Dangers Within: DAMP Responses to Damage and Cell Death in Kidney Disease

Abstract

The response to exogenous pathogens leads to activation of innate immunity through the release of pathogen-associated molecular patterns (PAMPs) and their binding to pattern recognition receptors. A classic example is septic shock where Toll receptor 4 recognizes PAMPs. Although well accepted, this concept does not explain the activation of innate immunity and inflammation occurs with transplantation, autoimmunity, or trauma. Increasingly recognized is that endogenous molecules released by dying cells (damage-associated molecular patterns; DAMPs) activate cellular receptors leading to downstream inflammation. Thus endogenous danger signals and exogenous PAMPs elicit similar responses through seemingly similar mechanisms. Also emerging is our understanding that normal repair processes benefit from dampening the immune response to these endogenous danger molecules. Here we focus on the role of DAMPs and their putative receptors in the pathogenesis of acute and chronic kidney diseases.

The inflammatory response to acute or chronic tissue injury engages the immune system. What are the initial activators of injury or disease? How does the immune system discriminate between live versus dead cells and know whether to respond? What factors regulate the inflammatory response to clear injury without causing excessive tissue damage and then initiate repair? We now recognize that the well-known activation of the immune system in response to foreign pathogens is recapitulated in an immune response to endogenous molecules released from necrotic, and perhaps apoptotic, cells after tissue injury or trauma related to hypoxia, ischemia, mechanical stress, or pathogen-induced inflammation.

Matzinger¹ originally proposed the danger model to clarify exceptions to Janeway's model² of the immune response to foreign antigens, which did not at the time explain autoimmunity or the immune response to transplantation. The danger model suggests that damaged or dying cells release endogenous molecules called damage/danger-associated molecular patterns (DAMPs) that activate the immune system in a fashion analogous to pathogen-associated molecular patterns (PAMPs), molecules released by pathogenic bacteria or viruses. These endogenous or self-molecules (extracellular matrix proteins (ECM), calcium-binding proteins, and structural proteins) typically function in normal cell homeostasis but are also recognized as danger signals when released into the extracellular space³ exposing hydrophobic portions of the molecules that are normally hidden in healthy living cells.⁴ Matzinger^4–6 and others have extended the danger model as more has been learned about the role of danger signals in tissue injury^7,8 and diseases such as arthritis⁹ and cancer^10,11 and the body's need for mechanisms that dampen the immune response and initiate repair. Here we focus on the role of DAMPs and ligands of Toll receptors (TLRs) in renal disease^12–15 and expand recent interest to a broader view of the currently identified classes of DAMPs and their putative receptors.

DANGER-ASSOCIATED MOLECULAR PATTERNS: ENDOGENOUS DANGER SIGNALS

A consistent terminology has not been adopted for the endogenous molecules that convey a danger signal to the immune system. Some DAMPs that stimulate an immune response have been called adjuvant molecules to distinguish them from DAMPs that produce only acute pro-inflammatory effects,³ sometimes referred to as alarmins.^16,17 PAMPs and alarmins have been grouped together as subcategories of a large family of DAMPs,^16,17 whereas others consider alarmins and DAMPs to be related molecules that are clearly distinguished from PAMPs.⁹ Here we use the term DAMPS to describe endogenous danger molecules as a group that is separate from pathogen-derived PAMPs (Table 1 and Figure 1); their classification is predicated on direct evidence of involvement in the immune response to injury with a clear absence of confounding effects from potential bacterial contaminants, such as LPS.^3,16

Table 1.

DAMPs and receptors for DAMPs

DAMP	Putative Receptors	References
Ligands of RAGE		40,42
AGEs	RAGE	39–42
HMGB1	TLR2, TLR4, TLR9, CD44,	18–28
	RAGE
S100 proteins/calgranulins	RAGE, TLR4	29–33
amyloid-β	RAGE, NLRP3	160
HSPs	CD14, CD91, TLR2, TLR4, CD40	34–38
Chromatin and DNA	TLR9	43–47
Uric acid (MSU) crystals	TLR2, TLR4, CD14	48–52
Neutrophil-derived alarmins		16,17
cathelicidins	TLR7, TLR9, FPRL1, FPR2	54
defensins	TLR4, CCR6	53
lactoferrin	TLR4	55
Extracellular matrix proteins		3,56,57
proteoglycans
hyaluronan	TLR2, TLR4, NLRP3
biglycan	TLR2, TLR4, NLRP3
versican	TLR2
fibrinogen	TLR4, Integrins
heparan sulfate	TLR4
fibronectin extra domain A	TLR4
laminin	Integrins
elastin-derived peptides	Integrins
collagen-derived peptides	CXCR2
Galectins	Unknown	58,59
Thioredoxin	Unknown	65,66
Adenosine; ATP	P1, P2X, and P2Y receptors	60–64
IL-33	ST2/IL-1R	67
Tamm-Horsfall glycoprotein	TLR4	68–73

CXCR2, CXC-chemokine receptor 2; FPR2, formyl peptide receptor 2; FPRL1, formyl peptide receptor-like receptor 1; MSU, monosodium uric acid; NLRP3, NLR family, pyrin domain-containing 3.

Figure 1.

Danger and stranger models. Infections of pathogenic bacteria or viruses cause release of PAMPs that bind to pattern recognition receptors (PRRs), such as TLRs, on immune cells and stimulate an innate immune response that is accompanied by inflammation, activation of adaptive immunity, and eventually processes to resolve the infection and allow for tissue repair. The danger model recognizes that similar events occur when cells are stressed or injured and that necrotic cells release molecules that are normally hidden within the cell. In the extracellular space these DAMPs can bind to PRRs or to specialized DAMP receptors to elicit an immune response by promoting release of pro-inflammatory mediators and recruiting immune cells to infiltrate the tissue. The immune cells that participate in these processes include, for example, APC, such as dendritic cells and macrophages, as well as T cells and neutrophils (PMN). DAMPs may also stimulate adaptive immunity and participate in autoimmune responses and tissue repair. A wide variety of intracellular and extracellular molecules function as DAMPs when released from cells (Table 1). The functions of such a diverse group of molecules may not yet be fully elucidated; it is unknown whether different DAMPs have specific roles, whether specific functions are elicited in different cell types or conditions, or even whether immune responses to DAMPs can be distinguished from those of PAMPs.

High-mobility Group Box 1 Protein

Perhaps the most well characterized DAMP, high-mobility group box 1 (HMGB1) protein, is a ubiquitously expressed nonhistone DNA-binding protein that regulates chromosomal stability, stabilizes nucleosomes, and regulates transcription.^18–22 As an extracellular DAMP after secretion or passive release,²³ HMGB1 is a late-phase pro-inflammatory mediator in sepsis²⁴ and in sterile inflammation, such as hepatic ischemia-reperfusion injury (IRI).^18,25 HMGB1 is chemotactic for immune cells and stimulates dendritic cell (DC) maturation and migration.²⁶ HMGB1 binds to the receptor for advanced glycation end products (RAGE)²⁷ and stimulates NFκB-induced transcription through interactions with TLR2, TLR4,²⁸ and RAGE.

S100 Protein Family

The S100 proteins, a large family of calcium-binding proteins, are implicated in the inflammation or fibrosis associated with cancer or diseases of the kidney, heart, joints, and lungs.²⁹ When functioning extracellularly as DAMPs after release from phagocytes³⁰ and other cells in response to cell stress, S100 proteins bind to RAGE³¹ and other receptors and produce earlier pro-inflammatory effects like HMGB1.^32,33 S100A8 and S100A9 release from activated phagocytes activates TLR4 and amplifies lethal endotoxin-induced shock.³³

Heat-shock Proteins

In normal healthy cells, heat-shock proteins (HSPs) are intracellular protein chaperones that guide newly synthesized polypeptide chains to prevent aggregation and misfolding. During cell stress, induction and secretion of HSPs causes pro-inflammatory cytokine and chemokine release and activation and maturation of antigen-presenting cells (APCs) to produce a robust innate immune response.^34–36 Indeed, HSPs extend their role as chaperones intracellularly by binding and presenting antigens to cell surface MHC class I molecules.^37,38 Complexes of extracellular HSPs and antigens are taken up by APCs and presented to MHC I to activate T cells through cross-presentation.^36,37

Other Ligands of RAGE

A diverse group of ligands (including HMGB1, S100 proteins, and HPSs) bind to RAGE. Advanced glycation end products (AGE) and related molecules are glycation and oxidation products of proteins and lipids that are formed in hyperglycemic states by oxidative stress, such as hypoxia and ischemia/reperfusion, and in a number of diseases.^39–42

Genomic Double-stranded DNA

Microbial DNA is a ligand for TLR9, but TLR9 can also recognize self-DNA from injured mammalian cells.⁴³ Release of DNA from dying mammalian cells initiates an innate immune response by activation of TLR9 and the NALP3 (cryopyrin) inflammasome, release of activated IL-1β and IL-18,⁴⁴ and induction of DC maturation.⁴⁵ In addition to genomic DNA, injury-induced release of mitochondrial DNA can also cause inflammation by conserved pathogenic PAMP sequences.^46,47

Uric Acid

The accumulation of uric acid in tissues has long been known to cause gout, but soluble uric acid released by injured cells also acts as a danger signal.⁴⁸ The active moiety, monosodium urate crystals, forms in the extracellular space where it stimulates an immune response by activating the NALP3 inflammasome^49–51 and by stimulating DC maturation and T-cell responses.⁵²

Neutrophil-derived Alarmins

Neutrophils, an early leukocyte to infiltrate tissue after injury or infection, undergo degranulation and release immune modulatory proteins and polypeptides. Neutrophil-derived alarmins (such as α-defensins,⁵³ cathelicidin,⁵⁴ and lactoferrin⁵⁵), named for their response to danger signals, link innate and adaptive immunity by recruiting leukocytes and inducing maturation of APCs, thereby bridging neutrophil and DC function.¹⁷

Extracellular DAMPS

Damaged or dying cells can also generate danger signals by cleaving molecules that are structural components of the ECM. Cleaved ECM glycoproteins, such as fibronectin and fibrinogen, and proteoglycans, such as hyaluronan, biglycan, and versican, function as DAMPs by signaling through TLRs, the NLRP3 inflammasome, and other receptors such as CD44, promoting a vigorous immune response^3,56 and contributing to kidney disease.⁵⁷

Other DAMPs

Other candidate DAMPs include galectins^58,59; ATP and adenosine^60–64; thioredoxin, a ubiquitous antioxidant enzyme^65,66; the intranuclear cytokine, IL-33⁶⁷; and Tamm-Horsfall protein (THP; or uromodulin), a glycoprotein expressed in the thick ascending limb that is excreted into the urine after proteolytic cleavage. THP is the most abundant protein in urine, but its function remains unclear⁶⁸; it has been implicated as a biomarker of kidney disease and allograft rejection^69,70 and may be a damage signal in renal tubular injury. THP promotes cytokine release⁷¹ and immune cell activation, including DC maturation,⁷² but may also have a protective anti-inflammatory role.⁷³

RECEPTORS

DAMPs are recognized by pattern-recognition receptors such as certain TLRs, NOD-like receptors (NLRs, including the NLRP inflammasomes), and RLRs (RIG-I-like receptors), thereby overlapping with PAMPs,^74,75 and by specialized receptors such as RAGE.⁷⁶ Common signaling pathways shared by DAMPs and PAMPs may have an evolutionary basis.⁴⁷ Injury-induced release of bacterial-related DAMPs from mitochondria, evolutionary symbiotic descendents of bacteria, causes inflammation.⁴⁶ Other DAMPs receptors include CD91, scavenger receptors, CD2, integrins, chemokine receptors, and CD44.^3,11,16,74

RAGE is a multiligand receptor that binds AGE, some S100 proteins, amyloid protein, and HMGB1. It amplifies immune responses³⁹ and is implicated in diseases such as diabetes, Alzheimer's disease, cancer, and various inflammatory conditions.^40,76,77 A soluble form of RAGE (sRAGE) found in human and mouse plasma⁷⁸ may act as a ligand decoy to bind danger molecules and check the immune response. Circulating sRAGE and a splice variant (endogenous secretory RAGE) may be clinical indicators or biomarkers for the severity of disease.^41,79

Inflammasomes, large cytosolic complexes of NOD-like receptors, adaptor protein, and caspase-1, link pathogenic and endogenous danger signals to activation of caspase-1, and proteolytic processing and release of the proinflammatory cytokines, IL-1β, and IL-18.^50,80,81 Inflammasomes are important in gout,⁸² and their role in disease has been linked to several DAMPs, including the ECM molecule, biglycan.⁵⁷

CHECKS AND BALANCES

With such a vast array of stranger and danger signals and some crossover in their receptors, how does the immune system distinguish PAMPs from DAMPs? Furthermore, once an immune response is initiated, what mechanisms are engaged to prevent rampant tissue damage and begin the repair process? Although PAMPs and DAMPs bind to some of the same receptors, interaction with different coreceptors may account nt for a divergencefor a divergence in downstream effects⁸³ allowing for discrimination between PAMPs and DAMPs^84,85 and selective suppression of DAMP signaling (Figure 2).^85–87

Figure 2.

Proposed mechanisms for discriminating DAMPS from PAMPs and limiting the immune response. Several groups have recently proposed mechanisms that share a common theme involving coreceptors that pair with receptors for DAMPs and PAMPs to allow cells to distinguish between these related molecules and perhaps elicit downstream immune responses that may be specific for the ligand. For simplicity, these examples are illustrated only with inflammation as the target response, but selective signaling mechanisms may also result in varied inflammatory responses and provide specificity that may or may not include activation of adaptive immunity and tissue repair processes. (1) The CD24-Siglec pathway can distinguish DAMPs from PAMPs and can suppress DAMP signaling to prevent an unrestrained immune response and excessive collateral tissue damage.^85–87 By interacting with sialic acid-binding Ig-like lectins (Siglec-G (mouse) or Siglec-10 (human)),¹⁵⁷ CD24, a glycosylphosphatidylinositol (GPI)-anchored molecule (also known as heat-stable antigen) with diverse T-cell homeostatic functions,¹⁵⁸ negatively regulates the TLR- or NLR-mediated immune response to HMGB1 and heat-shock proteins but not PAMPs (LPS or poly(I:C)) perhaps by facilitating association with phosphatases, like SHP-1.⁸⁵ (2) Others suggest that although PAMPs and DAMPs bind to some of the same receptors, interaction with different coreceptors may account for a divergence in downstream effects.⁸³ (3) A mechanism for discriminating DAMPs from PAMPs has also been demonstrated for two other cell surface proteins, CD14 (which recognizes DAMPs in the absence of TLR2 and promotes both TLR2-DAMP and TLR2-PAMP responses) and myeloid differentiation protein 2 (MD2; responding to and enhancing only exogenous PAMP responses in complexes with CD14-TLR2 or CD14-TLR4), that form complexes with Toll receptors.⁸⁴ (4) Similarly, biglycan binding and induction of a multireceptor complex with TLR2/4 and purinergic P2×4 or P2×7 receptors activates the NLRP3 inflammasome,⁵⁷ and this complex may regulate immune cell infiltration and tissue injury in kidneys (not illustrated here).⁵⁶

Some mechanisms control damage-inducing immune responses through DAMPs inactivation. Anti-HMGB1 antibodies⁸⁸ and sRAGE reduce HMGB1 bioavailability.⁴⁰ Extracellular redox conditions may regulate the immune response by balancing early-stage reducing environment-promoting pro-inflammatory conditions with late-stage oxidation-induced DAMPs inactivation; disruption of this balance may contribute to chronic inflammation.^19,89,90 BCL2 proteins, a family of pro- or antiapoptotic moieties, may be tissue protective when released extracellularly in IRI.⁹¹

Some endogenous molecules contribute to the resolution of inflammation and initiation of renal repair.⁹² Weibel-Palade bodies are organelles of endothelial cells that contain bioactive substances such as von Willebrand factor, IL-8, P-selectin, angiopoietin-2, and eotaxin⁹³ that participate in hemostasis and inflammation. Uric acid release after IR mobilizes stem cells, protects kidneys from injury,^94,95 and through exocytosis of Weibel-Palade bodies and release of their constituent molecules may promote postischemic repair.⁹⁶ Some DAMPs, including HMGB1 and some ECM proteins, such as hyaluronan and heparin sulfate, initiate immune responses but are also necessary for recovery and healing.¹⁹

DAMPS AND KIDNEY DISEASE

Despite significant advances in understanding the important contribution of inflammation and immune mechanisms to the pathogenesis of a variety of kidney diseases, few specific or efficacious therapies exist. Understanding the role of candidate DAMPs released from somatic kidney or immune cells could reveal novel drug targets for inhibiting the inflammatory response or promoting repair processes in acute and chronic kidney disease (CKD). DAMPs contribute to multiple diseases. TLRs are important in IRI and various forms of glomerulonephritis including lupus nephritis.^12–15 Advanced glycation end products, DAMPs that bind to RAGE rather than TLRs, are increased in renal failure and are involved in the progression of CKD in diabetic and nondiabetic kidney disease.^97,98

Acute Kidney Injury

Acute kidney injury (AKI) results most commonly from IR, and although there are many causes, a common pathway leading to proximal tubule injury is activation of innate and adaptive immunity leading to inflammation.⁹⁹ Injury elicits release of pro-inflammatory DAMPs, which may be signals that engender the inflammatory response to IRI by binding to receptors such as TLRs and RAGE.¹⁰⁰ The best-characterized DAMPs in AKI are HSPs and HMGB1. Although their role in mediating renal injury in acute ischemic events is well documented, some studies do not establish that they function as DAMPs to stimulate the immune system. Nevertheless, we include them as viable DAMPs candidates on the basis of their role as DAMPs in other pathologies. In ischemic kidneys or hypoxic renal epithelial cells, stimulation of the TLR2-mediated ERK pathway is regulated by the HSP, gp96.¹⁰¹ Geldanamycin, an inhibitor of Hsp90 and gp96, protects mouse kidneys from IRI.¹⁰² Hsp70 and Hsp27 are up-regulated in rat kidneys after IRI¹⁰³; however, tissue-protective¹⁰⁴ and pro-inflammatory¹⁰⁵ roles for Hsp27 have also been reported in kidneys after IRI. Ethyl pyruvate, an inhibitor of HMGB1 release, protects kidneys from IRI and reduces the increase in TNFα in rat kidneys subjected to IRI.¹⁰⁶ TLR4- and MyD88-deficient mice showed protection from kidney IRI; HMGB1, hyaluronan, and biglycan expression increased in these null mice, suggesting that these DAMPs may be ligands for the observed role of TLR4 in IRI.¹⁰⁷

High concentrations of uric acid accumulate in ischemic tissues and precipitate to form monosodium urate crystals that elicit an immune response. However, a single treatment with uric acid or the enzyme uricase to produce an acute rise in blood levels of uric acid mobilized endothelial progenitor cells and protected kidneys from IRI; this effect was absent with hyperuricemia.⁹⁵

Kidney injury molecule-1 (Kim-1/Tim-1), one of a growing list of biomarkers of AKI and CKD, is a multifunction receptor protein expressed in proximal tubules that is released into the urine of patients with kidney disease.¹⁰⁸ In addition to imparting phagocytic properties to tubule epithelial cells,¹⁰⁹ Kim-1/Tim-1 may inhibit development of autoimmune responses and promote resolution of inflammation after kidney injury,¹⁰⁸ therefore suggesting its role as a possible DAMP molecule or receptor.

Diabetic Nephropathy and Nondiabetic Glomerular Diseases

There is expansive literature on the involvement of AGE and RAGE in podocytes, diabetes, and diabetic nephropathy.^{42,76,97,110–117} It is presumed but not always demonstrated that these molecules are important as DAMPs because of the involvement of inflammation and the immune system in diabetes.^{42,115,118–120} RAGE, functioning as an endothelial adhesion receptor, is implicated in a novel pathway for leukocyte recruitment in inflammatory disorders and diabetic mice, and this process is enhanced by S100, a proinflammatory RAGE ligand.¹²¹ Experiments using RAGE null or overexpressing mice, neutralizing RAGE antibodies, or sRAGE implicate RAGE in the pathophysiology of diabetic nephropathy and chronic inflammation.^76,112,122 In db/db mice, increased RAGE and S100 expression in podocytes associates both with renal pathology and with increased infiltration of mononuclear phagocytes to glomeruli; these effects are blocked by anti-RAGE antibody.¹¹² Serum RAGE levels, kidney damage, and inflammatory mediators in diabetic animals and humans are inversely related to levels of AGE receptor-1,^123,124 suggesting an anti-inflammatory protective role for AGE receptor-1 in diabetes.¹²³

AGE, RAGE,^97,117 and other DAMPs are also strongly implicated in the pathogenesis of nondiabetic glomerular diseases. Podocyte injury and adriamycin-induced glomerulosclerosis are accompanied by production of RAGE ligands; RAGE-deficient or sRAGE-treated mice are protected from glomerular injury.¹²⁵ Increased AGE and RAGE levels, macrophage infiltration, fibrosis, and inflammation are found in the kidneys of mice that develop glomerular disease from an atherogenic high-fat diet; this lipid-induced renal injury is regulated in part by galectin-3.¹²⁶

TLR4 is up-regulated in mouse models of cryoglobulinemic membranoproliferative glomerulonephritis and localizes to podocytes of nephritic glomeruli; stimulation of podocytes with either TLR4 ligands or fibrinogen results in similar patterns of enhanced chemokine expression, suggesting that TLR4 functions as a DAMPs receptor for endogenous ligands and may mediate glomerular injury by stimulating innate immunity.¹²⁷ In nephrotoxic nephritis, a model of rapidly progressive glomerulonephritis, Hsp60 is released from kidneys and excreted in urine, and administration of Hsp60 exacerbates disease in a T cell-dependent manner.¹²⁸

Expression of S100 proteins increases in kidneys in anti-Thy-1 antibody-induced glomerulonephritis, a rat model of mesangial proliferative glomerulonephritis.¹²⁹ Patients with glomerulonephritis have HMGB1 in serum, interstitial mononuclear cells, and glomeruli.¹³⁰ Two S100 proteins, myeloid-related proteins MRP8 and MRP14, are detected in macrophages in renal biopsies from patients with glomerulonephritis; the presence of these proteins and their heterodimeric complexes in macrophages infiltrating glomeruli correlates with the severity of the acute inflammatory process, and chronic inflammation associates with MRP8/MRP14 infiltrates without complex formation in the renal interstitium.¹³¹

Fibrosis

Expression of Hsp27, phospho-Hsp27,¹³² RAGE,¹³³ and calreticulin¹³⁴ is up-regulated in an animal model of tubulointerstitial fibrosis (unilateral ureteral obstruction [UUO]) and TGFβ-induced epithelial-to-mesenchymal transition in proximal tubular epithelial cells in vitro. The ECM protein, biglycan, increases in tubule epithelial cells 4 days after UUO and later in infiltrating and interstitial cells.¹³⁵ Biglycan is a pro-inflammatory DAMP in UUO-induced kidney injury; activation of the NLRP3 inflammasome and increases in levels of pro-inflammatory IL-1β after UUO are reduced in biglycan null mice.⁵⁷ Reactive oxygen species, oxidative stress, S100A4, and Hsp47 are involved in fibrosis after IRI, but their role in immune function has not been investigated.¹³⁶

Lupus Nephropathy and Autoimmune Disease

HMGB1 has been linked to the pathogenesis of a variety of proinflammatory and autoimmune diseases, including systemic lupus erythematosus (SLE).¹³⁷ Circulating HMGB1 levels increase in SLE patients and in mice,^138,139 and HMGB1 may be involved in antibody-induced kidney damage in SLE.¹⁴⁰ In lupus-prone MRL-Fas(lpr) mice, p38 MAPK activation-induced infiltration and maturation of DCs and secretion of HMGB1 from DCs have been implicated in autoimmune kidney disease.¹⁴¹ Inflammatory and immune responses in SLE, and particularly in lupus nephritis, can be induced by HMGB1-nucleosome complexes,^138,142 but necrotic nucleosomes also contain double-stranded DNA, which has long been recognized as a key mediator of lupus nephritis. Chromatin fragments can participate in the pathogenesis of kidney disease in SLE by stimulating the innate immune system and by engaging the adaptive immune response to produce antichromatin and anti-double-stranded DNA antibodies leading to glomerular deposits of immune complexes, which are the hallmark and likely the critical initiating events in lupus nephritis.¹⁴³

Transplantation

The contribution of DAMPs to IRI is likely also pertinent in kidney transplantation, because donor kidneys are susceptible to delayed graft function resulting from ischemia followed by reperfusion. Release of DAMPs, such as HMGB1, ATP, uric acid, and IL-1α, from allografts may induce pro-inflammatory effects and adaptive immune responses.¹⁴⁴ TLRs mediate the effects of some these endogenous molecules in transplant by signaling through type 1 interferons.¹⁴⁵ In human kidney transplants, increased expression of TLR4 and HMGB1 is found in deceased donor kidneys, and kidneys from patients with a TLR4 loss-of-function mutation have lower levels of inflammatory markers and improved graft function post-transplant.¹⁴⁶ Moreover, TLRs and their endogenous ligands may be important in chronic allograft dysfunction. Expression of DAMPs, including biglycan, HSPs, fibrinogen, and HMGB1, increases in the acute and chronic phases after kidney transplantation.¹⁴⁷ Kidney transplants to mice lacking TLR2, TLR4, or the adaptor proteins, MyD88 and TRIF, show improved graft function and morphology and decreased leukocyte infiltration and expression of fibrotic markers, cytokines, and chemokines.¹⁴⁷

Higher levels of S100A8 and S100A9 in renal biopsies taken in the acute phase after transplant are predictive of favorable graft outcome (patients that later had stable graft function compared with those progressing to chronic allograft nephropathy)¹⁴⁸; these S100 proteins are typically pro-inflammatory, but their beneficial role in wound repair may predominate in transplantation and rejection. Glycosylation of Tamm-Horsfall glycoprotein is altered, and its various immunomodulatory functions are diminished in renal transplant patients at least in part because of altered NF-κB p52 nuclear translocation.¹⁴⁹

Nondiabetic Chronic Kidney Disease

Many of the DAMPs molecules associated with AKI are also important in CKD. Increased HMGB1 levels correlate with pro-inflammatory markers and declining kidney function in CKD.¹⁵⁰ Endothelial dysfunction in CKD, which associates with elevated serum levels of AGE, may be due to AGE-induced suppression of endothelial NOS.¹⁵¹ RAGE is proinflammatory in hemodialysis patients¹⁵²; sRAGE and endogenous secretory RAGE levels inversely correlate with renal function and inflammatory state in hemodialysis patients.^153,154 Elevated TLR2 expression in patients with CKD and in mice with obstructive nephropathy associates with inflammation and increased expression of biglycan and HMGB1, but TLR2 may not be necessary for chronic kidney injury.¹⁵⁵ By contrast Tamm-Horsfall protein activates innate and adaptive immunity through TLR4⁷² and may play an inflammatory role in progression of CKD.⁷¹ Levels of HSPs, anti-HSP antibodies, and inflammatory markers are higher in children and young adults with CKD.¹⁵⁶

CONCLUSIONS

Processes that contribute to the pathogenesis of acute and chronic kidney disease involve the activation of innate and adaptive immunity. Release of endogenous molecules from dying cells leads to activation of innate immunity and downstream inflammation. By bridging to adaptive immunity, DAMPs also contribute to progressive injury or to attenuation of injury. As our understanding of these molecules and pathways evolves, information on precise therapeutic targets will likely lead to improved outcomes.

DISCLOSURES

This work was supported in part by grants from the National Institutes of Health (R01DK062324 and R01DK056223) and Genzyme (GRIP).