| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of Animal and Veterinary Basic Sciences3
and Large Animal Sciences,4 Royal Veterinary and Agricultural University, DK-1870 Frederiksberg C, Denmark
Monash Institute of Reproduction and Development,5 Monash Medical Centre, Clayton, Victoria 3168, Australia
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
bovine embryo, developmental biology, early development, epiblast, pregastrulation, pluripotency, OCT4
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Production and subsequent processing of Day 14 embryos was described previously [2]. Procedures for handling animals in the present study were conducted in accordance with the Guidelines for Care and Use of Agricultural Animals in Agricultural Research and Teaching (permission no. 2002/561-610). Unless otherwise indicated, the media and chemicals were from Sigma-Aldrich Denmark A/S (Vallensbaek, Denmark), and the plasticware was from VWR International (Nunclon, Albertslund, Denmark). The media were prepared with freshly produced Milli-Q water (Milli-RO Plus and Milli-Q PF Plus Water Purifications Systems; Millipore A/S, Hedehusene, Denmark). All incubations were carried out at 38.5°C.
Standard procedures were used for the production of abattoir-derived in vitro-matured (IVM), in vitro-fertilized (IVF), and IVC bovine embryos. Briefly, ovaries from Danish cows and heifers were collected and transported to the laboratory in thermo boxes in 0.9% sterile saline at 30–33°C. The cumulus-oocyte complexes (COCs) were aspirated from antral follicles (diameters, 3–15 mm). After isolation and a wash in Hepes-buffered TCM-199 (M 2520), aliquots of 50–60 COCs were submitted to IVM in 0.7 ml of medium without oil overlay in four-well dishes in 5% CO2 in humidified air. The IVM medium was Dulbecco modified Eagle medium (D 5523) supplemented with 50 ng/ml of epidermal growth factor (E 4127), 2 U/ml of Suigonan Vet (400 U of eCG and 200 U of hCG; Intervet Danmark A/S, Skovlunde, Denmark), 50 µg/ml of gentamicin (G 1264), 100 µM of ß-mercaptoethanol (31350-010; Gibco, Invitrogen A/S, Taastrup, Denmark), and 5% estrous cow serum (ECS; Danish Institute for Food and Veterinary Research, Copenhagen, Denmark). After 24 h of IVM, aliquots of 50–60 COCs were washed once in Tyrode albumin lactate pyruvate (TALP) medium and then placed in 0.5 ml of TALP medium without oil overlay in four-well dishes [10]. The TALP medium contained 6 mg/ml of BSA (A 4919), 30 µg/ml of heparin (H 3149), 0.25 mM sodium pyruvate (P 3662), 10 mM lactate, 20 µM penicillamine (P 4875), 10 µM hypotaurine (H 1384), and 1 µM epinephrine (E 4250). The COCs were then coincubated for 22 h with frozen-thawed, washed semen from a fertile bull (Taurus A.I. Station; Dansire, Aalborg, Denmark) at a final concentration of 2.5 x 106 spermatozoa/ml in 5% CO2 in humidified air. After IVF, the cumulus cells were removed by vortex agitation for 1 min. After washing, aliquots of 20–25 inseminated oocytes were transferred to 0.1-ml droplets covered with mineral oil (M 8410) in SOFaaci medium with 10% ECS for IVC [11] and incubated in 5% CO2, 5% O2, and 90% N2 in a Modular Incubator Chamber (patent no. 5352414; Billups-Rothenberg, Inc., CA).
At Day 7 postinsemination, grade A (International Embryo Transfer Society classification) compact morulas and blastocysts were loaded in numbers of 10–30 into 0.25-ml straws (IMV, 61302 L'Aigle Cedex, France) in Hepes-buffered TCM-199 (M 2520) with 10% ECS and transported to the barn in a portable incubator at 38.5°C. The embryos were transferred nonsurgically to the mid or distal part of each uterine horn of dairy heifers (age, 2–2.5 yr; n = 2) that were synchronized with two injections of cloprostenol (Estrumate Vet; Schering-Plough, Farum, Denmark) 11 days apart to a cycle stage of Day 8 after estrus. Five days later, the uterine horns of the heifers were flushed nonsurgically by an 18-gauge, two-way embryo collection catheter (Mini-Tüb, 84184 Tiefenbach, Germany) using approximately 250 ml of Dulbecco PBS (Pharmacy, Royal Veterinary and Agricultural University, Frederiksberg, Denmark) per horn, taking care to avoid disrupting the embryos. Embryos were identified in the flushing medium by stereomicroscopy and fixed for either transmission-electron microscopy (TEM) or immunohistochemistry. Specimens for TEM (n = 6) were fixed in 3% glutaraldehyde in 0.1 M PBS for 1 h at 4°C and stored in fresh 0.1 M PBS until further processing a few days later. Specimens for immunohistochemistry (n = 34) were fixed for 1 h at room temperature in 4% paraformaldehyde (Merck KGaA, Darmstadt, Germany) in 0.1 M PBS and stored in 1% paraformaldehyde in 0.1 M PBS at 4°C until processing within a week. Immunohistochemical specimens for dehydration and embedding in paraffin wax (n = 16) were fixed as complete embryos, whereas in specimens for confocal laser-scanning microscopy (CLSM; n = 18), the epiblast was isolated using hypodermic needles before fixation.
Processing of Day 12 Embryos for TEM
Specimens were embedded in 4% agar (Bacto-agar; Difco Laboratories, Detroit, MI) under stereomicroscopy and postfixed in 1% OsO4 in 0.1 M sodium phosphate buffer for 1 h at 4°C. Samples were stained en-bloc with uranyl acetate, dehydrated through ascending concentrations of ethanol, transferred to propylene oxide, embedded in Epon, and serially semithin sectioned (thickness, 2 µm). Semithin sections were stained with 1% basic toluidine blue and evaluated by bright-field light microscopy. Selected semithin sections were reembedded [12], ultrathin sectioned (thickness, 70 nm), contrasted with uranyl acetate and lead citrate, collected on copper grids, and examined using a TEM microscope (CM100; Philips, Darmstadt, The Netherlands).
Processing of Paraffin Sections for Immunohistochemistry
Day 12 embryos were completely cut into serial sections (thickness, 5 µm). Sections were collected on SuperFrost Plus Slides (Menzel GmbH & Co KG, Braunschweig, Germany) and were affixed to slides with an overnight incubation at 37°C. Selected paraffin wax sections were stained with hematoxylin-eosin for orientation purposes and analyzed by bright-field light microscopy. Representative sections (n = 100–300) had been cut from Day 14 embryos as described previously [2], and slides were immediately available. For immunohistochemical characterization, selected sections were dewaxed and rehydrated. Endogenous peroxidase activity was blocked with 3% (v/v) H2O2 in ethanol before antigen retrieval by microwave treatment [13] in 0.01 M sodium citrate buffer (pH 6.0) for 15 min. Nonspecific binding of antibodies was blocked with a ready-to-use, serum-free, protein-blocking solution (X0909; DakoCytomation, Glostrup, Denmark). Sections were labeled with rabbit polyclonal anti-human octamer-binding transcription factor 4 (OCT4; H-134; sc-9081; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), mouse monoclonal anti-human vimentin (18-0052; Zymed Laboratories, Inc., San Francisco, CA), rabbit polyclonal anti-human 
1-fetoprotein (A 00008; DakoCytomation), rabbit polyclonal anti-mouse (Englebreth Holm-Swarm mouse sarcoma) laminin (L9393; Sigma-Aldrich Denmark A/S), rabbit polyclonal anti-human zonula occludens (ZO)-1 (61-7300; Zymed Laboratories), and mouse monoclonal anti-human Ki-67 antigen, clone MIB-1 (M 7240; DakoCytomation). Incubation with the primary antibodies in 1% BSA in PBS was performed at room temperature for 2 h (OCT4, 1:250; vimentin, 1: 100; laminin, 1:50; Ki-67, 1:100) or overnight at 4°C (1-fetoprotein, 1: 5000; ZO-1, 1:3000). Negative-control slides were incubated with identical concentrations of nonimmune sera (vimentin and Ki-67: mouse immunoglobulin [Ig] G1 [X 0931; DakoCytomation]; 1-fetoprotein, laminin, and ZO-1: rabbit IgG [X 0936; DakoCytomation]). For OCT4, preabsorption for 2 h at room temperature with OCT4 protein (1-134; sc-4420 WB; Santa Cruz Biotechnology) was used. The primary antibodies were visualized by DAKO LSAB+ kit (K0690; DakoCytomation), and the sections counterstained with Mayer hematoxylin before mounting in Faramount aqueous mounting medium (S3025; DakoCytomation).
Alkaline phosphatase activity was visualized by use of a previously described protocol [14] with some modifications. All materials used were from Sigma-Aldrich Denmark A/S. Briefly, sections were incubated in Tris-maleate buffer, composed of 3.6 g/L of TrismaBase (T6066) and 1 M maleic acid (M0375), pH 9.0, for 30 min. Staining was performed by incubating sections in a solution composed of 9.8 ml of Tris-maleate buffer and 2 mg/ml of naphthol AS-MX phosphate (N4875) in 0.2 ml N-N-dimethylenformamide (40248) and 10 mg of fast blue (F3378) for 30 min. Staining was ended by incubating sections in PBS. Negative-control specimens were treated in a similar way, with 5 mg/ml of levamisol (L 9756) added to the staining solution.
Processing for Immunohistochemistry and Confocal Laser Scanning Microscopy (CLSM)
Specimens were permeabilized in 0.15 M PBS with 1% Triton X-100 and 0.2% BSA at room temperature for 1 h. To facilitate detection of filamentous actin and, thereby, cell shape [15], embryos were incubated for 1 h at room temperature in 10 IU/ml of Texas Red-X phalloidin (T-7471; Molecular Probes Europe BV, Leiden, The Netherlands) diluted in washing solution composed of 0.15 M PBS, 0.1% Triton X-100, and 0.2% BSA. After washing in washing solution, embryos were incubated in a blocking solution composed of 0.15 M PBS, 0.1% Triton X-100, and 5% inactivated goat serum for 2 h at room temperature. Embryos were then incubated overnight at 4°C, with primary antibodies diluted in blocking solution as follows: Ten embryos were incubated with rabbit polyclonal anti-mouse (Englebreth Holm-Swarm mouse sarcoma) laminin 1:25 (L9393; Sigma-Aldrich Denmark A/S), five embryos with rabbit polyclonal anti-bovine fibronectin antibody 1:500 (24921; Novotec, Saint Martin La Garenne, France); and three embryos with rabbit polyclonal anti-human ZO-1 1:3000 (61-7300; Zymed Laboratories, Inc.). After washing in washing solution, all embryos were incubated for 2 h at room temperature with biotinylated goat anti-rabbit antibody 1:100 (E 0432; DakoCytomation). Finally, all embryos were incubated for 1 h in the dark at room temperature in blocking solution with streptavidin fluorescein 1:100 (RPN 1232; Amersham Biosciences, Hilleroed, Denmark) before mounting on glass slides under coverslips in Vecta-Shield antifade medium (Vector Laboratories, Ltd., Peterborough, U.K.) with 0.1 µg/ml of 4',6'-diamidino-2-phenylindole (DAPI; D-1306; Molecular Probes Europe BV) for detection of nuclei within a plastic hole reinforcement placed between the glass slide and the coverslip. Control embryos were submitted to the same procedures with the omission of the primary-antibody incubation. Slides were stored in the dark at 4°C until examination by CLSM using sequential scanning on a Leica TCS SP2 microscope (Leica Microsystems GmbH, Heidelberg, Germany) and a 40x/1.25 oil objective. The DAPI was detected using an argon laser with excitation wavelengths at 351 and 364 nm. Emission was recorded at 409–468 nm. Fluorescein and Texas Red were detected using a helium/neon laser with excitation wavelengths at 488 nm and 543 nm, respectively. Emission was recorded at 500–535 nm (fluorescein) and 560–700 nm (Texas Red). Image stacks from CLSM were reconstructed with Leica Confocal Software Lite Version 2.05 1347a (Leica Microsystems).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The Day 12 embryos ranged in size from 0.4 x 0.5 to 1.7 x 2.1 mm. Trophectodermal cells were cuboidal in shape except over the well-defined epiblast, where they tended to flatten and become squamous, thus forming a Rauber layer (Fig. 1, a and b). Varying degrees of focal interruptions were found in this part of the trophectoderm in most embryos, either in the form of cell disintegration or presumptive apoptosis (or both). In some embryos, nascent amniotic folds were seen in the margins of the Rauber layer. The sprouting amniotic folds did not include the hypoblast and appeared to form by expansion of the trophectoderm only. Ultrastructurally, tight junctions and desmosomes were seen between adjacent trophectodermal cells.
|
In some embryos, the cells of the epiblast were seen to be organized in up to five more or less distinguishable, concentric cell layers (Fig. 1b). In other embryos, an apparently more developed epiblast seemed to have expanded and flattened (Fig. 1a). Size of the embryos as a whole and developmental stage of the epiblast did not seem to correlate. The organization of epiblast cells allowed for identification of two well-defined subpopulations: one constituting a distinctive basal layer of cells apposing the hypoblast (basal cells), and one composed of cells arranged inside or above the basal cells, including cells apposing the Rauber layer (Fig. 1a). With development, the shape of the basal cells was seen changing from pyramidal, with a broad base toward the closely apposed hypoblast, to cuboidal or columnar (Fig. 1c). The cells possessed interdigitating plasma membranes and were connected by desmosomes (macula adherens) and zonula adherens junctions (Fig. 1f). Cells located inside or above the basal cell layer generally were spherical in shape, without signs of polarization. In central parts of the epiblast, cells often were seen encircling cystic cavities (Fig. 1b), dividing or presumptive apoptotic cells. Among the epiblast cells underlying the Rauber layer, more variation was revealed. In areas underlying focal interruptions of the trophectoderm, cystic cavities between epiblast cells often were seen (Fig. 1b). In consecutive sections, some of these cavities appeared to be in direct contact with the "milieu exterieur" of the embryos, through gaps in the Rauber layer. Abundant microvilli were seen protruding into such cavities from surrounding, predominantly elongated epiblast cells. A special type of zonula adherens-like cell junction was seen connecting epiblast cells at these locations (Fig. 1d). In areas where the Rauber layer appeared to be intact, presumptive tight junctions often were seen connecting adjacent epiblast cells (Fig. 1e). Though closely apposed, cells of the two lineages (i.e., epiblast and trophectoderm) appeared to be loosely connected by interdigitating filapodia only (Fig. 1e). At the ultrastructural level, cells throughout the epiblast presented varying electron density, reflecting a different organelle composition, but not in an identifiable, systematic way. The most conspicuous organelles were polyribosomes, rough endoplasmic reticulum, and mitochondria, with polyribosomes dominating in the more electron-dense epiblast cells. In general, very few inclusions and vacuoles were noted in epiblast cells as compared with trophectodermal cells.
On the internal side of the epiblast and trophectoderm, a confluent squamous hypoblast was seen (Fig. 1a). With development, the hypoblast underlying the epiblast often demonstrated a cuboidal cell shape. Varying amounts of extracellular material, forming a basement membrane, were seen between the hypoblast and basal epiblast cells and between hypoblast and trophectoderm (Fig. 1f). In some embryos, basal epiblast cells and hypoblast cells were connected by interdigitating filapodia, whereas in other, apparently more developed embryos, these two cell lineages appeared to be completely separated by the basement membrane. Compared to areas between hypoblast and trophectoderm, deposits between basal epiblast cells and hypoblast cells appeared to be less densely packed. Ultrastructurally, adjacent hypoblast cells were seen connected by tight junctions and desmosomes.
Immunohistochemical Characterization of Day 12 Embryos
In all embryos, OCT4 labeling was specifically and exclusively localized to nuclei of the complete epiblast (Fig. 2, a and b). In mitotic cells, staining was seen in the entire cytoplasm. Unlabeled, presumptive apoptotic nuclei were found. An apparently nonspecific staining of microvilli of trophectodermal and hypoblast cells was noted. Embryos were seen displaying localization of vimentin to cytoplasmic foci in most epiblast cells (Fig. 2c). Examination of consecutive sections stained for OCT4 and vimentin, respectively, revealed that the two proteins often localized to the same cells. Alkaline phosphatase activity was detected apically in trophectodermal cells (Fig. 2d). No such activity was observed in the epiblast. The proliferation marker Ki-67 showed localization to most cells scattered throughout the embryo (data not shown). Occasionally, cytoplasmic localization of 1-fetoprotein throughout the hypoblast was observed (data not shown). In all embryos, laminin was found underlying the hypoblast throughout the embryo. In sections, the laminin sheet between the hypoblast and epiblast appeared like a "cup" investing the "ventral" portion of the latter. The ventrolateral sides of the cup seemed to disintegrate with development, thus leaving the epiblast covered by a continuous layer of laminin in the central lower parts and peripherally (Fig. 2, e and f). Fibronectin was found to be distributed diffusely between all cells throughout the embryo (data not shown). The most intense staining was found in the space between hypoblast and epiblast. In confocal images, ZO-1 could be detected as regions of continuous fluorescence between cells in both trophectoderm and hypoblast (data not shown). No similar staining was observed between epiblast cells, either in confocal images or in paraffin sections. In paraffin sections, a spot-like reactivity was revealed in trophectoderm and hypoblast only (data not shown).
|
Immunohistochemical Characterization of Day 14 Embryos
Labelings were performed on two ovoid and one elongated embryo ranging in length from 2.3 to 6.5 mm. In sections, all embryos presented a well-defined embryonic disk with no signs of an overlying Rauber layer. As compared with Day 12 embryos, in which the epiblast in some cases had started flattening, this process appeared to have continued. Thus, embryos presented an epithelial-like epiblast, with cells arranged in sheets having limited reminiscence of the concentric organization described in Day 12 embryos. The cell sheets ranged from being apparently bilayered or even pseudostratified to an estimated three to four layers, as seen in sections. As for the Day 12 embryos, the organization of epiblast cells allowed for identification of two well-defined subpopulations: one constituting a distinctive basal cell layer apposing the hypoblast, and one arranged above the former subpopulation. Basal cells at Day 14 bore close resemblance to the basal cells of the apparently most developed Day 12 epiblasts, being cuboidal or columnar in shape. Above this layer, in sections with two or more cell layers, the shape of the epiblast cells varied from spherical to elongated. In one of the embryos, the epiblast and the underlying hypoblast were separated completely by vast amounts of extracellular material, forming a diffuse web between the two cell lineages (Fig. 3, a and d). Labeling for laminin revealed no staining for the extracellular matrix protein at this location (data not shown). As for the Day 12 embryos, labeling for OCT4 revealed specific and exclusive localization to epiblast nuclei (Fig. 3, a and b). In one embryo, a decreasing intensity of staining was seen from the periphery of the embryonic disk inward (Fig. 3a). Consecutive sections labeled for vimentin revealed focal cytoplasmic labeling in most epiblast cells. Epiblast cells thus appeared to contain both vimentin and OCT4, as noted in the Day 12 embryos (Fig. 3, b and c). The Ki-67 labeling revealed proliferation throughout the epiblast, with no apparent dormant domains. As compared to the Day 12 embryos, a larger number of hypoblast cells in one embryo appeared to be nonproliferating (Fig. 3d). Labeling for laminin revealed a well-defined basement membrane between the hypoblast and trophectoderm and between the hypoblast and peripheral epiblast cells. In the middle of the embryonic disk, staining intensity was weaker, indicating a disintegration of the basement membrane at this location (Fig. 3e). As for the Day 12 embryos, labeling for ZO-1 did not reveal consistent staining in the epiblast, whereas spot-like reactivity was observed in trophectoderm and hypoblast (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the present study, the presence of at least two different cell subpopulations of the Day 12 epiblast was revealed. A distinctive layer of basal cells was identified relating to the underlying hypoblast. These cells appeared to undergo a change in shape from, initially, pyramidal to cuboidal or columnar. Cells composing the remainder of the epiblast generally were spherical in shape, forming different cell junctions and other membrane modifications. Our findings of at least two populations of cells within the epiblast suggest a certain level of differentiation at this developmental stage. In Day 14 embryos, the basal cells of the epiblast were cuboidal to columnar, whereas the remaining cells varied from spherical to elongated in shape. It is tempting to speculate that the basal cells at Day 12 and Day 14 constitute the same subpopulation. In Day 12 embryos, in areas with focal interruptions of the Rauber layer, underlying epiblast cells were polarized with apical microvilli and a special type of zonula adherens-like junction between them. This finding bears close resemblance to descriptions of porcine embryos at a similar developmental stage [16] and at gastrulation in the rabbit [18, 19]. Apparently, morphological changes within this part of the epiblast relates to loss of the Rauber layer. In the pig [16], the term junctional complex was used for the description of the zonula adherens-like cell junctions, indicating epithelial differentiation, within this part of the epiblast. In the present study, presumptive tight junctions often were seen between epiblast cells covered by a disintegrating Rauber layer (Fig. 1e). Similar findings were described in Day 11 bovine in vivo-derived embryos [2]. At Day 14 in that same previous study and at Day 15 in another study of bovine embryos [6], a marker of simple epithelia, cytokeratin 8 [20–22], was found in the epiblast. In the present study, we used immunohistochemical detection of a membrane protein specific for tight junctions in simple epithelia, ZO-1 [23], to further elucidate the extent and localization of differentiation within the epiblast. In spite of our ultrastructural observation of presumptive tight junctions, staining was not observed between epiblast cells in confocal images at Day 12, whereas the trophectoderm and the hypoblast displayed regions of continuous fluorescence between cells. Initial tight junction formation, as detected by weak punctate reactivity, has been described in the trophectoderm of murine [24] and bovine [25] morula-stage embryos. Findings in the pig [16, 26] and rabbit [19] have shown that the epiblast forms a functional seal with the trophectoderm, separating the uterine and blastocoelic compartments, before the Rauber layer disappears. Our findings concerning Day 12 embryos suggest that completion of tight junction formation in the bovine epiblast may be a late event, taking place after exposure of the epiblast to the milieu exterieur of the embryo. Thus, a seal toward the blastocoelic compartment may be provided by the hypoblast only, allowing all cell layers of the epiblast a more intense contact with the uterine secretions. Because the epiblast of Day 14 and Day 15 bovine embryos did not stain for cytokeratin 8 [2, 6], findings indicate that although cells of the bovine epiblast respond to changes in their surroundings, prominent differentiation within them has not yet taken place. We observed a spot-like reactivity for ZO-1 in both trophectoderm and hypoblast on paraffin sections, both at Day 12 and Day 14, in accordance with the ultrastructural finding of ZO junctions between cells of these two lineages. It may be noted that in some epithelia, zonula adherens junctions also present ZO-1 protein [27], suggesting that immunolocalization of this protein cannot be used as a specific marker for the presence of ZO junctions.
In the present study, we have been able, to our knowledge for the first time, to show specific and exclusive OCT4 staining of bovine epiblast nuclei in both Day 12 in vitro- and in vivo-produced embryos and Day 14 in vivo-derived embryos. Our findings suggest that around Day 14, an initial down-regulation of OCT4 RNA expression in the cell-rich central portions of the epiblast may have begun. Expression of OCT4, the gene of a POU domain transcription factor also known as POU5F1 [28–30], is considered to be a reliable marker of pluripotency in the mouse [31, 32] and in primates [33–35]. In contrast, in large domestic species, the specificity of this marker has been more ambiguous, because nuclear OCT4 staining has been demonstrated in both trophectodermal and inner cell mass cells of porcine and bovine blastocysts at Day 7 and Day 8, respectively [36]. The OCT4 RNA was coordinately down-regulated in both cell lineages between Day 10 and Day 14 [37]. Findings in the latter study were confirmed by whole-mount immunohistochemical evaluation of expressed protein using antibodies, the specificity of which was determined empirically by testing with differentiated and undifferentiated P19 cells. We used commercially available polyclonal rabbit antibodies on paraffin sections, which may explain, in part, the differences observed. Discrepancies between levels of OCT4 RNA transcripts and OCT4 protein in cells have been published [36, 38], indicating that although transcription has ceased, protein clearance may be delayed. Given the high sequence similarity between bovine, murine, and human genes encoding OCT4 [38], one may expect OCT4 to play a similar role in all mammalian species. Accordingly, bovine epiblast cells are expected to remain pluripotent and, thus, OCT4 positive, at least until the onset of gastrulation [39, 40], which has been observed around Day 14 [2, 8, 9]. Immunohistochemically, vimentin was detected already at Day 12, and on Day 12 and Day 14, this intermediate filament was localized to the same cells as OCT4. Recently, similar findings were made in porcine prestreak embryos [3]. The appearance of vimentin-positive epiblast cells has been associated previously with initial gastrulation in the chick [20], mouse [21], rabbit [41], and pig [42]. Because OCT4 RNA recently was shown to be expressed uniformly in both the epiblast and the mesoderm during gastrulation in the mouse [40], our findings regarding colocalization of a marker for pluripotency (i.e., OCT4) and one for initial differentiation (i.e., vimentin) do not appear to represent a paradox. Instead, these findings raise a question concerning the use of OCT4 products as markers for pluripotency during initial differentiation. According to our findings, differentiation within the bovine epiblast, characterized by vimentin expression, may already have initiated at Day 12. Taken together with the above discussion of morphological variation and presumptive epiblast tight junction formation, this adds to the general picture of a differentiating epiblast. In relation to the derivation of ES cells from isolated, undifferentiated embryonic cells, onset of vimentin expression has been suggested to represent an upper limit for such procedures in the pig [42]. In cattle, our results show that OCT4 may serve as a marker of epiblast cells. However, the period in which these cells are truly pluripotent may be rather short, and initial loss of pluripotency appears not to be reflected by a synchronous decrease in OCT4 protein levels. Thus, in this species, identification of a marker with a rapid down-regulation and protein clearance as the epiblast differentiates still remains to be done. In the mouse, however such a marker appears to be found in the transcript of Nanog [40, 43, 44].
During formation of the primitive streak, migrating epiblast cells may arise from different areas of proliferation in the embryonic disk [3, 45]. In porcine prestreak embryos, mitotic figures were found only in the anterior region of the epiblast [3]. A semicircular condensation of cells was seen between the epiblast and the hypoblast in one Day 14 embryo [2]. As shown in the present study, detection of the proliferation marker Ki-67 in both Day 12 and Day 14 embryos revealed an apparently similar rate of proliferation throughout the epiblast. The half-life of Ki-67 is very short, being an hour or less [46]. Thus, exclusively immediate postmitotic cells were expected to be detectable. Our findings suggest that particular proliferative regions have not yet been established in the bovine epiblast even at Day 14. However, we did not perform serial labeling throughout the epiblast. Detection of dormant or proliferative domains may require a more consistent investigation. Alkaline phosphatase activity was not found in bovine epiblast cells at Day 12 in the present study, confirming previous findings at Day 14 [2]. In another investigation concerning bovine blastocysts, alkaline phosphatase activity was found to be localized to the trophectoderm, and only after immunodissection staining could it be observed in the inner cell mass [47]. Findings on bovine ES cell-like cells are ambiguous [14, 48–50].
In the present study, we observed a continuous, laminin-containing basement membrane between the hypoblast and the trophectoderm and an incomplete membrane between the hypoblast and the epiblast. The former may constitute the equivalent to the membrane that in rodents develops into the so-called Reichert membrane [51–53]. Although chemically resembling other basement membranes [54], this is a more complex, multilayered membrane than the one that we have observed in bovine embryos. In different species of bats, it has been proposed that the Reichert membrane continues dorsal to the epiblast, either in connection with [55] or disconnected from [56] the hypoblast, depending on the species. In cattle, we did not see any signs of basement material dorsal to the epiblast; our results are in agreement with findings in the pig [16, 57]. These discrepant findings between domestic species and bats are not surprising and probably relate to the fact that both bovine and porcine polar trophectoderm (i.e., Rauber layer), for which cells a membrane dorsal to the epiblast would serve as a basement membrane, is eliminated well before gastrulation, which is not the case in rodents and bats. The other basement membrane described in the present study, between the hypoblast and the epiblast, appeared to form a cup that enveloped the ventral portion of the epiblast at Day 12. This membrane seemed to be in the process of disintegration, leaving the epiblast covered by a continuous layer of laminin in the central lower parts and peripherally. Our results are in good agreement with those of previous studies in bovine [6], porcine [57], and murine [58, 59] embryos. Discontinuity of the basement membrane, probably caused by failure of synthesis [60], may facilitate ingression of cells during the forthcoming gastrulation. Components of the basement membrane, fibronectin in particular, have been shown to provide a scaffold for migration of vimentin-positive cells [57]. The cystic cavities observed in the epiblast in Day 12 embryos (Fig. 1b) may be phylogenetic reminiscences of the cavities seen in rodents during amnion formation. In the mouse, the amniotic cavity is formed by apoptosis of cells lacking physical anchoring to the basement membrane between the hypoblast and the epiblast [61–65]. However, in the present study, the cystic cavities were observed at several cell layers of distance from the basement membrane, indicating that the mechanisms for their formation may be different from that in rodents.
In conclusion, Day 12 and Day 14 bovine epiblasts include at least two subpopulations of cells. Cells of the epiblast exclusively exhibit OCT4 staining, which is colocalized with staining for vimentin.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Morten Vejlsted, Department of Animal and Veterinary Basic Sciences, Royal Veterinary and Agricultural University, Groennegaardsvej 7, DK-1870 Frederiksberg C, Denmark. FAX: 45 3528 2547; mov{at}kvl.dk
Received: 12 July 2004.
First decision: 25 August 2004.
Accepted: 25 October 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. I Alexopoulos, P. Maddox-Hyttel, P. Tveden-Nyborg, N. T D'Cruz, T. R Tecirlioglu, M. A Cooney, K. Schauser, M. K Holland, and A. J French Developmental disparity between in vitro-produced and somatic cell nuclear transfer bovine days 14 and 21 embryos: implications for embryonic loss Reproduction, October 1, 2008; 136(4): 433 - 445. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME |
