Damage-associated molecular patterns and their receptors in upper airway pathologies

Abstract

Inflammation of the nasal (rhinitis) and sinus mucosa (sinusitis) are prevalent medical conditions of the upper airways that are concurrent in many patients; hence the terminology “rhinosinusitis”. The disease status is further defined to be “chronic” in case symptoms persist for more than 12 weeks without resolution. A diverse spectrum of external factors including viral and bacterial insults together with epithelial barrier malfunctions could be implicated in the chronicity of the inflammatory responses in chronic rhinosinusitis (CRS). However, despite massive research efforts in an attempt to unveil the pathophysiology, the exact reason for a lack of resolution still remains poorly understood. A novel set of molecules that could be implicated in sustaining the inflammatory reaction may be found within the host itself. Indeed, besides mediators of inflammation originating from outside, some endogenous intracellular and/or extracellular matrix (ECM) components from the host can be released into the extracellular space upon damage induced during the initial inflammatory reaction where they gain functions distinct from those during normal physiology. These “host-self” molecules are known to modulate inflammatory responses under pathological conditions, potentially preventing resolution and contributing to the development of chronic inflammation. These molecules are collectively classified as damage-associated molecular patterns (DAMPs). This review summarizes the current knowledge regarding DAMPs in upper airway pathologies, also covering those that were previously investigated for their intracellular and/or ECM functions often acting as an antimicrobial agent or implicated in tissue/cell homeostasis, and for which their function as a danger signaling molecule was not assessed. It is, however, of importance to assess these molecules again from a point of view as a DAMP in order to further unravel the pathogenesis of CRS.

Introduction

In analogy to pathogen-associated molecular patterns (PAMPs), some endogenous intracellular and/or extracellular matrix components from the host are collectively classified as damage-associated molecular patterns (DAMPs) as a consequence of the fact that they are able to modulate inflammatory responses under pathological conditions once released into the extracellular space, while under physiological conditions possessing intracellular and/or ECM functions often implicated in tissue/cell homeostasis or acting as an antimicrobial agent [1–5]. Intracellular molecules with an extracellular DAMP function are either passively released by damaged cells during the inflammatory and infectious process or are actively released by specific cell types via the so-called “alternative pathway of secretion” as they, by definition, lack a secretory leader sequence [5]. DAMPs derived from the extracellular matrix of the host can equally be released during damage or by regulated shedding mechanisms [4]. The list of molecules that can be classified as a DAMP is extensive (Table 1). They often execute their effect via activation of one or more Toll-like receptors (TLRs) or other members of the pattern recognition receptor (PRR) family including the receptor for advanced glycation end-products (RAGE) [4] and/or other receptor families such as the advanced glycation end-product-receptors AGE-R1–3 [6], the scavenger receptor CD36 [7], and purinergic P2Y_{1,2,4,6,11–14} and P2X1-7 receptors [8]. Moreover, several DAMPs are also known to bind ECM structures such as heparan sulfate proteoglycans [4, 5].

Table 1.

List of the known damage-associated molecular patterns (DAMPs)

	TLRs [19]							RAGE [21]	P2Y1,2,4,6,11–14 and P2X1-7 [8]	P2X7 [118]	CD36 [7]	AGE-R1–3 [6]
	TLR1	TLR2	TLR3	TLR4	TLR7	TLR8	TLR9
Proteins
S100A8, S100A9				Ѵ [52]				Ѵ [119, 120]			Ѵ [121]
S100A1–A2, S100A4–A7, S100A12–A13, S100B, S100P								Ѵ [21, 122]
HMGB1 (often in complex with other molecules; CpG-DNA, IL-1β, LPS…)		Ѵ [123]		Ѵ [123]			Ѵ [124, 125]	Ѵ [126, 127]
Heat shock proteins; HSP22, 60, 70, Gp96		Ѵ [128–131]		Ѵ [128–130, 132, 133]
Galectins		Ѵ [84]
Eosinophil-derived neurotoxin		Ѵ [70]
Fibronectin				Ѵ [77]
Tenascin-C				Ѵ [94, 95]							Ѵ [96]
AGEs (incl. N-(carboxymethyl)lysine, methylglyoxal…)								Ѵ [134]			Ѵ [135]	Ѵ [6]
Fibrinogen				Ѵ [136, 137]
Amyloid β peptide								Ѵ [138]
β-Defensins	Ѵ [110]	Ѵ [110]		Ѵ [109]
Annexin A2				Ѵ [139]
Thymosin α1							Ѵ [140, 141]
Surfactant protein A, D		Ѵ [59–61]		Ѵ [58, 59, 61, 62]
Antiphospholipid antibodies		Ѵ [142]		Ѵ [143]	Ѵ [144]	Ѵ [144]
Neutrophil elastase				Ѵ [145]
Lactoferrin				Ѵ [146]
Fatty acids, lipoproteins
Serum amyloid A		Ѵ [147]		Ѵ [148]				Ѵ [138, 149]		Ѵ [150, 151]	Ѵ [152]
Low density lipoproteins				Ѵ [153, 154]
Saturated fatty acids				Ѵ [155]
Proteoglycans, glycosaminoglycans (ECM components)
Hyaluronan		Ѵ [156]		Ѵ [157]
Heparan sulfate proteoglycans				Ѵ [158]				Ѵ [159]
Biglycan		Ѵ [160, 161]		Ѵ [160, 161]					Ѵ [160]
Versican		Ѵ [162]
Nucleic acids
mRNA			Ѵ [163]
Small nuclear RNA					Ѵ [164]	Ѵ [164]
IgG–chromatin complexes							Ѵ [165]
Nucleotides
Extracellular purine and pyrimidine nucleotides									Ѵ [166, 167]	Ѵ [166, 167]

Ѵ indicates that the respective DAMP acts as a ligand on the corresponding receptor(s)

Chronic rhinosinusitis (CRS) without nasal polyps (NP) (CRSsNP) and with nasal polyps (CRSwNP) are chronic airway diseases characterized by persistent inflammation of the nasal and sinus mucosa and displaying a high comorbidity with asthma [9, 10]. In Caucasians, CRSwNP is associated with a T-helper-2 (Th2) skewed inflammation with increased interleukin (IL)-5 and high eosinophil cationic protein (ECP)/myeloperoxidase (MPO) ratios in the polyps, while CRSsNP is characterized by a Th1 milieu with increased levels of interferon (IFN)-γ in the inflamed ethmoidal mucosa and low ECP/MPO ratios (Table 2) [11, 12]. CRSwNP is accompanied by reduced levels of transforming growth factor (TGF)-β1, a critical factor involved in the remodeling process, while CRSsNP shows increased TGF-β expression [13]. The findings regarding TGF-β levels are in agreement with the remodeling patterns observed, being reflected by edema formation and a lack of collagen production in CRSwNP and excessive collagen deposition in the extracellular matrix (ECM) associated with fibrosis in CRSsNP (Table 2) [14–16]. The inflammatory process is furthermore subject of interference by external inciters as evidenced by an increased colonization rate of Staphylococcus aureus (S. aureus) in human nasal polyps [17]. S. aureus-derived enterotoxin B (SEB) indeed further shifts the cytokine pattern in nasal polyps toward Th2 cytokines, and disfavors the T-regulatory cytokines [18].

Table 2.

Overview of the different characteristics of chronic rhinosinusitis (CRS) without nasal polyps (CRSsNP) and CRS with nasal polyps (CRSwNP) with regard to remodeling and inflammatory parameters, the expression of pattern recognition receptors, and alterations in the levels of damage-associated molecular patterns

CRSsNP	CRSwNP
Remodeling pattern	Remodeling pattern
Fibrosis [15]	Edema [15]
TGFβ ↑ [14]	TGFβ ↓ [14]
MMP9 ↑ [34]	MMP7 ↑, MMP9 ↑, TIMP1↑ [34]
Inflammatory pattern [74]	Inflammatory pattern [74]
Th1	Th2
Predominantly neutrophilic	Predominantly eosinophilic
Low ECP/MPO ratio, IFN-γ ↑, IL1β ↑	High ECP/MPO ratio, eotaxin ↑, IL5 ↑
	Total IgE ↑, SAE-IgE ↑
	Increased colonization rate of S. aureus
Pattern recognition receptors	Pattern recognition receptors
RAGE	RAGE
mRAGE ↓, sRAGE ↑ in tissue [22]	mRAGE ↓, sRAGE ↓ in tissue [22]
TLRs	TLRs
TLR1–10 expressed [47]	TLR1–10 expressed [40] with ambiguous data on TLR5 and 9 expression [43] and no functional response by TLR7, 8 and 9 [168]
TLR2 and 4: no different expression levels in CRSsNP vs. non-diseased status [42]; no data available for other TLRs vs. non-diseased status	TLR1–10: no different expression levels vs. non-diseased status [40], except for equivocal reports on increased TLR2 levels [40, 42] and reduced TLR9 levels [44] in CRSwNP
Molecules with DAMP function	Molecules with DAMP function
S100 proteins	S100 proteins
S100A7 ↑↑ in total homogenates, ↓ in nasal secretions, epithelium and glandular tissue [48]	S100A7 +/− ↑ in total homogenates, ↓ in nasal secretions, epithelium and glandular tissue [48]
S100A8/A9 ≈ in nasal secretions, ↓ in epithelium and glandular tissue [48]	S100A8/A9 ↑ in total homogenates, ↓in nasal secretions, epithelium and glandular tissue [48]
Galectins	Galectins
N.A.	Galectin-1 ↑ in tissue [87]
	Galectin-3 ↑ in tissue [88]
	Galectin-10 ↑ increased in nasal lavage fluid and tissue (explicitly pronounced in CRSwNP patients with comorbid aspirin-sensitive respiratory disease) [90]
Eosinophil-derived neurotoxin	Eosinophil-derived neurotoxin
↑ in nasal secretions (recurrent rhinosinusitis) [71]	N.A.
Fibronectin	Fibronectin
N.A.	↑ in asthmatics with nasal polyps compared with asthmatics without polyps.[78]
Tenascin-C	Tenascin-C
N.A.	↑ in nasal polyp tissue [97, 98]
Surfactant protein A, D	Surfactant protein A, D
↑ in tissue [63]	↑ in tissue [64, 65]
β-Defensin 2 and 3	β-Defensin 2 and 3
≈ in tissue vs. controls [111]	↓ in tissue [112]
↑ in tissue by TLR3 agonist but not in controls [111]	↓↓ in tissue with strong Th2 environment [112]

+/− ↑, moderately increased levels; ↑, increased levels; ↑↑, very pronounce increased levels; ↓, reduced levels; ↓↓, very pronounce decreased levels; ≈, similar levels; N.A. no data available

However, many unknowns remain regarding the pathophysiology of CRS including the exact cellular and tissue-specific structural patterns implicated in the immunological responses [10]. DAMPs potentially represent a novel class of molecules that might contribute to the persistence of the inflammatory response by synergizing with external insulting agents or by maintaining the inflammatory response even after external agents are eliminated. Therefore, in the next chapters we describe what is known regarding the DAMP-recognizing PRR family members RAGE and TLRs in CRS, followed by the different molecules that can act as a DAMP but for which this function as such has not been investigated or recognized yet in the upper airways. Because several of these molecules were previously assessed for their intracellular and/or ECM functions often acting as an antimicrobial agent or implicated in tissue/cell homeostasis, we describe in this review those molecules for which a deregulation has been reported in CRS with regard to their housekeeping function under normal physiology, and concurrently speculate on their potential second nature as a DAMP during chronic airway conditions in relationship to the expression status of the respective receptors that signal their presence.

Pattern recognition receptors in chronic rhinosinusitis

The sensing of conserved motifs derived from exogenous pathogens (PAMPs) infecting the host’s tissue is mediated by PRRs expressed on innate immune cells such as macrophages and dendritic cells, and on non-immune cells such as fibroblasts, epithelial cells, and endothelial cells. Four major classes of PRRs have been identified including TLRs, RIG-I-like receptors (RLRs), NOD-like receptors (NLRs), and C-type lectin receptors (CLRs) [19]. From these PRR members, the TLRs are able to equally recognize molecules released from cells of the host (DAMPs) and thereby signaling tissue damage and/or modulating inflammatory responses. Next to the traditional PRRs mentioned above, the prototypic DAMP receptor RAGE is also considered a PRR [20], as it recognizes common features rather than a specific ligand [21].

In this section, we will discuss what is known regarding RAGE and TLRs in the upper airways because adequate knowledge of these PRRs is a prerequisite in order to be able to fully recognize the possible signaling functions of the different molecules with DAMP function that will be discussed in the next section of this review.

Receptor for advanced glycation end products (RAGE)

Advanced glycation end-products (AGEs), being proteins and lipids that are modified by non-enzymatic glycoxidation and which are particularly of importance in age-associated and diabetic-related complications, are sensed by the AGE-binding receptors (AGE-Rs) AGE-R1 (oligosaccharyl transferase-48), AGE-R2 (80K–H phosphoprotein), and AGE-R3 (better known as galactin-3; see further) [6], but also other receptors including CD36 and RAGE serve in signaling AGE-mediated responses. With regard to the upper airways, of all known DAMPs and their specific receptors, up to now only the multi-ligand receptor RAGE has been assessed in view of the DAMP story [22].

Besides AGEs, RAGE binds to multiple other ligands such as the beta-integrin Mac-1 and the DAMPs high-mobility group box 1 (HMGB1), members of the S100/calgranulin family and amyloid β peptides [23] (Table 1). RAGE is a cell surface protein that belongs to the immunoglobulin superfamily, which is expressed as a full-length, membrane-bound receptor (mRAGE) but can also exist in two soluble forms lacking the transmembrane and cytoplasmic domains collectively termed sRAGE. sRAGE derived via alternative mRNA splicing is known as endogenous secretory RAGE, or esRAGE, whereas sRAGE derived from proteolytic cleavage of mRAGE is referred to as cRAGE; the latter being mediated by metalloproteinases (MMPs) and “a disintegrin and metalloproteinases” (ADAMs) [24–27]. sRAGE is often regarded to act as a decoy receptor with anti-inflammatory properties by scavenging away RAGE ligands from the cell surface mRAGE receptor that transduces proinflammatory responses via transcription factor NF-κB [20], but sRAGE is also described to be implicated in the development of inflammation [28, 29] while anti-inflammatory properties have been suggested for mRAGE as well [30].

These diverse effects of RAGE indicate that the tissue type and/or the overall inflammatory environment affects its final outcome towards a pro- or anti-inflammatory effect. Indeed, while RAGE expression is on most cell types and tissues undetectable (except under pathological conditions [20]), it is known that the lung forms a remarkable exception with high expression levels of RAGE during normal physiology [30–33]. Similar as for the lung, RAGE is also clearly expressed in the human upper airways under normal physiology where it is subject to differential regulation during chronic inflammatory conditions; in CRSsNP sRAGE protein levels are increased while mRAGE levels are reduced compared with controls. On the other hand, in CRSwNP, both tissue sRAGE and mRAGE protein levels are reduced (Table 2) [22]. The reduced levels of mRAGE protein in sinus mucosa from CRSsNP and CRSwNP patients are associated with an increased proteolytic cleavage of the full-length mRAGE protein which concurrently augments the levels of cleaved sRAGE. An imbalance between MMPs and their natural “tissue inhibitors of metalloproteinases” (TIMPs) was indeed previously reported to occur in CRSsNP and CRSwNP [16, 34], potentially yielding conditions of increased proteolytic cleavage (Table 2).

The lack of collagen in CRSwNP and the excessive collagen production with thickening of the collagen fibers in the ECM of CRSsNP [14, 16] might allow sRAGE to be retained in higher levels in the fibrotic tissue of CRSsNP patients than in the pseudocystic tissue of CRSwNP patients as sRAGE indeed binds to these structures [32, 35, 36]. Moreover, the increased colonization rate of S. aureus in the upper airways of CRSwNP patients [17] is a potential contributor to the overall reduction in sRAGE in this patient group as S. aureus induced the release of sRAGE from the ECM [22].

In CRSsNP, the increased levels of ECM-associated sRAGE likely contribute to the induction of a Th1-biased inflammation [22]. Indeed, RAGE is reported to be involved in the differentiation of T cells towards a Th1 phenotype, and RAGE deficiency decreases production of Th1-cytokines while producing more IL-4 and IL-5 as Th2 cytokines [37]. In CRSwNP, high ECP protein levels can break non-ECM-associated sRAGE down, preventing a redirection of T helper cells into the Th1 type [22].

Toll-like receptors

To date, ten Toll-like receptors (TLRs) have been identified in humans. Each TLR subtype is known to detect specific PAMPs derived from viruses, bacteria, fungi, and/or parasites. These include lipoproteins (recognized by TLR1, TLR2, and TLR6), double-stranded RNA (TLR3), lipopolysaccharide (TLR4), flagellin (TLR5), single-stranded RNA (TLR7 and TLR8), and DNA (TLR9) [38]. Among the human TLRs, TLR10 is the only family member without a defined signaling function [39]. For an overview regarding the TLR subtypes recognizing the different DAMP molecules that will be discussed in the next chapter of this review, we refer to Table 1.

In general, in human upper airway mucosa, most TLR subtypes seem to be expressed with a few equivocal findings in the literature (Table 2). Lane et al. [40] reported all TLR subtypes to be expressed in both healthy controls and CRSwNP patients. The authors observed no differences in the expression levels of the TLR subtypes between both subject groups except for a small increase of TLR2 mRNA levels in CRSwNP tissue [40], which could originate from increased influx of mast cells that were reported to express TLR2 in nasal polyps [41]. However, this increase in TLR2 levels was not in agreement with the findings of Claeys et al. [42]. While Pitzurra et al. [43] reported an inconsistent pattern of TLR expression in healthy controls, they confirmed the expression pattern found by Lane et al. in CRSwNP for most TLRs except for TLR5 and TLR9, which were absent in nasal polyp tissue. Reduced TLR9 expression was also reported in epithelial cells of CRSwNP patients [44]. In fibroblast of CRSwNP patients, no functional effect was observed by stimulation with TLR7/8 and TLR9 ligands, while TLR2, TLR3, TLR4, and TLR5 ligation yielded increases in downstream mediator production [45]. The non-effect of TLR7/8 ligands in nasal polyp fibroblasts might be explained by the absence of expression of these receptors in human nasal fibroblasts as TLR7 and TLR8 were previously reported not to be expressed in fibroblasts from non-polyposis patients with chronic sinusitis or allergic rhinitis [46]. On the other hand, the absence in effect of TLR9 ligands on nasal polyp fibroblasts [45] rather confirms the above-mentioned down-regulation in TLR9 expression as this receptor was clearly expressed in fibroblasts from non-polyposis patients [46].

Also in airway mucosa from CRSsNP patients, all TLR subtypes were reported to be expressed (Table 2) [47]. TLR2 and TLR4 were reported not to be differentially expressed between CRSsNP and control status [42]; no data are available for other TLR expression levels in CRSsNP vs. non-diseased status.

Damage-associated molecular patterns in chronic rhinosinusitis

Of the molecules that can be classified as a DAMP, several were previously investigated in the upper airways without the intention or knowledge to assess their function as a danger signal (Table 3). These molecules were, however, often the topic of previous research for their intracellular and/or ECM functions either acting as an antimicrobial agent or being implicated in tissue/cell homeostasis. It is nevertheless of importance to equally investigate these molecules from a point of view as a DAMP in order to further unravel the pathogenesis of CRS. In the next chapters, we give an overview of what is known regarding several of these molecules with potential DAMP function in upper airway pathologies such as S100 proteins, surfactant proteins, eosinophil-derived neurotoxin, fibronectin, galectins, tenascin-C, and β-defensins.

Table 3.

Overview of what is known regarding altered expression levels of molecules that can act as a damage-associated molecular pattern (DAMP) which were previously investigated for their intracellular and/or ECM functions in different upper airway pathologies without the intention or knowledge to assess their function as a danger signal. The receptors to which these molecules, when acting as a DAMP, ligate to are indicated behind the respective molecule

DAMPs	Receptors
S100A8/A9	TLR4, CD36, RAGE [52, 119–121]
Reduced in epithelium and glandular structures of nasal polyp tissue from CRSwNP patients [48]
Increased in total homogenates as a consequence of increased number of neutrophils containing S100A8/A9 [48]
S100A1–A2, S100A4–A7, S100A12–A13, S100B, S100P	RAGE [21, 122]
Reduced S100A7 levels in epithelium and glandular structures of nasal polyps tissue from CRSwNP [48] and nasal lavage fluid from CRSwNP patients [48] and allergic rhinitis patients [55]
Surfactant protein A, D (SP-A and SP-D)	TLR2 and TLR4 [58–62]
Increased in CRSsNP [63] and CRSwNP [64, 65] tissue
Compared to CRSsNP and CRSwNP, much more pronounced increase in SP-A and SP-D levels in sinus mucosal biopsies of patients with cystic fibrosis [65, 66]
Increased by fungal allergens in an in vitro model of human nasal explants [67]
Eosinophil-derived neurotoxin	TLR2 [70]
Elevated in patients with perennial allergic rhinitis and recurrent sinusitis [71, 72]
Fibronectin	TLR4 [77]
Induced sputum fibronectin levels are higher in asthmatics with nasal polyps compared with asthmatics without polyps [78]
Fibronectin accumulation correlates with the size of nasal polyps [79]
Galectins	TLR2 [83, 84]
Galectin-1 [87] and galectin-3 [88] increased in CRSwNP tissue
Galectin-3 increased in nasal lavage fluid of symptomatic allergic rhinitis patients [55]
Galectin-10 increased in nasal lavage fluid and nasal polyp tissue of CRSwNP patients and even more pronounced in CRSwNP patients with comorbid aspirin-sensitive respiratory disease [90]
Tenascin-C	TLR4 and CD36 [94–96]
Up-regulated in CRSwNP tissue [97, 98]
β-Defensins 2 and 3	TLR1, TLR2 and TLR4 [109, 110]
Increased in upper airways of cystic fibrosis patients [116]
Reduced in the lower airways of cystic fibrosis patients [169]
Reduced in a Th2-skewed inflammation with low TLR9 levels such as in CRSwNP [112]
Negligible or unchanged levels in CRSsNP [42, 111]

S100 proteins

The calcium-binding proteins S100A7 (psoriasin), S100A8 (MRP8 or calgranulin A), S100A9 (MRP14 or calgranulin B), and the latter’s heterodimeric form S100A8/A9 (calprotectin) were investigated most into detail in patients with CRSsNP and CRSwNP for their antimicrobial effects and their possible implication in deficiencies in the barrier function of airway mucosal epithelium in these patients [48]. Besides a direct antimicrobial effect [48–51], S100 proteins also show the typical parameters of DAMPs when released to the extracellular milieu where they are reported to activate TLR4 [52, 53] and are also (but not unambiguously) suggested to activate RAGE (Table 1) [54].

Expression of S100A7, S100A8, and S100A9 mRNA is decreased in CRSsNP and CRSwNP when compared with controls [49]. On protein level, S100A7 proteins are reduced in the epithelium and glandular structures of polyp tissue from CRSwNP patients as observed on immunohistochemical stainings, which was furthermore reflected in diminished levels of S100A7 in nasal lavage fluid of these patients [48]. S100 protein levels were also reported to be reduced in the nasal lavage fluid of allergic rhinitis patients compared to controls [55]. Similar diminished levels were observed for S100A8/A9 proteins but only in CRSwNP patients. However, despite the reduced levels in the epithelium and glandular structures, S100A8/A9 proteins are dramatically increased in total tissue homogenates of polyp tissue as a consequence of increased infiltration of neutrophils, which are a major source of S100A8/A9 proteins (Tables 2, 3) [48]. Taken together, these findings suggest a role for secreted S100A7 and S100A8/A9 from epithelial source as an antimicrobial substance in the sinonasal cavity and suggest that reduced levels can be considered as a possible contributing factor in CRS pathogenesis as a consequence of a reduced epithelial barrier defense mechanism. Moreover, the increased number of S100A8/A9-containing neutrophils in CRS tissue can release free S100 proteins which will subsequently target TLR4, CD36 and/or RAGE and consequently further modulate the inflammatory response during upper airway pathologies.

Surfactant proteins A and D

Surfactant proteins A (SP-A) and D (SP-D) are hydrophilic proteins that belong to the collectin family of innate immunity proteins and are secreted at epithelial surfaces. They exhibit antimicrobial properties by interacting with a wide range of pathogens including Staphylococcus through binding PAMPs located on microbial membranes via their calcium-dependent carbohydrate binding domains [56, 57]. Furthermore, independent of direct binding to various pathogens, SP-A and SP-D also deal with pathogens via other innate immune mechanisms such as the modulation and/or direct binding of TLR2 and TLR4 (Table 1) [58–62]; therefore also regarded as a DAMP [4].

Without assessing their potential function as a DAMP, surfactant proteins SP-A and/or SP-D have been reported to be clearly increased in the upper airways of CRSsNP [63] and CRSwNP [64, 65] patients (Tables 2, 3). In comparison to CRSsNP and CRSwNP, the increase in the levels of SP-A and SP-D were found to be much more pronounced in sinus mucosal biopsies of patients with cystic fibrosis (Table 3) [65, 66]. Similarly, SP-D levels were also increased by fungal allergen treatment in an in vitro model of human nasal explants [67], and in the upper and lower airway mucosa of influenza virus infected mice (Table 3) [68].

Eosinophil-derived neurotoxin

Eosinophils contain in their cytoplasmic granules toxic “basic granule proteins” such as ECP, eosinophil peroxidase (EPO), major basic protein (MBP), and eosinophil-derived neurotoxin (EDN). Of these, EDN can be categorized as a DAMP because besides its ribonuclease activity and neurotoxic effect [69], EDN can function as an endogenous ligand for TLR2 (Table 1) [70].

Levels of EDN were significantly elevated in patients with perennial allergic rhinitis and recurrent sinusitis compared with controls (Tables 2, 3) [71]. In allergic subjects with a clinical late response to nasal challenge with antigen, significant increases in EDN were reported in nasal-lavage fluid during the late response [72], which could be reduced by systemic [72] and topical corticosteroids (Table 3) [73]. To our knowledge, no reports are available on levels of EDN in CRSwNP. However, because CRSwNP is characterized as a typical T-helper-2 (Th2) skewed eosinophilic inflammation with high ECP concentrations in the polyps (Table 2) [74], it can be speculated that EDN might also play a role as a DAMP in CRSwNP.

Fibronectin

Fibronectin is a high molecular weight glycoprotein of the extracellular matrix that binds to cellular integrins and other extracellular matrix components such as fibrin, collagen, and heparan sulfate proteoglycans [75]. It plays a major role in wound healing, cell adhesion, matrix assembly, cell differentiation, cell cycle progression, and mitogenic signal transduction [76]. However, fibronectin is also implicated in immune processes via the activation of TLR4 though its extra domain A, hence its function as a DAMP (Table 1) [77].

Induced sputum fibronectin levels are higher in asthmatics with nasal polyps compared with asthmatics without nasal polyps (Tables 2, 3) [78]. It was furthermore demonstrated that fibronectin accumulation correlated with the size of nasal polyps (Table 3) [79].

Fibronectin is also recognized to be the target for a large number of bacterial proteins named bacterial fibronectin-binding proteins (FnBPs), for which those of S. aureus have been investigated most intensively [80]. The binding of FnBPs to fibronectin mediates not only the adherence of S. aureus to extracellular matrices but also to the surface of a number of host cell types, including endothelial cells, epithelial cells, and fibroblasts, which allows S. aureus to access the cytoplasm of these cells [80, 81]. In the intracellular space, S. aureus is protected from antibiotics and the host’s immune response [80, 81], which might be of relevance in view of the increased colonization rate of S. aureus in the upper airways of CRSwNP patients (Table 2) [17].

Galectins

Galectins are carbohydrate-binding animal lectins comprising at least 12 members in humans of which galectin-1 and galectin-3 have been investigated most extensively [82]. Although galectins lack a leader sequence, they can be released outside the cell; hence having functions both intracellularly and extracellularly [82, 83], making them a member of the DAMP family [1]. Intracellular galectins alter biological responses such as apoptosis, cell differentiation, motility, and growth, while extracellular galectins induce various cellular responses including production of cytokines and other inflammatory mediators, cell adhesion, migration, and apoptosis. For example, galectin-1 has been shown to induce apoptosis in both thymic and activated T cells, possesses anti-inflammatory activities by blocking or attenuating signaling events that lead to leukocyte infiltration, migration, and recruitment [82, 83]. Galectin-3 has a dual role in both pro- and anti-apoptotic pathways; cytosolic galectin-3 protects cells from apoptosis and promotes T cell proliferation while extracellular galectin-3 can induce apoptosis in activated T cells [82]. It furthermore shows pro-inflammatory activity, enhances macrophage survival, positively modulates macrophage recruitment and antimicrobial activity [83], and is reported to associate with TLR2 (Table 1) [84]. Besides its presence intracellularly and as a free extracellular protein, galectin-3 has also been found on the surface where it is reported to bind AGE-ligands, hence its alternative name AGE-R3 [85]. The lack of a transmembrane anchor sequence suggests that AGE-R3 is associated with other membrane structures, possibly AGE-R1 and AGE-R2, to form an AGE-receptor complex where it plays a role in the regulation of the turnover of AGE, and thus in the maintenance of tissue integrity as was reported in diabetic nephropathy [86].

Galectin-1 [87] and galectin-3 [88] expression is higher in nasal polyps compared to normal nasal mucosa (Tables 2, 3). Galectin-3 is furthermore reported to be upregulated in the nasal lavage fluid of patients with symptomatic birch and/or grass pollen-induced intermittent allergic rhinitis versus healthy volunteers (Table 3) [55]. The latter was confirmed in a mouse model of allergic rhinitis [89]. Galectin-10 is also reported to be increased in nasal lavage fluid and nasal polyp tissue of patients with aspirin-sensitive respiratory disease (Tables 2, 3) [90].

Tenascin-C

Tenascin-C is an ECM glycoprotein that is associated with tissue injury and repair. It is specifically and transiently upregulated during acute inflammation and persistently expressed in chronic inflammation [91, 92]. Under these conditions, tenascins have the potential to modify cell adhesion either directly or through interaction with fibronectin. Consequently, cell–tenascin interactions can affect subsequent cell motility and proliferation [93]. Besides cell adhesion, migration, and growth, tenascin-C induces synthesis of proinflammatory cytokines via activation of TLR4 [94], and stimulates foam cell formation from macrophages via TLR4 and the scavenger receptor CD36 (Table 1) [95]. Amplification of the inflammatory response by tenascin-C may thus lead to persistence of inflammation, in which damage created by inflammation leads to the induction of more endogenous activators of inflammation [94, 96].

Tenascin-C protein expression was significantly upregulated in nasal polyp tissue from CRSwNP patients compared with normal control inferior turbinate tissues (Tables 2, 3) [97]. Although the authors found the increased tenascin-C levels to be related to eosinophil-derived TGF-β1, others reported tenascin-C to be equally up-regulated in noneosinophilic chronic sinusitis with nasal polyps compared with ethmoid or sphenoid sinuses of controls [98], excluding eosinophils as a contributor in this process. Tenascin-C is indeed known to be upregulated by TGF-β1. However, because CRSwNP is characterized by an intense edema with low levels of TGF-β1, the latter seems unlikely to be implicated in the increase of tenascin-C in CRSwNP patients. Induction of tenascin-C was also reported for TGF-β2 [99], which was significantly higher in CRSwNP compared to controls [14]. Besides TGF-β, tenascin-C is also upregulated by fibroblast growth factor (FGF)1, FGF2, and FGF4 [99, 100]. Primary nasal polyp tissue-derived fibroblasts actively proliferate in vitro in response to FGF2 [101], which could also be clearly measured in nasal polyp tissues [102] and being increased versus control conditions [103]. The same applies to FGF1 [104]. Finally, as tenascin-C is also reported to be induced upon infection with S. aureus [105], the higher levels of tenascin-C reported in nasal polyp tissue might also be related to the increased colonization rate of S. aureus in CRSwNP patients (Table 2) [17].

β-Defensins

β-Defensins are small antimicrobial peptides produced in response to microbial infection of mucosal tissues and skin [106, 107]. While the human genome encompasses almost 40 potential coding regions for β-defensins, to date only the first four human β-defensin proteins have been characterized in detail [108]. The expression pattern of human β-defensins 1–4 shows features that are unique as well as being widely overlapping. One common pattern can be found in mucosal tissues where all four best characterized β-defensins are constitutively synthesized by epithelial cells being in direct contact with the environment such as is the case for the lungs. Their function and expression patterns can be altered by infectious agents and during inflammation [106, 108]. In addition to having direct antimicrobial activity, several β-defensins can also function as a DAMP. Indeed, β-defensin 2 is reported to activate dendritic cells (DCs) via binding to TLR4 [109], and β-defensin 3 similarly can activate the professional antigen-presenting DC’s and monocytes by interacting with TLR1 and -2 [110].

In human upper airway mucosa, levels of β-defensin 2 and β-defensin 3 have been assessed yielding equivocal results. Cultured sinonasal epithelial cells from CRSsNP patients showed equal levels of β-defensin 2 as observed in healthy controls [111]. However, in the presence of the TLR3 agonist dsRNA, β-defensin 2 levels were clearly increased in cultured sinonasal epithelial cells from CRSsNP patients but not in healthy controls [111]. In contrast, in CRSwNP, cultured sinonasal epithelial cells showed decreased expression of β-defensin 2 compared to control subjects [112]. These levels were further reduced in the presence of the Th2 cytokines IL-4 or IL-13 while the TLR9 ligand CpG DNA increased the levels of β-defensin 2 after TLR9 expression itself was being increased by adding the Th1 cytokine IFN-γ. The low β-defensin 2 levels in nasal polyp tissues are therefore in agreement with CRSwNP being a Th2-skewed disease for which reduced [44] or absent [43, 45] expression of TLR9 was reported. In another study, negligible levels of β-defensin 2 and -3 were reported in sinonasal samples of control subjects, CRSsNP and CRSwNP patients, while tonsillar tissue clearly expressed these defensins [42].

The reduced or absent levels of β-defensin 2 and -3 in CRSwNP gain even more immunological relevance in light of the proven antimicrobial activity against known pathogens such as S. aureus [113], which are frequent colonizers of human upper airway mucosa and increased in CRSwNP patients (Table 2) [17] in which they are implicated in the pathophysiology and severity of the disease and comorbid asthma [114, 115].

In contrast to nasal polyps from CRSwNP patients, in sinonasal polyp tissue from cystic fibrosis patients the levels of β-defensin 2 were reported to be clearly upregulated compared to healthy control tissue [116]. These higher levels of β-defensin 2 are in agreement with the reported increased expression of TLR9 in the sinonasal epithelium of cystic fibrosis patients [117].

Conclusions

The different molecules belonging to the DAMP family can be regarded as a mixed blessing; while implicated in defined housekeeping functions or host defense mechanisms such as tissue repair, cell differentiation, motility, growth, and/or antimicrobial activity, they can also play a role in the pathogenesis of human upper airway conditions via targeting pattern recognition receptors.

While the inflammatory microenvironment strives for resolution of disease via a coordinated elimination of pathogens or other injurious agents, the inability to completely eliminate noxious stimuli and/or an excessive derailed immune response towards them might lead to chronic inflammation. For upper airway pathologies such as CRSsNP and CRSwNP, the chronic inflammatory conditions have been extensively reported to be associated with increased cytokine, chemokine, eicosanoid and immunoglobulin production, altered patterns of inflammatory effector cells and T cells and tissue remodeling [74]. However, as covered in this review, the derailed immune response in chronically diseased upper airways could also be associated with persistent release of DAMPs, being generated in an attempt to achieve tissue healing during the host’s inflammatory response, but paradoxally yielding an autocrine loop of inflammation. Indeed, molecules such as RAGE, S100A8/A9 proteins, surfactant protein A and D, eosinophil-derived neurotoxin, fibronectin, galectin 1 and 3, tenascin-C, and β-defensins seem to be differentially present in the upper airways under pathological conditions and are therefore potentially implicated in the pathogenesis and/or etiology of the respective disease entities. All these molecules, except for RAGE, were, however, investigated apart from their DAMP function; therefore it is of importance to assess these molecules again from a point of view as an “alarming signal” from within the host in order to further unravel the pathogenesis of CRS. Unveiling the mechanisms or conditions by which these molecules, and others yet to be identified, switch from right to wrong may contribute to our knowledge with respect to prognosis and novel treatments to dampen inflammation.