CUL4B negatively regulates Toll-like receptor-triggered proinflammatory responses by repressing Pten transcription

Abstract

Toll-like receptors (TLRs) play critical roles in innate immunity and inflammation. The molecular mechanisms by which TLR signaling is fine-tuned remain to be completely elucidated. Cullin 4B (CUL4B), which assembles the CUL4B-RING E3 ligase complex (CRL4B), has been shown to regulate diverse developmental and physiological processes by catalyzing monoubiquitination for histone modification or polyubiquitination for proteasomal degradation. Here, we identified the role of CUL4B as an intrinsic negative regulator of the TLR-triggered inflammatory response. Deletion of CUL4B in macrophages increased the production of proinflammatory cytokines and decreased anti-inflammatory cytokine IL-10 production in response to pathogens that activate TLR3, TLR4, or TLR2. Myeloid cell-specific Cul4b knockout mice were more susceptible to septic shock when challenged with lipopolysaccharide, polyinosinic-polycytidylic acid or Salmonella typhimurium infection. We further demonstrated that enhanced TLR-induced inflammatory responses in the absence of CUL4B were mediated by increased GSK3β activity. Suppression of GSK3β activity efficiently blocked the TLR-triggered increase in proinflammatory cytokine production and attenuated TLR-triggered death in Cul4b mutant mice. Mechanistically, CUL4B was found to negatively regulate TLR-triggered signaling by epigenetically repressing the transcription of Pten, thus maintaining the anti-inflammatory PI3K-AKT-GSK3β pathway. The upregulation of PTEN caused by CUL4B deletion led to uncontrolled GSK3β activity and excessive inflammatory immune responses. Thus, our findings indicate that CUL4B functions to restrict TLR-triggered inflammatory responses through regulating the AKT-GSK3β pathway.

Introduction

Toll-like receptors (TRLs) are the primary sensors that detect conserved molecular patterns on microorganisms and thus act as important components of innate immunity against invading pathogens.^1,2 To date, 13 TLRs (TLR1-TLR13) in humans and mice have been identified.^3–5 By recognizing conserved pathogen components, TLRs recruit the adaptor proteins myeloid differentiation primary response 88 (MyD88) and/or TIR domain-containing adaptor-inducing interferon–β (TRIF) and activate multiple pathways to induce pathogen elimination by immune mediators.^3,6–8 Although TLR activation is important for host defense, excessive immune and inflammatory responses cause tissue damage and serious diseases, such as septic shock. Thus, the restriction of TLR signaling is essential to avoid uncontrolled inflammatory immune responses and maintain immune homeostasis.

Glycogen synthase kinase 3β (GSK3β) serves as the node of convergent signaling pathways that regulate a variety of cellular processes, such as metabolism, cell proliferation, differentiation, and development.^9–11 Unlike most kinases, GSK3β is constitutively active in resting cells, and its activity can be inhibited through N-terminal serine phosphorylation (ser 9) by the PI3K pathway.^9,12,13 Recently, GSK3β was identified as a key mediator of proinflammatory cytokine production in diverse TLR signaling pathways.^12,14,15 The inhibition of GSK3β potently suppressed the production of proinflammatory cytokines and concurrently augmented the production of anti-inflammatory IL-10 in response to several TLR signaling pathways.^16–18 Although these studies clearly documented the importance of GSK3β in TLR-mediated cytokine production, little is known about how GSK3β activity is regulated during TLR-mediated immune responses.

Cullin 4B (CUL4B) functions as a scaffold protein in the Cullin-4B-Ring E3 ligase complex (CRL4B) and participates in regulating diverse physiologically and developmentally controlled processes by targeting specific substrates for ubiquitin-dependent degradation or modification.^19–21 Mutations in human CUL4B are a common cause of X-linked mental retardation syndrome.^22–27 In addition to being mentally retarded, patients with CUL4B mutations also manifest short stature, the absence of speech, an elevated monocyte count, and other developmental defects.^23–26 The lack of CUL4B in mice markedly compromised extraembryonic development,^28,29 hematopoiesis,³⁰ neurogenesis,^31,32 and adipogenesis.³³ Mechanistically, CRL4B can catalyze either polyubiquitination for proteasomal degradation^33–36 or the monoubiquitination of H2A for epigenetic modification.³⁷ Recently, we showed that the deletion of CUL4B in the hematopoietic system led to the aberrant accumulation of myeloid-derived suppressive cells (MDSCs) mediated by downregulation of the AKT/β-catenin pathway.³⁰ CUL4B was shown to sustain the AKT/β-catenin pathway by repressing the phosphatases PP2A and PHLPP1/2, which dephosphorylate and inactive AKT.³⁰ However, the role of CUL4B in TLR-mediated innate immune responses remains unknown. In this study, we demonstrated that CUL4B functions as a negative regulator of TLR4/3/2-mediated immune and inflammatory responses via epigenetically repressing Pten transcription, thus restraining GSK3β activity. The depletion of CUL4B in macrophages augmented TLR4-mediated, TLR3-mediated, and TLR2-mediated proinflammatory cytokine production and decreased production of the anti-inflammatory cytokine IL-10. Myeloid cell-specific Cul4b knockout (MKO) mice exhibited an increased susceptibility to septic shock relative to wild-type (WT) control mice when challenged with lipopolysaccharide (LPS), polyinosinic-polycytidylic acid [poly(I:C)] or Salmonella typhimurium infection. The inhibition of GSK3β activity efficiently blocked the increase in TLR-triggered proinflammatory cytokine production and attenuated TLR-triggered death caused by Cul4b deletion. These findings not only provide mechanistic insight into the role of CUL4B in regulating innate immune responses but also suggest GSK3β as a target for sepsis therapy.

Results

Deletion of CUL4B in macrophages increases lipopolysaccharide-induced production of proinflammatory cytokines

To investigate the role of CUL4B in innate immune and inflammatory responses, we generated myeloid-specific Cul4b knockout mice (referred to as MKO mice) by crossing Cul4b-floxed mice with LysM-driven transgenic mice.²⁸ We examined the protein expression of CUL4B in peritoneal macrophages (PMs) and bone marrow-derived macrophages (BMDMs) and confirmed that CUL4B was effectively deleted in macrophages (Fig. 1a). Consistent with our previous observations in hematopoietic-specific Cul4b knockout mice, the MKO mice exhibited normal growth and survival and produced macrophages, neutrophils, and dendritic cells in bone marrow and spleens at frequencies comparable to those in wild-type mice (Fig. 1b and Suppl Fig. S1). To determine the role of CUL4B in regulating the LPS-induced inflammatory response, BMDMs were prepared from Cul4b MKO and control mice and exposed to the TLR4 agonist LPS, and the production of proinflammatory cytokines was determined. As shown in Fig. 1c, the mRNA levels of Tnfα, Il1β, and Il6 were significantly increased in CUL4B-deficient BMDMs treated with LPS compared to their WT counterparts. Consistent with their mRNA expression, ELISA showed that deletion of CUL4B significantly increased the secretion of these proinflammatory cytokines (Fig. 1d). These results suggest that CUL4B negatively regulates the TLR4-mediated production of proinflammatory cytokines.

Fig. 1.

Deletion of CUL4B in macrophages increases TLR4-mediated production of proinflammatory cytokines. a CUL4B levels in peritoneal macrophages (PMs) and bone marrow-derived macrophages (BMDMs). PMs and BMDMs were prepared from WT and MKO mice, and the CUL4B level was determined by immunoblot analysis. Band intensities given underneath gel images were determined using ImageJ software and are presented as the fold change. b Numbers of macrophages, neutrophils, monocytes, dendritic cells (DCs), conventional dendritic cells (cDCs), and plasmacytoid dendritic cells (pDCs) in bone marrow (left) and spleens (right) from WT and MKO mice determined by flow cytometry. c Effects of CUL4B deficiency on LPS-induced transcription of Tnfα, Il1β, and Il6 in BMDMs. WT and MKO BMDMs were stimulated with LPS (100 ng/ml) for 2 h before qRT-PCR was performed. d Effects of CUL4B deficiency on LPS-induced production of TNF-α, IL-1β, and IL-6 in BMDMs. WT and MKO BMDMs were stimulated with LPS (100 ng/ml) for 3 h, and then the concentrations of TNF-α, IL-1β, and IL-6 in the supernatant were determined by ELISA. The data in b are presented as the mean ± SEM. The data in c and d are presented as the mean ± SD. Data are representative of more than three independent experiments. NS not significant; *p < 0.05, **p < 0.01

Deletion of CUL4B in myeloid cells aggravates LPS-induced or gram-negative bacterium-induced septic shock

To determine the role of CUL4B in the TLR4-mediated immune response in vivo, age-matched and sex-matched WT and MKO mice were intraperitoneally injected with LPS, and the serum levels of proinflammatory cytokines were measured. We found that the serum levels of TNF-α, IL-6, and IL-1β were higher in Cul4b MKO mice than in control mice (Fig. 2a). Consistent with changes in cytokine production in the sera, Cul4b MKO mice died earlier and had lower survival rates than control mice when challenged with LPS (Fig. 2b). To confirm this phenotype, we also used the gram-negative bacterium S. typhimurium to induce sepsis in WT and MKO mice. Upon S. typhimurium infection, MKO mice also produced more proinflammatory cytokines (Fig. 2c) and exhibited a shorter survival time (Fig. 2d), as well as a higher bacterial load in the blood (Fig. 2e) than their WT counterparts, which is similar to the results of published reports showing that proinflammatory cytokines promote the dissemination of E. coli.^38,39 Furthermore, we observed more severe lung injury in MKO mice, as characterized by more immune cell infiltration after challenge with LPS or S. typhimurium (Fig. 2f and Suppl Fig. S2). These data indicate that the deletion of CUL4B aggravated LPS-induced or bacteria-induced septic shock, further supporting the notion that CUL4B negatively regulates innate inflammatory responses.

Fig. 2.

Deletion of CUL4B in myeloid cells aggravates LPS-induced or gram-negative bacterium-induced septic shock. a Effects of CUL4B deficiency on LPS-induced cytokine production in serum. Sex-matched and age-matched WT and MKO mice (n = 5 for each group) were intraperitoneally injected with LPS (10 μg/g) for 4 h. The serum concentrations of TNF-α, IL-1β, and IL-6 were determined by ELISA. b Effects of CUL4B deficiency on LPS-induced inflammatory death. WT and MKO mice (n = 10 for each group) were intraperitoneally injected with LPS (10 μg/g). Mouse survival was monitored over the following 60 h. c Effects of CUL4B deficiency on S. typhimurium-induced cytokine production in serum. WT and MKO mice (n = 5 for each group) were intravenously administrated S. typhimurium (5 × 10⁴ CFU/ml) for 6 h. Sera were collected, and the concentrations of TNF-α, IL-1β, and IL-6 were determined by ELISA. d Effects of CUL4B deficiency on S. typhimurium-induced septic shock. WT and MKO mice (n = 9 for each group) were infected with S. typhimurium (5 × 10⁴ CFU/ml), and survival was monitored every day for 20 days. e Effects of CUL4B deficiency on bacterial load in the blood. WT and MKO mice (n = 5 per genotype) were administrated S. typhimurium (5 × 10⁴ CFU/ml) for 24 h. The bacterial load was determined. f Effects of CUL4B deficiency on LPS-induced or S. typhimurium-induced lung injury. Hematoxylin and eosin staining of lung sections from WT and MKO mice after challenge with PBS, LPS or S. typhimurium for 6 h. Original magnification is ×200. Scale bars, 200 μm. The data in a, c and e are presented as the mean ± SEM. Data are representative of more than three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001

Increased production of proinflammatory cytokines in Cul4b-deficient macrophages is mediated by enhanced GSK3β activity

GSK3β functions as a negative regulator of the TLR4-mediated inflammatory response.¹⁶ Our recent study showed that CUL4B can regulate GSK3β activity.^30,40 Thus, we next examined the effect of CUL4B ablation on the activity of GSK3β in BMDMs. We observed that the levels of GSK3β phosphorylated at ser9, reflecting GSK3β inactivation, were significantly reduced in LPS-stimulated Cul4b-deficient BMDMs (Fig. 3a), indicating that the increased kinase activity of GSK3β could be responsible for the increased TLR4-mediated cytokine induction caused by CUL4B deletion. Indeed, pretreatment with the GSK3β inhibitor SB216763 efficiently blocked increased TLR4-mediated cytokine production in Cul4b-deficient BMDMs (Fig. 3b). Furthermore, the ability of SB216763 to decrease the TLR4-triggered production of proinflammatory cytokines was also demonstrated in MKO mice in vivo. As shown in Fig. 3c, d, SB216763 efficiently blocked increased proinflammatory cytokine production and increased the survival rate of LPS-challenged MKO mice. Taken together, these results indicate that increased GSK3β activity may mediate the enhanced production of TLR4-triggered proinflammatory cytokines in Cul4b MKO mice, suggesting that GSK3β is one of the downstream effectors of CUL4B in response to LPS stimulation or gram-negative bacterial infection.

Fig. 3.

Increased production of proinflammatory cytokines in Cul4b-deficient macrophages is mediated by enhanced GSK3β activity. a Effects of CUL4B deficiency on LPS-induced phosphorylation of GSK3β (Ser 9) in BMDMs. WT and MKO BMDMs were stimulated with LPS (100 ng/ml) for the indicated times before immunoblotting with the indicated antibodies. b Negative regulation of LPS-induced cytokine production in BMDMs with the GSK3β inhibitor SB216763. WT and MKO BMDMs were pretreated with or without SB216763 (10 μM) for 1 h and stimulated with LPS (100 ng/ml) for 3 h. The concentrations of TNF-α, IL-1β, and IL-6 in the medium were measured by ELISA. c Inhibition of LPS-induced cytokine production in the serum. WT and MKO mice (n = 10 for each group) were pretreated with or without the GSK3β inhibitor SB216763 (25 µg/g) for 2 h and then administered LPS (10 µg/g) for 3 h. Sera were collected to measure the TNF-α, IL-1β, and IL-6 levels by ELISA. d Inhibition of LPS-induced death with the GSK3β inhibitor SB216763. WT and MKO mice (n = 10 for each group) pretreated with or without the GSK3β inhibitor SB216763 (25 µg/g) for 1 h were injected with LPS (10 µg/g, intraperitoneal) and then monitored for survival. The data are presented as the mean ± SD in b and as the mean ± SEM in c. Data are representative of more than three independent experiments. *P < 0.05; **P < 0.01

Lack of CUL4B potentiates TLR-3-mediated and TLR2-mediated immune responses

In addition to regulating TLR4, GSK3β is critically involved in immune responses triggered by other TLRs.^{15–17,41,42} We therefore evaluated the role of CUL4B in TLR3-induced, TLR2-induced, or TLR7/8–induced inflammatory responses. We stimulated WT and Cul4b-deficient BMDMs with poly(I:C) (a ligand of TLR3), R848 (a ligand of TLR7 and TLR8), or PGN (a ligand of TLR2) and then examined the transcription of proinflammatory cytokines. We found that CUL4B deficiency potentiated poly(I:C)-induced or PGN-induced transcription of Tnfα, Il1β, and Il6, but not that induced by R848 (Fig. 4a). Consistently, poly(I:C)-induced or PGN-induced secretion of TNF-α, IL-1β, and IL-6, but not that induced by R848, was significantly increased in Cul4b-deficient BMDMs (Fig. 4b). To further confirm that TLR7/8 signaling is not affected by CUL4B deficiency, we tested another agonist of TLR7/8, CL097, and obtained similar results (Suppl Fig. 3). We then challenged MKO mice with poly(I:C) in vivo. Compared to WT mice, Cul4b-deficient mice produced significantly higher levels of IL-6, TNF-α, and IFN-β in the serum and developed a more severe innate inflammatory response after being challenged with poly(I:C) plus D-galactosamine (Fig. 4c, d). Similar to the TLR4-triggered response, the level of GSK3β phosphorylated at serine 9 was decreased in TLR3-triggered or TLR2-triggered MKO BMDMs (Fig. 4e, f). The administration of SB216763 efficiently blocked the increased proinflammatory cytokine production and rescued the reduced survival caused by CUL4B depletion (Fig. 4g, h). Together, these results suggest that CUL4B also negatively regulates TLR3-mediated and TLR2-mediated inflammatory responses via regulating GSK3β activity.

Fig. 4.

CUL4B deficiency potentiates TLR2-mediated and TLR3-mediated immune responses. a Effects of CUL4B deficiency on TLR-triggered transcription of Tnfα, Il1β, and Il6 in BMDMs. WT and MKO BMDMs were stimulated with PGN (20 μg/ml), poly(I:C) (50 μg/ml), or R848 (20 nM) for 2 h before qRT-PCR was performed. b Effects of CUL4B deficiency on PGN-induced, poly(I:C)-induced and R848-induced production of TNF-α, IL-6, and IL-1β in BMDMs. WT and MKO BMDMs were subjected to the indicated stimulation for 3 h. The concentrations of TNF-α, IL-6, and IL-1β in the supernatant were determined by ELISA. c Effects of CUL4B deficiency on poly(I:C)-induced cytokine production in the serum. WT and MKO mice (n = 5 for each group) were administered poly(I:C) (5 µg/g) for 4 h, and the concentrations of TNF-α, IL-6, and IFNβ in the serum were measured by ELISA. d Effects of CUL4B deficiency on poly(I:C)-induced inflammatory death. Sex-matched and age-matched WT and MKO mice (n = 11 for each group) were i.p. injected with poly(I:C) (5 µg/g) and monitored for survival. e–f Effects of CUL4B deficiency on (e) poly(I:C)-induced and (f) PGN-induced phosphorylation of GSK3β (Ser 9) in BMDMs. WT and MKO BMDMs were stimulated with poly(I:C) (50 μg/ml) or PGN (20 μg/ml) for the indicated times before immunoblotting with the indicated antibodies. g Inhibition of poly(I:C)-induced cytokine production in the serum. WT and MKO mice (n = 5 for each group) were pretreated with or without SB216763 (25 µg/g) for 1 h and then stimulated with poly(I:C) (5 µg/g) for 4 h. Levels of TNF-α, IL-6, and IFN-β in the plasma were determined by ELISA. h Inhibition of poly(I:C)-induced death with the GSK3β inhibitor SB216763. WT and MKO mice (n = 11 per group) were pretreated with the GSK3β inhibitor SB216763 (25 µg/g) for 1 h and then administered poly(I:C) (5 µg/g). Mouse survival was monitored. The data are presented as the mean ± SD in a and b and as the mean ± SEM in c and g. Data are representative of more than three independent experiments. NS not significant; *P < 0.05; **P < 0.01

CUL4B restricts NF-κB but promotes CREB signaling in macrophages

GSK3β was reported to positively regulate LPS-induced NF-κB activity, which controls proinflammatory immune responses, and suppress the activity of CREB, a critical component for anti-inflammatory cytokine IL-10 production.¹⁶ To further confirm that CUL4B regulates the inflammatory response, at least in part, by suppressing GSK3β activity, we next examined the key molecules in these signaling pathways. As expected, we observed the enhanced phosphorylation of IκB and NF-κB p65 in LPS-treated, poly(I:C)-treated, or PGN-treated Cul4b-deficient macrophages (Fig. 5a). Consistently, NF-kB activation was significantly increased in LPS-stimulated, poly(I:C)-stimulated, or PGN-stimulated Cul4b-deficient macrophages, as determined by transcription factor activity assays (Fig. 5b). In contrast, the levels of phosphorylated CREB and CREB activity were reduced in LPS-triggered, poly(I:C)-triggered, or PGN-triggered Cul4b-deficient macrophages (Fig. 5a, c). Correspondently, the level of IL-10 production was reduced in LPS-stimulated, poly(I:C)-stimulated, or PGN-stimulated Cul4b-deficient macrophages (Fig. 5d). These results demonstrate that CUL4B plays a regulatory role in the TLR4/3/2-mediated inflammatory response by suppressing the production of proinflammatory cytokines and promoting anti-inflammatory cytokine production.

Fig. 5.

Lack of CUL4B in macrophages results in upregulated NFκB and downregulated CREB signaling. a Effects of CUL4B deficiency on LPS-induced, poly(I:C)-induced and PGN-induced phosphorylation of IκBα, P65, and CREB in BMDMs. WT and MKO BMDMs were stimulated with LPS (100 ng/ml), poly(I:C) (50 μg/ml), or PGN (20 μg/ml) for the indicated times before immunoblotting was performed with the indicated antibodies. b–c Effect of CUL4B deficiency on the transcriptional activity of NF-κB and CREB by TLR agonists. WT and MKO BMDMs were stimulated with LPS (100 ng/ml), poly(I:C) (50 μg/ml), or PGN (20 μg/ml) for 20 or 40 min. Equal amounts of nuclear extracts were analyzed for NF-κB (b) and CREB (c) activation using an ELISA-based chemiluminescent kit as described in the Materials and methods section. d Effects of CUL4B deficiency on TLR-triggered IL-10 expression in BMDMs. WT and MKO BMDMs were stimulated with LPS (100 ng/ml), poly(I:C) (50 μg/ml) or PGN (20 μg/ml) for 2 h before qRT-PCR or ELISA was performed. The data are presented as the mean ± SD. Data are representative of more than three independent experiments. NS, not significant; *P < 0.05; **P < 0.01

Transcriptional upregulation of Pten contributes to enhanced GSK3β activity in CUL4B-deficient macrophages

As serine 9 of GSK3β is phosphorylated by AKT, we next determined the level of AKT phosphorylated at Thr308 and Ser473, which indicates maximal AKT activity. We found that the levels of phosphorylated AKT were significantly decreased in LPS-stimulated, poly(I:C)-stimulated, or PGN-stimulated Cul4b-deficient macrophages (Fig. 6a–c), indicating that the decreased GSK3β phosphorylation at ser9 is due to decreased AKT kinase activity. This decrease in the level of AKT phosphorylation could be attributed to the decreased activities of kinases upstream of AKT or increased activities of phosphatases that dephosphorylate AKT. While no significant difference in the levels of AKT kinases and phosphatases was detected (Suppl Fig. S4), PTEN, a negative regulator of AKT, was found to be significantly elevated in Cul4b-deficient macrophages (Fig. 6a–c).

Fig. 6.

Transcriptional upregulation of Pten contributes to enhanced GSK3β activity in CUL4B-deficient macrophages. a–c Effects of CUL4B deficiency on AKT activation and PTEN levels in BMDMs. WT and MKO BMDMs were treated with LPS (100 ng/ml) (a), poly(I:C) (50 μg/ml) (b), or PGN (20 μg/ml) (c) for the indicated times, and cells were then harvested and lysed for immunoblot analysis with the indicated antibodies. Band intensities given underneath gel images were measured using ImageJ software and are presented as a fold change. d Effects of CUL4B deficiency on Pten transcription in BMDMs. WT and MKO BMDMs were stimulated with LPS (100 ng/ml), poly(I:C) (50 μg/ml), or PGN (20 μg/ml) for 30 min, and the mRNA levels of Pten were analyzed by qRT-PCR. e ChIP assay of CUL4B binding to the Pten promoter. f The effect of TLR agonist stimulation on CUL4B binding to the Pten promoter. BMDMs were stimulated with LPS, poly(I:C) or PGN for 30 min, following which qChIP assays were performed. g ChIP analysis of binding of the indicated proteins to the Pten promoter. h qChIP analysis of binding of the indicated proteins to the Pten promoter in WT and MKO BMDMs. The data in d, f and h are presented as the mean ± SD. Data are representative of more than three independent experiments. *P < 0.05; **P < 0.01

Our previous studies showed that CUL4B functions as a transcriptional corepressor.^37,43 We thus next examined the mRNA level of Pten. Quantitative real-time PCR revealed that the Pten mRNA level was significantly higher in Cul4b-deficient BMDMs (Fig. 6d). To further elucidate the molecular mechanism by which CUL4B inhibits PTEN expression, we used the chromatin immunoprecipitation (ChIP) assay using primer pairs specific for a region of approximately 70 bp to 1500 bp upstream of the Pten transcription initiation site to confirm whether the Pten gene is bound by the CUL4B complex. The ChIP assay indicated that CUL4B binds the Pten promoter from the −1487 to −1344 region (Fig. 6e). Importantly, we detected decreased binding of CUL4B to the Pten promoter during TLR-triggered responses, which is consistent with increased Pten expression during TLR agonist stimulation (Fig. 6f). Notably, EZH2, H2AK119ub1, and H3K27me3 were also bound to the same region (Fig. 6g). To determine the effect of CUL4B deletion on the levels of CUL4B, EZH2, and H2AK119ub1 at the Pten promoter, quantitative ChIP assays were performed in CUL4B-null and control macrophages with antibodies specific for CUL4B, H2AK119ub1, EZH2, H3K27me3, and H3K4me3. While CUL4B, EZH2, H2AK119ub1, and H3K27me3 were enriched at the promoter of the Pten gene in cells in the control group, CUL4B deletion significantly decreased their occupancy at the promoter. Meanwhile, the levels of H2AK119ub1 and H3K27me3 at the Pten promoter were also significantly reduced, while the level of H3K4me3 on the Pten promoter was increased (Fig. 6h). Taken together, these results indicate that the CUL4B complex represses the transcription of Pten, which inhibits AKT and thus restrains GSK3β activity.

Discussion

Given the critical functions of TLR signaling in innate immunity, the identification and characterization of TLR signaling regulators is important for understanding how immune responses are tightly regulated. In this study, we demonstrate that CUL4B serves as a physiologic inhibitor to prevent excessive activation of TLR signaling and protect the host from being overwhelmed by inflammatory responses by restricting GSK3β activity and the subsequent production of pro-inflammatory and anti-inflammatory cytokines. We have provided several lines of evidence to support this claim. First, the lack of CUL4B in macrophages significantly augmented TLR-induced proinflammatory cytokine production and decreased IL-10 expression induced by TLRs. Second, myeloid cell-specific Cul4b knockout mice exhibited increased susceptibility to septic shock when challenged with LPS, poly(I:C) or S. typhimurium infection. Third, regulation of the TLR-triggered immune response by CUL4B was shown to be mediated by the suppression of proinflammatory GSK3β activity. The inhibition of GSK3β efficiently blocked the increase in TLR-triggered inflammation caused by CUL4B deletion and protected mice from S. typhimurium-induced endotoxin shock. Finally, the CRL4B complex was shown to regulate AKT-GSK3β activity by epigenetically repressing Pten transcription. However, our results differ from those of a report by Hung et al.,⁴⁴ in which the authors showed that the deletion of CUL4B in macrophages enhanced LPS-induced peritonitis but not production of the proinflammatory cytokines IL-6 and TNF-α.

Proper control of GSK3β activity appears to be quite important for a variety of cellular processes, including immune responses.^14,16,45 Accumulating evidence indicates that GSK3β regulates pro-inflammatory and anti-inflammatory cytokine production in TLR signaling.^15–17,41 Martin et al first demonstrated the regulatory roles of GSK3β in diverse TLR-mediated inflammatory cytokine production.¹⁶ TLR agonist stimulation led to the phosphorylation and inactivation of GSK3β through a PI3K/AKT-dependent pathway.^16,18,42,46 Consistent with these observations, we found that the dysregulation of pro-inflammatory and anti-inflammatory cytokine production in pathogen-triggered CUL4B-deficient macrophages was also mediated by enhanced GSK3β activity. Enhanced GSK3β activity in CUL4B-deficient BMDMs promoted the production of proinflammatory cytokines and inhibited production of the anti-inflammatory cytokine IL-10. Accordingly, we showed that the pharmacologic inhibition of GSK3β could efficiently block the enhanced TLR-triggered inflammatory response caused by CUL4B deletion in vitro and in vivo. Several mechanisms are involved in the regulation of GSK3β activity.^41,45,47 We found that decreased GSK3β phosphorylation at Ser9 is associated with decreased AKT kinase activity. Intriguingly, while CUL4B negatively regulates GSK3β activity by repressing the phosphatases PP2A and PHLPP1/2, which dephosphorylate and inactivate AKT to sustain Wnt signaling in myeloid-derived suppressive cells (MDSCs),³⁰ no significant difference in the levels of PP2A and PHLPP was found between Cul4b-deficient and control macrophages. However, PTEN, a negative regulator of AKT, was found to be significantly elevated in CUL4B-deficient macrophages. These findings suggest that AKT signaling is regulated differently in MDSCs and BMDMs. While decreased AKT activity was shown to affect other downstream targets, such as the mTOR signaling pathway and thereby NF-κB,^48,49 we detected no significant changes in these targets in CUL4B-deficient macrophages (data not shown). Importantly, the pharmacological inhibition of GSK3β activity efficiently blocked the increase in TLR-triggered proinflammatory cytokine production and attenuated TLR-triggered death caused by CUL4B deletion, indicating that GSK3β plays a major role in mediating negative regulation of the TLR-triggered inflammatory response by CUL4B.

Previous studies have shown that the CUL4B complex targets several substrates, such as cyclin E, PPARγ, PrxIII, and Jab1, for ubiquitin-dependent proteasomal degradation.^33,34,36,50 However, an examination of TLR and its downstream signaling molecules MYD88, TRAF6, TRAF3, TAK1, and TRIF revealed no difference in their levels between WT and MKO cells (data not shown). In addition to its role in promoting protein degradation, the CUL4B complex functions as a transcriptional corepressor of gene transcription.^37,43,51 We showed here that the CUL4B complex epigenetically represses the expression of Pten, which subsequently inhibits AKT-GSK3β signaling, subsequently increasing NF-κB activity and decreasing CREB activity. The AKT-GSK3β-CREB pathway has been shown to be critical for the regulation of IL-10 production during TLR activation.¹⁶ Consistent with these reports, elevated NF-kB activity and attenuated CREB activation were observed in TLR-induced CUL4B-deficient macrophages. Increased proinflammatory cytokine and decreased anti-inflammatory cytokine production is presumably responsible for the aggravated inflammatory responses observed in Cul4b-deficient mice.

Although CUL4B expression levels were not changed during the TLR response, we observed decreased binding of CUL4B to the Pten promoter during TLR agonist stimulation. The CUL4B complex does not contain a DNA-binding domain and may target specific genomic sites via its partner proteins, such as transcriptional factors. However, the identity of the transcription factor that recruits the CUL4B complex to the Pten gene promoter remains unknown. Future studies are needed to determine how CUL4B is recruited to the Pten promoter and which transcription factor(s) is used to specifically target the Pten promoter during the TLR response.

In summary, we demonstrated that CUL4B can restrict TLR-triggered inflammatory responses by regulating GSK3β activity. Our results reveal a CRL4B-based epigenetic mechanism by which TLR-mediated inflammatory responses are modulated and may have implications for the development of novel anti-inflammatory strategies to manage infections and other diseases.

Materials and methods

Mice

Cul4b floxed mice were developed as previously reported.²⁸ To produce conditional knockout mice in which Cul4b was specifically deleted in myeloid cells, Cul4b floxed mice were crossed with LysM-Cre transgenic mice, producing Cul4b^flox/YLysM-Cre^+/− (MKO) mice and Cul4b^flox/YLysM-Cre^−/− (WT) littermate controls. Mice were housed in a specific pathogen-free animal facility at Shandong University. Animal care and experiments were performed in accordance with protocols approved by the Animal Care and Use Committee of Shandong University School of Basic Medical Sciences.

Generation of bone marrow–derived macrophages and peritoneal macrophages

Bone marrow–derived macrophages (BMDMs) were generated as previously described.⁵² Bone marrow cells (1 × 10⁷) were cultivated in RPMI 1640 medium containing 10% FBS and 10 ng/ml recombinant murine M-CSF (PeproTech) in a 100-mm dish for 5 days to generate BMDMs.

Peritoneal macrophages (PMs) were obtained from thioglycollate-elicited mice and rinsed with ice-cold PBS. Cells were resuspended in the relevant medium (DMEM supplemented with 10% FBS, 2 g/l sodium bicarbonate, 100 mg/l sodium pyruvate, 10 mM HEPES (pH 7.4), 62.1 mg/l penicillin, and 100 mg/l streptomycin), seeded in plates with a density of 10⁶/ml, and subsequently cultured overnight at 37 °C. Nonadherent cells were discarded by rinsing with PBS.

In vivo endotoxic shock model

LPS (10 μg/g; Sigma-Aldrich) was intraperitoneally injected into age-matched and sex-matched WT and MKO mice to induce LPS shock. S. typhimurium strain SL1344 was diluted in PBS and administered to WT (5 × 10⁴ CFU/ml) or MKO (5 × 10⁴ CFU/ml) mice through tail vein injection. Colony-forming units (CFUs) were measured by counting viable bacteria on agar plates in blood samples. For poly(I:C) induction, age-matched and sex-matched WT mice and MKO mice were i.p. injected with poly(I:C) (5 μg/g body weight; ApexBio) and D-galactosamine (0.5 mg/g body weight; Sigma-Aldrich). Mouse survival was monitored, and serum and target organs were collected at the specified times.

Cell culture and cytokine assay

BMDMs were produced with recombinant murine M-CSF (50 ng/ml). After their incubation for 6 days, the cells were stimulated with LPS (100 ng/ml) or poly(I:C) (50 µg/ml). TNF-α, IL-6, IL-1β, IL-10, and IFN-β in supernatants and sera were quantified by sandwich ELISA (BioLegend) following the manufacturer’s instructions.

Real-time PCR

TRIzol reagent was applied to lyse the cells, and total RNA was collected according to the manufacturer’s instructions (Invitrogen). Real-time quantitative RT-PCR (qRT-PCR) was performed on a Roche 480 system. The results are expressed as the relative abundance of the specific mRNA normalized to that of Gapdh.

Western blot assays

Cells were lysed with SDS lysis buffer, lysates were obtained by centrifugation, and protein concentrations were detected with a BCA Protein Assay Kit (Thermo Scientific). Equivalent amounts of total proteins were separated by SDS-PAGE and subsequently transferred onto PVDF membranes. Proteins were detected with antibodies against specific proteins; anti-CUL4B antibody was purchased from Sigma-Aldrich, anti-PHLPP antibody was purchased from Proteintech Group, and antibodies against the rest of the proteins were purchased from CST.

Transcription factor activity assays

To prepare nuclear extracts, BMDMs from a single mouse were washed with precooled phosphatase inhibitor buffer (12.5 mM β-glycerophosphate, 6.25 mM NaF, 12.5 mM PNPP, 1.25 mM Na₃VO₄ in PBS) and centrifuged at 100 × g and 4 °C for 6 min. Nuclear extracts were prepared using Cayman’s nuclear extraction kit according to the manufacturer’s instructions. Activation of NF-κB and CREB was quantified using NF-κB and pCREB/CREB assay kits (Cayman), respectively, according to the manufacturer’s instructions.

Flow cytometric analysis

Single-cell suspensions were freshly prepared from mouse bone marrow and spleens. Cells were stained with anti-CD11b (BioLegend), anti-Gr-l (eBioscience), anti-Ly6C (BioLegend), and anti-F4/80 (BioLegend) antibodies. Data were analyzed with FlowJo 7.6.5 software.

Chromatin immunoprecipitation analysis

ChIP was performed as described previously.³⁷ Briefly, 1 × 10⁷ cells were crosslinked with 1% formaldehyde, sonicated, precleared and incubated with 5–10 mg of antibody per reaction. The complexes were rinsed with low-salt and high-salt buffers, and the relevant DNA was extracted and precipitated. Primer sequences are listed in Supplementary Tables S1.

Statistical analysis

The data are reported as the mean ± SEM from the indicated number of mice or biologic replicates and as the mean ± SD of the indicated number of technical replicates. Statistical analysis was performed with GraphPad Prism 5.0 software. Differences between means were analyzed by unpaired Student’s t-test. The statistical significance of survival differences was estimated with the Kaplan-Meier method, and curves were compared with the generalized Wilcoxon test. P values of less than 0.05 indicated statistical significance.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version of this article (10.1038/s41423-019-0323-0) contains supplementary material.