S-nitrosylation of Surfactant Protein D as a modulator of pulmonary inflammation

Abstract

Background

Surfactant protein D (SP-D) is a member of the family of proteins termed collagen-like lectins “collectins” that play a role in non-antibody-mediated innate immune responses [1]. The primary function of SP-D is the modulation of host defense and inflammation [2].

Scope of Review

This review will discuss recent findings on the physiological importance of SP-D S-nitrosylation in biological systems and potential mechanisms that govern SP-D mediated signaling.

Major Conclusions

SP-D appears to have both pro- and anti-inflammatory signaling functions. SP-D multimerization is a critical feature of its function and plays an important role in efficient innate host defense. Under baseline conditions, SP-D forms a multimer in which the N-termini are hidden in the center and the C-termini are on the surface. This multimeric form of SP-D is limited in its ability to activate inflammation. However, NO can modify key cysteine residues in the hydrophobic tail domain of SP-D resulting in a dissociation of SP-D multimers into trimers, exposing the S-nitrosylated N-termini. The exposed S-nitrosylated tail domain binds to the calreticulin/CD91 receptor complex and initiates a pro-inflammatory response through phosphorylation of p38 and NF-κB activation [3,4]. In addition, the disassembled SP-D loses its ability to block TLR4, which also results in activation of NF-κB.

General Significance

Recent studies have highlighted the capability of NO to modify SP-D through S-nitrosylation, causing the activation of a pro-inflammatory role for SP-D [3]. This represents a novel mechanism both for the regulation of SP- D function and NO’s role in innate immunity, but also demonstrates the S-nitrosylation can control protein function by regulating quaternary structure.

1.1. S-nitrosylation as a regulator of lung inflammation through TLR4 signaling pathway

Toll-like receptors (TLRs) are family of pattern recognition receptors that represent the first line of defense against many pathogens. Remarkably, TLRs are capable of discriminating between pathogenic and commensal microorganisms, and inducing appropriate and distinct antimicrobial response [5]. As regulators of the inflammatory response against pathogens, TLRs help to strengthen the processes of innate and adaptive immunity [6].

In mammals, the recognition of lipopolysaccharide (LPS) requires at least three proteins: TLR4, CD14 and myeloid differentiation factor 2 (MD-2) [7] (Figure 1). LPS is opsonized by LPS-binding protein (LBP), and the complex is recognized by the opsonic receptor CD14 on the macrophage surface and forms the ternary complex LPS–LBP–CD14. Transfer of LPS from CD14 to MD-2 is tied to binding to TLR4 [8]. Lipid chains of LPS shift into the hydrophobic pocket of MD-2 to maximize hydrophobic contact. This structural shift allows the phosphate groups of LPS to form ionic interactions with a cluster of positively charged residues in TLR4 and MD-2 and this, in turn, results in dimerization of the two LPS-MD-2-TLR4 complexes arranged symmetrically [9]. TLR dimerization results in the recruitment of the adaptor protein, MyD88. The intracellular of TLR4 (TIR) binds to a homologous domain in MyD88 to constitute the so-called “death domain”. This death domain undergoes homophilic interaction with a similar domain of the serine/threonine protein kinase, IRAK, resulting in IRAK phosphorylation. Auto-phosphorylated IRAK forms a complex with TRAF6, which initiates oligomerization of TRAF6 [10]. This results in association with the heterotrimer of TAK1-TAB1-TAB2 to form a multi-component complex. Upon phosphorylation of TAK1 and TAB2, IRAK dissociated from the complex, and the TRAF6–TAK1–TAB1–TAB2 complex is translocated to the cytosol where it activates IKK [10]. IKK phosphorylates IkB, leading to its proteolytic degradation and the translocation of NF-κB to the nucleus with the consequent transcription of immune response genes [11].

Figure 1. The “Dandelion Ball Model” of the immunomodulatory functions of SP-D in macrophages.

Under baseline conditions the hydrophobic N-terminus of SP-D is hidden in the center of the “dandelion ball” with exposed carboxy terminal domains outside. Numerous studies have shown that CRD-specific interactions of SP-D with immune cells may be mediated through a number of receptors. It has been proposed that binding of SP-D “dandelion ball” via its CRD domains with SIRP-1α induces the activation of tyrosine phosphatase SHP-1, resulting in blockage of the downstream signaling through p38 MAP kinase [51]. The interaction of CRD domain of SP-D “dandelion ball” with sCD14, sMD2 and TLR4 may blocks the binding of LPS to these receptors leading to inhibition of NF-κB activation [44–46]. Potentially, this may be achieved by SP-D inhibiting TLR4 dimerization in the presence of LPS.

Under pro-inflammatory circumstances S-nitrosylation of cysteine residues in the hydrophobic tail domain of SP-D results in disruption of multimeric SP-D structure (disassembling of a “dandelion ball”) and exposure of S-nitrosylated tail domain [3]. The exposed S-nitrosylated tail domain is now available to bind with the calreticulin/CD91 receptor complex and initiate a pro-inflammatory response through phosphorylation of p38 and NF-κB activation. [3,4].

In addition, within this model one can see that the disassembled “dandelion ball” of SP-D would have lost its ability to block LPS mediated activation of NF-κB. In the absence of multimeric SP-D, or in its absence, LPS forms a CD14-MD2-TLR4 complex which triggers TLR4 dimerization and recruitment of the adaptor protein, MyD88. Dimerization results in IRAK phosphorylation and oligomerization of TRAF6 Assembled TRAF6 –TAK1–TAB1–TAB2 complex is translocated to the cytosol where it phosphorylates IκB through activation of IKK and therefore results in translocation of NF-κB to the nucleus with the consequent transcription of immune response genes [8–11,45].

There is considerable experimental evidence indicating that TLR signaling cascade contains a number of proteins, including MyD88, IKKβ and NF-κB, whose activity is regulated via S-nitrosylation [12]. It has been shown that S-nitrosylation of the p50–p65 heterodimer inhibits its binding to iNOS promoter DNA at the NF-κB–DNA interface [13,14]. S-nitrosylation of the IkB kinase β, a major regulator of NF-κB, results in reduction of its kinase function, which in turn leads to a reduction in IkB ubiquitinylation and subsequent degradation [15]. S-nitrosylation of the adaptor protein MyD88, located upstream from the IKK complex, results in inhibition of its binding to the sorting adaptor TIRAP, and therefore decreases LPS-induced NF-κB activation [16]. All these studies demonstrate that S-nitrosylation inhibits TLR signaling through modification of components of the intracellular signaling cascade. Therefore, one can see that the generation of intracellular SNO is an important mechanism in the control of acute-phase inflammatory responses.

1.2. SP-D as a target for S-nitrosylation

SP-D is a Ca²⁺-binding lectin that is produced primarily as a multimer by alveolar type II cells and nonciliated bronchiolar cells in the lung [17]. SP-D shares considerable structural homology with other proteins of this type, including surfactant protein A (SP-A), conglutinin, bovine collectin-43, and mannose binding protein [18,19]. SP-D through its C-type carbohydrate-recognition domain (CRD) binds to carbohydrate structures present on a range of viruses, bacteria, yeasts and fungi [20,21]. Upon recognition of the infectious agents, SP-D initiates direct opsonization, neutralization, agglutination and phagocytosis [22]. In addition, it can interact with receptor molecules present on immune cells leading to enhanced microbial clearance and modulation of inflammation [23,24]. Targeted ablation of the SP-D gene results in chronic inflammation and an emphysema-like phenotype in the absence of infection [25,26]. The lungs of SP-D (−/−) mice are characterized by an inappropriate activation of alveolar macrophages with significant alterations in nitric oxide (NO) metabolism [27–30], indicating an important role for SP-D in regulating immune homeostasis and the function of the innate immune cells. Furthermore, selective inhibition of iNOS was shown to decrease inflammatory markers [31] and the additional ablation of the iNOS-gene in SP-D (−/−) mice attenuated the degree of pulmonary emphysema, indicating an important role of iNOS in the pathogenesis of this model [32].

The SP-D primary translation product is composed of three polypeptide chains that are held together by disulphide bonds and further assembled into a large cruciform dodecamer [33] that resembles a “dandelion ball”. Critical to this oligomerization are two cysteine residues at positions 15 and 20 in the hydrophobic N-terminus [34,35] (Figure 2). Mutation of these residues results in the secretion of trimers, which represent a single arm of the dodecamer [34]. SP-D possesses seven cysteine residues [36] but only two of them (cys¹⁵/cys²⁰) exist in S-nitrosylation-consensus sequences [37]. Therefore, the positioning of cysteines 15 and 20, within the most hydrophobic region of the N-terminal domain of the SP-D protein suggests that NO, through a covalent modification of these thiol groups, may control oligomerization of SP-D by S-nitrosylation [3] (Figure 2).

Figure 2. Schematic structure of the SP-D.

(A) A Kyle Doolittle hydropathy plot for the SPD sequence was constructed using an averaging sliding window of 9 residues; a positive value that indicates a region of hydrophobicity. (B) The amino acid sequences of mature SP-D polypeptide chain consist of four structural domains: an N-terminal domain; a collagen domain; a α-helical neck region; and a globular CRD involved in Ca²⁺-dependent binding of various groups of ligands. The three polypeptide chains combine to form a trimeric structure that is assembled into a large symmetrical cruciform dodecamer consisting of a tetramer of trimers. The position of reduced cysteines 15 and 20 buried within the most hydrophobic region of the N-terminal domain, suggest it as a motif for S-nitrosylation.

The dandelion ball structure of multimeric SP-D is too large (>1 MDa) to have any significant mobility on native gel electrophoresis. However, a mutant form of SP-D, termed the single-arm form due to its inability to form the multimeric dandelion ball structure, is easily separated by native gel. The two critical cysteine residues of the tail-domain, Cys 15 & 20, are mutated to Serine in this single-arm form of SP-D [34]. Treatment of rat recombinant SP-D (rrSP-D) with S-nitrosocysteine or free reduced cysteine results in disruption of its multimeric organization as shown by native gel electrophoresis (Figure 3A). That both of these treatments effectively disrupt SP-D’s dandelion ball structure indicates that reduced cysteine interaction, rather than disulfide bond formation, is critical in maintaining this multimer. The reduced nature of these cysteines is confirmed by their ability to be alkylated [3]. However, rrSP-D, but not the single arm mutant, was effectively nitrosylated by S-nitrosocysteine as determined by the biotin-switch assay (Figure 3, B). Therefore, one can see that under lung injury and/or inflammatory conditions, increased NO production could lead to modification of these cysteine residues resulting in a dissociation of SP-D multimers into trimers (similar to exhaling on a “dandelion ball”), exposing S-nitrosylated tail domains.

Figure 3. Signaling in SP-D is NO-dependent.

NO can modify SP-D by nitrosylating two critical cysteines in its tail domain resulting in disruption of SP-D oligomeric structure and the release of S-nitrosothiol containing SP-D trimers (SNO–SP-D) that stimulate macrophage migration.

(A) Disruption multimeric structure of SP-D. Recombinant rat SP-D (rrSP-D) or the mutant (Ser^15/20) was incubated with 200 μM saline, SNOC (S-nitrosocysteine) or L-cysteine for 30 min at room temperature and then the multimeric structure of SP-D was analyzed by native gel electrophoresis followed by western blotting with anti-SP-D antibody. (B) S-nitrosylated SP-D was analyzed by biotin switch assay followed by SDS-PAGE and Western blotting with anti-SP-D antibody. (C) Chemotactic activity of modified rrSP-D or the mutant Ser^15/20 was analyzed using Boyden chamber with RAW 264.7 cells.

SP-D, a modulator of immune cell responses to pathogens, binds to LPS, which is the major cell-wall component of gram-negative bacteria [38–40]. As a result of this binding, pathogens can be aggregated and/or opsonized, and this leads, in many cases, to enhanced killing and clearance by phagocytic cells; thereby preventing uncontrolled inflammation in the lung. Critical evidence for the significance of SP-D mediated phagocytosis of inhaled pathogens was provided by studies on bacterial infection. Recently Ikegami et al. demonstrated that SP-D deficient mice were more susceptible to intratracheal LPS than WT mice and that intratracheal administration of recombinant SP-D inhibited LPS-mediated lung inflammation in both SP-D deficient and WT mice [41]. Inhaled LPS activates the Toll-like receptor 4 (TLR4) signaling pathway, resulting in increased production of inflammatory cytokines and reactive species such as NO [42,43]. In order to understand the role that SP-D plays within immune signaling it is necessary to examine the mechanisms involved in innate immune activation.

Recently the native multimeric form of SP-D has been demonstrated to bind to TLR4 [44] CD14 [45] and sMD-2 [46], via its CRD domain inhibiting TLR4-mediated pro-inflammatory responses caused by both smooth and rough serotypes of LPS [47] (Figure 1). Since MD-2 is critical for triggering LPS signaling [48], the binding of SP-D to MD-2 could prevent TLR4 dimerization/activation and, therefore, inhibit LPS-induced inflammatory cell responses. Experiments with trimeric cys¹⁵/cys²⁰ mutant [49], SP-D/MBL chimera [50], SPA/SP-D chimera, and a collagenase-resistant fragment [47] demonstrated that the oligomeric structure of SP-D is a critical feature of its immunomodulatory function. It is worth noting that both SP-A and SP-D have been proposed to interact via their CRD domains with the inflammation inhibitory receptor, SIRP-1α [51]. This would provide another immunomodulatory mechanism for SP-D by activating SHP-1 and thus inhibiting NF-κB activation. Under baseline conditions, the hydrophobic N-terminal tail of SP-D exists in a reduced state hidden in the center of the dandelion ball multimer with the CRD domains exposed on the surface [3,52]. Pathogen recognition and binding by the SP-D dandelion ball leads to basal phagocytosis and a regulated release of inflammatory mediators, maintaining lung homeostasis in an infection and inflammation free state. In this way the degree of basal inflammation observed within a lung is dependent on the quantity of LPS that is found ubiquitously in the environment.

S-nitrosylation of SP-D to form SNO-SP-D trimers initiates a pro-inflammatory response (Figure 1) through calreticulin/CD91 binding and p38 activation [3,53]. Importantly, SNO-SP-D, but not trimeric SP-D, is chemoattractive for macrophages and induces p38 MAPK phosphorylation. In this context it is important that while SNO-SPD initiates macrophage chemotaxis and thus has a functional consequence, trimeric SP-D, either cysteine-treated rrSP-D or the single arm mutant, is not chemoattractive; despite the fact that the tail region of these molecules is exposed (Figure 3C). Therefore, S-nitrosylation not only results in structural disruption of SP-D but is also critical to the signaling function within macrophages. The concept of both structural disruption and NO-mediated post-translational modification provides an explanation of the many prior conflicting studies reporting either pro- or anti-inflammatory effects of SP-D depending on the model system or stimulus used.

1.3. SNO-SP-D in animal models of pulmonary inflammation

The mechanism of macrophage activation through p38 phosphorylation and NF-κB activation by SNO-SP-D has been observed in a variety of animal models. Using both mouse and rat models of bleomycin-induced lung injury, it has been shown that macrophage driven pulmonary inflammation is associated with formation of SNO-SP-D. Lung lavage fluid (BAL) from bleomycin-injured mice is a potent chemoattractant for RAW cells, however, treatment with either anti-SP-D or ascorbic acid, which selectively reduces S-nitrosylated proteins to release NO, blocked macrophage chemotaxis in vitro [3]. In addition, it has been shown that BAL from bleomycin-treated mice increases p38 phosphorylation within RAW264.7 cells in a SP-D dependent manner. In another mouse model of lung injury using LPS, post-translational modification of SP-D has also been observed [54]. In this study aerosolized LPS induced increases in airway NO levels, airway neutrophil numbers, lung neutrophil and CD8⁺ cell numbers, and BAL SP-D protein levels. Furthermore, SP-D recovered from the BAL of LPS-treated mice was covalently cross-linked and S-nitrosylated with concomitant disruption of its multimeric structure. Disruption of SP-D oligomeric structure was associated with worsened lung compliance and cytokine/chemokine production (KC, TNFα, IL-6) [54].

In an infection based murine model of lung inflammation we have demonstrated that CD4⁺ T cell immune-reconstituted (IRD) mice infected with Pneumocystis pneumonia (Pc) exhibit impaired pulmonary function with substantial increases in the BAL SP-D level [4]. S-nitrosylation of SP-D and therefore alteration in SP-D higher order structure were markedly enhanced by immune-reconstitution during Pc infection. BAL fluid of Pneumocystis infected mice during IRD exhibit enhanced chemotaxis in a macrophage cell line in vitro. Removal of the SNO moiety from SNO-SP-D abolished the enhancement of chemotaxis. Doubly homozygous mice with mutations in both pale ear and pearl (EPPE) provides a model of Hermansky-Pudlak Syndrome (HPS), a rare form of oculocutaneous albinism associated with progressive lung inflammation [55]. EPPE mice exhibit a dramatic age dependent increase in BAL SP-D levels when compared with WT mice. Furthermore, EPPE mice demonstrated a shift in alveolar nitrite/nitrate ratio favoring the formation of higher oxides of nitrogen in association with the presence of S-nitrosylated and cross-linked forms of SP-D coincident with the onset of macrophage predominant inflammation [55].

All these studies provide general mechanistic evidence that SNO-SP-D plays a role in recruitment of effector cells in vivo. Thus, lung inflammation in a variety of model systems represents a feed forward system in which additional inflammation leads to further modification of SP-D by NO and subsequent pro-inflammatory effects mediated by SNO-SP-D. These studies emphasize the delicate balance that exists between SP-D, the innate immune system, and pulmonary inflammation. In summary, these in vivo models demonstrate that NO is capable of controlling the dichotomous nature of SP-D and that post-translational modification by S-nitrosylation causing quaternary structural alterations, may be the key in switching its inflammatory signaling role [3].

1.4. SP-D in human studies

Human studies have found conflicting results for SP-D levels and lung disease. Several studies have demonstrated decreased SP-D levels in the BAL of children with cystic fibrosis (CF) [56–58]. Decreased SP-D levels are found in the BAL of children with RSV infection [59] and in patients with ARDS [60,61]. Several studies have assessed the value of plasma SP-D levels as a marker for human lung diseases. Eisner et al. reported significant increases in plasma SP-D during ALI/ARDS [62]. Conversely, Determann et al. have shown that SP-D level in the plasma of ALI/ARDS patients was not significantly different when compared to patients without lung injury [63]. Attempts have been made to correlate serum and BAL SP-D levels with human disease, including community-acquired pneumonia in adults, recurrent bronchitis and asthma in pediatric patients, COPD, and smoking [64–65], as well as suggesting the use of low serum SP-D levels as a biomarker for the development of bronchiolitis obliterans in hematopoietic stem cell transplant recipients [66].

However, the structural composition of the SP-D molecule may vary between healthy conditions and inflammatory lung diseases. During the inflammatory state post-translational modification of SP-D can result in alteration of their structure and function [53]. As in animal models, within human disease the native structure of SP-D has been found to be altered through nitration, S-nitrosylation, oxidation and/or crosslinking [3,4,55,67–71] and therefore may confound the measurement of total SP-D in BAL or plasma by commercially available ELISA kits.

Testing of a range of antibodies (in house and from different sources) against SP-D showed that all recognized a single band of approximately 43–50 kD under reduced conditions but some antibodies also recognized lower molecular weight proteins (~25–30 kD). Antibodies against SP-D from different sources demonstrated variability in the species binding affinity. For example, an antibody that did not react with mouse native SP-D demonstrated strong reactivity with mouse denatured isoforms of SP-D and human native SP-D. In addition, there was variability in how well tested antibodies bound to modified SP-D either through oxidation and/or crosslinking (unpublished data). Additionally, variations within the SP-D gene also result in alteration of its multimeric structure [72–76] and therefore may also contribute to conflicting results by measuring total SP-D level by ELISA. Finally, acknowledging the fact that different ELISAs use differing antibodies raised against various portions of the SP-D molecule suggesting some complexity in distinguishing between various structural forms of SP-D. Therefore, all these observations demonstrate that SP-D antibodies (in house and commercially available) should be used carefully and critically. In conclusion, the precise quantitation of SP-D and its reliability as a biomarker of various pulmonary diseases requires comprehension of the pro- or anti-inflammatory structure/function of the SP-D molecule.

Recently Ware et al. reported that two plasma biomarkers, SP-D and IL-8 are significantly increased during ALI [77]. In other cross-sectional cohort study, Todd et al. also have shown that SP-D is increased in both BAL and plasma during ALI and that there was significant increase in SP-D breakdown products in the lungs of these patients [78]. The elevated BAL SP-D level was also associated with respiratory dysfunction, inflammation and increase in plasma SP-D and IL-8 levels during ALI [78]. However, an enhanced level of inflammatory markers, together with detection of significant amounts of SP-D breakdown products, during ALI raises the question about which of the pro- or anti-inflammatory forms of SP-D are elevated.

The ability to predict patients at risk for these outcomes and effectively modify their treatment may lead to improvement in patient care. Asthma, Hermansky Pudlak Syndrome type 1 (HPS1) and COPD are all associated with amplification of inflammatory signals in the distal lung. Recently, has been reported that BAL from HPS1 patients [55], asthmatic patients after segmental challenge with allergen [68], and patients with COPD [70], demonstrated the presence of SNO-SP-D and dissociation of SP-D multimeric structure. Furthermore, half of the asthmatic patients developed a covalently cross-linked SP-D after segmental challenge with allergen which correlated positively both with BAL eosinophil counts and with total levels of nitrate suggesting a strong relationship between cross-linked SP-D and severity of allergic inflammation [68]. Clearly, within human disease, SP-D is a target for post-translational modifications such as S-nitrosylation, oxidant-mediated cross-linking, and thus altered quaternary structure. The present data suggests that SP-D isoforms, in addition to contributing to local immunomodulation in the lung, are potential candidates as biomarkers for of pulmonary inflammation.

1.5. SP-D directly influences alveolar macrophage polarization – a mechanism to modulate inflammatory responses?

Alveolar macrophages (AM) are the most abundant antigen-presenting cells in the airways and alveolar spaces, where they play a critical role in regulating immune responses and inflammation within the lung [79,80]. AM perform a number of important functions, including phagocytosis of particulate matter, secretion of cytokines and enzymes, and control of microbes. They are the first cells to contact many inhaled antigens, including infectious agents, allergens and small particulate debris and therefore, play a key role in initiating, progression and/or resolution of local pulmonary immune responses.

Several studies have reported that alveolar macrophages recovered from the lung lavage of normal mice have a unique phenotype when compared to typical tissue macrophages [81–83]. The unique features of AM could reflect the unusual environment in which they are found. Thus macrophages have the ability to modify/reprogramme their secretory activity, and acquire different phenotypes depending on the tissue environment and the concentration of pathogenic product [84]. The development of these studies has led to the identification of multiple macrophage phenotypes. Often these phenotypes are categorized into two alternative profiles: M1 and M2 (Figure 1).

Classic M1 phenotype is formed by the interaction of macrophages with bacteria, viruses, stimulation of IFN-γ or LPS. M1 phenotype is characterized by increased production of pro-inflammatory cytokines (Th1) such as TNF-α, IL-1ß, IL-6, IL-12, macrophage inflammatory protein 1α (MIP-1α), as well as increased generation of nitric oxide (NO) and reactive oxygen intermediates [85–86]. These cells operate to kill microorganisms and tumor cells and produce pro-inflammatory cytokines [86,87]. The M1 phenotype exhibit enhanced antigen presentation, intracellular killing, and have increased levels of nitric oxide.

M2 macrophages, or “alternatively activated” macrophages are formed by the interaction with extracellular parasites or stimulation of IL-4, IL-13, TGF-ß, or glucocorticoids [88]. M2 is characterized by the production of anti-inflammatory cytokines (Th2) such as IL-4; IL-13 and IL-10. M2 phenotype regulates the inflammatory response, and is also involved in angiogenesis, remodeling and repair tissue damaged by inflammation.

Although it has been shown that there are tissue specific signals that cause monocytes to differentiate into their appropriate macrophages [89] the signals that control switching between M1, “classically” activated or pro-inflammatory, and M2 “alternatively” activated or anti-inflammatory are less well-defined. SP-D has emerged as a potential regulator of macrophage phenotype and function [30,90]. Both increased relative SNO-SP-D content [68] and SP-D deficiency [25] predisposes the lung to excessive pro-inflammatory responses while over expression of native multimeric SP-D is protective [91]. Mice that are constitutively deficient in SP-D develop progressive lung inflammation and time-dependent airspace remodeling [25] with increased expression of iNOS [29] and reduced expression of Arg1, PPARy, Dectin1 and FIZZ1, suggesting that within these animals the macrophages phenotype is M1 favored [92]. Over expression of SP-D is protective against the excess inflammation that occurs as a result of acute lung injury either via intratracheal bleomycin administration [91] or from hyperoxia [93]. In vitro, using primary murine macrophages, preliminary observations have shown that SNO-SP-D can induce expression of M1 related genes, such as IL-1α, NOS2, and Ptgs2, while native SP-D inhibits the expression of these genes in the presence of LPS [94].

In conclusion, one can see that SP-D, a critical regulator of innate immune function within the lung, is regulated by S-nitrosylation of two key target cysteines within the tail domain. SNO-SP-D is altered in its multimeric state and the “dandelion ball” structure of SP-D is lost, resulting in exposure of tail domains. This exposure allows for activation of the p38 MAPK pathway via interaction with CD91, however, it may also alter the interaction of SP-D with both pathogens and other surface receptors. In this review I have proposed that SP-D can alter the interaction between LPS and TLR4 and that this signaling pathway is dependent upon SP-D multimerization. Therefore, S-nitrosylation of SP-D will switch this regulatory collectin from a reparative signaling molecule to an acute inflammatory mediator by favoring M1 over M2 macrophage differentiation. This represents a novel mechanism of SNO signaling by controlling quaternary structure, and demonstrates a role for S-nitrosylation in the regulation of innate immunity.

Highlights.

• Multimerization of SP-D is a critical feature of its function

• NO is capable of modifying SP-D through S-nitrosylation resulting in a disruption of the SP-D quaternary structure

• The formation of SNO-SP-D initiates a pro-inflammatory response through NFkB activation

• Disruption of the native multimeric state of SP-D results in a loss of its anti-inflammatory capability

Abbreviations

TLR

Toll-like receptor

TIR

Toll/IL-1 receptor domain

LPS

lipopolyssacharides

LBP

LPS-binding protein

MD-2

myeloid differentiation factor 2

MyD88

myeloid differentiation protein 88

IRAK

IL-1 receptor associated kinase

TRAF6

TNF receptor associated factor

MAPK

mitogen-activated protein kinase

TAK1

transforming growth factor b-associated kinase 1

TAB1

TAK1-binding protein 1

TAB2

TAK1-binding protein 2

IKK

IκB kinase

IκB

inhibitor of NF-κB

NF-κB

nuclear factor kappa B

SIRP-1α

signal inhibitory regulatory protein -1 α

SHP-1

tyrosine-protein phosphatase-1

MBL

mannose-binding lectin

ARDS

Acute respiratory distress syndrome

ALI

Acute lung injury

COPD

Chronic obstructive pulmonary disease