Duke University Dissertation Template

The Role of Irgm1 in Mitochondrial Dynamics and Metabolism

Elyse Schmidt

Department of Molecular Genetics and Microbiology Duke University

Date:______Approved:

______Gregory Taylor, Supervisor

______Jörn Coers

______Tso-Pang Yao

______Nancie MacIver

______David Pickup

Dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Molecular Genetics and Microbiology in the Graduate School of Duke University

2017

ABSTRACT

The Role of Irgm1 in Mitochondrial Dynamics and Metabolism

Elyse Schmidt

Department of Molecular Genetics and Microbiology Duke University

Date:______Approved:

______Gregory Taylor, Supervisor

______Jörn Coers

______Tso-Pang Yao

______Nancie MacIver

______David Pickup

An abstract of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Molecular Genetics and Microbiology in the Graduate School of Duke University

2017

Abstract

The Immunity-Related GTPases (IRG) are a family of proteins that are induced by interferon (IFN)-γ and play pivotal roles in immune and inflammatory responses. IRGs ostensibly function as dynamin-like proteins that bind to intracellular membranes, and promote remodeling and trafficking of those membranes. Prior studies have shown that loss of Irgm1 in mice leads to increased lethality to bacterial infections, as well as enhanced inflammation to non-infectious stimuli; however, the mechanisms underlying these phenotypes are unclear. In this dissertation, I studied the role of Irgm1 in mitochondrial biology and immunometabolism.

Past studies of Irgm1’s human orthologue, IRGM, reported that IRGM localized to mitochondria and induced mitochondrial fragmentation. Further, absence of IRGM inhibited the cell’s ability to undergo IFN-γ and starvation induced autophagy and promoted the formation of elongated mitochondrial networks. In the first chapter, I confirmed that mouse Irgm1 shares IRGM’s mitochondrial localization, induced mitochondrial fragmentation, and that its absence promoted a mitochondrial hyperfused state. The structural determinants required for Irgm1’s mitochondrial localization and

Irgm1-mediated mitochondrial fragmentation were also identified.

In chapter two, I studied the metabolism of Irgm1-deficient embryonic fibroblasts and macrophages through a series of bioenergetic and metabolomic assays, and found a

number of metabolic phenotypes in Irgm1-deficient cells suggesting enhanced proinflammatory activation.

In chapter 3, I describe the increased pro-inflammatory cytokine production observed in our Irgm1-deficient macrophages. A series of metabolic studies indicated that the enhanced cytokine production was associated with marked metabolic changes in the

Irgm1-deficient macrophages, including increased glycolysis and an accumulation of long chain acylcarnitines, whereas Irgm1-deficient macrophages exposed to the glycolytic inhibitor, 2-deoxyglucose, or fatty acid synthase inhibitors resulted in dampening of the excessive cytokine production. Finally, Irgm1-deficient mice displayed high levels of serum cytokines typifying profound auto-inflammation. Taken together, these results suggest that Irgm1-deficiency drives metabolic dysfunction in macrophages in a manner that is cell autonomous and independent of infectious triggers. This may be a significant contributor to excessive inflammation seen in Irgm1- deficient mice.

Dedication

This dissertation is dedicated first to my grandfather, who has weathered more challenges, and accomplished more in his life by the age of thirty with an eighth-grade education that I have up to this point in my life. This is for you Opa.

This manuscript is dedicated to my parents and my husband. Without their hard work and encouragement I would not be here.

Contents

Abstract ...... iv

List of Figures ...... xii

1. Introduction ...... 1

1.1 Interferon Signaling in Innate Immunity ...... 1

1.1.1 Introduction to Interferon ...... 1

1.1.2 Interferon Signaling ...... 2

1.1.2.1 Type I Interferon Signaling Pathway ...... 2

1.1.2.2 Type II Interferon Signaling Pathway ...... 4

1.1.3 Interferons in Immunity ...... 5

1.2 Immunity related GTPases ...... 6

1.2.1 Features of the IFN-Inducible GTPases ...... 6

1.2.2 The IRG Protein Family ...... 8

1.2.2.1 The Proteomic and Structural Features of the IRG Family ...... 8

1.2.2.2 IRG Subfamilies ...... 9

1.2.3 The Murine IRGs ...... 11

1.2.3.1 Murine IRGs Interactions in Lipid and Membrane Binding ...... 12

1.2.3.2 Murine IRGs in Intracellular Pathogen Defense ...... 13

1.2.3.3 Mechanisms of IRG-Mediated Pathogen Defense ...... 14

1.2.4 Functions of Murine Irgm1 ...... 16

1.2.5 The Human IRGs ...... 20

vii

1.2.5.1 Parity of Function between Mouse Irgm1 and Human IRGM ...... 23

1.3 Immunometabolism of Proinflammatory Macrophages ...... 25

1.3.1 Modulation of Immunometabolism in Proinflammatory Macrophages...... 26

1.3.2 Glycolytic Metabolism drives M1 Mitochondrial Polarization ...... 28

1.3.3 The Pentose Phosphate Pathway Acts as a Hub to Sustain Carbohydrate Metabolism and the Redox Demands of M1 Macrophages ...... 32

1.3.4 Fatty Acid Synthesis and Oxidation in Macrophage Polarization ...... 34

1.3.5 The Tricarboxylic Acid Cycle is Broken in Two Places During M1 Macrophage Polarization ...... 37

1.4 Mitochondrial Dynamics and Their Functions ...... 39

1.4.1 Mechanisms of mitochondrial fission and fusion ...... 40

1.4.1.1 Mitochondrial Fission ...... 40

1.4.1.2 Mitochondrial Fusion ...... 42

1.4.2 Mitochondrial Dynamics in Response to Stress ...... 43

1.4.3 Mitochondrial Dynamics in Metabolism ...... 45

1.5 Motivation for this Work ...... 47

2. Irgm1-Deficiency in MEFs Causes Changes in Mitochondrial Morphology ...... 50

2.1 Introduction ...... 50

2.2 Results ...... 53

2.2.1 Evidence of Altered Mitochondrial Morphology in Irgm1-Deficient Enterocytes ...... 53

2.2.2 A Subset of Irgm1 Localizes to the Mitochondria and Induces Mitochondrial Fragmentation ...... 54

viii

2.2.3 Characterization of the Structural Elements Required for Irgm1’s Mitochondrial Localization and Irgm1-Dependent Mitochondrial Fragmentation . 57

2.2.4 Irgm1-Dependent Mitochondrial Fragmentation is Dependent on Drp1 Activity...... 61

2.2.5 Mitochondrial Function in Irgm1-Deficient MEFs ...... 63

2.3 Discussion ...... 65

2.4 Methods ...... 68

3. Irgm1-Deficiency Causes Metabolic Alterations in MEFs and BMM ...... 74

3.1 Introduction ...... 74

3.2 Results ...... 76

3.2.1 Irgm1-Deficient MEFs Have a Different Metabolic Phenotype After Priming by IFN-γ ...... 76

3.2.2 Irgm1-Deficient MEFs Possess Bioenergetic Differences Compared to WT MEFs ...... 76

3.2.3 Key Metabolite Levels are Altered in Irgm1-Deficient MEFs ...... 78

3.2.4 – Mitochondrial Morphology and Metabolism in Irgm1-Deficient BMM ...... 80

3.2.5 IFN-γ Induces Mitochondrial Fission in Irgm1-Deficient Macrophages...... 82

3.2.6 Altered Bioenergetics in Irgm1-Deficient Macrophages ...... 85

3.2.7 Metabolomic Analysis of Irgm1-Deficient Macrophages Reveals Changes in Key Metabolites ...... 88

3.2.7.1 The Altered Metabolite Concentrations in Irgm1-Deficient Macrophages Primed with IFN-γ Suggests Activation of Pro-inflammatory Pathways ...... 88

3.2.7.2 Irgm1-Deficient Macrophages Stimulated with IFN-γ and LPS Suggests Metabolite Depletion ...... 91

3.2.8 The Absence of Irgm1 Induces Neutral Lipid Accumulation in IFN-γ-primed BMM ...... 91

3.2.9 Inhibiting Fatty Acid Synthesis or ROS Damage Mutes the Fragmented Mitochondrial Phenotype Seen in IFN-γ-Primed Irgm1-Deficient BMM ...... 94

3.3 Discussion ...... 96

3.4 Methods ...... 103

4. Metabolic alterations contribute to enhanced inflammatory cytokine production in Irgm1-deficient macrophages ...... 108

4.1 Introduction ...... 108

4.2 Results ...... 110

4.2.1 Secretion of Cytokines RANTES and MCP-1 are Elevated in Irgm1-Deficient Macrophages When Primed with IFN-γ ...... 110

4.2.2 Inhibition of Autophagic Flux Does Not Increase RANTES Secretion in IFN-γ- Primed Macrophages ...... 112

4.2.3 Effect of Metabolic Changes on Pro-Inflammatory Cytokine Production in Irgm1-Deficient Macrophages ...... 113

4.2.4 Effect of N-Acetyl Cysteine on Pro-Inflammatory Cytokine Production in Irgm1-Deficient Macrophages ...... 117

4.2.5 Secretion of TNFα and IL-1β Are Elevated in Irgm1-Deficient Macrophages Stimulated with IFN-γ and LPS ...... 117

4.2.6 TNFα Secretion is Slightly Muted in the Presence of ROS and Glycolytic Inhibitors...... 119

4.2.7 IL-1β Secretion is Muted in the Presence of ROS and Glycolytic Inhibitors ... 120

4.2.8 Enhanced Pro-Inflammatory Cytokine Production in Irgm1-Deficient Mice and Cells ...... 121

4.3 Discussion ...... 124

4.4 Methods ...... 127

5. Conclusion and Future Directions ...... 131

5.1 Overview ...... 131

5.2 Possible Mechanisms of Actions in the IFN-γ Activation of Irgm1-Deficient Cells ...... 135

Appendix A...... 144

Appendix B ...... 146

Appendix C ...... 148

Reference ...... 150

Biography ...... 192

List of Figures

Figure 1: Accumulation of swollen and elongated in the ileum of Irgm1-deficient mice 53

Figure 2: Irgm1 localizes to the mitochondria in IFN-γ primed WT MEFs ...... 55

Figure 3: Hyperfused mitochondrial morphology observed in Irgm1-deficient MEFs .... 56

Figure 4: Irgm1 structural determinants required for mitochondrial localization ...... 58

Figure 5: Irgm1 structural determinants required for increased mitochondrial fragmentation seen in WT MEFs primed with IFN-γ...... 60

Figure 6: Irgm1-mediated mitochondrial fragmentation is dependent on the activity of Drp1...... 62

Figure 7: Comparison of mitochondrial effectors and functions in IFN-γ primed WT and Irgm1-deficient MEFs...... 64

Figure 8: Irgm1-deficient MEFs show an increase in glycolytic activity ...... 77

Figure 9: Altered metabolite profiles in WT and Irgm1-deficient IFN-γ-primed MEFs ... 79

Figure 10: Increased mitochondrial fragmentation observed in Irgm1-deficient BMM ... 81

Figure 11: Timecourse quantification of IFN-γ-mediated mitochondrial fragmentation in Irgm1-deficient BMM ...... 83

Figure 12: Mitochondrial fragmentation remains unchanged in ATG7-deficient mice .... 85

Figure 13: Irgm1-deficient macrophages show an increase in glycolytic activity ...... 86

Figure 14: Altered metabolite profiles in WT and Irgm1-deficient IFN-γ-primed BMM . 90

Figure 15: Altered metabolite profiles in IFN-γ/LPS stimulated WT and Irgm1-deficient BMM ...... 92

Figure 16: Neutral lipid accumulation is increased in IFN-γ-primed, or IFN-γ/LPS stimulated macrophages lacking Irgm1 ...... 93

xii

Figure 17: The punctate mitochondrial morphology is increased in IFN-γ-primed macrophages lacking Irgm1, and is dependent on fatty acid synthesis and reactive oxygen species ...... 95

Figure 18: IFN-γ-primed macrophages show enhanced secretion of inflammatory cytokines in the absence of Irgm1 ...... 111

Figure 19: Inhibition of autophagy does not affect RANTES production in IFN-γ -primed macrophages ...... 112

Figure 20: Blocking glycolysis or fatty acid synthesis mitigates the increased RANTES and MCP-1 secretion in IFN-γ-primed macrophages lacking Irgm1 ...... 114

Figure 21: N-acetylcysteine mutes RANTES and MCP-1 production in IFN-γ-primed macrophages despite no difference in ROS levels ...... 116

Figure 22: Macrophages stimulated with IFN-γ/LPS show enhanced secretion of inflammatory cytokines in the absence of Irgm1 ...... 118

Figure 23: TNFα Secretion in Irgm1-deficient macrophages in the presence of glycolytic and ROS inhibitors ...... 119

Figure 24: 2-Deoxyglucose and N-acetylcysteine mute IL-1β secretion in Irgm1-deficient macrophages ...... 121

Figure 25: Irgm1-deficiency leads to enhanced cytokine production in mice ...... 123

Figure 26. Potential mechanisms of metabolic dysfunction and increased inflammatory cytokine production in Irgm1-deficient macrophages ...... 135

xiii

1. Introduction

1.1 Interferon Signaling in Innate Immunity

1.1.1 Introduction to Interferon

In 1957 Alick Isaacs and Jean Lindenmann discovered that cells treated with heat-inactivated influenza virus secreted a substance that protected against subsequent viral infection. They named this substance “interferon” for its ability to interfere with viral replication (1-3). This was the first cytokine discovered (4). Over the past sixty years, it was discovered that this substance was not a single protein, but a large family of pro-inflammatory cytokines that induce immune responses. Indeed, interferons have been present in chordates for the past 500 million years, and were part of the early development of the immune system (3,5).

Interferons (IFNs) can be divided into three categories: type I, type II, and type III. Type I IFNs are the most extensive group, including IFN-α, IFN-β, IFN-ε, IFN-κ, and IFN-ω, which are all found in humans, as well as many members found in other species (4,6). For the purposes of this dissertation, only the most commonly expressed type I IFNs, IFN-α and IFN-β, will be discussed. Type III IFNs, comprising of IFNλ1 (IL-

29), IFNλ2 (IL28a) and IFNλ3 (IL28b), are related functionally to type I IFNs, but are only expressed by epithelial cells (6). Type III interferons are not relevant to this dissertation and will not be further discussed. In contrast to the other classes, type II IFN has, to date, only one member: IFN-γ (7,8).

1.1.2 Interferon Signaling

The type I and type II interferons are quite similar in their mechanism of cell signaling. Both are induced by PRRs (pattern recognition receptors), and use members of the JAK/STAT pathway to induce gene expression. The key difference is that type II interferon’s mode of action is limited and discrete, while that of type I interferon is much more variable in scope.

1.1.2.1 Type I Interferon Signaling Pathway

Unlike types II and III, type I Interferon can be expressed by most cell types (9).

Type I interferon expression is classically induced through activation of extracellular and cytosolic PPRs, such as Toll-like receptors. While type I interferons were first identified by, and studied for, their anti-viral activity, more recent studies have shown that they are also involved in resistance against fungal, parasitic, and parasitic pathogens (6,9,10), as well as modulating the immune response of other immune cells.

The PPRs are classically activated by viral DAMPs (danger associated molecular patterns) and PAMPs (pathogen associated molecular patterns); including double stranded RNA and viral envelope proteins (6,11). Once activated, the PPRs induce distinct signaling pathways (described below), ending in the up-regulation of immune defense pathways, along with the secretion of cytokines IFN-α and IFN-β (6).

Every type I interferon binds to the same cell surface receptor, IFNAR (type I IFN receptor), to initiate a cell signaling cascade (4). IFNAR has two subunits, IFNAR1 and

IFNAR2, each of which associates with a member of the JAK (Janus-activated kinase) family: IFNAR1binds to TYK2 (Tyrosine Kinase 2), while IFNAR2 associates with JAK1

(12,13). Upon binding a type 1 IFN ligand, IFNAR1 and IFNAR2 dimerize, causing the autophosphorylation and activation of TYK2 and JAK1 (12-15). The two JAK kinases then induce other downstream signaling cascades.

In the canonical type I interferon downstream signaling pathway, the activated

JAK kinases phosphorylate STAT (signal transducer and activator of transcription) proteins, causing them to dimerize (12,13,15,16). Although the STAT protein family contains six members (STAT1-6), the classical type I IFN cascade involves the phosphorylation and dimerization of STAT1 and STAT2. The STAT1/2 dimer then binds with IRF9 (IFN regulatory factor 9), forming the ISGF3 (interferon-stimulated gene factor 3) complex, the main transcription factor for type I IFN genes (17,18). ISGF3 translocates to the nucleus, where it binds to IRSE (interferon-stimulated response elements) elements present upstream of the promoter of many interferon-inducible genes, inducing their expression (13,19).

Although the STAT1/2 dimer in the cascade described above is the classical example used in describing a type 1IFN signaling cascade, all 6 STAT proteins can form homo or hetero dimers with each other that are used in the immune response 3

(6,16,20,21). The type I interferon response is quite plastic; expression of certain STATs are restricted to specific cell types, and the formation of certain non-classical STAT dimers can activate other signaling pathways, causing the expression of genes outside the canonical IRSE set (16,20-23). Thus, the dimerization of the different STAT proteins expressed in a given cell type allows the body to modulate type 1 interferon response.

1.1.2.2 Type II Interferon Signaling Pathway

First described in the 1970s as an unidentified “macrophage activating factor” in lymph that is critical for antimicrobial function, IFN-γ was finally isolated in 1983

(7,13,24). Although the classical type II interferon signaling pathway is very similar to type I interferon, it shares little to no structural homology with them (8,13,25). As mentioned previously, there is only one type II interferon, interferon-gamma (IFN-γ), discovered to date (7,8). In stark contrast to type I interferons, which can be expressed by most cells in the body (9), IFN-γ expression is limited to antigen presenting cells

(APCs) (7,13).

APCs, such as natural killer (NK) cells, B cells, dendritic cells and macrophages, produce IFN-γ in response to stimulation by other APC-secreted cytokines, including

IL-12 and IL-18 (7,8,13). APC activation is thought to be important early in the innate immune response, activating nearby cells and recruiting others, most notably NK cells and T cells, to the site of infection (7,8). During the rest of the innate and adaptive

immune response, IFN-γ is produced primarily by CD4+ helper T cells, in response to antigen presentation and cytokine activation, and helps promote a Th1 cell response

(13,26).

The first step in the type II signaling cascade is the binding of IFN-γ to its cell surface receptor: IFNGR. Similar to the type I interferon receptor, IFNGR is also composed of two subunits, IFNGR1 and IFNGR2, which each associate with Jak1 and

Jak2, respectively (8,13). Upon binding to IFN-γ, the IFNGR subunits rearrange, inducing the activation and autophosphorylation of Jak2 (14,27-29), and the transphosphorylation of Jak1 by Jak2 (19,27). Activated Jak1 then phosphorylates a specific tyrosine residue on IFNGR1 (Tyr440) to induce the formation of two STAT1 binding sites. Once bound, the two STAT1 proteins are phosphorylated on Tyr701, dimerize, and are released into the cytoplasm (16,19,30). The STAT1 homodimer, also known as the’ gamma-IFN activated factor’ (GAF), then translocates to the nucleus, where it binds to specific promoter elements, called GAS (gamma activated sequences), and initiates or suppresses transcription (13,30-32).

1.1.3 Interferons in Immunity

The main purpose of interferons is to prime cells for immune defense. Upon induction, interferons up-regulate inflammatory pathways, like the production of cytokines, anti-virals and antimicrobials (33-36), pathways promoting anti-pathogen

defense, such as autophagic flux, cation sequestration, lysosomal degradation and nutrient depletion (10,37,38), and pathways that will alert the system to the infection, such as promoting T cell proliferation and antigen presentation pathways (37,38).

Similarly, interferons down regulate anabolic pathways, to conserve resources

(6,34,35,39). This result, at least in the case of type II interferon (IFN-γ), is achieved by a massive shift in the transcriptome. Interferons promote the transcription of hundreds of genes, modulate the transcriptional efficiency of genes involved in immune, metabolic, and synthetic pathways, and even induce large-scale changes in chromatin structure (39-

41). Thus, interferon activated cells, and the systems that they reside in, are able to mount a stronger, immediate response to an immune challenge.

1.2 Immunity related GTPases

1.2.1 Features of the IFN-Inducible GTPases

The ability to defend against microbial infection at the cell-to-cell level – a process termed cell-autonomous immunity – is essential for a functional innate immune system. Interferons induce the expression of cell-autonomous effectors, including four distinct classes of IFN-inducible GTPases (10,42). Comprised of the very large inducible

GTPases (VLIGs), the myxovirus resistance proteins (Mx), the guanylate-binding proteins (GBPs), and the immunity-related GTPases (IRGs), these proteins belong to the dynamin superfamily of GTPases, and are characterized by their ability to bind and alter

lipid membranes (43-46). Typical functions of IFN-inducible GTPases include vesicle formation, vesicle trafficking, and membrane remodeling, often in the context of membrane trafficking events (42,43,45-51).

The dynamin superfamily of GTPases, including the IFN-inducible GTPases, possesses unique structural characteristics and functional properties. Their distinct structural elements include a conserved G domain, responsible for binding GTP/GDP, and helical domains, known for mediating membrane association (42,52,53). Other distinguishing features include the ability to self-oligomerize into dimers and/or oligomers (48,54,55). Unlike canonical GTPases, such as the small Ras-like GTPase family, the main function of GTP hydrolysis and binding in dynamin-like GTPases is not signal transduction (56,57). Thus their biochemical activity cannot be explain using the simple ‘biochemical switch’ paradigm, where GDP bound proteins are thought to switch from an ‘inactive state’ to a ‘biologically active state’ upon binding GTP (57). Instead, dynamin-like GTPases are thought to utilize the energy of GTP hydrolysis to undergo conformational changes and oligomerization (58), a function critical to many of their membrane altering activities (48,50,53,59).

Of the IFN-inducible GTPases, the Mx, GBPs and IRGs are the best characterized.

The Mx GTPases are best characterized for their broad antiviral activity, and are only induced by type I and III interferons (60,61). Meanwhile, the IRGs and GBPs are best known for their antimicrobial activity against intracellular bacterial and protozoan

pathogens (62). Although they can be induced by either type I or type II interferon, they are primarily known as IFN-γ (type II) effectors. Indeed, GBPs and IRGs have been shown to be the most upregulated cellular transcripts following IFN-γ treatment in mouse embryonic fibroblasts (MEFs) and macrophages (51), underscoring their importance in innate immune defense. This dissertation will focus on studying IRGs, particularly Irgm1, and their functions in cells after IFN-γ stimulation.

1.2.2 The IRG Protein Family

1.2.2.1 The Proteomic and Structural Features of the IRG Family

The IRGs are a family of IFN-inducible GTPases that characteristically have a molecular mass between 47-48 kDa. IRGs were thought to originate from the common ancestor of chordates around 540 million years ago (63), and experienced dynamic evolution throughout vertebrate species (64,65). Indeed, mice and zebrafish have evolved complex antimicrobial mechanisms, mediated by up to 23 and 11 IRGs, respectively; while humans possess only 2 IRG genes, and birds seem to have lost the entire IRG lineage (62,64,65). The work presented in this dissertation is performed in a mouse model, and the implication of the results discussed in both mouse and human

IRG systems.

All vertebrates have a similar IFN system (4,66), which is thought to provide a conserved regulatory network that the IRG genes can operate in (64). The majority of

IRGs have multiple interferon-stimulated response elements (ISREs) and γ-activated sequence (GAS) elements in their promoter regions, and are therefore transcriptionally induced through interferon signaling pathways as a rule (51,65). However, the IRGs are known for being induced most by IFN-γ. Notable exceptions to IFN-γ induced expression include the human IRGM gene (67), and the members of the IRG “C” subfamily (65), which are constitutively expressed.

Only one IRG crystal structure has been solved to date, that of Irga6 (52).

However, it is thought to be a good structural model for most IRGs. Irga6 was isolated as a dimer, and consists of three domains: a short three-helix N-terminal domain, an eight helix C-terminal domain, and a “G” or GTPase binding domain. The G domain is in the center region of the protein, and contains three classical GTP binding motifs: the phosphate binding p-loop (G1), the DxxG region (G3), and the (N/T)(K/Q)xD motif (G4)

(52,68). The G domain is of particular importance; it is the most conserved domain across species and, as will be discussed further in a subsequent section, IRG function varies based on the peptide sequence in the p-loop motif (52).

1.2.2.2 IRG Subfamilies

Proteins of the IRG family are classified into nine subfamilies (A, B, C, D, E, F, G,

M and Q). This classification is based primarily on the sequence homologies of their G domain, as it is the most conserved domain across species and thought to be necessary

for their overall function (65). IRG nomenclature is based on the protein’s subfamily and the numerical position it was given within that subfamily. For example, Irgb6 is the 6th member of the ‘B’ subfamily, while Irgm1 is the first member of the “M” subfamily.

Five of these subfamilies (C, E, F, G and Q) are not relevant for this dissertation, as my work focuses only on IFN-γ-inducible IRGs in mammalian systems. IRG subfamilies E-G are only found in fish and the Q subfamily have no GTP binding motifs in their G domains. Meanwhile, the IRG C subfamily members are not induced by INF, have only been found to be expressed in the testes, and demonstrate no known functional relationship to the other IRG subfamilies (64,65). The remaining IRG subfamilies, A, B, D, and especially M, are present in either the mouse or human IRG models that will be described in this dissertation.

The IRG subfamilies with functional GTP binding domains can be further subdivided functionally into the GKS and GMS IRGs, based on their “p-loop” sequence.

The p-loop sequence, or G1 domain, is a classical peptide sequence present in the conventional dynamin GTP-binding domain (52). The majority of IRGs, including subfamilies A-G, contain the canonical GXXXXGKS p-loop sequence, and are classified as the “GKS” IRGs. The M IRG subfamily, alone, has the non-canonical GXXXXGMS p- loop sequence, where a lysine residue has been replaced with a methionine, and are classified as “GMS” IRGs (51,52,65). This singular p-loop sequence is thought to influence their nucleotide binding abilities, as compared to the GKS IRGs. Biochemical

studies of the murine GKS IRG Irga6 showed that it displayed a pronounced preference for binding GDP over GTP (69). In contrast, over 90% of the GMS IRG Irgm3 was found to be GTP-bound in vivo (70). This difference in nucleotide binding affinities implies that GMS IRGs are largely active within the cell once expression is induced, whereas

GKS IRGs, once expressed, remain in the inactive GDP-bound state unless activated by specific stimuli (71). The functional consequences of this shift in binding affinities will be discussed further in section 1.2.2.3.1 and 1.2.2.3.2.

1.2.3 The Murine IRGs

In C57B1/6 mice there are 16 IRG genes and 4 annotated IRG pseudogenes clustered on chromosomes 7, 11 and 18 (65,72). Three genes, Irgm1, Irgm2 and Irgm3, are GMS IRGs, while the rest, including Irga6, Irgb6, Irgb10 and Irgd, are GKS IRGs. The

GMS IRGs are commonly thought of as “regulatory” IRGs that control the localization of the “effector” GKS IRGs. As indicated above, the murine IRGs are some of the most upregulated cellular transcripts following IFN-γ treatment in mouse embryonic fibroblasts (MEFs) and macrophages (51) – a fact that accentuates that IRGs comprise a crucial innate immune system for defense against intracellular pathogens.

1.2.3.1 Murine IRGs Interactions in Lipid and Membrane Binding

The ability of IRGs to localize and bind intracellular lipid membranes constitutes a major component of their ability to defend against membrane bound intracellular pathogens. Highlighting the importance of membrane binding, some murine IRGs genes contain motifs for post-translational lipid modification. An N-terminal myristoylation motif mediates the binding of some of the murine GKS IRGs, including Irgb10 and Irga6 to the pathogen vacuole (73,74). As well, the GMS IRG Irgm1 has an amphipathic helix and several palmitoylation sites in its C-terminus responsible for its Golgi and mitochondrial localizations (75,76). When expressed in stimulated cells, the GKS IRGs localize predominantly to the cytosol, while the GMS IRGs associate with intracellular organelle membranes, including the ER, Golgi, mitochondria, peroxisomes and lipid droplets (70,73,77). This supports the theory expressed in the last section: that the GMS

IRGs, with high GTP-binding affinities are membrane bound and thus ‘constitutively active,’ while the cytosolic GKS IRGs remain in an ‘inactive state’ in the cytosol until activated.

Upon pathogen entry, GKS IRGs canonically localize to the pathogen-containing vacuole. In the case of T. gondii and C. trachomatis, Irga6, Irgb6, and Irgd migrate to the pathogen’s phagosome soon after infection, and, in doing so, are thought to embody anti-microbial activity. In the case of T.gondii, the localization of the GKS IRGs was correlated to the stripping of its phagocytic membrane, resulting in its death (78), while

in the case of C. trachomatis the GKS IRGs target the inclusion for destruction via the macroautophagy pathway(79). Regarding the GMS IRGs, there is some controversy on whether they also localize to a pathogen’s phagosome during infection. Early reports of the GMS IRG localizing to vacuolar membranes early during infection (76,80) were contested when the anti-Irgm1 antibody used in those studies was shown to also bind bacteria. Subsequent experimentation, with new anti-Irgm1 antibodies, found no association to the same bacterial phagosomes (81). While it has been suggested that phagosomal localization of GMS IRG to bacterial phagosomes could promote their swift fusion with lysosomes (76,80), it should be noted that a phagosomal localization of the

GMS IRGs has not been observed on many live microbial pathogens targeted by GKS

IRG systems (74,81,82). Furthermore, even in the instances that GMS IRGs were observed on phagocytic membranes (83-85), the amount of GKS IRGs that localize to the phagocytic membrane far outstrips that of the GMS IRGs, suggesting they have the primary function at these sites (74,86).

1.2.3.2 Murine IRGs in Intracellular Pathogen Defense

As their induction by cytokines and localization to membranes suggested that the murine IRG family might be involved in immune defense against membrane-bound intracellular pathogens, the effects of IRGs on pathogen susceptibility were extensively researched. Studies of mice with targeted deletions of Irgm1 and Irgm3, first published

in the early 2000s from our lab, established that pathogen resistance was a key function of IRGs (87,88). Since then, mice deficient in Irgm1, Irgm3, Irgd and Irga6 have been tested for susceptibility to many protozoan, bacterial and viral pathogens, in particular those that require IFN-γ signaling for pathogen restriction (Table 1) (89). While displaying resistance to most viral infections, IRG-deficient mice display a strikingly wide range of susceptibility to intracellular bacterial and protozoan pathogens (62,89).

For example, Irga6-deficient mice showed increased susceptibility to Toxoplasma gondii, but normal resistance to most pathogens, including Leishmania major (90) and Chlamydia trachomatis (79). In contrast Irgm1-deficent mice are susceptible to many intracellular pathogens, including Toxoplasma gondii (82,88), Leishmania major (90), Salmonella typhimurium (91), Listeria monocytogenes (88), Chlamydia trachomatis (92), and

Mycobacterium species (80,93). While the specific roles of individual IRGs in innate immunity remain unknown, the difference in patterns of resistance suggests that they have overlapping, but non-redundant, roles in defense against pathogens.

1.2.3.3 Mechanisms of IRG-Mediated Pathogen Defense

As indicated previously, the murine IRG system can be thought of as two functional subcategories in pathogen resistance: the “regulatory” GMS proteins, such as

Irgm1and the GKS “effector” proteins (59,65). When expression is induced by IFN, the

GKS IRGs reside mainly in the cytosol in their “inactive” GDP-bound form until the cell

is infected with an IRG-susceptible pathogen. In contrast, the murine GMS IRGs localize to intracellular lipid membranes, including the Golgi (75), the ER (90,94), mitochondria

(75,81), peroxisomes (76,95) and lipid droplets(74,96), independently of infection by intracellular pathogen. In doing so, the GMS IRGs block the cells’ “self” membranes from GKS localization through GDP dependent interactions, thus protecting these lipid- bound organelles from damage if/when the GKS IRGs are activated (74,96). In the context of a canonical GTPase paradigm, like the Ras family of GTPases, the GMS IRGs execute the same function as guanine dissociation inhibitors, blocking the premature activation and off-target localization of GKS IRGs. In the absence of GMS IRGs, the GKS

IRGs undergo GTP-induced oligomerization and form large aggregates in the cytosol, as well as mislocalization to lipid droplets (59,71,74,97).

Once the cells are infected with an IRG-susceptible pathogen, the GMS IRGs are chiefly absent from the pathogen vacuole, permitting the GKS IRGs to become activated and load onto the phagosomal membrane through an as-of-yet undetermined mechanism (74). The GKS IRGs load onto the phagosomal membrane in a coordinated and sequential manner that is dependent on their oligomerization and their GTP-bound state (59,69,74). The initial proteins that load onto the vacuole are thought to stabilize the recruitment of subsequent GKS IRGs (86). Although the mechanism governing the activation and translocation of the GKS IRG to the pathogen vacuole is currently unknown, it is thought that this process involves autophagic machinery. In support of

this theory, components of the autophagy system responsible for the lipidation of LC3, and other such proteins, have been shown to localize to GKS IRG-targeted pathogen vacuole (95,98,99), and enhance the translocation of the GKS IRGs to the pathogen vacuole of C. trachomatis and T. gondii (62,79,99,100).

In the case of T. gondii and C. trachomatis, the accumulation of GKS IRGs on the pathogen vacuole results in its lysis (83,95,101,102). The unenveloped microbe is then exposed to the cytosol of its host cell, leading to its death (101,102). The underlying mechanisms for the lysis of the pathogen vacuole and parasite death are still unknown, although in the case of T. gondii there are reports that the bare tachyzoites are enveloped by autophagosome-like vacuoles that then fuse with lyzosomes after lysis (102). Other studies suggest that the GKS proteins work in conjunction with other proteins at the pathogen vacuole, such as the GBPs (103) and ubiquitin E3 ligases (95), to induce lysis.

Still others theorize that one of the purposes of the GTP hydrolysis, required for the localization of the “effector” GKS IRGs, is to induce a higher order self-oligomerization and drive a dynamin-like mechanical constriction around the pathogen vacuole to induce rupture (42,101,104).

1.2.4 Functions of Murine Irgm1

This dissertation focuses on describing a novel function for the GMS IRG Irgm1.

The mouse Irgm1 gene is located on chromosome 11 and contains two introns and three

exons. Two mRNA Irgm1 transcripts are generated through alternative splicing, producing Irgm1 protein isoforms of 409 and 395 amino acids (81,105). Irgm1’s lipid binding is mediated by an amphipathic αK helix and palmitoylation domains on its C- terminus helical domain. Further studies demonstrated that Irgm1 binding affinities are quite specific, and include phosphoinositide-3,4-bisphosphate (PtdIns[3,4]P2), phosphoinositide-3,4,5-triphosphate( PtdIns(3,4,5)P3)and the mitochondrial lipid cardiolipin (75,76). In IFN-γ primed cells, Irgm1 localizes most strongly to the Golgi

(73,76,82), but a significant amount also localizes to peroxisomes (74,76), lipid droplets

(74), endolysosomes (82,106) and mitochondria (75,76,107). While both the long and short isoforms of Irgm1 accumulate on the Golgi and mitochondria, only the long isoform localizes to endolysosomes (81). As indicated above, loss of individual IRGs in mice produces a wide array of susceptibility phenotypes to IFN-associated pathogens, both in range and severity, although it generally results in a specific loss of function against a subset of pathogens (89,96). The loss of Irgm1 is the exception to this trend;

Irgm1-deficient mice present with a drastic phenotype, not only in susceptibility to pathogens, but in loss of general immune function.

As discussed in the previous section, the function the GMS IRGs, including

Irgm1, are best known for is acting as guanine dissociation inhibitors and regulating the activation and localization of the GKS IRGs to the pathogen vacuole (74,108,109).

However, loss of Irgm1 not only results in susceptibility to pathogens targeted by the

GKS IRGs, such as T. gondii (83,87), C. trachomatis (79,110) and E. cuniculi (111), but to a range of other intracellular microbial pathogens, including L. monocytogenes (88,112), M. tuberculosis (76,80), M. avium (93), T. cruzi (112) and L. major (90,113), whose phagosomes do not accumulate GKS IRGs during infection. Even in the absence of infection, Irgm1 deficient-mice reportedly have an augmented susceptibility to LPS injection alone (114). Furthermore, Irgm1-deficient mice are unusually susceptible to several autoimmune diseases, including animal models of colitis (115,116) and immune encephalitis (117,118), as well as the mouse model of stroke (119).

During microbial infection, it is very unlikely that Irgm1 acts as a direct effector at the phagosome / pathogen vacuole (see Section1.2.2.3.1), indicating its method of action must be indirect. Several groups have suggested a role for Irgm1 in autophagy and lysosomal degradation. Earlier studies described a defect in the acidification of lysosomes containing mycobacterial phagosomes (80,88). More recent works have presented evidence that Irgm1 plays a role in autophagy (108,120), and suggested that autophagic flux is impaired in its absence (97,121,122). Last year it was demonstrated that this was indeed the case – in the absence of Irgm1, the GKS IRGs mislocalize to lysosomes, causing a defect in their acidification. Furthermore, the GKS-loaded lysosomes had an impaired ability to process autophagosomes and other substrates, leading to a defect in catabolism and autophagic flux (96). Clearly, Irgm1 activity is

essential to several cellular functions and likely participates in many pathways in the presence or absence of infection.

Since its discovery, research surrounding Irgm1 focused mainly on its role in defense against intracellular bacteria and protozoa and thus it was largely thought of as a pathogen resistance protein. However, this interpretation is at best incomplete and fails to include key observations of the Irgm1’s role in the immune system as a whole.

Infection of Irgm1-deficient mice with pathogens that induce IFN-γ production, for example S. typhimurium or T. cruzi (91,112), results not only in reduced bactericidal activity at a cellular level, but a systemic lymphomyeloid collapse. Subsequent studies established that Irgm1 expression is essential to the health and expansion of many lymphomyeloid cell populations in response to IFN-γ, even in the absence of a bacterial challenge. Indeed, Irgm1-deficient mice suffer from a severe reduction in the baseline proliferative and self-renewal capacity of their hematopoietic stem cells (123,124), and induced massive IFN-γ-dependent autophagic cell death of effector CD4+ T lymphocytes (121), independent of infection. Further, the aberrant proliferative and autophagic phenotypes observed in Irgm1-deficient mice were abolished in either

Irgm1/IFNGR1 or Irgm1/STAT1 double-knockout mice (122), suggesting that dysregulated IFN signaling was the source of this defect.

Although IFN signaling represents a powerful activator for immune cells in the short term, this activation must be held in check, to avoid the harmful effects of

prolonged and uncontested pro-inflammatory effects of IFN stimulation (122,125). The studies presented above suggest that Irgm1 is an important systemic negative regulator of IFN signaling in lymphatic cell types. This role of Irgm1 correlates with the fact that, as noted above, lack of Irgm1 causes a greater susceptibility to autoimmune conditions in several mouse models (65,67,105,116-119). While it is true that proper autophagic regulation is extremely important to the health of lymphomyeloid cells (126-128), it remains to be seen whether the block in lysosomal degradation is the only or main trigger for the lymphopenia observed in Irgm1-deficient mice. Whatever the answer, research into Irgm1’s role as an immunomodulatory protein is far from complete.

1.2.5 The Human IRGs

The immunity-related GTPase system in humans stands in stark contrast to the one in mice. The number of IRG genes in the human genome, unlike the 20 or more genes present in the murine genome, is reduced to two: IRGC and IRGM (67). Of these two genes, only IRGC is not truncated compared to its murine counterpart, and is the sole GKS IRG present in the human genome. While human IRGC is 89% identical at the protein level to mouse Irgc, neither homologue is induced through IFN stimulation, both are expressed solely in the testes (105). Although their function remains unknown, it is presumed that the roles of human IRGC and mouse Irgc do not play a large role in immunity (62,65).

Conversely, the lone human GMS IRG, IRGM, is quite different from its mouse counterparts in form, regulation and function. Despite being having an orthologous G domain to the mouse GMS IRGs (Irgm1, Irgm2 and Irgm3), human IRGM is truncated at its N and C terminus. Human IRGM showed additional transcriptional differences to its mouse analogues. Five different splicing isoforms of IRGM mRNA (IRGM a-e) were cloned and expressed, with predicted molecular weights between 19-24kDa (65,105).

Moreover, further studies failed to demonstrate the induction of IRGM by IFNs (65).

As a result of all these differences, human IRGM was, upon discovery, thought to be a pseudogene. The evolutionary history of IRGM demonstrates that this assessment was not entirely incorrect. Briefly, analyses indicate that IRGM was pseudogenized around 50 mya due to an Alu transposon insertion, and remains so in the Old World and New World monkeys. Surprisingly, IRGM’s open reading frame is thought to have been restored by the insertion of an endogenous retrovirus element

(ERV9), which now functionally serves as IRGM’s promoter in the great apes, including humans (65,67,105).

Interest in IRGM was renewed when it was found to be a susceptibility locus for

Crohn’s disease, along with the autophagy gene ATG16L1, as well as for infection by M

.tuberculosis (129-131). These studies challenged IRGM’s status as a pseudogene, and were preceded by a functional study suggesting that IRGM expression played a role in regulating the autophagic processing of bacteria-containing phagosomes in M. bovis

BCG infected human macrophages (108,132). The assertion that human IRGM was involved with the autophagic immune response was bolstered by studies suggesting that IRGM expression was necessary for defense against two other intracellular pathogens: pathogenic adherent-invasive E. coli (133) and S. typhimurium (131).

Moreover, although IRGM is normally expressed at low levels, it is expressed broadly in cells types examined so far and its expression has been shown to be augmented when cells are primed with lipopolysaccharide (67,120,134). Recently, IRGM has been shown to interact with the core autophagy components ULK1 and Belclin 1, as well as complexing with Crohn’s disease risk factors NOD2 and ATG16L1, thereby regulating the assembly of the autophagic machinery (134). Furthermore, the interaction of IRGM with the autophagic core machinery is augmented by ubiquitination of the protein

(134,135). These results indicate that IRGM is a direct regulator of autophagy, although the direct implications IRGM deficiency in the human immune system still needs to be explored.

Modulating mitochondrial dynamics and function is postulated as a second function of human IRGM. One study showed that a portion of human IRGM isoform

“IRGMd” localized to the mitochondrial inner membrane, and binds to the mitochondrial lipid cardiolipin (120). Overexpression of the IRGMd in cell culture induced a punctate mitochondrial morphological phenotype, a decrease in mitochondrial membrane potential, and an increase in apoptotic cell death. Conversely,

when IRGM was knocked down, the cell’s mitochondria took on a more tubular, interconnected phenotype, resulting in a decrease in autophagosome formation and reactive oxygen species (ROS) production in the cell (120). As autophagy is believed to be induced by ROS (136), mitochondria are thought to be a primary source of ROS for induction of autophagy in the cell (137), these results suggest that IRGM’s mitochondrial function is tied to its autophagic function. Further study is required to define IRGM’s role in the human system.

1.2.5.1 Parity of Function between Mouse Irgm1 and Human IRGM

There is some evidence that the two clear functions of human IRGM, promoting autophagy and modulating mitochondrial function, may be shared with mouse Irgm1.

Initially, the N- and C- terminal truncations of IRGM were thought to be too severe for any parity of function to exist between it and the mouse GMS IRGs. That is, until a paper, cited in the previous section (108), not only reported that human IRGM induced the autophagic clearance of M. tuberculosis BCG, but asserted that this function was shared by mouse Irgm1. Indeed, previous work had indicated that overexpression of

Irgm1 led to increased autophagosome formation in an IFN-γ-stimulated immortalized mouse macrophage cell line, while knocking it down with siRNA resulted in a decrease in the number of autophagosomes per cell (108,138). However, a subsequent paper showed that the autophagic induction observed in Irgm1-deficient macrophages was a result of IFN-γ priming alone, and could be generated independently of Irgm1

expression (139). Still, the presence of Irgm1 on endolysosomes has been shown to be required for completion of autophagic degradation in IFN-γ macrophages (96), and the presence of Irgm1 and the LC3 lipidation machinery/ Atg8 proteins at lipid droplets is hypothesized to be essential in the activation of the murine GKS IRGs, both of which suggest Irgm1 may be intertwined with autophagic mechanisms. As of yet, further studies are required to determine what role Irgm1 has relative to autophagy.

The Deretic lab’s seminal paper describing IRGM’s mitochondrial function (120) was followed by papers from the Taylor lab (75,116,140) (that also, coincidently, make up the majority of this dissertation and will be discussed further in chapters 2, 3 and 4), and others (76,81), demonstrating that a subset of mouse Irgm1 also localized to the mitochondria and caused changes in mitochondrial dynamics. The modulation of mitochondrial function by IRGM and Irgm1 strongly suggests that these two proteins share some range of function. However, as the mechanism through which IRGM or

Irgm1 impact mitochondrial function remains unknown, more research is required to make a determination.

While it is tempting to assign parity of function between Irgm1 and IRGM, much more study is needed before a consensus can be reached. The three GMS murine IRGs,

Irgm1, Irgm2 and Irgm3 are all functionally distinct orthologues of human IRGM (65).

To date, little is known about the functions and mechanisms of actions of Irgm2 and

Irgm3. This gap in knowledge needs to be filled before definitive functional relationships among human and mouse IRGM proteins can be assigned.

In summation, the immunity related GTPases are a class of IFN-inducible

GTPases that is conserved, in some form, by a large portion of vertebrate species. This protein family is best known for their interferon induced, cell autonomous role in innate immunity to intracellular bacteria and protozoa. Functionally, the IRGs can be divided into the effector GKS IRGs and the regulatory GMS IRGs, so-called for the placement of a lysine or methionine in the p-loop sequence of their G domain. The functions and mechanisms of action of the IRG family varies between species, and encompasses modulation of pathogen-associated membranes, regulation of membrane trafficking and autophagy. Further study is required to determine the many mechanisms through which this family modulates innate immunity.

1.3 Immunometabolism of Proinflammatory Macrophages

This past decade has seen a surge in the investigation into the metabolism of immune cells, and popularized the field of study called immunometabolism. This advent came about as new metabolomic techniques and immunological models have allowed immunologists to measure flux through these metabolic pathways during infection or immunological challenge (141). This section will focus on the modulation of metabolic pathways seen in pro-inflammatory macrophages, and how they contribute to effector functions of the pro-inflammatory response.

1.3.1 Modulation of Immunometabolism in Proinflammatory Macrophages.

Oxygen levels and nutrient availability have historically been seen as the primary drivers of metabolic pathways. One model has suggested that cells encountering normal oxygen levels and the absence of stress favor high ATP production through the oxidative phosphorylation of sugars and beta-oxidation of lipids.

Conversely, when cells encounter hypoxia or undergo oncogenic transformation, the hypoxia-inducible factor 1α (HIF1α) becomes activated, leading the cells to use glycolysis as an energy source (142). However, recent discoveries have indicated that the activation of immune cells is also a critical axis of metabolic reprogramming. As the latter part of this dissertation investigates metabolic changes found in Irgm1-deficient cells, this section will focus on the modulation of metabolic pathways, and the phenotypes that result from these changes.

Macrophages are vital part of the response to immune challenges, such as pathogen infection or other stimuli, including debris from apoptotic or necrotic cells and tissue remodeling. Macrophage activation can be divided into two classes: M1 activation

(also known as ‘classical’ or proinflammatory macrophage activation) and M2 activation

(also known as ‘alternative’ or anti-inflammatory activation). Each activation state is incited by distinct molecular stimuli, usually cytokines, chemokines or pathogen PPRs, which cause specific metabolic and functional changes. M1 stimulation, including IFN-γ,

LPS, and IFN-γ/LPS, is associated with Th-1 related pathologies with each stimulus,

alone or in combination, resulting in a range of responses (143). The function of M2 stimulated macrophages is much broader, with stimuli including IL-4, IL-10, glucocorticoids, immune complexes and glucocorticoids/TGF-β. M2 activation is a response to diverse pathologies including parasitic infections (144), auto-immune reactions such as asthma (145), rheumatoid arthritis (146) and scleroderma (147) as well as in states, such as cancer (148,149) or late stage sepsis (150), where glucocorticoids are abundant (151,152). In an in vivo setting, activated macrophages don’t always fit discretely into the ‘M1’ or ‘M2’ state (153). In particular, different M2 stimuli can result in divergent ‘sub-phenotypes’ (151,154,155). However, as M1 macrophages, stimulated with IFN-γ and/or LPS, are much more distinct and set, the model of M1/M2 macrophage activation is valid in the case of this dissertation.

It is important to note that, for the purposes of this dissertation, full “M1” macrophage activation cannot be induced by IFN-γ alone - it requires LPS or IFN-γ/LPS stimulation. When in contact with IFN-γ, macrophages become “primed” for an immune challenge. While IFN-γ priming does result in the expression of many IFN-γ- inducible proteins through changes in transcriptional efficiency and chromatin structure, a second stimulation with LPS, or another bacterial agonist, is required for full macrophage activation with the associated metabolic changes and increases in effector functions. The current state of the field suggests that macrophages stimulated with either LPS or IFN-γ/LPS largely result in same mechanism of activation, although IFN-

γ/LPS stimulated macrophages give a stronger pro-inflammatory response (25,156-158).

Over the next sections, I will discuss the metabolic pathways that have important roles in macrophage function during activation.

1.3.2 Glycolytic Metabolism drives M1 Mitochondrial Polarization

Unlike most pathways discussed in this section, it has long been known that a high glycolytic rate is required for proinflammatory macrophage activation. Past studies established that, once activated, macrophages increase glucose consumption, while 2- deoxyglucose (2DG), a glycolytic inhibitor, inhibits macrophage activation and suppresses inflammatory macrophage function.

The glycolytic increase in activated macrophages may seem counterintuitive, as these cells require energy for their proinflammatory activities, and oxidative phosphorylation can generate 16X more ATP from a single molecule of glucose (159).

However, glycolysis has three main advantages for cell activation. First, it has a rapid response time, occurring within 30min, while oxidative phosphorylation requires boosting mitochondrial biogenesis, a slow, time-consuming process (142). Second, glycolysis allows for the reduction of NAD+ to NADH, which is used as a cofactor by numerous enzymes, and support’s the cell’s biosynthetic growth pathways (142). Third, glycolysis produces large amounts of intermediates for biosynthetic growth. These include intermediates for the anabolic production of nucleotides (by feeding glucose-6- phosphate into the pentose phosphate pathway) (160,161), amino acids (by feeding 3-

phosphogycerate into the serine biosynthetic pathway) (162) and fatty acids (by feeding pyruvate into the TCA cycle for citrate production) (163).

As mentioned above, in the context of the innate immune response to bacterial infection, macrophage activation must occur quickly. Within 30 minutes, the glucose uptake of LPS stimulated macrophages almost doubles, while their extracellular acidification rate (ECAR), an indirect measure of aerobic glycolysis, also increases

(164,165). From here, LPS or LPS/IFN-γ stimulated macrophages further entrench themselves into glycolytically programmed metabolism through distinct temporal steps

(166).

During early activation (1-2 hrs) the mRNA of the glucose transporter (GLUT1) is induced, and glucose intake further increases (165) in M1 activated macrophages.

Macrophage M2 activation pathways are also actively inhibited at this time by downregulating the sedoheptulose kinase Shpk (also known as CARKL), priming the pentose phosphate pathway to its M1 macrophage state (164,167). Mechanistically, glycolysis is further increased in macrophage and dendritic cells by the localization of hexokinase II to the mitochondria (161,168), where it funnels matrix-derived ATP into the formation of glucose-6-phosphate (160), a glycolytic and pentose phosphate pathway substrate. In addition, evidence suggests that mitochondrial-associated hexokinase II also interacts with NLRP3 at outer mitochondrial membrane in this time period, enabling NLRP3 inflammasome activation and IL-1β secretion (169).

Between 1-4 hours after LPS macrophage activation, changes occur to glycolytic enzymes at the transcriptional level to solidify the heightened glycolytic rate. Expression and phosphorylation of the protein pyruvate kinase M2 (PKM2) occurs during this period, and will be further augmented in the late phase of macrophage activation (170).

Phosphorylation of PKM2 induces dimerization and translocation into the nucleus, where it acts, in concert with hypoxia-inducible factor 1-alpha (HIF1α), as a transcriptional inducer of interleukin 1-beta (IL-1β) (170,171). More importantly, PKM2 also acts as a transcriptional inducer of glycolytic genes like 6-phosphofructo-2-kinase

(PFK), constituting an amplification loop important in the intermediate and late phase of macrophage activation (170,171). Within 4 hours of macrophage activation by LPS (or

LPS/ IFN-γ), transcription switches from the liver isoform of 6-phosphofructo-2-kinase

(PFKFB1/PFK2) to the most active PFK isoform, PFKFB3 (156). PFKFB3 produces augmented levels of fructose 2, 6-bisphosphate, which functions as an allosteric activator of 6-phosphofructo-1-kinase (PFK1), stabilizing and sustaining the high glycolytic rates

(172-174).

During the amplification phase, 4-20 hours after macrophage activation, this phenotypic shift towards a pro-glycolytic metabolism is further strengthened. Between

4-6 hours after activation, the pro-glycolytic macrophages also increase expression of monocarboxylate transporter 4 (MCT4), suggesting that the high glycolytic rate necessitates the export of its byproduct, lactate, for proper cellular function. It is

hypothesized that intracellular lactate accumulation may inhibit maximal PFK1 activity required between 6-12 h post LPS-induction (166,175). Finally, during this phase the tricarboxylic acid (TCA) cycle is fractured in two places, reconfiguring it to be an essential source of biosynthetic intermediates (140,142,166), a process explored in greater detail below.

Approximately 24 h after LPS-stimulation, metabolic reprograming is firmly entrenched: glycolytic gene expression and metabolites are increased, as well as lactate and ECAR (176). During the same time, the TCA-cycle becomes a mostly anabolic pathway, instead of a source of catabolic energy, leaving glycolysis to act as the cell’s main ATP source (166,171,177). This switch to a glycolytic-dependent metabolism enables macrophages to generate sufficient ATP, and, in concert with the TCA cycle and pentose phosphate pathway, to produce the biosynthetic intermediates needed to carry out their pro-inflammatory effector functions.

In contrast to the glycolic-driven M1 macrophage activation, alternative macrophage activation relies on oxidative metabolism, induced through signaling pathways involving STAT6, PPARγ and PGC1β. Accordingly, it results in the upregulation of mitochondrial biogenesis and increased beta /oxidative phosphorylation. Instead of glucose import, M2 macrophage activation is driven primarily by the lysosomal lipolysis of endocytosed lipoprotein particles. The metabolic

and mechanistic changes in M2 activated macrophages are just as transformative as M1 activation, and are detailed in many reviews (178-181).

1.3.3 The Pentose Phosphate Pathway Acts as a Hub to Sustain Carbohydrate Metabolism and the Redox Demands of M1 Macrophages

The pentose phosphate pathway (PPP) has many functions, although it is most well-known for facilitating carbon-flux between different metabolic pathways. In general, the PPP funnels carbon into pathways that will supply the cell’s redox and synthetic pathways, as well as biosynthetic precursors (162,166). Flux through the PPP is known to be elevated during M1 macrophage activation (164,182). This pathway is divided into the non-oxidative branch (non-oxPPP) and the oxidative (oxPPP) branch.

The non-oxPPP diverts intermediates from the glycolytic pathway, converting them, through a series of reversible reactions, to precursors of nucleotide and amino acid precursors. It can also recycle ribose-5-phosphate to glycolytic intermediates through reverse flux (164,166). The non-oxPPP pathway is known for functioning as a hub for primary carbohydrate metabolism by catalyzing the transfer of keto-groups to various aldose acceptors via the enzymes transketolase (TK) and transaldolase (TALDO)

(183,184). These enzymes act by interconverting carbohydrate-phosphates, with carbon chains of three to seven carbon-atoms, without the consumption of ATP (166,185).

However, it is important for another reason in macrophage polarization. As mentioned above, the enzyme ‘carbohydrate kinase-like protein’ (Shpk, formerly known as

CARKL) (164), a sedoheptulose kinase in the non-oxidative branch of the PPP, is a key control point in macrophage polarization (142,166). M1 macrophage activation results in the rapid degradation of Shpk mRNA within an hour of LPS stimulation in both mice and humans. Conversely, in IL-4 stimulated, M2 activated, macrophages, Shpk protein levels are steadily maintained, or even increased (164,167). If Shpk is expressed exogenously during LPS macrophage activation, it results in the accumulation of pentose phosphate intermediates, and oxidized redox couples, indicating an imbalance in the cell’s redox system, as well as blunted superoxide production (166,186).

The oxPPP decarboxylates glucose-6-phosphate (a glycolytic intermediate), through a series of irreversible reactions, to ribose 5-phosphate (187). The byproducts of this reaction are the reduction of two molecules of NADP+ to NADPH and the liberation of one molecule of CO2 (162,187,188). The NADPH produced by this arm of the pathway provides both redox power for the cell and as a requirement for de-novo fatty acid synthesis (166,189). During an infection, M1 macrophages use NADPH both to fuel

NADPH oxidase, producing reactive oxygen species (ROS) during respiratory bursts, and also to protect the cell from ROS damage by reducing oxidized redox-couples in the glutathione and thioredoxin systems (189-192).

The rate limiting enzyme of the oxidative branch of the PPP, glucose-6- phosphate dehydrogenase (G6PD), is also highly active in M1 activated macrophages

(164,182). As indicated above, during early macrophage activation (1-2 hours)

hexokinase II localizes to the mitochondria and uses mitochondria-derived ATP to produce glucose-6-phosphate, which is then funneled to G6PD for the production of

NADPH (166,188). In addition, elevated oxPPP also generates ribose sugars, such as ribose 5-phosphate, as byproducts (162,188). It is thought that these ribose sugars are critical, as M1 macrophage activation requires the production of enough pentose phosphates to sustain their elevated transcriptional activities (166,185). Although both the nonPPP and oxPPP branches of the PPP produce C5 sugars, and increase upon M1 macrophages activation, most of the ribose sugars are derived from the nonPPP branch during early activation period (156,193). Research indicates that this is another key control point in M1 macrophage activation. Overexpression of G6PD in RAW 264.7cells, an immortalized murine macrophage-like cell line, enhanced the activation of NFκB and p38-MAPK signaling pathways and potentiated the expression of pro-inflammatory cytokines as well as ROS production (194).

1.3.4 Fatty Acid Synthesis and Oxidation in Macrophage Polarization

To date, fatty acid oxidation does not seem to play an important role in macrophage polarization. Unlike glycolytic metabolism, which most often occurs in inflammatory, rapidly proliferating and short-lived cells, fatty acid oxidation is observed mainly in long-lived, immune-suppressive, cells, such as Treg cells, memory T cells and

M2 activated macrophages (142,180,181,195,196). In contrast, fatty acid synthesis, or the

accumulation of long chain intracellular fatty acids through lipid import, seems to positively regulate the activation of pro-inflammatory macrophages (197-201).

It’s only over the past ten years that studies have appeared suggesting that fatty acid synthesis promotes M1 macrophage responses. Transcriptional regulation of lipid metabolism is tightly controlled by two sets of competing transcription factor families: the SREBPs (sterol regulatory element-binding transcription factors) and LXRs (liver X receptors) (142,202). Both families have isoforms that are highly expressed in macrophages (SREBP-1c, SREBP-1a and LXRα) (203,204). In general, SREBP activity is associated with M1 macrophage activation, while LXR activity is linked to M2 macrophage activation (185).

The expression of SREBP-1c (sterol regulatory element-binding transcription factor 1c) was found to be upregulated during the differentiation of monocytes into macrophages following treatment with macrophage colony-stimulating factor (M-CSF), leading to increased expression of fatty acid synthesis-related target genes such as FASN and a functional increase in lipid synthesis. Increased fatty acid synthesis in this setting was found to be essential for the generation of M1 macrophages (205,206).

Furthermore, stimulation of macrophages with LPS was shown to upregulate the lipogenic transcription factor SREBP-1a, a known regulator of M1 macrophage cytokine secretion (204,207). When macrophages lacking SREBP-1a were stimulated with LPS, they failed to activate lipid synthesis, as well as IL-1β secretion, and SREBP-1a-deficient

mice were protected against LPS-induced endotoxic shock (204,207). In addition, research showed that activation of SREBP-1a induces the expression of the core inflammasome components and induced the cleavage of pro-IL-1β and pro-IL-18. These findings suggest that SREBP activation is linked to M1 macrophage activation

(204,207,208).

Conversely, mice deficient in the M2-phenotype associated LXRα presented with accelerated atherosclerosis from the accumulation of cholesterol in macrophages (209).

LXRs regulate cholesterol homeostasis, lipid efflux, lipid transport and fatty acid remodeling (210,211). When overexpressed, or pharmacologically activated, LXRα suppresses LPS-induced inflammation and TNFα secretion in macrophages by inhibiting NFκB and AP-1signalling pathways (211,212).

Finally, it was also found that expression of the mitochondrial uncoupling protein 2 (UCP2) stimulates de-novo fatty acid synthesis through fatty acid synthase

(FASN) in LPS stimulated macrophages, a process that activates both ATK / p38 MAPK signaling and inflammasome activity, and which leads to a harmful inflammatory response during sepsis (208). As fatty acid synthesis is an essential metabolic pathway connected with the M1 macrophage phenotype, it is of note that fatty acid synthesis requires NADPH production from the PPP, phosphoenolpyruvate/lactate production from glycolysis and citrate from the metabolic pathway examined next: the tricarboxylic acid cycle (166,169,208).

1.3.5 The Tricarboxylic Acid Cycle is Broken in Two Places During M1 Macrophage Polarization

Maintaining an intact tricarboxylic acid (TCA) cycle is critical for cells, like M2 activated macrophages, that rely on oxidative phosphorylation as the main energy source of the cell. This is not the case for M1 activated macrophages. In these cells an intact TCA cycle serves different purposes: it provides a source of biosynthetic intermediates and anti-bacterial effectors, as well as a mechanism for activating HIF1α- mediated inflammatory functions. To perform these functions, the TCA cycle has can be broken in two places: after citrate synthesis and after succinate synthesis (142,213).

After synthesis, citrate is exported from the mitochondria to the cytosol via the mitochondrial citrate carrier, solute carrier family 25 member 1 (Slc25a1), whose expression is upregulated via NFκB in LPS-stimulated macrophages (163). The cytosolic citrate is then converted in back to oxaloacetate and acetyl-CoA by ATP-citrate lyase

(ACLY) (163,191,214). Acetyl-CoA can then be funneled into multiple synthetic pathways, such as lipid membrane biogenesis, sterol and prostaglandin biosynthesis, and the generation of nitric oxide and reactive oxygen species production (191,214).

Further, the production of acetyl-CoA by ACYL is also an important source of acetyl groups for histone acetylation, and ACLY is a known upstream regulator of the transcription of many glycolytic enzymes, including hexokinase II, phosphofructokinase and lactate dehydrogenase (215). If acetyl-CoA synthesis is muted by silencing ACYL, the expression of these glycolytic genes is suppressed, decreasing the glycolytic rate

(191,214,216). Thus, this section of the TCA cycle not only drives fatty acid synthesis and accumulation, but is also essential for the shift to a glycolytic metabolism and amplification of the glycolytic rate – processes all critical for M1 macrophage effector functions.

The mechanism of this TCA cycle break is just beginning to be elucidated. In addition to being converted to Acetyl-CoA in M1 macrophage activation, citrate is also converted to the metabolite itaconate (217,218). Itaconate, one of the most highly induced metabolites in activated macrophages (219), is believed to act as a potent antimicrobial by inhibiting bacterial isocitrate lyase (218), and has been shown to have a substantial bactericidal effect against S. enterica, M. tuberculosis, and L. pneumonophilia

(218,220,221). The conversion of citrate to itaconate is mediated by mitochondria- associated enzyme Irg1 (immune-responsive gene 1) (217,222). The expression of Irg1, along with suppressed expression of the enzyme isocitrate lyase (catalyzing the conversion of isocitrate to α-ketoglutarate), occurs in response to LPS stimulation and is believed to cause the first break in the TCA cycle (217,218,221). In addition, a recent report demonstrated that itaconate production also functions as an inhibitor of succinate dehydrogenase, inducing a second break in the TCA cycle (219).

The second break in the TCA cycle causes a build-up of succinate, and results in the stabilization of the transcription factor HIF-1α, promoting the amplification of glycolytic flux during M1 macrophage activation and driving inflammation

(176,177,223). As citrate is diverted in the first break of the TCA cycle, the carbon requirement for this second portion of the cycle is replenished by glutamine. Glutamine is converted to α-ketoglutarate through glutaminolysis (193,224), fueling succinate production. The accumulation of succinate is thought to competitively inhibit the activity of prolyl hydroxylases, preventing them from hydroxylating the protein HIF-1α for proteomic degradation (219,225,226). Thus stabilized HIF-1α is able to travel to the nucleus and induce gene expression.

In conclusion, recent advances in immunometabolism have highlighted the various ways that metabolic pathways underpin M1 macrophage activation. Indeed, this analysis shows that M1 macrophage activation induces a complete re-organization of the major metabolic pathways of the cell, and that they are essential for macrophage activation.

1.4 Mitochondrial Dynamics and Their Functions

Contrary to their depiction in introductory biology textbooks, mitochondria do not form a population of discrete and autonomous organelles within the cell. Instead, the mitochondria in a cell should be considered as an interconnected, dynamic network that changes based on the needs and functions of the cell (227-229). Even during basic cellular functions, the cell’s mitochondrial network is undergoing constant restructuring.

For example, during the cell cycle mitochondria elongate during the G1/S transition and fragment at the onset of cell division (230,231). This section will explore the role that

mitochondrial dynamics play in cellular function and the consequences of aberrant or defective mitochondrial dynamics on the cell and the host.

1.4.1 Mechanisms of mitochondrial fission and fusion

Equilibrium between the opposing forces of mitochondrial fission and fusion is required for proper mitochondrial function and response to stress or immune challenge.

Fission and fusion are both carried out by large dynamin-like GTPases. As mentioned in section 1.2.1, the family of large dynamin-like GTPases are known for having a conserved G domain for binding GTP/GDP, the ability to self-oligomerize into dimers and/or oligomers (48,54) and for mediating membrane functions/tabulation

(42,43,46,54). All these functions hold true for the mediators of mitochondrial fission and fusion.

1.4.1.1 Mitochondrial Fission

The dynamin related GTPase Drp1 is the classical core component of mitochondrial division in mammalian cells. Drp1 oligomerizes into helical structures that wrap around mitochondria, mediating the scission of a single mitochondrion into two mitochondria (232,233). GTP binding drives Drp1 helical assembly, forming a catalytic interface between the GTPase domains of Drp1 molecules in adjacent helical rungs (234-236). The formation of the catalytic interface triggers GTP hydrolysis, driving conformational changes that result in helix constriction ending in the scission of mitochondrial membranes (233,237,238).

As one might expect, Drp1-dependent scission is regulated at many points during this process. In addition to the production of several alternative RNA splicing

Drp1 isoforms, Drp1 proteins themselves are subject to a large amount of post- translational modification, including phosphorylation, sumoylation, ubiquitination, nitrosylation and O-glycosylation, demonstrating that their activity is tightly controlled

(239-242). Recruitment and assembly of Drp1 to the mitochondrial outer membrane is mediated by many mitochondrial recruitment effectors, including Fis1, Mff, MiD49 and

MiD51, which provide additional spatial and temporal regulation of mitochondrial fission (243-247). Thus mitochondrial division is tailored to individual cellular function through distinct combinations of Drp1 post-translational modifications, effectors and isoforms.

While the process of mitochondrial fission is mediated by Drp1, the site selection of mitochondrial division occurs at contact sites between the ER and the mitochondria.

The early stages of mitochondrial division occur upstream of Drp1 recruitment, with the

ER physically wrapping itself around the mitochondrion (238,248). The ER mediates an initial stage of mitochondrial constriction, aided by the ER-associated formin INF2.

Research indicates that INF2 initiates actin polymerization and recruitment of myosin to mitochondria-ER contact sites called MAMs for mitochondria associated ER membranes, providing the force to drive the initial stage of mitochondrial constriction (238,249). This initial constriction is thought to provide a favorable nexus for the assembly of a Drp1

helix (239). Although there is not a distinct ER-mitochondrial tethering complex in mammalian cells, the MAM does possess a distinct proteome, including the presence of

Miro1, Miro2, and the mitochondrial fusion protein mitofusin 2, which together are required for mitochondrial fission (250,251) . The MAM is also recognized as a hub for nutrient sensing and lipid biosynthesis, linking mitochondrial fission to these processes

(250,252,253).

1.4.1.2 Mitochondrial Fusion

In mitochondrial fusion, the dynamin-like GTPases mitofusin 1 and mitofusin 2 mediate the fusion of the outer mitochondrial membrane, while Opa1 mediates inner membrane fusion (253-257). Although the mechanistic knowledge of Drp1-mediated fission is far from complete, still less is known about the physical process of mitochondrial fusion. Mitochondrial fusion is known to involve the sequential fusion of the outer and inner membranes (227).

The current state of the field suggests that each membrane fusion event occurs via two discrete stages – membrane tethering by the GTPases, followed by lipid mixing of the two mitochondria at the fusion site (254,255,258,259). Mitochondrial membrane tethering is known to be mediated by the oligomerization of the mitofusins or Opa1, while GTP hydrolysis is thought to drive lipid mixing by destabilizing the lipid bilayers of the adjacent mitochondria (254,258,260). Like Drp1, the functions of all three GTPases are highly regulated by physiological stimuli and cellular stresses through various

mechanisms, including alternative mRNA splicing, protein degradation, and proteolytic processing and post translational regulation (227,253).

1.4.2 Mitochondrial Dynamics in Response to Stress

Mitochondrial stress-induced pathways, often triggered by perturbations in electron transport chain function and/or reduction of membrane potential, routinely lead to changes in mitochondrial morphology. The mitochondrial inner membrane fusion

GTPase Opa1 is thought to act as a toggle between pro-fission and pro-fusion states

(253,260). In healthy respirating cells, Opa1 proteolytic processing occurs constitutively, to generate both long transmembrane anchored and short, soluble isoforms, both of which are required for proper fusion (258,260,261). However, a decrease in mitochondrial membrane potential results in the proteolytic processing of the long Opa1 isoforms by a metalloprotease, OMA1, blocking mitochondrial fusion and causing the mitochondrial network to fragment (261,262). This mitochondrial fragmentation induces autophagic degradation of damaged mitochondria (termed mitophagy) or potentiates cell death. Mitochondrial fragmentation serves to isolate damaged mitochondria that have a low membrane potential, as they are unable to fuse back into the mitochondrial network (239,263). These damaged mitochondria accumulate the kinase PINK1 on their outer mitochondrial membrane, recruiting the E3 ligase, Parkin, to ubiquitinate other mitochondrial outer membrane proteins specifically involved in motility and fusion, and enhancing the likelihood of their removal through autophagy (264-266). A more terminal

response to severe mitochondrial damage through fragmentation is cell death.

Mitochondrial fission actively promotes cell death, as Drp1 facilitates the recruitment and activation of the pro-apoptotic Bcl-2 protein BAX to the mitochondrial outer membrane (267-270).

Conversely, long OPA1 isoforms are also required for a different stress-induced response, termed mitochondrial hyperfusion (253). Mitochondrial hyperfusion causes the formation of a highly tubulated, interconnected mitochondrial network. This stress response is thought to buffer the mitochondrial network against more mild cellular stresses, such as ultraviolet radiation, transient depolarization or nutrient starvation

(271,272). Under nutrient starvation mitochondrial elongation is believed to protect the mitochondrial network from autophagic degradation through steric hindrance (271), while more recent work indicates it functions as a mechanism to maintain ATP production when the electron transport chain activity is mildly impaired (231,273). In addition, elevated levels of oxidized glutathione during oxidative stress has been shown to enhance fusion by promoting disulfide-mediated dimerization of MFN2 (274).

However, mitochondrial hyperfusion is a short-term adaptation to stress, and cannot buffer a long-term mitochondrial dysfunction. Therefore, if the mitochondrial stress is not resolved, the cells will eventually upregulate mitochondrial fragmentation (227,253).

1.4.3 Mitochondrial Dynamics in Metabolism

The energetic states of cells are often associated with specific mitochondrial morphologies. Mitochondrial fusion and elongated mitochondria have been associated with increases in OXPHOS activity (275). This may be due, in part, to the necessity of mitochondrial fusion for the maintenance of mitochondrial DNA (mtDNA), and therefore for respiration (276-279).

Ablation of mitochondria fusion, through deletion of both mitofusins or Opa1, results in a significant decrease of mtDNA, membrane potential, and respiratory chain activity in both cell culture and mouse tissue (278,279). Additionally, OXPHOS may be more efficient in elongated mitochondria. The fast oxidative fibers in skeletal muscle demonstrate increased mitochondrial fusion that presumably supports their highly active mitochondrial population (280,281). Moreover, elongated mitochondria have frequently been observed in conditions associated with an increase in ATP production

(231,273). Additionally, studies in cultured yeast and human cells suggest that mitochondrial elongation occurs when cells are grown in the absence of glucose, forcing the cells to rely more substantially on OXPHOS for ATP production (282,283).

Mitochondrial fission is equally important in cellular metabolism. Dysregulated mitochondrial fragmentation is associated with a lack of oxidative phosphorylation and membrane depolarization (241,284). This is due in part to that fact that fission is required to isolate dysfunctional mitochondria that are then degraded through mitophagy

(264,266,285). This mitochondrial quality control mechanism enriches the mitochondrial pool with healthy mitochondria that effectively increases the cell’s oxidative capacity

(240,241). Fission also facilitates the proper subcellular distribution of mitochondria, and thus ATP production (253). As an extreme example, fission is critical in neurons, where individual mitochondria travel to the axon terminal to drive synaptic processes, and therefore need to be small enough to fit through the narrow axon (286,287).

In conclusion, although mitochondria have long been recognized for their roles in metabolism, the importance of their cellular distribution and morphology in cellular function has only been recognized in the past 15 years. Since then, a substantial amount of progress has been made in defining the mechanisms of mitochondrial fission and fusion, even though our knowledge of these pathways is far from complete. Similarly, the role of mitochondrial dynamics is far from complete, despite the advances made in understanding how mitochondrial morphology contributes to distinct cellular functions, and adapts to cellular stress. Further research is needed to determine how mitochondrial dynamics contribute to generating appropriate responses to environmental signals.

1.5 Motivation for this Work

Interferon gamma plays a critical role in restricting infection of microbial organisms (288). One of the mechanisms through which it accomplishes this is by inducing the expression of IFN-γ-inducible effectors, including the immunity related

GTPases. The IRGs are best known for providing a cell-autonomous innate immune defense against intracellular pathogens (51,65,89,109). The lack of one particular IRG,

Irgm1, leads to a striking susceptibility to multiple intracellular pathogens, indicating that it is uniquely important in innate immune restriction (65,106,109,114). In addition, two studies investigating the lymphopenia observed in Irgm1-deficient mice imply that it may negatively regulate IFN-γ signaling in lymphocytes. In humans, the single functional IRG, IRGM, is an orthologue of mouse Irgm1 and is thought to share some of its properties, including a role in autophagy (108,120,134,289). Similarly, IRGM has been identified as a susceptibility locus for Crohn’s disease (129,130), while Irgm1-deficient mice demonstrate increased acute intestinal inflammation and worsened clinical responses in response to exposure to the intestinal irritant dextran sodium sulfate

(115,116). Indeed, during the analysis of electron micrographs, our lab noticed a buildup of swollen, elongated mitochondria in the ileal tissue of Irgm1-deficient mice (116).

Around this same time, a paper was published by the Deretic lab demonstrating that a human IRGM isoform localizes to the mitochondria, causing fragmentation of the mitochondrial network and mitochondrial depolarization (120). The impetus for this

dissertation was to determine if there was a similar mitochondrial phenotype in our

Irgm1-deficient mice, and, if so, to uncover the mechanism causing it.

In chapter 2, I sought to identify if Irgm1 played a role in modulating mitochondria in our murine embryonic fibroblast cell culture model. I identified a tubular, interconnected mitochondrial phenotype in Irgm1-deficient MEFs stimulated with IFN-γ, and confirmed that Irgm1 localizes to mitochondria. In addition, I characterized which structural determinants were required for Irgm1’s mitochondrial localization and modulation of mitochondrial dynamics (75).

No drastic phenotypes in mitochondrial biology were observed in the absence of

Irgm1 in chapter 2. Therefore, in chapter 3 I hypothesized that Irgm1-deficiency caused a shift in cellular metabolism. I found that Irgm1 had an increased glycolytic rate. Further metabolic studies showed that Irgm1-deficient MEFs, in addition to a metabolic shift towards glycolysis, also had metabolite deficiencies. At this point I shifted the project to our more functionally-relevant bone marrow macrophage cell culture model. I found that Irgm1-deficient BMM either primed with IFN-γ, or stimulated with IFN-γ/LPS had a similar increase in glycolytic rate, a buildup of neutral lipids and altered levels of the tricarboxylic acid cycle metabolites, as well as urea cycle intermediates metabolites (140).

Surprisingly, these aberrant shifts in the metabolism of Irgm1-deficient cells were most evident in macrophages primed only with IFN-γ, suggesting that Irgm1 expression negatively regulates immunometabolic activation.

As shifts in immunometabolism often result in altered cytokine secretion, in chapter 4 I examined whether Irgm1-deficient macrophages had altered cytokine secretion in the presence of IFN-γ alone, or both IFN-γ and LPS. In both cases we found altered cytokine secretion that depended on the immunometabolic changes in these cells

(140), and overexpression of pro-inflammatory cytokines in the serum of both uninfected and S. typhimurium infected Irgm1-deficient mice (114,140).

Taken together, these changes define a novel role of Irgm1 in modulating immunometabolism, and gives insight into how the severe lymphopenia observed in

Irgm1-deficient mice occurs.

2. Irgm1-Deficiency in MEFs Causes Changes in Mitochondrial Morphology

[Note: this chapter contains research and excerpted text from the following publications:

Liu B, Gulati AS, Cantillana V, Henry SC, Schmidt EA, et al. (2013) Irgm1- deficient mice exhibit Paneth cell abnormalities and increased susceptibility to acute intestinal inflammation. Am J Physiol Gastrointest Liver Physiol 305: G573–584

Henry S. C., Schmidt E. A., Fessler M. B. & Taylor G. A. (2014) Palmitoylation of the immunity related GTPase, Irgm1: impact on membrane localization and ability to promote mitochondrial fission. PLoS ONE 9, e95021]

2.1 Introduction

Immunity related GTPases (IRGs) are a family of IFN-γ-inducible large GTPases critical for vertebrate innate immunity and are implicated in diverse intra-cellular processes, including vesicle formation, transport, cell motility, and resistance against intracellular protozoan and bacterial pathogens (42,96,290).

The IRG family is particularly extensive in the mouse, with the C57Bl/6 strain containing 23 genes. These encode proteins around 47kDa that can be subdivided into the GKS and GMS protein families, based on the presence of canonical GX4GKS sequence, or a non-canonical GX4GMS sequence in their GTP-binding motif (65).

Transcription of all mouse IRGs - with the exception of Irgc which is constitutively expressed in the testes - is stimulated by IFN-γ in a STAT1-dependent manner. The three murine IRGs containing the GMS sequence (Irgm1, Irgm2, Irgm3) are particularly 50

important in resistance against intracellular pathogens, and seem to control the localization of the GKS IRGs (65,109).

While most mouse IRGs are required for resistance against some intracellular pathogens, the GMS Irgm1 protein is unique in that it is required for host resistance to the widest range of intracellular pathogens (89). Irgm1 is also implicated in the expansion of hematopoietic stem cell and progenitor cell populations and the survival of

IFN-γ primed CD4 + T lymphocytes (122,123). Collectively, these results suggest that

Irgm1 has distinct immune functions that do not appear to be shared by other mouse

IRGs.

The human IRG family has only two members: IRGM, that is orthologous to the mouse Irgm genes, and IRGC, which is only produced in the testes. Despite being severely truncated in both the N- and C-terminal, IRGM was shown to be implicated in resistance against M. tuberculosis (291), and was identified as a susceptibility locus in

Crohn’s disease (129,130,292). At the beginning of this project a third function was assigned to human IRGM: the Deteric lab found that IRGM localized to mitochondria and affects mitochondrial fission. When IRGM was knocked down in U937 cells, elongated networks of mitochondria were observed, indicating a lack of mitochondrial fission. Moreover, this lack of IRGM mediated fission also inhibited the cell’s ability to undergo IFN-γ and starvation induced autophagy (120).

It had been reported previously that mouse Irgm1 also localized to mitochondria

(76,293), but this mitochondrial localization had not been characterized, and the function of Irgm1 in mitochondrial biology had never been studied. The final impetus to investigate Irgm1’s role at the mitochondrial level came when our lab discovered swollen and distended mitochondria in electron micrographs of Irgm1-deficient ileal enterocytes (116).

I hypothesized that mouse Irgm1 played a similar role to its human orthologue, and influenced mitochondrial biology. In this chapter I describe a novel mitochondrial phenotype in Irgm1-deficient MEFs - that of the formation of a hyperfused mitochondrial network after priming with IFN-γ. Further, I characterize the structural determinants that determine Irgm1’s mitochondrial localization and modulation of mitochondrial dynamics (75), and examine mitochondria-associated properties and functions in MEFs with the goal of finding a shift in function in the absence of Irgm1.

No striking difference in mitochondria-associated function was discovered, other than small, but significant, increases in reactive oxygen species, and decreases in IFN-γ

Irgm1-deficient MEFs. In addition, increased mitochondrial fragmentation was observed in Irgm1-deficient MEFs in response to oxidative stress, indicating a potential shift in the redox regulation of Irgm1-deficient cells. Curiously, Irgm1’s GTPase activity and mitochondrial localization were both required for the mitochondrial fragmentation

observed in WT MEFs, suggesting that the mitochondrial hyperfusion in Irgm1- deficient MEFs was more than a mere artifact.

2.2 Results

2.2.1 Evidence of Altered Mitochondrial Morphology in Irgm1- Deficient Enterocytes

Figure 1: Accumulation of swollen and elongated in the ileum of Irgm1-deficient mice

Representative transmission electron micrographs from (A) mouse ileal enterocytes, (B) Irgm1 KO ileal enterocytes, and (C) Irgm1-deficient goblet cells (C). Magnification ×20,000. Note swollen mitochondria (arrowheads) and tubular mitochondria (outlined with dashed line) in Irgm1- deficient cells in B and C, respectively. These phenotypes were seen in multiple Irgm1 KO enterocytes and goblet and Paneth cells and were not cell-specific.

As human IRGM was found to be a susceptibility locus for Crohn’s disease, our lab, in conjunction with Dr. Sartor’s lab, examined the pathogenesis of dextran sodium- induced colitis on Irgm1-deficient mice. Irgm1-deficient mice showed increased acute inflammation in the colon and ileum, worsened clinical responses to dextran sulfate,

impaired autophagy in intestinal cells and abnormalities in the mitochondria of Irgm1- deficient Paneth and other intestinal cells (116).

Mitochondria normally cycle between punctate and tubular forms, with this cycle being highly regulated by a network of mitochondrial fission and fusion proteins, as well as being linked to the process of mitophagy (294,295). In Irgm1-deficient ileal cells, our lab commonly observed elongated/tubular mitochondria (Figure 1). Moreover, we found numerous swollen mitochondria in Irgm1-deficient intestinal tissue, likely representing defective organelles that had not been removed from the cell, that were absent in WT tissue. The accumulation of damaged and elongated mitochondria in the intestine of Irgm1-deficient mice, suggesting a defect in mitochondrial equilibrium and/or function in the cell, prompted us to examine the role of Irgm1 in mediating mitochondrial biology.

2.2.2 A Subset of Irgm1 Localizes to the Mitochondria and Induces Mitochondrial Fragmentation

In addition to the mitochondrial abnormalities observed in Irgm1-deficient intestinal cells, a paper came out implicating human IRGM, an orthologue of mouse

Irgm1, in mitochondrial function. IRGM was shown to localize to mitochondria and regulate mitochondrial fission, potentially influencing the removal of damaged mitochondria through mitophagy (120). To begin, our lab explored whether Irgm1 possessed similar function in cultured mouse embryonic fibroblasts (MEFs).We verified 54

that, just like previous reports, Irgm1 localized strongly to the mitochondrial surface

(Figure 2) (120,296). Indeed, Irgm1 displayed a very punctate staining pattern that overlaid or closely opposed to discontinuous areas of the mitochondria (76,296).

Figure 2: Irgm1 localizes to the mitochondria in IFN-γ primed WT MEFs

Representative WT mouse embryonic fibroblasts transfected with pMito to label mitochondria, exposed to IFN-γ for 24 h, and then processed for staining with anti- Irgm1 antibodies.

Mitochondria normally cycle between punctate and tubular forms, with this cycle being highly regulated by a network of mitochondrial fission and fusion proteins.

The equilibrium between mitochondrial fission and fusion present in a population of healthy WT cells can be qualitatively measured as the percent of cells that present with a 55

Figure 3: Hyperfused mitochondrial morphology observed in Irgm1-deficient MEFs

(A) Representative images of the mitochondrial morphologies observed in a population of mouse embryonic fibroblasts exposed to IFN-γ for 24 hours, stained with the mitochondrial outer membrane protein TOM20 to label mitochondria. Magnification ×1,000. (B) Quantified punctate, elongated, or mixed mitochondrial morphologies in WT and Irgm1-deficient MEFs exposed to IFN-γ for 24 hours. (C) Mitochondrial morphology of Irgm1-deficient MEFs transfected with either Irgm1 or a control plasmid. MEF mitochondrial morphology was assessed blindly from >50 cells per treatment group per experiment, with the averages of 4 experiments shown. Values are means ± SE. *P < 0.05; **P < 0.01

punctate mitochondrial network, an elongated/tubular network or, most often, a mix of both phenotypes (263,297,298) (Figure 3 A). Our data showed that in the absence of

Irgm1, IFN-γ-primed MEFs contain elongated networks of mitochondria (Figure 3 B), suggesting that a similar defect to the aberrant mitochondrial equilibrium and/or function found in ileal cells was also present in our MEF model. These results indicate that Irgm1 expression may drive mitochondrial fission in MEFs. To address this potential activity for Irgm1, we overexpressed Irgm1 in primary fibroblasts. Irgm1 expression promoted mitochondrial fragmentation, pushing the mitochondrial equilibrium in the cell toward more punctate forms in both WT and Irgm1-deficient

MEFs (Figure 3 C). Collectively, these data suggest another function of Irgm1 may be to mediate the mitochondrial fragmentation seen in IFN-γ-primed MEFs, or mediate the mechanisms responsible for this fragmentation.

2.2.3 Characterization of the Structural Elements Required for Irgm1’s Mitochondrial Localization and Irgm1-Dependent Mitochondrial Fragmentation

As stated previously, IRG proteins are large, dynamin-like GTPases, containing a helical domain with N- and C-terminal elements, and a Ras-like G domain (52).

Although Irga6 is the only IRG with a crystal structure, its structural elements, other than the highly conserved GxxxxGKS/T P-loop sequence in the first nucleotide-binding motif, are quite similar to Irgm1 (Figure 4 A) (52,65,105). To address the importance of structural features of Irgm1 in mitochondrial localization and mitochondrial fragmentation, our lab generated several Irgm1 mutants. For mitochondrial localization,

Figure 4: Irgm1 structural determinants required for mitochondrial localization

(A) Representative Irga6 crystal structure taken from Ghosh et al. Mol Cell. 2004 Sep 10;15(5):727-39. (B) Representative gene map denoting Irgm1 structural determinants. Irgm1-deficient MEFs were transfected with plasmids expressing wild-type or mutant Irgm1 proteins, as indicated. The cells were exposed to IFN-γ for 24 h, stained with anti- Irgm1 and anti-Tom20 antibodies, and used for immunofluorescence analysis. The experiment was performed 3 times, with at least 20 cells analyzed per group in each experiment. (C) Shown are images from representative cells. The scale bar represents 20 µm. (D) Co-localization analysis was performed to quantify overlap between the Irgm1 and Tom20 signals. In each experiment, the degree of co-localization was averaged for cells within an experimental group, and these values were then averaged across the three experiments, with error bars representing standard error the mean, and * representing p<0.05. 58

We generated a mutant Irgm1 with alanine substitutions of cysteines known to be palmitoylated, Irgm1(C371,373,374,375A), a mutant Irgm1 containing an insertional disruption of the predicted amphipathic α-helix, αK, which spans residues 356–369 in the C-terminal portion of Irgm1, Irgm1(ins362,367E), and a mutant Irgm1 containing contained both the palmitoylation and αK mutations,

Irgm1(ins362,367E;C371,373,374,375A) (Figure 4 B) (75,76). Wild-type Irgm1 and the three mutants described above were expressed in Irgm1-deficient fibroblasts, and those cells were then used for immunofluorescence analysis (Figure 4 A, B). The

Irgm1(C371,373,374,375A) palmitoylation mutant showed nearly as strong localization to the mitochondria as wild-type Irgm1 (Figure 4C), although a small but reproducible loss of localization in the mutant was measured by co-localization analysis (Figure 2.4D).

The Irgm1(ins362,367E) mutant lacking a functional αK domain displayed a more obvious decrease in localization, by both subjective image analysis (Figure 4C) and quantitative co-localization analysis (Figure 4D). The decrease in localization was even more pronounced in the mutant lacking the palmitoylation and the αK domains

Irgm1(ins362,367E;C371,373,374,375A), with little apparent mitochondrial localization remaining with subjective image analysis (Figure 2.4C). Thus, as with the Golgi, the αK and palmitoylation domains cooperate to enable Irgm1 association with mitochondria.

Figure 5: Irgm1 structural determinants required for increased mitochondrial fragmentation seen in WT MEFs primed with IFN-γ.

Irgm1-deficient MEFs were transfected with plasmids expressing wild-type or mutant Irgm1 proteins, as indicated. The cells were exposed to IFN-γ for 24 h, stained with anti- Irgm1 and anti-Tom20 antibodies, and used for immunofluorescence analysis. Images were collected from cells that expressed the Irgm1 proteins; the mitochondria in these images were scored in a blinded fashion as being punctate, tubular, or mixed phenotype. 50 cells were scored per experimental group, and the results displayed as percent of the total. Shown is the average of four separate studies, ± standard error, * p<0.05, ** p<0.01, *** p<0.001

We next addressed which structural elements were necessary for Irgm1’s functioning in mitochondria by examining the mitochondrial morphologies of Irgm1- deficient MEFs overexpressing the Irgm1 mutants described above (Figure 5). The palmitoylation mutant, Irgm1(C371,373,374,375A), displayed a modest but statistically significant decrease in its ability to shift the mitochondrial equilibrium toward punctate forms and away from tubular forms. The activity was more dramatically undermined in

the αK mutant Irgm1(ins362,367E) and the combined αK/palmitoylation mutant

Irgm1(ins362,367E;C371,373,374,375A). Additionally, an Irgm1 (S90N) mutant was tested that has a greatly reduced affinity for GTP and thus impaired GTPase functioning

(70,73). The GTPase domain in IRG proteins has previously been shown to be necessary for their dimerization (74), and in the present studies, it was also required for the Irgm1- driven promotion of mitochondrial fission. These results suggest that palmitoylation alone does have a small impact on the ability of Irgm1 to function in the mitochondria, while the αK motif has a more dominant effect (as does the GTPase activity of Irgm1).

The data underscore that Irgm1 functioning in the mitochondria tracks closely with its ability to associate with mitochondrial membranes.

2.2.4 Irgm1-Dependent Mitochondrial Fragmentation is Dependent on Drp1 Activity

Drp1 is a large dynamin-like GTPase that is required for the majority of mitochondrial fission in the cell. During mitochondrial fission, Drp1 oligomerizes around the fission site, and contracts around the mitochondria, aiding to scission it into two separate mitochondria. In the absence of Drp1, mitochondria take on elongated networks, like those observed in our Irgm1-deficient MEFs.

To determine whether the mitochondrial fragmentation observed after Irgm1 expression is dependent on Drp1 function, Irgm1-deficient MEFs were transfected with plasmids expressing WT Irgm1 and either wild-type (Drp1) or a dominant negative 61

Drp1 (Drp1 K38A), and mitochondrial phenotypes were measured (Figure 6). Cells expressing Irgm1 and Drp1 K38A revert back to the Irgm1-deficient MEFs elongated mitochondrial phenotype. Thus Irgm1-mediated mitochondrial fission is dependent on

Drp1 function.

Figure 6: Irgm1-mediated mitochondrial fragmentation is dependent on the activity of Drp1.

MEFs primed with IFN-γ. Irgm1-deficient MEFs were transfected with plasmids expressing a control plasmid, WT Irgm1, WT Drp1 and/or dominant negative Drp1 K38A, as indicated. The cells were exposed to IFN-γ for 24 hours, stained with anti- Irgm1 and anti-Tom20 antibodies, and used for immunofluorescence analysis. Images were collected from cells that expressed the Irgm1/Drp1 proteins; the mitochondria in these images were scored in a blinded fashion as being punctate, tubular, or mixed phenotype. 50 cells were scored per experimental group, and the results displayed as percent of the total. Shown is the average of four separate studies, ± standard error, * p<0.05, ** p<0.01, *** p<0.001

We first looked at whether there was a change in the expression level of key mitochondrial fission and fusion factors (Figure 7A), ultimately finding no difference in

expression levels between WT and Irgm1-deficient MEFs. We also found no change in rate of cell death between WT and Irgm1-deficient MEFs primed with IFN-γ, measured by LDH release assay (Figure 7B).

Some slight changes were found between the two cell types. A decrease in membrane potential, measured by the dye TMRE, as well as an increase in total ROS production were observed in Irgm1-deficient MEFs, compared to WT (Figure 7 C,D).

One caveat to these changes in membrane potential and total ROS is that they were present even without IFN-γ stimulation, therefore before Irgm1 was induced. We attribute this shift to the small amount of Irgm1 background expression observed in our

WT MEFs (data not shown), probably induced by the secretion of type I IFNs into the media.

2.2.5 Mitochondrial Function in Irgm1-Deficient MEFs

In addition to characterizing the elongated mitochondrial phenotype in Irgm1- deficient MEFs, I searched for differences in mitochondrial function, or changes typically caused by mitochondrial dysfunction.

Another area of interest explored was the response Irgm1-deficient MEFs to mitochondrial stressors, using mitochondrial morphology as a proxy for mitochondrial distress. Again, only small changes were observed in the rate of mitochondrial fragmentation

Figure 7: Comparison of mitochondrial effectors and functions in IFN-γ primed WT and Irgm1-deficient MEFs.

(A) Western blot of the major mitochondrial fission/fusion factors in IFN-γ primed WT and Irgm1-deficient MEFs. (B) LDH release in the supernatant of MEFs. (C) Total ROS levels measured in MEFs with the fluorescent dye CM-CH2DFDA. (D) Mitochondrial

membrane potential measured with the fluorescent dye TMRE. (E) Percent of cells with elongated mitochondria, using the same mitochondrial morphology assay as above, in WT and Irgm1-deficient MEFs over a time course with the mitochondrial uncoupler CCCP (20μM). (F) Mitochondrial morphologies observed in a population of WT and Irgm1-deficient MEFs exposed to IFN-γ for 24 hours, followed by a 1 hour incubation with 300 μM H2O2. The mitochondrial morphologies of each cell population was measured as described above. Each figure experiment, except for the western blots, represents the averages of 3 or more experiments. Values are means ± SE. *P < 0.05; **P < 0.01 in response inhibitors that dissipate the mitochondrial membrane potential, such as

CCCP (Figure 7 E). However, when the cells were stressed with 300 uM H2O2 (Figure 7

F), the mitochondria fragmentation in IFN-γ-primed Irgm1-deficient MEFs far exceeded that of WT MEFs, suggesting that Irgm1 may be involved in maintaining the redox homeostasis of the cell.

2.3 Discussion

In this chapter, I presented evidence that mouse Irgm1 modulates mitochondrial dynamics and influences mitochondrial biology, indicating a potential parity of function with its human orthologue IRGM. I described a novel mitochondrial phenotype in

Irgm1-deficient MEFs, namely that mitochondria take on an elongated network morphology when primed with IFN-γ in the absence of Irgm1. I also determined that

Irgm1’s amphipathic αK helix and palmytoylation domains mediate Irgm1’s localization, while these two domains, in addition to GTPase activity, are required to prevent the hyperfused mitochondrial network seen in Irgm1-deficient MEFs.

The mitochondrial fission activity seen in WT MEFs is dependent on the mitochondrial fission protein Drp1. Although both Irgm1 and Drp1 are large dynamin- like GTPases with roles in the modulation of lipid membrane binding, unpublished results from our lab failed to show any co-localization of Irgm1 and Drp1 on the mitochondrial membrane. Further, the Deretic lab showed that IRGMd, the isoform localizing to mitochondria, was present on the inner mitochondrial membrane, while

Drp1 localizes to its receptors on the outer mitochondrial membrane (120). Although I have not successfully shown the same of mouse Irgm1, given that I was unable to purify mitochondria without Golgi membrane contamination, it is probable that it also localizes to the inner mitochondrial membrane.

Unlike lipid droplets, or lysosomal membranes, severe mis-localization of the

GKS IGS to the mitochondrial membrane has not yet been described in Irgm1-deficient cells (96). This suggests that preventing aberrant GKS IRG activation may not be the sole function of Irgm1 at the mitochondrial membrane. One area of research that might be fruitful is to determine if Irgm1 is enriched in mitochondria at the contact interface of the mitochondrial and ER membranes (MAM). As Irgm1 is implicated in membrane modulation, and functions of the MAM include lipid import, lipid biosynthesis and the regulations of mitochondrial function(299), it is possible that Irgm1 has a function at this interface.

A shift in mitochondrial dynamics is well known to occur in response to cellular stress (300-303), therefore the research in the latter part of this chapter was an attempt to uncover the source of this stress, or at least the mechanism through which the stress was operating. However, while I did observe a slight decrease in membrane potential, an increase in ROS production and an oversensitivity to oxidative stress, I was unable to determine a consequential loss of function phenotype in the mitochondria of Irgm1- deficient MEFs such as an increase in cell death or modulation of mitochondrial fission and fusion factors.

As described previously, mitochondrial fission is a well-known mechanism for segregating dysfunctional mitochondrial components from the mitochondrial network, and tagging them for degradation through mitophagy (284,304). The Deretic lab’s study suggested that human IRGM was involved in mitochondrial degradation through autophagy (120). However, unpublished results in our lab found no evidence of mitophagy upregulation in our WT MEFs, or of dysfunctional mitochondrial accumulation in Irgm1-deficient MEFs.

Instead of viewing increased mitochondrial fission as a consequence of Irgm1 expression, I considered that perhaps the absence of Irgm1 was driving aberrant mitochondrial hyperfusion. Indeed, mitochondrial hyperfusion is seen in cells undergoing mild stress (253), and is thought of as a way to maximize the output of ATP.

It is seen in cells undergoing senescence (305), or under starvation conditions, as a way

to prevent mitochondrial degradation during autophagy (271). The small, but significant increase in ROS, and decrease in TMRE, in addition to the hyperfused mitochondria, suggests that our Irgm1-deficient MEFs are under mild stress, causing this hyperfusion.

In addition, it suggests that Irgm1-deficient IFN-γ-primed MEFs may have decreased oxidative respiration capacity, thus causing the mitochondria to hyperfuse in order maintain ATP production. These results prompted me to examine the metabolic function in Irgm1-deficient MEFs, which is explored in chapter 3.

2.4 Methods

Mice

Knockout C57Bl/6 mice deficient for Irgm1 (Irgm1−/−) were generated as previously described [5]. Mice were housed and maintained under procedures approved by the Institutional Animal Care and Use Committee at Duke University and the

Durham VA Medical Centers. Mouse embryonic fibroblasts (MEF) were isolated from mice and frozen or immortalized by the standard 3T3 procedure as previously described

(88). All primary cells and cell lines were grown and maintained in a humidified atmosphere of 5% CO2 at 37°C in Dulbecco’s Modified Eagle Medium (DMEM, GIBCO,

Life Technologies) supplemented with 2 mM L-glutamine, 4.5 g/L D-glucose, 110 mg/L sodium pyruvate, 10% (v/v) fetal bovine serum (FBS, HyClone, Logan, UT), 100 units/ml penicillin and 100 µg/ml streptomycin (GIBCO, Life Technologies). Where appropriate,

1000 U/ml interferon (IFN)-γ (Calbiochem, EMD Biosciences, San Diego, CA, USA) was included in the growth medium.

DNA Constructs, Mutagenesis and Transfection

The mouse retroviral vector pRV-GFP was a gift from Dr. Carl Feng (sequence deposited with Addgene). Plasmids pGW1H/Irgm1 and pGW1H/Irgm1(ins 363,357E) were a gift from Dr. Jonathan Howard and have been described previously [17]. All restriction endonucleases were obtained from New England Biolabs (Ipswich, MA). To construct pRV-GFP/Irgm1, the full-length Irgm1 coding sequence was excised from another plasmid (pCWX200/Irgm1) as a BamHI/XhoI fragment and inserted into BglII and XhoI sites of pRV-GFP, under control of the viral LTR and upstream of the IRES and

GFP cassettes. Mutagenesis of putative palmitoylation sites in Irgm1 retroviral and plasmid constructs was done using a site-directed mutagenesis kit (QuickChange II XL,

Agilent Technologies, Santa Clara, CA) following the manufacturer’s protocol.

Mutagenic oligonucleotides were obtained from Integrated DNA Technologies

(Coralville, IA) - (Irgm1 (C8A): 5′catcacacagttccgccgaggctgctccac-3′, 5′- gtggagcagcctcggcggaactgtgtgatg-3′; Irgm1 (C257/258A): 5′- ctccaacatcagggccgctgaacccttaaagac-3′, 5′-gtctttaagggttcagcggccctgatgttggag-3′; Irgm1

(C371/373/374/375A): 5′-gagatttctcccagccgtagccgctgctttaagacgcttg-3, 5’- caagcgtcttaaagcagcggctacggctgggagaaatctc-3′). The DNA sequences of all plasmid and

mutant constructions were verified at the Duke University DNA Sequencing Facility

(Durham, NC).

Transfection of primary MEFs and cell lines was done using XtremeGENE 9 reagent (Roche, Indianapolis, IN) following the manufacturer’s recommendations.

Electron Microscopy

Tissue was fixed in 2% paraformaldehyde-2.5% glutaraldehyde in 0.15 M sodium phosphate, pH 7.4. The University of North Carolina-Chapel Hill Microscopy Services

Laboratory processed tissues for transmission electron microscopy (TEM) according to standard techniques. Ultrathin (70-nm) sections were cut with a diamond knife and collected on 200-mesh copper grids. TEM grids were observed and photographed using a transmission electron microscope (Zeiss EM-910, LEO Electron Microscopy,

Thornwood, NY) and photographed using a digital camera (Bioscan, Gatan, Pleasanton,

CA).

Immunofluorescence and Co-Localization

Primary WT and Irgm1-deficient MEFs and were grown at a density of

2×104 cells on poly- D-lysine-coated glass coverslips in 24 well tissue culture plates.

When appropriate, the Irgm1-deficient MEF were transfected with the plasmids indicated in the text. All cells were treated with 100 U/ml IFN-µ for 24 hours prior to

fixation. Cells were fixed in 4% paraformaldehyde (Sigma) in PBS for 15 minutes, permeabilized with 0.2% saponin (Sigma) in PBS for 10 minutes, and blocked for a minimum of 60 minutes in 0.2% saponin/PBS with 10% FBS. Cells were stained with primary antibodies for 60 minutes, washed 3 times with 0.2% saponin, and stained with appropriate fluorochrome-tagged secondary antibodies (Molecular Probes, Life

Technologies, Eugene, OR) for 60 minutes. Monoclonal antibody 1B2 (Irgm1) was used as an undiluted hybridoma culture supernatant with saponin added to 0.2%. Primary antibody FL-145 (TOM20) was used at a dilution of 1∶250 in blocking buffer. Secondary antibodies were used at a dilution of 1∶750 in blocking buffer as recommended by the supplier. Cells were imaged on an Olympus IX70 inverted fluorescence microscope equipped with a Hamamatsu C8484-03G01 digital camera and ASI MS2000 XY Piezo Z stage. Images were collected as z-stacks with a plane thickness of 0.2 µm using

Metamorph version 7.7.5.0. Z-stacks were deconvolved using Auto Quant X3 software.

Co-localization measurements were performed using the Metamorph co-localization application. For co-localization analyses, in at least some experiments, the images were randomized by an independent party, and then assessed blindly. The co-localization analyses were performed on the indicated number of cells within each experiment; the average values from each of three separate experiments were then averaged to produce the displayed values.

Mitochondrial Morphology Assays

Cells were plated on coverslips at a density of 1x104 cells/coverslip. The next day, cells were primed with IFN-γ (Calbiochem) for 24 hours, and then incubated with 20μM

CCCP (Sigma) or 300μM H2O2 (ThermoFisher) where indicated. Cells were then fixed on coverslips with 4% paraformaldehyde (w/v) in PBS for 15 min and permeabilized with

0.2% (w/v) saponin in PBS for 10 min. The cells were then stained for 60 min with anti-

TOM20 rabbit polyclonal antibody (FL-145, Santa Cruz Biotechnologies) at a 1:500 dilution, followed by Alexa Fluor conjugated secondary antibody (Molecular

Probes/Invitrogen) at a 1:750 dilution for an additional 60 min. Cells were imaged on an

Olympus IX70 inverted fluorescence microscope equipped with a Hamamatsu C8484-

03G01 digital camera and ASI MS2000 XY Piezo Z stage. Cells were magnified ×1000.

Fifty wide-field fluorescence images were collected per coverslip using Metamorph. All images in an experiment were pooled and randomized in a blinded fashion before being classified as having a tubular, punctate, or mixed mitochondrial phenotype. The images were decoded, and mitochondrial morphologies expressed as percent cells per mitochondrial phenotype.

Western Blotting of Electron Transport Chain Components

Western blot analyses were performed according to standard protocols (71).

Electron transport chain proteins were probed using the Total OXPHOS Rodent WB

Antibody Cocktail (Abcam, ab110413) at a 1:250 dilution, anti-actin mouse monoclonal antibody (MAB1501, Millipore) at 1:1500, and goat anti-mouse (HL) HRP-conjugated

IgG (AP308P, Millipore) at 1:1000. The blots were imaged on a Kodak Image Station

4000R using Carestream Molecular imaging software. The sum intensity of the dot blots were calculated using the Carestream software, and normalized to protein content.

Statistical Analysis

As indicated for the particular figure, the Z-test or Student's t test, as calculated by Excel, was used to assess statistical significance. The significance threshold set a priori to be p < 0.05.

3. Irgm1-Deficiency Causes Metabolic Alterations in MEFs and BMM

3.1 Introduction

In chapter 2 of this dissertation, I determined that lack of the GMS IRG Irgm1 caused a phenotypic shift in mitochondrial morphology towards a hyperfused state.

However, I was unable to determine the cause, or functional significance, of this phenotype in Irgm1-deficient MEFs. One avenue that was not explored in that chapter was measuring the metabolic flux in Irgm1-deficient cells.

As stated previously, mitochondrial hyperfusion is commonly seen in cells that are maximizing their ATP output through oxidative phosphorylation (305,306). Thus in this chapter I initially examined the bioenergetics of Irgm1-deficient MEFs after priming with IFN-γ.

My results showed that IFN-γ induced an increase in the glycolytic rate in Irgm1- deficient MEFs. Further analysis through metabolomics showed that our Irgm1- deficient MEFs had a decrease in overall TCA cycle intermediates, as well as decreased acylcarnitine levels and amino acids from the urea cycle. Although this change in metabolism was intriguing in MEFs, at this point I shifted the research into a more functionally relevant phenotype: bone marrow macrophages.

Immunometabolism has become something of a hot research topic over the past five years. In macrophages, this involves either activation to a proinflammatory (or M1) state or anti-inflammatory (or M2) state. Classical (or M1) macrophage activation, 74

induced by exposure to LPS with or without IFN-γ, results in a proinflammatory phenotype that, functionally, requires a very different metabolic profile than the anti- inflammatory phenotype of alternatively (or M2) activated macrophages(151,176,181).

Briefly, when stimulated with either LPS or IFN-γ/LPS, M1 activated macrophages are switched to a glycolytic metabolism, the TCA cycle is broken, and lipid synthesis occurs.

This dramatic shift is required for proper cytokine secretion, and other effector functions of pro-inflammatory macrophages (186,206,228). Priming macrophages with IFN-γ alone does not result in these metabolic shifts. Although IFN-γ priming does alter chromatin structure, acetylation, transcription efficiency and mTOR signaling, macrophages remain dependent on oxidative phosphorylation (39,307).

Thus it was extremely relevant to macrophage function when I observed hallmarks of M1 activation in Irgm1-deficient macrophages primed only with IFN-γ, including a metabolic shift towards glycolysis, a buildup of acylcarnitines and neutral lipids, and a broken TCA cycle. Furthermore, this metabolic change in Irgm1-deficient macrophages resulted in a mitochondrial morphological phenotype. However, Irgm1- deficient macrophages showed increased mitochondrial fragmentation after priming with IFN-γ or IFN-γ/LPS, in line with the observed glycolytic shift, instead of the elongated mitochondrial phenotype of Irgm1-deficient MEFs. My findings in this chapter are consistent with the hypothesis that Irgm1 is a negative regulator of IFN-γ

signaling (122). Furthermore my findings point toward a mechanism through which this occurs.

3.2 Results

3.2.1 Irgm1-Deficient MEFs Have a Different Metabolic Phenotype After Priming by IFN-γ

In the previous chapter I attempted to identify a change in mitochondrial function that would explain the fused mitochondrial morphology present in Irgm1- deficient MEFs primed with IFN-γ. Given that the mitochondrial fission machinery expression did not significantly change despite changes in membrane potential and ROS activity, I hypothesized that the mitochondrial phenotype seen may be driven by metabolic changes. I tested this hypothesis by studying the energy homeostasis and metabolomics in our Irgm1-deficient MEFs primed with IFN-γ.

3.2.2 Irgm1-Deficient MEFs Possess Bioenergetic Differences Compared to WT MEFs

To analyze the bioenergetics of our cell culture MEF model, I used the Seahorse glycolytic stress kit and mitochondrial stress kit to measure glycolysis and oxidative phosphorylation on the Seahorse bioscience extracellular flux analyzer. Oxidative phosphorylation was examined in MEFs by measuring their rates of oxygen consumption (OCR). A marked decrease in OCR, although not statistically significant,

Figure 8: Irgm1-deficient MEFs show an increase in glycolytic activity

WT and Irgm1-deficient MEFs were primed with IFN-γ or were maintained under control conditions for 24 hours. Cells were subjected to a Seahorse mitochondrial stress test and the oxygen consumption rate (OCR) was measured as a proxy for oxidative phosphorylation. Shown are (A) representative OCR measurement from a stress test and (B) the average basal OCR from 4 experiments. The cell’s glycolytic rate was measured by a Seahorse glycolytic stress test, using the measurement of the extracellular acidification rate (ECAR) as a proxy for glycolysis. Shown are (C) representative ECAR measurements from a stress test, and (D) the average basal ECAR from 4 experiments. Shown in (E) are the average ECAR/OCR ratios over 4 experiments. Values are means ± SEM. *P < 0.05

was observed in IFN-γ-primed, Irgm1-deficient MEFs 1 Figure 8 A, B). Given that the decrease in OCR suggests reduced ATP production, I hypothesized that Irgm1-deficient

MEFs might compensate by increasing glycolysis.

As expected, glycolytic activity, measured through the rate of extracellular acidification (ECAR), was significantly elevated in IFN-γ-stimulated Irgm1-deficient

MEFs compared to WT (Figure 8 C, D). In addition, I found a large increase in the ECAR to OCR ratio (Figure 8 E), indicating that IFN-γ-primed Irgm1-deficient MEFs rely more heavily on glycolysis as a source of energy for the cell.

3.2.3 Key Metabolite Levels are Altered in Irgm1-Deficient MEFs

After observing the bioenergetics differences in Irgm1-deficient MEFs, I consequently asked whether this might drive, or be driven by, other alterations in metabolism. We therefore performed metabolic profiling using a mass-spectrometry based approach, measuring amino acids, organic acids, and acylcarnitines (ACC) in IFN-

γ WT and Irgm1-deficient MEFs. As expected, we found changes in metabolite levels in all three areas (Figure 9; Appendix A).

The most striking difference was a significant decrease in the mitochondrial citric acid cycle intermediates in Irgm1-deficient MEFs (Figure 9A). These results correlate with the bioenergetic findings in the previous section, suggesting that the rate of oxidative phosphorylation is muted in Irgm1-deficient MEFs in the presence of IFN-γ. In contrast, the majority of amino acids concentrations fluctuated very little between WT and Irgm1-deficient cells. The exceptions (asparagine/aspartic acid, glutamate/glutamic acid, ornithine, citrulline and arginine), are noteworthy, as they are all decreased in IFN-

Figure 9: Altered metabolite profiles in WT and Irgm1-deficient IFN-γ-primed MEFs

Groups of 3 WT and 3 Irgm1-deficient MEFs isolated from separate mice were primed with IFN-γ for 24 hours. Lysates were prepared from the cells and were sent for metabolic profiling of organic acids, amino acids and acylcarnitine levels at the Duke Sarah W. Stedman Nutrition and Metabolism Center Mass Spectrometry Laboratory. (A) Measurements of the levels of organic acids in the TCA cycle, and a heatmap of their fold changes between Irgm1-deficient and WT MEFs. (B) Measurements of the levels of the amino acids in the urea cycle, and a heatmap of the amino acid fold change between Irgm1-deficient and WT MEFs. (C) Heatmap of the relative fold changes in acylcarnitine concentration between Irgm1-deficient and WT MEFs. Values are means ± SEM. *P < 0.05

γ-primed Irgm1-deficient MEFs (Figure 9B), and are all involved in the urea cycle, hinting that nitrogen cycling may be dysregulated in these cells. Finally, I also observed an overall decrease in long-chain acylcarnitines (LC-ACC) and very long-chain acylcarnitines (VLC-ACC), beginning with ACC-C12, in our IFN-γ primed, Irgm1- deficient MEFs (Figure 9C). Decreased acylcarnitine availability suggests either that these cells have an increased rate of β-oxidation, a decreased amount of fatty acids available for β-oxidation, or a decreased rate of lipid synthesis.

The metabolic and bioenergetic shifts fit with the tubular mitochondrial phenotype observed in our Irgm1-deficient MEFs. Hyperfused mitochondria are commonly seen in response to specific cellular stresses, such as starvation (271,272) or decreased electron transport chain activity (193,231,268,308). Under these conditions, increasing mitochondrial interconnectivity through fusion can inhibit mitochondrial membrane depolarization (309), as well as prevent cytochrome c release (268), and resistance to apoptosis (267,310,311). A hyperfused, tubular mitochondria network allows cells to escape autophagic degradation and provides them with an increased number of cristae to maintain ATP production (240,271,301,304,308).

3.2.4 – Mitochondrial Morphology and Metabolism in Irgm1-Deficient BMM

Emerging lines of research show that metabolic changes are critical for macrophage function. It is well established that classical macrophage (or M1) activation 80

involves a shift from energy production being driven primarily by oxidative phosphorylation and β-oxidation to one primarily driven by glycolysis (179,312).

Moreover, this glycolytic shift is required for robust induction of the proinflammatory effector functions that characterize classical macrophage activation, including production of iNOs (313,314), and proinflammatory cytokine secretion (40,176,177,315).

Figure 10: Increased mitochondrial fragmentation observed in Irgm1-deficient BMM

(A) Representative images of the mitochondrial morphologies observed in a population of bone marrow macrophages, stained with the mitochondrial outer membrane protein TOM20 to label mitochondria, and then imaged. (B) Quantified punctate, elongated, or mixed mitochondrial morphologies in WT and Irgm1-deficient BMMs primed with either IFN-γ, alone, for 24 hours, or primed with IFN-γ for 24 hours and stimulated with 100nM LPS for the last 16 hours. (C) Western blot of MFN2 WT or Irgm1-deficient BMM, stimulated as indicated, and quantification of the band intensities. of three experiments. Each experiment, represents the averages of 3 or more experiments. Values are means ± SEM. *P < 0.05; **P < 0.01

I reasoned that, given the shifts in mitochondrial dynamics and metabolism observed in our Irgm1-deficient MEFs, similar changes may be occurring in the immune cells of our Irgm1-deficient mice. Consequently, I shifted our research focus to our more functionally relevant bone marrow macrophage (BMM) cell culture model and undertook mitochondrial and metabolic analyses of M1 activated Irgm1-deficient macrophages.

3.2.5 IFN-γ Induces Mitochondrial Fission in Irgm1-Deficient Macrophages

My previous results suggested that the altered mitochondrial morphology observed in Irgm1-deficient MEFs is a consequence of their altered metabolism, thus, I started my studies by addressing whether mitochondrial morphology was similarly altered in Irgm1-deficient macrophages.

To begin, we examined the mitochondrial morphology of WT and Irgm1- deficient bone marrow macrophages (BMM) under “control” (M0) conditions, after priming with IFN-γ for 24 hours, after stimulating the cells with LPS, or after stimulating the cells with IFN-γ and LPS. My results showed that, while Irgm1- deficiency altered mitochondrial morphology in macrophages, the effect was surprisingly opposite of that seen in fibroblasts. Irgm1-deficient BMM primed with IFN-

γ displayed a much more punctate mitochondrial morphology than that seen in WT

BMM (Figure 10 A-C). In addition, mitochondrial fragmentation was only observed in 82

Irgm1-deficient macrophages primed with IFN-γ, with or without LPS, but not LPS alone, suggesting that induction of Irgm1 negatively regulates mitochondrial fragmentation even in the absence of a bacterial challenge.

IFN-γ priming also induced a decrease of the mitochondrial fusion factor Mfn2 in our Irgm1-deficient BMM, strengthening the assertion that these cells are in a pro- fission state (Figure 10 C). Further, it was recently shown that MFN2 is phosphorylated by JNK and subsequently degraded in response to cellular stress, leading to mitochondrial fragmentation and enhanced apoptotic cell death (316).

Figure 11: Timecourse quantification of IFN-γ-mediated mitochondrial fragmentation in Irgm1-deficient BMM

Mitochondrial morphology of WT and Irgm1-deficient BMM primed with IFN-γ were assessed over a 24 hour timecourse. As above, the cells were primed with IFN-γ for the indicated time period, stained with the mitochondrial outer membrane protein TOM20 to label mitochondria, and then imaged, 50 cells per condition. The cells with punctate, elongated, or mixed mitochondria were quantified. Shown is the average of 3 separate studies, ± SEM, * p<0.05

To determine if the mitochondrial fragmentation in Irgm1-deficient cells was occurring simply because of the lack of IFN-γ-induced Irgm1 expression in WT BMM, I examined the accumulation of Irgm1-deficient BMM with punctate mitochondria over a

24 hour IFN-γ timecourse (Figure 11). In RAW 264.7 macrophages, Irgm1 mRNA was detectable at 2 hours after the induction of IFN-γ, reached a maximum at 4 hours and remains detectable for at least 24 hours, while 8 hours of LPS stimulation were required to detect Irgm1 transcripts (70,317). Our results indicate that mitochondrial fragmentation begins between 8 and 16 hours after IFN-γ induction in Irgm1-deficient cells. Thus mitochondrial dysfunction does not rapidly occur in the absence of Irgm1, suggesting that mitochondrial fragmentation either occurs due to an IFN-γ induced event between 8-16 hours that requires Irgm1, or that the deficiencies caused by the absence of Irgm1 build-up gradually and are only severe enough to cause mitochondrial dysregulation after 8 hours.

Finally, I verified that this change in mitochondrial morphology was likely independent of any modulation of autophagy in Irgm1-deficient BMM, as no increase in mitochondrial fragmentation occurred in Atg7-deficient BMM after IFN-γ priming

(Figure 12).

Figure 12: Mitochondrial fragmentation remains unchanged in ATG7-deficient mice

Representative images of the mitochondrial morphologies observed in WT, Irgm1- deficient of ATG7-deficient BMM primed with IFN-γ for 24 hours, or IFN-γ/LPS for 24 hours, stained with the mitochondrial outer membrane protein TOM20 to label mitochondria, and then imaged. Quantified punctate, elongated, or mixed mitochondrial morphologies in WT and Irgm1-deficient MEFs exposed to IFN-γ for 24 hours.

3.2.6 Altered Bioenergetics in Irgm1-Deficient Macrophages

The alternation of mitochondrial morphology in Irgm1-deficident IFN-γ-primed

BMM, supported my theory that, similar to our MEF model, metabolic changes may be occurring in Irgm1-deficient macrophages. Indeed, glycolytic activity, as measured by the extracellular acidification rate (ECAR), was markedly elevated in IFN-γ-stimulated

Irgm1-deficient BMM compared to IFN-γ-stimulated WT BMM (Figure 13 A-B). There was also a small increase in ECAR in unstimulated Irgm1-deficient BMM compared to unstimulated WT cells. This may be due to absence of the small levels of Irgm1 that is expressed in WT macrophages in the absence of IFN-γ stimulation (unpublished data).

Figure 13: Irgm1-deficient macrophages show an increase in glycolytic activity

BMM from WT and Irgm1-deficient mice were primed with IFN-γ or were maintained under control conditions for 24 h. The cells were then subjected to a glycolytic stress test

and measurement of the extracellular acidification rate (ECAR) as a proxy for glycolysis. The experiment was repeated 4 times, using cells isolated from different mice each time. Shown are (A) representative ECAR measurements from a stress test, and (B) representative ECAR measurement. The cells were also used for (C) metabolomic measurement of lactate levels, analyzing cells isolated from 3 separate mice per genotype. Cells were subjected to a mitochondrial stress test and measurement of the oxygen consumption rate (OCR) as a proxy for oxidative phosphorylation. The experiment was repeated 4 times, using cells isolated from different mice each time. Shown are (D) the average OCR from a representative experiment. Shown in (E) are the average ECAR/OCR ratios over 4 experiments. In other experiments, BMM from WT and Irgm1-deficient mice were maintained under control conditions, were primed with IFN-γ or were activated with IFN- γ for 24 h and LPS for the final 16 h. Protein lysates were isolated and used for immunoblotting with an antibody cocktail of electron transport chain (ETC) components. Shown are (F) a representative blot and (G) sum intensities of the bands normalized to actin and averaged over 3 experiments. Error bars indicate SEM. *p<0.05

These results were corroborated by lactate measurements taken in our metabolomic studies (described in the next section), in which concentrations were 36% higher in IFN-

γ-primed Irgm1-deficient BMM compared to IFN-γ-stimulated WT cells (Figure 13 C), while this difference was not present in BMM stimulated with IFN-γ and LPS

(Appendix B).

To determine whether the increased glycolytic activity in Irgm1-deficient macrophages corresponded with decreased oxidative phosphorylation, I subsequently measured the oxygen consumption rate (OCR) in macrophages to gain insight into the capacity for mitochondrial oxidative phosphorylation (OXPHOS). In IFN-γ-stimulated

Irgm1-deficient BMM, relative to IFN-γ-stimulated WT BMM, I observed a trend toward decreased maximal respiration rates, as well as a substantial increase in the ECAR to

OCR ratio (Figure 13 D-F), indicating that IFN-γ-primed Irgm1-deficient BMM are relatively more reliant on glycolysis to meet the energy demands of the cell. I also observed a significant decrease in several essential electron transport chain proteins in our IFN-γ-primed Irgm1 deficient macrophages, reflecting the decrease in OXPHOS activity (Figure 13 G). Taken together, these results suggest that IFN-γ-primed, Irgm1- deficient BMM demonstrate a marked shift toward a glycolytic phenotype. Such a shift toward glycolysis is generally seen in WT macrophages after full activation with LPS but not IFN-γ alone (39,315,318).

3.2.7 Metabolomic Analysis of Irgm1-Deficient Macrophages Reveals Changes in Key Metabolites

I further explored metabolic changes by measuring levels of key metabolites using a mass-spectrometry based metabolic profiling approach (Table S2), and detected profound metabolic changes in Irgm1-deficient BMM.

3.2.7.1 The Altered Metabolite Concentrations in Irgm1-Deficient Macrophages Primed with IFN-γ Suggests Activation of Pro-inflammatory Pathways

The changes observed in organic and amino acid metabolite levels in IFN-γ primed Irgm1-deficient BMM indicate a shift in the function of the citric acid cycle and urea cycle. A 50% drop in citrate, compared to WT BMM, was observed, along with increased accumulation of the downstream metabolites, succinate, fumarate, malate, and

α-ketoglutarate (Figure 14A), suggest a break in the citric acid cycle, reminiscent of the one seen in fully M1 activated macrophages (142,176,319). These results correlate with the glycolytic switch, and decreased oxidative phosphorylation, observed in these cells

(Figure 3.6). Altered function of the urea cycle, and perhaps dysfunction in the crosstalk between this metabolic pathway and the citric acid cycle, is suggested by the 700% increase in citrulline (Figure 14 B), and 40% decrease in aspartate (Figure 14 C). No nitric oxide production was detected in IFN-γ primed Irgm1-deficient BMM (Figure 14 D), supporting the theory that the buildup of citrulline is not a byproduct of nitric oxide production (320-322), but is instead caused by metabolic dysfunction.

However, the most striking metabolic phenotype observed in Irgm1-deficient

BMM primed with IFN-γ was the marked increase in most long-chain acylcarnitines

(LC-ACC) with individual LC-ACC levels increased by as much as 15.7 fold (Figure 14

E) relative to WT BMM. In contrast, short chain acylcarnitines were relatively decreased in IFN-γ-stimulated Irgm1-deficient BMM.

Taken together, the metabolite analysis suggests metabolic changes are occurring in these cells that are characteristic of M1 marcophage activation. The accumulation of

LC-ACC,, usually observed only in LPS-stimulated BMM, with or without IFN-γ

(206,208,323-325), is known to have pro-inflammatory properties, while the buildup of succinate (176,177) and fumarate (193,219,223,224,326), are known to engage proinflammatory pathways elicited in M1 macrophages.

Figure 14: Altered metabolite profiles in WT and Irgm1-deficient IFN-γ-primed BMM

Groups of 3 WT and 3 Irgm1-deficient BMM isolated from separate mice were maintained under control conditions, or primed with IFN-γ for 24 hours. Lysates were prepared from the cells and were sent for metabolic profiling of organic acids, amino acids and acylcarnitine levels at the Duke Sarah W. Stedman Nutrition and Metabolism Center Mass Spectrometry Laboratory. (A) Measurements of the levels of organic acids in the TCA cycle (B) Measurements of the levels of the amino acids in the urea cycle. (C) Nitrite measurments in WT and Irgm1-deficient BMM under control conditions, IFN-γ- primed or IFN-γ/LPS stimulated BMM. (D) Heatmap of the relative fold changes in acylcarnitine concentrations between Irgm1-deficient and WT BMM primed with IFN-γ. Values are means ± SEM. *P < 0.05 90

3.2.7.2 Irgm1-Deficient Macrophages Stimulated with IFN-γ and LPS Suggests Metabolite Depletion

Although the same shifts in metabolite levels, relative to control conditions, are observed in both cell types, an overarching reduction in metabolite levels is observed in

Irgm1-deficient macrophages. Indeed, 85% of all ACCs, and 60% of organic and amino acids showed a decrease of 20% or greater in Irgm1-deficient BMM stimulated with IFN-

γ and LPS, compared to WT cells. More specifically, under IFN-γ and LPS stimulation, citric acid cycle intermediates (Figure 15 A), excluding succinate, and many urea cycle intermediates (Figure 15 B) (asparagine/ aspartic acid, glutamate/ glutamic acid, ornithine and citrulline) were decreased compared to WT. Similarly, a widespread decrease in ACC is observed, especially MC-LCC and SC-LCC, where over 75% of species are decreased by 50% or more (Figure 15 C). All these shifts suggest that Irgm1- deficient BMM stimulated with LPS and IFN-γ have an overall depletion of metabolites, and a distinct metabolic phenotype.

3.2.8 The Absence of Irgm1 Induces Neutral Lipid Accumulation in IFN-γ-primed BMM

The finding that IFN-γ-primed Irgm1-deficient cells contained striking increases in LC-ACC prompted me to determine whether this correlated with general lipid increase in these cells. Neutral lipid content can be increased by lipid import or by lipid biosynthesis. As both methods are upregulated during M1 macrophage activation

Figure 15: Altered metabolite profiles in IFN-γ/LPS stimulated WT and Irgm1- deficient BMM

Groups of 3 WT and 3 Irgm1-deficient BMM isolated from separate mice were maintained under control conditions, or activated with IFN-γ for 24 h and LPS for the final 16 h. Lysates were prepared from the cells and were sent for metabolic profiling of organic acids, amino acids and acylcarnitine levels at the Duke Sarah W. Stedman Nutrition and Metabolism Center Mass Spectrometry Laboratory. (A) Measurements of the levels of organic acids in the TCA cycle. (B) Measurements of the levels of the amino acids in the urea cycle. (C) Heatmap of the relative fold changes in acylcarnitine concentration between Irgm1-deficient and WT BMM after stimulation with IFN-γ/LPS. Values are means ± SEM. *P < 0.05 92

(142,206,327), BMM neutral lipid content was measured under two different conditions: 1) in medium with BSA, to measure biosynthesized lipid, and 2) with 2.5mM oleate/palmitate mixture added to the medium to measure lipid import.

As expected, a large neutral lipid increase was observed in Irgm1-deficient BMM primed with IFN-γ, compared to WT, in the lipid-rich media. More surprising was the similar increase of neutral lipid content observed in Irgm1-deficient cells in the absence of extracellular lipids (Figure 16), suggesting de-novo lipid synthesis. Indeed, the

Figure 16: Neutral lipid accumulation is increased in IFN-γ-primed, or IFN-γ/LPS stimulated macrophages lacking Irgm1

WT and Irgm1-deficient BMM were grown in control medium with BSA, or in medium with added oleate/palmitate, and primed with IFN-γ for 24 hours, or IFN-γ/LPS for 24 hours. The cells were stained with the neutral lipid stain AdipoRed, and fluorescence measured. Shown is the average of 3 separate studies, ± standard error, * p<0.05

intracellular neutral-lipid content in Irgm1-deficient cells does not significantly change between IFN-γ and IFN-γ/LPS treated Irgm1-deficient cells in BSA medium. Nor is 93

there a statistically significant difference between IFN-γ-primed Irgm1-deficient BMM cultured in either BSA medium or lipid rich medium, although lipid content does increase by 60%, indicating that a substantial portion of the neutral lipid increase observed in Irgm1-deficient BMM is formed by de-novo synthesis, even in the presence of extracellular lipids.

Only when BMM are stimulated with both IFN-γ and LPS in lipid-rich medium is a substantial neutral lipid increase observed in both WT and Irgm1-deficient BMM, which is not present in cells cultured in BSA medium, likely representing extracellular lipid import.

3.2.9 Inhibiting Fatty Acid Synthesis or ROS Damage Mutes the Fragmented Mitochondrial Phenotype Seen in IFN-γ-Primed Irgm1- Deficient BMM

Mitochondrial fragmentation is known to occur after stress or damage, often leading to apoptosis (267,311). In macrophages, it is known that M1 activation with LPS increases mitochondrial fragmentation (109,328,329), with the fragmentation resulting in part from the high production of reactive oxygen species that occurs with activation

(329-332). I reasoned that the fragmented mitochondrial morphology in Irgm1-deficient macrophages may be a result of the metabolic changes in those cells in the context of high ROS levels. Previous work supported this hypothesis. As demonstrated in chapter

2, Irgm1-deficient fibroblasts exposed to H2O2 displayed a reversion of the

mitochondrial phenotype in those cells; that is, they no longer showed a more fused mitochondrial phenotype compared to WT fibroblasts, but a relatively more punctate phenotype (Figure 7).

Figure 17: The punctate mitochondrial morphology is increased in IFN-γ-primed macrophages lacking Irgm1, and is dependent on fatty acid synthesis and reactive oxygen species

BMM of the indicated genotype were plated on coverslips. Following various treatments, the cells were immunostained with antibodies to the mitochondrial marker TOM20. The mitochondrial morphology of cells was assessed in a blinded fashion in 50 cells per genotype per experiment. Shown are the average percent of cells with an overall punctate mitochondrial morphology. In (A) the cells were maintained under control conditions, or were primed with IFN-γ for 24 h with or without the fatty acid synthesis inhibitors, cerulenin or C75. In (B) the cells were maintained under control conditions, or primed with IFN-γ for 24 h with or without the reactive oxygen species quencher NAC. For each figure, shown are average values from experiments performed at least 3 times. Error bars indicate SEM. *p<0.05. **p<0.001, ***p<0.0001

To test this possibility, I examined the mitochondrial morphology of cells exposed to the fatty acid synthase (FAS) inhibitors, cerulenin and C75. Both fatty acid synthase inhibitors markedly decreased the punctate character of the mitochondria in

IFN-γ-primed, Irgm1-deficient BMM (Figure 3.10A). Further, Irgm1-deficient cells treated with the ROS quencher, N-acetyl cysteine (NAC), also displayed reduced mitochondrial fragmentation (Figure 17 B). Taken together, these studies suggest that mitochondrial fragmentation in Irgm1-deficient macrophages is, in fact, not a primary consequence of Irgm1-deficiency in those cells, but rather a downstream result driven by diverse metabolic changes in these cells.

3.3 Discussion

Specific metabolic changes are required for the effector functions of many immune cells. Up regulation of aerobic glycolysis often provides a source of biosynthetic materials in inflammatory and/or rapidly proliferating immune cells, while non- inflammatory immune cells that typically have longer life spans commonly rely on fatty acid oxidation and oxidative metabolism for energy production (141). This is true of macrophages in which activation to the M1 inflammatory state following exposure to

LPS (alone or in combination with IFN-γ) is driven by a shift from energy metabolism dominated by oxidative phosphorylation and β-oxidation, to one dominated by aerobic glycolysis (333).

Despite the substantial transcriptional and epigenetic-driven changes in gene expression that are known to be induced by IFN-γ in macrophages (25,39,41,334-336),

IFN-γ-primed WT macrophages do not display the dramatic shifts in energy

metabolism, nor the accumulation of long chain fatty acids seen in LPS-activated macrophages. This is in keeping with previously published data showing that IFN-γ induces translational modifications that, in fact, up-regulate pathways associated with oxidative phosphorylation and mitochondrial function while suppressing translational efficiency in pathways associated with biosynthetic activity (39,336). Given this context, it is striking that Irgm1 deficiency induces increased glycolysis in macrophages that have only been primed with IFN-γ, which in effect, partially mimics the metabolic shifts occurring in WT cells fully activated with LPS.

The changes in metabolite levels in IFN-γ primed Irgm1-deficient macrophages also mimic the metabolic rewiring during M1 activation. As described in the introduction, the increased glycolytic flux observed in M1 macrophages occurs alongside a broken TCA cycle, which is needed to support the production of critical M1 cellular products, including acetyl CoA, succinate, itaconate and nitric oxide (193). The metabolite signature observed in IFN-γ-primed Irgm1-deficient macrophages fits within the context of this broken TCA cycle. Citrate levels are decreased compared to WT cells, consistent with the first break in the TCA, while lipid content is increased, suggesting citrate is being funneled into lipid biosynthesis (142). Further, the drop in citrate also supports the production of itaconate, inhibiting succinate dehydrogenase, thought to cause the second break in in the TCA cycle (219). The second break of the TCA cycle causes a build-up of succinate – just like the accumulation of succinate observed in our

IFN-γ Irgm1-deficient macrophages (219,221,337). Further, in M1 macrophages, evidence suggests the conversion of succinate to fumarate is inhibited, and the remaining portion of the TCA cycle is fueled through the aspartate-arginosuccinate shunt, a set of transformations connecting the TCA cycle with the urea cycle (224), with active ornithine-to-citrulline conversion (159). Given the metabolites associated with the urea cycle are altered in both Irgm1-deficient macrophages, stimulated with IFN-γ alone or IFN-γ/LPS, as well as INF-γ induced Irgm1 deficient MEFs, aberrant activation of the aspartate-arginosuccinate shunt is a possible mechanism through which this occurs.

Mitochondrial dynamics have also generally been associated with different metabolic states: Cells relying on fatty acid oxidation or oxidative phosphorylation often display fused networks of mitochondria, while those relying on glycolysis tend to have more punctate mitochondria (275,298,338-340). Further, promoting mitochondrial fission and/or blocking fusion can stimulate pro-inflammatory cytokine production

(198,329,330,341). Nevertheless, my results suggest that the punctate mitochondrial morphology seen in Irgm1-deficient cells is secondary to accumulation of LC-ACC and neutral lipids seen in those cells, as treating of the cells with fatty acid synthase inhibitors to reduce fatty acid concentrations eliminated the punctate mitochondrial phenotype.

In WT BMM, the build-up of LC-ACC is only observed after LPS stimulation.

Moreover, the accumulation of neutral lipids is observed only in the presence of

extracellular lipids, implying that, unlike in Irgm1-deficient BMM, the majority of the increased neutral lipid content in WT IFN-γ/LPS stimulated BMM is derived from lipid import. In contrast, Irgm1-deficient BMM primed with IFN-γ show a dramatic increase in LC-ACC and a 3X increase neutral lipid content in the presence of extracellular lipids.

However, with IFN-γ /LPS stimulation, the overall neutral lipid content doubles, once again implying an increase in lipid import, but LC-ACC levels remain similar to those in cells incubated with IFN-γ alone. In addition, extracellular lipids are not required for the increase in neutral lipid accumulation in Irgm1-deficient cells – suggesting this increase is driven by de novo lipid biosynthesis. The accumulation of neutral lipids observed in M1 macrophages most likely correlates to an increase in the number and/or size of lipid droplets – an organelle consisting of a core of neutral lipids and steryl esters, surrounded by a phospholipid monolayer and a heterogenous group of proteins, comprising a key mechanism of cellular energy storage in the Irgm1-deficient cells

(342,343). Given the high level of LC-ACC in IFN-γ stimulated cells, it is likely that the concentration of free fatty acids in the cytoplasm is also increased. Elevated levels of long chain fatty acids, like palmitate, is sufficient to induce an M1-like phenotype in macrophages, and has even been reported to induce NF-κB and JNK signaling pathways, known to be active both during IFN-γ priming and LPS stimulation (344-346).

This also might explain the mitochondrial fragmentation seen in these cells, as high free fatty acid levels induces lipid toxicity, leading to mitochondrial dysfunction and

mitochondrial fragmentation (308,347,348). Long chain saturated fatty acids are also known to induce mitochondrial depolarization - the increased length of the hydrocarbon chain in the fatty acids, along with the absence of double bonds, increases their capacity to depolarize mitochondria (349) – by inducing production of the uncoupling protein

UCP2 (208). These data indicate that there may be impaired or dysfunctional lipid flux in Irgm1-deficient cells (185,350), and that is likely a large component in driving the observed M1 phenotype. This notion is contrary to previous studies that have suggested a direct role for Irgm1 in both associating with the mitochondria and controlling their dynamics (74,75,81,96,120).

Dysregulation of lipid flux may also be implicated in the mitochondrial hyperfusion seen in Irgm1-deficient MEFs. Like Irgm1-deficient BMM stimulated with

IFN-γ/LPS, I observed a dearth of overall acylcarnitines levels. Irgm1-deficient MEFs likely also have increased neutral lipid content, although this has not yet been shown.

Mitochondrial hyperfusion, as described previously is often seen in cells during starvation, to avoid autophagic degradation (308) and to maximize ATP production

(253), but also as a phenotype in cells with high rates of β-oxidation under starvation conditions. Cells undergoing β-oxidation can use lipid droplets as a conduit for supplying mitochondrial fatty acids for β-oxidation (338,348). Further, a study in MEFs showed that the transfer of fatty acid from lipid droplets required the mitochondria to be in a highly tubulated network, enabling lipids to homogeneously distribute inside

100

mitochondria after import, ensuring β-oxidation and that downstream oxidative phosphorylation reactions were optimized (348). Given the increase in glycolytic rate in

IFN-γ-primed Irgm1-deficient MEFs, it is unlikely that increased β-oxidation is occurring, however, if IFN-γ is causing dysregulation of lipid flux, their hyperfused mitochondrial state may be an attempt to optimize dysregulated lipid import into the mitochondria.

While my data does not in any way identify the primary and/or initial driver of this phenotype in Irgm1-deficient cells, it is plausible that Irgm1 could play a direct role in lipid homeostasis. It is a dynamin-like GTPase that associates with specific intracellular membrane compartments, including the Golgi, where it may alter membrane trafficking (73-75,81,82,97,351). Additionally, IRGM proteins have been reported to play roles in lipid droplet maintenance (74,96,352). However, more data is required to either confirm or rule out this hypothesis.

Alternatively, the underlying impetus for the metabolic changes seen in Irgm1- deficient macrophages could be the alterations in autophagy that my lab and others have documented as occurring with Irgm1-deficiency. That work has suggested an impairment in autophagic flux in Irgm1-deficient macrophages (96,97,108), fibroblasts

(75), and intestinal enterocytes (116). Broadly speaking, autophagy is known to promote oxidative respiration, and conversely, its inhibition leads to increased glycolysis

(141,176). Nevertheless, our results demonstrate that lack of Atg7 did not affect

101

mitochondrial morphology, in contrast to the strong promotion of mitochondrial fission that I saw in Irgm1-deficient macrophages.

It still remains possible that signaling upstream of autophagy may still drive these changes – for instance, perturbation of mTOR and altered activation of mTOR- regulated pathways that are distinct from autophagy (353). While we currently have no evidence for this, IFN-γ-priming of macrophages has been reported to inhibit mTOR activity (39,354), while in contrast, activation with LPS exposure activates mTOR

(353,355). Indeed, activation of mTOR is a key mechanism through which M1 inflammatory macrophages meet the high biosynthetic activity required for a sustained inflammatory response (39,177,353).

In sum, the identified metabolic alterations likely contribute to the enhanced inflammatory responses that have been observed in Irgm1-deficient mice in the intestine

(116) and in response to bacterial infections (91,114,121). The novel role for Irgm1, and perhaps other IRG proteins, in modulating fatty acid accumulation and metabolism establishes a new direction for research to unravel the roles of IRG proteins in the regulation of inflammation. Further, the M1-like activation phenotype in IFN-γ primed

Irgm1-deficient macrophages strongly supports the idea that Irgm1 plays a critical role as a negative regulator of IFN-γ signaling.

102

3.4 Methods

Mice

Irgm1-deficient (88) and Atg7f/fLysM-Cre mice (356) have been described previously. All mice were housed and maintained under procedures approved by the

Institutional Animal Care and Use Committees at the Duke University and Durham VA

Medical Centers.

Cell Culture

Primary murine bone marrow derived macrophages (BMM) were isolated from the tibia and femurs of 2- to 4-month-old mice and cultured according to standard procedures described previously (91). The bone marrow was flushed from the bones using a 27G needle fitted to a syringe filled with DMEM (Life Technologies); the marrow was dispersed by drawing through the needle three to four times; and red cells were lysed with ACK lysing buffer (Life Technologies). Adherent cells were cultured for

6 days in BMM medium [DMEM supplemented with 10% (v/v) FBS (Hyclone) and 30%

(v/v) L929 cell-conditioned medium]. The cells were cultured on Petri dishes that were not cell culture-treated, resulting in cultures that were loosely adherent and easily removed from the plates with cell dissociation buffer (13150-016, Gibco/Thermo Fisher).

Twenty-four hours prior to all experiments, the cells were placed in medium lacking

L929-conditioned media [DMEM (11995, Gibco ThermoFisher), supplemented with 10%

103

(v/v) FBS and 100 units/ml penicillin/100 µg/ml streptomycin (15140, GIBCO

ThermoFisher)] on coverslips or cell culture plates. All primary cells and cell lines were grown and maintained in a humidified atmosphere of 5% CO2 at 37°C. Where appropriate, 0.2 μg/ml interferon (IFN)-γ (IF005, EMD Millipore) and/or 100ng/ml LPS

(L2387, Sigma) was included in the growth medium.

Primary mouse embryonic fibroblasts were isolated and cultured according to standard procedures (357,358) in media containing DMEM supplemented with 10% (v/v) FBS. The cells were used for experiments between passages 2 and 5.

Mitochondrial Morphology Assay

Cells were fixed on coverslips with 4% paraformaldehyde (w/v) in PBS for 15 min and permeabilized with 0.2% (w/v) saponin in PBS for 10 min. The cells were then stained for 60 min with anti-TOM20 rabbit polyclonal antibody (FL-145, Santa Cruz

Biotechnologies) at a 1:500 dilution, followed by Alexa Fluor conjugated secondary antibody (Molecular Probes/Invitrogen) at a 1:750 dilution for an additional 60 min.

Cells were imaged on an Olympus IX70 inverted fluorescence microscope equipped with a Hamamatsu C8484-03G01 digital camera and ASI MS2000 XY Piezo Z stage. Cells were magnified ×1000. Fifty wide-field fluorescence images were collected per coverslip using Metamorph. All images in an experiment were pooled and randomized in a blinded fashion before being classified as having a tubular, punctate, or mixed

104

mitochondrial phenotype. The images were decoded, and mitochondrial morphologies expressed as percent cells per mitochondrial phenotype.

Western Blotting of Electron Transport Chain Components

Western blot analyses were performed according to standard protocols (97).

Electron transport chain proteins were stained using the Total OXPHOS Rodent WB

Antibody Cocktail, (Abcam, ab110413) at a 1:250 dilution, anti-mitofusin 2 mouse monoclonal antibody at a 1:500 dilution, anti-actin mouse monoclonal antibody

(MAB1501, Millipore) at 1:1500, and goat anti-mouse (HL) HRP-conjugated IgG

(AP308P, Millipore) at 1:1000. The blots were imaged on a Kodak Image Station 4000R using Carestream Molecular imaging software. The sum intensity of the dot blots were calculated using the Carestream software and normalized to protein content.

Metabolic Seahorse Assays

OCR (oxygen consumption rate) and ECAR (extracellular acidification rate) were measured with a XF24 extracellular flux analyzer (Seahorse Bioscience) using kits and protocols provided by the manufacturer. 1x105 cells/well were plated in the Seahorse 24- well plate and incubated with 0.2 ng/ml interferon (IFN)-γ for 24 hours. Cells were washed and incubated in 600μL of Seahorse media in a CO2-free incubator for 30 minutes prior to the assay. The OCR was then measured over time, following injection

105

of 1μM oligomycin, 3μM FCCP, and 100nM rotenone/antimycinA, as described in the

Seahorse Mitochondrial Stress Test Kit (103015-100, Seahorse XF). The OCR was measured without glutamine or glucose added to the media. The ECAR was measured following injection of 10 mM D-glucose, 1 μM oligomycin, and 20 mM 2DG as described in the Seahorse Glycolytic Stress Test Kit (103020-100, Seahorse XF). The ECAR was also measured without glucose or glutamine added to the media. ECAR and OCR measurements were normalized to protein content.

Metabolite Measurements

Primary murine BMM or fibroblasts were cultured in triplicate on 100mm cell culture plates, at a density of 3.4 x 106 cells/plate. The media [DMEM supplemented with

10% (v/v) FBS] was supplemented with 0.5mM L-carnitine, and 100μM of a 1:1 oleate/palmitate stock complexed to 0.14% (w/v) BSA. The cells were maintained under those conditions, or additionally exposed to ng/ml IFN-γ and/or 100ng/mL LPS for 16 hours. The cells were then scraped and collected in 0.3mL of 0.6% (v/v) formic acid. An equal volume of acetonitrile was then added to the lysates, which were then stored at -

80C. Targeted mass spectrometry-based metabolic profiling was performed at the Duke

Sarah W. Stedman Nutrition and Metabolism Center Mass Spectrometry Laboratory, as previously described (359-361). Free carnitine, acylcarnitines, and amino acids levels from macrophage samples were measured by direct-injection electrospray tandem mass

106

spectrometry (MS/MS), using a Micromass Quattro Micro LC-MS system (Waters-

Micromass, Milford, MA) equipped with a model HTS-PAL 2777 autosampler (Leap

Technologies, Carrboro, NC), a model 1525 HPLC solvent delivery system (Agilent

Technologies, Palo Alto, CA), and a data system running MassLynx 4.0 software

(Waters, Milford, MA)(359,360). Organic acids were quantified using methods described previously employing Trace Ultra GC coupled to a Trace DSQ MS operating under

Excalibur 1.4 (Thermo Fisher Scientific, Austin, TX)(361). Metabolite data was normalized to the total protein content in each sample, as determined by Pierce BCA

Protein Assay Kit (23225, Pierce Thermo Fisher Scientific).

Statistical Analysis

As indicated for the particular figure, the Z-test or Student's t test, as calculated by Excel, was used to assess statistical significance. The significance threshold set a priori to be p < 0.05.

107

4. Metabolic alterations contribute to enhanced inflammatory cytokine production in Irgm1-deficient macrophages

4.1 Introduction

In chapter 3 of this dissertation, I established that loss of Irgm1 induced a shift towards a glycolytic metabolism, increased accumulation of neutral lipids and long- chain acylcarnitines, and marked changes in metabolic pathways in IFN-γ primed BMM.

Conversely, Irgm1-deficient BMM stimulated with both IFN-γ and LPS demonstrated an overall decrease in metabolites compared to WT. Discouragingly, other than demonstrating that the elevated mitochondrial fragmentation observed in Irgm1- deficient BMM primed with IFN-γ can be muted by inhibiting lipid synthesis, my research until this point has yet to yield a functional output caused by the observed metabolic changes. Yet the shifts observed in immunometabolism in the absence of

Irgm1 lead me to re-interpret the increased pathogen susceptibility observed in our

Irgm1-deficient mice. Previous work has attributed the bacterial susceptibility in Irgm1- deficient mice to defective processing of bacteria-containing phagosomes in macrophages and other cells (76,81,96,108,109). That reduced capacity to manifest IFN-γ- induced killing of S. typhimurium and M. tuberculosis has been further linked to altered autophagic function in Irgm1-deficient macrophages (75,96,108,120,362-364). Impaired autophagy ostensibly leads to a reduced capacity to restrict bacterial growth, increased inflammation, and ultimately death of the host, though these linkages have not been

108

formally established. In this chapter, I address an alternative hypothesis – that a major contributor to increased inflammation and lethality in Irgm1-deficient mice is increased inflammatory cytokine production that directly results from cell intrinsic alterations in

Irgm1-deficient macrophages and other cells.

Activation of macrophages leading to production of inflammatory cytokines is a well-studied process. M1 (or classical) activation of macrophages results from their priming with IFN-γ predominantly derived from by NK and Th1 cells, followed by their activation with exposure to bacterial products such as lipopolysaccharide (LPS)

(142,143,153,365). While IFN-γ-priming increases expression of some cytokines, full activation of macrophages with LPS is needed to trigger robust production of the array of inflammatory cytokines characteristically secreted by classically-activated macrophages (25,318,366). Along those lines, when perturbations occur in processes linked to energy metabolism, including autophagy, production of inflammatory cytokines is markedly affected (141,176).

Indeed, the many metabolic perturbations that were observed in Irgm1-deficient

BMM in chapter 3, including increased neutral lipid content (185,202,367), increased long-chain acylcatnitine accumulation (324,325,368), decreased citrate content (219,337), increase succinate content (177,226), and an increase in the glycolytic rate (169,314,369), are critical for proinflammatory cytokine production. Therefore, it was not surprising to uncover increased secretion of the cytokines MCP-1 and RANTES in IFN-γ primed

109

Irgm1-deficient macrophages, nor that the secretion of those cytokines could be muted by inhibiting fatty-acid synthesis, glycolysis or reactive oxygen species. Likewise, increased cytokine secretion of IL-1β and TNFα in Irgm1-deficient cells induced with both IFN-γ and LPS is consistent with the metabolic phenotypes observed in chapter 3.

These results, along with the increased auto-inflammation in Irgm1-deficient mice, support the hypothesis that the increased lethality observed in Irgm1-deficient mice after infection stems not from unrestricted pathogen grown, but from cell intrinsic properties of Irgm1-deficient immune cells.

4.2 Results

4.2.1 Secretion of Cytokines RANTES and MCP-1 are Elevated in Irgm1-Deficient Macrophages When Primed with IFN-γ

As discussed in chapter 3, classical activation to the pro-inflammatory state of

M1 macrophages with exposure to lipopolysaccharide (LPS), alone or in combination with IFN-γ involves a shift from metabolism driven by oxidative phosphorylation and lipid oxidation, to one primarily driven by glycolysis (142,315,370). Metabolic reprogramming after immune cell activation often coincides with a change in cell signaling pathways, leading to distinct response patterns, including cytokine secretion.

Given the metabolic data described in chapter 3, I hypothesized that the metabolic shifts seen in our IFN-γ-primed, Irgm1-deficient macrophages drive pro- inflammatory cytokine secretion in our macrophage model. To explore this hypothesis, I

110

Figure 18: IFN-γ-primed macrophages show enhanced secretion of inflammatory cytokines in the absence of Irgm1

Irgm1. (A) Bone marrow-derived macrophages (BMM) from WT and Irgm1-deficient mice were primed with IFN-γ for 24h. Cytokine levels were measured in conditioned media using a cytokine array dot-blot. (B and C) BMM from WT and Irgm1-deficient mice were primed with IFN-γ or were maintained under control conditions for 24h. Levels of RANTES (B) and MCP-1 (C) were measured in conditioned media using ELISA. Shown are average relative levels measured in triplicate samples for the indicated cytokines. Error bars indicate SEM. *p<0.05. **p<0.001, ***p<0.0001

exposed conditioned media from IFN-γ-primed, WT and Irgm1-deficient macrophages to a dot blot cytokine array. In agreement with our hypothesis, I found several significantly elevated cytokines in our Irgm1-deficient BMM, including MCP-1/CCL2,

MIG, RANTES/CCL5 and TNFα (Figure 18 A). The increases in RANTES and MCP-1 111

were replicated using ELISAs (Figure 18 B-C). In subsequent studies described below, I examined the underlying mechanism that may drive increased pro-inflammatory cytokine production in Irgm1-deficient cells, focusing on RANTES and MCP-1 as representative, IFN-γ-induced, cytokines.

Figure 19: Inhibition of autophagy does not affect RANTES production in IFN-γ - primed macrophages

BMM of the indicated genotypes were plated and maintained under control conditions or primed with IFN-γ for 24 h. Conditioned media was collected and used for ELISAs to measure (A) RANTES or (B) IL-1β. Shown are averages of at least 3 experiments. Error bars indicate SEM. *p<0.05. **p<0.001, ***p<0.0001

4.2.2 Inhibition of Autophagic Flux Does Not Increase RANTES Secretion in IFN-γ-Primed Macrophages

As alluded to above, Irgm1-deficiency has been linked to inhibition of autophagic flux in macrophages (62,97,108,109), fibroblasts (75), and enterocytes (116).

Thus it is plausible that this impairment in autophagy may be involved in the metabolic

112

changes in Irgm1-deficient macrophages, and that it may drive changes in cytokine expression. To explore this possibility, I examined BMM from Atg7f/fLysM-Cre mice that have greatly reduced expression of the essential autophagy protein, Atg7, in macrophages. However, when autophagy was blocked via the Atg7 deletion, RANTES production induced by IFN-γ-priming was not affected (Figure 19 A). These results suggest that an autophagy-independent mechanism drives increased production of a subset of cytokines elevated in Irgm1-deficient cells including RANTES. It remains possible that impaired autophagy may be involved in the increases in others cytokines seen in Irgm1-deficient macrophages. In support of this, I also measured production of

IL-1β in macrophages, finding that cytokine to be substantially elevated in both Irgm1- deficient and Atg7-deficient BMM once activated with LPS (Figure 19 B) (328).

4.2.3 Effect of Metabolic Changes on Pro-Inflammatory Cytokine Production in Irgm1-Deficient Macrophages

As the metabolic changes seen in chapter 3 seemed to be a primary consequence of Irgm1-deficiency in IFN-γ-primed macrophages, I subsequently hypothesized that these metabolic changes were necessary for the marked increases in inflammatory cytokine production.

I hypothesized that the overproduction of RANTES and MCP-1 would be blunted, if we treated the IFN-γ-primed macrophages with the glycolytic inhibitor 2- deoxy glucose, or fatty acid synthesis inhibitors cerulenin or C75. As predicted, I 113

Figure 20: Blocking glycolysis or fatty acid synthesis mitigates the increased RANTES and MCP-1 secretion in IFN-γ-primed macrophages lacking Irgm1

BMM of the indicated genotypes were plated in triplicate and maintained under control conditions, primed with IFN-γ, primed with IFN-γ and simultaneously exposed to glycolytic inhibitor 2-DG (A and B), primed with IFN-γ and simultaneously exposed to the fatty acid synthase inhibitors the fatty acid synthase inhibitor C75 or Cerulenin (C and D), or primed with IFN-γ, primed with IFN-γ and simultaneously exposed to the carnitine palmitoyltransferase-1 inhibitor Etomoxir (E) for 24 h. Conditioned media were isolated and used to perform ELISA to measure either RANTES (A, C, E) or MCP-1(B and D). Shown are average values over at least three separate experiments using BMM cultures from separate mice. Error bars indicate SEM. *p<0.05. **p<0.001, ***p<0.0001

114

observed muted secretion of RANTES and MCP-1 in the presence of all three inhibitors

(Figure 20 A-D). These results indicate that the striking metabolic and mitochondrial changes displayed in Irgm1-deficient macrophages following exposure to IFN-γ are key drivers of the pro-inflammatory phenotype of those cells.

The accumulation of acylcarnitines, shown in Chapter 3 (Figure 17) in our IFN-γ- primed, Irgm1-deficient BMM was thought to be caused by either a decrease of fatty acid oxidation, the classical energy source of IFN-γ-primed macrophages, or an increase of fatty acid synthesis. To gain greater insight into the buildup of fatty acid intermediates, I observed RANTES secretion with etomoxir, an irreversible inhibitor of carnitine palmitoyltransferase I (CPT1). CPT1 is the mitochondrial enzyme that catalyzes the transfer of the acyl group of long-chain fatty acids from coenzyme A to carnitine for mitochondrial import (371), and thus inhibits catabolism of long-chain fatty acids via β- oxidation. If the acylcarnitine buildup was caused by decreased β-oxidation in Irgm1- deficient BMM, I would expect to see an increase in RANTES in our WT BMM IFN-γ- primed macrophages incubated etomoxir. However, no such change is observed (Figure

20 E), suggesting that the buildup of LC-ACC observed in Irgm1-deficient BMM primed with IFN-γ results from a surfeit of fatty acids rather than a decrease in β-oxidation.

115

Figure 21: N-acetylcysteine mutes RANTES and MCP-1 production in IFN-γ-primed macrophages despite no difference in ROS levels

BMM of the indicated genotypes were maintained under control conditions, or primed with IFN-γ for 24 h, with or without the reactive oxygen species quencher, NAC. (A) RANTES or (B) MCP-1 was measured in conditioned media from the cells using ELISA. (C) BMM of the indicated genotypes were maintained under control conditions, were primed with IFN-γ for 24 h, or were activated with IFN-γ and LPS for 16 h with or without the reactive oxygen species quencher, NAC. ROS levels were measured with the fluorescent probe CM-CH2DCFDA. For each figure, shown are average values from experiments performed at least 3 times. Error bars indicate SEM. *p<0.05. **p<0.001, ***p<0.0001

116

4.2.4 Effect of N-Acetyl Cysteine on Pro-Inflammatory Cytokine Production in Irgm1-Deficient Macrophages

Despite observing no increase in reactive oxygen species (ROS) levels in IFN-γ-primed,

Irgm1-deficient macrophages, I demonstrated (Chapter 3, Figure 21) that the ROS quencher N-acetyl cysteine (NAC) reversed the punctate mitochondrial phenotype in these cells. I predicted that this inhibitor would also mute the increased RANTES and

MCP-1 secretion in these cells (Figure 4.4 A-C). When treated with NAC, secretion of both cytokines was decreased in the presence of NAC, indicating, a shift in the redox capacity of Irgm1-deficient cells.

4.2.5 Secretion of TNFα and IL-1β Are Elevated in Irgm1-Deficient Macrophages Stimulated with IFN-γ and LPS

While IFN-γ is thought to ‘prime’ cells for infection, LPS, or another TLR4 agonist, it is required to activate most ‘classical’ M1 macrophage responses (176,372,373).

One of the main signaling pathways activated by TLR4 stimulation is MyD88 signaling transduction, inducing the nuclear transduction of NF-κB and phosphorylation of

MAPKs, resulting in the induction of many M1 macrophage responses, including the secretion of pro-inflammatory cytokines TNFα and IL-1β (114,296,372). While the

MyD88 pathways induced by TLR4 activation have been shown to be distinct from the

IFN-γ mediated induction of Irgm1 expression (114), Irgm1 expression is required for proper immune defense against bacterial pathogens (372,374,375). Thus, we next

117

Figure 22: Macrophages stimulated with IFN-γ/LPS show enhanced secretion of inflammatory cytokines in the absence of Irgm1

Bone marrow-derived macrophages (BMM) from WT and Irgm1-deficient mice were primed with IFN-γ/LPS for 24 h. Levels of (A) IL-1β (B) and TNFα were measured in conditioned media using ELISA. Shown are average relative levels measured in triplicate samples for the indicated cytokines. Error bars indicate SEM. *p<0.05. **p<0.001

observed the secretion of the LPS-dependent cytokines, TNFα and IL-1β in Irgm1- deficient macrophages.

Previous reports, from our lab and others, have shown increased secretion of the pro-inflammatory cytokines TNF-α in Irgm1-deficient macrophages treated with IFN-γ and LPS (Figure 22 A) (114). Interestingly, I also found elevated levels of IL-1β secreted by Irgm1-deficient macrophages (Figure 22 B), strengthening the theory that Irgm1 acts as a negative regulator of pro-inflammatory cytokines, in the presence of bacterial pathogens.

118

4.2.6 TNFα Secretion is Slightly Muted in the Presence of ROS and Glycolytic Inhibitors

As inhibiting glycolysis with 2DG served to mute increased cytokine secretion in

IFN-γ-primed, Irgm1-deficent, macrophages, I sought to determine whether it could dampen increased TNFα secretion as well. Indeed, I found TNFα secretion decreased a

Figure 23: TNFα Secretion in Irgm1-deficient macrophages in the presence of glycolytic and ROS inhibitors

Bone marrow-derived macrophages (BMM) from WT and Irgm1-deficient mice were primed with IFN-γ/LPS for 24 hours in the presence of (A) 2-DG or (B) N-acetylcysteine. Levels of TNFα were measured in conditioned media using ELISA. Shown are average relative levels measured in triplicate samples for the indicated cytokines. Error bars indicate SEM. *p<0.05

small, but significant, amount (Figure 23 A) in Irgm1-deficent macrophages, but remained unchanged in our WT cells. Subsequently, I asked whether quenching ROS would dampen TNFα secretion. As previously shown, Irgm1-deficient macrophages produce significantly more ROS in the presence of IFN-γ/LPS than WT. As with 2DG,

NAC muted TNFα secretion a small, but significant, amount. Although increased 119

glycolysis and ROS in Irgm1-deficient macrophages may be factors in increased TNFα secretion, the relatively small effects of the inhibitors suggest that they may not be the primary driving factors in the enhanced TNFα secretion seen in these cells.

4.2.7 IL-1β Secretion is Muted in the Presence of ROS and Glycolytic Inhibitors

IL-1β secretion requires stimuli that activate the inflammasome, most commonly the NLRP3 inflammasome. Activation of the inflammasome is a two-step process.

NLRP3 is prototypically primed by the binding of LPS to TLR4, and activated by a second challenge (such as mitochondrial ROS, potassium efflux through ion channels or cathepsin release from lysosomal membranes) supplied by the metabolic changes occurring during M1 macrophage activation (62,376).

As with TNFα, I was interested in determining if muting pro-inflammatory stimuli, such as glycolysis and ROS production, would decrease the elevated IL-1β secretion in IFN-γ-primed, Irgm1-deficent macrophages. I observed a significant decrease in IL-1β secretion when either glycolysis or ROS was inhibited via 2DG or

NAC, respectively (Figure 24 A-B). This suggests that, unlike TNFα, the metabolic changes in IFN-γ-primed, Irgm1-deficient macrophages may be primary driving factors in the over expression of IL-1β.

120

Figure 24: 2-Deoxyglucose and N-acetylcysteine mute IL-1β secretion in Irgm1- deficient macrophages

Bone marrow-derived macrophages (BMM) from WT and Irgm1-deficient mice were primed with IFN-γ/LPS for 24 hours in the presence of (A) 2-DG or (B) N-acetylcysteine. Levels of IL-1β were measured in conditioned media using ELISA. Shown are average relative levels measured in triplicate samples for the indicated cytokines. Error bars indicate SEM. *p<0.05. **p<0.001

4.2.8 Enhanced Pro-Inflammatory Cytokine Production in Irgm1- Deficient Mice and Cells

My next objective was to verify that our observations of increased pro- inflammatory cytokine secretion in our macrophage cell culture model could be replicated in our mouse model. To do so, our lab used a multiplex approach to measure production of an array of pro-inflammatory cytokines in the serum of Irgm1-deficient mice that were either uninfected, or injected with S.typhimurium for two days (Appendix

3).

Comparing the data from uninfected WT and Irgm1-deficient mice I observed a dramatic increase in pro-inflammatory cytokine secretion. Of the 23 cytokines that were assessed, 21 were elevated at least two fold in Irgm1-deficient mice as compared to 121

122

Figure 25: Irgm1-deficiency leads to enhanced cytokine production in mice

Sera collected from 3 WT and 3 Irgm1-deficient mice were used for cytokine measurements using a multiplex assay. (A) Shown are average levels of the indicated cytokines in uninfected Irgm1-deficient mice relative to levels in uninfected WT mice. (B) Shown are average levels of the indicated cytokines in two-day S. typhimurinum infected Irgm1-deficient mice relative to levels in two-day S. typhimurinum infected WT mice. (C) Cytokine fold increases between uninfected and infected WT and Irgm1- deficient mice. Error bars indicate SEM. *p<0.05

levels in infected WT mice (Figure 25 A). The majority of those cytokines were elevated at least as much in naïve Irgm1-deficient mice, as they were in those mice following infection with Salmonella typhimurium.

In infected Irgm1-deficient mice, the overproduction of pro-inflammatory cytokines was present, but to a much lower degree. Only 7 of the cytokines assessed were elevated two fold or greater, as compared to levels in infected WT mice (Figure 25

B). As expected, I observed that 20 of the 23 cytokines in WT mice increased after infection at least 2-fold or greater. This stands in stark contrast to the Irgm1-deficient mice, where only 7 cytokines were increased 2-fold or greater, and 11 cytokines decreased, after infection. These results correlate with our macrophage data, and indicate that there is a robust auto-inflammation in Irgm1-deficient mice, even in absence of infection. As stated earlier, this data implies that Irgm1 negatively regulates pro-inflammatory cytokine secretion.

123

4.3 Discussion

Traditionally, immunology and metabolism were thought to be distinct fields of study that rarely intersected. However, the past couple decades have produced overwhelming evidence to the contrary, demonstrating not only that altered metabolism can cause immune cell dysfunction, but that specific immune cell functions, such as activation, are dependent on metabolic shifts. In this chapter, I demonstrated that the metabolic changes observed in chapter 3 drive IL-1β secretion in Irgm1-deficient macrophages stimulated with both IFN-γ and LPS, and RANTES and MCP-1in Irgm1- deficient macrophages primed with IFN-γ alone.

TNFα was the sole cytokine overexpressed in Irgm1-deficient macrophages that did not demonstrate a large cytokine decrease upon either inhibition of ROS or glycolysis. However, at least regarding glycolysis, our result is in agreement with other reports from the literature that also indicate TNFα does not appear to be directly dependent on aerobic glycolysis (166,171).

The enhanced secretion of the chemokines RANTES and MCP-1 are often observed in diseased states in which macrophages are exposed to, or are forced to import, high levels of lipids. As an example, take the macrophage phenotype of adipose tissue macrophages (ATMs) (377). Macrophages in lean adipose tissue possess a M2-like phenotype, characterized by the expression of anti-inflammatory proteins, including the mammalian lectin Ym1, arginase 1, and cytokines IL-10 and IL-4 (378,379). On the other

124

hand, obese adipose tissue has been characterized by an increase in the population of

M1-like, pro-inflammatory macrophages, characterized by increased expression levels pro-inflammatory cytokines, including TNFα and MCP-1, and inducible nitric oxide synthase, and is suggested to contribute to insulin resistance in type II diabetes

(379,380). Moreover, enhanced cytokine production is often observed in obese adipose tissue macrophages, including MCP-1, RANTES and IL-1β (381,382). Similar lipid induced dysfunction, leading to enhanced cytokine secretion, is also observed in foam cells in atherosclerosis (330,350,383,384), and Kupffer cells in fatty liver disease

(185,303,385). In lipid-enriched macrophages from these diseases, RANTES and MCP-1 secretion has been reported to be induced through any number of pathways, including

JNK, NF-κB and Hif1α, all of which are probably active in our Irgm1-deficient cells primed with IFN-γ and IFN-γ/LPS. Although Hif1α typically requires LPS stimulation, it has been shown recently that second TCA cycle break in macrophages with an M1-like metabolic phenotype induces succinate accumulation, which in turn stabilizes Hif1α, allowing it to travel to the nucleus and activate transcription of pro-inflammatory genes

(177,226).

Although succinate accumulation is observed in IFN-γ primed Irgm1-deficient macrophages, I have not verified Hif-1α activation. Therefore, as of yet, I do not know whether RANTES and MCP-1 expression is induced through this pathway. Given that little to no reactive oxygen species are observed in IFN-γ primed, Irgm1-deficient

125

macrophages, the inflammasome activation required to generate active IL-1β likely cannot occur in the absence of LPS.

Finally, the pattern of increased inflammatory cytokine levels in the serum of

Irgm1 deficient mice mirrors the one in the metabolite analysis. In uninfected mouse serum, or IFN-γ primed macrophages, where little to no change in cytokine/metabolism is expected, a massive shift is observed, whereas in IFN-γ/LPS stimulated cells/ S. typhimurinum infected mice, the differences observed between Irgm1-deficient and WT macrophages/mice is significant, but relatively weak. These results further support the theory that, contrary to past studies, underlying metabolic changes are driving a state of hyper-inflammation in of Irgm1-deficient mice and cells. Although the studies in this chapter, and this dissertation, focus on macrophages, it is unlikely that they are the sole driver of inflammation in a whole animal. However, many immune cells beyond macrophages are known to shift towards a pro-inflammatory activation state under metabolic dysfunction, most notably T-cells (shifting toward an effector T-cell phenotype) (142) and dendritic cells (40). A model where Irgm1-deficient mice exist in a state of Th1-like response, driven through the cell-autonomous activation of various immune cells, both agrees with the role of Irgm1 as a negative regulator of IFN-γ responses, and provides a framework to begin investigating how this might occur.

126

4.4 Methods

Mice

Irgm1-deficient (88) and Atg7f/fLysM-Cre mice (356) have been described previously. All mice were housed and maintained under procedures approved by the

Institutional Animal Care and Use Committees at the Duke University and Durham VA

Medical Centers.

Cell Culture

Primary murine bone marrow-derived macrophages (BMM) were isolated from the tibia and femurs of 2- to 4-month-old mice and cultured according to standard procedures described previously (91). The bone marrow was flushed from the bones using a 27G needle fitted to a syringe filled with DMEM (Life Technologies); the marrow was dispersed by drawing through the needle three to four times; and red cells were lysed with ACK lysing buffer (Life Technologies). Adherent cells were cultured for

6 days in BMM medium [DMEM supplemented with 10% (v/v) FBS (Hyclone) and 30%

Twenty-four hours prior to all experiments, the cells were placed in medium lacking

L929-conditioned media [DMEM (11995, Gibco/ThermoFisher), supplemented with 10%

127

(v/v) FBS and 100 units/ml penicillin/100 µg/ml streptomycin (15140,

GIBCO/ThermoFisher)] on coverslips or cell culture plates. All primary cells and cell lines were grown and maintained in a humidified atmosphere of 5% CO2 at 37°C. Where appropriate, 0.2 ng/ml interferon (IFN)-γ (IF005, EMD Millipore) and/or 100ng/ml LPS

(L2387, Sigma) was included in the growth medium.

Serum Cytokine Array

Sera were isolated, immediately frozen, and stored at −80 °C. Serum cytokine levels were determined using a Bio-Plex Mouse Cytokine 23-plex Assay (Bio-Rad). All bead assay samples were quantified on the BioPlex protein array reader (Bio-Rad) in the

Laboratory Immunology Unit (Duke Human Vaccine Institute Durham, NC).

Bacterial Infections

Salmonella typhimurium SL1344 (386) was cultured overnight at 37 C in Luria-

Bertani (LB) broth without shaking. Mice were injected iv with 6×105 bacteria in a volume of 0.1 ml of PBS as described previously (91).

Cytokine Dot Blot

A Proteome Profiler Mouse Cytokine Array Kit, Panel A (ARY006, R&D Systems) was used to compare levels of a small array of cytokines. Cells were plated at 3x105

128

cells/well in six-well plates, and primed with IFN-γ for 24 hours. The conditioned media from triplicate wells were collected and pooled, and then used to carry out the assay according to the manufacturer’s instructions. The dot blots were imaged on a Kodak

Image Station 4000R using Carestream Molecular imaging software. The sum intensity of the dot blots were calculated using the Carestream software, and were normalized to protein content.

ELISAs for Cytokine Measurement

ELISA kits were used to measure levels of individual cytokines (mouse

RANTES/CCL5, DY478, R&D Systems; mouse MCP-1/CCL2, DY479, R&D Systems; mouse IL-1β ELISA Set, 559603, BD Biosciences; mouse TNFα ELISA kit, 560478, BD

Biosciences). Cells were plated in triplicate at 1x105 cells/well in 24-well plates; 24 h later, the media were changed and the cells incubated for an additional 24 hours in 0.5mL media containing as appropriate: 0.2 ng/ml interferon (IFN)-γ (IF005, EMD Millipore),

100ng/mL LPS, 100 μM Etomoxir (236020 Calbiochem), 20μM Cerulenin (C2389, Sigma),

10 μM C75 (C5490, Sigma), 1mM 2-deoxy-D-glucose (D6134, Sigma), and/or 15mM N- acetyl-L-cysteine (A7250, Sigma). Conditional media were collected from each well, and used neat for the MCP-1 ELISA, or at a 1:4 dilution for the RANTES ELISA according to the manufacturer’s instructions. The absorbance was read using Gen5 software on a

BioTek Synergy 2 plate reader. IL-1b and TNF ELISAs.

129

Statistical Analysis

As indicated for the particular figure, the Z-test or Student's t test, as calculated by Excel, was used to assess statistical significance. The significance threshold set a priori to be p < 0.05.

130

5. Conclusion and Future Directions

5.1 Overview

The innate immune system is equipped with many mechanisms to counter the onslaught of microbial pathogens that it encounters on a daily basis. One of these mechanisms is the induction of a family of IFN-γ dependent immune effectors, the immunity related GTPases or IRGs. The IRGs are best known for their ability to mount a cell-autonomous defense against pathogens even before PPRs, such as TLR4, are engaged (62). In mice, who possess 16 IRG genes and 4 annotated IRG pseudogenes (65), the IRGs comprise an important innate immune defense against intracellular bacterial and protozoan pathogens (89). Three mouse IRG genes, Irgm1, Irgm2 and Irgm3, are functionally classified as ‘GMS’ IRGs, so-named as they contain a lysine to methionine amino acid substitution in their p-loop GTP-binding motif (GX4GMS), while the rest are considered ‘GKS’ IRGs and contain the canonical GX4GKS p-loop motif. The working model of IRG function indicates that, upon direct infection of a cell by an invasive pathogen, the cytosolic GKS IRGs are activated and localize to the pathogen-containing compartment where they act against the pathogen through unknown mechanisms.

Meanwhile, the GMS IRGs coat the cell’s intracellular ‘self’ membranes upon expression, blocking the GKS IRGs from localizing to these membranes and activating prematurely

(71,74).

131

However, many studies indicate that the IRGs have functions beyond direct intracellular pathogen restriction. Studies of one particular IRG, Irgm1, have implicated it in lysosomal function (96), autophagy regulation (100,108), and negative regulation of

Th1 inflammatory responses and survival of immune cells (122,293). Furthermore, mice deficient in Irgm1 are susceptible to infection by many intracellular pathogens, including as T. gondii (83,87), C. trachomatis (79,110), E. cuniculi (111), L. monocytogenes

(88,112), M. tuberculosis (76,80), M. avium (93), T. cruzi (112) and L. major (90,113), despite the fact that many of these pathogens are not targeted by GKS IRG upon infection, as well as to several autoimmune diseases, including animal models of colitis (115,116), immune encephalitis (117,118), and stroke (119). Thus, Irgm1 seems to mediate unique functions in innate immunity through unknown mechanisms.

In this dissertation, I focused on discovering metabolic pathways that are engaged by mouse Irgm1, and the immune phenotypes that resulted from Irgm1’s absence. My research was initiated by the discovery of the phenotype of hyperfused mitochondria observed in Irgm1-deficient MEFs after IFN-γ induction. Although only a few subtle mitochondria-related phenotypic deficiencies were observed initially, I discovered that IFN-γ primed Irgm1-deficient MEFs had distinct changes in metabolic functions, including an increased glycolytic rate and an overall depletion of key mitochondrial metabolites.

132

As immunometabolism has emerged as a critical component in immunity, I subsequently began to study the modulation of metabolic pathways in bone marrow macrophage (BMM) function during activation. I discovered that, in the absence of

Irgm1, priming Irgm1-deficient cells with IFN-γ alone induced metabolic shifts typically associated with M1macrophage activation involving LPS stimulation. Further, when

Irgm1-deficient macrophages were stimulated with both IFN-γ and LPS, the cells showed an overall decrease in metabolites compared to WT cells. Finally, I discovered that the changes in metabolism were causing increased pro-inflammatory cytokines, both in Irgm1-deficient macrophages induced by either IFN-γ alone, or with IFN-γ and

LPS, and in our Irgm1-deficient mouse model.

The work presented in this dissertation adds immunometabolism to the laundry list of pathways that are implicated in Irgm1 function. However, this new function of

Irgm1 corroborates its other functions in immunity. As mentioned above, Irgm1 has been described as a negative regulator of IFN-γ signaling in lymphocytes, as animals lacking Irgm1 present with severe lymphopenia. A shift toward a glycolytic phenotype in immune cells is known to promote and sustain Th1-like responses. Similarly, aberrant metabolic activation of immune cells is implicated in many autoimmune diseases, including those observed in Irgm1-deficient mice. To conclude, I will discuss what topics need to be addressed with future studies as a result of this dissertation.

133

134

Figure 26. Potential mechanisms of metabolic dysfunction and increased inflammatory cytokine production in Irgm1-deficient macrophages

The data presented in this dissertation suggest three hypothetical, non-exclusive mechanisms through which Irgm1 deficiency may lead to increased inflammatory cytokine production in macrophages. A) The inflammatory phenotypes observed in Irgm1-deficient BMM result from the premature activation of inflammatory cell signaling pathways, including MAPK kinase and NF-κB. It is possible that Irgm1 acts as a direct inhibitor of these pathways, or as an upstream inhibitor, acting against the Akt or PIK3 signaling pathways. B) Irgm1-deficiency may lead to lysosomal dysfunction. Normal lysosomal function is crucial for lipid signaling in the cell, while the buildup of dysfunctional lysosomes is known to instigate inflammatory cell signaling. C) The accumulation of lipids in Irgm1 deficient cells may be the precipitating event that secondarily leads to decreased lysosomal function. Irgm1, which localizes to lipid droplets, may promote normal lipid trafficking, lipolysis, and/or lipophagy. Aberrant lipid homeostasis alone, or the downstream lysosomal damage, could lead to proinflammatory signaling.

5.2 Possible Mechanisms of Actions in the IFN-γ Activation of Irgm1-Deficient Cells

My exploration of the metabolic phenotype of Irgm1-deficient macrophages in this thesis is far from exhaustive. One of the most glaring gaps in knowledge from this project is that it does not address how this metabolic phenotype observed in Irgm1- deficient cells is established. This is of particular interest in Irgm1-deficient immune cells that are only primed with IFN-γ. There are three main areas of research that need to be addressed in Irgm1-deficient cells in order to determine this mechanism of action: the inhibition of lysosomal/autophagosomal fusion, IFN-γ signaling, and lipid metabolism.

IFN-γ Signaling

135

Surprisingly, I have been able to find very little research about IFN-γ signaling in

Irgm1-deficient cells. A 2007 article by Bafica et al presented some evidence of altered cell signaling in the context of LPS stimulated Irgm1-deficient macrophages. It was shown that Irgm1-deficient macrophages display enhanced activation of the NF-κB and p38 MAPK signaling pathways after LPS stimulation. No definitive mechanism was found, but the author hypothesized that these enhanced signaling pathways may be induced through Akt (114). Further research by the Macmicking lab demonstrated that

Irgm1 binds the bioactive lipids PtdIns[3,4]P2, PtdIns(3,4,5)P3 and cardiolipin (76), and that IFN-γ can induce autophagy in an Irgm1-independent manner via the p38

MAPK/PI3K pathways (139). Finally, the Goodell lab demonstrated that IFN-γ signaling was dysregulated in Irgm1-deficient hematopoietic stem cells, leading to enhanced hyperproliferation and defects in self-renewal, and autophagy (122). Taken together these studies suggest that the absence of Irgm1 induces hyper-activation of IFN-γ induced signaling pathways, or dampens the mechanisms responsible for the negative regulation of these pathways.

This hypothesis is supported by the results of a previous TUBE1 pulldown, a process that isolates K63-linked-ubiquitinated proteins, performed previously in our lab with IFN-γ induced WT and Irgm1-deficient MEFs. The results showed an over eighty percent increase in the ubiquitination of SOCS1, and a total lack of STAT1 ubiquitination in Irgm1-deficient cells compared to WT. The over-ubiquitination of SOCS1, a potent

136

negative regulator of IFN-γ (387), suggests it is either being overexpressed in these cells, or its degradation is increased. Similarly, given that the dysregulated IFN-γ signaling observed in Irgm1-deficient hematopoietic stem cells is dependent on STAT1, decreased ubiquitination of STAT1 is consistent with increased IFN-γ signaling in these cells.

Moreover the enhanced activation of the NF-κB and p38 MAPK signaling pathways in

Irgm1-deficient macrophages after LPS stimulation strongly suggests that this may be true in IFN-γ primed macrophages as well (114).

As indicated above, one signaling pathway that has not been investigated in the context of Irgm1-deficiency (to my knowledge) is the PIK3/Akt pathway, despite it being critical in both restricting and promoting anti-inflammatory responses in macrophages

(152). The PI3K/Akt pathway is activated by TLR4 and other PPRs, modulating downstream signals that control cytokine production. Activated PI3K type I phosphorylates phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P2] to generate phosphatidylinositol 3,4,5-triphosphate [PtdIns(3,4,5)P3] at the plasma membrane.

Finally, PtdIns(3,4,5)P3 recruits Akt and facilitates its activation by mTORC2 (388,389).

This is noteworthy, as Irgm1is also known to specifically bind PtdIns(4,5)P2 and

PtdIns(3,4,5)P3, and therefore potentially participates in crosstalk with this pathway (76).

Further, the PIK3/Akt1 axis upregulates IL-1 receptor associated kinase M (IRAK-M), a suppressor of TLR4 signaling via TRAF6 (E3 ubiquitin ligase tumor necrosis factor receptor associated factor 6) inactivation (390). Given that the Coers lab demonstrated

137

that activated GKS IRGs recruit TRAF6 to C. trachomatis’s or T. gondii‘s pathogen vacuole, this represents a second possible crosslink between these pathways.

Recent studies have shown that Akt1 and Akt2 kinase isoforms hold a key role in the regulation of macrophage activation. Deletion of Akt1 promotes upregulation of M1- macrophage responses, abrogating endotoxin tolerance (391-393), and activates autophagy (394). Akt1 also mediates suppression of TLR4-induced macrophage activation by adipokines and neuropeptides (380,395,396). Conversely, deletion of Akt2 promotes an M2-like phenotype, and express elevated levels of the M2 transcription factor C/EBPβ and the M2 markers Arg1, Ym1 and Fizz1 (393), and results in negative regulation of the TLR4 signaling pathway (391). Interestingly, in the absence of Akt1,

(which negatively regulates LPS inflammation), DSS-induced intestinal inflammation was exacerbated in a mouse model of colitis, and greater CNS damage was observed in the mouse model of experimental autoimmune encephalomyelitis – both mouse models of Th1-mediated inflammatory diseases that are also exacerbated in the absence of Irgm1

(116,397). Given preceding paragraphs, I hypothesize that there is crosstalk between

Irgm1 and Akt negatively regulating M1 macrophage inflammatory responses.

Irgm1-Mediated Lysosomal Damage

A paper published last year by the Howard lab established that in the absence of

Irgm1 activated GKS IRGs are loaded onto lysosomes, resulting in reduced lysosomal acidity and a failure of autophagosomal processing (96). Lysosomal damage is known

138

for inducing a pro-inflammatory phenotype in immune cells, therefore it is possible that the lysosomal damage accounts for the M1-like activation observed in our Irgm1- deficient macrophages. However, these studies were performed in MEFs, not macrophages, therefore further validation is required. Further, lipotoxicity is typically the inducer of lysosomal damage in pro-inflammatory macrophages (398,399), therefore, more research is needed to infer whether GKS IRG-induced lysosomal damage is the primary, or only, inducer of the M1-like phenotype observed in Irgm1-deficient macrophages.

Lipid Metabolism

The metabolic phenotype of Irgm1-deficient cells presented suggests that lipid flux is dysregulated in these cells. The lipid build-up, and M1-like phenotypic shift in

IFN-γ-primed Irgm1-deficient macrophages is reminiscent of the altered macrophage phenotypes observed in metabolic disorders characterized by altered lipid homeostasis, such as foam cell macrophages in atherosclerosis, adipose macrophages found in the fatty tissue of obese patients, or Kupffer cells in fatty liver disease (185). All three conditions are marked by elevated lipid uptake in macrophages, inducing an M1-like phenotypic change, resulting in the secretion of pro-inflammatory cytokines. Indeed, the intake of saturated fatty acids, such as palmitate and stearate, is strongly correlated with the inflammation and mitochondrial dysfunction seen in these disorders. Thus, the

139

build-up of lipids in immune cells can drive systemic effects, such as the rampant inflammation observed in Irgm1-deficient mice.

In Irgm1-deficient macrophages treated with IFN-γ/LPS, an overall decrease in acylcarnitine levels are observed compared to WT, yet they contain twice the level of neutral lipids. Further, even without LPS (normally required to induce an M1 phenotype, including lipid synthesis) Irgm1-deficient macrophages, primed solely with

IFN-γ, induced the accumulation of neutral lipids and long chain acylcarnitines. The difference in lipid utilization between Irgm1-deficient and WT macrophages is even more striking when it is considered that a large portion of the neutral lipid increase in

IFN-γ Irgm1-deficient macrophages is observed with or without the presence of extracellular lipids, implying that this is the result of de-novo lipid synthesis. Moreover, neutral lipid increase is not observed in WT macrophages in the absence of extracellular lipids, even when stimulated with both IFN-γ and LPS, indicating that its increase in neutral lipid mass results mainly from lipid import.

The accumulation of neutral lipids observed in Irgm1 macrophages most likely corresponds to an increase in the number and/or size of lipid droplets: an organelle consisting of a core of neutral lipids and steryl esters, surrounded by a phospholipid monolayer and a heterogenous group of proteins, comprising a key mechanism of cellular energy storage- in the Irgm1-deficient cells (342,343). This is of significance, as the mouse Irgm proteins, including Irgm1, are known to localize to lipid droplets, and

140

the GKS IRGs aberrantly localize to lipid droplets in the absence of Irgm1. Further, a recent study demonstrated that human IRGM influences hepatic lipid droplet metabolism by modulating lipophagy (400) - a newly-identified selective form of autophagy that degrades lipid-droplet sequestered lipids (202). This study showed that

IRGM knockdown increases lipid droplet accumulation, while its overexpression reduced lipid droplet content in human hepatoma HepG2 cells in an autophagy dependent manner. Another study suggested that Irgm1 enhanced autophagy in a melanoma cell line through interaction with the membrane curvature protein Bif-1 (401), a protein recently implicated in the lipophagy (202) and that this interaction was conserved in human IRGM. Thus Irgm1may have a role in the autophagic lipolysis. In such a case, the absence of Irgm1 may result in a block of lipid transport and restrict lipid droplet usage.

In addition, lipid droplets serve as a scaffold (or signaling platform) that recruits proteins involved in lipid metabolism, and other metabolic processes (402). Therefore

Irgm1 may also be enhancing, or inhibiting signaling pathways on lipid droplets. One example would, once again, be autophagy. Members of the ubiquitin-like Atg8 protein family, including microtubule-associated protein 1 A/1B light chain 3 (LC3), which is recruited to the surface of lipid droplets and conjugated with phosphatidylethanolamine

(403). Independent of its role in macroautophagy, the enzymatic machinery that controls the lipidation of LC3 and other Atg8 proteins also boosts the association of GKS IRGs

141

with T. gondii and C. trachomatis phagosomes, leading some groups to hypothesize that

Atg8-decorated membranes induce GKS IRG activation (68). One possible mechanism is that Irgm1 prevents GKS IRG activation at lipid droplet membranes. However, in the absence of Irgm1, activation of GKS IRGs occurs, potentially leading to lysosomal dysfunction, and resulting in the inhibition of autophagic lipolysis and lipid droplet growth.

This is potentially of even greater importance in Irgm1-deficient cells, where the

GKS IRGs aberrantly localize to lipid droplets in the absence of Irgm1. Given the abundance of neutral lipids in Irgm1-deficient macrophages, a greater portion of the

GKS IRGs may localize to lipid droplets, inhibiting them from associating with a pathogen’s phagosomal membrane during infection. Furthermore, lipid droplets are known signaling platforms. Aberrant accumulation of the GKS IRGs may disrupt signaling at this interface. If the lipid droplets are larger or more numerous in Irgm1- deficient cells, the surface area of this signaling platform, potentially magnifies any aberrant signaling. Therefore this also supports the hypothesis of increased signaling in

Irgm1-deficient cells. Moreover, Akt signaling, which may be increased in these cells, also occurs on lipid droplets, and upstream of the NF-κB and p38 MAPK signaling pathways. Another article reported lipophagy functions as a mechanism for lipid efflux from the cell (404). Inhibition of lipophagy, and subsequent efflux of excess lipid mass,

142

could be part of the cause of the accumulation of neutral lipid mass that may be involved in IFN-γ primed Irgm1-deficient cells.

The cause of intracellular pathogen susceptibility may also have to be re- examined if it shown that Irgm1-deficiency causes lipid flux. There is increasing evidence that protozoan and bacterial parasites may target host-derived lipid droplets either for gaining nutrients or for escaping the host immune response (405). Many of these pathogens are known for modulating lipid droplet biology for their own use, including T. cruzi (406-408), M. tuberculosis (409,410), C. trachomatis (411), L.major (412) and T. gondii (405). Therefore, the greater mass of lipid droplets in IFN-γ-primed Irgm1- deficient macrophages would be a potential boon for these infectious parasites.

In summary, the research presented in this thesis increases the understanding of how Irgm1modulates the innate immune response. My findings have expanded the knowledge of Irgm1function in relation to mitochondria, and discovered that Irgm1 regulates immunometabolism in macrophages, and, in doing so modulate proinflammatory cytokine secretion at both the level of the cell and in mice. Together, these findings expand the role of Irgm1 in immunity and provide new avenues of research.

143

Appendix A

Select metabolites WT and Irgm1-deficient murine embryonic fibroblasts as characterized by metabolomic profiling. MEF Metabolite Concentration [pg/mg protein]

WT Irgm1-deficient

Meanɤ SEM Meanɤ SEM C2 141.800 1.870 120.414 1.870 C3 76.966 4.496 81.803 4.496 C3-DC 0.554 0.142 0.492 0.142 C4 27.238 1.003 18.663 1.003 C4-OH 4.918 0.464 3.203 0.464 C4-DC 1.101 0.203 0.866 0.203 C5 16.699 1.513 16.229 1.513 C5:1 0.640 0.022 0.765 0.022 C5-OH 1.204 0.152 1.140 0.152 C5-DC 0.242 0.119 0.279 0.119 C6 6.175 0.967 5.128 0.967

C7-DC 0.000 0.000 0.006 0.000 Short Short Chain C8 0.941 0.189 0.552 0.189 C8:1 0.056 0.015 0.082 0.015 C8:1-OH/C6:1- 0.189 0.015 0.145 0.015 DC C8-OH/C6-DC 1.118 0.118 0.748 0.118 C8:1-DC 0.029 0.017 0.082 0.017 C8-DC 0.358 0.062 0.328 0.062 C10 0.190 0.190 0.073 0.190 C10:1 0.131 0.023 0.082 0.023 C10:2 0.019 0.019 0.050 0.019 C10:3 0.010 0.010 0.012 0.010 C12 0.739 0.077 0.444 0.077 C12:1 0.350 0.021 0.203 0.021 C12-OH/C10- 0.254 0.028 0.102 0.028

DC Medium Chain Medium C14 1.185 0.025 0.970 0.025 C14:1 0.969 0.128 0.804 0.128 C14:2 0.115 0.031 0.076 0.031 C14:1-OH 1.127 0.058 0.620 0.058 C14-OH/C12- 0.653 0.062 0.394 0.062 DC

C16 3.739 0.390 2.737 0.390

Acylcarnitines LongChain

144

C16:1 1.129 0.137 0.897 0.137 C16:2 0.143 0.035 0.057 0.035 C16:1- 0.395 0.135 0.296 0.135 OH/C14:1-DC C16-OH 0.346 0.062 0.203 0.062 C18 0.854 0.055 0.581 0.055 C18:1 2.516 0.124 2.131 0.124 C18:2 0.196 0.071 0.248 0.071 C18:2-OH 0.274 0.030 0.146 0.030 C18:1- 0.465 0.075 0.383 0.075 OH/C16:1-DC C18-OH/C16- 1.131 0.047 0.776 0.047 DC C18:1-DC 0.166 0.100 0.119 0.100 C18-DC/C20- 0.037 0.025 0.044 0.025 OH C20 0.114 0.019 0.069 0.019

C20:4 0.217 0.055 0.083 0.055

C22 0.094 0.006 0.038 0.006 VLC Glycine 13295.6 795.4 14737.8 795.4 Alanine 8167.4 301.6 9020.2 301.6 Serine 3573.3 99.7 3476.9 99.7 Proline 3773.3 183.4 3884.6 183.4 Valine 1609.8 24.8 1545.3 24.8 Leucine/Isoleucine 2122.0 94.5 2102.9 94.5 Methionine 566.6 31.1 563.4 31.1 Histidine 471.7 18.6 481.8 18.6 Phenylalanine 1319.0 53.7 1323.1 53.7 Tyrosine 1553.7 40.7 1526.9 40.7 Asparagine/Aspartic 4542.1 109.9 3572.8 109.9 Acid

Glutamine/Glutamic 33139.9 1559.9 25973.3 1559.9 Acid Ornithine 239.2 61.1 158.2 61.1 Citrulline 38.3 10.8 19.0 10.8

Arginine 746.8 211.1 493.8 211.1 Amino Acids Amino Lactate 148939.4 14093.0 126488.0 14093.0 Pyruvate 9151.3 738.2 8663.0 738.2

Succinate 1793.6 193.3 1233.0 193.3 Fumarate 2747.6 109.4 1858.0 109.4 Malate 20470.9 1123.9 14961.4 1123.9 α-Ketoglutarate 6450.9 251.1 5117.0 251.1

Citrate 12448.9 510.4 9480.2 510.4 Organic Acids Organic 145

Appendix B

Select metabolites WT and Irgm1-deficient macrophages as characterized by metabolomic profiling. Macrophage Metabolite Concentration [pg/mg protein]

Control IFN-γ IFN-γ/LPS Irgm1- Irgm1- Irgm1- WT WT WT deficient deficient deficient Mean SEM Mean SEM Mean SEM Mean SEM Mean SEM Mean SEM C0 27995.7 1231.5 28348.7 389.0 19646.3 868.2 13999.8* 131.5 12264.4 1752.1 10021.1 629.0 C2 8049.0 1271.2 8384.8 169.9 14064.1 1862.0 12936.8 251.1 6823.6 2682.9 4007.0 698.9 C3 1729.6 276.4 2073.4 229.3 3472.3 271.0 895.6*¥ 31.0 356.7 126.3 110.2 ¥ 21.8

C3-DC 23.5 1.3 27.9 3.2 16.5 2.2 10.0* 1.1 9.0 2.2 3.8 ¥ 0.5 C4 117.7 8.6 147.0 8.4 256.3 22.7 180.3* 5.2 112.5 31.6 55.1 ¥ 8.5 C4-OH 186.0 15.2 217.3* 20.8 122.1 12.0 189.2 9.2 214.7 84.9 100.7 ¥ 13.2 C4-DC 64.8 3.7 83.9 8.2 31.5 2.6 22.9 1.6 14.7 5.5 6.1 ¥ 1.2 C5 158.9 4.1 176.2 10.8 303.7 35.5 135.8*¥ 6.2 28.7 8.6 19.5 1.4 C5:1 16.7 0.1 20.5 1.4 14.4 2.7 12.4 1.8 12.6 3.2 7.2 1.2

Short Short Chain C5-DC 5.0 2.5 11.0 1.2 3.9 1.9 4.0 0.6 3.4 0.6 3.4 0.4 C5-OH 22.8 2.4 23.0 3.3 15.5 2.8 9.4 0.7 12.5 3.3 4.6 ¥ 0.3 C6 47.3 3.3 65.9* 4.3 82.4 8.9 78.9 2.8 27.0 12.3 5.7 ¥ 2.9 C7-DC 9.3 1.4 7.3 4.0 0.5 0.5 0.3 0.3 0.0 0.0 1.8 1.8 C8 28.4 3.5 37.4 1.7 25.6 1.2 32.7* 0.7 21.2 7.8 5.9 ¥ 0.8 C8:1 3.4 0.7 5.7 0.6 3.8 0.5 4.7 0.2 3.5 1.8 2.0 0.5 C8:1- OH/C6:1- 3.2 0.8 5.0 0.6 3.4 0.6 3.6 0.8 2.1 0.3 1.4 0.0

DC C8-OH/C6- DC 36.7 2.5 45.8 6.4 21.7 2.1 13.2 1.6 11.0 2.3 6.3 0.4 C8:1-DC 2.0 0.3 3.6* 0.2 3.2 0.2 1.6 0.4 1.6 0.6 0.8 0.1 C8-DC 20.7 3.3 28.0 4.0 14.6 3.3 14.3 2.3 14.9 5.8 3.0 ¥ 0.4 C10 8.3 4.1 9.7 4.9 14.7 4.0 21.0 1.7 26.1 8.5 5.7 ¥ 0.7 C10:1 3.2 0.3 3.1 1.4 2.8 0.3 3.9 1.0 4.3 1.5 1.8 ¥ 0.3 C10:2 1.7 0.4 3.1* 0.4 0.5 0.2 1.0 ¥ 0.6 0.6 0.3 0.3 0.3

Medium Medium Chain C10:3 11.4 1.0 14.9 0.6 7.4 2.4 6.6 1.1 6.6 3.7 3.3 ¥ 1.0 C12 32.1 1.8 36.5 1.9 47.9 8.4 78.5 2.4 141.7 43.9 42.2 ¥ 4.8 C12:1 8.9 1.1 12.7* 1.4 11.3 1.9 15.7 1.2 20.4 7.9 8.1 ¥ 1.2 C12- OH/C10-DC 12.8 3.0 15.9 1.8 7.9 1.4 12.3 0.8 21.1 6.8 5.9 ¥ 0.7

Acylcarnitines C14 86.5 8.1 111.7* 7.8 112.0 20.5 342.4*¥ 26.8 463.0 70.9 256.3 38.1 C14:1 33.5 3.3 42.3 3.9 52.0 8.8 96.2 3.7 132.5 38.6 53.6 ¥ 10.3 C14:2 1.8 0.9 2.9 0.4 3.7 0.3 7.8*¥ 0.7 7.7 2.0 3.8 ¥ 0.6 C14:1-OH 32.1 8.7 30.2 1.8 27.6 4.9 65.8 5.3 87.7 9.5 53.8 7.6 C14- OH/C12-DC 13.3 1.3 16.7* 0.6 13.6 2.6 28.8 3.1 85.2 12.3 30.8 ¥ 7.1 C16 251.9 24.7 323.0 18.7 303.2 67.1 2279.8*¥ 331.2 3331.7 471.1 3458.0 466.5

C16:1 45.0 8.5 53.4 4.0 54.4 11.6 287.3*¥ 26.6 654.0 26.7 391.8 67.9

C16:2 4.9 1.1 6.3 0.3 3.6 0.9 16.6*¥ 1.8 48.9 2.9 23.9*¥ 2.1 C16:1- OH/C14:1- 18.3 1.9 25.0* 1.4 26.1 4.9 95.3*¥ 7.6 189.9 14.7 99.2 16.9 DC C16-OH 26.9 0.7 34.2 3.1 27.8 6.0 79.2*¥ 8.0 303.7 5.0 135.1*¥ 23.7 C18 75.2 9.3 86.4 5.5 68.0 9.3 355.2*¥ 39.0 523.0 12.3 501.6 89.1

C18:1 156.4 15.2 202.1 4.6 227.5 46.2 1577.3*¥ 213.0 2703.3 296.7 2259.8 347.1 Long Chain Long C18:2 8.9 1.1 14.4* 1.7 12.1 1.0 69.1*¥ 8.5 226.0 24.1 135.7 22.3 C18:2-OH 3.7 0.6 5.4 1.2 5.1 0.9 16.0*¥ 0.7 30.1 1.7 16.2* 1.9 C18:1- OH/C16:1- 26.5 1.4 31.5 2.0 33.6 5.5 93.3*¥ 8.6 403.0 26.3 150.4*¥ 26.4 DC C18- OH/C16-DC 21.1 1.3 28.2* 1.7 20.7 0.4 26.0 1.9 83.9 10.8 36.6 ¥ 6.3 C18:1-DC 4.0 0.6 3.5 1.9 4.6 0.5 5.0 0.5 19.0 4.6 6.1 ¥ 0.8 C18- DC/C20-OH 3.7 1.2 4.5 0.8 0.8 0.2 2.5 ¥ 0.7 4.2 1.1 3.1 0.4

146

C20:4 6.5 0.7 11.3 0.9 1.9 0.2 3.6 0.7 1.1 0.3 1.4 0.5 C20 2.7 0.3 3.2 1.6 2.7 0.2 14.4*¥ 0.5 10.9 2.2 9.9 2.0

VLC C22 1.6 0.3 2.4 0.5 0.8 0.1 13.1*¥ 2.4 0.9 0.3 7.9 2.7 Glycine 26937.3 337.8 23060.4 1116.0 34369.3 866.0 40766.7 1619.8 28069.5 3094.7 22719.4 641.8 Alanine 22780.0 2436.7 20395.9 592.9 18780.4 2099.0 20723.9 553.8 20053.8 2974.2 15240.3 1943.4 Serine 40501.3 2704.0 40163.0 1665.0 39177.2 3244.6 41999.7 404.2 40540.6 845.5 32299.8 7197.3 Proline 5373.3 306.3 4857.6 275.4 3069.9 48.1 4028.5* 77.1 5518.2 320.8 4548.3 805.5 Valine 18322.2 1096.6 16393.3 1501.8 9388.8 380.6 10574.5 374.8 11947.0 1234.0 11918.4 785.1

Leucine/Isoleu 42023.9 3398.8 40087.3 1058.3 25255.9 2222.6 24466.9 724.3 25260.8 3514.3 20951.6 1967.5 cine Methionine 10939.5 1154.6 9983.3 349.5 6547.5 813.3 6002.4 304.4 5686.5 985.0 4993.7 320.7 Histidine 7323.2 538.7 5795.5 285.3 4038.4 271.9 3505.1 138.9 5020.3 664.6 3849.0 329.5 Phenylalanine 21346.8 1051.4 21030.8 765.8 13043.1 1024.7 12401.9 603.0 11771.6 1877.0 9670.4 1352.2 Tyrosine 16416.5 926.9 15259.4 1079.7 9130.7 605.7 9203.2 292.5 9782.2 1559.9 8892.4 1093.7 Asparagine/As Amino Acids Amino 28570.7 1610.4 32727.6 1899.8 23653.4 550.9 14344.5* 260.4 11309.0 1580.1 8974.6 1583.6 partic Acid Glutamine/Glut 90681.7 4034.3 102810.7 3863.1 121380.9 5735.4 125783.6 5520.4 88965.9 6143.1 77478.5 8829.4 amic Acid Ornithine 1175.9 48.1 924.6 143.3 538.8 30.8 536.7 34.5 852.2 98.2 583.9 89.6 35756.1* 656.2 149.5 637.3 86.4 538.8 30.8 3796.7*¥ 552.4 67897.8 5832.8 4890.9 Citrulline ¥ Arginine 14682.2 1014.4 13200.9 564.3 5891.0 597.0 5234.2 238.0 2726.8 490.3 3953.4 771.4 102209. 147392.8 49830. 10883. 7026.7 86473.2 14121.9 94116.7 1502.7 9008.8 337724.1 266825.0 Lactate 5 * 3 0

ids Pyruvate 8261.6 1872.6 6340.3 1427.0 3706.4 988.2 3438.2 251.9 3700.3 1241.5 3785.2 1047.2 Succinate 2209.5 193.1 1546.4 187.4 5071.4 442.6 6078.5 227.8 2772.5 249.8 3129.0 259.5 Fumarate 840.9 94.7 1030.5 52.1 764.8 167.2 1446.3 400.7 2934.0 506.9 2217.4* 461.8 Malate 3558.5 67.8 3826.1 249.1 3789.5 216.3 4706.2 339.6 9330.5 399.7 7081.9 1008.9 α- 823.8 38.2 792.3 134.9 941.1 201.4 1449.7 2.1 1604.7 291.9 962.1 101.2

Ketoglutarate Organic Ac Organic Citrate 8947.3 586.8 9006.7 296.7 6996.7 160.9 3767.8*¥ 60.0 6829.0 1473.2 3149.6 ¥ 253.8

147

Appendix C

Serum cytokine concentrations in uninfected, or S. typhimurium infected, WT and Irgm1- deficient mice measured by luminex multiplex assay. Units: Fluorescence/ mg Protein serum cytokine concentration

Control Two Day S.typhimurium Infection

WT Irgm1-deficient WT Irgm1-deficient

Meanɤ SEM Meanɤ SEM Mean SEM Meanɤ SEM ɤ Pro- IL-1α 206.2 31.7 413.3¥* 33.0 426.7 66.8 414.7 96.5 inflammatory IL-1β 275.0 38.8 618.1¥ 70.1 226.9 48.3 373.3 98.3 Cytokines ¥ ¥ IL-2 8.2 4.1 94.2 52.1 48.6 22.0 227.2 224.6 IL-3 28.6 26.0 421.9¥* 57.8 128.3 48.6 211.9 93.1 IL-5 5.4 2.1 27.6¥* 6.5 4.6 1.6 18.1* 4.5 IL-6 37.3 13.6 671.6¥ 412.3 158.4 44.8 208.1* 61.1 IL-9 274.7 73.2 535.2 230.5 216.6 99.2 679.9¥ 585.8 IL-12 489.5 158.8 1365.8¥* 103.8 1041.3 200.2 589.1 125.5 (p40) IL-12 161.2 69.4 418.0¥ 167.5 441.8 150.1 494.4 177.2 (p70) IL-17 17.7 9.7 120.7¥* 23.5 35.0 14.6 89.0¥ 31.1 G-CSF 68.5 6.4 322.7¥* 93.8 282.9 39.9 757.4¥ 398.5 GM-CSF 10.8 2.5 186.5¥ 92.5 76.6 54.4 133.8 125.9 IFN-γ 19.6 11.7 886.6¥ 686.1 544.7 292.8 902.8 923.1 TNF-α 517.5 232.0 6136.2¥ 2273.0 1426.0 612.1 2754.2¥* 828.7 Anti- IL-4 3.9 3.1 7.4 3.3 5.5 3.8 2.4 0.5 inflammatory IL-10 137.4 110.0 382.8¥ 188.0 303.0 172.7 273.2 52.3 Cytokines IL-13 722.1 414.0 1754.8¥ 352.9 1589.7 472.2 2640.3 1476.0 Chemokines Eotaxin/ 2316.9 893.5 6604.5¥ 1906.2 5291.6 3688.5 4370.8 1923.7 CCL11 KC/ 13.9 4.5 115.5¥ 68.6 87.6 62.3 138.7 134.1 CXCL1 MCP-1/ 43.5 13.9 102.9¥* 17.4 483.2 129.0 1622.0¥ 628.6 CCL2 MIP- 33.5 33.5 1087.8¥ 974.2 130.6 23.5 498.0¥ 410.4

148

1α/CCL3 MIP-1β/ 141.5 103.8 1264.5¥* 126.2 384.9 98.0 836.6¥* 223.1 CCL4 RANTES/ 473.6 236.1 1648.8¥* 71.1 1040.0 139.2 1162.1 86.8 CCL5

ɤ Fluorescence/ mg Protein serum cytokine concentration ¥ Two-fold change in cytokine concentration compared to WT * p<0.05 compared to WT

149

Reference

1. Isaacs, A., and Lindenmann, J. (1957) Virus interference. I. The interferon. Proc R Soc Lond B Biol Sci 147, 258-267

2. Isaacs, A., Lindenmann, J., and Valentine, R. C. (1957) Virus interference. II. Some properties of interferon. Proc R Soc Lond B Biol Sci 147, 268-273

3. Pestka, S. (2007) The interferons: 50 years after their discovery, there is much more to learn. J Biol Chem 282, 20047-20051

4. Pestka S, K. C., Walter MR. (2004) Interferons, interferon-like cytokines, and their receptors. Immunological Reviews 202, 8-32

5. Krause, C. D., and Pestka, S. (2005) Evolution of the Class 2 cytokines and receptors, and discovery of new friends and relatives. Pharmacol Ther 106, 299-346

6. McNab, F., Mayer-Barber, K., Sher, A., Wack, A., and O'Garra, A. (2015) Type I interferons in infectious disease. Nat Rev Immunol 15, 87-103

7. Goodbourn S, D. L., Randall RE. (2000) Interferons- cell signalling, immune modulation, antiviral response and virus countermeasures. J Gen Virol 81, 2341- 2364

8. Schroder, K., Hertzog, P. J., Ravasi, T., and Hume, D. A. (2004) Interferon- gamma: an overview of signals, mechanisms and functions. J Leukoc Biol 75, 163- 189

9. Ivashkiv, L. B., and Donlin, L. T. (2014) Regulation of type I interferon responses. Nat Rev Immunol 14, 36-49

10. MacMicking, J. D. (2012) Interferon-inducible effector mechanisms in cell- autonomous immunity. Nat Rev Immunol 12, 367-382

11. Ozato, K., Tsujimura, H., and Tamura, T. (2002) Toll-like receptor signaling and regulation of cytokine gene expression in the immune system. Biotechniques Suppl, 66-68, 70, 72 passim

12. Darnell JE Jr, K. I., Stark GR. (1994) Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264, 1415-1421

150

13. Platanias, L. C. (2005) Mechanisms of type-I- and type-II-interferon-mediated signalling. Nat Rev Immunol 5, 375-386

14. Silvennoinen O, I. J., Schlessinger J, Levy DE. (1993) Interferon-induced nuclear signalling by Jak protein tyrosine kinases. Nature 366, 583-585

15. Decker, T., Muller, M., and Stockinger, S. (2005) The yin and yang of type I interferon activity in bacterial infection. Nat Rev Immunol 5, 675-687

16. Aaronson DS, H. C. (2002) A road map for those who don't know JAK-STAT. Science 296, 1653-1655

17. Fu XY, S. C., Improta T, Aebersold R, Darnell JE Jr. (1992) The proteins of ISGF-3, the interferon alpha-induced transcriptional activator, define a gene family involved in signal transduction. Proc Natl Acad Sci USA 89, 7840-7843

18. Schindler, C., Shuai, K., Prezioso, V., and Darnell, J. (1992) Interferon-dependent tyrosine phosphorylation of a latent cytoplasmic transcription factor. Science 257, 809-813

19. Jr, D. J. (1997) STATs and Gene Regulation. Science 277, 1630-1635

20. Abdul-Sater, A. A., Majoros, A., Plumlee, C. R., Perry, S., Gu, A. D., Lee, C., Shresta, S., Decker, T., and Schindler, C. (2015) Different STAT Transcription Complexes Drive Early and Delayed Responses to Type I IFNs. J Immunol 195, 210-216

21. Gonzalez-Navajas, J. M., Lee, J., David, M., and Raz, E. (2012) Immunomodulatory functions of type I interferons. Nat Rev Immunol 12, 125-135

22. Farrar JD, S. J., Murphy TL, Murphy KM. (2000) Recruitment of Stat4 to the human interferon-alpha-beta receptor requires activated Stat2. J Biol Chem 275, 2693-2697

23. Torpey, N., Maher, S. E., Bothwell, A. L., and Pober, J. S. (2004) Interferon alpha but not interleukin 12 activates STAT4 signaling in human vascular endothelial cells. J Biol Chem 279, 26789-26796

24. Nathan CF, M. H., Wiebe ME, Rubin BY. (1983) Identification of interferon- gamma as the lymphokine that activates human macrophage oxidative metabolism and antimicrobial activity. J Exp Med 158, 670-689

151

25. Hu, X., and Ivashkiv, L. B. (2009) Cross-regulation of signaling pathways by interferon-gamma: implications for immune responses and autoimmune diseases. Immunity 31, 539-550

26. Schoenborn, J. R., and Wilson, C. B. (2007) Regulation of Interferon‐γ During Innate and Adaptive Immune Responses. 96, 41-101

27. Bach EA, A. M., Schreiber RD. (1997) The IFN gamma receptor- a paradigm for cytokine receptor signaling. Annu Rev Immunol 15, 563-591

28. Pestka S, K. S., Muthukumaran G, Izotova LS, Cook JR, Garotta G. (1997) The interferon gamma (IFN-gamma) receptor- a paradigm for the multichain cytokine receptor. Cytokine Growth Factor Rev 8, 189-206

29. Watling D, G. D., Müller M, Silvennoinen O, Witthuhn BA, Quelle FW, Rogers NC, Schindler C, Stark GR, Ihle JN, et al. (1993) Complementation by the protein tyrosine kinase JAK2 of a mutant cell line defective in the interferon-gamma signal transduction pathway. Nature 366, 166-170

30. Boehm, U., Klamp, T., Groot, M., and Howard, J. C. (1997) Cellular responses to interferon-gamma. Annu Rev Immunol 15, 749-795

31. Shuai K, S. C., Prezioso VR, Darnell JE Jr. (1992) Activation of transcription by IFN-gamma- tyrosine phosphorylation of a 91-kD DNA binding protein. Science 258, 1808-1812

32. Au-Yeung, N., Mandhana, R., and Horvath, C. M. (2013) Transcriptional regulation by STAT1 and STAT2 in the interferon JAK-STAT pathway. JAKSTAT 2, e23931

33. Ehrt S, S. D., Bekiranov S, Drenkow J, Shi S, Gingeras TR, Gaasterland T, Schoolnik G, Nathan C. (2001) Reprogramming of the macrophage transcriptome in response to interferon-gamma and Mycobacterium tuberculosis- signaling roles of nitric oxide synthase-2 and phagocyte oxidase. J Exp Med 194, 1123-1140

34. de Veer, M. J., Holko, M., Frevel, M., Walker, E., Der, S., Paranjape, J. M., Silverman, R. H., and Williams, B. R. (2001) Functional classification of interferon-stimulated genes identified using microarrays. J Leukoc Biol 69, 912-920

35. Kota, R. S., Rutledge, J. C., Gohil, K., Kumar, A., Enelow, R. I., and Ramana, C. V. (2006) Regulation of gene expression in RAW 264.7 macrophage cell line by interferon-gamma. Biochem Biophys Res Commun 342, 1137-1146

152

36. Stuehr, D. J., and Marletta, M. A. (1987) Induction of nitrite/nitrate synthesis in murine macrophages by BCG infection, lymphokines, or interferon-gamma. J Immunol 139, 518-525

37. Maldonado, R. A., Irvine, D. J., Schreiber, R., and Glimcher, L. H. (2004) A role for the immunological synapse in lineage commitment of CD4 lymphocytes. Nature 431, 527-532

38. Rosa, F., and Fellous, M. (1984) The effect of gamma-interferon on MHC antigens. Immunol Today 5, 261-262

39. Su, X., Yu, Y., Zhong, Y., Giannopoulou, E. G., Hu, X., Liu, H., Cross, J. R., Ratsch, G., Rice, C. M., and Ivashkiv, L. B. (2015) Interferon-gamma regulates cellular metabolism and mRNA translation to potentiate macrophage activation. Nat Immunol 16, 838-849

40. Kelly, B., and O'Neill, L. A. (2015) Metabolic reprogramming in macrophages and dendritic cells in innate immunity. Cell Res 25, 771-784

41. Qiao, Y., Giannopoulou, E. G., Chan, C. H., Park, S. H., Gong, S., Chen, J., Hu, X., Elemento, O., and Ivashkiv, L. B. (2013) Synergistic activation of inflammatory cytokine genes by interferon-gamma-induced chromatin remodeling and toll-like receptor signaling. Immunity 39, 454-469

42. Kim, B. H., Shenoy, A. R., Kumar, P., Bradfield, C. J., and MacMicking, J. D. (2012) IFN-inducible GTPases in host cell defense. Cell Host Microbe 12, 432-444

43. Yoon Y, P. K., McNiven MA. (2001) Mammalian dynamin-like protein DLP1 tubulates membranes. Mol Biol Cell 12, 2894-2905

44. van der Bliek AM, R. T., Damke H, Tisdale EJ, Meyerowitz EM, Schmid SL. (1993) Mutations in human dynamin block an intermediate stage in coated vesicle formation. J Cell Biol 122, 553-563

45. Jones SM, H. K., Henley JR, Cao H, McNiven MA. (1998) Role of dynamin in the formation of transport vesicles from the trans-Golgi network. Science 279, 573-577

46. Nicoziani P, V. F., Llorente A, Hilout L, Courtoy PJ, Sandvig K, van Deurs B. (2000) Role for Dynamin in Late Endosome Dynamics and Trafficking of the Cation-independent Mannose 6-Phosphate Receptor. Mol Biol Cell 11, 481-495

153

47. McNiven MA, C. H., Pitts KR, Yoon Y. (2000) The dynamin family of mechanoenzymes: pinching in new places. Trends Biochem Sci 25, 115-120

48. Praefcke, G. J., and McMahon, H. T. (2004) The dynamin superfamily: universal membrane tubulation and fission molecules? Nat Rev Mol Cell Biol 5, 133-147

49. Roux, A., Uyhazi, K., Frost, A., and De Camilli, P. (2006) GTP-dependent twisting of dynamin implicates constriction and tension in membrane fission. Nature 441, 528-531

50. Morlot, S., and Roux, A. (2013) Mechanics of dynamin-mediated membrane fission. Annu Rev Biophys 42, 629-649

51. Boehm U, G. L., Klamp T, Ozbek K, Schaub A, Fütterer A, Pfeffer K, Howard JC. (1998) Two families of GTPases dominate the complex cellular response to IFN- gamma. J Immunol 161, 6715-6723

52. Ghosh, A., Uthaiah, R., Howard, J., Herrmann, C., and Wolf, E. (2004) Crystal structure of IIGP1: a paradigm for interferon-inducible p47 resistance GTPases. Mol Cell 15, 727-739

53. Prakash B, P. G., Renault L, Wittinghofer A, Herrmann C. (2000) Structure of human guanylate-binding protein 1 representing a unique class of GTP-binding proteins. Nature 403, 567

54. Schafer, D. A. (2004) Regulating actin dynamics at membranes: a focus on dynamin. Traffic 5, 463-469

55. Danino D, H. J. (2001) Dynamin family of mechanoenzymes. Curr Opin Cell Biol 13, 454-460

56. Sever S, M. A., Schmid SL. (1999) Impairment of dynamin's GAP domain stimulates receptor-mediated endocytosis. Nature 398, 481-486

57. Cherfils, J., and Zeghouf, M. (2013) Regulation of small GTPases by GEFs, GAPs, and GDIs. Physiol Rev 93, 269-309

58. Marks B, S. M., Vallis Y, Mills IG, Gibson A, Hopkins CR, McMahon HT. (2001) GTPase activity of dynamin and resulting conformation change are essential for endocytosis. Nature 410, 231-235

154

59. Hunn JP, K.-W. S., Papic N, Schroeder N, Pawlowski N, Lange R, Kaiser F, Zerrahn J, Martens S, Howard JC. (2008) Regulatory interactions between IRG resistance GTPases in the cellular response to Toxoplasma gondii. EMBO J 27, 2495-2509

60. Haller, O., Staeheli, P., Schwemmle, M., and Kochs, G. (2015) Mx GTPases: dynamin-like antiviral machines of innate immunity. Trends Microbiol 23, 154-163

61. Holzinger, D., Jorns, C., Stertz, S., Boisson-Dupuis, S., Thimme, R., Weidmann, M., Casanova, J. L., Haller, O., and Kochs, G. (2007) Induction of MxA gene expression by influenza A virus requires type I or type III interferon signaling. J Virol 81, 7776-7785

62. Pilla-Moffett, D., Barber, M. F., Taylor, G. A., and Coers, J. (2016) Interferon- Inducible GTPases in Host Resistance, Inflammation and Disease. J Mol Biol 428, 3495-3513

63. Lipps, J. H., Waggoner, B. M., America, G. S. o., and Society, P. (2004) Neoproterozoic-Cambrian Biological Revolutions: Presented as a Paleontological Society Short Course at the Annual Meeting of the Geological Society of America, Denver, Colorado, November 6, 2004, Yale University Reprographics & Imaging Service

64. Li, G., Zhang, J., Sun, Y., Wang, H., and Wang, Y. (2009) The evolutionarily dynamic IFN-inducible GTPase proteins play conserved immune functions in vertebrates and cephalochordates. Mol Biol Evol 26, 1619-1630

65. Bekpen C, H. J., Rohde C, Parvanova I, Guethlein L, Dunn DM, Glowalla E, Leptin M, Howard JC. (2005) The interferon-inducible p47 (IRG) GTPases in vertebrates- loss of the cell autonomous resistance mechanism in the human lineage. Genome Biol 6

66. Robertsen, B. (2006) The interferon system of teleost fish. Fish Shellfish Immunol 20, 172-191

67. Bekpen, C., Marques-Bonet, T., Alkan, C., Antonacci, F., Leogrande, M. B., Ventura, M., Kidd, J. M., Siswara, P., Howard, J. C., and Eichler, E. E. (2009) Death and resurrection of the human IRGM gene. PLoS Genet 5, e1000403

68. Pilla-Moffett, D. (2015) Mechanisms by Which Guanylate Binding Proteins Target Pathogen Vacuoles and Promote Caspase-11 Dependent Pyroptosis. Doctorate of Philosophy Doctoral, Duke University

155

69. Uthaiah, R. C., Praefcke, G. J., Howard, J. C., and Herrmann, C. (2003) IIGP1, an interferon-gamma-inducible 47-kDa GTPase of the mouse, showing cooperative enzymatic activity and GTP-dependent multimerization. J Biol Chem 278, 29336- 29343

70. Taylor GA, S. R., Rulong S, Hudson E, Pei V, Pavlakis GN, Resau JH, Vande Woude GF. (1997) The Inducibly Expressed GTPase Localizes to the Endoplasmic Reticulum, Independently of GTP Binding. J Biol Chem 272, 10639-10645

71. MK, T. (2013) The Effects of GMS Immunity-Related GTPases on Guanylate-Binding Proteins, Protein Aggregate Formation, and Macroautophagy. Doctoral Thesis Doctoral thesis, Duke University

72. Lilue, J., Muller, U. B., Steinfeldt, T., and Howard, J. C. (2013) Reciprocal virulence and resistance polymorphism in the relationship between Toxoplasma gondii and the house mouse. Elife 2, e01298

73. Martens, S., Sabel, K., Lange, R., Uthaiah, R., Wolf, E., and Howard, J. C. (2004) Mechanisms Regulating the Positioning of Mouse p47 Resistance GTPases LRG- 47 and IIGP1 on Cellular Membranes: Retargeting to Plasma Membrane Induced by Phagocytosis. The Journal of Immunology 173, 2594-2606

74. Haldar, A. K., Saka, H. A., Piro, A. S., Dunn, J. D., Henry, S. C., Taylor, G. A., Frickel, E. M., Valdivia, R. H., and Coers, J. (2013) IRG and GBP host resistance factors target aberrant, "non-self" vacuoles characterized by the missing of "self" IRGM proteins. PLoS Pathog 9, e1003414

75. Henry, S. C., Schmidt, E. A., Fessler, M. B., and Taylor, G. A. (2014) Palmitoylation of the immunity related GTPase, Irgm1: impact on membrane localization and ability to promote mitochondrial fission. PLoS One 9, e95021

76. Tiwari, S., Choi, H. P., Matsuzawa, T., Pypaert, M., and MacMicking, J. D. (2009) Targeting of the GTPase Irgm1 to the phagosomal membrane via PtdIns(3,4)P(2) and PtdIns(3,4,5)P(3) promotes immunity to mycobacteria. Nat Immunol 10, 907- 917

77. Zerrahn, J., Schaible, U. E., Brinkmann, V., Guhlich, U., and Kaufmann, S. H. E. (2002) The IFN-Inducible Golgi- and Endoplasmic Reticulum- Associated 47-kDa GTPase IIGP Is Transiently Expressed During Listeriosis. The Journal of Immunology 168, 3428-3436

156

78. Henry, S. C., Daniell, X. G., Burroughs, A. R., Indaram, M., Howell, D. N., Coers, J., Starnbach, M. N., Hunn, J. P., Howard, J. C., Feng, C. G., Sher, A., and Taylor, G. A. (2009) Balance of Irgm protein activities determines IFN-gamma-induced host defense. J Leukoc Biol 85, 877-885

79. Al-Zeer, M. A., Al-Younes, H. M., Braun, P. R., Zerrahn, J., and Meyer, T. F. (2009) IFN-gamma-inducible Irga6 mediates host resistance against Chlamydia trachomatis via autophagy. PLoS One 4, e4588

80. MacMicking, J. D., Taylor, G. A., and McKinney, J. D. (2003) Immune control of tuberculosis by IFN-gamma-inducible LRG-47. Science 302, 654-659

81. Springer, H. M., Schramm, M., Taylor, G. A., and Howard, J. C. (2013) Irgm1 (LRG-47), a regulator of cell-autonomous immunity, does not localize to mycobacterial or listerial phagosomes in IFN-gamma-induced mouse cells. J Immunol 191, 1765-1774

82. Butcher, B. A., Greene, R. I., Henry, S. C., Annecharico, K. L., Weinberg, J. B., Denkers, E. Y., Sher, A., and Taylor, G. A. (2005) p47 GTPases regulate Toxoplasma gondii survival in activated macrophages. Infect Immun 73, 3278- 3286

83. Martens, S., Parvanova, I., Zerrahn, J., Griffiths, G., Schell, G., Reichmann, G., and Howard, J. C. (2005) Disruption of Toxoplasma gondii parasitophorous vacuoles by the mouse p47-resistance GTPases. PLoS Pathog 1, e24

84. Melzer, T., Duffy, A., Weiss, L. M., and Halonen, S. K. (2008) The gamma interferon (IFN-gamma)-inducible GTP-binding protein IGTP is necessary for toxoplasma vacuolar disruption and induces parasite egression in IFN-gamma- stimulated astrocytes. Infect Immun 76, 4883-4894

85. Halonen, S. K., Taylor, G. A., and Weiss, L. M. (2001) Gamma interferon-induced inhibition of Toxoplasma gondii in astrocytes is mediated by IGTP. Infect Immun 69, 5573-5576

86. Khaminets, A., Hunn, J. P., Konen-Waisman, S., Zhao, Y. O., Preukschat, D., Coers, J., Boyle, J. P., Ong, Y. C., Boothroyd, J. C., Reichmann, G., and Howard, J. C. (2010) Coordinated loading of IRG resistance GTPases on to the Toxoplasma gondii parasitophorous vacuole. Cell Microbiol 12, 939-961

87. Taylor, G. A., Collazo, C. M., Yap, G. S., Nguyen, K., Gregorio, T. A., Taylor, L. S., Eagleson, B., Secrest, L., Southon, E. A., Reid, S. W., Tessarollo, L., Bray, M., 157

McVicar, D. W., Komschlies, K. L., Young, H. A., Biron, C. A., Sher, A., and Vande Woude, G. F. (2000) Pathogen-specific loss of host resistance in mice lacking the IFN-gamma-inducible gene IGTP. Proc Natl Acad Sci U S A 97, 751- 755

88. Collazo, C. M., Yap, G. S., Sempowski, G. D., Lusby, K. C., Tessarollo, L., Vande Woude, G. F., Sher, A., and Taylor, G. A. (2001) Inactivation of LRG-47 and IRG- 47 reveals a family of interferon gamma-inducible genes with essential, pathogen-specific roles in resistance to infection. J Exp Med 194, 181-188

89. Hunn, J. P., Feng, C. G., Sher, A., and Howard, J. C. (2011) The immunity-related GTPases in mammals: a fast-evolving cell-autonomous resistance system against intracellular pathogens. Mamm Genome 22, 43-54

90. Taylor, G. A., Feng, C. G., and Sher, A. (2004) p47 GTPases: regulators of immunity to intracellular pathogens. Nat Rev Immunol 4, 100-109

91. Henry, S. C., Daniell, X., Indaram, M., Whitesides, J. F., Sempowski, G. D., Howell, D., Oliver, T., and Taylor, G. A. (2007) Impaired Macrophage Function Underscores Susceptibility to Salmonella in Mice Lacking Irgm1 (LRG-47). The Journal of Immunology 179, 6963-6972

92. Coers, J., Bernstein-Hanley, I., Grotsky, D., Parvanova, I., Howard, J. C., Taylor, G. A., Dietrich, W. F., and Starnbach, M. N. (2008) Chlamydia muridarum evades growth restriction by the IFN-gamma-inducible host resistance factor Irgb10. J Immunol 180, 6237-6245

93. Feng, C. G., Collazo-Custodio, C. M., Eckhaus, M., Hieny, S., Belkaid, Y., Elkins, K., Jankovic, D., Taylor, G. A., and Sher, A. (2004) Mice Deficient in LRG-47 Display Increased Susceptibility to Mycobacterial Infection Associated with the Induction of Lymphopenia. The Journal of Immunology 172, 1163-1168

94. Liu, Z., Zhang, H. M., Yuan, J., Ye, X., Taylor, G. A., and Yang, D. (2012) The immunity-related GTPase Irgm3 relieves endoplasmic reticulum stress response during coxsackievirus B3 infection via a PI3K/Akt dependent pathway. Cell Microbiol 14, 133-146

95. Haldar, A. K., Foltz, C., Finethy, R., Piro, A. S., Feeley, E. M., Pilla-Moffett, D. M., Komatsu, M., Frickel, E. M., and Coers, J. (2015) Ubiquitin systems mark pathogen-containing vacuoles as targets for host defense by guanylate binding proteins. Proc Natl Acad Sci U S A 112, E5628-5637

158

96. Maric-Biresev, J., Hunn, J. P., Krut, O., Helms, J. B., Martens, S., and Howard, J. C. (2016) Loss of the interferon-gamma-inducible regulatory immunity-related GTPase (IRG), Irgm1, causes activation of effector IRG proteins on lysosomes, damaging lysosomal function and predicting the dramatic susceptibility of Irgm1-deficient mice to infection. BMC Biol 14, 33

97. Traver, M. K., Henry, S. C., Cantillana, V., Oliver, T., Hunn, J. P., Howard, J. C., Beer, S., Pfeffer, K., Coers, J., and Taylor, G. A. (2011) Immunity-related GTPase M (IRGM) proteins influence the localization of guanylate-binding protein 2 (GBP2) by modulating macroautophagy. J Biol Chem 286, 30471-30480

98. Shibata, M., Yoshimura, K., Furuya, N., Koike, M., Ueno, T., Komatsu, M., Arai, H., Tanaka, K., Kominami, E., and Uchiyama, Y. (2009) The MAP1-LC3 conjugation system is involved in lipid droplet formation. Biochem Biophys Res Commun 382, 419-423

99. Choi, J., Park, S., Biering, S. B., Selleck, E., Liu, C. Y., Zhang, X., Fujita, N., Saitoh, T., Akira, S., Yoshimori, T., Sibley, L. D., Hwang, S., and Virgin, H. W. (2014) The parasitophorous vacuole membrane of Toxoplasma gondii is targeted for disruption by ubiquitin-like conjugation systems of autophagy. Immunity 40, 924- 935

100. Haldar, A. K., Piro, A. S., Pilla, D. M., Yamamoto, M., and Coers, J. (2014) The E2- like conjugation enzyme Atg3 promotes binding of IRG and Gbp proteins to Chlamydia- and Toxoplasma-containing vacuoles and host resistance. PLoS One 9, e86684

101. Zhao, Y. O., Khaminets, A., Hunn, J. P., and Howard, J. C. (2009) Disruption of the Toxoplasma gondii parasitophorous vacuole by IFNgamma-inducible immunity-related GTPases (IRG proteins) triggers necrotic cell death. PLoS Pathog 5, e1000288

102. Ling, Y. M., Shaw, M. H., Ayala, C., Coppens, I., Taylor, G. A., Ferguson, D. J., and Yap, G. S. (2006) Vacuolar and plasma membrane stripping and autophagic elimination of Toxoplasma gondii in primed effector macrophages. J Exp Med 203, 2063-2071

103. Yamamoto, M., Okuyama, M., Ma, J. S., Kimura, T., Kamiyama, N., Saiga, H., Ohshima, J., Sasai, M., Kayama, H., Okamoto, T., Huang, D. C., Soldati-Favre, D., Horie, K., Takeda, J., and Takeda, K. (2012) A cluster of interferon-gamma-

159

inducible p65 GTPases plays a critical role in host defense against Toxoplasma gondii. Immunity 37, 302-313

104. Ferguson, S. M., and De Camilli, P. (2012) Dynamin, a membrane-remodelling GTPase. Nat Rev Mol Cell Biol 13, 75-88

105. Bekpen, C., Xavier, R. J., and Eichler, E. E. (2010) Human IRGM gene "to be or not to be". Semin Immunopathol 32, 437-444

106. Zhao, Y. O., Konen-Waisman, S., Taylor, G. A., Martens, S., and Howard, J. C. (2010) Localisation and mislocalisation of the interferon-inducible immunity- related GTPase, Irgm1 (LRG-47) in mouse cells. PLoS One 5, e8648

107. Chang, C. P., Yang, M. C., and Lei, H. Y. (2011) Concanavalin A/IFN-gamma triggers autophagy-related necrotic hepatocyte death through IRGM1-mediated lysosomal membrane disruption. PLoS One 6, e28323

108. Singh, S. B., Davis, A. S., Taylor, G. A., and Deretic, V. (2006) Human IRGM induces autophagy to eliminate intracellular mycobacteria. Science 313, 1438-1441

109. Howard, J. (2008) The IRG proteins: a function in search of a mechanism. Immunobiology 213, 367-375

110. Haldar, A. K., Piro, A. S., Finethy, R., Espenschied, S. T., Brown, H. E., Giebel, A. M., Frickel, E. M., Nelson, D. E., and Coers, J. (2016) Chlamydia trachomatis Is Resistant to Inclusion Ubiquitination and Associated Host Defense in Gamma Interferon-Primed Human Epithelial Cells. MBio 7

111. Ferreira-da-Silva Mda, F., Springer-Frauenhoff, H. M., Bohne, W., and Howard, J. C. (2014) Identification of the microsporidian Encephalitozoon cuniculi as a new target of the IFNgamma-inducible IRG resistance system. PLoS Pathog 10, e1004449

112. Santiago, H. C., Feng, C. G., Bafica, A., Roffe, E., Arantes, R. M., Cheever, A., Taylor, G., Vieira, L. Q., Aliberti, J., Gazzinelli, R. T., and Sher, A. (2005) Mice deficient in LRG-47 display enhanced susceptibility to Trypanosoma cruzi infection associated with defective hemopoiesis and intracellular control of parasite growth. J Immunol 175, 8165-8172

113. Sacks, D., and Sher, A. (2002) Evasion of innate immunity by parasitic protozoa. Nat Immunol 3, 1041-1047

160

114. Bafica, A., Feng, C. G., Santiago, H. C., Aliberti, J., Cheever, A., Thomas, K. E., Taylor, G. A., Vogel, S. N., and Sher, A. (2007) The IFN-inducible GTPase LRG47 (Irgm1) negatively regulates TLR4-triggered proinflammatory cytokine production and prevents endotoxemia. J Immunol 179, 5514-5522

115. Deretic, V., Master, S., and Singh, S. (2008) Autophagy gives a nod and a wink to the inflammasome and Paneth cells in Crohn's disease. Dev Cell 15, 641-642

116. Liu, B., Gulati, A. S., Cantillana, V., Henry, S. C., Schmidt, E. A., Daniell, X., Grossniklaus, E., Schoenborn, A. A., Sartor, R. B., and Taylor, G. A. (2013) Irgm1- deficient mice exhibit Paneth cell abnormalities and increased susceptibility to acute intestinal inflammation. Am J Physiol Gastrointest Liver Physiol 305, G573-584

117. Xu, H., Wu, Z. Y., Fang, F., Guo, L., Chen, D., Chen, J. X., Stern, D., Taylor, G. A., Jiang, H., and Yan, S. S. (2010) Genetic deficiency of Irgm1 (LRG-47) suppresses induction of experimental autoimmune encephalomyelitis by promoting apoptosis of activated CD4+ T cells. FASEB J 24, 1583-1592

118. Wang, C., Wang, C., Dong, H., Wu, X. M., Wang, C., Xia, F., Li, G., Jia, X., He, S., Jiang, X., Li, H., and Xu, H. (2013) Immune-related GTPase Irgm1 exacerbates experimental auto-immune encephalomyelitis by promoting the disruption of blood-brain barrier and blood-cerebrospinal fluid barrier. Mol Immunol 53, 43-51

119. He, S., Wang, C., Dong, H., Xia, F., Zhou, H., Jiang, X., Pei, C., Ren, H., Li, H., Li, R., and Xu, H. (2012) Immune-related GTPase M (IRGM1) regulates neuronal autophagy in a mouse model of stroke. Autophagy 8, 1621-1627

120. Singh, S. B., Ornatowski, W., Vergne, I., Naylor, J., Delgado, M., Roberts, E., Ponpuak, M., Master, S., Pilli, M., White, E., Komatsu, M., and Deretic, V. (2010) Human IRGM regulates autophagy and cell-autonomous immunity functions through mitochondria. Nat Cell Biol 12, 1154-1165

121. Feng, C. G., Zheng, L., Jankovic, D., Bafica, A., Cannons, J. L., Watford, W. T., Chaussabel, D., Hieny, S., Caspar, P., Schwartzberg, P. L., Lenardo, M. J., and Sher, A. (2008) The immunity-related GTPase Irgm1 promotes the expansion of activated CD4+ T cell populations by preventing interferon-gamma-induced cell death. Nat Immunol 9, 1279-1287

122. King, K. Y., Baldridge, M. T., Weksberg, D. C., Chambers, S. M., Lukov, G. L., Wu, S., Boles, N. C., Jung, S. Y., Qin, J., Liu, D., Songyang, Z., Eissa, N. T., Taylor, G. A., and Goodell, M. A. (2011) Irgm1 protects hematopoietic stem cells by negative regulation of IFN signaling. Blood 118, 1525-1533 161

123. Feng, C. G., Weksberg, D. C., Taylor, G. A., Sher, A., and Goodell, M. A. (2008) The p47 GTPase Lrg-47 (Irgm1) links host defense and hematopoietic stem cell proliferation. Cell Stem Cell 2, 83-89

124. Chen, C., Liu, Y., Liu, Y., and Zheng, P. (2010) Mammalian target of rapamycin activation underlies HSC defects in autoimmune disease and inflammation in mice. J Clin Invest 120, 4091-4101

125. Hartner, J. C., Walkley, C. R., Lu, J., and Orkin, S. H. (2009) ADAR1 is essential for the maintenance of hematopoiesis and suppression of interferon signaling. Nat Immunol 10, 109-115

126. Lapaquette, P., Guzzo, J., Bretillon, L., and Bringer, M. A. (2015) Cellular and Molecular Connections between Autophagy and Inflammation. Mediators Inflamm 2015, 398483

127. Jia, W., He, M. X., McLeod, I. X., Guo, J., Ji, D., and He, Y. W. (2015) Autophagy regulates T lymphocyte proliferation through selective degradation of the cell- cycle inhibitor CDKN1B/p27Kip1. Autophagy 11, 2335-2345

128. Jia, W., and He, Y. W. (2011) Temporal regulation of intracellular organelle homeostasis in T lymphocytes by autophagy. J Immunol 186, 5313-5322

129. Massey, D. C., and Parkes, M. (2007) Genome-wide association scanning highlights two autophagy genes, ATG16L1 and IRGM, as being significantly associated with Crohn's disease. Autophagy 3, 649-651

130. Parkes, M., Barrett, J. C., Prescott, N. J., Tremelling, M., Anderson, C. A., Fisher, S. A., Roberts, R. G., Nimmo, E. R., Cummings, F. R., Soars, D., Drummond, H., Lees, C. W., Khawaja, S. A., Bagnall, R., Burke, D. A., Todhunter, C. E., Ahmad, T., Onnie, C. M., McArdle, W., Strachan, D., Bethel, G., Bryan, C., Lewis, C. M., Deloukas, P., Forbes, A., Sanderson, J., Jewell, D. P., Satsangi, J., Mansfield, J. C., Wellcome Trust Case Control, C., Cardon, L., and Mathew, C. G. (2007) Sequence variants in the autophagy gene IRGM and multiple other replicating loci contribute to Crohn's disease susceptibility. Nat Genet 39, 830-832

131. McCarroll, S. A., Huett, A., Kuballa, P., Chilewski, S. D., Landry, A., Goyette, P., Zody, M. C., Hall, J. L., Brant, S. R., Cho, J. H., Duerr, R. H., Silverberg, M. S., Taylor, K. D., Rioux, J. D., Altshuler, D., Daly, M. J., and Xavier, R. J. (2008) Deletion polymorphism upstream of IRGM associated with altered IRGM expression and Crohn's disease. Nat Genet 40, 1107-1112

162

132. Huett, A., McCarroll, S. A., Daly, M. J., and Xavier, R. J. (2009) On the level: IRGM gene function is all about expression. Autophagy 5, 96-99

133. Lapaquette, P., Bringer, M. A., and Darfeuille-Michaud, A. (2012) Defects in autophagy favour adherent-invasive Escherichia coli persistence within macrophages leading to increased pro-inflammatory response. Cell Microbiol 14, 791-807

134. Chauhan, S., Mandell, M. A., and Deretic, V. (2015) IRGM governs the core autophagy machinery to conduct antimicrobial defense. Mol Cell 58, 507-521

135. Chauhan, S., Mandell, M. A., and Deretic, V. (2016) Mechanism of action of the tuberculosis and Crohn disease risk factor IRGM in autophagy. Autophagy 12, 429-431

136. Filomeni, G., De Zio, D., and Cecconi, F. (2015) Oxidative stress and autophagy: the clash between damage and metabolic needs. Cell Death Differ 22, 377-388

137. Lee, J., Giordano, S., and Zhang, J. (2012) Autophagy, mitochondria and oxidative stress: cross-talk and redox signalling. Biochem J 441, 523-540

138. Gutierrez, M. G., Master, S. S., Singh, S. B., Taylor, G. A., Colombo, M. I., and Deretic, V. (2004) Autophagy is a defense mechanism inhibiting BCG and Mycobacterium tuberculosis survival in infected macrophages. Cell 119, 753-766

139. Matsuzawa, T., Kim, B. H., Shenoy, A. R., Kamitani, S., Miyake, M., and Macmicking, J. D. (2012) IFN-gamma elicits macrophage autophagy via the p38 MAPK signaling pathway. J Immunol 189, 813-818

140. Schmidt, E. A., Fee, B. E., Henry, S. C., Nichols, A. G., Shinohara, M. L., Rathmell, J. C., MacIver, N. J., Coers, J., Ilkayeva, O. R., Koves, T. R., and Taylor, G. A. (2017) Metabolic alterations contribute to enhanced inflammatory cytokine production in Irgm1-deficient macrophages. J Biol Chem

141. Rathmell, J. C. (2012) Metabolism and autophagy in the immune system: immunometabolism comes of age. Immunol Rev 249, 5-13

142. O'Neill, L. A., Kishton, R. J., and Rathmell, J. (2016) A guide to immunometabolism for immunologists. Nat Rev Immunol 16, 553-565

163

143. Juhas, U., Ryba-Stanislawowska, M., Szargiej, P., and Mysliwska, J. (2015) Different pathways of macrophage activation and polarization. Postepy Hig Med Dosw (Online) 69, 496-502

144. Thomas, G. D., Ruckerl, D., Maskrey, B. H., Whitfield, P. D., Blaxter, M. L., and Allen, J. E. (2012) The biology of nematode- and IL4Ralpha-dependent murine macrophage polarization in vivo as defined by RNA-Seq and targeted lipidomics. Blood 120, e93-e104

145. Girodet, P. O., Nguyen, D., Mancini, J. D., Hundal, M., Zhou, X., Israel, E., and Cernadas, M. (2016) Alternative Macrophage Activation Is Increased in Asthma. Am J Respir Cell Mol Biol 55, 467-475

146. Vogelpoel, L. T., Hansen, I. S., Rispens, T., Muller, F. J., van Capel, T. M., Turina, M. C., Vos, J. B., Baeten, D. L., Kapsenberg, M. L., de Jong, E. C., and den Dunnen, J. (2014) Fc gamma receptor-TLR cross-talk elicits pro-inflammatory cytokine production by human M2 macrophages. Nat Commun 5, 5444

147. Kato, A., Yutani, M., Terao, M., Kimura, A., Itoi, S., Murota, H., Miyoshi, E., and Katayama, I. (2015) Oligosaccharide modification by N- acetylglucosaminyltransferase-V in macrophages are involved in pathogenesis of bleomycin-induced scleroderma. Exp Dermatol 24, 585-590

148. Sica, A., Larghi, P., Mancino, A., Rubino, L., Porta, C., Totaro, M. G., Rimoldi, M., Biswas, S. K., Allavena, P., and Mantovani, A. (2008) Macrophage polarization in tumour progression. Semin Cancer Biol 18, 349-355

149. Dominguez-Soto, A., Sierra-Filardi, E., Puig-Kroger, A., Perez-Maceda, B., Gomez-Aguado, F., Corcuera, M. T., Sanchez-Mateos, P., and Corbi, A. L. (2011) Dendritic cell-specific ICAM-3-grabbing nonintegrin expression on M2-polarized and tumor-associated macrophages is macrophage-CSF dependent and enhanced by tumor-derived IL-6 and IL-10. J Immunol 186, 2192-2200

150. Pena, O. M., Pistolic, J., Raj, D., Fjell, C. D., and Hancock, R. E. (2011) Endotoxin tolerance represents a distinctive state of alternative polarization (M2) in human mononuclear cells. J Immunol 186, 7243-7254

151. Murray, P. J., Allen, J. E., Biswas, S. K., Fisher, E. A., Gilroy, D. W., Goerdt, S., Gordon, S., Hamilton, J. A., Ivashkiv, L. B., Lawrence, T., Locati, M., Mantovani, A., Martinez, F. O., Mege, J. L., Mosser, D. M., Natoli, G., Saeij, J. P., Schultze, J. L., Shirey, K. A., Sica, A., Suttles, J., Udalova, I., van Ginderachter, J. A., Vogel, S.

164

N., and Wynn, T. A. (2014) Macrophage activation and polarization: nomenclature and experimental guidelines. Immunity 41, 14-20

152. Vergadi, E., Ieronymaki, E., Lyroni, K., Vaporidi, K., and Tsatsanis, C. (2017) Akt Signaling Pathway in Macrophage Activation and M1/M2 Polarization. J Immunol 198, 1006-1014

153. Martinez, F. O., and Gordon, S. (2014) The M1 and M2 paradigm of macrophage activation: time for reassessment. F1000Prime Rep 6, 13

154. Periyalil, H. A., Gibson, P. G., and Wood, L. G. (2013) Immunometabolism in obese asthmatics: are we there yet? Nutrients 5, 3506-3530

155. Xue, J., Schmidt, S. V., Sander, J., Draffehn, A., Krebs, W., Quester, I., De Nardo, D., Gohel, T. D., Emde, M., Schmidleithner, L., Ganesan, H., Nino-Castro, A., Mallmann, M. R., Labzin, L., Theis, H., Kraut, M., Beyer, M., Latz, E., Freeman, T. C., Ulas, T., and Schultze, J. L. (2014) Transcriptome-based network analysis reveals a spectrum model of human macrophage activation. Immunity 40, 274-288

156. Rodriguez-Prados, J. C., Traves, P. G., Cuenca, J., Rico, D., Aragones, J., Martin- Sanz, P., Cascante, M., and Bosca, L. (2010) Substrate fate in activated macrophages: a comparison between innate, classic, and alternative activation. J Immunol 185, 605-614

157. Ivashkiv, L. B. (2011) Inflammatory signaling in macrophages: transitions from acute to tolerant and alternative activation states. Eur J Immunol 41, 2477-2481

158. Schroder, K., Sweet, M. J., and Hume, D. A. (2006) Signal integration between IFNgamma and TLR signalling pathways in macrophages. Immunobiology 211, 511-524

159. Berg JM, T. J., Stryer L. (2002) Biochemistry 5th edition, W H Freeman, New York

160. Mathupala, S. P., Ko, Y. H., and Pedersen, P. L. (2006) Hexokinase II: cancer's double-edged sword acting as both facilitator and gatekeeper of malignancy when bound to mitochondria. Oncogene 25, 4777-4786

161. Shulga, N., and Pastorino, J. G. (2016) Hexokinase II binding to mitochondria is necessary for Kupffer cell activation and is potentiated by ethanol exposure. J Biol Chem 291, 12574

165

162. Stincone, A., Prigione, A., Cramer, T., Wamelink, M. M., Campbell, K., Cheung, E., Olin-Sandoval, V., Gruning, N. M., Kruger, A., Tauqeer Alam, M., Keller, M. A., Breitenbach, M., Brindle, K. M., Rabinowitz, J. D., and Ralser, M. (2015) The return of metabolism: biochemistry and physiology of the pentose phosphate pathway. Biol Rev Camb Philos Soc 90, 927-963

163. Infantino, V., Iacobazzi, V., Menga, A., Avantaggiati, M. L., and Palmieri, F. (2014) A key role of the mitochondrial citrate carrier (SLC25A1) in TNFalpha- and IFNgamma-triggered inflammation. Biochim Biophys Acta 1839, 1217-1225

164. Haschemi, A., Kosma, P., Gille, L., Evans, C. R., Burant, C. F., Starkl, P., Knapp, B., Haas, R., Schmid, J. A., Jandl, C., Amir, S., Lubec, G., Park, J., Esterbauer, H., Bilban, M., Brizuela, L., Pospisilik, J. A., Otterbein, L. E., and Wagner, O. (2012) The sedoheptulose kinase CARKL directs macrophage polarization through control of glucose metabolism. Cell Metab 15, 813-826

165. Fukuzumi, M., Shinomiya, H., Shimizu, Y., Ohishi, K., and Utsumi, S. (1996) Endotoxin-induced enhancement of glucose influx into murine peritoneal macrophages via GLUT1. Infect Immun 64, 108-112

166. Nagy, C., and Haschemi, A. (2015) Time and Demand are Two Critical Dimensions of Immunometabolism: The Process of Macrophage Activation and the Pentose Phosphate Pathway. Front Immunol 6, 164

167. Blagih, J., and Jones, R. G. (2012) Polarizing macrophages through reprogramming of glucose metabolism. Cell Metab 15, 793-795

168. Everts, B., Amiel, E., Huang, S. C., Smith, A. M., Chang, C. H., Lam, W. Y., Redmann, V., Freitas, T. C., Blagih, J., van der Windt, G. J., Artyomov, M. N., Jones, R. G., Pearce, E. L., and Pearce, E. J. (2014) TLR-driven early glycolytic reprogramming via the kinases TBK1-IKKvarepsilon supports the anabolic demands of dendritic cell activation. Nat Immunol 15, 323-332

169. Moon, J. S., Hisata, S., Park, M. A., DeNicola, G. M., Ryter, S. W., Nakahira, K., and Choi, A. M. (2015) mTORC1-Induced HK1-Dependent Glycolysis Regulates NLRP3 Inflammasome Activation. Cell Rep 12, 102-115

170. Luo, W., Hu, H., Chang, R., Zhong, J., Knabel, M., O'Meally, R., Cole, R. N., Pandey, A., and Semenza, G. L. (2011) Pyruvate kinase M2 is a PHD3-stimulated coactivator for hypoxia-inducible factor 1. Cell 145, 732-744

166

171. Palsson-McDermott, E. M., Curtis, A. M., Goel, G., Lauterbach, M. A., Sheedy, F. J., Gleeson, L. E., van den Bosch, M. W., Quinn, S. R., Domingo-Fernandez, R., Johnston, D. G., Jiang, J. K., Israelsen, W. J., Keane, J., Thomas, C., Clish, C., Vander Heiden, M., Xavier, R. J., and O'Neill, L. A. (2015) Pyruvate kinase M2 regulates Hif-1alpha activity and IL-1beta induction and is a critical determinant of the warburg effect in LPS-activated macrophages. Cell Metab 21, 65-80

172. Mattevi, A., Bolognesi, M., and Valentini, G. (1996) The allosteric regulation of pyruvate kinase. FEBS Lett 389, 15-19

173. Jurica, M. S., Mesecar, A., Heath, P. J., Shi, W., Nowak, T., and Stoddard, B. L. (1998) The allosteric regulation of pyruvate kinase by fructose-1,6-bisphosphate. Structure 6, 195-210

174. Ruiz-Garcia, A., Monsalve, E., Novellasdemunt, L., Navarro-Sabate, A., Manzano, A., Rivero, S., Castrillo, A., Casado, M., Laborda, J., Bartrons, R., and Diaz-Guerra, M. J. (2011) Cooperation of adenosine with macrophage Toll-4 receptor agonists leads to increased glycolytic flux through the enhanced expression of PFKFB3 gene. J Biol Chem 286, 19247-19258

175. Tan, Z., Xie, N., Banerjee, S., Cui, H., Fu, M., Thannickal, V. J., and Liu, G. (2015) The monocarboxylate transporter 4 is required for glycolytic reprogramming and inflammatory response in macrophages. J Biol Chem 290, 46-55

176. O'Neill, L. A., and Pearce, E. J. (2016) Immunometabolism governs dendritic cell and macrophage function. J Exp Med 213, 15-23

177. Tannahill, G. M., Curtis, A. M., Adamik, J., Palsson-McDermott, E. M., McGettrick, A. F., Goel, G., Frezza, C., Bernard, N. J., Kelly, B., Foley, N. H., Zheng, L., Gardet, A., Tong, Z., Jany, S. S., Corr, S. C., Haneklaus, M., Caffrey, B. E., Pierce, K., Walmsley, S., Beasley, F. C., Cummins, E., Nizet, V., Whyte, M., Taylor, C. T., Lin, H., Masters, S. L., Gottlieb, E., Kelly, V. P., Clish, C., Auron, P. E., Xavier, R. J., and O'Neill, L. A. (2013) Succinate is an inflammatory signal that induces IL-1beta through HIF-1alpha. Nature 496, 238-242

178. Webb, P. (2009) Alternative macrophage activation and the regulation of metabolism. F1000 Biol Rep 1, 2

179. Odegaard, J. I., and Chawla, A. (2011) Alternative macrophage activation and metabolism. Annu Rev Pathol 6, 275-297

167

180. Roszer, T. (2015) Understanding the Mysterious M2 Macrophage through Activation Markers and Effector Mechanisms. Mediators Inflamm 2015, 816460

181. Hume, D. A. (2015) The Many Alternative Faces of Macrophage Activation. Front Immunol 6, 370

182. Board, M., Humm, S., and Newsholme, E. A. (1990) Maximum activities of key enzymes of glycolysis, glutaminolysis, pentose phosphate pathway and tricarboxylic acid cycle in normal, neoplastic and suppressed cells. Biochem J 265, 503-509

183. Samland, A. K., and Sprenger, G. A. (2009) Transaldolase: from biochemistry to human disease. Int J Biochem Cell Biol 41, 1482-1494

184. Zhao, J., and Zhong, C. J. (2009) A review on research progress of transketolase. Neurosci Bull 25, 94-99

185. Hubler, M. J., and Kennedy, A. J. (2016) Role of lipids in the metabolism and activation of immune cells. J Nutr Biochem 34, 1-7

186. Langston, P. K., Shibata, M., and Horng, T. (2017) Metabolism Supports Macrophage Activation. Front Immunol 8, 61

187. Kruger, N. J., and von Schaewen, A. (2003) The oxidative pentose phosphate pathway: structure and organisation. Curr Opin Plant Biol 6, 236-246

188. Wamelink, M. M., Struys, E. A., and Jakobs, C. (2008) The biochemistry, metabolism and inherited defects of the pentose phosphate pathway: a review. J Inherit Metab Dis 31, 703-717

189. Smirnov, A. N. (2010) Lipid signaling in the atherogenesis context. Biochemistry (Mosc) 75, 793-810

190. Ray, P. D., Huang, B. W., and Tsuji, Y. (2012) Reactive oxygen species (ROS) homeostasis and redox regulation in cellular signaling. Cell Signal 24, 981-990

191. Infantino, V., Convertini, P., Cucci, L., Panaro, M. A., Di Noia, M. A., Calvello, R., Palmieri, F., and Iacobazzi, V. (2011) The mitochondrial citrate carrier: a new player in inflammation. Biochem J 438, 433-436

168

192. Pollak, N., Dolle, C., and Ziegler, M. (2007) The power to reduce: pyridine nucleotides - small molecules with a multitude of functions. Biochemical Journal 402, 205-218

193. Jha, A. K., Huang, S. C., Sergushichev, A., Lampropoulou, V., Ivanova, Y., Loginicheva, E., Chmielewski, K., Stewart, K. M., Ashall, J., Everts, B., Pearce, E. J., Driggers, E. M., and Artyomov, M. N. (2015) Network integration of parallel metabolic and transcriptional data reveals metabolic modules that regulate macrophage polarization. Immunity 42, 419-430

194. Ham, M., Lee, J. W., Choi, A. H., Jang, H., Choi, G., Park, J., Kozuka, C., Sears, D. D., Masuzaki, H., and Kim, J. B. (2013) Macrophage glucose-6-phosphate dehydrogenase stimulates proinflammatory responses with oxidative stress. Mol Cell Biol 33, 2425-2435

195. Saucillo, D. C., Gerriets, V. A., Sheng, J., Rathmell, J. C., and Maciver, N. J. (2014) Leptin metabolically licenses T cells for activation to link nutrition and immunity. J Immunol 192, 136-144

196. Freemerman, A. J., Johnson, A. R., Sacks, G. N., Milner, J. J., Kirk, E. L., Troester, M. A., Macintyre, A. N., Goraksha-Hicks, P., Rathmell, J. C., and Makowski, L. (2014) Metabolic reprogramming of macrophages: glucose transporter 1 (GLUT1)-mediated glucose metabolism drives a proinflammatory phenotype. J Biol Chem 289, 7884-7896

197. Hoo, R. L., Shu, L., Cheng, K. K., Wu, X., Liao, B., Wu, D., Zhou, Z., and Xu, A. (2017) Adipocyte Fatty Acid Binding Protein Potentiates Toxic Lipids-Induced Endoplasmic Reticulum Stress in Macrophages via Inhibition of Janus Kinase 2- dependent Autophagy. Sci Rep 7, 40657

198. Namgaladze, D., Lips, S., Leiker, T. J., Murphy, R. C., Ekroos, K., Ferreiros, N., Geisslinger, G., and Brune, B. (2014) Inhibition of macrophage fatty acid beta- oxidation exacerbates palmitate-induced inflammatory and endoplasmic reticulum stress responses. Diabetologia 57, 1067-1077

199. Ishola, D. A., Jr., Post, J. A., van Timmeren, M. M., Bakker, S. J., Goldschmeding, R., Koomans, H. A., Braam, B., and Joles, J. A. (2006) Albumin-bound fatty acids induce mitochondrial oxidant stress and impair antioxidant responses in proximal tubular cells. Kidney Int 70, 724-731

169

200. Lei, S., Sun, R. Z., Wang, D., Gong, M. Z., Su, X. P., Yi, F., and Peng, Z. W. (2016) Increased Hepatic Fatty Acids Uptake and Oxidation by LRPPRC-Driven Oxidative Phosphorylation Reduces Blood Lipid Levels. Front Physiol 7, 270

201. Steinberg, G. R., and Schertzer, J. D. (2014) AMPK promotes macrophage fatty acid oxidative metabolism to mitigate inflammation: implications for diabetes and cardiovascular disease. Immunol Cell Biol 92, 340-345

202. Cingolani, F., and Czaja, M. J. (2016) Regulation and Functions of Autophagic Lipolysis. Trends Endocrinol Metab 27, 696-705

203. Joseph, S. B., Bradley, M. N., Castrillo, A., Bruhn, K. W., Mak, P. A., Pei, L., Hogenesch, J., O'Connell R, M., Cheng, G., Saez, E., Miller, J. F., and Tontonoz, P. (2004) LXR-dependent gene expression is important for macrophage survival and the innate immune response. Cell 119, 299-309

204. Im, S. S., Yousef, L., Blaschitz, C., Liu, J. Z., Edwards, R. A., Young, S. G., Raffatellu, M., and Osborne, T. F. (2011) Linking lipid metabolism to the innate immune response in macrophages through sterol regulatory element binding protein-1a. Cell Metab 13, 540-549

205. Feingold, K. R., Shigenaga, J. K., Kazemi, M. R., McDonald, C. M., Patzek, S. M., Cross, A. S., Moser, A., and Grunfeld, C. (2012) Mechanisms of triglyceride accumulation in activated macrophages. J Leukoc Biol 92, 829-839

206. Ecker, J., Liebisch, G., Englmaier, M., Grandl, M., Robenek, H., and Schmitz, G. (2010) Induction of fatty acid synthesis is a key requirement for phagocytic differentiation of human monocytes. Proc Natl Acad Sci U S A 107, 7817-7822

207. Posokhova, E. N., Khoshchenko, O. M., Chasovskikh, M. I., Pivovarova, E. N., and Dushkin, M. I. (2008) Lipid synthesis in macrophages during inflammation in vivo: effect of agonists of peroxisome proliferator activated receptors alpha and gamma and of retinoid X receptors. Biochemistry (Mosc) 73, 296-304

208. Moon, J. S., Lee, S., Park, M. A., Siempos, II, Haslip, M., Lee, P. J., Yun, M., Kim, C. K., Howrylak, J., Ryter, S. W., Nakahira, K., and Choi, A. M. (2015) UCP2- induced fatty acid synthase promotes NLRP3 inflammasome activation during sepsis. J Clin Invest 125, 665-680

209. Tangirala, R. K., Bischoff, E. D., Joseph, S. B., Wagner, B. L., Walczak, R., Laffitte, B. A., Daige, C. L., Thomas, D., Heyman, R. A., Mangelsdorf, D. J., Wang, X., Lusis, A. J., Tontonoz, P., and Schulman, I. G. (2002) Identification of macrophage 170

liver X receptors as inhibitors of atherosclerosis. Proc Natl Acad Sci U S A 99, 11896-11901

210. Hong, C., Walczak, R., Dhamko, H., Bradley, M. N., Marathe, C., Boyadjian, R., Salazar, J. V., and Tontonoz, P. (2011) Constitutive activation of LXR in macrophages regulates metabolic and inflammatory gene expression: identification of ARL7 as a direct target. J Lipid Res 52, 531-539

211. Zelcer, N., and Tontonoz, P. (2006) Liver X receptors as integrators of metabolic and inflammatory signaling. J Clin Invest 116, 607-614

212. Ghisletti, S., Huang, W., Jepsen, K., Benner, C., Hardiman, G., Rosenfeld, M. G., and Glass, C. K. (2009) Cooperative NCoR/SMRT interactions establish a corepressor-based strategy for integration of inflammatory and anti- inflammatory signaling pathways. Genes Dev 23, 681-693

213. Murray, P. J., Rathmell, J., and Pearce, E. (2015) SnapShot: Immunometabolism. Cell Metab 22, 190-190 e191

214. O'Neill, L. A. (2011) A critical role for citrate metabolism in LPS signalling. Biochem J 438, e5-6

215. Wellen, K. E., Hatzivassiliou, G., Sachdeva, U. M., Bui, T. V., Cross, J. R., and Thompson, C. B. (2009) ATP-citrate lyase links cellular metabolism to histone acetylation. Science 324, 1076-1080

216. Iacobazzi, V., Invernizzi, F., Baratta, S., Pons, R., Chung, W., Garavaglia, B., Dionisi-Vici, C., Ribes, A., Parini, R., Huertas, M. D., Roldan, S., Lauria, G., Palmieri, F., and Taroni, F. (2004) Molecular and functional analysis of SLC25A20 mutations causing carnitine-acylcarnitine translocase deficiency. Hum Mutat 24, 312-320

217. Strelko, C. L., Lu, W., Dufort, F. J., Seyfried, T. N., Chiles, T. C., Rabinowitz, J. D., and Roberts, M. F. (2011) Itaconic acid is a mammalian metabolite induced during macrophage activation. J Am Chem Soc 133, 16386-16389

218. Michelucci, A., Cordes, T., Ghelfi, J., Pailot, A., Reiling, N., Goldmann, O., Binz, T., Wegner, A., Tallam, A., Rausell, A., Buttini, M., Linster, C. L., Medina, E., Balling, R., and Hiller, K. (2013) Immune-responsive gene 1 protein links metabolism to immunity by catalyzing itaconic acid production. Proc Natl Acad Sci U S A 110, 7820-7825

171

219. Lampropoulou, V., Sergushichev, A., Bambouskova, M., Nair, S., Vincent, E. E., Loginicheva, E., Cervantes-Barragan, L., Ma, X., Huang, S. C., Griss, T., Weinheimer, C. J., Khader, S., Randolph, G. J., Pearce, E. J., Jones, R. G., Diwan, A., Diamond, M. S., and Artyomov, M. N. (2016) Itaconate Links Inhibition of Succinate Dehydrogenase with Macrophage Metabolic Remodeling and Regulation of Inflammation. Cell Metab 24, 158-166

220. Naujoks, J., Tabeling, C., Dill, B. D., Hoffmann, C., Brown, A. S., Kunze, M., Kempa, S., Peter, A., Mollenkopf, H. J., Dorhoi, A., Kershaw, O., Gruber, A. D., Sander, L. E., Witzenrath, M., Herold, S., Nerlich, A., Hocke, A. C., van Driel, I., Suttorp, N., Bedoui, S., Hilbi, H., Trost, M., and Opitz, B. (2016) IFNs Modify the Proteome of Legionella-Containing Vacuoles and Restrict Infection Via IRG1- Derived Itaconic Acid. PLoS Pathog 12, e1005408

221. Nemeth, B., Doczi, J., Csete, D., Kacso, G., Ravasz, D., Adams, D., Kiss, G., Nagy, A. M., Horvath, G., Tretter, L., Mocsai, A., Csepanyi-Komi, R., Iordanov, I., Adam-Vizi, V., and Chinopoulos, C. (2016) Abolition of mitochondrial substrate- level phosphorylation by itaconic acid produced by LPS-induced Irg1 expression in cells of murine macrophage lineage. FASEB J 30, 286-300

222. Degrandi, D., Hoffmann, R., Beuter-Gunia, C., and Pfeffer, K. (2009) The proinflammatory cytokine-induced IRG1 protein associates with mitochondria. J Interferon Cytokine Res 29, 55-67

223. O'Neill, L. A. (2015) A broken krebs cycle in macrophages. Immunity 42, 393-394

224. Allen, A. E., Dupont, C. L., Obornik, M., Horak, A., Nunes-Nesi, A., McCrow, J. P., Zheng, H., Johnson, D. A., Hu, H., Fernie, A. R., and Bowler, C. (2011) Evolution and metabolic significance of the urea cycle in photosynthetic diatoms. Nature 473, 203-207

225. Benit, P., Letouze, E., Rak, M., Aubry, L., Burnichon, N., Favier, J., Gimenez- Roqueplo, A. P., and Rustin, P. (2014) Unsuspected task for an old team: succinate, fumarate and other Krebs cycle acids in metabolic remodeling. Biochim Biophys Acta 1837, 1330-1337

226. Selak, M. A., Armour, S. M., MacKenzie, E. D., Boulahbel, H., Watson, D. G., Mansfield, K. D., Pan, Y., Simon, M. C., Thompson, C. B., and Gottlieb, E. (2005) Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha prolyl hydroxylase. Cancer Cell 7, 77-85

172

227. Lackner, L. L. (2014) Shaping the dynamic mitochondrial network. BMC Biol 12, 35

228. Tur, J., Vico, T., Lloberas, J., Zorzano, A., and Celada, A. (2017) Macrophages and Mitochondria: A Critical Interplay Between Metabolism, Signaling, and the Functional Activity. Adv Immunol 133, 1-36

229. Monlun, M., Hyernard, C., Blanco, P., Lartigue, L., and Faustin, B. (2017) Mitochondria as Molecular Platforms Integrating Multiple Innate Immune Signalings. J Mol Biol 429, 1-13

230. Taguchi, N., Ishihara, N., Jofuku, A., Oka, T., and Mihara, K. (2007) Mitotic phosphorylation of dynamin-related GTPase Drp1 participates in mitochondrial fission. J Biol Chem 282, 11521-11529

231. Mitra, K., Wunder, C., Roysam, B., Lin, G., and Lippincott-Schwartz, J. (2009) A hyperfused mitochondrial state achieved at G1-S regulates cyclin E buildup and entry into S phase. Proc Natl Acad Sci U S A 106, 11960-11965

232. Smirnova, E., Griparic, L., Shurland, D. L., and van der Bliek, A. M. (2001) Dynamin-related protein Drp1 is required for mitochondrial division in mammalian cells. Mol Biol Cell 12, 2245-2256

233. Ingerman, E., Perkins, E. M., Marino, M., Mears, J. A., McCaffery, J. M., Hinshaw, J. E., and Nunnari, J. (2005) Dnm1 forms spirals that are structurally tailored to fit mitochondria. J Cell Biol 170, 1021-1027

234. Wenger, J., Klinglmayr, E., Frohlich, C., Eibl, C., Gimeno, A., Hessenberger, M., Puehringer, S., Daumke, O., and Goettig, P. (2013) Functional mapping of human dynamin-1-like GTPase domain based on x-ray structure analyses. PLoS One 8, e71835

235. Frohlich, C., Grabiger, S., Schwefel, D., Faelber, K., Rosenbaum, E., Mears, J., Rocks, O., and Daumke, O. (2013) Structural insights into oligomerization and mitochondrial remodelling of dynamin 1-like protein. EMBO J 32, 1280-1292

236. Chappie, J. S., Acharya, S., Leonard, M., Schmid, S. L., and Dyda, F. (2010) G domain dimerization controls dynamin's assembly-stimulated GTPase activity. Nature 465, 435-440

173

237. Mears, J. A., Lackner, L. L., Fang, S., Ingerman, E., Nunnari, J., and Hinshaw, J. E. (2011) Conformational changes in Dnm1 support a contractile mechanism for mitochondrial fission. Nat Struct Mol Biol 18, 20-26

238. Legesse-Miller, A., Massol, R. H., and Kirchhausen, T. (2003) Constriction and Dnm1p recruitment are distinct processes in mitochondrial fission. Mol Biol Cell 14, 1953-1963

239. Otera, H., Ishihara, N., and Mihara, K. (2013) New insights into the function and regulation of mitochondrial fission. Biochim Biophys Acta 1833, 1256-1268

240. Friedman, J. R., and Nunnari, J. (2014) Mitochondrial form and function. Nature 505, 335-343

241. Labbe, K., Murley, A., and Nunnari, J. (2014) Determinants and functions of mitochondrial behavior. Annu Rev Cell Dev Biol 30, 357-391

242. Strack, S., Wilson, T. J., and Cribbs, J. T. (2013) Cyclin-dependent kinases regulate splice-specific targeting of dynamin-related protein 1 to microtubules. J Cell Biol 201, 1037-1051

243. Loson, O. C., Liu, R., Rome, M. E., Meng, S., Kaiser, J. T., Shan, S. O., and Chan, D. C. (2014) The mitochondrial fission receptor MiD51 requires ADP as a cofactor. Structure 22, 367-377

244. Palmer, C. S., Osellame, L. D., Laine, D., Koutsopoulos, O. S., Frazier, A. E., and Ryan, M. T. (2011) MiD49 and MiD51, new components of the mitochondrial fission machinery. EMBO Rep 12, 565-573

245. Gandre-Babbe, S., and van der Bliek, A. M. (2008) The novel tail-anchored membrane protein Mff controls mitochondrial and peroxisomal fission in mammalian cells. Mol Biol Cell 19, 2402-2412

246. Otera, H., Miyata, N., Kuge, O., and Mihara, K. (2016) Drp1-dependent mitochondrial fission via MiD49/51 is essential for apoptotic cristae remodeling. J Cell Biol 212, 531-544

247. Otera, H., Wang, C., Cleland, M. M., Setoguchi, K., Yokota, S., Youle, R. J., and Mihara, K. (2010) Mff is an essential factor for mitochondrial recruitment of Drp1 during mitochondrial fission in mammalian cells. J Cell Biol 191, 1141-1158

174

248. Friedman, J. R., Lackner, L. L., West, M., DiBenedetto, J. R., Nunnari, J., and Voeltz, G. K. (2011) ER tubules mark sites of mitochondrial division. Science 334, 358-362

249. Korobova, F., Ramabhadran, V., and Higgs, H. N. (2013) An actin-dependent step in mitochondrial fission mediated by the ER-associated formin INF2. Science 339, 464-467

250. Stroud, D. A., Oeljeklaus, S., Wiese, S., Bohnert, M., Lewandrowski, U., Sickmann, A., Guiard, B., van der Laan, M., Warscheid, B., and Wiedemann, N. (2011) Composition and topology of the endoplasmic reticulum-mitochondria encounter structure. J Mol Biol 413, 743-750

251. Kornmann, B., Osman, C., and Walter, P. (2011) The conserved GTPase Gem1 regulates endoplasmic reticulum-mitochondria connections. Proc Natl Acad Sci U S A 108, 14151-14156

252. Theurey, P., and Rieusset, J. (2017) Mitochondria-Associated Membranes Response to Nutrient Availability and Role in Metabolic Diseases. Trends Endocrinol Metab 28, 32-45

253. Nunnari, J., and Suomalainen, A. (2012) Mitochondria: in sickness and in health. Cell 148, 1145-1159

254. DeVay, R. M., Dominguez-Ramirez, L., Lackner, L. L., Hoppins, S., Stahlberg, H., and Nunnari, J. (2009) Coassembly of Mgm1 isoforms requires cardiolipin and mediates mitochondrial inner membrane fusion. J Cell Biol 186, 793-803

255. Chen, H., Detmer, S. A., Ewald, A. J., Griffin, E. E., Fraser, S. E., and Chan, D. C. (2003) Mitofusins Mfn1 and Mfn2 coordinately regulate mitochondrial fusion and are essential for embryonic development. J Cell Biol 160, 189-200

256. Santel, A., and Fuller, M. T. (2001) Control of mitochondrial morphology by a human mitofusin. J Cell Sci 114, 867-874

257. Hermann, G. J., Thatcher, J. W., Mills, J. P., Hales, K. G., Fuller, M. T., Nunnari, J., and Shaw, J. M. (1998) Mitochondrial fusion in yeast requires the transmembrane GTPase Fzo1p. J Cell Biol 143, 359-373

258. Ban, T., Heymann, J. A., Song, Z., Hinshaw, J. E., and Chan, D. C. (2010) OPA1 disease alleles causing dominant optic atrophy have defects in cardiolipin-

175

stimulated GTP hydrolysis and membrane tubulation. Hum Mol Genet 19, 2113- 2122

259. Song, Z., Ghochani, M., McCaffery, J. M., Frey, T. G., and Chan, D. C. (2009) Mitofusins and OPA1 mediate sequential steps in mitochondrial membrane fusion. Mol Biol Cell 20, 3525-3532

260. Mishra, P., Carelli, V., Manfredi, G., and Chan, D. C. (2014) Proteolytic cleavage of Opa1 stimulates mitochondrial inner membrane fusion and couples fusion to oxidative phosphorylation. Cell Metab 19, 630-641

261. Ehses, S., Raschke, I., Mancuso, G., Bernacchia, A., Geimer, S., Tondera, D., Martinou, J. C., Westermann, B., Rugarli, E. I., and Langer, T. (2009) Regulation of OPA1 processing and mitochondrial fusion by m-AAA protease isoenzymes and OMA1. J Cell Biol 187, 1023-1036

262. Head, B., Griparic, L., Amiri, M., Gandre-Babbe, S., and van der Bliek, A. M. (2009) Inducible proteolytic inactivation of OPA1 mediated by the OMA1 protease in mammalian cells. J Cell Biol 187, 959-966

263. Liu, X., Weaver, D., Shirihai, O., and Hajnoczky, G. (2009) Mitochondrial 'kiss- and-run': interplay between mitochondrial motility and fusion-fission dynamics. EMBO J 28, 3074-3089

264. Chan, N. C., Salazar, A. M., Pham, A. H., Sweredoski, M. J., Kolawa, N. J., Graham, R. L., Hess, S., and Chan, D. C. (2011) Broad activation of the ubiquitin- proteasome system by Parkin is critical for mitophagy. Hum Mol Genet 20, 1726- 1737

265. Narendra, D. P., Jin, S. M., Tanaka, A., Suen, D. F., Gautier, C. A., Shen, J., Cookson, M. R., and Youle, R. J. (2010) PINK1 is selectively stabilized on impaired mitochondria to activate Parkin. PLoS Biol 8, e1000298

266. Matsuda, N., Sato, S., Shiba, K., Okatsu, K., Saisho, K., Gautier, C. A., Sou, Y. S., Saiki, S., Kawajiri, S., Sato, F., Kimura, M., Komatsu, M., Hattori, N., and Tanaka, K. (2010) PINK1 stabilized by mitochondrial depolarization recruits Parkin to damaged mitochondria and activates latent Parkin for mitophagy. J Cell Biol 189, 211-221

267. Karbowski, M., Norris, K. L., Cleland, M. M., Jeong, S. Y., and Youle, R. J. (2006) Role of Bax and Bak in mitochondrial morphogenesis. Nature 443, 658-662

176

268. Cassidy-Stone, A., Chipuk, J. E., Ingerman, E., Song, C., Yoo, C., Kuwana, T., Kurth, M. J., Shaw, J. T., Hinshaw, J. E., Green, D. R., and Nunnari, J. (2008) Chemical inhibition of the mitochondrial division dynamin reveals its role in Bax/Bak-dependent mitochondrial outer membrane permeabilization. Dev Cell 14, 193-204

269. Whelan RS, K. K., Wei AC, Chen Y, Reyna DE, Jha S, Yang Y, Calvert JW, Lindsten T, Thompson CB, Crow MT, Gavathiotis E, Dorn GW 2nd, O'Rourke B, Kitsis RN. (2012) Bax regulates primary necrosis through mitochondrial dynamics. PNAS 109, 6566

270. Renault, T. T., Floros, K. V., Elkholi, R., Corrigan, K. A., Kushnareva, Y., Wieder, S. Y., Lindtner, C., Serasinghe, M. N., Asciolla, J. J., Buettner, C., Newmeyer, D. D., and Chipuk, J. E. (2015) Mitochondrial shape governs BAX-induced membrane permeabilization and apoptosis. Mol Cell 57, 69-82

271. Gomes, L. C., Di Benedetto, G., and Scorrano, L. (2011) During autophagy mitochondria elongate, are spared from degradation and sustain cell viability. Nat Cell Biol 13, 589-598

272. Rambold, A. S., Kostelecky, B., Elia, N., and Lippincott-Schwartz, J. (2011) Tubular network formation protects mitochondria from autophagosomal degradation during nutrient starvation. Proc Natl Acad Sci U S A 108, 10190-10195

273. Tondera, D., Grandemange, S., Jourdain, A., Karbowski, M., Mattenberger, Y., Herzig, S., Da Cruz, S., Clerc, P., Raschke, I., Merkwirth, C., Ehses, S., Krause, F., Chan, D. C., Alexander, C., Bauer, C., Youle, R., Langer, T., and Martinou, J. C. (2009) SLP-2 is required for stress-induced mitochondrial hyperfusion. EMBO J 28, 1589-1600

274. Shutt, T., Geoffrion, M., Milne, R., and McBride, H. M. (2012) The intracellular redox state is a core determinant of mitochondrial fusion. EMBO Rep 13, 909-915

275. Mishra, P., and Chan, D. C. (2016) Metabolic regulation of mitochondrial dynamics. J Cell Biol 212, 379-387

276. Stumpf, J. D., and Copeland, W. C. (2011) Mitochondrial DNA replication and disease: insights from DNA polymerase gamma mutations. Cell Mol Life Sci 68, 219-233

277. Copeland, W. C. (2012) Defects in mitochondrial DNA replication and human disease. Crit Rev Biochem Mol Biol 47, 64-74 177

278. Chen, H., Vermulst, M., Wang, Y. E., Chomyn, A., Prolla, T. A., McCaffery, J. M., and Chan, D. C. (2010) Mitochondrial fusion is required for mtDNA stability in skeletal muscle and tolerance of mtDNA mutations. Cell 141, 280-289

279. Parone, P. A., Da Cruz, S., Tondera, D., Mattenberger, Y., James, D. I., Maechler, P., Barja, F., and Martinou, J. C. (2008) Preventing mitochondrial fission impairs mitochondrial function and leads to loss of mitochondrial DNA. PLoS One 3, e3257

280. Skulachev, V. P. (2001) Mitochondrial filaments and clusters as intracellular power-transmitting cables. Trends Biochem Sci 26, 23-29

281. Amchenkova, A. A., Bakeeva, L. E., Chentsov, Y. S., Skulachev, V. P., and Zorov, D. B. (1988) Coupling membranes as energy-transmitting cables. I. Filamentous mitochondria in fibroblasts and mitochondrial clusters in cardiomyocytes. J Cell Biol 107, 481-495

282. Rossignol, R., Gilkerson, R., Aggeler, R., Yamagata, K., Remington, S. J., and Capaldi, R. A. (2004) Energy substrate modulates mitochondrial structure and oxidative capacity in cancer cells. Cancer Res 64, 985-993

283. Jakobs, S., Martini, N., Schauss, A. C., Egner, A., Westermann, B., and Hell, S. W. (2003) Spatial and temporal dynamics of budding yeast mitochondria lacking the division component Fis1p. J Cell Sci 116, 2005-2014

284. Chan, D. C. (2012) Fusion and fission: interlinked processes critical for mitochondrial health. Annu Rev Genet 46, 265-287

285. Twig, G., Elorza, A., Molina, A. J., Mohamed, H., Wikstrom, J. D., Walzer, G., Stiles, L., Haigh, S. E., Katz, S., Las, G., Alroy, J., Wu, M., Py, B. F., Yuan, J., Deeney, J. T., Corkey, B. E., and Shirihai, O. S. (2008) Fission and selective fusion govern mitochondrial segregation and elimination by autophagy. EMBO J 27, 433-446

286. Chen, Y., and Sheng, Z. H. (2013) Kinesin-1-syntaphilin coupling mediates activity-dependent regulation of axonal mitochondrial transport. J Cell Biol 202, 351-364

287. Saxton, W. M., and Hollenbeck, P. J. (2012) The axonal transport of mitochondria. J Cell Sci 125, 2095-2104

178

288. Manry, J., Laval, G., Patin, E., Fornarino, S., Tichit, M., Bouchier, C., Barreiro, L. B., and Quintana-Murci, L. (2011) Evolutionary genetics evidence of an essential, nonredundant role of the IFN-gamma pathway in protective immunity. Hum Mutat 32, 633-642

289. Deretic, V. (2016) Autophagy in leukocytes and other cells: mechanisms, subsystem organization, selectivity, and links to innate immunity. J Leukoc Biol 100, 969-978

290. Steinfeldt, T., Konen-Waisman, S., Tong, L., Pawlowski, N., Lamkemeyer, T., Sibley, L. D., Hunn, J. P., and Howard, J. C. (2015) Correction: Phosphorylation of Mouse Immunity-Related GTPase (IRG) Resistance Proteins Is an Evasion Strategy for Virulent Toxoplasma gondii. PLoS Biol 13, e1002199

291. Lu, Y., Li, Q., Peng, J., Zhu, Y., Wang, F., Wang, C., and Wang, X. (2016) Association of autophagy-related IRGM polymorphisms with latent versus active tuberculosis infection in a Chinese population. Tuberculosis (Edinb) 97, 47-51

292. Lu, X. C., Tao, Y., Wu, C., Zhao, P. L., Li, K., Zheng, J. Y., and Li, L. X. (2013) Association between variants of the autophagy related gene--IRGM and susceptibility to Crohn's disease and ulcerative colitis: a meta-analysis. PLoS One 8, e80602

293. Feng, C. G., Zheng, L., Lenardo, M. J., and Sher, A. (2009) Interferon-inducible immunity-related GTPase Irgm1 regulates IFN gamma-dependent host defense, lymphocyte survival and autophagy. Autophagy 5, 232-234

294. Wang, L., Gui, Y. S., Tian, X. L., Cai, B. Q., Wang, D. T., Zhang, D., Zhao, H., and Xu, K. F. (2011) Inactivation of mammalian target of rapamycin (mTOR) by rapamycin in a murine model of lipopolysaccharide-induced acute lung injury. Chin Med J (Engl) 124, 3112-3117

295. Mann-Chandler, M. N., Kashyap, M., Wright, H. V., Norozian, F., Barnstein, B. O., Gingras, S., Parganas, E., and Ryan, J. J. (2005) IFN-gamma induces apoptosis in developing mast cells. J Immunol 175, 3000-3005

296. Liu L, L. Y., Martinez J, Bi Y, Lian G, Wang T, Milasta S, Wang J, Yang M, Liu G, Green DR, Wang R. (2016) Proinflammatory signal suppresses proliferation and shifts macrophage metabolism from Myc-dependent to HIF1α-dependent. Proc Natl Acad Sci U S A 113, 1564-1569

179

297. Otera, H., and Mihara, K. (2011) Molecular mechanisms and physiologic functions of mitochondrial dynamics. J Biochem 149, 241-251

298. Wai, T., and Langer, T. (2016) Mitochondrial Dynamics and Metabolic Regulation. Trends Endocrinol Metab 27, 105-117

299. Vance, J. E. (2014) MAM (mitochondria-associated membranes) in mammalian cells: lipids and beyond. Biochim Biophys Acta 1841, 595-609

300. Campello, S., and Scorrano, L. (2010) Mitochondrial shape changes: orchestrating cell pathophysiology. EMBO Rep 11, 678-684

301. Galloway, C. A., Lee, H., and Yoon, Y. (2012) Mitochondrial morphology- emerging role in bioenergetics. Free Radic Biol Med 53, 2218-2228

302. Hwang, S. J., and Kim, W. (2013) Mitochondrial dynamics in the heart as a novel therapeutic target for cardioprotection. Chonnam Med J 49, 101-107

303. Galloway, C. A., and Yoon, Y. (2013) Mitochondrial morphology in metabolic diseases. Antioxid Redox Signal 19, 415-430

304. Gomes, L. C., and Scorrano, L. (2013) Mitochondrial morphology in mitophagy and macroautophagy. Biochim Biophys Acta 1833, 205-212

305. Sgarbi, G., Matarrese, P., Pinti, M., Lanzarini, C., Ascione, B., Gibellini, L., Dika, E., Patrizi, A., Tommasino, C., Capri, M., Cossarizza, A., Baracca, A., Lenaz, G., Solaini, G., Franceschi, C., Malorni, W., and Salvioli, S. (2014) Mitochondria hyperfusion and elevated autophagic activity are key mechanisms for cellular bioenergetic preservation in centenarians. Aging (Albany NY) 6, 296-310

306. Silva Ramos, E., Larsson, N. G., and Mourier, A. (2016) Bioenergetic roles of mitochondrial fusion. Biochim Biophys Acta 1857, 1277-1283

307. Kroczynska, B., Blyth, G. T., Rafidi, R. L., Majchrzak-Kita, B., Xu, L., Saleiro, D., Kosciuczuk, E. M., Jemielity, J., Su, B., Altman, J. K., Eklund, E. A., Fish, E. H., and Platanias, L. C. (2017) Central Regulatory Role for SIN1 in Interferon gamma (IFNgamma) Signaling and Generation of Biological Responses. J Biol Chem

308. Rambold AS, K. B., Elia N, Lippincott-Schwartz J. (2011) Tubular network formation protects mitochondria from autophagosomal degradation during nutrient starvation. Proc Natl Acad Sci U S A 108, 10190-10195

180

309. Benard, G., Bellance, N., James, D., Parrone, P., Fernandez, H., Letellier, T., and Rossignol, R. (2007) Mitochondrial bioenergetics and structural network organization. J Cell Sci 120, 838-848

310. Zhao, N., Zhang, Y., Liu, Q., and Xiang, W. (2015) Mfn2 Affects Embryo Development via Mitochondrial Dysfunction and Apoptosis. PLoS One 10, e0125680

311. Frank, S., Gaume, B., Bergmann-Leitner, E. S., Leitner, W. W., Robert, E. G., Catez, F., Smith, C. L., and Youle, R. J. (2001) The role of dynamin-related protein 1, a mediator of mitochondrial fission, in apoptosis. Dev Cell 1, 515-525

312. Pearce, E. L., and Pearce, E. J. (2013) Metabolic pathways in immune cell activation and quiescence. Immunity 38, 633-643

313. Mills, E. L., and O'Neill, L. A. (2016) Reprogramming mitochondrial metabolism in macrophages as an anti-inflammatory signal. European Journal of Immunology 46, 13-21

314. Everts, B., Amiel, E., van der Windt, G. J. W., Freitas, T. C., Chott, R., Yarasheski, K. E., Pearce, E. L., and Pearce, E. J. (2012) Commitment to glycolysis sustains survival of NO-producing inflammatory dendritic cells. Blood 120, 1422-1431

315. Schertzer, J. D., and Steinberg, G. R. (2014) Immunometabolism: the interface of immune and metabolic responses in disease. Immunol Cell Biol 92, 303

316. Leboucher, G. P., Tsai, Y. C., Yang, M., Shaw, K. C., Zhou, M., Veenstra, T. D., Glickman, M. H., and Weissman, A. M. (2012) Stress-induced phosphorylation and proteasomal degradation of mitofusin 2 facilitates mitochondrial fragmentation and apoptosis. Mol Cell 47, 547-557

317. Sorace JM, J. R., Howard DL, Drysdale BE. (1995) Identification of an endotoxin and IFN-inducible cDNA- possible identification of a novel protein family. J Leukoc Biol 58, 477-484

318. Schwartz, Y., and Svistelnik, A. V. (2012) Functional phenotypes of macrophages and the M1-M2 polarization concept. Part I. Proinflammatory phenotype. Biochemistry (Mosc) 77, 246-260

319. Torres, A., Makowski, L., and Wellen, K. E. (2016) Immunometabolism: Metabolism fine-tunes macrophage activation. Elife 5

181

320. Hurst, R. D., and Clark, J. B. (1997) Nitric oxide-induced blood-brain barrier dysfunction is not mediated by inhibition of mitochondrial respiratory chain activity and/or energy depletion. Nitric Oxide 1, 121-129

321. Zhang, C., Hein, T. W., Wang, W., Ren, Y., Shipley, R. D., and Kuo, L. (2006) Activation of JNK and xanthine oxidase by TNF-alpha impairs nitric oxide- mediated dilation of coronary arterioles. J Mol Cell Cardiol 40, 247-257

322. Zhao, H., Ma, J. K., Barger, M. W., Mercer, R. R., Millecchia, L., Schwegler-Berry, D., Castranova, V., and Ma, J. Y. (2009) Reactive oxygen species- and nitric oxide- mediated lung inflammation and mitochondrial dysfunction in wild-type and iNOS-deficient mice exposed to diesel exhaust particles. J Toxicol Environ Health A 72, 560-570

323. Kien, C. L., Matthews, D. E., Poynter, M. E., Bunn, J. Y., Fukagawa, N. K., Crain, K. I., Ebenstein, D. B., Tarleton, E. K., Stevens, R. D., Koves, T. R., and Muoio, D. M. (2015) Increased palmitate intake: higher acylcarnitine concentrations without impaired progression of beta-oxidation. J Lipid Res 56, 1795-1807

324. McCoin, C. S., Knotts, T. A., and Adams, S. H. (2015) Acylcarnitines--old actors auditioning for new roles in metabolic physiology. Nat Rev Endocrinol 11, 617-625

325. Rutkowsky, J. M., Knotts, T. A., Ono-Moore, K. D., McCoin, C. S., Huang, S., Schneider, D., Singh, S., Adams, S. H., and Hwang, D. H. (2014) Acylcarnitines activate proinflammatory signaling pathways. Am J Physiol Endocrinol Metab 306, E1378-1387

326. Mills, E. L., and O'Neill, L. A. (2016) Reprogramming mitochondrial metabolism in macrophages as an anti-inflammatory signal. Eur J Immunol 46, 13-21

327. Johnson, A. R., Qin, Y., Cozzo, A. J., Freemerman, A. J., Huang, M. J., Zhao, L., Sampey, B. P., Milner, J. J., Beck, M. A., Damania, B., Rashid, N., Galanko, J. A., Lee, D. P., Edin, M. L., Zeldin, D. C., Fueger, P. T., Dietz, B., Stahl, A., Wu, Y., Mohlke, K. L., and Makowski, L. (2016) Metabolic reprogramming through fatty acid transport protein 1 (FATP1) regulates macrophage inflammatory potential and adipose inflammation. Mol Metab 5, 506-526

328. Saitoh, T., Fujita, N., Jang, M. H., Uematsu, S., Yang, B. G., Satoh, T., Omori, H., Noda, T., Yamamoto, N., Komatsu, M., Tanaka, K., Kawai, T., Tsujimura, T., Takeuchi, O., Yoshimori, T., and Akira, S. (2008) Loss of the autophagy protein Atg16L1 enhances endotoxin-induced IL-1beta production. Nature 456, 264-268

182

329. Park, J., Choi, H., Min, J. S., Park, S. J., Kim, J. H., Park, H. J., Kim, B., Chae, J. I., Yim, M., and Lee, D. S. (2013) Mitochondrial dynamics modulate the expression of pro-inflammatory mediators in microglial cells. J Neurochem 127, 221-232

330. Y, W. (2014) Roles of Macrophage Mitochondrial Oxidative Stress and Mitochondrial Fission in Atherosclerosis.

331. Wang, Y., and Tabas, I. (2014) Emerging roles of mitochondria ROS in atherosclerotic lesions: causation or association? J Atheroscler Thromb 21, 381-390

332. Dan Dunn, J., Alvarez, L. A., Zhang, X., and Soldati, T. (2015) Reactive oxygen species and mitochondria: A nexus of cellular homeostasis. Redox Biol 6, 472-485

333. Freitag, J., Berod, L., Kamradt, T., and Sparwasser, T. (2016) Immunometabolism and autoimmunity. Immunol Cell Biol

334. Glass, C. K., and Natoli, G. (2016) Molecular control of activation and priming in macrophages. Nat Immunol 17, 26-33

335. Perkins, D. J., Patel, M. C., Blanco, J. C., and Vogel, S. N. (2016) Epigenetic Mechanisms Governing Innate Inflammatory Responses. J Interferon Cytokine Res 36, 454-461

336. Li, P., Zhao, Y., Wu, X., Xia, M., Fang, M., Iwasaki, Y., Sha, J., Chen, Q., Xu, Y., and Shen, A. (2012) Interferon gamma (IFN-gamma) disrupts energy expenditure and metabolic homeostasis by suppressing SIRT1 transcription. Nucleic Acids Res 40, 1609-1620

337. Meiser, J., Kramer, L., Sapcariu, S. C., Battello, N., Ghelfi, J., D'Herouel, A. F., Skupin, A., and Hiller, K. (2016) Pro-inflammatory Macrophages Sustain Pyruvate Oxidation through Pyruvate Dehydrogenase for the Synthesis of Itaconate and to Enable Cytokine Expression. J Biol Chem 291, 3932-3946

338. Buck, M. D., O'Sullivan, D., Klein Geltink, R. I., Curtis, J. D., Chang, C. H., Sanin, D. E., Qiu, J., Kretz, O., Braas, D., van der Windt, G. J., Chen, Q., Huang, S. C., O'Neill, C. M., Edelson, B. T., Pearce, E. J., Sesaki, H., Huber, T. B., Rambold, A. S., and Pearce, E. L. (2016) Mitochondrial Dynamics Controls T Cell Fate through Metabolic Programming. Cell 166, 63-76

339. Yu, T., Robotham, J. L., and Yoon, Y. (2006) Increased production of reactive oxygen species in hyperglycemic conditions requires dynamic change of mitochondrial morphology. Proc Natl Acad Sci U S A 103, 2653-2658

183

340. Ayanga, B. A., Badal, S. S., Wang, Y., Galvan, D. L., Chang, B. H., Schumacker, P. T., and Danesh, F. R. (2016) Dynamin-Related Protein 1 Deficiency Improves Mitochondrial Fitness and Protects against Progression of Diabetic Nephropathy. J Am Soc Nephrol 27, 2733-2747

341. Hahn, W. S., Kuzmicic, J., Burrill, J. S., Donoghue, M. A., Foncea, R., Jensen, M. D., Lavandero, S., Arriaga, E. A., and Bernlohr, D. A. (2014) Proinflammatory cytokines differentially regulate adipocyte mitochondrial metabolism, oxidative stress, and dynamics. Am J Physiol Endocrinol Metab 306, E1033-1045

342. Hashemi, H. F., and Goodman, J. M. (2015) The life cycle of lipid droplets. Curr Opin Cell Biol 33, 119-124

343. Wilfling, F., Haas, J. T., Walther, T. C., and Farese, R. V., Jr. (2014) Lipid droplet biogenesis. Curr Opin Cell Biol 29, 39-45

344. Anderson, E. K., Hill, A. A., and Hasty, A. H. (2012) Stearic acid accumulation in macrophages induces toll-like receptor 4/2-independent inflammation leading to endoplasmic reticulum stress-mediated apoptosis. Arterioscler Thromb Vasc Biol 32, 1687-1695

345. Leroux, A., Ferrere, G., Godie, V., Cailleux, F., Renoud, M. L., Gaudin, F., Naveau, S., Prevot, S., Makhzami, S., Perlemuter, G., and Cassard-Doulcier, A. M. (2012) Toxic lipids stored by Kupffer cells correlates with their pro- inflammatory phenotype at an early stage of steatohepatitis. J Hepatol 57, 141-149

346. Kennedy, A., Martinez, K., Chuang, C. C., LaPoint, K., and McIntosh, M. (2009) Saturated fatty acid-mediated inflammation and insulin resistance in adipose tissue: mechanisms of action and implications. J Nutr 139, 1-4

347. Unger, R. H., Clark, G. O., Scherer, P. E., and Orci, L. (2010) Lipid homeostasis, lipotoxicity and the metabolic syndrome. Biochim Biophys Acta 1801, 209-214

348. Rambold, A. S., Cohen, S., and Lippincott-Schwartz, J. (2015) Fatty acid trafficking in starved cells: regulation by lipid droplet lipolysis, autophagy, and mitochondrial fusion dynamics. Dev Cell 32, 678-692

349. Penzo, D., Tagliapietra, C., Colonna, R., Petronilli, V., and Bernardi, P. (2002) Effects of fatty acids on mitochondria: implications for cell death. Biochim Biophys Acta 1555, 160-165

184

350. Freigang, S., Ampenberger, F., Weiss, A., Kanneganti, T. D., Iwakura, Y., Hersberger, M., and Kopf, M. (2013) Fatty acid-induced mitochondrial uncoupling elicits inflammasome-independent IL-1alpha and sterile vascular inflammation in atherosclerosis. Nat Immunol 14, 1045-1053

351. Xia, F., Li, R., Wang, C., Yang, S., Tian, L., Dong, H., Pei, C., He, S., Jiang, P., Cheng, H., Fang, S., Li, H., and Xu, H. (2013) IRGM1 regulates oxidized LDL uptake by macrophage via actin-dependent receptor internalization during atherosclerosis. Sci Rep 3, 1867

352. Lin, Y. C., Chang, P. F., Lin, H. F., Liu, K., Chang, M. H., and Ni, Y. H. (2016) Variants in the autophagy-related gene IRGM confer susceptibility to non- alcoholic fatty liver disease by modulating lipophagy. J Hepatol

353. Byles, V., Covarrubias, A. J., Ben-Sahra, I., Lamming, D. W., Sabatini, D. M., Manning, B. D., and Horng, T. (2013) The TSC-mTOR pathway regulates macrophage polarization. Nat Commun 4, 2834

354. Wei, W. C., Liu, C. P., Yang, W. C., Shyur, L. F., Sheu, J. H., Chen, S. S., and Yang, N. S. (2015) Mammalian target of rapamycin complex 2 (mTORC2) regulates LPS-induced expression of IL-12 and IL-23 in human dendritic cells. J Leukoc Biol 97, 1071-1080

355. Weichhart, T., Hengstschlager, M., and Linke, M. (2015) Regulation of innate immune cell function by mTOR. Nat Rev Immunol 15, 599-614

356. Jia, W., Pua, H. H., Li, Q. J., and He, Y. W. (2011) Autophagy regulates endoplasmic reticulum homeostasis and calcium mobilization in T lymphocytes. J Immunol 186, 1564-1574

357. Xu, J. (2005) Preparation, culture, and immortalization of mouse embryonic fibroblasts. Curr Protoc Mol Biol Chapter 28, Unit 28 21

358. Todaro, G. J., and Green, H. (1963) Quantitative studies of the growth of mouse embryo cells in culture and their development into established lines. J Cell Biol 17, 299-313

359. An, J., Muoio, D. M., Shiota, M., Fujimoto, Y., Cline, G. W., Shulman, G. I., Koves, T. R., Stevens, R., Millington, D., and Newgard, C. B. (2004) Hepatic expression of malonyl-CoA decarboxylase reverses muscle, liver and whole-animal insulin resistance. Nat Med 10, 268-274

185

360. Wu, J. Y., Kao, H. J., Li, S. C., Stevens, R., Hillman, S., Millington, D., and Chen, Y. T. (2004) ENU mutagenesis identifies mice with mitochondrial branched-chain aminotransferase deficiency resembling human maple syrup urine disease. J Clin Invest 113, 434-440

361. Jensen, M. V., Joseph, J. W., Ilkayeva, O., Burgess, S., Lu, D., Ronnebaum, S. M., Odegaard, M., Becker, T. C., Sherry, A. D., and Newgard, C. B. (2006) Compensatory responses to pyruvate carboxylase suppression in islet beta-cells. Preservation of glucose-stimulated insulin secretion. J Biol Chem 281, 22342-22351

362. Delgado, M., Singh, S., De Haro, S., Master, S., Ponpuak, M., Dinkins, C., Ornatowski, W., Vergne, I., and Deretic, V. (2009) Autophagy and pattern recognition receptors in innate immunity. Immunol Rev 227, 189-202

363. Deretic, V. (2011) Autophagy in immunity and cell-autonomous defense against intracellular microbes. Immunol Rev 240, 92-104

364. Gregoire, I. P., Richetta, C., Meyniel-Schicklin, L., Borel, S., Pradezynski, F., Diaz, O., Deloire, A., Azocar, O., Baguet, J., Le Breton, M., Mangeot, P. E., Navratil, V., Joubert, P. E., Flacher, M., Vidalain, P. O., Andre, P., Lotteau, V., Biard- Piechaczyk, M., Rabourdin-Combe, C., and Faure, M. (2011) IRGM is a common target of RNA viruses that subvert the autophagy network. PLoS Pathog 7, e1002422

365. Galvan-Pena, S., and O'Neill, L. A. (2014) Metabolic reprograming in macrophage polarization. Front Immunol 5, 420

366. Ivashkiv, L. B. (2013) Epigenetic regulation of macrophage polarization and function. Trends Immunol 34, 216-223

367. Yin, J., Wang, Y., Gu, L., Fan, N., Ma, Y., and Peng, Y. (2015) Palmitate induces endoplasmic reticulum stress and autophagy in mature adipocytes: implications for apoptosis and inflammation. Int J Mol Med 35, 932-940

368. McCoin, C. S., Knotts, T. A., Ono-Moore, K. D., Oort, P. J., and Adams, S. H. (2015) Long-chain acylcarnitines activate cell stress and myokine release in C2C12 myotubes: calcium-dependent and -independent effects. Am J Physiol Endocrinol Metab 308, E990-E1000

369. Cao, Y., Rathmell, J. C., and Macintyre, A. N. (2014) Metabolic reprogramming towards aerobic glycolysis correlates with greater proliferative ability and resistance to metabolic inhibition in CD8 versus CD4 T cells. PLoS One 9, e104104 186

370. Artyomov, M. N., Sergushichev, A., and Schilling, J. D. (2016) Integrating immunometabolism and macrophage diversity. Semin Immunol

371. Jogl, G., Hsiao, Y. S., and Tong, L. (2004) Structure and function of carnitine acyltransferases. Ann N Y Acad Sci 1033, 17-29

372. Lu, Y. C., Yeh, W. C., and Ohashi, P. S. (2008) LPS/TLR4 signal transduction pathway. Cytokine 42, 145-151

373. Venter, G., Oerlemans, F. T., Wijers, M., Willemse, M., Fransen, J. A., and Wieringa, B. (2014) Glucose controls morphodynamics of LPS-stimulated macrophages. PLoS One 9, e96786

374. Martinez, F. O., Gordon, S., Locati, M., and Mantovani, A. (2006) Transcriptional Profiling of the Human Monocyte-to-Macrophage Differentiation and Polarization: New Molecules and Patterns of Gene Expression. The Journal of Immunology 177, 7303-7311

375. Kratz, M., Coats, B. R., Hisert, K. B., Hagman, D., Mutskov, V., Peris, E., Schoenfelt, K. Q., Kuzma, J. N., Larson, I., Billing, P. S., Landerholm, R. W., Crouthamel, M., Gozal, D., Hwang, S., Singh, P. K., and Becker, L. (2014) Metabolic dysfunction drives a mechanistically distinct proinflammatory phenotype in adipose tissue macrophages. Cell Metab 20, 614-625

376. Guo, H., Callaway, J. B., and Ting, J. P. (2015) Inflammasomes: mechanism of action, role in disease, and therapeutics. Nat Med 21, 677-687

377. Boutens, L., and Stienstra, R. (2016) Adipose tissue macrophages: going off track during obesity. Diabetologia 59, 879-894

378. Lumeng, C. N., Bodzin, J. L., and Saltiel, A. R. (2007) Obesity induces a phenotypic switch in adipose tissue macrophage polarization. J Clin Invest 117, 175-184

379. Lumeng, C. N., DelProposto, J. B., Westcott, D. J., and Saltiel, A. R. (2008) Phenotypic switching of adipose tissue macrophages with obesity is generated by spatiotemporal differences in macrophage subtypes. Diabetes 57, 3239-3246

380. Zacharioudaki, V., Androulidaki, A., Arranz, A., Vrentzos, G., Margioris, A. N., and Tsatsanis, C. (2009) Adiponectin promotes endotoxin tolerance in macrophages by inducing IRAK-M expression. J Immunol 182, 6444-6451

187

381. Yeligar, S. M., Machida, K., Tsukamoto, H., and Kalra, V. K. (2009) Ethanol augments RANTES/CCL5 expression in rat liver sinusoidal endothelial cells and human endothelial cells via activation of NF-kappa B, HIF-1 alpha, and AP-1. J Immunol 183, 5964-5976

382. Panee, J. (2012) Monocyte Chemoattractant Protein 1 (MCP-1) in obesity and diabetes. Cytokine 60, 1-12

383. Getz GS, R. C. (2014) The mutual interplay of lipid metabolism and the cells of the immune system in relation to atherosclerosis. Clin Lipidol 9, 657

384. Dushkin, M. I. (2012) Macrophage/foam cell is an attribute of inflammation: mechanisms of formation and functional role. Biochemistry (Mosc) 77, 327-338

385. Das, K., and Kar, P. (2005) Non-alcoholic steatohepatitis. J Assoc Physicians India 53, 195-199

386. Dougan, G., Chatfield, S., Pickard, D., Bester, J., O'Callaghan, D., and Maskell, D. (1988) Construction and characterization of vaccine strains of Salmonella harboring mutations in two different aro genes. J Infect Dis 158, 1329-1335

387. Linossi, E. M., Babon, J. J., Hilton, D. J., and Nicholson, S. E. (2013) Suppression of cytokine signaling: the SOCS perspective. Cytokine Growth Factor Rev 24, 241- 248

388. Covarrubias, A. J., Aksoylar, H. I., and Horng, T. (2015) Control of macrophage metabolism and activation by mTOR and Akt signaling. Semin Immunol 27, 286- 296

389. Song, G., Ouyang, G., and Bao, S. (2005) The activation of Akt/PKB signaling pathway and cell survival. J Cell Mol Med 9, 59-71

390. Coers, J., and Haldar, A. K. (2015) Ubiquitination of pathogen-containing vacuoles promotes host defense to Chlamydia trachomatis and Toxoplasma gondii. Commun Integr Biol 8, e1115163

391. Lee, Y. G., Lee, J., Byeon, S. E., Yoo, D. S., Kim, M. H., Lee, S. Y., and Cho, J. Y. (2011) Functional role of Akt in macrophage-mediated innate immunity. Front Biosci (Landmark Ed) 16, 517-530

392. Androulidaki, A., Iliopoulos, D., Arranz, A., Doxaki, C., Schworer, S., Zacharioudaki, V., Margioris, A. N., Tsichlis, P. N., and Tsatsanis, C. (2009) The

188

kinase Akt1 controls macrophage response to lipopolysaccharide by regulating microRNAs. Immunity 31, 220-231

393. Arranz, A., Doxaki, C., Vergadi, E., Martinez de la Torre, Y., Vaporidi, K., Lagoudaki, E. D., Ieronymaki, E., Androulidaki, A., Venihaki, M., Margioris, A. N., Stathopoulos, E. N., Tsichlis, P. N., and Tsatsanis, C. (2012) Akt1 and Akt2 protein kinases differentially contribute to macrophage polarization. Proc Natl Acad Sci U S A 109, 9517-9522

394. Levine, B., Mizushima, N., and Virgin, H. W. (2011) Autophagy in immunity and inflammation. Nature 469, 323-335

395. Arranz, A., Androulidaki, A., Zacharioudaki, V., Martinez, C., Margioris, A. N., Gomariz, R. P., and Tsatsanis, C. (2008) Vasoactive intestinal peptide suppresses toll-like receptor 4 expression in macrophages via Akt1 reducing their responsiveness to lipopolysaccharide. Mol Immunol 45, 2970-2980

396. Larson-Casey, J. L., Deshane, J. S., Ryan, A. J., Thannickal, V. J., and Carter, A. B. (2016) Macrophage Akt1 Kinase-Mediated Mitophagy Modulates Apoptosis Resistance and Pulmonary Fibrosis. Immunity 44, 582-596

397. Xu, Y., He, Z., Li, Z., Fang, S., Zhang, Y., Wan, C., Ma, Y., Lin, P., Liu, C., Wang, G., Li, R., Zhu, J., Li, Y., Mu, L., Zhang, Y., Wang, J., Kong, Q., Li, H., and Sun, B. (2017) Irgm1 is required for the inflammatory function of M1 macrophage in early experimental autoimmune encephalomyelitis. J Leukoc Biol 101, 507-517

398. Weber, K., and Schilling, J. D. (2014) Lysosomes integrate metabolic- inflammatory cross-talk in primary macrophage inflammasome activation. J Biol Chem 289, 9158-9171

399. Emanuel, R., Sergin, I., Bhattacharya, S., Turner, J. N., Epelman, S., Settembre, C., Diwan, A., Ballabio, A., and Razani, B. (2014) Induction of lysosomal biogenesis in atherosclerotic macrophages can rescue lipid-induced lysosomal dysfunction and downstream sequelae. Arterioscler Thromb Vasc Biol 34, 1942-1952

400. Lin, Y. C., Chang, P. F., Lin, H. F., Liu, K., Chang, M. H., and Ni, Y. H. (2016) Variants in the autophagy-related gene IRGM confer susceptibility to non- alcoholic fatty liver disease by modulating lipophagy. J Hepatol 65, 1209-1216

401. Dong, H., Tian, L., Li, R., Pei, C., Fu, Y., Dong, X., Xia, F., Wang, C., Li, W., Guo, X., Gu, C., Li, B., Liu, A., Ren, H., Wang, C., and Xu, H. (2015) IFNg-induced

189

Irgm1 promotes tumorigenesis of melanoma via dual regulation of apoptosis and Bif-1-dependent autophagy. Oncogene 34, 5363-5371

402. Arrese, E. L., Saudale, F. Z., and Soulages, J. L. (2014) Lipid Droplets as Signaling Platforms Linking Metabolic and Cellular Functions. Lipid Insights 7, 7-16

403. Singh, R., and Cuervo, A. M. (2012) Lipophagy: connecting autophagy and lipid metabolism. Int J Cell Biol 2012, 282041

404. Ouimet, M., Franklin, V., Mak, E., Liao, X., Tabas, I., and Marcel, Y. L. (2011) Autophagy regulates cholesterol efflux from macrophage foam cells via lysosomal acid lipase. Cell Metab 13, 655-667

405. Toledo, D. A., D'Avila, H., and Melo, R. C. (2016) Host Lipid Bodies as Platforms for Intracellular Survival of Protozoan Parasites. Front Immunol 7, 174

406. Padilla, A. M., Simpson, L. J., and Tarleton, R. L. (2009) Insufficient TLR activation contributes to the slow development of CD8+ T cell responses in Trypanosoma cruzi infection. J Immunol 183, 1245-1252

407. Toledo, D. A., Roque, N. R., Teixeira, L., Milan-Garces, E. A., Carneiro, A. B., Almeida, M. R., Andrade, G. F., Martins, J. S., Pinho, R. R., Freire-de-Lima, C. G., Bozza, P. T., D'Avila, H., and Melo, R. C. (2016) Lipid Body Organelles within the Parasite Trypanosoma cruzi: A Role for Intracellular Arachidonic Acid Metabolism. PLoS One 11, e0160433

408. Cestari, I., Ansa-Addo, E., Deolindo, P., Inal, J. M., and Ramirez, M. I. (2012) Trypanosoma cruzi immune evasion mediated by host cell-derived microvesicles. J Immunol 188, 1942-1952

409. Lovewell, R. R., Sassetti, C. M., and VanderVen, B. C. (2016) Chewing the fat: lipid metabolism and homeostasis during M. tuberculosis infection. Curr Opin Microbiol 29, 30-36

410. Singh, V., Jamwal, S., Jain, R., Verma, P., Gokhale, R., and Rao, K. V. (2012) Mycobacterium tuberculosis-driven targeted recalibration of macrophage lipid homeostasis promotes the foamy phenotype. Cell Host Microbe 12, 669-681

411. Recuero-Checa, M. A., Sharma, M., Lau, C., Watkins, P. A., Gaydos, C. A., and Dean, D. (2016) Chlamydia trachomatis growth and development requires the activity of host Long-chain Acyl-CoA Synthetases (ACSLs). Sci Rep 6, 23148

190

412. Rabhi, S., Rabhi, I., Trentin, B., Piquemal, D., Regnault, B., Goyard, S., Lang, T., Descoteaux, A., Enninga, J., and Guizani-Tabbane, L. (2016) Lipid Droplet Formation, Their Localization and Dynamics during Leishmania major Macrophage Infection. PLoS One 11, e0148640

191

Biography

Elyse Anne Schmidt was born on September 30th, 1986 in Montreal, Quebec,

Canada. She grew up with her younger brother, David, and her parents, John and

Nancy Schmidt, in the borough L’Ile Bizard on the west side of the island of Montreal. In

2006 she attended Concordia University in downtown Montreal, where she earned her bachelors of science in honors biochemistry. During her degree, she participated in the work/study (or co-op) program at Concordia University. Her stages included working at a pulp and paper research institute, FP Innovations, optimizing a paper pulp bleaching technique, working at Boheringer Ingleheim on HIV binding assays, and at the National

Research Institute Armand Frappier, where she studied protein splicing in the VZV and

HSV1 viruses. She graduated as the valedictorian of the Arts and Sciences Faculty in the spring of 2010.

Elyse was accepted into the Department of Molecular Genetics and Microbiology at Duke University and moved to Durham, NC in August 2010. She joined the lab of Dr.

Gregory Taylor to study the immunity related GTPases, an important innate immune system in mice. She received a pre-doctoral fellowship from the National Science and

Engineering Council of Canada during her tenure. In 2017 she published a paper entitled “Metabolic alterations contribute to enhanced inflammatory cytokine production in Irgm1-deficient macrophages” in the Journal of Biological Chemistry.

192