The mechanism of yolk consumption was studied morphologically and biochemically in Japanese quail Coturnix japonica. The amount of yolk granules in the yolk (or ‘yolk cell’) decreased in two steps during embryonic development. In the first step, during days 0–4 of incubation, the yolk-granule weight decreased at a rate of 13 mg/day. This decrease was due to segregation by endodermal cells that were newly formed in the developing yolk sac. In the second step after day 6, the decrease was drastic at a rate of 29.8 mg/day during days 6–12 and very slow thereafter. The decrease at the second step was due to the enzymatic digestion of yolk granules by cathepsin D that coexisted in yolk spheres. This digesting reaction was triggered by the solubilization of the granules with high concentrations of salts that were supplied after disruption of the limiting membrane of yolk spheres. The ‘yolk cell’ seemed to die around day 5 of incubation. Thus the digestion products might be taken up together with yolk lipids by endocytosis into the endodermal cells and transported to blood vessels.
In fully matured oocytes of laying hens, the yolk mass consists of tightly packed, membrane-bound yolk spheres in which morphologically two components were identified, aggregates of 11.4 nm granules and a matrix formed of 26.7 nm particles (Perry & Gilbert 1985). As the yolk of a fertile egg is being enclosed by the developing yolk sac, absorption of the yolk occurs through the endoderm layer of the yolk sac (Burley & Vadehra 1989). On the basis of the similar morphological features of yolk material outside and inside endodermal cells, early researchers suggested that the extracellular yolk is phagocytosed in endodermal cells of the yolk sac (Lambson 1970; Mobbs & McMillan 1979, 1981). It was supposed that the yolk is digested by enzymes such as cathepsin D (Wouters et al. 1985), cathepsins B and D (Gerhartz et al. 1997, 1999), or alkaline and acid phosphatases (Juurlink & Gibson 1973) in the yolk sac. However, it is strange that the extracellular yolk treated was not the yolk spheres but the yolk granules, meaning that the yolk spheres may have lost their limiting membrane naturally or artificially.
A similar question arises regarding the presence of the cell membrane of avian yolk. There is no doubt it exists in growing oocytes. But some researchers believe it is lost after the establishment of a blastoderm (Bellairs 1964; a review by Wishart & Horrocks 2000), because of a reported lack of cell membrane in the vegetal region of ovulating eggs (Bakst & Howarth 1977). They speculate that an acellular yolk is enclosed by the yolk sac, consisting of three germ layers, with the progression of epiboly, and is absorbed. There may be some misunderstanding over these structures, because of an inferior methodology for the fixation of the large eggs of hens.
In avian eggs, the cleavage furrows separate the daughter blastomeres from each other but not from the yolk, so that the central blastomeres are continuous with the yolk at their lower ends, and the blastomeres lying on the circumference are continuous with the uncleaved cytoplasm at their outer edges (Patterson 1910; Balinsky 1965). In the present study, we refer to the ‘yolk’ as the ‘yolk cell’ when it is apparently alive or just the yolk when it is dead. The present study will show the sequence of yolk consumption that is intimately associated with the development of endodermal cells of the yolk sac and with the disintegration of the yolk cell in quail embryos.
Materials and Methods
Procurement of embryos
Fertilized eggs of Japanese quail Coturnix japonica were collected around 10.00 hours and put into an incubator at 39°C and in 60% humidity where they were automatically subjected to a see-saw motion. The embryonic development was measured in days of incubation.
Determination of yolk volume and weight
The albumen (egg white) around the embryo was removed manually in a 0.25 mol sucrose solution. After several washes with the solution, the yolk was punctured and suspended in 10 mL of the sucrose solution. The volume of the yolk suspension was measured in a 25 mL messcylinder and the yolk volume was determined by subtracting 10 mL from it.
To determine the yolk weight, the yolk suspension prepared as mentioned above in 5 mL of the sucrose solution was once treated with an ultrasonic disruptor UR-200P (Tomy-Seiko, Japan) and centrifuged at 1000 g for 10 min at 4°C. The precipitate (yolk granules) was freeze-dried and the weight was measured with a balance Mettler AE240 (Mettler-Toredo, Japan).
Embryos were immersed in 2.5% glutaraldehyde in 0.1 mol cacodylate buffer at pH 7.2 for 3 h at 4°C and the embryonic tissues that were isolated were further fixed in the same solution for 3 h. The specimens were rinsed in the buffer, post-fixed with 1% osmium tetroxide in the same buffer for 1 h, dehydrated in acetone, and embedded in epoxy resin. Thin sections were stained with uranyl acetate and lead citrate. Some thick sections were stained with toluidine blue.
Embryos from day 2–8 of incubation were rinsed several times with the 0.25 mol sucrose solution and exposed to air in Petri dishes. The transparent ‘thin-yolk’ solution in the yolk was collected directly with a syringe attached to a 27-G needle. The sample was then centrifuged at 10 000 g for 10 min at 4°C and the supernatant was diluted 1000× for potassium measurements or 10 000× for sodium measurements with distilled water. The samples each were dried at 100°C for 1 h, then hydrolyzed with 1 mL of 10 mol sulfuric acid and 0.2 mL of 30% hydrogen peroxide at 150°C for 1 h. The hydrolysis was repeated with 0.2 mL of 30% hydrogen peroxide at 200°C for 1 h, until the samples became transparent. After a final drying, the samples were dissolved in 0.1 mL of distilled water. The measurement of element concentrations was done with an atomic absorption spectrophotometer, the Hitachi 170–40 (Hitachi, Japan).
Yolks were obtained from infertile eggs by removing the egg white and vitelline membrane manually. Isolated yolks were mixed with two volumes of acetone and chilled to -20°C. The supernatant was removed following centrifugation at 10 000 g for 10 min. One hundred milligrams of the precipitate was suspended in 1 mL of 0.6 mol NaCl. Insoluble debris was removed by centrifugation. The supernatants were mixed with distilled water to adjust the sodium concentration with various pH solutions or with various inhibitors. The buffer system was constructed with 50 mmol citric acid and 100 mmol Na2HPO4, and the buffers were diluted five times in the experiments. The inhibitors used were obtained from suppliers as follows: Z-Phe-Phe-CHN2 and E-64 from the Peptide Institute, Japan; aprotinin from Nacalai Tesque, Japan; soybean trypsin inhibitor and NaF from Wako, Japan; leupeptin, phenylmethylsulfonyl fluoride (PMSF), chymostatin, bestatin and pepstatin A from Sigma, USA. After incubation at 37°C, the reaction mixtures were analyzed by SDS-PAGE on a 7.5% gel. Gels were stained with Coomassie Brilliant Blue.
Data were analyzed with Fisher's protected least significant difference. Statements of significance were based on P ≤ 0.01.
Changes of yolk volume and weight during the development
The volume of the yolk cell drastically increased (P = 0.0000) from 3.14 ± 0.31 mL (mean ± SD; n = 10) on day 0 of incubation to 6.86 ± 0.58 mL on day 4 (Fig. 1). During this period, an extraembryonic tissue, the yolk sac, extended in between the vitelline membrane and the surface of the yolk cell. On day 4 the leading edges of the yolk sac met at the vegetal pole and thus the yolk cell was completely enclosed. The volume of the yolk, now in the yolk sac, decreased drastically (P = 0.0000) to 3.13 ± 0.38 mL on day 8 and thereafter gradually to 0.37 ± 0.27 mL on day 17, the hatching day. The area vasculosa, which contains a network of blood vessels and advances following the area vitellina in the yolk sac, developed along with the extension of the yolk sac and completely surrounded the yolk on day 9.
The dry weight of yolk granules decreased in two steps (Fig. 1). It decreased from 352 ± 27 mg (n = 10) on day 0 to 300 ± 26 mg on day 4 in the first step; did not change significantly during days 4–6 (P = 0.7286); and again decreased drastically from 293 ± 30 mg on day 6 to 114 ± 21 mg on day 12 (P = 0.0000) and thereafter slowly to 42 ± 7 mg on day 17 in the second step. The rate was 13 mg/day during the first step and 29.8 mg/day in the second step of days 6–12.
Morphological changes of yolk cell
Figure 2a shows a light microscopical image of the vegetal pole of the yolk cell of a day 0 embryo. The cytoplasm was filled with yolk spheres in which two components were identified: electron-dense yolk granules and an electron-lucent matrix (Fig. 2b). The cell membrane of the yolk cell was apparently organized.
On day 3, the yolk sac covered approximately two-thirds of the surface of the yolk cell. The yolk sac at the area vitellina consisted of two cell types, flat ectodermal cells and tall endodermal cells (Fig. 2c). The endodermal cells possessed many yolk spheres, whereas the ectodermal cells had no yolk spheres (Fig. 2e). Since the cell boundaries of endodermal cells were irregular, a clear contour of the cells could not be defined. At the leading edge of the yolk sac, however, no endodermal cells were present and thus the ectodermal cells directly faced the yolk cell (Fig. 2d). A description of the area vasculosa of the yolk sac will be made later.
On day 5, the yolk sac at the area vitellina around the vegetal pole consisted of flat ectodermal cells and tall endodermal cells (Fig. 3a,d), similar to embryos on day 3. In the area vasculosa, many mesenchymal cells and blood vessels extended in between the ectodermal and endodermal cells at the periphery of the area (Fig. 3b), but few in its principal part (Fig. 3c). Ultrastructural features of the endodermal cells at the periphery were the same as those in the area vitellina (Fig. 3e) whereas in the principal part they were quite different (Fig. 3f). The cells in the principal part possessed vesicles, which were filled with yolk granules and small multivesicles, in addition to yolk spheres (Fig. 3g). A tall columnar shape was apparent in the cells. The orientation of the cells was also apparent (Fig. 3f); the apical portion facing the yolk and the basal portion facing the mesenchyme. The nucleus tended to reside in the basal portion. The cell membrane of the yolk cell was obscure.
On day 7 of incubation, the area vitellina of the yolk sac was limited to around the vegetal pole. It again consisted of ectodermal and endodermal cells only (Fig. 4a). The endodermal cells still possessed yolk spheres but had decreased in height, thus forming a cuboidal shape. The yolk was mostly surrounded by the yolk sac of the area vasculosa. The endodermal cell layer often folded toward the yolk (Fig. 4b). The cytoplasm of these cells was occupied mostly by vesicles which contained small multiparticles (Fig. 4c.d). Yolk granules were scarce in the cells. Prominent features were microvilli on the apical cell membrane and small vesicles with the same electron density as the yolk matrix in the apical cytoplasm (Fig. 4e). The yolk matrix seemed to be taken up in the cells by the indentation of cell membranes (Fig. 4f).
Breakdown of yolk spheres
In the yolk cell of embryos around day 4 of incubation, the mass of yolk spheres was maintained in the watery cytoplasm. Each yolk sphere apparently possessed a limiting membrane until day 5 (Fig. 5a). By day 7, some yolk spheres had lost the limiting membrane and their yolk granules and matrix were dispersed among still-intact yolk spheres (Fig. 5b). From day 9, the yolk mass dissociated and the yolk granules dispersed in the yolk with the breakdown of the sphere.
Concentration of Na and K in the watery cytoplasm of the yolk cell
The sodium concentration of the watery cytoplasm of the yolk cell was kept at around 170 mmol from days 2–5 of incubation (Fig. 6). The concentration gradually decreased thereafter to be 117 ± 15 mmol (n = 10) on day 8. The potassium concentration increased linearly from 12 ± 1.2 mmol on day 2 through to 29 ± 4.8 mmol on day 5 and 40 ± 4.9 mmol on day 8 (Fig. 6).
Digestion of yolk proteins
Acetone-treated yolk proteins, once solubilized with 0.6 mol NaCl, were incubated at pH 3.0 in various concentrations of NaCl adjusted with distilled water. SDS-PAGE analyses showed that the bands of yolk proteins decreased in intensity as the concentration of NaCl increased above 0.1 mmol (Fig. 7a). This result further suggested the inclusion of a digesting enzyme in the preparations of yolk proteins.
The yolk preparations were next incubated at various pHs with 0.3 mmol NaCl. Yolk proteins were affected by the enzymes at a pH lower than 6.0 and optimally digested at pH 3.0 (Fig. 7b). Among various protease inhibitors tested, pepstatin A completely inhibited the digestion (Fig. 7c), suggesting that the enzyme involved is an aspartic one.
Avian yolk spheres morphologically consist of two components, an aggregate of 11.4 nm yolk granules and a matrix formed of 26.7 nm particles (Perry & Gilbert 1985). The present study revealed a two-step decrease in the amount of yolk granules in a yolk cell, traditionally called the yolk, during embryonic development in quail. The first step, which occurs during day 0–4 of incubation, is apparently related to the formation of endodermal cells of the yolk sac as schematically represented in Fig. 8. The endodermal cells are produced by unequal cleavage at multiple sites at the periphery of the developing yolk sac (Patterson 1910). Then, some yolk spheres might be segregated by the newly formed endodermal cells, thus leading to a decrease of yolk granules in the yolk cell. The ectodermal cells in that portion seem not to participate in the segregation since we could not find any yolk spheres in them. They might multiply themselves and extend along the surface of the yolk cell with the progression of so-called epiboly. The second step of the granular decrease, which occurs mostly during the period from days 6–12 and to a lesser extent thereafter, is brought about by an enzymatical digestion of the yolk granules in the yolk which is now a mixture of the components of the broken yolk cell as discussed below.
Although researchers agree that growing oocytes in the ovarian follicle of aves possess a cell membrane around the yolk (Bellairs 1965; Burley & Vadehra 1989; Ito et al. 2003), there is some doubt as to its existence once the blastoderm has been established (Bellairs 1964; Mobb & McMillan 1979, 1981), presumably because of a reported lack of cell membrane in the vegetal region of ovulating eggs (Bakst & Howarth 1977). Thus it has been said that the yolk granules are phagocytozed and subsequently digested within the lysosomes of the yolk sac endodermal cells (Juurlink & Gibson 1973; Mobb & McMillan 1979, 1981). However, Gerhartz et al. (1999) recently reported that the quail embryo yolk sac stops metabolizing yolk on day 8 of incubation, the day when the rapid decrease of yolk granules occurs. Their study strongly suggests that there are two pathways of yolk metabolism depending on developmental stage before and after day 8.
During the first week of embryogenesis in quails, the enzymatic degradation of yolk proteins that are sequestered in endodermal cells of the yolk sac seems to be done by cathepsins D and B (Gerhartz et al. 1997, 1999). At least cathepsin D remains in the yolk spheres after participating in the processing of vitellogenin and plasma lipoproteins to yolk proteins (Ito et al. 2003), and still maintains its activity in oviposited eggs as shown in the present study. But we could not find cathepsin B activity in the yolk cell, suggesting that the enzyme is synthesized after the cellularization of endodermal cells. In Xenopus embryos, newly synthesized cathepsin B was transported to yolk platelets after the gastrular stages, where it completely digested yolk proteins together with the pre-existing cathepsin D (Yoshizaki & Yonezawa 1998). Some researchers advocate that the yolk proteins are degraded in lysosomes as mentioned earlier.
The present study showed that the cell membrane of the yolk cell is apparently maintained in the oviposited egg. During the first 4 days of incubation in quails, fluid accumulates in the yolk cell with the movement from the egg white through the blastoderm (Babiker & Baggott 1995; Latter & Baggott 2002). Thus, the volume of the yolk cell doubled on day 4 of incubation. On day 5, the yolk cell was completely surrounded by the yolk sac and its cell membrane became obscure. Thus we speculate that the yolk cell might be dead at this stage (Fig. 8). From day 4 onward, the volume of the yolk cell decreased. The fluid might have moved to other compartments, the allantoic cavity, the amniotic cavity and the growing tissues (Simkiss 1980).
From day 7, the limiting membrane of some yolk spheres broke down and the yolk granules mixed with the surrounding medium (Fig. 8). It is known in Xenopus that yolk solubilization is prerequisite to digestion by enzymes (Yoshizaki & Yonezawa 1996; Yoshizaki 1999) and yolk proteins can be solubilized with salt concentrations higher than the physiological level (Essner 1954). The concentrations of major elements in the watery cytoplasm of the yolk cell in quails were 170 mmol for sodium and 29 mmol for potassium on day 5. These values changed to 117 mmol for sodium and 40 mmol for potassium on day 8, presumably with the addition of the yolk matrix after disruption of the limiting membrane of yolk spheres. These values remained at the same level thereafter (unpublished data). Thus in quail also, the salt concentrations are maintained high enough to solubilize and digest the yolk proteins with the aid of cathepsin D (Ito et al. 2003) in the second step of the granular decrease. Although it has been reported in aves that the yolk granules were phagocytosed by endodermal cells (Lambson 1970) and digested in lysosomes also at these stages (Juurlink & Gibson 1973; Mobb & McMillan 1979, 1981), such digestion might not occur, or be rare, because few yolk granules were observed in the cells at these stages, and the rate of decrease in yolk-granule weight, 29.8 mg/day, in the second step was much higher than the 13 mg/day in the first step where the yolk granules were digested in endodermal cells. This statement however, does not exclude the endocytotic uptake and the digestion of lipids and some lipid-associated proteins by endodermal cells. Notably, endocytotic pits and vesicles were often observed on the apical surface of the cells. Another notable point is the sheer intensity of lipid transfer from the yolk to the embryo during the second half of the avian embryonic period (Speake et al. 1998).
The optimal pH of cathepsin D was around 3.0 for the digestion of yolk proteins in vitro. It is known in Xenopus that the yolk platelets show a constant pH of approximately 5.7 during oogenesis and become progressively more acidic (pH < 5) during embryogenesis (Fagotto & Maxfield 1994). The pH system for yolk digestion of aves seems much more complex in vivo since the yolk spheres contain large amounts of lipids in the matrix (Perry & Gilbert 1985; Burley & Vadehra 1989). Measurements of a simple yolk-supernatant showed a neutral pH (unpublished data). A precise measurement of the pH in the microenvironment around the yolk granules is awaited.
The rate of decrease of yolk-granule weight became low after day 12 of incubation. At just this stage, the egg white that is transported through the albumen sac via the extraembryonic cavity, the amniotic cavity, and the intestinal lumen, is added to the yolk mixture inside the yolk sac (Yoshizaki et al. 2002). It was also digested by the same cathepsin D (Yoshizaki et al. 2002). Thus the lowering of the rate might be induced by the common usage of the enzyme between the yolk proteins and egg white.
The digestion products of yolk granules as well as the egg white directly faced the endodermal cells after day 5 when the cell membrane of the yolk cell was destroyed. They might be taken up into endodermal cells together with yolk lipids via endocytosis, and transported through the cells to the blood vessels and circulated throughout the embryo.
We wish to thank Dr S. Ito for supplying the quail eggs and Dr T. Hara for allowing us to use an atomic absorption spectrophotometer.