- Email: [email protected]
Studies on the differentiation of egg envelopes
DEVELOPMENTAL
BIOLOGY
99,
473-481 (1983)
Studies on the Differentiation
of Egg Envelopes
I. The Starfish, Astropecten LUIGIA
SANTELLA,
ALBERTO
Stazime Received
January
MONROY,
aurantiacus AND FLORIANA
ROSATI
Zoologica, 80121 Naples, Italy 6, 1983;
accepted
in revised
.forrn May
11, 1983
We have studied the differentiation of the oocyte vitelline coat (VC) and jelly coat (JC) of the starfish, Astropecten The precursor material of both envelopes is secreted by the oocyte while the follicle cells do not appear to participate in the secretory process. The first indication of differentiation of the VC is the deposition of a fine fibrillar material between the microvilli which emerge from the oocyte surface. External to this, a more loosely organized material becomes the precursor of the JC. At this time both layers are periodic acid-Schiff (PAS)-positive. In a later stage, the material between the microvilli acquires a more compact organization, looses its PAS-positivity while acquiring fucose binding protein (FBP) affinity. On the contrary, the JC remains PAS-positive and FBP-negative. In the full grown oocytes the VC is made up of densely packed fibrils oriented tangentially to the oocyte surface and is tightly bound to the microvilli. The observations are discussed in connection with the problem of the role of the egg envelopes in sperm-egg recognition and in the induction of the acrosome reaction. aurantiacus.
INTRODUCTION
In this paper we address the question of the differentiation of the vitelline coat (VC) and of the jelly coat (JC) in the starfish oocyte. It has been shown that the VC of the ascidians (Cotelli et al., 1981; Rosati et aL, 1982) and the zona pellucida (which is the equivalent of the VC) of the mouse (Bleil and Wassarman, 1980; Greve et al., 1982) are products of the oocyte. These observations are important in connection with the fact that in both these animals the VC is the carrier of the sperm receptors and is the site of the acrosome reaction (see review by Monroy and Rosati, 1983), two properties on which the specificity of the sperm-egg interaction depends. In the echinoderms the situation is complicated by the presence of a jelly layer which is external to the VC. While also in the sea urchin the sperm receptors appear to be components of the VC (Glabe and Vacquier, 1978; Glabe and Lennarz, 1981; Rossignol et cd., 1981; Glabe et al., 1982) whether the site of the acrosome reaction is the VC or the JC is still a debated question (see Monroy and Rosati (1983) for a discussion). On the other hand, it is well documented that in the starfish the acrosome reaction is elicited upon contact of the spermatozoon with the JC (Colwin and Colwin, 1956; Tilney et ah, 1973). This raises the question of the respective roles of the VC and of the JC in starfish and in sea urchin. In view of the fact that asteroids and echinoids diverged some 500 million years ago, it would not be surprising if the respective roles of the two oocyte envelopes in sperm-egg interaction were somewhat dif-
ferent. Accordingly, we have undertaken a study of the differentiation of the VC and of the JC in the starfish and in the sea urchin. Here we report observations on the starfish, Astropecten aurantiacus, while the results of the work in progress on the sea urchin will be described in a forthcoming paper. Specifically, the question we address is whether or not the VC and the JC of the starfish oocyte originate from a common source; i.e., from the oocyte or from the follicle cells, or alternatively, the former from the oocyte and the latter from the follicle cells. Should indeed both envelopes originate from a common source then the components responsible for the specific sperm recognition and for triggering the acrosome reaction may be supposed to be present in both of them, and their preferential (or exclusive) localization in one of them may be the result of evolutionary conditions. The observations described in this paper suggest that in Astropecten the VC and the JC are a product of the oocyte with little if any participation of the follicle cells. MATERIALS
AND METHODS
Ovarian fragments of the starfish, A. aurantiacus, were fixed 6 hr at 4°C in Karnowsky’s fixative modified for marine animals: 5 parts (w/v) paraformaldehyde + 1 part 5% (w/v) glutaraldehyde in filtered seawater + 4 parts 0.2 M sodium cacodylate (pH 7.3). After washing, the material was postfixed 2 hr at 4°C in 1% 0~0~ in 0.2 M cacodylate (pH 7.3). The samples were dehydrated in alcohol at -20°C and embedded in Spurr’s
474
DEVELOPMENTAL
BIOLOGY
resin (TAAB). The sections were cut with a Reichert ultramicrotome and observed with a Philips 400 electron microscope. For the localization of polysaccharides, the periodic acid thiosemicarbazide-silver protein method (Thiery, 1967) was employed. The sections were floated for 2025 min on 1% periodic acid and then treated with 1% semicarbazide in 10% acetic acid for 40 min for glycogen staining and for 24 hr for polysaccharide staining. The sections were then floated on 1% silver protein for 30 min. For the detection of fucosyl sites oocytes obtained by shaking ovarian fragments in sea water were incubated in 125 ~1 sea water containing 62 pg FBP (fucose binding protein, derived from Lotus tetragmolobus, and purchased from Miles & Co.). After 30 min at 0°C the eggs were washed several times in sea water, resuspended in 175 ~1 of sea water containing 100 pg of fucosyl ferritin, and incubated at 0°C for 20 min. The eggs were then washed and processed for electron miroscopy. For the light microscope observations, l-pm sections were stained with toluidine blue. RESULTS
Figures 1 and 2 show young oocytes at a stage of their growth prior to the appearance of the extracellular envelopes. The oocytes are surrounded by a single layer of follicle cells. The plasma membranes of the oocyte and of the follicle cells are closely apposed and junctional complexes are occasionally present between them (Fig. 5). At this stage the oocyte cytoplasm is rich in ribosomes, and Golgi complexes are present in the cortical region (Fig. 5). Small flattened vesicles are also abundant near the Golgi complexes whereas no such vesicles are found in the rest of the cytoplasm (Fig. 5). In contrast, the cytoplasm of the follicle cells is very poor in endoplasmic reticulum as well as in ribosomes and Golgi complexes (Fig. 5). Their body is often filled with large vesicles. Some of them are empty and others contain a coarse granular material (Fig. 5). The secretion of the egg envelope precursor material begins at the site where the oocyte is attached to the ovarian wall. In this area there are large irregular interstices between the oocyte surface and the follicle cells (Fig. 3). These interstices appear to derive from the confluence of vacuoles, also visible in the light microscope, which are concentrated mainly in the subcortical region of the oocyte (Fig. 3); in later stages they are also present throughout the cytoplasm (Fig. 4). At the electron microscope level these vacuoles appear as empty vesicles with a bilaminar wall; the inner wall is distinctly visible, being partially detached from the inner one (Fig. 6), probably as a result of shrinkage. The vesicles’appear
VOLUME
99, 1983
to be extruded as “balloons” between the oocyte surface and the follicle cells thus pushing the latter away from the oocyte (Fig. 6). The walls of the balloons then break open; their fragmented remnants are occasionally to be seen in the interstices between the oocyte and the follicle cells (Fig. 7). In larger oocytes the interstices extend all around the oocyte thus separating it from the follicle cells (Figs. 7 and 8). Concurrently with the growth of the oocyte and the enlargement of the interstices between the oocyte and the follicle cells, the latter become very thin and their cytoplasm is still very poor in organelles (Figs. 7 and 8). Follicle cell processes making contact with the oocyte surface have never been observed. After the detachment of the follicle cells from the oocyte, sparse and irregularly shaped microvilli emerge from the oocyte surface (Figs. 7 and 8) and a fluffy, delicate fibrillar material becomes detectable between them (Fig. 9). The cortical region of the oocyte is now rich in vesicles of different sizes (Figs. 8 and 9). Numerous flattened vesicles similar to those described in the proximity of the Golgi complexes are also present (Fig. 9). They contain a fibrillar material similar in appearance to that observed between the microvilli. In sections submitted to Thiery’s reaction the fluffy material is periodic acid-Schiff (PAS)-positive (Fig. 10). A similar PAS-positive material is present also in some of the subcortical vesicles (Fig. 10). In later stages of the oocyte growth the fluffy material, now forming a fine meshwork, fills the entire space between the oocyte and the follicle cells (Figs. 10 and 12). At this time the material between the microvilli is much more compact than the outer material and the fibrils become more orderly oriented with respect to the oocyte surface. This layer is the VC precursor, whereas the more loosely arranged material external to it is the JC precursor. Furthermore, whereas the inner compact layer is PASnegative the outer more loosely organized envelope still retains its PAS positivity (Fig. 13). In contrast, Ferritinconjugated FBP binds quite strongly to the dense layer between the microvilli, but fails to bind to the external loose layer or to the plasma membrane of the oocyte (Fig. 14). The final stage of the differentiation of the oocyte envelopes is observed in full grown oocytes, which can be operatively described as those that are released when ovarian fragments are shaken in sea water. In these oocytes the microvilli span the entire depth of the VC (which is about 1 pm thick); their tips are, however, always covered by a thin layer of fibrillar material (Fig. 15). The VC appears to be tightly bound to the plasma membrane. Figure 15 also shows that although the VC has become detached from the oocyte surface it is firmly bound to the plasma membrane of the microvilli. In
FIG. 1. Thick section of a young oocyte close to the ovarian wall (ow). arrows, follicle cells. X1500. FIG. 2. Electron micrograph of an oocyte at the same stage as in Fig. 1. arrows, follicle cells; ow, ovarian wall. X4200. FIG. 3. Thick section of an oocyte in a later stage of growth. Clear spaces (arrows) between the oocyte and the ovarian wall (ow); es in the oocyte cytoplasm adjacent to the ovarian wall. X540. FIG. 4. Larger oocyte. Vacuoles are present in the whole cytoplasm; their concentration is greater at the side adjacent to the ovarian (ow). >(1500.
475
many wall
Fl :G. 5. Electron micrograph of a young oocyte (0) with a follicle cell (f). Their plasma membranes are still closely apposed and a juncti plex is seen between them (large arrow). A large Golgi complex and some flattened vesicles are present in the cortical region of ’ the te (small arrows). Many big vacuoles (v) containing coarse material occupy the cytoplasm of the follicle cell. X8550. Fl .G. 6. Electron micrograph showing vacuoles both in the cytoplasm of the oocyte and between the oocyte (0) and the follicle cells w. The vacuoles have a bilaminar wall and appear empty. X14,250.
SANTELLA,
MONROY,
AND
RosATI
addition,
the fibrillar organization of the VC, which is clearly visible between the microvilli, is less distinct where the VC is attached to the microvilli, possibly because of the denser packaging of the fibrils and of their more irregular orientation. At several points this material is detached from the surface of the microvilli, although connection is retained by thin threads. Follicle cells processes crossing the JC and making contact with the oocyte surface-such as those described in Patiria miniata (Schroeder et al., 19’79) have never been observed in Astropecten. In the full-grown oocytes we have observed occasional local thickenings (m-30-40 nm) between the microvilli (Fig. 12). These may correspond to the postjunctional components described in another starfish, Pisaster ochraceus, and interpreted as the 1-methyladenine receptor sites (Schroeder, 1981). DISCUSSION
The observations described in this paper suggest that in the starfish, A. aurantiacus, the oocyte is the primary-and possibly the only-source of material from which the oocyte envelopes, the VC and the JC, differentiate. However, a small contribution from the follicle cells, cannot entirely be ruled out, as the enormous secretory activity of the oocyte would obscure a comparatively minute contribution of the follicle cells. Also, an extremely rapid secretory cycle by the follicle cells, though unlikely, would go undetected in the static electron microscopic pictures. Indeed, a limitation of this work is that it is based exclusively on ultrastructural evidence, as we have not yet succeeded in satisfactorily labeling the ovary of Astropecten with radioactive precursors. With this limitation in mind our observations lend themselves to some considerations. The origin of the VC and of the JC from the oocyte suggests that the two envelopes should be considered a functional unit; in fact., they can be viewed as the system involved in the processes of species-specific sperm recognition and of triggering the acrosome reaction. Starfishes and holothurians are the only echinoderms in which there is unequivocal evidence that the acrosome reaction is triggered upon contact of the spermatozoon with the JC (see Monroy and Rosati (1983), for a discussion). Hoshi and colleagues (Uno and Hoshi, 1978; Ikedai and Hoshi, 1981a,b) have isolated and purified a glycoprotein component from the JC of Asterias amurensis, which induces the acrosome reaction. It is likely that in the asteroids the JC is responsible not, only for
Egg Envelopes in StarJish
477
triggering the acrosome reaction but also for the speciesspecific sperm recognition (see Metz, 1945). This raises the question of the role of the VC in these animals. In tunicates and mammals the sperm receptors are cornponents of the VC, which is also the site of the acrosome reaction; and this is likely to be the case also in the sea urchin, though the question is still controversial (see Monroy and Rosati, 1982,1983). We suggest that in the asteroids, the VC contains receptors which interact with complementary groups of the acrosomal process or with a “bindin-like” protein (for a review on bindin see Vacquier, 1983). However, as in the asteroids and holothurians the acrosome reaction occurs far from the oocyte surface, one should assume that the acrosomal process spans the whole thickness of the JC ensheathed with bindin. This would be interesting to investigate. We further surmise that the molecules involved in the speciesspecific recognition between gametes and in the triggering of the acrosome reaction were originally present throughout the oocyte envelopes and that they gradually localized in the VC in the course of evolution (see also Monroy and Rosati, 1983). Our observations throw no light on the question as to whether the VC and the JC originate from a common precursor material or whether they are the products of two independent secretory processes. The histochemical tests show that the material which is secreted first is PAS-positive. In a later stage, i.e., in the larger oocytes, the material between the microvilli acquires a more compact organization with the fibrils running parallel to the oocyte surface: this layer is the VC primordium. At this point the histochemical tests we have used show a clear-cut. distinction between this VC primordium and the outer material: the VC has lost its PAS positivity and has become strongly FBP-positive while the outer layer, which can be indicated as the JC, is still PASpositive but fails to bind the FBP. Two interpretations of these observations are possible. (1) The material that is secreted first gives rise to the JC, the precursor of the VC being the product of a later secretory wave. (2) Both the VC and the JC originate from a common precursor material secreted by the oocyte and are differentially acted upon by the follicle cells and/or the oocyte. This would imply that the difference between the two envelopes is essentially in their supramolecular organization. Finally, our observations show that in the full grown oocytes the connections between the VC and the microvilli are very intimate. Whether or not they are com-
FIG. 7. An oocyte in a more advanced stage of growth. Two large interstices between the oocyte (0) and the follicle cells (fc) containing remnants of the walls of the extruded vacuoles. Part of a follicle cell still remains apposed to the oocyte. X8550. FIG. 8. Many interstices between the oocyte (0) and thin stretched follicle cells (fc). Irregular microvilli (small arrows) emerged from the plasma membrane of the oocyte. ow, ovarian wall. X17,100.
FI G. 9. A large interstice between the oocyte (0) and the follicle cells. Fluffy material begins to be deposited between the micro\ rilli. Nun lerous vesicles containing fibrillar material are present in the cortical region of the oocyte (arrows). X5320. A :G. 10. An oocyte stained with Thy&y’s method. Silver grains are present in the plasma membrane of the oocyte (0) and in the fl’UffY erial around the microvilli. A vacuole containing fluffy PAS-positive material close to the oocyte plasma membrane (*). Silver grains are also present in the cortical granules and in other cytoplasmic components. X25,320. FI :G. 11. An oocyte in a more advanced stage of growth. The space between the oocyte and the very thin follicle cells (fc) is filled with the while a more compact fibrillar material can be seen between the microvilli. X15,675. jelly ’ coat (jc) material 478
FIG. the fit FIG. to the FIG. no gri
12. Higher magnification of an oocyte (0) at the same stage of growth. The jelly coat (jc) appears as a fine fibrillar meshwork while nillar material between the microvilli appears more compact and begins to acquire a more orderly oriented organization. X27,790. 13. A final stage of the differentiation of the egg envelopes: treatment with FBP followed by fucosyl-ferritin. The labeling is limited vitelline coat (vc); no labeling of the jelly coat (jc). o, oocyte. X62,700. 14. An oocyte at the same stage. The Thy&y’s reaction is positive in the plasma membrane; the jelly coat (jc) is weakly positive and rins are present in the vitelline coat (vs). o, oocyte. X44,415. 479
480
DEVELOPMENTAL BIOLOGY VOLUME99,1983
FIG. 15. In a full grown oocyte (0) the fibrillar organization of the vitelline coat (vc) is distinctly visible; the fibrils are oriented tangentially to the oocyte surface and are attached to the plasma membrane of the microvilli. In the jelly coat (jc) the fibrillar meshwork is much more loose and the flbrils are irregularly oriented. X82,800.
parable to the “vitelline posts” described in the sea urchin egg (Kidd, 1978; Chandler and Heuser, 1980) cannot be decided on the basis of the present observations. Work is in progress to study the changes that occur in these connections in the course of oocyte maturation and as a result of fertilization. We thank Professor F. Kellenberg, Basel, for suggesting the lowtemperature dehydration procedure. We also thank Dr. B. Dale for his critical reading of the paper and G. Princivalli for typing. F.R. is on leave from the University of Siena. REFERENCES BLEIL, J. D., and WASSARMAN, P. M. (1980). Synthesis of zona pellucida proteins by denuded and follicle-enclosed mouse oocytes during culture in vitro. Proc Natl Ad Sti USA 77,1029-1033. CHANDLER, D. E., and HEUSER, J. (1980). The vitelline layer of the sea urchin egg and its modification during fertilization. J. CeU BioL 84, 618-632.
COLWIN, L. H., and COLWIN, A. L. (1956). The acrosome filament and sperm entry in Thyme tiara (Holothuria) and As-. BioL Bull 110.243-257. COTELLI, F., ANDRONICO, F., DE SANTIS, R., MONROY, A., and ROSATI, F. (1981). Differentiation of the vitelline coat in the ascidian Ciona intestinalis: An ultrastructural study. Roux’s Arch Lkv. BioL 190, 252-258. GLABE, C. G., and VACQUIER, V. D. (1978). Egg surface glycoprotein receptor for sea urchin bindin. Proc. NatL Acad Sti USA 75, 881885. GLABE, C. G., and LENNARZ, W. J. (1981). Isolation and partial characterization of a high molecular weight glyconjugate derived from the egg surface, which is implicated in sperm-egg adhesion. .I. .%pram.& S-u& C& Biochem 15,387-394. GLABE, C. G., GRABEL, L. B., VACQUIER, V. D., and ROSEN, S. D. (1982). Carbohydrate specificity of sea urchin sperm bindin: A cell surface lectin mediating sperm-egg adhesion. J. Cell BioL 94, 123-128. GREVE, J. M., SAHAMANN, G. S., ROLLER, R. J., and WASSARMAN, P. M. (1982). Biosynthesis of the major zona pellucida glycoprotcin secreted by oocytes during mammalian oogenesis. Cell 31,749-‘759. IKADAI, H., and HOSHI, M. (1981a). Biochemical studies on the acrosome
SANTELLA,
MONROY,
AND
ROSATI
reaction of the starfish, Aster& amurensis. 1. Factors participating in the acrosome reaction. Dev. Growth D@w. 23,73-80. IKADAI, H., and HOSHI, M. (1981b). Biochemical studies on the acrosome reaction of the starfish Aster& amurensti. 2. Purification and characterization of acrosome reacting-inducing substance. Dev. Grcxdh D$er. 23, 81-88. KIDD, P. (1978). The jelly and vitelline coats of the sea urchin egg: New ultrastructural features. J. Ultrastruct. Res. 64, 204-215. METZ, C. B. (1945). The agglutination of starfish sperm by fertilizing. Biol BulL 89, 84-94. MONROY, A., and ROSATI, F. (1982). On the molecular mechanisms of sperm-egg interaction. Cell Di&r. 11.299-301. MONROY, A., and ROSATI, F. (1983). A comparative analysis of spermegg interaction. Gamete Rex 7, 85-102. ROSATI, F., COTELLI, F., DE SANTIS, R., MONROY, A., and PINTO, M. R. (1982). Synthesis of fucosyl-containing glycoproteins of the vitelline coat in oocytes of Ciona intestinalis (Ascidia). Proc. NatL Acd Sci. USA 79, 1098-1911. ROSSIGNOL, D. P., ROSCHELLE, A. J., and LENNARZ, W. J. (1981). Spermegg binding: Identification of a species-specific sperm receptor from
Egg
Envelopes
in
481
Star$sh
eggs of Strongylocentrotus purpwatus. J. SupramoL Struct. Cell Biochem. 15, 347-358. SCHROEDER, T. E. ( 1981). Microfilament mediated surface change in starfish oocytes in response to 1-methyladenine: Implications for identifying the pathway and receptor sites for maturation-inducing hormones. J. Cell BioL 90, 362-371. SCHROEDER P. C., LARSEN, J. H. and WALDO, A. E. (1979). Oocytefollicle cell relationships in a starfish. Cell Tissue Res. 203.249-256. THICRY, J. P. (1967). Mise en 6vidence des polysaccharides sur coupes fines en microscopic ilectronique. J. Microsc. 6, 987-1018. TILNEY, L. G., HATANO, S., ISHIKAWA, H., and MOOSEKER, M. S. (1973). The polymerization of actin: Its role in the generation of the acrosomal process in certain echinoderm sperm. J. Cell BioL 59,109126. UNO, Y., and HOSHI, M. (1978). Separation of the sperm and the acrosome reaction-inducing substance in egg jelly Science 200, 58-59. VACQUIER, cellulose
V. D. (1983). chromatography.
Purification Anal
of sea urchin bindin B&hem. 129,497-501.
agglutinin of starfish. by DEAE
BIOLOGY
99,
473-481 (1983)
Studies on the Differentiation
of Egg Envelopes
I. The Starfish, Astropecten LUIGIA
SANTELLA,
ALBERTO
Stazime Received
January
MONROY,
aurantiacus AND FLORIANA
ROSATI
Zoologica, 80121 Naples, Italy 6, 1983;
accepted
in revised
.forrn May
11, 1983
We have studied the differentiation of the oocyte vitelline coat (VC) and jelly coat (JC) of the starfish, Astropecten The precursor material of both envelopes is secreted by the oocyte while the follicle cells do not appear to participate in the secretory process. The first indication of differentiation of the VC is the deposition of a fine fibrillar material between the microvilli which emerge from the oocyte surface. External to this, a more loosely organized material becomes the precursor of the JC. At this time both layers are periodic acid-Schiff (PAS)-positive. In a later stage, the material between the microvilli acquires a more compact organization, looses its PAS-positivity while acquiring fucose binding protein (FBP) affinity. On the contrary, the JC remains PAS-positive and FBP-negative. In the full grown oocytes the VC is made up of densely packed fibrils oriented tangentially to the oocyte surface and is tightly bound to the microvilli. The observations are discussed in connection with the problem of the role of the egg envelopes in sperm-egg recognition and in the induction of the acrosome reaction. aurantiacus.
INTRODUCTION
In this paper we address the question of the differentiation of the vitelline coat (VC) and of the jelly coat (JC) in the starfish oocyte. It has been shown that the VC of the ascidians (Cotelli et al., 1981; Rosati et aL, 1982) and the zona pellucida (which is the equivalent of the VC) of the mouse (Bleil and Wassarman, 1980; Greve et al., 1982) are products of the oocyte. These observations are important in connection with the fact that in both these animals the VC is the carrier of the sperm receptors and is the site of the acrosome reaction (see review by Monroy and Rosati, 1983), two properties on which the specificity of the sperm-egg interaction depends. In the echinoderms the situation is complicated by the presence of a jelly layer which is external to the VC. While also in the sea urchin the sperm receptors appear to be components of the VC (Glabe and Vacquier, 1978; Glabe and Lennarz, 1981; Rossignol et cd., 1981; Glabe et al., 1982) whether the site of the acrosome reaction is the VC or the JC is still a debated question (see Monroy and Rosati (1983) for a discussion). On the other hand, it is well documented that in the starfish the acrosome reaction is elicited upon contact of the spermatozoon with the JC (Colwin and Colwin, 1956; Tilney et ah, 1973). This raises the question of the respective roles of the VC and of the JC in starfish and in sea urchin. In view of the fact that asteroids and echinoids diverged some 500 million years ago, it would not be surprising if the respective roles of the two oocyte envelopes in sperm-egg interaction were somewhat dif-
ferent. Accordingly, we have undertaken a study of the differentiation of the VC and of the JC in the starfish and in the sea urchin. Here we report observations on the starfish, Astropecten aurantiacus, while the results of the work in progress on the sea urchin will be described in a forthcoming paper. Specifically, the question we address is whether or not the VC and the JC of the starfish oocyte originate from a common source; i.e., from the oocyte or from the follicle cells, or alternatively, the former from the oocyte and the latter from the follicle cells. Should indeed both envelopes originate from a common source then the components responsible for the specific sperm recognition and for triggering the acrosome reaction may be supposed to be present in both of them, and their preferential (or exclusive) localization in one of them may be the result of evolutionary conditions. The observations described in this paper suggest that in Astropecten the VC and the JC are a product of the oocyte with little if any participation of the follicle cells. MATERIALS
AND METHODS
Ovarian fragments of the starfish, A. aurantiacus, were fixed 6 hr at 4°C in Karnowsky’s fixative modified for marine animals: 5 parts (w/v) paraformaldehyde + 1 part 5% (w/v) glutaraldehyde in filtered seawater + 4 parts 0.2 M sodium cacodylate (pH 7.3). After washing, the material was postfixed 2 hr at 4°C in 1% 0~0~ in 0.2 M cacodylate (pH 7.3). The samples were dehydrated in alcohol at -20°C and embedded in Spurr’s
474
DEVELOPMENTAL
BIOLOGY
resin (TAAB). The sections were cut with a Reichert ultramicrotome and observed with a Philips 400 electron microscope. For the localization of polysaccharides, the periodic acid thiosemicarbazide-silver protein method (Thiery, 1967) was employed. The sections were floated for 2025 min on 1% periodic acid and then treated with 1% semicarbazide in 10% acetic acid for 40 min for glycogen staining and for 24 hr for polysaccharide staining. The sections were then floated on 1% silver protein for 30 min. For the detection of fucosyl sites oocytes obtained by shaking ovarian fragments in sea water were incubated in 125 ~1 sea water containing 62 pg FBP (fucose binding protein, derived from Lotus tetragmolobus, and purchased from Miles & Co.). After 30 min at 0°C the eggs were washed several times in sea water, resuspended in 175 ~1 of sea water containing 100 pg of fucosyl ferritin, and incubated at 0°C for 20 min. The eggs were then washed and processed for electron miroscopy. For the light microscope observations, l-pm sections were stained with toluidine blue. RESULTS
Figures 1 and 2 show young oocytes at a stage of their growth prior to the appearance of the extracellular envelopes. The oocytes are surrounded by a single layer of follicle cells. The plasma membranes of the oocyte and of the follicle cells are closely apposed and junctional complexes are occasionally present between them (Fig. 5). At this stage the oocyte cytoplasm is rich in ribosomes, and Golgi complexes are present in the cortical region (Fig. 5). Small flattened vesicles are also abundant near the Golgi complexes whereas no such vesicles are found in the rest of the cytoplasm (Fig. 5). In contrast, the cytoplasm of the follicle cells is very poor in endoplasmic reticulum as well as in ribosomes and Golgi complexes (Fig. 5). Their body is often filled with large vesicles. Some of them are empty and others contain a coarse granular material (Fig. 5). The secretion of the egg envelope precursor material begins at the site where the oocyte is attached to the ovarian wall. In this area there are large irregular interstices between the oocyte surface and the follicle cells (Fig. 3). These interstices appear to derive from the confluence of vacuoles, also visible in the light microscope, which are concentrated mainly in the subcortical region of the oocyte (Fig. 3); in later stages they are also present throughout the cytoplasm (Fig. 4). At the electron microscope level these vacuoles appear as empty vesicles with a bilaminar wall; the inner wall is distinctly visible, being partially detached from the inner one (Fig. 6), probably as a result of shrinkage. The vesicles’appear
VOLUME
99, 1983
to be extruded as “balloons” between the oocyte surface and the follicle cells thus pushing the latter away from the oocyte (Fig. 6). The walls of the balloons then break open; their fragmented remnants are occasionally to be seen in the interstices between the oocyte and the follicle cells (Fig. 7). In larger oocytes the interstices extend all around the oocyte thus separating it from the follicle cells (Figs. 7 and 8). Concurrently with the growth of the oocyte and the enlargement of the interstices between the oocyte and the follicle cells, the latter become very thin and their cytoplasm is still very poor in organelles (Figs. 7 and 8). Follicle cell processes making contact with the oocyte surface have never been observed. After the detachment of the follicle cells from the oocyte, sparse and irregularly shaped microvilli emerge from the oocyte surface (Figs. 7 and 8) and a fluffy, delicate fibrillar material becomes detectable between them (Fig. 9). The cortical region of the oocyte is now rich in vesicles of different sizes (Figs. 8 and 9). Numerous flattened vesicles similar to those described in the proximity of the Golgi complexes are also present (Fig. 9). They contain a fibrillar material similar in appearance to that observed between the microvilli. In sections submitted to Thiery’s reaction the fluffy material is periodic acid-Schiff (PAS)-positive (Fig. 10). A similar PAS-positive material is present also in some of the subcortical vesicles (Fig. 10). In later stages of the oocyte growth the fluffy material, now forming a fine meshwork, fills the entire space between the oocyte and the follicle cells (Figs. 10 and 12). At this time the material between the microvilli is much more compact than the outer material and the fibrils become more orderly oriented with respect to the oocyte surface. This layer is the VC precursor, whereas the more loosely arranged material external to it is the JC precursor. Furthermore, whereas the inner compact layer is PASnegative the outer more loosely organized envelope still retains its PAS positivity (Fig. 13). In contrast, Ferritinconjugated FBP binds quite strongly to the dense layer between the microvilli, but fails to bind to the external loose layer or to the plasma membrane of the oocyte (Fig. 14). The final stage of the differentiation of the oocyte envelopes is observed in full grown oocytes, which can be operatively described as those that are released when ovarian fragments are shaken in sea water. In these oocytes the microvilli span the entire depth of the VC (which is about 1 pm thick); their tips are, however, always covered by a thin layer of fibrillar material (Fig. 15). The VC appears to be tightly bound to the plasma membrane. Figure 15 also shows that although the VC has become detached from the oocyte surface it is firmly bound to the plasma membrane of the microvilli. In
FIG. 1. Thick section of a young oocyte close to the ovarian wall (ow). arrows, follicle cells. X1500. FIG. 2. Electron micrograph of an oocyte at the same stage as in Fig. 1. arrows, follicle cells; ow, ovarian wall. X4200. FIG. 3. Thick section of an oocyte in a later stage of growth. Clear spaces (arrows) between the oocyte and the ovarian wall (ow); es in the oocyte cytoplasm adjacent to the ovarian wall. X540. FIG. 4. Larger oocyte. Vacuoles are present in the whole cytoplasm; their concentration is greater at the side adjacent to the ovarian (ow). >(1500.
475
many wall
Fl :G. 5. Electron micrograph of a young oocyte (0) with a follicle cell (f). Their plasma membranes are still closely apposed and a juncti plex is seen between them (large arrow). A large Golgi complex and some flattened vesicles are present in the cortical region of ’ the te (small arrows). Many big vacuoles (v) containing coarse material occupy the cytoplasm of the follicle cell. X8550. Fl .G. 6. Electron micrograph showing vacuoles both in the cytoplasm of the oocyte and between the oocyte (0) and the follicle cells w. The vacuoles have a bilaminar wall and appear empty. X14,250.
SANTELLA,
MONROY,
AND
RosATI
addition,
the fibrillar organization of the VC, which is clearly visible between the microvilli, is less distinct where the VC is attached to the microvilli, possibly because of the denser packaging of the fibrils and of their more irregular orientation. At several points this material is detached from the surface of the microvilli, although connection is retained by thin threads. Follicle cells processes crossing the JC and making contact with the oocyte surface-such as those described in Patiria miniata (Schroeder et al., 19’79) have never been observed in Astropecten. In the full-grown oocytes we have observed occasional local thickenings (m-30-40 nm) between the microvilli (Fig. 12). These may correspond to the postjunctional components described in another starfish, Pisaster ochraceus, and interpreted as the 1-methyladenine receptor sites (Schroeder, 1981). DISCUSSION
The observations described in this paper suggest that in the starfish, A. aurantiacus, the oocyte is the primary-and possibly the only-source of material from which the oocyte envelopes, the VC and the JC, differentiate. However, a small contribution from the follicle cells, cannot entirely be ruled out, as the enormous secretory activity of the oocyte would obscure a comparatively minute contribution of the follicle cells. Also, an extremely rapid secretory cycle by the follicle cells, though unlikely, would go undetected in the static electron microscopic pictures. Indeed, a limitation of this work is that it is based exclusively on ultrastructural evidence, as we have not yet succeeded in satisfactorily labeling the ovary of Astropecten with radioactive precursors. With this limitation in mind our observations lend themselves to some considerations. The origin of the VC and of the JC from the oocyte suggests that the two envelopes should be considered a functional unit; in fact., they can be viewed as the system involved in the processes of species-specific sperm recognition and of triggering the acrosome reaction. Starfishes and holothurians are the only echinoderms in which there is unequivocal evidence that the acrosome reaction is triggered upon contact of the spermatozoon with the JC (see Monroy and Rosati (1983), for a discussion). Hoshi and colleagues (Uno and Hoshi, 1978; Ikedai and Hoshi, 1981a,b) have isolated and purified a glycoprotein component from the JC of Asterias amurensis, which induces the acrosome reaction. It is likely that in the asteroids the JC is responsible not, only for
Egg Envelopes in StarJish
477
triggering the acrosome reaction but also for the speciesspecific sperm recognition (see Metz, 1945). This raises the question of the role of the VC in these animals. In tunicates and mammals the sperm receptors are cornponents of the VC, which is also the site of the acrosome reaction; and this is likely to be the case also in the sea urchin, though the question is still controversial (see Monroy and Rosati, 1982,1983). We suggest that in the asteroids, the VC contains receptors which interact with complementary groups of the acrosomal process or with a “bindin-like” protein (for a review on bindin see Vacquier, 1983). However, as in the asteroids and holothurians the acrosome reaction occurs far from the oocyte surface, one should assume that the acrosomal process spans the whole thickness of the JC ensheathed with bindin. This would be interesting to investigate. We further surmise that the molecules involved in the speciesspecific recognition between gametes and in the triggering of the acrosome reaction were originally present throughout the oocyte envelopes and that they gradually localized in the VC in the course of evolution (see also Monroy and Rosati, 1983). Our observations throw no light on the question as to whether the VC and the JC originate from a common precursor material or whether they are the products of two independent secretory processes. The histochemical tests show that the material which is secreted first is PAS-positive. In a later stage, i.e., in the larger oocytes, the material between the microvilli acquires a more compact organization with the fibrils running parallel to the oocyte surface: this layer is the VC primordium. At this point the histochemical tests we have used show a clear-cut. distinction between this VC primordium and the outer material: the VC has lost its PAS positivity and has become strongly FBP-positive while the outer layer, which can be indicated as the JC, is still PASpositive but fails to bind the FBP. Two interpretations of these observations are possible. (1) The material that is secreted first gives rise to the JC, the precursor of the VC being the product of a later secretory wave. (2) Both the VC and the JC originate from a common precursor material secreted by the oocyte and are differentially acted upon by the follicle cells and/or the oocyte. This would imply that the difference between the two envelopes is essentially in their supramolecular organization. Finally, our observations show that in the full grown oocytes the connections between the VC and the microvilli are very intimate. Whether or not they are com-
FIG. 7. An oocyte in a more advanced stage of growth. Two large interstices between the oocyte (0) and the follicle cells (fc) containing remnants of the walls of the extruded vacuoles. Part of a follicle cell still remains apposed to the oocyte. X8550. FIG. 8. Many interstices between the oocyte (0) and thin stretched follicle cells (fc). Irregular microvilli (small arrows) emerged from the plasma membrane of the oocyte. ow, ovarian wall. X17,100.
FI G. 9. A large interstice between the oocyte (0) and the follicle cells. Fluffy material begins to be deposited between the micro\ rilli. Nun lerous vesicles containing fibrillar material are present in the cortical region of the oocyte (arrows). X5320. A :G. 10. An oocyte stained with Thy&y’s method. Silver grains are present in the plasma membrane of the oocyte (0) and in the fl’UffY erial around the microvilli. A vacuole containing fluffy PAS-positive material close to the oocyte plasma membrane (*). Silver grains are also present in the cortical granules and in other cytoplasmic components. X25,320. FI :G. 11. An oocyte in a more advanced stage of growth. The space between the oocyte and the very thin follicle cells (fc) is filled with the while a more compact fibrillar material can be seen between the microvilli. X15,675. jelly ’ coat (jc) material 478
FIG. the fit FIG. to the FIG. no gri
12. Higher magnification of an oocyte (0) at the same stage of growth. The jelly coat (jc) appears as a fine fibrillar meshwork while nillar material between the microvilli appears more compact and begins to acquire a more orderly oriented organization. X27,790. 13. A final stage of the differentiation of the egg envelopes: treatment with FBP followed by fucosyl-ferritin. The labeling is limited vitelline coat (vc); no labeling of the jelly coat (jc). o, oocyte. X62,700. 14. An oocyte at the same stage. The Thy&y’s reaction is positive in the plasma membrane; the jelly coat (jc) is weakly positive and rins are present in the vitelline coat (vs). o, oocyte. X44,415. 479
480
DEVELOPMENTAL BIOLOGY VOLUME99,1983
FIG. 15. In a full grown oocyte (0) the fibrillar organization of the vitelline coat (vc) is distinctly visible; the fibrils are oriented tangentially to the oocyte surface and are attached to the plasma membrane of the microvilli. In the jelly coat (jc) the fibrillar meshwork is much more loose and the flbrils are irregularly oriented. X82,800.
parable to the “vitelline posts” described in the sea urchin egg (Kidd, 1978; Chandler and Heuser, 1980) cannot be decided on the basis of the present observations. Work is in progress to study the changes that occur in these connections in the course of oocyte maturation and as a result of fertilization. We thank Professor F. Kellenberg, Basel, for suggesting the lowtemperature dehydration procedure. We also thank Dr. B. Dale for his critical reading of the paper and G. Princivalli for typing. F.R. is on leave from the University of Siena. REFERENCES BLEIL, J. D., and WASSARMAN, P. M. (1980). Synthesis of zona pellucida proteins by denuded and follicle-enclosed mouse oocytes during culture in vitro. Proc Natl Ad Sti USA 77,1029-1033. CHANDLER, D. E., and HEUSER, J. (1980). The vitelline layer of the sea urchin egg and its modification during fertilization. J. CeU BioL 84, 618-632.
COLWIN, L. H., and COLWIN, A. L. (1956). The acrosome filament and sperm entry in Thyme tiara (Holothuria) and As-. BioL Bull 110.243-257. COTELLI, F., ANDRONICO, F., DE SANTIS, R., MONROY, A., and ROSATI, F. (1981). Differentiation of the vitelline coat in the ascidian Ciona intestinalis: An ultrastructural study. Roux’s Arch Lkv. BioL 190, 252-258. GLABE, C. G., and VACQUIER, V. D. (1978). Egg surface glycoprotein receptor for sea urchin bindin. Proc. NatL Acad Sti USA 75, 881885. GLABE, C. G., and LENNARZ, W. J. (1981). Isolation and partial characterization of a high molecular weight glyconjugate derived from the egg surface, which is implicated in sperm-egg adhesion. .I. .%pram.& S-u& C& Biochem 15,387-394. GLABE, C. G., GRABEL, L. B., VACQUIER, V. D., and ROSEN, S. D. (1982). Carbohydrate specificity of sea urchin sperm bindin: A cell surface lectin mediating sperm-egg adhesion. J. Cell BioL 94, 123-128. GREVE, J. M., SAHAMANN, G. S., ROLLER, R. J., and WASSARMAN, P. M. (1982). Biosynthesis of the major zona pellucida glycoprotcin secreted by oocytes during mammalian oogenesis. Cell 31,749-‘759. IKADAI, H., and HOSHI, M. (1981a). Biochemical studies on the acrosome
SANTELLA,
MONROY,
AND
ROSATI
reaction of the starfish, Aster& amurensis. 1. Factors participating in the acrosome reaction. Dev. Growth D@w. 23,73-80. IKADAI, H., and HOSHI, M. (1981b). Biochemical studies on the acrosome reaction of the starfish Aster& amurensti. 2. Purification and characterization of acrosome reacting-inducing substance. Dev. Grcxdh D$er. 23, 81-88. KIDD, P. (1978). The jelly and vitelline coats of the sea urchin egg: New ultrastructural features. J. Ultrastruct. Res. 64, 204-215. METZ, C. B. (1945). The agglutination of starfish sperm by fertilizing. Biol BulL 89, 84-94. MONROY, A., and ROSATI, F. (1982). On the molecular mechanisms of sperm-egg interaction. Cell Di&r. 11.299-301. MONROY, A., and ROSATI, F. (1983). A comparative analysis of spermegg interaction. Gamete Rex 7, 85-102. ROSATI, F., COTELLI, F., DE SANTIS, R., MONROY, A., and PINTO, M. R. (1982). Synthesis of fucosyl-containing glycoproteins of the vitelline coat in oocytes of Ciona intestinalis (Ascidia). Proc. NatL Acd Sci. USA 79, 1098-1911. ROSSIGNOL, D. P., ROSCHELLE, A. J., and LENNARZ, W. J. (1981). Spermegg binding: Identification of a species-specific sperm receptor from
Egg
Envelopes
in
481
Star$sh
eggs of Strongylocentrotus purpwatus. J. SupramoL Struct. Cell Biochem. 15, 347-358. SCHROEDER, T. E. ( 1981). Microfilament mediated surface change in starfish oocytes in response to 1-methyladenine: Implications for identifying the pathway and receptor sites for maturation-inducing hormones. J. Cell BioL 90, 362-371. SCHROEDER P. C., LARSEN, J. H. and WALDO, A. E. (1979). Oocytefollicle cell relationships in a starfish. Cell Tissue Res. 203.249-256. THICRY, J. P. (1967). Mise en 6vidence des polysaccharides sur coupes fines en microscopic ilectronique. J. Microsc. 6, 987-1018. TILNEY, L. G., HATANO, S., ISHIKAWA, H., and MOOSEKER, M. S. (1973). The polymerization of actin: Its role in the generation of the acrosomal process in certain echinoderm sperm. J. Cell BioL 59,109126. UNO, Y., and HOSHI, M. (1978). Separation of the sperm and the acrosome reaction-inducing substance in egg jelly Science 200, 58-59. VACQUIER, cellulose
V. D. (1983). chromatography.
Purification Anal
of sea urchin bindin B&hem. 129,497-501.
agglutinin of starfish. by DEAE