PROTAC-Mediated HDAC7 Protein Degradation Unveils Its Deacetylase-Independent Proinflammatory Function in Macrophages

Abstract

Class IIa histone deacetylases (Class IIa HDACs) play critical roles in regulating essential cellular metabolism and inflammatory pathways. However, dissecting the specific roles of each class IIa HDAC isoform is hindered by the pan-inhibitory effect of current inhibitors and a lack of tools to probe their functions beyond epigenetic regulation. In this study, a novel PROTAC-based compound B4 is developed, which selectively targets and degrades HDAC7, resulting in the effective attenuation of a specific set of proinflammatory cytokines in both lipopolysaccharide (LPS)-stimulated macrophages and a mouse model. By employing B4 as a molecular probe, evidence is found for a previously explored role of HDAC7 that surpasses its deacetylase function, suggesting broader implications in inflammatory processes. Mechanistic investigations reveal the critical involvement of HDAC7 in the Toll-like receptor 4 (TLR4) signaling pathway by directly interacting with the TNF receptor-associated factor 6 and TGFβ-activated kinase 1 (TRAF6-TAK1) complex, thereby initiating the activation of the downstream mitogen-activated protein kinase/nuclear factor-κB (MAPK/NF-κB) signaling cascade and subsequent gene transcription. This study expands the insight into HDAC7's role within intricate inflammatory networks and highlights its therapeutic potential as a novel target for anti-inflammatory treatments.

Keywords: HDAC7; PROTAC; TLR4 signaling; anti‐inflammatory drug development; proinflammatory cytokines.

Figure 1

HDAC7 knockdown significantly attenuates proinflammatory cytokine production in LPS‐stimulated macrophages. A) Differential expression of the HDAC gene family in association with SLE (left) and SCAP (right). Transcriptome data for SLE (GSE616352, n = 99) and SCAP (GSE1963993, n = 56) were obtained from the Gene Expression Omnibus (GEO) database, applying a significance threshold of p < 0.05 and an absolute log fold change (|log₂FC|) greater than 0.2 for the SLE dataset and 0.5 for the SCAP dataset. B) Comparative analysis of HDAC gene expression across the two datasets indicates that significantly changed genes belong to class IIa HDACs. C) Illustration depicting the secretion of cytokines (TNF‐α, IL‐6) by macrophages following LPS stimulation and subsequent measurement via ELISA. D) The effective knockdown of class IIa HDAC isoforms in RAW264.7 cells was validated by immunoblotting. E,F) Measurement and analysis of IL‐6 (E) and TNF‐α (F) levels in RAW264.7 cells after class IIa HDAC knockdown, using an ELISA assay. G) Chemical structure of the class IIa HDAC inhibitor TMP269. H) Quantification of IL‐6 and TNF‐α secretion levels in RAW264.7 cells following TMP269 treatment, determined using ELISA kits. All experimental results are based on at least three biological replicates, with statistical significances denoted as not significant (n.s.), ^** for p < 0.01, and ^*** for p < 0.001.

Figure 2

Design and optimization of selective HDAC7 degraders. A) Schematic representation of the mechanism underlying PROTAC‐induced protein degradation. B) Computational modeling displaying the binding mode of TMP269 to the crystal structure of HDAC7 (PDB code: 3ZNS) using PyMOL software and the chemical structures of M1, M2, and M3. C) Assessment of in vitro HDAC7 enzymatic activity inhibition when treating with tested compounds at various concentrations and using Ac‐Leu‐Gly‐Lys(TFAc)‐AMC (TFAc: trifluoroacetyl; AMC: 7‐amino‐4‐methylcoumarin) as a substrate. D‐H) Immunoblot analysis of HDAC7 and HDAC9 in 293T cells upon treatment with different PROTAC compounds and TMP269 (1 and 5 µm, 24 h). I) Immunoblotting results illustrating the impact on HDAC7 levels in 293T cells treated with A5 (top) and B4 (bottom) at varying concentrations for 12 h.

Figure 3

A5 and B4 induce isoform‐selective, UPS‐dependent HDAC7 degradation. A) Assessment of HDAC4, HDAC5, and HDAC9 protein levels in 293T cells following treatment with A5 (left) and B4 (right) at concentrations of 1 and 5 µm for 12 h. B) Immunoblot analysis of HDAC7 levels in 293T cells treated with MG132 (10 µm) along with either A5 (5 µm, top) or B4 (5 µm, bottom) for 12 h. C) Analysis of HDAC7 protein levels in CRBN‐WT and CRBN‐KO cells treated with A5 (5 µm) for 12 h. D) Analysis of HDAC7 protein levels in VHL‐WT and VHL‐KO cells treated with B4 (5 µm) for 12 h. E, F) Validation of A5 or B4‐induced HDAC7 ubiquitination. Transient co‐transfection of 293T cells with HDAC7‐Flag and Ub‐His plasmids (1:1 ratio) followed by treatment with A5/B4 (5 µm), MG132 (10 µm), and TMP269 (5 µm) for 12 h. Co‐immunoprecipitation using anti‐NTA beads performed post‐transfection and treatment. G) Immunoblot analysis of HDAC7 in 293T cells after co‐treatment with various concentrations of TMP269 (25, 50, 100, and 150 µM) and A5 (5 µM, top) or B4 (5 µm, bottom) for 12 h.

Figure 4

B4‐mediated HDAC7 degradation reduces inflammatory cytokine levels in LPS‐stimulated macrophages. A) Immunoblot analysis of HDAC7 protein levels in RAW264.7 cells after treatment with varying concentrations of B4 for 12 h (left) and with 5 µm B4 at different time points (right). B) Immunoblot assessment of HDAC7 protein levels in RAW264.7 and BMDM cells following treatment with B4 (5 µm), VH032 (5 µm), B6 (5 µm), and TMP269 (5 µm). C) Overview of the proteomic analysis workflow (left), and the analysis results illustrating changes in HDAC protein abundances in RAW264.7 cells after B4 treatment displayed on a log₂FC scale (right). HDAC7 was undetected in this experiment, likely due to its low expression levels. D) Immunoblot analysis of HDAC4, HDAC7, and HDAC9 in RAW264.7 cells after treatment with B4 (5 µm) and TMP269 (5 µm). E) Immunoblot analysis of HDAC6 and HDAC8 in RAW264.7 cells after treatment with various concentrations of B4 up to 20 µm. F‐I) ELISA analysis of IL‐6 and TNF‐α secretion levels in RAW264.7 and BMDM cells. Cells were treated with tested compounds (5 µm) for 12 h, followed by LPS stimulation at 10 ng mL⁻¹ for 24 h. All data presented are based on a minimum of three replicates, with statistical significance denoted as follows: not significant (n.s.), ** for p < 0.01, and ^***for p < 0.001.

Figure 5

B4 selectively suppresses the transcription of a group of proinflammatory cytokines. A) Workflow illustrating the Antibody‐array method for detecting inflammatory cytokines and chemokines. B) Analysis of inflammatory cytokines and chemokines utilizing the Mouse Inflammation Antibody Array‐Membrane Kit. RAW264.7 cells were treated with B4 (5 µm) and TMP269 (5 µm) for 12 h, followed by exposure to LPS (10 ng mL⁻¹) for an additional 24 h. C) ELISA measurements of GM‐CSF (top) and IL‐1β (bottom) secretion levels. RAW264.7 cells were treated with B4 (5 µm) and TMP269 (5 µm) for 12 h, followed by LPS stimulation at 10 ng mL⁻¹ for another 24 h. D) Venn diagram depicting differentially expressed genes in RAW264.7 cells. Cells were treated with DMSO, B4, and TMP269 for 12 h, followed by LPS stimulation at 10 ng mL⁻¹ for 24 h. Differential gene identification criteria were set at FC ≥ 2 or FC ≤0.5 (equivalent to |log₂(FC)| ≥ 1) with a p‐value < 0.05 for screening differential genes. E) Analysis of inflammation‐related genes in the B4L group compared to DML and TML groups. F) qRT‐PCR analysis of Il6, Il1b, and Csf2 mRNA expression levels. Data were obtained from three independent experiments. Statistical significance is indicated as follows: not significant (n.s.), ^** for p < 0.01, ^*** for p < 0.001, and ^**** for p < 0.0001. G) GSEA algorithm analysis of inflammation response processes in RAW264.7 cells.

Figure 6

The regulatory role of HDAC7 in macrophage cytokine transcription is independent of chromatin opening. A) Graphic representation of the ATAC‐seq process. B) Enrichment of ATAC‐seq signals around TSSs after treatment with various compounds in RAW264.7 cells for 12 h. C) ATAC‐seq signal profiles in representative regions displaying consistent accessibility status across three groups. The chromatin coordinates of each region are indicated at the top left of each plot, with the TSS locations and transcription directions of target genes (Il6, Il1b, Csf2, and Tnf) denoted by black arrows.

Figure 7

HDAC7 mediates proinflammatory cytokines transcription via the TLR4‐TRAF6‐TAK1 signaling pathway. A) Simplified illustration of the LPS‐induced TLR4 signaling pathway. B) Immunoblot analysis depicting protein levels of p‐NF‐κB(S536), NF‐κB, and GAPDH in RAW264.7 cells. Cells were pretreated with B4 (5 µm) or TMP269 (5 µm) for 12 h, followed by LPS stimulation for 2 h. C) Immunoblot analysis illustrating protein levels of p‐c‐JUN, c‐JUN, p‐JNK1/2/3, JNK1/2/3, and GAPDH in RAW264.7 cells. For p‐c‐JUN and c‐JUN, cells were pre‐treated with B4 (5 µm) or TMP269 (5 µm) for 12 h, then stimulated by LPS for 2 h; for p‐JNK1/2/3 and JNK1/2/3, cells were pretreated with B4 (5 µm) or TMP269 (5 µm) for 12 h, followed by LPS stimulation for 0.5 h. D) 293T cells were transfected with HDAC7‐Flag, then treated with B4 (5 µm) or TMP269 (5 µm) for 12 h. Cell lysates prepared using RIPA lysis buffer underwent immunoprecipitation with Flag beads, followed by immunoblotting with anti‐Flag, anti‐TAK1, and anti‐TRAF6 antibodies. Representative immunoblot data were presented.

Figure 8

HDAC7 degrader B4 effectively reduces inflammatory cytokine levels in vivo. A) Outline depicting the experimental procedures conducted in mice. Eight‐week‐old ICR mice in each group (n = 8) received pretreatment with B4 (i.v., 12.5 mg kg⁻¹), TMP269 (i.v., 12.5 mg kg⁻¹), or Hexadecadrol (i.g., 2.0 mg kg⁻¹) for 6 h before LPS administration (i.p., 10 mg kg⁻¹). Serum samples were collected from each group 2 h post‐LPS exposure. Note: the administrated dosages were determined through preliminary studies designed to ascertain non‐toxicity levels (Table S8, Supporting Information). B–E) Measurement of IL‐6, GM‐CSF, TNF‐α, and IL‐1β secretion levels was performed using the ELISA assay.