Platelets Are Novel Regulators of ... - Semantic Scholar

0013-7227/07/$15.00/0 Printed in U.S.A.

Endocrinology 148(7):3056 –3064 Copyright © 2007 by The Endocrine Society doi: 10.1210/en.2006-1687

Platelets Are Novel Regulators of Neovascularization and Luteinization during Human Corpus Luteum Formation Kazumi Furukawa, Hiroshi Fujiwara, Yukiyasu Sato, Bin-Xiang Zeng, Haruko Fujii, Shinya Yoshioka, Eiichiro Nishi, and Takeshi Nishio Department of Gynecology and Obstetrics (K.F., H.Fujiw., B.-X.Z., H.Fujii, S.Y.), and Molecular Pathology Unit (E.N.), Horizontal Medical Research Organization, and Department of Integrative Brain Science (T.N.), Faculty of Medicine, Kyoto University, Kyoto 606-8507, Japan; and Department of Obstetrics and Gynecology (Y.S.), Osaka National Hospital, Osaka 540-0026, Japan The human corpus luteum is a unique endocrine organ that is periodically constructed from the ovulated follicle. During human corpus luteum formation, which is well known as a pathophysiological model for tissue remodeling, the precise mechanisms by which centripetal vascular development is regulated remain unknown. Recently platelets were reported to contain chemoattractive substances with the potential to induce endothelial migration. In this study, we examined the involvement of platelets in the early tissue remodeling process of the human corpus luteum. An immunohistochemical study demonstrated that considerable amounts of red blood cells and CD41-positive platelets were localized at extravascular sites among luteinizing granulosa cells after ovulation. Platelet deposition gradually decreased and became limited near the central cavity toward which microvessels were extending. Platelets were hardly observed in the midluteal

T

HE HUMAN CORPUS luteum (CL) is a unique endocrine organ that is newly constructed from the ovulated follicle during the menstrual cycle. This process is well known as a physiological model for tissue remodeling. The CL produces progesterone, which is essential for inducing and maintaining embryo implantation in the uterus early in pregnancy. To supply this hormone to the systemic circulation, two major phenomena are accomplished during CL formation, i.e. granulosa cell luteinization and neovascularization. Just before follicular rupture, granulosa cells in the follicular fluid, which contains anticoagulant substances (1), proceed to luteinization and shift their main products from estrogen to progesterone. After ovulation, these granulosa cells undergo hypertrophy to differentiate into large luteal cells, being in contact with migrating endothelial cells (2) and producing extracellular matrix (ECM) around the luteal cells (3–5). First Published Online April 19, 2007 Abbreviations: CL, Corpus luteum; ECM, extracellular matrix; FCS, fetal calf serum; FITC, fluorescein isothiocyanate; HCG, human chorionic gonadotropin; HUVEC, human umbilical vein epithelial cell; IVF, in vitro fertilization; mAb, monoclonal antibody; MCAM, melanoma cell adhesion molecule; pAb, polyclonal antibody; VEGF, vascular endothelial growth factor. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

phase when the vascular network had already been established. These platelets expressed CD62P/P-selectin and were colocalized with extracellular matrix, suggesting that platelets had been activated by the extracellular matrix. Progesterone production by luteinizing granulosa cells that were isolated from patients undergoing in vitro fertilization therapy was significantly promoted by direct contact with platelets during 4-d culture. Platelet-derived soluble factors induced spreading in granulosa cell morphology. These factors also increased the migration of human umbilical vein endothelial cells, whereas luteinizing granulosa cells attenuated platelet-induced endothelial cell migration. These findings lead us to propose the novel concept that platelets are regulators of endothelial cell migration and granulosa cell luteinization in the remodeling process of the human corpus luteum. (Endocrinology 148: 3056 –3064, 2007)

Dramatic centripetal angiogenesis also occurs from the vascular network surrounding the follicle, although follicular fluid contains antiangiogenetic factors (6). During ovulation, the follicular basement membrane is destroyed and endothelial cells just outside the membrane migrate into the inner granulosa cell layer. In the human CL, it takes several days to complete mature vascular networks among the luteal cells (2), finally achieving vascular anastomosis in the central cavity area, which is the luteal remnant of the antral cavity of a ruptured follicle. To induce neovascularization, luteinizing granulosa cells have been proposed to secrete several soluble angiogenic factors such as vascular endothelial growth factor (VEGF), angiogenin, endocrine gland-VEGF, and angiopoietin (7–10). Luteinizing theca cells were also proposed to play some role in angiogenesis in human CL (11). In addition, we previously reported that luteinizing granulosa cells increased the cell surface expression of ephrin B1 and melanoma cell adhesion molecule (MCAM), which are reported to regulate endothelial migration and vessel formation by cell-to-cell contact (12–14). Furthermore, ECM produced by luteinizing granulosa cells is considered to modulate the migration and outgrowth of endothelial cells (15). However, there is no definite evidence demonstrating that there is local dominance of angiogenic factors in the central cavity to maintain endothelial migration toward the area until final the anastomosis is

3056

Furukawa et al. • Platelets Regulate CL Formation

achieved. Thus, the precise mechanisms by which centripetal vascular development is regulated remain unknown. During ovulation, a decrease of vascular stability beneath the follicular basement membrane is evident, and consequently blood cells migrate into extravascular spaces around luteinizing granulosa cells (2, 16). In addition, blood vessels begin to penetrate into the granulosa cell layer and some vessels open into the antral cavity, filling it with blood (17). Accordingly, fresh bleeding toward the central cavity is often observed for 4 d in CL in the early stage (2, 18). Thus, blood plasma fluid and blood cells, including red cells, normally flow among luteinizing granulosa cells that are surrounded by ECM in the extravascular spaces (2, 16), and the CL in this stage is occasionally called the corpus rubrum. However, the fibrin net was mainly observed in the central cavity area, and only sparsely among luteinizing granulosa cells (2), theoretically suggesting that there is some anticoagulant system(s) operating around granulosa cells. This condition is considered essential for maintaining the local circulation of tissue fluid throughout fresh CL and recruiting this progesteronecontaining tissue fluid into the systemic circulation. However, very little attention has been given to this paradoxical issue, and it remains unclear how the dynamic kinetics of tissue fluid are controlled, regulating the coagulation systems throughout the process of corpus luteum formation until the establishment of a viable vascular network. Together with red blood cells, platelets, which are a type of blood cells that plays an important role in coagulant systems, are likely to be exuded into extravascular spaces among luteinizing granulosa cells. Recently platelets were reported to contain chemoattractive substances potentially capable of inducing endothelial migration (19) and to play an important role in pathological processes such as atherosclerosis, wound healing, and tissue remodeling (20, 21). Therefore, we examined the precise spatiotemporal distribution of platelets in human CL and estimated their possible roles in CL formation, which is one of the dynamic and physiological phenomena involving in tissue remodeling. Materials and Methods Reagents The mouse antihuman integrin ␣IIb/CD41 (clone M148) and thrombomodulin/CD141 (clone 1009) monoclonal antibodies (mAbs) were obtained from Novocastra Laboratories Ltd. (Newcastle, UK). Fluorescein isothiocyanate (FITC)-conjugated mouse antihuman CD41 mAb (clone M148) was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The mouse antihuman P-selectin/CD62P (clone AK-4) mAb was obtained from BD Biosciences-PharMingen (Tokyo, Japan). The mouse antihuman collagen type IV (clone IV-3A9) mAb and antihuman fibronectin (clone 96 –21F2) mAb were purchased from Daiichi Fine Chemical (Takaoka, Japan), and antihuman fibrin mAb (clone E8) was obtained from Chemicon (Temecula, CA). Antihuman 3␤-hydroxysteroid dehydrogenase rabbit polyclonal antibody (pAb) was purchased from Oxygene (Dallas, TX). FITC-conjugated and nonconjugated mouse IgG1 (clone DAK-GO1) and IgG2b (clone DAK-GO9) mAbs and rabbit Ig for negative controls were all obtained from Dako (Glostrup, Denmark). For the secondary antibody, FITC-conjugated rabbit antimouse Ig pAb (Dako), FITC-conjugated swine antirabbit Ig pAb (Dako) and rhodamine-conjugated goat antimouse Ig pAb (Santa Cruz) were used. The mouse antihuman MCAM (clone S-Endo1; Alexis Biochemicals, San Diego, CA) and antihuman integrin ␣5 (clone SAM-1; Chemicon) mAbs were used for flow cytometry.

Endocrinology, July 2007, 148(7):3056 –3064

3057

Tissues CL of the early (CL d 2–5, n ⫽ 17) and midluteal (CL d 7– 8, n ⫽ 5) phases were obtained from 22 women, aged between 25 and 43 yr. All women had undergone unilateral ovarian cystectomy or oophorectomy and contralateral wedge resection to treat benign ovarian tumors. All of the women had a history of regular menstrual cycles (28 –30 d), and their ovulatory basal body temperature charts were consistent with normal luteal phase length. No patient used contraceptives or GnRH analogs within at least 3 months before the operation. The CL day was reevaluated according to histological dating, using hematoxylin and eosinstained tissue sections that were fixed with 10% formalin and embedded in paraffin (2). The migration of endothelial cells and the size of luteal cells were used for this classification. In the present study, the term CL day was used according to this definition. For example, CL d 2 was the day after ovulation, which was confirmed by transvaginal ultrasonography and histological dating. Informed consent for the use of these tissues was obtained from each donor. Use of the materials was also approved by the Ethics Committee of Kyoto University Hospital.

Immunohistochemistry Double-immunofluorescence staining was performed as previously described (22, 23). Frozen tissues were sliced to 7-␮m thickness using a cryostat microtome (Cryocut 1800; Reichert-Jung, Heidelberg, Germany), immediately air dried on Neoplene (Nisshin EM, Tokyo, Japan)coated glass slides, and fixed in acetone at ⫺20 C for 5 min. The frozen sections were incubated with antihuman collagen type IV mAb [5 ␮g/ ml, diluted in culture medium containing 10% fetal calf serum (FCS; Equitech-Bio, Inc., Kerrville, TX) and 0.1% NaN3], antihuman fibronectin mAb (5 ␮g/ml), antihuman fibrin mAb (5 ␮g/ml), antihuman CD41 mAb (5 ␮g/ml), antihuman CD62p mAb (5 ␮g/ml), or mouse negative control IgG1 (5 ␮g/ml). After the slides were washed in PBS, they were incubated with rhodamine-conjugated goat antimouse immunoglobulin. The washed slides were blocked with mouse anti-TNP (trinitrophenyl) mAb (unrelated mAb; 20 ␮g/ml) and then incubated with FITCconjugated antihuman integrin ␣IIb/CD41 mAb (10 ␮g/ml). Otherwise, for secondary staining, the washed slides were incubated with antihuman 3␤-hydroxysteroid dehydrogenase rabbit pAb (10 ␮g/ml) or control rabbit Ig (10 ␮g/ml), followed by FITC-conjugated swine antirabbit Ig. The slides were washed, mounted with a mounting agent (Perma Fluor aqueous mounting medium; Immunon, Pittsburgh, PA), and examined under a confocal laser-scanning microscope (Carl Zeiss Inc., Jena, Germany).

Isolation of human luteinizing granulosa cells, platelets, and umbilical vein epithelial cells (HUVECs) Fresh human luteinizing granulosa cells were obtained from 38 patients aged from 25 to 39 yr who had undergone treatment for in vitro fertilization (IVF) as previously described (3). Human platelets were also isolated from patients undergoing IVF treatment as described (23, 24). Whole blood was obtained from patients undergoing IVF treatment, immediately mixed with 3.8% (vol/wt) trisodium citrate dihydrate (ratio of blood to citrate was 9:1) in polypropylene tubes, and centrifuged at 200 ⫻ g for 15 min at 22 C. The platelet-rich plasma was centrifuged after adding a mixture of 4.5% wt/volcitric acid and 6.6% wt/voldextrose at 50 ␮l/ml platelet-rich plasma. The sedimented platelets were resuspended in RPMI 1640 containing 5.4 mm EDTA, stabilized for 10 min at room temperature, centrifuged at 980 ⫻ g for 10 min at 22 C, and then was suspended in RPMI 1640 (2 ⫻ 108/ml). HUVECs were separated from the umbilical cord as described previously (25). After washing the inner wall of the umbilical vein with PBS to remove fetal blood, the lumen was filled with PBS containing Ca2⫹ and Mg2⫹ as well as 0.05% collagenase (Wako Pure Chemical Industries Ltd., Osaka, Japan) and incubated for 30 min at room temperature. After the detached cells were collected, the remaining endothelial cells in the inner layer of the umbilical vein were collected by further washing with RPMI 1640 containing 15% FCS. The isolated HUVECs were cultured using HuMedia-EB2 (Kurabo, Osaka, Japan) containing 2% FCS, human epithelial growth factor (10 ng/ml), hydrocortisone (1 ␮g/ml), human basic fibroblast growth factor (3 ng/ml), and heparin (10 ␮g/ml) as well as gentamicin (50 mg/ml) and amphotericin B (50 ␮g/ml). Immuno-

3058


cytochemical staining using anti-CD141/thrombomodulin mAb confirmed the greater than 95% purity of endothelial cells in the above preparation. Informed consent for the use of granulosa cells, platelets and HUVECs in this study was obtained from all donors. Use of the materials was also approved by the Ethics Committee of Kyoto University Hospital.

Luteinizing granulosa cell culture with platelets Isolated human granulosa cells were resuspended in culture medium consisting of RPMI 1640 medium supplemented with 10% FCS. These cells (1 ⫻ 105 cells/ml /well) were cultured in each well of 24-well plates (Becton Dickinson, Lincoln Park, NJ) in triplicate in the presence or absence of recombinant human chorionic gonadotropin (HCG) (5 IU/ ml; Rhoto Pharmaceutical Co. Ltd., Osaka, Japan) and isolated platelets (2 ⫻ 107 and 2 ⫻ 108 cells/ml). Granulosa cells were also incubated with platelets (2 ⫻ 108 cells/ml) in 0.45-␮m-pore culture chambers (Intercell, Kurabo Co. Ltd., Osaka, Japan), which prevented direct contact between granulosa cells and platelets. The culture medium was gently replaced with fresh medium every 2 d, and the collected medium was subjected to a RIA. Morphological changes were observed under a phase-contrast microscope and recorded using a digital camera (Camedia C5050; Olympus, Tokyo, Japan). The average length to width ratio of 30 cells in each well and the mean values of triplicate wells were calculated. The cell circumference and size (area) were calculated using National Institutes of Health Image 1.63 (n ⫽ 5).

Assay of progesterone in culture media The concentrations of progesterone in the culture medium were measured using RIA kits (Immunotech, Marseille, France). Inter- and intraassay coefficients of variation were 5.7 and 5.3%, respectively.

Matrigel invasion assay The invasion assay was carried out as previously described (26). A 6.4-mm-diameter culture insert with an 8-␮m-pore membrane filter (Becton Dickinson) was placed in collagen type I-coated 24-well plates (Asahi Techno Glass, Tokyo, Japan). The upper surface of the membrane filter was precoated with diluted Matrigel (Becton Dickinson; 300 ␮g/ ml). The lower well was filled with 700 ␮l RPMI 1640 (1% FCS) with or without (control) platelets (2 ⫻ 108 cells/well) and/or granulosa cells (1 ⫻ 105 cells/well) in the presence or absence of HCG (5 U/ml). Then isolated HUVECs (2.5 ⫻ 105 cells per 300 ␮l of RPMI 1640 with 1% FCS) were inoculated into the upper chamber. After a 3-h incubation at 37 C, HUVECs that reached the lower surface were fixed with 100% methanol at ⫺20 C for 5 min and were FITC stained using anti-CD141 mAb. The stained filters were examined under a confocal laser-scanning microscope and the numbers of CD141-positive cells were counted for quantification using National Institutes of Health Image 1.63 (26). These experiments were performed in triplicate (n ⫽ 7), and the average was defined as the invading cell number. Each result was expressed as the percentage of invading cell numbers found in the control (without coculture or additives).

Proliferation assay HUVECs were cultured in the intercell chambers for 48 h in the presence or absence of granulosa cells (1 ⫻ 105 cells/well), HCG (5 U/ml), and/or platelets (1 ⫻ 108 cells/well). The number of HUVECs in each intercell chamber was assessed using the Premix WST-1 cell proliferation assay system (Takara, Kusatsu, Japan) and ELISA plate reader (Molecular Device, Menlo Park, CA) according to the manufacturer’s instructions.

Flow cytometry Flow cytometry was performed as described previously (3). Detached granulosa cells (n ⫽ 5) cultured with or without HCG (5 U/ml) or platelets (1 ⫻ 108 cells/ml) were reacted with antihuman MCAM, integrin ␣5, or control mAb (100 ␮g/ml, 10 ␮l) and then with FITCconjugated rabbit pAb. Cell surface labeling was analyzed using a FACScalibur (Becton Dickinson).


Statistics Data are shown as means ⫾ sem. The concentration of progesterone in the culture medium, average length to width ratio, cell circumference and size of cultured luteinizing granulosa cells, cell numbers of HUVECs, and the mean intensity in flow cytometry were analyzed by ANOVA, followed by Scheffe´’s F test for multiple comparison. The difference was considered significant at P ⬍ 0.05.

Results Immunohistochemical localization of CD41-positive platelets

In CL on d 2 (n ⫽ 3), edematous changes induced by inflammatory reaction during ovulation remained evident, and red blood cells were located among luteinizing granulosa cells in all samples (Fig. 1, A–C). In the CL on d 3 (n ⫽ 6), extravascular blood was still evident among the sparsely lining granulosa cells that were not yet fully luteinized (Fig. 1D). In another sample, fresh bleeding into the antral cavity was observed from some vessels that had started to penetrate into the granulosa cell layer (Fig. 1, E and F). In this stage, vascular structures were detected in the luteinizing granulosa cell layer (Fig. 2, A and B). In the later stage of CL d 3, although intraantral bleeding was limited to within the peripheral site of the cavity by fibrin network, extravascular blood was still observed among the granulosa cells that became more luteinized (Fig. 1, G and H). To coincide with the above observation, CD41-positive platelets were diffusely observed among 3␤-hydroxysteroid dehydrogenase-positive luteinizing granulosa cells (Fig. 3, A and B). In the CL on d 4 (n ⫽ 4), endothelial cell migration almost reached the central area, and abundant extravascular blood was observed among luteinizing granulosa cells (Fig. 2, C and D). In the later stage of CL d 4, uncoagulated blood red cells were still observed at extravascular sites among luteinizing granulosa cells, which had considerably increased cytoplasmic volume and came into contact with each other (Fig. 1, I and J). In this stage, platelet deposition among luteinizing granulosa cells was mainly observed in the central area of the CL (Fig. 3, C and D). In the CL on d 5 (n ⫽ 4), granulosa cell luteinization proceeded further, and small luteal cells could be distinguished from large luteal cells (Fig. 1K). Extravascular blood lakes that contained degenerative materials were sparsely observed among large luteal cells (Fig. 1L). In this stage, platelet deposition was almost limited to within the central cavity (Fig. 3, E and F). In the CL on d 7– 8 (n ⫽ 5), vascular networks had been established (Fig. 2E), and vascular anastomosis in the central cavity was observed (Fig. 2F). In this stage, there were few extravascular blood lakes and no platelet deposition among large luteal cells or in extravascular spaces was observed (Fig. 3G). In the CL on d 3, CD 41-positive platelets were deposited in the ECM showing fibronectin and collagen type IV around luteinizing granulosa cells (Fig. 3H). In the CL on d 4, a fibrin net was mainly observed in the central cavity but was sparsely detected among luteinizing granulosa cells, as described previously (Fig. 3I) (2). In the CL on d 5, thrombomodulin/CD141-positive endothelial cells were observed to



3059

FIG. 1. Extravascular blood among luteinizing granulosa cells during human CL formation. A–C, Two samples of CL on the day after ovulation (CL d 2); D–H, Three samples of CL d 3; I and J, CL d 4; K and L, CL d 5. C, F, H, J, and L are magnified images of the boxed areas in B, E, G, I, and K. A–C, Inflammatory edema remained in the peripheral stroma (St) and abundant red blood cells were observed around luteinizing granulosa cells (LGC). D, Extravascular blood was still evident among the sparsely lining granulosa cells that had not yet undergone luteinization. E and F, Fresh bleeding into the antral cavity was observed. G and H, Intraantral bleeding was limited within the peripheral site of the cavity. Extravascular blood was still observed among the granulosa cells that became more luteinized. I and J, Uncoagulated blood red cells were still observed at extravascular sites among luteinizing granulosa cells that were considerably increased in cytoplasmic volume, and in contact with each other. K and L, Granulosa cell luteinization proceeded further and small luteal cells (SL) could be distinguished from large luteal cells (LL). Extravascular blood lakes that contained degenerative materials were sparsely observed among large luteal cells. Bar 100 ␮m. CC, Central cavity.

migrate through the luteinizing granulosa cell layer into the central cavity, forming vascular anastomoses (Fig. 3, J and K). Platelets that were deposited in the central region expressed P-selectin/CD62P (Fig. 3, L and M), showing that these platelets were activated. The effects of platelets on the morphology of luteinizing granulosa cells

In the presence of HCG, granulosa cells became round or oval, a change that resembles luteal cell transformation, FIG. 2. Neovascularization during human CL formation. A and B, CL d 3; C and D, CL d 4; E and F, CL d 7. B, D, and F are magnified images of the boxed areas in A, C, and E. A and B, Red blood cells at extravascular sites were observed around luteinizing granulosa cells (LGL), peripheral stroma (St), and central cavity (CC). B, Microvessels (arrows) were observed in the middle area. C and D, Note red blood cells at extravascular sites around the further luteinized granulosa cells. D, Microvessels (arrow) were observed in the central area. E and F, Granulosa cells were already transformed into large luteal cells (LL), being enlarged and tightly contacting each other. There were no red cells at extravascular sites and microvessels (arrows) were observed around the large luteal cells. Anastomoses of microvessels were established in the central cavity (arrowheads). Bar, 100 ␮m. LGC, Luteinizing granulosa cells; SL, small luteal cells.

whereas in the presence of platelets, spreading of luteinizing granulosa cells was enhanced during 48 h of culture (Fig. 4, A–C). Supporting this observation, both the calculated cell areas and circumferences of cultured granulosa cells were significantly reduced by HCG, whereas these parameters were promoted by platelets (Fig. 4, E and F), indicating that there are functional differences between HCG and platelets in the effects on the morphological changes of luteinizing granulosa cells. Under conditions of direct contact with platelets, similar microscopic morpho-

3060



FIG. 3. Immunofluorescence staining of human CL. A–G, Double staining using anti-CD41 mAb (red stained using rhodamine) and anti-3␤-hydroxysteroid dehydrogenase pAb (green stained using FITC). In the CL on d 3 (A and B), CD-41positive platelets were diffusely observed among 3␤-hydroxysteroid dehydrogenase-positive luteinizing granulosa cells. On d 4 (C and D), platelet deposition among luteinizing granulosa cells was observed only in the central area. On d 5 (E and F), platelets were mainly observed in the central cavity. In the CL on d 7 (G), platelets were hardly observed. On d 3 (H), platelets (green stained) were deposited in the collagen type IV (red stained) around luteinizing granulosa cells. On d 4 (I), the fibrin net (red stained) was mainly observed in the central cavity but was also sparsely detected among the luteinizing granulosa cells. J–M, On d 5 (J and K) thrombomodulin/ CD141-positive endothelial cells (red stained) migrated through the luteinizing granulosa cell layer into the central cavity, forming vascular anastomoses. L, Platelets (green stained) deposited in the central region expressed P-selectin/ CD62P (red stained). M is a higher magnification of the boxed area in L. Bar, 100 ␮m. CC, Central cavity; LGC, luteinizing granulosa cells; LL, large luteal cells.

logical changes were induced in luteinizing granulosa cells (Fig. 4D). Flow cytometry showed that the expression levels of MCAM and integrin ␣5 cell surface markers for luteinization, on granulosa cells were increased by HCG, as described previously (4, 14), but not by platelets (Fig. 4, G and H). The effects of platelets on progesterone production by luteinizing granulosa cells

During 2 d of culture, although the level of progesterone production in the groups that were treated with HCG and platelets seemed higher than that in the control group (without HCG or platelets), the differences were not significant (Fig. 5A). During an additional 2-d culture, progesterone production by luteinizing granulosa cells was significantly elevated under conditions of direct contact with platelets, as observed in the group treated with HCG (Fig. 5B). Under conditions of indirect contact using intercell chambers, there

was no significant enhancement of progesterone production (Fig. 5B). The effects of platelets and luteinizing granulosa cells on endothelial cell migration and proliferation

To investigate the in vitro effects of both luteinizing granulosa cells and platelets on endothelial cell function we used HUVECs because it is very difficult to obtain a sufficient amount of migrating endothelial cells from human CL tissue. In cocultures with platelets, the number of migrated HUVECs was significantly enhanced (Fig. 6A). However, luteinizing granulosa cells, which had been reported to produce angiogenic factors, showed little effect on HUVEC migration, even with stimulation by HCG. Unexpectedly, the enhancing effect of platelets was attenuated by luteinizing granulosa cells. Endothelial cell proliferation was promoted by both lu-



3061

FIG. 4. Platelet-induced morphological changes in cultured granulosa cells. Human granulosa cells were cultured for 48 h in the absence (A) or presence of HCG (B), and platelets without (C) or with (D) direct interaction. Granulosa cells became round in the presence of HCG, whereas platelets induced spreading in granulosa cells with or without direct interaction. Bar, 50 ␮m. Both calculated cell areas (E) and circumferences (F) were significantly reduced by HCG but increased by platelets under indirect contact conditions. As shown by flow cytometry, the expression levels of MCAM (G) and integrin ␣5 (H) were enhanced by HCG treatment, whereas these expression levels were not affected by platelets. **, P ⬍ 0.01.CTR, Control.

teinizing granulosa cells and platelets. There were no significant differences observed among the treated groups (Fig. 6B). Discussion

In this study, immunohistochemical examination indicated that platelet deposition was highly correlated with the angiogenic and luteinizing processes in human CL, suggesting that platelets are involved in human CL formation. Uncoagulated blood was observed at extravascular sites among luteinizing granulosa cells in all specimens derived from the early luteal phase, confirming that human CL formation is an intriguing model of tissue remodeling. The distribution of fibrin, which reflects the local status of the coagulatory system, was also positively correlated with platelet localization. Platelets deposited in the extravascular spaces were shown to be in contact with the ECM. In general, direct interaction with the extracellular matrix induces platelets to release biologically active substances in cytoplasmic granules (27). In support of this, the deposited platelets showed coexpression of CD62P/P-selectin. From these findings, it is speculated that platelets are activated during CL formation in vivo and that substances derived from platelets play some roles in constructing a new endocrine organ. To investigate the physiological role of platelets in the luteinization of granulosa cells, we examined the effects of platelets on progesterone production by cultured luteinizing

granulosa cells. In the presence of HCG, progesterone production was significantly enhanced during 4-d culture. Notably, coculture with platelets also promoted progesterone production by granulosa cells, suggesting that platelets facilitate luteinization of granulosa cells. This study also showed that coculturing with platelets induced marked spreading of granulosa cell morphology, suggesting that platelets enhanced the adhesive property of granulosa cells. Previously we reported that luteinizing granulosa cells showed increased expression of MCAM (14), which was reported to regulate cell attachment to endothelial cells (13), and integrin ␣5, which is a receptor for fibronectin (4). Although HCG enhances the expression of these molecules, there was no significant change in MCAM or integrin ␣5 expression by coculturing with platelets in this study. Thus, although the increase in size induced by platelets is compatible with the morphological change during the luteinization process, the mechanism may differ from that induced by HCG. In contrast to their effects on progesterone production, morphological changes elicited by platelets did not require direct contact with granulosa cells, suggesting that platelets affect granulosa cell luteinization through more than a single pathway. For more than a decade, the contribution of white blood cells to CL function has been proposed (28). However, there has not been any report concerning the direct effects of platelets on human luteal function. Although this study did not

3062



FIG. 5. Platelet-induced progesterone production in cultured granulosa cells. During the 2-d culture (A), progesterone production tended to be increased by HCG and platelet treatment. During the additional 2-d culture (B), progesterone production was significantly promoted in the groups cultured with HCG and by direct contact with platelets. However, under conditions of indirect contact using intercell chambers, there was no significant enhancement. *, P ⬍ 0.05, **, P ⬍ 0.01.

provide precise information about the mechanism by which platelets promote progesterone production, it can be proposed that platelets are novel local regulators of the luteinization of human granulosa cells. Then we examined the effects of platelets on endothelial cell migration using HUVECs and a Matrigel invasion assay. In the human early CL, it was reported that angiogenic factors such as VEGF and endocrine gland-VEGF were produced by luteinizing granulosa cells, and their production was promoted by LH/HCG stimulation (29, 30). Although these factors were proposed to contribute to endothelial cell proliferation and maintenance of the vascular structure in human CL, there is no evidence showing that granulosa cells enhance endothelial cell migration. In this study, endothelial cell migration was not enhanced by granulosa cells. Even in the presence of HCG, there was no significant effect on endothelial migration. However, when platelets were cocultured with endothelial cells, endothelial cell migration was promoted (31). Although there are some functional differences between endothelial cells in umbilical veins and ovulating follicles (32), these findings suggest that platelets are more potent stimulators of endothelial cell migration than luteinizing granulosa cells. Platelets contain several factors such as VEGF and sphin-

FIG. 6. The effects of platelets on endothelial cell migration and proliferation. In cocultures with platelets, the number of migrated endothelial cells was enhanced (A). These enhancing effects of platelets were attenuated in the presence of luteinizing granulosa cells (LGC). In contrast, endothelial cell migration was not affected by coculturing with granulosa cells in the presence or absence of HCG. The proliferation of endothelial cells was slightly promoted by both luteinizing granulosa cells and platelets in the presence or absence of HCG (B). There were no significant differences observed among the treated groups. *, P ⬍ 0.05, **, P ⬍ 0.01.

gosine 1-phosphate, which may contribution to vascular extension (33). When cocultured with platelets, granulosa cells unexpectedly showed inhibition of endothelial cell migration induced by platelets in the presence or absence of HCG. These findings may be accorded with a recent report that culturing of HUVECs with conditioned medium from cultured human luteinized granulosa cells lead to the expression of antiangiogenic factors at the transcript level in endothelial cells (6). To achieve fine and mature vascular networks among fully luteinized luteal cells, the processes of vascularization and luteinization as well as the arrangement of ECM should be synchronized. These findings may reflect a crucial role of granulosa cells in regulating adequate neovascularization by protecting against excessive stimulatory effects of platelets on endothelial migration. Ovulation and subsequent corpus luteum formation are considered to mimic an inflammatory reaction (34). Accumulating evidence shows that there is a cross-talk between inflammation and coagulation systems (35), whereby inflammation not only leads to the activation of coagulation, but coagulation also considerably affects inflammatory activity (36). This inflammatory reaction is also considered to elicit cell migration and proliferation in cooperation with the coagulation system (35). This study provides further evidence



3063

FIG. 7. The proposed roles of platelets in human CL formation. Platelets are mainly deposited at the central areas facing the central cavity in which a fibrin net has been formed. These platelets are activated and create a gradient of angiogenic factor concentration from the central cavity to the peripheral stromal layer, inducing endothelial cell migration toward the central cavity. Platelets are also directly in contact with luteinizing granulosa cells, contributing to luteinization. Until the establishment of fine vascular networks among luteinizing granulosa cells, extravascular fluid is speculated to circulate around granulosa cells and recruit progesterone hormone to the systemic blood circulation although it remains unknown how anticoagulant mechanisms are operating in CL during the tissue remodeling process.

that platelets, which are the main contributors to the coagulation system, induce cell differentiation and migration in the tissue remodeling process in the adult human ovary. In conclusion, this study showed that the extravascular localization of platelets accords with the neovascularization process during human CL formation. The direct interaction of platelets with granulosa cells was demonstrated to promote progesterone production by granulosa cells. In addition, platelet-derived soluble factors induced morphological changes in granulosa cells, suggesting that platelets are involved in the process of differentiation of human granulosa cells toward large luteal cells. Furthermore, platelet-derived soluble factors were shown to be a greater stimulant of endothelial migration than granulosa cells. Although it remains unknown how the coagulation system is controlled to maintain adequate kinetics of tissue fluid among luteinizing granulosa cells, the present results lead us to propose a novel concept whereby platelets regulate spatiotemporal construction of vascular networks in the early human CL (Fig. 7). These findings also support the recent concept that platelets play an important role in wound healing processes and will contribute to clarifying the mechanism of extravascular circulation in inflammatory lesions. Acknowledgments The authors are grateful to Ms. Mizuho Takemura for excellent technical assistance. Received December 15, 2006. Accepted April 9, 2007. Address all correspondence and requests for reprints to: Hiroshi Fujiwara, M.D., Department of Gynecology and Obstetrics, Faculty of Medicine, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan. E-mail: [email protected]. This work was supported by Grants-in-Aid for Scientific Research (16390474 and 17591731). Disclosure Statement: The authors have nothing to disclose.

References 1. Shimada H, Kasakura S, Shiotani M, Nakamura K, Ikeuchi M, Hoshino T, Komatsu T, Ihara Y, Sohma M, Maeda Y, Matsuura R, Nakamura S, Hine C, Ohkura N, Kato H 2001 Hypocoagulable state of human preovulatory ovarian follicular fluid: role of sulfated proteoglycan and tissue factor pathway inhibitor in the fluid. Biol Reprod 64:1739 –1745 2. Corner GW 1956 The histological dating of the human corpus luteum of menstruation. Am J Anat 98:377– 401 3. Fujiwara H, Honda T, Ueda M, Nakamura K, Yamada S, Maeda M, Mori T 1997 Laminin suppresses progesterone production by human luteinizing granulosa cells via interaction with integrin ␣6␤1. J Clin Endocrinol Metab 82: 2122–2128 4. Honda T, Fujiwara H, Yamada S, Fujita K, Nakamura K, Nakayama T, Higuchi T, Ueda M, Maeda M, Mori T 1997 Integrin ␣5 is expressed on human luteinizing granulosa cells during corpus luteum formation, and its expression is enhanced by human chorionic gonadotrophin in vitro. Mol Hum Reprod 3:979 –984 5. Yamada S, Fujiwara H, Honda T, Higuchi T, Nakayama T, Inoue T, Maeda M, Fujii S 1999 Human granulosa cells express integrin ␣2 and collagen type IV: possible involvement of collagen type IV in granulosa cell luteinization. Mol Hum Reprod 5:607– 617 6. Gruemmer R, Klein-Hitpass L, Neulen J 2005 Regulation of gene expression in endothelial cells: the role of human follicular fluid. J Mol Endocrinol 34: 37– 46 7. Yan Z, Weich HA, Bernart W, Breckwoldt M, Neulen J 1993 Vascular endothelial growth factor (VEGF) messenger ribonucleic acid (mRNA) expression in luteinized human granulosa cells in vitro. J Clin Endocrinol Metab 77:1723– 1725 8. Koga K, Osuga Y, Tsutsumi O, Momoeda M, Suenaga A, Kugu K, Fujiwara T, Takai Y, Yano T, Taketani Y 2000 Evidence for the presence of angiogenin in human follicular fluid and the up-regulation of its production by human chorionic gonadotropin and hypoxia. J Clin Endocrinol Metab 85:3352–3355 9. Fraser HM, Bell J, Wilson H, Taylor PD, Morgan K, Anderson RA, Duncan WC 2005 Localization and quantification of cyclic changes in the expression of endocrine gland vascular endothelial growth factor in the human corpus luteum. J Clin Endocrinol Metab 90:427– 434 10. Sugino N, Suzuki T, Sakata A, Miwa I, Asada H, Taketani T, Yamagata Y, Tamura H 2005 Angiogenesis in the human corpus luteum: changes in expression of angiopoietins in the corpus luteum throughout the menstrual cycle and in early pregnancy. J Clin Endocrinol Metab 90:6141– 6148 11. Kamat BR, Brown LF, Manseau EJ, Senger DR, Dvorak HF 1995 Expression of vascular permeability factor/vascular endothelial growth factor by human granulosa and theca lutein cells. Role in corpus luteum development. Am J Pathol 146:157–165 12. Egawa M, Yoshioka S, Higuchi T, Sato Y, Tatsumi K, Fujiwara H, Fujii S 2003 Ephrin B1 is expressed on human luteinizing granulosa cells in corpora lutea of the early luteal phase: the possible involvement of the B class Eph-ephrin

3064

13. 14.

15. 16. 17. 18. 19. 20. 21. 22.

23.


system during corpus luteum formation. J Clin Endocrinol Metab 88:4384 – 4392 Xie S, Luca M, Huang S, Gutman M, Reich R, Johnson JP, Bar-Eli M 1997 Expression of MCAM/MUC18 by human melanoma cells leads to increased tumor growth and metastasis. Cancer Res 57:2295–2303 Yoshioka S, Fujiwara H, Higuchi T, Yamada S, Maeda M, Fujii S 2003 Melanoma cell adhesion molecule (MCAM/CD146) is expressed on human luteinizing granulosa cells: enhancement of its expression by hCG, interleukin-1 and tumour necrosis factor-␣. Mol Hum Reprod 9:311–319 Davis GE, Senger DR 2005 Endothelial extracellular matrix: biosynthesis, remodeling, and functions during vascular morphogenesis and neovessel stabilization. Circ Res 97:1093–1107 Brewer JI 1942 Studies of the human corpus luteum. Am J Obstet Gynecol 44:1048 –1062 Balboni GC 1983 Structural changes: ovulation and luteal phase. In: Serra GB, ed. The ovary. New York: Raven Press; 123–141 Woodruff JD, Parmley TH, Kurman RJ 1988 The ovary. In: Woodruff JD, Parmley TH, Kurman RJ, eds. Atlas of gynecologic pathology. Philadelphia: Lippincott Co.; 7.1–7.47 Boehlen F, Clemetson KJ 2001 Platelet chemokines and their receptors: what is their relevance to platelet storage and transfusion practice? Transfus Med 11:403– 417 Anitua E, Andia I, Ardanza B, Nurden P, Nurden AT 2004 Autologous platelets as a source of proteins for healing and tissue regeneration. Thromb Haemost 91:4 –15 Weber C 2005 Platelets and chemokines in atherosclerosis: partners in crime. Circ Res 96:612– 616 Fujiwara H, Yoshioka S, Tatsumi K, Kosaka K, Satoh Y, Nishioka Y, Egawa M, Higuchi T, Fujii S 2002 Human endometrial epithelial cells express ephrin A1: possible interaction between human blastocysts and endometrium via Eph-ephrin system. J Clin Endocrinol Metab 87:5801–5807 Sato Y, Fujiwara H, Zeng BX, Higuchi T, Yoshioka S, Fujii S 2005 Plateletderived soluble factors induce human extravillous trophoblast migration and differentiation: platelets are a possible regulator of trophoblast infiltration into maternal spiral arteries. Blood 106:428 – 435


24. Walkowiak B, Kralisz U, Michalec L, Majewska E, Koziolkiewicz W, Ligocka A, Cierniewski CS 2000 Comparison of platelet aggregability and P-selectin surface expression on platelets isolated by different methods. Thromb Res 99:495–502 25. Jaffe EA, Nachman RL, Becker CG, Minick CR 1973 Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest 52:2745–2756 26. Sato Y, Higuchi T, Yoshioka S, Tatsumi K, Fujiwara H, Fujii S 2003 Trophoblasts acquire a chemokine receptor, CCR1, as they differentiate towards invasive phenotype. Development 130:5519 –5532 27. Nieswandt B, Watson SP 2003 Platelet-collagen interaction: is GPVI the central receptor? Blood 102:449 – 461 28. Adashi EY 1990 The potential relevance of cytokines to ovarian physiology: the emerging role of resident ovarian cells of the white blood cell series. Endocr Rev 11:454 – 464 29. Otani N, Minami S, Yamoto M, Shikone T, Otani H, Nishiyama R, Otani T, Nakano R 1999 The vascular endothelial growth factor/fms-like tyrosine kinase system in human ovary during the menstrual cycle and early pregnancy. J Clin Endocrinol Metab 84:3845–3851 30. Fraser HM, Wulff C 2003 Angiogenesis in the corpus luteum. Reprod Biol Endocrinol 1:88:1– 8 31. Kim HK, Song KS, Chung JH, Lee KR, Lee SN 2004 Platelet microparticles induce angiogenesis in vitro. Br J Haematol 124:376 –384 32. Davis JS, Rueda BR, Spanel-Borowski K 2003 Microvascular endothelial cells of the corpus luteum. Reprod Biol Endocrinol 1:89:1–15 33. Lee OH, Kim YM, Lee YM, Moon EJ, Lee DJ, Kim JH, Kim KW, Kwon YG 1999 Sphingosine 1-phosphate induces angiogenesis: its angiogenic action and signaling mechanism in human umbilical vein endothelial cells. Biochem Biophys Res Commun 264:743–750 34. Espey LL 1994 Current status of the hypothesis that mammalian ovulation is comparable to an inflammatory reaction. Biol Reprod 50:233–238 35. Esmon CT 2005 The interactions between inflammation and coagulation. Br J Haematol 131:417– 430 36. Levi M, van der Poll T, Buller HR 2004 Bidirectional relation between inflammation and coagulation. Circulation 109:2698 –2704

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.