Examining the role of ABC lipid transporters in pulmonary lipid homeostasis and inflammation

The critical role of ABCA1 in lung inflammation is evident from murine knockout models. Abca1-knockout mice exhibit interrupted lipid export, and a 70% reduction in plasma cholesterol and phospholipids due to virtually undetectable HDL and apoA-I levels, compared to wild-type [10]. Phenotypically, these mice at 4 months of age exhibit massive cholesterol and phospholipid accumulation in alveolar macrophages and ATII pneumocytes, as compared with wild-type littermates [10]. Furthermore, oil red-O staining of Abca1 ^−/− lungs revealed pulmonary focal lesions, and microscopic analysis revealed enlarged foamy alveolar macrophages and ATII cell hyperplasia, indicating alveolar proteinosis [10]. Physiologically, these mice experienced reduced tidal volume and hyperventilation, although it was unclear whether this respiratory distress was directly attributable to abnormalities in the lung or in other organs [10]. Nevertheless, these findings, coupled with similar but progressively severe observations in 7-, 12- and 18-month-old mice [31], suggest that absence of ABCA1 results in age-dependent progressive pulmonary disease. Supporting the importance of this ABCA1-mediated pathway in normal lung physiology is the finding that apoA-I^−/− mice have elevated inflammatory cell infiltration (particularly neutrophils and leukocytes), collagen deposition and airway hyper-responsiveness, and impaired pulmonary vasodilatation [32]. These observations strongly implicate a role of ABCA1-mediated RCT in inflammatory lung diseases.

The mechanism by which ABCA1 exports lipids has been predominantly studied in the context of its role in macrophage lipid homeostasis and the development of atherosclerosis. ABCA1 is thought to export lipids from the inner to the outer leaflet of the cell membrane via an ATP-dependent mechanism [12]. The subsequent method of transfer of these lipids to apoA-I is still open for debate. One model proposes that hydrolysis of ATP causes a conformational change in the transporter, enabling apoA-I to bind [33]; alternatively, apoA-I may bind and solubilise lipids in exovesiculated membrane lipid domains that arise from membrane asymmetry [34]. It is worth noting that evidence for the presence of apoA-I in the lung is currently limited. Bates and colleagues [28] were able to stain cryosections of murine lungs with anti-mouse apoA-I antibody, to verify that adequate levels of apoA-I exist for interaction with pneumocyte ABCA1. Meanwhile, Delvecchio and colleagues [30] performed in vitro experiments involving incubation of ASM cells with exogenous apoA-I, and were able to confirm the capacity of ASM ABCA1 to efflux lipids to apoA-I. Of further interest was the finding that apoA-I is reduced in sputum samples of patients with COPD compared to controls (which were smokers), with the authors suggesting that apoA-I could potentially be investigated as a biomarker [35]. Inhaled apoA-I mimetic peptides, which are under development as treatments for atherosclerosis, have been suggested as a potential treatment for inflammatory lung conditions [36].

Studies addressing the mechanisms by which ABCA1 activity affects inflammatory processes in the lung have investigated various pathways. Firstly, it has been proposed that excess cholesterol itself is pro-inflammatory and acts as the trigger for the cellular inflammatory response [37], and hence ABCA1 activity may simply be anti-inflammatory by removing excess cholesterol. Alternatively, other studies have investigated its transcriptional upregulation via LXR. In human alveolar macrophages, LXR-agonists that increased ABCA1 mRNA expression were shown to reduce LPS-induced airway inflammation and neutrophil production. However, explanations for the underlying mechanisms are conflicting. In one study, this was attributed to a NF-κB-dependent pathway [38], whereas another study found no involvement of either of the pro-inflammatory transcription factors NF-κB or AP-1 [29]. However, it is worthwhile noting that LXR also affects ABCG1 expression (covered in the next section), effects on which were not accounted for in the latter study.

Lastly, Dai and colleagues have investigated effects of ABCA1 overexpression in alveolar macrophages and showed that the anti-inflammatory effects of ABCA1 may occur via suppression of granulocyte-colony stimulating factor, which subsequently could reduce the involvement of inflammatory cells such as neutrophils [39]. Evidently, further research is required to clarify the mechanisms by which ABCA1 activity affects inflammatory pathways.

The association between ABCA1 function and asthma is still a novel and underdeveloped area. Mice overexpressing human ABCA1 exhibited reduced neutrophil count, IgE levels, peri-bronchial inflammation and airway epithelial thickness in response to daily ovalbumin challenges when compared to wildtype mice [39]. Furthermore, treatment of house dust mite-challenged mice with intranasal apoA-I was able to minimise eosinophilia, neutrophilia, interleukin (IL)-17E, IL-33 and airway hyperresponsiveness [40]. The same authors also revealed that apoA-I levels in bronchoalveolar lavage fluid of asthmatic patients were significantly lower than that from healthy subjects [40]. These findings suggest that upregulation of ABCA1 and/or apoA-I activity could be beneficial in patients with inflammatory lung diseases.

In summary, ABCA1 plays crucial roles in the maintenance of lipid homeostasis and modulation of inflammatory responses in lung cells. However, further studies are necessary to elucidate the exact mechanisms by which these effects occur and whether therapeutic targeting of these pathways will be effective.

There is strong evidence derived from Abcg1-knockout murine models to suggest that this transporter also plays an essential role in maintaining pulmonary lipid homeostasis. ABCG1 is widely expressed in various lung cell types, including alveolar macrophages, epithelial cells, ATII pneumocytes, ASM cells, T-lymphocytes and dendritic cells [43, 46, 47]. Several studies in Abcg1-knockout mice have demonstrated disrupted lipid homeostasis in murine lungs. Compared to heterozygote Abcg1 ^+/− mice, homozygote knockouts revealed massive accumulation of lipids in the alveolar macrophages. The subpleural region of Abcg1 ^−/− lungs also revealed build-up of lymphocytes, multinucleated giant cells, and macrophage foam cells containing an abundance of cholesterol clefts that indicated impaired lipid processing. These observations were progressive and age-dependent, and were exacerbated upon administration of a high-fat high-cholesterol diet [16, 43]. Furthermore, human ABCG1-transgenic mice overexpressing the ABCG1 transporter were protected from this diet-induced pulmonary lipidosis [43].

Absence of ABCG1 also results in progressive and chronic pulmonary inflammation [19]. Immunofluorescence studies and bronchoalveolar lavages obtained from chow-fed Abcg1 ^−/− mice revealed multiple markers of inflammation, including elevated levels of foamy macrophages, lymphocytes and pro-inflammatory cytokines. None of these features were observed in lungs of wild-type littermates [19]. Coupled with the observations that administration of a high-fat high-cholesterol diet significantly accelerated lipid deposition and induced macrophage cytokine expression in wild-type and (to an even greater extent in) Abcg1 ^−/− lungs, these findings imply that ABCG1 indirectly controls inflammation by modifying intracellular cholesterol levels [19]. However, the timeline of events was less clear in a separate study, which also reported inflammatory cytokine, neutrophil and lymphocyte infiltration alongside lipidosis in the lungs of Abcg1 ^−/− mice [48]. Wojcik and coauthors proposed that one of two mechanisms could be occurring: i) lipid accumulation causes inflammation or ii) absence of Abcg1 triggers inflammation that subsequently leads to the observed lipidosis [48]; which one of these events predominates is still unclear. Nevertheless, pulmonary ABCG1 expression also appears to protect against exogenous inflammation-inducing factors. Abcg1 ^−/− mice challenged with LPS or gram-negative bacteria displayed an exaggerated pulmonary response, characterised by elevated neutrophil, leukocyte, cytokine and chemokine levels in the airspace and tissue remodelling as compared to wild-type mice [49]. These effects were attributed specifically to disrupted Abcg1 ^−/− activity in alveolar macrophages. Despite the enhanced pulmonary bacterial clearance in these knockout mice, the excessive inflammation meant they also had a much higher mortality rate [49]. In a murine model of allergic asthma, ovalbumin-challenged Abcg1 ^−/− mice displayed increased airway neutrophils and IL-17, which are implicated in severe forms of asthma [46]. These findings suggest that ABCG1 plays a critical role in regulating the host defence response in the lung. Overall, results from murine models provide strong evidence for the critical role of ABCG1 in maintaining cellular lipid homeostasis and controlling pulmonary inflammation.

Amongst the myriad of inflammatory cytokines and chemokines released in asthma and COPD, several have also been identified in Abcg1-knockout mice. Levels of the inflammatory cytokines tumour necrosis factor (TNF)-α and IL-1β were markedly elevated in lungs of Abcg1-deficient chow-fed mice [19]. TNF-α is expressed in multiple lung cell types, including macrophages, lymphocytes, mast cells and ASM cells, and promotes oxidative damage that subsequently activates NF-κB and AP-1. Results of this activation include recruitment of adhesion molecules, lymphocytes, and other inflammatory cytokines, and increased airway hyperresponsiveness [50]. IL-1β has been shown to cause inflammation, airway fibrosis, bronchiolar thickening and mucus metaplasia – features that are prominent in COPD and asthma. Furthermore, IL-1β enhanced production of matrix metalloproteinases (MMP)-9 and −12 in neutrophils and macrophages of murine airways respectively [51]. MMPs that degrade extracellular matrix are overexpressed in patients with COPD and asthma, and have been associated with airway inflammation and remodelling [52]. Elevated levels of MMP-8 and MMP-12 have also been reported in lungs of Abcg1 ^−/− mice [19]. Alveolar macrophages also play a pivotal role in orchestrating the inflammatory processes in COPD and asthma by secreting numerous pro-inflammatory cytokines and chemokines [3]. However, they also secrete inhibitory mediators such as IL-10 that dampen inflammation, but this secretion is thought to be impaired in patients with asthma [53]. These activities were reflected in the alveolar macrophages of Abcg1 ^−/− mice, with Wojtik and coauthors reporting significantly elevated pro-inflammatory IL-1, IL-6 and IL-12 levels, and decreased expression of the anti-inflammatory cytokine IL-10 [48]. Another study indicated that following an asthma allergen challenge in Abcg1 ^−/− mice, airway neutrophil and IL-17 levels were elevated compared to wild-type [46]. Although it was not specified which IL-17 subtype was elevated, a separate study showed that IL-17A contributes to glucocorticoid-insensitivity by decreasing the activity of histone deacetylase-2, an enzyme that normally mediates the anti-inflammatory effects of glucocorticoids [54]. Taken together, evidence from Abcg1 ^−/− mouse models suggest that disruption to ABCG1 activity results in a phenotype reflective of inflammatory lung disease.

ABCG1 deficiency has been identified in patients with pulmonary alveolar proteinosis (PAP), a condition characterised by pulmonary surfactant accumulation [18]. Pulmonary surfactant is produced in lamellar bodies of ATII pneumocytes, and degraded by alveolar macrophages. Studies in chow-fed Abcg1 ^−/− mice showed accumulation of ATII pneumocytes that were enlarged and engorged with surfactant-rich lamellar bodies, suggesting abnormal surfactant clearance. Esterified cholesterol levels were also elevated in alveolar macrophages, which occurred despite increased Abca1 expression, which was thought to be due to compensatory upregulation. The authors hypothesised that ABCG1 disruption in both these cell types resulted in defective lipid/surfactant secretion, thereby resulting in compensatory cell hypertrophy and severe pulmonary lipidosis [16]. In 90% of PAP cases, autoantibodies that neutralise granulocyte macrophage-colony stimulating factor (GM-CSF) impair alveolar macrophage maturation, consequently hindering their ability to clear surfactant [55]. Thomassen and colleagues recognised that both PAP patients and GM-CSF-knockout mice demonstrate surfactant accumulation in alveolar macrophages, and ABCG1 deficiency, despite increased ABCA1 activity [18]. Induction of ABCG1 expression via LXRα was able to reverse intracellular lipid accumulation and restore lung compliance in GM-CSF-knockout mice [56, 57]. These results demonstrate that ABCG1 is critical for normal surfactant metabolism.

Altogether, the importance of ABCG1 in regulating lipid levels and inflammation in the lung is clear from murine studies, as well as the phenotype of patients with PAP. However, the underlying mechanisms and pathways by which ABCG1 mediates its protective effects are currently unclear.

Dysfunctional ABCA3 disrupts lamellar body phospholipid export, causing respiratory distress syndrome (RDS), and pediatric and adult interstitial lung disease [14, 60], which is a family of related conditions characterised by inflammation and fibrosis of the pulmonary interstitium [60]. RDS typically affects neonates, and is caused by surfactant deficiency that results in alveolar collapse, hypoxaemia and pulmonary oedema [14], hence this condition is often lethal. Inflammation accompanying RDS is fuelled by neutrophils and cytokines that are likely controlled via the NF-κB pro-inflammatory signalling pathway [64]. Corticosteroids can be used for their ability to stimulate surfactant phospholipid synthesis in surviving infants, but can have significant side effects. These compounds have also been found to upregulate ABCA3 expression and modulate transcription factors including NF-κB in chronic inflammatory lung diseases [58, 65]; these may be additional mechanisms by which they facilitate foetal lung maturation.

Several murine knockout studies have also confirmed the critical role for ABCA3 in surfactant production. Absence of mature lamellar bodies and surfactant phospholipids in the alveolar space of Abca3 ^−/− mice resulted in surfactant deficiency and fatal respiratory failure [15, 61]. One proposed mechanism for the disrupted surfactant production seen in RDS suggests that ABCA3 mutations elevate endoplasmic reticulum stress and trigger apoptosis of ATII cells [62]. Elaborating on this finding, a more recent study showed that ABCA3 expression protects ATII cells from free cholesterol-induced cytotoxicity by exporting free cholesterol. Cells transfected with a non-functional mutant of ABCA3, ABCA3-Q215K, displayed excessive accumulation of cholesteryl esters and total phospholipids. The study showed that the excess lipids in these cells were rerouted away from lamellar bodies towards the ER, where the subsequent lipid accumulation resulted in endoplasmic reticulum stress and thus ATII apoptosis [60]. There is also increasing evidence that ATII cell dysfunction and surfactant abnormalities are implicated in the pathogenesis of COPD. Surfactant dysfunction is thought to contribute to airway resistance, oxidative stress, inflammation and increased protease activity that promotes tissue remodelling [66]. A mouse model expressing a common ABCA3 mutation observed in patients with diffuse parenchymal lung disease, the ABCA3^E292V mutation, reported spontaneous lung remodelling in these mice with an emphysema-like phenotype [67]. However, no clinically significant ABCA3 mutations have been identified in COPD patients as yet [68].

Evidence is currently limited with respect to a potential role for ABCA3 in inflammation, despite suggestions that it may be involved in transport of cholesterol. Bronchoalveolar lavage cells from lungs of Abca3 ^−/− mice showed no statistically significant increase in expression of inflammatory cytokines, although measurements were only made for five cytokines [69]. In patients with ABCA3 mutations, levels of the representative cytokine IL-8 were significantly elevated in ATII cells as compared to normal cells [17]. It is unknown whether the increase in IL-8 was due to lipid accumulation within the ATII cells.

An alternate mechanism by which ABCA3-deficiency may lead to inflammation may be due to the associated reduction in the secretion of surfactant proteins. The hydrophilic collectins SP-A and SP-D have immunomodulatory functions, and have been suggested to reduce inflammation in the lungs by enhancing phagocytosis of apoptotic cells and pathogens, and inhibiting T-cell function. A deficiency in their secretion has been associated with the pathogenesis of COPD and asthma, although previous attempts at administering surfactants to asthma patients had mixed results [70‐72]. Chiba and colleagues also demonstrated that both SP-A and phosphatidylglycerol in surfactant have anti-inflammatory effects [72]. Since it is clear that ABCA3 is critical for lamellar body formation, it is possible that ABCA3 indirectly modulates lung inflammation by helping to form the organelles that are necessary to produce these anti-inflammatory surfactant components.

In summary, our understanding of a role for ABCA3 in regulating lung lipid levels and inflammation is not as expansive as that for ABCA1 and ABCG1. Its role in the lung has been well defined in the context of surfactant production, however how it modulates pulmonary inflammation warrants further investigation.

Recently, attention has been given to statins to explore their effect on lung inflammation in asthma and COPD. In addition to their lipid-lowering effects, statins have pleiotropic anti-inflammatory, antioxidant and anti-proliferative properties. In vivo studies in rodents have been promising, with simvastatin attenuating tobacco-induced infiltration of macrophages, neutrophils and leukocytes into the airways [9], and reducing the expression of macrophages, neutrophils, eosinophils, MMPs and various inflammatory cytokines in bronchoalveolar lavage fluid in ovalbumin-induced allergic asthma [73]. However, the benefits of statins in humans with inflammatory lung conditions have so far been inconsistent. Observational studies in humans have shown that statins can reduce exacerbation and mortality rates, and improve symptom control in patients with asthma and COPD [8, 74, 75]. Furthermore, a randomised placebo-controlled study in 12 COPD patients reported significantly reduced levels of inflammatory cells and mediators in the lung following atorvastatin treatment, although this did not translate into improved lung function [76]. On the other hand, one randomised double-blind crossover trial found no anti-inflammatory activity in asthma patients given simvastatin [77], while a large-scale randomised placebo-controlled trial in 884 patients concluded that simvastatin was ineffective in reducing exacerbations in moderate-to-severe COPD [78]. Two systematic reviews have reported statistically insignificant improvements in symptom control or lung function in asthma patients, however data did support statin-associated reductions in airway inflammation [79, 80]. A possible explanation for the inconsistent effects of statins may be their negative effects on ABC transporter expression. It has been well established that statin treatment in cells negatively affect ABCA1 and ABCG1 expression via an LXR dependent mechanism [81, 82]. This may have offset some of the benefits associated with their cholesterol-lowering and pleiotropic anti-inflammatory effects.