Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Mar 19.
Published in final edited form as: Immunity. 2019 Feb 12;50(3):576–590.e6. doi: 10.1016/j.immuni.2019.01.007

O-GlcNAc transferase suppresses inflammation and necroptosis by targeting receptor-interacting serine/threonine-protein kinase 3

Xinghui Li 1,13, Wei Gong 2,13, Hao Wang 3, Tianliang Li 1, Kuldeep S Attri 4, Robert E Lewis 4, Andre C Kalil 5, Fatema Bhinderwala 6, Robert Powers 6, Guowei Yin 7, Laura E Herring 8, John M Asara 9, Yu L Lei 10, Xiaoyong Yang 11, Diego A Rodriguez 12, Mao Yang 12, Douglas R Green 12, Pankaj K Singh 4, Haitao Wen 1,14,*
PMCID: PMC6426684  NIHMSID: NIHMS1519255  PMID: 30770249

SUMMARY

Elevated glucose metabolism in immune cells represents a hallmark feature of many inflammatory diseases such as sepsis. However, the role of individual glucose metabolic pathways during immune cell activation and inflammation remains incompletely understood. Here we demonstrate a previously unrecognized anti-inflammatory function of the hexosamine biosynthesis pathway (HBP)-associated O-linked β-N-acetylglucosamine (O-GlcNAc) signaling. Despite elevated activities of glycolysis and the pentose phosphate pathway, activation of macrophages with LPS resulted in attenuated HBP activity and protein O-GlcNAcylation. Deletion of the O-GlcNAc transferase (OGT), a key enzyme for protein O-GlcNAcylation, led to an enhanced innate immune activation and exacerbated septic inflammation. Mechanistically, OGT-mediated O-GlcNAcylation of the serine-threonine kinase RIPK3 on threonine 467 (T467) prevented RIPK3-RIPK1 hetero- and RIPK3-RIPK3 homo-interaction and inhibited downstream innate immunity and necroptosis signaling. Thus, our study identifies an immuno-metabolism crosstalk essential for fine-tuning innate immune cell activation and highlights the importance of glucose metabolism in septic inflammation.

Graphical Abstract

graphic file with name nihms-1519255-f0001.jpg

In Brief (eTOC Blurb)

The role of individual glucose metabolic pathways in the innate immunity remains largely unknown. Xinghui et al. demonstrate that attenuated O-linked β-N-acetylglucosamine (O-GlcNAc) signaling enhances TLR-induced innate immune response and necroptosis. Mechanistically, O-GlcNAcylation of the kinase RIPK3 blocks RHIM domain-mediated protein interaction and downstream signaling activation.

INTRODUCTION

Reprogramming of cellular metabolic activities has recently been demonstrated to play a critical role in the activation of the immune system and hyperinflammation (Buck et al., 2017; O’Neill et al., 2016). Increased glucose uptake and glycolysis occur in classically activated innate immune cells in vitro and in vivo (Everts et al., 2014). A widely accepted concept in the immunometabolism research field is that elevated catabolic activity in activated immune cells is required to meet the increased demand of biomolecules and energy for effective immune functions. Those functions including cell migration, phagocytosis and cytokine production are necessary for host response against invading pathogens or tissue injury during inflammation. Recent progress has broadened our understanding of how metabolic reprogramming modulates immune functions in multiple aspects. For example, a variety of metabolic enzymes involved in the glycolysis and mitochondrial metabolic pathways have been identified to play essential roles in affecting innate immune cell function (O’Neill et al., 2016). Moreover, many intermediate metabolites such as succinate (Tannahill et al., 2013), fumarate (Arts et al., 2016), itaconate (Bambouskova et al., 2018; Mills et al., 2018) and α-ketoglutarate (Liu et al., 2017) have recently been reported to participate in immune activation or modulation. Therefore, metabolic system regulates immune cell function and inflammation through combined strategies.

Glucose serves as a major nutrient to fuel cellular metabolic activities. Three major glucose metabolic pathways, namely glycolysis, the pentose phosphate pathway (PPP), and the hexosamine biosynthesis pathway (HBP) collaboratively determine how glucose is processed. HBP is a unique glucose metabolism pathway leading to the generation of its end product uridine diphosphate N-acetylglucosamine (UDP-GlcNAc), which is further utilized by the O-GlcNAc transferase (OGT) for protein modification, namely O-GlcNAcylation (Levine and Walker, 2016). Many proteins involved in various fundamental biological processes have been identified as O-GlcNAcylation targets, including transcription factors, kinases and enzymes (Yang and Qian, 2017). Recent studies have identified several molecules involved in the innate immune signaling as O-GlcNAcylation targets. For example, O-GlcNAcylation of IKKβ, NF-κB p65, c-Rel and TAB1 enhances their activities and promotes the transcription of NF-κB target genes. (Ozcan et al., 2010; Pathak et al., 2012; Yang et al., 2015c). We recently identified the transcription factor STAT3 O-GlcNAcylation as an important mechanism antagonizing its activation (Li et al., 2017). However, the specific role of OGT in innate immune function and inflammation remains poorly defined.

RIPK3 is a member of the RIP family of serine/threonine kinases and contains an N-terminal kinase domain and a C-terminal RIP homotypic interaction motif (RHIM) (Silke et al., 2015). Through RHIM-mediated protein interaction, RIPK3 forms a necrosome complex with RIPK1 that is required for the induction of necroptosis, an inflammatory form of cell death (Galluzzi et al., 2017; Weinlich et al., 2017). Both RHIM and kinase activity of RIPK3 are essential for activation of downstream effector protein MLKL and execution of necroptosis (Wallach et al., 2016). In addition to its central role in necroptosis, elevated RIPK3 activation has been shown to promote inflammatory responses in both cell death-dependent and -independent manner (Alvarez-Diaz et al., 2016; Moriwaki et al., 2017; Najjar et al., 2016). Despite a well-studied signaling pathway leading to necrosome formation, intrinsic mechanism modulating RIPK3 activation is not well understood. In this study, we identified OGT-mediated RIPK3 O-GlcNAcylation at T467 as a key mechanism to block RHIM-mediated RIPK3-RIPK1 and RIPK3-RIPK3 interaction. Removal of OGT or RIPK3 O-GlcNAcylation promoted macrophage inflammatory response and necroptosis, both of which are dependent on RIPK3 RHIM domain and kinase activity. As a result, genetic deletion of Ogt in myeloid cells markedly exacerbated cytokine storm and host mortality in experimental sepsis. Therefore, our findings demonstrate a check mechanism against overzealous innate immune activation through OGT-mediated RIPK3 O-GlcNAcylation.

RESULTS

Lipopolysaccharide attenuates HBP activity and protein O-GlcNAcylation in macrophages

Increased glucose uptake and glycolysis has been well-documented in activated immune cells (Everts et al., 2014). We compared intracellular metabolite profiles between mock-treated and LPS-stimulated mouse bone marrow-derived macrophages (BMMs) by performing metabolomics analysis. Principal component analysis revealed a markedly altered metabolic profile upon LPS stimulation (Figure 1A). Pathway-enrichment analysis identified several sugar-related metabolic pathways (fructose and mannose, pyruvate and glycolysis) among the most differentially regulated pathways (Figures 1B and 1C). Many intermediate metabolites involved in glycolysis (Figure 1D) and the PPP (Figure 1E) showed increased abundance upon LPS challenge (Table S1). Several HBP metabolites were decreased after LPS stimulation, including HBP end product UDP-GlcNAc (Figures 1F and S1). Metabolic tracer analysis using 13C-glucose tracing also revealed an increased incorporation of glucose-derived carbon in glycolysis (Figure 1G) and decreased incorporation into HBP metabolites (Figure 1H and Table S2), which suggests that LPS attenuated HBP activity.

Figure 1. LPS stimulation affects glucose metabolism in mouse macrophages.

Figure 1.

Open in a new tab

(A and B) Total metabolite profiling in mouse BMMs stimulated with or without LPS (200 ng/ml) for 6 h was determined by the liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based metabolomics assay, assessed by principle component analysis (A) and pathway-enrichment analysis (B).

(C) A schematic of three glucose metabolic pathways, including the glycolysis (middle), PPP (upper) and HBP (lower).

(D–F) LPS-induced fold changes in intermediate metabolites of the glycolysis (D), PPP (E), and HBP (F) in mouse BMMs.

(G and H) LPS-induced fold changes in 13C-intermediate metabolites of the glycolysis (G) and HBP (H) in mouse BMMs in the presence of 13C6-glucose.

(I and J) Immunoblotting of OGT and total O-GlcNAc in mouse BMMs (I), or peritoneal macrophages isolated from mice 24 h after sham or cecal ligation and puncture (CLP) procedure (J). * P < 0.05, versus controls (two-tailed Student’s t-test (E–H)). Data are from one experiment representative of three experiments (A, B, D–F; mean ± s.d. of four biological replicates) or two experiments (G and H; mean ± s.d. of three biological replicates) or four experiments (I and J). Please also see Figure S1.

The availability of UDP-GlcNAc is an important determinant of OGT enzymatic activity (Hart et al., 2011). Consistent with decreased UDP-GlcNAc abundance, LPS-stimulated BMMs exhibited attenuated total protein O-GlcNAcylation without affecting OGT protein amount (Figure 1I). In an experimental sepsis model induced by the cecal ligation and puncture (CLP) procedure (Wen et al., 2010), peritoneal macrophages isolated from septic mice 24 hours after CLP exhibited markedly attenuated protein O-GlcNAcylation compared to those from sham mice (Figure 1J). The O-GlcNAc signal was abolished when anti-O-GlcNAc antibody was pre-incubated with 500 mM N-acetylglucosamine (GlcNAc), indicating the specificity of O-GlcNAc signal. These findings demonstrate that classical activation of macrophages leads to attenuated HBP activity and protein O-GlcNAcylation in vitro and in vivo.

OGT inhibits activation of the innate immune response

We examined the function of OGT-mediated O-GlcNAc signaling in the activation of the innate immune responses. Upon LPS challenge, Ogtfl/flxLyz2-cre BMMs (Li et al., 2018) produced significantly higher amounts of inflammatory mediators at transcript (Figure 2A) and protein (IL-6 and TNF-α) (Figure 2B) concentrations. Induction of Nos2 protein and nitrite production by LPS was also enhanced in Ogtfl/flxLyz2-cre BMMs (Figures 2C and 2D). Treatment with TLR2 (Pam3Cys) or TLR9 (CpG) agonists showed similar phenotype (Figures 2E and 2F). M2-associated gene transcripts (Figure S2A) and arginase-1 protein (Figure S2B) were normally induced in IL-4 treated Ogtfl/flxLyz2-cre BMMs, indicating no defect in Ogtfl/flxLyz2-cre BMMs M2 polarization. Furthermore, OGT deficient human monocyte-like THP-1 cells (Li et al., 2018) produced significantly higher amounts of inflammatory cytokines in response to TLR2, 4 or 9 agonists, suggesting that OGT negatively regulates cytokine production both in mouse and human cells (Figure S2C).

Figure 2. OGT deficiency enhances activation of the innate immune responses.

Figure 2.

Open in a new tab

(A-F) BMMs generated from Ogtfl/fl and Ogtfl/flxLyz2-cre mice were left untreated or stimulated with LPS (A–D, G–I) or Pam3Cys or CpG (E and F) for indicated periods. Transcripts of inflammatory genes (A and E), IL-6 and TNF-α proteins (B and F), and nitrite concentrations (D) in the supernatants were measured with RT-PCR, ELISA and Griess assay, respectively. Nos2 protein was assayed by immunoblotting (C) (G and H) Immunoblotting for NF-κB (G, left), and MAPK (H, left) signaling molecules and densitometric analysis (G and H, right).

(I) Immunoblotting of NF-κB p65, RelB and p50 in the cytosolic (left) and nuclear (right) compartments. * P < 0.05, versus controls (two-tailed Student’s t-test). Data are from one experiment representative of five experiments (A, B, D–F; mean ± s.d. of four biological replicates) or four experiments (C, G, H and I). Please also see Figure S2 and S3.

We have recently reported that OGT-mediated STAT3 O-GlcNAcylation antagonizes STAT3 phosphorylation and IL-10 production (Li et al., 2017). Further assays with the use of Ogtfl/flxLyz2-cre macrophages revealed that Ogt deletion indeed resulted in an enhanced Stat3 phosphorylation (Figure S3A) and IL-10 production (Figures S3B and S3C) upon TLR activation. Pretreatment of cells with a specific Stat3 inhibitor S31–201 (Siddiquee et al., 2007) completely abolished the increased IL-10 production in Ogtfl/flxLyz2-cre macrophages; however, increased IL-6 and TNF-α production still maintained in Ogtfl/flxLyz2-cre macrophages (Figure S3D). These results indicate that the hyperinflammatory response in Ogtfl/flxLyz2-cre macrophages is caused by Stat3-independent mechanism.

Activation of innate immune signaling such as the NF-κB and MAPK pathways is essential for TLR-induced cytokine production. We observed increased activation of the NF-κB pathway evidenced by phosphorylation of IKKα/β, IκBα and p65 in LPS-challenged Ogtfl/flxLyz2-cre BMMs (Figure 2G), as well as enhanced phosphorylation of Erk, but not p38 or Jnk (Figure 2H). Furthermore, by isolating macrophage cytosolic and nuclear compartments, we found a markedly increased nuclear translocation of p65, RelB and p50 in LPS-stimulated Ogtfl/flxLyz2-cre BMMs, lending further support for increased NF-κB activation (Figure 2I). In sum, these findings collectively demonstrate that OGT deficiency leads to the hyperactivation of TLR-mediated innate immune signaling.

Myeloid Ogt deletion exacerbates septic inflammation

To examine the function of myeloid-derived OGT in the innate immune response in vivo, we employed two septic inflammation models, endotoxin shock induced by intraperitoneal LPS injection and CLP-induced polymicrobial peritonitis (Wen et al., 2010). After administration of LPS at 15 mg per kg body weight, 60% of WT mice survived within 48 hours, whereas all Ogtfl/flxLyz2-cre mice died over the same period (Figure 3A). Analyses of inflammatory cytokines in the peritoneal lavage fluid or serum revealed an exacerbated cytokine storm in Ogtfl/flxLyz2-cre mice (Figures 3B and 3C). During a mild experimental sepsis model induced by two-puncture CLP procedure, Ogtfl/flxLyz2-cre mice were significantly susceptible to CLP-induced lethality in sepsis (Figure 3D), accompanied by significantly elevated inflammatory cytokine production in the peritoneal lavage fluid (Figure 3E), serum (Figure 3F) and lung homogenate (Figure 3G). Therefore, OGT in myeloid cells is crucial to limit hyperactivation of the innate immune response and protects hosts from sepsis-induced lethality. It has been well known that sepsis causes massive T cell apoptosis, which plays an important role to promote immunosuppression following sepsis (Hotchkiss et al., 2013). However, we found no significant difference in T cell-associated cytokines IL-2 and IFN-γ in serum (Figure S4A), or the numbers of CD3+CD4+ or CD3+CD8+ T cells in spleen (Figure S4B) or inguinal lymph nodes (Figure S4C) between septic WT and Ogtfl/flxLyz2-cre mice. Therefore, increased mortality in septic Ogtfl/flxLyz2-cre mice is less likely due to altered T cell responses.

Figure 3. Myeloid-derived OG protects mice from experimental sepsis.

Figure 3.

Open in a new tab

Ogtfl/fl and Ogtfl/flxLyz2-cre mice were injected intraperitoneally with 15 mg/kg body weight LPS (n=10 for each group) (A–C) or were subjected to sham or CLP procedure (n=8 to10 for each group) (D–G). Survival were recorded (A and D). IL-6, TNF-α, and IL-1β protein concentrations in peritoneal lavage (B and E), serum (C and F), and lungs homogenates (G) were measured by ELISA 24 h after LPS injection or CLP procedure. * P < 0.05, versus controls (two-tailed Student’s t-test). Data are from one experiment representative of three experiments (B, C, E–G; mean ± s.d.) or two experiments (A and F). Please also see Figure S4.

Deletion of Ripk3 abolishes hyperinflammation in Ogtfl/flxLyz2-cre macrophages and mice

We next sought to determine at which level OGT inhibits TLR-induced innate immune signaling by performing NF-κB-driven luciferase assay. OGT inhibited NF-κB-dependent luciferase gene transcription induced by MYD88, TRAF6, RIPK1 and RIPK3, but showed no inhibitory effect on IKK1-, IKK2-, or p65-driven NF-κB activation (Figure 4A). These results suggest that OGT is functioning at the RIPK1 and/or RIPK3 level. Nec-1 (Degterev et al., 2008) and GSK-872 (Mandal et al., 2014) are well-defined kinase activity inhibitors for RIPK1 and RIPK3, respectively. Treatment with GSK-872, but not Nec-1, abolished increased cytokine production in Ogtfl/flxLyz2-cre BMMs (Figures S5A and S5B), as well as in OGT deficient THP-1 cells (Figure S5C). We next tested BMMs generated from Ripk1K45A (Berger et al., 2014) or Ripk3K51A (Mandal et al., 2014) kinase-dead mice. Pharmacological inhibition of OGT by OSMI-1 (Li et al., 2018) promoted LPS-induced cytokine production in WT BMMs to the same concentrations as those in Ogtfl/flxLyz2-cre BMMs, indicating that OSMI-1 treatment caused an OGT-dependent hyperinflammatory response (Figures S5D and S5E). Importantly, OSMI-1 treatment selectively increased LPS-induced cytokine production in Ripk1K45A BMMs, but not in Ripk3K51A BMMs (Figures S5F and S5G). Together, these findings suggest that the kinase activity of RIPK3, but not RIPK1, is required for increased cytokine production in LPS-treated Ogtfl/flxLyz2-cre BMMs.

Figure 4. OGT inhibits the innate immune responses through RIPK3.

Figure 4.

Open in a new tab

(A) NF-κB-driven luciferase activities by co-expression of MYD88, TRAF6, RIPK1, RIPK3, IKK1, IKK2, or p65, in the presence or absence of the expression plasmid for OGT.

(B and C) IL-6 and TNF-α proteins (B) and nitrite (C) produced by Ogtfl/fl, Ogtfl/flxLyz2-cre, Ripk3−/− or Ogtfl/flxLyz2-creRipk3−/− BMMs stimulated with or without LPS.

(D−G) Survival rate (D), IL-6, TNF-α and IL-1β proteins in peritoneal lavage (E), serum (F), or lungs (G) in Ogtfl/fl (n=14), Ogtfl/flxLyz2-cre (n=10), Ripk3−/− (n=12) or Ogtfl/flxLyz2-creRipk3−/− mice (n=8) subjected to CLP procedure.

(H and I) IL-6 and TNF-α proteins (H) and nitrite (I) produced by Ogtfl/fl, Ogtfl/flxLyz2-cre, Mlkl−/− or Ogtfl/flxLyz2-creMlkl−/− BMMs stimulated with or without LPS.

(J–L) Survival rate (J), IL-6, TNF-α and IL-1β proteins in serum (K) or peritoneal lavage in Ogtfl/fl (n=12), Ogtfl/flxLyz2-cre (n=10), Mlkl−/− (n=15) or Ogtfl/flxLyz2-creMlkl−/− mice (n=13) mice subjected to CLP procedure. * P < 0.05, versus controls (two-tailed Student’s t-test (A–C, E–I, K and L)). The results shown are representative of four independent experiments (A–C; mean ± s.d. of four biological replicates) or two experiments (D–L). Please also see Figure S5.

It has been well established that genetic deletion of apoptosis-associated effector molecules such as caspase-8 promotes RIPK3-dependent necroptotic and inflammatory responses (Galluzzi et al., 2017; Weinlich et al., 2017). We found that compared to WT BMMs, Ogtfl/flxLyz2-cre BMMs generated higher amounts of both IL-1β and IL-6 in response to the agonists of either canonical (LPS plus ATP, nigericin, silica or alum) (Figure S5H) or noncanonical (LPS plus cholera toxin B (CTB)) inflammasomes (Figure S5I). Since LPS stimulation alone caused no IL-1β release in Ogtfl/flxLyz2-cre BMMs, we conclude that OGT deficiency enhances the “priming” phase, as evidenced by increased Il1b gene transcription (Figure 2A), but not the “activating” phase of inflammasome activation. Furthermore, we found no difference in caspase-8 protein amount between WT and Ogtfl/flxLyz2-cre BMMs (Figures S5J and S5K) and detected no O-GlcNAc signal on exogenously expressed caspase-8 (Figure S5L). Therefore, it seems less likely that OGT affects RIPK3 activity through caspase-8.

We generated Ogtfl/flxLyz2-creRipk3−/− mice to test whether additional deletion of Ripk3 could reverse the hyperinflammatory phenotype in Ogtfl/flxLyz2-cre macrophages and mice. Compared to Ogtfl/flxLyz2-cre BMMs, additional deletion of Ripk3 (Ogtfl/flxLyz2-creRipk3−/− ) completely abolished the increased cytokine generation (Figures 4B and S6A), as well as of Nos2 transcript (Figure S6A) and NO production (Figure 4C) upon LPS treatment. Furthermore, Ogtfl/flxLyz2-creRipk3−/− mice were completely rescued from sepsis-induced lethality compared to Ogtfl/flxLyz2-cre mice (Figure 4D). Consistently, increased production of inflammatory cytokines in septic Ogtfl/flxLyz2-cre mice was also abolished in septic Ogtfl/flxLyz2-creRipk3−/− mice (Figures 4E–4G). In sum, these results suggest that RIPK3 is a key effector that mediates the hyperinflammatory response in Ogtfl/flxLyz2-cre cells and mice. We further generated Ogtfl/flxLyz2-creMlkl−/−mice to examine whether the hyperinflammatory phenotype in Ogtfl/flxLyz2-cre macrophages was due to RIPK3-mediated innate immune signaling or secondary to MLKL-mediated necroptosis. Ogtfl/flxLyz2-creMlkl−/− BMMs produced cytokines (Figures 4H and S6B) and NO (Figure 4I) at the same concentrations as those in Ogtfl/flxLyz2-cre macrophages. No improvement in survival rate (Figure 4J) or inflammatory response (Figures 4K and 4L) was observed in Ogtfl/flxLyz2-creMlkl−/− mice compared to Ogtfl/flxLyz2-cre mice during CLP-induced sepsis. In sum, these findings indicate that OGT negatively regulates macrophage cytokine production and septic inflammation by restraining RIPK3-mediated inflammatory response.

Increased necroptosis response in Ogtfl/flxLyz2-cre macrophages due to RIPK3 hyperactivation

We next asked whether OGT affects necroptosis signaling through RIPK3. When stimulated with LPS plus a pan-caspase inhibitor zVAD, Ogtfl/flxLyz2-cre BMMs exhibited significantly enhanced necroptosis shown by the increased release of lactate dehydrogenase (LDH) (Figure 5A), HMGB1 and IL-1α, as well as augmented phosphorylation of RIPK3 and MLKL (Figure 5B). RIPK1 phosphorylation was similarly induced between WT and Ogtfl/flxLyz2-cre BMMs, indicating that OGT affected activation of RIPK3, but not RIPK1. Ogtfl/flxLyz2-cre BMMs also showed markedly increased SYTOX Green staining, a cell-impermeable DNA-binding fluorescence dye (Orozco et al., 2014), upon LPS plus zVAD treatment compared to WT BMMs, suggesting increased necroptosis (Figure 5C). We observed a slightly induced LDH release (Figures S6C and S6D) and a low amount of phosphorylated RIPK3 (Figures S6E and S6F) in Ogtfl/flxLyz2-cre BMMs upon LPS challenge alone. Pretreatment with GSK-872, but not Nec-1, completely abolished cell death in both WT and Ogtfl/flxLyz2-cre BMMs (Figure 5H). OSMI-1 treatment increased necroptosis, which phenocopied Ogt genetic deletion (Figure S6G) and required the kinase activity of RIPK3, but not RIPK1 (Figure S6H). Together, these findings suggest that RIPK3 kinase activity is responsible for increased necroptosis in Ogtfl/flxLyz2-cre BMMs.

Figure 5. OGT inhibits necroptosis through RIPK3.

Figure 5.

Open in a new tab

(A–C) Cell death assessed by LDH release (A) or Sytox Green staining (C), and phosphorylation of necroptosis signaling molecules (B) in Ogtfl/fl or Ogtfl/flxLyz2-cre BMMs left untreated or stimulated with LPS (200 ng/ml) plus zVAD (10 µM) for indicated periods.

(D) Immunoblotting of p-RIPK3 and p-MLKL in Ogtfl/fl or Ogtfl/flxLyz2-cre BMMs placed in a transwell system stimulated with LPS and zVAD.

(E) Immunoblotting of HMGB1 and IL-1α in the peritoneal lavage fluids of Ogtfl/fl and Ogtfl/flxLyz2-cre mice 24 h after sham or CLP procedure.

(F–H) Necroptosis signaling molecules in the NP-40-insoluble fractions (F), total or phosphorylated RIPK1 in RIPK3 immunoprecipitates (G), and cell death assessed by LDH release in the absence or presents of RIPK1 inhibitor Nec-1 (20 µM) or RIPK3 inhibitor GSK-872 (10 µM) (H).

(I–M) LDH release (I and K) and phosphorylation of necroptosis signaling molecules (J and L) in Ogtfl/fl, Ogtfl/flxLyz2-cre, Ripk3−/− , Ogtfl/flxLyz2-creRipk3−/− (I and J), Mlkl−/− or Ogtfl/flxLyz2-creMlkl−/− (K and L) BMMs left untreated or stimulated with LPS plus zVAD, or placed in culture medium with 2 mM glucose (M). * P < 0.05, versus controls (two-tailed Student’s t-test (a, g, h)). Data are from represent of four independent experiments (A, H, I, K and M; mean ± s.d. of four biological replicates) or three experiments (B–G, J and L). Please also see Figure S6.

We next used a transwell coculture system to examine whether enhanced necroptosis was caused by an altered extracellular cytokine environment for Ogtfl/flxLyz2-cre macrophages. Despite the genotypes of cells in the upper chambers, Ogtfl/flxLyz2-cre BMMs in the lower chambers showed increased phosphorylation of RIPK3 and MLKL upon LPS plus zVAD stimulation, supporting a cell-intrinsic mechanism of OGT in affecting the necroptosis (Figure 5D). Following CLP procedure, septic Ogtfl/flxLyz2-cre mice contained markedly elevated amounts of HMGB1 and IL-1α in the peritoneal lavage fluids compared to septic WT mice (Figure 5E). Upon necroptosis induction, RIPK3 and RIPK1 form an amyloid signaling complex, known as necrosome (Li et al., 2012), which can be biochemically enriched in a NP-40-insoluble cellular fraction (Najjar et al., 2016). We compared necrosome formation between WT and Ogtfl/flxLyz2-cre BMMs and found that appearance of RIPK1, RIPK3 and MLKL in the detergent insoluble cellular fractions was heavily augmented in Ogtfl/flxLyz2-cre BMMs (Figure 5F). Coimmunoprecipitation assays revealed an increased association between endogenous RIPK3 and RIPK1, either in a phosphorylated or total form, in necroptotic Ogtfl/flxLyz2-cre BMMs (Figure 5G). These findings support an anti-necroptotic effect of OGT in macrophages.

Compared to Ogtfl/flxLyz2-cre BMMs, Ogtfl/flxLyz2-creRipk3−/− BMMs showed a complete rescue from cell death in response to LPS plus zVAD (Figure 5I) or LPS alone (Figure S6I), accompanied by the loss of MLKL phosphorylation (Figure 5J). Similarly, Ogtfl/flxLyz2-creMlkl−/− BMMs were also rescued from cell death (Figures 5K and S5J). Importantly, Ogt−/− still caused increased RIPK3 phosphorylation in the Mlkl−/− genetic background (Ogtfl/flxLyz2-creMlkl−/− versus Mlkl−/− BMMs) during necroptosis, indicating that OGT affects RIPK3 activity independently of MLKL-mediated cell death (Figure 5L). In sum, these findings suggest that Ogtfl/flxLyz2-cre BMMs exhibits an increased necroptotic response, which is dependent on RIPK3. We also tested whether OGT affects apoptosis. Hallmarks of apoptosis including the drop of intracellular ATP concentration, cleavage of poly (ADP-ribose) polymerase (PARP) and caspase 3 were comparable between WT and Ogtfl/flxLyz2-cre BMMs in response to TNF-α plus cycloheximide (Figures S6K and S6L) or staurosporine (Figures S6M and S6N), suggesting no involvement of OGT in either extrinsic or intrinsic apoptotic pathway. We finally asked whether the inhibitory effect of OGT on RIPK3 phosphorylation and activation occurred in a physiological condition by lowering the glucose concentration in culture medium. Switching the glucose concentration from 25 to 2 mM increased necroptosis in WT BMMs to those observed in Ogtfl/flxLyz2-cre BMMs, but showed no effect in the cells lack of RIPK3 or MLKL (Figure 5M). These results indicate that glucose metabolism affects the RIPK3-MLKL necroptosis pathway through OGT.

O-GlcNAcylation of RIPK3 on T467 inhibits inflammation and necroptosis

We sought to determine whether OGT directly O-GlcNAcylates RIPK3 to affect its function. Immunoprecipitated endogenous RIPK3 exhibited a positive O-GlcNAc signal in WT mouse BMMs, but not Ogtfl/flxLyz2-cre BMMs (Figure 6A). Reciprocal precipitation with the use of succinylated wheat germ agglutinin (sWGA) beads, a widely used biochemical strategy to enrich O-GlcNAcylated proteins (Hart et al., 2011), also detected RIPK3 as an O-GlcNAcylated protein (Figure 6B). Both RIPK3 O-GlcNAcylation and the association between RIPK3 and OGT increased upon LPS stimulation (Figure 6C), despite attenuated total protein O-GlcNAcylation (Figure 1I), which suggests that OGT actively and specifically promotes RIPK3 O-GlcNAcylation and prevents its activation. In contrast, necroptosis induction by LPS plus zVAD caused the opposite effects (Figure 6D). As expected, lowering medium glucose concentration from 25 to 2 mM markedly decreased RIPK3 O-GlcNAcylation with or without LPS treatment (Figure 6E). These findings explain the lack of necroptosis in LPS-stimulated WT macrophages and identify diminished RIPK3 O-GlcNAcylation as an important step mediating necroptosis. Indeed, treatment with PUGNAc, a widely used inhibitors of the O-GlcNAcylase (OGA) which potently enhances protein O-GlcNAcylation (Li et al., 2017), rescued RIPK3 O-GlcNAcylation and attenuated RIPK3 phosphorylation (Figure 6F) and cell necroptosis (Figure 6G).

Figure 6. O-GlcNAcylation of RIPK3 on T467 suppresses inflammation and necroptosis.

Figure 6.

Open in a new tab

(A–G) Total cell lysates of Ogtfl/fl or Ogtfl/flxLyz2-cre BMMs left untreated (A and B) or stimulated with LPS alone (C) or LPS plus zVAD (D) with or without PUGNAc pretreatment (E). Immunoprecipitation with anti-RIPK3 antibody (A, C–E) or succinylated wheat germ agglutinin (sWGA) beads to pull down O-GlcNAcylated proteins (B).

(H) FLAG-tagged RIPK3 overexpressed in 293T cells together with OGT WT or enzyme-dead K908A mutant was immunoprecipitated with anti-FLAG beads, followed by immunoblotting with anti-O-GlcNAc antibody.

(I) Domain organization of human RIPK3 and the RHIM sequence.

(J) O-GlcNAcylation of the full-length, N-terminal or C-terminal fragments of RIPK3 overexpressed in 293T cells in the presence or absence of WT OGT.

(K and L) O-GlcNAcylation of FLAG-tagged RIPK3 WT or mutants overexpressed in 293T cells together with OGT was analyzed as described in (H).

(M–P) RIPK3-silenced THP-1 cells were virally transfected with RIPK3 WT or T467A mutant, followed by LPS (M and N) or LPS plus zVAD (O and P) stimulation. Transcripts of RIPK3, IL6, TNFA, and IL1B (M), total and phosphorylated RIPK3, Erk, IKKα/β, IκBα and p65 (N) or necroptosis signaling molecules (P), and LDH release (O) was assayed. *P < 0.05, versus controls (two-tailed Student’s t-test (L and N)). Data are from one experiment representative of three independent experiments (A–G, I, J, K, M and O) or four experiments (L and N; mean ± s.d. of three biological replicates). Please also see Figure S7.

We next sought to identify the O-GlcNAcylation site(s) on RIPK3. O-GlcNAcylation of exogenously expressed human RIPK3 in 293T cells was sharply enhanced by WT OGT, but not enzyme-dead mutant of OGT (K908A) (Lazarus et al., 2011), indicating a requirement of OGT enzymatic activity for RIPK3 O-GlcNAcylation (Figure 6H). O-GlcNAcylation of exogenously expressed mouse Ripk3 could also be detected in 293T cells (Figure S7A). RIPK3 contains an N-terminal kinase domain and a C-terminal RHIM motif, both of them are necessary for necroptosis signaling. We therefore generated N-terminal (1–310 aa) and C-terminal (311–518 aa) truncated fragments of RIPK3 (Figure 6I) and detected the O-GlcNAcylation on its C-terminal fragment (Figure 6J). Through mass spectrometry (MS) analysis, several potential O-GlcNAcylation sites were detected on both full-length and C-terminal fragment of RIPK3 (Table S3). A follow-up site-directed mutagenesis strategy allowed us to reveal that a single mutant (T467A) inside the RHIM motif (Figure 6I) lost O-GlcNAc signal without affecting RIPK3 protein abundance (Figures 6K and 6L). It should be noted that T467 on human RIPK3 is only partially conserved among mammalian species (Figure S7B), suggesting a possibility that additional functional O-GlcNAcylation site(s) may exist in other species.

Given the hyperinflammatory and hyper-necroptotic phenotype of Ogt macrophages, we hypothesized that loss of RIPK3 O-GlcNAcylation causes the same effect. Human RIPK3 cannot induce necroptosis in mouse cells because it fails to interact with mouse MLKL (Sun et al., 2012). We therefore used short hairpin RNA (shRNA) to generate RIPK3-silenced THP-1 cells, followed by the reconstitution with either RIPK3 WT or T467A mutant. In response to LPS challenge, RIPK3-silenced cells reconstituted with RIPK3 T467A showed significantly increased cytokine production (Figure 6M), as well as enhanced NF-κB and Erk activation (Figure 6N), compared to those reconstituted with WT RIPK3. Since Ogtfl/flxLyz2-cre BMMs failed to show robust increase in the activation of NF-κB and Erk signaling upon LPS stimulation compared to WT BMMs (Figures 2G and 2H), it remains to be determined how RIPK3 activation upon OGT loss triggers increased LPS-induced cytokine responses. Furthermore, RIPK3 T467A reconstitution resulted in an augmented necroptosis (Figure 6O) and phosphorylation of RIPK3 and MLKL (Figure 6P) compared to WT RIPK3. These results demonstrate that RIPK3 O-GlcNAcylation on T467 negatively regulates the innate immune activation and necroptosis response.

O-GlcNAcylation of RIPK3 on T467 suppresses its RHIM functions

We finally explored the molecular mechanism by which O-GlcNAcylation of RIPK3 antagonizes its functions. RIPK3 RHIM region has been demonstrated to be essential for both inflammation and necroptosis pathway (Silke et al., 2015; Wallach et al., 2016), but the kinase activity of RIPK3 is not absolutely required for inflammation (Najjar et al., 2016). We found that overexpressed OGT inhibited RIPK3 phosphorylation in 293T (Figure 7A). Since RIPK3 T467 is localized in the RHIM region, we reasoned that O-GlcNAcylation of RIPK3 on T467 may function by suppressing RHIM-mediated effects such as RIPK3-RIPK1 hetero- and RIPK3-RIPK3 homo-interaction, as well as RIPK3 kinase activity. Indeed, overexpression of WT OGT, but not OGT enzyme-dead mutants (K908A, H508A) (Lazarus et al., 2011), efficiently abolished both RIPK3-RIPK1 hetero-(Figure 7B), RIPK3-RIPK3 homo-interaction (Figure 7C), and RIPK3 phosphorylation under both conditions. Loss-of-O-GlcNAcylation mutation of RIPK3 (T467A) generated opposite results (Figure 7D). Importantly, RIPK3 T467A mutant was resistant to OGT-inhibited RIPK3 phosphorylation and RIPK3-RIPK1 interaction. Furthermore, RIPK3 RHIM mutant (RHIMmut) showed defects in both RIPK3-RIPK1 interaction and RIPK3 phosphorylation (Figure 7E), but RIPK3 kinase-dead mutants (D160N, S199A) (McQuade et al., 2013) still interacted with RIPK1 (Figure 7F), supporting the notion that RHIM-mediated RIPK3-RIPK1 interaction is an upstream event leading to RIPK3 phosphorylation. In sum, these findings suggest that OGT-mediated RIPK3 O-GlcNAcylation on T467 suppresses RHIM-mediated RIPK3-RIPK1 interaction and downstream RIPK3 kinase activation.

Figure 7. O-GlcNAcylation of RIPK3 inhibits RHIM-mediated RIPK3-RIPK1 interaction.

Figure 7.

Open in a new tab

(A–D) FLAG-tagged RIPK3 was overexpressed in 293T cells together with indicated expression plasmids. Immunoprecipitation with anti-FLAG beads, followed by immunoblotting with specific antibodies against O-GlcNAc, p-RIPK3, RIPK1 and GFP.

(E–G) GFP-tagged RIPK3 WT or mutants was overexpressed in 293T cells. Immunoprecipitation with anti-GFP beads, followed by immunoblotting with specific antibodies against O-GlcNAc, p-RIPK3 or RIPK1.

(H and I) Transcripts of IL6 and TNFA (H) and LDH release (I) from RIPK3-silenced THP-1 cells virally transfected with RIPK3 WT or various mutants, followed by LPS (H) or LPS plus zVAD (I) stimulation.

(J) Structural modeling of RIPK3 by using the I-TASSER program. O-GlcNAcylated T467 is red and the amino acids VQVG of the RHIM motif is highlighted as pink sphere. * P < 0.05, versus controls (two-tailed Student’s t-test (g, h)). Data are from one experiment representative of three experiments (A–F) or four experiments (H and I; mean ± s.d. of four biological replicates).

We next examined if RIPK3 T467A-induced hyperinflammatory and hyper-necroptotic responses were dependent on its RHIM and/or kinase activity. Both T467A&RHIMmut or T467A/S199A double mutant lacked RIPK3 phosphorylation (Figure 7G). RIPK3-silenced THP-1 cells reconstitution with either RIPK3 T467A&RHIMmut or T467A/S199A abolished increased cytokine production in response to LPS (Figure 7H) and elevated necroptotic response (Figure 7I), compared to those reconstituted with RIPK3 T467A. These findings demonstrated that loss of O-GlcNAcylation on RIPK3 enhances inflammatory and necroptosis signaling, dependent on RHIM-mediated protein interaction and the resultant kinase activation (Figures S7C–S7F). We further performed structural modeling of RIPK3 by using the I-TASSER program (Yang et al., 2015a) and found that the sidechain of O-GlcNAc at T467 stays in proximity of the RHIM region core amino acids VQVG and likely perturbs RHIM-mediated protein interaction through steric hinderance (Figure 7J).

DISCUSSION

Previous studies have provided biochemical evidence to support the notion that the O-GlcNAc signaling promotes inflammation. Several key molecules in the TLR-NF-κB signaling pathway have been identified to be O-GlcNAcylated. O-GlcNAcylation of these proteins has been shown to enhance their functional activities and promotes the transcription of NF-κB target genes (Ozcan et al., 2010). Our recent study also observed an enhanced inflammatory response and disease severity in chemically induced colitis, due to attenuated STAT3-IL-10 signaling, when OGT expression and enzymatic activity were upregulated in myeloid cells (Li et al., 2017). In this study, we demonstrated an unexpected inhibitory effect of OGT on innate immune activation and necroptosis signaling through RIPK3 O-GlcNAcylation. Therefore, the net effect of OGT-mediated O-GlcNAc signaling in the immune system and inflammation seems to be multifaceted due to the involvement of a variety of target proteins in different immune signaling pathways. Both the increase or decrease in OGT-mediated protein O-GlcNAcylation resulted in a similar hyperinflammatory response through distinct mechanisms. On one hand, elevated O-GlcNAc signaling promoted activation of the innate immune cells by increasing NF-κB signaling, as well as counteracting the anti-inflammatory STAT3-IL-10 signaling. On the other hand, loss of O-GlcNAc signaling removed an inhibitory mechanism of RIPK3 activation and consequently led to enhanced inflammatory responses and inflammation-associated necroptosis, despite increased IL-10 production. We therefore propose that loss of homeostasis in the O-GlcNAc signaling, instead of a simple one-way increase or decrease, is an important metabolic mechanism underlying the pathogenesis of inflammatory diseases.

The aforementioned studies raise an important question on how to regulate OGT-mediated O-GlcNAc signaling under pathophysiological conditions. Previous studies have shed light on two important mechanisms: control of UDP-GlcNAc availability through regulation of HBP activity and targeting OGT gene transcription and protein degradation. First, the availability of UDP-GlcNAc produced through HBP metabolic process is an important determinant of OGT enzymatic activity (Hart et al., 2011). The first and rate-limiting enzyme in HBP, glutamine fructose-6-phosphate transaminase (GFPT), represents a key mediator to affect downstream OGT signaling by determining UDP-GlcNAc concentration. A recent study reports that GFPT is transcriptionally upregulated under ER stress, leading to increased protein O-GlcNAcylation and providing a protective effect against ischemia-reperfusion injury in heart (Wang et al., 2014). Our study observed a decreased HBP activity and protein O-GlcNAcylation by TLR4 activation, despite increased glycolysis and PPP metabolic activities. In contrast, in a parallel study, we observed an increased HBP flux activity and protein O-GlcNAcylation when the retinoic-acid inducible gene I (RIG-I) was activated by intracellular RNA ligands (Li et al., 2018). The opposite changes in the O-GlcNAc signaling due to activation of distinct innate immune sensors highlight that GFPT-mediated HBP activity is an important mechanism regulating OGT function. How GFPT-controlled HBP activity is differentially regulated during activation of distinct pathogen recognition receptor signaling requires further investigation. Second, the abundance of OGT protein is negatively regulated by a ubiquitination-mediated protein degradation process (Yang et al., 2015b), whereas OGT gene expression is promoted by the transcriptional activity of Nrf2 (Li et al., 2017). Whether OGT function is affected by K63-linked protein ubiquitination, a well-established protein modification for signal transduction, has not been determined yet. In sum, similar to diverse functions of OGT by targeting multiple downstream targets, upstream mechanisms which modulate OGT function are also diversified.

Recent studies have identified several intracellular amyloid signaling complexes, including necrosome (Li et al., 2012), ASC inflammasome (Lu et al., 2014), and MAVS signalosome (Cai et al., 2014), which initiate distinct important innate immune signaling. One common feature of these high-molecular weight complexes is well-characterized protein-protein binding patterns depending on unique domains on each protein component. The RHIM region of RIPK3, one best-studied protein-binding motif, has been shown to be required for the formation of necrosome and execution of necroptosis. Depending which initial sensors that receive extracellular or intracellular signals, RIPK3 can directly binds to TRIF (Kaiser et al., 2013), RIPK1 (Li et al., 2012; Mompean et al., 2018) or ZBP1 (Z-DNA binding protein 1; also known as DAI) (Lin et al., 2016; Newton et al., 2016) through RHIM motif to activate necroptosis. In addition to its central role in necroptosis, recent studies suggest that RIPK3 also contributes to activation of immune response in cell death-dependent and -independent manner (Alvarez-Diaz et al., 2016; Moriwaki et al., 2017; Najjar et al., 2016). RIPK3 RHIM motif is critical for both necroptosis and inflammatory response both in vitro and in vivo. Our study identified RIPK3 T467, which is located in proximity of the RHIM region core amino acids VQVG, as an O-GlcNAcylation site. Loss of this O-GlcNAcylation promoted RHIM-mediated RIPK3-RIPK1 hetero- and RIPK3-RIPK3 homo-interaction, RIPK3 activation, downstream inflammatory and necroptosis signaling, and cytokine storm in experimental sepsis. These findings support an essential role of RHIM-mediated assembly of necrosome signaling complex in promoting immune activation and tissue damage in inflammatory diseases such as sepsis.

Previous studies report that activated RIPK3 in macrophages deficient of apoptosis-associated molecules (caspase-8 or IAP) could activate inflammasome in response to LPS alone (Galluzzi et al., 2017; Weinlich et al., 2017). Inflammasome activation leads to caspase-1 activation and release of proinflammatory cytokines IL-1β and IL-18 among other substrates (Davis et al., 2011). Both canonical and noncanonical inflammasomes have been indicated to play important role promoting septic inflammation. We observed that upon LPS stimulation, loss of OGT-mediated O-GlcNAcylation caused RIPK3 phosphorylation and activation, indicating that OGT deficiency may increase RIPK3-dependent IL-1β production. Indeed, we observed increased IL-1β concentration in septic Ogtfl/flxLyz2-cre mice, which could be reversed by Ripk3 deletion. Since IL-1β is a prototypical pyrogen to drive cytokine storm and tissue damage in sepsis, excessive IL-1β may contribute to elevated mortality in septic OgtΔmye mice. In sum, our results provide a mechanistical link between OGT-mediated glucose metabolism and key immune signaling in the innate immune system and expand our current understanding of metabolic regulation of the immune function and inflammation-associated diseases. Targeting RIPK3 O-GlcNAcylation presents a potential therapeutic strategy for combating multiple inflammatory diseases.

STAR★METHODS

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources and reagents should be directed to will be fulfilled by the Lead Contact, Haitao Wen (haitao.wen@osumc.edu).

EXPERIMENTAL MODELS AND SUBJECT DETAILS

Mice.

Ogtfl/flxLyz2-cre mice were generated by crossing the Ogtfl/fl mice (Shafi et al., 2000) with lysosome M-Cre mice. C57BL/6 mice and lysosome M-Cre mice were purchased from Jackson Laboratories. Ripk3−/− (Newton et al., 2004) and Mlkl−/− (Murphy et al., 2013) mice have been previously described. Ogtfl/flxLyz2-creRipk3−/− and Ogtfl/flxLyz2-creMlkl−/− mice were generated by crossing Ogtfl/flxLyz2-cre mice with Ripk3−/− mice and Mlkl−/− mice, respectively. All mice were housed in SPF facilities at 21°C and 31% humidity. All mice were maintained und er general housing environment and fed sterilized food (Chow TD. 7912; Harlan Teklad) and autoclaved water ad libitum. Mice were weaned at 28 days after birth, separated into same sex groups and all in vivo experiments were conducted in accordance with the guidelines established by the Ohio state University and the National Institute of Health Guide for the Care and Use of Laboratory Animals and the Institutional Animal Care and Use Committee (IACUC).

Reagents and antibodies.

LPS and CpG were from Invivogen. Pam3Cys was from EMC Microcollections. Nec-1 was from Enzo. GSK-872 and zVAD were from Millipore. S3I-201, cycloheximide, PUGNAc, 13C6-glucose and N-acetylglucosamine (GlcNAc) were from Sigma-Aldrich. Lipofectamine 2000 and the Griess Reagent Kit for nitrite quantification were from Thermo Fisher Scientific. Antibodies for immunoblotting included anti-OGT, anti-O-GlcNAc, anti-mouse p-RIPK1, anti-RIPK1, anti-p-IKKα/β (S176/180), anti-p-IκBα (S32), anti-IκBα, anti-p-p65 (S536), anti-p65, anti-p-ERK1/2 (T202/Y204), anti-ERK1/2, anti-p-JNK (T183/Y185), anti-p-p38 (T180/Y182), anti-p38, anti-p-STAT3, anti-STAT3, anti-Histone H3 (Cell Signaling Technology), anti-JNK1, anti-GFP, HRP-conjugated anti-β-actin (Santa Cruz Biotechnology), HRP-conjugated anti-Flag (Sigma-Aldrich), HRP-conjugated anti-Myc (Roche), anti-IKKα, anti-NOS2, anti-GAPDH (Millipore), anti-RIPK3 (Novus Biologicals), anti-mouse MLKL (Abgent), anti-human p-RIPK3, anti-human p-MLKL, anti-mouse p-MLKL, anti-HMGB1 (Abcam), anti-IL-1α (R&D Systems) and anti-mouse p-RIPK3 (Genentech). Antibody-conjugated agarose for immunoprecipitation included anti-Flag agarose (Sigma-Aldrich), anti-c-Myc agarose (Thermo Scientific), GFP-Trap agarose (Chromotek), and sWGA agarose (Vector Laboratories).

Cell culture and stimulation.

292T cells, L929 and THP-1 cells were purchased from ATCC and maintained under a humidified atmosphere of 5% CO2 at 37°C in DMEM or RPMI 1640 supplemented with 10% fetal bovine serum (Sigma-Aldrich), respectively. To generate RIPK3-silenced THP-1 cells, GIPZ lentivector carrying shRNA targeting human RIPK3 (V2LHS_77679, Open Biosystems) were transfected into THP-1 cells. Infected cells were selected with 2 µg/ml puromycin for 5 d. BMMs were generated from mice in the presence of L-929 conditional medium, as previously described (Li et al., 2017). Peritoneal macrophages were collected from peritoneal lavage with 10-ml sterile DPBS containing 2% FBS. Macrophages were stimulated with LPS (200 ng/ml), Pam3Cys (1 µg/ml), CpG (2 µg/ml), or LPS plus zVAD (10 µM) for various periods as indicated in the Figure Legends. Cell culture supernatants were collected for ELISA or LDH release assay. Cells were collected for immunoblotting.

RT-PCR.

Total RNA was extracted from in vitro cultured cells using Trisure (Bioline). cDNA was synthesized with Moloney murine leukemia virus reverse transcriptase (Invitrogen) at 38°C for 60 min. RT-PCR was perform ed using SYBR Green PCR Master Mix (Applied Biosystems) in StepOnePlus detection system (Applied Biosystems). The fold difference in mRNA expression between treatment groups was determined by ∆∆Ct method. β-actin was used as an internal control. The primer pair sequences of individual genes are listed in the Supplemental Table 4.

Immunoblotting.

Electrophoresis of proteins was performed by using the NuPAGE system (Invitrogen) according to the manufacturer’s protocol. Briefly, cultured BMMs were collected and lysed with RIPA buffer. Proteins were separated on a NuPAGE precast gel and were transferred onto nitrocellulose membranes (Bio-Rad). Appropriate primary antibodies and HRP-conjugated secondary antibodies were used and proteins were detected using the Enhanced Chemiluminescent (ECL) reagent (Thermo Scientific). The images were acquired with ChemiDoc MP System (Bio-Rad).

ELISA.

Cytokines in supernatant from in vitro cultured cells or cytokines in the peritoneal lavage fluids, serum or lung homogenates from animal experiments were quantified using the ELISA Set for mouse IL-6, IL-1β and TNF-α (BD Biosciences) according to the manufacturer’s protocol.

Nitrite quantitation.

Nitrite generated from LPS-stimulated BMM were measured using the Griess Reagent Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol.

Experimental sepsis animal models.

For endotoxin shock model, mice were intraperitoneally administrated with LPS at 15 mg per kg body weight. CLP-induced polymicrobial peritonitis has been previously described (Wen et al., 2010). Briefly, mice were anesthetized with a combination of ketamine HCL and xylazine administrated intraperitoneally. A 1-cm midline incision was made to the ventral surface of the abdomen under sterile surgical conditions to expose the cecum. The cecum was partially ligated with a 3.0 silk suture at its base and punctured two times with a 21-G needle. Sham-operated mice underwent the identical operation except for the ligation and puncture. The cecum was returned to the abdomen cavity and closed with a surgical staple. Mice in both groups were treated with antibiotics intraperitoneally 6 and 24 hours after surgery.

Plasmids and molecular cloning.

pCMV vector expressing Myc-tagged human OGT has been described (Chen et al., 2013). FLAG-tagged RIPK3 (#78815), GFP-tagged RIPK3 (#41387), RIPK3 RHIMmut (#41385) and RIPK3 D160N (#41386) were purchased from Addgene. GFP-tagged RIPK3 S199A was kindly provided by Dr. Francis Chan (McQuade et al., 2013).

To generate the N- and C-terminal fragments of RIPK3, pcDNA3 vector expressing full-length RIPK3 was used as a template for PCR. To generate the lentivector expressing RIPK3, full-length RIPK3 WT or mutants were subcloned into the pWPXLd-EGFP lentivector (Addgene #12258). All primers used for cloning are listed in the Supplemental Table 5. To generate a series of RIPK3 mutant constructs and OGT K908A, H508A mutant, Phusion Site-Directed mutagenesis Kit was used according to the manufacturer’s instructions (Thermo Scientific). Primers for mutagenesis PCR are listed in the Supplemental Table 6. All cloned genes were double-checked by sequencing.

Cell transfection.

As indicated in the Figure Legends, 293T cells were transfected for 30 h with a combination of expression plasmids for OGT WT, K908A or H508A mutant, and RIPK3 WT or mutants with X-tremeGENE HP DNA Transfection Reagent (Roche). For lentivector-based transduction, 293T cells were employed to package the pWPXLd lentivirus expressing RIPK3 WT or various mutants, which were further used to transduce RIPK3-silenced THP-1 cells.

Luciferase assay.

293T cells were transfected with NF-κB-driven luciferase reporter construct together with expression plasmids for MYD88, TRAF6, RIPK1, RIPK3, IKK1, IKK2 or p65, in the presence or absence of OGT using FuGENE6 (Roche). Empty pcDNA3 was used to maintain equal DNA amounts for transfection. Cells were harvested at 30 h post transfection. Luciferase units were measured as previously described (Lei et al., 2012).

Metabolomics.

3 × 10 6 BMMs left untreated or stimulated with LPS for 4 h were harvested. The metabolite extraction was performed as previously described (Gunda et al., 2016). The media was aspirated, and the cells were washed twice with LC-MS grade water before lysing the cells. The metabolites were extracted using cold 80% methanol/water mixture and resuspended in 50% methanol/water mixture for further analysis using LC-MS/MS. A selected reaction monitoring (SRM) LC-MS/MS method with positive and negative ion polarity switching on a Xevo TQ-S mass spectrometer was used for analysis. Peak areas integrated using MassLynx 4.1 (Waters Inc.) were normalized to the respective protein concentrations. The resultant peak areas were subjected to relative quantification analyses with MetaboAnalyst 3.0. Further, principal component analysis, pathway impact analysis and heatmap were performed using MetaboAnalyst 3.0 software.

Nuclear Magnetic Resonance (NMR) Spectroscopy analysis of 13C-glucose metabolism.

13C-labeling of UDP-GlcNAc from U-13C6-Glucose was detected from cell lysates using NMR spectroscopy-based metabolomics analysis. Briefly, samples were dried and then reconstituted using 550 µL of 50 mM phosphate buffer in 100% D2O at pH 7.2 (uncorrected) with 500 µM 3-(tetramethysilane) propionic acid-2,2,3,3-d4 (TMSP) as a chemical shift reference. 500 µL of the supernatant was transferred to a 5 mm NMR tube for data acquisition. The NMR data were collected at 298K on an AVANCE III-HD 700 MHz spectrometer (Bruker) equipped with 5 mm quadruple resonance QCI-P cryoprobe (1H, 13C, 15N and 31P) using ICON-NMR software (Bruker). A (2D) 1H-13C heteronuclear single quantum coherence (HSQC) spectra was collected for each of the samples. The 2D 1H-13C HSQC spectra were collected with 1K data points and a spectrum width of 11160 Hz in the direct dimension and 128 data points and a spectrum width of 29052 Hz in the indirect dimension. The 2D 1H-13C HSQC spectra were processed with NMRPipe (Delaglio et al., 1995). The spectra were Fourier-transformed, manually phased, zero-filled, apodized with a sine-bell window function, and baseline-corrected following solvent subtraction. All spectra were referenced to TMSP at 0 ppm. The processed spectra were then analyzed using NMRViewJ Version 8.0 (Johnson, 2018).

RIPK3 O-GlcNAcylation site mapping.

MS strategy was employed to identify RIPK3 O-GlcNAcylation sites, as described in our recent study (Li et al., 2017). Briefly, immunoprecipitated RIPK3 full length or C-terminal fragment from 293T cells was subjected to SDS-PAGE. The corresponding bands were excised and the proteins were reduced with DTT, alkylated with iodoacetamide, and digested with trypsin overnight, then subjected to LC-MS/MS analysis using a nanoAcquity (Waters Corp) coupled to an LTQ Orbitrap Velos (Thermo Scientific). The LTQ Orbitrap Velos was operated in data-dependent mode, and the 10 most intense precursors were selected for collision-induced dissociation (CID) fragmentation. Raw data files were processed using Proteome Discoverer (PD) version 2.0 (Thermo Scientific). Peak lists were analyzed using Sequest against a Homo sapiens Uniprot database. The following parameters were used to identify tryptic peptides for protein identification: 0.6 Da product ion mass tolerance; 10 ppm precursor ion mass tolerance; up to two missed trypsin cleavage sites; hexNAc (+203.0794 Da) of N/S/T; carbamidomethylation of Cys was set as a fixed modification; oxidation of M and phosphorylation of S/T/Y were set as variable modifications. The Percolator node was used to determine false discovery rates (FDR) and a peptide FDR of 5% were used to filter all results.

RIPK3 structure prediction.

The structure of human RIPK3 is predicted by the program I-TASSER (Yang et al., 2015a). The O-GlcNAcylation was added onto the predicted structure using PyMOL 2.1 software.

Statistics.

All experiments were performed a minimum of three independent replications. Statistical analysis was carried out with Prism 5.0 for Macintosh. All data are shown as mean ± s.d.. The mean values for biochemical data from each group were compared by two-tailed Student’s t-test. Comparisons between multiple time points were analyzed by repeated-measurements analysis of variance (ANOVA) with Bonferroni post-tests. In all tests, p values of less than 0.05 were considered statistically significant.

Supplementary Material

1
2

Table S1. Related to Figure 1. List of metabolites in WT macrophages treated with or without LPS

Highlights.

  • LPS treatment causes a decrease in HBP activity and protein O-GlcNAcylation

  • OGT deficiency increases activation of innate immune response and necroptosis

  • O-GlcNAcylation of RIPK3 on T467 inhibits RIPK3-RIPK1 and RIPK3-RIPK3 interaction

ACKNOWLEDGMENTS:

We thank A.S. Baldwin for the NF-κB-driven luciferase reporter construct, the expression plasmids for IKK1, IKK2, and RELA; F.K. Chan for RIPK3 WT and mutant expression plasmids; H.K. Lin for TRAF6 expression plasmid; X. Yu for OGT expression plasmid; E.S. Mocarski for Ripk3K51A bone marrow cells; V.M. Dixit for Ripk3−/− mice; J.M. Murphy for Mlkl−/− mice; and M. Yuan for the metabolomics assay. Supported by the National Institutes of Health R01GM120496 (H.W.), 5P01CA120964 (J.M.A.), 5P30CA006516 (J.M.A.), P30CA016086 (UNC Lineberger Comprehensive Cancer Center), R01CA163649 (P.K.S.), R01CA210439 (P.K.S.), R01CA216853 (P.K.S.), P30GM103335 (R.P.), P20GM113126 (R.P.), R00DE024173 (Y.L.L.), R01DE026728 (Y.L.L.), and R01DK102648 (X.Y). Metabolic tracer experiments were performed in facilities renovated under NIH grant RR015468–01.

Footnotes

COMPETING FINANCIAL INTERESTS: The authors declare no competing financial interests.

References

  1. Alvarez-Diaz S, Dillon CP, Lalaoui N, Tanzer MC, Rodriguez DA, Lin A, Lebois M, Hakem R, Josefsson EC, O’Reilly LA, et al. (2016). The Pseudokinase MLKL and the Kinase RIPK3 Have Distinct Roles in Autoimmune Disease Caused by Loss of Death-Receptor-Induced Apoptosis. Immunity 45, 513–526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arts RJ, Novakovic B, Ter Horst R, Carvalho A, Bekkering S, Lachmandas E, Rodrigues F, Silvestre R, Cheng SC, Wang SY, et al. (2016). Glutaminolysis and Fumarate Accumulation Integrate Immunometabolic and Epigenetic Programs in Trained Immunity. Cell Metab 24, 807–819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bambouskova M, Gorvel L, Lampropoulou V, Sergushichev A, Loginicheva E, Johnson K, Korenfeld D, Mathyer ME, Kim H, Huang LH, et al. (2018). Electrophilic properties of itaconate and derivatives regulate the IkappaBzeta-ATF3 inflammatory axis. Nature 556, 501–504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Berger SB, Kasparcova V, Hoffman S, Swift B, Dare L, Schaeffer M, Capriotti C, Cook M, Finger J, Hughes-Earle A, et al. (2014). Cutting Edge: RIP1 kinase activity is dispensable for normal development but is a key regulator of inflammation in SHARPIN-deficient mice. J Immunol 192, 5476–5480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Buck MD, Sowell RT, Kaech SM, and Pearce EL (2017). Metabolic Instruction of Immunity. Cell 169, 570–586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cai X, Chen J, Xu H, Liu S, Jiang QX, Halfmann R, and Chen ZJ (2014). Prion-like polymerization underlies signal transduction in antiviral immune defense and inflammasome activation. Cell 156, 1207–1222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen Q, Chen Y, Bian C, Fujiki R, and Yu X (2013). TET2 promotes histone O-GlcNAcylation during gene transcription. Nature 493, 561–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Davis BK, Wen H, and Ting JP (2011). The inflammasome NLRs in immunity, inflammation, and associated diseases. Annu Rev Immunol 29, 707–735. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Degterev A, Hitomi J, Germscheid M, Ch’en IL, Korkina O, Teng X, Abbott D, Cuny GD, Yuan C, Wagner G, et al. (2008). Identification of RIP1 kinase as a specific cellular target of necrostatins. Nat Chem Biol 4, 313–321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Delaglio F, Grzesiek S, Vuister GW, Zhu G, Pfeifer J, and Bax A (1995). NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J Biomol NMR 6, 277–293. [DOI] [PubMed] [Google Scholar]
  11. Everts B, Amiel E, Huang SC, Smith AM, Chang CH, Lam WY, Redmann V, Freitas TC, Blagih J, van der Windt GJ, et al. (2014). TLR-driven early glycolytic reprogramming via the kinases TBK1-IKKvarepsilon supports the anabolic demands of dendritic cell activation. Nat Immunol 15, 323–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Galluzzi L, Kepp O, Chan FK, and Kroemer G (2017). Necroptosis: Mechanisms and Relevance to Disease. Annu Rev Pathol 12, 103–130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gunda V, Yu F, and Singh PK (2016). Validation of Metabolic Alterations in Microscale Cell Culture Lysates Using Hydrophilic Interaction Liquid Chromatography (HILIC)-Tandem Mass Spectrometry-Based Metabolomics. PLoS One 11, e0154416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hart GW, Slawson C, Ramirez-Correa G, and Lagerlof O (2011). Cross talk between O-GlcNAcylation and phosphorylation: roles in signaling, transcription, and chronic disease. Annu Rev Biochem 80, 825–858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hotchkiss RS, Monneret G, and Payen D (2013). Sepsis-induced immunosuppression: from cellular dysfunctions to immunotherapy. Nat Rev Immunol 13, 862–874. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Johnson BA (2018). From Raw Data to Protein Backbone Chemical Shifts Using NMRFx Processing and NMRViewJ Analysis. Methods Mol Biol 1688, 257–310. [DOI] [PubMed] [Google Scholar]
  17. Kaiser WJ, Sridharan H, Huang C, Mandal P, Upton JW, Gough PJ, Sehon CA, Marquis RW, Bertin J, and Mocarski ES (2013). Toll-like receptor 3-mediated necrosis via TRIF, RIP3, and MLKL. J Biol Chem 288, 31268–31279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lazarus MB, Nam Y, Jiang J, Sliz P, and Walker S (2011). Structure of human O-GlcNAc transferase and its complex with a peptide substrate. Nature 469, 564–567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lei Y, Wen H, Yu Y, Taxman DJ, Zhang L, Widman DG, Swanson KV, Wen KW, Damania B, Moore CB, et al. (2012). The Mitochondrial Proteins NLRX1 and TUFM Form a Complex that Regulates Type I Interferon and Autophagy. Immunity 36, 933–946. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Levine ZG, and Walker S (2016). The Biochemistry of O-GlcNAc Transferase: Which Functions Make It Essential in Mammalian Cells? Annu Rev Biochem 85, 631–657. [DOI] [PubMed] [Google Scholar]
  21. Li J, McQuade T, Siemer AB, Napetschnig J, Moriwaki K, Hsiao YS, Damko E, Moquin D, Walz T, McDermott A, et al. (2012). The RIP1/RIP3 necrosome forms a functional amyloid signaling complex required for programmed necrosis. Cell 150, 339–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Li T, Li X, Attri KS, Liu C, Li L, Herring LE, Asara JM, Lei YL, Singh PK, Gao C, and Wen H (2018). O-GlcNAc Transferase Links Glucose Metabolism to MAVS-Mediated Antiviral Innate Immunity. Cell host & microbe 24, 791–803 e796. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Li X, Zhang Z, Li L, Gong W, Lazenby AJ, Swanson BJ, Herring LE, Asara JM, Singer JD, and Wen H (2017). Myeloid-derived cullin 3 promotes STAT3 phosphorylation by inhibiting OGT expression and protects against intestinal inflammation. J Exp Med 214, 1093–1109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lin J, Kumari S, Kim C, Van TM, Wachsmuth L, Polykratis A, and Pasparakis M (2016). RIPK1 counteracts ZBP1-mediated necroptosis to inhibit inflammation. Nature 540, 124–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu PS, Wang H, Li X, Chao T, Teav T, Christen S, Di Conza G, Cheng WC, Chou CH, Vavakova M, et al. (2017). alpha-ketoglutarate orchestrates macrophage activation through metabolic and epigenetic reprogramming. Nat Immunol 18, 985–994. [DOI] [PubMed] [Google Scholar]
  26. Lu A, Magupalli VG, Ruan J, Yin Q, Atianand MK, Vos MR, Schroder GF, Fitzgerald KA, Wu H, and Egelman EH (2014). Unified polymerization mechanism for the assembly of ASC-dependent inflammasomes. Cell 156, 1193–1206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Mandal P, Berger SB, Pillay S, Moriwaki K, Huang C, Guo H, Lich JD, Finger J, Kasparcova V, Votta B, et al. (2014). RIP3 induces apoptosis independent of pronecrotic kinase activity. Mol Cell 56, 481–495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. McQuade T, Cho Y, and Chan FK (2013). Positive and negative phosphorylation regulates RIP1- and RIP3-induced programmed necrosis. The Biochemical journal 456, 409–415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mills EL, Ryan DG, Prag HA, Dikovskaya D, Menon D, Zaslona Z, Jedrychowski MP, Costa ASH, Higgins M, Hams E, et al. (2018). Itaconate is an anti-inflammatory metabolite that activates Nrf2 via alkylation of KEAP1. Nature 556, 113–117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mompean M, Li W, Li J, Laage S, Siemer AB, Bozkurt G, Wu H, and McDermott AE (2018). The Structure of the Necrosome RIPK1-RIPK3 Core, a Human Hetero-Amyloid Signaling Complex. Cell 173, 1244–1253 e1210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Moriwaki K, Balaji S, Bertin J, Gough PJ, and Chan FK (2017). Distinct Kinase-Independent Role of RIPK3 in CD11c(+) Mononuclear Phagocytes in Cytokine-Induced Tissue Repair. Cell Rep 18, 2441–2451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Murphy JM, Czabotar PE, Hildebrand JM, Lucet IS, Zhang JG, Alvarez-Diaz S, Lewis R, Lalaoui N, Metcalf D, Webb AI, et al. (2013). The pseudokinase MLKL mediates necroptosis via a molecular switch mechanism. Immunity 39, 443–453. [DOI] [PubMed] [Google Scholar]
  33. Najjar M, Saleh D, Zelic M, Nogusa S, Shah S, Tai A, Finger JN, Polykratis A, Gough PJ, Bertin J, et al. (2016). RIPK1 and RIPK3 Kinases Promote Cell-Death-Independent Inflammation by Toll-like Receptor 4. Immunity 45, 46–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Newton K, Sun X, and Dixit VM (2004). Kinase RIP3 is dispensable for normal NF-kappa Bs, signaling by the B-cell and T-cell receptors, tumor necrosis factor receptor 1, and Toll-like receptors 2 and 4. Mol Cell Biol 24, 1464–1469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Newton K, Wickliffe KE, Maltzman A, Dugger DL, Strasser A, Pham VC, Lill JR, Roose-Girma M, Warming S, Solon M, et al. (2016). RIPK1 inhibits ZBP1-driven necroptosis during development. Nature 540, 129–133. [DOI] [PubMed] [Google Scholar]
  36. O’Neill LA, Kishton RJ, and Rathmell J (2016). A guide to immunometabolism for immunologists. Nat Rev Immunol 16, 553–565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Orozco S, Yatim N, Werner MR, Tran H, Gunja SY, Tait SW, Albert ML, Green DR, and Oberst A (2014). RIPK1 both positively and negatively regulates RIPK3 oligomerization and necroptosis. Cell Death Differ 21, 1511–1521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ozcan S, Andrali SS, and Cantrell JE (2010). Modulation of transcription factor function by O-GlcNAc modification. Biochim Biophys Acta 1799, 353–364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Pathak S, Borodkin VS, Albarbarawi O, Campbell DG, Ibrahim A, and van Aalten DM (2012). O-GlcNAcylation of TAB1 modulates TAK1-mediated cytokine release. EMBO J 31, 1394–1404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Shafi R, Iyer SP, Ellies LG, O’Donnell N, Marek KW, Chui D, Hart GW, and Marth JD (2000). The O-GlcNAc transferase gene resides on the X chromosome and is essential for embryonic stem cell viability and mouse ontogeny. Proc Natl Acad Sci U S A 97, 5735–5739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Siddiquee K, Zhang S, Guida WC, Blaskovich MA, Greedy B, Lawrence HR, Yip ML, Jove R, McLaughlin MM, Lawrence NJ, et al. (2007). Selective chemical probe inhibitor of Stat3, identified through structure-based virtual screening, induces antitumor activity. Proc Natl Acad Sci U S A 104, 7391–7396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Silke J, Rickard JA, and Gerlic M (2015). The diverse role of RIP kinases in necroptosis and inflammation. Nat Immunol 16, 689–697. [DOI] [PubMed] [Google Scholar]
  43. Sun L, Wang H, Wang Z, He S, Chen S, Liao D, Wang L, Yan J, Liu W, Lei X, and Wang X (2012). Mixed lineage kinase domain-like protein mediates necrosis signaling downstream of RIP3 kinase. Cell 148, 213–227. [DOI] [PubMed] [Google Scholar]
  44. Tannahill GM, Curtis AM, Adamik J, Palsson-McDermott EM, McGettrick AF, Goel G, Frezza C, Bernard NJ, Kelly B, Foley NH, et al. (2013). Succinate is an inflammatory signal that induces IL-1beta through HIF-1alpha. Nature 496, 238–242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wallach D, Kang TB, Dillon CP, and Green DR (2016). Programmed necrosis in inflammation: Toward identification of the effector molecules. Science 352, aaf2154. [DOI] [PubMed] [Google Scholar]
  46. Wang ZV, Deng Y, Gao N, Pedrozo Z, Li DL, Morales CR, Criollo A, Luo X, Tan W, Jiang N, et al. (2014). Spliced X-box binding protein 1 couples the unfolded protein response to hexosamine biosynthetic pathway. Cell 156, 1179–1192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Weinlich R, Oberst A, Beere HM, and Green DR (2017). Necroptosis in development, inflammation and disease. Nat Rev Mol Cell Biol 18, 127–136. [DOI] [PubMed] [Google Scholar]
  48. Wen H, Lei Y, Eun SY, and Ting JP (2010). Plexin-A4-semaphorin 3A signaling is required for Toll-like receptor- and sepsis-induced cytokine storm. J Exp Med 207, 2943–2957. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Yang J, Yan R, Roy A, Xu D, Poisson J, and Zhang Y (2015a). The I-TASSER Suite: protein structure and function prediction. Nat Methods 12, 7–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Yang X, and Qian K (2017). Protein O-GlcNAcylation: emerging mechanisms and functions. Nat Rev Mol Cell Biol 18, 452–465. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Yang Y, Yin X, Yang H, and Xu Y (2015b). Histone demethylase LSD2 acts as an E3 ubiquitin ligase and inhibits cancer cell growth through promoting proteasomal degradation of OGT. Mol Cell 58, 47–59. [DOI] [PubMed] [Google Scholar]
  52. Yang YR, Kim DH, Seo YK, Park D, Jang HJ, Choi SY, Lee YH, Lee GH, Nakajima K, Taniguchi N, et al. (2015c). Elevated O-GlcNAcylation promotes colonic inflammation and tumorigenesis by modulating NF-kappaB signaling. Oncotarget 6, 12529–12542. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1519255-supplement-1.pdf (3.6MB, pdf)
2

Table S1. Related to Figure 1. List of metabolites in WT macrophages treated with or without LPS

NIHMS1519255-supplement-2.xlsx (45.3KB, xlsx)

RESOURCES