Alcohol-induced lipid dysregulation impairs glycolytic responses to LPS in alveolar macrophages

Abstract

Several conditions are marked by increased susceptibility to, and enhanced severity of, bacterial infections. Alcohol use disorder, one of these conditions, is known to predispose to bacterial pneumonia by suppressing the lung’s innate immune system, and more specifically by disrupting critical alveolar macrophage (AM) functions. Recently, we established that chronic ethanol consumption also perturbs surfactant lipid homeostasis in the lung and that elevated concentrations of free fatty acids contribute to blocking essential AM functions, such as agonist-induced cytokine expression. In this study, we extend these observations by showing that elevated free fatty acid levels impair metabolic responses to lipopolysaccharide (LPS) in AMs. In particular, we show that the glycolytic reprogramming characteristic of LPS-stimulated AMs is blunted by the saturated fatty acid palmitate, whereas oleate, an unsaturated fatty acid, or ethanol alone, had no effect on this adaptive metabolic response. Additionally, we found that elevated concentrations of palmitate induced mitochondrial oxidative stress and that glycolytic reprogramming and cytokine production to LPS could be partially restored in AMs by either pharmacologically blocking palmitate entry into mitochondria or administering a mitochondrial specific antioxidant. Taken together, these findings suggest that alcohol and elevated levels of saturated fatty acids conspire to impair pulmonary innate immunity by altering metabolic responses in AMs. Additionally, our findings suggest that targeting the mechanisms involved in fatty acid metabolism can restore pulmonary immunity and limit bacterial pneumonia in individuals with alcohol use disorder.

Introduction:

Bacterial pneumonia remains a frequent cause of prolonged hospitalization and death in alcoholics^1,2,3,4. A correlation between chronic alcohol use disorder and both the incidence and severity of lung infections has been established^5,6. One of the first lines of defense against respiratory pathogens is the alveolar macrophage (AM)⁷. AMs reside in the alveolus among type I and II alveolar epithelial cells (AEI, AEII).⁸ These cells are also bathed in a mixture of lipids, proteins and carbohydrates known as surfactant^9,10. The composition of surfactant is critical both for maintaining proper surface tension in the distal pulmonary airspaces and for metabolic signaling to other resident cells^11,12. For the most part, surfactant components are produced by AEIIs, while AMs and AEIIs work together to clear surfactant components from the alveolar space¹³.

Excessive consumption of ethanol (EtOH) leads to hepatic steatosis, also known as alcoholic fatty liver¹⁴. While most EtOH consumed by humans is metabolized to acetaldehyde through alcohol dehydrogenase (ADH) and cytochrome P450 (CYP2E1) in the liver, the distal lung is also a significant processor of EtOH due to the presence of these enzymes in AMs and AEIIs¹⁵. Similar to the effects in the liver, we and others have shown that chronic alcohol consumption disturbs the balance of lipids in lung surfactant, causing an increase of 3-fold or more in the levels of free fatty acids¹⁶, with the major species (~80%) being palmitate, a 16-carbon fatty acyl chain. Moreover, we showed that lipid accumulation leads to a “foamy” macrophage appearance and that uptake of lipids by AMs impairs cytokine responses to lipopolysaccharide (LPS). Together, these findings suggest that EtOH-induced pulmonary steatosis alters immune homeostasis in the lung, perhaps contributing to the increased incidence and severity of bacterial pneumonia in alcoholics.

AM behavior is believed to exist within a spectrum, and at its extremes, ranges from a pro-inflammatory M1 phenotype to an anti-inflammatory and pro-repair M2 phenotype. In response to M1 stimuli, such as the bacterial surface molecule LPS, AMs increase glycolysis and decrease mitochondrial respiration. The importance of these metabolic changes has been firmly illustrated by the ability of glycolytic inhibitors to blunt cytokine responses in various in vitro models^17,18. In contrast to M1 agonists, M2 promoting agents, such as the cytokines IL4 and IL13 have been shown to augment mitochondrial respiration and this is important for the activation of anti-inflammatory and pro-remodeling responses. While little is known about the effects of EtOH on metabolic reprogramming of alveolar macrophages it is known that these cells have blunted M1 responses to LPS and produce higher than normal levels of TGF-β at baseline^19,20.

In response to LPS stimulation, hypoxia-inducible factor-1α (HIF-1α) levels can also sometimes increase in cells. HIF-1α is normally an unstable protein, and is quickly degraded in the cytosol by prolyl hydroxylase (PHD) under most cellular conditions²¹. However, in response to LPS, succinate accumulates in the cytosol, in part from decreased activity of succinate dehydrogenase (SDHA), inhibiting PHD and allowing HIF-1α to stabilize and migrate to the nucleus²². Once in the nucleus, HIF-1α has been shown to drive the transcription of various factors, including many involved in glycolysis, such as the glucose transporters 1 and 4, lactate dehydrogenase, and an array of glycolytic enzymes that serve to enhance glycolytic capacity²³.

Metabolic reprogramming in the context of the chronic EtOH model has largely been overlooked as a potential cause for impaired immunoregulation. Here we explored the role that EtOH and elevated levels of free fatty acids have on metabolic responses to LPS in AMs. We have developed an in vitro model for examining the relationship between metabolic regulation and the M1 phenotype of LPS-stimulated AMs. This work also explores some of the potential pathways for restoring immune response by targeting mechanisms downstream of fatty acid β-oxidation.

Materials and Methods

NR8383 Cell Culture and LPS Stimulation:

Rat AMs NR8383 cells were purchased from ATCC (Manassas, VA). Cells were cultured in F-12K medium containing 15% FBS (Gemini, West Sacramento, CA) and 1% penicillin/streptomycin (Gibco, Gaithersburg, MD) and one or more of the following; 1) 0.1% v/v ethanol; 2) 25 to 125 μM BSA-conjugated palmitate (Sigma-Aldrich, St. Louis, MO); and/or 3) fatty-acid free BSA vehicle (Roche, Mannheim, Germany). LPS (1μg/mL, H₂O vehicle; Sigma-Aldrich) was administered for 5 hours before cell harvest. Treatments with MitoTempo (10μM, DMSO vehicle, Sigma-Aldrich, SML0737), Etomoxir (10 μM, DMSO vehicle, Sigma-Aldrich, E1905), or 2-DG (3 or 7 mM, H₂O vehicle; Sigma-Aldrich, D8375 were given concurrently with Palmitate/BSA treatment. NR8383 cells are seeded in 10cm dishes, 96-well plates or 8-well XFp cell culture miniplate and 0.10 % EtOH was added to the medium 72 hours before cell harvest or metabolic assay. In all scenarios, culture vessels were sealed with Parafilm (Bemis, Oshkosh, WI) to prevent evaporation of ethanol. The medium and Parafilm were changed every 24 hours. Fatty acids (palmitate or oleate) were conjugated to fatty-acid-free BSA at a 6:1 ratio before addition to culture medium.

Cellular respiration and glycolysis studies:

NR8383 cells (10×10⁴) were plated in an 8-well XFp cell culture miniplate (Agilent, Santa Clara, CA.) for 48 hours at 37°C in a 5% CO2 (v/v) incubator; Cells were exposed to palmitate for 24 hours prior to metabolic assays. In case of 72 hour ethanol exposure, cells were first exposed to ethanol (0.1% v/v) for 24 hours in regular culture vessels before being transferred to miniplates with medium containing 0.1% v/v ethanol for the final 48 hours.

Mito stress test:

One hour prior to the assay, growth medium was replaced with assay medium (XF basal medium containing 1 mM sodium pyruvate, 25 mM D-Glucose pH 7.4), and incubated in a non-CO2 incubator at 37°C. The stock compounds rotenone/antimycin A, carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone (FCCP) and oligomycin were prepared by diluting the compounds with the assay medium. The stock compounds were then loaded into the cartridge (Seahorse Bioscience, Agilent) hydrated with the XF calibrant to achieve final concentrations of 4.0 μM oligomycin, 0.75 μM FCCP and 4.0/2.0 μM rotenone/antimycin A. Oxygen consumption rate (OCR) and extracellular acidification rates (ECAR) were assessed using Seahorse XFp Extracellular Flux Analyzer (SeahorseBioscience, Agilent).

Glycolysis stress test:

Stock compounds D-glucose, oligomycin, and 2-deoxy-d-glucose (2-DG) were prepared by diluting with assay medium. The stock compounds were then loaded into the cartridge (Seahorse Bioscience, Agilent) hydrated with the XF calibrant to achieve final concentrations of 25 mM Glucose, 10 μM oligomycin, and 50 mM 2-DG. Oxygen consumption rate (OCR) and extracellular acidification rates (ECAR) were assessed using Seahorse XFp Extracellular Flux Analyzer (SeahorseBioscience, Agilent). Assessment of glucose uptake was performed using a kit purchased from Promega (Madison, WI) and following manufacturer’s instructions.

RNA Isolation and Real-Time Quantitative Polymerase Chain Reaction (RT-qPCR):

Gene transcript levels for IL-1β, IL-6, TNF-α and iNOS were measured using standard methods and HPRT1, β-Actin and GAPDH were used as housekeeping genes. See the MATERIALS AND METHODS section of the online supplement for specific details regarding RNA isolation and methods for quantification of transcript levels.

Western Blot (WB) Analysis:

Western blot analysis was performed for HIF-1α, using β-Actin as a housekeeping protein, and the electron transport chain protein complexes I-IV, using TOMM20 as a housekeeping protein. Detailed protocols are provided in the MATERIALS AND METHODS section of the online supplement.

MitoSOX Assay of mitochondrial oxidative state:

Cells were seeded in four-well culture slides (Corning Life Sciences, Corning, NY) and treated as indicated for 72 hours. One vial (50 μg) MitoSOX Red (Molecular Probes, Invitrogen) was dissolved in 13 μL DMSO to make a 5 mM stock solution. 12 μL of the stock solution was dissolved in 30 mL HBSS (Gibco, Gaitherburg, MD) to make a 2μM working solution. Medium was removed and cells were washed twice with PBS. 1 mL of the working solution was added to each well, and the slides were covered and incubated for 5 minutes at 37C, after which each well was washed three times with HBSS. The wells were removed and mounting medium with DAPI (Prolong Gold with DAPI, Invitrogen, Carlsbad, CA) was applied with a coverslip. Imaging was performed at 510/590 for MitoSOX and 340/380 for DAPI.

Glucose uptake assay:

Glucose uptake was quantified using a commercially available kit (Promega, Madison, WI). In brief, cells were seeded in 96-well plates and exposed to 2DG6P detection reagent made by combining 2500 μL luciferase reagent, 25 μL NADP+, 62.5 μL G6PDH, 12.5 μL reductase and 1.5625 μL reductase substrate. 100 mM stock 2-DG was then diluted in PBS to make a 1 mM solution. Standards were made by diluting 7.5 L of the 1mM 2-DGP standard in 142.5 μL MilliQ water to make a 50 μM solution. Standard wells were filled with 0, 10, 20 or 30 μL of the 50 μM solution and 50, 40, 30, or 20 μL MilliQ water. After treatment, medium was removed and cells were washed with PBS and 1 mM 2-DG solution was added to each well and incubated for 10 min at room temperature. Next, 25 μL of stop buffer was added, followed by 25 μL of neutralization buffer, and 100 μL 2DG6P detection reagent. Luminescence was recorded with a 0.8 s integration on a Biotek Synergy 2 plate reader. Upon completion of assay, cells were counted to normalize data.

Results

Culture of AMs with high concentrations of palmitate blunts the response to LPS.

To evaluate the effects of elevated free fatty acid levels on metabolic responses to LPS we employed the NR8383 rat alveolar macrophage cell line. This cell line and treatment protocol were selected based on its known expression of ADH and CYP2E1 enzymes and the fact that elevated levels of free fatty acids are seen in the rat lung after chronic EtOH exposure. Using the Seahorse XFp Bioanalyzer, we found, as expected, that exposure to LPS induced a dramatic downregulation in basal and maximal mitochondrial oxygen consumption rate (OCR) (Figure 1A) and a marked upregulation in glycolytic activity, as measured by extracellular acidification rate (ECAR) (Figure 1B). Having confirmed metabolic reprogramming to LPS in this macrophage cell line, we next designed a model of EtOH and free fatty acid exposure to simulate the environment experienced by AMs in the lung of humans with alcohol use disorder. A diagram depicting our approach is shown in Figure 2A. Consistent with the importance of glycolysis in driving cytokine production to LPS, we found that inhibition of glycolysis with 2-DG blocked pro-inflammatory cytokine gene transcription for IL1β, TNFα and IL-6 at 5 hours after LPS stimulation (Figure 2B).

Figure 1:

Metabolic reprogramming of NR8383 cells in response to LPS. A) Mitochondrial OCR is reduced at 5 hours after lipopolysaccharide (LPS) stimulation (1μg/mL). B) Glycolysis is upregulated in response to LPS as measured by extracellular acidification rate (ECAR).

Figure 2:

High levels of palmitate suppress LPS induced cytokine production in NR8383 cells. A) Diagram depicting our experimental model of alcohol and fatty acid exposure. B) Gene transcription for IL-1β, IL-6 and TNF-α in cells exposed to LPS in the presence or absence of 2-Deoxy-D-Glucose (2DG). C) High-dose palmitate (HDPA) reduces LPS-induced IL-1β and IL-6 gene transcription at 5 hours. TNF-α is unaffected by HDPA exposure. D) Transcript levels for M2 markers arginase 1 (Arg1), monocyte chemoattractant protein 1 (MCP1) and transforming growth factor β (TGFβ). E) Exposure to oleate (20μM) has not effect on LPS-induced cytokine gene transcription (N=6, normalized to HPRT). Note: b significantly different from a, c significantly different from b, p<0.05, Holm-Sidak pairwise method)

To confirm that elevated levels of free fatty acids alter cytokine response to LPS, we next cultured NR8383 cells in the presence of absence of EtOH plus either low (25 μM) or high concentrations (75 μM) of palmitate. The concentration of high dose palmitate selected for these studies was based on previous work showing a roughly three-fold increase in free fatty acid levels in bronchial alveolar lavage fluid (BALF) in rats chronically exposed to ethanol and work by others showing a similar increase in lipid levels in the alcohol exposed human lung^1,24. Importantly, the doses selected for our studies did not cause significant toxicity to cells, as measured by Trypan Blue Exclusion assay (Supplemental Figure 1). As shown in Figure 2C, we found that elevated concentrations of palmitate markedly inhibited gene transcription for IL-1β and IL-6 whereas physiological levels of palmitate alone had little to no effect on cytokine expression. Similarly, EtOH at various concentrations did not alter LPS-induced cytokine production in our cells (Supplemental Figure 2). Of note, the blunting effect to high dose palmitate was also associated with an increase in transcript levels for the M2 marker arginase 1 (Figure 2D). Interestingly, TNF-α gene transcription was not affected by high dose palmitate exposure, and levels increased in the context of combined EtOH plus high dose palmitate exposure.

To determine whether suppression of cytokine production to LPS by fatty acids was specific to palmitate, we next exposed NR8383 cells to oleate, an unsaturated (18:1) fatty acid, and the most abundant unsaturated fatty acid species in pulmonary surfactant. In contrast to findings with palmitate, oleate did not alter LPS-induced cytokine gene transcription in our model (Figure 2E). Taken together, our findings suggest that exposure to elevated levels of specific fatty acid species may be partially responsible for disrupting macrophage function in the alcoholic lung.

Elevated levels of palmitate inhibit the upregulation of glycolysis by LPS.

Next, to assess whether elevated concentrations of palmitate alter glycolytic responses to LPS, we measured ECAR in NR8383 cells exposed to varying concentration of palmitate (25–125 μM). As shown in Figure 3A, we found that glycolytic response to LPS was similar in cells exposed to BSA or to normal physiologic concentrations of palmitate. However, we observed a dose dependent suppression of glycolysis with increasing concentrations of palmitate (Figure 3A) and this associated with a significant reduction in glycolytic flux (Figure 3B, 3C, 3D), as measured by ECAR and confirmed by the glycolytic stress test. Interestingly, we found that EtOH exposure alone had no effect on glycolysis in our cells (Figure 3B). Importantly, decreased glycolysis due to high dose palmitate was not associated with changes in glucose uptake (Figure 3E), indicating that metabolic changes were not due to a decrease substrate availability. Notably, we found that HIF-1α levels were markedly reduced in cells exposed to high concentrations of palmitate plus EtOH, supporting the notion that these combined insults might act to reduce glycolytic enzymes in our cells (Figure 3F).

Figure 3:

Palmitate alters glycolytic responses to LPS in NR8383. A) Palmitate causes a dose-dependent reduction in glycolytic response to LPS in NR8383 cells. B) EtOH has no effect on extracellular acidification rate in NR8383 cells, although the combined effects of ethanol (0.10% v/v) plus palmitate (75 μM) suppressed glycolytic activity (N=9). C,D) Maximal and spare glycolytic capacity are reduced in response to combined EtOH and palmitate exposure. E) High dose palmitate alone had no effect on glucose uptake, although a trend toward reduced uptake was observed in association with combined palmitate and EtOH exposure. F) Western blot for HIF-1α in NR8383 cells exposed to either low or high dose palmitate in the presence or absence of EtOH (N=12, compiled from 4 blots, normalized to β-Actin). (b significantly different from a, p<0.05, Holm-Sidak pairwise method)

Palmitate exposure alone does not alter mitochondrial oxygen consumption due to LPS in NR8383 cells.

As already mentioned and shown in Fig 1, the metabolic response to LPS in macrophages is to augment glycolysis and suppress mitochondrial respiration. To test whether palmitate also affected mitochondrial respiration in our cells we measured OCR in NR8383 cells exposed to EtOH, palmitate or combined treatments. As shown in Fig 4., we found that high concentrations of palmitate did not affect basal or maximal mitochondrial OCR in unstimulated cells, although OCR was significantly reduced in response to combined palmitate and EtOH exposures (Figure 4A), especially in association with LPS (Figure 4B). Importantly, changes in OCR were not associated with altered levels of electron transport chain complexes (Figure 4C), suggesting that changes in OCR were due to functional rather than quantitative defects in the electron transport chain. Consistent with this, we found that transcript levels of inducible nitric oxide synthase (iNOS), a known suppressor of mitochondrial respiration, were markedly increased in NR8383 cells exposed to a combination of palmitate and EtOH (Figure 4D)²⁵.

Figure 4:

Combined effects of high dose palmitate and EtOH reduce mitochondrial OCR in NR8383 AMs. (A) Basal metabolism and maximal respiratory capacity are severely reduced by the combination of EtOH and HDPA. B) Basal metabolism and maximal respiratory capacity are further reduced by the combination of EtOH, HDPA and LPS exposure C). WB for the electron transport chain (ETC) protein complexes in NR8383 cells exposed to low or high dose palmitate in the presence or absence of EtOH. D). Transcript levels for inducible nitric oxide synthase (iNOS) are upregulated by HDPA (N=6, normalized to β-Actin), as measured by RT-qPCR. (b significantly different from a, c significantly different from a and b, p<0.05, Holm-Sidak pairwise method)

Fatty acid oxidation contributes to impairing LPS-induced cytokine production.

β-oxidation is a potent inducer of oxidative stress by delivering high concentrations of electrons to the ETC in response to fatty acid breakdown. With this understanding in mind, we hypothesized that elevated levels of palmitate might alter LPS induced metabolic responses in AMs by increasing the redox state in cells. To test this, we compared mitochondrial ROS levels in NR8383 cells exposed to either low or high concentrations of palmitate. As shown in figure 5A, we found that mitochondrial ROS levels were significantly increased (25%) in NR8383 cells exposed to elevated palmitate concentrations. Moreover, we found that treatment with etomoxir, a pharmacological inhibitor of the fatty acid transporter CPT-1, effectively restored mitochondrial redox state (Figure 5A) and glycolytic reprogramming (Figure 5B, left) while also enhancing LPS-induced glycolytic responses and partially rescuing cytokine production, as demonstrated by a small, but significant increase in IL-6 levels (Figure 5B, right). Lastly, to more specifically test the hypothesis that reducing oxidative stress restores LPS-induced glycolytic responses we treated cells with a mitochondrial specific antioxidant MitoTEMPO in the presence of palmitate. As shown in figure 6, we found that MitoTEMPO both restored glycolytic responses and significantly increased LPS-induced cytokine production, as evidenced by the increase in IL-6 and IL-1β transcript levels. Notably, MitoTEMPO reduced ROS levels to both EtOH and palmitate while, as expected, etomoxir reduced ROS levels only due to palmitate exposure alone. Finally, we confirmed the effects of MitoTEMPO were due to its anti-oxidant properties rather than to the effects of TPP on mitochondrial membrane potential (Figure 6B).

Figure 5:

Mitochondrial ROS levels are increased in NR8383 cells exposed to high concentrations of palmitate. A) Superoxide levels in cells exposed to LDPA or HDPA. Some cells exposed to HDPA were also treated with treatment with the CPT-1 inhibitor Etomoxir (ETO) at 10μM. B) Etomoxir partially restores the glycolytic response (left) and cytokine response (right) to LPS in NR8383 cells. (N=6, Normalized to GAPDH). (b significantly different from a, p<0.05, Holm-Sidak pairwise method)

Figure 6:

Antioxidant treatment restores glycolytic activity and cytokine responses to LPS in NR8383 cells exposed to EtOH and high dose palmitate. A) MitoTEMPO (10μmM) reducs superoxide levels in cells exposed to EtOH plus HDPA (left, N=6). B) MitoTEMPO restores LPS-induced glycolytic responses (left) and cytokine production (right) blunted by EtOH and HDPA (N=6, Normalized to GAPDH). (b significantly different from a, p<0.05, Holm-Sidak pairwise method)

Discussion

In this study, we confirmed that cytokine responses to LPS in AMs are significantly blunted by high levels of fatty acids characteristic of the chronic alcohol exposed lung. Moreover, we show that EtOH exposure alone or normal physiological levels of fatty acids did not alter LPS-induced Th1 cytokine production. Further, we found that suppression of cytokine production appeared to be dependent on the type of lipid species, and that cytokine suppression was associated with a downregulation in glycolytic reprogramming, mimicking the findings seen with known glycolytic inhibitors, such as 2-DG. Lastly, our work indicates that alcohol-induced AM dysfunction can be restored, at least partially, by inhibiting the uptake of fatty acids into mitochondrial or treating cells with a mitochondrial specific antioxidant, suggesting similar approaches might be effective in restoring immune homeostasis in the lungs of humans with alcohol use disorder.

A key finding in this study is the observation that elevated levels of free fatty acids impair glycolytic responses to LPS in AMs. Our findings suggest this is due to the consequences of metabolizing fatty acids in mitochondria rather than to toxic effects in the cytoplasm. More specifically, our findings suggest this relates to the induction of mitochondrial oxidative stress after fatty acid breakdown. That said, we recognize that β-oxidation alone does fully explain our findings, since similar effects were not seen when cells were cultured with an unsaturated fatty acid. With this in mind, we postulate that saturated fatty acids elicit other deleterious effects on cells. For example, in the obesity field elevated levels of saturated fatty acids are incorporated into cellular membranes, including the plasma, mitochondrial and endoplasmic reticulum membranes, which in turn disrupts membrane fluidity and trafficking of intracellular proteins^26,27,28. Time-dependent labeling studies will be important for ultimately uncovering how elevated levels of saturated fatty acids alter alveolar macrophage function in the chronic alcohol exposed lung.

Another finding observed in conjunction with a decrease glycolysis in palmitate-exposed cells was a dramatic reduction in HIF-1α levels. Interestingly, since HIF-1α is a known driver of the expression of many glycolytic genes²⁹, these findings suggest that low levels of HIF-1α might also contribute to the reduction in glycolysis in our model. However, in preliminary investigations (Supplemental 3) we found that treatment of AMs with dimethyloxaloylglycine, a cell permeable prolyl-4-hydroxlyase inhibitor, increased HIF1α levels but did not restore glycolytic responses to LPS, suggesting that defects in glycolytic responses in our cells may be mediated by mechanisms upstream of defects in HIF-1α signaling.

The primary focus of our investigation was on palmitate as a driver of AM immune dysfunction. We chose to focus on this fatty acid because of its known abundance (~80%) in the lung, comprising the vast majority of free fatty acyl chains in bronchoalveolar lavage fluid and the majority of fatty acyl chains incorporated into surfactant phospholipids. However, we recognize that other lipid species might also accumulate in the lungs of individuals with alcohol use disorder and contribute to immune dysfunction.

Analogous to the liver, the mechanisms contributing to lipid accumulation in the chronic alcohol exposed lung have not been elucidated³⁰. Previously, we showed that various key components of the lipid synthesizing machinery were increased in the rat lung after chronic alcohol exposure and in lung epithelial cells after prolonged alcohol exposure in vitro (>48 hours). Importantly, individuals with alcohol use disorder are also known to develop dyslipidemia, and it is possible that leakage of lipids from the circulation might also contribute to changes in surfactant lipids in the lung. This concept is only further bolstered by work in the Burnham and Koval laboratories^{31,32,33,34,35}, demonstrating marked reductions in the alveolar-capillary barrier in both human and rodent lungs after chronic alcohol exposure.

Antioxidant therapy showed considerable success in restoring the M1 cytokine response to LPS in our model. This finding is also supported by existing data showing that antioxidants are effective at ameliorating lung injury in rodents chronically exposed to alcohol^36,37,38. However, we recognize that antioxidant therapies have not been effective in the treatment of any form of human lung disease, making it somewhat premature to promote this type of therapy for alcohol-related lung complications. Notably, the failure of antioxidant treatments might be explained by the fact that oxidative stress can also have beneficial effects on cells through the activation of cytoprotective stress responses. Along these lines, this might explain why cytokine response to LPS were blunted in AMs exposed to mitoTempo alone.

In summary, our findings suggest that EtOH and lipids conspire to compromise immune homeostasis in the lung and that targeting the mechanisms involved in lipid metabolism and processing may be a novel strategy for restoring immune homeostasis and reducing the incidence and severity of bacterial pneumonia in individuals with alcohol use disorder.

Highlights:

1. Elevated fatty acid levels alter glycolytic reprogramming to LPS in alcohol-exposed alveolar macrophages.

2. Palmitate exacerbates mitochondrial oxidative stress in alcohol-exposed alveolar macrophages.

3. Blocking the entry of fatty acids into mitochondria or reducing mitochondrial oxidative stress can restore glycolytyic and cytokine responses to LPS in alcohol-exposed alveolar macrophages.