Junctional Complexes in the Preimplantation Rabbit Embryo RICHARD A. HASTINGS I1 AND ALLEN C. ENDERS Anatomy Department, Washington University Medical School, St. Louis, Missouri 631 10
ABSTRACT
The morphology and development of junctional complexes between blastomeres of the preimplantation rabbit embryo were investigated using several approaches. Electron microscopic examination of embryos stained en bloc with uranyl acetate, and the study of junction permeability using horseradish peroxidase and lanthanum nitrate provided information on structure, intermembrane spacing and permeability of the junctional complexes. In addition, the freeze fracture technique was used with day 5 and day 6 blastocysts, since the large size of these embryos facilitated use of this method. These experiments showed that although rudimentary junctions were present between blastomeres of the early cleavage stages, effective tight junctions were not present until the blastocyst stage. Electron microscopic examination of thin sections revealed apical foci of membrane approximation or “fusion” between trophoblast cells by day 4. Freeze fracturing revealed a lattice of interconnecting ridges (on the A face) and grooves (on the B face) in the apical region between trophoblast cells of the day 5 blastocyst. This lattice formed a continuous band along the apical margin of each cell, and therefore constituted a zonula occludens. The zonula occludens of the day 5 blastocyst averaged 2-3 ridges per lattice, while day 6 blastocysts had lattices that averaged 5-6 ridges. Also seen in the freeze fracture replicas from the day 5 and day 6 blastocysts were local accumulations of intramembranous particles on the A face. These particles were often observed in aggregates similar to those of previously described gap junctions. It could not be determined whether these small regions of particles were true gap junctions or a possible primitive form of gap junction because the complementary pitted surfaces ( B face pits) were not demonstrated.
In preimplantation development in the rabbit, the period from day 3 to day 4 is a time when there are many morphogenetic and biochemical changes. The rate of protein synthesis increases 10 fold per cell (Manes and Daniel, ’69 ) , and the ingestion of exogenous protein by the blastomeres is accelerating (Hastings and Enders, ’74). Also during this period morphogenetic movements and fluid accumulation lead to the formation of the blastocyst cavity. Blastocyst cavity formation is generally considered to occur on the 4th day (Lutwak-Mann, ’71) ; blastocyst diameter can increase from as small as 1.3 mm on day 5 to as large as 4.8 mm on day 6 (Adams, Hay and Lutwak-Mann, ’61). Abundant mitotic activity as well as increasing fluid accumulation accompany the 50-100 fold increase in the size of the ANAT. REC., 181: 17-34.
blastocyst (Hafez, ’71). The extensive expansion of the blastocyst is dependent upon mechanisms for the acquisition and retention of fluid. It has been demonstrated by Daniel (’63) that collapsed rabbit blastocysts will reexpand in vitro and also that blastocyst fragments will form exoblastocysts in vitro. The possible role of pinocytosis in the expansion of the rabbit blastocyst has not been explored, but evidence that a solute pumping system is present in the rabbit blastocyst was provided by Cross and Brinster (’70). Regardless of the mechanism of acquisition, fluid retention within the blastocyst cavity against an Received Jan. 14, ‘74. Accepted July 15, ’74. 1 Supported by grant HD 09462 from the National Institute of Child Health and Human Development. We would especially like to thank Dr. Karl Pfenninger for extensive assistance in freeze-cleave techniques with the blastocysts.
17
18
RICHARD A. HASTINGS I1 AND ALLEN C. ENDERS
osmotic or hydrostatic pressure gradient ranging from two-cell embryos to blastomust depend in part on the relative per- cysts were prepared for electron microsmeability of the trophoblast cells per se copy. Fixation was initiated at room temand the nature of the junctional complexes perature with either a 3% solution of glubetween these cells, The more complete the taraldehyde in 0.1 M cacodylate buffer, pH seal, the greater the relative efficiency of 7.3, or a modification of the Karnovsky the mechanism of accumulation and reten- fixative using a mixture of 2.5% glutaraltion of fluid within the cavity. Conversely, dehyde and 2% formaldehyde in 0.1 M the less complete the tight junctions, the cacodylate buffer for 2-3 hours. Following greater the amount of energy the blasto- an overnight wash in 0.1 M cacodylate cyst would have to expend to achieve and buffer, fixation was continued in a 2% maintain the same expansion. solution of collidine-buffered osmium teThe importance of junctional complexes troxide for two hours at 4°C. The tissues was recognized in earlier studies using were then either rapidly dehydrated in a transmission electron microscopy (Schlafke series of graded alcohols or were treated and Enders, '67). More recently Calarco with a 0.5% solution of uranyl acetate for and Epstein ('73) observed external ridges 90 minutes in the dark at 4"C, following at the cell junctions of mouse blastocysts which the tissues were washed in pH 5.2 using scanning electron microscopy. How- Maleate buffer and dehydrated. Subseever, neither freeze fracture nor lanthanum quently, all tissues were placed in propyltracer studies, which have been so useful ene oxide, then embedded in Durcupan in the characterization of gap and tight (ACM Fluka). The sections (silver to gold) junctions (Goodenough and Revel, '70), were obtained using a Porter-Blum Sorvall have been applied to the early stages of MT-2 microtome, and staining was with mammalian development. either alkaline lead alone or with a satuThis study uses the rabbit blastocyst as rated solution of aqueous uranyl acetate well as selected earlier preimplantation followed by the lead. The thin-sections embryos to characterize the junctional were examined with either a RCA EMU3-G complexes of the trophoblast, the first com- or a Philips 300 electron microscope. Secplete epithelial layer formed in the mam- tions from preimplantation embryos from malian embryo. 21 animals were examined for junctional complex structures. MATERIALS AND METHODS
Horseradish peroxidase Materials Horseradish peroxidase was used as a Female New Zealand white rabbits were marker of extracellular space, since its mated with fertile male rabbits. The age reaction product is electron dense. This of the selected preimplantation embryos tracer is a protein with a molecular weight was designated by the time elapsed from of approximately 40,000 (Keilin and coitus. To obtain the embryos, the preg- Hartree, '51). Peroxidase was used at a nant rabbits were anesthetized with ether, concentration of 10 mg/ml in 0.9% NaC1, and the uteri and oviducts removed and and the selected embryos were exposed to flushed with 0.9% NaCl. The embryos obthis tracer in vitro at 37°C. The exposure tained were then used for the experiments periods varied from 5 to 60 minutes. Foldescribed in the subsequent sections. The lowing initial fixation demonstration of reagents employed were : horseradish perthe peroxidase was accomplished by the oxidase (Type 11, Sigma Chemical Co., St. method of Graham and Karnovsky ('66) Louis, Mo.), lanthanum nitrate (Fisher including controls lacking either the H202 Scientific Co., Pittsburgh, Pa.), and 3,3' di- or the 3,3' diaminobenzidine. Preimplantaaminobenzidine (Sigma Chemical CO., St. tion embryos that had not been exposed to Louis, Mo.). tracer were also incubated in complete diaminobenzidine medium to exclude the Methods possibility of endogenous peroxidase activPreparation for electron microscopy ity. Fixation of peroxidase-treated emwas identical to that described above Embryos at preimplantation periods hrvos .~ . ,
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES
for thin section preparation. Sections of preimplantation embryos varying in age from two cell to late blastocyst from 15 animals were examined following peroxidase exposure.
Lanthanum Embryos selected for these experiments ranged in developmental age from day 3 to day 6 post-coitus (P.c.). Embryos were exposed to lanthanum nitrate ( LaN03),pH 7.6, by addition of an aliquot of the lanthanum solution (Revel and Karnovsky, '67) to both the aldehyde and osmium so that the final concentration of the lanthanum was 1 % . This exposure was carried out in two different ways. Either the embryos were exposed to the lanthanum in watchglasses or the lanthanum was injected into the cavity of the blastocysts. The latter method was employed only on days 5 and 6 using micropipettes (five blastocysts from three animals) while the external exposure in the watchglasses was used (eight embryos from five animals) on embryos from day 3 to day 6. Subsequent processing of the tissue was identical to that described for the thin section preparation.
19
onto the fractured surfaces, and the residual tissue was removed by a 24 hour wash in dilute clorox. The cleaned replicas were picked up on formvar coated grids and viewed in a RCA EMU3-G or a Philips 300 electron microscope. In the micrographs the shadows are toward the top of the print. Sixteen blastocysts from five animals on day 5 and 21 blastocysts from seven animals on day 6 were used to obtain the junctional complexes studied. Data on the number of strands or ridges comprising a tight junction was obtained in preference to information on the diameter of a particular ridge or strand due to the variations that can result in size measurements depending on amount of shadowing and the angle of casting. The number of strands were counted along the apical regions between two trophoblast cells and the average number calculated. This was done for both day 5 and day 6 blastocysts following the guidelines of Claude and Goodenough ('73). RESULTS
Junctional regions of trophoblast cells of the blastocyst
Characterization o f the Freeze fracture zonula occludens Day 5 and 6 p.c. rabbit blastocysts were The junctional complexes between troobtained by flushing the uterus and were phoblast cells of rabbit blastocysts obtained immediately placed in a solution of 2% from the uterus on days 4, 5 and 6 p.c. formaldehyde and 2.5% glutaraldehyde in were studied. The spherical blastocyst has 0.1 M cacodylate buffer for 20 minutes. a single, cellular layer of trophoblast comThe extracellular coats were then dissected posed of thin, polygonal cells. These cells away from the trophoblast using care not have an elevated central region which conto damage the integrity of the spherical tains the nucleus, and they taper to become blastocyst. The blastocysts were then thinner at their circumference. Due to the washed in 0.1 M cacodylate buffer for one polygonal shape, each trophoblast cell is in hour before being immersed in 20% glyc- contact with 5 or 6 other cells (fig. 1 ) . erol in 0.1 M cacodylate buffer for 40 min- There is much variation in the shape of utes just prior to freezing. Blastocysts were the adjoining edges of adjacent trophothen mounted into teflon coated micro- blast cells. Often stepwise overlapping chambers designed for the fracturing of flanges and flat overlapping edges that very small pieces of tissue, and the blasto- adjoin at acute angles to the vertical apicalcysts were either cut or collapsed to form basal axis are observed (figs. 2-4). Occaa flat mount during the loading. The tissue sionally, the contact regions resemble the was rapidly frozen in liquid nitrogen- configuration noted by Farquhar and Pacooled Freon 22 (Chlorodifluoromethane) lade ('63) for normal adult epithelia, in and fracturing was accomplished on a that the contact regions are oriented vertiBalzers apparatus at a stage temperature cally in an apical-basal plane and extend of - 120°C. Replicas were formed by the over long areas in a very regular manner. evaporation of platinum and carbon coats The junctional complex forms at the most
20
RICHARD A. HASTINGS I1 A N D ALLEN C. ENDERS
apical contact region between trophoblast cells. Within these junctional complexes the most apical components are areas where the two unit membranes of adjacent trophoblast cells converge until they approximate one another excluding the extracellular space (fig. 5). Even though the tissues were treated en bloc with uranyl acetate, the individual leaflets of the “unit membrane” structure could not be discerned with sufficient clarity to demonstrate “fusion” of the outer leaflets per se. These apparently fused areas, or tight junctions, consist of one or more punctate regions within individual junctional complexes using transmission electron microscopy. Although the presence of such regions in all micrographs of apical junctional complexes suggests that these tight junctions are in the form of a continuous band, the poor resolution of the membrane leaflets and the difficulty of obtaining more than a few thin sections of the same fused area precludes positive determination of continuity. When lanthanum is placed in the external medium, it penetrates the apical intercellular space between trophoblast cells until it is abruptly stopped by the region of apical membrane fusion (fig. 6). Where the lanthanum appears in adequate concentration, the pointed or V-shaped tip of the advancing front of the tracer demonstrates the convergence of the unit membranes to the point of fusion. Lanthanum was not observed basal to the areas of fusion, which implies that these tight junctions are sufficiently continuous along the apical region between trophoblast cells to exclude this tracer under the conditions employed. The study of freeze fracture replicas of day 5 and day 6 blastocysts revealed additional information on the nature of these apical tight junctions. Along the apical region of trophoblast cells (both day 5 and 6 ) a lattice of anastomosing ridges or strands was demonstrated on the A or inner membrane face, while a series of connecting grooves was found in the B or outer membrane face (figs. 7, 8). This lattice of ridges and grooves was found to be continuous along the apical region between trophoblast cells but was usually just a few ridges deep. The average number of
strands per lattice increased from the day 5 blastocysts, which averaged 2-3 ridges, to the day 6 blastocysts, which averaged 5-6 ridges per lattice. The areas immediately adjacent to regions where three cells adjoin were not included in the averages as these regions generally exhibit more extensive lattices (Claude and Goodenough, ’73).
Characterization of the gap junction Transmission electron microscopy revealed occasional areas between trophoblast cells where the unit membranes were not fused but were separated by an extracellular space of only 20-40 A (fig. 9). These regions were encountered only occasionally and were located basal to the tight junctions. As mentioned previously lanthanum was halted by the apical tight junctions. Consequently this tracer was injected into the blastocyst cavity of day 5 and 6 blastocysts which allowed it to penetrate the spaces between cells from the basal side. Where the section is perpendicular to the unit membranes, the lanthanum appears as a dark material within the extracellular space. The periodic spacing of this dark material gives a beaded appearance to this region (fig. 9). Obliquely sectioned regions of these lanthanum infiltrated junctions have a roughly polygonal substructure. Freeze fracture replicas from day 5 and 6 blastocysts revealed local aggregates of particles that were occasionally located among the ridges of the tight junctions but more commonly were found basal to the lattice. These aggregates (A face) were usually composed of relatively few particles; however, several more extensive polygonal aggregates were observed (fig. 10). Some of the larger polygonal aggregates are similar in structure to previously described gap junctions in the chick embryo epiblast (Revel et al., ’73) and early adult mammalian tissue (Goodenough and Revel, ’70). However, the aggregates observed here occupied only small areas, and the complementary pitted surfaces were not demonstrated. Macula adhaerens Desmosomes were a commonly observed component of the apical junctional com-
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES
plex, but in addition they could be found at different levels along the adjoining edges of the trophoblast cells. Desmosomes showed a wide variation in structure. The desmosomes located proximal to the apical junctional complex were usually extensive and displayed the morphological features of mature desmosomes. These desmosomes had a concentration of amorphous dense material located in the cytoplasm adjacent to the inner leaflet of the unit membrane and associated fibrillar material projecting into the adjacent cytoplasm (fig. 11). In addition, some of these desmosomes had dense material within the intercellular space that was generally 220-250 A wide. Even the larger desmosomes were punctate (macula adhaerens). Other desmosomes observed basally were variable in structure. Many lacked the fibrillar material, and had only a thin accumulation of amorphous material. Desmosomes of this type were commonly found adjacent to expanded extracellular spaces (extracellular “lakes”).
Morphology of junctional complexes during preblastocyst development Two-cell to morula Extensive regions of association between the blastomeres of the two cell embryo were observed; however this association did not include distinct junctional complexes. (See Longo and Anderson, ’69 for a description of the formation of the twocell rabbit embryo). Microvilli projected into the intercellular space and were often so numerous that the region between the blastomeres was filled with microvilli, allowing many areas of close membrane association. In addition, non-microvillous regions were found where the membranes of the two blastomeres were also closely apposed. The first specialized structures between blastomeres were observed in the four-cell embryo. These structures consisted of regions of parallel opposing membranes, separated by an intercellular space of less than 100 A, with localized variable accumulations of amorphous material adjacent to the inner aspect of each membrane. These punctate structures (fig. 12) were termed rudimentary desmosomes as a convenience for description and discus-
21
sion, but their role in subsequent junction formation has not yet been established. Peroxidase penetrated between cells without apparent restriction in all the cleavage stages examined (fig. 14).
Morula to blastocyst The morula is composed of a solid sphere of blastomeres, lacking the central cavity of the blastocyst. This solid, spherical arrangement of the blastomeres creates two populations of cells with regard to their position within the embryo. The outer group of blastomeres comprising the external layer of the sphere is unique in that a portion of each cell surface is exposed to the exterior. The central group of cells is composed of blastomeres that do not directly border the exterior. In older morulae (day 3 P.c.) the two groups showed distinct differences in the nature of the junctions between blastomeres. The blastomeres of the central group were generally separated by a wide extracellular space, and junctions were infrequent. The most commonly observed evidence of junction formation in this group of cells was the rudimentary desmosome. These punctate structures, as in earlier embryos, lacked fibrillar material and had only moderate amounts of amorphous, dense material subjacent to the inner leaflet. Junctional complexes between blastomeres of the outer group was found, with few exceptions, at the apical region. These junctions were characterized by local accumulations of amorphous material in a thin band along the inner leaflet of the unit membrane (fig. 13). They differed from the rudimentary desmosomes in their apical location, and in the extensive nature of the amorphous material. Focal points where the width of the intercellular space between the adjoining blastomeres was reduced to less than the thickness of either apposed cell membrane were occasionally seen in these complexes. The intercellular spaces between the blastomeres of the early morulae (day 2-2.5 P.c.) were small and were usually angular in shape. At all these early developmental periods prior to the blastocyst the tracer, horseradish peroxidase, could penetrate the intercellular spaces between blastomeres. Peroxidase could pass be-
22
RICHARD A. HASTINGS I1 AND ALLEN C. ENDERS
tween blastomeres of the morula (day 2-2.5 P.c.) but was restricted from intercellular passage by trophoblast cells of the day 3 blastocyst. With increasing development, the size of the intercellular spaces increased, and they lost their angular shape, becoming oval or elliptical in outline. The earliest embryos obtained as blastocysts were recovered from the oviduct 3 days 12 hours p.c. These embryos had a single, well formed cavity and a distinct embryonic cell mass beneath the trophoblast. The central cavity of these very early blastocysts was quite small, and the apical junctional complexes were identical to those described for the late morula ( 3 days 5 hours P.c.). The earliest blastocysts examined with the lanthanum method ( 3 days 12 hours P.c.) consistently had lanthanum deposits for nearly the entire length of the area of reduced intercellular space at the apical junctional complex. Blastocysts obtained 3 days 22 hours p.c. consistently exhibited the focal points of approximation of cell membrane or “fusion .” DISCUSSION
The increase in the size of the intercellular spaces between the early and late morula (2-3 days P.c.) and the change in shape of these spaces from irregular and angular to oval or elliptical are the early physical evidence of fluid retention. Since fluid retention implies restricted access to the exterior, Schlafke and Enders (’67) suggested that apical junctional complex formation is a necessary prerequisite to establishment of the cavity of the blastocyst. However, not until the zonula occludens portion of the junctional complex is relatively complete would the restriction of flow be expected to be significant. The progressive nature of apical junction complex formation is not always understood. Hesseldahl (’71 ) interpreted the earlier statements on junctional complexes to mean that whenever apical junctions could be seen in a given section, i t was presumptive evidence of the blastocyst stage. The presence of a single cavity is by definition the criterion of blastocyst formation. The smallest blastocysts obtained in the present study were removed from the oviduct 3 days 12 hours p.c. Even in these early
blastocysts, horseradish peroxidase did not penetrate the intercellular spaces, although focal points of “fusion” of adjacent cell membranes were rare in sections of the junctional complexes. However, by day 4 blastocysts commonly exhibited “fusion” within the apical junctions. The exclusion of lanthanum indicates that these junctions were zonular rather than punctate structures. Unfortunately due to the small size of the younger stages, and the consequent difficulty in getting a relatively flat appropriately oriented preparation following removal of extracellular coats, the earliest embryos that could be readily freeze fractured were day 5 blastocysts. Replicas obtained from the fracturing of day 5 blastocysts revealed the lattice of ridges and grooves extending continuously along the apical region between cells. This observation was particularly useful in demonstrating the zonular nature of the tight junctions (zonulae occludentes) present by day 5 of development. The blastocysts showed an increase in the average number of ridges per lattice from 2-3 on day 5 to 5-6 on day 6. This increase indicates that the development of the apical junctional complexes is progressive. Claude and Goodenough (’73) have recently examined a variety of zonulae occludentes of tight and leaky epithelia using freeze fracture, and have classified these junctions with respect to transepithelial permeability. They stated that the number of ridges per lattice relates more closely to the restriction of transepithelial permeability than do other factors, including junctional depth. Applying their criteria, the greater average number of ridges in the day 6 blastocyst would indicate a decreased intercellular transepithelial permeability compared to the day 5 blastocyst. Since there is an increase in cell numbers and consequently in potentially leaky cell junctions, this reduced permeability would be especially important in the retention of fluid necessary for the continued expansion. Boving (’56) has stated that the transport of the rabbit blastocysts along the uterus and the unique equidistant spacing of these blastocysts depends on having expanded, spherical blastocysts. In addition,
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES
in species whose blastocysts undergo a
large expansion, the increase in size helps bring the trophoblast into apposition with the uterine epithelium (Enders, '71). The mechanism of fluid movement into the cavity appears to involve an active transport system (Cross and Brinster, '70). Cross ('73) has calculated that the net ionic movement of solutes into the cavity of the day 6 blastocyst could be responsible for the influx of 60 ,L/day of fluid into this cavity, if an isosmotic solution were absorbed. He argues that this volume of fluid movement would be sufficient to account for all the fluid accumulation in the cavity prior to implantation, and that therefore an active water transport system as postulated by Tuft and Boving ('70) would be unlikely. Although the calculations of Cross do not include some of the variations in in situ conditions (hydrostatic pressure from uterine contraction, pulsation of the blastocysts), they establish the major role that solute transport can play in blastocyst expansion if this structure is considered as a continuous semipermeable membrane. The zonulae occludentes not only restrict the escape of fluid from the cavity of the blastocyst by an intercellular pathway, but also limit access to the blastocyst cavity. Large molecules must gain access to the cavity of the blastocyst and cells of the embryonic cell mass through the trophoblast cells. Evidence for discriminatory protein ingestion and transport system in the rabbit blastocyst has come from an in vitro study of pinocytosis using protein tracers (Hastings and Enders, '73). Numerous physiological studies have been reported demonstrating electrotonic coupling between cells of adult tissues (Hagiwara and Morita, '62; Kaneko, '71; Weidmann, '66) as well as embryonic tissues (Bennett and Trinkaus, '70; Sheridan, '71; Ito and Loewenstein, '69). Passage of tracers, such as fluorescein, procion yellow, and microperoxidase, between electrotonically coupled adult cells has been used as evidence that the gap junction is the specialized pathway mediating this coupling (Bennett and Dunham, '70; Payton et al., '69; Reese et al., '71). However, fluorescein did not pass between reag-
23
gregated and coupled cells of the teleost embryo (Bennett et al., '72), or between coupled cells of the intact cleavage and blastulae embryos of Xenqus (Slack and Palmer, '69). Although Bennett and Trinkaus ('70) demonstrated electrotonic coupling in Fundulus heteroclitus blastulae, Lentz and Trinkaus ('71) did not find gap junctions until the gastrula stage of this species. The chick (one of the earlier species in which junction formation was studied (Trelstad et al., '67) has recently been shown by freeze fracture to have gap junctions by stage four (Revel et al., '73). Although the small, scattered regions of close membrane association (20-40 A ) described in this study had most of the features of gap junctions, the freeze fracture identification was incomplete since none of the areas of aggregated particles had demonstrable complementary pitted surfaces. Further study will be necessary to determine whether such regions are developmental precursors to more typical gap junctions, are situated such that the complementary face is rarely cleaved, or are even unrelated structures. Since the rabbit blastocyst is large, data concerning electrotonic coupling of the trophoblast cells could be obtained from this species to accompany further freeze fracture information. LITERATURE CITED Adams, C. E., M. F. Hay and C. Lutwak Mann 1961 The action of various agents upon the rabbit embryo. J. embryo. exp. Morph., 9: 468491. Bennett, M. V. L., and P. B. Dunham 1970 Sucrose permeability of junctional membrane at an electrotonic synapse. Biophysical J., 10: 117a. Bennett, M. V. L., M. E. Spira and G. D. Pappas 1972 Properties of electrotonic junctions between embryonic cells of Fundulus. Dev. Biol., 29: 419-435. Bennett, M. V. L., and J. P. Trinkaus 1970 Electrical coupling between embryonic cells by way of extracellular space and specialized junctions. J. Cell Biol., 44: 592-610. Boving, B. G. 1956 Rabbit blastocyst distribution. Am. J. Anat., 98: 403-417. Calarco, P. G., and C. J. Epstein 1973 Cell surface changes during preimplantation development in the mouse. Dev. Biol., 32: 208-213. Claude, P., and D. A. Goodenough 1973 Fracture faces of zonulae occludentes from "tight" and "leaky" epithelia. J. Cell Biol., 58: 3 9 0 4 0 0 . Cross, M. H. 1973 Active sodium and chloride
24
RICHARD A. HASTINGS I1 AND ALLEN C. ENDERS
transport across the rabbit blastocoel wall. Biol. Reprod., 8: 566-575. Cross, M. H., and R. L. Brinster 1970 Influence of ions, inhibitors and anoxia on transtrophoblast potential of rabbit blastocyst. Exp. Cell Res., 62: 303-309. Daniel, J. C. 1963 Some kinetics of blastocyst formation as studied by the process of reconstitution. J. Exp. Zool., 154: 231-238. Enders, A. C. 1971 The fine structure of the blastocyst. In: Biology of the Blastc-yst, R. J. Blandau, ed. Univ. of Chicago Press. pp. 71-94. Farquhar, M. G., and G. E. Palade 1963 Junctional complexes in various epithelia. J. Cell Biol., 17: 375-412. Goodenough, D. A., and J. P. Revel 1970 A fine structural analysis of intercellular junctions in the mouse liver. J. Cell Biol., 45: 273290. Graham, R. C., and M. J. Karnovsky 1966 The early stages of absorption of injected horseradish peroxidase in the proximal tubules of the mouse kidney: Ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem., 14: 291-302. Hafez, E. S . E. 1971 Some maternal factors affecting physicochemical properties of blastocysts. In: Biology of the Blastocyst. R. J. Blandau, ed. Univ. of Chicago Press. pp. 113124. Hagiwara, S . , and H. Morita 1962 Electrotonic transmission between two nerve cells in leech ganglion. J. Neurophysiol., 25: 721-731. Hastings, R. A., and A. C. Enders 1974 Uptake of exogenous protein by the preimplantation rabbit. Anat. Rec., 179: 311-330. Hesseldahl, H . 1971 Ultrastructure of early cleavage stages and preimplantation i n the rabbit. Z . Anat. mEntwick1.-Gesh., 135: 139-155. Ito, S., and W. R. Loewenstein 1969 Ionic communication between eirly embryonic cells. Dev. Biol., 19: 228-243. Kaneko, A. 1971 Electrical connections between horizontal cells i n the dogfish retina J. Physiol. (London), 213: 95-105. Keilin, D., and E. F. Hartree 1951 Purification of horseradish peroxidase and comparison of its properties with those of catalase and methaemoglobin. Biochem. J., 49: 88-104. Lentz, T. C., and J. P. Trinkaus 1971 Differentiation of the junctional complexes of sur-
face cells in the developing Fundulus blastoderm. J. Cell Biol., 48: 455472. Longo, F. J., and E. Anderson 1969 Cytological events leading to the formation of the two cell stage in the rabbit: Association of maternally and paternally derived genomes. J. Ultrastructural Res., 29: 86-118. Lutwak-Mann, C. 1971 The rabbit blastocyst and its environment: Physiological and biochemical aspects. In: Biology of the Blastocyst, R. J. Blandau, ed. Univ. of Chicago Press. pp. 243-260. Manes, C., and J. C. Daniel 1969 Quantitative and qualitative aspects of protein synthesis in the preimplantation rabbit embryo. Exp. Cell Res., 55: 261-268. Payton, B. W., M. V. L. Bennett and G. D. Pappas 1969 Permeability and structure of junctional membranes at a n electrotonic synapse. Science, 166: 1641-1643. Reese, T. S., M. V. L. Bennett and N. Feder 1971 Cell-to-cell movement of peroxidases injected into the septate axon of crayfish. Anat. Rec., 169: 409-420. Revel, J. P., and M. J. Karnovsky 1967 Hexagonal array of subunits in intercellular junctions of the mouse heart and liver. J. Cell Biol., 33: c7.
Revel, J. P., P. Yip and L. L. Chang 1973 Cell junctions i n the early chick embryo - A freeze etch study. Dev. Biol., 35: 302-317. Schlafke, S . , and A. C. Enders 1967 Cytological changes during cleavage and blastocyst formation in the rat. J. Anat., 102: 13-32. Sheridan, J. D. 1971 Dye movement and low resistance junctions between reaggregated embryonic cells. Dev. Biol., 26: 627-632. Slack, C., and J. F. Palmer 1969 Permeability of intercellular junctions in the early embryo of Xenopus Eaevis, studied with a fluorescein tracer. Exp. Cell Res., 55: 416-419. Trelstad, R. L., E. D. Hay and J. P. Revel 1967 Cell contact during early morphogenesis in the chick embryo. Dev. Biol., 16: 78-106. Tuft, P. H., and B. G. Boving 1970 The forces involved i n water uptake by the rabbit blastocyst. J. EXP. ZOO^., 174: 165-172. Weidmann, S. 1966 The diffusion of radiopotassium across the intercalated disks of mammalian cardiac muscle. J. Physiol. (London), 187: 323-341.
PLATES
PLATE 1 EXPLANATION OF FIGURES
1 This micrograph shows cells of an intact day 5 blastocyst viewed with Nomarski phase interference optics. The polygonal outline of the trophoblast cells ( T ) is observed and the area of contact between two cells is shown (arrow). Each trophoblast cell has five to six other cells in contact with it to form this polygonal arrangement. x 1,270.
26
2
The apical region of contact between two day 6 trophoblast cells is shown i n this micrograph. The junctional region consists of parallel membranes bound basally by a desmosome (arrow) that is composed of small, dense plaques sectioned obliquely. x 42,000.
3
The apical junctional complex between two trophoblast cells from a day 5 blastocyst is observed in this micrograph. The cells are adjoined by slightly overlapping processes, with a desmosome (arrow) basal to the apical contact region. x 22,500.
4
This micrograph, of a day 5 blastocyst, shows the small junctional complex (arrow) between the two trophoblast cells. This contact region occurs at the apical end of extensive overlapping flanges from the adjacent cells. A n endoderm cell is observed beneath the trophoblast cells (lower right). x 13,000.
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES Richard A. Hastings I1 and Allen C. Enders
PLATE 1
PLATE 2 EXPLANATION O F FIGURES
28
5
The apical junctional complex between trophoblast cells of a day 4 blastocyst is shown in this micrograph. A thin band of dense material is situated along the cytoplasmic side of each membrane, and one of the points of fusion or loss of the intercellular space is noted (arrow). X 60,500.
6
This micrograph shows the apical junction between two trophoblast cells from a day 5 blastocyst after exposure to lanthanum nitrate for six minutes. The lanthanum which was added to the external medium, coats the cell surface and penetrates between the trophoblast cells until it is stopped by membrane fusion (arrow). x 62,000.
7
This micrograph of a freeze fracture replica from a day 5 blastocyst shows the apical region between trophoblast cells. The lattice of interconnecting ridges (large arrows ) and grooves is continuous along the apical region. This continuous lattice, even though narrow, constitutes a zonula occludens. The small cluster of particles (small arrow) noted within the lattice might represent a small primitive or atypical gap junction. X 54,000.
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES Richard A. Hastings I1 and Allen C. Enders
PLATE 2
PLATE 3 EXPLANATION OF FIGURES
8
This micrograph shows the apical junctional region between two trophoblast cells from a freeze fractured day 6 blastocyst. Ridges (large arrows) and the complementary grooves (small arrow) form the lattice, which is more extensive than the lattice on day 5 . Particles that might have broken from the complementary face are seen lying in the grooves (crossed arrow) in the upper right of the micrograph. X 54,000.
9 The intercellular space in the gap junction in this micrograph has been penetrated by lanthanum which was injected into the blastocyst cavity. The lanthanum demonstrates the polygonal intercellular substructure of the gap junction where the section passes obliquely (large arrow). Note the beaded appearance (small arrow) of the perpendicularly sectioned region of the junction. x 105,000. 10 A local aggregation of particles (arrow) from a freeze fracture replica of a day 5 blastocyst is shown in this micrograph. The A face morphology is similar to that of previously described gap junctions. However the complementary pitted surface (B-face) was not demonstrated. x 72,500.
30
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES Richard A. Hastings I1 and Allen C. Enders
PLATE 3
PLATE 4 EXPLANATION O F FIGURES
11 This micrograph shows a region basal to the apical junctional complex between two day 6 trophoblast cells. A small intercellular space ("lake") is delimited by two desmosomes (arrows), which have filamentous components associated with their dense plaques. Crystalloids ( C ) and mitochondria ( M ) are present in the adjacent cytoplasm. x 52,000.
12 This micrograph of two cells from a four-cell embryo shows a rudimentary desmosome (arrow). The closely apposed dense plaques of these structures are usually without the associated filamentous material seen in the desmosomes of the blastocyst stage. x 40,000. 13 This micrograph shows a typical junctional complex between two blastomeres of a rabbit morula (3 days 5 hours). The membranes are roughly parallel and a slight accumulation of dense material is noted i n the cytoplasm adjacent to the inner leaflet of each membrane. x 145,000.
14
32
This micrograph shows a n area of apposition between two cells of an eight-cell embryo following exposure to horseradish peroxidase for 15 minutes. In this stage and all other cleavage stages prior to the blastocyst, the tracer can penetrate the intercellular space between the blastomeres ( B ) . The electron dense reaction product coats the microvilli i n the intercellular space. X 38,000.
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES Richard A. Hastings I1 and Allen C. Enders
PLATE 4
ABSTRACT
The morphology and development of junctional complexes between blastomeres of the preimplantation rabbit embryo were investigated using several approaches. Electron microscopic examination of embryos stained en bloc with uranyl acetate, and the study of junction permeability using horseradish peroxidase and lanthanum nitrate provided information on structure, intermembrane spacing and permeability of the junctional complexes. In addition, the freeze fracture technique was used with day 5 and day 6 blastocysts, since the large size of these embryos facilitated use of this method. These experiments showed that although rudimentary junctions were present between blastomeres of the early cleavage stages, effective tight junctions were not present until the blastocyst stage. Electron microscopic examination of thin sections revealed apical foci of membrane approximation or “fusion” between trophoblast cells by day 4. Freeze fracturing revealed a lattice of interconnecting ridges (on the A face) and grooves (on the B face) in the apical region between trophoblast cells of the day 5 blastocyst. This lattice formed a continuous band along the apical margin of each cell, and therefore constituted a zonula occludens. The zonula occludens of the day 5 blastocyst averaged 2-3 ridges per lattice, while day 6 blastocysts had lattices that averaged 5-6 ridges. Also seen in the freeze fracture replicas from the day 5 and day 6 blastocysts were local accumulations of intramembranous particles on the A face. These particles were often observed in aggregates similar to those of previously described gap junctions. It could not be determined whether these small regions of particles were true gap junctions or a possible primitive form of gap junction because the complementary pitted surfaces ( B face pits) were not demonstrated.
In preimplantation development in the rabbit, the period from day 3 to day 4 is a time when there are many morphogenetic and biochemical changes. The rate of protein synthesis increases 10 fold per cell (Manes and Daniel, ’69 ) , and the ingestion of exogenous protein by the blastomeres is accelerating (Hastings and Enders, ’74). Also during this period morphogenetic movements and fluid accumulation lead to the formation of the blastocyst cavity. Blastocyst cavity formation is generally considered to occur on the 4th day (Lutwak-Mann, ’71) ; blastocyst diameter can increase from as small as 1.3 mm on day 5 to as large as 4.8 mm on day 6 (Adams, Hay and Lutwak-Mann, ’61). Abundant mitotic activity as well as increasing fluid accumulation accompany the 50-100 fold increase in the size of the ANAT. REC., 181: 17-34.
blastocyst (Hafez, ’71). The extensive expansion of the blastocyst is dependent upon mechanisms for the acquisition and retention of fluid. It has been demonstrated by Daniel (’63) that collapsed rabbit blastocysts will reexpand in vitro and also that blastocyst fragments will form exoblastocysts in vitro. The possible role of pinocytosis in the expansion of the rabbit blastocyst has not been explored, but evidence that a solute pumping system is present in the rabbit blastocyst was provided by Cross and Brinster (’70). Regardless of the mechanism of acquisition, fluid retention within the blastocyst cavity against an Received Jan. 14, ‘74. Accepted July 15, ’74. 1 Supported by grant HD 09462 from the National Institute of Child Health and Human Development. We would especially like to thank Dr. Karl Pfenninger for extensive assistance in freeze-cleave techniques with the blastocysts.
17
18
RICHARD A. HASTINGS I1 AND ALLEN C. ENDERS
osmotic or hydrostatic pressure gradient ranging from two-cell embryos to blastomust depend in part on the relative per- cysts were prepared for electron microsmeability of the trophoblast cells per se copy. Fixation was initiated at room temand the nature of the junctional complexes perature with either a 3% solution of glubetween these cells, The more complete the taraldehyde in 0.1 M cacodylate buffer, pH seal, the greater the relative efficiency of 7.3, or a modification of the Karnovsky the mechanism of accumulation and reten- fixative using a mixture of 2.5% glutaraltion of fluid within the cavity. Conversely, dehyde and 2% formaldehyde in 0.1 M the less complete the tight junctions, the cacodylate buffer for 2-3 hours. Following greater the amount of energy the blasto- an overnight wash in 0.1 M cacodylate cyst would have to expend to achieve and buffer, fixation was continued in a 2% maintain the same expansion. solution of collidine-buffered osmium teThe importance of junctional complexes troxide for two hours at 4°C. The tissues was recognized in earlier studies using were then either rapidly dehydrated in a transmission electron microscopy (Schlafke series of graded alcohols or were treated and Enders, '67). More recently Calarco with a 0.5% solution of uranyl acetate for and Epstein ('73) observed external ridges 90 minutes in the dark at 4"C, following at the cell junctions of mouse blastocysts which the tissues were washed in pH 5.2 using scanning electron microscopy. How- Maleate buffer and dehydrated. Subseever, neither freeze fracture nor lanthanum quently, all tissues were placed in propyltracer studies, which have been so useful ene oxide, then embedded in Durcupan in the characterization of gap and tight (ACM Fluka). The sections (silver to gold) junctions (Goodenough and Revel, '70), were obtained using a Porter-Blum Sorvall have been applied to the early stages of MT-2 microtome, and staining was with mammalian development. either alkaline lead alone or with a satuThis study uses the rabbit blastocyst as rated solution of aqueous uranyl acetate well as selected earlier preimplantation followed by the lead. The thin-sections embryos to characterize the junctional were examined with either a RCA EMU3-G complexes of the trophoblast, the first com- or a Philips 300 electron microscope. Secplete epithelial layer formed in the mam- tions from preimplantation embryos from malian embryo. 21 animals were examined for junctional complex structures. MATERIALS AND METHODS
Horseradish peroxidase Materials Horseradish peroxidase was used as a Female New Zealand white rabbits were marker of extracellular space, since its mated with fertile male rabbits. The age reaction product is electron dense. This of the selected preimplantation embryos tracer is a protein with a molecular weight was designated by the time elapsed from of approximately 40,000 (Keilin and coitus. To obtain the embryos, the preg- Hartree, '51). Peroxidase was used at a nant rabbits were anesthetized with ether, concentration of 10 mg/ml in 0.9% NaC1, and the uteri and oviducts removed and and the selected embryos were exposed to flushed with 0.9% NaCl. The embryos obthis tracer in vitro at 37°C. The exposure tained were then used for the experiments periods varied from 5 to 60 minutes. Foldescribed in the subsequent sections. The lowing initial fixation demonstration of reagents employed were : horseradish perthe peroxidase was accomplished by the oxidase (Type 11, Sigma Chemical Co., St. method of Graham and Karnovsky ('66) Louis, Mo.), lanthanum nitrate (Fisher including controls lacking either the H202 Scientific Co., Pittsburgh, Pa.), and 3,3' di- or the 3,3' diaminobenzidine. Preimplantaaminobenzidine (Sigma Chemical CO., St. tion embryos that had not been exposed to Louis, Mo.). tracer were also incubated in complete diaminobenzidine medium to exclude the Methods possibility of endogenous peroxidase activPreparation for electron microscopy ity. Fixation of peroxidase-treated emwas identical to that described above Embryos at preimplantation periods hrvos .~ . ,
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES
for thin section preparation. Sections of preimplantation embryos varying in age from two cell to late blastocyst from 15 animals were examined following peroxidase exposure.
Lanthanum Embryos selected for these experiments ranged in developmental age from day 3 to day 6 post-coitus (P.c.). Embryos were exposed to lanthanum nitrate ( LaN03),pH 7.6, by addition of an aliquot of the lanthanum solution (Revel and Karnovsky, '67) to both the aldehyde and osmium so that the final concentration of the lanthanum was 1 % . This exposure was carried out in two different ways. Either the embryos were exposed to the lanthanum in watchglasses or the lanthanum was injected into the cavity of the blastocysts. The latter method was employed only on days 5 and 6 using micropipettes (five blastocysts from three animals) while the external exposure in the watchglasses was used (eight embryos from five animals) on embryos from day 3 to day 6. Subsequent processing of the tissue was identical to that described for the thin section preparation.
19
onto the fractured surfaces, and the residual tissue was removed by a 24 hour wash in dilute clorox. The cleaned replicas were picked up on formvar coated grids and viewed in a RCA EMU3-G or a Philips 300 electron microscope. In the micrographs the shadows are toward the top of the print. Sixteen blastocysts from five animals on day 5 and 21 blastocysts from seven animals on day 6 were used to obtain the junctional complexes studied. Data on the number of strands or ridges comprising a tight junction was obtained in preference to information on the diameter of a particular ridge or strand due to the variations that can result in size measurements depending on amount of shadowing and the angle of casting. The number of strands were counted along the apical regions between two trophoblast cells and the average number calculated. This was done for both day 5 and day 6 blastocysts following the guidelines of Claude and Goodenough ('73). RESULTS
Junctional regions of trophoblast cells of the blastocyst
Characterization o f the Freeze fracture zonula occludens Day 5 and 6 p.c. rabbit blastocysts were The junctional complexes between troobtained by flushing the uterus and were phoblast cells of rabbit blastocysts obtained immediately placed in a solution of 2% from the uterus on days 4, 5 and 6 p.c. formaldehyde and 2.5% glutaraldehyde in were studied. The spherical blastocyst has 0.1 M cacodylate buffer for 20 minutes. a single, cellular layer of trophoblast comThe extracellular coats were then dissected posed of thin, polygonal cells. These cells away from the trophoblast using care not have an elevated central region which conto damage the integrity of the spherical tains the nucleus, and they taper to become blastocyst. The blastocysts were then thinner at their circumference. Due to the washed in 0.1 M cacodylate buffer for one polygonal shape, each trophoblast cell is in hour before being immersed in 20% glyc- contact with 5 or 6 other cells (fig. 1 ) . erol in 0.1 M cacodylate buffer for 40 min- There is much variation in the shape of utes just prior to freezing. Blastocysts were the adjoining edges of adjacent trophothen mounted into teflon coated micro- blast cells. Often stepwise overlapping chambers designed for the fracturing of flanges and flat overlapping edges that very small pieces of tissue, and the blasto- adjoin at acute angles to the vertical apicalcysts were either cut or collapsed to form basal axis are observed (figs. 2-4). Occaa flat mount during the loading. The tissue sionally, the contact regions resemble the was rapidly frozen in liquid nitrogen- configuration noted by Farquhar and Pacooled Freon 22 (Chlorodifluoromethane) lade ('63) for normal adult epithelia, in and fracturing was accomplished on a that the contact regions are oriented vertiBalzers apparatus at a stage temperature cally in an apical-basal plane and extend of - 120°C. Replicas were formed by the over long areas in a very regular manner. evaporation of platinum and carbon coats The junctional complex forms at the most
20
RICHARD A. HASTINGS I1 A N D ALLEN C. ENDERS
apical contact region between trophoblast cells. Within these junctional complexes the most apical components are areas where the two unit membranes of adjacent trophoblast cells converge until they approximate one another excluding the extracellular space (fig. 5). Even though the tissues were treated en bloc with uranyl acetate, the individual leaflets of the “unit membrane” structure could not be discerned with sufficient clarity to demonstrate “fusion” of the outer leaflets per se. These apparently fused areas, or tight junctions, consist of one or more punctate regions within individual junctional complexes using transmission electron microscopy. Although the presence of such regions in all micrographs of apical junctional complexes suggests that these tight junctions are in the form of a continuous band, the poor resolution of the membrane leaflets and the difficulty of obtaining more than a few thin sections of the same fused area precludes positive determination of continuity. When lanthanum is placed in the external medium, it penetrates the apical intercellular space between trophoblast cells until it is abruptly stopped by the region of apical membrane fusion (fig. 6). Where the lanthanum appears in adequate concentration, the pointed or V-shaped tip of the advancing front of the tracer demonstrates the convergence of the unit membranes to the point of fusion. Lanthanum was not observed basal to the areas of fusion, which implies that these tight junctions are sufficiently continuous along the apical region between trophoblast cells to exclude this tracer under the conditions employed. The study of freeze fracture replicas of day 5 and day 6 blastocysts revealed additional information on the nature of these apical tight junctions. Along the apical region of trophoblast cells (both day 5 and 6 ) a lattice of anastomosing ridges or strands was demonstrated on the A or inner membrane face, while a series of connecting grooves was found in the B or outer membrane face (figs. 7, 8). This lattice of ridges and grooves was found to be continuous along the apical region between trophoblast cells but was usually just a few ridges deep. The average number of
strands per lattice increased from the day 5 blastocysts, which averaged 2-3 ridges, to the day 6 blastocysts, which averaged 5-6 ridges per lattice. The areas immediately adjacent to regions where three cells adjoin were not included in the averages as these regions generally exhibit more extensive lattices (Claude and Goodenough, ’73).
Characterization of the gap junction Transmission electron microscopy revealed occasional areas between trophoblast cells where the unit membranes were not fused but were separated by an extracellular space of only 20-40 A (fig. 9). These regions were encountered only occasionally and were located basal to the tight junctions. As mentioned previously lanthanum was halted by the apical tight junctions. Consequently this tracer was injected into the blastocyst cavity of day 5 and 6 blastocysts which allowed it to penetrate the spaces between cells from the basal side. Where the section is perpendicular to the unit membranes, the lanthanum appears as a dark material within the extracellular space. The periodic spacing of this dark material gives a beaded appearance to this region (fig. 9). Obliquely sectioned regions of these lanthanum infiltrated junctions have a roughly polygonal substructure. Freeze fracture replicas from day 5 and 6 blastocysts revealed local aggregates of particles that were occasionally located among the ridges of the tight junctions but more commonly were found basal to the lattice. These aggregates (A face) were usually composed of relatively few particles; however, several more extensive polygonal aggregates were observed (fig. 10). Some of the larger polygonal aggregates are similar in structure to previously described gap junctions in the chick embryo epiblast (Revel et al., ’73) and early adult mammalian tissue (Goodenough and Revel, ’70). However, the aggregates observed here occupied only small areas, and the complementary pitted surfaces were not demonstrated. Macula adhaerens Desmosomes were a commonly observed component of the apical junctional com-
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES
plex, but in addition they could be found at different levels along the adjoining edges of the trophoblast cells. Desmosomes showed a wide variation in structure. The desmosomes located proximal to the apical junctional complex were usually extensive and displayed the morphological features of mature desmosomes. These desmosomes had a concentration of amorphous dense material located in the cytoplasm adjacent to the inner leaflet of the unit membrane and associated fibrillar material projecting into the adjacent cytoplasm (fig. 11). In addition, some of these desmosomes had dense material within the intercellular space that was generally 220-250 A wide. Even the larger desmosomes were punctate (macula adhaerens). Other desmosomes observed basally were variable in structure. Many lacked the fibrillar material, and had only a thin accumulation of amorphous material. Desmosomes of this type were commonly found adjacent to expanded extracellular spaces (extracellular “lakes”).
Morphology of junctional complexes during preblastocyst development Two-cell to morula Extensive regions of association between the blastomeres of the two cell embryo were observed; however this association did not include distinct junctional complexes. (See Longo and Anderson, ’69 for a description of the formation of the twocell rabbit embryo). Microvilli projected into the intercellular space and were often so numerous that the region between the blastomeres was filled with microvilli, allowing many areas of close membrane association. In addition, non-microvillous regions were found where the membranes of the two blastomeres were also closely apposed. The first specialized structures between blastomeres were observed in the four-cell embryo. These structures consisted of regions of parallel opposing membranes, separated by an intercellular space of less than 100 A, with localized variable accumulations of amorphous material adjacent to the inner aspect of each membrane. These punctate structures (fig. 12) were termed rudimentary desmosomes as a convenience for description and discus-
21
sion, but their role in subsequent junction formation has not yet been established. Peroxidase penetrated between cells without apparent restriction in all the cleavage stages examined (fig. 14).
Morula to blastocyst The morula is composed of a solid sphere of blastomeres, lacking the central cavity of the blastocyst. This solid, spherical arrangement of the blastomeres creates two populations of cells with regard to their position within the embryo. The outer group of blastomeres comprising the external layer of the sphere is unique in that a portion of each cell surface is exposed to the exterior. The central group of cells is composed of blastomeres that do not directly border the exterior. In older morulae (day 3 P.c.) the two groups showed distinct differences in the nature of the junctions between blastomeres. The blastomeres of the central group were generally separated by a wide extracellular space, and junctions were infrequent. The most commonly observed evidence of junction formation in this group of cells was the rudimentary desmosome. These punctate structures, as in earlier embryos, lacked fibrillar material and had only moderate amounts of amorphous, dense material subjacent to the inner leaflet. Junctional complexes between blastomeres of the outer group was found, with few exceptions, at the apical region. These junctions were characterized by local accumulations of amorphous material in a thin band along the inner leaflet of the unit membrane (fig. 13). They differed from the rudimentary desmosomes in their apical location, and in the extensive nature of the amorphous material. Focal points where the width of the intercellular space between the adjoining blastomeres was reduced to less than the thickness of either apposed cell membrane were occasionally seen in these complexes. The intercellular spaces between the blastomeres of the early morulae (day 2-2.5 P.c.) were small and were usually angular in shape. At all these early developmental periods prior to the blastocyst the tracer, horseradish peroxidase, could penetrate the intercellular spaces between blastomeres. Peroxidase could pass be-
22
RICHARD A. HASTINGS I1 AND ALLEN C. ENDERS
tween blastomeres of the morula (day 2-2.5 P.c.) but was restricted from intercellular passage by trophoblast cells of the day 3 blastocyst. With increasing development, the size of the intercellular spaces increased, and they lost their angular shape, becoming oval or elliptical in outline. The earliest embryos obtained as blastocysts were recovered from the oviduct 3 days 12 hours p.c. These embryos had a single, well formed cavity and a distinct embryonic cell mass beneath the trophoblast. The central cavity of these very early blastocysts was quite small, and the apical junctional complexes were identical to those described for the late morula ( 3 days 5 hours P.c.). The earliest blastocysts examined with the lanthanum method ( 3 days 12 hours P.c.) consistently had lanthanum deposits for nearly the entire length of the area of reduced intercellular space at the apical junctional complex. Blastocysts obtained 3 days 22 hours p.c. consistently exhibited the focal points of approximation of cell membrane or “fusion .” DISCUSSION
The increase in the size of the intercellular spaces between the early and late morula (2-3 days P.c.) and the change in shape of these spaces from irregular and angular to oval or elliptical are the early physical evidence of fluid retention. Since fluid retention implies restricted access to the exterior, Schlafke and Enders (’67) suggested that apical junctional complex formation is a necessary prerequisite to establishment of the cavity of the blastocyst. However, not until the zonula occludens portion of the junctional complex is relatively complete would the restriction of flow be expected to be significant. The progressive nature of apical junction complex formation is not always understood. Hesseldahl (’71 ) interpreted the earlier statements on junctional complexes to mean that whenever apical junctions could be seen in a given section, i t was presumptive evidence of the blastocyst stage. The presence of a single cavity is by definition the criterion of blastocyst formation. The smallest blastocysts obtained in the present study were removed from the oviduct 3 days 12 hours p.c. Even in these early
blastocysts, horseradish peroxidase did not penetrate the intercellular spaces, although focal points of “fusion” of adjacent cell membranes were rare in sections of the junctional complexes. However, by day 4 blastocysts commonly exhibited “fusion” within the apical junctions. The exclusion of lanthanum indicates that these junctions were zonular rather than punctate structures. Unfortunately due to the small size of the younger stages, and the consequent difficulty in getting a relatively flat appropriately oriented preparation following removal of extracellular coats, the earliest embryos that could be readily freeze fractured were day 5 blastocysts. Replicas obtained from the fracturing of day 5 blastocysts revealed the lattice of ridges and grooves extending continuously along the apical region between cells. This observation was particularly useful in demonstrating the zonular nature of the tight junctions (zonulae occludentes) present by day 5 of development. The blastocysts showed an increase in the average number of ridges per lattice from 2-3 on day 5 to 5-6 on day 6. This increase indicates that the development of the apical junctional complexes is progressive. Claude and Goodenough (’73) have recently examined a variety of zonulae occludentes of tight and leaky epithelia using freeze fracture, and have classified these junctions with respect to transepithelial permeability. They stated that the number of ridges per lattice relates more closely to the restriction of transepithelial permeability than do other factors, including junctional depth. Applying their criteria, the greater average number of ridges in the day 6 blastocyst would indicate a decreased intercellular transepithelial permeability compared to the day 5 blastocyst. Since there is an increase in cell numbers and consequently in potentially leaky cell junctions, this reduced permeability would be especially important in the retention of fluid necessary for the continued expansion. Boving (’56) has stated that the transport of the rabbit blastocysts along the uterus and the unique equidistant spacing of these blastocysts depends on having expanded, spherical blastocysts. In addition,
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES
in species whose blastocysts undergo a
large expansion, the increase in size helps bring the trophoblast into apposition with the uterine epithelium (Enders, '71). The mechanism of fluid movement into the cavity appears to involve an active transport system (Cross and Brinster, '70). Cross ('73) has calculated that the net ionic movement of solutes into the cavity of the day 6 blastocyst could be responsible for the influx of 60 ,L/day of fluid into this cavity, if an isosmotic solution were absorbed. He argues that this volume of fluid movement would be sufficient to account for all the fluid accumulation in the cavity prior to implantation, and that therefore an active water transport system as postulated by Tuft and Boving ('70) would be unlikely. Although the calculations of Cross do not include some of the variations in in situ conditions (hydrostatic pressure from uterine contraction, pulsation of the blastocysts), they establish the major role that solute transport can play in blastocyst expansion if this structure is considered as a continuous semipermeable membrane. The zonulae occludentes not only restrict the escape of fluid from the cavity of the blastocyst by an intercellular pathway, but also limit access to the blastocyst cavity. Large molecules must gain access to the cavity of the blastocyst and cells of the embryonic cell mass through the trophoblast cells. Evidence for discriminatory protein ingestion and transport system in the rabbit blastocyst has come from an in vitro study of pinocytosis using protein tracers (Hastings and Enders, '73). Numerous physiological studies have been reported demonstrating electrotonic coupling between cells of adult tissues (Hagiwara and Morita, '62; Kaneko, '71; Weidmann, '66) as well as embryonic tissues (Bennett and Trinkaus, '70; Sheridan, '71; Ito and Loewenstein, '69). Passage of tracers, such as fluorescein, procion yellow, and microperoxidase, between electrotonically coupled adult cells has been used as evidence that the gap junction is the specialized pathway mediating this coupling (Bennett and Dunham, '70; Payton et al., '69; Reese et al., '71). However, fluorescein did not pass between reag-
23
gregated and coupled cells of the teleost embryo (Bennett et al., '72), or between coupled cells of the intact cleavage and blastulae embryos of Xenqus (Slack and Palmer, '69). Although Bennett and Trinkaus ('70) demonstrated electrotonic coupling in Fundulus heteroclitus blastulae, Lentz and Trinkaus ('71) did not find gap junctions until the gastrula stage of this species. The chick (one of the earlier species in which junction formation was studied (Trelstad et al., '67) has recently been shown by freeze fracture to have gap junctions by stage four (Revel et al., '73). Although the small, scattered regions of close membrane association (20-40 A ) described in this study had most of the features of gap junctions, the freeze fracture identification was incomplete since none of the areas of aggregated particles had demonstrable complementary pitted surfaces. Further study will be necessary to determine whether such regions are developmental precursors to more typical gap junctions, are situated such that the complementary face is rarely cleaved, or are even unrelated structures. Since the rabbit blastocyst is large, data concerning electrotonic coupling of the trophoblast cells could be obtained from this species to accompany further freeze fracture information. LITERATURE CITED Adams, C. E., M. F. Hay and C. Lutwak Mann 1961 The action of various agents upon the rabbit embryo. J. embryo. exp. Morph., 9: 468491. Bennett, M. V. L., and P. B. Dunham 1970 Sucrose permeability of junctional membrane at an electrotonic synapse. Biophysical J., 10: 117a. Bennett, M. V. L., M. E. Spira and G. D. Pappas 1972 Properties of electrotonic junctions between embryonic cells of Fundulus. Dev. Biol., 29: 419-435. Bennett, M. V. L., and J. P. Trinkaus 1970 Electrical coupling between embryonic cells by way of extracellular space and specialized junctions. J. Cell Biol., 44: 592-610. Boving, B. G. 1956 Rabbit blastocyst distribution. Am. J. Anat., 98: 403-417. Calarco, P. G., and C. J. Epstein 1973 Cell surface changes during preimplantation development in the mouse. Dev. Biol., 32: 208-213. Claude, P., and D. A. Goodenough 1973 Fracture faces of zonulae occludentes from "tight" and "leaky" epithelia. J. Cell Biol., 58: 3 9 0 4 0 0 . Cross, M. H. 1973 Active sodium and chloride
24
RICHARD A. HASTINGS I1 AND ALLEN C. ENDERS
transport across the rabbit blastocoel wall. Biol. Reprod., 8: 566-575. Cross, M. H., and R. L. Brinster 1970 Influence of ions, inhibitors and anoxia on transtrophoblast potential of rabbit blastocyst. Exp. Cell Res., 62: 303-309. Daniel, J. C. 1963 Some kinetics of blastocyst formation as studied by the process of reconstitution. J. Exp. Zool., 154: 231-238. Enders, A. C. 1971 The fine structure of the blastocyst. In: Biology of the Blastc-yst, R. J. Blandau, ed. Univ. of Chicago Press. pp. 71-94. Farquhar, M. G., and G. E. Palade 1963 Junctional complexes in various epithelia. J. Cell Biol., 17: 375-412. Goodenough, D. A., and J. P. Revel 1970 A fine structural analysis of intercellular junctions in the mouse liver. J. Cell Biol., 45: 273290. Graham, R. C., and M. J. Karnovsky 1966 The early stages of absorption of injected horseradish peroxidase in the proximal tubules of the mouse kidney: Ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem., 14: 291-302. Hafez, E. S . E. 1971 Some maternal factors affecting physicochemical properties of blastocysts. In: Biology of the Blastocyst. R. J. Blandau, ed. Univ. of Chicago Press. pp. 113124. Hagiwara, S . , and H. Morita 1962 Electrotonic transmission between two nerve cells in leech ganglion. J. Neurophysiol., 25: 721-731. Hastings, R. A., and A. C. Enders 1974 Uptake of exogenous protein by the preimplantation rabbit. Anat. Rec., 179: 311-330. Hesseldahl, H . 1971 Ultrastructure of early cleavage stages and preimplantation i n the rabbit. Z . Anat. mEntwick1.-Gesh., 135: 139-155. Ito, S., and W. R. Loewenstein 1969 Ionic communication between eirly embryonic cells. Dev. Biol., 19: 228-243. Kaneko, A. 1971 Electrical connections between horizontal cells i n the dogfish retina J. Physiol. (London), 213: 95-105. Keilin, D., and E. F. Hartree 1951 Purification of horseradish peroxidase and comparison of its properties with those of catalase and methaemoglobin. Biochem. J., 49: 88-104. Lentz, T. C., and J. P. Trinkaus 1971 Differentiation of the junctional complexes of sur-
face cells in the developing Fundulus blastoderm. J. Cell Biol., 48: 455472. Longo, F. J., and E. Anderson 1969 Cytological events leading to the formation of the two cell stage in the rabbit: Association of maternally and paternally derived genomes. J. Ultrastructural Res., 29: 86-118. Lutwak-Mann, C. 1971 The rabbit blastocyst and its environment: Physiological and biochemical aspects. In: Biology of the Blastocyst, R. J. Blandau, ed. Univ. of Chicago Press. pp. 243-260. Manes, C., and J. C. Daniel 1969 Quantitative and qualitative aspects of protein synthesis in the preimplantation rabbit embryo. Exp. Cell Res., 55: 261-268. Payton, B. W., M. V. L. Bennett and G. D. Pappas 1969 Permeability and structure of junctional membranes at a n electrotonic synapse. Science, 166: 1641-1643. Reese, T. S., M. V. L. Bennett and N. Feder 1971 Cell-to-cell movement of peroxidases injected into the septate axon of crayfish. Anat. Rec., 169: 409-420. Revel, J. P., and M. J. Karnovsky 1967 Hexagonal array of subunits in intercellular junctions of the mouse heart and liver. J. Cell Biol., 33: c7.
Revel, J. P., P. Yip and L. L. Chang 1973 Cell junctions i n the early chick embryo - A freeze etch study. Dev. Biol., 35: 302-317. Schlafke, S . , and A. C. Enders 1967 Cytological changes during cleavage and blastocyst formation in the rat. J. Anat., 102: 13-32. Sheridan, J. D. 1971 Dye movement and low resistance junctions between reaggregated embryonic cells. Dev. Biol., 26: 627-632. Slack, C., and J. F. Palmer 1969 Permeability of intercellular junctions in the early embryo of Xenopus Eaevis, studied with a fluorescein tracer. Exp. Cell Res., 55: 416-419. Trelstad, R. L., E. D. Hay and J. P. Revel 1967 Cell contact during early morphogenesis in the chick embryo. Dev. Biol., 16: 78-106. Tuft, P. H., and B. G. Boving 1970 The forces involved i n water uptake by the rabbit blastocyst. J. EXP. ZOO^., 174: 165-172. Weidmann, S. 1966 The diffusion of radiopotassium across the intercalated disks of mammalian cardiac muscle. J. Physiol. (London), 187: 323-341.
PLATES
PLATE 1 EXPLANATION OF FIGURES
1 This micrograph shows cells of an intact day 5 blastocyst viewed with Nomarski phase interference optics. The polygonal outline of the trophoblast cells ( T ) is observed and the area of contact between two cells is shown (arrow). Each trophoblast cell has five to six other cells in contact with it to form this polygonal arrangement. x 1,270.
26
2
The apical region of contact between two day 6 trophoblast cells is shown i n this micrograph. The junctional region consists of parallel membranes bound basally by a desmosome (arrow) that is composed of small, dense plaques sectioned obliquely. x 42,000.
3
The apical junctional complex between two trophoblast cells from a day 5 blastocyst is observed in this micrograph. The cells are adjoined by slightly overlapping processes, with a desmosome (arrow) basal to the apical contact region. x 22,500.
4
This micrograph, of a day 5 blastocyst, shows the small junctional complex (arrow) between the two trophoblast cells. This contact region occurs at the apical end of extensive overlapping flanges from the adjacent cells. A n endoderm cell is observed beneath the trophoblast cells (lower right). x 13,000.
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES Richard A. Hastings I1 and Allen C. Enders
PLATE 1
PLATE 2 EXPLANATION O F FIGURES
28
5
The apical junctional complex between trophoblast cells of a day 4 blastocyst is shown in this micrograph. A thin band of dense material is situated along the cytoplasmic side of each membrane, and one of the points of fusion or loss of the intercellular space is noted (arrow). X 60,500.
6
This micrograph shows the apical junction between two trophoblast cells from a day 5 blastocyst after exposure to lanthanum nitrate for six minutes. The lanthanum which was added to the external medium, coats the cell surface and penetrates between the trophoblast cells until it is stopped by membrane fusion (arrow). x 62,000.
7
This micrograph of a freeze fracture replica from a day 5 blastocyst shows the apical region between trophoblast cells. The lattice of interconnecting ridges (large arrows ) and grooves is continuous along the apical region. This continuous lattice, even though narrow, constitutes a zonula occludens. The small cluster of particles (small arrow) noted within the lattice might represent a small primitive or atypical gap junction. X 54,000.
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES Richard A. Hastings I1 and Allen C. Enders
PLATE 2
PLATE 3 EXPLANATION OF FIGURES
8
This micrograph shows the apical junctional region between two trophoblast cells from a freeze fractured day 6 blastocyst. Ridges (large arrows) and the complementary grooves (small arrow) form the lattice, which is more extensive than the lattice on day 5 . Particles that might have broken from the complementary face are seen lying in the grooves (crossed arrow) in the upper right of the micrograph. X 54,000.
9 The intercellular space in the gap junction in this micrograph has been penetrated by lanthanum which was injected into the blastocyst cavity. The lanthanum demonstrates the polygonal intercellular substructure of the gap junction where the section passes obliquely (large arrow). Note the beaded appearance (small arrow) of the perpendicularly sectioned region of the junction. x 105,000. 10 A local aggregation of particles (arrow) from a freeze fracture replica of a day 5 blastocyst is shown in this micrograph. The A face morphology is similar to that of previously described gap junctions. However the complementary pitted surface (B-face) was not demonstrated. x 72,500.
30
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES Richard A. Hastings I1 and Allen C. Enders
PLATE 3
PLATE 4 EXPLANATION O F FIGURES
11 This micrograph shows a region basal to the apical junctional complex between two day 6 trophoblast cells. A small intercellular space ("lake") is delimited by two desmosomes (arrows), which have filamentous components associated with their dense plaques. Crystalloids ( C ) and mitochondria ( M ) are present in the adjacent cytoplasm. x 52,000.
12 This micrograph of two cells from a four-cell embryo shows a rudimentary desmosome (arrow). The closely apposed dense plaques of these structures are usually without the associated filamentous material seen in the desmosomes of the blastocyst stage. x 40,000. 13 This micrograph shows a typical junctional complex between two blastomeres of a rabbit morula (3 days 5 hours). The membranes are roughly parallel and a slight accumulation of dense material is noted i n the cytoplasm adjacent to the inner leaflet of each membrane. x 145,000.
14
32
This micrograph shows a n area of apposition between two cells of an eight-cell embryo following exposure to horseradish peroxidase for 15 minutes. In this stage and all other cleavage stages prior to the blastocyst, the tracer can penetrate the intercellular space between the blastomeres ( B ) . The electron dense reaction product coats the microvilli i n the intercellular space. X 38,000.
PREIMPLANTATION RABBIT JUNCTIONAL COMPLEXES Richard A. Hastings I1 and Allen C. Enders
PLATE 4
