novel serine phosphorylation sites of irs2 mediate 14-3-3 ... · cora weigert. \identi cation of...

$: Novel serine phosphorylation sites of IRS2 mediate 14-3-3 ... · Cora Weigert. \Identi cation of the amino acids 300-600 of IRS2 as 14-3-3 binding region with the importance of insulin/IGF-1-regulated$
Novel serine phosphorylation sites of

IRS2 mediate 14-3-3 binding and

regulate insulin signal transduction

Dissertation zur Erlangung des

Doktorgrades der Naturwissenschaften (Dr. rer. nat.)

Fakultat Naturwissenschaften

Universitat Hohenheim

Institut fur Biologische Chemie und Ernahrungswissenschaft

und

Universitatsklinikum Tubingen

Abteilung fur Innere Medizin IV

Pathobiochemie und Klinische Chemie

vorgelegt von

Sabine Sarah Neukamm

aus Filderstadt

2012

Dekan: Prof. Dr. rer. nat. Heinz Breer

1. berichtende Person: Prof. Dr. rer. nat. Erwin D. Schleicher

2. berichtende Person: Prof. Dr. rer. nat. Lutz Graeve

Eingericht am: 15.08.2012

Mundliche Prufung am: 15.11.2012

Die vorliegende Arbeit wurde am 24.10.2012 von der Fakultat Naturwissen-schaften der Universitat Hohenheim als “Dissertation zur Erlangung des Dok-torgrades der Naturwissenschaften” angenommen.

i

Contents

List of publications vi

Summary vii

Zusammenfassung ix

1 Introduction 11.1 Diabetes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Insulin signal transduction . . . . . . . . . . . . . . . . . . . . . . 3

1.2.1 Insulin and insulin like growth factor (IGF)-1 . . . . . . . 31.2.2 Insulin receptor and IGF-1 receptor . . . . . . . . . . . . 31.2.3 General characteristics of IRS proteins . . . . . . . . . . . 41.2.4 Physiological importance of IRS1 and IRS2 . . . . . . . . 51.2.5 IRS1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61.2.6 IRS2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.2.7 PI 3-kinase pathway . . . . . . . . . . . . . . . . . . . . . 101.2.8 Ras/MAPK pathway . . . . . . . . . . . . . . . . . . . . . 10

1.3 14-3-3 proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.3.1 Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . 121.3.2 Binding to phosphorylated serine/threonine residues . . . 131.3.3 Regulation of 14-3-3 proteins . . . . . . . . . . . . . . . . 141.3.4 Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

1.4 Aims of the thesis . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2 Materials and Methods 182.1 Chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.2 Consumables/Kits . . . . . . . . . . . . . . . . . . . . . . . . . . 232.3 Laboratory devices . . . . . . . . . . . . . . . . . . . . . . . . . . 252.4 Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272.5 Antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.5.1 Primary antibodies . . . . . . . . . . . . . . . . . . . . . . 272.5.2 Secondary antibodies . . . . . . . . . . . . . . . . . . . . . 282.5.3 Generation of phospho-specific antibody against p-Ser573 29

2.6 Tissue culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292.6.1 Cell lines used in this work . . . . . . . . . . . . . . . . . 292.6.2 Media and conditions of cultivation . . . . . . . . . . . . . 292.6.3 Cell splitting and seeding . . . . . . . . . . . . . . . . . . 302.6.4 Transient transfection of HEK293 cells . . . . . . . . . . . 30

ii

CONTENTS iii

2.6.5 Stable transfection of Flp-In HEK293 cells . . . . . . . . . 302.7 Protein Biochemistry . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.7.1 Cell lysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 322.7.2 Protein determination and sample preparation for SDS-

PAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322.7.3 SDS-PAGE . . . . . . . . . . . . . . . . . . . . . . . . . . 342.7.4 Western blotting . . . . . . . . . . . . . . . . . . . . . . . 352.7.5 Immunodetection . . . . . . . . . . . . . . . . . . . . . . . 362.7.6 Overlay assay . . . . . . . . . . . . . . . . . . . . . . . . . 372.7.7 Stripping . . . . . . . . . . . . . . . . . . . . . . . . . . . 372.7.8 Immunoprecipitation . . . . . . . . . . . . . . . . . . . . . 372.7.9 GFP pulldown . . . . . . . . . . . . . . . . . . . . . . . . 382.7.10 GST pulldown . . . . . . . . . . . . . . . . . . . . . . . . 38

2.8 Molecular Biology . . . . . . . . . . . . . . . . . . . . . . . . . . 392.8.1 Plasmid generation at the University of Dundee . . . . . . 392.8.2 Additional plasmids used in this thesis . . . . . . . . . . . 402.8.3 PCR mutagenesis . . . . . . . . . . . . . . . . . . . . . . . 412.8.4 Transformation . . . . . . . . . . . . . . . . . . . . . . . . 422.8.5 MiniPrep/MaxiPrep for isolation of DNA . . . . . . . . . 422.8.6 Sequencing . . . . . . . . . . . . . . . . . . . . . . . . . . 432.8.7 Restriction Digest . . . . . . . . . . . . . . . . . . . . . . 442.8.8 Propagation of E. coli . . . . . . . . . . . . . . . . . . . . 452.8.9 Propagation of GST-14-3-3 expression vectors . . . . . . . 45

2.9 Mass spectrometry (Dundee) . . . . . . . . . . . . . . . . . . . . 472.9.1 Sample preparation for mass spectrometry . . . . . . . . 472.9.2 Mass spectrometric analysis of IRS2 tryptic digests . . . . 47

2.10 Mass spectrometry (Tubingen) . . . . . . . . . . . . . . . . . . . 482.10.1 Sample preparation for mass spectrometry . . . . . . . . . 482.10.2 Mass spectrometric analysis of IRS2 tryptic digests . . . . 48

2.11 Animal studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492.11.1 Isolation of mice liver protein . . . . . . . . . . . . . . . . 492.11.2 Isolation of primary hepatocytes . . . . . . . . . . . . . . 49

2.12 Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 50

3 Results 513.1 Identification of signalling pathways that increased 14-3-3 binding

to IRS2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513.2 IGF-1-mediated 14-3-3 interaction with IRS2 . . . . . . . . . . . 52

3.2.1 Interaction of 14-3-3 with IRS2 upon IGF-1 or insulinstimulation in different cell systems . . . . . . . . . . . . . 52

3.2.2 Co-immunoprecipitation of endogenous 14-3-3 . . . . . . . 533.2.3 14-3-3 interacts with IRS2 from mouse liver . . . . . . . . 543.2.4 Mass spectrometry for identification of phosphorylated

serine/threonine residues on IRS2 after IGF-1 stimulation 553.2.5 Generation of IRS2 fragments . . . . . . . . . . . . . . . . 583.2.6 The area between amino acids 301 and 600 on IRS2 is

important for 14-3-3 binding . . . . . . . . . . . . . . . . 593.2.7 Identification of Ser573 as IGF-1-dependent 14-3-3 binding

site on IRS2 . . . . . . . . . . . . . . . . . . . . . . . . . . 603.2.8 Characterization of p-Ser573 antibody . . . . . . . . . . . 61

CONTENTS iv

3.2.9 Time course of Ser573 phosphorylation in Fao cells uponinsulin stimulation . . . . . . . . . . . . . . . . . . . . . . 65

3.2.10 Inhibition of Akt/PKB leads to a reduced phosphory-lation of Ser573 and reduced binding of 14-3-3 . . . . . . . 66

3.2.11 Positive influence of S573A mutant on Akt/PKB phos-phorylation . . . . . . . . . . . . . . . . . . . . . . . . . . 67

3.2.12 Ser573 is not the only IGF-1-dependent 14-3-3 binding site 673.3 cAMP-mediated 14-3-3 interaction with IRS2 . . . . . . . . . . . 70

3.3.1 The PKA substrate antibody as a convenient tool for theidentification of PKA phosphorylation sites on IRS2 . . . 71

3.3.2 PKA-mediated phosphorylation of IRS2 is not influencedby Akt/PKB or kinases of the MAPK pathway . . . . . . 73

3.3.3 Investigation of IRS2 and its fragments by overlay assayand PKA substrate antibody leads to opposing results . . 75

3.3.4 GST pulldown experiments to further elucidate 14-3-3binding to IRS2 . . . . . . . . . . . . . . . . . . . . . . . . 76

3.3.5 Generation of fragment 601-952 to further narrow downthe PKA-dependent 14-3-3 binding region . . . . . . . . . 78

3.3.6 Mass Spectrometry for the identification of potential PKAphosphorylation and 14-3-3 binding sites . . . . . . . . . . 79

3.3.7 Mutation of Ser1137 and Ser1138 on IRS2 reduces interac-tion with 14-3-3 . . . . . . . . . . . . . . . . . . . . . . . 83

3.3.8 Forskolin stimulation prevents protein degradation of IRS2wild type but not of IRS2-S1137A/S1138A . . . . . . . . 84

4 Discussion 874.1 Application of truncated proteins . . . . . . . . . . . . . . . . . . 884.2 Insulin and IGF-1-dependent interaction between 14-3-3 and IRS2 88

4.2.1 Identification of phosphorylated serine/threonine residueson IRS2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

4.2.2 Regulation and function of insulin/IGF-1-dependent 14-3-3/IRS2 interaction . . . . . . . . . . . . . . . . . . . . . 90

4.2.3 Potential relevance of insulin/IGF-1-dependent phosphory-lation of Ser573 and interaction with 14-3-3 proteins . . . 93

4.3 cAMP-dependent binding of 14-3-3 to IRS2 . . . . . . . . . . . . 944.3.1 cAMP and PKA . . . . . . . . . . . . . . . . . . . . . . . 944.3.2 Regulation and function of cAMP-dependent 14-3-3/IRS2

interaction . . . . . . . . . . . . . . . . . . . . . . . . . . 95

5 Conclusion 99

Bibliography 101

List of Figures 121

List of Tables 123

Abbreviations 124

CONTENTS v

A Appendix 126A.1 Phosphopeptides detected by mass spectrometry . . . . . . . . . 126A.2 Sequence alignment mouse and human IRS1/2 . . . . . . . . . . 129

Erklarung 133

Curriculum vitae 134

Acknowledgment 135

List of publications

Peer reviewed journal papers

Louise Fritsche, Sabine S. Neukamm, Rainer Lehman, Elisabeth Kremmer, AnitaM. Hennige, Andrea Hunder-Gugel, Martin Schenk, Hans-Ulrich Haering, Er-win D. Schleicher, and Cora Weigert. “Insulin-induced serine phosphorylationof IRS2 via ERK1/2 and mTOR: studies on the function of Ser675 and Ser907.”Am. J. Physiol. Endocrinol. Metab. 300 (2011), pp. E824-E836

Data concerning insulin/IGF-1-dependent phosphorylation of IRS2 and 14-3-3binding in this thesis are published in the following paper:

Sabine S. Neukamm, Rachel Toth, Nick Morrice, David G. Campbell, CarolMacKintosh, Rainer Lehmann, Hans-Ulrich Haering, Erwin D. Schleicher, andCora Weigert. “Identification of the amino acids 300-600 of IRS2 as 14-3-3binding region with the importance of insulin/IGF-1-regulated phosphorylationof Ser-573.” PLoS ONE 7(8) (2012)

Poster presentations

“The role of serine phosphorylation of IRS2 in insulin signalling”DFG Joint Meeting of the GRK880 & GRK1302 in Maurach, Germany, 2009

“Identification of novel phosphorylation sites on IRS2 and interaction with 14-3-3 proteins”XI International Symposium on Insulin Receptors and Insulin Action, Naples,Italy, 2010

“Identifizierung von 14-3-3 Bindungsstellen im Insulinrezeptorsubstrat 2”46. Jahrestagung der Deutschen Diabetes Gesellschaft, Leipzig, Germany, 2011

“Identification of a 14-3-3 binding site on insulin receptor substrate (IRS) 2”ADA 2011, San Diego, USA, 2011

“Novel serine phosphorylation sites of IRS2 mediate 14-3-3 binding and regulateinsulin signal transduction”DZD Workshop, Tubingen, Germany, 2012

vi

Summary

Insulin and insulin like growth factor (IGF)-1 mediate their metabolic and mi-togenic effects on target tissues through activation of the insulin and IGF-1receptor. Insulin receptor substrate (IRS) proteins function as intermediatedocking platforms to transduce the insulin/IGF-1 signal to intracellular effectormolecules that regulate glucose homoeostasis, lipid metabolism, cell prolifera-tion and β-cell survival. The activated receptors function as tyrosine kinasesthat phosphorylate IRS proteins on tyrosine residues, thereby generating in-teraction motifs for Src homology (SH) 2 domain containing proteins. Signaltransduction via IRS proteins is furthermore regulated by their serine/threoninephosphorylation by several kinases and this can result in inhibitory or stimu-latory consequences for downstream signalling. Hyperphosphorylation of IRS1has been shown to be involved in the desensitization process that can resultin insulin resistance. Both IRS1 and IRS2 undergo proteasomal degradationwhile particularly IRS2 levels are additionally regulated by cAMP-dependentgene activation.

14-3-3 proteins are versatile regulators of a variety of intracellular processeslike control of cell cycle, cell growth, gene transcription and apoptosis. Ser-ine/threonine phosphorylation within distinct motifs on the interaction partneris a prerequisite for 14-3-3 binding. IRS1 and IRS2 have been described as14-3-3 interaction proteins and interaction of IRS2 with 14-3-3 proteins wasspecifically characterized in this thesis.

Insulin/IGF-1-dependent PI 3-kinase stimulation as well as elevated cAMPlevels were identified to modulate 14-3-3 binding to IRS2. IGF-1 stimulation ledto increased binding of 14-3-3 to IRS2 in transfected HEK293 cells and this bind-ing was prevented by inhibition of the PI 3-kinase pathway and an Akt/PKBinhibitor. Insulin-stimulated interaction between endogenous IRS2 and 14-3-3was observed in rat hepatoma cells and in mice liver after an acute insulin stimu-lus or refeeding. Application of different IRS2 fragments enabled localization ofthe IGF-1-dependent 14-3-3 binding region spanning amino acids 300-600. Massspectrometric analysis produced a total of 24 phosphorylated serine/threonineresidues on IRS2 after IGF-1 stimulation with 12 sites unique for IRS2 whilethe other residues are conserved in IRS1 and IRS2. The 24 identified phos-phorylated residues on IRS2 included several 14-3-3 binding candidates in theregion 300-600 and single alanine mutants of these candidates led to the identi-fication of Ser573 as 14-3-3 binding site by overlay assays. A phosphosite specificantibody was generated to further characterize Ser573. IGF-1-dependent phos-phorylation of Ser573 was reduced by inhibition of PI 3-kinase and Akt/PKB.The alanine mutant of Ser573 showed enhanced phosphorylation of Akt/PKB in

vii

viii

an IGF-1 time course experiment. In summary, the data presented in this thesisindicate a negative impact of Ser573 phosphorylation on downstream signalling.

Binding of 14-3-3 to IRS2 upon stimulation with forskolin and the cAMPanalogue CPT-cAMP (8-(4-chlorophenylthio) adenosine 3’,5’-cyclic monophos-phate) was demonstrated in HEK293 cells, that was prevented with the PKAinhibitor H89. The amino acid region behind position 952 on IRS2 was identifiedas cAMP/PKA-dependent 14-3-3 binding region by GST-14-3-3 pulldown as-says. Mass spectrometric analyses revealed Ser1137/Ser1138 as cAMP-dependent,potential PKA phosphorylation sites. Inhibition of Akt/PKB or ERK didnot prevent the cAMP-dependent phosphorylation of IRS2 on PKA consensusmotifs. Mutation of Ser1137/Ser1138 to alanine strongly reduced the cAMP-dependent 14-3-3 binding as shown by GST-14-3-3 pulldown experiments andco-immunoprecipitation assays. IRS2 protein degradation was demonstrated bythe application of cycloheximide and an increased IRS2 protein stability wasobserved when HEK293 cells stably expressing IRS2 or primary hepatocyteswere incubated with forskolin. This reduced IRS2 protein degradation was de-pendent on the presence of Ser1137/Ser1138, since stimulation with forskolindid not increase protein stability of the double Ala1137/Ala1138 mutant. Toconclude, Ser1137/Ser1138 are presented as novel cAMP-dependent phosphory-lation sites on IRS2 and their importance in 14-3-3 binding and IRS2 proteinstability is demonstrated. This represents a novel mechanism for the cAMP-dependent upregulation of IRS2 protein levels that can be of importance forhepatic metabolism and β-cell survival.

Zusammenfassung

Insulin und Insulin like growth factor (IGF)-1 vermitteln ihre metabolischenund mitogenen Effekte in Zielgeweben durch die Aktivierung des Insulin- unddes IGF-1-Rezeptors. Die Insulin Rezeptor Substrat (IRS) Proteine fungierenals intermediare Bindungsstationen, uber die das Insulin/IGF-1 Signal an in-trazellulare Effektormolekule, welche die Glukosehomoostase, den Fettstoffwech-sel, das Zellwachstum und das Uberleben der β-Zelle regulieren, weitergeleitetwird. Die Bindung von Insulin/IGF-1 aktiviert die Tyrosinkinaseaktivitat desRezeptors und fuhrt zur Phosphorylierung der IRS Proteine an Tyrosinresten,wodurch Bindungsstellen fur Proteine mit einer Src homology (SH) 2 Domanegeneriert werden. Phosphorylierung von IRS Proteinen durch zahlreiche Serin/Threoninkinasen kann stimulatorische oder hemmende Auswirkungen auf dienachfolgende Signalkaskade zur Folge haben. Es konnte gezeigt werden dass dieHyperphosphorylierung von IRS1 zur Abschaltung des Insulinsignals beitragt,was zur Insulinresistenz fuhren kann. Sowohl IRS1 als auch IRS2 werden pro-teasomal degradiert, wobei vor allem IRS2 zusatzlich durch cAMP-abhangigeGenaktivierung reguliert wird.

14-3-3 Proteine regulieren so unterschiedliche intrazellulare Prozesse wie denZellzyklus, das Zellwachstums, die Gentranskription und die Apoptose. DieSerin/Threoninphosphorylierung in konservierten Motiven des Interaktionspart-ners ist eine Voraussetzung fur die Bindung von 14-3-3 Proteinen. IRS1 undIRS2 wurden bereits als 14-3-3 Bindungspartner beschrieben und die Interaktionvon IRS2 und 14-3-3 Proteinen wurde in der vorliegenden Dissertation genaueruntersucht.

Sowohl Insulin/IGF-1 Stimulation als auch erhohte cAMP-Spiegel konntenals Modulatoren der 14-3-3 Bindung an IRS2 identifiziert werden. IGF-1 Stim-ulation in transfizierten HEK293 Zellen fuhrte zur Bindung von 14-3-3 an IRS2wobei diese Bindung durch Inhibierung des PI 3-Kinase Signalweges und einenAkt/PKB-Kinase Inhibitor gehemmt werden konnte. Die Insulin-vermittelteInteraktion von 14-3-3 und endogenem IRS2 konnte in Ratten Hepatomzellenund in Mauselebern nach akuter Insulinstimulation oder Nahrungsaufnahmebeobachtet werden. Durch die Anwendung von verschiedenen Fragmenten desIRS2 Proteins konnte die IGF-1-abhangige 14-3-3 Bindungsregion im IRS2 inden Bereich der Aminosauren 300 – 600 lokalisiert werden. Eine massenspek-trometrische Analyse ergab 24 phosphorylierte Serin-/Threoninreste im IRS2Protein nach IGF-1 Stimulation, wovon 12 Stellen nur im IRS2 Protein ge-funden wurden, wahrend die anderen auch im IRS1 Protein konserviert sind.Unter den 24 identifizierten Phosphorylierungsstellen im IRS2 Protein befan-den sich auch einige Kandidaten in der 14-3-3 Bindungsregion. Die Mutationen

ix

x

dieser Bindungskandidaten zu Alanin fuhrte zur Identifizierung von Ser573 als14-3-3 Bindungsstelle und es wurde ein phosphospezifischer Antikorper zur weit-eren Charakterisierung der Stelle Ser573 hergestellt. Die Phosphorylierung vonSer573 durch IGF-1 Stimulation war nach Hemmung der PI 3-Kinase und derAkt/PKB-Kinase reduziert. Die Alaninmutante von Ser573 zeigte eine verstarktePhosphorylierung der Akt/PKB-Kinase in einer IGF-1 Zeitreihe. Zusammen-fassend deuten die Daten, welche in dieser Dissertation vorgestellt werden, aufeinen negativen Einfluss der Phosphorylierung von Ser573 auf die nachfolgendeSignalkaskade hin.

Bindung von 14-3-3 an IRS2 durch Stimulation mit Forskolin oder demcAMP Analogon CPT-cAMP (8-(4-chlorophenylthio) adenosin 3’,5’-zyklischesmonophosphat) konnte in HEK293 Zellen gezeigt werden. Diese Bindung konn-te mit dem PKA Inhibitor H89 gehemmt werden. Mittels GST pulldown Ex-perimenten wurde die cAMP/PKA-abhangige 14-3-3 Bindungsregion in denBereich der Aminosauren 952 − 1321 im IRS2 lokalisiert. Eine massenspek-trometrische Analyse identifizierte Ser1137/Ser1138 als cAMP-abhangige und po-tentielle PKA Phosphorylierungsstellen. Eine Hemmung der Akt/PKB-Kinaseoder der ERK-Kinase konnte die cAMP-abhangige Phosphorylierung von IRS2an PKA-Konsensusstellen nicht verhindern. Die Mutation von Ser1137/Ser1138

zu Alanin fuhrte zu einer starken Reduzierung der cAMP-abhangigen 14-3-3Bindung in GST pulldown Experimenten und Ko-Immunprazipitationen. DieDegradation des IRS2 Proteins wurde durch die Anwendung von Cyclohex-imid sichtbar gemacht und eine erhohte Stabilitat konnte beobachtet werden,wenn HEK293 Zellen die stabil IRS2 exprimierten, oder primare Hepatozytenmit Forskolin stimuliert wurden. Die reduzierte Proteindegradation war auf dieStellen Ser1137/Ser1138 zuruckzufuhren, da die Proteinstabilitat der entsprechen-den Ala1137/Ala1138-Mutante nicht durch Forskolinstimulation erhoht werdenkonnte.

Ser1137/Ser1138 werden in dieser Dissertation als neue cAMP-abhangige IRS2Phosphorylierungsstellen vorgestellt. Sowohl ihre Bedeutung bei der Interaktionvon 14-3-3 und IRS2 als auch bei der IRS2 Proteinstabilitat werden dargestellt.Die Ergebnisse deuten auf einen neuen Mechanismus fur die cAMP-abhangigeErhohung der IRS2 Proteinlevel hin, welche fur den Metabolismus der Leberund das Uberleben der β-Zelle von Wichtigkeit sein konnte.

Chapter 1

Introduction

Already in the nineteenth century scientists speculated that there must be aconnection between the pancreas and diabetes. In 1921/22 material extractedfrom the islets of Langerhans of a healthy dog lowered high blood glucose levelsfrom a diabetic dog and after treating a diabetic boy with that purified extractand the resulting reduction of blood glucose Frederick Banting and John Mcleodreceived the Nobel Prize for medicine in 1923 for their discovery of insulin [168].

Insulin administration is one of the standard treatments for insulin-dependentdiabetes and today we now a lot more about the molecular mechanisms that areresponsible for the blood glucose lowering effects of insulin. However, they arestill not completely understood. In this thesis the interaction of insulin receptorsubstrate protein (IRS) 2 and 14-3-3 proteins was investigated. IRS proteinshave been established as important adapter proteins in the early steps of insulinsignal transduction. 14-3-3 proteins also function as adapter proteins and arepart of numerous intracellular processes.

1.1 Diabetes

There are two major types of diabetes. Type I diabetes (T1D) is a chronicdisease characterized by an autoimmune destruction of the pancreatic β-cellsresulting in an absolute insulin deficiency with indispensable dependence on ex-ogenous administration of insulin. Type II diabetes (T2D) is characterized asa metabolic disorder with disturbed carbohydrate, fat and protein metabolismdue to overweight, obesity and physical inactivity. These characteristic lifestylefeatures can be found mostly in industrialized countries with a western lifestyle.Since this kind of living is also desired by people in developing countries, dia-betes has become a global epidemic. According to the World Health Organi-zation more than 346 million people worldwide suffer from diabetes and 80 %occurs in low- and middle income countries1. The metabolic syndrome with adi-positas, hypertension, dyslipidaemia and insulin resistance often precedes T2D.The term “insulin resistance” denominates a status, where the body becomesincapable to react metabolically adequate to an insulin stimulus. Pathologiclong-term effects of diabetes include diabetic nephropathy and retinopathy with

1http://www.who.int/features/factfiles/diabetes/facts/en/index9.html

1

http://www.who.int/features/factfiles/diabetes/facts/en/index9.html

1.1. DIABETES 2

neuronal and peripheral vascular complications. In the early stages of insulin re-sistance the β-cell reacts with increased insulin secretion and β-cell hyperplasia.Life-style intervention with a modest reduction in body weight, moderate phys-ical activity and sensible food-intake at this stage can improve insulin resistanceand T2D patients might not obligatory need to administer exogenous insulin. IfT2D is left disregarded by the patient the β-cells of the pancreas will undergoapoptosis and the exogenous administration of insulin becomes inevitable.

The genetic influence for the susceptibility to develop T2D is not clear yet.No disease-causing mutations have been discovered to date, but there are atleast 17 genetic loci associated with T2D [45]. Jin and Patti state in their reviewthat in addition to genetic factors, pre- and postnatal factors like suboptimalintrauterine environment, obesity, inactivity, gestational diabetes and advancingage can lead to insulin resistance and β-cell dysfunction [78]. Taken together,type II diabetes is a complex disorder arising from genetic as well as epigeneticfactors.

The exact molecular mechanisms of the aforementioned risk factors for thedevelopment of T2D are still not completely understood, but there is emergingknowledge from tissue culture, animal and human studies. Several physiologicaldefects could be identified which account for the development of insulin resis-tance and T2D. The largest insulin-responsive tissue in humans is the skeletalmuscle and a reduced rate of glucose disposal is the first step to insulin re-sistance. A decreased glucose uptake leads to hyperglycemia, whereupon theβ-cells secrete more insulin to counteract the increased blood glucose levels,thereby creating hyperinsulinaemia. This deleterious circle eventually leads toβ-cell hyperplasia and later to apoptosis of the β-cells. Adipose tissue was re-garded only as fat depot for a long time but in recent years it became clearthat adipose tissue depending on the location in the body secretes importanthormones and cytokines like TNFα and IL-6 [116, 157] and this secretion isaltered in insulin resistance. Of note, adipocytes release free fatty acids intothe circulation due to absent inhibition of lipolysis by insulin. In concert withincreased glucose levels glucolipotoxicity further impairs insulin signalling andworsens insulin resistance.

The liver as central organ for carbohydrate, fat and protein metabolism hasa special role in insulin resistance and T2D. Insulin usually suppresses hepaticglucose production by reducing the expression of gluconeogenic enzymes and alack of inhibition is a key feature in T2D [86], thereby the already increasedsystemic blood glucose levels are further increased by glucose molecules fromthe liver. Of note, the expression of SREBP-1c (sterol response element bindingprotein 1c) protein, which acts as a transcription factor for fatty acid synthesisis still stimulated by insulin [73]. In addition, hyperinsulinaemia and insulin re-sistance are discussed as pathogenic factors for the development of nonalcoholicfatty liver disease (NAFLD) [160, 32].

In the following sections the insulin signalling cascade with two of the mostimportant signalling pathways activated by insulin and the intermediate sig-nalling proteins IRS1 and IRS2 are introduced. Furthermore, an overview about14-3-3 proteins is given.

1.2. INSULIN SIGNAL TRANSDUCTION 3

1.2 Insulin signal transduction

1.2.1 Insulin and insulin like growth factor (IGF)-1

Insulin is synthesized in the β-cells of the islets of Langerhans in the pancreas.Initially, preproinsulin which consists of 107 amino acids is synthesized at theendoplasmic reticulum (ER). Two disulfide bonds between the A and the B-peptide and an internal disulfide bond within the A-peptide enable folding andwhen preproinsulin enters the ER 24 amino acids are removed and the generatedmolecule is named proinsulin. The A (21 amino acids) and B-peptide (30 aminoacids) are connected through the 31 amino acid comprising C-peptide. Proin-sulin is stored as hexamer, stabilized by a zinc ion in vesicles of the Golgi appara-tus and upon demand the C-peptide is proteolytically cleaved and active insulinis secreted into the blood stream. It is released into the blood stream in responseto elevated blood glucose concentrations. Fatty acids, amino acids and in-cretins like glucagon like peptide (GLP)-1 and glucose-dependent insulinotropicpolypeptide (GIP) also lead to insulin secretion, whereas glucagon functions asthe major physiological antagonist. Insulin acts on insulin-sensitive tissues suchas liver, skeletal muscle, adipose tissue and brain and exerts metabolic and mito-genic effects like increased glucose uptake, synthesis of glycogen, protein, lipidsand proliferation. The phosphatidylinositol 3-kinase (PI 3-kinase) and Rat sar-coma/mitogen activated protein kinase (Ras/MAPK) pathways represent themajor signalling pathways initiated by insulin and are depicted in Figure 1.4.

Insulin signalling via the insulin receptor (IR) is a very complex process. Toadd another level of complexity, insulin can also signal via the IGF-1 receptorand IGF-1 and -2 can also signal through the IR and IR/IGF hybrid recep-tors. IGF-2 for example can bind to the IR with an affinity of approximately1 % of insulin [137]. IGF-1 and -2 are peptides that show a high homology toinsulin and are thought to function as growth factors in most tissues whereasinsulin regulates glucose and cellular metabolism. Insulin binds with 100-foldhigher affinity to the IR than the IGF receptor and vice versa. Due to the highhomology hybrid receptors can be formed when IR and IGF-1 receptors areco-expressed and they are often the most abundant receptor type and differ intheir ligand binding affinity [173, 155, 13].

Exon 11 of the IR gene encodes for a 12 amino acid region at the C-terminalend of the β-subunit, therefore two isoforms either lacking (IR-A) or possessingthe 12 amino acids (IR-B) are transcribed. Both transcripts show a tissue-specific distribution with IR-A being expressed mainly in the central nervoussystem and hematopoietic cells and IR-B being found in adipose tissue andskeletal muscle [156, 110, 112]. The affinity of insulin for IR-A is 2-fold higherthan for IR-B and the ligand-induced internalization rate for IR-A is about 25 %higher than for IR-B [215].

1.2.2 Insulin receptor and IGF-1 receptor

The IR and the IGF receptor belong to the family of receptor tyrosine kinasesand are structurally similar. They show 50 % overall amino acid sequence ho-mology, their kinase domains share 84 % sequence homology and many commonsteps in their signalling pathways. As mentioned in subsection 1.2.1 insulin andIGF-1 can bind to both receptors and hybrid receptors with differing affinities.


The insulin receptor consists of two α-subunits that build the extracellularpart of the receptor and two β-subunits spanning the membrane with an in-tracellular extension. The α- and β-subunits are covalently connected to eachother via disulfide bonds and binding of an insulin molecule to the α-subunitsleads to a conformational change that brings the β-subunits in close proximity,thus enabling autophosphorylation on distinct tyrosine residues. Autophospho-rylation of the IR on Tyr1146, Tyr1150 and Tyr1151 (human insulin receptornumbering) in the regulatory loop is indispensable for the full activation of thetyrosine kinase activity [70]. The activated IR phosphorylates further tyrosineresidues on the β-subunits, namely Tyr953, Tyr960, Tyr1316 and Tyr1322. Thesephosphorylated tyrosine residues represent binding sites for intracellular Srchomology 2 (SH2) domain containing proteins that serve as substrates for theIR like Shc (Src homology s/α-collagen-related protein) [170], Grb associatedbinder (Gab)-1 [68], APS (adapter protein with a PH and SH2 domain) [4] andthe IRS (insulin receptor substrate) proteins [209]. The adapter proteins showpreferences concerning the amino acid composition surrounding the phospho-rylated tyrosine residues. Recruited substrates get phosphorylated on tyrosineresidues themselves and a subset of signalling proteins with SH2-domains canbind. Of note, the IR and the IGF-1 receptor phosphorylate IRS proteins onthe same tyrosine residues [138, 125, 89]. Insulin stimulation also leads to theactivation of signalling events that terminate the signalling cascade after anappropriate time. Lipid and protein phosphatases and serine/threonine kinasesuncouple and dephosphorylate signalling elements and the IR itself. In addition,the IR is rapidly internalized upon ligand binding. The ligand-receptor complexis internalized into endosomes, thereby other substrates that are not located atthe outer plasma membrane of the cell can be phosphorylated by the IR. Uponacidification of the endosome due to proton pumps, insulin is removed fromthe IR and signalling events are attenuated. Protein tyrosine phosphatases inaddition dephosphorylate the IR and it can be either degraded or relocated tothe cell surface to start a new signalling cycle [19]. Despite their high homology,IR and IGF-1 receptors are not completely functionally redundant. IR-deficientmice are slightly smaller than wild type littermates but die a few days afterbirth from diabetic ketoacidosis [80, 1]. IGF-1 receptor-deficient mice on theother hand show severe growth retardation and die shortly after birth fromrespiratory complications [99, 207].

Taken together, the tissue-specific expression of IR-A, IR-B and hybrid re-ceptors with their differing affinities for insulin, IGF-1 and IGF-2 account forthe diverse actions of insulin, IGF-1 and IGF-2.

1.2.3 General characteristics of IRS proteins

The IRS proteins are involved in the signalling process at a very early stage.They function as multiadapter proteins providing interfaces between activatedreceptors for insulin/IGF-1, growth hormones, several interferons and inter-leukins to various intracellular signalling proteins. In addition, extracellularstimuli like glucose and amino acids can also modulate IRS serine/threoninephosphorylation. The IRS family comprises 6 isoforms varying in their tissueexpression pattern, their biological function and the degree of sequence homol-ogy [181]. They all share the same configuration of the N-terminus: a pleckstrinhomology (PH) domain, followed by a phosphotyrosine binding (PTB) domain


what accounts for the high affinity for the insulin receptor. The C-terminal partcontains several tyrosine residues within distinct motifs that allow interactionwith SH2 domain containing proteins upon phosphorylation. The IRS proteinsfeature unique tyrosine phosphorylation sites but also three well conserved mo-tifs that are responsible for interaction with the p85 regulatory subunit of PI3-kinase (YMXM, multiple motifs are present), with Grb2 (YXNX) and withthe protein tyrosine phosphatase SHP-2 (YIDL and YASI) [115].

IRS1 and IRS2 are widely expressed, whereas IRS3 is mainly limited to thebrain and adipocytes. IRS3 knock-out mice demonstrated that the deletionof IRS3 did not result in a different phenotype compared to wild type mice,probably due to compensatory IRS1 and IRS2 expression [101]. IRS4 expressioncan be seen in embryonic tissues and could be detected in mice in skeletal muscle,liver, heart, brain and kidney [40]. The characterization of IRS4 knock-out miceshowed a relatively mild phenotype of knock-outs compared to their wild typelittermates, they showed growth reduction of about 10 %, produced less welland had decreased blood glucose levels [41]. Wauman et al. could show thatIRS4 couples the leptin receptor to multiple signalling pathways, indicating apotential role of IRS4 in the brain [40]. Tsuruzoe et al. suggested that IRS3 andIRS4 could act as negative regulators of IGF-1 signalling due to the suppressionof IRS1 and IRS2 [187]. Sequence comparisons from IRS5/DOK4 (downstreamof kinase) and IRS6/DOK5 revealed that they are more related to one anotherthan to other IRS proteins, they are both phosphorylated on tyrosine residuesupon IGF-1 or insulin stimulation and could potentially play roles in IGF-1 orinsulin action [23], but so far no important function has been reported yet.

Tyrosine phosphorylation of IRS proteins is the crucial step in the activationof the PI 3-kinase pathway and the Ras/MAPK pathway, since only tyrosinephosphorylation enables the regulatory p85 subunit of PI 3-kinase or Grb2 tobind to IRS proteins [12, 170]. That is why tyrosine phosphatases have an im-portant influence on the regulation of the PI 3-kinase and Ras/MAPK pathway[190]. Reduced tyrosine phosphorylation of IRS proteins in pathologies like obe-sity impairs activation of the aforementioned signalling pathways. Goodyear etal. showed reduced IRS1 tyrosine phosphorylation with decreased PI 3-kinaseactivity in skeletal muscle biopsies from obese subjects compared to samplesfrom lean subjects [56]. Kanety et al. provided data where they showed dimin-ished IRS1 tyrosine phosphorylation by TNFα [81].

A remarkable feature of the IRS proteins in general is the presence of nu-merous serine/threonine residues on the C-terminus that serve to modulate theadapter function of IRS proteins and to integrate different intra- and extracel-lular signals. The different IRS isoforms with varying protein length and thecommon PH and PTB domain are depicted in Figure 1.1. Of note, the kinaseregulatory loop binding domain (KRLB) is unique to IRS2.

Taken together, phosphorylation of tyrosine residues is crucial to activatethe signalling pathways and serine/threonine phosphorylation integrates andmodulates the signalling pathways initiated by the activated receptors.

1.2.4 Physiological importance of IRS1 and IRS2

The most important representatives of the IRS family are IRS1 and IRS2 andtheir biological significance has been demonstrated by knock-out mice mod-els. IRS1 knock-out mice have a metabolic syndrome-like phenotype and show


PH PTB N C KRLB

1 1321

PH PTB

PH PTB

PH PTB

PH PTB

N

N

N

N C

C

C

C

1

1

1

1

1233

495

326

306

IRS-1

IRS-2

IRS-3

IRS-5/DOK4

IRS-6/DOK5

PH PTB N C

1 1216

IRS-4

Figure 1.1: Schematic presentation of IRS isoforms. The individual protein length withregard to the amino acids is shown (PH, pleckstrin homology domain; PTB, phosphotyrosinebinding domain; KRLB, kinase regulatory loop binding domain). Sequences according toUniProt accession numbers for Mus musculus IRS1 (P35569), Mus musculus IRS2 (P81122),Mus musculus IRS3 (O54718), Mus musculus IRS4 (Q9Z0Y7), Homo sapiens IRS5/DOK4(Q8TEW6) and Homo sapiens IRS6/DOK5 (Q9P104).

growth retardation, impaired glucose tolerance and decreased glucose uptakebut do not develop diabetes due to compensatory hepatic IRS2 expression, β-cell hyperplasia and increased insulin secretion [10, 180]. IRS2 knock-out micein contrast display hepatic and skeletal muscle insulin resistance and the β-cellmass continuously decreases with the inevitable development of diabetes [208,88]. Of note, IRS2 knock-out mice do not show compensatory upregulation ofIRS1 and if the diabetes of IRS2 knock-out mice is left untreated these mice dieat 16 weeks of age from diabetic coma. A central role of IRS2 expression in theβ-cell is emphasized by studies from Hennige et al. who showed that reexpres-sion of IRS2 in β-cells in IRS2−/− mice resulted in β-cell growth, survival andinsulin secretion that prevented diabetes in these mice [65]. The involvement ofIRS1 and IRS2 in hepatic metabolism is controversially discussed. The wholebody knock-out models for IRS1 and IRS2 implicate a prominent role for IRS2in the liver [134, 164, 207], but liver-specific knock-out models imply a rathercomplementary role for IRS1 and IRS2 [33, 182, 167]. IRS1 and IRS2 share ahigh sequence identity, resulting in overlapping but also distinct functions andwill be described in more detail in the following sections.

1.2.5 IRS1

IRS1 is the predominant isoform in adipose tissue and skeletal muscle. IRS1was cloned for the first time in 1991 and Sun et al. demonstrated that IRS1gets phosphorylated on tyrosine residues upon insulin stimulation in CHO cellsstably expressing IRS1 and IR and that IRS1 binds PI 3-kinase, thereby actingitself as a docking molecule [177]. Phosphorylation of Tyr960 within the motifNPEY on the IR is essential for the binding of IRS1 to the IR [203].

IRS1 abundance is regulated on the protein level and on the transcriptionallevel, but differing results depending on cell type and tissue are published. IRS1mRNA expression in 3T3-L1 and 3T3-F442A adipocytes can be reduced by dex-


amethasone treatment [188, 142] whereas it increased IRS1 protein expressionin rat hepatoma cells [143]. IRS1 was found to be expressed in β-cells of thepancreas [61] and high fat feeding decreased IRS1 protein levels in epididymaladipocytes [9]. Short-term insulin stimulation in rat hepatoma cells as well asrefeeding in mice increases hepatic IRS1 levels [141], whereas fasting of micedoes not change hepatic IRS1 protein levels [87]. The short-term administrationof insulin in humans results in increased IRS1 levels in muscle [141].

Regulatory subunits of PI 3-kinase interact with tyrosine phosphorylatedIRS1 and this interaction can be modulated negatively by the SH2 domain con-taining protein tyrosine phosphatase SHP-2 [115]. Phosphorylation of Tyr895 ofIRS1 couples insulin signalling to Ras signalling via binding of Grb2 to IRS1[170]. The regulatory subunits of PI 3-kinase also interact with tyrosine phos-phorylated IRS1, as well as Nck [91] and c-fyn [176].

Multiple serine/threonine sites on IRS1 have been described in the litera-ture whose phosphorylation mainly seems to negatively regulate signalling, butalso positive regulation has been observed. For example, Giraud et al. showedthat Ser302 phosphorylation is inducible by stimulation with IGF-1 or insulinand treatment of cells with glucose or amino acids and stimulation of this sitepromoted mitogenesis and cell growth [54]. An inhibitory effect on tyrosine phos-phorylation by JNK phosphorylation of Ser307 has been established by Aguirreet al. [3]. Nevertheless, one should always keep in mind that the aforementionedsites cannot be looked at in an isolated manner, the overall phosphorylation pat-tern instead is important for the signalling process. Single alanine mutants areused to investigate the influence of single sites on the signalling cascade, butin vivo the entity of serine/threonine residues work in concert to regulate sig-nalling – they can depend on each other and can be phosphorylated be severalkinases at different time points. Werner et al. established phosphorylation ofSer302 and Ser307 by JNK1 is necessary to disrupt IR/IRS1 interaction [202].The phosphorylation of Ser302 is a prerequisite for phosphorylation of Ser318 byPKCζ as has been shown by Weigert et al. [201]. Further, they could show thatphosphorylation events occur at different time points of the signalling process.Ser302 and Ser318 were involved in the first 10 min of signalling, whereas Ser307

and Ser318 were maximally phosphorylated at 30 and 60 min of insulin stimu-lation. Ser612, Ser622, Ser632 and Ser731 modulate the tyrosine phosphorylationof IRS1 and its association with p85 of PI 3-kinase [113]. S6K1 phosphorylatesSer270 on IRS1 upon TNFα stimulation and this is the prerequisite for furtherphosphorylation of IRS1 by S6K1 on residues Ser307, Ser636 and Ser1101 [222].

In Figure 1.2 the IRS1 protein structure is displayed in a simplified way.Serine and tyrosine phosphorylation sites that were experimentally tested withfor example loss-of-function mutants are indicated. If the responsible kinase isknown this is indicated by an arrow above the IRS1 protein scheme and bindingpartners are shown below. For a further overview of IRS1 phosphorylation sitessee the review from Taniguchi et al. [181].

Serine/threonine phosphorylation not only modulates signalling, but can alsobe used to mark the IRS proteins for degradation. When 3T3-L1 adipocytes areexposed to insulin for several hours the IRS1 protein content decreases and this isnot due to changed protein synthesis but increased degradation [135]. Pedersonand co-workers showed insulin-mediated IRS1 protein degradation in adipocytesthrough a rapamycin-sensitive pathway [127]. However, a partial participationof the PI 3-kinase pathway in insulin-induced IRS1 degradation in adipocytes


PH PTB N C IRS-1

Y1222 Y1172 Y895

Grb2

S302

JNK

S307 S318

PKCζ

S6K1

S270

S636

S1101

S6K1

S622 S632 S612 S731 S332

GSK3

S6K1

S6K1

p85

Y460 Y546

Y608

Y628 Y658 Y690 Y935

SHP2 c-fyn

Figure 1.2: Collection of reported phosphorylated serine and tyrosine residues on IRS1.Kinases are shown above the IRS1 protein scheme whereas binding partners are shown below.

was shown by Haruta et al. although they also showed a rapamycin-sensitivepathway to be involved in IRS1 degradation [62]. Insulin-independent IRS1 pro-tein degradation can be mediated by glycogen synthase kinase 3β (GSK3β). Itphosphorylates IRS1 at Ser332 and thereby induces ubiquitinylation with sub-sequent proteasomal degradation without the participation of IR signalling andPI 3-kinase in CHO cells treated with high glucose concentrations [93]. GSK3βactivity is increased in type II diabetes in basal and insulin-stimulated muscleand thereby could play a causative role in the development of insulin resistance[117].

1.2.6 IRS2

In 1995 Tobe et al. and Patti et al. discovered IRS2 when they investigated IRS1knock-out mice [184, 125]. Since these mice displayed only mild insulin resistanceboth groups figured that there must be an alternate substrate for the IR tomediate insulin signalling, resulting in the identification of IRS2. Similar toIRS1, IRS2 contains numerous serine/threonine residues that potentially couldbe phosphorylated by a variety of kinases like Akt/PKB, S6K, JNK, GSK3,PKA, PKC, IKKβ, MAPK and AMPK [158, 20]. Both proteins display a highhomology and show overlapping yet also distinct functions, but the molecularmechanisms that distinguish IRS1 from IRS2 function are not fully understood.In their C-terminus IRS1 and IRS2 only share 35 % of sequence identity withsimilar tyrosine phosphorylation motifs that are surrounded by variable aminoacid sequences [178].

The most important structural feature that distinguishes IRS2 from IRS1and all other IRS proteins is the presence of a domain reaching from aminoacids 591-733 (according to IRS2 mouse numbering) what was named the ki-nase regulatory loop binding (KRLB) domain [153, 152, 210]. It was definedby yeast two-hybrid studies and serves as negative regulatory element to con-trol the extent of IRS2 tyrosine phosphorylation. IRS2 can bind to the IR viaphosphorylated Tyr960 in the NPEY motif, but it is not essential for the inter-action with the IR [152]. The two tyrosine residues Tyr624 and Tyr628 withinthe motif PYPEDY in the KRLB domain mediate binding to the IR. Althoughthe corresponding region in IRS1 shows quite sequence similarity, mutations toconstitute the PYPEDY motif in IRS1 does not lead to stable interaction withthe IR.

IRS2 is expressed in skeletal muscle, liver, brain, β-cells of the pancreas


and to a lesser extent in testis and adipocytes. IRS2 expression and functionis especially important in the liver and β-cells of the pancreas. Hepatic IRS2expression is tightly regulated by fasting and refeeding [24, 48]. Fasting inducesIRS2 mRNA expression through activation of the transcription factor CREB.In addition, TORC2 (transducer of regulated CREB activity 2) is an impor-tant co-activator for CREB-dependent IRS2 expression in hepatocytes [149].IRS2 expression is suppressed by sterol regulatory element binding proteins(SREBPs) that control lipid synthesis. They directly repress IRS2 expressionby replacing forkhead transcription factors on the IRS2 promotor. RepressedIRS2 expression by SREBPs serves as a switch to lipogenesis from glycogensynthesis. This could contribute to hepatic insulin resistance which is associ-ated with hepatosteatosis [73]. Insulin also suppresses IRS2 protein expressionby reducing the transcription rate of the IRS2 gene [73]. Simpson et al. couldshow that the phosphatase and tensin homolog (PTEN) apparently can regulateIRS2 expression in a positive manner and suggest the PTEN substrate PIP3 canregulate the amount of IRS2 available for insulin signalling [169]. In pancreaticβ-cells IRS2 expression is crucial for survival and function [154]. β-cell specificexpression of IRS2 in IRS2 knock-out, obese and streptozotocin-treated miceprevents diabetes in these mice [65]. IRS2 turnover is rapid, mainly regulatedat the transcriptional level, predominantly by FoxO3a (forkhead box class O)and can be augmented by elevated glucose and cAMP levels [76, 97, 185].

Similar to IRS1, IRS2 can also be degraded via the proteasome. Prolongedinsulin stimulation and refeeding leads to proteasomal degradation of IRS2 [87,48, 164]. Rui et al. showed that prolonged insulin/IGF-1 stimulation led todegradation of IRS2 via the proteasome in a PI 3-kinase and mTOR-dependentmanner in 3T3 cells, rat hepatoma cells and mouse embryo fibroblasts [139].

Compared to IRS1 less is known about individual IRS2 serine/threoninephosphorylation sites and their potential inhibitory or stimulatory effects. Ser-ine/threonine phosphorylation of IRS2 has been found to exert negative effectson hepatic insulin signalling. An example is the phosphorylation of IRS2 onSer488 by JNK and Ser484 by GSK3 [159]. Phosphorylation by JNK as thepriming kinase with subsequent phosphorylation by GSK3 reduced IRS2 tyro-sine phosphorylation and the association with p85 in H4IIE cells overexpressingJNK and GSK3.

Gurevitch et al. mutated five serine residues on IRS2 simultaneously (Ser303,Ser343, Ser362, Ser381 and Ser480) to alanine. This alanine mutant showed pro-longed tyrosine phosphorylation compared to wild type IRS2 and increasedbinding to the IR in CHO cells. Murine islets adenovirally infected with thisalanine mutant showed reduced apoptosis when treated with pro-inflammatorycytokines [60]. Our laboratory could show that Ser675 and Ser907 of IRS2 arephosphorylation sites for mTOR and ERK respectively and mutation of Ser675

to alanine increased the IRS2 protein half-life [48].In Figure 1.3 the IRS2 protein scheme is displayed in a simplified way. Ex-

perimentally reported serine and tyrosine phosphorylation sites are indicated. Ifthe responsible kinase is known this is indicated by an arrow above the proteinscheme and binding partners are shown below.

Due to the fact that several serine/threonine residues present in IRS1 canalso be found in IRS2 and the overlapping but also distinct functions of IRS1and IRS2 it is a major challenge to dissect the IRS2 serine/threonine phosphory-lation sites. The knowledge about the precise mechanisms that regulate IRS2


PH PTB N C KRLB IRS-2

Y628 Y624

IR

S484 S488

GSK3 JNK

S343 S381

S480

S303 S362 S675 S907

mTOR ERK

Y671 Y536 Y538

Y594

Y628

Y649 Y734

Y758

Y814

Y1061

Y1242 Y1303

SHP2

p85 Y911

Grb2

Figure 1.3: Collection of reported phosphorylated serine and tyrosine residues on IRS2.Kinases are shown above the IRS2 protein scheme wheras binding partners are shown below.

function could open up novel possibilities for drug development and treatmentof pathologies like type 2 diabetes.

1.2.7 PI 3-kinase pathway

Other growth factors and hormones can activate PI 3-kinase, but only insulinhas the ability to effectively induce glucose metabolism. There are three differ-ent functional classes of PI-kinases that differ in their protein domain structure,lipid substrate specificity and the associated regulatory subunits. The class IPI 3-kinases transduce signals from receptor tyrosine kinases and G-protein cou-pled receptors [46]. Class IA PI 3-kinase catalyzes the formation of PI-3,4,5-P3

(PIP3) from PI-4,5-P2. Five regulatory isoforms (p85α, p85β, p55α, p55γ andp50α) and two catalytic isoforms (p110α and p110β) have been discovered todate [124, 130, 75, 74]. All regulatory subunits have the ability to recruit p110with subsequent PI 3-kinase activation [162]. The p85 regulatory subunit inter-acts via its two SH2 domains with YXXM or YMXM motifs on IRS proteins[12]. For full activation of PI 3-kinase it is important that both SH2 domainsof the regulatory subunit bind to IRS [11]. PIP3 acts as a membrane tether forexample for PDK (phosphoinositide-dependent kinase) 1 and Akt/PKB. By in-teraction of Akt/PKB with PIP3 via its PH domain it translocates to the plasmamembrane where it gets phosphorylated, and thereby activated, on Thr308 andSer473 through PDK1 and mTOR respectively [150]. Activation of the PI 3-kinase pathway leads to GLUT4 translocation to the plasma membrane withsubsequently increased glucose uptake, glycogen synthesis, induction of triglyc-eride synthesis, suppression of gluconeogenesis by inhibition of gluconeogenicenzyme gene expression and DNA synthesis [145, 147, 16]. For a schematicoverview see Figure 1.4.

1.2.8 Ras/MAPK pathway

Via activation of the Ras/MAPK pathway insulin mediates its mitogenic effects.The Ras/MAPK pathway can be initiated by the recruitment of Grb2 (Growthfactor receptor-bound 2) and Sos (son of sevenless). This can be achieved byphosphorylation of tyrosine residues within YVNI motifs on IRS1 resulting in arecognition motif for SH2-domain containing protein Grb2 that associates with

1.3. 14-3-3 PROTEINS 11

IRS1. Grb2 contains a SH3-domain which can interact with the guanine nu-cleotide exchange factor Sos. Grb2 and Sos can be alternatively recruited to anactivated receptor by Shc which interacts with the receptor via its PTB domain.Sos exchanges GDP for GTP on Ras, whereupon Ras can bind and thereby ac-tivate Raf (MAPKKK). Raf activates MEK (MAPKK) by phosphorylation of aserine residue within its activation loop. MEK1/2 then activates ERK (MAPKor extracellular regulated kinase) by phosphorylation. ERK itself phosphory-lates cellular proteins [219], but also translocates into the nucleus where it ac-tivates various transcription factors. Taken together, the Ras/MAPK pathwaycontrols cellular processes like proliferation, migration and differentiation (re-viewed in [82]. In Figure 1.4 the PI 3-kinase and the Ras/MAPK kinase pathwayare depicted.

insulin

IRS-1/-2 IRS-1/-2 p85

p110

PI 3-kinase

PI 3-kinase PIP3

PDK1

Akt/PKB

mTOR

glucose uptake

protein biosynthesis

DNA synsthesis

glycogen synthesis

gluconeogenesis

Grb2

Sos Ras

Raf

MEK1/2

ERK

proliferation

cytosol

GβL

Rictor Sin1

mTORC2

GβL Raptor

mTOR

mTORC1

GLUT4

FOXO GSK3

Figure 1.4: Schematic illustration of the PI 3-kinase and Ras/MAPK pathway. Abbrevia-tions: ERK, extracellular signal regulated kinase; FOXO, forkhead box O; Grb2, growth recep-tor bound protein-2; GSK3, glycogen synthase kinase-3; IRS, insulin receptor substrate; MEK,MAP kinase ERK kinase; mTORC, mammalian target of rapamycin; PDK1, phosphoinositide-dependent kinase 1; PI 3-kinase, phosphatidylinositol 3-kinase; Akt/PKB, protein kinase B;Sos, son of sevenless.

1.3 14-3-3 proteins

Posttranslational modification of serine and threonine residues by phosphory-lation is regarded as mechanism to regulate catalytic activity, but these modifi-cations also allow the formation of multimolecular signalling complexes. This isachieved by a multitude of proteins that can recognize modified serine/threonineresidues. These recognition elements are domains in proteins like WW domains,forkhead-associated (FHA) domains and WD40 repeats/LRR modules in F-boxproteins. 14-3-3 proteins represent a protein family also capable of recognizingphosphorylated serine/threonine residues. Over the past two decades it becameclear that 14-3-3 proteins are part of virtual all important cellular processes.

1.3. 14-3-3 PROTEINS 12

Figure 1.5: The two 14-3-3 monomers β and τ build a heterodimer. All seven 14-3-3 isoformsare superimposed and the other isoforms beside β and τ are shaded in grey [217].

They are versatile regulators of numerous cellular processes with no intrinsicenzymatic activity, highly conserved and expressed in all eukaryotes [49]. Thehigh level of sequence homology and functional conservation is emphasized bythe fact that 14-3-3 isoforms from yeast, plants and mammals are functionallyinterchangeable [161]. Expression of various 14-3-3 isoforms in different numberscan be found in different species. The mammalian family of 14-3-3 proteins com-prises seven isoforms encoded by seven genes. The nomenclature was based on14-3-3 isolated from bovine brain and separated by reversed phase chromatog-raphy. These analysis revealed a complex mixture consisting of 7 polypeptideswhich were assigned seven greek letters, namely α, β, γ, δ, ε, ζ and η [72]. Afew years later Aitken et al. discovered that α and δ are the phosphorylated βand ζ isoforms [8]. Two other isoforms, τ and σ, were found to be expressed inT cells and epithelial cells [5]. Drosophila and yeast express two genes whereasArabidopsis plants contain 15 genes for 14-3-3 proteins [171, 195, 44].

1.3.1 Structure

A special feature of the 14-3-3 proteins is their high sequence conservation.Maximal isotype diversification can only be found in the acidic stretch of theC-terminus [213].

14-3-3 proteins are synthesized as ∼ 30 kDa proteins and built-up from 9 an-tiparallel α-helices (αA-αI) with a 40 A channel in the center of the cup shapedprotein, where the first four are crucial for the dimer formation and helices αC,αE, αG and αI form the conserved peptide binding groove [212, 100]. Two 14-3-3 proteins contact each other with their N termini, thereby forming a centralgroove while the C-terminal helices form a wall around the central groove asshown in Figure 1.5 [98, 212]. The central groove is also named dimer interface

1.3. 14-3-3 PROTEINS 13

Figure 1.6: 14-3-3 cladogram. Shown are 14-3-3 isoforms from fission yeast (spRad25p andspRad24p), budding yeast (scBMH1 and scBMH2) and the human isoforms [205].

and the amino acid residues located within are strictly conserved throughoutyeast and mammalian species. Liu et al. and Silhan et al. established thatthe C-terminal residues are acidic and therefore can interact with the basicresidues in the dimer interface. This interaction is mainly based on two glu-tamate residues that are negatively charged, thereby mimicking the negativecharge of a phosphate group. This brings the C-terminus on top of the centralgroove thereby masking it and upon ligand binding the C-terminus adopts anew conformation and is ejected from the central groove [98, 166]. In Figure 1.5a 14-3-3 heterodimer consisting of the β and τ isoforms are shown to illustratethe cup shape. In addition, the high similarity between the different isoformsis expressed by the superimposition of all isoforms which conform to nearly thesame structure.

14-3-3 proteins form homo- and heterodimers depending on steric compati-bility [25]. 14-3-3σ preferentially forms homodimers [205], whereas 14-3-3ε seemsto preferentially form heterodimers [25, 217]. From an evolutionary point of view14-3-3σ is the most divergent isoform in comparison to the other six family mem-bers, shown in the cladogram from Wilker et al. in Figure 1.6 [205]. This is inline with distinct functions in vivo. Of note, the key residues that coordinate thephosphate group of the binding protein are the same as in other 14-3-3 isoforms.There are four stabilizing interactions in the σ homodimer that cannot occur inheterodimers. In addition, three amino acids (Met202, Asp204 and His206) areunique to 14-3-3σ and enable 14-3-3σ to specifically bind its substrates [205].

1.3.2 Binding to phosphorylated serine/threonine residues

Studies with dimerization-deficient 14-3-3 mutants revealed a significant roleof dimerization for phosphorylation-dependent target binding [161]. Muslin etal. discovered that the recognition of a target protein by 14-3-3 proteins isdependent on a phosphoserine and identified a putative motif for 14-3-3 binding(RSXpSXP, p indicates phosphorylated serine, X denotes any amino acid) andshowed that phosphorylated serine is crucial for recognition by 14-3-3 [114].

1.3. 14-3-3 PROTEINS 14

They established the importance of the arginine in position -3 relative to thephosphorylated serine on the target protein, since it is required for recognitionby kinases for phosphorylation. Further studies led to the nomination of two 14-3-3 binding motifs, namely mode I (RSXpSXP) and mode II (RX(Y/F)pSXP)[214, 136]. Additionally, Yaffe et al. showed binding of a 14-3-3 dimer with a30-fold increase in affinity for a peptide containing two phosphoserine motifs[214].

The phosphopeptide bonds are built with the three absolutely conserved re-sidues Lys49, Arg56 and Arg127 along with Tyr128, by formation of a basic pocketwhereas the surrounding amino acids are negatively charged [213, 114]. 14-3-3binding to a phosphorylated target is a two-step process where in the first stepthe binding of the phosphorylated peptide to the central groove is dependenton general protein-protein interaction motifs and the second step involves inter-action of the globular domain of the target protein with the remaining sectionsof the 14-3-3 dimer [217]. Both mode I and mode II binding motifs include aproline at position +2 relative to the phosphorylated serine/threonine. Yaffe etal. hypothesized that the proline allows the remainder of the peptide/proteinto exit the binding cleft [214], but an extensive analysis of > 200 reported 14-3-3 binding targets by Johnson et al. revealed that a proline in position +2is only found in about 50 % of the target proteins thus raising the question ifthis is always essential for 14-3-3 binding [79]. In addition, Aitken states thatthe defined binding sequences surrounding the phosphorylated serine/threonineresidues do often not correspond to the 14-3-3 binding motifs defined by peptidelibrary studies [6]. This could be due to affinity, because if they would exactlymatch the binding motif, the strength of the binding would be too strong tocarry out a regulatory role.

It is now well established that phosphorylation of serine/threonine is a pre-requisite for recognition by 14-3-3 proteins, but there is one report where phos-phorylation of the target protein is not required for 14-3-3 binding [108]. Ofnote, a report from Waterman and co-workers shows that dephosphorylation ofa serine residue on the tumor suppressor p53 creates a 14-3-3 binding motif [198].Some studies report 14-3-3 binding to only one phosphorylated serine/threonineresidue [17, 27, 128], whereas binding to two simultaneously phosphorylated ser-ine/threonine residues is often reported [133, 51, 29, 118]. The question ariseshow the 14-3-3 isoforms can exert different binding specificities, when they areso similar? Yaffe et al. propose that the binding specificity is not derived fromdifferences in phosphopeptide binding motifs, but from variable contacts fromthe surface of 14-3-3 proteins outside the binding cleft [214]. Noteworthy, Yanget al. propose a certain conformational flexibility in the peptide binding siteand the dimer interface that allows binding of peptides and proteins of varyingsize and sequence [217]. Another important fact concerning 14-3-3 specificityis the tissue [129, 197, 174] and temporal distribution [183, 196] as well as thesubcellular localization [50, 163] of 14-3-3 proteins.

1.3.3 Regulation of 14-3-3 proteins

14-3-3 proteins are regulated by phosphorylation and isoform-specific regula-tion is possible because the phosphorylation sites are not conserved among theisoforms [120]. Phosphorylation can influence the dimer and monomer statusas well as target specificity. For example does phosphorylation of Ser184 and

1.3. 14-3-3 PROTEINS 15

Ser/Thr232, which are in close proximity of the ligand binding groove, reduce lig-and binding [220, 186, 179, 36, 7]. Ser58 is located within the dimer interface andits phosphorylation promotes monomer formation [58, 131]. These data implythat phosphorylation of 14-3-3 proteins serves as a rather negative regulatorymechanism, since the outcome is reduced interaction with target proteins.

Beside serine phosphorylation of 14-3-3 proteins to regulate their functiontyrosine phosphorylation has also been described. Phosphorylation of 14-3-3ζon Tyr179 leads to a scaffolding complex where via Tyr179 Shc bound with itsSH2 domain and this complex is indispensaple for formation of a PI 3-kinase sig-nalling complex [15]. There are other studies where tyrosine phosphorylation of14-3-3 proteins were reported, but the exact sites were not determined [52, 102].Also acylation, acetylation, polyglycation, oxidation and proteolytic processinghas been described and is reviewed by Aitken [6].

1.3.4 Function

14-3-3 interactome studies show the involvement of 14-3-3 proteins in cell migra-tion and shape change and the participation of 14-3-3σ in cell cycle regulation[77, 18, 132, 109]. Yaffe proposed that by the rigid structure of the 14-3-3 pro-teins with only minor possible movement between free and peptide bound forms,14-3-3 dimers can provide themselves as a molecular anvil whereupon the tar-get protein can be reshaped [213]. It is suggested by Yaffe et al. that 14-3-3proteins and the family of AGC kinases coevolved, since basic residues 3 or 4positions N-terminal to the phosphoserine are the consensus motif required forphosphorylation by these kinases [214] and the basic residue arginine is crucialin both 14-3-3 binding motifs.

The precise role of 14-3-3 proteins is hard to define, since the number ofbinding partners increases constantly and involves very divergent proteins likekinases, phosphatases, receptors, transcription factors and regulatory proteins.According to Kjarland et al. do 14-3-3 proteins provide a mechanism to simulta-neously influence several intracellular processes [84]. Binding of 14-3-3 proteinsto target proteins can result in several possible outcomes: conformational changeof the target protein that influences the activity of the target protein; actingas scaffold to connect proteins; obscuring a nearby binding/interaction motifthus modulating interaction of the target protein with other binding partners;regulation of translocation upon binding.

The serotonin N-acetyltransferase (AANAT) is a good example for confor-mational change of the target protein with subsequently altered activity. Uponbinding of 14-3-3 the affinity and activity of AANAT for its substrates is in-creased [119].

Braselman and McCormick presented an example for 14-3-3 proteins actingas a scaffold where Bcr and Raf form a complex in vivo via 14-3-3 proteins [22].Another example for a scaffold function was presented by Agarwal-Mawal andco-workers who showed 14-3-3ζ simultaneously bound GSK3β and τ , therebybringing them into close proximity resulting in phosphorylation of tau by GSK3β[2].

14-3-3 binding to the cytoplasmic adapter 3BP2 (c-Abl Src homology 3domain-binding protein-2, also SH3BP2) upon phosphorylation of two serineresidues modulates NFAT (nuclear factor of activated T-cells)/AP (activatorprotein)-1 transcriptional activities in T cells [47]. A 3BP2 mutant incapable of

1.4. AIMS OF THE THESIS 16

14-3-3 binding showed an increased capacity to stimulate transcriptional activ-ities, thus 14-3-3 binding modulates 3BP2 function.

The regulation of cellular localization by 14-3-3 proteins has been shownfor example for forward transport from proteins from the endoplasmic reticu-lum (ER). Usually the proteins contain a β-COP (coatomer protein) bindingsite that functions as retention signal for the ER. Upon phosphorylation inthe “release site” 14-3-3 proteins bind and suppress β-COP binding, therebyovercoming the retention signal and proteins are transported forward [122]. Nu-clear/cytoplasmic shuttling from FOXO proteins and carbohydrate responseelement-binding protein (ChREBP) is also mediated by 14-3-3 proteins [223,146].

Since 14-3-3 proteins are part of various cellular processes it is not surprisingthat they are involved in the pathogenesis of many diseases. Isotype-specificinvolvement in diseases includes the σ isoform that acts as a negative regulatorof cell cycle and has been associated with breast cancer [192] and is reportedto be down-regulated or completely lost in breast and prostate cancer cells [42,69]. The ζ isoform has been found to be elevated in the cerebrospinal fluid ofneuroleptic patients [221]. Umahara et al. investigated the localization of 14-3-3isoforms in atherosclerotic lesions from human cerebral and carotid arteries andshow that the different isoforms differed in their expression levels in the diversecell types that are found in atherosclerotic lesions [191].

Over the past few years 14-3-3 proteins and their expression levels seem tobe investigated as potential biomarkers for cancer or as predictors for tumorresponse. Hodgkinson et al. for example report 14-3-3 proteins to be differen-tially expressed in chemotherapy-resistant and chemotherapy-sensitive oestro-gen receptor-positive breast cancer [67]. The analysis of 14-3-3 protein expres-sion therefore could function to predict beneficial chemotherapy in oestrogenreceptor-positive breast cancer patients. Another example is the reduced 14-3-3ε expression in gastric cancer samples that was associated with diffuse-typegastric cancer and an early onset of this cancer type [90].

1.4 Aims of the thesis

The importance of the IRS proteins in the insulin/IGF-1 signalling cascade andthe manifoldness of 14-3-3 protein function are well established. The multipleserine/threonine residues that can modulate signalling positively or negativelyand the still incomplete understanding of the signalling processes make it in-evitable to further investigate the IRS proteins and IRS2 in particular. Dys-regulated serine/threonine phosphorylation patterns of IRS proteins can lead topathophysiological changes. Additional information about inhibitory or stimula-tory serine/threonine residues on IRS2 could potentially lead to the developmentof new pharmaceuticals to treat insulin resistance and T2D.

IRS1 and IRS2 have been shown to interact with 14-3-3 proteins. A clear rolefor 14-3-3 binding to IRS1 could not be established yet, although data indicatea rather negative regulatory role.

The aim of this thesis was to characterize the interaction between 14-3-3proteins and IRS2. This included identification of stimuli that led to interac-tion between 14-3-3 and IRS2. With fragments of IRS2 that encompass differentregions of the IRS2 protein the 14-3-3 binding regions were localized. Mass spec-

1.4. AIMS OF THE THESIS 17

trometry was applied to identify residues that were phosphorylated by stimulithat also enabled 14-3-3 binding to IRS2. These candidate residues were testedfor their importance in 14-3-3 binding and their potential kinases were identi-fied. A physiological occurrence and importance for the interaction of 14-3-3and IRS2 was also investigated.

Chapter 2

Materials and Methods

2.1 Chemicals

Table 2.1: Chemicals

Substance Company

12-O-tetradecanoylphorbol13-acetate (TPA)

MerckDarmstadt, Germany

14-3-3ζ-HRP, human recombinant R&D SystemsWiesbaden-Nordenstadt, Germany

2-[bis(2-hydroxyethyl)ammonio]ethanesulfonate (BES)

Sigma-AldrichMunich, Germany

3-aminophthalhydrazide (luminol) Sigma-AldrichMunich, Germany

4 − 12 % Bis-tris mini gel Invitrogen,Karlsruhe, Germany

4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid(HEPES)

RothKarlsruhe, Germany

8-(4-chlorophenylthio) adenosine3′, 5′-cyclic monophosphate(CPT-cAMP)

Sigma-AldrichMunich, Germany

A23187 CalbiochemSchwabach, Germany

A-769662 Division of Signal Transduction andTherapy, University of DundeeDundee, UK

acetic acid, 100 %, waterfree Sigma-AldrichMunich, Germany

18

2.1. CHEMICALS 19

Substance Company

acrylamide 30 (37.5:1) RothKarlsruhe, Germany

agarose, peqgold universal PeqlabErlangen, Germany

Akt inhibitor Akti-1/2 CalbiochemSchwabach, Germany

ammonium persulfate (APS) Sigma-AldrichMunich, Germany

ampicillin Sigma-AldrichMunich, Germany

BID 1870 Division of Signal Transductionand Therapy, University ofDundeeDundee, UK

bovine serum albumin (BSA) RocheMannheim, Germany

bromophenol blue Sigma-AldrichMunich, Germany

calcium chloride (CaCl2) Sigma-AldrichMunich, Germany

Calyculin A Cell Signaling TechnologyFrankfurt, Germany

complete tablet RocheMannheim, Germany

Coomassie Brilliant Blue G-250 MerckDarmstadt, Germany

disodium hydrogen phosphate (Na2HPO4) MerckDarmstadt, Germany

DMEM high glucose (4.5g/l) LonzaCologne, Germany

dimethylsulfoxide (DMSO) RothKarlsruhe, Germany

DPBS LonzaCologne, Germany

E. coli (DH5α) New England BiolabsSchwalbach, Germany

epidermal growth factor (EGF) Sigma-AldrichMunich, Germany

ethanol p.A. MerckDarmstadt, Germany

ethidium bromide Sigma-AldrichMunich, Germany

2.1. CHEMICALS 20

Substance Company

ethylenediaminetetraacetic acid (EDTA) Sigma-AldrichMunich, Germany

ethylenglycol tetraacetic acid (EGTA) Sigma-AldrichMunich, Germany

fetal calve serum (FCS) HyClone Thermo Fisher ScientificSchwerte, Germany

forskolin ApplichemDarmstadt, Germany

gelatine MerckDarmstadt, Germany

GFP Trap R© ChromotekMartinsried, Germany

glutathione sepharose 4B (suspension) GE HealthcareMunich, Germany

glycerol MerckDarmstadt, Germany

glycine MerckDarmstadt, Germany

H2O, HPLC-grade MerckDarmstadt, Germany

H89 CalbiochemSchwabach, Germany

hepatocyte maintenance medium withsupplements

ProvitroBerlin, Germany

Hind III (10 U/µl) RocheMannheim, Germany

hydrochloric acid (HCl) ApplichemDarmstadt, Germany

hydrogen peroxide (H2O2) MerckDarmstadt, Germany

hygromycin B solution PAColbe, Germany

insulin like growth factor 1 (IGF-1) Sigma-AldrichMunich, Germany

insulin, human recombinant Sigma-AldrichMunich, Germany

Insuman rapid (insulin for mice studies) Sanovi AventisFrankfurt, Germany

isopropyl-β-D-thiogalactopyranoside (IPTG)


2.1. CHEMICALS 21

Substance Company

LB agar Sigma-AldrichMunich, Germany

LB broth Sigma-AldrichMunich, Germany

L-glutamine (200mM in 0.85 % NaCl) LonzaCologne, Germany

magnesium chloride (MgCl2) Sigma-AldrichMunich, Germany

magnesium sulfate (MgSO4) MerckDarmstadt, Germany

methanol Normapur VWRDarmstadt, Germany

N,N,N′,N′-tetramethyl-ethylenediamine (TEMED)


nonfat dried milk powder ApplichemDarmstadt, Germany

NuPAGE R© MOPS SDS runningbuffer

InvitrogenKarlsruhe, Germany

PEG 3350 Sigma-AldrichMunich, Germany

penicillin/streptomycin LonzaCologne, Germany

PI-103 Division of Signal Transductionand Therapy, University ofDundeeDundee, UK

p-iodophenole Sigma-AldrichMunich, Germany

polyethylenimine (MW 40000) Polysciences Inc.Eppelheim, Germany

Ponceau S Sigma-AldrichMunich, Germany

potassium chloride (KCl) Sigma-AldrichMunich, Germany

protein A sepharose (suspension) GE HealthcareMunich, Germany

protein G sepharose (suspension) GE HealthcareMunich, Germany

quickload 1 kb DNA ladder New England BiolabsBeverly, USA

RPMI 1640 PAAColbe, Germany

2.1. CHEMICALS 22

Substance Company

Sephadex G-50 superfine GE HealthcareMunich, Germany

sequencing buffer 5× Applied BiosystemsFoster City, CA, USA

sodium chloride (NaCl) MerckDarmstadt, Germany

sodium dihydrogen phosphate (NaH2PO4) MerckDarmstadt, Germany

sodium dodecyl sulfate (SDS) BioradMunich, Germany

sodium fluoride (NaF) Sigma-AldrichMunich, Germany

sodium hydrogen carbonate (NaHCO3) MerckDarmstadt, Germany

sodium-ortho-vanadate (Na3VO4) Sigma-AldrichMunich, Germany

sodium-pyrophosphate (Na4P2O7) Sigma-AldrichMunich, Germany

tris(hydroxymethyl)aminomethane (TRIS)ultrapure for ECL

MP Biomedicals Inc.Solon, OH, USA

Triton X-100 Sigma-AldrichMunich, Germany

Trizma base (TRIS) Sigma-AldrichMunich, Germany

trypsin gold, mass spectrometry grade PromegaMadison, USA

trypsin/EDTA LonzaCologne, Germany

Tween 20 Sigma-AldrichMunich, Germany

wortmannin CalbiochemSchwabach, Germany

Xba I (10 U/µl) RocheMannheim, Germany

xylene cyanole Sigma-AldrichMunich, Germany

β-glycerophosphate Sigma-AldrichMunich, Germany

β-mercaptoethanol Sigma-AldrichMunich, Germany

2.2. CONSUMABLES/KITS 23

2.2 Consumables/Kits

Table 2.2: Consumables/Kits

Consumable/Kit Company

0.1 − 10µl, 2 − 100µl, 100 − 1000µl epT.I.P.S. Reloads

EppendorfHamburg, Germany

0.5ml, 1.5ml, 2.0ml safe-lock reactiontubes

EppendorfHamburg, Germany

0.22µm filter unit (sterile) MilliporeSchwalbach, Germany

5ml, 10ml, 25ml, 50ml polystyrene,sterile pipettes (Stripette)

Corning B. V. Life SciencesAmsterdam Netherlands

15 and 50ml polypropylene conical tube BD BiosciencesErembodegem, Belgium

96-well ELISA microplate, PS, flatbottom

Greiner Bio OneFrickenhausen, Germany

Amersham Hyperfilm ECL GE HealthcareMunich, Germany

Biorad Protein Assay BioradMunich, Germany

blotting membrane Protran BA85(nitrocellulose)

WhatmanDassel, Germany

C57Bl/6 mice Charles River LaboratoriesSulzfeld, Germany

cell scraper Corning B. V. Life SciencesAmsterdam, Netherlands

Centri Sep Spin Column Princeton SeparationsAdelphia, USA

Costar 150mm, tissue culture treated Corning B. V. Life SciencesAmsterdam, Netherlands

Costar 6-well plates, tissue culture treated Corning B. V. Life SciencesAmsterdam, Netherlands

2.2. CONSUMABLES/KITS 24

Consumable/Kit Company

dounce homogenizer (2ml) SartoriusGttingen, Germany

drigalski applicator NeolabHeidelberg, Germany

Flp-In System InvitrogenKarlsruhe, Germany

gel blotting paper GB005 WhatmanDassel, Germany

Immobilon Western HRP SubstrateLuminol Reagent

MilliporeSchwalbach, Germany

inoculation loop SarstedtNmbrecht, Germany

Plasmid Maxi Kit QiagenHilden, Germany

plunger for dounce homogenizer, size S SartoriusGttingen, Germany

Precision Plus Protein Dual Color Standard(1610374)

BioradMunich, Germany

Qiaprep Spin Miniprep Kit QiagenHilden, Germany

QIAquick PCR Purification Kit100bp − 10kb

QiagenHilden, Germany

Quikchange II XL mutagenesis kit Stratagene/AgilentWaldbronn, Germany

quickload 1kb DNA ladder New England BioLabsBeverly, USA

tissue culture dishes 10cm and 15cm TPPTrasadingen, Switzerland

tissue culture plates 6-well, 12-well TPPTrasadingen, Switzerland

2.3. LABORATORY DEVICES 25

2.3 Laboratory devices

Table 2.3: Laboratory devices

Laboratory device Company

agarose gel chamber bsb11 blueSchauenburg, Germany

Agfa Curix 60 Developermachine

Agfa Healthcare GmbHBerlin, Germany

centrifuge Biofuge Fresco HeraeusHanau, Germany

centrifuge Heraeus Pico 17 Thermo Fisher ScientificSchwerte, Germany

centrifuge Hettich Rotanta RPC Andreas Hettich GmbH & Co KGTuttlingen, Germany

electrophoresis chamber forSDS-PAGE

bsb11 blueSchauenburg, Germany

CO2-incubator Binder GmbHTuttlingen, Germany

ELISA reader Model 680 BioradMunich, Germany

gradient gel maker Pharmacia LKB Biotechnology ABBromma, Sweden

Hera Safe Biological Safety Hood Thermo Fisher ScientificSchwerte, Germany

HP scanjet 4670 Hewlett Packard GmbHBerlin, Germany

incubator for bacteria HeraeusHanau, Germany

laboratoy scale SartoriusGttingen, Germany

magnetic stirrer IKA Labortechnik,Staufen, Germany

Mastercycler Gradient EppendorfHamburg, Germany

2.3. LABORATORY DEVICES 26

Laboratory device Company

microscope Axiovert 40 ZeissOberkochen, Germany

Neubauer chamber for cell counting Marienfeld GmbH & Co. KGLauda-Konigshofen, Germany

power supply consort E 802 Consort nvTurnhout, Belgium

Powershot camera A710IS CanonKrefeld, Germany

precision scale ALJI60-4NM Gottl. Kern & Sohn GmbHBalingen, Germany

Printer Selphy CP 510 CanonKrefeld, Germany

rotator wheel Grunewald GmbH & Co KGPSI MedizintechnikLaudenbach, Germany

Scepter Cell Counter MilliporeSchwalbach, Germany

semi-dry western blotting chambers HoelzelWorth/Horlkofen, Germany

Systec DX-65 autoclave SystecWettenberg, Germany

table shaker Hecht AssistentSondheim, Germany

ThermoStat Plus Hot/Cold Incubator EppendorfHamburg, Germany

UV-Transilluminator 254nm LTF Labortechnik, GmbH & Co KGWasserburg, Germany

Vortex Genie 2 Scientific industriesNew York, USA

waterbath MemmertSchwabach, Germany

2.4. SOFTWARE 27

2.4 Software

Table 2.4: Software

Software Company

JMP 10.0.0 SAS Institute Inc.NC, USA

Gelscan Professional V5.1 BioSciTec GmbHFrankfurt, Germany

2.5 Antibodies

2.5.1 Primary antibodies

Table 2.5: Primary antibodies

Primary antibody Dilution Company

p-Thr308 Akt/PKB 1:500 − 1:1000 Cell Signaling TechnologyFrankfurt, Germany

p-Ser473 Akt/PKB 1:1000 Cell Signaling TechnologyFrankfurt, Germany

p-Ser9/p-Ser21 GSK3α/β 1:1000 Cell Signaling TechnologyFrankfurt, Germany

p-Thr202/p-Tyr204 ERK 1:1000 Cell Signaling TechnologyFrankfurt, Germany

p-Ser157 VASP 1:500 in 5 % milkin TBS-T

AbcamCambridge, UK

p-Ser573 IRS2 1:2000 − 1:5000 University of DundeeDundee, UK

Akt/PKB 1:1000 BDHeidelberg Germany

GST 1:20000 GE HealthcareMunich, Germany

IRS2 (polyclonal) 1:1000 − 1:5000 Millipore,Schwalbach, Germany

IRS2 (monoclonal) 1:500 − 1:5000 Millipore,Schwalbach, Germany

http://www.bd.com/de

2.5. ANTIBODIES 28

Primary antibody Dilution Company

IRS1 1:1000 Millipore,Schwalbach, Germany

PKA substrate antibody 1:1000 Cell Signaling TechnologyFrankfurt, Germany

GSK3β 1:1000 Cell Signaling TechnologyFrankfurt, Germany

GFP 1:1000 − 1:5000 Santa CruzSanta Cruz, CA, USA

β-actin 1:1000 Cell Signaling TechnologyFrankfurt, Germany

insulin receptor (β-chain) 1:1000 Santa CruzSanta Cruz, CA, USA

ERK 1:1000 Cell Signaling TechnologyFrankfurt, Germany

myc 1:500 Rainer Lammers, UKTTubingen, Germany

14-3-3ζ (C-17) 1:1000 Santa CruzSanta Cruz, CA, USA

pan 14-3-3 (K-19) 1:1000 Santa CruzSanta Cruz, CA, USA

2.5.2 Secondary antibodies

Table 2.6: Secondary antibodies

Secondary antibody Dilution Company

goat anti-rabbit-HRP 1:3000 − 1:5000 Santa Cruz BiotechnologySanta Cruz, CA, USA

goat anti-mouse-HRP 1:1000 − 1:3000 Santa Cruz BiotechnologySanta Cruz, CA, USA

donkey anti-goat-HRP 1:3000 − 1:5000 Santa Cruz BiotechnologySanta Cruz, CA, USA

rabbit anti-sheep-HRP 1:3000 − 1:5000 Thermo Fisher ScientificBonn, Germany

2.6. TISSUE CULTURE 29

2.5.3 Generation of a phospho-specific antibody at theUniversity of Dundee

Sequence of the peptide used to generate a phospho-specific antibody againstphosphorylated residue Ser573 of mouse IRS2:

CLRKRTYS*LTTPAR (residues 567 − 579, * indicates phosphorylated serine)

The phosphopeptide was conjugated to keyhole limpet haemocyanin (KLH)and BSA using m-maleimidobenzoyl-N-hydroxysuccinimide ester (MBS) accord-ing to the method of Harlow and Lane [48]. Five aliquots containing 0.5mg ofeach peptide conjugate were sent to the Scottish Blood Transfusion Service,Penicuik, Scotland, where they were injected into sheep at monthly intervals.Serum was collected from the sheep at monthly intervals and purified by affinitychromatography with the appropriate phospho-peptide.

2.6 Tissue culture

2.6.1 Cell lines used in this work

Fao and HEK293 cells were purchased from The European Collection of CellCultures (Salisbury, UK) and Flp-In HEK293 cells from Invitrogen (Karlsruhe,Germany). Fao cells were cultivated on tissue culture plates from Corning andHEK293, Flp-In HEK293 and primary hepatocytes were cultivated on tissueculture plates from TPP.

2.6.2 Media and conditions of cultivation

All cell lines were cultivated in an incubator with 5 % CO2-atmosphere and 95 %humidity at 37◦C. HEK293 and Flp-In HEK293 cells were maintained in thefollowing growth medium: DMEM high glucose (4.5g/l), supplemented with10 % FCS, 100 U/ml penicillin, 100µg/ml streptomycin and 2mM glutamine.Fao cells were cultivated in RPMI 1640 containing 10 % FCS, 100 U/ml peni-cillin and 100µg/ml streptomycin. Regular change of growth medium tookplace every 2 − 4 days, depending on the cell line. For stimulation experi-ments cells were starved in their respective medium without any supplementsfor at least three hours. Primary hepatocytes were cultured in hepatocyte main-tenance medium with supplements (albumin, 50nM dexamethason, 5mg/l in-sulin, 2.75mg/l transferrin, 3.35µg/l selen, 10µg/l hepatocyte growth factor,20µg/l EGF). Hepatocytes were starved for serum overnight before stimulationexperiments.

For long term propagation cells were maintained in 15 cm dishes with 20mlof growth medium. If not otherwise indicated, cells were seeded in 6-well platesin 2ml of growth medium or on 10cm plates with 10ml of growth medium.

For long-term storage cells were resuspended in 1ml 10 % DMSO in FCS,transferred into cyrotubes and stored at either −140◦C or in liquid nitrogen.


2.6.3 Cell splitting and seeding

When cells reached confluency they were split and seeded onto new tissue culturedishes. This required detachment of the cells from the tissue culture plates. Cellswere washed once with PBS and incubated with trypsin/EDTA for 5 − 10minat 37◦C to detach cells. Trypsin activity was blocked by adding growth medium,cells were pelleted at 900rpm for 4min at room temperature and medium wasaspirated. The cell pellet was resuspended in fresh growth medium and appro-priate dilutions onto fresh plates were prepared.

For transfection purposes and stimulation experiments cells were countedbefore seeding to guarantee equal cell numbers in each well. After centrifugationthe cell pellet was resuspended in 1ml growth medium, filled up to 10ml withgrowth medium and cells were counted using the Scepter Cell Counter fromMillipore.

2.6.4 Transient transfection of HEK293 cells

Plasmid DNA was introduced into the cells by using polyethylenimine (PEI).A PEI stock was prepared as follows: PEI was dissolved in 25mM HEPES,the solution was sterile filtered (0.22µm filter unit) and 1mg/ml aliquots werestored at -80◦C. The day before transfection 0.4×106 cells were seeded per wellof a 6-well plate.

The transfection mix was prepared in medium without any supplements.2µg DNA and 8µg PEI per well were mixed with 1ml medium, the solution wasslightly mixed and pipetted drop-wise onto the cells in addition to the 2ml ofgrowth medium. In cases of co-transfection of two plasmids, 1µg DNA of eachplasmid was used respectively. After 45h cells were starved for serum for atleast 3h before stimulation experiments were conducted.

2.6.5 Stable transfection of Flp-In HEK293 cells

For stable expression of proteins the Flp-In System from Invitrogen was used.The DNA of the desired protein was cloned into the pcDNA5/FRT/TO vector.Flp-In HEK293 cells, which possess a Flp Recombination Target (FRT) site intheir genome were transfected with the vector containing the desired DNA andin addition with the vector pOG44, encoding for the Flp recombinase, arrangingthe homologous recombination event between the FRT sites (one FRT site inthe host cell line, one in the pcDNA5/FRT/TO vector). This system allowedgeneration of stable cell lines with equal expression levels, since the vector DNAwas always inserted at the same site of the genome, allowing reproducibility. Toensure sufficient levels of Flp recombinase, the pOG44:pcDNA5/FRT/TO ratiowas 9:1. The day before transfection 0.4×106 Flp-In HEK293 cells were seededin a well of a 6-well plate. On the day of transfection the growth medium wasexchanged against medium without antibiotics. For transfection the calciumphosphate method was used. For one well of a 6-well plate the transfection mixwas prepared as shown in Table 2.7 and 2.8.


Table 2.7: Components of stable transfection mix

1µl pcDNA5/FRT or pcDNA5/FRT/TO containingDNA for desired protein (c = 1µg/µl)

9µl pOG44 (c = 1µg/µl)

90µl 0.25 M CaCl2

100µl 2×BBS

Table 2.8: BBS solution (pH 6.96)

50mM BES

280mM NaCl

1.5mM Na2HPO4

The solution was prepared as 2× BBS and stored at -20◦C. After adding theDNA to the calcium chloride solution, the reaction tubes were slightly vortexed.Addition of 2× BBS followed flipping of the reaction tube for 20 seconds andfurther incubation for 20min at room temperature, before the transfection mixwas pipetted drop-wise onto the cells. 24h after transfection the cells fromone well were splitted in equal amounts onto a 12-well plate. One day laterthe antibiotic hygromycin B was added to the medium at a concentration of50µg/ml to select for positive clones. After several days expanding colonieswere picked and propagated. Cells were checked for expression of the desiredprotein by western blot. For experiments cells were seeded at a density of 106

cells/well in a 6-well plate for two days and starved for serum at least 3h.

2.7. PROTEIN BIOCHEMISTRY 32

2.7 Protein Biochemistry

2.7.1 Cell lysis

Cells were washed once with ice cold PBS, followed by adding 175µl lysis bufferwith phosphatase inhibitors for one well of a 6-well plate, 400µl for a 10cm dishor 900µl for a 15cm dish. HEK293 and Flp-In HEK293 cells were incubatedwith lysis buffer on a shaker at 4◦C for 15min, whereas Fao cells were incubatedfor 20−30min. Cells were scraped off the well on ice by using a cell scraper andtransferred into 1.5ml reaction tubes. Lysates were cleared by centrifugationwith 13000rpm for 5min at 4◦C. The supernatant was transferred into freshtubes and stored at −80◦C. If not otherwise indicated the following lysis bufferand phosphatase inhibitors were used:

Table 2.9: Lysis buffer

50mM HEPES

150mM NaCl

1.5mM MgCl2

1mM EGTA

10 % glycerol

1 % Triton X-100

100mM NaF

10mM Na4P2O7

Lysis buffer was freshly prepared with phosphatase inhibitors. The phos-phatase inhibitors were stored as 10× solution at −20◦C and diluted 1:10 inlysis buffer.

Table 2.10: Phosphatase Inhibitors

10 mM NaF

5 mM sodium pyrophosphate

10 mM sodium-ortho-vanadate

10 mM β-glycerophosphate

Phosphatase inhibitors were diluted in bidest water, stored at −20◦C andthawed shortly before use.

2.7.2 Protein determination and sample preparation forSDS-PAGE

The protein determination according to Bradford is based on the dye CoomassieBrilliant Blue G-250 that can form a complex with proteins in acidic solutions.Upon complex formation the dye is stabilized in its unprotonated form and theabsorbance maximum is shifted from 470nm to 595nm. This shift in absorption


can be measured with a spectrophotometer and comparison with a standardcurve allows calculation of the relative protein concentration in a sample. Bovineserum albumin (BSA) standards and samples were diluted in HPLC-grade H2O.

Table 2.11: BSA concentrations used for standard curve

Standard BSA [µg/µl]

1 0

2 0.05

3 0.1

4 0.2

5 0.3

6 0.4

7 0.5

Samples and standards were measured in triplicate. In case of cell lysatessamples were diluted 1:20 and lysates from mice liver were diluted 1:100 inHPLC-grade H2O. 40µl of samples and standards were prepared and 3×10µlwere pipetted in a 96-well microtiter plate. 200µl of 1:5 in water diluted andfiltered dye reagent was added and absorbance was measured at 595 nm usinga Biorad Microplate Reader, Model 680.

After protein determination varying amounts of protein were mixed with5×Lammli buffer and denaturated at 95◦C for 5min before loading onto SDSgels. The exact amount of protein used in the different experiments is indicatedin the Figure legends.

Table 2.12: 5×Lammli

1M Tris-HCl, pH 6.8

50 % glycerol

10 % SDS

5.5 % β-mercaptoethanol

1 % bromphenol blue


2.7.3 SDS-PAGE (sodium dodecyl sulfate polyacrylamidegel electrophoresis)

By applying a negative charge to proteins with sodium dodecyl sulfate, proteinscan be separated in an electric field according to their molecular weight. Inaddition, the density of the polyacrylamide gel can be varied and influencesthe migration behaviour of the protein as well. Proteins with a high molecularweight migrate slower, whereas proteins with a low molecular weight migratefaster through the gel.

Two glass plates were separated by a spacer of 1mm thickness and sealedwith a rubber band. The separating gel was mixed and poured between theglass plates, filled up with water to avoid drying-out and left to polymerize for20min at room temperature. After removing the water, the stacking gel wasprepared and poured on top of the separation gel and a comb with an appropri-ate number of pockets was plugged into the stacking gel. After polymerizationof the stacking gel, the comb was removed and the pockets were rinsed withwater. The rubber band was removed and the glass plates with the polymer-ized separation and stacking gel in between were placed in an electrophoresisapparatus. Samples and a molecular weight marker were loaded and gels wererun overnight with 50V. The next day the voltage was increased to 200V untilthe separation process was completed.

Table 2.13: Electrophoresis buffer

25mM Tris

200mM glycine

0.1 % SDS

Table 2.14: Stacking gel and separating gel buffer

Separating gel buffer(pH 8.8)

Stacking gel buffer(pH 6.8)

Component

1.5M 0.5M Tris

2 % 2 % SDS

Table 2.15: Pipetting scheme for separation and stacking gels

7.5 % Separating gel 5 % Stacking gel Component

20ml 7.05ml H2O

10ml 2.55ml separating gel buffer

10ml 1.35ml acrylamide 30 (37.5 : 1)

66µl 15µl TEMED

270µl 112.5µl 10 % APS


For a 5− 15 % gradient gel, 5 % separation gel and 15 % separation gel weremixed in a gradient gel maker simultaneously while pouring between the glassplates, therefore the 15 % “heavy” gel forms the bottom part of the gel, whilethe 5 % “light” gel forms the upper part of the gel. As stacking gel, the recipefor the 5 % stacking gel was used.

Table 2.16: Pipetting scheme for gradient gel

5 % gel 15 % gel Component

10ml 2.2ml H2O

5ml 5ml separating gel buffer

3.3ml 9ml acrylamide 30 (37.5 : 1)

33µl 33µl TEMED

135µl 135µl 10 % APS

- 2ml glycerol

2.7.4 Western blotting

As with SDS-PAGE, the characteristic that proteins migrate in an electric fieldis used by the western blotting technique. The polyacrylamide gel is placedbetween wet filter papers and an electric field is applied. The still negativelycharged proteins are transferred onto a nitrocellulose or polyvinylidene fluoridemembrane.

In this thesis the semi-dry western blotting technique was used. Gel blottingpaper and a nitrocellulose membrane with a size of 14×16cm were soaked inblotting buffer. The blot was assembled carefully to avoid enclosure of airbubbles and proteins were transferred for 2h with 0.8 mA/cm2. The membranewas reversibly stained with Ponceau S (0.1 % Ponceau S in 5 % acetic acid) tocontrol the quality of the protein transfer and to mark the bands of the molecularweight marker.

Table 2.17: 10× western blot buffer

480mM Tris

390mM glycine

0.4 % SDS

The blot buffer was set up as 10× solution and diluted to a 1× solution withwater and 20 % methanol.


2.7.5 Immunodetection

To block unspecific binding sites membranes were incubated 3×15min in NET-Gat room temperature. Blocking in 5 % milk in TBS-T for 1h at room tempera-ture was necessary for detection of GST.

Table 2.18: NET-G (pH 7.4)

10× NET-G 1× NET-G

500mM Tris 50mM Tris

50mM EDTA 5mM ETDA

1.5 NaCl 150mM NaCl

2.5 % gelatine 0.25 % gelatine

0.5 % Triton X-100 0.05 % Triton X-100

Mebranes were incubated overnight at 4◦C with primary antibody dilutedin NET-G. The next day membranes were washed 3 × 15min with NET-G andincubated with the appropiate secondary antibody diluted in NET-G for 45minat room temperature followed by three final washes with NET-G for 15min.Visualization of proteins was carried out using the enhanced chemiluminescencesystem (ECL). For this, solution A and B were mixed in equal volumes andmembranes were incubated in the mixture for approximately 1min. HyperfilmECL films were exposed for various time points and developed using an AgfaCurix 60 machine.

Table 2.19: Solution A for ECL

0.1M Tris-HCl ultrapure, pH 9.35

0.44mM p-iodphenol

0.43mM luminol

Table 2.20: Solution B for ECL

0.1M Tris-HCl ultrapure, pH 9.35

0.0075 % H2O2


2.7.6 Overlay assay

The overlay assay, sometimes referred to as far western blotting technique can beused to visualize protein-protein interactions. After separation of the proteinson SDS-PAGE and transfer onto nitrocellulose membranes, membranes wereblocked for 1h at room temperature in 5 % milk in TBS-T. Incubation with14-3-3ζ-HRP diluted 1:1000 in 5 % milk in TBS-T was carried out overnight at4◦C. The following day the membrane was washed 5 × 5min with TBS-T andinteraction of 14-3-3 with proteins on the membrane was visualized using ECL.

Table 2.21: TBS-T buffer

25mM Tris-HCl, pH 7.4

150mM NaCl

0.1 % Tween 20

2.7.7 Stripping

After detection of phospho-antibody signals on the membrane or performingoverlay assay, the corresponding proteins had to be detected to check for ex-pression and equal loading. Therefore, membranes were shortly washed withNET-G to remove residual ECL solution. The antibodies were removed by in-cubating the membranes in stripping buffer in a water bath at 56◦C for 30min.Afterwards, the membrane was blocked with NET-G and immunodetection withprotein antibodies took place.

Table 2.22: Stripping buffer (pH 6.8)

66mM Tris

0.5 % β-mercaptoethanol

2 % SDS

2.7.8 Immunoprecipitation

Protein lysates were incubated with the appropriate antibody and protein A orG sepharose (depending on the antibody) for 3−4 hours at 4◦C in reaction tubeson an end-over-end wheel. To ensure homogenous incubation of the samplesHNTG buffer was used to fill up the reaction tubes to at least 500µl of totalvolume. The beads were washed 2 − 3 times with 200µl HNTG buffer withcentrifugation steps in between at 4◦C with 5000rpm for 1min. After aspirationof the washing buffer after the last wash, beads were denaturated in 20µl of5× Lammli buffer at 95◦C for 5min before loading onto SDS gels.


Table 2.23: HNTG buffer (pH 7.5)

20mM HEPES

10mM NaF

150mM NaCl

0.1 % Triton X-100

10 % glycerol

2.7.9 GFP pulldown

All centrifugation steps were carried out at 4◦C for 5000rpm for 1min. Lysateswere precleared with 50 % slurry of protein A sepharose for 30min at 4◦C andafter centrifugation the supernatant was transferred into fresh tubes containing50 % slurry of a mixture of protein A sepharose and GFP binder. Up to 500µgprotein was pulled down with 2.5µl GFP binder, 5-10mg of protein was pulleddown with 15µl GFP binder without protein A sepharose. Samples were mixedfor 2h at 4◦C, followed by washing twice with 500µl high salt buffer and twicewith 500µl no salt buffer. 20µl 5× Lammli was added to the beads, sampleswere denaturated for 5min at 95◦C and separated by SDS-PAGE.

Table 2.24: High salt washing buffer


150mM NaCl

Table 2.25: No salt washing buffer


1mM EGTA

0.1 β-mercaptoethanol

2.7.10 GST pulldown

Varying amounts of protein were used as indicated in the Figure legends. Proteinlysates were incubated with 2µg of GST-14-3-3 protein and 500µl of HNTGbuffer to ensure adequate mixing for 2h on a rotating wheel at 4◦C. This wasfollowed by 3 times washing with 200µl HNTG buffer with centrifugation stepsin between with 5000rpm for 1min at 4◦C. 20µl 5× Lammli was added after thelast wash and samples were denaturated at 95◦C for 5min before SDS-PAGE.As a control for efficient and comparable pulldown of GST fusion protein, themembrane was incubated with GST antibody.

2.8. MOLECULAR BIOLOGY 39

2.8 Molecular Biology

2.8.1 Plasmid generation at the University of Dundee

The following expressions plasmids were generated at the University of Dundee:

• pcDNA5/FRT/TO-GFP-IRS2

• pcDNA5/FRT/TO-GFP-IRS2-1-R300

• pcDNA5/FRT/TO-GFP-IRS2-1-T600

• pcDNA5/FRT/TO-GFP-IRS2-S301-E1321

• pcDNA5/FRT/TO-GFP-IRS2-F601-E1321

• pcDNA5/FRT/TO-GFP-IRS2-S303A

• pcDNA5/FRT/TO-GFP-IRS2-T401A

• pcDNA5/FRT/TO-GFP-IRS2-T517A



Recombinant DNA procedures, restriction digests and ligations were per-formed using standard protocols. All PCR and mutagenesis reactions werecarried out using KOD Hot Start DNA polymerase (Novagen). The coding re-gion for mouse IRS2 (NM 001081212.1) was amplified from a pRK5-IRS2 vector(originally from Morris White, Boston, USA) using primers 5’-gaggatccatggctagcgcgcccctgcctg-3’ and 5’-ctgcggccgctcgagtcactctttcacgactgtggcttcc

ttc-3’, cloned into vector pSC-B (Stratagene) and sequenced. IRS2 was sub-cloned from this plasmid into pcDNA5/FRT/TO-GFP as a BamH1/Not1 in-sert to give vector pcDNA5/FRT/TO-GFP-IRS2 in which IRS2 is tagged withGFP at the N-terminus. This vector was used to generate various mutantsby PCR mutagenesis. All clones were fully sequenced. DNA sequencing wasperformed by The Sequencing Service, College of Life Sciences, University ofDundee (www.dnaseq.co.uk).

Truncated IRS2 versions coding for IRS2 residues 1-R300 and 1-T600 weremade by introducing a stop codon behind position 300 or 600 respectively byPCR mutagenesis. The following mutagenesis primers were used:

Table 2.26: Mutagenesis primers for fragments 1-300 and 1-600

Mutation Primer pair

R300 Stop 5’-gagttccggcctcgctgaaagagtcagtcgtcc-3’

3’-ggacgactgactctttcagcgaggccggaactc-5’

T600 Stop 5’-ctcatgagggccacctgatctggtagttcaggtc-3’

3’-gacctgaactaccagatcaggtggccctcatgag-5’


Constructs coding for residues S301-E1321 and F601-E1321 were made byamplifying the appropriate fragment from the full length pSC-B IRS2 vector,cloning into pSC-B and sequencing. S301-E1321 and F601-E1321 fragmentswere sub-cloned into pcDNA5/FRT/TO-GFP as a BamH1/Not1 or BamH1/BamH1 insert respectively.

For point mutations the following primers were used:

Table 2.27: Mutagenesis primers for single point mutations


S303A 5’-cggcctcgcagcaaggctcagtcgtccgggtcg-3’

3’-cgacccggacgactgagccttgctgcgaggccg-5’

T401A 5’-cttagccgctcgcacgccctgagcgccggctg-3’

3’-cagccggcgctcagggcgtgcgagcggctaag-5’

T517A 5’-gtagccacaggagcaacgcacccgagtcaatagc-3’

3’-gctattgactcgggtgcgttgctcctgtggctac-5’

S556A 5’-ccttaccgtagggtcgcaggggatggggccc-3’

3’-gggccccatcccctgcgaccctacggtaagg-5’

S573A 5’-gaggaagaggacttatgccctaaccacgcctg-3’

3’-caggcgtggttagggcataagtcctcttcctc-5’

In addition to the generated plasmids at the University of Dundee a vectorcoding for the human IRS2 sequence used in this work was a kind gift fromCalum Sutherland (Ninewells Hospital, Dundee, UK).

2.8.2 Additional plasmids used in this thesis

The myc-14-3-3γ expression vector was provided by Prof. Reiner Lammers(University Hospital Tuebingen).

Mutagenesis primers were synthesized at TIB Molbiol (Berlin, Germany).Based on the pcDNA5/FRT/TO-GFP-IRS2 expression vector additional mu-tants were generated during this thesis. For the S556A/S573A double mu-tant pcDNA5/FRT/TO-GFP-IRS2-S573A was used as template. The frag-ment F601-S952 was generated by mutating S952 into a stop codon and thepcDNA5/FRT/TO-GFP-IRS2-F601-E1321 construct served as template.

Table 2.28: Mutagenesis primers for generation of a double and a truncated mutant


S556A 5’-ccttaccgtagggtcgcaggggatggggcc-3’

3’-ggccccatcccctgcgaccctacggtaagg-5’

S952 stop 5’-cttcatccctgggttgaggaaccccagg-3’

3’-cctggggttcctcaacccagggatgaag-5’


For single point mutations expression vector pcDNA5/FRT/TO-GFP-IRS2served as template with the used mutagenesis primers as follows:

Table 2.29: Mutagenesis primers for generation of single point mutations


S1137A 5’-ggacgtcgtcgccacgcttcagagacc-3’

3’-ggtctctgaagcgtggcgacgacgtcc-5’

S1138A 5’-gtcgtcgccacagtgcagagaccttttcc-3’

3’-ggaaaaggtctctgcactgtggcgacg-5’

S1163A 5’-ccagcgccacaatgcggcctctgtg-3’

5’-cacagaggccgcattgtggcgcttg-3’

S1137A/S1138A 5’-gtcgtcgccacgctgcagagaccttttc-3’

5’-gaaaaggtctctgcagcgtggcgacgac-3’

2.8.3 PCR mutagenesis

Mutagenesis was performed with the Quickchange II XL site-directed Mutage-nesis Kit from Stratagene. All components were pipetted in a 0.5ml reactiontube as shown in Table 2.30. All components were kept on ice.

Table 2.30: Pipetting scheme for site-directed mutagenesis

2.5µl 10× reaction buffer

0.5µl template plasmid (c = 10 ng/µl)

3.6µl mutagenesis primer up (c = 1 pmol/µl)

3.6µl mutagenesis primer down (c = 1 pmol/ml)

0.5µl dNTP mix

1.5µl Quicksolution

0.5µl Pfu Ultra Polymerase

12.3µl PCR-grade H2O

Depending on the G/C content of the mutagenesis primers different tempera-tures were chosen in the gradient cycler (58.1◦C, 58.4◦C, 59◦C, 59.8◦C, 60.8◦C,61.8◦C, 63◦C, 64◦C, 65◦C, 65.7◦C or 66.2◦C). The reaction tubes were placedin a gradient cycler and the program depicted in Table 2.31 was carried out.To remove the template DNA a DpnI digest was performed. DpnI only cutsmethylated and hemimethylated, thus parental DNA and leaves only the mu-tated, unmethylated DNA. 20µl of the sample were incubated with 0.5µl DpnI(10 U/µl) for 1h at 37◦C. Afterwards, the residual 5µl were filled up with 15µlHPLC-grade H2O to get an equal volume of 20µl. The DpnI-treated and theDpnI-untreated samples were purified with the QIAquick PCR Purification Kitfrom Qiagen according to the manufacturer’s instructions and purified DNA waseluted with 30µl HPLC-grade H2O.


Table 2.31: Gradient cycler protocol for site-directed mutagenesis

Temperature [◦C ] Duration Repeats

95 1min 1

95 50 sec 18

58.1-66.2 50 sec 18

68 10min 18

68 7min 1

DNA was transformed into DH5α E. coli using the KCM method as describedin subsection 2.8.4.

2.8.4 Transformation

5×KCM solution was stored at −20◦C. 30µl of the purified DNA were mixedwith 20µl 5×KCM and 50µl HPLC-grade H2O in a reaction tube, vortexed andplaced on ice for 5min. 100µl competent DH5α E. coli were added, the reactiontube was vortexed again and incubated on ice for 15min, followed by a tempera-ture shock at 42◦C for 2min in a water bath with subsequent addition of 1ml LBmedium without ampicillin and incubation at 37◦C for 30min. Bacteria werepelleted by centrifugation for 2min with 6500rpm at room temperature. Thesupernatant was removed and the bacterial pellet resuspended in the residualLB medium (∼ 50µl) and plated onto LB agar plates with 0.1mg/ml ampicillin.After overnight incubation at 37◦C colonies were picked for MiniPrep isolationof DNA. In case of a MaxiPrep preparation 1µg DNA was mixed with 20µl 5×KCM and filled up with HPLC-grade H2O to 100µl of total volume.

Table 2.32: KCM solution

5×KCM 1×KCM

2M KCl 0.4M KCl

1M MgCl2 0.25M MgCl2

1M CaCl2 0.25M CaCl2

2.8.5 MiniPrep/MaxiPrep for isolation of DNA

For MiniPrep isolation of DNA several colonies were picked from the LB agarplates. Each colony was mixed with 2ml LB medium containing 0.1mg/mlampicillin. For MaxiPrep isolation a single colony was mixed with 100ml LBmedium containing 0.1mg/ml ampicillin. All cultures were incubated at 37◦Covernight in a bacteria shaker. The next day DNA was isolated by using theQIAprep Spin Miniprep Kit or the Plasmid Maxi Kit from Qiagen accordingthe to manufacturer’s instructions.


2.8.6 Sequencing

Mutagenesis was verified by sequencing MiniPrep DNA to confirm successfulmutagenesis. MaxiPrep preparations were also sequenced to ensure that thedesired mutation was still present. The sequencing primers listed in Table 2.33were synthesized at TIB Molbiol (Berlin, Germany).

Table 2.33: Sequencing primers

Construct Sequencing primer

pcDNA5/FRT/TO-GFP-IRS2-F601-S952 cctctacccacagagcccaagagc

pcDNA5/FRT/TO-GFP-IRS2-S556A gcgacctgagagccttcagtagccaca

pcDNA5/FRT/TO-GFP-IRS2-S573A ggagaccccgccagccagagatg

pcDNA5/FRT/TO-GFP-IRS2-S1137A gctaagtctcatggatcaggtatctgg



Sequencing Mix was prepared in a 0.5ml reaction tube on ice as shown inTable 2.34.

Table 2.34: Pipetting scheme for sequencing PCR

0.5µl plasmid (c = 1µg/µl)

1µl sequencing primer (c = 100pmol/µl)

2µl sequencing mix

1µl 5×buffer

5.5µl HPLC-grade H2O

The reaction was carried out in a Mastercycler from Eppendorf with thefollowing program:

Table 2.35: Cycler programme for sequencing

Step Temperature [◦C ] Time Repeats

denaturation 96 1min 1

denaturation 96 10 sec 24

annealing 50 10 sec 24

elongation 60 4min 24

cool 4 overnight -

10µl of HPLC-H20 was added to the PCR product and purifed by usingsephadex columns. The Centri-Sep Spin Columns already contained sephadex,which was discarded and replaced by 650µl of 5 % pre-swelled sephadex at room


temperature. Columns were centrifuged at 5000rpm for 4min at room temper-ature. The columns were placed in a 1.5ml reaction tube and 20µl of the PCRproduct were pipetted onto the sephadex column. Centrifugation at 5000rpmfor 4min at room temperature led to elution of the purified PCR product. Se-quencing was carried out by the Genotyping Facility Tubingen (Department ofInternal Medicine IV, University Hospital Tubingen).

2.8.7 Restriction Digest

Restriction digests were routinely performed after MaxiPrep preparation to en-sure amplification of the right construct. 1µg of DNA was incubated with theappropriate restriction enzyme and buffer for 1h at 37◦C.

Table 2.36: Pipetting scheme for restriction digest

1µl DNA (c = 1µg/µl)

0.5µl XhoI

0.5µl BamHI

1µl buffer B

7µl HPLC-grade H2O

After the digest fragments were mixed with gel loading buffer and loaded ona 1 % agarose gel which contained 0.01 % ethidium bromide. For gel preparation,agarose was dissolved in 1× TAE buffer by heating the mixture in a microwave,after cooling down the agarose to approximately 60◦C ethidium bromide wasadded, the solution was poured into an agarose gel chamber and a comb to formpockets for sample loading was added. Samples were run at 80mA and DNAbands were visualized with a UV lamp.

Table 2.37: 50× TAE buffer

1.25M Tris

625mM acetic acid

50mM EDTA

Table 2.38: 10× gel loading buffer

0.1 % bromphenolblue

0.1 % xylene cyanole

60 % glycerol

20 % 50× TAE


2.8.8 Propagation of E. coli

2ml of LB medium were inoculated with E. coli and incubated overnight at37◦C. The next day 150ml LB medium were inoculated with 1ml of the overnightculture and incubated at 37◦C until an optical density of 0.6 at 600nm. Bacteriawere harvested with 4000rpm for 10min at room temperature. The pellet wasresuspended in 15ml TSB for 10min on ice. TSB medium was freshly preparedand DMSO was added in a final concentration of 5 % shortly before the E. coliwere aliquoted and stored at -80◦C.

Table 2.39: TSB medium

50 % LB medium

10mM MgCl2

10mM MgSO4

2.5 % PEG 3350

40 % H2O

2.8.9 Propagation of GST-14-3-3 expression vectors

GST-14-3-3β and GST-14-3-3ε constructs in pGEX vectors were a kind gift fromAngelika Hausser (University of Stuttgart). The GST-14-3-3 plasmids weretransformed into DH5α E. coli and 100ml LB medium containing 0.1mg/mlampicillin was inoculated with one colony, followed by overnight incubation at37◦C. The next day the bacterial culture was diluted 1:10 in fresh LB mediumwith ampicillin and cultured for 1h at 37◦C. Induction of protein expressionwas achieved by adding 1mM IPTG (isopropyl β-D-thiogalactopyranoside) for5h. Bacterial cells were harvested at 4◦C by centrifugation with 5000rpm for15min. The supernatant was aspirated and the pellet resuspended in 10ml PBS.Bacterial cells were disrupted by applying 40 strokes with a probe-type sonica-tor. After adding Triton X-100 to a final concentration of 1 % the suspensionwas centrifuged at 5000rpm for 20min at 4◦C. The supernatant was transferredinto a fresh 15ml tube and prewashed 50 % slurry of glutathione sepharose 4Bwas added. After incubation for 1.5h on a rotating wheel at 4◦C, beads werewashed 3 times with PBS and stored as 50 % slurry at 4◦C for further use.


To determine the amount of GST-14-3-3 protein which was purified from thebacteria a 5 − 15 % gradient SDS gel was prepared. The gel was loaded with astandard curve prepared from a BSA stock solution ranging from 1µg to 10µgof BSA and with different amounts of purified GST-14-3-3. Coomassie stainingvisualized the protein bands. BSA has a molecular weight of approximately66 kDa and the GST fusion proteins run at 50 kDa. A representative gel isshown in Figure 2.1.

250

150

100

75

50

37

25

GST-14-3-3ε GST-14-3-3β BSA

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Figure 2.1: SDS gel stained with Coomassie G-250 to visualize the amount of preparedGST-14-3-3 proteins. On the left side next to the gel the molecular weights in kDa of themarker is shown. Lanes 1-6 show the BSA concentrations in µg used as standard, lanes 7 and12 show the molecular weight marker bands, lanes 8-11 show the different amounts loaded ofGST-14-3-3ε in µl and lanes 13-16 show the different amounts loaded of GST-14-3-3β in µl.

2.9. MASS SPECTROMETRY (DUNDEE) 47

2.9 Mass spectrometry performed at the Uni-versity of Dundee (Scotland, UK)

2.9.1 Sample preparation for mass spectrometry

HEK293 cells were seeded onto 10cm tissue culture plates, transfected with10µg DNA of GFP empty vector or GFP-IRS2 and 30µg polyethylenimine in25mM HEPES/plate. 45h later cells were starved 3h for serum and stimulatedwith 50ng/ml IGF-1 and after preincubation with 1µM PI-103 for 30min. Afterlysis with buffer consisting of 50mM Tris, 1mM EGTA, 1mM EDTA, 1 % TritonX-100, 1mM sodium-ortho-vanadate, 50mM sodium fluoride, 5mM sodium py-rophospate and one Protease Inhibitor Cocktail tablet (complete, Mini, EDTA-free, Roche, West Sussex, UK), the protein concentration was determined usingCoomassie Protein Assay Kit (Pierce, Rockford, USA). 10mg of total proteinwere pulled down using 15µl pure GFP binder and separated on a precast 4-12 %Bis-tris mini gel (Invitrogen, Paisley, UK). The gel was stained overnight us-ing the Colloidal Blue Staining Kit (Invitrogen, Paisley, UK) according to themanufacturer’s instructions. IRS2 bands were cut and transferred into 1.5mlreaction tubes. For washing steps 0.5ml per gel band was used for 10min andall liquid was discarded between the washes if not otherwise indicated. Gelpieces were washed with ultrapure H2O, 50 % methanol/H2O, 0.1M NH4HCO3,methanol/50mM NH4HCO3. Proteins were incubated with 10mM DTT/0.1MNH4HCO3 on a hot block at 65◦C for 45min. After removal of the liquid sampleswere alkylated by adding 50mM iodacetamide/0.1M NH4HCO3 and incubationin the dark for 20min at room temperature. Two times washing with 50mMNH4HCO3 and 50mM NH4HCO3/50 % methanol was followed by shrinking thegel pieces with methanol for 15min. Liquid was removed and samples were driedwith a Speed-Vac. Gel pieces were soaked in 20mM triethylammonium bicar-bonate containing 5µg/ml trypsin and shaken at 30◦C overnight. An equivalentvolume of methanol was added the next day and samples were shaken for 15min,before the supernatant was removed and the samples were dried in a Speed-Vac.Residual peptides were extracted from the gel pieces with 50 % methanol/2.5 %formic acid and the supernatant was combined with the dried first extract anddried completely.

2.9.2 Mass spectrometric analysis of IRS2 tryptic digests

IRS2 tryptic digests were analysed by LC-MS (liquid chromatography massspectrometry) on a LTQ-orbitrap classic mass spectrometer system (ThermoFisher Scientific, Schwerte, Germany) coupled to a Proxeon Easy-LC HPLCsystem. The peptide mixtures were loaded onto a nanoseparations C18 guardcolumn (0.1 × 20mm) equilibrated in 0.1 % formic acid/water at 5µl/min andthen separated on a 0.075×150mm PepMap C18 column, equilibrated in 0.1 %formic acid/water (LC Packings, Amsterdam, Netherlands). Peptides wereeluted with a 100min discontinuous gradient of acetonitrile/0.1 % formic acid ata flow rate of 300nL/min. The column outlet was connected to a ThermoFisherNanospray 1 source fitted with a New Objective FS360 20-10 uncoated emitterand a voltage of 1.3kV was applied to the emitter. The orbitrap was set to anal-yse the survey scans (m/z 350-2000) at 60.000 resolution and top 5 ions in eachduty cycle (minimum ion intensity of 50000cps), were selected for MS/MS in

2.10. MASS SPECTROMETRY (TUBINGEN) 48

the LTQ linear ion trap with multistage activation. Ions were excluded for 30safter 2 occurrences. The raw files were converted into mascot generic files usingRaw2msm (a gift from M. Mann, Martinsried, Germany). The MSMS spectra(mgf file) were searched against SwissProt database using the Mascot searchengine (Matrix Science) run on an in-house server using the following criteria:peptide tolerance = 10 ppm; trypsin as the enzyme; carboxyamidomethylationof cysteine as a fixed modification with oxidation of methionine and phosphory-lation of serine, threonine and tyrosine as a variable modification. Any MS/MSspectra that could be assigned to a phosphopeptide were inspected manuallyusing QualBrowser software. IRS2 tryptic digests were also analysed by LC-MSwith precursor of 79 scanning on a 4000 Q-Trap system (Applied Biosystems,Carlsbad, USA) as described by [206]. Extracted ion chromatograms for alldetected phosphopeptides were generated using Analyst 1.4.1 software.

2.10 Mass spectrometry performed at the Insti-tute for Ophtalmology at the University ofTubingen

2.10.1 Sample preparation for mass spectrometry

1 × 107 HEK293 cells were seeded 24h prior transfection onto 15cm tissue cul-ture plates. 10µg DNA and 80µg PEI were used to transfect HEK293 cells and45h later cells were serum starved for 3h. Cells were left unstimulated, stimu-lated with 20µM forskolin for 30min or after preincubation with 30µM H89 for30min. Cells were lysed with 900µl lysis buffer and precleared with 30µl of a50 % protein A sepharose slurry. GFP and GFP-IRS2 was purified from 5mgtotal protein by using 15µl GFP binder and samples were separated on 4-12 %Bis-tris mini gels. Protein bands were visualized by staining gels as follows: fix-ing of protein bands by washing 3× 10min with fixer/destainer solution (50 %methanol, 6 % acetic acid), staining with 0.4 % Coomassie G-250 in H2O for20-30min and destaining with fixer solution 3× 10min. Bands corresponding toIRS2 were cut and subjected to tryptic proteolysis as described by Gloeckneret al. [55] prior to LC-MS/MS analysis.

2.10.2 Mass spectrometric analysis of IRS2 tryptic digests

Peptides were separated using the UltiMate 3000 nano HPLC system. Theywere eluted from the trap column onto the analytical column (Acclaim PepMapRSLC 75µl x 25cm C18, 2µm, 100A) applying an acetonitrile gradient of 2 %to 35 % of eluent (80 % acetonitrile; 0.08 % formic acid) in 33min. The elutedpeptides were analyzed by mass spectrometry (LTQ Orbitrap Velos, ThermoFisher Scientific, Waltham, USA). Mass spectra were extracted by Mascot Dae-mon without charge state deconvolution and deisotoping and results were an-alyzed using Mascot (Matrix Science version 2.3.0, Boston, USA). Mascot wasset up to search the Mus musculus subset of the SwissProt database (Release2011 12, 533657 entries) assuming trypsin as the digestion enzyme. Mascot wassearched with a fragment ion mass tolerance of 1Da and a parent ion toleranceof 10ppm. Oxidation of methionine, deamidation of asparagines and glutamine

2.11. ANIMAL STUDIES 49

and phosphorylation of serine, threonine and tyrosine were specified in Mascotas variable modifications. Carboxyamidomethylation of cysteine was specifiedas fixed modification. The Mascot result files were loaded into Scaffold soft-ware (version 3.3.1, Proteome Software Inc., Portland, USA) to validate MS/MSbased peptide identifications. Peptide identifications were accepted if they couldbe established at greater than 80 % probability as specified by the Mascot Dae-mon. Further analysis of the MS/MS data with regard to posttranslationalmodifications was performed using the Proteome Discoverer software (version1.3, Thermo Fisher Scientific, San Jose, USA) and Skyline (version 1.2) [106].

2.11 Animal studies

All procedures were approved by the local Animal Care and Use committee.Male C57Bl/6 mice were obtained from Charles River Laboratories and main-tained on a normal 12h dark-light cycle with standard chow.

2.11.1 Isolation of mice liver protein

For the experiments shown in Figure 3.4 mice were fasted overnight and injectedintraperitoneally with 2 IU (international units) insulin for 10min. For the ex-periments shown in Figure 3.5 mice were also fasted overnight, refed ad libitumfor 5h or injected intravenously with 0.75U/kg insulin for 30min before liverwas taken. Liver tissue was homogenized on ice with a dounce homogenizerand 2ml lysis buffer containing phosphatase inhibitors and protease inhibitors(Complete tablet from Roche). After incubation on ice for 30min lysates werecleared by centrifugation for 3× 10min at 4◦C with 13000rpm. Cleared lysateswere transferred to a fresh reaction tube and stored at -80◦C.

Table 2.40: Lysis buffer for protein isolation of mouse liver (pH 7.6)

50mM Tris

150mM NaCl

1 % Triton X-100

Phosphatase and protease inhibitors were added directly before use.

2.11.2 Isolation of primary hepatocytes

14 week old male C57Bl/6 mice were anaesthetized and the portal vein wascatheterized. Liver was perfused with Krebs-Henseleit buffer I (Table 2.41) andsubsequently with Krebs-Henseleit buffer II (Table 2.42) containing 1mg/mlcollagenase II (Biochrom, Berlin, Germany). Isolated hepatocytes were washedwith Krebs-Henseleit buffer II without collagenase and counted with TrypanBlue. 5×105 cells were seeded in hepatocyte maintenance medium with supple-ments on 6-well plates precoated with Collagen R (Serva, Heidelberg, Germany).24h after isolation cells were starved for serum overnight and stimulated as in-dicated in the Results/Figure legends.

2.12. STATISTICAL ANALYSIS 50

Table 2.41: Krebs Henseleit buffer I (pH 7.4), sterile filtered

20 mM HEPES

115 mM NaCl

25 mM NaHCO3

5.9 mM KCl

1.18 mM MgCl2

1.23 mM NaH2PO4

1.195 mM Na2SO4

0.5 mM EGTA

Table 2.42: Krebs Henseleit buffer II (pH 7.2), sterile filtered

20 mM HEPES

115 mM NaCl

25 mM NaHCO3

5.9 mM KCl

1.18 mM MgCl2

1.23 mM NaH2PO4

1.195 mM Na2SO4

2.5 mM CaCl2

2.12 Statistical analysis

If not otherwise indicated, 3 independent biological replicates were made.Western blots were digitalized with a HP Scanjet 4670 to enable quantifi-

cation of signal intensities. Phosphorylation intensities were normalized to thecorresponding protein intensity. For the degradation experiments the IRS2 pro-tein signal was related to β-actin. In co-immunoprecipitation and GST pull-down experiments the protein that was used for immunoprecipitation or GSTrespectively was used to normalize signal intensities.

Data are presented as mean ± SEM, unpaired student’s t-test was used totest for significant differences and p values < 0.05 were considered significant.

Chapter 3

Results

3.1 Identification of signalling pathways that in-creased 14-3-3 binding to IRS2

To identify conditions for increased 14-3-3 interaction with IRS2, HEK293 cellswere transfected transiently with GFP-IRS2 and incubated with different sub-stances, which activate or inhibit major intracellular signal transduction path-ways: IGF-1 leads to activation of the PI 3-kinase pathway; PI-103 is an ATP-competitive compound targeting the catalytic subunit p110 of PI 3-kinase; EGF(epidermal growth factor) leads to activation of the Ras/MAPK signalling path-way; TPA (phorbol-12-myristate-13-acetate), also called PMA, activates PKCisoforms and other kinases; BID 1870 is an inhibitor against the RSK (90 kDaribosomal S6 kinase) family of kinases which are downstream effectors of theRas/MAPK signalling cascade; forskolin, a diterpene derived from the plantColeus forskohlii, acts as an activator of adenylylcylase, thus leading to in-creased intracellular cAMP levels which result in activation of PKA (proteinkinase A); H89 is an inhibitor of PKA, but can also inhibit other kinases suchas MSK1, S6K1 and ROCK-II (153); A23187 is also known as Calmycin and isa calcium ionophore increasing intracellular calcium concentrations; A-769662is an AMPK (AMP-activated protein kinase) activator; Calyculin A is a potentserine/threonine phosphatase inhibitor. After cell lysis, purification of GFP-IRS2, SDS-PAGE and western blot the membrane was incubated with humanrecombinant HRP-labelled 14-3-3ζ, in the latter named overlay assay, to visu-alize direct interaction of IRS2 and 14-3-3. The upper part of Figure 3.1 showsthe 14-3-3 overlay, the lower part the corresponding GFP reblot to ensure equalexpression and loading of GFP-IRS2.

In the serum starved condition a 14-3-3 interaction with IRS2 could be ob-served. A strong increase in the 14-3-3 interaction with IRS2 was observed inthe Calyculin A-treated sample. There was a slight increase observable whencells were treated with IGF-1 or forskolin, but marked decreases were detectablewhen the PI 3-kinase pathway was inhibited by PI-103 or PKA activity was in-hibited by H89. The reblot showed an augmented GFP signal for the CalyculinA-treated sample which would implicate unequal loading, unequal expressionor incomplete stripping of the overlay membrane. From this experiment it wasconcluded that two pathways could trigger 14-3-3 binding to IRS2: IGF-1 with

51

3.2. IGF-1-MEDIATED 14-3-3 INTERACTION WITH IRS2 52

14-3-3-HRP

GFP

IGF-1 (PI 3-kinase pathway activator) - + + - - - - - - - -

- - + - - - - - - - -

- - - + - - - - - - -

- - - - + + - - - - -

- - - - - + - - - - -

- - - - - - + + - - -

- - - - - - - + - - -

- - - - - - - - + - -

- - - - - - - - - + -

- - - - - - - - - - +

PI-103 (PI 3-kinase inhibitor)

EGF (Ras/MAPK pathway activator)

TPA (PKC activator)

BID 1870 (RSK inhibitor)

forskolin (adenylycyclase activator)

H89 (PKA inhibitor)

A23187 (calcium ionophore)

A-769662 (AMPK activator)

Calyculin A (Ser/Thr phosphatase inhibitor)

Figure 3.1: Kinase panel assay. HEK293 cells were transfected transiently with GFP-IRS2,starved for serum 48h later and incubated with the following agents: 50ng/ml IGF-1 for30min, 1µM PI-103 for 30min, 100ng/ml EGF for 15min, 100ng/ml TPA for 30min, 10µMBID 1870 for 30min, 20µM forskolin for 30min, 30µM H89 for 30min, 10µM A23187 for15min, 50µM A-769662 for 60min, 50nM Calyculin A for 10min. GFP-IRS2 was purifiedusing the GFP-Trap R© from 250µg of total protein, separated on 4-12 % Bis-tris mini gel andoverlay assay was performed. Membrane was stripped afterwards and reprobed with GFPantibody to ensure equal expression and loading.

involvement of the PI 3-kinase pathway and elevated cAMP levels with subse-quent PKA activation.

Based on these findings and the subsequent experiments the results are par-titioned in two sections, the first section shows the results of the characterizationof IGF-1-mediated 14-3-3 interaction with IRS2 and the second section showsthe results of the characterization of PKA-mediated interaction of IRS2 and14-3-3.

3.2 IGF-1-mediated 14-3-3 interaction with IRS2

3.2.1 Interaction of 14-3-3 with IRS2 upon IGF-1 or in-sulin stimulation in different cell systems

To verify the results of the kinase panel assay HEK293 cells were transfectedwith GFP-IRS2, stimulated with IGF-1 and PI 3-kinase activity was inhibitedby pretreating cells with PI-103. In addition, Fao rat hepatoma cells as a cellsystem that expresses IRS2 endogenously were tested if interaction of 14-3-3with IRS2 could be observed.

Since HEK293 cells express adequate amounts of IGF-1 receptor but notinsulin receptor [37] IGF-1 was used to stimulate HEK293 cells. The overlayassay showed that upon stimulation with IGF-1 for 30min the 14-3-3 bindingto GFP-IRS2 was increased, whereas it was decreased by treating cells 30min


14-3-3-HRP

GFP

PI-103

a) b)

14-3-3-HRP

IRS2

IGF-1

p-Thr308

Akt/PKB

- + +

- - +

IGF-1 - + -

- - +

insulin

IP: IRS2

lysate

Figure 3.2: 14-3-3 can bind to transiently and endogenously expressed IRS2. a) HEK293 cellswere transfected transiently with GFP-IRS2 and incubated with 50ng/ml IGF-1 for 30 minor with 1µM PI-103 for 30min prior IGF-1 stimulation. GFP-IRS2 was purified from 250µgtotal lysate using GFP-Trap R©, samples were separated on 4-12 % Bis-tris mini gels and overlayassay was performed. GFP reblot is shown as loading control. b) Fao cells were incubated witheither 10nM insulin or 50ng/ml IGF-1 for 30min. IRS2 was immunoprecipitated (IP) from250µg total lysate, separated on 5-15 % gradient SDS gel and overlay assay was performedwith subsequent IRS2 reblot as loading control. Lysates containing 80µg total protein werechecked for p-Thr308 phosphorylation and Akt/PKB reblot is shown as loading control.

prior IGF-1 stimulation with PI-103 as shown in Figure 3.2a. Figure 3.2b showsthat insulin stimulation led to a strong interaction of 14-3-3 with IRS2 andincreased Akt/PKB phosphorylation in Fao cells. Stimulation with IGF-1 didnot result in activation of the PI 3-kinase pathway, shown as missing Akt/PKBphosphorylation, and did not increase 14-3-3 binding. The finding that IGF-1stimulation in Fao cells does not lead to activation of the PI 3-kinase pathwayis similar to previous data from Fritsche et al. [48], probably due to low IGF-1receptor expression levels. Therefore, insulin stimulation was used to investigateIRS2 and 14-3-3 interaction in Fao cells. The data from Figure 3.2 suggest thatinsulin/IGF-1 stimulation induced phosphorylation on the IRS2 molecule onserine or threonine residues within one or more 14-3-3 binding motif(s) withinvolvement of the PI 3-kinase pathway.

3.2.2 Co-immunoprecipitation of endogenous 14-3-3

Since the overlay assay shows binding of 14-3-3 to denaturated IRS2 on a nitro-cellulose membrane after SDS-PAGE, a co-immunoprecipitation experiment wasconducted to confirm interaction of 14-3-3 and IRS2 in situ. HEK293 cells weretransfected transiently with GFP or GFP-IRS2, starved for serum and eitherincubated with IGF-1 alone or after preincubation with wortmannin. Endoge-nously expressed 14-3-3 was immunoprecipitated and IRS2 was detected.

Figure 3.3 shows the results for the co-immunoprecipitation experiment. Im-munoprecipitation of endogenous 14-3-3 revealed interaction with IRS2 alreadyat basal levels, but IGF-1 stimulation increased that amount. Incubation withthe covalent PI 3-kinase inhibitor wortmannin decreased binding of 14-3-3 andIRS2. Of note, the two different PI 3-kinase inhibitors used in this work, namelyPI-103 and wortmannin, led to similar results.


IRS2

14-3-3

IP: 14-3-3

- + + - + +

- - + - - +

IGF-1

wortmannin

lysate GFP

GFP GFP-IRS2

Figure 3.3: Co-immunoprecipitation of 14-3-3 and IRS2. HEK293 cells were transientlytransfected with GFP or GFP-IRS2 and starved for serum 48h after transfection. Cells wereeither left untreated, incubated with 50ng/ml IGF-1 for 30min alone or after preincubationwith 100nM wortmannin for 30min. 400µg protein were used for immunoprecipitation with2µg 14-3-3 C-17 antibody, samples were separated on 5-15 % SDS gel and after western blottingupper membrane was incubated with IRS2 antibody, whereas the membrane in the middlewas incubated with 14-3-3 K-19 antibody. 40µg protein was separated by 7.5 % SDS-PAGEand membrane was incubated with GFP antibody as expression control. Experiment wasperformed once.

3.2.3 14-3-3 interacts with IRS2 from mouse liver

To evaluate the physiological importance of 14-3-3 binding to IRS2 it was testedif the overlay assay could show 14-3-3 binding to IRS2 in mouse liver after anacute insulin stimulus. Mice were fasted overnight, injected intravenously withinsulin and after 10min liver was taken and IRS2 was immunoprecipitated.Figure 3.4 shows the overlay assay and the corresponding densitometric analysisthereof.

- - + +

14-3-3-HRP

IRS2

insulin a)

- + insulin

0.2

0.4

0.6

0.8

14

-3-3

-HR

P/I

RS2

* b)

0

Figure 3.4: Interaction of 14-3-3 with IRS2 upon insulin stimulation in mice liver. a) 14 weekold male C57Bl/6 mice were fasted overnight and injected intravenously with 2 IU insulin.After 10min liver was taken and 500µg of total protein was immunoprecipitated with 2µg IRS2antibody. Separation of proteins was carried out on a 5−15 % gradient gel and membranewas incubated overnight with 14-3-3-HRP. As loading control membrane was stripped andreprobed with IRS2 antibody. Two mice of each group are shown. b) Densitometric analysisof 14-3-3 interaction with IRS2. Overlay signal was normalized for total IRS2 protein content(mean ± SEM; n = 4; * p < 0.05 fasted vs. insulin).

Intravenous application of insulin for 10min led to a markedly increased14-3-3 binding to IRS2. The overlay assay showed a possible interaction of14-3-3 and IRS2 after an acute insulin stimulus and the next question was if theacute insulin-stimulated interaction could be compared to a physiological refedcondition.

For this experiment, male C57Bl/6 mice were fasted overnight, refed for 4hor injected intraperitoneally with insulin for 30min before liver was taken.

Figure 3.5 shows the increased binding of 14-3-3 to IRS2 after the acuteinsulin stimulus in the overlay assay, but also the refeeding condition led to


fasted refed insulin

14-3-3-HRP

IRS2

0

0.1

0.2

0.3

0.4

0.5

fasted refed insulin

14

-3-3

-HR

P/I

RS2

*

*

a)

b)

Figure 3.5: 14-3-3 binds to IRS2 after refeeding and insulin stimulation in mice. a) MaleC57Bl/6 mice were fasted overnight, refed for 5h or injected i.p. with insulin for 30min.1mg of total protein was immunoprecipitated with 2µg IRS2 antibody, samples were loadedtwice on a 5-15 % gradient gel and one membrane was incubated overnight with 14-3-3-HRP,whereas the other membrane was directly incubated with IRS2 antibody. Four mice of eachgroup are shown. b) 14-3-3 interaction with IRS2 was quantified by scanning densitometry ofimmunoblots and normalization for IRS2 protein (mean ± SEM; n = 4; * p < 0.05 fasted vs.refed and insulin stimulation).

an increased binding of 14-3-3 to IRS2 when the overlay signal was normalizedagainst total protein content. In conclusion, an acute insulin stimulus as wellas refeeding could lead to increased interaction between IRS2 and 14-3-3, thusimplicating a physiological importance of this interaction.

3.2.4 Mass spectrometry for identification of phosphory-lated serine/threonine residues on IRS2 after IGF-1stimulation

To identify serine/threonine residues on IRS2 that are involved in the insulin/IGF-1 signalling cascade GFP or GFP-IRS2 encoding mouse IRS2 was ex-pressed transiently in HEK293 cells and stimulated with IGF-1 alone or afterpre-treatment with PI-103. GFP and GFP-IRS2 was purified from cell lysatesand after SDS-PAGE and Coomassie staining, bands corresponding to IRS2were cut and prepared for mass spectrometry according to section 2.9. Thecomplete experiment including data analysis was performed at the Universityof Dundee.

Figure 3.6 shows the Coomassie stained SDS gel where the bands for massspectrometric analysis were cut out from. A vector containing solely GFP wastransfected as a control for evaluation of the pulldown. This would be of majorimportance when the lanes with the different conditions used for stimulation


250

150

100

75

50

37

25 20

- - + +

- - - +

IGF-1

PI-103

GFP GFP-IRS2

Figure 3.6: Coomassie stained SDS gel. HEK293 cells were transfected transiently withexpression vectors coding for GFP or GFP-IRS2. Cells were left untreated or stimulatedwith 50ng/ml IGF-1 for 30min or incubated prior IGF-1 stimulation for 30min with 1µMPI-103. GFP and GFP-IRS2 were purified from 10mg protein by GFP-Trap R© and sampleswere separated on 4-12 % Bis-tris mini gels. The SDS gel was stained with the Colloidal BlueStaining Kit overnight. The first lane shows the molecular weight marker, the numbers onthe left side depict the individual molecular weights in kDa.

would be analyzed for interaction partners. A prominent band between 150 and250kDa could be assigned to IRS2 in every GFP-IRS2 lane, since at the corre-sponding height in the GFP control lane no such prominent band was visible.In this thesis only the bands corresponding to IRS2 were cut and analyzed forphosphorylated residues. Tryptic digests of IRS2 protein bands followed liq-uid chromatography mass spectrometry (LC-MS). The sequence coverage was92 % and a total of 24 phosphorylated serine/threonine residues were identified.According to IRS2 mouse numbering phosphorylations were detected at resi-dues Ser66, Ser303, Ser305, Ser347, Ser385, Ser388, Thr401, Thr517, Ser556, Ser573,Ser675, Ser722, Ser727, Ser728, Ser762, Ser907, Ser968, Ser977, Ser999, Ser1089,Ser1151, Ser1165, Ser1190 and Ser1266 (see Figure A.1 and Figure A.2 for theidentified phosphopeptides). In Figure 3.7 the sections of the IRS2 mouse se-quence are shown that contain the 24 phosphorylated residues (labeled in boldred). Corresponding residues found in human IRS2 and mouse/human IRS1 arealso shown. A complete alignment of mouse and human IRS1 and IRS2 can befound in Figure A.5.


IRS species numbering sequence

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

42

42

24

24

HGHKRFFVLRGPGTGGDEASAAGGSPPQPPRLEYYESEKKWR

HGHKRFFVLRGPGAGGDEATAGGGSAPQPPRLEYYESEKKWR

SMHKRFFVLRAASEAG-----------GPARLEYYENEKKWR

SMHKRFFVLRAASEAG-----------GPARLEYYENEKKWR

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

291

294

255

260

LKELFEFRPRSKSQSSGSSATHPISVPGARRHHHLVNLPPSQ

LKELFEFRPRSKSQSSGSSATHPISVPGARRHHHLVNLPPSQ

MSDE--FRPRSKSQSSS-SCSNPISVPL--RRHHLNNPPPSQ

MSDE--FRPRSKSQSSS-NCSNPISVPL--RRHHLNNPPPSQ

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

333

336

292

297

TGLVRRSRTDSLAATPPAAKC----TSCRVRTASEGDGGAAG

TGLVRRSRTDSLAATPPAAKC----SSCRVRTASEGDGGAAA

VGLTRRSRTESITATSPASMVGGKPGSFRVRASSDGEGT---

VGLTRRSRTESITATSPASMVGGKPGSFRVRASSDGEGT---

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

371

374

331

336

GAGTAGGRPMSVAGSPLSPGPVRAPLSRSHTLSAGCGGRPSK

GAAAAGARPVSVAGSPLSPGPVRAPLSRSHTLSGGCGGRGSK

-----MSRPASVDGSPVSPSTNRTHAHRHR---------G--

-----MSRPASVDGSPVSPSTNRTHAHRHR---------G--

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

497

500

423

428

LDEYGSSPGDLRAFSSHRSNTPESIAETPPARDGS-GGELYG

LDEYGSSPGDLRAFCSHRSNTPESIAETPPARDGGGGGEFYG

SDEYGSSPCDFRS--SFRSVTPDSLGHTPPARGEE---ELSN

SDEYGSSPCDFRS--SFRSVTPDSLGHTPPARGEE---ELSN

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

554

558

502

507

-RVSGDGAQDLDRGLRKRTYSLTTPA---RQRQVPQPSSASL

-RVSGDAAQDLDRGLRKRTYSLTTPA---RQRPVPQPSSASL

PGDEAAGAADLDNRFRKRTHSAGTSPTISHQKTPSQSSVASI

AGDEAASAADLDNRFRKRTHSAGTSPTITHQKTPSQSSVASI

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

667

671

624

628

KSDDYMPMSPTSVSAPKQILQPRLAA----ALPPSGAAVPAP

RSDDYMPMSPASVSAPKQILQPRAAAAAAAAVPSAGPAGPAP

GNGDYMPMSPKSVSAPQQIINPIRRHPQR--VDPNGYMMM--

GSGDYMPMSPKSVSAPQQIINPIRRHPQR--VDPNGYMMM--

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

705

713

662

666

PSGVGRTFPVNGGG-YKASSPAESSPEDSGYMRMWCGSK---

TSAAGRTFPASGGG-YKASSPAESSPEDSGYMRMWCGSK---

-SPSGSCSPDIGGGSSSSS-SISAAPSGSSYGKPWTNGVGGH

-SPSGGCSPDIGGGPSSSSSSSNAVPSGTSYGKLWTNGVGGH

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

743

751

702

707

---------LSMENPDPKLLP-NGDYLNMSPSEAGTAGTPPD

---------LSMEHADGKLLP-NGDYLNVSPSDAVTTGTPPD

HTHALPHAKPPVESGGGKLLPCTGDYMNMSPVGDSNTSSPSE

HSHVLPHPKPPVESSGGKLLPCTGDYMNMSPVGDSNTSSPSD

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

889

897

860

865

GLQ--TL-----------PSMQEYPLPTEPKSPGEYINIDFG

GLP--SL-----------PSMHEYPLPPEPKSPGEYINIDFG

DPKASTLPRVREQQQ----QQQSSLHPPEPKSPGEYVNIEFG

DPKASTLPRAREQQQ----QQQPLLHPPEPKSPGEYVNIEFG

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

960

968

926

931

DSRQRSPLSDYMNLDFSSPKSPKPSTRSGDTVGSMDGLLSPE

DSRQRSPLSDYMNLDFSSPKSPKPGAPSGHPVGSLDGLLSPE

PRE-ETGSEEYMNMDLGPGRRATWQESGGVELGRIGPA-PPG

PREEETGTEEYMKMDLGPGRRAAWQESTGVEMGRLGPA-PPG

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

1079

1090

1028

1034

PPKPEGARV---ASPTSGL-------KRLSLMDQVSGVEAFL

PPKPEAARV---ASPTSGV-------KRLSLMEQVSGVEAFL

PPPSSTASSSASVTPQGATAEQATHSSLLGGPQGPGGMSAFT

ASSSSAASAS-PTGPQ-GAAELAAHSSLLGGPQGPGGMSAFT

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

1111

1122

1070

1074

QVSQPPDPHRGAKVIRADPQGGRRRHSSETFSSTTTVTPVSP

QVSQPPDPHRGAKVIRADPQGGRRRHSSETFSSTTTVTPVSP

RVNLSPNHNQSAKVIRADTQGCRRRHSSETFSA---PTRAGN

RVNLSPNRNQSAKVIRADPQGCRRRHSSETFSSTPSATRVGN

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

1153

1164

1109

1116

SFAHNSKRHNSASVENVSLRKSSEGSSTL--GGGDEPPTSPG

SFAHNPKRHNSASVENVSLRKSSEGGVGVGPGGGDEPPTSPR

TVPFG-------------------AGAAVGGSGGGGGGGSED

TVPFG-------------------AGAAVGGG-GGSSSSSED

mouse IRS-2

human IRS-2

mouse IRS-1

human IRS-1

1264

1281

1205

1214

KNSWSRTRSLGGLLGTVGGSGASGVCGGPGTGALPSASTYAS

KSSWGRTRSLGGLISAVGVGSTGGGCGGPGPGALPPANTYAS

NEGNSPRR------------------------SSEDLSNYAS

GESSSTRR------------------------SSEDLSAYAS

Figure 3.7: Sequence alignments of IRS1 and IRS2. Protein sequences from mouse IRS2(NP 001074681.1), human IRS2 (NP 003740.2), mouse IRS1 (NP 034700.2) and human IRS1(NP 005535.1) were aligned. The sequences were searched for homologue residues of mouseIRS2 that were identified by mass spectrometry and marked in bold red. Shown are only partsof the sequences that contain an identified phosphorylated serine/threonine residue. The term“numbering” indicates the amino acid position of the first amino acid in every line.


Of note, Ser305 and Ser1266 were the only sites that were unique to theIGF-1-treated sample. However, the mass spectrometric approach was suitableto identify phosphorylation sites, but did not allow evaluation of changes inthe phosphorylation levels due to IGF-1 treatment. Interestingly, some resi-dues are unique for IRS2 (Ser66, Thr401, Ser556, Ser727, Ser728, Ser968, Ser977,Ser999, Ser1089, Ser1151, Ser1165 and Ser1266), whereas some are conserved inIRS1 and IRS2 (Ser303, Ser305, Thr347, Ser385, Ser388, Thr517, Ser573, Ser675,Ser722, Ser762, Ser907 and Ser1190).

3.2.5 Generation of IRS2 fragments to identify the insu-lin/IGF-1-dependent 14-3-3 binding region

Having identified at least 24 potential candidate residues to be phosphorylatedby IGF-1, the next step was to choose sensible candidates for further testing. Bycomparing the 14-3-3 binding motifs RSXpS/pTXP and RX(F/Y)XpS/pTXPwith the sequences surrounding the identified phosphorylated residues it becameapparent that potential 14-3-3 binding motifs frequently appeared in an aminoacid area between the PTB (phosphotyrosine binding) and the KRLB (kinaseregulatory loop binding) domain on IRS2 stretching from amino acid position300-600. To prove that region 300-600 was important for 14-3-3 binding and toexclude possible other regions IRS2 fragments spanning different sections of theprotein were generated. A schematic illustration of the generated fragments isdisplayed in Figure 3.8. Two fragments from position 1-300 and 1-600 encom-passed the N-terminal part of the IRS2 molecule including the PH and the PTBdomain. The C-terminal part was also divided into two fragments spanningamino acids 301-1321 and 601-1321, both containing the KRLB domain. The5th construct comprised amino acids 301-600. All fragments were GFP-taggedat the N-terminus to allow purification with the GFP-Trap R©.

PH PTB KRLB N C

31 140 191 294 591 735

1-300 GFP

601-1321 GFP

1-600 GFP

301-1321 GFP

GFP 301-600

Figure 3.8: Schematic illustration of truncated IRS2 constructs to identify the 14-3-3 bindingregion. Shown are the five fragments of IRS2 and full length IRS2 with amino acid numberingaccording to mouse IRS2.


3.2.6 The area between amino acids 301 and 600 on IRS2is important for 14-3-3 binding

HEK293 cells were transfected transiently with GFP-IRS2 full length or withIRS2 fragments and stimulated with IGF-1 or after preincubation with PI-103.While equal amounts of DNA were used to transfect the cells, the protein ex-pression levels differed between IRS2 wild type and truncated IRS2 forms asshown in Figure 3.9.

GFP-IRS2- 1-300

GFP-IRS2-301-600

GFP

GFP-IRS2 GFP-IRS2- 1-600

GFP-IRS2-301-1321

GFP-IRS2-601-1321

250

150

100

75

50

Figure 3.9: Cell lysates of IRS2 wild type and fragments separated by SDS-PAGE. HEK293cells were transiently transfected with GFP-IRS2, GFP-IRS2-1-300, GFP-IRS2-1-600, GFP-IRS2-301-1321, GFP-IRS2-601-1321 or GFP-IRS2-301-600. Cells were either left unstimu-lated, stimulated with 50ng/ml IGF-1 for 30min or with 1µM PI-103 for 30min prior IGF-1stimulation. 20µg of total protein was separated on a 5-15 % SDS gel and after western blot-ting the membrane was incubated with GFP antibody. The arrow indicates a longer exposuretime and the numbers on the left side denote the molecular weight marker (molecular weightsin kDa).

Figure 3.10 shows the results for the transiently transfected fragments thatwere tested for residual 14-3-3 binding. GFP-IRS2 and fragments were purifiedfrom the cell lysate and tested for residual 14-3-3 binding.

GFP-IRS2 GFP-IRS2- 1-300

GFP-IRS2- 1-600

GFP-IRS2-301-1321

GFP-IRS2-601-1321

14-3-3-HRP

GFP

IGF-1

PI-103

-

-

+ +

+ -

-

-

+ +

+ -

-

-

+ +

+ -

-

-

+ +

+ -

-

-

+ +

+ -

-

-

+ +

+ -

GFP-IRS2-301-600

Figure 3.10: Overlay assay of IRS2 wild type and truncated versions of IRS2 to check forresidual 14-3-3 binding. Lysates were prepared as described in Figure 3.9. 250µg of totalprotein was incubated with GFP-Trap R©, samples were separated on 4-12 % Bis-tris mini gelsand overlay assay was performed. Stripping of the membranes followed reprobing with GFPantibody as loading control.

The overlay assay showed regulated binding of 14-3-3 to wild type IRS2, butno interaction between 14-3-3 and the constructs GFP-IRS2-1-300 and GFP-IRS2-601-1321. A strong and almost unregulated interaction was observed with


the construct GFP-IRS2-1-600. On the contrary, the construct GFP-IRS2-301-1321 showed regulated 14-3-3 binding upon IGF-1 stimulation and reducedbinding by PI-103 treatment. These data showed the importance of the region301-600 for 14-3-3 binding. To test if this region alone was capable of bindingto 14-3-3 the fragment 301-600 was tested. This fragment also showed almostunregulated binding to 14-3-3 similar to GFP-IRS2-1-600.

3.2.7 Identification of Ser573 as IGF-1-dependent 14-3-3binding site on IRS2

Having localized the 14-3-3 binding region to amino acids 301-600, the phospho-rylated serine/threonine residues that were identified by mass spectrometry andcorresponded to a 14-3-3 binding motif as shown in Table 3.1 were investigated.

Table 3.1: With mass spectrometry identified residues with surrounding amino acids thatcorrespond to a 14-3-3 binding motif. Shown are the identified residues indicated in bold withpS or pT and the surrounding six amino acids.

Residue Peptide

303 RSKRSKpSQSSGSS

401 PLSRSHpTLSAGCG

517 SSHRSNpTPESIAE

556 RPYRRVpSGDGAQD

573 LRKRTYpSLTTPAR

After single mutation of each serine/threonine to alanine residual 14-3-3binding upon IGF-1 stimulation was tested in transiently transfected HEK293cells by overlay assay. The mutation of Ser573 to alanine blocked the basaland IGF-1-dependent 14-3-3 interaction completely, while the other mutantsshowed similar interaction with 14-3-3 when compared with IRS2 wild type.An exemplary MS/MS spectrum for a phosphorylated peptide corresponding toSer573 is shown in Figure A.3.

- + + - + + - + +

- - + - - + - - +

- + + - + + - + +

- - + - - + - - +

GFP-IRS2 GFP-IRS2-S573A

GFP-IRS2-T401A

GFP-IRS2-T517A

GFP-IRS2-S556A

GFP-IRS2-S303A

14-3-3-HRP

GFP

PI-103

IGF-1

Figure 3.11: Identification of Ser573 as 14-3-3 binding site by overlay assay. HEK293 cellswere transiently transfected with GFP-IRS2, GFP-IRS2-S303A, GFP-IRS2-T401A, GFP-IRS2-T517A, GFP-IRS2-S556A or GFP-IRS2-S573A and stimulated with 50ng/ml IGF-1for 30min or after preincubation with 1µM PI-103 30min prior IGF-1 stimulation. 250µgof total protein were used for GFP pulldown and overlay assay was performed. Strippingfollowed reprobing of the membrane with GFP antibody as loading and expression control.Experiment was performed once.


Ser573 as 14-3-3 binding site was again confirmed in HEK293 cells transientlyexpressing GFP-IRS2-S573A and also Flp-In HEK293 cells stably expressingGFP-IRS2-S573A showed similar results. No difference in the ability of the twoPI 3-kinase inhibitors used (wortmannin or PI-103) to decrease IGF-1-mediated14-3-3 binding to IRS2 was detected.

GFP-IRS2

GFP-IRS2-S573A

a) b)

0.25

0.20

0.15

0.10

0.05

*

GFP-IRS2 GFP-IRS2- S573A

GFP

14-3-3-HRP

wortmannin

14

-3-3

-HR

P/G

FP

d)


GFP

14-3-3-HRP

IGF-1

0

GFP

β-actin

GFP-IRS2

GFP-IRS2-S573A

c)

p-Thr308 Akt/PKB

- + + - + +

- - + - - + wortmannin

IGF-1 - + + - + +

- - + - - +

wortmannin

IGF-1 - + + - + +

- - + - - +

wortmannin

IGF-1 - + + - + +

- - + - - +

Figure 3.12: Confirmation of Ser573 of IRS2 as an IGF-1-dependent 14-3-3 binding site. a)Transiently with GFP-IRS2 or GFP-IRS2-S573A transfected HEK293 cells were stimulatedwith 50ng/ml IGF-1 alone for 30min or after preincubation with 100nM wortmannin for30min. 100µg of total protein was used for GFP pulldown, separated on 5-15 % SDS gels andoverlay assay was performed. Corresponding GFP reblot as expression and loading controlis shown. b) Flp-In HEK293 cells stably expressing GFP-IRS2 or GFP-IRS2-S573A werestimulated as in a) and cell lysis was followed by pulldown of 400µg total protein. Sampleswere divided into two equal volumes and separated on SDS-PAGE. One membrane was usedfor overlay assay, the other one was directly incubated with GFP antibody. c) 100µg proteinfrom lysates in b) were separated on 7.5 % SDS gel and membranes were incubated with GFPantibody as expression control, with p-Thr308 Akt/PKB antibody as stimulation control andβ-actin antibody as loading control. d) Interaction of 14-3-3 with IRS2 from b) was quantifiedscanning densitometry of western blots (mean ± SEM; n = 3; * p < 0.05 IRS2 wild typeIGF-1 vs. IRS2 wild type wortmannin/IGF-1).

Noteworthy, the Flp-In HEK293 cells showed phosphorylation of Thr308

Akt/PKB in the serum starved condition, which may influence basal phosphory-lation of Ser573. Since Ser573 was crucial for insulin/IGF-1-mediated binding of14-3-3 to IRS2, sequence alignments were performed and these are shown inTable 3.2 and Table 3.3.

3.2.8 Characterization of p-Ser573 antibody

To characterize a phosphorylation site the use of a phospho-specific antibody isindispensable. For this reason an antibody was raised against a phosphorylatedpeptide that corresponded to phosphorylated Ser573 of IRS2. Before the anti-body was used to monitor Ser573 phosphorylation, the antibody itself had to becharacterized and tested for its suitability.


Table 3.2: Sequence alignments concerning the surrounding amino acids of Ser573. Sequencealignment of the amino acids adjacent to the 14-3-3 binding site Ser573 of IRS2 in differ-ent species. Shown are IRS2 sequences from Homo sapiens (NP 003740.2), Mus musculus(NP 001074681.1), Rattus norvegicus (NP 001162104.1) and Xenopus laevis (AAH72768.1).

Homo sapiens 571LRKRTYSLTTPAR584

Mus musculus 567LRKRTYSLTTPAR579

Rattus norvegicus 568LRKRTYSLTTPAR580

Xenopus laevis 519LRKRTYSLTKHTN531

Table 3.3: Shown is the amino acid sequence surrounding Ser573 in mouse IRS2 and thesequence surrounding the homologue position Ser522 in mouse/rat IRS1.

mouse IRS2 567LRKRTYSLTTPAR579

mouse IRS1 516FRKRTHSAGTSPT528

Sequence specificity of p-Ser573 antibody

To test the specificity of the antibody Flp-In HEK293 cells stably expressingGFP-IRS2 or GFP-IRS2-S573A were incubated with IGF-1 or TPA alone orwith wortmannin before the IGF-1 stimulation. A representative blot is shownin Figure 3.13.

TPA

GFP-IRS2 GFP-IRS2- S573A

p-Ser573

GFP

- + + - - + + -

- - + - - - + -

- - - + - - - +

wortmannin

IGF-1

Figure 3.13: Specificity test of the p-Ser573 antibody. Flp-In HEK293 cells stably expressingGFP-IRS2 or GFP-IRS2-S573A were stimulated with 50ng/ml IGF-1 for 30min, 0.5µM TPAfor 30min and 100nM wortmannin for 30min. 200µg lysate was separated by SDS-PAGE,membrane was incubated with p-Ser573 antibody (diluted 1:5000 in NET-G) and reprobedwith GFP antibody as loading control. Shown is a representative blot from two independentexperiments.

While the antibody did not detect any phosphorylation signal in the 573 ala-nine mutant, it detected IGF-1 and TPA-stimulated phosphorylation of Ser573

in wild type IRS2. Wortmannin blocked the IGF-1-mediated phosphorylation ofSer573. The antibody recognized different bands, which all corresponded to theIRS2 protein according to the reblot. This recognition of different sub-speciesof the IRS2 protein by immunoblotting is well in line with data from Fritscheet al. [48].


p-Ser573 antibody does recognize its homologue position Ser577 in hu-man IRS2

Sequence alignments showed Ser522 of rat/mouse IRS1 (same numbering in ratand mouse IRS1 sequence) and Ser577 of human IRS2 to be homologue posi-tions of Ser573 of mouse IRS2. To test if the antibody against phosphorylatedSer573 of IRS2 recognizes the homologue positions, HEK293 cells were trans-fected transiently with either rat IRS1 or human IRS2. In addition, the insulinreceptor was co-transfected to enable stimulation of HEK293 cells with insulin.

p-Ser573

IRS2

IRS1

IR

mIRS2 hIRS2 rIRS1

- + - + - + insulin

Figure 3.14: p-Ser573 antibody is specific for IRS2. HEK293 cells were co-transfectedtransiently with mouse IRS2 / IR (insulin receptor), human IRS2 / IR and rat IRS1 / IRand stimulated with 10nM insulin for 30min. Membranes were incubated with p-Ser573

antibody (diluted 1:5000 in NET-G) and reprobed with the corresponding protein antibodies.IR expression was also checked. Experiment was performed once.

Mouse IRS2 was transfected as a positive control. The antibody detectedphosphorylation of mouse IRS2 Ser573 and human IRS2 Ser577, but not IRS1Ser522. The p-Ser573 signal for human IRS2 seemed to be stronger than thecorresponding signal for mouse IRS2, but the IRS2 reblot showed that thehuman IRS2 construct was expressed at a higher level than the mouse IRS2construct. From this result it could be concluded that the antibody is spe-cific for the sequence LRKRTYpS573/577LTTPAR, while the IRS1 sequenceFRKRTHpS522AGTSPT is not recognized.

Determination of antibody dilution to detect phosphorylation of Ser573

on IRS2 in Fao cells

Since the p-Ser573 antibody recognized the fusion protein GFP-IRS2 in HEK293cells, it was further investigated, if the antibody detected Ser573 phosphorylationof endogenous expressed IRS2. Fao cells were starved for serum and incubatedwith insulin or TPA to induce phosphorylation of Ser573. Since the expressionlevels of endogenous and transiently overexpressed proteins differ very much,several dilutions of the p-Ser573 antibody were tested to establish the optimaldilution factor for the antibody to detect IRS2 in Fao cells.

Figure 3.15 shows that a strong signal intensity in every condition was ob-servable on the membrane piece that was incubated with an antibody dilutionof 1:1000, wheras with higher dilutions of the antibody the signal got more dis-tinct. It was decided to use the antibody within a dilution range from 1:2000


1:1000 1:2000 1:5000 1:10000

p-Ser573

IRS2

- - + - - + - - + - - + - + - - + - - + - - + - insulin

TPA

Figure 3.15: p-Ser573 antibody test in Fao cells. Fao cells were starved for serum andincubated with 10nM insulin or 0.5µM TPA for 30min. 35µl cell lysate was separated onSDS-PAGE and the membrane was cut in 4 parts. Membrane pieces were incubated withdifferent antibody dilutions (1:1000, 1:2000, 1:5000, 1:10000). Stripping and reprobing of themembrane with IRS2 antibody was performed as loading control. Experiment was performedonce.

to 1:5000, because these dilutions allowed a clear differentiation to the serumstarved condition and insulin as well as TPA stimulation was observable.

Immunoprecipitation of IRS2 with p-Ser573 antibody

The antibody was used to immunoprecipitate IRS2 species that were specifi-cally phosphorylated on Ser573. Two different cell systems were used for thisapproach, Flp-In HEK293 cells stably expressing either GFP or GFP-IRS2 andFao cells.

- 5 60 - 5 60 IGF-1

GFP

GFP GFP-IRS2

- 5 60 240 insulin

IRS2

a) b)

Figure 3.16: Immunoprecipitation of IRS2 with p-Ser573 antibody. a) Flp-In HEK293 cellswere starved for serum and incubated with 50ng/ml IGF-1 for the indicated time points inmin. 400µg of total protein was used for immunoprecipitation with 1.3µg p-Ser573 antibodyand membrane was incubated with GFP antibody. b) Fao cells were starved for serum andincubated with 10nM insulin for the indicated time points in min. 400µg of total protein wereused for immunoprecipitation with 1.3µg p-Ser573 antibody and membrane was incubated withIRS2 antibody. Experiments were performed once.

p-Ser573 antibody successfully immunoprecipitated IRS2 from Flp-In HEK293and Fao cells. In both approaches there was no difference in the band intensitiesobservable. Of note, in both cell lines did the antibody precipitate IRS2 fromthe serum starved condition. This was not unexpected from the serum starvedsample of Flp-In HEK293 cells that stably expressed GFP-IRS2, since this cellline showed a permanent active Akt/PKB kinase (see Figure 3.12c), that couldhave influenced phosphorylation of Ser573 in the serum starved condition. Inaddition, it seemed that the antibody precipitated two different IRS2 species inthe Flp-In HEK293 cells as can be seen by the two bands in every lane.


In Fao cells the band shift from IRS2 caused by insulin stimulation andsubsequent posttranslational modifications that increased its molecular weightcould be nicely seen, but it was unexpected that the antibody precipitated IRS2from Fao cells deprived of serum. This could indicate a permanent phosphory-lation of Ser573 and the antibody failed to precipitate increased amounts ofphosphorylated Ser573 upon IGF-1 or insulin stimulation. In conclusion, thep-Ser573 antibody was specific for phosphorylated Ser573 of mouse IRS2, rec-ognized Ser577 of human IRS2 and did not recognize its homologue Ser522 ofmouse/rat IRS1.

3.2.9 Time course of Ser573 phosphorylation in Fao cellsupon insulin stimulation

The phosphorylation pattern of Ser573 of endogenous expressed IRS2 was mon-itored in Fao cells and is shown in Figure 3.17. Cells were starved for serumand incubated with insulin for several time points.

p-Ser573

IRS2

p-Thr308

Akt/PKB

- 5 30 60 90 120 240 360 insulin

b)

- 5 30 60 90 120 240 360 insulin

0

0.1

0.2

0.3

0.4

0.5

p-S

er-5

73

/IR

S2

a)

Figure 3.17: Ser573 phosphorylation upon insulin stimulation in Fao cells. Cells were starvedfor serum and incubated with 10nM insulin for the indicated time points in min. a) 100µgprotein was separated on 7.5 % SDS gel, membrane was incubated with p-Ser573 antibodydiluted 1:2000 in NET-G. Phosphorylation of p-Thr308 of Akt/PKB is shown as stimulationcontrol. Corresponding protein reblots are shown. b) Densitometric analysis of Ser573 phos-phorylation. The p-Ser573 signal was normalized for IRS2 protein content (mean ± SEM;n = 3).


Ser573 showed a slight basal phosphorylation which modestly increased after5 and 30 min and peaked at 60, 90 and 120 min of insulin stimulation beforedecreasing again. The time course experiment showed that phosphorylation ofSer573 phosphorylation was a rather late event in insulin signalling.

3.2.10 Inhibition of Akt/PKB leads to a reduced phos-phorylation of Ser573 and reduced binding of 14-3-3

Comparison of the sequence surrounding Ser573 (RKRTYpS573) with the Akt/PKB consensus motif (RXRXXpS/T) revealed concordance. A literature searchfor publications concerning the homologue position of Ser522 in IRS1 broughtup the publication from Giraud et al. [53]. They described Ser522 of IRS1 as aninhibitory serine phosphorylation site and a substrate for Akt/PKB kinase or akinase downstream of Akt/PKB. It was a matter of course to test if Akt/PKBwas involved in phosphorylation of Ser573. Flp-In HEK293 cells stably express-ing GFP-IRS2 were incubated with an Akt/PKB inhibitor (Akti) before IGF-1stimulation and phosphorylation status of Ser573 was investigated. A represen-tative blot is shown in Figure 3.18.

p-Ser573

GFP

p-Thr308

Akt/PKB

p-Ser-9

GSK3β

a)

Akti

b)

0

0.05

0.1

0.15

0.2

- IGF-1 Akti IGF-1/Akti

p-S

er-5

73

/GF

P

* - + - +

- - + +

IGF-1

Figure 3.18: Phosphorylation of Ser573 after inhibition of Akt/PKB. a) Flp-In HEK293cells stably expressing GFP-IRS2 were starved for serum and incubated with 50ng/ml IGF-1for 30min, 1µM Akti for 30min or Akti incubation prior IGF-1 stimulation. 100µg proteinwas separated on 7.5 % SDS gel and membranes were incubated with p-Ser573 antibody (di-luted 1:5000 in NET-G) and reprobed with GFP antibody. To check inhibition of Akt/PKBphosphorylation of Thr308 of Akt/PKB and its substrate Ser9 of GSK3β were detected. Inaddition, corresponding protein reblots are shown. b) Effect of Akt/PKB inhibition on Ser573

phosphorylation was assessed by scanning densitometry of blots and normalization for GFP-IRS2 protein (mean ± SEM, n = 3, * p < 0.05 IGF-1 vs. Akti/IGF-1).

The compound Akti-1/2 reduced the phosphorylation of Thr308 Akt/PKB, ofits substrate Ser9 GSK3β and the IGF-1-stimulated phosphorylation of Ser573.Densitometric analysis revealed that the reduction in Ser573 phosphorylation bypretreating cells with Akti before IGF-1 stimulation was significant comparedto IGF-1 treatment alone. Again, similar to Figure 3.12, Flp-In HEK293 cellsshowed phosphorylation of Thr308 Akt/PKB in the serum starved condition,which may influence phosphorylation of Ser573 as already described in subsec-tion 3.2.7.

To further test whether Akt/PKB inhibition also influences 14-3-3 binding toIRS2, a GFP pulldown and subsequent overlay assay to visualize 14-3-3 bindingto IRS2 was performed. See Figure 3.19 for the result of the experiment. Ofnote, Flp-In HEK293 cells stably expressing GFP-IRS2 were used.


p-Thr308

Akt/PKB

- + - +

- - + +

IGF-1

Akti

14-3-3-HRP

GFP pulldown

lysate

IRS2

Figure 3.19: Akt/PKB kinase is part of IGF-1-mediated 14-3-3 binding to IRS2. Flp-InHEK293 cells stably expressing GFP-IRS2 were starved for serum and incubated for 30minwith 50ng/ml IGF-1, 1µM Akti-1/2 alone for 30min or Akti-1/2 preincubation followed IGF-1 stimulation. GFP pulldown was performed with 200µg of total protein and overlay assayfollowed reprobing with IRS2 antibody. 100µg of total protein was separated on 7.5 % SDSgel and membrane was checked for p-Thr308 phosphorylation and reprobed with Akt/PKBprotein antibody. Experiment was performed once.

IGF-1 stimulation resulted in increased binding of 14-3-3 to IRS2, whichwas prevented by incubation with Akti-1/2 before IGF-1 stimulation. Thesefindings led to the conclusion that Akt/PKB activity was necessary for Ser573

phosphorylation and the IGF-1-stimulated 14-3-3 binding to IRS2.

3.2.11 Positive influence of S573A mutant on Akt/PKBphosphorylation

To investigate if the phosphorylation of Ser573 had an influence on downstreamsignalling of the insulin/IGF-1 signalling cascade, the phosphorylation of Thr308

and Ser473 of Akt/PKB in HEK293 cells transfected transiently with GFP-IRS2or GFP-IRS2-S573A was assessed after incubation with IGF-1 for several timepoints.

Figure 3.20 shows the induction of Thr308 and Ser473 phosphorylation ofAkt/PKB after 5min of IGF-1 stimulation in HEK293 cells expressing wildtype IRS2 and this stimulation was decreased after 120min. In contrast, thisdownregulation was impaired in the S573A mutant expressing cells. This datasuggest that phosphorylation of Ser573 negatively regulated IGF-1 signalling.

3.2.12 Ser573 is not the only IGF-1-dependent 14-3-3 bind-ing site

The results of the overlay assay using the S573A mutant did not indicate a sec-ond site relevant for insulin/IGF-1-dependent 14-3-3 binding to IRS2. However,with a differential 14-3-3 affinity approach using insulin-stimulated HeLa cellsand the yeast 14-3-3 isoforms BMH1 and BMH2 as capture proteins, Dubois etal. identified IRS2 as a 14-3-3 interaction protein and also identified an IRS2derived peptide containing a phosphorylated serine corresponding to Ser556 [35]by mass spectrometry. Our mass spectrometry experiment also identified aphosphorylation site that corresponded to Ser556. As shown in Figure 3.11 theS556A mutant did not show reduced binding of 14-3-3 in the overlay assay. Nev-ertheless, to study the potential involvement of this site together with Ser573 in


p-Thr308

Akt/PKB

p-Ser473

Akt/PKB

GFP

- 5 15 30 60 120 240 - 5 15 30 60 120 240 min IGF-1


a)

b)

- 5 15 30 60 120 240 min IGF-1

0.5

1.0

1.5

2.0

2.5

0.5

1.0

1.5

2.0

2.5

* *

* *

p-T

hr3

08/A

kt

p-S

er4

73/A

kt

0

0

IRS2 wild type IRS2-S573A

- 5 15 30 60 120 240 min IGF-1

Figure 3.20: Ser573 influences phosphorylation of Akt/PKB. a) HEK293 cells transientlyexpressing GFP-IRS2 or GFP-IRS2-S573A were stimulated with 50ng/ml IGF-1 for the in-dicated time points. 40µg of total protein was separated on 7.5 % SDS gels and membraneswere incubated with GFP antibody to ensure equal expression levels and with p-Thr308 andp-Ser473 antibody respectively. Corresponding Akt/PKB reblots are shown. b) Densitomet-ric analysis of Akt/PKB phosphorylation. Black diamonds represent IRS2 wild type, whitesquares IRS2-S573A mutant. Phosphorylation was normalized against total protein and IRS2wild type protein stimulated with IGF-1 for 5min was set as 1 (mean ± SEM; n = 3; * p <0.05 IRS2 120min vs. S573A 120min; IRS2 240min vs. S573A 240min).


14-3-3 binding GST pulldown assays were applied, where cell lysates were mixedwith GST tagged 14-3-3β or 14-3-3ε. Western blot membranes were incubatedwith GST antibody to ensure equal pulldown and with GFP to visualize theamount of protein that interacted with GST-14-3-3. First, the GST pulldownwas applied for the single Ser573 to alanine mutant of IRS2 to confirm Ser573 as14-3-3 binding site.

GFP

GFP

GST

GST

pulldown:

GST-14-3-3β

pulldown:

GST-14-3-3ε

IGF-1

wortmannin

GFP-IRS2-S556A/S573A

- + +

- - +


- + + - + +

- - + - - +

b) a)

Figure 3.21: Ser573 is not the only IGF-1-dependent 14-3-3 binding site. a) and b) GFP-IRS2, GFP-IRS2-S573A and GFP-IRS2-S556A/S573A were expressed transiently in HEK293cells and cells were stimulated with 50ng/ml IGF-1 for 30min or after preincubation with100nM wortmannin for 30min. Lysates were incubated with 2µg GST-14-3-3β or ε for 2h.Samples were separated on 7.5 % SDS gels and membranes were probed with GFP and GSTantibodies.

Unexpectedly, Figure 3.21a shows that the GST pulldown experiments stillresulted in interaction of IRS2-S573A with 14-3-3 in contrast to the results ofthe overlay assays. The interaction was also not reduced using the single S556Amutant (data not shown) and also the double GFP-IRS2 mutant S556A/S573Afailed to disrupt 14-3-3 binding (Figure 3.21b).

Together, these results may either reflect restrictions of the GST pulldownin the context of the protein constructs of IRS2 and 14-3-3 or give a hint foran additional binding site of 14-3-3 in IRS2 that does not influence the bindingdetectable in the overlay assay but the binding visualized in the GST pulldown.

3.3. CAMP-MEDIATED 14-3-3 INTERACTION WITH IRS2 70

3.3 cAMP-mediated 14-3-3 interaction with IRS2

With the kinase panel assay (Figure 3.1) forskolin and H89 were identified asmodulators of 14-3-3 binding to IRS2. The interaction was induced by theadenylylcyclase activator forskolin. Increased cAMP levels activate PKA andcompound H89 inhibits PKA. The second aim of this thesis was to identifythe cAMP/PKA-dependent phosphorylation sites on IRS2 and the biologicaloutcome of this interaction. First, the kinase panel data were reproduced toverify the cAMP-dependence of 14-3-3 binding to IRS2. Beside forskolin the cell-permeable cAMP analogue CPT-cAMP (8-(4-chlorophenylthio) adenosine 3’,5’-cyclic monophosphate) was used to trigger 14-3-3 binding to IRS2. HEK293 cellstransiently expressing GFP-IRS2 were incubated with forskolin or CPT-cAMPalone or after preincubation with H89.

14-3-3-HRP

IRS2

forskolin

H89

b) a)

H89

CPT-cAMP

14-3-3-HRP

IRS2

- + + - + +

- - + - - +

Figure 3.22: 14-3-3 interacts with IRS2 upon elevated cAMP levels. a) and b) HEK293 cellswere transiently transfected with GFP-IRS2, starved for serum and incubated with 20µMforskolin or 100µM CPT-cAMP for 30min or with 30µM H89 for 30min prior forskolin/CPT-cAMP stimulation. a) 200µg total protein were pulled down and sample was divided in twoequal volumes and separated by SDS-PAGE. One membrane was used for overlay assay, theother one was directly incubated with IRS2 antibody as expression and loading control. b)200µg of total protein were pulled down and after overlay assay membrane was stripped andreprobed with IRS2 antibody as expression and loading control.

Forskolin as well as CPT-cAMP stimulation led to increased binding of 14-3-3to IRS2 and this interaction was completely abolished when cells were pretreatedwith compound H89. H89 is described as PKA inhibitor, although it inhibitsother kinases like MSK1 (mitogen- and stress-activated protein kinase 1), S6K1or ROCK-II (Rho-associated protein kinase II) with a similar potency [30]. Thepotential additional inhibition of other kinases was of minor relevance in the per-formed experiments due to the fact that they are not cAMP-dependent. Sincephosphorylation of serine/threonine residues is a prerequisite for 14-3-3 bindingin most cases, further experiments were conducted to confirm that the observedeffect by overlay assay was mediated by phosphorylation of IRS2 on PKA mo-tifs. For this, an experimental tool to monitor PKA-dependent phosphorylationof IRS2 was searched.


3.3.1 The PKA substrate antibody as a convenient toolfor the identification of PKA phosphorylation siteson IRS2

The PKA substrate antibody from Cell Signalling Technology R© provided thepossibility to monitor PKA-dependent phosphorylation of IRS2. The antibodyis advertised to recognize PKA phosphorylation sites and was tested for itssuitability before the actual experiments.

The PKA substrate antibody

A peptide containing an arginine at position -3 relative to a phosphorylatedserine or threonine was used to generate the PKA substrate antibody. It rec-ognizes the following motifs: RXXpS or RXXpT (X denotes any amino acidand pS or pT indicates the phosphorylated serine or threonine). This enablesthe antibody to identify substrates of the AGC (arginine-directed) kinase familythat comprises PKA, Akt/PKB, PKC, ROCK-II and CamK (Ca2+/Calmodulin-dependent kinases). When the recognition motifs of the AGC kinases were com-pared with 14-3-3 binding motifs it became obvious that the motifs show partialsequence accordance. The AGC kinases had the probability to phosphorylateserine or threonine residues which resided in 14-3-3 binding motifs on IRS2. Ta-ble 3.4 shows the recognition motifs of Akt/PKB and PKA and the two 14-3-3binding motifs.

Table 3.4: Presentation of the recognition motifs of Akt/PKB, PKA and 14-3-3 (R - arginine;X - any amino acid; F - phenylalanine; Y - tyrosine; P - proline; φ - bulky, hydrophobic residue)

Akt/PKB R X R X X S/T φ

PKA R X X S/T

14-3-3 mode I R S X S/T X P

14-3-3 mode II R F/Y X S/T X P

The recognition motifs for Akt/PKB and PKA show partial concordancein the arginine at postion -3 relative to the phosphorylated serine/threonine.Due to the fact that the PKA substrate antibody was raised to recognize argi-nine at -3 from a phosphorylated serine or threonine and the knowledge thatinsulin/IGF-1 stimulation led to activation of the PI 3-kinase pathway withsubsequent Akt/PKB activation resulting in phosphorylation of Ser573 of IRS2and possibly other sites as well, the antibody had to be evaluated if it coulddistinguish between PKA and IGF-1-mediated phosphorylation of IRS2.


Forskolin stimulation induces phosphorylation of IRS2 detected bythe PKA substrate antibody

In a first test (Figure 3.23) it was evaluated what the PKA substrate antibodydetected when cells were incubated with forskolin and H89. Therefore HEK293cells transiently expressing either GFP as control or GFP-IRS2 were starvedfor serum and stimulated afterwards with forskolin or after preincubation withH89.

GFP GFP-IRS2 GFP GFP-IRS2

PKA substrate antibody GFP

-

- -

+ +

+

-

- -

+ +

+

-

- -

+ +

+

-

- -

+ +

+

forskolin

H89

Figure 3.23: PKA substrate antibody recognizes phosphorylation on IRS2 after forskolinstimulation. HEK293 cells were transfected transiently with a vector containing either GFP orGFP-IRS2. 48h later cells were stimulated with 20µM forskolin for 30min or after incubationwith 30µM H89 for 30min. Samples were loaded twice (40µg lysate/lane) on a 7.5 % SDSgel and after western blot the membrane was cut in halve as indicated by the vertical line.The left part was incubated with PKA substrate antibody, the right membrane with GFPantibody. Experiment was performed once.

The left membrane was incubated with the PKA substrate antibody, theright membrane was incubated directly with the GFP antibody to have a controlwhere GFP-IRS2 migrated in the SDS gel. It can be clearly seen that theantibody detected several PKA sites in several proteins of the cell lysates, butthere was only one prominent band at the same height where the correspondingIRS2 band on the right side of the blot migrated. This band was only seen inthe forskolin-treated sample, indicating that forskolin induced phosphorylationof one or more serine/threonine sites on the IRS2 molecule and that these sitesdid not appear when PKA was inhibited by H89 treatment.

IGF-1-induced phosphorylation of IRS2 is not recognized by the PKAsubstrate antibody

To test if the PKA substrate antibody could distinguish between IGF-1 stim-ulation and PKA-mediated phosphorylation of IRS2, HEK293 cells transientlyexpressing GFP-IRS2 were stimulated with IGF-1 or forskolin.

Samples were loaded twice onto a SDS gel and the left membrane was incu-bated with PKA substrate antibody, whereas the right membrane was incubatedwith GFP antibody. The western blot in Figure 3.24 showed the expected signal


- + + - - -

- - + - - -

- - - - + +

- - - - - +

- + + - - -

- - + - - -

- - - - + +

- - - - - +

forskolin

H89

IGF-1

wortmannin

PKA substrate antibody GFP

Figure 3.24: IGF-1-induced phosphorylation of IRS2 is not recognized by PKA substrateantibody. HEK293 cells transiently expressing GFP-IRS2 were incubated for 30min with20µM forskolin, 50ng/ml IGF-1, 30µM H89 or 100nM wortmannin as indicated. 30µg oftotal protein was separated on 7.5 % SDS gel and membrane was cut in halve as indicated bythe vertical line. Left side of the membrane was incubated with PKA substrate antibody, theright side with GFP antibody. Experiment was performed once.

on IRS2 caused by forskolin and no relevant signals from the samples treatedwith IGF-1. Reduced expression of GFP-IRS2 in the H89 pretreated sampleswas not only seen in this blot but appeared frequently in the experiments. Thisphenomenon will be explained later in subsection 4.3.2. From the experimentsshown in Figure 3.23 and Figure 3.24 it was concluded that the PKA substrateantibody only detected forskolin-induced serine/threonine phosphorylation onIRS2 and therefore was suited for the experiments concerning the identificationof PKA phosphorylation sites on IRS2.

3.3.2 PKA-mediated phosphorylation of IRS2 is not influ-enced by Akt/PKB or kinases of the MAPK path-way

Akt/PKB and PKA show overlapping recognition motifs and regulation ofAkt/PKB activity by cAMP has been described before [224]. To rule out a pos-sible influence of Akt/PKB on cAMP-dependent phosphorylation of IRS2, cellstransiently expressing GFP-IRS2 were stimulated with IGF-1 and the Akt/PKBinhibitor Akti. After GFP pulldown the membrane was incubated with PKAsubstrate antibody.

In Figure 3.25 successful IGF-1 stimulation is shown as phosphorylation ofThr308 of Akt/ PKB, but this did not lead to phosphorylation of sites recognizedby the PKA substrate antibody. As a read-out for stimulation with forskolinphosphorylation of Ser157 of VASP (vasodilator-stimulated protein) was moni-tored. Only stimulation with forskolin led to phosphorylation of PKA sites onIRS2 and pretreatment of cells with the Akt/PKB inhibitor had no influence.


Akti

IGF-1

forskolin

PKA substrate antibody

IRS2

p-Thr308

Akt/PKB

- + - - + +

- - + - - +

- - - + + -

GFP pulldown

lysate

p-Ser157 VASP

Figure 3.25: Akt/PKB has no influence on forskolin-stimulated phosphorylation of IRS2.GFP-IRS2 was transiently expressed in HEK293 cells and cells were incubated for 30min with20µM forskolin or 50ng/ml IGF-1. Incubation with 1µM Akti was performed 30min priorIGF-1/forskolin stimulation. GFP-IRS2 was precipitated from 200µg total protein and mem-brane was incubated with PKA substrate antibody. Phosphorylation of Thr308 of Akt/PKBwas assessed as IGF-1 stimulation control and phosphorylation of Ser157 of VASP as forskolinstimulation control. Corresponding IRS2 and Akt/PKB reblots are also shown.

ERK has been shown to phosphorylate IRS2 [48] and kinases downstream ofERK (e.g. p90RSK) could probably also influence cAMP-dependent phosphory-lation of IRS2. To exclude this, HEK293 cells transiently expressing GFP-IRS2were pretreated with MEK1 inhibitor PD98059 before forskolin treatment. GFPpulldown followed incubation of the membrane with the PKA substrate anti-body. The results of this experiment are depicted in Figure 3.26.

PD98059

EGF forskolin


IRS2

p-Thr202/p-Tyr204

Erk

- + - - + +

- - + - - +

- - - + + -

GFP pulldown

p-Ser157 VASP

lysate

Figure 3.26: MAPK pathway kinases have no influence on forskolin-stimulated phosphory-lation of IRS2. GFP-IRS2 was transiently expressed in HEK293 cells and cells were incu-bated for 30min with 20µM forskolin or 100ng/ml EGF. Cells were incubated with 50µMPD98059 for 30min prior EGF/forskolin stimulation. GFP-IRS2 was precipitated from 200µgtotal protein and membrane was incubated with PKA substrate antibody. Phosphorylationof Thr202/Tyr204 of ERK was monitored as control for successful EGF stimulation. Corre-sponding IRS2 and ERK reblots are also shown.


MEK1 inhibitor PD98059 did not influence cAMP-dependent phosphory-lation of PKA sites on IRS2. Moreover, stimulation with EGF as an activatorof the MAPK pathway did not lead to phosphorylation of sites recognized bythe PKA substrate antibody. As a positive control for EGF stimulation phos-phorylation of ERK on Thr202 and Tyr204 was assessed.

To conclude, these results indicated a regulated interaction between IRS2and 14-3-3 upon elevated cAMP levels and this interaction may be mediated byphosphorylation of IRS2 on PKA sites.

3.3.3 Investigation of IRS2 and its fragments by overlayassay and PKA substrate antibody leads to opposingresults

To locate the PKA-dependent 14-3-3 binding region on IRS2 the already de-scribed fragments of IRS2 were transfected transiently in HEK293 cells andtested for residual 14-3-3 binding after stimulation with forskolin and H89.

14-3-3-HRP

GFP

- + +

- - +

- + +

- - +

- + +

- - +

- + +

- - +

- + +

- - +

forskolin

H89

GFP-IRS2 GFP-IRS2-1-300

GFP-IRS2- 1-600

GFP-IRS2-301-1321

GFP-IRS2-601-1321

Figure 3.27: Overlay assay from IRS2 wild type and fragments after forskolin and H89 stim-ulation. HEK293 cells were transiently transfected with GFP-IRS2 wild type or fragmentsGFP-IRS2-1-300, GFP-IRS2-1-600, GFP-IRS2-301-1321 and GFP-IRS2-601-1321. Stimula-tion with 20µM forskolin for 30min and with 30µM H89 prior forskolin stimulation followedpulldown of 200µg of total protein and samples were divided in two equal volumes and sepa-rated on 5-15 % gradient gel. Membranes were incubated either with 14-3-3-HRP for overlayassay or with GFP as expression and loading control.

Wild type IRS2 showed binding of 14-3-3 in the serum starved conditionand after forskolin stimulation, but was completely blocked by inhibiting PKAactivity with H89. Fragment 1-300 did not show any binding at all, whereasfragment 1-600 displayed a similar binding pattern compared to wild type IRS2.For the fragments 301-1321 and 601-1321 a regulated binding of 14-3-3 to IRS2 inthe forskolin-stimulated condition was detected, which was completely abolishedwhen cells were pretreated with H89. Based on these results it was concludedthat PKA-mediated 14-3-3 binding of IRS2 occurred somewhere after position300. The different signal intensities from the fragments in the overlay assaywere caused by the unequal expression of GFP-IRS2 and the fragments thereof.The fragments showed a differential expression pattern on western blot (seeFigure 3.28), although the same amount of plasmid DNA was used to transfectthe cells. This had been observed before, when the IGF-1-mediated 14-3-3binding to IRS2 had been investigated (Figure 3.9).

To verify the overlay data shown in Figure 3.27 with the PKA substrateantibody, the experiment was repeated and the membrane was incubated withthe PKA substrate antibody (Figure 3.29).

The PKA substrate antibody gave comparable results concerning GFP-IRS2,



GFP-IRS2- 1-600

GFP-IRS2-301-1321

GFP-IRS2-601-1321

GFP

Figure 3.28: Cell lysates from transiently transfected IRS2 wild type and fragments. Lysateswere prepared as described in Figure 3.27. 20µg of total protein was separated on 5-15 %gradient gel and membrane was incubated with GFP antibody. The arrow indicates a longerexposure time.

fragment 1-300, fragment 301-1321 and fragment 601-1321. A clear differencecompared to the overlay data was observed with fragment 1-600 since the PKAsubstrate antibody detected no signal in any condition. This opposing resultfor the fragments led to the following hypothesis: a serine or threonine betweenamino acids 300-600 was probably phosphorylated constitutively or phosphory-lated by another kinase than PKA, and resided within a 14-3-3 binding motifwhich explained the overlay assay data. The residue(s) that can be phosphory-lated by forskolin stimulation and recognized by the PKA substrate antibodyare located behind amino acid position 601.

- + + - + +

- - + - - +


GFP


GFP-IRS2-1-600

GFP-IRS2-301-1321

GFP-IRS2-601-1321

- + + - + + - + +

- - + - - + - - +

forskolin

H89

Figure 3.29: Detection of GFP-IRS2 wild type and fragments with PKA substrate antibody.HEK293 cells were transiently transfected with GFP-IRS2 or GFP-IRS2-1-300, GFP-IRS2-1-600, GFP-IRS2-301-1321 and GFP-IRS2-601-1321. Cells were incubated with 20µM forskolinfor 30min or after preincubation with 30µM H89 for 30min. 100µg of total protein wasused for GFP pulldown, samples were separated on a 5-15 % SDS gel and membrane wasincubated with PKA substrate antibody. Stripping of the membrane followed reprobing withGFP antibody as expression and loading control.

3.3.4 GST pulldown experiments to further elucidate 14-3-3 binding to IRS2

The opposing results from subsection 3.3.3 led to the application of GST pull-down experiments as another methodic approach than the overlay assay to in-vestigate which fragments were capable of interacting with a GST-14-3-3 fusionprotein. HEK293 cells were transfected transiently with GFP-IRS2 wild type


and fragments, stimulated with forskolin and H89 and lysates were incubatedwith either GST-14-3-3 isoform β or ε. Membranes were incubated with GSTantibody to ensure equal pulldown and with GFP antibody to visualize inter-action between GST-14-3-3 and IRS2 fragments.

GST

GFP

GST

GFP


GFP-IRS2-1-600

GFP-IRS2-301-1321

GFP-IRS2-601-1321

pulldown: GST-14-3-3β

pulldown: GST-14-3-3ε

- + + - + +

- - + - - +

- + + - + + - + +

- - + - - + - - +

forskolin

H89

Figure 3.30: GST pulldowns from GFP-IRS2 wild type and fragments. HEK293 cells weretransfected transiently with GFP-IRS2 or fragments GFP-IRS2-1-300, GFP-IRS2-1-600, GFP-IRS2-301-1321 and GFP-IRS2-601-1321. Cells were stimulated with 20µM forskolin for 30minor with 30µM H89 for 30min prior forskolin stimulation. 250µg of total protein was incubatedwith GST-14-3-3β or ε and samples were separated on 5-15 % gradient gels and membraneswere incubated with GFP and GST antibody.

Wild type IRS2 interacted nicely with the GST-14-3-3 fusion proteins uponstimulation with forskolin and this interaction was completely inhibited whencells were treated with H89 prior forskolin stimulation. Comparable to the over-lay assay data did the GST-14-3-3 fusion proteins not at all interact with thefragment 1-300. Fragment 1-600 showed increased interaction upon forskolinstimulation and reduced interaction by H89 pretreatment in contrast to theoverlay data, where only an unregulated binding in every condition was ob-servable. Fragments 301-1321 and 601-1321 showed similar results comparedto the overlay data. The GST pulldowns supported the hypothesis drawn insubsection 3.3.3 that in the amino acid area 300-600 a serine/threonine residueresided in a 14-3-3 binding motif. This residue was phosphorylated in the basalstate, forskolin stimulation could increase its phosphorylation and compoundH89 could block this phosphorylation, hence reduce binding of 14-3-3 in theGST pulldown. In addition did the GST pulldown verify the finding that aPKA-dependent phosphorylation site on IRS2 resides behind amino acid posi-tion 601 and the phosphorylation of this serine/threonine residue is importantfor 14-3-3 binding.


3.3.5 Generation of fragment 601-952 to further narrowdown the PKA-dependent 14-3-3 binding region

Due to the fact that the area from amino acids 601-1321 still comprised 720amino acids another fragment was generated to further narrow down the areafor forskolin-dependent 14-3-3 binding. The additionally generated fragment isshown together with the already described fragments in Figure 3.31.

PH PTB KRLB N C

31 140 191 294 591 735

1-300 GFP

601-1321 GFP

1-600 GFP

301-1321 GFP

GFP 601-952

Figure 3.31: Scheme of IRS2 fragments used for identification of the PKA-dependent 14-3-3binding region. Shown are 5 fragments of IRS2 according to mouse numbering. All fragmentswere GFP-tagged at the N-terminus.


GFP


GFP-IRS2-1-600

GFP-IRS2-301-1321

GFP-IRS2-601-1321

GFP-IRS2-601-952

- + +

- - +

- + +

- - +

- + +

- - +

- + +

- - +

- + +

- - +

- + +

- - +

forskolin

H89

Figure 3.32: Detection of GFP-IRS2 and fragments with PKA substrate antibody afterGFP pulldown. HEK293 cells were transfected transiently with GFP-IRS2, GFP-IRS2-1-300,GFP-IRS2-1-600, GFP-IRS2-301-1321, GFP-IRS2-601-1321 or GFP-IRS2-601-952. Starva-tion for serum followed incubation with 20µM forskolin for 30min or after preincubation with30µM H89 for 30min. After cell lysis 100µg of protein was used for GFP pulldown and af-ter separation on a 7.5 % SDS gel membrane was incubated with PKA substrate antibody.Stripping of the membrane followed reprobing with GFP antibody as expression and loadingcontrol.

Again, fragments and wild type IRS2 were expressed transiently in HEK293cells, stimulated with forskolin and H89, followed by GFP pulldown and incuba-tion of the membranes with PKA substrate antibody (Figure 3.32). The PKAsubstrate antibody did not recognize any signal in the fragment GFP-IRS2-601-952.


To test if fragment GFP-IRS2-601-952 also did not interact with 14-3-3 aGST pulldown assay was performed to test for residual 14-3-3 binding. Frag-ments were transfected transiently in HEK293 cells and after stimulation withforskolin and H89 GST pulldowns were performed. A representative blot forpulldown with GST-14-3-3β is shown in Figure 3.33.


GFP-IRS2-1-600

GFP-IRS2-301-1321

GFP-IRS2-601-1321

GFP-IRS2-601-952


GST

GFP

- + + - + + - + + - + + - + + - + +

- - + - - + - - + - - + - - + - - + H89

forskolin

Figure 3.33: GST pulldown of IRS2 wild type and fragments including fragment 601-952.HEK293 cells were transfected transiently with GFP-IRS2 wild type or fragments, stimulatedwith 20µM forskolin for 30min or after preincubation with 30µM H89 for 30min. Lysatescontaining 250µg protein were incubated with 2µg GST-14-3-3β for 2h. After SDS-PAGE andwestern blot membrane was incubated with PKA substrate antibody and stripped afterwardsfor GFP reblot. Lower part of the membrane was incubated with GST antibody to ensureequal pulldown.

Incubation with the PKA substrate antibody confirmed the presence of PKAphosphorylation sites on IRS2. Fragment 601-952 could not interact with GST-14-3-3β, thus the PKA substrate antibody could not detect a phosphorylationon this fragment. From this experiment it was concluded that one or morePKA-dependent phosphorylation site(s) on IRS2 were located behind aminoacid position 952.

3.3.6 Mass Spectrometry for the identification of poten-tial PKA phosphorylation and 14-3-3 binding sites

In collaboration with the Institute for Experimental Ophtalmology in Tubingenmass spectrometry was applied to identify candidates for phosphorylation byPKA. HEK293 cells were transfected transiently with GFP or GFP-IRS2 andstimulated with forskolin or after preincubation with H89. GFP-IRS2 was puri-fied from 5mg of total protein with the GFP Trap R© and run on a precast Bis-trismini gel. After staining with Coomassie the corresponding IRS2 bands were cutand processed according to section 2.10 for mass spectrometric analysis.


250

150

100

75

50

37

25

GFP GFP-IRS2

- + +

- - +

- + +

- - +

forskolin

H89

Figure 3.34: Coomassie staining of a SDS gel to cut bands for mass spectrometry. HEK293cells were transfected transiently with GFP or GFP-IRS2, starved for serum and incubatedwith 20µM forskolin for 30min or after preincubation with 30µM H89 for 30min. 5mg of totalprotein were used for purification with GFP-Trap R©. Samples were run on a 4-12 % Bis-trismini gel and protein bands were visualized by staining with Coomassie G-250. The first laneshows the molecular weight marker and the corresponding molecular weights are depicted onthe left side of the figure in kDa.

The Coomassie staining (Figure 3.34) showed successful purification of GFP-IRS2 as can be seen by the prominent bands that migrated between 150 and250kDa in the GFP-IRS2 lanes. Only the bands corresponding to IRS2 werecut and analyzed by mass spectrometry for phosphorylated residues. Otherbands visible in the GFP and the GFP-IRS2 lanes indicate unspecific interac-tions, but several bands are unique to the GFP-IRS2 lanes. These bands werenot analyzed by mass spectrometry but may represent interaction partners orIRS2 degradation products. Three biological replicates were prepared, analyzedand evaluated. The previous experiments with the different IRS2 fragments re-vealed the region behind amino acid position 952 to contain one or more PKAphosphorylation sites. That is why only sites that matched the following crite-ria were considered for further testing: present in all three replicates, locatedbehind position 952, not present in the H89 pretreated sample, match the con-sensus sequence for PKA (RXXpS) and match a 14-3-3 binding motif (modeI: RSXpS/pTXP; mode II: RX(F/Y)XpS/pTXP). From that evaluation threecandidates emerged that are presented in Table 3.5.

Table 3.5: Sequences surrounding the potential PKA phosphorylation sites Ser1137, Ser1138

and Ser1163 are shown according to mouse IRS2 numbering.

Ser1137 1131GGRRRHSSETFSS1143

Ser1138 1131GGRRRHSSETFSS1143

Ser1163 1157NSKRHNSASVENV1169


Of note, phosphorylated Ser1137 and Ser1138 could be found in peptides assingle phosphorylated residues but also doubly phosphorylated peptides contain-ing phosphorylated Ser1137 and Ser1138 were detectable as shown in Figure 3.35.

m/z [amu]

rela

tive

ab

un

dan

ce

400 600 800 1000 1200 1400

Precursor- H3PO4 +++

918.7555

Precursor-2×H3PO4 +++

886.0965

y11+

1170.5902

y12+

1271.6379

IRS2_MOUSE R H pS pS E T F S S T T T V T P V S P S F A H N S K

y23++

1280.0380

y23++-2×H3PO4

1182.0611

b3+-H3PO4

363.1888

b4+-2×H3PO4

432.2102

b16++-2×H3PO4

841.9103

b13++-2×H3PO4

693.3258

y9+

974.4690

y11++

585.7987

y22++

1196.5389

y23 y22

b3 b4

y12 y11 y9

b16 b13

Figure 3.35: MS/MS analysis of a tryptic peptide. The peptide with the parent mass(M+3H) = 951.4145 reveals a mass spectrum consisting of a single- and double-charged frag-ment mass (preferably ions most relevant to phosphorylation sites are annotated) matchingthe doubly phosphorylated IRS2 peptide sequence: RHpSpSETFSSTTTVTPVSPSFAHNSK.

When the surrounding amino acids of Ser1137 and Ser1138 were inspected itbecame obvious that Ser1137 (GRRRHS1137SETFSS) and Ser1138 (GRRRHSS1138

ETFSS) could probably compensate for each other, because in the case of sub-stitution of either residue with an alanine there would still be the same motifpresent that could be recognized by PKA as shown in Table 3.6.

Table 3.6: Comparison of the recognition motif of the PKA substrate antibody with theamino acids surrounding Ser1137 and Ser1138

PKA substrate ab R X X pS/pT

Ser1137 G R R R H Ser1137 Ser1138 E T F S

Ser1138 R R R H Ser1137 Ser1138 E T F S S

To exclude the possibility of compensation not only the single residues weremutated to alanine, but also an additional double mutant S1137A/S1138A wasgenerated. Transient expression of GFP, GFP-IRS2, GFP-IRS2-S1137A, GFP-IRS2-S1138A, GFP-IRS2-S1163A or GFP-IRS2-S1137A/S1138A in HEK293cells followed stimulation with forskolin and H89, subsequent GFP pulldownand incubation of the membrane with the PKA substrate antibody.

Figure 3.36 shows that the PKA substrate antibody only detected forskolin-induced phosphorylation on GFP-IRS2 wild type and the GFP-IRS2-S1163Amutant. This argued against Ser1163 as a PKA phosphorylation site and forSer1137 and Ser1138 as PKA phosphorylation sites. Due to the detection of apeptide phosphorylated on both Ser1137 and Ser1138 residues and the PKA motif



IRS2

GFP GFP-IRS2 GFP-IRS2-S1137A

GFP-IRS2-S1138A

GFP-IRS2-S1163A

GFP-IRS2- S1137A/ S1138A

forskolin

H89

- + + - + +

- - + - - +

- + + - + +

- - + - - +

- + + - + +

- - + - - +

Figure 3.36: Mutation of Ser1137 and Ser1138 abolishes recognition by PKA substrateantibody. HEK293 cells were transfected transiently with GFP, GFP-IRS2, GFP-IRS2-S1137A, GFP-IRS2-S1138A, GFP-IRS2-S1163A, GFP-IRS2-S1137A/S1138A and incubatedwith 20µM forskolin for 30min or after preincubation with 30µM H89 for 30min. 200µg totalprotein were used for GFP pulldown, samples were separated on a 7.5 % SDS gel and afterwestern blotting the membrane was incubated with PKA substrate antibody. Stripping andreprobing with IRS2 antibody was performed as expression and loading control.

still present in the single mutants GFP-IRS2-S1137A and GFP-IRS2-S1138Athe double mutant GFP-IRS2-S1137A/S1138A was used for further studies.Sequence alignments, as shown in Table 3.7, of IRS2 protein sequences fromhuman, rat, mouse and frog revealed sequence conservation.

Table 3.7: Sequence alignment from different species in the area surrounding Ser1137/Ser1138.Shown are the amino acid areas surrounding Ser1137 and Ser1138 from IRS2 in differentspecies: Homo sapiens (NP 003740.2), Mus musculus (NP 001074681.1), Rattus norvegicus(NP 001162104.1), Xenopus laevis (AAH72768.1).

Mus musculus 1131GGRRRHSSETFSST1148

Rattus norvegicus 1133GGRRRHSSETFSST1151

Homo sapiens 1142GGRRRHSSETFSST1155

Xenopus laevis 957G-RRRHSSETFAS-


3.3.7 Mutation of Ser1137 and Ser1138 on IRS2 reduces in-teraction with 14-3-3

If Ser1137 and Ser1138 are important for cAMP-mediated interaction with 14-3-3 proteins, mutation of these sites should reduce interaction. This hypothe-sis was tested by expressing GFP-IRS2 and GFP-IRS2-S1137A/S1138A tran-siently in HEK293 cells and subsequent stimulation with forskolin and H89prior forskolin stimulation. Pulldown experiments were performed with eitherGST-14-3-3ε or GST-14-3-3β and membranes were incubated with GFP anti-body to visualize the amount of protein that interacted with GST-14-3-3 (Fig-ure 3.37a) and in Figure 3.37b the densitometric analysis of IRS2 interactionwith GST-14-3-3ε is shown.

GFP

GST

GFP

GST p

ulld

ow

n:

GST

-14

-3-3ε

pu

lldo

wn

: G

ST-1

4-3

-3β

- + + - + +

- - + - - +

forskolin

H89

GFP-IRS2 GFP-IRS2- S1137A/ S1138A

a) b)


- + + - + +

- - + - - +

forskolin

H89

0

0.25

0.5

0.75

1.0

GF

P/G

ST

*

Figure 3.37: Double mutant GFP-IRS2-S1137A/S1138A shows reduced binding to GST-14-3-3. a) GFP-IRS2 and GFP-IRS2-S1137A/S1138A were expressed transiently in HEK293cells and cells were either left untreated, stimulated with 20µM forskolin alone for 30min orafter preincubation with 30µM H89 for 30min. Lysates containing 250µg of total protein wereincubated with 2µg GST-14-3-3β or GST-14-3-3ε for 2h and afterwards separated by 7.5 %SDS PAGE. Transfer onto nitrocellulose membranes followed incubation with GFP antibodyto visualize the amount of protein that interacted with GST-14-3-3 and with GST to ensureequal pulldown and loading. b) Densitometric analysis of pulldown with GST-14-3-3ε. GFPsignal intensities were normalized to GST and GFP-IRS2 stimulated with forskolin was setas 1 (mean ± SEM, n = 3, * p < 0.05 GFP-IRS2 forskolin vs. GFP-IRS2-S1137A/S1138Aforskolin).

IRS2 wild type slightly interacted with both GST-14-3-3ε and GST-14-3-3βfusion proteins in the serum starved condition. Forskolin stimulation increasedthe interaction and treatment of cells with H89 prior forskolin stimulation com-pletely abolished it. The double mutant showed no interaction with GST-14-3-3εin the basal condition and minor interaction with GST-14-3-3β. There was alsoan interaction detectable in the forskolin-treated samples, but the extent wassignificantly reduced compared to IRS2 wild type.


To further confirm these results co-immunoprecipitation assays were per-formed where HEK293 cells were transiently transfected with GFP-IRS2 orGFP-IRS2-S1137A/S1138A in addition with myc-tagged 14-3-3. Cells were sti-mulated with forskolin and H89 and lysates were incubated with myc antibody.Western blot membranes were incubated with myc antibody to ensure equalimmunoprecipitation and expression of myc and with GFP antibody to detectco-precipitated GFP-IRS2 or mutant IRS2.

Figure 3.38a shows the results of the western blot analysis and the signalintensities of GFP were normalized for myc and analyzed statistically in Fig-ure 3.38b. The double mutant showed interaction with myc-14-3-3 to a verymuch lesser extent compared to GFP-IRS2.

Taken together, it could be concluded that Ser1137/Ser1138 were crucial forPKA-mediated interaction between IRS2 and 14-3-3.

GFP

myc

GFP-IRS2 GFP-IRS2-S1137A/ S1138A

my

c-IP

- + + - + +

- - + - - +

forskolin

H89

a)

0

0.25

0.5

0.75

1.0

*

- + + - + +

- - + - - +

forskolin

H89


GF

P/m

yc

b)

Figure 3.38: Co-immunoprecipitation of myc-14-3-3 and IRS2. a) HEK293 cells were tran-siently co-transfected with GFP-IRS2 and myc-14-3-3γ or GFP-IRS2-S1137A/S1138A andmyc-14-3-3γ and were incubated with 20µM forskolin for 30min or after pretreatment with30µM H89 for 30min. Lysates containing 200µg protein were incubated with myc antibodyand protein G sepharose for 3h at 4◦C. Samples were separated on 5-15 % SDS gels and afterwestern blotting membrane was incubated with GFP and myc antibody. b) Densitometricanalysis of co-immunoprecipitated GFP. Band intensities were normalized for myc and GFP-IRS2 treated with forskolin was set as 1 (mean ± SEM, n = 3, * p < 0.05, GFP-IRS2 forskolinvs. GFP-IRS2-S1137A/S1138A forskolin).

3.3.8 Forskolin stimulation prevents protein degradationof IRS2 wild type but not of IRS2-S1137A/S1138A

Based on the strong effects of elevated cAMP levels on IRS2 protein amount [76]a direct effect on IRS2 protein stability from forskolin-induced 14-3-3 bindingwas hypothesized. Therefore, HEK293 cells stably expressing IRS2 were gener-ated. The advantage of stable transfection is the continuous expression of thedesired protein at considerably lower levels than in transiently transfected cells,thus preventing an obscuring of IRS2 protein half-life due to permanently highexpression levels.

IRS2 protein degradation in Flp-In HEK293 cells

Figure 3.39 shows the visualization of IRS2 protein degradation. For this, cellswere incubated with either cycloheximide to inhibit protein translation alone or


in combination with forskolin for 1, 3 and 6h and the IRS2 protein content wasassessed by western blot.

After 3 and 6h of incubation with cycloheximide IRS2 protein content wasreduced by 80 %. Concomitant incubation with forskolin reduced IRS2 proteincontent only by 25 %.

- 1 3 6 1 3 6 h

CHX CHX/FSK

IRS2

p-Ser157 VASP

β-actin

20

40

60

80

100

1 3 6

* * *

CHX CHX/FSK IR

S-2

/β-a

ctin

(%

of

con

tro

l)

0

[h]

a) b)

Figure 3.39: IRS2 protein stability is increased upon forskolin stimulation in Flp-In HEK293cells. a) Flp-In HEK293 cells stably expressing IRS2 were starved 3h for serum and incubatedwith 25µg/ml cycloheximide (CHX) alone or in combination with 20µM forskolin (FSK) for 1,3 and 6h. 10µg of total protein was separated on 7.5 % SDS gel and membrane was incubatedwith IRS2 antibody. Detection of p-Ser157 of VASP was performed to ensure successfulstimulation and β-actin was detected as loading control and to enable normalization of IRS2protein content. b) Densitometric analysis of IRS2 protein, IRS2 content from untreated cellswas set as 100 % (mean ± SEM, n = 8, * p < 0.05 CHX vs. CHX/FSK). White diamondsrepresent treatment with CHX, black squares represent treatment with CHX and FSK.

IRS2 protein degradation in primary mouse hepatocytes

To evaluate if the results of the experiment in Figure 3.39 could be verified in asystem that is closer to the in vivo situation, the experiment was also performedwith primary mouse hepatocytes.

- 1 3 6 1 3 6 h

CHX CHX/FSK

IRS2

β-actin

p-Ser157 VASP

Figure 3.40: Visualization of IRS2 protein degradation in primary hepatocytes. a) Hepato-cytes were isolated from male C57Bl/6 mice, plated onto 6-well plates and starved for serumovernight 24h later. Cells were incubated with 25µg/ml cycloheximide (CHX) alone for 1, 3and 6h or in combination with 20µM forskolin (FSK). 100µg protein was separated by 7.5 %SDS gel and membranes were incubated with antibodies against IRS2, p-Ser157 VASP andβ-actin. Experiment was performed once.

Similar to the results from the Flp-In HEK293 cells stably expressing IRS2,IRS2 protein content decreased over time with cycloheximide treatment in pri-mary hepatocytes. Additional incubation with forskolin prevented the IRS2protein from degradation.


Ser1137 and Ser1138 are important for increased IRS2 protein stabilityupon PKA activation

Having identified Ser1137 and Ser1138 as important sites on IRS2 for phosphory-lation by PKA and subsequent 14-3-3 binding the question arised if the IRS2S1137A/S1138A double mutant is sensitive to PKA-dependent protein stabiliza-tion. Therefore, Flp-In HEK293 cells were stably transfected with GFP-IRS2or GFP-IRS2-S1137A/S1138A and incubated with cycloheximide and forskolinfor 1, 3 and 6h.

GFP-IRS2 CHX/FSK

GFP-IRS2-S1137A/S1138A CHX/FSK

20

0

40

60

80

100

1 3 6 [h] IR

S-2

/β-a

ctin

(%

of

con

tro

l)

* * a)

IRS2

β-actin

- 1 3 6 1 3 6 - 1 3 6 1 3 6 h

CHX CHX/FSK CHX CHX/FSK b)

GFP-IRS2 GFP-IRS2-S1137A/S1138A

p-Ser157 VASP

Figure 3.41: Ser1137 and Ser1138 are crucial for IRS2 protein stability. a) Flp-In HEK293cells were stably transfected with GFP-IRS2 or GFP-IRS2-S1137A/S1138A. Cells were starvedfor serum and incubated with 25µg/ml cycloheximide (CHX) and 20µM forskolin (FSK) for 1,3 and 6h. 50µg protein was separated on 7.5 % SDS gels and membranes were incubated withIRS2 antibody, p-Ser157 VASP antibody as stimulation control and β-actin antibody as loadingcontrol. b) Extent of IRS2 protein degradation was quantified by scanning immunoblots andnormalization of IRS2 signal intensities for β-actin. Untreated cells were set as 100 % (mean± SEM, n = 4, * p < 0.05 GFP-IRS2 CHX vs. GFP-IRS2 CHX/FSK at 3h and 6h).

GFP-IRS2 wild type protein showed increased protein stability upon stimu-lation with forskolin compared to the S1137A/S1138A mutant. Thus, the lackof cAMP-dependent phosphorylation of IRS2 on Ser1137/Ser1138 abolished theability of forskolin to increase IRS2 protein stability.

Chapter 4

Discussion

IRS1 and IRS2 as intracellular signalling nodes contain numerous serine andthreonine residues beside tyrosine residues that can be modified by protein ki-nases and protein phosphatases. Especially for IRS1 this has been shown ex-tensively [201, 113, 159, 202, 200]. An increasing number of studies revealed theimportance of 14-3-3 proteins in controlling cellular processes like cell cycle, cellgrowth, gene transcription and apoptosis as reviewed in [120, 105, 43, 49]. Lessis known about individual serine/threonine phosphorylation sites on IRS2, theinteraction of 14-3-3 with IRS2 and the influence on cellular metabolism. Thisthesis aimed to characterize the interaction between 14-3-3 and IRS2.

Interaction of IRS1 and 14-3-3 has been investigated in some studies, butwith partially opposing results. Isotype- and cell line-specific differences mayaccount for the different results in the different publications. The overall ten-dency seems to be a rather negative modulation of binding partners and sig-nalling. Ogihara et al. identified three potential 14-3-3 binding sites on IRS1[121]. Additional experiments led to the conclusion that 14-3-3 binding to IRS1is not insulin-dependent since the amount of 14-3-3 bound to IRS1 was similarwith or without insulin stimulation in HepG2 cells, but co-immunoprecipitatedIRS1 was less phosphorylated on serine and tyrosine residues in comparison toimmunoprecipitated IRS1 alone. They suggested that 14-3-3 binding to IRS1may regulate insulin sensitivity by interrupting IRS1 association with the in-sulin receptor. On the contrary, Kosaki et al. described that 14-3-3β binds toIRS1 in 3T3-L1 adipocytes and this interaction could be increased with insulinstimulation [85]. Since PI 3-kinase isolated from a complex consisting of 14-3-3β/IRS1/PI 3-kinase showed only half of the activity of PI 3-kinase isolatedfrom a complex of IRS1/PI 3-kinase by sequential immunoprecipitation, theyconcluded an inhibitory function on PI 3-kinase activity. Oriente et al. de-scribed a regulatory function of 14-3-3ε in 3T3-hIR cells [123]. In this study theauthors showed that upon insulin stimulation a multimolecular complex consist-ing of IRS1, PKCα and 14-3-3ε is formed, in which 14-3-3ε is not necessary forthe formation of the complex, but for regulating the function of PKCα. Capraroet al. showed in vitro interaction of 14-3-3ε with the IGF-1 receptor and IRS-1,but not the insulin receptor [26] and Xiang et al. demonstrated a role for 14-3-3binding to IRS-1 in intracellular trafficking of IRS-1 [211].

A clear role for IRS1 and 14-3-3 interaction cannot be established from thesepublications, thus extrapolating a possible outcome for the interaction between

87

4.1. APPLICATION OF TRUNCATED PROTEINS 88

IRS2 and 14-3-3 can not be made. This stresses the importance of the generalcharacterization of IRS2 and 14-3-3 interaction. The IRS2/14-3-3 interactionhas been documented in large-scale proteomic studies, but only as one of manyinteraction partners without any further characterization [39, 77, 109]. Ogiharaet al. showed interaction of overexpressed IRS2 and 14-3-3 in Sf9 cells [121] andDubois et al. showed IRS2 as 14-3-3 interaction partner upon IGF-1 stimulationinvolving the PI 3-kinase pathway in HEK293 cells [35].

This thesis aimed to characterize the insulin/IGF-1-dependent 14-3-3 bind-ing to IRS2 and to discover additional, novel modulating stimuli. ElevatedcAMP levels could be identified as novel modulator of 14-3-3 binding to IRS2. Asin the results section the first part of the discussion is dealing with insulin/IGF-1-dependent 14-3-3 binding to IRS2 and the second part is concentrating on thecAMP-dependent 14-3-3 binding to IRS2.

4.1 Application of truncated proteins as a help-ful tool to track down important binding re-gions

IRS2 contains > 100 serine/threonine residues and since 14-3-3 binding motifsdo not necessarily conform to the exact consensus sequence described by Yaffeet al. and Rittinger et al. it was reasonable to narrow down the area of 14-3-3binding and then search within that area for the exact binding sites [214, 136].The time consuming generation of random single alanine mutations would havebeen unreasonable. Although very artificial, truncated versions of the IRS2protein transiently transfected in HEK293 cells were expressed quite well andwere folded properly enough that Ser1137/Ser1138 could be recognized by PKA.Truncated versions of proteins are used regularly to identify binding regions asshown by [121, 64, 152].

4.2 Insulin and IGF-1-dependent interaction be-tween 14-3-3 and IRS2

4.2.1 Identification of phosphorylated serine/threonine re-sidues on IRS2

The mouse IRS2 sequence was used to identify possibly novel and already de-scribed phosphorylation sites on IRS2. Mass spectrometric analysis revealed 24phosphorylated serine/threonine residues and a sequence alignment of mouseand human IRS2 showed that these sites seem to be conserved in human IRS2as well. Certain sites could also be found in the mouse and human IRS1 sequencewith a high degree of conservation. Due to the high sequence conservation be-tween IRS2 in different species and with the IRS1 amino acid sequence, theliterature and internet databases were screened for any possible reported phos-phorylation site that was also identified in the mass spectrometric approachdescribed in this thesis. The IRS2 protein sequence was scanned for modifiedresidues by using the online tool PhosphositePlus R©1. This database shows if a

1http://www.phosphosite.org

http://www.phosphosite.org

4.2. INSULIN AND IGF-1-DEPENDENT INTERACTION BETWEEN14-3-3 AND IRS2 89

modified site or a homologue of it was already reported. It distinguishes betweenrecords where the sites were experimentally determined with site-specific meth-ods, or if the site was assigned by proteomic discovery-mode mass spectrometry.Ser66, Ser968, Ser977, Ser999 and Ser1266 were not found in this Scan, whereasSer303, Ser305, Thr347, Ser385, Ser388, Thr401, Thr517, Ser556, Ser573, Ser675,Ser722, Ser727, Ser728, Ser762, Ser907, Ser1089, Ser1151, Ser1165 and Ser1190 werelisted. However, beside Thr347 all sites were identified in big mass spectromet-ric screens without any further confirmation and some sites were not annotatedthemselves but the homologue of human or rat IRS2. Of note, several addi-tional sites that were not identified in this mass spectrometric approach arelisted on the PhosphositePlus R© website. Reported homologues are described inthe following passage.

Ser303 of mouse IRS2 has a homologue in the human IRS1 sequence – Ser270.This site has been described as substrate for S6K1 and as part of TNFα-inducedinsulin resistance [222]. Thr347 of mouse IRS2 has a homologue in mouse IRS1that is Thr306. This site has not been experimentally described, but it is locatednext to Ser307 which has been shown to be phosphorylated by JNK, S6K1 andmTOR [59, 28, 202, 201]. One could speculate that since IRS1 Ser307 seems toplay an important role in insulin signalling maybe Thr347 of IRS2 could be animportant residue as well. And indeed, Solinas et al. showed phosphorylationof the rat IRS2 homologue Thr348 by JNK and suggest that this residue mightbe equivalent to Ser307 of mouse/rat IRS1 [172]. Ser573 was identified and char-acterized as insulin/IGF-1-dependent phosphorylation and 14-3-3 binding sitein this thesis. Ser522 is the mouse IRS1 homologue to Ser573 and was exper-imentally described and characterized by Giraud et al. [53]. Phosphorylationof Ser675 and Ser907 after insulin stimulation has already been described byFritsche et al. from our lab [48]. The mutation of Ser675 to alanine led toan increased protein stability of IRS2 and phosphorylation of this site couldbe mediated by mTOR. Ser675 has an IRS1 homologue – Ser632 in mouse/ratIRS1 and Ser636 in human IRS1. Mouse IRS1 Ser632 has been shown to bephosphorylated by ERK [59, 104], ROCK I [92] and to modulate PI 3-kinaseactivity together with Ser612 [113]. Tzatsos et al. showed the phosphorylationof human IRS1 Ser636 by mTOR [189]. Phosphorylation of human IRS1 Ser636

was increased in myotubes from patients with diabetes [21] and also in musclebiopsies of insulin-resistant offspring of type II diabetic parents [111]. The IRS2Ser762 homologue in mouse IRS1 – Ser731 was shown to modulate PI 3-kinaseactivity together with Ser662 [113] and was required together with Ser632 andSer662 for negative regulation by PDGF (platelet derived growth factor) [94].Ser907 was shown to be phosphorylated by ERK [48].

For an overview of the just described homologue residues see Table 4.1 andthe complete sequence alignment of mouse and human IRS1 and IRS2 withindication of the identified modified residues by mass spectrometry can be foundin Figure A.5.

Taken together, with a sequence coverage of 92 % the mass spectrometricanalysis was very successful and suitable to identify several phosphorylated re-sidues that have already been listed in big mass spectrometric approaches. Thecorresponding residues in IRS1 do have an important influence on the intracel-lular signalling cascade. It is therefore of uttermost importance to investigatethe corresponding IRS2 serine/threonine residues since a biological impact ofthese sites is also very likely.


Table 4.1: Experimentally reported serine/threonine residues of IRS1 and IRS2, that werealso identified in the mass spectrometric approach for identification of IGF-1-dependent phos-phorylation sites on mouse IRS2.

mouseIRS2

Correspondingresidue

Description Reference

Ser303 human IRS2 Ser270 S6K1 substrate [222]

Thr347 next to mouse/ratIRS1 Ser307

JNK/S6K1/mTORsubstrate

[59, 28, 202, 201]

Thr347 rat IRS2 Thr348 JNK substrate [172]

Ser573 mouse/rat IRS1Ser522

Akt/PKB substrate [53]


ERK/ROCK I substrate [59, 104, 92]

Ser675 mouse/rat IRS1Ser632 together withSer612

modulation of PI 3-kinaseactivity

[113]

Ser675 human IRS1 Ser636 mTOR substrate [189]

Ser675 human IRS1 Ser636 MAPK-dependent,increased in T2D patients

[21]

Ser675 human IRS1 Ser636 increased in muscle frominsulin-resistant offspringof T2D

[111]

Ser762 mouse/rat IRS1Ser731 together withSer662

modulation of PI 3-kinaseactivity

[113]


PDGF-dependent [94]

4.2.2 Regulation and function of insulin/IGF-1-dependent14-3-3/IRS2 interaction

The kinase panel assay revealed increased binding of 14-3-3 to transiently ex-pressed IRS2 upon IGF-1 stimulation. Co-immunoprecipitation of endogenousexpressed 14-3-3 also showed an increased interaction with IRS2 after IGF-1stimulation (Figure 3.3). Overlay assays performed on IRS2 immunoprecipi-tates from insulin-stimulated Fao rat epatoma cells which expressed IRS2 en-dogenously (Figure 3.2) and from insulin-treated and refed mice (Figure 3.4and Figure 3.5) confirmed this data and implicated a physiological occurrencewith potentially important consequences. The dependence on the PI 3-kinasepathway was demonstrated by two chemically different PI 3-kinase inhibitors− PI-103 and wortmannin. PI-103 is an ATP-competitive inhibitor targetingthe catalytic subunit p110 of PI 3-kinase. Wortmannin irreversibly inhibitsPI 3-kinase since it covalently binds in the ATP binding site of p110. Accordingto Bain and co-workers PI-103 and wortmannin are recommended to inhibit PI


3-kinase in biological processes [14].From all five tested single alanine mutants only mutation of Ser573 to ala-

nine did abrogate the 14-3-3 interaction with IRS2 in the overlay assay (Fig-ure 3.11). This result could be confirmed in Flp-In HEK293 cells stably express-ing the S573A mutant (Figure 3.12). Ser573 and the surrounding amino acids(RKRTYS573LTT) do not completely conform to a 14-3-3 binding motif withthe lack of proline at position +2 relative to the phosphorylated serine beingthe most prominent feature to be different. However, a proline in position +2is found in only 50 % of the target sites as has been reported by Johnson etal. [79]. In addition, if the target sequence would match the consensus motifcompletely, the strength of binding would not give appropriate regulation ofthis protein [6]. The antibody against phosphorylated Ser573 of mouse IRS2 wasindispensable for the characterization of the phosphorylation site. Experimentswith the S573A mutant revealed the specificity of the antibody for the sequenceLRKRTYpS573LTTPAR (Figure 3.13). Although the antibody worked quitewell in transiently and stably transfected HEK293 and Flp-In HEK293 cells itwas not possible to detect increased Ser573 phosphorylation in mouse liver bywestern blot (data not shown). Of note, when the antibody was used to im-munoprecipitate IRS2 phosphorylated on Ser573 from total cell lysates as shownin Figure 3.16 or from mice liver protein lysates (data not shown), it becameevident that the antibody failed to precipitate increased amounts of phosphory-lated Ser573 upon insulin or IGF-1 stimulation. This discrepancy cannot be fullyexplained since attempts with non-phosphorylated peptides to block unspecificbinding during the immunoprecipitation process in the reaction tube were notsuccessful (data not shown) and in addition the antibody as polyclonal anti-body was already purified for the phosphopeptide. To conclude, the Ser573

antibody was specific for phosphorylated Ser573 and when used for western blotwas able to detect increased Ser573 phosphorylation of IRS2 upon stimulationwith insulin or IGF-1, but when the antibody was used for immunoprecipitationincreased amounts of IRS2 protein in insulin- or IGF-1-stimulated samples werenot observable.

The phosphorylation of Ser573 was increased upon insulin/IGF-1 stimulationin a PI 3-kinase-Akt/PKB-dependent manner, thus under the same conditionswhen the 14-3-3/IRS2 interaction occured. Notably, pharmacological inhibi-tion of PI 3-kinase using wortmannin or PI-103 or inhibition of Akt/PKB usingAkti-1/2 prevented the enhanced phosphorylation of Ser573 (Figure 3.18) andthe increased 14-3-3 binding of IRS2 (Figure 3.19). Phosphorylation of Ser573 ofmouse IRS2 with insulin and IGF-1 stimulation and also phosphorylation of thecorresponding residue Ser577 in human IRS2 could be observed (Figure 3.14). Atime course of Ser573 phosphorylation in Fao cells showed peak phosphorylationlevels after 60, 90 and 120 min of insulin treatment (Figure 3.17). A physiolog-ical importance of Ser573 is suggested by the fact that sequence alignments ofdifferent mammalian species showed complete sequence accordance, and evenan amphibian sequence showed partial accordance (Table 3.2).

Akt/PKB is one of the key serine/threonine kinases downstream of PI 3-kinase and is essential for adequate stimulation of glycogen synthesis in skeletalmuscle. Paz et al. described Akt/PKB phosphorylation of IRS1 as a posi-tive regulatory mechanism, since phosphorylation by Akt/PKB protected IRS1from rapid dephosphorylation of tyrosine residues, thus maintaining tyrosine-


phosphorylated conformation of IRS1 [126]. They searched the IRS1 sequencefor potential Akt/PKB recognition motifs and generated a mutant where Ser265,Ser302, Ser325 and Ser358 of mouse IRS1 were mutated to alanine and phosphory-lation of this mutant by Akt/PKB was markedly inhibited. A scan of the IRS1mouse sequence with the online tool Motif Scan2 showed Ser265, Ser302, Ser325

and Ser522 as Akt/PKB site. Ser522 was identified by Giraud et al. a few yearslater as potential Akt/PKB target [53]. They showed insulin-dependent phos-phorylation of Ser522 in L6 myoblasts and myotubes and provided evidence fora negative role of Ser522 in insulin signalling. Online tools to predict possiblerecognition motifs for kinases can be very helpful, but loss-of-function mutantsand inhibitor studies are indispensable to clearly identify a phosphorylation siteand the respective kinase. The quadruple mutant from Paz et al. showed a mea-surable effect, but since they not applied single mutants it cannot be excludedthat some of the tested sites were not targets for Akt/PKB. Ser573 as homologueposition of Ser522 also resides in an Akt/PKB consensus motif (RXRXXpS/pT).Treating cells with an Akt/PKB inhibitor resulted in significantly reduced phos-phorylation of Ser573, thus indicating Akt/PKB as kinase for Ser573 phosphory-lation (Figure 3.18). The Akt/PKB inhibitor Akti is a specific inhibitor of Akt1and Akt2 as has been shown by Logie et al. [103]. Interaction of Akti with thePH domain of Akt/PKB prevents the conformational change that is requiredfor phosphorylation and activation of Akt/PKB by the upstream kinase. Ofnote, downstream kinases of Akt/PKB and other kinases cannot be excluded tophosphorylate Ser573, due to the fact that Ser573 phosphorylation was markedlyincreased when cells were treated with TPA (Figure 3.13). Akt/PKB belongsto the family of AGC kinases and an arginine at position -3 relative to a serineor threonine corresponds to a sequence motif which correlates with recognitionmotifs of this kinase family. In line with Ser522 as phosphorylation site fornegative regulation of signalling were the results with the alanine mutant ofSer573 that increased Akt/PKB phosphorylation upon IGF-1 stimulation. IGF-1 stimulation was carried out over several time points and significantly increasedphosphorylation of Akt/PKB on Thr308 and Ser473 was observable after 120 and240 min of stimulation (Figure 3.20). In conclusion, the data suggested com-parable regulation and function of the homologue Ser573 and Ser522 in IRS2and IRS1, but the implication of Ser522 in 14-3-3 binding of IRS1 has not beenreported yet. It is imaginable that phosphorylation of Ser573 functions as anegative feedback loop for downregulation of Akt/PKB activity.

The data of the overlay assay indicated that only one phosphorylated residueon IRS2 was sufficient to trigger insulin/IGF-1-dependent 14-3-3 interactionwith IRS2. This would be in line with several other studies which also identifiedsingle phosphorylated residues as 14-3-3 interaction site [206, 27, 128]. However,there are also other reports that showed 14-3-3 binding to two simultaneouslyphosphorylated serine/threonine residues on the same target protein (see [79]for a collection of reported binding partners). Ser556 appeared in the study fromDubois et al. to be phosphorylated after insulin stimulation in HeLa cells andthe same residue was identified in the mass spectrometric approach from thisthesis. This opened the possibility that perhaps Ser556 and Ser573 were 14-3-3 binding sites. The S556A mutant showed regulated binding of 14-3-3 in the

2www.scansite.mit.edu/motifscan seq.phtml

www.scansite.mit.edu/motifscan_seq.phtml


overlay assay, hence arguing against Ser556 as a 14-3-3 binding site (Figure 3.11).To support these findings with another methodical approach GST pulldown ex-periments were conducted. Unexpectedly, both single mutants S556A (data notshown) and S573A did interact with GST-14-3-3 and also the double mutantS556A/S573A failed to disrupt the interaction (Figure 3.21). The cause of thisdiscrepancy between the overlay assay and GST pulldown cannot be explainedyet, but it can be speculated that the S573A mutant probably failed to dis-rupt interaction of 14-3-3 with IRS2 in the GST pulldown due to the presenceof other proteins stabilizing 14-3-3 binding or preventing regulated binding ofexogenously added GST-14-3-3 protein. How can the persuasive result of theoverlay assay be explained? The overlay assay was used to monitor interactionof 14-3-3 with IRS2. This type of assay is regularly used to directly visual-ize protein/protein interactions on western blot membranes [151, 175, 79, 35].However, this technique showed binding of 14-3-3 to denaturated IRS2 resolvedon SDS-PAGE which is quite different from the in vivo situation. Usually, forproteins to interact a native conformation is needed. By SDS treatment the pro-teins are denaturated, but upon transfer onto the nitrocellulose membrane andthe subsequent procedure for protein detection the SDS is removed and a partialrenaturation of the protein on the membrane can take place. A possibility isthat the partial renaturation of the IRS2 protein on the nitrocellulose membraneprobably was effective enough to enable binding of 14-3-3 to the phospho-Ser573

motif, whereas the second binding site might not have been easily accessible andfull renaturation of the IRS2 protein would have been required. This may haveled to abrogation of 14-3-3 binding in the S573A mutant in the overlay assay,despite the presence of a second binding site.

Limitations in either of the aforementioned techniques make it necessaryto employ at least two of these methodic approaches to ensure observation ofprotein/protein interaction.

4.2.3 Potential relevance of insulin/IGF-1-dependent phos-phorylation of Ser573 and interaction with 14-3-3 pro-teins

Experiments performed in this thesis concerning phosphorylation kinetics ofSer573 and Akt/PKB indicated a potential role of Ser573 phosphorylation in anegative feedback loop to downregulate insulin/IGF-1 signalling. This is sup-ported by the fact that Ser573 showed peak phosphorylation levels after 60, 90and 120 min of insulin stimulation in Fao cells (Figure 3.17) and the S573Amutant increased Akt/PKB phosphorylation significantly at late time points ofprolonged IGF-1 stimulation (120 and 240 min) (Figure 3.20).

How could the interaction of IRS2 with 14-3-3 proteins influence downstreaminsulin signalling? The rigid structure of the 14-3-3 proteins led to the hypoth-esis that 14-3-3 proteins provide themselves as an anvil whereon the targetprotein can be restructured/reshaped [161]. Experimental data from Datta etal. showed that upon 14-3-3 binding to Ser316 of BAD (Bcl-2-associated deathpromoter) the accessibility of Ser155 of BAD for other kinases increased [29].GlobPlot R© analysis revealed unstructured regions in the IRS2 protein, whichcould potentially be restructured/reshaped upon 14-3-3 binding. Since IRS2functions as a multi-adapter protein, thereby linking the insulin receptor and

4.3. CAMP-DEPENDENT BINDING OF 14-3-3 TO IRS2 94

other receptor tyrosine kinases with intracellular effector molecules, 14-3-3 pro-teins could modulate the interactions with binding partners by binding to IRS2upon phosphorylation of Ser573 and a possible yet to be discovered second bind-ing site.

In summary the presented results include the identification of 24 phos-phorylation sites with not yet described sites on IRS2 and mapping of theinsulin/IGF-1-dependent 14-3-3 binding region. Ser573 as novel phosphory-lation site was shown to be part of insulin/IGF-1-dependent interaction be-tween 14-3-3 and IRS2 and its phosphorylation may have an inhibitory role ininsulin/IGF-1 signalling. The evidence of the interaction between 14-3-3 pro-teins and the signal transducer IRS2 in vivo opens several novel perspectives inthe (patho)physiological regulation of the biological function of IRS2 includingits role in metabolic disorders such as insulin resistance and type 2 diabetes.

4.3 Interaction between 14-3-3 and IRS2 uponelevated cAMP levels

4.3.1 cAMP and PKA

Glucagon as fasting hormone and physiological antagonist of insulin is secretedfrom the α-cells of the pancreas to ensure adequate blood glucose levels dur-ing fasting. It activates the G-protein-coupled glucagon receptor which in turnactivates adenylylcyclase. Adenylylcyclase catalyzes the formation of cAMPfrom ATP. However, elevated cAMP levels can also be generated by the cat-echolamines adrenaline and noradrenaline. They activate β-adrenergic recep-tors that are also G-protein-coupled and stimulate the adenylylcyclase. Cate-cholamines are incrased during exercise and when the body is in a stress situ-ation for preparation of the so-called fight or flight response. Elevated cAMPlevels lead to activation of PKA which has long been thought to be the mostimportant cAMP-dependent kinase, but cAMP can also activate protein kinaseG (PKG) [204] and exchange proteins directly activated by cAMP (EPAC) [83,31]. How can these different pathways that all rely on elevated cAMP levelsbe distinguished? Inhibitor studies can help to dissect the different signallingpathways. H89 as non-selective PKA inhibitor was used in the experimentsof this thesis. Of note a cAMP analogue, 8-(chloro-phenylthio)-2’-O-methyl-adenosine-3’,5’-cyclic monophosphate (8CPT-2MecAMP) that specifically acti-vate EPACs [38] could have also been used, but other inhibitor studies wereapplied to rule out an influence of EPACs, the PI 3-kinase or Ras/MAPK path-way on cAMP-dependent effects observed in the experiments. Of note, cAMPcan protect various cell types from apoptosis by synthesis of antiapoptotic pro-teins, inactivation of proapoptotic proteins and activation of the prosurvival PI3-kinase/Akt/PKB pathway [107, 199, 144], but it can also exert proapototiceffects in certain cancer cell lines [57, 148].


4.3.2 Regulation and function of cAMP-dependent 14-3-3/IRS2 interaction

The kinase panel assay showed reduced binding of 14-3-3 to IRS2 in sampleswhich were pretreated with the PKA inhibitor H89 before stimulation withforskolin. PKA as modulator of interaction between 14-3-3 and IRS2 had notbeen described before, so the first task concerning this part of the thesis was toconfirm this interaction and its dependence on cAMP/PKA. There are severalindications that the observed effects of increased 14-3-3 binding to IRS2 weremediated by PKA. Stimulation of HEK293 cells transiently expressing GFP-IRS2 with the adenylylcyclase stimulating agent forskolin or the cAMP ana-logue CPT-cAMP caused increased binding of 14-3-3 to IRS2 in overlay assayswhat was preventable by pretreating cells with H89 (Figure 3.22). The kinaseinhibitor H89 is not solely inhibiting PKA as has been shown by Davies et al.[30]. It can inhibit MSK1, S6K1 or ROCK-II with a similar potency, but in theexperiments conducted in this thesis these potential additional inhibitions of thementioned kinases should be of minor relevance due to the fact that they are notcAMP-dependent kinases as PKA is. EPAC proteins as Rap1 GTP exchangefactors are upstream of the prominent serine kinases ERK and Akt/PKB. In-hibition of these kinases did not reduce the forskolin-induced phosphorylationof IRS2 on PKA motifs and direct stimulation of ERK and Akt/PKB did notactivate phosphorylation of these sites (Figure 3.25 and Figure 3.26). Togetherthese are good indications for the involvement of the cAMP-PKA pathway asmediator of the forskolin-induced IRS2 phosphorylation and binding of 14-3-3proteins.

Having established that the observed 14-3-3 binding to IRS2 was cAMP-dependent truncated versions of the IRS2 protein were applied to identify theamino acid area behind amino acid position 952 as being crucial for 14-3-3binding. Mass spectrometric analysis led to the identification of IRS2 ser-ine/threonine residues as potential candidates for phosphorylation by PKA.Only 3 candidates matched the criteria to reside in a PKA consensus motif, a14-3-3 binding motif and were present in the forskolin-stimulated sample in all3 biological replicates. The single alanine mutant of Ser1163 did not alter thesignal detected by the PKA substrate antibody and was therefore excluded to bea PKA site (Figure 3.36). Single mutation of either Ser1137 (RRHA1137SETF)or Ser1138 (RRHSA1138ETF) to alanine still resulted in a PKA phosphorylationmotif (RXXpS). Although the PKA substrate antibody only detected a faint sig-nal in both single S1137A and S1138A mutants, subsequent experiments wereperformed with the double GFP-IRS2-S1137A/S1138A mutant to circumventany possible compensatory effects that could not be detected with the PKAsubstrate antibody and due to the fact that peptides phosphorylated on bothresidues 1137 and 1138 at the same time were detected by mass spectrometry(Figure 3.35). The human IRS1 protein sequence contains homologue positionsto Ser1137 and Ser1138 – Ser1100 and Ser1101. Yi et al. expressed GST-IRS1fusion proteins (with only certain areas of the human IRS1 protein) and phos-phorylated them in vitro with PKA or ERK2 followed by HPLC-ESI-MS/MSto detect phosphorylated residues [218]. Among other residues they could ob-serve PKA phosphorylation of residues on peptides corresponding to Ser1100 orSer1101 of human IRS1. Due to the lack of additional definitive spectra the au-thors were not able to clearly state if only Ser1100 or Ser1101 was phosphorylated


Table 4.2: Reported homologue residues of mouse IRS2 Ser1137/Ser1138

mouse IRS2 Corresponding Description Reference

Ser1137 human IRS1 Ser1100 PKA substrate [218]

Ser1138 human IRS1 Ser1101 PKA substrate [218]

Ser1138 human IRS1 Ser1101,human IRS2 Ser1149

PKCθ substrate [96]

or if maybe a doubly phosphorylated peptide could also have been possible.According to the data acquired in this thesis Ser1137/Ser1138 seem to be

the only cAMP-dependent phosphorylation sites on the IRS2 molecule thatare located in a PKA motif and can mediate cAMP-induced 14-3-3 binding toIRS2. However, other stimuli could lead to the phosphorylation of Ser1137 orSer1138. The Ser1138 homologues in human IRS1 (Ser1101) and human IRS2(Ser1149) have been investigated by Li et al., who showed Ser1101 and Ser1149

to be phosphorylated by insulin treatment of C2C12 cells and PKCθ was estab-lished as kinase for this phosphorylation [96]. They suggest a role for Ser1101

phosphorylation as part of an inhibitory feedback signal. This could implicatethat Ser1138 also probably could be phosphorylated by insulin or IGF-1 stim-ulation. Insulin/IGF-1-dependent phosphorylation of Ser1138 was not tested inthe experiments of this thesis, but the data from Li et al. should be kept inmind when further experiments concerning Ser1138 are performed. In Table 4.2the just described homologue residues are listed again for clarity and a sequencealignment of mouse and human IRS1 and IRS2 to visualize the sequence con-servation of Ser1137/Ser1138 can be found in Figure A.5.

The importance of Ser1137/Ser1138 was verified by three different experi-mental approaches: overlay assay, GST pulldown and co-immunoprecipitationof transfected myc-14-3-3 protein. In some experiments interaction betweenGFP-IRS2 wild type and GST-14-3-3 fusion proteins could also be found atbasal conditions. This could indicate the existence of further 14-3-3 bindingsites on IRS2 that can be phosphorylated at basal conditions. These sites couldbe phosphorylated for example by other family members of the AGC kinasefamily. The completely abolished binding of 14-3-3 to IRS2 by treating cellswith H89 can be explained with the already mentioned fact that H89 can alsoinhibit MSK1, S6K1 or ROCK-II. There was still observable interaction betweenGST-14-3-3 and the double mutant S1137A/S1138A in the forskolin-stimulatedcondition (Figure 3.37). It is imaginable that PKA phosphorylated a substratethat itself is a kinase and phosphorylated IRS2 on another 14-3-3 binding motif.A possible candidate in this case for example would be Ser573, since this site wasshown to be a 14-3-3 binding site and can be phosphorylated by insulin/IGF-1-dependent pathways as well as TPA stimulation. Forskolin stimulation led to aphosphorylation of Ser573 and this was blockable with pretreatment of the cellswith H89 (data not shown). A triple GFP-IRS2-S573A/S1137A/S1138A mutantwas generated and also tested in GST pulldown and co-immunoprecipitationexperiments to test this assumption, but failed to further suppress 14-3-3 inter-action with IRS2 (data not shown).

In summary, forskolin could induce phosphorylation of Ser573 but this site


seems not to be a relevant second 14-3-3 binding site beside Ser1137/Ser1138 incAMP-dependent 14-3-3 binding to IRS2. Sequence alignments of different IRS2species revealed complete sequence accordance of the amino acids surroundingSer1137/Ser1138. A high degree of conservation may implicate an important rolefor Ser1137/Ser1138.

The data presented in this thesis show an increased IRS2 protein half-lifeupon forskolin stimulation in Flp-In HEK293 cells stably expressing either IRS2or GFP-IRS2 and also in primary hepatocytes. Crucial for this increased stabil-ity were Ser1137/Ser1138, because a double GFP-IRS2-S1137A/S1138A mutantdid not show increased protein stability upon forskolin stimulation (Figure 3.41).This is no direct proof that binding of IRS2 to 14-3-3 proteins was responsiblefor the increased half-life of IRS2 protein, but based on the present data it couldbe concluded that phosphorylation on Ser1137/Ser1138 is necessary for both, thecAMP-enhanced binding to 14-3-3 proteins and the stabilization of the protein.Other groups also reported increased protein stability associated with 14-3-3binding. Short et al. showed for p27 an AMPK-dependent phosphorylationof Thr197 which increased 14-3-3 binding and protein half-life [165]. The his-tone deacetylase HDAC7 associated with 14-3-3 proteins upon phosphorylationby calcium/calmodulin-dependent kinases, thus preventing ubiquitinylation andsubsequent degradation via the 26S proteasome [95]. Finally, Yang et al. pro-vided evidence for 14-3-3σ-dependent stabilization of the cell cycle checkpointprotein p53 [216]. 14-3-3 binding blocked Mdm2-dependent ubiquitinylation andnuclear export of p53.

The amount of IRS2 protein present in cells is dependent on mRNA expres-sion and protein stability [87]. The stability of the IRS2 protein is regulated byposttranslational modifications that designate the molecule for ubiquitinylationand subsequent proteasomal degradation [139, 140]. To designate a protein forproteasomal degradation 3 classes of enzymes are needed: ubiquitin-activatingenzyme, ubiquitin-conjugating enzyme and ubiquitin ligase [66]. The presenteddata suggest a protection from proteasomal degradation upon cAMP-mediated14-3-3 binding to IRS2. It is imaginable that this protein/protein interactionresults in a conformational change that obscures the recognition motif for anE3 ubiquitin ligase, thus protecting IRS2 from proteasomal degradation. Thishypothesis is supported by GlobPlot R© analysis that predicted the C-terminalpart of the IRS2 molecule to be unstructured. Unstructured regions can of-ten be found in proteins that serve as regulatory and signalling proteins [71],because these intrinsically unstructured regions are suitable for restructuringupon interaction with a binding partner. 14-3-3 proteins itself display a veryrigid structure and therefore can provide themselves as an anvil whereupon thebinding partner can be restructured [213]. Conformational change upon 14-3-3binding has been shown in several examples: the phosphatidylinositol 4-kinaseIIIβ is protected from dephosphorylation in the active site by 14-3-3 binding [63],14-3-3 binding to BAD (Bcl-2-associated death promoter) enhances accessibilityof Ser155 for phosphorylation by prosurvival kinases [29] and AANAT (serotoninN-acetyltransferase) shows an increased affinity for its substrates upon 14-3-3binding [119].

The experimental data indicate an additional mechanism for the upregula-tion of IRS2 protein levels in physiological conditions with elevated cAMP con-centrations. During fasting blood glucagon levels are increased and are known to


induce IRS2 mRNA expression through glucagon-dependent G protein-coupledincreases in cAMP levels, activation of PKA and subsequent activation of thetranscription factor CREB and its coactivator TORC2 [24, 76]. Of note, IRS2expression in the fasting state is important to negatively regulate glucose out-put from the liver to avoid hyperglycemia [24]. Increased IRS2 protein levels inhepatocytes during the fasting state ensure fast and effective signalling upon in-sulin stimulation in the transition from the fasted to the fed state. In pancreaticβ-cells IRS2 expression is crucial for growth and survival [65]. Increased bloodglucose levels in the refed state are followed by insulin secretion of the β-cell.Glucose itself can induce IRS2 mRNA and protein expression in β-cells by acalcium-dependent mechanism [97]. In addition, glucagon like peptide (GLP) 1is released into the systemic circulation from enteroendocrine cells during feed-ing and potentiates insulin secretion. GLP1 increases cAMP levels in β-cells,thus promoting β-cell viability via increased IRS2 expression [76]. The findingof increased IRS2 protein stability due to elevated cAMP levels could provide anadditional mechanism for the β-cell to protect itself from reduced IRS2 proteinlevels. Of note, Van de Velde et al. observed rapid gene expression of CREBtarget genes by forskolin stimulation that decreased to baseline levels after 4h,whereas elevated IRS2 protein expression was measurable by western blot upto 16 hours [193]. This discrepancy in decreased IRS2 gene activation but ele-vated IRS2 protein levels could be explained with the finding of increased IRS2protein stability due to cAMP-induced 14-3-3 binding to IRS2. To conclude,Ser1137/Ser1138 could be identified as PKA phosphorylation sites on IRS2 thatmediate 14-3-3 binding and regulate IRS2 protein stability. The increased IRS2protein stability upon elevated cAMP levels provides an additional mechanismto cAMP-induced IRS2 mRNA and subsequent protein expression to ensuresufficient amounts of IRS2 protein.

Chapter 5

Conclusion

IRS proteins represent a central signalling node in insulin/IGF-1 signalling andtyrosine as well as serine/threonine phosphorylation of IRS proteins modulatedownstream signalling with inhibitory or stimulatory consequences.

Serine/threonine phosphorylation of IRS2 was investigated in this thesis.With mass spectrometry a total of 24 phosphorylated residues were identified,among them Ser573 was identified within a 14-3-3 binding region. Anothermass spectrometric approach for the identification of cAMP/PKA-dependentserine/threonine residues revealed at least three candidate residues that wereinvestigated further, namely Ser1137, Ser1138 and Ser1163.

Insulin/IGF-1-dependent 14-3-3 binding to IRS2 is partially mediated bySer573, since the alanine mutant abrogated 14-3-3 binding in overlay assays.GST pulldown experiments indicated a second 14-3-3 binding site. Of note,to randomly mutate single serine/threonine residues to alanine for studies on14-3-3 binding might not be a successful strategy, when the second site has onlya weak affinity for 14-3-3 proteins there might be no differences observable inoverlay assays or GST pulldown and co-immunoprecipitation studies [213]. Aprobably reasonable approach would be further sequence alignments with evo-lutionary old species to find highly conserved residues as candidates for 14-3-3binding and to perform direct phosphorylation analysis of these sites by apply-ing phosphorylated peptides and corresponding non-phosphorylated peptides.In addition, the presented data suggest a negative role for Ser573 phosphory-lation in insulin/IGF-1 signalling, Ser573 is probably part of a negative feedbackloop for downregulation of Akt/PKB phosphorylation.

cAMP-dependent phosphorylation of IRS2, presumably mediated by PKA,represents a novel mechanism to upregulate IRS2 protein levels with possibleimplications for hepatic metabolism and β-cell survival, where IRS2 expression isessential for growth, survival and function [154, 65]. Stimulation with forskolinincreased the phosphorylation of Ser1137/Ser1138, 14-3-3 binding and proteinstability of IRS2, whereas the S1137A/S1138A mutant did not show increasedIRS2 protein stability. Whether the cAMP-dependent 14-3-3 binding to IRS2is responsible for the protein stabilizing effect needs to be proven.

99

100

Figure 5.1 summarizes the serine residues that were identified as phosphory-lation sites during this thesis. In addition, kinases and stimuli that probablylead to phosphorylation of Ser573 and Ser1137/Ser1138 of IRS2 and the postulatedbiological outcome is shown.

PH PTB N C KRLB

S573 S1137 S1138

other AGC kinases? Akt/PKB

TPA-dependent

PKA

kinases PKA

Akt/PKB phosphorylation IRS2 protein stability

cAMP

Figure 5.1: Kinases and stimuli that can phosphorylate Ser573 and Ser1137/Ser1138 of IRS2.

14-3-3 interaction with IRS2 was investigated in this thesis from a mecha-nistic point of view with further experiments to elucidate the biological functionof insulin/IGF-1-dependent and cAMP-dependent 14-3-3 binding to IRS2. Thenext steps would be to confirm the data from this thesis in primary cell cul-tures and phosphosite-specific knock-out mice. Special attendance should bepaid to the β-cell, since profound effects of the S1137A/S1138A mutant couldbe expected. First, experiments should be performed to see if the results of thisthesis in HEK293 cells with transient and stable expression can be confirmedfor example in the rat insulinoma cell line INS1.

The ongoing challenge to develop new strategies and pharmacological treat-ments for the prevention and treatment of insulin resistance and T2D needs afull understanding of the cellular pathways and their dysregulation in pathologicstates. Characterization of the 14-3-3 binding to IRS2 is another step in theunderstanding of the cellular network and potentially opens new possibilities indrug development for the treatment of insulin resistance and T2D.

Bibliography

[1] D. Accili, J. Drago, E. J. Lee, M. D. Johnson, M. H. Cool, P. Salvatore,L. D. Asico, P. A. Jose, S. I. Taylor, and H. Westphal. “Early neonataldeath in mice homozygous for a null allele of the insulin receptor gene”.Nat. Genet. 12.1 (1996), pp. 106–109.

[2] A. Agarwal-Mawal, H. Y. Qureshi, P. W. Cafferty, Z. Yuan, D. Han,R. Lin, and H. K. Paudel. “14-3-3 connects glycogen synthase kinase-3beta to tau within a brain microtubule-associated tau phosphorylationcomplex”. J. Biol. Chem. 278.15 (2003), pp. 12722–12728.

[3] V. Aguirre, T. Uchida, L. Yenush, R. Davis, and M. F. White. “The c-Jun NH(2)-terminal kinase promotes insulin resistance during associationwith insulin receptor substrate-1 and phosphorylation of Ser(307)”. J.Biol. Chem. 275.12 (2000), pp. 9047–9054.

[4] Z. Ahmed, B. J. Smith, K. Kotani, P. Wilden, and T. S. Pillay. “APS,an adapter protein with a PH and SH2 domain, is a substrate for theinsulin receptor kinase”. Biochem. J. 341.Pt 3 (1999), pp. 665–668.

[5] A. Aitken. “Functional specificity in 14-3-3 isoform interactions throughdimer formation and phosphorylation. Chromosome location of mam-malian isoforms and variants”. Plant Mol. Biol. 50.6 (2002), pp. 993–1010.

[6] A. Aitken. “Post-translational modification of 14-3-3 isoforms and regu-lation of cellular function”. Semin. Cell Dev. Biol. 22.7 (2011), pp. 673–680.

[7] A. Aitken, H. Baxter, T. Dubois, S. Clokie, S. Mackie, K. Mitchell, A. Pe-den, and E. Zemlickova. “Specificity of 14-3-3 isoform dimer interactionsand phosphorylation”. Biochem. Soc. Trans. 30.4 (2002), pp. 351–360.

[8] A. Aitken, S. Howell, D. Jones, J. Madrazo, and Y. Patel. “14-3-3 alphaand delta are the phosphorylated forms of raf-activating 14-3-3 beta andzeta. In vivo stoichiometric phosphorylation in brain at a Ser-Pro-Glu-Lys MOTIF”. J. Biol. Chem. 270.11 (1995), pp. 5706–5709.

[9] M. Anai, H. Ono, M. Funaki, Y. Fukushima, K. Inukai, T. Ogihara, H.Sakoda, Y. Onishi, Y. Yazaki, M. Kikuchi, Y. Oka, and T. Asano. “Dif-ferent subcellular distribution and regulation of expression of insulin re-ceptor substrate (IRS)-3 from those of IRS-1 and IRS-2”. J. Biol. Chem.273.45 (1998), pp. 29686–29692.

101

BIBLIOGRAPHY 102

[10] E. Araki, M. A. Lipes, M. E. Patti, J. C. Bruning, B. I. I. I. Haag, R. S.Johnson, and C. R. Kahn. “Alternative pathway of insulin signallingin mice with targeted disruption of the IRS-1 gene”. Nature. 372.6502(1994), pp. 186–190.

[11] T. Asano, M. Fujishiro, A. Kushiyama, Y. Nakatsu, M. Yoneda, H. Ka-mata, and H. Sakoda. “Role of phosphatidylinositol 3-kinase activationon insulin action and its alteration in diabetic conditions”. Biol. Pharm.Bull. 30.9 (2007), pp. 1610–1616.

[12] J. M. Backer, M. G. J. Myers, S. E. Shoelson, D. J. Chin, X. J. Sun,M. Miralpeix, P. Hu, B. Margolis, E. Y. Skolnik, and J. Schlessinger.“Phosphatidylinositol 3’-kinase is activated by association with IRS-1during insulin stimulation”. EMBO J. 11.9 (1992), pp. 3469–3479.

[13] E. M. Bailyes, B. T. Nave, M. A. Soos, S. R. Orr, A. C. Hayward, and K.Siddle. “Insulin receptor/IGF-I receptor hybrids are widely distributedin mammalian tissues: quantification of individual receptor species byselective immunoprecipitation and immunoblotting”. Biochem. J. 327.Pt1 (1997), pp. 209–215.

[14] J. Bain, L. Plater, M. Elliott, N. Shpiro, J. Hastie, H. McLauchlan, I.Klevernic, J. S. Arthur, D. Alessi, and P. Cohen. “The selectivity ofprotein kinase inhibitors: a further update”. Biochem. J. 408 (2007),pp. 297–315.

[15] E. F. Barry, F. A. Felquer, J. A. Powell, L. Biggs, F. C. Stomski, A.Urbani, H. Ramshaw, P. Hoffmann, M. C. Wilce, M. A. Grimbaldeston,A. F. Lopez, and M. A. Guthridge. “14-3-3:Shc scaffolds integrate phos-phoserine and phosphotyrosine signaling to regulate phosphatidylinosi-tol 3-kinase activation and cell survival”. J. Biol. Chem. 284.18 (2009),pp. 12080–12090.

[16] A. Barthel and D. Schmoll. “Novel concepts in insulin regulation of hep-atic gluconeogenesis”. Am. J. Physiol. Endocrinol. Metab. 285.4 (2003),E685–E692.

[17] T. Benzing, M. Kottgen, M. Johnson, B. Schermer, H. Zentgraf, G. Walz,and E. Kim. “Interaction of 14-3-3 protein with regulator of G proteinsignaling 7 is dynamically regulated by tumor necrosis factor-alpha”. J.Biol. Chem. 277.36 (2002), pp. 32954–32962.

[18] A. Benzinger, N. Muster, H. B. Koch, J. R. I. I. I. Yates, and H. Hermek-ing. “Targeted proteomic analysis of 14-3-3 sigma, a p53 effector com-monly silenced in cancer”. Mol. Cell Proteomics. 4.6 (2005), pp. 785–795.

[19] P. Bevan. “Insulin signalling”. J. Cell Sci. 114.Pt 8 (2001), pp. 1429–1430.

[20] S. Boura-Halfon and Y. Zick. “Phosphorylation of IRS proteins, insulinaction, and insulin resistance”. AJP - Endocrinology and Metabolism296.4 (2009), E581–E591.

BIBLIOGRAPHY 103

[21] K. Bouzakri, M. Roques, P. Gual, S. Espinosa, F. Guebre-Egziabher,J. P. Riou, M. Laville, Y. Le Marchand-Brustel, J. F. Tanti, and H. Vidal.“Reduced activation of phosphatidylinositol-3 kinase and increased serine636 phosphorylation of insulin receptor substrate-1 in primary culture ofskeletal muscle cells from patients with type 2 diabetes”. Diabetes. 52.6(2003), pp. 1319–1325.

[22] S. Braselmann and F. McCormick. “Bcr and Raf form a complex in vivovia 14-3-3 proteins”. EMBO J. 14.19 (1995), pp. 4839–4848.

[23] D. Cai, S. Dhe-Paganon, P. A. Melendez, J. Lee, and S. E. Shoelson. “Twonew substrates in insulin signaling, IRS5/DOK4 and IRS6/DOK5”. J.Biol. Chem. 278.28 (2003), pp. 25323–25330.

[24] G. Canettieri, S. H. Koo, R. Berdeaux, J. Heredia, S. Hedrick, X. Zhang,and M. Montminy. “Dual role of the coactivator TORC2 in modulatinghepatic glucose output and insulin signaling”. Cell Metab. 2.5 (2005),pp. 331–338.

[25] M. Chaudhri, M. Scarabel, and A. Aitken. “Mammalian and yeast 14-3-3 isoforms form distinct patterns of dimers in vivo”. Biochem. Biophys.Res. Commun. 300.3 (2003), pp. 679–685.

[26] A. Craparo, R. Freund, and T. A. Gustafson. “14-3-3 (epsilon) interactswith the insulin-like growth factor I receptor and insulin receptor sub-strate I in a phosphoserine-dependent manner”. J. Biol. Chem. 272.17(1997), pp. 11663–11669.

[27] X. Cullere, P. Rose, U. Thathamangalam, A. Chatterjee, K. P. Mullane,D. C. Pallas, T. L. Benjamin, T. M. Roberts, and B. S. Schaffhausen.“Serine 257 phosphorylation regulates association of polyomavirus mid-dle T antigen with 14-3-3 proteins”. J. Virol. 72.1 (1998), pp. 558–563.

[28] C. D’Alessandris, R. Lauro, I. Presta, and G. Sesti. “C-reactive proteininduces phosphorylation of insulin receptor substrate-1 on Ser307 and Ser612 in L6 myocytes, thereby impairing the insulin signalling pathway thatpromotes glucose transport”. Diabetologia. 50.4 (2007), pp. 840–849.

[29] S. R. Datta, A. Katsov, L. Hu, A. Petros, S. W. Fesik, M. B. Yaffe,and M. E. Greenberg. “14-3-3 proteins and survival kinases cooperate toinactivate BAD by BH3 domain phosphorylation”. Mol. Cell. 6.1 (2000),pp. 41–51.

[30] S. P. Davies, H. Reddy, M. Caivano, and P. Cohen. “Specificity andmechanism of action of some commonly used protein kinase inhibitors”.Biochem. J. 351.Pt 1 (2000), pp. 95–105.

[31] R. J. de, F. J. Zwartkruis, M. H. Verheijen, R. H. Cool, S. M. Nijman,A. Wittinghofer, and J. L. Bos. “Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP”. Nature. 396.6710(1998), pp. 474–477.

[32] R. A. DeFronzo. “Insulin resistance: a multifaceted syndrome responsiblefor NIDDM, obesity, hypertension, dyslipidaemia and atherosclerosis”.Neth. J. Med. 50.5 (1997), pp. 191–197.

[33] X. Dong, S. Park, X. Lin, K. Copps, X. Yi, and M. F. White. “Irs1 andIrs2 signaling is essential for hepatic glucose homeostasis and systemicgrowth”. J. Clin. Invest. 116.1 (2006), pp. 101–114.

BIBLIOGRAPHY 104

[34] X. Du, J. E. Fox, and S. Pei. “Identification of a binding sequence for the14-3-3 protein within the cytoplasmic domain of the adhesion receptor,platelet glycoprotein Ib alpha”. J. Biol. Chem. 271.13 (1996), pp. 7362–7367.

[35] F. Dubois, F. Vandermoere, A. Gernez, J. Murphy, R. Toth, S. Chen,K. M. Geraghty, N. A. Morrice, and C. Mackintosh. “Differential 14-3-3affinity capture reveals new downstream targets of phosphatidylinositol3-kinase signaling”. Mol. Cell Proteomics. 8.11 (2009), pp. 2487–2499.

[36] T. Dubois, C. Rommel, S. Howell, U. Steinhussen, Y. Soneji, N. Morrice,K. Moelling, and A. Aitken. “14-3-3 is phosphorylated by casein kinase Ion residue 233. Phosphorylation at this site in vivo regulates Raf/14-3-3interaction”. J. Biol. Chem. 272.46 (1997), pp. 28882–28888.

[37] H. M. El Shewy, M. H. Lee, L. M. Obeid, A. A. Jaffa, and L. M. Luttrell.“The insulin-like growth factor type 1 and insulin-like growth factor type2/mannose-6-phosphate receptors independently regulate ERK1/2 activ-ity in HEK293 cells”. J. Biol. Chem. 282.36 (2007), pp. 26150–26157.

[38] J. M. Enserink, A. E. Christensen, R. J. de, T. M. van, F. Schwede, H. G.Genieser, S. O. Doskeland, J. L. Blank, and J. L. Bos. “A novel Epac-specific cAMP analogue demonstrates independent regulation of Rap1and ERK”. Nat. Cell Biol. 4.11 (2002), pp. 901–906.

[39] R. M. Ewing et al. “Large-scale mapping of human protein-protein in-teractions by mass spectrometry”. Mol. Syst. Biol. 3 (2007), p. 89.

[40] V. R. Fantin, B. E. Lavan, Q. Wang, N. A. Jenkins, D. J. Gilbert, N. G.Copeland, S. R. Keller, and G. E. Lienhard. “Cloning, tissue expres-sion, and chromosomal location of the mouse insulin receptor substrate4 gene”. Endocrinology. 140.3 (1999), pp. 1329–1337.

[41] V. R. Fantin, Q. Wang, G. E. Lienhard, and S. R. Keller. “Mice lack-ing insulin receptor substrate 4 exhibit mild defects in growth, repro-duction, and glucose homeostasis”. Am. J. Physiol. Endocrinol. Metab.278.1 (2000), E127–E133.

[42] A. T. Ferguson, E. Evron, C. B. Umbricht, T. K. Pandita, T. A. Chan, H.Hermeking, J. R. Marks, A. R. Lambers, P. A. Futreal, M. R. Stampfer,and S. Sukumar. “High frequency of hypermethylation at the 14-3-3sigma locus leads to gene silencing in breast cancer”. Proc. Natl. Acad.Sci. U.S.A. 97.11 (2000), pp. 6049–6054.

[43] R. J. Ferl. “14-3-3 proteins: regulation of signal-induced events”. Physiol.Plant. 120.2 (2004), pp. 173–178.

[44] R. J. Ferl, M. S. Manak, and M. F. Reyes. “The 14-3-3s”. Genome Biol.3.7 (2002), REVIEWS 3010.1–3010.7.

[45] J. C. Florez. “Clinical review: the genetics of type 2 diabetes: a realisticappraisal in 2008”. J. Clin. Endocrinol. Metab. 93.12 (2008), pp. 4633–4642.

[46] F. M. Foster, C. J. Traer, S. M. Abraham, and M. J. Fry. “The phos-phoinositide (PI) 3-kinase family”. J. Cell Sci. 116.15 (2003), pp. 3037–3040.

BIBLIOGRAPHY 105

[47] I. Foucault, Y. C. Liu, A. Bernard, and M. Deckert. “The chaperoneprotein 14-3-3 interacts with 3BP2/SH3BP2 and regulates its adapterfunction”. J. Biol. Chem. 278.9 (2003), pp. 7146–7153.

[48] L. Fritsche, S. S. Neukamm, R. Lehmann, E. Kremmer, A. M. Hennige,A. Hunder-Gugel, M. Schenk, H. U. Haring, E. D. Schleicher, and C.Weigert. “Insulin-induced serine phosphorylation of IRS-2 via ERK1/2and mTOR: studies on the function of Ser675 and Ser907”. Am. J. Phys-iol. Endocrinol. Metab. 300.5 (2011), E824–E836.

[49] H. Fu, R. R. Subramanian, and S. C. Masters. “14-3-3 proteins: structure,function, and regulation”. Annu. Rev. Pharmacol. Toxicol. 40:617-47.(2000), pp. 617–647.

[50] A. K. Gardino, S. J. Smerdon, and M. B. Yaffe. “Structural determinantsof 14-3-3 binding specificities and regulation of subcellular localizationof 14-3-3-ligand complexes: a comparison of the X-ray crystal structuresof all human 14-3-3 isoforms”. Semin. Cancer Biol. 16.3 (2006), pp. 173–182.

[51] K. M. Geraghty, S. Chen, J. E. Harthill, A. F. Ibrahim, R. Toth, N. A.Morrice, F. Vandermoere, G. B. Moorhead, D. G. Hardie, and C. Mack-intosh. “Regulation of multisite phosphorylation and 14-3-3 binding ofAS160 in response to IGF-1, EGF, PMA and AICAR”. Biochem. J. 407.2(2007), pp. 231–241.

[52] S. Giacometti, L. Camoni, C. Albumi, S. Visconti, M. I. De Michelis, andP. Aducci. “Tyrosine phosphorylation inhibits the interaction of 14-3-3proteins with the plant plasma membrane H+-ATPase”. Plant Biol. 6.4(2004), pp. 422–431.

[53] J. Giraud, M. Haas, E. P. Feener, K. D. Copps, X. Dong, S. L. Dunn,and M. F. White. “Phosphorylation of Irs1 at Ser-522 inhibits insulinsignaling”. Mol. Endocrinol. 21.9 (2007), pp. 2294–2302.

[54] J. Giraud, R. Leshan, Y. H. Lee, and M. F. White. “Nutrient-dependentand insulin-stimulated phosphorylation of insulin receptor substrate-1on serine 302 correlates with increased insulin signaling”. J. Biol. Chem.279.5 (2004), pp. 3447–3454.

[55] C. J. Gloeckner, K. Boldt, A. Schumacher, R. Roepman, and M. Ueff-ing. “A novel tandem affinity purification strategy for the efficient isola-tion and characterisation of native protein complexes”. Proteomics. 7.23(2007), pp. 4228–4234.

[56] L. J. Goodyear, F. Giorgino, L. A. Sherman, J. Carey, R. J. Smith,and G. L. Dohm. “Insulin receptor phosphorylation, insulin receptorsubstrate-1 phosphorylation, and phosphatidylinositol 3-kinase activityare decreased in intact skeletal muscle strips from obese subjects”. J.Clin. Invest. 95.5 (1995), pp. 2195–2204.

[57] O. Grbovic, V. Jovic, S. Ruzdijic, V. Pejanovic, L. Rakic, and S. Kanazir.“8-Cl-cAMP affects glioma cell-cycle kinetics and selectively induces apop-tosis”. Cancer Invest. 20.7-8 (2002), pp. 972–982.

[58] Y. M. Gu, Y. H. Jin, J. K. Choi, K. H. Baek, C. Y. Yeo, and K. Y. Lee.“Protein kinase A phosphorylates and regulates dimerization of 14-3-3epsilon”. FEBS Lett. 580.1 (2006), pp. 305–310.

BIBLIOGRAPHY 106

[59] P. Gual, T. Gremeaux, T. Gonzalez, Y. Marchand-Brustel, and J. F.Tanti. “MAP kinases and mTOR mediate insulin-induced phosphory-lation of insulin receptor substrate-1 on serine residues 307, 612 and632”. Diabetologia. 46.11 (2003), pp. 1532–1542.

[60] D. Gurevitch, S. Boura-Halfon, R. Isaac, G. Shahaf, M. Alberstein, D.Ronen, E. C. Lewis, and Y. Zick. “Elimination of negative feedback con-trol mechanisms along the insulin signaling pathway improves beta-cellfunction under stress”. Diabetes. 59.9 (2010), pp. 2188–2197.

[61] M. C. Harbeck, D. C. Louie, J. Howland, B. A. Wolf, and P. L. Rothen-berg. “Expression of insulin receptor mRNA and insulin receptor sub-strate 1 in pancreatic islet beta-cells”. Diabetes. 45.6 (1996), pp. 711–717.

[62] T. Haruta, T. Uno, J. Kawahara, A. Takano, K. Egawa, P. M. Sharma,J. M. Olefsky, and M. Kobayashi. “A rapamycin-sensitive pathway down-regulates insulin signaling via phosphorylation and proteasomal degra-dation of insulin receptor substrate-1”. Mol. Endocrinol. 14.6 (2000),pp. 783–794.

[63] A. Hausser, G. Link, M. Hoene, C. Russo, O. Selchow, and K. Pfizen-maier. “Phospho-specific binding of 14-3-3 proteins to phosphatidylinosi-tol 4-kinase III beta protects from dephosphorylation and stabilizes lipidkinase activity”. J. Cell Sci. 119.Pt 17 (2006), pp. 3613–3621.

[64] W. He, A. Craparo, Y. Zhu, T. J. O’Neill, L. M. Wang, J. H. Pierce,and T. A. Gustafson. “Interaction of insulin receptor substrate-2 (IRS-2) with the insulin and insulin-like growth factor I receptors. Evidencefor two distinct phosphotyrosine-dependent interaction domains withinIRS-2”. J. Biol. Chem. 271.20 (1996), pp. 11641–11645.

[65] A. M. Hennige, D. J. Burks, U. Ozcan, R. N. Kulkarni, J. Ye, S. Park,M. Schubert, T. L. Fisher, M. A. Dow, R. Leshan, M. Zakaria, M. Mossa-Basha, and M. F. White. “Upregulation of insulin receptor substrate-2 inpancreatic beta cells prevents diabetes”. J. Clin. Invest. 112.10 (2003),pp. 1521–1532.

[66] A. Hershko and A. Ciechanover. “The ubiquitin system”. Annu. Rev.Biochem. 67:425-79. (1998), pp. 425–479.

[67] V. C. Hodgkinson, V. Agarwal, D. ELFadl, J. N. Fox, P. L. McManus,T. K. Mahapatra, P. J. Kneeshaw, P. J. Drew, M. J. Lind, and L. Cawk-well. “Pilot and feasibility study: comparative proteomic analysis by 2-DE MALDI TOF/TOF MS reveals 14-3-3 proteins as putative biomark-ers of response to neoadjuvant chemotherapy in ER-positive breast can-cer”. J. Proteomics. 75.9 (2012), pp. 2745–2752.

[68] M. Holgado-Madruga, D. R. Emlet, D. K. Moscatello, A. K. Godwin, andA. J. Wong. “A Grb2-associated docking protein in EGF- and insulin-receptor signalling”. Nature. 379.6565 (1996), pp. 560–564.

[69] K. Horie-Inoue and S. Inoue. “Epigenetic and proteolytic inactivation of14-3-3sigma in breast and prostate cancers”. Semin. Cancer Biol. 16.3(2006), pp. 235–239.

BIBLIOGRAPHY 107

[70] S. R. Hubbard. “Crystal structure of the activated insulin receptor tyro-sine kinase in complex with peptide substrate and ATP analog”. EMBOJ. 16.18 (1997), pp. 5572–5581.

[71] L. M. Iakoucheva, C. J. Brown, J. D. Lawson, Z. Obradovic, and A. K.Dunker. “Intrinsic disorder in cell-signaling and cancer-associated pro-teins”. J. Mol. Biol. 323.3 (2002), pp. 573–584.

[72] T. Ichimura, T. Isobe, T. Okuyama, N. Takahashi, K. Araki, R. Kuwano,and Y. Takahashi. “Molecular cloning of cDNA coding for brain-specific14-3-3 protein, a protein kinase-dependent activator of tyrosine and tryp-tophan hydroxylases”. Proc.Nat. Acad. Sci. U.S.A. 85.19 (1988), pp. 7084–7088.

[73] T. Ide, H. Shimano, N. Yahagi, T. Matsuzaka, M. Nakakuki, T. Ya-mamoto, Y. Nakagawa, A. Takahashi, H. Suzuki, H. Sone, H. Toyoshima,A. Fukamizu, and N. Yamada. “SREBPs suppress IRS-2-mediated insulinsignalling in the liver”. Nat. Cell Biol. 6.4 (2004), pp. 351–357.

[74] K. Inukai, M. Anai, B. E. Van, T. Hosaka, H. Katagiri, M. Funaki, Y.Fukushima, T. Ogihara, Y. Yazaki, Kikuchi, Y. Oka, and T. Asano. “Anovel 55-kDa regulatory subunit for phosphatidylinositol 3-kinase struc-turally similar to p55PIK Is generated by alternative splicing of thep85alpha gene”. J. Biol. Chem. 271.10 (1996), pp. 5317–5320.

[75] K. Inukai, M. Funaki, T. Ogihara, H. Katagiri, A. Kanda, M. Anai, Y.Fukushima, T. Hosaka, M. Suzuki, B. C. Shin, K. Takata, Y. Yazaki, M.Kikuchi, Y. Oka, and T. Asano. “p85alpha gene generates three isoformsof regulatory subunit for phosphatidylinositol 3-kinase (PI 3-Kinase),p50alpha, p55alpha, and p85alpha, with different PI 3-kinase activityelevating responses to insulin”. J. Biol. Chem. 272.12 (1997), pp. 7873–7882.

[76] U. S. Jhala, G. Canettieri, R. A. Screaton, R. N. Kulkarni, S. Krajew-ski, J. Reed, J. Walker, X. Lin, M. White, and M. Montminy. “cAMPpromotes pancreatic beta-cell survival via CREB-mediated induction ofIRS2”. Genes Dev. 17.13 (2003), pp. 1575–1580.

[77] J. Jin, F. D. Smith, C. Stark, C. D. Wells, J. P. Fawcett, S. Kulkarni,P. Metalnikov, P. O’Donnell, P. Taylor, L. Taylor, A. Zougman, J. R.Woodgett, L. K. Langeberg, J. D. Scott, and T. Pawson. “Proteomic,functional, and domain-based analysis of in vivo 14-3-3 binding proteinsinvolved in cytoskeletal regulation and cellular organization”. Curr. Biol.14.16 (2004), pp. 1436–1450.

[78] W. Jin and M. E. Patti. “Genetic determinants and molecular pathwaysin the pathogenesis of Type 2 diabetes”. Clin. Sci. 116.2 (2009), pp. 99–111.

[79] C. Johnson, S. Crowther, M. J. Stafford, D. G. Campbell, R. Toth, and C.Mackintosh. “Bioinformatic and experimental survey of 14-3-3-bindingsites”. Biochem. J. 427.1 (2010), pp. 69–78.

[80] R. L. Joshi, B. Lamothe, N. Cordonnier, K. Mesbah, E. Monthioux,J. Jami, and D. Bucchini. “Targeted disruption of the insulin receptorgene in the mouse results in neonatal lethality”. EMBO J. 15.7 (1996),pp. 1542–1547.

BIBLIOGRAPHY 108

[81] H. Kanety, R. Feinstein, M. Z. Papa, R. Hemi, and A. Karasik. “Tu-mor necrosis factor alpha-induced phosphorylation of insulin receptorsubstrate-1 (IRS-1). Possible mechanism for suppression of insulin-stimu-lated tyrosine phosphorylation of IRS-1”. J. Biol. Chem. 270.40 (1995),pp. 23780–23784.

[82] M. Katz, I. Amit, and Y. Yarden. “Regulation of MAPKs by growthfactors and receptor tyrosine kinases”. Biochim. Biophys. Acta. 1773.8(2007), pp. 1161–1176.

[83] H. Kawasaki, G. M. Springett, N. Mochizuki, S. Toki, M. Nakaya, M.Matsuda, D. E. Housman, and A. M. Graybiel. “A family of cAMP-binding proteins that directly activate Rap1”. Science. 282.5397 (1998),pp. 2275–2279.

[84] E. Kjarland, T. J. Keen, and R. Kleppe. “Does isoform diversity ex-plain functional differences in the 14-3-3 protein family?” Curr. Pharm.Biotechnol. 7.3 (2006), pp. 217–223.

[85] A. Kosaki, K. Yamada, J. Suga, A. Otaka, and H. Kuzuya. “14-3-3betaprotein associates with insulin receptor substrate 1 and decreases insulin-stimulated phosphatidylinositol 3’-kinase activity in 3T3L1 adipocytes”.J. Biol. Chem. 273.2 (1998), pp. 940–944.

[86] A. Kotronen, A. Seppala-Lindroos, R. Bergholm, and H. Yki-Jarvinen.“Tissue specificity of insulin resistance in humans: fat in the liver ratherthan muscle is associated with features of the metabolic syndrome”. Di-abetologia. 51.1 (2008), pp. 130–138.

[87] N. Kubota, T. Kubota, S. Itoh, H. Kumagai, H. Kozono, I. Takamoto, T.Mineyama, H. Ogata, K. Tokuyama, M. Ohsugi, T. Sasako, M. Moroi, K.Sugi, S. Kakuta, Y. Iwakura, T. Noda, S. Ohnishi, R. Nagai, K. Tobe, Y.Terauchi, K. Ueki, and T. Kadowaki. “Dynamic functional relay betweeninsulin receptor substrate 1 and 2 in hepatic insulin signaling duringfasting and feeding”. Cell Metab. 8.1 (2008), pp. 49–64.

[88] N. Kubota, K. Tobe, Y. Terauchi, K. Eto, T. Yamauchi, R. Suzuki, Y.Tsubamoto, K. Komeda, R. Nakano, H. Miki, S. Satoh, H. Sekihara, S.Sciacchitano, M. Lesniak, S. Aizawa, R. Nagai, S. Kimura, Y. Akanuma,S. I. Taylor, and T. Kadowaki. “Disruption of insulin receptor substrate2 causes type 2 diabetes because of liver insulin resistance and lack ofcompensatory beta-cell hyperplasia”. Diabetes. 49.11 (2000), pp. 1880–1889.

[89] B. E. Lavan, V. R. Fantin, E. T. Chang, W. S. Lane, S. R. Keller, andG. E. Lienhard. “A novel 160-kDa phosphotyrosine protein in insulin-treated embryonic kidney cells is a new member of the insulin receptorsubstrate family”. J. Biol. Chem. 272.34 (1997), pp. 21403–21407.

[90] M. F. Leal, D. Q. Calcagno, S. Demachki, P. P. Assumpcao, R. Cham-mas, R. R. Burbano, and M. A. Smith. “Clinical implication of 14-3-3 ep-silon expression in gastric cancer”. World J. Gastroenterol. 18.13 (2012),pp. 1531–1537.

BIBLIOGRAPHY 109

[91] C. H. Lee, W. Li, R. Nishimura, M. Zhou, A. G. Batzer, M. G. J. Myers,M. F. White, J. Schlessinger, and E. Y. Skolnik. “Nck associates withthe SH2 domain-docking protein IRS-1 in insulin-stimulated cells”. Proc.Natl. Acad. Sci. U.S.A. 90.24 (1993), pp. 11713–11717.

[92] D. H. Lee, J. Shi, N. H. Jeoung, M. S. Kim, J. M. Zabolotny, S. W.Lee, M. F. White, L. Wei, and Y. B. Kim. “Targeted disruption ofROCK1 causes insulin resistance in vivo”. J. Biol. Chem. 284.18 (2009),pp. 11776–11780.

[93] S. Leng, W. Zhang, Y. Zheng, Z. Liberman, C. J. Rhodes, H. Eldar-Finkelman, and X. J. Sun. “Glycogen synthase kinase 3 beta mediateshigh glucose-induced ubiquitination and proteasome degradation of in-sulin receptor substrate 1”. J. Endocrinol. 206.2 (2010), pp. 171–181.

[94] J. Li, K. DeFea, and R. A. Roth. “Modulation of insulin receptor substrate-1 tyrosine phosphorylation by an Akt/phosphatidylinositol 3-kinase path-way”. J. Biol. Chem. 274.14 (1999), pp. 9351–9356.

[95] X. Li, S. Song, Y. Liu, S. H. Ko, and H. Y. Kao. “Phosphorylation of thehistone deacetylase 7 modulates its stability and association with 14-3-3proteins”. J. Biol. Chem. 279.33 (2004), pp. 34201–34208.

[96] Y. Li, T. J. Soos, X. Li, J. Wu, M. Degennaro, X. Sun, D. R. Littman,M. J. Birnbaum, and R. D. Polakiewicz. “Protein kinase C Theta inhibitsinsulin signaling by phosphorylating IRS1 at Ser(1101)”. J. Biol. Chem.279.44 (2004), pp. 45304–45307.

[97] M. K. Lingohr, I. Briaud, L. M. Dickson, J. F. McCuaig, C. Alarcon, B. L.Wicksteed, and C. J. Rhodes. “Specific regulation of IRS-2 expression byglucose in rat primary pancreatic islet beta-cells”. J. Biol. Chem. 281.23(2006), pp. 15884–15892.

[98] D. Liu, J. Bienkowska, C. Petosa, R. J. Collier, H. Fu, and R. Liddington.“Crystal structure of the zeta isoform of the 14-3-3 protein”. Nature.376.6536 (1995), pp. 191–194.

[99] J. P. Liu, J. Baker, A. S. Perkins, E. J. Robertson, and A. Efstratiadis.“Mice carrying null mutations of the genes encoding insulin-like growthfactor I (Igf-1) and type 1 IGF receptor (Igf1r)”. Cell. 75.1 (1993), pp. 59–72.

[100] M. Y. Liu, S. Cai, A. Espejo, M. T. Bedford, and C. L. Walker. “14-3-3 interacts with the tumor suppressor tuberin at Akt phosphorylationsite(s)”. Cancer Res. 62.22 (2002), pp. 6475–6480.

[101] S. C. Liu, Q. Wang, G. E. Lienhard, and S. R. Keller. “Insulin receptorsubstrate 3 is not essential for growth or glucose homeostasis”. J. Biol.Chem. 274.25 (1999), pp. 18093–18099.

[102] Y. Liu, J. F. Ross, P. V. Bodine, and J. Billiard. “Homodimerizationof Ror2 tyrosine kinase receptor induces 14-3-3(beta) phosphorylationand promotes osteoblast differentiation and bone formation”. Mol. En-docrinol. 21.12 (2007), pp. 3050–3061.

[103] L. Logie, A. J. Ruiz-Alcaraz, M. Keane, Y. L. Woods, J. Bain, R. Mar-quez, D. R. Alessi, and C. Sutherland. “Characterization of a proteinkinase B inhibitor in vitro and in insulin-treated liver cells”. Diabetes.56.9 (2007), pp. 2218–2227.

BIBLIOGRAPHY 110

[104] M. Luo, P. Langlais, Z. Yi, N. Lefort, E. A. De Filippis, H. Hwang,C. Y. Christ-Roberts, and L. J. Mandarino. “Phosphorylation of humaninsulin receptor substrate-1 at Serine 629 plays a positive role in insulinsignaling”. Endocrinology. 148.10 (2007), pp. 4895–4905.

[105] C. Mackintosh. “Dynamic interactions between 14-3-3 proteins and phos-phoproteins regulate diverse cellular processes”. Biochem. J. 381.Pt 2(2004), pp. 329–342.

[106] B. MacLean, D. M. Tomazela, N. Shulman, M. Chambers, G. L. Finney,B. Frewen, R. Kern, D. L. Tabb, D. C. Liebler, and M. J. MacCoss.“Skyline: an open source document editor for creating and analyzingtargeted proteomics experiments”. Bioinformatics. 26.7 (2010), pp. 966–968.

[107] M. C. Martin, I. Dransfield, C. Haslett, and A. G. Rossi. “Cyclic AMPregulation of neutrophil apoptosis occurs via a novel protein kinase A-independent signaling pathway”. J. Biol. Chem. 276.48 (2001), pp. 45041–45050.

[108] S. C. Masters, K. J. Pederson, L. Zhang, J. T. Barbieri, and H. Fu. “In-teraction of 14-3-3 with a nonphosphorylated protein ligand, exoenzymeS of Pseudomonas aeruginosa”. Biochemistry. 38.16 (1999), pp. 5216–5221.

[109] S. E. Meek, W. S. Lane, and H. Piwnica-Worms. “Comprehensive pro-teomic analysis of interphase and mitotic 14-3-3-binding proteins”. J.Biol. Chem. 279.31 (2004), pp. 32046–32054.

[110] D. E. Moller, A. Yokota, J. F. Caro, and J. S. Flier. “Tissue-specificexpression of two alternatively spliced insulin receptor mRNAs in man”.Mol. Endocrinol. 3.8 (1989), pp. 1263–1269.

[111] K. Morino, K. F. Petersen, S. Dufour, D. Befroy, J. Frattini, N. Shatzkes,S. Neschen, M. F. White, S. Bilz, S. Sono, M. Pypaert, and G. I. Shulman.“Reduced mitochondrial density and increased IRS-1 serine phosphory-lation in muscle of insulin-resistant offspring of type 2 diabetic parents”.J. Clin. Invest. 115.12 (2005), pp. 3587–3593.

[112] L. Mosthaf, K. Grako, T. J. Dull, L. Coussens, A. Ullrich, and D. A.McClain. “Functionally distinct insulin receptors generated by tissue-specific alternative splicing”. EMBO J. 9.8 (1990), pp. 2409–2413.

[113] I. Mothe and E. Van Obberghen. “Phosphorylation of insulin receptorsubstrate-1 on multiple serine residues, 612, 632, 662, and 731, modulatesinsulin action”. J. Biol. Chem. 271.19 (1996), pp. 11222–11227.

[114] A. J. Muslin, J. W. Tanner, P. M. Allen, and A. S. Shaw. “Interactionof 14-3-3 with signaling proteins is mediated by the recognition of phos-phoserine”. Cell. 84.6 (1996), pp. 889–897.

[115] M. G. J. Myers, R. Mendez, P. Shi, J. H. Pierce, R. Rhoads, and M. F.White. “The COOH-terminal tyrosine phosphorylation sites on IRS-1bind SHP-2 and negatively regulate insulin signaling”. J. Biol. Chem.273.41 (1998), pp. 26908–26914.

BIBLIOGRAPHY 111

[116] I. Nieto-Vazquez, S. Fernandez-Veledo, D. K. Kramer, R. Vila-Bedmar,L. Garcia-Guerra, and M. Lorenzo. “Insulin resistance associated to obe-sity: the link TNF-alpha”. Arch. Physiol. Biochem. 114.3 (2008), pp. 183–194.

[117] S. E. Nikoulina, T. P. Ciaraldi, S. Mudaliar, L. Carter, K. Johnson, andR. R. Henry. “Inhibition of glycogen synthase kinase 3 improves insulinaction and glucose metabolism in human skeletal muscle”. Diabetes. 51.7(2002), pp. 2190–2198.

[118] T. Obsil, R. Ghirlando, D. E. Anderson, A. B. Hickman, and F. Dyda.“Two 14-3-3 binding motifs are required for stable association of Fork-head transcription factor FOXO4 with 14-3-3 proteins and inhibition ofDNA binding”. Biochemistry. 42.51 (2003), pp. 15264–15272.

[119] T. Obsil, R. Ghirlando, D. C. Klein, S. Ganguly, and F. Dyda. “Crystalstructure of the 14-3-3zeta:serotonin N-acetyltransferase complex. a rolefor scaffolding in enzyme regulation”. Cell. 105.2 (2001), pp. 257–267.

[120] V. Obsilova, J. Silhan, E. Boura, J. Teisinger, and T. Obsil. “14-3-3proteins: a family of versatile molecular regulators”. Physiol. Res. 57Suppl 3:S11-21. Epub 2008 May 13 (2008), S11–S21.

[121] T. Ogihara, T. Isobe, T. Ichimura, M. Taoka, M. Funaki, H. Sakoda, Y.Onishi, K. Inukai, M. Anai, Y. Fukushima, M. Kikuchi, Y. Yazaki, Y.Oka, and T. Asano. “14-3-3 protein binds to insulin receptor substrate-1, one of the binding sites of which is in the phosphotyrosine bindingdomain”. J. Biol. Chem. 272.40 (1997), pp. 25267–25274.

[122] I. O’Kelly, M. H. Butler, N. Zilberberg, and S. A. Goldstein. “Forwardtransport. 14-3-3 binding overcomes retention in endoplasmic reticulumby dibasic signals”. Cell. 111.4 (2002), pp. 577–588.

[123] F. Oriente, F. Andreozzi, C. Romano, G. Perruolo, A. Perfetti, F. Fiory,C. Miele, F. Beguinot, and P. Formisano. “Protein kinase C-alpha regu-lates insulin action and degradation by interacting with insulin receptorsubstrate-1 and 14-3-3 epsilon”. J. Biol. Chem. 280.49 (2005), pp. 40642–40649.

[124] M. Otsu, I. Hiles, I. Gout, M. J. Fry, F. Ruiz-Larrea, G. Panayotou, A.Thompson, R. Dhand, J. Hsuan, N. Totty, and . “Characterization oftwo 85 kd proteins that associate with receptor tyrosine kinases, middle-T/pp60c-src complexes, and PI3-kinase”. Cell. 65.1 (1991), pp. 91–104.

[125] M. E. Patti, X. J. Sun, J. C. Bruening, E. Araki, M. A. Lipes, M. F.White, and C. R. Kahn. “4PS/insulin receptor substrate (IRS)-2 is thealternative substrate of the insulin receptor in IRS-1-deficient mice”. J.Biol. Chem. 270.42 (1995), pp. 24670–24673.

[126] K. Paz, Y. F. Liu, H. Shorer, R. Hemi, D. LeRoith, M. Quan, H. Kanety,R. Seger, and Y. Zick. “Phosphorylation of insulin receptor substrate-1(IRS-1) by protein kinase B positively regulates IRS-1 function”. J. Biol.Chem. 274.40 (1999), pp. 28816–28822.

[127] T. M. Pederson, D. L. Kramer, and C. M. Rondinone. “Serine/threoninephosphorylation of IRS-1 triggers its degradation: possible regulation bytyrosine phosphorylation”. Diabetes. 50.1 (2001), pp. 24–31.

BIBLIOGRAPHY 112

[128] C. Y. Peng, P. R. Graves, R. S. Thoma, Z. Wu, A. S. Shaw, and H.Piwnica-Worms. “Mitotic and G2 checkpoint control: regulation of 14-3-3 protein binding by phosphorylation of Cdc25C on serine-216”. Science.277.5331 (1997), pp. 1501–1505.

[129] L. Perego and G. Berruti. “Molecular cloning and tissue-specific expres-sion of the mouse homologue of the rat brain 14-3-3 theta protein: char-acterization of its cellular and developmental pattern of expression in themale germ line”. Mol. Reprod. Dev. 47.4 (1997), pp. 370–379.

[130] S. Pons, T. Asano, E. Glasheen, M. Miralpeix, Y. Zhang, T. L. Fisher,M. G. J. Myers, X. J. Sun, and M. F. White. “The structure and functionof p55PIK reveal a new regulatory subunit for phosphatidylinositol 3-kinase”. Mol. Cell Biol. 15.8 (1995), pp. 4453–4465.

[131] D. W. Powell, M. J. Rane, B. A. Joughin, R. Kalmukova, J. H. Hong,B. Tidor, W. L. Dean, W. M. Pierce, J. B. Klein, M. B. Yaffe, and K. R.McLeish. “Proteomic identification of 14-3-3zeta as a mitogen-activatedprotein kinase-activated protein kinase 2 substrate: role in dimer forma-tion and ligand binding”. Mol. Cell Biol. 23.15 (2003), pp. 5376–5387.

[132] R. M. Pozuelo, K. M. Geraghty, B. H. Wong, N. T. Wood, D. G. Camp-bell, N. Morrice, and C. Mackintosh. “14-3-3-affinity purification of over200 human phosphoproteins reveals new links to regulation of cellularmetabolism, proliferation and trafficking”. Biochem. J. 379.Pt 2 (2004),pp. 395–408.

[133] R. M. Pozuelo, M. Peggie, B. H. Wong, N. Morrice, and C. Mackintosh.“14-3-3s regulate fructose-2,6-bisphosphate levels by binding to PKB-phosphorylated cardiac fructose-2,6-bisphosphate kinase/phosphatase”.EMBO J. 22.14 (2003), pp. 3514–3523.

[134] S. F. Previs, D. J. Withers, J. M. Ren, M. F. White, and G. I. Shulman.“Contrasting effects of IRS-1 versus IRS-2 gene disruption on carbo-hydrate and lipid metabolism in vivo”. J. Biol. Chem. 275.50 (2000),pp. 38990–38994.

[135] K. M. Rice, M. A. Turnbow, and C. W. Garner. “Insulin stimulates thedegradation of IRS-1 in 3T3-L1 adipocytes”. Biochem. Biophys. Res.Commun. 190.3 (1993), pp. 961–967.

[136] K. Rittinger, J. Budman, J. Xu, S. Volinia, L. C. Cantley, S. J. Smerdon,S. J. Gamblin, and M. B. Yaffe. “Structural analysis of 14-3-3 phospho-peptide complexes identifies a dual role for the nuclear export signal of14-3-3 in ligand binding”. Mol. Cell. 4.2 (1999), pp. 153–166.

[137] R. A. Roth, G. Steele-Perkins, J. Hari, C. Stover, S. Pierce, J. Turner,J. C. Edman, and W. J. Rutter. “Insulin and insulin-like growth factorreceptors and responses”. Cold Spring Harb. Symp. Quant. Biol. 53 Pt1:537-43. (1988), pp. 537–543.

[138] P. L. Rothenberg, W. S. Lane, A. Karasik, J. Backer, M. White, andC. R. Kahn. “Purification and partial sequence analysis of pp185, themajor cellular substrate of the insulin receptor tyrosine kinase”. J. Biol.Chem. 266.13 (1991), pp. 8302–8311.

BIBLIOGRAPHY 113

[139] L. Rui, T. L. Fisher, J. Thomas, and M. F. White. “Regulation ofinsulin/insulin-like growth factor-1 signaling by proteasome-mediated de-gradation of insulin receptor substrate-2”. J. Biol. Chem. 276.43 (2001),pp. 40362–40367.

[140] L. Rui, M. Yuan, D. Frantz, S. Shoelson, and M. F. White. “SOCS-1and SOCS-3 block insulin signaling by ubiquitin-mediated degradationof IRS1 and IRS2”. J. Biol. Chem. 277.44 (2002), pp. 42394–42398.

[141] A. J. Ruiz-Alcaraz, H. K. Liu, D. J. Cuthbertson, E. J. McManus, S.Akhtar, C. Lipina, A. D. Morris, J. R. Petrie, H. S. Hundal, and C.Sutherland. “A novel regulation of IRS1 (insulin receptor substrate-1) ex-pression following short term insulin administration”. Biochem. J. 392.Pt2 (2005), pp. 345–352.

[142] M. J. Saad, F. Folli, E. Araki, N. Hashimoto, P. Csermely, and C. R.Kahn. “Regulation of insulin receptor, insulin receptor substrate-1 andphosphatidylinositol 3-kinase in 3T3-F442A adipocytes. Effects of dif-ferentiation, insulin, and dexamethasone”. Mol. Endocrinol. 8.5 (1994),pp. 545–557.

[143] M. J. Saad, F. Folli, and C. R. Kahn. “Insulin and dexamethasone reg-ulate insulin receptors, insulin receptor substrate-1, and phosphatidyli-nositol 3-kinase in Fao hepatoma cells”. Endocrinology. 136.4 (1995),pp. 1579–1588.

[144] C. Sakamoto, K. Suzuki, F. Hato, M. Akahori, T. Hasegawa, M. Hino, andS. Kitagawa. “Antiapoptotic effect of granulocyte colony-stimulating fac-tor, granulocyte-macrophage colony-stimulating factor, and cyclic AMPon human neutrophils: protein synthesis-dependent and protein synthesis-independent mechanisms and the role of the Janus kinase-STAT path-way”. Int. J. Hematol. 77.1 (2003), pp. 60–70.

[145] K. Sakamoto and G. D. Holman. “Emerging role for AS160/TBC1D4and TBC1D1 in the regulation of GLUT4 traffic”. Am. J. Physiol. En-docrinol. Metab. 295.1 (2008), E29–E37.

[146] H. Sakiyama, R. M. Wynn, W. R. Lee, M. Fukasawa, H. Mizuguchi, K. H.Gardner, J. J. Repa, and K. Uyeda. “Regulation of nuclear import/exportof carbohydrate response element-binding protein (ChREBP): interac-tion of an alpha-helix of ChREBP with the 14-3-3 proteins and regulationby phosphorylation”. J. Biol. Chem. 283.36 (2008), pp. 24899–24908.

[147] A. R. Saltiel and C. R. Kahn. “Insulin signalling and the regulation ofglucose and lipid metabolism”. Nature. 414.6865 (2001), pp. 799–806.

[148] T. Sandal, C. Stapnes, H. Kleivdal, L. Hedin, and S. O. Doskeland. “Anovel, extraneuronal role for cyclin-dependent protein kinase 5 (CDK5):modulation of cAMP-induced apoptosis in rat leukemia cells”. J. Biol.Chem. 277.23 (2002), pp. 20783–20793.

[149] M. M. Santoro, G. Gaudino, and P. C. Marchisio. “The MSP receptorregulates alpha6beta4 and alpha3beta1 integrins via 14-3-3 proteins inkeratinocyte migration”. Dev. Cell. 5.2 (2003), pp. 257–271.

[150] D. D. Sarbassov, D. A. Guertin, S. M. Ali, and D. M. Sabatini. “Phos-phorylation and regulation of Akt/PKB by the rictor-mTOR complex”.Science. 307.5712 (2005), pp. 1098–1101.

BIBLIOGRAPHY 114

[151] J. Satoh, T. Yamamura, and K. Arima. “The 14-3-3 protein epsilon iso-form expressed in reactive astrocytes in demyelinating lesions of multiplesclerosis binds to vimentin and glial fibrillary acidic protein in culturedhuman astrocytes”. Am. J. Pathol. 165.2 (2004), pp. 577–592.

[152] D. Sawka-Verhelle, V. Baron, I. Mothe, C. Filloux, M. F. White, andE. Van Obberghen. “Tyr624 and Tyr628 in insulin receptor substrate-2mediate its association with the insulin receptor”. J. Biol. Chem. 272.26(1997), pp. 16414–16420.

[153] D. Sawka-Verhelle, S. Tartare-Deckert, M. F. White, and E. Van Ob-berghen. “Insulin receptor substrate-2 binds to the insulin receptorthrough its phosphotyrosine-binding domain and through a newly iden-tified domain comprising amino acids 591-786”. J. Biol. Chem. 271.11(1996), pp. 5980–5983.

[154] G. T. Schuppin, S. Pons, S. Hugl, L. P. Aiello, G. L. King, M. White,and C. J. Rhodes. “A specific increased expression of insulin receptorsubstrate 2 in pancreatic beta-cell lines is involved in mediating serum-stimulated beta-cell growth”. Diabetes. 47.7 (1998), pp. 1074–1085.

[155] B. L. Seely, D. R. Reichart, Y. Takata, C. Yip, and J. M. Olefsky. “Afunctional assessment of insulin/insulin-like growth factor-I hybrid re-ceptors”. Endocrinology. 136.4 (1995), pp. 1635–1641.

[156] S. Seino and G. I. Bell. “Alternative splicing of human insulin recep-tor messenger RNA”. Biochem. Biophys. Res. Commun. 159.1 (1989),pp. 312–316.

[157] J. J. Senn, P. J. Klover, I. A. Nowak, and R. A. Mooney. “Interleukin-6induces cellular insulin resistance in hepatocytes”. Diabetes. 51.12 (2002),pp. 3391–3399.

[158] M. H. Shanik, Y. Xu, J. Skrha, R. Dankner, Y. Zick, and J. Roth. “In-sulin resistance and hyperinsulinemia: is hyperinsulinemia the cart or thehorse?” Diabetes Care. 31 Suppl 2:S262-8. (2008), S262–S268.

[159] H. Sharfi and H. Eldar-Finkelman. “Sequential phosphorylation of in-sulin receptor substrate-2 by glycogen synthase kinase-3 and c-Jun NH2-terminal kinase plays a role in hepatic insulin signaling”. Am. J. Physiol.Endocrinol. Metab. 294.2 (2008), E307–E315.

[160] S. Sharma, F. Barrett, J. Adamson, A. Todd, I. L. Megson, P. L. Zentler-Munro, and S. M. MacRury. “Diabetic fatty liver disease is associatedwith specific changes in blood-borne markers”. Diabetes Metab. Res. Rev.28.4 (2012), pp. 343–348.

[161] Y. H. Shen, J. Godlewski, A. Bronisz, J. Zhu, M. J. Comb, J. Avruch, andG. Tzivion. “Significance of 14-3-3 self-dimerization for phosphorylation-dependent target binding”. Mol. Biol. Cell. 14.11 (2003), pp. 4721–4733.

[162] P. R. Shepherd, D. J. Withers, and K. Siddle. “Phosphoinositide 3-kinase:the key switch mechanism in insulin signalling”. Biochem. J. 333.Pt 3(1998), pp. 471–490.

[163] S. Shikano, B. Coblitz, M. Wu, and M. Li. “14-3-3 proteins: regulation ofendoplasmic reticulum localization and surface expression of membraneproteins”. Trends Cell Biol. 16.7 (2006), pp. 370–375.

BIBLIOGRAPHY 115

[164] I. Shimomura, M. Matsuda, R. E. Hammer, Y. Bashmakov, M. S. Brown,and J. L. Goldstein. “Decreased IRS-2 and increased SREBP-1c lead tomixed insulin resistance and sensitivity in livers of lipodystrophic andob/ob mice”. Mol. Cell. 6.1 (2000), pp. 77–86.

[165] J. D. Short, R. Dere, K. D. Houston, S. L. Cai, J. Kim, J. M. Bergeron, J.Shen, J. Liang, M. T. Bedford, G. B. Mills, and C. L. Walker. “AMPK-mediated phosphorylation of murine p27 at T197 promotes binding of14-3-3 proteins and increases p27 stability”. Mol. Carcinog. 49.5 (2010),pp. 429–439.

[166] J. Silhan, V. Obsilova, J. Vecer, P. Herman, M. Sulc, J. Teisinger, and T.Obsil. “14-3-3 protein C-terminal stretch occupies ligand binding grooveand is displaced by phosphopeptide binding”. J. Biol. Chem. 279.47(2004), pp. 49113–49119.

[167] M. Simmgen, C. Knauf, M. Lopez, A. I. Choudhury, M. Charalambous, J.Cantley, D. C. Bedford, M. Claret, M. A. Iglesias, H. Heffron, P. D. Cani,A. Vidal-Puig, R. Burcelin, and D. J. Withers. “Liver-specific deletionof insulin receptor substrate 2 does not impair hepatic glucose and lipidmetabolism in mice”. Diabetologia. 49.3 (2006), pp. 552–561.

[168] R. D. Simoni, R. L. Hill, and M. Vaughan. “The Discovery of Insulin:the Work of Frederick Banting and Charles Best”. Journal of BiologicalChemistry 277.26 (2002), e15–e15.

[169] L. Simpson, J. Li, D. Liaw, I. Hennessy, J. Oliner, F. Christians, andR. Parsons. “PTEN expression causes feedback upregulation of insulinreceptor substrate 2”. Mol. Cell Biol. 21.12 (2001), pp. 3947–3958.

[170] E. Y. Skolnik, C. H. Lee, A. Batzer, L. M. Vicentini, M. Zhou, R.Daly, M. J. J. Myers, J. M. Backer, A. Ullrich, M. F. White, and .“The SH2/SH3 domain-containing protein GRB2 interacts with tyrosine-phosphorylated IRS1 and Shc: implications for insulin control of ras sig-nalling”. EMBO J. 12.5 (1993), pp. 1929–1936.

[171] E. M. Skoulakis and R. L. Davis. “14-3-3 proteins in neuronal develop-ment and function”. Mol. Neurobiol. 16.3 (1998), pp. 269–284.

[172] G. Solinas, W. Naugler, F. Galimi, M. S. Lee, and M. Karin. “Satu-rated fatty acids inhibit induction of insulin gene transcription by JNK-mediated phosphorylation of insulin-receptor substrates”. Proc. Natl.Acad. Sci. U.S.A. 103.44 (2006), pp. 16454–16459.

[173] M. A. Soos, J. Whittaker, R. Lammers, A. Ullrich, and K. Siddle. “Recep-tors for insulin and insulin-like growth factor-I can form hybrid dimers.Characterisation of hybrid receptors in transfected cells”. Biochem. J.270.2 (1990), pp. 383–390.

[174] R. R. Subramanian, S. C. Masters, H. Zhang, and H. Fu. “Functional con-servation of 14-3-3 isoforms in inhibiting bad-induced apoptosis”. Exp.Cell Res. 271.1 (2001), pp. 142–151.

[175] W. Suginta, N. Karoulias, A. Aitken, and R. H. Ashley. “Chloride intra-cellular channel protein CLIC4 (p64H1) binds directly to brain dynaminI in a complex containing actin, tubulin and 14-3-3 isoforms”. Biochem.J. 359.Pt 1 (2001), pp. 55–64.

BIBLIOGRAPHY 116

[176] X. J. Sun, S. Pons, T. Asano, M. G. J. Myers, E. Glasheen, and M. F.White. “The Fyn tyrosine kinase binds Irs-1 and forms a distinct signal-ing complex during insulin stimulation”. J. Biol. Chem. 271.18 (1996),pp. 10583–10587.

[177] X. J. Sun, P. Rothenberg, C. R. Kahn, J. M. Backer, E. Araki, P. A.Wilden, D. A. Cahill, B. J. Goldstein, and M. F. White. “Structure ofthe insulin receptor substrate IRS-1 defines a unique signal transductionprotein”. Nature. 352.6330 (1991), pp. 73–77.

[178] X. J. Sun, L. M. Wang, Y. Zhang, L. Yenush, M. G. J. Myers, E.Glasheen, W. S. Lane, J. H. Pierce, and M. F. White. “Role of IRS-2in insulin and cytokine signalling”. Nature. 377.6545 (1995), pp. 173–177.

[179] J. Sunayama, F. Tsuruta, N. Masuyama, and Y. Gotoh. “JNK antago-nizes Akt-mediated survival signals by phosphorylating 14-3-3”. J. CellBiol. 170.2 (2005), pp. 295–304.

[180] H. Tamemoto, T. Kadowaki, K. Tobe, T. Yagi, H. Sakura, T. Hayakawa,Y. Terauchi, K. Ueki, Y. Kaburagi, S. Satoh, and . “Insulin resistance andgrowth retardation in mice lacking insulin receptor substrate-1”. Nature.372.6502 (1994), pp. 182–186.

[181] C. M. Taniguchi, B. Emanuelli, and C. R. Kahn. “Critical nodes in sig-nalling pathways: insights into insulin action”. Nat. Rev. Mol. Cell Biol.7.2 (2006), pp. 85–96.

[182] C. M. Taniguchi, K. Ueki, and R. Kahn. “Complementary roles of IRS-1 and IRS-2 in the hepatic regulation of metabolism”. J. Clin. Invest.115.3 (2005), pp. 718–727.

[183] A. C. Tien, H. Y. Hsei, and C. T. Chien. “Dynamic expression andcellular localization of the drosophila 14-3-3epsilon during embryonic de-velopment”. Mech. Dev. 81.1-2 (1999), pp. 209–212.

[184] K. Tobe, H. Tamemoto, T. Yamauchi, S. Aizawa, Y. Yazaki, and T. Kad-owaki. “Identification of a 190-kDa protein as a novel substrate for theinsulin receptor kinase functionally similar to insulin receptor substrate-1”. J. Biol. Chem. 270.11 (1995), pp. 5698–5701.

[185] S. Tsunekawa, D. Demozay, I. Briaud, J. McCuaig, D. Accili, R. Stein,and C. J. Rhodes. “FoxO feedback control of basal IRS-2 expressionin pancreatic beta-cells is distinct from that in hepatocytes”. Diabetes.60.11 (2011), pp. 2883–2891.

[186] F. Tsuruta, J. Sunayama, Y. Mori, S. Hattori, S. Shimizu, Y. Tsuji-moto, K. Yoshioka, N. Masuyama, and Y. Gotoh. “JNK promotes Baxtranslocation to mitochondria through phosphorylation of 14-3-3 pro-teins”. EMBO J. 23.8 (2004), pp. 1889–1899.

[187] K. Tsuruzoe, R. Emkey, K. M. Kriauciunas, K. Ueki, and C. R. Kahn.“Insulin receptor substrate 3 (IRS-3) and IRS-4 impair IRS-1- and IRS-2-mediated signaling”. Mol. Cell Biol. 21.1 (2001), pp. 26–38.

[188] M. A. Turnbow, S. R. Keller, K. M. Rice, and C. W. Garner. “Dex-amethasone down-regulation of insulin receptor substrate-1 in 3T3-L1adipocytes”. J. Biol. Chem. 269.4 (1994), pp. 2516–2520.

BIBLIOGRAPHY 117

[189] A. Tzatsos and K. V. Kandror. “Nutrients suppress phosphatidylinositol3-kinase/Akt signaling via raptor-dependent mTOR-mediated insulin re-ceptor substrate 1 phosphorylation”. Mol. Cell Biol. 26.1 (2006), pp. 63–76.

[190] S. Ugi, H. Maegawa, A. Kashiwagi, M. Adachi, J. M. Olefsky, and R.Kikkawa. “Expression of dominant negative mutant SHPTP2 attenuatesphosphatidylinositol 3’-kinase activity via modulation of phosphorylationof insulin receptor substrate-1”. J. Biol. Chem. 271.21 (1996), pp. 12595–12602.

[191] T. Umahara, T. Uchihara, S. Koyama, T. Hashimoto, J. Akimoto, J.Haraoka, and T. Iwamoto. “Isoform-specific immunolocalization of 14-3-3 proteins in atherosclerotic lesions of human carotid and main cerebralarteries”. J. Neurol. Sci. 317.1-2 (2012), pp. 106–111.

[192] T. Urano, T. Saito, T. Tsukui, M. Fujita, T. Hosoi, M. Muramatsu,Y. Ouchi, and S. Inoue. “Efp targets 14-3-3 sigma for proteolysis andpromotes breast tumour growth”. Nature. 417.6891 (2002), pp. 871–875.

[193] d. V. Van, M. F. Hogan, and M. Montminy. “mTOR links incretin sig-naling to HIF induction in pancreatic beta cells”. Proc. Natl. Acad. Sci.U.S.A. 108.41 (2011), pp. 16876–16882.

[194] C. Vincenz and V. M. Dixit. “14-3-3 proteins associate with A20 in anisoform-specific manner and function both as chaperone and adaptermolecules”. J. Biol. Chem. 271.33 (1996), pp. 20029–20034.

[195] W. Wang and D. C. Shakes. “Molecular evolution of the 14-3-3 proteinfamily”. J. Mol. Evol. 43.4 (1996), pp. 384–398.

[196] M. Watanabe, T. Isobe, T. Ichimura, R. Kuwano, Y. Takahashi, and H.Kondo. “Molecular cloning of rat cDNAs for beta and gamma subtypes of14-3-3 protein and developmental changes in expression of their mRNAsin the nervous system”. Mol. Brain Res. 17.1-2 (1993), pp. 135–146.

[197] M. Watanabe, T. Isobe, T. Ichimura, R. Kuwano, Y. Takahashi, H.Kondo, and Y. Inoue. “Molecular cloning of rat cDNAs for the zeta andtheta subtypes of 14-3-3 protein and differential distributions of theirmRNAs in the brain”. Mol. Brain Res. 25.1-2 (1994), pp. 113–121.

[198] M. J. Waterman, E. S. Stavridi, J. L. Waterman, and T. D. Halazonetis.“ATM-dependent activation of p53 involves dephosphorylation and as-sociation with 14-3-3 proteins”. Nat. Genet. 19.2 (1998), pp. 175–178.

[199] C. R. Webster, P. Usechak, and M. S. Anwer. “cAMP inhibits bile acid-induced apoptosis by blocking caspase activation and cytochrome c re-lease”. Am. J. Physiol. Gastrointest. Liver Physiol. 283.3 (2002), G727–G738.

[200] C. Weigert, A. M. Hennige, T. Brischmann, A. Beck, K. Moeschel, M.Schauble, K. Brodbeck, H. U. Haring, E. D. Schleicher, and R. Lehmann.“The phosphorylation of Ser318 of insulin receptor substrate 1 is not perse inhibitory in skeletal muscle cells but is necessary to trigger the atten-uation of the insulin-stimulated signal”. J. Biol. Chem. 280.45 (2005),pp. 37393–37399.

BIBLIOGRAPHY 118

[201] C. Weigert, M. Kron, H. Kalbacher, A. K. Pohl, H. Runge, H. U. Har-ing, E. Schleicher, and R. Lehmann. “Interplay and effects of temporalchanges in the phosphorylation state of serine-302, -307, and -318 of in-sulin receptor substrate-1 on insulin action in skeletal muscle cells”. Mol.Endocrinol. 22.12 (2008), pp. 2729–2740.

[202] E. D. Werner, J. Lee, L. Hansen, M. Yuan, and S. E. Shoelson. “Insulinresistance due to phosphorylation of insulin receptor substrate-1 at serine302”. J. Biol. Chem. 279.34 (2004), pp. 35298–35305.

[203] M. F. White, J. N. Livingston, J. M. Backer, V. Lauris, T. J. Dull, A.Ullrich, and C. R. Kahn. “Mutation of the insulin receptor at tyrosine960 inhibits signal transmission but does not affect its tyrosine kinaseactivity”. Cell. 54.5 (1988), pp. 641–649.

[204] R. E. White, J. P. Kryman, A. M. El-Mowafy, G. Han, and G. O. Car-rier. “cAMP-dependent vasodilators cross-activate the cGMP-dependentprotein kinase to stimulate BK(Ca) channel activity in coronary arterysmooth muscle cells”. Circ. Res. 86.8 (2000), pp. 897–905.

[205] E. W. Wilker, R. A. Grant, S. C. Artim, and M. B. Yaffe. “A struc-tural basis for 14-3-3sigma functional specificity”. J. Biol. Chem. 280.19(2005), pp. 18891–18898.

[206] B. L. Williamson, J. Marchese, and N. A. Morrice. “Automated identifi-cation and quantification of protein phosphorylation sites by LC/MS ona hybrid triple quadrupole linear ion trap mass spectrometer”. Mol. CellProteomics 5.2 (2006), pp. 337–346.

[207] D. J. Withers, D. J. Burks, H. H. Towery, S. L. Altamuro, C. L. Flint,and M. F. White. “Irs-2 coordinates Igf-1 receptor-mediated beta-celldevelopment and peripheral insulin signalling”. Nat. Genet. 23.1 (1999),pp. 32–40.

[208] D. J. Withers, J. S. Gutierrez, H. Towery, D. J. Burks, J. M. Ren, S.Previs, Y. Zhang, D. Bernal, S. Pons, G. I. Shulman, S. Bonner-Weir,and M. F. White. “Disruption of IRS-2 causes type 2 diabetes in mice”.Nature. 391.6670 (1998), pp. 900–904.

[209] G. Wolf, T. Trub, E. Ottinger, L. Groninga, A. Lynch, M. F. White,M. Miyazaki, J. Lee, and S. E. Shoelson. “PTB domains of IRS-1 andShc have distinct but overlapping binding specificities”. J. Biol. Chem.270.46 (1995), pp. 27407–27410.

[210] J. Wu, Y. D. Tseng, C. F. Xu, T. A. Neubert, M. F. White, and S. R.Hubbard. “Structural and biochemical characterization of the KRLB re-gion in insulin receptor substrate-2”. Nat. Struct. Mol. Biol. 15.3 (2008),pp. 251–258.

[211] X. Xiang, M. Yuan, Y. Song, N. Ruderman, R. Wen, and Z. Luo. “14-3-3facilitates insulin-stimulated intracellular trafficking of insulin receptorsubstrate 1”. Mol. Endocrinol. 16.3 (2002), pp. 552–562.

[212] B. Xiao, S. J. Smerdon, D. H. Jones, G. G. Dodson, Y. Soneji, A. Aitken,and S. J. Gamblin. “Structure of a 14-3-3 protein and implications forcoordination of multiple signalling pathways”. Nature. 376.6536 (1995),pp. 188–191.

BIBLIOGRAPHY 119

[213] M. B. Yaffe. “How do 14-3-3 proteins work?– Gatekeeper phosphorylationand the molecular anvil hypothesis”. FEBS Lett. 513.1 (2002), pp. 53–57.

[214] M. B. Yaffe, K. Rittinger, S. Volinia, P. R. Caron, A. Aitken, H. Leffers,S. J. Gamblin, S. J. Smerdon, and L. C. Cantley. “The structural basisfor 14-3-3:phosphopeptide binding specificity”. Cell. 91.7 (1997), pp. 961–971.

[215] Y. Yamaguchi, J. S. Flier, A. Yokota, H. Benecke, J. M. Backer, and D. E.Moller. “Functional properties of two naturally occurring isoforms of thehuman insulin receptor in Chinese hamster ovary cells”. Endocrinology.129.4 (1991), pp. 2058–2066.

[216] H. Y. Yang, Y. Y. Wen, C. H. Chen, G. Lozano, and M. H. Lee. “14-3-3sigma positively regulates p53 and suppresses tumor growth”. Mol. CellBiol. 23.20 (2003), pp. 7096–7107.

[217] X. Yang, W. H. Lee, F. Sobott, E. Papagrigoriou, C. V. Robinson, J. G.Grossmann, M. Sundstrom, D. A. Doyle, and J. M. Elkins. “Structuralbasis for protein-protein interactions in the 14-3-3 protein family”. Proc.Natl. Acad. Sci. U.S.A. 103.46 (2006), pp. 17237–17242.

[218] Z. Yi, M. Luo, C. A. Carroll, S. T. Weintraub, and L. J. Mandarino.“Identification of phosphorylation sites in insulin receptor substrate-1 byhypothesis-driven high-performance liquid chromatography-electrosprayionization tandem mass spectrometry”. Anal Chem. 77.17 (2005),pp. 5693–5699.

[219] S. Yoon and R. Seger. “The extracellular signal-regulated kinase: multi-ple substrates regulate diverse cellular functions”. Growth Factors. 24.1(2006), pp. 21–44.

[220] K. Yoshida, T. Yamaguchi, T. Natsume, D. Kufe, and Y. Miki. “JNKphosphorylation of 14-3-3 proteins regulates nuclear targeting of c-Ablin the apoptotic response to DNA damage”. Nat. Cell Biol. 7.3 (2005),pp. 278–285.

[221] G. Zanusso, M. Fiorini, A. Farinazzo, M. Gelati, M. D. Benedetti, S. Fer-rari, L. A. Dalla, S. Capaldi, H. L. Monaco, N. Rizzuto, and S. Monaco.“Phosphorylated 14-3-3zeta protein in the CSF of neuroleptic-treatedpatients”. Neurology. 64.9 (2005), pp. 1618–1620.

[222] J. Zhang, Z. Gao, J. Yin, M. J. Quon, and J. Ye. “S6K directly phos-phorylates IRS-1 on Ser-270 to promote insulin resistance in responseto TNF-(alpha) signaling through IKK2”. J. Biol. Chem. 283.51 (2008),pp. 35375–35382.

[223] X. Zhao, L. Gan, H. Pan, D. Kan, M. Majeski, S. A. Adam, and T. G.Unterman. “Multiple elements regulate nuclear/cytoplasmic shuttling ofFOXO1: characterization of phosphorylation- and 14-3-3-dependent and-independent mechanisms”. Biochem. J. 378.Pt 3 (2004), pp. 839–849.

[224] E. Zmuda-Trzebiatowska, V. Manganiello, and E. Degerman. “Novelmechanisms of the regulation of protein kinase B in adipocytes; impli-cations for protein kinase A, Epac, phosphodiesterases 3 and 4”. CellSignal. 19.1 (2007), pp. 81–86.

List of Figures

1.1 Schematic presentation of IRS isoforms . . . . . . . . . . . . . . . 61.2 Collection of reported phosphorylated serine and tyrosine resi-

dues on IRS1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81.3 Collection of reported phosphorylated serine and tyrosine resi-

dues on IRS2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101.4 Illustration of the PI 3-kinase and Ras/MAPK pathway . . . . . 111.5 14-3-3 monomers β and τ as heterodimer . . . . . . . . . . . . . 121.6 14-3-3 cladogram . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

2.1 SDS gel stained with Coomassie G-250 to visualize the amountof prepared GST-14-3-3 proteins . . . . . . . . . . . . . . . . . . 46

3.1 Kinase panel assay . . . . . . . . . . . . . . . . . . . . . . . . . . 523.2 14-3-3 can bind to transiently and endogenously expressed IRS2 . 533.3 Co-immunoprecipitation of 14-3-3 and IRS2 . . . . . . . . . . . . 543.4 Interaction of 14-3-3 with IRS2 upon insulin stimulation in mice

liver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543.5 14-3-3 binding to IRS2 after refeeding and insulin stimulation . . 553.6 Coomassie stained SDS gel . . . . . . . . . . . . . . . . . . . . . 563.7 Sequence alignments of IRS1 and IRS2 . . . . . . . . . . . . . . . 573.8 Schematic illustration of truncated IRS2 constructs to identify

the 14-3-3 binding region . . . . . . . . . . . . . . . . . . . . . . 583.9 Cell lysates of IRS2 wild type and fragments separated by SDS-

PAGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 593.10 Overlay assay of IRS2 wild type and truncated versions of IRS2

to check for residual 14-3-3 binding . . . . . . . . . . . . . . . . . 593.11 Identification of Ser573 as 14-3-3 binding site by overlay assay . . 603.12 Confirmation of Ser573 of IRS2 as an IGF-1-dependent 14-3-3

binding site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 613.13 Specificity test of the p-Ser573 antibody . . . . . . . . . . . . . . 623.14 p-Ser573 antibody is specific for IRS2 . . . . . . . . . . . . . . . . 633.15 p-Ser573 antibody test in Fao cells . . . . . . . . . . . . . . . . . 643.16 Immunoprecipitation of IRS2 with p-Ser573 antibody . . . . . . . 643.17 Ser573 phosphorylation upon insulin stimulation in Fao cells . . . 653.18 Phosphorylation of Ser573 after inhibition of Akt/PKB . . . . . . 663.19 Akt/PKB is part of IGF-1-mediated 14-3-3 binding to IRS2 . . . 673.20 Ser573 influences phosphorylation of Akt/PKB . . . . . . . . . . 683.21 Ser573 is not the only IGF-1-dependent 14-3-3 binding site . . . . 69

120

LIST OF FIGURES 121

3.22 14-3-3 interacts with IRS2 upon elevated cAMP levels . . . . . . 703.23 PKA substrate antibody recognizes phosphorylation on IRS2 af-

ter forskolin stimulation . . . . . . . . . . . . . . . . . . . . . . . 723.24 IGF-1-induced phosphorylation of IRS2 is not recognized by PKA

substrate antibody . . . . . . . . . . . . . . . . . . . . . . . . . . 733.25 Akt/PKB has no influence on forskolin-stimulated phosphory-

lation of IRS2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 743.26 MAPK pathway kinases have no influence on forskolin-stimulated

phosphorylation of IRS2 . . . . . . . . . . . . . . . . . . . . . . . 743.27 Overlay assay from IRS2 wild type and fragments after forskolin

and H89 stimulation . . . . . . . . . . . . . . . . . . . . . . . . . 753.28 Cell lysates from transiently transfected IRS2 wild type and frag-

ments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 763.29 Detection of GFP-IRS2 wild type and fragments with PKA sub-

strate antibody . . . . . . . . . . . . . . . . . . . . . . . . . . . . 763.30 GST pulldowns from GFP-IRS2 wild type and fragments . . . . 773.31 Scheme of IRS2 fragments used for identification of the PKA-

dependent 14-3-3 binding region . . . . . . . . . . . . . . . . . . 783.32 Detection of GFP-IRS2 and fragments with PKA substrate anti-

body after GFP pulldown . . . . . . . . . . . . . . . . . . . . . . 783.33 GST pulldown of IRS2 wild type and fragments including frag-

ment 601-952 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 793.34 Coomassie staining of a SDS gel to cut bands for mass spectrometry 803.35 MS/MS analysis of a tryptic peptide . . . . . . . . . . . . . . . . 813.36 Mutation of Ser1137 and Ser1138 abolishes recognition by PKA

substrate antibody . . . . . . . . . . . . . . . . . . . . . . . . . . 823.37 Double mutant GFP-IRS2-S1137A/S1138A shows reduced bind-

ing to GST-14-3-3 . . . . . . . . . . . . . . . . . . . . . . . . . . 833.38 Co-immunoprecipitation of myc-14-3-3 and IRS2 . . . . . . . . . 843.39 IRS2 protein stability is increased upon forskolin stimulation in

Flp-In HEK293 cells . . . . . . . . . . . . . . . . . . . . . . . . . 853.40 Visualization of IRS2 protein degradation in primary hepatocytes 853.41 Ser1137 and Ser1138 are crucial for IRS2 protein stability . . . . . 86

5.1 Kinases/stimuli that phosphorylate Ser573 and Ser1137/Ser1138 . . 100

A.1 Phosphopeptides detected by mass spectrometry . . . . . . . . . 126A.2 Phosphopeptides detected by mass spectrometry that are unique

to IRS2 isolated from IGF-1-treated cells . . . . . . . . . . . . . 127A.3 Exemplary MS/MS spectrum and fragmentation table of the pep-

tide TY(pS)LTTPAR . . . . . . . . . . . . . . . . . . . . . . . . 127A.4 b- and y-ion series and fragmentation table of TY(pS)LTTPAR . 128A.5 Sequence alignment of mouse and human IRS1 and IRS2 . . . . 129

List of Tables

2.1 Chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.2 Consumables/Kits . . . . . . . . . . . . . . . . . . . . . . . . . . 232.3 Laboratory devices . . . . . . . . . . . . . . . . . . . . . . . . . . 252.4 Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272.5 Primary antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . 272.6 Secondary antibodies . . . . . . . . . . . . . . . . . . . . . . . . . 282.7 Components of stable transfection mix . . . . . . . . . . . . . . . 312.8 BBS solution (pH 6.96) . . . . . . . . . . . . . . . . . . . . . . . 312.9 Lysis buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322.10 Phosphatase Inhibitors . . . . . . . . . . . . . . . . . . . . . . . . 322.11 BSA concentrations used for standard curve . . . . . . . . . . . . 332.12 5×Lammli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332.13 Electrophoresis buffer . . . . . . . . . . . . . . . . . . . . . . . . 342.14 Stacking gel and separating gel buffer . . . . . . . . . . . . . . . 342.15 Pipetting scheme for separation and stacking gels . . . . . . . . . 342.16 Pipetting scheme for gradient gel . . . . . . . . . . . . . . . . . . 352.17 10× western blot buffer . . . . . . . . . . . . . . . . . . . . . . . 352.18 NET-G (pH 7.4) . . . . . . . . . . . . . . . . . . . . . . . . . . . 362.19 Solution A for ECL . . . . . . . . . . . . . . . . . . . . . . . . . . 362.20 Solution B for ECL . . . . . . . . . . . . . . . . . . . . . . . . . . 362.21 TBS-T buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372.22 Stripping buffer (pH 6.8) . . . . . . . . . . . . . . . . . . . . . . . 372.23 HNTG buffer (pH 7.5) . . . . . . . . . . . . . . . . . . . . . . . . 382.24 High salt washing buffer . . . . . . . . . . . . . . . . . . . . . . . 382.25 No salt washing buffer . . . . . . . . . . . . . . . . . . . . . . . . 382.26 Mutagenesis primers for fragments 1-300 and 1-600 . . . . . . . . 392.27 Mutagenesis primers for single point mutations . . . . . . . . . . 402.28 Mutagenesis primers for generation of a double and a truncated

mutant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402.29 Mutagenesis primers for generation of single point mutations . . 412.30 Pipetting scheme for site-directed mutagenesis . . . . . . . . . . 412.31 Gradient cycler protocol for site-directed mutagenesis . . . . . . 422.32 KCM solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422.33 Sequencing primers . . . . . . . . . . . . . . . . . . . . . . . . . . 432.34 Pipetting scheme for sequencing PCR . . . . . . . . . . . . . . . 432.35 Cycler programme for sequencing . . . . . . . . . . . . . . . . . . 432.36 Pipetting scheme for restriction digest . . . . . . . . . . . . . . . 442.37 50× TAE buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

122

LIST OF TABLES 123

2.38 10× gel loading buffer . . . . . . . . . . . . . . . . . . . . . . . . 442.39 TSB medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452.40 Lysis buffer for protein isolation of mouse liver (pH 7.6) . . . . . 492.41 Krebs Henseleit buffer I (pH 7.4), sterile filtered . . . . . . . . . . 502.42 Krebs Henseleit buffer II (pH 7.2), sterile filtered . . . . . . . . . 50

3.1 With mass spectrometry identified residues with surrounding aminoacids that correspond to a 14-3-3 binding motif. Shown are theidentified residues indicated in bold with pS or pT and the sur-rounding six amino acids. . . . . . . . . . . . . . . . . . . . . . . 60

3.2 Sequence alignments concerning the surrounding amino acids ofSer573. Sequence alignment of the amino acids adjacent to the14-3-3 binding site Ser573 of IRS2 in different species. Shown areIRS2 sequences from Homo sapiens (NP 003740.2), Mus muscu-lus (NP 001074681.1), Rattus norvegicus (NP 001162104.1) andXenopus laevis (AAH72768.1). . . . . . . . . . . . . . . . . . . . 62

3.3 Shown is the amino acid sequence surrounding Ser573 in mouseIRS2 and the sequence surrounding the homologue position Ser522

in mouse/rat IRS1. . . . . . . . . . . . . . . . . . . . . . . . . . . 623.4 Presentation of the recognition motifs of Akt/PKB, PKA and

14-3-3 (R - arginine; X - any amino acid; F - phenylalanine; Y -tyrosine; P - proline; φ - bulky, hydrophobic residue) . . . . . . . 71

3.5 Sequences surrounding the potential PKA phosphorylation sitesSer1137, Ser1138 and Ser1163 are shown according to mouse IRS2numbering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

3.6 Comparison of the recognition motif of the PKA substrate anti-body with the amino acids surrounding Ser1137 and Ser1138 . . . 81

3.7 Sequence alignment from different species in the area surround-ing Ser1137/Ser1138. Shown are the amino acid areas surround-ing Ser1137 and Ser1138 from IRS2 in different species: Homosapiens (NP 003740.2), Mus musculus (NP 001074681.1), Rattusnorvegicus (NP 001162104.1), Xenopus laevis (AAH72768.1). . . 82

4.1 Experimentally reported serine/threonine residues of IRS1 andIRS2, that were also identified in the mass spectrometric ap-proach for identification of IGF-1-dependent phosphorylation siteson mouse IRS2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

4.2 Reported homologue residues of mouse IRS2 Ser1137/Ser1138 . . . 96

Abbreviations

A alanine

A Angstrom

Ala alanine

APS adapter protein with a PH and SH2 domain

BSA bovine serum albumin

cAMP cyclic adenosine monophosphate

CREB cAMP response element binding protein

DNA desoxyribonucleic acid

DPBS Dulbecco’s phosphate buffered saline

ECL enhanced chemiluminescence

EDTA ethylenediaminetetraacetic acid

EGTA ethyleneglycol tetraacetic acid

ERK extracellular regulated kinase

FCS fetal calve serum

FoxO3a forkhead box class O

Gab1 Grb associated binder

GDP guanosine diphosphate

GIP glucose dependent insulinotropic polypeptide

GLP glucagon-like peptide

GLUT4 glucose transporter 4

Grb2 growth receptor bound 2

GSK glycogen synthase kinase

GTP guanosine triphosphate

H2O water

h hour

HEK human embryonic kidney cells

HPLC high performance liquid chromatography

HRP horseradish peroxidase

IGF-1 insulin like growth factor 1

124

ABBREVIATIONS 125

IR insulin receptor

IRS insulin receptor substrate

i.p. intraperitoneally

JNK c-Jun N-terminal kinase

kDa kilo Dalton

KRLB kinase regulatory loop binding

MAPK mitogen activated protein kinase

MAPKK mitogen activated protein kinase kinase

MAPKKK mitogen activated protein kinase kinase kinase

MEK MAP kinase ERK kinase

mTOR mammalian target of rapamycin

p phospho

PCR polymerase chain reaction

PDGF platelet derived growth factor

PDK phosphoinositide-dependent kinase

PEG polyethylene glycol

PH pleckstrin homology

PIP2 phosphatidylinositol (4, 5)-bisphosphate

PIP3 phosphatidylinositol (3, 4, 5)-trisphosphate

PI 3-kinase phosphatidylinositol 3-kinase

PKB protein kinase B

PKC protein kinase C

PTB phosphotyrosine binding

S serine

S6K1 S6 ribosomal protein kinase 1

SDS-PAGE sodium dodecylsulfate polyacrylamide gelelectrophoresis

Ser serine

SH2 Src homology domain 2

Shc Src homology s/?-collagen-related protein

SHIP2 SH2-containing inositol phosphatase 2

Sos son of sevenless

SREBP sterol regulatory element-binding protein

T threonine

T1D type I diabetes mellitus

T2D type II diabetes mellitus

TEMED N,N,N’,N’-tetramethylethylenediamine

Thr threonine

TNFα tumor necrosis factor α

TPA 12-O-tetradecanoylphorbol 13-acetate

Y tyrosine

Appendix A

Appendix

A.1 Phosphopeptides detected by mass spectrom-etry

residue m/z M sequence (ion score > 20) 66 971,9192 1941,8218 GPGTGGDEASAAGGpSPPQPPR (105) 303 1010,9786 2019,9375 SKpSQSSGSSATHPISVPGAR (107) 347 611,7894 1221,5642 TDSLAApTPPAAK (50) 385/388 1038,1247 3111,3482 TASEGDGGAAGGAGTAGGRPMSVAGpSPLpSPGPVR (98) 388 1011,4705 3031,3819 TASEGDGGAAGGAGTAGGRPMSVAGSPLpSPGPVR (97) 401 747,8305 1493,6446 SHpTLSAGCGGRPSK (57) 517 775,3495 1548,6820 SNpTPESIAETPPAR (55) 556 684,7994 1367,5830 RVpSGDGAQDLDR (62) 573 623,3026 1244,5914 RTYpSLTTPAR (38) 675 904,8615 1807,7045 SDDYMPmpSPTSVSAPK (54) 722 883,8343 1765,6502 ApSSPAESSPEDSGYmR (73) 727/728 915,8195 1829,6216 ASSPAEpSpSPEDSGYMR (60) 762 1522,1939 3041,3841 LLPNGDYLNMpSPSEAGTAGTPPDFSAALR (82) 907 903,3926 1804,7669 pSPGEYINIDFGEAGTR (72) 968 898,8778 1795,7375 SPLpSDYmNLDFSSPK (67) 977 930,8636 1859,7089 SPLpSDYMNLDFSpSPK (75) 999 1632,7888 4895,3365 SGDTVGSMDGLLpSPEASSPYPPLPPRPSTSPSSLQQPLPPAPGDLYR (50) 1089 470,2310 938,4474 VApSPTSGLK (61) 1151 1308,5862 2615,1541 HSSETFSSTTTVTPVpSPSFAHNSK (86) 1165 696,8173 1391,6194 HNSApSVENVSLR (84) 1190 1117,5376 3349,5827 SSEGSSTLGGGDEPPTpSPGQAQPLVAVPPVPQAR (101)

Figure A.1: GFP-IRS2 was expressed transiently in HEK293 cells and cells were either leftunstimulated or stimulated with IGF-1 for 30 minutes. Using GFP-Trap R© IRS-2 was purifiedfrom total cell lysate and after SDS-PAGE and Coomassie staining, the bands correspondingto IRS2 were cut and peptides were digested using trypsin (M = molecular mass of the peptidein Dalton, m/z = mass to charge). A mass accuracy of 10ppm was used in database searchesand phosphosites were identified by MS/MS analysis and manual inspection of the spectra.The sequence coverage was 92 %. The phosphorylated residues are indicated in bold with pSor pT.

126

A.1. PHOSPHOPEPTIDES DETECTED BY MASS SPECTROMETRY 127

Residue m/z M Sequence (ion score > 20) 305 903,4133 1804,8105 SQpSSGSSATHPISVPGAR (89) 1266 779,6802 2336,0182 GEQGSLAQSQPQPGDKNpSWSR (34)

Figure A.2: Phosphopeptides detected by mass spectrometry that are unique to IRS2 iso-lated from IGF-1-treated cells. Samples were prepared as in Figure A.1. The phosphorylatedresidues are indicated in bold with pS or pT.

SNeukamm_100222_06 #1744 RT: 29.99 AV: 1 NL: 4.86E4T: ITMS + c NSI d Full ms2 [email protected] [140.00-1105.00]

200 300 400 500 600 700 800 900 1000 1100

m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

825.33

536.28

444.27

343.28

527.31

364.29

727.41

459.50

265.12487.31

658.40

646.34334.18

711.06

545.35807.31419.11175.20 326.25

192.08 304.33 897.13789.43 974.54 1051.00

Figure A.3: Exemplary MS/MS spectrum and fragmentation table of the peptideTY(pS)LTTPAR from a tryptic digest of transiently expressed IRS2 in HEK293 cells treatedwith IGF-1. Positive-ion MS/MS spectra that enabled identification of Ser573 as phosphory-lated residue in this peptide (m/z = mass to charge).

A.1. PHOSPHOPEPTIDES DETECTED BY MASS SPECTROMETRY 128

# b b++ b0 b0++ Seq. y y++ y* y*++ y0 y0++ #

1 102.0550 51.5311 84.0444 42.5258 T 9

2 265.1183 133.0628 247.1077 124.0575 Y 988.4499 494.7286 971.4234 486.2153 970.4394 485.7233 8

3 432.1166 216.5620 414.1061 207.5567 S 825.3866 413.1969 808.3601 404.6837 807.3760 404.1917 7

4 545.2007 273.1040 527.1901 264.0987 L 658.3883 329.6978 641.3617 321.1845 640.3777 320.6925 6

5 646.2484 323.6278 628.2378 314.6225 T 545.3042 273.1557 528.2776 264.6425 527.2936 264.1504 5

6 747.2961 374.1517 729.2855 365.1464 T 444.2565 222.6319 427.2300 214.1186 426.2459 213.6266 4

7 844.3488 422.6781 826.3383 413.6728 P 343.2088 172.1081 326.1823 163.5948 3

8 915.3859 458.1966 897.3754 449.1913 A 246.1561 123.5817 229.1295 115.0684 2

9 R 175.1190 88.0631 158.0924 79.5498 1

Figure A.4: b- and y-ion series and fragmentation table of the peptide TY(pS)LTTPAR.

A.2. SEQUENCE ALIGNMENT MOUSE AND HUMAN IRS1/2 129

A.2 Sequence alignment of mouse and humanIRS1 and IRS2

Figure A.5: Complete sequence alignment of mouse and human IRS1 and IRS2 sequence.Identified phosphosites that are common to IRS2 isolated from serum starved and IGF-1-treated cells are labeled in red, whereas the phosphosites that were detected by stimulationof cells with forskolin are labeled in green.

mouse IRS2 1 MASAPLPGPPASAGGDGPNLNNNN-NNNNHSVRKCGYLRKQK 41

human IRS2 1 MASPPRHGPPGPASGDGPNLNNNN-NNNNHSVRKCGYLRKQK 41

mouse IRS1 1 -------------------MASPPDTDGFSDVRKVGYLRKPK 23

human IRS1 1 -------------------MASPPESDGFSDVRKVGYLRKPK 23

mouse IRS2 42 HGHKRFFVLRGPGTGGDEASAAGGSPPQPPRLEYYESEKKWR 83

human IRS2 42 HGHKRFFVLRGPGAGGDEATAGGGSAPQPPRLEYYESEKKWR 83

mouse IRS1 24 SMHKRFFVLRAASEAG-----------GPARLEYYENEKKWR 54

human IRS1 24 SMHKRFFVLRAASEAG-----------GPARLEYYENEKKWR 54

mouse IRS2 84 SKAGAPKRVIALDCCLNINKRADAKHKYLIALYTKDEYFAVA 125

human IRS2 84 SKAGAPKRVIALDCCLNINKRADAKHKYLIALYTKDEYFAVA 125

mouse IRS1 55 HKSSAPKRSIPLESCFNINKRADSKNKHLVALYTRDEHFAIA 96

human IRS1 55 HKSSAPKRSIPLESCFNINKRADSKNKHLVALYTRDEHFAIA 96

mouse IRS2 126 AENEQEQEGWYRALTDLVSEGRSGEGGSG---TTGGSCSASL 164

human IRS2 126 AENEQEQEGWYRALTDLVSEGRAAAGDAPPAAAPAASCSASL 167

mouse IRS1 97 ADSEAEQDSWYQALLQLHNRAKAHHDGA------GGGCGGSC 132

human IRS1 97 ADSEAEQDSWYQALLQLHNRAKGHHDGAA--ALGAGGGGGSC 136

mouse IRS2 165 PGVLGGSAGAAGCDDNYGLVTPATAVYREVWQVNLKPKGLGQ 206

human IRS2 168 PGALGGSAGAAGAEDSYGLVAPATAAYREVWQVNLKPKGLGQ 209

mouse IRS1 133 SGSS--GVGEAGEDLSYD-TGP-GPAFKEVWQVILKPKGLGQ 170

human IRS1 137 SGSS--GLGEAGEDLSYGDVPP-GPAFKEVWQVILKPKGLGQ 175

mouse IRS2 207 SKNLTGVYRLCLSARTIGFVKLNCEQPSVTLQLMNIRRCGHS 248

human IRS2 210 SKNLTGVYRLCLSARTIGFVKLNCEQPSVTLQLMNIRRCGHS 251

mouse IRS1 171 TKNLIGIYRLCLTSKTISFVKLNSEAAAVVLQLMNIRRCGHS 212

human IRS1 176 TKNLIGIYRLCLTSKTISFVKLNSEAAAVVLQLMNIRRCGHS 217

mouse IRS2 249 DSFFFIEVGRSAVTGPGELWMQADDSVVAQNIHETILEAMKA 290

human IRS2 252 DSFFFIEVGRSAVTGPGELWMQADDSVVAQNIHETILEAMKA 293

mouse IRS1 213 ENFFFIEVGRSAVTGPGEFWMQVDDSVVAQNMHETILEAMRA 254

human IRS1 218 ENFFFIEVGRSAVTGPGEFWMQVDDSVVAQNMHETILEAMRA 259

mouse IRS2 291 LKELFEFRPRSKSQSSGSSATHPISVPGARRHHHLVNLPPSQ 332

human IRS2 294 LKELFEFRPRSKSQSSGSSATHPISVPGARRHHHLVNLPPSQ 335

mouse IRS1 255 MSDE--FRPRSKSQSSS-SCSNPISVPL--RRHHLNNPPPSQ 291

human IRS1 260 MSDE--FRPRSKSQSSS-NCSNPISVPL--RRHHLNNPPPSQ 296


mouse IRS2 333 TGLVRRSRTDSLAATPPAAKC----TSCRVRTASEGDGGAAG 370

human IRS2 336 TGLVRRSRTDSLAATPPAAKC----SSCRVRTASEGDGGAAA 373

mouse IRS1 292 VGLTRRSRTESITATSPASMVGGKPGSFRVRASSDGEGT--- 330

human IRS1 297 VGLTRRSRTESITATSPASMVGGKPGSFRVRASSDGEGT--- 335

mouse IRS2 371 GAGTAGGRPMSVAGSPLSPGPVRAPLSRSHTLSAGCGGRPSK 412

human IRS2 374 GAAAAGARPVSVAGSPLSPGPVRAPLSRSHTLSGGCGGRGSK 415

mouse IRS1 331 -----MSRPASVDGSPVSPSTNRTHAHRHR---------G-- 356

human IRS1 336 -----MSRPASVDGSPVSPSTNRTHAHRHR---------G-- 361

mouse IRS2 413 VTLAPAGGALQHSRSMSMPVAHSPPAATSPGSLSSSSGHGSG 454

human IRS2 416 VALLPAGGALQHSRSMSMPVAHSPPAATSPGSLSSSSGHGSG 457

mouse IRS1 357 --SSRLHPPLNHSRSIPMPSSRCSPSATSPVSLSSSSTSGHG 396

human IRS1 362 --SARLHPPLNHSRSIPMPASRCSPSATSPVSLSSSSTSGHG 401

mouse IRS2 455 SYPLPPGSHPHLPHPLHHPQGQRPSSGSASASGSPSDPGFMS 496

human IRS2 458 SYPPPPGPHPPLPHPLHHGPGQRPSSGSASASGSPSDPGFMS 499

mouse IRS1 397 STSDCLFPRRSSASVSGSPSDGGFIS---------------- 422

human IRS1 402 STSDCLFPRRSSASVSGSPSDGGFIS---------------- 427

mouse IRS2 497 LDEYGSSPGDLRAFSSHRSNTPESIAETPPARDGS-GGELYG 537

human IRS2 500 LDEYGSSPGDLRAFCSHRSNTPESIAETPPARDGGGGGEFYG 541

mouse IRS1 423 SDEYGSSPCDFRS--SFRSVTPDSLGHTPPARGEE---ELSN 459

human IRS1 428 SDEYGSSPCDFRS--SFRSVTPDSLGHTPPARGEE---ELSN 464

mouse IRS2 538 YMSMDRP-------------LSHCGRPYR------------- 553

human IRS2 542 YMTMDRP-------------LSHCGRSYR------------- 557

mouse IRS1 460 YICMGGKGASTLAAPNGHYILSRGGNGHRYIPGANLGTSPAL 501

human IRS1 465 YICMGGKGPSTLTAPNGHYILSRGGNGHRCTPGTGLGTSPAL 506

mouse IRS2 554 -RVSGDGAQDLDRGLRKRTYSLTTPA---RQRQVPQPSSASL 591

human IRS2 558 -RVSGDAAQDLDRGLRKRTYSLTTPA---RQRPVPQPSSASL 595

mouse IRS1 502 PGDEAAGAADLDNRFRKRTHSAGTSPTISHQKTPSQSSVASI 543

human IRS1 507 AGDEAASAADLDNRFRKRTHSAGTSPTITHQKTPSQSSVASI 548

mouse IRS2 592 DEYTLMRATFS----GSSGRLCPSFPASSPKVAYNPYPEDYG 629

human IRS2 596 DEYTLMRATFS----GSAGRLCPSCPASSPKVAYHPYPEDYG 633

mouse IRS1 544 EEYTEMMPAAYPPGGGSGGRLP-GYR-HSAFVPTHSYPEEGL 583

human IRS1 549 EEYTEMMP-AYPPGGGSGGRLP-GHR-HSAFVPTRSYPEEGL 587

mouse IRS2 630 D-----IEIGSHKSSSSNLGADDGYMPMTPGAALRSGGPNSC 666

human IRS2 634 D-----IEIGSHRSSSSNLGADDGYMPMTPGAALAGSGSGSC 670

mouse IRS1 584 EMHHLERRGGHHRPDTSNLHTDDGYMPMSPGVAPVPSN--RK 623

human IRS1 588 EMHPLERRGGHHRPDSSTLHTDDGYMPMSPGVAPVPSG--RK 627

mouse IRS2 667 KSDDYMPMSPTSVSAPKQILQPRLAA----ALPPSGAAVPAP 704

human IRS2 671 RSDDYMPMSPASVSAPKQILQPRAAAAAAAAVPSAGPAGPAP 712

mouse IRS1 624 GNGDYMPMSPKSVSAPQQIINPIRRHPQR--VDPNGYMMM-- 661

human IRS1 628 GSGDYMPMSPKSVSAPQQIINPIRRHPQR--VDPNGYMMM-- 665


mouse IRS2 705 PSGVGRTFPVNGGG-YKASSPAESSPEDSGYMRMWCGSK--- 742

human IRS2 713 TSAAGRTFPASGGG-YKASSPAESSPEDSGYMRMWCGSK--- 750

mouse IRS1 662 -SPSGSCSPDIGGGSSSSS-SISAAPSGSSYGKPWTNGVGGH 701

human IRS1 666 -SPSGGCSPDIGGGPSSSSSSSNAVPSGTSYGKLWTNGVGGH 706

mouse IRS2 743 ---------LSMENPDPKLLP-NGDYLNMSPSEAGTAGTPPD 774

human IRS2 751 ---------LSMEHADGKLLP-NGDYLNVSPSDAVTTGTPPD 782

mouse IRS1 702 HTHALPHAKPPVESGGGKLLPCTGDYMNMSPVGDSNTSSPSE 743

human IRS1 707 HSHVLPHPKPPVESSGGKLLPCTGDYMNMSPVGDSNTSSPSD 748

mouse IRS2 775 F-SAALRGGSEGLKGIPGHCYSSLPRSYKAPCSCSG----DN 811

human IRS2 783 FFSAALHPGGEPLRGVPGCCYSSLPRSYKAPYTCGG----DS 820

mouse IRS1 744 CYYGPEDPQ-----HKPVLSYYSLPRSFKHTQRPGEPEEGAR 780

human IRS1 749 CYYGPEDPQ-----HKPVLSYYSLPRSFKHTQRPGEPEEGAR 785

mouse IRS2 812 DQYVLMSSPVGRILEEERLEPQAT------PGAGTFG-AAGG 846

human IRS2 821 DQYVLMSSPVGRILEEERLEPQAT------PGPSQAASAFGA 856

mouse IRS1 781 HQHLRLSSSSGRLRYTATAEDSSSSTSSDSLGGGYCGARPES 822

human IRS1 786 HQHLRLSTSSGRLLYAATADDSSSSTSSDSLGGGYCGARLEP 827

mouse IRS2 847 SHTQPHHSAVPSSMRPSAIGGRPEGFLGQRCRAVRPTRLSLE 888

human IRS2 857 GPTQPPHPVVPSPVRPS--GGRPEGFLGQRGRAVRPTRLSLE 896

mouse IRS1 823 SLTHPHHHVLQPHL-P----RKVDTAAQTNSRLARPTRLSLG 859

human IRS1 828 SLPHPHHQVLQPHL-P----RKVDTAAQTNSRLARPTRLSLG 864

mouse IRS2 889 GLQ--TL-----------PSMQEYPLPTEPKSPGEYINIDFG 917

human IRS2 897 GLP--SL-----------PSMHEYPLPPEPKSPGEYINIDFG 925

mouse IRS1 860 DPKASTLPRVREQQQ----QQQSSLHPPEPKSPGEYVNIEFG 897

human IRS1 865 DPKASTLPRAREQQQ----QQQPLLHPPEPKSPGEYVNIEFG 902

mouse IRS2 918 EAGTRLSPPAPPLLASAASSSSLLSASSPASSLGSGTPGTSS 959

human IRS2 926 EPGARLSPPAPPLLASAASSSSLLSASSPASSLGSGTPGTSS 967

mouse IRS1 898 SGQPGYLAGPATSR-SS-PSVRCPPQLHP------------A 925

human IRS1 903 SDQSGYLSGPVAFH-SS-PSVRCPSQLQP------------A 930

mouse IRS2 960 DSRQRSPLSDYMNLDFSSPKSPKPSTRSGDTVGSMDGLLSPE 1001

human IRS2 968 DSRQRSPLSDYMNLDFSSPKSPKPGAPSGHPVGSLDGLLSPE 1009

mouse IRS1 926 PRE-ETGSEEYMNMDLGPGRRATWQESGGVELGRIGPA-PPG 965

human IRS1 931 PREEETGTEEYMKMDLGPGRRAAWQESTGVEMGRLGPA-PPG 971

mouse IRS2 1002 ASSPYPPLPPRPSTSPSSLQQP--LPPAPGDLYRLPPASA-A 1040

human IRS2 1010 ASSPYPPLPPRPSASPSSSLQPPPPPPAPGELYRLPPASAVA 1051

mouse IRS1 966 SATVCRP--------------TRSVPNSRGDYMTMQIG---- 989

human IRS1 972 AASICRP--------------TRAVPSSRGDYMTMQMS---- 995

mouse IRS2 1041 TSQGPTAGSSMSSEPGDNGDYTEMAFGVAATP----PQPIVA 1078

human IRS2 1052 TAQGPGAASSLSSDTGDNGDYTEMAFGVAATP----PQPIAA 1089

mouse IRS1 990 ----CPRQSYVDTSPVAPVSYADMRTGIAAEKASLPRPTGAA 1037

human IRS1 996 ----CPRQSYVDTSPAAPVSYADMRTGIAAEEVSLPRATMAA 1033


mouse IRS2 1079 PPKPEGARV---ASPTSGL-------KRLSLMDQVSGVEAFL 1110

human IRS2 1090 PPKPEAARV---ASPTSGV-------KRLSLMEQVSGVEAFL 1121

mouse IRS1 1028 PPPSSTASSSASVTPQGATAEQATHSSLLGGPQGPGGMSAFT 1069

human IRS1 1034 ASSSSAASAS-PTGPQ-GAAELAAHSSLLGGPQGPGGMSAFT 1073

mouse IRS2 1111 QVSQPPDPHRGAKVIRADPQGGRRRHSSETFSSTTTVTPVSP 1152

human IRS2 1122 QVSQPPDPHRGAKVIRADPQGGRRRHSSETFSSTTTVTPVSP 1163

mouse IRS1 1070 RVNLSPNHNQSAKVIRADTQGCRRRHSSETFSA---PTRAGN 1108

human IRS1 1074 RVNLSPNRNQSAKVIRADPQGCRRRHSSETFSSTPSATRVGN 1115

mouse IRS2 1153 SFAHNSKRHNSASVENVSLRKSSEGSSTL--GGGDEPPTSPG 1192

human IRS2 1164 SFAHNPKRHNSASVENVSLRKSSEGGVGVGPGGGDEPPTSPR 1205

mouse IRS1 1109 TVPFG-------------------AGAAVGGSGGGGGGGSED 1131

human IRS1 1116 TVPFG-------------------AGAAVGGG-GGSSSSSED 1137

mouse IRS2 1193 QAQPLVAVPPVPQARPWNPGQPGALIGCPGGSSSPMRRETSV 1234

human IRS2 1206 QLQPA--PPLAPQGRPWTPGQPGGLVGCPGSGGSPMRRETSA 1245

mouse IRS1 1132 VKR----HSSASFENVWL--RPGDLGGVSKES--APVCGAAG 1165

human IRS1 1138 VKR----HSSASFENVWL--RPGELGGAPKEP--AKLCGAAG 1171

mouse IRS2 1235 GFQNGLNYIAIDVRGEQGSLAQ-----SQ--------PQPGD 1263

human IRS2 1246 GFQNGLNYIAIDVREEPGLPPQ-----PQP--PPPPLPQPGD 1280

mouse IRS1 1166 GLEKSLNYIDLDLAKE---RSQDCPSQQQSLPPPPPHQPLGS 1204

human IRS1 1172 GLENGLNYIDLDLVKDFKQCPQECTPEPQPPPPPPPHQPLGS 1213

mouse IRS2 1264 KNSWSRTRSLGGLLGTVGGSGASGVCGGPGTGALPSASTYAS 1305

human IRS2 1281 KSSWGRTRSLGGLISAVGVGSTGGGCGGPGPGALPPANTYAS 1322

mouse IRS1 1205 NEGNSPRR------------------------SSEDLSNYAS 1222

human IRS1 1214 GESSSTRR------------------------SSEDLSAYAS 1231

mouse IRS2 1306 IDFLSHHLKEATVVKE 1321

human IRS2 1323 IDFLSHHLKEATIVKE 1338

mouse IRS1 1223 ISFQKQPEDRQ----- 1233

human IRS1 1232 ISFQKQPEDRQ----- 1242

Erklarung

Erklarung gemaß §8 Absatz 2 Aufzahlungspunkt 2 der Promotionsor-dnung zum Dr. rer. nat. der Universitat Hohenheim

Hiermit erklare ich, dass ich die vorgelegte Dissertation selbstandig angefertigtund nur die angegebenen Quellen und Hilfsmittel verwendet habe. Wortlichoder inhaltlich ubernommene Stellen sind als solche gekennzeichnet.

Sabine Neukamm

Stuttgart, den 15.08.2012

133

Curriculum vitae

Sabine Sarah NeukammIm Asemwald 56/1470599 Stuttgart

geboren am 04.06.1982 in Filderstadt

Schulausbildung

1988 – 1992 Bruckenacker Grundschule Bernhausen

1992 – 1998 Fleinsbach Realschule BernhausenAbschluss: Mittlere Reife

1998 – 2001 Hedwig-Dohm Schule Stuttgart, ErnahrungswissenschaftlichesGymnasiumAbschluss: Allgemeine Hochschulreife (Note: 1,6)

Auslandserfahrung

2001 – 2002 Work & Travel Aufenthalt in Australien

Wissenschaftlicher Werdegang

2002 – 2008 Studium der Ernahrungswissenschaften an der UniversitatHohenheimAbschluss: Diplom-Ernahrungswissenschaftlerin (Note: 1,0)Diplomarbeit: “Expression and functional characterization ofCNTF receptor mutants in polarized cells”

2008 – 2012 Anfertigung der Dissertation am UniversitatsklinikumTubingen, Abteilung Innere Medizin IV

2010 5-monatiger Forschungsaufenthalt an der Universitat Dundee,Schottland

134

Acknowledgment

There are some persons I want to thank for their help and support throughoutthe process of generating this thesis.

First of all I want to thank Prof. Dr. Erwin Schleicher from the Departmentof Internal Medicine IV at the University Hospital Tubingen for the provisionof a very challenging project. My thank also goes to Prof. Dr. Lutz Graevefrom the Institute of Biological Chemistry and Nutrition from the University ofHohenheim for his willingness to co-supervise this thesis.

A tremendously big thank goes to Prof. Dr. Cora Weigert and Prof. Dr.Rainer Lehmann, since they always believed in my project (or at least left mewith that impression) and never gave up encouraging me to perform furtherexperiments, try new approaches and methods. Cora in particular was alwaysapproachable for discussions about any kind of problems that occur more or lessfrequently in an experimental thesis.

I spent almost five months in the laboratory of Prof. Carol MacKintosh atthe University of Dundee and acquired a lot of data there. Many thanks toCarol for the possibility to work in her group and all my colleagues in Dundee.

Emma – thank you so much for your positive vibes, the coffee times, goingout for dinner, shopping in Edinburgh and all the laughter!

No one works alone in a lab, that’s why it is so important to have goodcolleagues, who can help out with ideas and experimental knowledge. My thanksgo to Louise Fritsche, Miriam Hoene, Andrea Drescher and Andreas Peter.

Special thanks to my other colleagues – the technical assistants Ann KathrinPohl, Heike Runge and Mareike Walenta, since without them the lab would notrun as smoothly as it does.

During the years there were also some very nice diploma students with ex-cellent baking/cooking abilities (very much appreciated!) in the lab – thanks toMagnus Wolf, Christian Klingler and Tamara Suhm.

Thanks to Simone for being my car sharing partner and a good example innot taking everything too seriously!

I very much appreciate the continuing good atmosphere in the lab duringwork or while drinking coffee, eating cookies, chocolate, cake, ice cream andfunny Chinese sweets.

For their endless love, support and believe in me that I can achieve whateverI want I have to thank my family and especially my lovely parents.

Bernd, you are my “Helddesk”, IT-guru, motivator, supporter, ski hero andmy resting point who always cheers me up. I cannot imagine accomplishing thisthesis without you.

135

novel serine phosphorylation sites of irs2 mediate 14-3-3 ... · cora weigert. \identi cation of...

Documents