| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Animal Reproduction and Biotechnology Laboratory, Department of Biomedical Sciences, Colorado State University, Fort Collins, Colorado 80523-1683
ABSTRACT
The mechanisms responsible for the increased basal rates of progesterone secretion from large steroidogenic luteal cells (LLC) relative to small steroidogenic luteal cells (SLC) have not been clearly defined. To determine if protein kinase A (PKA) is tonically active in LLC, the adenylate cyclase activator forskolin and a specific PKA inhibitor (PKI) were utilized in a 2 x 2 factorial treatment with each steroidogenic cell type. Progesterone and cAMP production were quantified after the different treatments. In addition, the effects of the treatments on the concentrations and relative phosphorylation status of the steroidogenic acute regulatory (STAR) protein in the two cell types were determined as a measure of PKA activity. Treatment with PKI blocked forskolin-induced increases in progesterone secretion by SLC without affecting the production of cAMP. The treatment of LLC with PKI significantly decreased basal progesterone secretion in the presence or absence of forskolin, indicating that the high level of steroidogenesis in this cell type requires PKA activity. There were no differences in the steady-state concentrations of STAR protein in either cell type after treatment. However, the percentage of relative STAR phosphorylation was higher in the LLC than in SLC, and PKI treatment significantly decreased the phosphorylation of STAR in the LLC. The relative phosphorylation status of STAR and the concentrations of progesterone in the media were significantly correlated with the treatments in both cell types. The amount of progesterone secreted per picogram of cAMP was higher in the LLC than in the SLC, and this was accompanied by a significant increase in the ratio of relative STAR phosphorylation to the steady-state concentration of STAR protein. These data are compatible with the theory that LLC are constitutively steroidogenic, partly because they have tonically active PKA. In addition, the phosphorylation of STAR appears to be a primary activity of PKA in both types of ovine steroidogenic luteal cells.
corpus luteum,, kinases,, mechanisms of hormone action,, phosphorylation,, progesterone,, steroidogenesis,, steroidogenic acute regulatory protein
An intriguing, unanswered question regarding progesterone production by the corpora lutea (CL) of ruminants is why large steroidogenic luteal cells (LLC) appear to be constitutively steroidogenic, while small steroidogenic luteal cells (SLC) have lower basal rates of production of progesterone that increase dramatically in response to hormonal stimulation [1]. The transport of cholesterol to the mitochondria and across the mitochondrial membrane is the rate-limiting step in steroidogenesis and is the step that is most acutely influenced by second messengers [reviewed in 2]. Factors that activate protein kinase A (PKA) increase the production of progesterone, while the activation of protein kinase C (PKC) is inhibitory for progesterone production [3]. Therefore, factors that confer a constitutively active steroidogenic phenotype are likely to influence the transport of cholesterol across the mitochondrial membrane.
The addition of cAMP stimulates progesterone secretion by small but not large ovine luteal cells [4]. Thus, steroidogenesis by LLC appears to be independent of the cAMP/PKA pathway or PKA is tonically active in LLC. The latter possibility seems the most likely, as treatment with a specific PKA inhibitor decreased the basal rates of steroidogenesis by LLC of the sheep [5]. While the notion of tonically active PKA in LLC is not new [6–8], the available information is limited. Therefore, an aim of the current study was to investigate whether LLC contain tonically active PKA. Steroidogenic acute regulatory protein (STAR) is responsible for the transport of cholesterol to the mitochondria and across the mitochondrial membrane, and STAR is regulated by the cAMP/PKA pathway through both transcriptional and posttranslational events [2]. Therefore, a second aim of the current study was to quantify the concentration and phosphorylation status of STAR in the two steroidogenic cell types.
Tissue culture supplies (plates, pipettes, etc.) were purchased from Sarstedt Inc., Dulbecco modified Eagle medium (DMEM) and antibiotic/antimycotic concentrate (penicillin G, streptomycin, and amphotericin B) were from Mediatech Inc., and fetal bovine serum (FBS) was from Gemini Bio-Products. Chemicals for cell treatments included: myristoylated protein kinase inhibitor (PKI) 14–22 amide from BioSource Inc., forskolin and high-density lipoprotein (HDL) from Calbiochem/Novabiochem Corp., and phorbol 12-myristate 13-acetate (PMA) from Sigma-Aldrich. For the preparation of cellular homogenates, protease inhibitor cocktail was purchased from Roche Diagnostics Inc., and phosphatase inhibitor cocktail 1 from Sigma-Aldrich. Radioiodinated cAMP was purchased from Perkin Elmer Life Sciences. Supplies used for protein purification included: dialysis membrane (6–8000 molecular weight cut-off) from Spectrum Labs, Centricon-10 filter devices and 0.2-µm syringe filters from Millipore Corp., isopropyl ß-D-1-thiogalactopyranoside (IPTG) from Gold Biotechnology Inc., Affi-Gel 10 gel from Bio-Rad, and nickel-nitrilotriacetic acid (Ni-NTA) agarose was purchased from Qiagen. Supplies for the ELISAs included black 96-well tissue culture-treated plates from Corning Inc., calf intestinal alkaline phosphatase (CIAP) from MBI Fermentas, and Supersignal ELISA Pico chemiluminescent substrate and Supersignal ELISA Femto chemiluminescent substrate from Pierce Biotechnology. For the immunoprecipitations, protein A-agarose was obtained from Invitrogen Life Sciences, protein G-agarose and x-ray film were from Pierce Biotechnology, and enhanced chemiluminescent substrate (ECL) was purchased from Amersham Biosciences.
Antibody Production and Purification
The modified sandwich ELISA involved: 1) a STAR-specific capture antibody; 2) a secondary antibody that was either STAR-specific (for determining STAR protein concentration) or phosphoprotein-specific (for measuring phosphorylation state); and 3) an antibody specific for immunoglobulin (IgG) of the species used to produce the secondary antibody, conjugated to horseradish peroxidase (HRP). An affinity-purified rabbit anti-STAR polyclonal antibody (Affinity Bioreagents) was used as the capture antibody. The secondary STAR-specific antibody was produced in a goat that was immunized with amino acids 115–127 of the ovine STAR sequence conjugated to BSA (Affinity Bioreagents). The anti-STAR serum (Affinity Bioreagents) was antigen affinity-purified prior to use [9]. A mouse monoclonal anti-serine/threonine phosphoprotein antibody (BD Biosciences) served as the secondary antibody for the STAR phosphoprotein ELISA. Antibodies conjugated to HRP were rabbit anti-goat IgG (Santa Cruz Biotechnology) and goat anti-murine IgG1 preadsorbed with human IgG (Santa Cruz Biotechnology). The HRP-conjugated goat anti-rabbit IgG (Santa Cruz Biotechnology) antibody was used for Western analysis.
Recombinant STAR Protein Production
The cDNA that encodes the 37-kDa form of ovine STAR [10] was cloned into the pRSET A expression vector and transformed into Escherichia coli strain BL21(DE3) pLysS (Invitrogen Life Sciences). Expression from the pRSET A vector resulted in the expression of a protein with six histidine residues (6x-His) at the N-terminus, which allowed purification of the STAR protein based on the high affinity of the 6x-His tag for nickel [11]. The transformed cells were cultured in a volume of approximately 1.4 L to an optical density at 600 nm of 0.5–0.6. The expression of STAR protein was induced by the addition of 2 mM IPTG for 4 h at 37°C. Cells were harvested by centrifugation at 5000 x g for 15 min, and the pelleted cells were stored overnight at –70°C. Cells were lyzed in 12.5 ml of 8 M urea, 10 mM Tris-HCl (pH 8.0), 25 mM imidazole, and incubated at room temperature for 1 h. The lysate was cleared by centrifugation at 10 000 x g for 30 min at 4°C. The supernatant was collected, 1.5 ml of 50% Ni-NTA agarose was added, and the mixture was incubated at room temperature for 1 h. The Ni-NTA agarose was pelleted by centrifugation at 5000 x g for 5 min, washed twice with 4 ml of 8 M urea, 10 mM Tris-HCl (pH 6.3), and twice with 4 ml of 300 mM NaCl, 20 mM imidazole (pH 8.0). The recombinant STAR protein was then eluted in four fractions (500 µl each) with 300 mM NaCl, 250 mM imidazole (pH 8.0). Eluates were pooled and dialyzed against Tris-buffered saline (TBS) overnight at 4°C. The eluate was concentrated by centrifugation through Centricon-10 filter devices, the protein concentration was determined, 25 µl of freezing buffer were added per 75 µl of eluate, and small aliquots were stored at –20°C. In addition, 40 µl of the pooled eluate collected prior to the dialysis step was resolved by 12% SDS-PAGE, and the gel was stained with Coomassie Brilliant Blue. Protein bands were visualized using a NucleoTech UV gel box, and the purity of the band that corresponded to STAR was determined using the Gel Expert software.
Sandwich ELISA for STAR Protein
All volumes are 100 µl unless otherwise noted, and all washing steps were performed by filling the plate wells four times with TBS. Standards were developed from recombinant STAR protein (0–300 ng/well), and quality controls (QCs) consisted of 15 µg of SLC protein without (low QC), with 25 ng (medium QC), or with 75 ng (high QC) of recombinant STAR protein added. The capture antibody (rabbit anti-STAR) was diluted in TBS to 2.5 µg/ml, and used to coat the wells of black 96-well plates by passive adsorption for approximately 24 h at 4°C. The capture antibody solution was removed, and the wells were washed, followed by blocking with 300 µl of TBS with 5% non-fat dried milk for 4 h at room temperature. Recombinant STAR protein standards, QCs, and unknowns were added to triplicate wells and incubated at 37°C for 3 h. The solution was decanted, the plate was washed, and 75 ng/well of secondary antibody (goat anti-STAR) was added and incubated overnight at 4°C. After removal of the secondary antibody and washing, 75 ng/well of HRP-conjugated rabbit anti-goat IgG were added and incubated at room temperature for 4 h. Control wells were included that were not coated with capture antibody and did not receive sample, but were incubated with the secondary and HRP-conjugated antibodies. The average luminescence values of these control wells were subtracted from the average values for each standard, QC or unknown, to control for background luminescence. Following the final wash, SuperSignal ELISA Pico chemiluminescent substrate was added. Luminescence measurements were carried out in a Mithras LB 940 plate reader (Berthold Technologies Inc.), which was programmed to shake for 1 min followed by measurement of luminescence with a 425 ± 20-nm filter for 0.2 sec/well. In addition, the plate was exposed to x-ray film for 2–5 min to provide visual verification of the machine readings.
Sandwich ELISA for STAR Phosphorylation
The procedure to measure the degree of phosphorylation utilized 3 µg/ml of the capture antibody, 0.75 µg/ml of the anti-serine/threonine phosphoprotein secondary antibody, and 0.75 µg/ml of HRP-conjugated goat anti-mouse IgG1 antibody that was preabsorbed with human IgG. The substrate for detection was the SuperSignal ELISA Femto chemiluminescent substrate. The amount of protein analyzed was adjusted based on the results from the ELISA determining the concentration of STAR protein, so that 20 ng of STAR were added per well.
To create a standard curve for STAR phosphorylation, protein from forskolin/PMA-treated mixed luteal cells was designated as the 100% phosphorylated STAR standard, since STAR is potentially phosphorylated by PKA and PKC [1, 12]. Forskolin activates the cAMP/PKA pathway [13], and PMA activates PKC [14]. Bacterially produced recombinant STAR protein was used as the nonphosphorylated standard. The maximally phosphorylated and nonphosphorylated stocks were mixed to produce a set of standards with 100%, 75%, 50%, 25%, and 0% STAR phosphorylation, while holding constant the total amount of STAR protein. Quality controls with different levels of relative STAR phosphorylation were developed based on the findings that STAR is partially phosphorylated under basal conditions [15], pharmacological activation of PKA enhances phosphorylation of STAR [15–17], and activation of PKC is most effective at inducing the phosphorylation of STAR [18]. Therefore, three QCs were included in each assay, which consisted of mixed luteal cells treated with nothing (low phosphorylation), 10 µM forskolin (medium phosphorylation) or 100 nM PMA (high phosphorylation). Control wells that received all three antibodies but no sample were included, and their average luminescence values were subtracted to correct for background. The percentages of STAR phosphorylation in the samples and QCs were determined using the background-corrected average luminescence compared to the standard curve.
Preparation of Crude Corpora Lutea Homogenate
CL were collected from normally cycling ewes on Day 11 postovulation and were dissected into eight parts; one-eighth of a CL from three different ewes was homogenized in 3 ml of homogenization buffer (250 mM sucrose, 10 mM Tris-HCl [pH 7.4], 10 mM EDTA, 1 mM sodium orthovanadate, protease inhibitor cocktail, phosphatase inhibitor cocktail 1). The homogenates were cleared by centrifugation (5000 x g for 5 min), and the protein concentration was determined by a protein assay [19] using BSA as the standard. Aliquots were frozen at –20°C until used.
For quantitative Western blot analysis, standards (3–300 ng/lane of recombinant STAR) and unknowns were resolved by 12% SDS-PAGE, and transferred to a nitrocellulose membrane. Unbound sites were blocked with TBS that contained 0.1% Tween-20 and 2% (w/v) BSA at room temperature for 1 h. The membranes were probed with an IgG-purified rabbit anti-STAR antiserum (1:1500), which was previously raised in our laboratory against amino acids 123–134 of the ovine STAR sequence. The anti-STAR serum was diluted in TBS that contained 0.1% Tween-20 and 1% BSA, and incubated with the membranes overnight at 4°C. The membranes were washed with one change of TBS and three changes of TBS that contained 0.1% Tween-20. An anti-rabbit IgG conjugated to HRP was diluted 1:3500 and incubated with the membranes for 1.5 h at room temperature. The membranes were washed as before, and the blots were developed with enhanced chemiluminescent substrate for 5 min, exposed to x-ray film for 1–3 min, and scanned using the STORM imager. The intensities of the bands detected by STORM imaging were quantified with the ImageQuant 5.2 software.
To demonstrate the specificity and accuracy of the ELISA used to quantify the relative STAR phosphorylation states, immunoprecipitation (IP) followed by Western analysis was performed with mouse Leydig tumor cells (MA-10). Mouse Leydig tumor cells were used for IP in preference to purified luteal cells due to the large amount of protein needed (200 µg/reaction), and the limited availability of purified luteal cells. The cells were treated with 10 µM forskolin or 100 nM PMA. In addition, a portion of the cellular protein isolated from PMA-treated cells was dephosphorylated with alkaline phosphatase and used in the IP reactions, along with negative controls that consisted of the lysate of forskolin-treated cells without antibody. Dephosphorylation was accomplished by incubating the cellular protein with 1 U of CIAP per 3 µg of total protein at 37°C for 75 min, followed by heat inactivation of CIAP at 65°C for 20 min. The IP reactions contained 200 µg of cellular protein, 3 µg of antibody, 50 µl of protein A-agarose, 25 µl of protein G-agarose (for IP reactions with mouse IgG1), and TBS to a final volume of 500 µl. Different reactions were performed using either the rabbit anti-STAR or the anti-serine/threonine phosphoprotein antibody in the IP reaction. The reactions were incubated overnight at 4°C on an end-over-end shaker. The protein A-agarose and G-agarose were pelleted by centrifugation at 5000 x g for 3 min at 4°C. The rabbit anti-STAR IP reactions were washed three times with 500 µl of wash buffer (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1 mM EDTA, 5 µg/ml aprotinin, 5 µg/ml leupeptin, 0.1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS), and the anti-serine/threonine phosphoprotein IP reactions were washed three times with 500 µl of TBS. After washing, 40 µl of TBS were added, and the tubes were heated at 95°C for 3 min. The agarose was pelleted by centrifugation at 5000 x g and the supernatant was collected. The supernatants were subjected to Western blot analysis using the rabbit anti-STAR capture antibody (8 µg) or goat anti-STAR antiserum (1:1500) to probe the membranes.
All experimental protocols involving the use of animals were approved by the Colorado State University Institutional Animal Care and Use Committee. Western range ewes were superovulated [20]. Ten days after hCG treatment, the ewes were anesthetized with pentobarbital (11 mg/kg body weight), and CL were collected by midventral laparotomy [20]. The CL were pooled from 2–3 ewes per day, and partially purified preparations of SLC and LLC were obtained as previously described [21] on five separate days (n = 5).
SLC (2–3 x 106/plate) and LLC (5–6 x 105) were plated on 60-mm tissue culture plates overnight in DMEM that was supplemented with 10% heat-inactivated FBS, 100 IU/ml penicillin G, 100 µg/ml streptomycin, and 250 ng/ml amphotericin B, and cultured at 37°C in 5% CO2.
The PKA inhibitor PKI was derived from amino acids 14–22 of the skeletal muscle cAMP-dependent protein kinase inhibitor and was myristoylated at its N-terminus to enhance cell permeability [22–23]. Solutions of PKI were prepared fresh each day by diluting lyophilized PKI in serum-free DMEM to a final concentration of 500 µM. Forskolin was prepared in dimethyl sulfoxide (DMSO) at a concentration of 25 mM. The day after cell purification and plating, the media were aspirated and replaced with serum-free culture media. Six hours later, purified preparations of LLC and SLC were pretreated for 1 h with nothing or 50 µM PKI, followed by a 2 x 2 factorial treatment that resulted in each cell type being treated with: 1) 0.04% DMSO (vehicle); 2) 10 µM forskolin; 3) 50 µM PKI in 0.04% DMSO; and 4) 10 µM forskolin plus 50 µM PKI. Each treatment was added to duplicate plates, and all groups received 50 µg/ml human HDL, to supply the cholesterol needed for steroidogenesis [24–25]. Cells were incubated for 4 h, the media were collected, and the cells were washed twice with TBS. Both the media and cells were frozen at –20°C until analyzed.
At the end of the treatment period, cells were collected in 200–500 µl per plate of cold homogenization buffer. Cells from duplicate plates were pooled and lyzed by sonication with a Branson Sonifier using four 10-sec bursts on ice. The lysates were boiled at 100°C for 4 min immediately following sonication, concentrated in a Savant Speed Vac for 20–40 min, and the total protein concentration was determined [19].
The concentrations of progesterone in the media were determined by radioimmunoassay [26]. The interassay coefficients of variation (CVs) were 15.6%, 3.4%, and 7.6% for the low, medium, and high concentration QCs, respectively. The intraassay CVs averaged 8.9% and 17.8% at 20% and 80%, respectively, on the theoretical standard curve for the three assays. The sensitivity of the assays averaged 980 fg per tube.
The concentrations of cAMP in media were determined by radioimmunoassay [27] following acetylation of the samples [28–29]. The assay had average intraassay CVs of 7.9% and 15.7% at 20% and 80% on the theoretical standard curve, respectively, and interassay CVs of 6.9% and 4.6%, respectively, for the low and high concentration QCs. The average assay sensitivity was 92 fg per tube.
The data for the concentrations of progesterone and cAMP in the media, the concentrations of STAR protein, and the percentages of phosphorylation of STAR were analyzed using one-way ANOVA in the SAS program. The raw data for progesterone and cAMP in the media were normalized to the number of cells (number of cells plated x 0.65 for SLC due to nonsteroidogenic cell contamination [21], LLC preparations were considered to be 100% steroidogenic) and log-transformed prior to statistical analysis. Potential outliers were excluded from analysis if their values lay outside the 95% confidence interval calculated from the remaining values. Only 4/40 values for relative STAR phosphorylation were excluded. The Duncan multiple range test was used to compare the means for all four responses measured within each cell type and across treatments. Least significant differences were used to compare means when only 1–4 comparisons were made. A correlation analysis among the four responses measured in the experiment was performed using the average value for each response, and the data from the LLC and SLC were pooled. In addition, correlations were plotted within individual cell types, to verify that the average correlations were representative of each cell type.
Validation of ELISA for the Quantification of STAR Protein
A band of approximately 40 kDa (37-kDa STAR + polyhistidine tag + linker amino acids) was detected by Coomassie Brilliant Blue staining of the purified preparation of recombinant protein, and this band was confirmed as the STAR protein by Western analysis using two different STAR-specific antibodies (Fig. 1). A standard curve for STAR protein pooled from four assays had a linearity of R2 = 0.99 over a range of 0–300 ng of STAR protein per well (Fig. 2A). The intraassay CVs averaged 17.4%, 15.5%, and 13.9%, and the interassay CVs were 12.7%, 13.5%, and 4.8% for the low, medium, and high QCs, respectively. The intraassay variation was higher because it was calculated from the luminescence of individual wells, whereas the interassay variation was determined from the mean value of triplicate wells. Recombinant STAR (0–300 ng/well) and CL homogenate (10, 30, and 60 µg/well) were analyzed in three independent assays. The quantity of STAR in the CL homogenate was calculated from the standard curve, and the curves were demonstrated to be parallel (Fig. 2B). The accuracy of the ELISA was tested against quantitative Western analysis. The concentration of STAR protein was quantified by both methodologies in 10 µg of CL homogenate with 0, 10 or 30 ng of recombinant STAR added. The two methods yielded similar results (Fig. 2C).
|
|
Validation of ELISA to Quantify STAR Phosphorylation
The standard curve for the relative phosphorylation of STAR pooled from four assays had a linearity of R2 = 0.97 (Fig. 3A). The intraassay CVs were 14.7%, 4.1%, and 15.2%, and the interassay CVs were 5.5%, 12.2%, and 24% for the low, medium, and high QCs, respectively. Alkaline phosphatase sensitivity was used to demonstrate specificity of the assay for quantifying changes in phosphorylation status. As seen in Figure 3B, treatment with alkaline phosphatase caused an approximately 90% decrease in the luminescence measured with the STAR phosphorylation ELISA but did not affect the luminescence values measured with the ELISA used to quantify the concentration of STAR protein.
|
Immunoprecipitations (IPs) were utilized to demonstrate further the specificity of the ELISA for quantifying STAR phosphorylation, and to test the accuracy of the assay. Using the anti-serine/threonine phosphoprotein antibody in the IP reaction, both the goat (Fig. 4A) and rabbit (Fig. 4B) anti-STAR antibodies detected bands at 
30 kDa by Western analysis following forskolin or PMA treatment of MA-10 cells. MA-10 cells were used for these studies because they represent a homogenous population that expresses the STAR protein and responds to forskolin and PMA. Alkaline phosphatase treatment of protein from PMA-treated cells decreased the band intensity to control levels (Fig. 4, A and B) when the anti-serine/threonine phosphoprotein antibody was used in the IP reaction. When the rabbit anti-STAR antibody was used in the IP reaction, bands of similar intensity at 30 kDa were detected by Western analysis with the goat anti-STAR antibody (Fig. 4A) in all groups, except for the control. The predominant form of STAR was the 30-kDa form, probably because it had a 4-h half-life, compared to the 15-min half-life of the 37-kDa form [30].
|
Concentrations of P4 and cAMP in Culture Media
Treatment of SLC with forskolin was followed by a 7.8-fold increase (P < 0.05) in the concentration of progesterone in the medium (Fig. 5A), while treatment with PKI alone significantly (P < 0.05) decreased basal progesterone synthesis. Cotreatment with PKI and forskolin limited (P < 0.05) the forskolin-induced increase in progesterone secretion to 1.1-fold of the nontreated control. For the LLC (Fig. 5A), forskolin treatment did not induce a significant increase in progesterone secretion. The addition of PKI to untreated or forskolin-treated LLC resulted in significant (P < 0.05) decreases in basal progesterone secretion of 84% and 77% decreases, respectively, compared to untreated cells. Based on equal numbers of cells, LLC had an approximately 11-fold higher basal production of progesterone (Fig. 5A) than SLC under control conditions.
|
Forskolin treatment of SLC was followed by an approximately 5-fold increase in the concentration of cAMP in the medium (Fig. 5B; P < 0.05). The basal concentrations of cAMP in the media were not different between nontreated SLC and SLC that received PKI alone, and cotreatment with PKI did not significantly inhibit the forskolin-induced increase in cAMP. In the LLC, there was an approximately 5.7-fold increase (P < 0.05) in the concentration of cAMP in the medium following forskolin treatment. Basal secretion of cAMP into the medium was not different between non-treated and PKI-treated cells. Forskolin- and PKI-cotreated cells had significantly (P < 0.05) lower concentrations of cAMP in the medium than forskolin-treated cells, whereas they had significantly (P < 0.05) higher concentrations of cAMP in the medium than nontreated or PKI-treated cells. The LLC had 2.3-fold higher basal concentrations of cAMP in the medium than the equivalent number of SLC. The ratio of progesterone to cAMP secreted was calculated for each cell type and treatment; untreated LLC had significantly (P < 0.05) higher basal secretion of progesterone per picogram of cAMP than untreated SLC (Fig. 5C).
STAR Protein Concentrations and Relative STAR Phosphorylation Status
There were no significant differences in the steady-state concentrations of STAR protein between the treatment groups for either cell type (Fig. 6A). Ten luteal cell samples, consisting of both cell types and all treatments, were reanalyzed for concentrations of STAR protein after being diluted for quantification of relative STAR phosphorylation. The 10 samples averaged 19.7 ± 2.9 ng of STAR protein per well, which was close to the targeted amount of 20 ng per well. This confirmed that quantification of the phosphorylation state of STAR was performed with approximately equal quantities of STAR protein.
|
SLC treated with PKI or PKI and forskolin had significantly (P < 0.05) lower levels of relative STAR phosphorylation than forskolin-treated cells (Fig. 6B). In LLC (Fig. 6B), there was no change in STAR phosphorylation following forskolin treatment. Treatment with PKI in the presence or absence of forskolin resulted in a significant (P < 0.05) decrease in STAR phosphorylation compared to nontreated or forskolin-treated cells. The LLC also had a significantly (P < 0.05) higher ratio of relative STAR phosphorylation to steady-state concentrations of STAR protein than SLC under nontreated conditions (Fig. 6C).
Correlation analysis was performed between the four responses analyzed in these experiments, using data from both LLC and SLC (Table 1). In addition, correlations between responses were plotted individually for the SLC and LLC (data not shown), to verify that the average correlations were representative of each cell type. The concentrations of cAMP and progesterone in the media were significantly correlated at r = 0.78 (P < 0.05). The relative STAR phosphorylation and the progesterone concentrations in the media were also significantly correlated (r = 0.71, P < 0.05). There was no correlation between the concentrations of STAR protein and progesterone (r = –0.29, P = 0.49). There was a significant negative correlation between the concentrations of STAR protein and the relative STAR phosphorylation status (r = –0.76, P < 0.05), although this was specific for SLC.
|
Ligand binding to the LH receptor stimulates steroidogenesis in SLC, and this effect can be mimicked by pharmacologic activation of PKA [3, 21]. Therefore, progesterone production by SLC served as a control for PKI-induced inhibition of PKA activity. The addition of PKI to SLC resulted in a significant decrease in basal progesterone secretion, which suggests that an active PKA system is present in these cells 24–30 h after tissue collection and culture in the absence of continued activation of the PKA system. Treatment with PKI prevented forskolin-induced increases in progesterone synthesis by SLC, which would be expected if this increase was due to increased PKA activity. In a previous study, PKI was used to inhibit PKA in SLC and LLC, although there was no control for possible toxic effects of PKI [5]. Presumably, PKA is important for cellular functions other than steroidogenesis, so blocking PKA activity may decrease progesterone synthesis simply by causing cellular toxicity. The concentrations of cAMP in the media were analyzed to test for possible toxicity due to the PKI treatment. If cAMP production is unaffected by PKI treatment, it seems likely that everything upstream of PKA activation is normal. The concentrations of cAMP have been correlated with the intracellular concentrations of cAMP in ovine luteal cells [4]. There was a significant increase in cAMP secretion by SLC following forskolin treatment, and this effect was not influenced by PKI treatment. Thus, there is no evidence for PKI-induced toxicity in SLC.
The present report describes novel methods for quantifying the concentrations and relative phosphorylation states of STAR in biological samples. These methods allow rapid quantification and analysis of multiple samples, and require small amounts of material. There were no significant differences between the treatments in terms of the concentrations of STAR protein on a per microgram of protein basis in the SLC. The treatment time (4 h) may have been too short to induce changes in the steady-state concentrations of STAR. In addition, the ELISA to quantify the concentrations of STAR protein predominantly quantifies the 30-kDa form of STAR due to its greater abundance, which is the result of its longer half-life relative to the 37-kDa form [30–31]. Any changes in expression of the STAR protein induced by the different treatments would likely be more apparent in the short-lived 37-kDa form. A shortcoming of the present study is that the steady-state concentrations of STAR were measured, so no information was obtained regarding the rates of synthesis and/or degradation of this protein. However, the steady-state concentrations should be a good indicator of the quantities present for steroidogenesis. The relative phosphorylation state of STAR was significantly inhibited by PKI in forskolin-stimulated SLC compared to SLC treated with forskolin alone, which indicates that PKI blocks the ongoing PKA-induced phosphorylation of STAR, which presumably originates from both tonic and forskolin-stimulated PKA activities. Although PKI-treated SLC had less than half of the relative phosphorylation of the STAR protein in the untreated control cells, this difference was not significant. The relatively low levels of ongoing steroidogenesis in these unstimulated cells may make the small differences in the phosphorylation state of STAR difficult to detect reliably.
In the LLC, there was a dramatic decrease in basal progesterone secretion (84% decrease, P < 0.05 compared to nontreated LLC) following PKI treatment. Therefore, it appears that nearly all the progesterone produced by LLC under basal conditions is due to PKA-induced events. The dose of PKI used in the present study was twice that used by Diaz et al. [5], who reported a 50–60% decrease in the secretion of progesterone. The nonstatistically significant increase in progesterone secretion by LLC following forskolin treatment may be due to SLC contamination, dramatically decreased responsiveness of the LLC to cAMP production or both of these factors. Purified LLC may contain up to 20% SLC, which exist in clumps or attached to LLC, while SLC preparations exhibit no significant contamination with LLC [21]. The basal concentrations of cAMP were similar between nontreated and PKI-treated LLC. Forskolin caused a significant increase in cAMP secretion, while cells treated with PKI in the presence of forskolin had concentrations of secreted cAMP that were intermediate between the nontreated and forskolin-treated cells. Although PKI inhibited the forskolin effect on cAMP secretion in LLC, there was still a significant increase in cAMP secretion in response to forskolin in the presence of PKI. Thus, the effect of PKI on progesterone production by LLC appears to be primarily due to a specific negative effect on PKA activity, which consequently inhibits steroidogenesis.
There were no differences between the treatments in the steady-state concentrations of STAR protein in LLC. This may have been due to the fact that the 30-kDa form of STAR was the primary form detected, as discussed for SLC. The basal phosphorylation state of STAR in the LLC was high and did not increase further with forskolin treatment, although cAMP production increased several-fold. This suggests that either the PKA activity in this cell type does not respond to increased levels of cAMP [4] or that the STAR protein in this cell type is already maximally phosphorylated. The fact that the addition of PKI to LLC had such a dramatic effect in reducing the secretion of progesterone indicates that the elevated levels of progesterone synthesis are dependent upon PKA activity and appear to involve increased phosphorylation of STAR.
A significant correlation was found between the relative percentage of STAR phosphorylation and progesterone synthesis, which indicates that phosphorylation of STAR is a key event in PKA-induced steroidogenesis in ovine luteal cells. It is important to note that activators and inhibitors of PKA were used in the present experiments. Therefore, the significant correlation between STAR phosphorylation and the concentrations of progesterone in the media is most likely due to changes in phosphorylation at PKA motifs. Ovine STAR has potential PKC phosphorylation sites [1], and since PKC down-regulates steroidogenesis in luteal cells [3], it seems reasonable to expect that changes in STAR phosphorylation induced by PKC would be negatively correlated with progesterone synthesis.
The data for progesterone and cAMP secretion were normalized to the cell number, to facilitate direct comparisons between the two cell types. LLC had an approximately 11-fold higher basal secretion of progesterone but only 2.3-fold higher basal secretion of cAMP than SLC. The results from the analyses of progesterone production and relative STAR phosphorylation in the presence of PKI indicate that tonically active PKA influences basal steroidogenesis in both SLC and LLC. The ratio of progesterone to secreted cAMP and the ratio of relative STAR phosphorylation to the steady-state concentration of STAR protein were used to examine the importance of tonic PKA activity in the two cell types. LLC secreted significantly more progesterone per picogram of cAMP than SLC under nontreated conditions, while there were no differences between the two cell types following other treatments. The LLC also had a significantly higher ratio of relative STAR phosphorylation to steady-state concentration of STAR protein under basal conditions. These results indicate that although both cell types have tonic PKA activities that contribute to basal steroidogenesis, the relationship between cAMP and progesterone production is closer in SLC than in LLC, which indicates that tonically active PKA is a key modulator of steroidogenesis in LLC.
FOOTNOTES
2Correspondence: Gordon D. Niswender, Colorado State University, Animal Reproduction and Biotechnology Laboratory (ARBL), 1683 Campus Delivery, Fort Collins, CO 80523. 
3Current address: Oregon Regional Primate Research Center, Beaverton, OR 97006.
1Supported by a grant from the Colorado Agricultural Experiment Station and by NIH training grant HD07031 (to R.L.B.).
Correspondence: FAX: 970 491 3557; e-mail: Gordon.niswender{at}colostate.edu
Received: 19 December 2006.
First decision: 7 January 2007.
Accepted: 3 April 2007.
REFERENCES
production in the CL. Mol Cell Endocrinol 2002; 191:65–80[CrossRef][Medline]. Biol Reprod 1990; 42:239–245[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
