University of Massachusetts Medical School eScholarship@UMMS
GSBS Dissertations and Theses Graduate School of Biomedical Sciences
2021-07-21
IFNγ Mediated Monocyte Metabolic Reprogramming
Katelyn J. McCann University of Massachusetts Medical School
Let us know how access to this document benefits ou.y Follow this and additional works at: https://escholarship.umassmed.edu/gsbs_diss
Part of the Allergy and Immunology Commons, Endocrinology, Diabetes, and Metabolism Commons, Immunology and Infectious Disease Commons, and the Infectious Disease Commons
Repository Citation McCann KJ. (2021). IFNγ Mediated Monocyte Metabolic Reprogramming. GSBS Dissertations and Theses. https://doi.org/10.13028/4zcp-ht79. Retrieved from https://escholarship.umassmed.edu/ gsbs_diss/1146
This material is brought to you by eScholarship@UMMS. It has been accepted for inclusion in GSBS Dissertations and Theses by an authorized administrator of eScholarship@UMMS. For more information, please contact [email protected]. 1
IFNγ MEDIATED MONOCYTE METABOLIC REPROGRAMMING A Dissertation Presented By KATELYN JOSEPHINE MCCANN This work was undertaken in the Graduate School of Biomedical Science MD/PhD program in collaboration with the NIH/NIAID Graduate Partnership Program
Under the mentorship of: Dr. Beth A. McCormick and Dr. Steven M. Holland, Thesis Advisors Dr. Christopher Sassetti, Chair of Thesis Committee Dr. Kate Fitzgerald, Chair of Defense Committee Dr. Javier Irazoqui, Member of Committee Dr. Read Pukkila-Worley, Member of Committee Dr. Peter Newburger, Member of Committee Dr. Mihalis Lionakis, External Member of Committee
Dr. Mary Ellen Lane Dean of the Graduate School of Biomedical Sciences
July 21, 2021
2
Reviewer Page
Dr. Kate Fitzgerald: ______Dr. Javier Irazoqui: ______Dr. Read Pukkila-Worley: ______Dr. Peter Newburger: ______Dr. Mihalis Lionakis: ______
3
Dedication:
For my family and friends, without whom, this would not have been possible.
4
Acknowledgements:
Reflecting on my past 7 years as an MD/PhD student at UMass and the past 5 years as a graduate student at NIH, I couldn’t possibly thank all the people who have helped me along the way. I have had the opportunity to start this program with an amazing group of people, some of whom I will be going back to medical school with very soon! I was also fortunate enough to be a member of 2 wonderful labs. The McCormick lab has welcomed me as one of their own for the past 5 years despite my being hundreds of miles away. I can’t imagine a group of people more kind, generous and thoughtful to join from afar and I look forward to seeing you all in person when I am back in Worcester. Being a member of the Holland lab has been a once in a lifetime opportunity to work with and learn from some of the most amazing scientists and clinicians who are not only experts in their field, but also the best teachers and collaborators I could have hoped for. I was also fortunate to be welcomed into the Sher/Mayer-Barber lab which became a 3rd home. Thank you to Alan, Kat, Dragana, Edu, Logan, Kate and Caio for your generous support, advice, and teaching. Thank you for introducing me to the world of mouse immunology and for the opportunity to join your lab meetings which were always a helpful, collaborative, and exciting experience. I also have to thank my TRAC committee members who have been advising me since day 1. I’m not sure if anyone looked forward to their TRAC meetings as much as I did. Every meeting was full of discussion that I learned so much just from. I received the most insightful and practical feedback, developed new ideas and enjoyed talking to a group of mentors who were always engaged and encouraging. Chris, thank you for leading my TRAC, keeping my project focused and for all the interesting discussion and collaboration related to metabolism. Read, I’m lucky to have had the opportunity to work with you in the clinical and research worlds. Thank you for your honest and thoughtful advice and your constant willingness to teach which has helped me in all aspects of my training. Javier, thank you for always being so engaged and encouraging. Your suggestions made obstacles that seemed like huge disappointments feel like exciting new questions to answer and I might have given up a long time ago without your motivation. Kate, thank you for your input on everything from the big picture questions to the smallest technical nuances, your advice and expertise has been invaluable. Most importantly, I would like to thank my mentors who made it possible for me to do this work and helped me all along the way. Beth, your efforts to make this whole arrangement work have been so generous. From unruly Skype calls (before everyone was Zooming) to moments of panic, to the many times my project felt stuck, you have bent over backwards to support me and you’ve always done it with a smile. It has been such a pleasure to learn from you and I hope I have the opportunity to pass along your words of wisdom as a mentor some day. Steve, I can’t thank you enough for the opportunity to join your lab and work with you for the past 5 years. I don’t think there is a better physician scientist role model than you. I have learned so much and been so inspired just witnessing how you think and work. And somehow, despite having the schedule of the scientific director of NIAID during a global pandemic, you have always made time for my questions, concerns and thoughts, no matter how insignificant or metabolism-related they were. Your thoughtful and insightful advice encouraged me to keep going even when it felt like nothing was making sense. Thank you for always encouraging me to work on the things I was interested in and for sharing in my excitement to study patients even when the results were complicated. You have supported my work, my training and my career goals and you have given me opportunities that I did not deserve but truly appreciate. 5
Abstract:
IFNγ is an essential and pleiotropic activator of monocytes, but little is known about the effects IFNγ on cellular metabolism. Therefore, we sought to characterize and elucidate the mechanisms by which IFNγ reprograms monocyte metabolism to support its immunologic activities. First, we identified a critical role for IFNγ in the induction of immunoresponsive gene
1 (IRG1) and its product, itaconate. The immunometabolite, itaconate, has been reported to have antibacterial, anti-inflammatory and antioxidant activity. Irg1-/- mice, lacking itaconate, are highly susceptible and phenotypically similar to IFNγ knock out (GKO) mice upon infection with Mycobacterium tuberculosis. Therefore, we assessed the role of IRG1/itaconate in the context of non-tuberculous mycobacterial (NTM) infection, the most common type of infection in patients with immunodeficiencies caused by defects in IFNγ signaling. Our data suggest that impaired induction of itaconate in the context of mycobacterial infection may contribute to mycobacterial susceptibility and immune dysregulation in patients with defects in IFNγ signaling. Next, we evaluated the metabolic phenotype of IFNγ-stimulated human monocytes and found that IFNγ increased oxygen consumption rates (OCR), indicative of reactive oxygen species generation by both mitochondria and NADPH oxidase. Transcriptional profiling of human macrophages revealed that this oxidative phenotype was dependent on IFNγ-induced, nicotinamide phosphoribosyltransferase (NAMPT)-mediated NAD+ salvage to generate NADH and NADPH for oxidation by mitochondrial complex I and NADPH oxidase, respectively. These data identify an IFNγ-induced, NAMPT-dependent, NAD+ salvage pathway that is critical for complete induction of the respiratory burst in IFNγ stimulated human monocytes.
6
Table of Contents Dedication: ...... 3 Acknowledgements: ...... 4 Abstract: ...... 5 List of Tables: ...... 8 List of Figures: ...... 8 List of copyrighted materials produced by the author ...... 9 List of third party copyrighted material ...... 9 List of symbols, Abbreviations or Nomenclature ...... 9 CHAPTER I: Introduction ...... 11 Identification, structure and function of IFNγ and the IFNγ Receptor: ...... 11 IFNγ-regulated gene expression: ...... 13 Clinical Phenotypes of IFNγ Dysregulation: ...... 17 Regulation and function of Immunoresponsive gene 1 (IRG1): ...... 22 The Immunometabolite, itaconate, and its proposed functions: ...... 25 Macrophage metabolic reprogramming and the relationship between metabolism and immunity: .. 30 CHAPTER II: IRG1 and Itaconate contribute to the Interferon Gamma-Induced Host Defense Against Mycobacterium avium ...... 37 Introduction: ...... 38 Results: ...... 40 IRG1 Expression and Itaconate Production are synergistically enhanced by IFNγ in the Context of Toll- Like Receptor (TLR) stimulation ...... 40 Cell Intrinsic Responses to M. avium Infection are Not Significantly Altered in Irg1-/- Bone Marrow Derived Macrophages ...... 46 Irg1-/- mice have increased mycobacterial burdens following in vivo infection with M. avium ...... 50 Discussion: ...... 54 CHAPTER III: IFNγ Regulates NAD+ Metabolism in Human Monocytes ...... 58 Introduction: ...... 59 Results: ...... 61 IFNγ increases monocyte oxygen consumption rates (OCR) ...... 61 IFNγ-induced oxygen consumption is dependent on NAMPT mediated NAD+ salvage ...... 64 STAT1 is required for IFNγ-induced oxygen consumption and regulates oxygen consumption via mitochondrial complex I ...... 69 PMA-stimulated Oxygen Consumption Rates measure NADPH oxidase activity ...... 73 IFNγ coordinately regulates transcription of multiple pathways to promote both NAM-dependent and independent NAD salvage ...... 75 Discussion: ...... 80 7
CHAPTER IV: Discussion ...... 84 IRG1 and Itaconate Contribute to the Interferon Gamma-Induced Host Defense Against Mycobacterium avium ...... 86 Summary of major results and conclusions: ...... 86 Additional experiments and future directions: ...... 92 IFNγ Regulates NAD+ Metabolism in Human Monocytes ...... 96 Summary of major results and conclusions: ...... 96 Additional experiments and future directions: ...... 101 CHAPTER V: Materials and Methods ...... 105 Ethics Statement: ...... 105 Mice: ...... 105 In vivo infections: ...... 105 Harvest and differentiation of murine bone marrow derived macrophages: ...... 106 In vitro macrophage infections: ...... 106 Human PBMC collection, isolation and in vitro stimulation with M. avium: ...... 107 RNA isolation, cDNA synthesis and RT-PCR: ...... 107 LC-MS/MS Metabolic Analysis: ...... 107 Seahorse Metabolic Rate Assays with Bone Marrow Derived Macrophages: ...... 109 Human monocyte collection, isolation and stimulation: ...... 109 Seahorse Metabolic Rate Assays with Primary Human Monocytes: ...... 110 RNA sequencing: ...... 111 Statistical analyses: ...... 112 APPENDIX I: Clinical Presentation and Evaluation of a Patient with a Variant of Uncertain Significance in Immunoresponsive Gene 1 ...... 114 CHAPTER VI: References ...... 123
8
List of Tables: • Chapter III o Table 3.1: Patients assayed for metabolic dysregulation • Appendix I o Table A1: Detailed Course of Disease from Presentation Until Patient was Lost to Follow-up o Table A2: Gram Positive, Gram Negative and Anaerobic Bacteria Cultured from Patient’s Abscesses
List of Figures: • Chapter I o Figure 1.1: Metabolic pathways leading to IRG1 mediated itaconate biosynthesis. o Figure 1.2: IFNγ regulated metabolic pathways and their immunologic effects. • Chapter II o Figure 2.1: Irg1 mRNA Expression and Intracellular Itaconate Levels in Healthy Controls and Patients. o Figure 2.2: Irg1 mRNA Expression and Intracellular Itaconate Levels are Synergistically Induced by the Combination of LPS and IFNγ. o Figure 2.3: Irg1 Expression is Dysregulated in the Context of Primary Immunodeficiency Syndromes Affecting IFNγ Signaling. o Figure 2.4: IRG1 expression is a biomarker of IFNγ signaling activity in primary patient PBMC’s. o Figure 2.5: Cell Intrinsic Responses to M. avium Infection are Not Significantly Altered in Irg1-/- Bone Marrow Derived Macrophages. o Figure 2.6: In vitro assessment of IRG1 knock out in a human monocyte (THP-1) cell line. o Figure 2.7: Irg1-/- Mice have Higher Mycobacterial Burdens Following In Vivo Infection with Mycobacterium avium. o Figure 2.8: In vivo infections of Wild Type and Irg1-/- mice with M. avium • Chapter III o Figure 3.1: IFNγ Increases Monocyte Oxygen Consumption Rate (OCR). o Figure 3.2: Immediate and prolonged metabolic responses to LPS and/or IFNγ. o Figure 3.3: Mechanism of IFNγ induced OCR. o Figure 3.4: NAMPT is induced by IFNγ and is required for increases in basal and PMA stimulated oxygen consumption. o Figure 3.5: Figure 3.5: FK866 inhibits both OCR and ECAR in IFNγ stimulated monocytes. o Figure 3.6: STAT1 is required for IFNγ induced oxygen consumption and regulates oxygen consumption via mitochondrial complex I. o Figure 3.7: IFNγ induced increases in OCR are pSTAT1, NAMPT and Complex I dependent. o Figure 3.8: Chemical inhibition of mitochondrial complex I normalizes elevated IFNγ-induced PD-L1 and CD40 expression. 9
o Figure 3.9: PMA-stimulated Oxygen Consumption Rates Measure NADPH Oxidase Activity. o Figure 3.10: IFNγ coordinately regulates transcription of multiple pathways to promote both NAM-dependent and independent NAD salvage. o Figure 3.11: IFNγ regulates expression of genes involved in NAD(P)H oxidation and biosynthesis. o Figure 3.12: IFNγ-induced increases in OCR depend on electron transport, not ATP production. • Appendix I: o Figure A1: Severe pyomyositis caused by multiple bacteria, requiring fasciectomies o Figure A2: Amino Acid Sequence Alignment of IRG1 and Orthologous Decarboxylase Enzymes from Other Species. o Figure A3: Irg1 mRNA Expression and Intracellular Itaconate Levels are Synergistically Induced by the Combination of LPS and IFNγ. o Figure A4: Intracellular itaconate concentrations from a patient with and IRG1 variant (A214V) and healthy controls.
List of copyrighted materials produced by the author o none
List of third party copyrighted material o none
List of symbols, Abbreviations or Nomenclature o APECED: autoimmune polyendocrinopathy candidiasis ectodermal dystrophy o BCG: Bacillus Calmette–Guérin o BMDM: bone marrow derived macropahge o CMC: chronic mucocutaneous candidiasis o GAF: gamma-interferon activation factor o GAPDH: Glyceraldehyde-3-Phosphate Dehydrogenase o GAS: interferon gamma activated site o GOF: gain of function o HSCT: hematopoietic stem cell transplantation o IFNγ: interferon gamma o IFNGR: interferon gamma receptor o IL-12(R): interleukin-12 (receptor) o ISGF3: Interferon stimulated gene factor 3 o LC-MS/MS: liquid chromatography-tandem mass spectrometry o LOF: loss of function o (m)ROS: (mitochondrial) reactive oxygen species o Mtb: Mycobacterium tuberculosis o NAM: Nicotinamide o NAMPT: Nicotinamide Phosphoribosyltransferase 10 o NLRP3: nucleotide-binding domain, leucine-rich repeat-containing, pyrin domain containing 3 o NTM: non-tuberculous mycobacteria o OxPhos: oxidative phosphorylation o PBMC: peripheral blood mononuclear cells o PID: primary immunodeficiency syndrome o PMA: Phorbol myristate acetate o PPP: pentose phosphate pathway o RNS: reactive nitrogen species o SDH: succinate dehydrogenase o STAT: signal transducer and activator of transcription o TCA: tricarboxylic acid cycle
11
CHAPTER I: Introduction
Identification, structure, and function of IFNγ and the IFNγ Receptor:
Interferon gamma (IFNγ) was initially described as part of a broad class of biologically active molecules that interfere with viral replication [1]. These molecules were later defined as interferons and separated into types I and II based on their sequence homology and receptor specificity [2]. The type I interferon family includes IFNα, IFNβ, and IFNω which all signal through the IFNα receptor (IFNAR), whereas IFNγ is the only member of the type II interferon family and signals through the IFNγ receptor (IFNGR). [3]. IFNγ protein is encoded by a single
IFNG gene located on human chromosome 12, containing four exons and three introns [4]. IFNγ is highly conserved across mammalian species at the DNA level [5] and demonstrates even higher sequence homology in the promoter region than in protein coding regions [6].
IFNγ binds its receptor as a homodimer [7], facilitating its signaling activity [8, 9]. The
IFNγ receptor is composed of two ligand binding, IFNGR1, subunits and two signal transducing,
IFNGR2, subunits. IFNGR1 is ubiquitously and constitutively expressed on all nucleated cells and is poorly inducible [10]. The gene coding for IFNGR1 is located on chromosome 6 and consists of seven exons. The first five exons code for the extracellular domain with IFNγ binding capacity, while exon 6 codes for the transmembrane domain and exon 7 codes for the intracellular domain. IFNGR1 protein is a total of 472 amino acids and is N-glycosylated making its molecular weight approximately 90kDa [9, 11]. The intracellular domain of IFNGR1 contains
3 critical functional regions: 1) Jak1 binding site, 2) endocytosis/recycling domain, 3) STAT1 binding site [12].
The IFNGR2 gene is located on chromosome 21 and its expression is inducible but more limited than IFNGR1 both in terms of cell surface expression levels and cell types on which it is 12 expressed [13] [14]. IFNGR2 also has seven exons. The first 5 exons code for the extracellular domain (though there is no ligand binding region), exon 6 codes for the transmembrane domain and exon 7 codes for the intracellular domain. IFNGR2 protein is 316 amnio acids in total with a smaller intracellular domain than IFNGR1. Like the IFNGR1, IFNGR2 has distinct regions important for its function: 1) Jak2 binding site and 2) transmembrane domain involved in receptor endocytosis in a ligand independent manner [15].
In contrast to the early understanding of IFNGR signaling, the complete heterotetrameric
IFNγ receptor complex is assembled independent of IFNγ ligand binding [16] [8, 14, 17-19] [20].
With the extracellular domains of IFNGR2 bound to the IFNGR1 subunits, IFNγ binding to
IFNGR1 subunits induces a conformational change in the receptor complex that bring the cytoplasmic signal transducing domains into close proximity to initiate intracellular signaling
[16, 21, 22].
IFNGR1 is constitutively associated with inactive Janus kinase 1 (Jak1), bound to a four
amino acid sequence (266LPKS269) in its the intracellular domain [9, 21] and IFNGR2
constitutively binds JAK2 at a non-contiguous binding site (263PPSIP267 and 270IEEYL274) in its intracellular domain [9]. Upon ligand binding to the receptor complex, receptor associated Jak2 is activated by autophosphorylation, then transphosphorylates Jak1, activating Jak1 to
phosphorylate IFNGR1 at the Y440 site. Phosphorylation of IFNGR1 at Y440 allows for recruitment of one latent STAT1 molecule which binds to each phosphorylated IFNGR1 subunit in the receptor through an interaction between the SH2 domain of STAT1 and the IFNGR1
binding motif initiated by Y440 (440YDKPH444) [23, 24]. Receptor associated STAT1 molecules are then phosphorylated (probably by Jak2) at the STAT1 C-terminal Y701 site and form a 13 homodimer which dissociates from the receptor and translocates to the nucleus where it binds to
IFNγ activation sequences (GAS), regulating the expression of many genes [25, 26]
IFNγ-regulated gene expression:
The expression of IFNγ by T cells and NK cells is induced by interleukin 12 (IL-12) signaling. IL-12 is produced in response to toll-like receptor (TLR) signaling and is then secreted and acts to augment IFNγ production by signaling through the IL-12 receptor expressed on T cells and NK cells. IL-12 induced IFNγ then promotes enhanced IL-12 expression by innate cells and the positive feedback cycle continues. [27, 28] This process not only promotes IFNγ-induced cell mediated immunity, characteristic of the Th1 response, but also actively inhibits the Th2 response characterized by IL-4 production as opposed to IFNγ. [29-32]. The transcriptional changes induced by IFNγ are vast and pleiotropic and regulate a variety of cellular processes.
Binding of STAT1 to GAS elements regulates the expression of many genes, including some transcription factors which go on to regulate the expression of more genes. [2]. One subset of genes regulated by IFNγ are genes related to the antiviral response. IFNγ has been shown to increase MHC I gene expression (HLA-A, B and C) although the effect of IFNγ on expression of other effectors of the antiviral response are weak compared to that of type I interferons [33-35].
Furthermore, Groettrup et al. demonstrated that IFNγ also augments proteosome activity and expression levels to produce more peptides available for loading onto MCH I to improve antigen presentation [35]. These findings illustrate a unique role for IFNγ compared to type I interferons in the context of the antiviral response, where IFNγ (type II IFN) preferentially regulates genes involved in antigen processing and presentation while type I interferons regulate anti-viral effector processes more broadly. Additionally, compared to type I interferons, IFNγ has the 14 unique ability to upregulate MHC II (HLA-D) to promote antigen specific CD4+ T cell activation, and also induces expression of CIITA which serves as a regulator of MHC II complex gene expression, again promoting antigen presentation and facilitating the Th1 response [36-40].
IFNγ also has anti-proliferative and pro-apoptotic effects. It inhibits cellular proliferation through the regulation of cyclin-CDK complexes by IFNγ-induced p21 and p27. Cyclin-CDK complexes are inhibited by p21 and p27 resulting in hypophosphorylation of Rb and subsequent sequestration of the E3F family of transcription factors which regulate the expression of genes required for cell cycle progression such as c-myc. [41-43]. IFNγ has also been reported to regulate apoptosis in T lymphocytes in an IRF1-dependent manner that depends on the expression of IFNGR2 [17, 44]. The ability of IFNγ to induce apoptosis in the context of infection could be particularly advantageous for macrophages so that they die by a process of programmed cell death, as opposed to necrotic/lytic cell death, allowing them to contain their contents which may include pathogens that could propagate infection if released.
The inflammatory and antibacterial response to infection is also dependent on IFNγ- induced gene expression. IFNγ induces expression of pro-inflammatory cytokines and chemokines which augment the monocyte/macrophage response to infection and coordinate the adaptive immune response. As described above, the cross regulation of IFNγ and IL-12 induces a positive feedback cycle that augments and perpetuates the Th1 immune response [27, 28]. The
IFNγ-induced activation of macrophages results in the production of several important monocyte and T cell chemoattractants. CXCL9, also known as monokine induced by IFNγ (MIG) is a highly inducible, T cell chemoattractant [45], and CXCL10, also known as IFN-inducible protein
10 (IP10), acts as a chemoattractant for both monocytes and T cells [46]. Monocyte chemoattractant protein-1 (MCP-1), MIP1α (CCL3), MIP1ß (CCL4), and Regulated on 15 activation, normal T expressed and secreted (RANTES or CCL5) all similarly facilitate recruitment of monocytes and T cells to regions of active inflammation [47-49].
In addition to coordinating and enhancing innate and adaptive responses to infection, monocytes and macrophages also play a critical role in the clearance of pathogens through phagocytosis and intracellular killing, processes mediated by IFNγ signaling. IFNγ upregulates the expression of receptors that promote phagocytosis such as the FcγR1, which is important for antibody-dependent, cell mediated cytotoxicity [50]. Complement secretion and complement receptor expression are also induced by IFNγ as another means of promoting phagocytosis [51].
Once phagocytosed, IFNγ also acts to upregulate processes involved in pathogen killing and clearance. For example, in combination with a TLR signal, IFNγ induces the natural resistance- associated macrophage protein (NRAMP1) which has been associated with phagolysosome maturation and effective pathogen clearance, specifically in the context of non-tuberculous mycobacterial infection [52-56]. Polymorphisms in NRAMP1 have been associated with susceptibility to mycobacterial disease, an infection susceptibility also associated with IFNγ deficiency [57, 58].
The primary microbicidal mechanisms of monocytes and macrophages induced by IFNγ involve the production of reactive oxygen and nitrogen species. Inducible nitric oxide synthase
(iNOS/Nos2) is strongly induced by IFNγ in the context of infection and catalyzes the production of microbicidal nitric oxide (NO) in an NADPH-dependent process through the conversion of L- arginine to citrulline, yielding NO as a by-product [59, 60]. Mice lacking iNOS have increased susceptibility to viral and bacterial infections, including mycobacterial infections, as well as altered responses to endotoxin shock [60-62]. Importantly, despite extensive characterization of the role of iNOS and NO in the murine macrophage response to infection, there is still 16 controversy about the role of NO in the human macrophage response to infection, as iNOS is not induced in human macrophages in vitro and in vitro infected human macrophages do not produce detectable NO [63].
The immunologic effects of iNOS have also been studied in combination with the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complex, which generates
- superoxide (O2 ) to facilitate the respiratory burst. Both of these systems generate microbicidal metabolites which have been shown independently to confer protection against infection.
However, in the absence of both iNOS and gp91phox (a critical component of the NADPH oxidase), mice spontaneously develop abscesses containing commensal organisms and have further impairment of their response to infection with pathogens such as Salmonella typhimurium or Listeria monocytogenes, suggesting that one system may compensate for the other in cases where only one is deficient [64]. While the iNOS and NADPH oxidase seem to carry out similar and likely overlapping antimicrobial and immunoregulatory functions, loss of either of these
IFNγ inducible systems results in immune deficiency. While the role of iNOS in the human immune response remains unclear [65], the role of NADPH oxidase is obvious as patients with loss of function mutations in any of the NADPH oxidase complex subunits present with chronic granulomatous disease (CGD), characterized by recurrent infections, chronic granulomatous inflammation and immune dysregulation [66, 67]. IFNγ has been shown to transcriptionally induce critical components of the NADPH oxidase complex including gp91phox and p67phox[68-
71]. The impressive induction of the NADPH oxidase-mediated respiratory burst by IFNγ treatment in vitro prompted trials of IFNγ as a therapeutic to augment NADPH oxidase activity/superoxide production in chronic granulomatous disease [72, 73]. IFNγ did have a 17 protective effect in the treatment of CGD with a 70% reduction in infections, however, it did not augment superoxide production in these patients in vivo [74, 75].
While the precise mechanisms have not been completely elucidated, the antimicrobial effects of
IFNγ have been demonstrated in the context of many different types of infections and disease states. For these reasons and with an understanding of the pleiotropic effects of IFNγ, it has been approved for use as a therapeutic in chronic granulomatous disease and has also been used for the treatment of severe, refractory mycobacterial infection in patients with defects in IFNγ production or signaling who retain some capacity to respond to adequate IFNγ (discussed below)
[76].
Clinical Phenotypes of IFNγ Dysregulation:
Clinically significant mutations in various components of the IFNγ signaling pathway cause disease characterized by infection susceptibility and immune dysregulation and have informed much of the current understanding about the role of IFNγ in the human immune response. Here we review the genetics and molecular and clinical phenotypes of patients with defects in the IFNγ signaling pathway.
IFNγR1: Mutations of the IFNγ receptor complex can occur in the IFNGR1 or IFNGR2 and with dominant or recessive inheritance patterns. Autosomal recessive mutations in IFNGR1 cause complete loss of cellular response to IFNγ stimulation in vitro. Most of these mutations result in no detectable IFNGR1 surface expression. Those mutations that do produce detectable
IFNGR1 protein on the cell surface cannot bind IFNγ and have similarly absent cellular response to IFNγ stimulation, as measured by STAT1 phosphorylation and induction of IL-12p70 production by IFNγ [77]. Complete IFNGR1 deficiency causes a severe clinical phenotype with 18 disseminated, early-onset infections with Bacillus Calmette-Guerin (BCG) vaccine strain or other non-tuberculous mycobacteria, which often become life-threatening [77]. Antibiotics alone do not effectively control these disseminated infections and these patients have no response to exogenous IFNγ treatment. Treatment with exogenous IFNα was trialed in these patients but demonstrated variable effects including exacerbation of existing mycobacterial disease in some cases [78]. Therefore, the only curative treatment for these patients remains hematopoietic stem cell transplantation (HSCT), which is not without its own risks including a high rate of graft rejection, likely associated with high levels of plasma IFNγ [78-81].
Autosomal dominant mutations in IFNGR1 cause partial loss of the cellular response to
IFNγ stimulation in vitro due to heterozygous truncations in the cytoplasmic domain of the protein. The most common dominant IFNGR1 mutations are caused by a small (4-bp) deletion shortly after the transmembrane domain which prevents appropriate endocytosis and recycling of the protein. This results in a dominant negative effect as mutant IFNGR1 proteins accumulate on the surface and interfere with the activity of wild type IFNGR1, resulting in diminished IFNγ signaling [82, 83]. Patients with partial IFNGR1 deficiency experience infections with a similar spectrum of pathogens compared to complete IFNGR1 deficiency, but their disease differs in that their infections are usually responsive to prolonged treatment with both antibiotics and IFNγ and therefore, HSCT is not usually indicated [84].
IFNγR2: Mutations in IFNGR2 are less common than IFNGR1 mutations but include both complete and partial recessive mutations. The first IFNGR2 mutation was identified in a child with disseminated Mycobacterium avium and Mycobacterium fortuitum who was found to have a homozygous dinucleotide deletion causing a frameshift and premature stop codon resulting in a truncated, and non-functional IFNGR2 protein [85]. At least 9 other patients have 19 since been identified with this mutation which results in absent IFNGR2 cell surface expression.
Other loss-of-function (LOF) mutations have been identified in which cell surface expression of
IFNGR2 is maintained, but alterations in glycosylation sites result in a loss of signaling capacity
[86, 87]. Autosomal recessive complete IFNGR2 deficiency, with or without cell surface expression, causes a severe clinical phenotype with disseminated, early childhood nontuberculous mycobacterial or BCG infection with no response to IFNγ and [88] [89]. Partial
IFNGR2 deficiency presents with milder disease and can be caused by autosomal recessive hypomorphic mutations [90] or autosomal dominant haploinsufficiency [91].
IFNγ: Recently, the first documented mutation in IFNγ was identified in 2 related patients presenting with disseminated BCG following vaccination at 3 months of age. Both patients were found to have a small, homozygous, deletion in the IFNG gene, resulting in a frameshift and premature stop codon located 48 amino acids upstream of the canonical stop codon.
Interestingly, this is the first and only disease causing IFNγ mutation that has been identified, compared to many disease-causing mutations in the IFNγ receptor complex [92].
IL-12/IL-12Rβ1: Clinical IFNγ deficiency can also be caused by mutations in IL12B which codes for the p40 subunit of the IL-12p70 cytokine, the primary inducer of IFNγ production from T and NK cells. All identified mutations in IL12B result in a complete loss of detectable IL-12p40, IL-12p70 as well as IL-23 (which shares the p40 subunit with IL-12) proteins [93]. While these patients have a milder clinical phenotype than IFNγR deficient patients, they typically present with disseminated BCG following vaccination and also develop disseminated infection from other non-tuberculous mycobacteria and salmonella [94-96].
This clinical phenotype closely resembles that of patients with mutations in IL12RB1, a subunit of the IL-12 receptor complex. The IL-12 receptor complex is a heterodimer composed 20 of the IL-12Rß1 subunit, which is shared with the IL-23 receptor, and IL-12Rß2 subunit which is unique to the IL-12 receptor. The functional receptor complex is primarily expressed on activated T and NK cells. In patients with IL-12Rß1 deficiency, otherwise normal T and NK cells do not respond to IL-12 and produce minimal IFNγ [93, 97]. Patients with IL-12Rß1 deficiency often present with disseminated BCG (64%) as well as salmonellosis (22%), other
NTM infection (9%) or tuberculosis (4%) [93]. Interestingly, early NTM or BCG infection in these patients is protective against further NTM infection, whereas infections with Salmonella can recur repeatedly [98]. IL-12Rß1 deficiency is the most common cause of inherited mycobacterial infection susceptibility, with at least 70 unique disease-causing variants reported, all of which result in complete loss of signaling through the IL-12 receptor. Clinical disease is only apparent with biallelic mutations and family members who are heterozygous carriers of
IL12RB1 mutations are clinically healthy with normal IL-12 signaling [99].
Acquired Anti-Cytokine Autoantibodies: In addition to primary immunodeficiency syndromes caused by germline mutations in components of the IL-12-dependent IFNγ-mediated signaling pathway, acquired immunodeficiency syndromes caused by neutralizing anti-cytokine autoantibodies have also been identified [100-103]. While genetic associations with human leukocyte antigen (HLA) loci have been identified [104], this acquired autoimmune disease resulting in immunodeficiency does not follow a Mendelian inheritance pattern and the precise etiology is not well understood. The clinical phenotype amongst different patients and even in the same patient over time can vary based on the evolving titer and avidity of the autoantibody
[101]. Many cases of neutralizing anti-IFNγ autoantibodies have been identified to date and these patients’ clinical phenotype resembles that of patients with IFNGR1/2 mutations, except for the adult-onset presentation [100-102]. At least one patient with neutralizing anti-IL-12p70 21 autoantibodies has been identified and her presentation with recurrent Burkholderia gladioli infection was unique compared to patients with genetic defects in IL-12 or IL-12Rß1 [105].
Anti-IL-12 autoantibodies have also been identified in patients with thymoma, but their role in causing infection susceptibility remains unclear. In sum, immunodeficiency caused by neutralizing anti-IFNγ autoantibodies phenocopies disease associated with genetic defects in
IFNγ or its receptor, while the clinical phenotype of patients with neutralizing anti-IL-12 autoantibodies is incompletely understood. Importantly, treatment for the patients with mycobacterial disease secondary to anti-IFNγ autoantibodies includes the use rituximab to address the underlying etiology of the immunodeficiency in addition to prolonged antibiotic therapy [106].
STAT1: Downstream of IL-12-induced IFNγ production and signaling through the IFNγ receptor, STAT1 is the next critical mediator of IFNγ signaling. Mutations in STAT1 have also been associated with infection susceptibility and immune dysregulation, which can be caused by either gain or loss of function of STAT1. Loss of function mutations in STAT1 present with a clinical phenotype similar to that of other mutations in the IL-12-dependent IFNγ-mediated signaling pathway. Autosomal recessive mutations in STAT1 result in a complete loss of function with susceptibility to both weakly virulent mycobacteria and viruses, as both type I and type II interferon signalings are disrupted. Dupuis et al. reported 2 unrelated infants with homozygous mutations in STAT1 that prevented activation of the STAT1-containing transcription factors: IFN-stimulated gene factor 3 (ISGF3) composed of STAT1, STAT2 and
IRF9 (activated by IFNα and IFNß) and GAF/STAT1 homodimers (activated by IFNγ) [107].
Hypomorphic autosomal recessive mutations in STAT1 can also cause a partial loss of STAT1 function with milder bacterial and viral susceptibility phenotype [108]. Dominant negative 22
STAT1 deficiency caused by a heterozygous mutation in STAT1 (p.L706S), also results in a partial loss of function, but primarily affects IFNγ signaling. These patients presented with disseminated BCG following vaccination as well as other severe NTM infections in childhood but did not experience severe viral infections. The mutation, p.L706S, was found to interfere with STAT1 phosphorylation at tyrosine 701, resulting in impaired nuclear accumulation of GAF but not ISGF3, explaining the phenotypic similarities to diseases associated with IFNγ but not type I interferon deficiency [109].
Interestingly, heterozygous, autosomal dominant variants in the coiled-coil domain of
STAT1 can cause a gain-of-function (GOF) phenotype. These mutations have also been reported to have a preferential effect on GAF nuclear localization but little effect on STAT1-containing
ISGF3, effectively resulting in enhanced responsiveness to IFNγ. This enhancement of STAT1 mediated signaling promotes the development of Th1 CD4+ T cells at the expense of Th17 CD4+
T cells resulting in a relative Th17 deficiency which may contribute to the susceptibility to fungal disease and chronic mucocutaneous candidiasis (CMC) observed in these patients. [110].
Autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED) caused by LOF mutations in AIRE also causes CMC through a mechanism that was recently determined to be dependent on excessive IFNγ production, resulting in disruption of the epithelial barrier and increased susceptibility to C. albicans invasion. [111]. These diseases associated with increased
IFNγ signaling illustrate that both diminished and excessive IFNγ activity can be pathologic, suggesting that IFNγ controls cellular processes that must be finely tuned and highly regulated.
Regulation and function of Immunoresponsive gene 1 (IRG1):
Itaconic acid (itaconate) was identified in 2011 by metabolic profiling of lungs from
Mycobacterium tuberculosis (Mtb) infected mice [112] and in supernatants from LPS-stimulated 23
RAW264.7 cells [113]. Strelko et al. determined that itaconate was synthesized through the decarboxylation of the tricarboxylic acid (TCA) cycle intermediate, cis-aconitate. The production of itaconate was then linked to the decarboxylase enzyme, immunoresponsive gene 1 (IRG1) also known as aconitate decarboxylase 1 (ACOD1) or cis-aconitate decarboxylase (CAD)
(Figure 1.1) [114]. IRG1 had initially been identified in 1995 when it was isolated from a cDNA library of LPS-stimulated macrophages [115], but its specific function was not known.
Michelucci et al. demonstrated that knock down of IRG1 results in reduced itaconate production and overexpression of IRG1 in A549 cells increases itaconate expression, suggesting that itaconate levels are completely dependent on IRG1 decarboxylase activity [114]. Similarly, we used a method of simultaneous extraction of nucleic acids and metabolites [116] to assess both
IRG1 mRNA expression and intracellular itaconate levels in primary human macrophages and found a strong correlation between IRG1 mRNA expression levels and intracellular itaconate levels (Figure 2.2A-B).
24
Figure 1.1:
Glycolysis Glucose
Glucose-6-Phosphate
Pyruvate Lactate
Acetyl-CoA
Oxaloacetate Citrate
Malate Cis-aconitate TCA Cycle
IRG1 Itaconate
Fumarate Isocitrate
SDH
Succinate α-Ketoglutarate
Figure 1.1: Metabolic pathways leading to IRG1 mediated itaconate biosynthesis. IRG1 expression and its function to catalyze the decarboxylation of the TCA cycle intermediate, cis- aconitate, to produce itaconate.
Lampropoulou et al. then showed that Irg1 is one of the most highly induced transcripts and itaconate is the most highly induced metabolite in M1 (LPS+IFNγ) stimulated murine bone marrow derived macrophages [117] and Basagoudanavar et al. identified Irg1 as the primary
RelA-dependent interferon stimulated gene (ISG) in IFNβ stimulated mouse embryonic fibroblasts [118]. Degrandi et al. showed that Irg1 is highly inducible in murine bone marrow derived macrophages in response to a variety of TLR ligands as well as in the spleens and lungs of Listeria monocytogenes and Toxoplasma gondii infected mice [119]. The cell type specific expression of IRG1 is almost exclusively restricted to immune cells, specifically myeloid cells.
Hall et al. demonstrated that in zebra fish Irg1 is specifically expressed in macrophage-lineage cells [120]. Consistent with these findings, Nair et al. showed that a LysM-Cre Irg1fl/fl mouse model almost entirely recapitulated the phenotype of a total Irg1 knock out in the context of Mtb 25 infection [121]. Our group has also investigated the relative levels of IRG1 expression in different peripheral blood mononuclear cell (PBMC) subsets from healthy human subjects and found that IRG1 is predominantly expressed in the CD14+ fraction of PBMC’s and itaconate is only detectable in the CD14+ fraction of PBMC’s (Figure 2.1A-B).
The Immunometabolite, itaconate, and its proposed functions:
Antimicrobial Activity of Itaconate: With a better understanding of the regulation and function of IRG1 and its expression restricted almost exclusively to M1 activated macrophages, the immunologically relevant functions of itaconate emerged as a central question. Itaconate was first identified as a competitive inhibitor of bacterial isocitrate lyase (ICL) purified from
Pseudomonas indigofera in 1974 [122, 123] and was shown to inhibit growth of P. indigofera on ethanol or butyrate but not on glucose [124]. Isocitrate lyase (ICL) is a key enzyme of the glyoxylate shunt, which is a metabolic pathway required for mycobacterial (and other intramacrophagic pathogens) persistence within macrophages or in other glucose-deplete environments [125, 126]. Therefore, itaconate-mediated inhibition of ICL could be detrimental to the survival of mycobacteria and other intracellular pathogens.
A role for itaconate in control of intracellular infections was not proposed until 2011, when Strelko et al. identified itaconate as a metabolite that is highly induced in mammalian macrophages upon activation [113]. Naujoks et al. then demonstrated the bactericidal activity of itaconate against a variety of human pathogens including L. pneumophila, S. aureus and A. baumannii in liquid culture (without glucose) [127]. The broad inhibitory activity of itaconate on bacteria with and without ICL suggests alternative mechanisms of action, maybe related to the inhibition of propionyl-CoA, as proposed by Berg et al. [128]. Knock down of Irg1 in murine macrophages results in impaired control of the intracellular pathogens, L. pneumophila or S. 26 enterica, following in vitro infection [114, 127]. Recently, Chen et al. demonstrated that IRG1 interacts with the guanosine triphosphatase (GTPase), Rab32, which is required for effectively restricting replication of intravacuolar pathogens. They propose that in the context of Salmonella infection, Rab32 facilitates the delivery of IRG1, and therefore itaconate, to the Salmonella- containing vacuole thereby restricting its replication through direct antimicrobial activity [129].
Furthermore, itaconate degradation pathways have been identified in certain pathogens, including Salmonella, and are associated with pathogen virulence, suggesting co-evolution of pathogens to subvert effective host defense strategies [130]. Beyond the development of pathways to degrade the potentially microbicidal host metabolite, new evidence suggests that some pathogens have evolved to benefit from host-produced itaconate. Pseudomonas aeruginosa can utilize itaconate as a substrate for energy production and the metabolic reprogramming induced by itaconate in Staphylococcus aureus promotes biofilm formation [131, 132]. These findings suggest that itaconate can have direct antimicrobial activity and supports a role for itaconate as an important effector of cell intrinsic control of intracellular pathogens.
Anti-inflammatory Activity of Itaconate: In contrast, Nair et al. proposed a primarily anti-inflammatory role for itaconate. They assessed susceptibility of Irg1-/- (itaconate deficient) mice to Mtb and found that while Irg1-/- mice were significantly more susceptible to Mtb infection in vivo, there was no difference in in control or killing of Mtb when bone marrow derived macrophages were infected in vitro. Importantly, they also assessed infection of Irg1-/- mice with ICL-deficient Mtb which revealed no difference in susceptibility compared to wild type Mtb, suggesting that ICL inhibition may not be the mechanism by which itaconate is protective against mycobacterial infection. Nair et al. went on to demonstrate that susceptibility of Irg1-/- mice to Mtb is a result of neutrophil mediated immunopathology caused by 27 hyperinflammatory Irg1-/- macrophages [121]. These findings suggest an alternative function of itaconate as an anti-inflammatory metabolite in which itaconate does not contribute to control of infection through direct antimicrobial activity but rather by modulating host the inflammatory response.
One mechanism for the anti-inflammatory effects of itaconate was proposed by
Lampropoulou et al. who demonstrated that bone marrow derived macrophages stimulated with
LPS or LPS+IFNγ accumulated high levels of itaconate upon stimulation, which was associated with succinate accumulation. In contrast, Irg1-/- macrophages under the same stimulation conditions did not accumulate itaconate or succinate. Irg1-/- macrophages also produced higher levels of certain pro-inflammatory cytokines including IL-1ß, IL-6 and IL-12. Interestingly, they noted that TNFα levels were not altered. The proposed mechanism for the hyperinflammatory state of Irg1-/- macrophages was related to the ability of itaconate to act as a competitive inhibitor of succinate dehydrogenase (SDH), leading to the accumulation of succinate. Succinate oxidation has been associated with reverse electron transport in the mitochondria, a process which generates increased levels of mitochondrial reactive oxygen species (mROS), which in turn act to stabilize hypoxia inducible factor α (HIF1α) and induce IL-1ß production [133].
Consistent with this succinate-dependent mechanism for the anti-inflammatory effects of itaconate, Daniels et al. demonstrated a role for itaconate in the control of neuronal infection with Zika Virus. Their findings suggested that itaconate mediated inhibition of SDH resulted in a neuronal metabolic state that suppressed viral replication [134]. Indeed, enzymatic studies clearly demonstrated that itaconate acts as a competitive inhibitor or succinate dehydrogenase.
However, the precise consequences of this inhibition and subsequent accumulation of succinate 28 are not completely understood and the effect may be context specific, for example serving different functions in viral infection compared to bacterial infection.
Itaconate also has the capacity to post-translationally modify proteins by cysteine alkylation [135, 136], which has been linked to its anti-inflammatory effects through two different mechanisms. Mills et al. showed that itaconate directly alkylates Kelch-like ECH- associated protein 1 (KEAP1), the constitutive inhibitor of nuclear factor erythroid 2-like 2
(Nrf2). Itaconate mediated alkylation of KEAP1 targets it for proteosomal degradation, allowing for the activation and nuclear translocation of Nrf2, a transcription factor regulating the expression of many anti-oxidant response genes [137]. This Nrf2 activation was also shown to inhibit IL-1ß and type I interferon production through a mechanism that is not completely understood, but suggests a negative feedback cycle in which interferons augment itaconate production which, in turn, increases Nrf2 activation to limit further interferon production [136].
Liao et al. showed that itaconate also specifically alkylates cysteine 22 on the critical glycolytic enzyme, glyceraldehyde 3-phosphate dehydrogenase (GAPDH). In this case, itaconate-mediated alkylation of GAPDH inhibited glycolysis and therefore, reduced glycolysis-induced expression of genes such as IL-1ß, iNOS and TNFα [135].
Endogenous Itaconate vs. Exogneous Itaconate Derivatives: It is important to note that many studies of itaconate do not directly assess the activity of endogenous itaconate, and rather employ commercially available itaconate derivatives such as 4-octyl-itaconate (4-O-I) or dimethyl itaconate (DI). Conclusions from studies using exogenous itaconate derivatives have generated some controversy in the field as they are often implied to act by the same mechanism as endogenous itaconate. These differences are significant in that the mechanisms for transport or permeability of itaconate and its derivatives across cellular membranes are not well understood. 29
Furthermore, one study demonstrated that exogenous DI was capable of inducing an accumulation of intracellular succinate but was not metabolized into itaconate intracellularly
[138]. These findings suggest that exogenous forms of itaconate could act through alternative mechanisms such as signaling through yet unidentified cell surface receptors to affect the intracellular phenotypes observed. Additionally, itaconate is presumably produced within the inner matrix of the mitochondria, as its substrate is a TCA cycle intermediate and IRG1 has been shown to associate with mitochondria [113, 127]. The subcellular localization, compartmentalization/concentration and transport of itaconate are not well characterized, but likely differ significantly when produced endogenously by mitochondrial IRG1 versus when given exogenously by adding itaconate derivatives to the culture media or delivering it intravenously or intraperitoneally in vivo [117, 136, 139-143].
Itaconate and Cellular Electrophilic and Oxidative Stress: An alternative mechanism has also been proposed for itaconate-induced activation of Nrf2. Itaconate can form an adduct with glutathione (GSH), depleting a primary cellular antioxidant and inducing a cellular electrophilic stress response that activates both Nrf2-dependent antioxidant transcriptional programs as well as Nrf2-independent programs through ATF3 and IκBζ [141]. These findings link the anti- inflammatory activity of itaconate to cellular electrophilic and oxidative stress. There is also in vivo evidence to suggest that itaconate is an important mediator of protective anti-oxidant activities. Cordes et al. demonstrated that itaconate has protective effects in a murine model of cerebral ischemia/reperfusion injury which drives oxidative stress and ultimately, damaging inflammation. They hypothesized that the protection conferred by itaconate was mediated through its inhibitory effects on SDH which modulates the return of mitochondrial function (and mROS production) upon reperfusion [144]. Itaconate-mediated protection against sterile 30 inflammation caused by ischemia/reperfusion injury was also demonstrated in a model of liver injury in which Irg1 expression in both the hematopoietic and non-hematopoietic compartments were shown to contribute to the protective effects [145].
As an abundant intracellular metabolite in activated macrophages, the functional effects of itaconate have been associated with a variety its characteristics. Its structural similarity to other metabolites makes it a potential competitive inhibitor for many enzymes, including bacterial ICL and mammalian SDH. Its ability to post-translationally modify cellular proteins by alkylation has been demonstrated to affect at least Nrf2 and GAPDH. Finally, the electrophilic properties of itaconate give it the potential to react with any sulfhydryl containing proteins and likely affect many cellular processes through this process. However, like IFNγ, it is difficult to determine exactly which of the many activities of itaconate are important in any specific biological context.
Macrophage metabolic reprogramming and the relationship between metabolism and immunity:
The strong association of IRG1 and itaconate with “M1 polarized” macrophages raises questions about the ever evolving paradigm for characterizing the polarization states of macrophages. Mantovani et al. reviewed the spectrum of macrophage polarization with the understanding that these phenotypes are plastic and represent benchmarks on a continuum of immunologic, metabolic and cell physiologic activities [146]. Polarized macrophages are divided into 2 primary subsets: M1, classically activated macrophages with a generally pro-inflammatory phenotype and M2, alternatively activated macrophages with a generally anti-inflammatory phenotype. M1 macrophages express high levels of pro-inflammatory cytokines and chemokines such as IL-12 and CXCL9/10 and very high levels of IRG1. M2 macrophages can then be further categorized based on their specific function as M2a, M2b or M2c. M2a macrophages are responsible for coordinating the Th2 response, mediating allergic inflammation and killing of 31 parasites. M2b (type II) macrophages are immunoregulatory, expressing high levels of IL-10 and low levels of IL12. Finally, M2c (deactivated) macrophages are similarly immunoregulatory but additionally play an important role in tissue remodeling and repair, secreting high levels of matrix proteins [147].
Importantly, the metabolic program associated with each macrophage polarization state is distinct. Briefly, M1 macrophage activation by TLR ligands (with or without IFNγ) induces aerobic glycolysis and causes 2 break points in the Krebs cycle, whereas M2 macrophage polarization promotes fatty acid oxidation and oxidative phosphorylation [148-150]. While the expression of the metabolic enzyme, IRG1, has long been used as a marker of M1 macrophage activation, its role in promoting and maintaining this phenotype is still unclear. Expression of
Irg1 was found to be increased in a knockout model of peroxisome proliferator-activated receptor-γ (PPARγ), an IL-4-induced, regulator of fatty acid metabolism. These findings suggest that signaling pathways promoting M2 polarization, such as IL-4 signaling through PPARγ, also actively repress Irg1 expression [151]. Similarly, Ganta et al. found that the expression of the microRNA, mir93, which interacts with IRF9 and decreases Irg1 expression, also enhanced M2 macrophage polarization in an ischemia-reperfusion model [152]. Itaconate has also been proposed to mediate the metabolic reprogramming in M1 macrophages through its inhibition of
SDH, which disrupts its role in the TCA cycle and as complex II of the mitochondrial electron transport chain. This inhibition results in increased mROS production and may force these macrophages to rely more heavily on glycolysis for energy production [136, 153]. The most recent assessment of the role of itaconate in macrophage polarization revealed that itaconate can react with coenzyme A (CoA) to form itaconyl-CoA which broadly affects the mitochondrial
CoA pool and can inhibit enzymes involved in branch chain amino acid and fatty acid 32 metabolism, again suggesting a fundamental but pleiotropic role for itaconate in cellular metabolism and immunologic activity [154].
Beyond the IRG1-itaconate pathway, the regulation of macrophage cellular metabolism by immunologic signaling pathways and the immunologic effects of metabolic processes has been the focus of much investigation in recent years [155]. The defining metabolic feature of M1 activated macrophages is the Warburg effect, defined by a metabolic switch to glycolysis despite sufficient availability of oxygen, also known as aerobic glycolysis [156]. This switch to aerobic glycolysis is induced by macrophage pattern recognition receptors (PRR), activated by pathogens or purified TLR ligands. A proposed mechanism for LPS induction of glycolysis suggests that
LPS signaling through toll like receptor 4 (TLR4) activates HIF1α, independent of hypoxia
(possibly in response to cellular ROS production) to induce expression of genes that increase glucose uptake and glycolytic flux [157-160]. Increased flux through glycolysis is required for complete induction of LPS-inducible gene expression and inhibition of glycolysis by 2- deoxyglucose (2-DG) results in reduced LPS-induced IL-1ß production [133]. This process of infection-induced HIF1α stabilization and increased glycolytic flux has been shown to play an important role in the immunologic response to infection. Macrophages lacking HIF1α demonstrate impaired bacterial killing capacity and have impaired IFNγ induced immunity in vitro and in vivo to Mtb [161-163].
A few explanations have been proposed for the paradoxical switch to glycolysis, which produces less ATP per glucose, despite available oxygen to fuel the much more efficient process of oxidative phosphorylation. First, it has been proposed that in glucose replete environments, glycolysis can actually generate more ATP/time than oxidative phosphorylation, despite consuming significantly more glucose to do so [164]. Second, flux of carbon through glycolysis 33 generates the precursors required for flux through the pentose phosphate pathway (PPP). Flux through the PPP is required for the reduction of NADP+ to NADPH which can be oxidized to produce microbicidal ROS during the respiratory burst [165]. The PPP is also required for purine and pyrimidine biosynthesis, required for increased transcriptional activity associated with activation. Production of adenine, in particular, is induced and can act to augment LPS induced signals and globally regulate cellular metabolism through its effects on AMP-activated protein kinase (AMPK) [156, 166-168]. Finally, activation induced disruptions in the TCA cycle and mitochondrial electron transport chain may leave cells dependent on glycolysis for ATP production [133]. As described earlier, induction of IRG1 results in diversion of cis-aconitate into itaconate rather than isocitrate (the next step in the TCA cycle), causing a break in the TCA cycle that is further enhanced by the transcriptional downregulation of isocitrate dehydrogenase
(IDH) [169]. Flux through the TCA can proceed through an anapleurotic pathway by which glutamine supplies α-ketoglutarate downstream of the break point at IDH/isocitrate. However, itaconate mediated inhibition of SDH causes another break that inhibits the conversion of succinate to fumarate [169]. In addition to a disrupted TCA cycle, LPS/IFNγ-induced nitric oxide has also been shown to disrupt the electron transport chain by nitrosylating iron-sulfur proteins in complexes I and IV resulting in increased ROS production but reduced ATP production [136,
170, 171]. Again, there is evidence to suggest that the metabolic adaptations observed in murine macrophages may not translate to human macrophages. For example, the NO mediated effects that inhibit mitochondrial respiration under the same conditions that augment glycolysis are not observed in human monocytes, which augment both glycolysis and mitochondrial respiration upon stimulation with M1 polarizing stimuli or infection [172]. 34
While our understanding of the metabolic underpinnings of macrophage activation have advanced dramatically, many new questions have been raised about the specific responses to unique stimuli or the combination of multiple stimuli and how the kinetics of these responses evolve throughout the immune response [172, 173]. We focused our investigation on the metabolic reprogramming that occurs in response to IFNγ to understand how cellular metabolic processes adapt to support the significant changes in macrophage activity induced by IFNγ alone and in the context of infection. While IFNγ is known to broadly affect cellular metabolism through transcriptional regulation and induce the production of microbicidal ROS/RNS, the precise mechanism by which IFNγ acts and the other metabolic pathways that are affected are not well understood. Here, we first investigated the complex roles of itaconate as a metabolic effector of the IFNγ-mediated immune response. In Chapter II we assessed the mechanisms by which IRG1 and itaconate are regulated in the normal human immune response and how they are dysregulated in the context of primary immunodeficiency syndromes caused by genetic or acquired defects in IFNγ signaling. We then characterized the relative contribution of itaconate to the IFNγ-induced immune response to non-tuberculous mycobacteria (NTM). In the process of assessing the metabolic phenotype of IFNγ-induced itaconate in the context of infection, we also observed a unique metabolic phenotype of monocytes treated with IFNγ alone. In Chapter III we characterized the metabolic phenotype of IFNγ stimulated primary human monocytes and demonstrated how it differs from monocytes stimulated with IFNγ in the presence of a TLR signal. We found that IFNγ induces high levels of oxygen consumption in monocytes and identified a mechanism by which IFNγ signaling reprograms monocyte metabolism to promote a highly oxidative phenotype, critical for their immunologic activity. Figure 1.2 provides an 35 overview of these distinct, IFNγ regulated, metabolic pathways and their convergent effects on cellular processes involved in microbicidal metabolite production and protective anti-oxidant and anti-inflammatory activities.
36
Figure 1.2:
LPS LPS+IFN! IFN!
NAD Salvage
NAM NR
Glycolysis NAMPT NRK2 Glucose NMN
Pentose Phosphate Pathway Glucose-6-phosphate NADK/NNT NADH NAD+ Pool NADP(H) NADPH Pyruvate Glutathione/ NADPH NADH Thioredoxin Oxidase Nucleotide Biosynthesis/ Glutathione Depletion Reductases Transcriptional Response Acetyl-CoA Itaconate-GSH adduct + Oxaloacetate Citrate KEAP1 alkylation Nrf2 Activation Antioxidant activity NADP +Superoxide Citrulline NADH IRG1 Itaconate Inhibition of TCA Cycle ATF3/I"B# and Anti-inflammatory activity Ornithine Malate Isocitrate NLRP3 NO iNOS Inhibition ICL or Arginosuccinate Fumarate IDH NADH propionyl-CoA Antibacterial activity Arginine carboxylase SDH !-KG mitochondrial Urea Cycle Succinate NADH +mROS SOD/aconitase
OxPhos and Outer Mitochondrial HIF1α mitochondrial Electron Transport Membrane stabilization RET/mROS 2H+ 4H+ 2H+ 3H+ Intermembrane Space
CoQ Cyt C Complex I Complex III Complex IV Inner Mitochondrial Membrane Complex II Complex V Mitochondrial Matrix
NAD+ H2O NADH ½ O2
3H+ ATP + H2O ADP + Pi Figure 1.2: IFNγ regulated metabolic pathways and their immunologic effects. Schematic representation of IFNγ-mediated regulation of central carbon metabolism and NAD-dependent redox pathways important for IFNγ-induced monocyte activation.
37
CHAPTER II: IRG1 and Itaconate contribute to the Interferon Gamma-Induced Host Defense Against Mycobacterium avium
Katelyn J. McCann1,2, Logan Fisher3, Eduardo Amaral3, Kate Aberman3, Anuj Kayshap1,
Benjamin Schwarz4, Catharine Bosio4, Ian Moore5, Kelly M. Stewart6, Hyeryun Kim6, Alan
Sher3, Beth A. McCormick2, Steven M. Holland1
1 Immunopathogenesis Section, Laboratory of Clinical Immunology and Microbiology, National
Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892
2 Department of Microbiology and Physiological Systems, University of Massachusetts Medical
School, Worcester, MA, USA
3 Immunobiology Section, Laboratory of Parasitic Diseases, National Institute of Allergy and
Infectious Diseases, National Institutes of Health, Bethesda, MD 20892
4 Laboratory of Bacteriology, National Institute of Allergy and Infectious Diseases, National
Institutes of Health, Hamilton, MT 59840
5 Infectious Disease Pathogenesis Section (IDPS), Comparative Medicine Branch, National
Institute of Allergy and Infectious Diseases, National Institutes of Health, US Department of
Health and Human Services, Rockville, Maryland 20852, United States
6 Agios Pharmaceuticals, 88 Sidney Street, Cambridge, MA 02139, USA
Parts of this chapter are part of a manuscript submitted for review: Journal of Immunology manuscript identification number 21-00563-FL Funding for this study was provided by the Division of Intramural Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health.
38
Introduction:
IFNγ is an essential component of the host defense against mycobacterial infection.
However, the precise mechanism(s) by which IFNγ promotes control and clearance of mycobacteria remain unknown. Much of the current understanding of the pleiotropic roles of
IFNγ in humans has come from studying patients with either primary or acquired immunodeficiencies affecting the IL-12 dependent, IFNγ mediated signaling pathway. These patients share a common susceptibility to disseminated infection with environmentally ubiquitous, nontuberculous mycobacteria (NTM), which are typically avirulent in immune competent hosts [20, 85, 100-102, 174-176]. Consistent with the clinical manifestations in patients, IFNγ knock out (GKO) mice demonstrate impaired control of NTM infection. Although
GKO mice show no difference in mortality compared to wild type mice, they survive with persistently high mycobacterial loads in multiple organs [177-181].
In contrast to Mycobacterium tuberculosis (Mtb), which can infect immune competent and immunodeficient hosts, NTM are typically avirulent in immune competent hosts [182].
Therefore, restoring normal host defense in immunodeficient patients could rescue their ability to control NTM infections and their associated complications. However, the current paradigm for treatment of mycobacterial infections has focused on the development of antimicrobial therapies that often have dose-limiting toxicities and become ineffective with the emergence of resistant organisms [183-187]. For these reasons, targeted, host-directed therapies require a detailed understanding of the IFNγ-dependent mechanisms of host defense against mycobacterial infection.
Recent studies in macrophage immunometabolism suggest that the Immunoresponsive Gene
1 (IRG1)-itaconate pathway is part of an essential effector pathway of the innate immune 39 response to mycobacteria. IRG1 is well known as a transcriptionally induced marker of classically activated (M1) macrophages [113, 114]. However, its function was only recently elucidated as the enzyme that catalyzes the decarboxylation of the Kreb’s cycle intermediate, cis- aconitate, to produce itaconate [113]. Lampropoulou et al. identified Irg1 as one of the most highly induced transcripts in M1 (LPS+IFNγ stimulated) murine bone marrow derived macrophages, and its product, itaconate, as the most highly induced metabolite. In addition to its significance as the most abundant metabolite in M1 macrophages [117], itaconate has been reported to have direct antimicrobial activity against a variety of human pathogens [123, 125,
127] as well as anti-inflammatory and antioxidant activities [136, 141, 188, 189].
Nair et al. demonstrated that Irg1-/- mice, which cannot produce itaconate, display mortality kinetics similar to that of GKO mice in response to Mtb infection [121], suggesting that IRG1 is an essential effector of the IFNγ-mediated host defense programs induced in response to Mtb infection. Here we show that the IRG1-itaconate pathway is dysregulated in the context of several immunodeficiency syndromes caused by defects in the IFNγ pathway, likely contributing to their susceptibility to mycobacterial infections (Figure 2A). However, because IFNγ is known to regulate many cellular processes that contribute to host defense against NTM infection [190,
191] we sought to elucidate the specific contributions of IRG1/itaconate to the infection susceptibility and immune dysregulation observed in patients with broader IFNγ signaling defects. To do so we used the Irg1-/- mouse model previously described as highly susceptible to
Mtb infection [121] and investigated whether IRG1 and itaconate play a similar role in the context of infection with a more ubiquitous but less virulent mycobacterial infection,
Mycobacterium avium.
40
Results:
IRG1 Expression and Itaconate Production are synergistically enhanced by IFNγ in the Context of Toll-Like Receptor (TLR) stimulation
We began by confirming that IRG1 expression and itaconate production are almost
exclusively detected in the CD14+ fraction of human peripheral blood mononuclear cells
(PBMCs) (Figure 2.1A-B).
Figure 2.1:
IRG1 18h ItaconateIRG1 18h 18h Itaconate 18h A IRG1 mRNAIRG1IRG1 Expression 18h18h B IntracellularItaconateItaconate Itaconate 18h18h 150000 1500002500 2500 Unstimulated 150000150000 Unstimulated 25002500 Unstimulated 100000 100000 Unstimulated Unstimulated +LPS +LPS Unstimulated 1000001000002000 +LPS+LPS 2000 +LPS+ LPS 20002000 +LPS+ LPS 50000 +IFNg 50000 +IFNg +LPS 5000050000 +IFNg+ IFN+IFNg+IFNgg +IFNg+ IFNg +LPS+IFNg 1500 +LPS+IFNg 1500 +IFNg 5000 5000 +LPS+IFNg+LPS+IFNgg 15001500 g 50005000 +LPS+IFNg+ LPS + IFN +LPS+IFNg+ LPS + IFN
RQ IRG1 4000 RQ IRG1 4000600 600 +LPS+IFNg
RQ IRG1 4000 RQ IRG1 4000 600600 3000 3000400 400 2000 300030002000 400400 20002000200 200 1000 1000 200200 0 100010000 0
Intracellular Itaconic Acid (ng/mL) 00 Intracellular Itaconic Acid (ng/mL) 00 Intracellular Itaconic Acid (ng/mL) Intracellular Itaconic Acid (ng/mL)
CD14 (-) CD14 (+) CD14 (-) CD14 (+) CD14 (-) CD14 (+) CD14CD14 (-)(-) CD14 (-) Total PBMC Total PBMC IRG1 ExpressionCD14CD14 (+)(+) Total PBMC CD14 (-) CD14 (+)Intracellular Itaconate IRG1 Expression TotalTotal PBMCPBMC IntracellularTotalTotal PBMCPBMC Itaconate 5000 2000000 5000 2000000Healthy Control Healthy Control Healthy Control Healthy Control 4000 Monocytes Monocytes 4000 C IRG1 ExpressionMonocytes 1500000 Monocytes 1500000IFNGR1-/- Patient IFNGR1-/- Patient IRG1 6h 1.5 ItaconateIRG1IFNGR1-/- 6h 6h Patient Itaconate 6h IFNGR1-/- Patient IRG1 Expression 3000 Intracellular ItaconateHealthy ControlsMonocytes Monocytes IRG1 Expression 3000 Intracellular Itaconate IRG1IRG1 6h6hMonocytes ItaconateItaconate 6h6h Monocytes ** -/- 1000000 200000 5000 200000500 2000000 1000000IL-12Rß1 Patients 500 5000 2000000 Unstimulated Healthy2000200000200000 Control ** Unstimulated-/- 500500Healthy Control 150000 Healthy Control 2000 RQ IRG1 150000450 Healthy Control UnstimulatedIFNGR1 Patients 450
RQ IRG1 Monocytes1.0 Monocytes Monocytes4000 +LPS 150000150000 Monocytes Unstimulated+LPS 450450 Unstimulated 4000 100000 100000400 1500000 +LPS+LPS 400 500000 Unstimulated 1500000 IFNGR1-/-1000100000100000 Patient +LPS 500000 400400IFNGR1-/- Patient +LPS
IFNGR1-/- Patient 1000 +IFNg IFNGR1-/- Patient +IFNg Itaconate Peak Area 50000 3000 Monocytes50000350 350 Monocytes +LPS 3000 Monocytes 5000050000 Monocytes +IFNg+IFNgItaconate Peak Area 350350 +LPS+IFNg RQ IRG1 300 1000000 +IFNg+LPS+IFNg 300 +IFNg 1000000 0.50 0 +IFNg 3000 2000 0 1503000 +LPS+IFNg+LPS+IFNg0 300300150 +IFNg 2000 RQ IRG1 30003000 +LPS+IFNg 150150 +LPS+IFNg RQ IRG1 RQ IRG1 RQ IRG1 US US +LPS+IFNg
2000 RQ IRG1 1002000 100 US RQ IRG1 500000LPS US LPS 1000 500000 LPS 20002000 100100 LPS
1000 Itaconate Peak Area
1000 Itaconate Peak Area 100050 50 0.0 LPS+IFNg LPS+IFNg 0 LPS+IFNg10001000 0 5050 LPS+IFNg 0 0 0 0 0 0
US Intracellular Itaconic Acid (ng/mL) 0 US Intracellular Itaconic Acid (ng/mL) 0 US LPS US 0 LPS 0 Intracellular Itaconic Acid (ng/mL)
LPS LPS Intracellular Itaconic Acid (ng/mL) -/- Patients -/- Patients LPS+IFNg LPS+IFNg LPS+IFNg CD14 (-) CD14 (+) LPS+IFNg CD14CD14 (-) (-) CD14CD14 (+) (+) CD14 (-) CD14 (+) CD14CD14 (-)(-) CD14CD14 (+)(+) CD14CD14 (-)(-) CD14 (+) Healthy Controls IFNGR1 IL-12Rß1
IRG1 ExpressionD IRG1 Expression E Intracellular Itaconate Intracellular Itaconate IRG1 Expression IRG1 ExpressionIntracellular Itaconate 5000 Intracellular Itaconate 5000000 5000 5000000Healthy Control Healthy Control 5000 5000 5000000 5000000Healthy Control Healthy Control Healthy Control Healthy Control Healthy Control MonocytesHealthy Control Monocytes 4000 Monocytes 4000000 Monocytes Monocytes4000 4000 Monocytes Monocytes4000000 4000000 Monocytes 4000 4000000 IFNGR1-/- Patient IFNGR1-/- Patient IFNGR1-/- Patient IFNGR1-/- Patient IFNGR1-/- Patient IFNGR1-/- Patient IFNGR1-/- Patient 3000 IFNGR1-/- Patient Monocytes 3000000 Monocytes 3000 Monocytes3000 30000003000 Monocytes Monocytes3000000 Monocytes 3000000 Monocytes Monocytes
2000 2000 2000000 2000 2000000 2000000 RQ IRG1 RQ IRG1 2000 RQ IRG1 2000000 RQ IRG1 1000 1000 1000000 1000000 Itaconate Peak Area Itaconate Peak Area 1000 1000000
1000 1000000 Itaconate Peak Area
0 0 0 0 Itaconate Peak Area γ γ γ 0 γ 0 US US US US LPS LPS0 LPS LPS 0 γ γ γ US LPS+IFN LPS γ US LPS+IFN US LPS+IFN LPS+IFN US LPS LPS LPS LPS+IFN LPS+IFN LPS+IFN LPS+IFN 41
Figure 2.1: Irg1 mRNA Expression and Intracellular Itaconate Levels in Healthy Controls and Patients. (A-B) Total peripheral blood mononuclear cells (PBMC) were isolated and stimulated with media alone (unstimulated), LPS (200ng/mL), IFNγ (1000U/mL), or the combination of LPS and IFNγ for 24 hours. The same PBMC’s were also separated into CD14+ and CD14- fractions by positive selection for CD14+ monocytes with magnetic beads and stimulated under the same conditions. RNA and intracellular metabolites were isolated by simultaneous extraction of polar metabolites and nucleic acids from the same cells. (A) isolated RNA was used to quantify Irg1 expression by qPCR. (B) Intracellular itaconate levels were quantified by LC-MS based on a standard curve and normalized to packed cell volume. (C) Total PBMC’s from 4 healthy donors, 2 IL-12Rß1 deficient patients and 3 IFNGR1 deficient patients and stimulated with M. avium (MOI = 10), IFNγ (1000U/mL), and the combination of M. avium and IFNγ for 6 hours. (D-E) CD14+ monocytes from a healthy donor and IFNGR1 deficient patient stimulated with LPS (200ng/mL), or the combination of LPS and IFNγ (1000U/mL) for 24 hours. RNA and intracellular metabolites were isolated by simultaneous extraction of polar metabolites and nucleic acids from the same cells. Irg1 expression measured by qPCR and reported as relative quantity (RQ) normalized to unstimulated samples using housekeeping gene ß-actin. Data in (A-B and D-E) were collected from only 1 healthy donor, therefore statistical analysis could not be performed. Data were analyzed in (C) by one-way ANOVA with Tukey’s multiple comparisons test Error bars indicate mean ± s.e.m. ** p < 0.01.
When we assessed purified CD14+ PBMC-derived monocytes we found that IFNγ alone induced minimal IRG1 mRNA expression and intracellular itaconate production. LPS alone also induced low levels of IRG1 expression and intracellular itaconate. In contrast, the combination of
LPS and IFNγ induced 3-fold more IRG1 expression and intracellular itaconate than LPS alone
(Figure 2.2B). Since IRG1 mRNA expression levels correlated with intracellular itaconate levels as measured by LC-MS (Figure 2.2A and B) we subsequently used IRG1 mRNA expression as a surrogate for intracellular itaconate levels.
Similar to the responses observed with LPS and IFNγ, M. avium induced low levels of IRG1 mRNA expression in primary human monocytes while the combination of M. avium and IFNγ induced significantly more IRG1 mRNA expression (Figure 2.2C and D). This synergistic effect was also observed in murine bone marrow derived macrophages (Figure 2.2E). These data suggest that induction of IRG1 expression and itaconate production is dramatically increased when both a TLR and a type II interferon signal are triggered. 42
Figure 2.2:
Unstimulated IFNgA LPS LPS+IFNg Unstimulated +IFNγ +LPS +LPS+IFNγ Donor: 10000 10000 UnstimulatedUnstimulatedUnstimulatedUnstimulated 10000 IFNg IFNg IFNg IFNg LPS LPS 10000LPS LPS LPS+IFNgLPS+IFNgLPS+IFNgLPS+IFNg 3L 3L 3L 3L1 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 3L 3L 3L 3L 3L 3L 3L 3L 8000 3M 8000 3M 8000 3L 3L 3L 3M3L 8000 3L3M2 3L 3L 3L 8000 8000 8000 8000 3M 3M8000 3M8000 3M8000 8000 3M 3M8000 3M8000 3M8000 8000 3M 3M8000 3M8000 3M8000 8000 3M 3M 3M 3M 3N 3N3N 3N 3N 3N 3N 3N 3N 3N3N 3N 3N 3N 3N 3N3N3 3N 3N 3N 6000 6000 6000 6000 6000 6000 6000 60006000 6000 6000 6000 6000 6000 6000 6000 6000 6000 6000 6000 3O 3O 3O 3O 3O 3O 3O 3O 3O3O 3O 3O 3O 3O 3O 3O 3O 3O 4000 4000 4000 4000 3O 4000 4000 4000 4000 4000 4000 4000 4000 3O4 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 3Q 3Q 3Q 3Q RQ IRG1 3Q 3Q4000 3Q4000 3Q4000 4000 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 3Q 3Q 3Q 3Q RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 3Q 3Q 3Q 3Q 4000 4000 4000 RQ IRG1 3Q 3R 3R RQ IRG1 3R 3R 3Q 4000 3R 3R 3R 3R RQ IRG1 2000 2000 2000 2000 3Q 3R 3R2000 3R2000 3R2000 2000 2000 2000 2000 2000 RQ IRG1 2000 2000 2000 2000 3R3Q5 3R 3R 3R IRG1mRNA expression 3R 0 0 0 0 3R 0 0 3R 0 0 2000 0 2000 0 0 0 0 0 0 0 3R 2000 1h 3h 6h1h 12h3h 18h6h1h 24h12h3h 48h18h6h1h 24h12h3h 48h18h6h 24h12h 48h18h 24h 48h 1h 3h 6h1h 12h3h 18h6h1h 24h12h3h 48h18h6h1h200024h12h3h 48h18h6h 24h12h 48h18h 24h 48h 6 1h 3h 6h1h 12h3h 18h6h1h 24h12h3h 48h18h6h1h 24h12h3h 48h18h6h 24h12h 48h18h 24h 48h 1h 3h 6h1h 12h3h 18h6h1h 24h12h3h 48h18h6h1h 24h12h3h 48h18h6h 24h12h 48h18h 24h 48h Time (hours)Time (hours)Time (hours)Time (hours) Time (hours)Time (hours)Time (hours)Time (hours) Time (hours)Time (hours)Time (hours)Time (hours) 0 0 Time (hours)Time (hours)Time (hours)Time (hours) 0 0 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h Time (hours) Time (hours) Time (hours) Time (hours)
Unstimulated IFNg LPS LPS+IFNg B Donor: 10000 10000 Unstimulated 10000 +IFNγ +LPS 10000 +LPS+IFNγ 3L UnstimulatedUnstimulatedUnstimulatedUnstimulated3L IFNg IFNg IFNg IFNg 3LLPS LPS LPS LPS LPS+IFNg LPS+IFNg LPS+IFNg LPS+IFNg3L1 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 8000 3M 8000 3L 3M 3L 80003L 3L 3L 3L 3L 3M3L 8000 3L 3L 3L 3L 3L 3M2 3L 3L 3L 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3N 3N 3N 3N 6000 6000 100 100 100 100 3N 3N100 60003N 100 3N 100 100 3N 3N100 3N 100 3N 100 100 6000 3N 3N100 3N100 3N 100 100 3N 3 3N 3N 3N 3O 3O 3O 3O 3O 3O 3O 3O 3O 3O3O 3O 3O 3O 3O 3O 3O4 3O 3O 3O 4000 4000 50 50 50 50 3Q 3Q 50 40003Q 50 3Q 50 50 3Q 3Q 50 3Q 50 3Q 50 50 3Q 3Q 50 3Q 50 3Q 50 50 3Q 3Q 3Q 3Q RQ IRG1 3Q RQ IRG1 3Q 4000
RQ IRG1 3Q 3R 3R 3R 3R 3R 3R 3R 3R RQ IRG1 3R 3R 3R 3R 3R 3Q5 3R 3R 3R 3R Itaconate 3R 2000 2000 Intracellular 3R 2000 3R Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 0 0 0 0 2000 0 0 0 0 6 1h 3h 6h1h 12h3h 18h6h1h24h12h3h48h18h6h1h24h12h3h48h18h6h 24h12h 48h18h 24h 1h48h 3h 6h1h 12h3h 18h6h1h24h12h3h48h18h6h1h24h12h3h48h18h6h 24h12h 48h18h 24h 1h48h 3h 6h1h 12h3h 18h6h1h24h12h3h48h18h6h1h24h12h3h48h18h6h 24h12h 48h18h 24h 1h48h 3h 6h1h 12h3h 18h6h1h24h12h3h48h18h6h1h24h12h3h48h18h6h 24h12h 48h18h 24h 48h 0 Time (hours)Time (hours)Time (hours)Time (hours) 0 Time (hours)Time (hours)Time (hours)Time (hours) Time (hours)Time (hours)Time (hours)Time (hours) Time (hours)Time (hours)Time (hours)Time (hours) 0 0 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h Time (hours) Time (hours) Time (hours) Time (hours) C LPS +/- IFNγ D M. avium +/- IFNγ ✱✱ ✱✱ 4000 ✱✱ 300 ✱✱ ✱✱
3000 200
2000 RQ IRG1 RQ IRG1 100 1000
0 0
LPS - IFNg- LPS+ + M. avium - IFNg- LPS+ +
LPS+IFNg LPS+IFNg IFNγUnstimulated- + - + IFNγ Unstimulated- + - +
E IRG1 Expression ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ 50 ✱✱✱✱ 50 WT 40 WTIRG1 KO
40 30 IRG1 KO
20
30 RQ IRG1
10 20 RQ IRG1 0 10 M. avium - - + +
MAC+IFNg 0 IFNγ Unstimulated- + - + MAC (MOI = 25)IFNg (200U/mL)
Figure 2.2: Irg1 mRNA Expression and Intracellular Itaconate Levels are Synergistically MAC+IFNg Unstimulated InducedMAC (MOI =by 25)IFNg the (200U/mL) Combination of LPS and IFNγ. (A-B) Elutriated primary human monocytes from 6 healthy donors were stimulated with media alone (unstimulated), LPS (200ng/mL), IFNγ (1000U/mL), or the combination of LPS and IFNγ for time points indicated on X axis. RNA and 43 intracellular metabolites were isolated by simultaneous extraction of polar metabolites and nucleic acids from the same cells. (A) isolated RNA was used to quantify Irg1 expression by qPCR. (B) Intracellular itaconate levels were quantified by LC-MS based on a standard curve and normalized to packed cell volume. (C-D) CD14+ monocytes were isolated from peripheral blood mononuclear cells from 3 healthy donors and stimulated with (C) LPS (200ng/mL), IFNγ (1000U/mL), and the combination of LPS and IFNγ or (D) M. avium (MOI = 10), IFNγ (1000U/mL), and the combination of M. avium and IFNγ for 3 hours. (E) Irg1 mRNA expression in murine bone marrow derived macrophages from wild type and Irg1-/- C57B/6N mice stimulated with M. avium (MOI = 25), IFNγ (200U/mL), or the combination of M. avium and IFNγ for 24 hours. (C-E) Irg1 expression was measured by qPCR. Gene expression measured by qPCR and reported as relative quantity (RQ) normalized to unstimulated samples using housekeeping gene ß-actin. Data were analyzed in (C) and (D) by one-way ANOVA with Tukey’s multiple comparisons test, and in (E) by two-way ANOVA with Sidak’s multiple comparisons test. Error bars indicate mean ± s.e.m. ** p < 0.01. **** p < 0.0001.
Given the critical nature of intact TLR and interferon signaling, we hypothesized that patients with immunodeficiency syndromes caused by defects in IL-12-dependent IFNγ signaling might have diminished infection-induced IRG1 expression (Figure 2.3A). Since these patients share a common susceptibility to NTM infection, we assessed IRG1 expression in patient PBMCs stimulated in vitro with the most common NTM pathogen, M. avium (Figure 2.3B). As anticipated, patients with primary immunodeficiency syndromes caused by mutations in IFNγ
Receptor 1 (IFNGR1) had significantly lower IRG1 expression than healthy controls following stimulation with M. avium (2.5-fold; Figure 2.3B). Similarly, two patients with biallelic mutations in the IL-12 Receptor ß1 (IL12RB1) had 3-fold less IRG1 expression than healthy controls in response to M. avium (Figure 2.1C).
We also assessed patients with acquired immunodeficiency syndromes caused by neutralizing anticytokine autoantibodies targeting IFNγ, resulting in susceptibility to NTM infection. PBMCs from a healthy donor were stimulated with M. avium in the presence of plasma from healthy controls or patients with neutralizing anti-IFNγ autoantibodies. IRG1 expression in cells stimulated in the presence of patient plasma was significantly reduced compared to the same cells stimulated in the presence of healthy control plasma (Figure 2.3C). These data suggest 44
that both genetic and acquired immunodeficiency syndromes affecting the IL-12 dependent,
IFNγ mediated signaling pathway result in reduced IRG1 expression and itaconate production in
response to M. avium (Figure 2.1D-E). These low levels of IRG1 could potentially contribute to
the infection susceptibility and immune dysregulation in patients with IFNγ deficiency.
Figure 2.3: Y
A IFN# IFN# Y M.MAC/LPS avium/LPS
X IFNGR TLR2/4/6
JAK1/2 MyD88/IRAK
T cell STAT1 STAT1 NF!B/I!B" STAT4 STAT4
IRF1/8 NF!B JAK1/Tyk2
IL12R X Monocyte/ IRG1 Macrophage
IL-12 IL-12
Compiled IRG1 6h B IRG1 ExpressionIRG1IRG1 ExpressionExpression C IRG1 Expression 1.5 1.5 1.5 Healthy HealthyControls Controls ns ns ✱✱✱ ✱✱✱ ✱✱✱✱ ✱✱✱✱ IFNGR1IFNGR1-/- Patients-/- Patients 1.0 1.0 1.0 RQ IRG1 RQ IRG1 0.5 RQ IRG1 0.5 0.5
0.0 0.0 0.0 Unstimulated M. avium Plasma M. avium UnstimulatedUnstimulated Aab Unstimulated ! MAC (MOI =MAC 10) (MOI = 10) Unstimulated + Control Plasma MAC (MOI = 10)+ IFN
M. avium M. avium
MAC+HealthyMAC+IFNg Control Aab plasma patient plasma Figure 2.3: Irg1 Expression is Dysregulated in the Context of Primary Immunodeficiency Syndromes Affecting IFNγ Signaling. (A) Schematic of IFNγ and LPS signaling pathways required for maximal induction of Irg1 expression and the primary immunodeficiency syndromes that disrupt this signaling. (B) Irg1 mRNA expression in total peripheral blood mononuclear cells (PBMC’s) from 4 healthy controls and 3 IFNGR1 deficient patients following 45 stimulation with M. avium (MOI = 10) for 6 hours. (C) Irg1 mRNA expression in total peripheral blood mononuclear cells from a healthy donor following stimulation with M. avium (MOI = 10) for 6 hours in the presence of plasma from 5 healthy donors or 3 patients with neutralizing anti-IFNγ autoantibodies. (B-C) Gene expression measured by qPCR and reported as relative quantity (RQ) normalized to healthy control, MAC stimulated (B) or healthy control MAC stimulated without plasma (C) samples using housekeeping gene ß-actin. Data were analyzed in (B) by two-way ANOVA with Sidak’s multiple comparisons test and (C) by one- way ANOVA with Tukey’s multiple comparisons test. Error bars indicate mean ± s.e.m. *** p < 0.005. **** p < 0.0001.
While we cannot determine the specific effects of low IRG1 and itaconate in the context of infection in patients with IFNγ deficiency, we can conclude that M. avium induced IRG1 expression in patient PBMC’s serves as a good biomarker for the amount of IFNγ signaling activity (Figure 2.4), accounting for the amount of endogenous or exogenous IFNγ ligand, the signaling capacity of the receptor and the activity of downstream signaling components such as
STAT1.
Figure 2.4:
IRG1 Expression 3 HealthyHealthy ControlsControls ILIL-12Rß1-/--12Rß1-/- patients Patient IFNGR1IFNGR1-/--/- patients Patient 2 (Partial)Partial IFNGR1-/+ IFNGR1-/+ Patientpatients GATA2GATA2-/--/+ patients Patient
RQ IRG1 NEMONEMO patient Patient 1 STAT1STAT1 GOF GOF patient Patient
Subcutaneous IFN! 0 Tofacitinib
Unstimulated
M. avium (MOI = 10)
Figure 2.4: IRG1 expression is a biomarker of IFNγ signaling activity in primary patient PBMC’s. Irg1 mRNA expression in total peripheral blood mononuclear cells (PBMC’s) from 4 healthy controls and various immunodeficiency patients following stimulation with M. avium (MOI = 10) for 6 hours. Red circles indicate that the patient was receiving subcutaneous IFNγ at the time of sample collection and the red square indicates that the patient was receiving treatment with the Jak inhibitor, tofacitinib at the time of sample collection. Gene expression measured by qPCR and reported as relative quantity (RQ) normalized to ß-actin and healthy control, MAC stimulated samples. 46
Cell Intrinsic Responses to M. avium Infection are Not Significantly Altered in Irg1-/- Bone Marrow Derived Macrophages
To elucidate the specific contribution of IRG1 deficiency to the broader condition of IFNγ deficiency, we used bone marrow derived macrophages (BMDM) from Irg1-/- mice as a model to define the extent to which isolated IRG1 deficiency contributed to the increased susceptibility to
M. avium infection observed in patients with defects in IFNγ signaling. Nair et al. demonstrated that the impaired survival in Irg1-/- mice infected with Mtb could be almost entirely recapitulated in a cell type specific knock out of Irg1 in LysM+ cells, suggesting that monocyte/macrophage
IRG1 expression almost exclusively accounted for the susceptibility phenotype [121]. Therefore, we assessed the phenotype of BMDM from wild type and Irg1-/- mice infected in vitro with M. avium. We found no significant differences in colony forming units (CFU) following in vitro infection of BMDM, consistent with published data from in vitro infections with Mtb [121]
(Figure 2.5A). Despite significantly higher levels of Irg1 expression, we did not observe a significant reduction in CFU in wild type BMDM treated with recombinant murine IFNγ (Figure
2.2E and 2.5A). These data suggest that IRG1 and itaconate are not required for the cell intrinsic control of M. avium.
Figure 2.5: 47
IRG1 mRNA Expression in vitro CFU's A In Vitro CFU’s 50 8 ns ns WT 40 IRG1 KO 7 30 CFU 10 20 RQ IRG1
Log 6 10
0 5 γ γ γ IFN
M. avium M. avium Unstimulated RNA plate (48h) M. avium+IFN RNA plate (48h) M. avium+IFN LDH release B LDH release 4 4 UnstimulatedUnstimulated Unstimulated M.M. avium avium(MOI (MOI= =25 25)) 3 M. avium (MOI = 25) 3 IFNIFNgg (200U/mL) (200U/mL) IFNg (200U/mL) 2 M.M. a aviumvium + + IFNIFNgg 2 M. avium + IFNg OD (490)
OD (490) 1 1
0 0 CompiledWT WT Mito WTStressWT Test OCR WT WT WT WT IRG1 KO IRG1 KO IRG1 KO IRG1 KO 3 CompiledIRG1 KO IRG1 Mito KO IRG1Stress KO IRG1 KOTest OCR
C CFU plate (48h) Compiled Mito Stress Test OCR 3 CFU plate (48h) Oligomycin FCCP Rot/AA 3 LDH release LDH release 4 IRG1 KO Unstimulated 4 2 IRG1 KO M. avium M. avium (MOI = 25) 3 IRG1 KO Unstimulated M. avium (MOI = 25) IRG1IRG1 KO KO Unstimulated IFNγ 2 3 2 IRG1IRG1 KO KO M. aviumM. avium + IFNγ M. avium + IFNg 2 IRG1 KO IFNγ 2 M. avium + IFNg WTIRG1 Unstimulated KO IFNγ
OD (490) IRG1 KO M. avium + IFNγ
OD (490) IRG1 KO M. avium + IFNγ 1 WTWT Unstimulated M. avium 1 1 WT Unstimulated WTWT M. IFN aviumγ 0 1 WTWT IFN M.γ avium 1 0 WT M. avium + IFNγ WT WT WTWT M. IFN aviumγ + IFNγ OCR (pmol/min/Norm. Unit) WT WT OCR (pmol/min/Norm. Unit) IRG1 KO IRG1 KO WT M. avium + IFNγ IRG1 KO IRG1 KO OCR (pmol/min/Norm. Unit) 0 0 0 200 20 40 40 60 60 8080 Time (minutes) 0 Time (minutes) 0 20 40 60 80 Figure 2.4: CellTime Intrinsic (minutes) Responses to M. avium Infection are Not Significantly Altered in Irg1-/- Bone Marrow Derived Macrophages. (A) CFU’s quantified from bone marrow derived macrophages from wild type and Irg1-/- mice infected in vitro with M. avium (MOI = 25) or M. avium with exogenous IFNγ (200U/mL) for 48 hours. Data were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. (B-C) Bone marrow derived macrophages from wild type and Irg1-/- mice cultured in vitro and left unstimulated or stimulated with M. avium (MOI = 25), with or without exogenous IFNγ (200U/mL), or IFNγ alone. (B) LDH release assay assessing viability in bone marrow derived macrophages from wild type and Irg1-/- mice from 48 indicated stimulation conditions for 48 hours. (C) Real time monitoring of oxygen consumption rates (OCR) in wild type and Irg1-/- murine macrophages following 24h of stimulation with indicated stimuli. Changes in OCR in response to intra-assay injection of Oligomycin (2uM), FCCP (0.5uM) and Rotenone/Antimycin A (0.5uM) according to the manufacturer’s protocol for the mitochondrial stress test.
We next assessed whether the reported anti-inflammatory and antioxidant roles of IRG1 and itaconate might alter the viability of host cells during infection, resulting in loss of control of the pathogen secondary to host cell death. While there was more cell death when infected cells were also treated with exogenous IFNγ, there were no differences in viability between the wild type and Irg1-/- BMDM (Figure 2.5B). In order to assess the effects of IRG1/itaconate deficiency in a human monocyte model we used wild type and IRG1-/- THP-1 cells. Similar to the case in mice, we found no significant difference in CFU’s or viability following in vitro infection (Figure 2.6
A-B). However, we also noted that THP-1 cells had fundamentally altered cellular metabolism compared to primary human monocytes, as a result of their transformed state. Specifically, while itaconate levels were induced comparably to primary human monocytes, succinate levels followed an opposite trend, increased in primary monocytes when itaconate levels were high but decreased in THP-1 cells with similar levels of itaconate (Figure 2.6 C-D).
Figure 2.6: 49
A In vitro CFU’s (THP-1 cells) B Viability (LDH)
ns ns 0.60 nsns ns ns ns
7 7 0.55 WT WT WT1 IRG1 KO IRG1 KO 0.50 IRG1-/-
0.45 CFU CFU 6 6 10 10 LDH (OD) 0.40 Log Log
0.35
5 5 0.30 M. avium M. avium + IFNγ NoM. aviumInfection M.aviumM. aviumM.avium + IFNγ + IFNγ
Intracellular Itaconate Intracellular Succinate IntracellularC ItaconateIntracellularIntracellular Itaconate D IntracellularIntracellular SuccinateIntracellular Succinate 300 1.5×107 300 300 WT THP-1 1.5×107 1.5×107 WTWT THP THP-1-1 WT THP-1 WT THP-1 WT THP-1 WT THP-1 KO THP-1 IRG1KO KO THP-1 THP- 1 KO THP-1 KO THP-1 KO THP-1 KO THP-1 7 200 200 200 Elu Monos Elu1.0 Monos×10 1.0Elu×10 Monos7 1.0×107 PrimaryElu MonosElu Monocytes Monos Elu Monos
100 100 5.0×106 5.0×106 100 5.0×106 Succinate Peak Area Succinate Peak Area Succinate Peak Area
Itaconate(ng/mL/10^6 cells) 0 Itaconate(ng/mL/10^6 cells) 0 0.0 0.0
Itaconate(ng/mL/10^6 cells) 0 IFNg LPS IFNg LPS 0.0 IFNg LPS IFNg LPS
LPS+IFNg LPS+IFNg LPS+IFNg LPS+IFNg LPS LPS IFNg Unstimulated Unstimulated UnstimulatedIFNg Unstimulated LPS+IFNg LPS+IFNg Unstimulated Unstimulated Figure 2.6: In vitro assessment of IRG1 knock out in a human monocyte (THP-1) cell line. (A) CFU’s quantified from WT and IRG1 KO THP-1 cells infected in vitro with M. avium (MOI = 10) or M. avium with or without exogenous IFNγ (1000U/mL) for 48 hours. Data were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. (B) supernatants from the same samples were used for an LDH release assay to assess viability in different infection conditions. (C-D) Intracellular itaconate levels normalized to cell number (C) and relative values of intracellular succinate (D) in WT and IRG1 KO THP-1 cells compared to primary human monocytes. Data in (A-B) were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. Error bars indicate mean ± s.e.m.
Finally, IRG1 has been shown to be associated with mitochondria, and its substrate (cis-
aconitate) is a Kreb’s cycle intermediate, localized to the inner mitochondrial matrix [127].
Within mitochondria, itaconate can act as a competitive inhibitor of succinate dehydrogenase
(SDH)/mitochondrial complex II, which has been shown to play a role in infection with
Francisella tularensis [117, 192]. Therefore, we assessed the effects of IRG1/itaconate
deficiency on mitochondrial oxidative metabolism in the context of in vitro infection. Again,
while there were differences between different stimulation conditions, we did not observe 50 significant differences in mitochondrial oxygen consumption rates between wild type and Irg1-/-
BMDM (Figure 2.5C). This finding was striking considering that the mitochondrial metabolite, itaconate, is one of the most abundant metabolites in M1 macrophages, yet its absence caused minimal change in mitochondrial metabolism. Interestingly, stimulation with either M. avium or
IFNγ individually in wild type and Irg1-/- BMDM resulted in increased oxygen consumption rates
(OCR), whereas stimulation with M. avium and IFNγ simultaneously resulted in lower OCR, similar to OCR levels in unstimulated macrophages (Figure 2.5C).
Irg1-/- mice have increased mycobacterial burdens following in vivo infection with M. avium
Although we did not observe any differences in the cell intrinsic effects of M. avium on Irg1-
/- BMDMs, we assessed the in vivo response to infection in Irg1-/- mice to look for more complex effects. To mimic the disseminated M. avium infections that are pathognomonic for IFNγ signaling defects in patients, we infected wild type and Irg1-/- C57B/6N mice with M. avium intravenously, as described previously in GKO mice [177] (Figure 2.7A). We first confirmed that in vivo infection with M. avium induced Irg1 expression in wild type, but not Irg1-/- mice.
Strikingly, there was no itaconate induction in the infected Irg1-/- animals (Figures 2.8A-B).
We next assessed survival in Irg1-/- mice. Nair et al. reported susceptibility of Irg1-/- mice to
Mtb infection, in which Irg1-/- mice succumbed to Mtb infection significantly earlier than wild type mice, with 75% mortality in Irg1-/- mice 30 days post infection. Therefore, we assessed survival following in vivo infection with M. avium at 60 days post infection, a time when GKO mice have been reported to show significant differences in pathology and mycobacterial burden following M. avium infection. Both wild type and Irg1-/- mice survived in a relatively healthy condition throughout the infection, consistent with previous reports of NTM infections, even in 51 genetically susceptible mouse models [177, 180]. Weights of all animals were collected weekly and did not change significantly over the 60-day course of infection (Figure 2.8C-D). In fact, there was a slight increase in weight in all animals, potentially caused by the impressive mycobacterial burden in the spleen, liver and lungs (Figure 2.7B and Figure 2.8E). Of note, the
Irg1-/- mice had slightly but consistently lower weights from pre-infection throughout the entire infection (Figure 2.8C).
The phenotypes exhibited by both wild type and Irg1-/- mice in this NTM infection model highlight the significant differences in virulence, pathogenicity and inflammatory response to
NTM (M. avium) compared to those previously reported with Mtb. The response to NTM is marked by minimal inflammation, as evinced in the histology from infected animals (Figure
2.7C). Histopathology in infected animals was predominantly granulomatous (macrophage- heavy) with mixed neutrophils (granulomatous to pyogranulomatous), pneumonia, hepatitis and splenitis, respectively (Figure 2.7C). While studies using Mtb identified a dramatic and pathologic increase in neutrophil infiltration in the lungs of infected Irg1-/- mice, we observed only subtle differences in the amount of neutrophil infiltration in the Irg1-/- animals.
Histopathologic differences between wild type and Irg1-/- mice were observed only at 60 days post infection and were most notable in the spleen (Figure 2.7C).
M. avium CFU were assessed at 30 and 60 days post infection in each organ. M. avium CFU in Irg1-/- mice were significantly increased compared to wild type mice in the lungs and liver at
60 days post infection. However, there was no significant difference in CFU between wild type and Irg1-/- mice in the spleen (Figure 2.7B). It is possible that the slight increase in neutrophil numbers in the spleens of Irg1-/- mice compensated for the lack of IRG1/itaconate and contributed to the control of M. avium. The approximately 10-fold difference in CFU in the liver 52
and lungs is consistent with CFU differences reported in Mtb and BCG infected Irg1-/- mice
[180]. However, despite this excess of mycobacteria, Irg1-/- mice infected with M. avium showed
no other signs of severe illness or overt inflammation.
Figure 2.7:
A Day 0: Day 30: Day 60:
C57BL/6N WT
1.5x10^6 CFU C57BL/6N Irg1-/- M. avium (IV)
30 Days post infection: 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60
B Spleen Liver Lung
✱✱ ✱ Spleen 10 Liver 10 8 Lung 0.0013 7 0.0196 10 10 8 WT WT WT CFU CFU CFU 9 9 6 10 10 10 IRG1-/- IRG1-/- IRG1-/- Log Log Log 7 5 CFU CFU 9 CFU 9 8 8 6 4 10 10 10 30 60 30 60 30 60 Log Log Log Days Post Infection Days Post Infection Days Post Infection 5
8 8 4 #14 WT d60 C #14 WT d60 #14 WT d60 30 60 spleen 30 Spleen* spleenlung 60 Liver spleenliverlung Lung liver30lung 60 liver Days Post Infection Days Post Infection Days Post Infection WT MPO MPO MPO
#710x KO d60 10x#710X (MPO) KO d60 10x#7 KO d60 spleen spleenlung spleenliverlung liverlung liver IBA1 IBA1 IBA1 MPO MPO MPO IRG1 KO Lung 10x10x 10x10x 10x10x
Figure 2.7: Irg1-/- Mice have Higher Mycobacterial Burdens Following In Vivo Infection with Mycobacterium avium. (A) Schematic timeline of in vivo infection model. (B) M. avium IBA1 IBA1 CFU’s from spleens, liversIBA1 and lungs harvested from 5 WT and Irg1-/- mice at 30 and 60 days post in vivo infection with M. avium. (C) Representative images of MPO staining of fixed tissue 10x from10x the spleens, livers and10x lungs of 5 WT and Irg1-/- mice 60 days post in vivo infection with M. avium. (*) indicates significant difference in MPO staining between WT and Irg1-/-. In vivo infection experiments were performed twice. Data were analyzed in (B) by two-way ANOVA 53
with Sidak’s multiple comparisons test. Error bars indicate mean ± s.e.m. * p < 0.05. ** p <
0.001. Spleen Liver Lung 4 4 4 WT Lung WT Liver -/- 3 3 3 IRG1 Lung Figure 2.8: WT Spleen IRG1-/- Liver 2 IRG1-/- Spleen 2 2 RQ Irg1 RQ Irg1 RQ Irg1 1 1 1 A Spleen Liver Lung 4 4 4 0 0 0 WT Lung 0 20 40 60 0 20 40 60 0 20 40 60 WT Liver -/- 3 Time (Days) 3 Time (Days) 3 Time (Days) IRG1 Lung WT Spleen IRG1-/- Liver 2 IRG1-/- Spleen 2 2 RQ Irg1 RQ Irg1 RQ Irg1 1 1 1
0 0 0 0 20 40 60 0 20 40 60 0 20 40 60 Time (Days) Time (Days) Time (Days)
B Spleen Liver Lung 6 10 0.15 Spleen Weight Liver Weight Lung Weight WT Lung 8 WT Liver IRG1-/- Lung 6 0.0002 14 2.0 WT 4 WT Spleen WT IRG1-/- Liver 0.10 WT 0.0400 6 IRG1-/- Spleen
-/- g/mL -/- -/- g/mL g/mL µ µ IRG1 12 4 IRG1 1.5 µ IRG1 4 2 0.05 Spleen 2 Liver Lung 610 10 1.0 0.15 0 0 0.00 WT Lung 0 20 40 60 0 20 40 60 0 20 40 60 2 8 WT Liver IRG1-/- Lung Time (Days) Time (Days) Time (Days) 4 8 WT Spleen 0.5 IRG1-/- Liver 0.10 6 -/- % Total Body Weight % Total Body Weight % Total Body Weight IRG1 Spleen g/mL g/mL g/mL µ µ 4 µ 0 Total Body2 6Weight Total Body Weight 0.0 Weight 0.05 Weight 2 28 60 28 60 28 60 C D Days Post Infection 35 0 35 TotalDays BodyPost Infection Weight (g) 0 140 %Initial Body140 DaysWeight Post0.00 (g) Infection 0 20 40 60 C57BL/6N 0 20 C57BL/6N40 60 0 20 40 60C57BL/6N C57BL/6N Time (Days) Time (Days) Time (Days) -/- 130 -/- 130 -/- -/- 30 30 C57BL/6N Irg1 C57BL/6N Irg1 C57BL/6N Irg1 C57BL/6N Irg1 120 120 25 25 110 110 Weight (g) Weight (g) Weight 20 20 100 100 % Initial Body Weight % Initial Body Weight 15 15 90 90 0 20 040 2060 40 60 0 20 040 2060 40 60 Days Post Infection Days Post Infection Days Post Infection Days Post Infection
Spleen Weight Liver Weight Lung Weight E Spleen Weight Liver Weight Lung Weight 6 13 1.6 WT 6 ns ns 13 ns WT ns 1.6 ns ns WT -/- 12 -/- 1.4 -/- IRG1 12 IRG1 1.4 IRG1 4 4 11 11 1.2 1.2
10 10 1.0 1.0 2 2 9 0.8 % Total Body Weight % Total Body Weight % Total Body Weight 9 0.8 % Total Body Weight % Total Body Weight % Total Body Weight 0 8 0.6
0 8 30 60 30 600.6 30 60
28 58 Days28 Post Infection 58 Days Post Infection 28 Days 58Post Infection Days Post Infection Days Post Infection Days Post Infection Figure 2.8: In vivo infections of Wild Type and Irg1-/- mice with M. avium. (A) Irg1 mRNA expression levels in spleen, liver and lung of WT and Irg1-/- mice pre-infection and 30 and 60 days post infection. (B) itaconate levels in spleen, liver and lung tissue measured by LC-MS. (C) total body weights of WT and Irg1-/- mice measured every 2 weeks throughout the 60 day course of infection. (D) total body weights displayed as percent of initial body weight. (E) organ weights of spleen, liver and lung displayed as organ weight/animal weight (% total body weight) at 30 and 60 days post infection. Data in (E) were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. Error bars indicate mean ± s.e.m. 54
Discussion:
That there is a vital role for IFNγ-enhanced immunity is evidenced by the severe infection susceptibility associated with genetic or acquired immunodeficiencies impairing IFNγ signaling, specifically the dramatic susceptibility of these patients to otherwise avirulent nontuberculous mycobacteria. Our findings in both human monocytes and murine macrophages demonstrate that IFNγ acts synergistically with TLR signals to maximize induction of IRG1 expression and itaconate production. The absence of IRG1 and itaconate induction in response to IFNγ alone supports the hypothesis that itaconate evolved as part of the innate immune response to infection that is enhanced by IFNγ.
Furthermore, we demonstrate that M. avium-induced IRG1 expression in patients with immunodeficiencies affecting IFNγ signaling is significantly reduced compared to healthy controls. However, the extent to which this IRG1 and itaconate deficiency contributes to the infection susceptibility and immune dysregulation associated with defects in IFNγ signaling in these patients remains unclear. We addressed this question using models of isolated IRG1 deficiency including an in vivo model of disseminated M. avium infection in Irg1-/- mice. While the phenotype of Irg1-/- mice infected with M. avium was less dramatic than the phenotype reported for Irg1-/- mice infected with Mtb [121], we observed a significantly increased mycobacterial burden in the liver and lungs of Irg1-/- mice. However, unlike Mtb infected Irg1-/- mice, we did not observe any significant differences in mortality or tissue pathology. NTM are known to be less virulent than Mtb, even in genetically susceptible models of infection, due to multiple genetic, structural and metabolic features that distinguish them from Mtb [193-195].
Hoffman et al., described similar findings comparing Mtb to BCG-infected Irg1-/- mice. 55
Intranasal infection of Irg1-/- mice with Mycobacterium bovis BCG did not cause immunopathology in the lungs but did result in persistently increased mycobacterial burdens
(about 10-fold higher than wild type mice) throughout the 86-day course of infection [180]. In humans, despite a milder acute inflammatory response than Mtb, M. avium infection causes diffusely disseminated, persistent infection in susceptible patients that is difficult to eradicate even with multiple antimycobacterial drugs. This phenotype is consistent with our findings in
Irg1-/- mice and suggests that IRG1 likely contributes to the broader infection susceptibility associated with defects in IFNγ signaling.
Murine knock out animals have some limitations as models of human mycobacterial infection. There is evidence to suggest that knock out animals may respond to infection with compensatory mechanisms which may obscure our understanding of the normal activity of the gene of interest [196]. Additionally, Price et al. described the complex and redundant mechanisms by which IFNγ induces effective antimicrobial responses against Legionella pneumophila. That study specifically investigated the role of IRG1 in combination with other
IFNγ inducible genes in L. pneumophila infection and concluded that knock out of IRG1 along with CASP11, IRGM1, IRGM3, and NOX2 reduces, but does not completely abrogate the L. pneumophila growth restriction conferred by IFNγ treatment. A 6-gene KO including iNOS (in addition to IRG1, CASP11, IRGM1, IRGM3 and NOX2) resulted in a complete loss of the protective effects of IFNγ treatment, suggesting redundancy between iNOS and IRG1 [197].
While Price et al. clarified the complex, redundant roles of IRG1 and iNOS in murine macrophages, the role of iNOS in human monocytes remains controversial. iNOS expression is not induced in human monocytes or macrophages in vitro as it is in murine macrophages [63].
Therefore, there are limitations to using mouse models to understand the cell intrinsic response 56 to macrophage infection with mycobacteria. For this reason, we also assessed a human model of
IRG1 deficiency using an IRG1 knock out THP-1 cell line. This model proved to be significantly different from primary human monocytes and macrophages at a metabolic level. These metabolic differences present similar limitations for extending our interpretation of results beyond the in vitro setting.
One “in vivo” method to explore a role for IRG1/itaconate in human health and disease is through investigation of the many variants in IRG1 reported in gnomAD
(https://gnomad.broadinstitute.org). Chen at al. (2019) demonstrated that most of the reported
IRG1 variants are extremely rare and lead to dramatically reduced itaconate production.
Conversely, the few variants that are reported to occur with higher frequencies do not affect itaconate production. Furthermore, the most common missense variant (Asn152Ser) functionally increases itaconate production and therefore, the authors suggest that it may be protective against certain pathogens such as mycobacteria [198]. These findings suggest a selective advantage for
IRG1 variants that augment or maintain normal itaconate production, supporting a critical role for itaconate in the context of infection.
In sum, our study demonstrates that IRG1 and itaconate are effectors of host defense that are synergistically induced by IFNγ in combination with a TLR signal, in both humans and mice in the setting of M. avium infection. The specific mechanism of action of itaconate remains unclear, but our results suggest that its effects are not exclusively cell intrinsic.
The use of primary human cells allowed us to avoid the pitfalls associated with alternative models such as murine macrophages which differentially regulate and utilize iNOS in response to mycobacterial infection and human myeloid cell lines which may have fundamentally altered cell metabolism as a result of their immortalized state. With these limitations in mind, it was 57 striking to find that IRG1 and itaconate were highly and synergistically induced in response to stimulation with a TLR signal and IFNγ in both human and murine macrophages. However, despite this dramatic induction, complete loss of itaconate had little effect on cell intrinsic killing of mycobacteria, cellular viability during in vitro infection or mitochondrial oxidative metabolism. These findings suggest a complex role for itaconate, on the one hand tied closely to interferon activation and on the other important but somewhat redundant for antimicrobial effect.
Future work is required to assess the precise mechanisms of the IRG1-itaconate pathway and other metabolic pathways in primary human immune cells as possible targets for host directed therapies for infectious and immunologic diseases. These data identify IRG1 as a sensitive biomarker of IFNγ signaling in the context of bacterial infection and as a means of identifying patients with infection susceptibility caused by defective IFNγ signaling. Furthermore, these data suggest a partial role for IRG1 and itaconate in both murine and human IFNγ mediated defense against M. avium and support efforts to target this pathway with therapeutics aimed at restoring or augmenting itaconate levels in patients who may have reduced itaconate production.
58
CHAPTER III: IFNγ Regulates NAD+ Metabolism in Human Monocytes
Katelyn J. McCann1,2, Stephen M. Christensen3, Peter J. McGuire4, Ian A. Myles5, Christa S. Zerbe1, Clifton L. Dalgard6,7, Gauthaman Sukumar6,8, Warren J. Leonard3, Beth A. McCormick2, Steven M. Holland1 1 Immunopathogenesis Section, Laboratory of Clinical Immunology and Microbiology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA 2 Department of Microbiology and Physiological Systems and Program in Microbiome Dynamics, University of Massachusetts Medical School, Worcester, MA, USA 3 Laboratory of Molecular Immunology and the Immunology Center, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD, USA 4 Metabolism, Infection and Immunity Section, National Human Genome Research Institute, National Institutes of Health, Bethesda, MD, USA 5 Epithelial Therapeutics Unit, Laboratory of Clinical Immunology and Microbiology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA 6The American Genome Center, Uniformed Services University of the Health Sciences, Bethesda, MD, USA 7Department of Anatomy, Physiology & Genetics, Uniformed Services University of the Health Sciences, Bethesda, MD, USA 8Henry M. Jackson Foundation for the Advancement of Military Medicine, Bethesda, MD, USA
Parts of this chapter are part of a manuscript submitted for review: Blood manuscript identification number BLD-2021-012943 Funding for this study was provided by the Division of Intramural Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health.
59
Introduction:
Interferon gamma (IFNγ) is an essential activator of macrophages, inducing pro- inflammatory cytokines, augmenting intracellular microbial killing and stimulating the production of microbicidal and signal-propagating reactive oxygen/nitrogen species [2, 199].
IFNγ is dysregulated in myriad diseases. Functional IFNγ deficiency is caused by genetic mutations in IFNγ, its receptor, or by acquired neutralizing anti-IFNγ autoantibodies, impairing control of intracellular bacteria and causing systemic immune dysregulation [20, 85, 92, 100-
102, 174-176]. Conversely, states of excessive IFNγ activity, such as signal transducer and activator of transcription 1 (STAT1) gain of function (GOF), have inflammatory complications
[111, 200-203]. The effects of IFNγ on macrophages are pleiotropic and likely energetically expensive, requiring metabolic reprogramming to support this increased demand.
Much work has focused on the early metabolic changes in macrophages in response to lipopolysaccharide (LPS), known as the Warburg effect (aerobic glycolysis) [157, 204-206], but fewer data exist for metabolic responses to other stimuli. Cameron et al. demonstrated that M1
(classically activated macrophages) depend on the nicotinamide adenine dinucleotide (NAD(H)) salvage pathway to sustain aerobic glycolysis and their M1 phenotype following stimulation with
LPS or LPS+IFNγ [207]. However, the metabolic reprogramming in macrophages induced by
IFNγ alone, independent of LPS, is not well characterized.
IFNγ is known to regulate phagocyte activation through metabolic adaptations that produce reactive oxygen species (ROS) [69, 71-73, 208]. Superoxide generation through oxidation of phosphorylated NADH (NADPH) by the NADPH oxidase complex is known as the respiratory burst [209]. The same NAD+ salvage pathways that were identified as critical for maintaining aerobic glycolysis in LPS-stimulated macrophages also supply the intracellular 60
NAD+ required to produce NADP(H) to support NADPH oxidase superoxide production [210].
Therefore, we hypothesized that IFNγ-mediated activation of the respiratory burst might also depend on NAD+ salvage pathways.
We used primary human monocytes to characterize the metabolic reprograming that occurs in response to stimulation with IFNγ. We identified an IFNγ-induced, nicotinamide phosphoribosyltransferase (NAMPT)-dependent pathway of NAD+ salvage that is required for the highly oxidative metabolic phenotype induced by IFNγ. We used primary monocytes from normal donors and patients with genetic defects in various steps of this pathway as well as chemical inhibitors to demonstrate that IFNγ augments NAD+ biosynthesis and reduction/oxidation (redox) metabolism to increase oxygen consumption. Our findings suggest that the IFNγ-induced respiratory burst in monocytes is dependent on NAMPT-mediated NAD+ salvage. This metabolic pathway may include novel targets for therapeutic modulation of monocyte activation.
61
Results:
IFNγ increases monocyte oxygen consumption rates (OCR)
To further elucidate the metabolic processes that support IFNγ-mediated monocyte
activation, we first assessed mitochondrial oxidative phosphorylation by monitoring the real time
oxygen consumption rate (OCR) by Seahorse assay in primary human monocytes stimulated
with IFNγ for 24 hours and then subjected to the Mito Stress Test. We observed a significant
increase (p<0.0001) in basal OCR in monocytes stimulated with IFNγ for 24 hours (Figure
3.1A).
We modified this assay to assess how IFNγ reprograms monocyte metabolism and primes
cells metabolically to respond to the robust activating stimulus, phorbol myristate acetate (PMA).
6 1.0 Unstimulated Unstimulated Oligomycin Oligomycin Monocytes were stimulated with IFNγ for 24 hours prior to starting the assay. Basal0.8 IFNγ Glucose IFNγ 4 2-DG FCCP Rot/AA 0.6 measurements of OCR were collected, then PMA (100ng/mL) was injected, and the real time 0.4 2 0.2 6 1.0 response was monitored.Unstimulated Monocytes stimulated with IFNγ demonstrated a significantUnstimulated increase in
OCR (pmol/min/Norm. Unit) 0 0.0 Oligomycin Oligomycin ECAR (mpH/min/Norm. Unit) 0 0.8 20 40 60 80 0 20 40 60 80 IFNγ Glucose IFNγ Time (minutes) Time (minutes) 4 2-DG FCCPboth basal (p=0.003) and PMA-stimulated (p<0.0001) OCR (Figure 3.1B-C). Rot/AA 0.6
0.4 2 Figure 3.1: 0.2 A Mito Stress Test B Modified OCR Assay C Basal and PMA-Stimulated
OCR (pmol/min/Norm. Unit) 0 0.0 ECAR (mpH/min/Norm. Unit) OCR Values Rot/AA 0 20 40 60 80 0 20 40 60 15 80 6 15 PMA1.0 - 1.5 ✱✱✱✱ Unstimulated Unstimulated Rot/AA Unstimulated Unstimulated Unstimulated Time (minutes) Oligomycin stimulatedTime (minutes)Oligomycin 0.8 IFNγ IFNγ Glucose IFNγ IFNγ PMA IFNγ 4 10 102-DG 1.0 FCCP Rot/AA PMA 0.6 ✱✱ 0.4 2 5 Basal 5 0.5 0.2 OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 OCR (pmol/min/Norm. Unit) 0.0
0 0.0 ECAR (mpH/min/Norm. Unit) ECAR (mpH/min/Norm. Unit) 0 0 20 40 60 80 0 20 40 600 8020 100 40 60 Basal80 0PMAPMA-induced-stimulated20 40 60 80 100 Time (minutes) Time (minutes) TimeRot/AA (minutes) Time (minutes) 15 1.5 Rot/AA Unstimulated Unstimulated 24 25 24 25 Time post IFN! stimulationIFN (hours)γ Time post IFN! stimulationPMA (hours) IFNγ 10 1.0 PMA Figure 3.1: IFNγ Increases Monocyte Oxygen Consumption Rate (OCR). (A-B) Real time 5 0.5 Rot/AA 15 1.5 changes in OCR measuredRot/AA using SeahorseUnstimulated extracellular flux analyzer in primary humanUnstimulated
IFNγ PMA IFNγ
OCR (pmol/min/Norm. Unit) 0 0.0 10 ECAR (mpH/min/Norm. Unit) 1.0 0 20 40 60 80 PMA100 0 20 40 60 80 100 Time (minutes) Time (minutes) 5 0.5
OCR (pmol/min/Norm. Unit) 0 0.0 ECAR (mpH/min/Norm. Unit) 0 20 40 60 80 100 0 20 40 60 80 100 Time (minutes) Time (minutes) 62 monocytes treated with IFNγ or media alone for 24 hours prior to the start of the assay (n = 5 technical replicates). Stimuli/inhibitors were injected during the assay according to the Seahorse mitochondrial stress test (Oligomycin, FCCP, Rotenone/Antimycin A) (A), or a modified assay (PMA, Rotenone/Antimycin A) (B). Data in (A) are representative tracings of three independent experiments. Data in (B) are representative of more than fifteen independent experiments. (C) The last measurement before PMA injection is used to quantify “basal” OCR, and the last measurement before Rotenone/Antimycin A is used to quantify “PMA-stimulated” OCR throughout the study. Data in (C) were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. Error bars are mean ± s.e.m **p<0.01 and ****p<0.0001.
We also measured the immediate changes in glycolysis by extracellular acidification rate
(ECAR) monitoring in response to stimulation with LPS, IFNγ or the combination of LPS and
IFNγ (LPS+IFNγ). Consistent with studies describing a switch to aerobic glycolysis in response to LPS[133, 188, 211], we found that LPS and the combination of LPS+IFNγ robustly increased
ECAR within 20 minutes of stimulation. IFNγ alone had no immediate effect on ECAR (Figure
3.2A) and there were no significant changes in OCR within 1 hour of stimulation (Figure 3.2A).
Interestingly, both ECAR and OCR were increased after 24 hours of stimulation with IFNγ alone, but not with LPS or LPS+IFNγ (Figures 3.2B-E). These findings confirmed that our methods for metabolic monitoring in primary human monocytes can detect rapid and stimulus- specific changes in metabolism and that IFNγ significantly increased both basal and PMA- stimulated OCR.
63
Figure 3.2:
Immediate Response/Acute Injection A OCR ECAR 4 4 1.0 1.0 Stimulation Stimulation Unstimulated UnstimulatedStimulation Stimulation Medium Medium LPS0.8 0.8 LPS LPS 4 LPS 1.0 3 Stimulation 3 IFNγUnstimulated IFNγ Stimulation IFNMediumγ IFNγ 0.6 0.6 LPS 0.8 LPS 2 3 2 LPS+IFNγ LPS+IFNγ LPS+IFNγ LPS+IFNγ IFNγ IFNγ 0.4 0.6 0.4 LPS+IFNγ LPS+IFNγ 1 2 1 0.2 0.4 0.2 1 OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0.0 0 0.0 0.2 ECAR (mpH/min/Norm. Unit) ECAR (mpH/min/Norm. Unit) 0 20 0 40 20 60 40 60 0 20 0 40 20 60 40 60
OCR (pmol/min/Norm. Unit) 0 0.0 Time (minutes) Time (minutes) ECAR (mpH/min/Norm. Unit) Time (minutes) Time (minutes) 0 20 40 60 0 20 40 60 Time (minutes) Time (minutes) B C 6 6 Mito Stress Test 1.0 Glycolysis Stress1.0 Test 6 6 Unstimulated Unstimulated1.0 1.0 Unstimulated Unstimulated Oligomycin Oligomycin Unstimulated Unstimulated Oligomycin Oligomycin Unstimulated Unstimulated LPS LPS0.8 0.8 LPS LPS 4Oligomycin Oligomycin 1.0 Glucose Oligomycin Glucose Oligomycin Stimulation 0.8 0.8 2-DG 4 4 LPS Unstimulated IFNLPSγ StimulationGlucose 2-DG Glucose MediumLPS IFNLPSγ FCCP FCCP IFNγ IFNγ 4 4 Rot/AA Rot/AA 0.6 0.6 2-DG 2-DG 6 FCCP 6 FCCP IFNγLPS IFN0.8γ1.0 1.0 LPSIFNγ IFNγ 3 Rot/AA LPS+IFNRot/AAUnstimulatedγ LPS+IFN0.6Unstimulatedγ 0.6 LPS+IFNUnstimulatedγ LPS+IFNUnstimulatedγ 6Oligomycin Oligomycin LPS+IFNIFNγγ LPS+IFN0.4 1.0γ Oligomycin0.4 OligomycinIFNLPS+IFNγ γ LPS+IFNγ Unstimulated 0.60.8 0.8 Unstimulated 2 2 LPS 0.4LPS Glucose 0.4 Glucose LPS LPS 2 Oligomycin LPS+IFNγ Oligomycin LPS+IFNγ 2 4 24 LPS 0.8 0.2 2-DG 2-DGLPS FCCP FCCP IFNγ 0.2IFNγ Glucose IFNγ IFNγ Rot/AA Rot/AA 0.40.6 0.6 4 IFNγ 0.2 0.2 2-DG IFNγ FCCP LPS+IFNγ LPS+IFN0.6 γ LPS+IFNγ LPS+IFNγ OCR (pmol/min/Norm. Unit) Rot/AA OCR (pmol/min/Norm. Unit) 0 0.0 0 1 0.0 ECAR (mpH/min/Norm. Unit) LPS+IFNγ ECAR (mpH/min/Norm. Unit) 0.20.4 0.4 LPS+IFNγ OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0 20 40 60 80 0.0 0 20 40 60 80
0 0 20 40 60 80 0.0 0 20 40ECAR (mpH/min/Norm. Unit) 60 80 2 2 ECAR (mpH/min/Norm. Unit) 0 20 40 0 60 20 Time80 (minutes)40 60 80 0 0.4 20 40 0 60 20 Time80 (minutes)40 60 80 OCR (pmol/min/Norm. Unit) 0 2 Time (minutes) 0.0 Time (minutes) ECAR (mpH/min/Norm. Unit) 0.2 0.2 0 Time (minutes)20 40 Time60 (minutes) 0 20Time (minutes)40 60Time (minutes) 0.2 Time (minutes) Time (minutes) OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0.0 0 0.0 ECAR (mpH/min/Norm. Unit) ECAR (mpH/min/Norm. Unit)
OCR (pmol/min/Norm. Unit) 0 0.0 0 20 40 0 6020 8040 60 80 ECAR (mpH/min/Norm. Unit) 0 20 40 0 60 20 80 40 60 80 0 20 24 40 60 80 25 24 0 20 40 25 60 80 Time (minutes) Time (minutes) Time (minutes) Time (minutes) Time Time(minutes) post IFN! stimulation (hours) Time post IFN!Timestimulation (minutes) (hours)
D OCR E ECAR Rot/AA Rot/AA 6 15 1.0 15 Unstimulated 1.5 1.5 Rot/AA UnstimulatedRot/AA 4 UnstimulatedRot/AA Unstimulated1.0 15 StimulationOligomycin 15 Rot/AA Unstimulated 1.5 Stimulation Oligomycin1.5 MediumUnstimulated Unstimulated Unstimulated0.8 Rot/AA Rot/AA Rot/AA UnstimulatedRot/AALPS LPS Glucose Rot/AA LPSUnstimulated Unstimulated LPSLPS 0.8 PMA PMA LPSLPS LPS 15 43 15 15 1.5 1.5 1.5 2-DG 10 FCCP10 Rot/AARot/AA LPSIFNRot/AAγ LPSUnstimulated 1.0 PMA IFNLPSγ LPSUnstimulated Rot/AA IFNγIFNUnstimulatedγ Unstimulated IFN1.00.6γ PMA IFNIFNγ γUnstimulated IFNγ 10 PMA 10 PMA LPS+IFNγ IFN1.00.6γ 1.0 LPS+IFNγ IFNγ LPS LPS+IFNLPS γ PMA IFNγLPS IFNLPSγ 2 PMA PMA LPS+IFNLPS+IFNγLPSγ PMAPMA LPS+IFNLPS+IFNLPSγ γ LPS+IFNγ LPS+IFN0.4 γ 10 2 10 10 LPS+IFNIFNγIFNγ 0.4IFN1.0γ1.0 1.0 LPS+IFNIFNIFNγγ LPS+IFNIFNγ γ 5 PMA PMA5 PMA 0.5 0.5 5 1 5 LPS+IFNLPS+IFNγ γ 0.50.2LPS+IFNγ 0.5 LPS+IFNγγ LPS+IFNγ 0.2
OCR (pmol/min/Norm. Unit) 0 5 0.0 0.5 5 5 ECAR (mpH/min/Norm. Unit) 0.5 0.5 OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0.0 0 0.0 ECAR (mpH/min/Norm. Unit)
0 ECAR (mpH/min/Norm. Unit) 0.0 0 20 40 60 80 ECAR (mpH/min/Norm. Unit) 0 20 40 60 80 OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0 20 4020 0 060 40 2080 4060 60 80 0 20 20 40 0.040060 2080 60 40 60 80 0 0.0 ECAR (mpH/min/Norm. Unit) Time (minutes) ECAR (mpH/min/Norm. Unit) Time (minutes) 0 20 Time40 (minutes)Time (minutes)060 2080 Time40 (minutes)60 80 0 20 TimeTime40 (minutes)(minutes)060 2080 Time40 (minutes)60 80 OCR (pmol/min/Norm. Unit) 0 0.0 ECAR (mpH/min/Norm. Unit) OCR (pmol/min/Norm. Unit)
OCR (pmol/min/Norm. Unit) 0 0.0 0 0.0 ECAR (mpH/min/Norm. Unit) 0Time (minutes)20 40 60 Time80 (minutes)100 ECAR (mpH/min/Norm. Unit) 0 20Time (minutes)40 60 80 Time100 (minutes) 0 20 40 060 2080 40100 60 80 100 0 20 40 060 2080 40100 60 80 100 Time (minutes) Time (minutes) Time (minutes)24 Time (minutes)25 24 Time (minutes)25 Time (minutes) Time post IFN! stimulation (hours) Time post IFN! stimulation (hours) Rot/AA 15 1.5 6 Rot/AA Unstimulated 1.0 Unstimulated Unstimulated Unstimulated Oligomycin LPS PMAOligomycin LPS LPS 0.8 LPS 10 Figure 3.2: ImmediateIFNγ and prolonged1.0 Glucose metabolic responsesIFNγ to LPS and/or IFNγ. (A) Real 4 PMA 2-DG FCCP IFNγ IFNγ Rot/AA LPS+IFNγ 0.6 LPS+IFNγ time changes in OCRLPS+IFN andγ ECAR measured using SeahorseLPS+IFN extracellularγ flux analyzer in primary 5 0.40.5 2 human monocytes immediately following treatment with media alone, LPS, IFNγ or LPS+IFNγ 0.2
OCR (pmol/min/Norm. Unit) 0 0.0 (n = 5 technical replicates). DataECAR (mpH/min/Norm. Unit) in (A) are representative tracings of three independent
OCR (pmol/min/Norm. Unit) 0 0.0 0 20 40 60 80 100 ECAR (mpH/min/Norm. Unit) 0 20 40 60 80 100 0 20 40 60 80 0 20 40 60 80 Time (minutes) Time (minutes) experimentsTime (minutes) . Primary human monocytesTime were (minutes) stimulated with media alone, LPS, IFNγ or LPS+IFNγ prior to the start of the assay then (B) OCR was measured at baseline and following injection of mitochondrial inhibitors according to the Seahorse mitochondrial stress test
(Oligomycin, FCCP, Rotenone/Antimycin A) orRot/AA (C) ECAR was measured following injection of 15 1.5 glucose, oligomycin,Rot/AA Unstimulated 2-deoxyglucose according to the SeahorseUnstimulated glycolysis stress test. Data in (B- LPS PMA LPS 10 C) are representativeIFN tracingsγ of1.0 three independent experiments.IFNγ (D-E) Primary human PMA monocytes were stimulatedLPS+IFNγ with media alone, LPS, IFNγ orLPS+IFN LPS+γ IFNγ prior to the start of the 5 assay then PMA (100ng/mL) was0.5 injected followed by Rotenone/Antimycin A. Data in (D-E) are
OCR (pmol/min/Norm. Unit) 0 0.0 representative of ten independentECAR (mpH/min/Norm. Unit) experiments. 0 20 40 60 80 100 0 20 40 60 80 100 Time (minutes) Time (minutes) 64
IFNγ-induced oxygen consumption is dependent on NAMPT mediated NAD+ salvage
Wu et al. reported increased OCR and fatty acid oxidation in response to type I interferons in plasmacytoid dendritic cells[212]. Therefore, we first compared metabolic responses to IFNγ (type II interferon) to IFNα and IFNβ (type I interferons). We found that
IFNα and IFNβ induced significantly less basal and PMA-stimulated OCR (~1.5-fold increase) than IFNγ, (2-3-fold increase; Figure 3.3A). These data suggest a distinct metabolic effect of
IFNγ that drives oxygen consumption.
We determined that IFNγ-induced OCR was not a result of mitochondrial biogenesis leading to increased total mitochondrial mass (mitochondrial mass was significantly lower after
IFNγ stimulation, Figure 3.3B) and glucose uptake was only modestly increased (Figure 3.3C).
When we inhibited the three primary carbon sources fueling the tricarboxylic acid (TCA) cycle and mitochondrial oxidative metabolism (Figure 3.3D) we found that none of the individual inhibitors tested significantly reduced IFNγ-induced increases in OCR (Figure 3.3E). Therefore, we concluded that IFNγ-induced oxygen consumption was not exclusively dependent on the oxidation of pyruvate, glutamine, or fatty acids.
65
Figure 3.3:
A Oxygen Consumption Rate Basal and PMA-stimulated OCR Induced by Type I vs. Type II IFN Induced by Type I vs. Type II IFN 8 8 8 Medium MediumMedia IFNγ (50ng/mL) IFNIFNγ γ(50ng/mL) (50ng/mL) 6 6 6 PMA IFNγ (5ng/mL)PMA IFNIFNγ γ(5ng/mL) (5ng/mL) 4 4 4 IFNα (50ng/mL) IFNIFNαα (50ng/mL) (50ng/mL) IFNα (5ng/mL) IFNIFNαα (5ng/mL) (5ng/mL) 2 2 2 IFNβ (50ng/mL) IFNIFNββ (50ng/mL) (50ng/mL) IFNβ (5ng/mL) IFNIFNββ (5ng/mL) (5ng/mL) OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0 OCR (pmol/min/Norm. Unit) 0 0 20 40 60 0 Basal20 PMA-stimulated40 60 Time (minutes) Time (minutes)
Total Mitochondrial Mass Outer Mitochondrial B D Membrane 2H+ 4H+ 2H+ 3H+ ✱✱ Intermembrane Space 8000 ✱✱ ✱✱ 8000 CoQ Cyt C Complex I Complex III Complex IV Inner Mitochondrial ✱✱ Membrane 6000 Complex II Complex V 6000 Mitochondrial Matrix 4000 NAD+ ½ O H2O 4000 NADH 2
2000 TCA 3H+ NADH ATP + H2O 2000 Cycle UK5099 ADP + Pi 0 X Pyruvate γ γ
Geometric MFI (MitoTracker Green) 0 LPS IFN γ γ Geometric MFI (MitoTracker Green) LPS IFN LPS+IFN Acetyl-CoA Fatty Acid Oxidation Unstimulated X LPS+IFN Glutamine Glutamate Etomoxir Unstimulated X C Glucose✱✱✱✱ Uptake BPTES ✱✱✱✱ ✱
✱ 2500 ✱✱✱✱ E Mito Fuel Flex Test 2500 ✱✱✱✱ Rot/AA No Treatment + Medium 2000 PMA 2000 20 No Treatment + IFNγ 1500 BPTES + Medium 1500 15 1000 BPTES + IFNγ 1000 500 10 UK5099 + Medium 500 Geometric MFI (2-NBDG) UK5099 + IFNγ 0 5 Geometric MFI (2-NBDG) 0 γ γ Etomoxir + Medium LPS IFN
γ γ OCR (pmol/min/Norm. Unit) 0 Etomoxir + IFNγ LPS IFN LPS+IFN 0 20 40 60 80 100 Unstimulated LPS+IFN Time (minutes) Unstimulated
Figure 3.3: Mechanism of IFNγ induced OCR. (A) Real time changes OCR measured by Seahorse following treatment with media alone or high (50ng/mL) or low (5ng/mL) dose of IFNγ, IFNα, or IFNß for 24 hours prior to the start of the assay. Basal OCR was measured then PMA (100ng/mL) was injected during the assay and OCR was monitored (n = 5 technical replicates). Data in (A) are representative tracings with bar graphs indicating basal and PMA- stimulated OCR of three independent experiments. (B-C) Primary human monocytes were stimulated with media alone, LPS, IFNγ or LPS+IFNγ prior to the start of the assay then (B) total mitochondrial mass was measured using MitoTracker Green staining and (C) glucose uptake was measured using 2-NBDG by flow cytometry and the geometric mean fluorescent intensities are indicated in the bar graphs. (D) Schematic representation of the primary mitochondrial fuel sources and their respective inhibitors used in (E). Primary human monocytes were stimulated 66 with or without IFNγ and in the presence of absence of metabolic inhibitors (BPTES, UK5099, Etomoxir) for 24 hours prior to the start of the assay then PMA (100ng/mL) was injected followed by Rotenone/Antimycin A and OCR was monitored. For data in A and E, n=5 technical replicates and the tracings are representative of three independent experiments.
Using RNA sequencing (RNA-Seq) data from IFNγ-stimulated, human monocyte derived macrophages, unbiased assessment of the metabolic pathways transcriptionally regulated by
IFNγ[213] identified “NAD and NADP metabolism” as the most significantly upregulated metabolic pathway (Figure 3.4A). We found that IFNγ both upregulated NAMPT, the rate limiting enzyme in nicotinamide (NAM)-dependent NAD+ salvage,[207, 214] and simultaneously downregulated nicotinamide riboside kinase 2 (NRK2) (Figure 3.4B), which mediates the other arm of the NAD+ salvage pathway from nicotinamide ribose (NR) (Figure
3.4C).
Figure 3.4:
A NAD and NADP metabolism B C + IFN! Stimulated Human Monocyte NAD Salvage Pathway Immmune processes 30 Eukaryotic secretion sorting Derived Macrophages NAM NR Eukaryotic information processing Cancer panels NAMPT Biosynthesis of GTP derived cofactors 20 NRK2 NAMPT X FK866 NRK2 Amino acid metabolism Cellular processes Phosphorpus metabolism NMN 10 Lipid metabolism -log(adj.P.Val) Membrane transport NMNAT Uncategorized Carbohydrate metabolism 0 NAD+ One carbon metabolism -10 -5 0 5 10 15 Human signal transduction Cell cycle logFC Purine biosynthesis
Pyrimidine biosynthesis ✱✱ -4 -2 0 2 E ✱✱✱✱ ✱✱✱✱ 8 Mean Log2 Fold Change Unstimulated D IFNg 6 ✱✱✱✱ 10 FK866 Unstimulated 10 Medium ✱✱✱✱ ✱✱✱✱ Medium 4 FK866 + IFNg IFNγ PMA IFNγ PMA Medium + FK866 Medium + FK866 2
5 IFNγ + FK866 Fold Change OCR 5 IFNγ + FK866 0 IFNγ - + - + - + - + FK866 OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0 - - + + - - + + 0 0 20 20 40 40 60 60 PMA TimeTime (minutes) (minutes) - - - - + + + + PMA- Basal stimulated 67
Figure 3.4: NAMPT is induced by IFNγ and is required for increases in basal and PMA stimulated oxygen consumption. (A) Metabolic pathway analysis determined by the ERGO analysis platform based on differential gene expression from RNA sequencing data shown in (B). (B) RNA sequencing of human monocyte derived macrophages stimulated with IFNγ for 24 hours. (C) Diagram representing both NAM/NAMPT and NR/NRK2 dependent NAD+ salvage pathways. (D) OCR measured by Seahorse in primary human monocytes treated with media alone or IFNγ with or without NAMPT inhibitor, FK866 (5nM), for 24 hours prior to the start of the assay (n = 5 technical replicates). Data in (D) are from a representative tracing from three independent experiments. (E) Basal and PMA-stimulated OCR values from 3 independent experiments are summarized following normalization to the basal OCR values from samples without IFNγ, FK866, or PMA treatment. Data in (E) were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. Error bars are mean ± s.e.m **p<0.01, ***p<0.001 and ****p<0.0001.
Addition of exogenous NAMPT substrate, NAM, did not augment IFNγ-induced increases in oxygen consumption (Figure 3.5A), consistent with the understanding that NAMPT activity is the rate limiting step in NAM-dependent NAD+ salvage[215]. We also assessed the dose-dependent inhibitory effects of the NAMPT inhibitor, FK866 (APO866, Daporinad), and selected a dose of 5nM FK866 to minimize off target effects (Figure 3.5B).
FK866 completely blocked the IFNγ-induced increases in basal and PMA-stimulated
OCR. Interestingly, NAMPT inhibition preferentially affected IFNγ stimulated cells (Figure
3.4D), likely because IFNγ actively downregulates NRK2 expression, making IFNγ stimulated cells even more dependent on NAMPT for NAD+ salvage (Figure 3.4B-C). IFNγ stimulation increased basal OCR 2.2-fold and PMA-stimulated OCR 1.4-fold compared to monocytes without IFNγ stimulation. However, when FK866 was added throughout the 24-hour IFNγ stimulation period, the IFNγ-induced increases in basal and PMA-stimulated OCR were completely abrogated (Figure 3.4E).
Glycolysis is an NAD+-dependent pathway in which GAPDH activity depends on reduction of NAD+ to NADH[207]. Consistent with this finding, we observed an inhibition of
IFNγ-induced glycolysis with NAMPT inhibition (Figure 3.5C)[207]. Other inhibitors of OCR 68 such as diphenyleneiodonium (DPI) cause a compensatory increase in basal ECAR when OCR is blocked (Figure 3.5D). Following PMA stimulation, DPI treated monocytes achieve a similar maximum ECAR, while ECAR in IFNγ+FK866 treated monocytes remains significantly reduced
(Figure 3.5C-D). Because FK866 uniquely inhibited both OCR and ECAR, we assessed cytotoxicity induced by these inhibitors. While there was a significant increase in cytotoxicity with FK866 treatment, it was similar to the level of cytotoxicity associated with DPI treatment
(Figure 3.5E). Therefore, we concluded that cell death was not responsible for the reduction in
OCR and ECAR in response to FK866 treatment. Taken together, these results suggest that IFNγ reprograms NAD+ salvage and renders monocytes dependent on NAMPT activity to maintain elevated levels of basal and PMA-stimulated OCR induced by IFNγ.
Figure 3.5: A NAM B FK866 15 15
Medium Medium 10 IFNγ 10 IFNγ PMA PMA Medium + NAM Medium + FK866 IFN NAM 1mM 5 γ + 5 IFNγ + FK866 500nM IFNγ + NAM 10uM IFNγ + FK866 5nM IFNγ + NAM 1uM IFNγ + FK866 0.5nM OCR (pmol/min/Norm. Unit) 0 OCR (pmol/min/Norm. Unit) 0 0 20 40 60 0 20 40 60 Time (minutes) Time (minutes)
C FK866 ECAR D DPI ECAR
1.5 1.5 PMA
PMA Medium Medium 1.0 IFNγ 1.0 IFNγ Medium + FK866 Medium + DPI IFNγ + FK866 IFN + DPI 0.5 0.5 γ
0.0 0.0 ECAR (mpH/min/Norm. Unit) ECAR (mpH/min/Norm. Unit) 0 20 40 60 0 20 40 60 Time (minutes) Time (minutes)
✱✱ E ✱✱ 100 ✱✱
✱✱✱ ✱✱✱ 50 ✱✱✱
ns Cytotoxicity (%)
0
+Vehicle γ +DPI 10mMTotal Lysis +FK866 5uM+FK866 5nMγ IFN γ γ +FK866 0.5nM Media+Vehicle +FK866 500nMγ IFN γ IFN IFN Media+FK866IFN 5nM IFN 69
Figure 3.5: Figure 3.5: FK866 inhibits both OCR and ECAR in IFNγ stimulated monocytes. (A-B) Real time changes in OCR measured by Seahorse following treatment for 24 hours prior to the start of the assay with media alone or IFNγ with (A) exogenous NAM or (B) FK866 at indicated doses. (C-D) Real time changes in ECAR measured by Seahorse following treatment for 24 hours prior to the start of the assay with media alone or IFNγ with or without FK866 (C) or DPI (D). Cytotoxicity measured by LDH in supernatant from experiments in (B) and (D). Data in (D) were analyzed by one-way ANOVA with Dunnett’s multiple comparisons test. Error bars are mean ± s.e.m **p<0.01, ***p<0.001 and ****p<0.0001.
STAT1 is required for IFNγ-induced oxygen consumption and regulates oxygen consumption via mitochondrial complex I
To further elucidate the signaling pathways that mediate IFNγ-induced increases in OCR and the physiologic relevance of our assay, we assessed 2 patients with STAT1 GOF mutations
(Table 3.1) and observed higher basal OCR (p=0.192) and significantly higher PMA-stimulated
OCR (p=0.027) following IFNγ stimulation. Basal OCR levels in STAT1 GOF patient monocytes without IFNγ stimulation were even higher than basal OCR levels in healthy control monocytes stimulated with IFNγ (Figure 3.6A). We also observed that expression of genes involved in NAD+ salvage (NAMPT, P2RX7 and CD38) were elevated in monocytes from a
STAT1 GOF patient, and subsequently reduced to normal or below normal levels, after the patient was treated with the JAK inhibitor, ruxolitinib (Figure 3.7A-C). These data show that
STAT1 mediates IFNγ-induced increases in OCR and likely acts through modulation of NAD+ salvage gene expression.
The primary connection between NAD+ metabolism and cellular oxygen consumption is mediated by mitochondrial complex I (NADH ubiquinone oxidoreductase), where NADH is oxidized, regenerating NAD+ for use in vital cellular reactions and initiating mitochondrial electron transport. Mitochondrial oxygen consumption occurs as a product of electron transport by two mechanisms: 1) production of water as molecular oxygen acts as the final electron acceptor and 2) electron leak, primarily from complexes I and III, which can generate 70 superoxide[216-223]. Treatment with the chemical inhibitor of mitochondrial complex I, rotenone, throughout the 24-hour IFNγ stimulation period abrogated IFNγ-induced increase in
OCR (Figure 3.6B).
In addition to chemical inhibition of mitochondrial complex I, we also assessed primary monocytes from a patient with a clinical diagnosis of Leigh Syndrome, caused by mitochondrial
MT-ND1 (m.3697 G>A; Table 1) resulting in reduced function of mitochondrial complex I[224-
227]. Because this gene is maternally inherited, we assessed the proband’s father as an unaffected healthy control. One of the hallmarks of mutations in mitochondrial DNA (mtDNA) is heteroplasmy, in which an individual may possess both wild type and mutant mtDNA in the same cell[228], the balance of which determines mitochondrial function and cellular health. The proband was homoplasmic (100%) for m.3697 G>A, whereas her sibling was 80% heteroplasmic. As expected, the proband had significantly lower basal and PMA-stimulated OCR than her unaffected father (Figure 3.6C), while her sibling had an intermediate level of OCR
(Figure 3.7D-F). Therefore, both chemical and genetic disruption of mitochondrial complex I inhibits IFNγ-induced OCR.
The precise mechanisms by which IFNγ-induced NAD(H) metabolism and resultant increases in OCR affect monocyte activation and immunologic functions remain unclear.
Preliminary data comparing gene expression in a STAT1 GOF patient (Table 3.1) to 3 healthy controls subjects resulted in differential regulation of IFNγ-induced transcripts in the patient compared to healthy controls. When STAT1 GOF patient monocytes were stimulated with IFNγ for 24 hours, expression of PD-L1 and CD40 transcripts were increased compared to healthy controls. However, when stimulated with IFNγ in the presence of rotenone, there was no difference in PD-L1 or CD40 between the STAT1 GOF patient and healthy controls. 71
Interestingly, rotenone treatment throughout the IFNγ stimulation had no effect on gene expression in healthy control monocytes (Figure 3.8A-B). Therefore, chemical inhibition of
10 Medium IFNγ mitochondrialPMA complex I with rotenone normalized the high levels of IFNγ induced gene Medium + Ruxolitinib 5 IFNγ + Ruxolitinib expression in a STAT1 GOF patient.
FigureOCR (pmol/min/Norm. Unit) 0 3.6: 0 20 40 60 Time (minutes)
A STAT1 GOF Patients STAT1 GOF Patients ✱ ✱ ✱✱✱✱ ✱✱✱✱ Healthy Control + Medium 8 6 ✱✱✱✱ ✱✱✱✱✱ Healthy Control Healthy Control + IFNγ 8 HealthyUnstimulated Control 6 ✱✱✱✱ ✱✱✱✱ PMA STAT1 GOF + Medium 8 HealthyUnstimulated Control +IFNγ 4 6 Healthy Control STAT1Healthy GOF Control Unstimulated +IFNγ STAT1 GOF + IFNγ 4 Unstimulated 6 STAT1 GOF Unstimulated+ IFNγ 4 Healthy Control +IFNγ 2 2 STAT1 GOF Unstimulated+ IFNγ
Fold Change OCR 4 2 STAT1 GOF + IFNγ Fold Change OCR
OCR (pmol/min/Norm. Unit) 0 0 2 0 20 40 60 Basal PMA-stimulated Fold Change OCR 0 Basal PMA-stimulated Time (minutes) Basal PMA-stimulated 0 Basal PMA-stimulated✱✱✱✱ Rotenone Rotenone B ✱✱✱✱ 8 ✱✱✱✱ ✱✱✱✱ 10 ✱✱✱✱ Unstimulated Medium 8 ✱✱✱✱ ✱✱✱✱ UnstimulatedIFNg 6 ✱✱✱✱ ✱✱✱✱ ✱✱✱✱✱✱✱ PMA IFNγ 8 UnstimulatedIFNgRotenone Unstimulated Medium + Rotenone 6 ✱✱✱✱ ✱✱✱ 4 ✱✱✱ IFNgRotenone Unstimulated+ IFNg 5 IFNγ + Rotenone 6 ✱✱✱✱ ✱✱✱ 4 ✱✱✱ Rotenone Unstimulated+ IFNg 2 ✱✱✱ Fold Change OCR 4 Rotenone + IFNg 2
Fold Change OCR 0 2 Fold Change OCR OCR (pmol/min/Norm. Unit) 0 0 0 20 40 60 0 Basal PMA-stimulated Time (minutes) ✱✱✱✱
8 ✱✱✱✱ ✱✱✱✱ C Leigh Syndrome (ND1) Patient Leigh Syndrome (ND1)✱ Patient Healthy Control 8 ✱✱✱✱✱✱✱✱ Unstimulated 10 ✱✱✱✱ ✱ Healthy Control 6 ✱✱✱✱ 8 UnstimulatedHealthy Control +IFNγ ✱✱✱✱ ✱✱✱✱ 6 ✱ ✱✱✱✱ Healthy Control Healthy Father + Medium ✱✱✱✱ HealthyND1 Unstimulated Control +IFNγ 4 Unstimulated ✱✱✱✱ PMA Healthy Father + IFNγ 6 ✱✱✱✱ ✱✱✱✱ ND1 UnstimulatedIFNg 4 Healthy Control +IFNγ ND1 IFNg 5 ND1 Patient + Medium 2 ✱✱✱✱ ND1 Unstimulated Fold Change OCR 4 ND1 Patient + IFNγ 2 ND1 IFNg Fold Change OCR 0 2
Fold Change OCR 0
OCR (pmol/min/Norm. Unit) 0 0 0 20 40 60 Time (minutes) Basal PMA-stimulated
Figure 3.6: Figure 3: STAT1 is required for IFNγ induced oxygen consumption and regulates oxygen consumption via mitochondrial complex I. (A) OCR measured by Seahorse in primary human monocytes treated with media alone or IFNγ for 24 hours prior to the start of the assay (n = 5 technical replicates). (A) primary monocytes from 2 healthy controls and 2 STAT1 GOF patients are represented in the composite tracing from two independent experiments with values normalized to the unstimulated healthy control samples run in the same experiment. (B) primary monocytes from a healthy control were stimulated with media alone or IFNγ with or without Rotenone (10µM) treatment throughout stimulation. Data are representative of three independent of experiments and summarized in the bar graphs with basal and PMA-stimulated OCR values normalized to the healthy control unstimulated samples run in 72
the same experiment. (C) primary monocytes from 1 Leigh syndrome (MT-ND1) patient and the patient’s unaffected father as a control. Data represent OCR values from a single patient in a single experiment. Basal and PMA-stimulated OCR values from respective experiments are summarized in bar graphs and data were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. Error bars are mean ± s.e.m *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001.
Figure 3.7:
STAT1 NAMPT NAMPT NAMPTCD38 CD38 P2RX7CD38 P2RX7 P2RX7 A B 40 C 8 8 8008 800 20800 Healthy20 Controls 20 Healthy Controls Healthy Controls HealthyHealthy Controls Controls Healthy Controls HealthyHealthy Controls Controls Healthy Controls Healthy Controls STAT1 GOF Patient (Pre-Ruxolitinib) STAT1 GOF Patient (Pre-Ruxo) STAT1 GOF Patient (Pre-Ruxo) STAT1STAT1 GOF GOF Patient Patient (Pre-Ruxo) (Pre-Ruxo) 30 STAT1 GOF Patient (Pre-Ruxo) STAT1STAT1 GOF GOF Patient Patient (Pre-Ruxo) (Pre-Ruxo) STAT1 GOF Patient (Pre-Ruxo) STAT1 GOF Patient (Pre-Ruxo) 6 6 6006 600 15600 15 15 STAT1 GOF Patient (Post-Ruxo) STAT1 GOF Patient (Post-Ruxo) STAT1STAT1 GOF GOF Patient Patient (Post-Ruxo) (Post-Ruxo) STAT1 GOF Patient (Post-Ruxo) STAT1STAT1STAT1 GOF GOF GOF Patient Patient Patient (Post-Ruxo)(Post-Ruxolitinib) (Post-Ruxo) STAT1 GOF Patient (Post-Ruxo) STAT1 GOF Patient (Post-Ruxo) 4 4 4004 400 20 10400 10 10 RQ CD38 RQ CD38 RQ CD38 RQ P2RX7 RQ P2RX7 RQ P2RX7 RQ STAT1 RQ NAMPT RQ NAMPT RQ NAMPT 2 2 2002 200 10 5200 5 5
0 0 00 0 0 0 0 0 Unstimulated IFNγ Unstimulated IFNγ UnstimulatedUnstimulated IFNIFNγγ Unstimulated IFNγ 0 UnstimulatedUnstimulated IFNIFNγ γ Unstimulated IFNγ Unstimulated IFNγ Unstimulated IFNγ
D Leigh Syndrome (ND1) Patient
6 Father + Medium PMA Father + IFNγ
4 Proband + Medium Proband + IFNγ Sibling + Medium 2 Sibling + IFNγ
OCR (pmol/min/Norm. Unit) 0 0 20 40 60 Time (minutes)
E + Medium F + IFN! 8 8 ✱✱✱ Father (Unaffected) Father (Unaffected) ✱ ✱✱✱✱ 6 Proband (100% Homoplasmy) 6 Proband (100% Homoplasmy) ✱
✱✱✱✱ Sibling (80% Heteroplasmy) Sibling (80% Heteroplasmy) 4 4
2 2 OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0 Basal PMA-stimulated Basal PMA-stimulated
Figure 3.7: IFNγ induced increases in OCR are pSTAT1, NAMPT and Complex I dependent. (A-C) Gene expression measured by qPCR in primary monocytes from 3 healthy controls, and a STAT1 GOF patient before and after initiating treatment with Ruxolitinib. (D-F) Primary monocytes from a homoplasmic Leigh syndrome patient (proband), a heteroplasmic Leigh syndrome patient (sibling) and their unaffected father were stimulated with media alone or IFNγ for 24 hours and OCR was measured by Seahorse. (E-F) Basal and PMA-stimulated OCR values from (D) are summarized in bar graphs and data were analyzed by two-way ANOVA with Sidak’s multiple comparisons test. Error bars are mean ± s.e.m *p<0.05, ***p<0.001 and ****p<0.0001.
Figure 3.8: 73
8 80 15 A STAT1 GOF Patient B STAT1 GOF Patient STAT1 GOF Patient Healthy ControlsPD-L1 CD40 Healthy Controls Healthy Controls 6 8080 6088 10 6060 66 4 40
4040 44 RQ CD40 RQ PD-L1 5 RQ HLA-DR RQ CD40 RQ CD40 RQ PD-L1 2 RQ PD-L1 20 2020 22
0 00 000 0
VehicleVehicle VehicleVehicle Vehicle RotenoneRotenone Vehicle RotenoneRotenone Vehicle Rotenone IFNg+VehicleIFNg+Vehicle Rotenone IFNg+VehicleIFNg+Vehicle Rotenone IFNg+RotenoneIFNg+Rotenone IFNg+RotenoneIFNg+Rotenone IFNg+Vehicle PAM+Vehicle IFNg+Vehicle PAM+Vehicle IFNg+Vehicle PAM+Vehicle IFNg+Rotenone PAM+Rotenone IFNg+Rotenone PAM+Rotenone IFNg+Rotenone PAM+Rotenone Figure 3.8: Chemical inhibition of mitochondrial complex I normalizes elevated IFNγ- induced PD-L1 and CD40 expression. Gene expression of (A) PD-L1 and (B) CD40 measured by qPCR in primary monocytes from 3 healthy controls, and a STAT1 GOF patient treated with IFNγ (1000U/mL) in the presence or absence of rotenone (10uM) for 24 hours.
PMA-stimulated Oxygen Consumption Rates measure NADPH oxidase activity
Besides oxidation by mitochondrial complex I, NAD(H) can also be converted into
NADP(H) by cytosolic and mitochondrial NAD+ kinases (NADKs)[214, 229] or the
mitochondrial enzyme nicotinamide nucleotide transhydrogenase (NNT), which transfers a
hydride ion from NADH to NADP+ [230, 231]. Oxidation of NADPH by the NADPH oxidase is
the primary source of superoxide production during the respiratory burst, required for effective
clearance of some pathogens by phagocytes. Patients with chronic granulomatous disease (CGD)
have genetic defects in one of 5 NADPH oxidase components, and characteristically present with
recurrent infections and immune dysregulation including granulomatous inflammation[232-234].
Therefore, we assessed 4 patients with loss-of-function mutations in various NADPH oxidase
genes (Table 3.1). Interestingly, when compared to healthy controls, CGD patient cells
demonstrated no increase in OCR in response to PMA stimulation. Although basal OCR levels
(both with and without IFNγ stimulation) were similar to those of healthy controls, there was no
PMA-stimulated increase in OCR (Figure 3.9A), suggesting that the PMA-stimulated increases 74 in OCR measured by our assay represent superoxide production by the NADPH oxidase complex.
We also tested a chemical inhibitor of NADPH oxidase, diphenyleneiodonium
(DPI)[235]. Like CGD patient monocytes, healthy control monocytes treated with DPI showed no increase in OCR in response to PMA. DPI treatment also reduced basal OCR levels (Figure
3.9B), which is consistent with reports of off target effects of DPI, including inhibition of mitochondrial complex I which has a structure similar to NADPH oxidase[236-238].
Recognizing that the PMA-stimulated OCR in our assay represents NADPH oxidase-dependent oxygen consumption further clarifies the mechanism by which IFNγ acts to augment the monocyte respiratory burst[69, 72-74, 239, 240]. It also demonstrates that the reduction in PMA- stimulated OCR observed with NAMPT inhibition causes a diminished IFNγ-induced respiratory burst, comparable to that of CGD (Figure 3.9C).
10 Healthy Controls + Medium
Healthy Controls +IFNγ
PMA CGD Patients + Medium 5 CGD Patients + IFNγ
OCR (pmol/min/Norm. Unit) 0 0 20 40 60 Time (minutes) 75
Figure 3.9:
A Chronic Granulomatous Disease (CGD) Patients B Diphenyleneiodonium (DPI)
10 Healthy Controls + Medium 10 Medium
IFNγ Healthy Controls +IFNγ PMA
PMA CGD Patients + Medium DPI + Medium 5 5 CGD Patients + IFNγ DPI +IFNγ OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0 0 20 40 60 0 20 40 60 Time (minutes) Time (minutes)
C PMA-stimulated OCR **** 8 **** ****
6
10 4 Medium IFNγ PMA 2
Fold Change OCR DPI + Medium 5 0 DPI +IFNγ γ γ
+ FK866 γ OCR (pmol/min/Norm. Unit) 0 0 20 40 60 Time (minutes) CGD Patients + IFN Healthy Controls + IFN Healthy Controls + Medium Healthy Controls + IFN
Figure 3.9: PMA-stimulated Oxygen Consumption Rates Measure NADPH Oxidase Activity. (A-B) OCR was measured by Seahorse in primary human monocytes treated with media alone or IFNγ for 24 hours prior to the start of the assay (n = 5 technical replicates). (A) Monocytes from 4 healthy controls (black and green lines) and 4 CGD patients (red lines) were compared. (B) Monocytes were treated with (red lines) or without (black and green lines) DPI (100uM) in addition to IFNγ or media alone for 24 hours. (C) Bar graphs displaying PMA- stimulated OCR values normalized to the average of basal OCR in healthy control + medium samples run in the same experiment. Data are representative of at least three independent of experiments and were analyzed by one-way ANOVA with Tukey’s multiple comparisons test. Error bars are mean ± s.e.m ****p<0.0001.
IFNγ coordinately regulates transcription of multiple pathways to promote both NAM- dependent and independent NAD salvage
In addition to NAMPT, we also found that IFNγ transcriptionally regulated several other genes that promote NAD(H) import and reduction/oxidation. IFNγ induced P2RX7, which forms an ATP-gated pore that can import extracellular NADH, and CD38, an NAD+ consuming ectoenzyme that converts extracellular NAD+ into NAM, the membrane permeable substrate for 76
NAMPT (Figure 3.10A-B). IFNγ also downregulated PDK4, an inhibitor of pyruvate dehydrogenase (PDH), which could allow for increased PDH activity, potentially increasing flux through the TCA cycle and NAD+ reduction (Figure 3.10B and Figure 3.12A). Furthermore,
NADPH oxidase complex genes were upregulated by IFNγ, while nuclear encoded mitochondrial complex I genes were mostly downregulated (Figure 3.11A-B). Genes involved in de novo and Preiss-Handler pathways of NAD+ biosynthesis were broadly downregulated (Figure
3.11C), as opposed to genes involved in NAD+ salvage.
Figure 3.10:
A (NAD+ Consuming Ectoenzyme): CD38 NAD+ NAM NR
IFNGR pSTAT1 NAMPT X FK866 NRK2 GOF
NMN IFN! PDK4 Mitochondrial Superoxide P2RX7 NMNAT Complex I +H2O NADH NAD+ X NADH Rotenone ND1 Patient NADK NNT B NAD(H) Salvage
NADK 1.5 CGD Patients P2RX7 CD38 1 DPI NAMPT 0.5 NADP+ X NADPH NMNAT3 0 NRK1 -0.5 Superoxide NNT -1 NADPH PDK4 -1.5 Oxidase NADK2 NMNAT1 NMNAT2 NRK2 Media 24h ND1 Media 24h ND2 Media 24h ND3 Media 24h ND4 IFN IFN IFN IFN
24h ND1 24h ND2 24h ND3 24h ND4
Figure 3.10: IFNγ coordinately regulates transcription of multiple pathways to promote both NAM-dependent and independent NAD salvage. (A) Diagram of pathways by which IFNγ regulates cellular NAD+ metabolism. IFNγ activates STAT1 and this augments CD38, NAMPT, and P2RX7, but inhibits PDK4 and NRK2. (B) Heat map generated from RNA-seq NAD(P)H Redox Pathways
LDHA 1.5 PDHA1 1 MDH1 PDHB 0.5 IDH3B 0 GAPDH -0.5 MDH2 -1 DLD -1.5 G6PD 77 GLUD1 ME2 DLAT GLUD2 analysis of human monocyte-derived macrophages stimulated with IFNγ forOGDH 24 hours displaying PGD Media 24h ND1 Media 24h ND2 Media 24h ND3 Media 24h ND4 IFN IFN IFN IFN
genes involved in NAD(H) salvage. ND, normal donor.
24h ND1 24h ND2 24h ND3 24h ND4 Figure 3.11: A NADPH Oxidase Complex C NAD+ Biosynthesis Pathways 2 CYBB NAPRT 2 IDH1 1.5 NADSYN1 1.5 NCF1 1 QPRT NCF2 1 0.5 ACMSD CYBA 0.5 AFMID NCF4 0 Media 24h ND1 Media 24h ND2 Media 24h ND3 Media 24h ND4 IFN IFN IFN IFN KMO 0
-0.5 Media 24h ND1 Media 24h ND2 Media 24h ND3 Media 24h ND4 IFN IFN IFN IFN
24h ND1 24h ND2 24h ND3 24h ND4 -0.5 -1
24h ND1 24h ND2 24h ND3 24h ND4 -1
B Mitochondrial Complex I
NDUFV2 2 NDUFA9 NDUFB3 1 NDUFB4 NDUFS1 NDUFS2 0 NDUFV3 NDUFC1 -1 NDUFA4 NDUFA5 -2 NDUFA10 NDUFB9 NDUFS6 NDUFA12 NDUFB1 NDUFS4 NDUFV1 NDUFA13 NDUFA3 NDUFS8 NDUFB2 NDUFB7 NDUFS7 NDUFB8 NDUFA11 NDUFA6 NDUFA2 NDUFB11 NDUFB5 NDUFAB1 NDUFC2 NDUFA7 NDUFA8 NDUFB6 NDUFS3 NDUFA1 NDUFB10 NDUFS5 Media 24h ND1 Media 24h ND2 Media 24h ND3 Media 24h ND4 IFN IFN IFN IFN
24h ND1 24h ND2 24h ND3 24h ND4
Figure 3.11: IFNγ regulates expression of genes involved in NAD(P)H oxidation and biosynthesis. Heat map generated from RNA-seq analysis of human monocyte-derived macrophages stimulated with IFNγ for 24 hours displaying genes involved in (A) NADPH oxidase complex, (B) Mitochondrial complex I (nuclear encoded genes), and (C) De novo and Preiss-Handler pathways of NAD+ biosynthesis. ND, normal donor.
We then used several inhibitors to determine which mitochondrial processes were required to maintain IFNγ-induced increases in OCR (Figure 3.12A). Compared to inhibitors that block both mitochondrial ATP synthesis and electron transport (Rotenone, Antimycin A and
Oligomycin),[219-221, 241, 242] treatment with the mitochondrial ionophore, trifluoromethoxy 78 carbonylcyanide phenylhydrazone (FCCP), which uncouples ATP synthesis from electron transport, did not affect IFNγ-induced increases in basal or PMA-stimulated OCR (Figure 3.12D vs. B-C)[243]. Therefore, IFNγ-induced OCR is not dependent on mitochondrial ATP production, but rather on the capacity to support electron transport. These findings suggest that
IFNγ coordinately regulates a transcriptional program that increases cellular NAD(P)H available for oxidation. The result of this oxidation likely serves to both produce ROS and to recycle
NAD(P)H between its oxidized and reduced states for activity as co-factors in vital cellular reactions.
Figure 3.12:
NADP+ IFN" A NADPH Jak1/2 DPI Oxidase X pSTAT1 NADP(H) Glycolysis
(m)NADK/NNT NAMPT Pyruvate NAD+ NMN X NAM PDK4 PDH NADH NMAT FK866 Acetyl-CoA
Outer Mitochondrial Membrane Oxaloacetate Citrate 2H+ 4H+ 2H+ 3H+ NADH Intermembrane NADH Space FCCP FCCP FCCP Malate !-KG CoQ Cyt C Complex I Complex III Complex IV Inner Mitochondrial X X X Membrane NADH Complex II Complex V Mitochondrial Fumarate Succinate X X X Matrix Antimycin A Oligomycin NAD+ H2O NADH ½ O2 FADH2 Rotenone 3H+ ATP + H2O ADP + Pi
B Rotenone/AA C Oligomycin D FCCP
10 10 10 Medium Medium Medium
PMA IFNγ PMA IFNγ PMA IFNγ Medium + Rotenone/AA Medium + Oligomycin Medium + FCCP
5 IFNγ + Rotenone/AA 5 IFNγ + Oligomycin 5 IFNγ + FCCP OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) OCR (pmol/min/Norm. Unit) 0 0 0 0 20 40 60 0 20 40 60 0 20 40 60 Time (minutes) Time (minutes) Time (minutes)
Figure 3.12: IFNγ-induced increases in OCR depend on electron transport, not ATP production. (A) Diagram highlighting the metabolic pathways connecting NAD+ metabolism to oxygen consumption, including the inhibition of PDH by PDK4 and chemical inhibitors of mitochondrial electron transport chain components. (B-D) OCR measured by Seahorse in primary human monocytes treated with media alone or IFNγ with or without (B) mitochondrial complex I inhibitor, rotenone, and mitochondrial complex III inhibitor, antimycin A (0.5 µM), (C) complex V inhibitor, oligomycin (1µM), or (D) mitochondrial inner membrane 79 protonophore, FCCP (0.5 µM), for 24 hours prior to the start of the assay. Basal OCR was measured then PMA (100ng/mL) was injected during the assay and OCR was monitored (n = 5 technical replicates). Data in (B-D) are representative tracings of three independent experiments.
Table 3.1: Patient Patient Gene Gain/Loss of Figure Patient Diagnosis Age (years) Sex (M/F) Affected Function 3A STAT1 GOF 13 F STAT1 GOF 3A STAT1 GOF 34 M STAT1 GOF 3C Leigh Syndrome 5 F MT-ND1 LOF S4A-C STAT1 GOF (+/-Ruxolitinib) 34 M STAT1 GOF S4D-F Leigh Syndrome 10 M MT-ND1 LOF 4A CGD 57 F p47phox LOF 4A CGD 24 M gp91 LOF 4A CGD 38 M gp91 LOF 4A CGD 14 M gp91 LOF
Table 3.1: Patients assayed for metabolic dysregulation. Patients’ clinical diagnosis, age, sex, affected gene and functional consequence of their genetic mutation listed as they appear in the corresponding figures.
80
Discussion:
Metabolic reprogramming of immune cells in response to activation is essential for their immunologic activity[133, 156, 207], and disruption of these metabolic pathways has the potential to alter immune cell function. We investigated the metabolic reprogramming associated with IFNγ stimulation of human monocytes and identified a primary role for IFNγ in the regulation of NAD+ metabolism. This metabolic phenotype is not observed with type I interferons and is dependent on NAMPT. Inhibition of NAMPT with a specific chemical inhibitor, FK866, completely abrogated the IFNγ-induced increases in basal and PMA-stimulated
OCR (Figure 2D-E).
We also demonstrate that IFNγ-induced increases in OCR are dependent on both STAT1 and mitochondrial complex I [244]. Patients with STAT1 GOF mutations or MT-ND1 Leigh syndrome had altered OCR compared to healthy controls (Figure 3A-C). PMA-stimulated OCR levels measured in our assay represent oxygen consumption by the NADPH oxidase complex and were completely absent in CGD patient and DPI treated monocytes (Figure 4A-B).
Therefore, we have identified a metabolic pathway by which IFNγ-induced NAMPT augments
NAD+ salvage which is required for complete induction of the NADPH oxidase mediated respiratory burst.
Our data provide a more detailed understanding of how IFNγ acts to activate monocytes in health and disease. Increasing microbicidal ROS production through an NAD(P)H-dependent mechanism may provide the added benefit of maintaining NAD(H) and NADP(H) availability intracellularly as cofactors for many cellular reactions, including the essential antioxidant processes involving glutathione and thioredoxin reductases [166, 245-247]. Therefore, the regulation of NAD+ metabolism by IFNγ may serve to augment ROS production and 81 simultaneously enhance the activity of antioxidant processes required to protect the host cell from oxidative stress[248]. Consistent with this hypothesis we found that mitochondrial ATP synthesis was dispensable for maintaining the IFNγ-induced respiratory burst (Supplemental
Figure 6D), suggesting that IFNγ-induced NAD+ salvage does not primarily function to supply
NADH for the purpose of ATP synthesis. Rather, increased NAD+ salvage likely supports mitochondrial ROS (mROS) production and synthesis/recycling of NAD(H) and NADP(H).
These act as cofactors for critical oxidative and anti-oxidant processes [207, 210, 214, 249].
In contrast to superoxide production by the NADPH oxidase complex, less is known about the specific immunologic effects of mROS. Sander and Garaude recently reviewed the multifaceted relationship between innate immune responses and mitochondria [220, 250] and
Kiritsy et al. provided evidence of crosstalk between IFNγ signaling and mitochondria, demonstrating that IFNγ-induced activation of antigen presenting cells (APC) is dependent on mitochondrial complex I activity [208, 222]. Other reports suggest that mitochondrial superoxide serves a similar microbicidal function as NADPH oxidase derived superoxide [251].
Interestingly, NADPH oxidase produced ROS has been shown to induce mROS and mROS can, in turn, activate NADPH oxidase [252].
The NADPH oxidase complex and mitochondrial complex I (NADH ubiquinone oxidoreductase) are not only related in their ability to oxidize NAD(P)H to produce ROS but also in that their substrates are almost identical derivatives of NAD+. We now understand that oxygen consumption by both oxidases is regulated by IFNγ-induced NAMPT activity in human monocytes. This bidirectional relationship is likely important for both maximizing ROS production during the respiratory burst and for effectively inducing protective antioxidant 82 programs through 1) recycling NAD(P)H between its oxidized and reduced states and 2) by activating redox sensitive mitochondrial antioxidants [252].
Expression of NAMPT is precisely regulated and highly inducible in response to inflammatory stimuli, including IFNγ, compared to the resting state [207]. Impaired upregulation of NAMPT, as in the setting of IFNγ signaling defects, could therefore affect bacterial killing during infection through its effects on superoxide production. Conversely, sustained or dysregulated expression of NAMPT could maintain/promote excessive inflammation causing collateral tissue damage and chronic cellular stress. High levels of superoxide production have been associated with chronic inflammatory diseases such as atherosclerosis and are less prevalent in CGD patients who do not produce superoxide [253, 254]. For this reason, NADPH oxidase inhibitors have been trialed as immunomodulators to protect against oxidative stress [255-259].
Our data suggest that modulation of NAMPT activity could have a similarly protective effect in states of chronic inflammation. By contrast, induction of NAMPT activity could be beneficial in states of IFNγ deficiency to augment the respiratory burst for microbicidal purposes [260].
IFNγ was first trialed as a therapeutic in CGD with the aim of augmenting the respiratory burst. It was ultimately approved for clinical use to reduce the frequency and severity of infections associated with CGD, but did not significantly increase superoxide production in CGD patients[75]. While the mechanism of action of IFNγ in CGD still remains elusive, our data provide a metabolic mechanism by which IFNγ enhances NADPH oxidase superoxide production in normal human monocytes. This may explain why this effect is not observed in
CGD patients – namely that IFNγ acts to augment NADPH upstream of the defect in CGD.
In addition to NAMPT, we identified several other IFNγ inducible genes involved in
NAD(H) salvage, some of which have important clinical implications. P2RX7 has been 83 associated with susceptibility to mycobacterial infection, an infection for which increased susceptibility is also associated with defects in IFNγ signaling. The mechanism of P2RX7 associated mycobacterial infection has not been completely elucidated, but our data suggest it could be related to its role in regulating the import of extracellular NADH [261, 262]. Similarly, drugs targeting CD38, such as daratumumab [263-265], may have additional immunomodulatory effects through altered monocyte NAD+ metabolism. Finally, the effects of IFNγ on PDH via downregulation of PDK4 suggest that metabolic diseases such as PDH deficiency likely have immunologic manifestations that are not yet completely recognized [224].
Collectively, our data identify a novel immunometabolic phenotype of IFNγ-stimulated primary human monocytes, characterized by significantly increased oxygen consumption. We confirmed that these metabolic changes are physiologically relevant, as they are dysregulated in specific patients and reproducible. Further investigation is required to determine the specific effects of NAMPT-mediated metabolism on IFNγ-induced immunologic functions. This understanding of the metabolic foundation of IFNγ-induced monocyte activation identifies several metabolic genes as potential causes or modifiers of immunologic and inflammatory diseases that should be evaluated as therapeutic targets for the modulation of IFNγ mediated monocyte activation.
84
CHAPTER IV: Discussion
IFNγ functions as a master regulator of monocyte and macrophage activation by coordinating a positive feedback loop of innate and adaptive immune cell crosstalk. Expression of IFNγ by T and NK cells is induced by IL-12, primarily secreted by TLR-stimulated macrophages, and IFNγ then acts to further augment the activation of macrophages perpetuating the loop. IFNγ stimulation also enhances multiple pathways involved in the control and killing of intracellular pathogens [266]. In addition to regulating many genes independently, IFNγ has both agonistic and antagonistic effects on signals from other cytokines, growth factors and pathogen associated molecular patterns (PAMPs). In doing so, IFNγ facilitates the integration multiple signals to coordinate an immune response that is specific to the precise context of a given microenvironment [267-269].
The fundamental role IFNγ signaling in the human immune response is difficult to summarize as it regulates many genes in a time and context specific manner. However, the study of patients with primary immunodeficiency syndromes has expanded our understanding of how certain immunologic signaling pathways function to promote effective immune responses and maintain/restore homeostasis in the context of inflammation [270]. Thorough characterization of clinical disease and genotype-phenotype correlations in patients with various defects in IFNγ signaling has revealed the predominant consequence of loss of IFNγ signaling as a susceptibility to intramacrophagic pathogens, in particular, low virulence non-tuberculous mycobacteria
(NTM) [20, 271]. While the clinical disease associated with IFNγ deficiency is well understood, questions remain about the molecular mechanisms by which IFNγ mediates host control of intracellular pathogens [77]. Efforts to characterize the IFNγ response by immunologic and 85 transcriptomic analyses have resulted in the identification of hundreds of genes differentially regulated by IFNγ, but only a few pathways of known immunologic significance have been identified [266]. Recent advances in the field of immunometabolism have revealed immunologically active metabolites (e.g. itaconate and succinate) and metabolic pathways (e.g. glycolysis and oxidative phosphorylation) that directly affect immunologic responses to various stimuli [272]. Therefore, we sought to investigate how IFNγ regulates the most fundamental cellular metabolic processes in human monocytes and how this metabolic reprogramming could influence the function IFNγ stimulated cells.
Our work first identified a role for IFNγ in the induction of the immunometabolite, itaconate. Itaconate is synergistically induced by IFNγ in combination with a TLR stimulus and contributes to the host defense against Mycobacterium avium through a mechanism that likely involves a combination of antimicrobial, anti-inflammatory and antioxidant activities. We also identified the metabolic phenotype of IFNγ stimulated human monocytes, which is characterized by high levels of oxygen consumption. This phenotype is consistent with the well documented function of IFNγ as an inducer of microbicidal reactive oxygen/nitrogen species (ROS/RNS) [59,
68, 70, 273, 274]. We determined that this oxidative phenotype is dependent on NAMPT- mediated NAD+ biosynthesis and required for complete induction of the IFNγ-induced respiratory burst.
Taken together, our findings identify two metabolic pathways regulated by IFNγ, with immunologic consequences that contribute to the IFNγ-mediated response to infection.
Interestingly, both of these pathways produce metabolites with antimicrobial properties and activate metabolic pathways with the potential to promote cytoprotective antioxidant activity in the host cell during the process of microbial killing. Therefore, the metabolic reprogramming 86 induced by IFNγ may function to augment its microbicidal effector functions through metabolic mechanisms that simultaneously mitigate collateral host tissue damage.
IRG1 and Itaconate Contribute to the Interferon Gamma-Induced Host Defense Against Mycobacterium avium
Summary of major results and conclusions:
Our investigation into the regulation and function of itaconate in the host response to
Mycobacterium avium aimed to address whether itaconate acts an essential effector of the immune response to M. avium. We found that IFNγ is required for the synergistic induction of
IRG1 expression and itaconate production, in combination with a TLR signal, in the normal human immune response to M. avium. We also found that patients with susceptibility to M. avium caused by genetic or acquired IFNγ signaling defects have low levels of IRG1 expression and itaconate production in response to stimulation with M. avium, suggesting that itaconate deficiency may represent an IFNγ-inducible effector that could be repleted in the context of IFNγ deficiency.
Therefore, we sought to determine the extent to which itaconate deficiency contributed to the pathology and infection susceptibility associated with the broader IFNγ deficiency. We found that Irg1-/- mice demonstrated impaired control of M. avium following in vivo infection, with higher mycobacterial loads in the liver and lungs than wild type mice. Although, compared to the dramatic susceptibility of Irg1-/- mice to Mtb infection, the phenotype we observed in the context of NTM infection was mild. Interestingly, the mechanism of Mtb susceptibility in Irg1-/- mice was attributed to the lack of immunomodulatory activity of itaconate, resulting in a neutrophil mediated immunopathology and ultimately, increased mortality [121]. However, in our NTM 87 infection model, we did not observe any significant immunopathology, likely due to differences in the immunogenicity of NTM and Mtb.
To further elucidate the mechanism of itaconate mediated control of M. avium in vivo, we employed an in vitro model of M. avium infection of bone marrow derived macrophages
(BMDM). Interestingly, unlike the difference in mycobacterial burdens in vivo, we did not observe a difference in colony forming units (CFU) following in vitro infection of BMDM, despite the fact that monocytes and macrophages are the primary producers of itaconate. While the cell intrinsic control of M. avium did not seem to be affected by the complete loss of itaconate, we also assessed wild type and Irg1-/- BMDM for differences in cellular functions associated with itaconate activity. Again, complete loss of itaconate did not result in any significant differences in viability or mitochondrial oxidative metabolism despite high levels of
Irg1 expression.
Overall, our findings confirmed an important role for IFNγ in the induction of IRG1 expression and itaconate production and demonstrate that disease causing defects in IFNγ signaling result in diminished IRG1 expression in response to M. avium. In a mouse model of disseminated M. avium infection, loss of itaconate resulted in impaired control of mycobacteria.
However, this impaired control of M. avium was not associated with severe disease or significant immunopathology as has been reported in the context of Mtb infection. Our investigation of the molecular mechanisms of itaconate in the context of M. avium infection did not reveal an obvious cell intrinsic effect of itaconate. Therefore, we can conclude that itaconate plays some role in the control of M. avium in vivo, and likely contributes to the infection susceptibility in patients with IFNγ signaling defects. However, the lack of significant differences in macrophage 88 function in vitro and lack of pathology in vivo in the absence of itaconate prevented us from dissecting its precise molecular mechanism and may suggest redundancy in its cellular functions.
The identification of low levels of IRG1 and itaconate in patients with mycobacterial infection susceptibility is the first report of a clinical phenotype that correlates with the infection susceptibility of Irg1-/- mice. We found that both genetic and autoantibody-mediated defects in
IFNγ signaling result in decreased IRG1 expression in patient PBMC’s in response to M. avium.
Similar to the phenotype of Irg1-/- mice, these patients experience impaired control of M. avium resulting in high mycobacterial burdens and immune dysregulation [121, 275]. These findings suggest that itaconate deficiency may represent a potential effector of the host response to mycobacterial infection that is demonstrably low in patients with IFNγ signaling defects.
Therefore, itaconate could potentially be repleted, to circumvent the defect in IFNγ and restore some capacity of the immune response to mycobacterial infection in these patients.
Beyond itaconate’s potential as a therapeutic, our results also demonstrate that IRG1 expression in patient PBMC’s in response to stimulation with M. avium is a useful biomarker for
IFNγ signaling activity, accounting for the amount of endogenous and/or exogenous IFNγ ligand, the signaling capacity of the receptor and the activity of downstream signaling components such as STAT1. Similarly, other studies have identified IRG1/itaconate as biomarkers of disease activity or severity in human subjects. Li et al. reported elevated IRG1 expression in PBMC’s from septic patients and proposed that IRG1 mediates sepsis-associated immunosuppression through ROS-dependent induction of A20 [276]. This proposed immunomodulatory activity of
IRG1 and itaconate is consistent with recent findings demonstrating that plasma itaconate levels inversely correlate with disease severity in inflammatory diseases such as rheumatoid arthritis and COVID-19 [277, 278]. Furthermore, the most common polymorphism in IRG1 reported in 89 gnomAD (https://gnomad.broadinstitute.org) results in a gain-of-function missense change that augments itaconate production and is overrepresented in African/African American populations, possibly because it confers some protection against mycobacterial or other infections [198].
Taken together, these findings support a role for itaconate in the human immune response and our data add to a growing body of literature investigating IRG1 and itaconate as important diagnostic and therapeutic tools for both infectious and inflammatory diseases.
To further elucidate the specific contribution of itaconate to the IFNγ-induced response to mycobacterial infection we infected wild type and Irg1-/- mice with M. avium and found that
Irg1-/- were impaired in their ability to control M. avium in vivo. Therefore, we concluded that itaconate does contribute to control of M. avium in vivo. However, its mechanism of action is still unclear. Unlike previously published data, we did not observe the significant inflammatory pathology that was increased in the Irg1-/- mice, despite similarly significant differences in
CFU’s. The absence of inflammatory pathology in M. avium infection compared to Mtb infection may represent an important finding consistent with other data suggesting that the primary function of itaconate in the context of infection is not directly antimicrobial, but rather anti- inflammatory [121, 192]. Compared to the early mortality and severe immunopathology reported by Nair et al. following Mtb infection in Irg1-/- mice, M. avium induces a much weaker inflammatory response and therefore, we observed very few differences between the wild type and Irg1-/- mice, other than an increase in M. avium CFU’s in the liver and lungs of Irg1-/- mice
[121].
Our in vitro investigation into the mechanism of action of itaconate in M. avium infection was even less revealing, with no differences in CFU’s, viability or oxidative metabolism. These findings support a primarily anti-inflammatory role for itaconate that is less apparent in the 90 context of infection with a pathogen that is less virulent and less inflammatory than Mtb. Jessop et al. also observed differential induction of Irg1 when comparing the vaccine strain of F. tularensis with the more virulent SchuS2 stain. However, in this case the more virulent strain is less inflammatory as a means of subverting host defense mechanisms, and therefore induced less
Irg1 expression [192]. Several mechanisms for the anti-inflammatory effects of itaconate have been proposed, but only a few have assessed the effects of endogenous itaconate. Hooftman et al. reported that Irg1-/- BMDM’s have increased NLRP3 inflammasome activation and went on to use exogenous 4-octyl-itaconate (4-O-I) to suggest that itaconate facilitates dicarboxypropylation of cysteine 548 on NLRP3 preventing its activation. However, the absence of this NLRP3 modification was not demonstrated in Irg1-/- BMDM’s [279, 280]. Bambouskova et al. then identified cysteine 77 of gasdermin D (GSDMD) as a post-translationally modified target of endogenous itaconate. They proposed that “itaconation” of GSDMD Cys77 inhibits caspase-1 activation and promotes tolerance to NLRP3 inflammasome activation, protecting cells from pyroptotic cell death [280].
The strongest evidence exists to support an anti-inflammatory and anti-oxidant role for itaconate through activation of Nrf2 [136, 141]. Independently of itaconate, the Nrf2-mediated antioxidant response has been shown to play an important role in the pathogenesis of mycobacterial infection by multiple mechanisms. First, Nrf2 deficient mice are highly susceptible to intranasal M. avium infection, and their susceptibility was attributed to low expression of Nramp1 and HO-1 [281]. Additionally, Gpx4 is another Nrf2 inducible gene that has been associated with susceptibility to mycobacteria through its ability to regulate ferroptosis, a necrotic cell death process. Amaral et al. demonstrated that pulmonary necrosis in Mtb infected mice is caused by ferroptotic cell death, characterized by glutathione depletion, reduced Gpx4 91 expression and increased lipid peroxidation [282]. Inhibition of ferroptosis with ferrostatin-1, resulted in reduced necrosis and bacterial loads in the lungs of Mtb infected mice [282].
Similarly, treatment of Mtb infected human macrophages with N-acetylcysteine (NAC) reduces oxidative stress and cell death caused by infection [283]. These findings illustrate a mechanism by which cytoprotective antioxidant activity facilitates control of infection by limiting necrotic cell death which results in the release and possible dissemination of intracellular bacteria and promotes a damaging inflammatory response [284]. Conversely, apoptotic cell death results in containment of bacteria within apoptotic bodies and induces an anti-inflammatory response through the process of efferocytosis [285]. Our findings of increased mycobacterial burdens in the Irg1-/- mice could represent insufficient induction of Nrf2-mediated antioxidant responses in the absence of itaconate.
Our in vivo infection model did not reveal significant differences in necrosis, likely because the amount of necrosis induced by M. avium is significantly less than that induced by
Mtb. Consistent with the observation of minimal pathology in M. avium infection, differences in markers of oxidative stress and pro-inflammatory cytokine production were not even detectable in Nrf2-deficient mice [281]. In vitro infection of wild type and Irg1-/- BMDM’s allowed us to focus on the Irg1 expressing and M. avium containing cells of interest. Interestingly, even in this enriched cell population, we did not detect any difference in viability between wild type and
Irg1-/- macrophages following infection. There were also no differences in mycobacterial loads in vitro. Finally, since IRG1 and itaconate are presumably concentrated within mitochondria and have been shown to exert inhibitory effects on SDH which could have dramatic effects on mitochondrial oxidative metabolism, we assessed mitochondrial oxidative metabolism by
Seahorse and found no difference in oxygen consumption between wild type and Irg1-/- 92 macrophages. Interestingly, both wild type and Irg1-/- macrophages stimulated with M. avium alone or IFNγ alone, demonstrated increased oxygen consumption compared to unstimulated macrophages or macrophages simultaneously stimulated with M. avium and IFNγ. These findings raise more questions about the metabolic mechanisms responsible for augmented oxygen consumption, which we explore in Chapter III to determine how these metabolic phenotypes translate into immunologic activities that allow for control of mycobacterial infection.
Additional experiments and future directions:
Our finding that IRG1 expression is low in patients with susceptibility to mycobacterial infection caused by mutations in IFNγ signaling pathways is suggestive but not conclusive evidence of role for IRG1 and itaconate in the control of NTM infections in patients. The ideal assessment would be comparison of patients with IFNγ signaling defects to patients with loss of function mutations in IRG1, with the expectation that IRG1/itaconate deficiency would confer some, but not all, of the infection susceptibility and immune dysregulation that is associated with
IFNγ deficiency. As described in Appendix I, we assessed one patient who presented with atypical infections and was found to have a variant of uncertain significance in IRG1. However, functional evaluation of the patient’s monocytes demonstrated that this particular variant did not affect maximal itaconate production and was not likely related to her clinical phenotype.
However, there are many reported variants in IRG1, some of which are predicted to cause a loss of function. Interestingly, most IRG1 variants reported in gnomAD, a healthy control database, do not exist as homozygotes and those variants that are reported as homozygotes are all predicted to have benign or possible gain of function effects on IRG1 activity [198]. This suggests that, while rare, heterozygous loss of function alleles can occur in healthy individuals, but 93 homozygous loss of function alleles are not detected in healthy individuals and would likely result in a disease phenotype. Therefore, one future direction for this work is to screen for loss of function variants in IRG1 that result in diminished itaconate production in patients with both inflammatory and infectious diseases to further characterize the clinical phenotype and inform the study of itaconate activity in the human immune response.
While we were not able to compare patients with IFNγ deficiency to patients with isolated IRG1/Itaconate deficiency we were able to make this comparison using an Irg1-/- and itaconate deficient mouse model. This study revealed that isolated IRG1 deficiency results in a similar phenotype of increased mycobacterial burdens in an in vivo model of disseminated M. avium infection, compared to published data from the same model in IFNγ knock out (GKO) mice [177]. However, one significant caveat to the study of macrophage function in mouse models is the predominant role of iNOS and nitric oxide (NO) in murine macrophages that is absent in human macrophages (at least in vitro) [59, 63]. Multiple studies have demonstrated that iNOS/NO can compensate and act redundantly or synergistically with IRG1/Itaconate in the context of infection and inflammation [197, 280] and other studies have identified NO as a critical mediator of murine macrophage reprogramming [286-288].
To avoid the confounding effects of NO in our mechanistic assessment of the activity of itaconate, we also assessed a human monocytic cell line (THP-1 cells). In vitro infection of wild type and IRG1 knock out THP-1 cells with M. avium did not reveal any significant differences in mycobacterial control or viability. However, we also found that the metabolic reprogramming of
THP-1 cells is significantly different than that of primary cells. One important difference was the lack of succinate accumulation in THP-1 cells, despite high levels of itaconate production.
Succinate accumulation as a result of itaconate mediated inhibition of succinate dehydrogenase 94
(SDH) has been linked to many of the downstream functions of itaconate, therefore we did not pursue in depth characterization of the phenotype in IRG1 knock out THP-1 cells following in vitro infection with M. avium.
With these caveats in mind, our findings suggest a role for IRG1 and itaconate as effectors of the IFNγ induced host response to M. avium infection. Further investigation into the precise mechanism of itaconate in the context of NTM infection is still required, and typical model systems present unique challenges for the study of metabolically mediated pathologies.
Future assessments could employ strategies such as siRNA to knock down IRG1 in primary human macrophages to avoid the pitfalls associated with altered metabolism in transformed cell lines and the confounding effects of NO in murine macrophages. Alternatively, clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR associated protein 9 (Cas9) mediated knock out of IRG1 in human induced pluripotent stem cells been used to assess metabolic disease of human macrophages and could be a useful to tool [289], but should first be evaluated to compare their metabolic response to stimulation to primary human macrophages.
Loss of the immunomodulatory activity of itaconate proposed by Nair et al. resulted in early mortality and severe immunopathology in Irg1-/- mice characterized by increased mycobacterial burdens and excessive neutrophil infiltration [121]. This mechanism likely involves both cell intrinsic and cell extrinsic activities of itaconate. For this reason, in vivo models, which offer the ability to characterize systemic and tissue level interactions between different cell types, may represent a more physiologically relevant model for assessment of itaconate activity. Future directions for in vivo assessment of itaconate activity could directly compare the response of Irg1-/- to M. avium infection with other genetic models of NTM susceptibility with a. focus on a detailed assessment of tissue pathology including specific 95 markers of oxidative stress (lipid peroxidation, oxidized DNA, and mitochondrial integrity), cell death (apoptotic vs. necrotic) and pro/anti-inflammatory cytokine production.
Finally, our studies have focused on the activity of endogenous itaconate as a means of understanding the processes involved in the normal immune response to M. avium. However, there is abundant evidence to suggest that exogenous forms of itaconate such as 4-octyl-itaconate
(4-O-I) and dimethyl itaconate (DI) have anti-oxidant and immunomodulatory effects. Whether these exogenous metabolites function through the same mechanism as endogenous itaconate is unclear, but their activity as potential therapeutics should be assessed, both as it relates to the physiologic activity of endogenous itaconate, and independently as potent activators of Nrf2 and other immunomodulatory pathways.
96
IFNγ Regulates NAD+ Metabolism in Human Monocytes
Summary of major results and conclusions:
In this chapter, we aimed to 1) define the metabolic phenotype of IFNγ stimulated human monocytes, 2) identify the molecular mechanisms by which IFNγ mediates monocyte metabolic reprogramming and 3) determine how the immunologic activity of IFNγ stimulated monocytes is regulated by metabolism. An understanding of the fundamental metabolic effects of IFNγ is important in the context of many of diseases in which IFNγ signaling is dysregulated. Functional
IFNγ deficiency impairs control of intracellular bacteria and causes systemic immune dysregulation [271, 275]. Conversely, states of excessive IFNγ activity, such as signal transducer and activator of transcription 1 (STAT1) gain of function (GOF), have inflammatory complications [271, 290-292]. Defining the metabolic pathways involved in IFNγ mediated monocyte activation could reveal novel targets for therapeutic modulation of monocyte activation in states of dysregulated IFNγ signaling.
The majority of work in the field of macrophage immunometabolism has focused on early metabolic changes in murine macrophages in response to M1 stimuli (LPS or LPS+IFNγ) which induce a switch to aerobic glycolysis, known as the Warburg effect [157, 204-206].
However, our preliminary data from broad profiling metabolomics of primary human monocytes revealed that significant metabolic changes occur at later time points, 18-24 hours post stimulation. Therefore, we assessed both mitochondrial oxidative phosphorylation and glycolysis by monitoring the real time oxygen consumption rate (OCR) and extracellular acidification rate
(ECAR) respectively, in primary human monocytes following stimulation with LPS, IFNγ or the combination of LPS+IFNγ for 24 hours. Interestingly, following 24 hours of stimulation, 97 glycolysis was not significantly induced in LPS or LPS+IFNγ-stimulated monocytes. The most striking difference in metabolism was the significant increase in OCR in IFNγ stimulated monocytes. We then used a modified Seahorse extracellular flux assay to monitor basal OCR
(OCR measurement after 24 hours of IFNγ stimulation) and PMA-stimulated OCR (OCR measurements following intra-assay injection of PMA onto cell which have already been primed with IFNγ for 24 hours).
As described in Chapter III, we found that IFNγ significantly increased both basal and
PMA-stimulated OCR, while type I interferons induced only a modest increase in OCR, suggesting that this metabolic reprogramming is specific to the function of IFNγ stimulated monocytes. After ruling out several metabolic pathways that could support increased oxygen consumption, we found that IFNγ-induced increases in OCR are dependent on NAMPT- mediated NAD+ salvage, and inhibition of NAMPT with a specific chemical inhibitor, FK866, completely abrogated the IFNγ-induced increases in OCR. We determined that this metabolic phenotype is dependent on both on both STAT1 and mitochondrial complex I and is altered in patients with STAT1 GOF mutations or MT-ND1 Leigh Syndrome. Assessment of monocytes from patients with chronic granulomatous disease (CGD), caused by loss of function mutations in NADPH oxidase components, revealed that PMA-stimulated OCR levels measured in our assay represent oxygen consumption by the NADPH oxidase complex. PMA-stimulated increases in OCR were completely absent in CGD patient and monocytes treated with the
NADPH oxidase inhibitor diphenyleneiodonium (DPI). Furthermore, RNA sequencing revealed several other IFNγ regulated genes involved in NAD(H) biosynthesis, salvage, import and superoxide production, that could contribute to the oxidative metabolic phenotype induced by
IFNγ. Taken together, our work identifies a metabolic pathway by which IFNγ regulates 98 metabolically active genes, including NAMPT, to augment NAD+ salvage which is required for complete induction of the respiratory burst in monocytes.
We found that the metabolic phenotype induced by IFNγ stimulation is different than the metabolic phenotype induced by LPS or the combination of LPS+IFNγ. Additionally, it does not fit into the canonical paradigm of glycolytic, M1 metabolism or oxidative, M2 metabolism [32,
147, 150, 151, 188, 293-295]. In this paradigm, oxygen consumption in M2 macrophages is typically associated with increased fatty acid oxidation driving mitochondrial respiration [212].
However, our findings suggest that high levels of oxygen consumption can also represent the conversion of molecular oxygen into reactive oxygen species (ROS), a process that is known to be induced by IFNγ [69, 72, 191]. Wang et al. performed one of the only studies to assess metabolic reprogramming of macrophages induced by IFNγ alone [295]. Interestingly, they found that stimulation of murine bone marrow derived macrophages with IFNγ for 24 hours results in augmented glycolysis and decreased oxygen consumption, whereas we observed only a moderate increase in glycolysis and a dramatic increase in oxygen consumption in human macrophages. These discrepant findings may represent differences in the metabolic reprogramming of human vs. mouse macrophages. However, consistent with our findings in human monocytes, they also concluded that IFNγ stimulation repurposes mitochondria for ROS production rather than ATP production [295].
Kiritsy et al. recently reported that IFNγ-induced, STAT1-dependent regulation of mitochondrial complex I activity is required for effective antigen presenting cell function and T cell activation. They demonstrated that genetic or chemical inhibition of mitochondrial complex
I in macrophages or dendritic cells, significantly reduced expression of co-stimulatory molecules
(PD-L1, CD40, MHCII) and impaired their ability to activate T cells [222]. Similarly, we found 99 that expression of PD-L1 and CD40 were elevated in response to IFNγ stimulation in a STAT1
GOF patient’s monocytes compared to monocytes from healthy controls. Treatment of monocytes with the mitochondrial complex I inhibitor, rotenone, throughout the 24 hour stimulation with IFNγ normalized the expression of PD-L1 and CD40 in the STAT1 GOF patient’s monocytes to the levels observed in healthy controls, but did not affect expression in healthy control monocytes. These findings suggest that mitochondrial complex I inhibition has the capacity to normalized dysregulated IFNγ-induced gene expression in a STAT1 GOF patient.
While evidence defining the specific interactions between immunologic signaling pathways and mitochondria is still emerging, the concept of the immune-mitochondrial cross talk is widely accepted, and evolutionarily logical as eukaryotic mitochondria originated from endosymbiotic proteobacteria [296]. During eukaryotic evolution, what began as host-pathogen interactions likely evolved into a symbiotic relationship in which the pathogen’s response to host defense resulted in the mutually beneficial co-evolution of an energy producing organelle [297].
Recent evidence supports a reciprocal relationship in which mitochondria carry out important immunologic signaling and effector functions [158, 216, 220, 223, 298]. In response to activation by infection or inflammatory cytokines, macrophage mitochondria have been shown to directly activate the NLRP3 inflammasome [299], reorganize their electron transport chain to promote direct anti-microbial activity [220], generate signals to promote mitochondrial and cytosolic
(NADPH oxidase) ROS production and coordinate the redox homeostasis processes required to regulate these responses [192, 252, 300, 301].
Our novel methods for the simultaneous measurement of mitochondrial and NADPH oxidase induced oxygen consumption further elucidated the relationship between these two ROS producing processes [252, 302, 303]. As described in Chapter III, the NADPH oxidase complex 100 and mitochondrial complex I (NADH ubiquinone oxidoreductase) are not only related in their ability to oxidize NAD(P)H to produce ROS but also in that their substrates are structurally similar derivatives of NAD+. We now understand that oxygen consumption by both oxidases is regulated by IFNγ-induced NAMPT activity in human monocytes. This bidirectional relationship is likely important for both maximizing ROS production during the respiratory burst and for effectively inducing protective antioxidant programs through 1) recycling NAD(P)H between its oxidized and reduced states and 2) by activating redox sensitive mitochondrial antioxidants such as aconitase, manganese superoxide dismutase and NAD(P)H-dependent glutathione peroxidase
[223, 245, 301, 304].
Importantly, we identified IFNγ-induced NAMPT as the critical regulator of these processes. We demonstrated that NAMPT-mediated NAD+ salvage is required for both the basal and PMA-stimulated, IFNγ-induced increases in oxygen consumption. Therefore, the NADPH- oxidase dependent respiratory burst is fueled by NAMPT-dependent NAD+ salvage. Inhibition of this pathway impairs the respiratory burst to the same extent as CGD-causing mutations in
NADPH oxidase. This finding indicates that: 1) IFNγ-induced NAMPT is a critical component of the metabolic mechanism by which IFNγ augments the respiratory burst [69, 72], 2) modulation of NAMPT activity could augment the respiratory burst in patients with immunodeficiency or inhibit damaging ROS production in patients with chronic inflammatory diseases, 3) this mechanism of IFNγ-induced superoxide production is upstream of the defect in CGD and may explain why IFNγ therapy did not restore superoxide production in CGD patients, despite other protective effects [75].
Finally, the induction of microbicidal oxidant production through a mechanism that generates NAD(H) and NADP(H) may represent an important regulatory process. The generation 101 of ROS for the purpose of killing pathogens also has deleterious effects on host cells [208, 305].
Many of the antioxidant pathways responsible for detoxifying these damaging metabolites are dependent on NAD(H) and/or NADP(H) as essential cofactors [166, 245, 246]. A wide variety of enzymes rely on NAD(H) and/or NADP(H) to facilitate their activity and in doing so, act as cellular redox sensors, modulating their activity based on the redox balance of these metabolites
[306, 307]. Therefore, IFNγ may act to promote redox homeostasis during infection by inducing microbicidal oxidants through the oxidization of the same redox-regulating metabolites that facilitate the detection of oxidative stress and promote anti-oxidant activity [223, 298, 308, 309].
Additional experiments and future directions:
Our findings have identified a metabolic mechanism regulating IFNγ-induced mitochondrial and NADPH oxidase produced ROS in human monocytes that is dependent on
NAMPT-mediated NAD+ salvage. Additional work is required to determine the precise effects of
NAMPT inhibition on monocyte activation. Our study focused on the functional effects of
NAMPT inhibition on oxygen consumption but did not directly assess the primary source or relative abundance of NAD(H) and NADP(H) in the various stimulation conditions and disease states. Metabolic tracing of isotopically labeled nutrients could help to determine the source(s) of cellular NAD+ and its precursors. Additionally, our data do not directly address the extent to which cellular antioxidant activity is dependent on NAMPT-mediated NAD+ salvage.
Experiments comparing the redox state of IFNγ stimulated monocytes with and without NAMPT inhibition would be helpful in addressing this question.
We also identified an interesting discrepancy between the metabolic reprogramming of monocytes treated with IFNγ alone vs. IFNγ in combination with LPS. Our data suggests that the presence of LPS predominantly dictates the metabolic response over IFNγ-induced metabolic 102 effects. We hypothesize that LPS-induced glycolysis may divert pyruvate into lactate rather than the TCA cycle where it can be oxidized to regenerate NADH. Again, labeled glucose tracing experiments will help to determine how LPS signaling alters the metabolic response to IFNγ which has important implications in the context of infection.
We also demonstrated the physiological relevance of this pathway by evaluating patients with monogenic defects in various components of this pathway. However, because these monogenic diseases are rare, our ability to fully characterize their metabolic phenotype at baseline and in response to interventions was limited. We hope to further characterize the metabolic defects in a variety of immunologic diseases related to this pathway and use this understanding to identify metabolic targets for modulation of the immune response.
First, additional investigation into the effects of mitochondrial complex I inhibitors
(including the FDA-approved, metformin [310]) in STAT1 GOF patients may reveal a novel method to normalize aberrant gene expression, but requires comprehensive transcriptome analysis. In addition to STAT1 GOF, patients with dominant negative STAT3 mutations have similarly elevated levels of phosphorylated STAT1 and may benefit from similar interventions.
Excessive IFNγ is also implicated in the pathogenesis of several monogenic inflammatory diseases including Autoimmune Polyendocrinopathy Candidiasis Ectodermal Dystrophy
(APECED) [111], Hemophagocytic LymphoHistiocytosis (HLH) [200, 202], and Pyogenic
Arthritis, Pyoderma gangrenosum, and Acne (PAPA) syndrome [311, 312]. Our findings also suggest that excessive IFNγ signaling could cause sustained or dysregulated expression of
NAMPT, promoting superoxide production. High levels of superoxide production have been associated with chronic inflammatory diseases such as atherosclerosis, which are less prevalent in CGD patients who do not produce superoxide [253, 254]. For this reason, NADPH oxidase 103 inhibitors have been trialed as immunomodulators to protect against oxidative stress [255-259].
We plan to assess whether NAMPT inhibition in these patients could have a protective effect in limiting IFNγ-mediated macrophage activation. Our finding that the NAMPT inhibitor, FK866, has a preferential inhibitory effect on oxygen consumption in IFNγ stimulated monocytes, but does not affect oxygen consumption in unstimulated monocytes, makes it an appealing candidate for the treatment of IFNγ-mediated inflammatory diseases.
Second, the use of small molecules to augment NAMPT activity could potentially restore an important component of the IFNγ-induced response to infection in patients with infection susceptibility caused by defects in IFNγ signaling [260]. Similarly, we also plan to investigate some of the other NAD(H) metabolism related genes induced by IFNγ to determine if they contribute to the IFNγ-induced metabolic response to stimulation. Of particular interest is
P2RX7, a purinergic receptor which forms an ATP-gated pore that can import extracellular
NADH into the cytoplasm [261, 262]. Interestingly, variants in P2RX7 have been associated with susceptibility to mycobacterial infection, but the mechanism of P2RX7 associated mycobacterial susceptibility has not been completely elucidated [262, 313, 314]. We plan to assess its role in regulating the import of extracellular NADH and its impact on IFNγ-induced oxygen consumption using a chemical inhibitor of P2RX7 and evaluating monocytes from a patient with a P2RX7 variant and disseminated mycobacterial disease.
Finally, we found that IFNγ also regulates CD38, an NAD+ consuming ectoenzyme that converts extracellular NAD+ into NAM, the membrane permeable substrate for NAMPT and
PDK4, an inhibitor of pyruvate dehydrogenase (PDH) regulating flux of pyruvate into the TCA cycle. Additional experiments are required to determine the extent to which activity of these genes contributes to the IFNγ induced metabolic phenotype. For example, CD38 is the target of 104 daratumumab, an anti-neoplastic monoclonal antibody used for the treatment of multiple myeloma. Inhibition of this enzyme may have additional immunomodulatory effects on IFNγ activated monocytes by limiting NAM availability [263-265]. Additionally. IFNγ-induced downregulation of PDK4, could augment flux of pyruvate through the TCA cycle resulting in increased NADH regeneration. Isotopically labeled pyruvate tracing studies could confirm the effects of IFNγ on flux through PDH, and the PDH inhibitor, dichloroacetate, could be used to mimic the effects of PDK4 [315]. The effects of IFNγ on genes regulating pyruvate metabolism such as PDK4 also suggest that metabolic diseases such as PDH deficiency likely have immunologic manifestations that are not yet completely recognized [224]. We hope to further characterize the immune dysregulation in patients with a variety of inborn errors in metabolism, with a focus on mitochondrial diseases. Additionally, defects in the metabolic pathways discussed here should be considered in the evaluation of patients with inflammatory disease or infection susceptibility of unknown etiology.
105
CHAPTER V: Materials and Methods
Ethics Statement:
All patients and healthy volunteers signed informed consent for IRB approved NIH protocols.
All animal studies were conducted in Laboratory Animal Care accredited Biosafety Level 2 facilities at NIAID/National Institutes of Health (NIH), under approved protocols.
Mice:
C57BL/6N (WT) and Irg1-/- mice were generated at Washington University and have been previously described by Lampropoulou et al. [117]. Mice were housed under specific pathogen free conditions with ad libitum access to food and water. C57BL/6N WT mice from
Charles River Laboratories were used as age-matched controls.
In vivo infections:
WT or Irg1-/- mice (5-8 weeks old) were infected intravenously via tail vein injection with
M. avium (strain 2-151 SmT) (1.5x10^6 CFU per mouse). Both male and female mice were assessed with 5 mice per genotype per time point. Spleens, livers and lungs were harvested at 4 and 8 weeks post infection. The same sections of each organ were harvested from all mice for
CFU quantification, preparation for histology, and RNA, protein and metabolite extraction, respectively. Tissue for M. avium CFU quantification was homogenized in PBS then plated as described below. Tissue for histology was fixed in 10% formalin before embedding in paraffin wax for H&E and MPO staining. Tissue for RNA or protein was transferred immediately into
Trizol (ThermoFisher catalog# 15596-026) or Cell Lysis Buffer (Cell Signaling Technology catalog# 9803), respectively. 106
Harvest and differentiation of murine bone marrow derived macrophages:
Mice were euthanized in a chamber with CO2 gas at 4 and 8 weeks post infection. Bone marrow was harvested from WT and Irg1-/- mice by flushing both femora and tibiae with PBS (or
DMEM/F-12 supplemented with 10% HI-FBS). A single cell preparation was obtained by carefully streaming through a 26-guage needle. Cells were then seeded in 10cm dishes containing 10mL DMEM/F-12 supplemented with 2mM L-glutamine (Gibco), 10% FBS, 2%
HEPES (Life Technologies), 1mM sodium pyruvate (Gibco), 25ug/mL gentamicin (Gibco), and
20% L-929 conditioned media. After 3 days of incubation at 37°C with 5% CO2, fresh L929- conditioned media was added. On day 6 of culture, macrophages were detached and seeded in 24 well plates at 10^6 cells/well.
In vitro macrophage infections:
Bacterial preparation: M. avium (strain 2–151 SmT) was grown from frozen to mid-log phage in Middlebroook 7H9 media, washed with PBS, sonicated to disperse clumps and resuspended in culture media without antibiotics.
Infection of murine BMDMs: M. avium was added to BMDM’s with or without additional recombinant murine IFNγ (Peprotech catalog # 315-05) at 200U/mL for 3 hours. After
3 hours of infection media and extracellular bacteria were removed and cells were washed once with PBS. Fresh media with or without IFNγ was then replaced until cells were harvested for intracellular bacterial enumeration at the indicated time points.
Bacterial enumeration: Live intracellular CFU counts were determined at the indicated time points. BMDM’s were washed twice with 1x DPBS to remove extracellular bacteria and 107 dead cells. Cells were then lysed with 0.05% saponin for 10 minutes. Serial dilutions of cell lysates were plated onto Middlebrook 7H11 (Sigma-Aldrich) agar plates. Colonies were counted after 10-14 days of incubation at 35°C.
Human PBMC collection, isolation and in vitro stimulation with M. avium:
Peripheral blood was collected in sodium heparin tubes. PBMC’s were isolated by density gradient centrifugation using Lymphocyte Separation Medium (Corning catalog#
MT25072CI). PBMC’s in the buffy coat layer were collected, washed, counted and plated at
2x10^6 PBMC/mL/well in 12-well plates. M. avium was added to infected wells at an MOI of 10 for 6 hours. After 6 hours, infected cells were harvested for RNA extraction in RLT buffer
(Qiagen RNeasy Mini Kit Catalog# 74106).
RNA isolation, cDNA synthesis and RT-PCR:
RNA was extracted from cells or tissue Trizol or RLT buffer and the RNeasy kit (Qiagen
Catalog# 74106) per the manufacturer’s protocol. Purified RNA was used for cDNA synthesis using the SuperScript III First Strand Synthesis System kit (Invitrogen catalog# 18080-051).
Quantitative PCR was performed using TaqMan detection with the QuantStudio3 Real Time
PCR System. All qPCR assays were normalized to ß-actin transcript levels and the relative gene expression was determined using the ΔΔCT method. TaqMan primers/probes for Irg1 and ß-actin were pre-designed by Applied Biosystems.
LC-MS/MS Metabolic Analysis:
108
Primary human elutriated monocytes were seeded in 12-well plates at 2x106/well and stimulated with media alone, LPS (200ng/mL), IFNγ (1000U/mL) or the combination of LPS and
IFNγ. Following various durations of stimulation, monocytes were washed twice with ice cold
0.9% NaCl and the packed cell volume of the recovered cell pellets was recorded for normalization. Polar metabolites and nucleic acids were extracted by simultaneous extraction
[116].
Polar metabolite fractions were shipped on dry ice to Agios Pharmaceuticals for LC-
MS/MS analysis. Internal standard (13C5 Itaconate), Cambridge Isotope Laboratories, INC) solution was prepared at 1000 ng/mL in 50/50 MeOH/Water and the calibration curve (4.5 ng/mL ~10,000 ng/mL) was prepared in 50/50 MeOH/Water using 12C Itaconate (Millipore
Sigma, St.Louis, MO). 180 µL of polar metabolite extract was aliquoted to 96 well plate and 10
µL of 1000 ng/mL 13C5 Itaconate in 50/50 MeOH/Water was spiked into samples and standards.
Samples were dried down in a genevac and reconstituted with 50uL of 90/10 Mobile phase
A/Mobile phase B and 10 µL was injected for an LC-MS/MS analysis.
Experiments were carried out on a Vanquish liquid chromatography (LC) system coupled to a Q Exactive (Thermo Fisher Scientific, Framingham, MA). LC separation was performed on
Rezex™ RFQ-Fast Acid H+ (8%), LC Column 100 x 7.8 mm, 4um column with mobile phases consisting of 0.1% Acetic acid in Water (Mobile Phase A) and 0.1% Acetic acid in IPA/Water
(90:10, v/v) (Mobile Phase B). The sample injection was run isocratically at 90% A at a flow rate of 0.6 mL/min. Mass spectra were acquired using electrospray ionization in negative-ion mode.
The settings of the ESI source were as follows: 55 Sheath gas flow, 35 Aux gas flow, 3 sweep gas, 3.3 spray voltage, 350’C Capillary Temp, 50 S-lens RF and 475’C Aux gas heater temp
109
Seahorse Metabolic Rate Assays with Bone Marrow Derived Macrophages:
WT and Irg1-/- BMDM’s were seeded in a Seahorse cell culture plate at 105/well and stimulated for 24 hours with M. avium (MOI = 25) with or without additional recombinant murine IFNγ (200U/mL) or IFNγ alone. Cellular oxidative phosphorylation (OXPHOS) and glycolysis were measured using the Seahorse Bioscience Extracellular Flux Analyzer (XFe96,
Seahorse Bioscience Inc., North Billerica, MA, USA) by measuring OCR (indicative of respiration) and ECAR (indicative of glycolysis) in real time according to manufacturer’s protocol. Prior to measurements, culture media was removed and replaced with 180 µl pH ready
Seahorse Assay Media (Agilent; Catalog #103575-100) and incubated in the absence of CO2 for
1 hour in the Biotek Cytation1 instrument during which time pre-assay brightfield images were collected. Basal levels of OCR and ECAR were recorded, then OCR and ECAR levels following injection of compounds that inhibit the mitochondrial electron transport chain, or ATP synthesis were monitored. Per the manufacturer’s protocol for the Mito Stress Test, assay cells were sequentially treated with oligomycin (2 µM), carbonyl cyanide-4-
(trifluoromethoxy)phenylhydrazone (FCCP) (0.5 µM), and rotenone + Antimycin A (0.5 µM).
OCR and ECAR were then measured in a standard six-minute cycle of mix (2 min), wait (2 min), and measure (2 min). All OCR and ECAR values were normalized following the Seahorse
Normalization protocol. Briefly, after the assay cells were stained with 2μg/mL Hoechst 33342
(ThermoFisher Scientific) for 30 minutes while performing post-assay brightfield imaging. Cells were then imaged and counted using the Biotek Cytation1. Cell counts were calculated by Cell
Imaging software (Agilent) and imported into Wave (Agilent) using the normalization function.
Human monocyte collection, isolation and stimulation:
110
Peripheral blood was collected in sodium heparin tubes. PBMCs were isolated as descried above. CD14+ positive selection was performed using magnetic beads (Miltenyi Biotec
CD14 MicroBeads catalog #130-050-201) following the manufacturer’s protocol. CD14+ monocytes were plated at 105/50 ul/well in Seahorse XF96 V3 PS cell culture microplates
(Agilent, 101085-004) in serum-free media. After 3 hours, 50 μl of complete media with 20%
(2x) FBS was added, and cells were rested overnight. Cells were then stimulated with media alone, IFNγ (1000 U/mL; Actimmune, NDC number 75987-111-10) and/or select inhibitors:
Rotenone (Sigma, #R8875), FK866 (Selleckchem #S2799), Diphenyleneiodonium (Sigma,
#D2926), Rot/AA, Oligomycin, or FCCP (Agilent Seahorse Mito Stress Test, #103015-100)
Seahorse Metabolic Rate Assays with Primary Human Monocytes:
Oxygen consumption rates (OCR), indicative of mitochondrial respiration, and extracellular acidification rates (ECAR), indicative of glycolysis, were measured using the
Seahorse Bioscience Extracellular Flux Analyzer (XFe96, Seahorse Bioscience Inc., North
Billerica, MA, USA). Prior to measurements, culture medium was removed and replaced with
180 µl pH ready Seahorse Assay Media (Agilent; Catalog #103575-100) and incubated in the
absence of CO2 for 1 hour in the Biotek Cytation1 while pre-assay brightfield images were collected. For the Mito Stress Test, cells were sequentially treated with oligomycin (2 µM), carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP) (0.5 µM), and rotenone+Antimycin A (0.5 µM). OCR and ECAR were measured in a standard, six-minute cycle. For the Glycolysis Stress Test, cells were sequentially treated with glucose (10 mM), oligomycin (1 µM) and 2-deoxyglucose (50 mM). For our modified assay, cells were treated with PMA (100 µM), then, in some cases, Rotenone and Antimycin A (0.5 µM). 111
All OCR and ECAR values were normalized by staining cells with 2 μg/mL Hoechst
33342 (ThermoFisher Scientific) for 30 minutes while performing post-assay brightfield imaging. Cells were then imaged and counted using the Biotek Cytation1. Cell counts were calculated by Cell Imaging software (Agilent) and imported into Wave (Agilent) using the normalization function. OCR values were normalized to cell count data.
RNA sequencing:
RNA sequencing was performed on RNA isolated from primary CD14+ monocytes from
4 normal donors following differentiation with mCSF (30ng/mL) for 5 days and subsequent stimulation with recombinant human IFNγ (100ng/mL) for 24 hours.
Total RNA sample integrity was assessed using automated capillary electrophoresis on a
Fragment Analyzer (Agilent) using the HS RNA Kit (15NT). For all samples RQN >8.0, total
RNA amount of >75ng was used as input for library preparation using the TruSeq Stranded mRNA Library Preparation Kit (Illumina). Sequencing libraries were quantified by real-time
PCR using the KAPA Library Quantification Complete kit (Roche) and assessed for size distribution, absence of free adapters and adapter dimers on a Fragment Analyzer. Sequencing libraries were pooled and sequenced on a HiSeq 3000 System (Illumina) using a HiSeq
3000/4000 PE Cluster Kit and SBS Kit (150 cycles) with run conditions of paired-end reads at
75bp length.
Sequenced reads (100bp, paired-end) were mapped to the human genome hg19
(GRCh38) using Bowtie 2.2.6[316] and Tophat 2.2.1[317]. Uniquely mapped reads were retained and raw counts that fell in coding regions were calculated and normalized using reads per kilobase per million (RPKM) mapped reads using given gene lengths from the UCSC 112
genome browser. Quantile normalization was applied to all samples[318] and data were log2- transformed. Non-expressed and weakly expressed genes, defined as having less than 1 read per million in 4 of the samples[319], were removed prior to subsequent analyses, resulting in a count table of 13,873 genes. Limma (Bioconductor package) was used to conduct differential expression analyses[320]. The voom module was used to transform data based on observational level weights derived from the mean-variance relationship prior to statistical modeling[321].
Pairwise contrasts were done within limma to identify differentially expressed (DE) genes between conditions. Genes with a Benjamini-Hochberg (BH) multiple-testing adjusted P value of
<0.05 were defined as differentially expressed. RNAseq can be accessed at GEO, https://www.ncbi.nlm.nih.gov/geo/ (accession number: GSE176562).
Volcano plot was generated using GraphPad Prism9 software. Metabolic genes involved in NAD(P)H metabolism and related pathways were extracted from the KEGG pathway database. Metabolic genes that overlapped with RNA-Seq data are shown in heat maps generated using the web tool ClustVis[322]. Unbiased assessment of the metabolic pathways transcriptionally regulated by IFNγ was performed using the ERGO 2.0 analysis platform
(Ingenbio, Chicago, IL).
Statistical analyses:
Statistical analyses were performed using GraphPad Prism9 software. Data are expressed as mean ± s.e.m., and P values were calculated using one-way analysis of variance (ANOVA) with
Tukey’s multiple comparisons test or two-way ANOVA with Sidak’s multiple comparisons test unless otherwise indicated. A confidence interval of 95% was used for all statistical tests. Sample sizes were determined based on the experiment type and standard practice in the field. 113
114
APPENDIX I: Clinical Presentation and Evaluation of a Patient with a Variant of Uncertain Significance in Immunoresponsive Gene 1
As described in Chapter II, mammalian IRG1 protein catalyzes the decarboxylation of cis-aconitate to itaconate [113]. IRG1 is one of the most highly upregulated transcripts, and its product, itaconate, is one of the most highly induced metabolites in M1 (LPS+IFNγ) activated macrophages [117]. Functional studies in murine macrophages have implicated itaconate as a critical mediator of antimicrobial, anti-inflammatory, and antioxidant processes [114, 121, 136,
141]. Itaconate has antimicrobial activity against a variety of human pathogens [127], but itaconate deficiency has not been directly implicated as cause of immunodeficiency.
Little is known about the function of IRG1 and itaconate in the human immune response.
Chen et al. reported that IRG1 polymorphisms resulting in higher levels of itaconate production are maintained in populations from Mycobacterium tuberculosis endemic regions, possibly because of some protection conferred by these polymorphisms [198]. Liu et al. demonstrated that certain IRG1 polymorphisms are also associated with an enhanced response to hepatitis B vaccination [323]. Additionally, exogenous forms of itaconate, such as dimethyl itaconate and 4- octyl-itaconate, have been shown to have an anti-inflammatory effect on peripheral blood mononuclear cells (PBMC’s) from patients with systemic lupus erythematosus, and reduce IL-1ß secretion in PBMC’s from patients with cryopyrin-associated periodic syndrome (CAPS) [139,
279].
Many studies have reported a role for murine IRG1 in diseases ranging from arthritis to cancer as well as viral and bacterial infection [121, 134, 324, 325]. Nair et al. demonstrated that
IRG1 is required for survival following in vivo infection with Mycobacterium tuberculosis [121].
Basler et al. reported that Irg1 is also induced by non-tuberculous mycobacteria and LPS in murine macrophages [326] and Chen et al. identified a role for itaconate in control of 115 intracellular Salmonella enterica [129]. The proposed antimicrobial mechanism of itaconate is inhibition of isocitrate lyase which is required for survival of many intracellular pathogens [123,
125, 327, 328]. Consistent with the understanding that itaconate has direct antimicrobial activity, itaconate degradation pathways have been identified in pathogens and are associated with their virulence [130]. These data suggest that itaconate could play an important role in the context of human inflammatory and infectious diseases.
Here, we investigated the case of a 15-month-old girl who presented with severe and recurrent upper and lower extremity pyomyositis/abscesses (Figure A1 A-B) caused by multiple bacteria, including gas producing organisms (Figure A1 C), requiring prolonged hospitalization for IV antibiotics and fasciectomies (Figure A1 D).
Figure A1:
A B C D
Figure A1: Severe pyomyositis caused by multiple bacteria, requiring fasciectomies. (a-b) MRI of right lower extremity (a) and left upper extremity (b) showing diffuse inflammation of the muscle and soft tissue. (c) plain film radiograph of right lower extremity abscess with visible air as a result of gas producing organisms. (d) patient’s left arm status post fasciectomy.
The patient was initially evaluated in her home country of Cape Verde where she first presented at 2 months of age with myositis of her left upper extremity. Access to appropriate antibiotics in Cape Verde was limited, and the patient went on to experience 18 episodes of 116 myositis recurring in the same 2 locations in her left upper extremity and right lower extremity over the next 12 months. At 15 months of age she was transferred to Hospital Dona Estefania in
Lisbon, Portugal for further evaluation and primary immunodeficiency (PID) work up.
She had no known family history of immunodeficiency, no siblings at the time of work up and no known consanguinity in her family. Her mother denied any specific inciting event such as an injury or insect bite and denies any association with location and timing of vaccinations. She responded normally to all vaccinations, including BCG vaccination, and she has not had any other abnormal infections.
Her work up upon arrival in Lisbon revealed no evidence of gastrointestinal disease after multiple endoscopies, normal B and T cell numbers and functions, normal DHR and normal response to TLR ligands. During recurrences she experienced exacerbations of localized pain and swelling in the arm or leg, appropriately elevated inflammatory markers, and fever.
Importantly, her blood cultures were always negative and the inflammation was always limited to the soft tissue, with no evidence of osteomyelitis. Her disease course (Table A1) was marked by multiple episodes of recurrent pyomyositis which were severe, but always responsive to IV antibiotics, and on multiple occasions required surgeries to drain abscesses and relieve compartment syndrome. Drainage of abscesses was remarkable for copious amounts of purulent pus and cultures always revealed multiple bacteria, including gram negatives, gram positives and anaerobes (Table A2). Ultimately, she was treated with subcutaneous IFNγ which seemed to reduce the frequency of recurrence, but she did have at least 1 recurrence of pyomyositis in her arm while on IFNγ therapy.
Table A1: 117
2-15 Months 15-18 Months 17-26 Months 26 Months- 4.5 Years In Cape Verde: Transferred to Lisbon • 20M: Myositis + fasciitis thigh. • 26M: Myositis thigh • 18 episodes of fever • Myositis + fasciitis thigh Required fasciectomy, multiple • Required fasciectomy, multiple • Arthritis of the knee • Required fasciectomy drainages, 6W IV antibiotics drainages, 4W IV Ab • Arthritis of the elbow • Isolated B. fragilis, K. • Isolated E. cloacae, E. coli, F. • Added therapeutic IFN! (CGD pneumoniae, S. aureus necrophorum, K. pneumoniae dosing) • PID workup Normal • 20-26M: No episodes • Isolated F. necrophorum, E. coli, • 17M: Myositis arm B. fragilis, S. constellatus • 18M: Myositis leg • Discharged on prophylatic IFN!, • Discharged on Trimethoprim/ cotrimoxazole, cefuroxime Sulfamethoxazole • 39M: Myositis arm requiring fasciectomy • 40-55M: No episodes
Table A1: Detailed Course of Disease from Presentation Until Patient was Lost to Follow- up. Each column lists significant events in the patient’s disease course and management by age.
Table A2: Gram Positive Gram Negative Anaerobes Staphylococcus aureus (MSSA) Klebsiella pneumoniae Bacteroides fragilis Streptococcus constellatus Enterobacter cloacae Fusobacterium necrophorum Escherichia coli
Table A2: Gram Positive, Gram Negative and Anaerobic Bacteria Cultured from Patient’s Abscesses. List of organisms cultured at some point in the course of her disease. Some organisms were identified on multiple occasions.
IRG1 Variant of Uncertain Significance:
Whole exome sequencing was performed on this patient and revealed a rare, variant of uncertain significance (VUS) in IRG1. We investigated the functional effects of this variant on
IRG1 decarboxylase activity and the potential contribution of this variant to the patient’s clinical phenotype. The variant identified was a heterozygous missense variant, p.Ala214Val
(13/77531315 C/T), predicted to be deleterious or at least probably damaging by SIFT and
PolyPhen, respectively. This variant is very rare, with no reported homozygotes in the gnomAD database (https://gnomad.broadinstitute.org/variant/13-77531315-C-T). Only 3 heterozygotes are reported out of 180,540 alleles, resulting in an allele frequency of 1.662e-5.
This residue is also highly conserved across vertebrates, invertebrates and bacteria,
(Figure A2; NCBI blast; https://blast.ncbi.nlm.nih.gov/Blast.cgi) indicating a potentially important role for this residue in the structure and/or function of IRG1. 118
Figure A2:
IRG1 (Cis-aconitate decarboxylase) sequence conservation:
Protein Name Species A214 Cis-aconitate decarboxylase Homo sapiens Cis-aconitate decarboxylase Rattus norvegicus Cis-aconitate decarboxylase Mus musculus Immunoresponsive gene 1 Danio rerio Cis-aconitate decarboxylase Danio rerio 2-methylcitrate dehydratase Burkholderia sp. MmgE/PrpD family protein E. coli
Figure A2: Amino Acid Sequence Alignment of IRG1 and Orthologous Decarboxylase Enzymes from Other Species.
Before the crystal structure of mammalian IRG1 was solved, we relied on the crystal structure of bacterial IDS epimerase and other MmgE/PrpD protein family members [329] to determine the potential structural effects of this variant. Based on the IDS epimerase structure, and later confirmed by the mammalian IRG1 crystal structure, the enzyme likely forms a homodimer with each monomer consisting of 2 domains [198, 330]. The larger domain of the monomer forms a central helical region consisting of 6 α-helices and the smaller domain forms a mobile lid. The putative active site is located at the junction of the two domains, in a region comprised of basic residues, amenable to binding the negatively charged carboxyl groups on substrates iminodisuccinate for IDS epimerase or cis-aconitate for IRG1.
According to IDS epimerase structure, alanine 214 is located in the large subunit (Met1-
Lys266 and Leu400-Pro446), within an α-helical secondary structure. The variant we observed
(NM_001258406.2: c.641C>T) results in an amino acid change from alanine to valine, which are both non-polar, aliphatic amino acids, with a slightly different side groups (that of valine is larger). This type of missense change within an α-helix has a 50% chance of altering the secondary structure depending on whether this residue faces interiorly or exteriorly within the helix [331]. If this variant did alter the α-helical secondary structure, it is unlikely that it would 119
affect the active site based on its location, but could alter the dimerization domain located within
the larger helical domain [332]. A heterozygous variant that alters the structure of the
dimerization domain of IRG1 could decrease its enzymatic activity up to 75% if the variant has a
dominant negative effect in which homodimerization is maintained or enhanced, but the
functional activity of mutant dimers is reduced [333].
Therefore, we designed a functional assay to assess the enzymatic activity of a
heterozygous IRG1 A214V variant in primary cells from the patient. We first identified the
conditions under which IRG1 expression and itaconate were maximally induced so that we could
detect and quantify subtle changes in itaconate levels caused by the heterozygous variant of
interest. As described in chapter II, we found that the combination of LPS and IFNγ stimulation
for 24 hours yielded the highest levels of intracellular itaconate (Figure A3). We first generated a
“normal range” of itaconate levels based on the intracellular concentration of itaconate measured
by LC-MS/MS from elutriated monocytes from 6 healthy donors. The concentration of
intracellular itaconate was calculated using a standard curve and normalized to the packed cell
volume from which the polar metabolites were extracted.
Figure A3:
Unstimulated IFNg LPS LPS+IFNg Unstimulated +IFNγ +LPS +LPS+IFNγ Donor: 10000 10000 UnstimulatedUnstimulatedUnstimulatedUnstimulated 10000 IFNg IFNg IFNg IFNg LPS LPS 10000LPS LPS LPS+IFNgLPS+IFNgLPS+IFNgLPS+IFNg 3L 3L 3L 3L1 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 10000 3L 3L 3L 3L 3L 3L 3L 3L 8000 3M 8000 3M 8000 3L 3L 3L 3M3L 8000 3L3M2 3L 3L 3L 8000 8000 8000 8000 3M 3M8000 3M8000 3M8000 8000 3M 3M8000 3M8000 3M8000 8000 3M 3M8000 3M8000 3M8000 8000 3M 3M 3M 3M 3N 3N3N 3N 3N 3N 3N 3N 3N 3N3N 3N 3N 3N 3N 3N3N3 3N 3N 3N 6000 6000 6000 6000 6000 6000 6000 60006000 6000 6000 6000 6000 6000 6000 6000 6000 6000 6000 6000 3O 3O 3O 3O 3O 3O 3O 3O 3O3O 3O 3O 3O 3O 3O 3O 3O 3O 4000 4000 4000 4000 3O 4000 4000 4000 4000 4000 4000 4000 4000 3O4 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 3Q 3Q 3Q 3Q RQ IRG1 3Q 3Q4000 3Q4000 3Q4000 4000 RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 3Q 3Q 3Q 3Q RQ IRG1 RQ IRG1 RQ IRG1 RQ IRG1 3Q 3Q 3Q 3Q 4000 4000 4000 RQ IRG1 3Q 3R 3R RQ IRG1 3R 3R 3Q 4000 3R 3R 3R 3R RQ IRG1 2000 2000 2000 2000 3Q 3R 3R2000 3R2000 3R2000 2000 2000 2000 2000 2000 RQ IRG1 2000 2000 2000 2000 3R3Q5 3R 3R 3R IRG1mRNA expression 3R 0 0 0 0 3R 0 0 3R 0 0 2000 0 2000 0 0 0 0 0 0 0 3R 2000 1h 3h 6h1h 12h3h 18h6h1h 24h12h3h 48h18h6h1h 24h12h3h 48h18h6h 24h12h 48h18h 24h 48h 1h 3h 6h1h 12h3h 18h6h1h 24h12h3h 48h18h6h1h200024h12h3h 48h18h6h 24h12h 48h18h 24h 48h 6 1h 3h 6h1h 12h3h 18h6h1h 24h12h3h 48h18h6h1h 24h12h3h 48h18h6h 24h12h 48h18h 24h 48h 1h 3h 6h1h 12h3h 18h6h1h 24h12h3h 48h18h6h1h 24h12h3h 48h18h6h 24h12h 48h18h 24h 48h Time (hours)Time (hours)Time (hours)Time (hours) Time (hours)Time (hours)Time (hours)Time (hours) Time (hours)Time (hours)Time (hours)Time (hours) 0 0 Time (hours)Time (hours)Time (hours)Time (hours) 0 0 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h Time (hours) Time (hours) Time (hours) Time (hours)
Unstimulated IFNg LPS LPS+IFNg Donor: 10000 10000 Unstimulated 10000 +IFNγ +LPS 10000 +LPS+IFNγ 3L UnstimulatedUnstimulatedUnstimulatedUnstimulated3L IFNg IFNg IFNg IFNg 3LLPS LPS LPS LPS LPS+IFNg LPS+IFNg LPS+IFNg LPS+IFNg3L1 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 8000 3M 8000 3L 3M 3L 80003L 3L 3L 3L 3L 3M3L 8000 3L 3L 3L 3L 3L 3M2 3L 3L 3L 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3M 3N 3N 3N 3N 6000 6000 100 100 100 100 3N 3N100 60003N 100 3N 100 100 3N 3N100 3N 100 3N 100 100 6000 3N 3N100 3N100 3N 100 100 3N 3 3N 3N 3N 3O 3O 3O 3O 3O 3O 3O 3O 3O 3O3O 3O 3O 3O 3O 3O 3O4 3O 3O 3O 4000 4000 50 50 50 50 3Q 3Q 50 40003Q 50 3Q 50 50 3Q 3Q 50 3Q 50 3Q 50 50 3Q 3Q 50 3Q 50 3Q 50 50 3Q 3Q 3Q 3Q RQ IRG1 3Q RQ IRG1 3Q 4000
RQ IRG1 3Q 3R 3R 3R 3R 3R 3R 3R 3R RQ IRG1 3R 3R 3R 3R 3R 3Q5 3R 3R 3R 3R Itaconate 3R 2000 2000 Intracellular 3R 2000 3R Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 Intracellular Itaconate (uM) 0 0 0 0 0 2000 0 0 0 0 6 1h 3h 6h1h 12h3h 18h6h1h24h12h3h48h18h6h1h24h12h3h48h18h6h 24h12h 48h18h 24h 1h48h 3h 6h1h 12h3h 18h6h1h24h12h3h48h18h6h1h24h12h3h48h18h6h 24h12h 48h18h 24h 1h48h 3h 6h1h 12h3h 18h6h1h24h12h3h48h18h6h1h24h12h3h48h18h6h 24h12h 48h18h 24h 1h48h 3h 6h1h 12h3h 18h6h1h24h12h3h48h18h6h1h24h12h3h48h18h6h 24h12h 48h18h 24h 48h 0 Time (hours)Time (hours)Time (hours)Time (hours) 0 Time (hours)Time (hours)Time (hours)Time (hours) Time (hours)Time (hours)Time (hours)Time (hours) Time (hours)Time (hours)Time (hours)Time (hours) 0 0 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h 1h 3h 6h 12h 18h 24h 48h Time (hours) Time (hours) Time (hours) Time (hours) 120
Figure A3: Irg1 mRNA Expression and Intracellular Itaconate Levels are Synergistically Induced by the Combination of LPS and IFNγ. (a-b) Elutriated primary human monocytes from 6 healthy donors were stimulated with media alone (unstimulated), LPS (200ng/mL), IFNγ (1000U/mL), or the combination of LPS and IFNγ for time points indicated on X axis. RNA and intracellular metabolites were isolated by simultaneous extraction of polar metabolites and nucleic acids from the same cells. (a) isolated RNA was used to quantify Irg1 expression by qPCR. (b) Intracellular itaconate levels were quantified by LC-MS based on a standard curve and normalized to packed cell volume.
Peripheral blood from the patient was then shipped from Lisbon, Portugal for analysis. At the time the patient’s blood was collected, blood from a healthy donor on site was also collected.
Half of that sample was processed immediately (healthy control fresh) and the other half was left at room temperature to be processed when the patient sample arrived (healthy control 3 day old) to account for the degradation that may occur in the time it took the patient sample to arrive.
When the patient’s blood arrived, it was processed along with the remaining healthy control sample. PBMC-derived CD14+ monocytes were stimulated with LPS and IFNγ for 24 hours, then
RNA and polar metabolites were extracted using a method of simultaneous extraction of nucleic acids and metabolites [116].
We compared itaconate levels from the patient’s CD14+ monocytes to controls and found that although the concentration of intracellular itaconate in the patient’s monocytes was slightly lower than the healthy control cohort, it was comparable to healthy control monocytes that were processed 3 days after sample collection (Figure A4). Therefore, we concluded that despite the
A214V variant in IRG1, the patient was able to produce approximately normal levels of itaconate when maximally stimulated. While this does not rule out subtle defects in itaconate production kinetics or peak levels in response to other stimuli, it is unlikely that this variant would result in a physiologically significant reduction in itaconate production due to altered IRG1 enzymatic activity.
Figure A4: 121
Intracellular Itaconate 150
100
50
Intracellular Itaconate (uM) 0
A214V)
IRG1
Patient ( Healthy Control (Fresh) Healthy Control (3 day old)
Healthy Control Elutriated Monocytes Figure A4: Intracellular itaconate concentrations from a patient with and IRG1 variant (A214V) and healthy controls. Polar metabolites were isolated from CD14+ monocytes following stimulation with LPS (200ng/mL) and IFNγ (1000U/mL) for 24 hours. Itaconate levels were measured by LC-MS/MS and intracellular concentrations were calculated based on a standard curve and normalized to packed cell volumes.
Our data demonstrate that this specific IRG1 variant does not dramatically alter maximal itaconate production in primary monocytes and therefore, likely does not contribute to this patient’s clinical phenotype. However, Chen et al. showed that targeted disruption of residues in or near the active site of IRG1 can have functional consequences that significantly increase or reduce its ability to produce itaconate. Functional assessment of naturally occurring variants in human IRG1 revealed that all of the variants that have been experimentally determined to be loss of function are very rare (allele frequency < 2.22e-5) and only exist as heterozygotes while the 2 variants which have been experimentally determined to be gain of function or benign have the highest allele frequencies and occur as homozygotes [198].
The fact that expression of this gene is so highly induced and that its sequence is conserved across many species suggests an important role in the human immune response. However, 122 consistent with our findings in Irg1 knock out mice (Chapter II), there are likely redundant pathways that compensate for dysregulated IRG1 activity. While we cannot conclude that this patient’s phenotype is related to the IRG1 VUS, there is ample evidence to suggest a role for
IRG1 in the context of infection. These data define a process for functional assessment of IRG1 activity which could be considered in patients presenting with atypical infections and variants in
IRG1.
123
CHAPTER VI: References
1. Wheelock, E.F., Interferon-Like Virus-Inhibitor Induced in Human Leukocytes by Phytohemagglutinin. Science, 1965. 149(3681): p. 310-1. 2. Schroder, K., et al., Interferon-gamma: an overview of signals, mechanisms and functions. J Leukoc Biol, 2004. 75(2): p. 163-89. 3. Pestka, S., C.D. Krause, and M.R. Walter, Interferons, interferon-like cytokines, and their receptors. Immunol Rev, 2004. 202: p. 8-32. 4. Trent, J.M., S. Olson, and R.M. Lawn, Chromosomal localization of human leukocyte, fibroblast, and immune interferon genes by means of in situ hybridization. Proc Natl Acad Sci U S A, 1982. 79(24): p. 7809-13. 5. Savan, R., et al., Structural conservation of interferon gamma among vertebrates. Cytokine Growth Factor Rev, 2009. 20(2): p. 115-24. 6. Young, H.A. and K.J. Hardy, Role of interferon-gamma in immune cell regulation. J Leukoc Biol, 1995. 58(4): p. 373-81. 7. Fountoulakis, M., et al., Stoichiometry of interaction between interferon gamma and its receptor. Eur J Biochem, 1992. 208(3): p. 781-7. 8. Pestka, S., et al., The interferon gamma (IFN-gamma) receptor: a paradigm for the multichain cytokine receptor. Cytokine Growth Factor Rev, 1997. 8(3): p. 189-206. 9. Bach, E.A., M. Aguet, and R.D. Schreiber, The IFN gamma receptor: a paradigm for cytokine receptor signaling. Annu Rev Immunol, 1997. 15: p. 563-91. 10. Merlin, G., et al., The gene for the ligand binding chain of the human interferon gamma receptor. Immunogenetics, 1997. 45(6): p. 413-21. 11. Pfizenmaier, K., et al., High affinity human IFN-gamma-binding capacity is encoded by a single receptor gene located in proximity to c-ros on human chromosome region 6q16 to 6q22. J Immunol, 1988. 141(3): p. 856-60. 12. Rosenzweig, S.D. and S.M. Holland, Defects in the interferon-gamma and interleukin-12 pathways. Immunol Rev, 2005. 203: p. 38-47. 13. Rosenzweig, S.D., et al., A novel mutation in IFN-gamma receptor 2 with dominant negative activity: biological consequences of homozygous and heterozygous states. J Immunol, 2004. 173(6): p. 4000-8. 14. Soh, J., et al., Identification and sequence of an accessory factor required for activation of the human interferon gamma receptor. Cell, 1994. 76(5): p. 793-802. 15. Rosenzweig, S.D., et al., Characterization of a dipeptide motif regulating IFN-gamma receptor 2 plasma membrane accumulation and IFN-gamma responsiveness. J Immunol, 2004. 173(6): p. 3991-9. 16. Krause, C.D., et al., Seeing the light: preassembly and ligand-induced changes of the interferon gamma receptor complex in cells. Mol Cell Proteomics, 2002. 1(10): p. 805- 15. 17. Bernabei, P., et al., Interferon-gamma receptor 2 expression as the deciding factor in human T, B, and myeloid cell proliferation or death. J Leukoc Biol, 2001. 70(6): p. 950- 60. 18. Kotenko, S.V., et al., Identification and functional characterization of a second chain of the interleukin-10 receptor complex. EMBO J, 1997. 16(19): p. 5894-903. 19. Rhee, S., et al., The structure of the gene for the second chain of the human interferon- gamma receptor. J Biol Chem, 1996. 271(46): p. 28947-52. 124
20. Dorman, S.E. and S.M. Holland, Interferon-gamma and interleukin-12 pathway defects and human disease. Cytokine Growth Factor Rev, 2000. 11(4): p. 321-33. 21. Bach, E.A., et al., Ligand-induced assembly and activation of the gamma interferon receptor in intact cells. Mol Cell Biol, 1996. 16(6): p. 3214-21. 22. Kotenko, S.V., et al., Interaction between the components of the interferon gamma receptor complex. J Biol Chem, 1995. 270(36): p. 20915-21. 23. Briscoe, J., et al., Kinase-negative mutants of JAK1 can sustain interferon-gamma- inducible gene expression but not an antiviral state. EMBO J, 1996. 15(4): p. 799-809. 24. Greenlund, A.C., et al., Stat recruitment by tyrosine-phosphorylated cytokine receptors: an ordered reversible affinity-driven process. Immunity, 1995. 2(6): p. 677-87. 25. Decker, T., P. Kovarik, and A. Meinke, GAS elements: a few nucleotides with a major impact on cytokine-induced gene expression. J Interferon Cytokine Res, 1997. 17(3): p. 121-34. 26. Igarashi, K., et al., Interferon-gamma induces tyrosine phosphorylation of interferon- gamma receptor and regulated association of protein tyrosine kinases, Jak1 and Jak2, with its receptor. J Biol Chem, 1994. 269(20): p. 14333-6. 27. Kubin, M., J.M. Chow, and G. Trinchieri, Differential regulation of interleukin-12 (IL- 12), tumor necrosis factor alpha, and IL-1 beta production in human myeloid leukemia cell lines and peripheral blood mononuclear cells. Blood, 1994. 83(7): p. 1847-55. 28. Yoshida, A., et al., IFN-gamma induces IL-12 mRNA expression by a murine macrophage cell line, J774. Biochem Biophys Res Commun, 1994. 198(3): p. 857-61. 29. Macatonia, S.E., et al., Dendritic cells produce IL-12 and direct the development of Th1 cells from naive CD4+ T cells. J Immunol, 1995. 154(10): p. 5071-9. 30. Trinchieri, G. and P. Scott, Interleukin-12: a proinflammatory cytokine with immunoregulatory functions. Res Immunol, 1995. 146(7-8): p. 423-31. 31. Lehn, M., et al., IL-4 inhibits H2O2 production and antileishmanial capacity of human cultured monocytes mediated by IFN-gamma. J Immunol, 1989. 143(9): p. 3020-4. 32. Kang, K., et al., Interferon-gamma Represses M2 Gene Expression in Human Macrophages by Disassembling Enhancers Bound by the Transcription Factor MAF. Immunity, 2017. 47(2): p. 235-250 e4. 33. Wallach, D., M. Fellous, and M. Revel, Preferential effect of gamma interferon on the synthesis of HLA antigens and their mRNAs in human cells. Nature, 1982. 299(5886): p. 833-6. 34. Chang, C.H., et al., The activation of major histocompatibility complex class I genes by interferon regulatory factor-1 (IRF-1). Immunogenetics, 1992. 35(6): p. 378-84. 35. Groettrup, M., et al., Interferon-gamma inducible exchanges of 20S proteasome active site subunits: why? Biochimie, 2001. 83(3-4): p. 367-72. 36. Boss, J.M., Regulation of transcription of MHC class II genes. Curr Opin Immunol, 1997. 9(1): p. 107-13. 37. Mach, B., et al., Regulation of MHC class II genes: lessons from a disease. Annu Rev Immunol, 1996. 14: p. 301-31. 38. Figueiredo, F., T.J. Koerner, and D.O. Adams, Molecular mechanisms regulating the expression of class II histocompatibility molecules on macrophages. Effects of inductive and suppressive signals on gene transcription. J Immunol, 1989. 143(11): p. 3781-6. 39. Lah, T.T., et al., Gamma-interferon causes a selective induction of the lysosomal proteases, cathepsins B and L, in macrophages. FEBS Lett, 1995. 363(1-2): p. 85-9. 125
40. Cresswell, P., Assembly, transport, and function of MHC class II molecules. Annu Rev Immunol, 1994. 12: p. 259-93. 41. Xaus, J., et al., Interferon gamma induces the expression of p21waf-1 and arrests macrophage cell cycle, preventing induction of apoptosis. Immunity, 1999. 11(1): p. 103- 13. 42. Harvat, B.L., P. Seth, and A.M. Jetten, The role of p27Kip1 in gamma interferon- mediated growth arrest of mammary epithelial cells and related defects in mammary carcinoma cells. Oncogene, 1997. 14(17): p. 2111-22. 43. Matsuoka, M., I. Nishimoto, and S. Asano, Interferon-gamma impairs physiologic downregulation of cyclin-dependent kinase inhibitor, p27Kip1, during G1 phase progression in macrophages. Exp Hematol, 1999. 27(2): p. 203-9. 44. Novelli, F., et al., Switching on of the proliferation or apoptosis of activated human T lymphocytes by IFN-gamma is correlated with the differential expression of the alpha- and beta-chains of its receptor. J Immunol, 1996. 157(5): p. 1935-43. 45. Liao, F., et al., Human Mig chemokine: biochemical and functional characterization. J Exp Med, 1995. 182(5): p. 1301-14. 46. Taub, D.D., et al., Recombinant human interferon-inducible protein 10 is a chemoattractant for human monocytes and T lymphocytes and promotes T cell adhesion to endothelial cells. J Exp Med, 1993. 177(6): p. 1809-14. 47. Rollins, B.J., Monocyte chemoattractant protein 1: a potential regulator of monocyte recruitment in inflammatory disease. Mol Med Today, 1996. 2(5): p. 198-204. 48. Appay, V. and S.L. Rowland-Jones, RANTES: a versatile and controversial chemokine. Trends Immunol, 2001. 22(2): p. 83-7. 49. Taub, D.D., et al., Preferential migration of activated CD4+ and CD8+ T cells in response to MIP-1 alpha and MIP-1 beta. Science, 1993. 260(5106): p. 355-8. 50. Erbe, D.V., et al., The effect of cytokines on the expression and function of Fc receptors for IgG on human myeloid cells. Mol Immunol, 1990. 27(1): p. 57-67. 51. Strunk, R.C., et al., gamma-Interferon increases expression of class III complement genes C2 and factor B in human monocytes and in murine fibroblasts transfected with human C2 and factor B genes. J Biol Chem, 1985. 260(28): p. 15280-5. 52. Cuellar-Mata, P., et al., Nramp1 modifies the fusion of Salmonella typhimurium- containing vacuoles with cellular endomembranes in macrophages. J Biol Chem, 2002. 277(3): p. 2258-65. 53. Frehel, C., et al., Effect of Nramp1 on bacterial replication and on maturation of Mycobacterium avium-containing phagosomes in bone marrow-derived mouse macrophages. Cell Microbiol, 2002. 4(8): p. 541-56. 54. Gallant, C.J., et al., Reduced in vitro functional activity of human NRAMP1 (SLC11A1) allele that predisposes to increased risk of pediatric tuberculosis disease. Genes Immun, 2007. 8(8): p. 691-8. 55. Govoni, G., et al., Genomic structure, promoter sequence, and induction of expression of the mouse Nramp1 gene in macrophages. Genomics, 1995. 27(1): p. 9-19. 56. Jabado, N., et al., Natural resistance to intracellular infections: natural resistance- associated macrophage protein 1 (Nramp1) functions as a pH-dependent manganese transporter at the phagosomal membrane. J Exp Med, 2000. 192(9): p. 1237-48. 57. Koh, W.J., et al., NRAMP1 gene polymorphism and susceptibility to nontuberculous mycobacterial lung diseases. Chest, 2005. 128(1): p. 94-101. 126
58. Marquet, S., et al., Variants of the human NRAMP1 gene and altered human immunodeficiency virus infection susceptibility. J Infect Dis, 1999. 180(5): p. 1521-5. 59. MacMicking, J., Q.W. Xie, and C. Nathan, Nitric oxide and macrophage function. Annu Rev Immunol, 1997. 15: p. 323-50. 60. MacMicking, J.D., et al., Altered responses to bacterial infection and endotoxic shock in mice lacking inducible nitric oxide synthase. Cell, 1995. 81(4): p. 641-50. 61. Nicholson, S.C., et al., Lethality of endotoxin in mice genetically deficient in the respiratory burst oxidase, inducible nitric oxide synthase, or both. Shock, 1999. 11(4): p. 253-8. 62. MacMicking, J.D., et al., Identification of nitric oxide synthase as a protective locus against tuberculosis. Proc Natl Acad Sci U S A, 1997. 94(10): p. 5243-8. 63. Nathan, C., Role of iNOS in human host defense. Science, 2006. 312(5782): p. 1874-5; author reply 1874-5. 64. Shiloh, M.U., et al., Phenotype of mice and macrophages deficient in both phagocyte oxidase and inducible nitric oxide synthase. Immunity, 1999. 10(1): p. 29-38. 65. Drutman, S.B., et al., Fatal Cytomegalovirus Infection in an Adult with Inherited NOS2 Deficiency. N Engl J Med, 2020. 382(5): p. 437-445. 66. Bridges, R.A., H. Berendes, and R.A. Good, A fatal granulomatous disease of childhood; the clinical, pathological, and laboratory features of a new syndrome. AMA J Dis Child, 1959. 97(4): p. 387-408. 67. Holmes, B., et al., Fatal granulomatous disease of childhood. An inborn abnormality of phagocytic function. Lancet, 1966. 1(7449): p. 1225-8. 68. Cassatella, M.A., et al., Molecular basis of interferon-gamma and lipopolysaccharide enhancement of phagocyte respiratory burst capability. Studies on the gene expression of several NADPH oxidase components. J Biol Chem, 1990. 265(33): p. 20241-6. 69. Newburger, P.E. and R.A. Ezekowitz, Cellular and molecular effects of recombinant interferon gamma in chronic granulomatous disease. Hematol Oncol Clin North Am, 1988. 2(2): p. 267-76. 70. Newburger, P.E., et al., Induction of phagocyte cytochrome b heavy chain gene expression by interferon gamma. Proc Natl Acad Sci U S A, 1988. 85(14): p. 5215-9. 71. Ezekowitz, R.A., et al., Restoration of phagocyte function by interferon-gamma in X- linked chronic granulomatous disease occurs at the level of a progenitor cell. Blood, 1990. 76(12): p. 2443-8. 72. Ezekowitz, R.A., S.H. Orkin, and P.E. Newburger, Recombinant interferon gamma augments phagocyte superoxide production and X-chronic granulomatous disease gene expression in X-linked variant chronic granulomatous disease. J Clin Invest, 1987. 80(4): p. 1009-16. 73. Ezekowitz, R.A., et al., Partial correction of the phagocyte defect in patients with X- linked chronic granulomatous disease by subcutaneous interferon gamma. N Engl J Med, 1988. 319(3): p. 146-51. 74. Marciano, B.E., et al., Long-term interferon-gamma therapy for patients with chronic granulomatous disease. Clin Infect Dis, 2004. 39(5): p. 692-9. 75. A controlled trial of interferon gamma to prevent infection in chronic granulomatous disease. The International Chronic Granulomatous Disease Cooperative Study Group. N Engl J Med, 1991. 324(8): p. 509-16. 127
76. Holland, S.M., et al., Treatment of refractory disseminated nontuberculous mycobacterial infection with interferon gamma. A preliminary report. N Engl J Med, 1994. 330(19): p. 1348-55. 77. Dorman, S.E., et al., Clinical features of dominant and recessive interferon gamma receptor 1 deficiencies. Lancet, 2004. 364(9451): p. 2113-21. 78. Bax, H.I., et al., Interferon alpha treatment of patients with impaired interferon gamma signaling. J Clin Immunol, 2013. 33(5): p. 991-1001. 79. Chantrain, C.F., et al., Successful hematopoietic stem cell transplantation in a child with active disseminated Mycobacterium fortuitum infection and interferon-gamma receptor 1 deficiency. Bone Marrow Transplant, 2006. 38(1): p. 75-6. 80. Roesler, J., et al., Hematopoietic stem cell transplantation for complete IFN-gamma receptor 1 deficiency: a multi-institutional survey. J Pediatr, 2004. 145(6): p. 806-12. 81. Rottman, M., et al., IFN-gamma mediates the rejection of haematopoietic stem cells in IFN-gammaR1-deficient hosts. PLoS Med, 2008. 5(1): p. e26. 82. Waibel, K.H., et al., Fever and leg pain in a 42-month-old. Ann Allergy Asthma Immunol, 2002. 89(3): p. 239-43. 83. Jouanguy, E., et al., A human IFNGR1 small deletion hotspot associated with dominant susceptibility to mycobacterial infection. Nat Genet, 1999. 21(4): p. 370-8. 84. Sologuren, I., et al., Partial recessive IFN-gammaR1 deficiency: genetic, immunological and clinical features of 14 patients from 11 kindreds. Hum Mol Genet, 2011. 20(8): p. 1509-23. 85. Dorman, S.E. and S.M. Holland, Mutation in the signal-transducing chain of the interferon-gamma receptor and susceptibility to mycobacterial infection. J Clin Invest, 1998. 101(11): p. 2364-9. 86. Vogt, G., et al., Gains of glycosylation comprise an unexpectedly large group of pathogenic mutations. Nat Genet, 2005. 37(7): p. 692-700. 87. Vogt, G., et al., Gain-of-glycosylation mutations. Curr Opin Genet Dev, 2007. 17(3): p. 245-51. 88. Oleaga-Quintas, C., et al., A purely quantitative form of partial recessive IFN-gammaR2 deficiency caused by mutations of the initiation or second codon. Hum Mol Genet, 2018. 27(22): p. 3919-3935. 89. Bustamante, J., et al., Mendelian susceptibility to mycobacterial disease: genetic, immunological, and clinical features of inborn errors of IFN-gamma immunity. Semin Immunol, 2014. 26(6): p. 454-70. 90. Moncada-Velez, M., et al., Partial IFN-gammaR2 deficiency is due to protein misfolding and can be rescued by inhibitors of glycosylation. Blood, 2013. 122(14): p. 2390-401. 91. Kong, X.F., et al., Haploinsufficiency at the human IFNGR2 locus contributes to mycobacterial disease. Hum Mol Genet, 2013. 22(4): p. 769-81. 92. Kerner, G., et al., Inherited human IFN-gamma deficiency underlies mycobacterial disease. J Clin Invest, 2020. 130(6): p. 3158-3171. 93. de Beaucoudrey, L., et al., Revisiting human IL-12Rbeta1 deficiency: a survey of 141 patients from 30 countries. Medicine (Baltimore), 2010. 89(6): p. 381-402. 94. Altare, F., et al., Inherited interleukin 12 deficiency in a child with bacille Calmette- Guerin and Salmonella enteritidis disseminated infection. J Clin Invest, 1998. 102(12): p. 2035-40. 128
95. Frucht, D.M., et al., IL-12-Independent costimulation pathways for interferon-gamma production in familial disseminated Mycobacterium avium complex infection. Clin Immunol, 1999. 91(2): p. 234-41. 96. Jouanguy, E., et al., IL-12 and IFN-gamma in host defense against mycobacteria and salmonella in mice and men. Curr Opin Immunol, 1999. 11(3): p. 346-51. 97. van de Vosse, E., et al., Genetic variations in the interleukin-12/interleukin-23 receptor (beta1) chain, and implications for IL-12 and IL-23 receptor structure and function. Immunogenetics, 2003. 54(12): p. 817-29. 98. Fieschi, C., et al., Low penetrance, broad resistance, and favorable outcome of interleukin 12 receptor beta1 deficiency: medical and immunological implications. J Exp Med, 2003. 197(4): p. 527-35. 99. van de Vosse, E., et al., IL-12Rbeta1 deficiency: mutation update and description of the IL12RB1 variation database. Hum Mutat, 2013. 34(10): p. 1329-39. 100. Browne, S.K., et al., Adult-onset immunodeficiency in Thailand and Taiwan. N Engl J Med, 2012. 367(8): p. 725-34. 101. Browne, S.K. and S.M. Holland, Anticytokine autoantibodies in infectious diseases: pathogenesis and mechanisms. Lancet Infect Dis, 2010. 10(12): p. 875-85. 102. O'Connell, E., et al., The first US domestic report of disseminated Mycobacterium avium complex and anti-interferon-gamma autoantibodies. J Clin Immunol, 2014. 34(8): p. 928- 32. 103. Patel, S.Y., et al., Anti-IFN-gamma autoantibodies in disseminated nontuberculous mycobacterial infections. J Immunol, 2005. 175(7): p. 4769-76. 104. Ku, C.L., et al., Anti-IFN-gamma autoantibodies are strongly associated with HLA- DR*15:02/16:02 and HLA-DQ*05:01/05:02 across Southeast Asia. J Allergy Clin Immunol, 2016. 137(3): p. 945-8 e8. 105. Sim, B.T., et al., Recurrent Burkholderia gladioli suppurative lymphadenitis associated with neutralizing anti-IL-12p70 autoantibodies. J Clin Immunol, 2013. 33(6): p. 1057-61. 106. Browne, S.K., et al., Anti-CD20 (rituximab) therapy for anti-IFN-gamma autoantibody- associated nontuberculous mycobacterial infection. Blood, 2012. 119(17): p. 3933-9. 107. Dupuis, S., et al., Impaired response to interferon-alpha/beta and lethal viral disease in human STAT1 deficiency. Nat Genet, 2003. 33(3): p. 388-91. 108. Averbuch, D., et al., The clinical spectrum of patients with deficiency of Signal Transducer and Activator of Transcription-1. Pediatr Infect Dis J, 2011. 30(4): p. 352-5. 109. Dupuis, S., et al., Impairment of mycobacterial but not viral immunity by a germline human STAT1 mutation. Science, 2001. 293(5528): p. 300-3. 110. Liu, L., et al., Gain-of-function human STAT1 mutations impair IL-17 immunity and underlie chronic mucocutaneous candidiasis. J Exp Med, 2011. 208(8): p. 1635-48. 111. Break, T.J., et al., Aberrant type 1 immunity drives susceptibility to mucosal fungal infections. Science, 2021. 371(6526). 112. Shin, J.H., et al., (1)H NMR-based metabolomic profiling in mice infected with Mycobacterium tuberculosis. J Proteome Res, 2011. 10(5): p. 2238-47. 113. Strelko, C.L., et al., Itaconic acid is a mammalian metabolite induced during macrophage activation. J Am Chem Soc, 2011. 133(41): p. 16386-9. 114. Michelucci, A., et al., Immune-responsive gene 1 protein links metabolism to immunity by catalyzing itaconic acid production. Proc Natl Acad Sci U S A, 2013. 110(19): p. 7820-5. 129
115. Lee, C.G., et al., Cloning and analysis of gene regulation of a novel LPS-inducible cDNA. Immunogenetics, 1995. 41(5): p. 263-70. 116. Sapcariu, S.C., et al., Simultaneous extraction of proteins and metabolites from cells in culture. MethodsX, 2014. 1: p. 74-80. 117. Lampropoulou, V., et al., Itaconate Links Inhibition of Succinate Dehydrogenase with Macrophage Metabolic Remodeling and Regulation of Inflammation. Cell Metab, 2016. 24(1): p. 158-66. 118. Basagoudanavar, S.H., et al., Distinct roles for the NF-kappa B RelA subunit during antiviral innate immune responses. J Virol, 2011. 85(6): p. 2599-610. 119. Degrandi, D., et al., The proinflammatory cytokine-induced IRG1 protein associates with mitochondria. J Interferon Cytokine Res, 2009. 29(1): p. 55-67. 120. Hall, C.J., et al., Immunoresponsive gene 1 augments bactericidal activity of macrophage-lineage cells by regulating beta-oxidation-dependent mitochondrial ROS production. Cell Metab, 2013. 18(2): p. 265-78. 121. Nair, S., et al., Irg1 expression in myeloid cells prevents immunopathology during M. tuberculosis infection. J Exp Med, 2018. 215(4): p. 1035-1045. 122. Williams, J.O., T.E. Roche, and B.A. McFadden, Mechanism of action of isocitrate lyase from Pseudomonas indigofera. Biochemistry, 1971. 10(8): p. 1384-90. 123. Rittenhouse, J.W. and B.A. McFadden, Inhibition of isocitrate lyase from Pseudomonas indigofera by itaconate. Arch Biochem Biophys, 1974. 163(1): p. 79-86. 124. McFadden, B.A. and S. Purohit, Itaconate, an isocitrate lyase-directed inhibitor in Pseudomonas indigofera. J Bacteriol, 1977. 131(1): p. 136-44. 125. McKinney, J.D., et al., Persistence of Mycobacterium tuberculosis in macrophages and mice requires the glyoxylate shunt enzyme isocitrate lyase. Nature, 2000. 406(6797): p. 735-8. 126. Serafini, A., et al., Mycobacterium tuberculosis requires glyoxylate shunt and reverse methylcitrate cycle for lactate and pyruvate metabolism. Mol Microbiol, 2019. 112(4): p. 1284-1307. 127. Naujoks, J., et al., IFNs Modify the Proteome of Legionella-Containing Vacuoles and Restrict Infection Via IRG1-Derived Itaconic Acid. PLoS Pathog, 2016. 12(2): p. e1005408. 128. Berg, I.A., L.V. Filatova, and R.N. Ivanovsky, Inhibition of acetate and propionate assimilation by itaconate via propionyl-CoA carboxylase in isocitrate lyase-negative purple bacterium Rhodospirillum rubrum. FEMS Microbiol Lett, 2002. 216(1): p. 49-54. 129. Chen, M., et al., Itaconate is an effector of a Rab GTPase cell-autonomous host defense pathway against Salmonella. Science, 2020. 369(6502): p. 450-455. 130. Sasikaran, J., et al., Bacterial itaconate degradation promotes pathogenicity. Nat Chem Biol, 2014. 10(5): p. 371-7. 131. Riquelme, S.A., et al., Pseudomonas aeruginosa Utilizes Host-Derived Itaconate to Redirect Its Metabolism to Promote Biofilm Formation. Cell Metab, 2020. 31(6): p. 1091-1106 e6. 132. Tomlinson, K.L., et al., Staphylococcus aureus induces an itaconate-dominated immunometabolic response that drives biofilm formation. Nat Commun, 2021. 12(1): p. 1399. 133. Tannahill, G.M., et al., Succinate is an inflammatory signal that induces IL-1beta through HIF-1alpha. Nature, 2013. 496(7444): p. 238-42. 130
134. Daniels, B.P., et al., The Nucleotide Sensor ZBP1 and Kinase RIPK3 Induce the Enzyme IRG1 to Promote an Antiviral Metabolic State in Neurons. Immunity, 2019. 50(1): p. 64- 76 e4. 135. Liao, S.T., et al., 4-Octyl itaconate inhibits aerobic glycolysis by targeting GAPDH to exert anti-inflammatory effects. Nat Commun, 2019. 10(1): p. 5091. 136. Mills, E.L., et al., Itaconate is an anti-inflammatory metabolite that activates Nrf2 via alkylation of KEAP1. Nature, 2018. 556(7699): p. 113-117. 137. Kobayashi, A., et al., Oxidative and electrophilic stresses activate Nrf2 through inhibition of ubiquitination activity of Keap1. Mol Cell Biol, 2006. 26(1): p. 221-9. 138. ElAzzouny, M., et al., Dimethyl Itaconate Is Not Metabolized into Itaconate Intracellularly. J Biol Chem, 2017. 292(12): p. 4766-4769. 139. Tang, C., et al., 4-Octyl Itaconate Activates Nrf2 Signaling to Inhibit Pro-Inflammatory Cytokine Production in Peripheral Blood Mononuclear Cells of Systemic Lupus Erythematosus Patients. Cell Physiol Biochem, 2018. 51(2): p. 979-990. 140. Liu, H., et al., Four-octyl itaconate activates Keap1-Nrf2 signaling to protect neuronal cells from hydrogen peroxide. Cell Commun Signal, 2018. 16(1): p. 81. 141. Bambouskova, M., et al., Electrophilic properties of itaconate and derivatives regulate the IkappaBzeta-ATF3 inflammatory axis. Nature, 2018. 556(7702): p. 501-504. 142. Olagnier, D., et al., SARS-CoV2-mediated suppression of NRF2-signaling reveals potent antiviral and anti-inflammatory activity of 4-octyl-itaconate and dimethyl fumarate. Nat Commun, 2020. 11(1): p. 4938. 143. Olagnier, D., et al., Nrf2 negatively regulates STING indicating a link between antiviral sensing and metabolic reprogramming. Nat Commun, 2018. 9(1): p. 3506. 144. Cordes, T., et al., Itaconate modulates tricarboxylic acid and redox metabolism to mitigate reperfusion injury. Mol Metab, 2020. 32: p. 122-135. 145. Yi, Z., et al., Immune-Responsive Gene 1/Itaconate Activates Nuclear Factor Erythroid 2-Related Factor 2 in Hepatocytes to Protect Against Liver Ischemia-Reperfusion Injury. Hepatology, 2020. 72(4): p. 1394-1411. 146. Mantovani, A., et al., The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol, 2004. 25(12): p. 677-86. 147. Martinez, F.O. and S. Gordon, The M1 and M2 paradigm of macrophage activation: time for reassessment. F1000Prime Rep, 2014. 6: p. 13. 148. Van den Bossche, J. and G.J.W. van der Windt, Fatty Acid Oxidation in Macrophages and T Cells: Time for Reassessment? Cell Metab, 2018. 28(4): p. 538-540. 149. Van den Bossche, J., L.A. O'Neill, and D. Menon, Macrophage Immunometabolism: Where Are We (Going)? Trends Immunol, 2017. 38(6): p. 395-406. 150. Palsson-McDermott, E.M. and L.A. O'Neill, The Warburg effect then and now: from cancer to inflammatory diseases. Bioessays, 2013. 35(11): p. 965-73. 151. Nelson, V.L., et al., PPARgamma is a nexus controlling alternative activation of macrophages via glutamine metabolism. Genes Dev, 2018. 32(15-16): p. 1035-1044. 152. Ganta, V.C., et al., A MicroRNA93-Interferon Regulatory Factor-9-Immunoresponsive Gene-1-Itaconic Acid Pathway Modulates M2-Like Macrophage Polarization to Revascularize Ischemic Muscle. Circulation, 2017. 135(24): p. 2403-2425. 153. Cordes, T., et al., Immunoresponsive Gene 1 and Itaconate Inhibit Succinate Dehydrogenase to Modulate Intracellular Succinate Levels. J Biol Chem, 2016. 291(27): p. 14274-14284. 131
154. Cordes, T. and C.M. Metallo, Itaconate Alters Succinate and Coenzyme A Metabolism via Inhibition of Mitochondrial Complex II and Methylmalonyl-CoA Mutase. Metabolites, 2021. 11(2). 155. He, W., et al., Complexity of macrophage metabolism in infection. Curr Opin Biotechnol, 2021. 68: p. 231-239. 156. Krawczyk, C.M., et al., Toll-like receptor-induced changes in glycolytic metabolism regulate dendritic cell activation. Blood, 2010. 115(23): p. 4742-9. 157. Nishi, K., et al., LPS induces hypoxia-inducible factor 1 activation in macrophage- differentiated cells in a reactive oxygen species-dependent manner. Antioxid Redox Signal, 2008. 10(5): p. 983-95. 158. Chandel, N.S., et al., Reactive oxygen species generated at mitochondrial complex III stabilize hypoxia-inducible factor-1alpha during hypoxia: a mechanism of O2 sensing. J Biol Chem, 2000. 275(33): p. 25130-8. 159. Blouin, C.C., et al., Hypoxic gene activation by lipopolysaccharide in macrophages: implication of hypoxia-inducible factor 1alpha. Blood, 2004. 103(3): p. 1124-30. 160. Semenza, G.L., et al., Hypoxia response elements in the aldolase A, enolase 1, and lactate dehydrogenase A gene promoters contain essential binding sites for hypoxia- inducible factor 1. J Biol Chem, 1996. 271(51): p. 32529-37. 161. Braverman, J., et al., HIF-1alpha Is an Essential Mediator of IFN-gamma-Dependent Immunity to Mycobacterium tuberculosis. J Immunol, 2016. 197(4): p. 1287-97. 162. Peyssonnaux, C., et al., HIF-1alpha expression regulates the bactericidal capacity of phagocytes. J Clin Invest, 2005. 115(7): p. 1806-15. 163. Takeda, N., et al., Differential activation and antagonistic function of HIF-{alpha} isoforms in macrophages are essential for NO homeostasis. Genes Dev, 2010. 24(5): p. 491-501. 164. Vazquez, A., et al., Catabolic efficiency of aerobic glycolysis: the Warburg effect revisited. BMC Syst Biol, 2010. 4: p. 58. 165. Nagy, C. and A. Haschemi, Time and Demand are Two Critical Dimensions of Immunometabolism: The Process of Macrophage Activation and the Pentose Phosphate Pathway. Front Immunol, 2015. 6: p. 164. 166. Jain, M., et al., Glucose-6-phosphate dehydrogenase modulates cytosolic redox status and contractile phenotype in adult cardiomyocytes. Circ Res, 2003. 93(2): p. e9-16. 167. Ruiz-Garcia, A., et al., Cooperation of adenosine with macrophage Toll-4 receptor agonists leads to increased glycolytic flux through the enhanced expression of PFKFB3 gene. J Biol Chem, 2011. 286(22): p. 19247-58. 168. Belikoff, B.G., et al., A2B adenosine receptor blockade enhances macrophage-mediated bacterial phagocytosis and improves polymicrobial sepsis survival in mice. J Immunol, 2011. 186(4): p. 2444-53. 169. O'Neill, L.A., A broken krebs cycle in macrophages. Immunity, 2015. 42(3): p. 393-4. 170. Clementi, E., et al., Persistent inhibition of cell respiration by nitric oxide: crucial role of S-nitrosylation of mitochondrial complex I and protective action of glutathione. Proc Natl Acad Sci U S A, 1998. 95(13): p. 7631-6. 171. Drapier, J.C. and J.B. Hibbs, Jr., Differentiation of murine macrophages to express nonspecific cytotoxicity for tumor cells results in L-arginine-dependent inhibition of mitochondrial iron-sulfur enzymes in the macrophage effector cells. J Immunol, 1988. 140(8): p. 2829-38. 132
172. Lachmandas, E., et al., Microbial stimulation of different Toll-like receptor signalling pathways induces diverse metabolic programmes in human monocytes. Nat Microbiol, 2016. 2: p. 16246. 173. Galli, G. and M. Saleh, Immunometabolism of Macrophages in Bacterial Infections. Front Cell Infect Microbiol, 2020. 10: p. 607650. 174. Casanova, J.L. and L. Abel, Genetic dissection of immunity to mycobacteria: the human model. Annu Rev Immunol, 2002. 20: p. 581-620. 175. Rosenzweig, S.D. and S.M. Holland, Congenital defects in the interferon- gamma/interleukin-12 pathway. Curr Opin Pediatr, 2004. 16(1): p. 3-8. 176. Holland, S.M., Immune deficiency presenting as mycobacterial infection. Clin Rev Allergy Immunol, 2001. 20(1): p. 121-37. 177. Doherty, T.M. and A. Sher, Defects in cell-mediated immunity affect chronic, but not innate, resistance of mice to Mycobacterium avium infection. J Immunol, 1997. 158(10): p. 4822-31. 178. Pearl, J.E., et al., Inflammation and lymphocyte activation during mycobacterial infection in the interferon-gamma-deficient mouse. Cell Immunol, 2001. 211(1): p. 43-50. 179. van den Broek, M.F., et al., Immune defence in mice lacking type I and/or type II interferon receptors. Immunol Rev, 1995. 148: p. 5-18. 180. Hoffmann, E., Machelart, A., Belhaouane, I., Deboosere, N., Pauwels, A.-M., Saint- André, J.-P., Song, O.-R., Jouny, S., Poncet, A., Marion, S., et al. , IRG1 controls immunometabolic host response and restricts intracellular Mycobacterium tuberculosis infection. BioRxiv, 2019. 181. Appelberg, R., Protective role of interferon gamma, tumor necrosis factor alpha and interleukin-6 in Mycobacterium tuberculosis and M. avium infections. Immunobiology, 1994. 191(4-5): p. 520-5. 182. Cooper, A.M., R. Appelberg, and I.M. Orme, Immunopathogenesis of Mycobacterium avium infection. Front Biosci, 1998. 3: p. e141-8. 183. Ballarino, G.J., et al., Pulmonary nontuberculous mycobacterial infections: antibiotic treatment and associated costs. Respir Med, 2009. 103(10): p. 1448-55. 184. Aguilar, C., et al., Prevention of infections during primary immunodeficiency. Clin Infect Dis, 2014. 59(10): p. 1462-70. 185. Bonilla, F.A. and R.S. Geha, 12. Primary immunodeficiency diseases. J Allergy Clin Immunol, 2003. 111(2 Suppl): p. S571-81. 186. Holland, S.M., Cytokine therapy of mycobacterial infections. Adv Intern Med, 2000. 45: p. 431-52. 187. Holland, S.M., Therapy of mycobacterial infections. Res Immunol, 1996. 147(8-9): p. 572-81. 188. Jha, A.K., et al., Network integration of parallel metabolic and transcriptional data reveals metabolic modules that regulate macrophage polarization. Immunity, 2015. 42(3): p. 419-30. 189. Murphy, M.P. and L.A.J. O'Neill, Krebs Cycle Reimagined: The Emerging Roles of Succinate and Itaconate as Signal Transducers. Cell, 2018. 174(4): p. 780-784. 190. Ottenhoff, T.H., et al., Control of human host immunity to mycobacteria. Tuberculosis (Edinb), 2005. 85(1-2): p. 53-64. 191. Gallin, J.I., et al., Interferon-gamma in the management of infectious diseases. Ann Intern Med, 1995. 123(3): p. 216-24. 133
192. Jessop, F., et al., Interferon Gamma Reprograms Host Mitochondrial Metabolism through Inhibition of Complex II To Control Intracellular Bacterial Replication. Infect Immun, 2020. 88(2). 193. Abdissa, K., et al., Relevance of inducible nitric oxide synthase for immune control of Mycobacterium avium subspecies paratuberculosis infection in mice. Virulence, 2020. 11(1): p. 465-481. 194. Rojas, M., et al., Differential induction of apoptosis by virulent Mycobacterium tuberculosis in resistant and susceptible murine macrophages: role of nitric oxide and mycobacterial products. J Immunol, 1997. 159(3): p. 1352-61. 195. Brosch, R., et al., Comparative genomics of the mycobacteria. Int J Med Microbiol, 2000. 290(2): p. 143-52. 196. Resende, M., et al., TNF-Mediated Compensatory Immunity to Mycobacterium avium in the Absence of Macrophage Activation by IFN-gamma. J Immunol, 2019. 203(9): p. 2451-2458. 197. Price, J.V., et al., IRG1 and Inducible Nitric Oxide Synthase Act Redundantly with Other Interferon-Gamma-Induced Factors To Restrict Intracellular Replication of Legionella pneumophila. mBio, 2019. 10(6). 198. Chen, F., et al., Crystal structure of cis-aconitate decarboxylase reveals the impact of naturally occurring human mutations on itaconate synthesis. Proc Natl Acad Sci U S A, 2019. 116(41): p. 20644-20654. 199. Bhat, M.Y., et al., Comprehensive network map of interferon gamma signaling. J Cell Commun Signal, 2018. 12(4): p. 745-751. 200. Jordan, M.B., et al., An animal model of hemophagocytic lymphohistiocytosis (HLH): CD8+ T cells and interferon gamma are essential for the disorder. Blood, 2004. 104(3): p. 735-43. 201. Nau, G.J., et al., Human macrophage activation programs induced by bacterial pathogens. Proc Natl Acad Sci U S A, 2002. 99(3): p. 1503-8. 202. Pachlopnik Schmid, J., et al., Neutralization of IFNgamma defeats haemophagocytosis in LCMV-infected perforin- and Rab27a-deficient mice. EMBO Mol Med, 2009. 1(2): p. 112-24. 203. Pollard, K.M., et al., Interferon-gamma and systemic autoimmunity. Discov Med, 2013. 16(87): p. 123-31. 204. Cheng, S.C., et al., mTOR- and HIF-1alpha-mediated aerobic glycolysis as metabolic basis for trained immunity. Science, 2014. 345(6204): p. 1250684. 205. McGettrick, A.F. and L.A. O'Neill, How metabolism generates signals during innate immunity and inflammation. J Biol Chem, 2013. 288(32): p. 22893-8. 206. Fukuzumi, M., et al., Endotoxin-induced enhancement of glucose influx into murine peritoneal macrophages via GLUT1. Infect Immun, 1996. 64(1): p. 108-12. 207. Cameron, A.M., et al., Inflammatory macrophage dependence on NAD(+) salvage is a consequence of reactive oxygen species-mediated DNA damage. Nat Immunol, 2019. 20(4): p. 420-432. 208. Shadel, G.S. and T.L. Horvath, Mitochondrial ROS signaling in organismal homeostasis. Cell, 2015. 163(3): p. 560-9. 209. Bedard, K. and K.H. Krause, The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev, 2007. 87(1): p. 245-313. 134
210. Yang, Y. and A.A. Sauve, NAD(+) metabolism: Bioenergetics, signaling and manipulation for therapy. Biochim Biophys Acta, 2016. 1864(12): p. 1787-1800. 211. Rodriguez-Prados, J.C., et al., Substrate fate in activated macrophages: a comparison between innate, classic, and alternative activation. J Immunol, 2010. 185(1): p. 605-14. 212. Wu, D., et al., Type 1 Interferons Induce Changes in Core Metabolism that Are Critical for Immune Function. Immunity, 2016. 44(6): p. 1325-36. 213. Overbeek, R., et al., The ERGO genome analysis and discovery system. Nucleic Acids Res, 2003. 31(1): p. 164-71. 214. Xiao, W., et al., NAD(H) and NADP(H) Redox Couples and Cellular Energy Metabolism. Antioxid Redox Signal, 2018. 28(3): p. 251-272. 215. Yang, H., et al., Nutrient-sensitive mitochondrial NAD+ levels dictate cell survival. Cell, 2007. 130(6): p. 1095-107. 216. Banoth, B. and S.L. Cassel, Mitochondria in innate immune signaling. Transl Res, 2018. 202: p. 52-68. 217. Baradaran, R., et al., Crystal structure of the entire respiratory complex I. Nature, 2013. 494(7438): p. 443-8. 218. Barrientos, A. and C.T. Moraes, Titrating the effects of mitochondrial complex I impairment in the cell physiology. J Biol Chem, 1999. 274(23): p. 16188-97. 219. Ferguson, S.J., Similarities between mitochondrial and bacterial electron transport with particular reference to the action of inhibitors. Biochem Soc Trans, 1994. 22(1): p. 181- 3. 220. Garaude, J., et al., Mitochondrial respiratory-chain adaptations in macrophages contribute to antibacterial host defense. Nat Immunol, 2016. 17(9): p. 1037-1045. 221. Hatefi, Y., The mitochondrial electron transport and oxidative phosphorylation system. Annu Rev Biochem, 1985. 54: p. 1015-69. 222. Kiritsy, M.C., Mott, D., Behar, S.M., Sassetti, C.M., and Olive, A.J., Mitochondrial respiration contributes to the interferon gamma response in antigen presenting cells. BioRxiv, 2020. 223. Brand, M.D., Mitochondrial generation of superoxide and hydrogen peroxide as the source of mitochondrial redox signaling. Free Radic Biol Med, 2016. 100: p. 14-31. 224. Kapnick, S.M., S.E. Pacheco, and P.J. McGuire, The emerging role of immune dysfunction in mitochondrial diseases as a paradigm for understanding immunometabolism. Metabolism, 2018. 81: p. 97-112. 225. McGuire, P.J., Mitochondrial Dysfunction and the Aging Immune System. Biology (Basel), 2019. 8(2). 226. Ruhoy, I.S. and R.P. Saneto, The genetics of Leigh syndrome and its implications for clinical practice and risk management. Appl Clin Genet, 2014. 7: p. 221-34. 227. Lenaz, G., et al., Bioenergetics of mitochondrial diseases associated with mtDNA mutations. Biochim Biophys Acta, 2004. 1658(1-2): p. 89-94. 228. Stewart, J.B. and P.F. Chinnery, The dynamics of mitochondrial DNA heteroplasmy: implications for human health and disease. Nat Rev Genet, 2015. 16(9): p. 530-42. 229. Zhang, R., MNADK, a novel liver-enriched mitochondrion-localized NAD kinase. Biol Open, 2013. 2(4): p. 432-8. 230. Gameiro, P.A., et al., Cofactor balance by nicotinamide nucleotide transhydrogenase (NNT) coordinates reductive carboxylation and glucose catabolism in the tricarboxylic acid (TCA) cycle. J Biol Chem, 2013. 288(18): p. 12967-77. 135
231. Meimaridou, E., et al., Mutations in NNT encoding nicotinamide nucleotide transhydrogenase cause familial glucocorticoid deficiency. Nat Genet, 2012. 44(7): p. 740-2. 232. Okura, Y., et al., Monocyte/macrophage-specific NADPH oxidase contributes to antimicrobial host defense in X-CGD. J Clin Immunol, 2015. 35(2): p. 158-67. 233. Roos, D., et al., Hematologically important mutations: X-linked chronic granulomatous disease (third update). Blood Cells Mol Dis, 2010. 45(3): p. 246-65. 234. Segal, B.H., et al., Genetic, biochemical, and clinical features of chronic granulomatous disease. Medicine (Baltimore), 2000. 79(3): p. 170-200. 235. O'Donnell, B.V., et al., Studies on the inhibitory mechanism of iodonium compounds with special reference to neutrophil NADPH oxidase. Biochem J, 1993. 290 ( Pt 1): p. 41-9. 236. Zavadskis, S., et al., Effect of Diphenyleneiodonium Chloride on Intracellular Reactive Oxygen Species Metabolism with Emphasis on NADPH Oxidase and Mitochondria in Two Therapeutically Relevant Human Cell Types. Pharmaceutics, 2020. 13(1). 237. Aldieri, E., et al., Classical inhibitors of NOX NAD(P)H oxidases are not specific. Curr Drug Metab, 2008. 9(8): p. 686-96. 238. Majander, A., M. Finel, and M. Wikstrom, Diphenyleneiodonium inhibits reduction of iron-sulfur clusters in the mitochondrial NADH-ubiquinone oxidoreductase (Complex I). J Biol Chem, 1994. 269(33): p. 21037-42. 239. Gallin, J.I., Interferon-gamma in the treatment of the chronic granulomatous diseases of childhood. Clin Immunol Immunopathol, 1991. 61(2 Pt 2): p. S100-5. 240. Sechler, J.M., et al., Recombinant human interferon-gamma reconstitutes defective phagocyte function in patients with chronic granulomatous disease of childhood. Proc Natl Acad Sci U S A, 1988. 85(13): p. 4874-8. 241. Jastroch, M., et al., Mitochondrial proton and electron leaks. Essays Biochem, 2010. 47: p. 53-67. 242. Ruas, J.S., et al., Underestimation of the Maximal Capacity of the Mitochondrial Electron Transport System in Oligomycin-Treated Cells. PLoS One, 2016. 11(3): p. e0150967. 243. Smolina, N., et al., Assaying Mitochondrial Respiration as an Indicator of Cellular Metabolism and Fitness. Methods Mol Biol, 2017. 1601: p. 79-87. 244. Huffaker, T.B., et al., A Stat1 bound enhancer promotes Nampt expression and function within tumor associated macrophages. Nat Commun, 2021. 12(1): p. 2620. 245. Handy, D.E., et al., Glutathione peroxidase-1 regulates mitochondrial function to modulate redox-dependent cellular responses. J Biol Chem, 2009. 284(18): p. 11913-21. 246. Lubos, E., et al., Glutathione peroxidase-1 deficiency augments proinflammatory cytokine-induced redox signaling and human endothelial cell activation. J Biol Chem, 2011. 286(41): p. 35407-35417. 247. Zhang, Z., et al., Increasing glucose 6-phosphate dehydrogenase activity restores redox balance in vascular endothelial cells exposed to high glucose. PLoS One, 2012. 7(11): p. e49128. 248. Oka, S.I., et al., Nampt Potentiates Antioxidant Defense in Diabetic Cardiomyopathy. Circ Res, 2021. 249. Agledal, L., M. Niere, and M. Ziegler, The phosphate makes a difference: cellular functions of NADP. Redox Rep, 2010. 15(1): p. 2-10. 136
250. Sander, L.E. and J. Garaude, The mitochondrial respiratory chain: A metabolic rheostat of innate immune cell-mediated antibacterial responses. Mitochondrion, 2018. 41: p. 28- 36. 251. Shekhova, E., Mitochondrial reactive oxygen species as major effectors of antimicrobial immunity. PLoS Pathog, 2020. 16(5): p. e1008470. 252. Dikalov, S., Cross talk between mitochondria and NADPH oxidases. Free Radic Biol Med, 2011. 51(7): p. 1289-301. 253. Sibley, C.T., et al., Assessment of atherosclerosis in chronic granulomatous disease. Circulation, 2014. 130(23): p. 2031-9. 254. Violi, F., et al., NADPH Oxidase-2 and Atherothrombosis: Insight From Chronic Granulomatous Disease. Arterioscler Thromb Vasc Biol, 2017. 37(2): p. 218-225. 255. Kuai, Y., et al., An ultralow dose of the NADPH oxidase inhibitor diphenyleneiodonium (DPI) is an economical and effective therapeutic agent for the treatment of colitis- associated colorectal cancer. Theranostics, 2020. 10(15): p. 6743-6757. 256. Lambert, A.J., et al., Diphenyleneiodonium acutely inhibits reactive oxygen species production by mitochondrial complex I during reverse, but not forward electron transport. Biochim Biophys Acta, 2008. 1777(5): p. 397-403. 257. Nagel, S., et al., Suppression of the inflammatory response by diphenyleneiodonium after transient focal cerebral ischemia. J Neurochem, 2012. 123 Suppl 2: p. 98-107. 258. Wang, Q., et al., Subpicomolar diphenyleneiodonium inhibits microglial NADPH oxidase with high specificity and shows great potential as a therapeutic agent for neurodegenerative diseases. Glia, 2014. 62(12): p. 2034-43. 259. Wang, Q., et al., Post-treatment with an ultra-low dose of NADPH oxidase inhibitor diphenyleneiodonium attenuates disease progression in multiple Parkinson's disease models. Brain, 2015. 138(Pt 5): p. 1247-62. 260. Gardell, S.J., et al., Boosting NAD(+) with a small molecule that activates NAMPT. Nat Commun, 2019. 10(1): p. 3241. 261. Lu, H., et al., P2X7 receptors mediate NADH transport across the plasma membranes of astrocytes. Biochem Biophys Res Commun, 2007. 362(4): p. 946-50. 262. Di Virgilio, F., et al., The P2X7 Receptor in Infection and Inflammation. Immunity, 2017. 47(1): p. 15-31. 263. Kastritis, E. and M.A. Dimopoulos, Daratumumab combinations: what can we learn? Blood, 2017. 130(8): p. 957-958. 264. Kimmich, C.R., et al., Daratumumab for systemic AL amyloidosis: prognostic factors and adverse outcome with nephrotic-range albuminuria. Blood, 2020. 135(18): p. 1517- 1530. 265. Plesner, T., et al., Phase 1/2 study of daratumumab, lenalidomide, and dexamethasone for relapsed multiple myeloma. Blood, 2016. 128(14): p. 1821-1828. 266. Boehm, U., et al., Cellular responses to interferon-gamma. Annu Rev Immunol, 1997. 15: p. 749-95. 267. Jurkovich, G.J., et al., Interferon gamma increases sensitivity to endotoxin. J Surg Res, 1991. 51(3): p. 197-203. 268. Williams, J.G., et al., Macrophage priming by interferon gamma: a selective process with potentially harmful effects. J Leukoc Biol, 1992. 52(6): p. 579-84. 269. Sweet, M.J., et al., IFN-gamma primes macrophage responses to bacterial DNA. J Interferon Cytokine Res, 1998. 18(4): p. 263-71. 137
270. Notarangelo, L.D., et al., Human inborn errors of immunity: An expanding universe. Sci Immunol, 2020. 5(49). 271. Holland, S.M., Interferon gamma, IL-12, IL-12R and STAT-1 immunodeficiency diseases: disorders of the interface of innate and adaptive immunity. Immunol Res, 2007. 38(1-3): p. 342-6. 272. Diskin, C. and E.M. Palsson-McDermott, Metabolic Modulation in Macrophage Effector Function. Front Immunol, 2018. 9: p. 270. 273. Berton, G., et al., Molecular basis of macrophage activation. Expression of the low potential cytochrome b and its reduction upon cell stimulation in activated macrophages. J Immunol, 1986. 136(4): p. 1393-9. 274. Cassatella, M.A., et al., Interferon-gamma transcriptionally modulates the expression of the genes for the high affinity IgG-Fc receptor and the 47-kDa cytosolic component of NADPH oxidase in human polymorphonuclear leukocytes. J Biol Chem, 1991. 266(33): p. 22079-82. 275. Holland, S.M., et al., Abnormal regulation of interferon-gamma, interleukin-12, and tumor necrosis factor-alpha in human interferon-gamma receptor 1 deficiency. J Infect Dis, 1998. 178(4): p. 1095-104. 276. Li, Y., et al., Immune responsive gene 1 (IRG1) promotes endotoxin tolerance by increasing A20 expression in macrophages through reactive oxygen species. J Biol Chem, 2013. 288(23): p. 16225-16234. 277. Daly, R., et al., Changes in Plasma Itaconate Elevation in Early Rheumatoid Arthritis Patients Elucidates Disease Activity Associated Macrophage Activation. Metabolites, 2020. 10(6). 278. Song, J.W., et al., Omics-Driven Systems Interrogation of Metabolic Dysregulation in COVID-19 Pathogenesis. Cell Metab, 2020. 32(2): p. 188-202 e5. 279. Hooftman, A., et al., The Immunomodulatory Metabolite Itaconate Modifies NLRP3 and Inhibits Inflammasome Activation. Cell Metab, 2020. 32(3): p. 468-478 e7. 280. Bambouskova, M., et al., Itaconate confers tolerance to late NLRP3 inflammasome activation. Cell Rep, 2021. 34(10): p. 108756. 281. Nakajima, M., et al., Nrf2 Regulates Granuloma Formation and Macrophage Activation during Mycobacterium avium Infection via Mediating Nramp1 and HO-1 Expressions. mBio, 2021. 12(1). 282. Amaral, E.P., et al., A major role for ferroptosis in Mycobacterium tuberculosis-induced cell death and tissue necrosis. J Exp Med, 2019. 216(3): p. 556-570. 283. Amaral, E.P., et al., N-acetyl-cysteine exhibits potent anti-mycobacterial activity in addition to its known anti-oxidative functions. BMC Microbiol, 2016. 16(1): p. 251. 284. Jorgensen, I., M. Rayamajhi, and E.A. Miao, Programmed cell death as a defence against infection. Nat Rev Immunol, 2017. 17(3): p. 151-164. 285. Martin, C.J., et al., Efferocytosis is an innate antibacterial mechanism. Cell Host Microbe, 2012. 12(3): p. 289-300. 286. Palmieri, E.M., et al., Nitric oxide orchestrates metabolic rewiring in M1 macrophages by targeting aconitase 2 and pyruvate dehydrogenase. Nat Commun, 2020. 11(1): p. 698. 287. Palmieri, E.M., et al., Nitric Oxide in Macrophage Immunometabolism: Hiding in Plain Sight. Metabolites, 2020. 10(11). 138
288. Somasundaram, V., et al., Inducible nitric oxide synthase-derived extracellular nitric oxide flux regulates proinflammatory responses at the single cell level. Redox Biol, 2020. 28: p. 101354. 289. Zhang, H., et al., CRISPR/Cas9-Mediated Gene Editing in Human iPSC-Derived Macrophage Reveals Lysosomal Acid Lipase Function in Human Macrophages-Brief Report. Arterioscler Thromb Vasc Biol, 2017. 37(11): p. 2156-2160. 290. Sampaio, E.P., et al., A novel STAT1 mutation associated with disseminated mycobacterial disease. J Clin Immunol, 2012. 32(4): p. 681-689. 291. Toubiana, J., et al., Heterozygous STAT1 gain-of-function mutations underlie an unexpectedly broad clinical phenotype. Blood, 2016. 127(25): p. 3154-64. 292. Zimmerman, O., et al., STAT1 Gain-of-Function Mutations Cause High Total STAT1 Levels With Normal Dephosphorylation. Front Immunol, 2019. 10: p. 1433. 293. Divakaruni, A.S., et al., Etomoxir Inhibits Macrophage Polarization by Disrupting CoA Homeostasis. Cell Metab, 2018. 28(3): p. 490-503 e7. 294. Kabat, A.M. and E.J. Pearce, Inflammation by way of macrophage metabolism. Science, 2017. 356(6337): p. 488-489. 295. Wang, F., et al., Interferon Gamma Induces Reversible Metabolic Reprogramming of M1 Macrophages to Sustain Cell Viability and Pro-Inflammatory Activity. EBioMedicine, 2018. 30: p. 303-316. 296. Roger, A.J., S.A. Munoz-Gomez, and R. Kamikawa, The Origin and Diversification of Mitochondria. Curr Biol, 2017. 27(21): p. R1177-R1192. 297. Roger, A.J., Reconstructing Early Events in Eukaryotic Evolution. Am Nat, 1999. 154(S4): p. S146-S163. 298. Angelova, P.R. and A.Y. Abramov, Functional role of mitochondrial reactive oxygen species in physiology. Free Radic Biol Med, 2016. 100: p. 81-85. 299. Zhou, R., et al., A role for mitochondria in NLRP3 inflammasome activation. Nature, 2011. 469(7329): p. 221-5. 300. Fukai, T. and M. Ushio-Fukai, Cross-Talk between NADPH Oxidase and Mitochondria: Role in ROS Signaling and Angiogenesis. Cells, 2020. 9(8). 301. Handy, D.E. and J. Loscalzo, Redox regulation of mitochondrial function. Antioxid Redox Signal, 2012. 16(11): p. 1323-67. 302. Caceres, L., et al., NADPH oxidase and mitochondria are relevant sources of superoxide anion in the oxinflammatory response of macrophages exposed to airborne particulate matter. Ecotoxicol Environ Saf, 2020. 205: p. 111186. 303. Desouki, M.M., et al., Cross talk between mitochondria and superoxide generating NADPH oxidase in breast and ovarian tumors. Cancer Biol Ther, 2005. 4(12): p. 1367- 73. 304. Armstrong, J.S., et al., The redox regulation of intermediary metabolism by a superoxide- aconitase rheostat. Bioessays, 2004. 26(8): p. 894-900. 305. Jitschin, R., et al., Mitochondrial metabolism contributes to oxidative stress and reveals therapeutic targets in chronic lymphocytic leukemia. Blood, 2014. 123(17): p. 2663-72. 306. Sen, C.K., Redox signaling and the emerging therapeutic potential of thiol antioxidants. Biochem Pharmacol, 1998. 55(11): p. 1747-58. 307. Shi, Y. and Y. Shi, Metabolic enzymes and coenzymes in transcription--a direct link between metabolism and transcription? Trends Genet, 2004. 20(9): p. 445-52. 139
308. Ciccarese, F. and V. Ciminale, Escaping Death: Mitochondrial Redox Homeostasis in Cancer Cells. Front Oncol, 2017. 7: p. 117. 309. Grose, J.H., et al., Evidence that feedback inhibition of NAD kinase controls responses to oxidative stress. Proc Natl Acad Sci U S A, 2006. 103(20): p. 7601-6. 310. Fontaine, E., Metformin-Induced Mitochondrial Complex I Inhibition: Facts, Uncertainties, and Consequences. Front Endocrinol (Lausanne), 2018. 9: p. 753. 311. Boursier, G., et al., Phenotypic Associations of PSTPIP1 Sequence Variants in PSTPIP1- Associated Autoinflammatory Diseases. J Invest Dermatol, 2021. 141(5): p. 1141-1147. 312. Stone, D.L., Ombrello, A., Arostegui, J.I., Schneider, C., de Jesus, A., Girard- Guyonvarc’h, C., Gabay, C., Lee, W., Chae, J.J., Aksentijevich, I., Goldbach-Mansky, R., Kastner, D.L., Canna, S.W., Excess Serum Interleukin-18 Distinguishes Patients with Pathogenic Mutations in PSTPIP1. medRxiv, 2021. 313. Amaral, E.P., et al., Pulmonary infection with hypervirulent Mycobacteria reveals a crucial role for the P2X7 receptor in aggressive forms of tuberculosis. PLoS Pathog, 2014. 10(7): p. e1004188. 314. Haag, F., et al., Extracellular NAD and ATP: Partners in immune cell modulation. Purinergic Signal, 2007. 3(1-2): p. 71-81. 315. Holness, M.J. and M.C. Sugden, Regulation of pyruvate dehydrogenase complex activity by reversible phosphorylation. Biochem Soc Trans, 2003. 31(Pt 6): p. 1143-51. 316. Langmead, B., et al., Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol, 2009. 10(3): p. R25. 317. Kim, D., et al., TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol, 2013. 14(4): p. R36. 318. Bolstad, B.M., et al., A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics, 2003. 19(2): p. 185-93. 319. Anders, S., et al., Count-based differential expression analysis of RNA sequencing data using R and Bioconductor. Nat Protoc, 2013. 8(9): p. 1765-86. 320. Smyth, G.K., Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol, 2004. 3: p. Article3. 321. Law, C.W., et al., voom: Precision weights unlock linear model analysis tools for RNA- seq read counts. Genome Biol, 2014. 15(2): p. R29. 322. Metsalu, T. and J. Vilo, ClustVis: a web tool for visualizing clustering of multivariate data using Principal Component Analysis and heatmap. Nucleic Acids Res, 2015. 43(W1): p. W566-70. 323. Liu, X., et al., Polymorphisms in IRG1 gene associated with immune responses to hepatitis B vaccination in a Chinese Han population and function to restrain the HBV life cycle. J Med Virol, 2017. 89(7): p. 1215-1223. 324. Michopoulos, F., et al., Targeted Metabolic Profiling of the Tg197 Mouse Model Reveals Itaconic Acid as a Marker of Rheumatoid Arthritis. J Proteome Res, 2016. 15(12): p. 4579-4590. 325. Weiss, J.M., et al., Itaconic acid mediates crosstalk between macrophage metabolism and peritoneal tumors. J Clin Invest, 2018. 128(9): p. 3794-3805. 326. Basler, T., et al., Mycobacterium paratuberculosis, Mycobacterium smegmatis, and lipopolysaccharide induce different transcriptional and post-transcriptional regulation of the IRG1 gene in murine macrophages. J Leukoc Biol, 2006. 79(3): p. 628-38. 140
327. Kwai, B.X.C., et al., Itaconate is a covalent inhibitor of the Mycobacterium tuberculosis isocitrate lyase. RSC Med Chem, 2021. 12(1): p. 57-61. 328. Fang, F.C., et al., Isocitrate lyase (AceA) is required for Salmonella persistence but not for acute lethal infection in mice. Infect Immun, 2005. 73(4): p. 2547-9. 329. Lohkamp, B., et al., Three-dimensional structure of iminodisuccinate epimerase defines the fold of the MmgE/PrpD protein family. J Mol Biol, 2006. 362(3): p. 555-66. 330. Chun, H.L., et al., The crystal structure of mouse IRG1 suggests that cis-aconitate decarboxylase has an open and closed conformation. PLoS One, 2020. 15(12): p. e0242383. 331. Gregoret, L.M. and R.T. Sauer, Tolerance of a protein helix to multiple alanine and valine substitutions. Fold Des, 1998. 3(2): p. 119-26. 332. Chun, H.L., et al., Enzymatic reaction mechanism of cis-aconitate decarboxylase based on the crystal structure of IRG1 from Bacillus subtilis. Sci Rep, 2020. 10(1): p. 11305. 333. Barroso, I., et al., Dominant negative mutations in human PPARgamma associated with severe insulin resistance, diabetes mellitus and hypertension. Nature, 1999. 402(6764): p. 880-3.