Activation and Function of Interferon Regulatory Factor 5

REVIEWS

JOURNAL OF INTERFERON & CYTOKINE RESEARCH Volume 35, Number 2, 2015 ª Mary Ann Liebert, Inc. DOI: 10.1089/jir.2014.0023

Activation and Function of Interferon Regulatory Factor 5 Grigory Ryzhakov, Hayley L. Eames, and Irina A. Udalova

Interferon regulatory factor 5 (IRF5) is a crucial transcription factor in a number of immune and homeostatic processes, including host defense against pathogens, tumorigenesis, and autoimmunity. Upon induction of immune signaling pathways, IRF5 undergoes post-translational modifications such as phosphorylation and ubiqutination, which are believed to trigger IRF5 nuclear translocation from the cytosol, followed by recruitment to promoters where transcription of its gene targets is initiated. In this review, we systematically analyze the data published in the last decade on IRF5 activation, including the role of post-translational modifications and the proposed enzymes targeting IRF5 in this process. We discuss suggested models of IRF5 activation in connection to pathway-specific functions of IRF5. physical attributes of the IRF5 protein and discusses the latest insights into mechanisms of IRF5 activation, with particular emphasis on post-translational modifications.

Introduction

T

he interferon regulatory factors or IRFs constitute a family of transcription factors exclusively found in Metazoans (Nehyba and others 2009). The function of the IRFs has been mostly studied in mammals, where the 9 members of the IRF family are crucial regulators of immune cell differentiation and host defence against pathogens (Savitsky and others 2010). They have also been implicated in cancer and apoptosis (Tamura and others 2008). The IRFs are multi-domain proteins. Their amino-terminal DNA-binding domain (DBD) is involved in recognition of target DNA sequences, such as the positive regulatory domains (PRDs) I and III of the IFNb promoter, and the interferonstimulated response element (Levy and others 1988) or ISRE, with the putative binding motif G(A)AAASY GAAASY (Tanaka and others 1993). The IRF association domain (IAD), which is located toward the carboxy-terminal of the IRF proteins, is required for homo- and heterodimerization of IRF family members (Lin and others 1999; Mamane and others 1999) and interactions with other proteins, such as GRIP1 (Reily and others 2006). Both the DBD and IAD domains are evolutionarily conserved throughout the animal kingdom (Nehyba and others 2009). The extreme C-terminus of the IRFs has been shown to bind transcriptional co-activators such as CBP/p300 (Yang and others 2002). Upon activation with microbial stimuli, serine residues in the C-terminal region of IRF3/5/7 become phosphorylated (Honda and Taniguchi 2006). This modification promotes IRF dimerization and association with other transcriptional factors assembled on gene promoters (Honda and Taniguchi 2006). This process is relatively well characterized for IRF3 and IRF7, while the key steps of IRF5 activation, including the identity of IRF5 kinases, still remain to be elucidated. This review focuses on

Diverse Physiological Functions of IRF5 IRF5 was originally implicated in antiviral responses and type I interferon (IFN) production (Barnes and others 2001). It has since been demonstrated as a crucial regulator of the cell cycle and apoptosis (Barnes and others 2003b; Hu and others 2005), microbial infection (Paun and others 2011; del Fresno and others 2013), and inflammation (Takaoka and others 2005; Krausgruber and others 2011; Yang and others 2012; Courties and others 2014). Expression of IRF5 has been reported in numerous immune cell types, including macrophages, conventional dendritic cells (cDC) and plasmacytoid dendritic cells (pDC), and B-cells (Heng and others 2008). The essential role of IRF5 in multiple immune cell compartments during inflammation is highlighted by data generated with IRF5 - / - mice. Mice lacking IRF5 are resistant to lethal endotoxin-induced shock and show diminished production of pro-inflammatory mediators such as tumor necrosis factor (TNF), interleukin (IL)-6, and IL-12 in response to in vitro stimulation of pDCs/cDCs/macrophages with Toll-like receptor (TLR) agonists (Takaoka and others 2005). It can be said that IRF5 plays a central role in defining an inflammatory (M1) macrophage lineage: IRF5 expression is induced during differentiation of human monocytes and murine bone marrow into inflammatory macrophages to transcriptionally regulate characteristic pro-inflammatory markers such as cytokines (Krausgruber and others 2011; Weiss and others 2013). As a result, IRF5 - / - macrophages are polarized to an anti-inflammatory (M2) phenotype and IRF5 - / - mice are unable to mount essential Th1 responses

Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Kennedy Institute of Rheumatology, University of Oxford, Oxford, United Kingdom.

71

72

to chronic Leishmania donovani infection due to a shift toward increased Th2 cytokine expression (Paun and others 2011). Additionally, in response to lipopolysaccharide IRF5 - / - mice are unable to promote robust Th1/Th17 responses (Krausgruber and others 2011), which are especially prevalent in autoimmune conditions. In humans, IRF5 gene polymorphisms are associated with several inflammatory and autoimmune diseases, including systemic lupus erythematosus (SLE), inflammatory bowel disease, and rheumatoid arthritis (Dideberg and others 2007; Dieguez-Gonzalez and others 2008; Dawidowicz and others 2011). Consequently, absence of IRF5 expression in mice has phenotypic consequences for multiple murine models of autoimmune disease and infection. IRF5 expression has been identified in arthritic compared to control knees of the antigen-induced arthritis murine model (Weiss and others 2013). IRF5 is also crucial for development of murine SLE as demonstrated by the FcgRIIB - / - Yaa, FcgRIIB - / - , MLR/lpr, and pristane-induced models (Richez and others 2010; Tada and others 2011; Yang and others 2012). This reflects important roles for IRF5 in both myeloid cells and B-cells; indicated by reduced levels of type I IFN (Feng and others 2012), skewing toward Th2 responses (Xu and others 2012), impaired Ly6Chi monocyte recruitment (Yang and others 2012), reduced IgG class switching (Feng and others 2012), and attenuated plasma cell development (Lien and others 2010) in IRF5 - / - mice during these SLE models. Consistent with the importance of a type I IFN signature in SLE, IRF5 - / - mice also exhibit a marked reduction in serum type I IFN levels upon Newcastle disease virus (NDV) infection (Paun and others 2008) and poor survival of vesicular stomatitis virus (VSV) and Herpes simplex virus (HSV-1) infection associated with decreased IFNa production (Yanai and others 2007). IRF5-deficient B-cells also have attenuated class-switching to IgG2a/c—an antibody isotype that is prevalent in antiviral responses and autoimmunity (Fang and others 2012). Although the IRF5 gene can be induced by p53, IRF5 itself promotes pro-apoptotic gene expression and suppression of

FIG. 1. Signal transduction pathways that employ interferon regulatory factor 5 (IRF5). Pathogen-derived products and their mimetics [LPS, R-848, CpG, poly(I:C), b1,3-glucans, muramyl dipeptide (MDP), peptidoglycan (PGN), etc.] bind to pattern recognition receptors on the cell surface [Toll-like receptors (TLRs), Dectin-1] or in the cytosol (RIG-I, MDA-5, and NOD2) to trigger signaling cascades, which involve protein kinases [TBK1, IkB kinases (IKKs), RIP2, etc.] and result in IRF5 activation. Alternatively, IRF5 can be activated in response to DNA damage [CPT—camptoR (irinotecan)], viral infection [Newcastle disease virus (NDV)], or stimulation with cytokines (TRAIL and FASL).

RYZHAKOV, EAMES, AND UDALOVA

tumor cell growth in a p53-independent manner (Hu and others 2005). IRF5 inhibits the tumorigenic potential of Epstein Barr virus and Kaposi’s sarcoma-associated herpesvirus (Bi and others 2011; Xu and others 2011). Also, ectopic expression of IRF5 reduces proliferation of chronic myeloid leukaemia cells in vitro (Massimino and others 2014). Loss of IRF5 expression is observed in gastric cancer and correlates with progression and metastasis of breast cancer (Yamashita and others 2010; Bi and others 2011). Contrary to that, IRF5 promotes tumorigenesis in thyroid cancer (Massimino and others 2012). Such differences in IRF5 function may be attributed to cell type-specific roles of IRF5 and different genetic backgrounds of the cancers studied. In murine experiments, IRF5 behaves as a tumor suppressor gene. IRF5 deficiency results in tumorigenic transformation of murine embryonic fibroblasts in vitro or in athymic nude mice (Yanai and others 2007), while overexpression of IRF5 in a B-cell lymphoma cell line, BJAB, inhibits tumor growth in similar assays (Barnes and others 2003b). Problematically, it was identified in 2012 that a spontaneous genomic duplication and frame-shift mutation in the DOCK2 gene had been inadvertently bred to homozygosity in some IRF5 - / - colonies worldwide, leading to conflicting results within the field, including variation in data regarding type I IFN responses and antibody production. Phenotypes regarding pDC and B-cell development in IRF5 - / - mice were rescued by retroviral expression of wild-type DOCK2 (Purtha and others 2012), confirming the effect of this spontaneous mutation, and highlighting the necessity for rederivation of many IRF5 - / - colonies to ensure confidence in future data from IRF5 - / - mice (Yasuda and others 2013).

Signaling Pathways That Involve IRF5 The complex physiological functions of IRF5 described above reflect the involvement of this protein in several distinct signal transduction pathways in leukocytes and epithelial cells (Fig. 1). Several reports have characterized IRF5 activation downstream of TLRs and RIG-I-like RNA helicases (Barnes and others 2001, 2002b; Schoenemeyer

ACTIVATION AND FUNCTION OF IRF5

and others 2005; Chang Foreman and others 2012), involving interleukin-1 receptor-associated kinase (IRAK) and IkB kinase (IKK) (Fig. 1). As the same agonists are also well-known triggers of nuclear factor kappa B (NF-kB) activation, the canonical IKK complex might be the missing link. Chemically induced DNA damage can trigger IRF5dependent gene expression in tumor cell lines (Hu and others 2005; Couzinet and others 2008; Hu and Barnes 2009), however, the kinase mediators controlling this process are still unknown. This is also the case for Fas and TRAIL stimulation, which activate IRF5 upstream of caspase-8 (Hu and others 2005; Hu and Barnes 2009) leading to hepatic apoptosis and lethality that can be protected by absence of IRF5 expression in vivo, and is specific to activated dendritic cells (Couzinet and others 2008). Signaling triggered by stimulation of the intracellular receptor, NOD2, via the ligands muramyl dipeptide (MDP) or peptidoglycan (PGN), can lead to induction of IFNb expression via IRF5 and the protein kinase RIP2 (Pandey and others 2009), which associates with both IRF5 and NOD2 (Pandey and others 2009; Chang Foreman and others 2012). Finally, a recent report showed the critical requirement of IRF5 and the tyrosine kinase Syk in Dectin-1 signaling induced by stimulation with b1,3-glucans, indicating involvement of IRF5 in host defense against fungal pathogens and intracellular bacteria (del Fresno and others 2013). However, there is currently no evidence for Syk serving as an upstream kinase for IRF5 (del Fresno and others 2013).

73

Stages of IRF5 Activation As IRF5 is a close homologue of IRF3, similar modes of activation for IRF3 and IRF5 have been assumed (Barnes and others 2002a). The current model of IRF activation includes the following stages, each of which will be discussed in detail in the context of IRF5: (1) phosphorylation, ubiquitination, and possibly other post-translational modifications; (2) dimerization; (3) nuclear transport; (4) binding to gene promoters in complex with co-factors to regulate gene transcription (Fig. 2).

IRF5 Phosphorylation IRF5 phosphorylation was observed in cells infected with NDV by means of 32P metabolic labeling of untreated and infected cells (Barnes and others 2001, 2002b) or using a specific phospho-IRF5 antibody (Hu and others 2005). It was also detected in cancer cells stimulated with TRAIL (Hu and Barnes 2009) or subjected to DNA damage (Hu and others 2005). Residues S427 and S430 of IRF5 (numbered as in the human isoform v3/v4) were shown to be phosphorylated in cells infected with NDV but not upon stimulation with the DNA damaging agent CPT-11 (irinotecan); however, a pan-phospho-serine antibody did detect phosphorylation of IRF5 in response to CPT-11 treatment, which suggests that alternative serine residues are phosphorylated under these conditions (Hu and Barnes 2005).

FIG. 2. A schematic representation of the proposed mechanisms of IRF5 activation. Viral infection (NDV) or stimulation with pattern recognition receptor (PRR) agonists (R848) induces signaling cascades involving protein kinases (TBK1/IKKi) and the ubiquitin ligase TRAF6, which are proposed to phosphorylate and ubiquitinate IRF5. These modifications are suggested to trigger IRF5 dimerization and migration from the cytosol to the nucleus, where IRF5 binds to gene promoters in complex with co-activator proteins (p300) and activates transcription of antiviral or pro-inflammatory genes. An alternative model suggests that IRF5 is present in complex with the COP9 signalosome, which provides stability for IRF5. Upon stimulation with agonists such as TRAIL, an unknown kinase phosphorylates the complex, leading to IRF5 dissociation from COP9. This is followed by rapid ubiquitination and 26S proteasome-dependent degradation of IRF5. Some of the phosphorylated IRF5 migrates to the nucleus and activates transcription. DNA damage agents (CPT) are also able to induce IRF5 phosphorylation and transcriptional activation via an unknown kinase.

74


The protein kinase DNA-PK has been suggested to activate IRF5 in response to DNA damage (Hu and others 2005), though there are currently no reports providing additional experimental evidence to support this finding. In chronic myeloid leukemia cell lines, IRF5 was shown to be phosphorylated by the oncoprotein and tyrosine kinase BCR-ABL (Massimino and others 2014). The study suggested tyrosine Y104 of IRF5 as the target residue, though mutating it to phenylalanine had little impact on tyrosine phosphorylation of IRF5, which indicates other IRF5 sites may be modified by BCR-ABL (Massimino and others 2014). The mechanism of how a nuclear signal is transported to the cytosol to activate IRF5 is also still unknown. Multiple amino acid residues have been suggested as potential IRF5 phosphorylation sites, based on analyses of polypeptide sequence alignments (Barnes and others 2002b) and experimental approaches such as mass-spectrometry (Chang Foreman and others 2012) (Fig. 3). The initial approach was to shortlist candidate residues in the C-terminus of IRF5 that corresponded to known C-terminal phosphorylation targets in IRF3 (Yang and others 2002; Barnes and others 2002b). Serine residues S425, S427, and S430 of IRF5 (numbered as in the human isoform v3/v4) were mutated to alanine, as single substitutions or in combination, leading to suppression of NDV-induced activation of an IFNa promoter-controlled luciferase reporter in 2fTGH cells (Barnes and others 2002b). Mutation of the same residues plus serine S436 to aspartate, which has been shown to mimic a hyper-phosphorylated state of IRF5, resulted in constitutive activation and nuclear localization of IRF5 (Cheng and others 2006). This mutant also caused increased apoptosis compared to the wild-type protein when expressed in HeLa and HuH7 cells (Cheng and others 2006). Another report showed serine residues S156, S158, and the threonine residue T160 as important for IRF5 cellular localization as a part of its nuclear export motif (Lin and others 2005). Mutation of these residues to alanine had no effect on IRF5 subcellular localization, while mutating them to aspartate led to constitutive nuclear accumulation of IRF5 and loss of binding to the nuclear export machinery protein, Crm1 (Lin and others 2005). Based on these earlier studies, the crystal structure of an IRF5 protein dimer lacking the DBD domain and containing an S430D mutation has been solved (Chen and others 2008). According to this structure, the serine residues S425, S427, and S430 are not a part of the IRF5/p300 binding interface but are rather predicted to aid in unwrapping of the auto-inhibitory C-terminal structure of IRF5 upon their phosphorylation. At the

IAD

S457

CTD S436

S421 S425 S427 S430

SRR K401 K402

S309 S317

ID S156 S158 T160

T10

DBD

FIG. 3. Putative phosphorylation and ubiquitination sites of IRF5. DBD, DNA-binding domain; ID, intermediate domain; IAD, IRF association domain; SRR, serine-rich region; CTD, C-terminal domain. Circles indicate phosphorylation sites, triangles indicate ubiquitination sites. Residues numbered as in human isoform v3/v4. The suggested functions of the indicated residues in IRF5 activation are designated as: empty triangles (nuclear translocation), grey circles (nuclear retention), empty circles (release of autoinhibition), crossed circle (dimer stabilization), black (function unknown).

same time, the serine residue S436 appears to contribute to stabilization of the IRF5 dimer (Chen and others 2008). As a functional consequence, the S436D mutant triggers enhanced activation of an IFNb promoter-controlled luciferase reporter in HEK293 cells (Chen and others 2008). A recent study analyzed IRF5 phosphorylation in a systematic manner using mass spectrometry (Chang Foreman and others 2012). Briefly, IRF5 was co-expressed in human cells with the candidate kinases RIP2 or TBK1 or with an E3 ligase TRAF6. Alternatively, bacterially expressed IRF5, lacking its DBD, was incubated with immune-complexes containing TBK1. Following this, the IRF5 lysates or kinase reactions were immunoprecipitated with an IRF5 antibody. The eluted proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and the IRF5 bands were excised and analyzed by mass spectrometry (Chang Foreman and others 2012). Two IRF5 serine residues, S158 and S309 were found to be modified by TBK1 in the kinase reactions and 6 residues were pinned as IRF5 phosphorylation sites in the coexpression assays: T10, S158, S309, S317, S425, and S436 (the latter 2 are referred to as S451 and S462 of human IRF5 isoform v5 used in the study). The subsequent analyses showed that serine residue S436 had a major impact on RIP2-induced activation of an IL-12p40 promotercontrolled luciferase reporter (Chang Foreman and others 2012). Also, the S425A/S436A double serine mutant of IRF5 failed to support reporter construct activation in cells stimulated with either NOD2 ligands, peptidoglycan or muramyl dipeptide. This study did not address the impact of TBK1 target sites on TBK1-mediated IRF5 activation.

IRF5 Ubiquitination Many innate immune signaling pathways employ the E3 ligase TRAF6, which has been shown to K63 ubiquitinate IRF5 on lysine residues K401 and K402 (numbered as in human IRF5 isoform v3/v4; referred to as K410 and K411 of human IRF5 isoform v5 in the cited study) following overexpression in HEK293T cells (Balkhi and others 2008). IRAK1 kinase appears to bind to IRF5 and bring TRAF6 into the complex: TRAF6-mediated ubiquitination of IRF5 is not observed in IRAK1-deficient cells. On the other hand, ectopic expression of another IRF5-binding partner, MyD88, disrupts the IRAK1-IRF5 complex and suppresses ubiquitination of IRF5 (Balkhi and others 2008). The role of IRAK1 phosphorylation in IRF5 ubiquitination by TRAF6 has yet to be addressed. Mutation of lysine residues K401 and K402 (residue numbers as in human isoform v3/v4) to arginine resulted in impaired nuclear translocation of IRF5 (Balkhi and others 2008). As these residues are located close to the C-terminal phosphorylation sites of IRF5, their mutagenesis could potentially interfere with dimerization and other interactions with co-factors. These mutations, therefore, may not be used as sole evidence to support the significant role of IRF5 ubiquitination. Another study showed that co-expression of IRF5 with the de-ubiquitinating enzyme A20 inhibited IRF5-induced activation of an IL-12 promoter-controlled luciferase reporter (Chang Foreman and others 2012). However, A20 also possesses K48 ubiquitin ligase activity (Wertz and others 2004), the contribution of which was not addressed in this study. IRF5 ubiquitination triggered by its


co-expression with MyD88 and TRAF6 can be inhibited by ectopically expressed IKKa (Balkhi and others 2010). Moreover, in the same system IKKa also inhibited IRF5-induced gene expression as measured by monitoring induction of luciferase activities of IFNA4, TNFa, and RANTES-promoterdriven reporters (Balkhi and others 2010), which serves another indicator of functional significance of IRF5 ubiquitination. As there are common concerns that many posttranslational modifications are simply overexpression artifacts; ubiquitination of endogenous IRF5 under physiological conditions (in the presence of TLR agonists or other stimuli) needs to be explored. The relationship between phosphorylation and ubiquitination of IRF5 has not been fully elucidated. Poly-ubiquitination by TRAF6 alone appears to be insufficient to activate IRF5, as the mutated form of IRF5, with key serine residues S425 and S436 mutated to alanine, lacks transcriptional activity despite still being ubiquitinated by TRAF6 (Chang Foreman and others 2012). As overexpressed TRAF6 is prone to ubiquitinate protein targets nonspecifically, it would be interesting to see if the S425/S436 double mutant of IRF5 is ubiquitinated in response to a physiological stimulus. The ubiquitination-deficient K401R/K402R double mutant of IRF5 demonstrates diminished transcriptional activity in human cells co-expressing IRF5 and RIP2 kinase (Chang Foreman and others 2012). However, mutation of serine residues S425 and S436 to aspartate restores IRF5 activation on the K401R/K402R double mutant background (Chang Foreman and others 2012). It is possible that IRF5 ubiquitination simply precludes phosphorylation, and the IRF5 S425D/S436D double mutant mimicking the phosphorylated form of IRF5 no longer requires ubiquitination to be active. It could also be that this phenomenon is only limited to the case of RIP2 co-expression, and does not occur in cells stimulated with pathogen products. It would be interesting to determine whether or not the ubiquitination-deficient mutant of IRF5 can be phosphorylated at the indicated serine residues in response to pathogenic stimuli.

IRF5 Dimerization Phosphorylation is believed to cause nuclear translocation of members of the IRF family including IRF5 by inducing a conformational change in the C-terminus of the protein, which enables dimerization and interaction with co-factors (Barnes and others 2002b). Phosphorylation-dependent IRF5 dimerization and its binding to the transcription co-activator proteins CBP/p300 were demonstrated in vitro (Cheng and others 2006; Chen and others 2008). Purified recombinant IRF5 proteins (amino acids 222–467) with intact or mutated C-termini were analyzed by size-exclusion chromatography: the phospho-mimetic mutant of IRF5, S430D, showed strong dimer formation and association with p300 (Chen and others 2008). A crystal structure of the IRF5 S430D mutant was solved and demonstrated that IRF5 was present in the dimeric state, with the C-terminus of 1 subunit (the linker region connecting helix 4 and helix 5) binding to the IAD domain (helix 2) of the companion subunit (Chen and others 2008). Co-immunoprecipitation assays using cells overexpressing GFP- and FLAG-tagged IRF5 showed the formation of IRF5 dimers, which was enhanced in NDV-infected cells (Barnes and others 2002b). In addition, overexpressed FLAG-tagged IRF5 could be pulled out with GST-IRF5-labeled agarose beads more efficiently from NDV-infected versus noninfected

75

cells (Barnes and others 2003a). Another study observed IRF5 dimerization and its binding to p300 only in cells coexpressing one of the two protein kinases TBK1 or IKKi, but not in cells infected with NDV (Cheng and others 2006). A high molecular weight complex containing IRF5 was observed in HEK293T cells, in which IRF5 was co-expressed with the TLR adaptor protein MyD88 or the combination of MyD88 with IRAK1 or TRAF6 proteins (Paun and others 2008; Balkhi and others 2010). However, it is unclear whether this complex represents the putative IRF5 dimer or a single IRF5 subunit bound to another protein.

IRF5 Nuclear Translocation Once IRF5 is modified and dimerized, it is expected to go to the nucleus. Indeed, as is the case with IRF3, NDV infection of human 2fTGH cells triggered nuclear translocation of IRF5 (Barnes and others 2001), while another study discovered two putative nuclear localization sequences, PRRVRLK and PREKKLI, which the IRF5 protein requires for this process (Barnes and others 2002b). The lysine residues within the PREKKLI motif, K401 and K402 (residue numbers as in human IRF5 isoform v3/v4), were proposed to be ubiquitination sites of IRF5, as mutation of these residues to alanine abrogates both ubiquitination and nuclear import of IRF5 (Balkhi and others 2008). It is therefore possible that ubiquitination of these residues somehow triggers nuclear translocation of IRF5. Interestingly, purified recombinant nuclear import proteins, karyopherins a1 and b1, have been shown to bind to IRF5 (Yeon and others 2008). The exact IRF5-karyoferin interaction interface needs to be mapped, but the proposed NLS sequences in IRF5 could serve as initial candidates for this. It would also be interesting to investigate whether or not the interaction of karyoferins with IRF5 requires ubiquitination. A recent report hypothesized that IRF5 acts a chaperone for nuclear translocation of IRAK1 kinase (Inoue and others 2014). Using immunohistochemistry the study demonstrated LPS-induced IRF5 nuclear translocation in murine macrophages, while IRAK1 translocation required signals from both CD40 and TLR4, the former providing a signal to sumoylate IRAK1 (Inoue and others 2014). In addition to this, IRAK1 was shown to bind IRF5 in co-immunoprecipitation experiments in the same study. While it is not known whether or not IRAK1 translocation depends on the presence of IRF5, it is curious to assume that the sumoylated nuclear IRAK1 may phosphorylate IRF5 and contribute to its transcriptional activity. Phosphorylation of IRF5 was also suggested to inhibit nuclear export of IRF5: the Crm1 protein was shown to interact with a nuclear export sequence of IRF5, 150LQRMLPSLSLT160 (Lin and others 2005). Moreover, mutating serine/threonine residues in this motif to aspartate, which mimics a phosphorylated state of IRF5, prevented export of IRF5 from the nucleus (Lin and others 2005). The same study ruled out a role for TBK1 and IKKi in this process. Although, these kinases caused IRF5 phosphorylation and dimerization when co-expressed with IRF5 in HEK293 cells (Lin and others 2005; Schoenemeyer and others 2005), they failed to induce nuclear accumulation of IRF5 (Lin and others 2005). This suggests involvement of other kinases and as yet unidentified factors in IRF5 activation, with ubiquitination and acetylation of IRF5 playing important but yet to be understood roles (Balkhi and others 2008; Feng and others 2010).

76

IRF5 Gene Activation Nuclear IRF5 was shown to bind to gene promoters in chromatin immunoprecipitation experiments (Krausgruber and others 2010, 2011). While the exact transactivation mechanism of IRF5-mediated gene expression remains to be elucidated, IRF5 is likely to function by recruiting transcriptional co-activators that remodel chromatin and trigger activation of RNA polymerase II (Feng and others 2010; Krausgruber and others 2010). Indeed several reports have shown IRF5 associating with other transcription factors such as NF-kB, nuclear co-repressor proteins such as TRIM28 and HDAC1, and nuclear co-activating histone acetylase proteins such as p300, CBP, and PCAF (Feng and others 2010; Krausgruber and others 2010; Eames and others 2012; Steinhagen and others 2013). Furthermore, an interaction between phosphorylated IRF5 and p300 was shown in the nuclei of NDV-infected cells (Feng and others 2010). Interestingly, infection of cells with NDV in combination with co-expression of IRF5 with p300 and CBP but not PCAF induced acetylation of IRF5, which was detected using an acetyl-lysine-reactive antibody (Feng and others 2010). The identity of the modified residues and the role of IRF5 acetylation in transcription remain to be established.

IRF5 and COP9 Signalosome The current model of IRF5 activation has recently been challenged by a study showing the presence of a stable complex between IRF5 and the COP9 signalosome (CSN) in quiescent cells (Fig. 2, right panel) (Korczeniewska and Barnes 2013). The CSN is a multi-protein complex that is evolutionarily conserved in plants and animals and shares sequence homology to regulatory subunits of the 26S proteasome (Seeger and others 1998). COP9 participates in immune signaling pathways in mammals, such as NF-kB signaling, where COP9 controls stability of a number of the signaling components (Schweitzer and Naumann 2010). COP9 was also found to interact with IRF8, which leads to IRF8 phosphorylation by an unknown kinase associated with COP9 (Cohen and others 2000). The latter event is important for IRF8-IRF1 interactions and the repressor activity that IRF8 exerts on IRF1 (Cohen and others 2000). Of interest, a protein kinase CDK2 has been recently shown to bind COP9, which is important for COP9mediated cell senescence (Yoshida and others 2013). A component of the COP9 complex, CSN7, was found as an endogenous binding partner of IRF5 in a proteomic screen using overexpressed FLAG-tagged IRF5 in HEK293 cells as the pull-down bait (Korczeniewska and Barnes 2013). This finding was confirmed by immunoprecipitation assays where the interaction between endogenous IRF5 and the subunits of the COP9 signalosome, CSN1–8, were detected (Korczeniewska and Barnes 2013). The COP9 complex appears to regulate IRF5 protein stability and its ability to activate target genes (Korczeniewska and Barnes 2013). RNAi-assisted depletion of components of the CSN complex results in fast proteasome-mediated degradation of IRF5, while a deletion mutant of IRF5 that is unable to bind COP9, undergoes enhanced ubiquitination and rapid degradation in comparison (Korczeniewska and Barnes 2013). The study also reported quick dissociation of the IRF5/COP9 complex in response to TRAIL stimulation in THP-1 cells, followed by nuclear translocation of IRF5 and decreased stability of the IRF5 pro-


tein (Fig. 2, right panel). Overall, this intriguing study raises many questions. Is endogenous IRF5 protein bound to COP9 in nonstimulated cells? Does this happen in every cell type expressing IRF5? Does stimulation with viral and bacterial agonists also disrupt the IRF5/COP9 complex? What causes COP9 to dissociate from IRF5 on the structural level—are any posttranslational modifications involved in this? If so, then what is the identity of the kinases and/or other enzymes that target the COP9/IRF5 complex?

Future Research Over the last decade, the roles of phosphorylation and other modifications in IRF5 activation have been described using overexpression systems, recombinant purified proteins, and point mutants of IRF5, and utilizing tools such as mass spectrometry. The next step is to systematically explore posttranslational modifications of endogenous IRF5 in vivo and their contribution to IRF5-dependent transcription. Certain tools will be required for this research to progress. Generation of IRF5 phospho-specific antibodies would facilitate testing of candidate kinases for their ability to phosphorylate IRF5 at specific sites both in vivo and in vitro, alongside measurement of the impact of these modifications on IRF5-mediated gene expression. Ultimately, understanding how IRF5 activity is regulated in this manner would enable the design of strategies for pathway-specific modulation of IRF5 function within the context of cancer and inflammation.

Author Disclosure Statement No competing financial interests exist.

References Balkhi MY, Fitzgerald KA, Pitha PM. 2008. Functional regulation of MyD88-activated interferon regulatory factor 5 by K63-linked polyubiquitination. Mol Cell Biol 28(24):7296– 7308. Balkhi MY, Fitzgerald KA, Pitha PM. 2010. IKKalpha negatively regulates IRF-5 function in a MyD88-TRAF6 pathway. Cell Signal 22(1):117–127. Barnes B, Lubyova B, Pitha PM. 2002a. On the role of IRF in host defense. J Interferon Cytokine Res 22(1):59–71. Barnes BJ, Field AE, Pitha-Rowe PM. 2003a. Virus-induced heterodimer formation between IRF-5 and IRF-7 modulates assembly of the IFNA enhanceosome in vivo and transcriptional activity of IFNA genes. J Biol Chem 278(19):16630–16641. Barnes BJ, Kellum MJ, Field AE, Pitha PM. 2002b. Multiple regulatory domains of IRF-5 control activation, cellular localization, and induction of chemokines that mediate recruitment of T lymphocytes. Mol Cell Biol 22(16):5721–5740. Barnes BJ, Kellum MJ, Pinder KE, Frisancho JA, Pitha PM. 2003b. Interferon regulatory factor 5, a novel mediator of cell cycle arrest and cell death. Cancer Res 63(19):6424–6431. Barnes BJ, Moore PA, Pitha PM. 2001. Virus-specific activation of a novel interferon regulatory factor, IRF-5, results in the induction of distinct interferon alpha genes. J Biol Chem 276(26):23382–23390. Bi X, Hameed M, Mirani N, Pimenta EM, Anari J, Barnes BJ. 2011. Loss of interferon regulatory factor 5 (IRF5) expression in human ductal carcinoma correlates with disease stage and contributes to metastasis. Breast Cancer Res 13(6):R111. Chang Foreman HC, Van Scoy S, Cheng TF, Reich NC. 2012. Activation of interferon regulatory factor 5 by site specific phosphorylation. PLoS One 7(3):e33098.


Chen W, Lam SS, Srinath H, Jiang Z, Correia JJ, Schiffer CA, Fitzgerald KA, Lin K, Royer WE, Jr. 2008. Insights into interferon regulatory factor activation from the crystal structure of dimeric IRF5. Nat Struct Mol Biol 15(11):1213–1220. Cheng TF, Brzostek S, Ando O, Van Scoy S, Kumar KP, Reich NC. 2006. Differential activation of IFN regulatory factor (IRF)-3 and IRF-5 transcription factors during viral infection. J Immunol 176(12):7462–7470. Cohen H, Azriel A, Cohen T, Meraro D, Hashmueli S, BechOtschir D, Kraft R, Dubiel W, Levi BZ. 2000. Interaction between interferon consensus sequence-binding protein and COP9/signalosome subunit CSN2 (Trip15). A possible link between interferon regulatory factor signaling and the COP9/ signalosome. J Biol Chem 275(50):39081–39089. Courties G, Heidt T, Sebas M, Iwamoto Y, Jeon D, Truelove J, Tricot B, Wojtkiewicz G, Dutta P, Sager HB, Borodovsky A, Novobrantseva T, Klebanov B, Fitzgerald K, Anderson DG, Libby P, Swirski FK, Weissleder R, Nahrendorf M. 2014. In vivo silencing of the transcription factor IRF5 reprograms the macrophage phenotype and improves infarct healing. J Am Coll Cardiol 63(15):1556–1566. Couzinet A, Tamura K, Chen HM, Nishimura K, Wang Z, Morishita Y, Takeda K, Yagita H, Yanai H, Taniguchi T, Tamura T. 2008. A cell-type-specific requirement for IFN regulatory factor 5 (IRF5) in Fas-induced apoptosis. Proc Natl Acad Sci U S A 105(7):2556–2561. Dawidowicz K, Allanore Y, Guedj M, Pierlot C, Bombardieri S, Balsa A, Westhovens R, Barrera P, Alves H, Teixeira VH, Petit-Teixeira E, van de Putte L, van Riel P, Prum B, Bardin T, Meyer O, Cornelis F, Dieude P. 2011. The interferon regulatory factor 5 gene confers susceptibility to rheumatoid arthritis and influences its erosive phenotype. Ann Rheum Dis 70(1):117–121. del Fresno C, Soulat D, Roth S, Blazek K, Udalova I, Sancho D, Ruland J, Ardavin C. 2013. Interferon-beta production via Dectin-1-Syk-IRF5 signaling in dendritic cells is crucial for immunity to C. albicans. Immunity 38(6):1176–1186. Dideberg V, Kristjansdottir G, Milani L, Libioulle C, Sigurdsson S, Louis E, Wiman AC, Vermeire S, Rutgeerts P, Belaiche J, Franchimont D, Van Gossum A, Bours V, Syvanen AC. 2007. An insertion-deletion polymorphism in the interferon regulatory factor 5 (IRF5) gene confers risk of inflammatory bowel diseases. Hum Mol Genet 16(24):3008–3016. Dieguez-Gonzalez R, Calaza M, Perez-Pampin E, de la Serna AR, Fernandez-Gutierrez B, Castaneda S, Largo R, Joven B, Narvaez J, Navarro F, Marenco JL, Vicario JL, Blanco FJ, Fernandez-Lopez JC, Caliz R, Collado-Escobar MD, Carreno L, Lopez-Longo J, Canete JD, Gomez-Reino JJ, Gonzalez A. 2008. Association of interferon regulatory factor 5 haplotypes, similar to that found in systemic lupus erythematosus, in a large subgroup of patients with rheumatoid arthritis. Arthritis Rheum 58(5):1264–1274. Eames HL, Saliba DG, Krausgruber T, Lanfrancotti A, Ryzhakov G, Udalova IA. 2012. KAP1/TRIM28: an inhibitor of IRF5 function in inflammatory macrophages. Immunobiology 217(12):1315–1324. Fang CM, Roy S, Nielsen E, Paul M, Maul R, Paun A, Koentgen F, Raval FM, Szomolanyi-Tsuda E, Pitha PM. 2012. Unique contribution of IRF-5-Ikaros axis to the B-cell IgG2a response. Genes Immun 13(5):421–430. Feng D, Sangster-Guity N, Stone R, Korczeniewska J, Mancl ME, Fitzgerald-Bocarsly P, Barnes BJ. 2010. Differential requirement of histone acetylase and deacetylase activities for IRF5-mediated proinflammatory cytokine expression. J Immunol 185(10):6003–6012.

77

Feng D, Yang L, Bi X, Stone RC, Patel P, Barnes BJ. 2012. Irf5-deficient mice are protected from pristane-induced lupus via increased Th2 cytokines and altered IgG class switching. Eur J Immunol 42(6):1477–1487. Heng TS, Painter MW, Immunological Genome Project C. 2008. The Immunological Genome Project: networks of gene expression in immune cells. Nat Immunol 9(10):1091–1094. Honda K, Taniguchi T. 2006. IRFs: master regulators of signalling by Toll-like receptors and cytosolic pattern-recognition receptors. Nat Rev Immunol 6(9):644–658. Hu G, Barnes BJ. 2009. IRF-5 is a mediator of the death receptorinduced apoptotic signaling pathway. J Biol Chem 284(5): 2767–2777. Hu G, Mancl ME, Barnes BJ. 2005. Signaling through IFN regulatory factor-5 sensitizes p53-deficient tumors to DNA damage-induced apoptosis and cell death. Cancer Res 65(16): 7403–7412. Inoue M, Arikawa T, Chen YH, Moriwaki Y, Price M, Brown M, Perfect JR, Shinohara ML. 2014. T cells down-regulate macrophage TNF production by IRAK1-mediated IL-10 expression and control innate hyperinflammation. Proc Natl Acad Sci U S A 111(14):5295–5300. Korczeniewska J, Barnes BJ. 2013. The COP9 signalosome interacts with and regulates interferon regulatory factor 5 protein stability. Mol Cell Biol 33(6):1124–1138. Krausgruber T, Blazek K, Smallie T, Alzabin S, Lockstone H, Sahgal N, Hussell T, Feldmann M, Udalova IA. 2011. IRF5 promotes inflammatory macrophage polarization and T(H)1T(H)17 responses. Nat Immunol 12(3):231–238. Krausgruber T, Saliba D, Ryzhakov G, Lanfrancotti A, Blazek K, Udalova IA. 2010. IRF5 is required for late-phase TNF secretion by human dendritic cells. Blood 115(22):4421–4430. Levy DE, Kessler DS, Pine R, Reich N, Darnell JE, Jr. 1988. Interferon-induced nuclear factors that bind a shared promoter element correlate with positive and negative transcriptional control. Genes Dev 2(4):383–393. Lien C, Fang CM, Huso D, Livak F, Lu R, Pitha PM. 2010. Critical role of IRF-5 in regulation of B-cell differentiation. Proc Natl Acad Sci U S A 107(10):4664–4668. Lin R, Mamane Y, Hiscott J. 1999. Structural and functional analysis of interferon regulatory factor 3: localization of the transactivation and autoinhibitory domains. Mol Cell Biol 19(4):2465–2474. Lin R, Yang L, Arguello M, Penafuerte C, Hiscott J. 2005. A CRM1-dependent nuclear export pathway is involved in the regulation of IRF-5 subcellular localization. J Biol Chem 280(4):3088–3095. Mamane Y, Heylbroeck C, Genin P, Algarte M, Servant MJ, LePage C, DeLuca C, Kwon H, Lin R, Hiscott J. 1999. Interferon regulatory factors: the next generation. Gene 237(1):1–14. Massimino M, Consoli ML, Mesuraca M, Stagno F, Tirro E, Stella S, Pennisi MS, Romano C, Buffa P, Bond HM, Morrone G, Sciacca L, Di Raimondo F, Manzella L, Vigneri P. 2014. IRF5 is a target of BCR-ABL kinase activity and reduces CML cell proliferation. Carcinogenesis 35(5):1132–1143. Massimino M, Vigneri P, Fallica M, Fidilio A, Aloisi A, Frasca F, Manzella L. 2012. IRF5 promotes the proliferation of human thyroid cancer cells. Mol Cancer 11:21. Nehyba J, Hrdlickova R, Bose HR. 2009. Dynamic evolution of immune system regulators: the history of the interferon regulatory factor family. Mol Biol Evol 26(11):2539–2550. Pandey AK, Yang Y, Jiang Z, Fortune SM, Coulombe F, Behr MA, Fitzgerald KA, Sassetti CM, Kelliher MA. 2009. NOD2, RIP2 and IRF5 play a critical role in the type I interferon response to Mycobacterium tuberculosis. PLoS Pathog 5(7):e1000500.

78

Paun A, Bankoti R, Joshi T, Pitha PM, Stager S. 2011. Critical role of IRF-5 in the development of T helper 1 responses to Leishmania donovani infection. PLoS Pathog 7(1):e1001246. Paun A, Reinert JT, Jiang Z, Medin C, Balkhi MY, Fitzgerald KA, Pitha PM. 2008. Functional characterization of murine interferon regulatory factor 5 (IRF-5) and its role in the innate antiviral response. J Biol Chem 283(21):14295–14308. Purtha WE, Swiecki M, Colonna M, Diamond MS, Bhattacharya D. 2012. Spontaneous mutation of the Dock2 gene in Irf5-/- mice complicates interpretation of type I interferon production and antibody responses. Proc Natl Acad Sci U S A 109(15):E898–E904. Reily MM, Pantoja C, Hu X, Chinenov Y, Rogatsky I. 2006. The GRIP1:IRF3 interaction as a target for glucocorticoid receptormediated immunosuppression. EMBO J 25(1):108–117. Richez C, Yasuda K, Bonegio RG, Watkins AA, Aprahamian T, Busto P, Richards RJ, Liu CL, Cheung R, Utz PJ, MarshakRothstein A, Rifkin IR. 2010. IFN regulatory factor 5 is required for disease development in the FcgammaRIIB-/-Yaa and FcgammaRIIB-/- mouse models of systemic lupus erythematosus. J Immunol 184(2):796–806. Savitsky D, Tamura T, Yanai H, Taniguchi T. 2010. Regulation of immunity and oncogenesis by the IRF transcription factor family. Cancer Immunol Immunother 59(4):489–510. Schoenemeyer A, Barnes BJ, Mancl ME, Latz E, Goutagny N, Pitha PM, Fitzgerald KA, Golenbock DT. 2005. The interferon regulatory factor, IRF5, is a central mediator of toll-like receptor 7 signaling. J Biol Chem 280(17):17005–17012. Schweitzer K, Naumann M. 2010. Control of NF-kappaB activation by the COP9 signalosome. Biochem Soc Trans 38(Pt 1):156–161. Seeger M, Kraft R, Ferrell K, Bech-Otschir D, Dumdey R, Schade R, Gordon C, Naumann M, Dubiel W. 1998. A novel protein complex involved in signal transduction possessing similarities to 26S proteasome subunits. FASEB J 12(6):469– 478. Steinhagen F, McFarland AP, Rodriguez LG, Tewary P, Jarret A, Savan R, Klinman DM. 2013. IRF-5 and NF-kappaB p50 coregulate IFN-beta and IL-6 expression in TLR9-stimulated human plasmacytoid dendritic cells. Eur J Immunol 43(7):1896– 1906. Tada Y, Kondo S, Aoki S, Koarada S, Inoue H, Suematsu R, Ohta A, Mak TW, Nagasawa K. 2011. Interferon regulatory factor 5 is critical for the development of lupus in MRL/lpr mice. Arthritis Rheum 63(3):738–748. Takaoka A, Yanai H, Kondo S, Duncan G, Negishi H, Mizutani T, Kano S, Honda K, Ohba Y, Mak TW, Taniguchi T. 2005. Integral role of IRF-5 in the gene induction programme activated by Toll-like receptors. Nature 434(7030):243–249. Tamura T, Yanai H, Savitsky D, Taniguchi T. 2008. The IRF family transcription factors in immunity and oncogenesis. Annu Rev Immunol 26:535–584. Tanaka N, Kawakami T, Taniguchi T. 1993. Recognition DNA sequences of interferon regulatory factor 1 (IRF-1) and IRF2, regulators of cell growth and the interferon system. Mol Cell Biol 13(8):4531–4538. Weiss M, Blazek K, Byrne AJ, Perocheau DP, Udalova IA. 2013. IRF5 is a specific marker of inflammatory macrophages in vivo. Mediators Inflamm 2013:245804. Wertz IE, O’Rourke KM, Zhou H, Eby M, Aravind L, Seshagiri S, Wu P, Wiesmann C, Baker R, Boone DL, Ma A, Koonin EV, Dixit VM. 2004. De-ubiquitination and ubiquitin ligase domains of A20 downregulate NF-kappaB signalling. Nature 430(7000):694–699.


Xu D, Meyer F, Ehlers E, Blasnitz L, Zhang L. 2011. Interferon regulatory factor 4 (IRF-4) targets IRF-5 to regulate EpsteinBarr virus transformation. J Biol Chem 286(20):18261–18267. Xu Y, Lee PY, Li Y, Liu C, Zhuang H, Han S, Nacionales DC, Weinstein J, Mathews CE, Moldawer LL, Li SW, Satoh M, Yang LJ, Reeves WH. 2012. Pleiotropic IFN-dependent and -independent effects of IRF5 on the pathogenesis of experimental lupus. J Immunol 188(8):4113–4121. Yamashita M, Toyota M, Suzuki H, Nojima M, Yamamoto E, Kamimae S, Watanabe Y, Kai M, Akashi H, Maruyama R, Sasaki Y, Yamano H, Sugai T, Shinomura Y, Imai K, Tokino T, Itoh F. 2010. DNA methylation of interferon regulatory factors in gastric cancer and noncancerous gastric mucosae. Cancer Sci 101(7):1708–1716. Yanai H, Chen HM, Inuzuka T, Kondo S, Mak TW, Takaoka A, Honda K, Taniguchi T. 2007. Role of IFN regulatory factor 5 transcription factor in antiviral immunity and tumor suppression. Proc Natl Acad Sci U S A 104(9):3402–3407. Yang H, Lin CH, Ma G, Orr M, Baffi MO, Wathelet MG. 2002. Transcriptional activity of interferon regulatory factor (IRF)3 depends on multiple protein-protein interactions. Eur J Biochem 269(24):6142–6151. Yang L, Feng D, Bi X, Stone RC, Barnes BJ. 2012. Monocytes from Irf5-/- mice have an intrinsic defect in their response to pristane-induced lupus. J Immunol 189(7):3741–3750. Yasuda K, Nundel K, Watkins AA, Dhawan T, Bonegio RG, Ubellacker JM, Marshak-Rothstein A, Rifkin IR. 2013. Phenotype and function of B cells and dendritic cells from interferon regulatory factor 5-deficient mice with and without a mutation in DOCK2. Int Immunol 25(5):295–306. Yeon SI, Youn JH, Lim MH, Lee HJ, Kim YM, Choi JE, Lee JM, Shin JS. 2008. Development of monoclonal antibodies against human IRF-5 and their use in identifying the binding of IRF-5 to nuclear import proteins karyopherin-alpha1 and -beta1. Yonsei Med J 49(6):1023–1031. Yoshida A, Yoneda-Kato N, Kato JY. 2013. CSN5 specifically interacts with CDK2 and controls senescence in a cytoplasmic cyclin E-mediated manner. Sci Rep 3:1054.

Address correspondence to: Dr. Grigory Ryzhakov Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences Kennedy Institute of Rheumatology University of Oxford Roosevelt Drive Headington Oxford OX3 7FY United Kingdom E-mail: [email protected] Prof. Irina A. Udalova Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences Kennedy Institute of Rheumatology University of Oxford Roosevelt Drive Headington Oxford OX3 7FY United Kingdom E-mail: [email protected] Received 31 January 2014/Accepted 28 July 2014