.2 g/L HEPES, 1.0 g/L BSA, 2.2 g/L sodium bicarbonate, and 25 mg/L trypsin inhibitor]. Then the testes were decapsulated and incubated with dissociation buffer containing collagenase (0.25 mg/mL) at 34°C in a shaking water bath for 30 minutes. Dissociated testes were filtered through a 100-μm pore size nylon mesh to remove the seminiferous tubule mass. The filtrate was centrifuged at 1500 rpm for 5 minutes at RT and the pellet was resuspended in Percoll solution, which is mixed with Percoll buffer and iso-osmotic Percoll in Hanks balanced salt solution. After centrifugation (20,000g, 60 minutes, 4°C), the supernatant was discarded and the cell pellets were washed with dissociation buffer. The isolated Leydig cells were resuspended in M-199 medium containing 15 mM HEPES, 0.1% BSA, 5 μg/mL gentamicin, 50 U/mL penicillin, and 50 μg/mL streptomycin, which yielded 90% pure Leydig cells. For dbcAMP treatment experiments, the purified cells (1.0 × 10⁶) were plated onto a six-well plate and cultured without serum at 34°C in a humidified incubator with an atmosphere of 5% CO₂.

Measurement of mitochondrial mass

To determine mitochondrial mass, the cells were stained with 200 nM NAO, which binds specifically to cardiolipin at the IMM independently of membrane potential. After incubation at RT in the dark for 20 minutes, cells were washed and then analyzed by flow cytometry (Cytomics FC500; Beckman Coulter, Brea, CA). Mitochondrial mass was also measured by MTG staining. Cells were stained with 150 nM MTG at 37°C in the dark for 20 minutes and then washed in PBS followed by flow cytometric analysis. The fluorescence was excited with an argon laser (excitation wavelength, 488 nm) and analyzed in FL-1 (emission wavelength, 525 nm). A minimum of 10,000 events were collected in list mode and analyzed with CXP analysis software.

Evaluation of mitochondrial morphology

For evaluation of mitochondrial morphology, the mitochondrial-targeted form of Keima (mt-Keima) (49) was introduced into cultured MA-10 Leydig cells in this study. Keima is a fluorescent protein that emits different colored signals at acidic and neutral pHs. The mt-Keima lentiviral particles were provided by Prof. J. Yun (Department of Biochemistry, College of Medicine, Dong-A University, Busan, Korea) (50). Viral transduction was performed by incubating the MA-10 cells with the mt-Keima lentiviral particles supplemented with Polybrene (8 µg/mL) in a humidified incubator at 37°C for 18 to 20 hours. After puromycin (5 µg/mL) selection, the resistant cells were pooled and used for observation of mitochondrial morphology. The MA-10 cells infected with mt-Keima were cultured on poly-l-lysine–coated coverslips. After the cells were treated with 1 mM dbcAMP for 3 hours, immediately mitochondrial morphology was examined using a confocal laser scanning microscope (LSM 700; Carl Zeiss, Jena, Germany). The mt-Keima fluorescence signal was observed under 488 nm laser excitation (fluorescence signals of mt-Keima in mitochondria) and the status of mitochondria was quantified. Cells were classified according to mitochondrial length into three groups. Cells displaying either predominantly (>95%) short/punctiform (fragmented) or (>95%) elongated mitochondria were classified as cells with “fragmented” or “elongated” mitochondria, respectively. Cells containing both fragmented and elongated mitochondria were classified as “intermediate.”

Knockdown of Drp1 by siRNA transfection

MA-10 cells were transfected with Drp1 siRNA by using Lipofectamine RNAiMAX according to the manufacturer’s protocol. In brief, cells were grown to ∼50% to ∼60% confluency in six-well plates and transfected with 30 nM of either Drp1-specific siRNA or scrambled (control) siRNA in Opti-MEM reduced serum medium for 18 hours. Then the cells were incubated in complete medium for another 24 hours. After transfection, cells were treated with 1 mM dbcAMP. After dbcAMP treatment for 3 hours, the medium was collected to determine P4 concentration and cells were assayed by WBs.

Preparation of mitochondrial and cytosolic extracts

Crude mitochondria and cytosol were prepared as described by Liu et al. (51), with minor modifications. In brief, cells were harvested and immediately incubated in a digitonin buffer [5 mM NaAc, 2 mM MgAc₂, 1 mM EGTA, 110 mM KAc, 20 mM HEPES-KOH (pH 7.3), 0.2 mg/mL digitonin, phosphatase inhibitors, and protease inhibitors] on ice for 10 minutes. The permeabilized cells were centrifuged at 13,000 rpm for 5 minutes at 4°C, and the supernatants were collected as cytosolic fractions. The pellet including mitochondria and nuclei was resuspended in protein lysis buffer. Cytosolic and crude mitochondrial proteins separated by the digitonin permeabilization assay were used for WB analysis.

WB analysis

The cultured cells were washed with cold PBS and resuspended in lysis buffer [300 mM NaCl, 0.5% Triton X-100, 50 mM Tris-HCl (pH 7.4), 25 mM NaF, 1 mM Na₃VO₄, 10 mM Na₄P₂O₇, and protease inhibitor] for 40 minutes on ice. The lysates were centrifuged at 14,000 rpm at 4°C for 20 minutes, the supernatants were collected, and protein concentration was measured using a BCA protein assay kit. Twenty-five micrograms of protein extract with SDS loading buffer was electrophoretically separated on an 8% to 15% gradient SDS-PAGE gel and transferred onto a nitrocellulose membrane. The membranes were blocked with 5% nonfat dry milk dissolved in Tris-buffered saline buffer containing 0.05% Tween 20 at RT for 1 hour. The blots were incubated with primary antibodies, followed by incubation with appropriate horseradish peroxidate–conjugated secondary antibodies. The signals were detected with enhanced chemiluminescence detection reagent in an LAS-4000 (Fuji, Tokyo, Japan). Actin was used as internal control for total cellular proteins. Cox4 and Hsp60 were used as the internal control for mitochondrial proteins.

Immunocytochemistry

MA-10 cells were grown to ∼50% to ∼60% confluence on poly-l-lysine–coated glass coverslips. Cells were treated with 1 mM dbcAMP for 3 hours. The cells were fixed with 4% paraformaldehyde for 30 minutes and washed with PBS (pH 7.4) and permeabilized with 0.02% Triton X-100 in PBS for 10 minutes. After washing, cells were blocked with blocking buffer (l% BSA) in a humidified chamber for 1 hour. The appropriate primary antibodies (anti-Drp1, anti–pDrp1 Ser 616, anti–pDrp1 Ser 637) were applied overnight at 4°C. The next day, the wells were washed with PBS and incubated with the appropriate fluorescence-conjugated secondary antibodies at RT for 1 hours. When required, for mitochondrial staining, the above reactions were repeated with the Cox4 antibody. Cells were observed under a confocal laser scanning microscope.

Immunohistochemistry

For immunohistochemical staining of testes, the tissues were fixed in Bouin solution, embedded in paraffin, and sectioned at 5 μm thickness. Tissue sections were placed on glass slides, deparaffinized, hydrated, and treated with 1% H₂O₂ for 10 minutes to suppress endogenous peroxidase activity. Antigen retrieval was performed by heating the sections in 1 mM citric acid solution (pH 6.0). Sections were incubated with the primary antibody for Drp1 at 4°C overnight and reacted with biotinylated horse anti-mouse IgG for 1 hour at RT, as per the manufacturer’s instructions for the Vectastain Elite ABC kit. The stained sections were developed with liquid diaminobenzidine and counterstained with hematoxylin. The results were observed using a ScanScope digital slide scanning system (Aperio Technologies, Vista, CA).

Transmission electron microscopy

Primary Leydig cells were fixed with cold 2.5% glutaraldehyde at 4°C overnight. After washing three times with 0.1 M phosphate buffer, cells were postfixed in 1% osmium tetroxide. Cells were dehydrated in a graded series of ethanol and embedded in the mixture of embedding media (Epon 812, MNA, DDSA, and DMP-30). Semithin sections (1 μm) were obtained using glass knives and stained with toluidine blue. Ultrathin sections (0.06 to 0.09 μm) were obtained using diamond knives and double stained with uranyl acetate/lead citrate. Finally, the cells were observed using an H7650 transmission electron microscope (Hitachi, Tokyo, Japan).

Steroid hormone assays

P4 or T concentration was determined in serum, TIF, or culture medium using a corresponding ELISA kit, following the manufacturer’s instructions. The sensitivity of the P4 assay was 0.045 ng/mL, and the intraassay and interassay coefficients of variation were 6.4% and 6.6%, respectively. The sensitivity of the T assay was 0.083 ng/mL, and the intraassay and interassay coefficients of variation were 3.6% and 7.1%, respectively.

Statistical analysis

Data were expressed as the mean ± SD of at least three independent experiments. The difference in means between two groups was analyzed using the Student t test. For multiple group comparisons, one-way ANOVA followed by a Tukey test was performed by GraphPad Prism software (version 4.0; GraphPad Software, San Diego, CA). Mean values were considered significantly different at P < 0.05.

Results

Steroid production of Leydig cells is accompanied by increased mitochondrial mass

To investigate whether steroid production is associated with biophysical alterations of the mitochondria in Leydig cells, MA-10 Leydig cells were exposed to dbcAMP (1 mM) for 3 hours. The concentration of P4 in the medium was then determined, and mitochondrial mass was measured in the cells. Treatment with dbcAMP sufficiently induced P4 production (Fig. 1A). Simultaneously, mitochondrial mass significantly increased with exposure to dbcAMP (Fig. 1B). To ensure the accuracy of these results, two specific fluorescent dyes used for measurement of mitochondrial mass (MTG and NAO) were used in this experiment. Both dyes exhibited similar results (Fig. 1B).

Figure 1.

Changes in P4 production and mitochondrial mass in the MA-10 Leydig cells after exposure to dbcAMP. (A) P4 production. MA-10 Leydig cells were cultured for 3 h in the absence or presence of dbcAMP (1 mM). P4 levels in media were measured by ELISA. The data are expressed as mean ± SD of three independent experiments performed in triplicate. ***P < 0.001 compared with control (CON). (B) Mitochondrial mass. The MA-10 Leydig cells were treated with dbcAMP (1 mM) for 3 h, stained with either MTG or NAO, and measured by flow cytometry. Measurements of mitochondrial mass using MTG are shown. At least three independent experiments were performed and data shown are the mean ± SD. *P < 0.05, **P < 0.01 compared with CON.

Mitochondrial lengths are altered in Leydig cells during dbcAMP-induced steroid production

Under certain cellular and physiological conditions, the lengths of the mitochondria of cells are dynamically shortened or lengthened (15). In this context, the alterations of mitochondrial lengths were monitored in dbcAMP-stimulated Leydig cells. For accurate observation of mitochondria, MA-10 cells were virally infected with mt-Keima (49). As shown in Fig. 2A, whereas fragmented mitochondria were more readily observed in control cells, elongated mitochondria were more prevalent in dbcAMP-treated cells. The percentage of fragmented, intermediate, and elongated cells is plotted in Fig. 2B. In dbcAMP-treated cells, the percentage of cells undergoing mitochondrial fragmentation was significantly lower than in the control (Fig. 2B). In contrast, the proportions of elongated mitochondria in dbcAMP-treated cells were significantly higher than that of the control (Fig. 2B). Noticeably, the number of intermediate cells was also higher in dbcAMP-treated cells than in the control (Fig. 2B).

Figure 2.

Effect of dbcAMP on alterations in mitochondrial shaping in MA-10 Leydig cells. (A) Alterations in mitochondrial morphology in MA-10 Leydig cells after dbcAMP treatment. The cells were infected by mt-Keima, treated with dbcAMP (1 mM) for 3 h, and observed by confocal microscopy. Cells with intermediate morphology were images both in the (a) control (CON) and (b) dbcAMP-treated cells. Original magnification, ×800; scale bars, 5 μm. Rectangles shown in (a) and (b) were digitally enlarged to (c) (fragmented mitochondria) and (d) (elongated mitochondria), respectively. (B) Quantifications of the status of mitochondrial shapes in the CON and dbcAMP-treated cells. Each cell that contained predominantly (>95%) fragmented mitochondria, both fragmented and elongated mitochondria, and mostly (>95%) elongated mitochondria was counted as fragmented, intermediate, and elongated, respectively. Two hundred cells in each group were counted for evaluation. At least three independent experiments were performed, and data shown are the mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 compared with CON.

Mitochondrial Drp1 proteins are reduced in Leydig cells in response to dbcAMP stimulation

The fragmentation and elongation of mitochondria are associated with mitochondrial fission and fusion, respectively (13, 14). To demonstrate this relationship, proteins involved in mitochondrial fission (Drp1 and Fis1), mitochondrial fusion (Mfn1 and Mfn2), and steroidogenesis (StAR and CYP11A1) were comparably measured in Leydig cells by WB analysis. dbcAMP stimulation significantly increased StAR and CYP11A1 protein levels (Fig. 3A). However, Drp1, Fis1, Mfn1, and Mfn2 protein levels were not significantly affected by dbcAMP stimulation (Fig. 3A). Among these proteins, only Drp1 proteins appeared to increase compared with the control, but this change was not significant (Fig. 3A). Therefore, considering that most of these proteins are predominantly present in the mitochondria, except the cytosolic Drp1 (20), these proteins were further analyzed with proteins extracted from the cytosol-rich and mitochondria-rich fractions. This fractionated study demonstrated that whereas cytosolic Drp1 protein levels increased in dbcAMP-treated cells (Fig. 3B, low exposure blot), the levels of mitochondrial Drp1 were lower in the dbcAMP-treated cells compared with the control (Fig. 3B, high exposure blot). Although Fis1, Mfn1, and Mfn2 proteins were predominantly detected in the mitochondrial fraction, their protein levels did not considerably differ between the control and dbcAMP-treated groups (Fig. 3B). As expected, the StAR and CYP11A1 proteins were primarily present in the mitochondrial fractions, and their protein levels were increased by exposure to dbcAMP (Fig. 3B). Immunofluorescence colocalization of Drp1 and Cox4 demonstrated that Drp1 colocalization with Cox4 was more prominent in control cells (Fig. 3C) compared with dbcAMP-treated cells (Fig. 3D).

Figure 3.

Alterations in the levels of the mitochondrial shaping-related proteins and immunocytochemical localization of Drp1 in MA-10 Leydig cells after the dbcAMP treatment. (A) Cells were treated with dbcAMP (1 mM) for 3 h. Twenty-five micrograms of protein lysates was used in SDS-PAGE followed by western blotting. The levels of StAR and CYP11A1 were monitored for the evaluation of steroidogenic status. Cox4 and Hsp60 were used as internal controls for mitochondrial proteins. Actin was used as an internal control for total cellular proteins. (B) Cells were treated with dbcAMP (1 mM) for 3 h and proteins were extracted from the mitochondrial (Mito) and cytosolic (Cyto) fractions. The same amounts of proteins were applied to SDS-PAGE followed by western blotting. Cox4 and Hsp60 were used as internal controls for mitochondrial proteins. (C and D) Cells were treated with dbcAMP (1 mM) for 3 h. Drp1 and Cox4 were immunocytochemically colocalized and observed by confocal microscopy. Cox4 was stained for mitochondrial localization. In (C) (control) and (D) (dbcAMP-treated), (a)–(d) indicate Drp1 immunoreactivity, Cox4 immunoreactivity, merged Drp1 and Cox4 immunoreactivities, and their merged profiles, respectively. In (Cd) and (Dd), the green and red lines reflect the analyzed Drp1 and Cox4, respectively, indicated by the arrow lines in (Cc) and (Dc). Original magnification, ×800; scale bars, 10 mm.

Specific phosphorylations of Drp1 are crucial for the steroidogenic destiny of Leydig cells

It has recently been suggested that Drp1 phosphorylation events are key steps for the regulation of mitochondrial fission (20). Phosphorylation of Drp1 at Ser 637 or 616 each plays a role of switch for inhibition or trigger of mitochondrial fission, respectively (22–25). In this context, the status of Drp1 phosphorylation was examined in MA-10 Leydig cells after dbcAMP stimulation. Phosphorylation of Drp1 at Ser 637 and Ser 616 were detected by WB analysis using phosphorylation site–specific antibodies. Noticeably, dbcAMP stimulation resulted in an increase of Drp1 Ser 637 phosphorylation and decrease of Drp1 Ser 616 phosphorylation in the cytosol whereas the protein content of StAR was increased in the mitochondria (Fig. 4A). Consistent with the results seen in the WB analysis, the immunofluorescence of Drp1 Ser 637 was more intense in dbcAMP-treated cells (Fig. 4Bb) than in the control (Fig. 4Ba). In contrast, a reduction in immunofluorescence for Drp1 Ser 616 was clearly observed in dbcAMP-treated cells (Fig. 4Bd) compared with the control (Fig. 4Bc).

Figure 4.

Drp1 phosphorylation in MA-10 Leydig cells after exposure to dbcAMP. (A) Reciprocal relationship between Drp1 Ser 637 phosphorylation and Drp1 Ser 616 phosphorylation in the cytosol of MA-10 Leydig cells after dbcAMP treatment. Cells were treated with dbcAMP (1 mM) for 3 h and proteins were extracted from the mitochondrial (Mito) and cytosolic (Cyto) fractions. The same amounts of proteins were applied to SDS-PAGE followed by western blotting. Cox4 and Hsp60 were used as internal controls for Mito proteins. Actin was used as an internal control for total cellular proteins. (B) Cytochemical immunoreactivities of Drp1 Ser 637 and Drp1 Ser 616 phosphorylated proteins in MA-10 Leydig cells. The cells were treated with dbcAMP (1 mM) for 3 h and changes in the cellular localization of pDrp1 was monitored by immunocytochemistry using phosphorylation-specific antibodies for Drp1 Ser 637 and Drp1 Ser 616. The upper and lower panels indicate Drp1 Ser 637 and Drp1 Ser 616 immunoreactivities, respectively. Original magnification (confocal microscopy), ×800; scale bars, 10 μm.

Specific inhibition of Drp1 phosphorylation and activity directly affects steroid hormone production in Leydig cells

To examine whether inhibition of Drp1 phosphorylation and its activity influence steroid hormone production, their specific inhibitors were used in this study. PKA and Cdk1 (cdc2) are known to be responsible for the phosphorylation of Drp1 Ser 637 (22, 23) and Drp1 Ser 616 (24, 25), respectively. Therefore, the effects of H89 (PKA inhibitor) and RO3306 (Cdk1 inhibitor) on Drp1 phosphorylation and activity were monitored in dbcAMP-treated Leydig cells. Pretreatment of H89 resulted in decreased pDrp1 Ser 637 induced by dbcAMP (1 mM) (Fig. 5A). StAR protein levels were also simultaneously downregulated (Fig. 5A) and P4 production was significantly lowered (Fig. 5B). The levels of pDrp1 Ser 616 by dbcAMP (0.1 mM) were decreased in the presence of RO3306 (Fig. 5C). In contrast to H89, minimal P4 production was induced with a low concentration (0.1 mM) of dbcAMP, which is enhanced in the presence of RO3306 (Fig. 5D). The combined treatment of dbcAMP and RO3306 resulted in an increase in StAR protein levels compared with dbcAMP treatment alone (Fig. 5C).

Figure 5.

Alterations in dbcAMP-induced Drp1 phosphorylation in MA-10 Leydig cells either in the presence of PKA and Cdk1 inhibitors or by Drp1 knockdown. (A and B) Effect of PKA inhibitor on Drp1 Ser 637 phosphorylation and P4 production. Cells were treated with dbcAMP (1 mM) for 3 h in the absence or presence of H89 (20 μM). Proteins were extracted from the mitochondrial (Mito) and cytosolic (Cyto) fractions, and same amounts of proteins were applied to SDS-PAGE followed by western blotting. Cox4 and Hsp60 were used as internal controls for Mito proteins. Actin was used as an internal control for total cellular proteins. P4 levels in media were measured by ELISA. The data are expressed as mean ± SD of three independent experiments performed in triplicate. ***P < 0.001 compared with dbcAMP-treated cells. (C and D) Effect of Cdk1 inhibitor on Drp1 Ser 616 phosphorylation and P4 production. Cells were treated with dbcAMP (0.1 mM) for 3 h in the absence or presence of RO3306 (RO; 5 μM). Proteins were extracted from the Mito and Cyto fractions, and the same amounts of proteins were applied to SDS-PAGE followed by western blotting. P4 levels in media were measured by ELISA. The data are expressed as mean ± SD of three independent experiments performed in triplicate. *P < 0.05 compared with dbcAMP-treated cells. (E and F) Changes in Drp1 phosphorylation after Drp1 knockdown and P4 production. For knockdown of the Drp1 gene in MA-10 cells, siRNA for Drp1 (siD) or control siRNA were transfected for 24 h before dbcAMP treatment. Then, cells were treated with dbcAMP (1 mM) for 3 h. The same amounts of proteins were applied to SDS-PAGE followed by western blotting. P4 levels in media were measured by ELISA. The data are expressed as mean ± SD of three independent experiments performed in triplicate. **P < 0.01 compared with dbcAMP-treated cells.

Knockdown of Drp1 reduces dbcAMP-stimulated steroid hormone production in Leydig cells

To determine whether ablation of Drp1 in MA-10 cells reduces their capability to produce dbcAMP-P4, siRNA-mediated knockdown of Drp1 was applied. As shown in Fig. 5E, the expression of endogenous Drp1 was reduced by siRNA transfection. Additionally, both Drp1 Ser 637 and Ser 616 phosphorylations were also reduced by siRNA knockdown of Drp1 (Fig. 5E). Drp1 knockdown decreased dbcAMP-induced P4 production significantly but not sufficiently (Fig. 5F).

Phosphorylation of Drp1 at Ser 637 is increased during T production in primary Leydig cells

Although the MA-10 cells originated from mouse Leydig cells, it is presumed that the characteristics of immortalized MA-10 cells are not representative of Leydig cells isolated from mammalian testes. Actually, MA-10 cells are known to be able to produce P4 but not T because the amounts of P450c17 (17α-hydroxylase) mRNA and P450c17 activity are insignificant in this cell type (52). Therefore, to confirm whether Drp1 phosphorylation is altered during T production in primary Leydig cells, dbcAMP was added to Leydig cells isolated from adult rat testes. Primary Leydig cells produced significant levels of T in respond to dbcAMP (Fig. 6A). Similar to MA-10 cells, dbcAMP increased Drp1 Ser 637 phosphorylation and decreased Drp1 Ser 616 phosphorylation (Fig. 6B). Furthermore, dbcAMP-treated primary Leydig cells also retained lower Drp1 protein levels in the mitochondria-rich fractions than in the control (Fig. 6B, high exposure blot). After dbcAMP treatment, an increase in Drp1 Ser 637 phosphorylation and decrease in Drp1 Ser 616 phosphorylation were observed in cytosol-rich fractions (Fig. 6B). In addition to these biochemical analyses with primary Leydig cells, ultrastructural morphology of the mitochondria in control and dbcAMP-treated cells was observed by transmission electron microscopy (Fig. 6C). Whereas short (fragmented) and round-shaped mitochondria were frequently observed in control cells (Fig. 6Ca), long (elongated) and thread-like mitochondria were more readily observed in dbcAMP-treated cells (Fig. 6Cb).

Figure 6.

Involvement of Drp1 and its phosphorylation during T production in rat Leydig cells in vitro and in vivo. (A) T production in the adult primary Leydig cells after exposure to dbcAMP. Leydig cells were isolated from adult rat testes and cultured for 3 h in the presence of dbcAMP. T levels in media were measured by ELISA. The data are expressed as mean ± SD of three independent experiments performed in triplicate. ***P < 0.001 compared with control (CON). (B) Drp1 phosphorylation in the mitochondria and cytosol of adult primary Leydig cells after dbcAMP treatment. Leydig cells were isolated from adult rat testes and cultured for 3 h in the presence of dbcAMP. Proteins were extracted from the mitochondrial (Mito) and cytosolic (Cyto) fractions, and the same amounts of proteins were applied to SDS-PAGE followed by western blotting. The levels of StAR and CYP11A1 were monitored for the evaluation of steroidogenic status. Cox4 and Hsp60 were used as internal CONs for Mito proteins. Actin was used as an internal CON for total cellular proteins. (C) Changes in the ultrastructural Mito morphology in the adult primary Leydig cells after dbcAMP treatment. Leydig cells were isolated from adult rat testes and cultured for 3 h in the presence of dbcAMP. (a) CON and (b) dbcAMP-treated cells were fixed and observed with a transmission electron microscope; arrowheads and arrows indicate fragmented and elongated mitochondria, respectively. Original magnification, ×30,000; scale bars, 0.2 mm. (D) Localization of Drp1 in Leydig cells of rat testes of different ages. Rat testes aged (a) 4, (b) 7, (c) 9, and (d) 12 wk were fixed, paraffin embedded, and immunostained with Drp1 antibody and observed under a light microscope. Arrows indicate Drp1-immunoreactive Leydig cells in the interstitial compartment between seminiferous tubules. Original magnification, ×800. (E) T concentrations in immature rat serum and TIF after hCG injection in vivo. Immature rats (3 wk) were injected with 30 IU of hCG for 48 h. T levels in (a) serum and (b) interstitial fluid were measured by ELISA. The data are expressed as mean ± SD of three independent experiments performed in triplicate. *P < 0.05 compared with CON. (F) Effect of hCG on Drp1 level and phosphorylation in Leydig cells of immature rat testes. Immature rats (3 wk) were injected with 30 IU of hCG for 48 h. Proteins were extracted from isolated Leydig cells of the testes, and the same amounts of proteins were applied to SDS-PAGE followed by western blotting. The levels of StAR and CYP11A1 were monitored for the evaluation of steroidogenic status.

Drp1 is differentially expressed in Leydig cells during testicular development

Because the capability for T production of Leydig cells is developmentally acquired with increasing age of rat testes (i.e., progenitor Leydig cells from 3 weeks of age, immature Leydig cells from 5 weeks of age, adult Leydig cells from 12 to 13 weeks of age) (53), Drp1 protein expression was assessed by IH in rat testes of four different ages (4, 7, 9, and 12 weeks). Immunoreactivity for Drp1 was localized to the interstitial compartment of Leydig cells (Fig. 6D). Immature Leydig cells in 4-week-old testis did not show significant Drp1 immunoreactivity (Fig. 6Da). However, Leydig cells in 7-, 9-, and 12-week-old testes exhibited Drp1 immunoreactivity (Fig. 6Db, 6Dc, and 6Dd). In general, Drp1 immunoreactivity was intense in adult Leydig cells (9- and 12-week-old testes) (Fig. 6Dc and 6Dd).

Gonadotropin exposure alters the status of Drp1 phosphorylation in Leydig cells of immature rat testes

Finally, to investigate whether exogenous injection of gonadotroin (hCG) could stimulate the induction of Drp1 expression in Leydig cells of immature rat testes, hCG was administered to 3-week-old immature rats for 48 hours. As expected from the experimental immature rat model for induction of T production (54), the hCG injection was able to provoke increased T concentrations in serum (Fig. 6Ea) as well as in interstitial fluids (Fig. 6Eb). Compared with the control, exposure to hCG in immature rats resulted in increased Drp1 protein levels in Leydig cells isolated from the testes (Fig. 6F). Simultaneous increases in StAR and CYP11A1 protein levels were also observed (Fig. 6F). Most importantly, hCG injections led to an increase in Drp1 Ser 637 phosphorylation and a decrease of Drp1 Ser 616 phosphorylation in Leydig cells of immature testes (Fig. 6F).

Discussion

Mitochondria are semiautonomous organelles that have their own genome and protein synthesis machinery. The most prominent role for mitochondria is to generate ATP by oxidative phosphorylation. During the last few decades, dynamic aspects of mitochondrial behavior have been particularly studied for elucidating the varied functions of mitochondria. Among those, studies on mitochondrial fission and fusion (mitochondrial dynamics) have continuously contributed to the understanding of many biological processes (18, 19). With this progress, the cellular and molecular mechanisms involved in mitochondrial inheritance (55), mitochondrial distribution and morphology (56), mitochondrial quality control and aging (57), and mitochondrial fission-associated apoptosis (58) have been proposed. Furthermore, it has been shown that mitochondrial dynamics are associated with human neurodegenerative diseases such as Charcot–Marie–Tooth disease type 2a (59) and Parkinson disease (60). Therefore, it is expected that continuous studies on mitochondrial dynamics will elucidate its involvement in various human physiological and pathological conditions. In connection with this aspect, as well as the fact of mitochondrial participation during steroid hormone biosynthesis, the current study aimed to investigate the potential association of mitochondrial dynamics with steroidogenesis in testicular Leydig cells. In this study, it was demonstrated that mitochondrial dynamics are directly linked to steroid hormone production, and that the fission-related protein Drp1 plays an important regulatory role during steroidogenesis.

In the current study, stimulation of steroid hormone synthesis induced by dbcAMP in MA-10 Leydig cells was accompanied by physical alterations of mitochondria, such as increased mitochondrial mass, and decreased mitochondrial fragmentation. Although these changes were all significant, the intensities of the rate were indeed below assumptions. However, repeated experiments showed identical results, indicating that the extraordinary and extensive levels of morphological (shaping) alterations in mitochondria might not be required for successful completion of steroidogenesis. A stiff increase of mitochondrial mass is often observed in cells that are undergoing growth and division via a process called mitochondrial biogenesis (61). Mitochondrial biogenesis can also be initiated by responding to an oxidative stimulus, an increase in the energy requirements of the cells, exercise, an electrical stimulus, hormones, development, and certain diseases (62). Although this experiment was performed using dividing MA-10 Leydig cells, considering that Leydig cells in adult testis are rarely dividing, the increase (∼1.5-fold) of dbcAMP-provoked mitochondrial mass might not be due to the effect of cell proliferation within 3 hours. In fact, the doubling time of MA-10 cells is ∼16 to 20 hours. Furthermore, cAMP is known to suppress the G₂/M progression of cells through the inhibition of Cdk1 activity (63). Previously, it was demonstrated that energized, polarized, and actively respiring mitochondria are required for acute Leydig cell steroidogenesis (64). Under the observation of mitochondria-specific mt-Keima–infected MA-10 cells, the current study also showed that dbcAMP significantly decreases mitochondrial fragmentation but increases mitochondrial elongation. In control cells, the proportions of cells that contain fragmented, intermediate, and elongated mitochondria were ∼20%, ∼70%, and ∼10%, respectively. This means that most of this cell type consists of both fragmented (or shorter) and elongated mitochondria, assuming that mitochondria in these cells require normal maintenance of cellular homeostasis and/or quality control. However, after steroid hormone stimulation by dbcAMP, whereas the percentage of cells that only retain fragmented mitochondria was significantly reduced, the proportions of cells containing elongating and/or elongated mitochondria relatively increased. The important point of this observation is that mitochondrial elongations are needed for creating an atmosphere for steroidogenesis, but this is only achieved by reduction of mitochondrial fragmentation after dbcAMP stimulation. In the current study, we show that mitochondrial elongation may contribute to an increase in mitochondrial mass. Therefore, it is suggested that dbcAMP stimulates steroid hormone production via increased mitochondrial mass together with mitochondrial elongation. Taken together, these results suggest that physical alterations of mitochondria are prerequisites for steroid hormone synthesis in Leydig cells, and that mitochondrial dynamics are closely associated with steroidogenesis.

It is generally appreciated that mitochondrial fragmentation and elongation result from mitochondrial fission and fusion, respectively (13, 14). Drp1 and Fis1 participate in mitochondrial fission, and Mfn1 and Mfn2 function in mitochondrial fusion. In this study, these four proteins were monitored in dbcAMP-stimulated cells. The results showed that only Drp1 proteins noticeably responded to the dbcAMP, although Drp1 was increased in the proteins extracted from the cytosol-rich fractions; its contents were decreased in the proteins from the mitochondria-rich fractions. An inverse relationship in Drp1 protein levels between the cytosol and the mitochondria indicates the involvement of Drp1 translocation into the mitochondria. Unlike Fis1, Mfn1 and Mfn2 are primarily expressed in the mitochondria; cytosolic Drp1 may need to move into the mitochondria to perform its function. In this study, the difference of Drp1 protein levels in the mitochondria of control and dbcAMP-treated cells provides a critical clue for the direct involvement of Drp1 in steroidogenesis. Particularly, the decreased mitochondrial Drp1 protein levels (i.e., the reduced cytosolic Drp1 translocation into mitochondria) after dbcAMP indicate that mitochondrial fragmentation (fission) should be conditionally suppressed under this circumstance by modification of cytosolic Drp1 proteins. Indeed, it has been suggested that Drp1 is modified posttranslationally by protein phosphorylation, ubiquitylation, and sumoylation (25, 65, 66). Among those, Drp1 is often phosphorylated at several serine residues for determining cell fate (22–25). Related to the current study, particular interest has been paid to Drp1 Ser 637 and Ser 616 phosphorylation because these are modified by PKA and Cdk1, respectively. Particularly, PKA activation by cAMP is the most important step during steroidogenesis (31). In contrast, cAMP signaling decreases Cdk1 activity (63). Therefore, it could be presumed that cAMP can increase Drp1 Ser 637 phosphorylation and also can decrease Drp1 Ser 616 phosphorylation. Drp1 Ser 616 phosphorylation leads to its translocation into the mitochondria (25). However, this is blocked when Drp1 is phosphorylated at Ser 637 (23). Considering this, it can be postulated that cAMP-provoked PKA activation subsequently causes Drp1 Ser 637 phosphorylation, decreased Drp1 mitochondrial translocation, and reduced mitochondrial fission. In the current study, this hypothesis was tested by WB analysis as well as immunocytochemical observations using phosphorylation-specific antibodies for Drp1 Ser 637 and Ser 616. Accordingly, when StAR proteins were upregulated by dbcAMP, Drp1 Ser 637 phosphorylation increased. In contrast, Drp1 Ser 616 phosphorylation decreased in response to dbcAMP. This was more keenly demonstrated in the WB analysis with the proteins from fractionated mitochondria- and cytosolic-rich fractions. Most of the pDrp1 proteins were sensitively detected in the cytosol, confirming that Drp1 phosphorylation occurs in the cytosol of Leydig cells. Based on these results, it is suggested that cAMP-activated PKA is associated with Drp1 637 phosphorylation and subsequent attenuation of Drp1 mitochondrial translocation, and thus reduction of mitochondrial fission in Leydig cells.

The availability of PKA and Cdk1 enzyme inhibitors (H89 and RO3306) enabled confirmation of the association of Drp1 phosphorylation with steroidogenic activity of Leydig cells. Pretreatment of PKA inhibitor H89 diminished dbcAMP (1 mM)-induced Drp1 Ser 637 phosphorylation and StAR expression as well as steroid hormone (P4) production, indicating that PKA activation is responsible for Drp1 Ser 637 phosphorylation during steroidogenesis. In the presence of the Cdk1 inhibitor RO3306, Drp1 Ser 616 phosphorylation using a minimum concentration of dbcAMP (0.1 mM) was suppressed, and low levels of P4 production induced by 0.1 mM dbcAMP were significantly enhanced. However, the maximum produced P4 concentration with a higher dose of dbcAMP (1 mM) was not additively enhanced by various doses of RO3306 (data not shown). RO3306 experiments in this study demonstrated that inhibition of Cdk1 activity in Leydig cells is an important condition for steroid hormone production via reduced Drp1 Ser 616 phosphorylation. In addition to the kinase inhibitor experiments, the effects of Drp1 knockdown on Drp1 phosphorylation and steroid hormone production were examined in this study. When Drp1 was knocked down to considerably low levels in Leydig cells, dbcAMP still increased Drp1 Ser 637 phosphorylation and decreased Drp1 Ser 616 phosphorylation with a similar tendency as observed in the control (scrambled) siRNA transfected group, at which P4 production was significantly reduced. This implies that Drp1 is an important constituent of the steroidogenic machinery in Leydig cells.

Unlike MA-10 cells, adult Leydig cells of the testes rarely divide and finally produce T. Therefore, in the current study, the major findings that were acquired from MA-10 cells were verified in primary Leydig cells isolated from adult rat testes. Similar to MA-10 cells, among fission- and fusion-related proteins, only Drp1 proteins and their phosphorylation correlated with dbcAMP-induced T biosynthesis. The behaviors of Drp1 Ser 637 and Ser 616 phosphorylation were exactly identical to those observed in MA-10 cells, confirming that Drp1 phosphorylation is commonly involved in the regulation of T production in primary Leydig cells. Furthermore, the ultrastuctural morphology of primary Leydig cells depicted that round-shaped fragmented mitochondria were more frequently seen in control Leydig cells than in dbcAMP-treated cells, signifying that the cellular process involving mitochondrial fission is suppressed at the initial stage of steroidogenesis during T production. In the current study, immunohistochemical analysis of Drp1 proteins in each developmental stage from immature to adult testes demonstrated that Drp1 expression is indeed developmentally upregulated in the Leydig cells of the testis, which are comparative to the stages that are acquired during the steroidogenic activity in Leydig cell lineages (53). Finally, testicular T production was stimulated by exogenous injection of hCG to immature male rats to see whether Drp1 proteins in immature Leydig cells are regulated by gonadotropin in vivo. Surprisingly, Drp1 proteins in immature Leydig cells were remarkably upregulated in response to hCG, and the patterns of Drp1 Ser 637 and Ser 616 phosphorylation were similar to those observed in MA-10 cells as well as primary adult Leydig cells in vitro. These results suggest that Drp1 is regulated by gonadotropin in testicular Leydig cells and that Drp1 phosphorylation, particularly at Ser 637, is important for T production.

The involvement and importance of mitochondrial fusion in steroid biosynthesis has been demonstrated in MA-10 Leydig cells by Duarte et al. (67). They demonstrated that Mfn2 is upregulated with the increase in mitochondrial fusion after hormone stimulation by dbcAMP and hCG. Their findings and approaches are pioneering for understanding steroidogenesis in the aspect of linking mitochondrial dynamics with steroid hormone production. Consistent in part with their findings, an increase in mitochondrial fusion (elongation) was also observed after dbcAMP treatment in the current study. However, in that study, few missing links are found, especially in the regulation of Mfn2 by hormone stimulation. Although they showed that mitochondrial fusion induced by cAMP is reduced in the presence of H89 PKA inhibitor, the status (or downregulation) of Mfn2 proteins was not demonstrated. Based on the corresponding result (Fig. 5A) shown in the current study, dbcAMP-induced Drp1 Ser 637 phosphorylation via PKA activation is suppressed in the presence of H89, and thus increased mitochondrial fission. Therefore, their observations on the reduced mitochondrial fusion could be a result of increased mitochondrial fission by reduced Drp1 Ser 637 phosphorylation by H89 through PKA inhibition. In fact, several recent studies support the current concept that phosphorylation of Drp1 by PKA blocks its translocation to the IMM, leading to mitochondrial elongation rather than fission (22, 23, 68, 69). This is a very important point for understating mitochondrial dynamics involved with adequate preparations of the mitochondrial environment and balance between fission and fusion during steroidogenesis. There is no doubt that PKA is central to steroid hormone biosynthesis (31) as well as the regulation of mitochondrial function and shaping (22, 23). In this respect, there is a weak relationship between Mfn2 and PKA at present. To date, the cellular and molecular mechanisms by which Mfn2 is regulated by cAMP (or gonadotropin) remain undiscovered. However, it is considered that mitochondrial fusion by Mfn2 is phenomenally important for steroid biosynthesis.

In conclusion, this study shows that Drp1 is regulated by cAMP and that its phosphorylation by PKA activation plays a determinant role in mitochondrial shaping for offering the optimal environment for steroid hormone biosynthesis in Leydig cells. The importance of this study is that the potential involvement of Drp1 in mitochondrial dynamics was demonstrated both in in vitro and in vivo during T production. Therefore, it is suggested that Drp1 Ser 637 phosphorylation by PKA is indispensable for steroidogenesis in Leydig cells. It is thus thought that the Drp1 Ser 637 phosphorylation ultimately results in mitochondrial elongation via relative attenuation of mitochondrial fission during steroidogenesis. Further studies on the biochemical and molecular mechanisms underlying Drp1-mediated balancing events between mitochondrial fission and fusion in steroidogenic cells will be helpful to understand intangible steroid hormone–related endocrine and pathological disorders.

Abbreviations:

Abbreviations:
 
• 3β-HSD

  3β-hydroxysteroid dehydrogenase

 
• ACBD3

  acyl-coenzyme A binding domain containing 3

 
• Cdk1

  cyclin-dependent kinase 1

 
• CYP11A1

  cytochrome P450 11A1

 
• dbcAMP

  dibutyryl cAMP

 
• Drp1

  dynamin-associated protein 1

 
• ER

  endoplasmic reticulum

 
• Fis1

  mitochondrial fission 1 protein

 
• hCG

  human chorionic gonadotropin

 
• Hsp60

  60-kDa heat shock protein

 
• IC

  immunocytochemistry

 
• IMM

  inner mitochondrial membrane

 
• LHR

  LH receptor

 
• Mfn1

  mitofusin 1

 
• Mfn2

  mitofusin 2

 
• MTG

  MitoTracker Green FM

 
• mt-Keima

  mitochondrial-targeted form of Keima

 
• NAO

  10-N-nonyl-acridine orange

 
• OPA1

  optic atrophy type 1

 
• P4

  progesterone

 
• pDrp1

  phosphorylated Drp1

 
• PKA

  protein kinase A

 
• PRKAR1A

  protein kinase cAMP-dependent type I regulatory subunit α

 
• RT

  room temperature

 
• siRNA

  small interfering RNA

 
• StAR

  steroidogenic acute regulatory protein

 
• T

  testosterone

 
• TIF

  testicular interstitial fluid

 
• TSPO

  translocator protein

 
• VDAC1

  voltage-dependent anion channel 1

 
• WB

  western blot

Acknowledgments

Current Affiliation: Y.-J. Kim’s current affiliation is the Division of Medical Oncology, Department of Internal Medicine, Korea University College of Medicine, Korea University, Seoul 152-703, Republic of Korea. S. G. Lee’s current affiliation is the Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario K1H 8M5, Canada. J. Y. Kim’s current affiliation is the Division of Medical Oncology, Department of Internal Medicine, Korea University College of Medicine, Korea University, Seoul 152-703, Republic of Korea. J.-Y. Chung’s current affiliation is Wilmer Eye Institute, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205.

Financial Support: This work was supported by National Research Foundation of Korea Grant 2014 001483 (to J.-M.K.) funded by the Korean government.

Disclosure Summary: The authors have nothing to disclose.