| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Biomedical Sciences, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada N1G 2W1
b Michigan Reproductive and IVF Center, Grand Rapids, Michigan 49546
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
-glutamyl carboxylase, an enzyme essential for the vitamin K-dependent posttranslational modification of prothrombin, is expressed in granulosa cells in a manner similar to prothrombin mRNA. Thrombin's biological effects are mediated through selective proteolytic cleavage and activation of specific receptors. Bovine granulosa cells possess thrombin receptor (PAR-1) mRNA, and as seen with prothrombin mRNA and -glutamyl carboxylase mRNA, cells isolated from small follicles possess more PAR-1 mRNA than cells from large follicles. Thrombin receptor expression by cells in close proximity to an active thrombin-generating system suggests that these factors may be important mediators of cellular function in the ovarian follicle.
follicle, follicular development, granulosa cells, ovary
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
The hemostatic proteins involved in thrombin formation represent one group of proteins that have a distinct difference between the profiles in plasma and follicular fluid [7–9]. Thrombin is generated by the proteolytic cleavage of the zymogen, prothrombin, in a reaction catalyzed by the "prothrombinase complex," which consists of activated factor X, factor V, phospholipid, and calcium. The formation of this complex is preceded by the conversion of the zymogen, factor X, to its activated serine protease, factor Xa, by the tissue factor (TF)-factor VIIa complex [9, 10]. Tissue factor is a transmembrane phosphoprotein expressed on the surface of endothelial cells and white blood cells in response to vascular trauma and inflammatory mediators. Factor VII is one of the circulating vitamin K-dependent proteins also present in follicular fluid [7–9]. Preliminary evidence from our laboratory suggests that TF mRNA is present in rat ovarian tissues, but the conditions under which it is expressed by ovarian cells are not known.
In part because of the absence of any correlation in follicular fluid between the concentration and the molecular size of individual proteins involved in thrombin formation, we have proposed that this group of proteins is locally synthesized and not derived directly from plasma [9, 11]. One of the objectives of this study was to isolate granulosa cells from normal bovine follicles to determine whether they expressed prothrombin mRNA and produced the prothrombin protein. Prothrombin was selected for examination because it is the immediate precursor of thrombin and its presence has only been inferred in bovine follicular fluid, although it has been quantified in human follicular fluid [9]. The ability of TF to induce a functional prothrombinase complex depends on carboxylated glutamic acid residues being present in prothrombin as well as factors X and VII [12]. Thus, another approach to determining whether functionally active prothrombin could be generated within the follicle was to examine bovine granulosa cells for the presence of mRNA for -glutamyl carboxylase, the enzyme that is required for the carboxylation reaction. Although ovarian follicular cells appear not to have been examined for the presence of this enzyme, it has been found in essentially all mammalian tissues, including the testes and uterus [13].
In plasma, thrombin generation is a highly regulated process [14]. Because preliminary experiments indicated that the kinetics of thrombin formation appear to be slower in follicular fluid than in plasma [15], in this study we also compared the relative activity of the thrombin inhibitory proteins in follicular fluid and plasma. The inhibitors examined were TF pathway inhibitor (TFPI) and protein C, which both inhibit activity of the prothrombinase complex, and antithrombin, which inhibits both thrombin and its formation [14]. Antithrombin has been identified in follicular fluid from a number of species at levels close to those in plasma [7–9]. Protein C has been quantified in human follicular fluid [9], but to our knowledge, no previous reports have quantified TFPI.
In addition to its role in hemostasis, thrombin interacts with specific cell surface receptors to mediate a wide range of biological responses, including cell proliferation, extracellular matrix turnover, chemotaxis, and cytokine release [16, 17]. Thrombin receptors (ThRs) have been identified in numerous cell types, including fibroblasts, platelets, smooth muscle cells, stromal cells, and neuronal and glial cells [18–23]. In parts of the brain, prothrombin mRNA has been shown to colocalize with ThR [23], and a temporal correlation has been found between the levels of prothrombin and ThR during embryogenesis in mice [24]. Whether thrombin has a role in reproduction has not yet been defined.
One of the questions addressed in this study was whether ThRs are present within the ovarian follicle. The ThR is activated in a unique manner; thrombin cleaves a small peptide from the extracellular amino-terminal extension of the receptor and exposes a "tethered ligand," which then interacts with a separate region to induce cellular responses. Based on this method of activation, ThRs are also referred to as "protease-activated receptors" (PARs). A number of PARs have been identified. In this study, PAR-1 was selected for examination, because it is found in both vascular and extravascular cell types [18, 22, 25].
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
Bovine ovaries were obtained from a local slaughterhouse and transported in sterile PBS (pH 7.4) at 37°C. The ovaries were rinsed in 70% (v/v) ethanol and then washed in sterile PBS at 37°C. After elimination of grossly atretic follicles, follicular fluid was aspirated from the follicles with a 1-ml tuberculin syringe. The aspirated fluid was centrifuged at 11 000 x g for 20 min at 4°C to remove contaminating cells, and the clear supernatant was stored in small aliquots at -20°C. In some experiments, small (<0.5 cm), medium (0.5–1.0 cm), and large (>1.0 cm) follicles were aspirated. Criteria for sizing follicles were based on those of Einspanier et al. [26].
Blood was collected by jugular puncture from 10 healthy, mature Holstein cows maintained in the Ontario Veterinary College herd according to the guidelines of the Canadian Council on Animal Care. The blood was immediately transferred into plastic centrifuge tubes containing 0.17 M sodium citrate in the proportion of nine parts blood to one part anticoagulant. After centrifugation at 3000 x g for 20 min at 4°C, the top two-thirds of the platelet-poor plasma was removed. This plasma was pooled and stored in small aliquots at -20°C. Serum was prepared by allowing noncitrated whole blood to clot overnight at 37°C in a plain glass tube. After centrifugation at 2100 x g for 20 min at 4°C, the clear supernatant serum was stored in small aliquots at -20°C.
Preparation of Human Follicular Fluid
A total of 30 follicular fluid samples were collected from individual patients undergoing assisted reproductive technology procedures. Follicular fluid was prepared as previously described [9]. Briefly, fluid was aspirated from a specific follicle with an Echotip needle (16-gauge, 33 cm; Cook OB/Gyn, Spencer, IN) and collected in a polystyrene tube. After egg retrieval, the follicular fluid was centrifuged at 3000 x g for 15 min at 22°C, and clear, follicular fluid free of blood contamination was stored in small aliquots at -20°C. The protocol for this study was approved by the Internal Review Board of Spectrum Health-Downtown Campus.
Isolation of Bovine Granulosa Cells, Preparationof Lysates, and Isolation of RNA
Fluid was aspirated from follicles of various sizes as described above into Dulbecco modified Eagle medium:Ham F-12 (1:1 [v:v] ratio) buffer supplemented with 1% fungizone, 2% gentamicin, 1% heparin, and 2% penstrep. After centrifugation at 250 x g for 5 min, the supernatant was discarded and the pellet suspended in 1 ml of PBS buffer (10 mM, pH 7.2) before being recentrifuged at 5000 x g for 8 sec to repellet the granulosa cells. After draining the supernatant, the pellet was resuspended in 1.0 ml of lysis buffer containing a mixture of proteinase inhibitors that included 20 µl of 50 mM PMSF, 10 µl of aprotinin (2.0 mg/ml), and 20 µl of 0.1 M dithiothreitol. The solution was mixed and allowed to sit on ice for 5 min before a final centrifugation at 12 000 x g for 10 min at 4°C. The resulting supernatant was stored in small aliquots at -20°C. The isolation of RNA from the granulosa cells was achieved using Trizol (Gibco BRL, Life Technologies, Grand Island, NY) as described by the manufacturer.
Electrophoresis and Western Blot Analysis
Using 7.5% acrylamide gels, SDS-PAGE was performed according to the method of Laemmli [27]. Separated proteins were transferred to polyvinylidene difluoride membranes (Millipore Corp., Bedford, MA) at 300 mA for 2.5 h using Towbin transfer buffer [28]. After transfer, the membranes were washed with Tris-buffered saline (TBS; 0.05 M, pH 7.5) at room temperature and blocked with TBS containing 3% skim milk and 1% horse serum at 4°C for 15–16 h. The membranes were incubated with the primary antibody, which was antibody to bovine thrombin (known to recognize both thrombin and prothrombin) raised in sheep (Enzyme Research, South Bend, IN) at a 1:1000 dilution for 45–60 min at 37°C. Membranes were washed with TBS and incubated with peroxidase-labeled anti-sheep/goat IgG-POD (Boehringer-Mannheim, Laval, PQ, Canada) at a dilution of 1:500. Peroxidase activity was detected using the BM Chemiluminescence Kit (Boehringer-Mannheim). Prothrombin and thrombin standards were obtained from Enzyme Research (South Bend, IN). Before electrophoresis, the total protein content of each sample was determined using the method of Lowry [29], with bovine serum albumin as standard.
Northern Blot Analysis
The RNA samples (20 µg each) were separated on 1.0% agarose/formaldehyde gels. Gels were rinsed in 40 mM Tris-acetate and 1 mM EDTA (pH 8.0; TAE), and the RNA was electrotransferred to a nylon membrane (Hybond; Amersham Life Sciences, Oakville, ON, Canada) in 0.5x TAE for 24 h at 250 mA. After transfer, the membrane was rinsed in TAE, patted dry, and irradiated at 1200 µJ/cm2 in a UV Crosslinker (Fisher Scientific, Nepean, ON, Canada). Membranes were prehybridized for 3 h in 5x SSC (1x SSC:0.15 M sodium chloride and 0.015 M sodium citrate), 5x Denhardt reagent, 50 mM sodium phosphate, 50% formamide, 0.1% SDS, and 100 µg/ml of fish sperm DNA at 42°C. A cDNA for bovine prothrombin [30] kindly provided by Dr. Ross MacGillivray (University of British Columbia, Canada), a cDNA probe for human -glutamyl carboxylase [31] kindly provided by Dr. Darrel Stafford (University of North Carolina at Chapel Hill), and a cDNA for probe for murine ThR (i.e., PAR-1) [22] kindly provided by Dr. Shaun Coughlin (University of California, San Francisco) were each labeled with [
32P]dCTP (Amersham) using a Rediprime (Amersham) random-primer extension kit, added to the prehybridization solution, and incubated with the membrane overnight at 42°C. The membrane was then washed twice for 15 min in 2x SSC and 0.5% SDS at 45°C and twice for 15 min in 0.1x SSC and 0.5% SDS at 55°C, patted dry, and exposed using a Molecular Imager GS250 (BioRad, Richmond, CA). To determine equivalency of load and transfer, membranes were stripped by washing in 0.01x SCC and 0.5% SDS for 30 min at 90°C and rehybridized with a probe for murine 7S ribosomal RNA.
Prothrombin Activation and Thrombin Formation in Follicular Fluid
The ability of exogenous TF to induce thrombin formation in follicular fluid and plasma was determined with both a functional assay and an antigenic assay system. For the functional assay, the time of fibrin clot formation was determined following the addition of 0.2 ml of a commercial rhTF preparation supplemented with synthetic phospholipids and calcium (Dade-Behring, Miami, FL) to 0.1 ml of bovine follicular fluid or plasma. The time for fibrin clot formation was recorded on a Coag-A-Mate XM (Organon Teknika, Durham, NC). To ensure that fibrinogen, the thrombin substrate in the reaction, was not a rate-limiting factor, both follicular fluid and plasma were evaluated undiluted and following 1.25-, 1.67-, and 2.5-fold dilution with purified bovine fibrinogen (4 g/L; Sigma-Aldrich Chemical Co., St Louis, MO). Three separate trials were conducted, and all samples were assayed in duplicate. The results are expressed as mean ± standard deviation.
The antigenic determination of thrombin generation in bovine and human follicular fluids was performed using SDS-PAGE and Western blot analysis. Thrombin generation was initiated by addition of the rhTF preparation to pooled bovine follicular fluid samples diluted 4-fold with H2O. After 0, 1, and 2 min of activation, each sample was transferred to a conical polystyrene vial along with Tris buffer (10 mM, pH 6.8) containing 10% SDS, and the mixture was then placed in a boiling water bath for 5 min before electrophoresis. Nonactivated bovine plasma, purified bovine thrombin, and prothrombin were used as positive controls. The experiment was run in triplicate. A similar protocol was used to compare rhTF-induced thrombin formation in human follicular fluid and plasma.
The effect of exogenous bovine thrombin (Thrombostat; Parke Davis, Scarborough, ON, Canada) on the relative rates of fibrin formation in bovine follicular fluid and bovine plasma was assessed by adding various dilutions of a stock solution of thrombin (100 NIH units/ml) diluted with Tris buffer (10 mM, pH 7.4). To ensure that fibrinogen was not rate-limiting for the reaction, pooled follicular fluid was diluted 1.25- and 2.5-fold with a bovine fibrinogen solution (4 g/L). For the reaction, 0.1 ml of diluted follicular fluid or plasma was prewarmed to 37°C before the addition of 0.1 ml of thrombin, and the time for fibrin clot formation to occur was recorded on the Coag-A-Mate XM.
Activity of Individual Thrombin-Generating and Thrombin-Modulatory Proteins
A modification of the thrombin clotting time assay with the rhTF preparation as activator was used to quantify factors V, VII, and X in human and bovine follicular fluid as previously described [7–9]. The procedure involved mixing diluted test follicular fluid with a substrate human plasma (BioPool Canada, Inc., Burlington, ON, Canada) that was devoid of one of the factors. The clotting time reflects the ability of the test plasma to replace the protein that is absent in the substrate plasma and to correct the prolonged clotting time. The clotting times obtained with each sample for each protein were converted to percentage activity from standard curves prepared with human reference plasma or normal bovine plasma, depending on the source of the sample being analyzed. All samples were assayed in duplicate for each protein.
Prothrombin, antithrombin, TFPI, and protein C activities were determined with chromogenic substrates. Total prothrombin activity was determined by cleavage of prothrombin in follicular fluid with Ecarin (Sigma-Aldrich), a specific prothrombin activator, and with S-2238 (Dia Pharma, Inc., West Chester, OH) as previously described [9]. Antithrombin was assessed on the basis of residual activity toward S-2238 following thrombin neutralization by follicular fluid diluted in a heparin-Tris buffer (0.5 M, pH 8.8, 3 U of heparin [Glaxo Canada Ltd., Toronto, ON] per ml). The TFPI activity assay was used as described by the manufacturer (American Diagnostica, Inc., Greenwich, CT), as was the chromogenic protein C assay (Sigma-Aldrich). All samples were assayed in duplicate.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
The presence of prothrombin in follicular fluid was visualized by Western blot analysis using an anti-bovine thrombin antibody that recognizes both prothrombin and thrombin. The profiles in representative samples of several pooled follicular fluid samples were compared to those in bovine plasma, serum, and both prothrombin and thrombin standards. Results presented in Figure 1 illustrate the presence of an equivalent antigenic prothrombin band (solid double-headed arrow) in bovine plasma (lane 4), and three different representative samples of bovine follicular fluid (lanes 1–3) are shown. The 72-kDa prothrombin band is absent in serum (lane 5), where prothrombin depletion occurs due to plasma clotting, further supporting the identity of the band. The lower molecular weight bands in the prothrombin standard (lane 6) represent autoactivation products. The other bands observed in both plasma and follicular fluid samples at approximately 132 kDa (diamonds) and 50–55 kDa (circles) represent nonspecific reactivity of the secondary antibody, because they are present in samples in which the primary antithrombin antibody was omitted during the blocking procedure (follicular fluid, lane 8; plasma, lane 9).
|
Because an equal amount of total protein (4 µg) was used, the similarity of the intensity of prothrombin band staining in the follicular fluid and plasma samples suggests that the amount of prothrombin present in the two types of fluid is similar. This was confirmed by comparing the total prothrombin activity in follicular fluid relative to plasma. Using pooled bovine plasma as a 100% standard, the prothrombin activity in 10 representative follicular fluid samples was 105 ± 4% of plasma (range, 84.3%–125.4%) (Table 1).
|
Assessment of Prothrombin mRNA in Bovine Granulosa Cell Lysates
To determine whether granulosa cells have the capacity to synthesize prothrombin, Northern blot analysis was performed on total cellular RNA isolated from granulosa cells using a cDNA specific for prothrombin. Prothrombin mRNA was detectable as a single 2.0-kilobase (kb) band in granulosa cells from different-sized follicles (Fig. 2A, upper panel) in each of three separate experiments using RNA from cells isolated from pooled follicles. Signal intensity was strongest in the cell lysates from small follicles. Load equivalency was demonstrated by hybridization with a probe for 7S ribosomal RNA (Fig. 2A, lower panel). The amount of prothrombin mRNA appears to be higher in small and medium-sized follicles than in large follicles.
|
Antigenic Determination of Prothrombin in Bovine Granulosa Cells
The presence of the prothrombin protein was confirmed in granulosa cells with Western blot analysis using both a pooled sample of lysates obtained from granulosa cells isolated from follicles at various stages of development and the lysates obtained from individual follicles. Pooled lysate samples were used to determine whether a dose-dependent correlation existed between the amount of protein loaded onto the gel and the intensity of the chemiluminescent signal. A representative gel from three separate experiments is illustrated in Figure 3A. Not only is the 72-kDa prothrombin band present in the cell lysates (lanes 5–8, arrow), but the intensity of this band increases as the amount of protein loaded is increased incrementally from 4 to 15 µg. It would appear that the amount of prothrombin present in cell lysates is lower than that in follicular fluid, because in follicular fluid, the prothrombin band is visible with 4 µg of total protein (Fig. 1, lanes 1–4) but cannot be detected in the lysates when this amount of protein is loaded (Fig. 3A, lane 4). When the same amount of total protein is loaded (10 µg), the intensity of staining of the prothrombin band is noticeably lower in the lysates (Fig. 3, lane 6) than in follicular fluid (Fig. 3, lane 3). The total amount of prothrombin activity could not be accurately quantitated in the granulosa cell lysates, because proteinase inhibitors present in the lysate buffer interfere with the assay (unpublished observation).
|
In addition to prothrombin, other proteins in the granulosa cell lysates reacted with the primary antibody to thrombin. The band corresponding to a protein with a molecular weight of approximately 36 kDa is apparent in granulosa cells (Fig. 3A, lanes 6–8) and in some, but not all, follicular fluid samples (Fig. 1, lanes 1–4, and Fig. 3A, lane 3). This protein may represent a prothrombin activation product, because a similar band can also be seen in the prothrombin standard (Fig. 3A, lane 2). The identity of the high molecular weight bands that are visible in the lysates, but not in the follicular fluid samples, are not known and are presently under investigation.
Prothrombin content of granulosa cell lysates prepared from small, medium, and large follicles was evaluated to determine whether a correlation existed between the amount of prothrombin mRNA and protein. A representative profile from three separate experiments is shown in Figure 3B, with the profile of lysates (10 µg) from small (lane 9), medium (lane 10), and large follicles (lane 11) compared to a prothrombin standard. The signal for the 72-kDa prothrombin band appears with greatest intensity in the lysates prepared from the smallest follicles.
Identification of -Glutamyl Carboxylase in Granulosa Cell Lysates
To determine whether granulosa cells have the capacity to synthesize fully the functional form of prothrombin, RNA isolated from the granulosa cells of small, medium, and large follicles was probed for -glutamyl carboxylase mRNA. Figure 2B (upper panel) indicates that a single 3.7-kb band is visible in all samples. After stripping the membrane, reprobing with a radiolabeled probe for 7S ribosomal RNA demonstrated equivalent loading and transfer for all samples (Fig. 2B, lower panel).
Detection of PAR-1 mRNA in Granulosa Cells
Using a cDNA-specific probe for the ThR known as PAR-1 [22], Northern blot analysis was performed on total cellular RNA isolated from granulosa cells of follicles at various stages of development. Thrombin receptor mRNA (i.e., PAR-1 mRNA) was detectable as a single 4.5-kb band in all granulosa cell samples (Fig. 2C, upper panel) in each of three separate experiments and was absent from hepatocyte mRNA. Load equivalency was demonstrated by hybridization of the samples with a probe for 7S ribosomal RNA (Fig. 2C, lower panel).
Thrombin Formation in Follicular Fluid
The formation of thrombin in follicular fluid as a result of activation of the TF-factor VIIa-factor Xa (i.e., prothrombinase) complex following the addition of exogenous TF to follicular fluid was examined by Western blot analysis. To visualize thrombin formation, it was necessary to load 10 µg of total protein onto the gel. A representative profile is shown in Figure 4. The profiles of untreated follicular fluid (lane 4) and of follicular fluid exposed to rhTF for 1 min (lane 5) and 2 min (lane 6) were compared to those of bovine plasma (lane 3) and to both thrombin (lane 1) and prothrombin (lane 2) standards. After the addition of TF, the appearance of a 32-kDa band (circle) that corresponds to thrombin was clearly visible in the follicular fluid samples (lanes 5 and 6). Although a decrease in the intensity of the nonactivated prothrombin band was observed over the incubation period, it was apparent that not all the prothrombin had been converted to thrombin under the experimental conditions. A further change observed in follicular fluid after the addition of TF was the loss of a small-molecular-weight band of approximately 36 kDa (arrowhead) that corresponds to one of the autoactivation products visible in the prothrombin standard.
|
To examine the possibility that incomplete conversion of prothrombin to thrombin in bovine follicular fluid might be due to a lack of species specificity between bovine procoagulant proteins and human TF, the profile of thrombin formation in human follicular fluid (Fig. 5A) was compared to that of human reference plasma (Fig. 5C). As with bovine follicular fluid, thrombin formation (Fig. 5B, lane 9, circle and diamond) occurred within 1 min of the addition of TF to human follicular fluid (Fig. 5A, lanes 2, 4, and 6). Also, no apparent increase in the amount of thrombin generation was observed when the incubation was extended to 2 min (Fig. 5A, lanes 3, 5, and 7), despite the fact that prothrombin had not been depleted. Dilution of the follicular fluid before the addition of TF did not cause a consistent change in the prothrombin-to-thrombin conversion profile in follicular fluid (Fig. 5A, lane 2 vs. lane 4 vs. lane 6). This situation was also observed with a reference sample of human plasma diluted 8-fold (Fig. 5C, lanes 11 and 12) or 4-fold (Fig. 5C, lanes 13 and 14) before the addition of TF. It was not possible to compare the effects of higher amounts of human TF on thrombin formation in follicular fluid and plasma, because fibrin clots consistently developed in plasma samples.
|
Comparison of Fibrin Formation in Bovine Follicular Fluid and Plasma
Generation of functional thrombin activity in bovine follicular fluid relative to that in bovine plasma was assessed by determining the rate at which fibrinogen was converted to fibrin after the addition of exogenous rhTF. Fibrin clots formed in both follicular fluid and plasma, although the time for clot formation to occur in follicular fluid samples was consistently longer than that in plasma (Fig. 6). For example, the difference in fibrin clot formation with the follicular fluid samples diluted 1.25-fold was 22.3 ± 7.5 sec, and that of the samples diluted 2.5-fold was 20.1 ± 3.8 sec. The thrombin clotting times for bovine plasma found in this study are within the representative range of normal [32]. The profile of fibrin clot formation was similar in both follicular fluid and plasma, suggesting that the rate of thrombin formation is also similar. We have previously shown that fibrinogen levels are lower in bovine follicular fluid than in bovine plasma [7]. Because the follicular fluid samples had been supplemented with fibrinogen in this experiment, fibrinogen levels were not rate limiting for fibrin formation.
|
Activity of Thrombin-Generating and -Inhibitory Proteins
To determine whether the slower rate of fibrin clot formation was the result of a different profile of thrombin-generating or -inhibitory proteins in follicular fluid relative to plasma, human and bovine follicular fluid samples were assessed for relative amounts of prothrombin and factors V, VII, and X as well as antithrombin, TFPI, and protein C. To ensure that the activity measured in follicular fluid was similar to that in plasma, standard curves were prepared, as previously shown for prothrombin [9], with various dilutions of representative samples of pooled human and bovine follicular fluid and pooled normal plasma and human reference plasma before quantifying the activity of the individual proteins (data not shown). The similarity of the standard curves demonstrated that, with the exception of protein C, the activity of this group of proteins could be reliably determined in both human and bovine follicular fluids. Although protein C-like activity was detected in bovine follicular fluid, the protease inhibitory activity detected could not be attributed solely to protein C because of the lack of cross-reactivity between human protein C antibody in the commercial test kit and bovine protein C antigen. Hence, protein C could only be quantified in human follicular fluid.
All the proteins involved in the TF-initiated biochemical pathway of thrombin generation (i.e., factors VII, X, V, and prothrombin) were present in both bovine and human follicular fluid (Table 1). With the exception of factor VII, the average activity level of each protein was similar for each species and, for both prothrombin and factor X, was close to that found in plasma. Similarly, in human follicular fluid, the average activity for each of the inhibitory proteins evaluated was almost equivalent to, or higher than, that in the reference plasma (Table 1). For both antithrombin and TFPI, a broad range of values, determined as a percentage of human plasma, was evident among individual samples. For example, the range was 73.1%–278.0% of plasma for antithrombin and 54.0%–250.0% of plasma for TFPI.
Effect of Exogenous Thrombin on Fibrin Formation in Follicular Fluid
Although the Western blot analysis for both bovine and human follicular fluid indicated that thrombin is generated following the addition of TF, the relatively low levels of factors V and VII (Table 1) could account for a slower rate and level of thrombin generation in follicular fluid (Fig. 6). The anticoagulant properties of follicular fluid also might neutralize thrombin more effectively there than in plasma. To investigate this latter alternative, the ability of exogenous bovine thrombin to generate fibrin clots was compared in follicular fluid and plasma.
As shown in Figure 7, fibrin clot formation was not observed in follicular fluid at thrombin concentrations less than 40 U/ml. The time to fibrin clot formation dropped from 48.0 ± 7.3 sec at 60 U/ml to 27.5 ± 2.6 sec and 23.8 ± 5.4 sec at 90 and 100 U/ml, respectively. In contrast, the addition of 5 U/ml to plasma produced a fibrin clot within 49.2 ± 8.9 sec, whereas 60 U/ml caused fibrin clot formation in 7.0 ± 0.5 sec. To ensure that the availability of fibrinogen as a substrate for thrombin was not rate limiting in follicular fluid, the rate of fibrin formation was determined in follicular fluid that had been diluted 2-fold with bovine fibrinogen to yield a final fibrinogen concentration of approximately 3 g/L, which is within the range for bovine plasma. The rate of fibrin clot formation was similar in both the undiluted and the fibrinogen-supplemented follicular fluid samples.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
In bovine granulosa cells, prothrombin mRNA is expressed as a single 2.0-kb form. This agrees with the results found for prothrombin mRNA from bovine liver [34] and for bovine hepatocyte control RNA used in this study. Likewise, the size of the prothrombin molecules estimated from SDS-PAGE (72 kDa) agrees with previously published values for bovine prothrombin [35]. These results indicate that the prothrombin molecule synthesized by bovine ovarian granulosa cells is similar to that synthesized by hepatocytes. This conclusion is also consistent with the finding that mRNA for vitamin K-dependent -glutamyl carboxylase is expressed along with prothrombin mRNA in granulosa cells. The -glutamyl carboxylase is required for the posttranslational modification of prothrombin [36]. Without the carboxylation of glutamic acid residues, circulating prothrombin is functionally inactive, because it cannot interact with activated factor X, phospholipid, and calcium to form a reactive complex that permits thrombin formation to occur at a physiologically relevant rate [37]. Bovine follicular fluid, like plasma, contains adequate amounts of factor X [7], phospholipid [38], and calcium [39, 40] to permit rapid conversion of carboxylated prothrombin to thrombin.
If thrombin and its receptor have physiological roles in folliculogenesis, then thrombin must be generated within the follicle. Through a vitamin K-independent reaction with the specific snake venom-derived, prothrombin-cleaving enzyme, Ecarin, we have previously shown that the kinetics of activation of prothrombin are similar in human follicular fluid and human plasma [9]. Likewise, the pattern of thrombin generation observed in this study (Figs. 4 and 5) through the vitamin K-dependent TF pathway of prothrombin activation is similar in follicular fluid and plasma. The major difference between follicular fluid and plasma appears to be the slower rate of thrombin-mediated conversion of soluble fibrinogen to insoluble fibrin that occurs in follicular fluid (Fig. 6). This response could be due either to a difference in the extent to which thrombin generation can be sustained in follicular fluid after TF factor activation of the system or to a more effective thrombin-inhibitory system.
The activity levels of the thrombin-generating components determined in follicular fluid from individual women (Table 1) are in close agreement with those found in an earlier study [9], indicating that the profile is consistent and that the levels of zymogens, prothrombin, and factors X and VII are close to those in human plasma. The activity of the protein cofactor, factor V, is markedly lower in both human and bovine follicular fluid compared to plasma, as are levels of factor VII in bovine follicular fluid. However, these differences, by themselves, cannot account for the consistent differences of fibrin formation in follicular fluid in response to both endogenous and exogenous thrombin (Figs. 6 and 7). The shortest fibrin clotting times observed in this study with exogenous thrombin were in the range of 23.8–27.5 sec, which is similar to those ranges previously reported in bovine follicular fluid obtained from follicles at different stages of development [40].
The inhibitory activity of antithrombin is enhanced in the presence of heparin, which facilitates the formation of thrombin-antithrombin-heparin complexes. One way in which thrombin could be inhibited more effectively in follicular fluid compared to plasma is through the formation of these complexes. In the vascular system, heparan sulfates are localized to the surface of endothelial cells; however, soluble heparan sulfate has been identified as one of the glycosaminoglycans present in bovine follicular fluid [41]. This anticoagulant likely is locally derived within the follicle, because both porcine and rat granulosa cells can synthesize biologically active, heparin-like mucopolysaccharides [42].
The formation of a fibrin gel within the antrum of the follicle would be incompatible with sustained normal follicle development, because it would interfere with normal expansion of the extracellular fluid matrix and prevent normal diffusion of substrates between the follicular fluid, granulosa cells, and the developing oocyte. Thus, from a biological perspective, it appears to be logical that thrombin activity should be more tightly controlled in follicular fluid than in plasma, and that its activity should be directed toward a specific receptor rather than toward fibrinogen. It is possible that PAR-1 is the primary substrate for thrombin generated within the follicle. The presence of PAR-1 mRNA in granulosa cells indicates that the PAR-1 protein is also present.
The decrease in the relative amounts of prothrombin mRNA, prothrombin protein, and PAR-1 in granulosa cells as the follicle matures suggests that thrombin may have a role in the early stages of folliculogenesis. Thrombin is now recognized to have a number of biological functions, including cell proliferative effects, that are mediated through its specific cell surface receptors [16, 43]. Transgenic mouse studies have demonstrated that prothrombin deficiency results in embryonic and neonatal lethality [44]. In zebrafish embryos, prothrombin mRNA is expressed early in embryogenesis and accumulates in diverse tissues, including neuronal tissue, before the initiation of blood formation [45]. Although the local production of thrombin appears not to have been evaluated in the brain in situ, numerous in vitro studies have documented thrombin's effects on morphological changes, survival, and apoptosis of neuronal and glial cells [46–49]. It is possible that thrombin exhibits similar effects within the reproductive system. Thrombin has been shown to induce cell proliferation in rat stromal cell cultures [20] and bovine ovarian theca cells [49] and to inhibit cell growth in bovine endometrial cells [50].
During the female reproductive cycle, the ovary undergoes continuous growth, atresia, and repair, thus providing multiple stages at which thrombin could exert a biological effect. A role for thrombin and its potential interactions with other growth factors within the ovary remains to be determined. Figure 8 illustrates the profile of thrombin formation in follicular fluid as is currently understood. To understand the specific roles of the ThR interactions, it will be necessary to determine the effects of ThR activation on cellular growth and differentiated function.
|
|
NOTE ADDED IN PROOF |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by grants from the Natural Sciences and Engineering Research Council of Canada and the Ontario Ministry of Agriculture and Food (J.J.P., J.L., and P.A.G.). 
2 Correspondence. FAX: 519 767 1450; pgentry{at}uoguelph.ca
Accepted: November 28, 2001.
Received: September 19, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
-glutamyl carboxylase. Science 1991; 254:1634-1636This article has been cited by other articles:
|
|
 |
|
  A. I. de Agostini, J.-C. Dong, C. de Vantery Arrighi, M.-A. Ramus, I. Dentand-Quadri, S. Thalmann, P. Ventura, V. Ibecheole, F. Monge, A.-M. Fischer, et al. Human Follicular Fluid Heparan Sulfate Contains Abundant 3-O-Sulfated Chains with Anticoagulant Activity J. Biol. Chem., October 17, 2008; 283(42): 28115 - 28124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kisliouk, N. Levy, A. Hurwitz, and R. Meidan Presence and Regulation of Endocrine Gland Vascular Endothelial Growth Factor/Prokineticin-1 and Its Receptors in Ovarian Cells J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 3700 - 3707. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Hirota, Y. Osuga, O. Yoshino, K. Koga, T. Yano, T. Hirata, E. Nose, T. Ayabe, A. Namba, O. Tsutsumi, et al. Possible Roles of Thrombin-Induced Activation of Protease-Activated Receptor 1 in Human Luteinized Granulosa Cells J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 3952 - 3957. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
