Subpopulations Of Granulosa Cells Research Articles

EP3 receptor isoforms are differentially expressed in subpopulations of primate granulosa cells and couple to unique G-proteins

Prostaglandin E2 (PGE2) produced within the ovarian follicle is necessary for ovulation. PGE2 is recognized by four distinct G-protein-coupled receptors. Among them, PTGER3 (also known as EP3) is unique in that mRNA splicing generates multiple isoforms. Each isoform has a distinct amino acid composition in the C-terminal region, which is involved in G-protein coupling. To determine whether monkey EP3 isoforms couple to different G-proteins, each EP3 isoform was expressed in Chinese hamster ovary cells, and intracellular signals were examined after stimulation with the EP3 agonist sulprostone. Stimulation of EP3 isoform 5 (EP3-5) reduced cAMP in a pertussis toxin (PTX)-sensitive manner, indicating involvement of Gαi. Stimulation of EP3-9 increased cAMP, which was reduced by the general G-protein inhibitor GDP-β-S, and also increased intracellular calcium, which was reduced by PTX and GDP-β-S. So, EP3-9 likely couples to both Gαs and a PTX-sensitive G-protein to regulate intracellular signals. Stimulation of EP3-14 increased cAMP, which was further increased by PTX, so EP3-14 likely regulates cAMP via multiple G-proteins. Granulosa cell expression of all EP3 isoforms increased in response to an ovulatory dose of human chorionic gonadotropin. Two EP3 isoforms were differentially expressed in functional subpopulations of granulosa cells. EP3-5 was low in granulosa cells at the follicle apex while EP3-9 was high in cumulus granulosa cells. Differential expression of EP3 isoforms may yield different intracellular responses to PGE2 in granulosa cell subpopulations, contributing to the different roles played by granulosa cell subpopulations in the process of ovulation.

Plasticity of granulosa cells: on the crossroad of stemness and transdifferentiation potential

The ovarian follicle represents the basic functional unit of the ovary and consists of an oocyte, which is surrounded by granulosa cells (GCs). GCs play an important role in the growth and development of the follicle. They are subject to increased attention since it has recently been shown that the subpopulation of GCs within the growing follicle possesses exceptionally plasticity showing stem cell characteristics. In assisted reproduction programs, oocytes are retrieved from patients together with GCs, which are currently discarded daily, but could be an interesting subject to be researched and potentially used in regenerative medicine in the future. Isolated GCs expressed stem cell markers such as OCT-4, NANOG and SOX-2, showed high telomerase activity, and were in vitro differentiated into other cell types, otherwise not present within ovarian follicles. Recently another phenomenon demonstrated in GCs is transdifferentiation, which could explain many ovarian pathological conditions. Possible applications in regenerative medicine are also given.

EP3 Receptor Isoforms Differentially Expressed in Subpopulations of Primate Granulosa cells Couple to Unique G Proteins.

Prostaglandin (PG) E2 is necessary for ovulation of mammalian follicles. PGE2 acts through four distinct G protein coupled receptors (PTGER1-PTGER4, also called EP1-EP4). Among them, EP3 is unique in having multiple isoforms generated by alternative mRNA splicing. Each isoform has a distinct amino acid composition in the C-terminal region which couples to G proteins. Therefore, we hypothesized that each EP3 isoform couples to different G proteins to modulate intracellular signaling in primate granulosa cells. Using 3' rapid amplification of cDNA ends (3' RACE), six EP3 mRNAs were identified in cynomolgus monkey granulosa cells. Three of these isoforms shared significant sequence homology with EP3 isoform 5 (EP3-5), EP3-7, and EP3-9 as described previously in human tissues. In addition, three isoforms were identified which have not been previously reported. EP3-5, EP3-7, and EP3-9 were transiently transfected into Chinese hamster ovary (CHO) cells and were detected in a punctate pattern in the cytoplasm as well as in the plasma membrane. Stimulation of EP3-5 by sulprostone, an EP3 agonist, reduced cAMP compared to vehicle-treated cells. Pretreatment with pertussis toxin (PTX) increased sulprostone-stimulated cAMP compared to sulprostone alone, indicating that EP3-5 couples to the Gi protein. Addition of sulprostone to EP3-9 transfected cells increased cAMP, which was blocked by the general G protein inhibitor GDP-β-S. Sulprostone also increased intracellular Ca2+ in EP3-9 transfected cells, which was blocked by PTX or GDP-β-S. Therefore, EP3-9 likely couples to both Gs to increase cAMP and a PTX-sensitive G protein to regulate intracellular Ca2+. No G protein associated function was identified for EP3-7. Laser capture microdissection and qPCR were used to identify subpopulations of granulosa cells which preferentially expressed these EP3 isoforms 36 hours after an ovulatory dose of hCG, or just before ovulation. EP3-5 mRNA was significantly lower in apex granulosa cells compared to non-apex granulosa cells; EP3-9 mRNA was higher in cumulus cells when compared to mural granulosa cells. In summary, each EP3 receptor isoform may function through different G proteins. Differential expression of EP3 isoforms in subpopulations of granulosa cells may contribute to the different roles that these subpopulations play in the process of ovulation. Supported by NIH/NICHD HD054691.

Prostaglandin E2 Receptors Are Differentially Expressed in Subpopulations of Granulosa Cells from Primate Periovulatory Follicles1

Prostaglandin E2 (PGE2) mediates many effects of the midcycle luteinizing hormone (LH) surge within the periovulatory follicle. Differential expression of the four PGE2 (EP) receptors may contribute to the specialized functions of each granulosa cell subpopulation. To determine if EP receptors are differentially expressed in granulosa cells, monkeys received gonadotropins to stimulate ovarian follicular development. Periovulatory events were initiated with human chorionic gonadotropin (hCG); granulosa cells and whole ovaries were collected before (0 h) and after (24-36 h) hCG to span the 40-h primate periovulatory interval. EP receptor mRNA and protein levels were quantified in granulosa cell subpopulations. Cumulus cells expressed higher levels of EP2 and EP3 mRNA compared with mural cells 36 h after hCG. Cumulus cell EP2 and EP3 protein levels also increased between 0 and 36 h after hCG. Overall, mural granulosa cells expressed low levels of EP1 protein at 0 h and higher levels 24-36 h after hCG. However, EP1 protein levels were higher in granulosa cells away from the follicle apex compared with apex cells 36 h after hCG. Higher levels of PAI-1 protein were measured in nonapex cells, consistent with a previous study showing EP1-stimulated PAI-1 protein expression in monkey granulosa cells. EP4 protein levels were low in all subpopulations. In summary, cumulus cells likely respond to PGE2 via EP2 and EP3, whereas PGE2 controls rupture of a specific region of the follicle via EP1. Therefore, differential expression of EP receptors may permit each granulosa cell subpopulation to generate a unique response to PGE2 during the process of ovulation.

A Novel Three-Dimensional Culture System Allows Prolonged Culture of Functional Human Granulosa Cells and Mimics the Ovarian Environment

The development of techniques allowing the growth of primordial follicles to mature follicles in vitro has much potential for both reproductive medicine and developmental research. However, human primordial and preantral follicles fail to grow after isolation from the surrounding ovarian stroma. Granulosa cells, which normally undergo apoptosis after ovulation, contain a subpopulation of ovarian follicular cells remaining viable in vitro over prolonged periods when cultured in the presence of leukemia-inhibiting factor. However, when cultured as monolayers, they progressively lose all their characteristics, such as follicle-stimulating hormone receptor and cytochrome P450-aromatase. Here, we describe a three-dimensional culture system containing type I collagen, which, together with leukemia-inhibiting factor, allowed the survival and growth of a subpopulation of granulosa cells isolated from mature ovarian follicles and supported them to proliferate into spherical structures exhibiting steroidogenic capacity, as demonstrated by P450-aromatase and 3beta-hydroxysteroid dehydrogenase. After transplantation into the ovaries of immunodeficient mice, these cells became localized preferentially within antral follicles and the prolonged expression of follicle-stimulating hormone receptor was confirmed as well. With this optimization of the culture conditions, an environment was created, which acts as a niche closely mimicking the development of early ovarian follicles in vitro.

The Multipotency of Luteinizing Granulosa Cells Collected from Mature Ovarian Follicles

Graafian ovarian follicles consist of follicular fluid, one single mature oocyte, and several hundred thousands of granulosa cells (GCs). Until now, luteinizing GCs have been considered to be terminally differentiated, destined to undergo death after ovulation. Present concepts of luteal function, endocrine regulation of early pregnancy, and the recruitment of new ovarian follicles are all based on the cyclical renewal of the entire population of GCs. We now demonstrate that luteinizing GCs isolated from the ovarian follicles of infertile patients and sorted with flow cytometry based upon the presence of their specific marker, the follicle-stimulating hormone receptor (FSHR), can be maintained in culture over prolonged periods of time in the presence of the leukemia-inhibiting factor (LIF). Under those conditions the markers of GC function such as FSHR and aromatase gradually disappeared. POU5F1 (POU domain, class 5, homeobox 1), a typical stem cell marker, was expressed throughout the culture, but germ line cell markers such as nanog, vasa, and stellar were not. Mesenchymal lineage markers such as CD29, CD44, CD90, CD105, CD117, and CD166, but not CD73, were expressed by substantial subpopulations of GCs. The multipotency of a subset of GCs was established by in vitro differentiation into other cell types, otherwise not present within ovarian follicles, such as neurons, chondrocytes, and osteoblasts. Follicle-derived stem cells were also able to survive when transplanted into the backs of immunoincompetent mice, in vivo generating tissues of mesenchymal origin. The unexpected findings of multipotency of cells with prolonged lifespans originating from ovarian follicles are likely to have a significant impact on evolving theories in ovarian pathophysiology, particularly with reference to ovarian endometriosis and ovarian cancer.

Coordinated and cell-specific induction of both physiological plasminogen activators creates functionally redundant mechanisms for plasmin formation during ovulation.

Several lines of indirect evidence indicate that plasmin-mediated proteolysis plays a role in the breakdown of the follicle wall during ovulation. Consistent with this, the ovulation efficiency of mice lacking the two known physiological plasminogen activators (PAs), tissue-type PA (tPA) and urokinase-type PA (uPA), is reduced by 26%. Surprisingly, mice with a single deficiency of either tPA or uPA gene function were normal in their capacity to ovulate. In this study we used in situ hybridization and casein in situ zymography to localize the expression of messenger RNAs (mRNAs) encoding PAs and PA inhibitors and to examine the net PA activity in the mouse ovary at the time of ovulation. Although uPA mRNA expressed by granulosa cells is the most abundant and dramatically up-regulated PA before ovulation, a previously unnoticed coordinated induction oftPA mRNA was found in thecal-interstitial tissue. The existence of redundant mechanisms for plasmin production in the ovary may be the cause of the normal ovulation efficiency in single deficient mice lacking tPA or uPA. The expression of mRNAs for PA inhibitors, types 1 and 2, was low in the ovary, with minor inductions at restricted time points. In contrast, expression of protease nexin-1 (PN-1) by granulosa cells was high during the entire periovulatory period. Among subpopulations of granulosa cells, the expression of PN-1 and uPA was heterogeneous and complementary. Cumulus cells expressed high levels of PN-1 mRNA and low levels of uPA mRNA, thereby providing an inhibitory activity that may protect the mucified matrix of the cumulus oocyte complex from proteolytic degradation.

Coordinated and cell-specific induction of both physiological plasminogen activators creates functionally redundant mechanisms for plasmin formation during ovulation

Several lines of indirect evidence indicate that plasmin-mediated proteolysis plays a role in the breakdown of the follicle wall during ovulation. Consistent with this, the ovulation efficiency of mice lacking the two known physiological plasminogen activators (PAs), tissue-type PA (tPA) and urokinase-type PA (uPA), is reduced by 26%. Surprisingly, mice with a single deficiency of either tPA or uPA gene function were normal in their capacity to ovulate. In this study we used in situ hybridization and casein in situ zymography to localize the expression of messenger RNAs (mRNAs) encoding PAs and PA inhibitors and to examine the net PA activity in the mouse ovary at the time of ovulation. Although uPA mRNA expressed by granulosa cells is the most abundant and dramatically up-regulated PA before ovulation, a previously unnoticed coordinated induction oftPA mRNA was found in thecal-interstitial tissue. The existence of redundant mechanisms for plasmin production in the ovary may be the cause of the normal ovulation efficiency in single deficient mice lacking tPA or uPA. The expression of mRNAs for PA inhibitors, types 1 and 2, was low in the ovary, with minor inductions at restricted time points. In contrast, expression of protease nexin-1 (PN-1) by granulosa cells was high during the entire periovulatory period. Among subpopulations of granulosa cells, the expression of PN-1 and uPA was heterogeneous and complementary. Cumulus cells expressed high levels of PN-1 mRNA and low levels of uPA mRNA, thereby providing an inhibitory activity that may protect the mucified matrix of the cumulus oocyte complex from proteolytic degradation.

Expression and alternative splicing of the neural cell adhesion molecule NCAM in human granulosa cells during luteinization

Freshly aspirated human granulosa cells from pre-ovulatory follicles and granulosa cells luteinized in culture possess the neural cell adhesion molecule (NCAM) of approximate molecular mass of 140,000 and NCAM mRNA as confirmed by S1-nuclease protection assays and RT-PCR. Moreover, in the process of luteinization the NCAM isoform pattern is modified. Isoforms containing an insert of 10 amino acids (termed VASE) in the extracellular domain of NCAM were supplemented by alternatively spliced isoforms without this insert. NCAM immunoreactivity, at light and electron microscope levels, was associated with the cell membrane of most granulosa cells which formed clusters. During time in culture an increasing subpopulation of granulosa cells, devoid of NCAM immunoreactivity, spread out and formed monolayers. This differential expression and the alternative splicing of NCAM during luteinization of granulosa cells raise the possibility that NCAM could be involved in folliculogenesis and the formation of the corpus luteum in the human.

Flow cytometric determination of degraded deoxyribonucleic acid in granulosa cells to identify atretic follicles during preovulatory maturation in the pig.

Granulosa cells of individual follicles were analyzed by DNA fluorescence flow cytometry to determine how the percentage of cells with degraded DNA and the distribution of cells in the phases of the cell cycle (Go/G1, S1, G2/M) related to the incidence of morphological atresia and to changes in follicular steroid concentrations. Follicles were dissected from ovaries recovered at slaughter on Days 1, 3, 5, or 7 of altrenogest-synchronized preovulatory maturation. Twenty-one follicles with debris among their isolated granulosa cells were classified as morphologically atretic (MA); 92 follicles with debris-free granulosa cells were classified as morphologically healthy (MH). Granulosa cells were prepared for flow cytometry by fixation in 80% ethanol and staining with propidium iodide (PI) containing RNase. DNA fluorescence intensity was determined by use of the 488-nm line of an argon laser. A subpopulation of granulosa cells with degraded DNA (Ao cells), containing less fluorescence than the Go/G1 peak, was found in the DNA histogram of every follicle. The percentage of Ao cells ranged from 0.02 to 83.6% per follicle. The percentage of Ao cells was inversely related to the percentage of Go/G1 cells (r = -0.9611, p = 0.0001). The percentage of Ao cells (mean +/- SEM) was greater (p = 0.0001) in MA (45.9 +/- 6.3%) than in MH follicles (5.3 +/- 1.6%). Follicular estradiol-17 beta was less in MA than in MH follicles, but androstenedione or progesterone did not differ significantly.(ABSTRACT TRUNCATED AT 250 WORDS)

Localization of Luteinizing Hormone Receptor Messenger Ribonucleic Acid Expression in Ovarian Cell Types during Follicle Development and Ovulation*

The action of LH is mediated through specific plasma membrane receptors that are both up- and down-regulated in the ovary during the reproductive cycle. Using immature rats treated with PMSG and hCG as a model system, we have studied the regulation and distribution of LH receptor mRNA in different cell types during follicle development, ovulation, and luteinization by Northern blot and in situ hybridization. In untreated rats, LH receptor mRNA was below the detection level in granulosa cells, cumulus cells, and oocytes, while low levels of LH receptor mRNA were found in the thecal cells. After stimulation with PMSG, expression of LH receptor mRNA was enhanced in the thecal-interstitial cells, while a more dramatic increase in receptor mRNA abundance took place in granulosa cells of large tertiary follicles. In these follicles, the abundance of LH receptor mRNA varied among different subpopulations of granulosa cells, with mural granulosa cells close to the basement membrane exhibiting higher levels than granulosa cells located closer to the antrum, and cumulus cells and the oocyte lacking detectable hybridization signal. The uneven expression of LH receptor mRNA endows different ovarian cells with varying hormonal responsiveness. After an ovulatory dose of hCG, LH receptor mRNA levels were dramatically decreased, particularly in the granulosa cells of preovulatory follicles, to reach the lowest levels just before ovulation. During the transformation of ovulated follicles into corpora lutea, the expression of LH receptor message was again increased. Our results reveal that the previously documented regulation of the LH receptor-binding activity during ovarian development correlates with expression of the LH receptor transcripts, suggesting that the LH receptor gene is regulated in a complex manner during the periovulatory period to achieve cell-specific expression together with gonadotropin induction and suppression of receptor gene activity.

Subpopulations of granulosa cells within the human ovarian follicle

Human follicular cells were separated according to their isopycnic densities. Three populations were isolated and identified in terms of their secretion of progesterone and oestradiol. Cells in the least dense population secreted approximately 60% as much progesterone and 20% as much oestradiol per cell as did cells in the two denser bands. It is proposed that cumulus cells compose the least dense band and that another band may be made up of antral cells.

A monoclonal antibody to a mammary cell line recognizes two distinct subtypes of ovarian granulosa cells.

When screening supernatant fluids from hybridoma clones, Dulbecco and co-workers found that a mouse monoclonal antibody generated against a mammary tumor cell line showed rather striking high binding to rat oocytes and granulosa cells. In this study we have specifically investigated the reactivity of the monoclonal antibody (designated anti-OA-1) with granulosa cells during the differentiation process. This was accomplished using the two-step indirect immunocytochemical technique. When a primordial follicle is recruited to initiate growth, intense immunoreactivity appears in the surface membrane of the granulosa cells. As a follicle proceeds through the preantral stages, the plasma membrane of the granulosa cells is strongly positive for anti-OA-1 reactivity, and the granulosa appear as a homogeneous population. However, once Graafian follicle development is initiated, a major shift in anti-OA-1 immunoreactivity occurs among the granulosa cells. As the antrum expands, 75% of the granulosa in the mural region (those nearest the basal lamina) elongate and become negative for anti-OA-1. This is in contrast to the periantral and cumulus granulosa, which remain rounded and show strong anti-OA-1 reactivity up to the preovulatory stage. The disappearance of anti-OA-1 reactivity in the subpopulation of mural cells is specifically initiated by FSH and occurs very rapidly after a 12-h lag phase. After the loss of anti-OA-1 reactivity, the elongated mural granulosa cells express their terminal differentiated state by acquiring LH/hCG receptor, 3 beta-hydroxysteroid dehydrogenase activity and cytoplasmic lipid inclusions. By contrast, the periantral and cumulus granulosa, which remain positive for anti-OA-1, do not express these differentiated functions; however, they do differentiate ultrastructurally, indicating that they respond to the FSH signal. These results strongly suggest that a monoclonal antibody recognizes a major surface differentiation antigen in the granulosa cell. This antigen is under hormonal control and is inversely linked to expression of the terminal differentiation program in the granulosa cells. We anticipate that the monoclonal antibody will be a valuable probe to aide in the analysis of structure/function relationships in subpopulations of granulosa cells.

Identification of subpopulations of rat granulosa cells: sedimentation properties and hormonal responsiveness.

Ovarian granulosa cells from small follicles have generally been considered to comprise a homogeneous cell population; however, stratified arrangements of hormone receptors have been found in antral and mural granulosa cells of Graafian follicles. Using cultured granulosa cells derived from immature, hypophysectomized, estrogen-treated rats, we have previously shown that vasoactive intestinal peptide (VIP) as well as FSH stimulates steroid production by these cells. Dose-response analysis indicated that the actions of these hormones were additive, suggesting the presence of subpopulations of granulosa cells. Using a continuous (0-30%) Metrizamide density gradient, we have identified three populations of granulosa cells with different sedimentation properties. After centrifugation for 15 min at 1500 X g, cells sedimented at Metrizamide concentrations of 13%, 18%, and 20% (peaks A, B, and C, respectively). The subpopulation with lowest density (peak A) comprised 5% of the total cells, whereas the remaining cells were distributed approximately equally in the other two peaks. The profiles of estrogen and progesterone production by these cells in response to FSH and VIP indicated that FSH preferentially stimulated steroid production in cells with the highest density (peak C), whereas VIP mainly induced steroidogenic responses in cells of intermediate density (peak B). In contrast, cells with the lowest density (peak A) were unresponsive to either hormone. Treatment with forskolin, a universal adenylate cyclase activator, induced steroid production in both subpopulations B and C. Further studies demonstrated that LH/human CG receptors were induced by FSH and forskolin in cells from peak C, whereas VIP treatment did not induce LH/human CG receptors in cells from peak B. In unfractionated cultured cells, GnRH potently antagonized FSH- but not VIP-induced steroidogenesis. Upon density-gradient fractionation, the profile of GnRH receptor content correlated well with GnRH effects since FSH-responsive cells (peak C) contained the majority of GnRH receptors. The present results demonstrate that granulosa cells from immature follicles are heterogeneous and consist of two major subpopulations of cells with differential responsiveness to FSH and VIP. These findings provide the basis for further morphological and biochemical analysis of subpopulations of granulosa cells during follicular development.

The detection of subclasses of granulosa cells with differing responsiveness to EGF, FGF, and gonadotropin preparations using an anchorage-independent clonal agar assay.

Subpopulations of granulosa cells of differing responsiveness to epidermal growth factor (EGF), fibroblast growth factor (FGF), pituitary gonadotropin preparations, and rat erythrocyte suspensions (RBC) have been detected using an anchorage-independent clonal agar assay. All growth factor preparations were capable of enhancing colony formation when added alone, and elicited cloning efficiencies as high as 35% when added to the culture system at optimal concentrations in a variety of combinations. The FGF preparation was the single most effective stimulator of colony formation, augmenting both colony numbers and colony size at concentrations as low as 50 ng/ml. However, unlike the other growth factors in this assay system, a plateau in responsiveness could not be reached even at levels as high as 1 microgram/ml. NIH-FSH-P2 and NIAMMD-bLH-4 were considerably less potent than other growth factors. Both preparations were inactive at concentrations less than 1 microgram/ml and produced an optimal response at 10 micrograms/ml.