UvA-DARE (Digital Academic Repository)

Therapeutic agents and signal transduction in inflammatory bowel diseases

Löwenberg, M.

Publication date 2006 Document Version Final published version

Link to publication

Citation for published version (APA): Löwenberg, M. (2006). Therapeutic agents and signal transduction in inflammatory bowel diseases.

General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl) Download date:10 Oct 2021

Therapeutic agents and signal transduction in

inflammatory bowel diseases

Löwenberg, Mark “Therapeutic agents and signal transduction in inflammatory bowel diseases” Thesis University of Amsterdam

© M. Löwenberg, Amsterdam, The Netherlands. All rights reserved. No part of this publication may be reproduced or transmitted in any form by any means, electronic or mechanical, including photocopy, recording or any information storage and retrieval system, without written permission of the author.

Cover: Confocal images of immunofluorescent double stainings demonstrating the cellular distribution of Lck and the glucocorticoid receptor in activated human CD4+ T lymphocytes. Lck and the glucocorticoid receptor were stained with a red (TRITC) or green (FITC) fluorescent dye respectively, and merged images are shown. Yellow staining indicates glucocortoid receptor-Lck colocalization.

Sponsors: Abbott, Altana, AstraZeneca, Boston Scientific, Ferring, Giuliani Pharma, Janssen Cilag, Jurriaanse Stichting, Novartis, Schering Plough, Tramedico, Zambon.

Printed by Ponsen & Looijen b.v., Wageningen

Therapeutic agents and signal transduction in

inflammatory bowel diseases

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. mr. P.F. van der Heijden ten overstaan van een door het college voor promoties ingestelde commissie, in het openbaar te verdedigen in de Aula der Universiteit op woensdag 20 september 2006, te 12:00 uur door

Mark Löwenberg

geboren te Rotterdam Promotiecommissie

Promotores: Prof. dr. S.J.H. van Deventer Prof. dr. M.P. Peppelenbosch Co-promotor: Dr. D.W. Hommes

Overige leden: Prof. dr. F. Buttgereit Prof. dr. P.J. Coffer Prof. dr. E.J. Kuipers Prof. dr. M.M.Levi Prof. dr. R.A.W. van Lier Prof. dr. H. Spits

Faculteit der Geneeskunde

Contents

Chapter 1. General introduction p.7

Chapter 2. Update on biological and small molecule therapy for p.35 inflammatory bowel diseases Drugs (accepted for publication); Inflammatory Bowel Diseases. 2004; 10 Suppl 1:S52-7

Chapter 3. Specific inhibition of c-Raf activity by semapimod induces p.51 clinical remission in severe Crohn's disease Journal of Immunology. 2005; 175(4):2293-300

Chapter 4. Rapid immunosuppressive effects of glucocorticoids mediated p.71 through Lck and Fyn Blood. 2005; 106(5):1703-10

Chapter 5. Glucocorticoids cause rapid dissociation of a T cell receptor- p.91 associated protein complex containing Lck and Fyn EMBO reports. Accepted for publication

Chapter 6. Kinome analysis reveals nongenomic glucocorticoid receptor- p.107 dependent inhibition of insulin receptor signaling Endocrinology. 2006; 147(7):3555-62

Chapter 7. A role for STAT5 in steroid resistant ulcerative colitis? p.127 Inflammatory Bowel Diseases. 2006; 12(7):665

Chapter 8. Discussion p.133

Appendices 1. Summary p.142 2. Samenvatting voor niet ingewijden p.143 3. Androgen-mediated regulation and functional implications of p.147 FKBP51 expression in prostate cancer Journal of Urology. 2005;173(5): 1772-7 4. Abbreviations p.161 5. Dankwoord p.163 6. Curriculum Vitae p.164 7. Bibliography p.166

Chapter 1

General introduction

Summary p.8 1. Inflammatory bowel diseases (IBD) p.9

2. Immune responses in IBD p.10 2.1 Targeted immune therapies p.11 2.1.1 Conventional therapies p.12 2.1.2 Biologicals p.13

3. Glucocorticoids p.14 3.1 Genomic mode of glucocorticoid action p.15 3.2 Nongenomic mode of glucocorticoid action p.16

4. Signal transduction in IBD p.18 4.1 Mitogen activated protein kinases p.18 4.2 Signal transducers and activators of transcription p.18 4.3 Nuclear factor-κB p.19

5. The kinome p.21 5.1 Kinome profiling p.22 5.2 Small molecules p.23

6. Aim and outline of this thesis p.25

7 Chapter 1

Summary Crohn’s disease and ulcerative colitis, the two major forms of inflammatory bowel diseases (IBD), are chronic inflammatory disorders of the gastrointestinal tract. Current IBD therapy is based on ‘step-up’ algorithms, which initiate treatment with glucocorticoids, followed by immunomodulatory agents (including azathioprine and methotrexate), and defer therapy with biological therapy (such as anti-TNF) until patients become refractory to conventional therapeutics. As a consequence of the limited efficacy and significant toxicities of current IBD therapies, there is widespread interest in the development of novel drugs. Small molecules that specifically target intracellular signaling pathways have proven effective in experimental colitis and in severe Crohn’s disease. We here define a cellular target of the kinase inhibitor semapimod and describe a novel anti-inflammatory mechanism in Crohn’s disease (chapter 3). Furthermore, an unexpected immunosuppressive mechanism for glucocorticoids is identified in T cells (chapter 4 and 5). We demonstrate that glucocorticoids impair proximal T cell receptor signaling through rapid dissolution of membrane bound glucocorticoid receptor multi-protein complexes. Chapter 6 describes a genomic-independent mechanism underlying glucocorticoid-induced insulin resistance, a major side effect of glucocorticoid therapy. Overall, these studies provide novel insight into the nongenomic mode of glucocorticoid action. Lastly, we review the role of the transcription factor STAT5 in the pathogenesis of glucocorticoid resistant ulcerative colitis (chapter 7).

8 Introduction

1. Inflammatory bowel diseases (IBD) Inflammatory bowel diseases (IBD), which include Crohn’s disease (CD) and ulcerative colitis (UC), are chronic inflammatory diseases of the gastrointestinal tract characterized by relapses alternating with episodes of quiescent disease. Although UC and CD are generally accepted as clinically distinct conditions with distinguishing clinical, anatomical, and histological findings, a diagnostic gold standard remains elusive 1. CD may involve the entire gastrointestinal tract, is most commonly found in the ileocecal area and rectal sparing is commonly seen. UC usually starts in the rectum, progresses proximally in a continuous fashion and stays limited to the colon. A discontinuous transmural with granulomas is typical for CD, whereas inflammation in UC involves the superficial layers of the bowel. About 10% of patients have indeterminate features between UC and CD that cannot be clearly categorized 2. There is no established definition for this subset of patients and their condition is termed ‘indeterminate colitis’. These disorders probably represent a continuum of diseases, with UC and CD at the opposite ends. IBD typically presents early in adult life and occurs in approximately 0.2% of the western population 3-5. CD and UC are multifactorial diseases caused by the interplay of genetic, environmental and immunological factors, but the exact aetiology remains to be defined. There is sufficient evidence that CD and UC are, in part, the result of a genetic predisposition, with multiple susceptibility genes 6-16. In addition, environmental factors have been implicated in the initiation and maintenance of mucosal inflammation in IBD, such as microflora in the gut, dietary antigens and improved hygiene 17,18. The importance of environmental factors is supported by the striking increase in the frequency of CD in the more developed world over the past 50 years, and the increased recognition of the disease corresponding with progressive industrialisation in less developed countries. Lastly, the immune system plays a crucial role in IBD pathogenesis. The mucosal immune system has to respond effectively to pathogens, such as bacteria and viruses, whereas the presence of non- pathogenic pathogens has to be ignored. As a result, the intestinal immune system must tolerate the commensal flora which is constantly present in the lumen and maintain mucosal homeostasis by controlled inflammatory responses. It is generally accepted that CD and UC arise from a pathological antigen-driven inflammatory response that is triggered either by an unrecognized pathogen or by nonpathogenic bowel flora within a genetically susceptible

9 Chapter 1

individual 19,20. The complex interplay between genetic, microbial and environmental factors culminates in a sustained activation of the intestinal mucosal immune system eventually leading to tissue destruction. It is hoped that a better understanding of the environmental, genetic, and immunological mechanisms that produce UC and CD will lead to improved therapies for these diseases.

2. Immune responses in IBD There are many roads to colitis and different immune cells play a role in IBD pathogenesis (such as T and B lymphocytes, monocytes, , dendritic cells, eosinophils, neutrophils, granulocytes, mast cells, natural killer cells and epithelial cells) 21,22. During the initial innate immune response, neutrophils pass from the circulation through gaps in the vascular endothelium to infiltrate the intestinal tissue where they release antimicrobial peptides and reactive oxygen intermediates that cause tissue damage. Through the production of chemokines and proinflammatory (such as (TNF)-α, interleukin (IL)-1β, IL-6, and IL-8) macrophages and other white blood cells are recruited and become activated 23. Adaptive immune responses are characterized by the differentiation and activation of T and B cells 24,25. Microbes or antigens that enter the mucosa are taken up by antigen presenting cells, such as naïve dendritic cells (DCs) and macrophages, which present antigen to T and B lymphocytes. T cells, the principal conductors of adaptive immune responses, are subsequently triggered to mature towards Th1, Th2 or T-regulatory cells. Th1 cells mainly produce interferon (IFN)-γ, IL-2 and TNFα resulting in and B cell activation; important for the elimination of intracellular microbes and viruses. A Th2-type immune response is characterized by high IL-4, IL-5 and IL-13 levels leading to activation of mast cells, eosinophils and B cells; essential for the elimination of parasites and nematodes 26. CD is associated with a Th1 phenotype, which is illustrated by production of IFN-γ, IL-2, IL- 12 and TNFα 27,28, whereas high levels of IL-5 and IL-13 are seen in UC 29,30. The balance between Th1 and Th2-type immune responses is largely controlled by immuno-regulatory cytokines (i.e. transforming growth factor (TGF)-β and IL-10) produced by T-regulatory cells, that play an important role in maintaining gut homeostasis 31. It has been shown that gut DCs contribute to the dysregulated immune response that underlies IBD pathogenesis, either as a

10 Introduction

local effector cell or by modifying adaptive immune response via lymphocyte activation. Previous studies revealed enhanced Toll-like receptor (TLR)2 and TLR4 expression on DCs, as well as increased cytokine production and proliferation rates of DCs in the colonic mucosa of IBD patients 32. Furthermore, increased intestinal permeability has been demonstrated to be involved in IBD and evidence suggests that a leaky intestinal barrier intensifies antigen absorption leading to an exaggerated immune response 33,34. Intestinal epithelial cells play an important role in regulating immune tolerance, an essential mucosal defence mechanism maintaining hyporesponsiveness to harmless lumenal commensals and their products. Chapter 2 discusses several molecular mechanisms that ensure immune tolerance and possible mechanisms underlying loss of tolerance 35.

2.1 Targeted immune therapies CD and UC are characterized by a non-specific mucosal inflammatory response which is independent from the molecular mechanism of induction. Although curative treatment of IBD is still lacking (including total colectomy), conventional treatment therapy is based on ‘step- up’ algorithms, which initiate treatment with corticosteroids, followed by immunomodulatory agents (such as azathioprine, 6-mercaptopurine or methotrexate). Biologic therapies are introduced when patients become refractory to conventional drugs (Figure 1) 36. The choice of which depends on the clinical goal (i.e. induction or maintenance of remission), extent and severity of disease, response to current or prior medication and the presence of complications.

Figure 1. Step-up therapy for IBD. Conventional IBD therapy is based on step-up algorithms, which initiate treatment with conventional immunosuppressive agents (corticosteroids), followed by immunomodulators (such as azathioprine or methotrexate) and defer therapy with biologicals until patients become refractory to current treatment approaches.

11 Chapter 1

2.1.1 Conventional therapies 5-aminosalicylic acid (5-ASA) is considered as first-line therapy for mild to moderate UC (remission rates: 40-80%) 37. In contrast, efficacy of 5-ASA in CD is only slightly better compared to placebo 38-40. 5-ASA inhibits the production of cytokines and inflammatory mediators, but the underlying mechanism remains only partly understood. A recent study indicated that the nuclear hormone receptor peroxisome proliferator-activated receptor- gamma (PPAR-gamma) mediates 5-ASA activities in experimental colitis 41. These data underscored the importance of the development of better PPAR-gamma ligands with higher affinities. For several decades, glucocorticoids (GCs) were the only potent anti-inflammatory and immunosuppressive agents available for treatment of active IBD 42. GCs form the basis of current IBD treatment and provide response rates of 45-90% in UC and 60-78% in CD 42,43. Although GCs suppress active inflammation in the acute setting, they are ineffective maintenance agents and long-term use is associated with high relapse rates and side effects, such as insulin resistance, osteoporosis and hypertension. One year after starting GC treatment only 48% of UC and 32% of CD patients are steroid free without surgery 44. GC resistance represents a challenging problem in the clinical management of IBD patients. Approximately one-third of CD patients become GC dependent (i.e. disease responds to GC therapy, but relapses when GCs are withdrawn) and one-fifth become GC resistant (i.e. ≥ 30 mg prednisolone for ≥ 1 week without clinical or endoscopical improvements) 42,45-49. Moreover, one-quarter and one-sixth of UC patients become resistant to or dependent on GCs respectively 50,51. As a result, a significant number of IBD patients need additional immunomodulatory therapy as maintenance therapy. Immunomodulators have been used to achieve clinical remission in IBD and to accomplish withdrawal from GCs or reduction in dose (i.e. steroid-sparing effect) 44,52,53. Azathioprine and its metabolite 6-mercaptopurine (6-MP) are widely used as immunosuppressive agents in the treatment of IBD 54-58. Previous reports demonstrated that azathioprine induces T cell apoptosis via inhibition of Rac 59, but it’s use is limited by their slow onset of action and potential serious adverse events, such as leucocytopenia and liver toxicity 60. Calcineurin inhibitors, such as cyclosporine A and tacrolimus (FK506), are of limited value in CD and UC 61-69. Their immunosuppressive effects are mediated through impaired nuclear translocation of

12 Introduction

the transcription factor NFAT (nuclear factor of activated T cells) thereby preventing downstream initiation of transcription of T cell cytokines. More placebo-controlled trials are needed to evaluate efficacy and safety of cyclosporin therapy in IBD. Methotrexate, a folic acid antagonist, is an effective immunosuppressive agent in the treatment of IBD, but its use is limited by unpredictable efficacy and toxicity, such as myelosuppression and hepatotoxicity. Methotrexate acts rapidly and has been established as an induction agent for GC-dependent CD (remission rate: 40%), and for maintenance of remission after successful induction (65% relapse free after 40 weeks) 70,71. However, its role in UC is still unclear and is a matter for ongoing debate 72,73. Although it has been shown that methotrexate induces T cell apoptosis and reduces pro-inflammatory cytokine synthesis via the anti-inflammatory mediator adenosine 74-76, the precise underlying mechanism remains to be defined. The limitations in efficacy and safety encountered with current medical approaches for IBD continue to drive the search for better therapeutic agents.

2.1.2 Biologicals Current therapy for IBD has focused on non-specific suppression of the inflammatory process, which is characterized by overproduction of immune cells, inflammatory cytokines and tissue-destructive enzymes. The last decade, advances in our knowledge of IBD immunopathogenesis have opened a new era with the development of biologic therapies selectively interfering within inflammatory cascades. Identification of molecular targets has resulted in the development of recombinant cytokines, cytokine receptor antagonists, chimeric or humanized antibodies and antisense oligonucleotides 77. Blockade of inflammatory pathways through targeted immune therapies, most notably TNF inhibition, has revolutionized our approach to treat CD. TNF, an extra-cellular mediator of inflammation, plays a central role in the initiation and amplification of inflammatory reactions leading to tissue destruction. Antibodies against TNF-α, such as , have been proven efficacious in both inducing clinical remission and mucosal healing in CD 78-82. Although it has been reported that infliximab induces T cell apoptosis 83-86, the molecular mechanisms responsible for the observed clinical effects remain only partly understood. Most interesting, infliximab is also effective in UC 87, and additional studies are needed to define its underlying mechanism of action in this disease 88. The main side effects of anti-TNF therapies include

13 Chapter 1

infectious complications, such as opportunistic infections and reactivation of tuberculosis. With infliximab, a negative interaction with cardiovascular and autoimmune disorders have been described, as well as a 1-2% mortality rate 89,90. Several other antibodies that are directed against TNF (i.e. adalimumab, certolizumab) have been designed demonstrating clinical efficacy in IBD 77. However, immunogenicity limits the use of TNF inhibitors 91. Therapeutics that selectively block leukocyte adhesion and migration represent a novel and promising strategy in the treatment of IBD, such as natalizumab (α-4 integrin antagonist) 77,92- 94. However, it’s use has been associated with progressive multifocal leukoencephalopathy 95,96, an infrequent opportunistic infection of the nervous system that primarily affects individuals with suppressed immune systems. This drug has been recently reintroduced for IBD treatment and the risk-benefit evaluation will have to determine the eventual place in therapy for natalizumab 97. In addition, other cytokines represent promising future targets for IBD therapy. Visilizumab (anti-CD3 antibody), tocolizumab (anti-IL-6 receptor), daclizumab and basiliximab (anti-IL-2 receptor), ABT874 (anti-IL-12) and fontolizumab (anti-IFN-γ) demonstrated efficacy in preclinical and phase II/III clinical studies 77,98-100. Sargramostim (i.e. granulocyte-macrophage stimulating factor) was clinically efficient in moderate to severe CD, and has been shown to reduce neutrophil reponses and impair IL-8 synthesis 101,102. Topically applied ISIS-2303, an antisense oligonucleotide against intracellular adhesion molecule-1 (ICAM-1), demonstrated clinical efficacy in UC and CD 103-105. Hence, biologicals that target a wide range of molecules demonstrated clinical efficacy and are relatively well-tolerated. More clinical trials are currently being conducted exploring the safety and efficacy of biologic agents, and the search will certainly open new and exciting perspectives on the development of therapies for IBD. Although biologicals lighten many of the symptoms, they generally do not interfere with the underlying molecular mechanisms, have dose-limiting side effects and disease relapse remains a clinical problem. As a result, there is a continuous search for better therapy regimens to treat IBD.

3. Glucocorticoids (GCs) Steroid hormones, including GCs, estrogens, progesterone’s and androgens, freely penetrate the hydrophobic plasma membrane and bind to cytoplasmic receptors, which belong to the super family of nuclear hormone receptors. Chaperones, such as heat-shock proteins (Hsp)

14 Introduction

and immunophilins, are physically associated with nuclear hormone receptors and are important for receptor stability, ligand binding and nuclear translocation 106. Ligand binding induces a conformational change of these receptors resulting in receptor activation and nuclear translocation. Once nuclear, the activated receptor binds to specific DNA sequences (i.e. hormone response elements) resulting in transactivation or transrepression of target genes, thereby regulating numerous physiological processes, such as cellular proliferation, differentiation, apoptosis and metabolic processes. GCs are powerful anti-inflammatory and immunosuppressive agents that mediate their effects via the cytoplasmic GC receptor. Inactive GC receptors are associated with (co)chaperones such as Hsp, Hsp organizer protein (Hop) and immunophilins (i.e. FK506 binding protein or FKBP) 107. Upon GC receptor ligand binding, chaperones dissociate and the activated GC receptor dimer translocates into the nucleus where it binds to GC receptor responsive elements (GREs) in the promoter regions of target genes, leading to altered gene transcription (Figure 2) 108,109.

3.1 Genomic mode of glucocorticoid action GC receptors control transcriptional processes through binding to positive or negative GREs in the promoter regions of target genes (Figure 2). Examples of GC-induced enhanced gene expression, mediated by positive GREs, are the anti-inflammatory mediators IL-10 and lipocortin-1 110. Repression of negatively regulated target genes is mediated by negative GREs, such as genes encoding for inflammatory cytokines (IL-1, IL-2, TNF-α), pro- 110-112 inflammatory enzymes (phospholipase A2, synthase, cyclooxygenase-2) , or adhesion molecules (ICAM-1, VCAM-1, E-Selectin) 113-117. These genomic GC effects, which lead to anti-inflammatory and immunosuppressive effects, are based on impaired immune cell function 118-120, and induced immune cell apoptosis 121,122. Furthermore, GC-induced immunosuppressive action is mediated through direct protein-protein interaction between GC receptors and pro-inflammatory transcription factors, such as activator protein-1 (AP-1), nuclear factor-κB (NFκB), nuclear factor of activated T cells (NFAT), and signal transducers and activator of transcription (STAT) 123-125. Thus, GCs have a broad therapeutic spectrum and control inflammation through multiple transcriptional mechanisms acting on diverse target cells and proteins. GC-induced transactivation of target gene expression is associated

15 Chapter 1

with frequently occurring metabolic and endocrine side effects. Identification of GC receptor- mediated transcriptional mechanisms (i.e. functional analysis of negative and positive GREs that are associated with immunosuppressive action and side effects), may result in the development of more selective and safer GC receptor agonists.

3.2 Nongenomic mode of glucocorticoid action Using DNA microarray analysis, it has been demonstrated that approximately 20% of the expressed human leucocyte genome was positively or negatively affected by GC treatment 126. In addition to the classical mode of GC action, which arises within hours or days, there is increasing evidence for GC effects which occur within minutes 127-134. These rapid effects are too fast to be due to changes at the transcriptional level and are therefore termed nongenomic, to distinguish them from the traditional mechanism of GC action. High GC dosages are able to induce nongenomic effects resulting in clinical effects that occur within minutes 135, but the underlying mechanisms remain to be defined (Figure 2). Nongenomic GC effects on cellular physiology are mediated via GC receptor-dependent or GC receptor-independent mechanisms. There is evidence that the family of G protein-coupled receptors, major initiators of signaling pathways, mediate nongenomic GC receptor-independent effects 136. In addition, plasma membrane-bound GC receptors have been demonstrated in lymphocytes and peripheral blood mononuclear cells (PBMCs) that are able to trigger a chain of intracellular events 137, but their functional relevance remains to be defined. Saturation of cytosolic GC receptors is almost complete with 100 mg prednisone equivalent a day. However, the usage of much higher dosages is successful in daily clinical practice. It is speculated that these effects are due to additional nongenomic effects, mediated through cytosolic or membrane-bound GC receptors. Alternatively, it is possible that rapid GC effects are mediated through nonspecific physicochemical interactions with the plasma membrane. Previous studies have demonstrated nongenomic GC-induced inhibition of T cell receptor (TCR)-mediated intracellular calcium mobilization and suppressed phospholipase C (PLC)γ-1 tyrosine phosphorylation 133. Moreover, impaired epidermal growth factor (EGF)-mediated signaling due to short-term GC therapy has been shown, leading to reduced synthesis of the inflammatory mediator arachidonic acid 138. Pre-clinical observations showed that acute GC administration led to nongenomic activation of endothelial nitric oxide synthase (eNOS),

16 Introduction

leading to reduced vascular inflammation and decreased myocardial infarct size following ischemia and reperfusion injury 139. This study demonstrated beneficial early GC effects in the treatment of myocardial infarction, but the responsible mechanism remained undefined. In line with these observations, non-nuclear neuroprotective effects were seen following high- dose GC therapy in a cerebral ischemia model 140. These findings revealed that GCs rapidly activate eNOS through the phosphatidylinositol 3-kinase (PI3K) and protein kinase B (PKB) pathway resulting in acute neuroprotective effects through an increase in cerebral blood flow. Delineation of the molecular basis responsible for nongenomic GC effects may lead to the development of novel immunosuppressive therapies and may provide insight into the underlying inflammatory mechanisms.

Figure 2. Genomic and nongenomic glucocorticoid (GC) action. Left: GCs mediate genomic effects via cytoplasmic GC receptors (GR). Upon receptor ligand binding, the GC receptor becomes activated and dimerizes (not shown). GC receptor homodimers subsequently translocate into the nucleus leading to transrepression or transactivation of target gene expression. Transactivation (i.e. enhanced gene expression) is associated with metabolic or endocrine side effects, whereas transrepression (i.e. reduced gene expression) is associated with immunosuppressive and anti-inflammatory effects. Right: Nongenomic GC-induced effects are thought to be mediated through cytosolic (cGR) or membrane-bound glucocorticoid receptors (mGR) and/or via unspecific physicochemical interactions with membranes. In this simplified scheme only heat shock protein (HSP) is depicted as a GC receptor chaperone.

17 Chapter 1

4. Signal transduction in IBD Cellular signal transduction refers to the movement of signals from outside the cell to inside. The complex network of signal transduction pathways involves the coupling of ligand- receptor interactions to many intracellular events including phosphorylation by kinases that change enzyme activities and protein conformations. The eventual outcome is an alteration in cellular activity and changes in the program of genes expressed within the responding cells. Activation of intracellular signaling cascades is rapid and enables cells to respond to environmental changes in a fast and regulated manner.

4.1 Mitogen Activated Protein Kinases (MAPK) One of the best studied signaling routes represent mitogen activated protein kinase (MAPK) pathways; i.e. extra-cellular signal regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38 MAPK (Figure 3) 141. MAPKs were identified by virtue of their activation in response to growth factor stimulation of cells in culture; hence the name mitogen activated protein kinases. MAPKs form evolutionary well-covered cascades of signaling proteins that are found in all eukaryotes, and are principal regulators of cellular physiology. The classic MAPK cascade consists of three sequential intracellular protein kinase activation steps and is initiated by activation of MAPK kinase kinases (MAPKKKs). An activated MAPKKKs phosphorylates downstream MAPK kinases (MAPKKs). The next step in this intracellular phophorylation process is activation of a MAPK, which ultimately targets other kinases or transcription factors 142, thereby regulating the process of gene transcription. MAPKs are involved in many aspects of immune-mediated pathology, and increased activities of MAPKs in the inflamed colonic mucosa of IBD patients have been shown 143-145. However, in vivo regulation of MAPK pathways is extremely complex and the actual importance of these cellular processes in inflammatory pathology remains elusive 146.

4.2 Signal transducers and activators of transcription (STAT) The protein family of signal transducers and activators of transcription (STAT) consist of 7 STAT proteins (i.e. STAT 1, 2, 3, 4, 5A, 5B, 6) that transfer cytokine signaling from the cell surface to the nucleus 147. STATs are latent cytoplasmic transcription factors that become activated through receptor-associated Janus kinases (Jaks) upon signaling by extra-cellular

18 Introduction

cytokines. Upon dimerization, nuclear translocation takes place where activated STAT proteins interact with specific DNA binding elements thereby regulating transcription of immune response genes (Figure 3) 148,149. The Jak/STAT signal transduction pathway is regulated by specific suppressors, such as the SOCS (suppressors of cytokine signaling) family 150. SOCS proteins either directly inhibit Jak catalytic activities or compete with STATs for binding to specific cell-surface receptors. STATs play an important role in immune cell activation, such as T cells, macrophages and dendritic cells 147. In experimental colitis, STAT1, STAT3 and STAT4 have been demonstrated to be involved in the development and perpetuation of immune responses 151. Lamina propria mononuclear cells (LPMCs) isolated from CD patients displayed increased total STAT1 and STAT4 levels compared to UC and controls 151,152. Increased total and phosphorylated (i.e. biologically active) STAT3 levels were detected in LPMCs from CD and UC patients compared to non- inflammatory control cells, whereas no differences in STAT6 levels were seen between IBD and non-IBD subjects 151,153,154. Recent reports underscored the important role for STAT3 in mediating inflammation and fibrosis. Experimental studies showed that SOCS3 attenuated STAT-dependent pro-inflammatory signaling and reduced fibrosis 155,156. However, total STAT2 expression was reduced in LPMCs obtained from CD and UC compared to non-IBD subjects, which could be secondary to a continuous activation of mucosal immune cells 157. Further research is needed to define the role of different STAT proteins in IBD pathogenesis.

4.3 Nuclear factor-κB (NFkB) Different signal transduction pathways, originating from a wide variety of cellular stresses and stimuli, converge on a single target: the NF-κB (nuclear factor-κB)-IkappaB complex and its activating kinase IKK (inhibitor of kappaB kinase). NF-κB is a transcription factor which is present in the cytoplasm of most human cells 158. The NF-κB family consists of several proteins including p50, p52, p65, RelA, RelB and RelC. Activation of cells with various stimuli initiates a signaling cascade that leads to the release of NF-κB from an inactive complex followed by nuclear translocation of NF-κB where it binds to DNA and regulates gene transcription (Figure 3) 159. The prototypical NF-κB is a heterodimer composed of the p50 and p65 subunits and the latter is the most frequent component of active NF-κB in humans. NF-κB is a key regulator of many genes associated with immune function in the gut.

19 Chapter 1

For instance, NF-κB plays an important role in the transcriptional regulation of cytokine genes (i.e. IL-1, IL-2, IL-6, IL-8, IL-12) and adhesion molecules (i.e. E-selectin, ICAM-1) in lymphocytes, epithelial cells and monocytes 160. Previous work revealed increased NF-κB levels (in particular the p65 subunit) in epithelial cells and lamina propria macrophages in both CD and UC patients 161-165. Several anti-inflammatory drugs, such as aspirin-derivatives and anti-TNF antibodies, interfere with NF-κB activity 166. Although these drugs do not specifically target NF-κB, at least some of their anti-inflammatory effects are due to inhibition of NF-κB 167. GCs specifically target NF-κB through direct protein-protein interaction resulting in impaired NF-κB-mediated gene expression and reduced cytokine synthesis 168. It has been shown that p65 antisense oligonucleotide treatment led to an abrogation of chronic intestinal inflammation in experimental colitis and decreased proinflammatory cytokine production by lamina propria macrophages in active IBD patients 162. Although activation of NF-κB is not specific for IBD, its perpetuated activation makes it an attractive target for therapeutic intervention 169,170. A better understanding of NF-κB signaling pathways will be beneficial for the development of new generations of anti-inflammatory drugs (i.e. small molecules or oligonucleotides) that specifically target NF-κB, either by interfering with NF- κB DNA-binding activities or through inhibition of IKK.

Figure 3. MAPK, STAT5 and NFkB signal transduction cascades. MAPK signaling routes are discussed in detail in chapter 2. STAT5 proteins become tyrosine phosphory- lated by receptor associated Janus kinase 2 (JAK2), leading to dimerization and nuclear translocation. STATs bind to specific DNA elements thereby regulating gene transcrip- tion. NFkB is bound to IkB inhibitory proteins in unstimulated cells. Upon stimulation, NFkB is released from this complex, followed by NFkB nuclear translocation and DNA binding.

20 Introduction

5. The kinome The term signal transduction became popular in the early 1980s and now it is considered one of the most intensively studied areas of modern cell biology. Three different classes of signal transducing receptors have been identified, which are able to initiate intracellular signaling events. The first group consists of cell-surface receptors with intrinsic kinase activities, such as the insulin receptor. Membrane receptors with intrinsic phosphatase activities have also been identified, including the leucocyte surface receptor CD45. A second group includes G- coupled receptors (for example adrenergic or glucagon hormone receptors) that lack intrinsic enzymatic activity, yet are coupled inside the cell by direct protein-protein interactions to GTP-binding and hydrolyzing proteins. A third group consists of cytoplasmic signaling receptors that migrate to the nucleus upon ligand binding, thereby directly interfering with gene transcription (including the superfamily of nuclear hormone receptors). The responsible enzymes (kinases) facilitate the addition of a negatively charged phosphate group to a recipient serine, threonine or tyrosine residue on a target protein (Figure 4), thereby regulating cellular processes such as proliferation, maintenance of cell shape and apoptosis.

Figure 4. Protein phosphorylation and dephosphorylation. Kinases phosphorylate substrates by adding negatively charged phosphate groups to hydroxyl groups found on serine, threonine or tyrosine residues. Phosphatases in turn remove these phosphate groups. Phosphorylation and dephos- phorylation modulates activities of proteins thereby regulating cellular physiology. P, phosphate group.

The process of protein phosphorylation regulates protein function in both normal and disease states. All cellular kinases, collectively designated the ‘kinome’, consists of over 500 kinases (Table 1). The choice of target is decided by the catalytic domain of the kinase, which usually recognizes a docking domain on the substrate and a specific short sequence of amino acids (i.e. the consensus motif) that surrounds the residue to be phosphorylated. Each kinase can have several substrates and be itself a substrate for other kinases. The kinome signal

21 Chapter 1

transduction consists of a highly complex network of proteins and we are only beginning to understand some basic principles through which signal transduction is controlled. Important questions concern how we can study processes regulating these signaling networks, and how we can delineate interactions between different pathways.

Table 1. The humane kinome. More than 500 kinases have been identified and some examples are depicted here. Group names of different kinase families and members for each group are shown.

5.1 Kinome profiling Comprehensive cellular metabolism or signal transduction events can be analyzed in a fast and relatively easy way employing a recently developed array technique (i.e. kinome profiling) (Figure 5) 171. Peptide microarrays have been developed that contain several hundred specific kinase consensus peptide substrates. This experimental approach has several advantages compared to traditional in vitro enzyme activity assays. First, enzymatic activities of many different kinases in one single experiment can be analyzed, and the assay is performed under the same conditions using similar reagents. Second, it is possible to study numerous signal transduction pathways in a high throughput array-based manner, which opens the possibility to study downstream signaling events and feedback mechanisms in a complex web of interactions. Furthermore, signal transduction displays functional redundancy: i.e. the ability of a system to keep functioning normally in the event of a component failure, by having backup components that perform duplicate functions. Redundancy in signal transduction implicates that the degree of activation of a signaling pathway depends on stimulatory and inhibitory factors as well as on cross-talk between different pathways, but these phenomena are poorly defined. Kinome profiling might provide

22 Introduction

molecular insight into these complex mechanisms. Finally, as there is great potential for cross-reactivity of kinase inhibitors, this technique can be used to assess molecular specificity and to identify off-target interactions. Preliminary evidence suggests that this technique is a promising tool for kinome-wide analysis 171,172, but further research is needed to define its usefulness in order to identify new kinase targets or delineate novel signal tranduction routes.

Figure 5. Kinome analysis. Pepchips are incubated with cell or tissue lysates supplemen- ted with radioactive ATP (33-γATP). Arrays are scanned using a phospho- imager and analyzed with microarray software analysis. Active kinases specifically phosphorylate kinase consensus substrates incorporating radioactive ATP. A MAPK substrate is depicted, and a scanned picture of a pepchip is shown.

5.2 Small molecules Protein kinases are directly involved in many diseases, including cancer and inflammation, and have become important molecular targets for drug development nowadays 173,174. All of the more than 500 protein kinases identified in the human genome contain an ATP-binding pocket, an area within the activation loop of kinases in which phosphorylation takes place 175. ATP-binding sites are the predominant target for synthetic kinase inhibitors (i.e. small molecules), that may be easily administered as oral therapeutic compounds. These inhibitors can rapidly and often specifically alter the activation state of a target kinase. They have significant potential in the clinical treatment of major diseases and many candidates have been tested in (pre)clinical studies for anti-cancer strategies 176. Examples of small molecules that are approved for leukemia and lung cancer therapy include imatinib and gefitinib respectively. These drugs are currently used in the clinic and provided proof-of-principle that small molecule are effective therapies. In theory, blocking inflammatory signaling pathways may have profound effects on inflammatory pathology.

23 Chapter 1

Increased understanding of cellular signal transduction pathways involved in immunological processes has resulted in the development of small molecules specifically targeting inflammatory signaling molecules 177-179. In particular, pharmaceutical intervention of MAPK pathways has attracted widespread interest. P38 MAPK inhibitors (i.e. SB239063, RWJ67657, RPR203494) demonstrated anti-inflammatory effects in preclinical models 180-184. The p38 MAPK inhibitor BIRB796 diminished LPS-induced cytokine release and leukocyte responses in healthy subjects 185. However, a recent randomized controlled trial evaluating the efficacy of BIRB796 in CD patients revealed no clinical and endoscopical improvement at 8 weeks 186. Interestingly, the p38 MAPK and JNK inhibitor semapimod (CNI-1493) showed potent anti-inflammatory effects in severe CD patients 143. Semapimod therapy resulted in clinical remission and endoscopic improvement in all 12 patients with therapy-resistant CD, and no serious adverse events were reported. Although the cellular target of semapimod remained to be defined, parallel in vitro and in vivo studies supported the notion that MAPKs are involved in CD pathogenesis 143. Besides p38 MAPK, JNK plays a pivotal role in regulating activity of disease-associated genes and their products. Several JNK inhibitors demonstrated anti-inflammatory effects in preclinical settings (i.e. CEP1347, SP600125), whereas others failed 187. Finally, inhibitors of the ERK signal transduction pathway have been the focus of intense drug discovery as anti-cancer strategy. Small molecules targeting this signaling cascade (BAY43-9006, PD0325901) have proven a fruitful endeavour for anti- cancer therapies 188-190. Evidence suggests that this signaling route is also involved in regulating inflammatory responses, largely mediated through the upstream signaling molecule c-Raf. c-Raf has been shown to inhibit immune cell apoptosis as well as controlling NFκB- mediated synthesis of pro-inflammatory mediators 191-196. Up to now, there are no clinical data with ERK pathway inhibitors as anti-inflammatory therapy. Recent findings further supported a link between survival signaling cascades (such as ERK) and inflammation. It has been shown that inhibition of glycogen synthase kinase (GSK)-3 activity, a critical component of the Wnt and PI3K/PKB pathway, resulted in anti-inflammatory effects in a septic shock model 197. Inhibition of GSK-3 not only suppressed proinflammatory cytokine cascades, but also induced naturally suppressive or anti-inflammatory cytokines.

24 Introduction

6. Aim and outline of this thesis More potent and less toxic drugs are required in the IBD field. This emphasizes the need for understanding the underlying mechanisms of therapeutic interventions. The aim of this thesis was to characterize the interactions between cellular signal transduction and immunosuppressive agents in order to define molecular targets for new therapies. Conventional IBD treatment consists of glucocorticoids and immunomodulatory agents. More recently developed targeted immune therapies include biologicals and small molecules, the latter which specifically interfere with intracellular molecules. Recently, the treatment paradigm for CD was shifted demonstrating that induction therapy with infliximab and azathioprine is superior to the conventional step-up approach to induce clinical remission in recent-onset CD. In chapter 2, we discuss molecular mechanisms which could be responsible for the differences in clinical outcome between step-up and top-down therapy. In addition, we discuss advances in our knowledge of inflammatory signal transduction which has resulted in the development of small molecules. In chapter 3, we have defined c-Raf as a cellular target of semapimod, which demonstrated potent clinical effects in severe CD patients. We here describe a novel inflammatory signaling pathway in the pathogenesis of CD, which opens the possibility for the development of a new class of anti-inflammatory agents (chapter 3). GCs are powerful immunosuppressive drugs that mediate genomic effects via GC receptors. Evidence is increasing for clinically relevant nongenomic GC effects, but the responsible mechanisms remain to be characterized. The work presented in chapter 4, defined a nongenomic GC-induced immunosuppressive mechanism in T cells, mediated through Lck and Fyn. In chapter 5, the molecular mechanism is described responsible for these observations. We here show that GCs inhibit proximal TCR signaling through dissociation of membrane-bound GC receptor multiprotein complexes containing Lck and Fyn. Chapter 6 provides molecular insight into GC-induced insulin resistance, an important side effect of GC therapy. These studies indicated that the insulin receptor is a direct GC target which underscores the importance of the development of GC analogues which retain their immunosuppressive capacities without having an accompanying effect on insulin receptor activity. Chapter 7 reveals the presence of active STAT5 in GC resistant UC patients. This work suggests that STAT5 may represent a therapeutic target for the clinical management of steroid resistant UC.

25 Chapter 1

Reference List

(1) Sands BE. From symptom to diagnosis: clinical distinctions among various forms of intestinal inflammation. Gastroenterology. 2004;126:1518-1532. (2) Podolsky DK. Inflammatory bowel disease. N Engl J Med. 2002;347:417-429. (3) Chutkan RK. Inflammatory bowel disease. Prim Care. 2001;28:539-56, vi. (4) Hyams JS. Inflammatory bowel disease. Pediatr Rev. 2000;21:291-295. (5) Winesett M. Inflammatory bowel disease in children and adolescents. Pediatr Ann. 1997;26:227-234. (6) Bonen DK, Cho JH. The genetics of inflammatory bowel disease. Gastroenterology. 2003;124:521-536. (7) Hugot JP, Chamaillard M, Zouali H et al. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature. 2001;411:599-603. (8) Ogura Y, Bonen DK, Inohara N et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease. Nature. 2001;411:603-606. (9) Orholm M, Munkholm P, Langholz E et al. Familial occurrence of inflammatory bowel disease. N Engl J Med. 1991;324:84-88. (10) Orholm M, Fonager K, Sorensen HT. Risk of ulcerative colitis and Crohn's disease among offspring of patients with chronic inflammatory bowel disease. Am J Gastroenterol. 1999;94:3236-3238. (11) Satsangi J, Parkes M, Jewell DP. Genetics of ulcerative colitis. Lancet. 1996;348:624-625. (12) Satsangi J, Welsh KI, Bunce M et al. Contribution of genes of the major histocompatibility complex to susceptibility and disease phenotype in inflammatory bowel disease. Lancet. 1996;347:1212-1217. (13) Satsangi J, Parkes M, Louis E et al. Two stage genome-wide search in inflammatory bowel disease provides evidence for susceptibility loci on chromosomes 3, 7 and 12. Nat Genet. 1996;14:199-202. (14) Tysk C, Lindberg E, Jarnerot G, Floderus-Myrhed B. Ulcerative colitis and Crohn's disease in an unselected population of monozygotic and dizygotic twins. A study of heritability and the influence of smoking. Gut. 1988;29:990-996. (15) Tysk C. Genetic susceptibility in Crohn's disease--review of clinical studies. Eur J Surg. 1998;164:893- 896. (16) Weterman IT, Pena AS. Familial incidence of Crohn's disease in The Netherlands and a review of the literature. Gastroenterology. 1984;86:449-452. (17) Sartor RB. Current concepts of the etiology and pathogenesis of ulcerative colitis and Crohn's disease. Gastroenterol Clin North Am. 1995;24:475-507. (18) Shanahan F. Crohn's disease. Lancet. 2002;359:62-69. (19) D'Haens GR, Geboes K, Peeters M et al. Early lesions of recurrent Crohn's disease caused by infusion of intestinal contents in excluded ileum. Gastroenterology. 1998;114:262-267. (20) Laroux FS, Pavlick KP, Wolf RE, Grisham MB. Dysregulation of intestinal mucosal immunity: implications in inflammatory bowel disease. News Physiol Sci. 2001;16:272-277. (21) Mowat AM, Viney JL. The anatomical basis of intestinal immunity. Immunol Rev. 1997;156:145-166. (22) Mowat AM. Anatomical basis of tolerance and immunity to intestinal antigens. Nat Rev Immunol. 2003;3:331-341. (23) Cassatella MA. The production of cytokines by polymorphonuclear neutrophils. Immunol Today. 1995;16:21-26. (24) Iwasaki A, Kelsall BL. Mucosal immunity and inflammation. I. Mucosal dendritic cells: their specialized role in initiating T cell responses. Am J Physiol. 1999;276:G1074-G1078. (25) Neurath MF, Finotto S, Glimcher LH. The role of Th1/Th2 polarization in mucosal immunity. Nat Med. 2002;8:567-573. (26) Romagnani S. Th1/Th2 cells. Inflamm Bowel Dis. 1999;5:285-294. (27) Kakazu T, Hara J, Matsumoto T et al. Type 1 T-helper cell predominance in granulomas of Crohn's disease. Am J Gastroenterol. 1999;94:2149-2155. (28) Parronchi P, Romagnani P, Annunziato F et al. Type 1 T-helper cell predominance and interleukin-12 expression in the gut of patients with Crohn's disease. Am J Pathol. 1997;150:823-832. (29) Fuss IJ, Neurath M, Boirivant M et al. Disparate CD4+ lamina propria (LP) lymphokine secretion profiles in inflammatory bowel disease. Crohn's disease LP cells manifest increased secretion of IFN- gamma, whereas ulcerative colitis LP cells manifest increased secretion of IL-5. J Immunol. 1996;157:1261-1270. (30) Inoue S, Matsumoto T, Iida M et al. Characterization of cytokine expression in the rectal mucosa of ulcerative colitis: correlation with disease activity. Am J Gastroenterol. 1999;94:2441-2446.

26 Introduction

(31) Abreu MT. The pathogenesis of inflammatory bowel disease: translational implications for clinicians. Curr Gastroenterol Rep. 2002;4:481-489. (32) Hart AL, Al Hassi HO, Rigby RJ et al. Characteristics of intestinal dendritic cells in inflammatory bowel diseases. Gastroenterology. 2005;129:50-65. (33) Hollander D, Vadheim CM, Brettholz E et al. Increased intestinal permeability in patients with Crohn's disease and their relatives. A possible etiologic factor. Ann Intern Med. 1986;105:883-885. (34) Plevy S. The immunology of inflammatory bowel disease. Gastroenterol Clin North Am. 2002;31:77- 92. (35) Cario E, Podolsky DK. Intestinal epithelial TOLLerance versus inTOLLerance of commensals. Mol Immunol. 2005;42:887-893. (36) Travis SP, Stange EF, Lemann M et al. European evidence based consensus on the diagnosis and management of Crohn's disease: current management. Gut. 2006;55 Suppl 1:i16-i35. (37) Sandborn WJ, Hanauer SB. Systematic review: the pharmacokinetic profiles of oral mesalazine formulations and mesalazine pro-drugs used in the management of ulcerative colitis. Aliment Pharmacol Ther. 2003;17:29-42. (38) Camma C, Giunta M, Rosselli M, Cottone M. Mesalamine in the maintenance treatment of Crohn's disease: a meta-analysis adjusted for confounding variables. Gastroenterology. 1997;113:1465-1473. (39) Lochs H, Mayer M, Fleig WE et al. Prophylaxis of postoperative relapse in Crohn's disease with mesalamine: European Cooperative Crohn's Disease Study VI. Gastroenterology. 2000;118:264-273. (40) Sandborn WJ, Feagan BG. Review article: mild to moderate Crohn's disease--defining the basis for a new treatment algorithm. Aliment Pharmacol Ther. 2003;18:263-277. (41) Rousseaux C, Lefebvre B, Dubuquoy L et al. Intestinal antiinflammatory effect of 5-aminosalicylic acid is dependent on peroxisome proliferator-activated receptor-gamma. J Exp Med. 2005;201:1205-1215. (42) Summers RW, Switz DM, Sessions JT, Jr. et al. National Cooperative Crohn's Disease Study: results of drug treatment. Gastroenterology. 1979;77:847-869. (43) Stein RB, Hanauer SB. Medical therapy for inflammatory bowel disease. Gastroenterol Clin North Am. 1999;28:297-321. (44) Ardizzone S, Molteni P, Imbesi V, Bollani S, Bianchi PG. Azathioprine in steroid-resistant and steroid- dependent ulcerative colitis. J Clin Gastroenterol. 1997;25:330-333. (45) Landi B, Anh TN, Cortot A et al. Endoscopic monitoring of Crohn's disease treatment: a prospective, randomized clinical trial. The Groupe d'Etudes Therapeutiques des Affections Inflammatoires Digestives. Gastroenterology. 1992;102:1647-1653. (46) Malchow H, Ewe K, Brandes JW et al. European Cooperative Crohn's Disease Study (ECCDS): results of drug treatment. Gastroenterology. 1984;86:249-266. (47) Modigliani R, Mary JY, Simon JF et al. Clinical, biological, and endoscopic picture of attacks of Crohn's disease. Evolution on prednisolone. Groupe d'Etude Therapeutique des Affections Inflammatoires Digestives. Gastroenterology. 1990;98:811-818. (48) Modigliani R, Colombel JF, Dupas JL et al. Mesalamine in Crohn's disease with steroid-induced remission: effect on steroid withdrawal and remission maintenance, Groupe d'Etudes Therapeutiques des Affections Inflammatoires Digestives. Gastroenterology. 1996;110:688-693. (49) Munkholm P, Langholz E, Davidsen M, Binder V. Frequency of glucocorticoid resistance and dependency in Crohn's disease. Gut. 1994;35:360-362. (50) Farrell RJ, Kelleher D. Glucocorticoid resistance in inflammatory bowel disease. J Endocrinol. 2003;178:339-346. (51) Probert CS, Hearing SD, Schreiber S et al. Infliximab in moderately severe glucocorticoid resistant ulcerative colitis: a randomised controlled trial. Gut. 2003;52:998-1002. (52) Ardizzone S, Maconi G, Russo A et al. Randomised controlled trial of azathioprine and 5- aminosalicylic acid for treatment of steroid dependent ulcerative colitis. Gut. 2006;55:47-53. (53) Pearson DC, May GR, Fick GH, Sutherland LR. Azathioprine and 6-mercaptopurine in Crohn disease. A meta-analysis. Ann Intern Med. 1995;123:132-142. (54) Bouhnik Y, Lemann M, Mary JY et al. Long-term follow-up of patients with Crohn's disease treated with azathioprine or 6-mercaptopurine. Lancet. 1996;347:215-219. (55) Candy S, Wright J, Gerber M et al. A controlled double blind study of azathioprine in the management of Crohn's disease. Gut. 1995;37:674-678. (56) D'Haens G, Geboes K, Ponette E, Penninckx F, Rutgeerts P. Healing of severe recurrent ileitis with azathioprine therapy in patients with Crohn's disease. Gastroenterology. 1997;112:1475-1481.

27 Chapter 1

(57) Lewis JD, Schwartz JS, Lichtenstein GR. Azathioprine for maintenance of remission in Crohn's disease: benefits outweigh the risk of lymphoma. Gastroenterology. 2000;118:1018-1024. (58) Present DH, Korelitz BI, Wisch N et al. Treatment of Crohn's disease with 6-mercaptopurine. A long- term, randomized, double-blind study. N Engl J Med. 1980;302:981-987. (59) Tiede I, Fritz G, Strand S et al. CD28-dependent Rac1 activation is the molecular target of azathioprine in primary human CD4+ T lymphocytes. J Clin Invest. 2003;111:1133-1145. (60) Present DH, Meltzer SJ, Krumholz MP, Wolke A, Korelitz BI. 6-Mercaptopurine in the management of inflammatory bowel disease: short- and long-term toxicity. Ann Intern Med. 1989;111:641-649. (61) Brynskov J, Freund L, Rasmussen SN et al. A placebo-controlled, double-blind, randomized trial of cyclosporine therapy in active chronic Crohn's disease. N Engl J Med. 1989;321:845-850. (62) Campbell S, Travis S, Jewell D. Ciclosporin use in acute ulcerative colitis: a long-term experience. Eur J Gastroenterol Hepatol. 2005;17:79-84. (63) Durai D, Hawthorne AB. Review article: how and when to use ciclosporin in ulcerative colitis. Aliment Pharmacol Ther. 2005;22:907-916. (64) Egan LJ, Sandborn WJ, Tremaine WJ. Clinical outcome following treatment of refractory inflammatory and fistulizing Crohn's disease with intravenous cyclosporine. Am J Gastroenterol. 1998;93:442-448. (65) Hermida-Rodriguez C, Cantero PJ, Garcia-Valriberas R, Pajares Garcia JM, Mate-Jimenez J. High-dose intravenous cyclosporine in steroid refractory attacks of inflammatory bowel disease. Hepatogastroenterology. 1999;46:2265-2268. (66) Ierardi E, Principi M, Francavilla R et al. Oral tacrolimus long-term therapy in patients with Crohn's disease and steroid resistance. Aliment Pharmacol Ther. 2001;15:371-377. (67) Sandborn WJ, Present DH, Isaacs KL et al. Tacrolimus for the treatment of fistulas in patients with Crohn's disease: a randomized, placebo-controlled trial. Gastroenterology. 2003;125:380-388. (68) Santos JV, Baudet JA, Casellas FJ et al. Intravenous cyclosporine for steroid-refractory attacks of Crohn's disease. Short- and long-term results. J Clin Gastroenterol. 1995;20:207-210. (69) Stange EF, Modigliani R, Pena AS et al. European trial of cyclosporine in chronic active Crohn's disease: a 12-month study. The European Study Group. Gastroenterology. 1995;109:774-782. (70) Feagan BG, Rochon J, Fedorak RN et al. Methotrexate for the treatment of Crohn's disease. The North American Crohn's Study Group Investigators. N Engl J Med. 1995;332:292-297. (71) Feagan BG, Fedorak RN, Irvine EJ et al. A comparison of methotrexate with placebo for the maintenance of remission in Crohn's disease. North American Crohn's Study Group Investigators. N Engl J Med. 2000;342:1627-1632. (72) Cummings JR, Herrlinger KR, Travis SP et al. Oral methotrexate in ulcerative colitis. Aliment Pharmacol Ther. 2005;21:385-389. (73) Paoluzi OA, Pica R, Iacopini F, Paoluzi P. Low-dose methotrexate in ulcerative colitis: still a matter of debate. Eur J Gastroenterol Hepatol. 2004;16:1071-1072. (74) Chan ES, Cronstein BN. Molecular action of methotrexate in inflammatory diseases. Arthritis Res. 2002;4:266-273. (75) Genestier L, Paillot R, Fournel S et al. Immunosuppressive properties of methotrexate: apoptosis and clonal deletion of activated peripheral T cells. J Clin Invest. 1998;102:322-328. (76) Kozarek RA, Patterson DJ, Gelfand MD et al. Methotrexate induces clinical and histologic remission in patients with refractory inflammatory bowel disease. Ann Intern Med. 1989;110:353-356. (77) Ardizzone S, Bianchi PG. Biologic therapy for inflammatory bowel disease. Drugs. 2005;65:2253- 2286. (78) D'Haens G, Van Deventer S, Van Hogezand R et al. Endoscopic and histological healing with infliximab anti-tumor necrosis factor antibodies in Crohn's disease: A European multicenter trial. Gastroenterology. 1999;116:1029-1034. (79) Present DH, Rutgeerts P, Targan S et al. Infliximab for the treatment of fistulas in patients with Crohn's disease. N Engl J Med. 1999;340:1398-1405. (80) Rutgeerts P, D'Haens G, Targan S et al. Efficacy and safety of retreatment with anti-tumor necrosis factor antibody (infliximab) to maintain remission in Crohn's disease. Gastroenterology. 1999;117:761- 769. (81) Rutgeerts P. Infliximab is the drug we have been waiting for in Crohn's disease. Inflamm Bowel Dis. 2000;6:132-136.

28 Introduction

(82) Targan SR, Hanauer SB, van Deventer SJ et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn's disease. Crohn's Disease cA2 Study Group. N Engl J Med. 1997;337:1029-1035. (83) Kirman I, Whelan RL, Nielsen OH. Infliximab: mechanism of action beyond TNF-alpha neutralization in inflammatory bowel disease. Eur J Gastroenterol Hepatol. 2004;16:639-641. (84) Shen C, Maerten P, Geboes K et al. Infliximab induces apoptosis of monocytes and T lymphocytes in a human-mouse chimeric model. Clin Immunol. 2005;115:250-259. (85) Van den Brande JM, Braat H, van den Brink GR et al. Infliximab but not etanercept induces apoptosis in lamina propria T-lymphocytes from patients with Crohn's disease. Gastroenterology. 2003;124:1774- 1785. (86) Zeissig S, Bojarski C, Buergel N et al. Downregulation of epithelial apoptosis and barrier repair in active Crohn's disease by tumour necrosis factor alpha antibody treatment. Gut. 2004;53:1295-1302. (87) Rutgeerts P, Sandborn WJ, Feagan BG et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med. 2005;353:2462-2476. (88) ten Hove T, van Montfrans C, Peppelenbosch MP, van Deventer SJ. Infliximab treatment induces apoptosis of lamina propria T lymphocytes in Crohn's disease. Gut. 2002;50:206-211. (89) Colombel JF, Loftus EV, Jr., Tremaine WJ et al. The safety profile of infliximab in patients with Crohn's disease: the Mayo clinic experience in 500 patients. Gastroenterology. 2004;126:19-31. (90) Ljung T, Karlen P, Schmidt D et al. Infliximab in inflammatory bowel disease: clinical outcome in a population based cohort from Stockholm County. Gut. 2004;53:849-853. (91) Fefferman DS, Farrell RJ. Immunogenicity of biological agents in inflammatory bowel disease. Inflamm Bowel Dis. 2005;11:497-503. (92) Ghosh S, Goldin E, Gordon FH et al. Natalizumab for active Crohn's disease. N Engl J Med. 2003;348:24-32. (93) Gordon FH, Lai CW, Hamilton MI et al. A randomized placebo-controlled trial of a humanized monoclonal antibody to alpha4 integrin in active Crohn's disease. Gastroenterology. 2001;121:268-274. (94) Gordon FH, Hamilton MI, Donoghue S et al. A pilot study of treatment of active ulcerative colitis with natalizumab, a humanized monoclonal antibody to alpha-4 integrin. Aliment Pharmacol Ther. 2002;16:699-705. (95) Keeley KA, Rivey MP, Allington DR. Natalizumab for the treatment of multiple sclerosis and Crohn's disease. Ann Pharmacother. 2005;39:1833-1843. (96) Mathews AW, Westphal SP. Tricky FDA debate: should a risky drug be approved again? Wall St J (East Ed). 2006;B1, B4. (97) Chaudhuri A. Lessons for clinical trials from natalizumab in multiple sclerosis. BMJ. 2006;332:416- 419. (98) Creed TJ, Norman MR, Probert CS et al. Basiliximab (anti-CD25) in combination with steroids may be an effective new treatment for steroid-resistant ulcerative colitis. Aliment Pharmacol Ther. 2003;18:65- 75. (99) Mannon PJ, Fuss IJ, Mayer L et al. Anti-interleukin-12 antibody for active Crohn's disease. N Engl J Med. 2004;351:2069-2079. (100) Van Assche G, Dalle I, Noman M et al. A pilot study on the use of the humanized anti-interleukin-2 receptor antibody daclizumab in active ulcerative colitis. Am J Gastroenterol. 2003;98:369-376. (101) Korzenik JR, Dieckgraefe BK, Valentine JF, Hausman DF, Gilbert MJ. Sargramostim for active Crohn's disease. N Engl J Med. 2005;352:2193-2201. (102) Sainathan SK, Tu L, Bishnupuri KS et al. PEGylated murine Granulocyte-macrophage colony- stimulating factor: production, purification, and characterization. Protein Expr Purif. 2005;44:94-103. (103) Miner P, Wedel M, Bane B, Bradley J. An enema formulation of alicaforsen, an antisense inhibitor of intercellular adhesion molecule-1, in the treatment of chronic, unremitting pouchitis. Aliment Pharmacol Ther. 2004;19:281-286. (104) Travis S. Recent advances in immunomodulation in the treatment of inflammatory bowel disease. Eur J Gastroenterol Hepatol. 2003;15:215-218. (105) Yacyshyn BR, Chey WY, Goff J et al. Double blind, placebo controlled trial of the remission inducing and steroid sparing properties of an ICAM-1 antisense oligodeoxynucleotide, alicaforsen (ISIS 2302), in active steroid dependent Crohn's disease. Gut. 2002;51:30-36. (106) Pratt WB, Galigniana MD, Morishima Y, Murphy PJ. Role of molecular chaperones in steroid receptor action. Essays Biochem. 2004;40:41-58.

29 Chapter 1

(107) Morishima Y, Murphy PJ, Li DP, Sanchez ER, Pratt WB. Stepwise assembly of a glucocorticoid receptor.hsp90 heterocomplex resolves two sequential ATP-dependent events involving first hsp70 and then hsp90 in opening of the steroid binding pocket. J Biol Chem. 2000;275:18054-18060. (108) Bamberger CM, Schulte HM, Chrousos GP. Molecular determinants of glucocorticoid receptor function and tissue sensitivity to glucocorticoids. Endocr Rev. 1996;17:245-261. (109) Cato AC, Wade E. Molecular mechanisms of anti-inflammatory action of glucocorticoids. Bioessays. 1996;18:371-378. (110) Leung DY, Bloom JW. Update on glucocorticoid action and resistance. J Allergy Clin Immunol. 2003;111:3-22. (111) Barnes PJ, Adcock I. Anti-inflammatory actions of steroids: molecular mechanisms. Trends Pharmacol Sci. 1993;14:436-441. (112) Dostert A, Heinzel T.Negative glucocorticoid response elements and their role in glucocorticoid action. Curr Pharm Des. 2004;10:2807-2816. (113) Aziz KE, Wakefield D. Modulation of endothelial cell expression of ICAM-1, E-selectin, and VCAM-1 by beta-estradiol, progesterone, and dexamethasone. Cell Immunol. 1996;167:79-85. (114) Cronstein BN, Kimmel SC, Levin RI, Martiniuk F, Weissmann G. A mechanism for the antiinflammatory effects of corticosteroids: the glucocorticoid receptor regulates leukocyte adhesion to endothelial cells and expression of endothelial-leukocyte adhesion molecule 1 and intercellular adhesion molecule 1. Proc Natl Acad Sci U S A. 1992;89:9991-9995. (115) Osborn L, Hession C, Tizard R et al. Direct expression cloning of vascular cell adhesion molecule 1, a cytokine-induced endothelial protein that binds to lymphocytes. Cell. 1989;59:1203-1211. (116) van de SA, Caldenhoven E, Raaijmakers JA, van der Saag PT, Koenderman L. Glucocorticoid- mediated repression of intercellular adhesion molecule-1 expression in human monocytic and bronchial epithelial cell lines. Am J Respir Cell Mol Biol. 1993;8:340-347. (117) van de SA, Caldenhoven E, Stade BG et al. 12-O-tetradecanoylphorbol-13-acetate- and tumor necrosis factor alpha-mediated induction of intercellular adhesion molecule-1 is inhibited by dexamethasone. Functional analysis of the human intercellular adhesion molecular-1 promoter. J Biol Chem. 1994;269:6185-6192. (118) Barnes PJ, Adcock I. Anti-inflammatory actions of steroids: molecular mechanisms. Trends Pharmacol Sci. 1993;14:436-441. (119) Cohen JJ, Duke RC. Glucocorticoid activation of a calcium-dependent endonuclease in thymocyte nuclei leads to cell death. J Immunol. 1984;132:38-42. (120) Franchimont D. Overview of the actions of glucocorticoids on the immune response: a good model to characterize new pathways of immunosuppression for new treatment strategies. Ann N Y Acad Sci. 2004;1024:124-137. (121) Cidlowski JA, King KL, Evans-Storms RB et al. The biochemistry and molecular biology of glucocorticoid-induced apoptosis in the immune system. Recent Prog Horm Res. 1996;51:457-490. (122) Ramdas J, Harmon JM. Glucocorticoid-induced apoptosis and regulation of NF-kappaB activity in human leukemic T cells. Endocrinology. 1998;139:3813-3821. (123) Auphan N, DiDonato JA, Rosette C, Helmberg A, Karin M. Immunosuppression by glucocorticoids: inhibition of NF-kappa B activity through induction of I kappa B synthesis. Science. 1995;270:286-290. (124) Scheinman RI, Cogswell PC, Lofquist AK, Baldwin AS, Jr. Role of transcriptional activation of I kappa B alpha in mediation of immunosuppression by glucocorticoids. Science. 1995;270:283-286. (125) Wikstrom AC. Glucocorticoid action and novel mechanisms of steroid resistance: role of glucocorticoid receptor-interacting proteins for glucocorticoid responsiveness. J Endocrinol. 2003;178:331-337. (126) Galon J, Franchimont D, Hiroi N et al. Gene profiling reveals unknown enhancing and suppressive actions of glucocorticoids on immune cells. FASEB J. 2002;16:61-71. (127) Buttgereit F, Wehling M, Burmester GR. A new hypothesis of modular glucocorticoid actions: steroid treatment of rheumatic diseases revisited. Arthritis Rheum. 1998;41:761-767. (128) Buttgereit F. Mechanisms and clinical relevance of nongenomic glucocorticoid actions. Z Rheumatol. 2000;59 Suppl 2:II/119-II/123. (129) Buttgereit F, Scheffold A. Rapid glucocorticoid effects on immune cells. Steroids. 2002;67:529-534. (130) Cato AC, Nestl A, Mink S. Rapid actions of steroid receptors in cellular signaling pathways. Sci STKE. 2002;2002:RE9. (131) Goulding NJ. The molecular complexity of glucocorticoid actions in inflammation - a four-ring circus. Curr Opin Pharmacol. 2004;4:629-636.

30 Introduction

(132) Falkenstein E, Tillmann HC, Christ M, Feuring M, Wehling M. Multiple actions of steroid hormones--a focus on rapid, nongenomic effects. Pharmacol Rev. 2000;52:513-556. (133) Rhen T, Cidlowski JA. Antiinflammatory action of glucocorticoids--new mechanisms for old drugs. N Engl J Med. 2005;353:1711-1723. (134) Wehling M. Specific, nongenomic actions of steroid hormones. Annu Rev Physiol. 1997;59:365-393. (135) Buttgereit F, Straub RH, Wehling M, Burmester GR. Glucocorticoids in the treatment of rheumatic diseases: an update on the mechanisms of action. Arthritis Rheum. 2004;50:3408-3417. (136) Maier C, Runzler D, Schindelar J et al. G-protein-coupled glucocorticoid receptors on the pituitary cell membrane. J Cell Sci. 2005;118:3353-3361. (137) Bartholome B, Spies CM, Gaber T et al. Membrane glucocorticoid receptors (mGCR) are expressed in normal human peripheral blood mononuclear cells and up-regulated after in vitro stimulation and in patients with rheumatoid arthritis. FASEB J. 2004;18:70-80. (138) Croxtall JD, Choudhury Q, Flower RJ. Glucocorticoids act within minutes to inhibit recruitment of signalling factors to activated EGF receptors through a receptor-dependent, transcription-independent mechanism. Br J Pharmacol. 2000;130:289-298. (139) Hafezi-Moghadam A, Simoncini T, Yang E et al. Acute cardiovascular protective effects of corticosteroids are mediated by non-transcriptional activation of endothelial nitric oxide synthase. Nat Med. 2002;8:473-479. (140) Limbourg FP, Huang Z, Plumier JC et al. Rapid nontranscriptional activation of endothelial nitric oxide synthase mediates increased cerebral blood flow and stroke protection by corticosteroids. J Clin Invest. 2002;110:1729-1738. (141) Arbabi S, Maier RV. Mitogen-activated protein kinases. Crit Care Med. 2002;30:S74-S79. (142) Robinson MJ, Cobb MH. Mitogen-activated protein kinase pathways. Curr Opin Cell Biol. 1997;9:180- 186. (143) Hommes D, van den BB, Plasse T et al. Inhibition of stress-activated MAP kinases induces clinical improvement in moderate to severe Crohn's disease. Gastroenterology. 2002;122:7-14. (144) Waetzig GH, Seegert D, Rosenstiel P, Nikolaus S, Schreiber S. p38 mitogen-activated protein kinase is activated and linked to TNF-alpha signaling in inflammatory bowel disease. J Immunol. 2002;168:5342-5351. (145) Waetzig GH, Rosenstiel P, Nikolaus S, Seegert D, Schreiber S. Differential p38 mitogen-activated protein kinase target phosphorylation in responders and nonresponders to infliximab. Gastroenterology. 2003;125:633-634. (146) Hommes DW, Peppelenbosch MP, van Deventer SJ. Mitogen activated protein (MAP) kinase signal transduction pathways and novel anti-inflammatory targets. Gut. 2003;52:144-151. (147) Darnell JE, Jr. STATs and gene regulation. Science. 1997;277:1630-1635. (148) Leaman DW, Leung S, Li X, Stark GR. Regulation of STAT-dependent pathways by growth factors and cytokines. FASEB J. 1996;10:1578-1588. (149) Rane SG, Reddy EP. Janus kinases: components of multiple signaling pathways. Oncogene. 2000;19:5662-5679. (150) Starr R, Hilton DJ. Negative regulation of the JAK/STAT pathway. Bioessays. 1999;21:47-52. (151) Mudter J, Weigmann B, Bartsch B et al. Activation pattern of signal transducers and activators of transcription (STAT) factors in inflammatory bowel diseases. Am J Gastroenterol. 2005;100:64-72. (152) Schreiber S, Rosenstiel P, Hampe J et al. Activation of signal transducer and activator of transcription (STAT) 1 in human chronic inflammatory bowel disease. Gut. 2002;51:379-385. (153) Lovato P, Brender C, Agnholt J et al. Constitutive STAT3 activation in intestinal T cells from patients with Crohn's disease. J Biol Chem. 2003;278:16777-16781. (154) Musso A, Dentelli P, Carlino A et al. Signal transducers and activators of transcription 3 signaling pathway: an essential mediator of inflammatory bowel disease and other forms of intestinal inflammation. Inflamm Bowel Dis. 2005;11:91-98. (155) Jo D, Liu D, Yao S, Collins RD, Hawiger J. Intracellular protein therapy with SOCS3 inhibits inflammation and apoptosis. Nat Med. 2005;11:892-898. (156) Ogata H, Chinen T, Yoshida T et al. Loss of SOCS3 in the liver promotes fibrosis by enhancing STAT3-mediated TGF-beta1 production. Oncogene. 2006. (157) Dondi E, Rogge L, Lutfalla G, Uze G, Pellegrini S. Down-modulation of responses to type I IFN upon T cell activation. J Immunol. 2003;170:749-756.

31 Chapter 1

(158) Baeuerle PA, Henkel T. Function and activation of NF-kappa B in the immune system. Annu Rev Immunol. 1994;12:141-179. (159) Siebenlist U, Franzoso G, Brown K. Structure, regulation and function of NF-kappa B. Annu Rev Cell Biol. 1994;10:405-455. (160) Neurath MF, Becker C, Barbulescu K. Role of NF-kappaB in immune and inflammatory responses in the gut. Gut. 1998;43:856-860. (161) Jobin C, Haskill S, Mayer L, Panja A, Sartor RB. Evidence for altered regulation of I kappa B alpha degradation in human colonic epithelial cells. J Immunol. 1997;158:226-234. (162) Neurath MF, Pettersson S, Meyer zum Buschenfelde KH, Strober W. Local administration of antisense phosphorothioate oligonucleotides to the p65 subunit of NF-kappa B abrogates established experimental colitis in mice. Nat Med. 1996;2:998-1004. (163) Neurath MF, Fuss I, Schurmann G et al. Cytokine gene transcription by NF-kappa B family members in patients with inflammatory bowel disease. Ann N Y Acad Sci. 1998;859:149-159. (164) Rogler G, Brand K, Vogl D et al. Nuclear factor kappaB is activated in macrophages and epithelial cells of inflamed intestinal mucosa. Gastroenterology. 1998;115:357-369. (165) Schreiber S, Nikolaus S, Hampe J. Activation of nuclear factor kappa B inflammatory bowel disease. Gut. 1998;42:477-484. (166) Yamamoto Y, Gaynor RB. Therapeutic potential of inhibition of the NF-kappaB pathway in the treatment of inflammation and cancer. J Clin Invest. 2001;107:135-142. (167) Epinat JC, Gilmore TD. Diverse agents act at multiple levels to inhibit the Rel/NF-kappaB signal transduction pathway. Oncogene. 1999;18:6896-6909. (168) Bijlsma JW, Saag KG, Buttgereit F, da Silva JA. Developments in glucocorticoid therapy. Rheum Dis Clin North Am. 2005;31:1-17, vii. (169) Li Q, Verma IM. NF-kappaB regulation in the immune system. Nat Rev Immunol. 2002;2:725-734. (170) Jobin C, Sartor RB. NF-kappaB signaling proteins as therapeutic targets for inflammatory bowel diseases. Inflamm Bowel Dis. 2000;6:206-213. (171) Diks SH, Kok K, O'Toole T et al. Kinome profiling for studying lipopolysaccharide signal transduction in human peripheral blood mononuclear cells. J Biol Chem. 2004;279:49206-49213. (172) Houseman BT, Huh JH, Kron SJ, Mrksich M. Peptide chips for the quantitative evaluation of protein kinase activity. Nat Biotechnol. 2002;20:270-274. (173) Cohen P. Protein kinases--the major drug targets of the twenty-first century? Nat Rev Drug Discov. 2002;1:309-315. (174) Dancey J, Sausville EA. Issues and progress with protein kinase inhibitors for cancer treatment. Nat Rev Drug Discov. 2003;2:296-313. (175) Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The protein kinase complement of the human genome. Science. 2002;298:1912-1934. (176) Fabian MA, Biggs WH, III, Treiber DK et al. A small molecule-kinase interaction map for clinical kinase inhibitors. Nat Biotechnol. 2005;23:329-336. (177) Lowenberg M, Peppelenbosch MP, Hommes DW. Therapeutic modulation of signal transduction pathways. Inflamm Bowel Dis. 2004;10 Suppl 1:S52-S57. (178) Saklatvala J. The p38 MAP kinase pathway as a therapeutic target in inflammatory disease. Curr Opin Pharmacol. 2004;4:372-377. (179) Waetzig GH, Schreiber S. Review article: mitogen-activated protein kinases in chronic intestinal inflammation - targeting ancient pathways to treat modern diseases. Aliment Pharmacol Ther. 2003;18:17-32. (180) Lee JC, Kassis S, Kumar S, Badger A, Adams JL. p38 mitogen-activated protein kinase inhibitors-- mechanisms and therapeutic potentials. Pharmacol Ther. 1999;82:389-397. (181) Lee JC, Kumar S, Griswold DE et al. Inhibition of p38 MAP kinase as a therapeutic strategy. Immunopharmacology. 2000;47:185-201. (182) Salituro FG, Germann UA, Wilson KP et al. Inhibitors of p38 MAP kinase: therapeutic intervention in cytokine-mediated diseases. Curr Med Chem. 1999;6:807-823. (183) Underwood DC, Osborn RR, Bochnowicz S et al. SB 239063, a p38 MAPK inhibitor, reduces neutrophilia, inflammatory cytokines, MMP-9, and fibrosis in lung. Am J Physiol Lung Cell Mol Physiol. 2000;279:L895-L902. (184) Mclay LM, Halley F, Souness JE et al. The discovery of RPR 200765A, a p38 MAP kinase inhibitor displaying a good oral anti-arthritic efficacy. Bioorg Med Chem. 2001;9:537-554.

32 Introduction

(185) Branger J, van den BB, Weijer S et al. Anti-inflammatory effects of a p38 mitogen-activated protein kinase inhibitor during human endotoxemia. J Immunol. 2002;168:4070-4077. (186) Schreiber S, Feagan B, D'Haens G et al. Oral p38 mitogen-activated protein kinase inhibition with BIRB 796 for active Crohn's disease: a randomized, double-blind, placebo-controlled trial. Clin Gastroenterol Hepatol. 2006;4:325-334. (187) Manning AM, Davis RJ. Targeting JNK for therapeutic benefit: from junk to gold? Nat Rev Drug Discov. 2003;2:554-565. (188) Hilger RA, Scheulen ME, Strumberg D. The Ras-Raf-MEK-ERK pathway in the treatment of cancer. Onkologie. 2002;25:511-518. (189) Sebolt-Leopold JS, Dudley DT, Herrera R et al. Blockade of the MAP kinase pathway suppresses growth of colon tumors in vivo. Nat Med. 1999;5:810-816. (190) Thompson N, Lyons J. Recent progress in targeting the Raf/MEK/ERK pathway with inhibitors in cancer drug discovery. Curr Opin Pharmacol. 2005;5:350-356. (191) Egerton M, Fitzpatrick DR, Kelso A. Activation of the extracellular signal-regulated kinase pathway is differentially required for TCR-stimulated production of six cytokines in primary T lymphocytes. Int Immunol. 1998;10:223-229. (192) Liu Q, Fan J, McMahon M, Prince AM, Zhang P. Role of the oncogenic Raf-1 in orchestration of discrete nuclear factor-kappaB-activating pathways. Mol Cell Biol Res Commun. 2001;4:381-389. (193) Odabaei G, Chatterjee D, Jazirehi AR et al. Raf-1 kinase inhibitor protein: structure, function, regulation of cell signaling, and pivotal role in apoptosis. Adv Cancer Res. 2004;91:169-200. (194) Troppmair J, Rapp UR. Raf and the road to cell survival: a tale of bad spells, ring bearers and detours. Biochem Pharmacol. 2003;66:1341-1345. (195) van der BT, Nijenhuis S, van Raaij E, Verhoef J, van Asbeck BS. Lipopolysaccharide-induced tumor necrosis factor alpha production by human monocytes involves the raf-1/MEK1-MEK2/ERK1-ERK2 pathway. Infect Immun. 1999;67:3824-3829. (196) Xu XS, Vanderziel C, Bennett CF, Monia BP. A role for c-Raf kinase and Ha-Ras in cytokine-mediated induction of cell adhesion molecules. J Biol Chem. 1998;273:33230-33238. (197) Martin M, Rehani K, Jope RS, Michalek SM. Toll-like receptor-mediated cytokine production is differentially regulated by glycogen synthase kinase 3. Nat Immunol. 2005;6(8):751-2

33

Chapter 2

Update on biological and small molecule therapy for inflammatory bowel diseases

Mark Löwenberg1,3, Maikel P. Peppelenbosch2, Daniel W. Hommes3

1 Department for Experimental Internal Medicine, Academic Medical Center, Amsterdam, The Netherlands 2 Department of Cell Biology, University of Groningen, Groningen, The Netherlands 3 Department of Gastroenterology and Hepatology, Academic Medical Center, Amsterdam, The Netherlands.

Published, in part, in: Drugs (accepted for publication); Inflammatory Bowel Diseases. 2004; 10 Suppl 1:S52-7

35 Chapter 2

Summary Conventional therapies for inflammatory bowel diseases (IBD) consist of glucocorticoids (GCs), azathioprine and methotrexate. With the introduction of biologicals (such as anti- TNF/infliximab), the goals of therapy have advanced, including induction of remission with bowel healing as well as reduction in the rate of complications, surgeries, and mortality. Recently, it has been shown that induction therapy with infliximab and azathioprine (‘top- down’) in recent-onset Crohn’s disease (CD) is superior to current ‘step-up’ algorithms (i.e. repeated topical and systemic GC therapy) to induce clinical remission. Experimental studies have demonstrated that GCs are able to induce impaired apoptosis of immune cells resulting in ‘loss of tolerance’ and subsequent autoimmunity. Further research will have to determine whether GC therapy augments the mechanism of loss of tolerance in CD, which could complicate future clinical management. However, current therapies do not interfere with the underlying molecular mechanisms, are relatively non-selective and have dose-limiting side effects. For these reasons, new therapeutic modalities are needed in the IBD field. The last decade, attention has shifted towards modulation of intracellular signal transduction pathways, which play an important role in inflammatory pathology. Small molecules targeting signaling molecules are considered as a promising strategy for the clinical management of IBD.

36 Therapies for inflammatory bowel diseases

Conventional therapies and biologicals Conventional treatment of CD consists of GCs and immunomodulators, including azathioprine and methotrexate 1. GCs form the basis of current IBD therapy 2,3. However, GC therapy is complicated by frequently occurring side effects. Moreover, a significant number of patients become resistant to or dependent on GCs and need additional immunomodulatory therapy 4-8. Azathioprine and its metabolite 6-mercaptopurine (6-MP) are useful for the treatment of chronic active disease and for maintaining remission, but its use is limited by their slow onset of action and adverse events 9,10. Methotrexate has been established as an induction agent for GC-dependent CD and for maintenance of remission after successful induction 6,11, but myelosuppression and hepatotoxicity complicate the clinical use. Increased understanding of the immunopathogenesis of autoimmune diseases has resulted in the development of biologic therapies, which selectively interfere within inflammatory cascades. Several anti-tumor necrosis factor-α (TNF-α) compounds, such as infliximab, adalimumab and certolizumab, have become available for CD patients who do not respond to conventional therapies 12-15.

Loss of immune tolerance Although the pathogenesis of CD is still largely unclear, immune-mediated phenomena are obviously involved 16. The intestinal lumen contains huge quantities of non-pathogenic bacteria, which constantly interact with the host. The mucosal immune system must discriminate between commensal (harmless) and potential pathogenic micro-organisms. Hence, immune cells have to respond effectively to pathogens, whereas the presence of non- pathogenic micro-organisms has to be ignored, generally referred to as ‘immune tolerance’. This is accomplished by the innate immune system through cell surface structures that function as mammalian pattern-recognition receptors, such as toll-like receptors (TLRs) 17. TLRs specifically recognize microbes leading to adequate pathogen elimination. In order to regulate these TLR-mediated processes, several molecular mechanisms have been elucidated that ensure tolerance, such as decreased ligand recognition or inhibition of intracellular signaling 18. A fundamental aspect of commensal host-bacterial relationships in the gut is the development and maintenance of immune tolerance to the enteric flora 19. Accumulating evidence suggests that the normal indigenous flora of the intestine plays a crucial role in CD

37 Chapter 2

pathogenesis. It is currently believed that loss of tolerance against luminal commensals is a central event in this disease, which is due to altered pattern recognition 20-23, inadequate regulatory T cell function or excessive mucosal DC stimulation by the gut flora 24.

Glucocorticoid-induced loss of tolerance An alternative mechanism for maintaining immune tolerance and preventing autoimmunity is programmed cell death (apoptosis) of immune cells 25. This has been demonstrated in vitro and in vivo by systemic autoimmune diseases that result from mutations in the pro-apoptotic Fas receptor or Fas ligand genes 26. Previous studies revealed that loss of self-tolerance due to defective apoptosis of lymphocytes results in autoimmunity 27,28. In addition, it has been shown that DCs play a crucial role in maintaining immune tolerance 29-33. It was reported that DCs accumulate in autoimmune patients harbouring a deficiency in apoptosis 34, and significant expansion of DCs has been detected in Fas-deficient mice 35. These findings suggested that accumulation of DCs result in chronic lymphocyte activation and subsequent autoimmunity. Recently, it has been confirmed that DC apoptosis helps regulate self-tolerance 36. This study revealed that a defect in DC apoptosis can independently lead to autoimmunity, which is consistent with a central role for this cell-type in maintaining immune tolerance. GCs are effective as immunosuppressive therapy for a wide variety of inflammatory disorders and autoimmune pathology. Paradoxically, depending on their dose as well as on the activation state of target cells, GCs can either induce or inhibit apoptosis of immune cells 37. It has been reported that GC treatment inhibits IL-2-dependent activation-induced cell death of leukocytes, resulting in increased systemic leucocyte numbers and autoimmunity 38,39. Furthermore, GCs may increase de novo synthesis of the protein leucine zipper, which has been demonstrated to confer resistance to activation-induced apoptosis in hybridoma T cells 40. Importantly, CD pathogenesis is characterized by apoptosis defects in T cells 41,42. Altogether, GC-induced counterproductive effects on immune cell apoptosis may complicate the long-term course of CD pathology, and this could represent a clinical problem in the preferred first-line GC treatment of such patients. The responsible mechanisms for these GC- dependent effects remain only partly understood. Previous experimental studies have revealed a relationship between GCs and loss of tolerance in liver transplant and airway hyperreactivity models leading to autoimmunity. These studies indicated that immune

38 Therapies for inflammatory bowel diseases

tolerance in liver transplants is associated with apoptosis of infiltrating recipient leukocytes 43. Peri-transplant GC administration reduced apoptosis of infiltrating immune cells and prevented development of systemic tolerance resulting in reduced graft survival 43. These observations demonstrated that GC therapy is able to reduce apoptosis of immune cells resulting in loss of tolerance. Moreover, immune tolerance in respiratory organs has been described to be an important mechanism in preventing inflammation in the airway, which is mediated through antigen-specific adaptive regulatory T cells 44-46. Respiratory tolerance limits and controls immune responses against large quantities of innocuous antigens that enter the lungs. The effect of GC treatment on the development of immunological tolerance and airway hyperreactivity was studied in mice 47. It was shown that GCs prevent protective effects of respiratory tolerance on the development of airway hyperreactivity through reduced IL-10 synthesis by DCs and through impaired growth of T-regulatory cells. These outcomes indicated that GCs prevent the protective effects of respiratory tolerance on the development of airway hyperreactivity, resulting in an aggravated inflammatory response. Thus, preclinical evidence demonstrated that GC treatment is able to induce loss of tolerance (Figure 1).

Figure 1. Glucocorticoid (GC)-induced loss of tolerance. Gut flora is internalized, processed by dendritic cells (DCs), and presented to immune cells, such as T cells. The intestinal immune system tolerates the commensal flora and maintains mucosal homeostasis (immune tolerance), which is largely mediated by immune cells undergoing apoptosis. Loss of tolerance to commensal autologous flora results in enhanced reactivity against gut antigens and activation of the immune system. Evidence exists that GCs can break immune tolerance through interfering with immune cell apoptosis.

39 Chapter 2

In CD pathogenesis, steroid-induced loss of tolerance is a controversial issue. However, there is emerging insight into the importance of immune cell apoptosis in controlling immunological processes in the gut. The problem with GCs seems to lie in the sometimes negative effect on immune cell apoptosis (Figure 1), and further research will have to clarify the responsible molecular mechanisms. It is possible that strategies which avoid GC use and directly positively target apoptotic mechanisms may be superior compared to conventional therapeutic protocols.

Step-up versus top-down therapy for Crohn’s disease Current step-up algorithms to treat CD are based on initiation of treatment with immunosuppressives and immunomodulators (Figure 2). Biologicals are introduced when patients become refractory to conventional therapies. The advent of biologic therapy, in particular neutralizing antibodies against TNF, has dramatically changed the management of patients with more severe and/or refractory CD. A single infusion of infliximab (5 mg/kg) led to a response in more than 80% of CD patients 48. This therapy has been shown to be relatively safe and is ideally combined with continued immunomodulators in order to reduce immunogenicity 49,50. Up to now, anti-TNF therapy has been reserved for patients with refractory disease who have failed GCs and immunomodulators first (step-up approach). Recently, a more aggressive (top-down) therapeutic strategy was used early on in the disease course in order to find out whether this would lead to better clinical outcomes (Figure 2). One hundred thirty CD patients with moderate to severe disease diagnosed within 4 years and never treated with GCs, immunomodulators or biologicals, were randomized to receive either step-up treatment with repetitive topical or systemic GCs or top-down treatment with 3 infusions of infliximab (at week 0, 2, 6) together with azathioprine (2-2.5mg per kg per day) 51. In the top-down group, relapsing patients were given repeated infliximab and GCs when they failed to respond to infliximab. In the step-up group, azathioprine was added in case of repeated need for GCs or dependency and infliximab was given after failure of immunosuppression. This study demonstrated that the top-down approach in recent onset CD is superior to step-up treatment for inducing clinical remission, for avoiding steroid therapy, and for inducing endoscopic improvement and mucosal healing. Patients in the step-up arm needed prolonged exposure to GCs to control disease (25% at 6 months and 12.5% at 12

40 Therapies for inflammatory bowel diseases

months). Interestingly, it was only because of the introduction of immunomodulatory therapy that patients achieved clinical remission in the step-up arm: e.g. 40% at 6 months and 62.5% at 12 months were treated with azathioprine or methotrexate after two courses of steroids failed to control disease. In addition, two rather remarkable observations were made: 1) almost 100% of study patients, with newly-onset CD and naive to previous therapy, responded to infliximab versus only 60-70% of patients that have been treated with previous courses of GCs or immunomodulators; 2) the differences in clinical outcome of peri-anal fistulising disease was remarkable; in the top-down group the development of fistula’s was absent and in those patients presenting with fistula’s at baseline a long-lasting response was observed, as opposed to the step-up arm where fistulising disease was not controlled well. These findings suggest that treatment with anti-TNF therapy in naïve CD patients is preferable compared to an approach in which patients are repeatedly treated with conventional immunosuppressives before initiating biologic therapy. Hence, clinical observations and preclinical data suggest that GC therapy can induce loss of tolerance, which might complicate the clinical management of CD. Infliximab specifically targets one particular molecule (TNFα), a central player in the pathogenesis of CD, leading to immune cell apoptosis 52,53. Differences in mechanism of action between GCs and infliximab could contribute to the observed discrepancy in clinical outcome between step-up and top-down treatment protocols in recent-onset CD.

Figure 2. Step-up versus top-down therapy in recent-onset Crohn’s disease. Conventional therapy for Crohn’s disease is based on step-up algorithms, which initiate treatment with conventional immunosuppressives or immunomodulatory agents (i.e. corticosteroids, azathioprine (AZA), methotrexate (MTX)) and defer therapy with biologicals until patients become refractory to conventional therapies. A recent study demonstrated that induction therapy with infliximab (IFX) and azathioprine in recent-onset Crohn’s disease (top-down) is superior to the current step-up approach in order to induce clinical remission.

41 Chapter 2

Small molecules As a consequence of limited efficacy and toxicity of current IBD therapies, there is widespread interest in the development of novel drugs. The last decade, cellular signaling pathways have been elucidated that play a crucial role in IBD pathogenesis 54. As a result, attention has shifted towards modulation of signal transduction cascades resulting in the development of novel therapeutics (i.e. small molecules). Pharmaceutical intervention of mitogen activated protein kinase (MAPK) pathways has attracted widespread interest. In mammalian cells, three MAPK pathways are of importance: p38 MAPK, the cellular signal regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) (Table 1). MAPKs play an important role in cellular physiology, such as embryogenesis, cellular differentiation, proliferation and apoptosis. MAPK phosphorylation cascades consist of three sequential intracellular protein kinase activation steps and are initiated when the first member, a MAPK kinase kinase (MAPKKK) is activated upon receptor ligand interaction (Figure 3). Next, a downstream MAPK kinase (MAPKK) becomes activated, resulting in MAPK phosphorylation, and regulation of gene transcription.

Table 1. The MAPK family. Three MAPK families have been identified: i.e. ERK, JNK and p38 MAPK. ATF-2, activating transcription factor 2; ERK extra- cellular signal transduction kinase; JNK, c-Jun N terminal kinase; MAP, mitogen activated protein; MAPKAP, MAP kinase activated protein kinase; MEF, myocyte enhance factor; MNKs, MAPK- interacting kinases; MSKs, mitogen and stress activated protein kinases.

42 Therapies for inflammatory bowel diseases

P38 MAPK, JNK and ERK signal transduction pathways There have been identified five p38 MAPK isoforms sharing about 60% homology (p38α, p38β1, p38β2, p38γ, p38δ), that become activated by growth factors, lipopolysaccharide (LPS), osmotic stress and ultraviolet exposure 55. MEKK 3/6 and MKK3/6, acting as MAPKK kinases and MAPK kinases, are responsible for p38 MAPK activation (Figure 3). Downstream targets of p38 MAPK are transcription factors (such as ATF-2), other kinases or phospholipase A2 56. This pathway regulates different cellular processes, such as synthesis of inflammatory mediators, activation and recruitment of leukocytes, cellular growth, differentiation, and apoptosis. Previously, an alternative mechanism regulating the activity of MAPKs has been revealed 57. It was observed that the scaffolding protein transforming growth factor-beta-activated protein kinase 1 (TAB1), which lacks catalytic activity, could activate p38 MAPK, indicating that adaptor proteins are also able to regulate MAPK activity.

Figure 3. ERK, JNK and p38 MAPK pathways. MAPK pathways consist of 3 sequential activation steps (MAPKKK, MAPKK and MAPK), and there is considerable cross-talk between these cascades, that regulate gene transcription. MAPK, mitogen activated protein kinase; MAPKK, MAPK kinase; MAPKKK, MAPKK kinase; MEKK, MEK kinase; ASK, apoptosis signal-regulating kinase; MLK, mixed lineage kinase; MEK, mitogen-activated or extracellular signal-regulated protein kinase; MKK, MAP kinase kinase; ERK, extra-cellular signal regulated kinase; JNK, c-Jun N-terminal kinase; ATF-2, activating transcription factor 2.

43 Chapter 2

The JNK signaling pathway is activated by various cytokines, mitogens, osmotic stress and ultraviolet radiation 58. Three genes (i.e. JNK-1, JNK-2 and JNK-3) encode for JNK kinases, which are alternatively spliced to form ten different JNK isoforms. MKK4 and MKK7 function as upstream JNK activators. MAPKK kinase activators of JNK include MEKK, MLK and ASK (Figure 3). Activated JNK phosphorylates members of the activator protein-1 (AP-1) transcription factor family, such as ATF-2 and c-Jun. Besides playing an important role in TNF expression 59,60, JNK is important for cellular proliferation, differentiation and apoptosis 61. The Ras/Raf/MEK/ERK cassette was the first MAPK signal transduction pathway to be characterized (Figure 3) 62. ERK1 and ERK2, often referred to as p44 and p42 MAPK respectively, are transcribed from the same gene and become activated in response to mitogens, LPS and osmotic stress. MEK1/2 and Raf function as upstream MAPK kinases and as MAPKK kinases respectively. ERK1 and ERK2 target numerous transcription factors, thereby regulating cellular growth, differentiation, survival and cytokine production 63.

Modulation of MAPK signaling pathways as anti-inflammatory strategy MAPKs play an essential role in the pathogenesis of inflammatory responses, but their actual importance in human pathology remains poorly understood. Small molecules that impact on ERK, JNK and p38 MAPK cascades represent an attractive approach for the treatment of inflammatory pathology. Testing of selective p38 MAPK inhibitors (SB239063, RWJ67657, RPR203494) has progressed to animal models and clinical trials 64-68. The selective p38 MAPK inhibitor BIRB 796 has been tested in human endotoxemia, a model of acute inflammation. BIRB 796 administration resulted in inhibition of p38 MAPK activity and diminished experimentally induced parameters of inflammation (i.e. cytokine release and leukocyte responses) upon intravenous LPS application 69. In contrast, no clinical efficacy was seen for BIRB 796 compared to placebo in a randomized double-blind placebo-controlled trial in moderate to severe CD 70. SB203580, a pyridinyl imidazole inhibitor and widely used p38 MAPK inhibitor has been tested in TNBS-induced colitis. Remarkably, p38 MAPK inhibited animals did worse compared to solvent-treated controls 71. It was concluded that treatment of chronic inflammation requires broader strategies than p38 MAPK inhibition per se. Various JNK inhibitors (SP600125, CEP1347) showed anti-inflammatory effects in animal models of arthritis, asthma and pancreatitis 61,72-74. The anti-inflammatory small molecule

44 Therapies for inflammatory bowel diseases

semapimod (CNI-1493) suppressed JNK and p38 MAPK phosphorylation in vitro, and showed protective effects in endotoxic shock 75, allergic encephalitis 76, and arthritis 77 models. Semapimod has been tested in severe CD patients in an open label study, demonstrating clinical responses in 67% of patients with severe CD after 4 weeks and in 58% of the patients after 8 weeks 78. The drug was well-tolerated and side effects included local irritation at the infusion site (phlebitis) and mild increases in liver enzymes, both resolving spontaneously within weeks. Enhanced activity of the ERK signaling cascade has been demonstrated in cancer cells 79. MEK1 and MEK2 inhibitors (PD98059, U0126, PD184352, L783277, CI-1040) have been developed as anti-cancer therapeutics 80-82. Reduced tumor growth in mice with colon carcinomas of mouse and human origin after treatment with PD184352, an orally available MEK1/2 inhibitor, has been reported 83. A recent phase I study demonstrated that the MEK inhibitor CI-1040 was safe in patients with advanced malignancies 84. Parallel ex vivo studies revealed that the drug was able to suppress ERK phosphorylation in blood cells isolated from treated patients. Interestingly, the MEK inhibitor Ro092210 demonstrated anti-inflammatory effects. It was shown that Ro092210 treatment resulted in reduced T cell proliferation and impaired antigen induced IL-2 secretion 85. L783277 inhibited MEK activity in vitro and showed inhibitory effects on epithelial cell growth 86. In a preclinical model of brain ischaemia and LPS-induced inflammation, PD98059 administration led to significant reduced inflammatory responses 87,88. Finally, the MEK1/2 inhibitors PD98059 and U0126 demonstrated anti-inflammatory effects in cerulein-induced acute pancreatitis 89. Thus, the ERK signaling cascade is not only a survival pathway, but is also involved in inflammatory processes.

Conclusions CD has been treated for many decades using a step-up therapeutic approach, consisting of repetitive treatments with GCs, azathioprine or methotrexate. Biologicals, such as infliximab, are introduced when patients do not respond to conventional therapies. Preclinical studies suggest that GCs may limit beneficial immunological processes, such as immune tolerance (i.e. inactivation responses to commensal bacteria) through decreased immune cell apoptosis. Recently, the treatment paradigm for CD was shifted demonstrating that induction therapy

45 Chapter 2

with infliximab and azathioprine (top-down) is superior to the conventional step-up approach to induce clinical remission in recent-onset CD. Further studies are needed to find out whether GC therapy could complicate the long-term course of CD pathology. Increased understanding of inflammatory signaling pathways has led to the discovery of novel therapeutics. Experimental and clinical studies demonstrated potent anti-inflammatory effects of MAPK inhibitors. However, it is still unclear whether targeting a single signal transduction pathway will be efficacious in various disease states, and many of these emerging drugs will probably be used in combination therapy. Notably, small molecules are orally available, show selective inhibitory effects and are relatively cheap. Furthermore, no serious adverse events have been reported so far that excluded additional clinical development. For these reasons, MAPK inhibitors are considered as a promising approach for the treatment of IBD.

46 Therapies for inflammatory bowel diseases

Reference List

(1) Travis SP, Stange EF, Lemann M et al. European evidence based consensus on the diagnosis and management of Crohn's disease: current management. Gut. 2006;55 Suppl 1:i16-i35. (2) Malchow H, Ewe K, Brandes JW et al. European Cooperative Crohn's Disease Study (ECCDS): results of drug treatment. Gastroenterology. 1984;86:249-266. (3) Summers RW, Switz DM, Sessions JT, Jr. et al. National Cooperative Crohn's Disease Study: results of drug treatment. Gastroenterology. 1979;77:847-869. (4) Ardizzone S, Molteni P, Imbesi V, Bollani S, Bianchi PG. Azathioprine in steroid-resistant and steroid- dependent ulcerative colitis. J Clin Gastroenterol. 1997;25:330-333. (5) Ardizzone S, Maconi G, Russo A et al. Randomised controlled trial of azathioprine and 5- aminosalicylic acid for treatment of steroid dependent ulcerative colitis. Gut. 2006;55:47-53. (6) Feagan BG, Rochon J, Fedorak RN et al. Methotrexate for the treatment of Crohn's disease. The North American Crohn's Study Group Investigators. N Engl J Med. 1995;332:292-297. (7) Munkholm P, Langholz E, Davidsen M, Binder V. Frequency of glucocorticoid resistance and dependency in Crohn's disease. Gut. 1994;35:360-362. (8) Pearson DC, May GR, Fick GH, Sutherland LR. Azathioprine and 6-mercaptopurine in Crohn disease. A meta-analysis. Ann Intern Med. 1995;123:132-142. (9) Bouhnik Y, Lemann M, Mary JY et al. Long-term follow-up of patients with Crohn's disease treated with azathioprine or 6-mercaptopurine. Lancet. 1996;347:215-219. (10) Present DH, Meltzer SJ, Krumholz MP, Wolke A, Korelitz BI. 6-Mercaptopurine in the management of inflammatory bowel disease: short- and long-term toxicity. Ann Intern Med. 1989;111:641-649. (11) Feagan BG, Fedorak RN, Irvine EJ et al. A comparison of methotrexate with placebo for the maintenance of remission in Crohn's disease. North American Crohn's Study Group Investigators. N Engl J Med. 2000;342:1627-1632. (12) Present DH, Rutgeerts P, Targan S et al. Infliximab for the treatment of fistulas in patients with Crohn's disease. N Engl J Med. 1999;340:1398-1405. (13) Rutgeerts P, D'Haens G, Targan S et al. Efficacy and safety of retreatment with anti-tumor necrosis factor antibody (infliximab) to maintain remission in Crohn's disease. Gastroenterology. 1999;117:761- 769. (14) Rutgeerts P. Infliximab is the drug we have been waiting for in Crohn's disease. Inflamm Bowel Dis. 2000;6:132-136. (15) Ardizzone S, Bianchi PG. Biologic therapy for inflammatory bowel disease. Drugs. 2005;65:2253- 2286. (16) Laroux FS, Pavlick KP, Wolf RE, Grisham MB. Dysregulation of intestinal mucosal immunity: implications in inflammatory bowel disease. News Physiol Sci. 2001;16:272-277. (17) Takeda K, Kaisho T, Akira S. Toll-like receptors. Annu Rev Immunol. 2003;21:335-376. (18) Cario E, Podolsky DK. Intestinal epithelial TOLLerance versus inTOLLerance of commensals. Mol Immunol. 2005;42:887-893. (19) Hooper LV, Gordon JI. Commensal host-bacterial relationships in the gut. Science. 2001;292:1115- 1118. (20) Abreu MT, Arnold ET, Thomas LS et al. TLR4 and MD-2 expression is regulated by immune-mediated signals in human intestinal epithelial cells. J Biol Chem. 2002;277:20431-20437. (21) Cario E, Rosenberg IM, Brandwein SL et al. Lipopolysaccharide activates distinct signaling pathways in intestinal epithelial cell lines expressing Toll-like receptors. J Immunol. 2000;164:966-972. (22) Ortega-Cava CF, Ishihara S, Rumi MA et al. Strategic compartmentalization of Toll-like receptor 4 in the mouse gut. J Immunol. 2003;170:3977-3985. (23) Suzuki M, Hisamatsu T, Podolsky DK. Gamma interferon augments the intracellular pathway for lipopolysaccharide (LPS) recognition in human intestinal epithelial cells through coordinated up- regulation of LPS uptake and expression of the intracellular Toll-like receptor 4-MD-2 complex. Infect Immun. 2003;71:3503-3511. (24) Wen Z, Fiocchi C. Inflammatory bowel disease: autoimmune or immune-mediated pathogenesis? Clin Dev Immunol. 2004;11:195-204. (25) Rotrosen D, Matthews JB, Bluestone JA. The immune tolerance network: a new paradigm for developing tolerance-inducing therapies. J Allergy Clin Immunol. 2002;110:17-23. (26) Nagata S, Suda T. Fas and Fas ligand: lpr and gld mutations. Immunol Today. 1995;16:39-43.

47 Chapter 2

(27) Lenardo M, Chan KM, Hornung F et al. Mature T lymphocyte apoptosis--immune regulation in a dynamic and unpredictable antigenic environment. Annu Rev Immunol. 1999;17:221-253. (28) Rathmell JC, Thompson CB. Pathways of apoptosis in lymphocyte development, homeostasis, and disease. Cell. 2002;109 Suppl:S97-107. (29) Banchereau J, Pascual V, Palucka AK. Autoimmunity through cytokine-induced dendritic cell activation. Immunity. 2004;20:539-550. (30) Lanzavecchia A, Sallusto F. Regulation of T cell immunity by dendritic cells. Cell. 2001;106:263-266. (31) Liu YJ. Dendritic cell subsets and lineages, and their functions in innate and adaptive immunity. Cell. 2001;106:259-262. (32) Ludewig B, Odermatt B, Landmann S, Hengartner H, Zinkernagel RM. Dendritic cells induce autoimmune diabetes and maintain disease via de novo formation of local lymphoid tissue. J Exp Med. 1998;188:1493-1501. (33) Steinman RM, Hawiger D, Nussenzweig MC. Tolerogenic dendritic cells. Annu Rev Immunol. 2003;21:685-711. (34) Wang J, Zheng L, Lobito A et al. Inherited human Caspase 10 mutations underlie defective lymphocyte and dendritic cell apoptosis in autoimmune lymphoproliferative syndrome type II. Cell. 1999;98:47-58. (35) Fields ML, Sokol CL, Eaton-Bassiri A et al. Fas/Fas ligand deficiency results in altered localization of anti-double-stranded DNA B cells and dendritic cells. J Immunol. 2001;167:2370-2378. (36) Chen M, Wang YH, Wang Y et al. Dendritic cell apoptosis in the maintenance of immune tolerance. Science. 2006;311:1160-1164. (37) Kroemer G, Martinez C. Pharmacological inhibition of programmed lymphocyte death. Immunol Today. 1994;15:235-242. (38) Kelso A, Munck A. Glucocorticoid inhibition of lymphokine secretion by alloreactive T lymphocyte clones. J Immunol. 1984;133:784-791. (39) Arya SK, Wong-Staal F, Gallo RC. Dexamethasone-mediated inhibition of human T cell growth factor and gamma-interferon messenger RNA. J Immunol. 1984;133:273-276. (40) D'Adamio F, Zollo O, Moraca R et al. A new dexamethasone-induced gene of the leucine zipper family protects T lymphocytes from TCR/CD3-activated cell death. Immunity. 1997;7:803-812. (41) Ina K, Itoh J, Fukushima K et al. Resistance of Crohn's disease T cells to multiple apoptotic signals is associated with a Bcl-2/Bax mucosal imbalance. J Immunol. 1999;163:1081-1090. (42) Itoh J, de La MC, Strong SA, Levine AD, Fiocchi C. Decreased Bax expression by mucosal T cells favours resistance to apoptosis in Crohn's disease. Gut. 2001;49:35-41. (43) Sharland A, Yan Y, Wang C et al. Evidence that apoptosis of activated T cells occurs in spontaneous tolerance of liver allografts and is blocked by manipulations which break tolerance. Transplantation. 1999;68:1736-1745. (44) Akbari O, DeKruyff RH, Umetsu DT. Pulmonary dendritic cells producing IL-10 mediate tolerance induced by respiratory exposure to antigen. Nat Immunol. 2001;2:725-731. (45) Akbari O, Freeman GJ, Meyer EH et al. Antigen-specific regulatory T cells develop via the ICOS- ICOS-ligand pathway and inhibit allergen-induced airway hyperreactivity. Nat Med. 2002;8:1024-1032. (46) Stock P, Akbari O, Berry G et al. Induction of T helper type 1-like regulatory cells that express Foxp3 and protect against airway hyper-reactivity. Nat Immunol. 2004;5:1149-1156. (47) Stock P, Akbari O, DeKruyff RH, Umetsu DT. Respiratory tolerance is inhibited by the administration of corticosteroids. J Immunol. 2005;175:7380-7387. (48) Targan SR, Hanauer SB, van Deventer SJ et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn's disease. Crohn's Disease cA2 Study Group. N Engl J Med. 1997;337:1029-1035. (49) Modigliani R, Mary JY, Simon JF et al. Clinical, biological, and endoscopic picture of attacks of Crohn's disease. Evolution on prednisolone. Groupe d'Etude Therapeutique des Affections Inflammatoires Digestives. Gastroenterology. 1990;98:811-818. (50) Fefferman DS, Farrell RJ. Immunogenicity of biological agents in inflammatory bowel disease. Inflamm Bowel Dis. 2005;11(5):497-503. (51) Hommes D, Baert F, van Assche G, et al. The ideal management of Crohn's disease: Top down versus step up strategies, a randomized controlled trial. Gastroenterology. 2006;130:A-108, Abstract#749. (52) Shen C, Maerten P, Geboes K et al. Infliximab induces apoptosis of monocytes and T lymphocytes in a human-mouse chimeric model. Clin Immunol. 2005;115:250-259.

48 Therapies for inflammatory bowel diseases

(53) Van den Brande JM, Braat H, van den Brink GR et al. Infliximab but not etanercept induces apoptosis in lamina propria T-lymphocytes from patients with Crohn's disease. Gastroenterology. 2003;124:1774- 1785. (54) Hommes DW, Peppelenbosch MP, van Deventer SJ. Mitogen activated protein (MAP) kinase signal transduction pathways and novel anti-inflammatory targets. Gut. 2003;52:144-151 (55) Raingeaud J, Gupta S, Rogers JS et al. Pro-inflammatory cytokines and environmental stress cause p38 mitogen-activated protein kinase activation by dual phosphorylation on tyrosine and threonine. J Biol Chem. 1995;270:7420-7426. (56) Zhu X, Sano H, Kim KP et al. Role of mitogen-activated protein kinase-mediated cytosolic phospholipase A2 activation in arachidonic acid metabolism in human eosinophils. J Immunol. 2001;167:461-468. (57) Ge B, Gram H, Di Padova F et al. MAPKK-independent activation of p38alpha mediated by TAB1- dependent autophosphorylation of p38alpha. Science. 2002;295:1291-1294. (58) Davis RJ. Signal transduction by the JNK group of MAP kinases. Cell. 2000;103:239-252. (59) Hall JP, Merithew E, Davis RJ. c-Jun N-terminal kinase (JNK) repression during the inflammatory response? Just say NO. Proc Natl Acad Sci U S A. 2000;97:14022-14024. (60) Swantek JL, Cobb MH, Geppert TD. Jun N-terminal kinase/stress-activated protein kinase (JNK/SAPK) is required for lipopolysaccharide stimulation of tumor necrosis factor alpha (TNF-alpha) translation: glucocorticoids inhibit TNF-alpha translation by blocking JNK/SAPK. Mol Cell Biol. 1997;17:6274-6282. (61) Manning AM, Davis RJ. Targeting JNK for therapeutic benefit: from junk to gold? Nat Rev Drug Discov. 2003;2:554-565. (62) Boulton TG, Yancopoulos GD, Gregory JS et al. An insulin-stimulated protein kinase similar to yeast kinases involved in cell cycle control. Science. 1990;249:64-67. (63) Cobb MH, Goldsmith EJ. How MAP kinases are regulated. J Biol Chem. 1995;270:14843-14846. (64) Lee JC, Kassis S, Kumar S, Badger A, Adams JL. p38 mitogen-activated protein kinase inhibitors-- mechanisms and therapeutic potentials. Pharmacol Ther. 1999;82:389-397. (65) Lee JC, Kumar S, Griswold DE et al. Inhibition of p38 MAP kinase as a therapeutic strategy. Immunopharmacology. 2000;47:185-201. (66) Salituro FG, Germann UA, Wilson KP et al. Inhibitors of p38 MAP kinase: therapeutic intervention in cytokine-mediated diseases. Curr Med Chem. 1999;6:807-823. (67) Underwood DC, Osborn RR, Bochnowicz S et al. SB 239063, a p38 MAPK inhibitor, reduces neutrophilia, inflammatory cytokines, MMP-9, and fibrosis in lung. Am J Physiol Lung Cell Mol Physiol. 2000;279:L895-L902. (68) Underwood DC, Osborn RR, Kotzer CJ et al. SB 239063, a potent p38 MAP kinase inhibitor, reduces inflammatory cytokine production, airways eosinophil infiltration, and persistence. J Pharmacol Exp Ther. 2000;293:281-288. (69) Branger J, van den BB, Weijer S et al. Anti-inflammatory effects of a p38 mitogen-activated protein kinase inhibitor during human endotoxemia. J Immunol. 2002;168:4070-4077. (70) Schreiber S, Feagan B, D'Haens G et al. Oral p38 mitogen-activated protein kinase inhibition with BIRB 796 for active Crohn's disease: a randomized, double-blind, placebo-controlled trial. Clin Gastroenterol Hepatol. 2006;4:325-334. (71) ten Hove T, van den BB, Pronk I et al. Dichotomal role of inhibition of p38 MAPK with SB 203580 in experimental colitis. Gut. 2002;50:507-512. (72) Han Z, Chang L, Yamanishi Y, Karin M, Firestein GS. Joint damage and inflammation in c-Jun N- terminal kinase 2 knockout mice with passive murine collagen-induced arthritis. Arthritis Rheum. 2002;46:818-823. (73) Fleischer F, Dabew R, Goke B, Wagner AC. Stress kinase inhibition modulates acute experimental pancreatitis. World J Gastroenterol. 2001;7:259-265. (74) Wagner AC, Mazzucchelli L, Miller M, Camoratto AM, Goke B. CEP-1347 inhibits caerulein-induced rat pancreatic JNK activation and ameliorates caerulein pancreatitis. Am J Physiol Gastrointest Liver Physiol. 2000;278:G165-G172. (75) Bianchi M, Ulrich P, Bloom O et al. An inhibitor of macrophage transport and nitric oxide production (CNI-1493) prevents acute inflammation and endotoxin lethality. Mol Med. 1995;1:254- 266.

49 Chapter 2

(76) Martiney JA, Rajan AJ, Charles PC et al. Prevention and treatment of experimental autoimmune encephalomyelitis by CNI-1493, a macrophage-deactivating agent. J Immunol. 1998;160:5588-5595. (77) kerlund K, Erlandsson HH, Tracey KJ et al. Anti-inflammatory effects of a new tumour necrosis factor- alpha (TNF-alpha) inhibitor (CNI-1493) in collagen-induced arthritis (CIA) in rats. Clin Exp Immunol. 1999;115:32-41. (78) Hommes D, van den BB, Plasse T et al. Inhibition of stress-activated MAP kinases induces clinical improvement in moderate to severe Crohn's disease. Gastroenterology. 2002;122:7-14. (79) Hoshino R, Chatani Y, Yamori T et al. Constitutive activation of the 41-/43-kDa mitogen-activated protein kinase signaling pathway in human tumors. Oncogene. 1999;18:813-822. (80) Ahn NG, Nahreini TS, Tolwinski NS, Resing KA. Pharmacologic inhibitors of MKK1 and MKK2. Methods Enzymol. 2001;332:417-431. (81) Alessi DR, Cuenda A, Cohen P, Dudley DT, Saltiel AR. PD 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J Biol Chem. 1995;270:27489-27494. (82) Dudley DT, Pang L, Decker SJ, Bridges AJ, Saltiel AR. A synthetic inhibitor of the mitogen-activated protein kinase cascade. Proc Natl Acad Sci U S A. 1995;92:7686-7689. (83) Sebolt-Leopold JS, Dudley DT, Herrera R et al. Blockade of the MAP kinase pathway suppresses growth of colon tumors in vivo. Nat Med. 1999;5:810-816. (84) Lorusso PM, Adjei AA, Varterasian M et al. Phase I and pharmacodynamic study of the oral MEK inhibitor CI-1040 in patients with advanced malignancies. J Clin Oncol. 2005;23:5281-5293. (85) Williams DH, Wilkinson SE, Purton T et al. Ro 09-2210 exhibits potent anti-proliferative effects on activated T cells by selectively blocking MKK activity. Biochemistry. 1998;37:9579-9585. (86) Zhao A, Lee SH, Mojena M et al. Resorcylic acid lactones: naturally occurring potent and selective inhibitors of MEK. J Antibiot (Tokyo). 1999;52:1086-1094. (87) Gu Z, Jiang Q, Zhang G. Extracellular signal-regulated kinase and c-Jun N-terminal protein kinase in ischemic tolerance. Neuroreport. 2001;12:3487-3491. (88) Mori T, Wang X, Jung JC et al. Mitogen-activated protein kinase inhibition in traumatic brain injury: in vitro and in vivo effects. J Cereb Blood Flow Metab. 2002;22:444-452. (89) Clemons AP, Holstein DM, Galli A, Saunders C. Cerulein-induced acute pancreatitis in the rat is significantly ameliorated by treatment with MEK1/2 inhibitors U0126 and PD98059. Pancreas. 2002;25:251-259.

50

Chapter 3

Specific inhibition of c-Raf activity by semapimod induces clinical remission in severe Crohn’s disease

Mark Löwenberg1,4, Auke Verhaar1, Bernt van den Blink1, Fibo ten Kate2, Sander van Deventer1, Maikel Peppelenbosch3, Daniel Hommes4

1 Department for Experimental Internal Medicine, Academic Medical Center, Amsterdam, The Netherlands 2 Department of Pathology, Academic Medical Center, Amsterdam, The Netherlands 3 Department of Cell Biology, University of Groningen, Groningen, The Netherlands 4 Department of Gastroenterology and Hepatology, Academic Medical Center, Amsterdam, The Netherlands

Journal of Immunology. 2005; 175(4): 2293-300

51 Chapter 3

Abstract There is an unmet need for novel treatment strategies in Crohn’s disease, a chronic relapsing inflammatory disease of the gut. In an earlier study, we reported clinical efficacy of a two week treatment with semapimod (CNI-1493) in twelve patients with therapy resistant Crohn’s disease. The aim of this study was to identify the cellular target underlying semapimod action. Experiments with murine macrophages showed impaired MAPK signaling and decreased cytokine production due to semapimod treatment. In vitro kinase assays revealed c-Raf as a direct molecular target of semapimod, and semapimod did not affect b-Raf enzymatic activity. Immunohistochemistry performed on paired colon biopsies obtained from Crohn’s disease patients (n=6) demonstrated increased expression of phospho-MEK, the substrate of Raf. Strikingly, phospho-MEK levels were significantly decreased in patients with a good clinical response to semapimod, but no decrease in phospho-MEK expression was observed in a clinically non-responsive patient. In conclusion, this study identifies c-Raf as a molecular target of semapimod and suggests that decreased c-Raf activity correlates with clinical benefit in Crohn’s disease. Our observations indicate that c-Raf inhibitors are prime candidates for the treatment of Crohn’s disease.

52 Semapimod inhibits c-Raf

Introduction Inhibitors of intracellular signaling pathways have proven effective in a wide range of experimental inflammatory disorders, including experimental colitis 1. Small molecules targeting these signaling cascades are generally considered as a promising novel strategy for the clinical management of inflammatory bowel diseases (i.e. Crohn’s disease and ulcerative colitis). In particular, pharmaceutical intervention of the MAPK pathways of intracellular signaling mediators has attracted widespread interest 2-6. Three major MAPK cascades have been identified; ERK, JNK and p38 MAPK, and these pathways are critically involved in inflammatory pathology including Crohn’s disease (CD) 6-8. Selective MAPK inhibitors targeting the p38 MAPK, ERK and JNK pathway, demonstrated anti-inflammatory effects in pre-clinical models 1,9-13. Despite the fact that the impact of MAPK pathways on inflammatory pathology is profound, the molecular details of these signaling cascades in the pathogenesis of inflammatory disorders and their possible therapeutic value remain to be elucidated. In view of the redundancy of MAPK pathways and the extensive crosstalk between these and other routes of signal transduction (for example NFκB), such information is of great importance. We have reported that treatment of therapy resistant CD patients with the small molecule semapimod resulted in a reduction of disease activity and induction of clinical remissions 14. Although it has been demonstrated that semapimod interferes with the phophorylation of JNK and p38 MAPK 14, the exact underlying molecular mechanism of semapimod action remains to be characterized. The identification of the molecular target of semapimod has important clinical relevance, because it may prompt synthesis of a novel class of anti-inflammatory compounds. In this study we have identified macrophages as the target cells of semapimod action, and we characterized c-Raf as the cellular target. Reduced expression of phospho-MEK, a downstream target of c-Raf, in colon biopsies correlated with clinical benefit in semapimod-treated CD patients. In contrast, no reduction of phospho-MEK expression was observed in mucosal biopsies obtained from a non-responder. These results indicate that c-Raf activity is a critical mediator of disease progression in CD, and identify c- Raf as a novel therapeutic target for the clinical management of CD.

53 Chapter 3

Materials and Methods Antibodies and reagents. Phosphospecific antibodies (Abs) directed against p38Thr180/Tyr182, ERK1/2Thr202/Tyr204, MEK1/2Ser217/221, c-RafSer338, SAPK/JNKThr183/Tyr185, PAK1/2 Thr423/402, SEK1/MKK4Thr261, MKK3/pMKK6Ser189/207, as well as Abs specific for MKK4, MKK3 and PAK were purchased from Cell Signaling Technology (Westburg, Leusden, the Netherlands). Abs recognizing p38, ERK, JNK, MEK, b-Raf, c-Raf and phospho-JNKThr183/Tyr185 were from Santa Cruz Biotechnology Inc. (Heidelberg, Germany). HRP-conjugated Goat-anti-Rabbit, Goat-anti-Mouse and Rabbit-anti-Goat were from DakoCytomation (Heverlee, Belgium), and semapimod (CNI-1493) was acquired from Cytokine PharmaSciences Inc. (batch: 3/13/2004; Lot No: 08610302). The anti-CD68 monoclonal antibody was from DakoCytomation and anti-CD14 mAb was obtained from Becton Dickinson (Alphen a/d Rijn, the Netherlands). Anti-human CD3 (CD3 epsilon; mouse) was provided by Dr. A. te Velde (Academic Medical Center, Amsterdam, the Netherlands). Anti-CD28 was from Sanquin (Amsterdam, the Netherlands). The c-Raf and b-Raf kinase kits were obtained from Upstate (Veenendaal, the Netherlands).

CD4 purification and cell sorting. PBMCs were isolated from whole blood of healthy volunteers by Ficoll-Isopaque density gradient centrifugation (Amersham Biosciences, Roosendaal, the Netherlands). The monocytes present in the PBMC pellet were removed by an adherence procedure: cells were plated out in 6 well plates (CellStar, Greiner Bio-One, Alphen a/d Rijn, the Netherlands) at a final concentration of 5x106 cells per well for 1.5 hrs at 37 °C and non-adherent cells were subsequently harvested for magnetic cell sorting. CD4+ T cells were purified by depletion of non-CD4+ T cells (negative selection) using the MACS system. Non-CD4+ cells were indirectly magnetically labeled with a cocktail of biotin- conjugated monoclonal antibodies (against CD8, CD14, CD16, CD19, CD36, CD56, CD123, TCRγ/δ and Glycophorin A) bound to MicroBeads conjugated to a monoclonal anti-biotin Ab, as secondary labeling agent (Miltenyi Biotec Inc., Auburn, CA, USA). The magnetically labeled non-CD4+ T cells were depleted by retaining them on a MACS Column in the magnetic field of the autoMACS Separator (Miltenyi Biotec), while the unlabeled fraction of CD4+ T helper cells passed through the column. The sample purity was assessed by fluorescence-activated cell sorter (FACS) (Becton Dickinson, San Jose, CA, USA) with PE-

54 Semapimod inhibits c-Raf

conjugated CD4 and FITC-conjugated CD3 mAbs (Becton Dickinson) (purity >95% CD3+CD4+; not shown).

Generation of dendritic cells (DCs). DC generation from PBMCs (obtained form healthy volunteers) was performed as described previously 15,16. Briefly, PBMCs were resuspended in AIM-V medium (Invitrogen, Breda, the Netherlands), and allowed to adhere to 6 well plates (CellStar). After 2 hours at 37 °C, nonadherent cells were removed and the adherent cells were cultured in medium supplemented with 50 ng/ml GM-CSF and 1000 U/ml IL-4. Next, monocytes were incubated for 6 days in X-VIVO 15 medium (BioWhittaker, Heerhugowaard, the Netherlands) supplemented with 1000 U/ml GM-CSF (Berlex Laboratories Inc., Richmond, CA, USA) and 1000 U/ml IL-4 (R&D Systems, Minneapolis, MN, USA). The immature DCs were stimulated at day 6 in X-VIVO 15 medium supplemented with a cytokine cocktail containing TNF-α (10 ng/ml), PGE-2 (1 µg/ml), IL-1β (10 ng/ml), IL-6 (150 ng/ml), GM-CSF (800 U/ml) and IL-4 (500 U/ml). After 24 hours, mature DCs were harvested for phenotyping using a panel of mAbs and analyzed on a FACScan with cell quest software (Becton Dickinson), as previously described 17.

Cell culture. 4/4 macrophages (murine), which are phenotypically and functionally not different from primary isolated mature macrophages 18,19, were cultured in RPMI1640 (Life Technologies, Paisley, UK), supplemented with 10% heat-inactivated FCS, 2 mM L- glutamine and penicillin-streptomycin; ‘complete’, in a humidified 5% CO2 environment at 37 °C. Human CD4+ T cells were grown in IMDM (Gibco, Breda, the Netherlands), supplemented with 10% FCS, 2 mM L-glutamine and penicillin-streptomycin; ‘complete’, in a humidified 5% CO2 environment, 37 °C.

MTT viability assay. The cytotoxic effect of semapimod was studied in macrophages, which were incubated overnight with increasing concentrations of semapimod (0,01; 0,1; 1; 10; 100 µM, diluted in media) with or without LPS (100 ng/ml). Cell viability was assessed by MTT colorimetric assay. After overnight incubation, 0.5 mg/ml MTT (3-[4,5-dimethylthiazol-2-yl]- 2,5-diphenyltetrazolium bromide) was added to the media for 1-2 hours at 37 °C, and isopropanol/0.04N HCl was subsequently added. The OD560 was determined using an ELISA-

55 Chapter 3

plate reader (Bio-Rad Benchmark, Hercules, CA, USA). Treatment with semapimod concentrations ≤ 1 µM did not affect cell viability. A semapimod-induced cytotoxic effect was observed at semapimod concentrations > 1 µM (i.e. 10 and 100 µM); not shown.

Cytokine Bead Array. Macrophages and CD4+ T cells were pretreated for one hour with various concentrations of semapimod and cultured up to 24 hours in the presence of LPS (100 ng/ml) or anti-CD3 (immobilized on plastic)/anti-CD28 (3 µg/ml, soluble) respectively. Furthermore, mature DCs were pretreated for one hour with 0.1 and 1 µM semapimod. Media was removed and cells were cultured for 24 hrs in fresh media containing CD40L transfected J558 cells (1:1). The CD40L transfected mouse plasmacytoma cell line (J558), was a kind gift from dr. P. Lane (University of Birmingham, Birmingham, UK) 20. Cytokine levels were analyzed in supernatants of macrophages, DCs and T cells by Cytokine Bead Array (CBA) (BD Biosciences, Alphen a/d Rijn, the Netherlands) using a flow cytometer (Becton Dickinson), according to routine procedure.

Western blot analysis. MAPK signaling pathways were studied on Western blot using a panel of phosphospecific Abs. Macrophages were seeded in 6 well plates at a final concentration of 1-2x106 cells per well and grown overnight. Cells were pretreated for one hour with 0.1 and 1 µM semapimod, and subsequently stimulated with LPS (100 ng/ml) for 15 minutes. After washing with PBS, cells were harvested in sample buffer (150 mM Tris HCl, 6% SDS, 3% β- mercaptoethanol; 20% glycerol, 1 mg bromophenol blue, pH 6.8) and whole cell lysates were loaded on 10% SDS-PAGE and transferred to a polyvinylidene difluoride membrane (PVDF) (Immobilon-P; Millipore, Amsterdam, the Netherlands). Membranes were blocked with 1% Protifar in TBST (0.05M Tris, 150mM NaCl, 0.05% Tween-20). Primary and secondary HRP-conjugated antibodies were diluted in 1% Protifar TBST, and proteins were visualized using the Lumi-Light substrate (Roche, Woerden, the Netherlands). Blots were incubated in stripping buffer (62.5mM Tris-HCl pH6.8, 100mM β-mercaptoethanol, 2% SDS) for one hour at 50 °C, and reprobed with appropriate Abs to evaluate for equal loading. In addition, T cells (overnight cultured in 6 well plates, 3.106 cells/well) pretreated with 0.1 or 1 µM semapimod (1 hour) and subsequently activated with anti-CD3/anti-CD28 Abs (15 min.), were analyzed on Western blot using the indicated Abs.

56 Semapimod inhibits c-Raf

Raf in vitro kinase assays. Raf in vitro kinase assays were used according to the instructions of the manufacturer (Upstate). Truncated constitutively-active Raf (b-Raf and c-Raf) was diluted in a Mg/ATP cocktail and reaction buffer and incubated on ice with semapimod (1 µM) for 5 or 10 minutes. Next, recombinant inactive MEK was added and in vitro kinase assays were performed at 30 °C for 20 minutes. Active Raf together with MEK, and Raf without MEK served as a positive and negative control respectively. Samples were dissolved in sample buffer, heated at 95 °C for 5 min and immunoblotted using the anti-phospho- MEKser218/222 /MEK2Ser222/226 Ab.

Immunohistochemistry. To assess the amount of active MEK in the intestinal mucosa, screening and week 4 colon specimens were obtained from most affected regions of inflammation of CD patients (n=6) who participated in the semapimod study 14. Biopsies were analyzed for phospho-MEK expression. Paraffin sections (4 µm) were dewaxed, rehydrated in graded alcohols and endogenous peroxidase activity was quenched with 1.5% H2O2 in methanol (15 min, room temperature). Antigen retrieval was performed by heating for 10 minutes at 100 °C in 0.01M sodium citrate. After washing (PBS), non-specific staining was reduced by a blocking step with TENG-T (10 mM Tris, 5 mM EDTA, 0.15 M NaCl, 0.25% gelatin, 0.05% (v/v) Tween-20, pH 8.0) for 20 minutes at room temperature. Slides were washed and incubated overnight (4 °C) with an anti-phospho-MEKser218/222/MEK2Ser222/226 Ab diluted in 1% BSA 0.1% Triton-X-100. Next, slides were incubated for 20 minutes with a post-Ab blocking solution for powervision (Immunologic, Duiven, the Netherlands), followed by a 30 minute incubation with Poly-HRP-GAM/R/R IgG (Immunologic). Peroxidase activity was detected using diaminobenzidine (Fast DAB) (Sigma, St. Louis, MO, USA) in 0.05 M Tris, pH 7.4. Sections were counterstained with hematoxylin (Mayer, Fluka, Steinheim, Germany) when appropriate, dehydrated in graded alcohols, and mounted with Pertex (Histolab Products AB, Gothenburg, Sweden) under coverslips. Controls consisted of omitting the primary and secondary Ab and use of an appropriate Ig control (not shown).

Statistical analysis. Quantitative confirmation came from experiments in which the number of phospho-MEK positive cells was counted in sections in a blinded fashion. Two pictures of each section were taken at 200x magnification, and positive cells were counted, blind to

57 Chapter 3

treatment and day of endoscopy, in each microscope field with the use of an image analysis program (EFM Software, Rotterdam, the Netherlands). Pictures appeared randomly on a computer monitor and all intensely staining cells were marked positive by an observer, counted, and stored by the image analysis program for later data analysis. Statistical analysis was performed by use of the Wilcoxon test, and p-value < 0.05 was considered as statistically significant.

Results Semapimod does not affect MAPK signaling cascades in T cells It has been previously reported that T cell cytokine production is not influenced by semapimod 21, and this was confirmed in our laboratory (not shown). ERK, JNK and p38 MAPK signal transduction pathways were activated in T cells stimulated with anti-CD3/anti- CD28 Abs, and this was not affected by incubation with semapimod (Figure 1). Thus, these findings indicate that T cells are no direct target of semapimod action.

Figure 1. Semapimod does not influence phosphorylation of ERK, JNK or p38 MAPKs in T cells. Cells were pretreated with semapimod for 1 hour and subsequently stimulated for 15 minutes using anti-CD3 and anti-CD28 Abs. Cell lysates were immunoblotted using phosphospecific Abs against ERK, JNK and p38 MAPK. To test for equal loading blots were reprobed with appropriate Abs. Western blots represent three independent experiments, duplo conditions are shown.

Semapimod does not influence IL-12 cytokine production in activated mature DCs Recently, it has been reported that semapimod interferes with dendritic cell (DC) maturation 22, which prompted us to study the effect of semapimod on mature DCs. The effect of semapimod on IL-12 cytokine production was studied in CD40L activated mature DCs, because it is generally accepted that this is an important Th1 differentiation mechanism that is relevant for CD 23-26. Our data indicate that semapimod does not interfere with IL-12 cytokine

58 Semapimod inhibits c-Raf

production in activated mature DCs, suggesting that this compound does not influence the capacity of DCs to induce a Th1-type immune response (Figure 2).

Figure 2. Semapimod does not affect IL-12 cytokine production in activated mature dendritic cells (DCs). Mature DCs were incubated for 1 hour with semapimod and subsequently cultured for 24 hours in the presence of CD40L overexpressing cells. IL-12 cytokine levels were analyzed in supernatants by CBA. IL-12 levels in supernatants of control cells were not measurable and therefore error bars are not appropriate. Results are expressed as the mean ± SD of triplicate determinations.

Semapimod inhibits cytokine production by macrophages To determine the usefulness of in vitro stimulation of macrophages for studying the underlying molecular mechanism of semapimod action, the effect of semapimod on cytokine production in macrophages was investigated. Semapimod treatment resulted in a dose responsive reduction of LPS-induced TNFα, IL-1β and IL-6 protein levels (Figure 3). Decreased cytokine production observed upon treatment with 0.01, 0.1 and 1 µM semapimod was not a result of reduced cell viability, as MTT colorimetric assays revealed significant cytotoxicity of semapimod only at concentration of 10 and 100 µM (not shown). These observations confirm that semapimod effectively blocks cytokine synthesis in macrophages 27- 29, and indicate that incubation of macrophages with semapimod concentration of 0.1 and 1 µM constitutes an appropriate experimental system for identifying the molecular mechanism underlying semapimod-dependent inhibition of pro-inflammatory cytokine production. In vivo concentrations of 1 µM and 5 µM have been demonstrated in preclinical and clinical studies respectively 30,31. Therefore, we conclude that 0.1 and 1 µM semapimod concentrations used for our in vitro studies have clinical relevance.

59 Chapter 3

Figure 3. Semapimod inhibits TNF-α, IL-1β and IL-6 cytokine production in activated macrophages. Cells were cultured overnight with increasing semapimod concentrations in the absence or presence of LPS, and supernatants were analyzed by CBA. Decreased cytokine levels due to treatment with 0.01, 0.1 and 1 µM semapimod was not a result of reduced cell viability, as MTT colorimetric assays revealed significant toxicity of semapimod only at concentration of 10 and 100 µM. Results are expressed as the mean ± SD of triplicate determinations. LPS treated and control cells are indicated with black and open dots respectively.

Semapimod inhibits MAPK signaling pathways in macrophages As it has been previously shown that semapimod blocks p38 MAPK and JNK phosphorylation in vitro 14, this prompted us to study the effects of semapimod on MAPK signaling pathways in more detail. Therefore, the activation status of various kinases involved in MAPK signaling was analyzed by immunoblotting semapimod treated macrophages employing phosphospecific Abs against MAPK signal transduction molecules. LPS enhances phosphorylation of JNK, ERK and p38 MAPKs and semapimod treatment resulted in suppressed phosphorylation of JNK, ERK and p38 MAPKs (Figure 4). LPS-induced phosphorylation of upstream MAPK activators was observed, known as MAPK kinases (i.e. MEK1/2 for ERK, MKK4 for JNK and MKK3/6 for p38 MAPK). Impaired phosphorylation of MAPK kinases (i.e. MEK1/2, MKK4, MKK3/6) was seen upon pretreatment with semapimod. c-Raf phosphorylation was observed in stimulated and control cells (without LPS), in concordance with a previous report 32. Semapimod did not affect c-Raf phosphorylation, nor did it affect phosphorylation of PAK (p21-activated-protein kinase), an upstream c-Raf activator 33. These in vitro data suggest that semapimod interferes with MAPK activation upstream from MAPKK and downstream from PAK, thereby making c-Raf a likely candidate target.

60 Semapimod inhibits c-Raf

Figure 4. Semapimod inhibits MAPK phos- phorylation in macro- phages. The effect of semapimod on LPS- induced MAPK signaling cascades was studied in macrophages. Cells were pretreated with semapimod for 1 hr and stimulated with LPS for 15 min. MAPK signaling molecules were analyzed for their activation status on Western blot using phosphospecific Abs. Blots were reprobed with appropriate Abs to test for equal loading. The Western blots represent three independent experiments and duplo conditions are shown.

C-Raf is a molecular target of semapimod The observed inhibition of MEK phosphorylation by semapimod in LPS-stimulated macrophages without an apparent accompanying effect on c-Raf activation itself may indicate that semapimod is a direct inhibitor of c-Raf catalytic activity. To directly test this hypothesis, we employed two protein in vitro kinase assays in which the capacity of recombinant constitutively-active Raf to phosphorylate MEK was tested in the presence or absence of semapimod. Incubation of active c-Raf or b-Raf together with MEK in the absence of semapimod clearly induced MEK phosphorylation. Pretreatment of c-Raf with 1µM

61 Chapter 3

semapimod for 5 and 10 minutes abolished its potential to phosphorylate MEK in this two protein assay (Figure 5). Importantly semapimod treatment of b-Raf, an enzyme which is closely related to c-Raf, did not result in altered MEK phosphorylation (Figure 5), demonstrating the specificity of semapimod as an inhibitor of c-Raf enzymatic activity. These data reveal specific and direct inhibition of c-Raf enzymatic activity by semapimod.

Figure 5. Semapimod inhibits c-Raf activity. In vitro kinase assays were performed to study the effect of semapimod on c-Raf and b-Raf enzymatic activities using constitutively active c-Raf or b-Raf and MEK (a Raf substrate). C-Raf and b-Raf were pretreated for 5 or 10 minutes with semapimod (1µM) prior to in vitro kinase assays. Positive (i.e. active-Raf with MEK) and negative controls (i.e. active-Raf without MEK) are shown. Samples were immunoblotted for phospho-MEK1/2 expression. Western blots were reprobed with anti-b-Raf or anti-c-Raf Abs as a loading control. Similar results were obtained in three independent experiments. p, phosphorylated.

Semapimod inhibits c-Raf activity in vivo In an earlier study, patients with severe CD (mean Crohn’s Disease Activity Index (CDAI) of 380) received either 8 or 25 mg/m2 semapimod intravenously once daily for 12 consecutive days 14. Paired colon biopsies were available at baseline and after 4 weeks of treatment for 6 CD patients. Their mean age was 32 years, 2 were males, 5 were treated with infliximab (anti- TNF), 2 with steroids and 1 with mesalazine prior to semapimod treatment. Three patients received 8 mg/m2 semapimod, and the remaining 3 were treated with a 25 mg/m2 dose. Clinical response was defined by a CDAI reduction of ≥ 25% and ≥ 70 points compared to baseline or the occurrence of a clinical remission, as assessed by CDAI < 150 34. A clinical response was observed in 5 of 6 patients (mean CDAI reduction of 261 points at week 16), of whom 4 went into clinical remission at 16 weeks after initiation (Figure 6). 1 out of 6 patients

62 Semapimod inhibits c-Raf

did not respond to semapimod treatment. The observed clinical response rate correlated to a decrease in C-reactive protein (CRP) serum concentrations: all responders (5 patients) demonstrated decreased CRP levels and the single patient that did not show a decrease of the serum CRP did not respond clinically (Figure 6).

Figure 6. Clinical responses (i.e. CDAI reduction of ≥ 25% and ≥ 70 points compared to baseline) were seen in 5 out of 6 patients. 4 of 5 responders went into clinical remission (i.e. CDAI < 150) at 16 weeks after initiation. One responder, who did not go into clinical remission, demonstrated a 205 point decrease of CDAI at week 8 and a 117 point decrease at week 16 compared to baseline. The observed response rates correlated to decreased CRP serum levels: all responders demonstrated decreased CRP concentrations during follow-up, in contrast to the non-responder. CDAI, Crohn’s Disease Activity Index; CRP, C-reactive protein.

To establish the effect of semapimod treatment on c-Raf activity in vivo, colon biopsies were analyzed for phospho-MEK expression. Neutrophils and monocytes were detected as CD14+ cells (Figure 7 E) and macrophages as CD68+ cells (Figure 7 F) in adjoining sections. Lymphocytes (identified as CD3+ cells) had a different distribution pattern compared to phospho-MEK positive cells (not shown). Immunohistochemical analysis revealed high levels of phospho-MEK at baseline, which was mainly localized to macrophages and neuroendocrine cells in the crypts (Figure 7A and C). Faint phospho-MEK staining was seen in neutrophils and monocytes. The decrease in phospho-MEK expression after therapy, observed in 5 out of 6 patients, was statistically significant (* p = 0.0348) (Figure 7G). One patient did not show decreased phospho-MEK expression (Figure 7D). We next analyzed whether the reduction of phospho-MEK positive cells correlated with clinical outcome (defined by CDAI and CRP levels). The non-responder did not demonstrate decreased phospho-MEK expression in colon biopsies obtained at week 4 after treatment (Figure 7D)

63 Chapter 3

compared to baseline (Figure 7C). In contrast, all responders revealed significant reduced numbers of phospho-MEK positive cells after therapy (Figure 7G and H). These data indicate that semapimod inhibits c-Raf activity not only in vitro but also in vivo.

Figure 7. Semapimod treatment decreases phospho-MEK expression in vivo. Paired colon biopsies were obtained at screening (day 1) and at week 4 for 6 CD patients, who received 12 days of intravenous infusions with semapimod. (A) High phospho-MEK expression levels were seen before treatment, and this was mainly localized to macrophages and neuroendocrine cells (magnification, 200x). Inflammatory cells were identified as neutrophils or monocytes (E) (CD14+, closed arrows) or macrophages (F) (CD68+, dotted arrows), in adjoining sections (magnification, 400x). (B) Decreased cell numbers staining positive for phospho-MEK were seen in 5 out of 6 patients following treatment. In contrast, no difference in phospho-MEK expression was seen in one patient before (C) and after (D) semapimod therapy (magnification, 200x). (G) The decrease in phospho-MEK positive cells after treatment, observed in 5 responders, was statistically significant (* p = 0.0348). (H) One non- responder did not demonstrate a reduction of phospho-MEK positive cells after treatment.

64 Semapimod inhibits c-Raf

Discussion As a consequence of the limited efficacy and significant toxicity of current therapies, there is widespread interest in the development of novel drugs for the clinical management of CD. We have reported that semapimod, in a small and uncontrolled clinical trial in severe CD patients, showed significant clinical benefit, and that clinical responses correlated with an inhibitory effect on JNK and p38 MAPK signaling cascades 14. Despite this therapeutically relevant outcome, the molecular mechanism of semapimod action remains unexplained. We here report that c-Raf in macrophages is the molecular target of semapimod. Studies with LPS- stimulated macrophages show that this small molecule inhibits LPS signaling at the level of Raf, resulting in reduced pro-inflammatory cytokine production. Additional in vitro kinase assays revealed that semapimod is a specific c-Raf inhibitor. In agreement with a role for semapimod as an in vivo inhibitor of c-Raf, colon biopsies obtained from semapimod treated CD patients who responded to therapy showed significant decreased phospho-MEK expression, which was predominantly localized to macrophages and neuroendocrine cells. Interestingly, whereas semapimod is highly active in the macrophage compartment, our in vitro data confirm an earlier report that T cells are no direct target cells of semapimod 21. Semapimod treatment did not affect cytokine production neither did it affect MAPK signaling cascades in T cells. A likely explanation may be found in the relative importance of b-Raf in comparison to c-Raf in activating MAPK cascades in lymphocytes, further emphasizing the specificity of the c-Raf inhibitory effect observed 35-37. We evaluated whether semapimod could affect IL-12 cytokine production in activated mature DCs, a major pathogenic mechanism in T cell-mediated pathology, such as CD 26. Semapimod did not interfere with IL-12 cytokine production, suggesting that this compound does not influence Th1 mediated responses by mature DCs. All together, these observations indicate that the cell-specific effects of semapimod are related to Raf isotype specificity, and hypothesize that c-Raf inhibition in macrophages is the primary effector of semapimod action in CD. Macrophages play a major role in initiating, amplifying and perpetuating the inflammatory response by activating immune cells, including monocytes and T cells 38,39. However, we are not aware of data indicating that a therapeutic strategy that mainly targets macrophages has therapeutic efficacy in a chronic inflammatory disease in humans. Our data suggest that semapimod- induced inhibition of c-Raf in one particular immune cell (macrophage) results in a clinical

65 Chapter 3

response in severe CD, independent from an effect on T cells or dendritic cells. Hence, the present study provides novel evidence for a pivotal role of macrophages in the pathogenesis of CD 40-45. The identification of c-Raf as the molecular target of semapimod raises questions regarding the function of this molecule in the inflammatory process. Previous work has demonstrated that c-Raf is involved in inflammatory mechanisms by controlling downstream signaling molecules such as the pro-inflammatory transcription factor NFκB 46-48, thereby mediating cytokine synthesis and other pro-inflammatory mediators 49-55. In addition, various studies have identified c-Raf as an important anti-apoptotic molecule and its inhibition may well cause effector macrophages to undergo programmed cell death in the pro-apoptotic inflammatory environment present in the gut of CD patients 56,57. As a result, induced apoptosis of macrophages could lead to an attenuation of the inflammatory process. Further studies investigating apoptosis in the gut of semapimod treated CD patients may provide answers to this important issue. Our observations indicate that the pro-inflammatory effects of c-Raf include not only activation of ERK, but also JNK and p38 MAPK, and suggest that c- Raf may be an important target for anti-inflammatory small molecules. Clinical studies with semapimod demonstrated that the drug is relatively well tolerated 14,30,58. Side effects included local irritation at the infusion site (phlebitis) and mild increases in liver enzymes, both resolving spontaneously within weeks. Various Raf inhibitors have passed phase 1/2 as anticancer strategy showing a tolerable safety profile 59-67. To our knowledge, no clinical studies have been performed with Raf inhibitors as anti-inflammatory agents. A principal role for c-Raf in the pathogenesis of CD could have important clinical consequences, as these data indicate that semapimod and possibly other c-Raf inhibitors constitute novel candidates for the treatment of severe CD.

Acknowledgments We thank Meike Scheffer and Joyce Bilderbeek for technical support.

66 Semapimod inhibits c-Raf

Reference List

(1) Lowenberg M, Peppelenbosch MP, Hommes DW. Therapeutic modulation of signal transduction pathways. Inflamm Bowel Dis. 2004;10 Suppl 1:S52-S57. (2) Cobb MH, Goldsmith EJ. How MAP kinases are regulated. J Biol Chem. 1995;270:14843-14846. (3) Hommes DW, Peppelenbosch MP, Van Deventer SJ. Mitogen activated protein (MAP) kinase signal transduction pathways and novel anti-inflammatory targets. Gut. 2003;52:144-151. (4) Johnson GL, Lapadat R. Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science. 2002;298:1911-1912. (5) van den BB, Ten Hove T, Van Den Brink GR, Peppelenbosch MP, Van Deventer SJ. From extracellular to intracellular targets, inhibiting MAP kinases in treatment of Crohn's disease. Ann N Y Acad Sci. 2002;973:349-358. (6) Waetzig GH, Schreiber S. Review article: mitogen-activated protein kinases in chronic intestinal inflammation - targeting ancient pathways to treat modern diseases. Aliment Pharmacol Ther. 2003;18:17-32. (7) Kyriakis JM, Avruch J. Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation. Physiol Rev. 2001;81:807-869. (8) Waetzig GH, Seegert D, Rosenstiel P, Nikolaus S, Schreiber S. p38 mitogen-activated protein kinase is activated and linked to TNF-alpha signaling in inflammatory bowel disease. J Immunol. 2002;168:5342-5351. (9) English JM, Cobb MH. Pharmacological inhibitors of MAPK pathways. Trends Pharmacol Sci. 2002;23:40-45. (10) Hollenbach E, Neumann M, Vieth M et al. Inhibition of p38 MAP kinase- and RICK/NF-kappaB- signaling suppresses inflammatory bowel disease. FASEB J. 2004;18:1550-1552. (11) Kumar S, Boehm J, Lee JC. p38 MAP kinases: key signalling molecules as therapeutic targets for inflammatory diseases. Nat Rev Drug Discov. 2003;2:717-726. (12) Manning AM, Davis RJ. Targeting JNK for therapeutic benefit: from junk to gold? Nat Rev Drug Discov. 2003;2:554-565. (13) Saklatvala J. The p38 MAP kinase pathway as a therapeutic target in inflammatory disease. Curr Opin Pharmacol. 2004;4:372-377. (14) Hommes D, van den BB, Plasse T et al. Inhibition of stress-activated MAP kinases induces clinical improvement in moderate to severe Crohn's disease. Gastroenterology. 2002;122:7-14. (15) Sallusto F, Lanzavecchia A. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J Exp Med. 1994;179:1109-1118. (16) Berger TG, Feuerstein B, Strasser E et al. Large-scale generation of mature monocyte-derived dendritic cells for clinical application in cell factories. J Immunol Methods. 2002;268:131-140. (17) Romani N, Reider D, Heuer M et al. Generation of mature dendritic cells from human blood. An improved method with special regard to clinical applicability. J Immunol Methods. 1996;196:137-151. (18) Peppelenbosch MP, DeSmedt M, Ten Hove T, Van Deventer SJ, Grooten J. Lipopolysaccharide regulates macrophage fluid phase pinocytosis via CD14-dependent and CD14-independent pathways. Blood. 1999;93:4011-4018. (19) DeSmedt M, Rottiers P, Dooms H, Fiers W, Grooten J. Macrophages induce cellular immunity by activating Th1 cell responses and suppressing Th2 cell responses. J Immunol. 1998;160:5300-5308. (20) Lane P, Burdet C, McConnell F, Lanzavecchia A, Padovan E. CD40 ligand-independent B cell activation revealed by CD40 ligand-deficient T cell clones: evidence for distinct activation requirements for antibody formation and B cell proliferation. Eur J Immunol. 1995;25:1788-1793. (21) Bjork L, Tracey KJ, Ulrich P et al. Targeted suppression of cytokine production in monocytes but not in T lymphocytes by a tetravalent guanylhydrazone (CNI-1493). J Infect Dis. 1997;176:1303-1312. (22) Zinser E, Turza N, Steinkasserer A. CNI-1493 mediated suppression of dendritic cell activation in vitro and in vivo. Immunobiology. 2004;209:89-97. (23) Banchereau J, Briere F, Caux C et al. Immunobiology of dendritic cells. Annu Rev Immunol. 2000;18:767-811. (24) Hsieh CS, Macatonia SE, Tripp CS et al. Development of TH1 CD4+ T cells through IL-12 produced by Listeria-induced macrophages. Science. 1993;260:547-549.

67 Chapter 3

(25) Manetti R, Parronchi P, Giudizi MG et al. Natural killer cell stimulatory factor (interleukin 12 [IL-12]) induces T helper type 1 (Th1)-specific immune responses and inhibits the development of IL-4- producing Th cells. J Exp Med. 1993;177:1199-1204. (26) Trinchieri G. Interleukin-12 and its role in the generation of TH1 cells. Immunol Today. 1993;14:335- 338. (27) Bianchi M, Ulrich P, Bloom O et al. An inhibitor of macrophage arginine transport and nitric oxide production (CNI-1493) prevents acute inflammation and endotoxin lethality. Mol Med. 1995;1:254- 266. (28) Bianchi M, Bloom O, Raabe T et al. Suppression of proinflammatory cytokines in monocytes by a tetravalent guanylhydrazone. J Exp Med. 1996;183:927-936. (29) Cohen PS, Nakshatri H, Dennis J et al. CNI-1493 inhibits monocyte/macrophage tumor necrosis factor by suppression of translation efficiency. Proc Natl Acad Sci U S A. 1996;93:3967-3971. (30) Atkins MB, Redman B, Mier J et al. A phase I study of CNI-1493, an inhibitor of cytokine release, in combination with high-dose interleukin-2 in patients with renal cancer and melanoma. Clin Cancer Res. 2001;7:486-492. (31) Cerami C, Zhang X, Ulrich P et al. High-performance liquid chromatographic method for guanylhydrazone compounds. J Chromatogr B Biomed Appl. 1996;675:71-75. (32) Coles LC, Shaw PE. PAK1 primes MEK1 for phosphorylation by Raf-1 kinase during cross-cascade activation of the ERK pathway. Oncogene. 2002;21:2236-2244. (33) Morrison DK, Cutler RE. The complexity of Raf-1 regulation. Curr Opin Cell Biol. 1997;9:174-179. (34) Best WR, Becktel JM, Singleton JW, Kern F, Jr. Development of a Crohn's disease activity index. National Cooperative Crohn's Disease Study. Gastroenterology. 1976;70:439-444. (35) Brummer T, Shaw PE, Reth M, Misawa Y. Inducible gene deletion reveals different roles for B-Raf and Raf-1 in B-cell antigen receptor signalling. EMBO J. 2002;21:5611-5622. (36) Dillon TJ, Karpitski V, Wetzel SA et al. Ectopic B-Raf expression enhances extracellular signal- regulated kinase (ERK) signaling in T cells and prevents antigen-presenting cell-induced anergy. J Biol Chem. 2003;278:35940-35949. (37) Chadee DN, Kyriakis JM. MLK3 is required for mitogen activation of B-Raf, ERK and cell proliferation. Nat Cell Biol. 2004;6:770-776. (38) Grip O, Janciauskiene S, Lindgren S. Macrophages in inflammatory bowel disease. Curr Drug Targets Inflamm Allergy. 2003;2:155-160. (39) Springer TA. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell. 1994;76:301-314. (40) Hausmann M, Spottl T, Andus T et al. Subtractive screening reveals up-regulation of NADPH oxidase expression in Crohn's disease intestinal macrophages. Clin Exp Immunol. 2001;125:48-55. (41) Hausmann M, Obermeier F, Schreiter K et al. Cathepsin D is up-regulated in inflammatory bowel disease macrophages. Clin Exp Immunol. 2004;136:157-167. (42) Rogler G, Hausmann M, Spottl T et al. T-cell co-stimulatory molecules are upregulated on intestinal macrophages from inflammatory bowel disease mucosa. Eur J Gastroenterol Hepatol. 1999;11:1105- 1111. (43) Rogler G, Brand K, Vogl D et al. Nuclear factor kappaB is activated in macrophages and epithelial cells of inflamed intestinal mucosa. Gastroenterology. 1998;115:357-369. (44) Rogler G, Andus T, Aschenbrenner E et al. Alterations of the phenotype of colonic macrophages in inflammatory bowel disease. Eur J Gastroenterol Hepatol. 1997;9:893-899. (45) Ellis RD, Goodlad JR, Limb GA et al. Activation of nuclear factor kappa B in Crohn's disease. Inflamm Res. 1998;47:440-445. (46) Baumann B, Weber CK, Troppmair J et al. Raf induces NF-kappaB by membrane shuttle kinase MEKK1, a signaling pathway critical for transformation. Proc Natl Acad Sci U S A. 2000;97:4615- 4620. (47) Liu Q, Fan J, McMahon M, Prince AM, Zhang P. Role of the oncogenic Raf-1 in orchestration of discrete nuclear factor-kappaB-activating pathways. Mol Cell Biol Res Commun. 2001;4:381-389. (48) Pearson G, English JM, White MA, Cobb MH. ERK5 and ERK2 cooperate to regulate NF-kappaB and cell transformation. J Biol Chem. 2001;276:7927-7931. (49) Bruder JT, Kovesdi I. Adenovirus infection stimulates the Raf/MAPK signaling pathway and induces interleukin-8 expression. J Virol. 1997;71:398-404.

68 Semapimod inhibits c-Raf

(50) Egerton M, Fitzpatrick DR, Kelso A. Activation of the extracellular signal-regulated kinase pathway is differentially required for TCR-stimulated production of six cytokines in primary T lymphocytes. Int Immunol. 1998;10:223-229. (51) Ishizuka T, Terada N, Gerwins P et al. Mast cell tumor necrosis factor alpha production is regulated by MEK kinases. Proc Natl Acad Sci U S A. 1997;94:6358-6363. (52) van der BT, Nijenhuis S, van Raaij E, Verhoef J, van Asbeck BS. Lipopolysaccharide-induced tumor necrosis factor alpha production by human monocytes involves the raf-1/MEK1-MEK2/ERK1-ERK2 pathway. Infect Immun. 1999;67:3824-3829. (53) Xu XS, Vanderziel C, Bennett CF, Monia BP. A role for c-Raf kinase and Ha-Ras in cytokine-mediated induction of cell adhesion molecules. J Biol Chem. 1998;273:33230-33238. (54) Kawaguchi M, Kokubu F, Odaka M et al. Induction of granulocyte-macrophage colony-stimulating factor by a new cytokine, ML-1 (IL-17F), via Raf I-MEK-ERK pathway. J Allergy Clin Immunol. 2004;114:444-450. (55) Geppert TD, Whitehurst CE, Thompson P, Beutler B. Lipopolysaccharide signals activation of tumor necrosis factor biosynthesis through the ras/raf-1/MEK/MAPK pathway. Mol Med. 1994;1:93-103. (56) Odabaei G, Chatterjee D, Jazirehi AR et al. Raf-1 kinase inhibitor protein: structure, function, regulation of cell signaling, and pivotal role in apoptosis. Adv Cancer Res. 2004;91:169-200. (57) Troppmair J, Rapp UR. Raf and the road to cell survival: a tale of bad spells, ring bearers and detours. Biochem Pharmacol. 2003;66:1341-1345. (58) Sitaraman SV, Hoteit M, Gewirtz AT. Semapimod. Cytokine. Curr Opin Investig Drugs. 2003;4:1363- 1368. (59) Bollag G, Freeman S, Lyons JF, Post LE. Raf pathway inhibitors in oncology. Curr Opin Investig Drugs. 2003;4:1436-1441. (60) Cripps MC, Figueredo AT, Oza AM et al. Phase II randomized study of ISIS 3521 and ISIS 5132 in patients with locally advanced or metastatic colorectal cancer: a National Cancer Institute of Canada clinical trials group study. Clin Cancer Res. 2002;8:2188-2192. (61) Cunningham CC, Holmlund JT, Schiller JH et al. A phase I trial of c-Raf kinase antisense oligonucleotide ISIS 5132 administered as a continuous intravenous infusion in patients with advanced cancer. Clin Cancer Res. 2000;6:1626-1631. (62) Herrera R, Sebolt-Leopold JS. Unraveling the complexities of the Raf/MAP kinase pathway for pharmacological intervention. Trends Mol Med. 2002;8:S27-S31. (63) Hotte SJ, Hirte HW. BAY 43-9006: early clinical data in patients with advanced solid malignancies. Curr Pharm Des. 2002;8:2249-2253. (64) Lowinger TB, Riedl B, Dumas J, Smith RA. Design and discovery of small molecules targeting raf-1 kinase. Curr Pharm Des. 2002;8:2269-2278. (65) Mross K, Steinbild S, Baas F et al. Drug-drug interaction pharmacokinetic study with the Raf kinase inhibitor (RKI) BAY 43-9006 administered in combination with irinotecan (CPT-11) in patients with solid tumors. Int J Clin Pharmacol Ther. 2003;41:618-619. (66) Rudin CM, Holmlund J, Fleming GF et al. Phase I Trial of ISIS 5132, an antisense oligonucleotide inhibitor of c-raf-1, administered by 24-hour weekly infusion to patients with advanced cancer. Clin Cancer Res. 2001;7:1214-1220. (67) Strumberg D, Voliotis D, Moeller JG et al. Results of phase I pharmacokinetic and pharmacodynamic studies of the Raf kinase inhibitor BAY 43-9006 in patients with solid tumors. Int J Clin Pharmacol Ther. 2002;40:580-581.

69

Chapter 4

Rapid immunosuppressive effects of glucocorticoids mediated through Lck and Fyn

Mark Löwenberg1,4, Jurriaan Tuynman1, Joyce Bilderbeek1, Timo Gaber2, Frank Buttgereit2, Sander van Deventer1, Maikel Peppelenbosch3, Daniel Hommes4

1 Department for Experimental Internal Medicine, Academic Medical Center, Amsterdam, The Netherlands 2 Department of Rheumatology and Clinical Immunology, Charité University Hospital, Berlin, Germany 3 Department of Cell Biology, University of Groningen, Groningen, The Netherlands 4 Department of Gastroenterology and Hepatology, Academic Medical Center, Amsterdam, The Netherlands

Blood. 2005; 106(5):1703-10

71 Chapter 4

Abstract Glucocorticoids (GCs) are effective immunosuppressive agents and mediate well-defined transcriptional effects via GC receptors. There is increasing evidence that GCs also initiate rapid nongenomic effects. Employing activated human CD4+ lymphocytes and a peptide array containing 1176 different kinase consensus substrates, we generated a comprehensive profile of GC-induced rapid effects on signal transduction. The results showed marked early differences in phosphorylation between GC pretreated cells and control cells including impaired phosphorylation of Lck/Fyn consensus substrates. In vitro kinase assays confirmed rapid GC-induced inhibition of Lck and Fyn kinase activities employing SAM68 as a substrate. Immunoprecipitation experiments demonstrated reduced Lck-CD4 and Fyn-CD3 associations due to GC stimulation. Western blot analysis showed suppressed phosphorylation of a series of downstream TCR signaling intermediates following GC treatment, including PKB, PKC and MAPKs. Experiments with GC receptor negative Jurkat cells and a pharmacological GC receptor ligand (RU486) indicated that rapid inhibition of Lck and Fyn kinases is GC receptor dependent. Parallel experiments conducted following the application of GCs in healthy individuals confirmed suppression of Lck/Fyn in T cells within 1 hr in vivo. These results identify the inhibition of Lck and Fyn as rapid targets of GCs, mediated via a GC receptor dependent pathway.

72 Glucocorticoids impair TCR signaling

Introduction Glucocorticoids (GCs) are widely used therapeutically for immunosuppression. GC action is mediated through the intracellular GC-receptor (GR), present in the cytosol of T-cells. In the inactive state of the receptor, GRs are associated with heat shock proteins, which act as chaperones. Upon GC-binding this complex dissociates and the activated GR translocates into the nucleus where it binds to specific DNA motifs (glucocorticoid-responsive elements) and to transcription factors such as AP1 and NFκB, thereby regulating the expression of a number of genes involved in the immunological process 1-3. Through regulation of gene expression, GCs reduce the production of pro-inflammatory mediators including cytokines (IL-1, IL-2, TNF-α, IFN-γ, etc.), prostaglandins and nitric oxide. Moreover, GCs inhibit expression of adhesion molecules and may induce death of T-cells 4-6. However, the phenomenon of rapid GC-induced effects in cellular systems occurring within minutes is unlikely explained by the classical mode of GC action 7-18. The underlying mechanisms of rapid nongenomic GC- dependent immunosuppression remain to be elucidated. Among the earliest recognizable events after TCR stimulation are the activation of Lck and Fyn, resulting in TCR phosphorylation on tyrosine residues within immunoreceptor tyrosine- based activation motifs (ITAMs) 19. Lck and Fyn, members of the Src family of tyrosine kinases, are expressed in T cells and are critically involved in TCR-mediated signal transduction 20-23. Lck and Fyn associate with the TCR complex, and the cellular organization of these kinases is well coordinated and essential for efficient TCR signaling 24-28. Lck binds to CD4 or CD8 co-receptors, whereas Fyn associates with CD3 29-31. Alternatively, Lck and Fyn can physically interact with CD3 and CD4 co-receptors respectively, but these interactions are weak. The TCR-CD4 association largely determines the quality of TCR signaling, Lck being the critical mediator 32,33. As a subsequent step, Zap70 tyrosine kinase is recruited to the TCR where it is activated by Lck through tyrosine phosphorylation 34. Once activated, Zap70 phosphorylates LAT (linker for activation of T cells), an adapter molecule, which ultimately promotes activation of essential downstream signaling pathways, including PKC, PKB, and MAPKs (i.e. p38 MAPK, ERK and JNK), resulting in T cell activation 35-42. Rapid effects of GCs on these signaling events remain to be established. The importance of GCs in clinical immunosuppression combined with the unknown basis of the nongenomic GC-dependent actions, prompted us to investigate these early effects. Hence,

73 Chapter 4

we chose to screen rapid effects of the GC analogue dexamethasone (DEX) on the CD4+ T cell signal transduction kinome using a peptide array containing 1176 specific kinase pseudo- substrates 43. The results of these experiments show marked differences in phosphorylation patterns between DEX-treated and non-DEX-treated cells, providing proof for rapid DEX- dependent effects on signal transduction. Among the most prominent effects observed upon DEX treatment was reduced phosphorylation of Lck/Fyn substrates. Subsequent in vitro kinase assays and Western blot analysis revealed that DEX impairs activation of both kinases as well as recruitment to the TCR complex. Lck and Fyn kinase activities were found to be reduced in vivo at 1 hour following oral administration of prednisolon in human subjects. Experiments using GR negative Jurkat cells and a pharmacological GR ligand (RU486) indicate GR dependence of DEX-induced Lck and Fyn inhibition. Altogether, these results provide a molecular framework for understanding rapid GC mediated immunosuppression in T cells.

Material and Methods Cell culture. Human peripheral blood mononuclear cells (PBMCs) were isolated from whole blood of healthy volunteers by Ficoll-Isopaque density gradient centrifugation (Amersham Biosciences, Roosendaal, the Netherlands). The monocytes present in the PBMC pellet were removed by an adherence procedure: cells were plated out in 6 well plates (CellStar, Greiner Bio-One) at a final concentration of 5.106cells/well for 2 hrs at 37 °C and cells were harvested for magnetic cell sorting, described below. CD4+ T cells were cultured in Iscove’s Modified Dulbecco’s Medium/ IMDM (Gibco, Breda, the Netherlands), supplemented with 5% heat- inactivated fetal calf serum (FCS), 2mM L-glutamine and penicillin-streptomycin;

‘complete’, in a humidified 5% CO2 environment. E6-1 Jurkat T-lymphocytes (American Type Culture Collection) were cultured in RPMI 1640 media with 10% FCS, 2mM L- glutamine and penicillin-streptomycin; ‘complete’.

CD4+ purification. CD4+ T cells were purified by negative selection using the MACS system. Non-CD4+ cells were indirectly magnetically labeled with a cocktail of biotin-conjugated monoclonal antibodies bound to MicroBeads, as secondary labeling agent (Miltenyi Biotec Inc., Auburn, CA, USA). The magnetically labeled non-CD4+ T cells were depleted by

74 Glucocorticoids impair TCR signaling

retaining them on a MACS Column in the magnetic field of the autoMACS separator (Miltenyi Biotec) and the unlabeled CD4+ T helper cells were collected. Sample purity was assessed by fluorescence-activated cell sorter (Becton Dickinson, San Jose, CA, USA) with PE-conjugated CD3 and FITC-conjugated CD4 monoclonal antibodies (purity >95% CD3+CD4+; data not shown), according to routine procedures.

Reagents and antibodies. Phosphospecific antibodies (Abs) against p38Thr180/Tyr182, SrcTyr416, LATTyr171, PKBSer473, PKCSer660, ZAP70Tyr493, and an Ab against total PKB were purchased from Cell Signaling Technology (Beverly, CA, USA). Phosphospecific Abs obtained from Santa Cruz Biotechnology Inc. (Heidelberg, Germany) included Erk1/2Thr202/Tyr204, JnkThr183/Tyr185, FynThr12 and PY20. Abs recognizing total Lck, Fyn, Src, LAT, Jnk, p38, Erk, PKC, ZAP70, CD3, CD4, actin, and SAM68 were obtained from Santa Cruz Biotechnology. HIF-1α mAb was from Becton Dickinson (Heidelberg, Germany). HRP-conjugated Goat- anti-Rabbit, Goat-anti-Mouse and Rabbit-anti-Goat were purchased from DakoCytomation (Heverlee, Belgium). Anti-human CD3 (CD3-epsilon; mouse) was provided by the group of Prof. Dr. H. Spits (Academic Medical Center, Amsterdam, the Netherlands) 44, and anti-CD28 mAb (mouse IgG1) was from Sanquin (Amsterdam, the Netherlands). DEX, SU6656 (Src family kinase inhibitor), protein-A and G Sepharose, MTT (3-[4,5-dimethylthiazol-2-yl]-2,5- diphenyltetrazolium bromide) were from Sigma-Aldrich (Zwijndrecht, the Netherlands). RU486 (Mifepristone) was from LKT laboratories Inc. (Minnesota, MN, USA). γ-33P-ATP was from Amersham Biosciences. Lysis buffer and kinase buffer were from Cell Signaling Technology. Lysis buffer was supplemented with protease and phosphatase inhibitors (i.e.

1µg/ml NaF, 1µg/ml leupeptin, 1µg/ml aprotinin, 10mM Na3VO4 and 1mM pefabloc, obtained from Merck BV (Amsterdam, the Netherlands)).

Cytokine Bead Array. The effect of DEX on cytokine synthesis was evaluated in supernatants of stimulated T cells by Cytokine Bead Array (BD Biosciences, Alphen a/d Rijn, the Netherlands). Cells were plated out in a 96 well plate (CellStar) at a cell density of 1.105 cells per well, subsequently activated with anti-CD3 and anti-CD28 Abs and incubated overnight at 37 °C in the presence or absence of 10-6 M DEX. This experiment aimed at showing that activated human CD4+ cells indeed represent a valid model for studying GC effects.

75 Chapter 4

GC stimulation of T cells (in vitro and ex vivo). T cells and Jurkat cells were seeded in 6-well plates at a concentration of 5-10.106 cells per well and incubated at 37 °C for 2 hours in complete media. Cells were pretreated for 10 minutes with 10-6 M DEX dissolved in DMSO (Dimethyl Sulfoxide), or DMSO-supplemented media (control). Next, cells were incubated for 15 minutes with anti-CD3 Abs (immobilized on plastic) and soluble anti-CD28 (3µg/ml) and subsequently ice-cold PBS was added. After centrifugation, complete ice-cold lysis buffer was added and cell lysates were subjected to in vitro kinase assay or diluted in sample buffer for immunoblotting. To study the effects of GCs on Lck/Fyn kinase activities ex vivo, whole blood was collected from two healthy volunteers before and 1 hour after oral administration of 20 mg prednisolon. PBMCs were activated ex vivo for 15 minutes (anti-CD3/anti-CD28 Abs). Cell lysates were subjected to Lck and Fyn immunoprecipitation, and in vitro kinase assays were performed. Samples were analyzed on Western blot for phospho-SAM68 expression using PY20.

Kinome array analysis. The protocol of the kinome array is described in detail online: http://www.pepscan.nl. After 10 minute DEX treatment and incubation with anti-CD3 and anti-CD28 Abs (15 minutes), cells were washed in PBS and lysed in lysis buffer. Cell lysates were corrected for protein concentrations using Bradford analysis (Biorad, Veenendaal, the Netherlands). To study kinase activities, 50µl cell lysate was added to 10µl activation mix, containing 50% glycerol, 50µM ATP, 60mM MgCl2, 0.05% v/v Brij-35, 0.25mg/ml BSA and 2000µ Ci/ml γ-33P-ATP. The peptide arrays (Pepscan, Lelystad, the Netherlands), containing 1176 different kinase pseudo-substrates in duplo 43, were incubated with cell lysates for 2 hours in a humidified stove at 37 °C. The arrays were washed in 2M NaCl, 1% triton-x-100,

0.1% tween in H2O, where after slides were exposed to a phospho-imaging screen for 24-72 hours and scanned on a phospho-imager (Fuji, Stamford, CO, USA).

Data acquisition and statistical analysis of pepchip arrays. Acquisition of peptide arrays was performed with a phospho-imager (Fuji) and quantificated using ArrayVision 6.0 software (Molecular Dynamics, Sunnyvale, CA, USA). Subsequently, the data were exported to a spreadsheet program (Microsoft Excel 2002, Microsoft, Redmond, WA, USA). We corrected the spot density for the individual background to diminish interarray variances. In addition,

76 Glucocorticoids impair TCR signaling

the variation between arrays and individual experiments was reduced by normalization to the 75% percentile of the intensity of each array. Differential kinase activities in lysates from activated cells incubated in the presence or absence of DEX were determined by the comparison of the median density of the spots of each condition using the algorithm originally developed for microarray analysis (http://www.stat.stanford.edu/ ~tibs/SAM) and fold change ratios. In short, inconsistent data (i.e. SD between the data points > 1.96 of the mean value) were excluded from further analysis. Second, spots were averaged and included for dissimilarity measurement in order to extract kinases of which activity was either significantly induced or reduced. Alternatively, differential kinase activities were analyzed using a ranking method. The full list of peptides spotted on the peptide array can be found online: http://www.pepscan.nl

Preparation of cell lysates and Western blot analysis. CD4+ T cells and Jurkat cells were pretreated with 10-6 M DEX for 10 minutes and stimulated with anti-CD3/ anti-CD28 Abs for 15 minutes, where after ice-cold PBS was added. Cells were harvested in sample buffer (150mM Tris-HCl pH6.8, 30% glycerol, 6%SDS, 3%β-mercaptoethanol and broom phenol- blue), sonificated and heated to 90 ºC for five minutes. Whole cell extracts were loaded on 10% SDS-PAGE and subsequently transferred to a polyvinylidene difluoride membrane (PVDF, Immobilon-P; Millipore, Amsterdam, the Netherlands). The membranes were blocked with 1% Protifar (Nutricia, Zoetermeer, the Netherlands) in TBS/T (0.05M Tris, 150mM NaCl, 0.05% Tween-20). Primary antibodies and secondary HRP-conjugated antibodies were diluted in 1% Protifar in TBS/T. Proteins were visualized using the Lumi- LightPLUS substrate (Roche, Woerden, the Netherlands). Blots were subsequently stripped with strip buffer (62.5mM Tris-HCl pH6.8, 100mM β-mercaptoethanol, 2% SDS) and reprobed with adequate antibodies for evaluation of equal protein loading.

Immunoprecipitation and in vitro kinase assay. CD4+ T cells were lysed in complete non- denaturing lysis buffer after a 10 minute DEX treatment (10-6 M) and stimulation with anti- CD3, anti-CD28 Abs for 15 minutes. Lysates were subjected to immunoprecipitation with the indicated Abs for Lck and Fyn. First, a pre-clearance step was performed by incubating the samples with protein-A or protein-G Sepharose for 2 hours at 4 °C. Lysates were then

77 Chapter 4

centrifuged at 13.000 rpm for 10 minutes and supernatants were incubated overnight with Lck or Fyn specific Abs at 4C°. Samples were subsequently incubated for 2-3 hours with a Sepharose-conjugated polyclonal antibody at 4 °C. Next, kinase buffer supplemented with 200 µM ATP and 2 µg/ml SAM68 45 was added to the immunoprecipitates and in vitro kinase reactions were performed at 30 ºC for 30 minutes. Immunoprecipitates were dissolved in sample buffer, loaded on 10% SDS-PAGE, blocked with 2.5% BSA in TBS/T (0.05M Tris, 100mM NaCl, 0.05% Tween 20), and immunoblotted using PY20 and a secondary HRP- conjugated Ab. Alternatively, cell lysates were incubated with 2.10-6 M SU6656 (a selective Src family kinase inhibitor) for 45 minutes on ice prior to in vitro kinase assay.

Experiments with RU486, a pharmacological GR ligand. CD4+ lymphocytes were pretreated for 10 minutes with increasing concentrations of DEX (10-11; 10-8; 10-5 M) or RU486 (10-7; 10-6; 10-5; 10-4 M) in complete IMDM (4.106 cells/ml) at 37 ºC. Subsequently, cells were incubated for 6 hours in closed microtubes in order to induce hypoxia and compared with noncapped incubations (i.e. hypoxia versus control). After incubation, cells were centrifuged, and whole cell extracts were analyzed on Western blot using Abs against HIF-1α (Hypoxia Inducible Factor-1α) and actin. In addition, cells were pretreated (1hr, 37 °C) with increasing RU486 concentrations (50.10-9; 50.10-8; 50.10-7; 50.10-6 M), subsequently incubated with 10-6 M DEX (10 min.) and activated using anti-CD3, anti-CD28 Abs (15 min.). Lysates, subjected to immunoprecipitation using anti-Fyn mAbs, were then used for in vitro kinase assay and Western blotting. Finally, the effects of RU486 and DEX alone, as well as RU486 and DEX in combination, were compared in pre-incubations of CD4+ T cells (50 µM RU486, 1 hr, 37 ºC), followed by a 10 minute DEX treatment (10-6 M) and subsequent incubation with anti- CD3, anti-CD28 Abs for 15 minutes. After performing immunoprecipitation and in vitro kinase assays, Western blotting was conducted using PY20.

Supplemental Material. The supplemental information discloses the results obtained with the pepchip experiment (available at the Blood website: http://www.bloodjournal.org).

78 Glucocorticoids impair TCR signaling

Results DEX inhibits cytokine production in activated CD4+ T cells CD4+ cells were activated with anti-CD3 and anti-CD28 Abs, incubated with 10-6 M DEX or control media and supernatants were collected after overnight incubation. DEX treatment reduced the secretion of IL-2, IFN-γ and TNF-α (Figure 1), and this was not a consequence of reduced cell viability (as assessed by 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide [MTT] colorimetric assay; data not shown). Thus, incubation of activated human CD4+ cells with DEX has a potent effect on CD4+ T cell function.

Figure 1. DEX inhibits IL-2, IFN-γ, and TNF-α synthesis in activated CD4+ T cells. Cells were stimulated with anti-CD3 and anti-CD28 Abs and incubated in the presence or absence of 10-6 M DEX. Supernatants were collected after overnight culturing and cytokine levels were measured using cytokine bead array (CBA) analysis. Results are expressed as the mean ± SD of triplicate determinations. DEX, dexamethasone.

DEX rapidly alters kinomic profiles in activated CD4+ T cells To investigate the early effects of GCs on the kinome, CD4+ T cells were incubated for 10 minutes with DEX and then stimulated for 15 minutes with anti-CD3 and anti-CD28 Abs. Cell lysates were incubated on peptide arrays (pepchip) containing 1176 different kinase consensus substrates spotted in duplo. Analysis of kinomic profiles revealed 116 differential kinase-substrates with either significantly increased or decreased phosphorylation upon DEX treatment. The dot plot depicted in Figure 2A shows the median density of the spots for DEX versus control conditions. Figure 2B shows the same data set analyzed using a ranking method. Among the most prominent effects of DEX treatment, was significant decreased phosphorylation of Lck/Fyn-consensus substrates. A complete list of the peptide substrates

79 Chapter 4

with significantly altered phosphorylation upon DEX treatment can be found as supplemental information. Thus, these results show that short-term treatment with DEX significantly influences kinase activities upon TCR ligation and reveal that the rapid immunosuppressive effect of GC treatment correlates with profound alterations in cellular signaling.

Figure 2. Kinome analysis of rapid DEX effects in activated CD4+ T cells. (A) The dot plot represents the phosphorylation status of specific peptide-substrates spotted on the pepchips. Differential kinase activities in lysates from activated cells incubated in the presence or absence of DEX using median densities of the spots are shown. Each spot represents the amount of phosphorylation of a specific peptide substrate. Lck/Fyn kinase consensus substrates demonstrated significantly decreased phosphorylation (i.e. 0.32 and 0.51 fold changes) due to DEX. A full description of the substrates with significantly altered phosphorylation upon DEX treatment is provided as supplemental material. (B) The same data-set was analyzed using a ranking method; each spot representing the phosphorylation status of a specific kinase pseudo-substrate. Due to the ranking method, a bisymmetric distribution is generated of peptides of which phosphorylation was either significantly increased or decreased due to DEX. Peptide substrates demonstrating increased phosphorylation (corresponding with higher ranks of peptides) are reflected by an equivalent number of peptides of which phosphorylation was decreased due to DEX (lower ranks of peptides). Lck/Fyn kinase consensus substrates are marked by an arrow. Spots representing peptides of which phosphorylation is significant decreased (A), increased (C) or unaltered (B) due to DEX treatment are shown. DEX, dexamethasone.

DEX inhibits Lck and Fyn kinase activity in vitro To verify the data of the pepchip analysis, enzymatic activities of Lck/Fyn kinases were assessed using in vitro phosphorylation of the Src-family substrate SAM68 (Src-Associated in Mitosis) 45,46. Prior to DEX treatment and TCR stimulation (anti-CD3/ anti-CD28 Abs), cells were pretreated for 45 minutes with a Src family kinase inhibitor (SU6656). As shown in Figure 3A, TCR stimulation leads to SAM68 phosphorylation. SAM68 phosphorylation

80 Glucocorticoids impair TCR signaling

depended on Src family kinase activity, as treatment with SU6656 resulted in a complete abrogation of SAM68 phosphorylation. DEX treated cells were not capable of sustaining SAM68 phosphorylation. These assays provide direct evidence that short exposure to DEX interferes with Src family tyrosine kinase activation. Immunoprecipitation of Lck and Fyn and subsequent testing for in vitro enzymatic activities (Figure 3B) demonstrated the specific involvement of Lck and Fyn, two Src family members. SAM68 phosphorylation was impaired in Lck/Fyn immunoprecipitates prepared from DEX treated cells. Hence, DEX rapidly inhibits Lck and Fyn kinases in activated CD4+ T cells, corroborating the pepchip data.

Figure 3. DEX inhibits Lck and Fyn kinase activities. (A) T cells were pre-incubated for 45 minutes with the Src family kinase inhibitor SU6656, followed by 10 min treatment with DEX or DMSO (control), and 15 min activation (anti-CD3 and anti-CD28 Abs). Cell lysates were supplemented with SAM68 and in vitro kinase assays were performed. Src-like kinase activity was analyzed on Western blot using PY20. Equal loading was evaluated with an Ab against actin. (B) Cells were pretreated for 10 min with DEX and subsequently activated for 15 min. Cell lysates were subjected to immunoprecipitation using anti-Lck and anti-Fyn Abs, followed by in vitro kinase assays. Western blots were analyzed for SAM68 phosphorylation using PY20. Anti-Lck and anti- Fyn Abs were used to test for equal loading. Similar results were obtained in three independent experiments. DEX, dexamethasone; IB, immunoblotting; IP, immunoprecipitation.

DEX suppresses phosphorylation of signaling molecules downstream of the TCR To further investigate the effects of DEX treatment on early steps of TCR signal transduction, the activation status of several signaling molecules downstream of the TCR was analyzed on Western blot using a panel of phosphospecific Abs. As can be seen from Figure 4, DEX rapidly suppresses phosphorylation of Fyn, Src, LAT, PKB, PKC, Erk1/2, Jnk and p38 MAPK in activated T cells. These observations demonstrate that DEX interferes with Lck and Fyn-dependent TCR signal transduction.

81 Chapter 4

Figure 4. DEX inhibits TCR signaling. Cells were pre-incubated for 10 min with or without DEX and stimulated for 15 min (anti-CD3 and anti-CD28 Abs). Phosphorylation status of downstream TCR signaling intermediates was analyzed in whole lysates using phosphospecific Abs. An Ab against phosphorrylated Lck, associated with increased activity, was not available 47. The Ab we employed to detect phosphorylated-Src may cross-react with other Src kinase family members (i.e. Fyn, Lck, Lyn, Yes, Hck, Src) when phosphorylated at equivalent sites. To test for equal loading blots were reprobed with the indicated Abs. A representative experiment, out of 3, is shown. LAT, linker for activation of T cells; PKC, Protein Kinase C; PKB, Protein Kinase B; DEX, dexamethasone; p, phosphorylated.

DEX affects Lck-CD4 and Fyn-CD3 interactions Next, it was assessed whether DEX affects the spatial distribution of Lck and Fyn within the cell membrane, as translocation of these kinases to the TCR complex is crucial for Lck and Fyn activation and efficient TCR signaling 48,49. Lysates from DEX treated cells (CD3/CD28 activated) were subjected to Lck and Fyn immunoprecipitation and Western blot analysis using anti-CD4 and anti-CD3 Abs. Lysates prepared from activated cells displayed CD4 and CD3 expression in Lck and Fyn immunoprecipitates respectively (Figure 5). In contrast, CD4 expression was reduced in Lck-immunoprecipitates prepared from activated cells treated with DEX. Similarly, CD3 signals were suppressed in Fyn-immunoprecipitates following DEX exposure. Effects of DEX on Lck-CD3 and Fyn-CD4 associations were not evident; serving as a control experiment. Apparently, DEX treatment rapidly affects Lck-CD4 and Fyn-CD3 associations.

82 Glucocorticoids impair TCR signaling

Figure 5. DEX affects Lck-CD4 and Fyn-CD3 associations. Cells were pre-incubated with DEX (10 min) and stimulated for 15 min (anti-CD3/anti-CD28 Abs). Cell lysates were subjected to immunoprecipitation and Lck/Fyn immunoprecipitates were subsequently analyzed for the presence of CD4 or CD3 on Western blot. Immunoblots were analyzed for total Lck and Fyn to test for equal loading. To ensure specificity in these experiments, CD3 expression was analyzed in Lck-immunoprecipitates and CD4 was studied in Fyn- immunoprecipitates. Three experiments were performed and comparable results were obtained. DEX, dexamethasone; IB, immunoblotting; IP, immunoprecipitation.

DEX-induced inhibition of Lck/Fyn kinase activity is GR dependent To address the question as to whether DEX-inhibition of Lck/Fyn kinases is GR dependent or independent, GR negative Jurkat cells were pretreated with DEX for 10 minutes and incubated with anti-CD3/anti-CD28 Abs for 15 minutes. Western blot analysis was performed using Abs against phospho-Fyn and phospho-Zap70, a downstream target of Lck. DEX treatment did not affect phospho-Fyn nor phospho-Zap70 expression in activated Jurkat cells (Figure 6A), suggesting that GC effects on Lck/Fyn require a functional GR. In addition, a GR mimetic (RU486) was used to further investigate the role of the GR on Lck/Fyn activities. Depending on the concentration, RU486 may either stimulate or inhibit the GR (dualistic agonistic or antagonistic actions). The effects of RU486 on GR activation were studied in T cells activated with hypoxia. Hypoxia strongly induces HIF1α (Hypoxia Inducible Factor1α) expression (not shown), and DEX (at 10-8; 10-5 M) reduces HIF1α expression. At 10-7; 10-6; 10-5 M RU486 concentrations the effect of RU486 was neutral. Of note, cells incubated with the highest concentrations of RU486 (10-4 M) showed the opposite effect (i.e. reduced HIF1α protein levels). DEX and RU486 synergized in suppressing hypoxia-induced HIF1α expression at 10-11 M DEX and 10-5 M RU486. Additional evidence that RU486 has a dualistic nature was obtained in experiments in which activated CD4+ cells were pretreated with increasing RU486 concentrations together with DEX (10-6 M) followed by in vitro kinase assay. Pretreatment with the highest RU486 concentration (50 µM) showed an agonistic effect

83 Chapter 4

on Fyn kinase activity: i.e. reduced phospho-SAM68 phosphorylation on Western blot was seen (not shown). The lowest RU486 concentration (10-9 M) was not able to block the inhibitory effects of DEX on Fyn kinase activity: i.e. reduced phospho-SAM68 expression was observed (not shown). Cells pretreated with intermediate RU486 concentrations (10-8 M) together with DEX demonstrated phospho-SAM68 expression, suggesting an inhibitory effect of RU486 on DEX-dependent Fyn inhibition (i.e. antagonistic effect). Finally, we studied the effects of the highest RU486 concentration (50 µM, 1 hr) on Lck and Fyn kinases in activated T cells in the absence or presence of DEX (10-6 M). Cells incubated with RU486 (in the absence of DEX) exhibited strongly reduced phospho-SAM68 expression (Figure 6B) in Lck and Fyn immunoprecipitates (i.e. agonistic effect). Altogether, these experiments support rapid GR-mediated inhibition of Lck/Fyn kinase activities.

Figure 6. DEX-induced inhibition of Lck and Fyn activities is GR dependent. (A) GR negative Jurkat cells (confirmed by Western blot analysis; not shown) were pretreated with DEX for 10 min, and incubated with anti-CD3/anti-CD28 Abs for 15 min. Whole cell lysates were analyzed on Western blot. Since there are no Abs available against phospho-Lck, associated with increased activity, phospho-Zap70 expression (a downstream Lck target) was immunoblotted. Anti-Zap and anti-Fyn Abs were used to control for equal loading. (B) T cells were incubated for 1 hr with 50 µM RU486, a GR mimetic, which acts as a GR agonist at this concentration. Next, cells were treated with DEX (10 min.) and activated using anti-CD3/anti-CD28 Abs (15 min.). Lck and Fyn immunoprecipitates were subjected to in vitro kinase assay and phospho-SAM68 expression was immunoblotted using PY20. Appropriate Abs were used to evaluate for equal protein levels. Experiments were performed three times and representative outcomes are shown. DEX, dexamethasone; IB, immunoblotting; IP, immunoprecipitation; p, phosphorylated.

84 Glucocorticoids impair TCR signaling

Glucocorticoid-induced suppression of Lck and Fyn activities in vivo GC effects on Lck and Fyn kinase activities were studied in PBMCs isolated from 2 individuals before and at 1 hour following oral administration of a single dose of 20 mg prednisolon. In isolated cells stimulated ex vivo for 15 minutes (anti-CD3/ anti-CD28 Abs), Western blot analysis revealed suppressed SAM68 phosphorylation in Lck and Fyn immunoprecipitates at 1 hour following prednisolon treatment in both experiments (Figure 7). Altogether, these studies indicate that GCs rapidly inhibit Lck and Fyn activities both in vitro and in vivo.

Figure 7. Glucocorticoid-induced inhibition of Lck/Fyn activities in vivo. PBMCs were isolated from whole blood obtained from two healthy volunteers (#1; #2) before and 1 hr after oral administration of 20 mg prednisolon. Cells were stimulated ex vivo for 15 min using anti-CD3/CD28 Abs and cell lysates were subjected to Lck and Fyn immunoprecipitation. In vitro kinase assays were performed using SAM68 as a substrate and samples were immunoblotted using PY20. Anti-Lck and anti-Fyn Abs were used to evaluate for equal protein loading. IB, immunoblotting; IP, immunoprecipitation.

Discussion GCs form the basis of current immunosuppressive therapy. The classical mechanism of GC action involves the GR and modulation of transcriptional and translational events. Over the past years increasing evidence for rapid nongenomic GC action has accumulated, that cannot be explained by the traditional mechanistic model 8-11. The molecular mechanisms underlying the early effects of GCs are poorly characterized. The present study was undertaken to study rapid effects of GC action in activated human CD4+ T cells. Using an array of kinase pseudo-

85 Chapter 4

substrates we demonstrated that the activities of multiple kinases are rapidly altered in activated T cells following short term treatment with DEX, a synthetic GC analogue. Further analysis revealed that DEX inhibits Lck and Fyn kinases within minutes in these immune cells. This was demonstrated in vitro as well as in vivo in healthy individuals who had received GCs. Phosphorylation of Lck and Fyn are proximal events in T cell activation. Lck and Fyn positively regulate signaling initiated upon TCR stimulation through a variety of downstream pathways. Accordingly, reduced Lck and Fyn kinase activities may have an important role in the fast immunosuppressive effects of GCs in immune cells. In line with this notion, decreased activation of several signaling pathways downstream of the TCR upon DEX treatment was observed, including suppression of PKB, PKC, Erk, Jnk and p38 MAPKs. It is known that the biochemical and functional responses to TCR ligands are largely determined by Fyn-CD3 and Lck-CD4 associations 50-52. The data reported here show that DEX treatment rapidly alters the cellular distribution of Lck and Fyn which would likely result in decreased Lck/Fyn kinase activities and suppressed TCR signaling. Others have previously reported that DEX disturbs the submembrane localization of Lck and Fyn in a murine T cell hybridoma, since lipid rafts purified from DEX-treated cells displayed a decrease in Lck and Fyn protein concentration 53. However, in those experiments DEX treatment did not affect cellular expression levels of Lck and Fyn kinases, as measured by Western blot performed on whole cell extracts. Furthermore, in this particular study no effect of DEX on Lck/Fyn kinase activities was noted. We assume that these discrepant findings could be due to differences in experimental set up. Murine T cells were used in their study, and it is known that murine cells markedly differ from human T cells in their GC activation responses 54-58. Also the DEX incubation time of T cells was considerably different: they used a 16 hour DEX pre- incubation time in contrast to the 10 minutes in our study. It is well possible that feedback mechanisms upon GC treatment were activated within the 16 hour time frame, counteracting the negative effects on Lck/Fyn activation and the spatial distribution of these kinases. Finally, it is well possible that the observed early effects are short-lived and as a result come and go rapidly. Recent advances in nuclear hormone receptor biology provide evidence for novel types of receptors binding steroids and mediating rapid signaling events 59. Are the early GC-effects that we observed in activated T-cells, GR dependent or GR independent? Phospho-ZAP70

86 Glucocorticoids impair TCR signaling

(Lck substrate) and phospho-Fyn protein expression in Jurkat cells, which do not express GR, did not respond to short-term DEX treatment. These results are supported by experiments in which CD4+ T cells were treated with agonistic concentrations of RU486, a pharmacologic GR ligand. Again a strict correlation between GR activation and inhibition of Lck and Fyn kinase activities was observed. In conclusion, we have identified Lck and Fyn, key players in TCR activation, as rapid targets of GC action. These observations indicate that selective targeting of Lck and Fyn might constitute a potent immunosuppressive therapeutic approach.

Acknowledgments We acknowledge Inge Pronk, Auke Verhaar and Dennis van der Coelen for technical support.

87 Chapter 4

Reference List

(1) Boumpas DT, Paliogianni F, Anastassiou ED, Balow JE. Glucocorticosteroid action on the immune system: molecular and cellular aspects. Clin Exp Rheumatol. 1991;9:413-423. (2) Cato AC, Wade E. Molecular mechanisms of anti-inflammatory action of glucocorticoids. Bioessays. 1996;18:371-378. (3) Franchimont D. Overview of the actions of glucocorticoids on the immune response: a good model to characterize new pathways of immunosuppression for new treatment strategies. Ann N Y Acad Sci. 2004;1024:124-137. (4) Ashwell JD, Lu FW, Vacchio MS. Glucocorticoids in T cell development and function*. Annu Rev Immunol. 2000;18:309-345. (5) Barnes PJ, Adcock I. Anti-inflammatory actions of steroids: molecular mechanisms. Trends Pharmacol Sci. 1993;14:436-441. (6) Cohen JJ, Duke RC. Glucocorticoid activation of a calcium-dependent endonuclease in thymocyte nuclei leads to cell death. J Immunol. 1984;132:38-42. (7) Baus E, Andris F, Dubois PM, Urbain J, Leo O. Dexamethasone inhibits the early steps of antigen receptor signaling in activated T lymphocytes. J Immunol. 1996;156:4555-4561. (8) Buttgereit F, Scheffold A. Rapid glucocorticoid effects on immune cells. Steroids. 2002;67:529-534. (9) Croxtall JD, Choudhury Q, Flower RJ. Glucocorticoids act within minutes to inhibit recruitment of signalling factors to activated EGF receptors through a receptor-dependent, transcription-independent mechanism. Br J Pharmacol. 2000;130:289-298. (10) Falkenstein E, Wehling M. Nongenomically initiated steroid actions. Eur J Clin Invest. 2000;30 Suppl 3:51-54. (11) Falkenstein E, Tillmann HC, Christ M, Feuring M, Wehling M. Multiple actions of steroid hormones--a focus on rapid, nongenomic effects. Pharmacol Rev. 2000;52:513-556. (12) Schmidt BM, Gerdes D, Feuring M et al. Rapid, nongenomic steroid actions: A new age? Front Neuroendocrinol. 2000;21:57-94. (13) Wehling M. Specific, nongenomic actions of steroid hormones. Annu Rev Physiol. 1997;59:365-393. (14) Cato AC, Nestl A, Mink S. Rapid actions of steroid receptors in cellular signaling pathways. Sci STKE. 2002;2002:RE9. (15) Falkenstein E, Norman AW, Wehling M. Mannheim classification of nongenomically initiated (rapid) steroid action(s). J Clin Endocrinol Metab. 2000;85:2072-2075. (16) Hafezi-Moghadam A, Simoncini T, Yang E et al. Acute cardiovascular protective effects of corticosteroids are mediated by non-transcriptional activation of endothelial nitric oxide synthase. Nat Med. 2002;8:473-479. (17) Buttgereit F. Mechanisms and clinical relevance of nongenomic glucocorticoid actions. Z Rheumatol. 2000;59 Suppl 2:II/119-II/123. (18) Buttgereit F, Straub RH, Wehling M, Burmester GR. Glucocorticoids in the treatment of rheumatic diseases: an update on the mechanisms of action. Arthritis Rheum. 2004;50:3408-3417. (19) Zamoyska R, Basson A, Filby A et al. The influence of the src-family kinases, Lck and Fyn, on T cell differentiation, survival and activation. Immunol Rev. 2003;191:107-118. (20) Allison JP, Havran WL. The immunobiology of T cells with invariant gamma delta antigen receptors. Annu Rev Immunol. 1991;9:679-705. (21) Cooke MP, Perlmutter RM. Expression of a novel form of the fyn proto-oncogene in hematopoietic cells. New Biol. 1989;1:66-74. (22) Janeway CA, Jr. The T cell receptor as a multicomponent signalling machine: CD4/CD8 coreceptors and CD45 in T cell activation. Annu Rev Immunol. 1992;10:645-674. (23) Palacios EH, Weiss A. Function of the Src-family kinases, Lck and Fyn, in T-cell development and activation. Oncogene. 2004;23:7990-8000. (24) Collins TL, Uniyal S, Shin J et al. p56lck association with CD4 is required for the interaction between CD4 and the TCR/CD3 complex and for optimal antigen stimulation. J Immunol. 1992;148:2159-2162. (25) Ehrlich LI, Ebert PJ, Krummel MF, Weiss A, Davis MM. Dynamics of p56lck translocation to the T cell immunological synapse following agonist and antagonist stimulation. Immunity. 2002;17:809-822. (26) Rivas A, Takada S, Koide J, Sonderstrup-McDevitt G, Engleman EG. CD4 molecules are associated with the antigen receptor complex on activated but not resting T cells. J Immunol. 1988;140:2912-2918.

88 Glucocorticoids impair TCR signaling

(27) Rudd CE, Anderson P, Morimoto C, Streuli M, Schlossman SF. Molecular interactions, T-cell subsets and a role of the CD4/CD8:p56lck complex in human T-cell activation. Immunol Rev. 1989;111:225- 266. (28) Li QJ, Dinner AR, Qi S et al. CD4 enhances T cell sensitivity to antigen by coordinating Lck accumulation at the immunological synapse. Nat Immunol. 2004;5:791-799. (29) Dianzani U, Shaw A, al Ramadi BK, Kubo RT, Janeway CA, Jr. Physical association of CD4 with the T cell receptor. J Immunol. 1992;148:678-688. (30) Kanazawa S, Ilic D, Hashiyama M et al. p59fyn-p125FAK cooperation in development of CD4+CD8+ thymocytes. Blood. 1996;87:865-870. (31) Turner JM, Brodsky MH, Irving BA et al. Interaction of the unique N-terminal region of tyrosine kinase p56lck with cytoplasmic domains of CD4 and CD8 is mediated by cysteine motifs. Cell. 1990;60:755- 765. (32) van Oers NS, Killeen N, Weiss A. Lck regulates the tyrosine phosphorylation of the T cell receptor subunits and ZAP-70 in murine thymocytes. J Exp Med. 1996;183:1053-1062. (33) Veillette A, Bookman MA, Horak EM, Samelson LE, Bolen JB. Signal transduction through the CD4 receptor involves the activation of the internal membrane tyrosine-protein kinase p56lck. Nature. 1989;338:257-259. (34) Iwashima M, Irving BA, van Oers NS, Chan AC, Weiss A. Sequential interactions of the TCR with two distinct cytoplasmic tyrosine kinases. Science. 1994;263:1136-1139. (35) Denny MF, Kaufman HC, Chan AC, Straus DB. The lck SH3 domain is required for activation of the mitogen-activated protein kinase pathway but not the initiation of T-cell antigen receptor signaling. J Biol Chem. 1999;274:5146-5152. (36) Denny MF, Patai B, Straus DB. Differential T-cell antigen receptor signaling mediated by the Src family kinases Lck and Fyn. Mol Cell Biol. 2000;20:1426-1435. (37) DeSilva DR, Jones EA, Feeser WS, Manos EJ, Scherle PA. The p38 mitogen-activated protein kinase pathway in activated and anergic Th1 cells. Cell Immunol. 1997;180:116-123. (38) Janeway CA, Jr., Bottomly K. Signals and signs for lymphocyte responses. Cell. 1994;76:275-285. (39) Mustelin T, Tasken K. Positive and negative regulation of T-cell activation through kinases and phosphatases. Biochem J. 2003;371:15-27. (40) Nel AE, Slaughter N. T-cell activation through the antigen receptor. Part 2: role of signaling cascades in T-cell differentiation, anergy, immune senescence, and development of . J Allergy Clin Immunol. 2002;109:901-915. (41) Nel AE. T-cell activation through the antigen receptor. Part 1: signaling components, signaling pathways, and signal integration at the T-cell antigen receptor synapse. J Allergy Clin Immunol. 2002;109:758-770. (42) Salojin KV, Zhang J, Delovitch TL. TCR and CD28 are coupled via ZAP-70 to the activation of the Vav/Rac-1-/PAK-1/p38 MAPK signaling pathway. J Immunol. 1999;163:844-853. (43) Diks SH, Kok K, O'Toole T et al. Kinome profiling for studying lipopolysaccharide signal transduction in human peripheral blood mononuclear cells. J Biol Chem. 2004. (44) Yssel H, Spits H, de Vries JE. A cloned human T cell line cytotoxic for autologous and allogeneic B lymphoma cells. J Exp Med. 1984;160:239-254. (45) Fumagalli S, Totty NF, Hsuan JJ, Courtneidge SA. A target for Src in mitosis. Nature. 1994;368:871- 874. (46) Taylor SJ, Shalloway D. An RNA-binding protein associated with Src through its SH2 and SH3 domains in mitosis. Nature. 1994;368:867-871. (47) Hardwick JS, Sefton BM. The activated form of the Lck tyrosine protein kinase in cells exposed to hydrogen peroxide is phosphorylated at both Tyr-394 and Tyr-505. J Biol Chem. 1997;272:25429- 25432. (48) Ehrlich LI, Ebert PJ, Krummel MF, Weiss A, Davis MM. Dynamics of p56lck translocation to the T cell immunological synapse following agonist and antagonist stimulation. Immunity. 2002;17:809-822. (49) Xavier R, Brennan T, Li Q, McCormack C, Seed B. Membrane compartmentation is required for efficient T cell activation. Immunity. 1998;8:723-732. (50) Madrenas J, Chau LA, Smith J, Bluestone JA, Germain RN. The efficiency of CD4 recruitment to ligand-engaged TCR controls the agonist/partial agonist properties of peptide-MHC molecule ligands. J Exp Med. 1997;185:219-229.

89 Chapter 4

(51) Abraham N, Miceli MC, Parnes JR, Veillette A. Enhancement of T-cell responsiveness by the lymphocyte-specific tyrosine protein kinase p56lck. Nature. 1991;350:62-66. (52) Caron L, Abraham N, Pawson T, Veillette A. Structural requirements for enhancement of T-cell responsiveness by the lymphocyte-specific tyrosine protein kinase p56lck. Mol Cell Biol. 1992;12:2720-2729. (53) Van Laethem F, Baus E, Smyth LA et al. Glucocorticoids attenuate T cell receptor signaling. J Exp Med. 2001;193:803-814. (54) Ahmed SA, Sriranganathan N. Differential effects of dexamethasone on the thymus and spleen: alterations in programmed cell death, lymphocyte subsets and activation of T cells. Immunopharmacology. 1994;28:55-66. (55) Brunetti M, Martelli N, Colasante A et al. Spontaneous and glucocorticoid-induced apoptosis in human mature T lymphocytes. Blood. 1995;86:4199-4205. (56) Migliorati G, Nicoletti I, Nocentini G, Pagliacci MC, Riccardi C. Dexamethasone and interleukins modulate apoptosis of murine thymocytes and peripheral T-lymphocytes. Pharmacol Res. 1994;30:43- 52. (57) Ranelletti FO, Maggiano N, Aiello FB et al. Glucocorticoid receptors and corticosensitivity of human thymocytes at discrete stages of intrathymic differentiation. J Immunol. 1987;138:440-445. (58) Tuosto L, Cundari E, Montani MS, Piccolella E. Analysis of susceptibility of mature human T lymphocytes to dexamethasone-induced apoptosis. Eur J Immunol. 1994;24:1061-1065. (59) Revankar CM, Cimino DF, Sklar LA, Arterburn JB, Prossnitz ER. A Transmembrane Intracellular Estrogen Receptor Mediates Rapid Cell Signaling. Science. 2005.

90

Chapter 5

Glucocorticoids cause rapid dissociation of a T cell receptor-associated protein complex containing Lck and

Fyn

Mark Löwenberg1,5, Auke P Verhaar1, Joyce Bilderbeek1, Jan van Marle2, Frank Buttgereit3, Maikel P Peppelenbosch4, Sander J van Deventer1, Daniel W Hommes5

1 Department for Experimental Internal Medicine, Academic Medical Center, Amsterdam, The Netherlands 2 Department of Cell biology and Histology, Academic Medical Center, Amsterdam, The Netherlands 3 Department of Rheumatology and Clinical Immunology, Charité University Hospital, Berlin, Germany 4 Department of Cell Biology, University of Groningen, Groningen, The Netherlands 5 Department of Gastroenterology and Hepatology, Academic Medical Center, Amsterdam, The Netherlands

EMBO reports. Accepted for publication

91 Chapter 5

Abstract Glucocorticoid (GC)-induced nongenomic effects have been reported, but the responsible mechanisms remain unexplained. We previously described that Lck and Fyn mediate GC- induced inhibition of T cell receptor (TCR) signaling in vitro and in humans, and we here characterize the underlying molecular mechanism. The present study shows that the GC receptor is part of a TCR-linked multiprotein complex containing Hsp90, Lck and Fyn, which is essential for TCR-dependent Lck/Fyn activation. Experiments with GC receptor siRNA- transfected cells revealed that the GC receptor is an essential component of the TCR signaling complex. Short-term GC treatment induces dissociation of this protein complex resulting in impaired TCR signaling as a consequence of abrogated Lck/Fyn activation. Hsp90siRNA- transfected cells are not capable of assembling this TCR-associated multiprotein complex, and accordingly Hsp90siRNA treatment mimics GC effects on Lck and Fyn kinase activities. These observations support a model for nongenomic GC-induced immunosuppression based on dissolution of membrane-bound GC receptor multiprotein complexes following GC receptor ligation.

92 TCR-associated glucocorticoid receptors

Introduction Glucocorticoids (GCs) mediate their immunosuppressive effects through cytosolic ligand- inducible receptors 1. Inactive GC-receptors (GRs) are associated with (co)chaperones, such as Hsp90, which dissociate upon GR ligation, followed by nuclear translocation of GR and regulation of gene transcription 2,3. GCs also evoke nongenomic effects on cellular function, which occur within minutes 4-7. Cardiovascular protective effects of GCs have been reported that could not be mediated by genomic mechanisms, because they occurred too fast and could not be blocked by a transcription inhibitor 8. Recently, we generated a comprehensive profile of such GC-induced rapid effects on signal transduction using activated human CD4+ T lymphocytes and a peptide array for kinome analysis 9. These results revealed marked early effects of GC treatment, in particular suppressed phosphorylation of Lck/Fyn kinase consensus substrates. Additional studies revealed impaired recruitment of Lck and Fyn to the T cell receptor (TCR) complex resulting in reduced Lck and Fyn enzymatic activities and impaired TCR signaling following short-term GC treatment. Although these studies identified Lck and Fyn as cellular targets for nongenomic GC activities, the underlying mechanism remained unexplained. Lck and Fyn are members of the Src family of non-receptor tyrosine kinases and play a key role in TCR signaling. TCR activation results in membrane translocation of Lck and Fyn in a Hsp90-dependent fashion 10-12. Lck predominantly associates with CD4 or CD8 cell surface receptors and Fyn binds to CD3 co-receptors, resulting in Lck and Fyn kinase activation. Once activated, Lck and Fyn phosphorylate immunoreceptor tyrosine-based activation motifs (ITAM) on the TCR, permitting downstream signal transduction and eventually T cell cytokine production, proliferation and differentiation. GRs and Src-like kinases share a requirement for Hsp90 for proper functioning; opening the possibility that GR action on Lck and Fyn activities is mediated by Hsp90. The current study reveals that GCs cause disruption of TCR-associated multiprotein complexes containing GR, Hsp90, Lck and Fyn, leading to reduced Lck/Fyn enzymatic activities and impaired TCR signaling.

93 Chapter 5

Materials and Methods Cell culture. CD4+ T cells were purified from human peripheral blood mononuclear cells (PBMCs) and maintained as described earlier 9. In short, PBMCs were isolated from whole blood of healthy volunteers by Ficoll-Isopaque density gradient centrifugation (Amersham Biosciences, Roosendaal, the Netherlands). The monocytes present in the PBMC pellet were removed by an adherence procedure followed by magnetic cell sorting. CD4+ T cells were cultured in Iscove’s Modified Dulbecco’s Medium/ IMDM (Gibco, Breda, the Netherlands), supplemented with 5% heat-inactivated fetal calf serum (FCS), 2mM L-glutamine and penicillin-streptomycin, in a humidified 5% CO2 environment.

Reagents and antibodies. Abs against Lck, Fyn, Hsp90α/β, GRα, CD3, CD4, PY20, b-Raf, MEK, Src, ZAP70, PKB, ERK, JNK, p38 MAPK and actin were obtained from Santa Cruz Biotechnology Inc. (Heidelberg, Germany). MEKSer217/221, SrcTyr416, PKBSer473, p38Thr180/Tyr182, ERK1/2Thr202/Tyr204 and JNKThr183/Tyr185 Abs were from cell signaling technology (Beverly, CA, USA). Anti-human CD3 (CD3-epsilon; mouse) was provided by the group of Prof. Dr. H. Spits (Academic Medical Center, Amsterdam, the Netherlands), and anti-CD28 Ab (mouse IgG1) was from Sanquin (Amsterdam, the Netherlands). Dexamethasone (DEX) and geldanamycin were obtained from Sigma-Aldrich (Zwijndrecht, the Netherlands). Hsp90α/βsiRNA, GRsiRNA, control siRNA, SAM68, and GR immunizing peptide were purchased from Santa Cruz. Kinase buffer and lysis buffer (supplemented with 1µg/ml NaF,

1µg/ml leupeptin, 1µg/ml aprotinin, 10mM Na3VO4, 1mM pefabloc) were from Cell Signaling Technology.

Cell transfection. Cells were transfected by electroporation, according to routine procedures (Amaxa, Cologne, Germany). Each nucleofection sample contained 5.106 cells, 2 µg highly purified siRNA and 100 µl human T cell Nucleofector. To determine transfection efficiency, GFP-tagged DNA (Amaxa) was used for electroporations revealing cells that were 40-50% GFP positive 16 hrs post-transfection, as assessed by FACS analysis (data not shown).

Immunoprecipitation, in vitro kinase assay and immunoblotting. Cells were incubated at 37°C in 6-well plates (5-10.106 cells per well) for 2 hrs followed by a 10 min pretreatment with

94 TCR-associated glucocorticoid receptors

1µM DEX dissolved in DMSO, or DMSO-supplemented media (control). Next, cells were activated for 15 min with anti-CD3 Abs (immobilized on plastic) and soluble anti-CD28 Abs (3 µg/ml). Cells were centrifuged (1250 rpm, 5 min), lysed in non-denaturing lysis buffer and subjected to immunoprecipitation. First, a pre-clearance step was performed by incubating the lysates with protein-A Sepharose for 2 hrs. After centrifugation (14.000 rpm, 5 min), supernatants were incubated overnight at 4 °C with the indicated Abs, followed by a 2-3 hr incubation with a protein-A Sepharose conjugated polyclonal antibody. After centrifugation (14.000 rpm, 5 min), immunoprecipitates were used for Western blot analysis. Alternatively, Lck/Fyn immunoprecipitates were dissolved in kinase buffer supplemented with 200 µM ATP and 2 µg/ml SAM68 and in vitro kinase reactions were performed at 30 ºC for 30 min. Following centrifugation (14.000 rpm, 5 min), the pellets were dissolved in sample buffer (62.5 mM Tris-HCl (pH 6.8 at 25 °C), 2% (w/v) SDS, 10% glycerol, 50 mM dithiothreitol, 0.01% (w/v) bromphenol blue), heated for 5 min at 95°C, loaded on SDS-PAGE and transferred to a PVDF membrane (Immobilon-P; Millipore, Amsterdam, the Netherlands). The membranes were blocked with 5% BSA in TBS/T (0.05M Tris, 150mM NaCl, 0.05% Tween-20). Primary Abs and secondary HRP-conjugated Abs were diluted in 5% BSA TBS/T, and proteins were visualized using the Lumi-LightPLUS substrate (Roche, Woerden, the Netherlands).

Fluorescent double-stainings. After in vitro stimulations, cells were centrifuged for 5 min at 1750 rpm and pellets were resuspended in PBS. SuperFrost® plus microscope slides were pretreated with 0.01% poly-L-lysine for 10 min, cell suspensions were subsequently incubated on the slides for 30 min (final concentration: 2.105 cells/slide) and 3.7% paraformaldehyde was added for 30-60 min. Sections were washed in PBS-Triton X100 0.1% (PBS-T), followed by a 1 hr blocking step with 10% FCS in PBS-T. After overnight incubation with an anti-GR Ab at 4 °C, Abs against Lck or Fyn were added for 2-3 hrs. Slides were washed and incubated for 60 min with FITC and TRITC-conjugated secondary antibodies (Heverlee, Belgium) diluted in 3% BSA/PBS-T. Alternatively, cells were incubated overnight with an anti-Hsp90 Ab. Sections were mounted in Vectashield (Vector laboratories Inc., Burlingame, CA, USA) supplemented with Dapi, and sealed with cover slips.

95 Chapter 5

Confocal fluorescence microscopy. After immunostaining, cells were imaged with a Leica SP2 AOBS confocal microscope. Excitation/detection of FITC and TRITC was done with 488nm/500-550nm and 561nm/580-640nm respectively. Dapi was imaged with 405nm excitation and detected with 410-460nm. To avoid cross talk, detections were done in a sequentional scan mode. A HCX PLAN APO CS 63×/1.20 water immersion objective was used. All images were adapted to the full dynamic range of the system (8 bit). The pinhole size was set at 1 Airy to ensure an optimal z-resolution (approximately 600nm). The images were scanned with a pixel size of 75nm. The presence of FITC and TRITC was ascertained by spectral imaging of the samples. To be able to compare the amount of fluorescence in the illustrations, all cells were imaged with identical instrument settings (i.e. laserpower, pmt settings, pinhole diameter and pixelsize).

Results DEX inhibits GR-Lck-Hsp90 and GR-Fyn-Hsp90 interactions We examined the action of short-term treatment with the synthetic fluorinated GC dexamethasone (DEX) on Hsp90-Lck and Hsp90-Fyn physical interactions. Cell lysates were subjected to Lck or Fyn immunoprecipitation and analyzed on Western blot for the presence of Hsp90. These experiments revealed Hsp90-Lck and Hsp90-Fyn interactions (Fig. 1A), confirming previous studies 10,13,14. DEX treatment (30 min) resulted in disappearance of Hsp90-Lck and Hsp90-Fyn complexes, suggesting that DEX rapidly inhibits Hsp90-Lck and Hsp90-Fyn associations. Next, we investigated whether these DEX-sensitive Hsp90-Lck and Hsp90-Fyn complexes contained GR. Cells were pretreated for 10 min with DEX or vehicle, followed by 15 min activation using anti-CD3 and anti-CD28 Abs. Lck and Fyn immunoprecipitates were immunoblotted for the presence of GR, showing GR-Lck and GR- Fyn complexes in activated cells, and these associations disappeared after short-term DEX treatment (Fig. 1B). These observations indicate that Lck, Fyn, Hsp90 and GR are part of a multiprotein complex whose integrity is sensitive to DEX.

96 TCR-associated glucocorticoid receptors

Figure 1. DEX rapidly inhibits GR-Hsp90-Lck and GR-Hsp90-Fyn associations. (A) Cells were incubated for 10 or 30 min with 1µM DEX or solvent supplemented media (-). Lck and Fyn immunoprecipitates were analyzed on Western blot for the presence of Hsp90, and total Lck/Fyn levels were analyzed to test for equal protein loading. (B) Hsp90siRNA (non)transfected cells were pretreated for 10 min without (-) or with (+) DEX and activated for 15 min with anti-CD3 and anti-CD28 Abs. Lysates were subjected to Lck or Fyn immunoprecipitation followed by Western blot analysis for the presence of GR or Hsp90. Equal loading was verified using anti-Lck and anti-Fyn Abs. (C) The Hsp90siRNA-transfection procedure was evaluated by immunoblotting lysates prepared from quiescent (- anti-CD3/CD28 Abs) or activated (+ anti-CD3/CD28 Abs) cells. Hsp90 protein levels were determined by the ratio of Hsp90 signal intensity compared to actin; the data are expressed as means ± SEM. Experiments were performed three times and reproducible results were obtained. DEX, dexamethasone; GR, glucocorticoid receptor; IP, immunoprecipitation; IB, immunoblotting.

97 Chapter 5

Lck-CD4 and Fyn-CD3 associations are GR and Hsp90-dependent Hsp90-dependent cellular distribution of Lck and Fyn is essential for efficient TCR signaling. To determine the role of Hsp90 for Lck-CD4 and Fyn-CD3 associations, cells were transfected with Hsp90siRNA, followed by 10 min DEX pretreatment and 15 min activation. Lck-CD4 and Fyn-CD3 interactions were detected in CD4 and CD3 immunoprecipitates prepared from activated cells (Fig. 2A). These interactions were disrupted due to treatment with Hsp90siRNA, irrespective of DEX stimulation. Furthermore, GR-CD4 and GR-CD3 associations were seen in activated cells and treatment with DEX or Hsp90siRNA resulted in reduced GR-CD4 and GR-CD3 binding (Fig. 2A). Treatment with Hsp90siRNA also disrupted Lck-GR and Fyn-GR interactions (Fig. 1B). Cell lysates were subjected to GR immunoprecipitation confirming DEX and Hsp90 sensitive GR-Lck and GR-CD4 interactions (not shown). The transfection procedure was evaluated by immunoblotting cellular extracts prepared from cells transfected with or without Hsp90siRNA (Fig. 1C). In addition, cells were pretreated for 30 min with the Hsp90 inhibitor geldanamycin, followed by 10 min DEX treatment and 15 min stimulation. Lck and Fyn immunoprecipitates were immunoblotted for the presence of GR, confirming reduced Lck-GR and Fyn-GR bindings due to Hsp90 inhibition (Fig. 2B). Overall, these observations indicate that TCR-associated multiprotein complexes are formed in activated cells (containing GR, Hsp90, Lck and Fyn), that depend on the presence of Hsp90, and are disrupted upon short-term DEX incubation. Complex dissolution does not lead to Lck or Fyn degradation, as total Lck/Fyn protein levels were not affected by Hsp90siRNA or DEX (not shown). Next, we investigated whether the presence of GR is required for efficient TCR signaling. GRsiRNA (non)transfected cells were pretreated for 10 min with DEX followed by 15 min stimulation. Western blotting revealed impaired TCR signaling in GRsiRNA transfected cells, as indicated by suppressed phosphorylation of several downstream TCR signaling intermediates (Fig. 2C). These findings provide direct evidence that the GR plays an important role in mediating efficient TCR signaling. As GRsiRNA may induce non-specific effects, cells were transfected with control siRNA demonstrating that the observed effects on TCR signaling are not a result of non-specific GRsiRNA action (Fig. 2D).

98 TCR-associated glucocorticoid receptors

Figure 2. GR and Hsp90-dependent Lck-CD4 and Fyn-CD3 complex formations. (A) Hsp90siRNA transfected (+) and nontransfected (-) cells were pretreated for 10 min in the presence (+) or absence (-) of DEX (1µM) and incubated with (+) or without (-) anti-CD3/anti-CD28 Abs for 15 min. CD3 and CD4 immunoprecipitates were immunoblotted using the indicated Abs. To ensure specificity in this experiment, the presence of Fyn and Lck was analyzed in CD4 and CD3 immunoprecipitates respectively, demonstrating that specific complexes had been immunoprecipitated containing either Fyn and CD3 or Lck and CD4. (B) Cells were pretreated for 30 min with geldanamycin (5µM), followed by 10 min DEX treatment and 15 min activation. Lck and Fyn immunoprecipitates were analyzed on immunoblot for the presence of GR. Anti-Lck and anti-Fyn Abs were used to evaluate for equal loading. (C) GRsiRNA (non)transfected cells were pretreated for 10 min with DEX, followed by 15 min activation. Lysates were immunoblotted using phosphospecific Abs against TCR signaling intermediates, and appropriate Abs were used to test for equal loading. (D) Cells were transfected with GRsiRNA or with unrelated siRNA (negative control). After overnight incubation, cells were stimulated (anti- CD3/anti-CD28 Abs) for 15 min. Cell lysates were immunoblotted employing Abs against GR, phosphorylated- ERK and phosphorylated-Src. An Ab against actin was used to evaluate for equal loading. Experiments were performed three times and representative results are shown. DEX, dexamethasone; GR, glucocorticoid receptor; IP, immunoprecipitation; IB, immunoblotting; P, phosphorylated; Cntrl, control siRNA.

99 Chapter 5

Given that Hsp90 associates readily with unfolded proteins (such as ZAP70) and can therefore be non-specifically immunoprecipitated, we next studied whether DEX interferes with ZAP70-Hsp90 binding. These data indicated that the observed effects of DEX are specific for Src-family kinases, as DEX did not inhibit Hsp90-Zap70 interactions in activated T cells (Fig. 3A). Further confidence in these results was supported by the lack of effect of either DEX or Hsp90siRNA on the integrity and functional activity of b-Raf-MEK protein complexes (Fig. 3B). Hence, DEX treatment does not interfere with protein complex formation per se, but specifically targets Lck-Hsp90 and Fyn-Hsp90 formations.

Figure 3. DEX does not interfere with Hsp90-ZAP70 or MEK-Raf associations. (A) Hsp90siRNA (non)transfected cells were pretreated with DEX (1µM; 10 min) followed by 15 min stimulation (anti- CD3/CD28 Abs). Cell lysates were subjected to immunoprecipitation using anti-ZAP70 Ab and immunoprecipitates were subsequently immunoblotted employing anti-Hsp90 and anti-ZAP70 Abs (negative control). (B) To further examine the specificity of the observed effects of DEX and Hsp90siRNA on Lck and Fyn kinases, we investigated whether both treatments interfered with a different kinase cascade (i.e. b-Raf- MEK), serving as a control. Cell lysates, prepared as described above, were subjected to immunoprecipitation using an anti-MEK Ab, followed by immunoblotting for b-Raf and MEK. In addition, supernatants were immunoblotted for phosphorylated and total MEK, a downstream substrate of Raf, as well as for actin. DEX, dexamethasone; IB, immunoblot; IP, immunoprecipitation; P, phosphorylated.

Decreased Lck and Fyn kinase activities due to Hsp90siRNA treatment To investigate the functional consequences of dissolution of these multiprotein complexes on TCR-induced activation of Lck and Fyn, we examined whether Hsp90siRNA affects Lck or Fyn activities, using Lck and Fyn immunoprecipitates and in vitro phosphorylation of SAM68 (Src-Associated in Mitosis). TCR stimulation resulted in enhanced SAM68 phosphorylation

100 TCR-associated glucocorticoid receptors

compared to control cells, confirming TCR-dependent activation of Lck and Fyn in these experiments (Fig. 4). Similar to DEX treatment, Lck and Fyn immunoprecipitated from Hsp90siRNA transfected cells did not have the capacity to phosphorylate SAM68, indicating that Hsp90siRNA and DEX impair Lck and Fyn enzymatic activities. These observations suggest that Lck/Fyn-mediated TCR signaling depends on the presence of Hsp90 and GR.

Figure 4. Reduced Lck/Fyn activities due to Hsp90siRNA or DEX. Hsp90siRNA transfected (+) and non- transfected (-) cells were pretreated with (+) or without (-) 1µM DEX for 10 min and activated for 15 min using anti- CD3/anti-CD28 Abs or left unstimulated. Cell lysates were subjected to Lck or Fyn immunoprecipitation followed by in vitro kinase assay using SAM68 as a substrate. Phosphorylated-SAM68 was immuno- blotted with PY20, and equal loading was verified using anti-Lck and anti-Fyn Abs. DEX, dexamethasone; GR, glucocorticoid receptor; SAM68, Src- Associated in Mitosis; IP, immuno- precipitation; IB, immunoblotting; P, phosphorylated.

Reduced GR-Lck and GR-Fyn colocalization due to DEX or Hsp90siRNA Cellular localizations of GR, Lck and Fyn were studied using confocal fluorescence microscopy. Double-stainings for GR and Lck or GR and Fyn demonstrated cytoplasmic staining of Lck, Fyn and GR in quiescent cells (Fig. 5). Lck, Fyn and GR staining was mainly detected in the cell periphery upon TCR stimulation (Fig. 5; dotted lines). GR-Lck and GR- Fyn cytoplasmic colocalization (indicated by yellow staining) was seen in quiescent cells, which was increased upon TCR stimulation (Fig. 5; solid lines). Reduced GR-Lck and GR- Fyn colocalization was observed in activated cells treated with DEX or Hsp90siRNA, as indicated by diminished yellow staining. Hence, treatment with DEX or Hsp90siRNA reduces GR-Lck and GR-Fyn colocalization, supporting the immunoprecipitation data that DEX and Hsp90siRNA disrupt GR-Lck and GR-Fyn interactions. Due to the small cytoplasm volume compared to the nuclear volume, no conclusions can be drawn regarding the precise cytoplasmic or membrane localization of Lck or Fyn in the different conditions. Specificity controls for the immunofluorescent stainings are depicted in Figure 6.

101 Chapter 5

Figure 5. DEX and Hsp90siRNA interfere with GR-Lck and GR-Fyn colocalization. Hsp90siRNA (non)transfected cells were pretreated with or without DEX (1µM; 10 min) and activated with anti-CD3 and anti- CD28 Abs (15 min). Cells were studied by confocal fluorescence microscopy and immunofluorescent double- stainings of GR (FITC-labeled, green) together with Lck or Fyn (TRITC-labeled, red) are shown. Cytoplasmic colocalization between GR and Lck or GR and Fyn is indicated by yellow staining (solid lines). The nucleus was visualized with Dapi (not shown). Scanned area for all images: 9 µM x 9 µM. At least 20 cells of each condition were scanned and representative optical cross sections through T cells are shown. Three independent experiments were performed and reproducible results were obtained. DEX, dexamethasone; GR, glucocorticoid receptor.

102 TCR-associated glucocorticoid receptors

Figure 6. Immunofluorescent specificity controls. Immunofluorescent staining for Hsp90 (FITC-labeled, green) was performed to compare nontransfected cells (A) with Hsp90siRNA transfected cells (B). Staining with a goat secondary Ab in the absence of the primary Hsp90 Ab (negative control) (C). Cells stained with anti-GR Ab (FITC-labeled, green) (D). The specificity of the anti-GR Ab was confirmed using GR immunizing peptide: after preincubation with the immunizing peptide (i.e. anti-GR Ab versus immunizing peptide: 1:50), staining for GR remained negative (E). Cells were stained with the secondary Ab (rabbit) in the absence of anti-GR Ab (negative control) (F). Immunofluorescent stainings using anti-Fyn (G) and anti-Lck (H) Abs (TRITC-labeled, red) are shown. Appropriate controls for Lck and Fyn included stainings of cells using the mouse secondary Ab (anti-TRITC) in the absence of primary anti-Lck or anti-Fyn Abs (I). The nucleus was visualized with Dapi (blue). Panels of optical cross sections through T cells are shown; scanned area for all images: 9 µM x 9 µM. GR, glucocorticoid receptor.

Discussion GCs mediate well-defined genomic effects through GR-dependent transcriptional changes. Clinical and experimental evidence for rapid nongenomic GC action has accumulated, but the underlying molecular mechanisms remain unexplained. Estrogens can induce rapid signaling through transmembrane estrogen receptors, providing evidence for nongenomic effects of steroids on cellular physiology 15. Similarly, membrane-bound GRs have been identified in lymphocytes 16,17, and in human PBMCs 18, but their functional relevance remains unclear. It is known that at high concentrations, GCs increase GR saturation in a dose-dependent manner which intensifies the therapeutically relevant genomic GC activities 19. Although saturation of cytosolic GRs is almost complete with 100 mg prednisone equivalent a day, the usage of (much) higher dosages (100–1000 mg) is common and successful in daily clinical practice 1,

103 Chapter 5

possibly as a consequence of additional nongenomic effects 20. Nongenomic GC action is currently thought to be mediated through cytosolic or membrane-bound GRs or via unspecific physicochemical interactions with membranes. Nongenomic GC-induced inhibition of Lck/Fyn-mediated TCR signaling has been reported 9, but the responsible mechanism remained undefined. A relationship between GC stimulation and Lck/Fyn kinases is supported by previous work, showing that prolonged DEX stimulation (enabling GC-induced genomic effects) disturbed Lck and Fyn submembrane localization in murine T cells 21. It is possible that Hsp90-dependent regulation of Lck and Fyn contributes to these described effects, but further studies are needed to clarify the role of GR and Hsp90 on regulating TCR signaling. Based on the present work, we propose a model arguing that in the absence of ligand the GR sustains a TCR-associated complex containing Hsp90, Lck and Fyn (Fig. 7). Upon GR-ligand binding, this membrane-bound multiprotein complex is dissociated, leading to a cellular redistribution of Lck and Fyn and an inability of Lck/Fyn to participate in TCR signaling. Furthermore, our data reveal an essential role for Hsp90 and GR in the formation of TCR- associated protein complexes and the regulation of efficient TCR signaling. Overall, these results provide novel insight into the nongenomic mode of GC action in T cells. Selective inhibition of proximal TCR signaling through Lck/Fyn may represent novel therapeutic opportunities for the development of specific immunosuppressive agents.

104 TCR-associated glucocorticoid receptors

Figure 7. Model for nongenomic GC-induced immunosuppression in T cells. Left part: Nongenomic GR pathway. 1, 2) Upon TCR ligation, Lck and Fyn are recruited to the TCR complex resulting in Lck and Fyn kinase activation and initiation of downstream TCR signaling. The present study reveals TCR-linked GR multiprotein complexes containing Hsp90, Lck and Fyn in activated T cells. 3, 4) Upon GR-ligand binding this multiprotein complex rapidly dissociates. 5) Lck and Fyn are released from the TCR complex leading to impaired TCR signaling as a consequence of a cellular redistribution and abrogated activation of Lck and Fyn. Right part: Classical GR pathway: the genomic mode of GC action is well-defined and involves GR-mediated transcriptional effects. GC, glucocorticoid; GR, glucocorticoid receptor.

Acknowledgments We thank Meike Scheffer for technical support and Gijs van den Brink for critically reading this manuscript.

105 Chapter 5

Reference List

(1) Buttgereit F, Straub RH, Wehling M, Burmester GR. Glucocorticoids in the treatment of rheumatic diseases: an update on the mechanisms of action. Arthritis Rheum. 2004;50:3408-3417. (2) Franchimont D. Overview of the actions of glucocorticoids on the immune response: a good model to characterize new pathways of immunosuppression for new treatment strategies. Ann N Y Acad Sci. 2004;1024:124-137. (3) Pratt WB, Galigniana MD, Morishima Y, Murphy PJ. Role of molecular chaperones in steroid receptor action. Essays Biochem. 2004;40:41-58. (4) Baus E, Andris F, Dubois PM, Urbain J, Leo O. Dexamethasone inhibits the early steps of antigen receptor signaling in activated T lymphocytes. J Immunol. 1996;156:4555-4561. (5) Cato AC, Nestl A, Mink S. Rapid actions of steroid receptors in cellular signaling pathways. Sci STKE. 2002;2002:RE9. (6) Croxtall JD, Choudhury Q, Flower RJ. Glucocorticoids act within minutes to inhibit recruitment of signalling factors to activated EGF receptors through a receptor-dependent, transcription-independent mechanism. Br J Pharmacol. 2000;130:289-298. (7) Rhen T, Cidlowski JA. Antiinflammatory action of glucocorticoids--new mechanisms for old drugs. N Engl J Med. 2005;353:1711-1723. (8) Hafezi-Moghadam A, Simoncini T, Yang Z et al. Acute cardiovascular protective effects of corticosteroids are mediated by non-transcriptional activation of endothelial nitric oxide synthase. Nat Med. 2002;8:473-479. (9) Lowenberg M, Tuynman J, Bilderbeek J et al. Rapid immunosuppressive effects of glucocorticoids mediated through Lck and Fyn. Blood. 2005;106:1703-1710. (10) Bijlmakers MJ, Marsh M. Hsp90 is essential for the synthesis and subsequent membrane association, but not the maintenance, of the Src-kinase p56(lck). Mol Biol Cell. 2000;11:1585-1595. (11) Palacios EH, Weiss A. Function of the Src-family kinases, Lck and Fyn, in T-cell development and activation. Oncogene. 2004;23:7990-8000. (12) Zamoyska R, Basson A, Filby A et al. The influence of the src-family kinases, Lck and Fyn, on T cell differentiation, survival and activation. Immunol Rev. 2003;191:107-118. (13) Hartson SD, Barrett DJ, Burn P, Matts RL. Hsp90-mediated folding of the lymphoid cell kinase p56lck. Biochemistry. 1996;35:13451-13459. (14) Yun BG, Matts RL. Differential effects of Hsp90 inhibition on protein kinases regulating signal transduction pathways required for myoblast differentiation. Exp Cell Res. 2005;307:212-223. (15) Revankar CM, Cimino DF, Sklar LA, Arterburn JB, Prossnitz ER. A transmembrane intracellular estrogen receptor mediates rapid cell signaling. Science. 2005;307:1625-1630. (16) Gametchu B, Chen F, Sackey F, Powell C, Watson CS. Plasma membrane-resident glucocorticoid receptors in rodent lymphoma and human leukemia models. Steroids. 1999;64:107-119. (17) Gametchu B, Watson CS. Correlation of membrane glucocorticoid receptor levels with glucocorticoid- induced apoptotic competence using mutant leukemic and lymphoma cells lines. J Cell Biochem. 2002;87:133-146. (18) Bartholome B, Spies CM, Gaber T et al. Membrane glucocorticoid receptors (mGCR) are expressed in normal human peripheral blood mononuclear cells and up-regulated after in vitro stimulation and in patients with rheumatoid arthritis. FASEB J. 2004;18:70-80. (19) Buttgereit F, da Silva JA, Boers M et al. Standardised nomenclature for glucocorticoid dosages and glucocorticoid treatment regimens: current questions and tentative answers in rheumatology. Ann Rheum Dis. 2002;61:718-722. (20) Buttgereit F, Wehling M, Burmester GR. A new hypothesis of modular glucocorticoid actions: steroid treatment of rheumatic diseases revisited. Arthritis Rheum. 1998;41:761-767. (21) Van Laethem F, Baus E, Smyth LA et al. Glucocorticoids attenuate T cell receptor signaling. J Exp Med. 2001;193:803-814.

106

Chapter 6

Kinome analysis reveals nongenomic glucocorticoid receptor-dependent inhibition of insulin signaling

Mark Löwenberg1,2, Jurriaan Tuynman1, Meike Scheffer1, Auke Verhaar1, Louis Vermeulen1, Sander van Deventer1, Daniel Hommes2, Maikel Peppelenbosch3

1 Department for Experimental Internal Medicine, Academic Medical Center, Amsterdam, The Netherlands 2 Department of Gastroenterology and Hepatology, Academic Medical Center, Amsterdam, The Netherlands 3 Department of Cell Biology, University of Groningen, Groningen, The Netherlands

Endocrinology. 2006; 147(7):3555-62

107 Chapter 6

Abstract Glucocorticoids (GCs) are powerful immunosuppressive agents that control genomic effects through GC receptor-dependent transcriptional changes. A common complication of GC therapy is insulin resistance, but the underlying molecular mechanism remains obscure. Evidence is increasing for rapid genomic-independent GC action on cellular physiology. We here generated a comprehensive description of nongenomic GC effects on insulin signaling using a peptide array containing 1176 different kinase consensus substrates. Reduced kinase activities of the insulin receptor and several downstream insulin receptor signaling intermediates (i.e. p70S6k, AMPK, GSK-3 and Fyn) were detected in adipocytes and T lymphocytes due to short-term treatment with dexamethasone, a synthetic fluorinated GC. Western blot analysis confirmed suppressed phosphorylation of the insulin receptor and a series of downstream insulin receptor targets (i.e. IRS-1, p70S6k, PKB, PDK, Fyn and GSK- 3) following dexamethasone treatment. Dexamethasone inhibited insulin signaling through a GC receptor-dependent (RU486-sensitive) and transcription-independent (actinomycin D- insensitive) mechanism. We here postulate a molecular mechanism for GC-induced insulin resistance based on nongenomic GC receptor-dependent inhibition of insulin signaling.

108 Glucocorticoids inhibit insulin signaling

Introduction Glucocorticoids (GCs) are widely used therapeutically for their immunosuppressive and anti- inflammatory properties. Among the most common and serious clinical complications of GC therapy is the induction of insulin resistance 1-3. However, the molecular mechanism responsible for GC-induced insulin resistance remains to be defined and is of obvious clinical relevance. The insulin receptor signaling pathway regulates growth and metabolic responses in many cell types 4. Insulin receptor-mediated signaling is initiated by insulin binding to the α-subunit of the cell surface insulin receptor, which leads to auto-phosphorylation of the β- subunit and activation of insulin receptor tyrosine kinase activity. Downstream signaling targets are subsequently activated, including insulin receptor substrate (IRS), p70S6 kinase (p70S6k), AMP-activated protein kinase (AMPK), phosphatidylinositol 3 kinase (PI3K), protein kinase B (PKB) and glycogen-synthase kinase-3 (GSK-3) 5-10. Although it has been reported that prolonged treatment with the synthetic fluorinated GC dexamethasone (DEX) reduced IRS-1, PI3K and PKB cellular contents in adipocytes 11,12, the mechanism underlying GC-induced insulin resistance remains obscure at best. Because the kinome regulates virtually all major cellular metabolic pathways, it is reasonable to assume that a full description of the kinome should enable the identification of rapid DEX- induced effects on cellular metabolism possibly involved in mediating insulin resistance. The aim of this study was to investigate early effects of DEX on the adipocyte and T lymphocyte kinome. Here, we have generated a comprehensive description of DEX-induced effects on the kinome signal transduction, employing a peptide-array containing 1176 spatially addressed mammalian kinase consensus substrates 13-15. These results were validated with conventional techniques revealing nongenomic GC receptor-dependent inhibition of insulin signaling.

Materials and Methods Cell culture. 3T3-L1 cells (American Type Culture Collection; CCL 92.1) were provided by F. Falix (Liver Centre, Academic Medical Center, the Netherlands) and maintained in Dulbecco’s Modified Eagle Media/DMEM (Life Technologies, Rockville, MD, USA), supplemented with 10% FCS (Gemini Bio-Products, Woodland, CA, USA), 2 mM L- glutamine (Gibco, Breda, the Netherlands) and penicillin/streptomycin, in a humidified 5%

CO2 environment at 37 °C. Human peripheral blood mononuclear cells (PBMCs) were

109 Chapter 6

isolated from whole blood of healthy volunteers by Ficoll-Isopaque density gradient centrifugation (Amersham Biosciences, Roosendaal, the Netherlands). Monocytes present in the PBMC pellet were removed by an adherence procedure. Cells were plated out in 6 well plates (CellStar, Greiner Bio-One) at a final concentration of 5.106 cells per well for 2 hrs at 37 °C and soluble cells were harvested for subsequent magnetic cell sorting, described below. CD4+ T cells were cultured in Iscove’s Modified Dulbecco’s Medium/IMDM (Gibco, Breda, the Netherlands), supplemented with 10% FCS, 2 mM L-glutamine and penicillin/streptomycin.

CD4+ purification. CD4+ T cells were purified by negative selection using the MACS system. In short, non-CD4+ cells were indirectly magnetically labeled with a cocktail of biotin- conjugated monoclonal antibodies (mAbs) bound to MicroBeads as secondary labeling agent (Miltenyi Biotec Inc., Auburn, CA, USA). The magnetically labeled non-CD4+ T cells were depleted by retaining them on a MACS Column in the magnetic field of the autoMACS Separator (Miltenyi Biotec) and the unlabeled CD4+ T cells were collected. The sample purity was assessed by fluorescence-activated cell sorter (FACS) (Becton Dickinson, San Jose, CA, USA) with PE-conjugated CD3 and FITC-conjugated CD4 monoclonal antibodies (purity >95% CD3+CD4+; not shown), according to routine procedures.

Reagents and antibodies. Phospho-specific antibodies (Abs) against insulin receptorTyr1131/1146, IRS-1Ser636/639, GSK-3Ser21/9, PKBSer473, p70S6kThr389, PDK-1Ser241, GC receptorSer211, ERK1/2Thr202/Tyr204, STAT3Tyr705, JNK1/2/3Thr183//Tyr185, IKKα/βSer176/180, as well as Abs against total IRS-1, GSK-3α/β, PDK, p70S6k and PKB were purchased from Cell Signaling Technology (Beverly, CA, USA). Abs specific for the insulin receptor, GC receptor, Fyn, ERK1, STAT3, JNK, IKK, actin and phosphorylated-FynThr12 were obtained from Santa Cruz Biotechnology Inc. (Heidelberg, Germany). Abs against phosphorylated-JAK2Tyr1007/1008 and total JAK2 were from Abcam (Cambridge, UK). HRP-conjugated Goat-anti-Rabbit, Goat- anti-Mouse and Rabbit-anti-Goat were purchased from DakoCytomation (Heverlee, Belgium). Anti-human CD3 (CD3-epsilon; mouse) was provided by the group of Prof. Dr. H. Spits (Academic Medical Center, Amsterdam, the Netherlands), and anti-CD28 Ab (mouse IgG1) was from Sanquin (Amsterdam, the Netherlands). PE-conjugated CD3 and FITC-

110 Glucocorticoids inhibit insulin signaling

conjugated CD4 Abs were from BD Biosciences (Alphen a/d Rijn, the Netherlands). Dexamethasone (DEX), MTT and actinomycin D were obtained from Sigma-Aldrich (Zwijndrecht, the Netherlands). RU486 (Mifepristone) was from LKT laboratories Inc. (Minnesota, MN, USA). γ-33P-ATP was purchased from Amersham Biosciences. Insulin (Actrapid, 100U/ml) was from Novo Nordisk A/S (Copenhagen, Denmark). Recombinant interleukin-6 (IL-6) and TNFα was obtained from R&D systems (Minneapolis, MN, USA), and recombinant insulin growth factor-1 (IGF-1) was from Invitrogen (Breda, the Netherlands). Lysis buffer and kinase buffer were purchased from Cell Signaling Technology. Lysis buffer was supplemented with protease and phosphatase inhibitors, including 1 µg/ml NaF, 1 µg/ml leupeptin, 1 µg/ml aprotinin, 10 mM Na3VO4 and 1 mM pefabloc, obtained from Merck BV (Amsterdam, the Netherlands).

DEX in vitro stimulations and preparation of cell lysates. 3T3 adipocytes were cultured at 37 °C in 6 well plates (concentration: 5.106 cells per well) and serum starved overnight. Next, cells were pretreated for 1 hr with insulin (100 IE/ml) or IGF-1 (20 ng/ml), followed by DEX treatment for 10, 20 or 30 min (1 µM) dissolved in Dimethyl Sulfoxide (DMSO) or DMSO- supplemented media (control). To monitor the effect of prolonged DEX treatment on insulin signaling, insulin pretreated adipocytes (100 IE/ml, 1 hr) were incubated in the presence of 1 µM DEX up to 4 hrs. IL-6 stimulated adipocytes served as a control; e.g. cells were pretreated for 30 min with IL-6 (100 ng/ml) followed by 10, 20 or 30 min DEX incubation (1 µM). Furthermore, adipocytes were pretreated for 15 min with 0.1 µM actinomycin D 16-18, and insulin (100 IE/ml) was added for 45 min before addition of 1 µM DEX for 30 min. PBMC- derived CD4+ T cells were incubated at 37 °C in 6-well plates (final concentration: 5-10.106 cells per well) and pretreated for 10 min with DEX (1 µM). Cells were then stimulated for 15 min with anti-CD3 Abs (immobilized on plastic) and soluble anti-CD28 Abs (3 µg/ml). In vitro stimulations were terminated by an ice-cold phosphate-buffered saline (PBS) wash. Cell extracts were prepared by scraping cells into 100 µl lysis buffer supplemented with the indicated protease and phosphatase inhibitors. Cell lysates were used for pepchip array analysis or Western blotting. Support that 1 µM DEX did not exert a toxic effect in adipocytes and T cells was provided by tryptan blue exclusion tests verifying that cells were viable after culture (not shown).

111 Chapter 6

Cell transfection and reporter assay. 3T3 adipocytes were transfected using Effectene reagent (Qiagen, Hilden, Germany) and a plasmid containing a NFkB binding site enhancing expression of a firefly luciferase gene (0.5 µg/well) (Clontech laboratories Inc., Mountain View, CA, USA). Cells were incubated at 37 °C for 24 hrs, fresh medium was added to the wells and cells were pretreated with actinomycin D for 45 min (0.1 µM), followed by TNFα stimulation (100 ng/ml) for 45 min. Transfection efficiency was evaluated by cotransfecting cells with a plasmid encoding a Renilla luciferase gene under the cytomegalovirus promoter (Promega, Leiden, the Netherlands), after which cell lysates were recovered for luciferase assay. The firefly and Renilla luciferase activities were assayed on a Lumat Berthold LB 9501 Luminometer. Each firefly luciferase value was corrected for its cotransfected Renilla luciferase value to correct for transfection efficiency or dilution effects. Values are expressed as percent increase over non-TNF stimulated cells.

MTT viability assay. Cells were incubated at 37 °C for 3 hrs with or without 0.1 µM actinomycin D, and cell viability was assessed with MTT (3-[4,5-dimethylthiazol-2-yl]-2,5- diphenyltetrazolium bromide) colorimetric assay. In short, 0.5 mg/ml MTT was added to the media for 1-2 hours at 37 °C, and subsequently isopropanol/0.04N HCl was added. The OD560 was determined using an ELISA-plate reader (Bio-Rad Benchmark, Hercules, CA, USA). Actinomycin D treatment did not affect cell viability compared to control cells; 0.1 µM actinomycin D treated cells: +1.7% +/- 3.7% SEM, and control cells: 0% +/- 3.8% SEM

Peptide array analysis. The protocol of the kinome array is described in detail on the website: http://www.pepscan.nl/pdf/Manual%20PepChip%20Kinase%200203.pdf. Adipocyte and T cell extracts were corrected for protein concentrations using Bradford analysis (Biorad, Veenendaal, the Netherlands) and used for subsequent kinome array analysis. 10µl activation mix, containing 50% glycerol, 50 µM ATP, 60 mM MgCl2, 0.05% v/v Brij-35, 0.25 mg/ml BSA and 2000µ Ci/ml γ-33P-ATP was added to 60 µl cell lysate. The (second generation) peptide arrays, containing 1176 different kinase pseudo-substrates in duplo (Pepscan, Lelystad, the Netherlands) 15, were incubated with these mixtures (i.e. cell lysates together with activation mix) for 2 hrs in a humidified stove at 37 °C. The pepchips were washed (2 M

NaCl, 1% triton-x-100 and 0.1% tween in H2O), exposed to a phospho-imager plate for 24-72

112 Glucocorticoids inhibit insulin signaling

hrs (Fuji, Stamford, CO, USA) and the density of the spots was measured and analyzed with array software.

Data acquisition and statistical analysis of peptide arrays. Acquisition of the peptide arrays was performed with a phospho-imager (Fuji) and ArrayVision 6.0 software (Molecular Dynamics, Sunnyvale, CA, USA). After acquisition and quantification using median spot densities, the data were exported to a spreadsheet program (Microsoft Excel 2002, Microsoft, Redmond, WA, USA). Normalization was achieved by correction of the spot density for the individual background to diminish interarray variances. In addition, the variation between arrays and individual experiments was reduced by normalization to the 90% percentile of the intensity of each array. Inconsistent data (i.e. SD between the data points > 1.96 of the mean value) were excluded from further analysis. Spots were averaged and included for dissimilarity measurement in order to extract kinases of which activity was either significantly induced or reduced. Differential kinase activities in lysates from cells incubated in the presence or absence of DEX were determined by significant fold change ratios of the combined values of phosphorylated peptides resembling a substrate for kinase activity. Significance analysis was performed using a minimal modification for the algorithm originally developed for microarray analysis (http://www-stat.stanford.edu/~tibs/SAM/). The full list of peptides spotted on the peptide array can be found online: http://www.pepscan.nl /index5.htm

Western blot analysis. Total cell extracts were supplemented with SDS sample buffer (62.5 mM Tris-HCl (pH 6.8 at 25 °C), 2% (w/v) SDS, 10% glycerol, 50 mM dithiothreitol, 0.01% (w/v) bromphenol blue), sonificated and heated at 90 ºC for 5 min. Samples were loaded on 10% SDS-PAGE and transferred to a polyvinylidene difluoride (PVDF) membrane (Immobilon-P; Millipore, Amsterdam, the Netherlands). The membranes were blocked with 5% BSA in Tris-buffered saline supplemented with 0.1% tween-20 (TBST), and incubated overnight at 4 °C with the indicated primary Abs diluted in 5% BSA TBST. Next, the membranes were incubated for 1 hr at room temperature with horseradish peroxidase (HRP)- conjugated secondary Abs diluted in 5% BSA TBST. All Abs were used in accordance with the supplier’s protocol, and images were revealed with a Lumi-Imager (Roche Applied

113 Chapter 6

Science) using the chemoluminescence substrate LumilightPLUS (Roche, Woerden, the Netherlands). Blots were stripped with strip buffer (62.5mM Tris-HCl pH6.8, 100mM β- mercaptoethanol, 2% SDS) and reprobed with appropriate Abs for evaluation of equal protein loading.

Supplemental Material. The supplemental information discloses the results obtained with the pepchip experiments (available on The Endocrine Society’s Journals Online website at: http://endo.endojournals.org).

Results Short-term DEX treatment reduces kinase activities of insulin receptor signaling intermediates in adipocytes Since adipocytes are important in the pathogenesis of GC-induced insulin resistance, we studied rapid effects of DEX on the adipocyte kinome. 3T3 adipocytes were stimulated with insulin for 1 hr and treated for 30 min with DEX or solvent supplemented media (control). Kinome profiles of cell lysates revealed significant differences between DEX-treated and non- DEX-treated cells (Figure 1). In particular, it was found that short-term DEX treatment influenced insulin-induced kinase activities in adipocytes. We next analyzed the individual components of the kinome to examine the underlying determinants of the observed changes in kinase activities. Significant decreased phosphorylation of insulin receptor kinase consensus substrates was detected in DEX treated cells (Figure 2A), indicating reduced insulin receptor enzymatic activity. Furthermore, GSK-3 (Figure 2B), Fyn (Figure 2C), AMPK (Figure 2D) and p70S6k (Figure 2E) kinase consensus substrates showed suppressed phosphorylation in DEX treated adipocytes, pointing out suppressed kinase activities of several downstream insulin receptor targets. These results demonstrate rapid DEX-induced inhibition of kinase activities of the insulin receptor and downstream signaling intermediates, providing evidence for DEX-induced inhibition of insulin signaling.

114 Glucocorticoids inhibit insulin signaling

Figure 1. Rapid DEX-induced changes of kinome profiles in adipocytes and T cells. 3T3 adipocytes were stimulated with insulin (100 IE/ml) for 1 hr followed by a 30 min DEX treatment (1 µM). CD4+ T cells were pretreated for 10 min with or without DEX (1 µM) and activated for 15 min using anti-CD3 and anti-CD28 Abs. Cell lysates were incubated on peptide arrays in the presence of γ-33P-ATP. Differential kinase activities in lysates prepared from activated adipocytes (upper panel) or T cells (lower panel) incubated with or without DEX using median densities of the spots are shown. The dot plots represent the phosphorylation status of 1176 different kinase consensus substrates spotted in duplo on the pepchip arrays. Each spot reflects the amount of phosphorylation of a specific peptide substrate of which phosphorylation is significantly decreased (A), increased (B) or unchanged (C) due to DEX treatment. DEX, dexamethasone.

Rapid decreased catalytic activities of insulin receptor signaling intermediates in DEX- treated T cells In order to investigate whether the observed effects of DEX on adipocyte insulin receptor signaling were cell-type specific, we extended our studies to CD4+ T lymphocytes that are known to be important target cells of GC action. Cells were pretreated for 10 min with or without DEX and activated for 15 min with anti-CD3 and anti-CD28 Abs. Cell lysates were used for pepchip array analysis, and the phosphorylation patterns of DEX-treated and non- DEX-treated T cells were compared (Figure 1). In line with the adipocyte data, reduced phosphorylation of insulin receptor kinase consensus substrates was seen in activated T cells due to short-term DEX treatment (Figure 2A). Moreover, impaired phosphorylation of GSK-3 (Figure 2B), Fyn (Figure 2C) and AMPK (Figure 2D) consensus substrates was detected after DEX incubation. These findings indicate that DEX-induced inhibition of enzymatic activities of kinases involved in insulin signaling is not limited to adipocytes, but is of a more general significance.

115 Chapter 6

Figure 2. Reduced enzymatic activities of the insulin receptor and downstream signaling intermediates due to short-term DEX treatment. Adipocytes were stimulated with insulin (100 IE/ml; 1 hr) and incubated in the absence or presence of DEX (1 µM; 30 min). T cells were pretreated for 10 min with or without 1 µM DEX and activated with anti- CD3 and anti-CD28 Abs for 15 min. Cell lysates were incubated on pepchip arrays in the presence of γ-33P- ATP. The original scanned pictures were used for quantification and statistical analysis of the arrays. The graphs show the phosphorylation status of kinase consensus substrates which are specific for the insulin receptor (A), GSK-3 (B), Fyn (C), AMPK (D) and p70S6k (E). For each peptide substrate the annotation number is depicted and average changes of phosphory- lation are indicated by the dotted lines. Results for human kinase consensus substrates are shown, and peptide substrates obtained from other species are not indicated. DEX, dexamethasone; INSR, insulin receptor; IRS-1, insulin receptor substrate-1; AMPK, AMP-activated protein kinase; GSK-3, glycogen-synthase kinase-3.

DEX-induced inhibition of insulin receptor-mediated signal transduction To confirm that the observed changes in kinase activities indeed affect phosphorylation of insulin receptor signaling elements, the effect of short-term DEX treatment on insulin signaling was studied in 3T3 adipocytes and CD4+ T cells. Adipocytes were pre-incubated for

116 Glucocorticoids inhibit insulin signaling

1 hr with insulin and treated for 10, 20 or 30 min with DEX. T cells were pretreated for 10 min with DEX and stimulated for 15 min. Cellular extracts were immunoblotted employing phospho-specific Abs against the insulin receptor and several insulin receptor targets. Short- term DEX treatment resulted in reduced insulin receptor phosphorylation in both cell-types (Figure 3A,B). Effects of DEX on insulin signaling were investigated in more detail in adipocytes, demonstrating suppressed IRS-1, p70S6k, PKB, GSK-3, PDK and Fyn phosphorylation due to short-term DEX treatment (Figure 3A). These observations are in accordance with the pepchip results and indicate that DEX interferes with insulin signaling.

Figure 3. DEX inhibits insulin signaling. Adipocytes were stimulated with insulin (1hr, 100 IE/ml) and treated for 10, 20 or 30 min with DEX (1 µM). (A) Whole lysates were immunoblotted using phospho-specific Abs against the insulin receptor and downstream insulin receptor targets. Blots were analyzed for equal protein loading using appropriate Abs. (B) T cells were pretreated for 10 min with or without 1 µM DEX and activated for 15 min (anti-CD3/anti-CD28 Abs). Cell lysates were immunoblotted using Abs against phosphorylated and total insulin receptor. (C) Adipocyte lysates were immunoblotted for total and phosphorylated JNK and IKK. Experiments were performed three times and similar results were obtained. DEX, dexamethasone; INSR, insulin receptor; IRS-1, insulin receptor substrate-1; AMPK, AMP-activated protein kinase; GSK-3, glycogen-synthase kinase-3; PKB, protein kinase B; PDK, phosphoinositide-dependent protein kinase; JNK, c-Jun N-terminal kinase; IKK, IκB kinase; p, phosphorylated.

117 Chapter 6

DEX induces JNK phosphorylation, but does not interfere with IKK phosphorylation As previous reports have indicated that JNK and IKK, two serine kinases, are able to induce insulin resistance 16-21, the effect of short-term DEX treatment on JNK and IKK phosphorylation was studied. Adipocytes were pretreated for 1 hr with insulin followed by DEX treatment for 10, 20 or 30 min. Cell extracts were analyzed on Western blot using phospho-specific Abs against JNK and IKK (associated with activation). Increased phosphorylation of JNK was seen following short-term DEX treatment, whereas DEX did not interfere with IKK phosphorylation (Figure 3C). Altogether, these experiments demonstrate that DEX has opposing effects on insulin signaling, that is to say impaired phosphorylation of several insulin receptor signaling targets, but enhanced JNK phosphorylation.

DEX-induced inhibition of insulin signaling is independent of transcription In order to determine whether DEX-induced inhibition of insulin signaling is indeed dependent on a nongenomic (i.e. transcription-independent) mechanism, adipocytes were pretreated for 15 min with 0.1 µM actinomycin D, a transcription inhibitor. Next, insulin was added to the culture media for 45 min, followed by a 10, 20 or 30 min DEX stimulation. Cell lysates were immunoblotted using phospho-specific Abs against the insulin receptor and IRS- 1 (Figure 4A). Actinomycin D was found to be ineffective at preventing rapid DEX-induced inhibition of insulin receptor and IRS-1 phosphorylation. These results indicate that the inhibitory effects of DEX on insulin receptor kinase activity are not influenced by actinomycin D treatment, providing evidence for a transcription-independent or nongenomic mechanism. Control experiments were performed to ensure that actinomycin D treatment inhibited gene transcription under these conditions. Comparable time frames were used as to previous DEX stimulations: cells were pretreated for 45 min with actinomycin D followed by 45 min stimulation with TNF. Under these conditions, a TNF-dependent increase in the activity of a 3-kappaB sites containing reporter construct was seen, which was abolished in the presence of actinomycin D (Figure 4B). Hence, TNF-induced reporter activity was completely blocked in the presence of actinomycin D, demonstrating that actinomycin D inhibits transcription.

118 Glucocorticoids inhibit insulin signaling

Figure 4. Nongenomic DEX-induced inhibition of insulin signaling. (A) Adipocytes were pre- incubated for 15 min with actinomycin D (0.1 µM), where after insulin (100 IE/ml) was added for 45 min, followed by 10, 20 or 30 min DEX (1 µM) treatment. Cells were lysed and analyzed on Western blot in order to determine the phosphorylation levels of the insulin receptor and IRS-1. Total insulin receptor and IRS-1 expression levels were analyzed to evaluate for equal protein loading. A representative experiment (out of 3) is shown. (B) Inhibition of TNF-induced NFκB-reporter activity by actinomycin D treatment. 24 hrs after transfection with the NFκB-luciferase plasmid, actinomycin D (0.1 µM) was added to the cell culture media for 45 min, followed by 45 min stimulation with TNFα (100 ng/ml). Actinomycin D treatment abolished TNF-induced reporter gene activity and this was not a consequence of reduced cell viability as assessed by MTT viability assay (not shown). Activity was normalized to that of Renilla and expressed relative to the untreated control. Results are expressed as the mean ± SEM of triplicate determinations. DEX, dexamethasone; INSR, insulin receptor; IRS-1, insulin receptor substrate-1; p, phosphorylated; RLU, relative luciferase activity.

GC receptor-dependent inhibition of insulin receptor activity To find out whether the observed DEX-induced inhibition of insulin receptor tyrosine kinase activity is GC receptor-dependent, experiments were employed using a pharmacological GC receptor ligand (RU486, Mifepristone). Adipocytes were pre-incubated for 1 hr with insulin, treated for 5 min with increasing RU486 concentrations, and subsequently stimulated for 30 min with or without DEX. Cell lysates were immunoblotted to monitor GC receptor activity employing a phospho-specific Ab against the activated GC receptor. Because IRS-1 is a direct downstream insulin receptor target, phosphorylated IRS reflects insulin receptor kinase activity. Increased IRS-1 phosphorylation was detected upon insulin stimulation, indicating increased insulin receptor enzymatic activity (Figure 5). Cells incubated with combinations of high concentrations RU486 and DEX displayed decreased expression of phosphorylated GC receptor (indicating reduced GC receptor activation) and increased IRS-1 phosphorylation (reflecting increased insulin receptor activity). Thus, these results demonstrate that high RU486 concentrations antagonize the inhibitory effects of DEX on insulin receptor catalytic activity. In contrast, cells incubated with combinations of low RU486 concentrations together

119 Chapter 6

with DEX showed increased GC receptor phosphorylation and reduced phosphorylated IRS-1. Accordingly, RU486 at low concentration is not able to antagonize the inhibitory action of DEX on insulin receptor enzymatic activity. These experiments reveal a correlation between the activated GC receptor and reduced insulin receptor activity and therefore provide evidence for a GC receptor-dependent mechanism of DEX-induced inhibition of insulin receptor kinase activity.

Figure 5. GC receptor-dependent inhibition of insulin receptor kinase activity. (A) Adipocytes were pre- incubated for 1 hr with insulin (100 IE/ml), followed by a 5 min treatment with increasing RU486 concentrations (50 mM - 50 pM), where after 1 µM DEX was added for 30 min. Catalytic activities of the GC receptor and insulin receptor were determined on Western blot employing phospho-specific Abs against the activated GC receptor and IRS-1 respectively. Immunoblots were analyzed for total GC receptor, IRS-1 and actin to evaluate for equal loading. Three independent experiments were performed and comparable outcomes were obtained. (B) Western blots were quantitated and the phosphorylation status of the GC receptor and IRS-1 is depicted on the y- axis. DEX, dexamethasone; IRS-1, insulin receptor substrate-1; GR, glucocorticoid receptor; p, phosphorylated.

DEX inhibits IGF-1 induced signaling, but does not interfere with IL-6 signaling The kinases evaluated in this study act as intermediates for the signal transduction of insulin and IGF-1 (insulin-like growth factor 1) (i.e. two insulin receptor substrates), indicating that insulin and IGF-1 share common signal transduction. The effect of DEX on IGF-1-induced signaling was monitored in adipocytes which were pre-incubated for 1 hr with IGF-1 followed by DEX treatment for 10, 20 or 30 min. Cells were lysed and analyzed on Western blot employing the same panel of phospho-specific Abs which were used to study insulin signaling. DEX treatment suppressed IGF-1-induced phosphorylation of the insulin receptor, IRS-1, p70S6k, PKB, GSK-3, PDK and Fyn (Figure 6A). Altogether, these studies show that short-term DEX treatment interferes with insulin and IGF-1-induced signaling. In order to

120 Glucocorticoids inhibit insulin signaling

examine the specificity of the observed effects of DEX on insulin receptor-mediated signaling, we next investigated whether DEX interfered with a different kinase cascade. To this end, IL-6 (a gp130 receptor-ligand) was added to adipocytes for 30 min followed by 30 min DEX incubation. Cell lysates were analyzed on Western blot for phosphorylated STAT3, JAK2 and ERK1/2 (Figure 6B), key players in IL-6-induced signaling 22,23. These results indicate that DEX does not interfere with IL-6 signaling, further underscoring the specificity of the observed effects.

Figure 6. DEX inhibits IGF-1-induced signaling, but does not interfere with IL-6 signaling. (A) Adipocytes were pre-incubated for 1 hr with IGF-1 (20 ng/ml) and treated with 1 µM DEX for 10, 20 or 30 min. Whole lysates were immunoblotted using activation-state-specific Abs against several kinases involved in insulin signaling, including the insulin receptor, IRS-1, p70S6k, PKB, GSK-3, PDK and Fyn. Total protein levels were determined by Western blot employing the indicated Abs. (B) To ensure specificity in these experiments, adipocytes were pretreated with IL-6 (100 ng/ml) for 30 min and then incubated for 10, 20 or 30 min with 1 µM DEX (control). Cell lysates were analyzed on Western blot using phospho-specific Abs against STAT3, JAK2 and ERK1/2. Equal protein loading was verified employing Abs against proteins of interest. The immunoblots represent three independent experiments, and representative experiments are shown. DEX, dexamethasone; IGF-1, insulin-like growth factor 1; INSR, insulin receptor; IRS-1, insulin receptor substrate-1; GSK-3, glycogen- synthase kinase-3; PKB, protein kinase B; PDK, phosphoinositide-dependent protein kinase; STAT3, signal transducer and activator of transcription 3; JAK2, Janus kinase 2; ERK1/2, extracellular signal- regulated kinase 1/2; p, phosphorylated.

121 Chapter 6

Sustained DEX-induced inhibition of insulin signaling The present study shows that DEX inhibits insulin receptor-dependent signaling in adipocytes and T cells. These inhibitory effects of DEX are evident within the 30 min time frame. Finally, the effect of DEX at later time points was monitored to determine the temporal characteristics of DEX treatment on insulin receptor tyrosine kinase activation. Adipocytes were pretreated for 1 hr with insulin and subsequently stimulated up to 4 hrs with DEX. Cells were lysed at different time points and total cell extracts were used for Western blot analysis employing phospho-specific Abs against the insulin receptor and IRS-1 (Figure 7). These findings indicate that the inhibitory effects of DEX on insulin receptor and IRS phosphorylation are sustained for at least 4 hrs. Thus, these studies suggest that nongenomic GC receptor-dependent inhibition of insulin signaling is not a transient effect, although involvement of genomic effects at later time point cannot be excluded.

Figure 7. Sustained DEX-induced inhibition of insulin receptor activity. Insulin pretreated adipocytes (1 hr, 100 IE/ml) were stimulated with 1 µM DEX for 0, 30, 60 120 or 240 min. Cell lysates were immunoblotted using phosphospecific Abs against INSR and IRS-1. Total insulin receptor and IRS-1 expression was analyzed to evaluate for equal loading. Three independent experiments were performed and reproducible results were obtained. DEX, dexamethasone; INSR, insulin receptor; IRS-1, insulin receptor substrate-1; p, phosphorylated.

Discussion GCs form the basis of current anti-inflammatory and immunosuppressive therapy. GC treatment is significantly hampered by side effects, such as GC-induced insulin resistance, but the underlying molecular mechanisms remain unclear. Well-defined GC-induced effects are mediated via GC receptor-dependent transcriptional changes. Previous studies have explored long-term (i.e. ≥ 24 hrs) effects of DEX on glucose transport mechanisms and insulin signaling. It has been shown that DEX incubation results in decreased glucose uptake and

122 Glucocorticoids inhibit insulin signaling

impaired insulin signaling in 3T3 adipocytes and primary rat adipocytes 11,12,24-27. Although of significant importance for our understanding of the molecular mechanism underlying GC- induced insulin resistance, we are not aware of previous studies on rapid nongenomic (i.e. ≤ 30 min) effects of DEX on insulin signaling. More generally, evidence for early GC effects on cellular physiology is accumulating, but the responsible molecular mechanisms remain to be defined 28-30. We have recently reported nongenomic GC receptor-induced inhibition of T cell receptor signaling, providing further proof for rapid GC-effects on cellular function 15. In the current study, we demonstrate early DEX-induced effects on the signal transduction of adipocytes and CD4+ T cells using a peptide array approach. Phosphorylation of insulin receptor consensus substrates was strikingly inhibited due to short-term DEX treatment. Furthermore, reduced phosphorylation of AMPK, p70S6k, GSK-3 and Fyn kinase consensus substrates was detected in DEX treated cells, and Western blotting confirmed that DEX impairs insulin signaling. Experiments employing a pharmacological transcription inhibitor (actinomycin D) and a GC receptor-ligand (RU486) provided evidence for a nongenomic GC receptor-dependent mechanism of DEX-induced inhibition of insulin receptor activity. We conclude that short-term DEX treatment interferes with insulin receptor-mediated signaling and the fact that these effects were seen in two different cell-types suggests a general significance of DEX-induced inhibition of insulin signaling. We anticipate that such decreased insulin signaling might play an important role in GC-induced insulin resistance. The notion that DEX directly influences insulin receptor signaling rather than affecting cellular expression levels of the insulin receptor or downstream signaling targets, is supported by previous work 11. These authors showed that a 2 or 8 hr DEX treatment did not affect IRS- 1, PI3K and PKB expression in primary rat adipocytes. This is consistent with our observation that the total cellular content of insulin receptor and downstream signaling targets did not change due to 30 min DEX treatment and it provides further evidence that DEX directly inhibits insulin receptor-mediated signaling, rather than affecting absolute insulin receptor expression levels. An important question concerns the mechanism by which GC receptor engagement through DEX interferes with activity of the insulin receptor. Previous studies have shown that activation of JNK, a MAPK family member, inhibits insulin receptor-dependent signaling resulting in insulin resistance 18-20. We have investigated the effects of short-term DEX

123 Chapter 6

treatment on JNK phosphorylation. These findings showed increased JNK phosphorylation in DEX treated adipocytes compared to non-treated cells. Furthermore, the effect of DEX on IKK phosphorylation was analyzed, as this serine kinase has also been suggested to be involved in DEX-induced insulin resistance 16,17,21. However, we did not detect a change in IKK phosphorylation due to short-term DEX treatment (figure 3C). Further studies are needed to address the broader implications of JNK and IKK in generating DEX-induced insulin resistance. Based on the present study, we postulate a novel molecular mechanism for nongenomic GC receptor-mediated inhibition of insulin signaling that is independent of the classical (genomic) mode of GC action. These findings could have clinical implications as this further underscores the importance of the development of GC analogues, which retain their immunosuppressive and anti-inflammatory activities without having an accompanying effect on insulin receptor tyrosine kinase activity. Characterization of GC analogues could mark an important step towards the development of a new class of safer GC preparations.

Acknowledgments We thank Hans Verdeurmen and Maarten Bijlsma for technical support.

124 Glucocorticoids inhibit insulin signaling

Reference List

(1) Amatruda JM, Livingston JN, Lockwood DH. Cellular mechanisms in selected states of insulin resistance: human obesity, glucocorticoid excess, and chronic renal failure. Diabetes Metab Rev. 1985;1:293-317. (2) Gurwitz JH, Bohn RL, Glynn RJ et al. Glucocorticoids and the risk for initiation of hypoglycemic therapy. Arch Intern Med. 1994;154:97-101. (3) Pagano G, Cavallo-Perin P, Cassader M et al. An in vivo and in vitro study of the mechanism of prednisone-induced insulin resistance in healthy subjects. J Clin Invest. 1983;72:1814-1820. (4) Czech MP, Corvera S. Signaling mechanisms that regulate glucose transport. J Biol Chem. 1999;274:1865-1868. (5) Chan TO, Rittenhouse SE, Tsichlis PN. AKT/PKB and other D3 phosphoinositide-regulated kinases: kinase activation by phosphoinositide-dependent phosphorylation. Annu Rev Biochem. 1999;68:965- 1014. (6) Cheatham B, Vlahos CJ, Cheatham L et al. Phosphatidylinositol 3-kinase activation is required for insulin stimulation of pp70 S6 kinase, DNA synthesis, and glucose transporter translocation. Mol Cell Biol. 1994;14:4902-4911. (7) Hill MM, Clark SF, Tucker DF et al. A role for protein kinase Bbeta/Akt2 in insulin-stimulated GLUT4 translocation in adipocytes. Mol Cell Biol. 1999;19:7771-7781. (8) Proud CG, Denton RM. Molecular mechanisms for the control of translation by insulin. Biochem J. 1997;328 ( Pt 2):329-341. (9) Wang Q, Somwar R, Bilan PJ et al. Protein kinase B/Akt participates in GLUT4 translocation by insulin in L6 myoblasts. Mol Cell Biol. 1999;19:4008-4018. (10) White MF. The insulin signalling system and the IRS proteins. Diabetologia. 1997;40 Suppl 2:S2-17. (11) Buren J, Liu HX, Jensen J, Eriksson JW. Dexamethasone impairs insulin signalling and glucose transport by depletion of insulin receptor substrate-1, phosphatidylinositol 3-kinase and protein kinase B in primary cultured rat adipocytes. Eur J Endocrinol. 2002;146:419-429. (12) Turnbow MA, Keller SR, Rice KM, Garner CW. Dexamethasone down-regulation of insulin receptor substrate-1 in 3T3-L1 adipocytes. J Biol Chem. 1994;269:2516-2520. (13) Diks SH, Kok K, O'Toole T et al. Kinome profiling for studying lipopolysaccharide signal transduction in human peripheral blood mononuclear cells. J Biol Chem. 2004;279:49206-49213. (14) Houseman BT, Huh JH, Kron SJ, Mrksich M. Peptide chips for the quantitative evaluation of protein kinase activity. Nat Biotechnol. 2002;20:270-274. (15) Lowenberg M, Tuynman J, Bilderbeek J et al. Rapid immunosuppressive effects of glucocorticoids mediated through Lck and Fyn. Blood. 2005;106:1703-1710. (16) Arkan MC, Hevener AL, Greten FR et al. IKK-beta links inflammation to obesity-induced insulin resistance. Nat Med. 2005;11:191-198. (17) Cai D, Yuan M, Frantz DF et al. Local and systemic insulin resistance resulting from hepatic activation of IKK-beta and NF-kappaB. Nat Med. 2005;11:183-190. (18) Hirosumi J, Tuncman G, Chang L et al. A central role for JNK in obesity and insulin resistance. Nature. 2002;420:333-336. (19) Liu G, Rondinone CM. JNK: bridging the insulin signaling and inflammatory pathway. Curr Opin Investig Drugs. 2005;6:979-987. (20) Nguyen MT, Satoh H, Favelyukis S et al. JNK and tumor necrosis factor-alpha mediate free fatty acid- induced insulin resistance in 3T3-L1 adipocytes. J Biol Chem. 2005;280:35361-35371. (21) Yuan M, Konstantopoulos N, Lee J et al. Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted disruption of Ikkbeta. Science. 2001;293:1673-1677. (22) Heinrich PC, Behrmann I, Haan S et al. Principles of interleukin (IL)-6-type cytokine signalling and its regulation. Biochem J. 2003;374:1-20. (23) Kamimura D, Ishihara K, Hirano T. IL-6 signal transduction and its physiological roles: the signal orchestration model. Rev Physiol Biochem Pharmacol. 2003;149:1-38. (24) Garvey WT, Huecksteadt TP, Monzon R, Marshall S. Dexamethasone regulates the glucose transport system in primary cultured adipocytes: different mechanisms of insulin resistance after acute and chronic exposure. Endocrinology. 1989;124:2063-2073.

125 Chapter 6

(25) Lundgren M, Buren J, Ruge T, Myrnas T, Eriksson JW. Glucocorticoids down-regulate glucose uptake capacity and insulin-signaling proteins in omental but not subcutaneous human adipocytes. J Clin Endocrinol Metab. 2004;89:2989-2997. (26) Saad MJ, Folli F, Araki E et al. Regulation of insulin receptor, insulin receptor substrate-1 and phosphatidylinositol 3-kinase in 3T3-F442A adipocytes. Effects of differentiation, insulin, and dexamethasone. Mol Endocrinol. 1994;8:545-557. (27) Sakoda H, Ogihara T, Anai M et al. Dexamethasone-induced insulin resistance in 3T3-L1 adipocytes is due to inhibition of glucose transport rather than insulin signal transduction. Diabetes. 2000;49:1700- 1708. (28) Buttgereit F, Scheffold A. Rapid glucocorticoid effects on immune cells. Steroids. 2002;67:529-534. (29) Cato AC, Nestl A, Mink S. Rapid actions of steroid receptors in cellular signaling pathways. Sci STKE. 2002;2002:RE9. (30) Wehling M. Specific, nongenomic actions of steroid hormones. Annu Rev Physiol. 1997;59:365-393.

126

Chapter 7

A role for STAT5 in steroid resistant ulcerative colitis?

Mark Löwenberg1, Meike Scheffer1, Auke Verhaar1, Maikel Peppelenbosch2, Daniel Hommes1

1 Department of Gastroenterology and Hepatology, Academic Medical Center, Amsterdam, The Netherlands 2 Department of Cell Biology, University of Groningen, Groningen, The Netherlands

Published, in part, in: Inflammatory Bowel Diseases. 2006; 12(7):665

127 Chapter 7

Letter To The Editor:

Glucocorticoids (GCs) are powerful immunosuppressive agents and form the basis of current ulcerative colitis (UC) therapy. However, GC resistance has been reported in up to 30% of UC patients, thereby significantly complicating the clinical management of UC 1,2. Delineation of the molecular basis responsible for GC resistance is vital for the development of new therapeutic options and may provide novel insight into the underlying inflammatory mechanisms. A physical and functional interaction between the GC receptor and signal transducers and activators of transcription 5 (STAT5) has been demonstrated 3,4. Goleva et al. reported a defect in GC receptor nuclear translocation in response to GCs due phosphorylated STAT5- GC receptor heterodimer formation resulting in GC insensitivity 5. This observation supported previous work demonstrating STAT5-dependent inhibition of GC-induced transcriptional effects through direct protein-protein interaction 6. We have performed experiments in which GC sensitive Caco-2 cells were cotransfected (electroporation) with a glucocorticoid responsive element (GRE) and STAT5 construct. Control cells were cotransfected with GRE and an empty vector (PUC). 24 hrs post-transfection, cells were treated up to 48 hrs with 1µM dexamethasone (DEX), a synthetic GC. The GC receptor activation state was monitored in supernatants (Serum Excreted Alkaline Phosphatase reporter assay), which were collected at the indicated time points following DEX treatment. As can be seen from Figure 1, STAT5- transfected cells displayed reduced GC receptor reporter activity following DEX treatment compared to control cells, suggesting that STAT5 is able to induce GC insensitivity. In line with previous studies, these observations demonstrate STAT5-dependent inhibition of GC- mediated transcription upon heterologous STAT5 expression.

128 STAT5 and steroid resistant UC

Figure 1. STAT5-dependent inhibition of DEX GRE-driven reporter activity. A SEAP reporter assay was performed to investigate whether STAT5 could interfere with GC signaling. GC sensitive Caco-2 cells were cotransfected with a STAT5 construct or a control vector (PUC) together with a GRE plasmid. Cells were incubated with DEX (1 µM) and supernatant was collected at different time points for reporter gene activation measurement (shown on the y-axis). DEX-treated Caco-2 cells transfected with a control vector revealed GC receptor activity in time. STAT5 transfected cells (indicated by the black dots) demonstrated decreased GC receptor activity compared to control cells (as shown by the white dots) in the presence of DEX, as indicated by reduced reporter gene activity. Results are expressed as the mean ± SD triplicate determinations. Three independent experiments were performed and comparable results were obtained. GC, glucocorticoid; GRE, glucocorticoid responsive element; SEAP, Serum Excreted Alkaline Phosphatase; DEX, dexamethasone.

The STAT family consists of 7 STAT members (1, 2, 3, 4, 5A, 5B, 6) that play an important role in inflammatory responses 7. STAT proteins are cytoplasmic transcription factors that become phosphorylated upon signaling by extra-cellular cytokines, after which dimerization and nuclear translocation takes place, thereby regulating transcriptional processes. The clinical importance of GC resistance combined with the above mentioned considerations, prompted us to investigate the role of STAT5 in GC resistant UC. Colon biopsies from GC resistant (n=5) and non-GC resistant (n=4) active UC patients were analyzed with immunohistochemistry. Biopsies from patients who underwent colonoscopy for colon cancer screening (n=5) served as a control. These studies revealed the presence of phosphorylated (i.e. active) STAT5 in colon biopsies of GC resistant UC patients, which was localized to intestinal epithelial cells and lamina propria mononuclear cells (LPMCs) (Figure 2A, B). In contrast, almost no phosphorylated STAT5 was observed in biopsies obtained from non-GC resistant UC patients and in biopsies from control (non-UC) subjects (Figure 2 C,D). In addition, tissue lysates prepared from colon biopsies obtained from GC resistant UC patients were subjected to GC receptor immunoprecipitation using an antibody against the GC receptor. GC receptor immunoprecipitates were immunoblotted for the presence of STAT5,

129 Chapter 7

demonstrating phosphorylated STAT5 in these samples (not shown). These observations revealed an in vivo interaction between the GC receptor and phosphorylated STAT5 in steroid resistant UC. This work suggests that STAT5 might be involved in GC resistant UC, and identifies LPMCs and intestinal epithelial cells as target cells for GC therapy in this disease.

Figure 2. Increased phosphorylated STAT5 in GC resistant UC. Colon biopsies from GC resistant and non-GC resistant UC patients (obtained from affected regions of inflammation), and biopsies from control subjects were analyzed for the presence of phosphorylated STAT5. A, B: Phosphorylated STAT5 was detected in GC resistant UC, which was localized to intestinal epithelial cells and LPMCs. Almost no phosphorylated STAT5 was seen in non-GC resistant UC (C) and control patients (D); indicated with the black arrows. Magnification, 200x. GC, glucocorticoid; UC, ulcerative colitis; LPMC, lamina propria mononuclear cells.

Previous studies revealed a potential role for STAT1, STAT3 and STAT4 in experimental colitis 8,9. Correspondingly, increased amounts of total and phosphorylated STAT3 were detected in LPMCs from UC and Crohn’s disease patients compared to non-inflammatory control cells, as assessed by flow cytometry and immunohistochemistry 8,10,11. There are no data available on STAT5 in UC. Our study suggests that a subgroup of the UC population (i.e. GC resistant UC patients) have increased phosphorylated STAT5 levels in the colon. Hence, this work identifies STAT5 as a potential molecular target for the clinical management of GC resistant UC. Further studies will have to define the exact role of STAT5 in the pathogenesis of GC resistant UC.

130 STAT5 and steroid resistant UC

Reference List

(1) Flood L, Lofberg R, Stierna P, Wikstrom AC. Glucocorticoid receptor mRNA in patients with ulcerative colitis: a study of responders and nonresponders to glucocorticosteroid therapy. Inflamm Bowel Dis. 2001;7:202-209. (2) Kjeldsen J. Treatment of ulcerative colitis with high doses of oral prednisolone. The rate of remission, the need for surgery, and the effect of prolonging the treatment. Scand J Gastroenterol. 1993;28:821- 826. (3) Cella N, Groner B, Hynes NE. Characterization of Stat5a and Stat5b homodimers and heterodimers and their association with the glucocortiocoid receptor in mammary cells. Mol Cell Biol. 1998;18:1783- 1792. (4) Stocklin E, Wissler M, Gouilleux F, Groner B. Functional interactions between Stat5 and the glucocorticoid receptor. Nature. 1996;383:726-728. (5) Goleva E, Kisich KO, Leung DY. A role for STAT5 in the pathogenesis of IL-2-induced glucocorticoid resistance. J Immunol. 2002;169:5934-5940. (6) Biola A, Lefebvre P, Perrin-Wolff M et al. Interleukin-2 inhibits glucocorticoid receptor transcriptional activity through a mechanism involving STAT5 (signal transducer and activator of transcription 5) but not AP-1. Mol Endocrinol. 2001;15:1062-1076. (7) Darnell JE, Jr. STATs and gene regulation. Science. 1997;277:1630-1635. (8) Mudter J, Weigmann B, Bartsch B et al. Activation pattern of signal transducers and activators of transcription (STAT) factors in inflammatory bowel diseases. Am J Gastroenterol. 2005;100:64-72. (9) Schreiber S, Rosenstiel P, Hampe J et al. Activation of signal transducer and activator of transcription (STAT) 1 in human chronic inflammatory bowel disease. Gut. 2002;51:379-385. (10) Lovato P, Brender C, Agnholt J et al. Constitutive STAT3 activation in intestinal T cells from patients with Crohn's disease. J Biol Chem. 2003;278:16777-16781. (11) Musso A, Dentelli P, Carlino A et al. Signal transducers and activators of transcription 3 signaling pathway: an essential mediator of inflammatory bowel disease and other forms of intestinal inflammation. Inflamm Bowel Dis. 2005;11:91-98.

131

Chapter 8

Discussion

133 Chapter 8

The goal of this thesis was to further unravel signaling networks in physiology and IBD. The majority of the work presented here dealt with defining the interactions between therapeutic interventions and inflammatory signal transduction pathways in order to identify cellular targets for novel therapies. c-Raf inhibitors as a future therapy for inflammatory diseases Treatment with the synthetic guanylhydrazone semapimod (formerly known as CNI-1493) showed promising anti-inflammatory effects in severe CD 1. Although is has been shown that semapimod suppresses JNK and p38 MAPK phosphorylation leading to reduced production of pro-inflammatory cytokines and nitric oxide 1,2, the underlying mechanism remained elusive. In chapter 3, we have identified c-Raf as the cellular target of semapimod 3, and we here describe an association between semapimod-induced c-Raf inhibition and clinical responses in severe CD. MAPK signaling cascades cooperate in an orchestrated manner in inflammatory responses. Extensive cross-talk between ERK, JNK and p38 MAPK pathways, as well as between MAPK cascades and other inflammatory signaling routes (such as NFκB) have been reported 4-8. Our data indicate that the pro-inflammatory effects of c-Raf include not only activation of ERK, but also JNK and p38 MAPK. We demonstrate that semapimod specifically targets c-Raf leading to reduced ERK, JNK and p38 MAPK activities, and these findings could have important clinical implications. First of all, these observations indicate that c-Raf might be a cellular target for anti-inflammatory small molecules in the treatment of CD and possibly other inflammatory diseases. c-Raf inhibitors, which have been tested as anticancer strategies 9, may therefore have widespread applications in the treatment of inflammatory disorders. Second, these observations may help to redesign signal transduction intervention strategies, as upstream inhibition of MAPK signal transduction pathways could be the strategy of choice for pharmaceutical companies rather than inhibiting downstream targets. Several specific JNK and p38 MAPK inhibitors failed in preclinical studies or showed modest or no anti-inflammatory effects in clinical trials 10,11. Similar to the observation that semapimod-induced c-Raf inhibition results in clinical improvement in severe CD, recent findings provided insight into the immunosuppressive effect of azathioprine. It was shown that azathioprine and its metabolite metabolite 6- mercaptopurine inhibit activity of the proximal signaling molecule Rac, through inhibition of

134 Discussion

Vav guanosine exchange activity in T lymphocytes. Azathioprine-induced inhibition of Rac activity resulted in reduced downstream activation of MEK, NFκB and bcl-x(L), leading to induced T cell apoptosis 12,13. Thus, recent observations suggest that therapeutic interventions which specifically target proximal signaling molecules (such as c-Raf or Rac) might be a powerful strategy for combating CD.

A novel mechanism for an old drug: glucocorticoids inhibit proximal T cell receptor signaling through Lck/Fyn Glucocorticoids (GCs) form the basis of current anti-inflammatory and immunosuppressive therapy. GC-induced transcriptional action is responsible for immunosuppressive effects as well as for (unwanted) side effects, such as insulin resistance, hypertension and osteoporosis; associated with an increased mortality risk 14-16. As a result, there is an unmet need for novel treatment strategies with more potency and less toxicity. Some biologically important effects of GCs occur so rapid, that these cannot result from gene transcription, and therefore need to be caused by nongenomic effects 17-19. In chapter 4, we have identified a unique and unexpected role for GCs in the control of T cell receptor (TCR) signaling by nongenomic GC receptor-dependent modulation of Lck and Fyn kinase activities 20. Additional studies revealed the existence of a close physical and functional interaction between the TCR complex and the unligated membrane-bound GC receptor (chapter 5). We here demonstrated that GC receptors are important for efficient TCR signaling. Interestingly, GC treatment resulted in rapid dissolution of membrane-linked Lck/Fyn containing GC receptor protein complexes, resulting in impaired TCR signaling. Hence, these observations provide a new concept in cell biology, demonstrating a dichotomal functional role for GC receptors: unligated as part of the TCR complex and ligated as a transcription factor. GC analogues which selectively target proximal TCR signaling events through Lck/Fyn may represent novel therapeutic opportunities for more specific immunosuppression.

Kinomics: a useful tool for studying cellular signal transduction The kinome regulates virtually all major intracellular metabolic pathways, and is therefore largely responsible for cellular function in physiology and disease. DNA array techniques allow comprehensive analysis of the genome or transcriptome. However, high throughput

135 Chapter 8

array-based analysis of signal transduction pathways remains troublesome. One of the goals of this thesis was to shed more light on several mysterious GC-induced signaling effects, employing a novel array technique for kinome-wide analysis 21,22.

• The importance of GCs in clinical immunosuppression combined with the unknown basis of nongenomic GC-induced effects, prompted us to investigate rapid GC effects on the T lymphocyte kinome. Lysates, prepared from activated T cells pretreated for 10 minutes with or without the GC analogue dexamethasone, were supplemented with radioactive ATP and incubated on pepchip arrays. These studies showed marked early differences in phosphorylation patterns between GC-treated and non-GC-treated cells, providing proof for nongenomic GC effects on the T cell signal transduction kinome (chapter 4). Among the most prominent effects observed was reduced phosphorylation of Lck/Fyn kinase consensus substrates due to short-term GC treatment, pointing out reduced Lck and Fyn enzymatic activities. Hence, this non-biased kinome analysis approach identified Lck and Fyn as cellular targets for nongenomic GC activities 20, and this was confirmed with conventional laboratory techniques (chapter 4 and 5).

• GC-induced insulin resistance complicates the clinical management of GC therapy. Previous studies demonstrated that long-term (i.e. >24 hr) GC treatment impairs insulin signaling leading to insulin resistance 23,24, but the underlying mechanism remains obscure at best. Rapid GC effects on insulin signaling have not been studied so far, but are vital in order to unravel molecular mechanisms responsible for GC- induced insulin resistance. In chapter 6, we have investigated nongenomic GC effects on insulin receptor-mediated signaling in adipocytes. Kinome profiling revealed that short-term GC treatment significantly impaired insulin signaling, as indicated by reduced enzymatic activities of the insulin receptor and several downstream insulin receptor signaling intermediates. Conventional techniques confirmed these observations and revealed a nongenomic GC receptor-dependent mechanism for GC- induced inhibition of insulin signaling. These studies identified the insulin receptor as a direct GC target. This work could have clinical implications as this further underscores the importance of the development of GC analogues that retain their

136 Discussion

immunosuppressive activities without having an accompanying effect on insulin receptor tyrosine kinase activity.

Overall, these studies demonstrated that the kinome reacts dynamically to short-term GC stimulation and has helped in identifying novel GC targets. We feel that peptide arraying for kinome-wide analysis is a useful method to determine the enzymatic activities of a large group of kinases, offering high throughput analysis of cellular metabolism and signal transduction. Kinome profiling has several advantages compared to traditional in vitro enzyme activity assays. First of all, this approach enables high throughput array-based assessment of numerous signaling pathways, which opens the possibility to study downstream signaling events and feedback mechanisms. Second, redundancy has been described to be important in cellular metabolism and these phenomena can be investigated employing this technique. Furthermore, as there is great potential for cross-reactivity of therapeutics, this tool can be used to assess molecular specificity and to identify off-target interactions. Lastly, peptide arraying for kinome-wide analysis is a non-biased method which can be used for the identification of drug targets or delineation of signal transduction mechanisms.

Future directions: where to go next? The challenge now is to understand the functional implications of signal transduction pathways that are involved in specific immunological processes. Increased knowledge of intracellular signaling will be important for the identification of targets for new generations of small molecule inhibitors that have higher efficacy, fewer side effects and lower costs. Recent studies have shown that signaling molecules are attractive targets for therapeutic intervention in inflammatory diseases 11,25. On the other hand, signaling cascades are involved in cellular physiology, such as mounting effective immune responses. Inhibition of signal transduction pathways might therefore result in serious side effects, such as compromised normal host defences leading to opportunistic infections. However, evidence suggests that by selective targeting of signaling components it might be possible to minimize systemic toxicity 1,11,25. Therefore, it is crucial to identify individual key components for a specific disease in order to develop therapeutics that specifically interfere with signal transduction pathways.

137 Chapter 8

In this thesis, we have defined c-Raf as a cellular target for the anti-inflammatory small molecule semapimod (chapter 3). These observations open the possibility that c-Raf inhibitors constitute prime candidates for the treatment of IBD and possibly other inflammatory disorders, as upstream inhibition of MAPK signaling pathways could be a powerful approach for dampening immune responses. The results described in chapters 4 and 5 provide a molecular framework for understanding nongenomic GC-induced immunosuppression in T cells and suggest that GC analogues which specifically target Lck/Fyn-mediated TCR signaling may be useful for suppression of T cell-dependent pathogenic immune responses. Overall, this thesis provides insight into the underlying mechanisms of therapeutic interventions in physiology and IBD. A better understanding of these intracellular processes will allow us to develop novel drugs for the treatment of these disabling disorders.

138 Discussion

Reference List

(1) Hommes D, van den BB, Plasse T et al. Inhibition of stress-activated MAP kinases induces clinical improvement in moderate to severe Crohn's disease. Gastroenterology. 2002;122:7-14. (2) Bianchi M, Bloom O, Raabe T et al. Suppression of proinflammatory cytokines in monocytes by a tetravalent guanylhydrazone. J Exp Med. 1996;183:927-936. (3) Lowenberg M, Verhaar A, van den BB et al. Specific inhibition of c-Raf activity by semapimod induces clinical remission in severe Crohn's disease. J Immunol. 2005;175:2293-2300. (4) Adler V, Qu Y, Smith SJ et al. Functional interactions of Raf and MEK with JNK result in a positive feedback loop on the oncogenic Ras signaling pathway. Biochemistry. 2005;44:10784-10795. (5) Coso OA, Chiariello M, Yu JC et al. The small GTP-binding proteins Rac1 and Cdc42 regulate the activity of the JNK/SAPK signaling pathway. Cell. 1995;81:1137-1146. (6) Kyriakis JM, Avruch J. Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation. Physiol Rev. 2001;81:807-869. (7) Minden A, Lin A, Claret FX, Abo A, Karin M. Selective activation of the JNK signaling cascade and c- Jun transcriptional activity by the small GTPases Rac and Cdc42Hs. Cell. 1995;81:1147-1157. (8) Tibbles LA, Woodgett JR. The stress-activated protein kinase pathways. Cell Mol Life Sci. 1999;55:1230-1254. (9) Hilger RA, Scheulen ME, Strumberg D. The Ras-Raf-MEK-ERK pathway in the treatment of cancer. Onkologie. 2002;25:511-518. (10) Schreiber S, Feagan B, D'Haens G et al. Oral p38 mitogen-activated protein kinase inhibition with BIRB 796 for active Crohn's disease: a randomized, double-blind, placebo-controlled trial. Clin Gastroenterol Hepatol. 2006;4:325-334. (11) Lowenberg M, Peppelenbosch MP, Hommes DW. Therapeutic modulation of signal transduction pathways. Inflamm Bowel Dis. 2004;10 Suppl 1:S52-S57. (12) Poppe D, Tiede I, Fritz G et al. Azathioprine suppresses ezrin-radixin-moesin-dependent T cell-APC conjugation through inhibition of Vav guanosine exchange activity on Rac proteins. J Immunol. 2006;176:640-651. (13) Tiede I, Fritz G, Strand S et al. CD28-dependent Rac1 activation is the molecular target of azathioprine in primary human CD4+ T lymphocytes. J Clin Invest. 2003;111:1133-1145. (14) Leung DY, Bloom JW. Update on glucocorticoid action and resistance. J Allergy Clin Immunol. 2003;111:3-22. (15) Pisu M, James N, Sampsel S, Saag KG. The cost of glucocorticoid-associated adverse events in rheumatoid arthritis. Rheumatology (Oxford). 2005;44:781-788. (16) Schols AM, Wesseling G, Kester AD et al. Dose dependent increased mortality risk in COPD patients treated with oral glucocorticoids. Eur Respir J. 2001;17:337-342. (17) Buttgereit F, Straub RH, Wehling M, Burmester GR. Glucocorticoids in the treatment of rheumatic diseases: an update on the mechanisms of action. Arthritis Rheum. 2004;50:3408-3417. (18) Hafezi-Moghadam A, Simoncini T, Yang Z et al. Acute cardiovascular protective effects of corticosteroids are mediated by non-transcriptional activation of endothelial nitric oxide synthase. Nat Med. 2002;8:473-479. (19) Rhen T, Cidlowski JA. Antiinflammatory action of glucocorticoids--new mechanisms for old drugs. N Engl J Med. 2005;353:1711-1723. (20) Lowenberg M, Tuynman J, Bilderbeek J et al. Rapid immunosuppressive effects of glucocorticoids mediated through Lck and Fyn. Blood. 2005;106:1703-1710. (21) Diks SH, Kok K, O'Toole T et al. Kinome profiling for studying lipopolysaccharide signal transduction in human peripheral blood mononuclear cells. J Biol Chem. 2004;279:49206-49213. (22) Houseman BT, Huh JH, Kron SJ, Mrksich M. Peptide chips for the quantitative evaluation of protein kinase activity. Nat Biotechnol. 2002;20:270-274. (23) Buren J, Liu HX, Jensen J, Eriksson JW. Dexamethasone impairs insulin signalling and glucose transport by depletion of insulin receptor substrate-1, phosphatidylinositol 3-kinase and protein kinase B in primary cultured rat adipocytes. Eur J Endocrinol. 2002;146:419-429. (24) Turnbow MA, Keller SR, Rice KM, Garner CW. Dexamethasone down-regulation of insulin receptor substrate-1 in 3T3-L1 adipocytes. J Biol Chem. 1994;269:2516-2520. (25) Hommes DW, Peppelenbosch MP, van Deventer SJ. Mitogen activated protein (MAP) kinase signal transduction pathways and novel anti-inflammatory targets. Gut. 2003;52:144-151.

139

Appendices

141 Appendix 1

Appendix 1: Summary

During the last decade, important progress has been made in understanding cellular signal transduction that plays a crucial role in many aspects of immune-mediated inflammatory processes. The goal of this thesis was to study the interactions between signal transduction pathways and therapeutic interventions in order to create opportunities for the development of more powerful and less toxic therapies. Chapter 2 discusses molecular mechanisms that could be responsible for the striking difference in clinical outcome between step-up and top- down therapy in recent-onset Crohn’s disease (CD). Furthermore, we review the role of MAPK signaling pathways as potential targets for small molecule therapy in IBD. Chapter 3 identifies c-Raf as the cellular target of the anti-inflammatory small molecule semapimod, which demonstrated potent clinical effects in severe CD. We here discuss a principal role for c-Raf in CD pathogenesis which could have important clinical consequences, as these data suggest that c-Raf inhibitors constitute novel candidates for the treatment of CD. Chapter 4 and 5 dealt with defining a molecular mechanism responsible for nongenomic glucocorticoid (GC)-induced immunosuppression in T lymphocytes. We here demonstrate impaired TCR signaling due to dissociation of membrane-bound Lck/Fyn containing GC receptor multiprotein complexes following GC treatment. These findings suggest that inhibition of Lck/Fyn-mediated TCR signaling constitutes novel therapeutic opportunities for the development of specific immunosuppressive agents. In chapter 6, a mechanism is characterized underlying GC-induced insulin resistance, a major problem in the clinical management of GC therapy. These observations indicate that GCs directly inhibit insulin receptor activity leading to impaired insulin signaling. Finally, the results described in chapter 7 demonstrate increased intestinal activity of the transcription factor STAT5 in GC resistant ulcerative colitis (UC) patients. Previous findings together with our data suggest that STAT5 might be a therapeutic target for the clinical management of GC resistant UC.

142 Samenvatting voor niet ingewijden

Appendix 2: Samenvatting voor niet ingewijden

Inflammatoire darmziekten (IBD) De ziekte van Crohn en colitis ulcerosa zijn de meest voorkomende oorzaken van chronische ontstekingen van het maag-darm kanaal. In het Engels en in het Nederlands worden deze ziekten aangeduid als respectievelijk ‘inflammatory bowel diseases’ of inflammatoire darmziekten (afgekort IBD). Patiënten die lijden aan de ziekte van Crohn of colitis ulcerosa hebben te pas en te onpas ernstige ontstekingen in hun maag-darmstelsel. De oorzaak daarvan is nog steeds onbekend, maar er zijn duidelijke aanwijzingen dat naast aangeboren (genetische) factoren ook het afweersysteem een rol speelt.

In gezonde darmen bevinden zich miljoenen micro-organismen, die normaal gesproken met de ontlasting het lichaam verlaten. Soms passeren onverhoopt potentiële ziekteverwekkers (zoals bepaalde bacteriën en virussen) de darmwand. Dankzij een activatie van afweercellen worden deze micro-organismen door het lichaam opgeruimd. Na deze taak volbracht te hebben gaat het afweersysteem weer “op een lager pitje”. Ons afweersysteem is dus in staat deze micro-organismen te herkennen en daarop te reageren. Bij IBD is er sprake van een continu (over)actief afweersysteem waardoor ook gezonde lichaamscellen beschadigd raken, en dit leidt tot een chronische ontsteking van het maag-darmstelsel. IBD patiënten ondervinden klachten als hevige buikpijn, diarree, bloed en slijm bij de ontlasting, koorts, etc. Het moge duidelijk zijn dat deze ziekten veelal gepaard gaan met een afgenomen kwaliteit van leven.

Nieuwe inzichten in het ziekteproces van inflammatoire darmziekten Tegenwoordig zijn de ziekte van Crohn en colitis ulcerosa steeds beter met medicijnen te behandelen. Helaas gaat de behandeling met bestaande medicijnen vaak gepaard met ernstige bijwerkingen. Of erger nog: ze slaan niet aan. Er is dus een continue vraag naar nieuwe krachtigere behandelingen met minder bijwerkingen. Momenteel wordt IBD behandeld met medicijnen die het afweersysteem onderdrukken en op deze manier de ontsteking remmen. Glucocorticoiden, bijvoorbeeld prednison, dexamethason en hydrocortison, zijn dit soort medicijnen. Glucocorticoiden vormen de basis van de IBD behandeling. Daarnaast worden in

143 Appendix 2

de kliniek zgn. ‘immunomodulatoren’ gebruikt bij de behandeling van IBD, zoals methrotrexaat en azathioprine. Deze middelen remmen het ontstekingsproces via andere mechanismen. Er zijn de laatste 10 jaar moleculaire aangrijpingspunten ontdekt die de ontwikkeling van nieuwe medicijnen mogelijk hebben gemaakt. Hieronder vallen antistoffen: dit zijn eiwitten die specifiek het ontstekingsproces kunnen remmen. Een voorbeeld van een antistof dat vandaag de dag niet meer is weg te denken uit de kliniek is anti-Tumor Necrosis Factor (TNF). Dit medicijn blokkeert de werking van het eiwit TNF (vandaar de naam: anti-TNF) dat een centrale rol speelt in de ziekte van Crohn en waarschijnlijk ook belangrijk is voor colitis ulcerosa. Deze antistoffen bleken effectief bij de behandeling van de ziekte van Crohn en colitis ulcerosa. De introductie van anti-TNF heeft de aanzet gegeven tot een nieuwe groep van medicijnen. Helaas werken deze therapieën niet altijd en gaan de behandelingen ook weer gepaard met ernstige bijwerkingen. Geen van de huidige behandelingen voor de ziekte van Crohn en colitis ulcerosa leiden tot definitieve genezing. Vandaar dat er wereldwijd gezocht wordt naar betere therapieën. In dit proefschrift hebben wij het werkingsmechanisme van bestaande medicijnen (zoals glucocorticoiden) en nieuwe behandelvormen (bijvoorbeeld semapimod) voor IBD onderzocht. Het doel was om zgn. signaaltransductie processen in ontstekingscellen te ontdekken die de ontwikkeling van nieuwe, betere en veiligere therapieën mogelijk zou kunnen maken.

Signaaltransductie Signaaltransductie betekent letterlijk: ‘het doorgeven van signalen buiten de cel naar binnen in de cel’. Dit gebeurt met behulp van signaalstoffen die via receptoren op de buitenkant van de cel het signaal doorsturen (Figuur 1). Bepaalde eiwitten, kinases genaamd, spelen een centrale rol in signaaltransductie processen doordat ze een ander eiwit kunnen veranderen (phosphoryleren). Als je het systeem van signaaleiwitjes in een cel vergelijkt met een elektronische schakeling, dan zijn fosforgroepen de elektriciteit. Kinases zorgen er voor dat een signaal door de cel geleid wordt naar de cel kern toe, waarna DNA wordt afgelezen en vervolgens eiwitten worden gevormd. Dit heeft tot gevolg dat er iets met de cel gebeurt, bijvoorbeeld beweging of celdeling.

144 Samenvatting voor niet ingewijden

Figuur 1. Het doorgeven van signalen binnen in de cel (zgn. signaaltransductie) zorgt ervoor dat signalen van buiten de cel worden doorgegeven naar de celkern waar DNA wordt afgelezen. Dit alles heeft tot gevolg dat ‘het gedrag’ van de cel wordt beïnvloed, zoals het groeien of delen van cellen.

Mini-eiwitjes (‘small molecules’) Sinds het begin van de jaren 90 zijn er nieuwe medicijnen (zgn. small molecules) ontwikkeld die bepaalde signaaltransductie eiwitten binnen in ontstekingscellen kunnen remmen en op deze manier het ontstekingsproces kunnen tegen gaan. Small molecules hebben verschillende voordelen vergeleken met de huidige behandelmogelijkheden. Zo kunnen deze mini-eiwitten in pilvorm worden toegediend en zijn ze -vergeleken met o.a. antistof therapie- relatief goedkoop. Er lijken maar weinig bijwerkingen zijn, maar er moet nog veel aanvullend onderzoek worden gedaan naar de effectiviteit en veiligheid van deze medicijnen. Recent is aangetoond dat de behandeling van ernstig zieke Crohn patienten met het small molecule semapimod effectief was, maar hoe het werkt bleef onopgehelderd. Er is nog veel onduidelijkheid ten aanzien van de precieze werking van cellulaire signaaltransductie, waarbij eerder genoemde kinases een centrale rol spelen. Een recent ontwikkelde laboratorium techniek (zgn. kinoom analyse/pepchip) maakt het mogelijk de activiteit van vele honderden kinases tegelijkertijd te analyseren. Van deze techniek hebben wij in dit proefschrift gebruik gemaakt om meer inzicht te krijgen in signaal transductie processen in ontstekingscellen.

145 Appendix 2

In dit proefschrift hebben wij bestudeerd hoe het mini-eiwit semapimod werkt en wat het effect ervan is binnen in ontstekingscellen. In hoofdstuk 3 beschrijven wij het werkingsmechanisme van semapimod. Deze studies lieten zien dat semapimod specifiek een eiwit (c-Raf) remt binnen in een bepaald type ontstekingscel met als gevolg onderdrukking van het ontstekingsproces. Deze bevindingen kunnen de aanzet geven tot de ontwikkeling van nieuwe medicijnen voor de behandeling van IBD en mogelijk ook andere chronische ontstekingsziekten. Daarnaast suggereren deze observaties dat andere c-Raf remmers, die al als anti-kanker medicijn zijn getest, mogelijk effectief zouden kunnen zijn voor de behandeling van IBD.

Nieuwe inzichten in het werkingsmechanisme van glucocorticoiden Tot slot behandelt dit proefschrift nieuwe inzichten in het werkingsmechanisme van glucocorticoiden. Zoals eerder vermeld vormen glucocorticoiden de basis van de IBD behandeling. De werking van glucocorticoiden komt o.a. tot stand doordat ze invloed hebben op het proces dat genen omzet in eiwitten (zgn. gentranscriptie) met als gevolg remming van de aanmaak en functie van ontstekingscellen. Dit hele proces neemt wel wat tijd in beslag en verklaart dat het klinische effect van glucocorticoiden vaak pas na enkele uren of dagen optreedt. Glucocorticoiden grijpen echter ook in op signaaltransductie routes en deze snelle effecten kunnen niet door het bovengenoemde mechanisme worden verklaard. Tot op heden wisten we niet goed hoe deze snelle effecten kunnen ontstaan. Aangezien kinases waarschijnlijk hierbij een belangrijke rol spelen hebben wij voor onze studies gebruik gemaakt van de hierboven beschreven moleculaire chip-technologie. In onze experimenten is gebruik gemaakt van T lymfocyten, bepaalde ontstekingscellen die een belangrijke rol spelen in IBD. Deze studies lieten zien dat een korte behandeling (10 min) met glucocorticoiden tot gevolg heeft dat vele kinases geactiveerd of juist geremd worden. Verdere analyse bracht aan het licht dat glucocorticoiden in dit korte tijdsbestek de activatie van Lck en Fyn remmen, twee eiwitten (kinases) die van essentieel belang zijn voor het activeren van T cellen (hoofdstuk 4); en in hoofdstuk 5 hebben wij vervolgens het onderliggende mechanisme geïndentificeerd. Deze bevindingen bieden aanknopingspunten voor de ontwikkeling van nieuwe (meer specifieke) behandelingen.

146

Appendix 3

Androgen-mediated regulation and functional implications of FKBP51 expression in prostate cancer

Phillip G. Febbo 1,4,6, Mark Löwenberg 8, Aaron R. Thorner 1, Myles Brown 1,4,6, Massimo Loda 2,5,7, Todd R. Golub 3,9,10

Departments of 1 Medical Oncology, 2 Pathology and 3 Pediatric Oncology, Dana-Farber Cancer Institute, Boston, U.S.A. Departments of 4 Medicine and 5 Pathology, Brigham and Women’s Hospital, Boston, U.S.A. Departments of 6 Medicine and 7 Pathology, Harvard Medical School, Boston, U.S.A. 8 Department of Experimental Internal Medicine, Academic Medical Center, Amsterdam, the Netherlands 9 Broad Institute of MIT and Harvard, Cambridge, U.S.A. 10 Howard Hughes Medical Institute

Journal of Urology. 2005 May;173(5): 1772-7

147 Appendix 3

Abstract Androgen ablation continues to be the most effective therapy for metastatic prostate cancer even as the biologically active androgen receptor (AR) target genes remain largely unknown. As the AR signaling pathway continues to be important in hormone refractory disease, effector AR target genes may have therapeutic import. We used oligonucleotide microarrays to identify genes with expression both induced by androgen and associated with androgen independent growth. The androgen-induced expression of FKBP51, a steroid receptor chaperone, was further investigated in LNCaP cells by Northern and Western analysis and in primary prostate specimens using immunohistochemistry. We used stable clones over expressing FKBP51 to test the functional effects of FKBP51 expression in LNCaP cells. Many genes had expression that correlates with androgen stimulation in LNCaP cells but relatively few had reproducible androgen-mediated changes in expression across multiple prostate cancer cell lines. FKBP51 had androgen-induced RNA and protein expression in LNCaP cells and decreased expression in normal prostate epithelial cells following castration. Further work demonstrated that FKBP51 induction was not a generalized response to cell proliferation, that FKBP51 protein physically interacts with the AR, and that LNCaP cells constitutively over expressing FKBP51 have increased ligand-mediated AR activation of both an exogenous AR-reporter construct and endogenous PSA. In conclusion, these results confirm FKBP51 as an androgen-induced gene, demonstrate a physical interaction between FKBP51 and the AR, and suggest that FKBP51 over-expression increases AR transcriptional activity in prostate cancer.

148 FKBP51 and the androgen receptor

Introduction Androgen ablation remains the most effective treatment for metastatic prostate cancer and one of the only therapies proven to prolong survival 1-4. Even as metastatic tumors progress in men with castrate levels of circulating testosterone, the importance of androgen receptor (AR) signaling remains central 5. AR, a member of the steroid superfamily of nuclear transcription factors, is required for androgen activity 6 and a focus of genetic alteration during the progression of prostate cancer from androgen sensitive to androgen independent disease 7,8. Most recently, increased AR expression has been found to be the most common molecular alteration in androgen independent xenograft models and necessary for androgen independent growth 9. The identification of genes regulated by AR remains a focus of great interest as AR- regulated genes important to prostate cancer biology are potential therapeutic targets. Here, we report the androgen-induced gene expression and functional characterization of a particular AR target, the immunophilin FKBP51.

Material and methods Androgen and Epidermal Growth Factor (EGF) stimulation of LNCaP cells. Early passage (23 – 40) LNCaP cells (ATCC, Manassas, VA, USA) were grown in RPMI 1640 (Cellgro, Herndon, VA, USA) with 10% FBS (Sigma, St. Louis, MO, USA) with HEPES (Cellgro), Na+ Pyruvate (Cellgro), and L-Glutamine (Cellgro). Cell stimulations were performed according to the schema provided (Fig. 1, A) in RPMI 1640 with 10% charcoal/dextran treated FBS (Cellgro) at 3 - 4 million cells/10 cm plate with or without R1881 (0.1 nM, NEN, Boston, MA, USA) and/or epidermal growth factor (EGF, 10 µg/mL, Sigma, St. Louis, MO, USA). RNA was collected by direct lysis with Trizol (Invitrogen, Carlsbad, CA, USA). Triplicate plates were collected for cell counts.

Microarray analysis. Total RNA (20µg) was used to synthesize cRNA as previously described 10 and hybridized to U95Av2 microarrays (Affymetrix). Microarray data files were scaled together, expression values were set at upper and lower thresholds (lower 5, upper 16000), and genes were excluded that had less than a 3 fold difference in expression between any two samples using GeneCluster® software (available at http://www.broad.mit.edu/ cancer/software/software.html). The remaining 3040 genes were ranked according to the

149 Appendix 3

Pearson correlation coefficient between their expression and that of prostate specific antigen (PSA) (Affymetrix accession # 1805 g at, GenBank X07730). The 50 genes with highest correlation were analyzed in 3 independent experiments. First, their expression was determined in an androgen sensitive derivative cell line from CWR22 xenografts (“CWR14”) following dihydrotestosterone (DHT) stimulation. The fold change (FC) is the 20-hour expression value divided by the lower of the 0 or 14-hour expression values. Second, expression was compared between CWR22 xenografts growing in normal male mice (androgen dependent, n=3) and CWR22 xenografts growing in castrated mice (androgen independent, n = 3). Finally, the mean expression was compared between 50 benign prostate specimens, 52 local prostate cancer tumors, 9 metastatic, androgen-independent tumors from bone marrow metastases, and 2 samples of non-prostate cancer containing bone marrow samples 8.

Northern and Western analysis. Fifteen (PSA) or twenty-five (FKBP51) micrograms of total RNA were separated on a 1% denaturing agarose gel and transferred to nylon membranes. Northern probes were amplified from LNCaP cDNA using a previously published PSA- specific primer pair (“P1” and “P2”) 11 and a FKBP51-specific primer pair (sense 5'- GATGAAGGTGCCAAGAACAATG-3'; antisense 5'-CACAGTGAATGCCACATCTCTG- 3'), respectively. Whole cell lysates were separated on a denaturing 7.5% polyacrylamide gel, transferred to PVDF Immobilon-PTM membranes (Millipore, Bedford, MA, USA) and probed with antibodies detecting FKBP51 (murine monoclonal antibodies ("FF1" or “hi51”) and rabbit polyclonal antibodies (“purified51”), obtained from David Smith), AR (PG-21, UpstateBiolabs, Lake Placid, NY, USA), or alpha-tubulin (sc-8035, SantaCruz Biotechnology, Santa Cruz, CA, USA).

Immunohistochemistry. After antigen recovery, the “purified51” (at 1:100) was used to measure FKBP51 expression in prostate specimens removed without pre-surgical treatment (n=25) and those removed after androgen ablation (n=10) 12. Under IRB approval (M.L.), anonymized archival paraffin-embedded tissue blocks of prostate cancer from patients treated during the period from 1991 through 1996 were retrieved from the Departments of Pathology of the Beth Israel Deaconess Medical Center, Boston, and the University of Ancona, Italy for

150 FKBP51 and the androgen receptor

immunohistochemical analysis. Detection was performed using the MultiLink-HRP kit (BioGenex, San Ramon) and standardized 3,3 diaminobenzidine (DAB) development times allowed accurate comparison of all samples. Substitution of the primary antibody with phosphate buffered saline (PBS) served as the negative staining control. An expert in prostate pathology (M.L.), masked as to the identity of the samples, scored FKBP51 staining using a previously described 9-poing system 13. The difference in staining between hormonally intact and hormonally ablated tumors was determined using a two-tailed student’s T-test.

Immunoprecipitation. LNCaP cells were grown in media with charcoal/dextran stripped FBS and harvested in the presence or absence of additional androgen (1 nM R1881). Cells were lysed using ice-cold, low-salt RIPA buffer containing sodium molybdate (10 mM Tris pH 7.4, 50 mM NaCl, 1 mM EDTA, 10 mM Na2MoO4, 0.5% NP40 and 1X protease inhibitors). Cell lysates were pre-cleared with 30 µl of Protein G-plus agarose beads (Santa Cruz Biotech, Santa Cruz, CA) and incubated with either a monoclonal antibody directed against FKBP51 (hi51) or tubulin (Sigma) together with 30 µl of Protein-G plus agarose beads for 3 hours at 4oC with constant rotation. The beads were pelleted by centrifugation, and washed 4 times with 1 mL of ice-cold washing buffer (10 mM Tris pH 7.4, 50 mM NaCl, 10 mM Na2MoO4, 0.5% tween and 1X protease inhibitors). The beads were then re-suspended in 2X loading buffer, separated on a 7.5 % denaturing polyacrylamide gel, transferred to Immobilon-P membranes, and stained for AR as above.

FKBP51 over expression. FuGene 6 reagent (Roche) was used to transfect LNCaP cells with a pC1-neo expression vector containing the full human FKBP51 cDNA (pFKBP51) (provided by Dr. David Smith). Twenty-four LNCaP clones were selected and screened for stable over- expression of FKBP51. PSA protein levels in cell media were measured using a commercial immunofluorescence kit (Tosoh Medics, South San Francisco, CA, USA).

Luciferase reporter assays. A construct of the androgen-responsive MMTV promoter upstream to the firefly luciferase reporter gene was transiently co-transfected with a null- promoter Renilla luciferase plasmid into the LNCaP clones with either increased (n = 3) or wild type (n = 5) expression of FKBP51 with or without exogenous AR (transfected AR

151 Appendix 3

cDNA using the PCDNA3.1 expression plasmid). In triplicate, firefly luciferase activity was activated using a Dual Luciferase kit (Promega, Madison, WI, USA) and measured with a MicroLumat Plus 96 plate reader (EG&G Berthold, Bad Wildbad, Germany). Firefly luciferase activity was normalized by Renilla luciferase activity and the mean normalized values from 3 replicates for each plasmid and media condition were used as the observed value. The co-transfection experiments were repeated three times.

Results Identification of androgen induced genes Genes regulated by the AR in prostate cancer hold potential as therapeutic targets. In order to identify such genes, the androgen sensitive LNCaP cell line was stimulated with R1881 (a synthetic androgen) (Figure 1A). At 0.1 nM of R1881, LNCaP cells demonstrated both optimal proliferation and induction of prostate specific antigen (PSA) mRNA (Figure 1B and C). RNA was isolated from cells collected at 0, 6, 12, and 18 hours following androgen stimulation and processed for hybridization to U95Av2 human microarrays. As expected, androgen induced expression of PSA (a direct target of AR) was detected by the microarrays (Figure 1D). To identify additional androgen-induced genes, microarray features were ranked according to their correlation with PSA. The top 50 PSA-correlated genes are shown (Figure 2). This analysis readily identified two additional genes known to be androgen induced (Kallikrein 2 and spermidine/spermine n1-acetyltransferase) thus demonstrating the ability of this method to identify AR target genes. All array data are available at http://www.broad.mit.edu/MPR/Prostate/FKBP51.

152 FKBP51 and the androgen receptor

Figure 1. Androgen stimulation of LNCaP cells. (A) Schema used for androgen stimulation. (B) Growth curve in media containing charcoal/dextran treated FBS (c/s Media) with or without supplemental androgen (R1881). Error bars represent standard deviation (SD) across 3 experiments. (C) Northern blot for PSA. LNCaP cells were grown in c/s Media with R1881 (0.1 nM) or vehicle (0.001% EtOH). PC3 cells were growing in full media. Methylene blue staining of the 18S ribosomal RNA (18S) serves as a loading control. (D) Microarray detection of PSA expression across two independent R1881 stimulations (#1 and #2) at 0, 6, 12, and 18 hours. Average difference units calculated by MAS4 software. Microarray for stimulation #1, hour 12, was of poor quality and excluded from all analysis.

Refining the list of candidate genes We used data from three independent microarray experiments in order to determine which of the top PSA-correlated genes have 1) consistent androgen-induced expression and 2) expression associated with androgen independent growth (Figure 2). While this approach does not directly validate the microarray-detected changes in RNA expression, it identifies genes with expression changes in independent prostate cancer cell lines and tumors and excludes genes whose androgen-induced expression may be spurious or unique to the LNCaP cells. The expression of the top 50 genes was determined in 3 independent microarray experiments: 1) during dihydrotestosterone (DHT) stimulation of CWR14 cells, an in vitro androgen- sensitive cell line derived from the CWR22 xenograft, 2) in androgen-dependent (n = 3) and androgen-independent (n = 3) CWR22 xenograft tumors, and 3) the mean expression of each gene was determined in 52 local tumors and compared to the mean expression in 9 metastatic, androgen independent tumors 8, and their relative expression is shown (Figure 2). FKBP51 (*) had altered expression in each experiment and was chosen for further investigation.

153 Appendix 3

Figure 2. Expression of the 50 genes best correlating with androgen-induced PSA expression in LNCaP cells. Genes are ranked 1 – 50 according to their correlation with PSA in the LNCaP panel. Gene symbol and U95Av2 microarray accession number (“Probeset”) identify each gene. Mean normalized gene expression within each panel is depicted according to the legend with red signifying high expression and blue signifying low expression. Color intensity is determined by SD distance from mean expression within each panel. The four panels represent: (1) LNCaP panel: gene expression from two androgen stimulations of LNCaP cells (#1 and #2) at hours 0, 6, 12, and 18. The maximum fold change for each gene within each stimulation is provided (FC #1 and FC #2) (2) CWR14 Panel: gene expression following DHT stimulation of CWR14 at hours 0, 14, and 20. The maximum fold change within the experiment is provided (FC), (3) CWR22 Panel: gene expression in three androgen-dependent (AD) and three androgen- independent (AI) CWR22 xenografts. The mean expression in the AI tumors divided by mean expression in the AD tumors (AI/AD) is provided for each gene. (4) Patient Panel: mean gene expression in normal prostate samples (n = 50), local prostate tumors (n = 52), androgen-independent metastatic prostate tumors (“Met”, n = 9), and normal bone marrow samples (“BM”, n = 2). The ratio between the expression in the metastatic tumors and the local tumors (M/L) is provided. FKBP51 is identified (*).

154 FKBP51 and the androgen receptor

Confirmation of androgen regulation of FKBP51 expression in LNCaP cells FKBP51 is an immunophilin with peptidyl-prolyl isomerase activity that participates in the multi-protein chaperone complex that associates with steroid transcription factors prior to ligand binding 14. While the specific function of FKBP51 remains unclear, it interacts with chaperone complexes through a tetratricopeptide repeat domain (TPR) and has been found to exert functional effects on glucocorticoid and progesterone receptors (GR and PR, respectively) 15,16. While FKBP51 has been frequently identified as a target of AR in LNCaP cells 17-19, it remains unknown if FKBP51 interacts with or affects the function of the AR.

In vitro and in vivo androgen regulation of FKBP51 expression Northern and Western blots confirm androgen-induced expression of FKBP51 in LNCaP cells (Figure 3 A and C). To exclude the possibility that FKBP51 induction by androgens represents a non-specific response to mitogenic stimulation, LNCaP cells were stimulated with epidermal growth factor (EGF). While LNCaP cells demonstrated equal cell proliferation with either R1881 or EGF (Fig. 3, B), FKBP51 expression increased only in the presence of androgen (Figure 3C). To determine if FKBP51 expression is regulated by androgens in human tumors, immunohistochemistry for FKBP51 was performed on prostatectomy specimens from men prior to (non-castrate) and following (castrate) androgen withdrawal via castration. FKBP51 protein was expressed in the cytoplasm and nucleus of benign and malignant epithelial cells (Figure 3D). Little to no staining in was observed in basal or stromal cells. Staining was heterogeneous both between tumors and within individual tumor. There was higher expression of FKBP51 in the normal epithelial cells of untreated prostate samples compared to samples collected following castration (p = 0.002, two-tail student’s t-test), supporting androgen regulation of FKBP51 expression in vivo (Figure 3E). As castration results in decreased numbers of secretory epithelial cells in the prostate, decreased expression of FKBP51 could be partially explained by such cell loss. However, decreased FKBP51 staining was evident in the secretory epithelial cells that persisted following castration suggesting that the decreased expression of FKBP51 is due to loss of AR activity and not simply epithelial atrophy.

155 Appendix 3

Figure 3. FKBP51 expression and regulation in prostate cancer. (A) Northern blot for FKBP51 expression in LNCaP cells growing in c/sFBS containing media with androgen (R1881 #2 and R1881 #3) or vehicle (EtOH) at 0, 6, 12, and 18 hours and PC3 cells growing in full media. 18S ribosomal RNA demonstrates RNA loading. (B) Cell proliferation following the addition of R1881 (1 nM), EGF (10 µg/mL), both, or neither to LNCaP cells growing in c/sFBS containing media at days 0, 2 and 4. Error bars represent SD across triplicate plates. *p < 0.05, **p < 0.001, both by student’s t-test. The difference in proliferation between LNCaP cells stimulated with R1881 or EGF was not significant. (C) Western blot for FKBP51 and AR of total cell lysates following the addition of R1881 (1 nM), EGF (10 µg/mL), both or neither at 2 (2d) and 4 (4d) days. Tubulin expression demonstrates protein loading. (D) Immunohistochemistry for FKBP51 (αFKBP51). Representative sections from paraffin-embedded prostate cancer specimens are shown, one from a man castrated prior to prostatectomy (castrate) and one without prior castration (non-castrate). Control sections excluded the primary anti-FKBP51 antibody to assess for non-specific (control). (E) Mean FKBP51 staining intensity in normal (filled bars) and malignant (open bars) prostate epithelium in non-castrate (n = 22) and castrate (n = 10) men. A previously published 9-point system was used to assign a staining intensity to normal and malignant epithelium in each sample (see methods). Error bars represent SD. *p = 0.002 by student’s t-test. The difference in staining between castrate and non-castrate samples within the malignant prostate epithelium was not significant.

156 FKBP51 and the androgen receptor

AR - FKBP51 interaction While FKBP51 is known to bind GR and PR, it was unknown if FKBP51 is part of the AR chaperone complex. Here, FKBP51-specific antibodies co-immunoprecipitate AR (Figure 4A), confirming the association between the proteins. The addition of androgen to cells 3 hours prior to harvesting significantly decreased the amount of AR precipitated. While this does not prove a direct physical interaction, it demonstrates that FKBP51 is part of the chaperone complex interacting with AR prior to ligand binding and could potentially affect the function of the receptor.

Figure 4. FKBP51 associates with and affects the function of AR in LNCaP cells. (A) Western blot for AR (Western:AR) in total cell lysate proteins and immunoprecipitated proteins from LNCaP cells. Anti- FKBP51 (IP:FKBP51) or idiotype-matched anti-tubulin (IP:Tubulin) antibodies were used to immunoprecipitate proteins from the same total protein lysates. IP using αFKBP51 was also performed on cells lysed 3 hours after the addition of androgen (R1881). (B) Western blot for FKBP51 and AR of total cell lysates from LNCaP clones after selection. (C) Mean normalized luciferase activity (see methods) after transient transfection of LNCaP clones over-expressing FKBP51 (n = 3, open bars) and LNCaP clones with normal FKBP51 expression (n = 5, closed bars) with an AR-sensitive reporter plasmid (MMTV-luc) with or without co-transfection with an AR expression plasmid (pAR) and in the presence or absence of androgen (R1881). This experiment was performed three independent times and a representative experiment is presented. Error bars indicate SD. *p = 0.026, **p = 0.012 using a student’s t-test. (D) PSA levels in cell media from FKBP51 over-expressing LNCaP clones (n = 3, open bars) and LNCaP cells with normal FKBP51 expression (n = 5,closed bars) after 4 days of growth in media containing FBS (F), charcoal-dextran stripped FBS (CS), or charcoal-dextran stripped FBS with 0.1 nM R1881 (R1881). Error bars indicate SD. #p = 0.048, ##p = 0.104, ###p = 0.023, by student’s T-test.

157 Appendix 3

FKBP51 modulates AR-mediated transcription In order to assess the functional implications of FKBP51 expression, LNCaP cells were transfected with an FKBP51 expression plasmid and clones with consistently high expression levels of FKBP51 (n = 3) were compared to clones with wild type expression of FKBP51 (n = 5) (Figure 4B). In three independent experiments, clones over-expressing FKBP51 had consistently greater activation of an androgen-responsive MMTV-luciferase reporter compared to clones with normal FKBP51 expression (Figure 4C). These effects were amplified when exogenous wild-type AR was co-transfected into the cells. In addition, clones over-expressing FKBP51 secreted more PSA into culture media when compared to clones with normal FKBP51 expression (Figure 4D). These effects on AR activity, however, did not alter androgen-stimulated proliferation of LNCaP cells (data not shown). Taken together, these findings suggest that increased expression of FKBP51 augments AR-mediated transcription of both an exogenous AR-reporter construct and an endogenous AR-induced gene.

Discussion While the life sustaining role of hormonal ablation in men with metastatic prostate cancer has been known for more than half of a century, the specific target genes down-stream of the AR that are critical to growth remain largely unknown. In this report, we identified gene expression changes following androgen stimulation of LNCaP cells using microarrays. Similar to previous reports 17,18, many genes have reproducible and robust induction following androgen stimulation. We focused on FKBP51 because of its robust androgen induced expression and because FKBP51 belongs to a class of proteins called immunophilins that are part of a multi-protein chaperone complex for steroid receptors 20. Similar to previously published reports 17-19, we confirmed in vitro androgen-induced expression of FKBP51. Additionally, FKBP51 was found to be expressed in the secretory prostate epithelium, PIN, and in some prostate cancer tumors. The decreased expression of FKBP51 in prostate epithelium following androgen ablation suggests in vivo androgen regulation. While epithelial atrophy likely contributed to this decrease, secretory epithelium remaining after castration also had decreased expression. Interestingly, some prostate cancer tumors still expressed FKBP51 following androgen ablation. Continued activation of the AR despite castrate levels

158 FKBP51 and the androgen receptor

of androgen is the most consistent alteration detected in androgen independent prostate cancer 9 and may account for the continued expression of FKBP51 seen here in a few of prostate cancers following castration. The induction of FKBP51 by androgens (but not other mitogens) together with the physical association between FKBP51 and AR suggest a role for FKBP51 in the regulation of AR activity. While the specific functional roles of chaperone proteins such as FKBP51, FKBP52, Cyclophilin 40, and PP5 have yet to be fully described, they likely contribute to protein folding, ligand binding, and nuclear localization 21. Our data suggests that FKBP51 expression in LNCaP cells increases AR mediated transcription and therefore may be part of a feed-forward regulatory loop. FKBP51 may facilitate AR mediated transcription through direct (as with hsp90) or indirect means. For example, FKBP51 over expression has been found to increase STAT5 activity 22, which in turn increases AR-mediated transcription 23. In contrast, FKBP51 over expression decreases the transcriptional activity of glucocorticoid receptors 15 and progesterone receptors 16 and appears to participate in a negative feedback loop. Further experiments are clearly needed to resolve these differences and explore the potential mechanism by which FKBP51 augments AR transcription in LNCaP cells. In conclusion, this work demonstrates that FKBP51 is expressed in the prostate and prostate cancer, is regulated by androgens, and is physically associated with the AR prior to ligand binding. Additionally, over expression of FKBP51 increases AR-mediated transcription and, together, this work suggest that FKBP51 may represent a new therapeutic target to disrupt AR-pathway signaling in prostate cancer.

Acknowledgments The authors would like to thank David Smith for his contribution of the FKBP51 reagents. Additionally, we thank Karoline Nowillo for her technical assistance and Steven Balk and Glenn Bubley for RNA from the CWR22 derived cell lines and tumors and metastatic, androgen-independent prostate cancers.

159 Appendix 3

Reference List

(1) Crawford ED, Eisenberger MA, McLeod DG et al. A controlled trial of leuprolide with and without flutamide in prostatic carcinoma. N Engl J Med. 1989;321:419-424. (2) Huggins C, Hodges CV. Studies on prostatic cancer. I. The effect of castration, of estrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. CA Cancer J Clin. 1972;22:232-240. (3) Huggins C, Hodges CV. Studies on prostatic cancer: I. The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. 1941. J Urol. 2002;168:9-12. (4) Huggins C, Hodges CV. Studies on prostatic cancer. I. The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. 1941. J Urol. 2002;167:948-951. (5) Feldman BJ, Feldman D. The development of androgen-independent prostate cancer. Nat Rev Cancer. 2001;1:34-45. (6) Lubahn DB, Joseph DR, Sullivan PM et al. Cloning of human androgen receptor complementary DNA and localization to the X chromosome. Science. 1988;240:327-330. (7) Linja MJ, Savinainen KJ, Saramaki OR et al. Amplification and overexpression of androgen receptor gene in hormone-refractory prostate cancer. Cancer Res. 2001;61:3550-3555. (8) Taplin ME, Bubley GJ, Shuster TD et al. Mutation of the androgen-receptor gene in metastatic androgen-independent prostate cancer. N Engl J Med. 1995;332:1393-1398. (9) Chen CD, Welsbie DS, Tran C et al. Molecular determinants of resistance to antiandrogen therapy. Nat Med. 2004;10:33-39. (10) Singh D, Febbo PG, Ross K et al. Gene expression correlates of clinical prostate cancer behavior. Cancer Cell. 2002;1:203-209. (11) Seiden MV, Kantoff PW, Krithivas K et al. Detection of circulating tumor cells in men with localized prostate cancer. J Clin Oncol. 1994;12:2634-2639. (12) Signoretti S, Waltregny D, Dilks J et al. p63 is a prostate basal cell marker and is required for prostate development. Am J Pathol. 2000;157:1769-1775. (13) Gu Z, Thomas G, Yamashiro J et al. Prostate stem cell antigen (PSCA) expression increases with high gleason score, advanced stage and bone metastasis in prostate cancer. Oncogene. 2000;19:1288-1296. (14) Smith DF, Albers MW, Schreiber SL, Leach KL, Deibel MR, Jr. FKBP54, a novel FK506-binding protein in avian progesterone receptor complexes and HeLa extracts. J Biol Chem. 1993;268:24270- 24273. (15) Denny WB, Valentine DL, Reynolds PD, Smith DF, Scammell JG. Squirrel monkey immunophilin FKBP51 is a potent inhibitor of glucocorticoid receptor binding. Endocrinology. 2000;141:4107-4113. (16) Hubler TR, Denny WB, Valentine DL et al. The FK506-binding immunophilin FKBP51 is transcriptionally regulated by progestin and attenuates progestin responsiveness. Endocrinology. 2003;144:2380-2387. (17) Nelson PS, Clegg N, Arnold H et al. The program of androgen-responsive genes in neoplastic prostate epithelium. Proc Natl Acad Sci U S A. 2002;99:11890-11895. (18) Velasco AM, Gillis KA, Li Y et al. Identification and validation of novel androgen-regulated genes in prostate cancer. Endocrinology. 2004;145:3913-3924. (19) Zhu W, Zhang JS, Young CY. Silymarin inhibits function of the androgen receptor by reducing nuclear localization of the receptor in the prostate cancer cell line LNCaP. Carcinogenesis. 2001;22:1399-1403. (20) Smith DF, Albers MW, Schreiber SL, Leach KL, Deibel MR, Jr. FKBP54, a novel FK506-binding protein in avian progesterone receptor complexes and HeLa extracts. J Biol Chem. 1993;268:24270- 24273. (21) Ratajczak T, Ward BK, Minchin RF. Immunophilin chaperones in steroid receptor signalling. Curr Top Med Chem. 2003;3:1348-1357. (22) Komura E, Chagraoui H, Mansat dM, V et al. Spontaneous STAT5 activation induces growth factor independence in idiopathic myelofibrosis: possible relationship with FKBP51 overexpression. Exp Hematol. 2003;31:622-630. (23) Carsol JL, Gingras S, Simard J. Synergistic action of prolactin (PRL) and androgen on PRL-inducible protein gene expression in human breast cancer cells: a unique model for functional cooperation between STAT-5 and androgen receptor. Mol Endocrinol. 2002;16:1696-1710.

160 Abbreviations

Appendix 4: Abbreviations

Ab antibody AMPK AMP-activated protein kinase AP-1 activator protein-1 5-ASA 5-aminosalicylic acid ASK apoptosis signal-regulating kinase ATF-2, activating transcription factor 2 ATP adenosine triphosphate CBA cytokine bead array CD Crohn’s disease CDAI Crohn’s disease activity index CRP c-reactive protein; DC dendritic cell DEX dexamethasone DMEM dulbecco's modified eagle medium DMSO dimethyl sulfoxide EGF epidermal growth factor ERK extra-cellular signal-regulated kinase FACS fluorescence-activated cell sorter FCS fetal calf serum FITC fluorescein isothiocyanate FKBP FK506 binding protein GC glucocorticoid GR glucocorticoid receptor GRE glucocorticoid receptor responsive element GSK glycogen-synthase kinase HIF hypoxia inducible factor Hop HSP organizer protein HRP horseradish peroxidase HSP heat shock protein IBD inflammatory bowel diseases ICAM intracellular adhesion molecule IFN interferon IGF insulin-like growth factor IKK IκB kinase IL interleukin IMDM modified dulbecco’s medium INSR insulin receptor IRS insulin receptor substrate ITAM immunoreceptor tyrosine-based activation motif JAK janus kinase JNK jun N-terminal kinase LAT linker for activation of T cells LPMC lamina propria mononuclear cell LPS lipopolysaccharide MAP mitogen activated protein

161 Appendix 4

MAPK mitogen activated protein kinase MAPKAP MAP kinase activated protein kinase MAPKK MAPK kinase MAPKKK MAPK kinase kinase MEF myocyte enhance factor MEK MAP/ERK kinase MEKK MEK kinase MKK MAP kinase kinase MLK mixed lineage kinase MNK MAPK interacting kinases MSK mitogen and stress activated protein kinases MTT 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide NFAT nuclear factor of activated T cells NFkB nuclear factor kappa B NOS nitric oxide synthase PAK p21-activated kinase PBMC peripheral blood mononuclear cell PBS phosphate-buffered saline PDK phosphoinositide-dependent protein kinase PE phosphatidyletholamine PI3K phosphatidylinositol 3 kinase PKB protein kinase B PKC protein kinase C PLC phospolipase C PPAR peroxisome proliferator-activated receptor p70S6k p70 ribosomal S6 kinase PVDF polyvinylidene difluoride membrane SAM68 Src-Associated in Mitosis 68 SOCS suppressors of cytokine signaling STAT signal transducers and activators of transcription TAB transforming growth factor-beta-activated protein kinase TCR T-cell receptor TGF transforming growth factor TLR toll-like receptor TNF tumor-necrosis factor TRITC tetramethylrhodamine isothiocyanate UC ulcerative colitis

162 Dankwoord

Appendix 5: Dankwoord

Graag wil ik iedereen bedanken die een bijdrage heeft geleverd aan het tot stand komen van dit proefschrift. Tot een aantal personen wil ik mij persoonlijk richten.

Allereerst, mijn promotores (Sander van Deventer en Maikel Peppelenbosch) en mijn co- promotor (Daan Hommes). Sander: je enorme kennis op het gebied van de immunologie heeft mij zeer geïnspireerd. Maikel: bedankt voor je vele enthousiaste, creatieve en soms onnavolgbare ideeën. Daan: dank voor de bijzondere kansen die je mij de afgelopen jaren hebt geboden.

De leden van mijn promotie commissie bedank ik voor de bereidheid dit proefschrift op haar wetenschappelijke waarde te beoordelen.

Mijn collega’s van de maag-darm-lever afdeling en het laboratorium voor experimentele interne geneeskunde van het AMC bedank ik en natuurlijk Meike en Joyce die mij vele maanden geholpen hebben in het lab in het kader van hun wetenschappelijke stages. Tot slot, Auke: dank voor je hulp, al ons overleg en bovenal de prettige manier van samenwerken.

163 Appendix 6

Appendix 6: Curriculum Vitae (Dutch)

Mark Löwenberg werd geboren op 4 januari 1977 in Rotterdam. Na het behalen van het Atheneum diploma in 1995 aan het Rotterdams Montessori Lyceum begon hij aan de studie geneeskunde aan de Erasmus Universiteit in Rotterdam. Na zijn propedeuse in 1996 gehaald te hebben, werd de studie geneeskunde voortgezet aan de Universiteit van Amsterdam. In de doctoraalfase van de studie heeft hij in het jaar 2000 een onderzoekstage gedaan aan het Dana Farber Cancer Institute (Harvard Medical School) in Boston. Onder leiding van prof. dr. T.R. Golub en dr. P.G. Febbo richtte het onderzoek zich op het bestuderen van androgeen receptor- gemedieerde gentranscriptie en signaal transductie in prostaatkanker. In hetzelfde jaar deed hij een tweede onderzoekstage bij het biotechnologie bedrijf Crucell te Leiden (onderwerp: testen van de stabiliteit van virale vectoren voor gentherapie). In 2003 studeerde Mark af, waarna hij in april 2003 als artsonderzoeker begon aan zijn promotie-onderzoek bij de afdeling maag-darm-leverziekten (laboratorium voor experimentele interne geneeskunde). Onder leiding van prof. dr. S.J.H. van Deventer, prof. dr. M.P. Peppelenbosch en dr. D.W. Hommes heeft hij basaal en translationeel immunologisch onderzoek gedaan naar de interacties tussen therapeutische interventies en cellulaire signaal transductie in inflammatoire darmziekten; deze resultaten worden in dit proefschrift beschreven. In oktober 2006 start hij met de vooropleiding interne geneeskunde in het Onze Lieve Vrouwe Gasthuis in Amsterdam in het kader van de opleiding tot maag-darm-leverarts in het Academisch Medisch Centrum te Amsterdam.

164 Curriculum Vitae

Curriculum Vitae (English)

Mark Löwenberg was born on January 4, 1977 in Rotterdam. He graduated from highschool in 1995 at the Montessori Lyceum in Rotterdam. In 1995 he started medical school at the Erasmus University in Rotterdam. One year later, he passed his propaedeutics exam and moved to the University of Amsterdam to continue medical school. In 2000, he did a research internship at the Dana Farber Cancer Institute (Harvard Medical School) in Boston. He studied androgen receptor-mediated transcriptional and signaling processes in prostate cancer in the laboratory of prof. T.R. Golub, under the guidance of dr. P.G. Febbo. In the same year, he conducted another research project under supervision of dr. M. Havenga at the biotechnology company Crucell in Leiden were he tested the stability of viral vectors for gene therapy. In 2003, Mark passed his doctoral exam in Amsterdam and started his PhD project at the department of gastroenterology and hepatology (laboratory for experimental internal medicine) in the Academic Medical Center. Under supervision of prof. dr. S.J.H. van Deventer, prof. dr. M.P. Peppelenbosch and dr. D.W. Hommes he acquired skills in fundamental and translational immunological research. He studied the interactions between therapeutic interventions and cellular signal transduction in physiology and inflammatory bowel diseases; the results of these studies are described in this thesis. In October 2006 he will start his training to become a gastroenterologist. The first two years he will be working at the department of internal medicine in the Onze Lieve Vrouwe Gasthuis in Amsterdam. In October 2008, he will continue his clinical round at the department of gastroenterology and hepatology in the Academic Medical Center in Amsterdam.

165 Appendix 7

Appendix 7: Bibliography

Original Articles: P Febbo, M Löwenberg, A Thorner, M Brown, M Loda, T Golub. Androgen mediated regulation and functional implications of FKBP51 expression in prostate cancer. J Urol. 2005; 173(5):1772-7.

M Löwenberg, J Tuynman, J Bilderbeek, T Gaber, F Buttgereit, S van Deventer, M Peppelenbosch, D Hommes. Rapid immunosuppressive effects of glucocorticoids mediated through Lck and Fyn. Blood. 2005; 106(5):1703-10.

M Löwenberg, A Verhaar, B van den Blink, F ten Kate, S van Deventer, M Peppelenbosch, D Hommes. Specific inhibition of c-Raf activity by semapimod induces clinical remission in severe Crohn's disease. J Immunol. 2005; 175(4):2293-300.

M Löwenberg, J Tuynman, A Verhaar, M Scheffer, L Vermeulen, S van Deventer, D Hommes, M Peppelenbosch. Kinome analysis reveals nongenomic glucocorticoid receptor dependent inhibition of insulin signaling. Endocrinology. 2006; 147(7):3555-62

M Löwenberg, M Scheffer, A Verhaar, M Peppelenbosch, D Hommes. A role for STAT5 in steroid resistant ulcerative colitis? Inflamm Bowel Dis. 2006; 12(7):665

M Löwenberg, J Bilderbeek, A Verhaar, J van Marle, F Buttgereit, M Peppelenbosch, S van Deventer, D Hommes. Glucocorticoids cause rapid dissociation of a T cell receptor-associated protein complex containing Lck and Fyn. EMBO reports. Accepted for publication.

J Tuynman, M Löwenberg, L Vermeulen, H Verdurmen, S Diks, A Zwinderman, A Pandey, D Richel, M Peppelenbosch. Third generation kinomic profiling identifies Met dependent regulation of the Wnt pathway in APC mutation harboring colorectal cancer. Manuscript in preparation

K Parikh, J Tuyman, A Verhaar, M Löwenberg, D Hommes, J Joore, S van Deventer, M Peppelenbosch. Comparison of peptide array substrate phosphorylation of c-Raf and MAP3K8. Manuscript in preparation

Review Articles: M Löwenberg, M Peppelenbosch, D Hommes. Therapeutic modulation of signal transduction pathways. Inflamm Bowel Dis. 2004;10 Suppl 1:S52-7.

M Löwenberg, D Hommes. Glucocorticoid action and molecular mechanisms of steroid resistance: from molecules to patients. BTi. 2004; 16(4):6-9

M Löwenberg, M Peppelenbosch, D Hommes. Biological therapy in the management of recent-onset Crohn's disease: why, when and how? Drugs. Accepted for publication.

166