Molecular analysis of action using high throughput genetic screens

Poh Sim Khoo

Thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Garvan Institute of Medical Research University of New South Wales

31st March 2011

PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Khoo

First name: Poh Sim Other name/s:

Abbreviation for degree as given in the University calendar: PhD

School: St Vincent's Cllinical School Faculty:Medicine

Title: Molecular analysis of insulin action using high throughput genetic screens

Abstract 350 words maximum: (PLEASE TYPE)

The insulin-stimulated uptake of glucose by muscle and adipose is vital for the maintenance of glucose homeostasis in the body. This uptake primarily occurs through the action of the insulin-regulatable glucose transporter GLUT4, which is rapidly translocated to the plasma membrane in response to the insulin signal. However, in the insulin resistant state, insulin is unable to effect a normal biological response in its target tissues, characterized by decreased glucose uptake as a result of defects in insulin signaling and attenuated GLUT4 translocation. Elucidating the molecular causes underlying insulin resistance is important for the development of therapeutics for this disease. Identifying the components involved in the propagation or regulation of insulin signaling is an important step in understanding insulin resistance. To date, the upstream components of the signaling pathway are well established, demonstrating the importance of the insulin , IRS and PI3K/Akt signaling axis in this process. As a result much work has focused on defects at the point of IRS in the development of insulin resistance. However, it has recently been suggested that defects associated with insulin resistance occur independently of IRS. Therefore, identifying the sites that this dysfunction occurs at is of great interest in understanding this disease. The purpose of this study is to discover novel proteins involved in the regulation of the insulin signaling and GLUT4 translocation. A GLUT4-overexpressing HeLa cell line was developed and optimized for use in high throughput screening for regulators of insulin stimulated GLUT4 translocation. Insulin stimulation caused GLUT4 translocation to the plasma membrane, and the activation of the PI3K/Akt signaling pathway in these cells. Using this cell line, I performed an siRNA screen of and DUB libraries which identified a number of novel targets that may play a role in the regulation of insulin stimulated GLUT4 translocation. In conclusion, this assay can be used to identify novel regulators of insulin stimulated GLUT4 translocation, which may potentially represent targets for drug development in the treatment of insulin resistance

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

.... .... Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS

ORIGINALITY STATEMENT ‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’ Signed ...... Date ......

Acknowledgements

I would like to thank my supervisor at the Garvan, David James. I feel privileged to have been able to work on a challenging project, and it would not have been possible without his vision and guidance. I would also like to thank my co-supervisor Jackie Stoeckli for her great practical advice and for teaching me techniques in the lab that I have used during my PhD.

At Genentech, I would like to thank Richard Scheller for the opportunity to work in his lab. Thanks to my supervisor Jagath Junutula, I am extremely grateful for his help, advice and enthusiasm. I would also like to thank my colleagues in the lab; Karen Ervin, Sunil Bhakta, Siler Panowski and Coen Kuijl for the friendship and laughter, and for making it a great environment to be in.

Thanks to James Lee for his work with the optimization and execution of the siRNA transfections, Jean-Philippe Stephan for all of his help with running the screens and the image analysis and Eric Torres for his help with the data analysis. Their help with the siRNA screens was invaluable.

Finally thanks to my friends and family for believing in me and supporting me through this time, it is very much appreciated.

Table of Contents Summary 1 Chapter 1 – General Introduction 3 Insulin and the maintenance of glucose homeostasis 4 The GLUT4 glucose transporter 4 Insulin-stimulated GLUT4 translocation 6 PI3K-dependent insulin signaling 6 PI3K 7 Akt 7 AS160 8 aPKC (λ, ζ) 9 PI3K-independent signaling 10 Contraction stimulated GLUT4 translocation 12 Increased intracellular calcium 12 AMPK 13 Convergence of insulin and contraction signaling pathways leading to GLUT4 translocation 14 Insulin resistance 15 Inflammation and insulin resistance 16 JNK 16 IKKβ 16 ERK 17 P62 signaling adaptor 17 SOCs proteins 18 GSK3 18 PKCθ 19 Tyrosine and serine phosphatases 19 PIKfyve 20 Hypothesis and aims 21 Chapter 2 – Materials and Methods 22 Reagents, Antibodies and Constructs 23 Cell Culture and Transfections 23 Generation of HeLa cells stably expressing HA-GLUT4 24 siRNA Transfection (screening) 24 Quantitative GLUT4 translocation assay 24 IN Cell image acquisition 25 Image analysis 25 Transferrin uptake assay 25 Immunofluorescence microscopy 26 Taqman analysis for knockdown 26 Preparation of cell lysates 27 Western blotting 27 Chapter 3 – Development of a GLUT4 translocation assay in HeLa cells 29 Introduction 30 Results 34 HeLa cells stably expressing HA-GLUT4 exhibit insulin stimulated GLUT4 translocation 34 HA-GLUT4-HeLa cells express the majority of expressed in human adipose and muscle tissue 34 Localization of GLUT4 in the HA-GLUT4-HeLa cell line 37 Insulin, IGF1 and EGF treatment of HA-GLUT4-HeLa cells induces GLUT4 translocation 39 Comparison of insulin signaling in HA-GLUT4-HeLa cells and 3T3-L1 adipocytes 40 PI3K and Akt inhibitors block insulin-stimulated GLUT4 translocation in both 3T3-L1 adipocytes and HA-GLUT4-HeLa cells 41 siRNA-mediated knockdown of Akt and PI3K inhibits GLUT4 translocation in HA-GLUT4-HeLa cells 41 Development of GLUT4 translocation assay for high throughput screening using the HA-GLUT4-HeLa cell line 43 Discussion 47 Chapter 4 – Identifying kinase regulators of insulin signaling and GLUT4 translocation by siRNA screening 51 Introduction 52 The – kinase activity and regulation by 52 Phosphorylation of IRS proteins and modulation of the insulin signal 53 Kinase regulation of insulin signaling and GLUT4 translocation independent of IRS phosphorylation 54 Results 55 Primary kinase screen 55 Secondary screen 60 Tertiary screen 66 Discussion 75 known to function in insulin signaling and GLUT4 translocation 75 Kinases that interact with regulators of insulin signaling 77 Kinases related to ERK and JNK signaling 79 Kinases involved in regulation of trafficking 79 Metabolic proteins 80 Miscellaneous 81 Conclusion 81 Chapter 5 – Identifying DUB or ULP regulators of insulin signaling and GLUT4 translocation by siRNA screening 83 Introduction 84 Ubiquitin and Ubiquitin-like proteins 84 Ubiquitin system functions 85 Ubiquitin-like proteins and insulin signaling 87 Results 90 Primary screen 90 Secondary screen 96 Tertiary screen 100 Discussion 106 Proteins regulating stability and degradation 106 Proteins involved in mRNA processing and protein translation 107 Proteins involved in vesicle trafficking 109 Miscellaneous functions 109 Conclusion 110 Chapter 6 – General Discussion 112 Development and optimization of the HA-GLUT4-HeLa cell line 113 Involvement of the ERK/MAPK pathway in insulin signaling 115 TP53RK 117 USP36 118 Conclusion 119 Appendix 120 Bibliography 135

Table of Figures Chapter 1 – General Introduction 3 1.1 PI3K-Dependent insulin signaling 10 1.2 PI3K-independent insulin signaling 11 1.3 Contraction-stimulated GLUT4 translocation 14 Chapter 2 – Materials and Methods 22 Table 2.1 assays used for Taqman analysis of gene knockdown 27 Chapter 3 – Development of a GLUT4 translocation assay in HeLa cells 29 3.1 Affymetrix microarray expression profile of insulin receptor in various cell lines 33 3.2 Screening of single cell clones for HA-GLUT4 expression 35 3.3 Screening for GLUT4 translocation in HA-GLUT4 positive clones 35 3.4 Venn diagram of genes expressed in adipose, muscle and HA-GLUT4-HeLa cells 36 3.5 Intracellular localisation of GLUT4 in HA-GLUT4-HeLa cells 38 3.6 Effect of growth factors on GLUT4 translocation 39 3.7 Comparison of signaling in HA-GLUT4-HeLa cells and HA-GLUT4-3T3-L1 cells 40 3.8 PI3K and Akt inhibitors block insulin-stimulated GLUT4 translocation in HA-GLUT4-HeLa cells and HA-GLUT4 3T3-L1 adipocytes 42 3.9 Knockdown of PI3Ka and Akt1 + Akt2 inhibits IGF1-stimulated GLUT4 translocation 42 3.10 Image analysis for quantification of GLUT4 at the plasma membrane 45 3.11 GLUT4 translocation measured using the IN Cell analyzer 46 Chapter 4 – Identifying kinase regulators of insulin signaling and GLUT4 translocation by siRNA screening 51 4.1 Targets from primary kinase screen showing increased basal GLUT4 translocation or decreased IGF1 stimulated translocation 57 4.2 Primary screen data showing effects of knockdown of selected insulin pathway components 59 4.3a Deconvolution of increased basal hits from primary screen 61 4.3b Deconvolution of decreased stimulated hits from primary screen 62 4.4 Deconvolution of targets from primary screen 63 Table 4.1 Validated genes as based on deconvolution of siRNA pools 63 4.5 Total GLUT4 levels in validated hits from secondary screen 64 4.6 Increased hits from secondary screen – tertiary validation 68 4.7 Decreased stimulated GLUT4 translocation hits from secondary screen – tertiary validation 70 4.8. Gene expression analysis by qPCR 72 Table 4.2 Tertiary screen – Validation of hits 72 4.9 High confidence hits from DUB screen 73 Chapter 5 – Identifying DUB or ULP regulators of insulin signaling and GLUT4 translocation by siRNA screening 83 Table 5.1 Ubl family of proteins 84 5.1 The ubiquitin-proteasome system 85 Table 5.2 Results from primary DUB screen 91 5.2 Ranked Z scores from primary DUB screen 94 5.3 Targets from primary screen selected for deconvolution 95 5.4 Deconvolution of targets from primary screen 97 Table 5.3 Validated genes as based on deconvolution of siRNA pools 98 5.5 Total GLUT4 levels in validated hits from secondary screen 99 5.6 Tertiary screen validation – GLUT4 translocation 102 5.7 Tertiary screen validation – transferrin uptake 103 5.8 Gene expression analysis by qPCR 104 Table 5.3 Validation of hits in tertiary screen 104 5.9 High confidence hits from DUB screen 105 Appendix 120 Table A1. Results from primary kinase screen 121

Summary

1 The insulin-stimulated uptake of glucose by muscle and adipose is vital for the maintenance of glucose homeostasis in the body. This uptake primarily occurs through the action of the insulin-regulatable glucose transporter GLUT4, which is rapidly translocated to the plasma membrane in response to the insulin signal. However, in the insulin resistant state, insulin is unable to effect a normal biological response in its target tissues, characterized by decreased glucose uptake as a result of defects in insulin signaling and attenuated GLUT4 translocation. Elucidating the molecular causes underlying insulin resistance is important for the development of therapeutics for this disease. Identifying the components involved in the propagation or regulation of insulin signaling is an important step in understanding insulin resistance. To date, the upstream components of the signaling pathway are well established, demonstrating the importance of the insulin receptor, IRS proteins and PI3K/Akt signaling axis in this process. As a result much work has focused on defects at the point of IRS in the development of insulin resistance. However, it has recently been suggested that defects associated with insulin resistance occur independently of IRS. Therefore, identifying the sites that this dysfunction occurs at is of great interest in understanding this disease. The purpose of this study is to discover novel proteins involved in the regulation of the insulin signaling and GLUT4 translocation. A GLUT4-overexpressing HeLa cell line was developed and optimized for use in high throughput screening for regulators of insulin stimulated GLUT4 translocation. Insulin stimulation caused GLUT4 translocation to the plasma membrane, and the activation of the PI3K/Akt signaling pathway in these cells. Using this cell line, I performed an siRNA screen of kinase and DUB libraries which identified a number of novel targets that may play a role in the regulation of insulin stimulated GLUT4 translocation. In conclusion, this assay can be used to identify novel regulators of insulin stimulated GLUT4 translocation, which may potentially represent targets for drug development in the treatment of insulin resistance.

2

Chapter 1

General Introduction

3 Insulin and the maintenance of glucose homeostasis

The maintenance of glucose homeostasis is an important biological function. Typically, blood glucose concentration is maintained within a relatively narrow range despite fluctuations in glucose uptake and disposal (Gerich 2000). This control is required in order to avoid the complications related with hyper- and hypo- glycaemia. The maintenance of glucose homeostasis involves balancing glucose supply from the intestine and output from the liver, with glucose uptake by the peripheral tissues (Saltiel and Kahn 2001). Insulin is the primary regulator of glucose homeostasis, achieved in part by inhibiting glucose production by the liver, as well as stimulating the uptake of glucose by insulin responsive tissues such as muscle and fat (Schinner et al. 2005). Insulin also stimulates lipogenesis, glycogen and protein synthesis, as well as inhibiting lipolysis, glycogenolysis and protein breakdown (Saltiel and Kahn 2001). The main site of insulin action occurs in skeletal muscle, which is responsible for up to 75% of insulin stimulated glucose disposal. In contrast, adipose tissue accounts for less than 10% of glucose uptake (Klip and Paquet 1990). Despite this, mice with an adipose specific knockout of the insulin sensitive glucose transporter rapidly develop insulin resistance (Abel et al. 2001), whereas mice with a muscle-specific knockout of the insulin receptor have normal glucose tolerance (Bruning et al. 1998). These observations suggest that adipose tissue, while not the primary site of glucose uptake, plays an important role in the regulation of whole body homeostasis. Diabetes mellitus is a disease that is characterized by high blood glucose levels. There are two main forms of diabetes. Type I diabetes occurs due to insufficient production of insulin as a result of autoimmune destruction of insulin-producing β-cells of the pancreas. Type II diabetes is associated with insulin resistance, defined as an inability of insulin to promote normal cellular glucose uptake at a given insulin concentration (James and Piper 1994; Ducluzeau et al. 2002). Because insulin action is vital for the maintenance of glucose homeostasis, much work has been put into understanding how this hormone signals to initiate the uptake of glucose from the blood, and furthermore the mechanisms underlying the attenuated response to this signal in insulin resistance.

The GLUT4 glucose transporter

Glucose does not freely permeate the plasma membrane, and transport across the cell membrane requires specific transport proteins. The glucose transporter proteins can be

4 divided into two groups; the facilitative glucose transporter (GLUT) family and the Na+/glucose co-transporters (SGLT). The SGLT family consists of three isoforms. SGLT1 and SGLT2 play a role in absorption of glucose from the small intestine, and the re-absorption of glucose from urine in the kidney. The presence of Na+ facilitates the uptake of glucose against its concentration gradient, using the energy provided by co-transport of Na+ ions down their electrochemical gradient. In contrast, the GLUT proteins accelerate the transport of glucose by diffusion into the cell (Wood and Trayhurn 2003). The GLUT family belongs to the Major Facilitator Superfamily (MFS) of membrane transporters (Pao et al. 1998). To date, 14 GLUT proteins have been identified in humans, and include transporters for substrates other than glucose (Uldry and Thorens 2004). GLUTs are predicted to contain 12 transmembrane domains and a single N- linked oligosaccharide. Based on sequence similarity, the GLUT family members are grouped into three different classes (Joost et al. 2002). The GLUT proteins have unique tissue distribution, cell specific expression and biochemical properties, which determine their specific functions (Olson and Pessin 1996). The class I facilitative transporters (GLUT1-4), have been well characterized in terms of structure, function and tissue distribution (Wood and Trayhurn 2003). Of these, GLUTs 1-3 are constitutively targeted to the cell surface, thus allowing a constant influx of glucose into the cell (Bryant et al. 2002). In contrast, GLUT4 function is regulated in response to insulin (James et al. 1988). GLUT4 is expressed mainly in skeletal muscle, heart and adipose tissue, consistent with a function in insulin regulated glucose disposal (Bell et al. 1990). In the absence of insulin, GLUT4 is mainly sequestered in intracellular storage vesicles and is largely absent from the plasma membrane. However, in response to insulin stimulation, GLUT4 rapidly translocates to the plasma membrane resulting in a 10-20 fold increase of GLUT4 at the cell surface (Govers et al. 2004). A concomitant 10-40 fold increase in glucose uptake is observed, suggesting that insulin-stimulated glucose uptake is largely dependent on GLUT4 activity (Bryant et al. 2002). Several observations point to the importance of GLUT4 in the maintenance of glucose homeostasis. Heterozygous GLUT4+/- mice, which have decreased GLUT4 expression, display whole body insulin resistance (Rossetti et al. 1997; Stenbit et al. 1997; Li et al. 2000). Consistent with this selective disruption of GLUT4 in mouse muscle (Zisman et al. 2000) or adipose tissue (Abel et al. 2001) caused insulin resistance and glucose intolerance. In contrast, overexpression of GLUT4

5 improves insulin responsiveness in the db/db diabetic mouse model (Gibbs et al. 1995; Brozinick et al. 2001). Additionally, mice overexpressing GLUT4 in muscle (Tsao et al. 1996) or adipose tissue (Shepherd et al. 1993; Gnudi et al. 1995) show increased insulin sensitivity. Interestingly, overexpression of GLUT4 in adipose tissue of mice lacking GLUT4 selectively in muscle reverses insulin resistance and diabetes (Carvalho et al. 2005). It is suggested that GLUT4 may play a role in glucose sensing and regulation of whole body glucose homeostasis (Herman and Kahn 2006). As a result, studies have focused on elucidating the regulation of GLUT4 activity and translocation. Both insulin stimulation and physical exercise can stimulate the translocation of GLUT4 to the plasma membrane (PM) in muscle (Hayashi et al. 1997; Goodyear and Kahn 1998). The stimulatory effects of insulin and contraction on GLUT4 translocation are additive, suggesting that these two stimuli cause GLUT4 translocation via distinct signaling pathways (Garetto et al. 1984; Nesher et al. 1985). Interestingly, contraction-stimulated glucose transport is not inhibited in insulin resistant animals (Kusunoki et al. 1993; Brozinick et al. 1994). Due to the importance of GLUT4 to the maintenance of glucose homeostasis, many studies have focused on elucidating the different pathways leading to GLUT4 translocation.

Insulin-stimulated GLUT4 translocation

Insulin binding to its receptor triggers multiple signaling events that result in the translocation of GLUT4 from the cytoplasm to the PM. There are two pathways that have been proposed to connect insulin stimulation to GLUT4 translocation. The widely accepted and most firmly established path involves the action of IRS/PI3K/Akt. However, it has also been proposed that a PI3K-independent pathway involving APS/CAP/Cbl/TC10 action exists to regulate insulin-stimulated GLUT4 translocation.

PI3K-dependent insulin signaling

The insulin receptor is a heterotrimeric transmembrane receptor, consisting of two α and two β subunits (Kasuga et al. 1982). When insulin binds it triggers autophosphorylation of the receptor, resulting in phosphorylation of the cytoplasmic β subunits (Kahn and White 1988). This results in the recruitment of a number of proteins including the insulin receptor substrate (IRS) proteins, which interact with the phosphorylated residues via phosphotyrosine binding (PTB) domains (White 1998). There are several IRS proteins, of which IRS1 and IRS2 are the most important in terms

6 of insulin signaling (Myers and White 1993; Sun et al. 1995). The bound IRS proteins are subsequently tyrosine phosphorylated by the IR. These phosphotyrosine residues serve as docking sites for interacting proteins containing a Src Homology 2 (SH2) domain. One of the important SH2 containing proteins involved in insulin signaling is Phosphatidylinositol 3-kinase (PI3K). PI3K Class IA PI3K is a heterodimer consisting of an 85-kDa regulatory subunit (p85) and a 110-kDa catalytic subunit (p110). Phosphorylated IRS1/2 recruits PI3K, and via interaction with the SH2 domain of p85 activates the p110 subunit (Myers et al. 1992; Downes et al. 2005). The recruitment and activation of PI3K by IRS1/2 localizes PI3K at the PM, where it catalyses the formation of PI(3,4)-bisphosphate from PI(4)- phosphate and PI(3,4,5)-trisphosphate from PI(4,5)-bisphosphate (Fruman et al. 1998). The PI(3,4,5)-trisphosphate (PIP3) which is generated then recruits proteins containing a pleckstrin homology domain, including PDK1 and Akt. PDK1 then phosphorylates and activates a number of substrates including Akt and atypical PKC (aPKC) (Rameh and Cantley 1999). The importance of PI3K in insulin signaling was initially demonstrated through the use of PI3K inhibitors with broad specificity including wortmannin and LY294002 showing the requirement of PI3K to insulin signaling (Clarke et al. 1994) More recently, isoform specific inhibitors have been used to dissect the contributions of the different PI3K isoforms to insulin signaling. (Knight et al. 2006) Furthermore the expression of PI3K mutants, (Kotani et al. 1995; Katagiri et al. 1996; Tanti et al. 1996) has demonstrated the importance of PI3K in insulin signaling. Akt The insulin-stimulated increase of PI(3,4,5)P3 and PI(3,4)P2 recruits Akt isoforms to the PM where they undergo a conformational change and are activated by the phosphorylation of two residues (Lawlor and Alessi 2001). PDK1 phosphorylates Akt in the Thr308/309 residue of its activation loop, an important event in activation of (Filippa et al. 2000). The identity of the kinase that catalyses Ser473/474 of Akt has been more elusive, however recent evidence supports the idea that rictor/mTORC2 may play this role (Bayascas and Alessi 2005; Kumar et al. 2010). Following activation, Akt dissociates from the PM and phosphorylates numerous downstream substrates involved in insulin action (Lawlor and Alessi 2001).

7 Three mammalian isoforms of Akt exist (1-3). Of these, Akt1 and Akt2 are the isoforms that are predominantly expressed and activated in muscle and adipose tissue (Whiteman et al. 2002). Evidence for the involvement of Akt in the insulin-signaling pathway comes from multiple sources. Overexpression studies showed that a constitutively active Akt mutant stimulated GLUT4 translocation (Kohn et al. 1998), whereas dominant negative or kinase dead mutants inhibit translocation (Wang et al. 1999). Furthermore inhibition of insulin signaling by Akt ablation in mice (Cho et al. 2001; Bae et al. 2003), siRNA-mediated knockdown of Akt in cell culture (Jiang et al. 2003; Katome et al. 2003) and pharmacological inhibition of Akt (Gonzalez and McGraw 2006) provide further evidence for the role of Akt in insulin signaling. AS160 A direct substrate of Akt, called AS160 (Akt substrate of 160 kDa) has been identified that functions as a negative regulator of insulin signaling (Kane et al. 2002; Gridley et al. 2005). AS160 contains six putative Akt phosphorylation sites, five of which are phosphorylated in response to insulin stimulation (Kane et al. 2002). Expression of an AS160 mutant (4P) in which four of these phosphorylation sites are mutated to Ala, in adipocytes causes a reduction in insulin-stimulated GLUT4 translocation (Sano et al. 2003; Zeigerer et al. 2004). Knockdown of endogenous AS160 resulted in increased basal GLUT4 translocation, suggesting that AS160 is a regulator of basal GLUT4 exocytosis (Eguez et al. 2005). AS160 is localized to GLUT4-containing vesicles through its interaction with IRAP (Larance et al. 2005; Peck et al. 2006), and dissociates in response to insulin (Larance et al. 2005). Additionally, AS160 contains a GTPase- activating protein (GAP) domain, suggesting that AS160 may function to link insulin signaling to GLUT4 trafficking via regulation of the activity of target Rabs (Kane et al. 2002). 14-3-3 proteins interact with AS160 in an insulin- and Akt-dependent manner via an Akt phosphorylation site, Thr-642 (Ramm et al. 2006), and it is proposed that this interaction may inhibit AS160 GAP activity (Stockli et al. 2008). Consistent with this, it was recently reported that mice with an AS160 Thr649Ala knock-in mutation, which affects the interaction with 14-3-3, have altered glucose , glucose intolerance and reduced insulin sensitivity (Chen et al. 2011). Based on these observations, it is proposed that in the basal state, AS160 is associated with GLUT4 vesicles and functions to negatively regulate GLUT4 translocation by maintaining target Rabs in an inactive GDP-bound state. Upon insulin stimulation, phosphorylation of AS160 by Akt facilitates an interaction between 14-3-3 and AS160, leading to

8 inactivation of its GAP domain and dissociation of AS160 from GLUT4 vesicles. This leads to an increase of the active form of target Rabs on the GLUT4 vesicle, promoting GLUT4 translocation to the PM. aPKC (λ, ζ) There are three classes of PKC: conventional (α, ß, γ), novel (δ, ε, η, θ) and atypical (λ, ζ) (Way et al. 2000). In response to insulin stimulation, aPKC isoforms (λ/ζ) are activated via PDK1 phosphorylation of Thr410 in the activation loop of the enzyme, followed by autophosphorylation of Thr560 causing allosteric alterations that are required for full enzyme activation (Bandyopadhyay et al. 1999; Standaert et al. 1999; Standaert et al. 2001). The involvement of aPKCs in GLUT4 translocation was suggested as result of the following observations. Overexpression of wild type or constitutively active aPKCs in muscle and adipocytes stimulated GLUT4 translocation and glucose uptake, whereas kinase dead mutants of aPKCs inhibited this process (Kotani et al. 1998; Bandyopadhyay et al. 1999). Furthermore, the siRNA-mediated knockdown of PKCζ in muscle cells, and PKCλ in adipocytes resulted in inhibition of glucose transport, an effect that could be rescued by re-expression of the appropriate aPKC using adenovirus (Sajan et al. 2006). Consistent with this, mice with a muscle specific knockout of PKCλ displayed decreased GLUT4 transport and glucose uptake, as well as systemic insulin resistance (Farese et al. 2007). However, other studies report conflicting observations. For example, the siRNA- mediated knockdown of PKC λ/ζ in 3T3-L1 adipocytes had no effect on insulin- stimulated glucose uptake (Zhou et al. 2004). Another study reported that PKCλ knockdown or overexpression of a dominant negative mutant of PKCζ in muscle cells resulted in increased basal and insulin-stimulated insulin signaling and glucose uptake (Stretton et al. 2010). It has been suggested that the discrepancies may be due to comparing results between overexpression of mutants, with knockdown of endogenous protein. It has also been suggested that the differences between the siRNA knockdown results may be due to differing levels of protein knockdown, as well as methods of delivery.

9 Insulin

Insulin PIP2 PIP3 PIP3 receptor Akt PDK1 IRS PI3K

GLUT4 Rab AS160 Vesicle

Figure 1.1 PI3K-Dependent insulin signaling The PI3K-dependent pathway involves insulin binding at the insulin receptor, which is a that phosphorylates proteins like IRS. IRS then recruits other proteins like PI3K, which are activated and brought into proximity of their major substrates, in this case phosphoinositides. Their phosphorylation creates yet a platform to attract other molecules like Akt and its upstream kinases leading to its activation and regulation of downstream effectors leading to GLUT4 translocation.

PI3K-independent insulin signaling

It is proposed that PI3K is necessary but not sufficient to stimulate GLUT4 translocation. For example, activation of PI3K by PDGF or IL-4 stimulation did not cause GLUT4 translocation (Isakoff et al. 1995; Summers et al. 1999). Similarly, expression of constitutively active PI3K was not sufficient to induce full translocation of GLUT4 (Zeigerer et al. 2004). As a result, a second PI3K-independent insulin signaling has also been suggested. One proposed alternate pathway involves the recruitment of two adaptor proteins to the IR, APS and Cbl-associated protein (CAP). CAP contains three Src homology 3 (SH3) domains, which facilitate binding between Cbl and CAP (Ribon et al. 1998). APS functions as an adaptor linking Cbl to the IR, resulting in the tyrosine phosphorylation of Cbl by the IR (Ahmed et al. 2000; Ahn et al. 2004). Upon phosphorylation of Cbl, the CAP-Cbl complex dissociates from the insulin receptor and moves to plasma membrane lipid rafts (Baumann et al. 2000). This localization occurs through the sorbin homology domain of CAP, which binds flotillin, a component of lipid rafts (Bickel et al. 1997; Kimura et al. 2001). The CrkII/C3G complex is recruited to the lipid raft by binding to the phosphorylated residues of Cbl. This interaction occurs via the SH2 domain of CrkII (Tanaka et al. 1994). C3G functions as a guanine exchange factor for the small GTP binding protein TC10 (Baumann et al. 2000; Chiang et al. 2001). TC10 is involved in the regulation of actin dynamics, and is suggested to be involved in modulation of actin dynamics in 3T3-L1

10 adipocytes (Kanzaki et al. 2002). For example, overexpression of dominant negative mutants of CAP and TC10 inhibited actin dynamics and insulin stimulated GLUT4 translocation (Chiang et al. 2001).

Insulin

Flotillin Insulin APS CAP receptor Cbl

CrkII CG3

GLUT4 TC10 Vesicle

Figure 1.2 PI3K-independent insulin signaling The PI3K-independent pathway involves the action of APS, CAP and Cbl, which are recruited to the insulin receptor, leading to the phosphorylation of Cbl. The CAP-Cbl then localizes at lipid rafts, and regulates the activity of downstream effectors such as CRKII, CG3 and TC10 to regulate GLUT4 translocation.

However, the importance of this pathway for GLUT4 translocation has been questioned due to conflicting reports. The siRNA mediated knockdown of CAP, c-Cbl, Cbl-b and CrkII in 3T3-L1 adipocytes had no effect on insulin-stimulated GLUT4 translocation or glucose uptake (Mitra et al. 2004; Zhou et al. 2004). Furthermore, knockout of APS (Minami et al. 2003) and Cbl (Molero et al. 2004) genes in mice did not show inhibition of insulin action, but instead showed a phenotype of increased insulin sensitivity. It has also been reported that Cbl, along with IRS-1 can activate PI3K leading to activation of aPKC, suggesting that the two pathways may not be completely independent (Miura et al. 2004; Standaert et al. 2004). Similarly, it is reported that aPKCζ/λ may interact with TC10 resulting in its phosphorylation (Kanzaki et al. 2004; Saito et al. 2008). Recent studies have reported that activation of the PI3K/Akt pathway is sufficient to stimulated GLUT4 translocation. For example, it was reported that activation of Akt2 was sufficient to stimulate GLUT4 translocation in 3T3-L1 adipocytes to an extent similar to that obtained with insulin treatment (Ng et al. 2008). Similarly overexpression of the PDGF receptor (PDGFR) in 3T3-L1 adipocytes caused activation of PI3K in response to PDGF stimulation. In PDGFR overexpressing adipocytes, PDGF treatment resulted in the activation of PI3K, Akt and TBC1D4 and GLUT4 translocation to the same level as observed with insulin stimulation, suggesting that activation of PI3K is in fact

11 sufficient to stimulate GLUT4 translocation (Hoehn et al. 2008). Therefore in light of recent observations, the significance of this pathway for insulin-stimulated GLUT4 translocation is controversial (Fig 1.2).

Contraction stimulated GLUT4 translocation

In addition to insulin stimulation, it has been shown that contractile activity promotes glucose uptake and GLUT4 translocation in the absence of insulin (Douen et al. 1990; Goodyear et al. 1991; Goodyear et al. 1992; Hayashi et al. 1997; Goodyear and Kahn 1998; Ojuka 2004). In contrast to insulin signaling, contraction does not trigger tyrosine phosphorylation of the IR, IRS-1 or PI3K activity (Treadway et al. 1989; Goodyear et al. 1995). Consistent with this, contraction-stimulated GLUT4 translocation is not affected by the PI3K inhibitor wortmannin (Lund et al. 1995). Hence, it seems evident that contraction and insulin must stimulate glucose transport by distinct signaling pathways. It is proposed that separate calcium- and AMPK-dependent signals contribute to the translocation of GLUT4 in response to muscle contraction.

Increased intracellular calcium

Contraction raises intracellular calcium (Ca2+) concentrations, and it is proposed that it functions as a signal to trigger contraction-stimulated GLUT4 translocation (Holloszy et al. 1986; Holloszy and Hansen 1996). The increase in Ca2+ concentration is thought to act by stimulating the activity of a number of Ca2+-dependent kinases (Wright et al. 2004; Witczak et al. 2010). One such kinase is CaMKII, a protein that requires Ca2+/calmodulin for activation. Expression of a CaMKII inhibitory peptide did not affect insulin-stimulated glucose uptake, whereas contraction-induced glucose uptake was significantly decreased. As a result, it is proposed that in skeletal muscle CaMKII plays a critical role in the regulation of contraction-induced glucose, but does not regulate insulin-stimulated glucose uptake (Witczak et al. 2010). Another candidate for Ca2+ induced GLUT4 translocation is PKC, as muscle contraction has been shown to increase PKC activity, and trigger PKC translocation to the PM (Richter et al. 1987; Cleland et al. 1989; Perrini et al. 2004; Rose et al. 2004). Recently, it has been proposed that PKCε is involved in regulation of contraction-stimulated GLUT4 translocation. This is based on the observation that peptide inhibition and siRNA knockdown of PKCε decreased carbachol (which increased intracellular Ca2+)-stimulated GLUT4 translocation (Niu et al. 2011).

12 AMPK

During muscle contraction, ATP is hydrolyzed to ADP to provide energy. ADP also functions to replenish cellular ATP by donating a phosphate group to another ADP, forming ATP and AMP. As a result, ATP levels fall during contraction while the levels of AMP increase. The decrease in the levels of high-energy phosphates is proposed to induce mitochondrial biogenesis and increase glucose transport (Ojuka 2004). It is suggested that AMP-dependent (AMPK) acts as a "fuel gauge" in mammalian cells, and it has been shown that contraction also leads to activation of AMPK (Hardie et al. 1998; Kemp et al. 1999; Jensen et al. 2008; Lefort et al. 2008). Activation of AMPK is a two step process. First, binding of AMP to the γ subunit of AMPK is thought to trigger a conformational change, which exposes the (Thr172) on the α subunit to phosphorylation (Cheung et al. 2000). Second, phosphorylation of the Thr172 site on the α catalytic subunit activates AMPK (Crute et al. 1998; Stein et al. 2000). The kinase LKB1 in complex with STRAD and MO25, has been identified as an upstream kinase for AMPK (Hawley et al. 2003; Hong et al. 2003; Shaw et al. 2004). Interestingly, LKB1 is also a kinase for other members of the AMPK family (Spicer et al. 2003; Lizcano et al. 2004). Evidence for the involvement of AMPK with GLUT4 translocation comes from the following observations. Stimulation of 3T3-L1 adipocytes with activators of AMPK, namely 5-aminoimidazole-4-carboxamide-1-β-D-ribofuranoside (AICAR) and 2,4- dinitrophenol (DNP), caused GLUT4 translocation to the PM in a wortmannin- insensitive manner (Yamaguchi et al. 2005). However, it was also reported that AICAR treatment of 3T3-L1 adipocytes resulted in attenuated insulin-stimulated glucose transport (Salt et al. 2000). Although this discrepancy may result from the different ways in which GLUT4 translocation was measured. Mice overexpressing a dominant inhibitory mutant of the α2-catalytic subunit of AMPK showed complete inhibition of AICAR-stimulated glucose uptake, but only partially impaired contraction-stimulated glucose uptake. These data suggest that, although the AMPK pathway is involved in contraction-stimulated glucose uptake, other AMPK-independent pathways also contribute (Mu et al. 2001). In contrast, studies using muscle from whole-body α2- or α1-AMPK knockout mice showed no inhibition of contraction-stimulated glucose uptake, although AICAR-stimulated glucose was inhibited in α2 knockout mice (Jorgensen et al. 2004). It is proposed however that the lack of inhibition may be due to

13 compensation by the other remaining catalytic isoform, or it may be that AMPK- independent pathways are activated to compensate for the lack of AMPK. Furthermore, the finding that LKB1 is a kinase for 11 of 12 other members of the AMPK family suggests that members of this family may have a redundant function in contraction- stimulated signaling to GLUT4 translocation (Jessen and Goodyear 2005). It is also reported that AMPK isoforms are activated differently, depending on the duration and intensity of exercise (Jensen et al. 2009). Therefore this may explain some of the discrepancies observed between the different studies.

Contraction

Ca2+ AMP : ATP PKC CAMKII

LKB1 GLUT4 Vesicle AMPK

Figure 1.3 Contraction-stimulated GLUT4 translocation Contraction is proposed to stimulated GLUT4 translocation via calcium- and AMPK-dependent pathways. The rise in intracellular calcium is thought to activate Ca2+-dependent kinases that function to regulate GLUT4 translocation. Alternatively, changes in the levels of high-energy phosphates are thought to activate AMPK-dependent GLUT4 translocation.

Convergence of insulin and contraction signaling pathways leading to GLUT4 translocation

Studies have investigated whether the distinct signaling pathways initiated by insulin and contractions converge at some point in the regulation of GLUT4 translocation. It has been proposed that AS160 represents a potential convergence point of insulin- and contraction-stimulated signaling (Kramer et al. 2006; Cartee and Wojtaszewski 2007). For example, it was reported that both insulin and contraction result in phosphorylation of AS160 (Kramer et al. 2006). Studies to dissect the contributions of each pathway showed that insulin-stimulated AS160 phosphorylation was wortmannin-sensitive and Akt-dependent. In contrast, contraction-stimulated AS160 phosphorylation was only partially wortmannin sensitive, and Akt-independent. This suggests that, while the

14 contribution of AMPK activity to contraction stimulated AS160 phosphorylation is important there are likely to be additional regulatory mechanisms (Kramer et al. 2006). Recently, a novel site in AS160 at S711 was recognized as an AMPK phosphorylation site. It was shown that AMPK could phosphorylate this site in vitro, whereas Akt1, Akt2, or PKCζ did not. Furthermore, in muscle specific AMPK α2 kinase-dead transgenic mice, AICAR and contraction-stimulated phosphorylation of this site was inhibited. Intriguingly, AS160 S711 was also phosphorylated in response to insulin in an Akt2- and rapamycin-independent, but wortmannin-sensitive manner, suggesting this site is regulated by one or more additional upstream kinases. However, expression of an AS160 mutant in which S711 was mutated to (S711A), had no effect on glucose uptake in response to a range of stimuli, so the significance of AS160 S771 phosphorylation for GLUT4 translocation is still unknown (Treebak et al. 2010). AS160 has been suggested to play a role in the regulation of contraction-stimulated GLUT4 translocation through the activity of its calmodulin binding domain (CBD). Expression of a CBD mutant AS160 in mice inhibited contraction-stimulated glucose uptake. Intriguingly, this effect was not associated with aberrant AS160 phosphorylation. (Kramer et al. 2007) Another study compared the regulation of insulin and AMPK signaling pathways in the presence of insulin and the AMPK agonist berberine. In the presence of insulin, PKCζ regulated Akt activation and inhibited AMPK activity. Conversely, in the presence of berberine, PKCζ regulated AMPK activity and inhibited Akt activation. It is therefore suggested that PKCζ plays an important role in the integration of insulin and AMPK signaling and maintenance of glucose homeostasis (Liu et al. 2010). However, other work implicates AMPK upstream of PKCζ (Lee et al. 2011). As such, the evidence for a convergence of insulin and contraction signaling regulating GLUT4 translocation remains inconclusive, and requires further study to resolve this question.

Insulin resistance

Insulin resistance has been linked to multiple defects, which ultimately result in decreased GLUT4 translocation and glucose uptake in response to the insulin signal. Such defects include those which cause attenuation of the insulin signal, including increased inhibitory phosphorylation or dephosphorylation by phosphatases, altered expression of signaling components, or dysregulated trafficking.

15 A number of stimuli are associated with the induction of insulin resistance. These include inflammation and increased levels of inflammatory cytokines (Sethi and Hotamisligil 1999), elevated plasma levels of free fatty acids (Boden 1998) and hyperinsulinemia (Shanik et al. 2008). Due to the importance of the insulin receptor and the IRS proteins in the propagation of the insulin signal, studies have focused on these proteins as sites of dysfunction in insulin resistance. Indeed, it has been shown that in response to different stimuli, multiple kinases phosphorylate IRS1/2 causing either attenuation of the insulin signal, or degradation of IRS (Boura-Halfon and Zick 2009).

Inflammation and insulin resistance

JNK Proinflammatory cytokines such as TNFα, IL-6 and IL-1β activate a number of downstream kinases including JNK, IKKβ and NFκB (Shoelson et al. 2006). In response to TNFα, JNK phosphorylates Ser307 of IRS1 (Aguirre et al. 2000). This disrupts the interaction of IRS1 with the IR, and prevents IRS1 phosphorylation (Aguirre et al. 2000; Aguirre et al. 2002). Mice deficient in JNK are protected from obesity-induced insulin resistance (Hirosumi et al. 2002). Consistent with this, it was reported that knockdown of JNK rescues 3T3-L1 adipocytes from insulin resistance induced by mitochondrial dysfunction (Kim et al. 2009). Recently it was reported that (JNK) interacting protein 1 (JIP1) acts as a scaffold protein that mediates JNK activation. Transgenic mice with an inhibitory mutation in JIP1 show a defect in JNK activation and are protected against obesity-induced insulin resistance (Morel et al. 2010). It has also been shown that JNK activity is elevated in obesity, providing further evidence for the role of JNK in insulin resistance (Hirosumi et al. 2002). IKKβ The transcription factor nuclear factor κB (NF B) is a primary regulator of inflammatory responses (Barnes and Karin 1997). IKKβ activates NFκB by phosphorylating IκB (inhibitor of NFκB) resulting in its degradation (Karin 1999; Chen et al. 2003). Heterozygous deletion of IKKβ (Ikkβ +/−) protected against the development of insulin resistance during high fat feeding and in obese Lepob/ob mice (Yuan et al. 2001). Interestingly, the effect of IKKβ on insulin sensitivity is dependent on its site of action. Mice lacking IKKβ in hepatocytes retain liver insulin responsiveness, but develop insulin resistance in muscle and fat in response to high fat

16 diet, obesity or aging (Arkan et al. 2005). Conversely, overexpression of constitutively active IKKβ in hepatocytes caused profound hepatic insulin resistance, and moderate systemic insulin resistance (Cai et al. 2005). In contrast, mice with a depletion of IKKβ in myeloid cells display whole body insulin sensitivity and are protected from insulin resistance (Arkan et al. 2005). Consistent with this, siRNA mediated silencing of IKKβ in human skeletal muscle prevented TNFα-induced insulin resistance (Austin et al. 2008). IKKβ is reported to interact with IRS1 and is implicated in the Ser307 (equivalent of human Ser312) phosphorylation of rat IRS1 in response to TNFα (Gao et al. 2002). IKKβ, along with IKKα and the scaffold subunit nuclear factor κB essential modulator (NEMO) make up the IKK complex (Yamamoto et al. 2001). It was reported that the motor protein Myo1c and its receptor protein NEMO, act cooperatively to form the IKK-IRS1 complex and facilitate IRS1 Ser307 phosphorylation in response to TNFα stimulation (Nakamori et al. 2006). ERK TNFα also induces ERK activity and it is implicated in the development of insulin resistance. Overexpression of a constitutively active MEK1 mutant, which activates ERK, resulted in inhibition of insulin signaling as measured by tyrosine phosphorylation of IRS1/2 and PI3K activity (Fujishiro et al. 2003). Calorie restriction improves obesity-related insulin resistance through undefined molecular mechanisms. Using obese Zucker rats, it was reported that caloric restriction was associated with reduction of ERK activities in the liver, as well as improved insulin sensitivity (Zheng et al. 2009) P62 signaling adaptor The scaffold protein p62 was initially characterized for its function in localization of the atypical PKCs in signaling pathways (Sanchez et al. 1998). It was shown that p62 null mice develop mature-onset obesity and insulin resistance, and also show increased basal ERK activity. It is proposed that p62 normally inhibits ERK interaction and adipocyte differentiation, and that loss of p62 leads to hyperactivation of ERK resulting in adipogenesis and insulin resistance (Rodriguez et al. 2006). Consistent with this, the double knockout of ERK1 and p62 in mice results in a reversal of their increased adiposity and insulin resistance (Lee et al. 2010). This provides further evidence for the involvement of p62 and ERK in the development of insulin resistance.

17 SOCS proteins The suppressor of cytokine signaling (SOCS) proteins, SOCS1 and SOCS3 have been implicated as negative regulators of insulin signaling. The expression of SOCS1 and SOCS3 is increased in genetic models of obesity, as well as in humans with (Emanuelli et al. 2001; Rieusset et al. 2004). It has been shown that SOCS1 and SOCS3 can bind to the insulin receptor and down-regulate insulin signaling by inhibition of its kinase activity, or by blocking the binding of IRS proteins with the receptor (Emanuelli et al. 2000; Emanuelli et al. 2001; Mooney et al. 2001). SOCS1 and SOCS3 have been shown to target IRS1 and IRS2 for ubiquitin-mediated degradation (Rui et al. 2002). Overexpression of SOCS3 resulted in decreased tyrosine phosphorylation of both IRS1 and IRS2, while SOCS1 overexpression inhibited IRS2 phosphorylation (Ueki et al. 2004). Similarly, overexpression of SOCS1 or SOC3 resulted in reduced glucose uptake in 3T3-L1 adipocytes (Ueki et al. 2004). GSK3 GSK3 is a key regulator of glycogen synthesis (Woodgett and Cohen 1984). Insulin induces the inactivation of GSK3 by Akt mediated phosphorylation (McManus et al. 2005). It has been reported that GSK3 can phosphorylate IRS1 at Ser residues to attenuate the insulin signal, suggesting that GSK3 may induce insulin resistance (Eldar- Finkelman and Krebs 1997; Greene and Garofalo 2002; Liberman and Eldar-Finkelman 2005). GSK3 levels are increased in muscles in type 2 diabetes, and inhibition of GSK3 was shown to improve insulin action and glucose metabolism in human skeletal muscle (Nikoulina et al. 2002). Consistent with this, GSK-3α knockout mice display enhanced glucose tolerance and insulin sensitivity accompanied by reduced fat mass (MacAulay et al. 2007). RNAi reduction of GSK3α expression in human skeletal muscle cells resulted in increases in insulin stimulation and glucose uptake, whereas GSK3α overexpression caused insulin resistance (Ciaraldi et al. 2007). It was recently reported that cellular stress caused GSK3β phosphorylation, leading to phosphorylation of Rictor at Ser1235, a modification which blocked binding of Akt to Rictor thus inhibiting activation of Akt (Chen et al. 2011). Together, these observations suggest that GSK3 may be involved in the development of insulin resistance.

18 PKCθ PKCθ has also been implicated in the regulation of insulin action, however this is controversial. Some observations are consistent with a function of PKCθ in the negative regulation of insulin signaling. For example, PKCθ is reported to phosphorylate IRS1 at Ser 307 or Ser1101 (Yu et al. 2002; Gao et al. 2004; Li et al. 2004). PKCθ knockout mice are protected from fat-induced insulin resistance in skeletal muscle (Kim et al. 2004). However other studies suggest that PKCθ is required for insulin signaling. For example, insulin resistance is associated with reduced PKCθ expression levels in skeletal muscle, and it is suggested that PKCθ is required for the maintenance of insulin sensitivity (Chalfant et al. 2000; Itani et al. 2000; Itani et al. 2002). Consistent with this, mice expressing a muscle-specific dominant-negative PKCθ were glucose intolerant (Serra et al. 2003). Furthermore, PKCθ mice showed a phenotype of reduced energy expenditure and increased risk of dietary obesity and insulin resistance (Gao et al. 2007). Therefore, further work is required to determine the involvement of PKCθ in the regulation of insulin signaling. Tyrosine and serine phosphatases Protein phosphatases catalyze the dephosphorylation of tyrosine- and serine- phosphorylated proteins, thus functioning as negative regulators of signaling pathways (Tonks and Neel 2001; Pais et al. 2009). A number of protein phosphatases have been implicated in the negative regulation of the insulin signal. PTP1B is associated with attenuation of the insulin signal (Cicirelli et al. 1990; Maegawa et al. 1995; Chen et al. 1997). PTP1B directly interacts with the IR (Seely et al. 1996; Bandyopadhyay et al. 1997). Evidence for the role of PTP1B comes from several observations. Overexpression of PTP1B in rat adipocytes inhibited insulin-stimulated GLUT4 translocation, presumably by dephosphorylation of the IR and IRS1 (Seely et al. 1996; Galic et al. 2005). Conversely, PTP1B deficient mice show increased insulin sensitivity (Elchebly et al. 1999; Klaman et al. 2000). Moreover, PTP1B reduction in mice displaying a diabetic phenotype, partially overcame insulin resistance (Xue et al. 2007). Together, these findings suggest that PTP1B may be a target for treatment of type-2 diabetes. Phosphoinositides are important mediators of insulin signaling, due to their control of Akt localization and subsequent activation. Two phosphatases had been identified that modulate phosphoinositide levels, and in this way, are thought to negatively regulate insulin signaling. One such phosphatase is PTEN (phosphatase and tensin homolog

19 deleted on 10), which hydrolyzes PI(3,4,5)P3 to PI(4,5)P2 (Maehama and Dixon 1998). Overexpression of PTEN is reported to reduce insulin-stimulated GLUT4 translocation and glucose uptake (Nakashima et al. 2000; Ono et al. 2001). Adipose and muscle specific disruption of PTEN was found to be protective in animal models of diabetes, consistent with a phenotype of decreased insulin resistance (Kurlawalla- Martinez et al. 2005; Wijesekara et al. 2005). Similarly, siRNA-mediated knockdown of PTEN in 3T3-L1 adipocytes enhanced insulin action (Zhou et al. 2004). These data support the function of PTEN as a negative regulator of insulin signaling. Furthermore, the finding that PTEN expression is elevated in insulin resistance, suggests that it may play a role in the development of insulin resistance (Lo et al. 2004). SH2-containing inositol phosphatase 2 (SHIP2) functions as a lipid phosphatase, and hydrolyzes PI(3,4,5)P3 to PI(3,4)P2 (Wada et al. 2001). However, its relevance in terms of modulation of insulin signaling is controversial. Evidence for SHIP2 playing a role arose from the following observations. Overexpression of SHIP2 in 3T3-L1 adipocytes inhibits insulin-stimulated Akt2 phosphorylation as well as GLUT4 translocation and glucose uptake (Vollenweider et al. 1999; Wada et al. 2001; Sasaoka et al. 2004). Furthermore, SHIP2 deficient embryonic mouse fibroblasts showed up-regulation of

PI(3,4,5)P3 levels, and Akt2 activity in response to serum stimulation (Blero et al. 2005). Similarly, SHIP2-null mice have increased insulin sensitivity, and are protected from diet-induced obesity (Sleeman et al. 2005). However, it has also been reported that siRNA mediated knockdown of PTEN in 3T3-L1 adipocytes has no effect on insulin- stimulated glucose transport (Choi et al. 2002; Zhou et al. 2004). Therefore, further investigation is required to determine the contribution of SHIP2 to the regulation of insulin signaling. PIKfyve PIKfyve (Phosphoinositide Kinase for five position containing a fyve finger) is an enzyme that produces phosphatidylinositol PI5P and PI(3,5)P2 in mammalian cells (Sbrissa et al. 1999). PIKfyve expression is up-regulated during differentiation of 3T3- L1 fibroblasts to adipocytes (Ikonomov et al. 2002). PIKfyve is proposed to function as a positive regulator of GLUT4 translocation. In 3T3-L1 adipocytes, PIKfyve has been shown to co-localise with PI3K as well as GLUT4 (Shisheva et al. 2001; Berwick et al. 2004). Furthermore, insulin stimulation caused translocation of PIKfyve to the PM, as well as phosphorylation of PIKfyve in a PI3K- and Akt-dependent manner (Shisheva et al. 2001; Berwick et al. 2004; Hill et al. 2010). Overexpression of a dominant negative,

20 kinase-deficient mutant of PIKfyve in 3T3-L1 adipocytes reduced insulin-stimulated GLUT4 translocation (Ikonomov et al. 2002). Similarly, siRNA mediated knockdown of PIKfvye inhibited insulin-stimulated glucose uptake (Ikonomov et al. 2007). Therefore, it is proposed that the production of lipid messengers by PIKfyve is involved in regulation of insulin signaling. Recently, the phosphatase Sac3 was characterized and implicated in PtdIns(3,5)P2 turnover, thus counterbalancing the function of PIKfyve (Sbrissa et al. 2007). RNAi mediated knockdown of Sac3 in 3T3-L1 adipocytes increased GLUT4 translocation and glucose uptake, whereas overexpression of wild- type, but not phosphatase deficient Sac3 decreased GLUT4 translocation in response to insulin. Furthermore, the activity of Sac3 was reduced in response to insulin stimulation. Therefore it is proposed that down regulation of Sac3 increases insulin responsiveness, implicating Sac3 as a novel drug target in insulin resistance (Ikonomov et al. 2007). Much evidence exists to suggest that defects at the level of IR and IRS are involved with the development of insulin resistance. However, it has recently been suggested that many of the defects associated with insulin resistance occur independently of IRS (Hoehn et al. 2008). Therefore, elucidating the molecular components involved in the regulation of insulin signaling is required to fully understand insulin resistance.

Hypothesis and aims

To date many studies have focused on the IRS proteins as the main site of dysfunction in insulin resistance. However, it appears that effects downstream of these proteins may also be important in the pathogenesis of this disease. I propose that as yet unidentified proteins exist that modulate insulin signaling and GLUT4 translocation. Elucidating the various pathways that lead to GLUT4 translocation may lead to the identification of novel proteins that regulate this important process. Additionally, these proteins may be involved in insulin resistance, and as such these may represent potential targets for drug development for treatment of this disease. The specific aims of this study were: 1. To develop an appropriate screening method to facilitate large-scale investigation of GLUT4 translocation 2. To identify novel regulators of basal and insulin-stimulated GLUT4 translocation, through the use of genetic screens

21

Chapter 2

Materials and Methods

22 Reagents, Antibodies and Constructs

Mouse monoclonal HA antibody was purchased from Covance. Alexa488-conjugated transferrin, mouse monoclonal transferrin antibody and anti-mouse Alexa 488 secondary antibody were purchased from Molecular probes. HRP-conjugated β-actin antibody was purchased from Sigma-Aldrich. Mouse monoclonal antibodies against EEA1, GM130 and TGN38 were purchased from BD biosciences. Polyclonal rabbit antibodies against pThr642 AS160, pSer235/236 S6 Ribosomal Protein, Akt, monoclonal rabbit antibodies raised against pSer473 Akt, pSer21/9 GSK 3α/β, Akt1, Akt2 and PI3 Kinase p110α were purchased from Cell Signaling. Insulin was obtained from (Sigma Aldrich), Human recombinant IGF1, PDGF and EGF were obtained from R&D systems. Wortmannin was obtained from Sigma-Aldrich. Pan-Akt inhibitor (G-035608) and PI3K inhibitor (GDC-0941) were provided by Genentech. All siGENOME SMARTpool, ON-TARGETplus SMARTpool and single siGENOME oligo siRNAs were obtained from Dharmacon. The retroviral expression plasmid for HA-GLUT4 (pBabepuro HA-GLUT4), expressing the human GLUT4 with a HA-epitope tag at its first exofacial loop was previously described (Shewan et al. 2003)

Cell Culture and Transfections

3T3-L1 fibroblasts (ATCC) were cultured and passaged in DMEM (Invitrogen) supplemented with 10% v/v FCS (Sigma), penicillin/streptomycin (GIBCO), and

GlutaMAX (GIBCO) at 37°C with 10% CO2. Confluent fibroblasts were differentiated into adipocytes by supplementing standard media with IBMX, biotin, insulin and for 72 h, and then replacing with standard media containing insulin for a further 72 h (Ramm et al. 2006). 3T3-L1 fibroblasts were infected with pBabepuro-HA- GLUT4 retrovirus as described previously (Shewan et al. 2003). After 24 h, infected cells were selected in standard media with 2 μg/mL puromycin. HA-GLUT4 fibroblasts were then grown and differentiated into adipocytes (Shewan et al. 2003). Adipocytes at day 6-7 post differentiation were electroporated with DNA constructs, as described previously (Larance et al. 2005), and used 48-72 h after electroporation.

23 HeLa cells were cultured in DMEM supplemented with 10% v/v FCS (Sigma), 100 U/ml penicillin, 100 μg/ml streptomycin (GIBCO), and GlutaMAX (GIBCO) at 37°C with 10% CO2.

Generation of HeLa cells stably expressing HA-GLUT4

One well of a 6 well plate was seeded with 125,000 HeLa cells. The following day, the cells were transfected with 6 μl of Fugene HD (Invitrogen) and 2 μg linearised pBabepuro-HA-GLUT4. At 48 h after transfection, cells were serially diluted in 10 cm plates at 1:10, 1:100 and 1:1000. After 24 h, HA-GLUT4 expressing cells were selected in standard media with 1 μg/ml puromycin. Puromycin resistant single cell clones were screened for GLUT4 expression by Western blotting and analysis of insulin-stimulated GLUT4 translocation. HA-GLUT4-HeLa cells were maintained in standard media containing 1 μg/ml puromycin. siRNA Transfection (screening)

Dharmacon siGenome and On Target-plus pool and single oligo siRNAs were used to knock down genes of interest in the Kinase and DUB screens. Reverse transfection was performed using Dharmafect1 (Dharmacon), as per the manufacturer’s instructions. Briefly, siRNA transfection complexes containing 10 nM siRNA, 0.1 uL Dharmafect1 and 30 uL Optimem were incubated for 20 min at 25 ˚C. HA-GLUT4-HeLa cells (15,000 cells suspended in 120 uL DMEM, 10% FCS with no antibiotics) were plated onto siRNA complexes. Cells were incubated for 72 h at 37 ˚C before performing various experiments.

Quantitative GLUT4 translocation assay

HA-GLUT4 translocation to the plasma membrane was measured as described previously (Govers et al. 2004). Briefly, either 3T3-L1 adipocytes or HeLa cells stably expressing HA-GLUT4 in 96-well plates were starved for 2 h in serum free DMEM containing 0.1% w/v BSA. After stimulation, cells were fixed with 3% paraformaldehyde (PFA) and immunolabeled with monoclonal anti-HA antibody followed by Alexa 488-labeled secondary antibody in the absence or presence of saponin to analyse HA-GLUT4 at the plasma membrane, or the total HA-GLUT4 content, respectively. Nuclei were stained with Hoechst nuclear stain. After washing, fluorescence (emm 485 nm/exc 520 nm and emm 544 nm/exc 630 nm) was measured

24 using either a fluorescence plate reader or an automated fluorescent imager followed by image analysis using IN Cell Analyzer 2000 (GE).

IN Cell image acquisition

The plates were imaged using the IN Cell Analyzer 2000 (GE). For each well, 9 fields were taken at 20x magnification (Nikon 20x objective, 0.45, ELWD, Corr Collar 0-2.0, Plan Fluor). Acquisition parameters were as follows: DAPI and FITC channels were captured at 0.01-1.5 s, 2x2 binning, laser autofocus. Tiff images were generated and processed using the GE developer image analysis software.

Image analysis

Images were analysed using IN Cell developer toolbox 1.8. The first image analysis step involved nuclear localization as based on the DAPI signal (channel 2). No preprocessing was applied. Object segmentation was applied using kernel size of 19 and segmentation at 50. Post processing consisted of; binary sieving (targets with area greater than or equal to 22.8 pixels), followed by the fill holes function, and binary erosion (kernel size 7). A second round of segmentation was run (kernel size 7, sensitivity 50) in order to determine the nuclear count (sum). The second analysis step involved the HA-GLUT4 specific signal in the FITC channel (channel 1). No preprocessing was applied. Intensity segmentation was used with a minimum threshold in the range of 600-1160 and a maximum threshold in the range of 1900-1680. Post processing consisted of binary sieving (targets with area greater than or equal to 182.6 pixels). The output from measurements in channel 1 consisted of; cytoplasmic area (sum), cell intensity (sum). Data for all fields within a well were summed to calculate the summary output for each well. Data were further processed using Microsoft excel, and final screening analysis was made based on the ratio of (cytoplasmic area x cell intensity)/nuclear count. Values for duplicate wells were averaged and SD calculated for each duplicate well, as well as Z score.

Transferrin uptake assay HA-GLUT4-HeLa cells in 96-well plates were starved for 2 h in serum free DMEM containing 0.1% w/v BSA. Cells were incubated in the presence of 25 μg/mL Alexa488 conjugated transferrin for 30 min at 37 °C. Cells were then fixed in 3% PFA, nuclei

25 were stained with Hoechst nuclear stain, and washed twice in PBS. Cells were imaged using the IN Cell analyzer (GE) using the same settings as for measurement of GLUT4 translocation. Images were quantified using a similar analysis as for quantification of GLUT4 translocation, with the exception that the binary sieving step was excluded during analysis of the FITC channel.

Immunofluorescence microscopy

Cells grown on glass coverslips were serum starved in DMEM containing 0.1% w/v BSA for 2 h. Cells were then stimulated or not with 200 nM insulin for 15 min, washed with cold PBS, fixed with 3% paraformaldehyde, and quenched with 50 mM glycine in PBS and 2 % BSA. Cells were blocked in 50 mM glycine in PBS and 2 % BSA and immunolabeled with monoclonal anti-HA11 antibody. Coverslips were washed twice in PBS and incubated with the appropriate fluorophore conjugated antibody. Coverslips were washed twice and mounted on slides, or for intracellular probing were permeabilised with 0.1% saponin in 50 mM glycine in PBS and 2% BSA and labeled with the indicated antibodies.

Taqman analysis for gene knockdown

Messenger RNA was extracted using an oligo-dT based mRNA capture (mRNA Capture Kit; Roche). Cells in 96 well plates were lysed in 100 μl of lysis buffer per well and samples were processed according to the manufacturer’s instructions. The extracted mRNA was reverse transcribed into cDNA using the Transcriptor High Fidelity cDNA Synthesis Kit, (Roche) according to the manufacturer’s instructions. Quantitative PCR was carried out using sequence-specific, unlabeled primers and a FAM dye-labeled TaqMan MGB probe. The TaqMan gene expression assays were reconstituted to a 1 formulation in a 20 μL reaction containing 10 μL of Taqman Gene Expression master mix (Applied Biosystems) and 10 μL (75 ng) of cDNA template. The TaqMan Gene Expression Assay IDs (Applied Biosystems) used for analysis are listed in Table 2.1. Reactions were performed using an Applied Biosystems Model 7500

RT-PCR system. Quantification was performed using the comparative CT method

(ΔΔCT).

26 Gene Symbol Assay ID Gene Symbol Assay ID 18S Hs99999901_s1 PAK2 Hs02559219_s1 ALPK3 Hs00406434_m1 PAN2 Hs00208356_m1 AURKA Hs01597773_mH PANK4 Hs00217146_m1 BCR Hs01036532_m1 PFKFB3 Hs00190079_m1 BRAF Hs00269944_m1 PKN2 Hs00178944_m1 BUB1 Hs00177821_m1 PRKAG1 Hs00176952_m1 CNKSR1 Hs00178995_m1 PRPF8 Hs01556855_m1 COPB2 Hs00178076_m1 PSMD7 Hs00427396_m1 DAPK3 Hs00154676_m1 RIPK1 Hs00169407_m1 DUSP5 Hs00244839_m1 SENP6 Hs00210213_m1 EIF3F Hs02386975_gH SEPHS2 Hs00539041_s1 FASTK Hs00894816_g1 TNK1 Hs01005578_g1 HIPK3 Hs00178628_m1 TP53RK Hs00369266_m1 MAPK14 Hs00176247_m1 TPK1 Hs01558699_m1 MAST2 Hs00248380_m1 TTBK2 Hs00392032_m1 MED20 Hs00385586_m1 UCK2 Hs00367072_m1 MPZL1 Hs00827792_m1 USP10 Hs00382487_m1 NME1 Hs02621161_s1 USP36 Hs00228241_m1 PAG1 Hs00179693_m1 YOD1 Hs01596929_m1

Table 2.1 Gene expression assays used for Taqman analysis of gene knockdown

Preparation of cell lysates

After stimulation, cells were washed twice with ice-cold PBS and solubilised in M-PER Mammalian Protein Extraction Reagent (Thermo Scientific) with PhosSTOP phosphatase inhibitors (Roche) and Complete protease inhibitor mixture (Roche). Lysates were centrifuged at 14,000 x g for 15 min at 4 ˚C to remove insoluble material. Protein concentration for each sample was measured using the BCA protein assay (Pierce) according to the manufacturer’s procedures. Lysates were stored at -20 ˚C.

Western blotting Cell lysates were heated for 3 min at 95 ˚C, and equal amounts of protein were loaded for each sample in a single experiment on a 4-12% NuPAGE Bis-Tris gel (Invitrogen). Proteins were separated on SDS-PAGE gel at 180 V for ~1 h, and transferred onto nitrocellulose membranes using an iBlot gel transfer device (Invitrogen). Membranes were incubated for 20 min at room temperature in SuperBlock blocking buffer in TBS (Thermo Scientific). Membranes were incubated with primary antibody diluted in 5 % w/v BSA, TBST (1x TBS, 0.1 % Tween-20) at 4˚C with gentle shaking overnight. Following incubation, membranes were washed 3 x 5 min with TBST, and incubated with horseradish peroxidase-conjugated secondary antibodies in 5 % BSA, TBST for 1 h at room temperature. Membranes were washed 3 x 5 min with TBST. Immunoreactive

27 bands were detected by ECL Plus Western Blotting Detection Reagent (GE) and exposed to X-ray film (Kodak Biomax XAR).

28

Chapter 3

Development of a GLUT4 translocation assay in HeLa cells

29 Introduction

The discovery of gene silencing by RNA interference (RNAi) was of great biological importance, leading to the development of new technologies for probing gene function and pathway regulation. RNAi is a RNA-dependent gene silencing pathway in eukaryotic cells which is initiated by double stranded RNA (dsRNA) that is complementary to the target gene (Elbashir et al. 2001). In the cytoplasm, short fragments of double stranded RNA (siRNA) are incorporated into the RISC (RNA- induced silencing complex) protein complex, which facilitates the degradation of mRNA that is complementary to the antisense strand of the siRNA (Zamore et al. 2000; Rand et al. 2004; Ameres et al. 2007). Initially, the ability of RNAi to silence gene expression was demonstrated in Caenorhabditis elegans worms (Fire et al. 1998). Since then, RNAi studies have been conducted in a wide variety of organisms and cells, including Drosophila (Dasgupta and Perrimon 2004), Arabidopsis (Chuang and Meyerowitz 2000), and mammalian cells (Martin and Caplen 2007). Recently the availability of cDNA and genomic sequence data has facilitated the design and construction of genome-scale libraries of RNAi reagents for performing RNAi high- throughput screens in a wide variety of cell types (Echeverri and Perrimon 2006; Mohr et al. 2010). As a result, RNAi screening can now be performed to interrogate gene function in biological processes on much larger scale, allowing for the identification of multiple genes functioning within a pathway, as well as novel functions for genes. This strategy has great potential for drug discovery because it offers the ability to identify novel targets that would otherwise be hard to find. The overarching goal of my project was to use siRNA technology to further dissect the mechanism by which controls GLUT4 translocation. This seemed appropriate, as although we know a reasonable amount about this there are still major gaps. For instance we know that IR signaling involves the action of the IRS1/PI3K/Akt pathway but thereafter it becomes unclear. One substrate, AS160 has been identified but it is clearly not the whole story. Additionally, there may be parallel or feedback pathways that are not yet known and these could all provide novel inroads for drug development. An assay has been developed in our laboratory that quantitatively measures the percentage of GLUT4 translocation to the plasma membrane, thus providing a way to investigate changes in insulin signaling (Govers et al. 2004). In this assay, 3T3-L1 cells are infected with a retrovirus such that they stably express hemagglutinin (HA)-tagged

30 GLUT4. This HA tag is exposed to the outside of the cell only when GLUT4 is incorporated into the plasma membrane. The amount of GLUT4 at the plasma membrane of intact cells can be detected using an anti-HA antibody. This is compared with the total amount of GLUT4, as determined by the binding of anti-HA antibody in permeabilised cells. The important aspect of this assay is that it can be performed in a 96 well plate thus providing the potential to develop this into a medium throughout semi-automated screen. In designing a siRNA screen for G4 translocation I considered several important criteria. First, cells have to be easily transfectable. Studies of GLUT4 translocation have traditionally been conducted in cell lines expressing endogenous GLUT4, such as the murine 3T3-L1 adipocyte cell line, or rat L6 myoblast cell line. However, there are several limitations to using such cell lines, particularly with respect to high throughput screening. One issue is that these cell lines are typically difficult to transfect. Thus, when using these cell lines for conducting large scale screens, low transfection efficiency would be a concern, as even if knocking down the expression of a gene resulted in a phenotype of interest, this effect could potentially be masked against a background of untransfected cells. Second, is cost because if we are going to perform large numbers of assays it is important to use cell lines that do not require a lot of time for culturing. This is a problem for muscle and fat because these cell lines need to be differentiated. This leads to a longer experimental time frame as well as introducing the potential for variation between batches of cells. Third, in terms of druggablity and availability of reagents etc, use of a human cell line was seen to be an important although not essential criterion. A number of human adipocyte and muscle cell lines exist (Vogel 2008), however as with other adipose and muscle cell lines, transfection and culture of these cells may mean that these cell lines are not feasible for use in high throughput screening. Alternatively, there are many other human cell lines available, which although not expressing endogenous GLUT4, may be appropriate, for example HEK293 and HeLa cells. After considerable pilot testing neither 3T3-L1 nor L6 cells were deemed suitable for screening purposes. In the case of L6 they are rat cells and there are no good rat reagents available for screening. 3T3-L1 cells were not appropriate as they consistently showed poor transfectablity, as well as variability due to differences in differentiation between samples. To address this issue, we investigated the use of alternative cell lines. Cell lines that do not express endogenous GLUT4 have previously been used to study

31 GLUT4 translocation. These include CHO cells (Bogan et al. 2001), HEK293 cells (Liu et al. 2009) and HeLa cells (Hernandez et al. 2001). An important consideration was that in order to be insulin responsive, the cell line should express the insulin receptor. Microarray expression analysis revealed several cell lines that express the insulin receptor (Fig. 3.1). I selected HeLa cells as they are a human cell line that has been widely used for screening studies. They are very stable and easy to grow, easily transfected, and they show expression of the insulin receptor. Furthermore, this cell line has previously been used for siRNA screening purposes. One concern with these cells is that they do not normally express GLUT4 and so it is conceivable that they are missing essential machinery that is specific to the process. However they do possess all of the elements of the canonical PI3K/Akt pathway (Wang et al. 1998). Thus determining if HeLa cells expressing GLUT4 would display insulin stimulated GLUT4 translocation was important. In this chapter I describe the establishment of the assay in this cell line. As shown I was able to generate a HeLa cell line stably expressing HA-GLUT4. This cell line is insulin responsive and displays insulin stimulated GLUT4 translocation. Functional characterization of this cell line supports the idea that this could be an alternative method used to study insulin signaling and GLUT4 translocation. I optimized siRNA transfection and GLUT4 translocation assay methods such that this cell line would be of use for high throughput screening for regulators of GLUT4 translocation.

32

#*)1) *",, #*)1)

-,*$

#-+1

".01

 # .+)

#*)1)

 # *.0

"  

 #10

 # #+

#*

#+2. 

#,*  #*

 

01/#

!#,

#,2,

+.*

 

 #

#

10

 -

!#,"#

 ##

 **

#*

-0

 #10#

#./+

 # -/)

.-2

 ##*

#.0*.

.-2#

-21



 #,

 #*)

#.-2

.

 ##,

#-0-

#.,2

 

-1)

#+2

 

 # #+1

*+

*0+

#+.0

*-

 2

#*-.

in various cell lines

$#

0

##-,.

",

#/+)

#/+

 *

 "+).

 #-

#.

 # .++

#+/1

 # #.

 ##,

#.

 # +,

#)0

 

#, #/+

ile of insulin receptor 

#0.

 # ,++

 # ++/

#. 

#+221

#+

#*)

#1

 ##,#

 $#

#*



#2+

#*2

#-

*+

*

 "#+

##+,*

#,

 ##*/

#*

#+

#,

+/0#*

#*

#**/#

 ##*

 #1++/

#.#

#1

"**/

"0//

#,0,

#*.

#*

#1*+

#/)

#*+

#"-/1

#/)%&



 +   ) Panel: glgctissues Metric: affy_mas5sig Project: celllines Chip:HGU133P Probe: 238933_at Figure 3.1 Affymetrix microarray expression prof .)) ,))) +.)) +))) *.)) *)))

33 Results

HeLa cells stably expressing HA-GLUT4 exhibit insulin stimulated GLUT4 translocation.

HeLa cells were transfected with a linearised pBabepuro-HA-GLUT4 expression vector. This vector encodes recombinant GLUT4 containing an HA tag in the exofacial loop, as well as conferring puromycin resistance. Following transfection, cells were grown in media containing puromycin, and single cell clones were selected. These clones were screened for expression of the HA-GLUT4 construct using immunofluorescence microscopy (Fig. 3.2). From this first round of screening, several clones (clone 19, 27, 30 33) showed significant HA-GLUT4 positive staining. The HA-GLUT4 positive clones were re-screened to determine whether the recombinant GLUT4 underwent translocation to the plasma membrane upon stimulation with insulin. Clone 27 did not show any difference in surface GLUT4 staining between basal and insulin-stimulated conditions. Clones 19, 30 and 33 showed an increase in surface GLUT4 staining when stimulated with insulin. The difference between basal and stimulated GLUT4 surface levels was highest in clone 30 (Fig. 3.3). As a result, clone 30 was selected for further characterization and assay development.

HA-GLUT4-HeLa cells express the majority of genes expressed in human adipose and muscle tissue

To explore the applicability of HeLa cells for studying GLUT4 translocation, I compared gene expression between the genes that were expressed in the HA-GLUT4- HeLa cell line with those present in adipose and muscle tissue. To this end, mRNA was isolated from HA-GLUT4-HeLa cells and submitted for microarray analysis. HA-GLUT4-HeLa (n=5), normal muscle (n=35), and normal adipose (n=45) samples from Genelogic were analyzed using Affymetrix MAS5 present/absent p values to determine probe sets that were present in all samples of the respective tissues. A probe set was designated present if the p value was less than 0.02 in all samples of the dataset.

34                                               

Figure 3.2 Screening of single cell clones for HA-GLUT4 expression HA-GLUT4 transfected clones grown in a 96 well plate were incubated in serum free DMEM. Total HA-GLUT4 was determined by labeling with anti-HA followed by Alexa-488-conjugated secondary antibody in permeabilised cells. Plates were read using a fluorescence microtiter plate reader.

                             

Figure 3.3 Screening for GLUT4 translocation in HA-GLUT4 positive clones HA-GLUT4 positive clones were serum starved for 2 h, before treatment with or without 200 nM insulin for 15 min. The amount of HA-GLUT4 at the PM was determined by anti-HA fluorescence immunolabeling of non-permeabilised cells and expressed as a percentage of total cellular HA-GLUT4 as determined by labeling of permeabilised cells. Plates were read using a fluorescence microtiter plate reader.

35 [ Note - There are actually more "normal" Genelogic muscle and adipose samples, but some were removed based on annotation - some samples were noted to be from diabetic patients - or because the samples seemed to be outliers in hierarchical clustering ] Probesets were mapped to EntrezIDs using Bioconductor. Probesets without EntrezIDs were eliminated. The intersections were computed using unique EntrezIDs for each dataset. The significance of the intersection of EntrezIDs for each dataset (i.e. present in all three datasets) was calculating using a binomial test with the number of trials equal to the total number of unique EntrezIDs represented on the chip, the number of successes equal to the number of EntrezIDs present in all three datasets, and probability of success equal to the product of the marginal frequencies of being present in each dataset. As a result of the above calculations, it was determined that a total of 19622 unique EntrezIDs were represented above the cutoff level. Of these, 6823 were present in Adipose, 6037 were present in muscle and 9556 were present in HA-GLUT4-HeLa. It was determined that 4934 EntrezIDs were present in all three samples (Fig. 3.4).

Figure 3.4 Venn diagram of genes expressed in adipose, muscle and HA-GLUT4-HeLa cells HA-GLUT4-HeLa (n=5), normal muscle (n=35), and normal adipose (n=45) samples from Genelogic were analyzed using Affymetrix MAS5 present/absent p-values to determine probesets that were present in all samples of the respective tissues. A probeset was called present if the p-value was less than 0.02 in all samples of the dataset.

36 Localization of GLUT4 in the HA-GLUT4-HeLa cell line

The localization of GLUT4 in 3T3-L1 cells has been extensively studied (Piper et al. 1991; Holman and Cushman 1994; Guilherme et al. 2000). In these and many other cell lines GLUT4 is found in perinuclear structures that often co-localize with markers of the trans-golgi network (TGN). In addition it can be found scattered throughout the cytosol in small vesicles and tubules. These structures have been referred to as the GLUT4 storage vesicles from where GLUT4 translocates to the PM. Therefore it was of interest to investigate whether the localization of GLUT4 in HeLa cells was similar to that observed in 3T3-L1 adipocytes. The localization of the recombinant GLUT4 in the HA-GLUT4-HeLa cell line was investigated by comparing its localization with that of several intracellular markers. Strikingly the localization of GLUT4 in HeLa cells was very similar to that observed in adipocytes. There was concentrated labeling in a perinuclear area and this labeling did not correspond to TGN38 in these cells. Similarly it has been reported that G4 does not co-localize with TGN38 in adipocytes (Shewan et al. 2003). There was partial overlap with endosomal and Golgi markers at this location, such as transferrin receptor (TfR), EEA1 and GM130 but this was not comprehensive. Again, this is consistent with what has been reported for adipocytes. Notably there was also staining of small vesicles throughout the cytosol and these did not appear to co-localize with any of these markers. (Fig. 3.5).

37

Basal Insulin

Figure 3.5 Intracellular localisation of GLUT4 in HA-GLUT4-HeLa cells HA-GLUT4-HeLa cells were serum starved for 2 h before treatment with or without 100 ng/ml IGF1 for 15 min. A) Cells were co-stained with anti-HA and antibodies directed against EEA1, GM130, TfR, or TGN38. Scale bar 10 μm. B) Enlarged images of basal and insulin simulated HA-GLUT4-HeLa cells showing GLUT4 localization. Scale bar 10 μm

38 Insulin, IGF1 and EGF treatment of HA-GLUT4-HeLa cells induces GLUT4 translocation

We had demonstrated that insulin induced the translocation of GLUT4 to the plasma membrane in HeLa cells. This corresponded to roughly a 2-fold increase, which is significantly less than that observed in adipocytes where one typically sees a ~ 4-fold increase. For screening purposes, ideally the fold change between basal and stimulated surface GLUT4 levels should be as high as possible, thus providing an optimal dynamic range for screening. Therefore, I wanted to test if there were other growth factors that could trigger GLUT4 translocation to a level higher than that of insulin. The ability of several growth factors to induce the translocation of HA-GLUT4 to the plasma membrane was investigated. HA-GLUT4-HeLa cells were stimulated with 200 nM insulin, 100 ng/ml IGF1, 100 ng/ml PDGF or 100 ng/ml EGF for 15 min, and GLUT4 translocation to the surface was determined by surface HA staining. PDGF treatment of cells caused little translocation of GLUT4 to the PM. Insulin caused a ~ 2- fold increase in GLUT4 at the PM. Treatment with IGF1 or EGF caused the highest increase in GLUT4 surface levels (~2.5-fold) (Fig. 3.6). This may be related to the relative expression level of the respective receptors for these growth factors.









     

        

Figure 3.6 Effect of growth factors on GLUT4 translocation HA-GLUT4-HeLa cells were serum starved for 2 h before treatment with or without 200 nM insulin, 100 ng/ml IGF1, 100 ng/ml PDGF or 100 ng/ml EGF for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 staining. Plates were read using a fluorescence microtiter plate reader.

39 Comparison of insulin signaling in HA-GLUT4-HeLa cells and 3T3-L1 adipocytes

As previously mentioned, 3T3-L1 adipocytes are widely used to study insulin signaling. I wanted to compare phosphorylation of known targets of the insulin signaling pathway in HeLa cells and 3T3-L1 adipocytes. HA-GLUT4-HeLa cells and 3T3-L1 adipocytes were serum starved before treatment with or without, IGF1, insulin and wortmannin. Total cell lysates were prepared and immunoblotted with antibodies directed against pAS160 Thr642, pAkt Ser473, pGSK 3α/β Ser21/9, pS6 Ser235/236 or total Akt. β- actin was used as a loading control. Treatment with either insulin or IGF1 elicited increased phosphorylation of AS160, Akt, GSK3α/β, and S6 kinase in both HeLa cells and 3T3-L1 adipocytes. IGF1 treatment of HeLa cells showed increased phosphorylation of targets when compared with that caused by insulin. In 3T3-L1 cells however, there was no difference in phosphorylation levels of targets between insulin and IGF1 stimulated cells. In both HeLa cells and 3T3-L1 adipocytes, pre-incubation of cells with wortmannin inhibited the amount of phosphorylation in insulin and IGF1 stimulated cells. (Fig 3.7).

Figure 3.7 Comparison of signaling in HA-GLUT4-HeLa cells and HA-GLUT4-3T3-L1 cells HA-GLUT4-HeLa cells and HA-GLUT4-3T3-L1 adipocytes (serum starved for 2 h) were pre- incubated or not with 100 nM wortmannin for 30 min before treatment with or without 200 nM insulin or 100 ng/ml IGF1 for 15 min. Total cell lysates were prepared and immunoblotted with antibodies directed against pAS160 Thr642, pAkt Ser473, pGSK 3α/β Ser21/9, pS6 Ser235/236, total Akt or β-actin loading control.

40 PI3K and Akt inhibitors block insulin-stimulated GLUT4 translocation in both 3T3-L1 adipocytes and HA-GLUT4-HeLa cells

A significant amount of data exists which establishes PI3K and Akt as critical nodes early in the insulin signaling pathway (Taniguchi et al. 2006). Consistent with this, depletion or inhibition of either PI3K or Akt through siRNA knockdown or pharmacological inhibitors has been shown to decrease insulin-stimulated GLUT4 translocation (Jiang et al. 2003; Gonzalez and McGraw 2006; Chaussade et al. 2007). I wanted to determine if the insulin-stimulated GLUT4 translocation observed in HA- GLUT4-HeLa cells was also dependent upon these upstream components. HeLa cells and 3T3-L1 adipocytes expressing HA-GLUT4 (serum starved 2 h) were pre-incubated or not with 10 μM Akt inhibitor or 10 μM PI3K inhibitor for 1 h, before treatment with or without 200 nM insulin for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 staining. Pre-treatment with the PI3K inhibitor completely blocked insulin-stimulated GLUT4 translocation to (control untreated) basal levels in both HeLa and 3T3-L1 adipocytes. In HeLa cells, basal surface GLUT4 levels in the presence of the PI3K inhibitor were decreased below (control untreated) basal levels. Pre-treatment with the Akt inhibitor partially reduced insulin-stimulated GLUT4 translocation as compared to stimulated levels in control untreated cells. Basal levels of surface GLUT4 remained unchanged in 3T3-L1 cells in the presence of either the PI3K or Akt inhibitors. (Fig 3.8) siRNA-mediated knockdown of Akt and PI3K inhibits GLUT4 translocation in HA-GLUT4-HeLa cells

For siRNA screening purposes, it was important to determine if siRNA-mediated knockdown of known targets of the insulin signaling pathway had the expected downstream effects on GLUT4 translocation in HeLa cells. I had previously demonstrated that the use of chemical inhibitors of PI3K and Akt led to a decrease in insulin-stimulated GLUT4 translocation in HeLas. I next wanted to test if similar results could be observed using siRNA-mediated gene knockdown. Additionally, assay conditions with this cell line were being optimized such that transfections could be performed on a large scale manner. Therefore, I wanted to evaluate the knockdown efficiency when reverse transfection of cells was performed.

41

      

 

 

     



                          

      

Figure 3.8 PI3K and Akt inhibitors block insulin-stimulated GLUT4 translocation in HA- GLUT4-HeLa cells and HA-GLUT4 3T3-L1 adipocytes HeLa cells and 3T3-L1 adipocytes expressing HA-GLUT4 (serum starved 2 h) were pre- incubated or not with 10 μM Akt inhibitor or 10 μM PI3K inhibitor for 1 h, before treatment with or without 200 nM insulin for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 staining. Plates were read using a fluorescence microtiter plate reader.

         







    



       

Figure 3.9 Knockdown of PI3Kα and Akt1 + Akt2 inhibits IGF1-stimulated GLUT4 translocation HA-GLUT4-HeLa cells were transfected with non-targeting control siRNA, PI3Kα siRNA, and Akt1+Akt2 siRNA and incubated for 72 h. A) Cells were serum starved, and then treated with or without 100 ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 staining. Plates were read using a fluorescence microtiter plate reader. (B) Total cell lysates were prepared and immunoblotted with antibodies directed against Akt1, Akt2, PI3Kα or β-actin loading control.

42 HA-GLUT4-HeLa cells were transfected with non-targeting control (NTC) siRNA, PI3Kα siRNA, and Akt1+Akt2 siRNA. I had observed from the gene expression studies that HeLa cells express both Akt1 and Akt2 isoforms. In view of the redundancy that has previously been observed for these proteins in terms of their role in insulin action I decided to perform a double knock down strategy to test the role of Akt in the HeLa system. At 72 h following transfection, cells were serum starved, and then treated with or without 100 ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 staining. A significant reduction in protein levels for all three targets was observed by western blotting of lysates from cells transfected using the reverse transfection method (Fig. 3.9b). Importantly, knockdown of PI3Kα or Akt, inhibited IGF1-stimulated GLUT4 translocation as compared with NTC siRNA transfected cells (Fig. 3.9a). This is consistent with the data we obtained using the pharmacological Akt and PI3K inhibitors.

Development of GLUT4 translocation assay for high throughput screening using the HA-GLUT4-HeLa cell line

In the initial development of the GLUT4 assay in HeLa cells I routinely observed a 2-3 fold increase in GLUT4 translocation to the cell surface (Fig. 3.6, 3.8, 3.9). However, this degree of movement was deemed insufficient for screening purposes and so I next set out to optimize the assay even further. One of the problems with the current assay design is that I noted variability in the basal level of GLUT4 at the cell surface. When running the GLUT4 translocation assay, following staining of cells in 96 well plate, plates are read in a fluorescence microtiter plate reader. However, it is essential to correct for background and non-specific labeling. To achieve this, the assay is performed in the presence of excess HA peptide to compete specific labeling and the remaining signal is deemed the background labeling. However, there is often variability in this value, which is a potential problem because this value is subsequently subtracted from all assay values potentially introducing additional variability to the entire assay. A further problem is that in view of the configuration of most plate readers the sensitivity of measurement of fluorescence signals using these instruments is actually quite poor or certainly less than one typically observes using a fluorescence microscope. A further problem with the plate based assay currently used is that the cell number between wells is assumed to be constant. However, when performing a large scale siRNA screen, it is

43 to be expected that some of the gene knockdowns would affect cell viability or induce cell proliferation. Therefore, it would not be possible to simply compare total fluorescent readings between wells, as decreased signal may be due to lower cell numbers and increased signal may be due to a higher cell density as opposed to actual changes in GLUT4 translocation. To address this issue I investigated the use of automated imaging systems for high content screening of GLUT4 translocation in the HA-GLUT4-HeLa cell line. I tested the IN Cell Analyzer 2000 for this purpose. Previously I had tested the Image Express Micro (IXM) microscope (Molecular Devices) and associated software. However, the images captured using the image express were slightly blurry and had high background fluorescence, thus making it harder to analyze the images. As the images obtained using the INCELL were clearer, and the associated image analysis more reproducible, it was decided to use the INCELL for the screen. The image recognition software used for analysis of images was the software that is provided with the INCELL imager. The main reason being that the output from the machine was in a format that was easily manipulated by the software. In addition, the software was flexible in terms of defining the different ways in which the image could be analyzed, allowing control over different parameters of the analysis. Using this approach, the plate is imaged at 20x magnification (9 fields/well) in the DAPI and FITC channels. The resulting images are then analyzed using the IN Cell analysis toolbox. The DAPI channel is used to determine nuclear count (Fig. 3.10a). Object segmentation, and post-processing including binary sieving and erosion is applied to identify and count the nuclei in the images (Fig 3.10b). The HA-GLUT4 specific signal in the FITC channel is analyzed to measure surface GLUT4 staining (Fig 3.10c). Intensity segmentation is applied to exclude background fluorescence, and post-processing consisting of binary sieving is applied to measure the fluorescence intensity and the area that it covers (Fig 3.10d). Based on these data I was able to calculate the amount of surface HA-GLUT4 staining for each cell within a well. In summary, the Y-axis is a measure of the average amount of fluorescence at the cell surface for each cell within a sample, and correlates to the amount of GLUT4 at the plasma membrane. First, the total amount of fluorescence was determined for each field. This was calculated by multiplying the average fluorescence of the field by the area that was determined to be above background fluorescence. This value was then divided by the number of cells within the field, as determined by the nuclei count to give the average amount of fluorescence per cell. Using this method of

44 analysis, I was able to observe ~6-fold increase in surface GLUT4 in IGF1-stimulated wells when compared with basal (unstimulated) wells (Fig 3.11). This represents a significant improvement in the dynamic range of the assay bringing it into line with expected criteria for high throughput analysis.

Figure 3.10 Image analysis for quantification of GLUT4 at the plasma membrane Representative images from IN CELL Analyzer, captured at 20x magnification. HA-GLUT4- HeLa cells were fixed and stained with DAPI to visualize nuclei, and anti-HA antibody followed by fluorescent secondary antibody to visualize surface GLUT4. Images were analyzed using IN Cell developer toolbox 1.8. Scale bar 100 μm. A) The first image analysis step involves nuclear localization as based on the DAPI signal B) Following object segmentation and post processing the selected area for analysis is shown in red. A second round of segmentation is run in order to determine the nuclear count. C) The second analysis step involves the HA-GLUT4 specific signal in the FITC channel D) Following intensity segmentation and post processing the selected area for analysis is shown in red. The output from measurements in the FITC channel consist of; cytoplasmic area and cell intensity.

45

A)

B)     











                 

Figure 3.11 GLUT4 translocation measured using the IN Cell analyzer HA-GLUT4-HeLa cells were either transfected with non-targeting control siRNA, Akt1+Akt2 siRNA or not transfected (cell only), and incubated for 72 h. Cells were serum starved, and then treated with or without 100 ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 staining. Plates were read using an IN Cell analyzer, and immunofluorescence-based quantitative analysis was performed with the resulting images using the IN Cell Analyzer. A) Representative images taken by IN Cell Analyzer. Scale bar 70 μm B) GLUT4 translocation at the PM as determined by quantitative analysis.

46 Discussion

In this chapter I describe the development of a screen that would ultimately provide the foundation for the remainder of my PhD. After considerable preliminary investigation I opted to use HeLa cells for this purpose. These are human cells that do not express GLUT4 endogenously, however they do express the insulin receptor and elements of the PI3K/Akt pathway are conserved. I have been able to obtain a cell line that exhibits a 6-fold increase in cell surface GLUT4 levels in response to IGF1 stimulation. This process has all of the features of the process in the fat cell in that GLUT4 localization is comparable between these two cell lines, and insulin signaling via the PI3K/Akt axis is conserved. Considerable progress in our understanding of insulin action has been made over the past 50 years. It is now clear that glucose transport, which is one of the major actions of insulin in muscle and fat cells, is controlled by a process which involves the insulin stimulated translocation of GLUT4 to the PM. This involves binding of insulin to its receptor, which is a tyrosine kinase that phosphorylates proteins like IRS1. IRS1 then acts to scaffold other proteins like PI3K, which are activated and brought into proximity of their major substrates, in this case phosphoinositides. Their phosphorylation creates yet another platform to attract other molecules like Akt and its upstream kinases leading to its regulation. There is solid evidence to indicate that this cascade is fundamental to the regulation of GLUT4 translocation. While many Akt substrates have been found, a view has emerged that we still do not have the entire picture (Manning and Cantley 2007). For this reason development of an siRNA screen would be highly desirable. Due to the difficulties inherent in using cell lines that express GLUT4 endogenously, such as 3T3-L1 adipocytes or L6 myotubes, I determined that in order to perform such a screen, I would need to investigate the use of other cell lines. My aim was to develop a GLUT4 trafficking assay in a cell line that would be more amenable to high throughput screening. HeLa cells are a widely used cell line that expresses the insulin receptor. HeLa cells do not express GLUT4 endogenously, and as such may be considered an artificial system to study insulin signaling and GLUT4 translocation. However, microarray data indicates that HeLa cells express most of genes that are expressed in both muscle and adipose tissue. Thus, although HeLa cells may not have some of the tissue specific genes expressed in adipose and muscle, the vast majority of the genes from these tissues are

47 represented. With this in mind, HeLa cells appear to be a viable cell line for screening purposes, with the caveat that any results would subsequently need to be confirmed using a more biologically relevant cell line. I generated a stable single cell line expressing recombinant HA-GLUT4. Treatment of these cells with insulin stimulated the translocation of GLUT4 from the cytoplasm to the plasma membrane, leading to an increase in surface levels of GLUT4. Importantly, this translocation can be measured quantitatively by measuring surface HA-GLUT4 fluorescence. The clone that showed the largest difference between basal and insulin stimulated surface GLUT4 levels (typically ~2-fold increase) was selected for further characterization and assay development. Microscopy was performed to investigate the localization of HA-GLUT4 in HA-GLUT- HeLa cells. HA-GLUT4 showed partial co-localization with EEA1, GM130 and TfR. It has previously been demonstrated that GLUT4 trafficks from the cell surface through early endosomes. However, the subcellular localization of GLUT4 is not restricted to early endosomes, and GLUT4 shows only partial co-localization with EEA1 in both 3T3-L1 adipocytes and rat myocytes. (Slot et al. 1991; Patki et al. 1997) Similarly, GLUT4 has been reported to show partial co-localization with GM130 as well as the TfR in rat muscle fibers (Ploug et al. 1998). Finally, TGN38 is an integral membrane protein that has been shown to be predominantly localized to the trans-Golgi network (Stanley and Howell 1993). HA-GLUT4 showed little overlap with TGN38, which is consistent with previous studies carried out in 3T3-L1 adipocytes (Martin et al. 1994; Shewan et al. 2003). Based on these results, HA-GLUT4 expressed in HeLa cells displays similar cellular localization when compared with GLUT4 expressed in adipocytes or muscle cells. This provides confidence that in HeLa cells, GLUT4 is sorted correctly and potentially trafficks in a comparable way. Based on the results indicating HA-GLUT4-HeLa cells were insulin responsive, and displayed insulin stimulated GLUT4 translocation I then optimized the GLUT4 translocation assay conditions for high throughput screening of this cell line. The assay relies on the measurement of surface levels of GLUT4. Therefore a larger assay window between basal and insulin stimulated values provides a greater dynamic range, which is an important consideration when running large screens. I tested a number of growth factors to determine which would stimulate the greatest increase in surface GLUT4. As previously observed, insulin stimulated a roughly 2-fold increase of GLUT4 surface levels. However, treatment with either IGF1 or EGF resulted in a slightly greater

48 increase of GLUT4 surface levels when compared with insulin. IGF1 signals via both the insulin and IGF1 receptors and has previously been shown to increase GLUT4 translocation and glucose uptake in a similar manner to insulin (Weiland et al. 1991; Wilson et al. 1995). In order to balance the need for a larger assay window, yet retain relevance to studying the complete insulin signaling pathway and GLUT4 translocation we decided to use IGF1 when performing our library screens. To further characterize the HA-GLUT4-HeLa cell line, I compared signaling of insulin and IGF1 in these cells with that of 3T3-L1 adipocytes, which have typically been used to study insulin signaling and GLUT4 translocation. In both these cell lines, treatment with insulin and IGF1 lead to the phosphorylation of Akt and downstream targets AS160, GSK-3α/β, S6 ribosomal protein. Prior incubation with the PI3K inhibitor wortmannin inhibited phosphorylation of these targets in both HeLa and 3T3-L1 cells. Interestingly, in HeLa cells, IGF1 stimulation caused slightly higher levels of Akt, GSK-3α/β and S6 ribosomal protein phosphorylation. This may correlate with the increased GLUT4 translocation observed in HA-GLUT4-HeLa cells treated with IGF1 as compared with insulin treatment. Furthermore, treatment of HeLa cells and 3T3-L1 cells with Akt or PI3K inhibitors lead to the inhibition of insulin-stimulated GLUT4 translocation. Together with the signaling data, this suggests that components of the insulin signaling pathway are present and functional in the HA-GLUT4-HeLa cell line. In terms of high content siRNA screening, an important consideration is siRNA transfection efficiency and gene silencing. Secondly, getting the expected phenotype in siRNA transfected cells is vital. Transfection of HA-GLUT4-HeLa cells with siRNA targeting PI3Kα and Akt1 + Akt2 resulted in inhibition of IGF1 stimulated GLUT4 translocation as expected. Western blotting of cell lysates confirmed protein knockdown was effective. This provides confidence that this cell line is easily transfected with siRNA, and RNAi mediated knockdown of known components of the insulin signaling pathway inhibits IGF1 stimulated GLUT4 translocation as expected. This suggests that it would be an appropriate cell line for siRNA screening purposes. Use of the HA-GLUT4-HeLa cell line could enable the investigation of insulin stimulated GLUT4 translocation in a high throughput manner. I have shown that insulin and IGF1 stimulates GLUT4 translocation to the plasma membrane, demonstrating that much of the GLUT4 translocation machinery is conserved in this cell line. The inhibition of Akt and PI3K function via the use of inhibitors or RNAi knockdown,

49 caused inhibition of IGF1 stimulated GLUT4 translocation as expected, suggesting that components of the insulin signaling pathway are shared between both 3T3-L1 adipocytes and HeLa cells. I have performed experiments to optimize siRNA transfection efficiency and assay conditions in this cell line such that the screen can be performed in a high throughput manner. The benefit of using this cell line is that the cells are easier to transfect, do not require differentiation, and as they have been derived from a single cell clone, consist of a homogeneous population, therefore leading to less variation between experiments. A major adaptation to the assay was the use of automated imaging and image analysis to measure GLUT4 translocation. This resulted in an improvement to the dynamic range of the assay and I was able to observe an approximately 6-fold increase in surface GLUT4 in IGF1-stimulated wells when compared with basal wells. This is comparable to GLUT4 translocation that is observed in other assays. For example, in 3T3-L1 adipocytes an ~8-fold increase in cell surface GLUT4 levels is observed in response to insulin stimulation (Govers et al. 2004). Similarly, an assay using photolabeling of endogenous GLUT4 in rat and human skeletal muscle reports a six to seven-fold increase in GLUT4 exocytosis (Karlsson et al. 2009). Therefore, in the HA-GLUT4- HeLa cell line I am able to observe GLUT4 translocation, which is comparable to that which is observed in cell lines such as adipocytes and myotubes. Based on these results, we believe that this HA-GLUT4-HeLa cell line is suited for high throughput RNAi screening for regulators of insulin stimulated GLUT4 translocation.

50

Chapter 4

Identifying kinase regulators of insulin signaling and GLUT4 translocation by siRNA screening

51 Introduction

Protein kinases catalyze the reversible phosphorylation of target proteins and lipids. This phosphorylation serves to alter the target’s activity, subcellular localization, binding properties or association with other proteins. In this way, kinases are involved in mediating signal transduction and controlling a diverse range of cellular processes including metabolism, transcription, cell cycle-progression, cytoskeletal arrangement and cell movement, apoptosis and differentiation. (Manning et al. 2002) The signaling pathway that is activated upon insulin stimulation and which ultimately results in GLUT4 translocation and glucose uptake, involves the activity of multiple downstream targets, many of which have yet to be identified. Protein phosphorylation is a vital process in both the propagation of the insulin signal, as well as its regulation. As such, a number of kinases have been identified as important components of the insulin signaling pathway. Furthermore, several kinases have been found to function in regulatory feedback roles, or as mediators of insulin resistance.

The insulin receptor – kinase activity and regulation by phosphorylation

The insulin receptor (IR) is a tyrosine kinase receptor. Insulin binding to its receptor stimulates its intrinsic kinase activity resulting in receptor autophosphorylation, as well as phosphorylation of a number of insulin receptor substrate (IRS) proteins (Kahn and White 1988). Insulin receptor activity is regulated by multi site phosphorylation. Several tyrosine residues on the β subunit have been identified as autophosphorylation sites. A cluster of three tyrosines (1158, 1162 and 1163) located in the activation loop of the kinase domain, confer increased kinase activity towards substrates when they are phosphorylated (Hubbard 1997). Thus, even when insulin is no longer bound to the receptor, the kinase activity remains, as long as dephosphorylation of these residues does not occur. The insulin receptor also undergoes serine and threonine phosphorylation (Kasuga et al. 1982). In contrast to tyrosine phosphorylation, serine and threonine phosphorylation is typically associated with receptor kinase inactivation (Roth and Beaudoin 1987; Takayama et al. 1988).

52 Phosphorylation of IRS proteins and modulation of the insulin signal

The insulin receptor substrate proteins (IRS) are targets of the IR for phosphorylation. They have a highly conserved N-terminus, containing a PH (pleckstrin homology) domain that is flanked by a phospho-Tyr-binding (PTB) domain (Voliovitch et al. 1995; Eck et al. 1996). In contrast, the C-terminus is poorly conserved, although it contains multiple Tyr phosphorylation motifs that serve as binding sites for proteins containing SH2 domains (Le Roith and Zick 2001). Tyrosine phosphorylation of IRS1/2 by the IR facilitates the binding and activation of PI3K, and the subsequent activation of downstream insulin signaling components (Backer et al. 1992; Shoelson et al. 1993; Sun et al. 1993). In addition to Tyr phosphorylation, IRS proteins also contain a large number (~70) of potential Ser/Thr phosphorylation sites (Zick 2004). It has been shown that Ser/Thr phosphorylation of IRS proteins is involved in both positive and negative regulation of insulin signaling (Gual et al. 2005; Weigert et al. 2008). and it has been postulated that this may also be involved in insulin resistance (Qiao et al. 1999). Accordingly, much work has focused on IRS1 as a site of dysfunction in insulin resistance, and the identification of kinases that mediate the inhibitory Ser/Thr phosphorylation of IRS1 has been an area of concerted investigation (Sykiotis and Papavassiliou 2001). As a result, several kinases that phosphorylate IRS1 have been identified. A number of these kinases are mediators of insulin signaling, and act to negatively regulate signaling during prolonged insulin stimulation. Such kinases include mTor (Li et al. 1999; Gual et al. 2003; Hiratani et al. 2005; Shah and Hunter 2006), p70 S6 kinase (S6K1) (Um et al. 2004), MAPK (De Fea and Roth 1997) and PKCζ (zeta) (Liu et al. 2001; Ravichandran et al. 2001; Moeschel et al. 2004). Another group of kinases are activated along unrelated pathways to inhibit insulin action, and may also function as inducers of insulin resistance (Boura-Halfon and Zick 2009). These kinases include Inhibitory- kappa B kinase β (IKKβ) (Gao et al. 2002), c-Jun NH2-terminal kinase (JNK) (Lee et al. 2003), glycogen synthase kinase-3β (GSK3β) (Eldar-Finkelman and Krebs 1997; Liberman and Eldar-Finkelman 2005; Leng et al. 2010), extracellular signal regulated kinases (ERK) (Zheng et al. 2009), SIK-2 (Horike et al. 2003) and PKCθ (theta) (Griffin et al. 1999; Li et al. 2004). The IRS kinases play an important role in the control of insulin sensitivity, and may represent potential drug targets for the treatment of insulin resistance. However it is

53 suggested that many of the defects associated with insulin resistance occur independently of IRS (Hoehn et al. 2008). Indeed, several kinases have been shown to exert an effect on insulin signaling and GLUT4 translocation independent of IRS phosphorylation.

Kinase regulation of insulin signaling and GLUT4 translocation independent of IRS phosphorylation

Recently, Cdk5 has been implicated in the regulation of insulin-stimulated glucose transport and the pathogenesis of insulin resistance. Insulin stimulates the activation of Cdk5, which results in the phosphorylation of a number of membrane associated targets including E-Syt1 and TC10α (Okada et al. 2008; Lalioti et al. 2009). The siRNA mediated knockdown of Cdk5 in 3T3-L1 adipocytes causes altered GLUT4 translocation and glucose transport (Okada et al. 2008; Lalioti et al. 2009). Furthermore, it has been reported that obesity is linked with activation of Cdk5, which phosphorylates peroxisome proliferator-activated receptor gamma (PPARγ) leading to the altered expression of several genes including adiponectin. Thus, is it suggested that Cdk5- mediated PPARγ phosphorylation is involved in the development of insulin resistance (Choi et al. 2010). Another kinase that is reported to be a regulator of insulin signaling is G protein- coupled receptor kinase 2 (GRK2). In vivo, GRK2 levels correlate with insulin sensitivity, and in culture, altering GRK2 levels in adipocytes and myocytes modulates insulin signaling (Garcia-Guerra et al. 2010). For example, siRNA-mediated knockdown of GRK2 in 3T3-L1 adipocytes increased insulin-stimulated GLUT4 translocation, while overexpression of WT GRK2 inhibited GLUT4 translocation and glucose uptake. Notably, GRK2 effects are independent of IR, IRS1 and PI3K activity. GRK2 interacts with, and inhibits the activity of Gαq/11, a heterotrimeric G protein α subunit reported to mediate insulin-stimulated glucose transport. In this way, GRK2 is likely to function as a negative regulator of insulin signaling to GLUT4 translocation (Usui et al. 2004). The molecular scaffold, kinase suppressor of Ras 2 (KSR2) is also a regulator of energy balance and insulin sensitivity. KSR2 null mice are obese and insulin resistant. KSR2 interacts with AMPK, a master regulator of energy metabolism. KSR2 regulates AMPK, thus promoting glucose uptake and (FA) metabolism. These data

54 have led to a model whereby decreased AMPK activity leads to impaired FA oxidation and increased lipid storage contributing to obesity and insulin resistance (Costanzo- Garvey et al. 2009). Mutations in protein kinases affecting either their function or regulation are associated with a number of diseases. Identifying the kinases that are involved in GLUT4 translocation or insulin signaling is of interest as these may be affected in disease states such as insulin resistance. These kinases may be potential targets for the development of pharmaceutical agonists or antagonists for use in the treatment of this disease. To this end, we performed an siRNA screen to identify kinases that are involved in regulating this important process.

Results

Primary kinase screen

We screened a human kinase library to search for modulators of GLUT4 translocation. The library consisted of siRNA pools (Dharmacon siGENOME SMARTpool; each containing four distinct siRNA oligos) targeting 779 human kinases as well as several controls. The latter included cell only (no siRNA or transfection reagents), transfection reagent only (lipid only), non-targeting control siRNA (NTC), Akt1+2 siRNA and TBC1D4 siRNA. Knockdown of Akt1+Akt2 has previously been shown to inhibit insulin stimulated GLUT4 translocation (Jiang et al. 2003) while knockdown of TBC1D4 increases basal GLUT4 translocation (Eguez et al. 2005; Larance et al. 2005). The siRNAs were arrayed in 96 well plates. HA-GLUT4-HeLa cells were reverse transfected with siRNA pools. Each siRNA pool was transfected in replicates of four. After 72 h, cells were serum starved for 2 h before stimulation with or without IGF1 (100 ng/mL) for 15 min. Of the four replicates, two were left unstimulated (basal) and two were stimulated with IGF1. Cells were then fixed, and stained with anti-HA antibody followed by anti-mouse Alexa488 to visualize surface GLUT4. Nuclei were also stained using Hoechst nuclear stain. Plates were imaged using an IN Cell analyzer. Images were collected at 20x magnification and 9 fields per well. Data were collected in the FITC (HA-GLUT4 signal) and DAPI (nuclear signal) channels. The resulting images were processed by automated image analysis. First, images were analyzed in the DAPI channel to identify nuclei, which provided an estimate of cell number. Next, images were analyzed in the FITC channel to quantify the amount of

55 anti-HA staining (cytoplasmic area and cell intensity) thus indicating the amount of GLUT4 at the surface of the cells. The ratio of cytoplasmic area x cell intensity/nuclear count was calculated as a measurement of the amount of surface GLUT4 per cell. Values for duplicate wells within a treatment were averaged and SD was calculated for each duplicate well. Z scores were calculated in both the basal, and stimulated states. Results are presented in Appendix Table A1. In the basal state, the knockdown of TBC1D4 increased GLUT4 translocation (Zbasal = 0.9). There were 112 genes which, when knocked down, increased GLUT4 translocation in the basal state to this level or higher (Zbasal ≥ 0.9). We selected 30 candidates that increased basal GLUT4 translocation (Zbasal > 1.5), for secondary screening (Fig. 4.1a, 4.1b). Knockdown of Akt1+2 decreased IGF1-stimulated GLUT4 translocation (Zstimulated = -0.75). There were 189 genes, which when knocked down, decreased GLUT4 translocation in the IGF1-stimulated state to this level or lower (Zstimulated ≤ -0.75). We selected 34 candidates in which we observed decreased IGF1-stimulated GLUT4 translocation (Zstimulated < -0.8) for secondary screening (Fig. 4.1c, 4.1d). Importantly, knock down of several known insulin signaling components including the insulin receptor (INSR), IRS1, AKT1, AKT2 and PI3Kβ (PIK3CB), displayed decreased IGF1-stimulated GLUT4 translocation, supporting the integrity of the screen (Fig. 4.2). In total, 64 positive hits were selected in the primary screen; 30 that increased basal GLUT4 translocation, and 34 that decreased IGF1-stimulated GLUT4 translocation. These hits were selected for further characterization.

56





57

     

  

                                       

                             

%$   %$

%$

%$

%$

%$

%$

!%#

          %      "                           

   !            !        !    

Figure 4.1 Targets from primary kinase screen showing increased basal GLUT4 translocation or decreased IGF1 stimulated translocation HA-GLUT4-HeLa cells were transfected with siRNA pools targeting 779 human kinases. At 72 h following transfection, cells were serum starved, and then treated with or without 100 ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining followed by automated image analysis. A) Several candidates that increased basal GLUT4 translocation (Z > 1.5) were identified. B) HA-GLUT4 surface staining of candidate genes as compared to positive control TBC1D4. C) Candidates that decreased IGF1 stimulated GLUT4 translocation (Z < -0.8) were identified. D) HA-GLUT4 surface staining of candidate genes as compared to control Akt1/2 knockdown.

58

# ' &   "$' &

" ' &

!$' &

! ' &

$' &

 ' &

$ ' %

           '        !  "  "  "   !  #   

Figure 4.2 Primary screen data showing effects of knockdown of selected insulin pathway components HA-GLUT4-HeLa cells were transfected with siRNA pools targeting human kinases. At 72 h following transfection, cells were serum starved, and then treated with or without 100 ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining, followed by automated image analysis. Data from selected gene knockdowns is presented for comparison.

59 Secondary screen

To confirm the hits identified in the primary screen a secondary screen was performed. In this round of screening, in addition to basal and IGF1 stimulated conditions, we added replicates that were permeabilised with saponin. This would enable quantification of total GLUT4 levels in the cell so that we could determine if changes in GLUT4 translocation might be due to changes in the total amount of GLUT4 (e.g. transcriptional effect) as opposed to signaling or trafficking effects. As the primary screen was conducted using pooled siRNAs, we also wanted to exclude the possibility of off target effects. For each of the hits selected from the primary screen, we transfected HA-GLUT4-HeLa cells with four single oligos targeting the candidate gene. GLUT4 translocation was measured as in the primary screen. Single siRNA oligo knockdowns showing a phenotype of basal GLUT4 translocation greater than, or within 10% of that observed with TBC1D4 knockdown were considered as positive for increasing basal translocation (Fig. 4.3a). Single siRNA oligo knockdowns showing an effect on IGF1-stimulated GLUT4 translocation less than, or within 10% of that observed with Akt1+2 knockdown were considered as positive for decreasing stimulated GLUT4 translocation (Fig. 4.3b). Hits were considered validated if two or more single oligos gave the same phenotype as that observed in the primary screen (Fig. 4.4). Total GLUT4 levels were determined using the saponin permeabilised replicates (Fig. 4.5). For most targets, total GLUT4 protein levels were comparable to that of the controls. However, a number of knockdowns (COPB2, PRKCL2) showed reduced GLUT4 levels, which may explain the inhibition of GLUT4 translocation in those cells. As a result of secondary screening, 10 of the increased basal, and 23 of the decreased IGF1-stimulated hits were considered validated (Table 4.1).

60

 

   

 

   



  

   

 

 

 

 

 









 

 

 

 RI MPZL1 PAK2 

 O



 

 

 









 

 

 









 

 

 









 

 

42 MAK MAPK14 MBIP MID 

 9



 14 



 OC

  L

 





 0999 





 

 













 

 



 



 AMK1D FRDA KIA 

C

 

 





 B1

U 





 

 













 

 







 



  on of increased basal hits from primary screen

  

 

 



 

  

K2 AKAP1 AKAP11 AKAP7 B  

C   AD  PAK3 PDK4 PI4KII PRKAG1 PRKAR1A PRKCH PRKDC RIPK1 SEPHS2 SMAD7 STK25 STK35 TAO1 TOPK ZAK PAK3 PDK4 PI4KII PRKAG1 PRKAR1A PRKCH PRKDC RIPK1 SEPHS2 SMAD7 STK25 STK35 TAO1 TOPK 

!$# $# !$# $# !$# $# !$" $

          !$# $# !$# $# !$# $# !$" $           Figure 4.3a Deconvoluti

61

al



  

    



 ned by

10% of   

   

transfection,   

 L 

 



 

  

 

 



 

UMPK WEE1   

 

  

as positive for increasing bas

 

  

 





 anslocation less than, or within

 

 













 

 





 from primary screen. At 72 h following



 GLUT4 translocation to the PM was determi 

HIPK3 MAP2K1 MAP3K3 MAPK1 MAST 

 









 

 

 









 

10% of TBC1D4. were considered  

 











 otype on IGF1-stimulated GLUT4 tr

 

 









 

   

 hits from primary screen 

 

 

 

  DAPK3 DGKE DUSP5 EPHA3 FASTK

 

 ngle siRNA oligos targeting candidate genes 

 with or without 100 ng/ml IGF1 for 15 min. 

 

  ation greater than, or within

 

decreasing stimulated GLUT4 translocation. GLUT4 translocation. stimulated decreasing  

 

   

 

 

 

   



 on of decreased stimulated

 

 

   

 

 

 

   

 

 

  NME1 NRBP PAG PANK4 PFKFB3 PIP5K1A PRKCL2 STK6 TLR4 TNK1 TP53RK TPK1 TRIB2 TRIM TTBK2 ARAF1 BCR BRAF CNKSR1 COPB2 CSNK1E

$# !$# $# !$# $# !$" $ Akt1+2 were considered as positive for surface HA-GLUT4 immunofluorescence staining, followed by automated image analysis. Single siRNA oligo knockdowns showing a knockdowns showing oligo siRNA Single image analysis. by automated followed surface HA-GLUT4 immunofluorescence staining, phenotype on basal A) GLUT4 transloc a phen knockdowns showing oligo siRNA B) Single translocation. cells were serum starved, and then treated Figure 4.3b Deconvoluti HA-GLUT4-HeLa cells were transfected with si !$# $# !$# $# !$" $                    

62    

  

                                                          !                      !                          """ "

Figure 4.4 Deconvolution of targets from primary screen Secondary screening of candidates identified the number of oligos for each gene that either increased basal GLUT4 translocation (in blue), or inhibited IGF1 stimulated GLUT4 translocation (in red). Candidates were considered validated if two or more oligos showed the same phenotype as displayed in the primary screen.

Gene # oligos Increased basal BUB1 2 GLUT4 at PM MAK 2 MAPK14 2 MBIP 3 MIDORI 2 MPZL1 3 PAK2 3 PRKAG1 2 RIPK1 2 SEPHS2 2

Decreased stimulated BCR 2 GLUT4 at PM BRAF 2 CNKSR1 4 COPB2 4 DAPK3 2 DGKE 2 DUSP5 2 FASTK 3 HIPK3 2 MAPK1 2 MAST2 2 NME1 3 PAG 3 PANK4 2 PFKFB3 2 PRKCL2 3 STK6 2 TLR4 2 TNK1 2 TP53RK 3 TPK1 3 TTBK2 3 UMPK 3

Table 4.1 Validated genes as based on deconvolution of siRNA pools

63





!

!







 

    !

    ! 

!

!

 

 





  NME1

UMPK !

!

 

 









!

!

 

 









!

!

 

 









!

!

 

 









!

!

 

 







 TLR4 TNK1 TP53RK TPK1 TTBK2

!

!

   

 

 

! !

   

 

 

! !

   

 

 

! !

   

 

 

! !

   

 

BCR BRAF CNKSR1 COPB2 DAPK3 DGKE FASTK HIPK3 MAPK1 MAST2   PAG PANK4 PFKFB3 PRKCL2 STK6

!%$ "%$ %$ "%$ %$ "%$ %$ "%# %

!"%$ !%$ "%$ %$ "%$ %$ "%$ %$ "%# %                     

 

64



 #   ! 

  



 cates (yellow)

#

"

!  

SEPHS2

#

"

!

 

#

"

ence from saponin treated repli

!

 

#

"

!

  B) increased basal GLUT4 translocation

#

"

!  

otted with total GLUT4 fluoresc

# MPZL1 PAK2 PRKAG1 RIPK1

"

!

 

#

"

!

 

#

" !

ed IGF1-stimulated GLUT4 translocation, and  

ated GLUT4 surface staining is pl

#

"

!

 

#

" !

BUB1 MAK MAPK14 MBIP MIDORI  

#$' & # ' & "$' & " ' & !$' & ! ' & $' &  ' & $ ' %  '           targets that A) decreas for validated Figure 4.5 Total GLUT4 levels in validated GLUT4 levels hits from secondary screen Total 4.5 Figure Calculated basal and IGF1 stimul   

65

Tertiary screen For further validation of our targets, positive hits from the secondary screen were transfected with individual siGENOME siRNA oligos, siGENOME SMARTpool siRNA as well as ON-TARGETplus (OTP) SMARTpool siRNA to investigate whether the observed knockdown phenotype was reproducible. Furthermore, in this screen we also measured the degree of knock down for each gene and we also examined transferrin uptake to determine if the effects observed were specific to GLUT4 trafficking. At 72 h following transfection, cells were analysed to measure; GLUT4 translocation, transferrin uptake and target gene knockdown. Of the genes that previously increased basal GLUT4 translocation, BUB1, MAPK14, MPZL1 and PAK continued to show the same phenotype with siGENOME pool knockdown. Furthermore, BUB1 displayed the same phenotype with both pool siRNAs (Fig. 4.6a). Of the genes that decreased IGF1-stimulated GLUT4 translocation, all of the targets re-validated with siGENOME pool and at least two single oligos. Several of these (COPB2, FASTK, MAST2, NME1, PAG, PANK4, PFKFB3, TPK1, TTBK2 and UMPK) also showed a consistent phenotype with both pool siRNAs. These genes are considered high confidence hits (Fig. 4.7a, Table 4.2). To determine whether the observed effects on GLUT4 translocation were specific to insulin-stimulated GLUT4 translocation, or due to more generalized effects affecting vesicle transport, we measured transferrin uptake. At 72 h following siRNA transfection, cells were serum starved for 2 h and incubated with Alexa-488 labeled transferrin for 30 min. Cells were then fixed, washed and imaged. Automated image analysis was used to determine the amount of transferrin uptake. For many genes, there were slight variations in transferrin uptake between siRNA knockdowns, but generally uptake was comparable to controls. However, for a number of genes (COPB2, DAPK3, DUSP5, FASTK), transferrin uptake was drastically reduced suggesting that for these genes there is disruption to general trafficking in the cell (Fig. 4.6b, 4.7b). In addition, quantitative RT-PCR was performed to determine knockdown of target genes. Following transfection (72 h), mRNA was extracted from cells and reverse transcribed into cDNA. Quantitative RT-PCR was performed using this cDNA to quantify knockdown of target genes (Fig. 4.8). Several genes (MIDORI, MPZL1, PAG, TPK1, PFKFB3) did not show expression even in untransfected samples. Testing

66 different primers for these genes would resolve whether this is due to low expression levels, or primers with poor specificity. Of the genes tested here, transfection with siGENOME pool siRNA resulted in > 50% reduction in expression for the majority of genes, with the exception of B-Raf, CNKSR1, DUSP5 and PANK4. OTP siRNA transfection also resulted in efficient knockdown of many of our targets, however in several instances the knockdown was not as effective as siGENOME siRNA, and indeed for several genes (MAPK14, DUSP5, HIPK3, MAST2, PANK4, TNK1, UMPK) there did not appear to be any reduction in expression levels. This may explain the discrepancy between some of the siGENOME and OTP results. We did not have primers for MBIP and TLR4, and thus were not able to examine expression for these genes. As a result of tertiary screening, we have identified 1 hit that increases basal GLUT4 translocation, and 21 hits that consistently inhibit IGF1-stimulated GLUT4 translocation. These genes are of particular interest for further investigation (Fig 4.9).

67  &' &   $' &

#' &

!' &

 ' &

& ' %

$ ' %

# ' %

          ! ' %

 '        ! " #  ! " #  ! " #  ! " #   '!  #               BUB1 MAPK14 MBIP MIDORI

#' &   !' &

 ' &

& ' %

$ ' %

# ' %

          ! ' %

 '         ! " #   ! " #   ! " #   ! " #   ! " #   '!  #                  MPZL1 PAK2 PRKAG1 RIPK1 SEPHS2

68  %&%

$&%

#&%

"&%

!&%

&%

&%

&%

          &     &%     !    !    !    ! & !           BUB1 MAPK14 MBIP MIDORI

 $#

$#

$#

"$"

!$"

$"

$"

          $                           $             

MPZL1 PAK2 PRKAG1 RIPK1 SEPHS2

Figure 4.6 Increased hits from secondary screen – tertiary validation HA-GLUT4-HeLa cells were transfected with single and pool siRNA oligos targeting genes validated in the secondary screen. A) GLUT4 translocation was measured: 72 h following transfection, cells were serum starved, and then treated with or without 100ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining followed by automated image analysis. B) Transferrin uptake was analysed: 72 h following transfection, cells were serum starved, and then incubated with Alexa488-conjuugated transferrin for 30 min. Cells were fixed, washed and imaged in the IN CELL Analyzer. Transferrin uptake was determined by automated image analysis.

69  ' ( &   & ( & % ( & $ ( & # ( & " ( & ! ( &  ( &  (                     ! " #   ! " #   ! " #   ! " #   ! " #   ! " #   ! " #   (!  #                        BCR BRAF CNKSR1 COPB2 DAPK3 DUSP5 FASTK

#' &   !' &

 ' &

& ' %

$ ' %

# ' %

! ' %

 '                     ! " #  ! " #  ! " #  ! " #  ! " #  ! " #  ! " #   '!  #                        HIPK3 MAST2 NME1 PAG PANK4 PFKFB3 PRKCL2

 ) '  ( ) &  ' ) & & ) & % ) & $ ) & # ) & " ) & ! ) &  ) &  )                     ! " #   ! " #   ! " #   ! " #   ! " #   ! " #   ! " #   )!  #                        STK6 TLR4 TNK1 TP53RK TPK1 TTBK2 UMPK

70  $%$ #%$ "%$ !%$ %$ %$ %$ %$ %                                               %                 

$# BCR BRAF CNKSR1 COPB2 DAPK3 DUSP5 FASTK

$#

"$"

!$"

$"

$"

$                                               $                 

$%$ HIPK3 MAST2 NME1 PAG PANK4 PFKFB3 PRKCL2 #%$ "%$ !%$ %$ %$ %$ %$ %                                               %                  STK6 TLR4 TNK1 TP53RK TPK1 TTBK2 UMPK

Figure 4.7 Decreased stimulated GLUT4 translocation hits from secondary screen – tertiary validation HA-GLUT4-HeLa cells were transfected with single and pool siRNA oligos targeting genes validated in the secondary screen. A) GLUT4 translocation was measured: 72 h following transfection, cells were serum starved, and then treated with or without 100ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining followed by automated image analysis. B) Transferrin uptake was analyzed: 72 h following transfection, cells were serum starved, and then incubated with Alexa488-conjuugated transferrin for 30 min. Cells were fixed, washed and imaged in the IN CELL Analyzer. Transferrin uptake was determined by automated image analysis.

71  #   

#

     #

   $               !  "  ! #              #!   "

Figure 4.8 Gene expression analysis by qPCR HA-GLUT4-HeLa cells were transfected with single and pool siRNA oligos targeting genes validated in the secondary screen. 72 h following transfection mRNA was extracted from cells and reverse transcribed into cDNA. Quantitative RT-PCR was performed using this cDNA to quantify knockdown of target genes. Gene expression of siRNA-transfected cells is displayed relative to untransfected (Cell only) samples.

siRNA Gene siGENOME OTP single oligos BUB1 1 1 1 MAPK14 1 0 1 MBIP 0 1 0 Increased basal MIDORI 0 0 1 GLUT4 at PM MPZL1 1 0 1 PAK2 1 0 0 PRKAG1 0 0 1 RIPK1 0 0 1 SEPHS2 0 0 0 BCR 1 0 3 BRAF 1 0 3 CNKSR1 1 0 4 COPB2 1 1 4 DAPK3 1 0 4 DUSP5 1 0 3 Decreased stimulated FASTK 1 1 4 GLUT4 at PM HIPK3 1 0 4 MAST2 1 1 2 NME1 1 1 4 PAG 1 1 3 PANK4 1 1 4 PFKFB3 1 1 4 PRKCL2 1 0 4 STK6 1 0 2 TLR4 1 0 4 TNK1 1 0 3 TP53RK 1 0 3 TPK1 1 1 4 TTBK2 1 1 4 UMPK 1 1 3

Table 4.2 Tertiary screen – Validation of hits

72 

73



Figure 4.9 High confidence hits from DUB screen High throughput images of siRNA transfected HA-GLUT4-HeLa cells. HA-GLUT4-HeLa cells were transfected with; A) siGenome pools of the indicated genes and B) control siRNA. At 72 h following transfection, cells were serum starved for 2 h before treatment with or without 100 ng/ml IGF1 for 15 min. Cells were fixed and stained to detect nuclei (blue), and surface GLUT4 (green). Scale bar 70 μm.

74 Discussion

The kinase screen identified several kinases which, when knocked down, altered GLUT4 translocation. HA-GLUT4-HeLa cells were screened with a siGENOME pool kinase library to identify kinase regulators of insulin signaling and GLUT4 translocation. The primary screen identified 112 genes that increased basal GLUT4 translocation and 189 genes that decreased IGF1-stimulated GLUT4 translocation to the level of our controls (TBC1D4, Akt1+2 respectively). Of these genes, 64 genes were selected for secondary validation (30 increased basal, 34 decreased IGF1-stimulated GLUT4 translocation). Secondary screening, involved deconvolution of targets with the single siRNA oligos that comprised the original siGENOME pool. Of the 64 genes tested, 10 genes that increased basal translocation, and 23 genes that decreased IGF1-stimulated GLUT4 translocation validated with two or more siRNA oligos. Tertiary screening identified 21 genes that showed a strong phenotype of inhibited GLUT4 translocation when knocked down with multiple siRNAs. Furthermore, 1 gene; BUB1 showed increased basal GLUT4 translocation when either of the pool siRNAs was used. These high confidence hits, have varied biological functions and will be discussed in the following section. A recent screen identified components of the PI3K/mTOR pathway as positive regulators of transferrin uptake (Galvez et al. 2007). So as a measure of general trafficking, independent of GLUT4 trafficking, transferrin is probably not the best thing to measure, since many of the important upstream regulators of insulin signaling and GLUT4 translocation, are regulators of transferrin uptake as well.

Kinases known to function in insulin signaling and GLUT4 translocation

Several of the validated target genes have previously been shown to be involved with insulin signaling. This provides confidence that we are indeed identifying bona fide regulators of insulin signaling and GLUT4 translocation with this screen. Connector enhancer of kinase suppressor of Ras 1 (CNKSR1) is a scaffold protein implicated in the regulation of a number of signaling pathways, including Raf-1 activation, Rho mediated JNK MAP signaling and Ras signaling (Jaffe et al. 2004; Jaffe et al. 2005; Ziogas et al. 2005). Of particular interest, CNKSR1 has also been shown to

75 be involved in insulin-stimulated GLUT4 translocation. A proposed mechanism of action is as follows. Insulin stimulates the accumulation of CNKSR1 at the cell surface. Cytohesins are recruited to the CNKSR1 scaffold, stimulating Arf signaling to phosphatidylinositol 4-phosphate 5-kinases (PIP5Ks). This results in the generation of a PIP2 lipid rich membrane microenvironment, which is important for insulin signaling (Lim et al. 2010). For example, expression of PIP5K in 3T3-L1 adipocytes results in increased localization of GLUT4 at the plasma membrane (Kanzaki et al. 2004). CNKSR1 depletion was shown to inhibit insulin-mediated IRS1, PI3K and Akt signaling (Lim et al. 2010). This is consistent with our observation that knockdown of CNKSR1 inhibits IGF1-stimulated GLUT4 translocation. 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase-3 (PFKFB3) is an enzyme involved in control of glycolytic flux (Yalcin et al. 2009). PFKFB3 is highly expressed in adipose tissue, and is phosphorylated in response to insulin stimulation. Overexpression of PFKFB3 in 3T3-L1 adipocytes led to increased glucose uptake (Atsumi et al. 2005). Interestingly, PFKFB3 is a target of PPARγ, and is proposed to be involved in the anti-diabetic effect of PPARγ activation (Guo et al. 2010). In mice, disruption of PFKFB3 causes decreased insulin signaling and exacerbates high fat diet induced insulin resistance (Huo et al. 2010). Combined with the finding that prolonged insulin treatment results in reduction of PFKFB3 mRNA (Atsumi et al. 2005), this suggests that PFKFB3 may be an important site of dysfunction in insulin resistance. We also consistently observed reduced IGF1-stimulated GLUT4 translocation when PFKFB3 was knocked down. Toll-like receptor 4 (TLR) is widely reported to be involved in the pathogenesis of insulin resistance. TLR4 activates pro-inflammatory kinases including JNK, IKK and p38, which are linked to down-regulation of the insulin signal via phosphorylation of IRS1 (Kim and Sears 2010). Furthermore, disruption of TLR4 in mice resulted in improved glucose metabolism, and insulin sensitivity (Shi et al. 2006). However, we observed a decrease in IGF1-stimulated GLUT4 translocation when TLR4 was knocked down, which is not consistent with this model. This is perhaps not surprising since it is conceivable that the role of TLR4 in the whole animal is very different to that observed in individual cells and indeed HeLa cells may also display a unique response compared to adipocytes. Death-associated protein kinase 3 (DAPK3) is also identified as myosin light chain kinase (MLCK), and has been shown to phosphorylate the regulatory light chain (RLC)

76 of myosin II, resulting in activation of the actin-activated MgATPase activity of myosin II (Murata-Hori et al. 1999). Consistent with the results observed in the kinase screen, it has been reported that in 3T3-L1 adipocytes, inhibition of DAPK3 resulted in impaired insulin-stimulated glucose uptake (Fulcher et al. 2008). Recruitment of myosin IIA to the plasma membrane is dependent upon phosphorylation of its RLC by DAPK3 (Lodeiro et al. 2009). It was observed that GLUT4 translocation to the PM was independent of myosin IIA recruitment, however myosin IIA was required for GLUT4 vesicle fusion at the PM, as well as regulating the intrinsic activity of GLUT4 (Steimle et al. 2005; Fulcher et al. 2008). Therefore it is proposed that DAPK3 plays an important role in the activation of myosin IIA, thus regulating GLUT4 vesicle docking, and GLUT4 activity.

Kinases that interact with regulators of insulin signaling

A number of our kinase targets have not previously been linked to insulin signaling or GLUT4 translocation. However they have been shown to interact with proteins identified as regulators of insulin signaling or components of the insulin signaling pathway, such as Akt. Microtubule-associated serine/threonine-protein kinase 2 (MAST2) is a protein involved in the control of NF-kappa B activity (Zhou et al. 2004). I observed reduced stimulated G4 translocation in cells in which MAST2 expression was reduced. It has also been reported that MAST2 contains a PDZ domain that interacts with PTEN. This interaction was shown to increase PTEN stabilization, presumably by facilitating the phosphorylation of PTEN by the kinase domain of MAST2 (Valiente et al. 2005). PTEN is a known negative regulator of insulin signaling, and functions to down-regulate PI3K-dependent insulin signaling (Nakashima et al. 2000). The overexpression of PTEN in 3T3-L1 adipocytes leads to decreased glucose uptake and GLUT4 translocation (Nakashima et al. 2000). It is proposed that phosphorylation of PTEN results in an inactive conformation (Ross and Gericke 2009). One way to interpret our results, is that a reduction of MAST2 would result in reduced phosphorylation of PTEN, leading to an increase in the active, dephosphorylated form that may in turn, inhibit insulin signaling. TP53 regulating kinase (TP53RK) is an atypical protein kinase initially characterized for its ability to regulate p53 activity (Abe et al. 2001). I observed inhibition of IGF1- stimulated GLUT4 translocation in cells transfected with TP53RK siRNAs. Consistent with this, several lines of evidence suggest that TP53RK may be involved in GLUT4

77 translocation. TP53RK is phosphorylated and activated by Akt (Facchin et al. 2007).

Furthermore, TP53RK interacts with Rab35 (PMID:16600182), a Rab whose association with GLUT4 containing vesicles has been detected by mass spectrometry (Miinea et al. 2005; Abe et al. 2006). Therefore it will be of interest to determine if this interaction plays a role in insulin-stimulated GLUT4 translocation. Phosphoprotein associated with glycosphingolipid microdomains (PAG) is a transmembrane adaptor protein. One of the binding partners of PAG is the tyrosine kinase C-Src kinase (CSK). PAG promotes CSK activation and recruitment to lipid rafts (Brdicka et al. 2000). Intriguingly, CSK is reported to associate with IRS1 through its SH2 domain and promote dephosphorylation of the focal adhesion kinase (FAK) in an insulin-dependent manner (Tobe et al. 1996). FAK is an pathway protein shown to regulate the insulin-induced rearrangement of cytoskeletal components important for glucose transport (Baron et al. 1998; El Annabi et al. 2001; Huang et al. 2006). Thus it would be of interest to investigate if the effects of PAG knockdown on GLUT4 translocation are a result of a dysfunction in cytoskeletal arrangement. There also is evidence that CSK can phosphorylate and activate Akt (Lodeiro et al. 2009). Suggesting that PAG knockdown may result in reduced Akt activation. PAG has also been shown to interact with Fyn, a protein that is implicated in the development of insulin resistance (Bastie et al. 2007; Solheim et al. 2008). The genes PRKCL2 and STK6 are both reported to regulate Akt activity. For example, siRNA mediated knockdown of STK6 inhibits Akt activation and increased IkappaBα expression. In contrast, overexpression of STK6 results in up-regulation of Akt activity and down-regulation of IkappaBα. Interestingly, use of pharmaceutical inhibitors suggests that cross talk between STK6 and the PI3K pathway occurs, converging at Akt activation (Yao et al. 2009). Therefore it would be of interest to establish if STK6 represents a PI3K-independent pathway that is involved in insulin signaling. The involvement of PRKCL2 is less well defined. Upon binding a region of PRKCL2 termed the PDK1-interactive fragment (PIF), 3-phosphoinositide-dependent protein kinase-1 (PDK1) can phosphorylate Ser308 and Ser473 of Akt, up-regulating Akt activity (Balendran et al. 1999). However full length PRKCL2 is suggested to play negative role in Akt-mediated downstream events, as overexpression of full length PRKCL2 inhibited Akt phosphorylation by a constitutively active PDK1 mutant (Wick et al. 2000). Thus the role of PRCKL2 in insulin signaling still remains to be clarified.

78 Kinases related to ERK and JNK signaling

B-Raf functions in regulating the MAP/ERK signaling pathway, which affects cell growth, survival, differentiation and secretion (Wellbrock et al. 2004). ERK activity is stimulated by both insulin and contraction (Wojtaszewski et al. 1999). Furthermore, it is reported that glucose activates the ERK1/2 cascade through B-Raf (Duan and Cobb 2010). However, ERK activity does not appear to be required for glucose uptake (Wojtaszewski et al. 1999). Akt has also been shown to phosphorylate B-Raf, inhibiting its activity (Guan et al. 2000). Further investigation is required to establish if B-Raf functions in GLUT4 translocation or if our results are due to off target effects. DUSP5 has been reported to bind directly to ERK1/2 resulting in their dephosphorylation and inactivation, which in turn reduces Ser612 phosphorylation of IRS1 (Mandl et al. 2005; Fu et al. 2006). Therefore it is possible that DUSP5 knockdown results in increased ERK1/2 activity, resulting in increased inhibitory phosphorylation of IRS1, and inhibited insulin signaling. The genes FASTK and HIPK3 are both involved in Fas (TNF receptor superfamily, member 6) signaling. Fas activation inhibits Akt and impairs insulin sensitivity (Wueest et al. 2010). Consistent with this, Fas expression is increased in mouse models of obesity and insulin resistance (Wueest et al. 2010). HIPK3 is involved in mediating Fas signaling, and is reported to inhibit Fas-mediated JNK activation (Rochat-Steiner et al. 2000). FASTK (Fas-activated serine/threonine kinase) is a serine/threonine kinase that is rapidly activated during Fas-mediated apoptosis (Tian et al. 1995). Endogenous FASTK is associated with the mitochondria and is suggested to regulate mitochondrial metabolism in response to Fas induced apoptosis (Li et al. 2004). It has been reported that mitochondrial dysfunction is linked to insulin resistance, thus it may be that knockdown of FASTK affects mitochondrial function, leading to insulin resistance (Kim et al. 2008).

Kinases involved in regulation of trafficking

COPB2 is a subunit of the COPI (coatprotein I) coatomer complex which is essential for Golgi budding and vesicular trafficking (Salama and Schekman 1995). Active PKCε binds to COPB2 (Csukai et al. 1997). It is reported that PKCε is involved in contraction-stimulated GLUT4 translocation in muscle cells. Treatment of cells with carbachol, an acetylcholine analog, stimulates contraction and GLUT4 translocation.

79 Carbachol treatment also results in the concomitant phosphorylation of PKCγ and PKCε, and their translocation to membranes. Furthermore, knockdown of PKCε reduced carbachol-stimulated GLUT4 translocation (Niu et al. 2011). Thus it will be of interest to determine if COPB2 is involved in PKCε -dependent GLUT4 translocation. It should be noted that reduction of GLUT4 translocation may also be due to off target effects, as in other siRNA screens COPB2 knock down has been shown to affect cellular viability and the phenotype is reported to be quite severe in HeLa cells (Gilsdorf et al. 2010).

Metabolic proteins

NME1 is a subunit of the nucleoside diphosphate kinase (NDK), which exists as a hexamer composed of ‘A’ (encoded by NME1) and ‘B’ (encoded by NME2) isoforms. NME1 plays a role in the synthesis of nucleoside triphosphates other than ATP. NME1 is involved in cell proliferation, differentiation and development, signal transduction, G protein-coupled receptor endocytosis, and gene expression (Mehta and Orchard 2009). In this case, NME1 knockdown may result in a decrease of intracellular GTP or other NT pools, which may have a general inhibitory effect. In terms of a GLUT4 specific effect, this target may not be of interest. Thiamin pyrophosphokinase (TPK1) is involved in thiamine metabolism. It is widely expressed at a low level and is suggested to be a housekeeping gene (Nosaka et al. 2001). TPK1 catalyzes the formation of thiamine tyrophosphate (TDP) a signaling intermediate and precursor for thiamine triphosphate (TTP). TDP-dependent including pyruvate dehydrogenase, are essential in (Nosaka et al. 2001). Therefore it is conceivable that disruption of TDP production by knockdown of TPK1 could result in a dysfunction of glucose metabolism by altered GLUT4 translocation. (PANK4) is an essential regulatory enzyme in CoA biosynthesis (Jackowski and Rock 1981; Rock et al. 2000). PANK4 is ubiquitously expressed, particularly showing a high level of expression in muscle (Li et al. 2005). Interestingly, in glucose challenged rat muscle, PANK4 is upregulated (Li et al. 2005). Intriguingly, PANK4 has been reported to interact with M2-type (Pkm2), a regulatory glycolytic enzyme (Li et al. 2005). Therefore, it is proposed that the interaction of PANK4 with Pkm2 may be involved in the regulation of glucose metabolism.

80 Miscellaneous

Finally, a number of genes identified in the screen regulate diverse functions, and are not linked to insulin signaling or GLUT4 translocation. Further investigation is required to determine if these targets are valid regulators of GLUT4 translocation, or the observed changes in GLUT4 translocation are due to off target effects. Budding uninhibited by benzimidazoles 1 homolog (yeast) (BUB1) is a spindle checkpoint regulatory protein involved in control of cell cycle (Williams et al. 2007). In contrast, both non-receptor tyrosine-protein kinase (TNK1) (Hoare et al. 2008) and breakpoint cluster region (BCR) kinase (Radziwill et al. 2003) are involved in negative regulation of cell proliferation. Thus it may be that observed effects on GLUT4 translocation are due to cells undergoing aberrant growth conditions, as opposed to actual changes in insulin signaling. It is perhaps relevant that this screen was performed in proliferating cells as opposed to post mitotic cells. On the other hand one has to be mindful of the fact that kinases often perform diverse functions and so these observations should not be ignored. Tau-tubulin kinase 2 (TTBK2) is a Ser/Thr kinase that putatively phosphorylates residues Ser208 in the tau protein a phenomenon that occurs most prominently in the (Kitano-Takahashi et al. 2007). Current publications link it to the tau cascade and spinocerebellar degeneration. Interestingly, a link between insulin resistance and tau pathology has previously been suggested (Moloney et al. 2008). However, determining whether TTBK2 functions in insulin responsive tissues such as adipose and muscle would establish the relevance of this target for further study.

Conclusion

Kinases play an important role in insulin signaling and GLUT4 translocation, both in the propagation of the insulin signal, as well as regulation of the signaling pathway. Importantly, the identification of several genes with a known involvement in insulin signaling provides confidence in the screening method. Furthermore, our screen identified several genes, which interact with known regulators of insulin signaling, thus potentially expanding the way in which the insulin signaling network is regulated. The identification of genes involved in ERK/JNK/MAP signaling which is implicated in insulin resistance (Lee et al. 2003; Liu and Cao 2009; Zheng et al. 2009) is of interest, as these genes may represent points at which dysfunction may occur. Our screen also identified genes regulating diverse biological functions, some of which include

81 trafficking and metabolic proteins as well as others with functions not currently associated with insulin signaling or GLUT4 translocation. This screen of human kinases enhances our understanding of this complex pathway and how it might be regulated, and provides a list of novel genes for further investigation. Ultimately, this could lead to the identification of regulators of GLUT4 translocation that may be affected in insulin resistance, and which represent potential targets for pharmaceutical treatment of this disease.

82

Chapter 5

Identifying DUB or ULP regulators of insulin signaling and GLUT4 translocation by siRNA screening

83 Introduction

Ubiquitin and Ubiquitin-like proteins

The post-translational modification of proteins by members of the ubiquitin and ubiquitin like protein (Ubl) family is important in the regulation of a range of cellular processes. The modification of proteins by Ubls serves to modulate their function in the cell. This modification may alter interactions with other proteins, resulting in a change in the location, conformation, stability or activity of the target protein. In this way, the Ubl system impacts on the function and regulation of multiple biological pathways (Kerscher et al. 2006). The Ubl family comprises a group of small proteins related to ubiquitin (Table 5.1). The various members of the Ubl family share structural similarities, and are conjugated to their substrates via a related enzymatic pathway. This involves the sequential action of an E1 activating enzyme, an E2 conjugating enzyme and an E3 protein (Fig. 5.1)(Kerscher et al. 2006). Ubls are predominantly conjugated to substrate proteins on lysine residues. Additionally, because ubiquitin contains seven lysine residues, (Peng et al. 2003) this provides seven unique sites to which another molecule of ubiquitin can attach, and thus chain formation can occur, giving rise to linear and branched structures (Pickart and Fushman 2004). Similarly, SUMO-2 and SUMO-3 contain an internal sumoylation site, which enables the formation of polymeric SUMO chains (Tatham et al. 2001; Matic et al. 2008). Ultimately, protein modification by Ubls is potentially more complex than a simple ‘on/off’ phosphorylation event, as chains can have different configurations, depending on what subunits are involved and how they are linked. Indeed, different chain topologies are associated with diverse biological functions (Woelk et al. 2007).

Modifier Function Ubiquitin Substrate degradation, localization, protein interactions, other Substrate localization, protein interactions, endocytosis, DNA repair, Smt3/SUMO1-4 transcriptional regulation, chromatin structure regulation of E3 , transcriptional regulation of p53, Nedd8/Rub1 proteasomal degradation May act in transcription and pre-mRNA splicing during IFN response; ISG15 induced by IFN-α/β Atg8 Autophagy, cytoplasm to vacuole targeting Atg12 Autophagy, cytoplasm to vacuole targeting Urm1 Budding, nutrient sensing, oxidative-stress response Function unknown; Uba5 is induced during the unfolded protein Ufm1 response Fat10 Ubiquitin-independent substrate degradation, apoptosis FUBI/MNSFβ T cell activation Hub1/Ubl5 Pre-mRNA splicing Table 5.1 Ubl family of proteins This is a modified version of a table presented by Kerscher, O. (2006) Annu Rev Cell Dev Biol. 22:159-80

84 Ubl conjugation is reversible, and the removal is catalyzed by Ubl-specific proteases (ULPs) including deubiquitinating enzymes (DUBs), and sentrin-specific proteases (SENPs) that deconjugate the polypeptides from substrates (Gareau and Lima 2010). DUBs and SENPs are metallo- or cysteine proteases. Several DUBs are associated with ubiquitin recycling during protein degradation at the proteasome, but the majority of them have different roles in the cell (Nijman et al. 2005). ULPs balance the Ubl conjugation reactions in the cell, thus dynamically contributing to the regulation of various cellular processes (Clague and Urbe 2006).

AMP ATP + PPi Ub degraded substrate E1 Ub Ub Ub

Ub E2 Ub

26S proteasome

E2 Ub

E3 Non-proteolytic Ub Ub Ub Ub pathway

Target substrate

Figure 5.1 The ubiquitin-proteasome system

Ubiquitin system functions

Early studies of protein ubiquitination focused on its function in protein degradation (Hershko 2005). Typically, polyubiquitination via Lys48 of ubiquitin results in targeting of proteins to the proteasome for degradation (Pickart and Fushman 2004, Thrower, 2000 #94). Since then, it has been found that Ubl modifications play a role in many other biological processes (Mukhopadhyay and Riezman 2007). For example, Lys63 polyubiquitination is not generally associated with proteasomal degradation. Instead, the resultant change in protein structure alters the function of the substrate protein. Such polyubiquitination signals have been involved in DNA repair, signal transduction, protein trafficking and ribosomal protein synthesis (Chan and Hill 2001; Pickart and Fushman 2004). Monoubiquitination, or polyubiquitination (addition of multiple monomeric ubiquitin molecules) is involved in processes such as endocytosis, virus budding, nuclear shuttling and transcriptional regulation (Haglund et al. 2003; Huang and D'Andrea 2006; Mukhopadhyay and Riezman 2007; Salmena and Pandolfi 2007).

85 While phosphorylation plays a major role in regulating protein kinases and signaling cascades, (Yang et al. 2010) more recently ubiquitination has also been shown to play an important role in these processes. One of the best examples of this involves the regulation of nuclear factor kappa B (NF-κB) signaling (Wang et al. 2001; Mukhopadhyay and Riezman 2007). IκB is an inhibitor of NF-κB. Under basal conditions, IκB binds to NF-κB, retaining it in the cytoplasm. Upon activation of this pathway, IκB is degraded and NF-κB is then free to translocate into the nucleus to regulate gene transcription (Skaug et al. 2009). In this case, ubiquitination plays a fundamental role in propagation of the NF-κB signal, whereas deubiquitination, mediated via a DUB, inhibits and down-regulates NF-κB signaling. The ubiquitin- dependent regulation of the NF-κB pathway involves several steps. Briefly, receptor activation leads to the conjugation of Lys63 polyubiquitin chains to TRAF E3 ligases through autoubiquitination (Reyes-Turcu et al. 2009). The Lys63 polyubiquitin chain facilitates the assembly of a complex including TAK1/TAB2/3 protein kinase and its substrate IκB kinase (IKK). This complex catalyzes the phosphorylation and Lys63 polyubiquitination of IKK, causing its activation (Sun et al. 2004). The active IKK phosphorylates IkB, leading to Lys48 polyubiquitination of IkB by β-TrCP (beta- transducin repeat-containing protein), triggering its degradation and releasing its inhibition on NF-κB (Kroll et al. 1999). Down-regulation of the signal is achieved by the DUB A20, which removes the Lys63 polyubiquitin from NF-κB signaling components (Heyninck and Beyaert 2005; Shembade et al. 2010). Another DUB, CYLD has also been identified as in inhibitor of NF-kB signaling. Like A20, CYLD inhibits signaling by disassembling Lys63 polyubiquitin chains, and some of its substrates include TRAF2, TRAF6 and IKK (Kovalenko et al. 2003; Trompouki et al. 2003) Another example of ubiquitin regulation of signaling involves AMPK (AMP-activated protein kinase) activation. LKB1 phosphorylates and activates several AMPK-related kinases including AMPK (Lizcano et al. 2004). The ability of LKB1 to phosphorylate these kinases depends upon their ubiquitination status. Polyubiquitination of AMPK- related kinases via atypical Lys29/Lys33 linkages does not result in degradation, but inhibits LKB1 mediated phosphorylation. Many AMPK-related kinases contain an ubiquitin-associated (UBA) domain that is required for binding to LKB1. The theory is that the UBA domain may bind to the attached polyubiquitin chain, preventing AMPK-

86 related kinases from interacting with LKB1. Deubiquitination is thought to result in release of the auto-inhibitory interaction, enabling the UBA domain to interact with LKB1. USP9X deubiquitinates two AMPK-related kinases; MARK4 and NUAK, stimulating their LKB1 mediated phosphorylation (Al-Hakim et al. 2008). AMPK functions in regulating cellular metabolism (Hardie 2004), and this may be one way in which ubiquitination affects metabolic signaling.

Ubiquitin-like proteins and insulin signaling

Aberrations of the Ubl system are implicated in a number of pathologies, such as the neurodegenerative disorders Huntington’s disease (Steffan et al. 2004) and Alzheimer’s disease (Li et al. 2003). More recently, research has shown that components of the Ubl system affect the insulin signaling pathway, and may also be involved in the pathogenesis of type II diabetes. One mechanism for this involves the degradation of components of the insulin signaling pathway. For example, it has been shown that chronic insulin treatment results in the proteasomal degradation of IRS1 and IRS2 (Sun et al. 1999; Rui et al. 2001). Indeed, it has been reported that IRS1 and IRS2 are targeted for ubiquitin-mediated degradation by SOCS1/3 (suppressor of cytokines signaling protein 1/3), proteins that are up-regulated during inflammation. In this case, ubiquitination is thought to be catalyzed by the elongin BC ubiquitin ligase complex (Rui et al. 2002). IRS1 has also been identified as a target of the E3 ligase Cullin7, which mediates its ubiquitin-dependent degradation. Cullin7 deficient mouse embryonic fibroblasts (MEFs) accumulate IRS1 and display increased activation of IRS1 downstream signaling (Xu et al. 2008). Thus, the ubiquitin-mediated degradation of IRS1/2 contributes to the regulation of the insulin signaling pathway. It has also been demonstrated that modulation of insulin signaling by the Ubl system can occur in a proteasome-independent manner. TRAF6 has been identified as an E3 ligase for Akt, and was shown to induce Lys63 ubiquitination of Akt within its PH domain (Yang et al. 2009). The ubiquitination of Akt by TRAF6 promotes its translocation to the plasma membrane where it is subsequently phosphorylated and activated (Yang et al. 2009). In TRAF6-negative cells, the ubiquitination and -dependent phosphorylation of Akt was inhibited (Yang et al. 2009). This suggests that the ubiquitination of Akt by TRAF6 is an important step in growth factor- dependent Akt signaling.

87 Recently, the DUB ubiquitin carboxyl-terminal (UCH)-L3, has been shown to promote adipogenesis and insulin signaling. Cells lacking UCH-L3 display attenuated insulin signaling and phosphorylation of IR/IGFR and downstream targets were decreased in these cells. Ectopic expression of wild-type UCH-L3 restored phosphorylation of targets, whereas hydrolase activity-deficient UCH-L3 did not. It is proposed that UCH-L3 promotes adipocyte differentiation by promoting insulin signaling in a hydrolase-activity-dependent manner. (Suzuki et al. 2009) Interestingly, the insulin responsive glucose transporter GLUT4 itself is ubiquitinated. In 3T3-L1 adipocytes, a ubiquitin-resistant version of GLUT4 is not sorted from the trans-Golgi network (TGN) into GLUT4 storage vesicles. As a result, in basal conditions this ubiquitin null GLUT4 mutant is incorrectly localized, and does not translocate to the plasma membrane in response to insulin. Therefore, ubiquitination of GLUT4 is an essential step in its insulin-regulated trafficking. Intriguingly, it is suggested that this modification is transient, as it is estimated that approximately 0.1% of total GLUT4 is ubiquitinated in 3T3-L1 adipocytes under steady state conditions (Lamb et al. 2010). This is consistent with other post translational modifications like phosphorylation where under steady state conditions the amount of target that is modified is a function of both the addition and removal of the modifier. The sumoylation of GLUT4 has also been reported (Lalioti et al. 2002). The interaction of GLUT4 with the SUMO ligase Ubc9 has been observed by yeast two hybrid analysis. Overexpression of Ubc9, increased GLUT4 abundance in L6 myoblasts, concomitant with increased insulin-stimulated glucose uptake (Giorgino et al. 2000). Similarly, in 3T3-L1 cells, Ubc9 overexpression caused an increase in the amount of GLUT4 and insulin responsive glucose transport, presumably by inhibiting GLUT4 degradation and promoting GLUT4 targeting to GLUT4 storage vesicles (GSVs). Consistent with this, siRNA-mediated knockdown of Ubc9 accelerated GLUT4 degradation, resulting in reduced insulin-stimulated glucose transport. However, overexpression of catalytically inactive Ubc9 had the same effect as wild type Ubc9, suggesting that GLUT4 stabilization was not due to SUMO modification, but rather Ubc9 binding (Liu et al. 2007). Collectively, these data suggest that the Ubl system plays a major role in the regulation of insulin signaling and GLUT4 translocation. Several Ubl modifiers involved in this process have been identified, although their precise mechanism of action has yet to be fully elucidated. Therefore, the identification of other Ubl modifying family members

88 that modulate insulin signaling is of major interest, as this will provide a greater understanding of how this pathway is regulated, or dysregulated in the diseased state as well as provide novel targets for drug development. To this end, I screened a library of DUB and SUMO proteases to identify novel proteins that regulate GLUT4 translocation

89 Results

Primary screen

HA-GLUT4-HeLa cells were screened with an siRNA library targeting human DUBs and SUMO proteases to identify candidates involved in the regulation of GLUT4 translocation. The library consisted of 106 siRNA pools (siGenome SMARTpools, each containing four distinct siRNA oligos). Screen controls included cell only (no siRNA or transfection reagents), transfection reagent only (lipid only), Non-targeting Control siRNA (NTC), Akt1+2 siRNA and TBC1D4 siRNA. The screen was conducted using the same conditions that were used in the primary kinase screen. Briefly, HA-GLUT4- HeLa cells were transfected with siRNA pools in replicates of four. Following transfection (72 h), cells were serum starved, and then treated with or without 100 ng/ml IGF1 for 15 min (in duplicate). GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining followed by automated image analysis. The results are presented in Table 5.2 and Figure 5.2. In the basal state, knockdown of the control TBC1D4 increased GLUT4 translocation as expected (Zbasal = 0.9). There were 10 genes, which, when knocked down, increased GLUT4 translocation in the basal state to the same extent as observed with the TBC1D4 siRNA or higher (Zbasal ≥ 0.9). I selected 6 candidates that increased basal GLUT4 translocation (Zbasal > 1.5) for secondary screening (Fig. 5.3). In the IGF1-stimulated state knockdown of the Akt1 and Akt2 isoforms simultaneously decreased IGF1- stimulated GLUT4 translocation (Zstimulated = -0.75). There were 43 genes, which when knocked down, decreased GLUT4 translocation in the IGF1-stimulated state to the same or greater extent compared with Akt knock down (Zstimulated ≤ -0.75). I selected 10 candidates, that when knocked down resulted in a decrease in IGF1- stimulated GLUT4 translocation to less than 25% of NTC transfected cells (Zstimulated < -1.6), for secondary screening (Fig 5.3).

90

Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD Cell only 6.043E+07 3.642E+07 2.681E+08 5.425E+07 -0.30 0.41 0.94 0.55 Lipid Only 7.129E+07 4.240E+07 2.915E+08 5.190E+07 -0.14 0.38 1.15 0.50 NTC 4.420E+07 3.014E+07 1.893E+08 4.528E+07 -0.54 0.28 0.22 0.38 TBC1D4 1.395E+08 4.930E+07 2.430E+08 5.921E+07 0.90 0.60 0.69 0.49 AKT1/2 5.676E+07 4.817E+07 8.020E+07 5.271E+07 -0.37 0.42 -0.75 0.35 C15ORF16 1.699E+07 1.911E+07 6.515E+07 5.413E+06 -0.88 0.24 -1.49 0.05 CGI-77 8.545E+07 4.667E+07 2.999E+08 8.197E+06 -0.01 0.59 0.81 0.08 DKFZP761A052 5.218E+07 4.724E+07 1.853E+08 7.807E+07 -0.43 0.60 -0.31 0.76 HSHIN1 1.882E+07 1.740E+07 1.341E+08 2.865E+06 -0.85 0.22 -0.81 0.03 FLJ25831 1.402E+08 4.151E+07 3.108E+08 7.238E+07 0.68 0.53 0.92 0.71 OTUD1 1.127E+08 1.376E+07 2.494E+08 7.019E+06 0.34 0.17 0.32 0.07 OTUB1 1.377E+07 9.546E+06 7.757E+07 1.171E+07 -0.92 0.12 -1.36 0.11 OTUB2 1.043E+08 2.743E+07 2.475E+08 2.873E+06 0.23 0.35 0.30 0.03 TNFAIP3 2.970E+07 2.661E+07 1.722E+08 3.037E+07 -0.72 0.34 -0.44 0.30 VCIP135 9.686E+06 9.429E+06 7.435E+07 1.605E+07 -0.97 0.12 -1.40 0.16 ZA20D1 1.061E+08 1.952E+07 1.944E+08 4.047E+07 0.25 0.25 -0.22 0.40 ZRANB1 2.486E+07 9.895E+06 1.094E+08 1.638E+07 -0.78 0.13 -1.05 0.16 USP1 5.171E+07 3.537E+07 2.493E+08 2.588E+07 -0.44 0.45 0.32 0.25 USP2 6.311E+07 6.767E+07 1.684E+08 1.036E+08 -0.29 0.86 -0.48 1.01 USP3 1.485E+08 6.311E+07 2.683E+08 1.101E+08 0.79 0.80 0.50 1.08 USP4 1.142E+08 8.243E+07 2.636E+08 8.706E+06 0.35 1.04 0.46 0.09 USP5 7.740E+07 5.052E+07 2.321E+08 6.470E+07 -0.11 0.64 0.15 0.63 USP6 7.732E+07 5.023E+06 2.367E+08 1.177E+07 -0.11 0.06 0.19 0.12 USP7 5.042E+06 3.750E+06 4.533E+07 9.735E+06 -1.03 0.05 -1.68 0.10 USP8 1.759E+07 1.382E+07 1.324E+08 5.586E+06 -0.87 0.18 -0.83 0.05 USP9X 1.504E+08 3.015E+07 2.588E+08 1.121E+07 0.81 0.38 0.41 0.11 USP9Y 1.331E+08 9.793E+07 2.777E+08 8.249E+07 0.59 1.24 0.59 0.81 USP10 3.160E+08 4.061E+07 4.606E+08 3.070E+07 2.91 0.51 2.38 0.30 USP11 3.195E+07 1.422E+07 6.078E+07 1.459E+07 -0.69 0.18 -1.53 0.14 USP12 2.546E+08 5.177E+07 3.080E+08 9.758E+07 2.13 0.66 0.89 0.96 USP13 1.400E+08 3.654E+07 2.587E+08 6.230E+07 0.68 0.46 0.41 0.61 USP14 1.685E+07 1.629E+07 8.772E+07 1.480E+07 -0.88 0.21 -1.27 0.14 USP15 9.957E+07 4.323E+07 2.926E+08 6.581E+06 0.17 0.55 0.74 0.06 USP16 7.948E+07 7.576E+07 2.913E+08 5.042E+07 -0.09 0.96 0.73 0.49 DUB3 2.501E+07 2.607E+07 1.163E+08 1.636E+07 -0.78 0.33 -0.99 0.16 USP18 3.451E+07 3.409E+07 1.048E+08 6.325E+07 -0.66 0.43 -1.10 0.62 USP20 1.173E+08 4.720E+07 1.999E+08 2.154E+07 0.39 0.60 -0.17 0.21 USP21 1.469E+07 1.538E+07 1.041E+08 4.728E+07 -0.91 0.19 -1.11 0.46 USP22 1.659E+08 8.941E+07 2.529E+08 6.886E+07 1.01 1.13 0.35 0.67 USP25 1.197E+08 8.150E+07 3.265E+08 4.610E+07 0.42 1.03 1.07 0.45 USP26 1.566E+08 1.050E+07 3.223E+08 1.201E+07 0.89 0.13 1.03 0.12 USP28 2.961E+07 1.126E+07 1.724E+08 9.923E+06 -0.72 0.14 -0.44 0.10 USP29 8.788E+07 1.579E+07 2.544E+08 1.322E+07 0.02 0.20 0.37 0.13 USP30 1.268E+08 8.660E+07 2.480E+08 2.585E+07 0.51 1.10 0.30 0.25 USP31 5.080E+06 4.954E+06 3.656E+07 9.643E+05 -1.03 0.06 -1.77 0.01 USP32 7.333E+07 2.497E+07 2.252E+08 4.281E+07 -0.16 0.32 0.08 0.42 USP33 6.007E+07 5.520E+07 1.581E+08 6.333E+07 -0.33 0.70 -0.58 0.62 USP35 2.492E+08 5.254E+07 3.211E+08 6.319E+07 2.06 0.67 1.02 0.62 USP36 6.088E+06 5.497E+06 3.008E+07 1.388E+07 -1.02 0.07 -1.83 0.14 USP37 1.838E+08 3.021E+07 3.613E+08 8.827E+07 1.24 0.38 1.41 0.86 USP38 8.835E+07 3.240E+07 2.593E+08 7.430E+07 0.03 0.41 0.41 0.73 USP40 1.207E+07 1.241E+07 1.532E+08 7.579E+06 -0.94 0.16 -0.62 0.07 USP41 8.009E+06 8.051E+06 6.796E+07 4.185E+07 -0.99 0.10 -1.46 0.41

91

Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD USP43 3.040E+07 1.897E+07 1.299E+08 1.154E+07 -0.71 0.24 -0.85 0.11 USP44 1.179E+08 2.043E+07 2.924E+08 2.145E+07 0.40 0.26 0.74 0.21 USP45 2.478E+07 1.997E+07 9.649E+07 3.638E+07 -0.78 0.25 -1.18 0.36 USP46 3.736E+06 3.763E+06 7.575E+07 9.268E+06 -1.04 0.05 -1.38 0.09 USP47 5.957E+07 2.690E+07 2.527E+08 3.712E+07 -0.34 0.34 0.35 0.36 USP48 7.675E+07 4.497E+07 2.884E+08 1.642E+07 -0.12 0.57 0.70 0.16 MGC20741 4.191E+07 1.672E+07 1.817E+08 1.991E+07 -0.56 0.21 -0.35 0.19 USP51 7.646E+07 1.901E+07 2.553E+08 1.065E+06 -0.12 0.24 0.38 0.01 USP54 7.785E+07 5.017E+07 2.192E+08 2.761E+06 -0.11 0.64 0.02 0.03 C13ORF22 2.870E+07 2.637E+07 1.262E+08 5.744E+06 -0.73 0.33 -0.89 0.06 CYLD 3.260E+06 1.684E+06 4.546E+07 2.270E+06 -1.05 0.02 -1.68 0.02 BRAP 1.427E+07 6.000E+06 1.664E+08 3.460E+07 -0.91 0.08 -0.49 0.34 HDAC6 2.528E+08 5.739E+07 4.005E+08 4.219E+07 2.11 0.73 1.80 0.41 USP39 1.039E+08 6.256E+07 2.858E+08 1.495E+07 0.22 0.79 0.67 0.15 USP50 7.279E+07 4.530E+07 2.517E+08 2.463E+07 -0.17 0.57 0.34 0.24 USP52 4.236E+08 4.467E+07 4.631E+08 2.091E+07 4.27 0.57 2.41 0.20 BAP1 5.488E+07 2.341E+06 2.692E+08 1.886E+07 -0.40 0.03 0.51 0.18 UCHL1 1.034E+08 2.520E+06 2.601E+08 2.506E+07 0.22 0.03 0.42 0.25 UCHL3 1.543E+08 3.611E+07 3.298E+08 6.021E+06 0.86 0.46 1.10 0.06 UCHL5 5.793E+07 3.047E+06 2.002E+08 5.111E+06 -0.36 0.04 -0.16 0.05 LOC220594 7.308E+07 3.634E+06 2.622E+08 1.207E+07 -0.17 0.05 0.44 0.12 USP49 1.967E+06 8.019E+05 2.403E+07 8.483E+06 -1.07 0.01 -1.89 0.08 MJD 5.398E+07 2.918E+07 2.193E+08 6.577E+07 -0.88 0.52 -0.28 0.49 ATXN3L 7.155E+07 5.073E+07 1.207E+08 6.430E+07 -0.56 0.91 -1.02 0.48 KIAA0063 1.166E+08 1.953E+07 2.583E+08 1.172E+07 0.24 0.35 0.01 0.09 SBBI54 1.044E+08 3.381E+07 3.177E+08 3.974E+07 0.03 0.61 0.45 0.30 AMSH-LP 3.653E+07 1.011E+07 2.496E+07 2.126E+06 -1.19 0.18 -1.73 0.02 COPS5 6.335E+07 1.030E+07 9.813E+07 3.102E+07 -0.71 0.19 -1.18 0.23 PRPF8 2.625E+06 1.057E+06 2.086E+06 1.166E+06 -1.80 0.02 -1.90 0.01 PSMD14 6.655E+07 1.756E+07 4.507E+07 2.117E+07 -0.65 0.32 -1.58 0.16 CXORF53 5.560E+07 7.627E+06 1.139E+08 4.847E+07 -0.85 0.14 -1.07 0.36 FLJ14981 3.322E+07 1.788E+07 9.017E+07 3.775E+07 -1.25 0.32 -1.24 0.28 KIAA1915 5.064E+07 1.966E+07 1.158E+08 1.755E+07 -0.94 0.35 -1.05 0.13 STAMBP 3.423E+07 3.869E+04 9.766E+07 1.618E+07 -1.24 0.00 -1.19 0.12 IFP38 1.285E+08 1.571E+07 1.951E+08 8.308E+07 0.46 0.28 -0.46 0.62 EIF3S3 1.175E+08 7.534E+07 1.414E+08 8.673E+07 0.26 1.35 -0.86 0.65 EIF3S5 4.337E+07 2.071E+07 1.773E+07 4.115E+06 -1.07 0.37 -1.78 0.03 PSMD7 6.818E+07 1.363E+07 2.787E+07 2.179E+07 -0.63 0.24 -1.71 0.16 COPS6 1.223E+08 3.704E+07 3.404E+08 5.137E+07 0.35 0.67 0.62 0.38 YOD1 1.837E+07 7.807E+06 2.902E+07 3.001E+07 -1.52 0.14 -1.70 0.22 PARP11 4.624E+07 8.200E+06 1.865E+08 3.794E+07 -1.02 0.15 -0.53 0.28 KIAA0459 6.762E+07 1.534E+07 2.429E+08 9.971E+07 -0.64 0.28 -0.11 0.74 USP19 1.917E+07 4.573E+06 1.896E+07 1.701E+07 -1.51 0.08 -1.77 0.13 USP24 2.627E+08 5.469E+06 1.878E+08 2.977E+07 2.87 0.10 -0.52 0.22 USP27X 5.024E+07 6.616E+06 5.646E+07 4.113E+07 -0.95 0.12 -1.49 0.31 USP34 1.257E+08 2.155E+07 3.491E+08 6.815E+07 0.41 0.39 0.68 0.51 USP42 9.573E+07 3.497E+07 1.163E+08 3.499E+07 -0.13 0.63 -1.05 0.26 USP53 1.391E+08 7.773E+07 2.218E+08 1.053E+08 0.65 1.40 -0.26 0.78 LOC402168 1.018E+08 1.764E+07 1.253E+08 6.517E+07 -0.02 0.32 -0.98 0.48 SENP1 2.430E+08 6.223E+07 4.818E+08 5.436E+07 2.52 1.12 1.67 0.40 SENP2 4.947E+07 5.254E+06 6.929E+07 5.376E+07 -0.96 0.09 -1.40 0.40 SENP3 6.147E+07 1.793E+07 6.773E+07 5.281E+07 -0.75 0.32 -1.41 0.39 SENP5 8.259E+07 3.253E+07 1.767E+08 5.616E+07 -0.37 0.58 -0.60 0.42

92

Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD SENP6 2.336E+07 4.570E+06 1.789E+07 1.279E+07 -1.43 0.08 -1.78 0.10 SENP7 5.344E+07 1.881E+07 6.925E+07 7.356E+07 -0.89 0.34 -1.40 0.55 SENP8 2.098E+08 7.113E+07 1.968E+08 1.149E+08 1.92 1.28 -0.45 0.85 DUB3 1.192E+08 2.816E+07 1.395E+08 6.980E+07 0.29 0.51 -0.88 0.52 DUB3 1.486E+08 1.739E+07 2.445E+08 7.092E+07 0.82 0.31 -0.09 0.53

Table 5.2 Results from primary DUB screen Raw values and Z scores for the DUBs screened in the primary screen.

93

 

             

 "  "

 

 

   

& &

"'  

!$  '

 !# $

$  "'& 

 ' 

!    

    !#

% "'

   

% "$ taining followed by

  %

 " ! 

   

 %  !$ tarved, and then treated

 # $ " 

"    !#

!   

%  %  th control values at right.

    ' #

  !# "

"$  

  

 "   

 "'&  %

  

#  "

"#  # $

   &

 &   

#! #!

  "#

 ' #  

"  " !

!  !

!  !   !

" $& "  & 

 !   !  !

e s HA-GLUT4 immunofluorescence

! #

 !! "!

ated conditions are plotted wi "!  

&  &

    "

" "#'

# %

following transfection, cells were serum s !! ! 

   %"  

 "   

  %$ #   %$ #

    

!& #

&  #

 ! "%

 &  !!

 %"     

 %$ #   %$ # #

   #'" 

#! !  !

    "  "

   # 

 #

"& d in (A) basal and (B) stimul

"#' $

! #

$! #" by surfac the PM was determined

#"  $

!  ! " $& "

#  %%

$  ' 

    #"

! !&

   #

   

# !'

"%   

   $!

 '     

#   !!

 ! !

'  

RNA pools targeting 102 human DUBs. 72 h

!& $

  "

 #'"    

 #" ""

" !"

!  # 

  !&

' !

$ '

!' #!

!"  !

"&  #&!  

 $ !

"" '

 # !

 %%  !

    $ 

 #&!     

!# !%

 $  &

 #  !#

 ! $

!%   

   

$  " 

   

#  # #  #

!    # " !        !      !       

with or without 100ng/ml IGF1 for 15 min. GLUT4 translocation to GLUT4 translocation IGF1 for 15 min. 100ng/ml or without with automated image analysis. Ranked Z scores of all the DUBs screene Figure 5.2 Ranked Z scores from primary DUB screen DUB screen primary Ranked Z scores from 5.2 Figure HA-GLUT4-HeLa cells were transfected with si

94

          

 

     

                     " !

                   

           "        !   

#'%   "'%  !'% '% '%      '%

     '        $     %   #  &    #    "   !   " 

Figure 5.3 Targets from primary screen selected for deconvolution HA-GLUT4-HeLa cells were transfected with siRNA pools targeting 102 human DUBs. At 72 h following transfection, cells were serum starved, and then treated with or without 100 ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining followed by automated image analysis. A) Candidates that increased basal GLUT4 translocation (Z > 2). B) Candidates that decreased IGF1-stimulated GLUT4 translocation (Z < -1.6). C) HA-GLUT4 surface staining of candidate genes as compared to control TBC1D4 and Akt1/2 knockdown.

95 Secondary screen

A secondary screen was performed on the hits identified in the primary screen to confirm results and exclude positive hits resulting from off-target effects. In this round of screening, each of the targets was validated using the four individual siGenome siRNA oligos that comprised the screening pool. Furthermore, in addition to basal and IGF1-stimulated conditions, saponin permeabilised replicates were included to allow quantification of total GLUT4 levels in the cell. GLUT4 translocation was measured as in the primary screen. Single siRNA oligo knockdowns showing a phenotype of basal GLUT4 translocation greater than, or within 10% of that observed with TBC1D4 knockdown were considered as positive for increasing basal translocation (Fig. 5.4A, 5.4C). Single siRNA oligo knockdowns showing a phenotype of IGF1-stimulated GLUT4 translocation less than, or within 10% of that observed with Akt1+2 knockdown were considered as positive for decreasing stimulated GLUT4 translocation (Fig. 5.4B, 5.4C). Hits were considered validated if two or more single oligos gave the same phenotype as that observed in the primary screen. Total GLUT4 levels were determined using the saponin permeabilised replicates (Fig. 5.5). In most cases, total GLUT4 protein levels were comparable to that of the controls. GLUT4 levels were decreased with a number of gene knockdowns (EIF3S5, PRPF8, SENP6), which may account for the decreased GLUT4 surface staining observed in those samples. In contrast, there were a number of gene knockdowns, which showed decreased IGF1-stimulated surface GLUT4, yet no reduction in total GLUT4 levels (PSMD7, USP19, YOD1). As a result of secondary screening, 2 of the increased basal, and 7 of the decreased IGF1 stimulated hits were considered validated (Table 5.3).

96



 #   ! 

  



#

" 

!

 "

   

   

 #

 "

!

 

"

!

#

 

"

 !

 

"

#

!

"

 

!

  

# "

" !

!

 

 



#

" "

! !

 

 



#

"

" !

  !

 

#



"

!

"

 

!

#  

HDAC6 SENP1 USP10 USP12 USP24 USP52 " 

!

  #

!&% #&% &% #&% &% #&$ &

"          

!   YOD1 CYLD EIF3S5 PRPF8 PSMD7 SENP6 USP19 USP31 USP36 USP7 

&' & $' & #' & !' &  ' & & ' % $ ' % # ' % ! ' %  '                

97     

 



      !          "          

Figure 5.4 Deconvolution of targets from primary screen HA-GLUT4-HeLa cells were transfected with single siRNA oligos targeting candidate genes from primary screen that either A) increased basal GLUT4 translocation or B) inhibited IGF1 stimulated GLUT4 translocation. 72 h following transfection, cells were serum starved, and then treated with or without 100ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining followed by automated image analysis. Single siRNA oligo knockdowns showing a phenotype of basal GLUT4 translocation greater than, or within 10% of TBC1D4 were considered as positive for increasing basal translocation. Single siRNA oligo knockdowns showing a phenotype of IGF1-stimulated GLUT4 translocation less than, or within 10% of Akt1+2 were considered as positive for decreasing stimulated GLUT4 translocation. C) Secondary screening of candidates identified the number of oligos for each gene that either increased basal GLUT4 translocation (in blue), or inhibited IGF1-stimulated GLUT4 translocation (in red). Candidates were considered validated if two or more oligos showed the same phenotype as displayed in the primary screen.

Gene # oligos Increased basal USP10 2 GLUT4 at PM USP52 2

Decreased stimulated EIF3S5 4 GLUT4 at PM PRPF8 4 PSMD7 4 SENP6 4 USP19 2 USP36 2 YOD1 4

Table 5.3 Validated genes as based on deconvolution of siRNA pools

98

otted with total GLUT4 fluorescence from saponin treated repli cates (yellow). lidated hits from secondary screen ated GLUT4 surface staining is pl USP10 USP52 EIF3S5 PRPF8 PSMD7 SENP6 USP19 USP36 YOD1  Figure 5.5 Total GLUT4 levels in va Calculated basal and IGF1 stimul

99 Tertiary screen

To further validate our targets, positive hits from the secondary screen were transfected with different siRNAs, to determine if the observed phenotype was reproducible. For each gene, ON-TARGETplus SMARTpool, siGENOME SMARTpool, and four individual siGenome single siRNA oligos were used to transfect cells. At 72 h later, cells were analyzed to measure GLUT4 translocation as described above. In addition, I also wanted to determine if these manipulations had a specific effect on GLUT4 trafficking versus some global effect on vesicle transport. To test this I also measured transferrin uptake. Finally, I also measured target gene knockdown to validate the technique. GLUT4 translocation was measured under basal and IGF1-stimulated conditions. Of the two genes (USP10, USP52) that previously increased basal GLUT4 surface levels when knocked down, neither validated in this round of screening. The USP52 siGENOME pool siRNA transfection showed increased basal GLUT4 surface staining comparable to that observed with TBC1D4 knockdown, however this phenotype was not repeated in the single oligo, or OTP pool transfections (Fig. 5.6). The genes, EI3FS, PRPF8, PSMD7, SENP6, USP19, USP36 and YOD1 were validated with siGENOME pool siRNA and at least two individual oligos. Furthermore, EIF3S5, PRPF8, PSMD7, SENP6 and USP36 validated with both pool siRNAs (Fig. 5.6). The results are summarized in Table 5.3. To establish whether altered GLUT4 translocation was due to generalized trafficking defects or were GLUT4 specific, transferrin uptake assays were performed. At 72 h following siRNA transfection, cells were serum starved for 2 h and then incubated with Alexa-488 labeled transferrin for 30 min. Cells were fixed, washed and imaged. Automated image analysis was used to determine the amount of transferrin uptake. Interestingly, for all of the targets there was a general trend for reduced transferrin uptake, when compared with untreated or NTC siRNA transfected controls (Fig. 5.7). It should be noted that Akt1+2 knockdown also resulted in reduced transferrin uptake. As previously reported, components of the PI3K/mTOR pathway have been identified as positive regulators of transferrin uptake (Galvez et al. 2007). Therefore transferrin uptake and GLUT4 translocation share upstream signaling components, and as such, changes resulting in altered GLUT4 translocation may be expected to affect transferrin uptake as well. As such, transferrin uptake may not have been a good measure of

100 generic trafficking, and the DUBs that affected transferrin uptake are still potentially interesting. Taqman analysis was performed to determine the degree of knockdown for each target gene, as well as effects on GLUT4 levels relative to untransfected cells by qPCR. Following transfection (72 h) mRNA was extracted from cells and reverse transcribed into cDNA. Quantitative RT-PCR was performed using this cDNA to quantify knockdown of target genes (Fig. 5.8). Of the genes that were analyzed by this method, transfection with siGENOME pool siRNA typically yielded a significant reduction in mRNA levels. The OTP siRNA gave comparable knockdown in most genes, however there were a number of genes in which there was less effective knockdown (USP52, SENP6, YOD1). This may explain why in some cases the phenotype observed with OTP transfected cells is milder. Note: USP19 taqman primers were not available, and I was not able to test USP19 knockdown at this time. As a result of tertiary screening, we have identified seven high confidence regulators (EI3FS, PRPF8, PSMD7, SENP6, USP19, USP36 and YOD1) of GLUT4 translocation (Fig. 5.9). Further characterization of these hits is of primary interest.

101 USP10 USP52 EIF3S5 PRPF8

PSMD7 SENP6 USP19 USP36 YOD1

Figure 5.6 Tertiary screen validation – GLUT4 translocation HA-GLUT4-HeLa cells were transfected with single and pool siRNA oligos targeting genes validated in the secondary screen. 72 h following transfection, cells were serum starved, and then treated with or without 100ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining followed by automated image analysis.

102 USP10 USP52 EIF3S5 PRPF8

$%$

#%$

"%$

!%$

%$

%$

%$

%$

          %                                          %              

PSMD7 SENP6 USP19 USP36 YOD1

Figure 5.7 Tertiary screen validation – transferrin uptake HA-GLUT4-HeLa cells were transfected with single and pool siRNA oligos targeting genes validated in the secondary screen. 72 h following transfection, cells were serum starved, and then incubated with 25 μg/ml Alexa488-conjuugated transferrin for 30 min. Cells were fixed, washed and imaged in the IN CELL Analyzer. Transferrin uptake was determined by automated image analysis.

103

Figure 5.8 Gene expression analysis by qPCR HA-GLUT4-HeLa cells were transfected with single and pool siRNA oligos targeting genes validated in the secondary screen. 72 h following transfection mRNA was extracted from cells and reverse transcribed into cDNA. Quantitative RT-PCR was performed using this cDNA to quantify knockdown of target genes. Gene expression of siRNA-transfected cells is displayed relative to untransfected (Cell only) samples.

siRNA Gene siGENOME OTP single oligos Increased basal USP10 0 0 0 GLUT4 at PM USP52 1 0 0 EIF3S5 1 1 3 PRPF8 1 1 4 Decreased stimulated PSMD7 1 1 4 GLUT4 at PM SENP6 1 1 4 USP19 1 0 2 USP36 1 1 4 YOD1 1 0 4

Table 5.3. Validation of hits in tertiary screen

104 A)

B)

Figure 5.9 High confidence hits from DUB screen High throughput images of siRNA transfected HA-GLUT4-HeLa cells. HA-GLUT4-HeLa cells were transfected with; A) control siRNA and B) siGenome pools of the indicated genes. 72 h following transfection, cells were serum starved for 2 h before treatment with or without 100 ng/ml IGF1 for 15 min. Cells were fixed and stained to detect nuclei (blue), and surface GLUT4 (green). Scale bar 70 μm.

105 Discussion The current screen identified a number of ULPs, which may affect GLUT4 translocation. A siGENOME pool library of ~100 DUBs and SENPs was used to screen HA-GLUT4-HeLa cells to identify genes that regulated GLUT4 translocation. From the primary screen, approximately half of these genes either increased basal GLUT4 translocation or decreased IGF1-stimulated GLUT4 translocation to the level of our controls (TBC1D4, Akt1+2 respectively). From this initial screen, 16 genes were selected for further validation (6 increased basal, 10 decreased IGF1 stimulated GLUT4 translocation). Secondary screening and deconvolution of the 16 genes, resulted in validation of 2 genes that appeared to regulate GLUT4 trafficking under basal conditions, and 7 that when knocked down decreased IGF1-stimulated GLUT4 translocation. Interestingly, a number of the IGF1-stimulated GLUT4 translocation hits did not show reduced total levels of GLUT4, as determined by saponin permeabilisation. This is of interest because although the amount of GLUT4 in the cell was not reduced, IGF1-stimulated translocation was affected. Tertiary screening identified 7 genes, which consistently inhibited IGF1-stimulated GLUT4 translocation when knocked down with different siRNAs. As such, these genes represent high confidence hits from our screen. In summary, our primary and secondary screens identified several potential regulators of GLUT4 translocation. Of these, EI3FS, PRPF8, PSMD7, SENP6, USP19, USP36 and YOD1 are considered high confidence hits. These targets comprise a novel set of genes that represent potential new targets involved in insulin action that require more detailed analysis.

Proteins regulating protein stability and degradation

A number of targets (PSMD7 and YOD1) are involved in proteasomal function and protein degradation. PSMD7 is a subunit of the 26s proteasome (Gallastegui and Groll 2010). Interestingly, mass spectrometry analysis of L6 myotubes, indicates PSMD7 is associated with GLUT4. The association of PSMD7 with GLUT4 was decreased by insulin treatment. It is suggested that insulin may modulate the fate of GLUT4 by increasing its stabilization (Foster et al. 2006). YOD1 is a highly conserved deubiquitinating enzyme of the ovarian tumour (otubain) family, which is required for extraction of proteins from the ER (Ernst et al. 2009). Terminally misfolded proteins

106 that accumulate in the endoplasmic reticulum (ER) are dislocated and targeted for ubiquitin-dependent destruction by the proteasome (Claessen et al. 2010). YOD1 associates with the p97/valosin-containing protein (VCP), which is involved in the ubiquitin proteasomal degradation pathway (Messick et al. 2008; Ernst et al. 2009). In mammalian cells, a dominant-negative YOD1 variant stalls the dislocation of various misfolded, ER-resident proteins (Ernst et al. 2009). Intriguingly, YOD1 knockdown did not reduce the total amount of GLUT4, but severely inhibited IGF1-stimulated GLUT4 translocation. It will be of interest to determine if the defect in GLUT4 translocation is due to mislocalization or accumulation of GLUT4 within the cell, or some other effect on components of insulin signaling or GLUT4 trafficking. A number of intracellular stress pathways have been shown to impact on insulin action including ER stress, mitochondrial stress and oxidative stress. Therefore, it is possible to speculate that a reduction of the level of one of the proteasomal subunits might induce one or more of these intracellular stress pathways leading to impaired IGF1-stimulated GLUT4 trafficking.

Proteins involved in mRNA processing and protein translation

Several of the targets identified are involved in various aspects of mRNA processing, including splicing, translation and degradation. USP52, also known as PAN2 (poly(A) specific ribonuclease subunit homolog) is ubiquitously expressed and functions in mRNA decay (Quesada et al. 2004; Uchida et al. 2004). PRPF8 (Pre-mRNA-processing splicing factor 8) is a component of the spliceosome, and is essential for spliceosomal function (Valadkhan and Jaladat 2010). In vitro, PRPF8 has been shown to bind ubiquitin (Bellare et al. 2006), however based on structural analysis it is unlikely to function as a deubiquitinating enzyme (Komander et al. 2009). In eukaryotic cells, mRNA splicing is an important modification, which is required before protein translocation can occur. The inhibited GLUT4 translocation observed with knockdown of PRPF8 may be a result of defective mRNA splicing of either GLUT4 or components involved in GLUT4 translocation. EIF3S5 (eukaryotic translation initiation factor 3, subunit 5 epsilon)/EIF3F, is a subunit of the eukaryotic translation initiation factor 3 (eIF3) complex. Initiation of translation in eukaryotes depends on many eukaryotic initiation factors (eIFs) that assemble the various RNAs to the 40S ribosome, and scan the mRNA for the start codon. The eIF3 complex is one of the largest of these (Hinnebusch 2006). EIF3F is implicated in the

107 control of mTOR signaling, and has been found to interact with mTOR in yeast two hybrid as well as immunoprecipitation experiments. Insulin stimulates the association of eIF3 and eIF4G in a rapamycin-sensitive manner. The mTOR-induced hyperphosphorylation of 4E-BP family members is also stimulated by insulin, which promotes eIF4E binding to eIF4G, and the assembly of a translation initiation complex at the 5′ cap region of the mRNA. Ultimately, the increased association of eIF3 and eIF4 subunits with the 40s ribosome, results in an increase in the rate at which the small ribosomal subunit is positioned on the mRNA to begin scanning, thus starting protein synthesis (Harris et al. 2006). Furthermore, EIF3F is reported to play an important role in mTOR regulated muscle differentiation and cell growth (Csibi et al. 2010). It has been shown that components involved in mTOR dependent protein translation, can regulate energy and glucose metabolism (Le Bacquer et al. 2007; Rui 2007). Thus it would be important to resolve whether the observed effects on GLUT4 translocation are due to generalized effects, for example due to reduction of protein levels, or indicative of a real metabolic response. The deubiquitinating activity of USP36 has been associated with transcriptional regulation, and ribosome biogenesis. Furthermore, siRNA-mediated knockdown of USP36 resulted in reduced cell proliferation, suggesting that USP36 may function in the regulation of normal cellular proliferation (Endo et al. 2009). Additionally, USP36 has been shown to interact with SOD2 (superoxide dismutase 2, mitochondrial), a mitochondrial protein involved in oxygen radical detoxification. Deubiquitination of SOD2 by USP36 results in its stabilization. Consistent with this, expression of USP36 resulted in increased expression of endogenous SOD2 (Kim et al. 2010). Recent work has shown that mitochondrial oxidative stress is a common feature of multiple models of insulin resistance. Increased mitochondrial superoxide is linked with development of insulin resistance, and this is reversed by treatment with mitochondrial antioxidants (Hoehn et al. 2009). Importantly, overexpression of SOD2 is linked with insulin sensitization (Hoehn et al. 2009) as well as increased contraction-mediated 2-DG uptake (Sandstrom et al. 2006). Determining whether the decreased GLUT4 translocation in USP36 knockdown cells is a result of inhibited cell growth, or a result of its interaction with SOD2 is of interest.

108 Proteins involved in vesicle trafficking

USP10 is implicated in vesicular transport and trafficking of membrane proteins. In yeast, USP10 regulates trafficking between the ER and golgi (Cohen et al. 2003; Cohen et al. 2003). In polarized human airway epithelial cells, USP10 is located in early endosomes. In these cells, USP10 regulates the deubiquitination of cystic fibrosis transmembrane conductance regulator (CFTR), promoting endocytic recycling and trafficking of CFTR in the post endocytic compartment (Bomberger et al. 2009). Furthermore, USP10 has been shown to promote surface expression of ENaC (epithelial Na+ channel) in HEK293 cells. This effect was dependent upon USP10 interaction with SNX3, a member of the sorting nexin family, which is implicated in intracellular protein trafficking (Boulkroun et al. 2008). Another member of the sorting nexin family, SNX9 was shown to play a role in the trafficking of the insulin receptor, thus regulating insulin signaling (MaCaulay et al. 2003). The finding that USP10 interacts with SNX3 raises the possibility that it could interact with other members of this family. Determining if this is the case is a first step towards understanding the mechanism by which USP10 may regulate GLUT4 translocation. USP10 is ubiquitously expressed and is present in both adipose and skeletal muscle, however it has not previously been reported to regulate either insulin signaling or GLUT4 translocation. Further work is required to explore whether USP10 regulates either insulin receptor or GLUT4 vesicle trafficking in insulin sensitive cells.

Miscellaneous functions

Finally, the target genes USP19 and SENP6 regulate unrelated functions. USP19 is widely expressed in tissues, and has been shown to modulate cell proliferation (Lu et al. 2009), as well as the transcription of major myofibrillar proteins. USP19 mRNA is inversely correlated with muscle mass, and reduction of USP19 causes an increase in protein levels of myosin heavy chain (MHC), actin, troponin T and tropomyosin (Combaret et al. 2005; Sundaram et al. 2009). SENP6/SUSP1 is involved in regulating the activity of RXRα (Retinoid X receptor α). RXRα is a ligand activated transcription factor that dimerizes with a variety of hormone and orphan receptors, including peroxisome proliferator-activated receptor-γ (PPARγ). RXRα can be modified by SUMO-1. Prevention of this sumoylation by mutation of the sumo acceptor site led to an increase in the transcriptional activity of RXRα homo- and

109 hetero- dimers. SENP6 co-localizes with RXRα in the nucleus and desumoylates RXRα thus up-regulating its activity. Consistent with this, SENP6 knockdown caused increased levels of sumoylated RXRα and a significant decrease in transcriptional activity of RXRα (Choi et al. 2006). PPARγ was initially characterized for its role as master regulator of adipocyte differentiation and gene expression (Tontonoz and Spiegelman 2008). Thiazolidinediones (TZDs) are a class of drugs that are commonly used to treat type-2 diabetes and which function by activating the PPARγ receptor, resulting in an alteration of the transcription of several genes involved in glucose and lipid metabolism (Hauner 2002). PPARγ activation is linked to insulin sensitivity (Berger et al. 1996; Willson et al. 1996; Mukherjee et al. 1997; Altshuler et al. 2000; Miyazaki et al. 2001). Conversely, loss of PPARγ is associated with the development of insulin resistance (He et al. 2003; Hevener et al. 2003; Matsusue et al. 2003; Norris et al. 2003). SENP6 may indirectly regulate PPARγ activity, by modulating the activity of RXRα. In this way, SENP6 may represent a target linking the SUMO pathway with control of insulin sensitivity. It would be of interest to establish if the effect of SENP6 knockdown on GLUT4 translocation was due to down-regulation of PPARγ signaling.

Conclusion

The family of ubiquitin, and ubiquitin-like proteins are emerging as regulators of a diverse range of cellular processes. The current screen has identified several potential DUB and SENP regulators of insulin signaling and GLUT4 translocation. The validation of these targets in GLUT4 expressing cell lines, such as adipose or muscle is important to determine which of these hits is of greatest biological significance, and relevant for further study. The effect of USP10 knockdown in this screen is exciting in the context of its involvement in the control of vesicle trafficking. Furthermore, the identification of USP36 as a potential regulator of GLUT4 is of interest, as this raises the possibility that ubiquitin system control of mitochondrial oxidation could be linked with glucose metabolism and insulin resistance. Similarly, the involvement of SENP6 with regulation of RXRα activity may implicate SENP6 indirectly with regulation of PPARγ activity and insulin sensitivity. The identification of this novel set of genes expands our understanding of how insulin signaling and GLUT4 translocation may be regulated.

110 Further study of these proteins could identify their mechanism of action and potentially lead to the identification of new drug targets for the treatment of type-2 diabetes.

111

Chapter 6

General Discussion

112 The development of type 2 diabetes is preceded by insulin resistance (Warram et al. 1990). It is characterized by the inability of tissues such as adipose and muscle to respond to the insulin signal and dispose of glucose from the bloodstream. While the body may initially compensate for this, it can eventually lead to hyperglycaemia and a plethora of complications associated with this condition (Ramlo-Halsted and Edelman 1999). Type 2 diabetes is a growing problem, in 2007 it was estimated that approximately 197 million people worldwide had impaired glucose tolerance, a number that is predicted to increase (Hossain et al. 2007). Due to the human and financial costs associated with type 2 diabetes, much work has focused on understanding the underlying mechanisms and the development of therapeutics for the treatment of this disease. Elucidating the molecular components of the insulin-signaling pathway is an important step in understanding insulin resistance. The identification of the components involved in both the positive and negative regulation of this pathway, may lead to the identification of targets that are appropriate for drug development for the treatment of insulin resistance. Understanding how GLUT4 is translocated to the plasma membrane is no trivial task, as there is evidence for different stimuli initiating multiple signaling pathways to cause GLUT4 translocation. To complicate this further, it seems that crosstalk can occur between pathways, and so the function of any single component needs to be considered within the context of a network of signaling pathways (Durmus Tekir et al. 2010).

Development and optimization of the HA-GLUT4-HeLa cell line

The recent discovery of RNAi has enabled large scale functional genetic screening to be performed in order to understand complex biological processes. The main aim of my project was to identify novel targets involved in the regulation of insulin stimulated GLUT4 translocation through the use of siRNA screening. Studies involving insulin signaling and GLUT4 translocation are traditionally done in GLUT4 expressing cells such as adipose and muscle cells. However, after considerable preliminary investigation I determined that it was not feasible to perform high throughput siRNA screening in these cell lines, due to the difficulties associated with large-scale propagation and transfection of these cells. As a result I developed an assay in an alternate cell line (HA- GLUT4-HeLa).

113 HeLa cells do not express GLUT4 endogenously and as such could be considered a somewhat artificial system. They do however express the insulin receptor, and studies carried out to characterize this cell line show that stably expressed GLUT4 translocates to the PM in response to insulin stimulation and displays comparable cellular localization as observed in 3T3-L1 adipocytes. Furthermore, insulin stimulation results in signaling via the PI3K/Akt axis, and disruption of either PI3K or Akt by pharmacological inhibition or siRNA-mediated knockdown resulted in attenuated GLUT4 translocation. Using this cell line, assay conditions were optimized to obtain a dynamic range that was appropriate for high throughput analysis of GLUT4 translocation. Combined with the consideration that this cell line is able to be efficiently transfected and cultured, and also is a cell line of human origin, we determined that it was and appropriate assay to be used for screening for GLUT4 regulators. I used the HA-GLUT4-HeLa cell line to screen both kinase and DUB libraries to identify regulators of insulin stimulated GLUT4 translocation. The siRNA screens identified a number of genes, which when knocked down in HA-GLUT4-HeLa cells either increased basal GLUT4 translocation, or attenuated IGF1 stimulated GLUT4 translocation to levels comparable with the controls (Akt1+2, TBC1D4). When interpreting the data the following factors should be considered. First, off-target effects resulting from siRNA knocking down the expression of unintended genes, can give a false phenotype for the target gene (Jackson and Linsley 2004). Briefly, siRNA transfection can lead to off-target effects via different mechanisms. This can occur through miRNA-like inhibition of translation through partial sequence complementation (Saxena et al. 2003; Zeng et al. 2003), global up or down regulation of genes due to saturation of the RNAi machinery from high concentrations of siRNA (Persengiev et al. 2004; Schwartzenberg-Bar-Yoseph et al. 2004), and interferon response triggered by siRNAs or siRNA delivery reagents (Sledz et al. 2003; Jackson and Linsley 2010). One way of mitigating this effect is through the use of multiple siRNAs as well as deconvolution of siRNA pools, an approach that was used in this study to help to distinguish high confidence targets. The siRNAs used in these studies differed in terms of the oligo strand modification. This modification typically involves methylation of the oligo strands (Jackson et al. 2006; Chen et al. 2008). In some cases, the siGENOME siRNA is modified on the sense strand to promote guide (antisense) strand entry. In contrast, the OTP siRNA oligos are modified on both strands. The sense strand is modified to prevent interaction with RISC and favor antisense uptake, and the antisense

114 seed region is modified to minimize seed-related off targeting. Another approach that can be used to confirm the observed phenotype is to rescue the phenotype by expressing a functional version of the target gene that is resistant to the siRNA (Cullen 2006; Echeverri et al. 2006). This is an approach that I will take when performing further studies on these hits. Second, individual oligos give varying levels of knockdown, therefore further thorough study is required to exclude false positives, and retain true positives. Furthermore, although target mRNA levels may be reduced as a result of siRNA transfection, they do not necessarily correlate to protein levels, as the rate of target protein turnover is also important (Aleman et al. 2007). Western blotting of lysates would provide confirmation of protein knockdown. Finally, as the HA-GLUT4- HeLa cells are not bona fide insulin responsive tissues, it is important to carry out further studies to confirm the significance of these targets in appropriate cell types such as adipocytes or myotubes, which express endogenous GLUT4 and are the main sites of insulin stimulated glucose disposal in the body. In these cell lines, following either siRNA knockdown or over-expression of targets, GLUT4 translocation, glucose uptake and insulin signaling will be investigated to determine if the phenotype is also be observed in these cell lines. Once it is determined in these cell lines that the targets do function in insulin signaling or GLUT4 translocation, then it would be of interest to identify substrates and interacting partners, to further establish its function. Knockout mice could also be developed to further investigate the impact of disruption of targets on whole body insulin sensitivity and metabolism. Knockdown of known components or modulators of insulin signaling, including IR, IRS, Akt and TBC1D4 gave the expected phenotype in this cell line, giving confidence in the assay method. Indeed, a number of the identified hits (CNKSR1, PFKFB3) have previously been reported to play a role in insulin signaling. Expanding the net wider, several other hits (MAST2, TP53RK, PAG, PRKCL2, STK6) were proteins that interacted with known components of the insulin-signaling pathway, but were not previously studied in the context of insulin signaling or GLUT4 translocation. The other hits from the screen function in diverse pathways, and so identifying those of most interest and relevance will form the basis of further study.

Involvement of the ERK/MAPK pathway in insulin signaling

Interestingly, several proteins involved in ERK/MAPK signaling were amongst the hits identified in the screens. Insulin causes acute phosphorylation of ERK1/2, to stimulate

115 its function in promoting cell growth and differentiation (Seger and Krebs 1995). However the contribution of the ERK/MAPK signaling pathway to glucose uptake is somewhat unclear. By phosphorylating IRS proteins on specific serine residues, ERK1/2 is implicated in negative feedback regulation of the IR signaling pathway and it has been suggested that ERK plays a role in the development of insulin resistance (Tanti and Jager 2009; Fritsche et al. 2010). For example, the overexpression of a constitutively active MEK1 mutant, which activates ERK, resulted in reduced levels of IR and IRS12 protein expression as well as attenuated tyrosine phosphorylation of IR, IRS1/2 and PI3K activity (Fujishiro et al. 2003). It is also reported that RNAi knockdown of MAP4K4 prevents insulin resistance by preventing excessive JNK and ERK1/2 activation and IRS1/2 serine phosphorylation. Supporting the idea that ERK is involved in the negative regulation of insulin signaling in response to TNFα (Bouzakri and Zierath 2007). Similarly, long term treatment with another inflammatory cytokine, -1β (IL-1β) has been reported to induce insulin resistance in murine and human adipocytes (Lagathu et al. 2006). Chronic IL-1β treatment markedly reduced insulin stimulated GLUT4 translocation, presumably through the reduced expression of IRS1, resulting in inhibited insulin signaling to Akt and AS160. Treatment with ERK inhibitor partially rescued IRS-1 protein expression and insulin-induced Akt activation, AS160 phosphorylation, and GLUT4 translocation. This suggests a role for ERK in the development insulin resistance in response to chronic IL-1β treatment (Jager et al. 2007). It has also been reported that treatment of L6 myotubes with the peptide hormone Angiotensin II caused inhibited insulin stimulated glucose uptake. The use of PKC or p38 MAPK inhibitors did not improve insulin action, but treatment with ERK1/2 MAPK inhibitor reversed the insulin resistance, suggesting that the inhibition of the insulin stimulated glucose uptake by Angiotensin II treatment occurs via the action of ERK1/2 (Nazari et al. 2007). Recently a link between ERK and oxidative stress induced insulin resistance was reported. Chronic treatment of HL-1 adult cardiomyocytes with hydrogen peroxide lead to insulin resistance and decreased insulin stimulated glucose uptake, accompanied by hyperphosphorylation of ERK. Treatment of these cells with ERK inhibitor enhanced insulin sensitivity and decreased oxidative stress-induced insulin resistance (Tan et al. 2011). Therefore there is much evidence to support the idea that ERK is involved in the development of insulin resistance in response to a number of stimuli. Consistent with this, knockdown of two of the hits

116 identified in this study; HIPK3 and DUSP5 would be expected to increased ERK activity, and may be the reason attenuated IGF1-stimulated GLUT4 was observed. Determining if the knockdown of HIPK3 and DUSP5 is associated with increased ERK phosphorylation or activity will be an important first step in investigating how these genes may be involved in the modulation of GLUT4 translocation. However, it has also been suggested that ERK is important for insulin signaling. For example, it has been reported that ERK may play a role in PPARγ activity. Transfection of cultured cells with a dominant negative MEK, resulted in decreased ability of both insulin and TZDs to stimulate PPARγ activity, suggesting that ERK is involved in the cross-talk between insulin and PPARγ (Zhang et al. 1996). Intriguingly, a study of polycystic ovarian syndrome (PCOS) report interesting findings. Insulin resistance is a recognized feature of PCOS, however comparison of muscle isolated from PCOS and controls showed no difference between the two groups in the expression or activity of IRS or Akt. However, there was severe attenuation of insulin-stimulated ERK activation, and a trend toward increased basal ERK phosphorylation. Therefore, it is proposed that in PCOS, insulin resistance may not be due to a defect in the IRS/Akt pathway, but in the ERK pathway (Rajkhowa et al. 2009) This is an interesting observation as it suggests that in some cases, defective ERK activation may be involved in insulin resistance. The relevance of this finding to non-PCOS related insulin resistance remains to be determined however. The protein B-Raf was identified as a hit in my study, and knockdown would be expected to decrease ERK activity. Perhaps the resultant downstream decrease in ERK activity is related to the inhibited IGF1- stimulated GLUT4 translocation that was observed. However, it has also been reported that B-Raf knockdown results in defective cell growth, and thus it may be that the inhibition of GLUT4 translocation is more a result of aberrant cell growth.

TP53RK

Knockdown of TP53RK in the kinase screen resulted in attenuated IGF1-stimulated GLUT4 translocation. TP53RK is a particularly interesting hit, and is a high priority for further study. This is due to the observation that Akt phosphorylates and activates TP53RK placing it downstream of the major insulin signaling pathway (Facchin et al. 2007). Furthermore, TP53RK has been shown to interact with Rab35 (Abe et al. 2006). Rab35 is thought to play a role in vesicle docking or tethering (Hsu et al. 2010), and has also been reported to be associated with GLUT4 vesicles (Miinea et al. 2005). Together,

117 this data suggests that TP53RK may function to link upstream insulin signaling, with GLUT4 trafficking, and via its interaction with Rab35 may regulate docking of the GLUT4 vesicle at the PM. Future studies are required to investigate the function of TP53RK in adipocytes or myotubes. For example does insulin stimulate activation of TP53RK, or alter TP53RK localization? Furthermore, determining if TP53RK co- localizes with RAB35 and/or GLUT4 vesicles will help to establish the relevance of this protein to insulin stimulated GLUT4 translocation.

USP36

Recently it has been proposed that increased ROS is linked to the development of multiple forms of insulin resistance (Houstis et al. 2006; Hoehn et al. 2009). Therefore the identification of proteins that regulate intracellular ROS levels may be useful in understanding the molecular causes of insulin resistance. It has been shown that treatments that reduce ROS levels can ameliorate insulin resistance to varying levels (Houstis et al. 2006; Hoehn et al. 2009). For example, TNFα- or dexamethasone- induced insulin-resistant 3T3-L1 adipocytes showed a partial reduction in insulin resistance when treated with the antioxidant molecules N-acetylcysteine (NAC) or manganese (iii) tetrakis (4-benzoic acid) porphyrin (MnTBAP) (Houstis et al. 2006). Furthermore, the expression of ROS scavenging enzymes including CuZnSOD, MnSOD, cytocatalase and mitocatalase in 3T3-L1 adipocytes was able to prevent the development of TNFα- and dexamethasone-induced insulin resistance to varying levels (Houstis et al. 2006). Similarly it has also been reported that overexpression of MnSOD in mice confers partial protection against high fat diet-induced insulin resistance, and overexpression of MnSOD in L6 myotubes reversed several models of insulin resistance (Hoehn et al. 2009). USP36 is of interest because it has been shown to interact with SOD2 (superoxide dismutase 2, mitochondrial) aka MnSOD, a mitochondrial protein involved in oxygen radical detoxification. USP36 interaction with SOD2 resulted in its stabilization (Hoehn et al. 2009). Therefore it would be assumed that USP36 knockdown may cause reduced SOD2 levels, which may lead to increased intracellular ROS and development of insulin resistance. It would be of interest to measure if there are changes in ROS levels as a result of USP36 knockdown. If so, this would provide evidence for the role of USP36 in the maintenance of ROS levels and regulation of insulin sensitivity. Furthermore, identifying the ubiquitin ligase that

118 ubiquitinates SOD2 is of interest as inhibitors to this enzyme may aid in stabilization of SOD2 and reduction of ROS levels. .

Conclusion

In conclusion the HA-GLUT4-HeLa cell line is a useful tool to investigate insulin signaling and GLUT4 translocation. Using this cell line, I have performed functional screens to identify kinase and DUB regulators of insulin stimulated GLUT4 translocation and have identified novel proteins that may be involved in this process. These targets will form the basis of further studies to elucidate the insulin-signaling pathway. This will potentially expand our understanding of the players involved in this pathway, and may lead to the identification of targets for drug development for the treatment of type 2 diabetes.

119

Appendix

120 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD Cell only 6.043E+07 3.642E+07 2.681E+08 5.425E+07 -0.30 0.41 0.94 0.55 Lipid Only 7.129E+07 4.240E+07 2.915E+08 5.190E+07 -0.14 0.38 1.15 0.50 NTC 4.420E+07 3.014E+07 1.893E+08 4.528E+07 -0.54 0.28 0.22 0.38 TBC1D4 1.395E+08 4.930E+07 2.430E+08 5.921E+07 0.90 0.60 0.69 0.49 AKT1/2 5.676E+07 4.817E+07 8.020E+07 5.271E+07 -0.37 0.42 -0.75 0.35 AAK1 6.552E+07 3.570E+07 1.759E+08 8.402E+06 0.01 0.62 0.32 0.08 AATK 1.583E+08 5.499E+06 3.376E+08 5.048E+07 1.63 0.10 1.95 0.51 ABL1 9.317E+07 6.566E+07 1.522E+08 5.835E+07 0.49 1.15 0.09 0.59 ABL2 4.314E+07 1.870E+07 7.383E+07 4.607E+07 -0.38 0.33 -0.70 0.46 ACK1 5.545E+07 1.997E+07 8.703E+07 1.432E+07 -0.17 0.35 -0.57 0.14 ACVR1 3.412E+07 1.713E+07 1.536E+08 6.459E+07 -0.54 0.30 0.10 0.65 ACVR1B 3.572E+07 5.816E+06 1.115E+08 4.113E+07 -0.51 0.10 -0.32 0.41 ACVR2 4.398E+07 2.436E+07 2.319E+08 2.143E+07 -0.37 0.43 0.89 0.22 ACVR2B 1.026E+08 1.422E+07 3.190E+08 3.036E+07 0.65 0.25 1.76 0.30 ACVRL1 7.561E+07 1.728E+07 2.642E+08 2.324E+07 0.18 0.30 1.21 0.23 ADAM9 7.966E+07 2.905E+07 2.757E+08 3.198E+07 0.25 0.51 1.33 0.32 ADCK1 3.762E+07 2.174E+07 1.232E+08 7.543E+07 -0.48 0.38 -0.20 0.76 ADCK2 1.919E+08 3.197E+07 4.444E+08 5.457E+07 2.21 0.56 3.02 0.55 ADCK4 8.863E+07 1.883E+07 1.736E+08 1.454E+06 0.41 0.33 0.30 0.01 ADCK5 5.735E+07 3.207E+07 1.980E+08 2.449E+07 -0.14 0.56 0.55 0.25 ADK 3.616E+07 9.507E+06 7.272E+07 1.661E+07 -0.51 0.17 -0.71 0.17 ADRA1A 1.479E+08 2.895E+06 1.966E+08 5.639E+07 1.44 0.05 0.53 0.57 ADRA1B 1.002E+08 2.987E+07 3.074E+08 6.785E+07 0.61 0.52 1.64 0.68 ADRB2 1.198E+08 9.932E+06 1.916E+08 4.288E+06 0.95 0.17 0.48 0.04 ADRBK1 1.335E+07 1.166E+07 5.524E+06 5.592E+06 -0.91 0.20 -1.39 0.06 ADRBK2 5.696E+07 2.321E+07 1.402E+08 4.192E+06 -0.14 0.41 -0.03 0.04 AGTR2 6.175E+07 9.360E+06 1.110E+08 1.271E+07 -0.06 0.16 -0.33 0.13 AK1 3.064E+07 1.828E+07 5.879E+07 1.431E+07 -0.60 0.32 -0.85 0.14 AK2 6.266E+07 6.427E+05 1.693E+08 6.482E+07 -0.04 0.01 0.26 0.65 AK3 6.416E+07 1.533E+07 1.876E+08 2.588E+07 -0.02 0.27 0.44 0.26 AK3L1 3.726E+07 1.607E+07 6.232E+07 3.517E+06 -0.49 0.28 -0.82 0.04 AK5 4.503E+07 6.160E+06 8.563E+07 5.401E+07 -0.35 0.11 -0.58 0.54 AK7 2.246E+07 4.396E+06 7.197E+07 5.140E+07 -0.75 0.08 -0.72 0.52 AKAP1 3.135E+08 3.971E+07 3.827E+08 2.421E+07 4.34 0.69 2.40 0.24 AKAP11 1.675E+08 1.516E+06 2.749E+08 2.635E+07 1.79 0.03 1.32 0.26 AKAP13 6.115E+07 1.167E+07 2.270E+08 1.739E+07 -0.07 0.20 0.84 0.17 AKAP3 2.035E+07 8.475E+06 7.600E+07 2.123E+07 -0.78 0.15 -0.68 0.21 AKAP4 1.124E+08 2.162E+07 2.552E+08 2.232E+07 0.82 0.38 1.12 0.22 AKAP5 2.588E+08 1.664E+07 3.955E+08 1.202E+08 3.38 0.29 2.53 1.21 AKAP6 3.784E+07 1.866E+07 8.409E+07 3.193E+07 -0.48 0.33 -0.60 0.32 AKAP7 1.595E+08 3.637E+07 2.639E+08 6.345E+06 1.65 0.64 1.21 0.06 AKAP8 9.518E+07 3.130E+07 1.691E+08 3.662E+06 0.52 0.55 0.26 0.04 AKT1 3.202E+07 1.651E+07 5.435E+07 1.885E+07 -0.58 0.29 -0.90 0.19 AKT2 2.591E+07 1.980E+07 5.600E+07 3.348E+07 -0.69 0.35 -0.88 0.34 AKT3 5.185E+07 2.704E+07 1.079E+08 1.139E+06 -0.23 0.47 -0.36 0.01 ALK 5.328E+07 1.645E+07 1.768E+08 2.266E+06 -0.21 0.29 0.33 0.02 ACVR1C 3.603E+07 2.766E+07 1.406E+08 2.899E+06 -0.51 0.48 -0.03 0.03 ALS2CR2 7.094E+07 1.574E+07 2.701E+08 7.080E+06 0.10 0.28 1.27 0.07 ALS2CR7 3.298E+07 9.483E+06 9.269E+07 9.389E+06 -0.56 0.17 -0.51 0.09 AMHR2 9.458E+06 5.391E+06 2.003E+07 7.282E+06 -0.97 0.09 -1.24 0.07 ANGPT4 3.439E+07 2.405E+07 3.766E+07 1.333E+07 -0.54 0.42 -1.06 0.13 ANKK1 4.375E+07 1.005E+07 1.375E+08 1.320E+07 -0.37 0.18 -0.06 0.13 ANKRD3 7.573E+07 3.168E+07 1.380E+08 2.249E+07 0.18 0.55 -0.06 0.23 APEG1 5.733E+07 4.741E+06 9.757E+07 2.999E+07 -0.14 0.08 -0.46 0.30 APPL 4.593E+07 1.384E+07 8.023E+07 1.327E+07 -0.34 0.24 -0.64 0.13 ARAF1 1.289E+07 1.091E+07 5.105E+07 1.319E+07 -0.91 0.19 -0.93 0.13 ARK5 3.990E+07 1.321E+07 1.428E+08 4.278E+07 -0.44 0.23 -0.01 0.43 ASK 3.287E+07 3.567E+06 1.121E+08 2.670E+07 -0.56 0.06 -0.32 0.27

121 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD ASP 3.357E+07 1.127E+07 8.012E+07 5.973E+07 -0.55 0.20 -0.64 0.60 ATM 1.989E+07 1.031E+07 7.864E+07 1.636E+07 -0.79 0.18 -0.65 0.16 ATR 2.809E+07 9.118E+06 3.667E+07 2.171E+06 -0.65 0.16 -1.07 0.02 AURKB 2.879E+07 1.732E+07 6.081E+07 1.030E+07 -0.64 0.30 -0.83 0.10 AURKC 5.210E+07 7.573E+06 1.172E+08 1.471E+07 -0.23 0.13 -0.26 0.15 AVPR1A 5.106E+07 1.806E+07 1.149E+08 2.368E+07 -0.25 0.32 -0.29 0.24 AVPR1B 8.052E+06 5.219E+06 9.958E+06 9.353E+06 -1.00 0.09 -1.34 0.09 AXL 4.488E+07 2.016E+07 2.142E+08 3.147E+07 -0.35 0.35 0.71 0.32 AZU1 6.224E+07 8.430E+06 1.578E+08 2.176E+07 -0.05 0.15 0.14 0.22 BCKDK 4.821E+07 2.434E+07 7.988E+07 5.538E+07 -0.30 0.43 -0.64 0.56 BCR 4.263E+06 3.881E+06 1.126E+07 8.627E+06 -1.06 0.07 -1.33 0.09 BDKRB2 6.238E+07 3.984E+07 8.514E+07 1.563E+07 -0.05 0.70 -0.59 0.16 BLK 1.096E+08 1.186E+07 9.703E+07 9.920E+05 0.78 0.21 -0.47 0.01 BLNK 3.379E+07 1.823E+07 2.757E+07 2.562E+07 -0.55 0.32 -1.16 0.26 BMP2K 1.168E+08 4.597E+07 1.897E+08 7.047E+07 0.90 0.80 0.46 0.71 BMPR1A 6.628E+07 3.742E+06 1.241E+08 7.488E+06 0.02 0.07 -0.20 0.08 BMPR1B 8.461E+07 2.687E+07 1.704E+08 3.027E+06 0.34 0.47 0.27 0.03 BMPR2 6.637E+07 1.104E+07 2.756E+08 3.827E+07 0.02 0.19 1.33 0.38 BMX 2.652E+07 1.857E+06 1.274E+08 7.871E+07 -0.68 0.03 -0.16 0.79 BRAF 2.233E+06 1.163E+06 3.842E+06 7.533E+05 -1.10 0.02 -1.40 0.01 BRD2 3.287E+07 7.310E+06 1.123E+08 1.142E+07 -0.56 0.13 -0.31 0.11 BRDT 4.776E+07 6.065E+06 1.149E+08 4.408E+07 -0.30 0.11 -0.29 0.44 BTK 2.536E+07 1.151E+06 5.313E+07 1.848E+07 -0.70 0.02 -0.91 0.19 BUB1 2.125E+08 2.130E+07 2.652E+08 1.311E+08 2.57 0.37 1.22 1.32 BUB1B 3.380E+07 2.825E+05 8.650E+07 9.303E+06 -0.55 0.00 -0.57 0.09 C14ORF20 6.510E+07 3.606E+07 1.545E+08 1.094E+06 0.00 0.63 0.11 0.01 TP53RK 4.436E+06 2.511E+06 1.184E+07 8.874E+06 -1.06 0.04 -1.32 0.09 TRIB3 3.739E+07 3.422E+07 4.498E+07 3.951E+07 -0.53 0.45 -0.83 0.43 C6ORF199 2.353E+07 8.099E+05 6.229E+07 7.959E+07 -0.71 0.01 -0.64 0.87 C7ORF16 5.172E+07 5.237E+07 8.617E+07 8.685E+07 -0.34 0.70 -0.38 0.95 TRIB1 1.449E+08 1.118E+08 1.895E+08 5.364E+07 0.90 1.49 0.75 0.59 C9ORF12 2.353E+08 1.882E+08 2.495E+08 9.346E+07 2.10 2.50 1.41 1.03 CALM3 3.413E+07 1.919E+07 5.573E+07 6.850E+07 -0.57 0.26 -0.72 0.75 CAMK1 1.157E+07 8.014E+06 4.984E+07 6.379E+07 -0.87 0.11 -0.78 0.70 CAMK1D 2.223E+08 1.295E+08 3.172E+08 1.113E+08 1.93 1.72 2.16 1.22 CAMK1G 5.215E+07 3.061E+07 1.293E+08 7.226E+07 -0.33 0.41 0.09 0.79 CAMK2A 1.388E+08 1.483E+08 1.901E+08 9.069E+07 0.82 1.97 0.76 1.00 CAMK2B 1.042E+08 4.724E+07 1.142E+08 8.519E+06 0.36 0.63 -0.07 0.09 CAMK2D 1.116E+08 4.329E+07 2.185E+08 4.603E+07 0.46 0.58 1.07 0.51 CAMK2G 1.002E+08 2.984E+07 1.639E+08 7.904E+07 0.31 0.40 0.47 0.87 CAMK4 9.692E+07 8.138E+07 1.190E+08 3.937E+07 0.26 1.08 -0.02 0.43 CAMKK1 4.967E+07 4.568E+07 1.143E+08 4.356E+06 -0.37 0.61 -0.07 0.05 CAMKK2 1.085E+08 9.344E+07 1.322E+08 8.307E+07 0.42 1.24 0.12 0.91 CARD10 1.082E+06 8.178E+05 1.299E+07 1.835E+07 -1.01 0.01 -1.19 0.20 CARD14 2.260E+07 1.902E+07 4.183E+07 3.741E+07 -0.73 0.25 -0.87 0.41 TNNI3K 6.798E+07 5.524E+07 1.266E+08 9.654E+07 -0.12 0.73 0.06 1.06 CARKL 2.165E+08 9.713E+06 3.228E+08 2.445E+07 1.85 0.13 2.22 0.27 CASK 9.172E+07 4.041E+07 1.964E+08 1.338E+08 0.19 0.54 0.83 1.47 CCL2 1.868E+07 6.139E+06 4.466E+07 5.949E+07 -0.78 0.08 -0.84 0.65 CCL4 6.859E+07 4.685E+07 1.007E+08 3.920E+07 -0.11 0.62 -0.22 0.43 CCRK 9.183E+07 7.328E+07 1.107E+08 1.094E+08 0.19 0.97 -0.11 1.20 CD3E 1.083E+08 9.614E+07 1.020E+08 9.246E+07 0.41 1.28 -0.21 1.02 CD4 4.352E+07 1.117E+06 8.087E+07 1.048E+08 -0.45 0.01 -0.44 1.15 CD7 1.305E+08 1.188E+08 1.533E+08 1.236E+08 0.71 1.58 0.36 1.36 CDADC1 1.563E+08 8.160E+07 2.066E+08 2.119E+08 1.05 1.08 0.94 2.33 CDC2 4.938E+06 4.566E+06 4.244E+07 5.990E+07 -0.96 0.06 -0.86 0.66 CDC2L1 1.389E+08 4.423E+07 1.643E+08 1.172E+08 0.82 0.59 0.48 1.29 CDC2L2 1.795E+07 9.510E+06 5.616E+07 7.173E+07 -0.79 0.13 -0.71 0.79

122 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD CDC2L5 7.757E+07 8.008E+07 7.870E+07 3.096E+07 0.00 1.06 -0.46 0.34 CDC42BPA 1.186E+08 1.021E+08 1.429E+08 9.628E+07 0.55 1.36 0.24 1.06 CDC42BPB 1.588E+08 1.319E+08 1.376E+08 1.093E+08 1.09 1.75 0.18 1.20 CDC7 1.223E+08 1.027E+08 1.432E+08 8.105E+07 0.60 1.37 0.24 0.89 CDK10 6.479E+07 4.258E+07 6.324E+07 7.628E+07 -0.16 0.57 -0.63 0.84 CDK11 3.273E+07 1.893E+07 7.069E+07 6.989E+07 -0.59 0.25 -0.55 0.77 CDK2 1.916E+08 1.109E+08 2.275E+08 4.347E+07 1.52 1.47 1.17 0.48 CDK3 1.982E+07 4.212E+06 4.081E+07 5.136E+07 -0.76 0.06 -0.88 0.56 CDK4 1.429E+08 1.515E+08 2.156E+08 1.043E+08 0.87 2.01 1.04 1.15 CDK5 3.985E+07 4.716E+07 7.250E+07 8.132E+07 -0.50 0.63 -0.53 0.89 CDK5R1 2.888E+07 2.369E+07 4.530E+07 5.718E+07 -0.64 0.31 -0.83 0.63 CDK5R2 4.036E+07 1.772E+07 1.147E+08 1.594E+08 -0.49 0.24 -0.07 1.75 CDK5RAP1 4.721E+07 3.888E+07 6.924E+07 9.357E+07 -0.40 0.52 -0.57 1.03 CDK5RAP3 4.697E+07 4.281E+07 8.025E+07 9.035E+07 -0.40 0.57 -0.45 0.99 CDK6 1.578E+08 1.039E+08 2.140E+08 5.024E+07 1.07 1.38 1.02 0.55 CDK7 8.236E+07 8.980E+07 9.845E+07 6.362E+07 0.07 1.19 -0.25 0.70 CDK8 1.148E+08 9.179E+07 1.993E+08 1.938E+07 0.50 1.22 0.86 0.21 CDK9 1.320E+07 2.385E+06 5.722E+07 7.798E+07 -0.85 0.03 -0.70 0.86 CDKL1 8.849E+07 5.954E+07 1.039E+08 1.122E+08 0.15 0.79 -0.19 1.23 CDKL2 2.546E+07 1.763E+07 7.380E+07 1.031E+08 -0.69 0.23 -0.52 1.13 CDKL3 2.051E+08 2.125E+08 2.293E+08 1.345E+08 1.70 2.83 1.19 1.48 CDKL5 6.350E+07 6.009E+07 9.619E+07 9.638E+07 -0.18 0.80 -0.27 1.06 CDKN1A 5.861E+07 5.497E+07 1.239E+08 9.962E+07 -0.25 0.73 0.03 1.10 CDKN1B 5.930E+07 4.795E+07 7.605E+07 8.284E+07 -0.24 0.64 -0.49 0.91 CDKN1C 1.514E+07 7.188E+06 3.194E+07 3.739E+07 -0.83 0.10 -0.98 0.41 CDKN2B 3.504E+07 4.060E+07 6.226E+07 6.159E+07 -0.56 0.54 -0.64 0.68 CDKN2C 4.097E+07 2.418E+07 7.466E+07 7.259E+07 -0.48 0.32 -0.51 0.80 CDKN2D 8.094E+07 7.247E+07 1.161E+08 7.943E+07 0.05 0.96 -0.05 0.87 CDKN3 7.237E+07 7.816E+07 1.488E+08 2.574E+07 -0.06 1.04 0.31 0.28 CERK 9.363E+07 7.220E+07 1.612E+08 1.492E+08 0.22 0.96 0.44 1.64 RAPGEF4 2.982E+07 1.159E+07 6.798E+07 7.678E+07 -0.63 0.15 -0.58 0.84 CHEK1 6.102E+07 2.383E+07 1.297E+08 1.540E+08 -0.22 0.32 0.10 1.69 CHEK2 4.982E+07 2.241E+07 1.051E+08 6.386E+07 -0.36 0.30 -0.17 0.70 CHKA 4.768E+07 5.658E+07 6.383E+07 8.490E+07 -0.39 0.75 -0.63 0.93 CHKB 1.052E+08 5.258E+07 2.480E+08 8.543E+07 0.37 0.70 1.40 0.94 CHRM1 1.909E+08 1.327E+08 1.971E+08 7.051E+07 1.51 1.76 0.84 0.78 CHUK 2.460E+07 1.458E+07 7.692E+07 7.091E+07 -0.70 0.19 -0.48 0.78 CINP 1.584E+07 1.612E+07 3.775E+07 5.092E+07 -0.82 0.21 -0.91 0.56 CIT 1.847E+07 1.997E+07 5.268E+07 7.005E+07 -0.78 0.27 -0.75 0.77 CKB 3.696E+07 2.348E+07 1.517E+08 4.368E+07 -0.53 0.31 0.34 0.48 CKM 2.689E+07 1.103E+07 9.519E+07 9.230E+07 -0.67 0.15 -0.28 1.01 CKMT1 6.432E+07 4.430E+07 1.351E+08 1.377E+08 -0.17 0.59 0.16 1.51 CKMT2 4.255E+07 3.403E+07 1.428E+08 1.125E+08 -0.46 0.45 0.24 1.24 CKS1B 4.137E+07 4.004E+07 9.437E+07 1.069E+08 -0.48 0.53 -0.29 1.18 CKS2 5.863E+07 1.187E+07 1.308E+08 5.285E+07 -0.25 0.16 0.11 0.58 CLK1 1.363E+08 1.051E+08 1.924E+08 1.074E+08 0.79 1.40 0.79 1.18 CLK2 1.330E+07 9.711E+06 8.438E+07 1.023E+08 -0.85 0.13 -0.40 1.12 CLK3 4.526E+07 3.092E+07 1.280E+08 1.190E+08 -0.42 0.41 0.08 1.31 CLK4 1.248E+08 5.581E+06 1.719E+08 1.261E+08 0.63 0.07 0.56 1.39 PLK3 6.010E+07 4.601E+07 1.890E+08 4.545E+07 -0.20 0.68 0.37 0.50 CNKSR1 2.601E+06 2.239E+06 2.111E+07 1.308E+07 -1.05 0.03 -1.46 0.14 COL4A3BP 8.868E+07 8.509E+07 2.010E+08 1.108E+07 0.22 1.25 0.50 0.12 COPB2 2.312E+06 3.799E+05 1.110E+07 1.380E+07 -1.05 0.01 -1.57 0.15 CRK7 1.299E+08 1.135E+08 2.476E+08 7.792E+07 0.83 1.67 1.01 0.85 CRKL 2.061E+07 2.196E+07 4.318E+07 2.577E+07 -0.78 0.32 -1.22 0.28 CSF1R 5.532E+07 4.918E+07 1.435E+08 6.896E+06 -0.27 0.72 -0.12 0.08 CSK 4.265E+07 4.634E+07 1.133E+08 1.851E+07 -0.46 0.68 -0.45 0.20 CSNK1A1 1.055E+08 4.982E+07 2.764E+08 1.073E+07 0.47 0.73 1.32 0.12

123 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD CSNK1D 1.056E+08 9.444E+07 2.777E+08 1.055E+07 0.47 1.39 1.34 0.12 CSNK1E 8.687E+06 8.326E+06 3.807E+07 1.702E+07 -0.96 0.12 -1.27 0.19 CSNK1G1 5.739E+07 6.001E+07 1.082E+08 3.832E+07 -0.24 0.88 -0.51 0.42 CSNK1G2 9.614E+07 8.102E+07 2.960E+08 3.289E+07 0.33 1.19 1.54 0.36 CSNK1G3 4.754E+07 5.186E+07 9.100E+07 1.737E+06 -0.39 0.76 -0.70 0.02 CSNK2A1 8.968E+07 6.129E+07 2.192E+08 2.672E+07 0.23 0.90 0.70 0.29 CSNK2A2 2.022E+07 1.426E+07 8.579E+07 2.450E+06 -0.79 0.21 -0.75 0.03 CSNK2B 1.403E+08 5.241E+07 2.077E+08 6.897E+06 0.98 0.77 0.58 0.08 CXCL10 2.863E+07 3.017E+07 1.072E+08 2.639E+05 -0.67 0.44 -0.52 0.00 DAPK1 7.623E+07 4.690E+07 1.881E+08 2.796E+07 0.03 0.69 0.36 0.30 DAPK2 5.792E+05 7.237E+05 1.020E+06 5.872E+05 -1.08 0.01 -1.68 0.01 DAPK3 9.213E+06 8.562E+06 3.583E+07 2.775E+07 -0.95 0.13 -1.30 0.30 DCAMKL1 5.075E+07 5.496E+07 1.214E+08 2.163E+06 -0.34 0.81 -0.36 0.02 DCK 7.437E+07 6.815E+07 8.493E+07 2.840E+07 0.01 1.00 -0.76 0.31 DDR1 1.678E+08 9.290E+07 3.724E+08 8.111E+07 1.38 1.37 2.37 0.88 DDR2 1.074E+08 7.487E+07 2.000E+08 3.235E+07 0.49 1.10 0.49 0.35 DGKA 7.745E+07 3.225E+07 2.364E+08 4.715E+06 0.05 0.48 0.89 0.05 DGKB 1.170E+08 6.967E+07 2.952E+08 6.101E+06 0.64 1.03 1.53 0.07 DGKD 3.608E+07 2.676E+07 1.192E+08 1.144E+07 -0.56 0.39 -0.39 0.12 DGKE 6.733E+06 4.297E+06 1.835E+07 6.390E+06 -0.99 0.06 -1.49 0.07 DGKG 5.451E+07 3.689E+07 2.098E+08 4.335E+06 -0.29 0.54 0.60 0.05 DGKI 1.151E+08 1.081E+08 3.158E+08 2.017E+07 0.61 1.59 1.75 0.22 DGKQ 6.975E+07 7.235E+07 1.061E+08 6.615E+07 -0.06 1.07 -0.53 0.72 DGKZ 8.391E+07 5.156E+07 1.655E+08 4.602E+06 0.15 0.76 0.12 0.05 DGUOK 1.039E+08 8.476E+07 1.163E+08 5.528E+07 0.44 1.25 -0.42 0.60 DKFZP434C131 1.606E+08 1.159E+08 2.484E+08 4.112E+07 1.28 1.71 1.02 0.45 DKFZp434C1418 1.028E+08 3.888E+07 1.466E+08 4.738E+07 0.43 0.57 -0.09 0.52 DKFZP586B1621 1.392E+08 1.290E+08 2.447E+08 5.679E+07 0.96 1.90 0.98 0.62 DKFZP761P0423 9.084E+07 4.237E+07 1.581E+08 1.301E+07 0.25 0.62 0.03 0.14 STYK1 1.110E+08 3.582E+07 2.513E+08 5.642E+07 0.55 0.53 1.05 0.62 DLG1 5.420E+07 5.303E+07 1.658E+08 2.683E+07 -0.29 0.78 0.12 0.29 DLG2 1.031E+08 9.408E+07 2.748E+08 1.546E+07 0.43 1.39 1.31 0.17 DLG3 6.377E+07 5.917E+07 1.559E+08 2.531E+06 -0.15 0.87 0.01 0.03 DLG4 1.311E+08 5.692E+07 2.027E+08 3.903E+07 0.84 0.84 0.52 0.43 DMPK 1.568E+08 4.061E+07 2.368E+08 2.605E+07 1.22 0.60 0.89 0.28 DNAJC3 6.119E+06 3.169E+06 4.764E+07 4.000E+07 -1.00 0.05 -1.17 0.44 DOK1 1.668E+08 8.400E+07 2.479E+08 4.186E+07 1.37 1.24 1.01 0.46 DTYMK 1.062E+07 7.071E+06 5.378E+07 1.573E+07 -0.93 0.10 -1.10 0.17 DUSP1 2.762E+08 1.961E+08 3.836E+08 9.415E+06 2.98 2.89 2.49 0.10 DUSP10 8.683E+07 4.694E+07 1.544E+08 6.090E+07 0.19 0.69 -0.01 0.66 DUSP2 8.851E+07 7.680E+07 2.203E+08 1.745E+07 0.22 1.13 0.71 0.19 DUSP22 7.352E+07 4.117E+07 1.596E+08 3.565E+07 0.00 0.61 0.05 0.39 DUSP4 9.142E+07 6.730E+07 2.134E+08 2.736E+07 0.26 0.99 0.64 0.30 DUSP5 1.345E+07 6.819E+06 3.510E+07 3.058E+07 -0.89 0.10 -1.31 0.33 DUSP6 7.438E+07 7.440E+07 1.206E+08 3.936E+07 0.01 1.10 -0.37 0.43 DUSP7 3.517E+07 2.145E+07 1.271E+08 2.509E+07 -0.57 0.32 -0.30 0.27 DUSP8 9.287E+07 6.037E+07 1.411E+08 2.391E+07 0.28 0.89 -0.15 0.26 DYRK1A 9.811E+07 9.403E+07 1.543E+08 2.164E+07 0.36 1.39 -0.01 0.24 DYRK1B 8.969E+07 7.800E+07 1.738E+08 7.198E+06 0.23 1.15 0.21 0.08 DYRK2 1.326E+08 1.007E+08 2.303E+08 1.631E+07 0.87 1.48 0.82 0.18 DYRK3 9.422E+06 9.642E+06 2.649E+07 2.354E+07 -0.95 0.14 -1.40 0.26 DYRK4 5.851E+07 8.085E+06 2.065E+08 3.662E+06 -0.23 0.12 0.56 0.04 EDN2 4.014E+07 2.851E+07 7.322E+07 4.216E+07 -0.50 0.42 -0.89 0.46 EEF2K 1.259E+08 9.807E+07 2.400E+08 4.904E+07 0.77 1.45 0.93 0.53 EGFR 6.337E+07 5.348E+07 1.761E+08 7.080E+07 -0.15 0.79 0.23 0.77 EIF2AK3 1.088E+08 6.327E+07 1.630E+08 6.331E+06 0.51 0.93 0.09 0.07 EIF2AK4 2.113E+07 1.618E+07 5.449E+07 5.376E+07 -0.78 0.24 -1.10 0.59 EKI1 1.352E+07 1.102E+07 5.017E+07 4.660E+07 -0.89 0.16 -1.14 0.51

124 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD RAPGEF3 1.825E+07 3.543E+06 5.679E+07 1.004E+07 -0.82 0.05 -1.07 0.11 EPHA1 5.983E+07 4.947E+07 7.886E+07 3.141E+07 -0.21 0.73 -0.83 0.34 EPHA2 7.864E+07 5.927E+07 1.944E+08 1.711E+06 0.07 0.87 0.43 0.02 EPHA3 7.354E+06 9.038E+05 3.559E+07 2.022E+07 -0.98 0.01 -1.30 0.22 EPHA4 3.769E+06 5.682E+05 4.885E+07 1.498E+07 -1.03 0.01 -1.16 0.16 EPHA5 2.629E+07 5.646E+06 9.408E+07 1.602E+06 -0.70 0.08 -0.66 0.02 EPHA7 1.763E+08 7.003E+07 3.048E+08 3.951E+07 1.51 1.03 1.63 0.43 EPHA8 5.901E+07 3.922E+06 1.660E+08 1.927E+07 -0.22 0.06 0.12 0.21 EPHB1 5.545E+07 1.557E+07 1.130E+08 6.432E+07 -0.27 0.23 -0.46 0.70 EPHB2 1.500E+08 7.653E+07 1.405E+08 4.827E+07 1.12 1.13 -0.16 0.53 EPHB3 5.703E+07 2.699E+07 1.369E+08 1.382E+07 -0.25 0.40 -0.20 0.15 EPHB4 7.148E+06 5.093E+06 5.114E+07 1.209E+07 -0.98 0.08 -1.13 0.13 EPHB6 6.608E+07 4.941E+07 1.705E+08 1.003E+08 -0.11 0.73 0.17 1.09 ERBB2 2.111E+08 6.074E+07 3.652E+08 3.430E+07 1.13 0.67 1.44 0.28 ERBB3 1.754E+08 3.931E+07 2.538E+08 6.216E+07 0.74 0.43 0.53 0.51 ERBB4 1.162E+08 1.324E+07 2.075E+08 9.270E+07 0.09 0.15 0.15 0.76 ERK8 2.353E+07 1.558E+06 8.767E+07 8.312E+07 -0.94 0.02 -0.84 0.68 ERN1 2.434E+07 1.762E+07 1.198E+08 8.393E+07 -0.93 0.19 -0.57 0.69 EVI1 1.027E+08 2.798E+07 1.951E+08 2.274E+07 -0.06 0.31 0.04 0.19 FASTK 4.175E+06 1.689E+06 2.938E+07 2.753E+07 -1.15 0.02 -1.32 0.23 FER 1.034E+08 1.711E+07 2.299E+08 1.871E+07 -0.06 0.19 0.33 0.15 FES 1.901E+08 6.688E+07 3.075E+08 5.583E+07 0.90 0.74 0.96 0.46 FGFR1 2.000E+07 2.112E+07 3.774E+07 1.592E+07 -0.98 0.23 -1.25 0.13 FGFR2 3.682E+07 1.322E+06 9.433E+07 8.757E+07 -0.79 0.01 -0.78 0.72 FGFR3 7.718E+07 4.130E+07 1.342E+08 1.358E+08 -0.34 0.46 -0.46 1.11 FGFR4 1.095E+08 2.962E+07 1.566E+08 1.695E+08 0.01 0.33 -0.27 1.39 FGR 1.632E+07 7.235E+06 1.103E+08 1.375E+08 -1.02 0.08 -0.65 1.13 FLJ10074 2.226E+08 2.131E+07 3.991E+08 8.928E+07 1.26 0.23 1.72 0.73 FLJ10761 3.883E+07 4.594E+06 1.028E+08 8.256E+07 -0.77 0.05 -0.71 0.68 FLJ10842 1.792E+08 1.883E+07 2.847E+08 1.968E+07 0.78 0.21 0.78 0.16 RFK 1.831E+08 2.619E+07 3.605E+08 6.810E+07 0.82 0.29 1.40 0.56 FLJ12476 1.641E+08 5.120E+07 2.954E+08 7.330E+07 0.61 0.56 0.87 0.60 FLJ13052 1.378E+08 6.177E+07 2.544E+08 3.873E+07 0.32 0.68 0.53 0.32 FLJ20574 8.512E+07 7.076E+05 1.007E+08 9.235E+07 -0.26 0.01 -0.73 0.76 THNSL1 2.344E+08 1.475E+07 3.775E+08 3.043E+07 1.39 0.16 1.54 0.25 FLJ23074 1.788E+08 5.003E+07 2.928E+08 5.010E+07 0.77 0.55 0.84 0.41 LRRK1 4.104E+08 9.612E+06 4.569E+08 5.572E+07 3.33 0.11 2.19 0.46 FLJ23356 9.739E+07 7.016E+06 1.365E+08 1.090E+08 -0.12 0.08 -0.44 0.89 FLJ25006 1.991E+08 4.610E+07 2.631E+08 7.826E+07 1.00 0.51 0.60 0.64 FLJ32685 2.187E+08 6.116E+07 2.410E+08 6.389E+07 1.22 0.67 0.42 0.52 C9ORF98 1.851E+08 1.609E+07 2.919E+08 9.308E+07 0.85 0.18 0.84 0.76 FLJ34389 1.586E+08 3.321E+07 2.577E+08 1.152E+08 0.55 0.37 0.56 0.94 FLJ35107 3.066E+07 5.684E+06 1.234E+08 1.115E+08 -0.86 0.06 -0.54 0.91 FLT1 1.772E+08 7.631E+06 2.546E+08 1.200E+08 0.76 0.08 0.53 0.98 FLT3 1.633E+08 4.138E+06 2.190E+08 5.899E+07 0.60 0.05 0.24 0.48 FLT4 3.258E+08 1.567E+07 3.370E+08 3.110E+07 2.40 0.17 1.21 0.25 FN3K 8.563E+07 8.038E+06 1.492E+08 1.363E+08 -0.25 0.09 -0.33 1.12 FN3KRP 1.685E+07 1.411E+07 8.657E+07 1.047E+08 -1.01 0.16 -0.85 0.86 FRAP1 1.976E+08 5.699E+07 2.241E+08 4.599E+07 0.98 0.63 0.28 0.38 FRDA 2.695E+08 1.297E+07 3.657E+08 1.883E+07 1.78 0.14 1.44 0.15 FRK 6.139E+07 1.142E+07 2.205E+08 9.072E+07 -0.52 0.13 0.25 0.74 FUK 4.594E+07 1.722E+07 2.155E+08 1.693E+08 -0.69 0.19 0.21 1.39 FYB 2.322E+07 4.664E+05 8.899E+07 5.314E+07 -0.94 0.01 -0.83 0.44 FYN 7.105E+06 2.723E+06 4.421E+07 4.519E+07 -1.12 0.03 -1.19 0.37 GAK 1.573E+08 1.312E+06 2.687E+08 1.525E+08 0.54 0.01 0.65 1.25 GALK1 2.423E+08 1.153E+07 2.671E+08 5.842E+07 1.47 0.13 0.63 0.48 GALK2 6.383E+07 4.309E+05 1.662E+08 1.731E+08 -0.49 0.00 -0.19 1.42 GAP43 2.557E+08 1.992E+07 2.796E+08 4.418E+07 1.62 0.22 0.74 0.36

125 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD GCK 7.174E+06 3.008E+06 6.227E+07 7.036E+07 -1.12 0.03 -1.05 0.58 GFRA2 1.235E+08 4.353E+07 1.839E+08 8.965E+07 0.17 0.48 -0.05 0.74 GK 5.522E+07 2.226E+07 1.402E+08 7.109E+07 -0.59 0.25 -0.41 0.58 GK2 3.640E+07 1.861E+07 1.538E+08 8.895E+07 -0.79 0.21 -0.29 0.73 GMFB 2.126E+08 1.712E+07 3.601E+08 4.245E+07 1.15 0.19 1.40 0.35 GMFG 2.132E+07 1.188E+07 6.782E+07 7.152E+07 -0.96 0.13 -1.00 0.59 GNE 1.404E+08 5.722E+07 2.556E+08 5.859E+07 0.35 0.63 0.54 0.48 GRK4 9.362E+07 1.492E+07 1.637E+08 1.160E+08 -0.16 0.16 -0.21 0.95 GRK5 1.610E+08 4.240E+07 2.791E+08 5.407E+07 0.58 0.47 0.73 0.44 GRK6 6.047E+06 1.763E+05 3.503E+07 4.257E+07 -1.13 0.00 -1.27 0.35 GRK7 7.097E+07 1.223E+05 1.320E+08 7.588E+07 -0.41 0.00 -0.47 0.62 GSG2 1.549E+08 6.416E+07 2.760E+08 4.780E+07 0.51 0.71 0.71 0.39 GSK3A 7.624E+07 3.234E+07 1.876E+08 5.103E+07 -0.36 0.36 -0.02 0.42 GSK3B 1.364E+08 4.358E+07 3.023E+08 6.309E+07 0.31 0.48 0.92 0.52 GTF2H1 6.160E+07 2.331E+07 1.918E+08 3.591E+07 -0.52 0.26 0.02 0.29 GUCY2C 1.397E+08 5.682E+07 2.340E+08 1.255E+08 0.34 0.63 0.36 1.03 GUCY2D 3.461E+06 1.086E+06 1.126E+07 1.331E+07 -1.16 0.01 -1.46 0.11 GUCY2F 3.049E+07 1.357E+07 1.464E+08 8.739E+07 -0.86 0.15 -0.36 0.72 GUK1 2.220E+07 4.343E+06 1.070E+08 1.048E+08 -0.95 0.05 -0.68 0.86 HSPB8 2.459E+07 3.183E+04 7.728E+07 8.312E+07 -0.92 0.00 -0.92 0.68 HAK 2.579E+07 1.402E+07 9.998E+07 1.027E+08 -0.91 0.15 -0.74 0.84 HCK 2.940E+07 2.192E+07 4.827E+07 3.162E+07 -0.87 0.24 -1.16 0.26 HIPK1 1.963E+08 4.848E+07 3.392E+08 2.300E+07 0.97 0.53 1.23 0.19 HIPK2 4.950E+07 5.333E+06 9.146E+07 4.323E+07 -0.65 0.06 -0.81 0.35 HIPK3 2.177E+06 8.892E+05 2.546E+07 2.534E+07 -1.17 0.01 -1.35 0.21 HIPK4 1.099E+03 1.256E+03 3.639E+05 5.132E+05 -1.20 0.00 -1.55 0.00 HK1 1.831E+08 7.172E+07 3.295E+08 7.894E+06 0.82 0.79 1.15 0.06 HK2 1.783E+07 4.860E+06 9.180E+07 1.037E+08 -1.00 0.05 -0.80 0.85 HK3 1.062E+08 7.845E+06 1.988E+08 7.530E+07 -0.02 0.09 0.07 0.62 HRI 9.986E+07 1.248E+07 2.348E+08 1.251E+08 -0.09 0.14 0.37 1.03 STK32B 8.483E+07 1.263E+07 1.866E+08 6.854E+07 -0.26 0.14 -0.03 0.56 HSMDPKIN 9.220E+06 2.678E+06 6.565E+07 5.963E+07 -1.09 0.03 -1.02 0.49 HUNK 1.609E+07 6.198E+06 6.224E+07 5.599E+07 -1.02 0.07 -1.05 0.46 ITGB1BP1 2.453E+08 1.108E+08 2.582E+08 8.546E+07 1.51 1.22 0.56 0.70 MASTL 6.427E+06 1.579E+06 2.804E+07 1.568E+07 -1.12 0.02 -1.33 0.13 ICK 3.507E+07 5.097E+06 7.317E+07 7.921E+07 -0.69 0.09 -0.62 0.72 IGF1R 5.153E+07 3.327E+07 6.036E+07 5.313E+07 -0.40 0.57 -0.73 0.48 IHPK1 4.817E+07 5.909E+06 1.237E+08 3.597E+07 -0.46 0.10 -0.16 0.33 IHPK2 9.954E+07 4.429E+07 3.316E+08 3.803E+07 0.42 0.76 1.74 0.35 IHPK3 4.732E+07 2.943E+07 1.031E+08 1.203E+08 -0.48 0.51 -0.34 1.10 IKBKAP 5.534E+07 3.687E+06 2.255E+08 9.842E+07 -0.34 0.06 0.77 0.90 IKBKB 1.149E+08 1.398E+07 1.781E+08 1.099E+08 0.68 0.24 0.34 1.00 IKBKE 9.849E+07 1.232E+07 2.755E+08 3.292E+06 0.40 0.21 1.23 0.03 IL2 8.269E+07 3.417E+07 2.102E+08 2.004E+07 0.13 0.59 0.63 0.18 ILK 6.295E+07 3.245E+07 4.495E+07 1.083E+07 -0.21 0.56 -0.87 0.10 ILKAP 3.051E+07 1.346E+06 7.725E+07 8.372E+07 -0.76 0.02 -0.58 0.76 IMPK 1.175E+08 4.726E+07 1.584E+08 9.150E+07 0.73 0.81 0.16 0.83 INSR 3.631E+07 2.719E+07 5.928E+07 6.752E+07 -0.66 0.47 -0.74 0.62 INSRR 6.596E+07 4.423E+07 1.394E+08 1.346E+08 -0.16 0.76 -0.01 1.23 IRAK1 6.256E+07 4.034E+07 1.003E+08 1.249E+08 -0.21 0.69 -0.37 1.14 IRAK2 1.237E+08 5.594E+07 1.527E+08 1.817E+08 0.84 0.96 0.11 1.66 IRAK3 4.564E+07 1.690E+07 7.172E+07 9.521E+07 -0.50 0.29 -0.63 0.87 IRS1 3.325E+07 1.992E+07 5.446E+07 7.168E+07 -0.72 0.34 -0.79 0.65 ITK 4.015E+07 5.789E+06 1.638E+08 4.385E+07 -0.60 0.10 0.21 0.40 ITPK1 1.028E+08 4.066E+06 1.939E+08 1.100E+08 0.48 0.07 0.48 1.00 ITPKA 4.206E+07 2.535E+06 8.613E+07 1.016E+08 -0.57 0.04 -0.50 0.93 ITPKB 1.093E+08 3.000E+07 1.856E+08 1.016E+08 0.59 0.51 0.41 0.93 ITPKC 1.405E+08 5.921E+07 3.078E+08 2.345E+06 1.12 1.02 1.52 0.02

126 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD JAK1 8.051E+07 5.219E+07 7.494E+07 5.287E+07 0.09 0.90 -0.60 0.48 JAK2 1.034E+08 6.248E+06 2.669E+08 7.222E+07 0.49 0.11 1.15 0.66 JAK3 2.037E+08 1.203E+07 3.390E+08 1.706E+07 2.21 0.21 1.81 0.16 JIK 3.278E+07 2.024E+07 6.818E+07 6.205E+07 -0.72 0.35 -0.66 0.57 KDR 3.806E+07 1.485E+07 1.274E+08 7.875E+07 -0.63 0.25 -0.12 0.72 KHK 1.135E+08 3.663E+06 2.098E+08 1.933E+07 0.66 0.06 0.63 0.18 TNIK 4.005E+07 8.551E+06 9.696E+07 1.208E+08 -0.60 0.15 -0.40 1.10 MAST3 8.028E+07 2.919E+07 2.036E+08 1.831E+07 0.09 0.50 0.57 0.17 KIAA0999 1.683E+08 7.078E+06 3.127E+08 8.499E+07 1.60 0.12 1.57 0.77 KIAA1361 2.047E+07 1.295E+07 4.576E+07 5.489E+07 -0.94 0.22 -0.87 0.50 KIAA1399 4.823E+07 1.306E+07 9.669E+07 1.019E+08 -0.46 0.22 -0.40 0.93 KIAA1639 1.127E+08 6.036E+07 1.872E+08 8.848E+07 0.65 1.04 0.42 0.81 KIAA1765 1.534E+08 3.082E+07 3.185E+08 7.696E+07 1.34 0.53 1.62 0.70 KIAA1804 1.015E+08 1.229E+07 1.804E+08 1.013E+08 0.45 0.21 0.36 0.92 KIAA1811 8.681E+07 2.738E+07 2.430E+08 1.160E+08 0.20 0.47 0.93 1.06 LMTK3 4.395E+07 2.074E+07 1.355E+08 1.219E+08 -0.53 0.36 -0.05 1.11 KIF13B 6.922E+07 1.164E+06 1.844E+08 7.384E+07 -0.10 0.02 0.40 0.67 KIS 2.286E+07 1.930E+07 4.481E+07 5.956E+07 -0.90 0.33 -0.88 0.54 KIT 5.862E+07 7.348E+06 9.282E+07 7.365E+07 -0.28 0.13 -0.44 0.67 LMTK2 8.227E+07 5.525E+07 1.064E+08 1.149E+08 0.12 0.95 -0.31 1.05 KSR2 9.384E+07 1.160E+07 1.815E+08 3.208E+07 0.32 0.20 0.37 0.29 LAK 9.942E+07 8.128E+06 1.343E+08 5.502E+07 0.42 0.14 -0.06 0.50 LATS1 2.968E+07 2.192E+07 3.704E+07 4.311E+07 -0.78 0.38 -0.95 0.39 LATS2 3.281E+07 2.603E+07 7.665E+07 8.426E+07 -0.72 0.45 -0.59 0.77 7.242E+07 4.175E+06 1.720E+08 7.816E+07 -0.04 0.07 0.28 0.71 LCP2 1.469E+08 1.967E+07 3.208E+08 5.889E+07 1.23 0.34 1.64 0.54 LIM 1.495E+07 5.977E+06 6.074E+07 7.347E+07 -1.03 0.10 -0.73 0.67 LIMK1 3.548E+07 5.199E+06 6.130E+07 5.310E+07 -0.68 0.09 -0.73 0.48 LIMK2 7.110E+07 4.946E+05 1.040E+08 1.007E+08 -0.07 0.01 -0.34 0.92 LOC115704 6.198E+07 1.512E+07 1.012E+08 7.887E+07 -0.22 0.26 -0.36 0.72 LOC149420 1.940E+08 3.721E+07 3.958E+08 2.984E+07 2.04 0.64 2.32 0.27 PRPS1L1 5.160E+07 8.524E+06 1.171E+08 7.496E+07 -0.40 0.15 -0.22 0.68 LOC340371 6.246E+07 7.217E+06 1.145E+08 6.082E+07 -0.22 0.12 -0.24 0.55 LOC91807 4.332E+07 3.510E+07 3.851E+07 4.199E+07 -0.54 0.60 -0.93 0.38 LTK 3.310E+08 3.611E+07 3.362E+08 1.805E+08 4.39 0.62 1.78 1.64 LYK5 4.294E+07 2.155E+07 1.674E+08 1.171E+07 -0.55 0.37 0.24 0.11 LYN 4.201E+07 7.520E+06 8.697E+07 3.890E+07 -0.57 0.13 -0.49 0.35 SMAD7 1.682E+08 4.999E+07 3.050E+08 2.778E+07 1.60 0.86 1.49 0.25 MAGI-3 7.993E+07 1.185E+07 8.604E+07 8.239E+07 0.08 0.20 -0.50 0.75 MAK 2.146E+08 3.193E+06 1.965E+08 1.001E+08 2.40 0.05 0.51 0.91 MALT1 3.442E+07 5.635E+06 7.681E+07 8.947E+07 -0.70 0.10 -0.58 0.82 MAP2K1 1.730E+07 8.314E+06 2.687E+07 3.559E+07 -0.99 0.14 -1.04 0.32 MAP2K1IP1 2.922E+07 4.565E+06 6.281E+07 5.789E+07 -0.79 0.08 -0.71 0.53 MAP2K2 1.959E+07 7.171E+06 3.379E+07 2.012E+07 -0.95 0.12 -0.98 0.18 MAP2K3 5.773E+06 1.736E+06 1.389E+07 1.961E+07 -1.19 0.03 -1.16 0.18 MAP2K4 4.411E+07 1.271E+07 6.589E+07 1.728E+07 -0.53 0.22 -0.68 0.16 MAP2K5 8.269E+07 8.080E+06 1.466E+08 9.158E+07 0.13 0.14 0.05 0.83 MAP2K6 6.831E+06 4.014E+06 1.064E+07 1.342E+07 -1.17 0.07 -1.19 0.12 MAP2K7 4.250E+07 1.535E+07 3.718E+07 2.925E+07 -0.56 0.26 -0.95 0.27 MAP3K1 1.279E+07 3.739E+06 2.325E+07 1.642E+07 -1.07 0.06 -1.07 0.15 MAP3K10 1.109E+08 2.685E+07 2.491E+08 3.410E+06 0.62 0.46 0.99 0.03 MAP3K11 1.253E+08 1.003E+07 1.953E+08 1.281E+08 0.86 0.17 0.50 1.17 MAP3K12 1.549E+08 1.638E+07 3.128E+08 2.176E+07 1.37 0.28 1.57 0.20 MAP3K13 5.617E+07 1.864E+07 7.775E+07 4.997E+07 -0.32 0.32 -0.58 0.46 MAP3K14 3.310E+07 3.919E+06 6.292E+07 3.995E+07 -0.72 0.07 -0.71 0.36 MAP3K2 8.256E+06 4.075E+06 3.724E+07 3.039E+07 -1.15 0.07 -0.94 0.28 MAP3K3 2.165E+07 8.070E+06 3.344E+07 4.153E+07 -0.92 0.14 -0.98 0.38 MAP3K4 3.604E+07 3.211E+07 3.508E+07 1.179E+07 -0.63 0.46 -0.85 0.09

127 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD MAP3K5 1.053E+08 5.175E+07 2.625E+08 3.106E+07 0.36 0.74 0.89 0.24 MAP3K6 1.076E+08 7.309E+07 3.196E+08 9.420E+07 0.40 1.05 1.33 0.72 MAP3K7 1.145E+08 5.278E+07 2.471E+08 4.539E+07 0.50 0.76 0.77 0.35 MAP3K8 6.544E+07 3.587E+07 1.658E+08 1.905E+07 -0.21 0.51 0.15 0.15 MAP3K9 1.828E+07 1.344E+07 5.396E+06 6.384E+05 -0.88 0.19 -1.08 0.00 MAP4K1 5.735E+07 4.423E+07 1.740E+08 1.054E+06 -0.32 0.63 0.21 0.01 MAP4K2 1.109E+08 1.506E+07 2.693E+08 3.973E+07 0.44 0.22 0.94 0.30 MAP4K3 5.899E+07 3.744E+07 1.575E+08 5.973E+07 -0.30 0.54 0.09 0.46 MAP4K4 5.591E+07 2.233E+07 2.595E+08 2.457E+07 -0.34 0.32 0.87 0.19 MAP4K5 6.049E+07 1.491E+07 1.045E+08 4.703E+07 -0.28 0.21 -0.32 0.36 MAPK1 3.063E+07 7.094E+06 3.372E+07 3.277E+07 -0.71 0.10 -0.86 0.25 MAPK10 6.177E+07 7.932E+05 3.765E+07 1.046E+07 -0.26 0.01 -0.83 0.08 MAPK11 1.501E+06 3.010E+05 4.713E+06 5.155E+06 -1.12 0.00 -1.09 0.04 MAPK12 1.904E+07 5.245E+06 2.201E+07 2.727E+07 -0.87 0.08 -0.95 0.21 MAPK13 5.495E+07 4.402E+07 7.274E+07 5.643E+07 -0.36 0.63 -0.57 0.43 MAPK14 2.743E+08 1.812E+06 5.525E+08 1.289E+07 2.78 0.03 3.12 0.10 MAPK3 1.524E+08 2.082E+07 8.189E+07 9.293E+07 1.04 0.30 -0.50 0.71 MAPK4 8.746E+07 2.914E+07 1.517E+08 1.285E+08 0.11 0.42 0.04 0.99 MAPK6 1.172E+08 6.305E+07 2.024E+08 3.951E+06 0.53 0.90 0.43 0.03 MAPK7 3.064E+05 1.131E+05 1.147E+05 1.248E+05 -1.14 0.00 -1.12 0.00 MAPK8 8.445E+07 5.663E+07 7.387E+07 8.608E+06 0.07 0.81 -0.56 0.07 MAPK8IP1 1.644E+08 2.929E+07 2.733E+08 4.590E+07 1.21 0.42 0.97 0.35 MAPK8IP2 1.061E+08 4.731E+07 1.140E+08 5.113E+07 0.38 0.68 -0.25 0.39 MAPK8IP3 3.488E+07 5.565E+06 6.068E+07 4.575E+07 -0.64 0.08 -0.66 0.35 MAPK9 1.026E+08 2.226E+07 1.442E+08 1.820E+06 0.33 0.32 -0.02 0.01 MAPKAPK2 3.988E+06 2.007E+06 6.520E+06 8.702E+06 -1.09 0.03 -1.07 0.07 MAPKAPK3 9.170E+07 1.481E+07 1.524E+08 1.344E+08 0.17 0.21 0.05 1.03 MAPKAPK5 3.879E+07 1.051E+07 1.621E+08 7.574E+07 -0.59 0.15 0.12 0.58 MARK1 6.158E+07 1.488E+07 1.302E+08 7.639E+07 -0.26 0.21 -0.12 0.59 MARK2 7.904E+07 3.362E+07 8.092E+07 4.046E+07 -0.01 0.48 -0.50 0.31 MARK3 1.866E+08 2.391E+07 2.641E+08 7.492E+07 1.53 0.34 0.90 0.58 MARK4 2.734E+07 1.123E+07 2.381E+07 9.228E+06 -0.75 0.16 -0.94 0.07 MAST2 2.279E+07 6.799E+06 3.135E+07 2.256E+07 -0.82 0.10 -0.88 0.17 MATK 1.217E+07 5.777E+06 1.355E+07 1.260E+07 -0.97 0.08 -1.02 0.10 MBIP 2.347E+08 6.310E+07 3.533E+08 2.682E+07 2.22 0.90 1.59 0.21 MELK 8.499E+07 2.387E+07 2.565E+08 2.984E+07 0.07 0.34 0.84 0.23 MERTK 1.224E+08 7.621E+06 2.130E+08 1.626E+08 0.61 0.11 0.51 1.25 MET 3.516E+07 3.478E+06 6.641E+07 3.814E+07 -0.64 0.05 -0.61 0.29 MGC16169 3.464E+07 4.364E+05 5.497E+07 5.814E+07 -0.65 0.01 -0.70 0.45 STK32A 6.733E+07 1.037E+07 1.204E+08 2.160E+07 -0.18 0.15 -0.20 0.17 MGC26597 7.837E+07 2.507E+07 7.676E+07 8.075E+07 -0.02 0.36 -0.53 0.62 CSNK1A1L 2.373E+07 1.819E+07 2.265E+07 6.113E+06 -0.80 0.26 -0.95 0.05 MGC42105 2.000E+08 6.825E+07 3.086E+08 1.359E+07 1.72 0.98 1.25 0.10 C9ORF96 1.510E+08 7.832E+07 2.519E+08 2.802E+07 1.02 1.12 0.81 0.22 MGC45428 9.118E+07 1.594E+07 1.487E+08 4.517E+07 0.16 0.23 0.02 0.35 PIP5KL1 1.739E+08 3.561E+07 3.938E+08 2.026E+07 1.35 0.51 1.90 0.16 MGC4796 5.894E+07 9.412E+06 1.401E+08 6.794E+07 -0.30 0.13 -0.05 0.52 MGC5601 3.060E+07 9.460E+04 3.187E+07 3.608E+07 -0.71 0.00 -0.88 0.28 MGC8407 4.287E+07 2.119E+06 2.081E+08 9.800E+07 -0.53 0.03 0.47 0.75 MIDORI 2.117E+08 7.568E+07 3.580E+08 1.464E+07 1.89 1.08 1.62 0.11 MINK 4.349E+07 3.367E+07 5.714E+07 1.499E+07 -0.52 0.48 -0.69 0.12 MKNK1 6.251E+07 2.894E+07 6.872E+07 5.554E+07 -0.25 0.41 -0.60 0.43 MKNK2 1.426E+08 6.509E+07 3.185E+08 5.195E+07 0.90 0.93 1.32 0.40 MOS 3.437E+07 4.819E+06 6.501E+07 1.966E+07 -0.65 0.07 -0.62 0.15 MPP1 4.105E+07 1.635E+07 9.161E+07 6.364E+07 -0.56 0.23 -0.42 0.49 MPP2 4.834E+07 1.016E+07 1.101E+08 9.553E+07 -0.45 0.15 -0.28 0.73 MPP3 2.603E+07 8.245E+06 7.550E+07 6.239E+07 -0.77 0.12 -0.54 0.48 MPZL1 3.256E+08 8.907E+06 4.625E+08 2.159E+07 3.52 0.13 2.43 0.17

128 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD MRC2 1.391E+07 1.135E+07 5.270E+07 6.374E+07 -0.95 0.16 -0.72 0.49 MST1R 4.471E+06 5.315E+05 1.250E+06 1.033E+06 -1.08 0.01 -1.11 0.01 MST4 3.639E+07 1.012E+07 1.050E+08 3.050E+07 -0.62 0.14 -0.32 0.23 MUSK 5.780E+07 1.451E+07 5.175E+07 1.913E+07 -0.32 0.21 -0.73 0.15 MVD 2.103E+08 5.758E+07 2.936E+08 7.663E+07 1.87 0.82 1.13 0.59 MVK 7.496E+07 2.788E+06 1.687E+08 2.593E+07 -0.07 0.04 0.17 0.20 MYLK 1.851E+08 2.249E+07 3.726E+08 2.875E+07 1.51 0.32 1.74 0.22 MYLK2 3.298E+07 5.777E+06 2.837E+07 3.252E+07 -0.67 0.08 -0.91 0.25 MYO3A 9.095E+06 6.538E+06 1.558E+07 1.572E+07 -1.01 0.09 -1.00 0.12 MYO3B 5.193E+07 4.966E+05 1.275E+08 2.914E+07 -0.40 0.01 -0.15 0.22 NAGK 1.199E+07 4.453E+06 2.332E+07 1.136E+07 -0.97 0.06 -0.94 0.09 NBEA 2.438E+07 1.271E+06 7.859E+07 1.816E+07 -0.80 0.02 -0.52 0.14 COASY 7.114E+07 4.555E+06 3.887E+07 1.845E+07 -0.13 0.07 -0.83 0.14 NEK1 4.761E+07 3.323E+06 1.127E+08 4.746E+07 -0.46 0.05 -0.26 0.36 NEK11 6.752E+07 1.848E+07 1.133E+08 2.154E+07 -0.18 0.26 -0.25 0.17 NEK2 1.134E+08 3.616E+07 1.960E+08 2.874E+07 0.48 0.52 0.38 0.22 NEK3 4.726E+07 5.473E+06 9.498E+07 1.141E+07 -0.47 0.08 -0.39 0.09 NEK4 1.301E+08 4.920E+06 4.031E+08 6.872E+07 0.72 0.07 1.97 0.53 NEK6 1.816E+08 2.090E+07 4.079E+08 9.272E+06 1.46 0.30 2.01 0.07 NEK7 3.165E+07 1.244E+07 7.275E+07 4.662E+07 -0.69 0.18 -0.57 0.36 NEK8 1.541E+07 1.715E+07 9.857E+06 1.356E+07 -0.92 0.25 -1.05 0.10 NEK9 4.263E+07 3.801E+06 1.924E+08 7.043E+07 -0.52 0.05 -0.38 0.64 NLK 1.486E+08 1.730E+07 3.219E+08 1.237E+06 0.98 0.24 0.79 0.01 NME1 6.643E+06 7.580E+06 4.258E+07 2.681E+07 -1.02 0.11 -1.73 0.24 NME2 2.687E+07 1.344E+07 1.850E+08 9.888E+07 -0.74 0.19 -0.44 0.89 NME3 9.711E+07 2.337E+07 2.649E+08 1.516E+06 0.25 0.33 0.28 0.01 NME4 1.672E+08 2.619E+07 3.150E+08 1.602E+07 1.24 0.37 0.73 0.14 NME5 6.468E+07 6.601E+06 3.378E+08 2.580E+07 -0.21 0.09 0.94 0.23 NME6 7.015E+07 2.566E+07 2.157E+08 5.202E+06 -0.13 0.36 -0.17 0.05 NME7 4.960E+07 1.057E+07 2.257E+08 2.513E+06 -0.42 0.15 -0.08 0.02 NPR1 1.695E+07 4.899E+06 1.965E+08 4.717E+07 -0.88 0.07 -0.34 0.43 NPR2 1.686E+07 4.605E+06 1.601E+08 6.156E+07 -0.88 0.06 -0.67 0.56 NRBP 1.924E+06 2.381E+06 2.294E+07 2.791E+07 -1.09 0.03 -1.91 0.25 NRG3 1.635E+07 9.704E+06 1.464E+08 3.939E+06 -0.89 0.14 -0.79 0.04 NTRK1 3.497E+07 2.811E+06 2.456E+08 2.451E+07 -0.62 0.04 0.10 0.22 NTRK2 1.562E+08 1.444E+07 3.482E+08 6.673E+07 1.08 0.20 1.03 0.60 NTRK3 2.220E+08 3.156E+07 3.550E+08 5.254E+07 2.01 0.44 1.09 0.47 NYD-SP25 9.514E+07 5.285E+07 3.008E+08 5.589E+07 0.22 0.74 0.60 0.51 OSR1 1.835E+07 8.560E+05 1.615E+08 3.410E+07 -0.86 0.01 -0.66 0.31 P15RS 6.616E+07 2.595E+07 2.229E+08 4.130E+07 -0.18 0.37 -0.10 0.37 PACE-1 1.106E+07 5.037E+06 1.266E+08 1.293E+07 -0.96 0.07 -0.97 0.12 PACSIN1 5.748E+07 7.118E+05 3.067E+08 2.653E+07 -0.31 0.01 0.65 0.24 PAG 8.004E+06 9.421E+06 5.822E+07 2.239E+06 -1.00 0.13 -1.59 0.02 PAK1 1.460E+08 7.796E+06 3.457E+08 1.792E+07 0.94 0.11 1.01 0.16 PAK2 2.754E+08 7.053E+07 4.139E+08 3.204E+07 2.76 0.99 1.62 0.29 PAK3 2.079E+08 6.283E+07 3.887E+08 5.085E+07 1.81 0.88 1.40 0.46 PAK4 5.989E+07 1.654E+07 1.893E+08 1.542E+07 -0.27 0.23 -0.41 0.14 PAK6 9.924E+06 7.548E+06 7.541E+07 4.126E+07 -0.98 0.11 -1.44 0.37 PAK7 7.787E+07 1.986E+05 2.338E+08 5.651E+07 -0.02 0.00 0.00 0.51 PANK1 6.871E+07 2.259E+06 2.537E+08 9.362E+06 -0.15 0.03 0.18 0.08 PANK3 5.320E+07 4.987E+06 1.336E+08 5.809E+07 -0.37 0.07 -0.91 0.53 PANK4 2.481E+06 1.898E+06 2.234E+07 1.493E+06 -1.08 0.03 -1.92 0.01 PAPSS1 1.863E+08 2.737E+07 3.578E+08 3.364E+07 1.51 0.39 1.12 0.30 PAPSS2 8.571E+07 7.947E+06 2.560E+08 1.400E+07 0.09 0.11 0.20 0.13 PASK 8.113E+07 2.237E+07 1.956E+08 6.046E+07 0.03 0.32 -0.35 0.55 PCK1 1.679E+08 4.263E+06 3.568E+08 6.922E+07 1.25 0.06 1.11 0.63 PCK2 1.363E+08 9.832E+06 3.192E+08 2.232E+07 0.80 0.14 0.77 0.20 PCTK1 1.412E+08 2.447E+07 3.032E+08 3.045E+05 0.87 0.34 0.62 0.00

129 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD PCTK2 1.121E+08 2.319E+07 2.804E+08 2.465E+07 0.46 0.33 0.42 0.22 PCTK3 7.979E+07 1.007E+07 2.422E+08 4.371E+07 0.01 0.14 0.07 0.40 PDGFRA 1.388E+07 1.475E+06 1.104E+08 2.628E+07 -0.92 0.02 -1.12 0.24 PDGFRB 3.830E+07 4.735E+07 1.781E+08 3.073E+07 -0.58 0.67 -0.51 0.28 PDK1 4.194E+07 9.217E+06 2.091E+08 5.316E+06 -0.53 0.13 -0.23 0.05 PDK2 1.150E+08 3.178E+07 3.038E+08 1.165E+07 0.50 0.45 0.63 0.11 PDK3 7.497E+07 3.044E+06 1.794E+08 1.112E+07 -0.06 0.04 -0.50 0.10 PDK4 2.634E+08 7.316E+07 4.234E+08 1.785E+07 2.59 1.03 1.71 0.16 PDPK1 4.113E+07 3.339E+06 2.054E+08 1.901E+07 -0.54 0.05 -0.26 0.17 PFKFB1 1.960E+07 9.205E+06 1.399E+08 4.863E+07 -0.84 0.13 -0.85 0.44 PFKFB2 1.326E+08 3.328E+07 3.420E+08 5.404E+07 0.75 0.47 0.97 0.49 PFKFB3 1.788E+07 7.560E+06 6.348E+07 1.117E+07 -0.86 0.11 -1.54 0.10 PFKFB4 9.832E+07 5.153E+07 2.561E+08 3.488E+07 0.27 0.73 0.20 0.32 PFKL 9.088E+07 9.991E+06 4.014E+08 1.165E+07 0.16 0.14 1.51 0.11 PFKM 7.460E+07 2.385E+07 3.148E+08 2.684E+07 -0.07 0.34 0.73 0.24 PFKP 2.227E+07 1.459E+07 1.322E+08 4.037E+07 -0.80 0.21 -0.92 0.36 PFTK1 8.158E+07 1.653E+07 3.245E+08 3.654E+07 0.03 0.23 0.82 0.33 PGK1 1.321E+08 2.027E+07 3.259E+08 1.026E+07 0.74 0.29 0.83 0.09 PGK2 5.433E+07 7.489E+06 2.449E+08 6.143E+06 -0.35 0.11 0.10 0.06 PHKA1 1.171E+08 6.389E+07 3.124E+08 7.237E+07 0.53 0.90 0.71 0.65 PHKA2 9.794E+07 4.707E+07 3.082E+08 2.393E+07 0.26 0.66 0.67 0.22 PHKG1 8.462E+07 1.616E+07 2.626E+08 1.453E+07 0.08 0.23 0.26 0.13 PHKG2 2.794E+07 1.614E+06 1.430E+08 3.665E+07 -0.72 0.02 -0.82 0.33 PI4K2B 7.470E+06 8.210E+06 1.254E+08 2.218E+07 -1.01 0.12 -0.98 0.20 PI4KII 2.826E+08 7.331E+07 4.697E+08 3.071E+06 2.86 1.03 2.13 0.03 PIK3C2A 1.964E+07 2.722E+07 1.502E+08 2.746E+07 -0.84 0.38 -0.76 0.25 PIK3C2B 2.405E+08 8.168E+07 4.852E+08 7.272E+06 2.27 1.15 2.27 0.07 PIK3C2G 1.507E+08 4.205E+07 3.936E+08 5.742E+07 1.01 0.59 1.44 0.52 PIK3CA 4.549E+07 5.273E+06 1.730E+08 1.383E+07 -0.48 0.07 -0.55 0.13 PIK3CB 1.305E+07 1.340E+07 1.043E+08 3.926E+07 -0.93 0.19 -1.17 0.35 PIK3CG 4.324E+07 1.823E+07 2.235E+08 1.897E+07 -0.51 0.26 -0.10 0.17 PIK3R1 1.483E+08 9.898E+05 3.760E+08 4.060E+07 0.97 0.01 1.28 0.37 PIK3R2 1.562E+07 1.091E+06 1.027E+08 2.345E+07 -0.90 0.02 -1.19 0.21 PIK3R3 5.954E+06 6.661E+06 1.013E+08 9.914E+06 -1.03 0.09 -1.20 0.09 PIK3R4 3.430E+07 4.210E+07 1.345E+08 4.383E+06 -0.63 0.59 -0.90 0.04 PIK4CA 3.219E+07 2.543E+07 1.624E+08 6.761E+07 -0.66 0.36 -0.65 0.61 PIK4CB 3.908E+07 3.840E+07 1.400E+08 9.395E+06 -0.57 0.54 -0.85 0.08 PIM1 3.938E+07 2.661E+07 2.159E+08 3.298E+07 -0.56 0.37 -0.17 0.30 PIM2 9.411E+07 4.063E+06 3.157E+08 3.279E+07 0.21 0.06 0.74 0.30 PINK1 1.331E+08 2.328E+06 3.507E+08 2.183E+06 0.76 0.03 1.05 0.02 PIP5K1A 2.125E+07 1.960E+07 8.873E+07 1.079E+07 -0.82 0.28 -1.32 0.10 PIP5K2A 2.792E+07 1.090E+07 1.895E+08 4.647E+07 -0.72 0.15 -0.40 0.42 PIP5K2B 2.018E+07 5.168E+06 1.369E+08 4.625E+07 -0.83 0.07 -0.88 0.42 PIP5K2C 1.055E+07 5.603E+06 7.876E+07 1.043E+07 -0.82 0.10 -0.67 0.08 PITPNM3 5.568E+06 1.486E+06 1.229E+07 1.060E+07 -0.90 0.03 -1.18 0.08 STK32C 9.429E+06 8.186E+04 7.662E+07 1.254E+07 -0.84 0.00 -0.69 0.10 PKIA 9.375E+07 4.990E+06 2.406E+08 2.486E+07 0.61 0.09 0.56 0.19 PKIB 2.419E+07 2.162E+07 6.382E+07 1.158E+07 -0.59 0.37 -0.79 0.09 PKLR 2.523E+07 3.005E+07 3.959E+07 1.914E+05 -0.57 0.52 -0.97 0.00 PKM2 6.174E+07 3.078E+07 1.748E+08 2.724E+07 0.06 0.53 0.06 0.21 PKMYT1 1.864E+07 2.202E+05 1.476E+08 1.319E+07 -0.68 0.00 -0.15 0.10 PKN3 5.185E+06 2.902E+06 3.167E+07 1.531E+07 -0.91 0.05 -1.03 0.12 PLK1 5.760E+05 7.692E+05 3.877E+04 1.500E+04 -0.99 0.01 -1.27 0.00 EXOSC10 2.930E+06 2.085E+06 1.378E+07 1.095E+07 -0.95 0.04 -1.17 0.08 PMVK 2.983E+07 3.820E+07 4.284E+07 4.399E+07 -0.49 0.65 -0.95 0.34 PNKP 3.728E+04 2.545E+04 9.396E+04 2.415E+04 -1.00 0.00 -1.27 0.00 PPP1R1B 1.183E+08 7.628E+07 3.868E+08 3.916E+07 1.03 1.31 1.68 0.30 PPP2CA 5.038E+07 4.157E+07 1.672E+08 4.149E+07 -0.14 0.71 0.00 0.32

130 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD PPP2CB 5.588E+06 5.741E+06 1.935E+07 1.287E+07 -0.90 0.10 -1.13 0.10 PPP4C 4.840E+07 1.555E+07 1.798E+08 3.680E+07 -0.17 0.27 0.10 0.28 PRKAA1 2.411E+07 8.127E+06 7.610E+07 1.902E+06 -0.59 0.14 -0.69 0.01 PRKAA2 1.451E+08 8.011E+06 3.648E+08 6.124E+07 1.49 0.14 1.51 0.47 PRKACA 2.305E+07 2.860E+06 8.806E+07 5.161E+03 -0.60 0.05 -0.60 0.00 PRKACB 6.810E+07 3.303E+07 2.809E+08 3.651E+07 0.17 0.57 0.87 0.28 PRKACG 1.315E+08 2.878E+07 3.021E+08 9.005E+07 1.25 0.49 1.03 0.69 PRKAG1 1.901E+08 1.349E+07 4.227E+08 1.187E+07 2.26 0.23 1.95 0.09 PRKAG3 1.927E+06 1.957E+06 2.174E+06 5.283E+05 -0.97 0.03 -1.26 0.00 PRKAR1A 2.199E+08 5.967E+07 5.267E+08 1.607E+07 2.77 1.02 2.75 0.12 PRKAR2A 2.747E+07 1.430E+07 1.735E+08 1.238E+08 -0.53 0.25 0.05 0.95 PRKAR2B 2.635E+07 1.796E+06 5.355E+07 2.752E+07 -0.55 0.03 -0.86 0.21 PRKCA 1.090E+08 1.679E+07 2.968E+08 8.510E+07 0.87 0.29 0.99 0.65 PRKCABP 1.295E+08 6.941E+07 2.991E+08 1.941E+07 1.22 1.19 1.01 0.15 PRKCD 3.882E+07 1.736E+07 3.095E+08 4.184E+07 -0.33 0.30 1.09 0.32 PRKCE 5.399E+07 4.977E+07 1.603E+08 5.623E+07 -0.07 0.85 -0.05 0.43 PRKCG 3.532E+07 3.914E+06 1.696E+08 5.795E+07 -0.39 0.07 0.02 0.44 PRKCH 1.855E+08 7.482E+07 2.837E+08 4.767E+07 2.18 1.28 0.89 0.36 PRKCI 1.798E+07 8.148E+06 9.304E+07 2.944E+07 -0.69 0.14 -0.56 0.22 PRKCL1 3.477E+07 1.641E+07 1.222E+08 1.315E+07 -0.40 0.28 -0.34 0.10 PRKCL2 7.437E+06 3.086E+06 3.044E+07 3.050E+07 -0.87 0.05 -1.04 0.23 PRKCM 4.071E+07 1.545E+07 8.319E+07 7.082E+07 -0.30 0.26 -0.64 0.54 PRKCN 1.021E+08 1.126E+06 1.691E+08 5.485E+07 0.75 0.02 0.02 0.42 PRKCQ 1.113E+07 4.055E+06 4.249E+07 3.052E+07 -0.81 0.07 -0.95 0.23 PRKCSH 1.172E+07 7.890E+06 3.677E+07 3.468E+07 -0.80 0.14 -0.99 0.26 PRKCZ 7.420E+07 2.667E+06 2.499E+08 7.999E+07 0.27 0.05 0.64 0.61 PRKD2 9.941E+07 3.661E+07 2.665E+08 1.484E+07 0.70 0.63 0.76 0.11 PRKDC 1.976E+08 1.312E+06 3.197E+08 1.055E+07 2.39 0.02 1.17 0.08 PRKG1 2.825E+07 3.764E+06 1.403E+08 3.177E+07 -0.52 0.06 -0.20 0.24 PRKG2 8.045E+07 6.464E+06 2.297E+08 4.723E+07 0.38 0.11 0.48 0.36 PRKR 5.152E+07 3.446E+07 2.111E+08 4.779E+07 -0.12 0.59 0.34 0.36 PRKRA 2.791E+07 3.398E+06 6.215E+07 4.375E+07 -0.52 0.06 -0.80 0.33 PRKWNK1 3.040E+07 7.721E+05 8.308E+07 2.548E+07 -0.48 0.01 -0.64 0.19 PRKWNK2 8.876E+06 5.987E+06 8.964E+07 8.225E+07 -0.85 0.10 -0.59 0.63 PRKWNK3 7.036E+05 7.309E+05 3.656E+06 3.864E+06 -0.99 0.01 -1.25 0.03 PRKWNK4 7.836E+07 4.241E+07 2.497E+08 2.252E+07 0.34 0.73 0.63 0.17 PRKX 2.790E+07 4.658E+06 8.980E+07 3.781E+07 -0.52 0.08 -0.59 0.29 PRKY 9.437E+07 3.113E+07 2.055E+08 2.426E+07 0.62 0.53 0.30 0.19 PRPF4B 2.136E+07 2.379E+06 6.012E+07 6.025E+07 -0.63 0.04 -0.81 0.46 PRPS1 1.277E+08 1.752E+07 2.975E+08 8.782E+07 1.19 0.30 1.00 0.67 PRPS2 2.286E+07 1.073E+07 1.410E+08 7.567E+07 -0.61 0.18 -0.20 0.58 PRPSAP1 1.022E+07 5.226E+06 3.087E+07 3.500E+07 -0.82 0.09 -1.04 0.27 PRPSAP2 7.989E+07 3.337E+05 1.625E+08 5.841E+07 0.37 0.01 -0.03 0.45 TAO1 1.099E+08 5.830E+07 2.924E+08 6.396E+07 0.88 1.00 0.96 0.49 PSKH1 2.040E+08 7.181E+07 4.524E+08 4.268E+07 2.50 1.23 2.18 0.33 PSKH2 9.162E+06 4.781E+06 5.807E+07 5.298E+07 -0.84 0.08 -0.83 0.40 PTK2 1.628E+06 4.226E+05 5.132E+06 5.598E+06 -0.97 0.01 -1.23 0.04 PTK2B 1.463E+08 2.754E+07 3.814E+08 1.054E+08 1.51 0.47 1.64 0.80 PTK6 1.343E+08 9.036E+06 3.489E+08 5.699E+07 1.30 0.15 1.39 0.44 PTK7 4.524E+06 1.258E+06 4.655E+07 4.821E+07 -0.92 0.02 -0.92 0.37 PTK9 9.817E+07 5.326E+06 3.412E+08 5.935E+07 0.68 0.09 1.33 0.45 PTK9L 4.011E+07 9.108E+06 1.738E+08 9.497E+07 -0.31 0.16 0.05 0.73 PTPN5 8.032E+07 1.367E+07 2.994E+08 7.197E+07 0.38 0.23 1.01 0.55 PTPRG 2.022E+07 1.158E+07 1.725E+08 5.885E+07 -0.65 0.20 0.04 0.45 PTPRJ 4.138E+06 2.458E+06 2.757E+07 3.590E+07 -0.93 0.04 -1.06 0.27 PTPRR 1.999E+07 6.186E+06 7.428E+07 3.321E+07 -0.66 0.11 -0.71 0.25 PTPRT 1.097E+08 4.728E+07 3.099E+08 4.286E+07 0.88 0.81 1.09 0.33 PXK 8.227E+07 2.974E+07 2.257E+08 5.606E+07 0.41 0.51 0.45 0.43

131 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD PYCS 2.084E+07 6.858E+06 1.492E+08 1.200E+08 -0.64 0.12 -0.13 0.92 RAC1 8.397E+07 1.591E+07 2.037E+08 4.262E+07 0.44 0.27 0.28 0.33 RAF1 7.320E+07 1.797E+07 2.436E+08 1.437E+08 0.25 0.31 0.59 1.10 RAGE 1.341E+08 9.212E+06 3.089E+08 9.406E+07 1.30 0.16 1.09 0.72 RASGRF2 6.988E+06 3.635E+05 2.309E+07 2.837E+07 -0.88 0.01 -1.10 0.22 PDXK 3.404E+07 1.505E+07 1.616E+08 6.577E+07 -0.42 0.26 -0.04 0.50 PRKCB1 4.106E+07 1.230E+06 8.578E+07 3.790E+07 -0.30 0.02 -0.62 0.29 RBKS 1.005E+08 2.550E+07 3.222E+08 2.115E+07 0.27 0.41 1.33 0.18 RET 6.334E+07 4.376E+07 1.679E+08 4.935E+07 -0.32 0.70 0.01 0.42 RFP 3.045E+07 3.050E+07 9.832E+07 9.973E+07 -0.85 0.49 -0.59 0.86 GRK1 2.270E+08 5.410E+07 4.422E+08 1.451E+07 2.29 0.86 2.37 0.12 RIOK1 5.782E+07 5.735E+07 1.210E+08 1.017E+08 -0.41 0.91 -0.40 0.87 RIOK3 5.903E+07 5.218E+07 1.695E+08 7.198E+07 -0.39 0.83 0.02 0.62 RIPK1 1.987E+08 6.313E+07 2.927E+08 1.034E+08 1.84 1.01 1.08 0.89 RIPK2 8.174E+07 2.303E+07 2.877E+08 8.198E+07 -0.03 0.37 1.04 0.71 RIPK3 5.062E+07 2.429E+07 1.804E+08 5.066E+07 -0.52 0.39 0.11 0.44 RNASEL 2.161E+07 1.922E+07 1.010E+08 6.642E+07 -0.99 0.31 -0.57 0.57 ROCK1 1.378E+07 1.616E+07 3.728E+07 5.072E+07 -1.11 0.26 -1.12 0.44 ROCK2 1.347E+07 1.129E+07 3.258E+07 4.063E+07 -1.12 0.18 -1.16 0.35 ROR1 3.798E+07 2.372E+07 1.082E+08 8.669E+07 -0.73 0.38 -0.51 0.75 ROR2 7.502E+07 3.736E+07 2.131E+08 5.094E+07 -0.14 0.60 0.39 0.44 ROS1 5.280E+07 4.049E+07 6.429E+07 6.755E+07 -0.49 0.65 -0.89 0.58 RP2 1.674E+08 6.967E+07 1.885E+08 4.746E+07 1.34 1.11 0.18 0.41 RPS6KA1 8.009E+07 5.813E+07 2.518E+08 3.934E+07 -0.05 0.93 0.73 0.34 RPS6KA2 1.678E+08 7.921E+07 2.384E+08 1.501E+08 1.34 1.26 0.61 1.29 RPS6KA3 6.206E+07 5.270E+07 1.832E+08 1.297E+08 -0.34 0.84 0.14 1.12 RPS6KA4 2.344E+07 2.867E+07 2.994E+07 2.574E+07 -0.96 0.46 -1.18 0.22 RPS6KA5 2.268E+07 2.661E+07 5.933E+07 7.926E+07 -0.97 0.42 -0.93 0.68 RPS6KA6 1.085E+08 1.172E+07 1.932E+08 1.981E+07 0.40 0.19 0.22 0.17 RPS6KB1 1.122E+08 2.957E+07 1.763E+08 9.130E+07 0.46 0.47 0.08 0.79 RPS6KB2 1.015E+08 1.238E+07 1.922E+08 1.160E+08 0.29 0.20 0.22 1.00 RPS6KC1 9.637E+07 3.049E+07 1.748E+08 1.205E+08 0.20 0.49 0.07 1.04 RPS6KL1 8.058E+07 7.858E+07 9.797E+07 1.359E+08 -0.05 1.25 -0.60 1.17 RYK 1.077E+08 6.250E+07 1.868E+08 8.592E+07 0.39 1.00 0.17 0.74 SAST 1.733E+08 9.585E+07 2.242E+08 1.668E+08 1.43 1.53 0.49 1.44 SCAP1 4.769E+07 4.615E+07 1.012E+08 1.287E+08 -0.57 0.74 -0.57 1.11 SCYL1 5.037E+07 4.384E+07 1.808E+08 7.480E+07 -0.53 0.70 0.12 0.64 SEPHS1 7.011E+07 4.994E+07 8.878E+07 9.503E+07 -0.21 0.80 -0.67 0.82 SGK 8.773E+07 5.421E+07 2.277E+08 6.516E+07 0.07 0.86 0.52 0.56 SGK2 4.756E+07 3.642E+07 1.123E+08 6.383E+07 -0.57 0.58 -0.47 0.55 SGKL 5.466E+07 4.514E+07 1.447E+08 9.499E+07 -0.46 0.72 -0.19 0.82 SHC1 9.217E+07 1.187E+07 2.017E+08 9.967E+07 0.14 0.19 0.30 0.86 SIK2 1.664E+07 1.847E+07 6.617E+07 5.592E+07 -1.07 0.29 -0.87 0.48 SLK 1.087E+08 7.725E+07 1.545E+08 7.689E+07 0.40 1.23 -0.11 0.66 SMG1 3.750E+07 2.980E+07 4.814E+07 6.039E+07 -0.73 0.48 -1.02 0.52 SNARK 1.924E+07 1.647E+07 2.910E+07 3.575E+07 -1.03 0.26 -1.19 0.31 SNF1LK 1.260E+08 6.616E+07 2.775E+08 2.205E+07 0.68 1.06 0.95 0.19 PLK2 6.628E+07 6.049E+07 6.752E+07 3.327E+07 -0.28 0.96 -0.86 0.29 SNRK 5.969E+07 5.902E+07 9.004E+07 5.009E+07 -0.38 0.94 -0.66 0.43 SOCS1 6.643E+07 6.114E+07 1.452E+08 3.861E+07 -0.27 0.98 -0.19 0.33 SOCS5 4.454E+07 4.049E+07 8.240E+07 9.582E+07 -0.62 0.65 -0.73 0.82 SPA17 9.361E+07 6.772E+07 8.504E+07 4.351E+07 0.16 1.08 -0.71 0.37 SPEC2 8.795E+07 2.305E+07 2.172E+08 7.562E+07 0.07 0.37 0.43 0.65 SPHK1 5.908E+07 5.624E+07 1.218E+08 6.142E+07 -0.39 0.90 -0.39 0.53 SPHK2 1.223E+08 9.452E+07 1.774E+08 1.328E+08 0.62 1.51 0.09 1.14 SEPHS2 2.153E+08 1.291E+07 4.272E+08 9.453E+07 2.10 0.21 2.24 0.81 SQSTM1 5.279E+07 2.665E+07 1.284E+08 5.931E+07 -0.49 0.43 -0.33 0.51 SRC 3.301E+07 3.004E+07 1.190E+08 8.668E+07 -0.81 0.48 -0.41 0.75

132 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD SRMS 8.492E+07 5.134E+07 1.394E+08 1.851E+07 0.02 0.82 -0.24 0.16 SRPK1 6.421E+07 6.643E+07 1.257E+08 3.816E+07 -0.31 1.06 -0.36 0.33 SRPK2 1.053E+08 2.256E+07 2.254E+08 1.269E+08 0.35 0.36 0.50 1.09 ABI1 7.339E+07 6.425E+07 1.061E+08 8.839E+07 -0.16 1.02 -0.53 0.76 SSTK 2.555E+07 1.952E+07 6.368E+07 7.384E+07 -0.92 0.31 -0.89 0.64 STK10 8.636E+07 3.422E+07 2.625E+08 8.752E+07 0.05 0.55 0.82 0.75 STK11 2.986E+07 2.374E+07 5.326E+07 6.946E+07 -0.86 0.38 -0.98 0.60 STK16 4.697E+07 4.948E+07 8.942E+07 5.615E+07 -0.58 0.79 -0.67 0.48 STK17A 2.979E+07 1.869E+07 5.960E+07 6.429E+07 -0.86 0.30 -0.93 0.55 STK17B 3.107E+07 2.772E+07 5.172E+07 2.559E+07 -0.84 0.44 -0.99 0.22 PLK4 6.383E+07 3.709E+07 1.266E+08 8.873E+06 -0.31 0.59 -0.35 0.08 STK19 1.257E+08 4.809E+07 2.662E+08 8.674E+07 0.67 0.77 0.85 0.75 STK22B 1.017E+08 1.676E+07 1.207E+08 5.967E+07 0.29 0.27 -0.40 0.51 STK22C 7.176E+07 3.079E+07 1.254E+08 9.365E+07 -0.19 0.49 -0.36 0.81 STK22D 1.093E+08 3.779E+07 1.735E+08 1.514E+07 0.41 0.60 0.05 0.13 STK23 4.110E+07 3.693E+07 3.294E+07 2.396E+07 -0.68 0.59 -1.16 0.21 STK24 2.842E+07 2.560E+07 6.608E+07 5.460E+07 -0.88 0.41 -0.87 0.47 STK25 2.340E+08 7.097E+06 4.184E+08 2.388E+07 2.40 0.11 2.16 0.21 STK29 1.867E+08 2.391E+06 3.151E+08 3.046E+07 1.65 0.04 1.27 0.26 STK3 9.298E+07 5.832E+07 2.262E+08 5.546E+06 0.15 0.93 0.51 0.05 STK31 7.658E+07 1.572E+07 1.909E+08 4.728E+07 -0.11 0.25 0.20 0.41 STK33 1.606E+08 7.643E+07 3.605E+08 4.693E+07 1.23 1.22 1.66 0.40 STK35 2.178E+08 3.647E+07 4.122E+08 1.362E+07 2.14 0.58 2.11 0.12 STK36 5.918E+07 3.529E+07 1.605E+08 5.966E+07 -0.39 0.56 -0.06 0.51 STK38 1.363E+08 8.374E+06 3.607E+08 8.928E+06 0.84 0.13 1.66 0.08 STK38L 1.425E+08 7.120E+07 3.418E+08 7.295E+06 0.94 1.14 1.50 0.06 STK39 1.505E+08 4.492E+07 3.051E+08 7.656E+07 1.07 0.72 1.19 0.66 STK4 4.110E+07 3.486E+07 5.905E+07 1.577E+07 -0.68 0.56 -0.93 0.14 STK6 1.799E+07 2.464E+07 9.941E+06 6.087E+06 -1.05 0.39 -1.35 0.05 SYK 4.947E+07 5.170E+07 7.947E+07 7.468E+07 -0.79 0.86 -0.99 0.67 TAF1 1.669E+08 1.449E+07 2.836E+08 3.795E+07 1.15 0.24 0.84 0.34 TAF1L 1.196E+08 5.670E+07 2.833E+08 1.776E+04 0.37 0.94 0.84 0.00 TAO1 2.067E+08 5.518E+07 2.673E+08 7.104E+06 1.81 0.91 0.70 0.06 TBK1 9.100E+07 1.070E+08 6.831E+07 8.030E+07 -0.10 1.77 -1.09 0.72 TEC 1.295E+08 4.368E+07 2.562E+08 2.595E+07 0.54 0.72 0.60 0.23 TEK 1.091E+08 5.860E+07 2.280E+08 8.981E+07 0.20 0.97 0.34 0.81 TESK1 8.691E+07 5.810E+07 8.611E+07 6.547E+07 -0.17 0.96 -0.93 0.59 TESK2 2.271E+07 2.325E+07 2.474E+07 1.549E+07 -1.23 0.39 -1.48 0.14 TEX14 7.179E+06 9.768E+06 3.770E+06 4.227E+06 -1.49 0.16 -1.67 0.04 TGFBR1 1.939E+07 1.833E+07 1.138E+08 6.078E+07 -1.29 0.30 -0.68 0.55 TGFBR2 7.392E+07 6.631E+07 7.891E+07 1.877E+07 -0.39 1.10 -1.00 0.17 TIE 2.293E+08 4.641E+07 3.931E+08 1.714E+06 2.19 0.77 1.82 0.02 TJP2 1.485E+08 6.722E+07 2.231E+08 4.043E+07 0.85 1.11 0.30 0.36 TK1 5.870E+07 5.179E+07 5.656E+07 7.201E+07 -0.64 0.86 -1.20 0.65 TK2 6.099E+07 4.698E+07 1.022E+08 9.491E+07 -0.60 0.78 -0.79 0.85 TLK1 1.020E+08 5.396E+07 1.859E+08 2.987E+07 0.08 0.89 -0.04 0.27 TLK2 1.398E+08 4.358E+07 3.395E+08 1.116E+07 0.71 0.72 1.34 0.10 TLR1 7.630E+07 4.306E+07 2.870E+08 4.300E+07 -0.35 0.71 0.87 0.39 TLR3 1.323E+08 4.819E+07 3.369E+08 1.597E+07 0.58 0.80 1.32 0.14 TLR4 7.371E+07 6.276E+07 4.206E+07 1.188E+07 -0.39 1.04 -1.33 0.11 TLR6 3.420E+07 2.417E+07 2.872E+07 1.535E+07 -1.04 0.40 -1.45 0.14 TOPK 2.093E+08 4.947E+07 3.559E+08 6.097E+07 1.86 0.82 1.49 0.55 TPK1 7.869E+07 6.540E+07 3.892E+07 3.842E+07 -0.31 1.08 -1.35 0.34 TRAD 1.506E+08 6.993E+07 1.360E+08 9.035E+07 0.89 1.16 -0.48 0.81 TRIB2 9.178E+07 8.101E+07 4.397E+07 2.449E+07 -0.09 1.34 -1.31 0.22 TRIM 6.972E+07 4.773E+07 4.758E+07 5.111E+07 -0.45 0.79 -1.28 0.46 TRIO 6.166E+07 4.813E+07 1.086E+08 3.245E+07 -0.59 0.80 -0.73 0.29 TRPM6 1.190E+08 6.646E+07 2.505E+08 4.700E+07 0.36 1.10 0.54 0.42

133 Raw Data zScore (area x intensity / count) Basal Stimulated Basal Stimulated Gene Mean SD Mean SD Mean SD Mean SD TRPM7 7.716E+07 2.568E+07 2.848E+08 1.781E+07 -0.33 0.43 0.85 0.16 TTBK1 2.595E+08 7.452E+07 2.386E+08 2.650E+07 2.69 1.23 0.44 0.24 TTBK2 5.133E+07 4.604E+07 2.545E+07 1.963E+07 -0.76 0.76 -1.48 0.18 TTK 1.830E+08 5.418E+07 2.506E+08 3.753E+07 1.42 0.90 0.55 0.34 TTN 1.302E+08 5.941E+07 1.824E+08 9.166E+07 0.55 0.98 -0.07 0.82 TXK 1.379E+08 8.293E+07 1.189E+08 5.556E+07 0.67 1.37 -0.64 0.50 TXNDC3 1.339E+08 7.788E+07 2.946E+08 2.750E+07 0.61 1.29 0.94 0.25 TYK2 1.045E+08 6.559E+07 1.339E+08 2.037E+07 0.12 1.09 -0.50 0.18 TYRO3 8.316E+07 6.456E+07 7.289E+07 4.947E+07 -0.23 1.07 -1.05 0.44 UCK1 5.138E+07 3.061E+07 8.343E+07 2.335E+06 -0.76 0.51 -0.96 0.02 UGP2 1.056E+08 6.060E+07 2.630E+08 5.762E+07 0.14 1.00 0.66 0.52 ULK1 1.719E+08 1.702E+07 2.468E+08 8.031E+07 1.24 0.28 0.51 0.72 ULK2 4.173E+07 1.819E+07 7.020E+07 3.189E+07 -0.92 0.30 -1.07 0.29 UMP-CMPK 1.078E+08 6.050E+07 7.899E+07 3.778E+07 0.18 1.00 -1.00 0.34 UMPK 3.115E+07 2.432E+07 4.134E+07 2.145E+07 -1.09 0.40 -1.33 0.19 URKL1 7.868E+07 5.910E+07 1.326E+08 8.987E+07 -0.31 0.98 -0.51 0.81 VRK1 1.224E+08 4.969E+07 1.834E+08 3.093E+07 0.42 0.82 -0.06 0.28 VRK2 1.916E+08 3.070E+07 2.909E+08 2.120E+07 1.56 0.51 0.91 0.19 VRK3 9.564E+07 5.467E+07 2.696E+08 1.644E+07 -0.03 0.91 0.72 0.15 WEE1 8.504E+07 9.999E+07 1.045E+07 7.710E+05 -0.20 1.66 -1.61 0.01 WIF1 6.129E+07 2.624E+07 1.152E+08 1.561E+07 -0.59 0.43 -0.67 0.14 XYLB 2.214E+07 3.452E+06 8.294E+07 9.578E+07 -1.24 0.06 -0.96 0.86 YES1 1.499E+08 1.291E+07 2.147E+08 3.690E+07 0.87 0.21 0.22 0.33 YWHAH 5.062E+07 5.166E+06 8.669E+07 2.809E+07 -0.77 0.09 -0.93 0.25 YWHAQ 1.043E+08 4.816E+06 2.476E+08 6.449E+07 0.12 0.08 0.52 0.58 ZAK 2.041E+08 3.914E+07 2.803E+08 3.912E+07 1.77 0.65 0.81 0.35 ZAP70 5.922E+07 2.126E+07 6.516E+07 2.991E+07 -0.63 0.35 -1.12 0.27 TNFRSF10A 8.559E+07 2.928E+07 2.086E+08 6.662E+07 -0.19 0.48 0.17 0.60 TNK1 3.766E+07 3.321E+07 1.729E+07 1.445E+07 -0.99 0.55 -1.55 0.13 TSKS 5.865E+07 1.901E+07 5.374E+07 5.402E+06 -0.64 0.31 -1.22 0.05

Table A1 Results from primary kinase screen HA-GLUT4-HeLa cells were transfected with siRNA pools targeting 779 human kinases. 72 h following transfection, cells were serum starved, and then treated with or without 100ng/ml IGF1 for 15 min. GLUT4 translocation to the PM was determined by surface HA-GLUT4 immunofluorescence staining followed by automated image analysis.

134

Bibliography

135 Abe, Y., S. Matsumoto, et al. (2001). "Cloning and characterization of a p53-related protein kinase expressed in interleukin-2-activated cytotoxic T-cells, epithelial tumor cell lines, and the testes." J Biol Chem 276(47): 44003-11. Abe, Y., T. Takeuchi, et al. (2006). "A Small Ras-like protein Ray/Rab1c modulates the p53-regulating activity of PRPK." Biochem Biophys Res Commun 344(1): 377- 85. Abel, E. D., O. Peroni, et al. (2001). "Adipose-selective targeting of the GLUT4 gene impairs insulin action in muscle and liver." Nature 409(6821): 729-33. Aguirre, V., T. Uchida, et al. (2000). "The c-Jun NH(2)-terminal kinase promotes insulin resistance during association with insulin receptor substrate-1 and phosphorylation of Ser(307)." J Biol Chem 275(12): 9047-54. Aguirre, V., E. D. Werner, et al. (2002). "Phosphorylation of Ser307 in insulin receptor substrate-1 blocks interactions with the insulin receptor and inhibits insulin action." J Biol Chem 277(2): 1531-7. Ahmed, Z., B. J. Smith, et al. (2000). "The APS adapter protein couples the insulin receptor to the phosphorylation of c-Cbl and facilitates ligand-stimulated ubiquitination of the insulin receptor." FEBS Lett 475(1): 31-4. Ahn, M. Y., K. D. Katsanakis, et al. (2004). "Primary and essential role of the adaptor protein APS for recruitment of both c-Cbl and its associated protein CAP in insulin signaling." J Biol Chem 279(20): 21526-32. Al-Hakim, A. K., A. Zagorska, et al. (2008). "Control of AMPK-related kinases by USP9X and atypical Lys(29)/Lys(33)-linked polyubiquitin chains." Biochem J 411(2): 249-60. Aleman, L. M., J. Doench, et al. (2007). "Comparison of siRNA-induced off-target RNA and protein effects." Rna 13(3): 385-95. Altshuler, D., J. N. Hirschhorn, et al. (2000). "The common PPARgamma Pro12Ala polymorphism is associated with decreased risk of type 2 diabetes." Nat Genet 26(1): 76-80. Ameres, S. L., J. Martinez, et al. (2007). "Molecular basis for target RNA recognition and cleavage by human RISC." Cell 130(1): 101-12. Arkan, M. C., A. L. Hevener, et al. (2005). "IKK-beta links inflammation to obesity- induced insulin resistance." Nat Med 11(2): 191-8.

136 Atsumi, T., T. Nishio, et al. (2005). "Expression of inducible 6-phosphofructo-2- kinase/fructose-2,6-bisphosphatase/PFKFB3 isoforms in adipocytes and their potential role in glycolytic regulation." Diabetes 54(12): 3349-57. Austin, R. L., A. Rune, et al. (2008). "siRNA-mediated reduction of inhibitor of nuclear factor-kappaB kinase prevents tumor necrosis factor-alpha-induced insulin resistance in human skeletal muscle." Diabetes 57(8): 2066-73. Backer, J. M., M. G. Myers, Jr., et al. (1992). "Phosphatidylinositol 3'-kinase is activated by association with IRS-1 during insulin stimulation." Embo J 11(9): 3469-79. Bae, S. S., H. Cho, et al. (2003). "Isoform-specific regulation of insulin-dependent glucose uptake by Akt/protein kinase B." J Biol Chem 278(49): 49530-6. Balendran, A., A. Casamayor, et al. (1999). "PDK1 acquires PDK2 activity in the presence of a synthetic peptide derived from the carboxyl terminus of PRK2." Curr Biol 9(8): 393-404. Bandyopadhyay, D., A. Kusari, et al. (1997). "Protein-tyrosine phosphatase 1B complexes with the insulin receptor in vivo and is tyrosine-phosphorylated in the presence of insulin." J Biol Chem 272(3): 1639-45. Bandyopadhyay, G., M. L. Standaert, et al. (1999). "Effects of transiently expressed atypical (zeta, lambda), conventional (alpha, beta) and novel (delta, epsilon) protein kinase C isoforms on insulin-stimulated translocation of epitope-tagged GLUT4 glucose transporters in rat adipocytes: specific interchangeable effects of protein kinases C-zeta and C-lambda." Biochem J 337 ( Pt 3): 461-70. Bandyopadhyay, G., M. L. Standaert, et al. (1999). "Dependence of insulin-stimulated glucose transporter 4 translocation on 3-phosphoinositide-dependent protein kinase-1 and its target threonine-410 in the activation loop of protein kinase C- zeta." Mol Endocrinol 13(10): 1766-72. Barnes, P. J. and M. Karin (1997). "Nuclear factor-kappaB: a pivotal transcription factor in chronic inflammatory diseases." N Engl J Med 336(15): 1066-71. Baron, V., V. Calleja, et al. (1998). "p125Fak focal adhesion kinase is a substrate for the insulin and insulin-like growth factor-I tyrosine kinase receptors." J Biol Chem 273(12): 7162-8. Bastie, C. C., H. Zong, et al. (2007). "Integrative metabolic regulation of peripheral tissue fatty acid oxidation by the SRC kinase family member Fyn." Cell Metab 5(5): 371-81.

137 Baumann, C. A., V. Ribon, et al. (2000). "CAP defines a second signalling pathway required for insulin-stimulated glucose transport." Nature 407(6801): 202-7. Bayascas, J. R. and D. R. Alessi (2005). "Regulation of Akt/PKB Ser473 phosphorylation." Mol Cell 18(2): 143-5. Bell, G. I., T. Kayano, et al. (1990). "Molecular biology of mammalian glucose transporters." Diabetes Care 13(3): 198-208. Bellare, P., A. K. Kutach, et al. (2006). "Ubiquitin binding by a variant Jab1/MPN domain in the essential pre-mRNA splicing factor Prp8p." RNA 12(2): 292-302. Berger, J., P. Bailey, et al. (1996). "Thiazolidinediones produce a conformational change in peroxisomal proliferator-activated receptor-gamma: binding and activation correlate with antidiabetic actions in db/db mice." Endocrinology 137(10): 4189-95. Berwick, D. C., G. C. Dell, et al. (2004). "Protein kinase B phosphorylation of PIKfyve regulates the trafficking of GLUT4 vesicles." J Cell Sci 117(Pt 25): 5985-93. Bickel, P. E., P. E. Scherer, et al. (1997). "Flotillin and epidermal surface antigen define a new family of caveolae-associated integral membrane proteins." J Biol Chem 272(21): 13793-802. Blero, D., J. Zhang, et al. (2005). "Phosphatidylinositol 3,4,5-trisphosphate modulation in SHIP2-deficient mouse embryonic fibroblasts." Febs J 272(10): 2512-22. Boden, G. (1998). "Free fatty acids (FFA), a link between obesity and insulin resistance." Front Biosci 3: d169-75. Bogan, J. S., A. E. McKee, et al. (2001). "Insulin-responsive compartments containing GLUT4 in 3T3-L1 and CHO cells: regulation by amino acid concentrations." Mol Cell Biol 21(14): 4785-806. Bomberger, J. M., R. L. Barnaby, et al. (2009). "The deubiquitinating enzyme USP10 regulates the post-endocytic sorting of cystic fibrosis transmembrane conductance regulator in airway epithelial cells." J Biol Chem 284(28): 18778- 89. Boulkroun, S., D. Ruffieux-Daidie, et al. (2008). "Vasopressin-inducible ubiquitin- specific protease 10 increases ENaC cell surface expression by deubiquitylating and stabilizing sorting nexin 3." Am J Physiol Renal Physiol 295(4): F889-900. Boura-Halfon, S. and Y. Zick (2009). "Phosphorylation of IRS proteins, insulin action, and insulin resistance." Am J Physiol Endocrinol Metab 296(4): E581-91.

138 Bouzakri, K. and J. R. Zierath (2007). "MAP4K4 gene silencing in human skeletal muscle prevents tumor necrosis factor-alpha-induced insulin resistance." J Biol Chem 282(11): 7783-9. Brdicka, T., D. Pavlistova, et al. (2000). "Phosphoprotein associated with glycosphingolipid-enriched microdomains (PAG), a novel ubiquitously expressed transmembrane adaptor protein, binds the protein tyrosine kinase csk and is involved in regulation of T cell activation." J Exp Med 191(9): 1591-604. Brozinick, J. T., Jr., G. J. Etgen, Jr., et al. (1994). "Glucose uptake and GLUT-4 protein distribution in skeletal muscle of the obese Zucker rat." Am J Physiol 267(1 Pt 2): R236-43. Brozinick, J. T., Jr., S. C. McCoid, et al. (2001). "GLUT4 overexpression in db/db mice dose-dependently ameliorates diabetes but is not a lifelong cure." Diabetes 50(3): 593-600. Bruning, J. C., M. D. Michael, et al. (1998). "A muscle-specific insulin receptor knockout exhibits features of the metabolic syndrome of NIDDM without altering glucose tolerance." Mol Cell 2(5): 559-69. Bryant, N. J., R. Govers, et al. (2002). "Regulated transport of the glucose transporter GLUT4." Nat Rev Mol Cell Biol 3(4): 267-77. Cai, D., M. Yuan, et al. (2005). "Local and systemic insulin resistance resulting from hepatic activation of IKK-beta and NF-kappaB." Nat Med 11(2): 183-90. Cartee, G. D. and J. F. Wojtaszewski (2007). "Role of Akt substrate of 160 kDa in insulin-stimulated and contraction-stimulated glucose transport." Appl Physiol Nutr Metab 32(3): 557-66. Carvalho, E., K. Kotani, et al. (2005). "Adipose-specific overexpression of GLUT4 reverses insulin resistance and diabetes in mice lacking GLUT4 selectively in muscle." Am J Physiol Endocrinol Metab 289(4): E551-61. Chalfant, C. E., T. P. Ciaraldi, et al. (2000). "Protein kinase Ctheta expression is increased upon differentiation of human skeletal muscle cells: dysregulation in type 2 diabetic patients and a possible role for protein kinase Ctheta in insulin- stimulated glycogen synthase activity." Endocrinology 141(8): 2773-8. Chan, N. L. and C. P. Hill (2001). "Defining polyubiquitin chain topology." Nat Struct Biol 8(8): 650-2. Chaussade, C., G. W. Rewcastle, et al. (2007). "Evidence for functional redundancy of class IA PI3K isoforms in insulin signalling." Biochem J 404(3): 449-58.

139 Chen, C. H., T. Shaikenov, et al. (2011). "ER Stress Inhibits mTORC2 and Akt Signaling Through GSK-3{beta}-Mediated Phosphorylation of Rictor." Sci Signal 4(161): ra10. Chen, H., S. J. Wertheimer, et al. (1997). "Protein-tyrosine phosphatases PTP1B and syp are modulators of insulin-stimulated translocation of GLUT4 in transfected rat adipose cells." J Biol Chem 272(12): 8026-31. Chen, L. W., L. Egan, et al. (2003). "The two faces of IKK and NF-kappaB inhibition: prevention of systemic inflammation but increased local injury following intestinal ischemia-reperfusion." Nat Med 9(5): 575-81. Chen, P. Y., L. Weinmann, et al. (2008). "Strand-specific 5'-O-methylation of siRNA duplexes controls guide strand selection and targeting specificity." Rna 14(2): 263-74. Chen, S., D. H. Wasserman, et al. (2011). "Mice with AS160/TBC1D4-Thr649Ala knockin mutation are glucose intolerant with reduced insulin sensitivity and altered GLUT4 trafficking." Cell Metab 13(1): 68-79. Cheung, P. C., I. P. Salt, et al. (2000). "Characterization of AMP-activated protein kinase gamma-subunit isoforms and their role in AMP binding." Biochem J 346 Pt 3: 659-69. Chiang, S. H., C. A. Baumann, et al. (2001). "Insulin-stimulated GLUT4 translocation requires the CAP-dependent activation of TC10." Nature 410(6831): 944-8. Cho, H., J. Mu, et al. (2001). "Insulin resistance and a diabetes mellitus-like syndrome in mice lacking the protein kinase Akt2 (PKB beta)." Science 292(5522): 1728- 31. Choi, J. H., A. S. Banks, et al. (2010). "Anti-diabetic drugs inhibit obesity-linked phosphorylation of PPARgamma by Cdk5." Nature 466(7305): 451-6. Choi, S. J., S. S. Chung, et al. (2006). "Negative modulation of RXRalpha transcriptional activity by small ubiquitin-related modifier (SUMO) modification and its reversal by SUMO-specific protease SUSP1." J Biol Chem 281(41): 30669-77. Choi, Y., J. Zhang, et al. (2002). "PTEN, but not SHIP and SHIP2, suppresses the PI3K/Akt pathway and induces growth inhibition and apoptosis of myeloma cells." Oncogene 21(34): 5289-300.

140 Chuang, C. F. and E. M. Meyerowitz (2000). "Specific and heritable genetic interference by double-stranded RNA in Arabidopsis thaliana." Proc Natl Acad Sci U S A 97(9): 4985-90. Ciaraldi, T. P., S. E. Nikoulina, et al. (2007). "Role of glycogen synthase kinase-3 alpha in insulin action in cultured human skeletal muscle cells." Endocrinology 148(9): 4393-9. Cicirelli, M. F., N. K. Tonks, et al. (1990). "Microinjection of a protein-tyrosine- phosphatase inhibits insulin action in Xenopus oocytes." Proc Natl Acad Sci U S A 87(14): 5514-8. Claessen, J. H., B. Mueller, et al. (2010). "The transmembrane segment of a tail- anchored protein determines its degradative fate through dislocation from the endoplasmic reticulum." J Biol Chem 285(27): 20732-9. Clague, M. J. and S. Urbe (2006). "Endocytosis: the DUB version." Trends Cell Biol 16(11): 551-9. Clarke, J. F., P. W. Young, et al. (1994). "Inhibition of the translocation of GLUT1 and GLUT4 in 3T3-L1 cells by the phosphatidylinositol 3-kinase inhibitor, wortmannin." Biochem J 300 ( Pt 3): 631-5. Cleland, P. J., G. J. Appleby, et al. (1989). "Exercise-induced translocation of protein kinase C and production of diacylglycerol and phosphatidic acid in rat skeletal muscle in vivo. Relationship to changes in glucose transport." J Biol Chem 264(30): 17704-11. Cohen, M., F. Stutz, et al. (2003). "Ubp3 requires a , Bre5, to specifically de- ubiquitinate the COPII protein, Sec23." Nat Cell Biol 5(7): 661-7. Cohen, M., F. Stutz, et al. (2003). "Deubiquitination, a new player in Golgi to endoplasmic reticulum retrograde transport." J Biol Chem 278(52): 51989-92. Combaret, L., O. A. Adegoke, et al. (2005). "USP19 is a ubiquitin-specific protease regulated in rat skeletal muscle during catabolic states." Am J Physiol Endocrinol Metab 288(4): E693-700. Costanzo-Garvey, D. L., P. T. Pfluger, et al. (2009). "KSR2 is an essential regulator of AMP kinase, energy expenditure, and insulin sensitivity." Cell Metab 10(5): 366-78. Crute, B. E., K. Seefeld, et al. (1998). "Functional domains of the alpha1 catalytic subunit of the AMP-activated protein kinase." J Biol Chem 273(52): 35347-54.

141 Csibi, A., K. Cornille, et al. (2010). "The translation regulatory subunit eIF3f controls the kinase-dependent mTOR signaling required for muscle differentiation and hypertrophy in mouse." PLoS One 5(2): e8994. Csukai, M., C. H. Chen, et al. (1997). "The coatomer protein beta'-COP, a selective binding protein (RACK) for protein kinase Cepsilon." J Biol Chem 272(46): 29200-6. Cullen, B. R. (2006). "Enhancing and confirming the specificity of RNAi experiments." Nat Methods 3(9): 677-81. Dasgupta, R. and N. Perrimon (2004). "Using RNAi to catch Drosophila genes in a web of interactions: insights into cancer research." Oncogene 23(51): 8359-65. De Fea, K. and R. A. Roth (1997). "Modulation of insulin receptor substrate-1 tyrosine phosphorylation and function by mitogen-activated protein kinase." J Biol Chem 272(50): 31400-6. Douen, A. G., T. Ramlal, et al. (1990). "Exercise induces recruitment of the "insulin- responsive glucose transporter". Evidence for distinct intracellular insulin- and exercise-recruitable transporter pools in skeletal muscle." J Biol Chem 265(23): 13427-30. Downes, C. P., A. Gray, et al. (2005). "Probing phosphoinositide functions in signaling and membrane trafficking." Trends Cell Biol 15(5): 259-68. Duan, L. and M. H. Cobb (2010). "Calcineurin increases glucose activation of ERK1/2 by reversing negative feedback." Proc Natl Acad Sci U S A 107(51): 22314-9. Ducluzeau, P. H., L. M. Fletcher, et al. (2002). "Molecular mechanisms of insulin- stimulated glucose uptake in adipocytes." Diabetes Metab 28(2): 85-92. Durmus Tekir, S., P. Umit, et al. (2010). "Reconstruction of protein-protein interaction network of insulin signaling in Homo sapiens." J Biomed Biotechnol 2010: 690925. Echeverri, C. J., P. A. Beachy, et al. (2006). "Minimizing the risk of reporting false positives in large-scale RNAi screens." Nat Methods 3(10): 777-9. Echeverri, C. J. and N. Perrimon (2006). "High-throughput RNAi screening in cultured cells: a user's guide." Nat Rev Genet 7(5): 373-84. Eck, M. J., S. Dhe-Paganon, et al. (1996). "Structure of the IRS-1 PTB domain bound to the juxtamembrane region of the insulin receptor." Cell 85(5): 695-705. Eguez, L., A. Lee, et al. (2005). "Full intracellular retention of GLUT4 requires AS160 Rab GTPase activating protein." Cell Metab 2(4): 263-72.

142 El Annabi, S., N. Gautier, et al. (2001). "Focal adhesion kinase and Src mediate integrin regulation of insulin receptor phosphorylation." FEBS Lett 507(3): 247-52. Elbashir, S. M., J. Harborth, et al. (2001). "Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells." Nature 411(6836): 494-8. Elchebly, M., P. Payette, et al. (1999). "Increased insulin sensitivity and obesity resistance in mice lacking the protein tyrosine phosphatase-1B gene." Science 283(5407): 1544-8. Eldar-Finkelman, H. and E. G. Krebs (1997). "Phosphorylation of insulin receptor substrate 1 by glycogen synthase kinase 3 impairs insulin action." Proc Natl Acad Sci U S A 94(18): 9660-4. Emanuelli, B., P. Peraldi, et al. (2001). "SOCS-3 inhibits insulin signaling and is up- regulated in response to tumor necrosis factor-alpha in the adipose tissue of obese mice." J Biol Chem 276(51): 47944-9. Emanuelli, B., P. Peraldi, et al. (2000). "SOCS-3 is an insulin-induced negative regulator of insulin signaling." J Biol Chem 275(21): 15985-91. Endo, A., M. Matsumoto, et al. (2009). "Nucleolar structure and function are regulated by the deubiquitylating enzyme USP36." J Cell Sci 122(Pt 5): 678-86. Ernst, R., B. Mueller, et al. (2009). "The otubain YOD1 is a deubiquitinating enzyme that associates with p97 to facilitate protein dislocation from the ER." Mol Cell 36(1): 28-38. Facchin, S., M. Ruzzene, et al. (2007). "Phosphorylation and activation of the atypical kinase p53-related protein kinase (PRPK) by Akt/PKB." Cell Mol Life Sci 64(19-20): 2680-9. Farese, R. V., M. P. Sajan, et al. (2007). "Muscle-specific knockout of PKC-lambda impairs glucose transport and induces metabolic and diabetic syndromes." J Clin Invest 117(8): 2289-301. Filippa, N., C. L. Sable, et al. (2000). "Effect of phosphoinositide-dependent kinase 1 on protein kinase B translocation and its subsequent activation." Mol Cell Biol 20(15): 5712-21. Fire, A., S. Xu, et al. (1998). "Potent and specific genetic interference by double- stranded RNA in Caenorhabditis elegans." Nature 391(6669): 806-11. Foster, L. J., A. Rudich, et al. (2006). "Insulin-dependent interactions of proteins with GLUT4 revealed through stable isotope labeling by amino acids in cell culture (SILAC)." J Proteome Res 5(1): 64-75.

143 Fritsche, L., S. S. Neukamm, et al. (2010). "INSULIN-INDUCED SERINE PHOSPHORYLATION OF IRS-2 VIA ERK1/2 AND mTOR: STUDIES ON THE FUNCTION OF SER 675 AND SER 907." Am J Physiol Endocrinol Metab. Fruman, D. A., R. E. Meyers, et al. (1998). "Phosphoinositide kinases." Annu Rev Biochem 67: 481-507. Fu, Q., R. A. McKnight, et al. (2006). "Growth retardation alters the epigenetic characteristics of hepatic dual specificity phosphatase 5." Faseb J 20(12): 2127- 9. Fujishiro, M., Y. Gotoh, et al. (2003). "Three mitogen-activated protein kinases inhibit insulin signaling by different mechanisms in 3T3-L1 adipocytes." Mol Endocrinol 17(3): 487-97. Fulcher, F. K., B. T. Smith, et al. (2008). "Dual role for myosin II in GLUT4-mediated glucose uptake in 3T3-L1 adipocytes." Exp Cell Res 314(17): 3264-74. Galic, S., C. Hauser, et al. (2005). "Coordinated regulation of insulin signaling by the protein tyrosine phosphatases PTP1B and TCPTP." Mol Cell Biol 25(2): 819- 29. Gallastegui, N. and M. Groll (2010). "The 26S proteasome: assembly and function of a destructive machine." Trends Biochem Sci 35(11): 634-42. Galvez, T., M. N. Teruel, et al. (2007). "siRNA screen of the human signaling proteome identifies the PtdIns(3,4,5)P3-mTOR signaling pathway as a primary regulator of transferrin uptake." Genome Biol 8(7): R142. Gao, Z., D. Hwang, et al. (2002). "Serine phosphorylation of insulin receptor substrate 1 by inhibitor kappa B kinase complex." J Biol Chem 277(50): 48115-21. Gao, Z., Z. Wang, et al. (2007). "Inactivation of PKCtheta leads to increased susceptibility to obesity and dietary insulin resistance in mice." Am J Physiol Endocrinol Metab 292(1): E84-91. Gao, Z., X. Zhang, et al. (2004). "Inhibition of insulin sensitivity by free fatty acids requires activation of multiple serine kinases in 3T3-L1 adipocytes." Mol Endocrinol 18(8): 2024-34. Garcia-Guerra, L., I. Nieto-Vazquez, et al. (2010). "G protein-coupled receptor kinase 2 plays a relevant role in insulin resistance and obesity." Diabetes 59(10): 2407- 17.

144 Gareau, J. R. and C. D. Lima (2010). "The SUMO pathway: emerging mechanisms that shape specificity, conjugation and recognition." Nat Rev Mol Cell Biol 11(12): 861-71. Garetto, L. P., E. A. Richter, et al. (1984). "Enhanced muscle glucose metabolism after exercise in the rat: the two phases." Am J Physiol 246(6 Pt 1): E471-5. Gerich, J. E. (2000). "Physiology of glucose homeostasis." Diabetes Obes Metab 2(6): 345-50. Gibbs, E. M., J. L. Stock, et al. (1995). "Glycemic improvement in diabetic db/db mice by overexpression of the human insulin-regulatable glucose transporter (GLUT4)." J Clin Invest 95(4): 1512-8. Gilsdorf, M., T. Horn, et al. (2010). "GenomeRNAi: a database for cell-based RNAi phenotypes. 2009 update." Nucleic Acids Res 38(Database issue): D448-52. Giorgino, F., O. de Robertis, et al. (2000). "The sentrin-conjugating enzyme mUbc9 interacts with GLUT4 and GLUT1 glucose transporters and regulates transporter levels in skeletal muscle cells." Proc Natl Acad Sci U S A 97(3): 1125-30. Gnudi, L., E. Tozzo, et al. (1995). "High level overexpression of glucose transporter-4 driven by an adipose-specific promoter is maintained in transgenic mice on a high fat diet, but does not prevent impaired glucose tolerance." Endocrinology 136(3): 995-1002. Gonzalez, E. and T. E. McGraw (2006). "Insulin signaling diverges into Akt-dependent and -independent signals to regulate the recruitment/docking and the fusion of GLUT4 vesicles to the plasma membrane." Mol Biol Cell 17(10): 4484-93. Goodyear, L. J., F. Giorgino, et al. (1995). "Effects of contractile activity on tyrosine phosphoproteins and PI 3-kinase activity in rat skeletal muscle." Am J Physiol 268(5 Pt 1): E987-95. Goodyear, L. J., M. F. Hirshman, et al. (1991). "Exercise-induced translocation of skeletal muscle glucose transporters." Am J Physiol 261(6 Pt 1): E795-9. Goodyear, L. J., M. F. Hirshman, et al. (1992). "Glucose transporter number, function, and subcellular distribution in rat skeletal muscle after exercise training." Diabetes 41(9): 1091-9. Goodyear, L. J. and B. B. Kahn (1998). "Exercise, glucose transport, and insulin sensitivity." Annu Rev Med 49: 235-61.

145 Govers, R., A. C. Coster, et al. (2004). "Insulin increases cell surface GLUT4 levels by dose dependently discharging GLUT4 into a cell surface recycling pathway." Mol Cell Biol 24(14): 6456-66. Greene, M. W. and R. S. Garofalo (2002). "Positive and negative regulatory role of insulin receptor substrate 1 and 2 (IRS-1 and IRS-2) serine/threonine phosphorylation." Biochemistry 41(22): 7082-91. Gridley, S., W. S. Lane, et al. (2005). "Novel insulin-elicited phosphoproteins in adipocytes." Cell Signal 17(1): 59-66. Griffin, M. E., M. J. Marcucci, et al. (1999). "Free fatty acid-induced insulin resistance is associated with activation of protein kinase C theta and alterations in the insulin signaling cascade." Diabetes 48(6): 1270-4. Gual, P., T. Gremeaux, et al. (2003). "MAP kinases and mTOR mediate insulin-induced phosphorylation of insulin receptor substrate-1 on serine residues 307, 612 and 632." Diabetologia 46(11): 1532-42. Gual, P., Y. Le Marchand-Brustel, et al. (2005). "Positive and negative regulation of insulin signaling through IRS-1 phosphorylation." Biochimie 87(1): 99-109. Guan, K. L., C. Figueroa, et al. (2000). "Negative regulation of the serine/threonine kinase B-Raf by Akt." J Biol Chem 275(35): 27354-9. Guilherme, A., M. Emoto, et al. (2000). "Perinuclear localization and insulin responsiveness of GLUT4 requires cytoskeletal integrity in 3T3-L1 adipocytes." J Biol Chem 275(49): 38151-9. Guo, X., K. Xu, et al. (2010). "Involvement of inducible 6-phosphofructo-2-kinase in the anti-diabetic effect of peroxisome proliferator-activated receptor gamma activation in mice." J Biol Chem 285(31): 23711-20. Haglund, K., P. P. Di Fiore, et al. (2003). "Distinct monoubiquitin signals in receptor endocytosis." Trends Biochem Sci 28(11): 598-603. Hardie, D. G. (2004). "The AMP-activated protein kinase pathway--new players upstream and downstream." J Cell Sci 117(Pt 23): 5479-87. Hardie, D. G., D. Carling, et al. (1998). "The AMP-activated/SNF1 protein kinase subfamily: metabolic sensors of the eukaryotic cell?" Annu Rev Biochem 67: 821-55. Harris, T. E., A. Chi, et al. (2006). "mTOR-dependent stimulation of the association of eIF4G and eIF3 by insulin." Embo J 25(8): 1659-68.

146 Hauner, H. (2002). "The mode of action of thiazolidinediones." Diabetes Metab Res Rev 18 Suppl 2: S10-5. Hawley, S. A., J. Boudeau, et al. (2003). "Complexes between the LKB1 tumor suppressor, STRAD alpha/beta and MO25 alpha/beta are upstream kinases in the AMP-activated protein kinase cascade." J Biol 2(4): 28. Hayashi, T., J. F. Wojtaszewski, et al. (1997). "Exercise regulation of glucose transport in skeletal muscle." Am J Physiol 273(6 Pt 1): E1039-51. He, W., Y. Barak, et al. (2003). "Adipose-specific peroxisome proliferator-activated receptor gamma knockout causes insulin resistance in fat and liver but not in muscle." Proc Natl Acad Sci U S A 100(26): 15712-7. Herman, M. A. and B. B. Kahn (2006). "Glucose transport and sensing in the maintenance of glucose homeostasis and metabolic harmony." J Clin Invest 116(7): 1767-75. Hernandez, R., T. Teruel, et al. (2001). "Akt mediates insulin induction of glucose uptake and up-regulation of GLUT4 gene expression in brown adipocytes." FEBS Lett 494(3): 225-31. Hershko, A. (2005). "The ubiquitin system for protein degradation and some of its roles in the control of the cell division cycle." Cell Death Differ 12(9): 1191-7. Hevener, A. L., W. He, et al. (2003). "Muscle-specific Pparg deletion causes insulin resistance." Nat Med 9(12): 1491-7. Heyninck, K. and R. Beyaert (2005). "A20 inhibits NF-kappaB activation by dual ubiquitin-editing functions." Trends Biochem Sci 30(1): 1-4. Hill, E. V., C. A. Hudson, et al. (2010). "Regulation of PIKfyve phosphorylation by insulin and osmotic stress." Biochem Biophys Res Commun 397(4): 650-5. Hinnebusch, A. G. (2006). "eIF3: a versatile scaffold for translation initiation complexes." Trends Biochem Sci 31(10): 553-62. Hiratani, K., T. Haruta, et al. (2005). "Roles of mTOR and JNK in serine phosphorylation, translocation, and degradation of IRS-1." Biochem Biophys Res Commun 335(3): 836-42. Hirosumi, J., G. Tuncman, et al. (2002). "A central role for JNK in obesity and insulin resistance." Nature 420(6913): 333-6. Hoare, S., K. Hoare, et al. (2008). "Tnk1/Kos1 knockout mice develop spontaneous tumors." Cancer Res 68(21): 8723-32.

147 Hoehn, K. L., C. Hohnen-Behrens, et al. (2008). "IRS1-independent defects define major nodes of insulin resistance." Cell Metab 7(5): 421-33. Hoehn, K. L., A. B. Salmon, et al. (2009). "Insulin resistance is a cellular antioxidant defense mechanism." Proc Natl Acad Sci U S A 106(42): 17787-92. Holloszy, J. O., S. H. Constable, et al. (1986). "Activation of glucose transport in muscle by exercise." Diabetes Metab Rev 1(4): 409-23. Holloszy, J. O. and P. A. Hansen (1996). "Regulation of glucose transport into skeletal muscle." Rev Physiol Biochem Pharmacol 128: 99-193. Holman, G. D. and S. W. Cushman (1994). "Subcellular localization and trafficking of the GLUT4 glucose transporter isoform in insulin-responsive cells." Bioessays 16(10): 753-9. Hong, S. P., F. C. Leiper, et al. (2003). "Activation of yeast Snf1 and mammalian AMP- activated protein kinase by upstream kinases." Proc Natl Acad Sci U S A 100(15): 8839-43. Horike, N., H. Takemori, et al. (2003). "Adipose-specific expression, phosphorylation of Ser794 in insulin receptor substrate-1, and activation in diabetic animals of salt-inducible kinase-2." J Biol Chem 278(20): 18440-7. Hossain, P., B. Kawar, et al. (2007). "Obesity and diabetes in the developing world--a growing challenge." N Engl J Med 356(3): 213-5. Houstis, N., E. D. Rosen, et al. (2006). "Reactive oxygen species have a causal role in multiple forms of insulin resistance." Nature 440(7086): 944-8. Hsu, C., Y. Morohashi, et al. (2010). "Regulation of exosome secretion by Rab35 and its GTPase-activating proteins TBC1D10A-C." J Cell Biol 189(2): 223-32. Huang, D., M. Khoe, et al. (2006). "Reduced expression of focal adhesion kinase disrupts insulin action in skeletal muscle cells." Endocrinology 147(7): 3333-43. Huang, T. T. and A. D. D'Andrea (2006). "Regulation of DNA repair by ubiquitylation." Nat Rev Mol Cell Biol 7(5): 323-34. Hubbard, S. R. (1997). "Crystal structure of the activated insulin in complex with peptide substrate and ATP analog." Embo J 16(18): 5572-81. Huo, Y., X. Guo, et al. (2010). "Disruption of inducible 6-phosphofructo-2-kinase ameliorates diet-induced adiposity but exacerbates systemic insulin resistance and adipose tissue inflammatory response." J Biol Chem 285(6): 3713-21.

148 Ikonomov, O. C., D. Sbrissa, et al. (2007). "ArPIKfyve-PIKfyve interaction and role in insulin-regulated GLUT4 translocation and glucose transport in 3T3-L1 adipocytes." Exp Cell Res 313(11): 2404-16. Ikonomov, O. C., D. Sbrissa, et al. (2002). "Requirement for PIKfyve enzymatic activity in acute and long-term insulin cellular effects." Endocrinology 143(12): 4742-54. Isakoff, S. J., C. Taha, et al. (1995). "The inability of phosphatidylinositol 3-kinase activation to stimulate GLUT4 translocation indicates additional signaling pathways are required for insulin-stimulated glucose uptake." Proc Natl Acad Sci U S A 92(22): 10247-51. Itani, S. I., N. B. Ruderman, et al. (2002). "Lipid-induced insulin resistance in human muscle is associated with changes in diacylglycerol, protein kinase C, and IkappaB-alpha." Diabetes 51(7): 2005-11. Itani, S. I., Q. Zhou, et al. (2000). "Involvement of protein kinase C in human skeletal muscle insulin resistance and obesity." Diabetes 49(8): 1353-8. Jackowski, S. and C. O. Rock (1981). "Regulation of coenzyme A biosynthesis." J Bacteriol 148(3): 926-32. Jackson, A. L., J. Burchard, et al. (2006). "Position-specific chemical modification of siRNAs reduces "off-target" transcript silencing." Rna 12(7): 1197-205. Jackson, A. L. and P. S. Linsley (2004). "Noise amidst the silence: off-target effects of siRNAs?" Trends Genet 20(11): 521-4. Jackson, A. L. and P. S. Linsley (2010). "Recognizing and avoiding siRNA off-target effects for target identification and therapeutic application." Nat Rev Drug Discov 9(1): 57-67. Jaffe, A. B., P. Aspenstrom, et al. (2004). "Human CNK1 acts as a scaffold protein, linking Rho and Ras signal transduction pathways." Mol Cell Biol 24(4): 1736- 46. Jaffe, A. B., A. Hall, et al. (2005). "Association of CNK1 with Rho guanine nucleotide exchange factors controls signaling specificity downstream of Rho." Curr Biol 15(5): 405-12. Jager, J., T. Gremeaux, et al. (2007). "Interleukin-1beta-induced insulin resistance in adipocytes through down-regulation of insulin receptor substrate-1 expression." Endocrinology 148(1): 241-51.

149 James, D. E., R. Brown, et al. (1988). "Insulin-regulatable tissues express a unique insulin-sensitive glucose transport protein." Nature 333(6169): 183-5. James, D. E. and R. C. Piper (1994). "Insulin resistance, diabetes, and the insulin- regulated trafficking of GLUT-4." J Cell Biol 126(5): 1123-1126. Jensen, T. E., P. Schjerling, et al. (2008). "AMPK alpha1 activation is required for stimulation of glucose uptake by twitch contraction, but not by H2O2, in mouse skeletal muscle." PLoS One 3(5): e2102. Jensen, T. E., J. F. Wojtaszewski, et al. (2009). "AMP-activated protein kinase in contraction regulation of skeletal muscle metabolism: necessary and/or sufficient?" Acta Physiol (Oxf) 196(1): 155-74. Jessen, N. and L. J. Goodyear (2005). "Contraction signaling to glucose transport in skeletal muscle." J Appl Physiol 99(1): 330-7. Jiang, Z. Y., Q. L. Zhou, et al. (2003). "Insulin signaling through Akt/protein kinase B analyzed by small interfering RNA-mediated gene silencing." Proceedings of the National Academy of Sciences of the United States of America 100(13): 7569-7574. Joost, H. G., G. I. Bell, et al. (2002). "Nomenclature of the GLUT/SLC2A family of sugar/polyol transport facilitators." Am J Physiol Endocrinol Metab 282(4): E974-6. Jorgensen, S. B., B. Viollet, et al. (2004). "Knockout of the alpha2 but not alpha1 5'- AMP-activated protein kinase isoform abolishes 5-aminoimidazole-4- carboxamide-1-beta-4-ribofuranosidebut not contraction-induced glucose uptake in skeletal muscle." J Biol Chem 279(2): 1070-9. Kahn, C. R. and M. F. White (1988). "The insulin receptor and the molecular mechanism of insulin action." J Clin Invest 82(4): 1151-6. Kane, S., H. Sano, et al. (2002). "A method to identify serine kinase substrates. Akt phosphorylates a novel adipocyte protein with a Rab GTPase-activating protein (GAP) domain." J Biol Chem 277(25): 22115-8. Kanzaki, M., M. Furukawa, et al. (2004). "Phosphatidylinositol 4,5-bisphosphate regulates adipocyte actin dynamics and GLUT4 vesicle recycling." J Biol Chem 279(29): 30622-33. Kanzaki, M., S. Mora, et al. (2004). "Atypical protein kinase C (PKCzeta/lambda) is a convergent downstream target of the insulin-stimulated phosphatidylinositol 3- kinase and TC10 signaling pathways." J Cell Biol 164(2): 279-90.

150 Kanzaki, M., R. T. Watson, et al. (2002). "Small GTP-binding protein TC10 differentially regulates two distinct populations of filamentous actin in 3T3L1 adipocytes." Mol Biol Cell 13(7): 2334-46. Karin, M. (1999). "The beginning of the end: IkappaB kinase (IKK) and NF-kappaB activation." J Biol Chem 274(39): 27339-42. Karlsson, H. K., A. V. Chibalin, et al. (2009). "Kinetics of GLUT4 trafficking in rat and human skeletal muscle." Diabetes 58(4): 847-54. Kasuga, M., J. A. Hedo, et al. (1982). "The structure of insulin receptor and its subunits. Evidence for multiple nonreduced forms and a 210,000 possible proreceptor." J Biol Chem 257(17): 10392-9. Kasuga, M., Y. Zick, et al. (1982). "Insulin stimulation of phosphorylation of the beta subunit of the insulin receptor. Formation of both phosphoserine and phosphotyrosine." J Biol Chem 257(17): 9891-4. Katagiri, H., T. Asano, et al. (1996). "Overexpression of catalytic subunit p110alpha of phosphatidylinositol 3-kinase increases glucose transport activity with translocation of glucose transporters in 3T3-L1 adipocytes." J Biol Chem 271(29): 16987-90. Katome, T., T. Obata, et al. (2003). "Use of RNA interference-mediated gene silencing and adenoviral overexpression to elucidate the roles of AKT/protein kinase B isoforms in insulin actions." J Biol Chem 278(30): 28312-23. Kemp, B. E., K. I. Mitchelhill, et al. (1999). "Dealing with energy demand: the AMP- activated protein kinase." Trends Biochem Sci 24(1): 22-5. Kerscher, O., R. Felberbaum, et al. (2006). "Modification of proteins by ubiquitin and ubiquitin-like proteins." Annu Rev Cell Dev Biol 22: 159-80. Kim, J. A., Y. Wei, et al. (2008). "Role of mitochondrial dysfunction in insulin resistance." Circ Res 102(4): 401-14. Kim, J. J. and D. D. Sears (2010). "TLR4 and Insulin Resistance." Gastroenterol Res Pract 2010. Kim, J. K., J. J. Fillmore, et al. (2004). "PKC-theta knockout mice are protected from fat-induced insulin resistance." J Clin Invest 114(6): 823-7. Kim, M. S., S. Ramakrishna, et al. (2010). "Protein stability of mitochondrial superoxide dismutase SOD2 is regulated by USP36." J Cell Biochem.

151 Kim, T., J. Wayne Leitner, et al. (2009). "Knockdown of JNK rescues 3T3-L1 adipocytes from insulin resistance induced by mitochondrial dysfunction." Biochem Biophys Res Commun 378(4): 772-6. Kimura, A., C. A. Baumann, et al. (2001). "The sorbin homology domain: a motif for the targeting of proteins to lipid rafts." Proc Natl Acad Sci U S A 98(16): 9098- 103. Kitano-Takahashi, M., H. Morita, et al. (2007). "Expression, purification and crystallization of a human tau-tubulin kinase 2 that phosphorylates tau protein." Acta Crystallogr Sect F Struct Biol Cryst Commun 63(Pt 7): 602-4. Klaman, L. D., O. Boss, et al. (2000). "Increased energy expenditure, decreased adiposity, and tissue-specific insulin sensitivity in protein-tyrosine phosphatase 1B-deficient mice." Mol Cell Biol 20(15): 5479-89. Klip, A. and M. R. Paquet (1990). "Glucose transport and glucose transporters in muscle and their metabolic regulation." Diabetes Care 13(3): 228-43. Knight, Z. A., B. Gonzalez, et al. (2006). "A pharmacological map of the PI3-K family defines a role for p110alpha in insulin signaling." Cell 125(4): 733-47. Kohn, A. D., A. Barthel, et al. (1998). "Construction and characterization of a conditionally active version of the serine/threonine kinase Akt." J Biol Chem 273(19): 11937-43. Komander, D., M. J. Clague, et al. (2009). "Breaking the chains: structure and function of the deubiquitinases." Nat Rev Mol Cell Biol 10(8): 550-63. Kotani, K., A. J. Carozzi, et al. (1995). "Requirement for phosphoinositide 3-kinase in insulin-stimulated GLUT4 translocation in 3T3-L1 adipocytes." Biochem Biophys Res Commun 209(1): 343-8. Kotani, K., W. Ogawa, et al. (1998). "Requirement of atypical protein kinase clambda for insulin stimulation of glucose uptake but not for Akt activation in 3T3-L1 adipocytes." Mol Cell Biol 18(12): 6971-82. Kovalenko, A., C. Chable-Bessia, et al. (2003). "The tumour suppressor CYLD negatively regulates NF-kappaB signalling by deubiquitination." Nature 424(6950): 801-5. Kramer, H. F., E. B. Taylor, et al. (2007). "Calmodulin-binding domain of AS160 regulates contraction- but not insulin-stimulated glucose uptake in skeletal muscle." Diabetes 56(12): 2854-62.

152 Kramer, H. F., C. A. Witczak, et al. (2006). "Distinct signals regulate AS160 phosphorylation in response to insulin, AICAR, and contraction in mouse skeletal muscle." Diabetes 55(7): 2067-76. Kroll, M., F. Margottin, et al. (1999). "Inducible degradation of IkappaBalpha by the proteasome requires interaction with the F-box protein h-betaTrCP." J Biol Chem 274(12): 7941-5. Kumar, A., J. C. Lawrence, Jr., et al. (2010). "Fat cell-specific ablation of rictor in mice impairs insulin-regulated fat cell and whole-body glucose and lipid metabolism." Diabetes 59(6): 1397-406. Kurlawalla-Martinez, C., B. Stiles, et al. (2005). "Insulin hypersensitivity and resistance to streptozotocin-induced diabetes in mice lacking PTEN in adipose tissue." Mol Cell Biol 25(6): 2498-510. Kusunoki, M., L. H. Storlien, et al. (1993). "Muscle glucose uptake during and after exercise is normal in insulin-resistant rats." Am J Physiol 264(2 Pt 1): E167-72. Lagathu, C., L. Yvan-Charvet, et al. (2006). "Long-term treatment with interleukin- 1beta induces insulin resistance in murine and human adipocytes." Diabetologia 49(9): 2162-73. Lalioti, V., G. Muruais, et al. (2009). "The atypical kinase Cdk5 is activated by insulin, regulates the association between GLUT4 and E-Syt1, and modulates glucose transport in 3T3-L1 adipocytes." Proc Natl Acad Sci U S A 106(11): 4249-53. Lalioti, V. S., S. Vergarajauregui, et al. (2002). "The insulin-sensitive glucose transporter, GLUT4, interacts physically with Daxx. Two proteins with capacity to bind Ubc9 and conjugated to SUMO1." J Biol Chem 277(22): 19783-91. Lamb, C. A., R. K. McCann, et al. (2010). "Insulin-regulated trafficking of GLUT4 requires ubiquitination." Traffic 11(11): 1445-54. Larance, M., G. Ramm, et al. (2005). "Characterization of the role of the Rab GTPase- activating protein AS160 in insulin-regulated GLUT4 trafficking." J Biol Chem 280(45): 37803-13. Lawlor, M. A. and D. R. Alessi (2001). "PKB/Akt: a key mediator of cell proliferation, survival and insulin responses?" J Cell Sci 114(Pt 16): 2903-10. Le Bacquer, O., E. Petroulakis, et al. (2007). "Elevated sensitivity to diet-induced obesity and insulin resistance in mice lacking 4E-BP1 and 4E-BP2." J Clin Invest 117(2): 387-96.

153 Le Roith, D. and Y. Zick (2001). "Recent advances in our understanding of insulin action and insulin resistance." Diabetes Care 24(3): 588-97. Lee, J. O., S. K. Lee, et al. (2011). "Metformin induces Rab4 through AMPK and modulates GLUT4 translocation in skeletal muscle cells." J Cell Physiol 226(4): 974-81. Lee, S. J., P. T. Pfluger, et al. (2010). "A functional role for the p62-ERK1 axis in the control of energy homeostasis and adipogenesis." EMBO Rep 11(3): 226-32. Lee, Y. H., J. Giraud, et al. (2003). "c-Jun N-terminal kinase (JNK) mediates feedback inhibition of the insulin signaling cascade." J Biol Chem 278(5): 2896-902. Lefort, N., E. St-Amand, et al. (2008). "The alpha-subunit of AMPK is essential for submaximal contraction-mediated glucose transport in skeletal muscle in vitro." Am J Physiol Endocrinol Metab 295(6): E1447-54. Leng, S., W. Zhang, et al. (2010). "Glycogen synthase kinase 3 beta mediates high glucose-induced ubiquitination and proteasome degradation of insulin receptor substrate 1." J Endocrinol 206(2): 171-81. Li, J., K. DeFea, et al. (1999). "Modulation of insulin receptor substrate-1 tyrosine phosphorylation by an Akt/phosphatidylinositol 3-kinase pathway." J Biol Chem 274(14): 9351-6. Li, J., K. L. Houseknecht, et al. (2000). "Reduced glucose uptake precedes insulin signaling defects in adipocytes from heterozygous GLUT4 knockout mice." FASEB J 14(9): 1117-25. Li, W., N. Kedersha, et al. (2004). "FAST is a BCL-X(L)-associated mitochondrial protein." Biochem Biophys Res Commun 318(1): 95-102. Li, Y., Y. Chang, et al. (2005). "High glucose upregulates pantothenate kinase 4 (PanK4) and thus affects M2-type pyruvate kinase (Pkm2)." Mol Cell Biochem 277(1-2): 117-25. Li, Y., T. J. Soos, et al. (2004). "Protein kinase C Theta inhibits insulin signaling by phosphorylating IRS1 at Ser(1101)." J Biol Chem 279(44): 45304-7. Li, Y., H. Wang, et al. (2003). "Positive and negative regulation of APP amyloidogenesis by sumoylation." Proc Natl Acad Sci U S A 100(1): 259-64. Liberman, Z. and H. Eldar-Finkelman (2005). "Serine 332 phosphorylation of insulin receptor substrate-1 by glycogen synthase kinase-3 attenuates insulin signaling." J Biol Chem 280(6): 4422-8.

154 Lim, J., M. Zhou, et al. (2010). "The CNK1 scaffold binds cytohesins and promotes insulin pathway signaling." Genes Dev 24(14): 1496-506. Liu, F., Q. Dallas-yang, et al. (2009). "Development of a novel GLUT4 translocation assay for identifying potential novel therapeutic targets for insulin sensitization." Biochem J 418(2): 413-420. Liu, L. B., W. Omata, et al. (2007). "The SUMO conjugating enzyme Ubc9 is a regulator of GLUT4 turnover and targeting to the insulin-responsive storage compartment in 3T3-L1 adipocytes." Diabetes 56(8): 1977-85. Liu, L. Z., S. C. Cheung, et al. (2010). "The pivotal role of protein kinase C zeta (PKCzeta) in insulin- and AMP-activated protein kinase (AMPK)-mediated glucose uptake in muscle cells." Cell Signal 22(10): 1513-22. Liu, Y. F., K. Paz, et al. (2001). "Insulin stimulates PKCzeta -mediated phosphorylation of insulin receptor substrate-1 (IRS-1). A self-attenuated mechanism to negatively regulate the function of IRS proteins." J Biol Chem 276(17): 14459- 65. Liu, Z. and W. Cao (2009). "p38 mitogen-activated protein kinase: a critical node linking insulin resistance and cardiovascular diseases in type 2 diabetes mellitus." Endocr Metab Immune Disord Drug Targets 9(1): 38-46. Lizcano, J. M., O. Goransson, et al. (2004). "LKB1 is a master kinase that activates 13 kinases of the AMPK subfamily, including MARK/PAR-1." Embo J 23(4): 833- 43. Lo, Y. T., C. J. Tsao, et al. (2004). "Increase of PTEN gene expression in insulin resistance." Horm Metab Res 36(10): 662-6. Lodeiro, M., M. Theodoropoulou, et al. (2009). "c-Src regulates Akt signaling in response to ghrelin via beta-arrestin signaling-independent and -dependent mechanisms." PLoS One 4(3): e4686. Lu, Y., O. A. Adegoke, et al. (2009). "USP19 deubiquitinating enzyme supports cell proliferation by stabilizing KPC1, a ubiquitin ligase for p27Kip1." Mol Cell Biol 29(2): 547-58. Lund, S., G. D. Holman, et al. (1995). "Contraction stimulates translocation of glucose transporter GLUT4 in skeletal muscle through a mechanism distinct from that of insulin." Proc Natl Acad Sci U S A 92(13): 5817-21. MacAulay, K., B. W. Doble, et al. (2007). "Glycogen synthase kinase 3alpha-specific regulation of murine hepatic glycogen metabolism." Cell Metab 6(4): 329-37.

155 MaCaulay, S. L., V. Stoichevska, et al. (2003). "Insulin stimulates movement of sorting nexin 9 between cellular compartments: a putative role mediating expression and insulin action." Biochem J 376(Pt 1): 123-34. Maegawa, H., R. Ide, et al. (1995). "Thiazolidine derivatives ameliorate high glucose- induced insulin resistance via the normalization of protein-tyrosine phosphatase activities." J Biol Chem 270(13): 7724-30. Maehama, T. and J. E. Dixon (1998). "The tumor suppressor, PTEN/MMAC1, dephosphorylates the lipid second messenger, phosphatidylinositol 3,4,5- trisphosphate." J Biol Chem 273(22): 13375-8. Mandl, M., D. N. Slack, et al. (2005). "Specific inactivation and nuclear anchoring of extracellular signal-regulated kinase 2 by the inducible dual-specificity protein phosphatase DUSP5." Mol Cell Biol 25(5): 1830-45. Manning, B. D. and L. C. Cantley (2007). "AKT/PKB signaling: navigating downstream." Cell 129(7): 1261-74. Manning, G., D. B. Whyte, et al. (2002). "The protein kinase complement of the ." Science 298(5600): 1912-34. Martin, S., B. Reaves, et al. (1994). "Analysis of the co-localization of the insulin- responsive glucose transporter (GLUT4) and the trans Golgi network marker TGN38 within 3T3-L1 adipocytes." Biochem J 300 ( Pt 3): 743-9. Martin, S. E. and N. J. Caplen (2007). "Applications of RNA interference in mammalian systems." Annu Rev Genomics Hum Genet 8: 81-108. Matic, I., M. van Hagen, et al. (2008). "In vivo identification of human small ubiquitin- like modifier polymerization sites by high accuracy mass spectrometry and an in vitro to in vivo strategy." Mol Cell Proteomics 7(1): 132-44. Matsusue, K., M. Haluzik, et al. (2003). "Liver-specific disruption of PPARgamma in leptin-deficient mice improves fatty liver but aggravates diabetic phenotypes." J Clin Invest 111(5): 737-47. McManus, E. J., K. Sakamoto, et al. (2005). "Role that phosphorylation of GSK3 plays in insulin and Wnt signalling defined by knockin analysis." EMBO J 24(8): 1571-83. Mehta, A. and S. Orchard (2009). "Nucleoside diphosphate kinase (NDPK, NM23, AWD): recent regulatory advances in endocytosis, metastasis, psoriasis, insulin release, fetal erythroid lineage and heart failure; translational medicine exemplified." Mol Cell Biochem 329(1-2): 3-15.

156 Messick, T. E., N. S. Russell, et al. (2008). "Structural basis for ubiquitin recognition by the Otu1 ovarian tumor domain protein." J Biol Chem 283(16): 11038-49. Miinea, C. P., H. Sano, et al. (2005). "AS160, the Akt substrate regulating GLUT4 translocation, has a functional Rab GTPase-activating protein domain." Biochem J 391(Pt 1): 87-93. Minami, A., M. Iseki, et al. (2003). "Increased insulin sensitivity and hypoinsulinemia in APS knockout mice." Diabetes 52(11): 2657-65. Mitra, P., X. Zheng, et al. (2004). "RNAi-based analysis of CAP, Cbl, and CrkII function in the regulation of GLUT4 by insulin." J Biol Chem 279(36): 37431-5. Miura, A., M. P. Sajan, et al. (2004). "Insulin substrates 1 and 2 are corequired for activation of atypical protein kinase C and Cbl-dependent phosphatidylinositol 3-kinase during insulin action in immortalized brown adipocytes." Biochemistry 43(49): 15503-9. Miyazaki, Y., A. Mahankali, et al. (2001). "Improved glycemic control and enhanced insulin sensitivity in type 2 diabetic subjects treated with pioglitazone." Diabetes Care 24(4): 710-9. Moeschel, K., A. Beck, et al. (2004). "Protein kinase C-zeta-induced phosphorylation of Ser318 in insulin receptor substrate-1 (IRS-1) attenuates the interaction with the insulin receptor and the tyrosine phosphorylation of IRS-1." J Biol Chem 279(24): 25157-63. Mohr, S., C. Bakal, et al. (2010). "Genomic screening with RNAi: results and challenges." Annu Rev Biochem 79: 37-64. Molero, J. C., T. E. Jensen, et al. (2004). "c-Cbl-deficient mice have reduced adiposity, higher energy expenditure, and improved peripheral insulin action." J Clin Invest 114(9): 1326-33. Moloney, A. M., R. J. Griffin, et al. (2008). "Defects in IGF-1 receptor, insulin receptor and IRS-1/2 in Alzheimer's disease indicate possible resistance to IGF-1 and insulin signalling." Neurobiol Aging 31(2): 224-43. Mooney, R. A., J. Senn, et al. (2001). "Suppressors of cytokine signaling-1 and -6 associate with and inhibit the insulin receptor. A potential mechanism for cytokine-mediated insulin resistance." J Biol Chem 276(28): 25889-93. Morel, C., C. L. Standen, et al. (2010). "Requirement of JIP1-mediated c-Jun N- terminal kinase activation for obesity-induced insulin resistance." Mol Cell Biol 30(19): 4616-25.

157 Mu, J., J. T. Brozinick, Jr., et al. (2001). "A role for AMP-activated protein kinase in contraction- and hypoxia-regulated glucose transport in skeletal muscle." Mol Cell 7(5): 1085-94. Mukherjee, R., P. J. Davies, et al. (1997). "Sensitization of diabetic and obese mice to insulin by retinoid X receptor agonists." Nature 386(6623): 407-10. Mukhopadhyay, D. and H. Riezman (2007). "Proteasome-independent functions of ubiquitin in endocytosis and signaling." Science 315(5809): 201-5. Murata-Hori, M., F. Suizu, et al. (1999). "ZIP kinase identified as a novel myosin regulatory light chain kinase in HeLa cells." FEBS Lett 451(1): 81-4. Myers, M. G., Jr., J. M. Backer, et al. (1992). "IRS-1 activates phosphatidylinositol 3'- kinase by associating with src homology 2 domains of p85." Proc Natl Acad Sci U S A 89(21): 10350-4. Myers, M. G., Jr. and M. F. White (1993). "The new elements of insulin signaling. Insulin receptor substrate-1 and proteins with SH2 domains." Diabetes 42(5): 643-50. Nakamori, Y., M. Emoto, et al. (2006). "Myosin motor Myo1c and its receptor NEMO/IKK-gamma promote TNF-alpha-induced serine307 phosphorylation of IRS-1." J Cell Biol 173(5): 665-71. Nakashima, N., P. M. Sharma, et al. (2000). "The tumor suppressor PTEN negatively regulates insulin signaling in 3T3-L1 adipocytes." J Biol Chem 275(17): 12889- 95. Nazari, H., A. Takahashi, et al. (2007). "Angiotensin II inhibits insulin-induced actin stress fiber formation and glucose uptake via ERK1/2." J Med Invest 54(1-2): 19-27. Nesher, R., I. E. Karl, et al. (1985). "Dissociation of effects of insulin and contraction on glucose transport in rat epitrochlearis muscle." Am J Physiol 249(3 Pt 1): C226-32. Ng, Y., G. Ramm, et al. (2008). "Rapid activation of Akt2 is sufficient to stimulate GLUT4 translocation in 3T3-L1 adipocytes." Cell Metab 7(4): 348-56. Nijman, S. M., M. P. Luna-Vargas, et al. (2005). "A genomic and functional inventory of deubiquitinating enzymes." Cell 123(5): 773-86. Nikoulina, S. E., T. P. Ciaraldi, et al. (2002). "Inhibition of glycogen synthase kinase 3 improves insulin action and glucose metabolism in human skeletal muscle." Diabetes 51(7): 2190-8.

158 Niu, W., P. J. Bilan, et al. (2011). "PKCepsilon regulates contraction-stimulated GLUT4 traffic in skeletal muscle cells." J Cell Physiol 226(1): 173-80. Norris, A. W., L. Chen, et al. (2003). "Muscle-specific PPARgamma-deficient mice develop increased adiposity and insulin resistance but respond to thiazolidinediones." J Clin Invest 112(4): 608-18. Nosaka, K., M. Onozuka, et al. (2001). "Isolation and characterization of a human thiamine pyrophosphokinase cDNA." Biochim Biophys Acta 1517(2): 293-7. Ojuka, E. O. (2004). "Role of calcium and AMP kinase in the regulation of mitochondrial biogenesis and GLUT4 levels in muscle." Proc Nutr Soc 63(2): 275-8. Okada, S., E. Yamada, et al. (2008). "CDK5-dependent phosphorylation of the Rho family GTPase TC10(alpha) regulates insulin-stimulated GLUT4 translocation." J Biol Chem 283(51): 35455-63. Olson, A. L. and J. E. Pessin (1996). "Structure, function, and regulation of the mammalian facilitative glucose transporter gene family." Annu Rev Nutr 16: 235-56. Ono, H., H. Katagiri, et al. (2001). "Regulation of phosphoinositide metabolism, Akt phosphorylation, and glucose transport by PTEN (phosphatase and tensin homolog deleted on chromosome 10) in 3T3-L1 adipocytes." Mol Endocrinol 15(8): 1411-22. Pais, S. M., M. T. Tellez-Inon, et al. (2009). "Serine/threonine protein phosphatases type 2A and their roles in stress signaling." Plant Signal Behav 4(11): 1013-5. Pao, S. S., I. T. Paulsen, et al. (1998). "Major facilitator superfamily." Microbiol Mol Biol Rev 62(1): 1-34. Patki, V., J. Virbasius, et al. (1997). "Identification of an early endosomal protein regulated by phosphatidylinositol 3-kinase." Proc Natl Acad Sci U S A 94(14): 7326-30. Peck, G. R., S. Ye, et al. (2006). "Interaction of the Akt substrate, AS160, with the glucose transporter 4 vesicle marker protein, insulin-regulated aminopeptidase." Mol Endocrinol 20(10): 2576-83. Peng, J., D. Schwartz, et al. (2003). "A proteomics approach to understanding protein ubiquitination." Nat Biotechnol 21(8): 921-6. Perrini, S., J. Henriksson, et al. (2004). "Exercise-induced protein kinase C isoform- specific activation in human skeletal muscle." Diabetes 53(1): 21-4.

159 Persengiev, S. P., X. Zhu, et al. (2004). "Nonspecific, concentration-dependent stimulation and repression of mammalian gene expression by small interfering RNAs (siRNAs)." Rna 10(1): 12-8. Pickart, C. M. and D. Fushman (2004). "Polyubiquitin chains: polymeric protein signals." Curr Opin Chem Biol 8(6): 610-6. Piper, R. C., L. J. Hess, et al. (1991). "Differential sorting of two glucose transporters expressed in insulin-sensitive cells." Am J Physiol 260(3 Pt 1): C570-80. Ploug, T., B. van Deurs, et al. (1998). "Analysis of GLUT4 distribution in whole skeletal muscle fibers: identification of distinct storage compartments that are recruited by insulin and muscle contractions." J Cell Biol 142(6): 1429-46. Qiao, L. Y., J. L. Goldberg, et al. (1999). "Identification of enhanced serine kinase activity in insulin resistance." J Biol Chem 274(15): 10625-32. Quesada, V., A. Diaz-Perales, et al. (2004). "Cloning and enzymatic analysis of 22 novel human ubiquitin-specific proteases." Biochem Biophys Res Commun 314(1): 54-62. Radziwill, G., R. A. Erdmann, et al. (2003). "The Bcr kinase downregulates Ras signaling by phosphorylating AF-6 and binding to its PDZ domain." Mol Cell Biol 23(13): 4663-72. Rajkhowa, M., S. Brett, et al. (2009). "Insulin resistance in polycystic ovary syndrome is associated with defective regulation of ERK1/2 by insulin in skeletal muscle in vivo." Biochem J 418(3): 665-71. Rameh, L. E. and L. C. Cantley (1999). "The role of phosphoinositide 3-kinase lipid products in cell function." J Biol Chem 274(13): 8347-50. Ramlo-Halsted, B. A. and S. V. Edelman (1999). "The natural history of type 2 diabetes. Implications for clinical practice." Prim Care 26(4): 771-89. Ramm, G., M. Larance, et al. (2006). "A role for 14-3-3 in insulin-stimulated GLUT4 translocation through its interaction with the RabGAP AS160." J Biol Chem 281(39): 29174-80. Rand, T. A., K. Ginalski, et al. (2004). "Biochemical identification of Argonaute 2 as the sole protein required for RNA-induced silencing complex activity." Proc Natl Acad Sci U S A 101(40): 14385-9. Ravichandran, L. V., D. L. Esposito, et al. (2001). "Protein kinase C-zeta phosphorylates insulin receptor substrate-1 and impairs its ability to activate

160 phosphatidylinositol 3-kinase in response to insulin." J Biol Chem 276(5): 3543- 9. Reyes-Turcu, F. E., K. H. Ventii, et al. (2009). "Regulation and cellular roles of ubiquitin-specific deubiquitinating enzymes." Annu Rev Biochem 78: 363-97. Ribon, V., J. A. Printen, et al. (1998). "A novel, multifuntional c-Cbl binding protein in insulin receptor signaling in 3T3-L1 adipocytes." Mol Cell Biol 18(2): 872-9. Richter, E. A., P. J. Cleland, et al. (1987). "Contraction-associated translocation of protein kinase C in rat skeletal muscle." FEBS Lett 217(2): 232-6. Rieusset, J., K. Bouzakri, et al. (2004). "Suppressor of cytokine signaling 3 expression and insulin resistance in skeletal muscle of obese and type 2 diabetic patients." Diabetes 53(9): 2232-41. Rochat-Steiner, V., K. Becker, et al. (2000). "FIST/HIPK3: a Fas/FADD-interacting serine/threonine kinase that induces FADD phosphorylation and inhibits fas- mediated Jun NH(2)-terminal kinase activation." J Exp Med 192(8): 1165-74. Rock, C. O., R. B. Calder, et al. (2000). "Pantothenate kinase regulation of the intracellular concentration of coenzyme A." J Biol Chem 275(2): 1377-83. Rodriguez, A., A. Duran, et al. (2006). "Mature-onset obesity and insulin resistance in mice deficient in the signaling adapter p62." Cell Metab 3(3): 211-22. Rose, A. J., B. J. Michell, et al. (2004). "Effect of exercise on protein kinase C activity and localization in human skeletal muscle." J Physiol 561(Pt 3): 861-70. Ross, A. H. and A. Gericke (2009). "Phosphorylation keeps PTEN phosphatase closed for business." Proc Natl Acad Sci U S A 106(5): 1297-8. Rossetti, L., A. E. Stenbit, et al. (1997). "Peripheral but not hepatic insulin resistance in mice with one disrupted allele of the glucose transporter type 4 (GLUT4) gene." J Clin Invest 100(7): 1831-9. Roth, R. A. and J. Beaudoin (1987). "Phosphorylation of purified insulin receptor by cAMP kinase." Diabetes 36(1): 123-6. Rui, L. (2007). "A link between protein translation and body weight." J Clin Invest 117(2): 310-3. Rui, L., T. L. Fisher, et al. (2001). "Regulation of insulin/insulin-like growth factor-1 signaling by proteasome-mediated degradation of insulin receptor substrate-2." J Biol Chem 276(43): 40362-7.

161 Rui, L., M. Yuan, et al. (2002). "SOCS-1 and SOCS-3 block insulin signaling by ubiquitin-mediated degradation of IRS1 and IRS2." J Biol Chem 277(44): 42394-8. Saito, M., S. J. Lessard, et al. (2008). "Activation of atypical protein kinase Czeta toward TC10 is regulated by high-fat diet and aerobic exercise in skeletal muscle." Metabolism 57(9): 1173-80. Sajan, M. P., J. Rivas, et al. (2006). "Repletion of atypical protein kinase C following RNA interference-mediated depletion restores insulin-stimulated glucose transport." J Biol Chem 281(25): 17466-73. Salama, N. R. and R. W. Schekman (1995). "The role of coat proteins in the biosynthesis of secretory proteins." Curr Opin Cell Biol 7(4): 536-43. Salmena, L. and P. P. Pandolfi (2007). "Changing venues for tumour suppression: balancing destruction and localization by monoubiquitylation." Nat Rev Cancer 7(6): 409-13. Salt, I. P., J. M. Connell, et al. (2000). "5-aminoimidazole-4-carboxamide ribonucleoside (AICAR) inhibits insulin-stimulated glucose transport in 3T3-L1 adipocytes." Diabetes 49(10): 1649-56. Saltiel, A. R. and C. R. Kahn (2001). "Insulin signalling and the regulation of glucose and lipid metabolism." Nature 414(6865): 799-806. Sanchez, P., G. De Carcer, et al. (1998). "Localization of atypical protein kinase C isoforms into lysosome-targeted endosomes through interaction with p62." Mol Cell Biol 18(5): 3069-80. Sandstrom, M. E., S. J. Zhang, et al. (2006). "Role of reactive oxygen species in contraction-mediated glucose transport in mouse skeletal muscle." J Physiol 575(Pt 1): 251-62. Sano, H., S. Kane, et al. (2003). "Insulin-stimulated phosphorylation of a Rab GTPase- activating protein regulates GLUT4 translocation." J Biol Chem 278(17): 14599-602. Sasaoka, T., T. Wada, et al. (2004). "SH2-containing inositol phosphatase 2 predominantly regulates Akt2, and not Akt1, phosphorylation at the plasma membrane in response to insulin in 3T3-L1 adipocytes." J Biol Chem 279(15): 14835-43.

162 Saxena, S., Z. O. Jonsson, et al. (2003). "Small RNAs with imperfect match to endogenous mRNA repress translation. Implications for off-target activity of small inhibitory RNA in mammalian cells." J Biol Chem 278(45): 44312-9. Sbrissa, D., O. C. Ikonomov, et al. (2007). "Core protein machinery for mammalian phosphatidylinositol 3,5-bisphosphate synthesis and turnover that regulates the progression of endosomal transport. Novel Sac phosphatase joins the ArPIKfyve-PIKfyve complex." J Biol Chem 282(33): 23878-91. Sbrissa, D., O. C. Ikonomov, et al. (1999). "PIKfyve, a mammalian ortholog of yeast Fab1p lipid kinase, synthesizes 5-phosphoinositides. Effect of insulin." J Biol Chem 274(31): 21589-97. Schinner, S., W. A. Scherbaum, et al. (2005). "Molecular mechanisms of insulin resistance." Diabet Med 22(6): 674-82. Schwartzenberg-Bar-Yoseph, F., M. Armoni, et al. (2004). "The tumor suppressor p53 down-regulates glucose transporters GLUT1 and GLUT4 gene expression." Cancer Res 64(7): 2627-33. Seely, B. L., P. A. Staubs, et al. (1996). "Protein tyrosine phosphatase 1B interacts with the activated insulin receptor." Diabetes 45(10): 1379-85. Seger, R. and E. G. Krebs (1995). "The MAPK signaling cascade." Faseb J 9(9): 726- 35. Serra, C., M. Federici, et al. (2003). "Transgenic mice with dominant negative PKC- theta in skeletal muscle: a new model of insulin resistance and obesity." J Cell Physiol 196(1): 89-97. Sethi, J. K. and G. S. Hotamisligil (1999). "The role of TNF alpha in adipocyte metabolism." Semin Cell Dev Biol 10(1): 19-29. Shah, O. J. and T. Hunter (2006). "Turnover of the active fraction of IRS1 involves raptor-mTOR- and S6K1-dependent serine phosphorylation in cell culture models of tuberous sclerosis." Mol Cell Biol 26(17): 6425-34. Shanik, M. H., Y. Xu, et al. (2008). "Insulin resistance and hyperinsulinemia: is hyperinsulinemia the cart or the horse?" Diabetes Care 31 Suppl 2: S262-8. Shaw, R. J., M. Kosmatka, et al. (2004). "The tumor suppressor LKB1 kinase directly activates AMP-activated kinase and regulates apoptosis in response to energy stress." Proc Natl Acad Sci U S A 101(10): 3329-35. Shembade, N., A. Ma, et al. (2010). "Inhibition of NF-kappaB signaling by A20 through disruption of ubiquitin enzyme complexes." Science 327(5969): 1135-9.

163 Shepherd, P. R., L. Gnudi, et al. (1993). "Adipose cell hyperplasia and enhanced glucose disposal in transgenic mice overexpressing GLUT4 selectively in adipose tissue." J Biol Chem 268(30): 22243-6. Shewan, A. M., E. M. van Dam, et al. (2003). "GLUT4 recycles via a trans-Golgi network (TGN) subdomain enriched in Syntaxins 6 and 16 but not TGN38: involvement of an acidic targeting motif." Mol Biol Cell 14(3): 973-86. Shi, H., M. V. Kokoeva, et al. (2006). "TLR4 links innate immunity and fatty acid- induced insulin resistance." J Clin Invest 116(11): 3015-25. Shisheva, A., B. Rusin, et al. (2001). "Localization and insulin-regulated relocation of phosphoinositide 5-kinase PIKfyve in 3T3-L1 adipocytes." J Biol Chem 276(15): 11859-69. Shoelson, S. E., J. Lee, et al. (2006). "Inflammation and insulin resistance." J Clin Invest 116(7): 1793-801. Shoelson, S. E., M. Sivaraja, et al. (1993). "Specific phosphopeptide binding regulates a conformational change in the PI 3-kinase SH2 domain associated with enzyme activation." Embo J 12(2): 795-802. Skaug, B., X. Jiang, et al. (2009). "The role of ubiquitin in NF-kappaB regulatory pathways." Annu Rev Biochem 78: 769-96. Sledz, C. A., M. Holko, et al. (2003). "Activation of the interferon system by short- interfering RNAs." Nat Cell Biol 5(9): 834-9. Sleeman, M. W., K. E. Wortley, et al. (2005). "Absence of the lipid phosphatase SHIP2 confers resistance to dietary obesity." Nat Med 11(2): 199-205. Slot, J. W., H. J. Geuze, et al. (1991). "Translocation of the glucose transporter GLUT4 in cardiac myocytes of the rat." Proc Natl Acad Sci U S A 88(17): 7815-9. Solheim, S. A., E. Petsalaki, et al. (2008). "Interactions between the Fyn SH3-domain and adaptor protein Cbp/PAG derived ligands, effects on kinase activity and affinity." FEBS J 275(19): 4863-74. Spicer, J., S. Rayter, et al. (2003). "Regulation of the Wnt signalling component PAR1A by the Peutz-Jeghers syndrome kinase LKB1." Oncogene 22(30): 4752- 6. Standaert, M. L., G. Bandyopadhyay, et al. (2001). "Insulin and PIP3 activate PKC-zeta by mechanisms that are both dependent and independent of phosphorylation of activation loop (T410) and autophosphorylation (T560) sites." Biochemistry 40(1): 249-55.

164 Standaert, M. L., G. Bandyopadhyay, et al. (1999). "Insulin activates protein kinases C- zeta and C-lambda by an autophosphorylation-dependent mechanism and stimulates their translocation to GLUT4 vesicles and other membrane fractions in rat adipocytes." J Biol Chem 274(36): 25308-16. Standaert, M. L., M. P. Sajan, et al. (2004). "Requirements for pYXXM motifs in Cbl for binding to the p85 subunit of phosphatidylinositol 3-kinase and Crk, and activation of atypical protein kinase C and glucose transport during insulin action in 3T3/L1 adipocytes." Biochemistry 43(49): 15494-502. Stanley, K. K. and K. E. Howell (1993). "TGN38/41: a molecule on the move." Trends Cell Biol 3(8): 252-5. Steffan, J. S., N. Agrawal, et al. (2004). "SUMO modification of Huntingtin and Huntington's disease pathology." Science 304(5667): 100-4. Steimle, P. A., F. K. Fulcher, et al. (2005). "A novel role for myosin II in insulin- stimulated glucose uptake in 3T3-L1 adipocytes." Biochem Biophys Res Commun 331(4): 1560-5. Stein, S. C., A. Woods, et al. (2000). "The regulation of AMP-activated protein kinase by phosphorylation." Biochem J 345 Pt 3: 437-43. Stenbit, A. E., T. S. Tsao, et al. (1997). "GLUT4 heterozygous knockout mice develop muscle insulin resistance and diabetes." Nat Med 3(10): 1096-101. Stockli, J., J. R. Davey, et al. (2008). "Regulation of glucose transporter 4 translocation by the Rab guanosine triphosphatase-activating protein AS160/TBC1D4: role of phosphorylation and membrane association." Mol Endocrinol 22(12): 2703-15. Stretton, C., A. Evans, et al. (2010). "Cellular depletion of atypical PKC{lambda} is associated with enhanced insulin sensitivity and glucose uptake in L6 rat skeletal muscle cells." Am J Physiol Endocrinol Metab 299(3): E402-12. Summers, S. A., E. L. Whiteman, et al. (1999). "Differentiation-dependent suppression of platelet-derived growth factor signaling in cultured adipocytes." J Biol Chem 274(34): 23858-67. Sun, L., L. Deng, et al. (2004). "The TRAF6 ubiquitin ligase and TAK1 kinase mediate IKK activation by BCL10 and MALT1 in T lymphocytes." Mol Cell 14(3): 289- 301. Sun, X. J., D. L. Crimmins, et al. (1993). "Pleiotropic insulin signals are engaged by multisite phosphorylation of IRS-1." Mol Cell Biol 13(12): 7418-28.

165 Sun, X. J., J. L. Goldberg, et al. (1999). "Insulin-induced insulin receptor substrate-1 degradation is mediated by the proteasome degradation pathway." Diabetes 48(7): 1359-64. Sun, X. J., L. M. Wang, et al. (1995). "Role of IRS-2 in insulin and cytokine signalling." Nature 377(6545): 173-7. Sundaram, P., Z. Pang, et al. (2009). "USP19-deubiquitinating enzyme regulates levels of major myofibrillar proteins in L6 muscle cells." Am J Physiol Endocrinol Metab 297(6): E1283-90. Suzuki, M., R. Setsuie, et al. (2009). "Ubiquitin carboxyl-terminal hydrolase l3 promotes insulin signaling and adipogenesis." Endocrinology 150(12): 5230-9. Sykiotis, G. P. and A. G. Papavassiliou (2001). "Serine phosphorylation of insulin receptor substrate-1: a novel target for the reversal of insulin resistance." Mol Endocrinol 15(11): 1864-9. Takayama, S., M. F. White, et al. (1988). "Phorbol ester-induced serine phosphorylation of the insulin receptor decreases its tyrosine kinase activity." J Biol Chem 263(7): 3440-7. Tan, Y., T. Ichikawa, et al. (2011). "Diabetic downregulation of Nrf2 activity via ERK contributes to oxidative stress-induced insulin resistance in cardiac cells in vitro and in vivo." Diabetes 60(2): 625-33. Tanaka, S., T. Morishita, et al. (1994). "C3G, a guanine nucleotide-releasing protein expressed ubiquitously, binds to the Src homology 3 domains of CRK and GRB2/ASH proteins." Proc Natl Acad Sci U S A 91(8): 3443-7. Taniguchi, C. M., B. Emanuelli, et al. (2006). "Critical nodes in signalling pathways: insights into insulin action." Nat Rev Mol Cell Biol 7(2): 85-96. Tanti, J. F., T. Gremeaux, et al. (1996). "Overexpression of a constitutively active form of phosphatidylinositol 3-kinase is sufficient to promote Glut 4 translocation in adipocytes." J Biol Chem 271(41): 25227-32. Tanti, J. F. and J. Jager (2009). "Cellular mechanisms of insulin resistance: role of stress-regulated serine kinases and insulin receptor substrates (IRS) serine phosphorylation." Curr Opin Pharmacol 9(6): 753-62. Tatham, M. H., E. Jaffray, et al. (2001). "Polymeric chains of SUMO-2 and SUMO-3 are conjugated to protein substrates by SAE1/SAE2 and Ubc9." J Biol Chem 276(38): 35368-74.

166 Tian, Q., J. Taupin, et al. (1995). "Fas-activated serine/threonine kinase (FAST) phosphorylates TIA-1 during Fas-mediated apoptosis." J Exp Med 182(3): 865- 74. Tobe, K., H. Sabe, et al. (1996). "Csk enhances insulin-stimulated dephosphorylation of focal adhesion proteins." Mol Cell Biol 16(9): 4765-72. Tonks, N. K. and B. G. Neel (2001). "Combinatorial control of the specificity of protein tyrosine phosphatases." Curr Opin Cell Biol 13(2): 182-95. Tontonoz, P. and B. M. Spiegelman (2008). "Fat and beyond: the diverse biology of PPARgamma." Annu Rev Biochem 77: 289-312. Treadway, J. L., D. E. James, et al. (1989). "Effect of exercise on insulin receptor binding and kinase activity in skeletal muscle." Am J Physiol 256(1 Pt 1): E138- 44. Treebak, J. T., E. B. Taylor, et al. (2010). "Identification of a novel phosphorylation site on TBC1D4 regulated by AMP-activated protein kinase in skeletal muscle." Am J Physiol Cell Physiol 298(2): C377-85. Trompouki, E., E. Hatzivassiliou, et al. (2003). "CYLD is a deubiquitinating enzyme that negatively regulates NF-kappaB activation by TNFR family members." Nature 424(6950): 793-6. Tsao, T. S., R. Burcelin, et al. (1996). "Enhanced insulin action due to targeted GLUT4 overexpression exclusively in muscle." Diabetes 45(1): 28-36. Uchida, N., S. Hoshino, et al. (2004). "Identification of a human cytoplasmic poly(A) nuclease complex stimulated by poly(A)-binding protein." J Biol Chem 279(2): 1383-91. Ueki, K., T. Kondo, et al. (2004). "Suppressor of cytokine signaling 1 (SOCS-1) and SOCS-3 cause insulin resistance through inhibition of tyrosine phosphorylation of insulin receptor substrate proteins by discrete mechanisms." Mol Cell Biol 24(12): 5434-46. Uldry, M. and B. Thorens (2004). "The SLC2 family of facilitated hexose and polyol transporters." Pflugers Arch 447(5): 480-9. Um, S. H., F. Frigerio, et al. (2004). "Absence of S6K1 protects against age- and diet- induced obesity while enhancing insulin sensitivity." Nature 431(7005): 200-5. Usui, I., T. Imamura, et al. (2004). "GRK2 is an endogenous protein inhibitor of the insulin signaling pathway for glucose transport stimulation." Embo J 23(14): 2821-9.

167 Valadkhan, S. and Y. Jaladat (2010). "The spliceosomal proteome: At the heart of the largest cellular ribonucleoprotein machine." Proteomics. Valiente, M., A. Andres-Pons, et al. (2005). "Binding of PTEN to specific PDZ domains contributes to PTEN protein stability and phosphorylation by microtubule-associated serine/threonine kinases." J Biol Chem 280(32): 28936- 43. Vogel, H. G. (2008). Drug Discovery and Evaluation: Pharmacological Assays. Voliovitch, H., D. G. Schindler, et al. (1995). "Tyrosine phosphorylation of insulin receptor substrate-1 in vivo depends upon the presence of its pleckstrin homology region." J Biol Chem 270(30): 18083-7. Vollenweider, P., M. Clodi, et al. (1999). "An SH2 domain-containing 5' inositolphosphatase inhibits insulin-induced GLUT4 translocation and growth factor-induced actin filament rearrangement." Mol Cell Biol 19(2): 1081-91. Wada, T., T. Sasaoka, et al. (2001). "Overexpression of SH2-containing inositol phosphatase 2 results in negative regulation of insulin-induced metabolic actions in 3T3-L1 adipocytes via its 5'-phosphatase catalytic activity." Mol Cell Biol 21(5): 1633-46. Wang, C., L. Deng, et al. (2001). "TAK1 is a ubiquitin-dependent kinase of MKK and IKK." Nature 412(6844): 346-51. Wang, Q., R. Somwar, et al. (1999). "Protein kinase B/Akt participates in GLUT4 translocation by insulin in L6 myoblasts." Mol Cell Biol 19(6): 4008-18. Wang, Y., M. Falasca, et al. (1998). "Activation of the c-fos serum response element by phosphatidyl inositol 3-kinase and rho pathways in HeLa cells." Cell Growth Differ 9(7): 513-22. Warram, J. H., B. C. Martin, et al. (1990). "Slow glucose removal rate and hyperinsulinemia precede the development of type II diabetes in the offspring of diabetic parents." Ann Intern Med 113(12): 909-15. Way, K. J., E. Chou, et al. (2000). "Identification of PKC-isoform-specific biological actions using pharmacological approaches." Trends Pharmacol Sci 21(5): 181-7. Weigert, C., M. Kron, et al. (2008). "Interplay and effects of temporal changes in the phosphorylation state of serine-302, -307, and -318 of insulin receptor substrate- 1 on insulin action in skeletal muscle cells." Mol Endocrinol 22(12): 2729-40. Weiland, M., F. Bahr, et al. (1991). "The signaling potential of the receptors for insulin and insulin-like growth factor I (IGF-I) in 3T3-L1 adipocytes: comparison of

168 glucose transport activity, induction of oncogene c-fos, glucose transporter mRNA, and DNA-synthesis." J Cell Physiol 149(3): 428-35. Wellbrock, C., M. Karasarides, et al. (2004). "The RAF proteins take centre stage." Nat Rev Mol Cell Biol 5(11): 875-85. White, M. F. (1998). "The IRS-signalling system: a network of docking proteins that mediate insulin action." Mol Cell Biochem 182(1-2): 3-11. Whiteman, E. L., H. Cho, et al. (2002). "Role of Akt/protein kinase B in metabolism." Trends Endocrinol Metab 13(10): 444-51. Wick, M. J., L. Q. Dong, et al. (2000). "Mechanism of phosphorylation of protein kinase B/Akt by a constitutively active 3-phosphoinositide-dependent protein kinase-1." J Biol Chem 275(51): 40400-6. Wijesekara, N., D. Konrad, et al. (2005). "Muscle-specific Pten deletion protects against insulin resistance and diabetes." Mol Cell Biol 25(3): 1135-45. Williams, G. L., T. M. Roberts, et al. (2007). "Bub1: escapades in a cellular world." Cell Cycle 6(14): 1699-704. Willson, T. M., J. E. Cobb, et al. (1996). "The structure-activity relationship between peroxisome proliferator-activated receptor gamma agonism and the antihyperglycemic activity of thiazolidinediones." J Med Chem 39(3): 665-8. Wilson, C. M., Y. Mitsumoto, et al. (1995). "Regulation of cell surface GLUT1, GLUT3, and GLUT4 by insulin and IGF-I in L6 myotubes." FEBS Lett 368(1): 19-22. Witczak, C. A., N. Jessen, et al. (2010). "CaMKII regulates contraction- but not insulin- induced glucose uptake in mouse skeletal muscle." Am J Physiol Endocrinol Metab 298(6): E1150-60. Woelk, T., S. Sigismund, et al. (2007). "The ubiquitination code: a signalling problem." Cell Div 2: 11. Wojtaszewski, J. F., J. Lynge, et al. (1999). "Differential regulation of MAP kinase by contraction and insulin in skeletal muscle: metabolic implications." Am J Physiol 277(4 Pt 1): E724-32. Wood, I. S. and P. Trayhurn (2003). "Glucose transporters (GLUT and SGLT): expanded families of sugar transport proteins." Br J Nutr 89(1): 3-9. Woodgett, J. R. and P. Cohen (1984). "Multisite phosphorylation of glycogen synthase. Molecular basis for the substrate specificity of glycogen synthase kinase-3 and

169 casein kinase-II (glycogen synthase kinase-5)." Biochim Biophys Acta 788(3): 339-47. Wright, D. C., K. A. Hucker, et al. (2004). "Ca2+ and AMPK both mediate stimulation of glucose transport by muscle contractions." Diabetes 53(2): 330-5. Wueest, S., R. A. Rapold, et al. (2010). "Fas activation in adipocytes impairs insulin- stimulated glucose uptake by reducing Akt." FEBS Lett 584(19): 4187-92. Wueest, S., R. A. Rapold, et al. (2010). "Deletion of Fas in adipocytes relieves adipose tissue inflammation and hepatic manifestations of obesity in mice." J Clin Invest 120(1): 191-202. Xu, X., A. Sarikas, et al. (2008). "The CUL7 E3 ubiquitin ligase targets insulin receptor substrate 1 for ubiquitin-dependent degradation." Mol Cell 30(4): 403-14. Xue, B., Y. B. Kim, et al. (2007). "Protein-tyrosine phosphatase 1B deficiency reduces insulin resistance and the diabetic phenotype in mice with polygenic insulin resistance." J Biol Chem 282(33): 23829-40. Yalcin, A., S. Telang, et al. (2009). "Regulation of glucose metabolism by 6- phosphofructo-2-kinase/fructose-2,6-bisphosphatases in cancer." Exp Mol Pathol 86(3): 174-9. Yamaguchi, S., H. Katahira, et al. (2005). "Activators of AMP-activated protein kinase enhance GLUT4 translocation and its glucose transport activity in 3T3-L1 adipocytes." Am J Physiol Endocrinol Metab 289(4): E643-9. Yamamoto, Y., D. W. Kim, et al. (2001). "IKKgamma /NEMO facilitates the recruitment of the IkappaB proteins into the IkappaB kinase complex." J Biol Chem 276(39): 36327-36. Yang, W. L., J. Wang, et al. (2009). "The E3 ligase TRAF6 regulates Akt ubiquitination and activation." Science 325(5944): 1134-8. Yang, W. L., C. Y. Wu, et al. (2010). "Regulation of Akt signaling activation by ubiquitination." Cell Cycle 9(3): 487-97. Yao, J. E., M. Yan, et al. (2009). "Aurora-A down-regulates IkappaBalpha via Akt activation and interacts with insulin-like growth factor-1 induced phosphatidylinositol 3-kinase pathway for cancer cell survival." Mol Cancer 8: 95. Yu, C., Y. Chen, et al. (2002). "Mechanism by which fatty acids inhibit insulin activation of insulin receptor substrate-1 (IRS-1)-associated phosphatidylinositol 3-kinase activity in muscle." J Biol Chem 277(52): 50230-6.

170 Yuan, M., N. Konstantopoulos, et al. (2001). "Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted disruption of Ikkbeta." Science 293(5535): 1673-7. Zamore, P. D., T. Tuschl, et al. (2000). "RNAi: double-stranded RNA directs the ATP- dependent cleavage of mRNA at 21 to 23 nucleotide intervals." Cell 101(1): 25- 33. Zeigerer, A., M. K. McBrayer, et al. (2004). "Insulin stimulation of GLUT4 exocytosis, but not its inhibition of endocytosis, is dependent on RabGAP AS160." Mol Biol Cell 15(10): 4406-15. Zeng, Y., R. Yi, et al. (2003). "MicroRNAs and small interfering RNAs can inhibit mRNA expression by similar mechanisms." Proc Natl Acad Sci U S A 100(17): 9779-84. Zhang, B., J. Berger, et al. (1996). "Insulin- and mitogen-activated protein kinase- mediated phosphorylation and activation of peroxisome proliferator-activated receptor gamma." J Biol Chem 271(50): 31771-4. Zheng, Y., W. Zhang, et al. (2009). "Improved insulin sensitivity by calorie restriction is associated with reduction of ERK and p70S6K activities in the liver of obese Zucker rats." J Endocrinol 203(3): 337-47. Zhou, H., H. Xiong, et al. (2004). "Microtubule-associated serine/threonine kinase-205 kDa and Fc gamma receptor control IL-12 p40 synthesis and NF-kappa B activation." J Immunol 172(4): 2559-68. Zhou, Q. L., J. G. Park, et al. (2004). "Analysis of insulin signalling by RNAi-based gene silencing." Biochem Soc Trans 32(Pt 5): 817-21. Zick, Y. (2004). "Uncoupling insulin signalling by serine/threonine phosphorylation: a molecular basis for insulin resistance." Biochem Soc Trans 32(Pt 5): 812-6. Ziogas, A., K. Moelling, et al. (2005). "CNK1 is a scaffold protein that regulates Src- mediated Raf-1 activation." J Biol Chem 280(25): 24205-11. Zisman, A., O. D. Peroni, et al. (2000). "Targeted disruption of the glucose transporter 4 selectively in muscle causes insulin resistance and glucose intolerance." Nat Med 6(8): 924-8.

171