α1-Antitrypsin (AAT), a serine protease inhibitor, is the third most abundant protein in plasma. Although the best-known function of AAT is irreversible inhibition of elastase, AAT is an acute-phase reactant and is increasingly recognized to have a panoply of other functions, including as an anti-inflammatory mediator and a host-protective molecule against various pathogens. Although a canonical receptor for AAT has not been identified, AAT can be internalized into the cytoplasm and is known to affect gene regulation. Because AAT has anti-inflammatory properties, we examined whether AAT binds the cytoplasmic glucocorticoid receptor (GR) in human macrophages. We report the finding that AAT binds to GR using several approaches, including coimmunoprecipitation, mass spectrometry, and microscale thermophoresis. We also performed in silico molecular modeling and found that binding between AAT and GR has a plausible stereochemical basis. The significance of this interaction in macrophages is evinced by AAT inhibition of LPS-induced NF-κB activation and IL-8 production as well as AAT induction of angiopoietin-like 4 protein, which are, in part, dependent on GR. Furthermore, this AAT–GR interaction contributes to a host-protective role against mycobacteria in macrophages. In summary, this study identifies a new mechanism for the gene regulation, anti-inflammatory, and host-defense properties of AAT.

Severe α1-antitrypsin (AAT) deficiency most commonly occurs in individuals who possess the protease inhibitor ZZ genotype, resulting in a substantially increased risk of precocious emphysema. Other disorders associated with protease inhibitor ZZ include bronchiectasis, panniculitis, cirrhosis, and anti-neutrophil cytoplasmic Ab–mediated vasculitis (13). Additionally, AAT has potent anti-inflammatory properties and is host-protective against HIV, influenza, Pseudomonas aeruginosa, and nontuberculous mycobacteria (412); however, the molecular basis of these AAT-associated activities is not well understood.

AAT resides in the cytoplasm of monocytes and macrophages and may be produced by them (13). Alternatively, elastase-bound AAT may enter cells from the extracellular milieu through endocytosis by low-density lipoprotein receptor–related protein-1 or through clathrin-coated vesicles and caveolae-mediated endocytosis (1416). The scavenger receptor B type I has also been proposed as a mechanism for endocytosis of AAT (17). In the cytoplasm, an anti-inflammatory effect of AAT occurs via binding to NF of κ light polypeptide gene enhancer in B cells inhibitor, α (IκBα), the basal inhibitor of NF-κB. Such binding by AAT to IκBα prevents IκBα phosphorylation, ubiquitination, and targeted proteasome degradation (12). Through these actions, AAT inhibits NF-κB activation. Additional proposed anti-inflammatory functions of AAT include binding of IL-8 (18) and inhibiting the ability of a disintegrin and metalloprotease 17 to cleave membrane-bound TNF to soluble TNF (19). Via its function as a serine protease inhibitor (serpin), AAT also irreversibly binds and inhibits thrombin, plasmin, and caspase-3, with the lattermost interaction preventing the apoptosis of endothelial cells (14, 20, 21).

Independent of its serpin activity, the mechanisms underlying anti-inflammatory and disease-mitigating properties of AAT remain poorly understood. AAT regulates transcription and is found in the nucleus (22, 23); however, its ability to translocate into the nucleus is unknown. In contrast, the pleiotropic actions of the glucocorticoid (GC)–GC receptor (GR) complex have been extensively studied (24). GR regulates pro- and anti-inflammatory genes by binding to transcriptional activators, coactivators, repressors, and corepressors (24). In the absence of binding to its ligand, GR resides in the cytoplasm bound to the chaperones Hsp90 and Hsp70. Upon binding to its canonical ligand, cortisol, the chaperone molecules dissociate from GR, and the GC–GR complex enters the nucleus to inhibit or induce transcription. GR signaling in response to GC leads to anti-inflammatory effects and immune suppression. A known connection between AAT and GR is that GC activation of GR increases AAT production at the transcriptional level (25, 26).

Given that both AAT and GCs have anti-inflammatory properties and reside in the cytoplasmic compartment, we investigated the possibility that AAT interacts directly with GR. We found that AAT binds GR and that this molecular interaction has anti-inflammatory and anti-mycobacterial functions in macrophages. Furthermore, this AAT–GR interaction lends itself to manipulation within the context of the anti-inflammatory, immunosuppressive, and lesser well-known immune-enhancing activities of the GR. Potential prospects include the development of small peptides that serve as functional surrogates for AAT to bind GR but with more favorable benefit-to-risk ratio than GC. This notion is plausible given that the many synthetic GCs are known to have differences in potency and severity of adverse effects.

The human monocytic cell line THP-1 and Mycobacterium tuberculosis H37Rv were obtained from American Type Culture Collection (Manassas, VA). M. intracellulare NJH9141 is a clinical strain isolated at National Jewish Health. FBS was purchased from Atlanta Biologicals (Lawrenceville, GA) and heat-inactivated at 56°C for 30 min. Recombinant human GR protein (catalog no. A15663), GR-specific rabbit polyclonal Ab purified by Ag affinity chromatography (PA1-511A) with robust GR specificity in applications such as GR chromatin immunoprecipitation (IP) (27), AAT-specific rabbit polyclonal Ab purified by ammonium sulfate precipitation and dialysis (PA5-26439), AAT-specific mouse mAb purified by recombinant fragment of human AAT (clone 2B12), the human angiopoietin-like 4 protein (ANGPTL4) ELISA Kit, goat anti-mouse IgG (H+L) Alexa Fluor Plus 488, Lipofectamine RNAi-MAX Transfection reagent (catalog no. 13778075), and CyQUANT MTT Cell Viability Assay Kit (catalog no. V13154) were purchased from Invitrogen-Thermo Fisher Scientific (Carlsbad, CA). TLR2-specific polyclonal Ab (rabbit; catalog no. 12276), tubulin polyclonal Ab (catalog no. 2144), lamin B polyclonal Ab (catalog no. 12586), and β-actin polyclonal Ab (rabbit) were purchased from Cell Signaling Technology (Danvers, MA). Human IL-8/CXCL8 Quantikine ELISA kit and nonimmune human IgG control were purchased from R&D Systems (Minneapolis, MN). PMA, LPS (catalog no. L-2880), cortisol (catalog no. C-106), and dexamethasone (catalog no. 1756) were purchased from Sigma-Aldrich (St. Louis, MO). The TransAM NF-κB p65 kit was purchased from Active Motif (Carlsbad, CA). An aliquot of the GR (NR3C1) human short hairpin RNA (shRNA; shRNA-GR) Lentiviral Particle was a kind gift from Miles Pufall (University of Iowa Carver College of Medicine). Protein A Sepharose 4 Fast Flow Sepharose beads were purchased from GE Healthcare Life Sciences (Pittsburgh, PA). AAT (Glassia) was acquired from Kamada Pharmaceuticals (Rehovot, Israel). Cy3 goat anti-rabbit IgG (H+L) A10520 was purchased from Life Technologies (Grand Island, NY); GRa-small interfering RNA (siRNA; sc-35505) was purchased from Santa Cruz Biotechnology (Dallas, TX).

Human THP-1 cells were cultured in RPMI 1640 medium containing 2 mM l-glutamine (Life Technologies), 10% FBS, 100 U/ml penicillin, and 100 µg/ml streptomycin at 37°C and 5% CO2. THP-1 cells were differentiated into macrophages following incubation with 15 ng/ml PMA for 24 h. PBMCs from a healthy adult donor (known to possess the protease inhibitor MM [PiMM] genotype for AAT) were isolated from blood collected in CPT tubes after informed consent based on an approved Institutional Review Board protocol (HS-2651) that abides by the Declaration of Helsinki. Differentiation of human monocytes to monocyte-derived macrophages (MDM) was performed in the presence of 20 ng/ml monocyte-CSF and cultured in 10% autologous plasma as previously described (4).

Differentiated human THP-1 macrophages and MDM were seeded onto four-well chamber slides (Lab-Tek; Thermo Fisher Scientific) at a density of 0.5–1 × 105 cells/well. Cells were fixed with 4% paraformaldehyde for 30 min. Fixed cells were permeabilized for 20 min using 0.3% Triton X-100 in PBS. Nonspecific binding of Abs was blocked by incubating the cells for 1–1.5 h with 3% BSA dissolved in 1× PBS with Tween-20 (PBST). Cells were incubated with polyclonal anti-GR (1:300 dilution) and monoclonal anti-AAT (1:300 dilution) Abs overnight at 4°C, washed three times, and incubated with anti-rabbit Cy3 and anti-mouse Alexa Fluor Plus 488 secondary Abs, respectively (both 1:1000 dilution), at room temperature for 1–2 h. The slides were sealed with ProLong Gold Antifade Mountant with DAPI (Thermo Fisher Scientific). The THP-1 macrophages were imaged using a SP5 confocal microscope (Leica Microsystems) with a magnification of ×63 and a 1.4 numerical aperture oil immersion objective. The fluorescent labels DAPI, Cy3, and Alexa Fluor Plus 488 were excited using 359-, 552-, and 495-nm laser excitation, respectively. The MDM were analyzed at ×40 original magnification using a fluorescent microscope (Carl Zeiss Axiovert 200M) equipped with DAPI and fluorescent filters.

IP was performed according to a previously published protocol (28). Briefly, the cells were lysed with 50 mM Tris–HCl (pH 7.4) containing 0.5% (v/v) Nonidet P-40, 1 mM EDTA, 150 mM NaCl, 5 µg/ml aprotinin, 5 µg/ml leupeptin, 2 mM Na3VO4, and 1 mM PMSF. The Bradford protein assay (Bio-Rad Laboratories) was used to quantify the protein concentrations of the whole-cell lysates. Then, 300 μg of protein from each preparation was incubated with 3 μg of anti-GR polyclonal Abs, anti-AAT polyclonal Abs, or control nonimmune IgG at 4°C overnight with gentle rotary mixing. The protein–Ab complexes were isolated by incubation with 20 μl protein A-Sepharose beads with gentle mixing for 2 h at 4°C. The beads were washed three times with wash buffer (10 mM Tris buffer [pH 7.4] with protease inhibitor mixture).

Each IP pellet was resuspended in 30 μl 1× Laemmli/DTT buffer and heated at 95°C for 5 min for immunoblot analysis. The IP preparations and whole-cell lysates were fractionated using 12% SDS-PAGE and transferred onto polyvinylidene difluoride membranes using the iBlot 2 dry blotting System from Invitrogen (Thermo Fisher Scientific). For immunodetection of AAT, the membranes were blocked in PBST buffer containing 5% fat-free milk powder (blocking buffer) for 1 h, then incubated overnight at 4°C with anti-AAT polyclonal Ab, anti-AAT mAb, or anti–β-actin Ab (each at a dilution of 1:1000 [v/v]), followed by detection with HRP-conjugated anti-rabbit IgG (1:2000 dilution) or HRP-conjugated anti-mouse IgG (1:2000). To detect GR, polyclonal anti-GR polyclonal (1:1000) or anti–β-actin (1:1000) were diluted in PBST buffer with 5% BSA overnight at 4°C with light shaking, followed by detection using HRP-conjugated anti-rabbit IgG at 1:2000 to detect GR or 1:5000 to detect β-actin–Ab binding in blocking buffer at room temperature for 1–2 h. Immunoblotting for lamin and tubulin was performed with polyclonal anti-lamin (1:3000 dilution) and polyclonal anti-tubulin (1:3000), respectively, diluted in blocking buffer overnight at 4°C, and then incubated at room temperature for 1 h with anti–rabbit-HRP–conjugated Abs at 1:2000 dilution with blocking buffer. The bands were visualized using the SuperSignal West Femto Maximum Sensitivity Substrate System (Thermo Fisher Scientific). TLR2 immunoblotting was performed with anti-TLR2–specific polyclonal Ab (1:1000) and β-actin–specific polyclonal Ab (1:3000).

IP was performed as described above. The bound immune complexes were eluted by heating in 1× Laemmli/DTT loading buffer, separated with 12% SDS-PAGE, and the gel stained with Coomassie. The gel pieces were destained in 200 µl of 25 mM ammonium bicarbonate in 50% v/v acetonitrile for 15 min and washed with 200 µl of 50% (v/v) acetonitrile. Disulfide bonds in proteins were reduced by incubation in 10 mM DTT at 60°C for 30 min, and cysteine residues were alkylated with 20 mM iodoacetamide in the dark at room temperature for 45 min. Gel pieces were subsequently washed with 100 µl of distilled water followed by addition of 100 ml of acetonitrile and dried on SpeedVac (Savant Thermo Fisher). Then, 100 ng of trypsin was added to each sample and allowed to rehydrate the gel plugs at 4°C for 45 min and then incubated at 37°C overnight. The tryptic mixtures were acidified with formic acid up to a final concentration of 1%. Peptides were extracted twice from the gel plugs using 1% formic acid in 50% acetonitrile. The collected extractions were pooled with the initial digestion supernatant and dried on SpeedVac. Samples were desalted on Thermo Fisher Scientific Pierce C18 Tip.

Samples were analyzed on an LTQ Orbitrap Velos mass spectrometer (Thermo Fisher Scientific) coupled to an Eksigent NanoLC-2D system through a nanoelectrospray liquid chromatography-mass spectrometry interface. An autosampler injected 8 μl of sample into a 10-μl loop. To desalt the sample, the material was flushed out of the loop, loaded onto a trapping column (ZORBAX 300SB-C18; dimensions 5 × 0.3 mm × 5 μm), and washed with 0.1% formic acid at a flow rate of 5 μl/min for 5 min. The analytical column was then switched online at 0.6 μl/min over an in-house–made 100-μm internal diameter × 200-mm fused silica capillary packed with 4-μm 80 Å Synergi Hydro C18 resin (Phenomex, Torrance, CA). After 10 min of sample loading, the flow rate was adjusted to 0.35 ml/min, and each sample was run on a 90-min linear gradient of 4–40% acetonitrile with 0.1% formic acid to separate the peptides. Liquid chromatography mobile-phase solvents and sample dilutions used 0.1% formic acid in water (buffer A) and 0.1% formic acid in acetonitrile (buffer B) (Chromasolv LC-MS grade; Sigma-Aldrich).

Data were acquired using the instrument-supplied Xcalibur (version 2.1) software. The mass spectrometer was operated in the positive ion mode. Each survey scan of mass-to-charge ratio 400–2000 was followed by collision-assisted dissociation tandem mass spectrometry (MS/MS) of the 20 most intense precursor ions. Singly charged ions were excluded from collision-assisted dissociation selection. Doubly charged and higher ions were included. Normalized collision energies were employed using helium as the collision gas.

MS/MS spectra were extracted from raw data files and converted into mascot generic format (.mgf) files using a PAVA script (Mass Spectrometry Facility, University of California San Francisco, San Francisco, CA). These .mgf files were then independently searched against the human SwissProt database (13,188 entries) using an in-house Mascot server (Version 2.6; Matrix Science). Mass tolerances were ±10 ppm for MS peaks and ±0.6 Da for MS/MS fragment ions. Trypsin specificity was used, allowing for one missed cleavage. Met oxidation, protein N-terminal acetylation, and peptide N-terminal pyroglutamic acid formations were allowed as variable modifications, whereas carbamidomethyl of Cys was set as a fixed modification.

Scaffold (version 4.8; Proteome Software, Portland, OR) was used to validate MS/MS-based peptide and protein identifications. Peptide identifications were accepted if they could be established at >95.0% probability as specified by the Peptide Prophet algorithm. Protein identifications were accepted if they could be established at >99.0% probability and contained at least two identified unique peptides. All peptide sequences assigned are provided in Supplemental Table I. The full mass spectrometry data are available via ProteomeXchange with identifier PXD030989 (http://www.ebi.ac.uk/pride/archive/projects/PXD030989).

To measure the binding affinity between AAT and GR in a cell-free assay, microscale thermophoresis (MST) was performed using the Monolith NT.115 Pico instrument (NanoTemper Technologies, München, Germany) according to the manufacturer’s instructions. Instrument performance was confirmed prior to each experiment using the Monolith NT Control Kit RED Molecular Interaction Control Kit for Monolith NT.115 Blue/Red. AAT protein was fluorescently labeled using a Protein Labeling Kit RED-NHS 2nd Generation kit (Amine Reactive) (NanoTemper Technologies) according to the manufacturer’s specifications. Recombinant human GR was serially diluted 2-fold in nanopure water 2-fold from the stock concentration to create a 16-member concentration series (1.91 nM to 6.25 × 103 nM). Fluorescent-labeled 10 nM AAT, prepared in PBS buffer with 0.05% P20, was added to the different GR aliquots. Samples were then loaded into premium Monolith NT.115 capillary tubes, and the tubes were loaded onto the Monolith NT.115 Pico instrument. For all of the experiments, a Pico.RED detector was used at 20% laser power and medium MST power. Optimal laser and MST power settings, as well as protein stability, were confirmed by measuring a capillary tube containing only 10 nM of the fluorescent-labeled AAT diluted in PBS. Data were analyzed using the MO Affinity Analysis software suite (NanoTemper Technologies).

A model of GR was constructed using the known structures of DNA-binding domain (DBD; Protein Data Bank [PDB] identification no. 5HN5) and ligand-binding domain (LBD; PDB identification no. 1M2Z) and AlphaFold2 (29) to construct the hinge region connecting them. Structure refinement was performed using the Rosetta Relax protocol (30). Given the intrinsic disorder of the N-terminal activation function-1 domain (NTD), it was not modeled. Structural representations were produced using Schrödinger PyMOL version 2.5.2. Structural alignments were performed using MUSTANG-MR (31). Docking simulations were performed using AlphaFold2 and ClusPro (29, 32).

Nuclear and cytoplasmic fractions of THP-1 macrophages were isolated using the NE-PER Nuclear and Cytoplasmic Extraction Reagents kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. Briefly, THP-1 macrophages were washed with 1× PBS and resuspended in cell extraction buffer supplemented with 1 mM PMSF and a protease inhibitor mixture (Cell Signaling Technology). Nuclear and cytoplasmic protein fraction concentrations were quantified prior to coimmunoprecipitation (co-IP) of AAT and GR.

We employed shRNA-lentivirus technology to develop a pool of THP-1 cells stably depleted for GR. In brief, THP-1 cells were plated in 12-well tissue culture plates at a density of 1 × 104 cells/well and infected with GR-directed shRNA or scrambled shRNA lentiviral particles at a multiplicity of infection (MOI) of ≥1. THP-1 cells infected with the scrambled shRNA (THP-1control) and those stably depleted (knockdown [KD]) for GR (THP-1GR-KD) were selected by adding 1 µg/ml puromycin dihydrochloride to the cell culture medium.

PBMC were isolated from the same PiMM individual as described above differentiated to MDM. MDM (6.2 × 105) were transfected with one of two siRNA (scrambled as a control or targeting GR) using Lipofectamine RNAiMAX Transfection reagent (Thermo Fisher Scientific). After 24 h, the cells were harvested and immunoblotted for GR.

RNA sequencing (RNAseq) and immunoblotting were used to confirm the depletion of GR mRNA and protein, respectively. For the former approach, we analyzed the expression of the NR3C1 (GR) gene as part of another study of total RNA sequencing using the Illumina Next Generation Sequencing service from the Genomics Shared Resource at the University of Colorado Cancer Center in THP-1control and THP-1GR-KD cells. RNAseq data were analyzed using DAVID Bioinformatics Resources 6.8 Vision software.

The binding of the p65 subunit of NF-κB was quantified using the TransAM NF-κB p65 kit (Active Motif), as we previously reported (33).

Differentiated THP-1control and THP-1GR-KD macrophages were left untreated or treated with 1 µg/ml LPS alone or in the presence of 10 µM cortisol, 1 µM dexamethasone, or 3 or 5 mg/ml AAT at 37°C in a 5% CO2 incubator. After 6 and 24 h, the culture supernatants were quantified for IL-8 (R&D Systems) or ANGPTL4 (Thermo Fisher Scientific) expression according to the manufacturer’s instructions. MDMcontrol and MDMGR-KD were left untreated or treated with 1 µg/ml LPS alone or in the presence of either 10 µM cortisol or 3 or 5 mg/ml AAT. The cells were incubated at 37°C in a 5% CO2 incubator for 24 h and the supernatant assayed for IL-8.

Infection of THP-1 macrophages with M. tuberculosis or M. intracellulare and quantitation of cell-associated mycobacteria were performed as previously described (33). In brief, differentiated THP-1 macrophages in 24-well tissue culture plates were infected with M. tuberculosis H37Rv or M. intracellulare at an MOI of 10 bacilli to 1 macrophage. For cells infected for 1 h (day 0), the supernatants were collected after 1 h of infection, the cells were washed twice with a 1:1 solution of RPMI-1640/1× PBS, and the adherent cells were lysed with 250 µl of a 0.25% SDS/well, followed by addition of 250 µl of 7H9 plating broth. Serial dilutions of cell lysates were prepared, and then 5 µl of each dilution was plated on Middlebrook 7H10 agar. For day 2 and 4 infections, the cells were washed twice after the initial 1 h of infection and replaced with RPMI-1640 medium containing 10% FBS. After 2 or 4 d of incubation, the supernatants were centrifuged to recover any nonadherent macrophages; these macrophages were lysed together with the adherent macrophages and M. tuberculosis or M. intracellulare cultured as described above.

Replicate experiments (two for the MST experiments and three for the cellular experiments with the latter done in duplicate wells) are independent, and data are presented as means ± SEM or representative experiment. Group means were compared by repeated-measures ANOVA using Fisher least significant test or two-way ANOVA with Bonferroni post hoc test. Data were graphed in Prism 9, and comparisons were considered significant when p was <0.05.

The mass spectrometry proteomics data were deposited to the ProteomeXchange Consortium via the Proteomics Identification Database (PRIDE) partner repository under data set identifier PXD030989 and 10.6019/PXD030989 (http://www.ebi.ac.uk/pride/archive/projects/PXD030989).

Fluorescent immunocytochemical staining of macrophages was performed using anti-AAT and anti-GR Abs conjugated to different fluorochromes to determine whether AAT and GR are expressed under basal conditions. Both AAT and the GR colocalized in the cytoplasm and, to a lesser extent, within the nuclei of differentiated THP-1 macrophages by confocal microscopy (Fig. 1A). MDM from an individual who possesses two normal M alleles for AAT (PiMM) also showed that AAT and GR are found intracellularly by fluorescent microscopy (Fig. 1B). Similarly, AAT and GR expression in human macrophages were confirmed by immunoblotting whole-cell lysates (Fig. 1C).

FIGURE 1.

Immunofluorescence staining for GR and AAT. Human THP-1 differentiated macrophages (A) and human MDM (hMDM) (B) were immunostained using anti-AAT and anti-GR Abs conjugated to the fluorochromes Alexa Fluor Plus 488 and Cy3, respectively. THP-1 macrophages were imaged with a confocal microscope. Human MDM were imaged with a fluorescent microscope. Scale bars, 5 μm. (C) Immunoblotting for AAT and GR in human macrophages. Images shown are representative of three independent experiments. wcl, whole-cell lysate.

FIGURE 1.

Immunofluorescence staining for GR and AAT. Human THP-1 differentiated macrophages (A) and human MDM (hMDM) (B) were immunostained using anti-AAT and anti-GR Abs conjugated to the fluorochromes Alexa Fluor Plus 488 and Cy3, respectively. THP-1 macrophages were imaged with a confocal microscope. Human MDM were imaged with a fluorescent microscope. Scale bars, 5 μm. (C) Immunoblotting for AAT and GR in human macrophages. Images shown are representative of three independent experiments. wcl, whole-cell lysate.

Close modal

Co-IP and immunoblot approach

To investigate whether AAT binds to the GR in differentiated THP-1 macrophages, whole-cell lysates were IP for GR with polyclonal nonimmune IgG control or anti-GR Abs using protein A Sepharose beads. The beads were washed and centrifuged and the bound proteins extracted, separated by SDS-PAGE, and immunoblotted with an anti-AAT polyclonal Ab. In contrast to nonimmune IgG, IP for the GR followed by immunoblot for AAT revealed a distinct band at ∼52 kDa, which is consistent with the molecular mass of pharmacologic AAT (Glassia) and AAT from whole-cell lysates (Fig. 2A, left side of immunoblot). To corroborate the interaction between AAT and GR in primary human macrophages, MDMs were prepared from an individual with the PiMM AAT phenotype. Compared to the nonimmune IgG control, IP of MDM lysates with anti-GR revealed the presence of AAT within the IP lysates (Fig. 2A, right side of immunoblot). Conversely, THP-1 cell lysates were subjected to IP for AAT, and the separated proteins were immunoblotted for GR. As shown in (Fig. 2B, GR was detected in the fraction IP for AAT but not in the fraction IP with nonimmune IgG.

FIGURE 2.

AAT coimmunoprecipitates with GR. (A) THP-1 macrophage and human MDM lysates were subjected to IP with anti-GR polyclonal Ab, followed by immunoblotting of the IP fraction to detect AAT and β-actin. (B) THP-1 macrophage lysates were IP with anti-AAT polyclonal Ab, followed by immunoblotting of the IP fraction for GR. (C) Comparing the relative amount of IgG and anti-GR Ab alone (“No cell extract”) used in the IP experiments, with THP-1 macrophage lysates IP with nonimmune IgG or anti-GR, followed by immunoblotting for AAT. (D) THP-1 macrophage lysates were IP with anti-GR polyclonal Ab followed by immunoblotting of the IP fraction with a mouse anti-AAT mAb. (E) Densitometry of the co-IP protein using nonimmune IgG, anti-GR Ab, or anti-AAT Ab for immunoblots represented in (A)–(D) [letters “A” to “D” shown in the graph correspond to the IP bands in (A)–(D)]. The protein listed above each bar is the protein detected in the immunoblot. Images in (A)–(D) are representative of at least three independent experiments. Data shown in (E) are the mean ± SD of three independent experiments. Lanes labeled “No cell extract” contain only IgG or anti-GR Ab alone with no IP performed. **p < 0.01, ***p < 0.001. WB, Western blot; WCL, whole-cell lysates.

FIGURE 2.

AAT coimmunoprecipitates with GR. (A) THP-1 macrophage and human MDM lysates were subjected to IP with anti-GR polyclonal Ab, followed by immunoblotting of the IP fraction to detect AAT and β-actin. (B) THP-1 macrophage lysates were IP with anti-AAT polyclonal Ab, followed by immunoblotting of the IP fraction for GR. (C) Comparing the relative amount of IgG and anti-GR Ab alone (“No cell extract”) used in the IP experiments, with THP-1 macrophage lysates IP with nonimmune IgG or anti-GR, followed by immunoblotting for AAT. (D) THP-1 macrophage lysates were IP with anti-GR polyclonal Ab followed by immunoblotting of the IP fraction with a mouse anti-AAT mAb. (E) Densitometry of the co-IP protein using nonimmune IgG, anti-GR Ab, or anti-AAT Ab for immunoblots represented in (A)–(D) [letters “A” to “D” shown in the graph correspond to the IP bands in (A)–(D)]. The protein listed above each bar is the protein detected in the immunoblot. Images in (A)–(D) are representative of at least three independent experiments. Data shown in (E) are the mean ± SD of three independent experiments. Lanes labeled “No cell extract” contain only IgG or anti-GR Ab alone with no IP performed. **p < 0.01, ***p < 0.001. WB, Western blot; WCL, whole-cell lysates.

Close modal

We further confirmed the integrity of the AAT–GR binding interaction by excluding an experimental artifact resulting from the m.w. similarity between H chain Igs and AAT. We employed two approaches. In the first approach, the THP-1 macrophage lysates were IP as previously indicated with the anti-GR polyclonal Ab. In addition, 0.5 µg of anti-GR or nonimmune IgG Abs, the amount used for IP in each condition, was loaded onto separate wells of the SDS-PAGE gel. Although the migration of IgG H chain is similar to that of AAT, the intensities of the relevant bands in the lanes that contained only either nonimmune IgG or anti-GR (with no IP of cell extract) are significantly less than the IP band, indicating that the prominent band at 52 kDa following IP for GR mostly represents AAT and not the H chain of IgG (Fig. 2C). In the second approach, we conducted IP with rabbit anti-GR polyclonal Ab as before but used a mouse anti-AAT mAb for immunoblotting. This approach excluded detection of the anti-GR Ab (or the nonimmune IgG) used for the IP in the immunoblot. This strategy still demonstrated a prominent band at 52 kDa (representing co-IP AAT) following IP with anti-GR and immunoblotting for AAT but not with the control IgG (Fig. 2D). To semiquantify the co-IP proteins in these experiments, we performed densitometry of the relevant bands detected in the immunoblots following IP; the Abs used in the IP experiments are labeled on the x-axis and the protein probed by the immunoblot is labeled on top of each of the bars in (Fig. 2E. In summary, the data provide evidence that AAT binds the GR to form a complex in THP-1 macrophages and MDM.

Co-IP and mass spectrometry approach

We adopted a co-IP and mass spectrometry approach to further validate the physical interaction between AAT and the GR. THP-1 macrophage lysates were incubated with nonimmune IgG or anti-GR Ab as before, and the IP proteins were separated with SDS-PAGE under reducing conditions. After electrophoresis, the gel was stained with Coomassie blue. Protein bands that corresponded in m.w. to AAT following IP with nonimmune IgG and anti-GR Ab were excised from the gel and prepared for mass spectrometry (Fig. 3A, dashed and solid black line boxes). Because the m.w. of AAT may vary based on its glycosylation, we also excised adjacent areas of the gel just beneath the first excised sites and performed mass spectrometry of the samples (Fig. 3A, dashed and solid red line boxes).

FIGURE 3.

IP-mass spectrometry and MST demonstrate AAT–GR interaction. (A) THP-1 macrophages were IP with nonimmune IgG or anti-GR Ab, the IP fractions were separated by SDS-PAGE, and the gel was stained with Coomassie blue. Stained bands in the gel that ran similarly to AAT (dashed and solid black line boxes) were excised along with an area just beneath the aforementioned band (dashed and solid red line boxes) and subjected to mass spectrometry. (B) Graphical and numerical representation of the number of total and unique peptides that referenced the AAT protein in samples IP with control IgG (zero AAT peptides for either the top or bottom bands) or with anti-GR Ab (37 total AAT peptides with 12 unique AAT peptides in the top band and 21 total AAT peptides with 8 unique to AAT in the bottom band). Data shown are representative of two independent experiments. (C) MST time-course tracings following the mixing of fluorescent-tagged AAT with 16 different GR concentrations and subjected to a thermogradient. (D) The presence of a change in thermophoretic mobility of the fluorescent-tagged AAT with varying GR concentrations demonstrates that the two molecules interact in vitro. Data shown in (D) are the mean ± SD of two independent experiments.

FIGURE 3.

IP-mass spectrometry and MST demonstrate AAT–GR interaction. (A) THP-1 macrophages were IP with nonimmune IgG or anti-GR Ab, the IP fractions were separated by SDS-PAGE, and the gel was stained with Coomassie blue. Stained bands in the gel that ran similarly to AAT (dashed and solid black line boxes) were excised along with an area just beneath the aforementioned band (dashed and solid red line boxes) and subjected to mass spectrometry. (B) Graphical and numerical representation of the number of total and unique peptides that referenced the AAT protein in samples IP with control IgG (zero AAT peptides for either the top or bottom bands) or with anti-GR Ab (37 total AAT peptides with 12 unique AAT peptides in the top band and 21 total AAT peptides with 8 unique to AAT in the bottom band). Data shown are representative of two independent experiments. (C) MST time-course tracings following the mixing of fluorescent-tagged AAT with 16 different GR concentrations and subjected to a thermogradient. (D) The presence of a change in thermophoretic mobility of the fluorescent-tagged AAT with varying GR concentrations demonstrates that the two molecules interact in vitro. Data shown in (D) are the mean ± SD of two independent experiments.

Close modal

Mass spectrometry of the band corresponding to the pharmacologic control AAT (Glassia) (Fig. 3A, dotted black box, far right AAT lane) revealed 1762 total peptides and 62 unique peptides that referenced to the complete AAT protein. For the cell lysates IP with nonimmune IgG, mass spectrometry of both the upper band and lower adjacent area (Fig. 3A, dashed back and red line boxes, respectively) yielded zero peptides corresponding to AAT (Fig. 3B). For the cell lysates IP with anti-GR, mass spectrometry of the upper band (Fig. 3A, solid black line box) yielded 37 total peptides that corresponded to amino acid sequences of AAT, 12 of which are unique to the complete AAT protein (Fig. 3B). Similarly, the lower adjacent area (Fig. 3A, solid red line box) revealed a total of 21 peptides that referenced to AAT, 8 of which were unique peptides to AAT (Fig. 3B). As anticipated, the H chain of IgG also migrated to a similar position as AAT in that the upper band and the adjacent lower area of the migrated lysates IP with nonimmune IgG or anti-GR contained 540 and 1175 peptides (or 265 and 848 peptides) that referenced to their respective IgG H chains. These findings further corroborated that AAT co-IP with GR.

MST approach

To determine whether AAT binds to the GR in a cell-free system, we employed MST technology. MST is a highly sensitive method that measures the mobility of a fluorescently labeled protein in the presence of a thermal gradient, a process known as thermophoresis (34). The thermophoretic mobility of a molecule is dependent on multiple factors, including charge, hydrodynamic radius, and interactions with the solvent. If the fluorescently labeled molecule binds another molecule, the thermophoretic mobility of the labeled molecule increases by a measurable amount. For this specific experiment, 10 nM of fluorescently labeled AAT was combined with 16 GR concentrations ranging from 1.91 nM to 6.25 µM. The thermophoretic mobility of AAT changed in a GR concentration-dependent manner (Fig. 3C, 3D). We speculate that the partial reversal of the AAT–GR thermophoretic mobility at the highest GR concentrations suggests the possibility that more than one GR binding to AAT as seen biologically as well as the molecular modeling (detailed below). The calculated Kd of 4.62 × 10−8 ± 1.7 × 10−8 M indicates a strong interaction between the AAT and GR and validates in a cell-free system that AAT and GR bind to each other.

Molecular modeling of AAT–GR interaction

To begin to understand how AAT and GR may interact stereochemically, we performed in silico molecular modeling and docking calculations using two state-of-the-art approaches: AlphaFold2 (29) and ClusPro (32). GR is organized into three major domains: an intrinsically disordered NTD and two ordered domains comprised of a DBD and a C-terminal LBD (Fig. 4A). In addition to its role in ligand recognition, the LBD also contains a ligand-dependent activation function 2 (AF-2) region that is tightly regulated by hormone binding (dotted box). Structural modeling of the GR protein reveals that an unstructured hinge region links the LBD and DBD (Fig. 4B). Docking analyses revealed several plausible modes of interaction between AAT and GR both in the cytosol and nucleus. The top docking solution from AlphaFold2 shows AAT interacting with the LBD via its reactive center loop (RCL), in a binding mode compatible with the GR–Hsp90/p23 cochaperone complex (35) (Fig. 4C). The AAT–GR complex is also compatible with the known dimeric arrangement of LBD (36) (Fig. 4D). The RCL of AAT interacts hydrophobically with the ligand-dependent AF-2 region of the LBD in a similar fashion to that of the two nuclear coregulator proteins: nuclear receptor corepressor (NCoR) and transcriptional intermediary factor-2 (TIF2) (Fig. 4D). Superposed is a monomeric NCoR-bound LBD showing key structural differences within the AF-2 site that favor repression over activation. Once translocated to the nucleus, the DBD of GR recognizes GC response elements (GREs). The top docking result using ClusPro positions AAT at the opposite end of the LBD, again interacting via its RCL, bridging the interaction with a monomeric DBD, but showing some degree of steric overlap with the second DBD formed by d-loop–mediated dimerization (Fig. 4E). Thus, we hypothesize that the model shown in (Fig. 4D is stereochemically more plausible than the one shown in (Fig. 4E.

FIGURE 4.

Molecular modeling of AAT–GR complex. (A) GR is organized into three major domains: an intrinsically disordered NTD, a DBD, and a C-terminal LBD. (B) Structural modeling of the GR protein reveals the presence of the LBD (green) and DBD (magenta), linked by an unstructured hinge region (gray). The intrinsically disordered NTD was not modeled and is depicted by dotted lines. Dexamethasone (DEX) bound to the LBD is shown as steel blue spheres. (C) AlphaFold2 docking simulations of LBD and AAT (PDB 3NE4) superimposed on the GR-Hsp90-p23 cochaperone complex (PDB 7KRJ) show AAT interacting with the LBD via its RCL. (D) The same AAT–GR complex expanded to the LBD dimer (monomers labeled LBD-A and LBD-B). Superimposed are the nuclear coregulator proteins NCoR (derived from PDB 3H52) and TIF2 (derived from PDB 1M2Z) bound to the AF-2 site (broken rectangle) are shown as brown and slate-blue cartoons, respectively. (E) The second model using ClusPro docking shows the RCL of AAT interacting with the LBD of GR. In addition, the DBD is shown complexed with DNA (space-filling atoms), modeled using the GR DBD monomer: thymic stromal lymphopoietin (TSLP) nGRE complex (magenta; PDB 5HN5). Also shown (pink) is a second DBD formed by d-loop–mediated dimerization (PDB 5E69), having some degree of steric overlap with AAT. RCL shown in red.

FIGURE 4.

Molecular modeling of AAT–GR complex. (A) GR is organized into three major domains: an intrinsically disordered NTD, a DBD, and a C-terminal LBD. (B) Structural modeling of the GR protein reveals the presence of the LBD (green) and DBD (magenta), linked by an unstructured hinge region (gray). The intrinsically disordered NTD was not modeled and is depicted by dotted lines. Dexamethasone (DEX) bound to the LBD is shown as steel blue spheres. (C) AlphaFold2 docking simulations of LBD and AAT (PDB 3NE4) superimposed on the GR-Hsp90-p23 cochaperone complex (PDB 7KRJ) show AAT interacting with the LBD via its RCL. (D) The same AAT–GR complex expanded to the LBD dimer (monomers labeled LBD-A and LBD-B). Superimposed are the nuclear coregulator proteins NCoR (derived from PDB 3H52) and TIF2 (derived from PDB 1M2Z) bound to the AF-2 site (broken rectangle) are shown as brown and slate-blue cartoons, respectively. (E) The second model using ClusPro docking shows the RCL of AAT interacting with the LBD of GR. In addition, the DBD is shown complexed with DNA (space-filling atoms), modeled using the GR DBD monomer: thymic stromal lymphopoietin (TSLP) nGRE complex (magenta; PDB 5HN5). Also shown (pink) is a second DBD formed by d-loop–mediated dimerization (PDB 5E69), having some degree of steric overlap with AAT. RCL shown in red.

Close modal

Given that GR can localize to the nuclear compartment, our next step in characterizing the importance of the AAT–GR interaction was to identify whether the AAT–GR complex exists in the cytoplasm, nucleus, or both. We prepared nuclear and cytoplasmic fractions of the THP-1 macrophage lysates (Fig. 5A) and performed IP for GR, followed by AAT immunoblotting. We identified the presence of the AAT–GR complex in both the nuclear and cytoplasmic fractions, suggesting that this interaction may play a role in the transcriptional function of GR (Fig. 5B).

FIGURE 5.

The AAT–GR complex is found in both the nucleus and cytoplasm. (A) Pre-IP immunoblot (IB) experiment to detect lamin and tubulin to confirm specific isolation of nuclear and cytoplasmic fractions, respectively. (B) IP of the nuclear and cytoplasmic fractions for GR and immunoblot of immunoprecipitated lysates with an anti-AAT Ab. The same membrane was then immunoblotted for lamin and tubulin (two bottom panels). All data shown are representative of three independent experiments. CF, cytoplasmic fraction; NF, nuclear fraction.

FIGURE 5.

The AAT–GR complex is found in both the nucleus and cytoplasm. (A) Pre-IP immunoblot (IB) experiment to detect lamin and tubulin to confirm specific isolation of nuclear and cytoplasmic fractions, respectively. (B) IP of the nuclear and cytoplasmic fractions for GR and immunoblot of immunoprecipitated lysates with an anti-AAT Ab. The same membrane was then immunoblotted for lamin and tubulin (two bottom panels). All data shown are representative of three independent experiments. CF, cytoplasmic fraction; NF, nuclear fraction.

Close modal

We next investigated the biological importance of the AAT–GR interaction by generating a pool of transformed THP-1 macrophages in which GR was stably KD using a lentiviral vector that expressed an shRNA–GR construct. To confirm that GR was depleted, we immunoblotted the cell lysates of THP-1control and THP-1GR-KD for GR, demonstrating a significant downregulation of GR in the THP-1GR-KD cells (Fig. 6A, left panel). Semiquantitation of the relative KD of GR by densitometry measurement of the GR immunoband revealed ∼4-fold reduction of the GR protein in the THP-1GR-KD cells compared with the THP-1control cells (Fig. 6A, right panel). To further validate the depletion of GR expression in THP-1GR-KD macrophages, RNAseq for GR mRNA in THP-1control and THP-1GR-KD cells showed that THP-1GR-KD had a 2.5-fold reduction in mRNA for GR compared with that of THP-1control cells (Fig. 6B).

FIGURE 6.

GC or AAT inhibition of NF-κB activation, inhibition of IL-8 production, and induction of angiopoietin-like 4 is GR-dependent. (A) Western blot of whole-cell lysates of THP-1control and THP-1GR-KD macrophages for GR and β-actin. Densitometry of the GR band on immunoblot normalized for β-actin (***p < 0.001 compared with Scr-shRNA lentivirus). The immunoblot and densitometry shown are representative and the mean of three independent experiments, respectively. (B) RNAseq for the GR gene (NR3C1) transcript of the THP-1control and THP-1GR-KD cells using shRNA-lentivirus technology (***p < 0.001 compared with Scr-shRNA lentivirus). (C) THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with cortisol, AAT, or both at the indicated concentrations for 30 min, and, after stimulation with LPS for 6 h, p65–NF-κB binding assay to its consensus oligonucleotide was performed. THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with cortisol, AAT, or both at the indicated concentrations for 30 min, and, after stimulation with LPS for 6 h (D) or 24 h (E), the supernatants were assayed for IL-8 by ELISA. (F) GC (cortisol or dexamethasone [DEX]) or AAT induction of ANGPTL4 in THP-1control and THP-1GR-KD macrophages. Experiments in (C) and (F) and (D) and (E) are the mean ± SEM of three and four independent experiments, respectively, with each experiment done in duplicate. *p < 0.05, **p < 0.01, ***p < 0.001. Blue bars, control shRNA (THP-1control); red bars, GR shRNA (THP-1GR-KD). LV, lentivirus; Scr, scrambled.

FIGURE 6.

GC or AAT inhibition of NF-κB activation, inhibition of IL-8 production, and induction of angiopoietin-like 4 is GR-dependent. (A) Western blot of whole-cell lysates of THP-1control and THP-1GR-KD macrophages for GR and β-actin. Densitometry of the GR band on immunoblot normalized for β-actin (***p < 0.001 compared with Scr-shRNA lentivirus). The immunoblot and densitometry shown are representative and the mean of three independent experiments, respectively. (B) RNAseq for the GR gene (NR3C1) transcript of the THP-1control and THP-1GR-KD cells using shRNA-lentivirus technology (***p < 0.001 compared with Scr-shRNA lentivirus). (C) THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with cortisol, AAT, or both at the indicated concentrations for 30 min, and, after stimulation with LPS for 6 h, p65–NF-κB binding assay to its consensus oligonucleotide was performed. THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with cortisol, AAT, or both at the indicated concentrations for 30 min, and, after stimulation with LPS for 6 h (D) or 24 h (E), the supernatants were assayed for IL-8 by ELISA. (F) GC (cortisol or dexamethasone [DEX]) or AAT induction of ANGPTL4 in THP-1control and THP-1GR-KD macrophages. Experiments in (C) and (F) and (D) and (E) are the mean ± SEM of three and four independent experiments, respectively, with each experiment done in duplicate. *p < 0.05, **p < 0.01, ***p < 0.001. Blue bars, control shRNA (THP-1control); red bars, GR shRNA (THP-1GR-KD). LV, lentivirus; Scr, scrambled.

Close modal

Because GCs have potent anti-inflammatory function via the ability of the GC–GR complex to affect inflammatory gene transcription related to NF-κB targets (24, 37), we examined the role of AAT in NF-κB activation in the context of GR signaling. We pretreated THP-1control and THP-1GR-KD macrophages with cortisol, AAT, or both for 30 min, then stimulated the cells with LPS for 6 h, and quantified the binding of p65–NF-κB to its consensus oligonucleotide. In THP-1control cells, LPS-induced p65–NF-κB binding was significantly inhibited by cortisol or AAT (Fig. 6C, blue bars). However, although LPS-induced activation of p65–NF-κB in THP-1GR-KDmacrophages was comparable to THP-1control cells (Fig. 6C, second set of red and blue bars, respectively), there was less inhibition by cortisol or AAT in the THP-1GR-KD macrophages (Fig. 6C, compare the second set of blue and red bars to their corresponding last three sets of bars). Furthermore, whereas the inhibitory effect of cortisol was nearly completely abrogated in the THP-1GR-KD macrophages (Fig. 6C, second versus third red bars), the inhibitory effects of AAT were not completely attenuated in the THP-1GR-KD macrophages (Fig. 6C, second versus fourth and fifth red bars), suggesting that AAT also affects NF-κB signaling in a GR-independent manner.

To compare LPS-induced expression of IL-8 in the supernatants of THP-1control and THP-1GR-KD macrophage cultures, both cell types were stimulated with LPS for 6 and 24 h alone or in the presence of cortisol (10 µM) or AAT (3 and 5 mg/ml) and the supernatants measured for IL-8 expression. The normal AAT level in plasma is 1–2 mg/ml but may increase 3- to 5-fold in states of systemic inflammation and/or infection (38,39). There was robust induction of IL-8 by LPS alone at 6 h (Fig. 6D) and 24 h (Fig. 6E) in both THP-1control and THP-1GR-KD macrophages, with greater amounts accumulated at 24 h. In THP-1control macrophages, there was significant inhibition of LPS-induced IL-8 expression following incubation with cortisol or 5 mg/ml AAT, particularly at 24 h (Fig. 6D, 6E, blue bars). However, in THP-1GR-KD macrophages, neither cortisol nor AAT significantly inhibited LPS-induced IL-8 production (Fig. 6D, 6E, red bars).

Either AAT or GCs induce the expression and release of ANGPTL4 in human blood monocytes and human microvascular endothelial cells (4042). Thus, THP-1control and THP-1GR-KD macrophages were stimulated with cortisol (10 µM), dexamethasone (1 µM), AAT (3 and 5 mg/ml), or both cortisol and AAT for 24 h and the supernatants quantified for ANGPTL4 (43). Stimulating THP-1control macrophages with cortisol, dexamethasone, or AAT induced ANGPTL4 production, although stimulation with AAT was not as robust as the GCs (Fig. 6F, blue bars). However, GC- or AAT-induced production of ANGPTL4 was significantly less in the THP-1GR-KD macrophages (Fig. 6F, red bars).

To determine whether the biological effects of AAT–GR interaction are also seen in primary macrophages, transient KD of GR of MDM was performed using siRNA technology. Compared to MDM that were transfected with scrambled siRNA (MDMcontrol), MDM transfected with siRNA directed against GR transcript (MDMGR-KD) revealed there was a modest but significant reduction in GR protein as determined by immunoblotting (Fig. 7A). In MDMcontrol, LPS induced IL-8, which was inhibited by either cortisol or AAT (Fig. 7B, blue bars); this inhibition was partially but significantly abrogated in MDMGR-KD cells (Fig. 7B, red bars), similar to that seen with the THP-1 cells.

FIGURE 7.

AAT inhibition of IL-8 production is GR-dependent in MDM. (A) MDM differentiated from the PBMC from a PiMM individual were transfected with scrambled siRNA (control) or GR siRNA. Cell lysates from MDMcontrol or MDMGR-KD were separated by SDS-PAGE and immunoblotted for GR. Densitometry of the GR band on the immunoblot (mean density of two independent experiments). (B) MDMcontrol and MDMGR-KD were left untreated or pretreated with cortisol (10 µM) or AAT (3 mg/ml) for 30 min, followed by stimulation with LPS for 24 h, and supernatant quantified for IL-8 by ELISA. The immunoblot and densitometry analysis of GR are representative and the mean of three independent experiments, respectively. ELISA for IL-8 is the mean of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001. Blue bars, control siRNA (MDMcontrol); red bars, GR siRNA (MDMGR-KD). Control, scrambled (Scr).

FIGURE 7.

AAT inhibition of IL-8 production is GR-dependent in MDM. (A) MDM differentiated from the PBMC from a PiMM individual were transfected with scrambled siRNA (control) or GR siRNA. Cell lysates from MDMcontrol or MDMGR-KD were separated by SDS-PAGE and immunoblotted for GR. Densitometry of the GR band on the immunoblot (mean density of two independent experiments). (B) MDMcontrol and MDMGR-KD were left untreated or pretreated with cortisol (10 µM) or AAT (3 mg/ml) for 30 min, followed by stimulation with LPS for 24 h, and supernatant quantified for IL-8 by ELISA. The immunoblot and densitometry analysis of GR are representative and the mean of three independent experiments, respectively. ELISA for IL-8 is the mean of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001. Blue bars, control siRNA (MDMcontrol); red bars, GR siRNA (MDMGR-KD). Control, scrambled (Scr).

Close modal

The consensus among clinicians and scientists is that GCs increase susceptibility to M. tuberculosis (44, 45) and possibly nontuberculous mycobacteria (NTM) (46). In contrast, AAT enhances autophagy, an effector mechanism known to kill intracellular mycobacteria (4, 4749). Because canonical GC–GR signaling and the cellular effector functions induced by AAT appear to have opposing effects in controlling mycobacterial infection, we examined in macrophages the role of the AAT–GR interaction in the control of M. tuberculosis and M. intracellulare (an NTM) infection. THP-1control and THP-1GR-KD macrophages were initially infected with M. tuberculosis H37Rv or M. intracellulare at a MOI of 10 mycobacteria to 1 macrophage for 1 h, 2 d, and 4 d (without exogenous GC or AAT added) and then the mycobacteria quantified. We detected a productive M. tuberculosis and M. intracellulare infection of the THP-1control macrophages, peaking at day 2 (Fig. 8A, top and bottom panels, blue bars), which, unexpectedly, consistently increased in the THP-1GR-KD macrophages (Fig. 8A, top and bottom panels, red bars). Compared to uninfected THP-1control and THP-1GR-KD macrophages, there was ∼10–15% reduction in viability in both cell phenotypes (as measured by the MTT assay) after 4 d of infection with either Mycobacterium (data not shown), similar to what we previously reported (50).

FIGURE 8.

Mycobacterial infection of THP-1control and THP-1GR-KD macrophages in the absence or presence of AAT or GC. (A) THP-1control and THP-1GR-KD macrophages were infected with M. tuberculosis H37Rv or M. intracellulare for 1 h, 2 d, and 4 d. The cells were washed and intracellular mycobacteria quantified. (B) THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with dexamethasone (0.1 or 1 µM) for 60 min, followed by M. tuberculosis (top panel) or M. intracellulare (bottom panel) infection for 1 h, 2 d, and 4 d. The cells were washed and intracellular mycobacteria quantified. (C) THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with 3 mg/ml AAT for 30 min, infected with M. tuberculosis (top panel) or M. intracellulare (bottom panel) for 1 h, 2 d, and 4 d, the cells washed, and intracellular mycobacteria quantified. Experiments in (A) and (B) are the mean ± SEM of three independent experiments and in (C), the mean ± SEM of six independent experiments, with each experiment done in duplicate. *p < 0.05, **p < 0.01. The numbers shown above the bars at t = 0 in (B) and (C) are the mean cell-associated CFUs at 1 h postinfection. Blue or black bars, control shRNA (THP-1control); red bars, GR shRNA (THP-1GR-KD).

FIGURE 8.

Mycobacterial infection of THP-1control and THP-1GR-KD macrophages in the absence or presence of AAT or GC. (A) THP-1control and THP-1GR-KD macrophages were infected with M. tuberculosis H37Rv or M. intracellulare for 1 h, 2 d, and 4 d. The cells were washed and intracellular mycobacteria quantified. (B) THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with dexamethasone (0.1 or 1 µM) for 60 min, followed by M. tuberculosis (top panel) or M. intracellulare (bottom panel) infection for 1 h, 2 d, and 4 d. The cells were washed and intracellular mycobacteria quantified. (C) THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with 3 mg/ml AAT for 30 min, infected with M. tuberculosis (top panel) or M. intracellulare (bottom panel) for 1 h, 2 d, and 4 d, the cells washed, and intracellular mycobacteria quantified. Experiments in (A) and (B) are the mean ± SEM of three independent experiments and in (C), the mean ± SEM of six independent experiments, with each experiment done in duplicate. *p < 0.05, **p < 0.01. The numbers shown above the bars at t = 0 in (B) and (C) are the mean cell-associated CFUs at 1 h postinfection. Blue or black bars, control shRNA (THP-1control); red bars, GR shRNA (THP-1GR-KD).

Close modal

Because depletion of basal GR increased the burden of both mycobacterial species in THP-1 macrophages, we sought to determine the effects of GCs in the control of M. tuberculosis and M. intracellulare. THP-1control and THP-1GR-KD macrophages were preincubated with 0.1 and 1 µM dexamethasone for 60 min and then infected with M. tuberculosis or M. intracellulare. At 1 h, 2 d, and 4 d postinfection, the infected macrophages were washed, lysed, and quantified for each of the mycobacterial species by counting CFUs on solid agar plates. In THP-1control macrophages, dexamethasone reduced the burden of both M. tuberculosis and M. intracellulare in a dose-dependent fashion (Fig. 8B, top and bottom panels, blue bars). However, dexamethasone had no significant effect on the number of mycobacterial CFUs in the THP-1GR-KD macrophages (Fig. 8B, top and bottom panels, red bars).

To determine the effects of AAT on M. tuberculosis and M. intracellulare infection in the context of GR, THP-1control and THP-1GR-KD macrophages were then preincubated with AAT (3 mg/ml) for 30 min, infected with M. tuberculosis or M. intracellulare, and mycobacteria quantified 1 h, 2 d, and 4 d postinfection. In THP-1control macrophages, preincubation with AAT significantly reduced the burden of both M. tuberculosis and M. intracellulare (Fig. 8C, top and bottom panels, blue bars), whereas this did not occur in THP-1GR-KD macrophages (Fig. 8C, top and bottom panels, red bars). Although the ability of dexamethasone to enhance macrophage control of M. tuberculosis (or M. intracellulare) is not consistent with the paradigm that GC use may predispose patients to tuberculosis (TB) and NTM infections, other studies have shown that GCs can induce a host-protective phenotype in innate immune cells through increased expression of TLR2 and enhanced TLR4 signaling (5154). Thus, to determine whether GC alone can induce TLR2, a pattern-recognition receptor capable of recognizing lipoproteins of mycobacteria, we immunoblotted for TLR2 in unstimulated cells or with dexamethasone stimulation in THP-1control and THP-1GR-KD macrophages. Notably, dexamethasone induced TLR2 production in the THP-1control macrophages, but its expression was markedly diminished in the THP-1GR-KD macrophages (Figs. 9A, 9B). We have diagramatically summarized our findings in (Fig. 10.

FIGURE 9.

GC induces expression of a pattern-recognition receptor. (A) THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with dexamethasone (DEX; 1 µM) for 24 h, followed by immunoblotting of the whole-cell lysates for TLR2 (top). The same membrane was stripped and immunoblotted for β-actin (bottom). Data shown are representative of three independent experiments. (B) Relative densitometric measurement of TLR2 protein normalized for β-actin. Data shown are the mean density (arbitrary units) of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001. p values for the THP-1GR-KD cells are in comparison with their corresponding THP-1control cells.

FIGURE 9.

GC induces expression of a pattern-recognition receptor. (A) THP-1control and THP-1GR-KD macrophages were left untreated or pretreated with dexamethasone (DEX; 1 µM) for 24 h, followed by immunoblotting of the whole-cell lysates for TLR2 (top). The same membrane was stripped and immunoblotted for β-actin (bottom). Data shown are representative of three independent experiments. (B) Relative densitometric measurement of TLR2 protein normalized for β-actin. Data shown are the mean density (arbitrary units) of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001. p values for the THP-1GR-KD cells are in comparison with their corresponding THP-1control cells.

Close modal

In this study, we demonstrated that AAT binds directly to the GR in THP-1 macrophages and primary human MDMs. We further showed in THP-1GR-KD macrophages that AAT–GR interaction inhibited LPS-induced NF-κB activation and IL-8 production, induced ANGPTL4 expression, and reduced M. tuberculosis and M. intracellulare burden in macrophages. Similar to that seen in THP-1 cells, the ability of cortisol or AAT to inhibit LPS-induced IL-8 was abrogated in MDMGR-KD compared with MDMcontrol. We recognize that alternative pathways, such as Akt signaling, may also inhibit both LPS-induced NF-κB activation and IL-8 production through GR (55). Because AAT-mediated inhibition on NF-κB activation was less abrogated than the inhibition by GC in the THP-1GR-KD cells (Fig. 6C), it suggests that AAT affects signaling and gene regulation in both GR-dependent and GR-independent fashion (9, 56).

Individuals with AAT deficiency have increased neutrophilic pulmonary infiltration and proclivity to neutrophil-mediated panniculitis or vasculitis due to the ability of AAT to inhibit neutrophil infiltration by two reported mechanisms: 1) the ability of the glycan moieties of AAT to bind IL-8, preventing the neutrophil chemokine from binding to its receptor (CXCR1), and 2) the ability of AAT to inhibit a disintegrin and metalloprotease 17, a disintegrin capable of cleaving cell membrane–bound FcγRIIIb bound to immune complexes, which then binds complement receptor 3 to induce cytoskeletal rearrangements necessary for neutrophil chemotaxis (18, 57). We now demonstrate a third mechanism in which AAT promotes GR-mediated inhibition of IL-8 production via antagonizing proinflammatory NF-κB activation.

Ultimately, although we have established the AAT–GR interaction and the physiologic role of this interaction in antagonizing NF-κB–mediated inflammation, promoting the transcription of target genes, and mediating antimycobacterial function (Fig. 10A), the molecular mechanism by which AAT binds GR and regulates transcription remains to be defined. Although AAT is unlikely to bind to the canonical GC binding site of the LBD of GR (although molecular modeling predicts AAT binding to the AF-2 site of LBD of GR), several alternative avenues that are not necessarily mutually exclusive may explain its role in GR function, forming the basis for further investigations (Fig. 10B). First, our co-IP findings of the nuclear and cytoplasmic fractions suggest GR may shuttle between the two compartments (Fig. 10B, 1). A possible role of AAT binding to GR may be to increase GR localization to the nuclear compartment compared with absent or lower AAT levels. This hypothesis is consistent with the acute-phase reactant nature of both cortisol and AAT in states of systemic inflammation and infection. Second, exogenous or supraphysiologic AAT may act primarily in the cytoplasm to modulate the disassembly of GR from chaperone proteins and facilitate nuclear translocation and transactivation upon binding to canonical ligands (Fig. 10B, 2). We observed GR-dependent inhibition of NF-κB activity in response to AAT but in the absence of exogenous GC. This may be due to low concentrations of GC in the cell culture conditions potentiated by AAT. A less likely possibility is that of a direct transcriptional role for AAT that facilitates translocation of GR to the nucleus, where GR activates transcription in a GC-independent manner, although each is not necessarily mutually exclusive of the other. A third possibility of the cooperation between AAT and the GC–GR is that the serpin activity of AAT may exert an indirect, stabilizing effect on some transcriptional complexes that interact with GR, which may occur via its serpin or by a yet-uncharacterized mechanism (Fig. 10B, 3).

FIGURE 10.

Illustration of the key findings and hypotheses of the AAT–GR interactions. (A) Summarizing the key findings, AAT–GR interaction inhibits NF-κB activation and IL-8 production, induces ANGPTL4, and enhances macrophage control of mycobacteria. (B) Several hypotheses of how AAT–GR interaction may affect cellular function: 1) AAT may shuttle GR between the nuclear and cytoplasmic compartments, 2) AAT may facilitate disassembly of GR–chaperone complexes, and 3) AAT may stabilize transcriptional complexes of GR–transcription factor (TF; pink structure) or locally modulate the activity of proteases.

FIGURE 10.

Illustration of the key findings and hypotheses of the AAT–GR interactions. (A) Summarizing the key findings, AAT–GR interaction inhibits NF-κB activation and IL-8 production, induces ANGPTL4, and enhances macrophage control of mycobacteria. (B) Several hypotheses of how AAT–GR interaction may affect cellular function: 1) AAT may shuttle GR between the nuclear and cytoplasmic compartments, 2) AAT may facilitate disassembly of GR–chaperone complexes, and 3) AAT may stabilize transcriptional complexes of GR–transcription factor (TF; pink structure) or locally modulate the activity of proteases.

Close modal

We performed molecular modeling of AAT–GR complexes to explore these possibilities further. AAT binding to the LBD of GR is compatible with cochaperone–GR interactions in the cytosol (Fig. 4C), suggesting that AAT may play a role in the assembly/disassembly of the cochaperone–GR complex. Intriguingly, this complex positions the RCL of AAT interacting with the AF-2 region of the LBD of GR in a similar mode to the binding of the nuclear coregulator proteins NCoR and TIF2 to GR (Fig. 4D). Although the RCL does not contain the conserved Lxxx(I/L)xxx(I/L) motif found in corepressors (58), its hydrophobic nature and flexibility are compatible with the hydrophobic AF-2 site. Conformational dynamics of the AF-2 site dictate its function as a coactivator or a corepressor site (59, 60), and this is apparent from structural comparisons of LBD–ligand complexes (Fig. 4D). Coregulator molecules binding to AF-2 tune its conformational dynamics and its cooperativity with the GC ligand binding site and thus govern downstream events; for example, the release of bound chaperones, nuclear translocation, dimerization, DNA binding, and, ultimately, transcription. Therefore, it is plausible that AAT binding to the AF-2 site may similarly achieve allosteric regulation of downstream activation and repression pathways. An alternative docking solution positions AAT at the opposite end of the LBD, again interacting via its RCL and bridging the interaction with a monomeric DBD; however, this model showed some degree of steric overlap with the second DBD formed by d-loop–mediated dimerization (Fig. 4E). This mode of interaction within the nucleus prompts speculation that AAT binding may favor monomer over dimer and thus the repression of proinflammatory genes (61). However, our finding that AAT induces ANGPTL4 expression, together with the known dimeric GR DBD that binds to the ANGPTL4 GRE (41), is more consistent with a scenario in which AAT is released upon dimerization, which can then exert localized effects, be it protease inhibition or other binding activity that may be key for gene induction. However, the flexible nature of the hinge region limits the accuracy of the modeling, and therefore, such interpretations must be made with caution. Nevertheless, overall, modeling highlights several experimentally testable hypotheses on the role of AAT in GR biology.

Our findings suggest that short-term GR activation (whether by GC or AAT) in differentiated THP-1 macrophages directs protective responses against mycobacteria. On the surface, this notion is counterintuitive, given the well-described pharmacologic role of GC in suppressing immunity, resulting in predisposition to TB or NTM infection (44, 45). However, it is important to emphasize that GC-induced susceptibility to TB or NTM infection occurs with GC doses that are significantly greater than physiologic output and for an “extended period” of time in vivo. Yet, we demonstrate in two lines of evidence that GR activation enhances THP-1 macrophage control of M. tuberculosis and M. intracellulare: 1) the downregulation of GR in the absence of exogenous AAT (or GC), resulting in increased M. tuberculosis and M. intracellulare burden in the differentiated mononuclear cells, and 2) the treatment of THP-1 macrophages with either AAT or GC reduced the burden of M. tuberculosis and M. intracellulare in a GR-dependent fashion. Published literature supports this seemingly paradoxical finding. Tükenmez et al. (62) found that different GC formulations individually decreased M. tuberculosis burden in human MDMs. Bongiovanni and coworkers (63) reported that cortisol decreased phagocytosis of M. tuberculosis (by ∼10%) with further reduction in the intracellular burden of M. tuberculosis by day 4 of culture in THP-1 macrophages. In contrast, Wang et al. (64) found that dexamethasone enhanced the growth of M. bovis bacillus Calmette-Guérin in primary mouse and human macrophages, raising the possibility that the effects of GC on macrophage control of mycobacterial infection may be mycobacterial species- and host-specific. It cannot be overemphasized that such potential host-protective effects of GCs in pure macrophage cultures are likely to be different in vivo, in which, for example, GCs in pharmacologic doses induce apoptosis in lymphocytes, subsets of which are critical in controlling mycobacterial infections. Thus, future elucidation on how GC may be exploited to maximize innate immune function while minimizing their immunosuppressive effects on lymphocytes may bring new approaches to treating mycobacterial infections.

In support of the aforementioned findings, other studies have shown that low-dose GC has immunoenhancing effects in myeloid cells (51, 53, 6567). GCs have been shown to enhance immunity by inducing TLR2 expression, especially in the presence of TNF and IL-1β (51, 52). We also showed dexamethasone induced TLR2 expression in a GR-dependent fashion (Fig. 9). TLR2 activation may also induce autophagy (68), and various polymorphisms of TLR2 have been associated with an increased risk for TB and NTM lung disease (69, 70). Zhang and Daynes (53) demonstrated that during differentiation of murine myeloid progenitors into macrophages, incubation with corticosterone increased the cellular responsiveness to LPS through increased expression of a phosphatase that inhibits PI3K/Akt with subsequent increase in TLR4 signaling. Because we found that GC inhibited LPS-induction of NF-κB activation and IL-8 production, possible explanations that may account for these disparities are: 1) host species differences accounting for variations in the macrophage phenotypic response to GC, 2) differential responses to singular GC formulations, 3) different signaling pathways that may oppose each other may be similarly affected by GC, and 4) presence of a bimodal dose-response paradigm in which lower GC doses are immunopermissive, whereas higher doses are immunosuppressive (53). Our previous findings that inhibition of NF-κB activation in macrophages (by AAT or a small-molecule inhibitor of IκBα) enhanced autophagy and clearance of M. intracellulare or M. tuberculosis infections would lend further credence to the finding that GC inhibition of NF-κB activation augments macrophage activity against mycobacteria (4, 47).

We previously showed that AAT induces autophagosome formation and maturation in mycobacteria-infected macrophages, enhancing bacterial clearance (4). But could AAT have effects on other cell types that are beneficial in the context of TB? One possibility is that AAT antagonizes the damaging effects of neutrophils in the later stages of TB (71) by: 1) sequestering IL-8 (18), inhibiting neutrophil chemotaxis (57), and, as we have shown, inhibiting IL-8 production via binding to GR, and 2) inhibiting the formation of neutrophil extracellular traps, which are elastase- and IL-8–induced extruded DNA decorated with citrullinated histones, elastase, and other proteins, meant to capture extracellular bacteria but may also contribute to tissue injury (7275). AAT has also been shown to cause decreased shedding of macrophage mannose receptors, increasing their cell surface expression and reducing soluble macrophage mannose receptors, effects that may contribute to the anti-inflammatory effects of AAT (76, 77).

A limitation of this study is that we used mainly a differentiated monocytic cell line to KD GR to study the biological significance of the AAT–GR interaction, albeit similar findings were found with MDMGR-KD in the context of cortisol or AAT inhibition of IL-8 production. In contrast, the advantage of using THP-1 cells to create a pool of immortal cells stably KD for GR is that transient KD of primary macrophages would introduce more confounding variables such as nonclonality of cells, differential epigenetics among different individuals, the influence of AAT variants and GC resistance (latter often due to genotypic differences in GR subunits), and the acute cellular stress associated with transient transfection to accomplish gene KD. Nevertheless, additional future studies using primary macrophages and accounting for the aforementioned variables to determine the biological significance of the AAT–GR interaction are likely to lead to additional mechanistic insights to this paradigm. Another limitation is that we used a laboratory strain of M. tuberculosis, as clinical strains of varying virulence may have more robust immune evasive mechanisms. Although subjects with AAT deficiency are predisposed to NTM lung disease, perhaps a third limitation is that we are unaware such individuals are also predisposed to TB, as our findings may imply. Possible reasons for this lattermost point are that AAT anomalies are uncommon in most of the TB endemic areas of the world due mostly to racial differences (i.e., clinically significant AAT variants are more common in the white population), AAT phenotype or levels are not checked in the vast majority of cases of TB, and because M. tuberculosis is relatively more virulent than NTM and can affect healthy individuals, the number of TB cases that may be associated with AAT deficiency will be a small minority of the entire TB population.

In conclusion, AAT is a highly abundant biomolecule with a well-demonstrated protective role against multiple inflammatory disorders and disease states. The AAT–GR interaction we described intersects two crucial anti-inflammatory pathways that are highly amenable to pharmacologic manipulation. In an in vitro macrophage model, we showed that AAT–GR interaction can attenuate inflammatory markers and augment macrophage control of M. tuberculosis and M. intracellulare infections. A novel implication of AAT–GR biology is an alternative or adjunct to GC therapy in treating chronic inflammation due to noninfectious or infectious causes. Indeed, comparable induction of GR responses between exogenous AAT and GC may provide novel treatment approaches, such as the development of small peptides that could serve as surrogates for AAT to manipulate GR, a key target for the widely prescribed GCs. These potential benefits of the AAT–GR interaction in macrophages must be tempered by the need to corroborate the findings in vivo.

We thank Drs. Elena Goleva, Philippa Marrack, John Kappler, and Robert Sandhaus at National Jewish Health for intellectual input into this project. We also thank Dr. Miles Pufall at the University of Iowa Carver College of Medicine for his gift of a critical reagent.

This work was supported by the National Institutes of Health Grants R03 AI139822 and R01 GM135421 (to J.R.H.) but the funds were not used for this project.

The online version of this article contains supplemental material.

The mass spectrometry proteomics data presented in this article have been submitted to the ProteomeXchange Consortium via the Proteomics Identification Database (PRIDE) partner repository (http://www.ebi.ac.uk/pride/archive/projects/PXD030989) under data set identifier PXD030989.

Abbreviations used in this article:

     
  • AAT

    α1-antitrypsin

  •  
  • AF-2

    activation function 2

  •  
  • ANGPTL4

    angiopoietin-like 4 protein

  •  
  • co-IP

    coimmunoprecipitation

  •  
  • DBD

    DNA-binding domain

  •  
  • GC

    glucocorticoid

  •  
  • GR

    glucocorticoid receptor

  •  
  • GRE

    glucocorticoid response element

  •  
  • IκBα

    NF of κ light polypeptide gene enhancer in B cells inhibitor, α

  •  
  • IP

    immunoprecipitation

  •  
  • KD

    knockdown

  •  
  • LBD

    ligand-binding domain

  •  
  • MDM

    monocyte-derived macrophage

  •  
  • MOI

    multiplicity of infection

  •  
  • MS/MS

    tandem mass spectrometry

  •  
  • MST

    microscale thermophoresis

  •  
  • NCoR

    nuclear receptor corepressor

  •  
  • NTD

    N-terminal activation function-1 domain

  •  
  • NTM

    nontuberculous mycobacteria

  •  
  • PBST

    PBS with Tween-20

  •  
  • PDB

    Protein Data Bank

  •  
  • PiMM

    protease inhibitor MM

  •  
  • RCL

    reactive center loop

  •  
  • RNAseq

    RNA sequencing

  •  
  • shRNA

    short hairpin RNA

  •  
  • siRNA

    small interfering RNA

  •  
  • TB

    tuberculosis

  •  
  • TIF2

    transcriptional intermediary factor-2

1.
de Serres
F.
,
I.
Blanco
.
2014
.
Role of alpha-1 antitrypsin in human health and disease.
J. Intern. Med.
276
:
311
335
.
2.
Parr
D. G.
,
P. G.
Guest
,
J. H.
Reynolds
,
L. J.
Dowson
,
R. A.
Stockley
.
2007
.
Prevalence and impact of bronchiectasis in alpha1-antitrypsin deficiency.
Am. J. Respir. Crit. Care Med.
176
:
1215
1221
.
3.
Strnad
P.
,
N. G.
McElvaney
,
D. A.
Lomas
.
2020
.
Alpha1-antitrypsin deficiency.
N. Engl. J. Med.
382
:
1443
1455
.
4.
Bai
X.
,
A.
Bai
,
J. R.
Honda
,
C.
Eichstaedt
,
A.
Musheyev
,
Z.
Feng
,
G.
Huitt
,
R.
Harbeck
,
B.
Kosmider
,
R. A.
Sandhaus
,
E. D.
Chan
.
2019
.
Alpha-1-antitrypsin enhances primary human macrophage immunity against non-tuberculous mycobacteria.
Front. Immunol.
10
:
1417
.
5.
Bryan
C. L.
,
K. S.
Beard
,
G. B.
Pott
,
J.
Rahkola
,
E. M.
Gardner
,
E. N.
Janoff
,
L.
Shapiro
.
2010
.
HIV infection is associated with reduced serum alpha-1-antitrypsin concentrations.
Clin. Invest. Med.
33
:
E384
E389
.
6.
Chan
E. D.
,
A. M.
Kaminska
,
W.
Gill
,
K.
Chmura
,
N. E.
Feldman
,
X.
Bai
,
C. M.
Floyd
,
K. E.
Fulton
,
G. A.
Huitt
,
M. J.
Strand
, et al
2007
.
Alpha-1-antitrypsin (AAT) anomalies are associated with lung disease due to rapidly growing mycobacteria and AAT inhibits Mycobacterium abscessus infection of macrophages.
Scand. J. Infect. Dis.
39
:
690
696
.
7.
Harbig
A.
,
M.
Mernberger
,
L.
Bittel
,
S.
Pleschka
,
K.
Schughart
,
T.
Steinmetzer
,
T.
Stiewe
,
A.
Nist
,
E.
Böttcher-Friebertshäuser
.
2020
.
Transcriptome profiling and protease inhibition experiments identify proteases that activate H3N2 influenza A and influenza B viruses in murine airway.
J. Biol. Chem.
295
:
11388
11407
.
8.
Pott
G. B.
,
K. S.
Beard
,
C. L.
Bryan
,
D. T.
Merrick
,
L.
Shapiro
.
2013
.
Alpha-1 antitrypsin reduces severity of pseudomonas pneumonia in mice and inhibits epithelial barrier disruption and pseudomonas invasion of respiratory epithelial cells.
Front. Public Health
1
:
19
.
9.
Shapiro
L.
,
G. B.
Pott
,
A. H.
Ralston
.
2001
.
Alpha-1-antitrypsin inhibits human immunodeficiency virus type 1.
FASEB J.
15
:
115
122
.
10.
Whitney
J. B.
,
M.
Asmal
,
R.
Geiben-Lynn
.
2011
.
Serpin induced antiviral activity of prostaglandin synthetase-2 against HIV-1 replication.
PLoS One
6
:
e18589
.
11.
Zhou
X.
,
Z.
Liu
,
J.
Zhang
,
J. W.
Adelsberger
,
J.
Yang
,
G. F.
Burton
.
2016
.
Alpha-1-antitrypsin interacts with gp41 to block HIV-1 entry into CD4+ T lymphocytes.
BMC Microbiol.
16
:
172
.
12.
Zhou
X.
,
L.
Shapiro
,
G.
Fellingham
,
B. M.
Willardson
,
G. F.
Burton
.
2011
.
HIV replication in CD4+ T lymphocytes in the presence and absence of follicular dendritic cells: inhibition of replication mediated by α-1-antitrypsin through altered IκBα ubiquitination.
J. Immunol.
186
:
3148
3155
.
13.
Barbey-Morel
C.
,
J. A.
Pierce
,
E. J.
Campbell
,
D. H.
Perlmutter
.
1987
.
Lipopolysaccharide modulates the expression of alpha 1 proteinase inhibitor and other serine proteinase inhibitors in human monocytes and macrophages.
J. Exp. Med.
166
:
1041
1054
.
14.
Serban
K. A.
,
I.
Petrache
.
2016
.
Alpha-1 antitrypsin and lung cell apoptosis.
Ann. Am. Thorac. Soc.
13
(
Suppl 2
):
S146
S149
.
15.
Sohrab
S.
,
D. N.
Petrusca
,
A. D.
Lockett
,
K. S.
Schweitzer
,
N. I.
Rush
,
Y.
Gu
,
K.
Kamocki
,
J.
Garrison
,
I.
Petrache
.
2009
.
Mechanism of alpha-1 antitrypsin endocytosis by lung endothelium.
FASEB J.
23
:
3149
3158
.
16.
Zhou
X.
,
Z.
Liu
,
L.
Shapiro
,
J.
Yang
,
G. F.
Burton
.
2015
.
Low-density lipoprotein receptor-related protein 1 mediates α1-antitrypsin internalization in CD4+ T lymphocytes.
J. Leukoc. Biol.
98
:
1027
1035
.
17.
Lockett
A. D.
,
D. N.
Petrusca
,
M. J.
Justice
,
C.
Poirier
,
K. A.
Serban
,
N. I.
Rush
,
M.
Kamocka
,
D.
Predescu
,
S.
Predescu
,
I.
Petrache
.
2015
.
Scavenger receptor class B, type I-mediated uptake of A1AT by pulmonary endothelial cells.
Am. J. Physiol. Lung Cell. Mol. Physiol.
309
:
L425
L434
.
18.
Bergin
D. A.
,
E. P.
Reeves
,
P.
Meleady
,
M.
Henry
,
O. J.
McElvaney
,
T. P.
Carroll
,
C.
Condron
,
S. H.
Chotirmall
,
M.
Clynes
,
S. J.
O’Neill
,
N. G.
McElvaney
.
2010
.
α-1 Antitrypsin regulates human neutrophil chemotaxis induced by soluble immune complexes and IL-8.
J. Clin. Invest.
120
:
4236
4250
.
19.
de Queiroz
T. M.
,
N.
Lakkappa
,
E.
Lazartigues
.
2020
.
ADAM17-mediated shedding of inflammatory cytokines in hypertension.
Front. Pharmacol.
11
:
1154
.
20.
Petrache
I.
,
I.
Fijalkowska
,
T. R.
Medler
,
J.
Skirball
,
P.
Cruz
,
L.
Zhen
,
H. I.
Petrache
,
T. R.
Flotte
,
R. M.
Tuder
.
2006
.
alpha-1 antitrypsin inhibits caspase-3 activity, preventing lung endothelial cell apoptosis.
Am. J. Pathol.
169
:
1155
1166
.
21.
Gans
H.
,
B. H.
Tan
.
1967
.
α-1-antitrypsin, an inhibitor for thrombin and plasmin.
Clin. Chim. Acta
17
:
111
117
.
22.
Aggarwal
N.
,
J.
Koepke
,
N.
Matamala
,
B.
Martinez-Delgado
,
M. T.
Martinez
,
H.
Golpon
,
J.
Stolk
,
S.
Janciauskiene
,
R.
Koczulla
.
2016
.
Alpha-1 antitrypsin regulates transcriptional levels of serine proteases in blood mononuclear cells.
Am. J. Respir. Crit. Care Med.
193
:
1065
1067
.
23.
Janciauskiene
S.
,
S.
Wrenger
,
S.
Immenschuh
,
B.
Olejnicka
,
T.
Greulich
,
T.
Welte
,
J.
Chorostowska-Wynimko
.
2018
.
The multifaceted effects of alpha1-antitrypsin on neutrophil functions.
Front. Pharmacol.
9
:
341
.
24.
Necela
B. M.
,
J. A.
Cidlowski
.
2004
.
Mechanisms of glucocorticoid receptor action in noninflammatory and inflammatory cells.
Proc. Am. Thorac. Soc.
1
:
239
246
.
25.
Kadiyala
V.
,
S. K.
Sasse
,
M. O.
Altonsy
,
R.
Berman
,
H. W.
Chu
,
T. L.
Phang
,
A. N.
Gerber
.
2016
.
Cistrome-based cooperation between airway epithelial glucocorticoid receptor and NF-κB orchestrates anti-inflammatory effects.
J. Biol. Chem.
291
:
12673
12687
.
26.
Ni
K.
,
M.
Umair Mukhtar Mian
,
C.
Meador
,
A.
Gill
,
D.
Barwinska
,
D.
Cao
,
M. J.
Justice
,
D.
Jiang
,
N.
Schaefer
,
K. S.
Schweitzer
, et al
2017
.
Oncostatin M and TNF-α induce alpha-1 antitrypsin production in undifferentiated adipose stromal cells.
Stem Cells Dev.
26
:
1468
1476
.
27.
John
S.
,
P. J.
Sabo
,
R. E.
Thurman
,
M. H.
Sung
,
S. C.
Biddie
,
T. A.
Johnson
,
G. L.
Hager
,
J. A.
Stamatoyannopoulos
.
2011
.
Chromatin accessibility pre-determines glucocorticoid receptor binding patterns.
Nat. Genet.
43
:
264
268
.
28.
Bai
X.
,
D.
Silvius
,
E. D.
Chan
,
D.
Escalier
,
S. X.
Xu
.
2009
.
Identification and characterization of a novel testis-specific gene CKT2, which encodes a substrate for protein kinase CK2.
Nucleic Acids Res.
37
:
2699
2711
.
29.
Jumper
J.
,
R.
Evans
,
A.
Pritzel
,
T.
Green
,
M.
Figurnov
,
O.
Ronneberger
,
K.
Tunyasuvunakool
,
R.
Bates
,
A.
Žídek
,
A.
Potapenko
, et al
2021
.
Highly accurate protein structure prediction with AlphaFold.
Nature
596
:
583
589
.
30.
Nivón
L. G.
,
R.
Moretti
,
D.
Baker
.
2013
.
A Pareto-optimal refinement method for protein design scaffolds.
PLoS One
8
:
e59004
.
31.
Konagurthu
A. S.
,
C. F.
Reboul
,
J. W.
Schmidberger
,
J. A.
Irving
,
A. M.
Lesk
,
P. J.
Stuckey
,
J. C.
Whisstock
,
A. M.
Buckle
.
2010
.
MUSTANG-MR structural sieving server: applications in protein structural analysis and crystallography.
PLoS One
5
:
e10048
.
32.
Kozakov
D.
,
D. R.
Hall
,
B.
Xia
,
K. A.
Porter
,
D.
Padhorny
,
C.
Yueh
,
D.
Beglov
,
S.
Vajda
.
2017
.
The ClusPro web server for protein-protein docking.
Nat. Protoc.
12
:
255
278
.
33.
Bai
X.
,
J. A.
Stitzel
,
A.
Bai
,
C. A.
Zambrano
,
M.
Phillips
,
P.
Marrack
,
E. D.
Chan
.
2017
.
Nicotine impairs macrophage control of Mycobacterium tuberculosis.
Am. J. Respir. Cell Mol. Biol.
57
:
324
333
.
34.
Jerabek-Willemsen
M.
,
C. J.
Wienken
,
D.
Braun
,
P.
Baaske
,
S.
Duhr
.
2011
.
Molecular interaction studies using microscale thermophoresis.
Assay Drug Dev. Technol.
9
:
342
353
.
35.
Noddings
C. M.
,
R. Y.
Wang
,
J. L.
Johnson
,
D. A.
Agard
.
2022
.
Structure of Hsp90-p23-GR reveals the Hsp90 client-remodelling mechanism.
Nature
601
:
465
469
.
36.
Bledsoe
R. K.
,
V. G.
Montana
,
T. B.
Stanley
,
C. J.
Delves
,
C. J.
Apolito
,
D. D.
McKee
,
T. G.
Consler
,
D. J.
Parks
,
E. L.
Stewart
,
T. M.
Willson
, et al
2002
.
Crystal structure of the glucocorticoid receptor ligand binding domain reveals a novel mode of receptor dimerization and coactivator recognition.
Cell
110
:
93
105
.
37.
Van Bogaert
T.
,
K.
De Bosscher
,
C.
Libert
.
2010
.
Crosstalk between TNF and glucocorticoid receptor signaling pathways.
Cytokine Growth Factor Rev.
21
:
275
286
.
38.
Hazari
Y. M.
,
A.
Bashir
,
M.
Habib
,
S.
Bashir
,
H.
Habib
,
M. A.
Qasim
,
N. N.
Shah
,
E.
Haq
,
J.
Teckman
,
K. M.
Fazili
.
2017
.
Alpha-1-antitrypsin deficiency: genetic variations, clinical manifestations and therapeutic interventions.
Mutat. Res. Rev. Mutat. Res.
773
:
14
25
.
39.
de Serres
F. J.
,
I.
Blanco
,
E.
Fernández-Bustillo
.
2003
.
Genetic epidemiology of alpha-1 antitrypsin deficiency in North America and Australia/New Zealand: Australia, Canada, New Zealand and the United States of America.
Clin. Genet.
64
:
382
397
.
40.
Frenzel
E.
,
S.
Wrenger
,
B.
Brügger
,
S.
Salipalli
,
S.
Immenschuh
,
N.
Aggarwal
,
R.
Lichtinghagen
,
R.
Mahadeva
,
A. M.
Marcondes
,
C. A.
Dinarello
, et al
2015
.
α1-Antitrypsin combines with plasma fatty acids and induces angiopoietin-like protein 4 expression.
J. Immunol.
195
:
3605
3616
.
41.
Koliwad
S. K.
,
T.
Kuo
,
L. E.
Shipp
,
N. E.
Gray
,
F.
Backhed
,
A. Y.
So
,
R. V.
Farese
Jr.
,
J. C.
Wang
.
2009
.
Angiopoietin-like 4 (ANGPTL4, fasting-induced adipose factor) is a direct glucocorticoid receptor target and participates in glucocorticoid-regulated triglyceride metabolism.
J. Biol. Chem.
284
:
25593
25601
.
42.
Nakamoto
M.
,
K.
Ishihara
,
T.
Watanabe
,
A.
Hirosue
,
S.
Hino
,
M.
Shinohara
,
H.
Nakayama
,
M.
Nakao
.
2017
.
The glucocorticoid receptor regulates the ANGPTL4 gene in a CTCF-mediated chromatin context in human hepatic cells.
PLoS One
12
:
e0169225
.
43.
Yin
W.
,
S.
Romeo
,
S.
Chang
,
N. V.
Grishin
,
H. H.
Hobbs
,
J. C.
Cohen
.
2009
.
Genetic variation in ANGPTL4 provides insights into protein processing and function.
J. Biol. Chem.
284
:
13213
13222
.
44.
Jick
S. S.
,
E. S.
Lieberman
,
M. U.
Rahman
,
H. K.
Choi
.
2006
.
Glucocorticoid use, other associated factors, and the risk of tuberculosis.
Arthritis Rheum.
55
:
19
26
.
45.
Lai
C.-C.
,
M.-T. G.
Lee
,
S.-H.
Lee
,
S.-H.
Lee
,
S.-S.
Chang
,
C.-C.
Lee
.
2015
.
Risk of incident active tuberculosis and use of corticosteroids.
Int. J. Tuberc. Lung Dis.
19
:
936
942
.
46.
Xie
Y.
,
J.
Xie
,
A. H.
Meijer
,
M. J. M.
Schaaf
.
2021
.
Glucocorticoid-induced exacerbation of mycobacterial infection is associated with a reduced phagocytic capacity of macrophages.
Front. Immunol.
12
:
618569
.
47.
Bai
X.
,
N. E.
Feldman
,
K.
Chmura
,
A. R.
Ovrutsky
,
W.-L.
Su
,
L.
Griffin
,
D.
Pyeon
,
M. T.
McGibney
,
M. J.
Strand
,
M.
Numata
, et al
2013
.
Inhibition of nuclear factor-kappa B activation decreases survival of Mycobacterium tuberculosis in human macrophages.
PLoS One
8
:
e61925
.
48.
Deretic
V.
,
S.
Singh
,
S.
Master
,
J.
Harris
,
E.
Roberts
,
G.
Kyei
,
A.
Davis
,
S.
de Haro
,
J.
Naylor
,
H.-H.
Lee
,
I.
Vergne
.
2006
.
Mycobacterium tuberculosis inhibition of phagolysosome biogenesis and autophagy as a host defence mechanism.
Cell. Microbiol.
8
:
719
727
.
49.
Gutierrez
M. G.
,
S. S.
Master
,
S. B.
Singh
,
G. A.
Taylor
,
M. I.
Colombo
,
V.
Deretic
.
2004
.
Autophagy is a defense mechanism inhibiting BCG and Mycobacterium tuberculosis survival in infected macrophages.
Cell
119
:
753
766
.
50.
Bai
X.
,
S.-H.
Kim
,
T.
Azam
,
M. T.
McGibney
,
H.
Huang
,
C. A.
Dinarello
,
E. D.
Chan
.
2010
.
IL-32 is a host protective cytokine against Mycobacterium tuberculosis in differentiated THP-1 human macrophages.
J. Immunol.
184
:
3830
3840
.
51.
Busillo
J. M.
,
J. A.
Cidlowski
.
2013
.
The five Rs of glucocorticoid action during inflammation: ready, reinforce, repress, resolve, and restore.
Trends Endocrinol. Metab.
24
:
109
119
.
52.
Chinenov
Y.
,
I.
Rogatsky
.
2007
.
Glucocorticoids and the innate immune system: crosstalk with the toll-like receptor signaling network.
Mol. Cell. Endocrinol.
275
:
30
42
.
53.
Zhang
T. Y.
,
R. A.
Daynes
.
2007
.
Glucocorticoid conditioning of myeloid progenitors enhances TLR4 signaling via negative regulation of the phosphatidylinositol 3-kinase-Akt pathway.
J. Immunol.
178
:
2517
2526
.
54.
Bansal
A.
,
M. M.
Mostafa
,
C.
Kooi
,
S. K.
Sasse
,
A. N.
Michi
,
S. V.
Shah
,
R.
Leigh
,
A. N.
Gerber
,
R.
Newton
.
2022
.
Interplay between nuclear factor-κB, p38 MAPK, and glucocorticoid receptor signaling synergistically induces functional TLR2 in lung epithelial cells.
J. Biol. Chem.
298
:
101747
.
55.
Hattori
Y.
,
S.
Hattori
,
K.
Kasai
.
2003
.
Lipopolysaccharide activates Akt in vascular smooth muscle cells resulting in induction of inducible nitric oxide synthase through nuclear factor-kappa B activation.
Eur. J. Pharmacol.
481
:
153
158
.
56.
Chan
E. D.
,
G. B.
Pott
,
P. E.
Silkoff
,
A. H.
Ralston
,
C. L.
Bryan
,
L.
Shapiro
.
2012
.
Alpha-1-antitrypsin inhibits nitric oxide production.
J. Leukoc. Biol.
92
:
1251
1260
.
57.
McCarthy
C.
,
E. P.
Reeves
,
N. G.
McElvaney
.
2016
.
The role of neutrophils in alpha-1 antitrypsin deficiency.
Ann. Am. Thorac. Soc.
13
(
Suppl 4
):
S297
S304
.
58.
Frank
F.
,
E. A.
Ortlund
,
X.
Liu
.
2021
.
Structural insights into glucocorticoid receptor function.
Biochem. Soc. Trans.
49
:
2333
2343
.
59.
Köhler
C.
,
G.
Carlström
,
A.
Gunnarsson
,
U.
Weininger
,
S.
Tångefjord
,
V.
Ullah
,
M.
Lepistö
,
U.
Karlsson
,
T.
Papavoine
,
K.
Edman
,
M.
Akke
.
2020
.
Dynamic allosteric communication pathway directing differential activation of the glucocorticoid receptor.
Sci. Adv.
6
:
eabb5277
.
60.
Schoch
G. A.
,
B.
D’Arcy
,
M.
Stihle
,
D.
Burger
,
D.
Bär
,
J.
Benz
,
R.
Thoma
,
A.
Ruf
.
2010
.
Molecular switch in the glucocorticoid receptor: active and passive antagonist conformations.
J. Mol. Biol.
395
:
568
577
.
61.
Hudson
W. H.
,
C.
Youn
,
E. A.
Ortlund
.
2013
.
The structural basis of direct glucocorticoid-mediated transrepression.
Nat. Struct. Mol. Biol.
20
:
53
58
.
62.
Tükenmez
H.
,
I.
Edström
,
S.
Kalsum
,
C.
Braian
,
R.
Ummanni
,
S. B.
Fick
,
C.
Sundin
,
M.
Lerm
,
M.
Elofsson
,
C.
Larsson
.
2019
.
Corticosteroids protect infected cells against mycobacterial killing in vitro.
Biochem. Biophys. Res. Commun.
511
:
117
121
.
63.
Bongiovanni
B.
,
D.
Mata-Espinosa
,
L.
D’Attilio
,
J. C.
Leon-Contreras
,
R.
Marquez-Velasco
,
O.
Bottasso
,
R.
Hernandez-Pando
,
M. L.
Bay
.
2015
.
Effect of cortisol and/or DHEA on THP1-derived macrophages infected with Mycobacterium tuberculosis.
Tuberculosis (Edinb.)
95
:
562
569
.
64.
Wang
J.
,
R.
Wang
,
H.
Wang
,
X.
Yang
,
J.
Yang
,
W.
Xiong
,
Q.
Wen
,
L.
Ma
.
2017
.
Glucocorticoids suppress antimicrobial autophagy and nitric oxide production and facilitate mycobacterial survival in macrophages.
Sci. Rep.
7
:
982
.
65.
Barber
A. E.
,
S. M.
Coyle
,
M. A.
Marano
,
E.
Fischer
,
S. E.
Calvano
,
Y.
Fong
,
L. L.
Moldawer
,
S. F.
Lowry
.
1993
.
Glucocorticoid therapy alters hormonal and cytokine responses to endotoxin in man.
J. Immunol.
150
:
1999
2006
.
66.
Fantuzzi
G.
,
G.
Galli
,
M.
Zinetti
,
M.
Fratelli
,
P.
Ghezzi
.
1995
.
The upregulating effect of dexamethasone on tumor necrosis factor production is mediated by a nitric oxide-producing cytochrome P450.
Cell. Immunol.
160
:
305
308
.
67.
Smyth
G. P.
,
P. P.
Stapleton
,
T. A.
Freeman
,
E. M.
Concannon
,
J. R.
Mestre
,
M.
Duff
,
S.
Maddali
,
J. M.
Daly
.
2004
.
Glucocorticoid pretreatment induces cytokine overexpression and nuclear factor-kappaB activation in macrophages.
J. Surg. Res.
116
:
253
261
.
68.
Seto
S.
,
K.
Tsujimura
,
Y.
Koide
.
2012
.
Coronin-1a inhibits autophagosome formation around Mycobacterium tuberculosis-containing phagosomes and assists mycobacterial survival in macrophages.
Cell. Microbiol.
14
:
710
727
.
69.
Bento
C. F.
,
N.
Empadinhas
,
V.
Mendes
.
2015
.
Autophagy in the fight against tuberculosis.
DNA Cell Biol.
34
:
228
242
.
70.
Yim
J.-J.
,
H. J.
Kim
,
O. J.
Kwon
,
W.-J.
Koh
.
2008
.
Association between microsatellite polymorphisms in intron II of the human Toll-like receptor 2 gene and nontuberculous mycobacterial lung disease in a Korean population.
Hum. Immunol.
69
:
572
576
.
71.
Lowe
D. M.
,
P. S.
Redford
,
R. J.
Wilkinson
,
A.
O’Garra
,
A. R.
Martineau
.
2012
.
Neutrophils in tuberculosis: friend or foe?
Trends Immunol.
33
:
14
25
.
72.
de Melo
M. G. M.
,
E. D. D.
Mesquita
,
M. M.
Oliveira
,
C.
da Silva-Monteiro
,
A. K. A.
Silveira
,
T. S.
Malaquias
,
T. C. P.
Dutra
,
R. M.
Galliez
,
A. L.
Kritski
,
E. C.
Silva
;
Rede-TB Study Group
.
2019
.
Imbalance of NET and alpha-1-antitrypsin in tuberculosis patients is related with hyper inflammation and severe lung tissue damage.
Front. Immunol.
9
:
3147
.
73.
Frenzel
E.
,
E.
Korenbaum
,
J.
Hegermann
,
M.
Ochs
,
J.
Koepke
,
A. R.
Koczulla
,
T.
Welte
,
T.
Köhnlein
,
S.
Janciauskiene
.
2012
.
Does augmentation with alpha1-antitrypsin affect neutrophil extracellular traps formation?
Int. J. Biol. Sci.
8
:
1023
1025
.
74.
Wartha
F.
,
K.
Beiter
,
S.
Normark
,
B.
Henriques-Normark
.
2007
.
Neutrophil extracellular traps: casting the NET over pathogenesis.
Curr. Opin. Microbiol.
10
:
52
56
.
75.
Dömer
D.
,
T.
Walther
,
S.
Möller
,
M.
Behnen
,
T.
Laskay
.
2021
.
Neutrophil extracellular traps activate proinflammatory functions of human neutrophils.
Front. Immunol.
12
:
636954
.
76.
Serban
K. A.
,
D. N.
Petrusca
,
A.
Mikosz
,
C.
Poirier
,
A. D.
Lockett
,
L.
Saint
,
M. J.
Justice
,
H. L.
Twigg
III
,
M. A.
Campos
,
I.
Petrache
.
2017
.
Alpha-1 antitrypsin supplementation improves alveolar macrophages efferocytosis and phagocytosis following cigarette smoke exposure.
PLoS One
12
:
e0176073
.
77.
Embgenbroich
M.
,
H. J. P.
van der Zande
,
L.
Hussaarts
,
J.
Schulte-Schrepping
,
L. R.
Pelgrom
,
N.
García-Tardón
,
L.
Schlautmann
,
I.
Stoetzel
,
K.
Händler
,
J. M.
Lambooij
, et al
2021
.
Soluble mannose receptor induces proinflammatory macrophage activation and metaflammation.
Proc. Natl. Acad. Sci. USA
118
:
e2103304118
.

A.N.G. holds equity in Psammiad Therapeutics. The other authors have no financial conflicts of interest.

Supplementary data