Regulation of the fluid interface occurs primarily through regulating Na⁺ uptake via ENaC in both AT1 and AT2 cells. After ENaC-mediated entry of Na⁺ across the apical membrane, Na⁺ leaves the cell across the basolateral membrane via the Na⁺–K⁺ ATPase and enters the interstitium where it is in equilibrium with vascular Na⁺. Some investigators have suggested that regulation of the ATPase also plays a role in controlling trans-epithelial Na⁺ transport (11–15); however, we will not consider ATPase regulation in this review. The paradigm in which vectorial Na⁺ transport is considered a primary drive for fluid transport from the alveolar surface has been established by numerous studies where pharmacological inhibitors of apical Na⁺ channels have been shown to reduce the rate at which fluid is cleared (16–21).

Epithelial sodium channel is composed of three homologous subunits, such as α, β, and γ. Together, these subunits assemble in the endoplasmic reticulum and traffic to the apical membrane and are highly selective for Na⁺ (22). Using ENaCα-subunit knock-out mice, investigators first showed the importance of ENaCα for proper lung function: neonates lacking ENaCα died within 40 h of birth (23). The α subunit is the ionophoric component of the heteromultimer and is required for the expression and assembly of functional ENaCs at the apical membrane. The importance of ENaC to normal lung function is underscored by the phenotype of several monogenetic disorders that affect ENaC. Patients with pseudohypoaldosteronism (PHA), a condition resulting from ENaC partial loss-of-function, were found to have twice the volume of airway surface liquid compared normal levels (24). Mice lacking the ubiquitin ligase, NEDD4-2, had increased levels of ENaC expression and increased ENaC-mediated current in AT2 cells (25). Additionally, overexpression of ENaCβ in an ENaCβ transgenic mouse model leads to airway dehydration and mucous obstruction, comparable to many features observed in cystic fibrosis (CF) (26). Together, these studies highlight the importance of proper ENaC expression and regulation for the airways.

Understanding the regulation of ENaC is significant for understanding lung fluid balance, as ENaC dysregulation is the source of pathological lung edema. In recent years, probably because monogenetic disorders often alter ENaC trafficking, much of the focus has examined how regulation of the number of channels at the apical membrane of alveolar epithelial cells can alter Na⁺ transport. However, since ENaC is an ion channel, regulating how much of the time the channel spends open (the open probability, P_o) is also important. Both Liddle’s syndrome and PHA type 1 (PHA 1) are conventionally described as changes in channel density (an increase and decrease, respectively); however, examination of single ENaCs in these two syndromes shows that ENaC P_o also changes. There are an observed increase in channel activity in Liddle’s (27) and a decrease in activity in PHA I (28). Steroid hormones increase Na⁺ transport and are often thought to do so by increasing subunit transcription and translation. Although Frindt and Palmer (29) have shown in Na⁺-transporting epithelial tissue that this is indeed true, the increase in subunit density accounts for less than 25% of the increase in trans-epithelial Na⁺ current implying that the remaining 75% is due to an increase in single channel P_o. Single channel recordings show that acute application of steroids dramatically increases single channel P_o (30–32).

Kleyman, Hughey, and their co-workers have shown that the α and γ subunits of ENaC must be proteolytically cleaved to be active loops (33–37). Some investigators have suggested that such cleavage might be a mechanism by which ENaC in the apical membrane could be regulated. In fact, proteolysis does appear to be required for ENaC to have any appreciable activity, and may be required for it to reach the membrane. As such, proteolysis appears to be, more or less, an all-or-none phenomenon: channels that are uncleaved are capable of little if any activity. However, under conditions of normal Na⁺ transport most channels are cleaved. Under these conditions, cleaved channels are capable of a wide range of activity by changing their P_o.

Changes in membrane ENaC can occur by changing the rate of insertion into the membrane after transcription and translation (38). However, in any time frame less than 24 h, ENaC in the membrane is altered by recycling from intracellular pools into the membrane (22) or internalization of ENaC into recycling or degradative pools. Removal of ENaC occurs primarily via the ubiquitin ligase, NEDD 4-2, which targets ENaC for removal and proteosomal degradation (39, 40).

Therefore, in this review, we address both the regulation of P_o in cleaved channels and change in membrane channel protein density.

The regulation of ENaC occurs via multiple, redundant systems to ensure that Na⁺ transport is not limited. ENaC is regulated by a many agents including transmitters interacting with G-protein-coupled receptors (GPCRs), circulating hormones, cytokines and chemokines, and reactive oxygen and nitrogen species. The regulation of ENaC via hormones and GPCRs is not a primary focus of this review, but we briefly review ENaC activation and regulation via steroids since their actions often interact with the activities of cytokines and chemokines.

In the lung, the glucocorticoid receptor (GR) is the primary receptor for corticosteroids (41–43). Once activated, the GR activates response elements inducing the transcription of signaling kinases, such as the serum- and glucocorticoid-regulated kinase 1 (30, 44, 45). Ligand-mediated activation of the GR via corticosteroids is used clinically as an anti-inflammatory treatment. The positive effects of corticosteroid therapy lie in the ability of the GR to bind to and inhibit nuclear factor kappa-light-chain-enhancer of activated B cells (Nf-κB) (46–48). Nf-κB is an important mediator of cytokine signaling. This transcription factor increases cyclooxygenase 2-induced prostaglandin production, as well as increases other proinflammatory factors (49). Corticosteroids reduce inflammation propagated via Nf-κB-mediated mechanisms but may not affect inflammation mediated by other signaling pathways. Indeed, GR activation may actually augment some downstream signaling pathways, such as those mediated through Smad proteins (50). This distinction is important because of the many heterogeneous pathways activated by each cytokine (51–54).

During inflammation most cells are capable of secreting a variety of small molecular weight proteins, called cytokines and chemokines, which communicate the inflammatory signals. In the airway, resident immune cells are mostly the alveolar macrophages; however, during infection or inflammation, other mononuclear and granular immune cells infiltrate (56). Several studies have proposed a role for the airway epithelium in propagating the immune response, especially as a “first responder” since the airway is the first to sense viral and bacterial pathogens as they enter the body. This layer can be an active participant in the immune response, producing a variety of cytokines and chemokines, as well as exclusive epithelial-derived cytokines (55, 57, 58).

Direct interaction with pathogens, such as influenza, reduces ENaC activity (59). However, other evidence suggests that some of the more chronic effects of pathogens may be via noxae-stimulated chronic cytokine production (60–62). Additionally, inflammatory activation of the airway epithelium can result in local nitric oxide (NO) production, most likely via increased cytokine production, further reducing ENaC activity and fluid transport (59, 63–67). Cytokines frequently increase local levels of reactive oxygen species (ROS) as well. Interestingly, ROS has been shown to activate ENaC at relatively low concentrations but to inhibit ENaC at higher concentrations often associated with massive pathogen-induced cytokine production (40, 68, 69). The overall redox environment of the alveoli is crucial and can rapidly change, often driven by high levels of Rac1-NADPH oxidase activity in AT1 cells (67, 70).

The TNF super family comprises 19 members and was originally named for its role in apoptosis (53). The best-studied member of this family is TNF-α, which plays a role in propagating the immune response and secretion of other cytokines. TNF-α was implicated in the pathogenesis of pulmonary edema, and increased levels were observed in patients with ARDS (72, 73). Monocytes and macrophages produce significant TNF-α, but it is also produced by alveolar epithelial cells following lipopolysaccharide stimulation (74).

Although TNF-α can bind to two different receptors that are linked to separate signaling pathways, much of the work in the airway has focused on TNF receptor 1. The effect that TNF-α elicits on ENaC function, and alveolar liquid clearance, seems to be critically dependent upon receptor activation or receptor-independent mechanisms and has been shown using both in vitro and in vivo models (75–78).

Tumor necrosis factor receptor 1-mediated activation of NF-κB increases cytokine (IL-1, IL-8, IL-6) and chemokine production. It also increases the expression of adhesion molecules including selectins, vascular cell adhesion molecules, and intercellular adhesion molecule (ICAM)-1 (53, 79, 80). In freshly isolated AT2 cells, TNF-α decreased α- and γ-ENaC mRNA and protein levels and reduced amiloride-sensitive trans-epithelial current (75, 78).

Tumor necrosis factor-α also plays an especially important role in endothelial activation, as well as disturbing the epithelial tight junction barrier. Disruption of the tight junctions not only leads to respiratory distress and increased exudate but also may reduce alveolar fluid clearance as well (81). TNF-α reduces the expression of tight junction proteins, including the claudins and zonula occludens protein 1, thus increasing alveolar permeability (55, 82). Consequently, TNF-α has a critical and multi-faceted role in the development of ARDS. TNF-α not only regulates Na⁺ and water clearance but also disrupts tight junction barriers and endothelial integrity and contributes to a pro-inflammatory environment.

Interestingly, TNF-α contains not only a receptor-binding domain but also a lectin-like domain (referred to as a TIP domain) that is spatially distinct from the receptor-binding site (83, 84). TNF-α produces an opposite response when there is binding of the lectin-like domain, or TIP, to certain oligosaccharides at high concentrations of TNF-α. This process increases Na⁺ uptake in AT2 cells and may account for the differential responses to TNF-α (85–87). In vivo, a peptide analog of TIP increased clearance in a murine flooded-lung model (85). Czikora and colleagues also demonstrated that this TIP domain directly binds to and, then, activates ENaC (83). This implies an endogenous mechanism to limit the effects of high TNF-α concentrations. Use of this may become a novel method in counteracting reduced alveolar clearance.

Transforming growth factor-β1 is a pathogenic cytokine, which has been implicated in the early phase of acute lung injury (ALI) prior to ARDS (72, 88). TGF-β1 levels were increased in ARDS patients compared to healthy controls (89). Furthermore, active TGF-β1 levels were more than doubled in the epithelial lining fluid from ARDS patients (90). As mentioned earlier, corticosteroids are a common tool to reduce inflammation and aid in lung clearance. Interestingly, TGF-β actually reduces the ability to produce multiple steroids, possibly leading to the inability for self-healing and furthering inflammatory damage, in addition to the activation of multiple Smad pathways (50, 52). Some of these pathways may be insensitive to corticosteroid treatment. However, there is still much to learn regarding which Smad-mediated pathways are downstream of TGF-β signaling during ALI and ARDS.

Other studies have specifically have explored the role of TGF-β in alveolar flooding. Using a bleomycin-induced lung injury model, TGF-β1-inducible genes were dramatically increased as early as 2 days, suggesting that TGF-β1 may precede alveolar flooding (91). Of interest, TGF-β may actually remain latent locally, covalently attached to a latency-associated peptide (LAP); pulmonary epithelial cells can activate and cause dissociation of TGF-β from LAP (92–94). One member of the integrin family, ανβ6, was recently shown to be a ligand for LAP (93). ανβ6 is expressed normally at lower levels, yet increased significantly with injury revealing a novel mechanism for rapid and local TGF-β activation (95). TGF-β is also redox sensitive, and in vitro models of increased ROS via ionizing radiation revealed another mechanism for TGF-β activation (96). Together, these studies show multiple, redundant possibilities for systemic and paracrine TGF-β activation during lung injury.

One of the first studies to directly implicate TGF-β1 in regulating ENaC was by Frank and colleagues. They showed that TGF-β1 reduced amiloride-sensitive Na⁺ transport in lung epithelial cells. Additionally, TGF-β1 reduced αENaC mRNA and protein expression via an ERK1/2 pathway in a model of ALI, thus promoting alveolar edema (97). In vivo studies then showed that TGF-β1 reduces vectorial Na⁺ and water transport and that this process occurs independently from increases in epithelial permeability (97, 98). Interestingly, TGF-β was also found to have an integral role in ENaC trafficking. Peters and colleagues were the first to demonstrate this acute regulation of ENaC in the lung; they found that TGF-β induces ENaC internalization via interaction with ENaCβ (99). In summary, TGF-β has been implicated in multiple mechanisms reducing ENaC expression and apical localization, thus contributing to the pathophysiology of ARDS and pulmonary edema (100).

The interferons (IFN) are a family of proteins originally classified by their ability to reduce viral replication. This family consists of both Type I and Type II IFNs; INF-γ is the only member of the Type II IFN family and is structurally unrelated to the other IFNs. During inflammation, INF-γ is secreted by multiple immune cells, but mostly by T lymphocytes. INF-γ increases ICAM-1 levels and increases NO production via inducible nitric oxide synthase. Little is known about the role of INF-γ in ENaC regulation; however, studies using human bronchial epithelial cells (BECs) showed that INF-γ treatment significantly reduced trans-epithelial Na⁺ transport in normal human BECs (101).