Innate Immunity of Airway Epithelium and COPD

The mammalian immune system consists of two branchesinnate and adaptive immune systems and together they provide protection against infection. Innate immunity is a first line of host defense and is responsible for immediate recognition of pathogens to prevent microbial invasion. In addition innate immune responses also stimulate adaptive immune system (Medzhitov and Janeway, 1997). Cellular components of innate immune system include mucosal epithelial cells, macrophages, neutrophils, natural killer cells, basophils, eosinophils and others. The airway mucosa represents the body’s largest mucosal surface and is the first point of contact for inhaled microorganisms, environmental pollutants, airborne allergens and cigarette smoke (Diamond et al., 2000). Airway mucosa provides protection against potentially hazardous inhaled factors by multiple mechanisms. For instance, mucus secreted by the airway epithelium covers the apical surface of airway epithelium and traps inhaled microorganisms, allergens and particulate material. The trapped material is then cleared by mucociliary escalator away from lungs and towards the pharynx. Tight junctions between the polarized airway epithelial cells restrict the paracellular movement of solutes and ions, and prevent pathogens from gaining access to the submucosal compartment. In addition to its role as a physical barrier between environmental factors and internal milieu, airway epithelial cells also play a critical role in bridging innate and adaptive immune defenses (Hammad and Lambrecht, 2011; Kato and Schleimer, 2007). Airway epithelial cells express number of innate immune receptors also known as pattern recognition molecules, which recognizes pathogen-associated molecular patterns (PAMPS) or danger-associated molecular patterns (DAMPS) to initiate appropriate innate defense mechanisms. This includes elaboration of antimicrobial molecules, proinflammatory cytokines and chemokines that recruits and activates other mucosal innate immune cells. The responses of activated innate immune cells lead to recruitment of immune cells into epithelium or airway lumen and initiate adaptive immune responses. Continuous exposure to environmental stimuli, such as cigarette smoke, noxious gases or other environmental hazards may lead to prolonged and aberrant activation of airway epithelial cells resulting in excessive expression of pro-inflammatory cytokines and chemokines that recruit large number of inflammatory cells into airway lumen. This in turn leads to persistent inflammation, airway damage and abnormal repair, impaired innate immune responses. There are reports suggesting that exposure to cigarette smoke also

dampens the needed innate immune responses to infection, thereby promoting the persistence of infecting organism. This may result in delayed but sustained inflammation that can lead to progression of lung disease. In this chapter, we will discuss how the impaired innate immune defense mechanisms fail to provide protection against invading pathogens and its impact on progression of lung disease in patients with chronic obstructive pulmonary disease (COPD).

Barrier function of airway epithelium
Airway epithelium lines the entire airway mucosa. In normal adult human, the large airways are cartilaginous and mainly made up of ciliated cells, mucus producing goblet cells, undifferentiated columnar cells and basal cells with a capacity to multiply and differentiate into ciliated or goblet cells. Large airways are also surrounded by submucosal and serous glands. As the large airways branches out, it gradually becomes noncartilaginous, loses surrounding submucosal and serous glands, the cells become more columnar and cuboidal, and Clara secreting cells replace goblet cells in the small airways. Airway epithelium also consists of other minor cell types such as neuroendocrine cells, dendritic cells and others.
The three essential components that contributes to barrier function of airway epithelium are mucociliary apparatus (Knowles and Boucher, 2002), intercellular tight and adherens junctions (Pohl et al., 2009) that regulates epithelial paracellular permeability, and secreted antimicrobial products that kill the inhaled pathogens (Bals and Hiemstra, 2004).

Mucociliary clearance
The primary players of mucociliary apparatus are mucus produced by goblet cells and submucosal glands that overlay the airway epithelium and cilia. Mucociliary dysfunction results in recurrent and persistent respiratory infections as evidenced in patients with cystic fibrosis, ciliary dyskinesia and COPD (Bhowmik et al., 2009;Jansen et al., 1995;Livraghi and Randell, 2007;Sethi, 2000). In COPD patients, the dysfunction of mucociliary clearance is due to combined effect of mucus hypersecretion, increased viscosity of mucus and dysfunction or loss of cilia (Mehta et al., 2008). The airway mucus is a viscoelastic gel and contains more than 200 proteins, and it is secreted by goblet cells that are present in the airway epithelium and by submucosal glands. The main components of airway mucus are mucins, which are high molecular weight glycoproteins and cross link to form structural framework of mucus barrier (Rose et al., 2001;Thornton et al., 2008). At least 12 mucins are detected in human lungs, of these MUC5AC and MUC5B are the predominant mucins in normal airways . Airways infection with virus or bacteria, exposure to toxic agents such as cigarette smoke and pollutants that induce airway inflammation and oxidative stress have been shown to upregulate expression of MUC5AC and MUC5B (Borchers et al., 1999;Casalino-Matsuda et al., 2009;Dohrman et al., 1998;Gensch et al., 2004;Haswell et al., 2010;Shao et al., 2004). Cigarette smoke induces expression of number of inflammatory mediators including IL-1 , IL-8, TNF-, MCP-1, leukotrienes through oxidative stress-related pathways from airway epithelial cells, resident macrophages and infiltrated neutrophils, which can increase mucus secretion (Adcock et al., 2011;Choi et al., 2010;Cohen et al., 2009;Mebratu et al., 2011). Cigarette smoke also causes www.intechopen.com mucus hypersecretion by increasing expression of hypoxia-induced factor 1 and growth factors such as TGF-, and EGF ligands (Yu et al., 2011a, b). Smokers with COPD also show goblet cell metaplasia and submucosal gland hypertrophy (Innes et al., 2006). Increased EGF receptor expression and activation and increased expression of platelet activating factor caused by cigarette smoke are thought to play a role in development of goblet cell metaplasia (Curran and Cohn, 2010;Komori et al., 2001;O'Donnell et al., 2004). Cigarette smoke decreases water and ion transport by inhibiting apical chloride channel and basolaterally located potassium channel in primary human and mouse airway epithelial cells Savitski et al., 2009). This essentially reduces the periciliary liquid layer in which cilia can beat rapidly and also increases the viscosity of mucus resulting in reduced clearance of mucus from the airways. In addition, respiratory epithelial cells exposed to cigarette smoke extract or condensate showed 70% less cilia and shorter cilia compared to control cells ). Mice exposed to cigarette smoke although showed slight increase in ciliary beat frequency at 6 weeks and 3 months, it was significantly reduced at 6 months and these mice also showed significant loss of tracheal ciliated cells (Simet et al., 2010). Decreased number of cilia, reduced ciliary function combined with hypersecretion of mucin, increased viscoelasticity of secreted mucus in COPD patients can lead to airways obstruction and promote persistence of trapped pathogens in the airways Voynow et al., 2006). Persistence of bacteria or viruses can further increase production of mucus in the airways (Baginski et al., 2006). Fig. 1. Airway epithelial cells isolated from COPD patient cultured at air/liquid interface show more goblet cells (arrows) than the similarly grown normal airway epithelial cells.

Normal COPD
Another feature that is frequently noted in airways of COPD patients is squamous metaplasia (Araya et al., 2007) and it correlates with the severity of airway obstruction (Cosio et al., 1978). The airway epithelium exposed to cigarette smoke responds by secreting TGF- (de Boer et al., 1998), which is required for repair of injured epithelium and maintain homeostasis. However, chronic exposure to cigarette smoke can induce sustained production of TGF-and increased TGF-activation leading to expression of the 6 integrin, a TGF-responsive gene (Wang et al., 1996). This in turn contributes to a phenotypic switch from columnar ciliated to squamous epithelium (Masui et al., 1986a;Masui et al., 1986b). Squamous epithelial cells secrete increased amounts of IL-1 , which acts as a paracrine factor with adjacent airway fibroblasts to further activate TGF- (Araya et al., 2006), thereby increasing squamous metaplasia and further contributing to impaired barrier function and persistence of inhaled pathogens.
In our laboratory, we observed that cultured airway epithelial cells isolated from COPD patients show goblet cell metaplasia, decreased number of ciliated cells (Figure 1), and increased MMP activity suggesting that epigenetic changes that occur in vivo are maintained even when cells are expanded ex vivo (Schneider et al., 2010). COPD epithelial cells also showed increased viral load following rhinovirus challenge compared to normal cells. Similarly, we also found that elastase/LPS exposed mice which show typical features of COPD, including emphysema, airway remodeling, diffuse lung inflammation and goblet cell hypertrophy, also showed increased persistence of virus compared to normal mice following rhinovirus challenge and majority of the virus particles were observed in the airway epithelium (Sajjan et al., 2009). Rinovirus infection increased mucin expression further in these mice. Since goblet cells are the target for rhinovirus infection (Lachowicz-Scroggins et al., 2010) we suggest that COPD airway epithelial cultures which have increased number of goblet cells are more susceptible to rhinovirus infection than the controls. Patients with COPD, cystic fibrosis and asthma show goblet cell metaplasia and this may be one of the reasons these patients are more susceptible to rhinovirus infection. In addition, airway epithelial mucins also interact with several other respiratory pathogens including Pseudomonas aeruginosa, Staphylococcus aureus, Heamophilus influenza, Streptococcus pneumonia, Burkholderia cenocepacia, influenza virus, adenovirus and coronavirus (Landry et al., 2006;Matrosovich and Klenk, 2003;Plotkowski et al., 1993;Ryan et al., 2001;Walters et al., 2002). The bound pathogens which are cleared under normal conditions, persist in the airway lumen when the mucociliary clearance is impaired and initiate inflammatory response and damage the airway epithelium.

Junctional adherens complexes and airway epithelial permeability
Epithelial permeability is maintained through the cooperation of two mutually exclusive structural components: Tight junctions and adherence junctions on the lateral membranes (Pohl et al., 2009). While tight junctions regulate the transport of solutes and ions across epithelia, adherence junctions mediate cell to cell adhesion (Hartsock and Nelson, 2008;Schneeberger and Lynch, 2004;Shin et al., 2006). Under homeostatic conditions, these intercellular junctions prevent inhaled pathogens and also serve as signaling platforms that regulate gene expression, cell proliferation and differentiation (Balda and Matter, 2009;Koch and Nusrat, 2009). Therefore disassociation or sustained insult that affects junctional complex will disrupt not only barrier function, but also prevent normal repair of airway epithelium. Compared to control nonsmokers, airway epithelium is leaky, hyperproliferative and abnormally differentiated in smokers (Hogg and Timens, 2009). Consistent with this observation, various in vivo and in vitro studies showed that cigarette smoke increases airway epithelial permeability (Boucher et al., 1980;Gangl et al., 2009;Olivera et al., 2007;Serikov et al., 2006). Recently, transcriptome analysis of airway epithelial cells from normal and COPD patients revealed global down-regulation of physiological tight junction complex gene expression (Shaykhiev et al., 2011). Further, normal airway epithelial cells exposed to cigarette smoke extract also showed similar down-regulation of genes related to tight junction complex. This was associated with decreased expression of PTEN and FOXO3A, a transcriptional factor in the PTEN pathway, suggesting that cigarette smoke down-regulates expression of apical junctional complex genes by modulating PTEN signaling pathway. Consistent with this notion, cigarette smoke in combination with IL-1 has been shown to induce disassembly of tight junction complex in endothelial cells by suppressing PTEN activity (Barbieri et al., 2008). Chen et al showed that cigarette smoke also alters epithelial permeability by disrupting cell polarity via activation of EGFR, dissociation of -catenin and E-cadherin from adherence junctional complex and redistribution of apical MUC1 membrane bound mucin to cytoplasm (Chen et al., 2010). In a homestatic epithelium, -catenin cooperates with E-cadherin to form apical junctional complex and maintain cell polarity (Xu and Kimelman, 2007). In airway regeneration or oncogenic formation -catenin translocates to nucleus, and activates canonical Wnt signaling pathway (Mazieres et al., 2005;Tian et al., 2009). Similar to -catenin, the cytoplasmic tail of MUC1 also supports structural barrier during homeostasis (Chen et al., 2010). Since cigarette smoke causes aberrant activation of both EGFR and canonical Wnt/catenin signaling (Khan et al., 2008;Lemjabbar et al., 2003), it is plausible that chronic cigarette smoke exposure decreases barrier function and promote microbial invasion of airway epithelium.

Antimicrobial products of airway epithelium
In addition to acting as a physical barrier, airway epithelial cells also secrete antimicrobial substances, which include enzymes, protease inhibitors, oxidants and antimicrobial peptides. Lysozyme is an enzyme found in airway epithelial secretions and exerts antimicrobial effect against wide range of gram-positive bacteria by degrading peptidoglycan layer (Ibrahim et al., 2002). Lysozyme is also effective against gramnegative bacteria in the presence of lactoferrin, which disrupts the outer membrane allowing lysozyme to gain access to peptidoglycan layer (Ellison and Giehl, 1991). Lactoferrin is an iron-chelator and inhibit microbial growth by sequestering iron which is essential for microbial respiration (Ganz, 2002). Lactoferrin also display antiviral activity against both RNA and DNA viruses either by inhibiting binding of virus to host cells or by binding to virus itself (van der Strate et al., 2001;Laube et al., 2006). Lactoferrin levels increase in response to bacterial and viral infections. Epithelial cells produce protease inhibitors, such as secretory leukoprotease inhibitor (SLPI), elastase inhibitor, 1antiprotease and antichymotrypsin. These protease inhibitors mitigate the effects of proteases expressed by pathogens and recruited innate immune cells. Administration of SLPI decreased the levels of IL-8 and elastase activity in airway secretion of cystic fibrosis patients (McElvaney et al., 1992).
Human beta defensins (hBD) are the most abundant antimicrobial peptides expressed on the surface of airway epithelium and are effective against wide range of bacteria and viruses (Ganz, 2003;Kota et al., 2008;McCray and Bentley, 1997). While hBD1 is constitutively expressed, hBD2 to hBD4 expression is induced by LPS via NF-κB activation and also by IL-1 (Becker et al., 2000;Singh et al., 1998). hBD2 is induced by P. aeruginosa infection in normal but not in cystic fibrosis airway epithelia (Dauletbaev et al., 2002). Environmental factors such as air pollutants decrease defensin gene expression in the airways (Laube et al., 2006). In CF airway epithelia activity of hBD2 is also decreased due to increased salt concentration (Goldman et al., 1997). Cathelicidins are another class of antimicrobial peptides and LL37 is the only human cathelicidin identified to date. LL37 bind to LPS and inactivate its biological function. Overexpression of human LL37 in CF mouse model increased killing of P. aeruginosa and reduced the ability of this bacterium to colonize the airways (Bals et al., 1998).

www.intechopen.com
Airway epithelial cells also generate oxidants such as nitric oxide (NO) and hydrogen peroxide. Three NO synthases contribute to production of NO in airway epithelia: the constitutively expressed NOS1 and NOS3 and inducible NOS2. Viral infections and proinflammatory cytokines induce expression of NOS2 and defective NOS2 expression is responsible for increased viral replication in cystic fibrosis and overexpression of NOS2 provides protection against viral infection (Zheng et al., 2003;Zheng et al., 2004). Hydrogen peroxide is produced by dual oxidase 1 and 2. These belong to a family of NADPH oxidases and are located in the plasma membrane and secrete hydrogen peroxide to extracellular milieu. The dual oxidase-generated hydrogen peroxide in combination with thiocyanate and lactoperoxidase generates the microbicidal oxidant hypothiocyanite , which effectively kills both gram positive and gram negative bacteria and this innate defense mechanism is defective in cystic fibrosis airway epithelium due to impaired transport of thiocyanate (Moskwa et al., 2007).
In COPD patients, levels of lysozyme and SLPI decrease with bacterial infection, while lactoferrin levels remain unchanged (Parameswaran et al., 2011). Lower levels of salivary lysozyme in clinically stable COPD patients correlated with increased risk of exacerbations (Taylor et al., 1995). Reduced lysozyme levels in COPD is thought to be due to degradation by proteases elaborated by bacterial pathogens or neutrophils (Jacquot et al., 1985;Taggart et al., 2001). These proteases also inactivate SLPI (Parameswaran et al., 2009). In addtion, SLPI forms complexes with neutrophil elastase and binds to negatively charged membranes, thus decreasing the levels of SLPI further in the airway secretions during infection. In clinically stable patients however, the levels of SLPI were increased compared to smokers without COPD and never smokers (Tsoumakidou et al., 2010). In contrast, hBD2 was absent in COPD patients. Herr et al showed that hBD2 is significantly reduced in pharyngeal wash and suptum of current or former smokers compared to non-smokers, and exposure of airway epithelium to cigarette smoke in vitro inhibited induction of HBD2 by bacteria (Herr et al., 2009). Recently, we showed that COPD airway epithelial cells show a trend in decreased expression of NOS2 and Duox oxidases and this was associated with impaired clearance of rhinovirus (Schneider et al., 2010).

Innate immune receptors of airway epithelium
Airway epithelium in addition to providing a physical barrier, it also plays a pivotal role in recognition of pathogens and releasing appropriate chemokine and cytokines to initiate an inflammatory response. This inflammatory response includes recruitment of phagocytes to clear pathogens that are not cleared by barrier function of epithelium, and immune cells, such as dendritic cells and lymphocytes that initiate adaptive immune response. Airway epithelium recognizes pathogens or pathogen associated molecular patterns (PAMPS) by innate immune receptors also known as pattern recognition receptors (PRRs), which are germ-line encoded receptors. One of best characterized PRRs are Toll-like receptors (TLRs) Medzhitov, 2001).

Toll-like receptors
TLRs are type I transmembrane receptors with an extracellular domain that contains leucine-rich-repeat motifs, a transmembrane domain and a cytoplasmic domain known as www.intechopen.com the toll/interleukin-1 receptor (TIR) homology domain (Hoffmann, 2003) (Figure 2). To date thirteen TLRs have been identified in mammalian system. Only TLRs1 to 10 are expressed in humans. TLRs1, -2, -4, -5 and -6 are expressed on the cell surface and TLRs3, -7,-8 and -9 are expressed in the endosomes, lysozomes and the endoplastic reticulum. (Kawai and Akira, 2009). TLRs recognize a wide range of PAMPS-lipoproteins by TLRs 1, -2, and -6 (Aliprantis et Schwandner et al., 1999;Takeuchi et al., 2001;Takeuchi et al., 2002), LPS by TLR4 (Poltorak et al., 1998), flagella by TLR5 (Hayashi et al., 2001), DNA by TLR9 (Hemmi et al., 2000), and RNA by TLR3, -7 and -8 (Alexopoulou et al., 2001;Diebold et al., 2004;Heil et al., 2004). TLR4 also recognizes respiratory syncytial virus (Kurt-Jones et al., 2000).  Fig. 2. Impact of cigarette smoke on persistence of bacteria and inflammation. Under homeostasis, TLR4 recognizes infecting bacteria and activates both MAP kinase and NF-kB pathway to stimulate normal levels of CXCL-8, IL-6 and IL-1 to recruit neutrophils, which clear bacteria. Decreased expression of TLR4 caused by acute exposure to cigarette smoke attenuates release of CXCL-8, IL-6 and IL-1 , there by decreasing the neutrophil infiltration and increasing the bacterial persistence. Under chronic exposure as noted in COPD patients, if the TLR4 expression is increased, then chemokine and cytokine expression is increased leading to decreased bacteria coupled with increased inflammation.
The airway epithelium expresses all 10 TLRs, but the expression of TLR2 to TLR6 is stronger than the others. Expression of TLRs7 through -10 is variable depending on type of cells used (Mayer et al., 2007;Platz et al., 2004;Sha et al., 2004). Expression of TLRs 1 through -6 and -9 on the cell surface was confirmed by flow cytometry (Greene et al., 2005). However the signaling from these TLRs depends on the expression of adaptor molecules and coreceptors. Primary airway epithelial cells are hyporesponsive to LPS despite expressing TLR4 and this is because of reduced surface expression of co-receptor CD14 and low expression levels of co-stimulatory molecule MD2 (Jia et al., 2004). This may be necessary to restrict TLR4 activation under unstimulated conditions to prevent chronic inflammation of airways that is constantly exposed to inhaled bacteria and endotoxin. On the contrary, LPS was shown to activate TLR4 signaling in small airway and alveolar epithelial cells even though the TLR4 was localized to cytoplasmic compartment (Guillot et al., 2004). More recently John et al attributed chronic colonization of bacteria in CF airways to decreased expression of TLR4 in CF airway epithelial cells (John et al., 2010). TLR2, which is expressed on the apical surface of polarized airway cells is mobilized into an apical lipid raft receptor complex following P. aeruginosa infection and initiate signalling (Soong et al., 2004). TLR5 recgonizes flagella of P. aeruginosa and Burkholderia cenocepacia and activate NF-κB (Adamo et al., 2004;Urban et al., 2004;Zhang et al., 2005). Haemophilus infIuenzae traverses polarized airway epithelial cells by interacting with TLR2, which then activates p38 mitogen activated protein (MAP) kinase and TGF-Signalling (Beisswenger et al., 2007). TLR3 recognizes double stranded (ds)-RNA, an intermediate generated during RNA virus replication and elicits chemokine and type I IFN responses by MyD88-independent signaling mechanism (Gern et al., 2003;Wang et al., 2009). Upon ligation of ds-RNA, TRIF and TRAM (TRIFrelated adaptor molecule) are recruited to TIR domain of TLR3 and TRAM acts as a bridge between TLR and TRIF and this allows activation of TRIF-dependent signaling leading to activation of IRF3 via IKKε/TBK-1 to stimulate IFN production or activation of NF-κB via IKK /IKK to stimulate CXCL-8 expression . The recognition of double-stranded RNA by TLR3 also increases expression of hBD2 (Duits et al., 2003). Viral or bacterial infection transcriptionally upregulates TLR3 expression (Liu et al., 2007;Sajjan et al., 2006;Wang et al., 2009;Xing et al., 2011), thereby increasing viral induced cytokine and chemokine responses further. Stimulation of TLR2 or TLR3 also induces mucin expression by activating MAP kinases and inducing EGF receptor signaling (Chen et al., 2004;Kohri et al., 2002;Li et al., 1997;Zhu et al., 2009). MUC1, a transmembrane mucin is a negative regulator of TLRs and therefore may play an important role in limiting TLR-induced inflammatory responses (Ueno et al., 2008).
There are conflicting reports with regards to expression of TLRs and their role in innate immune responses in patients with COPD. Airway epithelial cells from patients with severe COPD showed decreased expression of TLR4, but not TLR2 (MacRedmond et al., 2007). In contrast, recently Pace et al observed increased neutrophils and decreased apoptosis of neutrophils in the bronchoalveolar lavage and increased expression of TLR4 in airway epithelium of COPD patients providing evidence that increased TLR4 may contribute to airway neutrophilia in COPD (Pace et al., 2011). Pace et al also demonstrated increased TLR4 expression and concurrent increased CXCL-8 in response to LPS challenge in cigarette smoke exposed airway epithelial cells (Pace et al., 2008), while other investigators showed decreased TLR4 expression which was associated with reduced CXCL-8 and hBD2 production (Kulkarni et al., 2010;MacRedmond et al., 2007). Our preliminary studies involving primary airway epithelial cells from COPD patients suggested heightened expression of CXCL-8 in responses to P. aeruginosa infection compared to normal airway epithelial cells (Ganesan and Sajjan, unpublished results). However, role of TLR in this context is yet to be established. Whether TLR4 expression is decreased or increased it has important implications in COPD airway inflammation and obstruction (Figure 2). The decreased expression of TLR4 may lead to decreased innate immune responses and increased persistence of infecting organism. On the other hand increased expression of TLR4 increases neutrophil recruitment and mucus production in response to bacterial or viral infection, thereby leading to increased airways inflammation and obstruction.

RIG-I like receptors
Another family of PRRs that play a role in innate defense mechanisms of airway epithelial cells is retinoic acid inducible (RIG)-I like receptors (RLR). This family of PRRs includes RIG-I, MDA-5 (melanoma differentiation associated protein 5) and LGP-2 (Laboratory of genetics and physiology 2). RLRs are the primary sensor molecules for detection of viral RNA in the cytoplasm (Meylan and Tschopp, 2006;Sun et al., 2006). Both RIG-I and MDA-5 contain a caspase recruitment domain (CARD) and a RNA helicase domain (Kang et al., 2002;Yoneyama et al., 2005;Yoneyama et al., 2004). On the other hand, LPG-2 has only RNA helicase domain but not CARD domain, which is required for recruiting adaptor protein MAVS (also known as VISA, Cardiff) . Therefore recognition of viral RNA by RIG-I and MDA-5 leads to IFN or chemokine response, and LPG-2 suppresses this response . RIG-I and MDA-5 recognize different RNA species. RIG-I recognizes single stranded (ss)RNA viruses, such as influenza virus, paramyxoviruses and deficiency in RIG-I increases the susceptibility of mice to RNA viruses . RIG-I specifically binds to the 5'-triphosphate moiety, the signature of which is exposed in the process of viral entry or replication. The host RNA which loses 5'triphosphate moiety during processing is therefore not recognized by RIG-I preventing cytokine and chemokine response due to self-recognition. RIG-I also recognizes short dsRNA (<1 kb) in 5'triphosphatewww.intechopen.com independent manner and induces IFN responses . On the other hand, MDA-5 recognizes long dsRNA that is >1 kb. Since viruses from picornaviridea family including rhinovirus generate long dsRNA in infected cells, innate immune responses to these viruses depends on recognition of viral RNA by MDA-5 (Kato et al., 2006;Wang et al., 2009). Mice deficient in MDA-5 show increased inflammatory response, delayed IFN response and significantly increased viral load up to 48 h after rhinovirus infection (Wang et al., 2011) . Both RIG-I and MDA-5 uses a common adaptor protein called interferon beta promoter stimulator-1 (IPS-1, also known as MAVS, VISA, CARDIF) (Kawai et al., 2005;Meylan et al., 2005;Seth et al., 2005;Xu et al., 2005). IPS-1 has a CARD domain which is homologous to RIG-I and MDA-5 and has a transmembrane domain at its C-terminal end that spans the mitochondrial membrane (Seth et al., 2005). IPS-1 after binding to RIG-I or MDA-5 through CARD-CARD interaction, activates IRF3 and NF-κB via TBK1/IKKε and RIP-1/IKK /IKK respectively. IPS-1 also interacts with receptor-interacting protein-1 (RIP-1), which is a death domain and is implicated in virus infection-induced IFN expression (Balachandran et al., 2004). However IPS-1 interaction with RIP-1 via the non-CARD region facilitates NF-κB activation, rather than IRF3 activation. Therefore IPS-1 regulates both IRF3 and NF-κB activation upon binding to RIG-I or MDA-5. IPS-1-deficient mice fail to activate IRF3 and NF-κB, with concomitant loss of type I IFN and inflammatory cytokine induction after viral infection and show increased persistence of virus . Recently, cigarette smoke extract was demonstrated to inhibit RIG-I-stimulated innate immune responses to influenza infection in bronchial organ culture model (Wu et al., 2011). Exposure to cigarette smoke extract also interfered with STAT1 activation by IFN-, a type II interferon which stimulates expression of various antiviral proteins (Modestou et al., 2010). Further, cigarette smoke also attenuated the inhibitor effect of IFN-on RSV mRNA and protein expression. Eddleston et al demonstrated that exposure of airway epithelial cells to cigarette smoke extract suppressed mRNA induction of CXCL-10 and IFN-by human rhinovirus and also viral dsRNA mimic polyinosinic:polycytidylic acid (poly I:C) (Eddleston et al., 2011). This was found to be due to decrease in activation of the IFN-STAT-1 and SAP-JNK pathways. Inhibition of antiviral responses, in particular IFN and CXCL-10 responses appear to be due to acute exposure to cigarette smoke that occurs in vitro, because the airway epithelial cells obtained from COPD patients showed antiviral responses to rhinovirus infection which was in fact significantly higher than the cells obtained from nonsmokers (Schneider et al., 2010). Similar to our observations, mice exposed to cigarette smoke and poly I:C or influenza virus showed increased IFN responses and this was attributed to pathogenesis of COPD (Kang et al., 2008).

NOD-like receptors
Nod-like receptors (NLR) are a family of proteins and sense microbial signatures in the cytosol. There are at least 22 identified NLRs in humans, although only few of them have been functionally characterized. All of them have a central nucleotide binding domain and C-terminal leucin-rich repeat domain, which possibly mediate ligand binding. In addition, they also contain different N-terminal effector domains such as CARD domain, pyrin domains or baculovirus inhibitor repeats and thus activate diverse downstream signaling pathways Fritz et al., 2006). The most widely studied among the CARD containing NLRs are NOD1 and NOD2. NOD1 primarily recognizes peptidoglycan (PGN) www.intechopen.com derivative, -D-glutamyl-mesodiaminopimelic acid from gram-negative bacteria (Chamaillard et al., 2003;Girardin et al., 2003a), whereas, NOD2 is considered as a general sensor of PGN through muramyl dipeptide (Girardin et al., 2003b). Upon recognizing PGN, both NOD1 and NOD2 activate NF-κB-mediated proinflammatory response via RIP-2 (Hasegawa et al., 2008). Both NOD1 and NOD2 are highly expressed in immune and inflammatory cells (Fritz et al., 2005;Kanneganti et al., 2007). These two NODs are also expressed in airway epithelium and are induced by bacterial stimuli (Bogefors et al., 2010;Mayer et al., 2007;Opitz et al., 2004;Travassos et al., 2005). NOD1 and NOD2 contribute to innate immune responses to different bacteria including Pseudomonas aeruginosa, Chlamydia pneumonia, Haemophilus influenza and L. pneumophila both in vivo and in vitro (Clarke et al., 2010;Frutuoso et al., 2010;Shimada et al., 2009;Zola et al., 2008).
NOD2 not only recognizes bacterial peptidoglycan, but also viral ssRNA. NOD2 deficiency results in impaired type I IFN expression in vitro upon stimulation with viral ssRNA (Sabbah et al., 2009). This was dependent on NOD2 interaction with IPS-1 and activation of IRF3, but not on activation of RIP-2. NOD2 deficient mice were also found to be more susceptible to infection with respiratory syncytial virus and influenza virus than the wild-type mice.
The widely expressed NLRX1 (NLR family member X1) is the only NLR receptor that is localized to mitochondria and it negatively regulates RIG-I and MDA-5 receptors. NLRX-1 mediates production of reactive oxygen species upon bacterial infection (Moore et al., 2008;Tattoli et al., 2008) and decreased dsRNA-stimulated IFN response.
Although, there is no evidence that NLRs play a role in innate immune responses to bacterial or viral infection in COPD so far, the emerging literature indicate inflammasome forming NLRs may contribute to COPD pathogenesis. Inhaled cigarette smoke, oxidative stress, necrotic cell death, hypoxia, hypercapnia may cause tissue injury and release of DAMPs (uric acid, ATP) and this in turn activates NLRP3 inflammasome (Wanderer, 2008). Consistent with this notion, uric acid concentration was increased in the bronchoalveolar lavage of COPD patients (Wanderer, 2008). COPD patients also had significantly increased amounts of IL-1 and this correlated with severity of the disease (Sapey et al., 2009). Mice exposed to cigarette smoke also showed increased IL-1 in their lungs (Doz et al., 2008) and finally mice overexpressing mature IL-1 in epithelial cells showed typical feature of COPD including emphysema, lung inflammation with increased neutrophils and macrophages and airway fibrosis (Lappalainen et al., 2005). ASC (inflammasome adaptor protein) null mice showed attenuated inflammation after exposing to elastase and less uric acid. Elastaseinduced inflammation was significantly reduced in wild-type mice treated with uricase or treated with IL-1R antagonist (Couillin et al., 2009). All these evidences suggest contribution of inflammasome forming NLRP3 to COPD pathogenesis.

Innate immunity and co-infections
Nontypeable H. influenzae (NTHi), S. Pneumoniae and P. aeruginosa are detectable in lower airways of appproximatley 25 to 50% of clinically stable COPD patients . Chronic colonization can alter the responses of airway epithelial cells and other innate and adaptive immune cells to subsequent viral or bacterial infections leading to increased severity of disease. Exacerbations due to concurrent or sequential infections was shown to be associated with increased severity of disease at least in one-quarter of COPD patient population (Papi et al., 2006;Sethi et al., 2006;Wilkinson et al., 2006). Risk of secondary bacterial infection following a viral infection dates back to 19 th century, when cases of pneumonia correlated with influenza (flu) epidemic (McCullers, 2006). Influenza infection increases risk of secondary bacterial infection by increasing binding or invasion of bacterial pathogen to airway epithelial cells, desensitizing innate immune receptors such as TLRs, and causing immunosuppression by increasing glucocorticosteriod expression (Beadling and Slifka, 2004;Hament et al., 1999;Jamieson et al., 2010;McCullers, 2006;Seki et al., 2004;Sun and Metzger, 2008). Respiratory syncytial virus infection increased persistence of P. aeruginosa in mice and increased P. aeruginosa and NTHi binding to airway epithelial cells (de Vrankrijker et al., 2009;Jiang et al., 1999;Van Ewijk et al., 2007). Respiratory syncytial virus also increased persistence of NTHi by dysregulating the expression of -defensin in chinchilla model of respiratory infection (McGillivary et al., 2009). Rhinovirus which causes common cold, in combination with S. pnuemoniae was associated with severe cases of community-acquired pneumonia in children (Honkinen et al., 2011). Various in vitro studies showed that rhinoviruses also increase bacterial binding to airway epithelial cells by increasing the expression of bacterial receptors on airway epithelial cells or by facilitating invasion of cells by bacteria (Ishizuka et al., 2003;Passariello et al., 2006). We demonstrated that rhinovirus infection also increases paracellular permeability and promote bacterial traversal across mucociliary-differentiated airway epithelium (Sajjan et al., 2008). Rhinovirus infection also decreases bacterial PAMPS-induced proinflammatory response by desensitizing TLRs (Oliver et al., 2008).  Impact of secondary viral or bacterial infection in patients colonized with bacteria is being increasingly recognized in recent years. For instance, despite chronic colonization with P. aeruginosa, cystic fibrosis patients show exacerbations periodically and some incidences are associated with acquiring secondary viral or bacterial infections (Ong et al., 1989;Ramsey et al., 1989;Wat et al., 2008). Similarly, in COPD patients who are chronically colonized with NTHi, exacerbations were associated with acquisition of new strain of NTHi, other species of bacteria or respiratory virus (Murphy, 2000;Murphy et al., 2008;Murphy et al., 2007;Papi et al., 2006;Sykes et al., 2007;Wilson, 2000). Recently, we showed that secondary bacterial infection in primary cystic fibrosis airway epithelial cells preinfected with P. aeruginosa increases C-X-C chemokine responses by increasing the load of planktonic bacteria which are more pro-inflammatory than their counterpart biofilm bacteria and also increased paracellular invasion of bacteria in differentiated airway epithelial cells (Chattoraj et al., 2011b). We also demonstrated that cystic fibrosis, but not normal airway epithelial cells infected with bacteria show suppressed type I IFN response to subsequent rhinovirus infection (Chattoraj et al., 2011a). This was due to increased oxidative stress in cystic fibrosis airway epithelial cells. Airway epithelial cells from COPD patients show increased oxidative stress similar to cystic fibrosis patients. Therefore we expect that bacterial preinfection may suppress innate immune responses to subsequent virus infection in COPD cells. Consistent with this notion, our preliminary studies indicate that infection with P. aeruginosa or NTHi infection increases oxidative stress further and decreases expression of antioxidant genes in COPD airway epithelial cells. In addition, we also observed suppression of IFN response in COPD airway epithelial cells infected with bacteria to subsequent rhinovirus infection (unpublished observations). Similar to our observations, LPS treatment was demonstrated to suppress IFN-production in response to dsRNA in mice as well as in monocytes and macrophages (Piao et al., 2009;Sly et al., 2009). This was due to increased expression of SHIP, a MPA kinase phosphatase in LPS treated monocytes. In airway epithelial cells however, P. aeruginosa infection induced suppression of IFN response to rhinovirus infection was not due to increased expression of SHIP, but rather due to decreased Akt phosphorylation (Chattoraj et al 2011) which is required for maximal activation of IRF3 (Dong et al., 2008;Sarkar et al., 2004). Previously, we have shown that expression of IFN response to rhinovirus infection requires activation of IRF3 in airway epithelial cells . Based on these experimental evidences, it is possible that 30% of COPD patients who are chronically colonized with NTHi or P. aeruginosa in their lower airways may show suppressed antiviral responses and increased chemokine expression (Figure 3). This may lead to increased lung inflammation and progression of lung disease in COPD patients following exacerbation due to co-infections.

Conclusion
The airway epithelium contributes significantly to innate immune system in the lungs. It acts as a physical barrier that protects against inhaled substances and pathogens. Airway epithelial cells also express plethora of innate immune receptors which recognizes both PAMPS and DAMPS and stimulate appropriate responses to either clear the infecting organism and to repair of injured epithelium. However in COPD, chronic exposure to cigarette smoke or environmental hazards causes airway remodeling and also modulate innate immune responses of airway epithelial cells to infection (Figure 4). This results in impaired clearance of infecting organisms and aberrant cytokine and growth factor expression and increased lung inflammation leading to progression of lung disease.

Acknowledgements
This work was supported by NIH AT4793 and HL897720 to U.S. We thank Marisa Lynn for assisting with processing cell cultures for histology and Adam Comstock for his assistance in culturing primary epithelial cells.

References
Adachi, O., Kawai, T., Takeda  Chronic Obstructive pulmonary disease (COPD) is an important cause of morbidity and mortality world-wide. The most common cause is chronic cigarette smoke inhalation which results in a chronic progressive debilitating lung disease with systemic involvement. COPD poses considerable challenges to health care resources, both in the chronic phase and as a result of acute exacerbations which can often require hospital admission. At the current time it is vital that scientific resources are channeled towards understanding the pathogenesis and natural history of the disease, to direct new treatment strategies for rigorous evaluation. This book encompasses some emerging concepts and new treatment modalities which hopefully will lead to better outcomes for this devastating disease.