Activation of Multiple Protein Kinases during the Burst in Protein

Vol. 263, No. 4, Issue of February 5, pp. 2009-2019,1988 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Q 1988 by The American Society for Biochemistry and Molecular Biology, Inc.

Activation of Multiple Protein Kinases duringthe Burst in Protein Phosphorylation That Precedes the First Meiotic CellDivision in Xenopus Oocytes* (Received for publication, April 21, 1987)

Michael F. Cicirelli, Steven L. PelechS, and EdwinG . Krebs From the Howard Hughes Medical Institute, University of Washington School of Medicine, Seattle, Washington 98195

A number ofdifferent protein and peptide substrates (2), human D98/AH2 cells (7), yeast cells (6), and Ch‘lnese were used to identify and characterize stimulatedki- hamster ovary cells (3), in addition to eggs and maturing nase activities in Xenopus oocyte extracts prepared oocytes from a wide variety of species (1, 4, 5, 8). MPFduring themajor burst in protein phosphorylation that containing extracts trigger a precocious meiotic cell division precedes meiotic cell division. While total CAMP-de- upon microinjection into Xenopus oocytes resulting in an pendent protein kinase activity in the cytosol was not easily observable breakdown of the germinal vesicle (nuclear stimulated, this kinase was the majorkinase phospho- envelope). Assayed in this manner, MPF activity has been rylating a number of the substrates and consequently had to be inhibited to prevent its masking CAMP-in- found to cycle during mitotic (1,4) and meiotic cell cycles( l ) , dependent protein kinase activities. Sizable stimula- reaching peak activity at each “phase. The purification of tions of kinase activities were then observed in ex- MPF has proven to be extremely difficult due to losses of tracts from progesterone-treated oocytes as compared activity during purification and the reliance on a bioassay to to controls when following the substrateswere utilized detect activity (9-12). The biochemical nature of MPF is Leu-Arg-Arg-Ala-Ser-Leu-Gly(Kemptide) (&fold); unknown. the synthetic peptide, Arg-Arg-Leu-Ser-Ser-Leu-Arg- Many studies imply that MPF activity and protein phosAla, the sequence of whichis based on that of a phos- phorylation are closely related. First, the appearance of MPF phorylation site in ribosomal protein S6 (%fold); ri- activity and a major burst in total protein phosphorylation bosomal protein56 (%fold); histoneH1 (5-fold); skel- (2-3-fold) during progesterone-induced Xenopus oocyte matetal muscleglycogensynthase(%fold);andmyelin uration (13) are approximately coincident. Furthermore, in a basic protein (30-fold). When these substrates were number of systems, changes in MPF activity during the cell used to assay extracts fractionatedon DEAE-Sephacel, cycle are accompanied by parallel changes in protein phosat least three distinct peaks of stimulated kinase activity were detected, eluting at 0.12,0.17, and 0.21 M phorylation (14-18). Second, microinjection of MPFinto NaCl.Thesepeaks were tentatively designated M- oocytes that have not been exposed to hormone leads to an immediate burst in protein phosphorylation, a response to phaseActivated Binases(s), MAK-H, MAK-S, and MAK-M, respectively. Using histone H1 as a selective progesterone that normally begins only after several hours probefor MAK-H and S6 peptide or Kemptide as (13). In addition, agents such as ,&glycerophosphate and yprobesforMAR-S,the kinase activities comprising thio-ATP, which would tend to maintain proteins in a phosthese peaks were found to cycle with the meiotic cell phorylated state, enhance the stability of MPF activity (10, cycle. 19). Finally, the finding that MPFactivity undergoes autoamplification upon injection into recipient oocytes (20) is consistent with the supposition that MPFis a protein kinase that Meiotic as well as mitotic cell division appears to be con- is activated by autophosphorylation. This possibility is further trolled by a factor, or factors, known as maturation or M- supported by observations that certain protein kinase activiphase promoting factor(s) (MPF)’ (1-6). MPF activity has ties correlate with MPF activity inextracts (7) and also been found in crude extracts obtained from mitotic HeLa cells copurify with MPF activity (10). Regardless of whether or not MPF is a kinase, the imme* The costs of publication of this article were defrayed in part by diate burst in protein phosphorylation following MPF actithe payment of page charges. This article must therefore be hereby vation represents a convenient starting point for tracing the marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 biochemical steps back to MPFactivation and elucidation of solely to indicate this fact. 4 Recipient of a Medical Research Council of Canada 1967 centen- the biochemical nature of MPF. While preliminary reports have indicated that therise in protein phosphorylation is due, nial fellowship. The abbreviations used are: MPF, maturation (or “phase) pro- at least in part, to the stimulationof protein kinase activity moting factor(s); PKI, heat-stable inhibitorof CAMP-dependentpro- (21, 22), questions as to the number of kinases involved, the tein kinase; PKI peptide, the peptide (Thr-Thr-Tyr-Ala-Asp-PheIle-Ala-Ser-Gly-Arg-Thr-Gly-Arg-Arg-Asn-Ala-Ile-His-Asp) which time courses of their activations, the mechanisms of activacontains the active inhibitory region of PKI; Kemptide, the peptide tion, their substrate specificities, and their relationships to MPF have remained largely unexplored. To help answer these (Leu-Arg-Arg-Ala-Ser-Leu-Gly) derived from the phosphorylation site on porcine liver pyruvate kinase; S6 peptide, the peptide (Arg- questions, it is necessary to identify a set of substrates that Arg-Leu-Ser-Ser-Leu-Arg-Ala) which is patterned after a phospho- can be used to study the kinases involved. Numerous proteins rylation site sequence in ribosomal protein S6; GVBD, germinal vesicle breakdown; MOPS, 4-morpholinepropanesulfonic acid; exhibit increased phosphorylation acid just prior to germinal EGTA, [ethylenebis(oxyethylenenitrilo)]tetraacetic acid SDS, so- vesicle (nuclear membrane) breakdown (GVBD) (23-25), but dium dodecyl sulfate. only 40 S ribosomal subunit protein S6 (26-28), nucleoplas-

2009

2010

ActivationMeiosis of during Kinases Protein

ent in both buffers) was the main component responsible for the preservation of stimulated kinase activity in the extracts (data not shown). Protein concentrations of the extracts were determined by the method of Bradford (43) with bovine serum albumin as the standard. Zon Exchange Chromatography-Control and GVBD extracts (33 mg of protein), prepared as described above, were diluted (1/10) with 10% buffer A (0.68 mmho;11mM salt) tolower the ionic strength for column binding. Eachextract was then loaded onto one of two identical 7-ml DEAE-Sephacel (Pharmacia LKB Biotechnology Inc.) columns (1X 9 cm) that had been equilibrated in 10%buffer A with flow rates adjusted to precisely 20 ml/cm2/h. The columns were then EXPERIMENTALPROCEDURES washed with 10% buffer A (21 ml each) and eluted with a common Materials-Peptides were synthesized by Dr. Patrick Chou and linear salt gradient of 0.011-0.5 M NaCl in 10% buffer A. For each Henry Zebroski of the Howard Hughes Medical Institute Chemical column, 160 fractions (1ml each) were collected, aliquoted, and stored Synthesis Facility at theUniversity of Washington and were purified at -70 "C until assay. No stimulated kinase activities were detected by Maria Harrylock of this laboratory. The peptides include the in the flow-through fractions. Protein and Peptide Kinuse Assays-All kinase assays were perfollowing: Arg-Arg-Leu-Ser-Ser-Leu-Arg-Ala (S6 peptide), which is based on a phosphorylation site sequence in ribosomal protein S6 formed a t 30 "C.The components of the assay mixture were made up (31) and has been used previously in this laboratory for studying an in assay buffer (buffer A). As indicated in the figure legends, the S6 kinase that is activated by treatment of cells with growth factors assay tubes contained the protein or peptide substrate, oocyte extract, and 10-15 PM (final concentration) [Y-~'P]ATP(approximately 15 (32, 33); Thr-Thr-Tyr-Ala-Asp-Phe-Ile-Ala-Ser-Gly-Arg-Thr-GlyArg-Arg-Asn-Ala-Ile-His-Asp (PKI peptide), which corresponds to dpm/fmol) to startthe reactions. In a number of cases, cAMP or PKI the active inhibitory region of PKI and inhibits CAMP-dependent peptide was also included in the reactions. To terminate thereactions, protein kinase with approximately the same Ki as the parent molecule aliquots were removed at specified time points, spotted on Whatman (34, 35); peptides Lys-Ala-Thr-Gly-Ala-Ala-Thr-Pro-Lys, Lys-Thr- P81 phosphocellulose paper (2.5 X 2.5 cm), washed 5 times (5 min each) with 600 ml of 180 mM phosphoric acid, and then once with Pro-Val-Lys, and Val-Ala-Lys-Ser-Pro-Lys, whichwere patterned after growth-associated histone kinase phosphorylation sites on his- 600mlof 95% ethanol. The papers were then placed into plastic tone H1 (36); Arg-Arg-Arg-Glu-Glu-Glu-Thr-Glu-Glu-Glu, regarded minivials, 5 ml of Aquamix (Westchem) were added, and the radioas a specific substrate for casein kinase I1 (37); and Ala-Ala-Ala-Ser- activity in the vials was counted. All values reported have been Phe-Lys-Ala-Lys-Lys-amide, patterned after the histone kinase I1 corrected for counts obtained in the absence of added extract. In phosphorylation site in histoneH1 (38). The classical CAMP-depend- some cases, phosphorylatable impurities in the substrate preparations ent protein kinase substrate, Kemptide (Leu-Arg-Arg-Ala-Ser-Leu-made it necessary to terminate the reactions with 5 X sodium dodecyl Gly), was purchased from Sigma along with histone fraction 111-S sulfate (SDS) sample buffer. The samples were then placed in a (enriched in histone H l ) , phosvitin, and casein. Whole histones and boiling water bath for 5 min and electrophoresed on SDS polyacrylhistone H4 were purchased from Boehringer Mannheim. Pure calf amide gels (44). The gels were stained with Coomassie Blue, dried, thymus mixed histones H1, H2a, H2b, and H3 were a gift from Dr. and placed under Kodak X-AR film with Hi plus Cronex intensifying Bassam Wakim (Department of Biochemistry, University of Wash- screens to visualize the radiolabeled protein bands. The bands of ington). Rabbit skeletal muscleglycogen synthase was purified by interest were excised and counted. Curtis Diltz and bovine brain myelin basic protein was prepared by Dr. Nicholas Tonks, both from the Department of Biochemistry, RESULTS University of Washington. Acetyl-coA carboxylase was purified by Levels of CAMP-dependent Protein Kinase in Control and Dr. James Sommercorn; smooth muscle myosin light chains by Dr. Arthur Edelman, Dr. Peter Kennelly, and Dr. Nicholas Tonks; gly- GVBD Extracts-There is considerable evidence that CAMPcogen phosphorylase b by Floyd Kennedy; and RIIsubunit by Barbara dependent protein kinase has an important role in triggering Flug, all from this laboratory. ATP citrate lyase was a gift from Dr. the maturation process caused by exposure of Xenopus ooPaul Srere (Veterans Administration Medical Center, Dallas, TX). Rat liver 40 S ribosomes were a gift from Dr. Robert Traut (Depart- cytes to progesterone. Microinjection of the regulatory subunit (R) of CAMP-dependent protein kinase or PKI hasbeen ment of Biological Chemistry, University of California, Davis). Oocyte Culture-Adult Xenopus heuis females were purchased from shown toinduce maturation in oocytes, while early injections Nasco (Fort Atkinson, WI) andmaintained as described (39). Ovaries of the catalytic subunit (C) inhibit maturation (45). There is were surgically removed followinghypothermic anesthesia and placed comparatively little evidence, however, that CAMP-dependent in OR-2 (40). Stag- VI (41) oocytes were then manually dissected protein kinase is involved in the later eventsof oocyte matufrom their follicles using watchmaker's forceps and cultured in OR2. In experiments involving treatment with progesterone, groups of ration including the burst in protein phosphorylation that oocytes received 1 pg/ml progesterone in OR-2 from a stock solution occurs shortly before GVBD (46). Nonetheless, it was considof 5 mg/ml progesterone in ethanol, while control groups received an ered important to examine CAMP-dependent protein kinase equivalent concentration of ethanol carrier.The extent of maturation activity carefully in the present study, if for no other reason was determined by scoring the percentage of oocytes that displayed a than to be certain that it was not interfering with the assay distinct white spot in the animal hemisphere, indicative of GVBD. In of other protein kinases. these experiments, GVBD (50%)occurred between 5 and 8 h following Two kindsof assays were used to examineCAMP-dependent progesterone addition. protein kinase activity oocyte in cytosolic extracts. Oneof the Preparation of Extracts-Oocytes were transferred to a 2-ml Potassays measured the extent of activation of the enzyme by ter-Elvehjem tissue grinder on ice, ice-cold buffer A (50 mM @glycerophosphate pH 7.3, 7 mM NaF, 0.3 mM EDTA, 15 mM MgCL, CAMP, i.e. an "activity ratio" (47, 48). Regretfully, these 2 mM dithiothreitol) was added, and theoocytes were rapidly homog- assays proved to be impossible to carry out,since the ratio of enized (5 strokes) and centrifuged for 15-20 min at 100,000 X g in a activity without and with added cAMP was sensitive to extract Beckman Airfuge at 5 "C. The clear supernatant layer was quickly dilution. This line of inquiry was not pursued further, since removed,divided into aliquots, and frozen at -70 "C until assay. of the kinase Extracts from progesterone-treated oocytes, prepared a t a time when procedures designedto preserve the activity state 50-100% of the oocytes underwent GVBD, are referred to as GVBD (47,49,50) remain questionable (51). The second type of ertracts. Corresponding control extracts were prepared at the same assay measured total CAMP-dependent protein kinase levels. time. In some cases, buffer B (60 mM @-glycerophosphate,30 mM p - Forthis,thepeptidesubstrate,Kemptide, was employed nitrophenyl phosphate, 25 mM MOPS, pH 7.2,15 mM EGTA, 15 mM together with PKI peptide. Tots1 CAMP-dependent protein MgCl,, 1 mM dithiothreitol, 0.1 mM Na3V04) was used instead of buffer A. Essentially the same kinetics of phosphorylation were found kinase activity was taken as activity toward Kemptide in the upon assay of kinase activity in extracts prepared with buffers A or presence of optimal concentrationsof cAMP minus the activB. As reported by Martin-Perez et al. (42), @-glycerophosphate(pres- ity in the presence of saturating concentrations of PKI pep-

min (29), and,tentatively,nuclearlamins (30) have been specifically identified. In the present study, we have employed a diverse set of exogenous substrates toreveal the existenceof several different protein kinases, the activities of which are enhanced in progesterone-treated oocytes as compared to nonstimulated controls. The time course of activation of some of these kinases approximates the cyclical changein MPF activity that occurs during oocyte maturation.

2011

Activation of Protein Kinasesduring Meiosis

“4L / A 80

E c t o

6 - 0 1

u

\\

ioo

0

cAMP concentrotton ( n M )

FIG. 1. Optimum cAMPconcentration for CAMP-dependent protein kinase stimulation in Xenopus oocyte extracts. The rates of phosphorylation of Kemptide (0.33 mM) by 21.4 pg of control (closed symbols) and GVBD (open symbols) oocyte extract protein were determined with increasing concentrations of cAMP in the absence (squares) or presence (circles) of 4.4 p~ PKI peptide. The reactions (60 pi, total volume) were started with the addition of 10 pM [Y-~*P]ATP, allowed to proceed for 10 min (linear incorporation), and were terminated using the P81 paperspotting technique as described under “Experimental Procedures.”

tide. As will be presented in thenext section, which describes experiments that utilize a different albeit similar peptide substrate for CAMP-dependent protein kinase, maximal inhibition of CAMP-dependent protein kinase by PKI peptide was achieved at concentrations greater than 1pM. Fig. 1shows the results of an experiment in which reaction rates for the phosphorylation of Kemptide by control and GVBD extracts were determined in the presence of increasing concentrations of cAMP in the presence and absence of a high concentration of PKI peptide. In the absence of PKI peptide, CAMP-dependentprotein kinase was fully stimulated in therange from 0.3 to 40 PM CAMP.cAMP at concentrations greater than 40 PM caused inhibition of CAMP-dependent protein kinase activity in both control and GVBD extracts, and CAMP-independent protein kinase activity (with PKI peptide) was inhibited inthe GVBD extracts. The same curves were obtained with or without the addition of theophylline to inhibit potentialphosphodiesterase activity (data notshown). With PKI peptide present in the assays, it can be seen that cAMP did not lead to an increase in activity in control or GVBD extracts, indicating that CAMP-dependent protein kinase activity was fully inhibited by PKI peptide. CAMPdependent protein kinase was the major kinase activity phosphorylating Kemptide in control andGVBD extracts (Fig. l), and thiswas also true for other substrates, such as S6 peptide and histone H1 (Sigma fraction 111-S). It is also evident from Fig. 1 that total CAMP-dependent protein kinase activity was the same in control and GVBD extracts. In fact, no change in total CAMP-dependentprotein kinase activity was detected in these extracts duringthe entirecourse of maturation, from progesterone addition to GVBD (data not shown). CAMP-independent Protein Kinase Actiuity in Control and GVBD Extracts as Measured Using S6 Peptide-S6 peptide contains one serine (position 4) that is an excellent site for phosphorylation by CAMP-dependent protein kinase and a second serine (position 5) that is readily phosphorylated by one or more CAMP-independent protein kinases (31-33, 52). In using this peptide as a probe for studying the latter, it is essential to block CAMP-dependent protein kinase-catalyzed phosphorylation of this substrate (32, 33). Accordingly, an experiment was carried out with control and GVBD extracts to determine the concentration of the inhibitory peptide that is required to achieve maximal inhibition of CAMP-dependent protein kinase when S6 peptide is used as a substrate(Fig. 2). With either type of extract, themaximal effect of PKI peptide was achieved at a concentration of approximately 1 p ~ Fig. . 2 also illustrates how instrumental PKI peptide can be in the

. . , . ,,,io,

PKI peptide concentrotion

1

, ,

I

‘To2

(pM)

FIG. 2. Inhibition of CAMP-dependent protein kinase activity. Control (0)and GVBD (0)extracts (21.4 pg of protein each) were assayed for their ability to phosphorylate 0.33 mM S6 peptide (RRLSSLRA) in the absence or presence of increasing concentrations of PKI peptide. The reactions were started and terminated as described in Fig. l.

3i A

0

30

60 Ti m e ( m i n )

FIG.3. Reaction time courses and dependence of reaction rates on protein concentration at high and low substrate levels. Control (0)and GVBD (0)extracts were assayed for kinase activities directed against S6 peptide (RRLSSLRA) in the presence of 0.8 p~ PKI peptide. In panels A and C, low substrate concentrations while in panels were used (0.25 mM S6 peptide, 15 pM [Y-~*P]ATP), E and D, higher substrate concentrationswere used (5 mM S6 peptide, 400 p~ [Y-~’P]ATP). Thereaction time courses (panels A and B ) were performed with 0.15 mg/ml extract protein, and the extract dilution reactions (panels C and D)were of 6-min duration. detection of stimulated CAMP-dependent protein kinases. Without PKI peptide present in the assays, there was approximately a 1.4-fold stimulation of kinase activity in GVBD extracts compared to controls. When concentrations of PKI peptide exceeding 1 p~ were added to the extracts to inhibit CAMP-dependent protein kinase, the apparent stimulation increased to 4.8-fold. Apparently CAMP-dependent protein kinase activity simply masked the stimulated CAMP-independent kinase activity(ies), since the difference between the control and GVBD phosphorylation rates remained the same (approximately 28 pmol/min/mg) when the CAMP-dependent protein kinase activity was inhibited by PKI peptide. If homogenates were assayed in the presence of PKI peptide, instead of 100,000 X g supernatants, the stimulation of S6 peptide phosphorylation seen in the GVBD extracts compared to controls was less than 2-fold, although both control and GVBD rates of S6 peptide phosphorylation were elevated considerably in homogenates compared to 100,000 x g supernatants (data not shown). Thus, in addition to using PKI

Activation of Protein Kinases during Meiosis

2012

TABLE I Phosphorylation of peptide substrates Reactions were carried out with 4.4 p~ PKI peptide, 10 pM [y3'P]ATP, and 0.33 mM substrate as described in Fig. 1. Abbreviations: CKII, casein kinase II; HKII, histone kinase 11; GAHK, growth-associated histone kinase. Control"

Substrate

GVBD"

Stimulation -fold

No substrate added 2.9 & 1.1 4.2 f 1.6 1.6 & 0.4 Kemptide (LRRASLG) 11.7 & 9.3 90.3 f 64.3 8.1 & 3.2 4.3 7.1 1.7 CKII peptide (RRREEETEEE) 70.9 f 31.8 7.8 & 1.9 S6 peptide (RRLSSLRA) 9.7 2 5.1 7.0 7.8 HKII peptide (AAASFKAKK-amide) 1.1 3.3 4.3 1.3 GAHK peptide (VAKSPK) GAHK peptide (KATGAATPK) 2.8 3.7 1.3 7.1 1.9 GAHK peptide (KTPVK) 3.8 "The reaction rates are expressed as pmol 32Pincorporated per min per mg of extract protein (+S.D. where shown, n 2 5).

peptide to eliminate CAMP-dependentprotein kinase activity, it also appears helpful to remove other kinase activities by centrifugation in order to observe large relative increases in kinase activity. Utilization of the S6 peptide for studying differences in the activities of CAMP-independent protein kinases in control and GVBD extracts was further facilitated by working at substrate concentrations that accentuated rate differences. This is illustrated in the experiments of Fig. 3A. With low concentrations of S6 peptide and ATP, 0.25 mM and 15 FM, respectively, there was a striking difference in the activities of the two extracts. This was less apparent when the concentrations of peptide and ATP were increased to 5 mM and 400 pM, respectively, approximately double the Km(app) values estimated for the stimulated kinase activity(ies) in these extracts. In this case, the stimulation in GVBD extracts was reduced to less than 2-fold (Fig. 3B). As would be anticipated, the kinase reactions were linear for a longer period of time at the high substrate concentrations (Fig. 3B as compared to Fig.3A). With low or high substrate levels, reaction rates were proportional to the concentration of enzymeover a readily workable range (Fig. 3, C and D). Kinase Activities Toward VariousSubstrates in Control and GVBD Oocyte Extracts-To help identify the kinase or kinases manifesting enhanced activity in GVBD extracts compared to control extracts, anumber of synthetic peptides were examined as potential substrates.As shown in Table I, Kemptide and S6 peptide consistently showed a large stimulation with GVBD extracts as compared to controls (on average 8fold). Over the course of our investigation with these peptides, experiments were performed on extracts prepared from the oocytes of several different Xenopus females. Thus, where possible, the phosphorylation rates are expressed as an average valueplus or minus the standarddeviation to indicate the amount of biological variability that can be expected when preparing extracts from the oocytes of different females. The peptide substrates for casein kinase I1 (Arg-Arg-Arg-Glu-GluGlu-Thr-Glu-Glu-Glu) (37) and histone kinase I1 (Ala-AlaAla-Ser-Phe-Lys-Ala-Lys-Lys-amide) (38) appeared to be phosphorylated poorly by control and GVBD extracts. The stimulation of peptide phosphorylation byGVBD extracts was not significantly greater than the stimulation with no substrate added (1.6-fold). Peptides that were patterned after growth-associated histone kinase phosphorylation sites on histone H1 (36) were also tested. Although these peptides have not been fully characterized as substratesfor the growthassociated histone kinase, the primary amino acid sequences of protein phosphorylation sites often provide sufficient determinants for kinases to phosphorylate them (31,33,38,53).

Furthermore, one of these peptides, Lys-Thr-Pro-Val-Lys, has been shown to be phosphorylated by a kinase in Xenopus oocyte extracts that is similar or identical to the growthassociated histone kinase (54). Thus, these peptides were considered potential substrates. As can be seen in Table I, none of these peptides were appreciably phosphorylated. In addition to synthetic peptides, a number of proteins were tested for their ability to serve as substrates for stimulated kinase activity(ies) in oocyte extracts. As presented in Fig. 4, preparations of the RII subunit of CAMP-dependent protein kinase, phosvitin, phosphorylase b, histones H2a, H2b, H3, and H4, and smooth musclemyosin light chains did not undergo increased phosphorylation by the GVBD extracts. However, preparations of acetyl-coA carboxylase and casein showed approximately 2-fold stimulations, while preparations of glycogen synthase, histone H1, ribosomal S6, and myelin basic protein exhibited more striking increases in phosphorylation by GVBD extracts over controls. With histones, a number of different commercial preparations were screened to find prominent, distinct, andeasily identifiable Coomassie Blue-stained bands on gels that could becut out and counted. Sigma histone 111-Sand Boehringer Mannheim whole histone preparations were found to be good sources of histone H1 and were used in subsequent assays. However, these preparations also contained lower molecular weight proteins, many of whichwere barely visiblebyCoomassieBlue staining but were highly phosphorylated and/or showed stimulated phosphorylation by the GVBD extracts (Fig. 4, histone H1 from Sigma histone 1113).None of these proteins could be clearly identified as histones H2a, HZb, H3, or H4. Therefore, pure calf thymus mixed histones (HZa, H2b, and H3, separated from H1 by perchloric acid extraction (55)) and aBoehringer Mannheim histone H4 preparation were used in which the histone protein bands (Coomassie Blue-stained) could be clearly identified and excised. With glycogen synthase and acetyl-coA carboxylase, which contained some contaminating kinase activity (Fig. 4B), the -fold stimulation could conceivably be even higher if this contaminating activity was subtracted out. However, this may not be a valid manipulation as exemplified in the similar case of ATP-citrate lyase (Fig. 4B). Here, the ATP-citrate lyase-associated kinase activity appeared to be completely inhibited by both control and GVBD extracts. The preparation of myelin basic protein that was used wassusceptible to partial degradation by the oocyte extracts. Therefore, the -fold stimulation reported included all degradation products. Ion Exchange Chromatography of Oocyte Extracts-Further information with respect to the protein kinases activated at the time of GVBD was obtained by ion exchange chromatog-

Activation of Protein KinasesMeiosis during

A 1

q. 'm

2

3

5

4

6

C G "-W .i.. I

-1;

L.

I

-rlrlJ

histone H1

histone H4

myelin basic protein

8

B 7

C G

C G

phosphorylase b

RII

H2b

s6

H3

10

11

C G

C G -

-

*

-w x

c

light choin substrate 14

l2 C G-

-

no

myosin

subunit

C G -

-0

L

glycogen synthase

ATP citrate

lyase

acetyl CaA carboxylase

hG. 4. Protein substrata! for stimulated kinase activities

in GVBD extracts. Control and GVBD extracts (0.36mg of protein/ ml) were incubated with various proteins, 4.4 PM PKI peptide, and 10 pM [r-SZP]ATPfor 10 min. The reactions were terminated by the addition of 5 X SDS sample buffer, and the reaction mixture was electrophoresed on 10-15% polyacrylamide gels and further processed aa described under "Experimental Procedures." The kft lane of each pairrepresents the control reaction (C), with the GVBD extract

2013

raphy. Control and GVBD extracts were fractionated on two identical DEAE-Sephacel columns and eluted in parallel from a common salt gradient. As shown in Fig. 5A, when the DEAESephacel fractions were assayed with Sigma histone 111-S (sourceof histone Hl), prominent peaks of stimulated activity appeared at fractions 44 and 62 in the GVBD extract profile. These peaks were partially masked by the trailing edges of much larger peaks of activity (fractions 22 and 34)when PKI peptide was not included in the assays (data not shown). The fractions were reassayed in the absence of exogenous substrates, so that the contribution of endogenous substrates in the various fractions could be examined (Fig. 5B). It is conceivable that the peak at fraction 44 from control extracts (Fig. 5A) was entirely the result of endogenous substrate phosphorylation at this fraction; thus, the -fold stimulation seen in the GVBD profile at fraction 44 mayhavebeen considerably larger than itfirst appears. Histone fraction 111-S contained a number of proteins, besides histone H1, that exhibited stimulatedphosphorylation with GVBD extracts upon SDS-polyacrylamide gel analysis (Fig. a), as did whole histones from Boehringer Mannheim (data not shown). This latterpreparation of histones was used to assay the DEAE-Sephacel fractions to ascertain whether one or both peaks of stimulated kinase activity were directed against histone H1. The phosphorylated histones were separated on SDS-polyacrylamidegels, and thehistone H1 bands were cut out andcounted. As shown in Fig. 5C, only the peak of stimulated kinase activity at fraction 44 was found to be directed against histone HI. When the other proteins from the histone preparation that exhibited stimulated phosphorylation in GVBD extracts (not identified) were excised from the gels and counted, fraction 62 was found to be the only peak of stimulated activity directed against them (data not shown). The peak at fraction 62 also represented the major stimulated phosphorylating activity directed against Kemptide'(Fig. 50), S6 peptide (Fig. 5E), ribosomal S6 (Fig. 5F), and glycogen synthase (Fig. 5G). With myelinbasic protein as substrate, in addition to fractions 44 and 62, DEAE-Sephacel fraction 76 and possibly fraction 92 from GVBD extracts were also peaks of stimulated phosphorylatingactivity (Fig. 5H). At fraction 92, endogenous substrate phosphorylation was high(Fig. 5B), so the -fold stimulation of this activity(ies) mighthave actually been greater. Minor stimulated peaks such as those at fraction 92 (Fig. 5H), fraction 30 (Fig. 5G), and fraction 120 (Fig. 5F) could conceivably showgreater stimulations if more suitable substrates for those kinase activities couldbefound. For example, when glycogensynthase (Fig. 5G) or S6 peptide (Fig. 5E)was used as theSubstrate, the peaks of stimulated phosreaction ( G ) on the right. In cases where the substrate preparation exhibited kinase activity, a reaction with substrate alone is shown (-). Substrate concentrations are given below with the average (n= 3) -fold stimulation (fS.D.) of phosphorylation rates in parentheses. To determine -fold stimulation, endogenous substrate phosphorylations in the corresponding gel positions from controls (control and GVBD extracts) with no exogenous substrate were subtracted. The arrowheads beside each gel lane indicate the positions of the various substrates. 1, 0.33 mg/ml Sigma histone fraction 111-S (histone H1 band, 5.3 f 1.3); 2,0.33 mgiml histone H4 (not quantitated); 3, 0.33 mg/ml myelin basic protein (30.4 f 15.1); 0.17 4, mg/ml phosvitin (1.2 f 0.3);5, 0.33 mg/ml histones H2a, H2b, H3 (1.0 f 0.3, 1.4 f 1.0,1.5 f 0.2); 6, 1.2 p M ribosomal protein S6 (40s ribosomes) (7.5 f 2.3); 7, 0.17 mg/ml phosphorylase b (1.1 f 0.5); 8, 0.17 mg/ml subunit of CAMP-dependent protein kinase (1.0 f 0.3);9, 0.17 mg/ ml casein (2.0 f 0.5);10, 0.17 mg/ml smooth muscle myosin light chain (not quantitated); 11, no exogenous substrate added, 12, 0.17 mg/ml glycogen synthase (3.2 f 0.3);13,O.l mg/ml ATP citrate lyase (not quantitated); 14,0.17mg/ml acetyl-coA carboxylase (2.1 f 0.9).

2014

Activation of Protein Kinasesduring Meiosis 0.5 0.4

FIG.5. Ion exchange chromatography of control and GVBD oocyte

0.3

extracts. Control (0)and GVBD (0) extracts were subjected to ion exchange chromatography using DEAE-Sephacel as described under "Experimental Procedures." Even-numbered fractions (30pl aliquots) were assayed in the presence ,. of 10 pM [y3'P]ATP, 4.4 p M PKI peptide, and the following substrates: in panel A, 1.9 mg/ml histone 111-S (Sigma); in panel B , no exogenous substrate was added; in panel C, 0.7 mg/ml 0.6 whole histones (Boehringer Mannheim); in panel D, 0.33 mM Kemptide; in panel 0.4 E, 0.33 mM S6 peptide; in panel F, 1.5 2 Aneounits/ml (0.1 1M) 40 S ribosomal $ 0.2 subunits; in panel G, 0.15 mg/ml glycogen synthase; and in panel H,0.33 mM 1 o myelin basic protein. In panels A, B, D. E E , and H,the reactions were terminated 2 E at 60 min by spotting onP81 paper, while p, in panels C, F, and G the reactions were $ l6 terminated at 60 min by addition of 5 X SDS sample buffer, subjected to electro- $ 12 phoresis on 10-15% polyacrylamide gels, H1, 8 and the substrate bands (histone ribosomal protein S6, and glycogen synthase, respectively) were excised and counted as described under "Experimental Procedures." Note differences in the scales used for the ordinates in the various panels. Substrate concentrations are not necessarily saturating, as exemplified in panel F,where it was necessary to use lower than I