Inflammasome links traumatic brain injury, chronic traumatic encephalopathy, and Alzheimer’s disease

Abstract

Traumatic brain injury, chronic traumatic encephalopathy, and Alzheimer’s disease are three distinct neurological disorders that share common pathophysiological mechanisms involving neuroinflammation. One sequela of neuroinflammation includes the pathologic hyperphosphorylation of tau protein, an endogenous microtubule-associated protein that protects the integrity of neuronal cytoskeletons. Tau hyperphosphorylation results in protein misfolding and subsequent accumulation of tau tangles forming neurotoxic aggregates. These misfolded proteins are characteristic of traumatic brain injury, chronic traumatic encephalopathy, and Alzheimer’s disease and can lead to downstream neuroinflammatory processes, including assembly and activation of the inflammasome complex. Inflammasomes refer to a family of multimeric protein units that, upon activation, release a cascade of signaling molecules resulting in caspase-induced cell death and inflammation mediated by the release of interleukin-1β cytokine. One specific inflammasome, the NOD-like receptor protein 3, has been proposed to be a key regulator of tau phosphorylation where it has been shown that prolonged NOD-like receptor protein 3 activation acts as a causal factor in pathological tau accumulation and spreading. This review begins by describing the epidemiology and pathophysiology of traumatic brain injury, chronic traumatic encephalopathy, and Alzheimer’s disease. Next, we highlight neuroinflammation as an overriding theme and discuss the role of the NOD-like receptor protein 3 inflammasome in the formation of tau deposits and how such tauopathic entities spread throughout the brain. We then propose a novel framework linking traumatic brain injury, chronic traumatic encephalopathy, and Alzheimer’s disease as inflammasome-dependent pathologies that exist along a temporal continuum. Finally, we discuss potential therapeutic targets that may intercept this pathway and ultimately minimize long-term neurological decline.

Introduction

Traumatic brain injury (TBI), Alzheimer’s disease (AD), and chronic traumatic encephalopathy (CTE) represent interrelated neurological disorders that share common epidemiological, molecular, and neuropathological features including dysregulation in protein folding and alteration in post-translation-modifications such as protein phosphorylation. One of these proteins includes the microtubule-associated protein tau, leading to tau phosphorylation (p-tau) (Katsumoto et al., 2019). Tau phosphorylation and hyperphosphorylation processes involve the phosphorylation of tau and its translocation into the cell body from the axons and aggregating intracellularly forming neurofibrillary tangles (NFTs) inclusions (Khlistunova et al., 2006; Alonso et al., 2008). While tauopathy is a common shared pathological finding, cytokine release, protease activation, neural cell death, and neuroinflammation involving microglia and astrocyte activation are also observed. Importantly, a major contributor to the inflammatory response experienced from TBI and AD is the inflammasome, a component of the innate immune response that activates the pro-inflammatory cytokines interleukin (IL)-1β and IL-18. Additionally, the inflammasome triggers the programmed cell death mechanism of pyroptosis (Chai et al., 2023). In this review, we discuss the integral role of the inflammasome in the formation of tauopathic deposits and provide an overview of TBI, CTE, and AD neuropathology involving inflammasome formation, maturation, and release.

Traumatic brain injury

Traumatic brain injury epidemiology and diagnosis

TBI is only recently beginning to receive the recognition it deserves as a global public health burden. TBI is a broad term that encompasses non-congenital neurological events caused by acceleration and deceleration forces, rotational forces, and/or blast waves acting on the head (Gean and Fischbein, 2010; Alluri et al., 2015; Dewan et al., 2018; Gu et al., 2022; Rossom et al., 2023). TBI may occur as a penetrating or non-penetrating trauma due to falls, motor vehicle accidents, explosions or blasts, and physical assaults such as those seen in contact sports (Hawkins et al., 2013; Michalicova et al., 2017). TBI is a complex, chronic pathology with heterogeneous manifestations that can occur via many processes, resulting in varied phenotypes including neuroinflammation, protein misfolding, and neuronal cell death (Golding, 2002; Alluri et al., 2015; Reddy et al., 2023).

TBI is on the rise with one 2016 study reporting 27 million new cases globally (GBD 2016 Traumatic Brain Injury and Spinal Cord Injury Collaborators, 2019) and a 2018 international epidemiological review estimating a prevalence of 69 million people affected yearly (Dewan et al., 2018). TBI is, therefore, a leading cause of morbidity and mortality, and in the United States, it cost the healthcare system 40.6 billion dollars in 2016 alone (Loane et al., 2009). Clinically, TBI is extremely diverse with no two TBI’s exhibiting the same phenotype. Most cases, 70%–90%, are classified as mild TBI according to the Glasgow Coma Scale (GCS). These TBIs are often colloquially referred to as “concussion” (Kattan et al., 2023). TBI can manifest as physical, emotional, cognitive, and/or functional impairments, depending on the location and force of impact (Corsellis, 1989; Alluri et al., 2015; Broshek et al., 2015; Acabchuk et al., 2021; Waithe et al., 2024). Examples of TBI symptomology include pain and fatigue, anger, anxiety and irritability, memory and attention deficits, dizziness and seizures, and difficulty with motor functioning (Lozano et al., 2015). As the consequences of repeated head trauma affect American National Football League (NFL) players, TBI has grasped the attention of media and scientists alike. Thus, TBI, and the neurodegeneration that may follow, are shifting away from being a “silent” epidemic. However, the precise mechanisms by which TBI leads to later onset cognitive and behavioral changes are still being elucidated.

Traumatic brain injury pathology

TBI induces neural injury as a result of mechanical head trauma in the primary phase. The acute and short-term outcomes of TBI are wide-ranging and patients may experience varied symptoms from a brief episode of dizziness or mild alterations in consciousness to coma or even death (Galgano et al., 2017; Sandsmark et al., 2019a; Bounajem et al., 2022). TBI pathogenesis occurs in a two-phased response. Immediately post-injury, tissue damage occurs resulting in the primary injury phase where there is neural injury and acute neuroinflammation, causing direct damage to neurons and supporting cells (Alluri et al., 2015; Sandsmark et al., 2019a, b; Bounajem et al., 2022). This may include parenchymal hemorrhages, focal contusions, intraparenchymal hemorrhages, axonal shearing, and cerebral edema (Jarrahi et al., 2020; Haidar et al., 2022; Kattan et al., 2023). The second phase occurs days, months to years following the initial injury and is characterized by chronic neuroinflammation, excitotoxicity, mitochondrial dysfunction, metabolic deficiencies, reactive oxygen species formation, and prolonged blood–brain barrier (BBB) disruption increased permeability; furthermore, at chronic phases, dysregulation of protein folding and phosphorylation modifications including the TDP-43, tau/p-tau proteins are observed as depicted in Figure 1A and B (Kobeissy et al., 2008; Zhang et al., 2011; Huang et al., 2017; Johnson et al., 2017; Wright et al., 2017; Abou-El-Hassan et al., 2020).

Figure 1.

Neuropathological and biochemical changes post-TBI.

TBI neuropathological sequelae are linked to several pathological processes including (A) NLRP3 inflammasome activation, neuroinflammation and microglial activation, tau hyperphosphorylation, oxidative stress as well as an axonal shearing as in diffused brain injury. This is coupled with (B) elevated mitochondrial dysfunction and ROS secretion. Created with BioRender.com. DAMPs: Damage association molecular patterns; NLRP: nucleotide-binding oligomerization domain-like receptor protein 3; ROS: reactive oxygen species; TBI: traumatic brain injury.

On the clinical level, some patients with TBI exhibit hyperphosphorylated tau protein and NFTs that are diagnosed postmortem and are typically seen in hippocampal and superficial cortical regions (Zanier et al., 2018; Kahriman et al., 2021). Tau density can be used to assess histological severity (Saman et al., 2014) Importantly, TBI pathology worsens temporally with several studies documenting increased p-tau deposition over time; hence, TBI is considered a chronic condition with longitudinal sequelae (Kim, 2011; Tomkins et al., 2011; Tagge et al., 2018; Amyot et al., 2021; Thakur et al., 2023). Intriguingly, these observations mirror the progressive clinical deterioration often noted among long-term TBI survivors.

TBI is a well-studied risk factor for the chronic neurodegeneration seen in both CTE and AD (Turner et al., 2012, 2016; Blennow et al., 2016; Ramos-Cejudo et al., 2018) with several cohort studies reporting significantly earlier AD onset in TBI survivors (Nemetz et al., 1999; Fleminger et al., 2003; Nguyen et al., 2018; Schaffert et al., 2018). One recent study looking at the brains of 205 male contact sport athletes reported a decrease in myelin-associated glycoprotein and proteolipid protein 1 that correlated with years of sport played (Alosco et al., 2023). While further clinical studies are needed, it is thought that this decrease in myelin-associated proteins may be responsible for the late-stage cognitive and personality aberrations seen in many contact sports athletes. Currently, there are no Food and Drug Administration–approved medications for TBI (Rabinowitz and Watanabe, 2020). However, given the temporal nature of TBI-induced neurodegeneration, timely therapeutic intervention is critically needed for post-TBI patient care which is also dependent upon elucidating the pathophysiological mechanisms involved.

Chronic traumatic encephalopathy

Chronic traumatic encephalopathy history and epidemiology

Dr. Harrison Martland first used the term punch drunk in 1928 to describe a clinical phenomenon commonly observed in American professional boxers (Castellani and Perry, 2017). Punch drunk syndrome, which later came to be known as dementia pugilistica, consisted of a medley of symptoms including tremors, gait changes, vertigo, and cognitive decline that result in institutionalization (Castellani and Perry, 2017; Changa et al., 2018). As technological advances were made, clinical symptomatology became better understood as brain changes that occur because of microtraumas from the force of mechanical head injury. Microhemorrhages and the shearing loss of neurons, via what is known as diffuse axonal injury, were first documented in the 1970s when Corsellis et al. (1973) published the first neuropathological study looking at the postmortem brains of 15 American boxers with documented clinical symptoms ranging from parkinsonism to changes in affect and cognition (Changa et al., 2018). Now termed chronic traumatic encephalopathy, CTE is best understood as a tauopathic disease where the accumulation of endogenous p-tau aggregates pathologically, resulting in neuronal toxicity (McKee et al., 2009, 2023a; Yi et al., 2013; Changa et al., 2018). Thus, CTE has been considered a chronic TBI neuropathological consequence.

The turn of the century proved a resurgence for CTE, where it held the spotlight following several public and tragic cases in retired, yet relatively young, NFL players. As one example, in 2017, an NFL player died by suicide at just 25 years old. His behavior was reported as uncharacteristic for a young athlete with aggressive behavioral changes that began after sustaining several football-related head injuries. Postmortem parenchymal analysis diagnosed this NFL player with CTE, showing tau depositions similar to what was documented in the American boxers studied decades earlier (Aaronson et al., 2021). In 2005 and then again in 2006, Turner et al. diagnosed the first known symptomatic NFL player with CTE (Turner et al., 2012). In 2013, McKee et al. studied the brains of 85 individuals with documented repeated head injury and symptoms of CTE. Their study found a significant association between CTE severity, and years of contact sports played (Lindsley, 2017; McKee et al., 2023a).

In addition to the classical TBI symptoms noted above, patients with CTE are known to exhibit aggressiveness, impulsivity, explosivity, emotional instability, substance use disorder, and suicidal ideation and behavior (Turner et al., 2012; Baugh et al., 2014; Asken and Bauer, 2018; Fesharaki-Zadeh, 2019; Stern et al., 2023). These symptoms progress in severity over time and some patients eventually develop severe cognitive decline including clinical AD. Of note, CTE typically develops in mid-life as opposed to late-life (Yi et al., 2013), heightening the impact of years lost.

Clinical chronic traumatic encephalopathy

The term traumatic encephalopathy syndrome is now used to describe the clinical features experienced by those with likely CTE, which can only be officially diagnosed postmortem. Classic symptoms include behavioral and personality changes, depression, anxiety, motor and cognitive changes, suicide ideation, and in severe cases, suicide (Asken et al., 2016). The most profound risk factor for CTE is repeated head impact, which may include multiple TBIs. In fact, over 97% of traumatic encephalopathy syndrome reports have a known history of head trauma (McKee et al., 2023a). Several recent studies have documented a dose-response relationship between years of contact sport played and CTE severity (Mez et al., 2020; LeClair et al., 2022). While most cases of CTE have been documented in contact sports and soldiers (Asken et al., 2017; Lindsley, 2017; Stern et al., 2023), CTE may occur due to motor vehicle accidents, violence, falls, and head-banging behaviors seen in some psychiatric conditions (Yi et al., 2013; GBD 2016 Traumatic Brain Injury and Spinal Cord Injury Collaborators, 2019). CTE is diagnosed histologically by the presence of misfolded p-tau that aggregate into oligomers and then into tau tangles or NFTs. Pathognomonic CTE p-tau tangles are distributed along the brain vasculature within the deep cortical sulci (Cherry et al., 2021; McKee et al., 2023a) with early cases showing depositions primarily in the neocortex and advanced cases showing spread to the medial temporal lobe. A recent study by McKee et al. (2023b) assessed the brains of 152 postmortem contact sport athletes with exposure to repeated head trauma. The study found that those diagnosed with CTE showed greater ventricular enlargement, cavum septum pellucidum, thalamic notching, and perivascular pigment-laden macrophage deposition in the frontal white matter. The leading cause of death in those diagnosed with CTE was suicide, followed by unintentional overdose (McKee et al., 2023b).

Alzheimer’s disease

Alzheimer’s disease epidemiology

AD, another distinct chronic neurodegenerative tauopathy, accounts for up to 75% of global dementia cases and affects roughly 5.5 million people worldwide. It describes a clinical syndrome characterized by memory and language difficulties, cognitive impairments, changes in behaviors, and a decline in the ability to carry out activities of daily living (Tahami Monfared et al., 2022). As population aging becomes more pronounced globally, the United Nations projects that approximately 1 billion elderly people globally will suffer from AD by the year 2030. In addition to the significant personal and familial burden AD places on patients, their families, and caregivers, AD presents a serious societal problem. A 2009 United States Alzheimer’s Association report estimates the annual cost to be upwards of $148 billion, citing that AD alone tripled health care costs for Americans aged 65 years and older (Lobo et al., 2000; Plassman et al., 2007). Given this impact, it is only appropriate that AD research remains on the rise.

Like CTE, AD can only definitively be diagnosed postmortem, with histological confirmation of amyloid-β (Aβ) and p-tau aggregates. The distribution of tau aggregates (Figure 2), distinguishes a CTE diagnosis from an AD diagnosis. CTE brains display perivascular p-tau depositions, particularly within the deep cortical sulci (McKee et al., 2023a). Brains with early-stage AD present with p-tau deposits within layer II of the entorhinal cortex (Shively et al., 2012; Reddy et al., 2023). As the disease progresses, toxic tau tangles propagate by way of neuronal connections, specifically along the limbic and association cortices (de Calignon et al., 2012; James et al., 2015; Michalicova et al., 2017). While natural aging also produces some degree of p-tau manifesting as NFTs, AD presents with higher densities, particularly with focal distributions affecting the hippocampus, amygdala, and the deep layers of the neocortex (Morrison et al., 1998; Turner et al., 2016; Otero-Garcia et al., 2022; Chen et al., 2023). Like CTE, AD presents with cortical atrophy and ventricular enlargement with AD neuropathology affecting the temporal lobe in the early stages of diseases with progression to frontal and parietal lobes with cortical involvement in late-stage diseases (Lemoine et al., 2017; Armstrong et al., 2019). Given that TBI is a risk factor for both CTE and AD and given their similar histopathological presentations, we propose a common mechanistic pathway governed by chronic and pathologic neuroinflammation. The next section will discuss neuroinflammation broadly, and the role of the inflammasome complex signaling pathway, in the context of TBI, CTE, and AD.

Figure 2.

Schematic of 2N4R Tau Protein Structure and the proposed epitope sites for anti-tau binding.

(A) Tau protein contains major structural domains including N-terminal domain with N1 and N2 inserts, proline-rich region, four major microtubule-binding repeats (R1–R4), anti-tau antibodies act at the step of oligomer formation from these aggregates, which are precursors to neurofibrillary tangles. (B) Monoclonal antibodies (listed) equipped with recognition sequences bind to indicated regions along the 441kd human tau protein, some at specified serine, threonine, and/or tyrosine residues, or along a range of residues along the total tau protein (epitopes corresponding to each antibody is listed in Additional Table 2). Created with BioRender.com. AD: Alzheimer’s disease; CTE: chronic traumatic encephalopathy; NFTs: neurofibrillary tangles.

Methodology Search Strategy and Database Search

This narrative review aims to provide up-to-date information on TBI, CTE, and AD within the context of neuroinflammation. A mechanism implicating the inflammasome was proposed to link these three neuropathologies in a time-dependent manner. A review of the literature was conducted between August 5, 2008 and December 12, 2023. Using the PubMed search engine, key words including “inflammasome,” “NLRP1,” “NLRP2,” “NLRP3,” “AIM2,” “ASC,” “PYRIN,” “NLRC4,” “pyroptosis,” “traumatic brain injury”, “chronic traumatic encephalopathy,” “Alzheimer’s disease” and “neuroinflammation” were utilized to generate relevant peer-reviewed data. Initially, studies were limited to include experimental studies and review articles published in or after the year 2018. The search was later expanded to include studies dating back to 2008 to gain perspective on how the field has changed over time. In addition to using the PubMed search engine, author references were used to confirm the accuracy of data and to provide further clarification on concepts described in original papers.

The main search words: (“brain injuries, traumatic”[MeSH Terms] OR (“brain”[All Fields] AND “injuries”[All Fields] AND “traumatic”[All Fields]) OR “traumatic brain injuries”[All Fields] OR (“traumatic”[All Fields] AND “brain”[All Fields] AND “injury”[All Fields]) OR “traumatic brain injury”[All Fields]) AND (“chronic traumatic encephalopathy”[All Fields] OR “CTE”[MeSH Terms] OR (“encephalopathy”[MeSH Terms] OR “Alzheimer’s”[All Fields] OR “Alzheimer’s disease”[All Fields] OR “inflammation”[All Fields]) OR “Boxer’s Dementia”[All Fields]) OR (“neuroinflammatory diseases”[MeSH Terms] OR (“neuroinflammatory”[All Fields] AND “diseases”[All Fields]) OR “neuroinflammatory diseases”[All Fields] OR “neuroinflammation”[All Fields])) OR (“brain injuries, traumatic”[MeSH Terms] OR (“brain”[All Fields] AND “injuries”[All Fields] AND “traumatic”[All Fields]) OR “traumatic brain injuries”[All Fields] OR (“traumatic”[All Fields] AND “brain”[All Fields] AND “injury”[All Fields]) OR “traumatic brain injury”[All Fields] OR (“neurodegenerative diseases”[MeSH Terms] OR (“neurodegenerative”[All Fields] AND “diseases”[All Fields]) OR “neurodegenerative diseases”[All Fields]) OR (“central nervous system diseases”[MeSH Terms] OR (“central”[All Fields] AND “nervous”[All Fields] AND “system”[All Fields] AND “diseases”[All Fields]) OR “central nervous system diseases”[All Fields] OR (“cns”[All Fields] AND “diseases”[All Fields]) OR “cns diseases”[All Fields])) AND (“inflammasomes”[MeSH Terms] OR “inflammasomes”[All Fields] OR “inflammasome”[All Fields] OR “NLRP3”[All Fields] OR “nlr family, pyrin domain containing 3 protein”[MeSH Terms] OR “NLRP”[All Fields] OR “Caspase-1”[All Fields] OR “Il-18”[All Fields] OR “IL-1β”[All Fields] OR (“pyroptosis”[MeSH Terms] OR “pyroptosis”[All Fields])).

Misfolded Proteins and Neurodegeneration

Amyloid-β protein

The cardinal histological signs of AD include an abundance of misconfigured Aβ presenting as extracellular senile plaques along with intracellular NFTs composed of misfolded p-tau aggregates (Williams, 2006; Zotova et al., 2013; Yamazaki et al., 2019; Hefter et al., 2020). Misfolded Aβ comes from the endogenous amyloid precursor protein (APP), a large, multimeric transmembrane protein unit (Hefter et al., 2020). In physiologic conditions, APP is cleaved by α-secretase and the downstream APP-α unit goes on to function in synaptic activity and plasticity. However, in a pathologic state, APP may be cleaved by β-secretase, resulting in APP-β, which gets misfolded into a toxic Aβ. Aβ translocates to the extracellular space where it may aggregate into neurotoxic oligomers and form senile plaque (Ricciarelli and Fedele, 2017). Extracellular Aβ aggregates also induce the formation of p-tau oligomers and have been shown to induce tau seeding, accelerating disease progression (Gomes et al., 2019). This model, by which Aβ initiates tauopathy is known as the amyloid cascade hypothesis and has traditionally been as the linchpin of AD pathology (Ricciarelli and Fedele, 2017).

The classical amyloid cascade hypothesis posits a linear relationship in which Aβ is an inciting event for the hyperphosphorylation of tau. While several genetic studies demonstrate that the removal of Aβ plaques is largely curative of AD in mice models, these results have not been translated to clinical trials. Despite high expectations, many of the clinical pharmaceutical trials targeting the amyloid cascade, by way of APP, Aβ, α-secretase, β-secretase, and Aβ solubility, have failed (Asher and Priefer, 2022). Only recently have two humanized monoclonal antibodies targeting Aβ been Food and Drug Administration-approved for patients with symptomatic AD (Ramanan and Day, 2023). The dearth of translational outcomes for AD patients in conjunction with emerging evidence on tau pathogenesis suggests a more complex model than proposed in the amyloid cascade hypothesis. New research proposes a synergistic relationship between Aβ and tau in the pathogenesis of AD (Busche and Hyman, 2020), providing a possible explanation as to why targeting only Aβ for AD therapeutics has thus far yielded null to limited results. Aβ-tau synergy is described in greater detail below.

Tau protein

Tau protein is an endogenous, relatively hydrophilic protein with 6 isoforms. It is abundantly present in neurons and less prominent in astrocytes and oligodendrocytes with its role being less studied in these cells (Williams, 2006; Wang and Mandelkow, 2012; Kent et al., 2020; Medina and Avila, 2020). The protein, in its 6 isoforms, functions to stabilize microtubules and has an important role in cell communication, plasticity, and genome regulation (Wang et al., 2015; Guo et al., 2016; Strang et al., 2019; Xia et al., 2019). Tau isoforms range from 37–46 kDa and are derived from alternative splicing of the microtubule-associated protein tau gene on chromosome 17 (Hanes et al., 2009; Zhang et al., 2009a; Schweighauser et al., 2023). Tau post-translation-modifications are developmentally regulated within the human fetal brain expressing only the 0N3R and the adult human brain expressing all the six isoforms (Guo et al., 2016).

The longest human tau isoform found in the human central nervous system (CNS) (2N4R) has 441 amino acids. It has 4 major domains including an acidic N-terminus, a proline-rich domain, a microtubule-binding domain, and a C-terminal tail (Mandelkow et al., 1995; Mukrasch et al., 2009; Avila et al., 2016). Due to its relative hydrophilicity, tau exists in a natively unfolded and highly flexible configuration with a poorly defined secondary structure (Jeganathan et al., 2008; Mukrasch et al., 2009; Guo et al., 2016). Despite this structure, its many charged domains allow room for secondary folding to occur under specific biochemical circumstances. Normally, tau binds to the tubulin heterodimers of microtubules to give structural stability along the neuronal axon. This binding occurs along the microtubule-binding domain where four imperfectly repeating motifs act as the primary binding sites, whereas the dividing flank regions play a role in binding regulation. The amino acid variation found amongst the imperfectly repeating motifs allows for selective binding. The amino acid variation found amongst the imperfectly repeating motifs allows for selective binding. Tau binds to F-actin (Correas et al., 1990), α-synuclein (Jensen et al., 1999), histone deacetylase-6 (Ding and Johnson, 2008), apolipoprotein E (Huang et al., 1995), presenilin-1 (Takashima et al., 1998), and actin (Elie et al., 2015). The affinity of tau for actin and microtubules provides a tethering system that brings stability and plays a role in maintaining healthy synapses throughout the brain (Guo et al., 2016).

Amyloid-β & tau synergy

While the classical amyloid cascade hypothesis importantly identifies Aβ as a key inciting event for AD, new research strongly suggests a critical role for p-tau oligomers in disease spread (Mandelkow and Mandelkow, 2012; Shi et al., 2017; Reddy et al., 2023). Bushe and Hyman (2020) propose a synergistic, rather than linear model, for understanding AD which may aid in our understanding of tauopathies more broadly. The authors have identified a useful framework in which AD can be understood in two phases: an Aβ-dependent phase, which dictates disease onset; and an Aβ-independent phase, which describes disease progression. The Aβ dependent phase relies on data garnered from familial AD, in which individuals with a genetic predisposition to AD (e.g., those with mutations in Preselin-1/2; APOE4, APP and those with Trisomy 21) begin accumulating Aβ early on in life and subsequently, develop AD much earlier on in life. Interestingly, the rate of disease progression is comparable to those with sporadic AD, suggesting that Aβ may play less of a role in disease progression, and as such, anti-Aβ therapies may be less effective during this secondary, Aβ-independent phase (Busche and Hyman, 2020). Experimental models utilizing 3×Tg-AD mice that develop both Aβ plaques and tau tangles show decreased development of tau tangles early in the disease. No changes, however, were observed later on in the disease, after initial tangles had already been formed (Busche and Hyman, 2020). Bennett et al. (2017) also demonstrated age-dependent interactions between and tau. They showed that Aβ increases the formation of tau seeds capable of propagating new tau aggregates. They saw this increase early in the disease (e.g., 4 months) and not in later stage disease (e.g., 12 months). Using APP/PS1-rTg4510 transgenic mice, a threefold increase was detected in tau seeding even before amyloid deposition was observed (Bennett et al., 2017) but not after tangles had already formed. While the details of this synergy still need to be clarified, several experimental studies have corroborated these early findings and a clearer understanding may provide insight into more efficient therapeutics for AD (Di Micco et al., 2016; Gomes et al., 2019).

Tau Phosphorylation

Tau protein can undergo several post-translation-modifications including acetylation, methylation, glycosylation, and ubiquitination (Qi et al., 2017; Xia et al., 2021; Stathas et al., 2022). By far the most examined enzymatic modification is phosphorylation (Noble et al., 2013; Wegmann et al., 2021; Chen et al., 2023) via kinase activity. Tau contains a total of 85 phosphorylation sites (45 serine, 35 threonine, 5 tyrosine) (Goedert et al., 1989), of which, when bound, can create chemical instability, reduced microtubule affinity, and cytoskeleton destabilization. The following sites have been shown to induce neurotoxicity in vitro and may lead to the development of pathologic filamentous tau: Thr175, Ser202, Thr205, and Ser422 (Rankin et al., 2005; Chang et al., 2011; Liu et al., 2016; Xia et al., 2021). Other sites, including Ser202/Thr205/Ser208, allow for triple phosphorylation, which is proposed to result in more rapid tau aggregation (Despres et al., 2017; Xia et al., 2020, 2021). Interestingly, phosphorylation at sites Ser214, Ser262, and Ser305 has been shown to inhibit tau aggregation in culture and via monoclonal antibody testing studies, demonstrating selective neuroprotective effects (Schneider et al., 1999; Zhang et al., 2009b; Strang et al., 2019). There is still much to uncover regarding tau phosphorylation, but current data overwhelmingly points towards hyperphosphorylation as a causal proponent of neurodegenerative pathology.

Tau hyperphosphorylation is known to induce tau aggregation (Thom et al., 2011; Putra et al., 2021). It was previously believed that NFTs comprised of p-tau were the causal source of pathology. However, more recent evidence indicates that low molecular weight oligomers, which aggregate in the early stages of pathology prior to NFT formation, are the causal proponents of neurodegeneration, inducing mitochondrial, synaptic, and memory deficits (Lasagna-Reeves et al., 2014; Hill et al., 2020; Cherry et al., 2021; Stathas et al., 2022). When p-tau detaches from microtubules, it mis-localizes from the axon to the cytosol where it accumulates into oligomers (Zhang et al., 2021). Oligomer initiation has been proposed to occur via two mechanisms: a cysteine-dependent polymerization that occurs by way of disulfide bonds (Schweers et al., 1995; Bhattacharya et al., 2001); and a second, cysteine-independent polymerization, occurring by way of hexapeptide motifs and non-covalent bonds (Goux et al., 2004; von Bergen et al., 2005). These interactions lead to a conformational change from one of relatively few secondary interactions to a disordered conformation rich in β-sheets that display minimal ellipticity when compared to their monomeric, unfolded counterparts (Friedhoff et al., 1998; Congdon et al., 2008; Lasagna-Reeves et al., 2010; Shafiei et al., 2017). These soluble tau oligomers mark the first steps in a pathological cascade that leads first to fibrils and then to NFTs (Avila et al., 2016; Hill et al., 2020). As noted above, p-tau oligomers and tangles are present in TBI, CTE, and AD, yet their distribution across the brain varies by pathology (Franzmeier et al., 2022).

Phosphorylated tau isoforms, distribution, and therapeutic targets

Tau aggregation, deposition, and spread are the hallmark features of tauopathies, including CTE and AD. It is also a prominent feature in TBI, which is a leading risk factor for tauopathy. An important distinguishing feature between TBI, CTE, and AD is the location and structure of such tau depositions (Chen et al., 2023). It is hypothesized that both the temporal and macro- and micro-spatial deposition within the brain as well as the degree of propagation of the tau accumulation may account for the different clinical symptoms that distinguish such diseases (Additional Table 1). This may be particularly pertinent when there are repetitive injuries. While the distribution of pathological tau in TBI varies widely, depositions are generally seen focally, surrounding the area of immediate injury (de Calignon et al., 2012; Liu et al., 2012; Ayers and Cashman, 2018). These depositions spread over time to adjacent or even contralateral brain regions (Asai et al., 2015; Goedert and Spillantini, 2017; Stancu et al., 2019a; Polanco et al., 2023). Conversely, CTE and AD have relatively well-defined distributions (Stathas et al., 2022; McKee et al., 2023a).

Additional Table 1.

Comparison of pathological features and presentations of Alzheimer’s disease, traumatic brain injury, and chronic traumatic encephalopathy

	Traumatic brain injury (TBI)	Chronic traumatic encephalopathy (CTE)	Alzheimer’s disease (AD)	References
Cause	Acceleration and deceleration forces, and/or blast waves on the head (i.e., penetrating, or non-penetrating trauma from falls, motor vehicle accidents, explosions, blasts, or physical assault).	Repeated TBI is the most well- documented risk factor	Multifactorial, risk factors include genetic predisposition and TBI.	Collins-Praino and Corrigan, 2017; McKee et al., 2023b; McKee et al., 2023a; Stein et al., 2014
Average age of onset (year)	Age range from 15-24 and 75+	Younger patients, ages 30+	Older patients, ages 65+	CDC, 2023
Clinical symptoms	Depression, increased risk-taking behavior, poor sleep stability, and cognitive defects.	Personality changes, memory loss, confusion, impaired judgment, loss of impulse control, aggression, depression, anxiety, and suicidality.	Memory loss, disorientation, impaired judgment, loss of motivation, mood and personality changes, and increased anxiety.	Stein et al., 2014; Tolnay and Probst, 1999
Diagnostic tool	Classified via the Glasgow Coma Scale as mild (14-15), moderate (9-13) or severe (3-8). Post-mortem histological confirmation of p-tau and NFTs in hippocampal and superficial cortical regions.	Diagnosed histologically (postmortem) by the presence of misfolded p-tau that aggregates into oligomers and then into NFTs. Clinically diagnosed as TES.	Diagnosed (post-mortem) by histological confirmation of Αβ and p-tau aggregates,	Braak and Braak, 1991; Cherry et al., 2021; McKee et al., 2023a; Stein et al., 2014
Pattern of tau deposition	Pathological tau distribution is focal. Spreads over time to adjacent and/or contralateral brain regions	Perivascular tau distribution in depths of cortical sulci. NFTS is initially seen in the superficial cortical layers (layers II and III), at the base of sulci and surrounding blood vessels. NFTs extend through the cerebral cortex. The hippocampus and entorhinal cortex affected in late-stage disease	First observed in the transentorhinal region (stages I & II), spreads to the limbic regions in stages III and IV, then extends uniformly throughout neocortex. More uniform cortical distribution than in CTE, with the involvement of layers V and VI of the cortex	Braak and Braak, 1991; Cherry et al., 2021; McKee et al., 2023a; Tolnay and Probst, 1999
Pattern of brain atrophy	Initially focal. Spreading occurs over time and includes compromise of the blood- brain-barrier. Widening of sulci, ventricular enlargement, cortical thinning, and hippocampal shrinkage.	Atrophy is pronounced in the frontal, temporal, and medial lobes. Atrophy of white matter, particularly corpus callosum, with thinning of the hypothalamic floor.	Marked atrophy of the medial temporal lobe, including the hippocampus, entorhinal cortex, and amygdala; eventual involvement of other cortical areas, including the parietal and frontal lobes.	Cherry et al., 2021; Collins- Praino and Corrigan, 2017; Tolnay andProbst, 1999

†Data gathered from Centers for Disease Control and Prevention Mortality Data on CDC WONDER. Table adapted from Collins-Praino and Corrigan (2017).

A recent study by Cherry et al. (2021) used multiplex immunofluorescence analysis to measure 3R and 4R tau isoforms in the postmortem brains of 41 diagnosed CTE cases and 50 diagnosed AD cases. Comparing relative levels across hippocampal subfields CA4, CA2/3, CA1, and the subiculum, as well as the temporal cortex grey matter and white matter, high levels of tau pathology diversity between the CTE and AD were observed. The authors documented increased 4R in the early stages of the disease and increased 3R in later stages. They found that CTE-diagnosed brains had increased total tau and increased p-tau in the CA2/3 and CA4 regions (Cherry et al., 2021). Conversely, AD cases showed lower overall tau levels and higher p-tau densities in the CA1 and subiculum regions. The authors also looked at correlations between tau and Aβ densities and concluded that Aβ density did not influence the deposition of tau aggregates. It is important to note that tau isoforms exist in a dynamic, rather than static state (Cherry et al., 2021; McKee et al., 2023a). As such, the mechanisms by which tau changes are induced may drive the development of AD from CTE. Thus, understanding the morphology and the factors that influence tau transformation and spread is critical to understanding how tau depositions in a post-TBI context may develop into CTE and AD. Recent research on tauopathy has provided a better understanding of the movement of tau across the brain.

Of interest, based on the biochemical-pathological characteristics of tau/p-tau protein, tau-targeted therapy has been an attractive approach for research and clinical trials in treating taupathic-related disorders involving AD and TBI among others (Jadhav et al., 2019). Approximately, more than 30 tau-related drugs have been clinically tested including microtubule stabilizers, glycogen synthase kinase-3β inhibitors, acetylation inhibitors, anti-tau monoclonal antibodies, and anti-tau active vaccines (Imbimbo et al., 2023). Among these approaches, immunotherapy via tau humanized monoclonal antibodies targeting different tau species, aiming for tau and phosphorylated tau aggregate clearance, has been assessed in several clinical trials (Ji and Sigurdsson, 2021). These antibodies have different binding epitopes targeting the amino-terminus carboxy-terminus, proline-rich area, or microtubule-binding domains (Congdon and Sigurdsson, 2018; Cummings et al., 2023). This therapeutic approach leverages the immune system’s ability to identify and eliminate different tau species, potentially slowing the progression of tau-associated neurodegeneration (Congdon et al., 2014). However, data show that these pharmacological approaches have limited clinical success in patients. For example, recent clinical trials have highlighted both the potential and the limitations of anti-tau antibodies. The phase 2 trial of Semorinemab antibody in 457 subjects with early AD did not show a significant impact on tangle accumulation as measured by tau positron emission tomography (PET) scans, despite some indications of plasma tau increase, suggesting some interaction with tau pathology (Teng et al., 2022). Another antibody, Gosuranemab, targeted the N-terminal region of tau, and failed to show efficacy in reducing tau pathology or slowing cognitive decline in a large phase 2 trial involving 654 subjects (Shulman et al., 2023). Interestingly, a trial of Tilavonemab involved 378 subjects with progressive supranuclear palsy showed no benefit, and the study was terminated early (Hoglinger et al., 2021). These examples underscore the complexities of tau pathology and the need for further research into the mechanisms of tau aggregation and propagation, as well as better trial designs and more sensitive clinical measures to assess the efficacy of these treatments (Imbimbo et al., 2023). A detailed list of the current anti-tau treatments with their clinical studies and their corresponding epitopes are summarized in Additional Table 2 and illustrated in Figure 3 (Ji and Sigurdsson, 2021; Xia et al., 2021; Imbimbo et al., 2023).

Additional Table 2.

List of anti-tau specific treatments, indication, clinical trials and their corresponding epitopes

Name/synonyms	Indication	Clinical trial	Phase	Epitope binding and mechanism
JNJ-63733657, (B296, PT3)	AD, dementia	NCT03689153 /NCT04619420	I/II	mAB that targets pThr217 (204-225) & (pT212/pT217)
PNT001	Patients with TBI	NCT04096287	I	Antibody against Cis isomer of pThr231
Lu AF87908	Healthy, AD	NCT04149860	I	Antibody against pSer396
RO6926496, (RG7345)	Healthy adults	NCT02281786	I	Antibody against pSer422 (416–430)
BIIB092/Gosuranemab, IPN007	Mild AD/MCI/PSP	NCT03352557/ NCT03068468	I/II	h-monoclonal IgG4 antibody targets extracellular N-terminal tau fragments (eTau). It targets tau aa 15–22
Aducanumab/BIIB076	AD/Healthy participants	NCT03056729	1	PAN- h-monoclonal anti-tau IgG1 antibody. It recognizes the middomain of tau. It targets monomeric and fibrillar forms of tau
LY3303560/ Zagotenemab	Mild-Mod. AD	NCT02754830/ NCT03518073	I/II	h-anti-tau antibody. It targets conformational epitope of tau. It binds aa 7–9 and 313–322 (Neutralizes soluble tau aggregates); probably 125–131
Bepranemab/UCBO 107/an tibody D	Mild MCI and mild AD/ Healthy participants/PSP	NCT04658199/NCT 04867616	I/II	h-monoclonal IgG4 antibody that targets amino acids 235-246 of tau. It binds mid-region central of tau (~ 100% inhibition of fibrillization at an antibody concentration of 300 nM)
E2814	Healthy/ Mild to Moderate MCI	NCT04231513/ NCT04971733	I	h-monoclonal IgG1 antibody recognizes an HVPGG epitope in the microtubule-binding domain in second (aa299–303) and fourth (aa362–366) repeat of the microtubule binding domain in 2N4R-tau
Semorinemab/RO7105705 /MTAU9937A/RG6100	AD with +ve Aβ PET or high cerebrospinal fluid AP42 levels	NCT03289143/ NCT03828747	I/II	IgG4 antibody that targets extracellular tau and has reduced effector function like all IgG4 antibodies. It binds N-terminus of all six tau isoforms (binds both monomeric and oligomeric tau)-binds aa 2-24
CN2-8E12 ABBV-8E12/ Tilavonemab /HJ9.3	AD/PSP	NCT02880956/NCT 03744546/ NCT03712787	I/II	h-IgG4, targets aggregated extracellular tau forms at aa 25-30 at the N- terminal sequence of tau (Does not require uptake into neurons)

Anti-Tau antibodies (with their proposed mechanisms and their corresponding phosphor-tau ortotal tau epitopes [adapted from Jadhav et al., 2019]. AP: Amyloid-β; AD: Alzheimer’s disease; MCI: mild cognitive impairment; PSP: progressive supranuclear palsy.

Figure 3.

Brain atrophy tau/p-tau distribution in CTE and AD.

Brain atrophy occurs in both CTE and AD and includes shrinking of functional parenchyma, depositions of misfolded tau protein, and ventricular enlargement. CTE is pathognomonic for p-tau distribution along the periphery of cerebral vessels and within the depths of cortical sulci. Atrophy typically begins with corpus callosum and other white matter structures and can progress to frontal and medial temporal lobe atrophy. AD is characterized by neurotoxic p-tau distributions in the frontal cortex and the CA1 region of the hippocampus. Tau pre-tangles and NFTs are observed in the cortical layers with resultant atrophy of temporal lobe structures. Created with BioRender.com. AD: Alzheimer’s disease; CTE: chronic traumatic encephalopathy; NFTs: neurofibrillary tangles; p-tau: phosphorylated tau; TBI: traumatic brain injury.

Traumatic brain injury and tau spreading

Current research posits two main mechanisms by which p-tau that originates in one region of the brain migrates to adjacent or even contralateral brain regions (Franzmeier et al., 2022; Chen et al., 2023). Important to note is that while these two mechanisms are distinct, they are not mutually exclusive. The first, which is well established, involves microglial phagocytosis and as such, is termed microglial spreading (Asai et al., 2015; Laurent et al., 2018). Microglia are resident macrophages and make up one of the two types of glial cells in the brain. They exist in a resting state where they surveil the brain parenchyma and become activated when danger signals are encountered in their environment. While more details on this process are provided in the section on neuroinflammation below, microglia are a key component to both the recycling and spreading of p-tau (Chen et al., 2023). As a CNS macrophage, microglia may phagocytize tau and secrete the toxic protein to neighboring neurons via exosomes (Chen et al., 2023). While the physiological function of phagocytosis is to clear pathogens and unwanted debris, microglia have been shown to perpetuate neurotoxicity in this manner, creating p-tau depositions along the pathway of neuronal circuits (Asai et al., 2015).

A second theory on how neurotoxic tau is spread to neighboring brain regions exists in concert with, rather than in opposition to, microglial spreading. As described in Figure 4, soluble tau oligomers are responsible for tau seeding and propagation along a pathway that follows neural synaptic connections by way of prion-like spreading (de Calignon et al., 2012; Liu et al., 2012; Stancu et al., 2015, 2019a; Ayers and Cashman, 2018; Kundel et al., 2018; Chen et al., 2023). A protein “seed” refers to an initiating protein that, upon misfolding and misconfiguration, can “infect” and propagate other tau entities found in neighboring neurons (Vasili et al., 2019). Protein seeds have traditionally been studied in prion diseases including the classic example of Creutzfeldt-Jakob disease. While tau is not considered a true prion protein, more and more evidence is surfacing that strongly suggests its ability to propagate via this mechanism. Tau seeds exist as either “naked tau” (vesicle-free oligomers or fibrils) or exosomal tau, encapsulated in an exosomal membrane. Several in vitro and in vivo studies demonstrate the plausibility of the tau seeding and propagation model, particularly in the context of tau oligomers (Lasagna-Reeves et al., 2015; DeVos et al., 2018; Stancu et al., 2019a; Vasili et al., 2019; Hussong et al., 2023; Polanco et al., 2023). Further studies show that TBI can initiate the development of the tau seeds; hence, its position as a risk factor for tauopathy (Clavaguera et al., 2009; Hawkins et al., 2013; Iba et al., 2013; Furman et al., 2017; Mudher et al., 2017; DeVos et al., 2018; Stancu et al., 2019a).

Figure 4.

Hyperphosphorylation of endogenous tau protein leads to microtubule instability and protein misfolding that results in the formation of neurotoxic tau aggregates, prolonging pathological neuroinflammation.

Tau is an endogenous microtubule-associated protein that functions to maintain cytoskeletal stability and structure. When kinase-mediated hyperphosphorylation occurs, tau protein detaches from microtubules and impairs cytoskeletal stability. Resulting in tau monomers losing their physiologically relaxed conformation and misfolding into pathological β-pleated sheets due to electrical charges generated by phosphorylation. Protein misfolding causes tau monomers to aggregate into tau oligomers, which can further aggregate into neurofibrillary tangles, ultimately leading to pyroptosis or cellular death by pore formation. Additionally, new research shows evidence for the parallel ability of tau oligomers to generate prion-like tau seeds, which propagate pathologically, causing the spread of neurotoxic tau to adjacent and even contralateral brain regions. Tau oligomers have been shown to activate pro-inflammatory mediators including interleukin-1β (IL-1β), that may further lead to physiological tau hyperphosphorylation and result in neurodegeneration. This figure depicts a cyclical pathway by which aberrant neuroinflammation is prolonged as is seen in traumatic brain injury, chronic traumatic encephalopathy, and Alzheimer’s disease. Created with BioRender.com.

Experiments by Hawkins et al. (2013) documented rapid accumulation and spreading of endogenous oligomeric tau in post-TBI rats modeled via parasagittal fluid percussion injury. Using AT8 and AT22 antibodies, they noted oligomeric tau development 4 hours following injury. These levels remained elevated up to 2 weeks post-injury. Sham rats, in comparison, showed no increase in total tau or tau tangles. Toxic p-tau aggregates were observed in the cortex and hippocampus, both ipsilateral and contralateral to the initial site of injury, indicating oligomeric spread following injury, across the midline of the brain. Newer research by Amorim et al. (2023) corroborates these findings using a seeding-based neuronal model in cortical human tau neurons from 9-month-old P301S mice. Cells incubated with 2N4R tau displayed a homogenous increase in tau that was not observed when the same cultures were incubated with P301S lysates alone. They concluded that the addition of P301S seeds is required to induce tau accumulation. Using a high-content imaging analysis algorithm, they quantified the density of tau and differentiated seeded from non-seeded conditions (Amorim et al., 2023). Finally, using an anti-aggregation compound, Anle 138b, they minimized tau spreading by selective inhibition of tau oligomerization, indicating that tau seeding requires a specific aggregational conformation for the spreading step.

An in vivo study by Mate De Gerando et al. (2023) show that mice injected with human tau isolates resulted in progressive tauopathy and regional spreading. Mice injected with senescence-associated receptor-like kinase (SARK) fibrillar tau or high-molecular-weight oligomeric tau in the dorsal hippocampus demonstrated an eightfold increase in AT8 and AT100 anti-tau staining when compared to PBS-injected mice that showed nearly no tau staining. When postmortem brains were studied, both experimental arms demonstrated the spontaneous appearance of tau pathology 3 months post-injection. SARK and high-molecular-weight mice also gave rise to tau propagation along adjacent brain regions, including the peri and entorhinal cortices, which are connected to the hippocampus via synapses. Induction with tau oligomers was also observed to accelerate tau pathology maturation in SARK and high-molecular-weight-induced mice, as observed by an increase in ThioS-positive neurons at 3 months post-injection. A better understanding of how aberrant tau proteins propagate, and spread is critical for identifying interventions in a post-TBI state. Several experimental models provide evidence that tau seeds (initiators of tauopathy) do occur post-TBI (Mate De Gerando et al., 2023).

Traumatic brain injury and tau accumulation

Several experimental models provide evidence that exposure to head trauma leads to pathologic p-tau deposition (Hoshino et al., 1998; Kane et al., 2012; Hawkins et al., 2013; Petraglia et al., 2014; McAteer et al., 2016; Collins-Praino and Corrigan, 2017; Lumba-Brown et al., 2018; Edwards et al., 2020; Jamjoom et al., 2021). Edwards et al. (2020) created a special heatmap approach to trace the spread of neurotoxic tau in TBI-induced mice and compared it to age-matched sham mice. The authors found increased pathologic tau, particularly in the ipsilateral cortex, hippocampus, brain stem, and overall cortex. The authors reported increased tau pathology in TBI-induced compared to sham mice as well as the spread of tau along ipsilateral brain regions. One-week post-impact, immunohistochemistry and heat map analysis showed increased tau deposition in the contralateral cortex and increased tau in the hippocampus and brainstem. At 1–2 months following injury, TBI mice presented increased insoluble tau in the cortex, brainstem, and ipsilateral hippocampus. At 6 months post-injury, TBI mice showed dramatic tau deposition compared to control mice, and NFT-like structure formation was seen profusely throughout the entorhinal cortex and pyramidal cells of the CA1 region. At 6 months, the areas of greatest tau deposition included the amygdaloid nucleus, the entorhinal cortex, the pons, and the medulla. They demonstrated clinical correlates by testing cognitive and behavioral effects, reporting deficits in both fields at 6 months post-injury (Edwards et al., 2020).

Years earlier, Petraglia et al. (2014) reported an increase in p-tau density after just a single TBI using a closed head injury model on 12-week-old mice. Changes were noted in the contralateral cortex, bilateral amygdala as well as the ipsilateral CA1 and CA3 regions. The same study found that repeated exposure to TBI led to further increased p-tau immunoreactivity at 7 days, 1, and 6 months post-injury, indicating that repeated head injuries increases tau density. Repeated TBI (rTBI) was found to induce a significant increase in the phosphorylation of tau at 7 days post-injury when compared to control and single TBI. They documented both ipsilateral and contralateral accumulations, particularly in the cortex, amygdala, dentate gyrus, CA1 and CA3 areas. Post-concussive behavioral abnormalities correlated with increased tau density in mice exposed to a single tau while repeated head injury was associated with more neurobehavioral symptoms including depression, increased risk-taking behavior, poor sleep stability, and cognitive deficits. These findings corroborate the clinical presentations that differentiate TBI, which often occurs in the context of one injurious event, from CTE, which occurs in the context of repeated TBI. Petraglia et al. (2014) concluded that increased exposure to head injury leads to rapidly progressing symptoms that begin as TBI but lead to CTE if TBI recurs repeatedly. This study lends plausibility to the theory that tau accumulation by way of repeated head injury drives the development of TBI into CTE.

Another study by McAteer et al. (2016) sought to develop a model that could reflect the progression of TBI to CTE. Using a modified rat Marmarou impact-acceleration diffuse-TBI model, they delivered 110G of linear force to simulate mild TBI 1 or 3 times spaced 5 days apart. When compared to rats that received no TBI, the authors reported increased p-tau after 24 hours in both single TBI and rTBI groups. Functionally, rats exposed to a single TBI and 3× rTBI showed increased maze latency at 6 weeks post-injury, with 3× rTBI rats exhibiting maze latency that persisted 12 weeks post-injury. The findings by McAteer et al. (2016) further bolster the theory that while a single TBI may lead to tau hyperphosphorylation, exposure to repeated injury lends itself to the development of long-term neurodegeneration induced by tau aggregation and spreading. Importantly, head injury, both single and repeated, occurs in the context of neuroinflammation, a complex and dynamic process that allows for the progression of symptoms at least in part by way of inflammasome and activation.

Interestingly, Zanier et al. (2018) characterized tau self-spreading and the generation of authentic tau prions following a single TBI. Using antibody staining for pathologically phosphorylated tau, the human paired helical filaments (PHF1-Tau), they compared samples from patients who had survived one moderate or one severe TBI to patients with no TBI and no history of neurological diseases. While some p-tau was present in controls, p-tau extent and distribution were significantly greater in TBI survivors with densities most prominent in the superficial cortical layers and hippocampus. Importantly, one-quarter of the TBI samples (3 of the 12 cases) exhibited “CTE-like” tauopathy with p-tau depositions occurring in clusters surrounding cortical vessels and spreading beyond the entorhinal cortex into hippocampal grey matter (Zanier et al., 2018). Zanier et al. (2018) used the controlled cortical impact model on mice to demonstrate that a single TBI can induce tauopathy and tauopathic changes increase over time. Accordingly, 2 of the 4 TBI mice exhibited ipsilateral p-tau immunoreactivity (assessed by AT8, AT180, and PHF1 antibodies) at 3 months post-injury. At 12 months, p-tau spreading occurred in the ispi- and contralateral brain regions. The experimenters showed evidence of the generation of novel tau prions and horizontal prion-like spreading following TBI. Zanier et al. (2018) inoculated 10% brain homogenates from TBI and sham-injured mice into the hippocampus and overlaying cerebral cortex of naive mice. Twelve months following inoculation, p-tau levels were quantified using Western blot analysis and AT8 staining, showing a significant increase in p-tau accumulation in the TBI group compared to sham. Furthermore, overt memory deficits were confirmed in TBI-inoculated mice that corresponded to hippocampal synaptic pathology (Zanier et al., 2018). Finally, in a follow-up experiment, they analyzed the effect on motor function in C. elegans and found that worms inoculated with oligomeric tau homogenates exhibit impaired motility (Zanier et al., 2021); hence, concluding that widespread CTE-like pathology is present in TBI survivors and establishing a direct connection between TBI-induced tau pathology and neurodegenerative symptomatology.

Neuroinflammation

Inflammation is a normal part of the body’s response to warning signals. It leads to the release of mediators such as cytokines (Stamouli and Politis, 2016; Bradburn et al., 2019; Foley, 2023; Karvandi et al., 2023; Kiraly et al., 2023) and chemokines (Heneka et al., 2015; Businaro et al., 2018; Skaper et al., 2018) that activate different pathways responsible for destroying pathogens and clearing out unwanted debris. Physiologic inflammation is a transient process that is turned off when the threat has been addressed and concludes to restore the body’s homeostatic balance (Onyango et al., 2021). Chronic inflammation is pathologic and involves prolonged misfiring of immune cells when there is no threat to be addressed. Neuroinflammation, or inflammation that occurs within the relatively sequestered brain and spinal cord, is common to many neurodegenerative diseases, including CTE and AD (Heppner et al., 2015; Singh, 2022; Zhang et al., 2023). Key cells involved in neurodegenerative inflammation include glial cells, which are comprised of microglia and astrocytes (Norden et al., 2016; Singh, 2022; Zhang et al., 2023). These cells communicate with each other as well as with neurons and other CNS cells to mediate inflammatory processes.

Microglia cells

Microglia are resident CNS cells derived from the monocyte lineage and as such, have a phagocytic role in the brain (Stoll et al., 2002; Felsky et al., 2019). They have a small cell body with highly motile and thin, ramified processes. In their resting state, microglia express a low level of surface antigens (Lee et al., 2014). They are plentiful throughout the brain and spinal cord and form the frontline defense for CNS immunity. Microglia exist in a constant state of immune surveillance, protecting the brain parenchyma and maintaining neuronal synapses by way of complement-dependent synapse elimination (Lee et al., 2014; Laurent et al., 2018; Puntambekar et al., 2018; Singh, 2022; Amanollahi et al., 2023). When microglia encounter a warning sign such as a pathogen-associated molecular pattern (PAMP), damage-associated molecular pattern (DAMP), or TNF from helper T cells, they become activated (Lull et al., 2010; Shabab et al., 2017; Amanollahi et al., 2023).

Chronic neuroinflammation occurs in the context of proteinopathy such as the tau deposition seen in CTE and AD. Several studies have demonstrated that microglia cells become activated following TBI. The process of microglial activation therefore represents a probable pathway for which TBI can develop into CTE and AD (Hou et al., 2022). The secondary phase gives rise to tau-hyperphosphorylation and the development of p-tau aggregates that lead to neurodegeneration over time (Shi et al., 2017; Graham and Sharp, 2019; Moreno-Garcia et al., 2021).

Several animal models dating back to 1929 reported the accumulation of activated microglia following head injury (Carmichael, 1929; Dunning and Stevenson, 1934). Recent studies corroborate these findings using more sophisticated models of induced head injury such as weight drop, the controlled cortical impact, fluid percussion, and blast brain injury models (Tran et al., 2011; Kokiko-Cochran et al., 2018; Kahriman et al., 2021). Meanwhile, more advanced technologies including PET and magnetic resonance imaging have been developed to identify microglial activation and subsequent p-tau deposition (Cernak et al., 2001, 2011; Lim et al., 2014; Roth et al., 2014; Baalman et al., 2015; Younger et al., 2019). These studies indicate that microglia activation follows a unique temporal pattern based on the type and severity of brain injury sustained by the subject (Jin, 2012; Younger et al., 2019).

In one study by Fan et al. (2018), authors compared 8 individuals with AD to 8 individuals with mild cognitive impairment (MCI) and 14 healthy, aged controls. Using C-PIB PET and magnetic resonance imaging scans, microglial activation was assessed longitudinally. They demonstrated elevated microglia activation at baseline for both MCI and AD patients when compared to healthy controls. After 14 months; however, MCI patients exhibited an 18% reduction in microglial activation while AD patients exhibited a 36% increase. While microglial phenotypes were not measured, the authors hypothesize that the activation presented at baseline may serve a more protective, physiological function, while the persistent activation observed in AD patients is likely a pathologic, chronic inflammation that mediates the transformation of MCI into AD (Fan et al., 2018).

Microglia receptors bind to inflammasome activation receptors

Several studies have shown that microglia activation leads to interactions between the nucleotide-binding oligomerization domain (NOD) found on microglia and the NOD-like receptor protein 3 (NLRP3) inflammasome (Gross et al., 2016; Chang et al., 2020; Chen et al., 2022; Hou et al., 2022; Kattan et al., 2023). This binding initiates the assembly of the inflammasome, resulting in inflammasome activation and cell death by pyroptosis (Chen et al., 2022). Pyroptosis causes leakage of cellular components into the cytosol, which is recognized by vigilant, resting microglia, further activating a pro-inflammatory response (Kattan et al., 2023). This cycle may persist as a positive feedback loop that results in chronic overactivity exacerbating maladaptive neuroinflammation.

Astrocytes

Along with microglia, astrocytes are a type of resident CNS glial cell. They are the most abundant CNS immune cells which exhibit a complex and dynamic morphology that depends on activity level and developmental stage. In general, they have highly detailed processes that provide a large surface area-to-volume ratio. They are found along neuronal synapses, blood vessels, and other glial cells where they have major structural and functional support roles, including the BBB maintenance, neuronal stability and plasticity, and regulation of extracellular fluids, ions, and neurotransmitters (Dong et al., 2021; Singh, 2022; Patani et al., 2023). Astrogliosis is a term used to describe the multifaceted and coordinated astrocytic response to CNS injury, which is characterized by functional, cellular, and molecular changes. Astrogliosis includes a spectrum of heterogenous alterations that occur within and around reactive astrocytes and are injury and environment-dependent (Kumar et al., 2023; Munoz-Ballester and Robel, 2023). These changes include both gain and loss of functions (Sofroniew and Vinters, 2010; Munoz-Ballester and Robel, 2023) and while diverse, are commonly characterized by cellular hypertrophy and proliferation, extension of processes towards the site of injury, tissue reorganization/scar formation, and changes in the expression of chemokines, cytokines, transcription factors, and other signaling molecules (Garcia et al., 2010; Hamby et al., 2012; Zamanian et al., 2012; Patani et al., 2023).

The cytoskeleton glial fibrillar astrocyte protein (GFAP) has historically been used as a marker of astrogliosis, as it is upregulated in astrocytes during cellular hypertrophy. While GFAP levels rise acutely following injury or infection, levels should return to baseline following insult (Sofroniew and Vinters, 2010; Onyango et al., 2021). When maladaptive astrogliosis persists, pathology leading to neuronal dysfunction, disease, and degeneration may result. Newer research indicates that measuring GFAP levels alone does not provide a comprehensive framework for understanding astrogliosis (Arneson et al., 2018, 2022; Witcher et al., 2018; Todd et al., 2021; Chen et al., 2022), as cellular hypertrophy is only one of its many components. While a more complete understanding of astrogliosis is underway, it is generally accepted that reactive astrocytic changes, measured by GFAP, occur in the context of TBI and neurodegenerative pathologies.

Astrocytes and tau

Current research indicates that tau is only subtly endogenous to astrocytes (Kovacs, 2020), yet astrocytes play a key role in tauopathic disease. Several studies demonstrate p-tau internalization by astrocytes under reactive conditions (Pascual et al., 2005; Martini-Stoica et al., 2018; Perez-Nievas and Serrano-Pozo, 2018; Reid et al., 2020) from neighboring neurons and projecting synapses. Homeostatic disturbances due to BBB dysfunction, mechanical impacts such as head trauma, perfusion disturbances, and neurodegenerative events can lead to tau upregulation, hyperphosphorylation, and intracellular accumulation (Reid et al., 2020).

Astroglial internalization of p-tau limits neuronal interaction with the toxic protein and as such, may be neuroprotective. However, at this time, more studies indicate that p-tau internalization by astroglia promotes toxicity (Gaikwad et al., 2021; Kumar et al., 2023) and is even involved in tau seeding and propagation. The data show that that p-tau accumulation in astrocytes disrupts intracellular Ca²⁺ signaling resulting in synaptic dysfunction (Pleiss et al., 2016). Richetin et al. (2020) show that astroglia p-tau accumulation in the dentate gyrus induces memory deficits. Gaikwad et al. (2021) published data indicating that oligomeric tau is associated with a senescence-like phenotype in astrocytes. Importantly, new experimental models show astrocytes that have internalized p-tau have the capability of seeding and spreading tau to neighboring neurons and glial cells (Wharton et al., 2016; Ferrer et al., 2018). Further inquiry into the mechanism by which astrocytic internalization exacerbates the spread of neurotoxic tau represents a new lens through which to understand tauopathies.

Astrogliosis in traumatic brain injury and Alzheimer’s disease

The role of astrocytes in response to TBI is now known to be heterogeneous, with varied responses pending injury type (focal or diffuse), severity (mild, moderate, or severe), and astrocyte distance to the lesion (adjacent or distant). Increased levels of GFAP have been reported in several in vivo models of TBI (Marmarou et al., 1994; Jellinger, 2009; Chandel et al., 2016; Nichols et al., 2016; Shandra et al., 2019; Munoz-Ballester and Robel, 2023) and are recommended for use as a biomarker for TBI (Bogoslovsky and Diaz-Arrastia, 2016; Bogoslovsky et al., 2016, 2017; Papa et al., 2016). Ondruschka et al. (2018) conducted a monocentric, prospective study examining 84 postmortem blood samples to determine the usefulness of liquid biomarkers in differentiating TBI fatality from non-TBI fatality. It was found that while many immune markers decline over time, GFAP levels remained persistently elevated up to 48 hours following death and were reliably able to distinguish TBI fatality from non-TBI fatality. These findings have been further validated by Zwirner et al. (2021) who showed that cerebrospinal fluid GFAP levels, in combination with IL-6, can predict TBI-related death in over 90% of retrospective cases; for a detailed review on astrocytes and TBI please refer to the review by He et al. (2023).

Similarly, studies have demonstrated astrogliosis in the context of AD pathology. Kumar et al. (2023) used in vivo PET and postmortem imaging to show a first and second wave of reactive astrogliosis unique to AD patients. They found the first wave of astrogliosis in prodromal AD stages, before the onset of tau deposition, and a second wave of astrogliosis in advanced AD pathology. Their analysis revealed astrocyte reactivity in close proximity to elevated GFAP levels in the temporal cortex of both patients with sporadic and genetic AD but not in geriatric controls. Their work corroborates previous studies suggesting a link between reactive gliosis and AD pathology (Kadir et al., 2011; Lemoine et al., 2017, 2021; Malarte et al., 2021; Ni and Wu, 2021). Understanding whether astrocyte proliferation in the context of AD is physiologic and responding to disease or pathologic and causing disease remains to be determined.

Inflammasomes activate astrocytes

The first inflammasome to be characterized in astrocytes was the NLRP2 inflammasome which is activated by ATP (Minkiewicz et al., 2013). Since then, other inflammasomes have been described (Freeman et al., 2017; Voet et al., 2018; Kattan et al., 2023). In one in vitro study, Western blot analysis demonstrated the activation of inflammasomes and caspase-1 in cortical astrocytes following toxically elevated levels of extracellular potassium at 30 minutes, 1 hour, and 2 hours post-exposure (Silverman et al., 2009). One study has used immunofluorescent labeling of astrocytes and microglia to show elevated levels of NLRP3, ASC, and caspase-1 expression in the ipsilateral cortex of TBI-induced rats (Liu et al., 2013). Thus, while astrocytes are critical in ameliorating CNS injury, like microglia, they can play a role in neuropathology. One such mechanism for this maladaptive process is mediated via the inflammasome complex (Freeman et al., 2017; Du et al., 2022). While inflammasome activation represents a relatively new frontier for understanding neuropathology, the next section describes our current understanding of its relationship to TBI and the neurodegenerative processes that may follow.

Peripheral immune cell

Despite the highly selective permeability of a healthy BBB, peripheral immune cells have a role to play in neuropathology. Peripheral immune cells, such as neutrophils, monocytes, natural killer (NK) cells, dendritic cells, T cells, and B cells, contribute to neurodegenerative processes by activating microglia and astrocytes, inciting toxicity, and releasing cytokines that can cross the BBB to regulate inflammation. One study by Dong et al. (2018) demonstrates that the percentage of aged neutrophils increases in patients with AD and suggests that neutrophil phenotype may be used as a prognostic biomarker for AD. Pathologic levels of neutrophils are associated with BBB dysfunction (Baik et al., 2014) and are directly toxic to neurons via the release of pro-inflammatory molecules neutrophil extracellular traps and IL-17 (Kolaczkowska and Kubes, 2013), and even temporary depletions (1 month) of neutrophils in early stages of AD have been shown to protect against cognitive decline in transgenic mice (Zenaro et al., 2015).

NK cells have also been suggested as a potential biomarker for AD diagnostics. NK cells are effector lymphocytes of the innate immune system that can cross the BBB. Lu et al. (2021) used bioinformatic analysis of scRNA and bulk sequencing data to show that NK cells infiltrate the CNS and contribute to neurotoxic inflammatory changes. These results have been corroborated by subsequent studies reporting improved cognitive function in AD mouse models following NK cell depletion (Zhang et al., 2020b). The role of the adaptive immune system, specifically the role of T cells and B cells, is currently conflicting. However, given that BBB disruption is a possible sequela of TBI, CTE, and AD, it is highly plausible that these cells play a role in the complex interactions of neurodegeneration.

Inflammasome Activation as a Driver of Traumatic Brain Injury and Neurodegeneration

Inflammasome activation represents one pathway by which neuronal cell death occurs in the context of TBI, particularly in the delayed phase of neuroinflammation (Galgano et al., 2017; Onyango et al., 2021; Chen et al., 2023), as well as in tauopathies such as CTE and AD. Recent research focuses on illuminating the complex activation mechanism for the inflammasome and its downstream causative effects.

Inflammasome structure and signaling pathways

Inflammasomes are intracellular multiprotein complexes that activate a series of pro-inflammatory mediators in response to foreign and pathogenic warning signals in the context of neuroinflammatory processes (Kattan et al., 2023). They are composed of three distinct units: a sensor component, an effector component, and an adaptor component. The sensor component can be one of two classes: a nucleotide-binding domain or a leucine-rich repeat, which contains NOD-like receptors (NLRs) or AIM-like receptors (Simonato et al., 2021; Kattan et al., 2023). When the NLR or AIM-like receptors (Cunha et al., 2017) sense an incoming pathogen via DAMP or PAMP signaling, the adaptor component apoptosis-associated speck-like protein containing a caspase-activation recruitment domain (ASC) oligomerizes and triggers the recruitment of effectors (Simonato et al., 2021). Effectors such as pro-caspases-1, -4, -8, or -11 lead to their maturation as activated caspases (Kigerl et al., 2014; de Rivero Vaccari et al., 2016), which subsequently cleave pro-inflammatory cytokines into their active forms, resulting in cell membrane pore formation and pyroptosis (Kattan et al., 2023; Figure 5). Thus, inflammasome activation serves as a green light for an inflammatory signaling cascade that ultimately results in cell death (Adamczak et al., 2014).

Figure 5.

NLRP3 inflammasome activation and assembly pathway.

Traumatic brain injury-induced brain damage and other neuroinflammatory triggers involve the release of damage-associated molecular patterns (DAMPs) that act as priming signals and trigger transcriptional upregulation of NLRP3 and pro-IL-1β through the TLR/NF-κβ pathway. Following priming, several activating signals induce the oligomerization of the inflammasome sensors (NLRP3), followed by the recruitment of ASC and pro-caspase-1, resulting in a complete NLRP3 inflammasome complex. The second step in deploying the inflammasome complex is activation, which can occur by several biochemical mechanisms including K⁺ efflux, Ca²⁺ influx, mitochondrial injury-induced ROS release, and lysosomal destabilization. Downstream caspase-1 subsequently cleaves pro-GSDMD into active GSDMD, allowing for pore formation and pyroptosis. Cell death results in the leakage of intracellular contents into the extracellular space. This process triggers the release of pro-inflammatory cytokines that further aggravate the extracellular environment and induce kinase-mediated hyperphosphorylation of tau protein. Created with BioRender.com. ASC: Adaptor molecule apoptosis-associated speck-like protein; GSDMD: gasdermin-D; IL-18: interleukin-18; IL-1β: interleukin-1β; NF-κβ: nuclear factor kappa-light-chain-enhancer of activated B cells; NLRP3: nucleotide-binding oligomerization domain, leucine-rich repeat and pyrin-domain 3; ROS: reactive oxygen species; TLR: Toll-like receptor.

Inflammasome activation and traumatic brain injury

In the context of TBI, the two most studied inflammasomes are NLRP1 and NLRP3 (de Rivero Vaccari et al., 2009). Both go on to activate pro-inflammatory IL-1β and IL-18, resulting in membrane pore formation and cell death. NLRP3 must first be primed by stimulation of pattern recognition receptors by DAMPs and PAMPs (Kigerl et al., 2014; Voet et al., 2018; Inokuchi et al., 2022) to initiate activation.

The priming signal upregulates the transcription of NLRP3 and pro-IL-1β via NF-κB, allowing for a robust immune response. NLRP1, on the other hand, does not require priming for activation. Rather, NLRP1 induces inflammasome activation and neuronal cell death in an NF-κB-independent manner (Lamkanfi and Dixit, 2014). While both NLR protein complexes have been shown to increase post-TBI, current data point towards multiple inflammasomes contributing to the secondary injury cascade in different cell types (Liu et al., 2013; de Rivero Vaccari et al., 2014; Liu and Chan, 2014; Russo et al., 2016; Ismael et al., 2018, 2021; Xu et al., 2018a, b; Irrera et al., 2020; Du et al., 2022).

Liu et al. (2013) conducted an in vivo experimental study documenting NLRP3 activation in cortical neurons following TBI. Using a modified Feeney model on Sprague-Dawley rats, Feeney et al. (1981) found that activated NRLP3 up-regulated the transcription of caspase-1 and ASC, with increased downstream processing of IL-1β and IL-18, particularly in the cortical areas adjacent to the initial lesion. They reported a rapid increase of IL-1β at 6 hours post-trauma and a gradual increase of IL-18 at 6 hours to 7 days post-trauma, suggesting differing roles in the primary and secondary post-TBI response. Similarly, the inflammasome proteins caspase-1, ASC, IL-18, and IL-1β have been described to be elevated in the serum and cerebrospinal fluid of patients after TBI (Adamczak et al., 2012; Kerr et al., 2018; Perez-Barcena et al., 2020, 2022; Johnson et al., 2022). In another study, Lin et al. (2017) used human tissue samples to study the neuroinflammatory pathway in post-TBI patients. Immunohistochemistry and western blot analysis showed a significant increase in NLRP3, caspase-1, IL-1β, and IL-18 in the brains of TBI-exposed patients when compared to control post-surgical brains lacking TBI exposure. Chen et al. (2019) also reported elevated levels of NLRP3 mRNA 6 hours following injury and peaking at 24 hours. Several studies have assessed inflammasome activation in the context of TBI utilizing WT and genetic manipulation of inflammasomes or their components as discussed in Additional Tables 3 and 4.

Additional Table 3.

Experimental studies reporting inflammasome activation post-TBI

Species	TBI model	Inflammaso me identified	Components of inflammasomes identified	Method of detection	Time points	Site of detection	Major findings	References
Male C57/BL6 mice	Moderate penetrating CCI At 4.5 m/s velocity with a 200 ms dwell time and 2.0 mm depression using a 3 mm diameter impactor	NLRP3, NLRC4, NLRP1, AIM2	NLRP3,NLRC4, NLRP1, AIM2, Caspase-1	1.WesternBlot 2. Immunofluoresce nce assays	1 d, 2 d, 3 d, 7 d	Microglia in the peri-injury cortex at 3- day	1. Long-term elevation of NLRP3 and caspase-1 from 1 d to 7 d after TBI unlike NLRP1, NLRC4, and AIM2 elevated at 1 day after TBI	Du et al., 2022
Male C57/BL6 mice	Left lateral skull weight drop-200 g steel & a height of 5 cm onto the	NLRP3	NLRP3, ASC, Caspase-1	1. Western Blot 2. Immunohistochem istry 3. qRT-PCR, ATP Assay 4. Fluorescence spectrophotometry 5. ELISA, Mitochondrial membrane potential	6 h, 12 h, 24 h, 2 d, 3 d, 7 d	Cortical neurons	1. Significant ↑mRNA and protein expression of NLRP3 and Caspase-1 at 6 h post-TBI (P < 0.01) 1. Mitochondrial dysfunction, and reduction of mtDNA copy number, ATP concentration, and MMP number, and increased ROS	Chen et al., 2019
Male C57/BL6 mice	1. Moderate penetrating CCI 2. at 4.5 m/s velocity with a 200 ms dwell time and 2.0 mm depression using a 3 mm diameter impactor	NLRP3	NLRP3, ASC, Caspase-1	1. WesternBlot 2. Flow Cytometry 3. Immunofluoresce nce assays	1 d, 3 d, 7d	Microglia in the pericontusiona l cortex a 3- day	2. CCI increased the expression of NLRP3, ASC, and caspase-1 from 1 d to 7 d, peaking at 3 d post-TBI 3.Flow cytometry and IF staining showed that microglia NLRP3	Xu et al., 2018a, b inflammasome-are Microglial Derived
Adult 3- month-old C57BL/6 J male mice	1. Moderate penetrating CCI 2. At 4.5 m/s velocity with a 200 ms dwell time and 2.0 mm depression using a 3mm diameter impactor	NLRP3	NLRP3, ASC, Caspase-1	1. WesternBlot 2. qRT-PCR	1 d, 2 d, 4 d, 7 d, 14 d	Cerebral cortex	1.TBI elevated NLRP3, ASC, and caspase-1 from 1 d to 14 d in a time-dependent manner 2. TBI-related NLRP3 inflammasome activation is associated with increased TXNIP	Ma et al., 2017
Male C57/BL6 mice	1. Moderate penetrating CCI 2.4.5 m/s velocity with a 200 ms dwell time and 2.0 mm depression using a 3mm diameter impactor	NLRP3	NLRP3, ASC, Caspase-1	1. WesternBlot 2. Immunohistochem istry	1 d and 3 d postinjury	Pericontusiona l cortex	1.CCI significantly upregulated the expression of NLRP3, cleaved caspase-1, and IL-1 β in the perilesional area
Adult male Sprague-Dawley rats	1. Closed head weight drop	NLRP3	NLRP3, ASC, Caspase-1	1. Western Blot 2. Quantitative RealTime PCR (qRT-PCR) 3. Immunohistochem istry	6 h, 24 h, 3 d, 7 d	Neurons, astrocytes, and microglia Note: More NLRP3, ASC, and caspase-1-positive cells were detected in the pericontusiona l cortex than in the contralateral at 3-day	1.TBI stimulated NLRP3-inflammasome activation and up-regulated ASC and caspase-1 followed by IL-1β andIL-18 2. NLRP3 expression observed in cortical neurons. 3.IL-1 β rapidly increased at 6 h post-injury and began to decrease at 24 h (ELISA) 2. IL-18 increased from 6 h to 7 days suggesting its involvement in the late phase post-TBI	Liu et al., 2013
Adult male Sprague-Dawley rats	Moderate parasagittal FPI closed	NLRP1	NLRP1, ASC, Caspase-1, Caspase-11	1. WesternBlot 2. Coimmunoprecipi tation 3. Immunohistochem istry	15 min, 30 min, 1 h, 3 h, 6 h, and 24 h	Cortical neurons at 4 h	2. FPI stimulated XIAP cleavage andNLRP1 inflammasome & caspase-1 activation, increased caspase-11, ASC, and maturation of IL-1β 3. Co-IP of NLRP1 inflammasome showed pannexin 1, as a novel upstream target for therapeutic interventions 3.NLRP1 inflammasome expression in cortical neurons suggests that microglia may not be the major source of IL-1 β in the brain	de Rivero Vaccari et al., 2009

Additional Table 4.

Experimental studies assessing the effects of genetically manipulating inflammasome-associated proteins on TBI-induced inflammatory response and brain damage

Species	TBI model	Manipulation	Experimental procedure	Behavioral outcomes	Molecular outcomes	Biomarkers	Comments	References
Male C57BL6/J mice	50 g weight dropped D= 34 cm (right hemispher e)	- NLRP3-/- vs WT mice - Intraperitone al injection with NLRP3 inhibitor MCC950 30 min (pre) trauma or 1 h (post) trauma Naïve (Lesion) vs. no lesion	- Neurological severity score (NSS) 1h and 24h post-TBI - Blood barrier integrity (2% Evans Blue tracer) - qRT-PCR - Western Blot	Significant↑neurological scores over time post-TBI, while NLRP3 showed insignificant improvement	- ↑ mRNA of NLRP3 and proinflammatory cytokines (IL-1 β) post-TBI in WT mice ↑ Inflammatory response in NLRP3-/- mice	↑ NLRP3, AIM2 and pro-inflammatory cytokine (IL-1 β, TNFα, IL6, CCL2) mRNA expression in WT and NLRP3-/-mice post-TBI	- Primed microglial in NLRP3-/- mice lead to ↑ cytokine expression and astrocyte response post-TBI - Time-dependence of NLRP3’s involvement in the inflammatory response and recovery, as well as regulation of NLRP3, shown in NLRP3genetic ablation & its pharmacological inhibition. - MCC950 (a selective inhibitor of NLRP3) may be a promising therapeutic approach for TBI patients within the optimal window of treatment	Lopez- Rodriguez et al., 2022
C57BL/ 6 J mice	Moderate penetrating CCI Mice were impacted at 3.5 m/s velocity with a 2 mm depression using a 3-mm diameter metal tip	GSDMD -/- NLRP3-/-	- NSS, beam walk test, accelerating rotarod test, and open field test a 3 dpost-TBI. - RNAseq	- ↓ NSS (p < 0.01) - ↓ Foot faults in beam walk test. - ↑ Latency to falls in accelerating rotarod test. - ↑ the total distance in open field test - ↓ NSS - ↓ foot faults in beam walk test - ↑ Latency to falls in accelerating rotarod test (p < 0.05) - ↑ The total distance in the open field test		↑ Synaptic proteins PSD95, VAMP and SNAP25 ↓ Pro-inflammatory cytokines Il-1 β and TNF-α ↑ Anti-inflammatory cytokines IL-10 and TGF-βΙ ↑ Synaptic proteins PSD95, VAMP, and SNAP25 ↓GSDMD and N- GSDMD ↑ Synaptic proteins (PSD95, SNAP25, and VAMP)	Outcomes suggest that NLRP3 primarily stimulates GSDMD cleavage after TBI. This study confirmed that the inhibition of GSDMD is a promising therapeutic target mediated by the NLRP3 inflammasome. Further studies for GSDMD inhibition are needed in chronic TBI.	Du et al., 2022
Adult male mice	Closed head weight drop using a 200 g steel weight	Caspase-1-/-	- mNSS, open-field test and Rotarod test at 12, 24, and 48 h post-TBI	- ↓ mNSS compared to WT-TBI (P < 0.01) - ↑ Latency to falls in rotarod test vs. WT-TBI at 12 h, and 24 h) - ↑ Total distance in open field test vs. WT- TBI		- ↑ Anti-inflammatory cytokines IL-10 and TGF-βΙ compared to WT-TBI - ↓ Pro-inflammatory cytokines Il-1β, IL-18 (p 0.001), and INFγ - ↓ Pyroptosis-related proteins caspase-1, p45, caspase-11, and GSDMD compared to WT-TBI - ↓ Serum LDH compared to WT-TBI	-	Liu et al., 2018
Male mice	Impact acceleratio n model of diffuse TBI	NLRP3 -/-	- Novel object recognition test 24 h post-TBI	- ↑ Behavioral test score compared to WT-TBI and sham NLRP3-/-		- Il-1 β and caspase-1 levels showed no increase post-TBI - ↓pro-apoptotic protein BAX activation vs. WT-TBI	- This is the first evaluates the effects of the NLRP3 using a genetic approach (NLRP3 KO mice) - Caspase-1 is a potent therapeutic target as its inhibition may induce TBI- induced neuroinflammation and pyroptosis.	Irrera et al., 2017
Male mice	Moderate penetrating CCI at 3.5 m/s velocity with a 150 ms dwell time and 0.5 mm depression using a 4-mm impactor	NLRP1 -/- - ASC -/-	- Rotarodtest at 3 d, 7 d, and 14 d post- TBI - Evaluation of cell death using TUNEL at 3 d post-TBI	- Genetic deletion of NLRP1 and ASC has not affected motor function post-TBI	Genetic deletion of NLRP1 has not affected contusion volume post-TBI Genetic deletion of NLRP1 or ASC has not produced any difference in the histopathological outcomes post-TBI No significance in TUNEL was seen in the cortex of NLRP1-/-, ASC-/- mice, and wild-type mice. ↓ CD68 + cells and GFAP+ cells - ↑MAP2+ cells and NeuN+	↓ Ipsilateral IL-1 β levels in CCI injured NLRP1 -/- and ASC -/- mice ↓ Ipsilateral IL-6 levels NLRP1-/- but not in ASC -/- Compared to wild type - No difference in IL-18 levels	- NLRP1 inflammasome showed no changes postmoderate CCI model using a genetic approach, highlighting the need to explore the role of NLRP1 in other TBI models to develop an effective therapeutic strategy	Brickler et al., 2016

Irrera et al. (2017) showed that NLRP3 deficiency led to significantly reduced pathology including decreased levels of pro-apoptotic protein BAX in TBI-induced mice when compared to controls. Wild-type mice expressed increased levels of anti-apoptotic proteins BCL-2 and BLC-XL, suggesting that TBI activates the inflammasome by way of ASC and caspase-8 recruitment. The authors reported overall preserved brain parenchyma in knockout mice up to 7 days following injury. Given the abundance of new data, Kerr et al. (2018) sought to investigate if NLRP3 could be used as a biomarker for TBI. In 2018, the authors published results where they used serum and cerebrospinal fluid samples from severe TBI patients at 15 separate time points (3 times daily for 5 days) to calculate the sensitivity and specificity of inflammasome activation as a TBI marker (Kerr et al., 2018). The sensitivity of the inflammasome proteins caspase-1 and ASC was above 80% and presented area under the curve values above 0.9 within the first 24 to 48 hours after TBI, indicating that inflammasome signaling proteins are promising biomarkers of the inflammatory response associated with TBI. These findings further highlight the potential of inflammasome proteins as therapeutic targets in TBI patients. ASC presents similar area under the curve values in the blood patients in the early stages of cognitive decline (mild cognitive impairment), further highlighting the role of the inflammasome in the inflammatory response in cognitive decline and AD (Scott et al., 2020).

Du et al. (2022) published the results of their moderate penetrating controlled cortical impact on male C57/BL6 mice. The authors found immediate increases in NLRP1, NLRP3, and AIM-2 receptors 1-day post-injury (Du et al., 2022). However, only NLRP3 and caspase-1 remained elevated through day 7, indicating a key role of NLRP3 in secondary and sustained post-TBI neuroinflammation. Their data further supports the temporal expression pattern of pro-inflammatory mediators, concluding a regulatory role for NLRP3 in long-term sequelae. Another study conducted by Lopez-Rodriguez et al. (2022) documented increased mRNA expressions of NLRP3 and IL-1β 24 hours following cortical impact on TBI-induced mice. No inflammatory markers were reported in sham mice.

The team then used a pharmacological approach, inhibiting NLRP3 inflammasome by injecting 3 mg/kg of MCC950 30 minutes prior to TBI. The data show that administration of MCC950 inhibitor given before head injury preserves BBB integrity and produces less neurological damage according to histological and behavioral analysis. The authors validated their pharmacological findings using a genetic model with NLRP3 knockout mice. Interestingly, the research team found no difference in IL-1β production between wild-type and genetically altered mice; however, they did observe significant increases in pro-inflammatory microglia and astrocyte markers. The authors conclude their discussion by stating that NLRP3 is the canonical sensor of sterile injury, where microglia and astrocytic interactions mediate the post-TBI inflammatory response. Most recently, the pyrin inflammasome has also been shown to be elevated in the cortex acutely after TBI. However, how the pyrin inflammasome becomes activated in the brain after injury is yet to be determined (Keane et al., 2023). Several studies have evaluated the utility of inflammasomes as biomarkers (Additional Table 5).

Additional Table 5.

Human studies highlighting the use of inflammasome-related proteins as promising biomarkers of TBI

Biomarker	Study design	Sample size	Control	Measured outcomes	Clinical significance	Limitations	References
ASC, Caspase- 1, IL-18	- Moderate and severe TBI (GCS score of ≤12) - TBI samples were collected 3 times a day for the first 5 days - Samples were analyzed for the 1st, and 2nd collection (Day 1) as well as the 4th and 6th collections (Day 2). - This study used the Ella Simple Plex system (Protein Simple) to assess the sensitivity and specificity of inflammasome proteins as TBI biomarkers	Control CSF samples: n =120 TBI CSF samples: n = 21 Control serum samples: n = 30 TBI serum samples: n =18	Mean ACS serum level: 236.6 pg/mL Mean Caspase-1 serum level: 1,436 pg/mL Mean IL- 18 serum level: 213.5 pg/mL	- ROC curves and confidence intervals for Serum inflammasome signaling proteins showed that caspase-1 and ASC had the highest AUC values of 0.93 and 0.9 respectively. - ROC curves and confidence intervals for CSF inflammasome signaling proteins in CSF showed that ASC and Il- 18 had the highest AUC values of 1 and 0.84 respectively. - ASC protein level is higher in the serum of TBI patients with unfavorable outcomes compared to favorable outcomes patients. No statistical difference was obtained when comparing caspase-1 and IL-18 between the 2 groups.	- At a cut-off point of 1.943 pg/mL, Casepase- 1 yielded 94% sensitivity and 89% specificity. At a cut-offpoint of 451.3 pg/mL, ASC yielded 85% specificity. - At a cut-offpoint of 74.33 pg/mL, ASC yielded 100% sensitivity and 100% specificity. For IL-18, acut-off points of2.722 pg/mL gave 80% sensitivity. - ROC curves and confidence intervals for ASC in serum of favorable vs. unfavorable outcomes showed that AUC of ASC is 0.9167 (95% CI 0.7194-1.042, P = 0.0039) making ASC levels a promising predictive serum biomarker of TBI	- Serum samples are higher than CSF samples. - Most samples were collected from patients with severe TBI. - Some patients had polytrauma, - Apart from whites & blacks. No Hispanics or other races were included.	Kerr et al., 2018
NLRP3, Caspase- 1, IL-Ιβ, IL-18	- Severe TBI (GCS ≤ 8) - Contusional brain tissues resected from TBI and control surgical patients without TBI - 5 TBI patients with severe TBI and 3 other control surgical patients without TBI - This study used immunohistochemistry and western blot to assess inflammasome activation in human brains following TBI	Control brain tissues: n = 3 TBI brain tissues: n=5		- NLRP3, Caspase-1, IL- 1β, and IL-18 are significantly higher in the brain of TBI patients vs. controls. - TBI stimulates NLRP3 inflammasome activation and pro-inflammatory cytokines release in human brains			Lin et al., 2017
ASC, Caspase- 1,NLRP1	- Moderate and severe TBI (CGS score of ≤12) - 45 CSF samples were collected from all TBI patients within 12 h of injury and up to three days post-injury. - This study carried out immunoblot analysis of the CSF samples and used the Glasgow Outcome scale (GOS) to correlate the levels of the biomarkers with the outcome five months post-injury	Control CSF samples: n = 9 TBI CSF samples: n = 23		- ASC, Caspase-1, and NLRP1 are expressed in CSF of TBI patients. - ASC, caspase 1, and NLRP1 were significantly higher in the CSF of patients with unfavorable outcomes, including death and severe disability, vs. moderate to no disability patients. - Linear regression analysis showed ASC, caspase-1 (p20), and NLRP1 expression significantly correlate with the longterm functional outcome of TBI patients at 5 months	- ASC, caspase-1 and NLRP-1 are acutely elevated in the CSF of TBI patients compared to controls, which makes them promising CSF biomarkers for TBI - ASC, caspase-1, and NLRP1 were significantly higher in the CSF of patients with unfavorable outcomes, including death and severe disability, compared to patients with moderate to no disability.	- The inflammasomes might be high in the CSF due to brain parenchyma disruption (i.e., acute cerebrovascular accident, tumor resection, abscess). - The post-injury time points should be extended.	Adamczak et al., 2012

Inflammasome activation and Alzheimer’s disease

AD pathology in microglia seems to be modulated by the NLRP3 inflammasome, whereas neuronal pathology seems to be contributed to by the NLRP1 inflammasome (Vontell et al., 2023). Several studies using human samples have found increased levels of NLRP3, IL-1β, and caspase-1 in patients diagnosed with AD (Heneka et al., 2015; Piancone et al., 2021; Ravichandran and Heneka, 2021). Similarly, ASC has been shown to be elevated in the blood of patients with mild cognitive impairment, suggesting a role for the inflammasome in the early stages of the disease (Scott et al., 2020). Clinical assessments using validated cognitive and behavioral scales and postmortem analyses further corroborate these findings, as do murine models and in vitro culture studies (Pontillo et al., 2012; Saresella et al., 2014; Johann et al., 2015; Peelen et al., 2015; Kadhim et al., 2016; Sarkar et al., 2017; Keane et al., 2018; Piancone et al., 2018; Moreno-Garcia et al., 2021; Trudler et al., 2021; Jose et al., 2022; Ising and Heneka, 2023). Aβ, the first hallmark misconfigured protein of AD, is thought to be one of the primers for NLRP3 assembly (Nakanishi et al., 2018; Zheng et al., 2020). Detection of Aβ leads to binding of NLRP3 with ASC, initiating the first step in this process (Nakanishi et al., 2018).

Aggregated Aβ, particularly in the oligomeric and fibrillar forms, complexes with TLRs; hence, promoting the nuclear translocation of NF-κB, which can upregulate NRLP3 and IL-1β transcription (Venegas et al., 2017; Eren and Ozoren, 2019; Venegas and Heneka, 2019; Zhang et al., 2020a). Following priming, the secondary activation signal occurs when microglia phagocytose Aβ, consuming the toxic protein and engulfing it within its lysosomal structures. These lysosomes may then rupture via the reactive oxygen species generation pathway (Halle et al., 2008; Zhang et al., 2020a), releasing proteinase cathepsin-B into the cytoplasm, inducing NLRP3 and IL-1β activation (Halle et al., 2008; Rubartelli, 2012; Parajuli et al., 2013; Eren and Ozoren, 2019; Zhang et al., 2020a). Thus, while NLRP3 can be induced by the onset of AD by way of protein aggregates, it also perpetuates AD pathology (Shippy et al., 2020).

Evidence of elevated NLRP3 and NLRP1 levels in the temporal cortex was reported upon immunohistochemical evaluation of postmortem AD brains when compared to age-matched geriatric controls (Ahmed et al., 2017). Saresella et al. (2016) reported similar results. The authors used qPCR to measure the mRNA expression in patients diagnosed with mild (20) and moderate (21) AD compared to healthy, aged controls. They documented significant upregulation of NLRP1, NLRP3, caspase-1, IL-1β, and IL-18 in the peripheral monocytes of mild AD patients and significant increases in NLRP3 and caspase-1 in severe AD patients compared to controls. The study team verified their findings by performing RT-PCR on each specimen and concluded that the activation of at least two separate inflammasome complexes likely mediates neuroinflammation within an AD context (Saresella et al., 2016).

Identifying NLRP3 activation as a causal mechanism in AD has begun to show promise for therapeutic targets. Kim (2021) investigated the mechanism by which acetylcholinesterase inhibitor donepezil, a commonly prescribed anti-AD agent, interrupts AD pathology. The authors gave LPS-treated wild-type mice along with donepezil and using qPCR analysis, reported significantly reduced NLRP3 signaling, decreased NF-κB and STAT3 phosphorylation, decreased microglia activation, and decreased pro-inflammatory cytokine signaling. In murine AD models, donepezil showed reduced Aβ microglial and astrocytic activation, density, and morphology compared to vehicle-treated subjects. The authors conclude that donepezil downregulates LPS and Aβ-induced NLRP3 activation (Kim, 2021).

Other studies have validated these findings, linking NLRP3 activation to AD. Lonnemann et al. (2020) used a pharmacological NLRP3 inhibitor, OLT1177, in wild type and APP/PS1 AD mice. After 3 months of exposure, the 9-month-old mice exhibited improved memory via Morris water maze testing compared to baseline and non-treated controls. The authors also reported a dose-dependent reduction in microglia activation and reduced cortical Aβ plaques, corroborating the NLRP3 pathway in AD symptomatology (Lonnemann et al., 2020).

Inflammasome activation and chronic traumatic encephalopathy

A recent study has shown that following TBI in 3×Tg AD mice, inflammasome activation is exacerbated in the brain of the mice with a genetic predisposition towards AD when compared to the WT injured group. Thus, indicating that the inflammasome contributes significantly to the pathology present after TBI in AD (Johnson et al., 2023a, b). Moreover, treatment of mice with a genetic predisposition towards AD with IC100, a humanized monoclonal antibody against ASC, (de Rivero Vaccari et al., 2023), resulted in decreased IL-1β in the brain of mice after TBI when compared to IgG-treated controls (Johnson et al., 2023b). This is consistent with a previous study showing decreased inflammasome activation with IC100 in the cortex of aged mice (Cyr et al., 2022). CTE is a chronic, neurodegenerative disease marked by perivascular hyperphosphorylated tau in the deep cortical sulci. Nearly all CTE cases occur in a post-head trauma context (McKee et al., 2023a) Given the similar etiology and pathology for TBI, CTE, and AD, it is reasonable to hypothesize that CTE may also develop in an inflammasome-dependent manner. Several studies have shown that tauopathies in general are mediated by the inflammasome.

Inflammasome activation and tauopathy

Understanding the activation of inflammasomes has proved useful in better understanding the pathophysiology of neurodegeneration following head trauma. While the role of p-tau as pathogenic has been established, its connection to inflammasome complexes is only beginning to come to light. A 2019 Nature paper studied the role of inflammasomes in tauopathy and found that NLRP3 activation in mice was overexpressed in tauopathic mice (Ising et al., 2019). Immunoblot was used to detect caspase-1 along with immunohistochemical staining of microglia, ASC, and p-tau in WT mice compared to mice modeled with tauopathy. Using self-organizing map clustering, the authors identified genes upregulated in tauopathy. Their findings indicated areas of hyperphosphorylation linked to pathogenicity by over-expression of inflammatory signaling processes, including genes associated with inflammation such as Irak1, MyD88, and Il1rap. The authors were able to visualize neurodegenerative changes in microglia at 8 and 11 months old in areas of dense tauopathy. Importantly, NLRP3 knockout mice showed decreased levels of caspase-1, IL-1β, and ASC speck formation, as well as reduced levels of total tau and p-tau, suggesting a causal pathway. Interestingly, levels of misfolded tau protein were reduced in mice at 8 months but not at 11 months. The authors suspect that this may indicate a normal physiological plateau in tau expression during this time. In addition to NLRP3 activation producing p-tau, Heneka et al. (2015) demonstrated that oligomeric tau aggregates also activate NLRP3, suggesting a positive feedback loop that has been proposed before (Heneka et al., 2015; Kim, 2021). In addition, in a post-mortem study, caspase-1 is associated with tau aggregates in the brain of AD descendants (Vontell et al., 2023). However, the mechanism by which tau directly or indirectly activates the NLRP3 inflammasome remains to be determined. Mechanisms involving NEK7 or post-translational modification following alterations in homeostatic mechanisms within the cell are known to promote NLRP3 inflammasome activation, but how tau aggregates alter the NLRP3 cage to allow for binding to ASC and caspase-1 remains to be elucidated.

Stancu et al. (2019a) reported similar results. Accordingly, the authors injected transgenic mice with pre-aggregated tau seeds and LPS to initiate NLRP3 priming and measured the levels of pro-caspase-1 and ASC. As expected, tau seeding induced Il-1β expression by way of ASC oligomerization and pro-caspase-1 activation (Stancu et al., 2019a). Upon inhibition of NLRP3 with MCC950, levels of IL-1β and pro-caspase 1 were blunted. Tau transgenic mice deficient in ASC showed that, upon injection with exogenous tau seeds in the frontal cortex, tau pathology, measured by both AT88 and Thiol, failed to propagate. Stancu et al. (2019b) repeated this experiment using non-exogenously seeded tau and reported similar findings, indicating that tau propagation requires the assembly of NLRP3 via ASC activation. Furthermore, the authors used spatial-temporal progression analysis of microglial cells to show that microglial activation precedes the development of mature neurofibrillary tangles, corroborating previous findings that indicate tau oligomers as activators of tau spreading and toxicity (Clavaguera et al., 2009; Hawkins et al., 2013; Iba et al., 2013; Holmes and Diamond, 2014; Holmes et al., 2014; Furman et al., 2017; DeVos et al., 2018; Stancu et al., 2019a). By comparing tau pathology in knockout compared to wild-type murine, the authors documented significantly decreased tau pathology in the absence of NLRP3.

Pampuscenko et al. (2023) corroborated previous findings implicating NLRP3 activation in a post-TBI context using neuronal and microglial co-culture fluorescence studies. They report increased activation of labeled caspase-1 in microglia by 44% when treated with 3 µM of tau protein for 24 hours. Treatment with tau resulted in a 50% loss of the total neuronal population and a doubling of the number of microglial cells, indicating pyroptosis by way of astrogliosis (Pampuscenko et al., 2021). When two respective selective caspase-1 inhibitors were added (YVAD-CHO and VX-765), both culture samples showed complete inhibition of tau-induced neuronal loss and microglial proliferation. To better understand the effect of caspase-1 on microglial phagocytosis and neuronal cell death, the researchers used carbonylated beads to mimic the phosphatidylserine (PS)-exposing region of cells. When PS is exposed on the outer cell membrane, it allows for recognition by PS receptors on microglia, which subsequently engulf the apoptotic cell (Lemke et al., 2023). Thus, increased uptake of PS serves as a proxy for programmed cell death. Pampuscenko et al. (2021) found that microglia cultures increased their uptake of the PS-mimicking beads from 100% in the untreated arm to 214% in the tau-treated arm, showing that tau increases phosphatidyl serine exposure and risk for cell death. Treatment with caspase-1 inhibitor YVAD-CHO completely prevented neuronal cell death, indicating that the microglial phagocytosis of neurons is mediated via caspase-1 activity. These data strongly suggest that pyroptosis in tauopathy is mediated at least in part by the NLRP3 inflammasome and downstream caspase-1 activation (Pampuscenko et al., 2021). Given the growing evidence linking NLRP3 activation to tauopathies, it is reasonable to hypothesize that CTE, a similar, yet distinct tauopathy compared to AD, occurs along a similar mechanistic pathway.

Discussion

This review surveyed recent literature on TBI, CTE, and AD in the context of maladaptive neuroinflammation and tau protein changes. While TBI, CTE, and AD represent distinct diagnoses, these three neuropathologies share the histological and clinical features of tauopathy. As such, they are characterized by the accumulation of neurotoxic tau protein accumulations in brain parenchyma that over time, leads to neurodegeneration and brain atrophy with dose-dependent behavioral, motor, and cognitive dysfunctions. While the clinical presentations overlap, these diseases are distinguished primarily by the distribution of hyperphosphorylated tau, a once physiological protein that becomes pathologic upon inflammation-induced phosphorylation, misfolding, and aggregation.

TBI response occurs in a biphasic manner, with the initial response occurring at the onset of injury followed by a prolonged inflammatory response that leads to chronic sequelae. Pathological tau may be generated during the prolonged response and misfolded protein aggregates are generally confined to the focal area surrounding injury. Repeated TBI, however, may lead to the spreading of p-tau to adjacent and even contralateral brain regions. Over time, tau depositions increase in density and may develop into CTE. CTE is histologically defined by the presence of p-tau depositions along the periphery of the brain vasculature. These oligomeric depositions have been shown to spread from the initiating point of impact by way of microglial spreading, tau seeding, and prion-like spreading. Both TBI and CTE are documented risk factors that predispose individuals to develop AD, which is characterized by p-tau deposits that aggregate into neurofibrillary tangles, distinguished primarily by their accumulation in the entorhinal cortex and hippocampus. As is seen in both TBI and CTE, these tau depositions increase in density and spread over time, with correlating clinical symptomatology.

In vivo and in vitro studies have demonstrated that TBI can serve as a trigger for inflammasome activation and subsequent IL-1β and pyroptosis. Cell death leads to the leakage of cell contents, which activates pro-inflammatory signals that can further trigger the inflammasome pathway. When TBI occurs repeatedly, as is seen in CTE pathogenesis, NLRP3 hyperactivates and neuroinflammatory processes are prolonged. Similarly, experimental studies and postmortem human tissue analysis demonstrate NLRP3 activation and resultant cell death in AD pathology. P-tau depositions increase and spread over time, worsening the disease.

To date, there is no direct evidence linking inflammasome activation and CTE in animal models or post-mortem brain samples from humans. In part, this is due to the lack of animal models of CTE. However, findings from animal studies highlight that TBI increases the inflammasome response in mice with a genetic predisposition towards AD when compared to mice without such genetic predisposition and that the pathology after TBI in these mice is similar to what patients with CTE present. Current studies are under way to better characterize the role of the inflammasome in CTE pathogenesis in humans.

Considering the relevance of the inflammasome to a variety of diseases in the CNS and the periphery (de Rivero Vaccari et al., 2016), the therapeutic potential of targeting the inflammasome to improve outcomes in patient populations affected by TBI, AD, CTE, and other neurodegenerative diseases such as Parkinson’s disease (Cabrera Ranaldi et al., 2023) and multiple sclerosis (Desu et al., 2020) is of great interest. This interest has resulted in several compounds being in the late pre-clinical to early clinical stages of testing. These potential therapeutics include DFV890 (IFM-2427, Novartis), Inzomelid (IZD174, Roche), NT-0796 (Nodthera Limited), OLT1177 (Dapansutrile, Olatec Therapeutics), SELNOFLAST (RO-7486967, Roche), Rizaben (Tranilast, Nuon Therapeutics) and IC100 (ZyVersa Therapeutics), among others (McManus and Latz, 2024). Of these inflammasome inhibitors, many of them have been shown to penetrate the CNS such as INZOMELID, NT-0796, OLT1177, Rizaben, and IC100 (Desu et al., 2020; de Rivero Vaccari et al., 2023; McManus and Latz, 2024). These BBB-penetrant properties make these drugs suitable therapeutics for the treatment of the inflammatory response mediated by the inflammasome in indications such as TBI, AD, and CTE. In fact, drugs that currently have active clinical trials in the CNS inflammasome field include ZYIL1 (Zydus Lifesciences Limited) for the treatment of amyotrophic lateral sclerosis (ALS) (NCT05981040, phase 2) and SELNOFLAST (Roche) for the treatment of Parkinson’s disease (NCT05924243, phase 1) (McManus and Latz, 2024). Thus, in the recent future, it is anticipated that several compounds will be tested in clinical trials to inhibit the inflammasome either by interfering with the function of individual sensors such as NLRP1 or NLRP3 or by interfering with other inflammasome components that are present in several inflammasomes such as ASC. Based on the role of the inflammasome in AD, TBI, and CTE, and the role that the inflammasome plays in the early stages of cognitive decline (Scott et al., 2020), it is promising that inhibition of the innate immune response at the level of the inflammasome will provide therapeutic benefits in these CNS conditions and the periphery.

Given the dynamic processes involved in neuroinflammation and resulting p-tau deposition, it is reasonable to theorize that TBI, CTE, and AD occur along the same NLRP3-associated pathway, where TBI can develop into CTE and later, AD, given the right time and biochemical conditions. Further experimental studies are needed to confirm NLRP3 activation in CTE pathogenesis and explore inflammasome inhibitors as potential pharmacological targets for CTE mitigation. Given the temporal sequala by which TBI may lead to CTE and CTE may lead to AD, timely therapeutic intervention following head injury is critical. This updated review highlights the importance of further biochemical studies on NLRP3 inflammasome assembly and activation in neurodegenerative pathologies and calls for the prompt identification and treatment of TBI-induced injuries.

Conclusion and Proposed Framework

We propose a novel framework that identifies inflammasome activation and tau protein disbalance as key biochemical pathways that allow for the transformation of repeated TBI into CTE and AD, given the right environmental and genetic factors. While it is known that repeated TBI leads to CTE, this review proposes an inflammasome-dependent process that, when prolonged, has the potential to develop into AD over time. Future studies must include genetic and pharmacological in vitro and in vivo experimental studies to further establish this theory.

Additional files:

Additional Table 1: Comparison of pathological features and presentations of Alzheimer’s disease, traumatic brain injury, and chronic traumatic encephalopathy.

Additional Table 2: List of anti-tau specific treatments, indication, clinical trials, and their corresponding epitopes.

Additional Table 3: Experimental studies reporting inflammasome activation post-TBI.

Additional Table 4: Experimental studies assess the effects of genetically manipulating inflammasome-associated proteins on TBI-induced inflammatory response and brain damage.

Additional Table 5: Human studies highlight the use of inflammasome-related proteins as promising biomarkers of TBI.

Data availability statement:

All relevant data are within the manuscript and its Additional files.