Inflammaging, cellular senescence, and cognitive aging after traumatic brain injury

Abstract

Traumatic brain injury (TBI) is associated with mortality and morbidity worldwide. Accumulating pre-clinical and clinical data suggests TBI is the leading extrinsic cause of progressive neurodegeneration. Neurological deterioration after either a single moderate-severe TBI or repetitive mild TBI often resembles dementia in aged populations; however, no currently approved therapies adequately mitigate neurodegeneration. Inflammation correlates with neurodegenerative changes and cognitive dysfunction for years post-TBI, suggesting a potential association between immune activation and both age- and TBI-induced cognitive decline. Inflammaging, a chronic, low-grade sterile inflammation associated with natural aging, promotes cognitive decline. Cellular senescence and the subsequent development of a senescence associated secretory phenotype (SASP) promotes inflammaging and cognitive aging, although the functional association between senescent cells and neurodegeneration is poorly defined after TBI. In this mini-review, we provide an overview of the pre-clinical and clinical evidence linking cellular senescence with poor TBI outcomes. We also discuss the current knowledge and future potential for senotherapeutics, including senolytics and senomorphics, which kill and/or modulate senescent cells, as potential therapeutics after TBI.

Traumatic brain injury (TBI) is a significant worldwide public health issue, killing or debilitating individuals regardless of age, gender, race, or socioeconomic status (Centers for Disease C, Prevention, 2009). TBI, which is broadly defined as a blow or jolt to the head that disrupts neurological function, has a heterogenous presentation that complicates the development of efficacious therapies to improve short and long-term quality of life. As a result, up to 2% of the population currently lives with the neurological consequences of a prior TBI. Research spanning several decades has focused on defining and treating the acute pathophysiology after a moderate-severe TBI, with significant effort directed toward the identification of acute cerebroprotectants that limit neuronal loss, including blockade of excitatory pathways and quenching of free radicals; however, despite widespread pre-clinical success, these approaches failed to translate into better long-term clinical outcomes. Whether due to poor target selection, incorrect drug choices, limited therapeutic windows, and/or inadequate pre-clinical modeling, these clinical failures illustrate the daunting challenge that remains to develop neuroprotective therapies after TBI. As such, pre-clinical and translational research must continue to mechanistically define TBI pathophysiology with the hope that this knowledge will advance the development of innovative therapeutic approaches.

The preponderance of research to date has targeted axonal pathology, neuronal loss, cerebral edema, and elevated intracranial pressure within the acute and sub-acute periods after TBI. Conversely, compelling epidemiological data indicate TBI increases the lifelong dementia risk by up to 50%, as compared to non-traumatic, age-matched control cohorts, making neurotrauma the leading external risk factor for dementia (Fleminger et al., 2003; Mortimer et al., 1991; Molgaard et al., 1990; Sullivan et al., 1987; Gedye et al., 1989; Nemetz et al., 1999; Ross, 2011; Tomaiuolo et al., 2004; Warner et al., 2010; Lye and Shores, 2000; Plassman et al., 2000; Schofield et al., 1997; Guo et al., 2000; Graves et al., 1990; Mortimer et al., 1985). Thus, elucidating how acute brain injuries drive subsequent progressive neurodegeneration, which often manifests decades after the initial TBI, is imperative for improving quality of life in neurotrauma patients. The goal of this mini-review is to provide an overview of the emerging evidence suggesting TBI accelerates the aging progress and promotes neurodegeneration. We also speculate on potential new therapeutic targets/approaches, including the unexplored possibility that inflammaging is a primary driver of premature cognitive aging after TBI. This knowledge will provide a foundation for future experimental and clinical research to improve chronic outcomes after TBI.

1. TBI accelerates cognitive aging and elevates the risk of dementia

The term ‘aging’ encompasses both chronological and biological aging. Chronological aging describes the linear changes over the lifespan while biological aging may produce nonlinear changes due to the influence of genetics and/or the environment. Ideally, chronological and biological age match, whereas deviations between chronological and biological age reflect accelerated aging, as compared to a “typical” individual. As aging is associated with an increased risk for developing chronic diseases and incident multi-morbidity, particularly during the final 25% of the lifespan, understanding the factors that influence accelerated aging may reduce societal burden of age-associated chronic illnesses (St Sauver et al., 2015; Kennedy et al., 2014).

Dementia encompasses pathological impairments in memory, language, cognition, and decision making that interferes with quality of life. Compelling epidemiological data, including prospective clinical studies, indicate TBI accelerates brain aging and is the leading extrinsic risk factor for later onset dementia (Fleminger et al., 2003; Mortimer et al., 1991; Molgaard et al., 1990; Sullivan et al., 1987; Gedye et al., 1989; Nemetz et al., 1999; Ross, 2011; Tomaiuolo et al., 2004; Warner et al., 2010; Lye and Shores, 2000; Plassman et al., 2000; Schofield et al., 1997; Guo et al., 2000; Graves et al., 1990; Mortimer et al., 1985; Faden and Loane, 2015; de Freitas Cardoso et al., 2019; Irimia et al., 2014). Toward this end, post-traumatic neurological deficits in young adults often resemble cognitive decline in elderly populations with no neurological history while a single moderate-severe TBI or repetitive mild TBI is a risk factor for the development of neurodegenerative diseases, including Alzheimer’s disease (AD) (Fann et al., 2018; Tolppanen et al., 2017; Mackay et al., 2019; Lehman et al., 2012), frontotemporal dementia (Rosso et al., 2003; Kalkonde et al., 2012; Deutsch et al., 2015), Parkinson’s disease (Mackay et al., 2019; Lehman et al., 2012; Gardner et al., 2015), and amyotrophic lateral sclerosis (Mackay et al., 2019; Lehman et al., 2012; Pupillo et al., 2020; Chio et al., 2005). Both observational and case-control studies found a “dose-dependent” effect of TBI, with more severe injury associated with higher risks of dementia (Tolppanen et al., 2017; Barnes et al., 2018; Nordstrom and Nordstrom, 2018), although a dissenting study found no clear association between TBI and AD risk (Sugarman et al., 2019). In addition, a history of TBI may be a risk factor for earlier onset AD, although this remains a topic of debate (Mortimer et al., 1991; Nemetz et al., 1999; Plassman et al., 2000; Guo et al., 2000; LoBue et al., 2017; Schaffert et al., 2018). Data primarily collected in military personnel and contact sport athletes further suggest repetitive (sub)concussive hits may lead to chronic traumatic encephalopathy (CTE), a progressive neurodegenerative disease associated with a unique pathology and both cognitive and psychiatric deficits (Katz et al., 2021; Bieniek et al., 2021). Together, these findings suggest a poorly understood relationship between neurotrauma and premature brain aging (Wood, 2017).

In line with the clinical association between TBI and subsequent neurodegeneration (Crane et al., 2016), histopathological features of TBI frequently mirror the classical features of AD. Using experimental models of TBI, we and others reported β-amyloid accumulation in the thalamus one year after moderate/severe TBI (Iwata et al., 2002; Chen et al., 2004; Zhang et al., 2012). Moderate/severe TBI also induces chronic widespread neurofibrillary tangles (NFT), temporally paralleling dementia onset (Corsellis and Brierley, 1959; Rudelli et al., 1982; Smith et al., 2003a), while amyloid precursor protein (APP) and APP processing enzymes, presenilin-1 and BACE1, accumulate at points of disrupted axonal transport, persisting for over six months post-TBI in large animals and humans (Iwata et al., 2002; Gentleman et al., 1993; Sherriff et al., 1994; Smith et al., 1999; Smith et al., 2003b; Chen et al., 2009). Diffuse amyloid pathology is acutely observed in ~30% of moderate/severe TBI patients, resembling the pathology in early AD, whereas an elevated fibrillar plaque burden presents at >1 year post-TBI, resembling established AD (Johnson et al., 2010; Roberts et al., 1991; Roberts et al., 1994; Ikonomovic et al., 2004; Johnson et al., 2012); however, widespread accumulation of extracellular amyloid-beta (Aβ) plaques and intracellular tau tangles often appear in postmortem brain tissue from patients with a history of TBI, but without a clinical diagnosis of AD (Johnson et al., 2012; Hyman et al., 2012; Agrawal et al., 2022). Similarly, cognitive and psychiatric symptoms may partially, but not entirely, overlap between TBI and dementia patients, possibility attributed to divergent etiologies (McDonald et al., 2002; Bray et al., 2021; Rao et al., 2010). Thus, despite overlapping neuropathology between TBI and AD, mechanistic and clinical differences may exist between the conditions.

In addition to neuropathological changes, white matter (WM) pathology, including extensive demyelination, contributes toward the clinical presentation of dementia (Sullivan et al., 1987; Gedye et al., 1989; Nemetz et al., 1999; Ross, 2011). Impact and/or coup-contrecoup injuries produce immediate diffuse axonal shearing and myelin degradation, yet most demyelination and WM loss develops for years after severe TBI in both rodents and humans (Bramlett and Dietrich, 2002; Pierce et al., 1998; Johnson et al., 2013a; Ramlackhansingh et al., 2011; Buki and Povlishock, 2006; Green et al., 2014; Gale et al., 1995; Ng et al., 1994; Berger et al., 2005; Wakade et al., 2010; Farook et al., 2013; Liu et al., 2006; Ottens et al., 2008; Ross et al., 1993; Su et al., 2012; Johnson et al., 2013b; Smith et al., 1997). WM lesions present within the corpus callosum of up to 76% of severe TBI patients while 96% of moderate/severe TBI patients exhibit cerebral atrophy, including reduced callosal volume (Green et al., 2014; Gentry et al., 1988). These structural changes manifest as memory loss, mood and concentration impairments, and psychiatric deficits (Kraus et al., 2007; Kumar et al., 2009; Pischiutta et al., 2018; Yoshita et al., 2006; Kovari et al., 2007; Bagi et al., 2022; Hase et al., 2018). While no FDA-approved therapies effectively prevent, delay, or reverse post-traumatic neurodegeneration, the degree of chronic inflammation correlates with progressive, bilateral thalamo-cortical WM tract damage, including reduced callosal thickness, for years after a moderate/severe TBI in rodents and humans (Johnson et al., 2013a; Pischiutta et al., 2018; Scott et al., 2015), supporting a poorly defined relationship between axonal injury, chronic inflammation, and progressive neurodegeneration after TBI.

2. Inflammaging as a contributing factor in delayed neurodegeneration

Inflammation is critical for host defense against infection and injury. Conversely, prolonged immune activation contributes toward a multitude of chronic, age-related ailments while chronological aging is associated with immune dysfunction, including elevated levels of inflammatory modulators that may be adaptive and/or detrimental for longevity (Fulop et al., 2017). While the CNS is traditionally regarded as immune privileged, meningeal lymphatic vessels that drain cerebrospinal fluid into deep cervical lymph nodes provide a conduit for the bidirectional movement of immune cells into and out of the brain (Louveau et al., 2015; Aspelund et al., 2015). A temporally and spatially coordinated series of immune responses develop within the CNS after experimental TBI while chronic neuroinflammation, the complex integration of CNS resident cells and infiltrating immune cells, correlates with WM loss, dementia, and premature cognitive aging for decades after moderate/severe TBI in humans (Wood, 2017; Johnson et al., 2013a; Scott et al., 2015). Despite the long-held assumption that immunosuppression may be beneficial after CNS injury (Braughler and Hall, 1985), the 10,008 patient CRASH trial was prematurely halted due to worse outcomes in the corticosteroid arm, as compared to the placebo arm (Roberts et al., 2004), supporting the notion that inflammation is a double-edged sword after injury (Bramlett and Dietrich, 2002; Pierce et al., 1998; Johnson et al., 2013a; Ramlackhansingh et al., 2011; Buki and Povlishock, 2006; Green et al., 2014; Gale et al., 1995; Johnson et al., 2013b; Gentry et al., 1988; Kraus et al., 2007; Kumar et al., 2009; Pischiutta et al., 2018; Scott et al., 2015). Thus, research to define the initiation and perpetuation of cerebral immunity is critical to spur targeted therapeutic development to minimize the deleterious consequences of unrestrained immune activation without impacting the protective actions of inflammation after TBI.

Inflammaging, an age-related increase in the chronic production of pro-inflammatory mediators, is a risk factor for accelerated aging, oncogenesis, cardiovascular and neurological disease, and premature death (Fulop et al., 2017; Ruparelia et al., 2017; Leonardi et al., 2018; Miller and Raison, 2016; Gorelick, 2010; Fabbri et al., 2015; Volpato et al., 2001; Soysal et al., 2016). Potential mechanisms underlying the initiation of inflammaging remain elusive, but genetic susceptibility, microbiome changes, oxidative stress, immune dysregulation, and infection are implicated. In addition, cellular senescence is a potent stimulus for inflammaging and chronic inflammation (Di Micco et al., 2021) while the presence of senescent cells correlates with cognitive aging and neurodegeneration (Cole et al., 2019; Franke and Gaser, 2019). Cellular senescence, first described over six decades ago by Hayflick and Moorehead, is characterized by irreversible cell cycle arrest, sustained viability with resistance to cell death, and increased metabolic activity (Hayflick and Moorhead, 1961; Tchkonia and Kirkland, 2018). Senescent cells functionally contribute to biochemical changes associated with neurodegeneration, including aberrant protein aggregation, metabolic changes, autophagy deficits, mitochondrial dysfunction, oxidative stress, and impaired neurogenesis (Chini et al., 2019; Xu et al., 2015a; Chapman et al., 2019; Nacarelli et al., 2019; Musi et al., 2018; Ogrodnik et al., 2019). Thus, cellular senescence may be an underexplored, yet critical factor in facilitating inflammaging and progressive neurodegeneration after TBI.

3. Does glial senescence promote post-traumatic neurodegeneration after TBI?

Moderate-severe TBI increased cell cycle markers and the number of senescence associated beta-galactosidase (SA-β-Gal) positive cells, a gold standard marker of cellular senescence, in the ipsilateral hemisphere for one-month post-injury (Tominaga et al., 2019; Schwab et al., 2021). SA-β-Gal was elevated throughout the cortex, hippocampus, and thalamus at one day and one month following a single or repeated moderate blast TBI in rats (Arun et al., 2020). Repeated mild TBI, induced by a concussive-like, closed head injury for three consecutive days, showed gene expression profiles consistent with activation of cell cycle markers, genotoxic stress responses, and inflammation over the first week post-injury (Schwab et al., 2021). With respect to cell type(s), post-mitotic neurons do not undergo senescence whereas glial senescence more readily occurs (Fielder et al., 2017; Moreno-Blas et al., 2019; Chinta et al., 2013; Cohen and Torres, 2019; Han et al., 2020). Moderate TBI elevated the expression of senescence markers in microglia, with more pronounced effects on microglial responses, leukocyte invasion, and senescence observed in aged mice, as compared to young mice (Ritzel et al., 2019). Interestingly, microglial proliferation in response to facial nerve axotomy exhibits age-dependent changes while aged microglia exhibit defects in phagocytosis, disrupted homeostatic functions, increased senescence, and potentiated inflammation (Koellhoffer et al., 2017; Conde and Streit, 2006). Moreover, senescent astrocytes are observed in both aged and Alzheimer’s brain tissue (Pertusa et al., 2007; Limbad et al., 2020; Bhat et al., 2012). Similarly, SA-β-gal activity, increased for over two weeks post-TBI while elevated expression of p16^INK4a, a cyclin-dependent kinase inhibitor associated with senescence, was specifically elevated in astrocytes after TBI (Tominaga et al., 2019; LaPak and Burd, 2014). While temporally correlated with cognitive decline, it is unknown whether glial senescence causes neurodegeneration or reflects ongoing pathology. In support of the former possibility, genetic clearance of senescent glia slowed neurodegeneration and improved cognitive function in a murine tauopathy model (Bussian et al., 2018). In addition, aged astrocytes reduce neurotrophic support while astrocyte senescence is associated with an elevated risk of excitotoxicity, an important cause of neuronal loss and post-traumatic seizures, after TBI (Pertusa et al., 2007; Limbad et al., 2020). The mechanisms underlying the cell-type and age-dependent effects are unknown; however, a higher baseline of cellular senescence in the aged brain could explain why chronologically older individuals exhibit worse neurological outcomes after TBI (Thompson et al., 2006). Further analysis of cell-type specific mechanisms underlying the initiation of senescence may provide valuable insights into determining the wide range of outcomes after TBI.

4. Senescence associated secretory phenotype (SASP) and chronic inflammation

Depending on the cell type and initiating stimulus, approximately 30–70% of senescent cells undergo a phenotypic change to develop a senescence associated secretory phenotype (SASP). SASP consists of increased, cell-dependent secretion of interleukins, cytokines, chemokines, and matrix metalloproteinases to potentiate senescence and perpetuate inflammaging (Coppe et al., 2008; Kuilman and Peeper, 2009). Genotoxic stress, mTOR signaling, Jak2/STAT3 signaling, or mitochondrial dysfunction drive the expression of SASP-associated mediators, at least in part, via activation of the pro-inflammatory transcription factors, nuclear factor κB (NFκB) and CCAAT/enhancer-binding protein β (CEBP/β) (Hernandez-Segura et al., 2018).We and others reported elevated expression of SASP-associated factors, including interleukin-1β (IL-1β), IL-6, IL-8, tumor necrosis factor-α (TNF-α) and chemokines, such as C-X-C motif chemokine ligand 10 (CXCL10), C-C motif chemokine ligand 2 (CCL2), and CCL5 following TBI (Vaibhav et al., 2020; Woodcock and Morganti-Kossmann, 2013). These inflammatory mediators simultaneously enhance immune cell recruitment to eliminate senescent cells and perpetuate inflammation. Cells exhibiting the SASP also release matrix metalloproteinases (MMP), including MMP-3, MMP-9, MMP-12 and serpins, such as plasminogen activator inhibitor-1 (PAI-1) that modify the extracellular matrix and induce both tissue fibrosis and destruction (Jiang et al., 2017; Yamamoto et al., 2014). Interestingly, fibrosis and glial scarring persist for years after TBI (Di Giovanni et al., 2005; Burda et al., 2016) and contribute to chronic neurodegeneration (D’Ambrosi and Apolloni, 2020). Finally, release of exosomes and microRNA from senescent cells modulate stem cells and promote inflammation (Xu and Tahara, 2013). Thus, targeting the SASP may provide an innovative therapeutic approach to limit the long-term consequences of chronic inflammation after TBI.

5. Cellular senescence drives neurodegeneration

Cellular senescence drives tau aggregation (Musi et al., 2018) and is associated with Aβ pathology, the accumulation of the neurotoxic oligomer characteristic of AD and CTE (Stein et al., 2014). In a reciprocal manner, Aβ accelerated cellular senescence, as quantified by increased p16^INK4a expression, and worsened cognitive function in a 5xFAD mouse model of AD (Wei et al., 2016). AD patient brains showed increased senescence in oligodendrocyte precursor cells while administration of senolytic drugs reduced cellular senescence, decreased neuroinflammation, prevented neurofibrillary tangles, decreased neurodegeneration, and improved cognitive outcomes in rodent AD models (Musi et al., 2018; Zhang et al., 2019; Mendelsohn and Larrick, 2018). Similarly, genetic elimination of senescent cells prevented tau pathology and slowed cognitive decline in transgenic mouse models of tau-dependent neurodegeneration (Bussian et al., 2018). Consistent with pre-clinical modeling, evidence of glial senescence and SASP phenotypes were observed in post-mortem brains from contact athletes with a history of brain trauma (Schwab et al., 2019); however, many classic neuropathological signs of dementia were absent in this cohort, despite clinical symptoms and neurological deficits during life (Schwab et al., 2019), raising the intriguing and largely unexplored possibility that senescent cells drive neurodegeneration.

6. Genotoxic stress: a initiating factor in cellular senescence after TBI?

Cellular senescence is triggered by diverse stimuli, including genotoxic stress, telomere attrition, oncogenes, stress, protein aggregation, reduced autophagy, reactive metabolites, oxidative stress, mechanical stress, danger associated molecular patterns, pathogens, and exposure to inflammatory mediators (Fig. 1). Of these inducers, oxidative stress, a metabolic state whereby the production of reactive oxygen species overwhelms endogenous antioxidant defense, develops immediately after acute brain injuries and is believed to induce senescence, at least in part, via DNA damage. Genotoxic stress is associated with the induction of senescence pathways, including elevated expression of the cyclin-dependent kinase inhibitors, p16^INK4a and p21^WAF1/Cip1 (Rayess et al., 2012; Yosef et al., 2017), or the tumor suppressor gene, p53, which maintain cell cycle arrest (Rufini et al., 2013). While the precise mechanisms remain poorly defined, reduced DNA repair pathways may confer susceptibility to chronic neurodegeneration (Jeppesen et al., 2011; Madabhushi et al., 2014; Pessina et al., 2021), providing a potential candidate risk factor for further exploration.

Fig. 1.

Proposed mechanisms of senescence following TBI. A single moderate-severe TBI or repetitive mild TBI produces oxidative and mechanical stress, inflammation, and genotoxic stress. These events contribute toward the development of cellular senescence and a SASP, which potentiate cellular senescence, advance cognitive aging and drive inflammaging to accelerate progressive neurodegeneration. Administration of senolytic and/or senomorphic therapies removes or modulates senescence cells, respectively, to improve cognitive outcomes.

Oxidative genotoxic stress is present in multiple models of TBI (Morita-Fujimura et al., 1999; Zhang et al., 2002; Wang et al., 2014). Toward this end, moderate controlled cortical impact increased both single and double strand DNA breaks predominantly within neurons of the ipsilateral cortex and hippocampus over the first days post-injury (Clark et al., 2001). In parallel, diffuse staining for 8-hydroxy-2’-deoxyguanosine (8-OHdG), a marker of oxidative DNA damage, was elevated within both neurons and astrocytes of the ipsilateral cortex following severe TBI (Mendez et al., 2004). Moreover, double-strand DNA breaks were associated with early onset senescence and cognitive impairments after repeated mild TBI in mice (Schwab et al., 2021). In line with experimental data, a reduction in DNA repair pathways increased WM injury and axonal damage in professional athletes after mild TBI (Schwab et al., 2019) while genotoxic stress induced senescent oligodendrocytes, which reduced myelination and worsened axonal health, during chronological aging (Tse and Herrup, 2017). Consistent with these findings, mice with deficient DNA repair exhibited delayed neurological recovery after TBI (Tomasevic et al., 2012; Fujimura et al., 2000). Furthermore, elevated DNA damage markers in TBI patient blood correlated with injury severity while serum 8-OHdG levels associated with mortality after severe TBI (Shaked et al., 2014; Lorente et al., 2020).

Poly (ADP-ribose) polymerase (PARP) is a key DNA-binding enzyme that detects DNA strand breaks following genotoxic stress. Mice with genetic deletion of PARP exhibit increased genomic instability and elevated mortality after genotoxic stress (de Murcia et al., 1997). With respect to TBI, persistently elevated PARP expression is observed in brain mitochondria following controlled cortical impact, indicative of mitochondrial dysfunction and metabolic oxidative stress (Lai et al., 2008; Hill et al., 2017). Of note, PARP activation leads to energy depletion and activation of cell death pathways to exacerbate tissue damage (Andrabi et al., 2006). In support of this possibility, PARP inhibition attenuated microglial activation, reduced neurodegeneration, and improved functional outcomes after moderate TBI in mice (Stoica et al., 2014). Similarly, PARP inhibitors reduced injury volume after fluid percussion injury in rats (LaPlaca et al., 2001). Thus, early DNA damage responses may induce stress-induced cellular senescence to potentiate progressive neurodegeneration.

7. Senotherapeutics – the future of chronic TBI management

Senescent cells exhibit profound resistance to apoptosis, in part, due to upregulation in senescence cell associated anti-apoptotic pathways (SCAP) (Zhu et al., 2015). Interestingly, genetic elimination of as few as 30% of p16^INK4a positive senescent cells delayed the onset of aging-associated phenotypes and extended the healthy lifespan of progeroid mice (Zhu et al., 2015; Baker et al., 2011; Farr et al., 2017). Based on these exciting studies, a push to identify and/or develop senotherapeutic agents, a class of drugs that selectively target and remove senescent cells (senolytics), emerged to combat the detrimental aspects of aging. Senolytic agents, broadly divided into classes based on mechanism of action, include kinase inhibitors, inhibitors of Bcl-2 family members, natural compounds, p53 inhibitors, heat shock protein 90 inhibitors, and histone deacetylase inhibitors (HDAC). As TBI appears to prematurely accelerate aging, these approaches may provide an innovative approach to reduce the long-term deleterious consequences of TBI. Table 1 summarizes research to date studying senolytics in the pre-clinical and clinical management of TBI.

Table 1.

Senolytic treatments after pre-clinical and clinical TBI.

Drug	Dose/Route/Timing	Species	TBI Model	Main findings	Ref
Pre-clinical
Dasatinib	25 mg/kg, i.p., daily beginning at 2h post-injury	Mice	Moderate lateral fluid percussion	Reduced inflammation, neurodegeneration, and improved motor and cognitive outcomes after TBI	(Yu et al., 2016)
Quercetin	25 pmol/kg, i.p., 1h post-injury	Rats	Moderate anterior midline fluid percussion	Improved compound action potential amplitudes, reduced oxidative stress	(Schultke et al., 2005)
Quercetin	30 mg/kg i.p., daily for 3 days post-injury	Rats	Weight drop	Improved cognitive, reduced cell death, improved antioxidant enzymes expression	Yang et al., 2014
Quercetin	50 mg/kg, i.p., 0.5h post-injury	Mice	Weight drop	Improved mitochondrial function, increased antioxidant defense, reduced cell death, decreased gliosis, improved neurobehavioral outcomes	(Li et al., 2016a)- (Kosari-Nasab et al., 2019)
Quercetin	50 mg/kg, i.p., 0.5h, 12h, 24h post-injury	Rats	Weight drop	Reduced neuronal autophagy, decreased apoptosis, improved cognitive function	(Du et al., 2016)- (Du et al., 2018)
Quercetin	5, 20, 50 mg/kg i.p., 0.5h, 12h, 24h post-injury	Rats	Weight drop	Reduced edema, microgliosis, inflammation, and oxidative stress	(Song et al., 2021)
Navitoclax (ABT- 263)	1.5 mg/kg, i.p, 1 week post-injury	Mice	Repeated mild (controlled cortical impact) closed head injury	Reduced senescence markers, improved cognitive performance in males, but not females	(Schwab et al., 2022)
Fisetin	25, 50 or 75 mg/kg, i.p, 0.5h post-injury	Mice	Weight drop	Improved neurological function, reduced oxidative stress, decreased neuronal apoptosis	(Zhang et al., 2018)
Curcumin	500 ppm, p.o, 1 month pre-injury	Rats	Mild fluid percussion injury	Increased neurotrophin expression, reduced oxidative stress, improved cerebral energy homeostasis	(Sharma et al., 2009)
Curcumin	75, 150 mg/kg, i.p. 15 minutes pre-injury	Mice	Controlled cortical impact	Reduced inflammation, improved neurological outcomes	(Khayatan et al., 2022)- (Sun et al., 2020)
Curcumin	100 mg/kg, i.p., 5d pre-injury	Rats	Weight drop	Reduced oxidative stress, decreased lesion volume	(Samini et al., 2013)
Curcumin	100 mg/kg, i.p. 15 minutes pre-injury	Mice	Weight drop	Decreased microglial/macrophage activation, reduced neuronal apoptosis	(Zhu et al., 2014)
Clinical
Curcumin (Curcuminoids, C3 Complex)	500 mg, via enteral nutrition, daily for 7d	Human	ICU-admitted adult TBI patients	Improved nutritional status and reduced serum levels of IL- 6, TNF-alpha, MCP-1 and CRP compared to baseline	(Zahedi et al., 2021)
Luteolin (PEALut, Glialia)	700 mg palmitoyl ethanolamide+70 mg luteolin, oral, twice daily for 180d	Human	Moderate (GCS 9–13) TBI patients	Improved cognitive function assessed via mini mental state examination (MMSE) and brief neuropsychological cognitive examination (BNCE) and facilitated neural recovery on working memory	(Campolo et al., 2021)

Dasatinib, an FDA-approved tyrosine kinase inhibitor used to treat newly diagnosed Philadelphia chromosome-positive chronic myeloid leukemia, promotes apoptosis in part, via inhibition of Src kinase while quercetin is a naturally occurring flavonoid that gives apple peels a bitter taste. Combination treatment with dasatinib and quercetin, but not treatment with either compound alone, induced apoptosis in senescent human primary adipocyte progenitor cells and in human umbilical vein endothelial cells (HUVEC), but not in non-senescent adipocyte progenitor cells or HUVEC (Zhu et al., 2015). Administration of dasatinib and quercetin also removed senescent cells in vivo, improving outcomes in fibrotic pulmonary disease, reducing age-associated bone loss, improving physical function in old age, and improving outcomes after diabetic kidney disease (Schafer et al., 2017; Xu et al., 2018; Hickson et al., 2019).

Astrocytes acquire a senescent phenotype in Alzheimer’s disease patients while genetic clearance of senescent glia reduced gliosis, tau accumulation, and preserved cognitive function in a pre-clinical model of tauopathy (Bhat et al., 2012; Bussian et al., 2018). Similarly, administration of dasatinib and quercetin ablated senescent astrocytes and reversed tau-mediated neurodegeneration in experimental models of Alzheimer’s disease (Musi et al., 2018). In line with these findings, post-mortem analysis of athletes with a history of neurotrauma revealed widespread astrocyte senescence and activation of SASP pathways, which paralleled the development of neurodegenerative changes (Schwab et al., 2019). While the functional significance of this association remains unclear, intraperitoneal administration of dasatinib decreased inflammation, reduced neurodegeneration, and improved neurological outcomes, as compared to placebo control (Yu et al., 2016). Quercetin also provided beneficial effects in rodent models of TBI, reducing oxidative stress after moderate fluid percussion injury (Schultke et al., 2005), improving mitochondrial function and cognitive function after weight drop injury (Yang et al., 2014; Li et al., 2016a; Li et al., 2016b), and reducing gliosis and anxious behavior after repetitive mild TBI in mice and rats (Zabenko and Pivneva, 1994; Kosari-Nasab et al., 2019; Du et al., 2016; Du et al., 2018; Song et al., 2021); however, it remains undetermined whether these beneficial effects are due to the direct free radical scavenging properties of quercetin or whether these changes are due to downstream senolytic activity. The clinical utility of dasatinib and quercetin remains unexplored in TBI patients; however, Enzogenol, a quercetin containing extract derived from Pinus radiata bark, was safe, well tolerated, and appeared to improve cognitive function, enhance sleep quality, and reduce mental fatigue for up to one year after mild TBI (Theadom et al., 2013; Walter et al., 2017). Furthermore, dasatinib plus quercetin is currently under investigation in an exploratory open-label clinical study for early-stage Alzheimer’s disease (Gonzales et al., 2022). These promising pre-clinical and early-stage clinical studies suggest a potential benefit of senolytic therapy after TBI.

Curcumin, the principal curcuminoid of the spice turmeric, is a multi-functional compound with anti-aging properties that exhibits potent senolytic activity (Beltzig et al., 2021). Similarly, a putative curcumin metabolite, o-vanillin, and a curcumin analog, EF24, exhibit senolytic activity (Li et al., 2019; Zanzer et al., 2019). We and others reported that administration of curcumin reduced glial activation, decreased inflammation, and improved cognitive outcomes in rodent models of both mild and moderate-severe TBI (Khayatan et al., 2022; Laird et al., 2010; Li et al., 2022; Sun et al., 2020; Zhu et al., 2014; Wu et al., 2011; Wu et al., 2006; Sharma et al., 2009; Samini et al., 2013). Curcumin also exhibits protective effects against neurodegeneration in animal models (Ojha et al., 2012; Kim et al., 2012; Sundaram et al., 2017; Abrahams et al., 2019; Ulamek-Koziol et al., 2020; Nebrisi, 2021; Pluta et al., 2022) while curcuminoid supplementation exhibits beneficial effects on inflammation and clinical outcomes in critically ill TBI patients (Zahedi et al., 2021). Similarly, a randomized, placebo-controlled, double-blind study found curcumin supplementation slowed cognitive decline in aged adults (Rainey-Smith et al., 2016). In line with these data, administration of a cocktail containing luteolin, another natural compound with predicted senolytic activity (Vazquez et al., 2022), improved cognitive and neuropsychological function in moderate TBI patients (Campolo et al., 2021). While these promising experimental and early-stage clinical studies suggest a potential therapeutic role for senolytics, additional pre-clinical and randomized controlled trials are needed to define the precise mechanism of action and establish the efficacy of senolytics to improve cognitive outcomes after TBI.

Fisetin (7, 3’, 4’-flavon-3-ol), a natural flavonol present in strawberries, apples, persimmons, onions, and cucumbers, exhibits senolytic activity, at least in part, by inhibiting anti-apoptotic Bcl-2 family members and other targets within SCAP networks (Yousefzadeh et al., 2018; Zhu et al., 2017). While fisetin has not been explored after clinical TBI, it is noteworthy that fisetin suppressed neuronal loss after experimental TBI (Zhang et al., 2018) and reduced neurodegeneration in experimental models of neurodegeneration (Hassan et al., 2022; Nabavi et al., 2016; Ahmad et al., 2017; Patel et al., 2012). In addition to natural senolytics that exhibit broad spectrum effects in vivo, newer compounds are under development to specifically target senescence inducing pathways. Navitoclax (ABT263), an experimental oral Bcl-2/Bcl-xL inhibitor, reduced the expression of senescence markers and improved cognitive function in male, but not female mice after repetitive mild TBI (Schwab et al., 2022), suggesting a possible sex-difference in the mechanisms of neurodegeneration after head injury; however, the impact of inhibitors of Bcl-2 family members after moderate-severe TBI or in clinical populations remain untested to date.

In addition to senolytics, which directly kill senescent cells, senomorphics modulate senescent cells to interfere with senoinflammation and the SASP to prevent inflammaging. Putative senomorphic therapies include telomerase activators, caloric restriction, sirtuin activators, mTOR activators, antioxidants, autophagy activators, and proteasome activators (Hubbard and Sinclair, 2014; Wei et al., 2017; Lamming et al., 2013; Chondrogianni et al., 2015; Tilstra et al., 2012; Xu et al., 2015b; Kang et al., 2017; Cao et al., 2011; Lee et al., 2016). While less characterized as compared to senolytic agents, caloric restriction, a nutritional, non-pharmaceutic intervention, increases longevity via activation of sirtuin 1 and modulation of mTOR, which affects cellular metabolism and autophagy. Caloric restriction of 50% of daily food intake for three months, suppressed microglial activation and reduced neurodegeneration in a cortical stab wound model of TBI in rats (Loncarevic-Vasiljkovic et al., 2012; Loncarevic-Vasiljkovic et al., 2016). Similarly, chronic dietary caloric restriction of 70% of normal food intake decreased lesion volume and improved spatial memory following weight drop injury in rat (Rich et al., 2010; Rubovitch et al., 2019). Whether the benefits of caloric restriction are due to senomodulation and/or other mechanisms of action remain untested and will be the subject of exciting future investigations.

8. Conclusions and future perspectives

TBI remains a worldwide public health issue that places a major burden on society. Compelling epidemiological data suggests TBI is a distinct neurodegenerative disorder that may be initiated in otherwise healthy individuals following a single moderate-severe TBI or repetitive mild head trauma; however, the mechanisms underlying the development of progressive neurological injury after TBI remain poorly defined, which contributes to the lack of efficacious therapies to improve long-term quality of life. Cellular senescence is a conserved mechanism of inflammaging and may represent a cause of premature cognitive aging after all types of TBI. Oxidative stress and genotoxic damage are key initiators of senescence. Consistent with this possibility, the polymorphic nature of DNA repair factors and sex-differences in genomic instability may be responsible for the clinical heterogeneity in outcomes after TBI (Fischer and Riddle, 2018; Leung and Hazrati, 2021). Thus, proactive targeting of the early genotoxic stress response may restrain the initiation of cellular senescence/SASP as a “first domino” that potentiates progressive neurodegeneration at all levels of TBI. As such, senolytics may provide an innovative therapeutic approach to improve long-term neurological outcomes after TBI. First generation senolytics are largely natural compounds with pleiotropic mechanisms of action, including the removal of senescent cells. Newer generation senolytics will optimize and more precisely target senescence to potentially limit or reverse post-traumatic cognitive aging. In addition to the selective killing of senescent cells by senolytic agents, senomorphic agents act by blocking SASP in senescent cells to control the deleterious aspects of chronic inflammation. Future translational studies are also needed in humans to further refine and optimize the timing, dosing, and routes of administration of senotherapeutics. These exciting topics offer great promise in the future of TBI management.

Data availability

No data was used for the research described in the article.