Role of hyaluronan and hyaluronan-binding proteins in lung pathobiology

Frances E. Lennon, Patrick A. Singleton

Abstract

Hyaluronan (HA) has diverse functions in normal lung homeostasis and pulmonary disease. HA constitutes the major glycosaminoglycan in lung tissue, with HA degradation products, produced by hyaluronidase enzymes and reactive oxygen species, being implicated in several lung diseases, including acute lung injury, asthma, chronic obstructive pulmonary disease, and pulmonary hypertension. The differential activities of HA and its degradation products are due, in part, to regulation of multiple HA-binding proteins, including cluster of differentiation 44 (CD44), Toll-like receptor 4 (TLR4), HA-binding protein 2 (HABP2), and receptor for HA-mediated motility (RHAMM). Recent research indicates that exogenous administration of high-molecular-weight HA can serve as a novel therapeutic intervention for lung diseases, including lipopolysaccharide (LPS)-induced acute lung injury, sepsis/ventilator-induced lung injury, and airway hyperreactivity. This review focuses on the regulatory role of HA and HA-binding proteins in lung pathology and discusses the capacity of HA to augment and inhibit various lung diseases.

  • lung disease
  • therapy

hyaluronan (HA), the major nonsulfated glycosaminoglycan (GAG) in the lung, is a dynamic molecule that can differentially promote or inhibit lung pathology on the basis of its molecular weight and accessibility to various HA-binding proteins. HA comprises a linear repeat of disaccharide units consisting of d-glucuronic acid and N-acetylglucosamine (1, 61, 102, 149) (Fig. 1). The prevalent form of HA in vivo is high-molecular-weight (>1 × 106) HA (HMW-HA) (1, 61, 102). Structurally, HMW-HA exhibits a random coil structure that can expand in aqueous solutions (117, 141). Aqueous HMW-HA is highly viscous and elastic, properties that contribute to its space-filling and filtering functions (39, 117). In the lungs, HA is mainly located in the peribronchial and interalveolar/perialveolar tissue (62).

Fig. 1.

Structure of hyaluronan (HA). HA is composed of a linear repeat of disaccharide units consisting of d-glucuronic acid and N-acetylglucosamine. Molecular weight of HA is >1 × 106 in vivo (140, 141).

Proinflammatory cytokines, including TNFα, IL-1β, and LPS, induce HA production in vitro by various cell types, including endothelial (91), dendritic (8), and fibroblast (153) cells. Increased HA and its degradation products are observed in animal models of chronic obstructive pulmonary disease (COPD) and bleomycin-induced lung injury (31, 101, 160). Furthermore, increased HA levels are observed in bronchoalveolar lavage (BAL) fluid and/or plasma from patients with lung disorders such as pulmonary fibrosis (7), COPD (31), allergic alveolitis (128), asthma (115, 161), interstitial lung disease (14), sarcoidosis (89), and idiopathic pulmonary arterial hypertension (106). An increase in HA production has also been reported in airway epithelial cells in response to tunicamycin-induced endoplasmic reticulum stress (70). The levels of HA are regulated, in part, by the opposing activities of HA synthases and hyaluronidases (18, 141). HA can directly influence cell behavior through binding cell surface receptors. It has long been known that molecular weight is an important factor in regulation of HA's signaling activities. HMW-HA can induce cyclooxygenase-2 expression in endothelial cells via cluster of differentiation 44 (CD44) (94), mediate epithelial-mesenchymal transition during heart valve formation via ErbB2 (13), and enhance endothelial cell barrier function in the lung via CD44, the sphingosine 1-phosphate (S1P1) receptor, Akt, and Rac signaling (124, 125). Low-molecular-weight HA (LMW-HA), on the other hand, has been shown to decrease endothelial cell barrier function (124, 125), stimulate angiogenesis (30), and induce expression of a host of inflammatory mediators in alveolar macrophages, including macrophage inflammatory protein-1α (MIP-1α), regulated upon activation, normal T cell expressed and secreted (RANTES), Mig, IFN-γ-induced protein 10, and plasminogen activator inhibitor 1, via CD44 and Toll-like receptor (TLR) 2 and TLR4 (57, 58, 60, 86). Therefore, the effects of HA are wide-ranging and are dependent not only on its molecular weight, but also on the specific receptors expressed and the cell type involved.

HA Synthases and Lung Disease

HA is synthesized by at least three HA synthases (HAS1, HAS2, and HAS3), which are well conserved evolutionally, despite being located on separate chromosomes (27). The synthases comprise seven membrane-spanning regions (which are hypothesized to form a channel) and a large cytoplasmic loop (the putative substrate-binding domain). The mode of HA synthesis is unusual, in that it is synthesized at the inner face of the plasma membrane, and not inside the Golgi, as is usual for other GAGs. The growing HA molecule is extended at the reducing, rather than the nonreducing, terminus, and as the polymer grows, it is extruded into the extracellular space via the membrane-spanning domains of the HA synthase (98, 140). Although HA synthases catalyze the same reaction, the three enzymes differ in a number of ways: 1) Km values for their substrates (d-glucuronic acid and N-acetylglucosamine), leading to differential rates of HA synthesis (59) and 2) production of >5 × 105 molecular weight (Mr) HA by HAS1 and HAS2 and <5 × 105 Mr HA by HAS3 (152). These differences are believed to contribute to the multiple types of HA matrix secreted by different cell types (24, 59). Studies utilizing HA synthase gene knockout mice indicate that only HAS2 is required for viability, with HAS2 deletion resulting in lethal defects in cardiac development and vascular abnormalities that are rescued by addition of exogenous 7.5 × 105 Mr HA (12, 13).

HA synthase expression is altered in a number of lung pathologies. HAS2 expression is increased in an animal model of monocrotaline-induced pulmonary hypertension, leading to increased total lung HA concentration (105). In addition, increased HAS1 and decreased HAS2 levels are observed in pulmonary artery smooth muscle cells from patients with idiopathic pulmonary arterial hypertension, where total lung HA concentration was also increased (106). In a murine model of asthma, expression of HAS1 and HAS2 is increased in lung tissue (22). Furthermore, an increase in HA accumulation is associated with lung airways, blood vessels, and alveolar interstitium (22). In contrast, HAS1 and HAS2 are decreased in primary airway smooth muscle cells from asthma patients, resulting in secretion of lower levels of HA into the cell culture medium (67). HAS3 levels are increased in the lungs of mice in a cigarette smoke model of COPD (11). These mice also show increased HA deposition in the alveolar walls and peribronchial regions. Further characterization of the HA deposits by electrophoresis indicates that its molecular weight is lower (7 × 104) than that of HA from control lungs (5 × 105). Bai et al. (5) utilized a HA synthase knockout mouse to study the role of HAS3 and LMW-HA in ventilator-induced lung injury (VILI). While they observed an increase in LMW-HA and neutrophil infiltration in control animals at high-ventilation tidal volumes, no increase in LMW-HA was detected in the HAS3 knockout animals, and neutrophil infiltration was decreased (5). Lauer et al. (70) also showed that HA synthesis may be regulated by endoplasmic reticulum stress, as they observed an increase in the HA matrix secreted by airway epithelial cells following treatment with tunicamycin. This matrix stimulated the binding of a macrophage precursor cell line, U937, to the epithelial cell layer (70). Unfortunately, Lauer et al. did not examine the expression of HA synthase in these cells or its activity following tunicamycin treatment. These studies and others indicate that HA synthesis has a role in many lung pathologies (5). HA synthase levels are variable, depending on the pulmonary cell type and disease state. In most of these studies, an increase or a decrease in HA synthase expression is accompanied by the expected increase or decrease in HA deposition in the lung. Although HA synthesis and total HA concentration are important in regulating lung function, we must also consider that HA degradation products can alter downstream signaling pathways and directly affect lung function.

Hyaluronans and Lung Disease

HA is degraded by hyaluronidases to produce lower-molecular-weight (<5 × 105) fragments (47). Six hyaluronidase genes encode HYAL-1, -2, -3, and -4, PHYAL1 (a pseudogene), and PH-20 (27, 130132). A recent study by Hofinger et al. (54) revealed that HA degradation by hyaluronidase enzymes may be pH-dependent. HYAL-1 was found to have maximal and complete HA-degrading activity at pH 3.5–4.0, which is consistent with its role as a HA-degrading enzyme within the lysosome. PH-20 continuously degrades HMW-HA to small oligosaccharides at pH 4.5; at pH 5.5, HA is degraded to larger fragments (54). Unfortunately, the other hyaluronidase proteins were not examined in this study, but the authors speculate that the pH dependence of these enzymes could lead to different biological responses based on the production of different-sized HA fragments at the site of action. Extracellular acidosis is often observed in the airways of smokers and patients suffering from COPD and asthma (92). HA fragments are implicated in the progression of lung diseases (131). HYAL-1 expression is increased in a rat model of monocrotaline-induced pulmonary hypertension, leading to increased fragmentation of native HMW-HA and increased hyaluronidase activity in lung lysates (105). In addition, HYAL-1 is increased in primary airway smooth muscle cells from asthma and COPD patients, and these cells were found to degrade HMW-HA into 2.5 × 105 Mr fragments compared with 7 × 105 Mr fragments for control cells (67). Dentener and colleagues (31) found that HYAL-2 expression was also increased in the lungs of patients with COPD, while HAS2 was decreased. In contrast, HYAL-1 levels were decreased in the lungs of patients with idiopathic pulmonary arterial hypertension (106). However, hyaluronidases are not the only HA-degrading moiety in the lung, and other factors, including reactive oxygen species (ROS) production, can account for the presence and potential regulating activity of lower-molecular-weight HA.

Reactive Oxygen Species and Hyaluronan

In addition to the enzymatic activity of hyaluronidases, HA can be degraded to LMW species by ROS (92, 129, 132). ROS is increased in the lungs with many types of pulmonary diseases (107). Excess lung ROS can be generated from a variety of exogenous (particulate air pollution, cigarette smoke) and endogenous (activation of phagocytes, stimuli-induced NADPH oxidase, mitochondrial electron transport chain, xanthine oxidase, lipid peroxidation) sources (21, 25, 35, 112, 121, 144). Cigarette smoke-generated ROS can degrade HA (20, 28, 85). Recently, Monzon at al. (92) reported that ROS can regulate the expression of HYAL-2 in primary airway epithelial cells. This combination of ROS and HYAL-2 expression stimulates the appearance of ∼7.5 × 104 Mr HA fragments in lung secretions. Monzon at al. speculate that ROS exposure leads to an initial direct effect on HA degradation by ROS followed by sustained effects through the upregulation of hyaluronidase enzymes. Furthermore, in human airway epithelial cells exposed to xanthine/xanthine oxidase, HA fragment accumulation is blocked by addition of SOD or catalase (82). In addition, LMW-HA in the lung is increased in the extracellular SOD knockout mouse compared with wild-type controls (41). Intratracheal crocidolite asbestos exposure increases the appearance of <2.5 × 105 Mr lung HA, with exacerbation of asbestos-induced increases in lung HA production in the extracellular SOD knockout mouse. In addition to endogenous lung hyaluronidase production, many infectious microbes express HA lyases on their surface or secrete the protein. It is considered a virulence factor allowing for degradation of the host organism's extracellular matrix (ECM), thereby increasing permeability and facilitating invasion by the pathogen. Bacteria that express HA lyases responsible for human disease include Streptococcus, Staphylococcus, and Clostridium (48, 113).

Hyaluronic Acid-Binding Proteins in Lung Disease

HA and its degradation products bind to a variety of HA-binding proteins that exist in diverse locales, including the blood, ECM, cell plasma membrane, cytosol, and nucleus (29, 143). Many HA-binding proteins contain Link modules consisting of conserved clusters of positively charged amino acids that bind to HA (29, 99). The differential activities of HA are regulated in the lung, in part, through interactions with HA-binding proteins, including CD44, TLR4, HA-binding protein 2 (HABP2), and receptor for HA-mediated motility (RHAMM). The role of these proteins in lung disease is discussed below (also see Table 1).

View this table:
Table 1.

HA-binding proteins and their associated lung diseases

CD44.

CD44 is a type 1 transmembrane glycoprotein expressed in a variety of lung cell types, including pulmonary epithelial, fibroblast, endothelial, and hematopoietic cells (122, 123). Its expression can be regulated in response to inflammatory stimuli such as LPS and cytokines, including IL-1β and TNFα, and growth factors, such as basic fibroblast growth factor and VEGF (34, 37, 46, 50). Several CD44 isoforms result from alternative exon splicing (52, 77), often occurring between exons 5 and 15, leading to a tandem insertion of one or more variant exons (v1–v10, or exon 6–exon 14) within the membrane proximal region of the extracellular domain (10, 45). It has been reported that CD44 can be expressed in an “active” form, capable of binding HA, or an “inducible” form, capable of binding HA following treatment with a CD44-activating antibody or PMA (72). The exact mechanism by which CD44 may be activated or induced to bind HA is unclear but may involve clustering of the receptor or alterations in glycosylation state. CD44 is modified by extensive N- and O-glycosylations and GAG additions (10, 143). The extracellular domain of CD44 contains clusters of conserved basic residues that are part of a HA-binding Link module common to HA-binding proteins (99, 143). The cytoplasmic domain of CD44 functions to recruit regulatory proteins to the cell membrane and initiate HA-mediated intracellular signaling (10, 143). Some examples of signaling pathways and molecules activated by HA binding of CD44 include Rac activation, leading to lamellipodia formation (103); ERM and merlin proteins (93); Src (124); and Rho kinase (ROCK) (123, 124). The importance of CD44 in the lung has been demonstrated through the use of the CD44 knockout mouse in multiple models of lung disease, including inflammation, vascular leak syndromes, and noninfectious lung diseases, which are discussed below (45, 56, 111, 139, 145, 147).

CD44 AND LPS-INDUCED LUNG INJURY.

LPS is a potent endotoxin from gram-negative bacteria that, when administered intratracheally, produces an inflammatory reaction characterized by disruption of epithelial/endothelial barriers and leakage of fluid, protein, and immune cells into lung air spaces (6). Recently, our laboratory and others have demonstrated increased BAL protein and HA concentration and exaggerated inflammatory cell recruitment of macrophages and neutrophils with LPS-induced lung injury in CD44 knockout mice (56, 126). Increased NF-κB nuclear translocation and cytokine production have also been reported in CD44 knockout mice (56). In this model of intratracheal administration of LPS, it appears that CD44 acts as a negative regulator to limit the in vivo response to LPS and prevent excessive tissue damage. However, Hollingsworth et al. (56) observed decreased macrophage infiltration and chemokine secretion in their model of aerosolized LPS-induced inflammation. These differences are likely accounted for by the different modes of delivery (intratracheal vs. aerosolized) and lower LPS concentration, leading to a milder inflammatory response and faster resolution. One common finding in all these studies is the increased concentration of HA in the BAL fluid of CD44 knockout mice (56, 75, 125).

CD44 AND PNEUMONIA.

Pneumonia, a disease characterized by inflammation of the parenchyma of the lung and alveolar edema, is the sixth-leading cause of death in America (133, 151). In animal models of pneumonia using live Escherichia coli and Streptococcus pneumoniae, CD44-deficient mice exhibited increased expression of the neutrophil chemoattractant proteins KC and MIP-2 (150). However, CD44-deficient mice exhibited increased neutrophil migration and edema formation only in the E. coli model of pneumonia (150). Katoh et al. (65) reported increased levels of HA and soluble CD44 in BAL fluid and increased numbers of CD44-expressing eosinophils in BAL of patients with eosinophilic pneumonia. The increase in CD44 is reported to be due to a local increase in IL-5 production in the lung. Unfortunately, Katoh et al. did not examine the size of HA in the BAL or its effect on other cell types such as neutrophils or macrophages.

CD44 AND TUBERCULOSIS.

Tuberculosis (TB) is a serious lung disease caused by Mycobacteria infection. It is estimated that up to one-third of the world's population has been exposed to Mycobacterium tuberculosis, although only 5–10% of those exposed will develop TB and become infectious (136). CD44 knockout mice were used in two similar studies of TB infection. Both studies report increased neutrophil infiltration and increased lung lesion burden as the infection progressed in the CD44 knockout mice compared with control (66, 71). However, only Leemans et al. (71) reported a decrease in macrophage infiltration into the lung in the early stages of infection and increased bacterial load in the lungs and liver. They also observed increased mortality in the CD44 knockout mice. Kipnis et al. (66) did not observe significant differences in macrophage numbers. This could be due to the lower doses of bacilli used for initial infection than in the study of Leemans et al. Similarly, these lower doses may explain the differences in bacterial load in both studies. Given these results, it would appear that CD44 is necessary for successful control of TB infection at low and high doses. Recently, Hirayama et al. (53) reported that addition of HA to the culture medium of A549 cells infected with M. tuberculosis or Mycobacterium bovis bacillus (Calmette-Guerin) enhanced the growth rate of both bacterial strains. They reported that TB bacilli can use HA as a carbon source and that the Mycobacteria possess hyaluronidase activity (53). Furthermore, they found that mycobacterial growth could be reduced in vivo through the use of the hyaluronidase inhibitor Vcpal (53). One could also speculate that the action of mycobacterial hyaluronidase can also lead to an increase in LMW-HA fragments in the lungs of infected patients, leading to altered HA signaling and pathogenesis.

CD44 AND NONINFECTIOUS LUNG INJURY.

A prevalent animal model for noninfectious lung injury is intratracheal administration of bleomycin, which causes an acute pulmonary epithelial cell injury and an inflammatory response that later subsides and develops into lung fibrosis (83). Lung CD44 expression is increased in the initial inflammatory response, along with a transient increase in HA concentration in the lung interstitium (138). In the CD44 knockout mouse, the bleomycin-induced acute inflammatory response persists, leading to excess immune cell recruitment to the lungs, excess inflammatory cytokine production, decreased transforming growth factor-β activation, progressive HA fragment (<5 × 105 Mr) accumulation, and, ultimately, death (139). In humans, CD44 is upregulated in the lungs of patients with acute lung injury (ALI) (134). Treatment of lung mesenchymal cells isolated from these patients with anti-CD44 antibody attenuated migration and invasion into a fibrin matrix (134). This study did not examine HA localization, concentration, or molecular weight in the lungs of these patients. As the principal ligand for CD44, any alterations in HA size or concentration could greatly influence CD44 signaling.

CD44 AND ASTHMA.

Asthma is an increasingly common lung disease characterized by bronchial hyperresponsiveness and airway thickening due to scarring and inflammation (32, 135). CD44 is overexpressed in the lungs of rats with experimental asthma (73). In addition, CD44 variant 6 (CD44v6) is upregulated in bronchial smooth muscle of asthma patients (49). Antibody blockage of CD44 decreased mast cell adhesion to human bronchial smooth muscle cells in vitro (49), a process associated with airway hyperresponsiveness and remodeling. Klagas et al. (67) reported a decrease in HA secretion by airway smooth muscle cells from asthma patients. Also the molecular weight of the HA was lower than that of the HA secreted by airway smooth muscle cells from healthy volunteers. Klagas et al. reported that CD44 expression is also decreased on airway smooth muscle cells from asthma patients, although the decrease was only significant following 24 h of in vitro culturing.

CD44 AND HYPEROXIA.

Recently, van der Windt et al. (146) reported that CD44 has a protective role in hyperoxia-induced lung injury. Hyperoxia is often used as a treatment to increase tissue oxygenation during ALI or respiratory distress syndrome but can also lead to further lung damage, even in healthy tissue (2). van der Windt et al. reported that CD44 knockout mice have increased mortality compared with wild-type animals and exhibit higher levels of necrosis in their lungs, particularly, in the bronchiolar tissue. Although both groups of mice have increased numbers of neutrophils in BAL fluid after 24 h of hyperoxia, CD44 knockout mice have significantly higher numbers of neutrophils than control mice. CD44 knockout mice also have increased levels of HA in BAL fluid, but, unfortunately, the size of this HA was not determined. No changes were observed in the levels of osteopontin, another CD44 ligand. These data suggest that CD44 protects the lung epithelium during hyperoxia by limiting the neutrophil response and preventing HA buildup in the lung. The role of other HA-binding protein receptors, such as TLR4, which may modulate the functions of CD44 in the lung, should also be considered in this context, considering that TLR4 knockout mice are more susceptible to hyperoxia injury (159). In addition, hyperoxia-mediated pulmonary apoptosis is reduced in the inducible transgenic mouse with the human TLR4 signaling domain expressed in airways (110).

CD44 AND PULMONARY VASCULAR LEAKINESS.

Endothelial cells make up ∼30% of lung tissue and function as a semiselective cellular barrier to regulate the interface between the circulating blood and the vessel wall (26, 80, 108). Disruption of the endothelial cell barrier is a critical feature of inflammation, as well as an important contributing factor to ALI, an inflammatory condition that is a major cause of morbidity and mortality in critically ill patients, because it results in leakage of fluid, protein, and cells into lung air spaces. We have demonstrated that human pulmonary endothelial cells express the CD44 isoforms CD44s (standard form) and CD44v10 (124). In vitro models of pulmonary endothelial cell barrier function indicate that HMW-HA (∼1 × 106) activates CD44s signaling and promotes barrier enhancement through its interaction with the S1P1 receptor and activation of Rac1 signaling, leading to cytoskeletal reorganization, while HA fragments (∼2.5 × 106) activate CD44v10 signaling and induce barrier disruption via S1P3 and Rho signaling (124) (Fig. 2). In addition, we have demonstrated that targeted deletion of CD44 in the mouse pulmonary vasculature increases basal leakiness in the lungs (125). In contrast, pulmonary vascular leak caused by intraperitoneal administration of IL-2 is attenuated in CD44 knockout mice and by CD44 antibody blockage (51, 95, 111).

Fig. 2.

HA regulation of pulmonary vascular function. High-molecular-weight HA (HMW-HA) can be degraded to low-molecular-weight fragments (LMW-HA) by hyaluronidases and/or reactive oxygen species (ROS) production with lung disease (129132). HMW-HA activates the standard form of cluster of differentiation 44 (CD44s) signaling in pulmonary endothelial cells and inhibits HA-binding protein 2 (HABP2) protease activity (81, 124). These events promote increased pulmonary vascular integrity. LMW-HA activates cluster of differentiation 44 variant (CD44v) signaling and induces HABP2 protease activity (81, 124), events that lead to pulmonary vascular leakiness, which is a prominent feature of acute lung injury.

TLR4.

TLRs sense exogenous and endogenous danger-associated molecular motifs and produce inflammatory responses (104). Structurally, TLRs contain an extracellular leucine-rich repeat domain and a cytosolic Toll/IL-1 receptor homology domain (44). TLR4 is the major receptor for LPS and can also bind HA, high-mobility-group protein B1, oxidized lipoproteins, and oxidized phospholipids, since these molecules contain features of “pathogen-associated molecular patterns” (104). Interestingly, CD44 and TLR4 have been shown to be physically associated in a signaling complex following exposure to HA (137).

TLR4 AND LPS-INDUCED LUNG INJURY.

Intratracheal administration of gram-negative bacteria endotoxin, LPS, induces a lung inflammatory reaction (6, 142). Inhibition of TLR4 in animal models protects against LPS-induced lung injury (38, 79, 127). TLR4 knockout animals show decreased neutrophil infiltration and have decreased levels of TNFα, IL-1β, and IL-6 (55). A TLR4-blocking antibody has been reported to decrease lung inflammation in a rabbit model of LPS-induced injury in mechanically ventilated animals (127). In humans, TLR4 loss-of-function mutations attenuate inhaled LPS-induced lung injury (4, 87). Interestingly, CD44-deficient mice have decreased expression of negative regulators of TLR, including IL-1 receptor-associated kinase M, Toll-interacting protein, and TNFα-induced protein 3 (A20) (74). Muto et al. (96) go one step further in their model of the septic response to LPS by showing that pretreatment with HMW-HA (<5 × 105) is protective against LPS-induced shock. CD44 knockout mice are not protected in this model (96). This indicates that CD44 plays an active role in regulating TLR4 signaling events. Alveolar macrophages isolated from control treated and CD44 knockout mice have differences in TNFα and IL-6 expression, with HA-pretreated control macrophages showing decreased expression following LPS exposure. Also, expression of the TLR4-negative regulator TNFα-induced protein 3/A20 is increased in HA-treated macrophages (96).

TLR4 AND NONINFECTIOUS LUNG INJURY.

Intratracheal bleomycin treatment in the double TLR2/TLR4 knockout mouse causes enhanced pulmonary epithelial cell apoptosis, exaggerated lung injury, and impaired inflammatory cell migration, results similar to blocking HA with the Pep-1 peptide in bleomycin-treated wild-type mice (60). In addition, inflammatory cytokine expression by HA fragments (∼1.35 × 105 Mr) is completely blocked in double TLR2/TLR4 knockout mouse peritoneal macrophages and reduced in the TLR4 knockout mouse (60). Using a model of lung inflammation induced by LMW-HA (2 × 105) administered directly to the trachea, Zhao et al. (159) reported that TLR4 acts as a negative regulator. They found an increase in neutrophilic infiltration and red blood cells in BAL fluid of TLR4 knockout animals (159). They reported that TLR deficiency essentially increases LMW-HA-induced lung injury due to an imbalance in the ratio of pro- to anti-inflammatory mediators in the lungs of TLR4 knockout animals. In TLR4 knockout animals, IL-1β, MIP-2, TNFα, and IL-6 levels in BAL fluid are increased following LMW-HA administration and can be rescued by pretreatment with IL-1RA (159). These results are somewhat at odds with the findings of Jiang et al. (60), who reported that TLR4 knockout reduced MIP-2 expression by peritoneal macrophages. This difference may be accounted for by the cell-specific effects of TLR4 and HA. Scheibner et al. (116) reported that TLR2, but not TLR3 or TLR5 (TLR4 was not examined), is required for peritoneal macrophage activation and MIP-1α expression by LMW-HA (2 × 105), which can be blocked by HMW-HA (6 × 106) (116).

TLR4 AND AIRWAY HYPERREACTIVITY.

Ozone-exposed mice develop airway hyperreactivity and accumulate HA fragments (<2 × 105 Mr) in BAL fluid (42). CD44 and TLR4 knockout mice, as well as mice treated with a HA-binding peptide, are protected from ozone- and HA fragment (2–4 × 105 Mr)-induced airway hyperreactivity (42, 43). The effects of LMW-HA in airway hyperreactivity may be mediated through alveolar macrophages. Garantziotis et al. (43) reported that while there was no TLR4/HA colocalization on airway epithelial cells in their model of airway hyperresponsive, there were significant increases in HA and TLR4 staining on alveolar macrophages following ozone exposure.

HABP2.

HABP2, also called factor VII-activating protease, is a HA-binding extracellular serine protease involved in the extrinsic pathway of blood coagulation via activation of factor VII and fibrinolysis via activation of pro-urokinase-type plasminogen activator (23, 33, 64, 114). It is expressed as a single amino acid chain proenzyme that undergoes autocatalytic cleavage upon binding of a ligand (63). The mature enzyme consists of a trypsin-like catalytic domain, linked via a disulfide bond to the kringle domain, and three epidermal growth factor-like domains (64, 114). The second and third epidermal growth factor-like domains form the polyanion-binding domain (3).

HABP2 AND ACUTE RESPIRATORY DISTRESS SYNDROME.

HABP2 levels and activity are increased in the BAL fluid of mechanically ventilated patients with early acute respiratory distress syndrome (ARDS) compared with patients with cardiogenic pulmonary edema or healthy controls (155). In patients who died from ARDS, immunohistochemical analysis of excised lungs revealed increased HABP2 levels in alveolar macrophages and bronchial epithelial and pulmonary endothelial cells (155). This suggests a role for HABP2 in the pathogenesis of ARDS.

HABP2 AND LUNG INJURY WITH PULMONARY VASCULAR LEAKINESS.

As stated previously, intratracheal LPS administration produces an inflammatory reaction characterized by disruption of epithelial and endothelial cellular barriers, with leakage of fluid, protein, and immune cells into lung air spaces (6, 142). Although HABP2 is mainly produced in the liver, we and others have demonstrated that the pulmonary endothelium expresses HABP2, which is upregulated with lung injury (81, 155). HABP2 promotes LPS- and HA fragment (∼2.5 × 106 Mr)-mediated human pulmonary endothelial cell barrier disruption through a mechanism that involves protease-activated receptors and inhibits HMW-HA-mediated endothelial barrier protection in vitro (81) (Fig. 2). We determined the contribution of vascular HABP2 to lung injury in mice by inhibiting HABP2 through intravenous administration of HABP2 small interfering RNA and observed attenuation of LPS-induced ALI (81). In addition, vascular inhibition of HABP2 expression attenuates another mouse model of lung injury with pulmonary vascular hyperpermeability, VILI, demonstrating an important role of HABP2 in the development of lung disease (81).

RHAMM.

RHAMM is found in diverse cellular locales, including the cell surface, cytosol, mitochondria, and nucleus (143). RHAMM activates ERK1/2 and regulates mitotic-spindle integrity (84). RHAMM is alternatively spliced, similar to CD44, and these two HA-binding proteins are often coexpressed in pulmonary cells (78, 143). In some cases, RHAMM can compensate for CD44 function (100).

RHAMM AND NONINFECTIOUS LUNG INJURY.

As stated previously, one common animal model for noninfectious lung injury is intratracheal administration of bleomycin (83). RHAMM expression is increased in lung macrophages with bleomycin treatment (158). Intraperitoneal injection of anti-RHAMM antibody attenuates bleomycin-induced lung macrophage recruitment and reduction of alveolar septae thickening and early indications of lung fibrosis (158).

RHAMM AND AIRWAY CILIARY FUNCTION AND MUCOSAL HOST DEFENSE.

HA binding to RHAMM in ciliated airway epithelial cells stimulates ciliary beat frequency, suggesting a role for HA and RHAMM in airway mucosal host defense (36). In addition, RHAMM antibody blockage attenuates xanthine/xanthine oxidase-mediated increase in human airway epithelial ciliary beat frequency (82).

Protective Effects of Exogenous Administration of HMW-HA

Although HMW-HA (>1 × 106) is produced endogenously and is an integral component of the ECM, synovial fluid, and vitreous humor, recent attention has been focused on the use of exogenously administered HMW-HA in a variety of diseases, including lung disease (15, 40, 68, 156). In vitro, exogenous administration of HMW-HA inhibits ROS, nitrotyrosine, and inflammatory cytokine production and also promotes immune tolerance (9, 88, 157). In addition, excess production of endogenous HMW-HA in mice overexpressing HAS2 in airway epithelia protects against bleomycin-induced lung injury and ozone-induced airway hyperresponsiveness (42, 60).

Exogenous HMW-HA and LPS-induced lung injury.

We recently demonstrated that intravenous administration of HMW-HA (∼1 × 106) 4 h after intratracheal administration of LPS protects against lung injury in mice (125, 126). This finding is consistent with results from the study of Nadkarni et al. (97), who demonstrated that pretreatment of hamsters with aerosolized HMW-HA protects against endotoxin-induced lung injury. Interestingly, these authors noted that treatment with aerosolized HMW-HA after endotoxin treatment actually enhanced lung inflammation, indicating that the timing and route of administration are important determinants of HMW-HA's effectiveness.

Exogenous HMW-HA and sepsis/VILI.

Intraperitoneal administration of HMW-HA (1.6 × 106) 18 h before mechanical ventilation with a low tidal volume (7 ml/kg) and carotid artery administration of LPS (to induce sepsis) protects rats from lung injury (76). In these same studies, simultaneous intravenous administration of HMW-HA and initiation of ventilation also protected from lung injury (76). Interestingly, the use of 3.5 × 104 Mr HA showed partial protection in these models, but to a lesser extent than HMW-HA (76). Our laboratory had demonstrated that intravenous administration of HMW-HA protects from VILI in mice (Fig. 3).

Fig. 3.

HMW-HA protects against ventilator-induced lung injury. Male C57BL/6J wild-type mice (8–10 wk old; Jackson Laboratories, Bar Harbor, ME) were anesthetized with an intraperitoneal injection of ketamine (150 mg/kg) and acetylpromazine (15 mg/kg); then the right internal jugular vein was exposed via neck incision. Mice were allowed to spontaneously breathe or were ventilated (40 ml/kg tidal volume) for 4 h. After 1 h of ventilation, mice received HMW-HA (1.5 mg/kg) or saline control through the internal jugular vein. Mice were killed after 4 h of ventilator-induced lung injury (VILI), and bronchoalveolar lavage protein was analyzed (n = 5 mice per group). *Statistically significant difference (P < 0.05) from spontaneously breathing (SB) group. There is also a statistically significant difference (P < 0.05) between VILI alone and HMW-HA + VILI.

Exogenous HMW-HA and airway hyperreactivity.

Aerosolized HMW-HA reduces neutrophil elastase- and pancreatic elastase-induced bronchoconstriction in sheep (16, 118120). In addition, oropharyngeal administration of exogenous HMW-HA before or after ozone exposure significantly attenuates airway hyperreactivity in mice (42). Furthermore, pretreatment of aerosolized HMW-HA protects asthma patients from exercise-induced bronchoconstriction (109).

Exogenous HA and experimental emphysema.

Aerosolized 1.5 × 105 Mr HA given to mice 1 h before a model of cigarette smoke-induced pulmonary emphysema protected against lung injury (17, 19). Although this LMW-HA provides protection, it would be prudent to examine whether HMW-HA offers greater protection, as observed in the sepsis/VILI model (see above).

Exogenous HMW-HA and chronic bronchitis.

In patients with chronic bronchitis, subcutaneous administration of HMW-HA for 6 mo protects against acute exacerbations and results in less consumption of antibiotics (148).

Given the numerous animal and patient studies indicating that HMW-HA protects against a variety of lung diseases, the therapeutic potential of exogenously administered HMW-HA warrants further study. It should be noted, however, that intraperitoneal injection of IL-2, which causes systemic vascular leak, is attenuated in CD44 knockout mice and with intraperitoneal injection of human umbilical cord HA, anti-CD44 (9F3) mouse antibody or the HA-specific binding peptide Pep-1 (51, 95, 111). Whether intravenous (vs. intraperitoneal) administration of HMW-HA protects from IL-2-induced vascular leak syndrome remains to be determined. These studies by Mustafa et al. (95) with IL-2 use HA that is not purified to remove LMW-HA. We previously reported that LMW-HA induces human endothelial cell barrier disruption in vitro (124). The IL-2 model involves intraperitoneal injection three times a day for 3 days followed by a single dose on day 4 (51, 95, 111, 126). Therefore, the targeting of injury (lung vs. systemic model), the IL-2 receptor expression level of the pulmonary endothelium, the molecular weight of HA, and the route of HA delivery (intravenous vs. intraperitoneal) can be important factors in the effectiveness of HMW-HA.

Concluding Remarks

HA, its degradation products, and its target binding proteins are important regulators of a multitude of lung diseases. They are also involved in lung cancer oncogenic function, which is beyond the scope of this review. The effects of HA are complex and dependent on its concentration in the lung and molecular weight. Different-sized HA species are regulated by the opposing actions of HA synthases, hyaluronidases, and ROS production. The LMW-HA produced in pathological conditions of the lung can selectively bind to HA-binding proteins upregulated in specific pulmonary cell types in various disease states (81, 124, 125) (Fig. 4). Future areas of investigation will include elucidating the exact mechanisms by which HA and HA fragments can differentially regulate HA-binding protein function. Furthermore, although this review focused on four HA-binding proteins, several others, including versican, TNFα-induced protein-6, and stabilin (69, 90, 99, 154), could potentially regulate lung pathology. Although many HA-binding proteins can be expressed by the same cell type in the lung, certain HA-binding proteins, including HABP2 and CD44v10, are overexpressed in lung pathology and can account for the differential effects of HMW-HA vs. LMW-HA (81, 124, 125). Understanding the complexities of HA in lung disease will allow for future therapeutic exploration.

Fig. 4.

Schematic diagram of factors that regulate HA-mediated lung pathobiology. HA mainly exists as HMW-HA in normal lung. Total HMW-HA levels can change in lung pathology due, in part, to the levels and activity of HA synthases. In several lung diseases, HMW-HA is degraded to lower-molecular-weight HA by hyaluronidase enzymes and ROS. HMW-HA and LMW-HA can differentially bind to and have opposing effects on various HA-binding proteins that exist extracellularly (HABP2) on the cell surface (CD44 and Toll-like receptor 4), as well as in other cellular locales (receptor for HA-mediated motility). Therefore, the type of HA-binding protein(s) upregulated in various lung diseases can often determine the functional effects of HA. Furthermore, cell types (e.g., epithelial, endothelial, immune, smooth muscle, fibroblast) targeted by HMW-HA and LWM-HA will have a distinct impact on lung disease progression.

GRANTS

P. A. Singleton was supported in part by American Heart Association National Scientist Development Grant 0730277N, American Lung Association National Biomedical Research Grant RG-75229-N, and National Heart, Lung, and Blood Institute Grant RO1-HL-095723.

DISCLOSURES

F. E. Lennon has no conflict of interest. P. A. Singleton holds a provisional patent involving applications of hyaluronan with the University of Chicago and has received no financial gain.

REFERENCES

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.
  67. 67.
  68. 68.
  69. 69.
  70. 70.
  71. 71.
  72. 72.
  73. 73.
  74. 74.
  75. 75.
  76. 76.
  77. 77.
  78. 78.
  79. 79.
  80. 80.
  81. 81.
  82. 82.
  83. 83.
  84. 84.
  85. 85.
  86. 86.
  87. 87.
  88. 88.
  89. 89.
  90. 90.
  91. 91.
  92. 92.
  93. 93.
  94. 94.
  95. 95.
  96. 96.
  97. 97.
  98. 98.
  99. 99.
  100. 100.
  101. 101.
  102. 102.
  103. 103.
  104. 104.
  105. 105.
  106. 106.
  107. 107.
  108. 108.
  109. 109.
  110. 110.
  111. 111.
  112. 112.
  113. 113.
  114. 114.
  115. 115.
  116. 116.
  117. 117.
  118. 118.
  119. 119.
  120. 120.
  121. 121.
  122. 122.
  123. 123.
  124. 124.
  125. 125.
  126. 126.
  127. 127.
  128. 128.
  129. 129.
  130. 130.
  131. 131.
  132. 132.
  133. 133.
  134. 134.
  135. 135.
  136. 136.
  137. 137.
  138. 138.
  139. 139.
  140. 140.
  141. 141.
  142. 142.
  143. 143.
  144. 144.
  145. 145.
  146. 146.
  147. 147.
  148. 148.
  149. 149.
  150. 150.
  151. 151.
  152. 152.
  153. 153.
  154. 154.
  155. 155.
  156. 156.
  157. 157.
  158. 158.
  159. 159.
  160. 160.
  161. 161.
View Abstract