Targeting BRD4 ameliorates experimental emphysema by disrupting super-enhancer in polarized alveolar macrophage

Abstract

Background

Chronic obstructive pulmonary disease (COPD) is a progressive chronic lung disease characterized by chronic airway inflammation and emphysema. Macrophage polarization plays an important role in COPD pathogenesis by secreting inflammatory mediators. Bromodomain-containing protein 4 (BRD4), an epigenetic reader that specifically binds to histones, plays a crucial role in inflammatory diseases by regulating macrophage polarization. Herein, we attempted to examine the hypothesis that modulating alveolar macrophage polarization via BRD4 inhibitors might has a potential for COPD treatment.

Methods

We firstly analyzed BRD4 expression and its correlation with clinical parameters and macrophage polarization markers in sputum transcriptomes from 94 COPD patients and 36 healthy individuals. In vivo, BRD4 inhibitor JQ1 and degrader ARV-825 were intraperitoneally administrated into emphysema mice to assess their effects on lung emphysema and inflammation. In vitro, RNA-seq and CUT&Tag assay of BRD4 and H3K27ac were applied for elucidating how BRD4 regulates macrophage polarization.

Results

We found an increased expression of BRD4 in the induced sputum from patients with COPD and unveiled a strong correlation between BRD4 expression and clinical parameters as well as macrophage polarization. Subsequently, BRD4 inhibitor JQ1 and degrader ARV-825 significantly mitigated emphysema and airway inflammation along with better protection of lung function in mice. BRD4 inhibition also suppressed both M1 and M2 alveolar macrophage polarization. The CUT&Tag assay of BRD4 and H3K27ac, revealed that BRD4 inhibition disrupted the super-enhancers (SEs) of IRF4 (a crucial transcription factor for M2 macrophage), and subsequently affected the expression of matrix metalloproteinase 12 (MMP12) which is vital for emphysema development.

Conclusion

This study suggested that downregulation of BRD4 might suppress airway inflammation and emphysema through disrupting the SEs of IRF4 and alveolar macrophages polarization, which might be a potential target of therapeutic intervention in COPD.

A diagram of the mechanism by which BRD4 mediated super-enhancer of IRF4 in M2 AMs.

Graphic illustration showed targeting BRD4 in M2 polarized AMs lead to the downregulation of MMP12 expression, resulting in the amelioration of experimental emphysema by disrupting the super-enhancer of IRF4.

Graphical abstract

Introduction

COPD is a progressive and lethal chronic lung disease characterized by chronic airway inflammation and emphysema [1]. An aberrant inflammatory response that involves both innate and adaptive immunity is critical to the development of COPD. A variety of immune cells, including macrophages, neutrophils, and T lymphocytes, are involved in the pathogenesis of COPD [2]. Among all, alveolar macrophages (AMs) are abundantly found in patients with COPD and play essential roles in the inflammation and airway remodeling of this disease [3]. Macrophages are the most plastic immune cells present in all tissues and can be commonly polarized into classically activated (also called M1) macrophages and alternatively activated (also called M2) macrophages [4]. M1 polarized macrophages are stimulated by interferon-γ (IFN-γ) and toll-like receptor (TLR) ligands, which are recognized for their ability to elicit the production of cytokines such as IL‐6, IL-1β, and TNF-α, and play an essential role in Th1-type immune responses. M2 macrophages are generated by the stimulation of T-helper 2 (Th2) cytokine IL-4 or IL-13, and secrete anti‐inflammatory cytokines (e.g., IL-10, TGF-β) and chemokines (e.g., CCL17, CCL22) to participate in Th2-type immune responses [5].

In the context of COPD, AMs are cells of the innate immune system and represent a vital component of the first-line body defense against pathogens and inhaled particles [6]. On one hand, AMs exposed to cigarette smoke or microbes can polarize into M1 subtype and secrete the pro-inflammatory cytokines and chemokines, thereby play a crucial role in initiating the pro-inflammatory response [7, 8]. On the other hand, AMs adapt to M2 states to release active mediators likewise matrix metalloproteinases 12 (MMP12) which have been recognized as a key factor in causing emphysema [9,10,11]. Therefore, both types of macrophages are vital to the pathogenesis of COPD [12]. As such, it was widely believed that regulating the AMs polarization holds significant promise for advancing COPD therapy.

Notably, clinical cohort studies offer a powerful tool for discovering potential therapeutic targets. In China, a cohort of sputum transcriptomes was previously created, consisting of 94 patients diagnosed with COPD and 36 persons who were deemed healthy [13] and found that the transcriptional level of bromodomain containing protein 4 (BRD4) is significantly associated with clinical parameters and macrophage polarization. Hence, this observation suggests that it possesses the potential to serve as a therapeutic target for COPD. Numerous human diseases have been closely linked to aberrant epigenetic regulation [14]. BRD4 is the best-characterized member of the bromodomain and extra-terminal (BET) protein family and is an epigenetic reader of acetylated lysine residues to regulate the transcription of genes and chromatin landscape. It usually acts as a scaffold for transcription factors and localizes at gene promoters and enhancers [15]. In addition, BRD4 plays a crucial role in the organization of super-enhancers (SEs), which are critical in driving the high-level transcription of specific genes and are typically identified based on strong enrichment of H3K27ac [16]. Due to the vital role of BRD4 in various disease progression, including carcinomas, inflammatory diseases, fibrosis, and vascular diseases, several BRD4 inhibitors, such as JQ1, have been developed for disease intervention [17,18,19,20]. In recent years, BRD4 degraders like ARV-825 were also developed based on the proteolytic targeting chimera (PROTAC) technology and were tested for many disease treatments [21, 22]. Although BRD4 plays a critical role in regulating both M1 and M2 macrophage polarization, they have been reported to be effective in the treatment of tumors by modulating the function of tumor-associated macrophage [23,24,25]. However, little is known whether targeting BRD4 affects the AMs polarization and the therapeutic consequence for COPD. As such, we hypothesized that modulating macrophage polarization via BRD4 inhibitors is a potential therapeutic approach for COPD.

Material and methods

Patient cohorts and samples

This study utilizes human airway transcriptomic data from two separate cohorts of individuals with COPD in different cities in China. Specifically, the discovery cohort consisted of 70 patients with stable COPD and 18 healthy controls, from whom induced sputum samples were collected at the First Affiliated Hospital of Guangzhou Medical University in Guangzhou. The validation cohort included 24 patients with stable COPD and 18 healthy controls, from whom induced sputum samples were obtained at Shenzhen People’s Hospital in Shenzhen. The inclusion and exclusion criteria were described in previous study [13]. The study was approved by the ethics committee of two centers (reference no. 2017–22 and KY-LL-2020294-01). The inclusion criteria for patients with COPD consisted of being over the age of 40 and having a confirmed diagnosis of COPD based on the GOLD guideline. Specifically, a post-bronchodilator forced expiratory volume in 1 s (FEV1)/forced vital capacity ratio < 0.7. The exclusion criteria were: (1) a diagnosis of known respiratory disorders other than COPD; (2) COPD exacerbation within 4 weeks of enrolment; (3) history of lung surgery and tuberculosis; (4) diagnosis of cancer; (5) blood transfusion within 4 weeks of enrolment; (6) diagnosis of autoimmune diseases; (7) enrolment in a blinded drug trial; and (8) antibiotic usage within 4 weeks of enrolment. All participants provided written informed consent.

The chest CT data in Guangzhou cohort and Shenzhen cohort were imported separately into workstations of the VIDA software (version 2.2, Apollo; VIDA Diagnostics, Coralville, IA, USA) and NeuLungCare-QA (version 1.0, Neusoft Medical Systems Co., Ltd. Shenyang, Liaoning, China) to automatically analyze the extent of emphysema (LAA950%).

Murine model

To establish an emphysema model, male Bagg Albino (BALB/c) mice, aged 6 weeks and weighing 25–30 g, were procured from Gem Phamatech Co., Ltd., located in Guangdong, China. In order to mitigate agonistic behavior and prevent the occurrence of diseases, appropriate measures were implemented for the care of animals. Specifically, all mice were accommodated in a pathogen-free animal facility, where they were housed in isolated cages that provided sufficient ventilation. Additionally, a 12-h light/dark cycle was maintained, and the mice were given unrestricted access to food and water. It is important to note that all experiments conducted in this study were approved by the Animal Subjects Committee of Shenzhen People’s Hospital (AUP-220714-CRC-0599-01).

Establishment of a model with emphysema

The BALB/c mice were administered intratracheal instillation of a mixture containing 7 ug LPS purified from Escherichia coli O26:B6 (Sigma-Aldrich, St. Gallen, Switzerland) and 1.8 U of porcine pancreatic elastase (Elastin Products Company, Owensville, MO) in 50-μL PBS for a total of four times. Following each exposure to LPS/elastase, subgroups of mice were intraperitoneally injected with either the BRD4 inhibitor JQ1 (50 mg/kg), the BRD4 degrader ARV-825 (10 and 20 mg/kg), or a control Vehicle on days 14, 17, 21, 24, 28, and 31 [26].

Lung histopathology

The left lungs of all mice were aseptically collected and subsequently fixed in 4% paraformaldehyde overnight. The following day, the lung tissues were embedded in paraffin and sectioned into 4-um thickness. The selection criteria for the mouse sections were as follows: (1) The sections could display structures such as airways and alveoli; (2) The thickness of the sections was consistent. These sections were then H&E stained, sealed, and scanned at 40 × magnification using a multifunctional pathology scanner. The mean linear intercept (MLI) of the mouse sections was calculated. An average of five fields of view per section was selected, covering different areas of the lung tissue. The total length of all visible alveoli in these fields was measured using Aperio ImageScope software. The total length measured across the selected fields was then divided by the total number of observed alveoli to calculate the MLI for each mouse.

The degree of peri-bronchial inflammation was assessed using a scoring standard ranging from 0 to 4, as previously described [27]. In this scoring system, a score of 0 indicates the absence of inflammatory cells, while scores of 1, 2, 3, and 4 represent the presence of occasional, one layer, two layers, and more than two layers of inflammatory cells around the bronchi, respectively.

Lung function analysis in the emphysema model mice

The Forced Pulmonary Maneuver System (Buxco Research Systems, Wilmington, North Carolina, USA) was employed in accordance with the manufacturer’s guidelines to assess fluctuations in lung function. The respiratory rate of anesthetized mice was standardized to an average of 150 breaths per minute. The study involved the execution of three semiautomatic maneuvers: (a) Boyle’s law functional residual capacity (FRC), (b) quasi-static pressure volume (PV), and (c) fast flow volume (FV). The quasi-static PV maneuver was conducted to assess total lung capacity (TLC) and chord compliance (Cchord). Additionally, the fast FV maneuver was employed to measure forced expiration volumes (FEV50, FEV100, FEV200, and FEV300) in milliseconds, as well as forced vital capacity (FVC).

Bronchoalveolar lavage fluid (BALF) analysis

Following the extraction of blood, the trachea and right lung were subjected to in-situ lavage using a prewarmed sterile solution of 0.9% NaCl saline, with a volume of 0.5 mL. The fluid obtained from this procedure was then examined to determine the total cell count in the BALF and utilized for the preparation of cyto-spin samples. The enumeration of total inflammatory cells, macrophages, neutrophils, and lymphocytes in the BALF was conducted using a hemacytometer.

Protein extraction from lung tissues and western blotting

Total protein was extracted from lung homogenates and the protein concentration was determined using a BCA Protein Assay kit (Thermofisher, USA). Cell lysis was performed at a temperature of 100 °C for a duration of 10 min, followed by the loading of 20-μg of protein into a 10% SDS-PAGE gel. Subsequently, the protein was transferred onto a PVDF membrane (Merck Millipore, Bedford, MA, USA) and incubated overnight at 4 °C with antibodies against BRD4 (1:1000; A301985A100, Thermo-fisher, USA) and β-actin (1:3000; AB2001, AB Ways). The next day, following three rounds of washing with PBST, the samples underwent incubation with anti-rabbit-IgG horseradish peroxidase-conjugated secondary antibodies (CST, USA). Subsequently, the signals were detected through enhanced chemiluminescence (Merck Millipore, Bedford, MA, USA) and recorded using a Chemi Doc^™ MP Imager (Bio-Rad, USA). The Image J program was utilized to measure the integral optical density of each sample.

Immunofluorescence staining

The lung sections were subjected to overnight incubation at 4 °C with antibodies against CD68 (GB113109, Servicebio) and BRD4 (AB128874, Abcam). Subsequently, the lung sections were exposed to a fluorescein isothiocyanate-labeled secondary antibody (GB22303, Servicebio) at room temperature for a duration of 1 h. Furthermore, DAPI (G1012, Servicebio) was employed for nuclear staining, and anti-fluorescence quenching sealer (G1401, Servicebio) was administered. The visualization of staining was carried out using an inverted fluorescence microscope (Servicebio).

Isolation and culturing of alveolar macrophages

The mice underwent tracheal lavage with 1 ml of PBS seven times to obtain BALF. BALF cells were isolated by centrifugation at 1500 rpm for 10 min and subsequently washed with PBS. To promote cell adherence, 12-well plates were utilized, with each well containing 3 × 10⁵ cells, and were incubated for 12 h. Following incubation, the plates were washed twice with PBS to eliminate non-adherent cells. The adhered cells were cultured overnight in a solution containing 20 ng/mL IFN-γ (C746, Novo protein) and 1 μg/mL LPS (Sigma-Aldrich, St. Louis, MO, USA) to induce M1 polarization. Additionally, 20 ng/mL interleukin-4 (CK15, Novo protein) was used to induce M2 polarization. The M1 and M2 macrophages were subsequently harvested for quantitative PCR analysis, while the supernatant was collected for ELISA.

RNA extraction for qPCR analysis and RNA-seq followed analysis

Induced sputum samples were collected and preformed with good quality control as previously described [13]. Transcriptomic data from sputum cells for RNA-sequencing were obtained by using the Illumina NovaSeq platform. The normalized gene count matrix was used to next transcriptomic analysis as described [13]. For a full set of transcriptomic data see the previous report [13]. Differentially expressed genes (DEGs) data from sputum cells analysis was performed using limma in R vegan package, among genes in BET protein family, MMP12 and IRF4. Genes in BET protein family consists of BRD1, BRD2, BRD3, BRD4, BRD7, BRD8, BRD9. Sets of M1 and M2 genes were based on selected classic genes in human expression publications. M1 markers includes 18 genes (NOS2, IL1A, IL1B, IL6, IL12B, TNF, CXCL1, IL1F9, CXCL2, CXCL9, CXCL10, CCL2, IRF5, IRF7, IRF9, STAT1, P65, SPI1). M2 markers includes 12 genes (ARG1, MRC1, STAT6, IRF4, MMP12, ADAM8, ADAM9, IL10, CCL17, CCL22, TGFB1, TFRC). Gene Set Variation Analysis (GSVA) was used to calculate the enrichment score (ES) for each patient and for gene signatures of macrophages and neutrophils. ES values range from − 1 to 1. A linear model for transcriptomic data with Benjamini–Hochberg false discovery rate (FDR) correction was used in the analysis of the DEGs and for GSVA. Spearman Correlation between genes and genes, genes and clinical parameters, genes and ES were calculated.

In order to perform cells and animal model qPCR analysis, total RNA was extracted from AMs using a UNIQ-10 Column Total RNA extraction Kit (IA24KA6842, Sangon Biotech, China). RNA isolation from lung homogenates of LPS/elastase-induced COPD model was performed using the Spin Column Animal Total RNA Purification Kit (HA14KA1972, Sangon Biotech, China). gDNA was removed, and total RNA was reverse transcribed into cDNA using PrimeScript^™ RT reagent Kit supplemented with gDNA Eraser (RR047A, TAKARA, China). qPCR was performed using TB Green^® Premix Ex Taq^™ II (Tli RNase H Plus) (RR820A, TAKAR, China) on a CFX Connect^™ Real-Time PCR system (BIO-RAD, USA). The sequences of primers used were shown in (Table 1).

Table 1 Primers used in the studies

Total amounts and integrity of RNA from cells were assessed using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, CA, USA). The RNA-Seq library was prepared using NEBNext Ultra RNA Library Prep Kit for Illumina (NEB, USA). After the construction of the library, 150bp × 2 paired-end sequences were generated using the Illumina NovaSeq 6000 system at Novogene, Shenzhen.

For the RNA-seq analysis, raw reads were removed with the help of Trimmomatic v0.39 [28], and aligned to the mm10 mouse genome by Bowtie2 v2.2.5 [29]. Duplicate sequences were removed using MarkDuplicates v3.0.0 from Picard software (http://broadinstitute.github.io/picard/). Gene expression level was quantified by Salmon v0.12 [30] followed by normalization using trimmed mean of M-values (TMM) method in edgeR v3.32.1 [31]. DEGs analysis among all groups was performed using edgeR’s glmQLFTest. Significant DEGs were identified using a fold change > 1.5 and FDR < 0.05. Further, Gene Ontology and gene set enrichment analyses were performed using clusterProfiler v3.18.15 [32] and GSEApy v0.10.46 [33], respectively, to investigate the biological functions of DEGs. Heatmaps were generated with the help of ComplexHeatmap v2.6.27 to visualize the expression level of DEGs.

Cleavage under target and tagmentation (CUT and tag)-seq assay

CUT & Tag-seq assay were performed by using NovoNGS^® CUT&Tag^® 4.0 High-Sensitivity Kit (for Illumina^®) (N259-YH01-01B, Suzhou, China). 50,000 alveolar macrophages were bound by ConA-magnetic beads and were resuspended in primary antibody buffer (containing protease inhibitor cocktail, 5% digitonin and primary antibody BRD4 (1:100; A301985A100, Thermo-fisher, USA) overnight at 4 °C. After washing, AMs were incubated with a secondary antibody (1:200) at room temperature for 1 h. After washing again, AMs were incubated with transposome buffer at room temperature for 1 h and tagmentation buffer at 37 °C for 1 h respectively. And termination of tagmentation was performed by adding stop buffer at 50 °C for 10 min. Finally, DNA fragments were extracted by adding Tagment DNA extract beads. Amplified and cleaned DNA fragments were used for sequencing at the Illumina platform.

CUT&tag sequencing data analysis

Raw reads were trimmed and aligned to mm10 mouse genome as aforementioned mentioned. SEACR peak caller v1.3 with default parameters was employed to identify the binding peaks of BRD4 and H3K27ac [34]. Differential peaks between groups were caculated using DiffBind v3.8.4 [35], and the genomic region and genes around peaks were annotated by ChIPseeker v1.34.1 [36]. The coverage of reads for each genome bin of 50 bp was quantified and standardized to counts per million (CPM) using bamCoverage function of deepTools v3.5.1 [37]. Further, R v4.1.0 was employed to construct heatmaps enriched around transcription start sites (TSS), and visualized using EnrichedHeatmap v1.28.1 [38]. A binding density plot of gene body and enhancer regions was generated using ngsplot v2.61 [39]. The identification of super-enhancers (SEs) was carried out by initially combining H3K27ac and BRD4 binding peaks with the help of mergeBed [40]. Following this, the ROSE algorithm was employed to compute enriched H3K27ac signals within 12.5 kb windows surrounding the merged binding sites. By ranking these stitched enhancer regions based on the H3K27ac signal, SEs were identified, and associated target genes were annotated [41]. Genomic signal tracks of desired genes were visualized in IGV browser v2.15.4 [42].

Overexpression experiment

Primary AMs were isolated from mice. The cells were cultured in antibiotic-free growth medium supplemented with 10% fetal bovine serum (FBS) (BioWest). All cells were cultured in a humidified incubator containing 5% CO2 at 37 °C. For gene overexpression, the full-length mouse Irf4 sequence was cloned into a pReceiver-M02 Expression Clone (GeneCopoeia) which were transfected into cells using Lipofectamine 3000 reagents and Opti-MEM (Thermofisher) reagents.

Statistical analysis

Data were tested for normal distribution and homogeneity of variance. Differences were assessed with t test between 2 groups, and one-way analysis of variance (ANOVA) accompanied by Bonferroni’s difference post hoc test for ≥ 3 groups. Data are presented as mean ± SEM. p < 0.05 was regarded as statistically significant.

Results

BRD4 was associated with key clinical parameters during COPD progression

To identify the clinically relevant genes and pathways involved in COPD development, induced sputum was collected from 94 patients with COPD and 36 healthy controls for mRNA sequencing characterization [13]. Notably, BET family proteins mediated epigenetic regulation were known for their critical role in the pathogenesis of various inflammatory diseases [19, 43] and the differential expression of BRD1, BRD2, BRD3, BRD4, BRD7, BRD8, BRD9 in patients with COPD and healthy controls were analyzed. In this study, it was shown that the expression levels of BRD4 were significantly increased in the sputum samples obtained from patients with COPD compared to the healthy control group in the Guangzhou cohort. These findings were subsequently confirmed in the Shenzhen cohort as well (Fig. 1a). Besides, to investigate whether BRD expression was elevated in circulating monocytes, we analyzed single-cell RNA sequencing (scRNA-seq) data of peripheral blood mononuclear cells (PBMCs) from patients with COPD and healthy controls (GSE249584). Differential gene analysis showed no significant changes in BRDs expression between monocytes from COPD patients and healthy controls (Table S1).

Fig. 1

BRD4 was associated with key clinical parameters during COPD progression. a Box plots showed the abundance of BRD1, BRD2, BRD3, BRD4, BRD7, BRD8, BRD9 in COPD (n = 94) and control (n = 36). Significance was determined using a two-sided Wilcoxon rank-sum test. NS: Not significant; b Pairwise correlations between BRD1, BRD2, BRD3, BRD4, BRD7, BRD8, BRD9 and clinical parameters of COPD patients. Significant p-values are shown by color, and Spearman’s correlation analysis was used to estimate significant correlations. Clinical parameters included: BMI: body mass index; postFEV1pct: post FEV1 (% reference); pack year: cigarette consumption per year; CAT: COPD Assessment Test; mMRC: modified Medical Research Council; AE_year_0: acute exacerbation times per year; c Correlations of the expression of BRD4 between FEV1% predicted, mMRC and LAA_950 in Guangzhou and Shenzhen COPD cohorts. d Correlations of the expression of between BRD4 and Macrophage ES or Neutrophil ES in the Guangzhou cohort. e BRD4 expression levels in lung tissue were detected by western blotting in control and model mice (n = 3). f Semiquantitative analysis of western blotting results of BRD4 expression levels in the lung. g Immunofluorescence staining of CD68 and BRD4 in lung tissue of control and model mice. Original magnifications 200 × (left) and 630 × (right) (n = 5). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

We demonstrated that BRD4 exhibited stronger correlations with clinical indicators in comparison other genes within the same family (Fig. 1b). Specifically, an upregulation of BRD4 was associated with poorer lung function (negative correlation with post-bronchodilator forced expiratory volume in one second (FEV1) (% reference), R = − 0.49%, p < 0.001), more severe dyspnea symptoms (positive correlation with modified Medical Research Council (mMRC), R = 0.31, p < 0.05), a higher degree of pulmonary emphysema (positive correlation with LAA950%, R = 0.39, p < 0.005) in Guangzhou cohort and validated in Shenzhen cohort (Fig. 1c). Collectively, BRD4 exhibited stronger associations with clinical parameters.

Of note, COPD is a chronic inflammatory lung disease that involves a variety of immune cells, including macrophages, neutrophils, and lymphocytes [2]. It has been previously found that induced sputum comprises a great amount of innate immune cells like macrophages and neutrophils with few parenchymal pulmonary cells. GSVA was used to calculate the ES of macrophage and neutrophil for each patient based on the specific gene expression. We observed that BRD4 expression significantly positively correlated with macrophage ES and neutrophil ES both in Guangzhou cohort and Shenzhen cohort (Fig. 1d, Figure S1a, b).

In order to further validate the result in vivo, we established the elastase dosage dependent mouse model of emphysema using 7 ug LPS and 0 U, 0.6 U, 1.2 U, 1.8 U, 2.4 U, 3.0 U elastase intratracheal treatment for 4 weeks (Figure S3). Terminal readout was carried out 1 week after the last challenge. Using pulmonary function tests, initially, it was observed that 1.8 U elastase significantly change FEV100/FVC (%) in mice with LPS + elastase-induced emphysema (Figure S3a), whereas 1.2 U elastase induced significantly change on functional residual capacity (FRC) (Figure S2a). The number of inflammatory cells in BALF are upregulated even treatment with 0.6 U elastase (Figure S3b). H&E staining showed that 0.6 U elastase markedly increased lung inflammation in the vicinity of small airways and alveoli calculated according to inflammation score, and 1.2 U elastase treatment significantly expanded alveolar spaces in emphysema mice, which were validated by the statistical analysis of histological slides by calculating MLI (Figure S3c, d). In summary, we select 1.8 U elastase as the best amount to induce emphysema animal model. Subsequentially, the upregulation of BRD4 protein expression in LPS + elastase-induced experimental emphysema model was verified (Fig. 1e, f). The immunofluorescence staining assay showed that BRD4 co-localized with macrophage-specific marker CD68 in lung tissues and displayed increased expression in emphysema mice compared with control mice (Fig. 1g).

BRD4 targeting alleviated experimental emphysema

To evaluate the potential of BRD4 as a therapeutic target for COPD, the BRD4 inhibitor JQ1 (at a dosage of 50 mg/kg) and the degrader ARV-825 (at dosages of 10 and 20 mg/kg) were intraperitoneally administered into mice with induced emphysema. This administration occurred twice a week for four consecutive weeks, commencing 1 week after the initiation of LPS + elastase treatment. Using pulmonary function tests, initially, it was observed that ARV-825 significantly improved FEV100/FVC (%) and suppressed FRC in mice with LPS + elastase-induced emphysema (Fig. 2a), whereas no effect was observed on TLC. To investigate the impact of BRD4 inhibition on immune responses relevant to emphysema, the number of inflammatory cells in BALF was counted. It was found that treatment of JQ1 and ARV-825 significantly inhibited the infiltration of macrophages, neutrophils, and lymphocytes in BALF (Fig. 2b). Subsequently, H&E staining demonstrated that targeting BRD4 markedly alleviated lung inflammation in the vicinity of small airways and alveoli, and accompanied with reduction of the expanded alveolar spaces in emphysema mice, which were further validated by the statistical analysis of histological slides (Fig. 2c). It has been shown that the parameters of inflammation score and MLI were reduced by treatment of JQ1 and ARV-825 in emphysema mice (Fig. 2d, e). In summary, these results strongly indicated that targeting BRD4 holds significant therapeutic potential for emphysema treatment.

Fig. 2

Pharmacological targeting of BRD4 ameliorated LPS/elastase-induced airway inflammation and emphysema. a Measurement of obstructive airflow limitation with FEV100/FVC (%), FRC, and TLC b Measurement of inflammatory cells, macrophages, neutrophils, and lymphocytes in BALF. c Representative H&E-stained pulmonary sections were isolated from different groups. Original magnifications 400 ×. d Semiquantitative analysis of airway inflammation. e Semiquantitative analysis of MLI, n = 5–10 mice per group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

BRD4 inhibition suppressed both M1 and M2 AMs polarization

As shown above, both the COPD cohort and experimental model showed that BRD4 expression displayed a significant association with macrophages during COPD development. Notably, both M1 and M2 macrophage polarization have been reported to be important to COPD development at different stages [44, 45]. As such, the effect of BRD4 targeting on macrophage polarization was further investigated, we found that BRD4 expression were significantly positively correlated with M1 polarization markers (IL1A, IL1B, IL6, TNF, IL12B, IL36G, CXCL1, CXCL2, RELA, SPI1, IRF7, and IRF9) in Guangzhou cohort, and these trends were partially validated in the Shenzhen cohort (Fig. 3a). Subsequently, it was shown that the targeting BRD4 with JQ1 and ARV-825 resulted in a significant decrease in the mRNA expression of Tnf and Il1b in lung tissue, as well as a reduction in the release of IL-6 in BALF in the emphysema mice model (Fig. 3b).

Fig. 3

Transcriptome characterization of BRD4 inhibition in M1 AMs. a Pairwise correlations between BRD1, BRD2, BRD3, BRD4, BRD7, BRD8, BRD9, and M1 macrophage marker genes expression. b Barplot showed the expression of Tnf, Il1b and IL-6 (n = 5) in the lung of emphysema mice treated with or without DMSO, ARV825 (10 mg/kg), ARV825 (20 mg/kg), JQ1 (50 mg/kg). c Heatmap analysis of RNA-seq data change of M1 macrophage when using 3 different BRD4 inhibition drugs. d Venn diagrams showing overlapped genes between M1-specific genes and genes regulated by 3 different BRD4 inhibition drugs. e GSEA enrichment analysis of ARV-825 regulated genes against M1-specific genes in M1 AMs. f GO analysis of ARV825 downregulated genes in M1 AMs. g Heatmap of gene expression in the pathway of leukocyte migration, cytokine production, and IL-6 production of M1 macrophage. h Barplot showed expression of Tnf, Il1b, and Nos2 (n = 3) in the AMs stimulated with LPS/IFN-γ treated with or without DMSO, ARV825 (0.2 µmol), JQ1 (1 µmol). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

To further investigate if BRD4 inhibition can block the M1 AMs polarization, the primary AMs were isolated for M1 polarization in vitro by stimulating with LPS/IFN-γ for 24 h while pretreatment was done with pan-BET inhibitor JQ1, and BRD4 degrader ARV-825 2 h in advance for BRD4 inhibition. The application of RNA-seq analysis revealed that the inhibition of BRD4 using all three drugs resulted in significant suppression of genes associated with M1 polarization, as demonstrated by the heatmap representation (Fig. 3c) of gene expression levels, including Il1a, Il1b, Il6, Tnf, nitric oxide synthase 2 (Nos2), and prostaglandin-endoperoxide synthase 2 (PtgS2) (Table S2). Venn diagram analysis revealed that 55.68% (1108/1990) of M1 macrophage-specific genes were downregulated under the effect of three different drugs (Fig. 3d). In particular, 657 genes were down-regulated by all three drugs (Fig. 3d). Further, GSEA revealed a significant enrichment of M1 macrophage-specific genes among those that ARV-825 downregulated (normalized enrichment score (NES) = − 1.554, p < 0.001) (Fig. 3e). The results of the Gene Ontology (GO) enrichment analysis indicated that the downregulation of M1 macrophage-specific genes caused by ARV-825 primarily affects processes related to the positive regulation of cytokine production, cytokine-mediated signaling pathway, regulation of immune effector process, leukocyte migration, responses to LPS, and regulation of inflammatory response, among others (Fig. 3f). Heatmap analysis displayed a group of downregulated genes mediated by all three drugs, which mainly enriched in pathways of positive regulation of cytokine production, specifically IL-6 production (Fig. 3g). Finally, the verification of BRD4 inhibition-induced suppression of M1 macrophage-specific markers, including Tnf, Il1b, and Nos2, was conducted using quantitative polymerase chain reaction (qPCR) analysis (Fig. 3h).

Next, the functional role of BRD4 during the M2 macrophage polarization was analyzed in COPD patients. As shown in the heatmap, BRD4 expressions were positively correlated with M2 markers in the Guangzhou cohort (ADAM8, STAT6, TGFB, ARG1, and IL10) and the Shenzhen cohort (CCL22, ARG1, IL10, ADAM8, STAT6, IRF4 and TGFB) (Fig. 4a). Subsequently, it was observed that BRD4 targeting with JQ1 and ARV-825 markedly decreased the mRNA expression of Arg1, Ccl17 and Tfrc in lung tissue from the emphysema mouse (Fig. 4b).

Fig. 4

Transcriptome characterization of BRD4 inhibition in M2 AMs. a Pairwise correlations between BRD1, BRD2, BRD3, BRD4, BRD7, BRD8, BRD9, and M2 macrophage expression in Guangzhou and Shenzhen clinical cohorts. b Barplot showed expression of Arg1, Ccl17, and Tfrc (n = 5) in the lung of emphysema mice treated with or without DMSO, ARV825 (10 mg/kg), ARV825 (20 mg/kg), and JQ1 (50 mg/kg). c Heatmap of RNA-seq data of M2 macrophage when using 3 different BRD4 inhibition drugs. d Venn diagrams showing overlapped genes between M2-specific genes and genes downregulated using 3 different BRD4 inhibition drugs. e GSEA enrichment analysis of ARV825 regulated genes against M2-specific gene list. f GO analysis of the differentially expressed mRNAs in BRD4 inhibition drugs of ARV825 downregulated genes. (G) Heatmap analysis of gene expression change in myeloid leukocyte differentiation, leukocyte chemotaxis, and extracellular matrix organization of M2 macrophage when using 3 different BRD4 inhibition drugs. h Barplot showed expression of Arg1, Ccl17 (n = 3) in the AMs stimulated with IL-4 in the treatment with or without DMSO, ARV825 (0.2 µmol), JQ1 (1 µmol). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Further, we investigated the effect of inhibition of BRD4 targeting on transcriptional changes in IL-4 induced M2 AMs using JQ1, ARV-825. In total, 595 genes were upregulated in M2 macrophages, and the majority of these genes were suppressed by all three drugs (Table S3). Among these upregulated genes, marker genes of M2 macrophage, including Arg1, Il1r1, Ccr5, Sema4b, Ccl17, and Cish were all included (Fig. 4c). Venn diagram analysis revealed that 75.75% (421/595) of M2 macrophage-specific genes were downregulated under the effect of three different drugs (Fig. 4d). In particular, 281 genes were down-regulated by all three drugs (Fig. 4d). Further, GSEA revealed that M2 macrophage-specific genes were significantly enriched among those downregulated by ARV-825 (NES = − 1.418, p < 0.001; Fig. 4e). The GO enrichment analysis revealed that the genes particular to M2 macrophages, which were downregulated by the influence of ARV-825, had a strong enrichment in immunological functions and remodeling processes. These processes encompassed myeloid leukocyte differentiation, leukocyte chemotaxis, and extracellular matrix organization (Fig. 4f, g). Lastly, all three anti-BRD4 drugs induced downregulation of M2 macrophage-specific markers such as Arg1 and Ccl17 were significantly downregulated by all three anti-BRD4 drugs (Fig. 4h). These results collectively indicated that BRD4 expression was strongly associated with both M1 and M2 macrophage polarization in clinical cohort and experimental COPD animal model, and its inhibition is potent enough to suppress both M1 and M2 AMs polarization in vitro. We also found other BRDs were associated with M1 or M2 macrophage polarization markers (Figs. 3a, 4a). In order to investigate the effects of other BRDs (e.g. BRD7) on AMs polarization, we treated M1 and M2 AMs with a BRD7 inhibitor (BRD7-IN-2) and found that treatment with the BRD7 inhibitor (BRD7-IN-2) after M1/M2 polarization did not downregulate the transcription of M1/M2 polarization markers (Figure S4), suggesting other BRD inhibitor (e.g. BRD7 inhibitor) do not have similar effects on macrophages polarization as BRD4 degrader, ARV-825.

BRD4 regulated chromatin expression of M2 polarization-related genes by binding at promoter regions

Previous studies have already figured out that BRD4 promotes M1 macrophage polarization-related inflammatory response by facilitating the NF-κB pathway [46, 47]. However, there has been limited research on the role of BRD4 in promoting M2 AMs polarization. Therefore, the subsequent inquiry aimed to clarify the molecular mechanism by which BRD4 influences the polarization of M2 AMs, which contributing to the development of COPD. The interaction between BRD4 and H3K27ac occurs through competitive binding mechanisms, enabling BRD4 to act as a scaffold for transcription factors (TFs) at both promoters and enhancers, therefore modulating the levels of gene expression [48]. A comprehensive analysis was conducted to examine the occupancy of BRD4/H3K27ac throughout the whole genome in M2 macrophages. The study conducted CUT&Tag analysis on IL-4-driven M2 AMs that were treated with or without ARV-825 in vitro. This analysis involved the use of BRD4 and H3K27ac antibodies. In total, 38,800 peaks of BRD4 and 38,610 peaks of H3K27ac were obtained in M2 macrophages; among those, 96.4% of peaks were overlapped, which was slightly higher as in M0 (Fig. 5a). Peaks were widely distributed across the genome, including promoter, introns, and distal intergenic regions. Originally, around 25% of BRD4 and H3K27ac peaks were specifically located in the promoter region (≤ 1kb), which were decreased to < 20% upon treatment with ARV-825 (Fig. 5b). Notably, a significant decrease in BRD4 and H3k27ac binding was observed in the nearby transcription start site (TSS) (Fig. 5c, d).

Fig. 5

BRD4 inhibition decreased BRD4 and H3K27ac occupancy on promoter regions in vitro. a Venn diagram showing BRD4 and H3K27ac overlapped binding peaks in M0 and M2 AMs. b Genome-wide distribution of BRD4 and H3k27ac in AMs from M0 group, M2 group, ARV-825 + M2 group. c Heatmaps of BRD4 and H3k27ac occupancy on the promoter region (TSS ± 5 kb), aligned by the degree of BRD4 and H3k27ac signal intensity in AMs from M0 group, M2 group, ARV-825 + M2 group. d Binding intensity of BRD4 and H3k27ac across the gene body of the whole genome in AMs from M0 group, M2 group, ARV-825 + M2 group

BRD4 inhibition changed the dynamic enhancer epigenome in M2 AMs

To further investigate the regulatory mechanisms underlying the effect of BRD4 inhibition over dynamic enhancer epigenome in M2 AMs, genome-wide binding dynamics of BRD4 and H3K27ac were explored. In total, 24,071 BRD4-associated genes and 20,587 H3K27ac-associated genes were annotated. Among all, 86.89% (517/595) M2-specific genes were co-occupied with BRD4 and H3K27ac and only 5.04% (30/595) M2-specific genes were not associated with BRD4 and H3K27ac (Fig. 6a). Moreover, genes exhibiting reduced BRD4 binding demonstrated a significantly greater fold change decrease compared to other genes after ARV-825 treatment (Fig. 6b). This suggested that genes directly regulated by BRD4 experience the most potent regulatory effect. GSEA was conducted to examine the downregulated genes affected by ARV-825, which indicated significant enrichment of genes associated with the lower peaks of BRD4 and H3K27ac (Fig. 6c). BRD4 and H3K27ac-associated ARV-825 downregulated genes were primarily enriched in pathways related to immune response activation, regulation of immune response through cell surface receptors, signaling pathways mediated by cell surface receptors, leukocyte migration, positive regulation of innate immune response, regulation of DNA-binding transcription factor activity, myeloid leukocyte migration, and innate immune response activation pathways, etc. (Fig. 6d). In M2 AMs treated with ARV-825, CUT&Tag and RNA-seq analysis revealed genomic binding events of BRD4 and H3K27ac, and quantified gene expression levels in the whole genome. The data presented pertains to the gene loci associated with marker genes for M2 macrophage polarization, including Arg1, Ccl22, Ccr5, and Ccl17 (Fig. 6e). These results suggested that BRD4 degraders might disrupt the enhancers of specific genes to regulate the polarization of macrophages.

Fig. 6

BRD4 inhibition changed the dynamic enhancer epigenome in M2 AMs. a The left Venn diagram showed enriched genes annotated to BRD4 peaks, H3K27ac peaks in M2 AMs, and their overlapped genes with upregulated genes in M2 macrophage; the right Venn diagram showed overlapped genes between genes associated with ARV825 downregulated BRD4 peaks, genes associated with ARV825 downregulated H3K27ac peaks, and ARV825 downregulated genes in M2 AMs. b The boxplot showed significantly differential gene expression fold changes between genes associated with and not with downregulated BRD4 peaks and H3K27ac peaks in M2 AMs with ARV-825. c GSEA plots indicated genes associated with downregulated BRD4 and H3k27ac peaks are significantly downregulated by ARV-825 in M2 macrophage. d GO analysis of the ARV-825 downregulated genes associated with BRD4 and H3k27ac downregulated peaks in M2 AMs. e Gene tracks showed strong downregulated signals of the M2 marker genes Arg1, Ccl17, Ccr5, and Ccl22 after ARV-825 treatment. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

BRD4 inhibition disrupted IRF4 super-enhancer formation in M2 AMs

SEs, which regulate gene expression involved in multiple cellular processes including macrophage-associated immune responses, have broadly been identified as genomic regions with top-ranked enrichment(s) of H3K27ac and BRD4 [49]. To elucidate whether SEs are critical to the M2 AMs polarization, SEs were identified in M0, M2, and M2 macrophages treated with ARV-825 by analyzing BRD4 and H3K27ac CUT&Tag data with the help of the ROSE algorithm (Fig. 7a). The 952 M2 and 829 M2/ARV-825 associated SEs were identified. The findings of this study demonstrated considerable modifications in H3K27ac and BRD4 binding patterns at many SEs during polarization of M2 AMs. These alterations were significantly reversed by the administration of ARV-825, particularly at the SEs associated with Irf4 enhancer sites (Fig. 7b). This data uncovered a mechanism where the expression of Irf4 is increased in M2 due to the combined binding of BRD4 and H3K27ac to its SE site, while also being suppressed by ARV-825 interventions (Fig. 7c, d; Figure S5a). The binding sites of IRF4 were significantly overlapped with those of BRD4 and H3K27ac, and the binding capacity of IRF4 to BRD4 and H3K27ac peaks was significantly decreased following drug intervention (Fig. 7e, f; Figure S6a). As a result, the ability to bind to key downstream targets is diminished, leading to a significant decrease in Mmp12 which is linked with emphysema during COPD (Fig. 7g) [50]. The further functional rescue experiment showed that over-expression of Irf4 in the ARV-825-treated M2 AMs can partially recover the BRD4 inhibition mediated suppression of Mmp12 (Fig. 7h). In addition, it was observed that BRD4 targeting with JQ1 and ARV-825 markedly decreased both the mRNA expression of Irf4 and Mmp12 in lung tissue in the emphysema mouse (Fig. 7i). Moreover, the clinical cohort analysis from both Guangzhou and Shenzhen showed that the expression of both IRF4 and MMP12 was elevated in COPD patients compared to healthy control, and the expression of IRF4 was significantly positively correlated with MMP12 in both cohorts (Fig. 7j, k).

Fig. 7

BRD4 inhibition disrupted IRF4 SE formation in M2 AMs. a Super enhancers of M0, M2, and ARV825-M2 macrophage, which are ranked by their H3K27ac signaling using ROSE. b Venn diagram showed the overlap of genes associated with SEs sites, M2-specific genes, and ARV825 downregulated genes in M2. c Gene tracks of Irf4 associated SE in M2 AMs. The adjacent genes IRF4 showed strong SE peaks and RNA-seq signals in M2 AMs and intensively reduced when using ARV-825. d Volcano plot depicting significant downregulation of Irf4 in M2 macrophage by BRD4 inhibition drug ARV825. e Venn diagram showed overlapped peaks of BRD4, H3K27ac, and IRF4 in M2 AMs. f Binding intensity of IRF4 across the BRD4 binding sites of the whole genome in AMs from the M0 group, M2 group, and ARV-825 + M2 group. g Gene tracks of Mmp12 in the M2 AMs. The adjacent gene Mmp12 showed strong IRF4 peaks and RNA-seq signals in the M2 AMs. h Barplot showed expression of Irf4, Mmp12, and Tfrc in response to IRF4 transfection in M2 AMs with or without treatment with ARV825 (n = 5). i Transcription level of Irf4, Mmp12 was significantly downregulated by BRD4 inhibition drugs in pulmonary sections from LPS/elastase-induced COPD mice. The values were detected by quantitative PCR (n = 5). j Box plots showed the abundance of IRF4 (Guangzhou p = 0.006; Shenzhen p = 0.0012), and MMP12 (Guangzhou p = 0.00056; Shenzhen p = 0.00022) in COPD and healthy control. k The gene expression of IRF4 and MMP12 were significantly correlated in Guangzhou and Shenzhen COPD patients. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Discussion

The present study delves into the pivotal role of BRD4 within the immune system, elucidating their vital contribution to the pathogenesis of chronic airway inflammatory diseases, particularly in the context of emphysema. A significant increase in the expression of BRD4 has been seen in the produced sputum of patients with COPD, as well as in the macrophages of mice in the experimental model of emphysema. However, there were no significant changes in BRD expression between monocytes from COPD patients and healthy controls, suggesting that the BRD4 paradigm is specific to sputum/airway macrophages and not monocyte-derived macrophages. Additionally, through analysis of RNA-seq data from two separate clinical cohorts in China, it was found that the expression of BRD4 was strongly correlated with both clinical parameters and the expression of macrophage polarization markers. Crucially, the present findings provided evidence that targeting BRD4 using JQ1 and ARV-825 yielded substantial therapeutic benefits, including improved lung function, alleviation of lung inflammatory responses, reduction of air space enlargement in the experimental emphysema model, as well as inhibition of both M1 and M2 AMs polarization. In particular, the current research elucidated the regulatory mechanism of BRD4 inhibition mediated suppression of M2 AMs polarization, which at least partially involved the disruption of the SEs of Irf4, and thereby further impacted the expression of Mmp12, a vital pathogenic gene of emphysema. The aforementioned findings indicated that the epigenetic machinery of BETs in AMs has promise as a viable target for therapeutic interventions in COPD.

Initially, the BET family proteins were analyzed in clinical cohort RNA-seq data, the present study revealed a notable upregulation of BRD4 in the Guangzhou cohort, which also exhibited a statistically significant rise in the Shenzhen cohort. Further analysis revealed that BRD4 had a strong correlation with clinical index mMRC and post FEV1 (% reference), which are the key parameters for evaluating COPD severity. In particular, it was interesting to observe that LAA-950, a parameter to test the extent of emphysema generated by CT scanning, was significantly positively correlated to BRD4 expression. Moreover, the results exhibited a significant correlation between BRD4 expression and macrophage with the help of GSVA analysis, which was verified in the experimental emphysema mouse model. The present outcomes were also supported by a study from Duan et al. [51], which reported that BRD4 expression was strongly associated with COPD viral exacerbation and also demonstrated that BRD4 expression both in the blood and in the sputum were significantly correlated with FEV1% predicted in stable COPD patients. Tang et al. [52] reported a correlation between BRD4 expression and lung function by analyzing COPD lung tissue samples and detected an increased expression of BRD4 in the nuclei of bronchial epithelial cells by immuno-histofluorescence. In this study, the BET family proteins were analyzed in the sputum from two independent COPD cohorts. The proposal suggests that further investigation is needed to validate the functional significance of BRD4 in macrophages during the development of COPD, given that macrophages are a prominent immune cell type in sputum samples.

Before conducting the functional analysis of BRD4 targeting in the disease model, a precise intratracheal instillation approach was constructed to directly administer LPS/elastase into the lungs. This technique was used to produce COPD in a mouse model, following a procedure developed by Yadava et al. [26]. Furthermore, it has been determined that an elastase concentration of 1.8 U is suitable for conducting intervention studies (Figure S2). Previous research has been conducted to investigate the impact of BRD4 inhibition in cigarette-induced COPD animal models [53]. Insufficient research has been conducted to establish the therapeutic efficacy of targeting BRD4 in the treatment of emphysema. To bridge this knowledge gap, the current study was conducted to examine the effects of BRD4 targeting through the use of JQ1 and ARV-825 in a mouse model of emphysema. The findings revealed that BRD4 inhibition significantly reduced the lung inflammatory response in emphysema mice, improved lung function, and air space enlargement as well. Panagis Filippakopoulos et al. (2010) published a seminal study introducing JQ1, a small chemical compound with the ability to permeate cell membranes. JQ1 was found to competitively bind to acetyl-lysine recognition motifs, specifically targeting bromodomains. This compound was quickly acknowledged as a pioneering selective inhibitor of BRD4, making it a valuable tool for cancer biology research [54]. The first confidential evidence demonstrating that BRD4 inhibition with JQ1 was a potential therapeutic approach for inflammatory diseases involving macrophage-mediated immune responses was provided by the Gerald V. Denis laboratory in 2013. They showed that BRD2 and BRD4 physically regulated the activity at the promoters of inflammatory genes in macrophages, which were inhibited by the treatment with JQ1, and consequently protected mice from an LPS-induced cytokine storm and death [55]. Owing to the crucial role of BRD4 in various disease pathogenesis, a diverse type of BRD4 modulators have been designed for a variety of disease interventions including autoimmune diseases, organ fibrosis, vascular diseases, and inflammatory lung diseases. Within the field of COPD, Liu et al. (2021) conducted a study investigating the potential protective effects of JQ1 against COPD produced by cigarette smoke (CS) in mice [53]. In a similar vein, Zakarya and colleagues determined that JQ1 can suppress small airway fibrosis associated with COPD [19]. In 2015, Craig M Crews and his research group utilized PROTAC technology to develop ARV-825. This compound functions by recruiting BRD4 to the E3 ubiquitin ligase cereblon, leading to the rapid, efficient, and sustained degradation of BRD4 through the proteasome pathway. Notably, ARV-825 demonstrated superior efficacy in inhibiting c-MYC expression and suppressing cancer cell growth compared to other BRD4 inhibitors, such as JQ1 and OTX015 [56]. As anticipated, it was observed that the 20 mg/kg of ARV-825 exhibited a similar ability to 50 mg/kg of JQ1 in ameliorating the experimental emphysema. Additionally, a concentration of 0.2 μmol of ARV-825 was more effective than 1 μmol of JQ1 to suppress M2 AMs polarization in vitro. Hence, the findings indicated that the utilization of small compounds based on PROTAC exhibited significant potential for therapeutic intervention in chronic inflammatory lung diseases.

This study aimed to investigate the inhibitory effect of targeting BRD4 on the inflammatory response generated by LPS/Elastase. This choice was supported by recent reports indicating that BRD4 has the potential to influence the activation of inflammatory cells by disrupting the Janus kinase/signal transducers and activators of transcription (JAK/STAT) and nuclear factor kappa B (NF-κB) signaling pathways [57, 58]. Similarly, it was shown that BRD4 inhibition significantly blocked the M1 AMs polarization, however, it was more attractive to see the improvement of air space enlargement. It was speculated that BRD4 inhibition caused the suppression of M2 AMs, suggesting its significance in this particular process. Moreover, in this study, the underlying mechanism of how BRD4 regulated the M2 AMs polarization was also elucidated and thereby contributed to the development of COPD in the part of the mechanistic study.

Young et al. (2013), first proposed SEs, a large cluster of transcriptional enhancers comprising a complex array of sequence elements that drive the expression of specific genes. SEs are much more likely to work as modulators of the vital processes in normal cells and pathological processes compared to conventional enhancers [59]. Recently, there has been a growing body of researches that have been dedicated to investigating the impact of SEs on the development of abnormal transcriptional programs in immune cell dysfunction. Several studies have examined the functional involvement of BRD4 in the process of M2 macrophage polarization. However, one study conducted by Das et al. (2021) has reported on the promotion of M2 polarization during Leishmania donovani parasite infection through the transcription of miR146a-5p, which is controlled by SE [49]. Therefore, the role of SEs in M2 polarization is still largely unknown. The present study observed a high enrichment of SEs in M2 AMs in relation to H3K27ac/BRD4 co-occupied areas. Specifically, there were 952 instances of enrichment. However, when ARV-825 therapy was administered, the number of enriched SEs decreased to 829. This finding aligns with a previous investigation in 2023 conducted by Carelock ME, Master RP, Kim M-C, Jin Z, Wang L, Maharjan CK, Hua N, De U, Kolb R and Xiao Y [60]. Through the examination of RNA-seq data from M2 AMs subjected to treatment with BRD4 inhibitors and degraders, it was shown that IRF4, a crucial transcription factor necessary for M2 polarization, was the sole gene that experiences suppression by all three anti-BRD4 drugs. As expected, the administration of ARV-825 exhibited a notable reduction of SEs in the gene regulatory regions involving the transcription factor Irf4, as well as inhibition of BRD4. These combined effects had a discernible impact on the expression of downstream markers associated with M2 polarization, such as MMP12. MMP12 plays a key role in regulating emphysema in individuals suffering from COPD [50]. Subsequently, the data demonstrated that ARV-825 treatment reduced the binding sites of IRF4 on Mmp12, and similar results were presented in 2021 by Fu Y, Saraswat A, Wei Z, Agrawal MY, Dukhande VV, Reznik SE and Patel K [61]. This work offers valuable insights into the impact of anti-BRD4 medicines and their associated side effects, as well as the regulatory role of IRF4 in M2 AMs among patients with COPD. Furthermore, it identifies possible therapeutic targets for future studies. In addition, relationship between IRF4 and MMP12 were verified in the clinical cohort by a functional rescue experiment involving over expression of IRF4. It is worth mentioning that chromatin immunoprecipitation sequencing (CHIP-seq) is a well-documented epigenetic technique utilized for the identification of SEs through the profiling of H3K27ac and BRD4. Nevertheless, the current approaches frequently encounter difficulties related to the requirement for greater cell inputs, as well as the time-consuming and intricate nature of the operation. In the present work the CUT & Tag technique was applied to replace CHIP-seq as it is much faster than CHIP-seq and reduced sequencing depths. Specifically, this technique allows for the utilization of smaller cell inputs, which are very appropriate for the study of primary AMs separated by BALF. To the best of our knowledge, this research is the pioneering work which applied this technology for the analysis of AMs.

Conclusion

In conclusion, this study’s clinical cohort revealed a stronger correlation between BRD4 and COPD evaluating parameters. The in vivo experiments demonstrated that BRD4 inhibition mitigated mouse emphysema and suppressed both M1 and M2 AMs polarization. Lastly, a novel mechanism by which BRD4 regulates the M2 AMs polarization by disrupting the formation of IRF4 SEs was uncovered by using the novel CUT& Tag assay. Our bioinformatics investigations have delineated the critical involvement of super-enhancers in modulating alveolar macrophage polarization. However, the specific mechanisms underlying their regulatory influence remain to be elucidated, which will constitute the focus of our future research efforts. Taken together, the current study suggests that targeting BRD4 in macrophages holds potential as a therapeutic approach for COPD.