Resistin family proteins in pulmonary diseases

Abstract

The family of resistin-like molecules (RELMs) consists of four members in rodents (RELMα/FIZZ1/HIMF, RELMβ/FIZZ2, Resistin/FIZZ3, and RELMγ/FIZZ4) and two members in humans (Resistin and RELMβ), all of which exhibit inflammation-regulating, chemokine, and growth factor properties. The importance of these cytokines in many aspects of physiology and pathophysiology, especially in cardiothoracic diseases, is rapidly evolving in the literature. In this review article, we attempt to summarize the contribution of RELM signaling to the initiation and progression of lung diseases, such as pulmonary hypertension, asthma/allergic airway inflammation, chronic obstructive pulmonary disease, fibrosis, cancers, infection, and other acute lung injuries. The potential of RELMs to be used as biomarkers or risk predictors of these diseases also will be discussed. Better understanding of RELM signaling in the pathogenesis of pulmonary diseases may offer novel targets or approaches for the development of therapeutics to treat or prevent a variety of inflammation, tissue remodeling, and fibrosis-related disorders in respiratory, cardiovascular, and other systems.

INTRODUCTION TO RELM SIGNALING

Pulmonary diseases, including pulmonary hypertension (PH), asthma/allergic airway inflammation, chronic obstructive pulmonary disease, fibrosis, cancers, infection, and other lung injuries, are multifactorial diseases that cause morbidity and mortality worldwide. The pathogenesis of these lung disorders is tremendously complex, with many questions left unanswered. One potentially unifying factor is the resistin-like molecule (RELM) family of proteins. This family comprises pleiotropic cytokines and growth factors that have been linked to obesity, type 2 diabetes, and alternate (such as Th2) inflammation (30, 98). The role of these cytokines in many aspects of physiology and pathophysiology, especially in cardiothoracic diseases, is rapidly evolving in the literature. In this review, we summarize the involvement of RELM signaling in pulmonary diseases. We focus on the immunoregulatory activities of RELMs, the immunotherapies that target RELM signaling in a variety of lung inflammation-related pathologies, and the potential to use RELMs as biomarkers of these diseases.

Nomenclature of the Resistin-Family Proteins

Resistin was first discovered in rodents as an adipokine with insulin resistance properties (116). Proteins that comprise the RELM family are characterized by a unique cysteine-rich motif at the COOH terminus (9, 18, 58, 117). Four rodent and two human isoforms of the resistin family have been discovered to date (36, 44, 117, 122). Members of this protein family have been identified as “found in inflammatory zone” (FIZZ) in an airway reactivity model (44), as “hypoxia-induced mitogenic factor” (HIMF) in a model of PH (122), as adipokines (resistin) (116, 117), as insulin resistance proteins (ADSF) (58), and as chemokines (XCP) (18). The diverse nomenclature of RELMs is shown in Table 1. In this review, we will predominantly refer to this family of proteins as RELMα, RELMβ, resistin, and RELMγ. Given the species variants (30), human (h) and rodent/murine (m) RELMs also will be specifically indicated.

Table 1.

Nomenclature for the RELM/FIZZ/XCP family of proteins

RELM Family Members	Cellular Source (RELM-Producing Cells)
Murine/rodent (m) classification
FIZZ1 = RELMα = HIMF = XCP2	Macrophages (107), dendritic cells (20), and other immune cells (bone marrow cells and splenocytes) (30); bronchial epithelial cells, and type II pneumocytes
FIZZ2 = RELMβ = XCP3	Goblet cells (7), lung epithelial cells (78)
FIZZ3 = Resistin = ADSF = XCP4	Adipocytes (59)
FIZZ4 = RELMγ = XCP1	Immune cells (bone marrow and white blood cells, splenocytes and thymocytes) (36, 115), adipocytes, and pneumonocytes (36)
Human (h) classification
Resistin = RELMα = FIZZ1 = XCP1	Macrophages, monocytes, and neutrophils (50)
RELMβ = FIZZ2 = XCP2	SMCs, macrophages, and T cells (6), ECs and fibroblasts (33)

ADSF, adipose tissue-specific secretory factor; ECs, endothelial cells; FIZZ, found in inflammatory zone; HIMF, hypoxia-induced mitogenic factor; RELM, resistin-like molecule; SMCs, smooth muscle cells; XCP, 10-cysteine protein.

RELM Structure and Biochemistry

Proteins in the RELM family have a secretory signal peptide at the NH₂ terminus and a unique structure with no known homology to any other protein (44). Both human and rodent variants contain between 105 and 117 amino acids, with notable interspecies (33–59%) and intraspecies (21–51%) homology (36, 44, 117, 132), especially at the cysteine-rich COOH terminus. The common feature of these RELMs is the presence of five disulfide bonds formed by a group of 10 equally spaced cysteine residues in the COOH terminus (44). In mice, the genes for RELMα, β, and γ are aligned sequentially on chromosome 16B5 as Retnlb, Retnla, and Retnlg; the resistin gene is located separately on chromosome 8A1. In humans, resistin is located on chromosome 19p13.2, and RELMβ is located on chromosome 3q13.1 (30).

The Immunoregulatory Roles of RELM Signaling

RELM signaling is an important component of the Th2-activated macrophage inflammatory response to tissue injury in the lung and other organs (54). Rodent (m) RELMα has been shown to play a critical role in the development of pulmonary arterial hypertension in hypoxia models and Th2 inflammatory models induced by ovalbumin (OVA) sensitization, schistosomiasis, and HIV-related stimuli (4, 6, 22, 31, 54, 120, 122). Our laboratory has shown that mRELMα promotes lung inflammation by activating the vascular endothelial growth factor (VEGF) and canonical Th2 cytokine interleukin (IL)-4 pathways (129, 130). In the OVA-induced Th2 immune response, mRELMα upregulates inflammatory cytokines such as IL-1β, IL-1ra, IL-16, and IL-17 and chemokines CXCL-1, CXCL-2, CXCL-9, CXCL-10, and CXCL-13, MCP-1, M-CSF, TIMP-1, and TREM-1 (31). The mRELMα ortholog could not be identified in humans (132), and human (h) resistin shows a greater similarity in expression pattern and inflammation-modulatory functions to mRELMα than to mResistin (92, 95, 132). Using mRELMα knockout mice and humanized mice expressing hResistin, we and others have identified additional similarities in the immunoregulatory properties of hResistin and mRELMα (52, 76). Recently, researchers have proposed the idea that alarmins might initiate Th2 inflammation (11). Indeed, high mobility group box (HMGB) 1, a key damage-associated molecular pattern (DAMP) protein, serves as an alarmin to drive inflammation (41). HMGB1 polarizes and activates M2 macrophages through the receptor for advanced glycation end products (RAGE) around hypoxic tumors (47) and mediates Th2 airway inflammation in acute allergic asthma (82). We have identified mRELMα and its human homolog hResistin as the upstream activators of DAMPs HMGB1 and S100A11 in the development of pulmonary vascular remodeling (29, 76, 77). Hence, RELMs might initiate Th2 inflammation in the lungs though the DAMP-RAGE pathway. Th2 stimuli also can induce mRELMα expression through IL-4, IL-13, and STAT-6 pathways (93, 102, 118). Thus, RELM signaling may act as a crucial hub of a positive feedback loop to trigger, amplify, and sustain inflammation through its immunoregulatory activities in the pathogenesis of lung diseases (Fig. 1).

Fig. 1.

Schematic illustration of inflammatory signals regulated by resistin-like molecules (RELMs). Through immunoregulatory properties, RELMα or human resistin may create a positive feedback loop that mediates the amplification of lung inflammation via Th2 (IL-4) and damage-associated molecular pattern (DAMP) (high mobility group box 1, HMGB1) pathways.

Putative Receptors and Binding Partners of RELMs

Receptors for RELMs have not yet been identified. Reports in the literature have suggested an isoform of decorin (23), mouse receptor tyrosine kinase-like orphan receptor (ROR) 1 (113), and Toll-like receptor (TLR) 4 (121) as putative receptors for Resistin. However, neither decorin nor ROR1 have been shown to mediate the inflammatory effects of hResistin (71). Additionally, biochemical binding assays have not been conducted to show interaction between TLR4 and hResistin (71), and TLR4 knockdown did not significantly affect hResistin-mediated monocyte migration and inflammatory cytokine production (71). The adenylyl cyclase-associated protein 1 (CAP1) has been implicated as a functional receptor that mediates the inflammatory actions of hResistin in human monocytes (71). However, as CAP1 lacks a transmembrane domain, the biological mechanism underlying its membrane location is still unclear (71). m/hRELMβs were also identified as bactericidal proteins that bind to negatively charged bacterial lipids and form a membrane-permeabilizing pore to lyse the targeted cells (105). We previously identified Bruton’s tyrosine kinase (BTK) as the first known functional binding partner of mRELMα to mediate chemokine actions in myeloid cells (119). More recently, the extracellular calcium-sensing receptor (CaSR) was suggested as a binding partner for mRELMα in pulmonary artery smooth muscle cells (SMCs) (134). A limitation of that study was that the PH model was induced by intermittent, not continuous, hypoxia (134). Therefore, the results need validation.

Tissue Distribution of RELMs

In healthy humans, hResistin is expressed predominantly in bone marrow, monocytes, and leukocytes, whereas hRELMβ is found primarily in colon and small intestine and to a lesser extent in testis and spleen (18, 96). In normal mice, mRELMα is expressed predominantly in lung, bone marrow, and spleen (44, 80, 117); mRELMβ is expressed in intestinal epithelial cells (7, 42, 117); mResistin is expressed in white adipose tissue (44, 116); and mRELMγ is expressed in hematopoietic tissues and lung (18, 36). Under pathophysiologic conditions, hResistin and hRELMβ have been observed in the lungs of humans with asthma, scleroderma (SSc), idiopathic PH, and other Th2 inflammatory diseases (6, 45, 77, 89, 102). The main cellular source of hResistin is the myeloid cells, especially macrophages, and its expression pattern shows a greater similarity to that of mRELMα than to that of mResistin (92, 95, 132). We also have reported that mRELMα, mRELMβ, and mRELMγ are expressed in rodent models of allergic (asthmatic) inflammation and chronic hypoxia-induced PH (5, 30, 31, 122).

On the basis of the above-described RELM tissue distribution, and because lung is the primary location of most RELM isoforms (30, 36, 44, 122), research into the association between RELMs and the pathogenesis of cardiothoracic and respiratory diseases is now beginning to expand rapidly. We believe that summarizing the involvement and clinical implications of RELM signaling in pulmonary disease will be helpful to the development of novel therapeutic approaches for a variety of related thoracic diseases.

ASSOCIATION OF RELM SIGNALING WITH PULMONARY DISEASE

Pulmonary Hypertension

The pathogenesis of PH features pulmonary vascular remodeling, increases in pulmonary artery pressure, and right heart failure (133). In humans, severe PH is characterized by plexiform lesions that contain phenotypically altered pulmonary SMCs and endothelial cells (ECs) (125). Using gene array technology, we have shown that mRELMα is dramatically upregulated in hypoxic lungs and produces potent mitogenic effects (122). The expression of mRELMα is regulated by hypoxia, likely via the C/EBP pathway (18), as the mRELMα gene was shown to possess six C/EBP binding motifs (122). In rodent models, transtracheal delivery of mRELMα gene by adeno-associated virus causes vascular remodeling and hemodynamic changes like those of PH (4). Conversely, in vivo knockdown of mRELMα markedly reduces PH development caused by chronic hypoxia or Th2 inflammatory stimuli (4, 22, 89), indicating an etiologic role for mRELMα in PH. In vitro and in vivo gain- and loss-of-function studies carried out in PH-related human primary cells and rodent models with mRELMα knockout, knockdown, or overexpression have suggested mRELMα to be a proinflammatory, angiogenic, vasoconstrictive, fibrotic, and chemokine-like cytokine in hypoxic lungs (Fig. 2) (3, 4, 22, 29, 32, 55, 62, 76, 77, 119, 122, 129–131). These cytokine-like pro-PH properties of mRELMα are shared to a great extent with hResistin in humans (76, 77, 92, 95, 131, 132). It has been found that mRELMα promotes the proliferation of pulmonary vascular SMCs, microvascular (PMV) ECs, and fibroblasts in vitro via the Akt or ERK1/2 signal transduction pathway (19, 122, 123). mRELMα induces lung vascular inflammation via IL-4 and VEGF/VEGFR2 (129, 130). Systemic administration of mRELMα causes EC apoptosis and subsequent IL-4-dependent PH in mice (131). Interestingly in embryonic mouse lung, RELMα has antiapoptotic properties that may regulate lung alveolarization and maturation (127). HIF-1α has been identified as a downstream actor of RELMα based on its ability to mediate mRELMα-induced myeloid cell migration and vascular tube formation (55). In addition to its molecular mechanisms, we have uncovered the mRELMα-regulated stem cell biology. mRELMα activation may promote stem cell recruitment from bone marrow to lungs in hypoxic mice in vivo (3). In vitro, mRELMα maintains the multipotency of cultured mesenchymal stem cells by preserving their differentiation potential and phenotypic changes (62).

Fig. 2.

Schematic illustration of resistin-like molecule-α (RELMα)/resistin-governed cellular pathology and interaction in pulmonary arterial hypertension (PAH). Initiated by hypoxia, RELMα/resistin signaling regulates proinflammatory, proangiogenic, and promigratory pathways in PAH-related cells, including endothelial cells, smooth muscle cells, macrophages, and stem cells. By mediating vascular inflammation, RELMα/resistin drives pulmonary vascular remodeling and PAH development. BTK, Bruton’s tyrosine kinase; h, human; HIF, hypoxia-inducible factor; HMGB1, high mobility group box 1; m, mouse; MCP-1, monocyte chemoattractant protein-1; MMP-2, matrix metalloproteinase-2; PDGF, platelet-derived growth factor; RAGE, receptor for advanced glycation end products; RANTES, regulated on activation, normal T cell expressed and secreted, also known as chemokine (C-C motif) ligand 5 (CCL5); RHD, right heart dysfunction; SDF-1, stromal cell-derived factor-1.

The roles of mRELMα also have been interrogated in non-hypoxia-induced PH models. mRELMα was shown to be elevated in bronchoalveolar lavage fluid (BALF) of a mouse AIDS (acquired immune deficiency syndrome)-related PH model created by pneumocystis infection plus CD4 T-cell depletion (120). mRELMα expression also was increased in a schistosomiasis-induced PH mouse model, and the protein interacted with IL-13 signaling to exacerbate vascular remodeling in the infected lungs (40). In mice expressing a hypomorphic BMPR2 mutation, OVA-triggered PH-upregulated expression of mRELMα, but not mRELMγ, in the right ventricle (100). mRELMα was also upregulated in a mouse PH model induced by coexposure to OVA and urban ambient particulate matter (mimicking air pollution) (99), but mice lacking mRELMα also fully exhibited a severe remodeling response. These findings suggest that the role of mRELMα in arterial thickening and remodeling could vary between different types of PH inducers such as hypoxia and antigens (5).

In patients with SSc-associated PH, hRELMβ expression is upregulated in vascular cells and leukocytes, as well as in myofibroblast-like cells (6). Elevated serum hResistin levels in SSc patients also correlate with high right ventricular systolic pressure (RVSP), indicating that hResistin contributes to the pathogenesis of SSc-associated PH and might be useful as a biomarker (85). An in vitro study revealed that mRELMα stimulates intracellular Ca²⁺ release in human pulmonary artery (PA) SMCs through the phospholipase C signaling pathway involving Gα_q/11 protein-coupled receptors and ryanodine receptors (32). mRELMα also induces PASMC migration by regulating the EF-hand calcium-binding DAMP protein S100A11 (29). hResistin promotes the EC-mediated alteration of PASMCs to a proliferative phenotype (131). hRELMβ induces proliferation and activation of ERK1/2 in human PASMCs and PMVECs in vitro (6). Besides being upregulated in pulmonary vascular cells, hRELMβ was shown to be upregulated in hypoxia-exposed human lung A549 cells and primary adventitial fibroblasts, and its overexpression enhanced mitogenesis via PI3K activation (110). hRELMβ also increases HIF-1α and IL-6 expression through an NF-κB-mediated mechanism in human primary lung fibroblasts (55). Thus, PH-related vascular, inflammatory, and resident cells, as well as stem/progenitor cells, all might be regulated by RELM signaling under inflammatory conditions.

Recently, our mechanistic studies showed that signaling of mRELMα, as well as its human homolog hResistin (92, 132), triggered immune-vascular cell crosstalk by activating DAMP signaling during hypoxic inflammation (76, 77). Our human studies support the biomarker potential of hResistin based on the unpublished data that serum hResistin levels in patients with idiopathic pulmonary arterial hypertension and SSc-PH correlate directly with disease severity and survival. In addition, ongoing studies indicate that RELMs drive mitochondrial and metabolic dysfunction of PASMCs and cardiomyocytes, and modulate M1/M2 macrophage polarization in PH development. These findings provide deeper mechanistic insight into RELM-induced PH.

Asthma and Allergic Lung Diseases

Th2 allergic airway inflammation, especially asthma, shares some key pathological features with PH, including increased RVSP and induction of vascular inflammation and remodeling (31, 112). However, animal models of severe Th2-associated allergic inflammation still fail to develop PH (5). One mechanism that might underlie this distinction is a difference in mRELMα expression patterns: mRELMα, which is expressed in the lung vasculature of animals with PH, is missing in allergic models (5). In rodent models, mRELMα is still one of the most upregulated gene products in allergen-associated Th2 responses (91). Similar to the way that we discovered HIMF in hypoxia-exposed animals and PH-related cells (122), FIZZ1 was first described in the OVA-induced asthmatic mouse model (44). Our genomic analysis confirmed mRELMα to be the most consistently upregulated gene in OVA-challenged mouse lungs (31). FIZZ2 (mRELMβ) and FIZZ4 (mRELMγ), but not FIZZ3 (mResistin), also were upregulated by OVA challenge in the mouse acute pulmonary inflammation model (31, 118). Activation of IL-4/IL-13, STAT-6, and C/EBP mediates mRELMα induction by Th2 allergic inflammation (91, 93, 118). In contrast, OVA-induced mRELMβ production is entirely dependent on the IL-13 receptor-α1 chain (91). In OVA-treated rats, upregulated mRELMα expression and mRELMα-enhanced myofibroblast differentiation are characteristic of the early stages of asthmatic airway remodeling (26). Other studies of OVA-challenged rodents have identified mRELMα as a potential oxidative stress marker (136). In a prolonged allergic inflammation model induced by repeated OVA inhalation, lung mRELMα expression was implicated in the pulmonary arterial remodeling driven by the IL-25-natural killer T-cell pathway (57). Genetic ablation of mRELMα prevented OVA-induced inflammation, vascular remodeling, and cardiac hypertrophy, further indicating the proallergic properties of mRELMα (31).

RELM activation also displays pathogenicity in fungus-associated asthma. Mice challenged with the aeroallergen Aspergillus fumigatus (Asp) exhibit significantly augmented mRELMα expression in areas surrounding remodeled vessels (5). Similar to OVA-induced mRELMβ, Asp-triggered mRELMα production and secretion by epithelial cells is regulated by IL-13Rα1 (91). Mice sensitized to a combination of dust mite, ragweed, and Asp extracts exhibited mRELMα induction in lungs that was dependent on the transcription factor PU.1, a regulator of macrophage differentiation and maturation (106). In mice exposed to alternaria, but not Asp or candida, gene microarray analysis of airway epithelial cell brushings demonstrated that fungal allergen sensitization caused a STAT6-dependent increase in mRELMα expression of more than 20-fold (25). mRELMβ was shown to promote lung remodeling in mice via proinflammatory and fibroblast mitogenic activity after intratracheal delivery and after Asp challenge (89). In response to Asp, lung epithelial cells, macrophages, fibroblasts, and vascular endothelial cells produce mRELMβ, which precipitates airway remodeling by extracellular matrix protein deposition and mucosal hypertrophy, suggesting a target potential of hRELMβ in human asthma (33).

Clinically, plasma hResistin levels are elevated in patients with asthma and correlate with disease severity (68). Indeed, hResistin is elevated in patients with acute asthma exacerbation as well as in those with stable asthma (1), suggesting that it could be used as a disease activity marker and predictor of asthma risk. Another study found that the hResistin:adiponectin ratio is a negative predictor of lung function in patients with asthma (8). The ratio is elevated in patients with asthma, much higher in those with more severe disease, and highest in asthmatic patients who are obese and male (8), supporting the premise that hResistin may contribute to increased severity in the obese population. These findings bolster the idea that this ratio might be a potential therapeutic target for management of the obese asthma phenotype.

Nevertheless, data regarding the association between RELM signaling and asthma are conflicting. The mRELMα-overexpressing mice generated by Lee et al. (70) showed attenuated Th2 airway inflammation after OVA-sensitization. However, in another mouse study, mRELMα gene mutation did not significantly suppress Th2 immune responses induced by OVA or Asp (91), mirroring a redundant functional role for mRELMα in these acute allergic models. Clinically, high plasma hResistin levels in steroid-naïve women with asthma predicted a favorable anti-inflammatory effect of inhaled glucocorticoids (73). Another study of childhood asthma showed that serum hResistin levels negatively correlated with atopy parameters, including eosinophil count and total serum IgE (60), suggesting a protective role of hResistin with disease-modifying effects against asthma in children. In an atopic adult study, sensitization to allergen Dermatophagoides pteronyssinus transiently increased serum hResistin levels (63). Unexpectedly, patients with normal airway responsiveness had higher bronchoalveolar lavage and serum hResistin concentrations than did hyperresponsive individuals (63). These results are directly opposed to reports in the literature of elevated serum hResistin in those with moderate or severe asthma (68). Hence, additional studies are required to dissect the immune-modifying mechanism that underlies RELM-related allergic inflammation and resolve these conflicting findings.

Pulmonary Fibrosis

Lung fibrosis is often complicated by PH and promoted by Th2 and Th17 inflammation. The pro-PH (Fig. 2) (3, 4, 22, 29, 32, 55, 62, 76, 77, 119, 122, 129–131) and Th2/Th17 regulatory (Fig. 1) (31) activities of mRELMα, or its human homolog hResistin (92, 132), suggest that it might contribute to fibrogenesis. Indeed, profibrotic effects of mRELMα were first identified in the rat model of bleomycin-induced pulmonary fibrosis (79). cDNA microarray screening showed that of the 628 genes upregulated or downregulated in that model, mRELMα exhibited the most dramatic increase (79). Additional in vivo and in vitro analyses revealed that mRELMα derived from alveolar epithelial cells stimulates α-SMA and collagen I production in fibroblasts (79). IL-4 and IL-13 strongly induce epithelial mRELMα production via STAT6 in bleomycin-treated lungs (80). In turn, mRELMα reduces myofibroblast apoptosis by suppressing the ERK pathway (19). mRELMα also activates collagen-1-producing fibroblasts to promote fungal-induced airway fibrosis (25) and links M2/wound-healing macrophage development with lung fibrogenesis induced by herpesvirus (35). In a mouse Pneumocystis-infected model, BALF-derived macrophages showed consistent mRELMα upregulation concomitant with fibrosis development (120). CC10-driven mRELMα overexpression in transgenic mice was sufficient to recruit bone marrow-derived dendritic cells to lungs, but was not sufficient to cause or alter lung fibrosis induced by bleomycin or silica (84). In support of these findings, experiments using mRELMα knockout mice and adenoviral mRELMα-overexpressing rats showed that mRELMα has profibrogenic properties essential for bleomycin-induced pulmonary fibrosis in vivo (81).

Other RELMs also have been implicated in the pathogenesis of fibrotic lung diseases. mRELMβ is highly induced in bleomycin-treated fibrotic rodent lungs (78), and research with gene-deficient mice showed that mRELMβ is essential to pulmonary fibrosis development (33, 78). hRELMβ expression is upregulated in primary cultured human pulmonary artery adventitial fibroblasts (110) and in patients with idiopathic pulmonary fibrosis (78). The pulmonary fibrosis that develops after both asbestos and silica exposure is also associated with high plasma hResistin levels (72, 114). The data suggest that hResistin mediates immune responses in fiber- or particle-induced parenchymal fibrotic disorders (72, 114) and might be useful as a biomarker of silicosis (114). Recently, hResistin levels were found to be elevated in the plasma and sputum of patients with cystic fibrosis lung disease and to correlate with lung function impairment (34). As discussed above, it also has been suggested that hRELMβ may contribute to the inflammation-induced fibrosis and vascular remodeling in SSc-associated PH (6). In comparison to hResistin, hRELMβ may play a more direct role in matrix protein deposition during tissue and vascular remodeling by activating local fibroblasts (55). Additional mechanistic studies are needed to uncover the hRELMβ-modulated endothelium-mesenchymal transition in human primary pulmonary artery ECs (53).

Lung Cancer

PH bears some resemblance to cancer, as pulmonary vascular cells acquire cancer-like traits, including excessive proliferation and resistance to apoptosis (13). The pro-PH role of mRELMα thus suggests that it has protumor effects. Interestingly, mRELMα gene expression was highly induced in response to tobacco carcinogen nitrosamine in lungs of tumor-susceptible A/J mice but not in those of resistant C3H mice (39). The upregulated mRELMα exhibited immunomodulating action to create a protumorigenic environment in lungs (39).

hResistin has a closer relationship with pulmonary carcinoma in humans than mRELMα has in mice. Serum hResistin levels are significantly elevated in patients with non-small-cell lung carcinoma (NSCLC) (14) and are associated with weight loss (56). In the study by Karapanagiotou et al. (56), although baseline serum hResistin level did not correlate significantly with overall survival of patients with NSCLC, the data showed a trend toward an association between hResistin levels and time to disease relapse in the responders at the end of chemotherapy. A recent study in another cohort showed that serum hResistin has inflammation-modulating effects on cancer cachexia in patients with advanced-stage NSCLC (24), indicating that it has potential as an adjuvant diagnostic or prognostic biomarker for lung cancer (14, 24). Moreover, hResistin gene (RETN) polymorphisms have been correlated with lung cancer progression (46), and hResistin activation was shown to promote human lung adenocarcinoma metastasis (38). Kuo et al. (67) reported that hResistin was elevated in tumor-infiltrating dendritic cells and in human lung cancer samples. Interestingly, they also showed that local expression of mResistin was increased in mice transplanted with lung cancer cells (67). Mechanistically, they found that hResistin-expressing dendritic cells promote lung cancer progression by increasing the Wolf–Hirschhorn syndrome candidate 1/Twist pathway (67). Thus, targeting hResistin may constitute a novel therapeutic approach to manipulate immune cells in the cancer microenvironment. Moreover, Tsai et al. (124) showed that hResistin-enhanced lung cancer metastasis is dependent on miR-519d downregulation in human chondrosarcoma cells. Hence, hResistin-governed micro-RNAs may also represent a novel focus for the development of strategies to treat lung cancer.

Chronic Obstructive Pulmonary Disease

Chronic obstructive pulmonary disease (COPD), including chronic bronchitis and emphysema, is the most common chronic lung condition in the world (66) and the most common cause of secondary PH (37). Chronic cigarette smoking-induced airway hyperresponsiveness is the main cause of COPD. The underlying mechanism has been suggested to be the upregulation of mRELMα, as demonstrated in experimental rat lungs (83). Indeed, serum hResistin levels are elevated in COPD patients and are associated with smoking, inflammatory marker complement C3 and C4 (1), and hyperinflation (111). These findings are supported by the observation that proper exercise training prevents cigarette smoking-induced augmentation of mRELMα production and thereby reduces the risk of COPD development (83). Consistently, patients with COPD who benefit from high-intensity exercise training exhibit lower serum levels of hResistin (12). Thus, hResistin has the potential as a biomarker for inflammation, risk assessment, severity, and prognosis of COPD disease progression (66). In a study by Mineo et al. (88), lung volume reduction surgery in patients with moderate-to-severe emphysema restored respiratory dynamics and ameliorated cachexia. The postoperative reduction in peripheral blood hResistin levels in these patients was associated with favorable clinical change, further highlighting the biomarker potential of hResistin and suggesting its contribution to COPD pathogenesis (88). Additionally, because hResistin is proinflammatory, its levels also correlate with tumor necrosis factor receptors in COPD patients who develop osteoporosis (126).

Interestingly, in COPD patients with chronic Chlamydia pneumoniae infection who had frequent exacerbations, upregulated circulating hResistin was shown to positively correlate with the level of anti-C. pneumoniae IgA (97). In the group with serological features of chronic C. pneumoniae infection, serum IL-6 correlated with hResistin (97). Thus, hResistin is also a critical mediator of the immune response to lung infection, as discussed in the following section.

Lung Infection

We discussed above how RELMs mediate allergic lung inflammatory responses to viral or fungal infection and in infection-induced PH. mRELMα is one of the most upregulated gene products in parasite-associated Th2 responses (91). Pulmonary expression of mRELMα induced by Nippostrongylus brasiliensi (Nb) correlates with the extent of long-term lung damage (43, 109). Interestingly, malaria (Plasmodium chabaudi) coinfection suppresses this nematode-induced increase in mRELMα to alter pulmonary repair processes (43). The accumulation of mRELMα-positive cells in mouse lungs is also induced by Schistosoma mansoni (Sm) eggs and is associated with augmented granulomatous inflammation (101). Moreover, in combination with granuloma obstruction of local capillary beds, mRELMα is related to the development of plexiform-like lesions in a murine schistosomiasis model of lung vascular remodeling (21, 54).

Other studies have shown that RELM activation produces anti-inflammatory effects. In Sm egg-challenged mice, mRELMα deficiency led to exacerbated lung inflammation and fibrosis (94). In a study by Pesce et al. (102) that used three mouse parasite models, including acute and chronic infection with Sm, Sm eggs, and the nematode Nb, the authors reported that mRELMα mediated IL-4/IL-13-dependent Th2-suppressing actions in the inflamed lungs. The mRELMα-suppressed Th2 adaptive immunity in the parasite-infected lungs was believed to protect the host from excessive and potentially fatal lung inflammation (10, 17, 51, 64). Interestingly, M2 macrophages are the main cellular source of this mRELMα (64), suggesting a feedback mechanism to control the Th2 inflammation in anti-parasite responses. Moreover, the suppression of Th2 immunity by mRELMα was offset by an mRELMβ-enhanced immune response in a study that used mRELMβ single- and mRELMβ/α double-deficient mice infected with Nb (17). However, a side effect of increased mRELMα would be its secondary stimulation of vascular remodeling in these infected lungs, as mentioned above (54). Thus, the perceived benefits of enhancing mRELMα signaling must be balanced against the risks associated with the provasculogenic properties of this PH inducer.

As the human homolog of mRELMα, hResistin is also essential for the immune response to helminths. In humanized mice with hResistin gene knock-in, hResistin overexpression exacerbated Nb-induced lung inflammation (50). In individuals infected with soil-transmitted helminths or filarial nematode Wuchereria bancrofti, increased serum hResistin was associated with higher parasite load and augmented proinflammatory responses, identifying hResistin as detrimental in lungs with multiple helminth infections (50). Plasma hResistin is elevated in patients with pulmonary tuberculosis (TB) (15, 90) and can be used for early assessment of drug and vaccine efficacy (15, 28, 90). It correlates directly with TB disease stage and has been suggested to be a key player in TB-associated energy dysregulation (15). Moreover, in patients with type 2 diabetes who also have TB infection, the increased hResistin may suppress the mycobacterium-induced inflammasome activation by leukocytes (16). Thus, hResistin has not only biomarker potential, but also immune-regulatory activities in pulmonary TB.

Other Lung Diseases

Because of their pluripotent effects, RELMs have been implicated in other lung injury conditions. Researchers identified mRELMα as the most promising target in radiation-induced acute lung injury (ALI) by comparing differences in radiation-induced gene expression in radio-sensitive and radio-tolerant mouse strains (48). mRELMα also is a causal factor for acute pancreatitis-associated lung injury (APALI) (128). In a rat model of APALI, mRELMα lung expression was significantly upregulated and associated with pulmonary injury severity. Overexpression and knockdown experiments in rats revealed that mRELMα mediates inflammation in APALI pathogenesis by activating the PI-3K/Akt-NF-κB pathway (128).

Resistin has been implicated in severe pulmonary system dysfunction. hResistin levels are elevated in patients with end-stage lung disease (ESLD) and correlate with the pathogenesis of osteopenic syndrome in ESLD (61). In ALI and acute respiratory distress syndrome (ARDS), the presence of severe PH is associated with increased morbidity and mortality (104). In a model of lipopolysaccharide-induced ALI in humanized mice that exclusively express hResistin (hRTN^+/−/−) but not mResistin, hResistin overexpression promoted neutrophil-mediated inflammation involving DAMP molecules HMGB1 and TLR4 (52). In a rat model of ALI induced by blunt chest trauma, mResistin expression increased significantly in the injured lungs beginning in the initial posttraumatic inflammatory phase and remained upregulated throughout the experiment (27). These findings indicate that hResistin may be a potential target candidate for the development of therapeutic strategies in patients with life-threatening organ system failure such as ALI, ARDS, or sepsis. Lung mResistin levels were also increased in BALF of mice with acute exposure to ozone (O₃) (108). Data from gene knockout mice indicated that mResistin suppressed chemokine keratinocyte chemoattractant expression in the lung but did not promote O₃-induced lung pathology (108).

Baseline resistin expression in normal lungs has also been examined. The ozone challenge study described above was the first to demonstrate the presence of mResistin in mouse lung epithelial lining fluid in the absence of any inciting stimulus (108). Moreover, one human study has detected hResistin in the alveolar lining fluid of healthy adults (87). The authors of that study reported that the concentration of hResistin is significantly higher in alveolar lining fluid than in serum and that protein net charge may influence trafficking and compartmentalization to the alveolar airspace more than molecular weight or hydrophobicity (87). Data on baseline lung hResistin expression and localization in healthy subjects will be critical for understanding its homeostasis and for developing aerosolized drug delivery to or through the lungs.

CONCLUSIONS AND PERSPECTIVES

In this review, we have summarized the involvement of RELM signaling in pulmonary diseases (Table 2), with a focus on the immunoregulatory activities of m/hRELMs and the immunotherapies targeting RELM signaling in a variety of lung inflammation-related pathologies. Asthma, fibrosis, COPD, infection, and other lung injuries all are strongly related to each other and could result in or from immune dysfunction caused by RELM signaling (Fig. 3). Activation of RELMs may be a pivotal event that links these interconnected pulmonary disorders, contributing to or being implicated in vascular remodeling and PH development (Fig. 3).

Table 2.

Involvement of RELM family members in lung diseases

Lung Pathologies	Rodent/Murine (m) RELMs		Human (h) RELMs
	Increased expression/elevated level	Contribution to pathogenesis*	Elevated level/increased expression	Therapeutic target potential	Biomarker potential
Pulmonary hypertension	RELMα (5, 40, 120, 122)	RELMα (4, 55, 76, 77, 122, 129-131)	RELMβ (6); Resistin (85)	Resistin (76, 77, 131); RELMβ (6, 55)	Resistin (85)
Asthma/allergy	RELMα (5, 25, 31, 44, 91); RELMβ and RELMγ (31, 33, 118)	RELMα (25, 26, 31), (anti-inflammatory effect) (70)	RELMβ (33); Resistin (1, 68, 73)	Resistin (8), (anti-inflammatory effect) (73)	Resistin (1, 60, 68)
Fibrosis	RELMα (79, 80); RELMβ (78)	RELMα (79, 81); RELMβ (33, 78)	RELMβ (6, 78, 110); Resistin (34, 72, 114)	RELMβ (6)	Resistin (34, 114)
Cancer	RELMα (39, 67)	RELMα (39)	Resistin (14, 67)	Resistin (38, 124)	Resistin (14, 24, 46)
COPD	RELMα (83)	RELMα (83)	Resistin (1, 111)		Resistin (66, 88)
Infection	RELMα (parasite infection) (43, 91, 109)	RELMα (anti- inflammatory effect in parasite infection) (10, 17, 64, 94, 102)	Resistin (parasite infection) (50), (in TB) (15, 90)	Resistin (proinflammatory effect in parasite infection) (50), (immunosuppressive effect in TB with DM) (16)	Resistin (TB) (28)
Other lung diseases/injuries	RELMα (irradiated lung) (128); resistin (ALI) (27), (O3-induced injury)	RELMα (APALI) (128)	Resistin (ESLD) (61), (ALI/ARDS)	Resistin (ALI) (52)
Healthy lung	RELMα (detactable) (117), (highest in lung vs. other organs) (44); RELMγ (18, 36) and resistin (108) (detectable)		Resistin (detectable) (18, 87)

^*Contribution to pathogenesis primarily indicates the promotion of lung disease development unless specifically noted.

ALI, acute lung injury; APALI, acute pancreatitis-associated lung injury; ARDS, acute respiratory distress syndrome; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; ESLD, end-stage lung disease; N/A, not applicable; O₃, ozone; RELMs, resistin-like molecules; TB, (pulmonary) tuberculosis.

Fig. 3.

Overview of the involvement of the resistin-like molecule (RELM) family in the interconnected etiologies of pulmonary diseases. ALI, acute lung injury; ARDS, acute respiratory distress syndrome; COPD, chronic obstructive pulmonary disease.

To date, the tremendously complex pathogenesis of lung diseases such as PH remains unclear. The cytokines, chemokines, and other mediators, together with the cells that release and respond to them, form a complex network that involves multiple signaling pathways. Knowledge about the roles of RELMs is vital for sorting out these pathways and their regulation during lung inflammation, and for defining an effective therapy that addresses the mechanism of the disease rather than just the symptoms. RELMs also mediate the tumorigenicity of lung cancers. Inflammation in the tumor microenvironment affects every aspect of tumor development, including initiation, promotion, malignant conversion, invasion, and metastasis, as well as response to therapy. In turn, elucidating the signaling network of the inflammatory tumor microenvironment in lungs is crucial to developing novel therapies for complex multifactorial pulmonary inflammatory diseases, especially PH (75), given the striking pathogenic analogies between it and cancer (13). We have a long way to go before the etiologies of these pulmonary diseases become clear. Nevertheless, the advances in our understanding of RELM signaling recapped here are filling in the picture.

RELM-induced vascular lesions and inflammation might also lead to heart failure, atherosclerosis, thrombosis, angiogenesis, and other cardiovascular diseases (2, 49, 65, 135). Additionally, RELMs mediate a variety of pathological processes in fatty liver disease, autoimmune disease, diabetes, inflammatory bowel disease, and chronic kidney disease (49), as well as immune dysfunction in gut, spleen, pancreas (128), etc. (103). Most of these diseases involve the RELM-related lung pathologies discussed in this review (18, 35, 69, 86, 96, 102, 128). All of these inflammation-related diseases remain clinical challenges in current medical practice. Although the study of RELM proteins in physiology and disease has begun to expand rapidly and involvement has been identified in a wide range of inflammatory responses, little is yet known regarding the exact biological function, pleiotropic roles, tissue and cellular distribution, receptors, and downstream signaling of different RELM isoforms. Consequently, some of their reported contributions to lung disease progression and prediction remain controversial. These limitations urge us to define the detailed functions of these family members and further dissect their roles in human disease. The mechanistic findings have been applied to the development of anti-hResistin blocking antibodies for use in PH (74) (Q. Lin, J. T. Skinner, C. Fan, E. N. Hunter, J. J. Gray, N. Biswas, K. Yamaji-Kegan, R. A. Johns, unpublished observations) and other related disorders. These pharmacologic intervention studies are likely to help researchers acquire more critical insights into disease pathogenesis. Nevertheless, we believe that the information summarized here provides a deeper understanding of the mechanisms by which RELMs alter the inflammatory milieu, opening a door for the identification of precise therapeutic targets within the pulmonary microenvironment. Identification of such targets is vital to the development of novel treatment approaches for a variety of disorders in lung and other organs, such as diabetic retinopathy, rheumatoid arthritis, psoriasis, and cancers.

GRANTS

This work was supported by National Institutes of Health (NIH) Centers for Advanced Diagnostics and Experimental Therapeutics in Lung Diseases Stage II (CADET II) Grant 5UH2HL123827-02 and NIH National Heart, Lung, and Blood Institute Grant 1R01HL138497-01A1 (to R.A.J.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Q.L. and R.A.J. prepared figures; Q.L. and R.A.J. drafted manuscript; Q.L. and R.A.J. edited and revised manuscript; Q.L. and R.A.J. approved final version of manuscript.