Molecular Human Reproduction, Vol. 7, No. 5, 415-423,
May 2001
© 2001 European Society of Human Reproduction and Embryology
Ovary and oogenesis |
Distinct regulation of gene expression by prostaglandin F2
(PGF2) is associated with PGF2 resistance or susceptibility in human granulosa-luteal cells
1 Departments of Physiology and 2 Obstetrics & Gynecology, National Cheng Kung University Medical College, Tainan 701, Taiwan, Republic of China
Abstract
The effects of human chorionic gonadotrophin (HCG) and prostaglandin F2
(PGF2) on regulation of human granulosa-luteal cell (GLC) function at different stages of differentiation (day 2 versus day 8 of culture) were studied. Expression of LH receptor mRNA and biosynthesis of progesterone were HCG dependent in human GLC at all stages (n = 6, P < 0.05). Steady-state concentrations of mRNA encoding for FP (a specific high-affinity plasma membrane receptor for PGF2) were not dependent on, but were stimulated by, addition of HCG (10 IU/ml) or 8-bromo-cAMP (0.5 mmol/l) (n = 6, P < 0.05). Treatment with PGF2 (100 nmol/l) decreased FP mRNA concentration, but had no effect on LH receptor and cyclo oxygenase-2 (COX-2) expression on day 2 of cultured GLC (n = 8). As a result, the progesterone biosynthesis by GLC was not affected. On day 8, PGF2 induced FP and PGHS-2 expression and at the same time decreased LH receptor expression, resulting in inhibition of progesterone output by GLC. Our data demonstrated that early stage GLC (day 2 of culture) are resistant to PGF2-induced inhibition of progesterone synthesis but underwent further differentiation and acquired luteolytic capacity after 8 days culture in vitro. We conclude that, via distinct gene regulation at different stages of differentiation, human GLC may become resistant or susceptible to PGF2-induced luteolysis.
FP/granulosa-luteal cell/LH receptor/PGF2/PGHS-2
Introduction
Prostaglandin F2 (PGF2) plays important roles in regulation of ovarian functions in many mammalian species including humans. The luteolytic capability of PGF2 in ruminants has been clearly documented in the literature (Auletta and Flint, 1988
). In non-human primates and women, evidence has accumulated to demonstrate that PGF2 may have a luteolytic effect on corpus luteum. Infusion of PGF2 into the corpus luteum of rhesus monkey has been shown to cause a decrease in serum progesterone and menstrual bleeding (Auletta et al., 1984). Direct injection of PGF2 to mid-late stage human corpus luteum was shown to both reduce progesterone synthesis and shorten the luteal phase (Bennegard et al., 1991). An in-vitro study using human granulosa-luteal cells (GLC) retrieved from women undergoing IVF demonstrated that PGF2 inhibits LH-stimulated progesterone production (Abayasekara et al., 1993).
The action of PGF2 is mediated by binding to a specific high-affinity plasma membrane receptor termed PGF2receptor (FP). Binding of PGF2 to FP leads to increased intracellular free calcium concentrations and activation of protein kinase C (PKC). High-affinity FP has been identified on luteal tissues, especially granulosa cell-derived large steroidogenic cells (Sakamoto et al., 1995; Wiltbank et al., 1995). On the other hand, thecal cells and thecal cell-derived small luteal cells have little, if any, FP mRNA or PGF2 binding capability at any time of the entire cycle (Sakamoto et al., 1995; Wiltbank et al., 1995; Tsai et al., 1996). Expression of FP mRNA and PGF2 binding capacity have also been found in human corpus luteum and GLC (Ristimaki et al., 1997; Vaananen et al., 1998; Ottander et al., 1999). However, contradictory results about regulation of FP in human GLC have been reported (Ristimaki et al., 1997; Vaananen et al., 1998). The reason for the discrepancies in these studies is not clear but may be due to differences in culture conditions and stages of GLC being treated.
LH is the primary luteotrophic agent that plays a role in the development and/or maintenance of a fully functional corpus luteum. Pulses of LH during the luteal phase in ewes, monkeys and women have been shown to induce a concomitant increase in secretion of progesterone by corpus luteum (Backstrom et al., 1982; Filicori et al., 1984; Baird, 1992). Activation of LH receptor induces cAMP formation resulting in increased steroidogenesis by the ovary (Armstrong et al., 1964). Thus, expression of LH receptor in the corpus luteum is critical in maintaining the function and prolonging the life span of the corpus luteum. Although LH receptor has been identified in human corpus luteum as well as in in-vitro luteinized human GLC (Nishimori et al., 1995; Minegishi et al., 1997; Takao et al., 1997), regulation of LH receptors by luteotropic and/or luteolytic hormones at different stages of corpus luteum is largely unclear.
A growing body of evidence has demonstrated that, in many mammalian species including women, PGF2-induced luteal regression is stage-dependent (Rowson et al., 1972; Diehl and Day, 1974; Wright et al., 1980; Summers et al., 1985; Bennegard et al., 1991). Early phase corpus luteum is resistant to a single treatment with PGF2, while mid-late stage corpus luteum is susceptible to PGF2-induced luteolysis. The mechanisms responsible for different responsiveness to PGF2-induced luteolysis are not completely understood. In pigs, amounts of FP increase near the time of acquisition of luteolytic capacity (Gadsby et al., 1993), suggesting that FP may play a role in PGF2-induced luteal regression. In bovine corpus luteum, PGF2 has been shown to induce cyclooxygenase-2 (COX-2; also known as PGS-2) mRNA on day 11 (with luteolytic capacity) but not on day 4 (without luteolytic capacity), indicating that autoamplification of intraluteal PGF2 may be a key component of luteolytic capacity (Tsai and Wiltbank, 1998a). Other reports have demonstrated that elevated monocyte chemoattractant protein-1, which recruits the immune system to the luteolytic process, and endothelin-1, which inhibits progesterone production and induces vessel constriction, can be identified only in the mid cycle of bovine corpus luteum after PGF2 administration (Tsai et al., 1997; Levy et al., 2000). In women, to our knowledge, no similar study has been performed to elucidate the mechanisms responsible for distinct responsiveness to PGF2. Thus, the first objective of the present study was to determine whether human GLC need luteotrophic hormones (HCG in the current study) to maintain proper physiological function. The second objective was to determine if GLC would acquire luteolytic capacity after long-term culture, and if so, whether the actions of PGF2 on these two distinct stages could be evaluated.
Materials and methods
Chemicals and reagents
All chemicals used in this study, unless otherwise specified, were purchased from Sigma Chemical Company (St Louis, MO, USA). T7 RNA polymerase, M-MLV reverse transcriptase, and restriction enzymes were from Promega (Madison, WI, USA). PCR2.1TM cloning system was from Invitrogen (Carlsbad, CA, USA). Taq DNA polymerase and 1 kb DNA ladders were from Gibco/BRL (Gaithersburg, MD, USA). MagnetightTM Oligo(dT) particles were from Novagen (Madison, WI, USA).
Preparation of GLC
Human GLC were obtained from women receiving assisted reproduction treatment at The National Cheng Kung University Hospital. Patients entering the IVF/embryo transfer programme due to male infertility factors (30%), tubal ligation (35%), endometriosis (10%), or factors other than polycystic ovarian syndrome (25%) were included in this study. The cells were a by-product of the IVF/embryo transfer procedure and normally would have been discarded. They were provided for this study as coded samples, thus this research was granted an exemption from review by the university human subjects review committee. The patients received varying regimens of Metrodin (Serono, Randolph, MA, USA) and Pergonal (Serono) and received 10 000 IU of HCG 36 h prior to follicular aspiration. Approximately 1.0 ml modified human tubal fluid medium (Irvine Scientific, Santa Ana, CA, USA) containing HEPES buffer, antibiotics and heparin was added to each follicular fluid aspirate during the oocyte retrieval procedure. After oocytes and cumulus masses were removed, the follicular fluid containing granulosa cells was transported to the laboratory within 30 min for purification of GLC. Individual follicles were not distinguished as all granulosa cells from an individual were pooled, but cells from different subjects were not pooled.
Follicular fluid from each subject was divided equally into 15 ml disposable, sterile centrifuge tubes and centrifuged at 400 g for 10 min. This created a firm layer of GLC on top of a red blood cell pellet. The layers of GLC were resuspended in 2 ml Hanks' solution containing 50 µg/ml DNase I and 2 mg/ml collegenase (type IV) and combined in a sterile 50 ml centrifuge tube. The cell suspensions were shaken at 125 r.p.m. for 20 min at 37°C and then centrifuged at 400 g for 10 min at room temperature. The pellets were resuspended with Dulbecco's modified Eagle's medium/Ham's R-12 medium (DMEM/F-12) (6 ml) and layered onto 4.0 ml Ficoll-Paque (Pharmacia) in 15 ml centrifuge tubes and centrifuged at 600 g for 20 min. The cell layer was removed from each Ficoll-Paque column and washed twice with 10 ml fresh culture medium I (DMEM/F-12 supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin sulphate, 0.625 µg/ml fungizone, and 10% fetal bovine serum). Cells were finally suspended in 5 ml of culture medium I, counted on a haemacytometer, and brought to a final concentration of 1x105 cells/ml in culture medium I. Cell viability was determined by 0.04% Trypan Blue dye and 1x105 per well viable cells were plated down in 24-well culture plates. Cells attached after 16–18 h, and remaining debris was removed by washing with Hanks' solution. Low serum culture medium II [DMEM/F-12 supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin sulphate, 0.625 µg/ml fungizone, 2% fetal bovine serum, and 1x GlutaMAXTM (Gibco)] was added to each well. This was designated as day 0 of culture. Culture media were changed every other day in all experiments and stored frozen until the assay for progesterone concentrations.
Cell culture and treatments
Cells from different patients were not pooled together in most studies except in the third experiment where cells from two subjects were combined. In the first experiment, cells were treated with control medium, 10 IU/ml HCG, or 0.5 mmol/l 8-bromo-cAMP on day 0, 2, 4 or 6 and incubated for a further 24 or 48 h. In the subsequent studies, 1 IU/ml HCG was added since the result from the above study indicated that HCG and/or cAMP were necessary for the maintenance of LH receptor and progesterone production. In the second experiment, day 2 and day 8 GLC were treated with PGF2 (100 nmol/l) or control medium for various time periods (0–24 h). In the third experiment, 6x106 cells were cultured in 30 mm Petri dishes. On day 2 or day 8, cells were treated with PGF2 (100 nmol/l) or left untreated for 4, 12 or 24 h.
Preparation of native and competitor RNA
Specific primer pairs for LH receptor, FP, COX-2, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Table I) were designed to amplify a fragment of DNA from mRNA transcripts. Figure 1 illustrates the strategy for generating native and competitor RNA. Briefly, DNA fragments amplified by reverse transcription-polymerase chain reaction (RT-PCR) using specific primer pairs, were cloned into PCR cloning vectors (PCR2.1TM; Invitrogen, Madison, WI USA), amplified, and sequenced. Internal primers were designed to consist of a portion of complementary sequence to the mRNA at the 3' end of the internal primer and a sequence identical to the downstream primer at the 5' end of the internal primer. The internal primer was paired with the upstream primer to amplify a shorter fragment that also contains the downstream primer sequences. The amplified fragment was again cloned into a PCR cloning vector and a positive clone was identified. This type of clone served as the competitor plasmid for LH receptor, FP, COX-2 and GAPDH. All the plasmids containing native and competitor DNA were sequenced by ABI PrismTM terminator cycle sequencing kit (Perkin Elmer, Foster City, CA, USA) according to the manufacturer's protocol for verification of the sequences.
|
|
Plasmids containing native or competitor DNA were linearized by Hind III and transcribed in vitro using T7 RNA polymerase. Transcribed RNA were precipitated twice using 0.3 mol/l sodium acetate (pH 4.2) and 2.5 volumes of 100% ethanol after removal of DNA and protein from the solution. The concentrations of RNA were quantified by OD260 absorbance, aliquoted, and stored at –80°C. Each RNA aliquot was used only once to reduce variation due to potential degradation of RNA after freezing and thawing.
Standard curve, quantitative competitive (QC)-RT-PCR
The detailed procedure of standard curve QC-RT-PCR was described previously (Tsai and Wiltbank, 1996, 1998b). Briefly, a constant amount of competitor RNA was added into an RT master mix (50 mmol/l Tris–HCl, 75 mmol/l KCl, 3 mmol/l MgCl2, pH 8.3, 10 mmol/l dithiothreitol, 100 pmol random primer, 4 mmol/l dNTPs and 50 U MML-V reverse transcriptase). This mix was then dispensed into 0.2 ml thin wall PCR tubes and known amounts of native RNA in 2 µl of DEPC-treated water or 2 µl of unknown amounts of mRNA samples were added individually to each tube. The final volume of RT mix was 20 µl and RT was performed at 42°C for 60 min followed by heating to 95°C for 10 min and quick-chilled to 4°C in a programmable thermocycler (PTC-100; MJ Research, Watertown, MA, USA). Five µl of RT products were added to 15 µl of PCR mix (final concentration: 20 mmol/l Tris–HCl [pH 8.4 at 25°C], 50 mmol/l KCl, 1.5 mmol/l MgCl2, 0.2 mmol/l dNTPs, 0.5 U Taq polymerase, and 0.4 µmol/l of primers). This was subjected to 30 cycles of amplification (30 s denaturation at 95°C, 30 s annealing at 57°C, and 30 s elongation at 72°C) followed by final elongation at 72°C for 5 min. Ten µl of PCR products were directly separated on a 5% acrylamide gel with 1xTBE (0.09 mol/l Tris, 0.09 mol/l boric acid, 0.001 mol/l EDTA, pH 8.0) buffer at 110 V for 40 min using Mini-protein® II electrophoresis system (BioRad, Richmond, CA, USA). The gel was then stained with ethidium bromide and placed on a UV illuminator equipped with a camera connected to a Macintosh computer. The gel image was analysed using AlphaImagerTM software (Alpha Innotech Corp., San Leandro, CA, USA). A ratio was calculated for the intensity of native versus competitor bands on each lane of the gels. The logarithmic ratio of native to competitor was plotted against the logarithmic initial amounts of native to produce the standard curve (Figure 1
) and concentrations of specific mRNA transcripts were obtained by comparison to the standard curve as previously described (Tsai and Wiltbank, 1996).
Western blot for COX-2
Protein was extracted by addition of 200 µl lysis buffer (50 mmol/l Tris, 150 mmol/l NaCl, 1% Nonidt P-40, 0.5% sodium deoxycholate, 0.1% SDS, 1 µg/ml pepstatin A, and 0.1 mmol/l PMSF) into culture plates and concentrations were determined by the Lowry method (Lowry et al., 1951). Twenty-five µg of protein were loaded into each lane, separated using 10% sodium dodecyl sulphate (SDS)–polyacrylamide gel electrophoresis, and transferred onto a PVDF membrane (Millipore Co., Bedford, MA, USA). Non-specific binding was blocked by immersing the membrane in 2% BSA at 4°C overnight. The membrane was then incubated with rabbit anti-COX-2 polyclonal antibody (Cayman, Ann Arbor, MI, USA) at 1:1000 dilution for 1 h at 37°C. After washing with TBST (10 mmol/l Tris, pH 8.0, 150 mmol/l NaCl, 0.05% Tween-20) for 3x10 min, membrane was further incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG (Sigma, St Louis, MO, USA) at 1:25 000 dilution for 1 h at room temperature. The membrane was washed for 1 h with TBST and proteins were detected by ECL (Amersham, Little Chalfont, Bucks, UK). The blots were then stripped with stripping buffer (100 mmol/l 2-mercaptoethanol, 2% SDS, and 62.5 mmol/l Tris–HCl, pH 6.7) and re-used as described above except that mouse anti-ß-actin monoclonal antibody (Amersham) and HRP-conjugated goat anti-mouse IgG were used (Sigma).
Progesterone assay
Progesterone concentrations in the medium were determined by a competitive enzyme-linked immunosorbent assay procedure as previously described (Tsai and Wiltbank, 1998a). Briefly, primary antibody (mouse anti-P4 monoclonal antibody; Biostride Inc., Palo Alto, CA, USA) was added to 96-well plate precoated with goat anti-mouse antibodies (Calbiochem, San Diego, CA, USA) and incubated for 90 min at room temperature. After washing off excess primary antibody, samples were added to the plate and incubated for another 90 min at room temperature. Fifty µl of HRP-conjugated progesterone (made in our laboratory) were added into each well to compete for the primary antibody for 90 min at room temperature. The plate was then washed four times with washing buffer (20 mmol/l MOPS and 0.05% Tween-20, pH 7.2). Substrate solution (125 µl; 50 mmol/l sodium acetate, pH 4.4, 0.5 mol/l H2O2, and 20 mg/ml 3,3',5,5'-tetramethyl benzidine) was added to each well and incubated at 37°C for 15 min with shaking. Colour development was terminated by adding 50 µl of stop solution (0.5 mol/l H2SO4) to each well and optical density was determined by reading absorbance at 450 nm in an EIA plate reader. The sensitivity (80% bound) of the progesterone assay was 0.08 ng/ml and the intra- and inter-assay CV were 5.8 and 9.2% respectively.
Statistical analyses
Data were analysed through use of general linear model (GLM) of the Statistical Analysis System (SAS, 1987). The concentrations of each specific mRNA or progesterone concentrations were analysed by one-way analysis of variance followed by Duncan's multiple range test. In any given experiment, if there were no significant changes in control groups of any time points, data from these groups were combined as a single control group (0 h).
Results
Effect of HCG on LH receptor mRNA expression
Steady-state concentrations of mRNA encoding for the LH receptor was detected in freshly isolated (36 h post administration of ovulatory dose of HCG) GLC (20 ± 4 amol/µg DNA). On day 0 of culture, the amounts of LH receptor mRNA were 17 ± 5 amol/µg DNA and were not affected by treatment with HCG or 8-bromo-cAMP (Table II). Steady-state concentrations of mRNA encoding for LH receptor decreased 60–70% when GLC were cultured in basal medium containing 2% FBS but without luteotrophic hormone supplement for >2 days (Table II). Treatment of GLC with 10 IU/ml HCG or 0.5 mmol/l 8-bromo-cAMP for 24 h restored LH receptor mRNA concentrations to the pre-culture level (Table II). It was noted that on day 2 of culture, HCG did not completely restore LH mRNA concentrations. The reason for this was not clear, but restoration of the LH receptor (16 ± 5 amol/µg DNA) was observed with 48 h of treatment. The expression of LH receptor mRNA was maximal after 24 h of treatment, as evidenced by a longer treatment period (48 h) which did not increase LH receptor mRNA concentration (data not shown).
|
Effect of HCG on FP mRNA expression
Basal concentrations of FP mRNA were detected in freshly isolated (21 ± 4 amol/µg DNA) as well as cultured GLC (Table II
). In contrast with the LH receptor, FP mRNA expression was not decreased by days in culture in the absence of luteotrophic hormone or cAMP analogue. Administration of HCG or 8-bromo-cAMP caused a 2.5–4.5-fold increase in steady-state concentrations of mRNA encoding for FP on day 0, 2, 4 and 6 cultures (Table II). Steady-state concentrations of mRNA encoding for GAPDH were detected in all of the GLC examined and were not significantly different among cultures or treatment groups (average 245 ± 87, 366 ± 91 and 260 ± 50 amol/µg DNA for control, cAMP- and HCG-treated groups respectively).
Effect of HCG on progesterone production
GLC produced great amounts of progesterone after administration of the ovulatory dose of HCG in vivo. Production of progesterone was demolished when GLC were cultured in the media without exogenous HCG or cAMP (Table II). More than 90% of progesterone production was inhibited after 7 days of culture (day 6 plus 24 h) in the absence of HCG. Treatment with HCG or cAMP on any day of culture restored the progesterone production by GLC (Table II). Concentrations of progesterone measured after 48 h of HCG or cAMP treatment were almost double the concentrations measured after 24 h of treatment (data not shown).
Effect of PGF2 on LH receptor mRNA at different stages of GLC
When cultured in the continuous presence of 1 IU/ml HCG, GLC expressed greater concentrations of LH receptor as compared to that when treated with 10 IU/ml HCG after 48 h of treatment on day 0 (56 ± 5 versus 19 ± 10 amol/µg DNA). For this reason, cells were cultured in the presence of 1 IU/ml HCG for the second part of the study. Treatment of GLC with PGF2 (100 nmol/l) did not change the LH receptor mRNA concentration on day 2 (Figure 2). Steady-state concentrations of mRNA encoding for LH receptor were greater on day 8 in the presence of HCG as compared to that on day 2 (Figure 2, P < 0.05). On day 8, PGF2 reduced the steady-state concentrations of mRNA encoding for LH receptor after 12 and 24 h of treatment (Figure 2).
|
Effect of PGF2
on FP mRNA at different stages of GLCExpression of FP mRNA was similar on day 2 and day 8 cultured GLC in the presence of 1 IU/ml HCG (Figure 3
). Treatment of day 2 GLC with PGF2 decreased the steady-state concentrations of mRNA encoding for FP after 4 h (Figure 3A). In contrast, steady-state concentrations of mRNA encoding for FP on day 8 were transiently increased by PGF2 (at 0.5 and 1 h) and then returned to the pretreated level after 2 h (Figure 3B). Again, steady-state concentrations of mRNA encoding for GAPDH were not changed by treatment with PGF2 either on day 2 (control versus PGF2 199 ± 48 versus 161 ± 13 amol/µg DNA) or day 8 (control versus PGF2 207 ± 20 versus 206 ± 16 amol/mg DNA).
|
Effect of PGF2
on COX-2 mRNA and protein at different stages of GLCThe basal concentrations of mRNA encoding for COX-2 were not different between day 2 and day 8 GLC (P > 0.05). Treatment of day 2 GLC with PGF2
did not change steady-state concentrations of mRNA for COX-2 at any time point examined (Figure 4A). In contrast, COX-2 mRNA was quickly induced by PGF2 (in <1 h) before returning to the basal level (at 8 h of treatment) in day 8 GLC (P < 0.05, Figure 4B).
|
Figure 5
shows the Western blots for COX-2 and ß-actin protein. There was no substantial change (P > 0.05) in ß-actin protein in either day 2 or day 8 GLC treated with PGF2 or control medium (Figure 5A and B, lower panels). Due to variations between subjects, data were converted to `-fold' change relative to 0 h control. As seen in Figure 5C, PGF2 failed to stimulate COX-2 protein expression in day 2 GLC (P > 0.05). In contrast, in concordance with the mRNA pattern, there was a 60 and 260% increase in COX-2 protein after 4 and 12 h of PGF2 treatment in day 8 GLC respectively (Figure 5D).
|
Effect of PGF2
on progesterone production at different stages of GLCGLC produced less progesterone on day 2 as compared to day 8 in the presence of 1 IU/ml HCG (491 ± 161 and 1046 ± 258 ng/µg DNA 24 h respectively). On day 2, treatment of GLC with PGF2
did not change progesterone production (P > 0.05). On day 8, PGF2 significantly decreased the amounts of progesterone produced by GLC (Figure 6). Use of pharmacological PKC activator (PMA, 1 nmol/l) or calcium ionophore (ionomycin, 0.1 mmol/l) to treat GLC also led to inhibition of progesterone production by day 8 GLC, similar to that caused by PGF2 (Figure 6).
|
Discussion
The life span of the corpus luteum is primarily dependent on the balance of two counteracting signals, namely luteotrophic and luteolytic signals. LH, or HCG, represents the former whereas PGF2 is regarded as the latter. These two signals, through binding to their specific plasma membrane receptors on the luteal cells, activate downstream signalling cascades and subsequently regulate gene expression to antagonize each other. The current report aims to quantify the amounts of LH receptor, FP and COX-2 under the regulation of LH/HCG and/or PGF2, as well as to investigate their roles in PGF2-induced inhibition of progesterone production in human GLC at different stages. Herein, we present evidence linking distinct regulation of LH receptor, FP and COX-2 in human GLC with sensitivity of GLC to PGF2-induced decreases in progesterone production.
Our current results showed that progesterone concentrations were demolished after 2 days of GLC culture in the absence of HCG. Administration of HCG or 8-bromo-cAMP for 24 h restored progesterone production. This result indicates that GLC relied strictly on LH/HCG to maintain their luteal cell-like function. Indeed, continuous presence of 1 IU/ml HCG (one-tenth of the concentration) stimulated a greater amount of progesterone production as compared to that stimulated by 10 IU/ml HCG, but only for 24 h. In the absence of HCG, there was a 70% decrease in LH receptor transcript after 2 days culture. Treatment of GLC with HCG or cAMP analogue for 24 h significantly induced LH receptor mRNA expression, indicating that LH/HCG can up-regulate its own receptor. The induction of LH receptor transcripts by LH/HCG in GLC may increase sensitivity to LH, therefore increasing the utilization of the low concentration of LH during the luteal phase. Basal concentrations of mRNA encoding for FP were not changed throughout the entire culture period. Administration of HCG or 8-bromo-cAMP resulted in a 3-4-fold increase in steady-state concentrations of mRNA encoding for FP. This is consistent with one report (Ristimaki et al., 1997) but differs from another (Vaananen et al., 1998). In addition, the FP transcripts in the continuous presence of 1 IU/ml HCG for 8 days appeared to be the same as that in 10 IU/ml HCG-treated GLC.
The action of PGF2 in luteolysis has been the most extensively investigated role for PGs in regulating ovarian function. The results from numerous studies support the primary role for PGF2 in regression of the corpus luteum in many species (Auletta and Flint, 1988). However, it has been found that early phase corpora lutea of cows (Rowson et al., 1972), marmoset monkeys (Summers et al., 1985), pigs (Diehl and Day, 1974) or pseudopregnant rats (Wright et al., 1980) are resistant to PGF2-induced luteolysis. In studies that evaluated hundreds of women with apparently normal ovulatory cycles, the shortest luteal phases were 8 or 9 days (Lenton et al., 1984; Smith et al., 1984; Johannisson et al., 1987). This indirectly suggests that premature luteolysis may not occur before day 8 of the 14 day luteal cycle in women. A novel study by Bennegard et al. demonstrated that injection of PGF2 into the corpus luteum caused an acute decrease in serum progesterone, but only the mid phase corpus luteum (days 8–9) regressed. The early phase corpus luteum (days 3–6) recovered their progesterone concentration to the pre-injection level at 24 to 36 h after PGF2 injection (Bennegard et al., 1991). Thus, it is likely that human corpus luteum in its early phase, like other mammalian species, is PGF2 resistant, though the mechanisms are largely unknown. Our current study is the first report about distinct responses to PGF2-induced decrease in progesterone production by GLC at different culture stages and represents an excellent in-vitro model to study the stage-dependent responsiveness of corpus luteum to PGF2 in human beings.
The stage-dependent resistance to PGF2-induced luteolysis is a common scenario among mammals, but the mechanisms responsible for this insensitivity are not yet clear. The availability of PGF2 binding sites on the luteal cells has been suggested to account for the sensitivity of PGF2-induced luteolysis (Wright et al., 1980). However, in cattle, the FP transcript expression, binding capacity, and binding affinity appear to be the same between early and mid cycle corpus luteum (Wiltbank et al., 1995; Tsai et al., 1996). The FP mRNA concentration in early phase (day 2–5) human corpus luteum appeared to be less than that in mid (day 6–10) and late (day 11–14) phase corpus luteum, but only luteal cells from late phase corpus luteum responded to cloprostenol-induced inhibition of HCG-stimulated progesterone synthesis (Ottander et al., 1999). Our current data demonstrated that FP mRNA was not different in GLC with different times in culture. Thus, whether the amount of FP per se is the critical factor for determination of PGF2 sensitivity in luteal cells is debatable. Alternatively, the change of receptor after PGF2 administration may play a pivotal role in PGF2 resistance. In early stage GLC, steady-state concentrations of mRNA encoding for FP were decreased by treatment with PGF2. This receptor down-regulation may play a significant role in preventing overstimulation by its homologous ligand. In contrast, mid stage GLC not only lacked this safeguarding mechanism, but even expressed more FP after PGF2 treatment. Though the FP protein was not measured in this study, the assumption that FP mRNA was translated is highly likely given that PGF2 exerted its action in regulation of LH receptor, FP, COX-2 mRNA expression as well as progesterone output. Thus, the increase in FP mRNA may lead to a greater availability of PGF2 binding and result in more pronounced or even prolonged exposure to PGF2-mediated luteolytic signals.
The sources of PGF2 during natural luteolysis in primates may come from other tissues in addition to the uterine endometrium since hysterectomy did not prevent luteolysis in rhesus macaques (Neill et al., 1969) and women (Beling et al., 1970). Given the nature of the short half-life of PGF2 in circulation, it is generally believed that PGF2 may come from the ovary or even from within the corpus luteum. A recent study using dispersed human corpus luteum cells demonstrated that cells of late but not early or mid luteal phase were susceptible to nitric oxide-induced PGF2 production, resulting in decreased progesterone production (Friden et al., 2000). In addition, a PGF2 positive feedback loop has been identified in mid cycle corpus luteum of sheep (Tsai and Wiltbank, 1997), cows (Tsai and Wiltbank, 1998a), and pigs (Diaz et al., 2000). Interestingly, this intraluteal PGF2 autoamplification system was identified in mid cycle corpus luteum with luteolytic capacity, but not in early cycle corpus luteum without luteolytic capacity (Tsai and Wiltbank 1998a). COX-2 regulates the first committed step in PG biosynthesis and is believed to be the target isoform for acute regulation of PG synthesis. In this study, we used cultured human GLC and demonstrated that PGF2 induced 2- and 3-fold increases in COX-2 mRNA and protein respectively in day 8, but not in day 2, GLC. This is the first report to demonstrate that similar PGF2 autoamplification positive feedback is present specifically in human luteal cells with luteolytic capacity, as seen in the decrease in progesterone production. To put our result in a physiological perspective, the intraluteal production of PGF2 would maintain a high PGF2 concentration which may be critical for initiation and completion of luteolysis, especially in human and non-human primates where the uterus is not a necessity for triggering luteolysis.
In addition to luteolytic signals, luteotrophic signals also play important roles in controlling luteal function and life span. Here we provide evidence that PGF2 caused distinct regulation of the LH receptor in the early and mid stage human GLC that may be related to stage-dependent luteolysis, as observed by the decrease in progesterone production. In early stage GLC, PGF2 had no effect on LH receptor mRNA expression, but it decreased steady-state concentrations of mRNA encoding for LH receptor in mid stage GLC. Since the LH receptor in the early stage GLC was not affected by treatment with PGF2, the cell can still receive luteotrophic signal and thus antagonize the luteolytic effect of PGF2. In addition, FP was downregulated in the early stage and no PGF2 autoamplification was occurring, thus preventing GLC from over-stimulation by PGF2. Therefore the luteotrophic signal could out-compete the luteolytic signal and GLC could retain its function as evidenced by no decrease in progesterone production. In contrast, the LH receptor was down-regulated and FP was up-regulated by PGF2 in mid stage GLC when intraluteal production of PGF2, as indicated by induction of COX-2, would occur. This change could increase the local concentration of the luteolysin PGF2 and its receptor, FP, and at the same time could reduce the capability of GLC to respond to the luteotrophic signal, due to a decrease in LH receptor. As a result, the luteolytic signal at this stage is stronger than the luteotrophic signal and caused GLC to forfeit its function. Further studies are needed to characterize mechanisms responsible for such distinct responses at the signal transduction and gene transcription level.
Acknowledgements
This work was supported by grants from National Science Council of Taiwan, ROC (grant no. NSC89-2320-B-006-064 to S.J.T. and NSC89-2314-B-006–103 to M.H.W.).
Notes
3 To whom correspondence should be addressed. E-mail: Seantsai{at}mail.ncku.edu.tw 
References
Abayasekara, D.R.E., Michael, A.E., Webley, G.E. et al. (1993) Mode of action of prostaglandin F2 in human luteinized granulosa cells: role of protein kinase C. Mol. Cell. Endocrinol, 97, 81–91.[Web of Science][Medline]
Armstrong, D.T., O'Brien, J. and Greep, R.O. (1964) Effects of luteinizing hormone on progestin biosynthesis in the luteinized rat ovary. Endocrinology, 75, 488–500.
Auletta, F.J. and Flint, A.P. (1988) Mechanisms controlling corpus luteum function in sheep, cows, nonhuman primates, and women especially in relation to the time of luteolysis. Endocr. Rev., 9, 88–105.
Auletta, F.J., Kamps, D.L., Pories, S. et al. (1984) An intra-corpus luteum site for the luteolytic action of prostaglandin F2 alpha in the rhesus monkey. Prostaglandins, 27, 285–298.[Web of Science][Medline]
Backstrom, C.J., McNeilly, A.S., Leask, R.M. et al. (1982) Pulsatile secretion of LH, FSH, prolactin, oestradiol and progesterone during the menstrual cycle. Clin. Endocrinol., 17, 29–42.[Medline]
Baird, D.T. (1992) Luteotrophic control of the corpus luteum. Anim. Reprod. Sci., 28, 95–102.[Web of Science]
Beling, C.G., Marcus, S.L. and Markham, S.M. (1970) Functional activity of the corpus luteum following hysterectomy. J. Clin. Endocrinol. Metab., 30, 30–39.
Bennegard, B., Hahlin, M., Wennberg, E. et al. (1991) Local luteolytic effect of prostaglandin F2 in the human corpus luteum. Fertil. Steril., 56, 1070–1076.[Web of Science][Medline]
Diaz, F.J., Crenshaw, T.D. and Wiltbank, M.C. (2000) Prostaglandin F2 induces distinct physiological responses in porcine corpora lutea after acquisition of luteolytic capacity. Biol. Reprod., 63, 1504–1512.
Diehl, J.R. and Day, B.N. (1974) Effect of prostglandin F2 on luteal function in swine. J. Anim. Sci., 39, 392–396.
Filicori, M., Bulter, J.P. and Crowley, W.F. (1984) Neuroendocrine regulation of the corpus luteum in the human: evidence for pulsatile progesterone secretion. J. Clin. Invest., 73, 1638–1647.
Friden, B.E., Runesson, E., Hahlin, M. et al. (2000) Evidence for nitric oxide acting as a luteolytic factor in the human corpus luteum. Mol. Hum. Reprod., 6, 397–403.
Gadsby, J.E., Lovdal, J.A., Britt, J.H. et al. (1993) Prostaglandin F2 receptor concentrations in corpora lutea of cycling, pregnant, and pseudopregnant pigs. Biol. Reprod., 49, 604–608.[Abstract]
Johannisson, E., Landgren, B.-M., Rohr, H.P. et al. (1987) Endometrial morphology and peripheral hormone levels in women with regular menstrual cycles. Fertil. Steril., 48, 401–408.[Web of Science][Medline]
Lenton, E.A., Landgren, B.-M. and Sexton, L. (1984) Normal variation in the length of the luteal phase of the menstrual cycle: identification of the short luteal phase. Br. J. Obstet. Gynaecol., 91, 685–689.[Web of Science][Medline]
Levy, N., Kobayashi, S., Roth, Z. et al. (2000) Administration of prostaglandin F2 during the early bovine luteal phase does not alter the expression of ET-1 and its type A receptor: a possible cause for corpus luteum refractoriness. Biol. Reprod., 63, 377–382.
Lowry, D.H., Rosebrough, N.J., Farr, A.L. et al. (1951) Protein measurement with the folin phenol reagent. J. Biol. Chem., 193, 265–275.
Minegishi, T., Tano, M., Abe, Y. et al. (1997) Expression of luteinizing hormone/human chorionic gonadotrophin (LH/HCG) receptor mRNA in the human ovary. Mol. Hum. Reprod., 3, 101–107.
Neill, J.D., Johansson, E.D. and Knobil, E. (1969) Failure of hysterectomy to influence the normal pattern of cyclic progesterone secretion in the rhesus monkey. Endocrinology, 84, 464–465.
Nishimori, K., Dunkel, L., Hsueh, A.J. et al. (1995) Expression of luteinizing hormone and chorionic gonadotropin receptor messenger ribonucleic acid in human corpora lutea during menstrual cycle and pregnancy. J. Clin. Endocrinol. Metab., 80, 1444–1448.[Abstract]
Ottander, U., Leung, C.H.B. and Olofsson, J.I. (1999) Functional evidence for divergent receptor activation mechanisms of luteotrophic and luteolytic events in the human corpus luteum. Mol. Hum. Reprod., 5, 391–395.
Ristimaki, A., Jaatinen, R. and Ritvos, O. (1997) Regulation of prostaglandin F2 receptor expression in cultured human granulosa-luteal cells. Endocrinology, 138, 191–195.
Rowson, L.E., Tervit, R. and Brand, A. (1972) The use of prostaglandins for synchronization of oestrus in cattle. J. Reprod. Fertil., 29, 145.[Medline]
Sakamoto, K., Miwa, K., Ezashi, T. et al. (1995) Expression of mRNA encoding the prostaglandin F2 receptor in bovine corpora lutea throughout the oestrous cycle and pregnancy. J. Reprod. Fertil., 103, 99–105.
SAS (1987) SAS/STAT User's Guide, Version 6, vol. 2, 4th edn. SAS Institute, Inc., Cary, NC.
Smith, S.K., Lental, E.A., Landgren, E.M. et al. (1984) The short luteal phase and infertility. Br. J. Obstet. Gynaecol., 91, 1120–1121.[Web of Science][Medline]
Summers, P.M., Wennink, C.J. and Hodges, J.K. (1985) Cloprostenol-induced luteolysis in the marmoset monkey (Callithrix jacchus). J. Reprod. Fertil., 73, 133–138.
Takao, Y., Honda, T., Ueda, M. et al. (1997) Immunohistochemical localization of the LH/HCG receptor in human ovary: HCG enhances cell surface expression of LH/HCG receptor on luteinizing granulosa cells in vitro. Mol. Hum. Reprod., 3, 569–578.
Tsai, S.-J. and Wiltbank, M.C. (1996) Quantification of mRNA using competitive RT-PCR with standard-curve methodology. BioTechniques, 21, 862–866.[Web of Science][Medline]
Tsai, S.-J. and Wiltbank, M.C. (1997) Prostaglandin F2 induces expression of prostaglandin G/H synthase-2 in the ovine corpus luteum: A positive feedback loop during luteolysis. Biol. Reprod., 57, 1016–1022.[Abstract]
Tsai, S.-J., Wiltbank, M.C. and Bodensteiner, K.J. (1996) Distinct mechanisms regulate induction of messenger ribonucleic acid for prostaglandin (PG) G/H synthase-2, PGE (EP3) receptor, and PGF2 receptor in bovine preovulatory follicles. Endocrinology, 137, 3348–3355.[Abstract]
Tsai, S.-J., Juengel, J.L. and Wiltbank, M.C. (1997) Hormonal regulation of monocyte chemoattractant protein -1 messenger ribonucleic acid expression in corpora lutea. Endocrinology, 138, 4517–4520.
Tsai, S.J. and Wiltbank, M.C. (1998a) Prostaglandin F2 regulates distinct physiological changes in early and mid-cycle bovine corpora lutea. Biol. Reprod., 58, 346–352.
Tsai, S.J. and Wiltbank, M.C. (1998b) Standard curve quantitative competitive RT-PCR (SC-QC-RT-PCR): A simple method to quantify absolute concentration of mRNA from limited amounts of sample. In Siebert, P.D. and Larrick, J.W. (eds), Gene Cloning and Analysis by RT-PCR. Eaton, Palo Alto, CA, pp. 91–101.
Vaananen, J.E., Vaananen, C.C., Lee, S. et al. (1998) Regulation of prostaglandin F2 receptor mRNA in human granulosa-luteal cells by human chorionic gonadotrophin and prostaglandin F. Endocrine, 8, 261–267.[Web of Science][Medline]
Wiltbank, M.C., Shiao, T.F., Bergfelt, D.R. et al. (1995) Prostaglandin F2 receptors in the early bovine corpus luteum. Biol. Reprod., 52, 74–78.[Abstract]
Wright, K., Pang, C.Y. and Behrman, H.R. (1980) Luteal membrane binding of prostaglandin F2 and sensitivity of corpora lutea to PGF2-induced luteolysis in pseudopregnant rats. Endocrinology, 106, 1333–1337.
Submitted on December 20, 2000; accepted on March 2, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C.-C. Hsu, C.-W. Lu, B.-M. Huang, M.-H. Wu, and S.-J. Tsai Cyclic Adenosine 3',5'-Monophosphate Response Element-Binding Protein and CCAAT/Enhancer-Binding Protein Mediate Prostaglandin E2-Induced Steroidogenic Acute Regulatory Protein Expression in Endometriotic Stromal Cells Am. J. Pathol., August 1, 2008; 173(2): 433 - 441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Chandras, T E Harris, A L. Bernal, D R E Abayasekara, and A E Michael PTGER1 and PTGER2 receptors mediate regulation of progesterone synthesis and type 1 11{beta}-hydroxysteroid dehydrogenase activity by prostaglandin E2 in human granulosa lutein cells J. Endocrinol., September 1, 2007; 194(3): 595 - 602. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-H. Wu, K.-F. Chen, S.-C. Lin, C.-W. Lgu, and S.-J. Tsai Aberrant Expression of Leptin in Human Endometriotic Stromal Cells Is Induced by Elevated Levels of Hypoxia Inducible Factor-1{alpha} Am. J. Pathol., February 1, 2007; 170(2): 590 - 598. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-H. Wu, Y. Shoji, M.-C. Wu, P.-C. Chuang, C.-C. Lin, M.-F. Huang, and S.-J. Tsai Suppression of Matrix Metalloproteinase-9 by Prostaglandin E2 in Peritoneal Macrophage Is Associated with Severity of Endometriosis Am. J. Pathol., October 1, 2005; 167(4): 1061 - 1069. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-J. Tsai, S.-J. Lin, Y.-M. Cheng, H.-M. Chen, and L.-Y. C. Wing Expression and Functional Analysis of Pituitary Tumor Transforming Growth Factor-1 in Uterine Leiomyomas J. Clin. Endocrinol. Metab., June 1, 2005; 90(6): 3715 - 3723. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-H. Wu, C.-A. Wang, C.-C. Lin, L.-C. Chen, W.-C. Chang, and S.-J. Tsai Distinct Regulation of Cyclooxygenase-2 by Interleukin-1{beta} in Normal and Endometriotic Stromal Cells J. Clin. Endocrinol. Metab., January 1, 2005; 90(1): 286 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.-M. Huang, K.-Y. Hsiao, P.-C. Chuang, M.-H. Wu, H.-A. Pan, and S.-J. Tsai Upregulation of Steroidogenic Enzymes and Ovarian 17{beta}-Estradiol in Human Granulosa-Lutein Cells by Cordyceps sinensis Mycelium Biol Reprod, May 1, 2004; 70(5): 1358 - 1364. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L.-Y. C. Wing, P.-C. Chuang, M.-H. Wu, H.-M. Chen, and S.-J. Tsai Expression and Mitogenic Effect of Fibroblast Growth Factor-9 in Human Endometriotic Implant Is Regulated by Aberrant Production of Estrogen J. Clin. Endocrinol. Metab., November 1, 2003; 88(11): 5547 - 5554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. S. Sun, K.-Y. Hsiao, C.-C. Hsu, M.-H. Wu, and S.-J. Tsai Transactivation of Steroidogenic Acute Regulatory Protein in Human Endometriotic Stromal Cells Is Mediated by the Prostaglandin EP2 Receptor Endocrinology, September 1, 2003; 144(9): 3934 - 3942. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-H. Wu, H.S. Sun, C.-C. Lin, K.-Y. Hsiao, P.-C. Chuang, H.-A. Pan, and S.-J. Tsai Distinct mechanisms regulate cyclooxygenase-1 and -2 in peritoneal macrophages of women with and without endometriosis Mol. Hum. Reprod., December 1, 2002; 8(12): 1103 - 1110. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-J. Tsai, M.-H. Wu, H.-M. Chen, P.-C. Chuang, and L.-Y. C. Wing Fibroblast Growth Factor-9 Is an Endometrial Stromal Growth Factor Endocrinology, July 1, 2002; 143(7): 2715 - 2721. [Abstract] [Full Text] [PDF] |