Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies on protein phosphorylation in response to insulin in isolated cellular fractions reconstituted… Lew, Gregory John 1988

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

Item Metadata

Download

Media
831-UBC_1988_A6_7 L48.pdf [ 6.29MB ]
Metadata
JSON: 831-1.0097641.json
JSON-LD: 831-1.0097641-ld.json
RDF/XML (Pretty): 831-1.0097641-rdf.xml
RDF/JSON: 831-1.0097641-rdf.json
Turtle: 831-1.0097641-turtle.txt
N-Triples: 831-1.0097641-rdf-ntriples.txt
Original Record: 831-1.0097641-source.json
Full Text
831-1.0097641-fulltext.txt
Citation
831-1.0097641.ris

Full Text

^ / STUDIES ON PROTEIN PHOSPHORYLATION IN RESPONSE TO INSULIN IN ISOLATED CELLULAR FRACTIONS RECONSTITUTED WITH INSULIN RECEPTORS. By GREGORY JOHN LEW B.Sc.(Kin) Simon Fraser University, 1984. A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUTE STUDIES (Department of Biochemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April, 1988. (f) John Lew, 1988. 0 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of 'BfOC^&HfST^ The University of British Columbia Vancouver, Canada DE-6 (2/88) i i ABSTRACT The mechanism by which insulin and other polypeptide growth factors alter cellular metabolism is not fully understood. In the case of insulin, it is thought that phosphorylation/dephosphorylation mechanisms may play a central role in the signalling pathway. This is based on evidence which includes demonstration that the receptor for insulin is a tyrosine-specific protein kinase which is activated in response to insulin binding. Ultimately, insulin binding to its receptor on the surface of intact fat cells leads to altered levels of serine phosphorylation of several soluble proteins, including the phosphorylation of ATP-citrate lyase and acetyi-CoA carboxylase. Recently, studies involving site-specific mutagenesis have shown that the tyrosine kinase function of the insulin receptor is essential for insulin signalling. The studies described in this thesis have addressed the problem of how activation of the insulin receptor/tyrosine kinase results in the altered serine phosphorylation observed in intact cells in response to insulin. To gain further understanding of the cellular components required for insulin signalling, reconstitution experiments have been carried out mixing isolated cellular fractions with preparations of insulin receptors. The effects of insulin on altering protein-serine and protein-tyrosine phosphorylation have been determined in this reconstituted system. Results show that in a high-speed (100,000 x g) supernatant fraction prepared from rat adipose tissue endogenous protein-serine kinases are sensitive to conditions which are commonly employed for assaying insulin receptor/kinase activity. This includes inhibition by micromolar concentrations of MnCI2, by 40 mM NaF, and by low reaction temperature (0°C). When the insulin receptor, present in a WGA-Sepharose-purified preparation of detergent-solublized rat liver membranes, was assayed in the complete absence of both MnCI2 and NaF, receptor/tyrosine kinase activity was only slightly reduced with little or no decrease in the responsiveness to insulin. Furthermore, when the WGA-Sepharose-purified membrane fraction was incubated at 37°C in the presence of [ y -32P]ATP several endogenous proteins were observed to be phosphorylated in addition to the £-subunit of i i i the insulin receptor. These membrane proteins appear to be phosphorylated on tyrosine as indicated by their resistance to alkali hydrolysis. Upon reconstitution of the adipose tissue high-speed supernatant fraction with the WGA-Sepharose-purified preparation of insulin receptors the most striking effects observed were the phosphorylation of a 40 kd protein subunit (pp40) and the dephosphorylation of a 25 kd protein subunit (pp25) present in adipose tissue. The phosphorylation of pp40 occurs on tyrosine and is insulin-responsive, whereas the dephosphorylation of pp25 occurs following reconstitution with either untreated control, or insulin-activated insulin receptors. To assess the effect that reconstituted insulin receptors may have on the phosphorylation of endogenous ATP-citrate lyase in adipose tissue high-speed supernatant, it was found that a more pure preparation of insulin receptors was required. Further purification of the insulin receptor to homogeneity was therefore attempted using insulin-agarose affinity chromatography. However, difficulties including low yield and instability of the receptor through purification have prevented progress with these studies at present. In a separate study, highly purified acetyl-CoA carboxylase was reconstituted with a crude fraction consisting of total Triton-solublized membrane proteins. In this reconstituted system phosphorylation of acetyl-CoA carboxylase was enhanced to an extent greater than 6-fold after incubation with [ ^  - P]ATP. Following chromatography of the crude Triton-solublized extract over WGA-Sepharose this acetyl-CoA carboxylase kinase activity was found to be present in the flow-through void fraction and not in the N-acetylglucosamine eluted fraction. The acetyl-CoA carboxylase kinase, at present, does not appear to be insulin-responsive, but further studies are needed to confirm this observation. i v TABLE OF CONTENTS Abstract - ii List of Tables vii List of Figures viii List of Abbreviations x Acknowledgements xii INTRODUCTION 1 1. The Insulin Receptor. 1.1. Structure. 1 1.2. Tyrosine Kinase Activity. 2 2. The Role of Tyrosine Phosphorylation in Cells. 4 3. Role of the Insulin Receptor/Kinase Function in Insulin Action. 3.1. Substrates for the Insulin Receptor/Kinase. 6 3.2. Is Tyrosine Kinase Activity Required for Insulin Action? 7 4. Insulin Action. 8 4.1. Altered Serine Phosphorylation in Response to Insulin. 8 4.2. Substrates Phosphorylated in Response to Insulin. 4.2.1. Acetyl-CoA Carboxylase. 9 4.2.2. ATP Citrate Lyase. 10 4.2.3. Ribosomal Protein S6. 11 4.2.4. Insulin Receptor. 12 5. Thesis I nvestigations. 13 MATERIALS AND METHODS 1. Chemicals. 16 2. Tissue Sources. 16 3. Preparation of Wheat Germ Agglutinin-Sepharose. 16 V 4. Preparation of Partially Purified Insulin Receptors. 17 4.1. Preparation of Membranes. 17 4.2. Detergent Solublization. 18 4.3. Wheat Germ Agglutinin-Sepharose Affinity Chromatography. 18 5. Assays. 5.1. Protein Determination. 18 5.2. Preparation of Radiolabeled ATP. 19 5.3. Assay for Insulin Receptor/Kinase 5.3.1. Autophosphorylation. 19 5.3.2. Peptide Phosphorylation. 19 5.4. SDS-Polyacrylamide Gel Electrophoresis (SDS-PAGE). 20 5.5. Autoradiography. 21 5.6. Scanning Densitometry. 21 6. Phosphoamino Acid Analysis. 21 7. Preparation of Rat Adipose Tissue Cytosol. 22 8. Reconstitution of High-Speed Supernatant Proteins of Adipose Tissue with a WGA-Sepharose-Purified Preparation of Insulin Receptors. 23 9. Alkali Treatment of Gels. 23 RESULTS 1. Preparation of Partially Purified Insulin Receptors. 24 2. Protein Kinase Activity Associated with the Insulin Receptor. 24 3. Investigation of the Possible Effects of the Insulin Receptor/Tyrosine Kinase on the Activity of Protein Serine Kinases in a Reconstituted System. 31 3.1. Effect of Assay Components on Protein Kinase Activity. 3.1.1. Detergents. 36 3.1.2. Protein Phosphatase Inhibitors. 36 3.1.3. Divalent Metal Ions. 41 vi 3.2. Reconstitution of High-Speed Supernatant Fractions of Adipose Tissue with Partially Purified Insulin Receptors. 54 4. Purification of the Insulin Receptor to Homogeneity 62 4.1. Construction of Insulin-Bound Agarose 4.1.1. Synthesis of Boc-lnsulin 62 4.1.2. Coupling of Boc-lnsulin to Affigel 67 4.1.3. Deprotection of Boc-lnsulin Coupled to Agarose 67 4.2. Insulin-Agarose Affinity Chromatography 69 5. Investigation of Possible Effects of the Triton-Solublized Extract of Liver Membranes on the Phosphorylation of Acetyl-CoA Carboxylase in a Reconstituted System. 70 DISCUSSION 1. Preparation of Partially Purified Insulin Receptors. 74 2. Protein Kinase Activity Associated with the Insulin Receptor 76 3. Effect of Assay Components on Protein Kinase Activity 76 4. Reconstitution of High-Speed Supernatant Fractions of Adipose Tissue with Partially Purified Insulin Receptors. 79 5. Purification of the Insulin Receptor to Homogeneity 81 6. Investigation of Possible Effects of the Triton-Solublized Extract of Liver Membranes on the Phosphorylation of Acetyl-CoA Carboxylase in a Reconstituted System. 83 REFERENCES 85 vii LIST OF TABLES Table I Partial Purification of the Insulin Receptor. 25 Table II Deprotection of Boc-lnsulin Bound to Agarose. 63 Table III Effects of Insulin on the Incorporation of [32P] into the Insulin Receptor p -subunit, ATP-Citrate Lyase and pp40 68 v i i i LIST OF FIGURES Fig. 1. Model of the Insulin Receptor. 3 Fig. 2. Possible Mechanism by Which Insulin Binding to Cell Surface Receptors May Lead to Altered Serine Phosphorylation of IntracellularProteins. 14 qp Fig. 3. Time Course of [P]-Incorporation into Angiotensin Catalyzed by Insulin Receptor/Kinase. 26 Fig. 4. Phosphorylation of the Inulin Receptor -subunit. 29 Fig. 5. Phosphoamino Acid Analysis of the Insulin Receptor ^-subunit. 32 Fig. 6. Incorporation of [32P] from h/-32P]ATP into High-Speed Supernatant Proteins of Rat Adipose Tissue. 34 qp Fig. 7. Effects of Receptor Assay Components on the Incorporation of [ P] into High-Speed Supernatant Proteins of Rat Adipose Tissue. 37 Fig. 8. Effects of Triton X-100 on the Incorporation of [32P] into High-Speed Supernatant Proteins of Rat Adipose Tissue. 39 qp Fig. 9. Effects of NaF and 2-glycerophosphate on the Incorporation of [ P] into High-Speed Supernatant Proteins of Rat Adipose Tissue. 42 qp Fig. 10. Effects of 2 mM MnCI,, on the Incorporation of [ P] into High-Speed Supernatant Proteins of Rat Adipose Tissue. 44 qp Fig. 11. Incorporation of [ P] into High-Speed Supernatant Proteins of Rat Adipose Tissue with Varying Concentrations of MnCI2- 47 Fig. 12. Effects of MnCU on Protein Phosphatase Activity of Rat Adipose Tissue High-bpeed Supernatant. 49 Fig. 13. Effects of MnCI2 on Insulin Receptor/Kinase Activity. 51 Fig. 14. Incorporation of [32P] from [^-32P]ATP into Proteins Upon Reconstitution of the High-Speed Supernatant Fraction of Rat Adipose Tissue with the WGA-Sepharose-Purified Extract of Rat Liver Membranes. 55 Fig. 15. Effects of High-Speed Supernatant Fractions of Adipose Tissue on Insulin Receptor/Kinase Activity in a Reconstiuted System. 58 Fig. 16. Determination of Tyrosine Phosphorylated Proteins in High-Speed Supernatant Fractions of Adipose Tissue Reconstituted with WGA-Sepharose-Purified Extracts of Liver Membranes by Alkali Treatment of SDS-Polyacrylamide Gels. 60 Fig. 17. Insulin-Agarose Affinity Purified Insulin Receptor. 64 ix Fig. 18. Effects of Reconstitution of Purified Acetyl-CoA Carboxylase with Fractions Obtained From the Triton-Solublized Extract of Liver Membranes. X LIST OF ABBREVIATIONS ATP adenosine triphosphate CAMP cyclic adenosine monophosphate. cGMP cyclic guanine monophosphate. DTT dithiothreitol. EDTA ethylenediaminetetraacetic acid. EGTA ethyleneglycoltetraacetic acid. GP glycerophosphate. HEPES N-2-Hydroxyethylpiperazine-N'-2-ethanesulfonic acid. HSS high-speed supernatant fraction of rat adipose tissue. MOPS 3-[N-Morpholino]propanesulfonic acid. P inorganic phosphate. PAGE polyacrylamide gel electrophoresis. PMSF phenyl methylsulfonylfl uoride. SDS sodium dodecyl sulfate. TCA trichloroacetic acid. TFA trifluoroacetic acid. TLC thin layer chromatography TE Triton X-100-solubilized extract of rat liver membranes. Tris Tris(hydroxymethyf)aminomethane WGA wheat germ agglutinin WGE wheat germ agglutinin-Sepharose purified extract of rat liver membranes. Units of Measurement g gram 1 liter mol mole xi M molar min minutes d (or dal) daltons dpm disintegrations/minute k kilo m milli u micro n nano p pico ACKNOWLEDGEMENTS Any contribution that this thesis should make rests solely on the inspiration and support of Roger Brownsey, who was an an excellent supervisor and teacher. In addition, I would like to thank the people who I worked with most closely, and who were of invaluable help to me in the lab - Kath Quayle, Bob Winz, Gordon Dong, Connie Leung, Gordon Louie, Wendy Hutcheon, Chris Sherwood and Al Burton. Most deeply, I thank those of you with whom I shared an immeasurable friendship, a plentitude of laughs and a few hundred pints of Rickards Red. I sometimes think you people were what this thesis was all about. You know who you are. Finally, this thesis was made possible by support from the Medical Research Council of Canada operating grant of Dr. R.W. Brownsey, Department of Biochemistry, U.B.C. 1 INTRODUCTION The polypeptide hormone, insulin, exerts a spectrum of both short and long term physiological effects on cells of target tissues. The long term effects involve changes in both general and specific protein synthesis and breakdown, while the short term effects are brought about through changes in the activity of pre-existing enzymes and membrane transporters. The acute actions of insulin include the stimulation of glucose uptake into both muscle and adipose tissue where it is converted primarily to glycogen (in muscle), and to fatty acids and triacylglycerol (in fat cells). In liver, the hormone acts to prevent glucose release and stimulates glycogen storage and fatty acid and triacylglycerol synthesis. Through these actions insulin occupies a central role in maintaining blood glucose homeostasis. In addition, insulin promotes the uptake of amino acids, and the enhanced synthesis of DNA, RNA and protein (Jacobs and Cuatrecasus, 1981; Khan et al., 1981). Insulin therefore acts as a growth factor and is required for the maintenance of mammalian cells in culture. To date, the molecular mechanisms by which insulin exerts its biological effects are not fully understood. 1. The Insulin Receptor. 1.1. Structure. Of all the known receptors for polypeptide hormones, none have drawn more attention than the receptor for insulin. Its purification was first reported by Cuatrecasas (1972b), and the cDNA encoding the entire insulin receptor has now been cloned and sequenced (Ullrich et al., 1985; Ebina et al., 1985). Today we know much about its structure, properties and seqence homology with other proteins, which include the closely related receptor for insulin-like growth factor (IGF)-1 (Ullrich et al., 1986) and the family of src-related oncogene products (Ullrich et al., 1985). The predominant form of the insulin receptor is composed of two glycoprotein subunits («*,^) covalently linked by disulfide bonds in an s u b u n i t stoichiometry. This subunit arrangement 2 has been deduced from a range of cell surface and biosynthetic labelling studies (Kasuga et al., 1982; Van Obberghen et al., 1981; Hedo and Khan, 1985) (fig. 1). Cell surface labelling and protease-sensitivity studies have shown the <*-subunit (Mr=135 kd) to be extracellular and the p-subunit (Mr=95 kd) to have both extracellular and intracellular domains with a linking transmembrane region (Hedo and Simpson, 1984). The «-subunit can be labelled with 125l-insulin (Yip et al., 1980) whereas the /3-subunit is labelled very weakly or not at all with this technique. The stoichiometry of insulin binding is controversial. Pang and Schafer (1983) have used double-probe labelling to show that the receptor is monovalent with respect to insulin binding. Analysis by Scatchard plot shows the receptor to contain one high and one low affintiy binding site (Fujita-Yamaguchi et al., 1983). From the nucleotide sequence of the insulin receptor cDNA (Ullrich et al., 1985; Ebina et al., 1985), the corresponding deduced amino acid sequence shows that the mature insulin receptor arises from a single chain precursor containing both the «*- and /3-subunits, which is cleaved during post-translational processing (Hedo and Gorden, 1985). Both subunits are heavily glycosylated, containing N-linked oligosaccharides of the high-mannose type, which comprise approximately 27% and 16% of the molecular masses of the and /»-subunits, respectively (Herzberg et al., 1985). The mature receptor acquires sialic acid during processing and binds strongly to wheat germ agglutinin (Hedo et al., 1981). 1.2. Tyrosine Kinase Activity. The insulin receptor was originally shown to be closely associated with an insulin-sensitive protein-tyrosine kinase activity (Kasuga et al., 1982a). Purification studies showed that 125l-insulin binding activity copurified with tyrosine kinase activity to apparant homogeneity (Petruzzelli et al., 1984). In addition, the cytoplasmic domain of the ^-subunit has been affinity labelled with [<*-32P]8-azidoadenosine 5'-triphosphate (Roth and Cassell, 1983). These observations suggest that the receptor-associated tyrosine kinase activity is: i) intrinsic to the insulin receptor protein; ii) located 135 Kd 95 Kd S-S l-S-S Extracellular ligand-binding d o m a i n C y t o p l a s m i c tyrosine kinase d o m a i n _c O F i g . 1 Model of the I n s u l i n Receptor (Gammeltoft, 1986). 4 within the ^-subunit; and iii) activated by insulin binding. In a broader perspective, the insulin receptor is classified as belonging to the family of growth factor receptor/tyrosine kinases which also include the receptors for epidermal growth factor (EGF), platelet-derived growth factor (PDGF), and insulin-like growth factor (IGF)-I (for review, see White and Khan, 1986). The insulin receptor/kinase undergoes autophosphorylation both in intact cells and in pure preparations (Kasuga et al., 1982b). The stiochiometry of autophosphorylation is not precisely known, but there is some evidence to suggest that two moles of phosphate are incorporated per mole of insulin bound (White et al., 1984; Petruzzeili et al., 1984). In intact ceils, insulin-treatment results in the increased phosphorylation of serine and threonine as well as tyrosine residues on the insulin receptor p-subunit (Kasuga et al., 1982c; Pang et al., 1985). This is of interest since the receptor/kinase itself has been shown to be incapable of phosphorylating substrates on serine or threonine (Walker et al., 1987). As well, purification of the receptor to homogeneity results in a preparation which catalyzes autophosphorylation solely at tyrosine residues (Kasuga et al., 1983; White et al., 1985). This suggests that protein-serine/threonine kinases distinct from the insulin receptor are activated in response to insulin binding, and it has been postulated that such protein kinases may play a role in mediating some or all of insulin's actions (Denton et al., 1981; Yu et al., 1987) 2. The Role of Tyrosine Phosphorylation in Cells. The first description of protein-tyrosine phosphorylation was made by Eckhart et al. (1979) where phosphotyrosine was detected in hydrolysates of viral transforming proteins labelled with [ •{-qo P]ATP. It is now recognized that in all animal cells, protein phosphotyrosine comprises a small (<0.2%) but significant amount of total phosphorylated amino acids, the vast majority being phosphoserine (90%) and phosphothreonine (10%). By analogy with cellular protein serine and threonine phosphorylation, reversible tyrosine phosphorylation might be expected to modulate 5 protein function. Although substrates for tyrosine kinases have recently been identified, the physiological significance of tyrosine phosphorylation has yet to be discovered. The fact that insulin as well as other growth factors (EGF, PDGF, and IGF-I), stimulate autophosphorylation on tyrosine residues of their respective receptors (for review, see White and Khan, 1986), suggests that this phenomenon may serve to regulate some fundamental mechanism of cellular growth and/or differentiation. To extend this hypothesis, the virally encoded oncogene products, v-sis and v-erb-B, are derived from sequences of the genes which code for a PDGF (Doolittle et al., 1983) and the EGF receptor (Downward et al.,1984), respectively. The sequence of other oncogenes also correspond to other growth factor receptor gene sequences, such as the sequence of v-fms and macrophage colony-stimulating factor (CSF)-1 receptor, and v-kit and the PDGF receptor (Carpenter, 1987). It has also been shown that cells transformed by expression of v-abl oncogene show increased tyrosine phosphorylation of nuclear proteins which bind DNA, and which may effect altered gene transcription (Bell ef al., 1987). In addition, there have been examples where tyrosine phosphorylation of purified proteins in vitro has led to altered protein activity. Most notably, autophosphorylation has been shown to be necessary for activation of the insulin (Rosen et al.,1983; Yu et al., 1984) and IGF-I (Yu et al., 1986) receptor/tyrosine kinases towards exogenous substrates. In the case of the insulin receptor, activation can also be achieved through tyrosine phosphorylation of the receptor ^-subunit catalyzed by pp60 v" s r c (Yu et al., 1985). pp60 v" s r c will also phosphorylate and inhibit the activity of protein phosphatase 1 (Johansen and Ingebritsen, 1986). Likewise, the inhibitor protein of cAMP-dependent protein kinase is several fold less effective when phosphorylated on tyrosine by the activated EGF receptor/tyrosine kinase (Van Patten et al., 1987). 6 3. Role of the Insulin Receptor/Kinase Function in Insulin Action. 3.1. Substrates for the Insulin Receptor/Kinase. With the discovery that the receptors for EGF, PDGF, IGF-1 and insulin were all protein-tyrosine kinases, it was perhaps reasonable to anticipate that the mechanism of action of these hormones would soon be elucidated by identification of possible substrates for these receptors. Several years have passed, however, and as yet a concensus on the identities of physiologically significant substrates for these receptor/kinases has not been established. Very recently, however, a number of possible candidates have been described (Kadowaki et al., 1987; Haring et al., 1987; Reese-Jones and Taylor, 1985). Of particular note are two proteins of Mr=185 kd (White et al., 1985) and Mr=15 kd (Bernier et al., 1987) which are observed to be tyrosine phosphorylated in intact cells in response to insulin. The Mr =185 kd protein (pp185) is a soluble protein that is distinct from the insulin receptor judging by a number of biochemical and physiological criteria. The phosphorylation of pp185 follows similar kinetics to that of the insulin receptor and is increased in direct response to increased expression of cell-surface receptors in cells transfected with insulin receptor cDNA (White et al., 1987). Phosphorylation of pp185 is also brought about by the action of IGF-I (Izumi et al., 1987) but not by that of EGF (Kadowaki et al., 1987). There has been no clear evidence concerning the function of this newly described phosphoprotein, however. The 15 kd phosphoprotein (pp15) was observed by Bernier et al. (1987) in cells treated with phenylarsine oxide, an agent shown to interrupt signal transmission from the insulin receptor to the glucose transport system (Frost et al., 1987). Exposure of cells to phenylarsine oxide leads to the insulin-stimulated accumulation of pp15 which is phosphorylated solely on tyrosine. The observed kinetics of insulin receptor autophosphorylation, the phosphorylation of pp15, and activation of hexose transport suggests that pp15 may be an intermediate in the signalling of this pathway. This so far, has been the only example of a putative substrate described in context with any physiological action of insulin. 7 3.2. Is Tyrosine Kinase Activity Required for Insulin Action? The cloning and sequencing of the insulin receptor cDNA has provided the opportunity for many elegant studies which have attempted to determine whether the tyrosine kinase function of the insulin receptor is necessary for insulin action. Mutation of the receptor at either the site of autophosphorylation (Ellis et al., 1986) or the site of ATP binding (Ebina et al., 1987) yields a receptor that is processed normally during post-translational modification and binds insulin, but is tyrosine kinase deficient. Chou et al. (1987) have shown that several rapid effects of insulin, including glucose uptake, S6 kinase activation, endogenous substrate phosphorylation, glycogen synthesis, and thymidine incorporation into DNA, are mediated in cells expressed with wild type, but not mutant receptors. Mutant receptors expressed in cells treated with insulin also fail to be internalized and down-regulated (Ghertzi et al., 1987; Russell et al., 1987; Hari and Roth, 1987). However, insulin receptors have been shown to undergo internalization by an anti-insulin receptor monoclonal antibody that does not stimulate receptor autophosphorylation (Russell et al., 1987), and which is capable of internalizing kinase-deficient receptors. A different antibody has been shown to activate glucose uptake, and similarly, does so without stimulating insulin receptor autophosphorylation (Forsayeth etal., 1987). In Xenopus oocytes several acute (Morgan and Roth, 1987) and long term (Morgan et al., 1986) effects of insulin have been shown to be inhibited by antibodies that block receptor autophosphorylation following their microinjection. Furthermore, Ellis et al. (1987) have constructed a hybrid receptor consisting of the extracellular domain of the insulin receptor fused to the transmembranal and intracellular domain of the chicken sarcoma virus, v-ros, which has been shown to be a tyrosine kinase and with which the insulin receptor shares the greatest sequence homology (Ullrich et al., 1985). This hybrid molecule is expressed normally on the cell surface, binds insulin, undergoes autophosphorylation, and will phosphorylate exogenous substrates on tyrosine residues. However, the hybrid fails to mediate at least two classical effects of insulin, namely glucose uptake and thymidine incorporation into DNA. 8 In summary, the majority of evidence suggests that the tyrosine kinase function intrinsic to the insulin receptor is necessary for mediating the biological effects of insulin. 4. Insulin Action. Over the years, the evidence supporting a unified model of insulin action has been extremely fragmentary. Possible mechanisms of intracellular signal transduction include changes in intracellular Ca concentration, the release of low molecular weight mediators, interaction with guanine nucleotide regulatory proteins and internalization of receptor-hormone complexes. In addition, there is substantial evidence to strongly suggest a role for protein phosphorylation, which is the premise of this thesis. 4.1. Altered Serine Phosphorylation in Response to Insulin. Early studies on several well-characterized proteins revealed that enzyme activation, in response to insulin-treatment of tissue, correlated with net dephosphorylation. Thus, dephosphorylation has been observed with pyruvate dehydrogenase, glycogen synthase (both enzymes exhibiting concomittant activation) and triacylglycerol lipase (the activity of which is decreased; e.g. see Denton et al., 1984). In the case of the hormone-sensitive lipase, dephosphorylation is observed only after prior phosphorylation and activation by cAMP-dependent protein kinase. Recently, it has been recognized that there are a number of proteins which exhibit a net increase in their level of phosphorylation in response to insulin. In adipose tissue examples of this are ATP-citrate lyase, acetyl-CoA carboxylase, the ribosomal protein S6 and two proteins of subunit M r 22,000 and 61,000. Of these proteins, only the phosphorylation of acetyl-CoA carboxylase has been associated with a corresponding change (increase) in activity. 9 4.2. Substrates Phosphorvlated in Response to Insulin. 4.2.1. Acetvl-CoA Carboxylase. Acetyl-CoA carboxylase [EC 6.4.1.2] (Mr=230 kd) catalyzes the ATP-dependent carboxylation of acetyi-CoA to produce malonyl-CoA the sole fate of which is incorporation into fatty acids catalyzed by fatty acid synthetase. The first direct evidence for the phosphorylation of acetyl-CoA carboxylase in intact cells was obtained by Brownsey et al. (1977) in studies using intact epididymal fat cells labelled with [32P]. Incorporation of [32P] into acetyl-CoA carboxylase under these conditions reaches a steady state of about 1-2 mol/mol of subunit after 1 hr. Treatment of cells with either insulin or adrenaline results in modest increases in the level of phosphorylation of acetyl-CoA carboxylase compared to non-hormone-treated controls (Brownsey and Denton, 1982; Brownsey et al, 1979; Brownsey et al. 1977). This was surprising at the time of discovery since insulin acts to increase acetyl-CoA carboxylase activity, while adrenaline treatment of cells results in inactivation of the enzyme. Phosphopeptide analysis, however, revealed that the tryptic peptides phosphoryiated in response to insulin were not the same as those phosphorylated in response to adrenaline (Brownsey and Hardie, 1980; Brownsey and Denton, 1982) and that these different patterns of phosphorylation may be associated with opposite changes in enzyme activity. Phosphorylation of purified acetyl-CoA carboxylase by the catalytic subunit of cAMP-dependent protein kinase occurs mainly at the same sites as those seen in the enzyme isolated from fat cells treated with adrenaline (Brownsey and Hardie, 1980). The effect on enzyme activity is also similar (Brownsey and Denton, 1982; Brownsey et al, 1979), and it is therefore likely that adrenaline action on acetyl-CoA carboxylase is mediated through a mechanism leading to the activation of cAMP-dependent protein kinase. This is not the mechanism of insulin-stimulated acetyl-CoA carboxylase phosphorylation, however, since the phosphorylation of the insulin-directed peptide in vitro is not influenced by cAMP or by the specific Walsh inhibitor protein (which inhibits cAMP-dependent protein kinase, see Scott et al., 1985). 10 Purified casein kinases I and II are also able to catalyze the phosphorylation of purified acetyl-CoA carboxylase (Witters et al., 1983). Both casein kinases have been reported to phosphorylate a tryptic peptide that is also phosphorylated in the enzyme isolated from intact fat cells treated with insulin (Witters et al., 1983). However, this phosphorylation was not associated with altered acetyl-CoA carboxylase activity (for review, see Brownsey and Denton, 1986), It is not clear as to whether the peptide phosphorylated by purified casein kinase I and II is in fact the major insulin-directed site as seen when acetyl-CoA carboxlyase is phosphorylated in intact cells in response to insulin. Unpublished observations of Katherine Quayle in this lab suggest that the sites phosphorylated by casein kinases I and II are different from the major insulin-directed site. Acetyl-CoA carboxylase does not appear to be a good substrate for the Ca2+/phospholipid-dependent protein kinase (protein kinase C). 4.2.2. ATP Citrate Lyase. Insulin promotes the phosphorylation of ATP-citrate lyase [EC 4.1.3.8] (Mr=120 kd) in intact cells. This may or may not be via the same protein kinase associated with insulin-dependent acetyl-CoA carboxylase phosphorylation. Like acetyl-CoA carboxylase, ATP-citrate lyase also undergoes increased phosphorylation in response to hormones which increase cellular levels of cAMP. Thus, in vitro, purified ATP-citrate lyase is a substrate for cAMP-dependent as well as cGMP-dependent protein kinase, both kinases which phosphorylate ATP-citrate lyase on the same tryptic peptide (Swergold et al., 1982). Phosphorylation by cAMP-dependent kinase occurs to an extent of about 0.5 mol P/mol of subunit. Ramakrishna and Benjamin (1985) have purified a multifunctional cAMP-independent protein kinase that phosphorylates both ATP-citrate lyase and acetyl-CoA carboxylase. It is unlikely, however, that this kinase is the insulin-sensitive kinase which phosphorylates ATP-citrate lyase, since the multifunctional protein kinase phosphorylates purified ATP-citrate lyase specifically on a serine and threonine residue of a peptide designated "b", whereas insulin promotes the 11 dephosphorylation of peptide "b" and phosphorylation of a different peptide designated "a". It is puzzling that peptide "a" is the site of insulin-stimulated phosphorylation in intact cells as well as the site of phosphorylation by cAMP-dependent protein kinase, in vitro (Swergold et al., 1982). Amino acid sequence analysis by Pierce et al. (1982) has further shown that an identical serine residue within this peptide is phosphoryiated in cells exposed to both insulin and hormones which increase cellular levels of cAMP. With these observations in mind it is difficult to imagine the physiological significance of the hormone-dependent phosphorylation of ATP-citrate lyase. As yet there have been no studies demonstrating altered enzyme activity in response to phosphorylation, although ft has been suggested that insulin and ^-adrenergic agonists may alter the cellular localization of the enzyme (Stralfors, 1987). Irrespective of functional changes, however, the enhancement of phosphorylation by insulin suggests that this protein may be used as a molecular marker for studying protein kinase activity in response to insulin. 4.2.3. Ribosomal Protein S6. Of all protein-serine phosphorylation reactions that occur in response to insulin, perhaps the most studied to date has been the phosphorylation of the mammalian 40S ribosomal protein, S6 (Mr=31 kd). As is the case with ATP-citrate lyase the functional consequence of S6 phosphorylation is poorly understood. However, there is some evidence that S6 phosphorylation, in response to growth factors, correlates with an increased rate of protein synthesis (Cobb and Rosen, 1983). Insulin (Tabarini et al., 1985) and other growth factors (Novak-Hofer and Thomas, 1984; Pelech et al., 1986) stimulate S6 phosphorylation via a cAMP-independent mechanism. Although cAMP-dependent protein kinase will phosphoryiate S6 in vitro, two dimensional peptide analysis shows the pattern of phosphorylation to be clearly different from that observed in response to growth factor treatment (Lastic and McConkey, 1981). In vitro, protein kinase C will catalyze the 12 phosphorylation of S6 at sites reported to be identical with those phosphorylated by insulin in intact cells (Trevillyan et al., 1985). The insulin-dependent kinase activity and protein kinase C, however, appear to be distinct fjabarini et al., 1985). Treatment of intact cells with phorbol esters can mimic insulin-dependent S6 phosphorylation, suggesting that protein kinase C may be involved in activation of the S6 kinase (Tabarini et al., 1985; Pelech et al., 1986). Phosphorylation in response to growth factors, however, does not necessarily require protein kinase C (Pelech et al., 1986), which suggests that activation of S6 kinase by growth factors and phorbol esters may occur via separate mechanisms. Ribosomal protein S6 can be phosphorylated to a stoichiometry as high as 6 mol P/mol S6 in response to growth factor-treatment in intact cells. The S6 protein has been partially sequenced in a region that contains the major phosphorylation sites for cAMP-dependent protein kinase and one of the insulin-directed phosphorylation sites (Pelech etal., 1986; Wettenhall and Morgan, 1984). A synthetic peptide, Arg-Arg-Leu-Ser-Ser-Leu-Arg-Ala, based upon this sequence has been shown to be a substrate for cAMP-dependent protein kinase and fibroblast growth factor (FGF)-stimulated S6 kinase (Pelech et al., 1986). While the first serine residue (residue 4 from the N-terminus, as shown) is the predominant site for cAMP-dependent phosphorylation, the adjacent serine is the major site for phosphorylation in response to insulin (Wettenhall and Morgan, 1984). This peptide has provided a convenient assay of S6 kinase activity (Pelech et al., 1986). 4.2.4. Insulin Receptor. The insulin receptor itself undergoes serine phosphorylation in whole cells treated with insulin. Studies by Pang et al. (1985) have shown that insulin stimulates receptor phosphorylation ocurring initially on tyrosine with the subsequent appearance of phosphoserine and phosphothreonine. In addition, in transfection studies, wild type receptors expressed on the cell surface show serine/threonine phosphorylation in response to insulin binding, but under the same conditions mutant receptors which are tyrosine kinase-deficient do not autophosphorylate and are not 13 phosphorylated on serine or threonine (Ora Rosen, personal commun.). This suggests that these protein-serine/threonine kinases are activated only in response to insulin receptor tyrosine phosphorylation (or alternatively, act only on the tyrosine phosphorylated form of the receptor). As yet, the identity of the receptor-associated serine/threonine kinase activity is not known, however, evidence suggests that it is not protein kinase C (Jacobs and Cuatrecasas, 1986). Serine phosphorylation of the detergent-solubilized insulin receptor by cAMP-dependent protein kinase inhibits receptor autophosphorylation and its ability to phosphorylate exogenous substrates (Stadtmauer and Rosen, 1986; Tanti et al., 1987; Haring et al., 1986a). In intact cells, the unoccupied receptor also undergoes serine phosphorylation in response to treatment of cells with phorbol esters, similarly resulting in diminished receptor/kinase activity and decreased insulin responsiveness (Takayama et al., 1984; Haring et al., 1986b). This has also been observed in vitro where autophosphorylation and receptor/kinase activity is decreased following phosphorylation by purified protein kinase C (Bollag et al., 1986). In general, it appears that serine phosphorylation of the insulin receptor results in the inhibition of its tyrosine kinase function. 5. Thesis Investigations. In the studies described in this thesis I have attempted to reconstitute insulin signalling in a cell-free system by mixing isolated cellular fractions. This is based on the hypothesis that the insulin receptor/tyrosine kinase plays a vital role in regulating the much more abundant serine phosphorylation observed in whole cells (including the phosphorylation of acetyl-CoA carboxylase, ATP-citrate lyase and ribosomal protein, S6). A possible model of this scheme is shown in figure 2. Experiments revealed that the conditions required for co-expression of protein-serine kinases in a crude cytosol fraction and insulin receptor/tyrosine kinase in a membrane glycoprotein fraction are critical, and that the temperature of reaction, divalent metals, and the presence of phosphatase inhibitors must be carefully considered. Upon reconstitution I have shown the appearance of a 40 14 i n s u l i n \4) I \ i I i n s u l i n r eceptor out l l v v W l / plasma //////J|77^ 77l//////////////y tnUtr membrane m T y r - P 0 4 SECONDARY SUBSTRATES - S e r SECONDARY SUBSTRATES - S e r —P04 e.g. ACC, ACL or S6 F i g . 2 A P o s s i b l e Mechanism by Which I n s u l i n Binding to C e l l Surface Receptors May Lead to A l t e r e d Serine P h o s p h o r y l a t i o n of I n t r a c e l l u l a r P r o t e i n s . ACCj acetyl-CoA carboxylase; ACL, A T P - c i t r a t e l y a s e ; S6, 40S ribosomal p r o t e i n S6. 15 kd protein subunit that is most likely phosphorylated on tyrosine and may be an endogenous substrate for the insulin receptor/kinase. To better assess the possible effects of reconstitution on the serine phosphorylation of soluble proteins, I set out to reconstitute the cytosolic fraction of fat cells with highly purified insulin receptors. At this time, however, the yield of receptors following purification has not been satisfactory. The development of an improved purification technique will certainly take priority in future studies, at which time reconstitution and other investigations using purified receptors can be carried out. Finally, I have reconstituted a purified preparation of acetyl-CoA carboxylase with a crude detergent-solubilized membrane extract and shown the presence of a serine kinase that will catalyze the phosphorylation of acetyl-CoA carboxylase with high stoichiometry. 16 MATERIALS AND METHODS 1. Chemicals. All general laboratory chemicals were reagent grade and were purchased from Sigma, BDH or Fisher. Water used throughout was distilled and purified using a Millpore (milli-Q) water purification system. 2. Tissue Sources. Male Wistar rats were from the University of British Columbia Animal Care Center and were fed ad libitum within the Department of Biochemistry for 1-3 days prior to use. Whole human placenta was obtained from Grace Hospital, Vancouver, B.C., placed on ice immediately after delivery and used within 1 hr. 3. Preparation of Wheat Germ Agqlutinin-Sepharose. Wheat germ agglutinin (Boehringer Mannheim) was coupled to Sepharose CL-4B (Pharmacia) using the procedure described by Syska et al. (1974). Sepharose beads (20 ml packed vol) were washed thoroughly (10-20 volumes) with water in a buchner funnel and brought to pH 11.0 by dropwise addition of NaOH (5N). One gram of CNBr (in water) was then added, the pH maintained between 10.8-11.5 by the addition of NaOH, and the temperature maintained between 18-22°C by the addition of ice until the reaction was complete. Unreacted CNBr was then removed by washing with several volumes of ice cold 0.2M NaHC03 (pH 9.0). The activated Sepharose was mixed with 100 mg wheat germ agglutinin dissolved in 0.2M N-acetylglucosamine (Sigma), 0.2M NaHC03, pH 9.5 at 4°C, on a rotory shaker. Following incubation overnight the resin was washed with 5-10 volumes of phosphate buffer (10 mM sodium phosphate, pH 7) and then incubated with 1M ethanolamine-HCI (BDH) (pH 7) for 4 hr at 0°C. The resin was then washed again and stored in the same buffer with 0.02% NaN~ at 4°C. 17 Prior to use, WGA-Sepharose was washed at 4 C with 5 volumes of buffer (pH 7.6) containing 10 mM HEPES, 0.1% (w/v) Triton X-100, 150 mM NaCl, 0.01% SDS. The column was then washed overnight with 50 volumes of the same buffer from which SDS was omitted. 4. Preparation of Partially Purified Insulin Receptors. Insulin receptors were partially purified according to the published procedures of Pike et al. (1984, 1986) which is based on the protocol originally developed by Cuatrecasus (1972a,b). The procedure is outlined below and includes preparation of a crude cell membrane fraction, detergent solubilization of integral membrane proteins and isolation of the solubilized membrane glycoproteins. 4.1. Preparation of Membranes. Eight male wistar rats weighing between 220-250 g were stunned and decapitated, and the livers removed and placed in ice within 2 min. The livers were minced with scissors and then homogenized with 2 strokes of a Potter-Elvehjem homogenizer in 3 volumes of buffer (pH 7.4) containing 250 mM sucrose, 25 mM Tris, 25 mM MOPS, 5 mM EDTA, 2 mM EGTA, 2 ug/ml pepstatin A, 2 ug/ml leupeptin, 2.5 mM benzamidine, 1 mg/ml bacitracin, 0.1 mM PMSF and 5 mM 2-mercaptoethanol. During all procedures samples were kept on ice or at 4°C unless otherwise mentioned. The crude homogenate was centrffuged for 10 min at 600 x g. The pellets were washed once in 1 volume homogenization buffer using a Potter-Elvehjem, recentrifuged (10 min, 600 x g), and the initial and wash supernatants combined. The pooled supernatants were then centrffuged for 20 min at 15,000 x g and the mitochondrial pellet discarded. Following the addition of 100 mM NaCl and 200 mM MgCI2 the homogenate was rehomogenized with a Polytron PT 10-35 probe (setting 7) for 3 sees. The membrane fraction was then pelleted by centr'rfugation for 40 min at 40,000 x g in a Sorvall SS-34 rotor. The membranes were resuspended in approximately 10 volumes of homogenization buffer using a Polytron, as described above and recentrifuged at 40,000 x g for 40 min. The washing procedure was repeated a second time replacing the 18 homogenization buffer with 50 mM HEPES, pH 7.6. The membranes were then stored as a pellet at -20°C overnight. 4.2. Detergent Solubilization. Membranes were thawed at room temperature and the volume was brought to 50 mis with 50 mM HEPES (pH 7.6) and Triton X-100 at a final concentration of 2% (w/v). The suspension was homogenized using a Potter-Elvehjem homogenizer and stirred at 4°C for 1 hr. The supernatant, after centrifugation for 1 hr at 120,000 x g, is referred to as the Triton-solubilized extract of rat liver membranes" and was either purified further or stored at -70°C 4.3. Wheat Germ Agglutinin-Sepharose Affinity Chromatography. The Triton-solubilized membrane extract (30-35 mis) was diluted 4-fold with 50 mM HEPES (pH 7.6) and incubated with 20 mis of WGA-Sepharose for 1.5-2 hr on a rotory shaker at 4°C. The resin was then batch washed with 4 volumes of 50 mM HEPES, 0.1% (w/v) Triton X-100, pH 7.6, 150 mM NaCI, poured into a column and further washed free of unbound protein with 300 mis of the same buffer. Glycoproteins were eluted with the same buffer supplemented with 300 mM N-acetylglucosamine (Sigma) at a flow rate of 10 mls/hr. Fractions (2 ml) were collected and assayed for protein by the method of Bradford (1976). Fractions containing protein were pooled and dialyzed overnight against 100 volumes of 25 mM HEPES, 0.05% (w/v) Triton X-100, pH 7.6, at 4°C, and then concentrated against Aquacide III (Calbiochem) to a final protein concentration of 3 mg/ml. This fraction is referred to as the "WGA-Sepharose-purified extract of rat liver membranes" and was stored frozen at -70°C in 50 ul aliquots until use. 5. Assays. 5.1. Protein Determination. Protein was assayed by the method of Bradford (1976). One hundred microliter samples containing 10-50 ug of protein sample were mixed with 5.0 mis of Bradford reagent and the colour 19 allowed to develop at room temperature. After 10-30 min the absorbance was measured at 595 nm in a Beckman DU-8 spectrophotometer against a reagent blank consisting of 5 mis of Bradford reagent and 0.1 mis of the appropriate buffer. Protein concentration was estimated from a standard calibration curve, linear between 10-50 ug BSA, and which was prepared on each separate occasion. The values shown are the mean of duplicate samples. 5.2. Preparation of Radiolabelled ATP. W- P]ATP (1 mCi/ml, 10 Ci/mmole) was purchased from Amersham and dried under a stream of N 2 at room temperature. The radioisotope was then reconstituted in buffer (pH 7.2) containing 20 mM 2-glycerophosphate, 5 mM MgCI2, 1 mM EDTA and 2 mM ATP (Sigma). The [tf-32P]ATP preparation (500-1000 dpm/pmol) was kept frozen at -20°C until use. 5.3. Assay for Insulin Receptor/Kinase. 5.3.2. Autophosphorylation. To determine autophosphorylation of the insulin receptor ^-subunit, fractions containing insulin receptors were preincubated with or without insulin (Sigma) (10 ug/ml) for 45 min at 23°C in buffer (pH 7.6) containing 50 mM HEPES, 0.1% (w/v) Triton X-100, 10 mM MgCI2, 2mM DTT. Samples were brought to 37°C in a water bath and phosphorylation was initiated by the addition of 200 uM h/ -3 2P]ATP (500-1000 dpm/pmol). The reaction (total volume 20 ul) was allowed to continue for the time indicated in the figure legends and stopped by the addition 30 ul SDS-sample buffer and heating at 95°C for 5 min. Proteins were separated on SDS-PAGE (7.5% acrylamide w/v) and gels were fixed, stained, destained, dried and autoradiographed for 16 hrs at -70°C, as described below. 5.3.2. Peptide Phosphorylation. To determine the activity of the insulin receptor/kinase towards exogenous substrates, peptide phosphorylation was carried out as described under "Autophosphorylation" with the inclusion of 20 2.5 mM angiotensin II (Sigma) in the assay medium. Reactions were stopped by the addition of 30 ul ice cold TCA (5% w/v) and reaction tubes placed on ice, at which time 20 ug of BSA was added to each tube with vortexing and samples allowed to incubate for an additional 15-30 min. The samples were centrifuged at 15,000 x g for 3 min and supernatants spotted onto squares of Whatman P81 phosphocellulose paper that had been presoaked in 10 mM NaH 2P0 4, 2 mM ATP and then dried. The filter paper squares containing bound peptide were washed with 5 changes of distilled water (400-500 mis, 2-5 min per wash), 3 changes of 0.85% H 3 P0 4 (500 mis, 5 min per wash), dried, and then counted in 3.5 mis of ACS liquid scintillation fluid (Amersham). Background counts were estimated by counting filter papers that were spotted with samples containing [ ^ -qp PJATP without peptide, and that were treated in the same manner. Where indicated, following removal of the TCA supernatant, the protein pellets were dissolved in 30 ul of SDS sample buffer by heating at 95°C, subjected to SDS-PAGE and the gels fixed, stained, destained, dried and autoradiographed as described below. In some experiments, the phosphorylation of the synthetic polymer, Glu^-tyr 2 0 (Mr approx. 30 kd) was used instead of angiotensin to assay for receptor/kinase activity. Phosphorylation was carried out as described using 50 ug of Glu ^ -tyr 2 0 peptide (Sigma) per assay. The reaction was terminated by TCA precipitation, as described above for reactions with angiotensin. Precipitated protein pellets were washed once with 0.5 ml 5% TCA and once with the same volume of qp diethylether. Incorporation of [ P] into precipitated proteins was estimated by Cerenkov counting. 5.4. SDS-Polvacrvlamide Gel Electrophoresis (SDS-PAGE). Proteins were dissolved in 30 ul sample buffer (pH 6.8) containing 65 mM Tris, 10% (w/v) SDS, 20% (w/v) sucrose, 0.2 mg/ml bromophenol blue (BDH), 500 mM 2-mercaptoethanol and separated on 0.75 mm thick slab gels by the discontinuous buffer system of Laemmli (1970). Gels 21 (7.5% acyrlamide (Bio-Rad), w/v) were run under constant current at 10 mA per gel for 1 hr and then at 15-20 mA per gel until completion. The gel ahead of the dye front (containing most of the free ft-32P]ATP and [32P]) was excised and the gel was then fixed (20% TCA, 40% methanol, 10 min.), stained (2.5 g/l Coomassie Blue R-250 (Sigma) in 7% acetic acid, 45% methanol, 30 min), destained (7% acetic acid, 45% methanol, 2-3 hrs) and dried between cellophane either at room temperature or on a Bio-rad slab gel dryer. 5.5. Autoradiography. Kodak X-ray film (X-Omat XAR-5) was preflashed and exposed at -70°C to polyacrylamide gels or TLC plates containing radiolabeled samples in a Permacon cassette with intensifying screens (Dupont Hi-plus). After exposure, the film was developed in Kodak developer for 5 min., briefly washed in tap water, fixed for 5 min, washed again in water for 20 min and finally dried. 5.6. Scanning Densitometry. Scanning Densitometry of autoradiographs was performed using a Bio-Rad model 620 Video Densitometer and data was analyzed using the Bio-Rad 1-D Analyst (version 2.0) computer software package. 6. Phosphoamino Acid Analysis. Phosphorylated proteins were separated by SDS-PAGE and the gels directly dried without fixing or staining between cellophane (Bio-Rad) at room temperature overnight and autoradiographed as described, for 4 hrs. The band of interest was excised from the gel and rehydrated in 50 mM NaHC0 3 . The gel chips were then washed in 3 changes of buffer containing 50 mM NaHCOg (40 mis) over a period of 1 hr, cut into small cubes and incubated at 37°C in a minimal volume of the same buffer containing 0.1 mg/ml trypsin (Sigma). After 3 hr, fresh trypsin (1 mg/ml, 0.1 vol.) was added and the sample incubated for a further 12 hr at 37°C. The gel pieces were then removed and washed with a minimal volume of water by incubation for 1 hr at 37°C. The initial and wash 22 supernatants were combined and made acidic with 0.01 volumes of 6N HCL and heated at 95°C for 5 min to inactivate any phosphatase activity. The sample was lyophilized and resuspended in 250 ul 6N HCL, and acid hydrolysis was carried out at 110°C for 90 min. The sample was diluted 5 fold with water, lyophylized, reconstituted in 500 ul of water and relyophilized. The washing/drying procedure was repeated 3 times and the residue finally reconstituted in 10 ul 50% ethanoi and spotted on Kodak cellulose TLC plates together with phosphoamino acid standards (Sigma). The plates were equilibrated in electrode buffer (pyridine:acetic acid:water 1:10:189, pH 3.5) and electrophoresis was performed at 1000 V for 60 min. Phosphoamino acids were detected by autoradiography, and identified by comparison with the standards which were located by spraying the TLC plates with ninhydrin (0.25% w/v in acetone) followed by heating at 80°C for 5 min. 7. Preparation of Rat Adipose Tissue Cvtosol. Male Wistar rats weighing between 150-180 g were stunned and decapitated, and the perirenal and epididymal fat pads from both sides were removed and incubated for 30 min at 37°C in Krebs bicarbonate buffered medium (35 mis/gram of tissue) containing 1.25 mM CaCI2 and 11 mM glucose. The tissue was then transfered into fresh medium which was supplemented with or without insulin (Sigma) (0.5 ug/ml) and the tissue incubated for a further 30 min. Homogenization of tissue was carried out in ice-cold extraction buffer (pH 7.4) containing 250 mM sucrose, 10 mM Tris, 20 mM MOPS, 2 mM EGTA, 2 ug/ml pepstatin A, 2 ug/ml leupeptin, 2.5 mM benzamidine and 4 mM 2-mercaptoethanol. Extraction was carried out in 3 mis of buffer/gram of tissue using a polytron tissue homogenizer (setting 6 for 1-2 sec). The homogenate was centrifuged at 10,000 x g for 2 min, the lipid cake removed and the homogenate centrifuged again at 120,000 x g for 45 min. The resulting supernatant was either used immediately or stored at -70°C overnight and is referred to as the "high-speed supernatant fraction of rat adipose tissue". 23 8. Reconstitution of High-Speed Supernatant Proteins of Adipose Tissue with a WGA-Sepharose- Purified Preparation of Insulin Receptors. Samples of WGA-Sepharose-purified extracts of rat liver membranes (3 ul, containing 10 ug protein) were pre-incubated in the absence or presence of added insulin (Sigma) (10 ug/ml) for 45 min on ice, together with the high-speed supernatant fraction of untreated or insulin-treated rat adipose tissue (10 ul, containing 17 ug protein). Other components of the pre-incubation mixture were HEPES (50 mM, pH 7.6), Triton X-100 (Bio-Rad) (0.05%, w/v), MgCI2 (12 mM) and, where indicated, MnCI2 (2 mM) and NaF (100 mM). The total volume was 20 ul unless otherwise indicated. Samples were briefly (3-5 min.) warmed to 37°C and phosphorylation reactions were initiated by the addition of [ ^ -3 2P]ATP (800 dpm/pmol) to a final concentration of 200 uM. Phosphorylation reactions were allowed to continue for 10 min, at which time the reactions were stopped by the addition of 30 ul SDS sample buffer followed by heating to 95°C for 5 min. Proteins were separated by SDS-PAGE (7.5% acrylamide) and [32P]-labelled proteins identified by autoradiography. 9. Alkali Treatment of Gels. The presence of phosphotyrosine was assessed by phosphoamino acid analysis (described above) and also by alkali treatment of gels. Proteins were electrophoresed on gels (0.75 mm thickness), fixed in 10% glutaraldehyde (BDH) for 30 min and thoughly washed for 3-4 hrs in destaining solution (45% MeOH, 7% acetic acid w/v) with 3 changes of solution. The gel was then briefly rinsed with water and incubated with 1 N NaOH at 55-60°C for 1 hr on a rotory shaker water bath. Following alkali treatment, gels were washed in destaining solution and then stained, destained, dried and autoradiographed, as described. Control gels were subject to identical conditions minus treatment with 1 N NaOH. RESULTS 1. Preparation of Partially Purified Insulin Receptors. The receptor for insulin, as well as other protein-tyrosine kinases (Pike et al., 1984; Kong and Wang, 1987; Foulkes et al., 1985) has been shown to phosphorylate a number of exogenous substrates that include both synthetic (Stadtmauer and Rosen, 1983; Braun etal., 1984; Pike et al., 1982; Casnellie et al., 1982) and natural polypeptides (Kasuga et al., 1983). In the studies described below I have used angiotensin II as a substrate to assay and follow insulin receptor/kinase activity through purification. This peptide binds to phosphocellulose which can be washed free of unbound ATP and inorganic phosphate under acid conditions (Glass et al., 1978). Using untreated phosphocellulose paper the nonspecific binding of ATP represents as much as 50% of the radioactivity associated with insulin-stimulated peptide phosphorylation, even after extensive washing. However, this can be reduced to levels as low as 10% of the incorporation into angiotensin if the filter papers are presoaked with 10 mM NaH 2P0 4, 2 mM ATP and dried prior to use. Table I shows the specific activity of the insulin receptor/kinase following purification from rat liver by detergent-solubilization of isolated membranes and chromatography over wheat germ aggutinin bound to Sepharose. The receptor is typically enriched 10-20 fold after WGA-Sepharose affinity chromatography with approximately 50% yield. The yield may be increased by using a column of higher capacity. 2. Protein Kinase Activity Associated with the Insulin Receptor. Insulin receptors purified from rat liver through WGA-Sepharose chromatography are capable of catalyzing the ATP-dependent phosphorylation of angiotensin in both the absence or presence of insulin (fig. 3). Preincubation of the extract with insulin resulted typically in a 2-5 fold stimulation of receptor/kinase activity toward angiotensin at all time points and was observed consistantly in all preparations. At 37°C phosphorylation of peptide was linear for 5-10 min under both basal and 25 Table I Partial Purification of the Insulin Receptor. Protein Tot. Act. Spec. Act. Purification Yield (mg) (pmol/min) (pmol/min/mg) (-fold) (%) TE 312 12,936 41.5 1 100 WGE 8.3 6,080 732 17.6 47 TE- Triton-solublized extract. WGE- WGA-Sepharose-purified extract. 26 Fig. 3. Time Course of [P]-Incorporation into Angiotensin Catalyzed by Insulin Receptor/Kinase. The WGA-Sepharose-purified extract from rat liver membranes (40 ug protein) was preincubated without (A) or with (•) insulin (10 ug/ml) for 45 min at 23°C in the presence of buffer (pH 7.6) containing 50 mM Hepes,0.05% (w/v) Triton X-100, 2.5 mM angiotensin and 10 mM MgCI2 in a final volume of 80 ul. The samples were brought to 37°C and phosphorylation was initiated by the addition of [y-32P]ATP (850 dpm/pmol) to a final concentration of 200 uM. At the indicated time intervals 20 ul aliquots were removed, the reactions quenched, and the samples assayed as described in Materials and Methods. Values are the mean of two experiments, where 100% represents 45,281 cpm and 31,645 cpm. 27 2 8 insulin-treated conditions. This phosphorylation represents solely tyrosine kinase activity since the amino acid sequence of angiotensin includes neither serine nor threonine. As is the case with many protein kinases (Krebs, 1986; Flockhart and Corbin, 1982), the insulin receptor/kinase undergoes an autophosphorylation reaction which is insulin responsive (White et al., 1984; Gammeltoft and Van Obberghen, 1986). Fig. 4a shows an autoradiograph of proteins phosphorylated in the Triton-solubilized extract of rat liver membranes and in the same extract after WGA-Sepharose chromatography, following incubation at 0°C with []f-32P]-labelled ATP. In both fractions, a major phosphoprotein band appears at 95,000 daltons and undergoes an qp increase in [^Pj-labelling by 2-5 fold in response to insulin. It has previously been demonstrated that this 95 kd subunit represents the -^subunit of the insulin receptor (Avruch et al., 1982; Kasuga et al., 1982b; Kasuga et al., 1982c; Petruzzelli et al., 1982), which contains the insulin-dependent tyrosine kinase activity (Roth and Cassell, 1983). When phosphorylation was carried out at 37°C other phosphoprotein bands appeared in both the Triton-solubilized extract of liver membranes and the extract after WGA-Sepharose chromatography (fig. 4b), suggesting that temperature effects must be considered for maximal expression of protein kinase activity in these fractions. It is o interesting that at the higher temperature (37 C), the insulin receptor present in the Triton-solubilized extract shows only a moderate (if any) increase in autophosphorylation in response to insulin, when compared with the insulin response observed in the WGA-Sepharose purified extract (fig. 4b). This is also reflected in the lower insulin-sensitivity of the receptor when the Triton-solubilized membrane extract was assayed by the phosphorylation of angiotensin. The reason for this is unclear but may be attributable in part to a high degree of serine kinase activity which may act on the insulin receptor. Studies have shown that serine phosphorylation of the insulin receptor -^subunit inhibits the receptor/tyrosine kinase activity and diminishes its responsiveness to insulin (Stadtmauer et al., 1986; Bollag et al., 1986; Haring et al., 1986). The greater insulin responsiveness of the receptor at low temperature may reflect a temperature sensitivity of this serine kinase activity. 29 Fig. 4. Phosphorylation of the Insulin Receptor /J-Subunit. The Triton-solubilized extract from rat liver membranes (TE) (25 ug protein) or the WGA-Sepharose-purified extract (WGE) (10 ug protein) was preincubated without (-) or with (+) insulin (10 ug/ml) for 45 min at 23°C in the presence of buffer (pH 7.6), containing 50 mM Hepes, 0.05% (w/v) Triton X-100, 2 mM DTT and 10 mM MgCI2 in a final volume of 20 ul. The samples were brought to 0°C (A) or 37°C (B) and phosphorylation was initiated by the addition of b/-32P]ATP (850 dpm/pmol) to a final concentration of 200 uM. After 10 min the reaction was quenched, the samples run on SDS-PAGE and gels autoradiographed as described in Materials and Methods. The results of TE and WGE assayed at 0°C, and TE assayed at 37°C are typical of 3 experiments. Results of WGE assayed at 37°C are typical of 10 experiments. 30 31 Phosphoaminoacid analysis of the receptor -^subunit indeed confirms the presence of a high degree of serine as well as tyrosine phosphorylation when isolated from the extract of Triton-solubilized rat liver membranes incubated with [v/ -3 2P]ATP. However, if the Trfton-solubilized extract is partially purified over WGA-Sepharose prior to phosphorylation the ratio of phosphoserine to phosphotyrosine is decreased approximately 16-fold (8.8 to 0.56 arbitrary units) (fig. 5). These observations are in general agreement with those reported in the literature, where others have further shown that insulin receptors highly purified by insulin-Sepharose affinity chromatography, or anti-insulin receptor antibody precipitation, undergo autophosphorylation only on tyrosine residues (Pang et al., 1985; Kasuga et al., 1983a). 3. Investigation of the Possible Effects of the Insulin Receptor/Tyrosine Kinase on the Activity of Protein Serine Kinases in a Reconstituted System. Protein kinase activity towards a select number of substrates in the high-speed (100,000 x g) supernatant fraction prepared from rat epidiymal adipose tissue is increased when the intact tissue is pre-treated with insulin. This insulin-stimulated protein-serine kinase has been assayed by following °P-incorporation into phosphoproteins in high-speed supernatant fractions incubated 32 with [tf- P]ATP (Brownsey et al., 1984). In this system the most consistent and dramatic effects are seen with ATP-citrate lyase which migrates at approximately 120,000 daltons on SDS-PAGE, and which typically undergoes a 1.5-5 fold increase in phosphorylation in response to insulin (fig. 6). The phosphorylation of acetyl-CoA carboxylase (Brownsey et al., 1977) (220-240 kd) is also insulin responsive but to a maximum extent of 10-40%, which is often not detectable on SDS-PAGE. In the following experiments I have investigated the possibility that the insulin-dependent phosphorylation of ATP-citrate lyase may be observed by reconstituting a high-speed (100,000 x g) supernatant fraction prepared from non-hormone-treated adipose tissue with the WGA-Sepharose-32 Fig. 5. Phosphoamino Acid Analysis of the Insulin Receptor ^ -Subunit. The Triton-solublized extract of rat liver membranes (TE) (600 ug protein) and the extract after purification over WGA-Sepharose (WGE) (60 ug protein) was preincubated with insulin (10 ug/ml) for 45 min at 23°C in the presence of buffer (pH 7.6) containing 50 mM Hepes, 0.05% (w/v) Triton X-100 and 10 mM MgCI2 in a final volume of 120 ul. The reaction mixture was brought to 37°C and phosphorylation was initiated by the addition of [/-32P]ATP (3000 dpm/pmol) to a final concentration of 200 uM. After 20 min the reaction was stopped with 180 ul SDS sample buffer and heated at 95°C for 15 min. Phosphoamino acid analysis was carried out as described in Materials and Methods after excising the 95 kd band from SDS-polyacrylamide gels. 33 P-Ser • P-Tyr • 4 \< P-Ser < P-Tyr TE WGE < Orig 34 Fig. 6. Incorporation of [J P] from ^P]ATP into High-Speed Supernatant Proteins of Rat Adipose Tissue. The high-speed supernatant fraction (17 ug protein) from rat adipose tissue (pretreated without (-) or with (+) insulin) was incubated (10 min, 37 C) in extraction buffer (Materials and Methods) to which was added 10 mM MgCI2 and 200 uM h/-32P]ATP (850 dpm/pmol). The reaction was terminated by the addition of SDS-sample buffer and heating at 95°C for 5 min. Samples were subjected to SDS-PAGE (7.5% polyacrylamide, w/v) and autoradiography as described in Materials and Methods. ACC and ACL represent the migration of acetyl-CoA carboxylase and ATP-citrate lyase, respectively. The results shown are typical of at least 6 experiments. 35 + ins. 36 purified extract of liver membranes that contain insulin receptors. In fig. 7 it can be seen that mixing of the two fractions resulted in the marked inhibition of the activity of protein kinases in the adipose tissue high-speed supernatant fraction. Initially, therefore, it was essential to develop assay conditions that would allow optimal coexpression of protein-serine kinase activity in the high-speed supernatant fraction of adipose tissue, and insulin receptor/tyrosine kinase activity in the WGA-Sepharose-purified extract of solubilized rat liver membranes. 3.1. Effect of Assay Components on Protein Kinase Activity. 3.1.1. Detergents. The insulin receptor, being an integral membrane protein, requires the presence of small amounts of detergent for full expression of activity (Aiyer, 1983). In most studies of insulin receptor/kinase activity, Triton X-100 is commonly used at concentrations beween 0.02% and 0.05% (w/v) (Harrison et al., 1978). Figure 8 shows the effect of Triton X-100 on protein kinase activity in the high-speed supernatant fraction of adipose tissue. There is no apparant inhibition of phosphorylation using Triton X-100 at concentrations of 0.05% and 0.1% (w/v). Two other common detergents, Nonidet P-40 and octyl glucoside, were also tested, with similar results. Thus, the presence of detergent was unlikely to account for the observed inhibition of protein-serine kinase activity in the high-speed supernatant fractions of adipose tissue. 3.1.2. Protein Phosphatase Inhibitors. Both NaF and 2-glycerophosphate are commonly used in phosphorylation studies to inhibit protein phosphatase activity (Wettenhall and Morgan, 1984). In my hands, 100 mM NaF had no effect on the autophosphorylation of the insulin receptor/kinase. However, as shown in fig. 9, concentrations of NaF as low as 40 mM caused dramatic inhibition of the activity of protein kinases in the high-speed supernatant fraction of adipose tissue. In the presence of 100 mM NaF all protein kinase activity in the adipose tissue fraction, except for that represented by phosphorylation of one persistent band, was essentially abolished. 37 Fig. 7. Effects of Receptor Assay Components on the Incorporation of [ P] into High-Speed Supernatant Proteins of Rat Adipose Tissue. Incorporation of [ P] from [/-32P]ATP into high-speed supernatant proteins (17 ug) of untreated or insulin-treated adipose tissue was studied employing standard extraction buffer (Materials and Methods) + 10 mM MgCI2 (EB), or after addition of receptor assay components (50 mM Hepes pH 7.6, 0.05% (w/v) Triton X-100, 10 mM MgCI2, 2 mM MnCI2,100 mM NaF) (EB + RR), or in the presence of receptor assay components alone (RB). In the latter case, supernatant proteins were reconstituted in receptor buffer by exchange using a Sephadex G-50 mini-column. Reactions were initiated by the addition of 200 uM [Jf-32P]ATP and samples processed as described in the legend to fig. 7. ACL represents the migration of ATP-citrate lyase. Results shown are typical of two experiments. 38 39 Fig. 8. Effects of Triton X-100 on the Incorporation of [ P] into High-Speed Supernatant Proteins of Rat Adipose Tissue. Incorporation of [32P] from [/-32P]ATP into high-speed supernatant proteins (17 ug) of untreated or insulin-treated adipose tissue was studied employing standard extraction buffer (Materials and Methods) + 10 mM MgCL (left lanes), or with the addition of 0.05% (center lanes) or 0.1% (w/v) (right lanes) Triton )£l00. Reactions were initiated by the addition of 200 uM P]ATP and samples processed as described in the legend to fig. 7. ACL represents the migration of ATP-citrate lyase. Results shown are typical of two experiments. 41 The effect of 2-glycerophosphate on protein-serine kinase activity in the adipose tissue high-speed supernatant fraction was more selective, but nonetheless highly significant for the objective of op these studies. Its specific effect in inhibiting the [ PJ-labelling of ATP-citrate lyase can be seen in fig. 9. [ P]-incorporation into ATP-citrate lyase was completely inhibited at a concentration of 60 mM 2-glycerophosphate, with little or no apparant effect on other protein kinase activity. The results were similar when the concentration of 2-glycerophosphate was increased to 80 mM. 3.1.3. Divalent Metal Ions. Divalent cations are essential for the full activity of protein kinases (Krebs, 1986). The incorporation of [32P] from [ y'-32P]ATP into proteins of the high-speed supernatant fraction of adipose tissue incubated alone or with the addition of 10 mM MgCI2 is shown in fig 10 (left lanes vs center lanes). Thus, it can be seen that optimum expression of overall protein kinase activity in P i these fractions requires the presence of Mg . All assays, therefore, have been carried out with the inclusion of 10 mM MgCI2. The low level of phosphorylation that occurs without addition of metal ion is due to the small amount of MgCI2 introduced with the ATP (less than 500 uM final concentration), and is completely abolished by further addition of 1 mM EDTA. p i The presence of Mn has been reported to be required for optimum autophosphorylation of the receptors for EGF (King et al., 1980), PDGF (Ek and Heldin, 1982), and insulin (White et al., 1984; Nemenoff et al., 1984), and also for the pp60 v" s r c oncogene product (Richert et al., 1982) in vitro . For this reason concentrations of MnCI2 between 0.5 and 2.0 mM are normally included in assays of insulin receptor/kinase activity. However, when 2 mM MnCI2 was included in the high-speed supernatant fraction of adipose tissue it was found to dramatically inhibit protein phosphorylation, even in the presence of 10 mM MgCI2 (fig. 10, center lanes vs right lanes). To estimate the maximum concentration of MnCI2 that protein kinases in the high-speed supernatant fraction of adipose tissue can tolerate without apparant inhibition, several concentrations of MnCU were 42 Fig. 9. Effects of NaF and 2-glycerophosphate on the Incorporation of [° P] into High-Speed Supernatant Proteins of Rat Adipose Tissue. Incorporation of [ P] from [^-32P]ATP into high-speed supernatant proteins (17 ug) of untreated or insulin-treated adipose tissue was studied employing standard extraction buffer (Materials and Methods) + 10 mM MgCL (left lanes), or with the addidtion of NaF or 2-glycerophosphate at various concentrations: 10 mM NaF, (a,a'); 40 mM NaF, (b,b'); 100mM NaF, (c,c'); 10 mM 2-GP, (d,d'); 60 mM 2-GP, (e,e'); 80 mM 2-GP, (f,f). Reactions were initiated by the addition of 200 uM [v'-32P]ATP and samples processed as described in the legend to fig. 7. ACC and ACL represent the migration of acetyl-CoA carboxylase and ATP-citrate lyase, respectively. 43 I I 44 Fig. 10. Effects of 2 mM MnCI2 on the Incorporation of [ P I into High-Speed Supernatant Proteins of Rat Adipose Tissue. Incorporation of [32P] from [Y-2P]ATP into high-speed supernatant proteins (17 ug) of untreated or insulin-treated adipose tissue was studied employing standard extraction buffer (Materials and Methods) with the addition of no divalent metal ions (left lanes), 10 mM MgCI2 (center lanes), or 10 mM MgCI2 + 2 mM MnCI2 (right lanes). Reactions were initiated by the addition of 200 uM ry-32P]ATP and samples processed as described in the legend to fig. 7. ACL represents the migration of ATP-citrate lyase. Results are typical of two experiments. 45 46 tested (fig. 11). As shown, MnCI2 inhibited [ P]-incorporation into all major phosphoprotein bands, including acetyl-CoA carboxylase and ATP-citrate lyase even at 0.05 mM MnCI2. At a concentration of only 0.5 mM MnCI2 the labelling of many high molecular weight bands was completely abolished. Studies by Brownsey et al. (1984) have shown that protein phosphatase activity towards ATP-citrate lyase in the high-speed supernatant fraction of adipose tissue incubated with MgCI2 is very low over a 30 min period, compared to the activity of protein kinases which act on the substrate. However, in the studies described here it could be argued that decreased phosphorylation in the presence of M n 2 + may be due to the activation of a Mn2+-dependent phosphatase. To address this question, the experiment illustrated in fig. 12 was carried out. Following [32P]-incorporation into proteins of the adipose tissue high-speed supernatant fraction, subsequent dephosphorylation was observed to occur in both the absence (a-e) and presence (a'-e') of 2.0 mM MnCI2. However, the rate of dephosphorylation in the presence of MnCI2 was observed to be no different than that observed in the absence of added MnCI2, indicating that this low level of protein phosphatase O x activity is not Mn' -dependent over the time period tested (20 min). Given the effects of even small quantities of MnCI2 on the protein kinase activity in the high-speed supernatant fraction of adipose tissue, it was necessary to deterimine whether the insulin receptor/kinase could function in the complete absence of Mn . To ascertain this, both protein-tyrosine kinase activity toward angiotensin (fig. 13a) and receptor autophosphorylation (fig. 13b) were measured over a range of MnCI2 concentrations. As a result, several observations are worthy of mention. First, under these conditions maximum enzyme activity is apparant at 0.5 mM MnCI2. Secondly, enzyme activity and perhaps insulin responsiveness, is actually decreased at higher concentrations of MnCI2 (up to 2 mM). Most important is the finding that the insulin receptor/kinase is both active and insulin-responsive in the complete absence of MnCU-47 Fig. 11. Effects of Various Concentrantions of MnCI2 on Incorporation of [ P] into High-Speed Supernatant Proteins of Rat Adipose Tissue. Incorporation of [32P] from [/-32P]ATP into high-speed supernatant proteins (17 ug) of untreated or insulin-treated adipose tissue was studied employing standard extraction buffer (Materials and Methods) in the presence of 10 mM MgCI2 and various concentrations of MnCI2 as indicated. Reactions were initiated by the addition of 200 uM [jf-32P]ATP and samples processed as described in the legend to fig. 7. Lanes are arranged in pairs representing samples obtained from tissue that was either untreated (left) or treated with insulin (right), a) 0; b) 0.05; c) 0.1; d) 0.2; e) 0.5 mM MnCI2. ACL represents the migration of ATP-citrate lyase. Results are typical of two experiments. 48 ins. - + - + - + - + - + MnCI2 0 .05 J .2 .5 mM 49 Fig. 12 Effects of MnCI2 on Protein Phosphatase Activity of Rat Adipose Tissue High-Speed Supernatant. The high-speed supernatant fraction from insulin-treated adipose tissue was incubated in extraction buffer + 10 mM MgCI2 at 37°C as described in Materials and Methods with the addition of 50 uM [^-32P]ATP (850 dpm/pmol). After 5 min incubation, non-radiolabelled ATP was added to a final concentration of 500 uM followed by the addition of either 2 ul dH 2 0 (a-e), or 2 mM MnCI2 (final concentration) (a'-e'). The concentration of all other assay components were the same between control samples and samples containing M n 2 + . After addition of either dH 20 or MnCI2, as described, one 20 ul aliquot (17 ug protein) was immediately removed, added to SDS-sample buffer (30 ul) and heated at 95°C for 5 min (a,a'). Further aliquots were removed and reactions stopped at 2 (b,b'), 5 (c,c'), 10 (d,d') and 20 min (e.e'). Molecular weight markers are expressed in kd. 50 200 -b c d e a' b' c' d 51 Fig. 13. Effects of MnCI2 on Insulin Receptor/Kinase Activity. A) Insulin receptor/kinase activity in the WGA-Sepharose-purified extract of rat liver membranes was measured by [ P]-incorporation into angiotensin (2.5 mM) in the presence of 10 mM MgCL and the indicated concentrations of MnCI2 (as described in Materials and Methods under peptide phosphorylation"). Values represent the mean±S.D. of 3 independant experiments where 100% = 57,540; 38,980 and 24,740 cpm. B) Proteins in the TCA precipitated pellets from (A) were separated on SDS-PAGE, and the gels autoradiographed as described in Materials and Methods under "SDS-PAGE". Relative receptor autophosphorylation was estimated by scanning densitometry of the 95 kd band. Values represent the mean of two independant experiments, (A), - insulin; (•), + insulin. 52 MnCl2 Concentration (mM) 32p_incorporation into 95 kd band (Percent of Maximum) 54 3.2. Reconstitution of High-Speed Supernatant Fractions of Adipose Tissue with Partially Purified  Insulin Receptors. The studies described above have led to the development of assay conditions which allow co-expression of protein-serine kinase activity present in high-speed supernatant fractions of adipose tissue and insulin receptor/tyrosine kinase activity in the WGA-Sepharose-purified extract of solulbilized liver membranes (fig. 9-13). The effects of insulin receptor/tyrosine kinase activity on protein-serine phosphorylation in adipose tissue high-speed supernatant was therefore examined and is described below. The high-speed supernatant fraction of adipose tissue was mixed with the WGA-Sepharose purified extract of liver membranes, and the effects of activated insulin receptors on 32P-labelling of proteins was observed. Under these assay conditions, the addition of adipose tissue high-speed supernatant buffer medium to the WGA-Sepharose-purified extract had no effect on insulin receptor autophosphorylation (fig. 14, b vs a). Similarly, buffer components of the WGA-Sepharose-purified extract did not impair the functioning of protein-serine kinases in the high-speed supernatant fraction of adipose tissue (fig. 14, d vs c). Samples (e) and (f) of fig. 14 show high-speed supernatant fractions, prepared from adipose tissue treated without (e) or with (f) insulin, which have been reconstituted with a WGA-Sepharose-purified preparation of insulin receptors. The left lane of each pair represents reconstitution with untreated control receptors, while the right lane shows reconstitution with insulin-treated receptor fractions. The most notable effect of reconstitution was the altered level of phosphorylation of two low molecular weight protein subunits. Most obvious was the dephosphorylation of a 25 kd subunit present in the high-speed supernatant fraction of adipose tissue, upon reconstitution with WGA-Sepharose-purified insulin receptors (fig 14, c,d vs e,f). Of further interest was the qp appearance of a 40 kd [ P] -labelled protein upon reconstitution with the insulin-activated WGA-Sepharose-purified insulin receptor (fig. 14, right lanes of e,f). This band was not apparant in either 55 Fig. 14. Incorporation of [J P] from b/- P]ATP into Proteins Upon Reconstitution of the High-Speed Supernatant Fraction of Rat Adipose Tissue with the WGA-Sepharose-Purified Extract of Rat Liver Membranes. Samples of the WGA-Sepharose-purified extract of liver membranes (WGE) (10 ug protein) (a,b) were treated without or with insulin and then incubated (10 min, 37°C) with 12 mM MgCI2 and 200 uM Gf-32P]ATP as described in Materials and Methods under "Autophosphorylation". Samples of the high-speed supernatant fraction of adipose tissue (HSS) (17 ug protein) (c,d) were incubated with 200 uM LY- P]ATP as described in the legend to figure 7. Reconstitution of the two fractions (e,f) was carried out as described in Materials and Methods. Pairs of lanes are shown as untreated and insulin-treated fractions, respectively, a) WGE in receptor buffer (50 mM Hepes pH 7.6, 0.05% (w/v) Triton X-100, 10 mM MgCI2, 2 mM DTT). b) WGE in receptor buffer + HSS extraction buffer (see Materials and Methods), c) HSS in HSS extraction buffer, d) HSS in HSS extraction buffer + receptor buffer, e) WGE (-/+ insulin) + HSS prepared from untreated tissue, f) WGE (-/+ insulin) + HSS prepared from insulin-treated tissue. Results shown are typical of at least 5 separate experiments. 56 O a a • s 53 w to a c r * * t I l I I i * H i • CM l + + I 57 the WGA-Sepharose-purified preparation of insulin receptors or the high-speed supernatant fraction of adipose tissue when these fractions were assayed alone. Unfortunately, effects of insulin receptor/tyrosine kinase activity on the serine phosphorylation of ATP-citrate lyase in this system could not be interpreted, due to contaminating phosphoproteins present in the WGA-Sepharose-purified membrane extract which mask ATP-citrate lyase. It is of interest, however, that a non-buffer component in the adipose tissue high-speed supernatant has some inhibitory effect on insulin receptor/tyrosine kinase activity towards angiotensin (fig. 15). This inhibition may be correlated with the observed lower level of receptor autophosphorylation (as much as 3-fold, estimated by scanning densitometry) of the 95 kd band (fig. 14, a,b vs e,f), consistent with the theory that tyrosine phosphorylation of the insulin receptor is required for its activation (Rosen et al., 19983; Yu and Czech, 1984). The reconstitution experiment was repeated as described in the legend to fig. 14, but each sample was assayed in duplicate and electrophoresed on two separate gels under identical conditions. Thus, fig. 16a is a repeat of the experiment shown in fig. 14. However, fig. 16b represents the same gel following alkali digestion prior to autoradiography. The bands remaining after alkali-treatment represent proteins containing phosphotyrosine residues, as alkali digestion hydrolyzes most qp phosphoserine and phosphothreonine residues. Therefore, after [ P]-labelling, proteins in the WGA-Sepharose-purified membrane extract appear to be phosphorylated primarily on tyrosine residues, represented by alkali-resistant bands (fig. 16b a,b), whereas phosphoproteins in the high-speed supernatant fraction of adipose tissue are mainly the result of serine/threonine phosphorylation demonstrated by their alkali lability (fig. 16 c,d). When adipose tissue high-speed supernatant reconstituted with insulin receptor is analyzed in the same manner (fig. 16 e,f) the observed 40 kd phosphoprotein band is shown to be alkali resistant suggesting that it is phosphorylated on tyrosine (right lanes of e,f). Estimation by scanning densitometry reveals that qp qp the total [P]-incorporation into pp40 is typically 16% and 15% of the [ Reincorporation into the 58 Fig. 15. Effects of High-Speed Supernatant Fractions of Adipose Tissue on Insulin Receptor/Kinase Activity in a Reconstituted System. Reconstitution of the high-speed supernatant fraction of rat adipose tissue (HSS) (17 ug protein) with WGA-Sepharose-purified extracts of rat liver membranes (WGE) (10 ug protein) was carried out as described in Materials and Methods. Insulin receptor/kinase activity was determined by incorporation of [32P] into angiotensin (2.5 mM) in the presence of 10 mM MgCI2 and 200 uM f/-32P]ATP as described in Materials and Methods under "peptide phosphorylation". (a,b) insulin-treated WGE and untreated WGE, respectively. (c,d) insulin-treated and untreated WGE, respectively + HSS obtained from non-hormone-treated adipose tissue. Results shown are the mean of two independant experiments. 59 60 Fig. 16. Estimation of Protein Tyrosine Phosphorylation by Alkali Stability Following Reconstitution of Insulin Receptors with Adipose Tissue High-Speed Supernatant Fractions. WGA-Sepharose-purified extracts of rat liver membranes (WGE), high-speed supernatant fractions of rat adipose tissue (HSS) and the two fractions after reconstitution were incubated in the presence of 10 mM MgCI2 and 200 uM h/-32P]ATP as described in the legend to fig. 14, and electrphoresed in duplicate on separate gels under identical conditions. Gels were subjected to alkali digestion as described in Materials and Methods. A) Before alkali treatment. B) After alkali treatment, a) WGE in receptor buffer (50 mM Hepes pH 7.6, 0.05% (w/v) Triton X-100,10 mM MgCL, 2 mM DTT). b) WGE in receptor buffer + HSS extraction buffer (see Materials and Methods). cfHSS in HSS extraction buffer, d) HSS in HSS extraction buffer + receptor buffer, e) WGE (-/+ insulin) + HSS prepared from untreated tissue, f) WGE (-/+ insulin) + HSS prepared from insulin-treated tissue. IR represents migration of the insulin receptor f-subunit. Results shown are typical of at least 3 independant experiments. 61 6 2 insulin receptor -subunit, before and after alkali-treatment, respectively. The labelling of pp40 is not significantly reduced by alkali-treatment, suggesting that pp40 is phosphorylated mostly on op tyrosine. Table II summarizes the levels of [P]-incorporation into the insulin receptor j^-subunit, ATP-citrate lyase and pp40 as observed in fig. 16a and 16b. 4. Purification of the Insulin Receptor to Homogeneity. In order to detect any possible effects that solubilized insulin receptors might have on altering the serine phosphorylation of ATP-citrate lyase or other proteins in a cell-free system, receptors of a higher degree of purity were found to be required. Reconstitution of the high-speed supernatant fraction of adipose tissue with homogeneous insulin receptors might also reveal whether the 40 kd phosphoprotein is endogenous to the adipose tissue high-speed supernatant fraction or the WGA-Sepharose-purified membrane extract. I therefore attempted to purify the insulin receptor to homogeneity. Affinity purification of the insulin receptor using insulin immobilized on agarose was first achieved by Cuatrecasas (1972b). Over the years the method has been refined and is now routine in many laboratories. I have found however, the practical aspects of the technique to be less straight forward than that which many published methods describe (Fugita-Yamaguchi et al., 1983; Petruzzelli et al., 1984; Pike et al., 1986). Thus, at present, although I have been able to recover insulin receptor/kinase activity following insulin-agarose affinity chromatography (fig. 17), the yield of receptor following purification has been insufficient for extensive use. The following describes the technique which has been the most successful so far. 4.1 Construction of Insulin-Bound Agarose. 4.1.1. Synthesis of Boc-lnsulin. The insulin molecule contains three primary amine groups located at glycine A1, phenylalanine B1 and lysine B29, all of which can undergo esterification to form amide bonds. From studies of 63 Table II Effects of Insulin on the Incorporation of \° P] into the Insulin Receptor fl-Subunit, ATP- Citrate Lyase and pp40. Samples of the WGA-Sepharose-purified extract of rat liver membranes (WGE), high-speed supernatant fraction of rat adipose tissue (HSS) and the reconstituted fractions (WGE + HSS) were incubated with h/- P]ATP and assayed as described in the legend to fig. 14. [ P] -incorporation was quantitated by scanning densitometry of autoradiographs and expressed as the mean(n) or the mean ± S.D.(n) where n = no. of samples. Results of reconstitution are shown only for the untreated HSS reconstituted with untreated or insulin-treated WGE as indicated (fig. 14, e). Results were essentially the same for the insulin-treated HSS (fig. 14, f). ND, not determined; IR, Insulin Receptor; ACL, ATP-citrate lyase. Before Alkali Wash: Ins. Relative Incorporation of f°^P] into: IR/3-sub ACL pp40 WGE - 1 + 4.7*2.2(8) HSS - 1 + - 2.7 ±1.4(6) WGE + 1 ND 1 HSS + 3.2*0.9(4) ND 2.9±0.9(4) After Alkali Wash: WGE 1 + 3.2(2) HSS - - ND -+ - ND -WGE + - 1 ND 1 HSS + 7.1(2) ND 4.8(2) 64 Fig. 17. Insulin-Agarose Affinity Purified Insulin Receptor. Insulin receptor was purified from rat liver by detergent-solublization of a crude membrane fraction followed by WGA-Sepharose affinity chromatography, as described in Materials and Methods. The partially pure receptor was further purified over insulin-immobilized agarose (Affigel-10) and concentrated by adsorption to DEAE-cellulose as described in Results. Insulin receptor/kinase activity in the DEAE eluted fraction was assayed by [32P]-incorporation into the insulin receptor -^subunit after preincubation with 10 ug/ml insulin (45 min, 23 C) and subsequent incubation with 10 mM MgCI2 and 200 uM 3 P]ATP, as described in Materials and Methods under "autophosphorylation". [32P]-incorporation into the 95 kd band represents insulin receptor/kinase activity in 20 ul of extract out of a total of 100 ul obtained, as shown by subjecting SDS-polyacrylamide gels to autoradiography for 8 days at -70°C, as described in Materials and Methods. 120 95 66 modified insulins (Pullen et al., 1976), it has been found that substitutions at positions B1 and B29 exert minor changes in the biological activity of insulin, whereas substitution at position A1 renders the molecule virtually inactive. Studies involving the synthesis of biotinylinsulin describe the coupling of biotin to "boc-insulin", where positions A1 and B29 of the native insulin molecule are blocked by reaction with tert-n-butoxycarbonyl azide (Hoffman et al., 1977; Geiger et al., 1971). This treatment surprisingly leaves the amine group at B1 free, to undergo reaction with biotin or any other group (Geiger, 1971). To my knowledge, there have been no reports on the use of boc-insulin for the coupling of insulin to agarose. But after several unsuccessful attempts at binding receptor/kinase activity to insulin-agarose conventionally made according to published protocols, I decided to construct a column using the novel boc-insulin. To synthesize boc-insulin (Geiger et al., 1971), porcine insulin (40 mg, Sigma) was dissolved in a mixture containing 1.5 mis of dimethylformimide, 0.38 ml distilled H 2 0 and 0.11 ml 1N NaHC0 3 at 23°C. Following addition of tert-n-butoxycarbonyl azide (Aldrich, 77 ul) the mixture was stirred at 35°C for 5 hrs. The mixture was acidified with 50% acetic acid (133 ul), divided into 4 eppendorf microfuge tubes, and evaporated to dryness in a rotory Speedvac at 40°C. Each of the clear gelatinous pellets was then ground with a pestle in 0.5 ml diethyl ether and washed by vigorous vortexing. This was repeated three times. The ether was evaporated by placing samples in a water bath (35 C) and the pellets washed again twice, with 500 ul per tube of 1% acetic acid. The boc-insulin was then finally dried in a rotory Speedvac at 40°C. Overall recovery was near 100%. To test whether insulin had been fully blocked its reactivity towards ninhydrin (Sarin et al., 1981) before and after blocking was estimated. To 700 ug boc-insulin or 600 ug insulin was added 4 drops of a mixture consisting of phenol:ethanol (75:25 v/v), 8 drops 0.2 mM KCN in pyridine and 4 drops 5% ninhydrin in ethanol. The samples were heated at 100°C for 10 min and then brought to 3 mis with 60% ethanol in water. Any precipitate was allowed to settle. The number of free amino groups before and after blocking was estimated by measuring the absorbances at A c 7 n . correcting 67 for the amount of sample originally present. If insulin had been properly blocked the number of free amine groups before blocking should theoretically be 3 times that after blocking. Indeed, native insulin reacted 3.8 times more strongly with ninhydrin than the synthesized boc-insulin, indicating that it had been fully blocked. 4.1.2. Coupling of Boc-lnsulin to Affiael. Boc-insulin was coupled to Affigel-10 (Bio-Rad) according to the manufacture's instructions. Resin (25 mis packed vol) was washed with approximately 20 volumes of cold distilled water, and then with 5 volumes of coupling buffer (40 mM MOPS, 80 mM CaCI2, 6 M urea (Bio-Rad), pH 7.0) in a buchner funnel. Boc-insulin was dissolved in 25 mis of the same buffer and incubated with the washed resin on a rotory shaker for 15 hrs at 4oC. The resin was then poured into a column and washed free of unbound protein with appoximately 100 mis of 10 mM Hepes, pH 7.4, and then with 10 volumes of 10 mM Na2HP04, 0.02% NaN3, pH 7.4. The extent of coupling was 77% (1.6 mg/ml) estimated by Bradford assay. 4.1.3. Deprotection of Boc-lnsulin Coupled to Agarose. An important potential problem associated with this method of coupling insulin to agarose, might be the possible instability of the resin under the harsh conditions of deprotection. Complete deprotection is normally achieved using 50% trifluoroacetic acid for 10 min at 23°C (Ian Clark-Lewis, personal commun.). However, under these conditions consideration must be given to the glycosidic bonds of agarose which are susceptible to acid hydrolysis. Table III shows the extent of deprotection of boc-insulin-agarose after various times of incubation with 50% TFA. After 3 min of TFA treatment deprotection appears to have reached a maximum with no observable effects on the integrity of the resin. After 10 min of incubation, however, the number of ninhydrin reactable groups has actually decreased, indicating that some of the ligand had been lost, perhaps due to hydrolysis. Therefore, boc-insulin-agarose was deprotected as described in the legend to 68 Table III Deprotection of Boc-lnsulin Bound to Agarose. One ml of boc-insulin coupled to Affigel-10 was continuously stirred in 10 mis of 50% TFA on ice for the indicated time. The resin was washed extensively with water and then with 0.01 N NH^OH in a buchner funnel, and finally with ethanol to facilitate drying. The resin was then dried under vacuum for 45 min. Reactivity towards ninhydrin was carried out as described in the text (Sarin et al., 1981). The umoles of amino groups per gram of sample was calculated according to the equation: umoles NH2/gm = A 5 7 Q x 200 / ug sample Boc-lnsulin Treatment with TFA (min) umoles NH2/gm 0 3 5 10 23.5 49.3 45.3 34.7 69 Table III for 3 min, washed as described, and washed again with phosphate buffer containing azide (0.02%, w/v) and stored in the same buffer until use. 4.2. Insulin-Aqarose Affinity Chromatography. Approximately 20 mis of insulin-agarose was poured into a column and washed with 500 mis of acid buffer (0.1 M NaAcetate, 0.5 M NaCI, pH 4.0), 500 mis of basic buffer (0.1 M NaCOg, 0.5 M NaCI, pH 8.3) and finally equilibrated with 2 L of column buffer (25 mM Hepes, 0.2 M NaCI, 10% glycerol, 0.1% (w/v) Triton X-100, 2 mM DTT, pH 7.2). Washing was continued right up until the time of loading. The resin was then mixed with the WGA-Sepharose eluate for 4 hrs at 4°C, that had been previously dialyzed to remove N-acetylglucosamine. The resin containing bound insulin receptors was equally divided into four 20 ml columns (Konti) and washed with 10 column volumes of wash buffer (50 mM Hepes, 0.5 M NaCI, 10% glycerol, 0.1% (w/v) Triton X-100, 1 mM DTT, pH 7.2). The columns were then closed to await elution. Just prior to elution each column was washed with 4 column volumes of dilution buffer (50 mM Hepes, 10% glycerol, 0.1% (w/v) Triton X-100, 1 mM DTT, pH 7.2) and drained to bed level. The column was brought to room temperature and 3 volumes of elution buffer (50 mM NaAcetate, 0.5 M MgCI2, 10% glycerol, 0.1% (w/v) Triton X-100, 1 mM DTT, pH 5.0, 23 C) was placed on top of the resin and the column opened. Three column volumes of effluent were collected over a period of no more than 2 min., and immediately diluted 10-fold in ice cold dilution buffer. The receptors were then concentrated either by pouring the dilute eluate through a small syringe containing 250 ul of DEAE-cellulose (Whatman DE-52), or by batch adsorption to 1 ml of the same ion exchanger, stirring for approximately 30 min at 4°C. This whole procedure was then repeated for the remaining 3 columns. After adsorption to DEAE-cellulose the receptors were eluted with 2 column volumes of dilution buffer supplemented with 250 mM NaCI. The eluates were further concentrated (approximately 5-fold) in a Centricon 30 microconcentrator and stored at -70°C. Assay of insulin 70 receptor/kinase activity involving SDS-PAGE and autoradiography were carried out as described in Materials and Methods. 5. Investigation of Possible Effects of the Triton-Solubilized Extract of Liver Membranes on the Phosphorylation of Acetvl-CoA Carboxylase In a Reconstituted System. The crude extract prepared by Triton-solubilization of rat liver membranes contains protein-serine kinase activity (fig. 5) that is insulin responsive towards the insulin receptor j5-subunit (Pang et al., 1985). Tavare et al. (1985) have demonstrated protein-serine kinase activity that will act on acetyl-CoA carboxylase when purified acetyl-CoA carboxylase is reconstituted with the Triton-solubilized extract of human placental membranes. This acetyl-CoA carboxylase kinase activity was enhanced when the reconstituted fraction was treated with insulin. The effect observed by these workers was extremely variable, however; over five experiments the increase in acetyl-CoA carboxylase phosphorylation ranged from 10-460%. Using identical assay conditions to those used by Tavare et al. (1985) I was unable to reproduce the effects that they reported. In experiments using purified rat liver acetyl-CoA carboxylase (obtained from Katherine Quayle in this lab), I observed no increase in the phosphorylation of acetyl-CoA carboxylase upon reconstitution with the Triton-solubilized membrane extract of rat liver, nor following insulin-treatment of the reconstituted fraction. However, using conditions which I have shown to be essential for the expression of protein-serine kinases in the high-speed supernatant fraction of adipose tissue (ie. absence of both M n 2 + and NaF as well as temperature considerations), I observed an increase in acetyl-CoA carboxylase phosphorylation of approximately six-fold when purified acetyl-CoA carboxylase was reconstituted with the Triton-solubilized membrane extract of rat liver (fig. 18, a,b,c). This effect was reproduced in a system using the Triton-solubilized membrane extract prepared from rat adipose tissue (done by Katherine Quayle, data not shown). The phosphorylation of acetyl-CoA carboxylase was not enhanced, 71 however, when the reconstituted fraction was treated with insulin (measured by scanning densitometry; cf. left and right lanes of c, fig. 18). When the Triton-solubilized membrane extract was chromatographed over WGA-Sepharose the acetyl-CoA carboxylase kinase activity was found to be present in the flow-through void fraction and was not found in the WGA-Sepharose eluate, judged by reconstitution of acetyl-CoA carboxylase with either the WGA-Sepharose flow-through void (fig. 18, a,d,e), or the WGA-Sepharose eluate fractions (fig. 18, a,f,g). 72 Fig. 18. Effects of Reconstitution of Purified Acetyl-CoA Carboxylase with Fractions Obtained from the Triton-Solublized Extract of Liver Membranes. Acetyl-CoA Carboxylase (ACC) (2 ug) purified from rat liver was reconstituted with either the Triton-solublized extract of rat liver membranes (TE) (32 ug protein), the flow-through void fraction following chromatography of the Triton extract over WGA-Sepharose (WGV) (29 ug protein), or the WGA-Sepharose eluted fraction (WGE) (10 ug protein). Samples were preincubated for 30 min at 23°C in buffer (pH 7.6) containing 50 mM Hepes, 0.05% (w/v) Triton X-100,12 mM MgCI2 and 2 mM DTT, in the absence or presence of insulin (10 ug/ml). The samples were then incubated with [jf-32P]ATP (200 uM, 800 dpm/pmol) (final assay vol = 20 ul) for 10 min at 37°C and stopped by the addition of SDS-sample buffer (20 ul) with heating at 95°C for 5 min. Proteins were separated by SDS-PAGE and [32P]-labelled proteins identifed by autoradiography, as described in Materials and Methods, a) ACC alone; b) TE alone; c) ACC + TE; d) WGV alone; e) ACC + WGV; f) WGE alone; g) ACC + WGE. Left and right lanes of each sample pair (b-g) represent treatment without and with insulin, respectively. Molecular weight markers are expressed in kilodaltons. 73 74 DISCUSSION 1. Preparation of Partially Purified Insulin Receptors. The most commonly employed whole tissue sources for insulin receptor purification are human placenta, which is rich in both insulin and EGF receptors, and rat liver. In preliminary studies I have prepared insulin receptor fractions from both tissues and have found little difference between the two in their preparation and quality. Subsequently, all preparations have been from rat liver which, for my purposes, has been the most convenient. For example, the livers are easily excised from the animal and cooled to ice temperature within minutes, minimizing proteolysis. In most of the studies described in this thesis, the insulin receptor has been used in the form of a partially purified fraction prepared from rat liver by chromatography of detergent-solubilized membranes over WGA-Sepharose. Under defined conditions this WGA-Sepharose-purified extract displays phosphorylation of very few endogenous proteins after incubation with radiolabeled ATP. Thus, a preparation relatively free of phosphoproteins can be obtained without the use of harsh conditions required for elution of the receptor from agarose-bound insulin, and avoids the use of insulin altogether, ensuring that the receptors are completely inactive until the time of use. The preparation of wheat germ extract was found to be straightforward and was easily adopted from published procedures. The use of Triton X-100 at 2% (w/v) to solubilize membranes is based on the original protocol of Cuatrucasas (1972a) who reported that higher concentrations of detergent did not appreciably increase the recovery of receptors from plasma membranes. I have also found that re-extraction of the membranes with fresh detergent does not seem to further increase the yield of receptors. In general, detergents other than Triton X-100 have not been extensively tested aside from j}-octylglucoside, which has been employed in studies involving the reconstitution of insulin receptors into liposome vesicles. The usefulness of |3-octylglucoside for this purpose is largely 75 based on the fact that it can be easily removed by dialysis attributible to its high critical micellar concentration (30 mM). Triton X-100, on the other hand, is dialyzable only to an extremely limited extent. This is a noteworthy point when considering assaying the insulin receptor. Triton concentrations of 0.5% (w/v) have been found to decrease insulin binding to approximately 50% that observed at a detergent concentration of 0.12% (w/v) (Cuatrecasus, 1972a). The critical micellar concentration of Trition X-100 is about 0.01% and concentrations in the range of 0.02% to 0.1% (w/v) do not appreciably affect receptor autophosphorylation (White et al., 1984). Recovery of bound glycoproteins from WGA-Sepharose is achieved by elution with buffer containing 300 mM N-acetylglucosamine. On occassion this concentration was observed to inhibit insulin receptor/kinase activity, possibly by interfering with insulin binding. I have also found this concentration to affect the relative mobility of proteins on SDS-PAGE. Therefore, the WGA-Sepharose-purified extract was always dialyzed to reduce the concentration of N-acetylglucosamine to less than 30 mM before use. The insulin receptor present in this extract is remarkably stable, and can be stored at -70°C for several months with minimal loss of tyrosine kinase activity. The use of synthetic peptides for receptor kinase assays has been characterized by Stadtmauer and Rosen (1983) and Casnellie et al. (1982). In my studies the receptor was assayed by using a synthetic peptide with the sequence Asp-Arg-Val-Tyr-lle-His-Pro-Phe which corresponds to the naturally occuring form of human angiotensin II. This is a particularly convenient assay as it allows the measurement of protein-tyrosine kinase activity even in crude extracts, since background phosphorylation of endogenous substrates can be removed by precipitation with trichloroacetic acid under which conditions angiotensin is soluble. The synthetic peptide Glu SO-Tyr20 was also used in some experiments since it has been shown to be a better substrate for the insulin receptor than angiotensin (Braun et al., 1984). However, it is precipitated under conditions which also precipitate most phosphoproteins in a crude extract. 76 2. Protein Kinase Activity Associated with the Insulin Receptor. The activities shown in Table I are initial reaction rates obtained using a concentration of peptide substrate equal to the reported Km value of the receptor for angiotensin, which is approximately 2.5 mM. These values are therefore approximately one-half of the theoretical maximal activity. The high Km associated with the phosphorylation of angiotensin and other peptide substrates appears to be a common feature among many tyrosine kinases (see Foulkes et al., 1985). The reason for this is unclear, but may refect the small size of the substrate, since under the same conditions the Km for proctolin (Arg-Tyr-Leu-Pro-Thr) is 25 mM, and that for the dipeptide Tyr-Arg Is greater than 80 mM (Stadtmauer and Rosen, 1983). Certainly, other factors such as amino acid specificity and physical conformation may play a role. It is interesting that if the receptor is prephosphorylated prior to the addition of peptide, the Km of the receptor for angiotensin is dramatically decreased to approximately 420 uM (Walker et al., 1987). In whole cells (Pang et al., 1985) and crude homogenates (fig 5) the insulin receptor is phosphorylated on both serine and tyrosine residues after incubation with insulin. It should be recognized, however, that the insulin receptor is a tyrosine-specific protein kinase and will not catalyze the phosphorylation of substrates on serine or threonine residues. The receptor is also unable to phosphorylate synthetic substrates on d-tyrosine (Walker et al., 1987). Thus, phosphoserine present in the receptor -^subunit is most probably the result of exogenous serine kinase activity, which decreases as the receptor is purified to near homogeneity. This insulin-responsive serine kinase activity associated with the insulin receptor is worthy of attention as it may possibly play a role in mediating or regulating insulin action. 3. Effect of Assay Components on Protein Kinase Activity. When the buffer components of the WGA-Sepharose-purified preparation of insulin receptors are mixed with the high-speed supernatant fraction of adipose tissue there is marked inhibition of 77 protein kinase activtity (fig. 8). This can be attributed largely to the presence of NaF and/or MnCI2, either of which when tested alone cause dramatic reduction of protein phosphorylation. NaF has been shown to be a protein phosphatase inhibitor (Ballou et al., 1986). However, this is by no means the only possible effect of NaF, which has also been shown to inhibit protein kinase activity towards the ribosomal protein S6 in cells stimulated with various growth factors (Novak-Hofer and Thomas, 1985; Pelech et al., 1986), including insulin (Cobb and Rosen, 1983; Tabarini et al., 1985). NaF has also been reported to inhibit the activity of casein kinases I and II (see Cobb and Rosen, 1983). The use of 2-glycerophosphate was also tested since its presence has been shown to be essential for the recovery of insulin-stimulated (Cobb and Rosen, 1983; Tabarini et al., 1985) and EGF-stimulated (Novak-Hofer and Thomas, 1985) S6 kinase activity from 3T3-L1 and Swiss 3T3 cells, respectively. However, when tested in the high-speed supernatant fraction of adipose tissue, 2-glycerophosphate specifically inhibited the insulin-dependent phosphorylation of ATP-citrate lyase. Protein kinases require metal ions (usually M g 2 + or Mn 2 + ) for activation usually to form the metal-nucleotide complex which is the actual substrate (Krebs, 1986). The presence of M n 2 + has been shown to activate insulin receptor autophosphorylation in vitro by decreasing the Km for ATP (White et al., 1984). However, this occurs at concentrations of M n 2 + greater than that needed to form MnATP, and therefore suggests that the metal ion may act by binding to a separate site on the receptor protein (White et al., 1984). For a given concentration of ATP, it has been shown that insulin receptor activity is several fold greater in the presence of M n 2 + than M g 2 + (Zick et al., 1983; Nemenoff et al., 1984; Uhing and Exton, 1986). This is especially true at low ATP concentrations (20 uM ATP or less) (White et al., 1984; Zick et al., 1983; Nemenoff et al., 1984). As ATP concentration is increased, however, the enhanced effect of M n 2 + over M g 2 + is diminished, and at a concentration of 2 mM ATP receptor activity can be supported equally with 10 mM MnCI2 or 10 mM MgCU (Nemenoff et al., 1984). Thus, under cellular conditions where ATP concentration 78 is in the mM range and Mg^ predominates, the insulin receptor can be fully activated even when p . cellular Mn concentrations are less than 1 uM. In the complete absence of M n 2 + but in the presence of 10 mM MgCI2,1 have found 200 uM ATP to be of sufficient concentration to measure receptor activation in response to insulin treatment. The addition of MnCI2 up to 0.5 mM enhances receptor activity, but beyond this point further addition of the metal becomes inhibitory. These results were observed with both the 95 kd subunit autophosphorylation and the phosphorylation of angiotensin. This parallel relationship between insulin receptor autophosphorylation and the phosphorylation of exogenous substrates has also been shown by Nemenoff et al.(1984), and is consistent with the theory that autophosphorylation is required to activate the insulin receptor/kinase towards exogenous substrates (Yu and Czech, 1984; Rosen etal., 1983). Although Mn enhances insulin receptor/kinase activity in vitro, I have shown micromolar concentrations to severely inhibit protein-serine kinase function in the high-speed supernatant fraction of adipose tissue incubated with 10 mM MgCI2. Tabarini et al. (1985) have found 0.05 mM MnCI2 to inhibit an insulin-stimulated S6 kinase activity in 3T3-L1 cells, and similarly Pelech et al. (1986) have found 2.5 mM MnCI2 to inhibit a fibroblast growth factor (FGF)-sensitive S6 kinase in Swiss 3T3 cells. Since the affinity of ATP for M n 2 + and M g 2 + is similar (see Uhing and Exton, 1986), one may assume that at 10 mM MgCI2 and 0.05 mM MnCI2 the predominant MeATP complex would be MgATP (as is the case under cellular conditions). Thus, M n 2 + probably inhibits protein-serine kinase activity via a mechanisim independent of competition between different forms of MeATP and may involve a distinct metal ion binding site, similar to that proposed for the insulin receptor (White et al., 1984). 79 4. Reconstitution of High-Speed Supernatant Fractions of Adipose Tissue with Partially Purified  Insulin Receptors. The major changes in phosphorylation observed in high-speed supernatant fractions of adipose tissue reconstituted with partially pure insulin receptors are of two proteins of subunit Mr 40 kd (pp40) and 25 kd (pp25). Phosphorylation of the 40 kd protein is insulin-dependent and appears to occur largely on tyrosine residues. It is therefore probable that pp40 is a substrate for the insulin receptor. It is not possible to determine the stoichiometry to which pp40 is phosphorylated. With many of the putative receptor substrates that have been recently described (Kadowaki et al., 1987; Haring et al., 1987) the stoichiometry of phosphorylation is equally uncertain, but in almost all op O p cases [P]-incorporation into substrate is far less than the total [P]-incorporation into the p-subunit of the insulin receptor, itself, giving indication of the low abundance of the substrate. This raises the question of signal amplification and to what extent it is acheived at this step of the signal pathway. It is possible that the true physiological substrates for growth factor receptors are extremely rare proteins and that these systems achieve signal amplification subsequent to this event. The significance of pp40 must be considered carefully. It is a protein that is phosphorylated only in vitro and has not previously been observed in whole tissue treated with insulin. Nonetheless, it is an insulin responsive event that occurs with cellular components, in a cell-free environment. It is most likely a protein endogenous to the adipose tissue high-speed supernatant fraction since it does not appear in the WGA-Sepharose-purified membrane extract when incubated alone. It may still, however, be a membrane component phosphorylated only in the presence of some factor in the adipose tissue high-speed supernatant. The disappearance of pp25 present in the adipose tissue high-speed supernatant occurs as a result of reconstitution, and its dephosphorylation does not appear to be insulin dependent pp 25 does not exist as a high molecular weight form as suggested by some experiments involving gel 80 filtration chromatography (not shown). pp40 is unlikely to represent a dimer of pp25 since the latter is phosphorylated on serine, as shown by alkali digestion. There is strong evidence to suggest that insulin-dependent serine kinases can be activated by reconstitution with insulin receptor (Tavare et al., 1985; Mailer et al., 1986; Stefanovic et al., 1986) or other tyrosine kinases (Blenis and Erikson, 1985). Most of this evidence has been obtained using the phosphorylation of S6 as a model system in frog oocytes following microinjection of insulin receptor (Mailer et al., 1986; Stefanovic et al., 1986) or expression of pp60 v" s r c in chick embryo fibrblasts by infection with Rous sarcoma virus (Blenis and Erikson, 1985). By analogy, I have examined the possibility that protein-serine phosphorylation in the high-speed supernatant fraction of adipose tissue may be altered by reconstitution with insulin receptor/tyrosine kinase present in the WGA-Sepharose-purified extract of liver membranes. Pelech et al. (1986) have used a synthetic peptide to assay for S6 kinase activity in extracts from cells treated with growth factor. This peptide is patterned after the amino acid sequence that represents the major site of S6 phosphorylation in response to insulin (Whettenhall and Morgan, 1984). I have used this peptide to try to determine whether there is insulin-responsive S6 kinase activity in the high-speed supernatant fraction of adipose tissue, and if so to what extent the activity can be reconstituted with detergent-solubilized insulin receptors. Using the same conditions described in the legend to fig. 7,1 have indeed found that the S6 peptide is phosphorylated in the presence of the high-speed supernatant fraction of rat adipose tissue . However, I have observed no difference in the phosphorylation of S6 peptide in supernatant fractions prepared from control tissue or tissue treated with insulin (results not shown). The S6 peptide is a substrate for cAMP-dependent protein kinase, however, inclusion of the cAMP-dependent protein kinase inhibitor peptide (Pelech et al., 1986) reduces S6 phosphorylation only minimally, suggesting that cAMP-dependent protein kinase activity is low in these preparations. It is possible that the S6 peptide is a substrate for protein kinases which in vivo do not recognize 40S ribosomes as substrates, and that 81 insulin-dependent S6 kinase activity in this case, is masked by the abundant activity of other protein-serine kinases. Thus, it would be important in further studies to investigate insulin responsiveness of S6 peptide phosphorylation after partial purification. The effect of reconstitution on the phosphorylation of other endogenous proteins, such as ATP-citrate lyase, is presently unclear due to a phosphoprotein in the WGA-Sepharose-purified receptor preparation which comigrates and masks ATP-citrate lyase on SDS-PAGE. Increases in the phosphorylation of acetyl-CoA carboxylase are also difficult to detect in this system. 5. Purification of the Insulin Receptor to Homogeneity. The purification of insulin receptors to homogeneity, in my hands, has proved to be a difficult task and so far is still not achievable with satifactory yield. The major difficulty resides in the poor binding characteristics of an insulin-agarose column prepared by published procedures, using Affigel-10 (Bio-Rad). Poor binding characteristics have been suggested to be the result of insufficient washing of the column prior to its use, which may be necessary to remove ligand leached from the column (Linda Pike, Don Tinker, personal commun.). In addition to the extensive washing of conventionally made insulin-agarose, I have tried to improve on this technique by constructing an insulin-agarose column using a novel method involving the coupling of "boc-insulin". As described in Results, this method prevents coupling of insulin through glycine at position A1 and lysine at position B29. Therefore, this column should theoretically have 50% greater binding capacity per mg of bound insulin, considering that coupling through residue B29 does not appreciably alter the affinity of insulin for its receptor. When tested, the binding characteristists of this column, although better, were still not satisfactory and must be improved upon. In addition, methods for elution of the receptor from insulin-agarose have generally been unsatisfactory. Elution using 6M urea, as originally proposed for the purification of insulin-binding 82 activity (Cuatrecasas, 1972b) did not yield an active tyrosine kinase (Petruzzelli et al., 1982). However, using milder elution conditions (1 M NaCI, pH 5.5) recovery of both insulin receptor/kinase activity and maximum insulin binding was possible (Kasuga et al., 1983a; Rosen et al., 1983). But even under these conditions 80% of the receptor activity is lost after 20 min. at 25°C (Hoffman and Finn, 1985). Rosen et al.(1983) have reported that receptor/kinase activity is stable for only 30 min. at pH 5.0-6.0 at 23°C. It must be assumed then, that the receptor cannot tolerate such acidic conditions for more than several minutes. For this reason the elution procedure as described has been designed to minimize the time of exposure of receptors to acid conditions. To achieve this, insulin-agarose containing bound receptors was divided into four small columns. Each was eluted separately and rapidly, the eluate then immediately diluted and neutralized. Dilution of the high salt eluate is necessary for subsequent concentration on DEAE-cellulose which inturn is necessary for storage of the enzyme, since receptors frozen in dilute solution rapidly lose activity (Linda Pike, personal commun.). Working with such small quantities of protein, as is the case here, presents an additional problem in following the receptor through this stage of purification. Considering the time required for assaying and the low concentration of receptors this was not possible. To appreciate this, one must consider the proportion of cellular protein that the insulin receptor represents which is about 0.0002% of the total cellular protein in a liver homogenate and about 0.004% of the membrane protein (Cuatrecasas, 1972b). Thus, a total of approximately 40 ug of receptor protein would be expected to be present in the crude homogenate obtained from 8 rat livers, which contains about 18 grams of protein. Assuming that the purified receptor can be recovered with a yield of 10-30% (Petruzzelli et al., 1984; Pike et al., 1986) the maximum amount of receptor protein expected after insulin-agarose chromatography would be in the order of 13 ug. This represents an approximate 500-700 fold purification of the insulin receptor from the Trition-solubilized membrane extract. 83 6. Investigation of Possible Effects of the Triton-Solubilized Extract of Liver Membranes on the Phosphorylation of Acetvl-CoA Carboxylase in a Reconstituted System. Tavare et al. (1985) incubated purified acetyl-CoA carboxylase with the Triton-soluble fraction of human placental membranes plus lY-32P]ATP at 0°C in the presence of 2 mM MnCI2 and 50 mM NaF. Under these conditions they observed the phosphorylation of acetyl-CoA carboxylase in response to added insulin, however the effect was extremely variable (10-450% increase over 5 experiments). In my own hands I was not able to reproduce their observations using a Triton-solubilized extract from the membranes of rat liver. This may be attributable to inherent variabilities in the preparation of the detergent-extracted membrane fractions and/or the difference in tissue sources. However, in addition, I have shown in my own studies that Mn and NaF, at the concentrations used by Tavare and coworkers, are potent inhibitors of protein-serine kinase activity in the high-speed supernatant fraction of adipose tissue, and that low reaction temperature also influences the degree of phosphorylation. It seems most probable that the conditions employed in the studies carried out by Tavare et al. were almost certainly far from optimal for consistent expression of protein-serine kinase activity. When both MnCI2 and NaF were omitted from the assay medium, a six-fold increase in acetyl-CoA carboxylase phosphorylation was observed when purified acetyl-CoA carboxylase was reconstituted with the Triton-soluble fraction of rat liver membranes and assayed at 37°C. This protein kinase activity was not retained by chromatography of the detergent-solubilized membrane extract over WGA-Sepharose and was recovered almost completely in the WGA-Sepharose flow-through void fraction (greater than 90%, estimated by scanning densitometry). The phosphorylation of acetyl-CoA carboxylase does not appear to be insulin-responsive, however this cannot be assessed with certainty since under these assay conditions (ie. 37°C) I have also observed the 95 kd insulin receptor ^-subunit to be nonresponsive to insulin treatment, as previously discussed (see Results). In further studies, it will be important to determine the site(s) at which phosphorylation occurs on acetyl-CoA carboxylase, including analysis of tryptic 84 phosphopeptides and comparison with the phosphopeptides observed when the enzyme is isolated from intact cells treated with insulin. 85 REFERENCES Aiyer, R.A., (1983) J. Biol. Chem. 258: 14992-14999. Avruch, J., Nemenoff, R.A., Blackshear, P.J., Pierce, M. and Osathanondh, F., (1982) J. Biol. Chem. 257: 15162-15166. Ballou, LM., Fisher, E.H., (1986) Phosphoprotein Phosphatases. In: The Enzymes (PD Boyerand EG Krebs, ed.) vol. XVII Pt.A: 312-365. Bell, J.C., Mahadevan, L.C., Colledge, W.H., Frackelton Jr., A.R., Sargent, M.G., Foulkes, J.G., (1987) Nature 325: 552-554. Belsham, G.J., Brownsey, F.W. and Denton, R.M., (1982) Biochem. J. 204: 345-352. Bernier, M., Laird, D.M. and Lane, M.D., (1987) Proc. Natl. Acad. Sci. 84:1844-1848. Blenis, J. and Erikson, R.L, (1985) Proc. Natl. Acad. Sci. 82: 7621-7625. Bollag, G.E., Roth, R.A.. Beaudoin, J., Mochly-Rosen, D., Koshland, Jr., D.E., (1986) Proc. Natl. Acad. Sci. 38: 5822-5824. Bradford, M.M. (1976) Anal. Bioch. 72: 248-254. Braun, S., Raymond, W.E. and Racke, R.E., (1984) J. Biol. Chem. 259: 2051-2054. Brownsey, R.W., Hughes, W.A. and Denton, R.M., Mayer, R.J., (1977) Biochem. J. 168: 441-445. Brownsey, R.W., Hughes, W.A. and Denton, R.M., (1979) Biochem. J. 184: 23-32. Brownsey, R.W. and Hardie, D.G., (1980) FEBS Lett. 120: 67-70. Brownsey, R.W. and Denton, R.M., (1982) Biochem. J. 202 77-86. Brownsey, R.W., Edgell, N.J., Hopkirk, T.J. and Denton, R.M., (1984) Biochem. J. 218 733-743. Brownsey, R.W. and Denton, R.M., (1986) Acetyl Coenzyme A Carboxylase. In: The Enzymes (PD Boyer and EG Krebs, ed.) vol. XVII Pt.B 123-146. Brownsey, R.W., Quayle, K.A. and Mui, A.L-F., (1988) submitted for publication. Casnellie, J.E., Harrison, M.L, Pike, L.J., Hellstrom, K.E. and Krebs, E.G., (1982) Proc. Natl. Acad. Sci. 79 282-286. Carpenter, G., (1987) Ann. Rev. Biochem. 56 881-914. Chou, C.K., Dull, T.J., Russell, D.S., Gherzi, R., Lebwohl, D., Ullrich, A. and Rosen, O.M., (1987) J. Biol. Chem. 262 1842-1847. Cobb, M.C. and Rosen, O.M., (1983) J. Biol. Chem. 12472-12481. Cuatrecasas, P. (1972a) Proc. Natl. Acad. Sci. 69 318-322. 86 Cuatrecasas, P. (1972b) Proc. Natl. Acad. Sci. 691277-1281. Denton, R.M., Brownsey, R.W., and Belsham, G.J., (1981) Diabetologia 21 347-362. Denton, R.M., Brownsey, R.W., Hopkirk, T.J. and Belsham, G.J., (1984) Biochem. Soc. Trans. 12 768-771. Doolittle, R.F., Hunkapiller, M.W., Hood, L.E., Devare, S.G., Robbins, K.C., Aaronson, S.A. and Antoniades, H.N., (1983) Science 221 275-277. Downward, J., Yasden, Y., Mayes, E., Scrace, G., Totty, N., Stockwell, P., Ullrich, A., Schlessinger, J. and Waterfield, M.W., (1984) Nature 307 521-527. Ebina, Y., Ellis, L, Jarnagin, K., Edery, M., Graf, L, Clauser, E., Ou, J., Masiarz, F., Kan, Y.W., Goldfine, I.D., Roth, R.A. and Rutter, W.J., (1985) Cell 40 747-758. Ebina, Y., Araki, E., Taira, M., Shimada, F., Mori, M., Craik, C.S., Siddle, K., Pierce, S.B., Roth, R.A. and Rutter, W.J., (1987) Proc. Natl. Acad. Sci. 84 704-708. Eckhart, W., Hutchinson, M.A. and Hunter, T., (1979) Cell 18 925-933. Ek, B., and Heldin, C.H., (1982) J. Biol. Chem. 257 10486-10492). Ellis, L, Clauser, E., Morgan, D.O., Edery, M., Roth, R.A. and Rutter, W.J., (1986) Cell 45 721-732. Ellis, L, Morgan, D.O., Jong, S.-M., Wang, L-H., Roth, R.A. and Rutter, W.J., (1987) Proc. Natl. Acad. Sci. 84 5101-5105. Flocknart, D.A., and Corbin, J.D., (1982) CRC Crit Rev Biochem 13 133-186. Forsayeth, J.R., Caro, J.F., Sinha, M.K., Maddux, B.A. and Goldfine, I.D., (1987) Proc. Natl. Acad. Sci. 84 3448-3451. Foulkes, J.G., Chow M., Gorka, C , Frackelton, A.R., and Baltimore, D., (1985) J. Biol. Chem. 260 8070-8077. Frost, S.C., Kohanski, R.A. and Lane, M.D., (1987) J. Biol. Chem. 262 9872-9876. Fujrta-Yamaguchi, Y., Choi, S., Sakamoto, Y. and Itakura, K. (1983) J. Biol. Chem. 258 5045-5049. Gammeltoft, S. and van Obberghen, E., (1986) Biochem. J. 235 1-11. Geiger, R., (1971) diese Z. 352 7. Geiger, R., Schone, H.-H. and Pfaff, W., (1971) Hoppe-Seyler's Z. Physiol. Chem. 352 1487-1490. Gherzi, R., Russell, D.S., Taylor, S.I. and Rosen, O.M., (1987) J. Biol. Chem. 262 16900-16905. Glass, D.B., Masaracchia, R.A., Feramisco, J.R., Kemp. B.E., (1978) Analytical Biochem. 87 566-575. Hari, J. and Roth, R.A., (1987) J. Biol. Chem. 262 15341-15344. 87 Harrison, L.C., Billington, T., East, I.J., Nichols, R.J. and Clark, S., (1978) Endocrinol. 102 1485-1494. Haring, A., Kirsch, D., Obermaier, B., Ermel, B. and Machicao, F., (1986a) Biochem. J. 234 59-66. Haring, A., Kirsch, D., Obermaier, B., Ermel, B. and Machicao, F., (1986b) J. Biol. Chem. 261 3869-3875. Haring, H.U., White, M.F., Machicao, F., Ermel, B., Schleicher, E. and Obermaier, B., (1987) Proc. Natl. Acad. Sci. 84 113-117. Hedo, J.A., Harrison, LC. and Roth, J., (1981) Biochemistry 20 3385-3393. Hedo, J.A. and Simpson, I.A., (1984) J. Biol. Chem. 259 11083-11089. Hedo, J.A. and Gorden, P., (1985) Horm. Metabol. Res. 17 487-490. Hedo, J.A. and Kahn, C.R., (1985) Meth. Enz. 109 593-609. Herzberg, V.L, Grigorescu, F., Edge, A.S.B., Spiro, R.G. and Kahn, C.R., (1985) BBRC 129 789-796. Hofmann, K. and Finn, F.M., (1985) Annal. N.Y. Acad. Sci. 447 359-372. Hofmann, K., Finn, F.M., Friesen, H.-J., Diaconescu, C. and Zahn, H., (1977) Proc. Natl. Acad. Sci. 74 2697-2700. Izumi, T., White, M.F., Kadowaki, T., Takaku, F., Akanuma, Y., and Kasuga, M., (1987) J. Biol. Chem. 262 1282-1287. Jacobs, S. and Cuatrecasas, P., (1981) Endocrine Rev. 2 251-263. Jacobs, S. and Cuatrecasas, P., (1986) J. Biol. Chem. 261 934-939. Johansen, J.W. and Ingebritsen, T.S., (1986) Proc. Natl. Acad. Sci. 83 207-211. Kadowaki, T., Koyasu, S., Nishida, E., Tobe, K., Izume, T., Takaku, F., Sakai, H., Yahara, I. and Kasuga, M., (1987) J. Biol. Chem. 262 7342-7350. Kahn, C.R., (1981) Recent Prog. Horm. Res. 37, 477-438. Kasuga, M., Zick, Y., Blithe, D.L, Crettaz, M. and Kahn, C.R., (1982a) Nature 298 667-669. Kasuga, M., Karlsson, F.A., and Kahn C.R., (1982b) Science 215 185-186. Kasuga, M., Zick, Y., Blith, D.L, Karlsson, F.A., Haring H., and Kahn C.R., (1982c) J. Biol. Chem. 257 9891-9894. Kasuga, M., Hedo, J.A., Yamada, K.M. and Kahn, C.R., (1982d) J. Biol. Chem. 257 10392-10399. Kasuga, M., Fujita-Yamaguchi, Y., Blithe, D.L and Kahn, C.R., (1983a) Proc. Natl. Acad. Sci. 80 2137-2141. 88 Kasuga, M., Fujita-Yamaguchi, Y., Blithe D.L, White, M.F., Kahn C.R., (1983b) J. Biol. Chem. 258 10973-10980. King, Jr. L.E., Carpenter, G. and Cohen, S. (1980) Biochemistry 19,1524-1528. Krebs, E.G., (1986) The Enzymology of Control by Phosphorylation. In:: The Enzymes (PD Boyer and E G Krebs, ed.) vol. XVII Pt.A 3-20 Kohanski, R.A. and Lane, M.D., (1985) J. Biol. Chem. 260 5014-5025. Kong, S.K. and Wang, J.H., (1987) J. Biol. Chem. 262 2597-2603. Laemmli, U.K., (1970) Nature 227 680. Lastick, S.M. and McConkey, E.H., (1981) J. Biol. Chem. 256 583-585. Mailer, J.L, Pike, L.J., Freidenberg, G.R., Cordera, R., Stith, B.J., Olefsky, J.M. and Krebs, E.G., (1986) Nature 320 459-461. Morgan, D.O., Ho, L, Korn, LJ. and Roth, R.A., (1986) Proc. Natl. Acad. Sci. 83 328-332. Morgan, D.O. and Roth, R.A., (1987) Proc. Natl. Acad. Sci. 84 41-45. Nemenoff, R.A., Kwok, Y.C., Shulman, G.I., Blackshear, P.J., Osathanondh, R. and Avruch, J., (1984) J. Biol. Chem. 259 5058- 5065. Novak-Hofer, E. and Thomas, G., (1984) J. Biol. Chem. 259 5995-6000. Novak-Hofer, E. and Thomas, G., (1985) J. Biol. Chem. 260 10314-10319. Pang DT and Shafer JA (1983) J. Biol. Chem. 258 2514-2518. Pang, D.T., Sharma, B.R. and Shafer, J.A.; White, M.F. and Kahn, C.R., (1985) J. Biol. Chem. 260 7131-7136. Pierce, M.W., Palmer, J.L, Keutmann, H.T., Hall, and Avruch J., (1982) J. Biol. Chem. 257 10681-10686. Pelech, S.L, Oiwin, B.B. and Krebs, E.G., (1986) Proc. Natl. Acad. Sci. 83 5968-5972. Petruzzelli, L.M., Ganguly, S., Smith, C.J., Cobb, M.H., Rubin, C. and Rosen, O.M., (1982) Proc. Natl. Acad. Sci. 79 6792-6796. Petruzzelli, LM., Herrera R and Rosen OM (1984) Proc. Natl. Acad. Sci. 81 3327-3331. Pike, L.J., Eakes, A.T., and Krebs, E.G. (1986) J. Biol. Chem. 261 3782-3789. Pike, L.J., Gallis, B., Casnellie, J.E., Bomstein, P. Krebs, E.G., (1982) Proc. Natl. Acad. Sci. 79 1443-1447. Pike, LJ., Kuenzel, E.A., Casnellie, J.E., and Krebs.E.G. (1984) J. Biol. Chem. 259 9913-9921. Pullen, R.A., Lindsay, D.G., Wood, S.P., Tickle, I.J., Blundell, T.L., Wollmer, A., Krail, G., Brandenburg, D., Zahn, H., Gliemann, J. and Gammeltoft, S., (1976) Nature 259 369-373. 89 Ramakrishna, S. and Benjamin, W.B., (1985) J. Biol. Chem. 260 12280-12286. Rees-Jones, R.W. and Taylor, S.I., (1985) J. Biol. Chem 260 4461-4467. Richert, B.D., Blithe, D.L and Pastan, I., (1982) J. Biol. Chem. 257 7143-7150. Rosen, O.M., Herrera, R., Olowe, Y., Petruzzelli, LM, and Cobb, M.H., (1983) Proc. Natl. Acad. Sci. 80 3237-3240. Roth, F.R. and Cassell, D.J., (1983) Science 219 299-301. Russell, D.S., Gherzi, R., Johnson, E,L, Chou, C.-K. and Rosen, O.M., (1987) J. Biol. Chem. 262 11833-11840. Sarin, V.K., Kent S.B.H., Tarn J.P. and Merifield R.B., (1981) Anal. Biochem. 117 147-157. Scott, J.D., Fischer, E.H., Takio, K., Demaille, J.G. and Krebs, E.G., (1985) Proc. Natl. Acad. Sci. 82 5732-5736. Stadtmauer, L A. and Rosen, O. M , (1983) J. Biol. Chem. 258 6682-6685 Stadtmauer, L A. and Rosen, O. M , (1986) J. Biol. Chem. 261 3402-3407. Stefanovic, D., Erikson, E., Pike, LJ. and Mailer, J.L, (1986) EMBO J 5 157-160. Stralfors, P., (1987) J. Biol. Chem. 262 11486-11489. Swergold, G.D., Rosen, O.M. and Rubin, C.S., (1982) J. Biol. Chem. 257 4207-4215. Syska, H., Perry, S.V. and Trayer, I.P. (1974) FEBS Letts. 40 253-257. Tabarini, D., Heinrich, J. and Rosen, O.M., (1985) Proc. Natl. Acad. Sci. 82 4369-4373. Takayama, S., White, M.F., Lauris, B. and Kahn, R.C, (1984) Proc. Natl. Acad. Sci. 81 7797-7801. Tanti, J.-F., Gremeaux, T., Rochet, G., Van Obberghen, E. and Le Marchand-Brustel, Y., (1987) Biochem. J. 245 19-26. Tavare, J.M., Smyth, J,E,. Borthwick, A.C., Brownsey, R.W. and Denton, R.M. (1985) Biochem. Soc. Trans. 13 734-735. Trevillyan, J.M., Perisic, O., Traugh, J.A. and Byus, C.V., (1985) J. Biol. Chem. 260 3041-3044. Uhing, R.J. and Exton, J.H., (1986) Mol. Cell. Endocrinol. 47 137-143. Ullrich, A., Bell, J.R., Chen, E.Y., Herrera, R., Petruzzelli, L.M., Dull, T.J., Gray, A., Coussens, L, Liao, Y.-C, Tsubokawa, M., Mason, A., Seebury, P.H., Grunfeld, C , Rosen, O.M. and Ramachandran, J., (1985) Nature 313 756-761. Ullrich, A., Gray, A., Tarn, A.W., Yang-Feng, T., Tsubokawa, M., Collins, C , Henzel, W., Le Bon, T., Kathuria, S., Chen, E., Jacobs, S., Francke, U., Ramachandran, J. and Fujita-Yamaguchi, Y., (1986) EMBO J. 5 2503-2512. 90 Van Obberghen, E., Kasuga, M., LeCam, A., Hedo, J.A., Itin, A. and Harrison, L C , (1981) Proc. Natl. Acad. Sci. 78 1052-1056. Van Patten, S.M., Heisermann. G.J., Cheng H-J and Walsh DA (1987) J. Biol. Chem. 262 3398-3403. Walker, D.H., Kuppuswamy, D., Visvanathan, A., and Pike, L.J., (1987) Biochemistry 26 1428-1433. Wettenhall, R.E.H. and Morgan, F.J., (1984) J. Bid. Chem. 259 2084-2091. White, M.F., Haring, G., Kasuga, M., and Kahn, C.R., (1984) J. Biol. Chem. 259 255-264. White, M.F., Maron, R., Kahn, C.R., (1985) Nature 318 183-186. White, M.F., Takayama, S., Kahn, C.R., (1985) J. Biol. Chem. 260 9470-9478. White, M.F., Stegmann, E.W., Dull, T.J., Ullrich, A. and Kahn, C.R., (1987) J. Biol. Chem. 262 9769-9777. Witters, LA., Tipper, J.P. and Bacon, G.W., (1983) J. Biol. Chem. 258 5643. Yip, C.C, Yeung, CW. and Moule, M.L, (1980) Biochemistry 19 70-76. Yu, K.-T., Werth, D.K., Pastan, I.H. and Czech, M.P., (1985) J. Biol. Chem. 260 5838-5846. Yu, K.-T. and Czech, M.P., (1984) J. Biol. Chem. 259 527-5286. Yu, K.-T., Peters, M.A. and Czech, M.P., (1986) J. Biol. Chem. 261 11341-11349. Yu, K.-T., Khalaf, N. and Czech, M.P., (1987) Proc. Natl. Acad. Sci. 84 3972-3976. Zick, Y., Kasuga, M., Kahn, C.R. and Roth, J., (1983) J. Biol. Chem. 258 75-80 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097641/manifest

Comment

Related Items