The embryonic rat parietal yolk sac. Changes in the morphology and composition of its basement membrane during development.

The basement membrane (Reichert's membrane) of the entire capsular portion of the parietal yolk sac of rat embryos was examined both morphologically and chemically at various stages of gestation. The overall microscopic and compositional analyses showed Reichert's membrane to be typical of basement membranes isolated from other tissues and species. However, with increasing gestational age (from 11.5 to 17.5 days) a number of changes involving Reichert's membrane were noted: 1. The thickness increased rapidly then declined, while the surface area increased tenfold; 2. The total protein content increased twenty-fold while the collagen content increased eight-fold. As a result, the relative collagen content declined significantly; 3. The changes in the amino acid and carbohydrate composition were consistent with the latter finding.The observations listed above were evaluated in light of their possible relevance to an understanding of the morphogenesis of basement membranes during development, and to the possible mechanisms involved in pathogenesis of basement membrane dysfunction.     

Organ cultures of the embryonic rat parietal yolk sac. I. Morphologic and autoradiographic studies of the deposition of the collagen and noncollagen glycoprotein components of basement membrane.

The synthesis, deposition, and turnover of [³H]proline- and [³H]glucosamine-labeled basement membrane components have been studied in a system using organ cultures of embryonic rat parietal yolk sac tissues on a nutrient agar substrate. The morphologic and autoradiographic studies described in this report were correlated with the biochemical studies described in the companion paper (Minor et al., Develop. Biol.48, 1976). These studies showed that: (1) basement membrane (BM) was the only extracellular matrix synthesized in the cultures, (2) 10- to 30- μm thick layers of new BM were deposited during a 6-day culture period, (3) it was only the parietal endodermal cells that synthesized BM, (4) new BM was deposited only on the surface of existing BM, which was in contact with endodermal cells, (5) the amounts of new BM that accumulated were increased by a twice daily feeding schedule and decreased by the presence of trophoblast, (6) there was a notable difference in the redistribution of labeled components during a 6-day chase, and (7) [³H]glucosamine tended to localize in the Golgi complex, whereas [³H]proline was distributed throughout the cytoplasm of the endodermal cells.     

Organ cultures of the embryonic rat parietal yolk sac. II. Synthesis, accumulation, and turnover of collagen and noncollagen basement membrane glycoproteins.

Synthesis, accumulation, and turnover of basement membrane components have been studied in organ cultures of 13.5- and 14.5-day embryonic rat parietal yolk sac tissues on a nutrient agar substrate. The biochemical studies described in this report were correlated with morphologic and autoradiographic studies described in the companion paper (Minor et al., Develop. Biol.48, 1976). These studies showed that basement membrane is the only extracellular matrix synthesized, it is only synthesized by the parietal endodermal cells, and its synthesis is maintained for at least 6 days. In these cultures, synthesis and degradation of collagen and noncollagen proteins varied independently in response to environmental changes, such as the frequency of feeding or presence of trophoblast. The turnover of basement membrane collagen was much slower than that of the noncollagen proteins and this difference in the rate of turnover of the components had a major role in determining the composition of the newly synthesized basement membrane.     

Ultrastructure of Reichert's membrane, a multilayered basement membrane in the parietal wall of the rat yolk sac

The ultrastructure of Reichert's membrane, a thick basement membrane in the parietal wall of the yolk sac, has been examined in 13-14-d pregnant rats. This membrane is composed of more or less distinct parallel layers, each one of which resembles a common basement membrane. After routine fixation in glutaraldehyde followed by osmium tetroxide, the layers appear to be mainly composed of 3-8-nm thick cords arranged in a three-dimensional network. Loosely scattered among the cords are unbranched, straight tubular structures with a diameter of 7-10 nm, which mainly run parallel to the surface and to one another; they are referred to as basotubules. Permanganate fixation emphasizes the presence of a thick feltwork of irregular material around basotubules. Finally, minute dot-like structures measuring 3.5 nm and referred to as double pegs are present within the meshes of the cord network. Reichert's membranes have been treated for 2-48 h at 25 degrees C with plasmin, a proteolytic enzyme known to rapidly digest laminin and fibronectin. After a 2-h treatment, most of the substance of the cords is digested away leaving a three-dimensional network of 1.5-2.0-nm thick filaments. The interpretation is that the cords are formed of a plasmin-resistant core filament and a plasmin-extractable sheath. When plasmin treatment is prolonged for 15 h or longer, the filaments are dissociated and disappear, while basotubules are maintained. Plasmin digestion also reveals that basotubules are composed of two parts: a ribbon-like helical wrapping and tubule proper. Further changes in the tubule under plasmin influence are interpreted as a dissociation into pentagonal units suggestive of the presence of the amyloid P component. After 48 h of plasmin treatment, basotubules are further disaggregated and dispersed, leaving only linearly arranged double pegs. Reichert's membranes with or without a 2- hr plasmin treatment have been immunostained by exposure to antibodies against either laminin or type IV collagen with the help of peroxidase markers. The results indicate that the sheath of the cords contains laminin antigenicity, while the core filament contains type IV collagen antigenicity. It is proposed that Reichert's membrane consists mainly of a three-dimensional network of cords composed of a type IV collagen filament enclosed within a laminin-containing sheath. Also present are basotubules--which may contain the amyloid P component--and double pegs whose nature is unknown.     

Chondroitin sulfate proteoglycan is a constituent of the basement membrane in the rat embryo parietal yolk sac

Summary In addition to containing Type IV collagen, laminin and entactin, basement membranes contain small amounts of proteoglycans substituted primarily with heparan sulfate chains. We have previously shown, however, that parietal yolk sacs in organ culture synthesize predominantly chondroitin sulfate proteoglycan. In the present study, we have used histochemical and immunohistochemical techniques coupled with chondroitinase ABC digestion to provide evidence for the presence of chondroitin sulfate proteoglycan in the basement membrane (Reichert's membrane) of the 14.5-day rat embryo parietal yolk sac. The results revealed numerous cuprolinic blue-positive filaments and granules, 20–30 nm in greater length or diameter, dispersed throughout the thickness of the basement membrane. Both structures were removed by preincubating freshly isolated parietal yolk sacs with chondroitinase ABC. A similar labeling pattern was also obtained with immunoelectron microscopy using gold-labeled monoclonal anti-bodies directed against the three major isomers of protein-bound chondroitin sulfate. In contrast, coarser cuprolinic blue granules, 40–100 nm in diameter, were neither sensitive to chondroitinase ABC digestion nor labeled by the monoclonal antibodies. These results thus indicate that Reichert's membrane contains chondroitin sulfate proteoglycan in addition to heparan sulfate proteoglycan.     

Effects of L-asparaginase on rat embryonic development and yolk sac membrane in vitro

A comparative analysis of the teratogenic effects of L-asparaginase on 10.5- and 11.5-day rat embryos after 24 and 48 hours of exposure in vitro, respectively, were performed. Several medium concentrations of L-asparaginase (0.05, 0.25, and 1.5 IU/ml) were tested in both embryo series. Resulting embryos were submitted to morphological studies in a search for a specific route of pathogenesis. Morphological alterations of the visceral yolk sac were also studied to investigate its contribution to L-asparaginase teratogenicity in rats. Main embryonic malformations (open truncal neural tube, open encephalic vesicles, anophthalmia, lack of inversion, abnormal frontolateral protrusions, great vascular dilations at the cephalic level) and developmental retardation were already generated after the first 24 hours of culture (embryos of 10.5 days) and presented a dose-response relationship. Vascular dilations and neurulation disturbances seemed to be related to an early mesenchyme deficiency. Reduced number of mesenchymal cells was more evident in embryos of 10.5 days than those of 11.5 days, suggesting the existence of a later compensatory mechanism of cellular proliferation in the older embryo. Visceral yolk-sac endodermal cells at both embryonic stages were greatly deformed and enlarged by an increase of the high electron-dense vacuolar system. Therefore, both a blockage of the processes of lysosomal digestion and derived trophic deficiencies probably existed. A double teratogenic mechanism for L-asparaginase is postulated: a direct action mainly in younger embryos (before invagination of the embryo into the yolk sac) and a yolk sac-mediated one.     

Deposition of fibronectin and laminin in the basement membrane of the rat parietal yolk sac: immunohistochemical and biosynthetic studies

Rat parietal yolk sacs (PYS) at gestational ages 7.5, 9.5, 11.5, 13.5, 14.5, and 16.5 d were reacted with antibodies against laminin or plasma fibronectin. At all times studied, laminin consistently gave a positive reaction with Reichert's membrane and with the cytoplasm of PYS cells. In contrast, fibronectin gave a negative reaction with Reichert's membrane at day 7.5, was weakly positive at day 9.5, and from then on was increasingly positive with maximum reactivity at 14.5 d. By electron microscopic immunohistochemistry, antilaminin reacted strongly with 14.5-d Reichert's membrane and with the contents of the rough endoplasmic reticulum RER cisternae of the PYS cells. Antifibronectin had some spotty reactivity with Reichert's membrane, but the cytoplasm of the PYS cells was negative. The contents of the vitelline vessels and the interface between trophoblast and Reichert's membrane were strongly positive. Metabolic labeling of PYS cells in organ culture clearly demonstrated the presence of laminin, type IV procollagen, and entactin both in the medium and in tissues, but fibronectin was absent. No component in the medium bound to gelatin-Sepharose columns. These studies demonstrate that PYS cells, which actively synthesize and secrete basement membrane components, do not synthesize any detectable fibronectin. Furthermore, the anti-fibronectin staining pattern in the vitelline vessels and trophoblast-Reichert's membrane interface strongly suggests that the fibronectin present in Reichert's membrane is derived from the maternal circulation and is merely "trapped" in the membrane.     

Intracellular localization of basement membrane precursors in the endodermal cells of the rat parietal yolk sac. III. Immunostaining for laminin and its precursors

Reichert's membrane and the endodermal cells of the parietal yolk sac were examined for the presence of laminin antigenicity using anti-laminin antibodies and the peroxidase-antiperoxidase sequence. Immunostaining was observed through the full width of Reichert's membrane and within endodermal cells. In these cells immunostaining was observed in rough endoplasmic reticulum (rER) cisternae and Golgi apparatus. The Golgi staining could occur in any saccule, but predominated in components interpreted as the last saccule of the stack, the GERL element, and associated prosecretory granules. The secretory granules found in the ectoplasm were also immunostained. Finally, multivesicular bodies showed some staining. The immunostaining of Reichert's membrane indicates the presence of laminin itself, while that of rER cisternae and the Golgi apparatus is attributed to laminin precursors. Presumably the biosynthesis of laminin occurs along the usual protein pathway, that is, from rER through Golgi saccules and the GERL element to secretory granules, which release their content into Reichert's membrane. The laminin immunostaining of Reichert's membrane and endodermal cells is similar to that of type IV collagen. It is, therefore, likely that the two substances are processed and secreted simultaneously.     

Intracellular localization of basement membrane precursors in the endodermal cells of the rat parietal yolk sac. I. Ultrastructure and phosphatase activity of endodermal cells

The parietal layer of the rat yolk sac includes a 5 microliter thick sheet known as Reichert's membrane that exhibits properties of basement membranes. Its inner side is lined by a single layer of loosely distributed cells referred to as endodermal cells. Both Reichert's membrane and endodermal cells were examined at 13-14 days' gestation with emphasis on the ultrastructure of the Golgi apparatus, the identification of its component parts by specific phosphatase activities, and its possible role in the cells' secretory process. Reichert's membrane is composed of a series of stacked layers similar to basal laminae and composed of a network of fibrils with a diameter of 2-8 nm along which dots are located at irregular intervals. The endodermal cells contain the usual organelles, including interconnected rough endoplasmic reticulum (rER) cisternae and a prominent Golgi apparatus. With the help of phosphatase reactions, the stacks of Golgi saccules were divided into a) "phosphatase-free" saccules, the first ones on the cis or forming side, b) one or two "intermediate" saccules in the middle of the stacks, containing nicotinamide adenine dinucleotide phosphatase activity, c) one or two "last" saccules rich in thiamine pyrophosphatase activity on the trans or mature side, and d) continuing beyond the trans side, the GERL element displaying acid phosphatase activity. The latter is associated with profiles equally rich in acid phosphatase and tentatively considered to be prosecretory granules. Finally, the ectoplasm adjacent to Reichert's membrane displays large, acid phosphatase-containing structures tentatively considered to be secretory granules. Thus, the extensive rER network, the well-compartmentalized Golgi apparatus, and the presence of structures which may be prosecretory and secretory granules indicate that the endodermal cells are well-equipped for the secretion of the components of Reichert's membrane.     

Comparative enzyme histochemical study on the visceral yolk sac endoderm in the rat in vivo and in vitro

Summary Histochemical study of the visceral yolk-sac endoderm of the rat was performed in vitro (whole-embryo culture for 24, 48 and 72 h explanted at 9.5 days of gestation) and in vivo (10.5, 11.5 and 12.5 days of gestation) in order to compare the distribution and activity of various enzymes involved in the digestion and energy metabolism in both systems. It was shown that, both in vitro and in vivo -glytamyltransferase and dipeptidylpeptidase IV are demonstrable in the apical cell membranes (membranebound hydrolases), while acid phosphatase, dipeptidylpeptidases I, II and acid -galactosidase are concentrated in the supranuclear vacuoles (lysosomal hydrolases), and cytoplasmic lactate dehydrogenase and mitochondrial enzymes (succinate dehydrogenase, NAD-dependent isocitrate dehydrogenase, cytochrom oxidase) are localized in the whole cytoplasm and mainly in the apical cytoplasm, respectively, of the visceral yolk-sac epithelium. In vivo, the activity of all enzymes increased until 12.5 days, but in vitro, this activity increased only until 48 h after the start of culture (corresponding to 11.5 days in vivo). Comparison of the yolk sacs at 10.5 and 11.5 days in vivo with those after 24 and 48 h in vitro showed that the activities of all the investigated enzymes were almost identical. Yolk sacs which were cultured for 72 h showed lower activities of lysosomal and mitochondrial enzymes than those at 12.5 days in vivo. It is concluded that the digestive function and energy metabolism of the visceral yolk-sac epithelium are almost identical in vitro and in vivo at 10.5 and 11.5 days.Supported by the Deutsche Forschungsgemeinschaft 541/1-1)     

Comparative enzyme histochemical study on the visceral yolk sac endoderm in the rat in vivo and in vitro

Histochemical study of the visceral yolk-sac endoderm of the rat was performed in vitro (whole-embryo culture for 24, 48 and 72 h explanted at 9.5 days of gestation) and in vivo (10.5, 11.5 and 12.5 days of gestation) in order to compare the distribution and activity of various enzymes involved in the digestion and energy metabolism in both systems. It was shown that, both in vitro and in vivo gamma-glytamyltransferase and dipeptidylpeptidase IV are demonstrable in the apical cell membranes (membrane-bound hydrolases), while acid phosphatase, dipeptidylpeptidases I, II and acid beta-galactosidase are concentrated in the supranuclear vacuoles (lysosomal hydrolases), and cytoplasmic lactate dehydrogenase and mitochondrial enzymes (succinate dehydrogenase, NAD-dependent isocitrate dehydrogenase, cytochrom oxidase) are localized in the whole cytoplasm and mainly in the apical cytoplasm, respectively, of the visceral yolk-sac epithelium. In vivo, the activity of all enzymes increased until 12.5 days, but in vitro, this activity increased only until 48 h after the start of culture (corresponding to 11.5 days in vivo). Comparison of the yolk sacs at 10.5 and 11.5 days in vivo with those after 24 and 48 h in vitro showed that the activities of all the investigated enzymes were almost identical. Yolk sacs which were cultured for 72 h showed lower activities of lysosomal and mitochondrial enzymes than those at 12.5 days in vivo. It is concluded that the digestive function and energy metabolism of the visceral yolk-sac epithelium are almost identical in vitro and in vivo at 10.5 and 11.5 days.     

Sulfated glycosaminoglycan and collagen patterns in parietal yolk sac carcinoma (PYSC)

Electrophoretic analyses of collagenous material have shown that the parietal yolk sac carcinoma (PYSC) ascitic tumour synthesizes polypeptide chains that migrate as type IV procollagen. Having molecular weights of 185,000 and 160,000, these polypeptides are sensitive to collagenase. When the PYSC cells are injected subcutaneously, they form a solid tumour, and type I collagen predominates. The electrophoretic analyses of sulfated glycosaminoglycans and enzymatic degradation have shown a predominance of heparan sulfate in the ascitic tumour, and of chondroitin sulfate B in the solid tumour. Cells cultured from ascitic tumours have maintained the same collagen and sulfated glycosaminoglycan patterns as the original cells, whereas in the solid tumour culture only chondroitin sulfate AC has been detected.     

In vitro differentiation of mouse embryonic yolk sac cells

The embryonic yolk sac is the first site in the mammalian embryo in which cells are found that can carry out cell-mediated immune functions, yet the relation of cells of this primitive hematopoietic organ to the development of the mature immune system has not been established. We have initiated a series of experiments to determine the potential of cells of the mouse yolk sac to differentiate in vitro, in order to get an insight into the development of immunocompetence in this primary population of hematopoietic stem cells. The present paper describes the conditions promoting stem-cell differentiation and provides an initial characterization of cell surface phenotypes of the cell lineages established in vitro. Yolk sac cells obtained from 10- to 13-day mouse embryos were maintained in culture for more than 18 months, giving rise to a variety of cell types belonging to the hematopoietic lineages and culminating in the establishment of long-term cell lines. Supernatants of secondary mixed leukocyte cultures were found to be an effective source of growth factors promoting the initial differentiation as well as the maintenance of these cells. Flow-cytometric analysis showed that, in contrast to freshly obtained yolk sac cells, which had no detectable Thy 1 antigen, cells expressing significant levels of Thy 1 were obtained after 1 week or more of culture. Ly1 and Lyt 2 antigens were detected only rarely and the L3T4 (GK 1.5) antigen was never expressed.(ABSTRACT TRUNCATED AT 250 WORDS)     

The role of the yolk sac in copper metabolism during rat embryogenesis

Using the immunoblotting method, the synthesis of two copper-transporting P1-type ATPases, ATP7A (a candidate for the product of the Menkes disease gene) and ATP7B (presumed product of the Wilson disease gene), in the yolk sac cells of rat embryos at days 11 and 20 of embryogenesis was demonstrated. Concomitantly, yolk sac cells produce ceruloplasmin, a soluble copper-transporting glycoprotein, a proportion of which in secreted proteins progressively diminishes, attaining 5.2% at day 11 and 3.1% at day 20 of development. At different stages of embryogenesis, yolk sac cells synthesize two molecular forms of [14]C-ceruloplasmin, one of which is secreted towards the embryo, whereas the other, towards the decidual membrane. Two forms of ceruloplasmin secreted in polar directions differ in the rate of secretion. The role of the yolk sac as a key organ controlling the delivery and secretion of copper in the embryo during the postimplantation period is discussed.     

Radioautographic tracing of 3H-proline in the endodermal cells of the parietal yolk sac as an indicator of the biogenesis of basement membrane components

The biogenesis of basement-membrane components was investigated in the endodermal cells of the rat parietal yolk sac in 12.5-day pregnant rats; 3H-proline was injected into conceptuses. After various time intervals, the parietal yolk sac, including endodermal cells and the associated Reichert's membrane, was removed and processed for electron-microscopic radioautography. Silver grains were counted over endodermal cell organelles and Reichert's membrane. At 2 and 5 min after 3H-proline injection, endodermal cells showed heavy labeling in rough endoplasmic reticulum (rER). Silver grain density over the rER decreased from 2 to 20 min and then remained at a plateau. Grain density was moderate over the Golgi apparatus initially but rose to a peak at 2 hr and decreased by 4 hr and later. Grain density was negligible over secretory granules at 2 and 5 min and increased moderately with time to reach a maximum at 8 hr. Thus, radioautographic peaks occurred sequentially in rER, Golgi apparatus, and secretory granules. By 4 hr and later, silver grains accumulated over Reichert's membrane. These results indicated that endodermal cells incorporated labeled proline into substances which were processed from the rER through the Golgi apparatus, transported from there to the cell surface by secretory granules, and released for export to Reichert's membrane. To clarify the nature of the exported substances, the amount of label present in proline and hydroxyproline residues after 3H-proline injection was measured in Reichert's membrane with or without the associated endodermal cells. Within the cells, 61.8% of the labeled proteins were classified as "sedentary" and 38.2% as "exportable." Of the label exported to Reichert's membrane, 66.3% consisted of type IV collagen and the rest of other basement-membrane components. The results obtained with this model suggest that basement-membrane proteins, including type IV collagen, are elaborated by the associated cells through the classical pathway: rER-Golgi apparatus-secretory granules.     

A novel large-scale production system for modified basement membrane matrices using gene-swapped parietal endoderm cells.

Parietal endoderm-like cells, including Engelbreth-Holm-Swarm tumor and differentiated F9 embryonal carcinoma cells, produce huge amounts of basement membrane components, including laminin-1 (alpha1beta1gamma1). We employed a double-lox system-based gene-swapping strategy in F9 cells to replace the laminin alpha1 gene with a laminin alpha5 minigene. The gene-swapped F9 cells secreted laminin-10 (alpha5beta1gamma1) consisting of the exogenous alpha5 subunit and endogenous beta1 and gamma1 subunits on differentiation. The laminin-10 concentration in the conditioned medium exceeded 10 mg/l, which is 10-fold higher than the concentrations achieved by conventional recombinant expression systems. The gene-swapped F9 cells deposited basement membrane-like matrices containing laminin-10 on culture dishes, offering a novel microenvironment for in vitro cell manipulation.