Effects of brefeldin-A and monensin on organelle ... - Springer Link

Dev Genes Evol (1997) 206:503–514

© Springer-Verlag 1997

O R I G I NA L A RT I C L E

&roles:Paul A. De Sousa · Gerald M. Kidder

Effects of brefeldin-A and monensin on organelle distribution and morphology in the preimplantation mouse embryo

&misc:Received: 22 April 1996 / Accepted: 4 December 1996

&p.1:Abstract The intracellular trafficking of integral membrane and secreted proteins is likely to be a key element involved in the morphogenesis and differentiation of the early mammalian embryo. In this study, we used transmission electron microscopy (TEM) to analyse the effects of brefeldin-A (BFA) and monensin, well known inhibitors of vesicular protein trafficking in somatic cells, on the structure of preimplantation mouse embryos. Both BFA and monensin distinctively altered the morphology of Golgi compartments in the blastomeres of treated morulae. BFA-treated morulae lacked recognizable Golgi complexes but possessed heterogeneous organelle clusters consisting of an abundance of smooth tubular and vesicular membrane compartments in addition to mitochondria, endosomes and lysosomes. Treatment of morulae with monensin was associated with swelling of Golgi compartments in addition to altering the morphology of mitochondria, lysosomes and the plasma membrane. BFA, and to a lesser extent monensin, inhibited cytokinesis as evidenced by the detection of binucleate blastomeres. In addition, BFA induced morulae to decompact. These latter effects have not been reported previously for these agents in mammalian somatic cell lines or other vertebrate or invertebrate embryos. These results provide the first demonstration of the structural effects of BFA and monensin on cells of the early mammalian embryo, some of which are consistent with the known actions of these agents on components of the vesicular protein trafficking system in mammalian somatic Edited by R.P. Elinson P.A. De Sousa1 (✉) · G.M. Kidder2 Molecular Genetics Unit and Department of Zoology, The University of Western Ontario, London, Ontario, N6A 5B7, Canada 1 Present address: Department of Obstetrics and Gynaecology and Department of Physiology, Medical Sciences Building Rm 238, The University of Western Ontario, N6A 5C1, Canada 2 Present address: Department of Physiology, The University of Western Ontario, London, Ontario, N6A 5C1, Canada&/fn-block:

cells. This information serves as a foundation for the further use of these agents in studies of vesicular protein trafficking as an agent of preimplantation morphogenesis. &kwd:Key words Preimplantation mouse embryo · Brefeldin-A · Monensin · Golgi · Electron microscopy&bdy:

Introduction The synthesis, maturation, sorting and degradation of secretory and integral membrane proteins is normally accomplished through common organelles and compartments which include the endoplasmic reticulum, Golgi complexes, endosomes and lysosomes. Movement of membrane and proteins between these compartments and to the plasma membrane occurs by budding and fusion of transport vesicles (reviewed by Melancon et al. 1991; Sztul et al. 1992). Over the last decade, considerable insight into the mechanisms by which integral membrane and secretory proteins are trafficked has been obtained using the pharmacological agents brefeldin-A (BFA) and monensin on primary cultures and cell lines. Although both agents have been shown to possess a variety of cell type-specific effects, it has become well established that BFA interferes with the translocation of proteins from the endoplasmic reticulum to the Golgi complex, whilst monensin is a well known inhibitor of the movement of protein through the Golgi complex itself (reviewed in Mollenhauer et al. 1990; Klausner et al. 1992). Both BFA and monensin interfere with the developmental processes of several species, underscoring the importance of protein trafficking events in development. In sea urchins, both agents interfere with the ability of primary mesenchyme cells in the blastocoel to form a spicule, presumably by interfering with secretion of the extracellular matrix proteins and CaCO3 which contribute to this structure (Hwang and Lennarz 1993). Monensin blocks secretion of the hatching enzyme by which sea urchin embryos shed their fertilization envelopes (Roe and Lennarz 1990). Monensin also interferes with gas-

504

trulation in the sea urchin, Drosophila and the chick, probably by inhibiting secretion of extracellular matrix and the spreading of mesodermal cells (Sanders and Chokka 1987; Brachet and Pays 1988; Kuhtreiber et al. 1988; Callaerts and De Loof 1993). The importance of vesicular trafficking of integral membrane and secreted proteins to the success of preimplantation mammalian development is suggested indirectly both by evidence of growth factor-mediated autocrine signalling in early embryos (reviewed by Werb 1990; Pampfer et al. 1991) and by polarization of surface antigens during compaction (reviewed by Wiley et al. 1990). In the mouse, essentially all of the cytoplasmic organelles required for protein trafficking are either inherited in the egg or are formed during the early cleavage divisions following fertilization. Changes in the abundance, morphology and distribution of the endoplasmic reticulum, Golgi, endosomes and lysosomes throughout development are supportive of the notion that utilization of these organelles is both temporally and spatially regulated (Calarco and Brown 1969; Baranska 1980; Reeve 1981; Dvorak et al. 1985; Fleming and Pickering 1985; Maro et al. 1985). More direct evidence that morphogenesis during preimplantation development may be regulated at the level of vesicular protein trafficking comes from studies of the mechanism regulating de novo gap junction assembly during compaction. Gap junctional communication is involved in maintaining the compacted state wherein embryonic blastomeres are flattened against one another, which in turn is a prerequisite for continued development to the blastocyst stage (Buehr et al. 1987; Lee et al. 1987; Bevilacqua et al. 1989). The acquisition of gap junctional coupling during compaction, which in the mouse occurs at the 8-cell stage, can be inhibited by pretreatment of embryos with either BFA or monensin (De Sousa et al. 1993). This inhibition is correlated with an inhibition of the redistribution of a gap junction protein, connexin43, from the cytoplasm of precompacted embryos to the plasma membrane of compacted embryos. Like compaction, the onset of gap junctional communication is also linked to the second round of DNA replication (Smith and Johnson 1985; Valdimarsson and Kidder 1995). A delay of 10 h in DNA synthesis during the second cell cycle, caused by transient treatment with aphidicolin, inhibits the acquisition of gap junctional coupling. This inhibition is also correlated with the failure of nascent connexin43 to be inserted into plasma membranes (Valdimarsson and Kidder 1995). Thus, the temporally regulated trafficking of gap junctional precursors, and other proteins, is probably an essential element of the developmental program controlling preimplantation development. The objective of the present study was to determine by transmission electron microscopy (TEM) the structural effects of both BFA and monensin in the compacted embryos (morulae) of the mouse. In addition to the anticipated effects on the morphology of intracellular trafficking organelles, our results report for the first time an in-

hibitory effect of these agents on cytokinesis and, in the case of BFA, the ability to interfere with maintenance of the compacted state. These results provide a foundation for the further use of these agents in studies of vesicular trafficking as an agent of preimplantation morphogenesis.

Materials and methods Embryo isolation and culture Embryos were flushed from the reproductive tracts of CF1 female mice (Charles River Canada, St. Constant, PQ) which had been superovulated with 5 IU pregnant mare serum gonadotropin (PMSG) followed 46 h later with 5 IU human chorionic gonadotropin (hCG) and then mated with CB6F1/J males (The Jackson Laboratory, Bar Harbor, Me). Embryos were isolated as compacted morulae beginning at 70 h post-hCG in flushing medium-I (FM-I; Spindle 1980). Embryos were cultured in microdrops (50 µl) of standard egg culture medium (SECM; Spindle 1980) under heavy paraffin oil (BDH, Toronto, ON), at 37°C in 5% CO2, in the presence or absence of monensin or brefeldin-A. Monensin (Sigma, St. Louis, Mo.) and brefeldin-A (BFA; Epicenter Technologies, Madison, Wis.) were dissolved in ethanol to make a stock of 10 mg/ml, and used in culture at a concentration of 1–5 µg/ml, equivalent to 3–17 µM for BFA and 1–7 µM for monensin (Donaldson et al. 1990; Mollenhauer et al. 1990; De Sousa et al. 1993). The final concentration of ethanol in all culture conditions ranged from 0.01 to 0.05%. Nuclear staining for epifluorescence microscopy Embryos were fixed for 1 h in 1% paraformaldehyde in PHEM buffer (Schliwa and Van Blerkom 1981; 60 mM Pipes, 10 mM EGTA, 25 mM Hepes, 1 mM MgCl2, pH 6.9). Fixative was removed by two washes with PHEM and embryos were permeabilized for 20 min with PHEM containing 0.1% Tween-20 before staining for 1 h in the same buffer containing 5 µg/ml 4,6-diamidino-2-phenylindole (DAPI; Polysciences, Warrington, Pa.). Unincorporated DAPI was removed with two washes of permeabilization buffer. Embryos were mounted in fluorescein isothiocyanate (FITC) guard mounting medium (Testog, Chicago, Ill.) and viewed on an epifluorescence microscope. Transmission electron microscopy For TEM, embryos were embedded in Epon-Araldite (J.B.E.M, Dorval, PQ). All steps were performed at room temperature using filtered solutions. Embryos were fixed in 2% glutaraldehyde in 0.1 M phosphate buffer (pH 7.3), containing 2% w/v sucrose (PBS) for 1 h. All subsequent steps leading to tissue infiltration were performed on a Fisher clinical horizontal rotator set at a slow speed. Fixative was removed with three 10-min changes of PB-S, after which embryos were preembedded in molten 2% agar which was then trimmed to a 1–2 mm3 block. Embryos were post-fixed for 1 h with 2% OsO4 in PB-S. Following passage through three more 10-min changes of PB-S, embryos were next washed through four changes of sterile double-distilled water (ddH 2O) and left overnight in the last wash at room temperature. The next day, embryos were stained en bloc with saturated aqueous (5%) uranyl acetate for 2.5 h, which was then removed with three 20-min changes of ddH2O. Embryos were dehydrated by treatment with an ascending series of 20, 50, 70 and 90% acetone, for 10 min at each step. Dehydration was completed by two 30-min exposures to 100% acetone. Embryos were infiltrated by end-over-end rotation (0.5 revolutions per minute) for 1 h intervals through a graded series of acetone:resin, wherein the proportion of acetone was de-

505 creased from 3:1 to 1:1 to 1:3 acetone:resin. This was followed by overnight infiltration at room temperature in pure resin and 3–4 h infiltration and encapsulation in fresh resin. Resin was polymerized for 40 h at 60°C in gelatin capsules (size 00; Polysciences, Warrington, Pa.). Epon-Araldite resin was formulated as follows: 4.5 ml Epon (812), 3.5 ml Araldite (502), 18 ml dodecenyl succinic anhydride (DDSA), and 0.67 ml 2,4,6-tri- (dimethylaminomethyl) phenol (DMP-30). DMP-30 was only added after mixing the epoxy components. Silver and silver/gold (60 to 100 nm) sections were cut with glass knives on a Reichert-Jung Ultracut E ultramicrotome and collected on 200-hexagonal-mesh copper grids. Grids were counterstained by treatment for 1 h with saturated aqueous uranyl acetate at 37°C, followed by washing with ddH2O and 10 min with lead citrate. Lead citrate stain was prepared by dissolving approximately 35 mg lead citrate in 10 ml freshly boiled ddH 2O, to which 3 drops of 10 N NaOH was added to dissolve the lead. Undissolved lead was pelleted by centrifugation at 2,000 rpm before use. Lead staining was removed by washing with freshly boiled (CO2-free) ddH2O. Sections were viewed with either a Phillips EM 201 or a Phillips CM10 transmission electron microscope at 60 kV.

Fig. 1A–D Distribution of Golgi complexes in untreated morulae. Golgi complexes (GC) in cultured morulae appeared predominantly in the cortical cytoplasm near apical (A) and basolateral (C,D) membranes, although occasionally they were also seen near nuclei (B). Most Golgi consisted of only 4–6 flattened cisternae slightly dilated at their ends at which vesicles were situated. Also evident at this magnification were mitochondria (M) and lipid droplets (L). (N nuclei, APM apical plasma membrane, ZP zona pellucida, CI crystaline inclusions, scale bars 250 nm)&ig.c:/f

Results To investigate the effect of the trafficking inhibitors on cell and organelle structure, 8- to 16-cell morulae were treated for 4 h with 1–5 µg/ml of either monensin or BFA and examined by TEM. Previously, we showed that this treatment led to redistribution of nascent connexin43 into distinct cytoplasmic structures visualized by immunofluorescence (De Sousa et al. 1993). Sections from 13, 10 and 11 embryos cultured in control medium, monensin and BFA, respectively, were examined. For each inhibitor there was no difference in the effects observed on organelle structure and distribution following treatment with either 1 or 5 µg/ml of that agent. Furthermore, the structural effects observed for each inhibitor were observed in all cells of serially sectioned material and were never observed in the cells of control embryos. Cell and organelle structure in control cultured 8- to 16-cell morulae was similar to that reported previously by others (Calarco and Brown 1969; Dvorak et al. 1985; Fleming and Pickering 1985). Briefly, cytoplasmic organelles tended to be distributed singly or aggregated in small clusters throughout the cytoplasm. These included: mitochondria, lipid droplets, smooth tubular and vesicular membranes, lysosomes and endosomes (Figs. 1, 3A, B, 4E). Both lysosomes and endosomes in the preimplantation mouse embryo have been characterized cytochemically by the detection of endogenous acid hydro-

506 Fig. 2A,B Brefeldin-A (BFA) treatment inhibits cytokinesis. A Low magnification overview of morula treated for 4 h with 1–5 µg/ml BFA depicting an outer binucleate blastomere (with nuclei enclosed by white box) and an inner cell. The nuclei in these cells are juxtaposed by BFA-induced heterogeneous organelle clusters (HOC). B Higher magnification of area boxed in A obtained from adjacent serial section. (N nuclei, scale bars 5 µm)&ig.c:/f

507

Fig. 3A–D BFA-induced heterogeneous organelle clusters contain recognizable organelles. In addition to an abundance of tubular and vesicular membranes, membranous complexes in morulae treated for 4 h with 1–5 µg/ml BFA (C, D) contained endosomes (E), lysosomes (Ls) and mitochondria (M). In control cultured morulae (A, B), smooth tubular and vesicular membranes (SM) appeared as smaller clusters (Scale bars 500 nm)&ig.c:/f

lase activity and exogenous horseradish peroxidase tracer, respectively (Reeve 1981; Fleming and Pickering 1985). Golgi complexes were evident predominantly in the cortical cytoplasm near apical and basolateral membranes, although occasionally they were evident near the nucleus. All Golgi complexes consisted of several (4–6) flattened cisternae slightly dilated at their end parts at which minute vesicles were situated (Fig. 1). BFA alters organelle distribution and abolishes discernible Golgi compartments Treatment of morulae for 4 h with 1–5 µg/ml BFA profoundly altered the distribution, but not the structure, of many cytoplasmic organelles. In contrast to control embryos, where organelles were dispersed singly and in small clusters throughout the cytoplasm, most organelles in BFA-treated embryos were coalesced into large heterogeneous clusters (Fig. 2). Clusters could be found in

proximity to either the nucleus or plasma membrane and were rich in smooth tubular and vesicular membrane compartments in addition to a variety of recognizable organelles including mitochondria, endosomes and lysosomes (Fig. 3). No discernible Golgi complexes were observed in BFA-treated embryos either within or outside of the heterogeneous organelle clusters. Monensin alters the morphology of mitochondria and lysosomes, as well as Golgi compartments Treatment of morulae for 4 h with 1–5 µg/ml monensin resulted in multiple structural effects distinct from those caused by BFA, confirming that each drug possesses unique sites of action. Treatment with monensin dramatically altered the structure of many membranous organelles. Within cells, monensin increased the electron density of mitochondrial matrices, induced swelling of membranous compartments possibly derived from Golgi complexes as judged by their close association with flattened cisternae and increased the apparent size of lysosomes, which now also appeared to contain a medium electron dense material similar to ground cytoplasm (Fig. 4). Lastly, the plasma membrane of monensin-treated embryos examined by TEM appeared disrupted (Fig. 5).

508 Fig. 4A–E Monensin alters the morphology of membranous organelles in morulae. Culturing morulae for 4 h in the presence of 1–5 µg/ml monensin (A,B,C) had multiple effects on the structure of membranous organelles including increasing the electron density of matrices within mitochondria (M), increasing the size and internal complexity of lysosomes (Ls), disrupting plasma membranes along basolateral domains (B, between small arrows) and inducing the swelling of membranous compartments possibly derived from Golgi complexes (arrow in A and at higher magnification in C). Golgi complexes (GC) and endosome clusters (EC) in control cultured morulae (D,E) appeared smaller than putative swollen Golgi cisternae in monensintreated morulae visualized at a similar magnification (C). Also evident in monensin-treated morulae (A) were lipid droplets (L) whose size and distribution appeared unaltered by the drug compared with controls (N nuclei, scale bars 500 nm)&ig.c:/f

Effects on cytokinesis and compacted state of morulae Previously, we observed by light microscopy that treatment of embryos with BFA induced decompaction of morulae and inhibited cytokinesis (De Sousa et al. 1993).

In the present study, the inhibition of cytokinesis was supported by the observation of binucleate blastomeres in electron micrographs of serial sections of BFA-treated embryos (Fig. 2). To further document the frequency with which this phenomenon occurred as well as the in-

509 Fig. 5A, B Apposed basolateral membranes (between arrows) in morulae are disrupted following treatment with monensin. Extensive involution of apposed membrane regions was apparent in morulae cultured in the presence of 1–5 µg/ml monensin (A,B). (M mitochondria, N nuclei (N), scale bars 500 nm)&ig.c:/f

duction of decompaction, 8- to 16-cell morulae were treated blindly for 4 h with either control medium or 5 µg/ml of BFA or monensin. Fixed embryos were then scored for their state of compaction as well as their nuclear content following staining of chromatin using 4,6-

diamidino-2-phenylindole (DAPI; Table 1). An embryo was scored as decompacted if it exhibited rounding of one to several (partial) or all (fully) of its blastomeres in a way that obscured or abolished basolateral plasma membrane domains. Approximately two thirds of control

510 Fig. 6 Induction of decompaction and inhibition of cytokinesis in BFA-treated morulae. Morulae were treated with 5 µg/ml BFA and stained with DAPI to visualize chromatin following fixation. Stained nuclei were illuminated by irradiation with ultraviolet light whilst examining the embryo with dim visible light. Arrows denote binucleate blastomeres (Scale bar 20 µm)&ig.c:/f

Table 1 Frequency of binucleation and decompaction in morulae following 4 h treatment with the trafficking inhibitors BFA and monensin&/tbl.c:& Treatmenta

Control Monensin (5µg/ml) BFA (5µg/ml)

n

24 25 25

Embryos with binucleate blastomeres (%)

State of compaction Fully decompacted embryosb (%)

Partially decompacted embryosb (%)

Fully compacted embryos (%)

0 12 44

8 4 16

25 28 76

67 68 8

a

Fully compacted morulae were selected from a pool of embryos collected and cultured as described in the materials and methods. The experiment was performed blindly with the identity of the treatment revealed only after scoring. Control medium consisted of SECM containing 0.05% ethanol

b An embryo was scored as decompacted if it exhibited rounding of one to several (partial) or all (fully) of its blastomeres in a way that obscured or abolished basolateral plasma membrane domains

or monensin-treated embryos maintained a fully compacted state in contrast to less than a tenth of those embryos treated with BFA. The extent of decompaction was also more severe in BFA-treated embryos, affecting several blastomeres in each embryo, in contrast to usually only a single blastomere in the case of monensin-treated or control embryos. BFA, and to a lesser extent monensin, also inhibited cytokinesis as evidenced by the presence of binucleate blastomeres in 44 and 12% of treated embryos, respectively (Table 1). The number of binucle-

ate blastomeres per affected embryo ranged from one to three. In the case of BFA, binucleation was only observed in the rounded blastomeres of partially or fully decompacted embryos (Fig. 6). In contrast, the binucleate blastomeres observed in monensin-treated embryos remained compacted. Neither BFA nor monensin had any distinguishable effects on nuclear morphology, including the nucleolus, chromatin and nuclear envelope (Figs. 2, 3, 4, 5).

&/tbl.:

511

Discussion Putative targets of BFA in the preimplantation embryo In the present study, we show that treatment of morulae with the protein trafficking inhibitors BFA and monensin results in distinct alterations in the morphology of Golgi compartments as well as other organelles (summarized in Fig. 7). In BFA-treated embryos, no Golgi complexes were apparent. In addition, BFA altered organelle distribution, creating large heterogeneous organelle clusters rich in tubular and vesicular membranes. Although several different types of organelles were evident within these clusters, including mitochondria, lysosomes and endosomes, no effects on their morphology were apparent. The lack of Golgi and the preponderance of smooth tubular and vesicular membranes observed in BFA-treated embryos is consistent with the ability of this drug to cause collapse of Golgi complexes and their reabsorption

Fig. 7 Summary of effects of trafficking inhibitors on embryo organelle structure. Normally, cytoplasmic organelles in embryos are dispersed throughout the cell (Untreated). Readily identifiable organelles include the nucleus (N), mitochondria (M), small patches of tubular and vesicular membranes (SM), Golgi complex (GC) and associated trans-Golgi-like membrane networks (TGN), endosome clusters (EC) and lysosomes (L). Treatment of morulae for 4 h with 1–5 µg/ml BFA resulted in the creation of large heterogenous organelle clusters (HOC), which consisted of an abundance of smooth tubular and vesicular membrane in addition to readily identifiable organelles including endosomes, lysosomes and mitochondria. Binucleation was also observed in the rounded blastomeres of decompacted BFA-treated embryos. Treatment for 4 h with 1–5 µg/ml monensin caused swelling of putative Golgi cisternae, an increase in the size and complexity of lysosomes, an increase in the electron density of mitochondrial matrices and perturbation of apposed plasma membrane regions. Binucleation was also occasionally observed in the compacted blastomeres of monensin-treated embryos with a lesser frequency than that seen with BFA&ig.c:/f

into the endoplasmic reticulum. Trafficking from the endoplasmic reticulum to the Golgi and between Golgi cisternae is mediated by non-clathrin-coated vesicles (reviewed by Rothman and Orci 1992). Vesicle formation is dependent on the recruitment of a set of proteins from the cytoplasm to the budding membrane, the most wellstudied member of which is β-cop. In the presence of BFA, β-cop is redistributed to the cytosol and vesicle formation is inhibited (Donaldson 1990; Orci et al. 1991). When this occurs, Golgi cisternae tend to form tubules which are elongated along microtubules and rapidly fuse with the endoplasmic reticulum (LippincottSchwartz et al. 1989, 1990). As a result, proteins normally resident in the Golgi are redistributed into a mixed Golgi/endoplasmic reticulum compartment (Wood and Brown 1992). The detection of endosomes and lysosomes in the BFA-induced heterogeneous organelle clusters also suggests that the distribution of these organelles is BFA-sensitive. In normal rat kidney (NRK) and primary bovine testicular cells, BFA has been found to induce tubulation and fusing of the trans-Golgi network and early endosomes as well as tubulation of lysosomes, the former suggesting the existence of a transport pathway linking these organelles (Lippincott-Schwartz et al. 1991; Wood and Brown 1992). BFA-treatment of polarized Madin Darby canine kidney (MDCK) cells selectively alters the degree and directionality of endocytosis and transcytosis of vesicles (Hunziker et al. 1991; Low et al. 1992; Prydz et al. 1992). However, no effects on vesicular trafficking to lysosomes and lysosomal function have been observed (Wood and Brown 1992; Strous et al. 1993). The effects of BFA on the structure of the Golgi and the transport of endosomes support the existence of a general mechanism for regulating organelle structure and membrane traffic, which has been proposed to involve microtubules (Kelly 1990).

512

Putative targets of monensin in the preimplantation embryo At the level of the TEM, monensin displayed several effects on embryo organelle structure, including swelling of cisternae probably belonging to Golgi complexes, increasing the apparent size and vacuolar complexity of lysosomes, perturbing the plasma membrane and increasing the electron density of mitochondrial matrices (summarized in Fig. 7). In contrast to BFA, the effects of monensin are probably due to its activity as a Na+ ionophore and its capacity to collapse Na+ and H+ gradients across intracellular membranes (Mollenhauer et al. 1990). In coated vesicles, acidic endosomes, lysosomes and the cisternae of the trans-Golgi network, Na+/H+ gradients are created through the activity of H+-translocating ATP-hydrolyzing enzymes (H+-ATPases), which create acidic intracellular compartments through the transport of H+ into their lumen (Maxfield 1985). Monensin inhibits degradation in lysosomes and the processing and trafficking of proteins at the trans-Golgi network by neutralizing this environment (reviewed by Mollenhauer et al. 1990). In some cells monensin also inhibits endocytosis, presumably as a result of interference with membrane recycling stemming from the neutralization of acidic endosomes (Wilcox et al. 1982). Monensin is also reported to inhibit trafficking of receptor-mediated endocytosed ligand to lysosomes (Merion and Sly 1983). In the case of the trans-Golgi network, loss of function is also accompanied by swelling of its cisternae, although all cisternae of the Golgi complex may swell in response to monensin (Morre et al. 1983). Swelling is presumably due to the flow of water in the direction of osmotically active Na+ into the Golgi lumen. However, in bovine mammary epithelial cells examined using enhanced video microscopy, the swelling of the trans-Golgi network was only observed upon fixation (Morre et al. 1992). Monensin’s effect on Golgi-like compartments in embryos is consistent with this agents ability to interfere with the trafficking of nascent proteins at this organelle. The preponderance of swollen membranous compartments in relation to flattened cisternae in putative Golgi complexes affected by monensin suggests that the effects may not be confined to just the cisternae of the transGolgi network. Inhibition of lysosomal function without a concurrent inhibition in vesicular trafficking to lysosomes presumably would have resulted in an accumulation of material targeted for degradation. In contrast to the Golgi and lysosomes, the matrices of mitochondria appeared condensed in the presence of monensin. This is consistent with the findings of others on primary cultures, cell lines and embryonic cells (Callaerts and De Loof 1993; Sanders and Chokka 1987; reviewed in Mollenhauer et al. 1990) and is probably due to the fact that in these structures H+ are normally pumped in an outward direction (Darnell et al. 1990). The disruption of apposed plasma membrane regions in monensin-treated embryos may have stemmed from a disruption in the secretion of extracellular matrix. A sim-

ilar disruption was also observed in Drosophila embryos treated with monensin during cellular blastoderm formation, which resulted in abnormal cell morphology, disruption of intercellular contacts and the extracellular matrix and a subsequent block in gastrulation (Callaerts and De Loof 1993). Effect of trafficking inhibitors on cytokinesis and compaction in early mouse embryos Treatment of early embryos with BFA and, to a lesser extent, monensin inhibited cytokinesis. BFA also induced decompaction of previously compacted morulae. Interestingly, ours is the first report demonstrating an influence of these agents on cytokinesis. No such effect has been reported even in studies of cell lines cultured up to 24 h in the presence of BFA (Bershadsky and Futerman 1994). One possible explanation for this may be that the shorter cell cycle length of the cleaving blastomeres of a compacted embryo, which can be estimated to be approximately 10 h (Hogan et al. 1994), makes them more susceptible to the effects of these trafficking inhibitors. Since cell division in early murine embryos is asynchronous (Graham and Deussen 1978), this might also explain why the number of affected cells normally ranged from one to three, conceivably the range of cells which might have been dividing over the course of the 4 h treatments. BFA’s ability to inhibit cytokinesis and induce decompaction was unlikely to be a function of the concentration used or the duration of the treatment, which were both within limits routinely used by others (Lippincott-Schwartz et al. 1989; Donaldson et al. 1990; Hunziker et al. 1991). We have previously also shown that treatment of embryos with BFA and monensin is reversible, with development to the blastocyst stage proceeding upon removal of each drug (De Sousa et al. 1993). BFA’s interference with cytokinesis and maintenance of the compacted state may have resulted from inhibition of the delivery of membrane sterols to the surface plasma membrane. Preimplantation mouse embryos are capable of synthesizing membrane sterols from the 8cell stage onward and these are essential for normal cytokinesis and compaction (Pratt and Keith 1980). Given that the endoplasmic reticulum is the site of membrane sterol synthesis (Chesterton 1968) and that this organelle is more adversely affected by BFA treatment than monensin treatment, this may explain why the latter agent affected cytokinesis only minimally and compaction not at all. In summary, many of the morphological effects of BFA and monensin on preimplantation mouse morulae are consistent with previous findings on somatic cell lines and primary cultures, as well as with other invertebrate and vertebrate embryos. In contrast to other embryonic and somatic cell systems however, BFA and, to a lesser extent, monensin were also observed to inhibit cytokinesis, with the former also having the capacity to interfere with the maintenance of the compacted state of

513

morulae. One interpretation of this phenomenon is that proper vesicular trafficking is more essential for cleavage divisions of the early mammalian embryo than for cell division in other systems studied to date. In this regard, mouse embryos which have lost the ability to compact and communicate via gap junctions following a delay of the second round of DNA replication by 10 h (probably due to an interruption in delivery of connexin43 and other proteins to the plasma membrane) lose the ability to cleave further although karyokinesis still occurs (Smith and Johnson 1985; Valdimarsson and Kidder 1995). The present study therefore provides a foundation for the further use of BFA and monensin in studies investigating the contribution of vesicular trafficking to the events controlling preimplantation morphogenesis. &p.2:Acknowledgements The authors wish to acknowledge Mrs. Anita Caveney and Drs. G.L. Babalola, University of Pennsylvania, and A.J. Watson, University of Western Ontario, for critically reviewing the manuscript; Mr. Ron Smith, University of Western Ontario, for assistance and helpful discussions on the preparation of material for electron microscopy, and Mr. Ian Craig for assistance in the preparation of the final figures. This work was primarily conducted while P.D.S. was a doctoral student at the University of Western Ontario and supported by an Ontario Graduate Scholarship to P.D.S. and NSERC operating grant to G.M.K.

References Baranska W (1980) Morphometric ultrastructural analysis of mouse embryo during early development. Gegenbaurs Morphol Jahrb Leipzig 126:349–359 Bershadsky AD, Futerman A (1994) Disruption of the Golgi apparatus by brefeldin blocks cell polarization and inhibits directed cell migration. Proc Natl Acad Sci USA 91:5686–5689 Bevilacqua A, Loch-Caruso R, Erikson RP (1989) Abnormal development and dye coupling produced by antisense RNA to gap junctional protein in mouse preimplantation embryos. Proc Natl Acad Sci USA 86:5444–5448 Brachet J, Pays A (1988) Effects of the monovalent cation ionophore monensin on sea urchin eggs and embryos. Arch Biol 99:265–288 Buehr M, Lee S, McLaren A, Warner A (1987) Reduced gap junctional communication is associated with the lethal condition characteristic of DDK mouse eggs fertilized by foreign sperm. Development 101:449–459 Calarco PG, Brown EH (1969) An ultrastructural and cytological study of preimplantation development of the mouse. J Exp Zool 171:253–284 Callaerts P, De Loof A (1993) Developmental effects of monensin on Drosophila melanogaster. Roux’s Arch Dev Biol 203: 83–91 Chesterton CJ (1968) Distribution of cholesterol precursors and other lipids among rat liver intracellular structures. J Biol Chem 243:1147–1151 Darnell J, Lodish H, Baltimore D (1990) Molecular cell biology, 2nd edn. W.H. Freeman and Company, New York De Sousa PA, Valdimarsson G, Nicholson B, Kidder GM (1993) Connexin trafficking and the control of gap junction assembly in mouse preimplantation embryos. Development 117: 1355–1367 Donaldson JG, Lippincott-Schwartz J, Bloom GS, Kreis TE, Klausner RD (1990) Dissociation of a 110-kD peripheral membrane protein from the Golgi apparatus is an early event in brefeldin A action. J Cell Biol 111:2295–2306 Dvorak M, Cech S, Stastna J, Tesarik J, Travnik P (1985) The differentiation of preimplantation mouse embryos. Monograph, J.E. Purkyne University, Brunensis, Hungary

Fleming TP, Pickering SJ (1985) Maturation and polarization of the endocytotic system in outside blastomeres during mouse preimplantation. J Embryol Exp Morphol 89:175–208 Graham CF, Deussen ZA (1978) Features of cell lineage in preimplantation mouse embryos. J Embryol Exp Morphol 48:53–72 Hogan B, Beddington R, Constantini F, Lacy E (1994) Manipulating the mouse embryo: A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Hunziker W, Whitney JA, Mellman I (1991) Selective inhibition of transcytosis by brefeldin A in MDCK cells. Cell 67:617–627 Hwang S-PL, Lennarz WJ (1993) Studies on the cellular pathway involved in assembly of the embryonic sea urchin spicule. Exp Cell Res 205:383–387 Kelly RB (1990) Microtubules, membrane traffic, and cell organization. Cell 61:5–7 Klausner RD, Donaldson JG, Lippincott-Schwartz J (1992) Brefeldin A: Insights into the control of membrane traffic and organelle structure. J Cell Biol 116:1071–1080 Kuhtreiber WM, Til EH van, Dongen CAM van (1988) Monensin interferes with the determination of the mesodermal cell line in embryos of Patella vulgata. Roux’s Arch Dev Biol 197: 10–18 Lee S, Gilula NB, Warner AE (1987) Gap junctional communication and compaction during preimplantation stages of mouse development. Cell 51:851–860 Lippincott-Schwartz J, Yuan LC, Bonafacio JS, Klausner RD (1989) Rapid redistribution of Golgi proteins into the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell 56:801–813 Lippincott-Schwartz J, Donaldson JG, Schweizer A, Berger EG, Hauri H-P, Yuan LC, Klausner RD (1990) Microtubule-dependent retrograde transport of proteins into the ER in the presence of brefeldin A suggests an ER recycling pathway. Cell 60:821–836 Lippincott-Schwartz J, Yuan L, Tipper C, Amherdt M, Orci L, Klausner RD (1991) Brefeldin A’s effects on endosomes, lysosomes, and the TGN suggest a general mechanism for regulating organelle structure and membrane traffic. Cell 67:601–616 Low SH, Tang BL, Wong SH, Hong W (1992) Selective inhibition of protein targeting to the apical domain of MDCK cells by brefeldin A. J Cell Biol 118:51–62 Maro B, Johnson MH, Pickering S, Louvard D (1985) Changes in the distribution of membranous organelles during mouse early development. J Embryol Exp Morphol 90:287–309 Maxfield FR (1985) Acidification of endocytotic vesicles and lysosomes. In: Paston I, Willingham MC (eds) Endocytosis. Plenum Press, New York, pp 235–257 Melancon P, Franzusoff A, Howell KE (1991) Vesicle budding: insights from cell-free assays. Trends Cell Biol 1:165–171 Merion M, Sly WS (1983) The role of intermediate vesicles in the absorptive endocytosis and transport of ligand to lysosomes by Human fibroblasts. J Cell Biol 96:644–650 Mollenhauer HH, Morre DJ, Rowe LD (1990) Alteration of intracellular traffic by monensin; mechanism, specificity and relationship to toxicity. Biochim Biophys Acta 1031:225–246 Morre DJ, Boss WF, Grimes H, Mollenhauer HH (1983) Kinetics of Golgi apparatus membrane flux following monensin treatment of embryogenic carrot cells. Eur J Cell Biol 30:25–32 Morre DJ, Mollenhauer HH, Spring H, Trendelenburg M, Montag M, Mollenhauer BA, Morre DM (1992) Swelling of Golgi apparatus of bovine mammary epithelial cells in response to monensin treatment requires fixation. Eur J Cell Biol 57: 321–324 Orci L, Tagaya M, Amherdt M, Perrelet A, Donaldson JG, Lippincott-Schwartz J, Klausner RD, Rothman JE (1991) Brefeldin A, a drug that blocks secretion, prevents the assembly of non-clathrin-coated buds on Golgi cisternae. Cell 64:1183– 1195 Pampfer S, Arceci RJ, Pollard JW (1991) Role of colony stimulating factor-1 (CSF-1) and other lympho-hematopoietic growth factors on mouse preimplantation development. Bioessays 13 (10):535–540 Pratt HPM, Keith J (1980) Membrane sterols and the development of the preimplantation mouse embryo. J Embryol Exp Morphol 60:303–319

514 Prydz P, Hansen SH, Sandvig K, Deurs B van (1992) Effects of brefeldin A on endocytosis, transcytosis and transport to the Golgi complex in polarized MDCK cells. J Cell Biol 119: 259–272 Reeve WJD (1981) Cytoplasmic polarity develops at compaction in rat and mouse embryos. J Embryol Exp Morphol 62: 351–367 Roe JL, Lennarz WJ (1990) Biosynthesis and secretion of the hatching enzyme during sea urchin embryogenesis. J Biol Chem 265:8704–8711 Rothman JE, Orci L (1992) Molecular dissection of the secretory pathway. Nature 355:409–415 Sanders EJ, Chokka P (1987) Monensin inhibits secretion of extracellular matrix and the spreading of mesoderm cells in the early chick embryo. J Cell Science 87:389–398 Schliwa M, Van Blerkom J (1981) Structural interaction of cytoskeletal components. J Cell Biol 90:222–235 Smith RK, Johnson MH (1985) DNA replication and compaction in the cleaving embryo of the mouse. J Embryol Exp Morphol 89:133–148

Spindle AI (1980) An improved culture medium for mouse blastocysts. In Vitro 16:669–674 Strous GJ, Vankerhof P, Vanmeer G, Rijnboutt S, Stoorvogel W (1993) Differential effects of Brefeldin-A on transport of secretory and lysosomal proteins. J Biol Chem 268:2341–2347 Sztul ES, Melancon P, Howell KE (1992) Targeting and fusion of vesicular transport. Trends Cell Biol 2:381–386 Valdimarsson G, Kidder GM (1995) Temporal control of gap junction assembly in preimplantation mouse embryos. J Cell Sci 108:1715–1722 Werb Z (1990) Expression of EGF and TGF-α genes in early mammalian development. Mol Reprod Dev 27:10–15 Wilcox DK, Kitson RP, Widnell CC (1982) Inhibition of pinocytosis in rat embryo fibroblasts treated with monensin. J Cell Biol 92:859–864 Wiley L, Kidder GM, Watson AJ (1990) Cell polarity and development of the first epithelium. BioEssays 12:67–73 Wood SA, Brown W (1992) The morphology but not the function of endosomes and lysosomes is altered by brefeldin A. J Cell Biol 119:273–285