Modulation of M2 macrophage polarization by the crosstalk between Stat6 and Trim24

Abstract

Stat6 is known to drive macrophage M2 polarization. However, how macrophage polarization is fine-tuned by Stat6 is poorly understood. Here, we find that Lys383 of Stat6 is acetylated by the acetyltransferase CREB-binding protein (CBP) during macrophage activation to suppress macrophage M2 polarization. Mechanistically, Trim24, a CBP-associated E3 ligase, promotes Stat6 acetylation by catalyzing CBP ubiquitination at Lys119 to facilitate the recruitment of CBP to Stat6. Loss of Trim24 inhibits Stat6 acetylation and thus promotes M2 polarization in both mouse and human macrophages, potentially compromising antitumor immune responses. By contrast, Stat6 mediates the suppression of TRIM24 expression in M2 macrophages to contribute to the induction of an immunosuppressive tumor niche. Taken together, our findings establish Stat6 acetylation as an essential negative regulatory mechanism that curtails macrophage M2 polarization.

Fig. 1

Stat6 is acetylated at Lys383. a–c Immunoblot analysis of the lysine acetylation of Stat6 in primary murine macrophages (a) or 293T cells transfected with the indicated expression vectors (b), or in control and CBP-knockdown iBMDMs (c) that were (+) or were not (−) stimulated with IL-4 for 30 min; lysates were assessed by immunoprecipitation (IP) with anti-Stat6 or anti-Flag and immunoblotting with anti-Ac-Lys and anti-Stat6, or anti-Flag, and immunoblot of the lysates to detect input proteins and loading controls. d Luciferase assay showing Stat6 transcriptional activity in 293T cells transfected with the indicated vector that were pretreated with DMSO (DM) or trichostatin A (TSA, 5 μM) plus nicotinamide (NAM, 1 mM) (T/N) and then stimulated with (+) or without (−) IL-4 for 4 h before detection. Lysates were immunoblotted for acetylated (Ac−), phosphorylated (P−) Stat6, HA, Flag, and Hsp60 as controls. e, f Immunoblot of Stat6 and Lamin B in nuclear extracts and phosphorylated (P)-Stat6 and Hsp60 in whole-cell lysates (e) and ChIP–QPCR analysis of Stat6 binding to the promoters of Arg1 or Ym1 (f) in murine primary macrophages that were pretreated with DMSO or TSA plus NAM (T/N) and then stimulated with (+) or without (−) IL-4 for 1 h. g Mass spectrometry analysis showing potential acetylation sites in Stat6 after the immunoprecipitation of Stat6 in 293T cells transfected with Stat6 and CBP. h Schematic representation of the mouse Stat6 protein showing the DNA-binding domain (DBD) and its amino acid sequence with all lysine (K) residues highlighted in red. i Immunoblot analysis of the lysine acetylation of wild-type (WT) and KR mutant Stat6 in 293T cells transfected with the indicated expression vectors; lysates were assessed by immunoprecipitation (IP) with anti-Flag and immunoblotting with anti-Ac-Lys and anti-Flag. j Amino acid sequence alignment of Stat6 among the indicated species and different mouse STAT proteins showing Lys383 that are highlighted in red. Data with error bars are represented as mean ± SD. Each panel is a representative experiment of at least three independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 as determined by the unpaired Student’s t test. Source data are provided as a Source Data file

Fig. 2

Lys383 acetylation in Stat6 suppressed its transcriptional activity. a Dot-blot analysis to detect the efficiency of the antibody against Stat6 site-specific (Lys383) acetylation. b, c Immunoblot analysis of the Lys383 acetylation of Stat6 in 293T cells transfected with the indicated expression vectors; lysates were assessed by immunoprecipitation (IP) with anti-Flag and immunoblotting with anti-Ac-Stat6 (K383) and anti-Flag; lysates were immunoblotted to detect input proteins and loading controls before immunoprecipitation. d, e Flow cytometric analysis of the endogenous Stat6 acetylation at Lys383 in murine primary peritoneal macrophages isolated from naive C57BL/6 mice treated with DMSO or trichostatin A (TSA) (5 μM) plus nicotinamide (NAM) (1 mM) (T/N) for 2 h. Data are presented as a representative histography (d) and summary bar graph (e). f–h Luciferase assay to assess Stat6 transcriptional activity in 293T cells transfected with wild-type (WT) or KR mutant Stat6 (f), 293T cells transfected with WT or the K383R mutant Stat6 pretreated with DMSO (−) or with NAM plus TSA (g), and 293T cells transfected with WT or K383A or the K383Q mutant Stat6 (h) and then stimulated with (+) or without (−) IL-4 (20 ng mL⁻¹) for 4 h before detection. Immunoblots showing the protein expression levels in transfected cells are presented below each bar graph. i, j QPCR analysis of Arg1, Mrc1, and Ccl24 mRNA in Stat6-knockdown iBMDMs that were reconstituted with empty vector (EV), WT, or the K383R mutant Stat6, and then were left unstimulated (−) or stimulated (+) with IL-4 for 6 h (i). Immunoblot of Flag-Stat6, phosphorylated (P)-Stat6, and Hsp60 showing the reconstitution efficiency of Flag-Stat6 (j). Data with error bars are represented as mean ± SD. Each panel is a representative experiment of at least three independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 as determined by the unpaired Student’s t test. Source data are provided as a Source Data file

Fig. 3

Trim24 promoted CBP-induced Stat6 acetylation. a Venn diagram showing the number of genes up/downregulated in M2-polarized macrophages and those that encode CBP-associated proteins. b Confocal microscopy images showing the intracellular localization of Trim24 and CBP in peritoneal macrophages that were left nontreated (NT) or stimulated with IL-4 for 30 min, with DAPI staining used to indicate the nuclei of the cells. Scale bars: 5 μm. c, d Immunoblot analysis of the lysine acetylation of Stat6 in 293T cells transfected with the indicated expression vectors assessed by immunoprecipitation (IP) with anti-Flag and immunoblotting with anti-Ac-Lys and anti-Flag. e, f Interaction between CBP, Stat6, and Trim24 in 293T cells transfected with the indicated expression vectors (e) and control and Trim24-knockdown iBMDMs that were left unstimulated (−) or stimulated with IL-4 (+) for 30 min (f) assessed by immunoprecipitation (IP) with anti-Flag or anti-CBP and immunoblotting with anti-HA and anti-Flag, or anti-CBP, anti-Stat6, and anti-Trim24. g Immunoblot analysis of endogenous lysine acetylation of Stat6 in control and Trim24-knockdown iBMDMs stimulated with IL-4 for 30 min assessed by immunoprecipitation (IP) with anti-Stat6 and immunoblotting with anti-Ac-Stat6 (K383) and anti-Stat6. h, i Luciferase assay to assess Stat6 transcriptional activity in 293T cells transfected with the indicated expression vectors (h) and 293T cells transfected with scramble (siCtrl) or siTRIM24 (i) stimulated with (+) or without (−) IL-4 (50 ng mL⁻¹) for 4 h before detection. Immunoblot of acetylated (Ac−) Stat6 after immunoprecipitation; phosphorylated (P−) Stat6, HA-CBP, HA-Trim24, or Hsp60 in whole-cell lysates were applied as controls. j QPCR and immunoblot analysis of M2 gene expression in control and Trim24-knockdown iBMDMs that were left nontreated (NT) or stimulated with IL-4, and the immunoblot of Trim24 and Hsp60 to detect the knockdown efficiency. Data with error bars are represented as mean ± SD. Each panel is a representative experiment of at least three independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 as determined by the unpaired Student’s t test. Source data are provided as a Source Data file

Fig. 4

Trim24-mediated Lys63-linked ubiquitination of CBP at Lys119. a Structure schematic of full-length (FL) Trim24 and its truncations. b Immunoblot analysis of the lysine acetylation of Stat6 in 293T cells transfected with the indicated expression vectors assessed by immunoprecipitation (IP) with anti-Flag and immunoblotting with anti-Ac-Lys and anti-Flag; lysates were immunoblotted to detect input proteins and loading controls. c Endogenous ubiquitination of CBP in control and Trim24-knockdown iBMDMs that were unstimulated (−) or stimulated (+) with IL-4 for 30 min assessed by immunoblot analysis with anti-ubiquitin after immunoprecipitation with anti-CBP (top) and immunoblot analysis with input proteins and loading controls (below). d Ubiquitination of CBP in 293T cells transfected with the indicated expression vectors assessed by immunoblot analysis with anti-HA after immunoprecipitation with anti-Flag (top) or immunoblot analysis with input proteins in lysates without immunoprecipitation (below). e In vitro ubiquitination assay to determine CBP ubiquitination after a mixture reaction of ubiquitin-charged E2 (UbcH5a) and in vitro-translated Flag-CBP with or without HA-Trim24 proteins was assessed by immunoblot analysis with anti-CBP and anti-HA. f Mass spectrometry analysis of the CBP ubiquitination site after the immunoprecipitation of overexpressed Flag-CBP in 293T cells transfected with Flag-CBP and HA-Trim24 showing that Lys119 is the site of CBP ubiquitination by Trim24. g Ubiquitination of CBP in 293T cells transfected with the indicated expression vectors assessed by immunoblot analysis with anti-HA after immunoprecipitation with anti-Flag (top) or immunoblot analysis with input proteins in lysates without immunoprecipitation (below). h In vitro ubiquitination assay to determine CBP ubiquitination after a mixture reaction of ubiquitin-charged E2 (UbcH5a) and in vitro-translated wild-type (WT) or the K119R mutant Flag-CBP with or without HA-Trim24 proteins was assessed by immunoblot analysis with anti-CBP and anti-HA. i, j Immunoblot analysis of the lysine acetylation of Stat6 and the interaction of wild-type (WT) or the K119R mutant CBP with Stat6 in 293T cells transfected with the indicated expression vectors, assessed by immunoprecipitation with anti-Flag and immunoblot with anti-Ac-Lys, anti-Ac-Stat6 (K383), anti-HA, and anti-Flag. Each panel is a representative experiment of at least three independent biological replicates. Source data are provided as a Source Data file

Fig. 5

Trim24 deficiency enhanced DNA-binding activity of Stat6 in M2 gene promoters. a Endogenous ubiquitination of CBP in WT and Trim24-deficient macrophages that were unstimulated (−) or stimulated (+) with IL-4 for 30 min assessed by immunoblot analysis with anti-ubiquitin after immunoprecipitation with anti-CBP (top) and immunoblot analysis with input proteins and loading controls (below). b Immunoblot analysis of endogenous lysine acetylation of Stat6 in WT and Trim24-deficient peritoneal macrophages (PMs) that were unstimulated (−) or stimulated (+) with IL-4 for 30 min assessed by immunoprecipitation (IP) with anti-Stat6 and immunoblotting with anti-Ac-Lys, anti-Ac-Stat6 (K383), and anti-Stat6. c Electrophoretic mobility shift assay (EMSA) of nuclear extracts from WT and Trim24-deficient macrophages that were untreated (−) or stimulated (+) with IL-4 for 1 h assessed with ³²P-radiolabeled WT or mutant probes for Stat6. Immunoblotting of phosphorylated (P−) Stat6 and Lamin B in the nuclear extract were applied for control. d Average Stat6 ChIP signals in the promoter regions of M2 cluster genes in WT and Trim24-deficient macrophages stimulated with IL-4 for 1 h. e Snapshot of the Stat6 ChIP-Seq signals at the Irf4 gene loci in WT and Trim24-deficient macrophages stimulated with IL-4 for 1 h. f, g ChIP–QPCR analysis of Stat6 binding to the promoters of Arg1 and Irf4 in WT and Trim24-deficient macrophages that were left nontreated or stimulated with IL-4 for the indicated time points (f) or Stat6 binding to the promoters of Arg1 and Ym1 in Stat6-knockdown iBMDMs that were reconstituted with empty vector (EV), WT, or the K383R mutant Stat6 and then were left nontreated (NT) or stimulated with IL-4 for 1 h (g). Data with error bars are represented as mean ± SD. Each panel is a representative experiment of at least three independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 as determined by the unpaired Student’s t test. Source data are provided as a Source Data file

Fig. 6

Trim24 negatively regulated macrophage M2 polarization. a RNA-sequencing analysis showing differentiated gene expression profiles in WT and Trim24-deficient macrophages stimulated with IL-4 for 6 h. Representative M2 genes that were upregulated in Trim24-deficient macrophages are indicated in the scatter plot. b Gene set enrichment analysis (GSEA) of the M2 gene set in WT and Trim24-deficient macrophages stimulated with IL-4 for 6 h by using the RNA-sequencing data from (a). c QPCR and immunoblot analysis of M2 gene expression in WT and Trim24-deficient macrophages that were left nontreated (NT) or stimulated with IL-4. d, e Flow cytometric analysis of endogenous Stat6 acetylation at Lys383 in control and TRIM24-knockdown human peripheral blood mononuclear cell (PBMC)-derived primary macrophages that were stimulated with IL-4 for 30 min. Data are presented as a representative histogram (d) and summary bar graph (e). f QPCR analysis of TRIM24, TGM2, CCL17, and IRF4 mRNA in control and TRIM24-knockdown human PBMC-derived primary macrophages that were left nontreated (−) or stimulated (+) with IL-4 for 6 h. Data with error bars are represented as mean ± SD. Each panel is a representative experiment of at least three independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 as determined by the unpaired Student’s t test. Source data are provided as a Source Data file

Fig. 7

Trim24 deficiency in macrophages impaired antitumor immunity. a–c Tumor growth (a), representative tumor pictures (b), and tumor weight (c) of WT and Trim24^M−/− mice (n = 8) that were injected s.c. with B16 melanoma cells. d Confocal microscopy images showing Lys383-acetylated Stat6 in the nucleus of tumor-associated macrophages (TAMs) that isolated from B16 melanoma of WT and Trim24-deficient mice; DAPI staining indicated the nucleus of the cells. Scale bars: 5 μm. e, f Flow cytometric analysis of endogenous Stat6 acetylation at Lys383 in WT and Trim24-deficient TAM as described in (d). Data are presented as a representative histogram (e) and summary bar graph (f). g Tumor weight of WT (n = 6) and Trim24^M−/− mice (n = 7) that were injected s.c. with MC-38 colon adenocarcinoma cells. h–j RNA-sequencing analysis of gene profiles (h), M2 gene enrichment (i), and QPCR verification of M2 genes (j) in WT and Trim24^M−/− TAM that were isolated from B16 melanoma. The M2 signature genes were indicated in the plot. k, l Flow cytometric analysis of CD45⁺CD11b⁺F4/80⁺ TAM (k) and IFNγ-producing CD4⁺ and CD8⁺ T cells (l) in B16 melanoma of WT and Trim24^M−/− mice. Data are presented as a representative plot (left panel) and summary graph (right panel). m, n Tumor growth (m) and weight (n) in WT or Trim24^M−/− mice (n = 4) that were injected s.c. with B16 melanoma cells, and with intravenous injection of the anti-CSF1R antibody to deplete macrophages in vivo. Data with error bars are represented as mean ± SD. Each panel is a representative experiment of at least three independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 as determined by the unpaired Student’s t test. Source data are provided as a Source Data file

Fig. 8

Stat6-mediated direct transcriptional suppression of Trim24. a QPCR analysis of TRIM24, IRF4, or CCL17 mRNA expression in sorted CD11b⁺ macrophages that isolated from six pairs of breast tumor and adjacent normal tissues from six breast cancer patients. b–e QPCR and immunoblot analysis of Trim24 gene expression in murine primary macrophages or in human PBMC-derived macrophages or in human THP-1 cells that were left nontreated (−) or stimulated (+) with a conditioned medium of B16 melanoma cells or IL-4 for 6 h. f, g Structure schema of the constructed luciferase reporters by using different truncated promoter sequences of TRIM24 gene, which were then used to test the luciferase activity. h Immunoblot of Trim24, phosphorylated (P), and total Stat6 and Hsp60 (loading control) expression in macrophages isolated from wild-type (WT) and Stat6-deficient (Stat6-KO) mice. i Schematic representation showing a Stat6-conserved binding site located in the promoter region of TRIM24 gene between –1167 and –380. j ChIP–QPCR analysis of the binding of STAT6 in the promoter region of TRIM24 gene in the THP-1 cells that were stimulated with (+) or without (−) IL-4 for 1 h. k Luciferase assay of Stat6-induced transcriptional suppression of TRIM24 in 293T cells that were transfected with the indicated expression vector, and then were stimulated with (+) or without (−) IL-4 for 4 h. Data with error bars are represented as mean ± SD. Each panel is a representative experiment of at least three independent biological replicates. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001 as determined by the unpaired Student’s t test. Source data are provided as a Source Data file