MOLECULAR REPRODUCTION AND DEVELOPMENT 29:22-28 (1991)

Intercellular Communication in the Early Human Embryo B. DALE, R. GUALTIERI, R. TALEVI, E. TOSTI, L. SANTELLA, AND K. ELDER Stazione Zoologica, Villa Comunale, (B.D., E.T., L.S.), II Cattedra di Istologia ed Embriologia (R.G.), Dipartimento di Biologia Evolutiva e Comparata (R.T.), University of Naples, Naples Italy; Bourn Hall Clinic, Bourn, Cambridge, England (K.E.) A preliminary study on intercellular ABSTRACT communicative devices in the early human embryo has been made using dye-coupling techniques and electron microscopy (EM). Lucifer yellow injected into single blastomeres of embryos at the 4-cell stage u p to the late morula stage did not spread to neighbouring cells, indicating that gap junctions and cytoplasmic bridges are not significant pathways for information transfer. Dye spread was first observed in the blastocyst stage, where trophectoderm cells and inner mass cells were shown to be in communication through gap junctions. Studies at the EM level confirmed this finding. Tight junctions and desmosome-like structures, apparent from the 6-cell stage onward, were located both peripherally and centrally and were initially nonzonular. The role of intercellular devices in the primary differentiation of the human embryo is discussed.

tomeres and caused defects in later stages of development. In the mosaic ascidian embryo, early blastomeres are electrically coupled, but there is no evidence for functional gap junctions until the 8 to 16-cell stage (Dale et al., 1982; Serras et al., 1988, 1989). By contrast, gap junctions and electrical coupling first appear at the early morula stage in the regulative echinoderm embryo. (Tupper et al., 1970; Dale et al., 1982; Andreuccetti et al., 1987). Gap junctions in the mouse embryo are first detectable at the 8-cell stage, when blastomeres undergo compaction (Lo and Gilula, 1979). Early mouse blastomeres are also linked in a syncytial network (Goodall and Johnson, 19841, and slightly later tight junction and desmosome formation begins (Fleming and Johnson, 1988). Little is known about cell-cell interactions, or indeed the first stages of differentiation, in the early human embryo. Studies at the electron microscopic level (EM) Key Words: Human pre-embryo, Blastomeres, Dye have centered on cytoplasmic structures, rather than coupling, Intercellular communicative devices intercellular junctions, and material has often been suboptimal, when not scarce (Dvorak et al., 1982; INTRODUCTION Trounson and Sathananthan, 1984; Pereda and Coppo, Developmental information may be transferred 1987). The purpose of the present paper was to make a through direct cell-cell pathways, such as gap junc- preliminary study of intercellular communication detions and cytoplasmic bridges, or through ligand- vices in the early human embryo using techniques of receptor mechanisms at the cell surface (Slack, 1983). dye coupling and electron microscopy. Gap junctions are present both in vertebrate and in invertebrate embryos; however, their distribution and MATERIALS AND METHODS permeability properties vary according to developmenOocyte Collection and Embryo Culture tal stage and species (Caveney, 1985; Warner, 1988). Mature oocytes were obtained from IVF and GIFT For example, in many embryos, cells of specific devel- patients, who consented to donate them for research opmental fate often form communication compart- purposes. The following follicular stimulation protocol ments, thereby limiting information flow to other cell was used. Long Gn-rh analogue down-regulation groups (Warner and Lawrence, 1982; Weir and Lo, (Buserelin, Hoechst) and human menopausal gonado1982; Kimmel and Law, 1985; Serras and van den tropin daily, from day 2 to day 9 of stimulation (PerBiggelaar, 19871, while in others, such as the squid gonal, Serono, Rome), followed by 5,000 IU of human pre-organogenetic embryo, gap junctions are ubiqui- chorionic gonadotropin (hCG, Profasi, Serono, Rome), tously distributed throughout and between the germ for ovulation induction. layers (Marthy and Dale, 1989). Although information The oocytes were collected by laparoscopy or transis scant, gap junctions appear to play an important role vaginal ultrasonography about 34 hr after hCG adminin early embryogenesis. Microinjection of antibody, or antisense RNA, to a 27-kD gap junction protein into early amphibian (Warner et al., 1984) or mammalian Received May 11, 1990; accepted December 14, 1990. embryos (Lee et al., 1987; Bevilaqua et al., 19891, Address reprint requests to Dr. B. Dale, Stazione Zoologica, Villa reduced dye and electrical coupling between blas- Communale, Naples 80121, Italy.

0 1991 WILEY-LISS, INC.

CELL COUPLING IN THE HUMAN PRE-EMBRYO istration. Oocytes were cultured for 3-6 hr in minimum essential medium (Earle's, Gibco) + 10% heat-inactivated human serum and maintained at 37°C in a gas mixture of 5% CO, in air. The same culture medium was used for preparing semen, used at a final concentration of 105/mlfor insemination. Embryos were cultured in MEM + 20% HS, in 5% CO, in air, at 37°C until use.

23

a cell within 5 sec under constant fluorescent illumination. LY-CH was injected into one or two blastomeres of 13 human embryos, ranging in developmental age from the 4-cell stage to the 10-cell stage (days 2-3 postinsemination). There was no detectable spread of dye to neighbouring blastomeres from the injected cells within 1hr. Two of these injected embryos are shown in Figure la,b. The blastomeres in the 4-cell embryo Dye Coupling shown in Figure l a were deformed by mechanical stress Oocytes were scored for the presence of pronuclei the during the dissection procedure. Four morula stages day after insemination. Corona cells were then re- were also injected with LY, and again there was no dye moved by aspirating oocytes through a fine Pasteur spread. Figure l c shows an early compacted morula 20 pipette. The zona pellucida was removed manually min after microinjection of LY into one of the blasusing fine steel needles under a dissection microscope, tomeres. Later stage morulae also showed no evidence following exposure to 0.1% trypsin (Sigma, St. Louis) of LY coupling (day 4). Extensive dye spread was first for 15 min at 37°C to facilitate removal. detected in the late blastocyst stage (n = 2, days 5-6). Nude embryos were transferred into fresh MEM Figure Id shows one of these blastocysts 10 min after buffered with 10 mM Hepes in a petri dish for micro- injection of LY into one of the trophectoderm cells. Note injection at 25°C. Standard micropipettes of 1-2 km that the dye also flows into the larger inner mass cells. diameter and -10 MR resistance were used to inject The irregular shape of the embryo is attributable to the Lucifer yellow (LY-CH, Sigma St. Louis) into blas- collapse of the blastocoel cavity after removal of the tomeres. Following contact with the cell surface, a GR zona pellucida. seal was obtained, the pipette potential set to -80 mV, Morphological Studies and the patch ruptured. The holding potential was altered to give 0 current and the LY-CH (5% solution in Embryos at the 2-cell stage (n = 21, 6-cell stage 0.2 M LiC1) permitted to diffuse into the cell, rather (n = 3), 8-cell stage (n = 3), morula stage (n = 21, and than cause damage by pressure injection. After injec- blastocyst stage (n = 21, some previously injected with tion, the embryos were observed under epifluorescence LY, were fixed as described. Two-cell stage blastomeres on a ZEISS IM 35 inverted microscope (100 W Hg) and were loosely apposed with numerous microvilli extendphotographed with Kodak Ektachrome ASA 400. ing into the intercellular space (Fig. 2). Apposing plasma membranes were 100-250 nm apart. Neither Electron Microscopy adhesive nor communicative junctions were evident at this stage. At the 6-cell stage, blastomeres were in After dye injection, embryos were fixed in 3% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.3, for 1hr much closer apposition, with the reduced intercellular at room temperature. The embryos were then washed space ranging from 10 to 50 nm. In serial sections of in the same buffer, postfixed in 1% Os04, 0.8% these embryos, desmosome-like structures were found K,Fe(CN)Gin buffer for 1hr at 4"C, quickly washed in in deeply located planes, but not always peripherally distilled water, and stained en bloc in 0.5% uranyl (Fig. 3). These structures were found both basally, i.e., acetate for 1hr at room temperature. After dehydration toward the central cavity of the embryo (Fig. 3a,b), and embryos were embedded in Epon 812. Semithin sec- apically, i.e., toward the zona pellucida, as well as in tions were stained with 1% Toluidine blue in 1% the central regions of blastomeres. In the peripheral sodium borate. Serial thin sections were cut with regions, the desmosome-like structures appeared to be diamond knives (Diatome, Switzerland) on a Ultracut more developed than in other regions; bundles of mimicrotome (Reichert), collected on formvar-coated and crofilaments were seen to extend from the electronuncoated grids, and stained with alcoholic uranyl ace- dense plaques of the desmosome-like structures to the tate and Reynolds lead citrate. Thin sections were microvilli of the free surface of the blastomeres (Fig, observed with a Philips EM 400 electron microscope. 3b). Focal points of apparent membrane fusion, resembling tight junctions, were often seen in the 6- to 10-cell stages associated with the desmosome-like structures RESULTS (Fig. 3b). By serial sectioning, we estimate that the Dye-Coupling Experiments tight junctions and desmosomes were focal and < 5 km Only embryos showing normal developmental pro- in length, i.e., not in a continuous zonular belt. The gression and morphology were used for experiments. situation in the 8-cell stage embryo was similar to that Fragmented embryos were discarded. Blastomeres described above. In the early morula, tight junctions were selected at random for dye injection. Following are more frequent and are associated with a mass of rupture of the patch, measurement of a stable resting filaments on the cytoplasmic side of the plasma mempotential indicated access to the cytosol. Resting poten- brane (Fig. 3c). Gap junctions were not found in any of tials varied from -20 mV to -55 mV (n = 28). In the the early stages, including the late morula, but were whole cell clamp configuration, LY-CH was seen to fill frequent in blastocysts, particularly advanced stages

24

B. DALE ET AL.

Fig. 1. Half-light and half-fluorescence photographs of human pre-embryos after injection of Lucifer yellow (LY) into single blastomeres. a. A 4-cell stage 15 min postinjection, where the zona pellucida was removed manually with steel needles. b. LY injected into a single blastomere of a 10-cell stage human embryo. c. Early compacted morula 20 min after microinjection of LY into a cell. d. Late

blastocyst 10 min after injection of LY into a trophectoderm cell. The blur to the right is the LY-containing micropipette. icm: inner cell mass; te: trophectoderm cells b-d. Zona pellucida of embryos was removed with needles after prior exposure to trypsin as described under Materials and Methods. ~ 6 0 .

(Fig. 41, where they connected trophectoderm cells and inner mass cells and also bridged the two cell lineages.

the 16-cell stage (Dale et al., 1982; Andreuccetti et al., 1987), while in the mouse, dye and electrical coupling between blastomeres occurs at the %cell stage, when the inner cell mass and trophectoderm first develop (Lo and Gilula, 1979; Goodall and Johnson, 1984). In the present study, we show that these intercellular devices are expressed much later in the human pre-embryo, first appearing at the blastocyst stage. Owing to the

DISCUSSION In many embryos, gap junctions appear early in development and often consecutively with early events of differentiation. For example, in the holoblastic regulative sea urchin embryo, they are not expressed until

CELL COUPLING IN THE HUMAN PRE-EMBRYO

Fig. 2. a. Transmission electron micrograph (TEM) of adjacent blastomeres in a 2-cell-stage human embryo. ~3,800.Bar: 3 pm. b. Higher magnification showing the intercellular space of the same embryo devoid of communicative and adhesive structures. x20,400. Bar: 0.5 km. M, mitochondria; MV, microvilla.

25

B. DALE ET AL.

Fig. 3. a. Transmission electron micrograph (TEM) showing intercellular apposition of three blastomeres of a 6-cell stage human embryo. Desmosome-like structures are apparent (D). Bar: 0.5 pm. x20,400. MV, microvilla; MF, microfilament bundle. b. Higher magnification of the desmosome-like structures showing two microfilament bundles running tangential to the plasma membrane. Large

arrowheads indicate focal sites of tight membrane apposition. Bar: 0.5 pm. x43,OOO. c. TEM showing areas of close membrane apposition in a n early morula. Note the presence of a mass of filaments associated with the cytoplasmic side of the plasma membrane. Arrowhead points to sites of apparent membrane fusion. Bar = 100 nm. x 105,000.

scarcity of human material, we have not been able to determine whether there is a difference in the extent of coupling between the trophectoderm cells and those of the inner cell mass. Experiments in the mouse suggest

this to be the case (Lo and Gilula, 1979). We did not detect signs of cell polarization or differentiation in the early cleavage blastomeres that might indicate the initiation of compaction; explanations for the late ex-

CELL COUPLING IN THE HUMAN PRE-EMBRYO

27

Fig. 4. Contact area between a trophectoderm cell and inner mass cell in an advanced blastocyst. Asterisks indicate gap junctions. Bar: 2 pm. x6,OOO. M, mitochondria; N, nucleus; RB, residual body of lysosomal origin; ZP, zona pellucida. (See also Sathananthan et al., 1982.) Insets show high magnification of the gap junctions shown in the centre. Bar: 50 nm. ~220,000.

pression of these junctions in the human would therefore be speculative. Since dye transfer between blastomeres was not detected in the early cleavage and morula stages, and

we did not observe cytoplasmic bridges in serial sections, it seems unlikely that blastomere communication via bridges is a significant pathway for transfer of information. This is different to the mouse (Goodalland

28

B. DALE ET AL.

Johnson, 19841, where a syncytial arrangement exists at early stages. Tight junctions and desmosome-like structures were first observed at the 6-cell stage. This is in contrast to the study of Pereda and Coppo (1987), who observed desmosome-like structures in a 2-cell embryo, but supports the observations of Trounson and Sathananthan (1984). The focal points of apparent membrane fusion appear to be tight junctions, however definition will require freeze-fracture studies and lanthanium tracing. In the mouse, tight junctions are first assembled at the 8-cell stage (Ducibella and Anderson, 1975; Magnuson et al., 1977) and are initially focal. This appears to be the case in the human embryo. Another similarity between the mouse and the human embryo is the observation that focal tight junctions seem to assemble in regions with electron-dense plaques of microfilament-like composition (Fleming and Johnson, 1988). Compaction is a fundamental event in mammalian development that leads to the formation of two primary cell lines: the inner cell mass and the trophectoderm. The finding that gene expression starts at the 4- to 8-cell stage in the human (Braude et al., 1988) lends support to our observations that intercellular devices are not assembled before this stage. It will be of interest to determine whether intercellular devices play a causal role in the primary differentiation of the human embryo or are consequential.

ACKNOWLEDGMENTS We would like to thank Ares Serono for financial support (to B. Dale), Professors Grudzinskas, Chapman, Iaccarino, and Merlin0 for supplying human eggs, and C. Gargiulo and G. Falcone for photography. Experiments were performed according to the guidelines of the N.L.A.

REFERENCES Andreuccetti P, Barone Lumaga MR, Cafiero G, Filosa S, Parisi E (1987): Cell junctions during the early development of the sea urchin embryo (Parcentrotus lzuidus). Cell Diff 20:137-146. Bevilacqua A, Loch-Caruso R, Erickson RP (1989): Abnormal development and dye coupling produced by antisense RNA to gap junction protein in mouse preimplanatation embryos. Proc Natl Acad Sci USA 8654465448. Braude P, Bolton V, Moore S (1988): Human gene expression first occurs between the four and eight cell stages of preimplantation development. Nature 332:459461.

Caveney S (1985):The role of gap junctions in development. Annu Rev Physiol 47:319-335. Dale B, de Santis A, Ortolani G, Rasotto M, Santella L (1982): Electrical coupling of blastomeres in early embryos of ascidians and sea urchins. Exp Cell Res 140:457-461. Ducibella T, Albertini D, Anderson E, Biggers J (1975): The preimplantation mammalian embryo: Characterization of intercellular junctions and their appearence during development. Dev Biol 45:231-250. Dvorak M, Tesarik J , Pilka L, Travnik P (1982): Fine structure of human two-cell ova fertilized and cleaved in vitro. Fertil Steril 37:661-667. Fleming TP, Johnson MH (1988):From egg to epithelium. Annu Rev Cell Biol 4:459435. Goodall H, Johnson MH (1984): The nature of intracellular coupling within the preimplantation mouse embryo. J Embryol Exp Morphol 79:53-76. Kimmel C, Law R (1985): Cell lineage of Zebrafish blastomeres 11. Formation of the yolk syncytial layer. Dev Biol 108:86-93. Lee S, Gilula N, Warner A (1987):Gap junctional communication and compaction during preimplantation stages of mouse development. Cell 51:851-860. Lo C, Gilula N (1979): Gap junctional communication in the preimplantation mouse embryo. Cell 18:399-109. Magnuson T, Demsey A, Stackpole C (1977): Characterization of Intercellular junctions in the pre-implantation mouse embryo by freeze fracture and thin section electron microscopy. Dev Biol 61:252-261. Marthy H, Dale B (1989): Dye-coupling in the early squid embryo. Rouxs Arch Dev Biol 198:211-218. Pereda J, Coppo M (1987): Urastructure of a two cell human embryo. Anat Embryol 177:91-96. Sathananthan A, Wood C, Leeton J (1982):Ultrastructural evaluation of 8 to 16-cellhuman embryos cultured in vitro. Micron 13:193-203. Serras F, van den Biggelaar J (1987):Is a mosaic embryo also a mosaic of communication compartments? Dev Biol 120:132-138. Serras F, Baud C, Moreau M, Guerrier P, van den Biggelaar JAM (1988): Intercellular communication in the early embryo of the ascidian Ciona intestinalis. Development 102:55-63. Serras F, Notenboom RGE, Dictus WJAG, Marthy HJ, van den Biggelaar JAM (1989): Pattern of gap junctional communication in early embryos and tailbud stage of Phallusia mammillata. Slack J (1983): “From Egg to Embryo.” Cambridge: Cambridge University Press. Trounson A, Sathanathan A (1984): The application of electron microscopy in the evaluation of two to four-cell human embryos cultured in vitro for embryo transfer. JIVF ET 1:153-165. Tupper J , Saunders JW, Edwards C (1970): J Cell Biol 46:187-195. Warner A (1988): The gap junctions. J Cell Sci 89:l-7. Warner A, Lawrence P (1982): Permeability of gap junctions at the segmental border in insect epidermis. Cell 28:243-252. Warner AE, Guthrie SE, Gilula NB (1984): Antibodies to gap junctional protein selectively disrupt junctional communication in the early amphibian embryos. Nature 311:127-131. Weir M, Lo C (1982):Gap junctional communication compartments in the Drosophila wing imaginal disk. Proc Natl Acad Sci USA 79:3232-3235.

Intercellular communication in the early human embryo.

A preliminary study on intercellular communicative devices in the early human embryo has been made using dye-coupling techniques and electron microsco...
4MB Sizes 0 Downloads 0 Views