Encyclopedia
Early or primary injury due to brain aggression, such as mechanical trauma, hemorrhage or is-chemia, triggers the release of damage-associated molecular patterns (DAMPs) in the extracellular space. Some DAMPs, such as S100B, participate in the regulation of cell growth and survival but may also trigger cellular damage as their concentration increases in the extracellular space. When DAMPs bind to pattern-recognition receptors, such as the receptor of advanced glycation end-products (RAGE), they lead to non-infectious inflammation that will contribute to necrotic cell clearance but may also worsen brain injury.
- acute brain injuries
- damage-associated molecular pattern molecules
- receptor for advanced glycation end-products
- biomarkers
1. Introduction
The cells of the innate immune system search the extracellular environment for exogenous pathogens or self-molecules—either modified (e.g., oxidized lipids) or usually confined within cells via pattern recognition and scavenger receptors (PRRs) [1][2][3][1,2,3]. In the central nervous system these receptors are mainly expressed by microglial cells but also by astrocytes, neurons, endothelial cells, and infiltrating leukocytes upon vessel injury or blood–brain barrier (BBB) opening [1][3][4][1,3,4]. Nevertheless, resident microglial cells have a larger repertoire of Toll-like receptors (TLRs) and a greater amount of the receptor for advanced glycation end-products (RAGE) [1][5][6][1,5,6]. According to the context, PRRs may polarize resident microglial cells and infiltrating leukocytes toward a panel of pro- or anti-inflammatory phenotypes, playing a role in tissue damage clearance and repair but which may also trigger neuronal and glial degeneration [1][3][5][7][8][1,3,5,7,8]. Although acute brain injuries encompass a wide variety of mechanisms, they share a common feature as they all lead to acute cellular necrosis with the release of intracellular ions, adenosine di- or triphosphate (ATP), proteins, and nucleic acids in the extracellular space, and eventually the extravasation of red blood cells and hemoglobin [9][10][11][12][9,10,11,12]. Collectively, this wide variety of endogenous molecules released upon tissue injury are termed damage or danger-associated molecular patterns (DAMPs).
Neurological recovery following acute brain injury depends on the patient’s prior medical condition and primary injuries, but also on secondary damage [13][14][13,14]. Among other secondary insults, the inflammation that is triggered by the release of DAMPs plays a crucial role in the development of brain lesions which will be discussed in the present review. The levels of inflammatory markers released in the systemic blood, such as the C-reactive protein, or interleukin (IL) 6 and 12, are actually predictors of post-stroke neurological dysfunction [14][15][16][14,15,16]. Moreover, the concomitant occurrence of a systemic inflammation, such as during sceptic or anaphylactic shock, increases neurological damage at the acute phase of stroke [17].
The acute increase of extracellular potassium after ischemia or hemolysis leads to the depolarization and swelling of neighboring cells thereby starting propagating waves of spreading depolarization [18] (Figure 1A,B). Potassium and ATP release from dying cells also activate the inflammasome in neurons and astrocytes via pannexin1, purinoreceptor, and nucleotide oligomerization domain receptors (NOD-like receptors) [1][19][1,19]. Larger molecules, such as S100 proteins, hemoglobin derivatives or the high mobility group protein 1 (HMGB1), bind mainly to TLRs and RAGE [1][10][1,10]; the ligation of TLRs and RAGE triggers a series of cellular signaling events, including the activation of nuclear factor-kappa B (NF-κB), leading to the production of pro-inflammatory cytokines, and causing non-sceptic inflammation [1][20][21][1,20,21]. However, the consequences of DAMPs ligation to PRRs depends on the context and their concentration in the extracellular space. For instance, S100B, a calcium binding protein that has several intracellular actions in astrocytes, can promote cell growth in the nM extracellular range (i.e., physiological conditions) [22][23][24][22,23,24]; whereas at higher concentration S100B activates astrocytes with a pro-inflammatory phenotype and facilitate neuronal death [24][25][24,25]. HMGB1 expression may also promote neuroinflammation related to brain injury but the deletion of the encoding gene is not able to prevent such consequences [26][27][26,27]. Furthermore, sustained activation and up-regulation of RAGE on neurons has been reported to cause death via stimulating the production of reactive oxygen species [28], but also regulates neurite growth and cell survival [24], as well as apoptosis and autophagy [20][29][30][20,29,30]. However, some extracellular soluble truncated receptors, such as sRAGE, act as decoys for ligands, and thus have a cytoprotective effect against advanced glycation end-products (AGE) and RAGE interactions; serum levels of sRAGE have been investigated in pathological process and proposed as a biomarker of their intensity and severity of outcome [31]. It seems that the net effect of DAMPs and PRRs, such as RAGE, depends on the context, cell type, and the number of DAMPs and the level of expression of PRRs.
Figure 1. Damage-associated molecular patterns (DAMPs) and pattern-recognition receptors (PRRs) changes at the acute phase of brain injury. (A) Morphological changes of neurons (grey), astrocytes (green), microglial and infiltrating leukocytes (blue), and DAMPs release (orange); (B) local field potential (LFP) and extracellular KCl recordings during the spreading depolarization triggered by the primary injury: the direct current (DC; 0–0.5Hz) shift is associated with a decrease of neuronal activity (AC; >0.5 Hz) and a KCl release; (C) Pattern-recognition receptors (PRRs) activation pathways; (D) kinetics of DAMPs and PRR expression as well as the course of cell death mechanisms. DAMPs: Damage-associated molecular patterns; IL: Interleukin; iNOS: inducible nitric oxide synthase; LFP: local field potential; MLKL: Mixed-lineage kinase domain-like pseudokinase; MyD88: Myeloid differentiation primary response 88; NFκB: nuclear factor-kappa B; RAGE: Receptor for advanced glycation end-products; RIPK1: receptor-interacting protein kinase 1; TLR: Toll-like receptor; TNF: Tumor necrosis factor; TNFR1: Tumor necrosis factor receptor 1.
2. Acute Lesion Progression Pathophysiology
2.1. Kinetics of DAMPs and Consequences after the Primary Insult
In addition to mechanical cell destruction, necroptosis is the main cell death pathway at the acute phase of traumatic brain injury (TBI), intracerebral or subarachnoid hemorrhage (SAH), and ischemic stroke (IS) [9][20][32][9,20,32]. Necroptosis is a morphologically lytic form of cell death implicating the receptor-interacting protein kinase 1 and 3 (RIPK1-RIPK3) and the mixed-lineage kinase domain-like pseudokinase (MLKL) pathway, resulting in the release of the contents of the cell into the extracellular space [33] (Figure 1A). The subsequent release of DAMPs peaks around 24 h after the primary insult and decreases thereafter over several days [4][12][20][34][35][36][37][38][4,12,20,34,35,36,37,38]. There is a spillover of DAMPs into the core of the primary injury, with a drastic decrease of intracellular HMGB1 [37] while HMGB1 is translocated from the nucleus to the cytoplasm of neurons but not glial cells at the periphery [4][9][37][39][4,9,37,39]. This extracellular release of HMGB1 may also act as a chemoattractant to monocytes [40] that infiltrate the core of the lesion from the first hours following the injury [4][37][4,37]. After several days, HMGB1-positive cells are mainly phagocytic microglial cells [4] (Figure 1D). Some authors have also described a biphasic expression of HMGB1 and S100 proteins with a second peak 14 days later [34][35][34,35]. The cellular consequences of DAMPs depend on the expression of PRRs but also on their post-translational modification. For instance, HMGB1 contains three highly conserved cysteines that are readily oxidized by reactive oxygen species, forming an intramolecular disulfide bridge thereby changing its conformation [41][42][41,42]. The reduced form released upon the primary injury binds the chemokine ligands (CXCL) 2 and 4 on monocytes and RAGE but not TLR-4, thereby promoting autophagy and acting as a chemoattractant for monocytes [40][43][40,43]. Conversely, the oxidized form may promote cell death and, at the late phase, vascular remodeling and progenitor cell migration via TLR signaling [35][43][35,43].
The expression of RAGE on neurons and microglial cells follows the same time course, with a biphasic pattern peaking on day 1 and 14 [12][20][34][35][36][44][45][12,20,34,35,36,44,45]. The amount of TLRs, which are predominantly expressed on microglial cells, increases at the subacute phase (i.e., after day 1) when activated microglia are present [34] (Figure 1D). The ligation of DAMPs to RAGE in turn activates the nuclear factor-kappa B (NF-κB) which in a positive feedback loop will increase the expression of RAGE. The activation of PRRs is related to different signaling pathways in neurons and glial cells. In neurons RAGE promotes the expression of proteins involved in necroptosis (i.e., MLKL) and autophagy (i.e., Becline-1); accordingly, blocking RAGE reduces autophagy, but also increases neuronal sensitivity to injury and apoptosis [20]. Later, at the subacute phase, TLRs and RAGE activation will polarize microglial cells and infiltrating leukocytes toward a pro-inflammatory phenotype, termed M1, with the release of cytokines such as tumor necrosis factor (TNF), IL-6 or IL-1 and the up-regulation of the inducible NO synthase [5][11][21][34][37][39][5,11,21,34,37,39]. Experimental overactivation of RAGE or TLRs by DAMPs injection or an increase in glucose degradation products, such as during hyperglycemia, will in turn increase the cytokine levels and worsen neuronal injuries [5][11][37][5,11,37]. The progression of secondary lesions follows different pathways. The ligation of TNF to the TNF receptor 1 (TNFR1) expressed by neurons will induce both necroptosis (via RIP1-3 and MLKL) and apoptosis (via caspase 3 and 8) [9][33][9,33]. Moreover, the pro-inflammatory cocktail can also activate astrocytes with a destructive phenotype (termed A1) leading to neuronal death and fewer synapses [7] (Figure 1C).
2.2. DAMPs Clearence
Under normal conditions, there is a continuous flow of cerebrospinal fluid (CSF) allowing the renewal of extracellular medium and the clearance of solutes such as amyloid-β (Aβ), also called the glymphatic system [46]. the subarachnoid CSF circulates in para-vascular spaces and enters the brain along penetrating arteries reaching the capillary bed; the interstitial fluid is then cleared along a perivenous drainage pathway and cervical lymphatic structures [46][47][48][46,47,48]. The glymphatic flux is driven by arterial pulse [47], water flux through aquaporin-4 expressed on astrocytes [46][48][46,48], as well as sleep cycles [30]. The CSF convection in the glymphatic system participates in the clearance of DAMPs to the peripheral blood [48], but also exhibits several changes following brain injury. The spreading depolarization that occurs at the very beginning of brain injuries triggers transient cell swelling, vasoconstriction [49][50][49,50] (Figure 1B), and increases the CSF flux in the glymphatic system, participating in the increase in brain water content [51]. However, the glymphatic flux is then impaired from 30 min to several days after the injury thereby leading to an accumulation of extracellular proteins such as DAMPs and Aβ [52][53][54][55][52,53,54,55]. DAMPs are also actively cleared from the extracellular space by infiltrating myeloid cells from the peripheral blood and activated resident glial cells. Infiltrating myeloid cells penetrate the lesion then differentiate into phagocytes that express scavenger receptors such as the macrophage scavenger receptor 1 (MSR1) that can bind and internalize HMGB1 and S100 proteins, as well as peridoxine [56]. These MSR1+ cells are present up to several days after the injury and exhibit an M2 phenotype, as opposed to the pro-inflammatory M1 phenotype of activated microglia that also internalize DAMPs [3][4][56][3,4,56]. Astrocytes are also able to internalize S100B into lysosomes by a RAGE-dependent mechanism [57]. Activated phagocytes may release HMGB1 from secretory lysosomes when modified lipids such as lysophosphatidylcholine are present at the subacute phase of acute brain injury, which may explain its biphasic release [58][59][60][58,59,60]. DAMPs can also traffic across the BBB by transcytosis in endothelial cells via RAGE and eventually reenter the brain from the blood compartment [61][62][61,62].
Increased expression of RAGE has been associated with the promotion of neuroinflammation in the main brain injuries, downstream HMGB1 release and NF-κB pathway activation [36]. Such a positive feedback loop of neuroinflammation may be blocked by intervention with antagonist of RAGE or the release of a truncated soluble form of RAGE (sRAGE) which may act as a decoy receptor to mitigate the inherent consequences of the inflammatory cascade. sRAGE is released upon brain injury [12], either secreted by astrocytes or monocytes [63] or from the cleavage of the membrane-bound RAGE [64]. sRAGE can scavenge extracellular and circulating DAMPs thus preventing their refilling into the brain [61] and reduce brain lesion progression [12][37][65][12,37,65]. The protective effect of recombinant sRAGE in cerebral parenchyma has been recently depicted in an animal model of SAH [65]. Although the characteristics of sRAGE sound interesting for clinical objectives, it has been little studied in brain injuries, with inconsistent results for outcomes such as in acute IS [66], or aSAH [31][67][31,67]. These discrepencies in data may result from different types of sRAGE studied and timing of measurements. We can speculate that the plasma level of sRAGE may be either positively associated with the intensity of injury, or negatively because of the consumption by its ligants (i.e., HMGB1 or S100B) possibly reflecting the healing process. Sometimes differences in plasma sRAGE may result from previous patient conditions such as diabetes or renal dysfunction [68]. In such a complex context it might be particularly interesting to combine more than one of these biomarker measurements to characterize the entire process.
