Connexin trafficking and the control of gap junction assembly in mouse preimplantation embryos

Paul A. De Sousa 1 Gunnar Valdimarsson 1 Bruce J. Nicholson 0 Gerald M. Kidder 1 0 Department of Biological Sciences, State University of New York , Buffalo, New York 14260 , USA 1 Department of Zoology, The University of Western Ontario , London, Ontario, N6A 5B7 , Canada *Author for correspondence - Gap junction assembly in the preimplantation mouse embryo is a temporally regulated event, beginning a few hours after the third cleavage during the morphogenetic event known as compaction. Recently, we demonstrated that both mRNA and protein corresponding to connexin43, a gap junction protein, accumulate through preimplantation development beginning at least as early as the 4-cell stage. Using an antibody raised against a synthetic C-terminal peptide of connexin43, this protein was shown to assemble into gap junction-like plaques beginning at compaction (G. Valdimarsson, P. A. De Sousa, E. C. Beyer, D. L. Paul and G. M. Kidder (1991). Molec. Reprod. Dev. 30, 18-26). The purpose of the present study was to follow the fate of nascent connexin43 during preimplantation development, from synthesis to plaque insertion, and to learn more about the control of gap junction assembly during compaction. Cell fractionation and reverse transcription-polymerase chain reaction were employed to show that connexin43 mRNA is in polyribosomes at the 4-cell stage, suggesting that synthesis of connexin43 begins at least one cell cycle in advance of when gap junctions first form. The fate of nascent connexin43 was then followed throughout preimplantation development by means of laser confocal microscopy, using two other peptide (C-terminal)specific antibodies. As was reported previously, connexin43 could first be detected in gap junction-like plaques beginning in the 8-cell stage, at which time considerable intracellular immunoreactivity could be seen as well. Later, connexin43 becomes differentially distributed in the apposed plasma membranes of morulae and blastocysts: a zonular distribution predominates between outside blastomeres and trophectoderm cells whereas plaque-like localizations predominate between inside blastomeres and cells of the inner cell mass. The cytoplasmic immunoreactivity in morulae was deemed to be nascent connexin en route to the plasma membrane since it could be abolished by treatment with cycloheximide, and redistributed by treatment with monensin or brefeldin-A, known inhibitors of protein trafficking. Treatment of uncompacted 8-cell embryos with either monensin or brefeldin-A inhibited the appearance of gap junction-like structures and the onset of gap junctional coupling in a reversible manner. These data demonstrate that the regulated step in the onset of gap junction assembly during compaction is downstream of transcription and translation and involves mobilization of connexin43 through trafficking organelles to plasma membranes. INTRODUCTION Gap junctions are aggregations (plaques) of aqueous intramembranous channels that couple adjoining cells metabolically. Hemichannels in each cell are referred to individually as connexons, each of which is a hexamer of integral membrane proteins termed connexins (reviewed by Bennett et al., 1991). The connexins constitute a large family of related proteins, commonly identified according to their relative molecular masses as predicted from their cDNAs (e.g. connexin26 is 26 103, connexin31 is 31 103, etc; Haefliger et al., 1992). Although much is now known about the expression of connexin genes in various adult tissues, comparatively little is known about the regulation of nascent connexin trafficking, membrane insertion, and assembly into plaques. Nor is there much precise information about the role of gap junctional coupling in embryonic development. The preimplantation mouse embryo provides a unique opportunity to explore all of these aspects. The assembly of gap junctions occurs de novo in this system and is a temporally regulated event, being initiated during the process of compaction, which begins a few hours after completion of the third cleavage (reviewed by Kidder, 1987). The establishment of intercellular coupling is independent of cell flattening, one of the other components of compaction, but is required for maintenance of the compacted state, and hence for continued development (Buehr et al., 1987; Lee et al., 1987; Bevilacqua et al., 1989). Connexin43 (Cx43) is one member of the connexin gene family whose zygotic expression supplies subunits for gap junction assembly during early development in the mouse. Cx43 mRNA has been detected as early as the 4-cell stage, and accumulates steadily thereafter (Valdimarsson et al., 1991; Nishi et al., 1991). Likewise, Cx43 itself was found by western blotting to accumulate from the 4-cell stage onward (Valdimarsson et al., 1991). These findings explain the relative insensitivity of gap junction formation to inhibition of transcription and protein synthesis from the 4-cell stage (McLachlin et al., 1983; McLachlin and Kidder, 1986). Cx43 can also be detected in the cytoplasm of 4cell embryos, and in both cytoplasmic and gap junctionlike structures beginning with compaction in the 8-cell stage, using the same antibody used for western blotting (Valdimarsson et al., 1991). The expression of three other connexin genes (Cx26, Cx32, and Cx46) has been examined in preimplantation embryos, and none has been found to be zygotically transcribed or to contribute subunits to gap junctional plaques (Barron et al., 1989; Valdimarsson et al., 1991; Nishi et al., 1991). In the present study we have used two other antibodies specific for Cx43 to examine the distribution of this protein and its assembly into gap junctions during preimplantation development. Our goal was to learn more about the control of gap junction assembly during compaction and the contribution of Cx43 to gap junctional channels in later stages. Our evidence indicates that the regulated step in de novo gap junction assembly involves the transfer of nascent Cx43 from intracellular membranes to the plasma membrane, where it is quickly incorporated into plaques. Although Cx43 is inserted into the plasma membranes of all blastomeres, qualitative differences in its distribution become evident with cell polarization and the divergence of the inner cell mass from the trophectoderm. MATERIALS AND METHODS Embryo collection and culture Embryos were flushed from the reproductive tracts of superovulated CF1 mice (Charles River Canada Ltd., St. Constant, Qubec) mated with CB6F1/J males (The Jackson Laboratory, Bar Harbor, ME), as described previously (Barron et al., 1989). Collections were carried out at the following times (hours post-hCG): 1- to 2-cell, 24 hours; uncompacted 4- to 8-cell, 58-67 hours; 8- to 16cell compacted morulae, 74-76 hours; late morulae (up to 32 cells), 80 hours; blastocysts, 90-92 hours. Embryos were cultured in standard egg culture medium (SECM; Spindle, 1980), at 37C in 5% CO2, in the presence or (...truncated)