Common reagents were from Sigma-Aldrich (St. Louis, MO, USA). Alcohol or ethanol (200 proof, absolute ACS/USP Grade) was purchased from Pharmco Products Inc. (Shelbyville, KY, USA). Radioactive γ-³²P-ATP was obtained from Perkin Elmer (Billerica, MA, USA). All oligonucleotides were ordered from Integrated DNA Technologies (Coralville, IA, USA) and their sequences are displayed in Table 1.

A549 and Mono Mac 6 (MM6) cells were, respectively, cultured in Dulbecco’s Modified Eagle Medium (Gibco) and advanced RPMI-1640 (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) (HyClone, Logan, UT, USA), 2 mM GlutaMax (Gibco), 100 U/ml penicillin, 100 µg/ml streptomycin, and 0.25 µg/ml amphotericin B. Forty-eight hours before and during experiments, fresh media with 10% charcoal-stripped FBS (Atlanta Biological) were used to avoid GC interference in experiments. For alcohol exposure, various doses (0, 25, or 50 mM) of alcohol were added to the culture media, respectively, and the cells were cultured in the incubators presaturated with the corresponding levels of alcohol.

MM6 cells (1.5 × 10⁶) were exposed to alcohol (0, 25, or 50 mM) or 1 µM Dexamethasone (Dex) (Sigma-Aldrich) for 24 h in the presence or absence of 5 µM mifepristone (Sigma-Aldrich). The cells were harvested and washed with 1× PBS twice. Total RNAs were extracted using the Qiagen RNeasy Kit. The cDNAs were synthesized using the Quantitect Reverse Transcriptase Kit (Qiagen). Human GILZ primers (GILZ-F and GILZ-R) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) primers (GAPDH-F and GAPDH-R) were used at a final concentration of 500 nM. The final reaction for each sample was brought to a total volume of 20 µl with RT SYBR green qPCR mastermixes (Qiagen). All reactions were carried out in duplicate on a CFX96 system (Bio-Rad Laboratories, Hercules, CA, USA) for quantitative real-time PCR (qPCR) detection. The qPCR data were analyzed by the comparative Ct (ΔΔCT) method. The expression of GILZ of each treated group was compared to that of GAPDH and normalized to the non-treatment group.

The gilz promoter was PCR-amplified from the genomic DNA of primary human airway epithelial cells (Lonza Walkersville, Inc., Walkersville, MA, USA) using the primers (Long-GILZp-F and GILZp-R). The amplicon was cloned into pCR™-Blunt II-TOPO^® plasmid (Invitrogen), and the primary sequence was verified by DNA sequencing. Then, the obtained 1.94-Kb promoter was excised with KpnI/HindIII and subcloned into the promoter-less pGL.4.16[luc2CP/Hygro] plasmid (Promega, Madison, WI, USA), resulting in pGL.4.16-wtGILZ-2p-[luc2CP/Hygro] in which the GILZ promoter drives the expression of the rapid-response firefly luciferase (Fluc) gene. The cloned promoter contains five GREs. To remove GRE3-5, the region (−1,560 to −1,884) was deleted using the primers (Short-GILZp-F and GILZp-R). To achieve site-specific mutagenesis of the GRE1 and GRE2 core sequences, QuikChange II Site-Directed Mutagenesis Kit (Agilent Technologies) was used, as instructed by the manufacturer. Synthetic oligonucleotide primers (Mut-GRE1-F, Mut-GRE1-R, Mut-GRE2-F, and Mut-GRE2-R) were used to introduce mutations. All the plasmids with mutagenesis were verified by DNA sequencing.

Transfection was performed with Lipofectamine 2000 Transfection Reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instruction. Each of the constructed plasmids was cotransfected into A549 cells with the reference plasmid pGL4.73[hRluc/SV40] (Promega) that expresses Renilla luciferase. Forty-eight hours later, the transfected cells were exposed to 0 or 50 mM of alcohol or 1 µM Dex for 24 h. Activities of Firefly luciferase and Renilla luciferase were measured using the dual luciferase assay system (Promega). Potency of the wild-type or mutant gilz promoter to drive the reporter luciferase expression was determined by ratio of the activity of Firefly luciferase over that of Renilla luciferase.

Double-stranded oligonucleotides for GRE1 and its flanking region and for GRE2 and its flanking region were prepared and end-labeled with ³²P using T4 polynucleotide kinase. GMSA was performed by incubation of recombinant human GR protein (40 ng or 0.5 pmol) (Sigma-Aldrich), ³²P-labeled oligonucleotides (0.5 ng or 0.05 pmol) and 1 µg poly-deoxyinosinic-deoxycytidylic acid (Sigma-Aldrich), in a final volume of 25 µl for each reaction. The GR-GRE complex was resolved by non-denature polyacrylamide gel electrophoresis and detected by X-ray film exposure.

Lentiviral vectors expressing either siGILZ or siCNTL transgene were constructed as previously described (37) and were produced via triple-plasmid cotransfection of HEK 293T cells (41). Transduction of MM6 cells was carried out at MOI of 5 in the presence of Polybrene (Sigma-Aldrich, 8 µg/ml). Because the lentiviral vectors had an EGFP marker gene in an independent cassette, each set of vector-transduced cells was sorted through multiple rounds to a pure population.

The S. pyogenes Cas9 plasmid pSpCas9(BB)-2A-GFP (PX458) (Addgene plasmid # 48138) was a gift from Dr. Feng Zhang at the Massachusetts Institute of Technology. A target-specific sgRNA sequence for Exon 3 of human GR gene was identified using the CRISPR MultiTargeter¹ and the Zhang lab target finder² online tools. The target sgRNA sequence was cloned into the Cas9 vectors according to the Zhang lab protocol (42). In brief, two single-stranded DNA oligos were designed for the sgRNA target (oligo 1: 5′-CACCGCATTAT-GGAGTCTTAACTTG-3′; oligo 2: 5′-AAACCAAGTTAAGACTCCATAATGC-3′). The pair of oligos was phosphorylated and annealed with T4 Polynucleotide Kinase (NEB). The Cas9 plasmids were digested with BbsI (NEB) and then ligated with the annealed oligos using NEB Quick Ligase. The final constructed plasmids were sequenced for verification. MM6 cells (1.4 × 10⁶) in 300 µl of RPMI 1640 media were combined with 20 µg of Cas9-GFP-GR plasmid, incubated on ice for 20 min and electroporated via exponential decay pulse (200 V, 975 μF) using a Bio-Rad Gene Pulser Xcell system. Following electroporation, cells were incubated on ice again for 20 min and cultured in complete RPMI 1640 overnight. Then, the cells were sorted for GFP-positive population using FACS. The sorted cells were plated on 96-well culture plates at single-cell density. After 7–14 days of growth, cell clones were screened by PCR using GR-E3(+) and GR-E3(−) primers and DNA sequencing to confirm GR gene knockout.

For nuclear GR measurement, MM6 cells were exposed to alcohol (0 or 50 mM) or Dex (1 µM) for 24 h in the presence or absence of fomepizole (4 µM). The cell nuclei were isolated using Thermo Scientific™ NE-PER™ Nuclear and Cytoplasmic Extraction Reagents in accordance with the manufacturer’s instruction. The samples were then subjected to Western blot. Rabbit polyclonal antibodies against human GR (Santa Crus Technology, Santa Cruz, CA, USA) and Histone H3 (Cell Signaling Technology, Danvers, MA, USA) were used. Intensities of the GR and Histone H3 bands were estimated by densitometry using NIH Image-J system.

For GILZ measurement, parental MM6, siCNTL-MM6, and siGILZ-MM6 cells were, respectively, treated with 0 or 1 µM Dex for 24 h. Cells were collected, harvested, and lysed with 1× RIPA buffer (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 1× protease inhibitor cocktail (Roche, Indianapolis, IN, USA). The cell lysates were collected and subjected to BCA protein assay (Thermo Fisher Scientific, Waltham, MA, USA) for protein concentration. Proteins (50 µg) from each sample were loaded onto 10–20% Tris–Tricine precast gel (Bio-Rad, Hercules, CA, USA) in the Mini-Protein system (Bio-Rad) for electrophoresis. Rabbit polyclonal anti-human GILZ antibody at 1:200 final dilution (FL-134, Santa Cruz Biotechnology) and rabbit anti-β-Actin antibody at 1:5,000 final dilution (Cell Signaling Technology, Danvers, MA, USA) were used to probe GILZ and actin.

Different genotypes of MM6 cells, as indicated in each experiment, were exposed to alcohol (0 or 50 mM) 1 h before stimulation with 100 or 1,000 ng/ml LPS from Escherichia coli 0111:B4 (Sigma-Aldrich). The supernatants were collected at different time points. TNF-α was measured by ELISA (R&D Systems) following the manufacturer’s recommendations.

Statistical analysis was performed with Prism 7.0 (GraphPad Software, La Jolla, CA, USA). A 2-tailed, unpaired Student’s t-test was used for statistical comparisons. All the data are expressed as mean ± SD. Difference with p-value smaller than 0.05 is considered statistically significant.

Classical activation of GR signaling is initiated by GC binding to GR that resides in the cytoplasm in complex with various chaperone proteins. Then, GR undergoes a conformational change, dissociates from the complex, and translocates to the nucleus to modulate gene transcription (43). However, our previously published data demonstrated that alcohol upregulates gilz, a prominent GC-responsive gene, in the absence of GC (37), suggesting an alternative GR-activation mechanism under alcohol exposure. To delineate the mechanism, we exposed MM6 cells to various concentrations of alcohol (0, 25, or 50 mM) or dexamethasone (Dex, 1 µM) for 24 h in the presence or absence of the GC competitor mifepristone (5 µM). Transcription of gilz was assessed by quantitative real-time-PCR. As shown in Figure 1, alcohol (50 mM) significantly upregulated gilz by ~2-fold. Dex increased gilz expression by ~27-fold. Importantly, mifepristone completely blocked Dex-activation of gilz, but had no effect on alcohol-activation of gilz. Because Dex is a synthetic GC that activates GR signaling as a GR agonist, while mifepristone a potent competitive GR antagonist, the data indicate that the mechanism for alcohol activation of GR is different from that for GC activation of GR.

To define the cis-acting elements that are responsible for alcohol activation of gilz, we cloned the 1.94-kb promoter region of gilz. Primary sequence analysis by the CISTER algorithm method (44) revealed multiple clusters of cis-elements, including GRE, cAMP response element (CRE), Forkhead responsive element (FHRE-1), serum response element (SRE), signaling transducer and activator of transcription 6 (STAT-6), GATA Box (GATA), and TATA box (TATA) (Figure S1 in Supplementary Material). Noticeably, there are five GREs, three of which (GRE3, 4, and 5) are relatively distal, and the other two (GRE1 and 2) proximal to the transcriptional initiation site defined by a previous study (45).

To compare Dex-activation of gilz with alcohol activation of gilz, we subcloned the 1.94-kb gilz promoter into the promoter-less pGL.4.16[luc2CP/Hygro] construct, generating the pGL-Long-GILZp-Fluc plasmid (Figure 2A), in which the cloned promoter was designed to regulate the expression of the rapid-response firefly luciferase (Fluc) gene. Because GRE3, 4, and 5 were located closely, we decided to remove them altogether by deletion of the region (−1,560 to −1,884), resulting in a shorter gilz promoter and the pGL-Short-GILZp-Fluc plasmid. On the basis of this short promoter construct, we further site-mutated the proximal GRE1 and GRE2 core sequences, producing the pGL-Short-mutGILZp-Luc plasmid (Figure 2A) that had lost all five GREs. A549 cells were, respectively, transfected with one of the three generated plasmids in combination with the reference plasmid pGL4.75-hRluc that constitutively expressed the humanized Renilla luciferase. Forty-eight hours after transfection, the cells were exposed to 0 or 50 mM alcohol or 1 µM of Dex for 24 h. Potency of the wild-type or the mutated gilz promoter to drive the reporter gene expression was determined by the activity of the Firefly luciferase that had been normalized to that of the Renilla luciferase in each sample. The data (Figure 2B) demonstrate that the promoter-less pGL.4.16-[luc2CP/Hygro] vector (pGL-vector) did not respond to either Dex or alcohol stimulation. However, the long wild-type gilz promoter with all five GREs (pGL-Long-GILZp) faithfully responded to Dex and alcohol, displaying ~2.3- and ~1.6-fold higher firefly luciferase activity than the non-treatment control. Deletion of GRE3, 4, and 5 (pGL-Short-GILZp) significantly reduced the gilz promoter responsiveness to alcohol but not Dex. Further mutations of GRE1 and GRE2 (pGL-Short-mut-GILZp) totally abolished the responsiveness of the promoter to both Dex and alcohol. These data suggest that alcohol bypasses GC to activate gilz promoter through the GRE cis-elements, and that all five GREs appear to be needed for a full action of alcohol. In contrast, the GRE1 and GRE2 are sufficient to convey the full effect of Dex.

Previous publications from our laboratory and others have reported that alcohol induces nuclear translocation of non-ligand-bound GR (35, 36). However, it is unknown whether this effect is caused by alcohol itself or alcohol metabolites. We exposed MM6 cells to alcohol (0 or 50 mM) for 24 h in the presence or absence of the alcohol dehydrogenase (ADH) inhibitor fomepizole (4 µM). The levels of nuclear GR and Histone H3 for each condition were assessed by Western blot assay and densitometry quantification (Figure 3A). The ratio of nuclear GR over Histone H3 for each condition was calculated. Data from three separate experiments (Figure 3B) show that alcohol exposure increased nuclear GR as compared to that of the non-alcohol control. Strikingly, fomepizole enhanced GR nuclear translocation. Thus, alcohol-induced GR nuclear translocation is likely the effect of alcohol molecules instead of alcohol-derivatives.

Alcohol induces GR–GRE interaction in the absence of GC, predicting that non-ligand-bound GR can bind GREs. To prove the prediction, we synthesized and radioactively labeled the double-stranded oligonucleotides covering the core and flanking sequences of GRE1 and GRE2 (Figure 4A). Similarly, the corresponding core-sequence-mutated GRE1 and GRE2 oligonucleotides were also prepared. GMSA was undertaken by incubation of human GR protein with the radioactive oligonucleotides. Non-denaturing polyacrylamide gel electrophoresis revealed that human GR protein without GC coupling formed a GR–GRE complex with the wild-type GRE1 and GRE2 oligonucleotides, but not the mutant counterparts (Figure 4B). This result proves that non-ligand-bound GR is able to interact with GRE in the absence of ligands.

We further tested if the interaction between the unliganded-GR and GRE could withstand alcohol at clinically encountered levels. Human GR protein was incubated with the wild-type GRE1 or GRE2 probe in the presence of alcohol (0, 25, 50, or 100 mM). GMSA revealed that the non-canonical GR-GRE interaction was strong enough to withstand all the tested levels of alcohol (Figure 4C).

To test if gilz gene activation by alcohol through the GR-GRE interaction has any functional alteration, we created a GR−/− MM6 cell line by CRISPR/Cas9 gene targeting (Figure 5A), in which a 94-bp insertion mutation was made to GR Exon 3 (Figure S2 in Supplementary Material). Such an insertion led to a premature stop of GR translation and a truncated GR (Figure S3 in Supplementary Material). Then, both GR−/− and GR+/+MM6 cells were exposed to either 0 or 50 mM alcohol, followed by LPS challenge. Twenty-four hours after alcohol exposure, the medium from each condition was collected. Because TNF-α is the key hallmark cytokine induced by LPS (46, 47), we elected to use TNF-α as the representative cytokine to indicate the status of the cell inflammatory response. ELISA measurement of TNF-α revealed that without alcohol exposure, both genotypes of MM6 cells produced comparable levels of TNF-α after LPS stimulation (Figure 5B). Alcohol substantially decreased TNF-α production by GR+/+ cells as compared to their non-alcohol control. However, the alcohol suppressive effect on GR−/− cells was significantly attenuated, suggesting GR without ligand coupling is involved in alcohol suppression of cell inflammatory response. To confirm the fidelity of TNF-α in reflecting cell inflammatory response in our experimental system, we were asked to check more inflammatory genes. Toward this end, we checked three more cytokines: IL-1β, IL-6, and IL-10. Like TNF-α, IL-1β, and IL-6 are proinflammatory, while IL-10 is anti-inflammatory. The results demonstrate that alcohol exposure substantially suppressed proinflammatory cytokine (IL-1β and IL-6) production and substantially promoted anti-inflammatory cytokine production by GR+/+ cells. Strikingly, such alcohol effects were diminished in GR−/− cells (Figure S4 in Supplementary Material). These data confirm that TNF-α is a reliable marker to attest inflammatory response in our system and that GR is crucial in alcohol modulation of cell inflammatory response.

We further hypothesized that if GR-activation of gilz is critical to alcohol suppression of cell inflammatory response, depletion of GILZ should resemble deletion of GR, resulting in abrogation of the alcohol suppressive effect. Taking advantage of the already-made GILZ-knockdown (siGILZ) and control (siCNTL) lentiviral vectors (37), we transduced MM6 cells and created a pair of stable cell lines with expression of a normal level of GILZ (siCNTL-MM6) or a diminished level of GILZ (siGILZ-MM6). To verify GILZ expression in these cells, we performed Western blot assay on the parental MM6, siCNTL-MM6, and siGILZ-MM6 cells stimulated with or without Dex. Because GILZ is a highly GC-responsive gene, Dex application here was to maximally stimulate GILZ expression for the most stringent test of GILZ knockdown. The data (Figure 6A) demonstrate that GILZ (17 kDa) was highly induced by Dex (1 µM) in the parental MM6 and siCNTL-MM6 cells. However, the siGILZ-MM6 cells failed to express any detectable level of GILZ, regardless of Dex induction, indicating that the siRNA effectively eradicates GILZ expression in the GILZ-depleted cells.

To examine whether GILZ depletion in the MM6 cells affects alcohol suppression of the cell inflammatory response, we exposed the parental MM6, siCNTL-MM6, and siGILZ-MM6 cells to 0 or 50 mM alcohol, respectively, 1 h before challenge with LPS at either 100 ng/ml or 1 µg/ml dose. Twenty-four hours after alcohol exposure, the medium from each condition was collected for TNF-α measurement. As shown (Figure 6B), all three types of cells responded to LPS by producing abundant TNF-α in a LPS dose-dependent fashion. Interestingly, GILZ depletion rendered the siGILZ-MM6 cells more inflammatory, producing more TNF-α as compared to the parental MM6 and siCNTL-MM6 cells under an equal level of stimulation. Alcohol exposure at 50 mM substantially suppressed TNF-α production by the parental MM6 and siCNTL-MM6 cells. However, the alcohol-suppressive effect was abolished in the GILZ-depleted cells (siGILZ-MM6), suggesting that GILZ is the major player involved in alcohol suppression of cell inflammatory response to LPS.

Based on the obtained data, we put forward a working model for alcohol activation of GR signaling, as displayed (Figure 7). Conventional GC-activation of GR signaling starts from GR binding by GC ligands, e.g., cortisol. The GR–GC interaction unleashes GR from the confines of chaperone proteins, such as various heat-shock proteins (HSPs), p23, and others. Then, the GC-coupled GR migrates to the cell nucleus to modulate the expressions of GC-responsive genes. In contrast, alcohol interferes with the stability of the GR complex, from which the unliganded GR is released. Such a non-ligand-bound GR is able to interact with GREs in the promoters of GR-targeting genes including gilz. As compared to the conventional GR activation, the alcohol-triggered GR signaling does not involve GCs. This represents a novel non-canonical mechanism for GR activation by alcohol, which may play a pivotal role in alcohol anti-inflammation and immunosuppression.