| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RESEARCH COMMUNICATION |
a Reproductive Physiopathology Group, Departments of Obstetrics and Gynaecology, Birmingham Women's Hospital, University of Birmingham, Edgbaston, Birmingham, B15 2TG, U.K.
b Immunology Division, Faculty of Medicine, Université de Sherbrooke, Québec, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: PAF receptor • uterus • nitric oxide • menstruation • implantation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Actions of PAF are mediated by a specific PAF receptor (PAF-R). Molecular cloning of the PAF-R from a human leukocyte cDNA library has identified a seven-transmembrane spanning, G-protein-linked receptor superfamily that encodes a 342 amino acid protein with a calculated molecular mass of approximately 39 kDa (6). The human PAF-R gene contains no introns in its coding region and maps to chromosome 1 (7, 8). Although autoradiographic studies show binding of [3H]PAF to rabbit endometrial epithelium and to the embryonic disc of day 6 rabbit blastocysts (9), and reverse transcription-polymerase chain reaction (RT-PCR) demonstrated PAF-R mRNA in proliferative and secretory endometrium (10), the exact location of the receptors in human endometrium remains to be determined. This is especially important since the responsiveness of the endometrium to PAF appears to be confined to the secretory phase of the menstrual cycle, as demonstrated by PAF-stimulated prostaglandin E2 (PGE2) release (4), PAF-mediated activation of phospholipase C (11), and phospholipase D (12) activity in human secretory endometrium. Moreover, there is potentiation of PAF-evoked phospholipase D activity by 17ß-estradiol in the human endometrial cell line (13).
We provide the first evidence of PAF-R localization in uterine tissue and quantitative changes in mRNA encoding the PAF-R throughout the menstrual cycle, indicating sex steroid regulation of PAF-R in vivo. We also demonstrate that PAF may modulate neovascularization through the expression of vascular endothelial growth factor (VEGF) in human endometrium and that the potent vasodilatory properties of PAF may relate to its ability to stimulate the release of nitric oxide (NO), the formation of which is achieved by the conversation of L-arginine to L-citrulline by NO synthase (NOS) in the presence of molecular oxygen (14).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tissue collection
Endometrial tissue was rinsed in ice-cold phosphate-buffered saline (PBS) on ice to remove excess blood. Tissue was either immediately immersed in 4% formaldehyde and processed for paraffin wax embedding, rapidly frozen in liquid N2 for RNA and protein extraction, or rapidly frozen over dry ice, wrapped in Parafilm to prevent dehydration, and stored at -80°C until sectioned for in situ hybridization studies. For functional studies, endometrium was collected in sterile Hanks' balanced salt solution (HBSS, Gibco Ltd., Uxbridge, U.K.) and transported on ice to the laboratory.
Tissue explant procedure and cell line conditions
Endometrial biopsies (1 mm3 size) were placed on square sterile capillary matting in a multiwell incubating plate as previously described (11). Tissue was preincubated in Dulbecco's modified Eagle's medium (DMEM, Gibco Ltd., Uxbridge, U.K.) supplemented with Hepes buffer (20 mM; Gibco Ltd., Uxbridge, U.K.), L-glutamine (10 mM; Gibco Ltd., Uxbridge, U.K.), glucose (250 mM; Sigma Chemical Co. Ltd., Poole, Dorset, U.K.), and 0.1% bovine serum albumin (BSA, Sigma Chemical Co. Ltd., Poole, Dorset, U.K.) for 1 h at 37°C in 95% O2 and 5% CO2 at 95% humidity prior to agonist stimulation.
A spontaneously transformed human endometrial epithelial cell line (HES) generated by repeated passaging of cells obtained by routine dilatation and curettage was kindly provided by Dr. D. A. Kniss, Department of Obstetrics and Gynecology, Ohio State University, Columbus, Ohio. HES cells were maintained in 25 cm2 flasks in medium 199 (M199; Gibco Ltd., Uxbridge, U.K.) containing 10% (v/v) fetal calf serum (FCS; Gibco Ltd., Uxbridge, U.K.), 1% L-glutamine (Gibco Ltd., Uxbridge, U.K.), and 1% antibiotic mixture (10,000 U penicillin and 10 µg streptomycin; Gibco Ltd., Uxbridge, U.K.) at 37°C in 95% O2 and 5% CO2 at 95% humidity. Prior to agonist stimulation, growth medium was removed and subconfluent (85%) monolayers were serum starved for 24 h in serum-free M199 supplemented with 0.2% BSA.
Immunocytochemistry
Serial 4 µm sections of formalin-fixed, paraffin-embedded tissue were used for immunohistochemistry as previously described in detail (16). Briefly, sections were deparaffinized, hydrated, and endogenous peroxidase activity was quenched by the addition of 0.3% (v/v) of hydrogen peroxide for 10 min. Nonspecific protein binding sites were blocked with 10% normal goat serum in PBS for 30 min at room temperature in a humidified chamber. Sections were incubated with rabbit polyclonal anti-PAF-R antibody reported by Müller et al. (17) at 1:200 dilution in 10% goat serum and 0.3% Triton X-100 for 60 min at room temperature. Subsequently, the sections were incubated with a goat anti-rabbit IgG biotinylated secondary antibody and the localization of PAF-R was detected using a streptavidin-biotyn immunoperoxidase kit for rabbit IgG (DAKO Ltd. Bucks, U.K.). Diaminobenzidine was used as the chromogen. The immunostained sections were conterstained with Mayer's hematoxylin. Specificity of the immunoreactions was determined by comparing the reactivity of the antibodies with that of 10% goat serum in PBS in the absence of the primary antibody. Hematoxylin and eosin sections were dated, based on the criteria of Fox and Buckley (15).
The specificity of the anti-PAF-R antibody was confirmed by the fact that PAF-R transfected CHO cells showed labeling whereas untransfected cells did not. Western blots with anti-PAF-R do not work reliably; the receptor is probably too hydrophobic to migrate adequately on gels.
Preparation of PAF-R probe
Total RNA from endometrium was prepared using RNAzol and quantified spectrophotometrically at 260 nm. Ethidium bromide staining confirmed integrity and the quality of the RNA was confirmed on 1% agarose-formaldehyde gel electrophoresis. cDNA was synthesized from total RNA using the Superscript Preamplification System (GibcoBRL Life Technologies, Paisley, Scotland). RT-PCR was performed as described previously (10) and the RT-PCR product of 1.3 Kb was subcloned into the BamHI and HindIII sites of pBluescript KS. Sequencing analysis was performed on the PCR fragment to confirm that it encoded the human PAF-R.
In situ hybridization
Frozen tissues was surrounded in embedding medium (Tissue-Teck O. C. T. Compound; Agag Scientific, Stanstead, Essex, U.K.) before 10 µm sections were cut and thaw mounted onto poly-L-lysine-coated glass slides for in situ hybridization studies. Sections were stored (for less than 2 wk) and dessicated at -80°C until use. For transcription of the antisense or sense RNA probes, the plasmid-containing RT-PCR product was linearized with HindIII or BamHI and the transcripts were generated by [3535S]UTP (1500 Ci mmol/l, Dupont NEN, Boston, Mass.) incorporation using the T7 or T3 RNA polymerases, respectively. The probes thus generated were single-stranded RNA probes and had a specific activity of 1 x 107 dpm/µg plasmid template. In situ hybridization was performed as previously described (18). Pretreated sections were hybridized with either sense or antisense probes in hybridization buffer at a final concentration of 1 x 106 dpm/ml. Hybridization was carried out at 50°C for 16 h. Two adjacent endometrial sections from each sample were hybridized with either sense or antisense probe during the same experiment. After serial washing, the slides were dipped in photographic emulsion and stored in a lightproof box at 4°C for 5 wk. The sections were then developed and counterstained in Mayer's hematoxylin.
Semiquantitative RT-PCR for PAF-R
cDNA was synthesized from total RNA. For initial amplification by PCR, 2 µl of six RT products from throughout the menstrual cycle were amplified with 2.5 units of Taq DNA polymerase and 20 µM of each mouse ß-actin primer (5' ß-actin: 5' GTTACCAACTGGGACGACA 3'; 3' ß-actin: 5' TGGCCATCTCCTGCTCGAA 3') in 100 µl of reaction mix containing 500 mM KCl, 200 mM Tris-HCl, and 2 mM MgCl2 as follows: 94°C, 30s; 50°C, 40s; 72°C, 60s. Reactions were temporarily halted after 20–30 cycles by cooling to 4°C and 10 µl of PCR product was removed from each tube. Reactions were then restarted by heating to 94°C and amplification was allowed to proceed for three additional cycles before removal of a another 10 µl of product. This was continued until each PCR reaction had been sampled six times. All products (10 µl; six per initial RT product) were then analyzed by 2% agarose gel electrophoresis. Dilutions of RT products were made where necessary and the amplification procedure was repeated until all samples were standardized for ß-actin content. After standardization, 2 µl of appropriately diluted RT products were amplified with five units of Taq DNA polymerase and 10 µM of each human PAF-R primer (5' PAF-R: 5' TCTTCTGCAACCTGGTCATCA 3' [608 bp to 628 bp]; 3'PAF-R: 5' GATCTGGTTGAATGGCACAAC 3' [985 bp to 1005 bp]) in 100 µl of reaction mix containing 500 mM KCl, 200 mM Tris-HCl, and 1.5 mM MgCl2, as follows: 95°C, 30 s; 60°C, 50 s; 72°C, 60 s. Sampling was initiated two cycles after the initial sampling cycle for ß-actin and reactions were temporarily halted every four cycles for sampling purposes. All products were then analyzed by 2% agarose gel electrophoresis to obtain the PAF-R mRNA profile throughout the menstrual cycle. The entire procedure was repeated three more times.
RNAse protection assay
A shorter PAF-R cDNA template of 420 base pairs was obtained by restriction digestion of the original plasmid containing the insert with Blp1. The transcript was generated using T7 polymerase, which hybridizes with human total RNA to yield a 420 base pair protected fragment of human PAF-R mRNA. The pTRI-ß-actin control template (Ambion, Whitney, Oxon) contained a 247 base pair fragment of human cytoplasmic actin (ß-actin) gene. The TRIPLE-script vector was transcribed to a 298 base antisense RNA probe using T3 polymerase, which generated a 247 base pair protected fragment of human ß-actin mRNA. This was used as a constitutively expressed gene product to control for possible differences in RNA amounts between samples in order to allow accurate quantitation of PAF-R mRNA abundance between RNA samples.
RNAse protection assay was carried out on an endometrial sample throughout the cycle as described previously (19). Antisense RNA was transcribed from each plasmid using 
[32P]-CTP (Dupont NEN) to specific activities of 0.4–1.2 x 109 cpm/µg. Briefly, sample RNA (20 µg) was mixed with [32P]-labeled riboprobe (0.5–1.0 x 105 cpm), denatured at 85 ± 5°C for 5 min, and incubated overnight at 45°C. Nonhybridized RNA was removed by the addition of RNAse digestion; thereafter, RNAses were inactivated, and RNA was extracted and precipitated. RNA pellets were resuspended in loading buffer; samples were denatured and electrophoresed on 4% polyacrylamide/8 mol/l urea gels. The gels were washed in buffer containing 10% methanol and 10% acetic acid before transfer to Whatman 3 mm paper. The gels were covered with saran wrap and dried under vacuum on a gel drier (BioRad Gel Dryer 583; BioRad Laboratories, Inc. Hertfordshire, U.K.) at 80°C for 30–60 min. Autoradiography was performed using X-ray film (Kodak Biomax MR) with intensifying screens at -70°C for 2 to 6 days. Quantitation of autoradiographic signals was performed by densitometric analysis, using densitometry computer software (Gelbase). Transcripts detected were assigned relative densitometric units and the ratio of PAF-R/actin was calculated for each sample. Negative controls included 20 µg of yeast tRNA in two reaction tubes. Control reaction 1 included radiolabeled probes in the absence of RNAse treatment, which should yield the full-length probes. Control reaction 2 was treated in the same manner as the RNA samples, thereby testing for nonspecific hybridization of radiolabeled probes.
The VEGF antisense control template was formulated from a 600 base pair cDNA (EcoR1) and cloned into pBluescript SK+. Transcripts were generated using T3 polymerase, which hybridizes with human total RNA to yield a 570 base pair protected fragment of human VEGF mRNA. ß-Actin was used as a constitutively expressed gene product for comparison of VEGF mRNA abundance between RNA samples. Ribonuclease protection analyses were performed as described above.
Measurement of NO release by chemiluminesence
For concentration dependence experiments, increasing concentrations of PAF-acether (De-hydro-PAF [C18]; Sigma Chemical Co. Ltd., Poole, Dorset, U.K.) in serum-free M199 buffer containing 0.2% BSA were added to endometrial tissue explants from both proliferative and secretory phases of the menstrual cycle. For experiments with PAF or vechicle in the presence of the PAF-R antagonist, WEB 2170 (20), or the NO synthesis inhibitor NG-monomethyl-L-arginine (L-NMMA), tissues were pretreated with the antagonists for 1 h and then stimulated in a final volume of 1.0 ml at 37°C for a another 2 h. A similar procedure was adopted for experiments with low Ca2+ (150 nM) HBSS and protein kinase C inhibitor Ro 31–8220. Reactions were terminated by removal of the supernatant, which was subsequently stored at -80°C for NO analysis. Levels of NO were measured in the gas phase using a Sievers NOA 270B chemiluminescence analyzer as described previously (21). The total amount of NO present in conditioned media was expressed as picomoles of NO/mg protein. Protein was extracted from endometrial tissue explants in PBS with 5 M sodium hydroxide and 0.1% sodium dodecyl sulfate (SDS) at 85°C for 1 h.
Western blot analysis
After exposure of cells to either PAF (0.1 nM–1 µM) or vehicle, the conditioned medium was removed and stored at -70°C. The cells were washed with ice-cold PBS and transferred to prefrozen Eppendorfs. After centrifugation (13,000 rpm, 4°C, 10 min), the supernatant was discarded and the pellet was frozen in liquid nitrogen for 5 min.
Samples were then allowed to thaw on ice in 100 µl ice-cold high salt buffer (KCl 0.4 M, Hepes pH 7.4 20 mM, DTT 1 mM, glycerol 20%, bacitracin 0.5 mg/ml, PMSF 40 µg/ml, pepstatin 5 µg/ml, leupeptin 5 µg/ml) prior to protein estimation. Protein extraction from the medium was performed by ethanol precipitation. Media was mixed with three volumes (15 ml) of ice-cold ethanol and stored at -70°C for 24 h. After centrifugation (6000 rpm, 4°C, 10 min), the supernatant was discarded and pellets were resuspended in PBS. Protein concentration was determined using the Bio-Rad protein assay (Bio-Rad Laboratories Inc., Herts, U.K.), with BSA as standard.
The nonradioactive enhanced chemuluminascence Western blotting system (Amersham, Bucks, U.K.) was used to detect VEGF in cell lysates and conditioned medium as described previously (19). Briefly, 50 µg total protein was separated by electrophoresis on a polyacrylamide gel consisting of a stacking gel overlaying the separating gel. The samples were loaded in a final volume of 25 µl with sample buffer containing 0.002% bromophenol blue and electrophoresed at 50 V for approximately 2 h. A kaleidoscope protein marker (Bio-Rad Laboratories, Richmond, Calif.) was run alongside the samples. This protein marker is a protein standard calibrated molecular weight and prestained, which allows accurate molecular weight determination of unknown protein. After electrophoresis, the protein was transferred onto a nitrocellulose membrane in a cooling system (10°C) overnight at 36 V. The filter was blocked to reduce nonspecific binding of the antibody using Tris-buffered saline-Tween (TBS-T) containing 10% Marvel and 2.5% BSA for 4 h. After washing in TBS-T, the membrane was incubated overnight at 4°C with the VEGF antibody. The primary rabbit polyclonal anti-VEGF antibody raised against a 20 amino acid peptide corresponding to residues 1 to 20, mapping at the amino terminus of human VEGF, was used at a dilution of 1:1000. The filter was washed and incubated with the secondary anti-rabbit antibody for 1.5 h at room temperature. After a final wash in TBS-T, the filter was incubated for 1 min at room temperature in detection reagent, immediately wrapped in saran wrap, and exposed for periods of 30 s, 1 min, 5 min, and 10 min to X-ray film.
Tyrosine phosphorylation
Confluent monolayers of endometrial glandual epithelial cells in 25 cm2 flasks were incubated with either PAF (1 µM), VEGF (1 ng/ml), or vehicle in a final volume of 5 ml of serum-free M199 supplemented with 0.2% BSA. Reactions were terminated at the appropriate time (1, 5, 15, or 30 min after stimulation) by rapid aspiration and the cells were washed twice with ice-cold PBS. Modified radioimmunoprecipitation assay buffer (0.5 ml; RIPA: 50 mM Tris-HCL, pH 7.4, 1% NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EGTA, 1 mM PMSF, 1 µg/ml aprotinin, leupeptin, and pepstatin, 1 mM activated Na3VO4, and 1 mM NaF) was added and the flasks were rocked on an orbital shaker at 4°C for 15 min. Cell lysates were collected and centrifuged (13,000 rpm, 4°C, 10 min). Total protein (25 µg) was electrophoresed in a 10% SDS-polyacrylamide minigel at 50 V and proteins were electroblotted onto nitrocellulose for 3 h at 50 V. Blots were blocked for 2 h at room temperature in TBS-T containing 1% rabbit serum. After washing, the membranes were incubated overnight at 4°C with an anti-pTyr mAb (PY20: Affiniti, Exeter, U.K.) diluted 1:2000. Blots were washed and incubated with a secondary peroxidase-labeled anti-mouse antibody diluted 1:3000 for 1 h at room temperature. After a final wash in TBS-T, blots were incubated for 1 min at room temperature in detection reagent, immediately wrapped in saran wrap, and exposed to X-ray film.
Statistical analysis
Statistical analysis was performed using Student's paired t test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Fig. 1A).
In mid-proliferative endometrium (MP, n=6), however, staining was confined largely to endometrial glands with relatively no staining of stromal cells ( Fig. 1B). The endometrial blood vessels also showed little or no staining in the MP phase ( Fig. 1B). During the late proliferative phase (LP, n=4), the intensity of PAF-R immunostaining varied from light to moderate in the endometrial glands whereas the endometrial stroma appeared negative ( Fig. 1C). In mid-secretory (MS, n=5) endometrium, strong to moderate staining was observed in the endometrial glands and light staining was also detected in the cytoplasm of vascular endothelial cells in endometrial blood vessels. Negligible immunostaining was seen in the endometrial stromal cells ( Fig. 1D). In the late secretory (LS, n=6) endometrium, strong to moderate staining was seen in the glandular epithelium whereas discrete cells within the stroma showed weak staining ( Fig. 1E). No immunostaining was observed in secretory endometrium ( Fig. 1F) when using nonimmune serum instead of the primary antibody.
|
Localization of PAF-R mRNA
In situ hybridization using 35S-labeled antisense riboprobe demonstrated cyclic changes in the expression of PAF-R mRNA in endometrial tissue during the menstrual cycle, with the highest level in the LS (n=4) phase of the cycle. Comparison of the hybridization signal obtained with sections of EP (n=3) endometrium with either control sense ( Fig. 2A, B)
or the antisense ( Fig. 2C, D) probe demonstrated no expression of PAF-R mRNA in the early proliferative endometrium. In contrast, the level of hybridization signal was high in the late proliferative (n=4) endometrium and localized over the glands ( Fig. 2E, F). Similarly, PAF-R mRNA expression was high in the early secretory (n=3) endometrium, but was uniformly distributed over the glandular epithelium and stroma ( Fig. 3C, D). In the late secretory endometrium, the hybridization signal persisted over the glands and stromal cells and increased in intensity ( Fig. 3E, F). The degree of hybridization was consistently lower in the stroma compared with the glands throughout the menstrual cycle. A strong hybridization signal was seen around the blood vessels. The specificity of the signal was confirmed by the use of a sense probe ( Fig. 3A, B).
|
|
Semiquantitation of PAF-R mRNA throughout the menstrual cycle
To quantify changes in PAF-R gene expression in endometrium throughout the menstrual cycle, variations in transcript levels were determined using a semiquantitative RT-PCR by normalization of the amount of PCR product for the PAF-R against that for ß-actin. The overall trend in PAF-R mRNA expression throughout the menstrual cycle, despite considerable patient variation, showed that the transcript levels were highest in endometrium from the mid- to late proliferative and in the late secretory phases of the menstrual cycle. Amplification of cDNA with primers specific for mouse ß-actin produced a single band of the predicted size (460 bp). Figure 4A
shows amplification of appropriately diluted RT products from throughout the menstrual cycle standardized for ß-actin and on which amplification was subsequently performed, using primers specific for the PAF-R. Subsequent amplification of cDNA appropriately diluted with primers specific for the PAF-R similarly produced a single band of the predicted size (397 bp). Figure 4B demonstrates the cyclic changes in PAF-R mRNA expression in endometrial samples standardized for ß-actin, showing transcript levels increasing from the early to mid-proliferative, persisting in late proliferative endometrium, only to decrease in early secretory and then to increase progressively in the mid- and late secretory endometrium.
|
Quantitation of PAF-R mRNA throughout the menstrual cycle
Finally, levels of PAF-R mRNA were determined throughout the normal menstrual cycle using RNAse protection assay. Autoradiographic signals were apparent in all phases of the cycle; however, variation in signal intensity was observed between different phases ( Fig. 5A).
Quantitation of autoradiographic signals by densitometric analysis showed relatively low abundance of PAF-R mRNA throughout the cycle in EP, MP, and LP phases of the cycle, as shown in Fig. 5B. The expression of PAF-R remained low in the ES and MS phases, giving densitometric PAF-R/ß-actin ratios similar to those found for proliferative phase endometrium. In the LS phase, the abundance of PAF-R mRNA increased dramatically by sixfold relative to all other cycle phases.
|
Effect of PAF on NO release
Addtion of PAF to endometrial explants resulted in the rapid generation of NO consistent with activation of a constitutive isoform of NOS, with values exceeding basal release by 47.7±9.7% within 2 h of stimulation. No demonstrable difference was observed in the basal values of NO released by proliferative and secretory endometrium. The concentration of PAF required to evoke maximal NO release, however, appears to be governed by the phase of the menstrual cycle. Maximal NO release from proliferative (n=6) endometrial explants was observed ( Fig. 6A)
after the addition of 1 µM PAF, whereas maximal PAF-mediated NO release from secretory (n=6) endometrial explants required only 0.1 nM of PAF ( Fig. 6B).
|
At a range of PAF concentration (1 µM–1 pM) there was significant inhibition in total NO release in the presence of WEB2170 (10 µM), indicating that the response is mediated by the PAF-R. PAF-mediated NO release was attenuated by L-NMMA in a dose-dependent fashion, indicating NOS activation, with half-maximal inhibition (IC50) at approximately 5 µM ( Fig. 7).
To identify the regulatory mechanisms involved in PAF-stimulated NO release, the roles of external calcium and protein kinase C (PKC) were determined by conducting experiments in the presence of both low calcium buffer and Ro 31–8220, a PKC inhibitor. Stimulation of endometrial explants with either 0.1 nM or 1 µM PAF in the presence of external calcium evoked a significant increase in total NO, with values typically 61.1±16.0% above basal (n=4, P<0.0001). When stimulations were conducted in the presence of low calcium buffer (150 nM), however, PAF-mediated NO release was totally abolished, suggesting total dependency on extracellular calcium (n=4, P<0.0001; Fig. 8).
Moreover, the response was attenuated by pretreatment with, and stimulation in the presence of, 1 and 5 µM Ro 31–8220, inhibiting PAF-mediated (0.1 nM or 1 µM) NO release by 54.5±2.3% (n=3, P<0.001) and 82.5±8.0% (n=3, P<0.0001), respectively ( Fig. 8).
|
|
Effect of PAF on VEGF expression
To further define the role of PAF, we examined the effect of this phospholipid on VEGF mRNA and protein in the endometrial glandual epithelial cell line. After stimulation with PAF (1 µM PAF) or vehicle, the reactions were terminated at the appropriate time (1, 2, 4, 8, 12, 24, or 48 h) by removal of the conditioned medium from confluent cell monolayers, and total RNA or protein was extracted. Stimulation with PAF resulted in at least a twofold increase in VEGF mRNA at 24 h compared to control ( Fig. 9).
Prior to 24 h, mRNA levels were consistently lower in all samples; however, at 48 h VEGF mRNA was increased in both control and PAF-stimulated cells. PAF similarly evoked a dose-dependent up-regulation of VEGF protein in cell lysates after 24 h. The rabbit polyclonal anti-VEGF antibody raised against a 20 amino acid synthetic peptide, corresponding to the first 20 residues at the amino terminus of human VEGF, recognized one major band at 46 kDa ( Fig. 10).
This effect of PAF on VEGF protein expression was inhibited, in a dose-dependent manner, by the specific PAF-R antagonist WEB 2170 (0.1 µM–10 µM; data not shown). No VEGF was detected in the conditioned media of either control or PAF-stimulated samples by Western blotting.
|
|
Effect of PAF and VEGF on protein tyrosine phosphorylation
In unstimulated human endometrial epithelial cells, a number of tyrosine phosphorylated proteins were observed ( Fig. 11).
PAF (1 µM) stimulated a time-dependent increase in the tyrosine phosphorylation of two major proteins of approximately 125 and 42–44 kDa ( Fig. 11A). The phosphorylation state of pp125 was similarly increased by VEGF (1 ng/ml), but the response was more rapid than that observed with PAF; a significant increase in the phosphorylation state of this protein was apparent within 1 min rather than after 5 min of stimulation (Fig 11B). Reprobing of the anti-phosphotyrosine blot with anti-focal adhesion kinase anti-FLKpp125 antibody confirmed the identity of this band (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cyclic changes in the expression of PAF-R mRNA demonstrated when using semiquantitative RT-PCR reflect similar changes observed both in the pattern of immunostaining and the intensity of hybridization signal between different phases of the menstrual cycle. This demonstrates that expression of the PAF-R is modulated by sex steroids in vivo. The discrepancy between no hybridization signal in contrast to PAF-R-like immunoreactivity in the early proliferative endometrium most likely reflects increased turnover of steady-state PAF-R mRNA in this phase of the cycle. An increase in PAF-R mRNA expression from the early proliferative to the late proliferative endometrium suggests that the receptor mRNA may be up-regulated by estrogen. Indeed, positive transcriptional regulation of human PAF-R transcript 2 (tissue type) by estrogen has been demonstrated in JR-st cells, a human stomach cancer cell line that expresses both functional endogenous PAF-R transcripts (23), and, more recently, in endometrial cells in culture (22). Furthermore, PAF-R mRNA expression in the late proliferative endometrium was confined to individual endometrial glands, suggesting that the PAF-R gene is likely to be transcriptionally activated in these cells.
PAF production has been shown to be regulated by ovarian steroids in endometrium (3). Moreover, the observation that PAF-R mRNA expression is high in the secretory phase of the cycle suggests that progesterone does not antagonize PAF-R mRNA expression stimulated by estrogen, but most likely enhances preferential expression in the glands and endothelial cells of the blood vessels. PAF is known to stimulate synthesis of PGE2 by enriched glandular fractions of human secretory endometrium (4). Functional PAF-R are confined to the glandular epithelium, as demonstrated by us (for review, see ref 2) and confirmed by Sato and co-workers (22). A similar pattern of sex steroid regulation of expression was reported for VEGF (24, 25) and bradykinin (26) in endometrium. Moreover, PAF-evoked phospholipase D activity (13) and intracellular Ca2+ (22) was up-regulated by 17ß-estradiol. These results strongly suggest that PAF is involved in the physiological process of reproduction and in the preparation of a receptive endometrium for implantation.
PAF is a potent vasodilator, a response dependent on intact endothelium-derived NO synthesis (27, 28). PAF was reported to stimulate NO release in a human endometrial epithelial cell line, HEC-1B, suggesting that NO might augment the vasodilatory effects of PAF (2). The current study demonstrates not only the production of NO in human endometrium, but also shows that the vasodilatory action of PAF is via NO release. Implantation is characterized by vasodilation (1). PAF-evoked NO generation in human endometrial explants was rapid and calcium dependent, consistent with the activation of a constitutive NOS (14). PAF-evoked NO was blocked by PAF-R antagonist WEB2170, indicating a receptor-mediated response. The maximal endometrial NO release was governed not only by the concentration of PAF, but also by the phase of the menstrual cycle. At 1.0 µM, PAF evoked maximal NO release in the proliferative endometrium, whereas only 0.1 nM PAF was required to generate maximal NO release in the secretory endometrium. The different responses exhibited by the proliferative and secretory endometrium may reflect cyclic variation in PAF-R density and/or the activity of PAF acetylhydrolase, the enzyme responsible for degradation of PAF during the menstrual cycle. It is equally possible, however, that estrogen primes the uterus to become more responsive to PAF. Estradiol-17ß up-regulates PAF-evoked phospholipase D activity (13) and PAF-induced intracellular Ca2+ release (22). The observation that PAF-mediated NO release is attenuated by Ro 31–8220 suggests that PKC may be an important point of modulation in the release of NO from endometrial explants. Ro 31–8220 inhibits members of the conventional PKC family (PKC isoenzymes PKCs-, -ß, and -
) that are Ca2+ and phospholipid dependent (29). Use of Ro 31–8220 suggests a role for a member of the conventional PKC family in regulating PAF-mediated NO release. Indeed, PAF stimulates NO release from HEC-1B cells via PKC (30).
Increased formation of NO could lead to increased endometrial bleeding (31). PAF-R is localized around the endometrial blood vessels during the secretory phase of the menstrual cycle. Since PAF-R expression increases dramatically duning the late secretory endometrium and PAF stimulates NO release, it is possible that one physiological function of PAF may be to promote menstruation. After menstruation, repair of the endometrium involves profound angiogenesis. During the secretory phase of the menstrual cycle, the spiral artery development occurs under the influence of progesterone, a process during which spiral arterioles develop from straight arteries in the myometrium and basal endometrium into the thickened superficial zone of the secretory endometrium (32). Thus, the increase in PAF-R gene expression and protein observed in the late proliferative and late secretory endometrium suggests that PAF may modulate cell growth and neovascularization. Recent studies have confirmed that PAFs possess angiogenic activity (33). PAF stimulates directional migration of endothelial cells in vitro, consistent with the morphologic alterations and redistribution of cytoskeleton seen in previous studies (34, 35), by acting on specific PAF-Rs without affecting cell proliferation (33). Furthermore, PAF has been implicated as a mediator of the angiogenesis induced by tumor necrosis factor (36) and to cooperate with other angiogenic molecules and chemokines, including VEGF, acidic and basic FGF, and hepatocyte growth factor (HGF), in inducing vascular development in Kaposi's sarcoma (37). The fact that PAF can stimulate VEGF mRNA and protein in endometrium supports our notion that PAF may modulate endometrial remodeling after menses, probably by up-regulating VEGF expression. Increases in both VEGF gene expression (38) and protein production (25) have been demonstrated in the late proliferative and late secretory phases of the menstrual cycle.
PAF and VEGF both stimulate the tyrosine phosphorylation of a number of proteins, including focal adhesion kinase, which raises the possibility that PAF directly or via VEGF may regulate endometrial cell motility and adhesion by activating this pathway. PAF antagonists have previously been shown to inhibit trophoblast outgrowth in vitro (5), suggesting that PAF may be involved not only in the activation of blastocyst attachment, but also in the differentiation of the trophectoderm into invasive trophoblast. Together, these results suggests that functional PAF receptors may play a role in the regulation of growth and division in endometrial cell by coupling to a tyrosine kinase pathway. In summary, the cyclic changes in PAF-R-like immunoreactivity and PAF-R mRNA expression in human endometrium, together with the observed stimulatory effects of PAF on NO release, VEGF mRNA, and protein expression, suggest roles for PAF in the control of the uterine vascular bed as well as in regeneration of the endometrium in preparation for implantation after endometrial shedding.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication June 23, 1997. Accepted for publication January 20, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and PGE2 by separated cells of human endometrium. J. Reprod. Fertil. 82, 271–276[Abstract]
This article has been cited by other articles:
|
|
 |
|
  N. F. Voelkel, R. W. Vandivier, and R. M. Tuder Vascular endothelial growth factor in the lung Am J Physiol Lung Cell Mol Physiol, February 1, 2006; 290(2): L209 - L221. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. O'Neill The role of paf in embryo physiology Hum. Reprod. Update, May 1, 2005; 11(3): 215 - 228. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Chami, G. Evans, and C. O'Neill Components of a platelet-activating factor-signaling loop are assembled in the ovine endometrium late in the estrous cycle Am J Physiol Endocrinol Metab, August 1, 2004; 287(2): E233 - E240. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ogata, K. Nandate, T. Kawasaki, C. Kawasaki, M. Ozaki, and A. Shigematsu A Platelet Activating Factor Receptor Antagonist Inhibits Cytokine Production in Human Whole Blood by Bacterial Toxins and Live Bacteria Anesth. Analg., June 1, 2004; 98(6): 1767 - 1772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Maccarrone, M. DeFelici, F.G. Klinger, N. Battista, F. Fezza, E. Dainese, G. Siracusa, and A. Finazzi-Agro Mouse blastocysts release a lipid which activates anandamide hydrolase in intact uterus Mol. Hum. Reprod., April 1, 2004; 10(4): 215 - 221. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Y. Cheung Vascular Endothelial Growth Factor Activation of Intramembranous Absorption: A Critical Pathway for Amniotic Fluid Volume Regulation Reproductive Sciences, February 1, 2004; 11(2): 63 - 74. [Abstract] [PDF]
|
|
|
|
|
|
T. Sugano, H. Narahara, K. Nasu, K. Arima, K. Fujisawa, and I. Miyakawa Effects of platelet-activating factor on cytokine production by human uterine cervical fibroblasts Mol. Hum. Reprod., May 1, 2001; 7(5): 475 - 481. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. KENT, S. A VINORES, and P. A CAMPOCHIARO Macular oedema: the role of soluble mediators Br. J. Ophthalmol., May 1, 2000; 84(5): 542 - 545. [Full Text]
|
|
|
|
 |
|
 T. Sugano, K. Nasu, H. Narahara, Y. Kawano, Y. Nishida, and I. Miyakawa Platelet-Activating Factor Induces an Imbalance Between MatrixMetalloproteinase-1 and Tissue Inhibitor of Metalloproteinases-1 Expressionin Human Uterine Cervical Fibroblasts Biol Reprod, March 1, 2000; 62(3): 540 - 546. [Abstract] [Full Text]
|
|
|
