| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
in Bovine Theca Cells Stimulated with Luteinizing Hormone1
a Institute for Hormone and Fertility Research, University of Hamburg, D-22529 Hamburg, Germany
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(PKC) increase during the course of pregnancy. Because LPLD is involved in the production of the bioactive phospholipid lysophosphatidic acid (LPA), we examined whether stimulation with LPA would influence PKC in the ovary. We used immunoblotting and immunohistochemical methods to show that stimulation of bovine theca cells with LPA leads to an unexpected redistribution of PKC from the cytosol to the perinuclear area and that in the presence of LH, LPA induces a complete nuclear translocation of PKC. These effects of LPA are dose dependent, can be mimicked by phorbol ester, and are inhibited by a PKC inhibitor, rottlerin. Concomitantly, under the same experimental conditions both LPA and the phorbol ester PMA (4ß-phorbol-12-myristate-13-acetate) augment LH-stimulated progesterone accumulation in this cell system. This functional effect of LPA and PMA is abolished in cells pretreated with rottlerin. It is unclear whether the nuclear localization of PKC indicates a specific function of the enzyme in the bovine ovary. Because PKC supports a luteotropic function in rodent models, a similar role in the bovine ovary is also likely.
corpus luteum, growth factors, signal transducers, signal transduction, theca cells
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(PKC) during pregnancy [5–7], we investigated whether these two events are linked. A specific subcellular localization of a particular PKC isoform may indicate the specific function of this isoform. We therefore evaluated the presence and intracellular distribution of PKC in our cell system under the influence of LPA.
PKC is a calcium-independent protein kinase C involved in the regulation of cellular growth, differentiation, and carcinogenesis [8–10]. This kinase is activated in vivo by diacylglycerol generated from membrane phospholipids upon receptor-mediated stimulation of a phospholipase [9]. Although the genomic sequence of PKC remains unknown, PKC protein is the most thoroughly studied member of the nuclear PKC subfamily [8]. In recent years, attention has been focused on biological functions, substrate specificity, and the localization of PKC mainly in tissues outside of the reproductive system [8]. Special interest in ovarian PKC was generated after it was established that the amounts of PKC increase more than 30-fold during pregnancy [11]. Based on this data, Hunzicker-Dunn and her colleagues [5] proposed that this PKC isoform may be one of the key regulators of luteotropic function during pregnancy.
Luteotropic function of the corpus luteum is maintained by either luteotropic hormones or withdrawal of luteolytic factors and probably by additional rescue stimuli [12]. Gonadotropins such as LH or chorionic gonadotropin are classic luteotropic hormones, whereas other hormones such as growth hormone, prolactin, and estradiol may also play an important role [12]. In the rabbit, LH is simply an initiator, and maintenance of luteal function is taken over by estrogen. In the vast majority of mammals (including cattle), gonadotropins are needed for both the growth and function of the corpus luteum. The bovine corpus luteum remains under the influence of gonadotropins during early pregnancy, with LH regulating the process of remodeling [12].
LH is not able to induce the expression of PKC in rodent models [11, 13]. However, earlier results implicated the ovulatory LH surge in initiation of the expression of PKC in rabbits [5]. Further examination revealed that only stimulation with prolactin or rat placental lactogen (PRL/rPL-1) resulted in an acute activation of PKC [11, 13]. In a more recent study, PKC was involved in the PRL/rPL-1-dependent expression of relaxin [14]. Using luteinized granulosa cell culture, PRL/rPL-1 induces relaxin mRNA expression with Stat 3 (signal transducer and activator of transcription) as a possible target of PKC activity [14]. Estrogen synergizes with PRL/rPL-1 receptor agonists to promote this relaxin expression. Thus, it is unequivocal that both expression and function of PKC in rodents is regulated by PRL/rPL and by estrogen, but not by gonadotropin [5–7, 11, 13, 14]. It remains unclear whether such a regulatory mechanism exists for species other than rodents.
In early bovine corpus luteum, small luteal cells are considered theca derived and the large luteal cells are primarily of granulosa origin; however, the relative contribution of each cell type is still being debated [15–18]. Theca cells certainly play a significant role in follicular growth and development and in ovarian steroid production. Because a majority of LPA receptors known to be expressed in the ovary could be attributed to these cells (unpublished results) and small luteal cells in the bovine ovary possess specific LPA-binding sites [19], luteinized theca cells were chosen as an appropriate in vitro model system to examine the interaction between LPA and PKC in the ovary.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, PKC
, nucleoporin p62, Lamin A/C, and DNA topoisomerase II
were from Santa Cruz Biotechnology (Heidelberg, Germany). The antibody against connexin 43 was from Zymed (through Zytomed, Berlin, Germany), the anti-mitogen-activated protein kinase Erk 1/2-CT was from Upstate (through Biomol, Hamburg, Germany), and the Cy3-/Cy-2 and peroxidase-conjugated secondary antibodies were from Jackson Immunochemicals (through Dianova, Hamburg, Germany). Mitochondrial stain DePsipher was provided by R&D Systems (Wiesbaden, Germany). The peroxidase-based Western blot detection system was from Pierce (through KMF, Sankt Augustin, Germany), and bovine dermal collagen was from Cellon (Strassen, Luxemburg). Bovine LH was a gift from the NIADDK (Bethesda, MD), and Amicon Centricon filters were from Amicon-Millipore (Eschborn, Germany). All other reagents were obtained from commercial sources and were of the highest purity. Theca Cell Isolation, Induction of Luteinization, and Culture Procedures
Methods for isolation and purification of bovine theca cells have been published elsewhere [20]. Bovine ovaries were obtained from the local abattoir. For the isolation of theca cells, large antral follicles of healthy morphological appearance with follicular fluid volume of 1–4 ml and a diameter of 10–25 mm were obtained from ovaries without fresh or mature corpus luteum. Because dominant preovulatory follicles can only be obtained during the relatively short preovulatory period, the antral follicles used may have originated from both follicular and luteal phases. Before setting up the sterile cell preparation, the ovaries were soaked in 3% sodium hyperchloride for 20 min and extensively washed. Follicular fluid was aspirated, the follicles were opened with a scalpel, and the granulosa cells were carefully scraped from the theca layer. The theca interna layer was than carefully peeled off from the theca externa and surrounding stromal tissue using a pair of tweezers. The tissue was weighed, cut into pieces, and transferred into incubation tubes containing Moscona balanced salt solution (8 g/L NaCl, 0.3 g/L KCl, 0.05 g/L Na2PO4, 0.025 g/L KH2PO4, 1 g/L NaHCO3, 2 g/L D (+) glucose) with 0.5% collagenase, 0.1% hyaluronidase, 0.1% pronase, 0.1% BSA, and 1 mg DNase at 10 ml/g wet tissue weight. The enzyme-tissue mixture was incubated for 45 min at 37°C in a water bath with moderate shaking. In 15-min steps, the dispersion of the tissue was facilitated by aspiration of the suspension through a sterile Pasteur pipette. After 45 min, the dispersed mixture was filtered through nylon gauze and centrifuged for 15 min at 100 x g. The cell pellet was washed three times with Moscona balanced salt solution and then resuspended in cell culture medium (medium 1) containing Dulbecco modified Eagle medium/Ham F-12 medium, 2 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. The crude cell suspension was layered over Percoll solution (with a density of 1.07) and centrifuged at 900 x g for 25 min at room temperature. The theca cell layer was carefully aspirated, diluted with medium, and washed three times by centrifugation at 100 x g for 15 min. The purified cells were resuspended in medium 1, and the cell number was determined by counting on a hemocytometer. Cell viability was determined by the trypan blue exclusion test, and the cells were plated in duplicate on six-well plates (2 x 106 cells/well) or in 50-ml polysterol culture flasks (5 x 106 cells/flask) precoated with 0.3% Cellon collagen according to the procedure provided by the manufacturers. Cellon collagen is prepared from pepsin-treated bovine dermis and consists of 99.8% pure collagen (95% type I collagen and 5% type III collagen with no degradation products). In some experiments (i.e., the determination of steroids in culture medium), cells were grown on 24-well plates (250 000 cells/well). To induce the luteinization process in vitro, the theca cells were cultured in medium 1 containing 1.5% heat-inactivated fetal calf serum plus 25 µM forskolin for the first 48 h. From Day 3 on, the cells were maintained further in serum-free medium 1 containing 5 µg/ml BSA and 5 µg/ml insulin, 5 µg/ml transferrin, and 5 ng/ml sodium selenite. The morphology, diameter, and steroid production of these in vitro luteinized theca cells were comparable to those of small luteal cells and have been characterized elsewhere [18, 21]. On Day 6, when the cells had reached full confluence, they were washed with fresh serum-free medium 1 containing only 5 µg/ml BSA and were stimulated as indicated. The methods for isolation and culture of bovine luteal cells were published previously [21].
Cell Treatments
On Day 6, the cell monolayers (grown either in cell culture dishes or on chamber slides) were stimulated with agonists LPA, PMA, or bovine LH for indicated time periods (0–24 h) and were either subjected to subcellular fractionation or were used in immunohistochemical analysis. To monitor possible morphological or cell number changes during the incubation period, cells grown in culture plates were also morphologically examined. Medium was taken from the culture plates and used for the steroid measurements. The adherent cells were fixed with methanol (5 min), stained (May-Grünwald), and visualized using bright-field microscopy.
Subcellular Fractionation
Cells were washed with PBS and were harvested by gently scraping with a rubber policeman into ice-cold isotonic sucrose buffer A containing 10 mM Tris-HCl (pH 7.4) and 0.25 M sucrose plus 10 µg/ml leupeptin, 1 µg/ml pepstatin, 1 µg/ml amastatin, and 1 mM phenyl-methylsulfonic acid. The cells were allowed to swell for 5 min and were gently and carefully homogenized using an all-glass Dounce type homogenizer [22]. The homogenate was centrifuged first at 100 x g for 15 min to yield pellet 1 and supernatant 1. Supernatant 1, which contained cytosol plus particulate membrane fractions (but no nuclear fractions) was left on ice for further preparation, and pellet 1 was resuspended in lysis buffer containing 0.5% Nonidet P-40 and 10 mM Tris-HCl (pH 7.4) plus protease inhibitors (PI). From this suspension, the nuclei were pelleted at 200 x g (pellet 2) and further suspended in 10 mM Tris-HCl buffer (buffer C) containing PI. Pellet 2 was then homogenized in buffer C and centrifuged at 100 000 x g. This step produced a pellet enriched with nuclear membranes (pellet 3) and a nucleocytosol fraction (supernatant 3). Supernatant 1 was recentrifuged at 100 000 x g for 60 min to yield particulate non-nuclear membrane fractions (pellet 4, containing plasma, microsomal membranes, and organelles such as mitochondria and lysosomes) and the cytosol (supernatant 4). Subfractions were filtered through a Centricon Ultrafilter with 10-kDa cut (Amicon-Millipore), and protein content was determined according to the method of Bradford [23]. To measure nuclear integrity, a few drops of the nuclear fraction were released gently onto a microscope slide to measure the number of undisrupted cells (by exclusion of tryptan blue) and the number of intact nuclei (by phase contrast, Giemsa staining). To exclude cross-contamination between the fractions, several antibodies were used as individual markers for each fraction. For example, the application of antibodies against nuclear pore-p63 (nucleoporin) protein and anti-Lamin (Lamin A/C) proteins enabled us to monitor the enrichment of nuclear membrane fractions [24]. The antibody against DNA topoisomerase II served as a control for the nucleocytosol, and the antibody against connexin 43 was used to show the lack of contamination of the non-nuclear particulate fractions by the fractions enriched with nuclear membranes. There was no appreciable cross-contamination between the various fractions within the detection limit of our assay systems.
Immunoblot Analysis of PKC: Protein Expression and Translocation
Equal amounts of the cellular subfractions from stimulated cells (30–100 µg) were separated by SDS-PAGE, transferred to polyvinylidene fluoride (PVDF) membranes, and stained with Ponceau S [22, 25]. The membranes were incubated with affinity-purified polyclonal rabbit antibodies recognizing the PKC isoform (1:1000). The antibody corresponds to amino acids AFKGFSFVNPKYEQFLE of the C-terminus of PKC protein (with no cross-reactivity to other PKC isoforms) and recognizes both the holoenzyme and the phosphorylated form of PKC. The immunocomplex was detected using peroxidase-conjugated affinity-purified goat anti-rabbit secondary antibody (1:5000) and a luminol-based chemiluminescence system. Control blots were processed with nonimmune serum and the secondary antibody. Immunoanalysis of the control proteins (PKC, nucleoporin, Lamin A/C, etc.) was performed in a similar way using the respective affinity-purified secondary antibodies (Jackson Immunochemicals).
Immunohistochemical Analysis
The cells (25 000 cells/chamber) were grown on eight-chamber glass slides (LAB-TEK; Nunc, Roskilde, Denmark) coated with Cellon collagen. The cells and medium were dispersed onto the base gel surface, and the slides were stimulated with agonists. After the stimulation, the chambers were washed with PBS and fixed with 3% paraformaldehyde for 30 min. The slides were extensively washed (three times with PBS), permeabilized for 2 min with 0.5% Triton X-100, and blocked for 60 min with 5% nonimmune serum (species of the secondary antibody). The 60-min incubation with primary antibodies (anti-PKC or anti-PKC, 1:100) was followed by 60 min of treatment with the Cy3-conjugated secondary antibody (1:200). Slides were then viewed with an epifluorescence microscope (Nikon, Düsseldorf, Germany). To counterstain the cell nuclei, a fluorescent groove binding probe for DNA, 4',6'-diamidino-2-phenylindole (DAPI), was applied after the secondary antibody. The control analysis was performed for each single slide using different Nikon filter blocks with excitation wavelengths of 330–380 nm for DAPI staining and 450–490 nm for Cy3 immunofluorescence (using a fluorescein isothiocyanate filter).
Steroid Assays
The amounts of progesterone accumulated in medium were measured by a specific ELISA as described previously [18, 20, 25]. The ELISA is a competitive double-antibody enzyme immunoassay and provides accurate measurements of progesterone in the range of 0.14–34.02 ng/ml, corresponding to 1701 pg/well. Within this range, the interassay coefficient of variation for the lowest standard was <15%. Relative cross-reactivity with structurally related compounds was estimated from the concentration required to yield 50% suppression of the biotinylated progesterone tracer. Cross-reactivity was minor with 11-deoxycorticosterone (9.5%), corticosterone (4.9%), and 17--hydroxyprogesterone (1.1%) and was negligible with an array of other steroids tested. (Full details of all components and protocols can be obtained from Dr. M. Schumacher, Institute for Hormone and Retility Research, University of Hamburg, 22529 Hamburg, Germany.) Androstendione was measured by RIA using a commercial kit (Immunobiological Laboratories, Hamburg, Germany).
Mitochondrial Respiration Assay
Cells were grown on chamber slides, and after cell treatment, the medium was replaced with reaction buffer containing mitochondrial stain (dePsipher) according to the procedure described by the manufacturer. The chamber slides were incubated in the dark at 37°C for 20 min. The cells were then washed with reaction buffer without stain and immediately examined under the fluorescent microscope using filters for fluorescein and rhodamine. The accumulation of energy in healthy cells creates a mitochondrial transmembrane potential, called delta-psi. A lipophilic cation, dePsipher, can be used as a mitochondrial activity marker to differentiate cells with disturbed mitochondrial membrane polarization versus those with no change in the membrane potential. This mitochondrial activity marker can be visualized in these two types of cells as an orange-red aggregate (formed upon membrane polarization) or a green monomeric form (in those cells in which the mitochondrial potential is disrupted and the marker cannot access the transmembrane space). In cells with intact mitochondrial potential, the mitochondria appeared as red aggregates emitting at 590 nm. In cells with disrupted potential, the dye remained in its monomeric green form with emission at 530 nm.
Data Analyses
Each experiment was performed at least three times using the ovaries from different animals (n = 3). All blots and films were scanned, and specific bands were analyzed densitometrically using the NIH Scion Image Program (Scion Corp., Frederick, MD). Equal amounts of protein separated in each blot were stained as described above and documented. The immunoblots were quantitated with the computer-assisted analysis program using the threshold and density slicing modes to provide area and density measurements of the gray scale immunoreactive images. The background was substracted, and the values were corrected against the control bands (i.e., Lamin). The intergrated optical density bands were then analyzed statistically. The data are presented as mean values (±SD) from three experiments. Each representative data set (shown as inserts) are from the same individual representative experiment.
Changes in the intracellular protein relocalization (i.e., nuclear translocation) were monitored using digitized epifluorescence microscopy and were calculated as a percentage of cells showing redistribution of immunofluorescence stain versus the total number of cells (DAPI staining) in 500-µm2 areas. To monitor possible morphological changes (or changes in the number of cells) during the stimulation period, the cells grown for steroid production (in 24-well plates) were fixed and stained for control analysis.
Statistical analysis of the data was performed using Instat Prism Software (GraphPad Software Inc., San Diego, CA). The first analyses were performed using standard t-tests, and further analysis was carried out using repeated measures one-way ANOVAs and Dunnett posttests. Differences with P values >0.05 were considered not significant (NS), 0.01–0.05 were considered significant, 0.001–0.01 were considered very significant, and < 0.001 were considered extremely significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and whether treatment with LPA affected the cellular distribution of PKC enzyme.
Cytosolic and membrane fractions were prepared from theca cells following treatment for 240 min (or as indicated) with various concentrations (0–12 µM) of LPA. Immunoblotting with anti-PKC antibody was used to monitor the possible presence and relocalization of this PKC isoform in different subcellular fractions (Fig. 1). PKC was present in both cytosolic (C) and particulate non-nuclear membrane (P) fractions in unstimulated theca cells. Treatment of the cells with increasing concentrations of LPA resulted in a dose-dependent reduction of the immunoreactive 78-kDa PKC band from cytosolic fractions (Fig. 1, fraction C). However, this decrease was not accompanied by an expected increase in the amounts of PKC protein associated with the particulate membrane fractions (Fig. 1, fraction P). There was instead a marked increase in the amounts of PKC in the fractions enriched with nuclear membranes (Fig. 1, fraction Nm).
|
Our preliminary time-course experiments revealed no nuclear translocation of PKC in time periods up to 60 min (data not shown). Within the next 120 min, the enzyme appeared slowly in the Nm fraction and was abundant there after 180 min. Maximal nuclear appearance occurred within 180–240 min (Fig. 2). No nuclear staining was observed when the cells were stimulated for longer (>12 h) time periods (Fig. 2). Immunoreactive PKC protein appears as a doublet of 78/76-kDa proteins (as shown in Fig. 1). However, in some experiments the more quickly migrating 76-kDa form was hardly detectable (Fig. 2), which can be attributed to the presence of relatively small amounts of the 76-kDa form in theca cells (below the detection limit of the assay system).
|
Having demonstrated that LPA could cause translocation of PKC to the nuclear membrane fraction, we examined whether the action of LPA could be mimicked by PMA, an activator of PKC, or blocked by rottlerin, a specific inhibitor of PKC [26, 27]. Treatment with PMA (Fig. 3, lane 2) and LPA (Fig. 3, lane 3) increased the amounts of 78-kDa PKC in fractions enriched with nuclear membranes. Addition of rottlerin inhibited the translocation of PKC induced by either PMA or LPA (Fig. 3, lanes 4 and 5). The lower control panel shows the same PVDF membrane reblotted with an antibody against nuclear protein Lamin A/C.
|
For the next set of experiments, cells were grown on chamber slides to visualize the distribution of PKC in a single cell. The presence and localization of PKC within the cell was monitored using anti-PKC antibody followed by a fluorescence-labeled secondary antibody. A diffuse cellular (cytosolic and cytoskeleton bound) staining was observed in cells treated with either medium alone or with LH (Fig. 4, A and B). After stimulation of theca cells with LPA, no staining of the cytoplasm was seen because the majority of the cellular immunoreactive PKC was localized at the perinuclear area (Fig. 4C). In cells treated with LPA plus LH, the immunoreactive PKC was found in the perinuclear area and within the intranuclear space (Fig. 4D). Absolutely no effect of LPA on the cellular distribution of cellular PKC (negative control) either in the absence (Fig. 4E) or presence (Fig. 4F) of LH was observed. Additional control micrographs (Fig. 4, H–J) provided evidence that treatment with LPA alone (Fig. 4H) or with LH plus LPA (Fig. 4I) did not produce any observable morphological changes in the theca cells (Fig. 4, compare C and D with H and I).
|
The immunohistochemical data revealed that the addition of LH affected nuclear distribution of PKC. We investigated next the redistribution of PKC in different subcellular fractions of theca cells, using the immunoblotting procedure, to determine whether LH (alone or with LPA) could influence the nuclear relocalization of PKC. Because LH alone produced only minor (Fig. 4), if any, effects on the nuclear distribution of PKC, we performed statistical analysis employing data from several independent experiments (using different animals). Comparative data from individual subfractions obtained from cells stimulated with medium alone, LH alone, LPA, or LH plus LPA are shown in Figure 5. LH alone had no significant effect on the redistribution of PKC protein to the nuclear membranes, but in the presence of LPA the amount of PKC found in fractions enriched with nuclear membranes increased dramatically. There was no translocation to the particulate non-nuclear membranes following treatment with any combination of LH plus LPA (Fig. 5).
|
To assess the relative purity of the individual cellular subfractions, the blots were rehybridized with several marker proteins specific for each of these fractions. Figure 6 shows staining for nuclear pore protein p62 and Lamin (both nuclear envelope markers) in fractions enriched with nuclear membranes. The nonnuclear particulate fractions showed the presence of gap-junctional protein connexin 43 (Fig. 6, fraction P), which was absent from other fractions. The nuclear soluble fractions (nucleocytosol) showed the presence of DNA topoisomerse (Fig. 6, fraction Nc). This nucleus-specific protein was present in the nucleocytosol only and was absent from other fractions. Cytosolic fractions showed no staining for any of the nuclear marker proteins or for the gap-junctional protein connexin (Fig. 6, fraction C). As expected from the previous results, the treatment with LH alone did not translocate PKC to the nuclear membranes or to the nucleocytosol. Addition of LPA, however, not only relocalized this PKC isoform to the nuclear membrane fractions but also shifted it further into the nucleocytosol. Although under basal conditions in some cell preparations absolutely no PKC protein was detected in the nuclear fraction, a minor signal was detected in some other preparations. This variation might be due to a variability of the cell culture material used from different individual animals. To obtain a more pure nuclear fraction we had to sacrifice the amount of protein recovered in this fraction. Because in all lanes an equal amount of protein had to be loaded, the total amount of protein loaded was limited by the sample having the lowest protein content, which was often the nuclear fraction. Because the relative abundance of PKC in any cellular fraction was low, the ability to detect this enzyme in the Western blots was consequently limited by the amount of protein loaded.
|
In the presence of LH plus LPA, large amounts of PKC were found within the intranuclear space. Therefore, we examined further whether the presence of LH enhanced the amount of PKC associated with these fractions. Using nucleocytosol fractions from cells stimulated with LPA, LH, or LPA plus LH, we observed (Fig. 7) that LH alone had no appreciable effect on the nuclear relocalization of PKC. However, addition of LH facilitated the entrance of the enzyme to the nucleus in cells stimulated with LPA plus LH (showing larger amounts of 78-kDa PKC in the soluble nuclear fractions). Similar effects were observed when cells were stimulated with PMA instead of LPA. Closer examination of the lanes stimulated with PMA revealed an additional tiny band at 40–50 kDa, which might represent the known hydrolysis product of PKC [8]. Lower control panel shows the same PVDF membrane reblotted with an antibody against DNA topoisomerase (DNA Top).
|
Western blot data indicate that only in the presence of LPA plus LH was the nuclear translocation completed. Therefore, we performed additional control experiments to reexamine the effects of LPA at the single cell level. Using DAPI as a counterstain, each microscopic field was examined carefully with two different filters. Figure 8 shows identical cells analyzed using either a rhodamine or a DAPI filter (for the same fields). Closer examination of individual cells treated with LPA revealed some cells with nuclear staining for PKC and other cells that were not affected (Fig. 8A). The vast majority of individual cells examined showed staining limited to the perinuclear area, with a small empty space (no staining) in the intranuclear space. Only in those cells stimulated with LPA plus LH was the whole nucleus stained. (Fig. 8B). Stimulation with PMA mimicked the effects of LPA (Fig. 8C). Furthermore, treatment with the PKC inhibitor rottlerin prevented nuclear translocation of PKC in response to LPA plus LH (Fig. 8D). As expected there was no redistribution of PKC under basal conditions (Fig. 8F) or in cells treated with LH alone (Fig. 8E).
|
These findings indicate that rottlerin may prevent the nuclear distribution of PKC. However, although it is considered a specific PKC inhibitor, rottlerin may also alter mitochondrial respiratory metabolism in other cell systems [28]. To determine whether rottlerin could exert such an effect in theca cells, we attempted to directly measure the cellular energy produced during mitochondrial respiration. This energy is stored as an electrochemical gradient across the mitochondrial membrane, and this process can be visualized in a living cell using fluorescence microscopy. Theca cells pretreated in the absence and presence of rottlerin were stimulated for 240 min with LPA, LH, or LPA plus LH under the experimental conditions applied earlier. Incubation of the cells for 3 h under basal conditions resulted in a loss of mitochondrial potential, observed as a fluorescent monomeric dye form. This finding indicates that standard cell culture treatment (i.e., incubation with medium) was sufficient to induce disturbed mitochondrial potentials, even in cells not treated with agonist. Pretreatment with rottlerin, however, had only a minor effect on these changes (data not shown).
If rottlerin had produced any nonspecific effects on the function of mitochondria in theca cells, there would have been a direct effect on steroid production. Therefore, we examined whether rottlerin affected progesterone production in our cell system. The cells were pretreated with medium only or with 4 µM rottlerin and then treated with LH (or with LH in combination with LPA or PMA), and progesterone accumulated in medium was measured. The addition of rottlerin had no effect on either basal or LH-stimulated progesterone production in the theca cells (Fig. 9). However, both LPA and PMA did enhance gonadotropin-stimulated steroid production. Furthermore, pretreatment of the cells with 4 µM rottlerin almost completely abolished this stimulatory effect.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
protein found in the cytosolic fraction and a concomitant translocation of this enzyme to the perinuclear areas. This effect was dose dependent, could be mimicked by PMA, and was abolished by rottlerin, an inhibitor of PKC. LPA was able to complete the nuclear localization of this PKC isoform only in the presence of LH. Functionally, this translocation is correlated with an LPA-induced upregulation of LH-stimulated progesterone production. In rodent ovary, a luteotropic function of PKC has been suggested [5]. Despite a different regulatory pattern, the data presented herein allow us to presume a similar luteotropic function for the bovine ovary. However, although LPA augments steroid production in gonadotropin-treated theca lutein cells, it has a completely different effect on luteal cells from late luteal phase (unpublished data).
The nuclear localization of PKC in the ovary may indicate its specific function within the nucleus. Nuclear PKC isoforms appear to be involved in the regulation of several other critical biological functions, such as cell differentiation or neoplastic transformation in a variety of cell systems [29, 30]. Although the isoform has not been specified, LPA is known as an activator of PKC [31]. Nuclear translocation of PKC in response to cAMP has been reported recently [32]. One possible function of this nuclear PKC could be to directly modify the transcription factors that control cell proliferation and differentiation. Overexpression of PKC has been shown to modulate the differentiation of myeloid 32D cells into mature macrophages [33]. In other cell systems, the activation of PKC is known to block the progress of cell differentiation through phosphorylation of cell-specific transcription factors [32]. Recently, PKC was shown to bind to and phosphorylate heterogeneous nuclear ribonucleoprotein, k protein (hnRNP) [34]. This phosphorylation could possibly bridge PKC to other molecular partners in the nucleus. Some other substrate proteins such as MARCKS are not only phosphorylated by PKC but also bind to its regulatory domain with high affinity [8]. Because PKC does not itself have a nuclear localization signal, hnRNP may also be responsible for the transport of PKC across the nuclear envelope [8, 35, 36].
Information on specific nuclear substrates of PKC is scarce [8]. Within the nucleus, DNA regulatory proteins such as DNA topoisomerases, DNA polymerases, and transcription factors are the most prominent PKC substrates [8, 28, 29]. Using rat luteal cell lysates, Maizels et al. [37] showed that recombinant PKC effectively catalyzed the phosphorylation of small heat shock protein (sHSP) 25/27. The evaluation of the sHSP phosphorylation status during two stages of rat pregnancy (middle and late) showed that the sHSP-27 was more highly phosphorylated in vivo in corpora lutea in the second half of pregnancy [37]. Phosphorylation of sHSP-27 was clearly enhanced in particulate membrane fractions, which also contained nuclear membranes [37]. In other cell systems, the phosphorylated dimers of sHSP-27 inhibit Fas/Apo-1-induced apoptosis [38]. However, using ovarian BG1 cancer cells, Fujiwara et al. [39] showed that phosphorylated sHSP-27 itself was redistributed to the nucleus. It is therefore tempting to suggest a possible role for this PKC substrate in regulating apoptosis. Cytosolic and mitochondrial localization of PKC can be linked to the progression of apoptosis, whereas a nuclear localization of this enzyme is associated with the inhibition of apoptosis [40, 41].
There is strong experimental evidence to support the hypothesis that PKC is involved in the progression of pregnancy and that several hormones cooperate to regulate the expression of ovarian PKC [5–7, 11, 13, 14]. In a rodent model, the activation of prolactin receptor promoted the activation of PKC [13]. Stimulation of the rat corpus luteum with prolactin causes an increase in cellular content of diacylglycerol (because of the activation of either phospholipase Cß or phospholipase D in the absence of any increase in IP3 and Ca2+), which in turn leads to the activation of PKC [13]. Because LPA is also known to stimulate both phospholipase C and phospholipase D activities [42], a similar molecular mechanism is therefore likely. In the case of bovine luteal cells, stimulation with LH may also lead to enhanced phospholipase C activity [43, 44]. Because we did not find any direct effect of LH on the translocation of PKC, this possibility was discounted. Although LH itself has no effect on the translocation of PKC, it did facilitate the entrance of this PKC isoform into the nucleus (when added with LPA). Thus, LH may activate another unknown cofactor (anchoring protein?), which then activates the nuclear entrance of the PKC enzyme. Because PKC itself does not have a nuclear localization signal, induction or activation of an additional factor appears likely. This hypothesis is supported by the time course data, which indicate that the time period necessary to complete nuclear relocalization is rather long (120–240 min). PKC binding proteins play an important role in regulating the physiological effects of PKCs [45]. Without a nuclear localization signal, the necessity for such a PKC anchoring or carrier protein is more likely. Several PKC anchoring proteins plus their activators and inhibitors have been identified, adding to the complexity of various regulatory mechanisms [10, 45]. Unfortunately, very little is currently known about proteins that may specifically bind this particulate PKC isoform [10].
PKC seems to differ from other PKC isoforms in the way it undergoes posttranslational modification, but its relevance is still not fully understood. Olivier and Parker [46] characterized different PKC forms (76 kDa and 78 kDa) from rat brain and found no differences in their specific activities and responses to diacylglycerol. Kadotani et al. [47] showed that nonphosphorylated PKC may represent a doublet of both 76- and 78-kDa proteins, whereas the phosphorylated PKC isoform has only the 78-kDa protein. Apart from the two autophosphorylation sites (on serine and threonine residues), effector-activated tyrosine phosphorylation of PKC was also demonstrated [8]. This tyrosine phosphorylation, which takes place at more than one tyrosine residue, increases the apparent molecular mass of PKC and may determine the specificity of this kinase for a given substrate [8]. Rottlerin, a PKC inhibitor, was ineffective in blocking the enzymatic activity of PKC in vitro [48]. However, in our study rottlerin influenced the cellular redistribution of PKC protein. Others have suggested that rottlerin may block the biological effects of PKC rather than its enzymatic activity [28]. Rottlerin is reported to be able to uncouple mitochondria, resulting in a reduction of intracellular ATP necessary for the enzymatic activity of the enzyme [28]. With this scenario, mitochondrial events would be necessary for nuclear translocation of the PKC protein, which is an energy requiring process. Sufficient evidence to support either of the possibilities is not yet available. In spite of having no direct inhibitory effect on the enzymatic activity of PKC, rottlerin is in a position to inhibit biological actions of PKC in various cell systems [9, 48], which could also explain its effects on PKC redistribution. In luteinized rat granulosa cells, rottlerin blocks the ability of PRL/rPL-1 to induce relaxin mRNA expression [14]. Likewise in theca cells, treatment with rottlerin inhibits both LPA- and PMA-stimulated upregulation of LH-induced steroid production. This effect might be explained by possible abrogation of mitochondrial function in response to rottlerin. However, rottlerin affects neither basal nor LH-induced steroid production, which also require intact mitochondrial function.
Although the role of ovarian LPA remains obscure, its effects on PKC may provide an important clue to its function. At the moment we can only speculate that LPA-induced nuclear localization of PKC in LH-stimulated theca cells participates in the luteotropic function. This speculation is also consistent with the overall suggested role of LPA as a survival factor [49, 50]. LH-induced phenotype stabilization (i.e., during pregnancy) and inhibition of apoptosis during this period represent a part of the ovarian rescue mechanism. LPA may be one of the factors involved in this complex process. Data provided by Hunzicker-Dunn and her colleagues clearly established that PKC has an important part to play during pregnancy in rodents [5–7, 11, 13, 14]. Although in other species hormonal patterns regulating corpus luteum function are different, key regulatory proteins such as PKC may have a similar function determined by the physiological status of the ovary. However, any comparison between artificial long-term cultures and the in vivo situation is only speculative. Contributions of other cell types and intracellular communication should not be underestimated. Elevated plasma levels of LPA are consistently detected in patients with stage 1 ovarian cancer [51, 52], and a role in neoplastic transformation in the ovary can be imagined. Further research will be necessary to resolve these issues. Our results point to a new cellular mechanism of LPA action in the ovary. This finding may open important research avenues leading to a better understanding of the compex ovarian remodeling process.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Presented in part at the 34th Annual Meeting of the Society for the Study of Reproduction, 2001. 
2 Correspondence: L.T. Budnik, Institute for Hormone and Fertility Research, University of Hamburg, Grandweg 64, D-22529 Hamburg, Germany. FAX: 4940 561908 64; budnik{at}ihf.de
Accepted: April 16, 2002.
Received: December 28, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. J Biol Chem 1992 267:17061-17068 and - and not by protein kinase C-ßII, -
, -, and -
. J Biol Chem 1993 268:20110-20115 and - to the cytosol. Implications in apoptosis. J Biol Chem 1997 272:2452-2458 in phorbol ester-induced cytochrome c release and apoptosis. J Biol Chem 2000 275:21793-21796 isoform of protein kinase C isolated from Chinese hamster ovary cells. J Biochem 1997 121:1047-1053This article has been cited by other articles:
|
|
 |
|
  G. Song, K. A. Dunlap, J. Kim, D. W. Bailey, T. E. Spencer, R. C. Burghardt, G. F. Wagner, G. A. Johnson, and F. W. Bazer Stanniocalcin 1 Is a Luminal Epithelial Marker for Implantation in Pigs Regulated by Progesterone and Estradiol Endocrinology, February 1, 2009; 150(2): 936 - 945. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Liszewska, P. Reinaud, E. Billon-Denis, O. Dubois, P. Robin, and G. Charpigny Lysophosphatidic Acid Signaling during Embryo Development in Sheep: Involvement in Prostaglandin Synthesis Endocrinology, January 1, 2009; 150(1): 422 - 434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Ye Lysophospholipid signaling in the function and pathology of the reproductive system Hum. Reprod. Update, September 1, 2008; 14(5): 519 - 536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Mancuso, H. F. Sims, X. Han, C. M. Jenkins, S. P. Guan, K. Yang, S. H. Moon, T. Pietka, N. A. Abumrad, P. H. Schlesinger, et al. Genetic Ablation of Calcium-independent Phospholipase A2{gamma} Leads to Alterations in Mitochondrial Lipid Metabolism and Function Resulting in a Deficient Mitochondrial Bioenergetic Phenotype J. Biol. Chem., November 30, 2007; 282(48): 34611 - 34622. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. T. Budnik and B. Brunswig-Spickenheier Differential effects of lysolipids on steroid synthesis in cells expressing endogenous LPA2 receptor J. Lipid Res., May 1, 2005; 46(5): 930 - 941. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. T. Budnik, B. Brunswig-Spickenheier, and A. K. Mukhopadhyay Lysophosphatidic Acid Signals through Mitogen-Activated Protein Kinase-Extracellular Signal Regulated Kinase in Ovarian Theca Cells Expressing the LPA1/edg2-Receptor: Involvement of a Nonclassical Pathway? Mol. Endocrinol., August 1, 2003; 17(8): 1593 - 1606. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
