Lysophospholipids in Lung Inflammatory Diseases

Abstract

The lysophospholipids (LPLs) belong to a group of bioactive lipids that play pivotal roles in several physiological and pathological processes. LPLs are derivatives of phospholipids and consist of a single hydrophobic fatty acid chain, a hydrophilic head, and a phosphate group with or without a large molecule attached. Among the LPLs, lysophosphatidic acid (LPA) and sphingosine-1-phosphate (S1P) are the simplest, and have been shown to be involved in lung inflammatory symptoms and diseases such as acute lung injury, asthma, and chronic obstructive pulmonary diseases. G protein–coupled receptors (GPCRs) mediate LPA and S1P signaling. In this chapter, we will discuss on the role of LPA, S1P, their metabolizing enzymes, inhibitors or agonists of their receptors, and their GPCR-mediated signaling in lung inflammatory symptoms and diseases, focusing specially on acute respiratory distress syndrome, asthma, and chronic obstructive pulmonary disease.

20.1. Introduction

Lysophospholipids (LPLs) are derivatives of phospholipids. Much attention has been paid to phospholipids and their roles in maintaining biological membrane structure. Most phospholipids contain a glycerol backbone that has three carboxyl positions (sn). Two fatty acid chains are esterified to positions 1 (sn-1) and 2 (sn-2), while a phosphate group is attached to sn-3. Phosphatidic acid (PA) is the simplest phospholipid with only a phosphate group attached to its sn-3 position. Some other organic molecules can link to the phosphate group to generate phosphatidylcholine (PC), phosphatidylserine (PS), phosphatidylglycerol (PG), etc. The other phospholipid form is based on a sphingoid backbone, such as sphingomyelin. LPLs differ from phospholipids due to lack of a fatty acid chain in either sn-1 or sn-2. Lysophosphatidic acid (LPA) is the simplest LPL, consisting of one fatty acid chain and a hydroxyl group in sn-1 or sn-2, and a phosphate in the sn-3 position. Depending on the molecules attached to the phosphate, LPLs may contain LPC, lysoPS, LPG, sphingosine-1-phosphate (S1P), etc. Numerous enzymes are involved in the metabolism of LPA and S1P, which contribute to homeostatic regulation of LPLs. In this chapter, we will focus on LPA and S1P, while the role of other LPLs in lung inflammatory diseases will be briefly discussed.

Changes of LPLs in biological fluids, including plasma, have been reported (reviewed in [1–7]). Extracellular LPLs may trigger intracellular signaling pathways and a wide spectrum of biological activities through ligation to specific plasma membrane receptors. G protein–coupled receptors (GPCRs) mediate LPL-induced biological responses, although not every receptor for LPLs has been identified. So far, receptors for LPA and S1P have been well characterized and extensively studied. LPA receptors consist of LPA_1–6, and S1P receptors consist of S1P_1–5. LPA_1–3 and all S1P receptors belong to the endothelial cell differentiation gene (Edg) family. S1P₁ (also called Edg1), S1P₂ (Edg5), S1P₃ (Edg3), S1P₄ (Edg6), and S1P₅ (Edg8) bind to S1P, whereas LPA₁ (Edg2), LPA₂ (Edg4), and LPA₃ (Edg7) are specific receptors for LPA. Three non-Edg receptors (LPA_4–6) are members of P2Y family. Agonists and antagonists of LPA and S1P have been developed to activate or interfere LPA/LPA receptors or S1P/S1P receptors. In addition, intracellular LPA is an endogenous ligand for peroxisome proliferator-activated receptor gamma (PPARγ) (reviewed in [8–10]).

The lungs, as the pivotal organs in the respiratory system, facilitate gas exchange between the atmosphere and blood stream. Lungs consist of a conducting zone and respiratory zone. Trachea, bronchi, and bronchioles function as airways for bulk air flow. In addition, airway epithelial cells clear away inhaled pathogens or irritants through mucus secretion and cilia movement. Alveolus in the respiratory zone is the site for gas diffusion between lungs and capillaries. Alveolar epithelial cells release surfactant to maintain alveolar space. Capillary endothelial barrier integrity is important to prevent plasma and erythrocyte leakage into alveolar spaces. Interstitial fibroblast cells in the respiratory zone function as structural cells, and play a role in lung repair and remodeling after injury [11]. Impaired oxygen diffusion is a hallmark of lung disorders [12].

In this chapter, we will discuss role of LPLs in the pathogenesis of lung inflammatory diseases, such as mild acute respiratory distress syndrome [mild acute respiratory distress syndrome (ARDS), previously called acute lung injury (ALI)], asthma, and chronic obstructive pulmonary disease (COPD). LPA and S1P have gained attention due to their pathogenic role in pulmonary fibrosis; thus, we will include pulmonary fibrosis in this chapter.

20.2. Metabolism of LPA and S1P

20.2.1. Biosynthesis of LPA

LPA contains one fatty acid chain in either the sn-1 or sn-2 position. Commonly, saturated and monounsaturated fatty acids are in the sn-1 while polyunsaturated fatty acids are linked to the sn-2 position. LPA can be synthesized both inside and outside of cells.

Inside cells, three pathways regulate LPA synthesis that is achieved through breaking down PA, adding a phosphate group to monoacylglycerol (MAG), or de novo synthesis. Phospholipases are a group of enzymes that hydrolyze phospholipids. In the first pathway, phospholipase A1 (PLA₁) or PLA₂ converts PA to LPA by removing one of fatty acid chains at either the sn-1 or sn-2 positions of PA. It has been shown that two membrane-bound PA-specific mPLA₁α and mPLA₂β regulate LPA synthesis [13, 14]. Both enzymes are localized in the plasma membrane, especially in lipid rafts, suggesting PLA_1/2-mediated intracellular LPA generation’s involvement in signaling within protein-enriched membrane domains. Thus, increase in PA levels may lead to upregulation of intracellular LPA. Activations of phospholipase C (PLC), phospholipase D (PLD), or diacylglycerol kinase (DGK) generate PA, thus leading to the formation of LPA [15]. Another pathway for LPA production is phosphorylation of MAG by acylglycerol kinase (AGK), which is a novel lipid kinase in the mitochondria [16]. The molecular mechanism regarding LPA exports from mitochondria is not clear. De novo synthesis of LPA is the third pathway. LPA is formed by linking a fatty acid to glycerol 3-phosphate (G3P) in the sn-1 position by glycerol 3-phosphate acyltransferases (GPATs) [17, 18].Out of these three pathways, conversion from PA is the major contributor to intracellular LPA generation.

LPA can be generated outside of cells. Extracellular LPA triggers intracellular signaling pathways and is attributed in a wide range of cellular responses. LysoPLD is also known as autotaxin (ATX) or ectonucleotide pyrophosphatase/phosphodiesterases family member 2 (ENPP2) (reviewed in [19, 20]. ATX heterozygous knockout mice exhibit significantly reduced plasma LPA levels [21, 22], while recombinant ATX demonstrates increased extracellular LPA levels [23]. ATX generates LPA through hydrolyzing LPC, which is enriched in plasma and bronchoalveolar lavage. ATX is a secreted glycoprotein and its level changes in various human disorders including lung diseases and cancer [1, 24, 25]. Another source of extracellular LPA is from activation of secretory PLA₂ [26].

20.2.2. Catabolism of LPA

Two major pathways have been identified to limit LPA levels either inside or outside of cells. Inside the cells, a reversible reaction allows LPA conversion to PA by LPA acyltransferases (LPAATs), which include four isoforms [27, 28]. This pathway facilitates increases of intracellular PA levels, which is also a bioactive lipid and regulates signaling pathways.

Another limiting pathway is to eliminate extracellular LPA levels. A group of lipid phosphatases (LPPs) catalyze LPA dephosphorylation to generate MAG. LPP1–3 are three major isoforms of LPPs, which are transmembrane proteins with their enzymatic activity domain lying outside of the plasma membrane ([29], reviewed in [30, 31]).

20.2.3. Synthesis of S1P

S1P is based on a sphingoid backbone and is synthesized intracellularly. Sphingosine kinases (SphKs) utilize sphingosine as a substrate and transfer a phosphate to synthesize S1P. SphKs belong to the group of lipid kinases. Two isoforms of SphKs (SphK1 and SphK2) have been identified and well characterized (reviewed in [32, 33]). SphK1 is predominately localized in the cytoplasm, while SphK2 mainly resides in nucleus [4, 34, 35]. S1P is also generated in mitochondria by SphK2, suggesting that SphK2 is also expressed in mitochondria [36]. Both SphK1 and SphK2 are expressed in lungs [37, 38]. The two kinases compensate for each other to maintain normal S1P levels. S1P can be released outside of cells by membrane transporters. ABC transporters (ABCA1, ABCB1, ABCC1, and ABCG2) and spinster homolog 2 (Spns2) export intracellular S1P ([39], reviewed in [40, 41]. A recent study revealed that a new S1P transporter (named Mfsd2b) plays a critical role in export of S1P within erythrocytes and platelets. Deletion of Spns2 or Mfsd2b in mice significantly eliminates plasma S1P levels [42, 43].

20.2.4. Catabolism of S1P

Similar to LPA, catabolism of S1P occurs intracellularly and extracellularly. S1P lyase (SPL) is localized within the endoplasmic reticulum membrane and its activity site is on the cytoplasmic side. SPL plays a major role in the elimination of cellular S1P levels by degrading S1P to phosphoethanolamine and hexadecenal [44, 45]. Lack of SPL expression in erythrocytes and platelets causes higher concentrations of S1P compared to other cells. S1P is degraded by SPL to ethanolamine phosphate and hexadecenal [46]. This reaction is irreversible and dependent on pyridoxal phosphate.

In addition to dephosphorylating LPA, LPPs convert S1P to sphingosine. LPP functions are not only to eliminate extracellular S1P levels, but also to facilitate uptake of sphingosine by cells, sequentially increasing intracellular S1P by phosphorylation via SphKs. In addition to LPP1–3, two S1P phosphatases (SPPases) have been identified to dephosphorylate S1P and other sphingolipids ([47, 48], reviewed in [49]). Though both SPPases and LPPs catalyze the catabolism of S1P, there is very little homology between these two groups, except the conserved active sites.

20.3. LPA- and S1P-Mediated Signaling Pathways

20.3.1. LPA and S1P Receptors in Lungs

Extracellular LPA triggers a wide range of biological functions through ligating and activating its specific GPCRs. Among the Edg and non-Edg receptors, LPA₁ and LPA₂ have been well studied in various lung cell types. Heterotrimeric G proteins, including Gα, Gβ, and Gγ, couple to the intracellular potion of LPA receptors. Among the G proteins, Gα plays a major role in determining downstream signaling. Depending on which G subunit is bound to LPA receptors, LPA may exhibit various, even opposite, biological functions (reviewed in [6, 8, 50]). PPARγ has been identified as an intracellular LPA receptor [10], while the LPA/PPARγ pathways in lung diseases have not been revealed. This chapter will focus on LPA₁ and LPA₂ in lung diseases. LPA₁ and LPA2 are potential targets for treating lung diseases. Several antagonists of LPA_1–3, such as AM966, ki14625, BMS-986278, H2L5186303, VPC32183, and VPC12249, have been developed to inhibit LPA receptor-mediated signaling.

All S1P receptors (S1P_1–5) belong to the Edg family. Similar to LPA receptors, S1P receptors are coupled with G proteins [8]. Expression of S1P receptors in different cell types of lungs has been determined. Most studies have been focused on the role of S1P receptors in lung endothelial cells and immune cells. FTY720-phosphate, a phosphorylated chemical of a fungus metabolite, is a S1P mimetic. FTY720 has been approved by FDA as an immunomodulator drug in treating multiple sclerosis as it can be phosphorylated by SphK2 and activate S1P₁ [51–53].

20.3.2. LPA and S1P Receptor-Mediated Signaling

As discussed above, LPA and S1P receptors both utilize Gα-mediated signaling, thus, parts of LPA- and S1P-triggered signaling and biological functions are comparable. The role of LPA and S1P in tumorigenesis has been well documented (reviewed in [54–56]). They activate Ras-Raf-Erk1/3 and PI3K-AKT pathways, increase cell proliferation, and function as pro-oncogenic factors (reviewed in [54–56]). S1P antagonizes the proapoptotic action of ceramide and impairs the caspase-dependent proapoptotic pathway [57]. LPA reduces proapoptotic Bax protein levels in the cytosol and increases anti-apoptotic Bcl-2 expression [58]. In addition to the activation of GPCRs, LPA and S1P trigger cross-talking between their receptors with epidermal growth factor recetpor (EGFR) [59] and platelet-derived growth factor recetpor (PDGFR) [60], which contribute to cancer progression.

Both LPA and S1P regulate inflammatory responses. Transcriptional factor nuclear factor kappa B (NF-κB) is critical for cytokine gene expression (reviewed in [61]). Activation of LPA and S1P receptors induces phosphorylation of I-κB and degradation, thus leading to NF-κB nuclear translocation and transcriptional activation [29, 62–64]. Other transcriptional factors, such as AP1, p38 MAPK, and cAMP-response element binding protein (CREB), are common downstream molecules of LPA and S1P receptors (reviewed in [65–67]. LPA and S1P treatment induces cytokine release and MMPs expression in a variety of cell types including lung cells.

Although both LPA and S1P exposure triggers activation of the Rho family of GTPases, including Rho, Rac, and Cdc42 in various cells [67–70], the immediate effects on the cytoskeleton may differ between different cell types. It has been shown that S1P enhances endothelial barrier integrity through activation of Rac1 [71, 72], while LPA increases endothelial permeability through activation of myosin light chain (MLC) via RhoA-mediated phosphorylation [73, 74]. In contrast, in bronchial epithelial cells, LPA induces E-cadherin accumulation at cell–cell contacts and reduces cells’ permeability [75]. Despite the distinct effects on cell–cell contact in different cell types, both LPA and S1P have shown to increase cell migration in most cell types, including lung epithelial, endothelial, and fibroblast cells.

The molecular regulation of LPA and S1P receptors has not been well studied. Gene regulation of these receptors has been reported in lung diseases, but the molecular mechanisms remain unclear. Protein stability and internalization of LPA and S1P receptors were demonstrated. Ubiquitin E3 ligases, Nedd4L and WWP2, are responsible for LPA₁ and S1P₁ ubiquitination and degradation [76], while deubiquitinating enzyme USP11 stabilizes LPA₁ [76].

20.4. LPA and S1P in Acute Respiratory Distress Syndrome (ARDS)

20.4.1. Pathogenesis of ARDS

ARDS is a severe condition characterized by acute inflammation and alveolar-capillary barrier disruption, leading to edema and gas exchange failure in the lungs. ARDS can be induced by inhalation of airborne pathogens such as bacteria or viruses. SARS-CoV2-induced COVID-19 has a high association with ARDS. Systemic inflammatory diseases such as sepsis also lead to ARDS. There are no effective treatments for ARDS, and the mortality rate of ARDS remains 30–40% [77]. For the past several decades, supportive therapies such as mechanical ventilation through extracorporeal membrane oxygenation (ECMO) have been essential treatments for ARDS [77]. Acute lung injury is a mild form of ARDS. Researchers have focused on investigating the pathogenesis of acute lung injury and are seeking new therapeutic strategies to treat this severe lung disease.

20.4.2. Pro- and Anti-inflammatory Roles of ATX-LPA-LPA Receptor Axis in Experimental Acute Lung Injury

The role of ATX/LPA in the pathogenesis of acute lung injury is controversial; however, there is solid evidence that LPA receptors are pro-inflammatory in experimental acute lung injuries. The functional disconnect between LPA and its receptors is not clear. It is possible that other ligands for LPA receptors have not been identified.

The levels of LPA and ATX in bronchoalveolar lavage fluid (BALF) are increased in experimental acute lung injuries caused by inhalation of lipopolysaccharide (LPS), bleomycin, or exposure to hyperoxia [78–81]. ATX plays distinct roles in different lung injury models. Pulmonary NKT cells have been shown to be a source of ATX and LPA in hyperoxia-induced lung injury [82]. Injection of Brp-LPA, which is an ATX inhibitor and LPA receptor antagonist, significantly improved survival and alleviated lung injury [82]. Another study confirmed that ATX levels were increased in hyperoxia-challenged 4-day-old rat pups. These studies indicate the ATX/LPA/LPA receptor axis plays a proinflammatory role in hyperoxia-induced lung injuries [81]. However, this conclusion is not supported by studies in LPS-induced experimental acute lung injury. Mouratis M-A et al. showed that lung epithelial cells are a source of ATX in BALF in response to inhalation of LPS [78]. Overexpression of ATX in lung epithelial cells increased two-fold of LPA levels in BALF [78], suggesting that ATX released from epithelial cells plays a role in LPA generation within BALF during periods of inflammation. The release of ATX in lung epithelial cells was confirmed in an in vitro study [83]. Interestingly, modulation of ATX levels in lung epithelial cells did not seem to contribute to the pathogenesis of LPS-induced lung injury [83]. In contrast, the overexpression of systemic ATX increased susceptibility to LPS-induced lung injury, evidenced by increased BALF cellularity, protein levels, and neutrophil infiltration into lungs. However, pharmacologic targeting of ATX had minor effects in lung injury severity [83]. These data indicate that ATX plays a role in the process of LPS-induced lung injury independent of LPA generation. This conclusion has been confirmed by another study showing that increased ATX activity is not required for BAL LPA production following bleomycin-induced lung injury [84]. An explanation for the phenomenon is that the extracellular ATX can interact with LPA₁ and trigger LPA₁-mediated signaling in an ATX activity-independent manner [83]. Thus, the proinflammatory effects of ATX may be due to directly ligation to LPA₁. As discussed above, extracellular LPA also can be generated via the activation of secretory PLA₂ [26]. Secretory group V PLA₂ has been shown to play a critical role in LPS-induced acute lung injury [85], indicating that the secretory PLA₂/LPA axis plays a role in the pathogenesis of acute lung injury. Secretory PLA₂’s potential as a limiting enzyme in BAL LPA needs to be explored in the future. Evidence directly supporting the proinflammatory effect of LPA is that IT LPA induced neutrophil influx into lungs, though the effect is not comparable with the effects of IT LPS [86].

In addition to its proinflammatory properties, the role of LPA in attenuation of acute lung injury has been revealed. One study in which LPA was directly injected into murine lungs following LPS-induced acute lung injury showed that post-treatment LPA played a protective role [75]. Consistent with this conclusion, LPA administration has been reported to demonstrate a protective role in acute liver injury, and the effects of LPA were reported in an LPA receptor-independent manner [87]. The effect of LPA on lung epithelial barrier enhancement may explain the protective role of LPA in experimental acute lung injury.

In addition to changes in LPA and ATX, LPA receptor levels were increased in lung cells in both hyperoxia- and LPS-induced experimental acute lung injuries [79, 81]. The studies using LPA receptor-deficient mice or antagonists provided solid data to support that LPA receptors, especially LPA₁, act as proinflammatory GPCRs during the progression of hyperoxia- or LPS-induced acute lung injury and sepsis [79, 88, 89]. Interestingly, LPA₁ seems to have no effects on alveolar-capillary integrity as no changes of LPS-induced BAL protein were detected in LPA₁-deficient mice or ki16425-treated mice [79]. This may be due to the opposite effects of LPA on epithelial and endothelial barrier integrity [75]. Notably, all the studies used whole body LPA receptors knockout mice. Use of lung epithelial or endothelial cell specific knockout mice will clearly demonstrate in which cell type LPA receptors play roles in lung injury.

20.4.3. Role of ATX-LPA-LPA Receptor Axis in Biological Functions in Acute Lung Injury-Related Lung Cells

LPA treatment increases cytokine release in bronchial and alveolar epithelial cells. The proinflammatory effect of LPA occurs through activation of Gαi-coupled LPA receptors. Activation of transcriptional factors, NF-κB, AP-1, and p38 MAPK, by LPA regulates the expression of cytokines and MMPs. A variety of intracellular signaling pathways, such as activation of PLD or PKCδ, which results in increases of intracellular calcium, are involved in LPA-induced activation of transcriptional factors (reviewed in [6]). Intriguingly, LPA1 activation leads to phosphorylation of EGFR, and this crosstalk between GPCR and tyrosine kinase receptor contributes to IL-8 release through the activation of CREB [59].

Other major findings regarding the role of LPA in lung epithelial cells are that LPA induces lung epithelial repair and remodeling, such as epithelial barrier enhancement [75] and cell migration [83, 90]. PKCδ, PKCζ, and focal adhesion kinase (FAK) regulate LPA-induced E-cadherin accumulation at cell–cell contacts [75]. Rac1 is involved in LPA-induced lung epithelial cell migration [90]. The effects of LPA are consistent with its properties as a growth factor.

Vazquez-Medina JP et al. showed that LPA increases oxidant generation through the activation of NADPH oxidase type 2 (NOX2) in pulmonary microvascular endothelial cells [91]. A study demonstrated that LPA increases permeability in human pulmonary arterial endothelial cells (HPAECs), but not in human lung microvascular cells (HLMVECs) [92]. Cai J et al. showed that HLMVECs reduce barrier integrity in response to LPA treatment. Intriguingly [74], they also revealed that AM966, an LPA₁ antagonist, exhibits similar effects to LPA. Both LPA and AM966 activate RhoA and induce phosphorylation of MLC and VE-cadherin, which are critical factors for endothelial barrier dysfunction [74]. This warning study raised caution for using AM966 as an LPA receptor antagonist.

20.4.4. Protective Role of S1P in Experimental Acute Lung Injury

S1P levels in lung tissues and BALF, not in plasma, were upregulated in intratracheal (IT) LPS-induced acute lung injury [93]. Plasma S1P was increased in a two-hit model induced by intraperitoneal (IP) LPS combined with ventilation [94]. A recent study found that Pseudomonas aeruginosa challenge increases S1P levels in lungs and BALF [95]. However, a study using human samples showed an opposite phenomenon. Analysis of serum S1P from 121 ARDS patients and 100 healthy individuals revealed that serum S1P levels were decreased in ARDS patients [96]; however, the BAL S1P levels in patients were not measured in this study. The role of S1P in acute lung injury was studied by modulating expression of S1P metabolism enzymes or injection of S1P in various models of lung injury induced by LPS, P. aeruginosa, mechanical ventilation, or radiation [4, 93, 95, 97, 98]. As we discussed above, S1PL is a limiting enzyme of S1P degradation. S1PL expression is enhanced by LPS challenge and mechanical ventilation. S1PL heterozygous knockout mice increased S1P levels in lung tissues and BALF while reducing LPS- or mechanical ventilation-induced lung injury and inflammation [93, 98]. The effect in S1PL heterozygous knockout mice was confirmed by administration of S1PL inhibitor THI (2-acetyl-5-tetrahydroxybutyl imidazole) in an LPS-induced a murine model of acute lung injury [93].

Synthesis of S1P is catalyzed by SphK1/2. Severe Plasmodium falciparum malaria causes lung edema and an increase in SphK1 in lung tissues [99], suggesting a role of the SphK1/S1P axis in lung edema. The effects of SphK1/2 on acute lung injury were examined using SphK1/2 deficient mice. IT LPS increased SphK1/2 expression in lung tissues. SphK1 knockout mice exhibited more susceptibility to LPS-induced lung injury [100]. This was rescued by SphK1 overexpression, but not by overexpression of SphK2 [100]. In addition to its anti-acute lung injury property, SphK1 has been reported to contribute to the pathogenesis of lung injury. IP administration of SphK1 inhibitor exhibited protective effects on the two-hit (ventilation + IP LPS)-induced acute lung injury [94]. Gutbier B et al. reported that SphK1-deficient mice had reduced lung hyperpermeability in a P. aeruginosa-induced murine model of acute lung injury [101]; however, another controversial study demonstrated that SphK1 knockout had no effects in P. aeruginosa-induced acute lung injury [95]. Ebenezer DL et al. revealed that SphK2 deficiency attenuated P. aeruginosa-induced acute lung injury, indicating a role of nuclear S1P in the pathogenesis of lung injury as SphK2 is localized in the cell nuclei [95].

Similar to the effects of SphKs, controversial conclusions regarding the effects of administration of S1P on acute lung injury have been drawn by different studies. Intravenous (IV) or IT S1P reduces IT LPS-induced edema and neutrophil influx into lungs [102, 103]. In the canine model, IV S1P attenuated IT LPS- or ventilation-induced edema and neutrophil infiltration into the lungs without altering BAL cytokine profile [103, 104]. Intriguingly, IV S1P increased serum proinflammatory cytokines [104], which raised concerns about the use of S1P as a therapy.

With the controversial studies in SphK1/2 and administration of S1P, it is possible that S1P may exhibit distinct effects due to ligation to different receptors. S1P₁ heterozygous mice potentiated LPS-induced lung injury, while S1P₂ knockout mice or downregulation of S1P₃ exhibited an opposite response compared to S1P₁ heterozygous mice [102], indicating that S1P₁ exhibits a protective role, while S1P_2/3 promotes LPS-induced lung injury. FTY720, an analog of sphingosine, is an FDA-approved drug for treating multiple sclerosis. FTY720 can be phosphorylated to phospho-FTY720 by SphK1 and has an endothelial barrier protective effect (reviewed in [105]). In a hindlimb ischemia reperfusion (IR)-induced acute lung injury model, pretreatment with FTY720 attenuated lung injury [106]. However, similar to S1P, FTY720 caused barrier disruption at higher concentrations and increased airway hyper-responsiveness [107]. (S)-FTY720-phosphoate, an analog of FTY720, prolongs S1P₁ levels on the cell surface and exhibited more protective effects compared to FTY720 [108]; thus, (S)-FTY720-phosphate might be developed as a potential therapy to treat acute lung injury.

20.4.5. S1P Regulates Biological Functions in Acute Lung Injury-Related Lung Cells

The role of S1P in the regulation of lung endothelial barrier integrity has been well investigated. S1P treatment increases transendothelial mono-layer resistance, indicating that S1P is an enhancer of lung endothelial barrier integrity (reviewed in [105]. Rac1 plays a major role in the process through increasing cell spreading. Rac1 is activated by TIAM-1 (T-cell lymphoma invation and metastasis 1), the guanine exchange factor for Rac1. Various signaling pathways including PI3K/AKT, PKCζ, PKCε, and PLD are involved in S1P-activated Rac1 [109].

Several studies demonstrate that S1P could affect lung epithelial cell functions. Exogenous S1P induces intercellular adhesion molecule 1 (ICAM-1) expression in alveolar epithelial cells through activation of NF-κB [110]. S1P₃-mediated S1P signaling activates cytosolic PLA₂α in lung epithelial cells [111], which may explain the proinflammatory role of S1P₃ in lung injury. Plasma S1P₃ level is considered as a biomarker for acute lung injury since it is increased in human and experimental acute lung injury and associated with mortality rate [112]. Nuclear S1P is generated by SphK2. Pseudomonas aeruginosa treatment induced phosphorylation of SphK2 and association with HDAC1/2 in the nuclei, resulting in acetylation of histone 3 and 4 [95].

Neutrophil influx into lungs is a hallmark of pathogenesis of acute lung injury. S1P has been shown to increase IL-8 release in bronchial epithelial cells, and the effect was mediated by S1P ligation to S1P₂ [113], indicating that S1P/S1P₂ pathway contributes to lung inflammation. However, pretreatment with S1P reduced IL-8-or fMLP-induced neutrophil chemotaxis, suggesting an anti-inflammatory effect of S1P. The S1P effects on neutrophils may be through ligation to S1P₄ in the neutrophil [114]. The neutrophil specific S1P receptor knockout mice may identify that S1P receptor is responsible for the effect of S1P on neutrophil migration.

20.5. LPA and S1P in Asthma

20.5.1. Pathophysiology of Asthma

Asthma is a chronic airway inflammatory disease affecting at least 300 million people and causes more than 380,000 deaths per year worldwide. Asthma is characterized by reversible airflow obstruction in association with airway hyper-responsiveness, increased mucus generation, eosinophilia, and increased Th2 cells and Th2 cytokines. An increase in airway smooth muscle (ASM) mass and mucus glands lead to airway wall thickening and airway constriction. Th2-dominant inflammatory responses are a hallmark of allergic asthma. Lung IL-4, IL-5, IL-13, and IL-33 levels are increased in both human and experimental asthma models (reviewed in [115–117]). Other cytokines including IL-9, IL-17A, and tumor necrosis factor (TNF)-a are involved in the pathogenesis of asthma (reviewed in [116, 117]). Treatments for asthma focus on dilation of airways and suppression of lung inflammation. While most inflammatory responses can be diminished by corticosteroids, while severe asthma is resistant to such and has no effective treatment (reviewed in [116, 117]).

20.5.2. Role of LPA in Asthma

Several studies have shown that ATX protein and LPA species in BALF are increased in segmental allergen challenge of allergic subjects and asthma patients [118–120]. Park GY et al. showed that ATX-overexpressing transgenic mice exhibited increased severity of asthmatic responses, while ATX heterozygous knockout mice or administration of ATX inhibitor (GWJ-23) significantly attenuated Th2 cytokines and allergic lung inflammation in a triple-allergen murine model of asthma [119]. This study suggests that the ATX/LPA axis is a potential target for treating asthma.

To investigate the role of LPA in the pathogenesis of allergic asthma, LPA receptor deficient mice were sensitized and challenged with an allergen. LPA₁ heterozygous deficient mice did not show dramatic changes compared to wild-type mice, while LPA₂ heterozygous deficient mice showed reduced eosinophil influx into lungs and mucus glands in bronchi [86], suggesting that the LPA/LPA₂ pathway is implicated in the development of allergic asthma. Further, this conclusion was confirmed by another study showing that LPA₂-deficient mice exhibited reduced BAL total cell numbers, IL-4 and IL-5 levels within BAL, and severity of lung inflammation [119]. Interestingly, both studies revealed that downregulation of LPA₂ significantly diminished LPA levels in BAL [86, 119]. The molecular regulation of LPA₂ on LPA generation remains unclear; it is possible that LPA/LPA₂ reduces LPA synthesis enzymes, such as ATX or secreted PLA₂, by a negative feedback mechanism. On the contrary, in a murine model of ovalbumin (OVA)-sensitized and challenged allergic lung inflammation, LPA₂ promoted lung inflammation, evidenced by increased BAL eosinophil numbers and hyper-reactivity compared to wild-type mice [121]. The reason for the controversial conclusion from the distinct effects of LPA₂ in the different allergic challenges has not been well understood. Administration of LPA₂ specific antagonist is needed to understand the role of LPA₂ in the pathogenesis of asthma.

An interesting study by Jendzjowsky NG et al. showed that increased plasma LPA levels are implicated with carotid body activation-mediated vagal activity, which has been shown to trigger bronchoconstriction [122]. Administration of Brp-LPA, an ATX inhibitor and LPA receptor antagonist, prevents bradykinin-induced asthmatic bronchoconstriction [122], suggesting a role of LPA/LPA receptors in regulation of bronchoconstriction by activation of carotid bodies. Hashimoto T et al. showed that inhalation of oleoyl LPA induced airway hyper-responsiveness to acetylcholine, possibly through increasing release of histamine and activating the Rho/ROCK-mediated pathway [123]. LPA receptor deficient mice and ATX transgenic mice may be useful to investigate the role of LPA in activation of carotid body-regulated bronchoconstriction during an asthmatic attack.

20.5.3. Molecular Mechanisms by which LPA/LPA Receptors Contribute to the Pathogenesis of Allergic Asthma

Increase in ASM mass is implicated in airway bronchoconstriction in asthmatic patients. Proliferation of ASM isolated from asthmatic patients is increased compared to ASM from nonasthmatic patients [124]. LPA has been considered as a plasma growth factor that induces cell proliferation in a variety of cell types, including ASM. Coculture with EGF exhibited a markedly synergistic mitogenesis in human ASM [125]. Activation of Erk, Rho, and AP-1 is required for LPA-induced ASM cell growth [125]. In addition to increasing cell growth, LPA treatment facilitates methacholine-induced ASM contractility, attenuates isoproterenol-induced relaxation of ASM, and increase IL-6 release [126]. Notably, LPA alone did not increase ASM contraction [125].

The effects of LPA on immune cells have been reported. LPA increases Th17 differentiation in obesity [127], but the effect of LPA on IL-17A production in allergic asthma has not been studied. Here we focus on discussing the role of LPA in the release of Th2 cytokines and their signaling in the development of allergic asthma. Cocultured with T-cell activators, LPA induces IL-13 but not IL-4 production in human T cells [128]. In vitro chemotaxis assays showed that LPA induces CD4+ T cell and eosinophil migration [129]. Bronchial epithelial cells regulate Th2 cytokine expression and are also affected by Th2 cytokines. LPA treatment increased decoy receptors of Th2 type cytokine, such as IL-13Ra2 and soluble ST2 (sST2) [130], suggesting that LPA may attenuate IL-13- and IL-33-mediated signaling in bronchial epithelial cells, further suggesting an anti-Th2 response property of LPA in bronchial epithelial cells. As discussed by Kim S et al., these in vitro studies used LPA18:1, which is not the major LPA species in asthmatic patients (reviewed in [131]). The major species, LPA22:5 and 22:6, should be used to test and evaluate their effects on IL-13- and IL-33-mediated signaling in airway epithelial cells. In addition to regulating Th2 cytokine signaling, LPA induces cyclo-oxygenase (COX)-2 expression and prostaglandin E2 (PGE2) release in human bronchial epithelial cells [59]. Inhibition of COX-2 is an effective therapy to treat asthma, indicating that LPA receptors are therapeutic targets for treating asthma. Lundequist A and Boyce JA demonstrated that LPA₅ is highly expressed in human mast cells. LPA induces calcium flux, MIP-1β, and histamine release from mast cells [132]. Activation of mast cells is a hallmark of an allergic response. The LPA/LPA₅ axis is a potential target for diminishing IgE-mediated allergic responses.

20.5.4. Role of S1P in Asthma

S1P levels in BALF are increased in ragweed-allergic asthmatics and a murine model of OVA-challenged allergic asthma [133]. Exogenous S1P administration into isolated lungs increased mast cell number, IL-4, IL-13, and IL-17 production, as well as contraction of isolated bronchi [134], suggesting a proasthmatic role of S1P. Inhibition of SphK1 reduces both intracellular and extracellular S1P levels. Increased SphK1 and SphK2 levels in bronchial tissues were found in OVA-sensitized mice compared to control mice [135]. An inhibitor of SphK1, SK1-I, reduced activation of human and murine bone marrow-derived mast cells, as well as OVA challenge-induced cellular infiltration into lungs, goblet cell hyperplasia, and pulmonary eosinophilia in mice [133]. Consistent with these findings, administration of another potent SphK inhibitor, N,N-dimethylsphingosine (DMS), or SphK1 siRNA exhibited anti-inflammatory effects in OVA-challenged mice [37]. DL-threo-Dihydrosphingosine (DTD), another SphK inhibitor, inhibited acetylcholine-induced contraction of isolated bronchi harvested from OVA-sensitized mice [135].

S1P₂ and S1P₃, but not S1P₁, were increased in lung tissues from OVA-sensitized mice [135]. The effects of S1PR antagonists on attenuation of asthmatic responses have been studied. Administration of JTE013 (S1P₂ antagonist) attenuated Th2 type cytokines and eosinophil numbers in BALF of OVA-challenged asthmatic mice [136]. In support of the conclusion from the JTE013 treatment, S1P₂-deficient mice were found to have reduced IL-4, IL-5, and IL-13 expression in both lung tissues and inflammatory cells in BALF [136]. Polymorphism analysis showed that functional variants of the S1P₁ gene are associated with asthma susceptibility [137]. IT FTY720 reduced features of airway remodeling in an OVA-induced rat model of asthma. IT FTY720 diminished OVA challenge-induced increase in airway smooth muscle mass, airway hyper-responsiveness, BAL eosinophil and lymphocytes, and IL-5 and IL-13 expression in lung tissues [138]. Consistent with this study, Oyeniran C et al. demonstrated that intranasal FTY720 administration reduced OR-like protein isoform 3 (ORMDL3) expression, airway inflammation, and mucus hypersecretion in HDM-challenged mice. ORMDL3 is considered as a gene associated with susceptibility to asthma [139]. In contrast to these two studies, Ble F-X et al. showed that intranasal FTY720 had no effect on OVA challenge-induced immune cell influx into lungs, while it inhibited allergen-edema [140]. In the same study, the authors demonstrated that intranasal administration of S1P1-selective agonist, AUY954, prior to OVA challenge reduced lung edema without altering BAL eosinophil influx into lungs [140]. These controversial conclusions raise concerns in using FTY720 to treat allergic asthma. The role of S1P₁ in the development of allergic asthma can be further evaluated using S1P₁-deficient mice in the future.

20.6. LPA and S1P in COPD

20.6.1. Pathogenesis of COPD

COPD is a chronic lung inflammatory disease characterized by poorly reversible airflow obstruction, emphysema, and bronchiolitis. Cigarette smoking is the leading cause of the progressive airway inflammatory disease (reviewed in [141]). Neutrophils, macrophages, and T lymphocytes release inflammatory mediators including lysophospholipids in COPD. Airflow obstruction is caused by mucus hypersecretion and hypertrophy of smooth muscle and connective tissues. Imbalances of protease and antiprotease disrupt alveolar structure, leading to abnormal enlargement of alveolar spaces. Major medications for COPD open the airway through inhaled bronchodilators and diminish inflammatory responses via use of corticosteroids, phosphodiesterase-4 inhibitors, and so on (reviewed in [141]).

20.6.2. Role of LPA in COPD

Unlike the role of LPA in acute lung injury and asthma, the role of LPA in COPD has not been well demonstrated. An increase in plasma LPA was revealed in a tobacco smoke-induced rat model of chronic bronchitis [142]. Naz S et al. discovered that both serum LPA (16:0) and LPA (18:2) are increased in smokers with COPD [143]. Interestingly, they found that the levels of LPA are correlated with lung function in males with COPD, but not females [143]. The effect of increased LPA in the development of COPD remains unclear. LPA induces cytokine release, such as IL-8 and IL-6 in human bronchial epithelial cells, suggesting a proinflammatory role of LPA in COPD (reviewed in [6, 65]. Blanque R et al. attempted to reveal if reduction of LPA levels by inhibition of ATX alleviates severity of COPD. They showed that post-treatment with GLPG1690, an ATX inhibitor, dose-dependently reduced inflammatory cell influx into lungs [144], suggesting that ATX inhibitors can be developed as a novel therapeutic strategy to treat COPD. However, LPA has been shown to increase lung epithelial cell migration [90]. Analysis of LPA₁-deficient mice during lung development indicates a role of LPA/LPA1 in alveolarization[145]. To investigate whether LPA and LPA receptors contribute to the development of COPD, LPA receptor isotype deficient mice and antagonists should be used in experimental COPD models.

20.6.3. Role of S1P in COPD

The role of sphingolipids, including S1P, sphingosine, and ceramide, in the pathogenesis of COPD has been well studied. In this chapter, we will focus on discussing the discoveries of S1P, SphK, and S1P receptors in COPD. S1P levels are increased in lungs cigarette smoke-induced COPD mice [146]. The balance of S1P/ceramide in COPD has been well discussed [147]. Expressions of SphKs, S1P receptors, and SPL1 in lung tissues and alveolar macrophages were examined. Barnawi J et al. showed that SphK1/2, SIP₂, S1P₅, and SPL1 mRNA levels were increased in COPD patients compared to healthy controls [148]. The changes in mRNA expression within macrophages were confirmed in cigarette smoke extract-treated THP-1 macrophages [148], indicating that S1P signaling in macrophages may be involved with the development of COPD. Consistent with this conclusion, activation of S1P₂ and S1P₃, but not S1P₁, has been shown to stimulate macrophage migration[146]. In the study by Cunto GD et al., S1P challenge significantly increased contraction of isolated bronchi from cigarette smoke-exposed mice, and the effect was abrogated by pretreatment with S1P₂ and S1P₃ antagonists. Inhibition of SphK reversed carbachol-increased contractions in bronchi of mice challenged with cigarette smoke [146]. Taken together, these studies suggest that the inhibition of S1P signaling may alleviate the severity of COPD.

However, another study revealed that plasma S1P levels tended to be negatively correlated with emphysema and COPD exacerbations, supporting that S1P/ceramide ration plays a critical role in the pathogenesis of COPD [147]. Tran HB et al. showed that cigarette smoke extract exposure reduced activity of SphK1 but not SphK2 [149]. S1P is an endothelial barrier integrity enhancer, and FTY720 agonists attenuated nicotine-increased endothelial hyperpermeability[150]. Further, SEW2871, an S1P1 agonist, exhibited an antiapoptotic effect and blocked vascular endothelial growth factor (VEGFR) inhibition-induced emphysema [57], suggesting that S1P signaling suppresses the development of emphysema by maintaining lung epithelial and endothelial integrity. The dissimilar effects of S1P signaling on the development of COPD in the different studies will be investigated by using S1P receptor or SphKs global or cell-specific knockout mice.

20.7. LPA and S1P in other Lung Diseases

Both LPA and S1P exhibit pro-oncogenic properties by increasing cell proliferation, migration, antiapoptosis, and epithelial–mesenchymal transition (EMT) (reviewed in [151–153]). Thus, it is not surprising that LPA and S1P signaling play roles in the development of lung cancer. However, there is no clear preclinical evidence showing that targeting LPA and S1P signaling diminishes lung cancer. Serum S1P and LPA levels and their metabolic enzymes may be used as biomarkers for both prognosis and prediction for lung cancers.

The profibrotic effects of LPA and S1P have been well documented. Tager AM et al. demonstrated that LPA and LPA receptors are increased in lungs in pulmonary fibrosis patients and in a bleomycin-induced murine model of pulmonary fibrosis [80]. LPA₁ knockout mice exhibited significantly reduced severity of pulmonary fibrosis, indicating that LPA/LPA₁ signaling is a potential target for treating pulmonary fibrosis [80]. Many restudies followed the discovery and showed that activation of LPA_1/2 and ATX contributes to the pathogenesis of pulmonary fibrosis (reviewed in [154, 155]). Similar to LPA, a large amount of preclinical experimental pulmonary fibrosis studies showed that reduction of S1P levels and inhibition of S1P signaling alleviate the severity of pulmonary fibrosis (reviewed in [156, 157]). A recent study indicates that intracellular S1P may exhibit an opposite effect on the development of pulmonary fibrosis compared to extracellular S1P [158]. Thus, precise regulation of extracellular S1P levels needs to be further investigated to better understand the pathogenesis of pulmonary fibrosis.

20.8. Perspective

LPA and S1P, the simple bioactive lipid mediators, have been implicated in normal physiological functions and in the pathogenesis of a variety of human disorders. Thus, their metabolic enzymes and receptors have been targeted for drug development to treat lung inflammatory diseases. Since both LPA and S1P play a variety of cellular functions, completely blocking synthesis of LPA and S1P may cause unexpected side effects. To improve our understanding of the roles of LPA and S1P signaling play in the development of lung inflammatory diseases, cell-specific LPA and S1P receptor isotype knockout mice need to be developed to investigate their role in preclinical models of lung inflammatory diseases. Regulation of GPCRs homeostasis by the ubiquitin-lysosome or ubiquitin-proteasome systems has been given more attention. The molecular mechanisms of LPA and S1P receptors internalization and degradation are a new focus, with the intention to discover new targets in regulating LPA and S1P signaling. As discussed above, both LPA and S1P levels are increased in most lung diseases. Reduction of extracellular lysophospholipids is a potential therapy strategy that has not been tested. LPPs degrade both extracellular LPA and S1P, and LPPs are transmembrane proteins. Thus, LPPs are druggable targets for treating human disorders, including lung inflammatory diseases. Cell-specific LPP isoform knockout mice need to be used in the preclinical models of lung inflammatory diseases (Fig. 20.1).

Fig. 20.1.

Mechanisms of LPA and S1P generation and degradation and receptor-mediated cellular responses in lung diseases

Abbreviations

AGK

Acylglycerol kinase

ALI

Acute lung injury

ARDS

Acute respiratory distress syndrome

ASM

Airway smooth muscle

ATX

Autotaxin

BALF

Bronchoalveolar lavage fluid

COPD

Chronic obstructive pulmonary disease

DGK

Diacylglycerol kinase

DMS

Dimethylsphingosine

DTD

DL-threo-Dihydrosphingosine

ECMO

Extracorporeal membrane oxygenation

Edg

Endothelial cell differentiation gene

EMT

Epithelial–mesenchymal transition

ENPP2

Ectonucleotide pyrophosphatase/phosphodiesterases family member 2

G3P

Glycerol 3-phosphate

GPAT

Glycerol 3-phosphate acyltransferase

GPCR

G protein–coupled receptor

HLMVEC

Human lung microvascular cell

HPAEC

Human pulmonary arterial endothelial cell

IP

Intraperitoneal

IT

Intratracheal

IV

Intravenous

LPA

Lysophosphatidic acid

LPAATs

LPA acyltransferases

LPL

Lysophospholipid

LPP

Lipid phosphatase

LPS

Lipopolysaccharide

MAG

Monoacylglycerol

MLC

Myosin light chain

NOX2

NADPH oxidase type 2

ORMDL3

OR-like protein isoform 3

PA

Phosphatidic acid

PC

Phosphatidylcholine

PG

Phosphatidylglycerol

PLA

Phospholipase A

PLC

Phospholipase C

PLD

Phospholipase D

PS

Phosphatidylserine

S1P

Sphingosine-1-phosphate

SphK

Sphingosine kinase

Spns2

Spinster homolog 2

SPPase

S1P phosphatase