MOLECULAR REPRODUCTION AND DEVELOPMENT26199-210 (1990)

Changes in Protein Phosphorylation Associated With Compaction of the Mouse Preimplantation Embryo THEODORA BLOOM AND JOSIE MCCONNELL Department of Anatomy, Cambridge Uniuersity, Cambridge, United Kingdom In order to investigate the role ABSTRACT of protein phosphorylation in the early differentiative events of mouse preimplantation development, timed groups of embryos of various stages were incubated in medium containing [32P]orthophosphate and harvested immediately after labelling or following a chase period. The phosphoproteins obtained were separated by electrophoresis in one and two dimensions. While some of the phosphoproteins found were common to all the stages examined, the detection of many depended on both the combination of pulse-labelling and chase periods used and on the developmental stage examined. Some phosphoproteinswere only found in compacted 8-cell embryos, a correlation which suggests a possible link with the post-translational mechanisms which underlie compaction,

translation. Neither phenomenon is prevented by inhibition of the synthesis of DNA, RNA or proteins during the 4-cell stage (Smith and Johnson, 1985; Kidder and McLachlin, 1985).Elements of compaction may even be advanced when protein synthesis is inhibited (Levy et al., 1986). Possible mechanisms by which cell-cell contact could induce intracellular polarisation and alterations in cell adhesivity include the production of intracellular second messengers. In a previous study, 8-cell mouse embryos were exposed to phorbol esters which mimic some actions of the second messenger diacylglycerol and are potent stimulators of the Ca2+-and phospholipid-dependent protein kinase, protein kinase C (Bloom, 1989). Phorbol esters cause various changes to the cytoskeleton and cell surface of 8-cell blastomeres which can be interpreted as an extreme but spatially disorganised version of compaction. This response presumKey Words: Mouse embryo, Polarisation, Phosably arises as a result of stimulation of protein kinase phoproteins C activity. Phosphorylation of certain proteins by protein kinase C may therefore prove to have a role in the normal process of compaction. INTRODUCTION In this paper, the changes in phosphoprotein profile In the development of the mouse embryo, the first associated with passage through the 4-cell and 8-cell major morphological transition after fertilisation oc- stages are described, both after brief pulse labelling of curs a t compaction. This rearrangement of intracellu- embryos with [32Plorthophosphate and following lar structure and intercellular contacts occurs during longer incubations and “chase” periods. While many of the fourth cell cycle. Blastomeres flatten on each other the radiolabelled polypeptides detectable after electro(Ducibella and Anderson, 1975; Lehtonen, 1980) and phoresis in one or two dimensions are similar a t each their contents become polarised. The axis of polarisa- stage examined, there are some changes associated tion is determined by the pattern of cell-cell contacts specifically with passage through the %cell stage (Ziomek and Johnson, 1980; Johnson and Ziomek, which may be related to the cell flattening and polar1981a) and, like cell flattening, seems to involve the isation occurring at this time. Ca2 -dependent cell-cell adhesion molecule uvomoruMATERIALS AND METHODS lin (reviewed by Kemler et al., 1988). The asymmetry Recovery and Culture of Embryos generated in 8-cell blastomeres at compaction is imporMF1 female mice (3-4 weeks; Central Animal Sertant for the subsequent differentiation of two cell types which give rise to the trophectoderm and inner cell vices, Cambridge, U.K.) were superovulated by inmass of the blastocyst (Johnson and Ziomek, 1981b; Ziomek and Johnson, 1982; Balakier and Pedersen, 1982; Fleming et al., 1984; Pedersen et al., 1986; Johnson et al., 1986, 1988; Fleming and Johnson, Received November 28, 1989; accepted January 29, 1990. 1988). Theodora Bloom’s present address is Department of Anatomy & Cell The mechanisms initiating both intercellular flat- Biology, Harvard Medical School, 25 Shattuck Street, Boston, MA tening and intracellular polarisation are entirely post- 02115-6092. Address reprint requests there. +

0 1990 WILEY-LISS, INC.

200

T. BLOOM AND J. MCCONNELL

Fig. 1. Phase contrast micrographs of 8-cell embryos illustrating those scored as A) non-flattened, B)partially flattened, or C)fully flattened. Bar= 20 prn.

traperitoneal injections of 5 I.U.of pregnant mares' serum gonadotrophin (PMS) and human chorionic gonadotrophin (hCG, Intervet) 48 h apart. To obtain embryos, females were paired individually overnight with HC-CFLP males (Interfauna) and inspected for vaginal plugs the next day as an indication of successful mating. Late 2-cell and early 4-cell embryos were recovered a t 46-50 h post-hCG; 8-cell embryos were derived by overnight culture of 2-cell and 4-cell embryos. Two-cell and 4-cell embryos were flushed from oviducts into warmed (37°C) Medium 2 containing 4 mg ml-I BSA (M2 + BSA; Fulton and Whittingham, 1978) and washed before culturing in drops of Medium 16 containing 4 mg ml-' BSA (M16 + BSA; Whittingham and Wales, 1969) under paraffh oil (Martindale), in Fig. 2. One-dimensional SDS-PAGE separation of 1""SlrnethFalcon tissue culture dishes, in 5% CO, in air. ionine-labelled polypeptides synthesised during the 4-cell and %cell All manipulations were carried out a t 37°C on heated stages. Embryos were cultured in M16 + BSA containing l3'S1 methionine for 1h a t 0 h (e4 and e8) and 6 h (14 and 18)postdivision stages or in incubators.

Synchronisation of Embryos 4-cells.Populations of 2-cell embryos were inspected a t hourly intervals and any embryos with 3 cells were selected and cultured for up to 2 h. Those which had not completed division to 4-cells in this period were discarded. Any 4-cell embryos were cultured as synchronised groups with the time a t which the last blastomere was seen to have cleaved designated the time of division; times are expressed as hours postdivision to 4-cells. 8-cells. Populations of 4-cell embryos were inspected at hourly intervals and any embryos with 5-7 blastomeres were selected and cultured for up to 3 h. Those which did not complete division to 8-cells during that time were discarded. All 8-cell embryos were cultured together as synchronised groups with the time of last blastomere cleavage designating the time of division; times are expressed as hours postdivision to 8-cells.

to 4-cells or 8-cells. Relative molecular mass markers (m) 200,92,69, 46, 30 x 10'.

cells (6h postdivision or more) were all flattened completely. The degree of flattening of embryos of intermediate ages is indicated.

Radiolabelling of Polypeptides [35Slmethionine.Eggs and embryos were incubated for l h in 50p1 M16+BSA containing 1.5 mCi/ml of [35Slmethionine (specific activity: 1,100-1,300 mCi/ mM; Amersham) and then washed in large drops (2 x 500 p1) of protein-free M2 containing 6 mg/ml polyvinylpyrrolidone (PVP). For l-dimensional analysis, groups of 10 embryos were pipetted into 10 p1 SDS sample buffer (Laemmli, 1970), boiled for 1min and stored a t -70°C. For 2-dimensional analysis, groups of 50 embryos were transferred to 10 p1 sample lysis buffer (O'Farrell, Assessment of Intercellular Flattening 1975) and stored a t -70°C. The degree of flattening of embryos was determined [32Plorthophosphate. Eggs and embryos were by examination using a Wild dissecting microscope. washed through phosphate-free M16 + BSA (KH2P04 Embryos were compared with standards illustrated in omitted and osmolarity corrected with NaC1) and then Figure 1and scored as completely unflattened (A), par- incubated for 1 h in 50 p1 phosphate-free M16 + BSA tially flattened (B), or completely flattened (0.Em- containing lmCi/ml of [32Plorthophosphate ("32P-me,32P: carrier-free, Amersham; stocks of 32Pwere bryos taken for analysis as early %cells (0 h postdivision) were all unflattened and those taken as late 8- routinely used for up to two weeks after the assay

PROTEIN PHOSPHORYLATION AT COMPACTION

201

Fig. 3. Two-dimensional IEF/SDS-PAGE separation of [%l methionine-labelled polypeptides synthesised during 1 h pulse by A) early 4-cell embryos, 0 h postdivision; B) late 4-cell embryos, 6 h postdivision; C)early 8-cell embryos, 0 h postdivision; D) late 8-cell

embryos, 6 h postdivision. Isoelectric focussing is from the left (approx. pH 7.5)to right (approx. pH 5.0)as indicated and bars indicate the position of relative molecular mass markers (m) as in Figure 2.

date). Samples were collected as for ["%]methioninelabelled embryos. In cold chase experiments, embryos were transferred from 32P-medium via 3 x 300 p1 wash drops of M16 + BSA to 50 pl drops of M16 BSA for incubation, prior to harvest.

For two-dimensional analysis, polypeptides were first separated by isoelectric focussing (IEF). Samples were saturated with urea, frozen, and thawed three times and applied to pre-equilibrated cylindrical 4% acrylamide gels. These were run for a total of 6,000 .V h with the last hour at 800 V and then equilibrated with SDS sample buffer (O'Farrell, 1975) before being placed onto 10% acrylamide slab gels, as described above, to separate proteins by relative molecular mass. Gels were fixed for 30-60 min in 45% methanol, 10% acetic acid before drying down on to filter paper and exposure to Fuji RX X-ray film a t -70°C. Exposure times vary between gels, from 3 to 10 days, to give comparable images; exposure time was altered according to the duration of labelling in different experiments

+

Polyacrylamide Gel Electrophoresis and Autoradiography For one-dimensional analysis, polypeptides were separated using uniform 10% SDS polyacrylamide slab gels containing 0.1% SDS and 0.5M Tris-HC1 (pH 8.81, with a stacking gel of 4.5% acrylamide containing 0.1% SDS and 0.125M Tris-HC1(pH 6.8; method of Laemmli, 1970). Gels were run for 3.5-4h a t 175V.

4 14

3 14

D

18

I

t

Fig. 4.

PROTEIN PHOSPHORYLATION AT COMPACTION

203

Flach et al., 1982; Pratt et al., 1983; Howlett and Bolton, 1985; Howlett, 1986; Endo et al., 1986). To establish whether similar changes occur during subsequent cell cycles, newly synthesised proteins were labelled by incubating timed groups of 4-cell and &cell embryos in medium containing [35Slmethionine for 1 h and were separated by one-dimensional SDS-PAGE (Fig. 2) or in two dimensions by IEFISDS-PAGE (Fig. 3). Examination of replicate gels revealed no reproducible differences, using either method, between embryos labelled during similar periods of the third and fourth cycles. Radiolabelling Embryos With [32P10rthophosphate Preliminary experiments were undertaken to assess the effect of pre-incubation of embryos in medium depleted of phosphate for up to 12 h prior to radiolabelFig. 5. One-dimensional SDS-PAGE separation of [ 3 2 P l ~ r t h ~ -ling with L32P]orthophosphate.The results of these exphosphate-labelled polypeptides detectable in 8-cell embryos after periments showed no effect on the pattern of 32Pprolonged incubation or pulse-chase. &cell embryos were incubated in “P-medium for a) 1h, b) 1 h followed by 5 h chase, or c ) 6 h. Large labelled protein bands seen after one-dimensional arrowheads indicate bands Mr 32K, 35K, and 84K that reproducibly electrophoresis (data not shown). Incubating embryos increase in relative intensity after more than 1 h incubation. Small in phosphate-free medium for up to 12 h also did not arrowheads indicate the position of bands Mr 31K, 40K, and 54K that affect the timing or proportion of embryos developing reproducibly appear only after 6 h incubation or 5 h chase. Arrows indicate bands Mr 76K and 92K that reproducibly disappear (76K) or subsequently to blastocysts. To assess the toxicity of [32Plorthophosphate on emdecrease in relative intensity (92K) after chase compared to pulselabelling. All lanes are taken from a single gel. This was exposed to bryos in culture, groups of 50 embryos were cultured film for a shorter period for lane c than lanes a, b, and m (relative from the late 2-cell stage (50 h postHCG) for 0 h, 1h, 3 molecular mass markers, as in Fig. 2), to allow comparison of embryos h, 6 h, or 12 h in phosphate-free medium supplemented incubated for 6 h in See also Figure 7. with 1 mCi ml-’ [32Plorthophosphate (“32P-medi~m”) and then washed and transferred to non-labelled, comand to the elapsed time from the assay date of the plete medium. For incubations of 1h or 3 h, there was [32Plorthophosphate.Where greatly different exposure no apparent effect on the survival rate or timing of times have been used, this is noted in the figure legend. developing to blastocysts (controls, 1h and 3 h incubaFigures show representative gels, obtained on three tions: 90-98% blastocysts 90 h post-HCG). After 6 h in or more occasions from samples collected during sepa- 32P-medium,only 46% of embryos developed to blastorate experiments. Only those spots which appeared to cysts, the remainder arresting as 2- and 4-cells. After change reproducibly are indicated on 1D and 2D gels. 12 h incubation, no embryos divided beyond 2-cells. RESULTS The Pattern of Proteins Synthesised Does Not Appear to Change With Compaction There are major changes apparent in the protein synthetic profile of mouse embryos during the first and second cell cycles of development (Van Blerkom, 1981;

Fig. 4. a: Two-dimensional IEFISDS-PAGE separation of I”PI orthophosphate-labelled polypeptides detected after 1 h pulse-labelling of A) early 4-cell embryos, B) late 4-cell embryos, C) early 8-cell embryos, D) late 8-cell embryos. Timing, E F ,and markers Mr 92,69, 46,30 x lo’, as for Figure 3. The positions of phosphoprotein spots that reproducibly alter with developmental age are indicated and are illustrated diagrammatically in b. Small arrowheads indicate the position of phosphoproteins detectable in 8-cell (C,D) but not 4-cell (A,B) embryos, Mr 32K, 40K, 51K. Large arrowheads indicate a chain of phosphoproteins, Mr 35K, detectable in 4-cell embryos (A,B) that increase in relative intensity in 8 t e l l embryos (C,D). Reference polypeptides that do not appear to alter in intensity are marked with an asterisk. See also Figure 8.

Pulse-Labelling With [32P10rthophosphate Reveals Some Differences Between 4-Cell and 8-Cell Embryos Groups of 50 embryos of the same age postdivision were pulse-labelled in 32P-medium(1 mCi ml-’) for 1h and washed and harvested immediately. Groups of embryos timed at 0, 3, 6, and 9 h postdivision to 4-cells (Fig. 7A, lanes a-d) or postdivision to 8-cells (Fig. 7B, a-d) showed similar patterns of major phosphoprotein bands after electrophoresis in one dimension. After separation in two dimensions (Fig. 41, some phosphoprotein spots are detectable reproducibly a t both early and late 8-cell stages that are not present at the 4-cell stage (compare Fig. 4A and B with C and D). The positions of examples of such spots, Mr 32K, 40K, and 51K, are illustrated by small arrowheads in Figure 4. Additionally, a chain of phosphoprotein spots Mr 35K which are detectable at the early 4-cell stage show an increase in intensity relative to reference phosphoprotein spots by the early 8-cell stage (large arrow-

204

T. BLOOM AND J. MCCONNELL

medium decreases embryo viability, it is desirable to keep incubation periods to a minimum. 8-cell embryos (0 h postdivision) were therefore labelled for 1 h and “chased” in non-labelled, complete medium for a further 5 h before harvesting. Most of the pattern of phosphoprotein bands obtained on separation in one dimension after a 1h pulse and 5 h chase was similar to that obtained after a 6 h pulse (Fig. 5, compare lanes b and c). In particular, the phosphoprotein bands Mr 31K, 32K,35K,40K,54K,and 84K,referred to above (arrowheads, Fig. 51, all increased in relative intensity in the pulse-chased group as was the case after a 6 h pulse. However, a phosphoprotein band Mr 76K which is present in both 1 h and 6 h pulse-labelled samples was not detectable in the pulse-chased sample and a phosphoprotein band of 92K decreased in relative intensity in the pulse-chased samples (arrows, Fig. 5). The effects on the phosphoprotein profile of the prolonged presence of 32Pcan be distinguished from the effects of developmental age. When %cell embryos were labelled for prolonged periods or in pulse-chase regimes, many novel phosphoprotein bands were seen compared to 1h pulse-labelling (Fig. 5). However, labelling for 6 h at the 8-cell stage occunext division begins division pies half of the cell-cycle and spans the time in develFig. 6. Diagram to illustrate the protocol of an experiment to dis- opment during which compaction occurs. It is therefore tinguish phosphoprotein bands that appear with increasing chase duration from those that appear at a particular time of passage through not clear from these experiments whether the observed the &cell stage. The horizontal axis represents time in hours postdi- changes in phosphoprotein profile result directly from vision. Boxes indicate the 1 h labelling period and horizontal bars the longer labelling period or from changes during indicate the duration of the chase period. Letters and vertical bars compaction. indicate the time of harvesting and correspond to lanes in Figure 7. In order to differentiate changes in phosphoprotein band pattern resulting from the prolonged presence of 32Pfrom those due to arrival a t a particular stage of heads, Fig. 4; compare intensity with spots marked development, groups of timed 8-cell embryos (0, 3, 6, with asterisks). These changes in phosphoprotein pro- and 9 h postdivision) were pulse-labelled and either file are illustrated diagrammatically in Figure 4b.

t

t

Some Phosphoproteins are Detected Only After Prolonged Incubation in [3?P10rthophosphate or Pulse-Chase Longer pulses label more phosphoprotein bands. Incubation of cells in medium containing [“2P]orthophosphate does not cause all phosphoproteins t o become labelled immediately (see Discussion). Groups of &cell embryos (0h postdivision) were incubated in ”P-medium for 1 h (Fig. 5, lane a) or 6 h (Fig. 5, lane c), washed, and harvested immediately. As expected, longer incubation in 32P-medium is associated with a generally increased intensity of labelling. In addition, an increase in the relative intensity of several bands is detected. For example, phosphoprotein bands Mr 32K, 35K,and 84K (large arrowheads, Fig. 5) increased in relative intensity and novel bands Mr 31K,40K,54K appeared (small arrowheads, Fig. 5) after 6 h pulse compared to 1h pulse. “Pulse-chase” produces mostly similar phosphoprotein bands to a continuous pulse over the same time period. As prolonged incubation in 32P-

Fig. 7. One-dimensional SDS-PAGE separation of [”2Plorthophosphate-labelledpolypeptides synthesised by Al4-cell embryos and B)8-cell embryos after pulse-labelling or pulse-chase according to the protocol illustrated in Figure 6. Groups of embryos of 0, 3, 6,or 9 h post division were incubated for 1 h in 32P-mediumand immediately harvested a)0-1h; b) 3-4h,&cell embryos partially flattened; c ) 6-7 h; d)9-10h postdivision. Additional groups were further incubated for 2 h in a non-labelled chase: el labelled 0-1 h, harvested 3 h, embryos partially flattened; D labelled 3-4 h, harvested 6 h, %cell embryos partially flattened; f)labelled 3-4 h, harvested 6 h, 8-cell embryos fully flattened g) labelled 6-7 h, harvested 9 h; h) labelled 9-10 h, harvested 12 h postdivision. Some embryos were pulse-labelledfor 1h followed by 5 h in a non-labelled chase: i) labelled 0-1 h, harvested 6 h; j) labelled 3-4 h, harvested 9 h; k)labelled 5-6 h, harvested 12 h postdivision. Embryos taken as 12 h postdivision to 4-cells or 8-cells had divided to 8-cells and 16-cells, respectively. Spots indicate the position of a band Mr 76K that decreases in relative intensity with increasing chase duration (arrow, Fig. 5);closed arrowheads indicate bands that appear or increase in intensity with 2 h chase, Mr 35K, 37K, 84K,bars indicate bands that appear or increase in intensity with 5 h chase, Mr 32K,39K,54K; open arrowheads indicate bands appearing only in 8-cell embryos more than 3 h postdivision,Mr 31K, 40K.Relative molecular mass markers (m)were run with each gel as in Figure 2.

Fig. 7.

AO-lh

D 0-1-dh

I

B 6-7h

I

&

E 6-7-9h 92 69

,46

30

B Fig. 8.

PROTEIN PHOSPHORYLATION AT COMPACTION harvested immediately or cultured for a series of 2 h and 5 h chase periods, such that the fourth cell cycle was covered (sampling protocol illustrated in Fig. 6). As an additional control, the changing pattern of phosphoprotein bands during the 4-cell stage was mapped according to a similar schedule. Comparison of the 4-cell and 8-cell stages should allow phosphoprotein profile changes associated specifically with compaction to be distinguished from those occurring with passage through each cell cycle. The labelled phosphoproteins obtained from this experimental protocol were then separated in one dimension (Fig. 7) for preliminary analysis. Selected examples of similarly timed embryos were then further analysed by two-dimensional SDS/PAGE (see below, Fig. 8).

The Behaviour of Some Phosphoprotein Bands After Pulse-Labelling or Pulse-Chase Is Similar in 4-Cell and 8-Cell Embryos As described for &cell embryos above (Fig. 5), a phosphoprotein band Mr 76K (marked with dots, Fig. 7) also decreases in relative intensity in pulse-chased samples compared to pulse-labelled samples in 4-cell embryos (Fig. 7A, 4-cells and Fig. 7B, 8-cells, compare lanes e-k with lanes a-d). However, the majority of phosphoprotein bands detected in 4-cells did not alter in relative intensity after 1h pulse plus 2 h chase (Fig. 7A, lanes e-h) or 1 h pulse plus 5h chase (Fig. 7A, lanes l-k) compared to 1 h pulse-labelled samples (Fig. 7A, lanes a-d).

207

bryos. In Figure 7B, closed arrowheads (Mr 35K, 37K, 84K) indicate phosphoprotein bands detected after 1 h pulse plus 2 h or 5 h chase, but only faintly, or not a t all after 1 h pulse alone (compare lanes a-d with lanes e-k). Bars (Mr 32K, 39K, 54K) indicate the position of phosphoprotein bands that appear or increase in relative intensity only after l h pulse plus 5h chase (compare lanes i-k with lanes a-h). By contrast to these observations for 8-cell embryos, no marked increase in intensity of corresponding phosphoprotein bands is detectable in 4-cell embryos (Fig. 7A).

Some Phosphoprotein Bands Are Detected Only in Compacted 8-Cells In addition to phosphoprotein bands detected at the 8-cell stage, the intensity of which vary with chase duration, two phosphoprotein bands are detectable that only appear in pulse-chased samples more than 3h into the fourth cell cycle, independent of chase duration (Fig. 7B, compare lanes f-k with lanes a-e; open arrowheads, Mr 31K, 40K). There are no bands which behave in a corresponding way in the third cell cycle (Fig. 7A).

Two-DimensionalSeparation Reveals Additional Phosphoproteins Associated With the 8-Cell Stage and Pulse-Chase Most of the phosphoproteins that showed differences between labelling protocols or between different stages of development migrate in the same region of a onedimensional SDS-polyacrylamide gel (Mr 30-40K) and Several Phosphoprotein Bands Are Detected are therefore difficult to resolve with confidence. SeOnly in 8-Cells After Pulse-Chase lected groups of 8-cell embryos from similar labelling Most of the phosphoprotein bands that appear or in- regimes were therefore separated by two dimensional crease in relative intensity with a chase period com- electrophoresis (Fig. 8). Spots, the behaviour of which pared to pulse-labelling are detected only in 8-cell em- can be correlated with bands described above, are marked similarly in Figure 8 to Figure 7. Two phosphoprotein bands are described above as increasing in relative intensity with passage through Fig. 8. a: Two-dimensional IEFISDS-PAGE separation of L3*Plor- development (31K) or with chase duration (32K).These thophosphate-labelled polypeptides reproducibly detected in &cell may correspond to a single spot on a two-dimensional embryos, labelled A) pulse 0-1 h postdivision, as for Figure 7, lane a; B) pulse 6-7 h postdivision as for Figure 7 lane c; C) pulse 0-1 h, gel (Mr =32K, PI' = 5.75, marked with a large arrowchase to 6 h postdivision as for Figure 7, lane i; D)pulse 0-1 h, chase head in Fig. 8).This spot appears to increase in relative to 3 h postdivision, embryos partially flattened, as for Figure 7,lane intensity and to shift to a more acidic PI' both with e; E)pulse 6-7h, chase to 9 h postdivision as for Figure 7, lane g. Only increasing chase duration and with passage through part of each gel ( =Mr 30K-70K)is shown in A and B for comparison the 8-cell stage. This observation is consistent with the as the entire gels are shown in Figure 4 C,D, respectively. The posiphosphoprotein being more heavily phosphorylated untions of phosphoprotein spots described in the text are illustrated diagrammatically in b. The large arrowhead (Mr 32K)shows the der these two conditions. position of a spot that may correspond to bands marked with an arThere are no spots, the behaviour of which correrowhead and a bar in Figure 7, increasing in intensity with chase sponds clearly to that of the bands of 35K and 37K duration (see text). Small arrows show the position of spots Mr 42K, described after one-dimensional separation, increasing 44K that appear after 1 h pulse plus 5 h chase that were not detected on one-dimensional gels. Open arrows show the position of spots (Mr in relative intensity with increasing chase duration. 40K,41K)whose intensity increases in late 8-cells compared to early Their appearance on one-dimensional gels may corre8-cells after 2 h chase, which may correspond to a band marked with spond to the increase in relative intensity of the chain an open arrowhead in Figure 7. A large arrow (E)marks a diffuse spot of phosphoprotein spots Mr 35K indicated with large (Mr 40K)that appears only in 8-cell embryos more than 6 h postdivision. Asterisks mark reference spots that do not appear to alter in arrowheads in Figure 4. Such an additional modificaintensity (as in Fig. 4).The gel shown in C has been exposed for less tion is not easily detectable on two dimensional gels time than those in A,B,E,D.Timing and markers (m) as for Figure 7. but could cause an apparent shift in Mr on one-dimen~

~~

208

T. BLOOM AND J. MCCONNELL

sional SDS gels. Alternatively, the phosphoproteins increased phosphorylation of existing phosphoproteins seen after one-dimensional separation may be migrat- t o a level detectable by autoradiography. Different profiles of phosphoproteins were detected ing beyond the range of this IEF system. In addition to the bands that increase in relative in samples taken 1h after the addition of 32Pto the intensity described after one-dimensional separation, medium compared to samples taken some time later. two phosphoprotein spots Mr 42K and 44K are appar- As particular protein kinases use phosphate from difent only after 1h pulse and 5 h chase (arrows, Fig. 8C). ferent sources of nucleotide triphosphate, this observaTwo phosphoprotein spots Mr 40K and 41K, PI’ ~ 6 . 0 tion may reflect different rates a t which nucleotide appear more heavily labelled in late %cells than early triphosphate pools equilibrate with 32P(Cooper et al., 8-cells after 1h pulse and 2 h chase (open arrows, Fig. 1983). Different rates of turnover of phosphate on dif8, compare D and E). These two spots are detectable but ferent phosphoproteins have also been noted, in particat reduced relative intensity in pulse-labelled early ular between those containing phosphotyrosine and and late 8-cells (open arrows, Fig. 8A,B): the spot Mr those containing phosphoserine or phosphothreonine 40K is not detectable in pulse-labelled 4-cells (arrow- (Hunter et al., 1980). The results may, therefore, rehead, Fig. 4). This phosphoprotein migrates very close flect differences in rates of phosphatases. It is also posto, but is more basic than, the diffuse 40K phosphopro- sible that proteins are cleaved after phosphorylation, tein spot (PI‘-5.85, arrow, Fig. 8E) which appears only or are further post-translationally modified (Krebs, in 8-cells more than 6 h postdivision. This group of 1986). It is important to distinguish both those proteins dephosphoprotein spots may therefore be detectable for the first time in early &cell embryos and be increas- scribed here whose phosphorylation state varies with ingly phosphorylated or further modified with passage duration of pulse and chase (Mr 32K, 35K, and 84K, through the 8-cell stage. The behaviour of these phos- Figs. 5, 7) and those whose phosphorylation state varphoprotein spots is illustrated diagrammatically in ies with passage through each cell cycle (such as those of Mr 35K that have been described previously; Figure 8b. Howlett, 1986), from phosphoproteins unique to a parDISCUSSION ticular cell cycle. Most of the phosphoproteins reThis paper describes changes in phosphoprotein pro- stricted to the fourth cell cycle were only detectable file that accompany a major cellular rearrangement of after pulse-labelling embryos for a prolonged period of mouse embryonic development. Blastomeres of 8-cell time (which decreases embryo viability) or after pulsemouse embryos differ from those of 4-cell embryos by labelling for a short period followed by a longer “chase” their ability to flatten onto each other and to polarise in non-radioactive medium (which does not affect viatheir contents and microvilli along an axis determined bility). Chase periods of increasing length were associby the positions of cell contacts. Cell flattening and ated with increased labelling of certain phosphopropolarisation provide the first cellular asymmetries that teins and with the detection of novel phosphoproteins seem to be essential for the later differentiation of two in 8-cell embryos (Mr 32K, 37K, and 84K after 2 h or 5 cell types. These events seem to be initiated post-trans- h chase; Mr 40K, 42K, 44K, 54K in 8-cells only after 5 lationally, using proteins present in the embryo from h chase; Figs. 7, 8). These results suggest that specific pathways of proat least the 4-cell stage onwards (Kidder and McLachlin, 1985; Levy et al., 1986).By studying the changes in tein phosphate metabolism may be used a t the 8-cell protein phosphorylation that occur during this key stage. This may be of significance for the events of time in development, it may be possible to elucidate compaction occurring at this time in development. Adfurther the molecular events underlying these morpho- ditionally, some novel phosphoproteins were only detectable during the second half of the fourth cell cycle logical changes. The only previous study of phosphoproteins a t these (Mr 31K, 40K, Figs. 7,8). Such phosphoproteins might stages is that by Lop0 and Calarco (1982) who pulse- be involved in the late events of cell flattening and labelled large groups of relatively heterogeneous pre- polarisation such as stabilisation of the polar phenoimplantation embryos with $2PJorthophosphate.In the type (see Johnson and Maro, 19861, or in processes ocpresent study, smaller groups of embryos have been curring after, or as a consequence of, compaction. It is not unreasonable to suppose that changes in synchronised more precisely and exposed to 32P for varying periods and combinations of pulse and chase a t protein phosphorylation might mediate cellular events intervals through the third and fourth cell cycles. such as cell flattening and polarisation since it has Whilst no difference was observed in the patterns of been apparent for some time that changes in the phospolypeptides synthesised by embryos (as assessed by phorylation state of an enzyme can alter its activity pulse-labelling with [35Slmethionine,Figs. 2,3), pulse- fundamentally (Krebs, 1985). Many extracellular siglabelling 4-cell and 8-cell embryos with [32P]ortho- nals such as hormones and growth factors bind t o rephosphate revealed clear differences in phosphopro- ceptors in the plasma membrane and produce changes teins within and between cell cycles. These differences in protein kinase activity, directly or indirectly, which may represent the novel phosphorylation of proteins or are presumed t o mediate some or all of their physiolog-

PROTEIN PHOSPHORYLATION AT COMPACTION ical effects (reviewed by Hunter and Cooper, 1985; Krebs, 1985; Sibley et al., 1987). Changing patterns of cell-cell contact at the 8-cell stage of preimplantation development might produce alterations in protein phosphorylation by similar mechanisms. Indirect evidence for such a mechanism has been presented previously (Bloom, 1989). Of particular interest is the observation that phosphorylation of many cytoskeletal proteins alters their function significantly (reviewed by Backman, 1988). Thus, phosphorylation of actin binding proteins (Stossel et al., 1985; Pollard and Cooper, 19861,myosin (Citi and Kendrick-Jones, 19871, tubulin (Hargreaves et al., 1986), vimentin (Inagaki et al., 1987; Geiger, 1987), and nuclear lamins (Ottaviano and Gerace, 1985) are all associated with changes in function and/ or intracellular localisation. Less direct evidence for the role of phosphorylation in changing cell shape and cytoskeletal organisation comes from cells transformed with oncogenes (Jove and Hanafusa, 1987; Kellie, 1988) or using drugs such as phorbol esters to stimulate specifically protein kinase C (see Bloom, 1989). It seems feasible to propose that cell-cell contact causes cell polarisation and flattening via the extracellular activation of protein kinases which are either themselves localised a t cell contacts, or are locally activated or act on localised substrates. By investigating the changes in protein phosphorylation after manipulating the events of compaction, it may now be possible to establish whether protein phosphorylation controls compaction or is simply an early manifestation of the phenotype of flattened, polarised cells.

ACKNOWLEDGMENTS We thank colleagues in the Embryo & Gamete Research Group for their support, in particular, Martin George and Brendan Doe for technical assistance, Ian Bolton for excellent photographic work, and Martin Johnson, Lesley Clayton, and Hester Pratt for invaluable discussions and advice. The work was supported by grants from the Medical Research Council and Cancer Research Campaign to Martin Johnson and Josie McConnell and T.B. gratefully acknowledges receipt of studentships from the MRC and Cambridge Philosophical Society. REFERENCES Backman L (1988):Functional or futile phosphorus? Nature 334:653654. Balakier H, Pedersen RA (1982):Allocation of cells to inner cell mass and trophectoderm lineages in preimplantation mouse embryos. Dev Biol 90:352-362. Bloom TL (1989):The effect of phorbol esters on mouse blastomeres: a role for protein kinase C in compaction? Development 106159-171. Citi S, Kendrick-Jones J (1987): Regulation of non-muscle myosin structure and function. Bioessays 7:155-159. Cooper JA, Sefton BM, Hunter T (1983):Detection and quantitation of phosphotyrosine in proteins. Methods Enzymol 99:387-402. Ducibeila T, Anderson-E (1975):Cell shape and membrane changes in

209

the eight-cell mouse embryo. Prerequisites for morphogenesis of the blastocyst. Dev Biol 47:45-58. Endo Y, Kopf GS, Schultz RM (1986): Stage-specific changes in protein phosphorylation accompanying meiotic maturation of mouse oocytes and fertilisation of mouse eggs. J Exp Zoo1 239:401-409. Flach G, Johnson MH, Braude PR, Taylor RS,Bolton VN (1982):The transition from maternal to embryonic control in the 2-cell mouse embryo. EMBO J 1:681-686. Fleming TP, Johnson MH (1988): From egg to epithelium. Ann Rev Cell Biol 4:459-485. Fleming TP, Warren P, Chisholm JC, Johnson MH (1984): Trophectodermal processes regulate the expression of totipotency within the inner cell mass of the mouse expanding blastocyst. J Embryol Exp Morphol 84:63-90. Fulton BP, Whittingham DG (1978): Activation of mammalian oocytes by intracellular injection of calcium Nature 273:149-151. Geiger B (1987): Intermediate filaments: looking for a function. Nature 329:392-393. Hargreaves AJ, Wandosell F, Avila J (1986): Phosphorylation of tubulin enhances its interaction with membranes. Nature 3282327829. Howlett SK (1986): A set of proteins showing cell-cycle dependent modification in the early mouse embryo. Cell 45387-396. Howlett SK, Bolton VN (1985): Sequence and regulation of morphological and molecular events during the first cell cycle of mouse embryogenesis. J Embryol Exp Morphol87:175-206. Hunter T, Cooper JA (1985): Protein-tyrosine kinases. Ann F k v Biochem 54:897-930. Hunter T, Sefton BM, Beemon K (1980): Phosphorylation of tyrosine: a mechanism of transformation shared by a number of otherwise unrelated RNA tumour viruses. In Animal Virus Genetics (ed. B. N. Fields, R. Jaenisch & C. F. Fox) New York: Academic Press. Inagaki M, Nishi Y, Nishizawa K, Matsuyama M, Sat0 C (19871: Site-specific phosphorylation induces disassembly of vimentin filaments in vitro. Nature 328:649-652. Johnson MH, Mar0 B (1986): Time and space in the mouse early embryo: a cell biological approach to cell diversification. In J Rossant & R Pedersen (ed): “Experimental Approaches to Mammalian Embryonic Development.” Cambridge, UK. Cambridge University Press. Johnson MH, Ziomek CA (1981a): Induction of polarity in mouse 8cell blastomeres: specificity, geometry and stability. J Cell Biol 91: 303-308. Johnson MH, Ziomek CA (1981b): The foundation of two distinct cell lineages within the mouse morula. Cell 24:71-80. Johnson MH, Chisholm JC, Fleming TP, Houliston E (1986): A role for cytoplasmic determinants in the development of the mouse early embryo? J Embryol Exp Morphol. 99 Supplement 97-121. Johnson MH, Pickering SJ, Dhiman A, Radcliffe GS, Maro B (19881: Cytocortical organisation during natural and prolonged mitosis of mouse 8-cell blastomeres. Development 102143-158. Jove R, Hanafusa H (1987): Cell transformation by the viral src oncogene. Ann Rev Cell Biol 3:31-56. Kellie S (1988): Cellular transformation, tyrosine kinase oncogenes and the cellular adhesion plaque. Bioessays 8:25-30. Kemler R, Gossler A, Mansouri A, Vestweber D (1988): The cell adhesion molecule uvomorulin. In G Edelman (ed): “The Cell in Contact. Vol. 2.” New York: Rockefeller Institute Press. Kidder GM, McLachlin JR (1985):Timing of transcription and protein synthesis underlying morphogenesis in preimplantation mouse embryos Dev Biol 112:265-275. Krebs EG (1985): The phosphorylation of proteins: a major mechanism for biological regulation. Trends Biochem SOC13:813-820. Krebs EG (1986):The enzymology of control by phosphorylation In PD Boyer and EG Krebs (ed): “The Enzymes” Vol. 17. New York: Academic Press. Laemmli UK (1970): Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. Lehtonen E (1980): Changes in cell dimensions and intercellular con-

210

T. BLOOM AND J. MCCONNELL

tact during cleavage stage cell cycles in mouse embryonic cells. J Embryo1 Exp Morphol. 58:231-249. Levy JB, Johnson MH, Goodall H, Mar0 B (1986):The timing of compaction: control of a major developmental transition in mouse early embryogenesis. J Embryol Exp Morphol. 95:213-237. Lop0 AC, Calarco PG (1982):Stage-specific changes in phosphorylation during preimplantation development in the mouse. Gamete Res 5:283-290. 0'Farrell PH (1975): High resolution two-dimensional electrophoresis of proteins J Biol Chem 250:4007-4021. Ottaviano Y, Gerace L (1985):Phosphorylation of the nuclear lamin during interphase and mitosis. J Biol Chem 160524-632. Pedersen RA,Wu K,Balakier H 11986): Origin of the inner cell mass in mouse embryos: cell lineage analysis by microinjection. Dev Biol 117:581-595. Pollard TD, Cooper J A (1986): Actin and actin-binding proteins: a critical evaluation of mechanisms and functions. Ann Rev Biochem 55:987-1035. Pratt HPM, Bolton VN, Gudgeon KA (1983): The legacy from the oocyte and its role in controlling early development of the mouse embryo. CIBA Foundation Symposium 98 (Molecular biology of egg maturation; London: Pitman Publishing) pp 197-227.

Sibley DR, Benovic JL, Caron MG, Leflcowitz R J (1987):Regulation of transmembrane signalling by receptor phosphorylation. Cell 48: 913-922. Smith RKW,Johnson MH (1985): DNA replication and compaction in the cleaving embryo of the mouse. J Embryol Exp Morphol. 89: 133-148. Stossel TP, Chapponnier C, Ezzell RM,Hartwig JH, Janmey PA, Kwiatkowski DJ, Lind SE, Smith DB, Southwick FS, Yin HL, Zaner KS (1985):Non-muscle actin-binding proteins. Ann Rev Cell Biol 1:353-402. Van Blerkom J (1981):Structural relationship and post-translational modification of stage-specific proteins synthesised during preimplantation development in the mouse. Proc Natl Acad Sci USA 81: 7629-7633. Whittingham DG, Wales RG (1969): Storage of two-cell mouse embryos in vitro. Aust J Biol Sci 22:1065-1068. Ziomek CA, Johnson MH (1980): Cell surface interactions induce polarisation of mouse 8-cell blastomeres at compaction. Cell 21:935942. Ziomek CA, Johnson MH (1982): The roles of phenotype and position in guiding the fate of 16-cell mouse blastomeres. Dev Biol91:440442.