| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ovary |
Department of Medical Biochemistry, Division of Molecular Genetics, Max F. Perutz Laboratories, Medical Universityand Biocenter of Vienna, A-1030 Vienna, Austria
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
activin, follicle-stimulating hormone, granulosa cells, ovary, ovulation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Avian ovarian follicles contain concentric layers of membranes and cells surrounding the oocyte. The perivitelline layer (PVL), the avian homologue of the mammalian zona pellucida [4], organizes immediately adjacent to the oocyte and acts as the sperm receptor [5–7]. The PVL is produced at the lumenal face of chicken granulosa cells (cGCs), which are highly specialized epithelial cells of endodermal origin and organize along a basal membrane. Being the somatic cell type closest to the germ cell, cGC are thought to play important roles in the separation of somatic and germ cells as well as in establishing communication between them. On top of the basal membrane, two layers of mesenchymal theca cells contribute to the vascularized connective tissue in the follicle. The theca interna additionally serves important functions in growth factor secretion and steroidogenesis [8]. In contrast with mammalian species, where granulosa cells line the antrum of ovarian follicles maintaining follicular fluid homeostasis, avian ovarian follicles lack a comparable lumen. Instead, cGC engulf the whole germ cell, growth of which depends on rapid, efficient, and yet specific import of large amounts of liver-derived macromolecules through the unvascularized granulosa layer. Hence, lipoproteins may transverse cGC either by transcytosis or by paracellular transport routes. Earlier morphological studies provided evidence that the latter is responsible for satisfying the oocyte's enormous demand for yolk precursors during the rapid growth phase [9, 10]. Consequently, despite pronounced epithelial features of granulosa cells, at least a temporary absence of tight junctions has been proposed because a permanent one would certainly impair paracellular transport processes.
Tight junctions normally localize to the most apical region of lateral epithelial cell membranes [11]. They exert a barrier function by restricting paracellular permeability through epithelial layers and implement a fence by separating the plasma membrane into an apical and a basolateral domain [12]. A major component is occludin, an integral membrane protein with four predicted transmembrane domains localizing exclusively to tight junctional complexes [13, 14]. Occludin is directly involved in establishing cell- cell contacts because hydrophobic protein loops of adjacent tight junction strands on neighboring cells interact in a zipper-like fashion, bringing together neighboring plasma membranes so as to seal the intercellular space [15]. Despite its important function in tight junction organization [14, 16, 17], occludin displays pronounced interspecies sequence diversity, which made characterization in other species tedious [18]. The fairly conserved occludin carboxy- terminus binds to zonula occludens protein 1 (ZO-1) [19], a peripheral membrane protein that links occludin to intracellular actin filaments via 
-catenin molecules [20]. Although ZO-1 localization is used as a tight junction marker, it is not restricted to tight junctions, but is also present in adherens junctions, where it colocalizes with cadherins [20].
In this study, we provide the first data on the presence and regulation of the tight junction protein occludin in avian granulosa cells. We demonstrate that occludin expression is shut down in a stage-specific manner during follicular maturation in vivo. Further, our results provide evidence that high occludin expression in cGC requires the synergistic action of follicle-stimulating hormone (FSH) and activin A, two major regulators of ovarian development. According to our model, the high FSH and/or activin A responsiveness of cGC in SWF serves to maintain occludin expression and thus favors intact tight junctions, preventing paracellular transport processes. Upon decrease of cGC responsiveness to FSH and/or activin A, occludin expression disappears, tight junction complexes disintegrate, and paracellular transport can proceed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The following abbreviations are used: 8-Br-cAMP, 8-Bromo-cAMP; Na, salt; AG99, -cyano-(3,4-dihydroxy)cinnamide; Tyrphostin A46, Tyrphostin B40, CAS 118409-59-9; cGC, chicken granulosa cell; DAPI, 4,6- diamino-2-phenylindole; DMEM, Dulbecco modified Eagle medium; EGF, epidermal growth factor; EGFR, EGF receptor; FCS, fetal calf serum; FSHR, FSH receptor; INF
, interferon-; KT5720, potent, specific, cell- permeable inhibitor of protein kinase A CAS 108068-98-0; MDCK, Madin-Darby canine kidney cells; PD98059, 2'-amino-3'-methoxyflavone, CAS 167869-21-8; PKA, protein kinase A, cAMP-dependent protein kinase; SWF, small white follicle; SYF, small yellow follicle; TGF, transforming growth factor-; TNF, tumor necrosis factor-; VLDL, very low-density lipoprotein; VLDLR, very low-density lipoprotein receptor; ZPC, zona pellucida protein C; ZO-1, zonula occludens protein 1; PVL: perivitelline layer.
Animals, cGC Isolation, and Cell Culture
Derco Brown laying hens were maintained on layer's mesh with water and feed provided ad libitum under a daily light period of 16 h. All animal experimentation was in compliance with the regulations of the animal ethics committee of the University of Vienna. Hens were killed by decapitation and cGC were isolated from large preovulatory follicles as described previously [21] using minor modifications [22]. Dispersed cells were plated in DMEM (cat. no. 41965; GibcoBRL Life Technologies, Invitrogen Ltd., Paisley, UK), supplemented with 5% FCS (Invitrogen Ltd.), 50 U/ml penicillin, 50 µg/ml streptomycin, 50 ng/ml (1.4 nM) human pituitary FSH (Calbiochem, San Diego, CA), and 25 ng/ml (1.8 nM) human recombinant activin A (R&D Systems, Minneapolis, MN). The cGCs were cultivated under standard conditions (37°C, 5% CO2) and, after reaching confluency, further propagated by splitting 1:2. About 24 h before the start of an experiment, cells were growth-factor starved in medium lacking FSH and activin A. Growth-factor treatment was then carried out with the indicated amounts of FSH and activin A, while TGF (R&D Systems) was used at 10 ng/ml concentration (1.6 nM). Inhibitors and activators were added 30 min before growth factor supplementation at the following concentrations 3 µM KT5720, 10 µM PD98059, 250 µM AG99 (Calbiochem, San Diego, CA), and 100 µM for 8-Br-cAMP (Sigma-Aldrich, St. Louis, MO).
Protein Preparation, SDS-PAGE, and Immunoblotting
Cultured cGCs were lysed directly in culture dishes by addition of 30 µl/cm2 SDS-PAGE sample buffer (5% [v/v] glycerol, 1% [w/v] SDS, 2.5% [v/v] ß-mercaptoethanol, 30 mM Tris/HCl, pH = 6.8, 0.025% [w/v] bromphenol blue). After boiling extracts for 5 min, lysates were cleared and stored frozen at –20°C. Equal amounts of protein were separated through 10% SDS-PAGE gels and transferred to Protran nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany). Membranes were stained with Ponceau S (Sigma-Aldrich, Corp., St. Louis, MO) to verify protein transfer and equal protein loading, followed by blocking with BLOTTO (5% [w/ v] fat-free milk powder in PBS, 0.1% [v/v] Tween-20) for 1 h at room temperature. Antibody incubation was done overnight at 4°C using polyclonal rabbit anti-chicken occludin antibodies (cat. no. 71-1600; Zymed Laboratories, Inc.), polyclonal rabbit anti-ZO-1 antibodies (cat. no. 61- 7300; Zymed Laboratories Inc.) and mouse anti-pan-cadherin antibodies (Clone CH-19, cat. no. C1821; Sigma-Aldrich Corp.). After incubation with secondary goat anti-rabbit or goat anti-mouse IgG-HRP (Oncogene Research Products, EMD Biosciences, Inc., San Diego, CA) antibodies, each used at dilutions of 1:10 000, bands were visualized using ECL Western blotting detection system using conditions as suggested by the manufacturer (Amersham Pharmacia Biotech, Buckinghamshire, UK). All data were compiled from at least three independent and representative experimental datasets.
Immunofluorescence and Immunohistochemistry
Preovulatory follicles and whole ovaries from adult laying hens were embedded in freezing agent and immediately shock frozen on dry ice. Cryosections of 20-µm thickness were prepared using a cryomicrotome (HM 500 OM, Microm Laboratories, Walldorf, Germany) and transferred onto Superfrost Plus slides (Menzel, Braunschweig, Germany). Alternatively, intact cGC sheets were isolated from preovulatory follicles and immediately spread on microscope slides. Granulosa cells in culture were grown on collagen I-coated glass slides (BD Biosciences, Erembodegem, Belgium). After fixation in a mixture of equal volumes acetone and methanol at –20°C for 15 min, sections, sheets, or cultured cells were rehydrated for 15 min in PBS at 37°C. Primary mouse anti-pan-cadherin and rabbit anti-occludin antibodies were used at dilutions of 1:500. Secondary antibodies (Alexa Fluor series; Molecular Probes, Eugene, OR) were applied at dilutions of 1:500 in 1% (w/v) fat-free milk powder in PBS for 1 h at 37°C. After several PBS washes, the first of which contained DAPI to stain nuclear DNA, slides were mounted in fluorescence mounting medium (DakoCytomation A/S, Glostrup, Denmark) and inspected using an Axiovert 135 fluorescence microscope (Carl Zeiss, Oberkochen, Germany) or a TCP SP2 laser-scanning spectral confocal microscope (Leica Microsystems Heidelberg GmbH, Bensheim, Germany).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To follow expression of occludin in vitro and in vivo, we performed immunoblotting experiments using extracts from cultured cGC and freshly isolated cGC sheets. For technical reasons, it is very difficult to impossible to isolate cGC from early follicular stages. As a consequence, protein extracts from SWF and SYF contained whole follicle material, whereas extracts from preovulatory F1 follicles represent isolated GC sheets (Fig. 1). SWF contained high levels of occludin; SYF, by contrast, showed clearly decreased expression. Importantly, occludin was almost undetectable in cGC from F1 preovulatory follicles (Fig. 1). Cadherin immunostaining served as loading control because its expression levels remained unchanged during follicular maturation. Thus, these data demonstrate a pronounced decrease of occludin expression during follicular maturation.
|
To verify the assumption that occludin immunoreactivity on immunoblots arises mainly from granulosa cells and to determine the intercellular and subcellular distribution of occludin, immunohistochemistry on cryosections of intact follicles as well as isolated granulosa sheets was performed. The aforementioned occludin immunoblots confirmed that the occludin antibodies were also suitable for immunofluorescence studies. Immunohistochemistry revealed that occludin was most abundantly expressed in cGC of SWF (Fig. 2). In terms of subcellular distribution, occludin was strongly enriched in the apical region of the granulosa layer, immediately adjacent to the oocyte (Fig. 2, A and C). Pan- cadherin antibodies served as counterstain to demonstrate the epithelial nature of granulosa cells, which exhibited a cobblestone morphology typical for epithelial cell types. The overlay in yellow indicated that cadherins and occludin colocalized in the apical region of the epithelial granulosa cell layer (Fig. 2, arrow in B and D). These experiments clearly attributed occludin to the granulosa layer and further supported the strong decrease of occludin levels during follicle development as seen on immunoblots (Fig. 1). Occludin was abundantly expressed in cGC up to the SWF stage, whereas it became virtually undetectable in cGC of SYF (Fig. 2E). Isolated cGC sheets from F1 follicles displayed punctuate staining patterns rather than the usual belt-like fluorescence (Fig. 2F). To seal the intercellular space, tight junctions have to circumvent the outside of whole cells. Hence, the observed intracellular staining suggests the absence of functional tight junctions. Furthermore, immunohistochemistry staining of sections from small white follicles of 3–8-mm diameter demonstrated that occludin is also expressed in these follicular stages preceding the rapid growth phase (data not shown). These results demonstrate that granulosa cells express occludin predominantly in early follicular stages in vivo; its cell surface localization suggests the presence of functional tight junctions in these stages. Furthermore, occludin expression strongly decreased during the small white to small yellow transition and remained low up to the F1 stage.
|
Activin A and FSH Are Key Regulators of Occludin Expression in cGC
Tight junction disintegration may represent a functional adaptation of the epithelial barrier to satisfy altered permeability requirements. Nevertheless, important questions as to how these events are triggered at the molecular level in vivo and which signal transduction cascades are involved remained to be addressed. To identify growth factors implicated in occludin regulation, we took advantage of a recently established tissue culture system allowing for the long-term cultivation of primary untransformed cGC in a functionally differentiated state [22, 23]. Using this in vitro system, we have recently shown that activin A and FSH are required to maintain the functional polarity of cGC in vitro. Cells growing in the presence of these growth factors maintain an epithelial morphology seen in vivo, and they express granulosa cell markers at similar levels as in the in vivo situation [22, 23]. Immunoblotting of extracts from cultured cGC revealed that both FSH and activin A exerted a pronounced effect on occludin expression levels (Fig. 3). While activin A alone (Fig. 3, lane 2) and, to a lesser extent, FSH (Fig. 3, lane 5) elevated occludin levels, both hormones added simultaneously to cGC cultures acted in a synergistic manner (Fig. 3, lane 6), giving rise to high occludin levels. By contrast, TGF, which has been previously demonstrated to exert adverse effects on cGC morphology and differentiation [22], suppressed occludin expression, even in combination with activin A (Fig. 3, lanes 3 and 4). The observed effects were not brought about by a general induction or repression of junctional proteins due to a change in cellular morphology because the levels of the tight junction- and adherens junction-associated protein ZO-1, as well as cadherin, remained constant under all applied conditions (Fig. 3). Hence, FSH signaling in combination with activin A must play a key role in tight junction formation or maintenance in cGC in vivo.
|
To identify cellular signaling components involved in this process, specific inhibitors of protein kinase A (KT5720), the epidermal growth factor receptor (AG99), and the MAP kinase kinase (PD98059) were exploited. As shown in Figure 4A, the PKA activator 8-Br-cAMP mimicked FSH action, whereas inhibition of PKA attenuated occludin induction by activin A/FSH or activin A/8-Br- cAMP (Fig. 4A). Strikingly, occludin induction by activin A alone was completely abolished by KT5720, indicating that activin A signaling requires basal PKA activity to induce occludin. A similar regulation pattern was observed for the intracellular activin A signal transducer Smad2 [23]. Blocking the upstream epidermal growth factor receptor (EGFR) with AG99 or the MAP kinase kinase (MEK) with PD98059 augmented activin A-dependent occludin expression, fully consistent with a TGF-triggered occludin repression (Fig. 4B). These results demonstrate the involvement of the EGFR and the MEK/ERK signaling pathway in the regulation of occludin in granulosa cells.
|
Occludin Localizes to Granulosa Cell-Cell ContactsIn Vitro
To determine the subcellular localization of occludin in cultured cGC from preovulatory follicles, immunofluorescence experiments were carried out. In FSH-/activin A- treated cells, occludin was highly expressed and colocalized with cadherins to plasma membrane regions implicated in the formation of cell-cell contacts (Fig. 5). Such a distribution of occludin is consistent with the presence of intact and functional tight junctions. Control cells showed a similar morphology but a lack of occludin expression. Taken together, our studies clearly demonstrate that cGCs express occludin in a growth factor-dependent manner in vitro and in early follicular stages in vivo. Hence, our data provide compelling evidence for the existence of functional tight junctions that are regulated in granulosa cells during follicular maturation. In fact, the results suggest a simple model (Fig. 6) by which tight junction regulation occurs in vivo through the interplay of activin A and FSH. We propose that a decrease of cGC responsiveness toward these growth factors, as it occurs with progressing follicular maturation, leads to disruption of tight junctions, allowing for yolk precursors to reach the growing oocyte via paracellular transport, bypassing the transcytosis route through the epithelial granulosa layer (Fig. 6).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We show that occludin is abundantly expressed in cGC during early stages of follicle development. The in vivo distribution of occludin within granulosa cells closely resembles the staining pattern found in the ovarian surface epithelium or even endothelial cells. Early developmental stages up to SWF display the typical belt-like staining pattern in the apical plasma membrane domain. Hence, cGCs are likely to exhibit tight junctions during early follicular stages. However, upon the transition from SWF to SYF, occludin expression strongly decreases, indicating tight junction disintegration with progressing follicle maturation, perhaps allowing for high-capacity transport processes to occur independent of the typical cGC transcytosis route. Interestingly, the decrease of occludin levels seems to directly precede yellow yolk incorporation, which occurs during the final rapid growth phase of the oocyte. According to our model (Fig. 6), granulosa cell tight junctions prevent massive yolk incorporation into slow-growing oocytes. Disintegration of this intercellular seal, which is brought about by growth factor-induced downregulation of occludin expression, opens the way for bulk yolk components to enter the oocyte and may play a major role in the rapid growth phase once follicles have been committed into the ovulatory hierarchy.
Tight junctions present in the monocellular granulosa layer are likely to form a diffusion barrier between the somatic body fluids and the developing germ cell. Hence, the rapid uptake of huge amounts of very low-density lipoproteins (VLDL) and vitellogenin, two major yolk precursors, raises the question of how these large proteins and macromolecular complexes transverse the epithelial granulosa layer. There is morphological evidence that yolk precursors pass cGC paracellularly rather than being transported into the oocyte by transcytosis. The absolute mass of about 14 g of yolk precursors incorporated within only 7 days argues for paracellular transport due to its high capacity. Our data are supported by early electron microscopy studies of avian ovarian follicles in the hierarchical growth phase, which reported wide gaps between granulosa cells [9]. Moreover, numerous 20–30-nm particles, identified as VLDL particles, have been detected within the basal lamina, which is thought to act as a molecular sieve, as well as in spaces between granulosa cells [10].
The oocyte cell surface harbors receptors implicated in yolk lipoprotein precursor uptake, including the oocyte-specific 95-kDa VLDL receptor (VLDLR) [10, 32]. Genetic evidence for its role in uptake of two major yolk lipoprotein precursors, vitellogenin and very low-density lipoprotein, came from the nonlaying restricted ovulator (R/O) chicken strain, which carries a single mutation in the VLDL receptor gene [33]. As the VLDLR is only found in oocytes, but not in somatic cells such as cGCs, high efficient yolk uptake into the oocyte is likely to bypass the granulosa layer. No alternative receptor possibly mediating efficient or specific transcytosis in granulosa cells has been identified as yet. Hence, the absence of the VLDL receptor in cGC provides another strong argument for a paracellular lipoprotein transport route. The kinetics of lipoprotein uptake depend solely on the oocyte's receptor-mediated uptake capacity and the delivery of VLDL particles to the oocyte plasma membrane. It seems thus reasonable that a tightly sealed granulosa sheet serves an important function in maintaining small oocytes in a resting state by preventing immature follicles from yolk incorporation.
To explore tight junction regulation, we took advantage of a novel culture system for primary cGC [22]. We demonstrate that two major regulators of ovarian function, activin A and FSH, exert a strong effect on occludin regulation in vitro. Moreover, we show cooperation of FSH-triggered protein kinase A signaling with the activin pathway. Interestingly, the additive effect of FSH and activin A in occludin regulation appears as a common theme in the regulation of different genes within granulosa cells. In previous studies, we reported a similar regulation pattern of the intracellular signal transducer Smad2 [23], for the activin A antagonist inhibin , and for the PVL component zona pellucida protein C [22]. Moreover, similar mechanisms operate in mammalian granulosa cells, giving rise to synergistic induction of proliferating cell nuclear antigen and cyclin D2 [34].
Based on our data, we postulate two plausible, not necessarily mutually exclusive, mechanisms for occludin downregulation during follicular development in vivo. Notably, there is good evidence in the literature for both mechanisms. First, the levels of required growth factors FSH and activin A decrease with increasing maturation, or at least cGC responsiveness toward them decreases. In the case of FSH, the first scenario seems to be predominant. FSH represents a well-established regulator of ovarian development. The cognate FSH receptor (FSHR) is most abundant in granulosa cells of early follicular stages [35]. In fact, FSH is considered as the essential survival factor for immature follicles. However, FSHR expression levels sharply decline at later developmental stages, perhaps causing the decreased responsiveness of cGC to FSH [35]. The signals responsible for a decrease in FSHR expression in vivo remain to be elucidated. However, we and others have shown that TGF treatment decreases FSHR expression in chicken [22] and rat [36] granulosa cells in vitro. Hence, TGF may represent a likely candidate to induce disruption of tight junctions by decreasing granulosa cell responsiveness toward FSH. Importantly, TGF also interferes with activin A, the second growth factor required for high occludin expression. Activin A action is antagonized by TGF in cultured cGC in many different respects [22]. However, TGF additionally might exert more direct effects on occludin expression. As a ligand of the EGF-receptor, TGF promotes activation of Raf-family members. Oncogenic Raf-1 downregulates occludin levels and may therefore disrupt tight junctions in MDCK cells [24], providing another feasible explanation for TGF-induced repression of occludin expression in cultured cGC. Indeed, we show the direct involvement of this signaling pathway because inhibition, either at the level of the EGFR itself or blocking the downstream MAP kinases ERK1/ERK2, abrogates the inhibitory effects of TGF on activin A-dependent occludin induction.
In the case of activin A, there is good evidence for the second proposed mechanism, namely reduced growth factor availability in maturing cGC. The availability of activin A decreases with ongoing follicle development, whereas expression of inhibin A increases [37]. Because increasing amounts of inhibin A interfere with activin A signaling, activin A action might be significantly impaired in all follicular stages except for SWF. Moreover, in later stages, activin A seems present predominantly in theca cells, whereas granulosa cells produce mainly inhibin A in steadily increasing amounts [38]. The maximum is not reached before the F1 stage [39–41], arguing for an autocrine blockade of activin A action within granulosa cells. As free inhibin--subunit precursors compete with FSH for FSHR binding [42], production of high amounts of inhibin- might further impair FSH signaling. In summary, there is evidence for several distinct hormonal effectors that exert a negative influence on signaling cascades that control occludin expression and thus tight junction formation. A decrease in FSHR expression and an increase in inhibin-- subunit levels affect FSH signaling, while a decrease in activin A levels and an increase of inhibin A impair autocrine and paracrine signaling in follicles. Finally, TGF severely impairs both pathways and is hence another putative regulator of tight junction formation during follicular maturation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Karl Kuchler, Department of Medical Biochemistry, Division of Molecular Genetics, Max F. Perutz Laboratories, Medical University and Biocenter of Vienna, Dr. Bohr-Gasse 9/2, A-1030, Vienna, Austria. FAX: 43 1 4277 9618; karl.kuchler{at}univie.ac.at
Received: 15 October 2003.
First decision: 10 November 2003.
Accepted: 15 January 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. A. Adissu, E. K. Asem, and S. A. Lelievre Three-Dimensional Cell Culture to Model Epithelia in the Female Reproductive System Reproductive Sciences, December 1, 2007; 14(8_suppl): 11 - 19. [Abstract] [PDF]
|
|
|
|
 |
|
 J. B. Hoffman, A. P. Benson, V. L. Christensen, B. D. Fairchild, and A. J. Davis Follicular Development and Expression of the Messenger Ribonucleic Acid for the Inhibin/Activin Subunits in Two Genetic Lines of Turkey Hens that Differ in Total Egg Production Poult. Sci., May 1, 2007; 86(5): 944 - 952. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sato, Y. Kanai, T. Noma, M. Kanai-Azuma, S. Taya, T. Matsui, M. Ishii, H. Kawakami, M. Kurohmaru, K. Kaibuchi, et al. A close correlation in the expression patterns of Af-6 and Usp9x in Sertoli and granulosa cells of mouse testis and ovary Reproduction, November 1, 2004; 128(5): 583 - 594. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
