Timing of Glucocorticoid Treatment Dictates Glucocorticoid Receptor Actions Modulating the NLRP3‐Inflammasome Activation in Macrophages

ABSTRACT

Glucocorticoids (GCs) are lifesaving medicines prescribed to treat inflammatory diseases. Macrophages play a pivotal role during hyperinflammation and cytokine storm, two processes intricately linked to NLRP3 inflammasome activation. Macrophages undergo a process of transcriptional reprogramming to resolve inflammation and restore homeostasis. We hypothesized that glucocorticoid receptor (GR) signaling contributes to macrophage metabolic reprogramming to overcome the NLRP3‐inflammasome activation through genomic effects induced by GCs. Glucocorticoid administration following prolonged exposure to lipopolysaccharide (LPS) decreases the expression of iNOS and ACOD1 and their respective metabolic products, nitric oxide and itaconate, to maintain an intact tricarboxylic acid (TCA) cycle in WT mouse pro‐inflammatory macrophages. Glucocorticoids also antagonize the LPS‐induced glycolytic switch through their regulation of mitochondrial dynamics. In addition, we show that glucocorticoids inhibit NLRP3 inflammasome activation and subsequent pyroptosis following extended LPS priming. These suppressive glucocorticoid actions were associated with a marked expansion in the GR cistrome following LPS‐mediated changes in the chromatin landscape. The glucocorticoid inhibition of the late NLRP3 inflammasome activation also was preserved in human monocyte‐derived macrophages. The glucocorticoid effects were attenuated when LPS and glucocorticoids were added together, and they were largely absent in myeloid‐GR knockout mice. This study, employing a glucocorticoid treatment regimen with clinically relevant timing, suggests an underlying metabolic mechanism by which glucocorticoids preserve the integrity of the TCA cycle and inhibit the glycolytic switch induced by LPS. The glucocorticoid regulation precedes the second signal for the NLRP3 inflammasome activation to strategically prevent hyperinflammation and pyroptosis in macrophages.

Utilizing a clinically relevant course of treatment, our study reveals that glucocorticoids protect against the late NLRP3 inflammasome activation regulating the IL‐1β secretion, the processing of GSDM‐D and pyroptosis following signal 2 stimulation. These potent anti‐inflammatory effects are mediated by the activation of GR signaling and the expansion in the GR cistrome following LPS‐mediated changes in the chromatin landscape which effectively bridges metabolic and immunological processes through the inhibition of key metabolic enzymes iNOS and ACOD1 that influence the mitochondria function, the TCA cycle and the mitochondrial ROS production after LPS priming.

1. Introduction

Glucocorticoids act on numerous cell types and tissues of the body; however, one of their most important effects is to suppress the immune system [1, 2]. Synthetic glucocorticoids, such as dexamethasone and prednisone, rank among the most widely prescribed anti‐inflammatory and immunomodulatory drugs. The therapeutic efficacy of glucocorticoids extends across a broad spectrum of immune diseases, including acute and chronic inflammation, hematological cancer, and viral infections, notably exemplified by their role in managing conditions like COVID‐19 [2, 3, 4, 5]. The action of glucocorticoids in immune cells is mediated by the glucocorticoid receptor (hereafter GR; encoded by Nr3c1 gene) that is a member of the nuclear receptor superfamily of ligand‐dependent transcription factors [6, 7]. Macrophages emerge as central targets for the anti‐inflammatory actions of glucocorticoids during periods of inflammation [8, 9, 10]. They stand as pivotal cells in the orchestration of immune responses and maintenance of tissue homeostasis [11, 12]. Macrophages undergo a highly dynamic metabolic reprogramming in response to signals associated with infection or damage [13, 14]. Recently, it has been reported that glucocorticoids also exert regulatory control over a network of genes associated with cellular metabolism in macrophages [15]. This connection ties their anti‐inflammatory effects to mitochondrial function by modulating reactive oxygen species (ROS) production and the functionality of the tricarboxylic acid (TCA) cycle [15]. In addition to coordinating macrophage metabolism, glucocorticoids have also been shown to increase the production of itaconate by non‐genomic mechanisms that involve the interaction of GR with the enzyme pyruvate dehydrogenase (PDH) within the cytoplasm [16]. The enzyme aconitate decarboxylase 1 (ACOD1, encoded by Acod1 or immune‐responsive gene 1; Irg1) is responsible for itaconate production and the enzyme inducible nitric oxide synthase (iNOS, encoded by Nos2 gene) for nitric oxide production. Both itaconate and nitric oxide, recognized as immunomodulatory metabolites, have emerged as key players suppressing and/or establishing tolerance to the NLRP3‐inflammasome activation [17, 18, 19]. Given that glucocorticoids are known regulators of iNOS expression and nitric oxide (NO) production in macrophages [20, 21], it is plausible to anticipate their influence on the regulation of macrophage metabolic remodeling. Notably, macrophages are primary effectors of NLRP3‐inflammasome activation and play a pivotal role in releasing potent pro‐inflammatory mediators and triggering pyroptosis, a form of cell death induced by inflammation [22, 23]. The mechanism of NLRP3 inflammasome activation requires two distinct steps: priming (signal 1) and activation (signal 2) [24]. The inflammasome plays a critical role in the activation of inflammation in human diseases. The excessive activation of the NLRP3 inflammasome is the basis of many acute or chronic human inflammatory diseases [25]. GR orchestrates the regulation of conventional NLRP3‐inflammasome activation that occurs during a short‐term exposure to LPS by directly diminishing the priming step through its actions as a transcription factor on the mRNA levels of Nlrp3 and Il1b [26, 27, 28]. However, the actions and effectiveness of glucocorticoids to suppress the NLRP3 inflammasome following a prolonged LPS exposure are incompletely understood. We hypothesized that utilizing a clinically relevant glucocorticoid course of treatment, in which glucocorticoids are administered after a prolonged period of LPS priming, may play a unique role in suppressing NLRP3 inflammasome activation through the regulation of metabolic genes and mitochondrial dynamics in activated macrophages. Our studies, performed in macrophages from WT and myeloid GR knockout mice as well as primary human cells, provide compelling molecular evidence that glucocorticoids protect against overactivation of signal 2 by ATP in the late NLRP3‐inflammasome activation and pyroptosis pathway. Elucidating the mechanism by which glucocorticoids achieve their anti‐inflammatory effects on NLRP3‐mediated pyroptosis could pave the way for exploring novel therapeutic applications of glucocorticoids in clinical settings.

2. Materials and Methods

2.1. Reagents

Dexamethasone (Dex) and the GR‐antagonist RU486 were purchased from Steraloids Inc. (Newport, RI, USA). Heat‐inactivated fetal calf serum and charcoal‐stripped heat‐inactivated FBS were purchased from Gemini Bio‐Products (West Sacramento, CA, USA). RPMI medium, penicillin/streptomycin, and HEPES (pH 7.0) were purchased from Invitrogen (ThermoFisher Scientific, Carlsbad, CA, USA). Recombinant human M‐CSF was purchased from Biolegend (San Diego, CA, USA). Recombinants mouse M‐CSF, IFNg and IL‐4 were purchased from Miltenyi Biotec (Gaithersburg, MD, USA). Human and mouse anti‐GR (RRID:AB_11179215), anti‐IRG1/ACOD1 (RRID:AB_3064865), anti‐iNOS (RRID:AB_2687529), anti‐NLRP3 (RRID:AB_2722591), anti‐DRP‐1 (RRID:AB_10950498), anti‐MFN‐2 (RRID:AB_2716838), anti‐pro‐IL‐1β (RRID:AB_2721117), anti‐Caspase‐1 (RRID:AB_2890194), anti‐GSDM‐D (RRID:AB_2916333), anti‐cleaved IL‐1β (RRID:AB_2799639), anti‐NRF2 (RRID:AB_2715528), anti‐HIF‐1α (RRID:AB_2622225), and anti‐DUSP1/MKP1 (RRID:AB_3371713) were purchased from Cell Signaling Technology (Danvers, MA, USA). Lipopolysaccharide from E. coli 0111: B4 strain (LPS‐EB), NLRP3 inhibitor MCC950, Caspase‐1 inhibitor Ac‐Y‐VAD‐cmk, 4‐octyl‐itaconate (4‐OI), adenosine 5′‐triphosphate disodium salt (ATP), nigericin, and monosodium urate (MSU) crystals were purchased from InvivoGen (San Diego, CA, USA). The proteasome inhibitor MG‐132 was purchased from Millipore Sigma (St. Louis, MO, USA). The inhibitor of hypoxia‐inducible factor (HIF) prolyl hydroxylase Roxadustat was purchased from Selleckchem (Houston, TX, USA). The Succinate Dehydrogenase assay (Cat# MAK561) was purchased from Sigma‐Aldrich (Darmstadt, Germany). The TaqMan RT‐PCR primer probes were purchased from Applied Biosystems (Foster City, CA, USA).

2.2. Mouse Colony Maintenance

All studies on mice were approved and performed according to the guidelines of the Animal Care and Use Committee (NIEHS). The mice employed in this study were derived from a lineage harboring the floxed GR locus (GR flox/flox, GR^fl/fl), as previously detailed [29]. These mice were mated with counterparts expressing Cre recombinase under the direction of the CX3C chemokine receptor 1 (Cx3cr1) locus (The Jackson Laboratory, #025524). The resulting GRfl/fl Cx3cr1 Cre/+ mice, designed as GR ^Cx3cr1−Cre (specific myeloid‐GR knockout mice or GRKO), and their Cre^/− counterparts, the GR^fl/fl Cx3cr1^+/+ (GR^fl/fl, referred to as wild type, WT), served as controls. All mice were maintained on a C57BL/6NJ background. The data presented in this study encompass both female and male mice, depending on the specific procedure conducted. Genotypes were determined through real‐time PCR, utilizing specific probes designed for each gene (Transnetyx). Mice were maintained in a pathogen‐free facility with 12‐h day‐night cycles, receiving standard mouse chow and water ad‐libitum. Stringent measures were taken to minimize animal suffering, reduce the overall number of mice utilized, and employ alternative techniques to in vivo methods whenever possible.

2.3. Isolation of Bone Marrow Derived Macrophages (BMDM) and Polarization to Proinflammatory Profile

Bone marrow derived macrophages (BMDMs) were obtained from 8‐ to 12‐week‐old GR^fl/fl (WT), myeloid‐specific GRKO mice. The isolation involved flushing the femur and tibia with complete medium, comprising RPMI‐1640 medium supplemented with 10% heat‐inactivated fetal bovine serum (FBS, Hyclone, GE Healthcare, IL), 2 mM L‐glutamine, and 100 U/mL penicillin–streptomycin. Bone marrow monocytes (BMM) were purified through negative selection using the EasySep mouse monocytes Isolation kit (StemCell Tech, Vancouver, BC, Canada) and resuspended in complete medium supplemented with 75 ng/mL M‐CSF. The cells, at a density of 1.0 × 10⁶ cells/plate, were incubated in 10 cm plastic petri‐dishes plates for 6 days at 37°C and 5% CO₂, with a medium change every 3 days. On day 6 post‐isolation, non‐enzymatic harvesting using the macrophage detachment solution DXF1 (PromoCell, Heidelberg, Germany, RRID:SCR_023579) was performed, and cells were seeded in complete medium at a concentration of 1.0 × 10⁶ cells/mL in tissue culture plates. For Dex effects on LPS priming assays, BMDMs were initially stimulated with 100 ng/mL of LPS for 16 h. Subsequently, the medium was replaced with fresh 10% charcoal‐stripped medium, and treatment with 100 nM Dex was carried out for 4–24 h. Some experiments were carried out in the presence of LPS and Dex as a co‐treatment (LPS + Dex).

2.4. NLRP3 Inflammasome Assays

For NLRP3 inflammasome stimulations, we initiate their activation by adding LPS (100 ng/mL) for intervals as indicated, followed by the addition of ATP (5 mM) or Nigericin (10 μM, Invivogen). Depending on the procedure, co‐treatment with 100 nM Dex was added or Dex administered after LPS priming during the last 2 or 8 h. In experiments involving NLRP3 blockers such as MCC950 (50 μM), 1400 W (500 μM) and 4‐OI (150 μM), these agents were added during the last 2 or 8 h as Dex‐treatment. Additionally, Y‐VAD (10 μM) was added 1 h before the administration of ATP or Nigericin.

2.5. Seahorse Extracellular Flux Analysis

BMDMs derived from both WT and myeloid‐GRKO mice were subjected to different treatments: LPS for 24 h, LPS for 16 h and Dex for the last 8 h (16/8) and Dex alone for 8 h. Following treatment, the macrophages were seeded at a density of 150 000 cells/well in a Seahorse 24‐well plate for extracellular flux experiments using the Agilent Technologies platform. The macrophages underwent a XF Cell Mito Stress Test and XF Glycolysis Stress test in a XFe24 analyzer, following the manufacturer's protocol. The Oxygen Consumption Rate (OCR) and the Extracellular Acidification Rate (ECAR) were measured by the software, and the results were normalized to protein content per well, which was quantified using Nanodrop.

2.6. Reactive Oxygen Species (ROS) Detection and Mitochondrial Content

BMDMs obtained from WT and myeloid GRKO mice were seeded in 6 cm plastic petri dishes at a density of 1.0 × 10⁶ cells/well per well. Subsequently, the cells were treated with LPS for 24 h, LPS for 16 h and Dex for the last 8 h (16/8) and Dex alone for 8 h. Mitochondrial reactive oxygen species (mitROS) levels were assessed using MitoSOXRed staining (final concentration 5 μM, Invitrogen) and analyzed via flow cytometry following the manufacturer's protocols. To determine mitochondrial content, live cells were stained with 100 mM Mitotracker Green FM (ThermoFisher, M7514) and analyzed by flow cytometry according to the manufacturer's instructions.

2.7. Analysis of Gene Expression Using RT‐qPCR

The time‐course analysis of gene expression using RT‐qPCR was conducted on BMDM in both unpolarized and polarized macrophages with 100 ng/mL LPS. Cultures consisting of 1.0 × 10⁶/mL unpolarized and polarized macrophages were stimulated for 4 h with vehicle or 100 nM of Dex. Following Dex stimulation, cells were harvested and lysed to extract total RNA, employing the Qiagen RNeasy minikit (Qiagen, Hilden, Germany). Predesigned and validated TaqMan primer/probes sets for each analyzed transcript were employed (Applied Biosystems).

2.8. RNA Sequencing Analysis

Transcriptome analyses was performed on RNA from control and LPS‐primed BMDM from WT and GRKO in presence of vehicle or 100 nM Dex for 4 h. Following Dex stimulation, cells were harvested and lysed to extract total RNA, employing the Qiagen RNeasy minikit (Qiagen). The integrity of the total RNA was determined using the TapeStation (Agilent Technologies), and RNA concentration was assessed using a Qubit (ThermoFisher Scientific). A TruSeq RNA kit (Illumina, San Diego, CA) was utilized to prepare the poly(A)‐enriched RNA‐seq libraries that were sequenced on the Illumina NovaSeq 6000 in a 75‐base paired‐end mode according to the manufacturer's protocol. The RNA‐seq data will be available in the Gene Expression Omnibus repository at the National Center for Biotechnology Information (RRID:SCR_005012). Raw reads (69–105 million reads per sample) were multiplexed to each sample according to their barcode information. Trim_galore (V 0.4.4) was used to remove or trim adaptor‐containing reads and low‐quality reads. Remaining reads were aligned to the UCSC mm10 reference genome using STAR (V 2.5.1; RRID:SCR_004463). The quantification results from featureCount were then analyzed with the Bioconductor package DESeq2 (V 1.38.3; RRID:SCR_006442), which fits a negative binomial distribution to estimate technical and biological variability. We made comparisons for treatments and WT versus GRKO samples. A gene was considered differentially expressed if the p‐value for differential expression was < 0.01. The volcano plots were made using Partek Genomic Suite (Version 6.6). Pathway enrichment analysis was performed with GSEA against the selected databases includes Gene Ontology (GO) cellular component (CC), biological pathway (BP) and Molecular Function (MF) plus REACTOME.

2.9. Cleavage Under Targets and Tagmentation (CUT&Tag) Assay and Analysis

BMDM from vehicle, LPS treated for 4 h; 4 h Dex after LPS‐primed and 4 h LPS and Dex as a co‐treatment were harvested and subjected to Cut&Tag analysis using an improved published protocol for GR antibodies [30]. Trypsinized cells were washed once with PBS and incubated with NE nuclear extraction buffer (20 mM) HEPES‐KOH, pH 7.9, 10 mM KCl, 0.1% Triton X‐100, 20% Glycerol, 0.5 mM Spermidine, 100 μM PMSF, and 1× EDTA‐free protease inhibitors (Roche) for 10 min on ice. Samples were centrifuged for 3 min at 600  g , nuclei pellet resuspended in NE buffer supplemented with 0.1% formaldehyde and incubated at room temperature for 2 min followed by quenching with 0.125 M glycine. Following formaldehyde cross‐linking, nuclei were re‐centrifuged, resuspended in fresh NE buffer and counted with trypan blue dye using Bio‐Rad TC20 Automated Cell Counter. 100 000 total nuclei were incubated with activated concanavalin A‐coated (ConA) magnetic beads (EpiCypher) in 0.2 mL thin‐walled tubes for 10 min at room temperature. Nuclei‐bound ConA beads were resuspended in antibody incubation buffer (20 mM HEPES, pH 7.5, 150 mM NaCl, 0.01% Digitonin, 2 mM EDTA, 0.5 mM Spermidine, 100 μM PMSF, and 1× EDTA‐free protease inhibitor), 0.5 μg primary antibody added and samples incubated overnight at 4°C with gentle rotation. Samples were incubated with 0.5 μg species‐specific secondary antibody for 30 min at room temperature with gentle rotation followed by two washes with Digitonin150 wash buffer (20 mM HEPES, pH 7.5, 150 mM NaCl, 0.01% Digitonin, 0.5 mM Spermidine, 100 μM PMSF, and 1× EDTA‐free protease inhibitor) then incubation with pAG‐Tn5 in Digitonin300‐wash buffer (20 mM HEPES, pH 7.5, 300 mM NaCl, 0.01% Digitonin, 0.5 mM Spermidine, 100 μM PMSF, and 1× EDTA‐free protease inhibitor) for 1 h at room temperature. After pAG‐Tn5 incubation, bound ConA beads were washed twice with Digitonin300‐wash buffer then resuspended in Tagmentation buffer (Digitonin300 wash buffer containing 10 mM MgCl2) and incubated for 1 h at 37°C. Beads were then pelleted, resuspended in TAPS buffer (10 mM TAPS pH 8.5, 0.2 mM EDTA). TAPS buffer was removed and beads carefully resuspended in SDS Release buffer (10 mM TAPS pH 8.5, 0.1% SDS) and incubated for 1 h at 58°C. Following incubation, SDS Quench buffer (0.67% Triton X‐100) was added to each sample along with 2 μL of a universal i5 and a uniquely barcoded i7 primers (from 10 μM stocks). Equal volume (25 μL) NEBNext High‐Fidelity 2× PCR Master mix (New England Biolabs) was added to each sample and mixed by gentle pipetting. Samples were subjected to DNA amplification using a thermocycler and the CUT&TAG‐specific PCR cycling parameters: 58°C for 5 min; 72°C for 5 min; 98°C for 45 s; with 19 cycles of 98°C for 15 s and 60°C for 10 s; final extension at 72°C for 1 min; and hold at 4°C. DNA clean‐up performed by adding 1.3× AMPure XP beads (Beckman Coulter) to each sample, allowing libraries to incubate with beads for 10 min at room temperature, followed by two washes with 80% ethanol, then elution in 15 μL of 10 mM Tris pH 8.0. While on magnetic stand, supernatant was carefully taken and transferred to a fresh tube. Library size distribution was determined by Agilent TapeStation 4200 (Agilent Technologies) and libraries mixed to yield equal representation before paired end (2 × 50 bp) Illumina sequencing by the NIEHS Epigenomics and DNA Sequencing Core Facility. Reads were trimmed using Cutadapt and aligned to both hg19 and mm10 genomes using Bowtie2 (RRID:SCR_011841). Aligned reads were deduplicated with picardtools. Deduplicated read counts were used to calculate the human:mouse ratio, which was then used to calculate scaling factors for samples performed side by side with the same antibody. The scale factor was calculated by dividing the human:mouse read ratio for each sample using a given antibody by the maximum human:mouse read ratio from samples using the same antibody. Scaled and normalized coverage files were generated using deepTools (RRID:SCR_016366) with calculated scale factors and rpkm normalization. Meta‐profiles and heatmaps were generated using deepTools and ggplot2 (RRID:SCR_014601).

2.10. Protein Analysis

For the immunoblot assessment of signaling activation, supernatants were separated, and cells were washed once with PBS before being lysed in RIPA buffer (25 mM Tris–HCl (pH 7.6), 150 mM NaCl, 1% NP‐40, 1% sodium deoxycholate and 0.1% SDS) supplemented with a protease inhibitor cocktail (Roche, Rotkreuz, Switzerland). Equal amounts of total proteins were diluted in sample loading buffer, loaded, and separated in precast Novex 4%–12% Tris‐Glycine mini gels (ThermoFisher Scientific, Waltham, MA, USA). The proteins were then transferred to nitrocellulose membranes using a semi‐dry rapid transfer system (BioRad) and blocked with blocking buffer (LI‐COR, Lincoln, NE, USA) for 60 min at room temperature. Subsequently, the membranes were incubated overnight at 4°C with primary antibodies against ACOD1/IRG1 (1:1000 dilution), iNOS (1:1000 dilution), NLRP3 (1:1000 dilution), pro‐IL‐1β (1:1000 dilution), GSDMD (1:1000 dilution), DRP‐1 (1:1000 dilution), Mitofusin‐2 (1:1000 dilution), DUSP1/MKP‐1 (1:1000 dilution), SDHA (1:1000 dilution), NRF2 (1:1000 dilution), HIF‐1α (1:1000 dilution) and GR (1:1000 dilution), in 5% skimmed powdered milk in TBS‐T or 5% BSA in TBS‐T, accordingly. Blots were washed and incubated with goat anti‐rabbit IRDye680‐conjugated secondary antibody (RRID:AB_10956166) (LI‐COR) and/or goat anti‐mouse IRDye800‐conjugated secondary antibody (RRID:AB_621842) (LI‐COR) for 1 h at room temperature and visualized with the LICOR Odyssey Imaging scanner system (LI‐COR). The obtained immunoreactivity was normalized to β‐Actin and/or β‐Tubulin proteins as a loading control and was expressed relative to the protein level of the unstimulated condition. For the analysis of protein in western blots from supernatants, to concentrate supernatants, 5 μL Strataclean Resin (Agilent) was added to 500 mL of supernatant and vortexed for 60 s. Supernatants were then centrifuged, and the remaining pellet was resuspended in 30 μL lysis buffer, boiled, loaded and separated in precast Novex 14% Tris‐Glycine mini gels and then transferred to nitrocellulose membranes. The membranes were blocked and subsequently incubated with the mouse‐specific cleaved‐IL‐1β (p17) antibody (1:1000 dilution) overnight at 4°C. The following day, the blots were washed and incubated with goat anti‐rabbit IRDye680‐conjugated secondary antibody for 1 h at room temperature and visualized with LICOR Oddysey Imaging scanner system.

2.11. Immunofluorescence (IF) Staining

Cultures of 2.0 × 10⁵/mL unpolarized BMDM from WT and myeloid‐GRKO were cultivated in glass bottom culture dishes (MartTek corporation). Subsequently, the cells were washed with warm PBS and incubated with prewarmed (37°C) staining solution containing MitoTracker Green FM probe at 50 nM for 45 min. After the staining was completed, the cells were washed with fresh prewarmed buffer and analyzed. A Zeiss laser scanning confocal microscope (LSM 880; Carl Zeiss) was utilized to analyze MitoTracker Green staining and mitochondrial fitness. The samples were analyzed in a blinded manner.

2.12. Determination of Metabolites in Cell Media by Mass Spectrometry

Cultures of 1.0 × 10⁶/mL unpolarized and LPS‐polarized BMDM from WT and myeloid‐GRKO were subjected to 8‐h stimulation with either vehicle or 100 nM of Dex for the determination of itaconate, succinate, fumarate and lactate levels (n = 9–11 biological replicates/condition). Following Dex stimulation, media were collected and immediately processed. Itaconate and lactate content of media was determined by gas chromatography–tandem mass spectrometry as was described previously [31]. Duplicate 100 μL aliquots of collected media were transferred to 1.5 mL centrifuge tubes. Samples were frozen at −80°C and lyophilized to dryness. Dried samples were reconstituted with 50 μL ice cold 70% acetonitrile. After vortex mixing for 10 s, samples were centrifuged at 14 k rpm and 0°C for 3 min. The supernatant was then transferred to a glass centrifuge tube. Samples were then diluted with 100 μL methoxylamine hydrochloride derivatization solution followed by incubation at 60°C for 30 min. Samples were then cooled, frozen at −80°C, and lyophilized to dryness. Samples were again reconstituted in 60 μL 1:1 anhydrous acetonitrile: pyridine and 30 μL MTBSTFA, 1% t‐BDMCS. Samples were held at 60°C for 30 min for derivatization. After samples had returned to room temperature, a portion was transferred to autosampler vials for analysis. Samples were injected in triplicate followed by a solvent blank injection. The levels of metabolites in the supernatant were expressed relative to the internal standard and normalized to the untreated cells.

2.13. Determination of the Nitric Oxide by Griess Reaction

Nitrite levels in the cell supernatants were assessed using the Griess Reagent System (Promega), following the manufacturer's instructions. BMDMs from WT and myeloid GRKO mice were activated with 100 ng/mL LPS for 24 h. After 16 h LPS primming, 100 nM Dex or 500 μM of the specific iNOS inhibitor, 1400W, were added to the cells.

2.14. Determination of Cytotoxicity Levels

Lactate dehydrogenase (LDH) activity in cell supernatants was measured with CytoTOX 96 (Promega), following the manufacturer's instructions. BMDMs from WT and myeloid GRKO mice were seeded in OptiMEM before the NLRP3 inflammasome activation assay. Cell lysates used as control for total LDH content were prepared for each treatment/genotype condition.

2.15. Determination of SDH Activity

Succinate dehydrogenase (SDH) activity was measured in lysates from control BMDMs and LPS‐primed BMDMs that were stimulated for 8 h with Dex upon the NLRP3 inflammasome activation using the buffer provided in the kit according to the manufacturer's instructions for SDH enzymatic activity using colorimetric assays.

2.16. Isolation and Culture of Murine Peritoneal Macrophages

Female and male mice aged 8–12 weeks, including both WT and specific myeloid‐GRKO, were intraperitoneally injected with 3.8% Brewer thioglycollate medium. Four days post‐injection, macrophages elicited with thioglycollate were obtained through peritoneal lavage using 5 mL of PBS. After seeding for 1 h, macrophages became adherent, and unattached cells were washed off. Approximately 1.0 × 10⁶ macrophages were cultured overnight in charcoal‐stripped serum medium supplemented with 10 ng/mL M‐CSF. Subsequently, the isolated peritoneal macrophages were primed with 100 ng/mL LPS, followed by the addition of Dex after 16 h of LPS or as a co‐treatment. Additionally, 100 μg/mL monosodium urate (MSU) crystals were added during the last 3 h of NLRP3 inflammasome activation. Supernatants and lysates were collected and analyzed by western blot and ELISA.

2.17. Monosodium Urate (MSU) Crystal‐Induced Peritonitis for In Vivo Assessment of NLRP3 Inflammasome Activation

Female and male mice aged 8–12 weeks, including both WT and myeloid‐GRKO, were intraperitoneally injected with MSU crystals (30 mg/kg, Invivogen) suspended in PBS for 6 h [18]. Euthanasia was performed by CO₂ asphyxiation, followed by peritoneal lavage using 2.5 mL of sterile PBS. The cells and supernatants in the lavage fluid were centrifuged and separated for analysis. IL‐1β levels in the supernatants were directly analyzed by ELISA and concentrated by incubation with resin (Stratagene) for western blot analysis. The pelleted cells were washed, counted using an automated cell counter (Biorad), and then stained for flow cytometry analysis. After blocking with TruStain FcX antibody (Biolegend), the cells were incubated with 1 μg of PE‐Cy7 CD19 (RRID:AB_2927870), Brilliant Violet (BV)‐510 CD11b (RRID:AB_2561390), APC‐CD115 (RRID:AB_2085222), BV‐421 F4/80 (RRID:AB_10901171), PE‐MHC II (RRID:AB_313726) and APC‐Fire Ly6G (RRID:AB_2616732) antibodies (Biolegend). After a 20‐min incubation on ice, cells were washed, resuspended in staining buffer, and analyzed on a BD LSRFortessa Cell Analyzer (BD Biosciences). Macrophages subsets were gated from the CD115⁺/CD11b⁺ population and classified into large peritoneal macrophages (LPM; F4/80⁺/MHC II^dim) and small peritoneal macrophages (SPM, F4/80^low/MHC II⁺) [32]. The total number of neutrophils obtained from each mouse was calculated by multiplying the frequency of Ly6G+ cells with the total number of cells obtained, adjusted to the volume recovered. The number of macrophages for each population was calculated by determining the percentage of single live cells multiplied by the total number of cells divided by 100.

2.18. Isolation and Differentiation of Human Monocyte‐Derived Macrophages (MoDMs)

The NIEHS Institutional Review Board conducted a thorough review and granted approval for a study involving the collection of blood samples from healthy human subjects (IRB approval #000152). Written consent was obtained from all participants, who were six males aged 35–50 years. Venous blood was collected in 50 mL tubes supplemented with EDTA adjusted to the volume of blood. Plasma and blood cells were separated using RosetteSep Immunodensity cell separation (STEMCELL Technologies) following the manufacturer's protocol. Briefly, the procedure involved incubating whole blood with the RossetteSep antibody cocktail for 10 min, diluting it 1:1 with sterile DPBS‐2 supplemented with 2% FBS, and layering it on top of 15 mL of LymphoPrep (STEMCELL Technologies) in SepMate tubes (STEMCELL Technologies). Gradients were centrifuged at 1200 g for 10 min. The enriched peripheral blood mononuclear cell (PBMC) layer from the density medium‐plasma interface was transferred to a new tube and washed twice with cold DPBS‐2 with 2% FBS. CD14+ monocytes were purified using anti‐human CD14 magnetic beads (Cat# 130‐050‐201, Miltenyi Biotec). These purified monocytes were further differentiated into monocyte‐derived macrophages (MoDMs) by culturing in ImmunoCult‐SF Macrophage medium supplemented with human recombinant M‐CSF (50 ng/mL; Peprotech) for 6 days in a CO₂ incubator supplied with 5% CO₂ in a humidified atmosphere. Every 2 days, half of the medium was exchanged for fresh medium containing M‐CSF. On day 6, cells were non‐enzymatically harvested using the macrophage detachment solution DXF1 (PromoCell) and washed three times with PBS before being re‐seeded in RPMI charcoal stripped medium without M‐CSF. A total of 500 mL media was used for stimulations with LPS (100 ng/mL) plus Dex (100 nM), RU‐486 (1 nM) or RU‐486 plus Dex, and ATP (5 mM) was added during the final 60 min of the assay in 12‐well plates at density 5.0 × 10⁵ cells. The human MoDMs phenotype was analyzed by flow cytometry of the surface markers CD45‐AF700 (RRID:AB_2566373), CD3‐PE (RRID:AB_571912), CD14‐BV785 (RRID:AB_2810577), CD11b‐PE/Cy7 (RRID:AB_2734450) and CD36‐APC (RRID:AB_1279226) (Biolegend).

2.19. Determination of Cytokines by ELISA

Cytokines in the cell supernatants from mouse BMDM, thioglycolate‐elicited macrophages and peritoneal lavage from WT and myeloid GRKO mice, as well as from the supernatants of human MoDM, were analyzed using DuoSet ELISA (mouse IL‐1β, mouse TNF, and human IL‐1β R&D Systems). The analysis was conducted following the manufacturer's protocols. For IL‐1β assays, supernatants from cells were appropriately diluted at a 1:2 ratio.

2.20. Statistical Analysis

The GraphPad Prism version 9.0 (RRID:SCR_002798) was used to analyze the data. To determine the statistical significance of the results, the Two‐tailed unpaired Student's t‐test and One‐ or Two‐way ANOVA statistical test was performed with the ad/hoc post‐test according to the distribution of the data. For data containing 2 groups, significant differences between means were analyzed by the 2‐tailed, unpaired, Student's t‐test. For data containing more than 2 groups, significant differences between means were analyzed by 2‐way ANOVA test with correction for multiple comparisons (Tukey, Benferroni or Sydak, accordingly). Those comparisons whose value was p < 0.05 were considered statistically significant. In all the experiments the samples were analyzed in duplicate (considering the average of two values), values were expressed as the mean ± SEM from at least 3 independent experiments, unless otherwise indicated in figure legends. Statistical analysis of the microarray, RNA‐seq and CUT&Tag data sets are detailed in each section analysis.

3. Results

3.1. GR Repression of Pro‐Inflammatory Genes in Macrophages Is Modulated by LPS Induced Alterations in the Chromatin Landscape

The presence of an intricate network of gene regulation is a hallmark of macrophage polarization. The polarization process refers to the molecular mechanism by which macrophages produce distinct functional phenotypes as a reaction to specific microenvironmental signals. The polarization toward a pro‐inflammatory profile has been associated with the most dramatic changes in the transcriptome of macrophages, many of which are known to be under the influence of glucocorticoids [33]. Therefore, we evaluated the global transcriptome regulated by a 4‐h treatment with the synthetic glucocorticoid dexamethasone (Dex) through RNA‐sequencing (RNA‐seq) in both unpolarized and LPS‐primed pro‐inflammatory mouse macrophages derived from bone marrow (BMDM). In LPS‐primed WT macrophages, 404 genes were significantly repressed by Dex treatment, including the inflammasome‐associated genes Nos2, Nlrp3, Il1b, Hif1a, Acod1/Irg1 and P2ry2 (Figure 1A). Ingenuity pathway analysis of the 404 DEGs showed that Dex treatment of LPS‐primed WT macrophages inhibited many signaling molecules and pathways associated with inflammation, such as TLR4, MYD88, IFNG and RELA (Figure S1A). Since GR is a ligand‐activated nuclear receptor, it translocates from the cytoplasm to the nucleus within minutes upon activation to regulate gene expression. However, some reports have shown that Dex does not exert detectable effects within 4 h of treatment in pathways associated with anti‐inflammatory responses when co‐administered LPS [16].

FIGURE 1.

GR actions over the repression of pro‐inflammatory genes requires a more accessible chromatin landscape in response to glucocorticoids in macrophages. (A) Volcano plot from RNA‐seq data in which each dot in red represents the differentially expressed genes (DEG) by Dex in LPS‐primed macrophages versus LPS‐treated with adjusted p < 0.05. The circle highlights a group of inflammasome‐associated genes. Schematic representation of the experimental setup for Dex treatment after LPS‐priming (B) and LPS plus Dex as a co‐treatment (co‐tx) (C). (D) Assessment of mRNA levels of Nlrp3, Nos2, Acod1/Irg1, Il1b, Hif1a, P2ry2, Dusp1, and Tsc22d3/Gilz genes in BMDM from Vehicle, Dex, LPS, Dex after LPS‐priming (after), and LPS plus Dex as a co‐treatment (co‐tx). The relative mRNA expression levels of the target genes were normalized to Ppib. (E) Immunoblot and densitometry analysis representing the levels of NLRP3, iNOS, ACOD1, and Pro‐IL1B in cell lysates from BMDMs, comparing the stimulation by 100 nM Dex at 6, 12, and 24 h in LPS‐primed macrophages (LPS → Dex) and as a cotreatment with LPS (LPS + Dex) as depicted in (B) and (C). (F) Meta‐profiles of GR signals over active TSSs and heatmaps depicting relative changes in Cut&Tag signal for GR enrichment in LPS, LPS plus Dex, and Dex after LPS treatments. Venn diagram illustrating the numbers of GR peaks in LPS plus Dex (LPS + Dex) and Dex after LPS (LPS → Dex) regimens. (G) Browser image of Cut&Tag signal for GR peaks over Per1, Dusp1, Ccl3, Ccl4, and Nlrp3 genes as induced and reduced by Dex treatment. (H) Frequency distribution graph for motifs enrichment analysis. The x‐axis is the ratio of % GR peaks over % background regions, and the y‐axis represents the −log10 of the p value. NHR: Nuclear hormone receptor. Data are mean ± SEM. Blots are representative of a minimum of 3 independent experiments. Statistical analysis was performed using one‐way ANOVA with Tukey's multiple‐comparison test (A) and (B). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

To investigate the temporal sensitivity of these inflammatory genes to glucocorticoids, we examined their regulation in WT macrophages under two different experimental paradigms: Dex added for 4 h after a 16‐h LPS priming treatment (Figure 1B) and Dex added as a cotreatment with LPS for 4 h (Figure 1C). Quantitative real‐time RT‐PCR demonstrated repression of the mRNA levels of the inflammasome‐associated genes Nos2, Acod1/Irg1, Il1b, Nlrp3, Hif1a, P2ry2 by glucocorticoids when administered after LPS priming (Figure 1D). In contrast, when administered at the same time as LPS, glucocorticoid treatment had no effect on Nos2, Acod1/Irg1, Nlrp3, Hif1a, and P2ry2 expression and only promoted a small reduction in Il1b gene expression. Interestingly, upregulated genes such as Tsc22d3 (Gilz) and Dusp1 both were induced in a similar magnitude between both paradigms (Figure 1D). Similar data were found at the protein level, as Dex treatment after LPS priming resulted in a greater reduction in NLRP3, iNOS, ACOD1, and pro‐IL‐1β expression compared to the cotreatment condition (Figure 1E). These data indicate that the inhibitory actions of GR on pro‐inflammatory gene expression in macrophages can be modulated by the inflammatory stimulus to achieve a more widespread and robust repression.

To investigate the mechanism by which a pro‐inflammatory stimulus exposure can alter the activity of GR on target gene expression, we performed Cleavage Under Targets and Tagmentation (CUT&Tag) to evaluate the interaction of GR with chromatin in macrophages treated with Dex after LPS priming or with Dex and LPS added together. The Dex treatment after LPS priming resulted in a striking increase in the interaction of GR with chromatin compared to the cotreatment condition, with over 3000 statistically significant GR peaks observed exclusively when Dex treatment followed LPS (Figure 1F). For downstream analysis, we compared the localization and the intensity of the GR peaks in proximity to genes regulated by Dex treatment. For Per1 and Dusp1, two genes well known to be increased by glucocorticoids in macrophages, GR enrichment was observed at multiple sites downstream of the transcription start site (TSS) in response to either Dex treatment paradigm (Figure 1G). In contrast, at the Ttll5/Tgfb locus, which encodes two other genes induced by glucocorticoids, GR enrichment was observed predominantly when Dex treatment occurred after LPS priming (Figure S2A). At the Ccl3/Ccl4 locus, which encodes 2 genes well known to be decreased by glucocorticoids in macrophages, GR enrichment downstream of the TSS was more pronounced when Dex treatment occurred after LPS priming than when it was co‐administered with LPS (Figure 1G). Among the inflammasome associated genes, we observed Dex‐dependent GR enrichment in proximity to the locus of Nlrp3, Nos2, Il1b, and Acod1. For the Nos2, Il1b, and Acod1 genes, the extent of GR enrichment was similar in both Dex treatment conditions (Figure S2A). However, for the Nlrp3 gene, two GR peaks were identified in close proximity to the TSS and the more downstream site showed greater GR enrichment in the Dex after LPS priming compared to the cotreatment condition (Figure 1G). To further explore the genomic interactions of GR and the timing of Dex treatment, we examined the enrichment of motifs in the collection of called peaks for each treatment regimen. Nuclear Hormone Receptor (NHR) motifs, including the glucocorticoid responsive element (GRE), were among the top 5 enriched motifs in the GR peaks from both the Dex after LPS priming and the co‐treatment condition (Figure S2B). Interestingly, relative to NHR motifs, AP‐1 and PU.1/ETS were more strongly enriched in the Dex after LPS priming GR peaks than in the co‐treatment peaks (Figure 1H). These data suggests that LPS priming of macrophages opens the chromatin landscape to permit more GR binding directly to DNA and/or to other DNA bound transcription factors to expand the glucocorticoid transcriptome.

3.2. Glucocorticoids Reduce Expression of iNOS and ACOD1 in Macrophages to Preserve the Integrity of the TCA Cycle and Rescue the Glycolytic Switch Induced by LPS

To investigate whether these time‐dependent differences in glucocorticoid regulation of gene expression were dependent on GR, we utilized a mouse model with conditional knockout of GR in myeloid cells (myeGRKO) to generate GR‐deficient macrophages (GRKO). Analysis of GRKO derived from bone marrow cells showed a substantial reduction of over 80% in GR mRNA and protein levels compared to WT (GR^fl/fl) macrophages (Figure 2A,B). To evaluate at a global level whether the glucocorticoid mediated inhibition of the pro‐inflammatory signature in macrophages is dependent on the GR signaling pathway, we performed RNA‐seq on untreated and LPS‐primed BMDM from WT and myeGRKO mice that were treated with vehicle or Dex for 4 h. When we compared the transcriptome at baseline, very few differences were observed between WT and GRKO macrophages (Figure 2C). However, when we compared the transcriptome in treated cells, differences were observed. Of note, most of the inflammasome‐associated genes were insensitive to Dex in GRKO macrophages and most of the pro‐inflammatory pathways were predicted to be unaffected by Dex treatment in LPS‐primed GR‐deficient macrophages demonstrating the requirement for GR for their regulation by glucocorticoids (Figures 2D and S1B).

FIGURE 2.

Glucocorticoids keep the integrity of the TCA cycle and rescue the glycolytic metabolic switch induced by LPS reducing the expression of the enzymes iNOS and ACOD1 in macrophages. (A) Assessment of mRNA levels of GR in WT and GR‐deficient macrophages (GRKO) (n = 12). (B) Immunoblot analysis depicting the levels of GR in WT and GRKO macrophages. Densitometry values of the western blot bands quantified for protein content, normalized to β‐actin (lower part; n = 6). (C) Volcano plot from RNA‐seq data in which each dot in red represents the differentially expressed genes (DEG) at the baseline in pro‐inflammatory macrophages from WT and GRKO. (D) Volcano plot from RNA‐seq data in which each dot in red represents the differentially expressed genes (DEG) by Dex in LPS‐primed macrophages versus LPS‐treated from GRKO versus WT macrophages with adjusted p < 0.05. (E) Acod1/Irg1 mRNA levels in vehicle and response to Dex, LPS and Dex after LPS‐primed treatments in both WT and GRKO (n = 3). (F) Immunoblot analysis depicting the ACOD1 regulation by Dex in cell lysates from WT and GR‐deficient BMDM. Both 100 nM Dex over LPS‐primed macrophages (LPS → Dex) and Dex as a cotreatment with LPS (LPS + Dex) were analyzed. (G) Representative histogram displaying itaconate levels detected by mass spectrometry in the supernatant of unpolarized and LPS‐polarized BMDM from WT and GRKO. (H) Relative itaconate levels in the supernatant of BMDMs from WT and GRKO following treatment with LPS (16 h) and adding 100 nM Dex for 8 h (n = 9). (I) Nos2 mRNA levels in vehicle and response to Dex, LPS and Dex after LPS‐primed treatments in both WT and GRKO (n = 3). (J) Nitrite levels, indicative of Nitric Oxide production, measured by Griess reaction in the supernatant of BMDMs from WT and GRKO macrophages treated as depicted in (H) (n = 9). (K) Relative succinate, fumarate and lactate levels in the supernatant of BMDMs from WT and GRKO as depicted in (H). (L) Curves of GSEA enrichment scores for Reactome Glucose Metabolism in WT macrophages Dex after LPS‐primed over LPS, and Reactome Pyruvate Metabolism (right panel) in the comparison GRKO Dex after LPS‐primed over LPS against WT LPS‐primed over LPS. Assessment of metabolic flux OCR (M) and ECAR (N) in WT and GRKO macrophages analyzed by Seahorse XF Mito stress test and XF Glycolysis test flux analyzer. BMDMs were primed with vehicle or LPS for 16 h, followed by treatment with vehicle or 100 nM Dex for 8 h. (O) Representative immunofluorescence staining of MitoTracker Green (green) in unpolarized WT and GRKO BMDMs. (P) Levels of MitoTracker Green measured by flow cytometry in BMDMs from WT and GRKO following priming with LPS and then stimulated with 100 nM Dex for 8 h (n = 4). (Q) Levels of mitochondrial ROS production measured by the percentage of MitoSox using flow cytometry and treated as depicted in (P) (n = 4). (R) Assessment of mRNA levels of Mnf2 and Drp1/Dnm1l genes in WT and GRKO BMDM from vehicle, Dex, LPS only and 4 h Dex after LPS‐priming (LPS → Dex) (n = 4). (S) Immunoblot analysis depicting the levels of Mitofusin‐2 and DRP‐1 in cell lysates from WT and GRKO BMDMs Vehicle, Dex, LPS and 8 h Dex after LPS‐priming (LPS → Dex) (n = 3). The relative mRNA expression levels of the target genes were normalized to Ppib. Immunoblot data were normalized to protein content per well (n = 3). Blots are representative of a minimum of 3 independent experiments. Data are mean ± SEM. Statistical analysis was performed using two‐tailed unpaired t‐tests (A) and (B); 2‐way ANOVA with Sidak's multiple‐comparison test (E), (F), (H), (I), (J), (K), (M), (N), (P), (Q), (R) and (S). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Since both iNOS and ACOD1 and their respective metabolites itaconate and nitric oxide have emerged as key players linking macrophage metabolism and the inhibition of the NLRP3‐inflammasome activation [17], we investigated this association and its regulation by glucocorticoids further using the GR‐deficient macrophages. Dex treatment after LPS priming led to a decrease in Acod1/Irg1 mRNA levels in WT pro‐inflammatory macrophages (Figure 2E). Acod1/Irg1 mRNA levels were also decreased by Dex in unpolarized macrophages (Figure 2E). These glucocorticoid mediated effects were abolished in the GR‐deficient macrophages (Figure 2E). At the protein level, WT macrophages exposed to Dex after LPS priming but not coincident with LPS exhibited a reduction in ACOD1 (Figure 2F), and this effect was blocked in the GR deficient macrophages. We also measured itaconate, the main product of ACOD1 enzymatic activity. Interestingly, the GRKO macrophages exhibited higher levels of itaconate at baseline and upon activation compared to WT macrophages suggesting a deregulation of itaconate production with the loss of GR signaling (Figure 2G). The reduction of ACOD1 mRNA and protein by Dex in LPS‐primed WT pro‐inflammatory macrophages led to a significant decrease in itaconate levels in the supernatant (Figure 2H), and this reduction was not observed in GR‐deficient macrophages. Similarly, the decrease in Nos2 mRNA levels (Figure 2I) and the repression of iNOS by glucocorticoids after LPS priming resulted in lower nitric oxide levels (NO) in WT macrophages but not in the GR‐deficient macrophages (Figure 2J). These data demonstrate a critical role for GR in regulating the enzymes ACOD1 and iNOS and their respective metabolic products.

To determine whether the glucocorticoid‐dependent decrease in ACOD1 expression and itaconate production affected TCA cycle activity, we measured the levels of the enzyme succinate dehydrogenase (SDH) that catalyzes the mitochondrial conversion of succinate to fumarate and is targeted for inhibition by itaconate. LPS treatment strongly reduced SDH enzymatic activity in WT macrophages (Figure S3A). Dex treatment of LPS‐primed cells partially reversed this reduction, leading to a corresponding decrease in succinate levels and an increase in fumarate and lactate (Figure 2K). These glucocorticoid‐mediated effects were absent in GR‐deficient macrophages (Figures 2K and S3A).

Gene set enrichment analysis (GSEA) of the genes upregulated by Dex in the LPS‐primed WT macrophages through RNA‐seq also revealed a strong enrichment for Glucose Metabolism (Reactome_Glucose_Metabolism; Figure 2L). Conversely, genes upregulated in the GRKO macrophages were enriched for pathways involved in Pyruvate Metabolism (Reactome_Pyruvate_Metabolism). These transcriptomic changes suggest that GR, in addition to controlling immune cell activation, plays a critical role in regulating macrophage metabolism. To determine directly whether the glucocorticoid‐mediated changes in the expression of metabolic genes alter the metabolism of LPS‐primed macrophages, we performed an extracellular metabolic flux analysis in WT and GR deficient macrophages (Figures 2M,N and S3B,C). Dex‐treated LPS‐primed WT macrophages showed a reversion in oxygen consumption rate (OCR) given by the measurement of the mitochondrial maximal respiration and a decrease in the extracellular acidification rate (ECAR) given by the measurement of basal glycolysis compared to WT macrophages treated only with LPS. Conversely, GRKO macrophages showed an altered LPS‐induced glycolytic switch that could not be rescued by Dex. These findings are consistent with the gene enrichment predictions and reveal that both mitochondrial respiration and aerobic glycolysis are modulated by glucocorticoids in macrophages during pro‐inflammatory activation induced by LPS to preserve the integrity of the TCA cycle.

3.3. Glucocorticoids Rescue Mitochondrial Dysfunction Through the Regulation of Mitochondrial‐Shaping Proteins in LPS‐Primed Macrophages

Since glucocorticoid treatment reversed the onset of aerobic glycolysis and sustained the integrity of the TCA cycle in LPS primed macrophages, we evaluated mitochondria functionality. Reactome pathway analysis and GSEA of genes regulated by Dex in LPS‐primed WT pro‐inflammatory macrophages from the RNA‐seq data identified mitophagy (the clearing process of mitochondria through autophagy) as a significant downregulated pathway (Figure S4A). Conversely, GRKO pro‐inflammatory macrophages showed an opposite profile with mitophagy identified as a significant activated pathway (Figure S4B). Consistent with these gene enrichment predictions, GR‐deficient macrophages exhibited a dysfunctional mitochondria phenotype under steady‐state conditions, as evidenced by mitochondria fragmentation and increased mitochondrial mass (Figures 2O and S4C). Dex treatment decreased LPS‐induced mitochondria fragmentation and ROS production in WT macrophages but not in the GRKO macrophages (Figure 2P,Q). Since mitochondrial morphology, dynamics, and mitophagy are determined by the equilibrium between mitochondrial fusion and fission events [34], we evaluated the ability of glucocorticoids to regulate Mitofusin‐2 (MFN‐2) and DRP‐1, markers of mitochondria fusion and fission, respectively. Dex treatment of LPS‐primed WT macrophages enhanced MFN‐2 and decreased DRP‐1 mRNA and protein expression compared to LPS treatment alone, and these effects were abolished in GR deficient macrophages (Figure 2R,S). These data indicate that the metabolic effects of glucocorticoids on glucose metabolism and the TCA cycle in LPS primed macrophages are accompanied by improved mitochondrial dynamics through the GR dependent reduction in mitochondrial fission and fragmentation by DRP‐1 regulation and the rescue of mitochondrial fusion by Mitofusin‐2 regulation.

3.4. Glucocorticoids Limit Inflammation and Pyroptosis by Regulating the Late NLRP3‐Inflammasome Activation

Given the physiological and clinical relevance of glucocorticoids in established inflammatory processes and our finding that the proinflammatory stimulus and glucocorticoid exposure pattern can have an impact on GR recruitment to chromatin, we evaluated the effects of Dex under two distinct NLRP3 inflammasome activation conditions: the conventional inflammasome activation protocol (LPS 3 h) involving Dex treatment after 1 h of LPS priming and the late inflammasome activation protocol (LPS 24 h) involving Dex treatment after 16 h of LPS priming (Figure 3A). The effects of Dex were assessed by examining its ability to inhibit IL‐1β cleavage and subsequent secretion into the supernatant induced by ATP or Nigericin as a signal 2 during NLRP3 inflammasome activation in the presence or absence of the specific NLRP3 inhibitor MCC950 (Figure 3A). Analysis of the mature IL‐1β form (p17) in the supernatant via ELISA and western blot revealed that Dex treatment did not alter IL‐1β secretion induced by ATP or Nigericin in LPS‐primed WT macrophages or GRKO macrophages under conventional NLRP3 inflammasome activation (Figures 3B and S5). However, under these same conditions, MCC950 effectively blocked IL‐1β secretion induced by ATP or Nigericin in both LPS‐primed WT and GR‐deficient macrophages. In marked contrast to our findings with the conventional protocol, Dex treatment decreased IL‐1β secretion induced by ATP (72% ± 25.19% of reduction) or Nigericin (68% ± 24.13% of reduction) in LPS‐primed WT macrophages under late NLRP3 inflammasome activation (Figure 3C). This suppressive effect by Dex was comparable to that seen with MCC950 in presence of ATP (73% ± 25.65% of reduction) or Nigericin (71% ± 24.76% of reduction) and was not evident in GR‐deficient macrophages (Figure 3C), indicating GR signaling is required for limiting NLRP3 inflammasome activation under prolonged LPS stimulation in macrophages.

FIGURE 3.

Glucocorticoids modulate inflammation and pyroptosis by regulating late NLRP3‐inflammasome activation through ACOD1 and iNOS. (A) Schematic representation of the experimental setup for conventional NLRP3‐inflammasome activation (upper), and late NLRP3‐inflammasome activation regulated by Dex and MCC950 (lower). (B) and (C) Quantification of IL‐1β concentration determined by ELISA in the supernatant of WT and GRKO BDMDs stimulated with LPS for 3 h (conventional) and 24 h (late), with Dex and vehicle added during the final 2 and 8 h of treatment, followed by ATP (5 mM) or Nigericin (10 μM) (n = 3). (D) Immunoblot analysis depicting Dex effects on the protein levels of ACOD1, iNOS, NLRP3, pro‐IL‐1β, Caspase‐1 and GSDM‐D during late LPS‐induced NLRP3‐inflammasome activation and using ATP as signal 2 in WT and GRKO BMDMs lysates, detected with specific antibodies using western blotting. (E) Densitometry values of the immunoreactive bands quantified for protein content, normalized to β‐actin as depicted in (D) (n = 4). (F) Nitrite levels in the supernatants of WT and GRKO BMDMs as depicted in (D) determined by colorimetric assay (n = 4). (G) Assessment of lactate dehydrogenase (LDH) activity in the supernatants of WT and GRKO BMDMs as depicted in (D) (n = 4). (H) Immunoblot analysis depicting the protein levels of NRF2 as depicted in (D) in presence of the proteasome inhibitor MG‐132 (2 μM) 6 h before harvesting cells (n = 3). (I) Immunoblot analysis depicting the protein levels of HIF‐1α as depicted in (D) in presence of the prolyl‐hydroxylase inhibitor Roxadustat (RXD; 10 μM) 2 h before adding Dex (n = 3). (J) Quantification of IL‐1β concentration determined by ELISA in the supernatant of WT BDMDs as depicted in (H) and (I) (n = 3). (K) Assessment of lactate dehydrogenase (LDH) activity in supernatants of WT BMDMs subjected to the conditions described in (H) and (I) (n = 3). Data represented as mean ± SEM. Blots shown are representative of a minimum of 3 independent experiments. Statistical analysis was performed using 2‐way ANOVA with Sidak's multiple‐comparison test (B–I); one‐way ANOVA with Tukey's multiple‐comparison test (J) and (K). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Several studies have demonstrated the inhibitory effects of itaconate derivatives, such as 4‐octyl itaconate (4‐OI) and dimethyl itaconate, on NLRP3 inflammasome activation in macrophages [17, 18, 19, 35]. Additionally, a synergistic relationship between ACOD1 and iNOS has been identified, highlighting the importance of itaconate and nitric oxide as critical co‐regulators of NLRP3‐inflammasome activation during sustained LPS priming and tolerized inflammasome states [17]. Based on these findings, we examined glucocorticoid regulation of ACOD1 and iNOS under late NLRP3 inflammasome activation. Dex treatment decreased the expression of both ACOD1 and iNOS in WT macrophages but not GRKO macrophages (Figure 3D,E). In addition, the levels of NLRP3, Caspase‐1, pro‐IL‐1β, DRP‐1, and nitric oxide were each decreased by Dex in WT macrophages but not the GR deficient macrophages under late NLRP3 inflammasome activation (Figure 3D–F). In contrast, under conditions of conventional NLRP3 inflammasome activation, glucocorticoid treatment had minimal effects on ACOD1, iNOS, NLRP3, Mitofusin‐2, and DRP‐1 (Figure S6A,B). We also examined GSDMD cleavage (GSDMD‐NT) as the main executioner of pyroptosis (Figure 3D,E) and lactate dehydrogenase (LDH) activity (Figure 3G) in supernatants as an indicator of cell death induced by ATP since the observed reduction in secreted IL‐1β under late NLRP3 inflammasome activation could result from the inhibitory effects of Dex on pyroptosis. Dex treatment in the presence of ATP during the late NLRP3 inflammasome activation protocol reduced GSDMD cleavage in WT macrophages by 59.81% ± 17.2% but not GR‐deficient macrophages (Figure 3D,E). Under these conditions, Dex also attenuated LDH release in the presence of ATP by 54% ± 12.7% in WT macrophages, a protective effect that was not observed in GRKO macrophages (Figure 3G). During the conventional NLRP3 inflammasome activation in WT and GRKO macrophages, we did not observe significant Dex effects on GSDMD‐NT and LDH release (Figure S6B,D,E). Similarly, Dex showed inhibitory effects on GSDMD‐NT levels and LDH release upon Nigericin treatment, another potent pyroptosis inducer (Figure S7A,B), suggesting that a gradual establishment of the pro‐inflammatory phenotype is necessary for GR actions on metabolic target genes to reduce their expression over time. Collectively, these data demonstrate that glucocorticoids, in addition to reducing the expression of Nlrp3, Nos2, Acod1 and Il1b during the priming step of NLRP3 inflammasome activation, also mitigate events associated with ATP activation (signal 2) during late NLRP3 inflammasome activation and pyroptosis.

The increase in iNOS and ACOD1 and their products nitric oxide and itaconate by LPS can lead to the activation of hypoxia‐inducible factor 1‐α (HIF‐1α) [36] and the oxidative stress by the nuclear factor erythroid 2‐related factor 2 (NRF2) [17], and these two signaling pathways can modulate activation of the NLRP3 inflammasome. Therefore, we investigated whether Dex treatment in the late NLRP3 inflammasome activation protocol altered the levels of HIF‐1α and NRF2 in WT and GR‐deficient macrophages. For detection of these proteins, we utilized the proteosome inhibitor MG‐132 (NRF2 stabilizer) and the prolyl‐hydroxylase inhibitor Roxadustat (RXD) (HIF‐1α stabilizer). Dex treatment of WT macrophages during the late NLRP3 inflammasome activation reduced the levels of NRF2 and HIF‐1α in a GR‐dependent manner (Figure 3H,I). Importantly, neither MG‐132 nor RXD affected the Dex mediated inhibition of late NLRP3 inflammasome activation in WT macrophages as assessed by the levels of pro‐IL‐1β (Figure S8A–C), secretion of IL‐1β induced by ATP (Figure 3J), and release of LDH induced by ATP (Figure 3K). Thus, inhibition of late NLRP3 inflammasome activation by Dex may involve glucocorticoid mediated reductions in nitric oxide and itaconate that converge to decrease signaling of NRF2 and HIF‐1α.

We next compared the ability of glucocorticoids to inhibit activation of the NLRP3 inflammasome with well‐known NLRP3‐inflammasome inhibitors such as MCC950, the caspase‐1 inhibitor Y‐VAD, the cell permeable derivative of itaconate 4‐OI, and the iNOS inhibitor 1400 W, in WT and GR‐deficient macrophages under the same proinflammatory setting of late NLRP3‐inflammasome activation (Figure S9A–K). Dex effectively reduced IL‐1β release (59% ± 12.12%), GSDMD processing (56% ± 12.7%), and LDH release (58% ± 12.12%) in the presence of ATP in WT but not GRKO macrophages, and this inhibition was comparable to that measured for MCC950 (92% ± 2.3%, 74% ± 7.5% and 69% ± 8.94%, respectively), Y‐VAD (82% ± 3.46%, 81% ± 5.48% and 56% ± 12.7%, respectively) and 4‐OI (84% ± 6.53%, 77% ± 9.38% and 71% ± 11.83%, respectively) in WT macrophages (Figure S9H–J). These results indicate that glucocorticoids also modulate NLRP3 inflammasome activation at the level of caspase‐1 to provide further protection against overactivation of the signal 2. In addition, among these NLRP3 inhibitors, only Dex was able to inhibit the expression of pro‐IL‐1β (Figure S9F). Notably, the iNOS inhibitor 1400 W was the only agent that failed to decrease the levels of NLRP3 inflammasome activation effectors upon ATP stimulation in WT macrophages (Figure S9H,I). In fact, GSDMD‐NT levels were significantly increased in the 1400 W‐treated macrophages indicative of a more pro‐inflammatory phenotype (Figure S9H). This data indicates that a reduction in nitric oxide alone is not sufficient to inhibit late NLRP3 inflammasome activation.

3.5. Clinically Relevant Course of Glucocorticoid Plays a Role in Suppressing Sustained NLRP3‐Inflammasome Activation

Endogenous glucocorticoids play a role regulating the body's homeostatic response to stress. They effectively regulate the immune response to suppress potentially damaging inflammatory responses and avoid unwarranted autoimmunity and cellular damage. However, episodes of immune‐overactivation induced by infections and/or trauma disrupt the homeostatic response and endogenous glucocorticoids are not enough to cope with this disturbance. Therapeutic glucocorticoids are well established drugs to treat a myriad of inflammatory conditions and help restore homeostasis. To understand how glucocorticoids control late NLRP3 inflammasome activation in macrophages, we compared the actions of Dex administered as a cotreatment with LPS for 24 h with the actions of Dex administered for 8 h after 16 h of LPS priming. The latter condition more closely mimics the clinical situation in patients treated with glucocorticoids to combat existing inflammation. At the mRNA levels, Dex treatment in LPS‐primed macrophages reduced the levels of Nos2, Acod1, Il1b, Nlrp3, Hif1a, Nfr2, P2ry2 and Drp1 and increased Mfn2 and Dusp1 (Figure 4A). In contrast, Dex as a cotreatment with LPS only reduced the levels of Nos2, Il1b, Hif1a and P2ry2 and increased Dusp1 (Figure 4A), revealing a differential regulation of target gene expression by Dex depending on the inflammatory context. Similar results were found at the protein level, as the effects of Dex observed on ACOD1, NLRP3, DRP‐1, MFN‐2, and GSDMD‐NT when administered after LPS priming were blunted or absent when administered as cotreatment with LPS (Figure 4B). Importantly, Dex increased the levels of DUSP1 under both treatment protocols confirming that the direct effects of GR activation on some target genes may be independent of the chromatin landscape changes induced by the inflammatory context (Figures 1G and 4B). In regard to signal 2 and NLRP3 inflammasome activation, Dex after LPS attenuated the secretion of IL‐1β, GSDMD processing, and LDH release induced upon ATP whereas the Dex and LPS cotreatment had no significant effect on these events (Figure 4B–D). In addition, Dex inhibition of nitric oxide production occurred when administered after LPS but not as cotreatment with LPS (Figure 4E). Together, these experiments demonstrate that the timing of glucocorticoid treatment in relationship to the inflammatory stimulus has a major impact on the macrophage fate through the suppression of the inflammatory response and the prolonged NLRP3 inflammasome activation.

FIGURE 4.

Clinically relevant course of glucocorticoid plays a role in suppressing sustained NLRP3‐Inflammasome activation. (A) Assessment of mRNA levels of Nos2, Acod1/Irg1, Il1b, Nlrp3, Hif1a, Nrf2, P2ry2, Drp1, Mfn2, and Dusp1 genes during late LPS‐induced NLRP3‐inflammasome activation and using ATP as signal 2 in WT BDMDs using Dex after LPS‐priming (blue) and LPS plus Dex as a co‐treatment (red). (B) Immunoblot analysis depicting Dex effects when is added after LPS priming (LPS → Dex) or as a co‐treatment (LPS + Dex) on the protein levels of ACOD1, iNOS, NLRP3, DRP‐1, MFN2, NT‐GSDMD, pro‐IL‐1β and DUSP1/MKP1 during the late LPS‐induced NLRP3‐inflammasome activation post ATP. (C) Quantification of IL‐1β determined by ELISA in the supernatant of BMDMs as depicted in (B). (D) Assessment of lactate dehydrogenase (LDH) activity in the supernatants of BMDMs as depicted in (B) (n = 3). (E) Nitrite levels determined by colorimetric assay in the supernatants of BMDMs as depicted in (B) (n = 4). Data represented as mean ± SEM. Blots shown are representative of a minimum of 3 independent experiments. Statistical analysis was performed using one‐way ANOVA with Tukey's multiple‐comparison test for (A) and (C–E) and 2‐way ANOVA with Sidak's multiple‐comparison test (B). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

3.6. Depletion of GR in Myeloid Cells Increases the Inflammasome Activation Upon MSU Crystals and Alters the Small Peritoneal Macrophage Subsets

The influence of GR signaling pathway on NLRP3 inflammasome activation was further investigated in both ex vivo and in vivo models using the inflammasome activator monosodium urate (MSU) crystals. We first evaluated the role of GR signaling ex vivo in thioglycolate‐elicited peritoneal macrophages isolated from both WT and myeloid GRKO mice in the same late NLRP3‐inflammasome activation setting (Figure S10A,B). Dex was added to the macrophages after 16 h of LPS priming or as a co‐treatment with LPS for 24 h, followed by MSU crystals during the last 3 h as a signal 2 (Figure S10A). Consistent with our findings in BMDM, Dex effectively reduced the levels of pro‐IL‐1β and blocked IL‐1β secretion under both LPS priming and cotreatment paradigms in WT peritoneal macrophages (Figure S10C,D). The MSU crystals alone induced GSDMD processing (Figure S10C); however, the levels of GSDMD processing when Dex was administered after LPS priming were less than the levels observed under the LPS and Dex co‐treatment condition (Figure S10C). These glucocorticoid mediated effects were absent in peritoneal macrophages isolated from myeloid GRKO mice (Figure S10C,D).

To complement and consolidate the major findings from the in vitro BMDM and ex vivo thioglycolate‐elicited peritoneal macrophages, we assessed inflammasome activation through intraperitoneal injections of MSU crystals as a model of peritonitis in vivo, in both WT and myeloid GRKO mice (Figure 5A). After 6 h of MSU treatment, we observed elevated levels of IL‐1β by ELISA and western blot in the peritoneal lavage fluid from myeloid GRKO compared to WT littermates (Figures 5B and S10E). Additionally, myeloid GRKO mice exhibited a greater number of infiltrating neutrophils and a significantly higher population of infiltrating small peritoneal macrophages (SPM subset) characterized by F4/80^low/MHC‐II^high (Figures 5C,D, and S10F). Conversely, there was no significant difference in the subset F4/80^high/MHC‐II^dim associated with peritoneal resident macrophages (large peritoneal macrophages) between myeloid GRKO and WT littermate post‐MSU treatment (Figures 5E and S10F), suggesting that genetic deletion of GR in myeloid cells results in an enhanced pro‐inflammatory response, exacerbating inflammation and inflammatory cell infiltration in MSU‐induced peritonitis.

FIGURE 5.

The loss of Glucocorticoid Receptor (GR) in murine myeloid cells and the block of GR signaling in human monocytes‐derived macrophages exacerbates the NLRP3 inflammasome activation. (A) Schematic representation of the experimental setup for in vivo peritonitis model using MSU crystal. Peritoneal lavage fluid was collected 6 h post‐ intraperitoneal injection of 30 mg/kg MSU crystals in WT and myeloid‐GRKO mice. Peritoneal lavage involved the injection of 2.5 mL PBS and subsequent retrieval from the peritoneal cavity for supernatant harvesting and immune cells isolation. (B) Evaluation of IL‐1β concentration in peritoneal lavage. (C) Determination of absolute numbers of infiltrating neutrophils in peritoneal lavage, as was described in (A), calculated based on cell count per mL and the total recovered volume. (D) Relative percentage of small peritoneal macrophages (SPM) and (E) large peritoneal macrophages (LPM) expressing differentially the markers F4/80 and MHC‐II. These cell populations were gated from CD11b⁺/CD115⁺ cells. Data represented as mean ± SEM (n = 9, from B to E, where each dot representing one mouse, encompassing both female and male). (F) Schematic representation of the experimental setup for late NLRP3‐inflammasome activation in human monocytes‐derived macrophages (MDMs) modulated by Dex in the presence of the GR antagonist RU‐486. (G) Quantification of NLRP3 normalized to β‐actin in lysates from human MDMs stimulated with LPS for 24 h, with the addition of the antagonist RU‐486 and Dex during the final 8 h, followed by ATP (5 mM). (H) Quantification of ACOD1 protein as depicted in (G). (I) Quantification of pro‐IL‐1β protein as depicted in (G). (J) Assessment of IL‐1β secretion in human MDMs stimulated as described in (G), determined by ELISA (N = 6). Data representative of 4–6 healthy donors. Statistical analysis was performed using two‐tailed unpaired t‐tests (B–E) and one‐way ANOVA with Tukey's multiple‐comparison test (G–J) *p < 0.05; **p < 0.01.

3.7. The Relevant Course of Activation of the GR Signaling Reduces ACOD1 and Inhibits NLRP3‐Inflammasome Activation in Human Monocyte Derived‐Macrophages

Given the association between NLPR3 inflammasome activation and inflammation triggered by viral infection, such as occurs in response to SARS‐CoV2 [37], we sought to investigate whether Dex could also inhibit NLRP3 inflammasome activation following exposure to LPS and ATP in human monocyte‐derived macrophages (MoDMs) isolated from healthy donors and whether this effect was accompanied by a reduction in ACOD1. To inhibit GR signaling and thereby mimic the myeloid GRKO mouse, we treated human MoDMs with the GR antagonist RU‐486 for 30 min before adding 100 nM Dex following 16 h of LPS priming (Figures 5F and S11A,B). Consistent with observations in mouse WT and GRKO macrophages, Dex attenuated the LPS‐induced expression of NLRP3 and ACOD1. This effect was prevented by RU‐486 (Figures 5G,H and S11C). The reduction in NLRP3 and ACOD1 levels by Dex correlated with a decrease in the levels of pro‐IL‐1β, which was also prevented by RU‐486 (Figures 5I and S11C). Furthermore, Dex inhibited IL‐1β secretion triggered by ATP in human MoDMs, and this effect was blocked by the GR antagonist (Figure 5J). These findings establish that a clinically relevant course of Dex treatment effectively reduces ACOD1 expression and blocks human NLRP3‐inflammasome activation, suggesting a highly conserved mechanism between humans and mice.

4. Discussion

Synthetic glucocorticoids represent standard therapeutic options for managing acute and chronic diseases linked to sustained inflammation. Our study aimed to investigate whether glucocorticoids exert a regulatory effect on NLRP3 inflammasome activation in macrophages by modulating gene expression associated with metabolism. This study, employing a glucocorticoid treatment regimen which has clinically relevant timing, suggests that glucocorticoids regulate metabolism induced by inflammation to potentiate their anti‐inflammatory actions. We demonstrate that glucocorticoids mitigate NLRP3‐inflammasome and subsequent pyroptosis following extended LPS priming in macrophages. In this paradigm, the suppressive effect of glucocorticoid on the LPS‐sustained NLRP3 inflammasome activation is also associated with changes in the chromatin landscape induced by pro‐inflammatory signals. These alterations at the chromatin induced by LPS expand the action of GR as a transcription factor. Glucocorticoid effects are achieved by promoting mitochondrial function. Glucocorticoids also rescue the glycolytic switch induced by LPS. Glucocorticoids preserve the integrity of the TCA cycle while decreasing the expression and activity of ACOD1 and iNOS to control itaconate and nitric oxide, respectively. These transcriptional effects on metabolic genes potentiate the anti‐inflammatory effects of glucocorticoids during LPS‐sustained NLRP3 inflammasome activation.

Macrophages initiate pro‐inflammatory responses to microbial stimuli by activating the transcription of pro‐inflammatory genes, including those involved in the NLRP3 inflammasome pathway. Recent investigations have suggested a complex interplay between glucocorticoids and the NLRP3 inflammasome [15, 26, 28]. These studies also suggest that glucocorticoids are modulators of inflammatory responses and regulators of macrophage metabolism [15]. Acting as a transcription factor, GR exerts a regulatory influence on the mRNA levels of key genes such as Nlrp3, Nos2, and Il1b during the NLRP3 inflammasome activation. This regulation occurs through the inhibition of the NF‐κB and AP‐1 transcription factors [26, 28, 33]. Using WT and GR‐deficient macrophages, we shed light on the anti‐inflammatory mechanisms of glucocorticoids and their role in both reducing the expression and activity of iNOS, ACOD1, HIF‐1α, and NRF2 and altering mitochondrial fusion and fragmentation through changes in Mitofusin‐2 and DRP‐1. Since it has been demonstrated that the molecular mechanisms involved in mitochondrial dynamics and mitophagy regulate the fate of dysfunctional or damaged mitochondria, we have linked the glucocorticoids' action regulating metabolism with the RNA‐seq data affecting the gene expression associated with mitophagy to confirm a mechanism associated with mitochondrial quality control and cellular response stress induced by glucocorticoids (Figure S4A). Notably, dexamethasone not only promoted oxidative phosphorylation that attenuated LPS‐induced glycolysis and HIF‐1α and NRF2 expression but also contributed to the inhibition of mitochondrial ROS production and reversed the inhibitory effects of LPS on SDH activity in WT macrophages.

The induction of Acod1/Irg1 expression during LPS exposure also leads to the accumulation of itaconate, which inhibits the enzyme SDH [35]. Indeed, comprehensive transcriptomic and metabolomic profiles have been proposed as a mechanism by which glucocorticoids rewire macrophage metabolism during inflammation [15]. This regulatory action serves to prevent the intracellular accumulation of succinate by enhancing SDH activity and thus mitigating inflammation [15]. Glucocorticoids also regulated intracellular succinate levels in LPS‐activated macrophages [16]. This regulation does not appear to be directly associated with the modulation of ACOD1 and its product itaconate [16]. We identified the Acod1/Irg1 gene, together with the Nos2 gene, as two genes encoding for metabolic enzymes that were reduced early by glucocorticoids following macrophage pro‐inflammatory activation. We also validated the reduction in the expression of these genes at both mRNA and protein levels by glucocorticoids and confirmed decreased levels of nitric oxide and itaconate in a GR‐dependent manner. Interestingly, Dex treatment reduced the levels of nitric oxide to an extent comparable to that measured for the iNOS inhibitor 1400 W (Figure S9K), suggesting that the glucocorticoid‐mediated inhibition of late NLRP3 inflammasome activation requires alterations in other genes and/or metabolites.

Recently, a report employing a treatment paradigm where the inflammatory stimulus LPS and the anti‐inflammatory molecule Dex were administered together as a co‐treatment described how glucocorticoids affected the reprogramming of the mitochondrial metabolism of macrophages [16]. This study proposed a non‐genomic effect by which glucocorticoids induced reprogramming of metabolism via modulating the activity of the PDH enzyme [16]. This study also reported that the co‐treatment of LPS plus Dex resulted in the production of itaconate and inhibition of the inflammatory response independently of ACOD1 regulation [16]. Our experimental design adding Dex after onset of the inflammatory response emphasizes the importance and clinical relevance of studying GR action in the context of ongoing inflammation. This approach more closely simulates clinical scenarios where glucocorticoids are prescribed to treat pre‐existing inflammatory conditions. In addition, our studies examined the consequences of early GR activation on the regulation of inflammasome‐associated genes following inflammation as opposed to prolonged GR activation (after 24 h Dex treatment) when Dex and LPS are added as a co‐treatment that was utilized by these previous reports [15, 16, 38]. Our data analyzing GR accessibility to the chromatin by CUT&Tag support a direct GR‐mediated transcriptional transrepression of pro‐inflammatory genes by glucocorticoids following LPS priming, where GR gains access to repress genes located in a more proinflammatory chromatin landscape. These findings suggest that the temporal effects of glucocorticoid signaling have important clinical relevance and that GR as a ligand dependent transcription factor requires a more accessible chromatin landscape which expands the nuclear targets of GR to repress gene expression and promote anti‐inflammatory effects of glucocorticoids. Since the single Nr3c1 gene gives rise to multiple GR isoforms via alternative translation initiation [39], an important question for future studies will be to investigate whether these isoforms exhibit unique patterns of chromatin occupancy and target gene repression in an inflammatory context.

Our findings support the idea that glucocorticoids are an important component blocking the metabolic rewiring driven by the induction of iNOS and ACOD1 in pro‐inflammatory macrophages. In addition to antagonizing this pro‐inflammatory profile, glucocorticoids controlling ACOD1 expression and the production of itaconate could also safeguard and promote an anti‐inflammatory M2‐like phenotype. Supporting this assertion, macrophages polarized to M2‐like profile in the presence of IL‐4 or glucocorticoids revealed the presence of a shared core homeostatic program dependent on GR, KLF4, and the cofactor GRIP1 [40]. However, the underlying mechanisms for this phenomenon in an inflammatory context in vivo warrant further evaluation.

During inflammation, maintaining appropriate levels of itaconate and nitric oxide appears critical for suppressing NLRP3 inflammasome activation [18] and conferring tolerance to late NLRP3 inflammasome activation [17]. Regulation of ACOD1 and iNOS is necessary for averting subsequent proinflammatory responses and prevents life‐threatening secondary infections associated with immunotolerance [41, 42]. Since both ineffective mitophagy and abnormal mitochondrial dynamics increase NLRP3 inflammasome activation [43], our study reveals an alternative mechanism by which glucocorticoids, through the GR‐mediated regulation of ACOD1, iNOS, Mitofusin‐2, and DRP‐1 following sustained LPS stimulation, mitigate NLRP3 inflammasome activation and pyroptosis in macrophages. In addition to Dex reducing itaconate and succinate levels in the supernatant of LPS‐treated WT macrophages, we observed an increase in lactate accumulation. Both itaconate, succinate, and lactate are generated specifically during the immune response and are essential for fuelling and supporting immune cell activation [44]. Lactate also polarizes macrophages toward an M2 phenotype [45] and signals via the receptor GPR81 to reduce NLRP3 inflammasome activity in macrophages [46]. Thus, the observed glucocorticoid actions on NLRP3 inflammasome activation could be an alternative upstream event associated with lactate production to limit inflammation in macrophages.

Targeting of the NLRP3 inflammasome activation pathway in macrophages has been proposed as a key mechanism underlying the hyperinflammation and cytokine storm observed in SARS‐CoV‐2 infection [37]. Considering the COVID‐19 pandemic, emerging evidence underscores the pathological role of NLRP3 inflammasome activation in severe cases of the disease [37]. The anti‐inflammatory properties of Dex were shown to reduce mortality by dampening exaggerated inflammation and cytokine production by immune cells. Given that hyperinflammation and cytokine storm are intricately linked to NLRP3 inflammasome activation, the efficacy of Dex in the treatment of COVID‐19 may involve inhibiting the activation of the NLRP3 inflammasome [23, 47]. Overall, while seemingly unaffected under homeostatic conditions, myeloid knockout GR mice exhibit heightened susceptibility to stressful insults, particularly concerning NLRP3 inflammasome activation. Understanding the mechanism by which glucocorticoids achieve their anti‐inflammatory effects on NLRP3‐mediated pyroptosis could pave the way for exploring novel therapeutic applications of glucocorticoids in clinical settings.

Author Contributions

Conceptualization: D.D.‐J., R.H.O., J.A.C. Methodology: D.D.‐J., C.D.B., J.A.H., F.L., J.G.W., E.S. Investigation: D.D.‐J., C.D.B., J.A.H., E.S. Visualization: D.D.‐J., R.H.O., C.D.B., J.L., J.G.W., J.K.G., S.G. Funding acquisition: T.K.A., J.A.C. Project administration: D.D.‐J., J.A.C. Supervision: S.G., K.E.G., T.K.A., J.A.C. Writing – original draft: D.D.‐J. Writing – review and editing: D.D.‐J., R.H.O., C.D.B., K.E.G., S.G., J.C.A.

Funding

This research was supported by the Intramural Research Program of the National Institute of Health (NIH) at the NIEHS (ZIAES090057). The contributions of the NIH authors were made as part of their official duties as NIH federal employees, are in compliance with agency policy requirements, and are considered Works of the United States Government. However, the findings and conclusions presented in this paper are those of the authors and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Glucocorticoid treatment of LPS‐primed macrophages inhibits signaling pathways associated with inflammation.

Figure S2: Differential GR interaction with chromatin is govern by LPS and dexamethasone regimens in macrophages.

Figure S3: GR‐deficient macrophages exhibit hypersensitivity to LPS, and impaired glycolytic switch induced by inflammation.

Figure S4: GR‐deficient macrophages exhibit mitochondrial fragmentation and mitophagy deregulation.

Figure S5: Differences in levels of IL‐1β secretion regulated by glucocorticoids during the conventional and late NLRP3 inflammasome activation.

Figure S6: Effects of Glucocorticoids on conventional NLRP3 inflammasome activation.

Figure S7: Effects of Glucocorticoids on the pyroptosis induced by nigericin.

Figure S8: Effects of Glucocorticoids on late NLRP3 inflammasome activation in presence of NRF2 and HIF‐1α stabilizers.

Figure S9: Glucocorticoids attenuate the NLRP3‐inflammasome activation pathway dampening signal 1 and reinforcing protection against signal 2 overactivation.

Figure S10: Effect of Monosodium Urate Crystals on Thioglycolate‐elicited peritoneal macrophages from WT and myeloid specific GRKO mice.

Figure S11: Effect of GR antagonist RU‐486 on human monocytes‐derived macrophages (MDMs).

Diaz‐Jimenez D., Oakley R. H., Hoffman J. A., et al., “Timing of Glucocorticoid Treatment Dictates Glucocorticoid Receptor Actions Modulating the NLRP3‐Inflammasome Activation in Macrophages,” The FASEB Journal 40, no. 3 (2026): e71510, 10.1096/fj.202504083R.

Data Availability Statement

The data from RNA‐seq and CUT&Tag experiments (GSE306502) were deposited in a GEO database. The supporting raw mass spectrometry data have been uploaded to the MassIVE Data Repository (accession number MSV000099019).