Characterization of protein kinase C in early ... - Semantic Scholar

461

Development 110, 461-470 (1990) Printed in Great Britain © The Company of Biologists limited 1990

Characterization of protein kinase C in early Xenopus embryogenesis

ARIE P. OTTE, IJSBRAND M. KRAMER, MAURICE MANNESSE, CARO LAMBRECHTS and ANTONY J. DURSTON Hubrecht Laboratory, Netherlands Institute for Developmental Biology, Uppsalalaan 8, 3584 CT UTRECHT, The Netherlands

Summary

Recently, we presented evidence that protein kinase C (PKC) is involved in mediating the endogenous signals that induce competent Xenopus ectoderm to differentiate to neural tissue. We report here that PKC is already strongly activated in neural-induced ectoderm from midgastrula embryos and that this activation runs parallel with an increase in the level of inositol phosphates. We further identify several proteins that are phosphorylated, both in natural neural-induced ectoderm and in TPA-treated ectoderm, suggesting that they are phosphorylated through the PKC route. We found no major changes in PKC activity among different

pregastrula stages, including the unfertilized egg. However, PKC isolated from animal, ectodermal cells is highly sensitive to Ca2+ and can be activated by low concentrations, (6-25 /iM) of arachidonic acid, while PKC isolated from vegetal, endodermal cells is more insensitive to Ca 2+ and cannot be activated by arachidonic acid. These results suggest that different PKC isozymes are present in animal and vegetal cells.

Introduction

(Swann and Whitaker, 1985) and Xenopus eggs (Bement and Capco, 1989). In addition, TPA caused a cortical contraction when applied to unfertilized Rana pipiens eggs, suggesting that it causes Ca 2+ release (Zimmerman, 1985). In Xenopus oocytes, meiotic maturation is induced by TPA (Stith and Mailer, 1987; Smith, 1989) and facilitated by injection of protein kinase C (Kamata and Kung, 1990). These studies indicate a role for PKC and phosphoinositol phosphate turnover in the maturation of oocytes and in fertilization. Another clue about the roles of PKC and the inositol phosphates in early embryogenesis comes from experiments with Li + . This ion is thought to block the processing of inositol phosphates, resulting in depletion of the phosphatidylinositol phosphate pools (reviewed in Berridge et al. 1989). Short treatments of 32- to 64cell Xenopus embryos with Li + dorsalizes these embryos (Kao et al. 1986). Injection of Li + into a ventral blastomere also leads to dorsalization. Importantly, these teratogenic effects can be annihilated when myo-inositol is coinjected with Li + into a ventral blastomere (Busa and Gimlich, 1989; for an excellent review on this subject see Berridge et al. 1989). Busa and Gimlich also found that activation of PKC by TPA overrules the effect of Li + , indicating a role for PKC in establishing the dorsoventral polarity of the embryo. Another clue about the role of PKC in early embryogenesis comes from our own work. We found that PKC is translocated from the cytosol to the

Protein kinase C (PKC) plays an important role in the mechanisms by which external signals are transmitted to the interior of cells. Several growth factors and hormones that bind to, and activate, receptors on the cell surface, stimulate inositol phospholipid turnover via phospholipase C activation, thereby producing the second messengers diacylglycerol and inositol phosphates, of which diacylglycerol activates PKC (Nishizuka, 1984, 1986; Berridge, 1987; Berridge and Irvine, 1989). Several growth-factors and hormones that stimulate inositol phospholipid turnover, induce translocation of PKC from the cytosol to the membrane (Niedel and Blackshear, 1986). PKC also serves as the receptor for phorbol esters, a class of tumour promoters (Castagna et al. 1982). One of them, 12-0-tetradecanoyl phorbol-13-acetate (TPA), activates PKC efficiently and causes PKC to translocate from the cytosol to the membrane. There are clues suggesting that the PKC/inositol phosphate pathway plays an essential role in early embryogenesis. Fertilization increases the polyphosphoinositide content of sea urchin eggs (Turner et al. 1984) and induces changes in phosphatidylinositol turnover (Dworkin and Dworkin-Rastl, 1989). Iontophoresis of inositol trisphosphate activates Xenopus (Busa et al. 1985) and sea urchin eggs (Whitaker and Irvine, 1984). Also, TPA activates mouse oocytes (Cuthbertson and Cobbold, 1985), sea urchin eggs

Key words: protein kinase C, signal transduction, neural induction, Xenopus.

462

A. P. Otte and others

membrane in neural-induced ectoderm, but not in noninduced ectoderm. We also found that TPA causes translocation of Xenopus PKC and induces competent ectoderm to differentiate to neural tissue (Otte et al. 1988). In this paper, we show that there are several proteins that become strongly phosphorylated in neural-induced ectoderm, either after artificial PKC activation or during natural neural induction. An increased level of inositol phosphates was also detected in neural-induced ectoderm. We show further that PKC, which is present in animal, ectodermal cells, is highly sensitive to Ca 2+ and can be activated by low concentrations of arachidonic acid. PKC, which is present in the vegetal, presumptive endodermal cells, is more independent of Ca 2+ for its activation and cannot be activated by arachidonic acid. These observations suggest that there are different PKC isozymes present in early embryogenesis. Materials and methods Culture conditions Embryos were obtained by natural fertilization via standard procedures. Embryonic stages were determined according to Nieuwkoop and Faber (1967). Embryos were dejellied in 2 % cysteine-hydrochloride (Sigma) (pH8.0). Explants were cultured in 100% Flickinger solution (NaCl, 58 mM; KC1, IITIM;

NaHCO 3 ,

0.24 HIM;

Na 2 HPO 4 ,

IIUM;

KH 2 PO 4)

0.2mM, CaCl2) 0.5HIM; MgSO4, IITIM) (pH7.5). Protein kinase C assay Embryos or explants were washed and homogenized in icecold lysis buffer (20 mM Tris pH7.5, 2mM EDTA, 0.5 mM EGTA, 5mM /3-mercaptoethanol, 2mM phenylmethylsulphonyl fluoride and 0.01 mgml" 1 leupeptin). The homogenate was centrifuged at lOOOOg for 15s, the pellet was homogenized again, centrifuged and the supernatants were combined. The pellet, containing yolk platelets and organelles, was discarded, as it contained no PKC activity. The supernatant was spun at 100000g for l h . The 100000g supernatant, considered to be the cytosolic fraction, was removed. The pellet was used for the in vitro phosphorylation experiments described in Fig. 3. To isolate membrane-bound PKC, the pellet was resuspended and shaken regularly for 45 min in lysis buffer containing 1 % Nonidet-40 (Sigma). The mixture was centrifuged for 15 min at 10 000 £ and the pellet discarded. The supernatant was considered to be the paniculate or membrane fraction. When total (soluble and paniculate) PKC was isolated from tissue, the tissue was homogenized in 1 ml ice-cold lysis buffer containing 1 % Nonidet-40 and shaken regularly for 45 min. After that, the homogenate was spun at lOOOOg for lOmin. The supernatant was loaded onto a DEAE column. Partial purification of PKC with DEAE-cellulose was performed exactly as described (Otte et al. 1988). The PKC assay, in which histone III S was used as a substrate, was performed exactly as described previously (Otte et al. 1988). The PKC assay in which a nonapeptide, derived from the EGF receptor (Val-Arg-LysArg-Thr-Leu-Arg-Arg-Leu-NH2) (Auspep, South Melbourne, Australia) was used as substrate, was essentially performed as described by House et al. (1987) and House and Kemp (1987). 25/il of DEAE PKC eluate was added to 60^1 of reaction mix containing (as the final concentration) 20 mM

Tris pH7.5, 7.5 mM MgAc, lmgml l leupeptin, 10 ^M EGF receptor peptide, 0.25 mg ml" 1 BSA, 0.2 mM ATP, l.SxlO^ctsmin"1 [32P]>^ATP (New England Nuclear) and either imM EGTA (background) or lmin Ca 2+ , 8/igml"1 phosphatidylserine (PS) (Spinal cord, Lipid Products), 0.8/igml"1 diacylglycerol (Diolein, Sigma) (assay). Lipids were stored under nitrogen at — 20°C in chloroform/methanol 3:1. They were dried using a flow of nitrogen, suspended in 20 mM Tris pH7.5 and sonicated for 3 min to obtain liposomes. Different Ca2+-concentrations were obtained by using Ca2+-EGTAbuffers (Bartfai, 1979). The reactions were started by adding 25 /A PKC eluate to 60 /A reaction mix. The incorporation of 32PO4 at 30°C was linear with time up to 60min. After 15 min, the reactions were stopped by adding 25/A 25 % TCA. After 20min on ice, the samples were spun (4°C) at lOOOOg for lOmin and 90 fA of the supernatant was then spotted onto a Whatman P-81 phosphocellulosefilterand this was immediately transferred to 400 ml 75 mM phosphoric acid solution in which a magnetic stirrer rotated slowly. After changing the phosphoric acid solution 4 times (10-15 min rotation each), the filters were dried and counted. We found identical results using either histone HI S or the EGF-receptor nonapeptide as substrate. Only the absolute value of the specific activities differed (compare with Otte et al. 1988), due to different numbers of phosphorylation sites in histone III S and the nonapeptide. Using the nonapeptide as a substrate has the advantage that lower backgrounds can be obtained, leading to a better signal to noise ratio. Substrate phosphorylation and gel electrophoresis Explants were incubated for 3h in 5 ml Holtfreter solution (59mM NaCl, 0.7 mM KC1, 0.9 mM CaCl2) containing 80/*Ci [32P]orthophosphate. After incubation, the explants were washed and homogenized in sample buffer consisting of 8 M urea, 2 % Nonidet P40 and 0.1 % ampholytes pH3-10. Twodimensional electrophoresis was carried out according to O'Farrell (1975), using isoelectric focusing (2% ampholytes pH3-10, Sigma) in the first dimension and SDS-polyacrylamide gel electrophoresis (10 % polyacrylamide) in the second dimension. Measurement of inositolphosphates and phosphatidylinositolphosphates Determination of total inositol phosphates was performed essentially according to methods described by Tilly et al. (1987) and by Dworkin and Dworkin-Rastl (1989). Determination of the phosphatidylinositol phosphates was performed essentially according to methods described by Lacal et al. (1987) and by Dworkin and Dworkin-Rastl (1989). Fertilized eggs were microinjected with 40 nl of 20 mM Tris (pH7.2) containing 2>yC\ of [3H]myoinositol (Amersham, 3.5 Ci mmol"1). The eggs were cultured to stage 10 and entire ectoderm explants were dissected, either with or without a small piece (