Sequential Activation of MAP Kinase Activator, MAP Kinases, and S6 ...

6 downloads 0 Views 4MB Size Report
Jun 19, 1991 - Tashiro-Hashimoto, Y., Tobe, K., Koshio, O., Izumi, T., Takaku, F., Ak- anuma, Y., and Kasuga, M. (1989) J. Biol. Chem. 264,6879-6885. 12.
THEJOURNAL OF BIOLOGICAL CHEMISTRY Q 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 267, No. 29, Issue of October 15, pp. 21089-21097 1992 Printed in d..S.A.

Sequential Activation of MAP Kinase Activator, MAP Kinases, andS6 Peptide Kinasein Intact Rat Liver following Insulin Injection* (Received for publication, November 15, 1991)

Kazuyuki Tobe$, Takashi Kadowaki$$T,Kenta Hara$ 11, Yukiko Gotoh**, Hidetaka Kosako**, Satoshi Matsuda**, Hiroyuki Tamernoto$§, Kohjiro Ueki$, YasuoAkanumat, Eisuke Nishida**, and Yoshio YazakiS From the $Third Department of Internal Medicine, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, the Slnstitute for Diabetes Care and Research, Asahi Life Foundation, 1-6-1 Marunomhi, Chiyoda-ku, Tokyo 100, and the **DeDartment of. BioDhvsics and Biochemistry, - . Faculty - of. Science, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan ”

recombinant Xenopus MAP kinase. From these data, we demonstrate three tiers of a cascade composed of the MAP kinase activator, MAP kinases, and an S6 peptide kinase activity inrat liver under physiological conditions in the intactanimal.

An insulin-stimulated phosphorylation cascade was examined in rat liver after insulin injection via a portal vein by the use of immune complexkinase assays specific to the mitogen-activated protein (MAP) kinase and S6 kinase I1homologue (rsk) kinase. We have prepared anantibody against thepeptide consisting of a carboxyl-terminal portion of the extracellularsignalregulated kinase 1 (aC92), one of the MAP kinases, and an antibody against the peptide consisting of the carboxyl terminus of the mouse 56 kinase I1 homologue (arsk(m)C).In aC92 immune complex assay, maximal activation of rat liver MAP kinases (approximately 4.3-fold) were observed 4.5 min after insulin injection. We also observed an insulin-stimulated MAP kinase activity (approximately 3-fold) in liver extracts from insulin-treated rat in fractions eluted from phenylSepharose with 30-50% ethylene glycol. Kinase assay in myelin basic protein (MBP)-containing gel after sodium dodecylsulfate-polyacrylamide gel electrophoresis followed bydenaturation with6 M guanidine HCl, and renaturation revealed that insulin injection stimulated the kinase activityof the 42- and 44-kDa proteins, which corresponded to the two distinct MAP kinases. In arsk(m)Cimmune complexassay, maximal stimulation (approximately 5-fold) of the S6 peptide (Arg-Arg-Leu-Ser-Ser-Leu-Arg-Ala) kinaseactivity was observed 7.5 min after insulin injection. In addition, MAP kinases purified from insulin-treated rat liver were able to activate S6 peptide kinase activity in vitro in arsk(m)C immunoprecipitates from untreated rat liver, accompanied by the appearance of several phosphorylated bands including a major band at 88 kDa. We also examined whether insulin injection stimulates the MAP kinase activator (Ahn, N. G., Seger, R., Bratlien, R. L., Diltz, C. D., Tonks, N. K., and Krebs, E. G . (1991)J. Biol. Chem. 266, 4220-4227) in rat liver. Using recombinant Xenopus MAP kinase, fractions of Q-Sepharose eluted early in theNaCl gradient were found to have MAP kinase activator activity accompaniedby the phosphorylation of 42-kDa

The first stepof insulin action is the interaction of insulin with a receptor bound to the plasma membrane,which actifl subunit of vates the protein tyrosine kinase intrinsic to the the receptor (1, 2). Activation of the tyrosine kinase of the insulin receptor is thought tobe a necessary step for insulin’s signal transmission (3-6). The activated tyrosine kinase of the receptor subsequently phosphorylatescellular substrates on tyrosine residues (7-13). Although the roles of tyrosine phosphorylated substrates have not been clarified, it is reasonable to assume that the function of many of these substrates are intimately associated with the large network of protein serine/threonine kinases present in eukaryotic cells (14-16). In fact, insulin binding to thereceptor also leads to rapid stimulation of protein phosphorylation of an array of proteins on serine/threonine residues in target cells within minutes. To account for this serine/threonine phosphorylation inresponse to insulin, it has been proposedthat a cascade of several serine/threonine protein kinases may be activated by insulin treatment of cells. In a simplecase, we would expect that a tyrosine kinase would activate one or more intermediary serine/threonine kinases, which would then activate other serinelthreonine-specific enzymes. A protein kinase cascade was first demonstrated by Krebs and his co-workers (17), who reported that rabbit muscle phosphorylase kinase is phosphorylated and activated by the CAMP-dependent protein kinase. Thephosphorylation of phosphorylase kinase leads to a decrease in the K , for phosphorylase b (18).Since then,several other examples of protein kinase cascadeshavebeen demonstrated. Evidencefor an insulin-activatedserine/threonine kinasecascade was first shown by the activation of an S6 kinase I1 purified from Xenopus oocytes by an insulin-stimulated microtubule-associated protein 2 (MAP2)’ kinases or mitogen-activated pro-

*This work has been supported by a Grant 190831 from the Juvenile Diabetes Foundation International (to T. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ll To whom correspondence and reprint requestsshould be addressed. Tel.: 81-3-3815-5411 (ext. 8296); Fax: 81-3-3815-2087. I( Present address: Second Department of Internal Medicine, Kobe University School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobeshi 650, Japan.

The abbreviations used are: MAPS, microtubule-associated protein-2; MAP kinases, mitogen-activated protein kinases; ISPK, insulin-sensitive protein kinases; PPlG, glycogen-associated form of protein phosphatase-1; G subunit, glycogen-targeting subunit of protein phosphatase-1; DTT, dithiothreitol; MBP, myelin basic protein; TPA, 12-O-tetradecanoylphorbol13-acetate; CHO, Chinese hamster ovary; HIR, human insulin receptor; rsk, ribosomal S6 kinase; SDSPAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; MAPK(Xe), recombinant Xenopus MAP kinase; HEPES, 4-(2-hydroxyethy1)-l-piperazineethanesulfonic acid; ERK1, extracellularsignal-regulated kinase 1.

21089



21090

Phosphorylation CascadeRat in

Liver following Insulin Injection

tein (MAP)kinase, partially purified from differentiated 3T3- sham Corp. (Buckingham, United Kingdom). Gel reagents and AffiL1 adipocytes (19). Activation occurred concomitantly with Gel 15 were obtained from Bio-Rad. Actrapid monocomponent insulin from Novo (Copenhagen). Porcine insulin was a gift from Eli threonine phosphorylation of the S6 kinase 11. Homologues was Lilly Co. Pansorbin, okadaic acid, sodium orthovanadate, and dithiof S6 kinase 11, termed pp9Wk, have also been shown to exist othreitol (DTT) were from Seikagaku Kogyo (Tokyo, Japan). Myelin in chicken, mouse, and human (20),and tobe phosphorylated basic protein (MBP), 12-0-tetradecanoylphorbol-13-acetate(TPA), and activated by MAP kinases (21-28). MAP kinases have protein kinase inhibitor, and cycloheximidewere from Sigma. Qbeen shown to be activated incommon by a varietyof growth Sepharose and phenyl-Sepharose were from Pharmacia LKB Biofactors and phorbol esters (29-35). An upstream component technology Inc. (Uppsala, Sweden). Phosphatase 2A catalytic subwere kindly provided by Drs. Takeda and Usui of Hiroshima of this cascade, MAP kinase activator, was recently reported units University and were purified as described previously (47). S6 peptide to stimulate MAP kinase activity in vitro in the presence of (RRLSSLRA) was synthetized by M. Shiomi (Fujiya Co., Ltd., HaMg2+ and ATP, concomitantly with phosphate incorporation tano, Japan). Chinese hamster ovary (CHO) cells overexpressing the into both tyrosine and threonineresidues of MAP kinase (36). human insulin receptor (CHO-HIR)' were prepared as described Another group also reported this MAP kinase activator as previously (48, 53). Antibodies-Antibodies (aY91 and aC92) whichrecognizedtwo "MAP kinase kinases" (37), however it remains to be clarified distinct kinases were as described previously (35). aY91 was whether this activity is itself a kinase that catalyses the preparedMAP against the peptide containing the triple tyrosine residues phosphorylation of MAP kinases (38). of ERKl and(uC92 was prepared against the peptide of the carboxylLiver is one of the insulin's major target tissues, and to terminal region of ERKl (33,35). Thepeptide of PTPQLKPIESSIstudy the mechanism by which insulin stimulates glycogen LAQRR, which corresponded to residues 699-715 (peptide rsk(m)C) synthesis and inhibits glycogenolysis in this organ should be of the amino acid sequence deduced from the cDNAof mouse S6 physiologically relevant. Recently, Dent andCohen (39) iden- kinase I1 homologues (rskmo-l)(20), was also synthesized byM. Shiomi. The peptide was coupled to keyhole lympet hemocyanin tified an insulin-sensitive protein kinase (ISPK) inrabbit (Calbiochem, CA) as previously described. The peptide-keyhole lymskeletal muscle, which activates the glycogen associated form pet hemocyanin conjugate (1.5mg)was emulsified with complete of protein phosphatase-1 (PPlG) by phosphorylating its gly- Freund's adjuvants and injected simultaneously into multiple sites of cogen-targeting subunit (G subunit) on a particular serine female New Zealand White rabbits. Rabbit serum was collected after residue, termed site 1. This increases the rate at which P P l G several booster injections. The serum was affinity purified by passing dephosphorylates (activates) glycogen synthase and dephos- over an affinity column with the peptide coupled to Affi-Gel 15 (Bio-Rad) as previously described (35). phorylates (inactivates) phosphorylase kinase. This appears column Animals--Rats were anesthetized by the administration of 10-15 to represent the mechanism by which insulin stimulates gly- mg of pentobarbital sodium 10-15 min before experiment. The portal cogen synthesis and inhibits glycogenolysis in skeletalmuscle, vein was exposed, and insulin (3 mg insulin/kg body weight,Actrapid since the phosphorylation of site-1 on theglycogen-targeting monocomponent) was injected via the portal vein (10). After injection, subunit has also been shown to increase in vivo in response the liver of each animal was removed and quickly homogenized by a to insulin (39). The ISPK hasbeen shown to be immunolog- Polytron in 100 mlof buffer A (25 mM Tris-HC1, pH 7.4, 1 mM sodium orthovanadate, 10 mM sodium pyrophosphate, 10 mM sodium ically very similar to S6 kinase I1 from Xenopus eggs (40). fluoride, 1 mM molybdic acid, 1 mM EGTA, 1 mM EDTA, 1 mM These two kinases share various substrates including the long phenylmethylsulfonyl fluoride, and 10 nM okadaic acid) at 4 "C. The S6 peptide (the carboxyl-terminal 32 residues of ribosomal homogenate was centrifuged at 100,000 X g for 30 min at 4 "C. The protein S6), the shortS6 peptide (Arg-Arg-Leu-Ser-Ser-Leu-supernatants were subjected to immunoprecipitation or applied to Arg-Ala), Kemptide, and thus the ISPKseems to be a mam- phenyl-Sepharose or Q-Sepharose column equilibrated with buffer A. Kinase Assay-The supernatants of the liver homogenate were malian homologue of Xenopus S6 kinase I1 (40). with each antibody (aC92 or arsk(m)C) for 1 h at 4 "C. One of the goals of the research in insulin action is to incubated Pansorbin was then added and themixture was incubated for 45 min. investigate how insulin works in intact live animals under The immunoprecipitates were washed twice with buffer A, and resusphysiological conditions. We recently developed a method to pended in 40 pl of kinase buffer (25 mM Tris-HC1, pH 7.4, 10 mM detect phosphotyrosine-containing proteins, the insulin MgCl', 1 mM DTT, 40 p~ ATP, 2 pCi of [y3'P]ATP, 2 p~ protein receptor /3 subunit and a 170-kDa protein, inrat liver following kinase inhibitor peptide, 0.5 mM EGTA) and substrates (25 pg MBP insulin injection of a portal vein (< 2 min) (10). Others have or50 pgof S6 peptide). After 15 min at 25 "C, the reaction was by adding 10 pl of stopping solution containing 0.6% HC1, 1 reported several serine/threonine kinases in rat liver or in stopped mM ATP, 1%bovine serum albumin. After centrifugation, aliquot8 rabbit skeletal muscle of intact live animals which are acti- of the supernatant (15 p l ) were spotted on 1.5 X 1.5-cm squares of vated at 2-25 min following insulin stimulation (39-46). The P81 paper (Whatman), washed five times for a t least 5 min each in molecular link between tyrosine phosphorylation atthe 0.5% phosphoric acid, washed in acetone, dried, and counted by plasma membrane region and serine/threonine phosphoryla- Cerenkov counting (35). Chromatography-All chromatography was performed in a cold tion in the cytosol remains to be clarified.This study was at 4 "C. A phenyl-Sepharose (Pharmacia) column (1.5 X 5 cm) undertaken to examine the physiological relevance of an room was equilibrated with buffer A. Liver extracts from insulin-treated or insulin-stimulated phosphorylation cascade in rat liver, one untreated rats were applied to the column. The column was washed of the major insulin target tissues, under conditions most with 5 ml of buffer A and eluted with a 20-ml gradient of increasing germane to a normal insulin response. Using an antibody concentrations of ethylene glycol (0-60%). The aliquot fractions (5 against a peptide from the carboxyl terminus portion of one pl) were then incubated with 40 pl of kinase buffer (25 mM Tris-HC1, of the MAP kinases (or the extracellular signal-regulated pH 7.4,lO mM MgCl, 1 mM DTT, 40 p~ ATP, 2 pCi of [Y-~'P]ATP, p~ protein kinase inhibitor peptide, 0.5 mM EGTA) and 25pg kinase 1 (ERKl)) (aC92) (33, 35), and an antibody against 2MBP. After 15 min at 25 "C, the reaction was stopped by adding 10 the peptide from the mouse S6 kinase homologue (arsk(m)C) pl of stopping solution containing 0.6% HCl, 1 mM ATP, 1%bovine (20), we demonstrate that insulin injection stimulated the serum albumin. Aliquots of the supernatants were spotted on P81 MAP kinase activator, the MAP kinases, and the S6 peptide paper, washed, dried, and counted by Cerenkov counting (35). Kinase Assays in MBP-containing Polyacrylamide Gels-Liver exkinases in rat liver, and we were able to reconstitute three tiers of a cascade of activation accompanied by phosphoryla- tracts from untreated or insulin-treated rats for 4.5 min were applied to phenyl-Sepharose column and eluted with a 20-ml gradient of tion in vitro. increasing concentrations of ethylene glycol (0-60%). Aliquots of the MATERIALS ANDMETHODS

[y3'P]ATP (6000 Ci/mmol) was obtained from Du Pont-New England Nuclear, and '261-proteinA (30 mCi/mg) was from Amer-

eluted fractions were subjected to kinase assay in MBP-containing 'The sequence of the human insulin receptor used here corresponds to thatof the receptor of Ebina et al. (55).

Phosphorylation Cascade in Rat Liver

following Insulin Injection

21091

gels, after equalizing the amounts of MAP kinases by Western blot- min before experiment. Rats were injected intraperitoneally with 30 ting with aY91. Aliquots of the fractions were electrophoresed onan mg/kg body weight sodiumorthovanadate and 10 mg/kg body weight SDS-polyacrylamide gel containing 0.5 mg/ml MBP (35, 49). SDS cycloheximide dissolved in phosphate-buffered saline (pH 7.4).7.5 was removed fromthe gel by washing the gel with twochanges of 100 min later animals were sacrificed, and the liverswereremoved, ml each of 20% 2-propanol in50 mM Tris-HC1(pH 8.0) for 1 h, and quickly homogenized, and ultracentrifuged. The supernatants were then 250mlof50 mM Tris-HC1 (pH 8.0) containing 5 mM 2- subjected to immunoprecipitation witharsk(m)Cfollowed by immune mercaptoethanol for1 h at room temperature. The enzyme inthe gel complex kinase assay toward S6 peptide. was denaturated by treating the gel first with two changes of 100 ml of 6 M guanidine HClat room temperature for 1h and then renatured RESULTS withfivechangesof250mleachof 50 mM Tris-HC1 (pH 8.0) Two Distinct MAP Kinases-Using two specific antibodies, containing 0.04% Triton X-100 and 50 mM 2-mercaptoethanol, After of two immunologically renaturation, the gel was preincubated at 25 "C for 1 h with 5 ml of we have recently shown the presence 40 mM HEPES (pH 8.0) containing 2 mM DTT, and 10 mM MgCl,. similar but distinct MAP kinases with molecular masses of Phosphorylation of MBP was carried out by incubating the gel at 42 and 44 kDa in CHO-HIRcells and suggested that the 4425 "C for 1 h with 5 mlof40mM HEPES (pH 8.0), 0.5 mM EGTA, kDa MAP kinase is more highly related to the ERKl gene 10 mM MgC12,2 p~ protein kinase inhibitor, 40 p~ ATP, and 25 pCi of [Y-~~PIATP. After incubation, the gel was washed with a 5% (w/ product (35). The supernatants of liver extracts were subaY91 in the v) trichloroacetic acid solution containing 1%sodium pyrophosphate jected t o immunoprecipitation with aC92 and of 0.1%SDS followed by Western blotting until the radioactivity of the solution became negligible. The washed presence or absence gel was dried and then subjected to autoradiography. with aY91. In the presence of 0.1%-SDS, aY91 was able to Purification of MAP Kinases-Liver extracts from insulin-treated immunoprecipitate the 42- and 44-kDa proteins (Fig. lA, b ) . or untreated rats were applied to a Q-Sepharose (Pharmacia) column In the absence of SDS, aY91 failed to recognize these two (1.6 X 5 cm) equilibrated with bufferA, washed with 30 ml of buffer of SDS, aC92 A and eluted with a 50-ml increasing gradient of0-400mM NaCl. proteins (Fig. l A , a). In contrast, in the absence recognized the 44-kDa protein (Fig. lA, c); only after denaFractions were subjected to kinase assay toward MBP or Western blotting with aY91. Fractions containing MBP kinase activityor two turation with 0.1% SDS, aC92 was able torecognize the 42proteins with molecular masses of 42 and 44 kDa detected by aY91 kDa protein in addition to the 44-kDa protein (Fig. lA, d). were concentrated to 2 mlby Centriprep 10 and were chromato- These data are consistent with the results in CHO-HIR cells graphed over a phenyl-Sepharose column (1.6 X 5 cm). Aftera 20-ml (35), indicating that there are two immunologically similar wash with buffer A, the column was eluted with a 20-ml gradient of ethylene glycol (0-60%). Fractions containing MBP kinase activity but distinctMAP kinases withmolecular masses of 42 and 44 or two proteins with molecular massesof 42 and 44 kDa detected by kDa in ratliver. aY91 were concentrated to 2 ml by Centriprep 10 and were chromatInsulinStimulatesMAPKinase Activity in Rat Liver ographedover a SepharoseS-200 (Pharmacia) column (1.6 X 100 Shortly after Insulin Injection-We have recently shownthat cm). The fractions containing MAP kinases were collected and ap- aC92 specifically immunoprecipitates insulin- or TPA-stimplied to phenyl-Sepharose column (1.6 X 5 cm). After washing with 20mlofbuffer A, the columnwas elutedwith20-mlwashes of ulated 44-kDa phosphoprotein with MAP kinase activity and increasing concentrations of ethylene glycol (0-60%). The peak frac- that phosphotransferase activity toward MBP or MAP2 can tions werecollected and used as partially purifiedMAP kinases. be monitored through the useof immune complex assay (35). Protein phosphatase 2A-treated MAP kinases were prepared as fol- Using this system,we analyzed the insulin-stimulated kinase lows. The buffer in these fractions was changed into phosphatase activity toward MBP in rat liver from insulin-injected rats. buffercontaining 20 mM Tris-HC1 (pH 7.4), 2 mM EGTA, 2 mM Rats were anesthetized, and livers from insulin-treated or MgC12,0.1% 2-mercaptoethanol, 0.1% Triton X-100, and 1 mM phenylmethylsulfonyl fluoride, using PDlO (Pharmacia). This frac- untreated rats were removed, homogenized, and ultracentrition was incubated with 6.83 pg of protein phosphatase 2A catalytic fuged. The supernatantswere subjected to immunoprecipitasubunits for 30 min. By adding 2 p~ okadaic acid, the reaction was tion with aC92 followed by kinase assay toward MBP. We stopped (35). This fraction was changed into the buffer containing observed a maximal increase in kinase activity (about 4.320 mM Tris-HC1 (pH 7.4), 0.1 mM EGTA, and 1 mM phenylmethyl- fold) 4.5 minafterinsulin injection. The kinaseactivity sulfonyl fluoride and stored at -70 'C. decreased thereafter(Fig. 1B). We measured the MAP kinase Activation of S6 PeptideKinaseActivity by MAPKinasearsk(m)C or control preimmune serum immunoprecipitatesof liver activity in other systems. The supernatants of liver extracts extracts from untreated rats were incubated with or without 5 p1 of from rats were applied t o phenyl-Sepharose and eluted with purified MAP kinases from insulin-treated or untreated rat liver in a gradient of increasing concentration of ethylene glycol (0the presence of 25 mM Tris-HC1, pH 7.4, 10 mMMgC12, 1 mM DTT, 60%). Each fraction was subjected to kinase assay toward 40 p~ ATP, 1 pCiof [T-~'P]ATP,2 p~ protein kinase inhibitor peptide, 0.5 mM EGTA. After 15-min incubation at 25 "C, 500 pl of MBP. A maximally stimulatedMBP kinase activity (about3buffer A was added. Each fraction was divided into two aliquots (A fold) wasobserved at 4.5-7.5 min (Fig. IC). The insulinand B). In A, the mixture was centrifuged and washed with buffer A stimulated MBP kinase activity was decreased thereafter for and subjected to SDS-PAGE followed by autoradiography. In B, the up to 15 min. When each fraction was subjected to Western mixture was centrifuged and washed with buffer A and subjected to blotting with aY91, we observed twomajor bands molecular at S6 peptide kinase assay. mass of 42 and 44 kDa in fractions with MBP kinase activity MAPKinaseActivator-RecombinantXenopusMAPkinase (MAPK(Xe))was expressed in Escherichiacoli and purified to >90% (data not shown). To determine the proteinsresponsible for of liver exhomogeneity by a method as previously described (49). Purification the increased kinase activity, the supernatants of Xenopus M-phase MAP kinase (42 kDa) and isolation of its cDNA tracts from insulin-treated or untreated rats were applied to 51). The supernatant ofliver havebeendescribedpreviously(50, phenyl-Sepharose and eluted with increasing concentrations extracts from untreated or insulin-treated rats for 3 min was applied of ethylene glycol (0-60%). After equalizing the amounts of to a Q-Sepharose column, washed with5 ml of buffer A, and eluted MAP kinase proteins in fractions we compared by Western with a 30-ml gradient of 0-200 mM NaC1, and 1.4-ml fractions were collected. The aliquot fractions were incubated in kinase buffer con- blotting with aY91, the aliquots of each fraction were subtaining MgZ+and [yS2P]ATPin the presence or absence of about 1 jected to kinase assay in gels containing MBP (35, 52). We pg of MAPK(Xe)for 20 minat 25 "C.Then, 25 pg of MBP was added, observed two major insulin-stimulated kinase activities miand the incubation continued for another 12 min, after which ["PI grated at molecular mass of 42 and 44 kDa (Fig. 1D). phosphorylated MBPwas adsorbed to P81 paper, washed, dried,and Insulin Injection Stimulates S6 Peptide Kinase Activity in counted by Cerenkov counting (35). Aliquots were subjected to SDSRat Liver-We have prepared a rabbit polyclonal antiserum PAGE followed by pictography by Fuji image analyzer. Vanadate and Cycloheximide Treatment-Rats were anesthetized raised against a peptide consisting of the carboxyl terminus portion of mouse S6 kinase I1 homologue (rskmo-1) (20). by the administration of10-15mg of pentobarbital sodium 10-15

Phosphorylation Cascade in Rat Liverfollowing Insulin Injection

21092

B 500 control serum

0-0

c . re92

400

300 &

m 5:

.-E L

h

f

200

N

1OE

A

2

0

4.5

1

Time (min.)

C

D

-

e--* 0 mln 4.Smln p r d 7.5min Ismin

,44kDa '42kDa

0

2

4

6

8

IO 12 lk 16 Fraction No

18 20 22

Ins a b Injection (-1 (+)

FIG.1. Insulin stimulates two distinct MAP kinases in rat liver. A, immunologically distinct MAP kinases. A liver was removed from an anesthetized rat and homogenized by Polytron in 100 ml of buffer A at 4 "C. As described under "Materials and Methods," after centrifugation, the supernatantwithout (a, c) or with (b, d ) addition of 0.1% SDS was subjected to immunoprecipitation with aY91 (a, b ) or aC92 (c, d ) . The immunoprecipitates were subjected to SDS-PAGE followed by Western blotting with aY91. B, insulin injection stimulates MAP kinase activity in aC92 immunoprecipitates in ratliver. Liver extracts from untreated (time, 0) or insulin-treated rats for the indicated periods were subjected to immunoprecipitation followed by kinase assay toward MBP. C, insulin injection stimulates kinase activity toward MBP in the phenyl-Sepharose fraction. Liver extracts from untreated or insulin-treated rats for the indicated periods were applied to phenylSepharose column, eluted with a 20-ml gradient of increasing concentrations of ethylene glycol (0-60%). The aliquot fractions (5 pl) were subjected to kinase assay toward MBP without immunoprecipitation. D,insulin stimulates 42- and 44-kDa MBP kinase activity in rat liver. Liver extracts from untreated or insulin-treated rats for 4.5 min were applied to phenyl-Sepharose column, eluted as in C. The eluted fractions were subjected to kinase assay in MBP-containing gels. Before this assay, we equalized the amount of the MAP kinases in fractions and compared them by Western blotting with aY91.

Phosphorylation Cascade in Liver Rat following Insulin Injection Affinity-purified arsk(m)C antibody (cursk(m)C) was able to immunoprecipitate the insulin- and TPA-stimulated S6 peptide kinase activity from CHO-HIR cells, suggesting the existence of members of a pp9Wk family in this cell line (Fig. 24). Using thisimmune complex assay, we analyzed whether

r r

2.000~

.-I= h

1.000 v

LL N

m

500

(-)

INS

(-)

TPA (1 00ng/rnl)

(10-7~)

500

400

n

300

w

D

r n

co

= ._

h

zE 200

N

100

d

4.5

7.5

15

Time (min.)

21093

insulin injection was able to activate S6 peptide kinase activity in rat liver. Insulin was injected via a portal vein, and the liver of each animal was removed and homogenized. After centrifugation, the supernatants were subjected to immunoprecipitation with serumcontrol or ursk(m)C by followed S6 peptidekinaseassay.We observed an increase in insulinstimulated S6 peptide kinase activity 4.5 min after insulin injection (2.2-fold). A maximal increase (5-fold) was observed 7.5 min after insulininjection. The stimulated S6 peptide kinaseactivitydecreased thereafter for up to 15 min after insulininjection(Fig. 2B). We also observed aninsulinstimulated S6 peptide kinase activity in arsk(m)C immunoprecipitates in breakthrough fractions of Q-Sepharose (data not shown). M A P Kinases from Insulin-injected Rat Liver Can Activate S6 Peptide Kinase Activity in arsk(m)C Immunoprecipitates from Untreated Rat in Vitro-To determine whether MAP S6 peptidekinaseactivity kinase is able toactivatethe immunoprecipitated by arsk(m)C, arsk(m)Cimmunoprecipitates from untreated rat liver were incubated with partially purified MAP kinasespurified from insulin-treated or untreated ratliver and were subjected to S6 peptide kinase assay and SDS-PAGE. After incubation with MAP kinases from untreated liver, arsk(m)Cimmunoprecipitates showed approximately a 3-fold increase in S6 peptide kinase activity (Fig. 3A, lane e ) and increased phosphorylation of an 88-kDa protein (Fig. 3B, lane e ) . After arsk(m)C immunoprecipitates were incubated with partially purified MAP kinases from insulin-treated liver, we observed about a 10-fold increase in S6 peptidekinaseactivity(Fig. 3A, f ) and an increasein phosphorylation of the88-kDaproteinand several other proteins including the 40- and 50-kDa proteins (Fig. 3B, lane f ) . Control preimmune serum immunoprecipitates incubated with MAP kinases from insulin-treated or untreated rat liver showed neither increased S6 peptide kinase activity (Fig. 3A, lanes a, b, and c) nor increased phosphorylation of the 88kDa protein (Fig. 3B, lanes a, b, and c). Insulin Injection Rapidly Activates M A P Kinase Activator Activity in Rat Liver-Toinvestigate the mechanism by which insulin injection activated the MAP kinaseactivity in rat liver, we examined whether insulin injection activates MAP kinase activator which has been reported to stimulate the in vitro activity of the MAP kinase from quiescent cells in the presence of Mg”+and ATP (36).Recombinant Xenopus MAP kinase (MAPK(Xe)) having a molecular mass of 42 kDa by SDS-PAGE was expressed in E. coli and purified to >90% homogeneity (49).Thispreparation was found to possess detectable, albeit low, kinase activity toward MBP (Fig. 4B, a and c). Using this MAPK(Xe), liver extracts from insulin-treated or untreated rats were applied to Q-Sepharose and eluted a t a gradientof 0-200 mM NaCl, and fractionswere screened for activities which were able to activate the kinase activity of MAPK(Xe) toward MBP. In liver extracts from rats injected with insulin, each fraction was incubated withMg2+ and ATP in the absence or presence of MAPK(Xe) andsubjected to kinase assay toward MBp. We observed a synergistic of MBp phos-

FIG. 2. Detection of insulin-stimulated S6 peptide kinase activity in rat liver. A , detection of insulin-stimulated S6 peptide kinase activity in arsk(m)C immunoprecipitates in CHO-HIR cells. phorylation in fractions in thegradient (Fig. Serum-starved CHO-HIR cells were incubated without or with 10-~ MAPWXe) was subjected to kinase assay toward MBP in the M insulin for 10 min and 100 ng/ml TPA for 10 min a t 37 “C, lysed, absence or mesence of MAP kinase activator fraction (Fie.

. .

peptide. 0, control serbm; 0, cursk(m)C.

ation of MBP was stimulated (Fig. 4B,b). Next, we examined

21094

Phosphorylation Cascade

in Rat Liver following Insulin Injection

B

A FIG. 3. MAP kinases activate 56 peptide kinase activity in arsk(m)C immunoprecipitates and phosphorylate 88-kDa protein. MAP kinases were purified from untreated or insulintreated rat liver. Immunoprecipitates with control preimmune serum or arsk(m)C from the liver extracts of untreated rats were incubated without or with MAP kinases purified from untreated or insulin-treated rat liver in the presence of magnesium [y3*P]ATPfor 15 min at 25 “C. The reaction was stopped by adding 500 ~1 of buffer A, washed with buffer A twice, and divided into two aliquots equally. One immunoprecipitate was subjected to S6 peptide kinase assay. The other was subjected to SDS-PAGE followedby autoradiography.

1500

1000 c

B

-

116:

0

7i

CL N

500

whether insulin stimulates MAP kinase activator activity in rat liver. Liverextracts from untreated or insulin-treatedrats were applied to Q-Sepharose and eluted, and fractions were assayed for MAP kinase activator activity. The increase in stimulation of MBP phosphorylation was stimulated in fractions early in the gradient in insulin-treated rats (Fig. 4C). MAPK(Xe) was incubated without or with fractions eluted from Q-Sepharose (Fig. 4B, fractions 1, 10, and 12) in the presence ofMg2‘ and ATP and subjected to SDS-PAGE followed by pictography (Fig. 40). We observed an increased phosphorylation of the 42-kDa protein in peak fractions of MAP kinase activator (Fig. 4 0 , c and d ) . This 42-kDa protein was specifically immunoprecipitated by aY91(datanot shown), suggesting that thisprotein was MAPK(Xe). We performed similar experiments using protein phosphatase 2A-treated MAP kinases partially purified from untreated rat liver which contained both 42- and 44-kDa MAP kinases as judged by Western blotting using aY91. We have used protein phosphatase2A-treated MAP kinases from control animals, because MAP kinases from control animalswere in a slightly activated state, which may be partly due to circulating insulin or other growth factors in the blood, although the levels of these hormones are low. Liver extracts from 4.5-min insulin-treated or untreated rats were applied to Q-Sepharose, washed, and eluted with an increasing gradient of 0-100 mM NaCl. Each fraction or buffer alone was subjected to MAP kinase activator activityusing phosphatase PA-treated MAP kinases. We observed an insulin-induced synergistic stimulation of MBP phosphorylation in fractions eluted from Q-Sepharose early in theNaCl gradient (data not shown). We were not able to detect a significant MAP kinase activator activity in fractions from untreated rat liver. This synergistic stimulation was also observed in extracts of liver from 2-min insulin-treated rats (data notshown). Vanadate and Cycloheximide Can Also Activate S6 Peptide Kinase Activity-To determine whether vanadate or cycloheximide treatment of the ratstimulates theS6 peptide kinase activity, anesthetized rats were injected intraperitoneally with 30mg/kgbodyweight sodium orthovanadate or 10 mg/kg body weight cycloheximide. 7.5 min later, animals were sacrificed and the livers were removed,quickly homogenized, and ultracentrifuged. The supernatants were subjected to immunoprecipitation with arsk(m)C followed by immune complex kinase assay toward S6 peptide. We observed an increased S6 peptide kinase activity (3.5- and 4.5-fold, respectively) (Fig.

46:

5). Preliminary experiments have shown that vanadate or cycloheximide treatment caused a small or modest increase in MAP kinase activity (1.8-and 1.5-fold, respectively) (data not shown), which should be confirmed by future experiments. DISCUSSION

This study was undertaken to examine whether insulin stimulates a kinase cascade in intact animalliver, one of the major insulin target tissues, under physiological conditions and to clarify the role of this cascade in the insulin’s signal transduction. Using antibodies against MAP kinases and mouse S6 kinase I1 homologue, we were ableto detect insulinstimulated MAP kinase activity and S6 peptide kinase activity in rat liver. Using recombinant Xenopus MAP kinases, we were also able to detect MAP kinase activator activity in rat liver following insulin injection. In this study, we have demonstrated thefollowing: (i) Insulin injection rapidly stimulates activation of MAP kinase activity in aC92 antibody immune complex assay. (ii) Insulin injection also stimulates an S6 peptide kinase activity in arsk(m)C antibody immune complex assay. (iii) MAP kinase activity is maximally activated by 4.5-7.5 min, followed by S6 kinase activity (around 7.5 min). Both MAP kinase and S6 peptide kinase activity are gradually inactivated within 15 min of insulin injection. (iv) Activated MAP kinases are able to activate S6 peptide kinase activity in arsk(m)C immunoprecipitates and phosphorylated an 88-kDa protein in vitro. (v) Insulin injection rapidly stimulates MAP kinase activator activity (an activity which activates MAP kinase activity in vitro) in ratliver. (vi) Cycloheximide and vanadate treatment also cause a modest increase in S6 peptide kinase activity in rat liver. In our previous work, we demonstrated that insulin injection stimulates tyrosine phosphorylation of the insulin receptor fi subunit and a 170-kDa protein as early as 30 s, and a maximal tyrosine phosphorylation of pp170 was observed 2 min after insulin injection (10). We suggested that tyrosine phosphorylation might play some role in the initial steps of insulin’s signal transduction. Recently, we have shown that the tyrosine kinase-defective mutant insulin receptor ( L ~ S ’ ~ ’ ~ + Arg)3is not able to activate MAP kinase activity, demonstrating a requirement of receptor’s tyrosine kinase activity in the activation process of MAP kinase (53). Subsequent to The numbering system of the human insulin receptor sequence used here corresponds to thatof the receptor of Ullrich et al. (56).

Phosphorylation Cascade in Rat Liver

following Insulin Injection

21095

C

B

D

97.497.4-

66.2-

66.2. ..

45.0-

45-

42kDa

3131.01. '

21 -5-

-MBP+

21.5-

14.4-

14.4C

MAPK(Xe) Activator Fraction

T

-

T

+-

(No 10)

FIG. 4. Insulin injectionactivates MAP kinase activator activity in rat liver. A, identification of increased MBP kinase activity in insulin-injected rat liver in the presence or absence of MAPK(Xe). The supernatants of liver extracts from 3-min insulin-injected rats were applied to Q-Sepharose column, washed with 5 ml of buffer A, and eluted with a 30-ml gradient of 0-200 mM NaCI. Fractions of 1.4 ml were collected and assayed for MAP kinase activator activity. The aliquots of the fractions were incubated with M e and [y3*P]ATPfor 20 min in the absence (0)or presence (0)of MAPK(Xe) followed by adding MBP. The incubation continued for another 12 min, [32P]MBP was adsorbed to P81 paper, washed, dried, and counted by Cerenkov counting. B, phosphorylation of MBP was stimulated in the presence of activator fraction. Kinase activity of MAPK(Xe) toward MBP was measured without ( l a n e a ) or with ( l a n e b) prior incubation of MAP kinase activator fraction (fraction 10 in A ) . Instead of adsorbing to P81 paper, 5 X Laemmli buffer was added to thereaction mixture followed by SDS-PAGE. C, insulin injection activates MAP kinase activatoractivity in rat liver. Liver extracts from untreated or insulin-treated rats were applied to Q-Sepharose, eluted, and assayed for MAP kinase activator activity. Phosphorylation of MBP was measured in the absence or presence of MAPK(Xe). The figure shows that the MBP phosphorylation in the presence of MAPK(Xe) was subtracted by that in the absence of MAPK(Xe). D, fractions with MAP kinase activator activity stimulate phosphorylation of MAPK(Xe). Liver extracts from insulin-treated rats were applied to Q-Sepharose and eluted with a gradient of 0-200 mM NaCl. Aliquots of fractions (fraction 2, 10, and 12 in A ) or buffer alone were incubated with MAPK(Xe) in the presence of M$' and [y3'P]ATP for 20 min at 25 "C and were subjected to SDS-PAGE and kinase assay toward MBP.

the activation of the receptor's tyrosine kinase activity in- two divergent pathways that take place in parallel. Further duced byinsulin, tyrosine phosphorylation of pp170 maytake studies will be necessary to differentiate between these posplace first, andthen in turn, mayserve as an upstream sibilities. The MAP kinases and S6 peptide kinase appeared to be sequential activation of the MAP kinase activator activity, MAP kinases, and S6 peptide kinase in rat liver. Alternatively, sequentially activated with an activation ofMAP kinase it is also possiblethat tyrosine phosphorylation of pp170 and immediately precedingor paralleling that of S6 peptide kinase sequential activation of these serine/threonine kinases are activity in insulin-treated rat liver. The kinetics or sequence

21096

Phosphorylation Cascade in Rat Liver following Insulin Injection ity. It will be interesting toknow whether theinsulin-sensitive kinaseimmunoprecipitated by arsk(m)Cisabletophosphorylate the site-1residue of the G subunit of type-1 protein phosphatase and activate the enzyme both in vitro and in

vivo. We havealso shown that cycloheximide and vanadate stimulate S6 peptide kinaseactivity.Somehave reported that cycloheximide treatment activated the70-kDa S6 kinase (28, 41-43) and failed to activate the somatic cell homologues of Xenopus S6 kinase I1 in intact animal rat liver. However, in our system, cycloheximide treatment actually activates the S6 peptide kinase activity immunoprecipitatedby arsk(m)C to a small extent.

FIG. 5 . Vanadate and cycloheximide treatment activate S6 peptide kinase activityin rat liver. Liver extracts from untreated rats or rats injected with 3 mg of sodium orthovanadate/lOO-g body weight for 7.5 min or rats injected with 1 mg of cycloheximide/lOO-g body weight for 7.5 min were subjected to immune complex kinase assay toward S6 peptide with arsk(m)C.

Acknowledgments-We thank Drs. Masato Kasuga and Simeon Taylor for helpful discussions and critical reading of this manuscript and Drs. Takeda and Usui of Hiroshima University for providing the phosphatase 2A catalytic subunits. We also thank Dr. Morioka and personnel of the Radioisotope Center at theUniversity of Tokyo for support. We also appreciate the support offered by Dr. R. Hagura, Dr. A. Kawai, and Director H. Kawashima (Institute for Diabetes Care and Research, Asahi Life Foundation). We also thank Drs. K. Momomura, 0. Koshio, and R. Honda for helpful discussions. K. Toyoshima is acknowledged for typing this manuscript. Note Added in Proof-We have recently identified a MAP kinase activator as a serine/threonine/tyrosine kinase (Kosako, H., Gotoh, Y., Matsuda, S., Ishikawa, M., and Nishida, E. (1992) EMBO J. 11, 2903-2908).

of activation of MAP kinase activity and S6 peptide kinase activity suggested that these insulin-regulated protein kinases REFERENCES may be part of a single signaling pathway. In fact, purified 1. Kasuga, M., Karlsson, F. A,, and Kahn, C. R. (1982) Science 215, 185-187 MAP kinases were able to activate the S6 peptide kinase 2. Kasuga, M., Fujita-Yamaguchi, Y., Blithe, D. L., and Kahn, C. R. (1983) activity and to phosphorylate the 88-kDa protein several and Proc. Natll. Acad. Sci. U. S. A. 80, 2137-2141 proteins in arsk(m)C immunoprecipitates from untreated rat 3. Chou, C. K., Dull, T. J., Russel, D. S., Gherzi, R., Lebwohl, D., Ullrich, A,, and Rosen, 0.M. (1987) J. Biol. Chem. 262,1842-1847 liver extracts. It has been reported by Ahn and Krebs (25) 4. Ebina, Y., Araki, E., Taira, M., Shimada, F., Mori, M., Craik, C. S., Siddle, K., Pierce, S. B., Roth, R. A., and Rutter, W. J. (1987) Proc. Natl. Acad. that fractions containing activated MAP kinases were able to Sei: U. S. A. 8 4 , 704-708 activate the S6 peptide kinase activity in breakthrough frac- 5. Izuml, T., Saeki, Y., Akanuma, Y., Takaku, F., and Kasuga, M. (1988) J. Biol. Chem. 263,10386-10393 tions of Mono Q column accompanied by phosphorylation of 6. Odawara, M., Kadowaki, T., Yamamoto, R., Shibasaki, Y., Tobe, K., Accili, an 85-kDa protein. This is consistent with our datashowing D.,Bevins, C., Mikami, Y., Matsuura, N., Akanuma, Y., Takaku, F., Taylor, S. I., and Kasuga, M. (1989) Science 245,66-68 that 1) insulin-stimulated S6 peptide kinase activity was in 7. White, M. F., Maron, R., and Kahn, C. R. (1985) Nature 318,183-186 the breakthrough fractionsof Q anion-exchange column, and 8. Kadowaki, T., Koyasu, S., Nishida, E., Tobe, K., Izumi, T., Takaku, F., Sakal, H., Yahara, I., and Kasuga, M. (1987) J. Biol. Chem. 262, 73422) it was activated by activated MAP kinases accompanied by 7350 the phosphorylationof 85-90-kDa protein(s). 9. Momomura, K., Tobe, K., Seyama, Y., Takaku, F., and Kasuga, M. (1988) Biochem. Biophys. Res. Commun. 155, 1181-1186 MAP kinase activator stimulates MAP kinase activity in 10. Tobe, K., Koshio, O., Tashiro-Hashimoto, Y., Takaku, F., Akanuma, Y., vitro in the presence ofMg2+ and ATP, accompanied with and Kasuga, M. (1990) Diabetes 39,528-533 11. Tashiro-Hashimoto, Y., Tobe, K., Koshio, O., Izumi, T., Takaku, F., Akphosphateincorporationintobothtyrosineandthreonine anuma, Y., and Kasuga, M. (1989) J. Biol. Chem. 264,6879-6885 residues of the MAP kinases as reported by Ahn and Krebs 12. Perrotti, N., Accili, D., Marcus-Samuels, B., Rees-Jones, R. W., and Taylor, S. I. (1987) Proc. Natl. Acad. Sci. U. S. A. 84,1844-1848 (36). Phosphorylation of each type of residue was found to 13. Bernier. M.. Laird. D. M.. and Lane. M. D. (1987) . . Proc. Natl. Acad. Sci. U. contribute to the activity of the enzyme in vitro (54). The S. A. 84,’1844-1848 14. Pelech, S. L., Tinker, D. A,, Chan, C. P., and Krebs, E. G. (1987) Insulin, other group also reported MAP kinase activator activity as Insulin-like Growth Factors, and Their Receptors in the Central Nervous System (Raizada, M. K., Phillips, M. I., and LeRoith, D., eds) pp. 27-46, “MAP kinasekinases’’ (37). However, consideringthe finding Plenum Publishing Corp., New York that purified, low activity MAP kinase preparationsundergo 15. Weiel, J. E., Ahn, N. G., Seger, R., and Krebs, E. G. (1990) Adu. Second Messenger Phosphoprotein Res. 24,182-195 autophosphorylation on both tyrosine and threonine residues, 16. Denton, R. M. (1986) Adu. Cyclic Nucleotide Protein Phosphorylation Res. this MAP kinase activator need not necessarily possesskinase 2 0 , 293-401 17. Walsh, D. A., Perkins, J. P., and Krebs, E. G. (1968) J. Biol. Chem. 243, activity itself (38). 3763-376.5 Recently, Dent and Cohen (39,40) reported the which ISPK 18. Krebs, E. G., Love, D. S., Bratvold, G. E., Trayser, K.A., Meyer, W. L., and Fischer, E. H. (1964) Biochemistry 3,1022-1033 phosphorylates and activates the type-1 protein phosphatase 19. Sturgill, T. W., Ray, L. B., Erikson, E., and Maller, J. L. (1988) Nature that controls glycogen metabolism in rabbit skeletal muscle. 344, 7150-7158 20. Alcorta, D. A,, Crews, C. M., Sweet, L. J., Bankston, L., Jones, S. W., and They argue that ISPK isvery similar to the 90-kDa S6 kinase Erikson, R. L. (1989) Mol. Cell. Biol. 9,3850-3859 from Xenopus oocytes because they phosphorylate the same 21. Erickson. E.. Stefanovic. D.. Blenis. J.. Erikson. R. L.. and Maller. J. L. (1987) Moi.Cell.Biol. 7 , 3147-3155 substrates such as the 32-residue S6 peptide Kemptide,40 S 22. Sweet, L. J., Alcorta, D. A., Jones, S. W., Erikson, E., and Erikson, R. L. ribosomal subunits, and P P l G at similar relative rates (40). (1990) Mol. Cell. Biol. 10,2413-2417 L. J., Alcorta, D. A,, and Erikson, R. L. (1990) Mol. Cell Biol. 10, Furthermore, they are immunologically related (40). In addi- 23. Sweet, 2787-2792 tion, site-1 of the G subunit is the major residue on the G 24. Chen, R. H.,and Blenis, J. (1990) Mol. Cell. Biol. 10, 3204-3215 25. Ahn, N. G., and Krebs, E. G. (1990) J. Biol. Chem. 2 6 5 , 11495-11501 subunit phosphorylated by ISPK and 90-kDa S6 kinase 11. 26. Chen, R. H., Chung, J., and Blenis, J. (1991) Mol. CeU. Bgl. 11,1861-1967 Our data demonstrate that the antibody (arsk(m)C) against 27. Chung, J., Chen, R. H., and Blenis, J. (1991) Mol. Cell. Bcol. 11,1868-1874 28. R. L. (1991) J. Biol. Chem. 266,6007-6010 the mouse homologue of 90-kDa S6 kinaseI1 actually immu- 29. Erikson, Hoshi, M., Nishida, E., and Sakai, H. (1988) J. Bcol. Chem. 263, 53965401 noprecipitated a n insulin-dependent S6 peptide kinase activ’

-._I

~~

~

I

~

~~

~~~~



Phosphorylation Cascade in Rat Liver following Insulin Injection 30. Hoshi, M., Nikshida, E., and Sakai, H. (1989) Eur. J. Biochem. 184,477486 31. Ahn,N. G., Weiel, J. E., Chan, C. P., and Krebs, E. G. (1990) J . BWl. Chem. 2 6 6 , 11487-11494 32. Rossomando, A. J., Payne, D. M., Weber, M. J., and Sturgill, T. W. (1989) Proc. Natl. Acad Sci. I- J. S A RfL ", G94fl-G9A.? 33. Boulton, T. G., Yancopoulos, G. D., Gregory, J. S., Slaughter, C., Moomaw, C., Hsu, J., and Cobb. M. H. (1990) Science 249.64-67 34. Boulton, T. G., Nye, S. H., Robbins, D. J., Ip, N. Y.. Radzieiewska.E.. Morgenbesser, S. D., DePinho, R. A., Panayotatos, N., Cobb,k. H.,.and YancoDoulos. G. D. (1991) Cell 66.663-675 35. Tobe, K , Kadowaki, T., Tamemoto: H.. Ueki. K., Hara. K.. Koshio. 0.. Momomura, K., Gotoh, Y., Nishicia, E . , Akanuma, Y.,' Yazaki, Y.,'and Kasuca. M. (1991) J. Biol. Chem. 266.24793-24803 ~

~~

~

""

"" ""

E. G. (1991)-J. Biol. Chem. 266,4220-4227 37. Gomez. N.. and Cohen. P. (1991) Nature 363.,~ 170-173 ~. 38. Seger, R . , Ahn, N. G.,'BoUlton,'T. G., Yancopoulos, G. D., Panayotatos, N., Radiziejewska, E., Ericsson, L., Bratlien, R. L., Cobb, M., and Krebs, E. G. (1991) Proc. Nutl. Acud. Sei. U. S. A. 88,6142-6146 39. Dent, P., Lavoinne, A., Nakielny, S., Coldwell, F. B., Watt, P., andCohen, P. (1990) Nature 348,302-308 40. Levoinne, A,, Erickson, E., Maller, J. L., Price, D. J., Avruch, J., andCohen, P. (19911 Eur. .,722-728 ~ ~ " . . ,I .. ~ R i o.c h m .~ -19 _9 -, ." ." 41. Jeno, P., Ballou, L. M., and Thomas, G. (1988) Proc. Nutl. Acud. Sci. U. S. A. 86.406-410 42. Jeno, P., Jaggi, N., Luther, H., Siegmann, M., and Thomas, G. (1989) J. BWL Chem. 264,1293-1297 43. Price, D. J., Nemenoff, R. A,, and Avruch, J. (1989) J. Biol. Chem. 2 6 4 , 13825-13833 ~

21097

44. Gregory J. S. Boulton T. G., Sang, B.-C., and Cobb, M. H. (1989) J. Biol. Chep: 264: 18397-lk401 45. K riakls J. M., and Avruch, J. (1990) J.Biol. Chem. 2 6 6 , 17355-17363 46. KrarIund, J. K., Bradford, A. P., Milla, M. G., and Czech, M. P. (1990) J. Biol. Chem. 266.227-234 47. Brandt, H., Capuii6yZ.L.i and Lee, E. Y. C. (1975) J. Biol. Chem. 2 6 0 , 8038-8044 48. Kabura Y. Momomura K. Andoh A Honda R. Akanuma Y. Yazaki Y., K%u& M., and KLddwaki, T: (?991) A d s t r h of the i3rd Ann& $et?ting of the Endocrine Society, June 19-22, 1991, pp. 309, Wash. u. c;. 49. Matsuda S Kosako H. Takenaka K Mori ama K Sakai H.,Akiyama, T., Goto$, Y., and'Ni;hida, E. (lb92; E M 2 0 J. '1 973-982 50. Gotoh Y. Nishida, E. Matsuda, S. Shiina N. Kosako H. Shiokawa, K., Akiiama, T., Ohta, K., and Sakd, H. (199l)Nuture 349; 251-254 51. Gotoh, Y., Moriyama, K., Matsuda, S., Okuyama, E., Kishimoto, T., Kawasaki, H., Suzuki, K., Yahara, J., Sakal, H., and Nishida, E. (1991) E M R 0 .-. I I", n . 3fifil-3fifiA 52. Gotoh Y., Nishida, E., Yamashita, T., Hoshi, M., Kawakami, M., and Sakii, H.(1990) Eur. J. Biochem. 193,661-669 53. Ando, A., Momomura, K., Tobe, K., Yamamoto-Honda, R., Kaburagi, Y., Sakura, H., Koshio, O., Akanuma, Y., Yazaki, Y., Kasuga, M., and 267.1271U1-127Flfi - ., -- . - - -- . - Kadowakl. T.(1992) J . Biol. Chem. 54. Anderson, N. G.; Tonks, N. K., Maller, J. L., and Sturgill, T. W. (1990) Nature 343.651-653 55. Ebina Y. Ellis L Jarnagin K. Edery M. Graf L. Clauser E , Ou, J.H., haharz, k., k a n Y. d., Gbldfine,'I. d., Roth, A. A., and Rutter,W. J. (1985) Cell 4 0 , 743-758 56. Ullrich A., Bell, J.R. Chen E. Y. Herrera R. Petruzzelli L. M Dull T. J.,dreay, A.,CouAsens'L.,Li(ao,Y.-C.;Tkubokawa Mison'A,, Seeburg, P. H., Grunfeld: C., Rosen, 0. M., and Ramacdandran,J. (1'985) Nature 3 1 3 , 756-761

1;

" "

" "

~~

~

h.,