Molecular Mechanisms of Host Cell Response to Francisella Infection

MOLECULAR MECHANISMS OF HOST CELL RESPONSE

TO Francisella INFECTION

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of

Philosophy in the Graduate School of The Ohio State University

Laxmi Kishore Parsa Venkata, M.S.

********

The Ohio State University 2007

Dissertation committee

Susheela Tridandapani, Ph.D., Advisor Approved by

Clay B. Marsh, M.D.

Mark D. Wewers, M.D. Advisor

John S. Gunn, Ph.D. Ohio State Biochemistry Graduate Program

ABSTRACT

Francisella tularensis is a Gram-negative intra-cellular pathogen causing the

zoonotic disease tularemia. The mechanisms of host response to infection are poorly

understood. Francisella tularensis is considered a potential bio-weapon. Thus, currently

there is an increased focus to understand the regulatory mechanisms controlling the host

response and the strategies employed by the pathogen to evade the host-mounted immune

response. In this study, we examined the molecular mechanisms of i) phagocytosis of

Francisella ii) inflammatory response to infection and iii) subversion mechanisms

employed by Francisella against host IFNγ response.

In the first part of this dissertation we have identified the intracellular tyrosine

kinase Syk as a critical player in the engulfment of Francisella and the subsequent

cytokine release. Specifically, we established that Syk promotes Francisella phagocytosis and the ensuing cytokine response via the activation of Erk and PI3K/Akt pathways.

Unchecked cytokine production is deleterious to the host thus emphasizing the importance of negative regulators.

Therefore, in the second part of this dissertation, we focused on identifying

negative regulators of host cell response. These studies revealed that the inositol

phosphatase SHIP1 although dispensable for Francisella phagocytosis, is critical for

regulating pro- and anti-inflammatory mediator release in response to infection.

Molecular analyses of SHIP1-dependent cytokine regulation demonstrated that SHIP1 suppresses the activation of PI3K/Akt/NFκB signaling cascade. Cytokines produced during infection confer protection against infection. One major mechanism of cytokine- mediated host protection is through the induction NK cell and T cell IFNγ production, which in turn protects against infection by limiting phagosomal escape of the organisms.

However, some pathogens are known to subvert this host-protective response.

In the third part of this project, we examined whether Francisella subverts IFNγ signaling response. Here, we established that Francisella suppresses IFNγ-induced host response through the upregulation of a negative regulator, SOCS3. Functional analysis revealed that Francisella infection inhibits IFNγ-induced iNOS, a critical anti-microbial enzyme, leading to the enhanced intra-macrophage survival of the bacteria.

Collectively, these studies unravel signaling pathways that modulate host response against Francisella infection and identify potential targets for future therapeutic interventions.

iii

Dedicated to my dearest family

ACKNOWLEDGMENTS

I would like to use this opportunity to acknowledge the assistance that I have received during the course of my research and academic study.

I am greatly indebted to my mentor Dr. Susheela Tridandapani, Associate

Professor, department of Internal Medicine for providing excellent training, guidance, motivation and tireless help during this study. I also thank her sustained interest in assisting to improve my presentation and writing skills.

I am thankful to my advisory committee members, Drs Clay Marsh, Mark

Wewers and John Gunn, for providing me constructive suggestions, insights and critical reagents during the entire course of this investigation. I also thank them for motivating me to visualize the “the big picture’’ of this study. I am also grateful to Dr. Larry

Schlesinger, director of CMIB, for his valuable input, comments and also for providing critical reagents.

I also thank all the help that I have received from Dr. Mikhail Gavrilin when I started on this project. I am grateful to Dr. Melissa Hunter for providing antibodies. I also acknowledge the priceless efforts of my lab colleagues, both past and present. I am thankful to Anne-Sophie Wavreille, Donna Cain, Huiqing Fang, Jonathan P Butchar,

Latha P Ganesan, Murugesan Rajaram, Payal Mehta, Ruma Pengal, Thomas Cremer and

Trupti Joshi. Special thanks to Latha P Ganesan-for helping me with different techniques v

when I started in the lab, Murugesan Rajaram, Jonathan P Butchar and Thomas Cremer- for providing technical assistance with tedious experiments.

I also thank my other colleagues, Ashwin Balagopal, Nrusingh Mohapatra, Shilpa

Soni and Bret Betz for their technical assistance. I acknowledge all the assistance that I have received from Luke Davis with blood withdrawals.

I also thank my cousin, Suman, who helped me when I needed the most especially when I injured my toes. I also like to express my gratitude to my undergrad friends,

Vamsi, Narendranath, and Sambasivarao for providing technical inputs into the study.

This note of acknowledgements would carry no value if I failed to confess the help and assistance that I have received from my family. I strongly credit my success to my wife, Suneeta, my Son, Keshav Nimai and my parents-Sri. P.V.S. Rama sarma and

Smt. P.V.S. Usha. My wife was there with me all the time and extended her moral support through out our wedded life including this graduate study. I am also grateful to my parents who have put many efforts into my education. Thank you, you both strongly deserve my gratitude. I take this opportunity to express gratitude to my brothers-

Rajasekhar and Ravisankar, Sisters-in-law-Sujatha and Sindhu. I wish for the success of their kids-Nikhil, Vineel, Vaibhav and Vainavi.

I also acknowledge the help from the administrative staff of Davis Heart and

Lung Research Institute, especially Pulmonary division for all the organizational help. I deeply thank grants from National Institute of Health and OSBP program fellowship for providing me financial support during this study.

Laxmi Kishore Parsa Venkata

VITA

August 31st 1975 ……….…… Born, Vijayawada, India

2001 ………………………… M.VSc., A.N.G.R.A.U., India

2001-2004 ………….……….. Research Assistant, Texas A&M University-Kingsville

2003 …………………………. Teaching Assistant, Texas A&M University-Kingsville

2004 ……………………….... M.S., Texas A&M University-Kingsville

2004 ……...... Program Fellow, The Ohio State University

2005-Present ……………...… Graduate Research Associate, The Ohio State University

PUBLICATIONS

1. P.V.L. Kishore, G. Narasimha Rao, R.P. Sharma, N.K. Praharaj, B. Ramesh Gupta, A. Satyanarayana. Inheritance of body weights in synthetic broiler chickens, Indian J. Poult. Sci. (2002) 37(2): 175-178

2. P.V.L. Kishore, G. Narasimha Rao, R.P. Sharma, N.K. Praharaj, B. Ramesh Gupta, A. Satyanarayana. Inheritance of abdominal fat and serum lipids in broiler chickens. Indian J. of Anim Sci. 2002.

3. Vancha AR, Govindaraju S, Parsa KV, Gonzalez-Garcia M, Ballestero R.P. Use of polyethyleneimine polymer in cell culture as attachment factor and lipofection enhancer, BMC Biotechnol. 2004 Oct 15;4(1):23

4. Chintharlapalli SR, Jasti M, Malladi S, Parsa KV, Ballestero R.P. and Gonzalez- Garcia, M. BMRP is a Bcl-2 binding protein that induces apoptosis. J. Cell. Biochem 2005 Feb 15;94(3):611-26.

5. Parsa KV, Ganesan LP, Rajaram MV et al. Macrophage pro-inflammatory response to Francisella novicida infection is regulated by SHIP. PLoS Pathog. 2006;2:e71 vii

6. Rajaram MV, Ganesan LP, Parsa KV et al. Akt/Protein kinase B modulates macrophage inflammatory response to Francisella infection and confers a survival advantage in mice. J Immunol. 2006;177:6317-6324

7. Butchar JP, Parsa KV, Marsh CB and Tridandapani S. (2006) Negative regulators of Toll-like receptor 4-mediated macrophage inflammatory response. Current Pharmaceutical Design. 2006;12(32):4143-53

8. Butchar JP, Rajaram MV, Ganesan LP, Parsa KV, Clay CD, Schlesinger, LS, Tridandapani, S. Francisella tularensis induces IL-23 production in human monocytes. J Immunol. 2007;178:4445-4454.

FIELDS OF STUDY

Major Field: Ohio State Biochemistry Program

viii

TABLE OF CONTENTS

Page

Abstract…………………………………………………………………….…………….ii

Dedication………………………………………………………………………………..iv

Acknowledgments………………………………………………………………………..v

Vita………………………………………………………………………………………vii

List of Figures…………………………………..………………………………………xii

List of Abbreviations………………………………………………………………….xvii

Chapters:

1. Introduction

1.1 Macrophages and Innate Immunity……………………………...……………2

1.2 MAPKs and Innate Immunity…………………………………………………4

1.3 PI3K/Akt Pathway and Innate Immunity……………………………………...6

1.4 Src and Syk Kinases………………………………………………………….10

1.5 Inositol Phosphatases………………………………………………………...11

1.6 Classical Activation of Macrophages-Potent Innate Immune Response…….15

1.7 Tularemia…………………………………………………………………….17

1.8 Intra-cellular Lifestyle of Francisella tularensis…………………………….18

1.9 Francisella: Host Cell Response and Innate Immunity……………………...18

2. The tyrosine kinase Syk promotes Francisella phagocytosis and cytokine response to infection

2.1 Abstract………………………………………………………………………23

2.2 Introduction…………………………………………………………………..24

2.3 Materials and Methods……………………………………………………….26

2.4 Results………………………………………………………………………..30

2.5 Discussion……………………………………………………………………56

3. Macrophage inflammatory response to Francisella novicida infection is regulated by the SH2 domain-containing inositol 5’ phosphatase SHIP1

3.1 Abstract………………………………………………………………………63

3.2 Introduction…………………………………………………………………..64

3.3 Materials and Methods……………………………………………………….66

3.4 Results………………………………………………………………………..69

3.5 Discussion……………………………………………………………………91

4. Francisella gains a survival advantage within mononuclear phagocytes by suppressing host IFNγ response

4.1 Abstract…………………………..………………………………………..…97

4.2 Introduction………………..………………………………………………....98

4.3 Materials and Methods………………………………………………...……101

4.4 Results………………………..……………………………………….…….105

4.5 Discussion………………………………………..…………………………134

5. Summary and future perspectives...... 139

Bibliography……………………………………………………………………………151

LIST OF FIGURES

Figures Page

Chapter 1

1.1 Schematic representation of TLRs………………………………………………...3

1.2 Schematic representation of macrophage signaling and its role

in innate immunity………………………………………………………………..7

1.3 Domain structures of inositol phosphatases……………………………………...13

1.4 Intra-cellular lifestyle of Francisella…………………………………………….19

Chapter 2

2.1 Francisella induces tyrosine phosphorylation of proteins……………………….31

2.2 Francisella induces tyrosine phosphorylation of Syk…………………………...34

2.3 Syk inhibition suppresses the phagocytosis of Francisella……………………...36

2.4 Syk inhibition suppresses the phagocytosis of Francisella……………………...37

2.5 Syk overexpression enhances the phagocytosis of Francisella………………....38

2.6 Syk is required for the Francisella-induced phosphorylation

of Erk and Akt……………………………………………………………………40

2.7 Overexpression of Syk enhances the Francisella-induced

phosphorylation of Erk and Akt…….…………………………………………...42

xii

2.8 Erk but not PI3K/Akt pathway is required for the phagocytosis

of Francisella…………………………………………………………………….43

2.9 Erk but not PI3K/Akt pathway is required for the phagocytosis

of Francisella………………………………………………………………….....45

2.10 Erk overexpression enhances the phagocytosis of Francisella………………….46

2.11 Syk-dependent increase in the phagocytosis of Francisella is abrogated

by Erk inhibition………………………………………………………………....47

2.12 Erk-dependent increase in the phagocytosis of Francisella is abrogated

by Syk inhibition…………………………………………………………………49

2.13 Inhibition of Syk suppresses Francisella-induced TNFα and IL-10………….…51

2.14 Overexpression of Syk enhances Francisella-induced TNFα and IL-10………..52

2.15 Erk and PI3K/Akt pathways differentially regulate Francisella-induced

TNFα and IL-10………………………………………………………………….53

2.16 Syk, Erk, PI3K/Akt pathways promote Francisella-induced NFκB activation…55

2.17 NFκB differentially regulates Francisella-induced TNFα and IL-10…………...57

2.18 Proposed model for the Syk-dependent regulation of Francisella-induced

macrophage response…………………………………………………………….58

Chapter 3

3.1 SHIP1 is phosphorylated during F. novicida infection…………………………..71

3.2 SHIP1 is phosphorylated during Francisella infection………………………….72

3.3 SHIP1 is dispensable for the phagocytosis of F. novicida……………………….74

3.4 Representative images of TEM analysis…………………………………………75 xiii

3.5 SHIP1 downregulates macrophage pro-inflammatory response

to F. novicida infection………………………………………………………….77

3.6 SHIP1 downregulates macrophage pro-inflammatory response

to F. novicida independent of MOI………………………………………………78

3.7 Analysis of SHIP1 influence on macrophage inflammatory

response to F. novicida infection………………………………………………...79

3.8 Activation of signaling pathways during F. novicida infection………………….82

3.9 SHIP1 negatively regulates F. novicida-induced Akt activation………………...83

3.10 PI3K activation promotes F. novicida induced macrophage

inflammatory response…………………………………………………………...84

3.11 NFκB activation is required for the production of F. novicida-induced

IL-12, IL-6 and RANTES………………………………………………………..86

3.12 Surface contact may be sufficient to trigger F. novicida-induced

NFκB activation………………………………………………………………….87

3.13 SHIP1 overexpression downregulates F. novicida-induced NFκB activation…..89

3.14 Knockout of SHIP1 enhances F. novicida-induced NFκB activation…………...90

3.15 SHIP1 positively regulates F. novicida-induced IL-10………………………….92

3.16 Proposed model for SHIP1-dependent negative regulation of

F. novicida-induced IL-12, IL-6 and RANTES………………………………….93

Chapter 4

4.1 Schematic representation of IFNγ-mediated signaling pathway………………..100

4.2 Francisella suppresses IFNγ-induced STAT1 phosphorylation………………..107 xiv

4.3 Francisella suppresses IFNγ-induced STAT1 phosphorylation………………..108

4.4 Francisella-mediated suppression of IFNγ-induced STAT1

phosphorylation is not unique…………………………………………………..109

4.5 Phagosomal escape or the replication of bacteria is dispensable for

Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation…112

4.6 Viability or phagocytosis of Francisella is dispensable for

Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation…113

4.7 Francisella does not predominantly suppress positive mediators

of IFNγ signaling……………………………………………………………….115

4.8 Francisella induces the expression of SOCS3………………………………….117

4.9 Phagosomal escape or the replication of bacteria is dispensable for

Francisella-mediated induction of SOCS3……………………………………..118

4.10 Viability or phagocytosis of Francisella is dispensable for

Francisella-mediated induction of SOCS3……………………………………..119

4.11 Francisella suppresses IFNγ-induced STAT1 phosphorylation

through SOCS3…………………………………………………………………120

4.12 Francisella suppresses IFNγ-induced STAT1 phosphorylation

in human cells…………………………………………………………………..122

4.13 Phagosomal escape or the replication of bacteria is dispensable for

Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation…123

4.14 Francisella induces the expression of SOCS3 in human cells…………………124

4.15 Francisella suppresses IFNγ-induced IP-10 production………………………..126

4.16 Phagosomal escape, replication and viability of Francisella are

dispensable for the suppression of IFNγ-induced IP-10 production……………127

4.17 Co-stimulation of macrophages with Francisella and IFNγ induces

iNOS/nitric oxide……………………………………………………………....128

4.18 Francisella suppresses IFNγ-induced iNOS and IP-10………………………...130

4.19 Francisella opposes IFNγ-induced bacterial death……………………………..132

4.20 Proposed model of Francisella-mediated interference of IFNγ

signaling response……………………………………………………………...133

Chapter 5

5.1 Proposed model for the regulation of macrophage response to Francisella

infection………………………………………………….…………………...... 141

xvi

LIST OF ABBREVIATIONS

AP-1 activator protein-1 BMM bone marrow derived macrophages Btk Brutons’s tyrosine kinase CR3 complement receptor 3 DMSO dimethyl sulfoxide ELISA Enzyme-Linke Immunosorbent Assay Erk Extracellular signal-regulated kinase FcγR Fc gamma receptor Grb2 growth factor receptor-bound protein 2 GSK3β glycogen synthase kinase 3 beta GST glutathione S-transferase IFNβ interferon beta IFNγ interferon gamma IgG Immunoglobulin G IL-1β interleukin 1-beta IL-4 interleukin 4 IL-6 interleukin 6 IL-10 interleukin 10 IL-12 interleukin 12 iNOS inducible nitric oxide synthase ITAM immunoreceptor tyrosine-based activation motif ITIM immunoreceptor tyrosine-based inhibitory motif JAK Janus Associated Kinase JNK c-jun N-terminal kinase LPS lipopolysaccharide xvii

LRR leucine rich repeats LSP-1 leukocyte specific protein-1 MAPK mitogen activated protein kinase MOI multiplicity of infection Myr-Akt myristoylated Akt NADPH nicotinamide adenine dinucleotide phosphate NFAT nuclear factor of T cells NFκB nuclear factor kappa B NK natural killer NO nitric oxide PAMPs pathogen-associated molecular patterns PBMs peripheral blood monocytes PKCδ protein kinase C delta PRRs pattern recognition receptors PRD proline-rich domains PI3K Phosphatidylinositol 3-kinase

PtdIns3,4P2 Phosphatidylinositol 3,4 bisphosphate

PtdIns3,4,5P3 Phosphatidylinositol 3,4,5 trisphosphate PTB phosphotyrosine binding domains PTEN phosphatase and tensin homologue deleted on chromosome 10 PX phox homology RNAi RNA interference SDS-PAGE Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis SOCS Suppressor Of Cytokine Signaling Syk spleen tyrosine kinase ROS reactive oxygen species RNS reactive nitrogen species SH2 Src homology 2 SHIP1 SH2 domain-containing inositol 5’-phosphatase1 SHIP2 SH2 domain-containing inositol 5’-phosphatase2

xviii

SHP-1 SH2 domain-containing phosphatase-1 siRNA small interfering RNA Sos Son of sevenless SR scavenger receptor ΤΙR Toll-like/IL-1 receptor TLR toll-like receptor ΤΝFα tumor necrosis factor alpha

xix

CHAPTER 1

INTRODUCTION

In this chapter, an introduction of innate immune responses of the professional phagocytic cells is outlined, followed by a review of select signaling molecules that modulate these innate immune responses. The chapter will conclude with a brief description of the broad objectives and the critical findings of this dissertation.

Multi-cellular organisms are constantly exposed to infectious agents. Vertebrates have developed efficient mechanisms to combat these insults. Such effector mechanisms can be broadly categorized into two groups1: innate immunity and adaptive immunity

These two groups differ in many aspects. Innate immunity exists from birth where as adaptive immunity is acquired during life. Innate immunity comprises of rapid responses with limited diversity and no memory. In contrast, initial adaptive immune responses are delayed but they possess the ability to remember and respond more vigorously to repeated infections (of the same microbe)1.

The chief components of innate immunity are: a) protective physical and chemical barriers such as mucosal epithelia, anti-microbial peptides etc. b) blood proteins such as the complement c) cells such as macrophages and natural killer (NK) cells1. The Innate

immune system recognizes pathogens via germline-encoded pattern recognition receptors

(PRRs). There are several PRRs such as toll like receptors (TLRs), scavenger receptors,

complement receptors etc.TLRs are the most widely studied receptors. To date, 12 TLR

members have been identified in mammals2. These receptors have an extracellular

domain consisting of varying number of leucine rich repeat (LRR) motifs where as the

intracellular domain is homologous to that of IL-1 receptor (termed as Toll/IL-1R or TIR

domain). These receptors recognize several bacterial, viral, fungal and parasitic

components initiating the immune response. A schematic representation of selected TLRs

and their cognate ligands is shown in the Figure 1.1. The other PRRs are mainly

implicated in the phagocytosis of the pathogen3.

1.1 Macrophages and Innate Immunity

The innate immune responses are mainly mediated by cells such as macrophages

and NK cells. Eli Metchnikoff first described macrophages as phagocytic cells-cells

capable of ingesting and destroying potential pathogens4. Macrophages are terminally differentiated monocytes and have a long life span in tissues as resident macrophages.

Such resident macrophages are described by different names in different organs5. For example resident macrophages are referred to as kupffer cells in liver, alveolar macrophages in lungs and microglial cells in the central nervous system.

Macrophages contribute to innate immunity by several ways. These cells mediate

different effector functions such as phagocytosis, production of anti-microbial toxic metabolites, cytokines and chemokines. Macrophages are equipped with several surface receptors that aid in the recognition and phagocytosis of different microbial pathogens3.

Ligand LPS Mycoplasmal Bacterial dsRNA Flagellin Imidazoquinolines Lipopeptide Lipopolypeptide

CD14

MD2

TLR 4 2 6 2 1 3 5 7

Figure 1.1: Schematic representation of TLRs. All the TLRs possess an extracellular LRR, short transmembrane region and an internal TIR domain. TLR2 can heterodimerize with either TLR-6 or -1. TLR4 is associated with the co-receptor CD14. The cognate ligands for different TLRs are also depicted in the figure.

During this process of engulfment, the pseudopodia extend around the pathogen

and this is accompanied by actin cytoskeletal rearrangements leading to the formation of

the phagosome. After internalization, the phagosome fuses with lysosome and the

pathogen is destroyed by the acidic milieu and hydrolytic enzymes present in the

lysosomes3. This process of phagocytosis is usually accompanied by the production of

reactive oxygen species (ROS) and reactive nitrogen species (RNS) which possess anti-

microbial properties.

In addition, macrophages produce a variety of cytokines such as TNFα, IL-12, IL-

6 etc. which help in the elimination of pathogens and other host defense mechanisms3.

The macrophage effector functions are mediated by the activation of different signaling cascades such as the mitogen-activated protein kinase (MAPK) pathway, phosphatidylinositol 3 kinase (PI3K)/Akt pathway etc. These pathways relay the signal for the activation of transcription factors such as NFκB, AP-1 and NFAT leading to cytokine and chemokine gene transcription3.

In the following sections, a brief description of the contribution of MAPK and

PI3K/Akt pathways to innate immune responses and the regulatory mechanisms that

modulate these responses is presented.

1.2 MAPKs and Innate Immunity

MAPK signaling pathway is the one of the most ancient and evolutionarily

conserved pathway with a widely accepted role in both innate and adaptive immunity6.

Three main families of mammalian MAPKs exist: extracellular signal-related protein 4

kinases (Erk), p38 MAPK and c-Jun NH2 terminal kinases (JNK). MAPKs are activated by diverse stimuli. MAPKs are active when phosphorylated at both threonine and tyrosine residues of Thre-X-Tyr motif present in their activation loop6. The activity of

MAPKs is regulated by kinases that phosphorylate MAPKs (MAP kinase kinase; MKK) and by phosphatases that dephosphorylate MAPK. A schematic model describing the role

of MAPKs in innate immune responses is shown in Figure 1.2. MAPKs are activated by

different stimuli and once activated they mediate several effector functions such as

phagocytosis, cytokine production, Type I interferon generation, growth, development,

apoptosis, differentiation etc. suggesting that they possess a vital role in the able

functioning of the host system. A brief description of the role of MAPKs in inflammation is presented in the following paragraphs.

The role of MAPK in innate immunity has been emphasized by many studies.

Engagement of receptors such as TLRs or immune receptors results in MAPK-dependent

activation of transcription factors such as NFκB, AP-1 etc. leading to the transcription of

inflammatory genes. Thus, the ablation of MAPKs activation results in compromised

generation of inflammatory mediators. Few examples that highlight such defects are

described below6. First, mice that are deficient in Tp12/Cot, a MAP kinase kinase kinase

(MKKK) showed selective Erk activation defects and are compromised in the production

of lipopolysaccharide-induced (LPS-induced) TNFα because of a defect in the transport

of TNFα mRNA into the cytoplasm. Second, inhibition of immune complex-mediated

Erk activation resulted in suppressed IL-10 and IL-1β production7,8.

Further analysis of the Erk-dependent IL-10 production indicated that Erk is critical for

histone phosphorylation and subsequent binding of Sp1 transcription factor to IL-10

promoter7. Third, pharmacological inhibition of p38 or disruption of one of its activator,

MAPKK, resulted in decreased p38 activity and a selective defect in LPS-induced IL-12

production6. Finally, deficiency of JNK resulted in severe defect in the production of multiple cytokines by fibroblasts. Inhibition or siRNA-mediated knock down of JNK but not Erk or p38 resulted in the decrease of TNF-α-induced PTX3 (an acute phase protein mediating innate immune responses) in human lung epithelial cells9.

1.3 PI3K/Akt Pathway and Innate Immunity

PI3Ks are a conserved family of proteins involved in several different functions

such as inflammatory response, survival, cell proliferation and metabolism. Four different

classes of PI3K exist, class IA, IB, II and III based on structural characteristics and

substrate specificity. PI3K consists of two subunits- catalytic and regulatory. All the

members of PI3K phosphorylate phosphatidylinositides on 3’ position. Such

phosphorylated phosphatidylinositides serve as docking sites for other proteins10. For

example, phosphatdiylinosiol 3, 4 bisphosphate (PIP2) and phosphatidylinositol 3,4,5

trisphosphate (PIP3) bind to pleckstrin homology (PH) domain-containing proteins such

as Akt, Btk, Vav etc. Similarly phosphatidylinositol 3 phosphate (PI3P) serves as a docking site for proteins with phox homology (PX) domain such as p47phox, one of the

important components of phagocytic oxidase system.

Extra-cellular Stimuli

SHIP Tyrosine Kinases SHIP (eg. Syk)

MAPKs PI3K/Akt (eg. Erk, p38 and JNK)

Transcriptional Factors

Inflammatory Mediators Phagocytosis Phagocytosis

Innate Immunity Toxic Metabolites (eg. RNS, ROS)

Figure 1.2: Schematic representation of macrophage signaling and its role in innate immunity. Several signaling cascades including MAPKs and PI3K/Akt pathways are activated upon binding of cognate ligand to its receptor. In certain contexts, SHIP1, an inositol phosphatase, negatively regulates both MAPK and PI3K/Akt pathways activation. These pathways regulate several effector functions that contribute to the innate immunity. RNS, reactive nitrogen species. ROS, reactive oxygen species.

PI3Ks have been implicated in several effector functions of macrophages such as

phagocytosis, chemotaxis, and toxic metabolites production11. A schematic model

describing the role of PI3K in innate immune responses is shown in Figure 1.2. The

following paragraphs describe the role of PI3K in some of its effector functions. PI3K is

critical for phagocytosis initiated through receptors such as Fcγ and complement

receptors12,13. The importance of PI3K in phagocytosis is more prominent in the engulfment of larger particles (> 4.5μm)14. The function of PI3K in the phagocytic uptake

is interesting. Although PI3K is dispensable for the particle binding and subsequent

phagocytic cup formation but it is critical for pseudopod extension and the ensuing phagocytic cup closure14,15. Downstream of PI3K, Akt and Rac were reported to be

essential for the phagocytosis of IgG-opsonized Sheep-RBC (S-RBC)16,17. In addition to

the role of PI3K in FcγR-mediated phagocytosis, Allen et al reported that it is required

for the delayed of phagocytosis of ulcerogenic Helicobacter pylori18. However, PI3K-

independent phagocytosis was also reported in the internalization of Salmonella19,20 and

Shigella20. PI3K is also required for chemotaxis. Attesting to the importance of PI3K in

chemotaxis, the inhibition of Akt, a downstream-effector of PI3K, significantly

dampened fMLP-stimulated neutrophil chemotaxis21.

PI3K is also required for the generation of toxic metabolites such as superoxide

and nitric oxide. First, Rac2 (PI3K-dependent protein), an abundant isoform of Rac in

neutrophils, is an essential component of the phagocytic oxidase (or NADPH oxidase)

complex that catalyzes the production of superoxide22. Second, PI3K was reported to be

critical for the phosphorylation of p47phox, an essential component of phagocytic

oxidase. Specifically, PI3K effector Akt was demonstrated to interact with and 8

phosphorylate p47phox. Thus, Akt regulates the fMLP-induced respiratory burst in

human neutrophils23. Third, macrophages deficient in PI3K are impaired in the

production of LPS and IFNγ induced nitric oxide production24. In this study, Sakai et al. found that the expression of inducible nitric oxide synthase (iNOS), an enzyme involved in the generation of nitric oxide, is unchanged in PI3K-deficient macrophages but the dimerization of iNOS protein is significantly impaired.

PI3K has an indispensable role in modulating the production of cytokines in response to pathogenic stimuli25,26. The role of PI3K in inflammation is controversial.

Some studies demonstrate that PI3K promotes inflammation but few studies suggest that

PI3K/Akt pathway may act as a negative regulator and limit pro-inflammatory response

in response to infectious stimuli. Briefly, PI3K/Akt pathway was reported to promote

LPS-induced IL-10 production by macrophages27,28 suggesting a negative role for PI3K.

Also, in another study inhibition of PI3K using a wortmannin in vivo resulted in increased serum levels of IL-1β, TNFα, IL-6 etc. thereby enhancing the susceptibility to polymicrobial sepsis25,26. In contrast, blockade of PI3K/Akt pathway by specific inhibitor

inhibited immune complex-induced IL-6 production but not IL-1β production in bone

marrow derived macrophages suggesting that PI3K/Akt is a positive regulator of FcγR-

mediated cytokine response8. Further, Fukao et al reported that p85α (one of the subunits

of PI3K) knockout mice showed impaired clearance of enterobacteria injected into the

peritoneal cavity29. These discrepancies regarding the requirement of PI3K in inflammation may be due to the differences in the cellular context and/or the stimuli used. Future studies may enhance our understanding about the role of PI3K in inflammation. 9

In the following sections, the signaling pathways that either promote (Src and Syk

tyrosine kinases) or downregulate (inositol phosphatases) the activation of MAP kinase

and PI3K/Akt pathways are described.

1.4 Src and Syk Kinases

Stimulation of cells with immune complex or LPS leads to tyrosine

phosphorylation of several proteins suggesting a key role for tyrosine kinases such as Src

family of kinases, Syk, Btk etc. These tyrosine kinases with a prominent role in receptor

proximal signaling activate several downstream signaling pathways including MAPK and

PI3K/Akt pathways and ultimately regulate several functional outcomes such as

phagocytosis, cytokine response, chemotaxis and production of reactive intermediates.

Both Src and Syk kinases have indispensable role in signaling initiated through

FcγR clustering. For example, the ITAM of the receptor FcγIIa is phosphorylated by Src

kinases30. In many instances both Src and Syk kinases operate together and it is perceived

that Src kinases are upstream of Syk. Both Src and Syk kinases are essential components

of macrophage phagocytic process mediated by Fcγ receptors. Lyn, a Src kinase, was

shown to be important in the phagocytosis of microbes mediated by Fcγ receptors31. Both

Src and Syk are implicated in the actin skeletal rearrangements that are essential for the phagocytic process32-34. On the other hand, some of their effects appear to be non-

redundant. Loss of Src kinases decreased the phagocytic ability of macrophages but lack

of Syk completely abolished phagocytosis mediated by Fcγ receptors32. Src kinase-

deficient cells showed impaired actin polymerization below the phagocytic cup where as

Syk deficiency lead to inappropriate closure of phagocytic vesicle32. Such a defect in the 10

failure of the closure of phagocytic cup was also seen when PI3K pathway is blocked by

a specific inhibitor suggesting that Syk may be acting upstream of PI3K14,15. Indeed, Syk

is reported to be upstream of PI3K in B cells35. Src kinases are reported to be important in

the activation of PI3K/Akt pathway and Rac, which are essential for the actin cytoskeletal

rearrangements. The role of Syk in complement mediated phagocytosis appears to be

controversial. Kiefer et al reported that Syk is not important for CR3-mediated

internalization of serum opsonized zymosan36. However, recently Shi et al demonstrated

that Syk is indispensable for CR3-mediated phagocytosis34.

Syk is also important for other effector functions such as the production of ROS,

RNS and inflammatory mediators. First, Syk-deficient neutrophils are deficient in FcγR-

induced respiratory burst36. Further Syk is required for the ROS production induced by the engagement of Dectin-1 receptor in macrophages37. Second, inhibition of Syk by

siRNA or piceatannol (specific inhibitor of Syk) significantly downregulated the

production of TNFα induced iNOS and nitric oxide production in lung airway epithelial

cells38. Third, Syk is important for the inflammatory response induced by receptors such

as FcγR and Dectin-1. Thus, Syk is critical for various innate immune responses.

1.5 Inositol Phosphatases

Excessive production of inflammatory mediators may lead to collateral tissue

damage and undesired systemic effects. Thus, the inflammatory response needs to be

modulated. This is achieved by both positive mediators and negative regulators. One

group of such negative regulators are the inositol phosphatases. These proteins

differentially regulate MAPK and PI3K/Akt pathways under different stimulatory 11

conditions thereby influencing the final functional outcome. A brief description of such proteins is presented in the following paragraphs. Three main inositol phosphatases have

been described in immune cells namely SH2 domain containing 5’ inositol phosphatase 1

(SHIP1), SHIP2 and phosphatase and tensin homolog deleted on chromosome 10 (Pten).

A schematic model of the 3 inositol phosphatases is shown in the Figure 1.3. These

inositol phosphatases work as the negative regulators of the PI3K pathway but possess

different modes of actions. Both the SHIP proteins catalyze the hydrolysis of the 5’

phosphate on PIP3 and subsequently dampen the activation of PI3K-dependent enzymes.

However, Pten hydrolyzes the 3’ phosphate on PIP3. Further, Pten is also capable of

dephosphorylating tyrosine residues, and is therefore referred to as dual phosphatase39.

The expression of SHIP1 is restricted to hematopoietic cells where as SHIP2 is

ubiquitous. Both proteins have N-terminal SH2 domain, central catalytic domain and C-

terminal proline rich domain (PRD). SHIP1 and SHIP2 harbor two and one NPXY

motifs, respectively in their PRDs. SH2 domain, PRD and NPXY motifs are implicated in

protein-protein interactions whereas the catalytic domain is responsible for hydrolyzing

PIP3. Thus, SHIP1 influences the functional outcomes of immune cells via its catalytic

activity and through its interactions with other proteins39,40.

The SH2 domain of SHIP1 targets it to the membrane where it binds to tyrosine

phosphorylated receptors such as immune receptors and growth factor receptors39. At the membrane, tyrosine residues in the NPXY motifs of SHIP1 are phosphorylated in response to different stimuli. Subsequently it can bind to proteins containing phosphotyrosine binding domains such as Shc and Dok. The PRD of SHIP1 can associate with

SHIP1 SHIP2 NPAY NPLY NPNY

PROLINE PROLINE SH2 CATALYTIC SH2 CATALYTIC SAM RICH RICH

PI (3,4,5)P3 PI (3,4)P2 PI (3,4,5)P3 PI (3,4)P2

Serine phosphorylation PTEN sites

PDZ CATALYTIC C2 BD BD

PI (3,4,5)P3 PI (4,5)P2 Dual Phosphatase activity

pY Protein Y Protein

Figure 1.3: Domain structures of inositol phosphatases. Both SHIP1 and SHIP2 hydrolyze 5’ phosphate group where as PTEN hydrolyzes 3’ phosphate group of phosphatidylinositol 3,4,5 trisphosphate. PTEN also exhibits tyrosine phosphatase activity.

Grb2 thereby influence the activation of Erk pathway. Owing to its catalytic activity,

SHIP1 hydrolyzes PIP3 resulting in the decreased activation of PI3K-dependent proteins

such as Akt, Btk and Vav thereby affecting the downstream signaling41.

Targeted disruption of SHIP1 results in decreased life span, splenomegaly,

lymphoadenopathy, extensive proliferation of granulocytes and macrophages and

elevated serum levels of immunoglobulins. This phenotype emphasizes the role of SHIP1

in hematopoietic cells42.

SHIP1 is a critical regulator of signaling pathways induced by different stimuli.

Similar to the controversy about the role of PI3K in inflammation, the requirement of

SHIP1 in modulating inflammation is also contentious. Briefly, SHIP1 positively

regulates LPS-induced signaling but negatively regulates TLR2 and FcγR-mediated

signaling outcomes. SHIP1-deficient macrophages displayed enhanced LPS-induced Akt phosphorylation but diminished Erk and p38 phosphorylation suggesting that SHIP1 is a negative regulator of LPS-induced PI3K/Akt pathway and a positive regulator of LPS- induced MAPK activation43. Moreover, treatment of BMMs with LPS resulted in enhanced SHIP1 expression and this induction was found to be essential for endotoxin

tolerance44. In contrast, SHIP1-deficient macrophages displayed elevated levels of

phospho-Erk upon clustering of FcγR suggesting a negative role for SHIP1 in the

activation of Erk8. Similarly, SHIP1 was reported to negatively regulate TLR2-induced

signaling and functional outcomes45. Future studies will probably find an answer to these

controversies.

Due to its ability to regulate several key signaling pathways, SHIP1 modulates

several effector functions such as phagocytosis, inflammation, toxic metabolite

generation etc. For example, SHIP1 is a negative regulator of both FcγR and complement

receptor 3 (CR3)-induced phagocytosis12. It is believed that SHIP1 downregulates

phagocytosis by dampening PI3K/Akt pathway. Further SHIP1 is demonstrated to

effectively regulate inflammatory response induced by the ligation of TLRs or

FcγRs8,45,46.

Both SHIP2 and Pten are also reported to modulate immune receptor-induced

events. SHIP2 negatively regulates the phagocytosis of IgG-opsonized S-RBC

independently of SHIP1 via the inhibition of Rac activation47. Likewise, Pten-deficient

peritoneal macrophages displayed enhanced phagocytosis mediated through FcγR48. In addition, Pten also downregulated immune complex-induced cytokine production but promoted LPS-induced TNFα in BMMs48. These findings parallel to those obtained with

SHIP1. Consistently, Pten negatively regulated FcγR-induced activation of Akt and Erk

but promoted LPS-induced Erk phosphorylation48.

1.6 Classical Activation of Macrophages-Potent Innate Immune Response

Innate immune responses of macrophages are also modulated by the surrounding

cytokine milieu. For example, IFNγ, a cytokine produced predominantly by NK cells and

49 TH cells, activates macrophages and primes them for a potent TH1 response . Such activation of macrophages alters their phenotype and confers additional functional capabilities to eliminate the invading pathogen. On the other hand, IL-4 and IL-13 (TH2 cytokines) modify the macrophages to be anti-inflammatory and directs them to be 15

involved in wound healing and tissue repair49. Mainly, 3 different macrophage

populations with distinct biologic functions exist-classically activated macrophages,

alternatively activated macrophages and Type II-activated macrophages. In the following

paragraphs there is a brief description about classically activated macrophages.

Classically activated macrophages

Macrophages primed with IFNγ are referred to as classically activated

macrophages. However priming with IFNγ itself does not activate macrophages, they

must be stimulated by a second stimulus such as TNFα or a TNFα inducer49. Classically

activated murine macrophages are characterized by their ability to produce NO which has

anti-microbial properties49. However, the production of nitric oxide in human cells is

controversial. Activation of macrophages by IFNγ does not change the phagocytic ability

of macrophages but it enhances the expression of receptors such FcγRI and the

components of phagocytic oxidase and subsequently the respiratory burst49. Thus, IFNγ is very critical in enhancing innate immune responses in response to pathogenic stimuli. A new emerging field is that microbes such as Leishmania50 and Mycobacteria51,52 have

developed ways to circumvent the IFNγ-mediated protective effects. Understanding the

mechanism used by these pathogens to subvert IFNγ-mediated response will not only

help us understand their pathogenesis but it will further enhance our understanding about the functioning of IFNγ-mediated activation itself.

Alternative macrophages are anti-inflammatory and immunosuppressive in nature.

Alternative activated macrophages are generated by the exposure of cells to cytokine IL-4

or gluocorticoids. Alternatively activated macrophages express higher levels of

fibronectin and arginase (enzyme metabolizing arginine to urea, the precursor of proline

and polyamines), thus they are believed to function in tissue repair49. Type II activation

of macrophages requires two signals- clustering of FcγR coupled with the ligation of

TLR. Such macrophages produce increased amounts of IL-10 and drastically less IL-12

production conferring an anti-inflammatory phenotype. Such phenotype may be exploited

to prevent acute pathologies such as those associated with LPS endotoxemia49.

1.7 Tularemia

Francisella tularensis, a Gram-negative facultative organism, is the causative

agent of the zoonotic disease tularemia. It is most commonly reported in North America

and Eurasia. Humans are infected with this pathogen in different ways: arthropod bites,

handling infectious animal tissues, direct contact with contaminated water & food and

inhalation of infectious aerosols53. The incidence of tularemia is more common in June

through September when tick-borne transmission is high. Persons of all age groups are

affected but tularemia is more commonly seen in children less than 10 years54. Tularemia is manifested in different forms such as inhalational tularemia, typhoidal tularemia etc.

The signs of inhalational/pneumonia form include bronchiolitis, pleuropneumonitis and hilar lymphadenitis54. The major targets of Francisella are the lung, heart, spleen and

liver. Mortality due to tularemia is 30-60 % when not treated. Although tularemia is not a

commonly occurring disease, due to the possibility of the use of Francisella in bio-

warfare, there is currently an increasing thrust to understand its pathogenesis and host

response.

1.8 Intra-cellular Lifestyle of Francisella tularensis

The organism primarily infects immune cells such as monocytes, macrophages and neutrophils wherein Francisella replicates and establishes its niche in the host system53. A schematic model describing the intra-cellular lifestyle of Francisella tularensis is shown in the Figure 1.4. The molecular basis of the internalization of

Francisella is not identified except that the phagocytosis of Francisella requires actin cytoskeletal rearrangements55. Macrophage receptors for Francisella are beginning to be defined. Recent reports suggest that complement receptor355 (CR3), FcγR56, mannose receptor (MR)56 and scavenger receptor (SR)57 are important in the internalization of

Francisella. Once internalized, the phagosome containing the pathogen does not fuse with the lysosome and the bacteria escape into the cytosol of the host cell. In the cytosol, bacteria replicate and trigger apoptosis of the host cell. Subsequently, bacteria infect surrounding cells53. Recently it was reported that post-replication, the bacteria are sequestered into autophagic vacuoles which later fuse with lysosomes forming autophagolysosomes58.

1.9 Francisella: Host Cell Response and Innate Immunity

Host response and molecular mechanisms:

The host cell response to Francisella tularensis infection and its regulation is poorly defined. It primarily involves phagocytosis, subsequent production of inflammatory mediators (such as IL-12, TNFα, IL-6, IL-1β, IFNγ, ΜCP-1 etc), toxic metabolites (ROS and RNS) generation, apoptosis53 and autophagy58. 18

Figure 1.4: Intra-cellular lifestyle of Francisella bacteria. The receptors involved in phagocytosis of Francisella are beginning to be understood. Macrophages engulf the organism. Subsequently the bacteria escape into the cytosol of the host cell wherein they replicate and trigger apoptosis of the host cell.

There is preliminary evidence that Francisella in addition to macrophages,

monocytes and neutrophils can also replicate in vitro in dendritic cells and non-

phagocytic cells such as hepatocytes and epithelial cells59. One study suggests that

Francisella-infected dendritic cells fail to produce TNFα, IL-12 and IL-10 in response to

TLR ligands following the infection60. Further, the authors report that infected dendritic

cells produced TGFβ, a cytokine commonly associated with dampening pro- inflammatory cytokine response. Results from another study indicate that Francisella- infected dendritic cells readily produce TNFα and IL-1261. The discrepancy in the two

studies may be due to the bacterial culture conditions. Indeed, Loegering et al. reported

that Francisella grown in broth conditions elicited the production of multiple cytokines however bacteria harvested from macrophages failed to stimulate similar response62.

The molecular mechanisms regulating the inflammatory response induced by

Francisella are not clearly understood. Activation of MAPK and NFκB has been

reported63. TLR2 has been shown to play a critical role in Francisella-induced

inflammatory response. Francisella infected TLR2-deficient macrophages are impaired

in the activation of NFκB and subsequent production of pro-inflammatory cytokines such

as TNFα and IL-661. Further TLR2-/- mice were highly susceptible for intra-nasal infection of Francisella and showed higher bacterial burden in lungs, spleen and liver.

Innate immunity against Francisella and potential subversion mechanisms:

Innate immunity against Francisella infection is primarily mediated by cytokines,

ROS and RNS. Utilizing neutralizing antibody or gene knockout models IL-12,

TNFα and IFNγ have been shown to be important for the immunity against this bacterial

infection64-66. IFNγ is a major player for combating Francisella infection. Data obtained from various mouse models strongly support the protective role of IFNγ during tularemia.

Dramatic increase in the number of IFNγ-secreting NK cells have been observed 3 days post-infection67. Priming of the human macrophages with IFNγ increased the fusion of

Francisella-containing phagosome (FCP) to fuse with lysosomes68. Moreover, other

studies report that IFNγ-induced ROS contributes very little to innate immune responses

of the cytokine, but IFNγ-induced RNS (nitric oxide) is shown to be critical for the killing of Francisella in mouse macrophages69. The effector mechanisms mediating

Francisella-killing in human cells may not involve NO or ROS and need to be

characterized.

Since Francisella normally resides and replicates in immune cells that are

equipped to target invading pathogens, thus, it should be have been equipped with

effective defense mechanisms to permit the establishment of its niche in these cells.

These mechanisms include inhibiting phagolysosomal fusion70, suppressing inflammatory

response upon phagosomal escape63,71 and dampening T cell responses72. There may be

more subversion mechanisms that are not yet explored.

Although it is known that Francisella induces the production of inflammatory

mediators such as cytokines, chemokines and nitric oxide but the molecular mechanisms

of this host response is not well understood. Further, the regulatory mechanisms that are

in place to modulate this host response are poorly defined. Therefore, the main objective

of this study is to understand the regulatory mechanisms that control the macrophage

response to Francisella. Host response involves phagocytosis and production of

inflammatory mediators. The latter can feedback to other cells leading to production 21

of cytokines such as IFNγ which further boost the innate immune responses of the

macrophage. On the other hand, pathogens may subvert such host mounted immune

responses so that they can modify the cellular conditions to suit their needs.

Thus, in this study we investigated the roles of Syk, a tyrosine kinase and SHIP1,

an inositol phosphatase in the Francisella-induced innate immune responses of

macrophages. Also, we studied how the bacteria can subvert the IFNγ-mediated host

immune responses to survive better in the macrophages. A model describing the critical

findings of this study is depicted in the Figure 5.1.

Chapter 2 describes studies utilizing genetic and pharmacological approaches to

investigate the role of Syk in Francisella engulfment and cytokine response to infection

and demonstrates that Syk promotes the uptake and subsequent cytokine release through

the activation of Erk and PI3K/Akt pathways.

Chapter 3 describes the use of gene knockout and overexpression models to

examine the role of SHIP1 in the phagocytosis of Francisella and in the inflammatory

response to infection. These studies established that SHIP1 is dispensable for Francisella phagocytosis but is critical to temper the cytokine response to infection by antagonizing the PI3K/Akt pathway.

Chapter 4 describes studies examining how Francisella evades IFNγ-mediated

responses by suppressing the signaling cascade initiated by IFNγ. We also describe the

functional consequences of this subversion process and demonstrate that Francisella

prevents the induction of iNOS resulting in enhanced intra-macrophage survival of the

bacteria.

CHAPTER 2

THE TYROSINE KINASE SYK PROMOTES PHAGOCYTOSIS OF

Francisella AND CYTOKINE RESPONSE TO INFECTION

2.1 Abstract

Francisella tularensis is a highly virulent, Gram-negative intra-cellular pathogen

that can cause the zoonotic disease tularemia. Host response against Francisella infection

involves phagocytosis of the pathogen and subsequent generation of the inflammatory mediators and toxic metabolites. The molecular mechanisms regulating such crucial host response are not well understood. The receptors crucial for recognition and subsequent internalization of Francisella by macrophages are beginning to be defined, but the identity of the downstream signaling pathways that modulate the process of phagocytosis and ensuing cytokine response are not yet identified. In this study we tested the role of

Syk, a tyrosine kinase, in the Francisella phagocytosis and Francisella-induced cytokine response. We report that Syk is activated during Francisella infection and is critical for the uptake of the organisms. Pharmacologic inhibition of Syk almost completely inhibited uptake, whereas the overexpression of Syk significantly enhanced uptake. However, Syk appears to be dispensable during initial host-pathogen contact. Further analyses of the molecular mechanism of Syk-dependent Francisella uptake revealed that the MAPK Erk

but not the PI3K/Akt pathway is the downstream effector of Syk. Thus, the inhibition of 23

Erk in Syk-overexpressing cells or the suppression of Syk activation in Erk-

overexpressing cells led to a significant attenuation of uptake. Further, we observed that

Syk promotes the production of Francisella-induced TNFα and IL-10 via the activation of Erk and PI3K/Akt pathways. Collectively, these data identify Syk as the key player in the modulation of Francisella phagocytosis and the cytokine response induced by this pathogen.

2.2 Introduction

Francisella tularensis is a Gram-negative intra-cellular pathogen and causes the

zoonotic disease tularemia. Four sub-species of Francisella tularensis exist- tularensis

(Type A), holarctica (Type B), novicida and mediasiatica. Francisella novicida is

virulent in mice but not in humans53. Further, the intra-cellular life style of F. novicida is

similar to that of the highly virulent Type A strain68. Therefore, F. novicida is a frequently used experimental model for tularemia in the murine system. The infectious dose of F. tularensis tularensis is very low and the organism can be easily aerosolized, thus, it is considered a potential biological weapon.

Francisella tularensis predominantly infects immune cells such as macrophages,

monocytes and neutrophils. After bacterial internalization, the Francisella-containing

phagosomes do not fuse with the lysosome and the bacteria subsequently escape into the

cytosol. In the cytosol, the bacteria replicate and prompt apoptosis of the host cell53. The host response to Francisella infection is beginning to be understood. Host response involves key processes such as phagocytosis, production of inflammatory mediators and generation of toxic metabolites. The molecular mechanisms leading to the production of 24

inflammatory mediators are poorly defined. The activation of MAP Kinases and NFκB has been reported in Francisella infected cells63. Recent reports also indicate that

activation of the inflammasome complex is mediated by Francisella that escape into the

cytosol, resulting in the processing and release of IL-1β73,74.

Also, the mechanisms regulating the phagocytosis of Francisella are currently

unknown. Several host cell receptors including complement receptor3 (CR3)55, mannose

receptor56, class A scavenger receptor57, Toll-like receptor 261 and Fcγ receptors56 have been implicated in the recognition and subsequent host response against Francisella.

However, the downstream signals that coordinate phagocytosis of Francisella have not been identified.

Syk is a tyrosine kinase that has been shown to be crucial for various immune cell functions, including cytoskeletal rearrangements, phagocytosis, cytokine generation and production of toxic metabolites. Thus, in this study we specifically examined the role of

Syk in both Francisella phagocytosis and in the cytokine response induced by this pathogen. Our data demonstrate that Syk is phosphorylated in Francisella-infected cells.

Inhibition of Syk markedly suppressed uptake of Francisella, whereas overexpression of

Syk enhanced the uptake. Interestingly, this Syk-dependent uptake is mediated via Erk but not through the PI3K/Akt pathway. Further we found that inhibition of Syk suppressed Francisella-induced TNFα and IL-10. On the other hand, overexpression of

Syk significantly increased the production of these cytokines. Further analysis of this

Syk-dependent cytokine response suggested that Syk-effectors Erk and PI3K/Akt pathways differentially regulate the generation of TNFα and IL-10. Therefore, these data

suggest an indispensable role for Syk in Francisella phagocytosis and in the generation

of cytokines as well.

2.3 Materials and Methods

Cells, antibodies and reagents: Raw 264.7 and THP-1 cells were obtained from ATCC

and maintained in RPMI 1640 with 5% heat-inactivated fetal bovine serum (FBS).

Antibodies specific for phospho-tyrosine, Syk, Erk and actin were obtained from Santa

Cruz Biotechnology (Santa Cruz, CA). Antibodies against pErk and pSer Akt were

obtained from Cell Signaling Technology (Beverly, MA). Mouse anti-Francisella

lipopolysaccharide primary antibody was from Immune-Precise Antibodies Limited

(Victoria, B.C., Canada). Piceatannol, UO126 and LY294002 were purchased from

Calbiochem (San Diego, CA). F. novicida U112 (JSG1819), a generous gift of Dr. John

Gunn (The Ohio State University, OH) were used in all experiments. Bacteria were

streaked and grown overnight on Chocolate II agar plates (Becton, Dickinson and

Company, MD) at 370C.

Culture of murine bone marrow macrophages: Bone marrow macrophages (BMMs) were

derived from mice as previously described8. Briefly, bone marrow cells were cultured in

RPMI containing 10% fetal bovine serum plus 10 μg/ml polymixin B and supplemented with 20ng/ml CSF-1 for 7 days before they were used in experiments. BMMs derived in this manner were >99% positive for Mac-1, as determined by flow cytometry.

Cell stimulation, lysis, and Western blotting: Macrophages were infected with plate- grown (grown on Chocolate II agar plates for 16-18 h at 370C) F. novicida as previously

described73 at a multiplicity of infection (MOI) of 100. Briefly, RAW 264.7 cells were

plated in 12-well or 6 well plates and allowed to adhere. F. novicida resuspended in

RPMI medium containing 5% heat inactivated FBS was added to the adherent

0 macrophages and then incubated at 37 C and 5% CO2 for the indicated time points. In

parallel, the viability of bacteria was tested by plating the inoculum on Chocolate II agar

plates and bacterial numbers in the inoculum were quantified using the Petroff-Hauser

chamber. These data indicate that >98% of bacteria in the inoculum were viable. During

the infection, cells were not washed at any point unless indicated otherwise. Where

indicated, before infection, cells were incubated with specific pharmacological inhibitors

for 30 minutes. Post-infection, the cell supernatant was aspirated and uninfected and

infected cells were lysed in TN1 buffer (50mM Tris pH 8.0, 10mM EDTA, 10mM

Na4P2O7, 10mM NaF, 1% Triton-X 100, 125mM NaCl, 10mM Na3VO4, 10μg/ml each

aprotinin and leupeptin). Post-nuclear lysates were boiled in Laemmli Sample Buffer and

were separated by SDS/PAGE, transferred to nitrocellulose filters, probed with the

antibody of interest and developed by enhanced chemiluminescence (ECL).

Western blot data quantitation: The ECL signal was quantitated using a scanner and a

densitometry program (Scion Image) as previously described71,75. To quantitate the

phospho-specific signal in the activated samples, we first subtracted background,

normalized the signal to the amount of actin or total target protein in the lysate, and

plotted the values as arbitrary units (a.u). Statistical analysis was performed by unpaired

Student’s t-test. p<0.05 was considered significant.

Colony forming unit (CFU) assay: CFU assays were performed as we have previously reported76 with few modifications. Briefly, RAW 264.7 cells were infected with 100 MOI and then centrifuged at 650g for 4 minutes. Infection was allowed to occur for a total of

60 minutes, next infected cells were washed two times with sterile PBS and treated with

0 50 μg/ml of gentamicin for 30 min at 37 C and 5% CO2. Cells were then washed twice

and subsequently lysed in 0.1% SDS for 5 minutes. Immediately, 10 fold serial dilutions

were made and appropriate dilutions were plated on Chocolate II agar plates. Assays

were performed in triplicate for each test group. Statistical analysis was performed by

unpaired Student’s t-test. p<0.05 was considered significant.

Microscopy analysis of Francisella association with macrophages: Phagocytosis of

Francisella was measured by microscopy as previously described75, with a few

modifications. In brief, 60 minutes post-infection, cells were washed (with PBS) and

fixed in 4% paraformaldehyde for 20 minutes. The cells were washed again and one of

the two sets of samples was permeabilized with 100% methanol for 10 minutes and the

other set was left non-permeabilized. Immunostaining was then performed with mouse

anti-F. novicida LPS antibody (diluted 1/100; Immune Precise Antibodies) and the

bacteria were visualized by anti-mouse Alexa Fluor 488 secondary antibody. Bacteria

binding to macrophages were counted on non-permeabilized samples whereas total

number of bacteria associated (both attached and phagocytosed) with the cells were

counted on methanol permeabilized samples using the X100 oil immersion objective of a

BX40 Olympus fluorescence microscope. The number of bacteria phagocytosed was

obtained by subtracting bacterial numbers adhering to macrophages from the total number of bacteria associated with the cell. At least 100 cells per sample were examined

and three separate sets of infection were analyzed. Phagocytic and binding indices are

defined as the number of bacteria phagocytosed or adherent to 100 macrophages,

respectively. Statistical analysis was performed by unpaired Student’s t-test. p<0.05 was

considered significant.

Plasmids and transient transfection: The construct encoding Syk was a kind gift from

Dr. Axel Ullrich (Max-Planck Institute of Biochemistry, Germany). GST-Erk2 was a

generous gift from Dr. Mark Coggeshall (Oklahoma Medical Research Foundation, OK).

RAW 264.7 cells were transfected with the appropriate plasmid DNA using the Amaxa

Nucleofector apparatus (Amaxa biosystems, Germany) as previously described17. Briefly,

7 x106 cells were resuspended in 100 μl Nucelofector Solution V and were nucelofected with 8 μg of appropriate plasmid. Immediately post- nucleofection, 500 μl of pre-warmed

RPMI was added to the transfection mix before transferring to 12-well plates containing

1.5 ml pre-warmed RPMI per well. Plates were incubated for 16 hours at 37oC before

infections were performed.

ELISA measurement of cytokine production: RAW 264.7 cells and BMMs were infected with F. novicida for 8 hours. Cell supernatants were harvested, centrifuged to remove dead cells and analyzed by ELISA using cytokine specific kits from R & D Systems 29

(Minneapolis, MN). Data were analyzed using unpaired Student’s t-test. p value < 0.05

was considered as significant.

Luciferase Assays: RAW 264.7 cells were transfected with NFκB-luc plasmid DNA

using the Amaxa Nucleofector apparatus (Amaxa biosystems, Germany) as previously

described17,46. Briefly, 12x106 cells were resuspended in 100 μl Nucelofector Solution V and were nucelofected with 1 μg of NFκB-luc. Immediately post nucleofection, 500 μl of pre-warmed RPMI was added to the transfection mix before transferring to 12-well plates containing 1.5 ml pre-warmed RPMI per well. Plates were incubated for 16 hours at

37oC. Transfected cells were either left uninfected or were infected for 5 hours with F.

novicida. Cells were lysed in 100 μl of Luciferase Cell Culture Lysis Reagent (Promega).

Luciferase activity was then measured using Luciferase Assay Reagent (Promega), as previously described17,46.

2.4 Results

Tyrosine phosphorylation of cellular proteins during Francisella infection

As described in Chapter 1, tyrosine kinases are critical for modulating several effector functions of immune cells. Thus, we examined the potential role of tyrosine kinases in general during Francisella infection. For this mouse macrophages were infected with 100 MOI of F. novicida for indicated time points and the tyrosine phosphorylation of the proteins was assessed by Western blotting analysis with phospho tyrosine (pY) antibody. The results shown in the Figure 2.1 suggest that many proteins are tyrosine phosphorylated in the infected cells. 30

R 30’ 2h 5h 8h

98 kDa

Immunoblot: Anti-pY

Immunoblot: Anti-Actin

Figure 2.1: Francisella induces tyrosine phosphorylation of proteins. BMMs were infected with F. novicida for indicated time points. Protein-matched lysates were resolved by SDS-PAGE and tyrosine phosphorylation of cellular proteins was studied by Western blotting with phospho-tyrosine antibody. R, resting, uninfected cells.

Francisella infection activates Syk in both mouse and human cells

Several tyrosine kinases such as Syk, Src family of kinases and Btk are expressed in immune cells. As mentioned in the introduction, Syk tyrosine kinase is critical for several immune effector functions. Thus we wanted to examine the role of Syk during

Francisella infection. Specifically we examined the role of Syk in Francisella phagocytosis and Francisella-induced cytokine response. For this, we first tested if Syk is involved during Francisella infection. The activation of Syk is accompanied by autophosphorylation, so we used tyrosine phosphorylation of Syk as a measure of its activity. To examine whether Syk is activated during Francisella infection, RAW 264.7 cells were infected with 100 MOI of F. novicida for different time points indicated in

Figure 2.2A and the induction of Syk phosphorylation was examined by Western blot analysis. For this, Syk was immunoprecipitated from uninfected and infected cells and

Western blotting was performed using pY antibody (Figure 2.2A, upper panel). The same membrane was reprobed with Syk antibody to ensure equal loading in all the lanes

(Figure 2.2A, lower panel). Results indicated that Syk is phosphorylated as early as 30 seconds in infected cells.

Host responses in human and mouse cells may differ considerably during

Francisella infection. Thus, to test if Syk is phosphorylated in human cells also, THP1 cells, a human monocytic cell line, were infected for the indicated time points and the phosphorylation of Syk was assessed by Western blot analysis. Results shown in Figure

2.2B indicate that Syk is phosphorylated in infected human monocytic cells as well.

Collectively, these data demonstrate that Syk is activated in cells infected with

Francisella, suggesting a potential role for Syk in macrophage/monocyte response to

Francisella.

Francisella-induced Syk activation is dose dependent but strain independent

We next examined whether the phosphorylation of Syk was dose-dependent. To

study this, RAW 264.7 cells were infected with 1, 10 or 100 MOI of F. novicida for 1 minute and phosphorylation of Syk was assessed as described above. Results shown in

Figure 2.2C indicate that Syk phosphorylation is minimal at 1 MOI but is obvious at 10

MOI suggesting Francisella-induced Syk phosphorylation is dose dependent. Further, we found that Syk was phosphorylated in cells infected with F. tularensis LVS, the vaccine

strain of F. tularensis (Figure 2.2D) indicating the phosphorylation of Syk during

Francisella infection is strain independent.

Syk promotes the phagocytosis of Francisella

Having established that Syk is activated very early in infection, we next examined

whether Syk is required for the engulfment of Francisella. To test this, RAW 264.7 cells

were incubated with DMSO (vehicle control) or piceatannol, a specific Syk inhibitor, and

then infected with 100 MOI of F. novicida for 1 hour. Phagocytosis of the bacteria was

assessed by both colony forming unit (CFU) assay and by immunofluorescence

microscopy and the results are shown in Figures 2.3 and 2.4. Inhibition of Francisella

phagocytosis by the actin polymerization inhibitor cytochalasin-D served as our positive

control in the experiment.

A B

R 30’’ 1’ 5’ R 30’’ 1’

Immunoprecipitation: Syk Immunoprecipitation: Syk Immunoblot: pY Immunoblot: pY

Immunoprecipitation: Syk Immunoblot: Syk Immunoprecipitation: Syk Immunoblot: Syk

C D

MOI

R FN LVS R 1 10 100

Immunoprecipitation: Syk Immunoprecipitation: Syk Immunoblot: pY Immunoblot: pY

Immunoprecipitation: Syk Immunoprecipitation: Syk Immunoblot: Syk Immunoblot: Syk

Figure 2.2: Francisella induces tyrosine phosphorylation of Syk. A&B. RAW 264.7 (A) or THP-1 (B) cells were infected with 100 MOI of F. novicida for the indicated time points and Syk phosphorylation was studied by Western blotting analysis with phosphotyrosine (pY) antibody (upper panels). The same membrane was reprobed with Syk antibody to ensure equal loading (lower panels). C. RAW 264.7 cells were infected with 100 MOI of F. novicida for 1 minute and Syk protein was immunoprecipitated and phosphorylation of Syk was analyzed by Western blotting. D. RAW 264.7 cells were infected with 100 MOI of F. novicida (FN) or F. tularensis LVS (LVS) for 1 minute and phosphorylation of Syk was examined by Western blotting. Data represents a representative experiment of three similar and independent experiments. R, resting, uninfected cells.

As shown in Figure 2.3A, inhibition of Syk activation markedly suppressed the phagocytosis of Francisella. As expected, treatment of cells with cytochalasin-D also inhibited the engulfment of Francisella. At the concentrations used, piceatannol was not toxic to the cells as assessed by trypan blue exclusion (data not shown). In parallel, the

effectiveness of the inhibitor, piceatannol, was tested and results shown in the Figure

2.3B indicate that Francisella-induced Syk phosphorylation was inhibited by pre-

treatment of cells with piceatannol.

As an additional approach to measure phagocytosis, we examined the role of Syk

in Francisella phagocytosis by immunofluorescence microscopy and the results are

shown in Figure 2.4. Consistent with the results obtained with CFU assays, inhibition of

Syk significantly suppressed Francisella phagocytosis (Figure 2.4). But, the attachment

of the bacteria to macrophages remained unchanged due to Syk inhibition (Figure 2.4B).

To validate the findings obtained above with the inhibitor, the role of Syk in the

phagocytosis of Francisella was alternatively assessed using a genetic approach. Here,

RAW 264.7 cells were transiently transfected with vector or a Syk-encoding construct.

Sixteen hours post transfection, cells were infected with 100 MOI of F. novicida for 1 hour and phagocytosis was assessed by CFU assay. The results shown in Figure 2.5A indicate that overexpression of Syk significantly enhanced the phagocytosis of

Francisella. In parallel experiments, the transfectants were analyzed by Western blotting with Syk antibody to test the overexpression of Syk. The data shown in Figure 2.5B verify that Syk is indeed overexpressed in Syk transfected cells. Collectively, these experiments demonstrate that Syk promotes the phagocytosis of Francisella.

A B

15 R DMSO Pice.

)

4 10 Immunoprecipitation: Syk Immunoblot: pY 5 CFU (X 10 * * 0 Immunoprecipitation: Syk DMSO Pice. Cyt. D Immunoblot: Syk

Figure 2.3: Syk inhibition suppresses the phagocytosis of Francisella. A. RAW 264.7 cells were treated with vehicle control (0.1% DMSO) or 25 μg/mL piceatannol (Pice.) or 5 μg/mL of cytochalasin-D (Cyt. D) for 30 minutes; cells were infected with 100 MOI of F. novicida for 1 hour and the phagocytosis of Francisella was examined by CFU assay. Data are a representative of three independent experiments. *, p<0.05 compared to DMSO value. B. RAW 264.7 cells were treated with vehicle control (0.1% DMSO) or 25 μg/mL piceatannol (Pice.) for 30 minutes; cells were infected with 100 MOI of F. novicida for 1 minute and the phosphorylation of Syk was examined. Data represent a representative experiment of 3 independent experiments. R, resting, uninfected cells.

250

200

150

100 * Phagocytic index 50 *

0 DMSO Pice. Cyt. D

B 160 140

120 100 80 60 Binding index 40

20 0 DMSO Pice. Cyt. D

Figure 2.4: Syk inhibition suppresses the phagocytosis of Francisella. A&B. RAW 264.7 cells were treated with 0.1 % DMSO or 25 μg/mL piceatannol (Pice.) or 5 μg/mL cytochalasin-D (Cyt. D) for 30 minutes, cells were infected with 100 MOI of F. novicida for 1 hour, phagocytosis (A) and binding (B) of the bacteria was analyzed by immunofluorescence. Phagocytic and binding indices are defined as the number of bacteria phagocytosed or adherent to 100 macrophages, respectively. Data represent mean and standard deviation of 3 independent experiments. Representative images of permeabilized and non-permeabilized samples are shown in the insets. *, p<0.05 compared to DMSO value.

* 20 )

4 15

10 10 CFU (X 5

0 Vector Syk

B Vector Syk

Immunoblot: Anti-Syk

Immunoblot: Anti-Actin

Figure 2.5: Syk overexpression enhances the phagocytosis of Francisella. A. RAW 264.7 cells were transfected with vector or Syk encoding construct, ~16 hours post- transfection cells were infected with 100 MOI of F. novicida for 1 hour and the phagocytosis of Francisella was assessed by CFU assay. Data represent mean and standard deviation of 3 independent experiments. *, p<0.05 compared to vector value. B. Vector and Syk transfectants were lysed, protein-matched lysates were resolved by SDS PAGE and the expression of Syk was analyzed by Western blotting with Syk antibody (upper panel). The same membrane was reprobed with actin antibody (lower panel). Data represents a representative experiment of at least three independent experiments.

Syk is upstream of the Erk and Akt pathways

Having established that Syk is required for the engulfment of Francisella, we then

examined the molecular mechanism underlying Syk-dependent phagocytosis. For this,

first we screened for the activation of different signaling pathways. These experiments

revealed that in addition to the previously reported MAPK pathway63, PI3K/Akt pathway is also activated in mouse macrophages (Figure 2.6 and data not shown). Francisella- induced activation of PI3K/Akt pathway was not previously reported and observed for the first time in this study. The activation of Erk and PI3K/Akt pathways has been reported to be critical for phagocytosis under other contexts. Thus, we investigated whether Syk was upstream of either the Erk pathway and/or the PI3K/Akt pathway.

To test this, cells were treated with vehicle control or piceatannol, infected for 15 minutes

and the cell lysates were examined for the phosphorylation status of Erk and Akt by

Western blotting with phospho-specific antibodies (Figure 2.6, upper panels). The same membranes were reprobed with actin antibody to ensure equal loading (Figures 2.6, middle panels).The phosphorylation signals were quantitated and plotted (Figures 2.6, lower panels). Results show a significant decrease in the phosphorylation of Erk and Akt

when cells were treated with the Syk inhibitor, piceatannol.

To test the role of Syk in the activation of Erk and PI3K/Akt using another

approach, cells were transfected with either vector or a plasmid encoding Syk. The

transfectants were subsequently infected and cell lysates were analyzed by Western

blotting with phospho-specific antibodies.

A B

R DMSO Pice. R DMSO Pice.

Immunoblot: Anti-pErk1/2 Immunoblot: Anti-pSer Akt

Immunoblot: Anti-Actin Immunoblot: Anti-Actin

15 15

10 10

Erk (a.u.) 5 5 Akt (a.u.) *

of Phosphorylation * 0 Phosphorylation of Phosphorylation 0 R DMSO Pice. R DMSO Pice.

Figure 2.6: Syk is required for the Francisella-induced phosphorylation of Erk and Akt. A&B. RAW 264.7 cells were treated with vehicle control (0.1% DMSO) or 25 μg/mL piceatannol (Pice.) for 30 minutes; cells were infected with 100 MOI of F. novicida for 15 minutes and protein-matched lysates were resolved by SDS PAGE and analyzed by Western blotting with indicated phospho-specific antibodies (upper panels). The membranes were reprobed with actin antibody (middle panels). The phosphorylation signals were quantitated, normalized to actin in each lane and graphed (lower panels). Data are representative of mean and standard deviation of three independent experiments. R, resting, uninfected cells. *, p<0.05. a.u., arbitrary units.

The results shown in Figure 2.7 indicate that phosphorylation of Erk and Akt is significantly enhanced in the Syk-overexpressing cells compared to vector transfectants.

Collectively, these data demonstrate that Syk is upstream of and required for the activation of Erk and PI3K/Akt.

Erk but not PI3K/Akt is required for the phagocytosis of Francisella

Having established that Syk is upstream of Erk and PI3K/Akt, we next

investigated which of these pathways may be required for the phagocytosis of

Francisella. For this, cells were incubated with vehicle control or U0126 (MEK/Erk pathway inhibitor) or LY294002 (PI3K/Akt inhibitor), infected and phagocytosis of

Francisella was measured by CFU and immunofluorescence assays. The results shown in

Figure 2.8A indicate that pre-treatment of cells with the Erk but not the PI3K/Akt inhibitor before infection significantly decreased phagocytosis of Francisella by macrophages. In parallel experiments, we tested the efficacy of Erk and PI3K/Akt inhibitors by Western blotting analysis (Figure 2.8B). Treatment of cells with U0126 specifically blocked the phosphorylation of Erk proteins where as incubation of cells with

LY294002 inhibited the phosphorylation of Akt. Inhibition of Erk also significantly suppressed the phagocytosis of Francisella at a lower MOI, suggesting that the decrease in the phagocytosis from Erk blockade is dose independent (data not shown).

To study phagocytosis using additional approach, we treated the cells with vehicle

control or inhibitors specific for either Erk or PI3K/Akt pathway and measured the

phagocytosis by immunofluorescence. Results shown in the Figure 2.9 indicate that

A B

Vector Syk Vector Syk R FN R FN R FN R FN

Immunoblot: Anti-pErk1/2 Immunoblot: Anti-pSer Akt

Immunoblot: Anti-Actin Immunoblot: Anti-Actin

8 6 6 * *

4 4

2 2 Erk (a.u.)

0 Akt (a.u.) 0 Phosphorylation of R FN R FN of Phosphorylation R FN R FN Vector Syk Vector Syk

Figure 2.7: Overexpression of Syk enhances the Francisella-induced phosphorylation of Erk and Akt. A&B. Vector and Syk transfectants were infected with 100 MOI of F. novicida, phosphorylation levels of Erk and Akt in the transfectants were analyzed by Western blotting with phospho-specific antibodies (upper panels). The membrane was reprobed with actin antibody (middle panels). Phosphorylation signals were normalized to actin content (lower panels). Data are representative of mean and standard deviation of three independent experiments. R, resting, uninfected cells. *, p<0.05. a.u., arbitrary units.

) 4 4 3 * 2

10 CFU (X 1

0 DMSO UO LY

B R DMSO UO LY

Immunoblot: Anti-pErk1/2

Immunoblot: Anti-pSer Akt

Immunoblot: Anti-Actin

Figure 2.8: Erk but not PI3K/Akt pathway is required for the phagocytosis of Francisella. A. RAW 264.7 cells were treated with vehicle control (0.2% DMSO) or 2.5 μΜ UO126 (UO) or 20 μΜ of LY294002 (LY) for 30 minutes; cells were infected with 100 MOI of F. novicida for 1 hour and the phagocytosis of Francisella was examined by CFU assay. Data represent mean and standard deviation of 3 independent experiments. *, p<0.05 compared to DMSO value. B. RAW 264.7 cells were treated with vehicle control or inhibitors (as described above), infected for 1 hour and protein-matched lysates were analyzed by Western blotting with phospho-specific antibodies (upper and middle panels). The same membrane was reprobed with Actin antibody to ensure equal loading (lower panel). R, resting, uninfected cells.

Erk but not PI3K/Akt pathway is critical for the phagocytosis of Francisella. However

the binding of Francisella to the host macrophages was unaffected (Figure 2.9B). This

observation is consistent with the unchanged pathogen-host attachment when Syk is

inhibited (Figure 2.4B). We next examined the role of Erk in the phagocytosis of

Francisella using an overexpression system. Here, cells were transfected either with

vector or a GST-tagged Erk2 encoding construct. The transfectants were infected and

phagocytosis of Francisella was analyzed by CFU assays. Results shown in Figure 2.10A

indicate that GST-Erk2 transfected cells engulfed a significantly higher number of

bacteria than control transfectants. In parallel, the overexpression of GST-Erk2 protein

was verified by Western blotting (Figure 2.10B). Collectively these data demonstrate that

Erk but not PI3K/Akt is critical for the phagocytosis of Francisella by macrophages.

Syk-dependent increase in phagocytosis of Francisella is abrogated by Erk

inhibition

We next examined if the enhancement of Francisella phagocytosis due to Syk overexpression can be inhibited by blocking the downstream Syk-effector Erk. To test this, vector or Syk transfectants were incubated with vehicle control, Erk inhibitor or

PI3K/Akt inhibitor, subsequently infected and the phagocytosis of Francisella was examined by CFU assay. The results shown in Figure 2.11A indicate that the increase in phagocytosis due to Syk overexpression was lost when the Erk pathway but not

PI3K/Aktpathway was inhibited. In parallel experiments, the effectiveness of the inhibitors and Syk over-expression were tested by Western blotting analysis (Figures

2.11B and C).

A 180 160 140 120 100 80 * 60

Phagocytic index 40 20 0 DMSO UO LY

B 160 140 120

100

80 60 Binding index 40 20

0 DMSO UO LY

Figure 2.9: Erk but not PI3K/Akt pathway is required for the phagocytosis of Francisella. A&B. RAW 264.7 cells were treated with vehicle control (0.2% DMSO) or 2.5 μΜ UO126 (UO) or 20 μΜ of LY294002 (LY) for 30 minutes; cells were infected with 100 MOI of F. novicida for 1 hour and the phagocytosis (A) or adherence (B) of Francisella to macrophages was examined by microscopy assay. Data represent mean and standard deviation of 3 independent experiments. *, p<0.05 compared to DMSO value.

* 70 * 60 )

4 50

40 30 10 CFU (X 20 10 0 Vector GST-Erk

B Vector GST-Erk

GST-Erk

NS Endogenous Erk

Immunoblot: Anti-Erk

Immunoblot: Anti-Actin

Figure 2.10: Erk overexpression enhances the phagocytosis of Francisella. RAW 264.7 cells were transfected with vector or GST-Erk2 encoding construct, ~16 hours post-tranfection cells were infected with 100 MOI of F. novicida for 1 hour and the phagocytosis of Francisella was assessed by CFU assay. Data represent mean and standard deviation of 3 independent experiments. *, p<0.05 compared to vector value. B. Vector and GST-Erk2 transfectants were lysed, protein-matched lysates were resolved by SDS PAGE and the expression of Erk was analyzed by Western blotting with Erk antibody (upper panel). The same membrane was reprobed with Actin antibody (lower panel). Data represents a representative experiment of three independent experiments. NS, non-specific band.

A B

R DMSO UO LY 20 * 16 * ) 4 12 Vector Immunoblot: Anti-pErk1/2 8 Syk 4 CFU (X 10 CFU (X Immunoblot: Anti-pSer Akt 0 DMSO UO LY

Immunoblot: Anti-Actin

C Vector Syk

Immunoblot: Anti-Syk

Immunoblot: Anti-Actin

Figure 2.11: Syk-dependent increase in the phagocytosis of Francisella is abrogated by Erk inhibition. A. Vector and Syk transfectants were treated with 0.2% DMSO or 2.5 μΜ UO126 (UO) or 20 μΜ of LY294002 (LY) for 30 minutes; cells were infected with 100 MOI of F. novicida for 1 hour and the phagocytosis of F. novicida was examined by CFU assay. Data represent mean and standard deviations of three independent experiments. *, p<0.05, compared to corresponding vector transfectants. B. RAW 264.7 cells were treated with vehicle control or inhibitors (as described above), infected for 1 hour and protein-matched lysates were analyzed by Western blotting with phospho- specific antibodies (upper and middle panels). The same membrane was reprobed with Actin antibody to ensure equal loading (lower panel). R, resting, uninfected cells. C. Vector and Syk transfectants were lysed, protein-matched lysates were resolved by SDS PAGE and the expression of Syk was analyzed by Western blotting with Syk antibody (upper panel). The same membrane was reprobed with Actin antibody (lower panel). Data represents a representative experiment of three independent experiments.

The Erk-dependent increase in phagocytosis of Francisella is abrogated by Syk inhibition Finally, we tested whether the Erk-dependent phagocytosis of Francisella is suppressed if signaling through upstream Syk is inhibited. For this, vector- or GST-Erk2- transfected cells were treated with DMSO or the Syk inhibitor piceatannol. Cells were subsequently infected and phagocytosis was measured by CFU assays (Figure 2.12A).

Erk2 overexpression significantly increased the ingestion of Francisella, but treatment of

Erk2-overexpressing cells with Syk inhibitor significantly dampened the phagocytosis of

Francisella to control levels. In parallel experiments, the effect of Syk inhibition on the

phosphorylation status of both exogenous and endogenous Erk was examined by Western

blotting with phospho-Erk antibody. Results shown in Figure 2.12B indicate that

piceatannol inhibited phosphorylation of both exogenous and endogenous Erk proteins.

Syk is required for Francisella-induced cytokine production

Next, we examined the role of Syk tyrosine kinase in the inflammatory response

induced by Francisella. To test this, both RAW 264.7 cells and BMMs were treated with

DMSO or Syk inhibitor before infection with 100 MOI of F. novicida for 8 hours. Post

infection, cell supernatants were analyzed for TNFα and IL-10 production by ELISA.

Results displayed in the Figure 2.13 show significant reduction in Francisella-induced

TNFα and IL-10 production in both RAW 264.7 cells (Figure 2.13A) and in the primary

cells, BMMs (Figure 2.13B) upon inhibition of Syk activation. To verify these findings

using an additional approach, TNFα and IL-10 production by infected vector and Syk

transfectants was assessed. Here, we observed that Syk tranfectants produced

A 18 * 16 14

) 12

4 10 Vector 8 GST-Er k 2 6 CFU (X 10 4 2 0

DMSO Pice.

B Vector GST-Erk

Pice DMSO R UO Pice DMSO UO R

Immunoblot: Anti-pErk1/2 Endogenous Erk

Immunoblot: Anti-Erk

Immunoblot: Anti-pErk1/2 GST-Erk

Immunoblot: Anti-Erk

Figure 2.12: Erk-dependent increase in the phagocytosis of Francisella is abrogated by Syk inhibition. A. Vector and GST-Erk2 transfectants were treated with 0.1% DMSO or 25 μg/mL piceatannol for 30 minutes; cells were infected with 100 MOI of F. novicida for 1 hour and the phagocytosis of F. novicida was examined by CFU assay. Data represent mean and standard deviations of three independent experiments. *, p<0.05, compared to corresponding vector transfectants. B. RAW 264.7 cells were treated with DMSO or piceatannol (as described above), infected for 15 minutes and protein-matched lysates were analyzed by Western blotting with phospho-Erk antibody (upper panel). The same membrane was reprobed with Erk antibody to test the over-expression of Erk2 (lower panel). R, resting, uninfected cells. Data are representative of 2 independent experiments.

significantly enhanced amounts of TNFα and IL-10 than control tranfected cells

indicating that Syk promotes Francisella-induced cytokine response. In parallel

experiments, the overexpression of Syk was also verified (Figure 2.14B). Also, Syk

inhibition markedly suppressed the production of IL-12, IL-6 and RANTES in BMMs,

collectively, suggesting that Syk is critical for the inflammatory response in general (data

not shown).

Erk and PI3K/Akt differentially regulate Francisella-induced TNFα and IL-10

production

Since Syk is upstream of Erk and PI3K/Akt pathways we wanted to examine the

role of these pathways in Francisella-induced TNFα and IL-10 generation. For this,

RAW264.7 cells were treated with vehicle control or the corresponding inhibitors before

infection with 100 MOI of F. novicida for 8 hours. When the signaling through Erk is blocked there is a significant reduction in both Francisella-induced TNFα and IL-10 production (Figure 2.15A). Pharmacological inhibition of PI3K/Akt pathway suppressed

TNFα to a certain extent but in contrast to Erk inhibition, it significantly enhanced

Francisella-induced IL-10 production (Figure 2.15B) indicating that PI3K/Akt pathway

negatively regulates Francisella-induced IL-10 production. Similarly, we found that inhibition of Akt significantly dampened TNFα production but enhanced the generation of Francisella-induced IL-1071. Collectively, this suggests that Erk and PI3K/Akt pathways differentially regulate Francisella-induced TNFα and IL-10 production.

7000 80 α 6000 70 5000 60 4000 50 3000 40 2000 * 30 pg/ml TNF 20 * 1000 IL-10 pg/ml 10 0 0

R ol l n R n no DMSO a MSO n at D e ata Pic ice P

100 5000 90 80

α 4000 70 60 * 3000 50 2000 40 pg/ml TNF pg/ml IL-10 pg/ml 30 1000 20 * 10 0 0 R DMSO Piceatannol R DMSO Piceatannol

Figure 2.13: Inhibition of Syk suppresses Francisella-induced TNFα and IL-10. A&B. RAW 264.7 (A) or BMMs (B) are pre-treated with 0.1% DMSO or 25µg/ml piceatannol for 30 minutes before infection with 100 MOI of F. novicida for 8 hours. Post infection, cell supernatants were analyzed for TNFα and IL-10 by ELISA. R, resting, uninfected cells. Data are representative of 3 independent experiments. *, p<0.05.

8000 * 250 7000 * 200

α 6000 5000 R 150 R 4000 8 h 8 h

pg/ml TNF 3000 100 pg/ml IL-10 2000 50 1000 0 0 Vector Syk WT Vector Syk WT

B Vector Syk

Immunoblot: Anti-Syk

Immunoblot: Anti-Actin

Figure 2.14: Overexpression of Syk enhances Francisella-induced TNFα and IL-10. A. Vector and Syk transfectants were infected with 100 MOI of F. novicida for 8 hours. Post infection, cell supernatants were analyzed for TNFα and IL-10 by ELISA. *, p<0.05. B. Vector and Syk transfectants were lysed 16 hours after transfection and the expression of Syk was studied by Western blotting with Syk antibody. R, resting, uninfected cells. Data are representative of 3 independent experiments.

16000 14000 200 α 12000 10000 150 8000 * 100

pg/ml TNF 6000 4000 * pg/ml IL-10 50 2000 0 0 RDMSOUO RDMSOUO

16000 14000 350 * α 12000 300 10000 * 250 8000 200 6000 150 pg/ml TNF

4000 pg/ml IL-10 100 2000 50 0 0 RDMSOLy RDMSOLY

Figure 2.15: Erk and PI3K/Akt pathways differentially regulate Francisella-induced TNFα and IL-10. A&B. RAW 264.7 are pre-treated with 0.1% DMSO or 2.5µM UO126 (A) or 20µM LY294002 (LY) (B) for 30 minutes before infection with 100 MOI of F. novicida for 8 hours. Post infection, cell supernatants were analyzed for TNFα and IL-10 by ELISA. R, resting, uninfected cells. Data are representative of 3 independent experiments. *, p<0.05.

Syk promotes Francisella-induced NFκB activation through Erk and PI3K/Akt

pathway activation

We next investigated the molecular mechanism by which Erk and PI3K/Akt

pathways influence F. novicida-induced cytokine response. It has been previously reported that F. novicida induces the activation of NFκB63, and that activation of this

transcription factor may play an important role in the secretion of inflammatory

mediators. Thus, we hypothesized that Syk may promote Francisella-induced NFκB

activation that is necessary for cytokine production via the activation of Erk and

PI3K/Akt pathways. For this, we first transiently transfected RAW 264.7 cells with a

construct encoding luciferase reporter gene under the influence of NFκB binding (NFκB-

luciferase construct). Sixteen hours post-transfection, cells were infected with F. novicida

for the time points indicated in Figure 2.16A, and luciferase activity in the cell lysate was

determined. Results indicated that NFκB is activated in response to F. novicida infection.

Second, to test whether Erk and PI3K/Akt pathways are required for F. novicida-

induced activation of NFκB, transfected cells were treated with corresponding

pharmacological inhibitors and expression of luciferase enzyme in response to F.

novicida infection was measured. The results are shown in Figure 2.16B. Treatment of

cells with an inhibitor of NFκB SN50 served as our positive control. Inhibition of either

Erk or PI3K significantly decreased expression of the NFκB-driven luciferase in response

to F. novicida infection, suggesting that the activation of Erk and PI3K pathways may be

necessary for NFκB activation.

A B

25000 200000 20000 150000 15000 100000 RLUs

RLUs 10000 50000 *

0 5000 R 30' 2h 5h * * Time 0 RDMSOUOLYSN50

C 12000 10000

8000 6000 RLUs 4000 2000 * 0

l R O o MS nn D ta ea ic P

Figure 2.16: Syk, Erk and PI3K/Akt pathways promote Francisella-induced NFκB activation. A. RAW 264.7 cells transfected with plasmid encoding the luciferase gene driven by an NFκB binding element (NFκB-luc). Transfectants were infected with F. novicida for varying time points and analyzed for the luciferase activity as a measure of NFκB activation. B&C. NFκB-luc transfected RAW 264.7 cells were treated with 0.2% DMSO or 2.5µM UO126 or 20µM LY294002 (B) or 25 μg/mL piceatannol for 30 minutes (C). Cells were subsequently infected with 100 MOI of F. novicida for 5 hours and the luciferase activity in the lysates was measured. Data represent mean and standard deviations of three independent experiments. *, p<0.05, compared to DMSO value. RLUs, relative light units. R, resting, uninfected cells.

Consistently, inhibition of Syk also dampened the activation of Francisella-induced

NFκB (Figure 2.16C). These data indicate that in addition to PI3K/Akt pathway, both Erk and Syk are required for Francisella-induced NFκB activation.

Third, we inhibited the activation of NFκB and subsequently infected murine

BMMs with F. novicida. Results displayed in the Figure 2.17 indicate significant inhibition of TNFα production upon NFκB inhibition. In contrast, inhibition of NFκB

significantly enhanced Francisella-induced IL-10 generation suggesting NFkB-

dependent mechanisms may suppress the production of Francisella-induced IL-10

production.

In summary, we have demonstrated that Syk is required for the activation of both

Erk and PI3K/Akt pathways (Figure 2.18). We further established that Syk/Erk axis is

critical for the phagocytosis of Francisella by macrophages and that Syk modulates the generation of Francisella-induced TNFα and IL-10 production by regulating its

downstream effectors Erk and PI3K/Akt pathways. Collectively, the data obtained in this

study demonstrate an indispensable role for Syk in the phagocytosis of Francisella and

cytokine response induced by this microbe via its regulatory influence on Erk and

PI3K/Akt pathways.

2.5 Discussion

Phagocytosis is one of the earliest host immune responses against a pathogen and

the mechanisms of phagocytosis differ between pathogens. The molecular mechanisms

modulating the uptake of Francisella are not understood. Our current study demonstrates

a crucial requirement for Syk activation in the engulfment of Francisella. These findings

A 5000 4500 4000 α 3500 3000 2500 * 2000 pg/ml TNF 1500 1000 500 0 R Control SN50

B 800 * 600

400

pg/ml IL-10 200

0 R Control SN50

Figure 2.17: NFκB differentially regulates Francisella-induced TNFα and IL-10. A&B. BMMs were pre-treated with 75 μg/ ml of NFκB inhibitor (SN50) before infection with F. novicida for 8 hours. Cell supernatants from uninfected and infected cells were analyzed by ELISA for TNFα (A) and IL-10 (B). The graph represents mean and SD of values obtained from three independent experiments. Data were analyzed by paired student’s t-test (* indicates p value < 0.05). R, resting, uninfected cells.

Francisella

is tos ocy hag P

Syk

Erk PI3K/Akt

NF-κB

IL-10 TNFα

Figure 2.18: Proposed model for the Syk-dependent regulation of Francisella- induced macrophage response. Syk is activated during Francisella infection and activates downstream Erk and PI3K/Akt pathways and thereby subsequent NFκB activation and pro-inflammatory cytokine production. On the other hand, Syk via Erk may promote the activation of an unknown transcription factor and subsequent transcription of IL-10 gene. Also, Syk via mediating the activation of Erk promotes the phagocytosis of Francisella.

are consistent with the widely accepted role of Syk in phagocytosis of various agents

such as opsonized zymosan and erythrocytes34,36,77. Attesting to the importance of Syk

during phagocytosis, inhibition of Syk by RNAi suppressed the reorganization of actin around the nascent phagosomes in HL60 cells fed with opsonized zymosan34. Likewise,

Syk-deficient macrophages displayed defective phagosomal closure77. But, Syk did not

influence binding of either C3bi-opsonized zymosan or IgG-coated erythrocytes77 to macrophages in these studies indicating that Syk is dispensable in particle binding.

The requirement of Erk activation for the phagocytosis of Francisella is

mechanistically consistent with several other studies. For example, Kugler et al.

demonstrated that overexpression of an Erk phosphatase, MKP-1, significantly decreased

the phagocytosis of Listeria monocytogenes by macrophages indicating that Erk is critical

for the phagocytosis of this organism78. Further, Erk was reported to be critical for the

phagocytosis of IgG-coated erythrocytes by macrophages79. Both Erk and Syk were also found to be essential for the phagocytosis of E. coli by haemocytes of Manduca Sextata, a lepidopteran insect80. However, Erk was found to be dispensable for the phagocytosis by monocytes and microglia suggesting that Erk requirement for phagocytosis may differ depending upon the target and the immune cell type81.

Although Erk is reported to be critical, its precise role in the phagocytosis of

either pathogenic organisms or IgG-coated erythrocytes is poorly understood. It is

possible that Erk regulates phagocytosis via indirect and direct mechanisms. Though less

likely, Erk may induce the transcriptional synthesis of proteins involved in the

engulfment of Francisella. The more conceivable explanation for Erk-dependent

phagocytosis is that Erk, similar to Syk, may regulate the actin cytoskeletal

rearrangements that are crucial for the uptake of Francisella. Several pieces of

information support this hypothesis. First, MAPKs are reported to be important for the

82 activation of phoshpolipaseA2 which via profilin can modulate the actin cytoskeleton .

Second, Erk2 phosphorylation is reported to essential for the polarization of the microtubule organizing center in natural killer cells83. Third, a recent study demonstrated

that leukocyte specific protein-1, an F-actin binding cytoskeletal protein, interacts with

Erk2 and targets it to peripheral actin filaments84.

For the first time we have observed that PI3K/Akt pathway is activated during

Francisella infection. It is noteworthy that the uptake of Francisella is not influenced by

inhibition of the PI3K pathway. Although, the activation of PI3K has been shown to be

important for phagocytosis initiated through different receptors (such as FcγR15,

complement receptor12 and CD4485), PI3K activation has been found to be dispensable

for the phagocytosis of other intra-cellular pathogens such as Salmonella, Shigella, and

Type II Helicobacter pylori18-20. This reiterates the observation that intracellular signaling

events that regulate the functional responses of host cells are dependent upon the cellular

context and the nature of the stimulus.

Since the Syk/Erk axis is important for the phagocytosis of Francisella, how Syk activates Erk deserves attention. Syk is able to activate Erk through at least two independent pathways. First, Syk phosphorylates the Ras adapter Shc which can then associate with the Grb2/SoS complex, leading to the activation of Ras, an upstream activator of Erk86. Second, autophosphorylation of Syk can result in the activation of

protein kinase C (PKC), ultimately leading to the phosphorylation of Erk. Specifically,

using constitutively active and dominant-negative mutants of PKCδ, a key component in

the phagocytic pathway of neutrophils, Ueda et al have shown that PKCδ mediates

phorbol ester-induced activation of Erk in a Raf and MEK dependent but Ras-

independent manner87. We are currently testing the precise mechanism by which Erk is activated during Francisella infection.

Further, our data demonstrates that Syk is required for the Francisella-induced

TNFα and IL-10 production. We also found that Syk is required for the production of

Francisella-induced IL-12, IL-6, RANTES and IL-1β suggesting that Syk similar to

TLR2 may be required for Francisella-induced inflammatory response in general (data not shown). These findings are consistent with other studies which demonstrate a critical role for Syk in the regulation of cytokine production induced by immune complexes or

LPS88,89. Preliminary data also suggests that Syk is required for the induction of iNOS, which is important for innate immune responses against invading pathogens. This result is consistent with other studies90. In the future, the molecular mechanism underlying Syk- dependent induction of iNOS must be thoroughly examined. Therefore, Syk may be

important in regulating additional aspects of the innate immune response to Francisella.

In summary, we have identified that Syk regulates the activation of Erk and

PI3K/Akt, and we further demonstrate that Syk regulates phagocytosis and inflammatory response via the modulation of these two pathways. To our knowledge, this is the first study that reports the identity of 1) the upstream signaling events leading to the activation of the Erk and PI3K/Akt pathways and 2) the regulatory mechanisms that modulate the phagocytosis of Francisella.

In this chapter, we defined intracellular signaling pathways that promote host cell

inflammatory response to infection. Negative regulators such as SHIP1 have been shown

to temper inflammatory cytokine production by various immune cells. Thus, in the next chapter, the role of SHIP1 in modulating Francisella-induced inflammatory response was examined.

CHAPTER 3

MACROPHAGE INFLAMMATORY RESPONSE TO Francisella

novicida INFECTION IS REGULATED BY THE SH2 DOMAIN-

CONTAINING INOSITOL 5’ PHOSPHATASE SHIP1

3.1 Abstract

In the previous chapter we described how Syk promotes Francisella phagocytosis and Francisella-induced cytokine response. Although, the production of inflammatory mediators is critical for the resolution of infection, but this process must be tightly

regulated to prevent collateral damage. Thus, in this chapter, we examined the role of

SHIP1, an established negative regulator of immune responses, during Francisella infection. Herein we report that the inositol phosphatase SHIP1 is phosphorylated upon infection of primary murine macrophages with the genetically related F. novicida, and negatively regulates F. novicida-induced cytokine production. Analyses of the molecular details indicated that in addition to the activation of the MAP Kinases, PI3K/Akt pathway is also activated in Francisella novicida infected cells. Interestingly, SHIP1-deficient macrophages displayed enhanced Akt activation upon F. novicida infection, suggesting elevated PI3K-dependent activation pathways in the absence of SHIP1. Inhibition of

PI3K/Akt pathway resulted in suppression of F. novicida-induced cytokine production

through the inhibition of NFκB. Consistently, SHIP1 deficient macrophages displayed

enhanced NFκB-driven gene transcription, whereas overexpression of SHIP1 led to

decreased NFκB activation. Thus, we propose that SHIP1 negatively regulates F. novicida-induced inflammatory cytokine response by antagonizing the PI3K/Akt pathway and suppressing subsequent NFκB-mediated gene transcription. A detailed analysis of

phosphoinositide signaling may provide valuable clues for better understanding the

pathogenesis of tularemia.

3.2 Introduction

Francisella tularensis, the etiological agent of the zoonotic disease tularemia, is a

gram-negative intracellular pathogen. There are four different sub-species of Francisella

tularensis. The F. tularensis sub-species tularensis is the most virulent of the four with an

LD50 less than 10 colony forming units for humans. Other less virulent sub-species of F.

tularensis include novicida, holarctica and mediasiatica53.

Francisella tularensis principally infects immune cells such as monocytes,

macrophages and neutrophils. The survival strategy adopted by F. tularensis in the host

cell is to evade phago-lysosomal fusion. After a few hours of phagocytosis of the

organism, the membrane of phagosome ruptures and F. tularensis is released into the host

cell cytosol53. The release of the pathogen into the cytosol requires expression of bacterial

proteins such as IglC and MglA91,92. On the other hand, bacterial escape is suppressed by

IFNγ treatment of the cells infected with F. tularensis68.

The host cell responses to F. tularensis infection are poorly understood. IL-12,

TNFα and IFNγ have been reported to be critical for immunity against Francisella

infection66. While NK cells are thought to be the major source of IFNγ, IL-12 is produced

by infected macrophages67. Further, IL-12 also appears to strongly induce IFNγ

production. Attesting to the importance of IL-12 in immunity against F. tularensis LVS

infection, IL-12 knockout mice, or mice treated with IL-12 neutralizing antibodies are

unable to clear the bacteria64. Macrophages also produce several other pro-inflammatory

cytokines/chemokines upon infection. Intracellular signaling molecules involved in

regulating macrophage response to Francisella are not well defined. Although pathways

involving NFκB and the MAPKs p38 and JNK have been reported to be activated during

F. tularensis LVS infection, the exact role of these signaling pathways in macrophage

responses is still not clear63.

The SH2 domain-containing inositol 5’ phosphatase SHIP1 is a hematopoietic cell-specific phosphatase that negatively regulates PI3K pathway by consuming the lipid products of PI3K. SHIP1 is a constitutively active enzyme residing in cytosol. Membrane translocation of SHIP1 is required for access to its substrates. Upon stimulation of

immunoreceptors, growth factor/cytokine receptors or toll-like receptors, SHIP1

translocates to the membrane where it is phosphorylated by membrane-associated Src

kinases39. In addition to the central catalytic domain, SHIP1 also consists of an N- terminal SH2 domain, a C-terminal proline-rich domain and two NPXY motifs, which can all associate with other cellular signaling proteins. Thus, SHIP1 mediates its

functions by both its catalytic and protein-protein interaction domains. For example, the

catalytic domain of SHIP1 regulates cellular responses by hydrolyzing PI3,4,5P3 into 65

39 PI3,4P2, thus opposing the PI3K pathway . SHIP1 is also reported to suppress activation

of the Ras pathway, in some cases, by virtue of its interaction with Dok, a Ras GAP activator, and/or its interactions with the Ras adapter Shc93,94.

Herein, we investigated the role of SHIP1 in the regulation of F. novicida-induced

cytokine response in murine macrophages. We report that SHIP1 negatively regulates the production of IL-12, IL-6 and RANTES by F. novicida-infected macrophages. Our

current studies indicate that the production of these cytokines requires the activation of

the PI3K pathway and involves NFκB activation. Further, these studies demonstrate that

F. novicida induced pro-inflammatory cytokine production is negatively regulated by

SHIP1 by antagonizing the PI3K/Akt pathway and NFκB-driven transcriptional

activation. Thus, we conclude that SHIP1 is a regulator of macrophage innate immune responses to F. novicida infection.

3.3 Materials and Methods

Cells, antibodies and reagents: RAW 264.7 murine macrophage cells were obtained from

ATCC and maintained in RPMI with 3.5% heat-inactivated fetal bovine serum (FBS).

Antibodies specific for phospho-Erk, phospho-JNK, phospho-SHIP1, phospho-Akt and

phospho-p38 were purchased from Cell Signaling Technology (Beverly, MA). Actin,

phosphotyrosine and Akt antibodies were from Santa Cruz Biotechnology (Santa Cruz,

CA). Rabbit polyclonal SHIP1 antibody was a generous gift from Dr. K. Mark

Coggeshall (OMRF, Oklahoma City, OK). F. novicida U112 (JSG1819) was used in all

experiments. Bacteria were grown on Chocolate II agar plates at 370C.

Culture of murine bone marrow macrophages: SHIP1+/- animals were generously provided by Dr. G. Krystal (B.C. Cancer Agency, Vancouver, BC, Canada).

Heterozygotes were bred to obtain SHIP1+/+ and SHIP1-/- mice. Bone marrow

macrophages (BMMs) were derived from these animals as previously described8.

Briefly, bone marrow cells were cultured in RPMI containing 10% fetal bovine serum plus 10 μg/ml polymixin B and supplemented with 20ng/ml CSF-1 for 7 days before they were used in experiments. BMMs derived in this manner were >99% positive for Mac-1, as determined by flow cytometry.

Cell stimulation, lysis, and Western blotting: Bone marrow macrophages were plated in 6 well culture dishes. Macrophages were infected either with F. novicida (at 1 or 10 or 100

MOI) or with Francisella tularensis LVS (100 MOI) that were scraped from Chocolate II agar plates, resuspended and diluted in RPMI. For some experiments, bacteria were killed by heat at 980C for 10 minutes prior to adding to the macrophages. To prevent internalization of bacteria, cells were treated with 5 μg/ml of cytochalasin D for 30 min at

0 37 C 5 % C02 prior to infection. Uninfected and infected cells were lysed in TN1 buffer

(50mM Tris pH 8.0, 10mM EDTA, 10mM Na4P2O7, 10mM NaF, 1% Triton-X 100,

125mM NaCl, 10mM Na3VO4, 10μg/ml each aprotinin and leupeptin). Post-nuclear lysates were boiled in Laemmli Sample Buffer and were separated by SDS/PAGE, transferred to nitrocellulose filters, probed with the antibody of interest and developed by enhanced chemiluminescence (ECL).

Colony Forming Unit Assays: SHIP1+/+ and SHIP1-/- BMMs were infected with F.

novicida (100 MOI). Two hours post-infection, cells were washed two times and 67

0 incubated with 50 μg/ml of gentamicin for 30 min at 37 C and 5 % CO2. The cells were

subsequently washed twice and lysed in 0.1 % SDS for 5 minutes. Immediately, 10 fold

serial dilutions were made and appropriate dilutions were plated on Chocolate II agar plates. Assays were performed in triplicate for each test group.

Transmission Electron Microscopy (TEM): SHIP1+/+ and SHIP1-/- BMMs were infected

with F. novicida, and 2 hours post-infection the cells were washed, fixed and prepared for

TEM analysis as described previously95. Bacterial count in 20 SHIP1+/+ and SHIP1-/-

BMMs was assessed and averaged.

ELISA measurement of cytokine production: BMMs were infected with F. novicida for varying time points. Cell supernatants were harvested, centrifuged to remove dead cells and analyzed by ELISA using cytokine specific kits from R & D Systems (Minneapolis,

MN). Data were analyzed using a paired Student’s t-test. p value < 0.05 was considered as significant.

Transfection and Luciferase Assays: BMMs and RAW 264.7 cells were transfected with

the appropriate plasmid DNA using the Amaxa Nucleofector apparatus (Amaxa

biosystems, Germany) as previously described46. Briefly, 5x106 cells were resuspended in

100 μl Nucelofector Solution (T for BMMs and V for RAW 264.7 cells) and were

nucelofected with 1 ug of NFκB-luc alone or with 5 ug of empty vector or plasmid

encoding WT-SHIP1. Immediately post nucleofection, 500 μl of pre-warmed RPMI was

added to the transfection mix before transferring to 12-well plates containing 1.5 ml pre-

warmed RPMI per well. Plates were incubated for 12 hours at 37oC. Transfected cells 68

were either left uninfected or were infected for 5 hours with F. novicida. Cells were lysed in 100 μl of Luciferase Cell Culture Lysis Reagent (Promega). Luciferase activity was then measured using Luciferase Assay Reagent (Promega), as previously described46.

3.4 Results

SHIP1 is phosphorylated during Francisella infection

The inositol phosphatase SHIP1 plays a critical role in regulating macrophage

innate immune responses to IgG immune complexes and bacterial products. Recent

studies indicated that while SHIP1 negatively regulates TLR2 function45, it promotes

TLR4 function46. SHIP1 is a constitutively active, cytoplasmic enzyme that must

translocate to the membrane where it gains access to its substrate PI3,4,5P3. Membrane

translocation of SHIP1 is accompanied by tyrosine phosphorylation of SHIP1 by

membrane-associated Src kinases. Thus, tyrosine phosphorylation of SHIP1 is often used

as an indicator of SHIP1 translocation to the membrane. To investigate the role of SHIP1

in F. novicida-stimulated host cell response, RAW 264.7 murine macrophage cells were

infected with F. novicida, SHIP1 proteins were immunoprecipitated from uninfected and

infected cells and analyzed by Western blotting with anti-phosphotyrosine antibody

(Figure 3.1A, upper panel). To ensure equal loading the same membrane was reprobed

with SHIP1 antibody (lower panel). Results indicate that SHIP1 is tyrosine

phosphorylated in cells infected with F. novicida, and that robust phosphorylation of

SHIP1 occurs at the later time points (30 and 60 minutes post infection). Having

determined the time course of SHIP1 phosphorylation in Raw 264.7 cells, we next tested

whether these findings could be validated in primary cells. Here, murine BMMs were 69

infected with F. novicida for varying time points and tyrosine phosphorylation of SHIP1

was analyzed by Western blotting with a phospho-specific SHIP1 antibody (Figure 3.1B,

upper panel). Results suggest that infection of BMMs by F. novicida induces tyrosine

phosphorylation of SHIP1. The same membrane was reprobed with anti-SHIP1 antibody

to ensure equal loading (Figure 3.1B, lower panel).

Next, we tested two aspects: 1) if viability of Francisella is important for SHIP1

phosphorylation 2) if SHIP1 is also phosphorylated by Francisella tularensis LVS, the vaccine strain of F. tularensis. For this, RAW 264.7 cells were infected with 100 MOI of live or dead F. novicida or live F. tularensis LVS for 30 minutes and the phosphorylation

of SHIP1 was assessed by Western blotting. The results are shown in the Figure 3.2.

Results obtained indicate that both live and killed Francisella induce SHIP1

phosphorylation. Further, SHIP1 was also phosphorylated when macrophages were

infected with F. tularensis LVS. Collectively, these results suggest a potential

involvement of SHIP1 in Francisella -induced macrophage signaling.

SHIP1 is dispensable for the phagocytosis of Francisella

Having established that SHIP1 is phosphorylated during Francisella infection, we

next wanted to examine the role of SHIP1 in Francisella infection. First, we tested the

role of SHIP1 in Francisella phagocytosis by colony forming unit (CFU) assays, and the

results are shown in Figure 3.3. For this, SHIP1 WT and SHIP1 KO BMMs were infected

for 1 hour and the internalization of the bacteria was assessed. Results shown in Figure

3.3 suggest that there is no significant difference between the engulfment of F. novicida

by SHIP1+/+ and SHIP1-/- BMMs.

A B

R 5’ 15’ 30’ 60’ Ctrl R 30’ 2h 5h 8h

Immunoprecipitation: Anti-SHIP 1 Immunoblot: Anti-pSHIP1 Immunoblot: Anti-pY

Immunoblot: Anti-SHIP1 Immunoblot: Anti-SHIP1

Figure 3.1: SHIP1 is phosphorylated during F. novicida infection. A. RAW 264.7 murine macrophage cells were infected with F. novicida (MOI = 100). SHIP1 immunoprecipitates from uninfected and infected cells were analyzed by Western blotting with anti-phosphotyrosine antibody (upper panel). The lower panel is a reprobe of the same membrane with anti-SHIP1 antibody. The lane marked ‘Ctrl’ represents immunoprecipitate with control antibody. B. BMMs were infected with F. novicida for the time points shown in the figure. Protein-matched lysates were separated by SDS/PAGE and analyzed by Western blotting with antibodies specific for phosphorylated SHIP1 (upper panel). The lower panel is a reprobe with SHIP1 antibody. R, resting, uninfected cells.

FN R HKFN LVS

Immunoprecipitation: Anti-SHIP 1

Immunoblot: Anti-pY

Immunoblot: Anti-SHIP1

Figure 3.2: SHIP1 is phosphorylated during Francisella infection. A. RAW 264.7 murine macrophage cells were infected with live (FN) or heat killed (HKFN) F. novicida or F. tularensis LVS (MOI = 100). SHIP1 immunoprecipitates from uninfected and infected cells were analyzed by Western blotting with anti-phosphotyrosine antibody (upper panel). The lower panel is a reprobe of the same membrane with anti-SHIP1 antibody. R, resting, uninfected cells.

Next, we tested the role of SHIP1 in the phagocytosis of F. novicida by TEM analysis

and representative images of this analysis are shown in the Figure 3.4. Results obtained

by TEM analysis supported CFU data: SHIP1+/+ and SHIP1-/- macrophages ingested 3.95

and 3.2 bacteria/cell, respectively.

SHIP1 negatively regulates Francisella-induced pro-inflammatory response

In order to further probe the functional consequence of SHIP1 activation, we

assessed the role of SHIP1 in Francisella-induced inflammatory response. Here, F.

novicida-induced cytokine production was compared in BMMs obtained from SHIP1+/+

and SHIP1-/- littermate mice. For this, SHIP1+/+ and SHIP1-/- BMMs were infected with F.

novicida, and cell supernatants from uninfected and infected cells were harvested 8 hours

post infection and analyzed by ELISA for IL-12, IL-6 and RANTES. As shown in Figure

3.5A, SHIP1-/- BMMs produced significantly increased levels of pro-inflammatory

molecules compared to their wild type counterparts. Protein-matched cell lysates from

SHIP1+/+ and SHIP1-/- BMMs were analyzed by Western blotting with SHIP1 antibody to confirm the genotype of these cells (Figure 3.5B). These data indicate that SHIP1 is a

negative regulator of IL-12, IL-6 and RANTES production by F. novicida infected

macrophages.

Next we tested if the regulatory influence of SHIP1 on Francisella-induced

cytokines is dependent on the MOI used. For this, SHIP1+/+ and SHIP1-/- BMMs were infected for 8 hours and the cell supernatants were analyzed by ELISA.

3 CFU 10

2 Log

0 SHIP1+/+ SHIP1-/-

Figure 3.3: SHIP1 is dispensable for the phagocytosis of F. novicida. BMMs from SHIP1 +/+ and SHIP1 -/- littermate mice were infected with 100 MOI of F. novicida for 1 hour, then treated with gentamicin (50 μg/ml), lysed in 0.1 % SDS and appropriate dilutions of the lysates were plated on Chocolate II agar plates for enumeration of CFUs. The graph represents mean and SD of values obtained from three independent experiments. Data were analyzed by unpaired student’s t-test and no significant difference was observed.

A B

F F M

Figure 3.4: Representative images of TEM analysis. BMMs from SHIP1 +/+ (A) and SHIP1-/- (B) littermate mice were infected with 100 MOI of F. novicida for 1 hour, then washed and processed for TEM analysis. The original magnification of the images was 28000 X. F, Francisella. M, mitochondria.

We have observed that the regulatory influence of SHIP1 on F. novicida-induced

IL-12, IL-6 and RANTES was also evident at different multiplicities of infection (Figure

3.6) suggesting that the negative regulatory effect of SHIP1 is independent of bacterial numbers infecting the macrophages. In addition, the viability of the bacteria is not essential for the negative influence of SHIP1 as macrophages infected with heat-killed bacteria displayed similar cytokine responses as those infected with live bacteria (Figure

3.7). Likewise, SHIP1 also downregulated Francisella tularensis LVS-induced IL-12, IL-

6 and RANTES production (Figure 3.7).

We next examined whether internalization of bacteria is a requisite for the

downregulatory influence of SHIP1 on F. novicida-induced inflammatory mediator production. For this, cells were treated with cytochalasin D, an actin polymerization inhibitor, prior to infection and cell supernatants were assayed for the secretion of IL-12,

IL-6 and RANTES by ELISA. The results are shown in Figure 3.7. The release of

RANTES and the negative influence of SHIP1 on RANTES production were not influenced by failure of internalization. However treatment of either SHIP1+/+ or SHIP1-/-

BMM with cytochalasin D significantly suppressed the release of IL-12 and IL-6

suggesting that either internalization of bacteria or actin cytoskeletal rearrangements are

essential for the secretion of these two cytokines.

A SHIP1+/+ -/- SHIP1 * 700 1000 600 * * 500 600

12 800 500 400 IL- 600 400 300 ml 400 300 200 pg/ 200 IL-6 pg/ml 200 100 pg/ml RANTES 100 0 0 0 RFN RFN RFN

B SHIP1+/+ SHIP1-/-

Immunoblot: Anti-SHIP1

Immunoblot: Anti-Actin

Figure 3.5: SHIP1 downregulates macrophage pro-inflammatory response to F. novicida infection. A. BMMs from SHIP1+/+ and SHIP1-/- littermate mice were infected for 8 hours with F. novicida. Cell supernatants from uninfected and infected cells were analyzed by ELISA for IL-12, IL-6 and RANTES. The graphs represents mean and SD of values obtained from three independent experiments. Data were analyzed by paired student’s t-test (* indicates p value < 0.05). R, resting, uninfected cells. B. Protein- matched lysates from SHIP1+/+ and SHIP1-/- BMMs were analyzed by Western blotting with SHIP1 antibody (upper panel). The same membrane was reprobed with actin antibody (lower panel).

SHIP1+/+ -/- SHIP1

2000 2500 * * 2000 1500 1500 * 1000 * 1000 * pg/ml IL-6 pg/ml IL-12 500 500 * 0 0 R 1 10 100 R 1 10 100 MOI MOI

1200 * 1000 * 800 600 * 400 200 pg/ml RANTES 0

R 1 10 100 MOI

Figure 3.6: SHIP1 downregulates macrophage pro-inflammatory response to F. novicida independent of MOI. BMMs from SHIP1 +/+ and SHIP1 -/- littermate mice were infected at different MOI (indicated in figure) of F. novicida for 8 hours. Cell supernatants from uninfected and infected cells were assayed for IL-12, IL-6 and RANTES by ELISA. Data represent mean and SD of three independent experiments. Data were analyzed by paired student’s t-test (* indicates p value < 0.05). R, resting, uninfected cells.

SHIP1+/+ -/- SHIP1

2000 3000 * * * 2500 * 1500 2000 1500 * 1000

pg/ml IL-12 pg/ml * 1000 pg/ml IL-6 500 500 0 0 RFNHKFNCyt DLVS RFNHKFNCyt DLVS

5000 * * 4000 * * 3000

2000

pg/ml RANTES 1000

0 RFNHKFNCyt DLVS

Figure 3.7: Analysis of SHIP1 influence on macrophage inflammatory response to F. novicida infection. BMMs from SHIP1 +/+ and SHIP1 -/- littermate mice were infected with 100 MOI of live or heat killed F. novicida or with Francisella tularensis LVS for 8 hours. Cyt D represents samples that were treated with cytochalasin D (5 μg/ml) before infection with 100 MOI of live F. novicida. Cell supernatants from uninfected and infected cells were analyzed by ELISA for IL-12, IL-6 and RANTES. Data represent mean and SD of three independent experiments. Data were analyzed by paired student’s t-test (* indicates p value < 0.05). R, resting, uninfected cells.

MAP kinase and the PI3K/Akt pathways are activated upon F. novicida infection of

macrophages

To examine the molecular mechanism by which SHIP1 downregulates pro-

inflammatory cytokine production by F. novicida infected macrophages, cell signaling events activated in the infected macrophages were first analyzed. Here, murine BMMs were infected with F. novicida for indicated time points and the activation of the MAP

Kinases Erk, p38 and JNK was analyzed by Western blotting protein-matched lysates

with phospho-specific antibodies (Figure 3.8A). All membranes were reprobed with

antibody to Actin to ensure equal loading of protein in all lanes. Results indicated that all

three MAPKs were activated in response to F. novicida infection as previously

reported63. The phosphorylation levels of MAPKs peaked at 30 minutes and gradually

diminished but were still persistent even at 8 hours post-infection. Consistent with the

results described in the Chapter 2 we found that the PI3K/Akt pathway is also activated

during F. novicida infection of primary mouse macrophages (BMMs).

SHIP1 regulates F. novicida-induced activation of the PI3K/Akt pathway

Having identified cellular signaling events activated by F. novicida infection, we

next examined the influence of SHIP1 on these events. For this, BMMs from SHIP1+/+ and SHIP1-/- littermate mice were infected with F. novicida for the indicated time points

and activation of MAPKs and Akt was analyzed. Results are displayed in the Figure 3.9.

Infection by F. novicida induced robust phosphorylation of Erk1/2, JNK, p38 and Akt.

However, SHIP1+/+ and SHIP1-/- BMMs showed comparable levels of the

phosphorylation of the MAPKs at all time points tested. On the other hand, SHIP1-/-

BMMs displayed enhanced phosphorylation of Akt compared to SHIP1+/+ BMMs (Figure

3.9). These results indicate that while SHIP1 may not regulate F. novicida-induced

activation of the MAPKs, SHIP1 downregulates activation of Akt. Together, these data suggest that the PI3K/Akt pathway may play a critical role in F. novicida-induced

macrophage pro-inflammatory responses.

Influence of PI3K pathway on F. novicida-induced macrophage inflammatory

response.

A role for the PI3K/Akt pathway in macrophage response to F. novicida infection

has been reported for the first time in this study (Figure 2.6). The above observations of

enhanced cytokine production and enhanced activation of Akt in SHIP1-/- BMMs suggest

a potential role for the PI3K/Akt pathway in F. novicida-induced IL-12, IL-6 and

RANTES production. To test this notion, we next treated cells with pharmacologic

inhibitor of PI3K, LY294002, for 30 minutes before infecting with 100 MOI of F.

novicida for 8 hours and monitored the production of IL-12, IL-6 and RANTES by

ELISA. The results are shown in Figure 3.10. Inhibition of the PI3K pathway

significantly decreased the production of IL-12 (p < 0.009), IL-6 (p < 0.003) and

RANTES (p < 0.03) by F. novicida infected BMMs indicating that PI3K/Akt pathway is

involved in the production of pro-inflammatory mediators in response to F. novicida

infection.

A B R 30’ 2h 5h 8h R 30’ 2h 5h 8h

Immunoblot: Anti-pErk Immunoblot: Anti-pSer Akt

Immunoblot: Anti-pp38 Immunoblot: Anti-Akt

Immunoblot: Anti-pJNK

Immunoblot: Anti-Actin

Figure 3.8: Activation of signaling pathways during F. novicida infection. BMMs were infected with F. novicida for the time points shown in the figure. A. Protein- matched lysates were analyzed by Western blotting with phospho-specific antibodies to Erk, p38 and JNK. The membranes were reprobed with Actin antibody to ensure equal loading of protein in all lanes. B. Protein-matched lysates were analyzed by Western blotting with phospho-specific antibody to Akt (upper panel). The lower panel is a reprobe of the same membrane with total Akt antibody. R, resting, uninfected cells.

A B

+/+ -/- SHIP1 SHIP1 SHIP1+/+ SHIP1-/-

R 30’ 2h 5h 8h R 30’ 2h 5h 8h R 30’ 2h 5h 8h R 30’ 2h 5h 8h

Immunoblot: Anti-pErk Immunoblot: Anti-pSer Akt

Immunoblot: Anti-pp38 Immunoblot: Anti-Akt

Immunoblot: Anti-SHIP1 Immunoblot: Anti-pJNK

Immunoblot: Anti-Actin

Figure 3.9: SHIP1 negatively regulates F. novicida-induced Akt activation. BMMs from SHIP1+/+ and SHIP1-/- littermate mice were infected for the time points indicated in the figure. A. Protein-matched lysates were probed with antibodies specific for phosphorylated Erk, p38 and JNK. All membranes were reprobed with Actin antibody. B. Protein-matched lysates were probed with phospho-specific antibody for Akt (upper panel). The membranes were reprobed with total Akt antibody (middle panel), and with SHIP1 antibody (lower panel). These data are representative of three independent experiments. R, resting, uninfected cells.

3000 1400 2500 1200 1000 2000 * 800 1500 600 pg/ml IL-12 1000 pg/ml IL-6 400 * 500 200 0 0 RDMSOLY RDMSOLY

1400 1200 1000

800 *

600 400

pg/ml RANTES 200 0 RDMSOLY

Figure 3.10: PI3K activation promotes F. novicida induced macrophage inflammatory response. BMMs were pretreated for 30 minutes with PI3K inhibitor LY294002 (20uM), or with vehicle control (DMSO), and subsequently infected with F. novicida. Cell supernatants were harvested 8 hours post infection and assayed by ELISA for IL-12, IL-6 and RANTES. The graphs represent mean and SD of values obtained from three independent experiments. Data were analyzed by paired student’s t-test (* indicates p value < 0.05). R, resting, uninfected cells.

The PI3K/Akt pathway promotes macrophage inflammatory response to F. novicida through its influence on downstream NFκB activation.

We next investigated the molecular mechanism by which PI3K/Akt influences F.

novicida-induced cytokine response. Since (i) F. novicida induces the activation of NFκB

in a PI3K-depedent manner (Figure 2.16) and (ii) PI3K is required for the production of

pro-inflammatory cytokines (Figure 3.10), we have hypothesized that the activation of

this transcription factor may play an important role in the secretion of pro-inflammatory

cytokines. To test this, we inhibited the activation of NFκB using the pharmacologic

inhibitor SN50 and subsequently infected murine BMMs with F. novicida. Results

displayed in the Figure 3.11A indicate significant inhibition of IL-12 (p < 0.009), IL-6 (p

< 0.007) and RANTES (p < 0.02) production upon NFκB inhibition (Figure 3.11A). The

Western blots shown in Figure 3.11B demonstrate that while treatment of cells with the

PI3K inhibitor completely inhibited downstream Akt phosphorylation, Akt

phosphorylation was not influenced by the NFκB inhibitor SN50 indicating that SN50

inhibition has occurred downstream of Akt. Taken together, these data provide evidence

that PI3K influences F. novicida-induced macrophage pro-inflammatory response

through its influence on NFκB activation. Of note, the activation of NFκB is unaffected

in macrophages pre-treated with cytochalasin D, suggesting that internalization of F.

novicida is not required (Figure 3.12).

3000 500 2000 2500 400 1500 2000 300 1500 1000 200 1000 pg/ml IL-12 pg/ml IL-6 pg/ml 100 500

500 pg/ml RANTES * * * 0 0 0 R8h8h SN R8h8h SN R8h8h SN

B DMSO LY R SN

Immunoblot: Anti-pSer-Akt

Immunoblot: Anti-Akt

Figure 3.11: NFκB activation is required for the production of F. novicida-induced IL-12, IL-6 and RANTES. A. BMMs were pretreated for 30 minutes with either DMSO or the NFκB inhibitor SN50, and subsequently infected with F. novicida. Cell supernatants were harvested 8 hours post infection and assayed for IL-12, IL-6 and RANTES by ELISA. The graph represents mean and SD of values obtained from three experiments. Data were analyzed by paired student’s t-test (* indicates p value < 0.05). B. Protein-matched lysates from BMMs pretreated with inhibitors of PI3K or NFκB and infected with F. novicida were analyzed by Western blotting with phospho-specific antibody to Akt (upper panel). The lower panel is a reprobe of the same membrane with Akt antibody. R, resting, uninfected cells.

10000 8000 6000

RLUs 4000 2000 * 0

R SO M 5h SN 5h D 5h Cyt D

Figure 3.12: Surface contact may be sufficient to trigger F. novicida-induced NFκB activation. RAW 264.7 cells transfected with NFκB-luc plasmid were pretreated with either cytochalasin D or with vehicle control (DMSO), and subsequently infected with F. novicida. 5 hours post infection cells were lysed and assayed for luciferase activity. The data are representative of three independent experiments. RLUs, relative light units. R, resting, uninfected cells.

SHIP1 negatively regulates F. novicida induced NFκB activation

Requirement of PI3K activity for F. novicida-induced NFκB activation suggested

that SHIP1 may modulate F. novicida-stimulated NFκB activation. To test this prediction, RAW 264.7 cells were transfected with NFκB-luciferase construct alone along with either SHIP1 encoding construct or the empty vector. Transfectants were then infected with F. novicida, and luciferase activity in the cell lysates was measured as an indicator of NFκB activation. As anticipated, overexpression of SHIP1 significantly (p <

0.04) suppressed NFκB-dependent reporter gene expression (Figure 3.13A).

Overexpression of SHIP1 in the transfected cells was confirmed by Western blotting

SHIP1 immunoprecipitates with SHIP1 antibody (Figure 3.13B).

As an additional approach, SHIP1 +/+ and SHIP1 -/- BMMs were transiently

transfected with the NFκB-luciferase construct. Transfection of BMMs using the Amaxa nucleofector (Solution T, program T-20) yielded comparable transfection efficiency for

both SHIP1+/+ and SHIP1-/- cells, as determined with plasmid encoding EGFP (Figure

3.14A). Transfected BMMs were subsequently infected with F. novicida and assayed for

luciferase activity. BMMs deficient in SHIP1 displayed significantly higher levels (p <

0.05) of luciferase activity than the SHIP1+/+ BMMs, indicating that SHIP1 negatively

regulates F. novicida-induced activation of NFκB (Figure 3.14B).

A 500

400

300

200 *

100 0 Percent increaseover uninfected NFκB-luc+vector NFκB-luc+SHIP1

Vec. SHIP1 Ctrl

Immunoprecipitation: Anti-SHIP1

Immunoblot: Anti-SHIP1

Figure 3.13: SHIP1 overexpression downregulates F. novicida-induced NFκB activation. A. RAW 264.7 cells were transfected with NFκB-luc plasmid along with either empty vector (Vec.) or plasmid encoding SHIP1. Transfectants were infected with F. novicida and assayed for luciferase activity 5 hours post infection. B. SHIP1 immunoprecipitates from the RAW 264.7 transfectants were analyzed by Western blotting to verify the overexpression of SHIP1.

A SHIP1+/+

SHIP1-/-

300 * 250 200

150 100 50

Percent increase over resting SHIP1+/+ SHIP1-/-

Figure 3.14: Knockout of SHIP1 enhances F. novicida-induced NFκB activation. SHIP1+/+ and SHIP1-/- macrophages were transfected with construct encoding EGFP and images were taken 6 hours after transfection. The transfection efficiency in both these macrophages was approximately 25%. B. SHIP1+/+ and SHIP1-/- BMMs were transfected with NFκB-luc plasmid. Transfectants were infected with F. novicida and assayed for luciferase activity. The graph represents values from three independent experiments. Data were analyzed by student’s t-test (* indicates p value < 0.05).

SHIP1 positively regulates F. novicida-induced IL- 10 production.

Since PI3K antagonizes Francisella-induced IL-10 (Figure 2.15), we anticipated

that SHIP1 may promote IL-10 production. To test this, SHIP1+/+ and SHIP1-/- BMMs

were infected with F. novicida for 8 hours. Cell supernatants from uninfected and

infected cells were analyzed by ELISA for IL-10. The results shown in Figure 3.15

demonstrate that SHIP1+/+ BMMs produce significantly higher levels of IL-10 than the

SHIP1-/- cells.

Collectively the data obtained in this study demonstrate that SHIP1 regulates the

inflammatory response induced by Francisella by dampening the activation of PI3K/Akt

pathway and subsequent NFκB activation.

3.5 Discussion

Early protection against F. tularensis is dependent upon the production of IFNγ,

TNF-α and IL-12, all of which are produced within a day after infection65,96-98. Before the

start of this investigation, regulatory mechanisms controlling the production of pro-

inflammatory molecules have not been established. SHIP1 is a critical regulator of

hematopoietic cell functions; hence we examined the role of SHIP1 in F. novicida-

induced inflammatory response. Our data demonstrate that SHIP1 is a negative regulator

of Francisella-induced IL-12, IL-6 and RANTES production by BMMs.

SHIP1 has been shown to influence hematopoietic cell functions through its

inhibitory effect on the PI3K pathway as well as the MAPK pathways. A model describing the role of SHIP1 in Francisella-induced inflammatory response is presented in the Figure 3.16. Since previous studies and our current studies have demonstrated the 91

SHIP1+/+ -/- 600 SHIP1 500

400

300 * 200

pg/ml IL-10 100 0 RFN

Figure 3.15: SHIP1 positively regulates F. novicida-induced IL-10. BMMs from SHIP1+/+ and SHIP1-/- littermate mice were infected for 8 hours with F. novicida. Cell supernatants from uninfected and infected cells were analyzed by ELISA for IL-10. The graph represents mean and SD of values obtained from three independent experiments. Data were analyzed by paired student’s t-test (* indicates p value < 0.05). Data were analyzed by paired student’s t-test (* indicates p value < 0.05). R, resting, uninfected cells.

Francisella

MAPKs SHIP1 PI3K/Akt

NF-κB

IL-12, IL-6 IL-10 and RANTES

Figure 3.16: Proposed model for SHIP1-dependent negative regulation of F. novicida-induced IL-12, IL-6 and RANTES. SHIP1 negatively regulates the production of F. novicida-induced IL-12, IL-6 and RANTES by dampening the PI3K/Akt pathway and subsequent NFκB activation that is important for the synthesis of IL-12, IL-6 and RANTES. On other hand, SHIP1 promotes IL-10 by antagonizing Francisella-induced NFκB activation.

activation of MAPKs in Francisella-infected macrophages, we examined whether SHIP1

influenced activation of the MAPKs. However, our studies indicate that the molecular

mechanism by which SHIP1 influences macrophage response to Francisella does not

involve inhibition of the MAPKs. Thus, we did not observe any significant differences in

the activation levels of MAPKs at any of the time points tested ranging from 5 minutes to

8 hours. However, we observed enhanced Akt activation in SHIP1-deficient macrophages

infected with Francisella. Thus, the data from Chapters 2 and 3 taken together indicate

that SHIP1 and Syk modulate PI3K pathway which may be critical in the regulation of

macrophage response to Francisella. Consistently, we observed that pharmacological

inhibition of PI3K attenuated F. novicida-induced activation of NFκB and the subsequent

production of pro-inflammatory cytokines. On the other hand, inhibition of PI3K activity

enhanced anti-inflammatory IL-10 production suggesting PI3K may modulate the

balance between the levels of pro-inflammatory and anti-inflammatory cytokines.

Activation of the PI3K/Akt pathway by Francisella was reported for the first time in this study. However, other intracellular pathogens such as Salmonella enterica have

been shown to modulate host cell phosphoinositide pathways and downstream Akt

activation19. In the latter case the functional consequence of phosphoinositide signaling is unclear. Interestingly, SHIP1 did not influence the phagocytosis of F. novicida suggesting PI3K/Akt pathway may not be involved in the engulfment of the pathogen.

These results are consistent with the findings from Chapter 2 which indicate that

PI3K/Akt pathway is dispensable for the phagocytosis of Francisella. Such dispensability of PI3K/Akt pathway is also observed in the case of other pathogens. For example, inhibition of PI3K failed to prevent invasion of Salmonella20, which use a type III

secretion-mediated mechanism to gain entry into the host cell. This is in contrast to

receptor-mediated phagocytosis where PI3K activation is critical.

Macrophage receptors that sense Francisella and mediate phagocytosis are just

being defined, but the linkages between receptors and intracellular signaling pathways are essentially unknown. Earlier studies suggested that since there is a lack of macrophage inflammatory response to the LPS of Francisella it is unlikely that TLR4 is involved99-

101. Other studies demonstrate that TLR2 may be important61. The role of SHIP1 in pro-

inflammatory cytokine response elicited by TLR4 or TLR2 engagement has been studied.

Thus, studies by Strassheim et al. demonstrated that SHIP1 negatively regulates TLR2 signaling45. In contrast, the presence of SHIP1 appears to enhance TLR4-induced

inflammatory response. In an earlier study we have demonstrated that SHIP1-deficient

macrophages are hyporesponsive to TLR4 engagement compared to their wild type

counterparts46. These findings are supported by recent reports by Rauh et al102. Although

the latter group earlier reported a negative regulatory role for SHIP1, these findings were

later attributed by the same group to in vitro culture conditions of macrophages in their

study44. Our current findings that SHIP1 negatively regulates Francisella-induced

inflammatory response are consistent with a role for TLR2 in Francisella-induced signaling.

SHIP1 is a cytosolic enzyme that must translocate to the membrane to access its lipid substrates. Previous reports indicate that the N-terminal SH2 domain of SHIP1 is critical for this translocation under certain stimulation conditions, whereas the C-terminal region may be more important under other conditions. Indeed, the C-terminal proline- rich region of SHIP1 has been shown to be required for stabilization of SHIP1 at the

membrane. Additional studies are required to understand the mechanism by which SHIP1

translocates to the membrane in response to Francisella infection. A thorough

understanding of the role of SHIP1 in Francisella-induced signaling events may shed

light on regulatory mechanisms controlling the production of inflammatory mediators

that are essential for protection against Francisella infection. The production of pro-

inflammatory cytokines such as IL-12 by monocytes/macrophages contributes further to

immunity against Francisella infection by augmenting NK cell IFNγ production. In a

recent report, we demonstrated that production of IFNγ by NK cells in response to

stimulation by monokines (IL-12, IL-15 and IL-18) is augmented in the absence of

SHIP1103,104. Thus, it is conceivable that SHIP1 not only regulates macrophage responses

to Francisella infection, but also the subsequent NK cell responses, thereby regulating

overall defense against intracellular pathogens.

In conclusion, the findings in this chapter unravel novel roles for PI3K and SHIP1 in regulating intracellular signaling events involved in macrophage innate immune response to Francisella infection.

In the previous chapter we established that Syk promotes macrophage cytokine

response to infection. In the current chapter we demonstrated that the inflammatory

response is negatively regulated by an inositol phosphatase, SHIP1. Many of these

inflammatory mediators (ex. IL-12 and IL-23) can activate natural killer cells (NK cells)

and T cells resulting in the production of IFNγ which activates the macrophages to

enhance their innate immune response. So, successful pathogens have developed

strategies to subvert such IFNγ-induced immune response. Therefore, in the next chapter,

the influence of Francisella on IFNγ-induced host response will be examined. 96

CHAPTER 4

Francisella GAINS A SURVIVAL ADVANTAGE WITHIN

MONONUCLEAR PHAGOCYTES BY SUPPRESSING HOST IFNγ

RESPONSE

4.1 Abstract

Cytokines such as IL-12 and IL-23 produced during Francisella infection can initiate IFNγ production by NK cells which activates the macrophages to kill the invading pathogens. For example, IFNγ has been reported to suppress the intra-macrophage growth of Francisella through both nitric oxide-dependent and -independent pathways. Since

Francisella is known to subvert host mounted immune responses, we hypothesized that this pathogen could in turn interfere with IFNγ-induced signaling. Here, we report that

Francisella infection suppresses IFNγ-induced STAT1 phosphorylation in both human and murine mononuclear phagocytes. This suppressive effect of Francisella is not dependent on phagosomal escape or bacterial replication and appears to be mediated by a heat-stable and constitutively-expressed bacterial factor. An analysis of the molecular mechanism of STAT1 inhibition indicated that although positive mediators of IFNγ signaling are not influenced, expression of SOCS3, an established negative regulator of

IFNγ signaling, is highly upregulated during infection. Functional analyses revealed that

this interference with IFNγ signaling is accompanied by the suppression of IP-10 production and iNOS induction ultimately resulting in increased intracellular bacterial

survival. Importantly, the suppressive effect on IFNγ-mediated host cell protection is

most effective when IFNγ is added after infection, suggesting that the bacteria may

establish an inhibitory environment within the host cell.

4.2 Introduction

Francisella tularensis is a Gram-negative, intracellular, zoonotic pathogen that

causes the disease tularemia. F. tularensis subspecies tularensis (Type A) and subspecies

holarctica (Type B) cause disease in humans. F. tularensis has been categorized by the

CDC as a Category A Select agent and is commonly perceived as a potential biological

weapon. However very little is known about its pathogenesis. Other subspecies include F.

tularensis novicida and mediasiatica53. Although F. novicida is highly infectious to mice, but is attenuated to humans68. Thus F. novicida is a widely used experimental model for tularemia.

F. tularensis mainly infects monocytes and macrophages. The phagosomes

containing F. tularensis fail to fuse with lysosomes and the bacteria escape into the

cytosol where they replicate and subsequently trigger apoptosis of the host cell53. Escape of this pathogen into the cytosol requires the expression of bacterial proteins such as

IglC, MglA, PmrA and AcpA105-107. Priming of macrophages with IFNγ, however,

inhibits this escape68. IFNγ is a cytokine predominantly produced by natural killer and T

cells, although we have found that infected monocytes also secrete very low levels of

IFNγ75. The signaling cascade initiated by engagement of the IFNγ receptor (IFNγR) has been comprehensively examined108. A schematic model describing the IFNγ-induced

signaling response is shown in the Figure 4.1. Briefly, upon the ligand binding, JAKs are activated and they subsequently phosphorylate the receptor. Next, STAT1 transcription factor is recruited to the receptor, phosphorylated in a JAK-dependent manner.

Phosphorylated STAT1 dimerizes, translocates into nucleus wherein it transcribes IFNγ- responsive genes such as iNOS and IP-10. IFNγ signaling is negatively regulated by several mechanisms including the internalization of IFNγR, dephosphorylation of JAKs by the tyrosine phosphatase SHP-1 and the induction of the SOCS (Suppressors Of

Cytokine Signaling) proteins. SOCS proteins function by binding to JAK tyrosine kinases and inhibiting their phosphorylation, and thereby the JAK-dependent phosphorylation of the receptor and STAT1.

IFNγ plays a critical role in modulating host immune responses. In particular,

IFNγ is crucial for the activation of anti-microbial events such as the production of nitric

oxide and reactive oxygen species and the up-regulation of FcγRI, complement receptor

CR3 and NRAMP149. Several pathogens such as Leishmania donovani50, Mycobacterium

avium51, Mycobacterim tuberculosis52, and Listeria monocytogenes109 have been

equipped with mechanisms to evade IFNγ-mediated host responses. Since Francisella is

frequently referred to as a ‘stealth pathogen’ that effectively evades host immune responses, it is rational to postulate that this organism may also interfere with the host

IFNγ response.

IFNγ Positive Mediators

JAK 1 and JAK2 JAK1

JAK1 chain chain JAK2

α chain α chain

Y Y β β JAK2 p p

Negative Regulators SOCS1, SOCS3, SHP-1 p STAT1 STAT1 p

Gene Transcription Ex. iNOS Nitric Oxide IP-10

Microbial Death

Figure 4.1: Schematic representation of IFNγ-mediated signaling pathway. IFNγR has two sub-units, α and β. The ligand, IFNγ, binds to α chain of the receptor leading to the activation of tyrosine kinases, JAK1 and JAK2 which phosphorylate the receptor and recruit STAT1 transcriptional factor to the receptor. Next, STAT1 is phosphorylated, subsequently homodimerizes and translocates to nucleus wherein it transcribes IFNγ- responsive genes.

100

In this study we report that Francisella suppresses IFNγ-induced STAT1 tyrosine

phosphorylation in both human and murine mononuclear phagocytes. Examination of

downstream events shows that IFNγ-induced iNOS expression is reduced, suggesting a

mechanism for increased bacterial survival. This signaling interference is independent of internalization, phagosomal escape, replication and viability of the pathogen. Further, we demonstrate that the negative regulator SOCS3 is highly upregulated but positive mediators of IFNγ signaling remain unchanged. Finally, we show that administration of recombinant IFNγ leads to diminished intracellular bacterial survival as previously reported69,110, but that administration after infection offers little benefit. Collectively our data demonstrate that Francisella infection subverts IFNγ−mediated host response.

4.3 Materials and Methods

Cells, antibodies and reagents: Raw 264.7 and THP-1 cells were obtained from ATCC

and maintained in RPMI 1640 with 5% heat-inactivated fetal bovine serum (FBS).

Recombinant IFNγ (mouse and human) and mouse IFNβ were purchased from R& D

Systems (Minneapolis, MN). Antibodies specific for phospho-STAT1, phospho-JAK2,

phospho-JAK1, were purchased from Cell Signaling Technology (Beverly, MA). Actin

and SOCS3 antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA).

Antibodies against STAT1 and iNOS were obtained from BD Biosciences (San Jose,

CA). F. novicida U112 (JSG1819), an mglA mutant of F. novicida (JSG2250) and F.

tularensis LVS were used in all experiments. The iglD mutant of F. novicida bacteria

were a generous gift from Dr. Yousef Abu Kwaik (U. of Louisville, KY). Control siRNA

101

and SOCS3-specific siRNA (On-Targetplus SMARTpool) were obtained from

Dharmacon (Lafayette, CO). Francisella strains were streaked and grown overnight on

Chocolate II agar plates at 370C. E. coli K12 were a kind gift of Dr. Brian Ahmer (The

Ohio State University, OH) and grown overnight on Luria-Bertini agar plates.

Isolation of peripheral blood monocytes: Peripheral blood monocytes (PBMs) were isolated as previously described75. Briefly, PBMCs were first isolated by density gradient centrifugation over Histopaque (Sigma-Aldrich, St Louis, MO) and monocytes were then purified from PBMCs by negative selection using MACS Monocyte Isolation kit

(Miltenyi Biotec, Auburn, CA) according to the manufacturer’s instructions. The purity of the monocytes was more than 97% as determined by flow cytometry using CD14 antibodies.

Cell stimulation, lysis, and Western blotting: Macrophages were infected with F. novicida as previously described at an MOI of 10 unless described otherwise71. Briefly,

RAW 264.7 cells were plated overnight in 12-well or 6 well plates and allowed to adhere.

F. novicida resuspended in RPMI medium containing 5% heat inactivated FBS was

0 added to the adherent macrophages and then incubated at 37 C and 5% CO2 for the indicated time points. In parallel, the percent viability of bacteria was tested by plating the inoculum on chocolate II agar plates and comparing the obtained CFUs to the total number of bacteria counted using the Petroff-Hauser chamber. Results indicated that more than 98% of the bacteria in the inoculum were viable in all cases. During the infection shown here the cells were not washed at any point. However, similar responses

102

were seen when cells were extensively washed 1 hour post infection, treated with gentamicin and allowed to incubate (data not shown). Where indicated, cells were stimulated with 25 ng/ml of IFNγ. In various experiments IFNγ was added at different points -either 8 hours prior to infection, at the time of infection or 8 hours post infection.

These details are provided in the corresponding Figure legends. In some experiments, bacteria were killed either by heating the bacterial suspension at 980C for 10 min or by treating with 50 μg/ml of gentamicin for 60 min. In some experiments macrophages were treated with vehicle control (DMSO) or 5 μg/ml of cytochalasin D for 30 minutes before infection. Uninfected and infected cells were lysed in TN1 buffer (50mM Tris pH 8.0,

10mM EDTA, 10mM Na4P2O7, 10mM NaF, 1% Triton-X 100, 125mM NaCl, 10mM

Na3VO4, 10μg/ml each aprotinin and leupeptin). Post-nuclear lysates were boiled in

Laemmli Sample Buffer and were separated by SDS/PAGE, transferred to nitrocellulose filters, probed with the antibody of interest and developed by enhanced chemiluminescence.

Microarray analysis: PBMs were isolated from 4 separate donors and resuspended in

RPMI 1640 with 5% heat-inactivated FBS at 5 million cells per ml. The monocytes were then infected with F. novicida at 100 MOI for 24 hours. Bone marrow-derived macrophages were isolated and differentiated from 3 C57Bl/6 mice as described previously. Seven days after differentiation, macrophages from each mouse were plated into 2 wells of a 6-well dish at 2.5 million per well in RPMI containing 5% heat- inactivated FBS. F. novicida was added to each well at an MOI of 100. Cells were gently mixed and incubated at 37°C in 5% CO2 for 24 h. 103

Twenty four hours post-infection, RNA was extracted from both PBMs and

BMMs using TRIzol® Reagent (Invitrogen Life Technologies, Carlsbad, CA), column- purified using RNeasy columns (Qiagen, Valencia, CA) and hybridized to Affymetrix

GeneChip® Human Genome 133 plus 2.0 and Mouse Genome 430 2.0 Arrays

(Affymetrix, Santa Clara, CA). Expression values were calculated using the “gcrma” package in BioConductor (www.bioconductor.org) and the “limma” package (Smyth,

2004) was used to find genes significantly different (p≤0.05) between infected and uninfected samples.

siRNA transfection: RAW 264.7 cells were transfected with either control siRNA or

SOCS3-specific siRNA using the Amaxa Nucleofector (Amaxa biosystems, Germany) as previously described28. Briefly, 7x106 cells were resuspended in 100 μl Nucelofector

Solution V and were nucelofected with 10 μ1 of 100 μM siRNA. Immediately post- nucleofection, 500 μl of pre-warmed RPMI was added to the transfection mix before

transferring to 12-well plates containing 1.5 ml pre-warmed RPMI per well. Plates were

incubated for 16 hours at 370C before infections were performed.

ELISA measurement of cytokine production: Raw 264.7 cells were infected with F. novicida for varying time points. Cell supernatants were harvested, centrifuged to remove dead cells and analyzed by ELISA using an IP-10 specific kit from R & D Systems

(Minneapolis, MN). Data were analyzed using an unpaired Student’s t-test. A p value <

0.05 was considered significant.

104

Intracellular survival assay: This assay was performed as previously reported76 with a few modifications. Briefly, 90 min post-infection with 100 MOI of F. novicida, cells were washed twice with RPMI and further incubated for 22 h. The cell cultures were

O treated with 50 μg/ml of gentamicin for 60 min at 37 C and 5 % CO2, washed twice with

RPMI and subsequently lysed in 0.1 % SDS for 5 minutes. Immediately, 10 fold serial

dilutions were made and appropriate dilutions were plated on Chocolate II agar plates.

Assays were performed in triplicate for each test group. To maintain constant infection

and IFNγ exposure periods in “pre-, co- and post- groups” the following protocol was

followed. Briefly, within each group two parallel sets of macrophage cultures were infected for a constant infection period. One set of the samples in each group was exposed to 25 ng/ml of IFNγ for 24 hours to maintain constant IFNγ exposure across the groups. Subsequently, both sets in a group were processed to measure the intra-cellular number of the bacteria, and the bacterial count in IFNγ-treated samples was expressed as

a percentage of bacterial number obtained in untreated samples of that group.

4.4 Results

Francisella suppresses IFNγ-induced STAT1 phosphorylation

IFNγ orchestrates several immune processes that are crucial for efficient pathogen

clearance. Recently Santic et al. demonstrated that phagosomes containing Francisella

are able to fuse with lysosomes in IFNγ-activated human macrophages68. Thus, we

hypothesized that Francisella may hinder the IFNγ-mediated host response. To test this

hypothesis, we examined tyrosine phosphorylation of the downstream STAT1 (pY

105

STAT1). For this, RAW 264.7 cells were infected with 10 MOI of F. novicida in the

presence or absence of IFNγ. Unless indicated otherwise both bacteria and IFNγ were

added at the same time in all the experiments (co-stimulation). Protein-matched lysates

were analyzed by Western blotting with antibody specific for tyrosine phosphorylated

STAT1 (Figure 4.2A, upper panel). Treatment of the RAW 264.7 cells with IFNγ resulted

in a robust STAT1 phosphorylation which was dampened by infection with F. novicida

(lane 4). To ensure equal loading of protein in all lanes the same membranes were

reprobed with actin antibody (Figure 4.2A, lower panel). Next, to test whether the

suppressive effect of F. novicida on IFNγ-induced pY STAT1 levels is dose-dependent,

RAW 264.7 cells were infected with increasing MOI in the presence or absence of IFNγ.

The results indicated that Francisella suppressed IFNγ-induced STAT1 phosphorylation

at all the multiplicities of infection tested (Figure 4.2B). Further characterization revealed

that this suppressive effect of F. novicida on IFNγ-induced STAT1 phosphorylation was

evident as early as 1 hour (Figure 4.2C). Next we wanted to validate these findings in

primary cells. For this, BMMs were infected with 10 MOI of F. novicida for 24 hours

and phosphorylation of STAT1 was assessed by Western blotting analysis. The data

obtained indicates that Francisella suppresses IFNγ-induced STAT1 phosphorylation in

primary murine cells also (Figure 4.3A).

To test whether F. tularensis LVS (Type B, vaccine strain) could likewise

suppress IFNγ-induced STAT1 phosphorylation, similar infection experiments were

performed with this latter strain. The results shown in Figure 4.3B demonstrate that F.

tularensis LVS is also able to suppress IFNγ-induced STAT1 phosphorylation.

106

A γ

γ R IFN IFNFN + FN

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

B FN FN+IFNγ

R 1 10 100 IFNγ 1 10 100

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin C

FN IFNγ FN + IFNγ

R 1 6 12 24 1 6 12 24 1 6 12 24 (hours)

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

Figure 4.2: Francisella suppresses IFNγ-induced STAT1 phosphorylation. A. RAW 264.7 were infected with 10 MOI of F. novicida (FN) in the presence or absence of IFNγ (25 ng/ml) for 24 h. Protein-matched lysates were resolved by SDS PAGE and analyzed by Western blotting with phosphotyrosine STAT1(pY STAT1) antibody (upper panel). The same membranes were re-probed with actin antibody (lower panel). The results are representative of six independent experiments. B&C. Raw 264.7 cells were infected with 1, 10 or 100 MOI of F. novicida for 24 hours (B) or with 10 MOI of F. novicida for the indicated time points (C) in the presence or absence of IFNγ and the levels of pY STAT1 were analyzed by Western blotting (upper panels). The lower panels are the re-probes of the same membranes with actin antibody. In all the experiments bacteria and IFNγ were added to the macrophages at the same time (co-stimulation). These results are representative of two independent experiments. R, resting, uninfected cells. 107

A γ

R IFN FN + IFN FN

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

γ B γ LVS LVS + IFN FN + IFN IFN R FN

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

Figure 4.3: Francisella suppresses IFNγ-induced STAT1 phosphorylation in BMMs also. A&B. BMMs were infected with 10 MOI of F. novicida (FN) in the presence or absence of IFNγ (25 ng/ml) for 24 h or for the indicated h (A) or Raw 264.7 cells were infected with 10 MOI of F. novicida (FN) or F. tularensis LVS in the presence or absence of IFNγ (25 ng/ml) for 24 h (B). In both the experiments bacteria and IFNγ were added to the macrophages at the same time (co-stimulation). Protein-matched lysates were analyzed by Western blotting with pY STAT1 antibody (upper panels). The same membranes were re-probed with actin antibody (lower panels). These results are representative of at least two independent experiments. R, resting, uninfected cells.

108

γ γ A

γ γ

γ Killed K12 K12 + IFN Killed FN K12 Killed K12 + IFN R IFN FN FN + IFN Killed FN + IFN

Immunoblot: Anti-pYSTAT1

Immunoblot: Anti-Actin

B β

R FN IFN IFNFN +

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

Figure 4.4: Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation is not unique. A. RAW 264.7 cells were infected with either 10 MOI of E. coli K12 (live or heat killed) or F. novicida (live or heat killed) in the presence or absence of IFNγ for 24 hours and phosphorylation of STAT1 was assessed by Western blotting (upper panel). The same membrane was re-probed with actin antibody (lower panel).These results are representative of two independent experiments. B. RAW 264.7 cells were infected in the presence or absence of IFNβ (500 units/ml; added at the time of infection) and 24 hours after the infection, phosphorylation of STAT1 was assessed by Western blotting (upper panel). The same membranes were re-probed with actin antibody (lower panel). Results are representative of three independent experiments performed. R, resting, uninfected cells.

109

Similar suppression of STAT1 activation or its function has been reported with other

pathogens such as Leishmania donovani50, Mycobacterium avium51, Mycobacterim

tuberculosis52,111 , and was also seen with the other Gram-negative microbe, E.coli K12

(Figure 4.4A, this study) suggesting that some pathogenic bacteria have developed ways

to effectively subvert IFNγ-mediated immune response.

Recent studies have reported that Type 1 IFN is produced and is important for the

inflammasome (a complex critical for the Francisella-induced apoptosis and for the

processing of pro-IL1β) activation during Francisella infection112. STAT1 homodimers are also formed during Type 1 interferon signaling. Thus we wanted to examine the specificity of Francisella-mediated signaling interference. Interestingly, we found that

Francisella is capable of suppressing STAT1 phosphorylation induced by IFNβ as well

(Figure 4.4B). This indicates that the effect of Francisella on STAT1 phosphorylation

may be a general response against interferons.

Uptake, phagosomal escape, replication and viability of Francisella are dispensable

for its suppressive effect on IFNγ-induced STAT1 phosphorylation

We next investigated the bacterial factors that may be responsible for Francisella-

mediated downregulation of IFNγ-induced STAT1 phosphorylation. MglA is a global

transcriptional regulator of Francisella essential for the expression of many virulence

genes. Moreover, MglA is critical for the bacterial escape from the phagosome92. To test

the likely involvement of any MglA-dependent protein/event in the Francisella-mediated

suppression of IFNγ signaling, RAW 264.7 cells were infected with F. novicida or an

mglA mutant of F. novicida (FN mglA), in the presence or absence of IFNγ. In parallel, 110

we examined whether bacterial multiplication is required for the suppression of IFNγ-

induced STAT1 phosphorylation. For this, we used an iglD mutant of F. novicida. IglD is

an MglA-dependent protein that was recently demonstrated to be critical for the

replication of the bacteria within macrophages91. Protein-matched lysates were separated

by SDS PAGE and analyzed by Western blotting with pY STAT1 antibody (Figure 4.5A,

upper panel). Results indicated that both the mutants (FN mglA and FN iglD) similar to

the wild type effectively suppressed IFNγ-induced STAT1 phosphorylation, indicating

that the suppressive effect is not dependent on phagosomal escape nor intracellular

replication of Francisella. Of note, we also examined the number of viable bacteria in the

inoculum and in the macrophages immediately after infection to confirm that a

comparable MOI was achieved in these experiments (Figure 4.5B and data not shown).

To investigate the possible involvement of an inducible bacterial factor in the downregulation of IFNγ-induced STAT1 phosphorylation, macrophages were infected with live or killed bacteria in the presence or absence of IFNγ, and pY STAT1 levels

were analyzed by Western blotting (Figure 4.6A). Bacteria were killed by either heat

treatment (HKFN) or exposure to gentamicin (GKFN). Both live and killed bacteria

effectively suppressed IFNγ-induced pY STAT1. These results suggest that a

constitutively-expressed and heat-stable bacterial component is involved in the

suppression of IFNγ-induced STAT1 phosphorylation.

Next, we tested whether phagocytosis of bacteria is required for Francisella-

mediated interference of IFNγ signaling. For this, we treated RAW 264.7 cells with

vehicle control or cytochalasin D, an actin polymerization inhibitor, for 30 minutes

111

FN + IFN FN mglA + IFN FN iglD IFN R FN FN iglD + IFN FN mglA

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

Figure 4.5: Phagosomal escape or the replication of bacteria is dispensable for Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation. A. RAW 264.7 cells were infected with 10 MOI of wild type (FN), FN mglA or FN iglD in the presence or absence of IFNγ for 24 h (co-stimulation). Protein-matched lysates were analyzed by immunoblotting with pY STAT1 antibody (upper panel). The same membrane was re-probed with actin antibody to ensure equal loading (lower panel). R, resting, uninfected cells. These results are representative of four independent experiments. B. RAW 264.7 cells were infected with 10 MOI of for wild type (FN), FN mglA or FN iglD for 30 minutes and the viability of bacteria was determined by CFU assay. The data are representative of 3 independent experiments. R, resting, uninfected cells.

112

A γ γ

R IFN FN + IFN HKFN + IFN γ HKFN + FN GKFN + IFN GKFN HKFN

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

B Cyt.D.

γ γ

γ γ R IFN FN FN + IFN FN + IFN IFN

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

Figure 4.6: Viability or phagocytosis of Francisella is dispensable for Francisella- mediated suppression of IFNγ-induced STAT1 phosphorylation. A&B. RAW 264.7 cells were infected with live (FN), heat-killed (HKFN) or gentamicin-killed (GKFN) bacteria in the presence or absence of IFNγ for 24 h (A), or were prior treated with 0.1 % DMSO or 5 μg/ml cytochalasin D (Cyt.D) for 30 minutes and infected with F. novicida in the presence or absence of IFNγ for 6 hours (B). Protein-matched lysates were analyzed by immunoblotting with pY STAT1 antibody (upper panels). The same membranes were re-probed with actin antibody to ensure equal loading (lower panels). In all the experiments described above bacteria and IFNγ were added to the macrophages at the same time (co-stimulation). R, resting, uninfected cells. These results are representative of four independent experiments.

113

before infection with 10 MOI of F. novicida for 6 hours. Results obtained clearly show that F. novicida suppresses IFNγ-induced STAT1 phosphorylation even when the uptake of the bacteria is suppressed, suggesting that surface contact of the bacteria may be sufficient for Francisella-mediated IFNγ signaling interference (Figure 4.6B).

Francisella-induced SOCS3 expression dampens STAT1 phosphorylation

We next examined the molecular mechanism of Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation. Here, we examined two possibilities: a) that

Francisella downregulates positive regulators upstream of STAT1 phosphorylation, and/or b) that Francisella upregulates negative regulators of STAT1 phosphorylation. To test whether positive regulators are down-regulated, Raw 264.7 cells were infected with

F. novicida in the presence or absence of IFNγ and the expression levels of STAT1,

JAK1 and JAK2 were studied by Western blot analysis. In parallel, phosphorylation

(indicative of activation) of JAK1 and JAK2 were assessed by Western blotting with phospho-specific antibodies to JAK1 and JAK2. Results indicated that there is no suppressive effect on the expression or phosphorylation of JAK1 and JAK2 during infection (Figure 4.7). Further we observed that in addition to suppressing STAT1 phosphorylation, F. novicida also prevented the induction of IFNγ-mediated STAT1 protein to a certain extent (Figure 4.7). The decrease in the induction of IFNγ-mediated

STAT1 protein during Francisella infection may be a consequence of the suppressed activation of STAT1, which can feed back to upregulate itself.

Since the tyrosine phosphorylation level of the JAKs was not inhibited by F. novicida infection, we did not investigate the activation of SHP-1, the tyrosine

114

FN IFNγ FN + IFNγ R 1 6 12 24 1 6 12 24 1 6 12 24 (hours)

Immunoblot: Anti-STAT1

Immunoblot: Anti-JAK1

Immunoblot: Anti-JAK2

Immunoblot: Anti-pJAK1

Immunoblot: Anti-pJAK2

Immunoblot: Anti-Actin

Figure 4.7: Francisella does not predominantly suppress positive mediators of IFNγ signaling. RAW 264.7 cells were infected with 10 MOI of F. novicida for the indicated time in the presence or absence of IFNγ (co-stimulation). Protein-matched lysates were separated by SDS PAGE and analyzed by Western blotting with the indicated antibodies. R, resting, uninfected cells. These results are representative of two independent experiments.

115

phosphatase that dephosphorylates JAK kinases. However, we examined whether other

negative regulators of IFNγ signaling, the SOCS proteins, were upregulated during

infection. SOCS1 and SOCS3 are established negative regulators of the IFNγ signaling

pathway. Downregulation of IFNγ-induced signaling by several pathogens is strongly

associated with the upregulation of SOCS3 protein. To this end, we examined microarray

results from an experiment comparing infected and uninfected BMMs and analyzed the

expression of various SOCS proteins. Results shown in Figure 4.8A indicate that during

F. novicida infection SOCS3 but not SOCS1 mRNA was highly upregulated in murine

BMMs. Further the expression of SOCS3 was confirmed at the level of protein by

Western blotting in RAW 264.7 cells (Figure 4.8B). Infection alone induced the expression of SOCS3 as early as 6h post-infection and the protein levels persisted even at

24h post-infection (Figures 4.8B and 4.8C). Moreover, co-stimulation of macrophages

with F. novicida and IFNγ resulted in a further increase in the SOCS3 protein levels

(Figure 4.8C). Consistently, phagosomal escape, replication, viability and uptake of

Francisella, which were dispensable for the suppression of IFNγ-induced STAT1 phosphorylation, were found to be not essential for the Francisella-induced upregulation of SOCS3 protein as well (Figures 4.9 and 4.10).

We next examined whether Francisella-induced SOCS3 may contribute to the

suppression of IFNγ-induced STAT1 phosphorylation. For this, RAW 264.7 cells were

transfected with either control or SOCS3-specific siRNA by nucleofection.

Approximately 16 hours post transfection, cells were infected with 10 MOI of F.

novicida and the tyrosine phosphorylation of STAT1 was assessed by Western blotting

(Figure 4.11A). SOCS3-specific siRNA transfectants, when exposed to both IFNγ and 116

A B

4000 3500 BMM 3000 R 12 h 18 h 24 h 2500 R 2000 Immunoblot: Anti-SOCS3 1500 FN

Expression Values 1000 500 Immunoblot: Anti-Actin 0

1 2 3 4 7 CS CS CS CS CS6 CS O O O O O S S S SO SOCS5S S

FN IFNγ FN + IFNγ

R 1 6 12 24 1 6 12 24 1 6 12 24 (hours)

Immunoblot: Anti-SOCS3

Immunoblot: Anti-Actin

Figure 4.8: Francisella induces the expression of SOCS3. A. Bone marrow-derived macrophages (BMM) were infected with 100 MOI of F. novicida for 24 h. RNA was extracted, purified and microarray analysis was performed. Expression values for SOCS mRNA from three independent infections are shown. B. RAW 264.7 cells were infected with 10 MOI of F. novicida and the expression of SOCS3 protein was analyzed by Western blotting (upper panel). The same membrane was re-probed with actin antibody to ensure equal loading (lower panel). C. RAW 264.7 cells were infected with 10 MOI of F. novicida in the presence or absence of IFNγ (25 ng/ml) for the indicated hours (co- stimulation). Protein-matched lysates were analyzed by Western blotting with SOCS3 antibody (upper panel). Lower panel is the re-probe of the same membrane with actin antibody to ensure equal loading. R, resting, uninfected cells. These results are representative of three independent experiments.

117

γ γ

γ IFN FN mglA + IFN FN + IFN FN iglD R FN FN iglD + IFN FN mglA

Immunoblot: Anti-SOCS3

Immunoblot: Anti-Actin

Figure 4.9: Phagosomal escape or the replication of bacteria is dispensable for Francisella-mediated induction of SOCS3. RAW 264.7 cells were infected with 10 MOI of wild type (FN), FN mglA or FN iglD in the presence or absence of IFNγ for 24 h (co-stimulation). Protein-matched lysates were analyzed by immunoblotting with SOCS3 antibody (upper panel). The same membranes were re-probed with actin antibody to ensure equal loading (lower panel). R, resting, uninfected cells. These results are representative of four independent experiments.

118

γ IFN IFNFN + R IFN γ HKFN + FN IFNGKFN + GKFN HKFN

Immunoblot: Anti-SOCS3

Immunoblot: Anti-Actin

B Cyt.D. γ γ

γ γ FN + IFN R IFN FN + IFN IFN FN

Immunoblot: Anti-SOCS3

Immunoblot: Anti-Actin

Figure 4.10: Viability or phagocytosis of Francisella is dispensable for Francisella- mediated induction of SOCS3. A&B. RAW 264.7 cells were infected with live (FN), heat-killed (HKFN) or gentamicin-killed (GKFN) bacteria in the presence or absence of IFNγ for 24 h (A), or were prior treated with 0.1 % DMSO or 5 μg/ml cytochalasin D (Cyt.D) for 30 minutes and infected with F. novicida in the presence or absence of IFNγ for 24 hours (B). Protein-matched lysates were analyzed by immunoblotting with SOCS3 antibody (upper panels). The same membranes were re-probed with actin antibody to ensure equal loading (lower panels). In all the experiments described above bacteria and IFNγ were added to the macrophages at the same time (co-stimulation). R, resting, uninfected cells. These results are representative of four independent experiments.

119

A Ctrl siRNA SOCS3 siRNA

FN + IFNγ FN + IFNγ γ γ R IFN Post Co R IFN Co Post

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

B Ctrl siRNA SOCS3 siRNA γ γ

γ γ R IFN FN + IFN R IFN FN + IFN FN FN

Immunoblot: Anti-SOCS3

Immunoblot: Anti-Actin

Figure 4.11: Francisella suppresses IFNγ-induced STAT1 phosphorylation through SOCS3. A. RAW 264.7 cells were nucleofected with either control or SOCS3-specific siRNA. Twenty four hours after transfection, cells were infected with F. novicida in the presence or absence of IFNγ (25 ng/ml) for 16 hours and the tyrosine phosphorylation of STAT1 was assessed by Western blotting analysis. Co, IFNγ was added to the transfectants at the time of infection. Post, IFNγ was added 6 hours after infection. These results are representative of three independent experiments. B. RAW 264.7 cells were transfected with control or SOCS3-specific siRNA and 16 hours after transfection, cells were infected with F. novicida in the presence or absence of IFNγ (25ng/ml) for 16 hours and expression of SOCS3 was analyzed by Western blotting (upper panel). The same membrane is reprobed with actin antibody to ensure equal loading (lower panel). R, resting, uninfected cells. These results are representative of 4 similar and independent experiments.

120

F. novicida, displayed elevated STAT1 phosphorylation levels than control transfectants.

This suggests that SOCS3 at least in part may contribute to the Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation. In parallel, we also tested the levels of SOCS3 protein in the tranfectants by Western blotting with SOCS3 antibody and found that cells nucleofected with SOCS3-specific siRNA contained less SOCS3 protein than the control transfectants (Figure 4.11B).

Francisella suppresses IFNγ-induced STAT1 phosphorylation in human monocytes

To investigate whether Francisella can also downregulate IFNγ-mediated STAT1 phosphorylation in human cells, we primed human peripheral blood monocytes (PBM) or

THP-1 (human monocytic cell line) with IFNγ for 8 hours before infection with F. novicida. Similar to the results obtained in murine macrophages (RAW 264.7 and

BMMs), we found that Francisella infection suppressed IFNγ-induced STAT1 phosphorylation in both PBMs and THP-1 cells (Figure 4.12A and 4.12B). Similar results were obtained when THP-1 cells were infected with F. tularensis LVS (Figure 4.13A).

Moreover, the suppression in the IFNγ-induced STAT1 phosphorylation was seen as early as 2h post-infection of THP-1 cells (Figure 4.13B). Also consistent with the results obtained in RAW 264.7 cells, FN mglA or FN iglD effectively suppressed IFNγ-induced

STAT1 phosphorylation (Figure 4.13A). Finally, SOCS3 mRNA was also found to be upregulated during F. novicida infection of PBMs (Figure 4.14).

121

A γ

R IFN FN + IFN FN

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

γ IFN R IFN + FN B FN

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

Figure 4.12: Francisella suppresses IFNγ-induced STAT1 phosphorylation in human cells. A&B. PBMs (A) or THP-1 (B) were pre-treated with IFNγ for 8 hours and then infected with 10 MOI of F. novicida in the presence or absence of IFNγ for 24 h. Protein-matched lysates were analyzed by Western blotting with pY STAT1 antibody (upper panels). The same membranes were re-probed with actin antibody (lower panels).The results are representative of at least 4 independent experiments. R, resting, uninfected cells.

122

γ γ

FN mglA + IFN FN iglD FN + IFN LVS + IFN IFN R LVS FN FN iglD + IFN FN mglA

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

Figure 4.13: Phagosomal escape or the replication of bacteria is dispensable for Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation. THP-1 cells were pre-treated with IFNγ for 8 hours and then infected with 10 MOI of wild type (FN), FN mglA or FN iglD in the presence or absence of IFNγ for 24 h. Protein-matched lysates were analyzed by immunoblotting with pY STAT1 antibody (upper panel). The same membrane was re-probed with actin antibody to ensure equal loading (lower panel). R, resting, uninfected cells. These results are representative of three independent experiments.

123

1600 1400 1200 1000 R 800 FN 600 400

Expression Values 200 0 3 6 S S CS1 C CS4 C CS7 O O O O O S SOCS2S S SOCS5S S

Figure 4.14: Francisella induces the expression of SOCS3 in human cells. PBMs were infected with 100 MOI of F. novicida (FN) for 24 h. RNA was extracted, purified and microarray analysis was performed. Expression values for SOCS mRNA from four monocytes donors are shown. R, resting, uninfected cells.

124

Francisella suppresses STAT1-dependent nitric oxide and IP-10 production

We next examined the influence of Francisella on IFNγ-responsive and STAT1-

dependent genes such as IP-10 and iNOS. For this, we infected RAW 264.7 cells in the

presence or absence of IFNγ for the indicated time points (Figure 4.15A) and the cell

supernatants were analyzed by ELISA for IP-10 production.We found that infection of

macrophages with F. novicida significantly reduced the IFNγ-induced IP-10 production

at both 24h and 36h post-infection. Similar results were obtained in THP-1 cells (Figure

4.15B). Moreover, FN mglA, FN iglD and dead bacteria also significantly suppressed

IFNγ-induced IP-10 production further strengthening the observation that phagosomal

escape, replication and the viability of bacteria are dispensable for the bacterial

interference of IFNγ signaling pathway (Figure 4.16). It must be noted that although

though there is a statistically significant reduction in the production of IFNγ-induced IP-

10 but it was not a drastic reduction and may not correlate with the effective inhibition of

STAT1 phosphorylation observed in mouse macrophages (Figure 4.2A) and human monocytes (Figure 4.12B).

Treatment of macrophages IFNγ is reported to induce the expression of iNOS

leading to the production of NO, which is essential for killing microbes. In agreement

with previous reports, we observed that co-stimulation of macrophages with IFNγ and F.

novicida resulted in the generation of NO69,110 (Figure 4.17A). Similarly, co-stimulation of cells with IFNγ and F. novicida resulted in the upregulation of iNOS (Figure 4.17B).

The observation that Francisella downregulates IFNγ-induced STAT1 phosphorylation but does not suppress STAT1-dependent iNOS induction appeared paradoxical.

125

12000 RAW 264.7 10000 R FN IFNγ 8000 * FN + IFNγ 6000 * 4000

(pg/ml) IP-10 2000 0

24 h 36h

THP-1 20000

15000 FN * IFNγ * FN + IFNγ 10000

IP-10 (pg/ml) IP-10 5000

0 24 h 36h

Figure 4.15: Francisella suppresses IFNγ-induced IP-10 production. RAW 264.7 cells (A) or THP-1 cells (B) were infected with F. novicida in the presence or absence of IFNγ (25 ng/ml; co-stimulation) for the indicated time points and the amount of IP-10 in the cell supernatants was analyzed by ELISA. The graphs show mean and SD of values obtained from 3 independent experiments. * p<0.05 compared with the corresponding IFNγ value. R, resting, uninfected cells.

126

18000 16000 14000 * * 12000 10000 8000 * 6000 IP-10 (pg/ml) * * 4000 2000

0 γ γ γ D γ γ γ R N lA l N N F N N N N F N F N F F g F g F F F I I I i I K I K I m H + + N + + G + N F N A N N F l lD F F F g g K K i m H G N N F F

Figure 4.16: Phagosomal escape, replication and viability of Francisella are dispensable for the suppression of IFNγ-induced IP-10 production. RAW 264.7 cells were infected with wild type (FN) or the mutant bacteria (FN iglD or FN mglA), or heat- killed (HKFN) or gentamicin-killed (GKFN) bacteria in the presence or absence of IFNγ for 24 h. Cell supernatants were analyzed by ELISA for IP-10 production. The graph shows mean and SD of values obtained from 3 independent experiments. * p<0.05 compared with IFNγ value. R, resting, uninfected cells.

127

180 160 * 140 120 100 80 60

moles of NO μ 40 20 0

y γ γ R nl N N o F F I I N + F FN

B FN IFNγ FN + IFNγ R 1 6 12 24 1 6 12 24 1 6 12 24 (hours)

Immunoblot: Anti-iNOS

Immunoblot: Anti-Actin

Figure 4.17: Co-stimulation of macrophages with Francisella and IFNγ induces iNOS/nitric oxide. A. RAW 264.7 cells were infected with 10 MOI of F. novicida in the presence or absence of IFNγ (25 ng/ml; co-stimulation) for 24 h and the amount of NO in the cell supernatants was measured by the Griess reagent. The graph show mean and SD of values obtained from 3 independent experiments. * p<0.05 compared with IFNγ value. B. RAW 264.7 cells were infected in the presence or absence of IFNγ for the indicated time points (co-stimulation) and protein-matched cell lysates were analyzed for the expression of iNOS by Western blotting. R, resting, uninfected cells. These results are representative of three independent experiments.

128

We reasoned that Francisella may require some time to establish a permissive environment and be able to better counter IFNγ-induced iNOS expression. To test this notion, we compared the levels of iNOS induction under priming, co-stimulatory and post-infection conditions. For this, IFNγ was added to RAW 264.7 cells 8h prior to infection (priming), at the time of infection (co-stimulation) or 8h after infection (post- infection) and the expression of iNOS under these conditions was analyzed by Western blotting. Infection or IFNγ treatment alone induced very low levels of iNOS expression.

The induction of iNOS was highly enhanced when IFNγ-primed macrophages were infected with F. novicida. However, when IFNγ was administered at the time of infection

(co-stimulation) the induction of iNOS was dramatically reduced, and almost completely inhibited if IFNγ was administered after infection. Likewise, compared to the IFNγ priming condition, co-stimulation of macrophages with IFNγ and F. novicida resulted in lower levels of STAT1 phosphorylation, which is further reduced when cells were exposed to IFNγ after infection (Figure 4.18A, middle panel). A similar trend was seen with IFNγ-induced IP-10, an IFNγ-dependent chemokine (Figure 4.18B). Collectively, these data indicate that F. novicida establishes a permissive environment in the host cell over time to offset the host protective effects of IFNγ.

Establishment of Francisella infection in macrophages leads to resistance to IFNγ- induced bacterial death

Previous studies demonstrated that prior exposure of infected cells to IFNγ drastically reduces the intra-macrophage growth of bacteria69,110. This is consistent with

129

γ IFN Pre R FN Co Post

Immunoblot: Anti-iNOS

Immunoblot: Anti-pY701STAT1

Immunoblot: Anti-Actin

4500 4000 3500 * 3000 2500 * 2000

1500

(pg/ml) IP-10 1000 500 *

R FN IFNγ Pre Co Post FN + IFNγ

Figure 4.18: Francisella suppresses IFNγ-induced iNOS and IP-10. A. RAW 264.7 cells were exposed to IFNγ 8 h prior to infection (pre), at the time of infection (co) or 8 h post-infection (post) for a constant “IFNγ-exposure” period of 24 hours and the cellular levels of iNOS and pY STAT1 were analyzed by Western blotting. The same membranes were re-probed with actin antibody. These data are representative of three independent experiments. B. RAW 264.7 cells were treated as described above in A and the amount of IP-10 produced was measured by ELISA. The graph shows mean and SD of values obtained from triplicate samples. The data is representative of 3 independent experiments. * p<0.05 compared with the IFNγ value. R, resting, uninfected cells.

130

the induction of iNOS in our model. However, our data indicate that if IFNγ is

administered either at the time of infection or after infection Francisella can suppress

IFNγ-induced iNOS expression. Thus, to test if this reduction in iNOS levels correlates

with an increase in the bacterial survival, CFUs from infected cell lysates under various

conditions of IFNγ exposure (priming, co-stimulatory and post-infection) were compared.

The results are shown in Figure 4.19.

When IFNγ was added prior to or at the time of infection, there was ~ 90% decrease in the intracellular bacterial survival, compared to the survival in the absence of

IFNγ exposure. However, when IFNγ was added 8 hours post infection there was only a

30-40% reduction in intracellular bacterial survival. The reduced efficiency of IFNγ to

induce bacterial death when added 8 hours after infection correlates with the decrease in

the levels of iNOS under these conditions as shown in Figure 4.18A.

In summary, we demonstrated that in both mouse and human cells Francisella

suppresses IFNγ-induced STAT1 phosphorylation and this Francisella-mediated

interference of IFNγ signaling is associated with an induction of SOCS3 (Figure 4.20).

We also showed that SOCS3, at least in part, contributes to the Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation. Further, Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation resulted in lowered induction of

STAT1-dependent proteins such as iNOS, leading to increased intra-macrophage survival of the bacteria.

131

120 without IFNγ 100 with IFNγ

Bacteria Viable % 20

0 Pre Co Post

Figure 4.19: Francisella opposes IFNγ-induced bacterial death. RAW 264.7 cells were treated as described in the Materials and Methods section to maintain constant infection period within each of the three groups (pre, co and post) and constant IFNγ exposure period across the three groups. CFUs in samples that were not treated with IFNγ were set as 100%. The CFUs obtained from the samples treated with IFNγ were expressed as a percent of the corresponding non-IFNγ treated sample. The graph shows mean and SD of values obtained from 3 independent experiments.

132

Francisella

IFNγ Receptor

p STAT1 STAT1 p

SOCS3 Gene Transcription Eg. iNOS

SOCS3 Nitric Oxide

Microbial Death SOCS3 gene induction

Figure 4.20: Proposed model of Francisella-mediated interference of IFNγ signaling response. Surface contact of Francisella is sufficient to suppress IFNγ-induced STAT1 phosphorylation. Francisella infection up-regulates SOCS3 expression, which suppresses STAT1 phosphorylation potentially by binding to the IFNγ receptor and dampening subsequent recruitment and activation of STAT1. Ultimately, Francisella-mediated suppression of IFNγ response leads to the inhibition of IFNγ-induced iNOS and other anti-microbial events, resulting in enhanced bacterial survival in host cells.

133

4.5 Discussion

Phagocytic cells such as monocytes and macrophages ingest and subsequently

destroy pathogens through the phago-lysosomal pathway and via the production of

inflammatory mediators such as cytokines and chemokines. These microbicidal activities

are greatly increased by exposure of the cells to IFNγ. Such IFNγ-activated macrophages

are often referred to as classically activated or Type I activated macrophages. Highly

successful pathogens have developed various strategies to circumvent microbicidal

responses of the host cell and, instead, create a favorable intracellular environment to suit

their needs. Our current knowledge regarding the immune subversion mechanisms used

by Francisella remains limited. These mechanisms have been reported to include evasion of phago-lysosomal fusion70,113, dampening of the inflammatory response upon escape

into the cytosol63,71 and inhibition of T cell responses72. One strategy utilized by pathogens such as Leishmania donovani50, Mycobacterium avium51, Mycobacterium

tuberculosis52, and Listeria monocytogenes109 is to interfere with the IFNγ-mediated

activation of macrophages. In this process, pathogens such as Leishmania donovani

suppress IFNγ-induced STAT1 activation50 whereas Mycobacterium tuberculosis does not influence STAT1 activation but exerts a gene-selective inhibition of IFNγ-induced transcriptional responses by disrupting its interaction with transcriptional co- activators52,111. Results from the present study provide evidence that Francisella

suppresses the IFNγ-mediated signaling response and gene expression which results in decreased microbicidal activity of the host cell.

134

The inhibitory effect of Francisella on IFNγ-mediated signaling does not require the bacteria to be metabolically active. Our finding that killed Francisella also downregulates IFNγ signaling is consistent with the findings obtained with other pathogens such as L. donovani114 and Mycobacterium bovis115. Together, these data

provide evidence that a heat-stable bacterial and constitutive factor is responsible for the

downregulation of IFNγ-induced STAT1 tyrosine phosphorylation. This observation is in

accord with the results obtained in other studies which demonstrate that cells treated with

heat-stable bacterial components such as LPS116 or Mycobacterial TDM115 efficiently

inhibited IFNγ-induced STAT1 phosphorylation. Moreover, we observed that prevention

of bacterial phagocytosis by exposure of cells to cytochalasin D did not reverse the

Francisella-mediated inhibition of IFNγ signaling, suggesting that bacterial contact itself

is sufficient to cause mediate this signaling interference. Perhaps the ligation of a host

cell receptor by a surface bacterial component such as LPS may contribute to the

interference of IFNγ signaling. For example the inhibition of IFNγ-mediated responses by

Mycobacterium tuberculosis117,118 and Mycobacterium avium118 is at least partly

dependent upon TLR2 engagement. TLR2 is one of the major macrophage receptor

involved in the recognition of Francisella. This tempts us to speculate that TLR2 may

play a critical role in Francisella-mediated interference of IFNγ signaling as well.

In our model, Francisella infection did not decrease the IFNγ-induced

phosphorylation of JAK tyrosine kinases, which is in contrast with findings obtained with

other intracellular pathogens50,115. Instead, we observed that Francisella-mediated suppression of IFNγ signaling is associated with the upregulation of SOCS3 and that

135

knock-down of SOCS3 using RNAi technology partially restored the IFNγ-induced

STAT1 phosphorylation during Francisella infection. The interference of IFNγ signaling

by several pathogens has been correlated with expression of either SOCS1 or SOCS3 or

both. For example, Salmonella enterica serovar Typhi induces both SOCS1 and SOCS3

expression while Burkholderia pseudomallei induces only SOCS3 expression119. SOCS proteins are suppressors of cytokine signaling that form a vital part of the negative feedback of cytokine signal transduction. Our microarray data obtained with infected murine BMMs and human PBMs indicate that SOCS3 but not SOCS1 mRNA is highly upregulated during Francisella infection. SOCS1 can directly interact with JAK1 and

JAK2 and inhibit their phosphorylation and kinase activity120. Therefore, the unchanged

levels of IFNγ-induced JAK phosphorylation during Francisella infection correlate with

the lack of induction of SOCS1 in our experiments.

The mechanism of SOCS3 action is not clear. One study showed that although

SOCS3 associated with the JAK kinases, overexpression of SOCS3 did not inhibit the in vitro kinase activity of either JAK1 or JAK2121. Other studies showed that SOCS3 binds

to phosphorylated receptors including the leptin receptor122 and IL-2 receptor β chain123.

When bound to the activated receptors, SOCS3 may inhibit JAK2 activation. However, high levels of activated growth hormone receptor were found to be required for SOCS3- mediated inhibition of growth hormone induced JAK2 activation124.

Our data demonstrate that a heat-stable bacterial factor(s), but not live bacteria, is

essential for the induction of SOCS3. These results are in agreement with results from

other studies showing purified LPS116 of other bacteria or mycobacterial TDM115 could activate SOCS3 expression. Further, consistent with the hypothesis that a bacterial 136

surface component is involved in the inhibition of IFNγ signaling, failure of bacterial

internalization still resulted in the induction of SOCS3 protein. Ongoing experiments are

aimed at deciphering the molecular mechanism(s) of Francisella-induced SOCS3

expression.

The observation that iNOS induction is inhibited during Francisella infection is

consistent with previous findings that overexpression of SOCS3 results in the suppression

of the mouse iNOS gene promoter116. Interestingly, we found that Francisella requires some time (in the absence of IFNγ exposure) for the establishment of maximal inhibition of host-protective events. Such a requirement was also observed in the case of

Burkholderia pseudomallei infection119. This suggests that Francisella regulates the host

cell response to protective inflammatory mediators over time, the net result of which is

increased intracellular survival.

Data obtained from different mouse models strongly support the protective role of

IFNγ during murine tularemia125,126. Leiby et al. have shown that intra-peritoneal

injections of IFNγ neutralizing antibody at the time of infection resulted in the death of

mice even at a very low dose of infection. However, administration of the IFNγ

neutralizing antibody two days post infection did not significantly influence the progress

of the disease, suggesting that the protective effect of IFNγ is most effective at early time

points after infection.66 This is consistent with our infection model where IFNγ is not that

effective at protecting host cells if administered after the infection has occurred. Such

strict time dependence for IFNγ therapy is also seen with Listeria monocytogenes, a

pathogen previously reported to inhibit IFNγ response127. Further, Conlan et al observed

137

that neutralizing antibodies against IFNγ did not alter the progress of primary murine

tularemia initiated through intra-nasal administration of Francisella128. Thus, based on our studies it may be speculated that alveolar macrophages may be more effective in inhibiting IFNγ responses and may thus contribute to the severe lethality observed during the pneumonic form of tularemia.

Although IFNγ is produced during the progression of murine tularemia but mice

still succumb. This could be partly explained by the ability of Francisella to induce

SOCS3 and thereby potentially subvert the IFNγ-mediated survival benefit. To our knowledge, this is the first study to demonstrate Francisella-mediated suppression of

IFNγ signaling and provide evidence for the functional consequence of this signaling interference.

138

CHAPTER 5

SUMMARY AND FUTURE PERSPECTIVES

Innate immune responses are one of the most important lines of defense against invading pathogens. Various cell types such as macrophages, natural killer cells and neutrophils orchestrate different innate immune responses very effectively and target microbes for destruction. Although the effector functions of macrophages are well described the roles of different signaling pathways that modulate these functions are not comprehensively understood. Activation of these signaling pathways is stimulus-specific thereby further complicate the scenario. Thus, a thorough analysis is required to delineate the importance of the various regulatory mechanisms that modulate the host functional response.

Tularemia is a zoonotic disease that is readily treatable with the available antibiotics. Francisella tularensis, the etiological agent of tularemia, is highly infectious

(LD50 in humans-10 CFU) and because of its ability to be easily aerosolized there is a growing concern that this organism may be used as a biological weapon53. The host response towards Francisella is poorly understood. Thus, the broad objective of this study was to identity the activation of critical signaling pathways that regulate macrophage effector functions against Francisella. Specifically the roles of Syk and

139

SHIP1 in the uptake of Francisella and in the production of Francisella-induced

inflammatory mediators has been studied in the second and third chapters, respectively.

On the other hand, host response is also modulated by the pathogen to evade immune

response and to modify the cellular conditions its needs. Therefore, to better understand

the host response during Francisella infection, we also examined Francisella-mediated

subversion of IFNγ response in the fourth chapter.

Syk regulates macrophage response to Francisella

Our investigation of the role of Syk during Francisella infection demonstrated

that Syk is necessary for phagocytosis of Francisella through the activation of Erk MAP kinase. Further, we also found that Syk regulates the cytokine response by modulating both Erk and PI3K/Akt pathways and subsequent NFκB activation. Based on our data we propose the schematic model to describe the roles of Syk and SHIP1 in the macrophage response to Francisella infection (Figure 5.1 and discussed in the following paragraphs).

Both Syk and SHIP1 regulate the PI3K/Akt pathway albeit in opposite directions and thereby modulate Francisella-induced inflammatory response. Although, both Syk and

SHIP1 modulate PI3K/Akt pathway this pathway is not required for the uptake of

Francisella. Syk not SHIP1 controls the phosphorylation of Erk which is required for the

optimal uptake of Francisella. Thus, only Syk but not SHIP1 plays a role in the

phagocytosis of Francisella by macrophages.

140

s osi cyt ago Ph IFNγ Receptor SHIP1 Syk

p STAT1 STAT1 p Erk PI3K/Akt

IFNγ NF-κB Gene Transcription S O Eg. iNOS C S 3 Nitric Oxide NK cells Microbial Death T cells

Cytokines IL-12 and IL-23

Figure 5.1: Proposed model for the regulation macrophage response to Francisella. Both SHIP1 and Syk are activated during Francisella infection. Syk but not SHIP1 controls the activation of Erk pathway that is required for the phagocytosis of Francisella. Syk and SHIP and modulate the activation of PI3K/Akt pathway (in opposing directions) and thereby subsequent NFκB activation and cytokine production. Thus both SHIP1 and Syk regulate inflammatory response induced by Francisella but only Syk modulates the uptake of this microbe. Cytokines produced during Francisella infection can trigger NK cells to release IFNγ which can subsequently induce STAT1- dependent iNOS leading to microbial death. However, our results demonstrate that Francisella induces SOCS3 expression to dampen STAT1 activation and subsequent iNOS induction leading to enhanced intra-macrophage survival of the bacteria.

141

Clemens et al recently demonstrated that actin cytoskeletal reorganization is

necessary for the engulfment of Francisella55. Thus based on our findings it is possible

that the inhibition of either Syk or Erk could affect the reorientation of actin around the

nascent phagosomes thereby dampening the subsequent uptake of the bacteria.

In fact, inhibiting Syk by RNAi affected the accumulation of actin around the forming

phagosomes in HL60 cells fed with opsonized zymosan34. Thus, the role of Syk and Erk

proteins in Francisella-induced actin cytoskeletal rearrangements needs to be further

studied in the future. The potential redistribution of either Syk or Erk to the nascent

phagosomes and the regulatory mechanisms that modulate the relocation of these proteins

must be addressed to enhance our understanding of the signaling circuitry required for

phagocytosis of Francisella.

The upstream events leading to Syk activation have not been addressed. Since

Syk is a kinase that can phosphorylate itself it is conceivable that phosphorylation of Syk

is a result of autophosphorylation event. But, how is the signal transduced to Syk during

Francisella infection? It is necessary to identify the receptor which triggers the phosphorylation of Syk. It is possible that Syk may be activated in a TLR2-dependent manner. Several findings support this hypothesis as outlined below.

First, engagement of TLR2 can promote the activation of PI3K and Erk

pathways129. These two pathways are dependent upon Syk in our infection model.

Second, cytosolic tyrosine residues of TLR2 are phosphorylated upon ligation of the

receptor130. These sites may be potential docking sites for SH2 domain containing

proteins such as PI3K, Syk and SHIP1. Consistently, two recent studies reported that

PI3K was found in a multi-protein complex consisting of TLR2130,131. Third, recently it

142

was reported that Syk binds to a tyrosine phosphorylated residue of the cytosolic domain

of another toll-like receptor, TLR4132. Therefore, it is possible that during Francisella

infection Syk may associate with TLR2, the major TLR recognizing Francisella.

If Syk associates with TLR2, it is worth testing if this interaction is direct (SH2

domain of Syk binding to cytosolic phosphorylated tyrosine residue of TLR2) or if this association is bridged by an adaptor molecule. Interestingly, Syk can interact with and subsequently phosphorylate the p85 subunit of class I PI3K133. Therefore it is logical to

propose the existence of a multi-protein complex consisting of TLR2-Syk-PI3K during

Francisella infection. As discussed in Chapter 2, Syk can potentially activate the Erk

pathway in a Ras-dependent86 and/or a PKC-dependent manner87. Thus, how Syk

activates the Erk pathway during Francisella infection needs to be investigated. It is

possible that both these pathways are operational during Francisella infection.

SHIP1 and PI3K regulate Francisella-induced macrophage inflammatory response

Our studies emphasize the role of SHIP1 as a negative regulator of cytokine response during Francisella infection. SHIP1 is phosphorylated during Francisella

infection. The phosphorylation of SHIP1 indicates its translocation to the membrane

where it can access its lipid substrates. But several questions remain unanswered. First,

how does SHIP1 dock at the membrane? Recently, SHIP1 has been reported to be involved in TLR2-induced neutrophil activation and acute lung injury indicating a critical role for SHIP1 in TLR2-mediated signaling45. Also, as described above TLR2 is

phosphorylated on the cytosolic tyrosines providing docking sites for SH2-domain

containing proteins130. Moreover, TLR2 is one of the major receptor that recognizes

143

Francisella61. Thus, it is possible that SHIP1 directly or indirectly interact with this receptor. Second, how does SHIP1 get to the receptor? Association of SHIP1 to the membrane-bound receptors may be direct as in the case of FcγIIb or may be mediated by an adapter protein as seen in during growth factor receptor signaling39. Third, how is

SHIP1 phosphorylated during Francisella infection? It has been reported previously that

SHIP1 is phosphorylated by Src family of kinases during FcγR signaling39,40. It is quite possible that one of the members of this family phosphorylates SHIP1 at the membrane.

Finally, what domains of SHIP1 are important in modulating host response against

Francisella? Based on the data from this study it is possible that the catalytic domain may be critical because it antagonizes PI3K activity. However domains such as the SH2 may be important in directing SHIP1 to the membrane. Further, it was also reported that

PRD of SHIP1 is critical for stabilizing its interactions at the membrane134-137. Moreover, these domains and the NPXY motifs can associate with other proteins thereby influencing signaling pathways that are not investigated in the current study. These questions need to be addressed to fully understand the role of SHIP1 in Francisella infection.

Using a SHIP1 knockout model we have found that SHIP1 suppresses

Francisella-induced cytokine response by downregulating the activation of PI3K/Akt pathway and subsequent NFκB activation. Further, our ex-vivo data indicate that SHIP1 does not influence the activation of MAPK pathways. This is interesting because SHIP1 differentially regulates the activation of Erk through FcγR and TLR48,46. SHIP1 suppresses FcγR-mediated but promotes LPS-induced Erk phosphorylation. These stimulus-specific discrepancies in the functioning of SHIP1 may be due to differences in

144

its interaction with other proteins. Thus analysis of these stimulus-specific SHIP1 interactions will help us better understand the functioning of SHIP1 in general.

Akt, the downstream target of SHIP1, enhances the production of LPS-induced

anti-inflammatory IL-10 suggesting a negative role for SHIP1 in LPS-stimulated IL-10

production27,138. However data obtained in our study indicates that SHIP1 promotes

Francisella-induced IL-10 production (we also found that PI3K/Akt pathway inhibits

Francisella-induced IL-10). This apparent disconnect may be better explained by identifying the downstream targets of Akt under these stimulatory conditions. It is

recently reported that Akt promotes LPS-induced IL-10 by phosphorylating and thereby

inhibiting its downstream target GSK3. However the downstream targets of Akt through

which it inhibits Francisella-induced IL-10 production are not known and should be duly

explored to explain these stimulus-specific functional outcomes. In this direction, data

obtained in this study indicate that inhibition of NFκB suppresses Francisella-induced

pro-inflammatory cytokine production but enhances the production of IL-10 suggesting

that some NFκB-dependent mechanisms may feedback to the macrophage and inhibit

subsequent IL-10 production. Consistent with this notion, we have observed in our

laboratory that pharmacological inhibition of NFκB had enhanced the transcriptional

activity of Sp1, a well characterized transcription factor of the IL-10 gene (Rajaram et al,

unpublished observations). Also, blockade of Sp1 activity using a small molecule

inhibitor significantly dampened IL-10 production (Rajaram et al, unpublished

observations). Thus, understanding these NFκB-dependent feedback mechanisms such as

the activation of Sp1 may help us gain more insight into how the SHIP1/PI3K/Akt axis

145

regulates IL-10 production and how this axis may be altered to better resolve Francisella infection.

Based on our ex-vivo data it may be speculated that deletion of SHIP1 may

prolong the survival of mice. However we could not test this possibility because of the

phenotype of the SHIP1 deficient mice. Since the predominant effect observed in

SHIP1-/- macrophages was hyperactivation of Akt, results obtained in another (Myr-Akt)

model may be translatable to the SHIP1 model. Myr-Akt mice express constitutively

active Akt specifically in macrophages. We found in our laboratory that Myr-Akt mice

displayed a survival advantage and decreased bacterial organ burden than their wild type

counterparts when challenged with Francisella novicida. This suggests that

SHIP1/PI3K/Akt axis may play a role in the resolution of Francisella infection.

Although PI3K signaling has been widely believed to promote phagocytosis we

surprisingly found that in our infection model SHIP1 and PI3K/Akt pathway are not

involved in the engulfment of Francisella. Such PI3K-independent phagocytosis has

been described in the case of Salmonella19,20 and Shigella20. This observation further

strengthens the fact that the requirement of various signaling pathways for different

effector functions may be stimulus-specific.

PI3K is an important player in mediating various functions such as inflammation,

phagocytosis, chemotaxis, apoptosis etc. Thus, our data that SHIP1 negatively regulates

PI3K-dependent pathway suggests a broader role for SHIP1 during Francisella infection.

Data obtained from our laboratory suggest that indeed this may be true. We have observed that SHIP1 negatively regulates the production of Francisella-induced

RANTES, a chemokine. Thus, it is possible that during murine tularemia SHIP1 may

146

regulate chemotaxis of different cell types to the site of action and thereby may modulate both the magnitude and the nature of the functional responses. Furthermore, SHIP1 may also regulate the apoptosis of host cells. In this direction, data obtained in our laboratory indicate that SHIP1 regulates the expression of the death receptor Fas in the infected cells

(Rajaram et al., unpublished observations). Further, we observed that SHIP1 and its downstream target Akt modulate the phagolysosomal fusion events and subsequently the intra-macrophage survival of Francisella. All these observations point that SHIP1 has a vital role in modulating macrophage response against Francisella. Thus, it is rational to target SHIP1 for the development of novel and alternative therapeutic measures.

Recently, agonists of SHIP1 have been described. These compounds activate SHIP1 and dampen the PI3K pathway139. But, based on our ex-vivo and in-vivo data it is logical to inhibit SHIP1 but not to activate it. However no small molecule inhibitors of SHIP1 are available to date. Thus, several technical challenges must be addressed before the findings in the Chapter 3 can be translated to therapy.

Francisella subverts IFNγ-mediated host cell immune response

Cytokines such as IL-12 and IL-23 can feedback to cells such as NK cells and T cells to trigger the release of IFNγ which activates the macrophages to destroy invading pathogens (Figure 5.1). Also, IFNγ primes mononuclear phagocytes to produce enhanced amounts of IL-12 and IL-23 resulting in an amplification loop. Data obtained in our laboratory provide evidence for the existence of such amplification loop between

Francisella-infected monocytes and NK cells. IL-23 released from infected monocytes stimulated IFNγ production from NK cells75. Interestingly, infected human monocytes 147

secreted limited amounts of IFNγ as well. Moreover, IFNγ priming of human monocytes

enhanced the release of IL-23 during Francisella infection (Butchar et al, unpublished

observations). Attesting to the importance of this amplification loop, neutralization of IL-

23 or IL-12 or IFNγ led to enhanced survival of the bacteria (Butchar et al, unpublished

observations). Some pathogens have developed mechanisms to interfere with such IFNγ- mediated protective effects. Therefore, we examined the interference of IFNγ-mediated host response by Francisella and our results presented in Chapter 5 demonstrate that

Francisella suppresses IFNγ-induced host response. Specifically we have shown that

Francisella inhibits IFNγ-induced STAT1 phosphorylation and the subsequent induction of STAT1-dependent proteins such as iNOS, an antimicrobial enzyme, leading to enhanced survival of the pathogen inside the macrophages (Figure 5.1). We also showed that replication, phagosomal escape, viability and phagocytosis of bacteria are not essential for the Francisella-mediated signaling interference. Further using RNAi we demonstrated that Francisella-induced SOCS3 dampens IFNγ-mediated STAT1 phosphorylation (Figure 5.1).

Although this subversion mechanism is not unique to Francisella, understanding the basis for this immune evasion mechanism sheds light on the fundamental principles of the host-pathogen interactions. Moreover, understanding the mechanism of this subversion is important because striking differences may exist in the manner in which subversion is achieved under different contexts. Our data suggest that indeed this is true.

For example, in contrast to Francisella, Mycobacterium has been shown to inhibit IFNγ-

mediated response by disrupting interaction of STAT1 with its co-activators but not by

suppressing the activation of STAT1 itself111. Further, unlike Leishmania, Francisella 148

does not induce SHP-1 and fails to trigger the degradation of JAKs and STAT protein50.

On the other hand a few similarities also exist. For example, both Burkholderia and

Francisella induce SOCS3 but only the former induces SOCS1 another negative regulator of IFNγ signaling119. Thus a comprehensive examination of the subversion mechanisms will in general help us better understand the host-pathogen interactions and permit the development of effective and specific therapeutic strategies.

In this study, the effect of Francisella on the surface expression of IFNγR was not

addressed. It is possible that the expression of the receptor is altered during Francisella infection. However, since we have found that the phosphorylation of downstream JAK kinases is not suppressed in the infected cells we would propose that the receptor levels

may not have been changed. But still this is an open question that requires further

examination. The findings of this project did not rule out the possibility of the involvement of a tyrosine phosphatase of STAT1 in the suppression of STAT1 phosphorylation. Recently, a nuclear tyrosine phosphatase of STAT1, TC45, has been identified140. TC45 is the nuclear isoform of the T-cell PTP (TC-PTP). Our microarrary

data revealed that the TC-PTP gene (gene symbol PTPN2) is upregulated by

approximately 4.5 fold during Francisella infection. Thus, the role of this protein in

Francisella-mediated suppression of IFNγ-induced STAT1 phosphorylation needs to be

investigated to further delineate the molecular mechanisms.

Our data indicate that Francisella-induced SOCS3 expression dampens IFNγ- mediated STAT1 phosphorylation. However, the molecular mechanisms that regulate the

Francisella-induced SOCS3 protein are not addressed in this study. Recent reports indicate a potential role for the MAPKs Erk and p38 in the induction of SOCS3141,142. 149

MKK6/p38MAPK/MK2 cascade was demonstrated to be critical for TNFα-induced

SOCS3 expression in fibroblasts. Therefore, the potential role of these pathways in

Francisella-induced SOCS3 expression needs to be addressed.

Based on our in vitro data SOCS3 may be an important target to treat Francisella

infection. To date there are no pharmacological inhibitors available that target SOCS3.

Further, SOCS3 also regulates the signaling through other cytokines which may be

beneficial to the host. Thus a better understanding about the pathways inducing SOCS3

and the mechanism of action of SOCS3 is warranted before any therapy targeting SOCS3

is realized. Alternatively, the pathways that either trigger the induction of SOCS3 or

those that dictate the stability of SOCS3 may be targeted for an effective therapy.

In conclusion, the findings from these studies have significantly contributed to our

understanding of host cell response and also the modulation of this host response against

Francisella. Specifically, this study has identified three key players-Syk, SHIP1 and

SOCS3, which regulate different aspects of host response. Undoubtedly, the findings of these studies lay the foundation for further exciting and significant contributions to the field of tularemia in the future.

150

BIBLIOGRAPHY

1. Medzhitov, R. and C. Janeway, Jr. 2000. Innate immune recognition: mechanisms and pathways. Immunol.Rev. 173:89-97.:89-97.

2. Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu.Rev.Immunol. 21:335-76.:335-376.

3. Underhill, D. M. and A. Ozinsky. 2002. Phagocytosis of microbes: complexity in action. Annu.Rev.Immunol. 20:825-52. Epub;%2001 Dec 7.:825-852.

4. Heifets, L. 1982. Centennial of Metchnikoff's discovery. J.Reticuloendothel.Soc. 31:381-391.

5. Hume, D. A., I. L. Ross, S. R. Himes, R. T. Sasmono, C. A. Wells, and T. Ravasi. 2002. The mononuclear phagocyte system revisited. J.Leukoc.Biol. 72:621-627.

6. Zhang, Y. L. and C. Dong. 2005. MAP kinases in immune responses. Cell Mol.Immunol. 2:20-27.

7. Zhang, X., J. P. Edwards, and D. M. Mosser. 2006. Dynamic and transient remodeling of the macrophage IL-10 promoter during transcription. J.Immunol. 177:1282-1288.

8. Ganesan, L. P., T. Joshi, H. Fang, V. K. Kutala, J. Roda, R. Trotta, A. Lehman, P. Kuppusamy, J. C. Byrd, W. E. Carson, M. A. Caligiuri, and S. Tridandapani. 2006. Fc{gamma}R-induced production of superoxide and inflammatory cytokines is differentially regulated by SHIP through its influence on PI3K and/or Ras/Erk pathways. Blood 108:718-725.

9. Han, B., M. Mura, C. F. Andrade, D. Okutani, M. Lodyga, C. C. dos Santos, S. Keshavjee, M. Matthay, and M. Liu. 2005. TNFalpha-induced long pentraxin PTX3 expression in human lung epithelial cells via JNK. J.Immunol. 175:8303- 8311.

151

10. Itoh, T. and T. Takenawa. 2002. Phosphoinositide-binding domains: Functional units for temporal and spatial regulation of intracellular signalling. Cell Signal. 14:733-743.

11. Koyasu, S. 2003. The role of PI3K in immune cells. Nat.Immunol. 4:313-319.

12. Cox, D., B. M. Dale, M. Kishiwada, C. D. Helgason, and S. Greenberg. 2001. A regulatory role for Src homology 2 domain-containing inositol 5'-phosphatase (SHIP) in phagocytosis mediated by Fc gamma receptors and complement receptor 3 (alpha(M)beta(2); CD11b/CD18. J.Exp.Med. 193:61-71.

13. Leverrier, Y., K. Okkenhaug, C. Sawyer, A. Bilancio, B. Vanhaesebroeck, and A. J. Ridley. 2003. Class I phosphoinositide 3-kinase p110beta is required for apoptotic cell and Fcgamma receptor-mediated phagocytosis by macrophages. J.Biol.Chem. 278:38437-38442.

14. Cox, D., C. C. Tseng, G. Bjekic, and S. Greenberg. 1999. A requirement for phosphatidylinositol 3-kinase in pseudopod extension. J.Biol.Chem. 274:1240- 1247.

15. Araki, N., M. T. Johnson, and J. A. Swanson. 1996. A role for phosphoinositide 3-kinase in the completion of macropinocytosis and phagocytosis by macrophages. J.Cell Bio. 135:1249-1260.

16. Caron, E. and A. Hall. 1998. Identification of two distinct mechanisms of phagocytosis controlled by different Rho GTPases. Science 282:1717-1721.

17. Ganesan, L. P., G. Wei, R. A. Pengal, L. Moldovan, N. Moldovan, M. C. Ostrowski, and S. Tridandapani. 2004. The serine/threonine kinase Akt promotes Fc gamma receptor-mediated phagocytosis in murine macrophages through the activation of p70S6 kinase. J.Biol.Chem. ..

18. Allen, L. A., J. A. Allgood, X. Han, and L. M. Wittine. 2005. Phosphoinositide3- kinase regulates actin polymerization during delayed phagocytosis of Helicobacter pylori. J.Leukoc.Biol. 78:220-230.

19. Drecktrah, D., L. A. Knodler, and O. Steele-Mortimer. 2004. Modulation and utilization of host cell phosphoinositides by Salmonella spp. Infect.Immun. 72:4331-4335. 152

20. Brumell, J. H. and S. Grinstein. 2003. Role of lipid-mediated signal transduction in bacterial internalization. Cell Microbiol. 5:287-297.

21. Chen, Q., D. W. Powell, M. J. Rane, S. Singh, W. Butt, J. B. Klein, and K. R. McLeish. 2003. Akt phosphorylates p47phox and mediates respiratory burst activity in human neutrophils. J.Immunol. 170:5302-5308.

22. Hordijk, P. L. 2006. Regulation of NADPH oxidases: the role of Rac proteins. Circ.Res. 98:453-462.

23. Chen, Q., D. W. Powell, M. J. Rane, S. Singh, W. Butt, J. B. Klein, and K. R. McLeish. 2003. Akt phosphorylates p47phox and mediates respiratory burst activity in human neutrophils. J.Immunol. 170:5302-5308.

24. Sakai, K., H. Suzuki, H. Oda, T. Akaike, Y. Azuma, T. Murakami, K. Sugi, T. Ito, H. Ichinose, S. Koyasu, and M. Shirai. 2006. Phosphoinositide 3-kinase in nitric oxide synthesis in macrophage: critical dimerization of inducible nitric-oxide synthase. J.Biol.Chem. 281:17736-17742.

25. Williams, D. L., C. Li, T. Ha, T. Ozment-Skelton, J. H. Kalbfleisch, J. Preiszner, L. Brooks, K. Breuel, and J. B. Schweitzer. 2004. Modulation of the phosphoinositide 3-kinase pathway alters innate resistance to polymicrobial sepsis. J.Immunol. 172:449-456.

26. Williams, D. L., T. Ozment-Skelton, and C. Li. 2006. Modulation of the phosphoinositide 3-kinase signaling pathway alters host response to sepsis, inflammation, and ischemia/reperfusion injury. Shock 25:432-439.

27. Martin, M., K. Rehani, R. S. Jope, and S. M. Michalek. 2005. Toll-like receptor- mediated cytokine production is differentially regulated by glycogen synthase kinase 3. Nat.Immunol. 6:777-784.

28. Pengal, R. A., L. P. Ganesan, G. Wei, H. Fang, M. C. Ostrowski, and S. Tridandapani. 2005. Lipopolysaccharide-induced production of interleukin-10 is promoted by the serine/threonine kinase Akt. Mol.Immunol. ..

29. Fukao, T. and S. Koyasu. 2003. PI3K and negative regulation of TLR signaling. Trends Immunol. 24:358-363.

153

30. Ghazizadeh, S., J. B. Bolen, and H. B. Fleit. 1994. Physical and functional association of Src-related Protein Tyrosine kinases with FcgRII in monocytic THP-1 cells. J.Biol.Chem. 269:8878-8884.

31. Lorenzi, R., P. M. Brickell, D. R. Katz, C. Kinnon, and A. J. Thrasher. 2000. Wiskott-Aldrich syndrome protein is necessary for efficient IgG-mediated phagocytosis. Blood 95:2943-2946.

32. Crowley, M. T., P. S. Costello, C. J. Fitzer-Attas, M. Turner, F. Y. Meng, C. Lowell, V. L. Tybulewicz, and A. L. DeFranco. 1997. A critical role for Syk in signal transduction and phagocytosis mediated by Fcgamma receptors on macrophages. Journal of Experimental Medicine 186:1027-1039.

33. Fitzer-Attas, C. J., M. Lowry, M. T. Crowley, A. J. Finn, F. Meng, A. L. DeFranco, and C. A. Lowell. 2000. Fcgamma receptor-mediated phagocytosis in macrophages lacking the Src family tyrosine kinases Hck, Fgr, and Lyn. J.Exp.Med. 191:669-682.

34. Shi, Y., Y. Tohyama, T. Kadono, J. He, S. M. Shahjahan Miah, R. Hazama, C. Tanaka, K. Tohyama, and H. Yamamura. 2006. Protein-tyrosine kinase Syk is required for pathogen engulfment in complement-mediated phagocytosis. Blood 107:4554-4562.

35. Beitz, L. O., D. A. Fruman, T. Kurosaki, L. C. Cantley, and A. M. Scharenberg. 1999. SYK is upstream of phosphoinositide 3-kinase in B cell receptor signaling. J.Biol.Chem. 274:32662-32666.

36. Kiefer, F, Brumell, J., Al-Alawi, N., Latour, S., Cheng, A., Veillette, A., Grinstein, S., and Pawson, T. The Syk Protein Tyrosine Kinase Is Essential for Fc Receptor Signaling in Macrophages and Neutrophils. Molecular & Cellular Biology 18, No. 7(July 1998), 4209-4220. 4-20-1998. Ref Type: Generic

37. Underhill, D. M., E. Rossnagle, C. A. Lowell, and R. M. Simmons. 2005. Dectin- 1 activates Syk tyrosine kinase in a dynamic subset of macrophages for reactive oxygen production. Blood 106:2543-2550.

38. Ulanova, M., M. Marcet-Palacios, S. Munoz, S. Asfaha, M. K. Kim, A. D. Schreiber, and A. D. Befus. 2006. Involvement of Syk kinase in TNF-induced

154

nitric oxide production by airway epithelial cells. Biochem.Biophys.Res.Commun. 351:431-437.

39. Krystal, G. 2000. Lipid phosphatases in the immune system. Immunology 12:397- 403.

40. Liu, L., Damen, J. E., Ware, M., Hughes, M., and Krystal, G. SHIP, a new player in cytokine-induced signalling. Leukemia 11, 181-184. 1997. Ref Type: Journal (Full)

41. Ravetch, J. V. and J.-P. Kinet. 1991. Fc receptors. Ann.Rev.Immunol. 9:457-492.

42. Helgason, C. D., J. E. Damen, P. Rosten, R. Grewal, P. Sorensen, S. M. Chappel, A. Borowski, F. Jirik, G. Krystal, and R. K. Humphries. 1998. Targeted disruption of SHIP leads to hemopoietic perturbations, lung pathology, and a shortened life span. Genes Dev. 12:1610-1620.

43. Fang, H., R. A. Pengal, X. Cao, L. P. Ganesan, M. D. Wewers, C. B. Marsh, and S. Tridandapani. 2004. Lipopolysaccharide-Induced Macrophage Inflammatory Response Is Regulated by SHIP. J.Immunol. 173:360-366.

44. Sly, L. M., M. J. Rauh, J. Kalesnikoff, C. H. Song, and G. Krystal. 2004. LPS- induced upregulation of SHIP is essential for endotoxin tolerance. Immunity. 21:227-239.

45. Strassheim, D., J. Y. Kim, J. S. Park, S. Mitra, and E. Abraham. 2005. Involvement of SHIP in TLR2-induced neutrophil activation and acute lung injury. J.Immunol. 174:8064-8071.

46. Fang, H., R. A. Pengal, X. Cao, L. P. Ganesan, M. D. Wewers, C. B. Marsh, and S. Tridandapani. 2004. Lipopolysaccharide-Induced Macrophage Inflammatory Response Is Regulated by SHIP. J.Immunol. 173:360-366.

47. Ai, J., A. Maturu, W. Johnson, Y. Wang, C. B. Marsh, and S. Tridandapani. 2005. The inositol phosphatase SHIP-2 downregulates Fc{gamma}R-mediated phagocytosis in murine macrophages independently of SHIP-1. Blood ..

155

48. Cao, X., G. Wei, H. Fang, J. Guo, M. Weinstein, C. B. Marsh, M. C. Ostrowski, and S. Tridandapani. 2004. The inositol 3-phosphatase PTEN negatively regulates Fc gamma receptor signaling, but supports Toll-like receptor 4 signaling in murine peritoneal macrophages. J.Immunol. 172:4851-4857.

49. Mosser, D. M. 2003. The many faces of macrophage activation. J.Leukoc.Biol. 73:209-212.

50. Ray, M., A. A. Gam, R. A. Boykins, and R. T. Kenney. 2000. Inhibition of interferon-gamma signaling by Leishmania donovani. J.Infect.Dis. 181:1121- 1128.

51. Hussain, S., B. S. Zwilling, and W. P. Lafuse. 1999. Mycobacterium avium infection of mouse macrophages inhibits IFN-gamma Janus kinase-STAT signaling and gene induction by down-regulation of the IFN-gamma receptor. J.Immunol. 163:2041-2048.

52. Ting, L. M., A. C. Kim, A. Cattamanchi, and J. D. Ernst. 1999. Mycobacterium tuberculosis inhibits IFN-gamma transcriptional responses without inhibiting activation of STAT1. J.Immunol. 163:3898-3906.

53. Oyston, P. C., A. Sjostedt, and R. W. Titball. 2004. Tularaemia: bioterrorism defence renews interest in Francisella tularensis. Nat.Rev.Microbiol. 2:967-978.

54. Dennis, D. T., T. V. Inglesby, D. A. Henderson, J. G. Bartlett, M. S. Ascher, E. Eitzen, A. D. Fine, A. M. Friedlander, J. Hauer, M. Layton, S. R. Lillibridge, J. E. McDade, M. T. Osterholm, T. O'Toole, G. Parker, T. M. Perl, P. K. Russell, and K. Tonat. 2001. Tularemia as a biological weapon: medical and public health management. JAMA 285:2763-2773.

55. Clemens, D. L., B. Y. Lee, and M. A. Horwitz. 2005. Francisella tularensis enters macrophages via a novel process involving pseudopod loops. Infect.Immun. 73:5892-5902.

56. Balagopal, A., A. S. MacFarlane, N. Mohapatra, S. Soni, J. S. Gunn, and L. S. Schlesinger. 2006. Characterization of the receptor-ligand pathways important for entry and survival of Francisella tularensis in human macrophages. Infect.Immun. 74:5114-5125.

156

57. Pierini, L. M. 2006. Uptake of serum-opsonized Francisella tularensis by macrophages can be mediated by class A scavenger receptors. Cell Microbiol. 8:1361-1370.

58. Checroun, C., T. D. Wehrly, E. R. Fischer, S. F. Hayes, and J. Celli. 2006. Autophagy-mediated reentry of Francisella tularensis into the endocytic compartment after cytoplasmic replication. Proc.Natl.Acad.Sci.U.S.A 103:14578- 14583.

59. Elkins, K. L., S. C. Cowley, and C. M. Bosio. 2007. Innate and adaptive immunity to Francisella. Ann.N.Y.Acad.Sci. 1105:284-324.

60. Bosio, C. M. and S. W. Dow. 2005. Francisella tularensis induces aberrant activation of pulmonary dendritic cells. J.Immunol. 175:6792-6801.

61. Katz, J., P. Zhang, M. Martin, S. N. Vogel, and S. M. Michalek. 2006. Toll-like receptor 2 is required for inflammatory responses to Francisella tularensis LVS. Infect.Immun. 74:2809-2816.

62. Loegering, D. J., J. R. Drake, J. A. Banas, T. L. McNealy, D. G. Mc Arthur, L. M. Webster, and M. R. Lennartz. 2006. Francisella tularensis LVS grown in macrophages has reduced ability to stimulate the secretion of inflammatory cytokines by macrophages in vitro. Microb.Pathog. 41:218-225.

63. Telepnev, M., I. Golovliov, and A. Sjostedt. 2005. Francisella tularensis LVS initially activates but subsequently down-regulates intracellular signaling and cytokine secretion in mouse monocytic and human peripheral blood mononuclear cells. Microb.Pathog. 38:239-247.

64. Elkins, K. L., A. Cooper, S. M. Colombini, S. C. Cowley, and T. L. Kieffer. 2002. In vivo clearance of an intracellular bacterium, Francisella tularensis LVS, is dependent on the p40 subunit of interleukin-12 (IL-12) but not on IL-12 p70. Infect.Immun. 70:1936-1948.

65. Stenmark, S. and A. Sjostedt. 2004. Transfer of specific antibodies results in increased expression of TNF-alpha and IL12 and recruitment of neutrophils to the site of a cutaneous Francisella tularensis infection. J.Med.Microbiol. 53:501-504.

157

66. Leiby, D. A., A. H. Fortier, R. M. Crawford, R. D. Schreiber, and C. A. Nacy. 1992. In vivo modulation of the murine immune response to Francisella tularensis LVS by administration of anticytokine antibodies. Infect.Immun. 60:84-89.

67. Lopez, M. C., N. S. Duckett, S. D. Baron, and D. W. Metzger. 2004. Early activation of NK cells after lung infection with the intracellular bacterium, Francisella tularensis LVS. Cell Immunol. 232:75-85.

68. Santic, M., M. Molmeret, and K. Y. Abu. 2005. Modulation of biogenesis of the Francisella tularensis subsp. novicida-containing phagosome in quiescent human macrophages and its maturation into a phagolysosome upon activation by IFN- gamma. Cell Microbiol. 7:957-967.

69. Lindgren, H., S. Stenmark, W. Chen, A. Tarnvik, and A. Sjostedt. 2004. Distinct roles of reactive nitrogen and oxygen species to control infection with the facultative intracellular bacterium Francisella tularensis. Infect.Immun. 72:7172- 7182.

70. Clemens, D. L., B. Y. Lee, and M. A. Horwitz. 2004. Virulent and avirulent strains of Francisella tularensis prevent acidification and maturation of their phagosomes and escape into the cytoplasm in human macrophages. Infect.Immun. 72:3204-3217.

71. Rajaram, M. V., L. P. Ganesan, K. V. Parsa, J. P. Butchar, J. S. Gunn, and S. Tridandapani. 2006. Akt/Protein kinase B modulates macrophage inflammatory response to Francisella infection and confers a survival advantage in mice. J.Immunol. 177:6317-6324.

72. Woolard, M. D., J. E. Wilson, L. L. Hensley, L. A. Jania, T. H. Kawula, J. R. Drake, and J. A. Frelinger. 2007. Francisella tularensis-infected macrophages release prostaglandin E2 that blocks T cell proliferation and promotes a Th2-like response. J.Immunol. 178:2065-2074.

73. Gavrilin, M. A., I. J. Bouakl, N. L. Knatz, M. D. Duncan, M. W. Hall, J. S. Gunn, and M. D. Wewers. 2005. Internalization and phagosome escape required for Francisella to induce human monocyte IL-1{beta} processing and release. Proc.Natl.Acad.Sci.U.S.A. ..

158

74. Mariathasan, S., D. S. Weiss, V. M. Dixit, and D. M. Monack. 2005. Innate immunity against Francisella tularensis is dependent on the ASC/caspase-1 axis. J.Exp.Med. 202:1043-1049.

75. Butchar, J. P., M. V. Rajaram, L. P. Ganesan, K. V. Parsa, C. D. Clay, L. S. Schlesinger, and S. Tridandapani. 2007. Francisella tularensis induces IL-23 production in human monocytes. J.Immunol. 178:4445-4454.

76. Parsa, K. V., L. P. Ganesan, M. V. Rajaram, M. A. Gavrilin, A. Balagopal, N. P. Mohapatra, M. D. Wewers, L. S. Schlesinger, J. S. Gunn, and S. Tridandapani. 2006. Macrophage pro-inflammatory response to Francisella novicida infection is regulated by SHIP. PLoS.Pathog. 2:e71.

77. Crowley, M. T., P. S. Costello, C. J. Fitzer-Attas, M. Turner, F. Meng, C. Lowell, V. L. J. Tybuleewicz, and A. L. DeFranco. 1997. A critical role for Syk signal transduction and phagocytosis mediated by Fcy recptors on macrophages. J.Exp.Med. 186:1027-1039.

78. Kugler, S., S. Schuller, and W. Goebel. 1997. Involvement of MAP-kinases and - phosphatases in uptake and intracellular replication of Listeria monocytogenes in J774 macrophage cells. FEMS Microbiol.Lett. 157:131-136.

79. Garcia-Garcia, E., R. Rosales, and C. Rosales. 2002. Phosphatidylinositol 3- kinase and extracellular signal-regulated kinase are recruited for Fc receptor- mediated phagocytosis during monocyte-to-macrophage differentiation. J.Leukoc.Biol. 72:107-114.

80. de Winter, P., R. C. Rayne, and G. M. Coast. 2007. The effects of intracellular signalling pathway inhibitors on phagocytosis by haemocytes of Manduca sexta. J.Insect Physiol.

81. Garcia-Garcia, E., G. Sanchez-Mejorada, and C. Rosales. 2001. Phosphatidylinositol 3-kinase and ERK are required for NF-kappaB activation but not for phagocytosis. J.Leukoc.Biol. 70:649-658.

82. Lin, L. L., M. Wartmann, A. Y. Lin, J. L. Knopf, A. Seth, and R. J. Davis. 1993. cPLA2 is phosphorylated and activated by MAP kinase. Cell 72:269-278.

159

83. Chen, X., D. S. Allan, K. Krzewski, B. Ge, H. Kopcow, and J. L. Strominger. 2006. CD28-stimulated ERK2 phosphorylation is required for polarization of the microtubule organizing center and granules in YTS NK cells. Proc.Natl.Acad.Sci.U.S.A 103:10346-10351.

84. Harrison, R. E., B. A. Sikorski, and J. Jongstra. 2004. Leukocyte-specific protein 1 targets the ERK/MAP kinase scaffold protein KSR and MEK1 and ERK2 to the actin cytoskeleton. J.Cell Sci. 117:2151-2157.

85. Vachon, E., R. Martin, J. Plumb, V. Kwok, R. W. Vandivier, M. Glogauer, A. Kapus, X. Wang, C. W. Chow, S. Grinstein, and G. P. Downey. 2006. CD44 is a phagocytic receptor. Blood 107:4149-4158.

86. Jabril-Cuenod, B., C. Zhang, A. M. Scharenberg, R. Paolini, R. Numerof, M. A. Beaven, and J. P. Kinet. 1996. Syk-dependent phosphorylation of Shc. A potential link between FcepsilonRI and the Ras/mitogen-activated protein kinase signaling pathway through SOS and Grb2. J.Biol.Chem. 271:16268-16272.

87. Ueda, Y., S. Hirai, S. Osada, A. Suzuki, K. Mizuno, and S. Ohno. 1996. Protein kinase C activates the MEK-ERK pathway in a manner independent of Ras and dependent on Raf. J.Biol.Chem. 271:23512-23519.

88. Song, X., S. Tanaka, D. Cox, and S. C. Lee. 2004. Fcgamma receptor signaling in primary human microglia: differential roles of PI-3K and Ras/ERK MAPK pathways in phagocytosis and chemokine induction. J.Leukoc.Biol. 75:1147-1155.

89. Ulanova, M., S. Asfaha, G. Stenton, A. Lint, D. Gilbertson, A. Schreiber, and D. Befus. 2007. Involvement of Syk protein tyrosine kinase in LPS-induced responses in macrophages. J.Endotoxin.Res. 13:117-125.

90. Ulanova, M., M. Marcet-Palacios, S. Munoz, S. Asfaha, M. K. Kim, A. D. Schreiber, and A. D. Befus. 2006. Involvement of Syk kinase in TNF-induced nitric oxide production by airway epithelial cells. Biochem.Biophys.Res.Commun. 351:431-437.

91. Santic, M., M. Molmeret, J. R. Barker, K. E. Klose, Dekanic A., M. Doric, and Y. A. Kwaik. 2007. A Francisella tularensis pathogenicity island proteins essential for bacterial proliferation within the host cell cytosol. Cell Microbiol.In Press.

160

92. Santic, M., M. Molmeret, K. E. Klose, S. Jones, and Y. A. Kwaik. 2005. The Francisella tularensis pathogenicity island protein IglC and its regulator MglA are essential for modulating phagosome biogenesis and subsequent bacterial escape into the cytoplasm. Cell Microbiol. 7:969-979.

93. Tridandapani, S., G. W. Chacko, M. C. J. Van Bruggen, and K. M. Coggeshall. 1997. Negative signaling in B cells Causes reduced Ras activity by reducing Shc- Grb2 interactions. Journal of Immunology 158:1125-1132.

94. Tridandapani, S., M. Pradhan, J. R. LaDine, C. L. Anderson, and K. M. Coggeshall. 1998. Protein interactions of SHIP: association with Shc displaces SHIP from FcyRIIb in B cells. J Immunol. 162:1408-1414.

95. Horwitz, M. A. 1984. Phagocytosis of the Legionnaires' disease bacterium (Legionella pneumophila) occurs by a novel mechanism: engulfment within a pseudopod coil. Cell. 36:27-33.

96. Stenmark, S., D. Sunnemark, A. Bucht, and A. Sjostedt. 1999. Rapid local expression of interleukin-12, tumor necrosis factor alpha, and gamma interferon after cutaneous Francisella tularensis infection in tularemia-immune mice. Infect.Immun. 67:1789-1797.

97. Elkins, K. L., T. R. Rhinehart-Jones, S. J. Culkin, D. Yee, and R. K. Winegar. 1996. Minimal requirements for murine resistance to infection with Francisella tularensis LVS. Infect.Immun. 64:3288-3293.

98. Elkins, K. L., T. Rhinehart-Jones, C. A. Nacy, R. K. Winegar, and A. H. Fortier. 1993. T-cell-independent resistance to infection and generation of immunity to Francisella tularensis. Infect.Immun. 61:823-829.

99. Chen, W., R. KuoLee, H. Shen, M. Busa, and J. W. Conlan. 2004. Toll-like receptor 4 (TLR4) does not confer a resistance advantage on mice against low- dose aerosol infection with virulent type A Francisella tularensis. Microb.Pathog. 37:185-191.

100. Chen, W., R. KuoLee, H. Shen, M. Busa, and J. W. Conlan. 2005. Toll-like receptor 4 (TLR4) plays a relatively minor role in murine defense against primary intradermal infection with Francisella tularensis LVS. Immunol.Lett. 97:151-154.

161

101. Ancuta, P., T. Pedron, R. Girard, G. Sandstrom, and R. Chaby. 1996. Inability of the Francisella tularensis lipopolysaccharide to mimic or to antagonize the induction of cell activation by endotoxins. Infect.Immun. 64:2041-2046.

102. Rauh, M. J., J. Kalesnikoff, M. Hughes, L. Sly, V. Lam, and G. Krystal. 2003. Role of Src homology 2-containing-inositol 5'-phosphatase (SHIP) in mast cells and macrophages. Biochem.Soc.Trans. 31:286-291.

103. Parihar, R., R. Trotta, J. M. Roda, A. K. Ferketich, S. Tridandapani, M. A. Caligiuri, and W. E. Carson, III. 2005. Src homology 2-containing inositol 5'- phosphatase 1 negatively regulates IFN-gamma production by natural killer cells stimulated with antibody-coated tumor cells and interleukin-12. Cancer Res. 65:9099-9107.

104. Trotta, R., R. Parihar, J. Yu, B. Becknell, J. Allard, J. Wen, W. Ding, H. Mao, S. Tridandapani, W. E. Carson, and M. A. Caligiuri. 2005. Differential expression of SHIP1 in CD56bright and CD56dim NK cells provides a molecular basis for distinct functional responses to monokine costimulation. Blood 105:3011-3018.

105. Lindgren, H., I. Golovliov, V. Baranov, R. K. Ernst, M. Telepnev, and A. Sjostedt. 2004. Factors affecting the escape of Francisella tularensis from the phagolysosome. J.Med.Microbiol. 53:953-958.

106. Mohapatra, N. P., A. Balagopal, S. Soni, L. S. Schlesinger, and J. S. Gunn. 2007. AcpA is a Francisella acid phosphatase that affects intramacrophage survival and virulence. Infect.Immun. 75:390-396.

107. Mohapatra, N. P., S. Soni, B. L. Bell, R. Warren, R. K. Ernst, A. Muszynski, R. W. Carlson, and J. S. Gunn. 2007. Identification of an orphan response regulator required for Francisella virulence and transcription of pathogenicity island genes. Infect.Immun.

108. Schroder, K., P. J. Hertzog, T. Ravasi, and D. A. Hume. 2004. Interferon-gamma: an overview of signals, mechanisms and functions. J.Leukoc.Biol. 75:163-189.

109. Stoiber, D., S. Stockinger, P. Steinlein, J. Kovarik, and T. Decker. 2001. Listeria monocytogenes modulates macrophage cytokine responses through STAT serine phosphorylation and the induction of suppressor of cytokine signaling 3. J.Immunol. 166:466-472.

162

110. Lindgren, H., L. Stenman, A. Tarnvik, and A. Sjostedt. 2005. The contribution of reactive nitrogen and oxygen species to the killing of Francisella tularensis LVS by murine macrophages. Microbes.Infect. 7:467-475.

111. Kincaid, E. Z. and J. D. Ernst. 2003. Mycobacterium tuberculosis exerts gene- selective inhibition of transcriptional responses to IFN-gamma without inhibiting STAT1 function. J.Immunol. 171:2042-2049.

112. Henry, T., A. Brotcke, D. S. Weiss, L. J. Thompson, and D. M. Monack. 2007. Type I interferon signaling is required for activation of the inflammasome during Francisella infection. J.Exp.Med. 204:987-994.

113. Golovliov, I., V. Baranov, Z. Krocova, H. Kovarova, and A. Sjostedt. 2003. An attenuated strain of the facultative intracellular bacterium Francisella tularensis can escape the phagosome of monocytic cells. Infect.Immun. 71:5940-5950.

114. Bertholet, S., H. L. Dickensheets, F. Sheikh, A. A. Gam, R. P. Donnelly, and R. T. Kenney. 2003. Leishmania donovani-induced expression of suppressor of cytokine signaling 3 in human macrophages: a novel mechanism for intracellular parasite suppression of activation. Infect.Immun. 71:2095-2101.

115. Imai, K., T. Kurita-Ochiai, and K. Ochiai. 2003. Mycobacterium bovis bacillus Calmette-Guerin infection promotes SOCS induction and inhibits IFN-gamma- stimulated JAK/STAT signaling in J774 macrophages. FEMS Immunol.Med.Microbiol. 39:173-180.

116. Crespo, A., M. B. Filla, and W. J. Murphy. 2002. Low responsiveness to IFN- gamma, after pretreatment of mouse macrophages with lipopolysaccharides, develops via diverse regulatory pathways. Eur.J.Immunol. 32:710-719.

117. Banaiee, N., E. Z. Kincaid, U. Buchwald, W. R. Jacobs, Jr., and J. D. Ernst. 2006. Potent inhibition of macrophage responses to IFN-gamma by live virulent Mycobacterium tuberculosis is independent of mature mycobacterial lipoproteins but dependent on TLR2. J.Immunol. 176:3019-3027.

118. Lafuse, W. P., G. R. Alvarez, H. M. Curry, and B. S. Zwilling. 2006. Mycobacterium tuberculosis and Mycobacterium avium inhibit IFN- gamma - induced gene expression by TLR2-dependent and independent pathways. J.Interferon Cytokine Res. 26:548-561.

163

119. Ekchariyawat, P., S. Pudla, K. Limposuwan, S. Arjcharoen, S. Sirisinha, and P. Utaisincharoen. 2005. Burkholderia pseudomallei-induced expression of suppressor of cytokine signaling 3 and cytokine-inducible src homology 2- containing protein in mouse macrophages: a possible mechanism for suppression of the response to gamma interferon stimulation. Infect.Immun. 73:7332-7339.

120. Kile, B. T. and W. S. Alexander. 2001. The suppressors of cytokine signalling (SOCS). Cell Mol.Life Sci. 58:1627-1635.

121. Nicholson, S. E., T. A. Willson, A. Farley, R. Starr, J. G. Zhang, M. Baca, W. S. Alexander, D. Metcalf, D. J. Hilton, and N. A. Nicola. 1999. Mutational analyses of the SOCS proteins suggest a dual domain requirement but distinct mechanisms for inhibition of LIF and IL-6 signal transduction. EMBO J. 18:375-385.

122. Bjorbak, C., H. J. Lavery, S. H. Bates, R. K. Olson, S. M. Davis, J. S. Flier, and M. G. Myers, Jr. 2000. SOCS3 mediates feedback inhibition of the leptin receptor via Tyr985. J.Biol.Chem. 275:40649-40657.

123. Cohney, S. J., D. Sanden, N. A. Cacalano, A. Yoshimura, A. Mui, T. S. Migone, and J. A. Johnston. 1999. SOCS-3 is tyrosine phosphorylated in response to interleukin-2 and suppresses STAT5 phosphorylation and lymphocyte proliferation. Mol.Cell Biol. 19:4980-4988.

124. Hansen, J. A., K. Lindberg, D. J. Hilton, J. H. Nielsen, and N. Billestrup. 1999. Mechanism of inhibition of growth hormone receptor signaling by suppressor of cytokine signaling proteins. Mol.Endocrinol. 13:1832-1843.

125. Anthony, L. S., E. Ghadirian, F. P. Nestel, and P. A. Kongshavn. 1989. The requirement for gamma interferon in resistance of mice to experimental tularemia. Microb.Pathog. 7:421-428.

126. Pammit, M. A., E. K. Raulie, C. M. Lauriano, K. E. Klose, and B. P. Arulanandam. 2006. Intranasal vaccination with a defined attenuated Francisella novicida strain induces gamma interferon-dependent antibody-mediated protection against tularemia. Infect.Immun. 74:2063-2071.

127. Nakane, A., T. Minagawa, M. Kohanawa, Y. Chen, H. Sato, M. Moriyama, and N. Tsuruoka. 1989. Interactions between endogenous gamma interferon and tumor necrosis factor in host resistance against primary and secondary Listeria monocytogenes infections. Infect.Immun. 57:3331-3337. 164

128. Conlan, J. W., R. KuoLee, H. Shen, and A. Webb. 2002. Different host defences are required to protect mice from primary systemic vs pulmonary infection with the facultative intracellular bacterial pathogen, Francisella tularensis LVS. Microb.Pathog. 32:127-134.

129. Cario, E., G. Gerken, and D. K. Podolsky. 2007. Toll-like receptor 2 controls mucosal inflammation by regulating epithelial barrier function. Gastroenterology 132:1359-1374.

130. Arbibe, L., J. P. Mira, N. Teusch, L. Kline, M. Guha, N. Mackman, P. J. Godowski, R. J. Ulevitch, and U. G. Knaus. 2000. Toll-like receptor 2-mediated NF-kappa B activation requires a Rac1-dependent pathway. Nat.Immunol. 1:533- 540.

131. Liljeroos, M., R. Vuolteenaho, S. Morath, T. Hartung, M. Hallman, and M. Ojaniemi. 2007. Bruton's tyrosine kinase together with PI 3-kinase are part of Toll-like receptor 2 multiprotein complex and mediate LTA induced Toll-like receptor 2 responses in macrophages. Cell Signal. 19:625-633.

132. Chaudhary, A., T. M. Fresquez, and M. J. Naranjo. 2007. Tyrosine kinase Syk associates with toll-like receptor 4 and regulates signaling in human monocytic cells. Immunol.Cell Biol. 85:249-256.

133. Ravetch, J. V. and J.-P. Kinet. 1991. Fc receptors. Ann.Rev.Immunol. 9:457-492.

134. Baran, C. P., S. Tridandapani, C. D. Helgason, R. K. Humphries, G. Krystal, and C. B. Marsh. 2003. The inositol 5'-phosphatase SHIP-1 and the Src kinase Lyn negatively regulate macrophage colony-stimulating factor-induced Akt activity. J.Biol.Chem. 278:38628-38636.

135. Damen, J. E., M. D. Ware, J. Kalesnikoff, M. R. Hughes, and G. Krystal. 2001. SHIP's C-terminus is essential for its hydrolysis of PIP3 and inhibition of mast cell degranulation. Blood 97:1343-1351.

136. Cambier, J. C., D. Fong, and I. Tamir. 1999. The unexpected complexity of Fc gamma RIIB signal transduction. Curr.Top.Microbiol.Immunol. 244:43-55.:43- 55.

165

137. Aman, M. J., S. F. Walk, M. E. March, H. P. Su, D. J. Carver, and K. S. Ravichandran. 2000. Essential role for the C-terminal noncatalytic region of SHIP in FcgammaRIIB1-mediated inhibitory signaling. Mol.Cell Biol. 20:3576-3589.

138. Pengal, R. A., L. P. Ganesan, G. Wei, H. Fang, M. C. Ostrowski, and S. Tridandapani. 2006. Lipopolysaccharide-induced production of interleukin-10 is promoted by the serine/threonine kinase Akt. Mol.Immunol. 43:1557-1564.

139. Ong, C. J., A. Ming-Lum, M. Nodwell, A. Ghanipour, L. Yang, D. E. Williams, J. Kim, L. Demirjian, P. Qasimi, J. Ruschmann, L. P. Cao, K. Ma, S. W. Chung, V. Duronio, R. J. Andersen, G. Krystal, and A. L. Mui. 2007. Small-molecule agonists of SHIP1 inhibit the phosphoinositide 3-kinase pathway in hematopoietic cells. Blood 110:1942-1949.

140. ten Hoeve, J., J. de, I, Y. Fu, W. Zhu, M. Tremblay, M. David, and K. Shuai. 2002. Identification of a nuclear Stat1 protein tyrosine phosphatase. Mol.Cell Biol. 22:5662-5668.

141. Uchiya, K. and T. Nikai. 2005. Salmonella pathogenicity island 2-dependent expression of suppressor of cytokine signaling 3 in macrophages. Infect.Immun. 73:5587-5594.

142. Ehlting, C., W. S. Lai, F. Schaper, E. D. Brenndorfer, R. J. Matthes, P. C. Heinrich, S. Ludwig, P. J. Blackshear, M. Gaestel, D. Haussinger, and J. G. Bode. 2007. Regulation of suppressor of cytokine signaling 3 (SOCS3) mRNA stability by TNF-alpha involves activation of the MKK6/p38MAPK/MK2 cascade. J.Immunol. 178:2813-2826.

166