Phosphorylation of ribosomal proteins during meiotic maturation and following activation in starfish oocytes: Its relationship with changes of intracellular pH

Experimenral

Cell Research 174 (1988) 71-88

Phosphorylation of Ribosomal Proteins during Meiotic Maturation and Following Activation in Starfish Oocytes: Its Relationship with Changes of Intracellular pH GI?RARD PEAUCELLIER,* ANDRfi PICARD, t JEAN-JACQUES ROBERT,? JEAN-PAUL CAPONY ,$ JEAN-CLAUDE LABBE,* and MARCEL DOREES**’ *Station

Biologique, U.249

F29211 Roscoff; and CNRS-CRBM,

TLaboratoire BP 5051,

Arago, F34033

F66650 Banyuls SW Mer; and SINSERM Montpeilier Cedex, France

An increased phosphorylation of ribosomal protein S6 has been shown to be correlated with an increase of intracellular pH (pHi) and with stimulation of protein synthesis in many systems. In this research changes in ribosome phosphorylation following hormone-induced meiotic maturation and fertilization or activation by ionophore A23 187 were investigated in starfish oocytes. The hormone was found to stimulate, even in the absence of external Na+, the phosphorylation on serine residues of an M, 31,000 protein identified as S6, as well as that of an acidic M, 47,000 protein, presumably Sl, on threonine residues. Phosphorylation of ribosomes was an early consequence of hormonal stimulation and did not decrease after completion of meiotic maturation. Fertilization or activation by ionophore of prophase-arrested oocytes also stimulated ribosome phosphorylation. Only S6 was labeled in this case, but to a lesser extent than upon hormone-induced meiotic maturation. Changes in pHi were monitored with ion-specific microelectrodes throughout meiotic maturation and following either fertilization or activation. The pHi did not change before germinal vesicle breakdown (GVBD) following hormone addition, but it increased before first polar body emission. It also increased following fertilization or activation by ionophore or the microinjection of Ca-EGTA. In all cases, alkalinization did not depend on activation of an amiloride-sensitive Na+/H+ exchanger. Microinjection of an alkaline Hepes buffer or external application of ammonia, both of which increased pHi, prevented unfertilized oocytes from arresting after formation of the female pronucleus and induced chromosome cycling. Phosphorylation of S6 was still observed following fertilization or induction of maturation when pH, was decreased by external application of acetate, a treatment which suppressed the emission of polar bodies. Protein synthesis increased in prophase-arrested oocytes after fertilization or activation. It also increased after ammonia addition, although this treatment did not stimulate S6 phosphorylation. 0 1988 Academic Press. Inc.

Extensive work in sea urchins has shown that the increase in pHi which occurs following fertilization is at least partially responsible for the late events of activation, including stimulation of protein synthesis and release of the egg from cell cycle arrest (for review, see [I]). On the other hand, an increase in phosphorylation of ribosomal protein S6 has been shown to be correlated with an increase in pHi and with stimulation of protein synthesis in many cell division systems, including the fertilized sea urchin egg [2-51. This led to the proposal that in’ To whom reprint requests should be addressed at CRBM, BP 5051, F34033 Montpellier France. 71

Cedex,

Copyright 0 1988 by Academic Press, Inc. All rights of reproduction in any form reserved 0014~4827/88 $03.00

72

Peaucellier

et al.

creased pHi might cause S6 phosphorylation, which in turn would stimulate protein synthesis [6, 71. Two kinds of observations also suggested that a pH elevation might be involved in the hormone-dependent release of amphibian oocytes from cell cycle arrest at first meiotic prophase, as well as in stimulation of S6 phosphorylation and protein synthesis. First, various treatments which increase pHi in the absence of progesterone have been shown to induce meiotic maturation [8-lo] and the stimulation of both S6 phosphorylation and protein synthesis [I 1, 121. Second, an elevation of pHi (about 0.2 units) occurs during progesterone-induced meiotic maturation [IO]. Starfish oocytes are also arrested at first meiotic prophase. When treated with 1-methyladenine (I-MeAde), the relay hormone produced by follicle cells under the influence of a peptide hormone of neural origin, they resume complete meiotic maturation including the emission of two polar bodies and the formation of a haploid nucleus (for review, see [ 13, 141). Although I-MeAde has been shown to increase the incorporation of labeled amino acids into proteins [15-171 as well as protein phosphorylation [l&20] it is not known whether ribosomal proteins become more phosphorylated upon hormonal stimulation. On the other hand, starfish oocytes release acid upon fertilization, although to a lesser extent than sea urchin eggs [17, 211. A small increase in protein synthesis (14%) was reported to occur following fertilization [17] but its possible connection with S6 phosphorylation was not investigated. The aim of this work was to investigate possible changes in phosphorylation of ribosomal proteins during meiotic maturation and following activation by sperm or ionophore A23187. Intracellular pH was also monitored, using pH-specific microelectrodes, following activation and throughout the complete process of meiotic maturation. This provided us with the opportunity to discuss the possible relationships among pHi, S6 phosphorylation, and protein synthesis in the starfish system. We found that (1) 1-MeAde stimulates S6 phosphorylation and a pHi rise after GVBD, (2) the pHi rise was not necessary for S6 phosphorylation increase, and (3) increased S6 phosphorylation was not necessary for protein synthesis stimulation. MATERIAL

AND

METHODS

Handling of oocytes. Ripe females of Marthasterias glacialis were obtained from the Arago Laboratory at Banyuls or from the Biological Station of Roscoff. Fully grown prophase-blocked oocytes were prepared free of follicle cells by washing them several times in artificial Ca”-free seawater (ASW) [22]. Chemicals. Carrier-free [32P]orthophosphate and [y-‘*P]ATP were purchased from Amersham, England. Amiloride, 12-tetradecanoylphorbol 13-acetate (TPA), choline chloride, ionophore A23187, and Hoechst 33258 were obtained from Sigma. The basic formula of artificial seawater used in the experiments of Table 2 was 475 nn’t4 NaCl, 50 nnI4 MgCl*, 10 mM KCl, 10 m&I CaCl*, 10 mit4 Hepes, pH 8.0, whereas in those of Fig. 10 it was 460 mJ4 NaCI, 28 rnY14MgSOI, 25 mM MgC12, 10 ti KCI, 10 mM CaCI,. [32PlPhosphare incorporafion. A 20% suspension of M. glacialis prophase-blocked oocytes was incubated with gentle stirring in the presence of 0.2 mCi/ml of carrier-free [“Plorthophosphate (Amersham). After 30 min incubation, oocytes were washed three times with ASW and separated in

S6 phosphorylation

and pHi in starfish oocytes

73

several batches. Oocytes were activated either by fertilization (10 ul of 250xdiluted sperm suspension/ml) or by addition of 5 @/ml of ionophore A23187 (10m3M in DMSO), both of which resulted in 100% fertilization membrane elevation. Some batches were treated with 10 $/ml of 1 M NH,CI, pH 8.0, or 20 @/ml of 1-MeAde (10m4M in ASW). Preparation of ribosomal fractions. Ribosomes were prepared according to the procedure of Takeshima and Nakano [23]. Oocytes pelleted by low-speed centrifugation were suspended and homogenized in 5 vol of cold 5% sucrose buffer (50 mM Tris-HCl, pH 7.7; 7 nnI4 MgC&; 0.1 M KCI; 10 mM 2-mercaptoethanol). After centrifugation (lO,OOOg, 20 min), the postmitochondrial supernatant was mixed with Triton X-100 (0.5% final concentration) and layered on top of an equal volume of 10% sucrose buffer. Ribosomes were sedimented by centrifugation at 100,OOOgfor 2 h at 4°C. For separation of ribosomal subunits, the ribosomal pellet was resuspended in 0.7 M KC1 buffer and laid on a 15-40% linear sucrose gradient in 0.5 M KC1 buffer, followed by centrifugation at 100,000 g for 15 h. Fractions containing 40 and 60 S subunits, localized by their absorbance at 260 mm, were diluted with buffer without sucrose and centrifuged at 100,OOOgfor 15 h. Ribosomal proteins were extracted with acetic acid, precipitated with acetone and lyophilized, according to Thomas et al. [4]. Gel electrophoresis and autoradiography. Ribosomal pellets were dissolved in SDS-sample buffer (24), whereas postmitochondrial supematants were fixed by addition of 0.3 vol of 4~ concentrated SDS-sample buffer solution. After 4 min at lOO”C, samples were kept frozen at -30°C until use. For SDS-polyacrylamide slab gel electrophoresis (SDS-PAGE), linear gradients (10-20 % acrylamide, 0.3-0.05% bisacrylamide) were used with the discontinuous buffer system of Laemmli [24]. Molecular weight markers were purchased from Pharmacia. Gels were stained with Coomassie blue R-250, dried under vacuum, and processed for autoradiography using X-Omat AR films (Kodak). In control experiments, gels were immersed in 5 % TCA after destaining and kept for 15 min at 90°C: no difference was observed in Coomassie staining and in the pattern of protein phosphorylation. Bidimensional acid-urea gel electrophoresis was performed according to Kaltschmidt and Wittmann [25] with the modifications described by Thomas er al. [4]. Nonequilibrium pH-gradient electrophoresis (NEPHGE) was performed according to O’Farell et al. [26]. Analysis of labeled phosphoamino acids. Labeled phosphoamino acid analysis was carried out as described by Cooper ef al. [27]. Briefly, the gel band was cut out after SDS-PAGE and hydrolyzed for 3 h in 6 N HCl at 1lO”C, and the extract was dried under vacuum. The amino acids were extracted by ion-exchange chromatography on Dowex AGl-X8, eluted with 0.1 N HCl, and concentrated by lyophilization. After addition of unlabeled phosphoserine (P-Ser), phosphothreonine (P-Thr), and phosphotyrosine (P-Tyr), they were analyzed by two-dimensional thin-layer electrophoresis (TLE) on a lOO+m-thick cellulose plate (Merck) with migration at pH 1.9 (1.500 V for 25 min) in the first dimension and at pH 3.5 (1.500 V for 25 min) in the second dimension. The position of standards was determined by ninhydrin staining and radioactivity was scored by autoradiography. Intracellular pH measurements. The pH, measurements were performed as already described [28], using a neutral carrier-based hydrogen ion-selective microelectrode prepared according to Amman et al. [29]. Protein

synthesis. A 5 % suspension of M. glacialis prophase-blocked oocytes was incubated with gentle stirring in the presence of 7.5 @i/ml of [14C]arginine (10e5 M final concentration) at room temperature (20°C). After 30 min incubation, oocytes were washed three times with ASW containing 1 mg/ml unlabeled arginine, resuspended in ASW (1% cell volume/ml), and separated in several batches. Aliquots of oocytes suspension (0.5 ml) were taken at various times and injected into 4.5 ml of 10% TCA containing 1 mg/ml arginine. After standing 15 h at 4”C, samples were washed four times with 5 ml of 10% TCA. The pellets were dissolved in 0.8 ml of 1 N NaOH at room temperature for 15 min, neutralized with 0.2 ml of 4 N HCl, and mixed with 11 ml of scintillation fluid (Ria-luma, Lumac). Incorporation at the end of the experiment was found not to exceed 3 % of the label uptake.

RESULTS Changes in the Extent of Ribosomal during Meiotic Maturation

Protein

Phosphorylation

Prophase-arrested oocytes were preloaded for 3 h with 32P, and then 1-MeAde was added after elimination of external radioactivity. Ribosomes were prepared either just before or at various times after hormone addition (10 min: i.e., before

74

Peaucellier

et al.

Fig. 1. Autoradiographic analysis of “P incorporation Ribosomes were prepared from preloaded oocytes of M. MeAde (2x 10V6M) for 10 min (2), 40 min (3), or 180 min by SDS-PAGE and autoradiography. The migrations proteins are indicated on the left of the autoradiograms. undergone maturation before the I-MeAde addition.)

into ribosomal proteins during maturation. glacialis, before (I) or after treatment with l(4). Equal amounts of proteins were analyzed and molecular weights (X lo-‘) of markers (In this batch 9% of oocytes had already

GVBD; 40 min: i.e., after GVBD; 180 min: i.e., after formation of the female pronucleus). As shown in Fig. 1, ribosomes were strongly labeled after but not before hormone addition. The major labeled phosphoproteins were M, 3 1,000 and, to a lesser extent, M, 47,000 components. The labeling of both phosphoproteins

Fig. 2. Analysis of ‘*P incorporation into proteins of the 40 S ribosomal subunit by bidimensional acid-urea gel electrophoresis. Ribosomal subunits (40 S) were prepared from ‘*P-preloaded oocytes of M. glacialis, treated (A, B, 0 or not (D) with 2x 10m6M I-MeAde for 40 min. Samples were submitted to acid-urea bidimensional electrophoresis and autoradiography. (A) Coomassie blue-stained gel; (I?) corresponding autoradiogram, (C, D) detail of the S6 area from Coomassie blue-stained gels. The arrowheads show S6 position. The direction of electrophoresis in the first (1) and second (2) dimension is indicated at the top left, with the arrows pointing toward the negative electrodes.

S6 phosphorylation

and pHi in starfish

oocytes

75

Fig. 3. Analysis of 32P incorporation into ribosomal proteins by NEPHGE. (A) Coomassie blue staining of the proteins; (B) corresponding autoradiogram. Reference proteins used for the first dimension (nonequilibrium pH gradient) were soybean trypsin inhibitor (pZ 4.5), conalbumin (pZ 5.9), pancreatic ribonuclease (pZ 9.4). Molecular weight markers used for the second dimension (SDS-PAGE) and observable on the stained electrophoregram were, from top to bottom: glycogen phosphorylase (M, 94,000), bovine serum albumin (M, 67,000), ovalbumin (M, 43,000), carbonic anhydrase (M, 30,000), soybean trypsin inhibitor (M, 20, lOO), and a-lactalbumin (it& 14,400). After the gel was dried, dots were made with ‘%T-labeled ink on each molecular weight marker in order to localize them easily on the autoradiogram.

increased from 10 to 40 min, but did not change appreciably from 40 to 180 min after hormone addition. Both phosphoproteins were recovered in the 40 S subunit. Only one spot was observed in autoradiograms of acid-urea bidimensional gel electrophoresis: it was found at the classical location of the M, 31,000 S6 protein (Figs. 2A and 2B) [4, 30, 311. Moreover, the position of the stained protein as well as that of the radioactive spot was found to be significantly shifted toward the anode after lMeAde addition (Figs. 2 C and 20). This indicates that S6 becomes more acidic following hormonal stimulation, so that it can be assumed that its phosphate content actually increased. The ik& 47,000 component was resolved in NEPHGE bidimensional gels (Fig. 3) as an acidic protein at a position in the first dimension similar to that of soybean trypsin inhibitor (pZ about 4.5). In contrast to S6, the radioactivity of the M, 47,000 protein was found to be mostly resistant to a l-h

76

Peaucellier

et al.

D467-

302014-

SNl

P2

40s

Fig. 4. Autoradiographic analysis of 32Pincorporation into ribosomal proteins following activation. Preloaded oocytes were either treated for 25 min with sperm (F), .5x 10e6 M ionophore A23187 (0, 2~10~~ M 1-MeAe (M), or left untreated (C). After homogenization, 10,OOOg supematants (SNl), ribosomal pellets (P2), or 40 S subunits (40 S) were submitted to SDS-PAGE (SNl and P2) or acid-urea bidimensional electrophoresis (40 S). Only autoradiograms are shown from SDS-PAGE, while detail of the S6 area from the Coomassie blue-stained gel is shown on top of the corresponding autoradiogram.

treatment at 55°C with 1 N NaOH. After partial acid hydrolysis of this protein and bidimensional TL electrophoresis (data not shown), as much radioactivity was recovered in phosphothreonine as in phosphoserine (only traces of radioactivity were found to comigrate with phosphotyrosine). Changes in the Extent of S6 Phosphorylation Following Fertilization or Parthenogenetic

and Protein Activation

Synthesis

It is difficult in maturing or fully mature starfish oocytes to investigate whether fertilization or parthenogenetic activation stimulates ribosome phosphorylation and protein synthesis, since 1-MeAde by itself produces both effects. Although fertilization normally occurs between GVBD and first polar body emission, oocytes with an intact germinal vesicle can, however, be fertilized and elevate a fertilization membrane. They remain arrested at first meiotic prophase, unless lMeAde is subsequently added [30]. Therefore all the following experiments were performed using prophase-arrested oocytes. 32P-preloaded oocytes were either fertilized or activated by ionophore A23187 in the absence of external radioactivity. Twenty-five minutes later, postmitochondrial supematants were prepared and proteins were separated by SDS-PAGE. As shown in Fig. 4, labeling of most proteins did not change, or was only slightly increased, depending on the female, following fertilization or parthenogenetic activation. The single evident exception in all females was an M, 3 1,000 protein whose labeling increased markedly in fertilized or activated oocytes. We identified it as S6 because it was completely recovered in the 100,OOOg

S6 phosphorylation TABLE Effect of activation

or maturation

and pHi in star$sh oocytes

77

1

on protein synthesis of M. glacialis Incorporatior+

oocytes”

(min after onset of treatment)

Treatment

0

20

40

None Fertilization Ionophore (5 10m6M) Ammonia (lo-* M) I-MeAde (2 10e6 M)

203f13 203f 13 203213 203+13 203513

319f4 394f22 379f24 331+12 365f14

447+ 10 530+8 552f5 516f5 846+24

’ A suspension of prophase-blocked oocytes, preloaded with [14C]arginine, was divided in several batches and treated as indicated. Protein synthesis was scored on triplicate aliquots of oocyte suspension taken from each batch at the indicated time. b cpm incorporated per 0.5 ml of oocyte suspension (mean + SD of triplicates).

ribosomal pellet and migrated as S6 in acid-urea bidimensional gel electrophoresis of 40 S subunits. Finally we investigated whether the increased phosphorylation of the ribosoma1 protein S6 was correlated with an increased ability of the activated oocytes to incorporate amino acids into proteins. As shown in Table 1, protein synthesis was indeed enhanced following fertilization or activation by ionophore of prophasearrested oocytes, although stimulation was much less pronounced than that following hormonal stimulation. Changes in Intracellular Meiotic Maturation

pH during the Course of I-MeAde-Induced

In other systems, S6 phosphorylation appears to be correlated with an increase in intracellular pH. When fully grown oocytes were treated with 1-MeAde, no change in pHi was observed until GVBD, confirming earlier reports [17, 281. However, pHi was found to be about 0.2 units higher when oocytes had completed meiotic maturation, 120-150 min after hormone addition (six females, ApH=0.22, n=89, SD=0.06). When the time course of pHi changes was investigated during meiotic maturation, a slight and transient pHi decrease (0.08-0.09 units) was observed just after GVBD; then pHi rose before the first polar body emission (Fig. 5). In most cases pHi did not change after the first meiotic cleavage (Fig. 5A). In a few cases, however (3 out of 10 females), it continued to increase slowly until pronuclear formation (Fig. 5 B) . The above experiments were performed with Mediterranean starfish collected near Banyuls. Fully grown oocytes taken from Channel females collected near Roscoff have a somewhat lower pHi (6.92, n=30, SD=O.l). When such oocytes were treated with I-MeAde, pHi also increased by about 0.2 units before the first

78

Peaucellier

et al.

lb

2’0

30

40

50

60

70

so

90

loo

If0 min

a‘,.,

120 ,- Y.M.

130

I40

150

addition

Fig. 5. Time course of pHi changes during meiotic maturation. Two typical experiments using different females are shown (A and B). In each case I-MeAde was added at 0 time to an homogeneous suspension of fully grown prophase-arrested oocytes. Intracellular pH was measured at the indicated times during meiotic maturation, using a different oocyte for each measurement.

polar body emission, but the transient observed (data not shown). Changes in Intracellular pH Following or Parthenogenetic Activation

acidification

just after GVBD

was not

Fertilization

In a first set of experiments, maturing oocytes were fertilized after GVBD (30 min after 1-MeAde addition), and then pHi was measured as a function of time elapsed from fertilization in both the fertilized and the control unfertilized maturing oocytes. Intracellular pH was found to be about 0.15 units higher in fertilized eggs 10 min after sperm addition. Later, pHi changed essentially according to the same schedule in both fertilized and unfertilized eggs, at least until the first mitotic cleavage (Fig. 6). When oocytes with an intact germinal vesicle were fertilized, the pHi did not change for 3-6 min; then it rose progressively to reach a maximal and stable value about 10 min after sperm addition (Fig. 7). Alkalinization induced by fertilization in immature oocytes was about 0.14 pH units, which is not significantly different from alkalinization due to fertilization of maturing oocytes. Intracellular pH was also found to increase, even to a larger extent, following parthenogenetic activation due to the external application of 2 uJ4 ionophore A23187 or the intracellular microinjection of 40 pl of a 0.2 M Ca-EGTA solution adjusted at pH 7.2 with KOH (Fig. 7). The Requirement of an Elevated pHi for Complete Maturation and Initiation of Development

Meiotic

To investigate whether an elevated pHi was required for completion of meiotic maturation, prophase-arrested oocytes were stimulated with 2x 10e6M 1-MeAde, and Na-acetate (pH 6.5) was added 3 min later to a final concentration of 10 mM.

S6 phosphorylation

and pHi in star-sh

PB2

PBl ‘1

Fl P

Cl 1 l”

79

Cl2

1 .

oocytes 1 :AF

A-

?. 0

7-s:::’

”

.IGV 7. 10

so

100

150 min.

Fig. 6. Time at 0 time, and was measured (GV) refers to

after

.

I-MeAde

200

addition

course of pHi changes following fertilization of maturing oocytes. I-MeAde was added sperm was added (F, fertilized) or not (U, unfertilized) 30 min later. Intracellular pH at the indicated time using a different oocyte for each measurement. The vertical bar the initial value of pHi before 1-MeAde addition.

During the next 10 min, pHi dropped from 7.18 (n=lO, SD=0.02) to 6.75 (n=5, SD=O.OI), and then GVBD occurred without delay with respect to control oocytes. About 30 min after hormonal stimulation, pHi began to rise simultaneously in both acetate-treated and control maturing oocytes. Finally pHi leveled off 90 min after 1-MeAde addition to reach 7.04 in acetate-treated (n=lO, SD=0.04) and 7.45 in control mature oocytes. Thus the extent of 1-MeAdeinduced alkalinization was very similar in both cases (ApH=0.29 in acetatetreated oocytes; ApH=0.27 in control oocytes) but acetate-treated oocytes never reached the same absolute value of pHi. In this experiment performed with females captured near Banyuls, as in similar experiments performed with females

A

7.1.

A

A23187 0

A

BP2 10

0

CLEGTA 10

time

tmin)

7. Time course of pHi changes following fertilization or parthenogenetic activation of prophase-arrested oocytes. Three different experiments were performed using the same batch of oocytes; sperm (Spz) or ionophore A23187 was added from outside, while Ca-EGTA was microinjected. The initial value of pH, before activation was 7.2. A different oocyte was used for each measurement. Fig.

6-888331

80

Peaucellier

et al.

captured near Roscoff, polar bodies were never emitted in acetate-treated oocytes, which arrested with a large-sized tetraploid instead of the normal haploid pronucleus. Taken together our results show that a sufficiently alkaline pH is required for meiotic cleavage, but not for GVBD. We found that pHi increased when M. glacialis oocytes were activated by sperm or ionophore at the germinal vesicle stage or after hormone addition, between GVBD and first polar body emission. Since unactivated oocytes normally arrest after formation and migration of the female pronucleus, this suggested that elevation of pHi due to sperm or ionophore A23187 might prevent mature oocytes from arresting after pronucleus formation. Depending on the females and the period of the year, a small (< 10%) and variable proportion of hormonestimulated oocytes spontaneously escape cell cycle arrest. In agreement with the above hypothesis it was observed that pHi is always higher in such cycling cells than in normal mature arrested oocytes [A. Picard, unpublished results]. In the next experiment maturing unactivated oocytes were transferred 40 min after lMeAde addition to seawater containing NH&l and returned to normal seawater 15 min later. After the two polar bodies were emitted and a normal pronucleus was formed, the pronuclear envelop disappeared and chromosomes condensed in most NH&l-treated oocytes (greater than 90%). Several chromosome cycles occurred subsequently, without cleavage, producing eggs with large numbers of chromosomes. Similar results were obtained when oocytes were treated with NH&l after formation of the female pronucleus. In another experiment, maturing oocytes (intracellular volume about 2 nl) were injected with 40 pl of a 0.4 M Hepes buffer, pH 8.4, between GVBD and first polar body emission, which increased pHi by about 0.4 units. In this experiment, none of the control (injected with 40 pl distilled water) oocytes escaped cell cycle arrest after pronuclear formation. In contrast a high proportion of Hepes-injected oocytes (25/30) escaped cell cycle arrest and several chromosome cycles occurred without cleavage after completion of meiotic maturation. Mechanism Of PHi Changes during Meiotic Maturation and Following Fertilization or Parthenogenetic Activation Frequently intracellular alkalinization results from an enhanced activity of the Na+/H+ exchanger [31-331. On the other hand, the activity of the Na+/H+ exchanger is usually inhibited by amiloride [34, 351 and the amiloride-sensitive Na+/H+ exchanger has been reported to be stimulated by TPA in several systems, including sea urchin eggs [36, 371. As shown in Table 2, however, no change in pHi, was observed in prophase-arrested oocytes following the external addition of 100 nM TPA or 1 rnM amiloride, even in artificial seawater containing 5 mM Na+ (Na+ partially replaced either by impermeant cation choline or by K+) or in Na-free seawater (Na+ replaced by K+). The above data suggested that no Nat/H+ exchanger was at work in prophasearrested oocytes, but it remained possible that it was switched on following lMeAde addition. Thus 1 mM amiloride was added at the time of GVBD in normal seawater, 5 mM Na+ seawater (Na+ partially replaced by choline), or Na+-free

S6 phosphorylation

TABLE Effect of amiloride

and pHi in starfish

oocytes

81

2

and TPA on pHi in fertilized

and unfertilized

starfish

oocyte8

PH, Treatment Unfertilized oocytes Control in ASW Control in 5 n&f Na-ASWb Control in 5 mM Na-ASW 1 mM amiloride in ASW 1 mM amiloride in ONa-ASWb 1 mM amiloride in 5 r&f Na-ASW’ 100 ml4 TPA in ASW Fertilized oocytes Control in 5 mM Na-ASWb 1 m&f amiloride in 5 mit4 Na-ASWb

Immature oocytes

7.27 7.26 7.27 7.27 7.26 7.27 7.28

(n=6, SD=0.03) (n=5, SD=0.02) (n=4, SD=0.04) (n=9, SD=0.03) (n=4, SD=0.04) (n=8, SD=0.03) (n= 10, SD=0.03)

7.40 (n=6, SD=0.03) 7.41 (n=5, SD=0.03)

Mature oocytes

7.45 (n=6, SD=0.02) 7.46 (n=5, SD=0.03) 7.54 7.49 7.54 7.47

(n=5, (n=5, (n=6, (n=7,

SD=O.Ol) SD=O.OS) SD=0.02) SD=0.03)

7.60 (n=5, SD=0.04) 7.62 (n=S, SD=0.04)

’ Maturing oocytes were fertilized 40 min after 1-MeAde addition. In maturing oocytes, treatments were initiated at the time of GVBD (unfertilized oocytes) or 3 min after fertilization. Intracellular pH was measured from 30 to 120 min after starting each treatment (immature oocytes) or after formation of the female pronucleus (mature oocytes). b Kf substituted for Nat. ’ Choline substituted for Na’.

seawater (Naf replaced by K+). These treatments did not prevent the oocytes from completing meiotic maturation. When they had reached the pronucleus stage, pHi was measured and compared to that of control oocytes matured in normal seawater without amiloride. The pHi was not found to be lower in amiloride-treated oocytes (Table 2); in fact it was even somewhat higher. Presumably this small difference (not observed in immature oocytes) is due to the penetration of the weak base amiloride in maturing oocytes. Finally, fertilized eggs (hormone-treated or not) were washed and resuspended in 5 mM Na+ seawater containing 1 mM amiloride 3 min after fertilization. Intracellular pH was measured either 30 min (non-hormone-treated oocytes) or 90 min after fertilization (hormone-treated oocytes at the female pronucleus stage). No difference in pHi was observed whether fertilized eggs were treated with amiloride or not (Table 2). The above results suggest that intracellular alkalinization during meiotic maturation and upon fertilization in starfish does not result from an enhanced activity of the transmembrane Na+/H+ exchanger. Effect of External Na+ and pHi on S6 Phosphorylation

The above results show that external Na+ does not control pHi in starfish. Nevertheless, 1-MeAde increases Na+ uptake [38] and release of acid is associated with an enhanced Na+ uptake when starfish oocytes are parthenogenetically activated in natural seawater [21]. Since S6 phosphorylation has been proposed to depend on Naf influx independent of pHi in fertilized sea urchin (30), experi-

82

Peaucellier

et al.

9467-

N N ONa AC Fig. 8. Autoradiographic analysis of S6 phosphorylation in the absence of external sodium or presence of acetate. Preloaded oocytes of M. glacialis were transferred to Na-free seawater (ONa) or seawater acidified to pH 6.5 with 10 mM Na acetate (AC) or left in normal seawater (N). They were treated (M) or not (C) with 2 10e6 M I-MeAde for 25 min and homogenized. Samples of the 10,OOOg supematants were submitted to SDS-PAGE and autoradiography. The migrations and molecular weights (X lo-‘) of marker proteins are indicated on the left of the autoradiograms. The arrow points to S6.

ments were designed to investigate the possible involvement of external Na+ in S6 phosphorylation. As shown in Fig. 8, stimulation of S6 phosphorylation was, however, observed in maturing oocytes, even in the absence of external Na+ (Na+ replaced by choline or Kf before hormone addition). In the course of this study 100 % meiotic maturation was observed, even in the absence of 1-MeAde, upon transfer of follicle cell-free oocytes to artificial seawater with choline substituted for Na+. Time required for GVBD did not differ significantly whether oocytes were transferred to Na-free seawater in the absence of I-MeAde or stimulated with lMeAde in natural seawater. This result was obtained without exception using different females captured near Roscoff, and with choline chloride from different Substitution of K+ for Na+ origins (Merck, DAB8; Sigma, 3x cristallized). induced meiotic maturation to some extent in the absence of 1-MeAde; however, percentages of “spontaneously” maturing oocytes never exceeded 20% in such case. Due to spontaneous induction of meiotic maturation in Na-free seawater, we were prevented from investigating the effect of Na+ suppression on S6 phosphorylation upon activation of prophase-arrested oocytes. At present we have no explanation as to why Na-free seawater induces maturation. Our finding that pHi does not increase before GVBD, in contrast to S6 phosphorylation, suggested that a high pHi is not a prerequisite for S6 phosphorylation in starfish. To confirm this view, oocytes were transferred before I-MeAde addition in seawater containing 10 mM acetate, pH 6.5, which decreased pHi by

S6 phosphorylation

and pHi in starfish c

oocytes

83

A

94-

302014-

NAcN

NN

Fig. 9. Autoradiographic analysis of S6 phosphorylation upon activation by ionophore in the presence of acetate or by ammonia. Preloaded oocytes of M. glacialis were transferred to seawater acidified to pH 6.5 with 10 m&f Na acetate (AC) or left in normal seawater (N). They were treated for 25 min with 5x 10m6 M ionophore A23187 (I) or 10 mM NH,Cl (A) or left untreated (C) and homogenized. Samples of the 10,OOOgsupernatants were submitted to SDS-PAGE and autoradiography. The migrations and molecular weights (X 10-3) of marker proteins are indicated on the left of the autoradiograms. Activations by ionophore or by ammonia were performed on two different batches of oocytes.

about 0.45 units. As shown in Fig. 8 (AC), stimulation of S6 phosphorylation was still observed in such oocytes at the time of GVBD. Next, we investigated the possible effect of the increase of pHi due to fertilization or ionophore A23187 on S6 phosphorylation in prophase-arrested oocytes. First, oocytes were transferred to seawater containing 10 m&f acetate, pH 6.5, and activated: stimulation of S6 phosphorylation was still observed (Fig. 91). Next, oocytes were transferred to seawater containing 10 mM NH&l, which increases pHi even more than does fertilization or activation. Surprisingly, S6 phosphorylation was not observed in such oocytes (Fig. 9A), although NH&l significantly stimulated protein synthesis (Table 1).

DISCUSSION pHi during Oocyte Maturation

As shown previously [17, 281 and confirmed in the present work, pHi does not change until GVBD following hormonal stimulation of prophase-arrested starfish oocytes. Moreover, both external application of weak bases and intracellular microinjection of alkaline buffers have been shown to inhibit hormone-induced meiotic maturation in a dose-dependent manner [39, 281. Considered together, these findings rule out the possibility that an elevated pHi might be required to release oocytes from prophase block. This study even found GVBD to occur

84

Peaucellier

et al.

without delay following hormonal stimulation when pHi was decreased by 0.45 units due to the external application of acetate, as already reported [17]. Meiotic maturation is, however, a complex sequence of events, which includes not only GVBD but also the emission of two polar bodies and the formation of the female pronucleus. In the present work, pHi was about 0.2 units higher after completion of meiotic maturation than before hormone addition. The main increase in pHi occurs before the first polar body emission, in contrast to Xenopus where pHi starts rising before GVBD (10). This was observed using starfish collected either near Banyuls in the Mediterranea or near Roscoff in the Channel, although pHi differs by about 0.2-0.3 units in fully grown immature oocytes of both populations. External addition of acetate suppressed polar body emission in 1-MeAde-stimulated oocytes of both origins, whereas pHi was constantly lower than that in control oocytes, but still increased at the normal time after GVBD. Thus, as for early mitotic cleavages in sea urchin embryos [40], a sufficiently alkaline pHi seems to be required for meiotic cleavage to occur in starfish oocytes. In Xenopus oocytes, elevation of pHi during progesterone-induced meiotic maturation is associated with proton efflux (41) and is suppressed, or strongly reduced, in Na-free medium as well as upon addition of the Na+/H+ blocker amiloride [42]. In contrast, no significant acid release occurs in starfish during meiotic maturation [21, 173. Moreover, there is no decline in the elevation of pHi during meiotic maturation in amiloride-treated oocytes, even in Na-free artificial seawater. Clearly the increase in pHi in maturing oocytes cannot result from an enhanced activity of a transmembrane Na+/H+ exchanger. Its mechanism remains unknown. As shown previously [21] oocytes of the starfish Marthasterias glacialis and Asterias rubens release acid upon fertilization or ionophore addition, although to a lesser extent than sea urchin eggs. A small acid release upon fertilization was also observed by Johnson and Epel [17] in Pisaster ochraceus, although the authors could not decide whether it originated from oocytes or sperm. In the present work we show that acid efflux is correlated with intracellular alkalinization in both prophase-arrested and maturing oocytes of the starfish M. gEacialis. Although Na+ influx normally occurs simultaneously with H+ efflux in M. glacialis, other research has shown that acid release is not coupled to Na+ influx, since it also occurs in Na-free seawater. Our results agree with this view, since intracellular alkalinization upon fertilization was found not to differ in the presence or absence of amiloride. In Xenopus also, the increase in pHi following fertilization has been shown to be independent of extracellular Na+ [43]. In sea urchins, oocytes resume meiotic maturation in the ovary, arresting after formation of the female pronucleus. The cell cycle can be turned on either by external application of NH&l [44] or by fertilization, which also increases pHi. We found that external application of NH&l to mature starfish oocytes prevents them from arresting after formation of the female pronucleus, and turns subsequent chromosome cycling on, without cleavage, as has been observed in sea urchin eggs. Recently the question whether NH&l acts in sea urchin eggs only by

S6 phosphorylation TABLE

Treatment

Increases S6 phosphorylation

I-MeAde

Yes

Sperm A23187 NH3 Acetate

Yes Yes No Does not inhibit (I-MeAde, sperm, A23187) Does not inhibit ( I-MeAde)

OnaSW and/or amiloride

and pHi in starfish

85

3

Effect on protein synthesis

++

oocytes

Effect on PH,

+ + + ?

None before GVBD (while S6 phosphorylation and protein synthesis are increasing) Increases Increases Increases Decreases

?

No effect

Effect on polar body emission

Inhibits Inhibits No effect

increasing pHi has been challenged [45]. Therefore we investigated the effect of microinjecting the impermeant alkaline buffer Hepes, which increased pHi by about 0.4 units. Repeated chromosomes cycles were observed after pronucleus formation when unfertilized oocytes were injected with the alkaline buffer at the normal time of fertilization. These results give additional support to the view that the increase in pHi following fertilization plays a key role in the regulation of chromosome cycling. S6 Phosphorylation during Oocyte Maturation

In Xenopus, progesterone stimulation of prophase-arrested oocytes has been demonstrated to trigger [46] or to stimulate [31] phosphorylation of the 40 S ribosomal protein S6. Stimulation of phosphorylation was reported to begin well before GVBD [47]. The present work showed (Table 3) as well that incorporation of 32P into ribosomal proteins dramatically increases following hormonal stimulation of prophase-arrested oocytes. As in Xenopus, stimulation was found to begin well before GVBD. The main labeled ribosomal protein is a 3 1,OOO-Da component of the 40 S subunit, also identifiable as S6. Labeling of S6 was found not to decrease after formation of the female pronucleus in unfertilized starfish oocytes. This is of interest since S6 has been shown to be already phosphorylated before fertilization in fully mature oocytes of several sea urchin species, including Lytechinus pictus, Strongylocentrotus purpuratus [48] and Echinus esculentus 1301.

Although the cell cycle remains arrested and sperm chromatin does not decondense under such conditions, fertilization and ionophore A23187 were found to stimulate phosphorylation of S6 in prophase-arrested starfish oocytes. The ribosomal protein was phosphorylated to a lesser extent than following 1-MeAde stimulation, but very specifically. In fact several experiments determined it to be the only protein in oocytes whose labeling changed (increased) following fertilization.

86

Peaucellier

et al.

There are only a few reports that ribosomal proteins other than S6 are phosphorylated in response to activation stimuli in eucaryotic cells. This includes S2 in HeLa cells infected with vaccinia virus [49] and several unidentified ribosomal proteins in mouse L cells infected with vesicular stomatites virus [50]. In starfish oocytes, however, a 47,000-Da protein, apparently not present in 100,OOOg supernatants, was found to be appreciably labeled in ribosomal fractions prepared from hormone-stimulated starfish oocytes in addition to S6. This protein was not labeled before stimulation. As S6, the 47,000-Da phosphoprotein was quantitatively extracted from ribosomal pellets with acetic acid. It was resolved in NEPHGE bidimensional gels as an acidic protein (pZ ca 4.5). Thus it appears to be the largest and the most acidic component of the small ribosomal subunit. It closely corresponds to a protein identified as Sl in rat and Xenopus (51), to use the general nomenclature of MacConkey et al. [52]. The 47,000-Da ribosomal protein is largely phosphorylated on threonine residues, in contrast to S6, which is phosphorylated only on serine residues. This is of interest because we showed elsewhere that changes in threonine phosphorylation are better correlated with the cell cycle than those in serine phosphorylation (20). In numerous systems, an increased rate of protein synthesis appears to be correlated with an increased phosphorylation of S6, in response to natural or artificial activation stimuli [2, 4, 31, 46, 47, 53-551. It has been shown, however, that S6 phosphorylation is not stimulated in Arbacia when unfertilized eggs are treated with ammonia, although this weak base stimulates protein synthesis [30]. In the same way, priming Xenopus oocytes with PMSG increases the rate of protein synthesis without changing the basal level of S6 phosphorylation [55]. Finally, a dramatic increase in S6 phosphorylation is seen, but total protein synthesis is not stimulated, when injection of exogenous mRNA into stage 4 Xenopus oocytes is followed by an injection of MPF [56]. In M. glacialis, we found that incorporation of amino acids into proteins increased to some extent following fertilization, following activation by ionophore A23187, and, slightly less but significantly, following ammonia addition (Table 3). Similar results have been obtained for another starfish species [17]. On the other hand, a major acceleration in incorporation of amino acids into proteins occurs during meiotic maturation, beginning before [15, 171 or around the time of GVBD [16], depending on the species. Increased phosphorylation of S6 is not necessary for stimulation of protein synthesis in starfish, since we found that ammonia does not increase S6 phosphorylation, as also reported for Arbacia eggs [30]. Nevertheless, we observed that hormonal stimulation, fertilization, or activation by ionophore induced both events. Our finding that in several experiments S6 was even found to be the only protein whose labeling was increased following activation by sperm or ionophore in prophase-arrested oocytes, makes it implausible that simultaneous stimulation of S6 phosphorylation and protein synthesis in such oocytes is merely a coincidence. Considering the number of investigations appearing in the literature designed to elucidate the relationships between S6 phosphorylation and protein synthesis, there may be as many different purposes for synthesis (e.g., relative to growth) and relying upon more or less unrelated

S6 phosphorylation

and pHi in starfish

oocytes

87

control processes. Hence we favor the hypothesis that S6 phosphorylation still belongs to one of these control processes. The possibility cannot be ruled out, for instance, that S6 phosphorylation specifically exerts an effect on selectivity of message recruitment [57, 581. In several systems, increased pHi in response to physiological cell-activating stimuli has been shown to be temporally correlated with increased phosphorylation of S6 [5-7, 11, 12, 33,591. The hypothesis that increased pHi is necessary for ribosomal S6 phosphorylation has, however, been seriously questioned, at least in both Xenopus oocytes [55] and sea urchin eggs [30]. In starfish oocytes, we found that stimulation of S6 phosphorylation is not coupled to intracellular alkalinization (Table 3). Fertilization or parthenogenetic activation by ionophore A23187, which increases pHi, specifically stimulated phosphorylation of S6. However, S6 phosphorylation still occurred when oocytes were previously treated with acetate, which changed pHi in the opposite way and to a greater extent. Moreover, ammonia increased pHi but did not cause S6 phosphorylation in starfish oocytes, as also observed in sea urchin eggs. In addition, stimulation of S6 phosphorylation in I-MeAde-stimulated starfish oocytes is not a consequence of alkalinization, since it already occurs before GVBD, whereas pHi begins to increase only later, before the first polar body emission. Finally, acidification by acetate does not prevent S6 phosphorylation in response to 1-MeAde. Clearly, stimulation of S6 phosphorylation following fertilization, activation, or ionophore stimulation is not coupled to alkalinization, nor does it appear to be coupled with Na+ influx, at least following hormonal stimulation. We thank Madame A. Cueff for her skillful technical help.

REFERENCES 1. 2. 3. 4. 5.

Epel, D. (1978) Cut-r. Top. Dev. Biol. 12, 185. Ballinger, D., and Hunt, T. (1981) Dev. Biol. 87, 277. Moolenaar, W., Tsien, R., Van Der Saag, P., and De Laat, S. (1983) Nature (London) 304, 645. Thomas, G., Martin-Perez, J., Siegmann, M., and Otto, A. M. (1982) Cell 30, 235. Pouyssegur, J., Chambard, J., Franchi, A., Paris, S., and Van Obberghen-Schilling, E. (1982) Proc.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Natl.

Acad.

Sci. USA

79, 3935.

L’Allemain, G., Franchi, A., Crague, E., and Pouyssegur, J. (1984) J. Biol. Chem. 259, 4313. L’Allemain, G., Paus, S., and Pouyssegur, J. (1984) J. Biol. Chem. 259, 5809. Kofoid, E. C., Knauber, D. C., and Allende, J. E. (1979) Dev. Biol. 72, 374. Robinson, K. R. (1979) Proc. Natl. Acad. Sci. USA 76, 837. Lee, S., and Steinhardt, R. (1980) Dev. Biol. 85, 358. Wassermann, W. J., and Houle, J. G. (1984) Dev. Bio/. 101, 436. Houle, J. G., and Wasserman, W. J. (1983) Dev. Biol. 97, 302. Masui, Y., and Clarke, H. J. (1979) Int. Rev. Cyto/. 57, 185 Kanatani, H., and Nagahama, Y. (1980) Biomed. Res. 1, 273. Houk, M. S., and Epel, D. (1974) Dev. Biol. 40, 298. Guenier, P., and Doree, M. (1975) Dev. Biol. 47, 341. Johnson, C. H., and Epel, D. (1982) Dev. Biol. 92, 461. Guenier, P., Moreau, M., and Doree, M. (1977) Mol. Cell. Endocrinol. 7, 137. Doree, M., Peaucellier, G., and Picard, A. (1983) Dev. Biol. 99, 489. Capony, J. P., Picard, A., Peaucellier, G., Labbe, J. C., and Doree, M. (1986) Dev. Biol. 117, 1. Peaucellier, G., and Doree, M. (1981) Dev. Growth D$f. 23, 287.

88 22. 23. 24. 25. 26. 27. 28. 29. 29. 30. 30. 31. 31. 32. 33. 34. 35. 36. 37. 38.

39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.

Peaucellier

et al.

Doree, M., and Guerrier, P. (1975) Exp. Cell Res. 91, 296. Takeshima, K., and Nakano, E. (1982) Cell Differ. 11, 229. Laemmli, U. K. (1970) Nature (London) 227, 680. Kaltschmidt, E., and Wittman, H. (1970) Anal. Biochem. 36, 401. O’Farell, P. Z., Goodman, H. M., and O’Farell, P. H. (1977) Cell 12, 1133. Cooper, J. A., Sefton, B. M., and Hunter, T. (1983) in Methods in Enzymology (Corbin, J., and Hardman, J., Eds.), Vol. 99, p. 387, Academic Press, New York. Picard, A., and Doree, M. (1983) Dev. Biol. 97, 184. Amman, D., Lanter, F., Steiner, R. A., Schulthess, P., Shito, Y., and Simon, W. (1981) Anal. Chem. 53, 2267. Picard, A., Peaucellier, G., Le Bouffant, F., Le Peuch, C., and Doree, M. (1985) Dev. Biol. 10.5, 311. Schuetz, A. W. (1975) .I. Cell Biol. 66, 86. Ballinger, D. G., Bray, S. J., and Hunt, T. (1984) Dev. Biol. 101, 192. Johnson, J. D., Epel, D., and Paul, M. (1976) Nature (London) 262, 661. Kruppa, J., Darmer, D., Kalthoff, H., and Richter, D. (1983) Eur. J. Biochem. 129, 537. Shen, S. S., and Steinhardt, R. A. (1979) Nature (London) 282, 87. Pouyssegur, J., Franchi, A., L’Alleman, G., and Paris, S. (1984) J. Cell. Biochem. 8A, 260. Moore, R. (1981) Biophys. J. 33, 203. Benos, D. (1982) Amer. J. Physiol. 242, C131. Swann, K., and Whitaker, M. (1985) Nature (London) 314, 274. Moolenaar, W. H., Tertoolen, L. G. J., and De Laat, S. W. (1984) Nature (London) 312, 371. Doree, M. (1981) Exp. Cell Res. 131, 115. Doree, M., Sano, K., and Kanatani, H. (1982) Dev. Biol. 90, 13. Dube, F., Schmidt, T., Johnson, C. H., and Epel, D. (1985) Cell 40, 657. Cicirelli, M. F., Robinson, K. R., and Smith, L. D. (1983) Dev. Biol. 100, 133. Stith, B. J., and Maller, J. L. (1984) Deu. Biol. 102, 79. Webb, D. J., and Nuccitelli, R. (1981) J. Cell Biol. 91, 562. Mazia, D. (1974) Proc. Natl. Acad. Sci. USA 71, 690. Dube, F., and Epel, D. (1986) Exp. Cell Res. 162, 191. Hanocq-Quertier. J., and Baltus, E. (1981) Eur. J. Biochem. 120, 351. Nielsen, P. J., Thomas, G., and Maller, J. L. (1982) Proc. Natl. Acad. Sci. USA 79, 2937. Ward, G. E., Vacquier, V. D., and Michel, S. (1983) Dev. Biol. 95, 360. Kaerlein, M., and Horak, I. (1976) Nature (London) 259, 150. Marvaldi, J., and Lucas-Lenard, J. (1977) Biochemistry 16, 4320. Hugle, B., Hazan, R., Scheer, U., and Franke, W. W. (1985) J. Cell Biol. 100, 873. MacConkey, E. H., Bielka, H., Gordon, J., Lastick, S. M., Lin, A., Ogata, K., Reboud, J. P., Traugh, J. A., Traut, R. R., Warner, J. R., Welfle, H., and Wool, I. G. (1979) Mol. Gen. Genet. 169, 1. Thomas, G., Siegmann, M., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 3952. Kalthoff, P., Darmer, D., Towbin, H., Gordon, J., Amons, R., Moller, W., and Richter, D. (1982) Eur. J. Biochem. 122, 439. Stith, B. J., and Maller, J. L. (1985) Dev. Biol. 107, 460. Taylor, M., Robinson, K. R., and Smith, L. D. (1985) J. Embryol. Exp. Morphol. 89 (Suppl.), 35. Duncan, R., and McConkey, E. H. (1982) Eur. .I. Biochem. 123, 535. Duncan, R., and McConkey, E. H. (1982) Eur. J. Biochem. 123, 539. Chambard, J. C., and Pouyssegur, J. (1986) Exp. Cell Res. 164, 282.

Received February 24, 1987 Revised version received June 26, 1987

Printed

in Sweden