Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Purification and characterization of an insulin-stimulated kinase Zhande, Rachel 1991

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_spring_zhande_rachel.pdf [ 4.44MB ]
JSON: 831-1.0086465.json
JSON-LD: 831-1.0086465-ld.json
RDF/XML (Pretty): 831-1.0086465-rdf.xml
RDF/JSON: 831-1.0086465-rdf.json
Turtle: 831-1.0086465-turtle.txt
N-Triples: 831-1.0086465-rdf-ntriples.txt
Original Record: 831-1.0086465-source.json
Full Text

Full Text

PURIFICATION AND CHARACTERIZATION OF AN INSULIN-STIMULATEDKINASE.byRACHEL ZHANDEDIPLOM INGENIEURIN CHEMIE TECHNISCHE FACHHOCHSCHULE, BERLIN1985A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THEREQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Biochemistry)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIANovember 1991© Rachel Zhande, 1991In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(SignatureDepartment of ^The University of British ColumbiaVancouver, CanadaDate ^C3e-e. (R 191DE-6 (2/88)iiABSTRACTThe rapid effects of insulin action appear to be mediated atleast in part by the activation of a number of discreteprotein serine/threonine kinases. The cellular targets ofsome of these kinases include key metabolic enzymes such asacetyl-CoA carboxylase (ACC) and the ribosomal S6. While,the S6 kinases have now been purified and characterized,very little is known about the nature of the proteinkinase(s) able to phosphorylate ACC particularly on theinsulin-directed site. Previous studies demonstrated that amyelin basic protein-kinase (MBP-kinase) from sea-star isable to phosphorylate the insulin-directed site on ACC.Thus, it was interesting to find out whether a mammalianhomolog of this kinase existed. Rapid chromatography ofsupernatant fractions from rat adipose tissue by MonoQ ionexchange using fast protein liquid chromatography revealedseveral peaks of insulin-stimulated protein-serine/threoninekinase activity towards myelin basic protein. Progress inthe purification and characterization of these kinases wasinitially impeded by the rapid decay in protein kinaseactivity of these extracts. Thus, a major concern initiallywas stabilization of the insulin-stimulated proteinserine/threonine kinase activity which would enablesubsequent purification and characterization of the kinases.Stabilization was achieved by the use of phosphataseinhibitors, particularly in combination with a rapidiiiprocedure which included ammonium sulphate precipitation.This enabled storage of the activated kinases (withapparently very little loss in activity) and furtherpurification. After first examining the chromatographicproperties using individual column techniques, the insulin-activated serine/threonine kinases were purified by aprocedure which involved ammonium sulphate precipitation,and sequential chromatography on polylysine-agarose, MonoQ,heparin-agarose and a final MonoQ. The purified enzymeshowed only two major silver-stained polypeptides (withsubunit sizes of 200 and 44 kDa) as judged by SDS-PAGE.From analysis of Western blots performed with severaldifferent anti-kinase antibodies, the 44 kDa polypeptide wasfound to be immunologically related to a sea-star MBP-kinaseas well as to a family of mammalian mitogen-activatedkinases designated ERKs.It is concluded that the 44 kDa MBP-kinase in rat adiposetissue is a member of the family of mitogen-activatedkinases. However, the precise relationship to these kinasesas well as the regulatory properties of this insulin-activated kinase await further characterization includingmolecular cloning of the rat adipose enzyme.ivTABLE OF CONTENTSABSTRACT^  iiList of Figures^  viList of Tables  viiList of Abbreviations^  viiiAcknowledgements  ixINTRODUCTION^  11.1 Rapid Metabolic Effects of Insulin^ 11.2 Characteristics of the insulin receptor^ 31.2.1 Structure of the receptor  31.2.2 Receptor tyrosine kinase  41.3 Signalling mechanisms implicated in the mechanismof insulin action^  51.3.1 Protein phosphorylation/dephosphorylation(in insulin action) ^  51.3.2 Phospholipid (in insulin action) ^ 71.3.3 Glycosyl phosphatidylinositol (in insulinaction) ^  91.4 Insulin promotes increases and decreases inspecific protein-serine/threoninephosphorylation  101.5 Insulin-stimulated protein serine/threoninekinases^  161.5.1 S6 kinases^  181.5.2 p42 and p44 MAP (Mitogen-activated protein)kinases Activation of 90 kDA S6 kinases by p42MAP kinase^  231.6 The thesis investigations^  25MATERIALS AND METHODS  271.1 Materials^  271.2 Methods  281.2.1 Tissue incubation and preparation ofextracts  281.2.2 Ammonium sulphate precipitation^ 291.2.3 Assay of kinase activity^  301.2.4 SDS-Polyacrylamide gel electrophoresis^ 311.2.5 Silver staining of gels  321.2.6 Autoradiography^  321.2.7 Scanning Densitometry^  331.2.8 Chromatography  331.2.8.1 DEAE-cellulose ion exchange^ 331.2.8.2 Phenyl-sepharose chromatography ^ 341.2.8.3 MonoQ ion exchange. ^ 341.2.8.4 Affinity chromatography using MBP-Affigel column  351. Coupling of. MBP to Affigel^ 351. MBP column chromatography  361.2.9 Western blotting Analysis^ 361.2.9.1 SDS-PAGE and Transfer of proteins ^ 361.2.9.2 Antibody binding  371.2.10 Immunoprecipitation  38V1.2.11 Protein Assays^  39RESULTS^  401.0 Instability of protein kinase activity inextracts of rat fat pad tissue^  402.0 Stabilization of insulin-activated kinases^ 432.1 Effect of phosphatase inhibitors  432.1.1 Estimation of phosphatase activity incrude extracts of rat adipose tissue^ 442.1.2 Effects of microcystin on thephosphorylation of myelin basicprotein by protein kinases present inrat adipose tissue^  452.2 Ammonium sulphate precipitation^ 483.0 Purification of insulin-activated proteinserine/threonine kinases from rat adiposetissue^  503.1 Application of various chromatographictechniques for purification of insulin-activated serine/threonine kinases from ratadipose tissue^  533.1.1 MonoQ chromatography. ^  533.1.2 DEAE-cellulose chromatography^ 553.1.3 Phenyl-sepharose chromatography  573.2 Purification of insulin-activated proteinserine/threonine kinases using sequentialchromatography with polylysine-agarose, monoQ,heparin-agarose and monoQ^  594.0 Western blotting analyses  654.1 Immunoblottinq with antibody raised againstsea-star p44mPt^  664.2 Immunoblotting with R2 antibody^ 734.3 Immunoblotting with R1 antibody  734.4 Immunoblotting with anti-phosphotyrosineantibody^  744.5 Immunoblotting with GEGA antibody^ 745.0 Additional characterization of rat adiposeprotein serine/threonine kinases  765.1 Inhibition of endogenous proteinphosphorylation by added myelin basicprotein^  765.2 Immunoprecipitation of MBP-kinaseactivity  78DISCUSSION  79Stabilization of insulin-activated MBP-kinases ^ 81Purification of insulin-activated MBP-kinases ^ 82Relationship of adipose tissue p44 MBP-kinase toother insulin-stimulated protein serine/threoninekinases^  84Inhibition of endogenous protein phosphorylationby added myelin basic protein^  88Immunoprecipitation^  89Time course/Dose response for MBP-kinaseactivation^  90Conclusion  92viREFERENCES^  93LIST OF FIGURESFigure la.Figure lb.Figure 2Figure 10Figure 11Figure 12Figure 13Effect of microcystin on phosphorylation ofMBP by protein kinases in high speedsupernatant fractions^  46Effect of microcystin on phosphorylation ofmyelin basic protein  47MonoQ chromatography of insulin-activatedserine/threonine kinases from rat adiposetissue 54DEAE-cellulose ion exchange chromatography^ 56Phenyl-Sepharose chromatography^ 58Polylysine-agarose chromatography  61MonoQ ion exchange chromatography^ 62MonoQ ion exchange chromatography  63SDS-PAGE of purified insulin-stimulatedprotein serine/threonine kinase^ 64Chromatography of insulin-stimulatedserine/threonine kinases on MonoQ andimmunoblotting with various antibodies^ 67Immunoblotting of insulin-activated kinasespartially purified on polylysine-agarose^ 70Immunoblotting of purified insulin-activatedprotein serine/threonine kinases^ 71Immunoblotting of fractions from polylysine-agarose peak 2^  72Inhibition of endogenous proteinphosphorylation by added MBP^  77Figure 3Figure 4Figure 5Figure 6Figure 7Figure 8Figure 9viiLIST OF TABLESTable 1 Proteins exhibiting changes inphosphorylation in response to insulin^ 12Table 2 Insulin-sensitive serine/threonine kinases^ 17Table 3 Loss of insulin-stimulated kinase activityunder different storage conditions^ 42Table 4 Precipitation of myelin basic protein-kinase ^ 49Table 5 Various chromatographic matrices used inattempts to purify insulin-stimulated proteinserine/threonine kinases^  52viiiLIST OF ABBREVIATIONSACC^acetyl-CoA carboxylaseATP adenosine triphosphatecAMP cyclic adenosine 3',5'-monophosphateDEAE-cellulose^diethyl-amino-ethyl celluloseDTT^dithiothreitolEDTA ethylenediaminetetraacetic acidEGTA ethyleneglycoltetraacetic acidERK extracellular signal-regulated kinaseFig.^figureFPLC fast protein liquid chromatographyHSS high speed supernatant from rat adiposetissue.MPK^meiosis-activated protein kinaseMAP microtubule-associated proteinMAPK mitogen-activated protein kinaseMBP myelin basic proteinMOPS^3-[N-morpholino]propanesulfonic acidP inorganic phosphatePAGE polyacrylamide gel electrophoresisPMSF phenylmethylsulfonylfluorideRSK^ribosomal S6 kinaseSDS sodium dodecyl sulfateTCA trichloroacetic acidTris tris(hydroxymethyl)aminomethaneUNITS OF MEASUREMENTDa^ daltonDPM disintegrations per minuteg gramxg times gravityk^ kilo1 literm milliM molarmol^ moleA micromin minuten nanop^ picoixAKNOWLEDGEMENTSThe final work may bear one person's name but it is almostalways the result of collaborative efforts. This iscertainly true with this thesis. A number of people haveshared their insights, ideas and experiences with me and Iwould like to thank all of them. A very special thanks toDr. R.W. Brownsey for the excellent supervision, constantencouragement and support during the course of this study.Many thanks to Drs. S.L. Pelech and J.S. Sanghera for theirassistance in chromatography work as well as the supply ofantibodies. Also, the generous gifts of antiphosphotyrosineantibodies and myelin basic protein from Alice Mui and FrankNezil, respectively, are greatly appreciated. I also wish toextend my gratitude to Dr. Kath Quayle and Bob Winz and EdHanada for the support advice and technical help offered.Finally, the Commonwealth University Scholarship received isgratefully acknowledged.1INTRODUCTION. This thesis is concerned with studies designed tofurther our understanding of the biochemical basis ofinsulin action. In particular, the studies are concernedwith rapid metabolic responses of cells observed within afew minutes after exposure to the hormone.1.1 RAPID METABOLIC EFFECTS OF INSULIN. The hormone insulin regulates a wide spectrum ofmetabolic processes in a variety of mammalian tissues(Czech, 1977; Denton, 1986). Of particular importance is themodulation of metabolic pathways leading to increasedsynthesis of protein, glycogen and fat. Concomittantly, therates of the opposed, catabolic pathways of gluconeogenesis,lipolysis and glycogenolysis are coordinately inhibited inresponse to insulin. The signaling system or systems whichmediate these responses are initiated by a membrane-boundinsulin receptor and result in the regulation of keycontrolling enzymes in their respective metabolic pathways.In view of the wide range of changes in metabolisminduced by exposure of different cell types to insulin, acorrespondingly large number of possible experimentalsystems may be considered appropriate to investigate themechanism of action of the hormone. Indeed, a vastliterature has developed describing studies of insulinaction and the references quoted here are necessarily2restricted in scope. The thesis goals and the introductionhere have been motivated by earlier studies of insulinaction in rat adipose tissue, which has been an importantmodel system experimentally. In this tissue, effects ofinsulin may be divided into those effects which can bedetermined upon exposure of cells to insulin alone (such asstimulation of fatty acid biosynthesis) and alternatively,effects which are most apparent in the presence of hormoneswhich increase cellular cyclic AMP concentration (such asinhibition of catecholamine-stimulated lipolysis). Theselatter effects are likely to be even more complex thandirect effects of insulin alone and will not be described indetail here.Of the direct effects of insulin acting alone, thestimulation of fatty acid biosynthesis for example, involvesincreases in glucose transport (Simpson and Cushman, 1986),activation of enzymes including pyruvate dehydrogenase(Coore et al., 1971) and acetyl-CoA carboxylase (Denton etal., 1981; Halestrap and Denton, 1973; Lee et al., 1973).These changes appear to be brought about by alterations inspecific activity rather than protein synthesis ordegradation and are still apparent following tissuedisruption and subcellular fractionation.31.2 CHARACTERISTICS OF THE INSULIN RECEPTOR. 1.2.1 STRUCTURE OF RECEPTOR. The insulin receptor is an integral membrane glycoproteinwith an apparent relative molecular mass (Mr) of 350-400,000composed of two a-subunits (Mr 130,000) and two B-subunits(Mr 95,000). The a and B subunits are held together bystrong disulfide bridges. The insulin binding domain islocated on the a-subunit which is completely extracellular.The B-subunit possesses the single transmembrane domain anda cytoplasmic region containing a protein-tyrosine kinasedomain (including the putative ATP binding site) andtyrosine phosphorylation sites. It is the B-subunit which isautophosphorylated on tyrosine residues upon the binding ofinsulin to the a-subunit. This autophosphorylation in turnactivates the protein-tyrosine kinase activity of theinsulin receptor, rendering the receptor capable ofphosphorylating other substrates. The insulin receptor istherefore multifunctional; it contains insulin bindingactivity, autotyrosine kinase and protein-tyrosine kinaseactivity.41.2.2 RECEPTOR TYROSINE KINASE. The receptor tyrosine kinase region is structurallyrelated to other tyrosine kinases including the transformingproteins encoded by oncogenes and receptors for theepidermal growth factor (EGF), platelet derived growthfactor (PDGF), and insulin-like growth factor-I (IGF-I).Binding of insulin to its receptor causes theautophosphorylation and activation of its B-subunit tyrosinekinase activity (Kahn and White, 1988). Site-directedmutagenesis of the receptor ATP binding site abolishes bothreceptor autophosphorylation and biological activity,suggesting that the kinase is critical for signaltransmission (Chou et al., McClain et al., 1987). However,mutant receptors with defective tyrosine phosphorylationsites and hence reduced kinase activity have been shown toretain some biological activity (Ellis et al., 1986; Wildenet al., 1990). In view of these studies as well as othersconcerning anti-receptor antibodies which weakly stimulatereceptor autophosphorylation, the relation between kinaseactivity and biological activity of the insulin receptorremains a matter of debate.51.3 SIGNALLING MECHANISMS IMPLICATED IN THE MECHANISM OFINSULIN ACTION. 1.3.1 PROTEIN PHOSPHORYLATIONIDEPHOSPHORYLATION SIGNALLINGSYSTEM. Several studies suggest that the tyrosine kinaseactivity of the B-subunit of the insulin receptor isessential to mediate the biological effects of insulin.Deletions and truncations that result in loss of insulin-stimulated kinase activity or mutations in the ATP-bindingsite cause loss of some of the rapid effects of insulinaction (Wente and Rosen, 1990; Olefsky, 1990; Kasuga et al.,1990). Similarly, inhibition of kinase activity bymonoclonal antibodies impairs insulin effects (Kasuga etal., 1990; Avruch et al., 1990). This leads to thehypothesis that the effect of insulin is mediated throughtyrosine phosphorylation of cellular substrates. Onepotential substrate is the insulin receptor itself. Thisraises the possibility that a conformational change or otherchange induced in the receptor by autophosphorylation,rather than the increase in tyrosine kinase activity towardsexogenous substrates could be involved in the transmissionof the signal. This is supported by the finding thatmutations of the autophosphorylation domain alter insulinresponses (Shechter et al., 1989; Saperstein et al., 1989).In addition, this may also explain why some monoclonal6antibodies stimulate glucose and amino acid transport withno apparent activation of the tyrosine kinase (Hawley etal., 1990; Soos et al., 1989).A number of proteins are phosphorylated on tyrosineresidues in response to insulin. These cellular substratesinclude proteins (of unknown function) of subunit Mr 15,000,60,000, 120,000, 185,000, or 240,000 (Kasuga et a1.,1990;Avruch et al., 1990) and also recognized enzymes includingphosphatidylinositol 3-kinase ( Endemann et al., 1990) and amicrotubule associated protein serine/threonine kinase witha subunit Mr 42,000 (Ray and Sturgill, 1988). There are alsoa number of other proteins that are phosphorylated onserine/threonine residues in response to insulin. Theseinclude the receptor itself, ATP-citrate lyase, acetyl CoAcarboxylase, S6 ribosomal protein, (Avruch et al., 1990;Czech et al., 1988; Denton, 1986) an isozyme of cAMPphosphodiesterase, (Degerman et al., 1990) Raf-1 kinase(Blackshear et al., 1990; Kovacina et al., 1990) and someunidentified proteins (Avruch et a/.,1990; Denton, 1986).The functional significance of the insulin-inducedphosphorylation in most of these proteins is notparticularly clear, although it is often assumed that somechange in function occurs.Taken together, these findings suggest a phosphorylationcascade which is initiated by stimulation of the receptortyrosine kinase leading to the activation of one or moreprotein serine/threonine kinases. Although insulin treatment7of several cell lines results in the phosphorylation andsubsequent activation of various serine/threonine kinasesassayed in vitro, such as MAP kinases, S6 kinases and Raf-1kinase, none of these have been shown to be phosphorylatedby the insulin receptor tyrosine kinase.Insulin also promotes dephosphorylation of severalproteins (Table 1). This is probably mediated by activationof protein phosphatases. Recently, evidence has beenpresented that insulin stimulates type 1 phosphatase inskeletal tissue by phosphorylating its G-subunit on aspecific serine (Dent et a/., 1990). This results inenhanced association of the phosphorylase with the glycogenparticle and hence activation of glycogen synthase andinactivation of phosphorylase kinase and can thus explainthe effects of insulin on glycogen metabolism. Otherphosphatases are probably involved in the dephosphorylationof other proteins (for example, in the activation ofmitochondrial pyruvate dehydrogenase; Denton et al., 1984;Thomas and Denton, 1986) but the mechanisms by which thesephosphatases are activated are unknown.1.3.2 PHOSPHOLIPID-SIGNALLING SYSTEM. Evidence supporting a role for phospholipid orglycophospholipid hydrolysis in insulin action isfragmentary and controversial. Whereas, some authors havereported an elevation of diacylglycerol (DAG) in some cell8types treated with insulin (Saltiel, 1990), others havereported no changes in cells that are major targets of thehormone (Turinsky, 1990; Augert and Exton, 1988).Conflicting reports in support of or arguing againstinvolvement of protein kinase C (PKC) in insulin action havebeen also presented (Cooper et al., 1990; Blackshear et al.,1987). On the one hand, phorbol esters (the tumor-promotingactivators of PKC which can substitute for DAG) can mimicsome of the effects of insulin on glucose transport andphosphorylation of proteins (Farese et al., 1985; Graves andMcDonald, 1985; Trevillyan et al., 1985). On the other hand,phorbol esters inhibit insulin stimulated lipogenesis andantagonize insulin action on glycogen synthesis andglycogenolysis (Van de Werve et al., 1985; Ahman et al.,1984). Further, down-regulation of PKC by prolonged exposureto phorbol-l2-myristate-l3-acetate (PMA) does notnecessarily abolish or even diminish rapid responses toinsulin. The dual role played by phorbol esters, both asinsulinomimetic agents as well as inhibitors of insulinaction is indeed intriguing. However, it may be possible toaccommodate these apparently conflicting observations byinvoking the involvement of distinct second messengers inthe mode of insulin action. PKC activation is usuallyaccompanied with stimulation of the hydrolysis ofpolyphosphoinositides, leading to the generation of inositolphosphates and DAG that contains arachidonate in the C2position. The absence of PI turnover in response to insulin9and the scarcity of arachidonate in the insulin-generatedDAG suggest that this DAG may indeed be different, and mayarise from an alternate source.1.3.3 GLYCOSYL PHOSPHATIDYLINOSITOL (GPI)-SIGNALLING SYSTEM. It has been suggested that an inositol phosphate glycan(IPG) released from glycosyl phosphatidylinositol isinvolved in the insulin-signalling system (Saltiel, 1990).Although the insulin-sensitive GPI anchor is believed to bestructurally analogous to the GPIs that anchor many cellsurface proteins its chemical nature has remained elusive.In addition little is definitely known about the chemicalcomposition of the active material released from plasmamembranes by insulin action. For example, the form ofinositol (myo or chiro) is debated, as is the carbohydratecomposition (Saltiel, 1990; Mato et al., 1987). The releaseof IPG would require the action of a phospholipase, andsince the activity of certain phopholipases is regulated byG-proteins, it has been speculated that an insulin-sensitivephospholipase regulated via a G-protein mechanism, mayaccount for IPG release (Burdett et al., 1990; Davis andMcDonald, 1990). Data supporting this postulate is howeverstill fragmentary, and a final point of confusion is raisedby uncertainty about whether the IPG is released at theextracellular or cytoplasmic surface of the plasma membrane(Saltiel et a/., 1989; Saltiel, 1990).101.4 INSULIN PROMOTES INCREASES AND DECREASES IN SPECIFICPROTEIN SERINE/THREONINE PHOSPHORYLATION. Studies of proteins which change activity in response toinsulin have revealed that some of these proteins exhibitnet dephosphorylations (Table 1). Thus, dephosphorylation ofglycogen synthase increases the activity of the enzymeleading to increased glycogen deposition (Parker et al.,1983), whereas, dephosphorylation of hormone-sensitivelipase (Stralfors et a/., 1984) and phosphorylase kinase(Sheorain et al., 1984) inhibits their activities and therespective breakdown of triglyceride and glycogen stores.Similarly, pyruvate dehydrogenase is dephosphorylated andactivated in response to insulin (Martin et al., 1972),leading to increased synthesis of acetyl-CoA and fattyacids. Taken together, these and other observationsdemonstrate that a central role of insulin action ismediated by dephosphorylation of key regulatory enzymes inintermediary metabolism.Paradoxically, insulin action also leads to rapidincreases in phosphorylation of serine/threonine residues ofa family of proteins (distinct from those noted above) intarget cells within minutes of exposure to the hormone.Quantitatively, the insulin-stimulated proteinphosphorylations^are^more^extensive^than^thedephosphorylations described above. Several major proteins11within fat cells are targets of this effect (Table 1) andinclude ATP-citrate lyase (Alexander et al., 1979; Witters,1981), the ribosomal protein S6 (Smith et al., 1980),acetyl-CoA carboxylase (Brownsey and Denton, 1982) andproteins so far only identified according to their subunitrelative molecular weight of 22,000 and 61,000 (Belsham etal., 1980; Belsham and Denton, 1980). In addition, increasedphosphorylation of the B-subunit of the insulin receptor inresponse to insulin has also been demonstrated (Kasuga etal., 1982; Pang et al., 1985). It is important to stressthat increased phosphorylation of the insulin receptor B-subunit occurs at serine residues as well as tyrosine.12TABLE 1. PROTEINS EXHIBITING CHANGES IN PHOSPHORYLATION INFAT CELLS EXPOSED TO INSULIN. DECREASES IN PHOSPHORYLATION^INCREASES IN PHOSPHORYLATIONPyruvate dehydrogenaseTriglyceride lipaseGlycogen phosphorylasePhosphorylase kinaseGlycogen synthaseAcetyl-CoA carboxylaseATP-citrate lyaseInsulin receptorRibosomal protein S661 kD plasma membraneprotein22 kD cytoplasmic proteinInsulin-directed phosphorylation of ATP-citrate lyase(ACL) is site-specific and is apparently mediated by aserine/threonine kinase with a subunit Mr of 52,000 (Yu etal., 1990; Klarlund et al., 1990). In general insulin andhormones that elevate cAMP levels have opposite cellulareffects. Surprisingly, however, both insulin and hormoneswhich increase cAMP levels cause the phosphorylation of ACLon the same serine residue as determined by amino acidsequencing of phosphopeptides (Pierce et al., 1982;Ramakrishna et al., 1983). One possible caveat to thissystem is the appearance of additional phosphopeptides13including phosphopeptides reported by Benjamin and coworkers(Ramakrishna and Benjamin, 1985). Although insulin enhancesserine phosphorylation of ACL, no apparent change incatalytic activity of the enzyme has been detected andsubsequently no regulatory role has been assigned to thiscovalent modification, although an argument for analteration in subcellular distribution of ACL has beenproposed.The phosphorylation of the ribosomal protein S6 has beencorrelated with growth and increased protein synthesis(Krieg et al., 1988; Palen and Traugh, 1987). It is believedthat the phosphorylated epitope on the S6 polypeptide islocated within a cleft where mRNA binds (Bommer et al.,1980; Kozma et al., 1989), a location consistent with a rolein regulation of initiation. Direct evidence that S6phosphorylation leads to altered rates of initiation invitro however, is still missing, perhaps due to thecomplexity of experiments involved. Furthermore, it isbecoming evident that a number of proteins associated withinitiation and elongation steps of protein synthesis arealso substrates for protein kinases and phosphatases(reviewed by Hershey, 1991).Both the activation and inhibition of acetyl-CoAcarboxylase (ACC) mediated by insulin and adrenaline,respectively, are accompanied by modest overall increases inphosphorylation of the enzyme (Brownsey and Denton, 1982;Brownsey et al., 1977; Witters, 1981). An explanation for14this apparent paradox came about with the demonstration thatboth insulin and adrenalin brought about phosphorylation butof different sites, which could be resolved by 2-dimensionalmapping of tryptic phosphopeptides. The insulin-mediatedactivation of ACC is accompanied by phosphorylation at aserine residue located on a tryptic peptide (designated asthe I-peptide, Brownsey and Denton, 1982). Adrenalin on theother hand, induces phosphorylation of the "A-sites" whichare serine residues located on the tryptic peptidesdesignated A-peptides. The latter peptides have now beensequenced (Hardie and coworkers) and their location withinthe primary sequence determined. Serine residues 79, 1200and 1215 are the principal sites phosphorylated in responseto rising cell cAMP (Haystead et al., 1990; Davies et al.,1990).The greater abundance of insulin-stimulated proteinphosphorylations over dephosphorylations suggests theirpotential importance in the mechanism of insulin action.These insulin activated phosphorylations may be due toincreases in kinase activities or decreased phosphataseactivity, or both. In general, effects of insulin on proteinphophatases have been difficult or impossible to detect oncethe integrity of cells is disrupted by homogenizations. Twoimportant exceptions have been described. Firstly, withinintact mitochondria, evidence of persistent activation ofpyruvate dehydrogenase phosphate phosphatase has beenobserved following insulin treatment (Hughes and Denton,151976). Very recently, evidence for activation of glycogensynthase^phosphatase^has^also^been^demonstrated.Intriguingly, the activation of glycogen synthasephosphatase appears to be mediated by activation of aninsulin-sensitive protein kinase (Dent et al., 1990). Thus,in skeletal muscle, insulin appears to activate a proteinserine/threonine kinase (ISPK). ISPK phosphorylates site 1on the glycogen targeting subunit of protein phosphatase 1(PP1G), leading to the enhancement of glycogen synthasephosphatase activity of PPIG. The latter causes site-specific dephosphorylation at multiple serine residues onglycogen synthase and the 8-subunit of phosphorylase kinase,activating and inactivating these two enzymes, respectively.As a result, glycogen synthesis is stimulated andglycogenolysis inhibited. Clearly, in this case theactivation of a protein phosphatase (PPIG) by the ISPKexplains how insulin can simultaneously stimulate thephosphorylation of some proteins and the dephosphorylationof others using a common mechanism. An important element ofthis regulatory network is spatial separation, of keyproteins in this case by binding to or released from theglycogen particles of muscle cells.161.5 INSULIN-STIMULATED PROTEIN SERINE/THREONINE KINASES. A number of insulin stimulated protein kinases have beendescribed by various authors (see Table 2). The insulin-sensitive kinases generally exhibit the critical propertythat they are stable after tissue disruption under carefullycontrolled conditions which usually include the presence ofphosphatase inhibitors. This is consistent with the role ofphosphorylation as a mode of regulation of these activatedkinases and has facilitated further purification. Inaddition, the insulin-stimulated protein-serine/threoninekinases are distinct from a number of well-characterizedprotein kinases classically activated by a variety of secondmessengers. The important characteristics of some of thesekinases are now described below.17TABLE 2 INSULIN-SENSITIVE SERINE/THREONINE RINASESENZYMEp65-70 S6 kinaseS6 kinase,rskp52 Kemptide kinaseERK1Casein kinase IIp70 Raf-1Insulin receptorserine kinaseMicrosomal kinaseGlycogen synthasekinase 3Multifunctionalprotein kinase(inhibited)Insulin stimulatedprotein kinase.(very similarto p90s6K )SUBSTRATE40S ribosomes40S ribosomes,glycogen synthase,S6 peptide.lamin C,G-PP1kemptide, histones,ACLMAP-2, MBPRRREEETEEECaseinhistones, ACL,syntide 2,raf-1 peptideinsulin receptorhistone V-S,kemptideinhibitor-2glycogensynthase,ACLG-PPIglycogensynthaseREFERENCENovak-Hofer andThomas (1984);Cobb, 1986.Erikson andMailer, 1988;Ward andKirschner, 1990;Lavoinne et al.,1991.Klarlund et al.,1990.p42 Map-2 kinase^MAP-2, MBP(identical to ERK2)Ray and Sturgill1988; Ericksonet a/.,1990.Boulton et al.,1990.Sommercorn etal., 1987.Kovacina etal., 1990;Blackshearet al., 1990.Smith and Sale,1989.Yu et a/.,1987.Yang et al.,1988.RamakrishnaandBenjamin, 1988.Lavoinne et al.,1991.181.5.1 S6 KINASES. Insulin and various mitogens markedly increase thephosphorylation of ribosomal protein S6 in a variety of celltypes (Smith et al., 1980; Cobb, 1986; Erikson and Mailer,1986; Pelech et al., 1986). Phosphorylation of S6 iscatalyzed by at least two distinct mitogen activated S6kinases which are distinguishable by size, the RSK (forribosomal S6 kinase) family containing enzymes from 90,000to 92,000 Mr, denoted p90 rsk, initially isolated fromXenopus laevis, (Erikson and Mailer, 1985; 1986) and 60,000to 70,000-Mr kinase activities designated p70 s6K , purifiedfrom avian and mammalian sources (Blenis et al., 1987; Jenoet al., 1988; Price et al., 1989). Both enzymes are in turnactivated by serine/threonine phosphorylations in responseto mitogens, and are inactivated by protein phophatase 1 andphosphatase 2A (Ballou et al., 1988a; Price et al., 1990).The activated forms of both enzymes autophosphorylate onserine/threonine residues although this does not appear toalter kinase activity towards S6. When stimulated bymitogens, neither of the two enzymes containsphosphotyrosine. This suggests that serine/threonine kinasesin the upstream pathways may be responsible for theactivation of these S6 kinases. Intriguingly,phosphotyrosine has been detected in p90 rsk which wasactivated by the Rous sarcoma virus protein pp60 src (Sweetet al., 1990). The significance of this tyrosine19phosphorylation is unclear because of the absence ofphosphotyrosine in p90rsk stimulated with other mitogens.The structure of the p90 rsk enzymes is highly conservedin other species; cDNA clones have been used to identifysimilar enzymes expressed in chicken, mice and humans(Alcorta et al., 1989). These enzymes contain a unique twokinase catalytic domain motif. The N-terminal half of themolecule is related to protein kinase C, the catalyticsubunit of cAMP-dependent protein kinase and cGMP-dependentprotein kinase whereas, the C-terminal half is related tothe catalytic subunit of phosphorylase b kinase (Jones etal., 1988). While the functional significance of this motifremains unknown, it is conceivable that both domains arevital elements during the activation and function of thesekinases.p70S6K by contrast contains a single catalytic domain57% identical to the N-terminus of p90 rsk . The C-terminalregion of this enzyme shares some sequence homology with theS6 substrate itself and is proposed to present apseudosubstrate site (Banerjee et al., 1990) which mayprevent substrate-enzyme interaction and/or ATP binding. Itis also believed that phosphorylation of this region ofp7 0S6K by yet an unidentified mitogen-activated kinaserelieves the constrains of this inhibitory domain, leadingto activation of the enzyme (Price et al., 1991).Evidence for the regulation of p90rsk is also still veryscanty. Blenis and coworkers (Chung et al., 1991), have20recently reported the presence of a kinase activity in Swiss3T3 cells able to phosphorylate an inactive rsk-encodedprotein. It is believed that MAP kinases may be involved inthe upstream regulation of the p90 rsk kinases.Both purified S6 kinase enzymes phosphorylate 40Sribosomal subunits to high stoichiometries (4-5 mol P/molprotein), with high specific activities of hundreds ofnmol/min/mg (Kozma et al., 1990; Banerjee et al., 1990;Erikson and Mailer, 1985, 1986, 1988). Phosphopeptide mapsof S6 phosphorylated in vivo in response to mitogens and invitro by the purified enzymes are identical, implying thatboth kinases function in vivo as S6 kinases. p70s6K isprobably more specific for 40S ribosomes as a substratewhereas, p9O rsk phosphorylates several other substrates invitro with a consensus sequence of R-X-X-Ser (Erikson andMaller, 1988). Thus, the specificity of p90 rsk for S6 is notabsolute, implying a possible multifunctional, physiologicalrole for this enzyme.In terms of time-course, the activation of the S6kinases is biphasic or even multiphasic (Sweet et al., 1990;Ahn et al., 1990; Susa et al., 1989). p90 rsk is maximallyactivated within the first few minutes of mitogenicchallenge and thus, activation is often transient. Incontrast, the activation of p7 0S6K although initiatedrapidly, usually reaches a maximum in 30-60 minutes and theactivation is more sustained. According to some authors, themaintained activation of p70 s6K may require the activation21of protein kinase C (Susa et al., 1989), but this isdebated.Two partially purified EGF-stimulated MBP kinaseactivities from Swiss 3T3 cells identified as E3 and E4 withnative molecular weights of approximately 30 and 50 kDarespectively have been shown to stimulate an S6 peptidekinase activity (B1) in vitro that behaved as an 110 kDaprotein on Superose (Ahn and Krebs, 1990). Finally, usingrecombinant RSK peptide as an in vitro substrate, twochromatographically distinct, serum-activated RSK kinases,designated RSKI (p44) and RSK II (p42) were identified inSwiss 3T3 cells (Chung et al., 1991b). Both of these kinasesare tyrosine phosphorylated and are immunological related toa meiosis-activated protein kinase from maturing sea star(p44mPk). Furthermore, these protein kinases phosphorylatedand partially activated the ribosomal p90 rsk . The numbers ofand the precise inter-relationship of all these kinases toone another is still not entirely clear although therelationships are emerging rapidly as the cloning andsequencing of MAP kinase isoforms progresses.1.5.2 p42 and p44 MAP (MITOGEN-ACTIVATED PROTEIN) KINASES. Studies by Sato and coworkers (1985) demonstrated thatinsulin and other mitogens stimulated the phosphorylation ofcytoskeletal-associated proteins. Later studies usingmicrotubule-associated protein-2 (MAP-2) showed the presence22of kinase activities from cells stimulated by variousmitogenic agents (Ray and Sturgill, 1987; Hoshi et al.,1988). The acronym MAP initially taken from the preferred invitro substrate microtubule-associated protein-2, now standsfor mitogen-activated protein kinase, consistent with therelatively promiscous activation induced by diversemitogenic agents (Rossomando et al., 1989). At least twoclosely related murine MAP kinases are known to exist andthey exhibit apparent subunit Mr of 42,000 (p42 maPk) and44,000 (p44 maPk). Activation of these MAP kinases is rapidand transient reaching a maximum within the first 10 minutesof stimulation (Ahn et al., 1990; Ray and Sturgill, 1987;Ahn and Krebs, 1990) and subsequently becoming inactivewithin the next 60 minutes. Whereas, microtubule associatedprotein-2 and myelin basic protein are preferred substratesfor the MAP kinase, casein, histones, the S6 peptide, and40S ribosomal subunits are not appreciably phosphorylated(Ray and Sturgill, 1988; Erickson et al., 1990).M-phase arrested Xenopus eggs contain tyrosinephosphorylated 40-45 kD proteins (Cooper, 1989). Pelech andcoworkers have purified to homogeneity a 44 kDa MBP-kinase(designated p44 mPk for meiosis-activated protein kinase)from maturing sea star (Sanghera et al., 1990). p44mPk isimmunologically related to murine p42maPk and p44maPk .However based on molecular weight and copurification duringphenyl-Superose, polylysine and MonoQ chromatography p44mPk23and p44maPk seem even more closely related (Rossomando etal., 1991).In addition to those studies of protein kinases above,an additional molecular cloning approach by Cobb andcoworkers has led to the recognition of four or more proteinserine/threonine kinases (initially using a rat brain cDNAlibrary) which are highly related and termed "ERKs" forextracellular-signal regulated kinases (Boulton et al.,1991a; Boulton and Cobb, 1991). An insulin-stimulatedmicrotubule-associated protein-2 kinase sequenced by Boultonand coworkers termed ERK1 (Boulton et al., 1990) is alsoclosely related to p42 mal* and p44maPk. ERK1 has a subunitmolecular size of 43 kDa as judged by SDS-PAGE (Boulton etal., 1991b). In addition, it is inactivated completely byphosphatase 2A and substantially by CD45, suggestingregulation through serine/threonine and tyrosinephosphorylations. ACTIVATION OF 90 kDa S6 KINASE BY MAP KINASE. Although S6 kinases are activated by serine/threoninephosphorylations,^the kinases responsible for thisactivation have not been completely established. Mostintriguing is how signaling mechanisms from the insulinreceptor tyrosine kinase are linked with proteinserine/threonine kinases. A possible candidate for this linkis the family of MAP kinases. When phosphatase-inactivated24p 9 or sk from Xenopus oocytes was incubated with insulin-activated MAP kinase from 3T3-L1 cells, partial reactivationand phosphorylation of the enzyme was observed (Sturgill etal., 1988). Phosphorylation was found on threonine residues,at sites distinct from the p9Orsk autophosphorylation site,implying that these enzymes constitute a single step of aprotein serine/threonine kinase cascade. Although the MAPkinases do become insulin activated with parallel tyrosinephosphorylation a direct link of MAP kinase with thereceptor tyrosine kinase has, however, not been proved.Thus, purified MAP kinase has not been successfullyphosphorylated by preparations of insulin receptors.251.6 THESIS INVESTIGATION. The general hypothesis is that the actions of insulin oncell metabolism and function are mediated by mechanisminvolving reversible phosphorylation of proteins. It hasbeen recognized that insulin leads to increases in fattyacid biosynthesis through activation of several enzymesincluding ACC (which demonstrated insulin-inducedphosphorylation). Insulin activates protein-serine/threoninekinases in fat cells and this thesis describes my attempt topurify and characterize such insulin-activated fat cellprotein serine/threonine kinases. As model target substratesfor the kinases to be studied, myelin basic protein and ACChave been employed.Initial studies had revealed the presence of a number ofinsulin-sensitive serine/threonine kinases in extracts fromrat adipose tissue. Progress was severely hampered becausethe kinase activities proved to be very labile even afteremployment of a number of standard conditions for the stablestorage of proteins. This necessitated devising means ofstabilizing the kinase activities and particularly importantwas rapid initial partial purification and use of newprotein phosphatase inhibitors. In addition to the proteinpurification, parallel characterization employing severaldifferent antisera including antibodies, which recognizephosphotyrosine and antiserum raised against purified p44mPk26(a meiosis-activated protein kinase from sea star), hasprovided an important insight into the complexity of andsuccess in purification of the specific protein-serine/threonine kinases of interest. The development ofthis improved purification technique will provide a strongbasis for the future studies, which will enable furthermolecular characterization of the insulin-stimulated kinaseactivity.27MATERIALS AND METHODS. 1.1 MATERIALS. Male Wistar rats (140-160 g) supplied by the Universityof British Columbia (UBC), Animal Care Unit were housed inthe Department of Biochemistry 2-3 days prior to use. Theanimals maintained on a 12 hour light-dark cycle wereallowed free access to water and Purina rat chow until thetime of killing, usually between 8am and 9am.Most laboratory chemicals and solvents were obtainedfrom BDH Chemicals Canada Ltd. The biochemicals includingthe protease inhibitors (pepstatin A, leupeptin andbenzamidine) as well as bovine myelin basic protein (MBP),synthetic inhibitor of cAMP-dependent protein kinase andmouse monoclonal antiphosphotyrosine antibody (PT-66) werepurchased from Sigma Chemical Company (St. Louis, MO.,USA).Microcystin LR was bought from Calbiochem. Human MBP was agenerous gift from Frank Nezil and Dr. Myer Bloom of thePhysics department at UBC. The antibodies R1, R2, GEGA andanti-p44 mPk were generously supplied by Dr. S.L. Pelech atthe Biomedical Research Centre, UBC. Initial trial suppliesof antiphosphotyrosine antibodies and antiphosphotyrosine-agarose were kindly supplied by Alice Mui and Dr. JerryKrystal (Terry Fox lab., BC Cancer Control Agency,Vancouver, B.C.).28All radiochemicals including [gamma- 32P]ATP as well asACS scintillation fluid and the rainbow molecular weightmarkers for polyacrylamide gel electrophoresis (PAGE) werepurchased from Amersham International (Oakville, Ontario,Canada). Reagents for PAGE were bought from Bio-RadLaboratories (Canada) Ltd, Mississauga, Ontario.1.2 METHODS. 1.2.1 TISSUE INCUBATION AND PREPARATION OF EXTRACTS. Epididymal and perirenal fat pads were freshly isolatedfrom normally-fed male Wistar rats and immediately incubatedat 37°C in a bicarbonate buffered medium which had beengassed with 02:CO2 (19:1) for 20 min prior to warming to37°C. This was followed by an additional 10 min incubationin a fresh medium of the same composition with or withoutaddition of insulin (0.5 gg/ml). At the end of theincubation period, pads were removed from the medium,lightly and rapidly blotted on Whatman filter paper. Thetissue was then disrupted at 0°C with a Polytron PT-15-35tissue homogenizer (setting 6, 3s) into buffer, pH 7.2,containing 250 mM- sucrose, 50 mM-Mops, 40 mM-p-nitrophenylphosphate (pNPP), 5 mM-EDTA, 2 mM-EGTA, 1 mM-sodiumorthovanadate, 1 mM-DTT, 1AM-8-methylaspartic acid, 100 nM-microcystin LR, 1 mM-PMSF, 2.5 mM-benzamidine and 2 gg/ml29each of pepstatin A and leupeptin. The extraction used 4 mlbuffer per gram fresh weight of tissue.A soluble fraction was prepared from the adipose tissueextracts by centrifugation. This consisted of an initialcentrifugation at 15, 000 xg for 10 min (SS-34/Sorval). Thefat-free infranatant was further centrifuged at 360,000 xgavfor an additional 10 min in a benchtop ultracentrifuge (TLA100.1/Beckman) or alternatively using a conventionalultracentrifuge at 215,000 xgav for 1 hour (Ti 70/Beckman).The high speed supernatant (HSS) fraction obtained waseither used directly or was subsequently fractionated usingammonium sulphate or MonoQ ion exchange chromatography asdescribed below.1.2.2 AMMONIUM SULPHATE PRECIPITATION. Saturated ammonium sulphate solution (761 g/1; Scopes,1987) was adjusted to pH 7.5 at 4°C with Tris buffer. Anequal volume of this solution was added slowly with stirringto an equal volume of the HSS at 4°C. Stirring was continuedfor 1 h. This was followed by centrifugation at 27,000 xgfor 30 min (SS-34/Sorval). The pellets were eitherresuspended and tested directly or stored at -70°C untilfurther chromatography usually carried out the followingday. For resuspension, ammonium sulphate pellets weredissolved at 0°C in a minimum volume of appropriate bufferwith gentle mixing and after incubation at 0°C for 20-3030minutes were clarified by centrifugation (5 min, 10,000 xg,Eppendorf cetrifuge).1.2.3 ASSAY OF PROTEIN KINASE ACTIVITY. Protein kinase assays were carried out in a final volumeof 50 gl and contained 0.1 mg/ml MBP (or other proteinsubstrate as indicated), 5 mM-MgC12, 40 µM-[gamma- 321:]ATP(1000 dpm/pmole), 0.6 gM of the peptide inhibitor of cAMP-dependent protein kinase (3750 U/ml), 50 mM-Mops pH 7.5, 2mM-EGTA, 1 mM-DTT and 1 gM-B-methylaspartic acid. Allreaction preincubations were performed at 0°C. After briefincubation (1-2 min at 30°C), the kinase reactions wereinitiated upon addition of [gamma- 32P]ATP and proceeded for10 min at 30°C. The assay reactions were terminated byspotting 30 gl aliquots of reaction mixture onto Whatman P81phosphocellulose paper (Glass et al., 1978). The filterpapers were washed 10 times with gentle shaking on an icebath (over a total of approximately 2 h) in 0.85% phosphoricacid. Washed papers were then transferred into 20-m1 glassscintillation vials containing 10 ml distilled water and[ 32P] incorporation quantified by Cerenkov counting.Alternatively, some reactions were terminated by additionof 500 gl of pre-chilled acetone (-20°C). After incubationat -20°C for 2 h, precipitated proteins were recovered bycentrifugation (5 min, Eppendorf) and then subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE). The gels were31stained with Coomassie blue, destained and finally washedwith acetic acid (7% w/v) containing glycerol (4%) tostabilize and hence minimize cracking. The destained gelswere air-dried and subjected to autoradiography using KodakXAR film at -70°C in cassettes (Rigid form, Brooklyn, NY)containing Dupont Hi plus Cronex intensifier screens. Thiswas followed by densitometric scanning of the autoradiogramsusing a Biorad video densitometer model 620.1.2.4 SDS- POLYACRYLAMIDE GEL ELECTROPHORESIS (SDS-PAGE). Proteins were dissolved in 30-50 gl sample buffercontaining 62.5 mM-Tris pH 6.8, 5% SDS (w/v), 20% sucrose(w/v), 0.2 mg/ml bromophenol blue, 5% 2-mercaptoethanol(v/v) and separated on 1.5 mm thick slab gels by thediscontinuous buffer system of Laemmli (1970). Gels wereusually standard 7.5 (w/v) or gradient 7.5-15% (w/v) withrespect to acrylamide concentration and were run underconstant current 20 mA per gel. The gels were then fixed insolution containing trichloroacetic acid (20%, w/v) andmethanol (40%, w/v) for 10 min, stained with Coomassie blue-R250 in acetic acid (7% w/v), methanol (45% w/v) for 30minutes, destained (acetic acid:methanol, mixtures as above)and air-dried between clear membrane sheets (Bio-Rad), whichwere clipped to glass plates.321.2.5. SILVER STAINING OF GELS. Gels were soaked overnight in methanol (50% v/v) andthen washed in Milli-Q water with two changes over 30 min.The silver staining method was essentially adopted fromRabilloud et al., (1988). The gels were soaked for 1 min ina sensitizing solution containing sodium dithionite (0.25g/l) and rinsed in Milli-Q water for another minute. Thegels were then incubated in a staining solution containingsilver nitrate (0.2% w/v, in 1 mM formaldehyde) for 30 minwith shaking. After a final rinse in Milli-Q water the stainwas developed in a solution containing potassium carbonate(4% w/v) in 6 mM formaldehyde and 20 M sodium thiosulphate.After 10 min, the reaction was stopped by addition ofglacial acetic acid (3.5 ml/100 ml) directly to thedeveloper.1.2.6 AUTORADIOGRAPHY. Kodak X-ray film (X-Omat XAR-5) was preflashed andexposed at -70°C to polyacrylamide gels or TLC plates incassettes with intensifying screens (Dupont Cronex Hi-Plus).After exposure, the film was developed in an automaticdeveloper (Kodak M35A X-OMAT Processor).331.2.7 SCANNING DENSITOMETRY. Scanning densitometry of autoradiograms was performedusing a Biorad model 620 video densitometer and data wasanalyzed using the Biorad 1-D Analyst (version 2.0) computersoftware package.1.2.8 CHROMATOGRAPHY. In general, to facilitate rapid chromatographicseparation, a Fast Protein Liquid Chromatography (FPLC)system (Pharmacia) was used. The system employed anautomatic sample injection valve (MV-7), dual pumps forprogrammable isocratic or gradient elution and a detector toallow determination of the UV absorption of column effluent. DEAE-CELLULOSE ION EXCHANGE CHROMATOGRAPHY. High speed supernatant (HSS) fractions from rat adiposetissue stored at -70°C as ammonium sulphate pellets wereallowed to thaw and resuspended in 3 ml of buffer A (50 mM-Tris/HC1 pH 7.5 at 4°C, 1 gM B-methylaspartic acid, 2 mM-EGTA, 2.5 mM-benzamidine, 2 µg/ml each of pepstatin A andleupeptin, and 2 mM-DTT). The resuspended pellet was appliedto an Econo-Pac 10 DG desalting column (Bio-Rad) and thedesalted sample collected in a 5 ml-volume. The sample wasapplied to a DEAE-cellulose column (2.5 x 6 cm) at a flow3 4rate of 10 ml/h. After washing the column at 20 ml/h with 20ml of buffer A, the column was eluted at 10 ml/h with bufferA containing 0.5 M NaCl. Ten 2-ml fractions were collected.Fractions containing kinase activity were pooled,concentrated using a centricon 10 microconcentrator (Amicon)and stored at -70°C. PHENYL-SEPHAROSE CHROMATOGRAPHY. Ammonium sulfate was added to the DEAE-cellulose sampleup to 0.8 M and the sample applied to a phenyl-Sepharosecolumn (2.5 x 6 cm) previously equilibrated in buffer Acontaining 0.8 M ammonium sulphate. The column was washedwith 3-5 column volumes of the loading buffer and theneluted with three sequential steps: firstly with buffer A,secondly with buffer A containing 50% ethylene glycol andfinally with 2% (w/v) Nonidet P40 in buffer A. In somecases indicated in the results section, a different elutionprocedure was used in which the column was eluted only with3% Brij MONOO ION EXCHANGE CHROMATOGRAPHY. To ensure removal of all particulate material, thedesalted ammonium sulphate pellet was filtered through aSyrfil 25 mm 0.22 gm filter (Amicon) and applied to a MonoQanion exchange column (HR 5/5) in buffer A. To ensure the35removal of unbound protein, the column was washed with 5column volumes of buffer A at a flow rate of 1 ml/min. Thecolumn was developed by a linear gradient of 0-0.5 M NaC1also using buffer A and at the same flow rate. Forty 1-mlfractions were collected assayed for kinase activity andstored at at -20°C. AFFINITY CHROMATOGRAPHY USING MBP-AFFIGEL COLUMN. COUPLING OF MBP TO AFFIGEL 10. One milliliter of Affigel 10 (Bio-Rad) was washedsequentially with 3 ml each of isopropanol and colddeionized water. To the washed gel was added 5 mg of bovineMBP dissolved in 0.5 ml 50 mM Mops pH 7.5. The protein/gelmixture was incubated with shaking for 1 h at roomtemperature and following addition of 0.1 mlethanolamine.HC1, pH 8.0, the incubation was continued foranother 1 h in order to block unoccupied sites of theAffigel matrix. Blocking is particularly important when thegel is to be used immediately. After blocking withethanolamine, the gel was transferred to a column and theprotein solution, together with 3 subsequent washes withsolution containing 7 M urea/1 M NaC1 each were collected.The protein concentration in the wash fraction wasdetermined using the Bradford assay and on the basis of thisthe coupling efficiency was estimated to be 98%.361. MBP-COLUMN CHROMATOGRAPHY. A simple two-step batchwise elution procedure wasattempted to investigate the possible binding of proteinkinase activity to MBP-Affigel. One milliliter of proteinfraction containing MBP-kinase activity which had beenpartially purified by DEAE-anion exchange chromatography wasapplied to the MBP-column. After washing with 3 ml of bufferB (buffer B = buffer A with 50 mM Mops instead of 50 mMTris), the column was eluted with 0.5 ml buffer B containing0.5 M NaCl.1.2.9 WESTERN BLOTTING ANALYSIS. SDS-PAGE AND TRANSFER OF PROTEINS. One hundred microliter samples from each of thefractions obtained following chromatography with MonoQ,phenyl-Sepharose or other techniques as indicated wereprecipitated with 1 ml of cold acetone (-20°C). The sampleswere incubated at -20°C for 2 h and centrifuged for 5 min(Eppendorf). After removing the acetone by aspiration, thepellets were allowed to dry in order to remove the residualacetone. The pellets were dissolved in 30-50 gl of SDS-sample buffer by heating at 95°C for 5-10 min. The proteins37were separated on 7.5-15% gradient gels using SDS-PAGE asdescribed before.The separated proteins were electrophoretically transferredto Immobilon PVDF membrane (Millipore) in 20 mM-Tris/acetatepH 7.4, 2 mM-EDTA, 0.01% SDS (w/v) at 400 mA for 90 min. Theblots were washed briefly in KBS (8 g/1 NaC1, 0.2 g/1monobasic, potassium hydrogen phosphate, 1.93 g dibasic,sodium hydrogen phosphate, heptahydrate, 0.02% sodium azide(w/v), 0.2 g KC1 and 0.5 ml/L Tween 20), and then stained in0.2% Ponceau S / 3% TCA / 3% sulfosalicylic acid. The blotswere destained in KBS and blocked in KBS containing 0.5%Tween-20 for 1 h at room temperature or overnight at 4°C. ANTIBODY BINDING. The blots were removed from the blocking solution andwashed twice for 5 min each in KBS. Blots were probed withthe indicated antibodies for 2 h at room temperature. Theantibodies used included: affinity purified rabbitpolyclonal antibodies raised against purified sea starp44mPk (1:1000) (Sanghera et al., 1991), "R2," and "R1"rabbit antisera raised against the carboxy terminal region(with the sequence PFTFDMELDDLPKERLKELIFQETARFQPGAPEAP) andkinase subdomain III (with the sequencePFEHQTYCQRTLREIQILLGFRHENVIGIRDILRAP), respectively, of ERK1(1:1000), rabbit antiserum developed against a synthetic38peptide^(GEGA)^with the^sequence^GLAYIGEGAYGMVACcorresponding to protein subdomain I of sea star p44 mPk(1:1000) or anti-phosphotyrosine (PT-66, Sigma) (1:2000)for 2 h at room temperature. After washing with 5 changes ofKBS over 25 min, the blots were incubated with a secondaryantibody conjugated to alkaline phosphatase for 1 h. Thesecondary antibody was raised in goat against rabbit IgG.After washing 4 times in KBS as above and once in distilledwater, the membranes were incubated in substrate buffer(100 mM-NaCl, 5 mM-MgC12, 100 mM-Tris/HC1 pH 9.5), andfinally the blots were developed with bromochloroindolylphosphate/nitro blue tetrazolium (BCIP/NBT) substrate.1.2.10 IMMUNOPRECIPITATION. Two different kinds of mouse monoclonal antibodiesraised against phosphotyrosine, PT-AM (a gift from AliceMui, Terry Fox lab, Vancouver) and PT-66 (Sigma) were usedin an attempt to immunoprecipitate the MBP-kinase activity.200 Al of MBP-kinase activity partially purified by MonoQion exchange chromatography was incubated with 5-20 Al ofthe antibody overnight at 4°C. The incubation was performedin a buffer which contained 100 AM sodium orthovanadate. 10-50 Al recombinant protein G (Zymed) was then added and theincubation continued for 30 min. After the supernatant wasremoved, the protein G pellet was washed 3 times with 1 ml20 mM triethanolamine buffer pH 7.5, containing 0.5% (w/v)Nonidet P40 and 0.5M NaCl. After a final rinse in distilled39water, 50 Al of SDS sample buffer was added and the sampleheated at 95°C for 10 min. The proteins were separated usingSDS-PAGE followed by electrophoretic transfer to immobilonmembrane. The blot was probed with R2 and anti-p44 mPkantibodies.1.2.11 PROTEIN ASSAYS. Protein was assayed by the method of Bradford (1976).Two hundred microliter samples containing 1-20 Ag of proteinwere mixed with 1 ml of Bradford reagent and the colourallowed to develop at room temperature. After 15 min theabsorbance was measured at 595nm. Protein concentration wasestimated from a standard calibration curve, linear only upto 15 gg gamma-globulin, and which was prepared on eachseparate occassion. The values shown are the mean ofduplicate samples.40RESULTS1.0 INSTABILITY OF PROTEIN KINASE ACTIVITY IN EXTRACTS OFRAT FAT PAD TISSUE. Previous experiments showing the presence of insulinstimulated kinases were carried out using freshly-preparedfractions from rat adipose tissue. In such experiments ithad been possible to carry out rapid MonoQ chromatographyimmediately after preparation of high-speed supernatantfractions and then detect peaks of insulin-stimulatedprotein kinase activity directed against differentphosphoacceptor substrates. In these earlier experiments,all kinase assays were carried out on the same day as thetissue extractions. Further developments in the rapidpurification and characterization were retarded by the lossof protein kinase activity. The insulin-activated proteinserine/threonine kinases proved to be particularly labile,leading to loss of apparent insulin effect with storage offractions either at 4°C, 0°C, -20°C or -70°C as shown inTable 3. The decay in protein kinase activity occurreddespite inclusion of commonly used protein phosphataseinhibitors (EGTA, EDTA) and proteolytic enzyme inhibitors.Other common phosphatase inhibitors including sodiumfluoride and B-glycerolphosphate appeared to directlyinhibit adipose tissue kinase activities and were thereforeunsuitable. More specifically, extraction buffers those used for the preservation of ribosomal S6 kinases(used by Novak-Hofer and Thomas, 1984) also provedunsuccessful in stabilization. The immediate concern wastherefore, to devise means of preserving these insulinsensitive protein serine/threonine kinase activities ofinterest.42TABLE 3. THE LOSS OF INSULIN-STIMULATED KINASE ACTIVITYUNDER DIFFERENT STORAGE CONDITIONS.STORAGE CONDITION^ % LOSS OF ACTIVITY50% GLYCEROL, -70°C, 2 DAYS^ 924°C, 10 DAYS*^ 89CONCENTRATED, -70°C, 6 DAYS^ 67High speed supernatant fractions from rat adipose tissuewere fractionated with ammonium sulphate (50%), desalted andapplied on a MonoQ column. The column was developed with alinear NaC1 0-0.5 M gradient as described in Methodssection. Fractions were assayed for myelin basic proteinphosphotransferase activity before and after storage underdifferent conditions.*Similar loss is also found under these conditions within 1-2 days at 4°C.432.0 STABILIZATION OF INSULIN-ACTIVATED KINASES. 2.1 EFFECT OF PHOSPHATASE INHIBITORS. Within the last 3-4 years, it has become clear that theactivation of a number of mitogen-activated kinases can bereversed by treatment with purified protein phosphatases(Mailer, 1987; Ballou et al., 1988b; Anderson et al., 1990;Sanghera et al., 1991). These findings are consistent with arole of phosphorylation in the regulation of these kinasesand consequently when these kinases were purified, bufferscontaining phosphatase inhibitors were used. Thus, if theinsulin-activated kinases of adipose tissue are alsoregulated through protein phosphorylation, then the use ofbuffers containing protein phosphatase inhibitors would beexpected to have stabilizing effects. As noted above,initial studies had indeed taken account of this possibilitybut the precautions successful in other studies were notadequate in this case. Attempts were therefore made toinvestigate which of the phosphatase inhibitors were mosteffective in preserving the kinase activity in adiposetissue extracts. This was carried out in the following twoexperiments.442.1.1 ESTIMATION OF PHOSPHATASE ACTIVITY IN CRUDE EXTRACTSOF FAT PAD HIGH SPEED SUPERNATANTS. Fat pads treated with insulin were homogenized in abuffer containing Mops (50 mM), sucrose (250 mM), EGTA (2mM), EDTA (4 mM), DTT (2 mM) and the protease inhibitorspepstatin, leupeptin and PMSF. Variations of the abovebuffer were made by further additions of either 1 mM sodiumorthovanadate, 1 gM B-methylaspartic acid, or 10 mM B-glycerolphosphate. The breakdown of p-nitrophenylphosphate(pNPP) to p-nitrophenol with subsequent measurement at 405nm was used as an indicator of phosphatase activity.Unfortunately, this assay was not sensitive enough to detectchanges in phosphatase activity in the presence of differentinhibitors over time-course of less than 30-60 min. However,storage (even at 0°C) of supernatant fractions led to markedhydrolysis of pNPP over a period of several hours. For thepurposes of protein purification, therefore, the endogenousprotein phosphatases still displayed significant activity.452.1.2 EFFECTS OF MICROCYSTIN ON THE PHOSPHORYLATION OFMYELIN BASIC PROTEIN BY PROTEIN KINASES PRESENT IN RATADIPOSE TISSUE. It has been most fortuitous that two extremely potentinhibitors of protein phosphatases 1 and 2A (major fat cellprotein phophatases) have recently been discovered. Theseare okadaic acid and microcystin. The effects ofmicrocystin, a potent inhibitor of type I (PP1) and type 2A(PP2A) protein phosphatases (Honkanen et al., 1990) wereinvestigated in an experiment whereby the phosphorylation ofMBP by serine/threonine kinases present in fresh high-speedsupernatant fraction of rat adipose tissue was carried out.As shown in Figure la, the phosphorylation of MBP markedlyincreased in a microcystin dose-dependent manner. Sincemicrocystin inhibits protein phosphatase 1 and 2A, thisobservation implies that PP1 and PP2A may be the dominantprotein phosphatases acting on phosphoproteins in ratadipose tissue. The extent of phosphorylation wasquantitated by densitometric scanning and the results areillustrated in figure lb. Microcystin increased thephosphorylation of MBP by a factor of 2.5. The maximumphosphorylation was achieved at a microcystin concentrationof 100 nM.In view of the above findings, microcystin, B-methylasparticacid, B-glycerolphosphate, pNPP and orthovanadate weresubsequently included in the extraction buffer.46FIGURE la. EFFECT OF MICROCYSTIN ON PHOSPHORYLATION OF MBPBY PROTEIN KINASES PRESENT IN HIGH SPEED SUPERNATANTFRACTIONS.Rat fat pads were incubated with insulin (10 mU/m1 for 10min) and homogenized in the absence or presence of thephosphatase inhibitor microcystin at the indicatedconcentrations. Kinase assays were initiated by the additionof [gamma- 32P]ATP and allowed to proceed for 10 minutes. Thephosphoproteins were separated by SDS-PAGE followed byautoradiography. The autoradiograms shown, illustrate theincorporation of 32P into added MBP. The arrows indicate thetwo major bands of MBP which are phosphorylated.47FIGURE lb. EFFECT OF MICROCYSTIN ON THE PHOSPHORYLATION OFMBP.Rat fat pads were incubated with insulin (10 mU/ml for 10min) and homogenized in the absence or presence of thephosphatase inhibitor microcystin at various concentrationsas shown in Fig. la. Kinase assays were initiated by theaddition of [gamma- 32P]ATP and allowed to proceed for 10min. The phosphoproteins were separated by SDS-PAGE whichwas followed by autoradiography. J2P-incorporation into MBPwas assessed by densitometric scanning of the autoradiogramsas illustrated below: series 1 and 3 correspond to 32P-incorporation into lower and upper bands of MBPrespectively, series 2 corresponds to average values ofseries 1 and 3. In each case, the values are the average oftriplicate measurements in one experiment.Phosphorylation (% maximum)1201008060402000 nM^10 nM^100 nM^1000 nMMicrocystin ConcentrationMI Series 1 Series 2^P i Series 3482.2 AMMONIUM SULPHATE PRECIPITATION. To address further, the question of preserving thekinase activity, a rapid initial purification step employingammonium sulphate precipitation was considered. Thistechnique offers the benefits of speed, large volumecapability and a rather generalized stabilization ofproteins (Scopes, 1987). The high salt concentrationinhibits enzyme action and therefore, protects againstpossible proteolysis and/or phosphatase action. The actualpercentage saturation of ammonium sulphate required toprecipitate the kinase activity, was determined in a trialfractionation with three different cuts. As shown in Table4, the majority of the protein kinase activity precipitatedwith ammonium sulphate between 20-45%. The combination ofthe use of phosphatase inhibitors and rapid ammoniumsulphate precipitation was very effective in recovering andstabilizing protein kinase activity. Not only were theactivities enriched but they could be stored for weeks at -70°C with apparently very little loss in activity, as willbe demonstrated by the subsequent studies aimed at furtherpurification.49TABLE 4. PRECIPITATION OF MYELIN BASIC PROTEIN-KINASE.PERCENT^PERCENTSATURATION ENZYME^PURIFICATIONRANGE PRECIPITATED FACTOR0-20 4 <1.020-45 85 2.745-65 5 <1.065 supernatant 6 <1.0High speed supernatant fractions were prepared followinghomogenization of rat adipose tissue which had beenincubated with insulin (10 mU/ml for 10 min) and then werefractionated with 0-20, 20-45 and 45-65 % ammonium sulfate.The resulting protein fractions precipitated wereredissolved and then assayed for kinase activity able tophosphorylate myelin basic protein.503.0^PURIFICATION^OF^INSULIN-ACTIVATED^PROTEINSERINE/THREONINE KINASES FROM RAT ADIPOSE TISSUE. In the long term, a high priority is to obtain anadipocyte protein kinase preparation able to phosphorylatethe insulin-directed site (I-site) on acetyl-CoA carboxylase(Brownsey and Denton, 1982; 1987), since this has not beendescribed before. Furthermore, the relationship of thisactivity (or activities) with respect to those involved inphosphorylation of other insulin targets especially ATP-citrate lyase is also of major interest. The insulin-directed phosphorylation of acetyl-CoA carboxylase occursonly at one or two of eight possible phosphorylation sitesand therefore analysis requires very lengthy phosphopeptideseparation and quantitation. For convenience, we thereforechose to use myelin basic protein.(MBP) as a model substratefor protein kinase assays.The choice of this substrate was justified because(a) MBP (and MAP-2) appears to be an excellentsubstrate for mitogen-stimulated kinases in a number ofstudies.(b) a MBP kinase from sea-star oocytes has been shownto phophorylate acetyl-CoA carboxylase on the insulin-directed site (Pelech et al., 1991).Initially, our approach was to carry out the preliminaryisolation and stabilization of insulin-stimulated kinases asdescribed above (employing preparation of high-speed51supernatants followed by ammonium sulfate precipitation).Following this, a range of different columns were employedto test for their effectiveness in purifying kinases ofinterest. The objective was therefore to be able to design apurification protocol first, based on sequencial columnchromatography. Table 5 summarizes the various resins whichwere tested in an attempt to purify the insulin-activatedkinases. A number of these columns had pitfalls and althoughnot likely to be used in a final purification scheme, maynevertheless offer some important insights about thekinases. Chromatography on resins which provided separationand reasonable recovery of the activities of the insulin-activated kinases is now described below.52TABLE 5. THE VARIOUS CHROMATOGRAPHIC MATRICES USED INATTEMPTS TO PURIFY INSULIN-ACTIVATED PROTEINSERINE/THREONINE KINASES.MATRIXDEAE-CELLULOSEMONOQHYDROXYLAPATITETYROSINE-AGAROSECOMMENTUseful for rapid earlypurification.Excellent resolving powerand recovery of kinase activity.Good matrix but very slow flowrates.No distinct separation, enzymeactivity smeared in manyfractions.Fe-CHELATING^Low pH required to bind enzyme was(IDA-SEPHAROSE) denaturing.PHOSPHOCELLULOSE^Very weak interaction with thematrix.PHENYL-SEPHAROSE^Offered high recovery of enzymeactivity.MBP-AFFINITY^Not reproducible.SUPEROSE 6 Limited by the sample volume andSEPHACRYL S-200^enzyme diluted, but useful at theSEPHADEX G-100 qualitative level.POLYLYSINE-AGAROSE^Resolved two broad peaks ofkinase activity.HEPARIN-AGAROSE^Enzyme did not bind but manycontaminating proteins did.533.1 APPLICATION OF VARIOUS CHROMATOGRAPHIC TECHNIOUES FORPURIFICATION OF INSULIN-ACTIVATED SERINE/THREONINE KINASES FROM RAT ADIPOSE TISSUE. 3.1.1 MONOO CHROMATOGRAPHY OF INSULIN-ACTIVATEDSERINE/THREONINE KINASES FROM RAT ADIPOSE TISSUE. High speed supernatant fractions from rat adipose tissuetreated with or without insulin were subjected to 50%ammonium sulphate precipitation, desalted and applied to aMonoQ column. The column was washed with 5 column volumes ofthe equilibrating buffer. No detectable kinase activity waseluted in the flow-through and wash fractions. When thecolumn was developed by using a linear salt gradient withNaCl over the range 0-0.5 M, three peaks of MBP-kinaseactivities eluted at approximately 0.2 M, 0.3 M and 0.42 MNaC1 as shown in Fig. 2. The comparable fractions fromuntreated tissue (no insulin) contained very little MBP-kinase activity even though the amount of protein loadedonto the MonoQ column was very similar with the twodifferent extracts. Thus, the stimulatory effect of insulinon the protein kinases was preserved during thischromatographic step. Determination of the proteinconcentration in the NaCl-eluted fractions indicated thatthe kinase activities eluted after the major protein peak.^ 0.5• —• +insulin0-0 -insulin•**-0.4-0.3-0.2•/^•:\ - o• - • O54FIGURE 2. MONOQ CHROMATOGRAPHY OF INSULIN-ACTIVATEDSERINE/THREONINE KINASES FROM RAT ADIPOSE TISSUE.High speed supernatant fractions were prepared from ratadipose tissue which had been treated with (10 mU/m1 for 10min) or without insulin and were subjected to ammoniumsulphate precipitation, desalted using a Bio-Rad Econo-Pac10 DG desalting column and loaded onto a MonoQ column. Thefraction applied (4 ml) contained 1 mg protein per ml. Thecolumn was developed with a 30-m1 linear 0-0.5 M NaC1gradient. The column fractions were assayed for MBP-phosphorylating activity.1000_75o^50_c^\a^o_cCL^0_^250_0 -••••••••••••••••!00?-2--------1-04t°°°°°°°°o°°°°"^0.00^10^20^30^40FRACTION #•**553.1.2 DEAE-CELLULOSE ION EXCHANGE CHROMATOGRAPHYHigh speed supernatants precipitated with 50% ammoniumsulphate were desalted and applied to a DEAE column whichwas developed with a single step elution with buffercontaining NaCl at 0.5 M. Protein kinase activity able tophosphorylate MBP eluted in a single sharp peak as shown inFigure 3. This column did not resolve the different MBPkinases present in the fat pad extracts but did achieve arapid enrichment of kinases that was useful in some of thesubsequent column procedures which were attempted - namelythe use of phenyl-Sepharose, which requires the applicationof the protein fraction in the presence of high saltconcentrations.1000800 -02w a 600 -ccO EO w 400 -_ca 0m Eo a2000• —• +insulin0-0 -insulin 8^ 56FIGURE 3. DEAE-CELLULOSE ION EXCHANGE CHROMATOGRAPHY OFINSULIN-ACTIVATED SERINE/THREONINE KINASES FROM RAT ADIPOSETISSUE.High speed supernatant fractions were prepared from ratadipose tissue which had been treated with (10 mU/m1 for 10min) or without insulin and were subjected to ammoniumsulphate precipitation (50% saturation). The pellets werere-dissolved, desalted using a Bio-Rad 10DG Econo Pac columnand applied to a DEAE-cellulose column. The fraction applied(4 ml) contained 2 mg of protein per ml. After washing offunbound proteins, the column was eluted with 0.5 M NaC1 and2-ml fractions collected. The figure below shows the proteinkinase activity assessed by phosphorylation of MBP.3^5^9FRACTION #573.1.3 PHENYL-SEPHAROSE CHROMATOGRAPHY. In addition to ionic interactions, hydrophobicity wasanother physiochemical property of proteins which weconsidered might be exploited in the logical design of apurification scheme. The separation of proteins usinghydrophobic interaction (HIC) is based on a differentprinciple from most other separation techniques and canthus, in combination with these other methods, afford a highdegree of purification. In addition HIC is generally a mildmethod due to the stabilizing influence of salts andrecoveries are often high (Eriksson, 1989). It is atechnique which has proved very valuable in other studies ofmitogen-activated protein kinases.Rat adipose tissue MBP-kinase activity partially purifiedby DEAE-cellulose ion exchange chromatography was furtherfractionated using HIC using a phenyl-Sepharose column (2.5x 6 cm). The sample was applied in a buffer containing 0.8 Mammonium sulphate to promote the hydrophobic nature of theproteins. The column was eluted stepwise using a sequenceof buffers as follows: firstly, a buffer identical to theapplication buffer except for the omission of ammoniumsulphate, secondly, the same buffer containing ethyleneglycol (50%, w/v) and finally with buffer containing thedetergent Nonidet P40 (NP40, 1.5%, w/v). Three peaks ofkinase activities were eluted under these three differentelution conditions as shown in Figure 4.58FIGURE 4. PHENYL-SEPHAROSE CHROMATOGRAPHY OF INSULIN-ACTIVATED SERINE/THREONINE KINASES FROM RAT ADIPOSE TISSUE.High speed supernatant fractions were prepared from adiposetissue which had been treated with or without insulin (10mU/m1 for 10 min) and were then applied to a DEAE-cellulosecolumn. Protein kinase activity was eluted with a single-step buffer change (0.5 M NaC1). Ammonium sulphate (0.8M)was added to the pooled DEAE-cellulose fractions exhibitingMBP-phosphorylation activity. This pooled fraction 4 ml,containing 1.5 mg/ml protein was applied to a phenyl-Sepharose column and eluted sequentially as follows: (a)buffer identical to the application buffer except for theomission of ammonium sulphate (fractions 1-10), (b) the samebuffer containing 50% (v/v) ethylene glycol (fractions 11-20) and finally (c) with buffer containing the detergentNonidet P40 (1.5%, w/v; fractions 21-30).40>,._>-.7,,o,^•^ •Q_^ 17)E\• /a.) •(8 0.ci__c ,_, ./\\^A^000••I 0011^21^31FRACTION #ELo ----. 30 -w Ea)^a)cp_2 20 -o4-. cc• —• +insulin •0-0 -insulina b c/I •Io.. 10 -a_co ...,, 0,..0 0,, /0•oor 00!(!lmf P••tp.,,,,000593.2^PURIFICATION^OF^INSULIN-ACTIVATED^PROTEINSERINEJTHREONINE KINASES USING SEQUENTIAL CHROMATOGRAPHYWITH POLYLYSINE-AGAROSE, MONO() AND HEPARIN-AGAROSE COLUMNS. A high speed supernatant fraction was prepared from ratadipose tissue (obtained from 18 rats) which had beentreated with insulin and this supernatant was precipitatedwith ammonium sulphate as described above. The proteinpellet precipitated by ammonium sulfate (50% saturation) wasresuspended, diluted to a volume of 200 ml and applied to apolylysine-agarose column (2.5 x 6 cm) which was developedwith a linear salt gradient using NaCl over the range 0.2-0.6 M. Two broad peaks of MBP kinase activities eluted withpeak maxima eluting at approximately 0.23 M and 0.43 M NaCl(Fig. 5). These peaks are now designated PLI and PLIIrespectively, (polylysine peaks I and II respectively).Pooled fractions from PLI and PLII were diluted andseparately applied to a MonoQ column. The column wasdeveloped with a linear 0-0.8M NaCl gradient. In order notto exceed the capacity of the column, the pooled sampleswere applied in three separate runs and the column fractionscollected in the same set of tubes three times. The elutionprofiles of the kinase activities are shown in Fig. 6. Inboth cases, sharp peaks designated MQI and MQII wereobtained. MQI and MQII are the protein kinase activity peaksobtained by fractionating PLI and PLII, respectively, on theMonoQ column.60The kinase activity in MQII was more tightly bound to boththe polylysine-agarose and the MonoQ columns. Based onwestern blotting analysis to be described in detail below,MQI was further purified using heparin-agarose and then afinal MonoQ column fractionation (Fig. 7) and a silverstained gel of the final MonoQ preparation is illustrated inFig. 8.- 0.5 • —0.4/ ••0• •fbio^•^•0I^ 0.1•••••^0- 0.2 °61FIGURE 5. CHROMATOGRAPHY OF INSULIN-STIMULATED PROTEINSERINE/THREONINE KINASES ON POLYLYSINE-AGAROSE.High speed supernatant fractions from 25g insulin-treatedtissue (10 mU/m1 for 10 min) were precipitated usingammonium sulphate (50% saturation). The pellet wasresuspended, diluted to a volume of 200m1 and applied to apolylysine-agarose column which was developed with a linearsalt gradient using NaCl over the range 0.2-0.6 M (dottedline). Four-milliliter fractions were collected and assayedfor MBP-phosphorylating activity.2000 0.6NO I< 15000}25• '65 10000_c= —E5000a_• 0•0^10^20^30^40^50^60^70^80FRACTION #062FIGURE 6. CHROMATOGRAPHY OF INSULIN-STIMULATED PROTEINSERINE/THREONINE KINASES ON MONO Q.Active fractions containing MBP-kinase activity from thepolylysine-agarose illustrated in Fig. 5 were pooledseparately and applied to a MonoQ column. After washing offunbound protein, the column was eluted with a linear 0-0.8 MNaC1 gradient.(24-34)pooled fractionsPanel A,from polylysine(50-60)chromatography of pooled fractionspeak 1; panel B, chromatographyfrom polylysine peak 2.of0.8000or°mI2000A•20 1500 -0.60017)^xc2 Et-c^'1" 0- 1000 —0.400600 .cE •1. z0m a_ 500 - • 0.200* • • 00 ••----.111=-11-1-1—!"•°^••••••••••••••••••• 0.000• 0 0 1 0 20 30^40^50^60^70^80FRACTION #• 200047,0a)ca2 ° 1500q)• xc2o 1000_c0_ 0-o.Em a 500o•0•0.800- 0.600- 0.400• - 0.200k••■•.•.•• ►^I• 1Pe •••••••••♦•••••••••• 0.00010^20^30^40^50^60^70^80B0FRACTION #15000w IO 0wX 1000C2 EO S--C^(2)o_ 0-o• S= E 500co 0_o0.8000.6000.400 8(7)000.200063FIGURE 7. CHROMATOGRAPHY OF INSULIN-STIMULATED PROTEINSERINE/THREONINE KINASES ON MONO Q.Active fractions from sequential chromatography onpolylysine-agarose (peak 1), MonoQ (Fig. 6, panel A),heparin-agarose (results not shown) were finallyfractionated on MonoQ. The column was developed with alinear 0-0.8 M NaC1 gradient at a flow rate of 1 ml/min. Twohundred and fifty microliter fractions were collected andassayed for MBP-phosphorylating activity.00•••10^20•30^40^50FRACTION #0.00080•••60 7064FIGURE 8 SDS-POLYACRYLAMIDE GEL ELECTROPHORESIS OF PURIFIEDINSULIN-STIMULATED PROTEIN SERINE/THREONINE KINASE.One hundred and fifty microliters of each of the indicatedfractions from the MonoQ column illustrated in Fig. 7 wereprecipitated with prechilled acetone (-20°C). The recoveredprotein pellets were dissolved in 200 pl SDS-digestionbuffer and 40 pl of this sample electrophoresed on agradient (7.5-15%) polyacrylamide gel. The gel was stainedwith silver nitrate. Lane 32 contained the most highlypurified enzyme which corresponded with the peak of kinaseactivity. The arrow shows the position of the 44 kDa band.The position of the molecular weight markers (in kDa) isindicated.30^31^32 33^34200►97►46►30■22 ►654.0 WESTERN BLOTTING ANALYSES. Further characterization of the protein kinase activitiesseparated on different columns was carried out using Westernblotting analyses. Two of the antibodies used were raisedagainst peptide antigens based on the primary sequenceswithin different domains of the ERK1 kinase. Antiserum wasraised against "R2" within the domain near the C-terminus ofERK1 while "R1" was directed towards a sequence surroundingkinase subdomain III of ERK1. The anti-p44 mPk antibody wasraised against purified p44 mPk, a sea-star proteinserine/threonine kinase which was shown to phosphorylateacetyl-CoA carboxylase apparently on the insulin directedsite. (Pelech et al., 1991). This antiserum was subsequentlypurified by immunoaffinity chromatography. In addition, twoother antibodies were used, one was a monoclonalantiphosphotyrosine antibody, the other was designated"GEGA" antibody. The GEGA antiserum is antiserum raisedagainst a peptide in p44mPk which corresponds to part of theATP-binding region in conserved kinase subdomain I. Thesequence is generally designated GXGXXG and forms part ofthe nucleotide-binding fold of all known protein kinases. Itwas anticipated that the use of these antibodies wouldascertain whether homologs of ERKs and/or p44 mPk existed inextracts of insulin-stimulated rat adipose tissue extractsand the extent to which any immunoreactive bands copurifiedwith insulin-stimulated protein kinase activity. In addition66the GEGA and phosphotyrosine antibodies would confirmwhether the crossreacting proteins were kinases and/ortyrosine phosphorylated. Although fractions were analyzedfor immunoreactive bands throughout all the gradients, manyof the fractions which showed little or no reactivity areomitted for clarity.4.1 IMMUNOBLOTTING WITH ANTIBODY RAISED AGAINST SEA-STARD44mPk .The anti-p44 mPk antibody reacted with polypeptides ofapproximate molecular weights 40, 42, 44 and 70 kDa (Fig.9). These peptides were well resolved during MonoQ anionexchange chromatography that was shown in Fig 2. The 44 kDaprotein band in particular, copurified with the first peakof protein kinase activity from that column corresponding tofractions 22, 23 and 24. In contrast, the bands of 40 and 42kDa emerged in fractions 25 26 and 27 with a peak aroundfraction 26 and the 70 kD band appeared last of all, infractions 29,30 and 31. These observations suggest at least3 protein kinase activities which are separable by Mono Q,which are all related in some way (immunologically) top44mPk and which display distinct subunit Mr values.67FIGURE 9. CHROMATOGRAPHY OF INSULIN-STIMULATED PROTEINSERINE/THREONINE KINASES ON MONO Q AND IMMUNOBLOTTING WITHVARIOUS ANTIBODIES.Chromatography was performed as described in Fig. 2. Onehundred microliter samples from each fraction wereprecipitated with prechilled acetone (-20°C) and therecovered protein pellets dissolved in 40 Al SDS-digestionbuffer. These were then electrophoresed on gradient (7.5-15%) polyacrylamide gels. The separated proteins were thentransferred to immobilon membrane. Finally, the blots wereprobed with the anti-p44 mPk (p44), antiphosphotyrosine (PY),R2, R1 and GEGA (GE) antibodies as shown below. Arrows 1 and2 indicate the positions of the 44 and 42 kDa bands,respectively. The bars show fractions within the peaks ofMBP-kinase activity. The position of the molecular weightmarkers (in kDa) is indicated.19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 n 39 40 4168When fractions from phenyl-Sepharose column were probedwith anti-p44m1* antibody, all the above mentionedpolypeptides were detected. Most, if not all immunoreactiveprotein bound to the phenyl-Sepharose and very littleemerged in the void fraction. Indeed, very littleimmunoreactive protein emerged in the wash treatment usinglow salt buffer. When the column was eluted with 1.5% Brij35, a single peak of protein kinase activity was recovered.All the immunoreactive polypeptides copurified within thispeak of kinase activity. In contrast when the column waseluted sequentially with 50% ethylene glycol andsubsequently with 1% NP40, separation of these polypeptideswas achieved. The 44 kDa as well as a faint 42 kDa band weredetected within the ethylene glycol eluant. The 70 kDpolypeptide was identified within the fractions eluted withthe detergent.It is therefore clear that the immunoreactive proteinsand insulin-stimulated protein kinases do indeed bind verytightly to phenyl-Sepharose. Specific elution conditions canbe developed to achieve some separation of the differentimmunoreactive polypeptides recognized by anti-p44 mPkantiserum. It is worth pointing out that even if this latterresolution is not achieved, the phenyl-Sepharose columnnevertheless, provides considerable purification due to theremoval of a large amount of unbound protein in the voidfraction combined with the low-salt wash.69Polylysine-agarose chromatography resolved two peaks ofkinase activities as already described. The first peak wasshown to contain a 44 kDa polypeptide which crossreactedwith p44mPk (Fig. 10). Consequently, the polylysine-agarosefractions containing the kinase activity within the firstpeak were subjected to further fractionation employing threeconsecutive column steps - MonoQ, heparin-agarose (resultsnot shown) and finally a second MonoQ column. The anti-p44mPk antibody recognized a 44 kDa protein band whichcopurified with protein kinase activity (Fig. 11) throughoutthese three further purification steps.The second polylysine peak contained immunoreactivepolypeptides of approximate molecular weight 65 and 75 kDwhich reacted with the anti-p44 mPk antibody (Fig. 12). Basedon these results, polylysine-agarose fractions from thesecond peak containing protein kinase activity were notfurther purified, at the present time although this will beof interest in future studies.197146130R2 70FIGURE 10. IMMUNOBLOTTING OF INSULIN-ACTIVATED RINASESPARTIALLY PURIFIED ON POLYLYSINE-AGAROSE.Pooled fractions from the first peak of polylysinechromatography were further fractionated on MonoQ as shownin Fig. 6A. Two hundred microliter samples from each of theindicated fractions were precipitated with 1 ml ofprechilled (-20°C) acetone. The recovered pellets weredissolved in SDS-digestion buffer and electrophoresed ongradient (7.5-15%) polyacrylamide gels. The separatedproteins were transferred to immobilon membrane. The blotswere probed with the anti-p44nk (p44) and R2 antibodies asshown below. The peak of MBP-kinase activity was in fraction32 (Fig. 6A). The position of the molecular weight standards(s) in kDa is indicated.24 25 2627 28 29 30 31 32 33 34 35p444 9744643030 31 32 33 S71FIGURE 11. IMMUNOBLOTTING OF PURIFIED INSULIN-ACTIVATEDPROTEIN SERINE/THREONINE KINASE.High speed supernatant fractions from insulin-treatedadipose tissue (10 mU/ml for 10 min) were subjected tosequential purification steps to enrich for insulin-stimulated MBP-kinase. The steps involved were: polylysine-agarose (Fig. 5), MonoQ (Fig. 6A), heparin-agarose (resultsnot shown) and MonoQ (Fig. 7) and the results of the finalMonoQ step are illustrated here. One hundred and fiftymicroliter volumes from each of the indicated fractions fromthe final MonoQ column were precipitated with 1 ml ofprechilled (-20°C) acetone. The precipitated proteins weredissolved in 200 Al of SDS-digestion buffer and 40A1/laneelectrophoresed on gradient gels (7.5-15%). The separatedproteins were then transferred to immobilon and probed withthe anti-p44 mPk (p44), R1, R2, antiphosphotyrosine (PY) andGEGA (GE) antibodies as shown below. The position of themolecular weight standards (S, in kDa) is also indicated.S 12 13 14 15 16 S 30 31 32 33 34p442009746■ 30■ 14-4 200■ 97■ 46-4 30-4114R1 R2 PY GE S 72FIGURE 12. IMMUNOBLOTTING OF FRACTIONS FROM POLYLYSINE PEAK2.High speed supernatant fractions from insulin-treatedadipose tissue (10 mU/ml) were applied to polylysine-agaroseand fractions from peak 2 were pooled and furtherfractionated on the MonoQ as shown in Fig. 6B. Two hundredmicroliter samples from the latter purification step wereprecipitated with 1 ml of prechilled acetone (-20°C). Therecovered proteins were dissolved in 200 gl of SDS-digestionbuffer. Forty microliter samples from each of the indicatedfractions were electrophoresed on polyacrylamide gels andsubsequently transferred to immobilon membranes. The blotswere probed with the anti-p44 mPk (p44) and R2 antibodies asshown below. The position of the molecular weight markers(in kDa) is indicated.28 29 30 3132 33 34 35 36 3738 39 40ANTI-p4435 36 37 38 39 401 97R2^ has130734.2 IMMUNOBLOTTING WITH R2 ANTIBODY. R2 antibody reacted very strongly with a 40/42 kDadoublet in protein fractions separated on the MonoQ column(Fig 9). It appeared not to crossreact with the 44 kDa bandwhich was detected using anti-p44 mPk antibody. Furthermore,this doublet was well resolved during the denaturing SDS-PAGE analysis from the 44 kDa band. Perhaps paradoxically,the R2 antibody did appear to react with the 44 polypeptidewhich copurified with protein kinase activity which emergedin the earlier of the two peaks observed during polylysine-agarose chromatography. Therefore, I obtained apparentlydifferent reactivity between the fat pad 44 kDa band and theanti-p44mPk antibody depending upon the purification stepused.4.3 IMMUNOBLOTTING WITH Rl ANTIBODY. The Ri antibody also reacted with the 42 and 40 kDapolypeptides in protein fractions separated by Mono Q ionexchange chromatography (Fig. 9). This doublet wasespecially prominent in fractions 25 and 26. Incidentally,these are the same fractions where the 40/42 kD doubletreacted very strongly with R2 antibody. In contrast to R2,R1 reacted with a 70 kD band which was also detected byanti-p44mPk antibody. As with the R2 antibody, Ri reacted74with the 44 kDa band in the most highly purified fractioncontaining protein kinase activity (Fig. 11).4.4 IMMUNOBLOTTING WITH ANTIPHOSPHOTYROSINE ANTIBODY.Since other MAP kinases have been shown to be regulated bytyrosine phosphorylation, probing with anti-phosphotyrosineantibodies was an important avenue to persue. When proteinfractions separated by MonoQ anion exchange chromatographywere probed with the antiphosphotyrosine antibody, only the44 kDa and the upper band of the 42 kDa doublet weredetected (Fig. 9). In contrast, neither the 40 kDa nor the70 kDa bands showed any appreciable labelling in response toanti-phosphotyrosine antibodies. The antiphosphotyrosineantibody also reacted with the 44 kD band in the most highlypurified fraction containing protein kinase activity (Fig.11) showing that the tyrosine phosphorylation had beenpreserved at least partially throughout the extensivepurification.4.5 IMMUNOBLOTTING WITH GEGA ANTIBODY. The GEGA antibody crossreacted with the 42 kDa doublet andto a lesser extent with the 44 kDa band in MonoQ fractions(Fig. 9). It also reacted with a 44 kDa band in the purifiedfraction containing protein kinase activity (Fig. 11). It is75surprising that the 70 kDa band apparently reacted onlyweakly, if at all, with the GEGA antibody. Nevertheless, thecross reactivity of the 44 kDa band with anti-p44 mPk , anti-phosphotyrosine and GEGA argue strongly that the most highlypurified preparation does indeed contain an insulin-stimulated protein ser/thr kinase highly-related to otherMAP kinases.765.0 ADDITIONAL CHARACTERIZATION OF RAT ADIPOSE TISSUESER/THR KINASES. 5.1 INHIBITION OF ENDOGENOUS PROTEIN PHOSPHORYLATION BYADDED MYELIN BASIC PROTEIN. High speed supernatant fractions prepared from fat padstreated with insulin, were incubated with [gamma- 32P]ATP inthe presence or absence of purified myelin basic protein.The phosphoproteins were separated by SDS-PAGE which wasfollowed by autoradiography. As shown in Figure 13, additionof MBP specifically decreased 32P-incorporation into ATP-citrate lyase (ATP), another insulin-directed protein,suggesting a possible interaction with the ACL-kinase. Inthese experiments, very little competition with acetyl-CoAcarboxylase was observed. In a number of previousexperiments, employing other protein kinase substrates(including histones and casein) no such interactions wereobserved. Furthermore, when bradykinin (a model substratefor proline-directed protein kinases) was used instead of32MBP P-incorporation into ACL was not inhibited.77FIGURE 13. INHIBITION OF PROTEIN PHOSPHORYLATION IN HIGHSPEED SUPERNATANT FRACTIONS FROM RAT ADIPOSE TISSUE BY ADDEDMYELIN BASIC PROTEINHigh speed supernatant fractions from insulin treated fatpads (10 mU/m1 10 miA) were i ncubated (10 min., 30 C) in thepresence of [gamma- J4P]ATP-Mg with (lanes 1,3) or without(lanes 2,4) added myelin basic protein (5mg/m1). Theproteins were precipitated with acetone and separated bySDS-PAGE, followed by autoradiography. The figure shows thecoomassie blue stained gel (A) and the autoradiogram (B).The position of the molecular weight standards as well asthat of acetyl-CoA carboxylase (ACC) and ATP-citrate lyase(ACL) are indicated.A785.2 IMMUNOPRECIPITATION OF MBP-KINASE ACTIVITY. Attempts were made to immunoprecipitate MBP-kinase usingR2 and anti-p44mPk antibodies. Very little apparent decreasein supernant MBP-kinase activity was observed. Theimmunoprecipitates were analyzed by SDS-PAGE and showed amultitude of bands. It was therefore, difficult to draw anydefinitive conclusions.Another attempt at immunoprecipitating MBP-kinase wasmade using anti-phosphotyrosine antibodies. Only partialprecipitation of polypeptides of subunit molecular mass 45,42 and 30 kDa was achieved. When the immunoprecipitates wereblotted with anti-p44 mPk polypeptide bands of subunitmolecular weight of 45, 42 and 30 kDa were recognized. TheR2 antibody reacted with the 42, 40 and 30 kDa bands. Thisobservation lends further support to the idea that the 44kDa band reacts weakly with R2 antibody compared to 42 and40 kDa bands.79DISCUSSION. Within seconds after binding of insulin at the cellsurface, the B-subunit of the insulin receptor displaysincreased auto-phosphorylation on tyrosine residues.Autophosphorylation leads to enhanced protein-tyrosinekinase activity towards other cellular proteins - thefunctions of which are still being actively persued.Following these initial events, a wide range of subsequentresponses occur within target cells. Among these subsequenteffects of insulin action are rapid increases inphosphorylation of serine and/or threonine (ser/thr)residues on target proteins like ACC, ACL and the S6ribosomal protein. These changes in protein phosphorylationoccur as a result of the activation of protein-serine/threonine kinases which appear to serve as criticalsignaling intermediates, and as such their identificationand characterization is important in the elucidation of themultiple steps of the pathway, initiated by receptoractivation.Studies of insulin-activated ser/thr kinases has beentechnically simplified by using substrates which are moreconvenient than the anticipated more physiological substrateenzymes themselves, which must be isolated usually bylaborious, lengthy procedures. In addition, these cellularsubstrate enzymes often (like acetyl-CoA carboxylase, ofinterest in these studies) contain multiple phosphorylation80sites which mean that the determination of sitesphosphorylated would require additional 2-dimensionalphosphopeptide analysis. This would be extremely time-consuming and really not feasible as a routine assay to beused for following kinase activity during purificationprocedures.In these studies, myelin basic protein (MBP) was used asa model substrate for the insulin-activated kinases. Afterinitial studies with MBP purified by another research groupat UBC (Dr. Myer Bloom, Physics), MBP has becomecommercially available (Sigma). MBP binds strongly to P-81phosphocellulose paper and therefore is ideally suited forfilter paper assays. This choice was justified because MBPappears to be an excellent substrate for mitogen-stimulatedkinases in a number of studies and indeed was found tosuccessfully detect insulin-activated protein kinase inextracts of rat adipose tissue. Furthermore, a MBP-kinasefrom sea-star was shown to phosphorylate acetyl-CoAcarboxylase on the insulin-directed site.The tissues most commonly used to study insulin effectsare muscle, liver, adipose and lactating mammary glandswhich are all sensitive to the hormone. Of the four tissues,adipose tissue can be most easily manipulated, in vitro andshows the most striking changes in metabolism following invitro incubations with insulin. The major points addressedin these studies were preserving the insulin-activated81Ser/Thr kinases from adipose tissue, and then subsequentpurification and characterization using various antibodies.STABILIZATION OF INSULIN-ACTIVATED MBP-KINASES. Initial experiments to study insulin-stimulated Ser/Thrkinases were carried out using freshly-prepared high speedsupernatant fractions. Further progress in the purificationand characterization of these kinases of interest washampered by the rapid loss in enzyme activity upon storage.The insulin-stimulated Ser/Thr kinases proved to be labileunder many commonly used storage conditions such as at 0°C,4°C, -20°C or -70°C and despite the use of stabilizingagents such as glycerol, proteinase inhibitors and certainphosphatase inhibitors. The immediate concern was therefore,to try to stabilize these insulin-activated protein Ser/Thrkinases. This was ultimately achieved by employing a rapidprocedure which included ammonium sulphate precipitation andwhich stressed the use of a cocktail of phosphataseinhibitors in the extraction buffer.Several advantages are achieved by using ammoniumsulphate precipitation. Firstly large sample volumes couldbe employed and dual control and insulin-treated sampleseasily treated in parallel. Secondly, the resulting pelletscould be dissolved in a small amount of buffer, thusincreasing the protein concentration of the sample andprobably aiding in stabilization. Finally, the high salt82concentration employed is often inhibitory to enzyme action,therefore, the precipitated proteins could be stored in astable form with likely minimal proteolytic action. Inaddition, to these benefits of rapid precipitation withammonium sulfate, the stability of the insulin-activatedprotein kinases was aided by the presence of phosphataseinhibitors in the extraction buffer. In particular, aturning point seemed to be the adoption of a particularlypotent inhibitor of protein phosphatases I and 2A, namelymicrocystin. This observation reinforces the possibilitythat the insulin-activated kinases may in turn be regulatedby phosphorylation, a feature which has certainly emerged asbeing highly significant in the studies of other so-called"mitogen-activated kinases".PURIFICATION OF INSULIN-ACTIVATED MBP-KINASES. Having now overcome the problem of instability of thekinases the next priority was purification andcharacterization of the insulin-stimulated Ser/Thr kinases.Purification of the polylysine MBP-kinase was achieved usingammonium sulfate precipitation followed by sequentialchromatography on polylysine-agarose, MonoQ, heparin-agaroseand finally Mono Q for a second time. The silver-stained gelof the final purified fraction contained a major 44 kDa bandwhich copurified with the peak of MBP-kinase activity.83Assignment of this protein band as a kinase wassupported by its comigration with Ser/Thr kinase activity,immunoreactivity with antibodies which recognize ERK1 andp44mPk kinases. In addition the immunoreactivity of the 44kDa band with the GEGA antibody in particular, lends furthersupport to this conclusion, since it recognizes one of thehighly conserved regions involved in ATP-binding found inall known kinases and which includes the concesus GXGXXGmotif. The preparation of the insulin-activated proteinserine/threonine kinase so far achieved is not homogenous.One major (high molecular weight) band and several minorbands are still present and may need to be removed. However,based on immunoreactivity with the various antibodies aswell as the homogeneity of chromatographic peaks of kinaseactivity, it seems unlikely that any other protein kinasesremain in the preparation. Significantly, the gel isremarkably "clean" below 44 kDa, suggesting the completeremoval of p42 and other lower molecular kinases. Though notabsolutely homogenous, this preparation may yet besufficient for detailed analysis.It will be necessary to confirm that the 44 kDa band isitself a single protein by running 2-dimensional gelanalysis. If indeed this is the case, then the preparationcan immediately be scaled up to attempt protein sequencingdirectly from the gel. Another graduate student in our lab(Bob Winz) has already achieved this for another proteinwhile working under the guidance of Dr. Aebersold (BRC) and84so we are fortunate that these techniques are available.This is perhaps the single most critical objective.Following either N-terminal or partial internal sequencing,all the necessary information would be available to initiatemolecular cloning. This would therefore, allow us tounambigously establish the relationships of this importantkinase to other members of the family.RELATIONSHIP OF ADIPOSE TISSUE P44 MBP-KINASE TO OTHERINSULIN-STIMULATED PROTEIN SERINE/THREONINE KINASES. The p44 kDa MBP-kinase is clearly distinct from the S6kinases by virtue of its molecular weight and substratepreference. The S6 kinases are larger enzymes, exhibitingsubunit molecular weights of around 67-70 and 90 kDa underdenaturing conditions employed for SDS-PAGE, contrastingwith 44 kDa for MBP-kinase. In addition, chromatographicproperties of these enzymes are different. S6 kinases bindavidly to anion exchangers and therefore require higher saltconcentrations for elution. Furthermore, S6 kinases bind toheparin columns whereas, the p44 MBP-kinase described heredoes not bind at all to such a column.An insulin-stimulated protein-serine kinase (Kemptidekinase) from the cytosol of rat adipocytes has been reportedby Czech and coworkers (Klarlund et al., 1990). However, theinsulin-stimulated ser/thr kinase reported in this studyappears to be unrelated to this cytosolic kinase. This is85because of the difference in molecular weights. The MBP-kinase has a subunit molecular size of 44 kDa whereas, thatof the Kemptide kinase (KIK) is 52 kDa. With respect tosubstrate specificity, KIK exhibits preference to histone,kemptide and ACL. The site phosphorylated in kemptide hasthe sequence LRRASLG which is different from the substrateconcensus sequence of MAP kinases. The latter phosphorylateSer/Thr residues in close proximity to proline residues suchas Thr 97 in bovine MBP (PRTPPP, Erickson et al., 1990;Sanghera et al., 1990a; Clark-Lewis et al., 1991). In lightof the close resemblance of p44 MBP-kinase to other MAP-kinases, it is more likely that p44 MBP-kinasepreferentially recognizes sequence motifs with prolinedeterminants. Another striking difference between KIK andMBP-kinase is that the latter reacts with anti-phosphotyrosine antibodies whereas, KIK lacks thisimmunoreactivity. A further distinguishing feature betweenp44 MBP-kinase and KIK is the ability of KIK to bind toheparin columns.Casein kinase II (CKII) is another Ser/Thr kinasereported to be stimulated by insulin. However, p44 MBP-kinase differs markedly in both substrate specificity andstructural properties from CKII which exists as aheterotetramer of a and B subunits. CKII is "acidotropic",in that it phosphorylates sites in motifs containing eitherGlu, Asp or Ser(P) such as in the CKII peptide (-EEETEEE;Roach, 1991).86From the evidence we so far have available, it seemslikely that p44-MBP-kinase is related to a family of proteinSer/Thr kinases known as ERKs or MAP kinases. Thisresemblance is based on a range of properties includingimmunological, chromatographic, subunit molecular weight andphosphorylation on tyrosine residues. Similarities betweenp44 MBP-kinase and the recently purified ERK1 kinase includesubstrate specifity, molecular weight and crossreactivitywith R1, R2, anti-p44mPk and anti-phosphotyrosineantibodies. However, a subtle difference existed when R2 andR1 antibodies (raised against peptides patterned against theC-terminal and subdomain III regions in ERK1) were used toprobe blots which contained the 44, 42 and 40 kDapolypeptide bands. Only the 42 and 40 kDa bands wererecognized. Surprisingly, these antibodies appeared not toreact with the 44 kDa band corresponding to MBP-kinase. Itis possible that p44 MBP-kinase contains sequencedifferences that reduce its recognition by R2 antibody. Inlight of this, it would appear that ERK1 is more closelyrelated to the 42 and 40 kDA polypeptides than it is to p44MBP-kinase. A definitive assignment of how much ERK1 and p44MBP-kinase are related however,^awaits^furthercharacterization of MBP-kinase particularly determination ofthe primary sequence.Other members of the ERK family include ERK2 and ERK3.The predicted molecular weights of these kinases are 41.2for ERK2 and 62.6 kDA for ERK3, contrasting with 44 kDa for87MBP-kinase. It is therefore unlikely that the MBP-kinasecorresponds to either ERK3 or ERK2.In addition to the ERKs which are derived from molecularcloning using a rat brain cDNA library, other relatedkinases have been described, mostly using murine cells. Thenow-classical studies of Ray and Sturgill reported aninsulin-activated ser/thr kinase from 3T3-L1 cells(p42maPk), recognized by the ability to phosphorylate MAP-2.This kinase has a subunit molecular weight of 42 kDa and ishighly related to the ERKs, most likely the murineequivalent of ERK2. The published amino acid sequence of themurine p42 maPk (Her et al., 1991) is identical to that ofERK2 (Boulton et al., 1991) which supports the idea thatp42maPk is the murine equivalent of ERK2. There also appearsto be a murine kinase with subunit molecular weight of 44kDa which is highly related to the sea-star p44 mPk andpresumably very closely related to the rat enzyme studiedhere.Similar "pairs" of kinases of subunit molecular weight44 and 42 kDa have been recognized in murine 3T3 cells inresponse to EGF (Ahn and Krebs, 1990) and in growth-factorstimulated KB cells. In the latter case, the stimulatedkinase activity was shown to account for most of thephosphorylation of the EGF receptor (Northwood et al.,Alvarez et al., 1991). Studies with growth-factor stimulatedSwiss 3T3 cells have also demonstrated the existence of twoRSK protein kinases. Both RSK kinase I (44 kDa) and II (4288kDa) are related to sea-star p44 mPk and can partiallyactivate p9O rsk in vitro. Together, these findings areconsistent with the existence of a family of mitogenactivated protein-serine/threonine kinases. Further progresstoward the identification of the relationships between theseenzymes will require the molecular cloning of all of thesefamily members.Overall, one may deduce that the family of ERK/MAPkinases is complex. At least there appear to be distinct 44and 42 kDa proteins (and these differ somewhat betweenspecies as well as within species). In addition, we alsoobserve a 40 kDa band, so that we cannot tell yet therelationship of the ERKs with any of the kinases in ratadipose tissue.INHIBITION OF ENDOGENOUS PROTEIN PHOSPHORYLATION BY ADDEDMYELIN BASIC PROTEIN. Initial "competition" studies using purified MBP as acompetitive substrate in high speed supernatant fractionincubations indeed suggested a decline of phosphorylation ofACL and to a lesser extent ACC. Overall, however, afterseveral more experiments it seems this effect is notstriking. We have no clear understanding why some resultswere more positive than others. The only real difference wasin the species of origin of MBP, being human (supplied byFrank Nezil, Physics Dept., UBC) versus commercial bovine89MBP. It is possible that some inter-species differences inprimary sequence might explain the different results.IMMUNOPRECIPITATION.Attempts at immunoprecipitation of MBP-kinase from high-speed supernatant fractions with R2 and anti-p44 mPk was notsuccessful. It is possible that these antibodies have a lowaffinity for the MBP-kinase and therefore could notsuccessfully precipitate the kinase. Certainly, it is notunusual for antibodies to successfully recognize antigens onWestern blotting while producing little or noimmunoprecipitation of native antigen from solution.Presumably, critical epitopes are exposed followingdenaturation. In addition, if the antibody/antigen complexeswere formed as a result of single point attachments ratherthan multimeric, then immunocomplexes would have easilyremained in solution. This also suggests that forimmunoaffinity purification of active kinase, the anti-phosphotyrosine antibodies would be the best of thepresently available choices.Unlike R2 and anti-p44mPk antibodies, the anti-phosphotyrosine was able to precipitate polypeptides ofsubunit molecular weight 44, 42, and 40 kDa. However, therewas only partial recovery of these bands in theimmunoprecipitates in comparison to the residualsupernatants. This would be consistent with either partial90tyrosine phosphorylation, weak immunoreactivity or both ofthese phenomena. Since several different anti-phosphotyrosine antibodies are commercially available, itwould be worth checking the merits of each of them.TIME-COURSE/DOSE RESPONSE FOR MBP-KINASE ACTIVATION. Trials to determine the kinetics of activation of theMBP-kinase in response to insulin have been carried out butfurther work is required. High speed supernatant fractionsof rat adipose tissue contain several kinases able tophosphorylate MBP. It is therefore difficult to determinethe contribution made by the specific insulin-activatedkinases of interest with a simple assay procedure. It isimportant to stress that a rapid and simple procedure isnecessary to facilitate determination of activation of MBP-kinase in the multiple fractions generated during a time-course or dose-response study. In order to circumvent thisproblem Western blotting analysis using anti-phosphotyrosineantibodies was undertaken. This was based on the assumptionthat activation of the MBP-kinases is accompanied bytyrosine phosphorylation. The results obtained following gelanalysis of proteins in high speed supernatant fractionswere still complex and it was difficult to draw any firmconclusions. These experiments however, helped us to confirmthat the 44 kDa protein band is indeed a significanttyrosine-phosphorylated protein in insulin-treated fat91.tissue. When both^high speed supernatant and membranefractions were immunoblotted with the anti-phosphotyrosineantibodies, a "rich" assortment of bands was highlighted,suggesting miltiple proteins became tyrosine-phosphorylatedin response to insulin. A particularly, prominent 30 kDprotein band was labelled with anti-phosphotyrosineantibody. This intriguing, major band remained in theammonium sulphate supernatant when the high speedsupernatant extract was subjected to ammonium sulphateprecipitation (50% w/v). Since this band was alsoimmunoreactive when tested with R2 and anti-p44 mPkantibodies, it would appear to be itself a protein kinaseperhaps, related to p34 cdc2 , a protein-serine/threoninekinase that functions in the M-phase of the cell cycle. Itis clear, therefore, that to accurately follow the dose- andtime-dependency of p44 tyrosine phosphorylation,purification beyond the high speed supernatant or evenammonium sulphate precipitation will be required. Thepatterns are still too complex at this stage for easyinterpretation. It seems that rapid phenyl-Sepharosepurification will be one important possibility to be tested,as used by Anderson and coworkers (1991). Further, thesestudies have opened up an important avenue for furtherinvestigation namely the apparent complexity of thephosphotyrosine profile, the preponderance of membraneproteins affected and the striking effect on a 30 kDaprotein.92CONCLUSION. The present study has demonstrated the presence of aninsulin-stimulated protein-Ser/Thr kinase in rat adiposetissue. The MBP-kinase is very similar to other mammalianmitogen-activated kinases in terms of molecular weight,immunological properties and substrate preference. Thissuggests that the MBP-kinase studied here in response toinsulin belongs to the same family of Ser/Thr kinasesdescribed as MAP kinases. The finding that related kinasesexist in sea-star oocytes, yeast and Xenopus indicates howhighly conserved this family is, through evolution. Thisunderlies the likelihood of a crucial role for these kinasesin cell regulation, extending from regulation of cell cycle(meiosis, mitosis) to analogous responses to hormones inquiescent/fully differentiated mammalian cells.93REFERENCESAhman, Z., Lee, F-T., DePaoli-Roach, A., Roach, P.J. (1984):Phosphorylation of glycogen synthase by the Ca 2 '- andphospholipid-activated protein kinase (protein kinaseC). J. Biol. Chem. 259, 8743-8747.Ahn, N.G. and Krebs, E.G. (1990): Evidence for an epidermalgrowth factor-stimulated protein kinase cascade inSwiss 3T3 cells: J. Biol. Chem. 265, 11495-11501.Ahn, N.G., Weiel, J.E., Chan, C.P. and Krebs, E.G. (1990):Identification of multiple epidermal growth factor-stimulated protein serine/threonine kinases from Swiss3T3 cells. J. Biol. Chem. 265, 11487-11494.Alcorta, D.A., Crews, C.M., Sweet, L.J., Bankston, L.,Jones, S.W. and Erikson, R.L. (1989): Sequence andexpression of chicken and mouse rsk: homologs ofXenopus laevis ribosomal S6 kinase. Mol. Cell Biol. 9,3850-3859.Alexander, M.C. Kowaloff, E.M., Witters, L.A., Dennihy, D.T.and Avruch, J. (1979): Purification of a hapatic123,000-dalton hormone stimulated 32P-peptide and itsidentification as ATP-citrate lyase. J. Biol. Chem.254, 8052-8056.Alvarez, E., Northwood, I.C., Gonzalez, F.A., Latour, D.A.,Seth, A., Abate, C., Curran, T. and Davis, R.J.(1991): Pro-Leu-Ser/Thr-Pro is a consensus primarysequence for substrate protein phosphorylation.Characterization of the phosphorylation of c-myc andc-jun proteins by an epidermal growth factor receptorthreonine 669 protein kinase. J. Biol. Chem. 266,15277-15285.Anderson, N.G., Kilgour E. and Sturgill T.W. (1991):Activation of mitogen-activated protein kinase in BCmyocytes by fluoroaluminate. J. Biol. Chem. 266,10131-10135.Anderson, N.G., Mailer, J.L., Tonks, N.K. and Sturgill, T.W.(1990): Requirement for integration of signals fromtwo distinct phosphorylation pathways for activationof MAP kinase. Nature 343, 651-653.94Augert, G. and Exton, J.H. (1988): Insulin and oxytocineffects on phosphoinositide metabolism in adipocytes.J. Biol. Chem. 263, 3600-3609.Avruch, J., Tornqvist, H.E., Gunsalus, J.R., Yurkow, E.J.,Kyriakis, J.M., Price, D.J. (1990): Insulin regulationof protein phosphorylation. In Handbook ofExperimental Pharmacology. Vol. 92 Cuatrecasas P.,Jacobs, S., Eds., Berlin, Springer-Verlag, p. 313-366.Ballou, L.M., Jeno, P. and Thomas, G. (1988b): Proteinphosphatase 2A inactivates the mitogen-stimulated S6kinase from Swiss mouse 3T3 cells. J. Biol. Chem. 263,1188-1194.Ballou, L.M., Siegmann, M. and Thomas, G. (1988a): S6 kinasein quiescent Swiss mouse 3T3 cells is activated byphosphorylation in response to serum treatment. Proc.Natl. Acad. Sci. U.S.A. 85, 7154-7158.Banerjee, P., Ahmad, M., Grove, J.R., Kozlosky, C., Price,D.J. and Avruch, J. (1990): Molecular structure of amajor insulin/mitogen-activated 70-kDa S6 proteinkinase Proc. Natl. Acad. Sci. U.S.A. 87, 8550-8554.Belsham, G.J. and Denton, R.M. (1980): The effect of insulinand adrenaline on the phosphorylation of a 22,000-molecular weight protein within isolated fat cells;possible identification as the inhibitor-1 of the'general phosphatase'. Biochem. Soc. Trans. 8, 382-383.Belsham, G.J., Denton, R.M. and Tanner, M.J.A. (1980): Useof a novel rapid preparation of fat-cell plasmamembranes employing percoll to investigate the effectsof insulin and adrenaline on membrane proteinphosphorylation within intact fat-cells. Biochem. J.192, 457-467.Blackshear, P.J., Haupt, D.M., App, H., Rapp, U.R. (1990):Insulin activates the Raf-1 protein kinase J. Biol.Chem. 265, 12131-34.Blackshear, P.J., Nemenoff, R.A., Hovis, J.G., Halsey, D.L.,Stumpo, D.J., Huang, J.-K. (1987): Insulin action innormal and protein kinase C-deficient rat hepatomacells: effects on protein phosphorylation, proteinkinase activities and ornithine decarboxylaseactivities and messenger ribonucleic acid levels. Mol.End. 1, 44-52.95Blenis, J., Kuo, C.J. and Erikson, R.L. (1987):Identification of a ribosomal protein S6 kinaseregulated by transformation and growth-promotingstimuli. J. Biol. Chem. 262, 14373-14376.Bommer, U.-A., Noll, F., Lutsch, G. and Bielka, H. (1980):Immunochemical detection of proteins in the smallsubunit of rat liver ribosomes involved in binding ofthe ternary initiation complex. FEBS Lett. 111, 171-174.Boulton, T.G. and Cobb, M.H. (1991): Identification ofmultiple extracellular signal-regulated kinases (ERKs)with antipeptide antibodies. Cell Regulation 2, 357-371.Boulton, T.G., Gregory, J.S. and Cobb, M.H. (1991b):Purification and properties of extracellular signan-regulated kinase 1. and insulin-stimulatedmicrotubule-associated protein 2 kinase. Biochemistry30, 278-286.Boulton, T.G., Nye, S.H., Robbins, D.J., Ip, N.Y.,Radziejewska, E., Morgenbesser, S.D., DePinho, R.A.,Panayotatos, N., Cobb, M.H. and Yancopoulos, G.D.(1991a): ERKs: A family of protein/serine kinases thatare activated and tyrosine phosphorylated in responseto insulin and NGF. Cell 65, 663-675.Boulton, T.G., Yancopoulos, G.D., Gregory, J.S., Slaughter,C., Moomaw, C., Hsu, J. and Cobb, M.H. (1990): Aninsulin-stimulated protein kinase similar to yeastkinases involved in cell cycle control. Science 249,64-67.Bradford, M.M. (1976): A rapid and sensitive method for thequantitation of microgram quantities of proteinutilizing the principle of protein-dye binding. Anal.Biochem. 72, 248-254.Brownsey, R.W. and Denton, R.M. (1982) Evidence that insulinactivates fat cell acetyl-CoA carboxylase by increasedphosphorylation at a specific site. Biochem. J. 202,77-86.Brownsey, R.W. and Denton, R.M. (1987): Acetyl-CoAcarboxylase. In The Enzymes, 3rd ed., vol.18, pp. 123-146.Brownsey, R.W., Hughes, W.A. Denton, R.M. and Mayer, R.J.(1977): Demonstration of the phosphorylation ofacetyl-CoA carboxylase within intact rat epididymalfat cells. Biochem. J. 168, 441-445.96Burdett, E., Mills, G.B., Klip, A. (1990): Effect of GTP Son insulin binding and tyrosine phosphorylation inliver membranes and L6 muscle cells. Am. J. Physiol.258, C99-108.Chou, C.K., Dull, T.J., Russel, D.S., Gherzi, R., Lebwohl,D., Ullrich, A. and Rosen, O.M. (1987): Human insulinreceptors mutated at the ATP-binding site lack proteintyrosine kinase activity and fail to mediatepostreceptor effects of insulin. J. Biol. Chem. 262,1842-1847.Chung, J., Chen, R.-H and Blenis. (1991a): J. Coordinateregulation of pp90 and a distinct protein-serine/threonine kinase activity that phosphorylatesrecombinant pp90 in vitro. Mol. Cell. Biol. 11, 1868-1874.Chung, J., Pelech, S.L., and Blenis, J. (1991b): Mitogen-activated Swiss mouse 3T3 RSK kinases I and II arerelated to pp44 from sea star oocytes and participatein the regulation of pp90 activity. Proc. Natl. Acad.Sci. USA 88,4981-4985.Clark-Lewis, I., Sanghera, J.S. and Pelech, S.L. (1991):Definition of a consensus segupnce for peptidesubstrate recognition by p44 1u1D , the meiosis-activatedmyelin basic protein kinase. J. Biol. Chem. 266,15180-15184.Cobb, M.H. (1986): An insulin-stimulated ribosomal proteinS6 kinase in 3T3-L1 cells. J. Biol. Chem. 261, 12994-12999.Cooper, D.R., Hernandez, H., Kuo, J.Y., Farese, R.V. (1990):Insulin increases the sythesis of phospholipid anddiacylglycerol and protein kinase C activity in rathepatocytes. Arch. Biochem. Biophys. 276, 486-494.Cooper, J.A. (1989): Related proteins are phosphorylated ontyrosine in response to mitogenic stimuli and atmeiosis. Mol. Cell Biol. 9, 3143-3147.Coore, H.G., Denton, R.M., Martin, B.R. and Randle P.J.(1971): Regulation of adipose tissue pyruvatedehydrogenase by insulin and other hormones. Biochem.J. 125, 115-127.Czech, M.P. (1977): Molecular basis of insulin action. Ann.Rev. Biochem. 46, 359-384.97Czech, M.P., Klarlund, J.K., Yagaloff, K.A., Bradford, A.P.,Lewis, R.E. (1988): Insulin receptor signalling:activation of multiple serine kinases. J. Biol. Chem.263, 11017-11020.Davies, S.P., Alastair, T.R.S. and Hardie, G. (1990):Location and function of three sites phosphorylated onrat acetyl-CoA carboxylase by the AMP-activatedprotein kinase. Eur. J. Biochem. 187, 183-190.Davis, H.W., and McDonald, J.M. (1990): Insulin receptorfunction is inhibited by guanosine 5'-[ -thio]triphosphate (GTP[S]). Biochem. J. 270, 401-407.Degerman, E., Smith, C.J., Tornqvist, H., Vasta, V.,Belfrage, P., Manganiello, V.C.(1990): Evidence thatinsulin and isoprenaline activate the cDMP-inhibitedlow Km cAMP phosphodiesterase. Proc. Natl. Acad.Sci.USA 87, 533-537.Dent, P., Lavoinne, A. Nakielny, S., Caudwell, F.B., Watt,P. and Cohen, P. (1990): The molecular mechanism bywhich insulin stimulates glycogen synthesis inmammalian skeletal muscle. Nature 348, 302-308.Denton, R.M. (1986): Early events in insulin action. InAdvances in Cyclic Nucleotide and ProteinPhosphorylation Research Vol.20, Greengard P.Robinson G.A., Eds. New York Raven p. 293-342.Denton, R.M., Brownsey, R.W. and Belsham G.J. (1981): Apartial view of the mechanism of insulin action.Diabetologia 21, 347-363.Denton, R.M., McCormack, J.G., Marshall, S.E. (1984):Persistence of the effect of insulin on pyruvatedehydrogenase activity in rat white and brown adiposetissue during the preparation and subsequentincubation of mitochondria. Biochem. J. 217, 441-452.Ellis, L., Clausner, E., Morgan, D.O., Edery, M., Roth, R.A.and Rutter, W.J. (1986): Replacement of insulinreceptor tyrosine residues 1162 and 1163 compromisesinsulin-stimulated kinase activity and uptake of 2-deoxyglucose. Cell 45, 721.Endemann, G., Yonezawa, K., Roth, R.A. (1990):Phosphatidylinositol kinase or an associated proteinis a substrate for the insulin receptor tyrosinekinase. J. Biol. Chem. 265, 396-400.98Erickson, A.K., Payne, D.M., Martino, P.A., Rossomando,A.J., Shabanowitz, J., Weber, M.J., Hunt, D.F. andSturgill T.W. (1990): Identification by massspectrometry of threonine 97 in bovine myelin basicprotein as a specific phosphorylation site formitogen-activated protein kinase. J. Biol. Chem. 265,19728-19735.Erikson, E. and Mailer, J.L. (1985): A protein kinase fromXenopus eggs specific for ribosomal protein S6. Proc.Natl. Acad. Sci. USA 82, 742-746.Erikson, E. and Mailer, J.L. (1986): Purification andcharacterization of a protein kinase from Xenopus eggshighly specific for ribosomal protein S6. J. Biol.Chem. 261, 350-355.Erikson, E. and Mailer, J.L. (1988): Sec. Messeng.Phosphoprot. 12, 135-143.Eriksson, K.-0. (1989): in Protein Purification ed. byJanson, J.-C. and Ryden, L., VCH, pp. 207-225.Farese, R.U., Standaert, M.L., Barnes, D.E., David, J.S.,Pollet, R.J. (1985): Phorbol ester provokes insulin-like effects on glucose transport, amino acid uptakeand pyruvate dehydrogenase activity in BC3H1 culturedmyocytes. Endocrinology 116, 2650-2653.Glass, D.B., Masaracchia, R.A., Feramisco, J.R., Kemp, B.E.(1978): Isolation of phosphorylated peptides andproteins on ion exchange papers. Anal. Biochem 87,566-575.Graves, C.B. and McDonald, J.M. (1985): Insulin and phorbolester stimulated phosphorylation of a 40 kDa proteinin adipocyte plasma membranes. J. Biol. Chem. 260,11286-11292.Halestrap, A.P. and Denton R.M. (1973): Insulin and theregulation of adipose tissue acetyl-Coenzyme Acarboxylase. Biochem J. 132, 509-517.Hawley, D.M., Maddux, B.A., Patel, R.G., Wong, K.-Y, Mamula,P.W., Firestone, G.L., Brunetti, A., Verspohl, E.,Goldfine, I.D. (1990): Insulin receptor monoclonalantibodies that mimic insulin action withoutactivating tyrosine kinase. J. Biol. Chem. 264, 2438-2444.99Haystead, T.A.J., Moore, F., Cohen, P. and Hardie, G. (1990)Roles of the AMP-activated and cyclic-AMP-dependentprotein kinases in the adrenaline-induced inactivationof acetyl-CoA carboxylase in rat adipocytes. Eur. J.Biochem. 187, 199-205.Her, J.-H., Wu, J., Rall, T.B., Sturgill, T.W. and Weber,M.J. (1991): Sequence of pp42/MAP kinase, aserine/threonine kinase regulated by tyrosinephosphorylation. Nucleic Acids Research 19, 3743.Hershey, J.W.B. (1991): Translational control in mammaliancells. Annu. Rev. Biochem. 60, 717-755.Honkanen, R.E., Zwiller, J., Moore, R.E., Daily, S.L.,Khatra, B.S., Dukelow, M. and Boynton, A.L. (1990):Characterization of microcystin-LR, a potent inhibitorof type 1 and type 2A protein phosphatases. J. Biol.Chem. 265, 19401-19404.Hoshi, M., Nishida, E. and Sakai, H. (1988): Activation of aCa2+-inhibitable protein kinase that phosphorylatesmicrotubule-assiciated protein 2 in vitro by growthfactors, phorbol esters, and serum in quiescentcultured human fibroblasts. J. Biol. Chem. 263, 5396-5401.Hughes, W.A. and Denton, R.M. (1976) Incorporation of 32Pinto pyruvate dehydrogenase phosphate in mitochondriafrom control and insulin-treated adipose tissue.Nature 264, 471-473.Jeno, P., Ballou, L.M., Novak-Hofer, I. and Thomas, G.(1988): Identification and characterization of amitogen-activated S6 kinase. Proc. Natl. Acad. Sci.USA 85, 406-410.Jones, S.W., Erikson E., Blenis, J., Maller, J.L. andErikson R.L. (1988): A Xenopus ribosomal protein S6kinase has two apparent kinase domains that are eachsimilar to distinct protein kinases. Proc. Natl. Acad.Sci. U.S.A. 85, 3377-3381.Kahn, C.R. and White, M.F. (1988): The insulin receptor andthe molecular mechanisms of insulin action. J. Clin.Invest. 82, 1151-1156.Kasuga, M., Zick, Y., Blithe, D.L., Karlson, F.A., Haring,H.U. and Kahn, C.R. (1982): Insulin stimulates thephosphorylation of the 95,000-dalton subunit of itsown receptor. J. Biol. Chem. 257, 9891-9894.100Kasuga, M., Izumi, T., Tobe, K., Shiba, T., Momomura, K.,Tashiro-Hashimoto, Y., Kadowaki, T. (1990): Substratesfor insulin-receptor kinase Diabetes Care 13, 317-326.Klarlund, J.K., Bradford, A.P., Milla, M.G. and Czech, M.P.(1990): Purification of a novel insulin-stimulatedprotein kinase from rat liver. J. Biol. Chem. 265,227-234.Kovacina, K.S., Yonezawa, K., Brantigan, D.L., Tonks, N.K.,Rapp, U.R., Roth, R.A. (1990): Insulin activates thekinase activity of the Raf-1 proto-oncogene byincreasing its serine phosphorylation. J. Biol. Chem.265, 12115-12118.Kozma, S.C. Ferrari, S. and Thomas, G. (1989): CellSignalling 3, 219-225.Kozma, S.C., Ferrari, S., Bassand, P., Siegman, M., Totty,N. and Thomas, G. (1990): Cloning of the mitogen-activated S6 kinase from rat liver reveals an enzymeof the second messenger subfamily. Proc. Natl. Acad.Sci. USA 87, 7365-7369.Krieg, J., Hofsteenge, J. and Thomas, G. (1988):Identification of the 40S ribosomal protein S6phosphorylation sites induced by cycloheximide. J.Biol. Chem. 263, 11473-11477.Laemmli, U.K. (1970): Cleavage of structural proteins duringthe assembly of the head of bacteriophage T4. Nature227, 680-685.Lavoinne, A., Erikson, E., Mailer, J.L., Price, D.J.,Avruch, J. and Cohen P. (1991): Purification andcharacterisation of the insulin-stimulated proteinkinase from rabbit skeletal nuscle; Close similarityto S6 kinase II. Eur. J. Biochem. 199, 723-728.Lee, K.H., Thrall, T. and Kim K.H. (1973): Hormonalregulation of acetyl-CoA carboxylase. Effect ofinsulin and epenephrine. Biochem. Biophys. Res.Commun. 54, 1133-1138.Mailer, J.L. (1987): J. Cyclic Nucleotide Protein Phospho.Res. 11, 543-555.Martin, B.R., Denton, R.M., Pask, H. and Randle, P.J.(1972): Mechanisms regulating adipose tissue pyruvatedehydrogenase. Biochem. J. 129, 763-773.101Mato, J.M., Kelly, K.L., Abler, A., Jarett, L., Corkey,B.E., Cashel, J.A., Zopf, D. (1987): Partial structureof an insulin-sensitive glycophospholipid. Biochem.Biophys. Res. Commun. 146, 764-770.McClain, D.A., Maegawa, H., Lee, J., Dull, T.J., Ullrich, A.and Olefsky, J.M. (1987): A mutant insulin receptorwith defective tyrosine kinase displays no biologicactivity and does not undergo endocytosis. J. Biol.Chem. 262, 14663-14671.Northwood, I.C., Gonzalez, F.A., Wartmann, M., Raden, D.L.,Davies, R.J. (1991): Isolation and characterization oftwo growth-stimulated protein kinases thatphosphorylate the epidermal growth factor receptor atthreonine 669. J.Biol. Chem. 266, 15266-15276.Novak-Hofer and Thomas (1984): An activated S6 kinase inextracts from serum- and epidermal growth factor-stimulated Swiss 3T3 cells. J. Biol. Chem. 259, 5995-6000.Olefsky, J.M. (1990): The insulin receptor: amultifunctional protein. Diabetes 39, 1009-1016.Palen, E. and Traugh, J.A. (1987): Phosphorylation ofribosomal S6 by cAMP-dependent protein kinase andmitogen-stimulated S6 kinase differentially alterstranslation of globin mRNA. J. Biol. Chem. 262, 3518-3523.Pang, D.T., Sharma, B.R., Shafer, J.A., White, M.F. andKahn, C.R. (1985): Predominance of tyrosinephosphorylation of insulin receptors during theinitial response to intact cells to insulin. J. Biol.Chem. 260, 7131-7136.Parker, P.J., Caudwell, F.B. and Cohen, P. (1983): Glycogensynthase from rabbit skeletal muscle. Effect ofinsulin on the state of phosphorylation of the sevenphosphoserine residues in vivo. Eur. J. Biochem. 130,227-234.Pelech, S.L., Olwin, B.B. and Krebs, E.G. (1986): Fibroblastgrowth factor treatment of Swiss 3T3 cells activates asubunit S6 kinase that phosphorylates a syntheticpeptide substrate. Proc. Natl. Acad. Sci. USA 83,5968-5972.Pelech, S.L., Sanghera, J.S., Paddon, H.B., Quayle, K.A.,Brownsey, R.W. (1991): Identification of a majormaturation-activated acetyl-CoA carboxylase kinase insea star oocytes as p44 mpk. Biochem. J. 274, 759-767.102Pierce, M.W., Palmer, J.L., Kentmann, H.T., Hall, T.A. andAvruch, J. (1982); The insulin-directedphosphorylation sites on ATP-citrate lyase. J.Biol.Chem. 257, 10681-10686.Price, D.J., Gunsalus, J.R. and Avruch, J (1990): Insulinactivates a 70-kDa S6 kinase through serine/threonine-specific phosphorylation of the enzyme polypeptide.Proc. Natl. Acad. Sci. USA 87, 7944-7948.Price, D.J., Mukhopadhyay, N.K. and Avruch, J. (1991):Insulin-activated protein kinases phosphorylate apseudosubstrate synthetic peptide inhibitor of the p70S6 kinase. J. Biol. Chem. 266, 16281-16284.Price, D.J., Nemenoff, R.A., and Avruch, J. (1989):Purification of a hepatic S6 kinase fromcycloheximide-treated rats. J. Biol. Chem. 264, 13825-13833.Rabilloud, T., Carpentier, G., Tarroux, P. (1988):Improvement and simplification of low-backgroundsilver staining of proteins by using sodiumdithionite. Electrophoresis 9, 288-291.Ramakrishna, S. and Benjamin, W. B. (1985): Cyclicnucleotide-independent protein kinase from rat liver.Purification and characterization of a multifunctionalprotein kinase. J. Biol. Chem. 260, 12280-12286.Ramakrishna, S. and Benjamin, W.B. (1988): Insulin actionrapidly decreases multifunctional protein kinaseactivity in rat adipose tissue. J. Biol. Chem. 263,12677-12681.Ramakrishna, S., Pucci, D.L. and Benjamin, W.B. (1983):Dependence of ATP-citrate lyase kinase activity on thephosphorylation of ATP-citrate lyase by cyclic AMP-dependent protein kinase. J.Biol. Chem. 258, 4950-4956.Ray, L.B. and Sturgill, T.W. (1987): Rapid stimulation byinsulin of a serine/threonine kinase in 3T3-L1adipocytes that phosphorylates microtubule-associatedprotein 2 in vitro. Proc. Natl. Acad. Sci. U.S.A. 84,1502-1506.Ray, L.B. and Sturgill, T.W. (1988): Characterization ofinsulin-stimulated microtubule-associated proteinkinase: rapid isolation and stabilization of a novelserine/threonine kinase from 3T3-L1 cells. J. Biol.Chem. 263, 12721-12727.103Ray, L.B. and Sturgill, T.W. (1988):Insulin-stimulatedmicrotubule-associated protein kinase isphosphorylated on tyrosine and threonine in vivo.Proc. Natl. Acad. Sci. USA 85, 3753-3757.Roach, P.J. (1991): Multisite and hierarchal proteinphosphorylation. J. Biol. Chem. 266, 14139-14142.Rossomando, A.J., Payne, D.M., Weber, M.J. and Sturgill,T.W. (1989): Evidence that pp42, a major tyrosinekinase target protein, is a mitogen-activatedserine/threonine protein kinase.Proc. Natl. Acad. Sci.U.S.A. 86, 6940-6943.Rossomando, A.J., Sanghera, J.S., Marsden, L.A., Weber,M.J., Pelech, S.L., and Sturgill, T.W. (1991):Evidence for a family of serine/threoninephosphorylations: a murine 44 kDa protein kinaserelated to both murine 42 kDa mitogen-activatedprotein kinase (p42 ma173i, ) and a sea star 44 kDameiosis-activated protein kinase (p44 m13,. ). (in press,1991).Saltiel, A.R. (1990): Second messengers of insulin action.Diabetes Care 13, 244-256.Saltiel, A.R., Osterman, D.G., Darnell, J.C., Chan, B.L.,Sorbara-Cazan, L.R. (1989): The role ofglycosylphosphoinositides in signal transduction.Recent Prog. Horm. Res. 45,353-382.Sanghera, J.S., Paddon, H.B. and Pelech, S.L. (1991): Roleof protein phosphorylation in the maturation-inducedactivation of a myelin basic protein kinase from seastar oocytes. J. Biol. Chem. 266, 6700-6707.Sanghera, J.S., Paddon, H.B., Bader, S.A. and Pelech, S.L.(1990): Purification and characterization of amaturation-activated myelin basic protein kinase fromsea star oocytes. J. Biol. Chem. 265, 52-57.Sanghera, J.S., Aebersold, R., Morrison, H.D., Bures, E.J.and Pelech, S.L. (1990a): Identification of the sitesin myelin basic protein that are phosphorylated bymeiosis-activated protein kinase p44 m1". Febs Lett.273, 223-226.Saperstein, R., Vicario, P.P., Strout, H.V., Brady, E.,Slater, E.E., Greenlee, W.J., Ondeyka, D.L., Patchett,A.A., Hangauer, (1989): D.G. Design of a selectiveinsulin receptor tyrosine kinase inhibitor and itseffect on glucose uptake and metabolism in intactcells. Biochemistry 28, 5694-5701.104Sato, C., Nishizawa, K., Nakayama, T. and Kobayashi, T.(1985): Effect upon mitogenic stimulation of calcium-dependent phosphorylation of cytoskeleton-associated350,000- and 80,000-mol-wt polypeptides in quiescent3Y1 cells. J. Cell Biol. 100, 748-753.Scopes, R.K. (1987): in Protein Purification 2nd ed.,Springer-Verlag, p41-54.Shechter, Y., Yaish, P., Chorev, M., Gilon, C., Braun, S.,Levitzki, A. (1989): Inhibition of insulin-dependentlipogenesis and ant-lipolysis by protein tyrosinekinase inhibitors. EMBO J. 8, 1671-1676.Sheorain, V.S., Juhl, H., Bass, M., Soderling, T.R. (1984):Effects of epinephrine, diabetes and insulin on rabbitskeletal muscle glycogen synthase. Phosphorylationsite occupancies. J. Biol. Chem. 259, 7024-7030.Simpson, I.A. and Cushman, S.W. (1986): Hormonal regulationof mammalian glucose transport. Ann. Rev. Biochem. 55,1059-1089.Smith, C.J., Rubin, C.S. and Rosen, O.M. (1980): Insulin-treated 3T2-L1 adipocytes and 11-free extractsderived from them incorporate '`P into ribosomalprotein S6. Proc. Natl. Acad. Sci. 77, 2641-2645.Smith and Sale (1989): Characterization of sites of serinephosphorylation in human placental insulin receptorcopurified with insulin-stimulated serine kinaseactivity by two-dimensional thin-layer peptidemapping. Febs. Lett., 242, 301-304.Sommercorn, J., Mulligan, J.A., Lozeman, F.J. and Krebs,E.G. (1987): Activation of casein kinase II inresponse to insulin and to epidermal growth factor.Proc. Natl. Acad. Sci. USA 84, 8834-8838.Soos, M.A., O'Brien, R.M., Brindle, N.P.J., Stigter, J.M.,Okamoto, A.K., Whittaker, J., Siddle, K. (1989):Monoclonal antibodies to the insulin receptor mimicmetabolic effects of insulin but do not stimulatereceptor autophosphorylation in transfected NIH 3T3fibroblasts. Proc. Natl. Acad. Sci. USA 86, 5217-5221.Stralfors, P., Bjorgell, P. and Belfrage, P. (1984):Hormonal regulation of hormone-sensitive lipase inintact adipocytes. Identification of phosphorylatedsites and effects on the phosphorylation by lipolytichormones and insulin. Proc. Natl. Acad. Sci. U.S.A.81, 3317-3321.105Sturgill (1988): Insulin-stimulated MAP-2 kinasephosphorylates and activates ribosomal protein S6kinase II. Nature 334, 715-Sweet, L.J., Alcorta,D.A., Jones, S.W., Erikson, E. andErikson, R.L. (1990): Identification of mitogen-responsive ribosomal protein S6 kinase pp90 rsk, ahomolog of Xenopus S6 kinase II, in chicken embryofibroblasts. Molecular and Cellular Biology 10, 2413-2417.Thomas, A.P. and Denton, R.M. (1986): Use of toluene-permeabilised mitochondria to explore the regulationof the pyruvate dehydrogenase system by reversiblephosphorylation. Biochem. J. 238, 93-101.Trevillyan, J.M., Perisic, O., Traugh, J.A., Byus, C.V.(1985): Insulin and phorbol ester-stimulatedphosphorylation of ribosomal protein S6. J. Biol.Chem. 260, 3041-3044.Turinsky, J., Bayly, B.P., O'Sullivan, D.M. (1990): 1,2-Diacylglycerol and ceramide levels in rat skeletalmuscle and liver in vivo: Studies with insulin,exercise, muscle denervation, and vasopressin. J.Biol. Chem. 265, 7933-7938.Van de Werve, G., Proielto, J., Jenrenaud, B. (1985): Tumor-promoting phorbol esters incease basal and inhibitinsulin-stimulated lipogenesis in rat adipocyteswithout decreasing insulin binding. Biochem. J. 225,523-527.Ward, G.E. and Kirschner M.W. (1990): Identification of cellcycle-regulated phosphorylation on nuclear lamin C.Cell 61, 561-577.Wente, S.R., Rosen, O.M. (1990): Insulin-receptor approachesto studying protein kinase domain. Diabetes Care 13,280-287.Wilden, P.A., Backer, J.M., Kahn, C.R., Cahill, D.A.,Schroeder, G.J. and White, M.F. (1990): The insulinreceptor with phenylalanine replacing tyrosine-1146provides evidence for separate signals regulatingcellular metabolism and growth. Proc. Natl. Acad. Sci.USA 87, 3358-3362.Witters, L.A. (1981): Insulin stimulates the phosphorylationof acetyl-CoA carboxylase. Biochem. Biophys. Res.COmmun. 100, 872-878.106Yang, S.-D., Ho, L.-T. and Fung, T.-J. (1988): Insulininduces activation and translation of protein kinaseFA (a multifunctional protein phosphatase activator)in human platelet. Biochem. Biophys. Res. Comm. 151,61-69.Yu, K.-T., Benjamin, W.B., Ramakrishna, S., Khalaf, N. andCzech, M.P. (1990): An insulin-sensitive cytosolicprotein kinase accounts for the regulation of ATPcitrate-lyase phosphorylation. Biochem. J. 268, 539-545.Yu, K.-T., Khalaf, N. and Cech, M.P. (1987): Insulinstimulates a novel Mn4+-dependent cytosolic serinekinase in rat adipocytes. Proc. Natl. Acad. Sci. USA84, 3972-3976.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items