Although it is well recognized that alcohol abuse impairs alveolar macrophage immune function and renders patients susceptible to pneumonia, the mechanisms are incompletely understood. Alveolar macrophage maturation and function requires priming by GM-CSF, which is produced and secreted into the alveolar space by the alveolar epithelium. In this study, we determined that although chronic ethanol ingestion (6 wk) in rats had no effect on GM-CSF expression within the alveolar space, it significantly decreased membrane expression of the GM-CSF receptor in alveolar macrophages. In parallel, ethanol ingestion decreased cellular expression and nuclear binding of PU.1, the master transcription factor that activates GM-CSF-dependent macrophage functions. Furthermore, treatment of ethanol-fed rats in vivo with rGM-CSF via the upper airway restored GM-CSF receptor membrane expression as well as PU.1 protein expression and nuclear binding in alveolar macrophages. Importantly, GM-CSF treatment also restored alveolar macrophage function in ethanol-fed rats, as reflected by endotoxin-stimulated release of TNF-α and bacterial phagocytosis. We conclude that ethanol ingestion dampens alveolar macrophage immune function by decreasing GM-CSF receptor expression and downstream PU.1 nuclear binding and that these chronic defects can be reversed relatively quickly with rGM-CSF treatment in vivo.
For over a century, alcohol abuse has been well recognized as a significant risk factor for serious pulmonary infections. For example, alcoholic patients are at increased risk for infection with necrotizing Gram-negative pathogens such as Klebsiella pneumoniae (1) or to develop bacteremia and shock from typical pathogens, most notably Streptococcus pneumoniae (2). The mechanisms by which alcohol abuse increases the risk of pneumonia are likely multiple and include increased risk of aspiration of oropharyngeal flora, decreased mucociliary clearance of bacterial pathogens from the upper airway, and impaired pulmonary host defenses. Perhaps the most prominent effects on host defense involve the alveolar macrophage, the first cellular line of defense against pathogens within the lower airways. In experimental models, chronic ethanol ingestion suppresses chemokine responses and pathogen clearance from the airways (3, 4, 5, 6, 7, 8, 9) and impairs alveolar macrophage innate immunity, including phagocytic function and IL-12 secretion in response to endotoxin (10). Such studies support the evolving recognition that alcohol abuse has specific effects on innate immune function within the lower airways and that the increased risk of pneumonia in these patients cannot be ascribed solely to factors such as malnutrition, aspiration, or poor oral hygiene. However, the precise mechanisms by which chronic ethanol ingestion impairs alveolar macrophage function are poorly understood.
Within the alveolar space, relatively undifferentiated circulating monocytes are recruited and undergo terminal maturation and differentiation into alveolar macrophages in response to stimulation by GM-CSF. GM-CSF is a 23-kDa protein that was originally isolated from mouse lung extracts but was named because of its potent effects on bone marrow development (fully reviewed in Ref.11). However, when a GM-CSF knockout mouse was constructed a little more than a decade ago, the phenotype was unexpected (12). Specifically, the absence of GM-CSF expression had no discernible effect on hematopoiesis. However, the mice developed a severe pulmonary phenotype that closely resembled pulmonary alveolar proteinosis (PAP)3 in humans. Insights from the mouse studies ultimately led to the recognition that most patients with PAP have acquired Abs to GM-CSF that neutralize the protein within the alveolar space and prevent binding to its receptor on the alveolar macrophage membrane (13). Although PAP was first described based on the accumulation of surfactant proteins and phospholipids within the alveolar space, we now recognize that it is due to global defects in GM-CSF-dependent alveolar macrophage function that include impaired surfactant recycling, as well as depressed innate immune functions (13). Therefore, patients with PAP have an acquired, functional deficiency in GM-CSF (as opposed to a genetic mutation) that produces alveolar macrophage dysfunction. With this background, we hypothesized that alcohol-mediated suppression of alveolar macrophage function could involve a functional defect in GM-CSF expression and/or signaling within the alveolar space.
GM-CSF is produced by the alveolar epithelium and binds to specific GM-CSF receptors on the plasma membrane of the alveolar macrophage and thereby activates an intracellular signaling pathway that ultimately leads to expression and nuclear binding of the transcription factor PU.1 (13). PU.1 is a member of the ETS family of transcription factors previously identified as a master transcription factor in the proliferation and differentiation of myeloid cells (14), and its expression is lost in alveolar macrophages both in patients with PAP and in GM-CSF knockout mice (11, 15). Lung-specific transgenic expression of GM-CSF in the type II cells of these mice restores PU.1 expression and normalizes alveolar macrophage function (16). In fact, constitutive expression of PU.1 in alveolar macrophages of GM-CSF-deficient mice by transfection with a PU.1-containing vector completely normalizes alveolar macrophage function (17), confirming the critical role for PU.1 in GM-CSF signal transduction. Thus, GM-CSF-dependent expression of PU.1 appears to be absolutely required for terminal maturation and function of the alveolar macrophage. However, to our knowledge, the effects of ethanol ingestion on GM-CSF expression and/or signaling to the alveolar macrophage within the alveolar space have not been examined. Therefore, we examined GM-CSF expression and key elements of its signaling, namely GM-CSF receptor expression and PU.1 expression, in our rat model of chronic ethanol ingestion. We then determined the effects of rGM-CSF treatment in vivo on restoring GM-CSF signal responsiveness, as well as innate immune function, in the alveolar macrophages of ethanol-fed rats.
Materials and Methods
Adult Male Sprague-Dawley rats (initial weights, 150–200 g; Charles River Laboratory) were fed the Lieber-DeCarli liquid diet (Research Diets) containing either ethanol (36% of total calories) or an isocaloric substitution with maltin-dextrin ad lib for 6 wk as published previously (18). All work was performed with the approval of the Institutional Care and Use of Animals Committee at the Atlanta Veterans Affairs Medical Center.
rGM-CSF treatment via upper airway in vivo
In some experiments, control-fed and ethanol-fed rats were treated with recombinant rat GM-CSF (PeproTech) or PBS vehicle alone via intranasal instillation for 3 consecutive days as we published previously (18). Briefly, rats were anesthetized with 2% isofluorane before gently instilling 500 ng of GM-CSF in 100 μl of PBS or 100 μl of PBS alone into one nostril with a pipette, which is then delivered into the airway by reflex sniffing by the anesthetized rat. Rats were then sacrificed 24 h after the third treatment with GM-CSF to obtain alveolar macrophages as described below.
Isolation of alveolar macrophages
Following pentobarbital anesthesia (100 mg/kg i.p.), a tracheostomy tube was placed and rat lungs were lavaged four times with 10 ml of sterile cold PBS (pH 7.4). The recovered lavage solution was centrifuged at 1500 rpm for 7 min, and the cell pellet resuspended in sterile medium for functional studies. This procedure yielded >95% alveolar macrophages.
RNA extractions and RT-PCR for GM-CSF expression
Total RNA was extracted from lung tissue using Qiagen RNA extraction kit. RNA from each sample was reverse transcribed followed by PCR with gene-specific primers. The number of cycles (35 for G3PDH and 40 for GM-CSF) was chosen from our preliminary optimization experiments for each gene product. PCR conditions were as follows: 5 min of denaturation at 94°C followed by 35–40 cycles of 45 s of denaturation at 94°C, 45 s annealing at 60°C or 53°C, and 90-s extension at 72°C, followed by a final extension at 72°C for 7 min. PCR products were separated on a 2% agarose gel containing ethidium bromide. For quantitation, PCR bands were scanned using an imaging system linked to a computer with analysis software. Relative amounts of G3PDH (983 bp) and GM-CSF (300 bp) were quantitated and expressed as GM-CSF:G3PDH ratios. Specific primers were as follows: G3PDH, (sense) 5′-GAAGGTCGGTGTCAACGGATTGGC-3′, and (antisense) 5′-CATGTAGGCCATGAGGTCCACCAC-3′; and GM-CSF, (sense) 5′-TCTGAGCCTCCTAAATGAC-3′, and (antisense) 5′-CATTTCTGGACCGGCTTC-3′.
Rat GM-CSF primers were designed in our lab and were obtained from Sigma-Genosys. Rat G3PDH primers were purchased from Promega. Molecular mass marker HaeIII digest with fragment sizes 1358–72 bp was purchased from Amersham Biosciences.
Determination of GM-CSF protein levels in lung lavage fluid
In selected experiments, rat lungs were lavaged via a tracheostomy tube with saline (5 cc × 3). The recovered lavage fluid (12 ± 1 cc in all cases) was centrifuged at 1500 × g for 10 min, and GM-CSF levels in the supernatants were determined by a rat-specific ELISA (R&D Systems). The lower limit of detection was 10 pg/ml. Data are reported as total amount (in nanograms) of GM-CSF present in the lung lavage fluid.
Flow cytometric detection of membrane and intracellular receptor expression
Membrane and intracellular expression of GM-CSF receptors on alveolar macrophages were measured by an established protocol (19). Briefly, cells were incubated for 30 min at room temperature with rabbit polyclonal Abs (Santa Cruz Biotechnology) to either the rat GM-CSF receptor α or β subunit or to an isotype-matched control Ab. Cells were washed to remove unbound Ab followed by 30 min incubation at room temperature with secondary anti-rabbit Ab conjugated to FITC. For intracellular staining of the receptors, cells were first permeabilized with 0.1% saponin in PBS, followed by staining with the Ab. Cells were washed with PBS-saponin before adding FITC-conjugated secondary Ab (Santa Cruz Biotechnology). Cells were washed with PBS and were kept in the dark at 4°C until analyzed. The labeled cells were analyzed by FACScan flow cytometer (BD Biosciences). Data are expressed both as percentage of cells positive for the α subunit or the β subunit, as well as the mean channel fluorescence for positive cells in each group.
Western blot analyses of PU.1 protein expression
Cell lysates were prepared by adding lysing reagent to isolated alveolar macrophages. Fifty micrograms of protein from each sample were loaded onto a 12% acrylamide gel and electrophoresed at 150 V for 75 min as described previously (17). The separated proteins were transferred to a 0.45 μM polyvinylidene difluoride membrane at 15 V for 75 min. Membranes were blocked at room temperature for 1 h in TBS with 0.2% Tween 20 (TBS-T) containing 5% nonfat dry milk in TBS-T. Primary Ab for PU.1 (Santa Cruz Biotechnology) at 1/50 in 5% milk in TBS-T was added to the membranes and kept at 4°C overnight. After several washing steps to remove unbound primary Ab, membrane was incubated at room temperature with HRP-labeled anti-rabbit IgG secondary Ab in 5% milk in TBS-T for 2 h. After adding ECL chemiluminescence reagent (Amersham Biosciences) to the membranes, bands were detected using a Bio-Rad Imaging System. For those experiments involving prior treatment with GM-CSF, PU.1 expression was normalized to expression of the housekeeping protein G3PDH to control for any potential proliferative effects of GM-CSF.
PU.1 electromobility shift assay
Cells were washed with cold PBS, and nuclear binding proteins were extracted. Protein concentration was determined by the Bradford method using Bio-Rad protein assay reagent. A double-stranded PU.1 consensus oligonucleotide (5′-TGAAAGAGGAACTTGGT-3′) was radiolabeled with [32P]γ-ATP using T4 polynucleotide kinase enzyme. Nuclear protein (10 μg) was incubated with radiolabeled PU.1 for 30 min at room temperature. For competition reactions, nonradiolabeled consensus and mutated PU.1 double-stranded oligonucleotides (5′-TGAAAGAGCTACTTGGT-3′) were added to the reaction mixture at 50× molar concentration as a control to confirm the identity of the PU.1-DNA complexes. DNA-protein complexes were separated on 6% native polyacrylamide gel (20:1 acrylamide/bis ratio) for 2–3 h. Gels were fixed in a 10% acetic acid/10% methanol solution for 10 min, dried under vacuum, and exposed to phosphoscreen.
Alveolar macrophage bacterial phagocytosis
In some experiments, alveolar macrophages were isolated from control-fed and ethanol-fed rats that had been treated with either GM-CSF or vehicle via the upper airway as described above. In those experiments, the macrophages were incubated for 4 h with FITC-labeled Staphylococcus aureus (Wood strain without protein A; Molecular Probes) in a 1:1 ratio; after incubation, cells were washed several times with PBS and examined by confocal microscopy. Phagocytosis images were obtained by laser confocal microscopy with Fluoview analysis (Olympus). Representative photomicrographs at ×60 magnification were obtained at a depth of 3–5 m in the z-plane of the macrophage, and both fluorescent and Nomarski differential contrast images were obtained. The cell membranes in the differential contrast images were digitally outlined and then these digital outlines were superimposed on the corresponding fluorescent images. In other experiments, alveolar macrophages from control-fed and ethanol-fed rats were isolated and then incubated with the FITC-labeled S. aureus in a 1:1 ratio ± rGM-CSF (10 ng/ml) in vitro for 4 h.
TNF-α release from rat alveolar macrophages
Freshly isolated alveolar macrophages (106 cells/ml) were incubated overnight ± 100 ng/ml LPS (Escherichia coli 0111:B4). Supernatants were collected and frozen at −70°C. TNF-α in these supernatants was measured using a rat TNF-α ELISA kit from BioSource International.
Data are presented as mean ± SEM. Data analysis was done by ANOVA with Student-Newman-Keuls test for group comparison and were considered statistically significant at a value of p < 0.05.
Chronic ethanol ingestion had no apparent effect on pulmonary GM-CSF expression
The first potential mechanism we examined was whether chronic ethanol ingestion dampened GM-CSF-dependent macrophage function by inhibiting expression of GM-CSF within the lung. We determined that ethanol ingestion in fact had no apparent effect on GM-CSF expression. As shown in Fig. 1, GM-CSF gene expression, as determined by RT-PCR (Fig. 1,A shows a representative PCR gel and Fig. 1,B shows the summary data from all experiments), was the same (p > 0.05) in the lungs of control-fed and ethanol-fed rats. We next examined GM-CSF protein levels in the alveolar space where GM-CSF priming of alveolar macrophages occurs. As shown in Fig. 1 C, chronic ethanol ingestion had no effect (p > 0.05) on the levels of GM-CSF protein in the lung lavage fluid when compared with control-fed rats. Taken together, these initial studies indicate that chronic ethanol ingestion had no significant effect on GM-CSF expression within the lungs of ethanol-fed rats.
Chronic ethanol ingestion decreased membrane expression of the GM-CSF receptor in the alveolar macrophage
As ethanol ingestion did not appear to affect GM-CSF protein availability within the alveolar space, we next examined whether ethanol ingestion could be interfering with GM-CSF signaling to the alveolar macrophage. As a first step in these experiments, we examined membrane expression of the GM-CSF receptor in alveolar macrophages freshly isolated from control-fed and ethanol-fed rats. As shown in Fig. 2, chronic ethanol ingestion significantly (p < 0.05) decreased membrane expression of both the GM-CSF receptor α subunit (GM-CSFRα) and the GM-CSF receptor β subunit (GM-CSFRβ). Fig. 2,A shows the relative number of cells that were positive for the GM-CSFR α and β subunit, with cells from ethanol-fed rats expressed relative to cells from control-fed rats. Fig. 2,B shows the relative mean channel fluorescence per cell for those cells that were positive for the α and β subunit and again with the cells from ethanol-fed rats expressed relative to cells from control-fed rats. Although ethanol ingestion did not significantly decrease the percentage of alveolar macrophages that were positive for GM-CSFRα membrane expression, the relative expression (mean channel fluorescence) for GM-CSFRα per positive cell was decreased by ∼50% (p < 0.05). By comparison, ethanol ingestion not only decreased the percentage of alveolar macrophages that were positive for GM-CSFRβ membrane expression by ∼50% (p < 0.05), the relative expression for GM-CSFRβ per positive cell was likewise decreased by ∼50% (p < 0.05). Importantly, decreased membrane expression of the GM-CSF receptor was relatively specific, at least as reflected by our determination that membrane expression for the IL-6R was the same (p > 0.05) in alveolar macrophages from ethanol-fed and control-fed rats (Fig. 3). Therefore, chronic ethanol ingestion significantly decreased membrane expression for both subunits of the GM-CSF receptor in alveolar macrophages, and this effect was more pronounced for the β subunit, which is responsible for initiating intracellular signaling following GM-CSF binding.
In parallel, chronic ethanol ingestion decreased expression of the transcription factor PU.1 that is required for GM-CSF-dependent functions in alveolar macrophages
We next compared the expression of PU.1, the master transcription factor for GM-CSF-dependent functions, in alveolar macrophages from ethanol-fed and control-fed rats. Although ethanol ingestion significantly decreased GM-CSF receptor expression, these findings did not necessarily mean that GM-CSG signaling was sufficiently impaired to explain the dampened macrophage function. Therefore, we reasoned that the next target to examine was PU.1 expression because the loss of PU.1 expression in the alveolar macrophages of patients with alveolar proteinosis and in GM-CSF knockout mice is causally related to alveolar macrophage dysfunction. As shown in Fig. 4, chronic ethanol ingestion significantly (p < 0.05) decreased PU.1 protein expression in alveolar macrophages from ethanol-fed rats compared with control-fed rats. Shown in Fig. 4,A are representative Western blots for PU.1 in macrophages from two control-fed and two ethanol-fed rats, while shown in Fig. 4 B are the summary data for all of the experimental determinations. Importantly, decreased PU.1 expression by ethanol was associated with decreased nuclear binding of PU.1 as discussed later.
Treatment with rGM-CSF in vivo restores alveolar macrophage membrane expression of the GM-CSF receptor in ethanol-fed rats
We had shown previously that treatment with rGM-CSF via the upper airway restores alveolar epithelial barrier function in chronic ethanol-fed rats. Therefore, we reasoned that similar treatment could mitigate the dampening effects of chronic ethanol ingestion on GM-CSF-dependent functions of the alveolar macrophage. As a first step in these experiments, we examined the effects of rGM-CSF treatment in vivo on membrane expression of the GM-CSF receptor. As shown in Fig. 5, in these experiments ethanol ingestion again decreased membrane expression (as reflected by mean channel fluorescence) by ∼50% for both the α and the β subunits. However, rGM-CSF treatment significantly (p < 0.05) increased membrane expression for both the α subunit (GM-CSFRα; Fig. 5,A) and the β subunit (GM-CSFRβ; Fig. 5,B). In fact, alveolar macrophage membrane expression for each subunit was increased ∼3-fold by GM-CSF treatment in ethanol-fed rats. In contrast, GM-CSF treatment had no significant effect (p > 0.05) on membrane expression of either the α or the β subunit in control-fed rats. This GM-CSF-induced increase in membrane expression of the GM-CSF receptor in alveolar macrophages from ethanol-fed rats appeared to be mediated in significant part by increased translocation of the receptor subunits from intracellular pools to the membrane. Specifically, rGM-CSF treatment significantly (p < 0.05) increased the membrane to intracellular ratio by ∼2-fold for both the α subunit (Fig. 6,A) and the β subunit (Fig. 6,B) in alveolar macrophages from ethanol-fed rats. In contrast, rGM-CSF treatment had no effect (p > 0.05) on the relative cellular distribution of either subunit in alveolar macrophages from control-fed rats. Fig. 7 shows representative fluorescent images for GM-CSFRα on the cell membranes of an alveolar macrophage from an ethanol-fed rat (Fig. 7, left panel) and an alveolar macrophage from an ethanol-fed rat treated with rGM-CSF (Fig. 7, right panel). Consistent with the flow cytometry data in Fig. 6,A, there is visual evidence of increased GM-CSFRα expression on the cell membrane following GM-CSF treatment. Taken together, the results in Figs. 5–7 suggest that rGM-CSF restores membrane expression of the GM-CSF receptor in alveolar macrophages from ethanol-fed rats, at least in part, by mobilizing receptor subunits from the intracellular pool to the plasma membrane.
Treatment with rGM-CSF in vivo restores PU.1 protein expression as well as PU.1 nuclear binding in alveolar macrophages in ethanol-fed rats
To determine whether restoration of GM-CSF receptor membrane expression translated to a restoration of PU.1 expression and therefore signaling capability, we next examined PU.1 expression as well as nuclear binding in alveolar macrophages from control-fed and ethanol-fed rats with or without treatment with rGM-CSF in vivo. For these experiments, PU.1 protein expression was quantitated and expressed relative to the housekeeping protein G3PDH. This was done to verify that any GM-CSF-mediated increases in PU.1 protein expression were not due solely to generalized growth factor effects of GM-CSF on the alveolar macrophages. As shown in Fig. 8,A, rGM-CSF treatment in vivo significantly (p < 0.05) increased cellular PU.1 protein expression in alveolar macrophages from ethanol-fed rats by ∼44%. By comparison, GM-CSF treatment induced a much more modest, albeit significant (p < 0.05), increase in PU.1 protein expression in alveolar macrophages from control-fed rats. In parallel, rGM-CSF treatment increased nuclear binding of PU.1 in alveolar macrophages from ethanol-fed rats. As shown in Fig. 8,B, GM-CSF treatment in vivo increased PU.1 nuclear binding as determined by electromobility shift assay on nuclear extracts from freshly isolated alveolar macrophages in each experimental group. Also evident in this representative gel is that chronic ethanol ingestion decreased PU.1 nuclear binding in parallel to the decrease in cellular PU.1 protein expression shown in Fig. 4. In contrast, GM-CSF treatment in vivo increased PU.1 nuclear binding in alveolar macrophages from both control-fed and ethanol-fed rats, although in general, this effect was more dramatic in macrophages from ethanol-fed rats. Taken together, the results in Fig. 8 suggest that rGM-CSF treatment in vivo restores PU.1 protein expression and nuclear binding in alveolar macrophages from ethanol-fed rats, and this increased PU.1 expression corresponds to restoration of GM-CSF receptor expression (as shown in Figs. 5–7).
Treatment with rGM-CSF in vivo restores alveolar macrophage innate immune function in ethanol-fed rats
Our final step in this study was to determine whether rGM-CSF treatment in vivo could actually improve the functional status of the alveolar macrophage in ethanol-fed rats. Clearly, restoration of GM-CSF receptor and PU.1 protein expression in the alveolar macrophage of ethanol-fed rats would be of limited significance if this did not translate into improved immune function. We first examined endotoxin-induced secretion of TNF-α by freshly isolated alveolar macrophages in vitro. As shown in Fig. 9, basal TNF-α secretion was the same (p > 0.05) in alveolar macrophages from control-fed and ethanol-fed rats, and prior GM-CSF treatment had the same (p > 0.05) modest effect on increasing basal TNF-α secretion in each group. However, endotoxin-stimulated TNF-α secretion was significantly decreased (p < 0.05) in alveolar macrophages from ethanol-fed rats (Fig. 9, third bar in each group). However, alveolar macrophages from GM-CSF-treated, ethanol-fed rats had the same (p > 0.05) endotoxin-stimulated secretion as alveolar macrophages from control-fed rats (hatched gray line connects these two groups in Fig. 9). Notably, TNF-α secretion was greatest (p < 0.05) in endotoxin-stimulated macrophages from control-fed rats, indicating that even under “normal” conditions, GM-CSF stimulation augments the endotoxin response. We next examined the effects of GM-CSF treatment on the ability of alveolar macrophages isolated from ethanol-fed rats to phagocytose bacteria in vitro. We did not include macrophages from control-fed rats in these studies. The confocal microscopy images in Fig. 10 illustrate that GM-CSF treatment augmented the ability of macrophages from ethanol-fed rats to phagocytose the fluorescent bacteria. Fig. 10, A and B, shows the corresponding fluorescent and differential contrast images for a macrophage from an untreated, ethanol-fed rat, in which relatively few bacteria have been phagocytosed, and the majority of these remain in the periphery of the cell. In contrast, as shown in Fig. 10, C and D, alveolar macrophages from ethanol-fed rats treated with GM-CSF were able to ingest and internalize more of the fluorescent bacteria. Taken together, the results shown in Figs. 9 and 10 indicate that GM-CSF treatment improved innate immune functions in the alveolar macrophages of ethanol-fed rats.
Treatment with rGM-CSF in vitro restores bacterial phagocytic capacity in alveolar macrophages from ethanol-fed rats
Although the results shown in Figs. 9 and 10 are striking, it is possible that GM-CSF treatment in vivo recruited and primed a new population of alveolar macrophages from peripheral monocyte pools and had no effect on the existing alveolar macrophage pool. To test whether or not GM-CSF treatment could directly improve immune function in resident alveolar macrophages in the alcoholic lung, we performed additional experiments in which macrophages were isolated and then stimulated with GM-CSF in vitro. In these conditions, alveolar macrophages adhere tightly to plastic, and therefore, we could not perform flow cytometry to assess GM-CSF receptor membrane expression. However, adherent macrophages remain functional, and therefore, we assessed bacterial phagocytic capacity with and without GM-CSF treatment in vitro. We reasoned that if GM-CSF treatment could directly increase a relevant macrophage function in these conditions, this would provide further evidence that its effects in vivo could not be solely ascribed to recruiting an entirely new alveolar macrophage population from an extrapulmonary monocyte pool. In these experiments, isolated macrophages from control-fed and ethanol-fed rats were incubated with the fluorescent bacteria for 4 h, ± recombinant rat GM-CSF (10 ng/ml). The cells were examined by fluorescent microscopy and the percentage of macrophages that had ingested one or more bacteria determined. As shown in Fig. 11 (and consistent with the results shown in Fig. 4 above), the percentage of macrophages from ethanol-fed rats that had any detectable phagocytic function in vitro was decreased (56 ± 8% vs 92 ± 5%; p < 0.05) compared with macrophages from control-fed rats, and this was almost completely reversed with exogenous GM-CSF treatment (83 ± 6% vs 56 ± 8%; p < 0.05).
In this study, we determined that although chronic ethanol ingestion in rats did not affect GM-CSF expression within the alveolar space, it nevertheless dampened GM-CSF signaling capacity by decreasing expression of GM-CSF receptors on the surface of the alveolar macrophage. This appeared to be relatively specific, as membrane expression of the IL-6R was not affected by chronic ethanol ingestion. In parallel and likely as a consequence of decreased membrane receptors, the expression and nuclear binding of the GM-CSF-dependent transcription factor PU.1 was also decreased. Remarkably, treatment with rGM-CSF via the upper airway restored membrane expression of the GM-CSF receptor as well as the downstream expression and nuclear binding of PU.1 in the alveolar macrophages of ethanol-fed rats. Even more importantly in terms of potential clinical relevance, GM-CSF treatment restored innate immune functions in alveolar macrophages of ethanol-fed rats, as reflected by endotoxin-induced secretion of TNF-α and bacterial phagocytosis. Taken together, these results suggest that chronic ethanol ingestion inhibits alveolar macrophage immune function by dampening, but not completely blocking, macrophage responsiveness to the stimulatory effects of ambient GM-CSF within the alveolar microenvironment. Furthermore, this study raises the possibility that in the appropriate clinical context impaired pulmonary host defenses in alcoholic patients could be corrected with rGM-CSF treatment.
For more than a century, it has been recognized that chronic alcohol abuse is a major risk factor for the development of pneumonia. Although factors associated with alcoholism such as malnutrition, poor dentition, aspiration, smoking, and other drug likely exacerbate the risk, experimental animal models have demonstrated that ethanol ingestion alone (in the absence of these other factors associated with chronic alcohol abuse in humans) impairs alveolar macrophage innate immune function (5, 7, 8, 10, 20). However, the specific mechanisms by which ethanol ingestion down-regulates macrophage function have not been identified. This study provides a plausible and specific mechanism by which alveolar macrophage maturation and function is dampened during chronic ethanol ingestion. Specifically, ethanol ingestion interferes with GM-CSF priming within the alveolar space that is absolutely essential for the alveolar macrophage to acquire its full complement of immune functions.
These findings are important because they provide novel insights into the fundamental mechanisms by which chronic alcohol abuse impairs host defenses and renders patients susceptible to pulmonary infections. They are also important because they raise the provocative possibility that the alcoholic macrophage could be stimulated in vivo by exogenous GM-CSF treatment and thereby rapidly reacquire the innate immune function that protects the lower airways from microbial invasion. As GM-CSF has already been tested in a phase II clinical trial of sepsis and lung injury and was found to increase alveolar macrophage function (21), it is reasonable to speculate that treating alcoholic patients with rGM-CSF as adjunctive therapy for serious lung infections could augment their pulmonary host defense and improve outcome.
Although we have identified significant defects in GM-CSF receptor expression and parallel decreases in PU.1 expression in the alcoholic macrophage, we recognize that GM-CSF signaling is complex and that alcohol abuse likely perturbs other components of the GM-CSF signal transduction pathway. For example, while the GM-CSFRβ initiates intracellular signaling following GM-CSF binding, it contains no intrinsic catalytic activity. Rather, it is constitutively associated with a tyrosine kinase, JAK2, that when activated initiates the intracellular signaling cascade (22). We did not examine this kinase or any of the other intracellular components that transduce the GM-CSF signal from receptor binding to PU.1 expression and nuclear transcription activation, and one would expect that these components would likewise be dampened. Another complexity is that the GM-CSFRβ is actually common to the IL-3R and IL-5R (22, 23). However, IL-5R expression appears to be limited to eosinophils (23), whereas IL-3 expression is limited to eosinophils, basophils, and mast cells (23). Therefore, ethanol-mediated changes in β subunit expression in alveolar macrophages, as we observed, are likely specific to GM-CSF receptor expression in this cell type, particularly as they parallel changes in the unique α subunit. Whether alcohol abuse has any significant effects on IL-3 and/or IL-5 function in eosinophils and/or basophils is an open question.
Our findings provide important new insights but also raise new questions. In particular, how does ethanol ingestion inhibit mobilization and/or insertion of the GM-CSF receptor into the plasma membrane? In parallel, how does treatment with supraphysiological levels of GM-CSF correct this defect when physiologic levels, which are not inhibited by ethanol ingestion, do not? At present we can only speculate. We know from previous studies that chronic ethanol ingestion causes oxidative stress and severe glutathione depletion within the alveolar space in experimental animals (24) as well as in humans (25), leading to diverse abnormalities in alveolar epithelial function that are prevented by supplementing the ethanol-containing diet with glutathione precursors (26, 27, 28). It is certainly conceivable that ethanol-induced oxidative stress interferes with GM-CSF trafficking to the plasma membrane. Acetaldehyde, the first product of ethanol metabolism, produces endoplasmic reticulum stress in hepatocytes, thereby inhibiting mitochondrial glutathione transport (29). If similar endoplasmic reticulum stresses occurred within the alveolar macrophage, this could lead to misfolding of the GM-CSF receptor. However, this would have to be a relatively specific inhibition as we determined that membrane expression of the IL-6R was not affected. Although we did not find evidence that ethanol ingestion decreased gene expression of GM-CSF and/or its receptors, it is possible that supraphysiological levels of GM-CSF increased expression of one or more components of the pathway in addition to augmenting membrane expression of the receptors. Regardless of the mechanism, it is intriguing that rGM-CSF treatment restored GM-CSF receptor membrane expression, PU.1 expression, and immune function in the alcoholic macrophage. Clearly, the signaling cascade is dampened, but not completely blocked, by chronic ethanol ingestion. As GM-CSF protein levels in the alveolar space were not altered, one could speculate that the significant decrease in GM-CSF receptor expression alone can explain this dampening. Therefore, when the ambient level of GM-CSF is markedly increased by treatment with rGM-CSF, a higher percentage (if not all) of the available GM-CSF receptors could become activated, thereby amplifying the previously muted signaling cascade. The subsequent activation of PU.1-responsive genes could then further activate the macrophage, which in turn could drive trafficking of the GM-CSF receptor as well as other membrane components of a “mature” macrophage to the cell surface. Therefore, while the alcoholic macrophage may be relatively dormant in the face of physiologic GM-CSF stimulation, it appears that its immune function can be restored by pharmacological treatment with “supraphysiological” concentrations of GM-CSF. In parallel, we cannot exclude the possibility that GM-CSF treatment could have increased the recruitment of extrapulmonary monocytes to the alveolar space and their maturation into functional alveolar macrophages. We did not detect any increase in the alveolar macrophage population in the lung lavage fluid after GM-CSF treatment (data not shown), but this does not exclude the possibility that recruited cells replaced the dysfunctional pool.
Finally, it is important to note that GM-CSF treatment did not significantly increase GM-CSF receptor expression in the alveolar macrophages of control-fed rats. However, this treatment modestly increased PU.1 protein expression and nuclear binding, and LPS-stimulated TNF-α secretion, in alveolar macrophages from control-fed rats, albeit less dramatically than in ethanol-fed rats. This suggests that under normal conditions, GM-CSF receptor surface expression is at or near maximal density. However, it is likely that not all of the receptors are occupied at any one time under normal conditions, as supraphysiological levels of GM-CSF induced a response even in the macrophages of control-fed rats. Whether GM-CSF signaling is constitutive or regulated at the receptor or postreceptor level under normal conditions is unknown and is a ripe area for further investigation.
In summary, we report for the first time that chronic ethanol ingestion interferes with GM-CSF-dependent alveolar macrophage immune function by decreasing GM-CSF receptor expression and subsequent activation, of the master transcription factor PU.1, and that these defects are reversed by high-dose GM-CSF treatment delivered via the airway. These findings provide new insights into the potential mechanisms by which alcohol abuse suppresses pulmonary host immunity and renders patients susceptible to serious lung infections, including tuberculosis and bacterial pneumonias. Although the specific cellular mechanisms require further investigation, it is fascinating to consider that pulmonary host defenses could be rapidly augmented by acute treatment with rGM-CSF, even in the context of chronic alcohol abuse, and thereby decrease the morbidity and/or mortality from serious pulmonary infections in this vulnerable population.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work is supported by National Institutes of Health, National Institute on Alcohol Abuse and Alcoholism P50 AA013757 and a Veterans Affairs Merit Review (to D.M.G.).
Abbreviations used in this paper: PAP, pulmonary alveolar proteinosis; GM-CSFRα, GM-CSF receptor α subunit; GM-CSFRβ, GM-CSF receptor β subunit.