Macrophage Nogo-B Drives Liver Fibrosis

Abstract

Background & aims: Liver fibrosis is characterized by sustained injury stress, chronic inflammation, and repeated cell death and repair, all of which promote the progression of end-stage liver diseases (eg, liver cirrhosis and carcinoma). As an endoplasmic reticulum-residential protein, Nogo-B strongly regulates macrophage function, but whether Nogo-B-decorated macrophages affect inflammation and progression during liver fibrosis is unclear. The purpose of our current study was to elucidate the roles of Nogo-B^high macrophages during liver fibrosis development.

Methods: The expression and distribution of Nogo-B were analyzed in clinical specimens and animal models. By utilizing myeloid-specific Nogo-B knockout (Nogo-B^mko) mice, the mechanism and functionality of Nogo-B^high macrophages were investigated in 3 murine liver fibrosis models, which were induced separately by bile duct ligation, methionine- and choline-deficient diets, and carbon tetrachloride administration.

Results: Our study revealed the predominant expression of Nogo-B in fibrotic liver macrophages and its positive correlation with fibrosis stage. Myeloid-specific Nogo-B deficiency effectively alleviated liver inflammation, injury, and fibrosis in 3 liver fibrosis models. Importantly, Nogo-B deficiency inhibited NOD-like receptor protein 3 (NLRP3) inflammasome activation and necroptosis in macrophages both in vivo and in vitro. Notably, receptor-interacting serine-threonine kinase 3 (RIPK3) is vital for Nogo-B-driven NLRP3 inflammasome activation and necroptosis in macrophages. Additionally, adoptive transfer of macrophages revealed that the Nogo-B/RIPK3 axis promoted NLRP3 inflammasome activation and necroptosis and accelerated liver fibrosis. Mechanistically, Nogo-B-mediated recruitment of ubiquitin-specific protease 14 restricted the degree of RIPK3 ubiquitination and increased RIPK3 stabilization.

Conclusions: Nogo-B facilitates liver fibrosis by recruiting the deubiquitination enzyme USP14, which increases the stabilization of RIPK3 and promotes NLRP3 inflammasome activation and necroptosis in macrophages.

Keywords: NLRP3; Necroptosis; Nogo-B; RIPK3; USP14.

Graphical abstract

Figure 1

Nogo-B is hyperexpressed in fibrotic liver macrophages and is positively correlated with the degree of fibrosis. (A) Normal and fibrotic liver sections from patients were subjected to H&E and Sirius Red staining and Masson staining and α-SMA IHC analysis (original magnification 1×; scale bar = 2000 μm). (B) Nogo-B mRNA expression in human liver tissues was quantified via RT-PCR. (C) Nogo-B protein expression in human normal and fibrotic liver tissues was explored via Western blotting with respect to the loading control GAPDH. (D) Nogo-B mRNA expression in different types of human fibrotic liver cells was identified via qRT-PCR. (E) Nogo-B protein expression in different types of human fibrotic liver cells was explored via Western blotting with respect to the loading control GAPDH. (F–G) Dual IF staining of Nogo-B (green) and CD68 (red) in human liver tissues, and the ratio of double-positive cells per high-power field was calculated (original magnification 20×; scale bar = 100 μm). (H–I) Western blotting was used to measure the ER-stress markers XBP1s, ATF6, ATF4, and CHOP in normal and fibrotic human liver, and GAPDH served as the loading control. (J) Measurement of Nogo-B in serum of patients with hepatic fibrosis were measured by ELISA. (K) Nogo-B mRNA expression in murine livers analyzed by qRT-PCR. (L) Nogo-B protein levels in murine liver macrophages determined by Western blotting with respect to the loading control GAPDH. (M–N) Dual-IF staining of Nogo-B (red) and F4/80 (green) in murine livers, with quantification of double-positive cells per high-power field (n = 6/group; original magnification 20×; scale bar = 100 μm). (O) Multiple IF analysis revealed colocalization of CD68 (purple), Nogo-B (green), SOX9 (red), and α-SMA (yellow) in normal and fibrotic liver tissues (original magnification 20×; scale bar = 100 μm). (P) Multiplex IF revealed colocalization of F4/80 (purple), Nogo-B (green), SOX9 (red), and α-SMA (yellow) in livers from BDL mice and in control samples. (original magnification 10×; scale bar = 200 μm). Human samples: control (n = 18), mild fibrosis (n = 20), advanced fibrosis (n = 13); mice samples (n = 6 per group). The data are presented as the means ± SEMs; ∗∗P < .01, ∗∗∗P < .001.

Figure 2

Myeloid-specific Nogo-B deficiency alleviates liver injury and fibrosis. (A) Breeding scheme used to generate mice with myeloid-specific Nogo-B deletion. (B) Identification of myeloid-specific Nogo-B-deficient mice. (C) Serum Nogo-B concentrations in the 3 liver fibrosis models, as determined by ELISA. (D) Measurement of ALT and AST in the serum of the mice. (E) Collagen I and α-SMA mRNA expression in the murine livers was evaluated through qRT-PCR. (F–H) Liver sections from the mice were subjected to H&E and Sirius Red staining and Masson staining and α-SMA IHC analysis, and the proportions of Sirius red-positive and Masson-positive and a-SMA-positive regions were quantified. (I–K) Hepatic protein levels of collagen I, MMP-9, α-SMA, and TIMP-1 were assessed via Western blotting; GAPDH served as the loading control. The data are presented as the means ± SEMs; n = 6 per group; original magnification, 10×; scale bars, 200 μm; ∗∗P < .01.

Figure 3

Myeloid Nogo-B deficiency suppresses inflammation and NLRP3 inflammasome activation in fibrotic livers. (A) CD11b immunofluorescence staining with quantification of positive cells per high-power field. (B) Ly6G IHC and corresponding quantification. (C–D) Quantification of positive cells per high-power field (HPF). (E) mRNA levels of IL-1β, TNF-α, IL-6, and IL-10 were analyzed through qRT-PCR within liver tissues of mouse liver fiber models. (F) Measurement of serum IL-1β, TNF-α, and IL-6 levels within mice measured by ELISA. (G–I) protein expression of ASC, NLRP3, Pro-IL-1β, and IL-1β were carried out via Western blotting in liver tissues of mouse liver fiber models concerning the loading control GAPDH. (J–M) Liver sections from mice were exposed to NLRP3 and ASC IHC analysis, and the proportions of NLRP3- and ASC-positive regions were quantified. The data are expressed as the means ± SEMs; n = 6 per group; original magnification, 20×; scale bars, 100 μm; ∗P < .05, ∗∗P < .01.

Figure 4

Nogo-B deficiency inhibits NLRP3 inflammasome activation in macrophages in vitro. (A) Nogo-B mRNA expression in BMDMS extracted from Nogo-B^fl/fl and Nogo-B^mko mice was identified using qRT-PCR. (B) Nogo-B protein expression within BMDMs extracted from Nogo-B^fl/fl and Nogo-B^mko mice was performed by Western blotting, GAPDH served as the loading control. (C) Western blotting analysis of Nogo-B protein in LPS-stimulated BMDMs from Nogo-B^fl/fl and Nogo-B^mKO mice, with GAPDH as the loading control. (D) mRNA levels of IL-1β, TNF-α, IL-6, and IL-10 were quantified in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs by qRT-PCR. (E) Western blotting was used for exploring the levels of ASC, NLRP3, Pro-IL-1β, and IL-1β in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs, GAPDH was the loading control. (F) IF staining for NLRP3 (red) in LPS-stimulated and unstimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (G) IF staining for ASC (red) in LPS-stimulated and unstimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (H) Western blotting was adopted for exploring the levels of ASC, NLRP3, Pro-IL-1β, and IL-1β in LPS-stimulated Nogo-B^fl/fland Nogo-B^mko KCs, GAPDH served as the loading control. (I) Schematic drawing showed LX-2 cells exposed to treatment with CM of LPS-treated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (J) IF staining for a-SMA (red) in LX-2 cells subject to treatment with CM of LPS-treated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (K) Western blotting detected the expression of α-SMA, Collagen I, and TGF-β1 in LX-2 cells exposed to CM from LPS-treated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (L) Schematic drawing showed that Nogo-B^fl/fl and Nogo-B^mko BMDMs were cocultured with LX-2 cells with or without LPS stimulation. (M) Transwell assay analyzed the migratory ability of LX-2 cells (original magnification 40×; scale bars, 50 μm). Data were shown to be mean ± SEM; ∗P < .05, ∗∗P < .01.

Figure 5

RIPK3 is critical for Nogo-B-facilitated NLRP3 inflammasome activation in macrophages in vivo and in vitro. (A) An IP‒MS workflow identified Nogo-B-interacting proteins. (B) Co-IP demonstrated an interaction between Nogo-B and RIPK3 in LPS-stimulated Nogo-B^fl/fl BMDMs. (C) IF staining revealed Nogo-B (red) and RIPK3 (green) colocalization within LPS-stimulated Nogo-B^fl/fl BMDMs. (D) IF staining revealed the colocalization of RIPK3 (green) and NLRP3 (red) and detected the expression of RIPK3 and NLRP3 in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (E) Dual-IF staining of F4/80 (green) and RIPK3 (red) in liver tissues of mouse liver fiber models (n = 6 per group; original magnification 20×; scale bars, 100 μm). (F) Western blotting analysis measured RIPK3 in Nogo-B-deficient macrophages treated with Lv-RIPK3 or the RIPK3 inhibitor RIPK3-IN-4. (G) IF was used to determine the level of RIPK3 in LPS-stimulated Nogo-B^mko BMDMs transfected with Lv-RIPK3 or RIPK3-IN-4. (H) qRT‒PCR was used to quantify IL-1β, TNF-α, IL-6, and IL-10 mRNA in LPS-stimulated Nogo-B^mko BMDMs after Lv-RIPK3 transduction or RIPK3-IN-4 treatment. (I) IF staining for NLRP3 (red) in LPS-stimulated Nogo-B^mko BMDMs transfected with Lv-RIPK3 or treated with RIPK3-IN-4. (J) IF staining for ASC (red) in LPS-stimulated Nogo-B^mko BMDMs transfected with Lv-RIPK3 or RIPK3-IN-4 treatment. (K) Western blotting was used to detect ASC, NLRP3, pro-IL-1β, and mature IL-1β in LPS-stimulated Nogo-B^mko BMDMs following Lv-RIPK3 or RIPK3-IN-4 treatment. The data are shown as the means ± SEMs; original magnification, 40×; scale bar = 50 μm; ∗∗P < .01.

Figure 6

RIPK3 is core to Nogo-B-induced necroptosis of macrophages both in vivo and in vitro. (A) Double fluorescence of necroptosis in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs (original magnification, 40×; scale bars, 50 μm). (B) Flow cytometry was used to evaluate the viability of LPS-stimulated and unstimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (C) Western blotting was performed to evaluate p-RIPK3, RIPK3, p-MLKL, MLKL, C-caspase-1, and Caspase-1 expression in LPS-stimulated and unstimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (D) Liver sections from the mice were subjected to p-RIPK3 and p-MLKL IHC analysis. (E) The proportions of p-RIPK3-positive and p-MLKL-positive regions were quantified. (F) Western blotting was used to evaluate p-RIPK3, RIPK3, p-MLKL, MLKL, C-caspase-1, and Caspase-1 expression in the liver tissues of the mouse liver fiber models. (G) Western blotting was used to evaluate p-RIPK3, RIPK3, p-MLKL, MLKL, C-caspase-1, and Caspase-1 expression. (H) Flow cytometry was conducted to evaluate the cell viability of LPS-triggered Nogo-B^mko BMDMs after Lv-RIPK3 or RIPK3-IN-4 transfection. The data are presented as the means ± SEMs; n = 6 per group; original magnification, 10×; scale bars, 200 μm; ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001.

Figure 7

Nogo-B prevents RIPK3 ubiquitination-mediated degradation by recruiting USP14. (A) Western blotting was used to evaluate RIPK3 in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs treated with MG-132. (B) Western blotting was used to assess RIPK3 in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs treated with CQ. (C) RIPK3 protein stability was tracked via Western blotting in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs that were or were not transduced with Lv-Nogo-B or a negative control lentivirus (LV-NC) and then exposed to CHX (100 ng/mL) for the indicated times. (D) RIPK3 ubiquitination was examined via Western blotting in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (E) Co-IP was used to investigate the interactions between Nogo-B and USP14, RIPK3, TRIM28, and PRPF19 in LPS-stimulated Nogo-B^fl/fl BMDMs. (F) Western blotting was used to measure TRIM28, USP14, and PRPF19 in LPS-stimulated Nogo-B^fl/fl and Nogo-B^mko BMDMs. (G–H) Co-IP was used to explore the associations among Nogo-B, USP14, and RIPK3 in LPS-stimulated Nogo-B^fl/fl BMDMs. (I) Confocal IF confirmed the colocalization of Nogo-B (red), USP14 (purple), and RIPK3 (green) in LPS-stimulated Nogo-B^fl/fl BMDMs (original magnification 40×; scale bar = 50 μm). (J–M) In HEK-293T cells transiently transfected with Myc-RIPK3, Flag-Nogo-B, and Flag-USP14 or with HA-UB/HA-UB-K48/HA-UB-K63, cell lysates were immunoprecipitated with anti-Myc and analyzed via Western blotting with anti-Flag or anti-HA to verify ternary complex formation and ubiquitin linkage specificity. (N) The simulated 3D triad structures of Nogo-B, USP14, and RIPK3 are suggested to be cartoon models in green, blue, and purple. (O) Simulated predicted binding sites for Nogo-B, USP14 and RIPK3. (P) Western blotting was performed to evaluate the levels of NLRP3 and RIPK3 in BMDMs transfected shUSP14 stimulated by LPS.

Figure 8

The adoptive transfer of BMDMs revealed that the Nogo-B/RIPK3 axis promoted NLRP3 inflammasome activation in macrophages and accelerated liver fibrosis. (A) Schematic illustration of adoptive transfer of BMDMs from mice. (B) ALT and AST levels in the serum of mice. (C) liver sections from mice subjected to Masson staining, Sirius Red staining, and α-SMA IHC analysis (original magnification 10×; scale bars, 200 μm). (D) Collagen I and α-SMA mRNA expression in murine livers was examined through qRT-PCR. (E) mRNA expression of IL-1β, TNF-α, IL-6, and IL-10 was quantified in liver tissues. (F) Protein expression of ASC, NLRP3, Pro-IL-1β, and IL-1β was measured via Western blotting in liver tissues. Data are presented as the means ± SEMs; n = 6 per group; ∗∗P < .01. ∗P < .05.

Figure 9

Differential abundance of NLRP3⁺F4/80⁺ and ASC⁺F4/80⁺ macrophages in liver tissue following adoptive transfer of BMDMs. (A–C) Dual-IF staining of fibrotic liver sections from mice that received BMDM transfer showed co-localization of the macrophage marker F4/80 (green) with ASC or NLRP3 (red) in all 3 fibrosis models: (A) CCl₄, (B) BDL, and (C) MCD diet. Nuclei were counterstained with DAPI (blue). n = 6 per group; original magnification, 20×; scale bars, 100 μm.

Figure 10

Differences in Nogo-B⁺RIPK3⁺ macrophage populations within liver tissue after adoptive transfer of BMDMs. (A) Western blotting was used to evaluate the levels of p-RIPK3, RIPK3, p-MLKL, MLKL, C-caspase-1, and Caspase-1 in liver tissues. (B–D) Multiplex IF microscopy demonstrated triple colocalization of the macrophage markers F4/80 (yellow), Nogo-B (green), and RIPK3 (red) in fibrotic liver sections from mice receiving BMDMs in 3 independent models: (B) CCl₄-induced, (C) BDL-induced, and (D) MCD-induced fibrosis. Nuclei were counterstained with DAPI (blue). n = 6 per group; original magnification, 20×; scale bars, 100 μm.

Figure 11

Nogo-B drives RIPK3 activation and NLRP3 inflammasome assembly in macrophages from patients with hepatic fibrosis. (A–B) Western blotting was used to quantify RIPK3 and NLRP3 protein levels in fibrotic human liver samples vs nonfibrotic control tissue, and GAPDH served as the loading control. (C) Dual-IF staining for NLRP3 (red) and the macrophage marker CD68 (green) in human liver sections. (D) Dual-IF staining for RIPK3 (red) and the macrophage marker CD68 (green) in liver sections from patients with hepatic fibrosis. (E) Multiplex IF demonstrating colocalization of CD68 (red), Nogo-B (yellow), USP14 (purple), and RIPK3 (green) in both normal and fibrotic liver tissues; original magnification of 10×; scale bar = 200 μm. data are presented as the means ± SEMs; n = 6 per group; ∗∗P < .01.