Mesenchymal stem cell therapy alleviates the neuroinflammation associated with acquired brain injury.

Abstract Ischemic stroke and traumatic brain injury (TBI) comprise two particularly prevalent and costly examples of acquired brain injury (ABI). Following stroke or TBI, primary cell death and secondary cell death closely model disease progression and worsen outcomes. Mounting evidence indicates that long‐term neuroinflammation extensively exacerbates the secondary deterioration of brain structure and function. Due to their immunomodulatory and regenerative properties, mesenchymal stem cell transplants have emerged as a promising approach to treating this facet of stroke and TBI pathology. In this review, we summarize the classification of cell death in ABI and discuss the prominent role of inflammation. We then consider the efficacy of bone marrow–derived mesenchymal stem/stromal cell (BM‐MSC) transplantation as a therapy for these injuries. Finally, we examine recent laboratory and clinical studies utilizing transplanted BM‐MSCs as antiinflammatory and neurorestorative treatments for stroke and TBI. Clinical trials of BM‐MSC transplants for stroke and TBI support their promising protective and regenerative properties. Future research is needed to allow for better comparison among trials and to elaborate on the emerging area of cell‐based combination treatments.

142 000 lives every year, and is also the leading cause of preventable long-term disability. 1 Moreover, the United States spends over $45 billion dollars every year on medications and healthcare services to treat and care for those affected. 1 Stroke patients also display an increased risk of developing dementia, which, in turn, may amplify their health and economic burdens. 2 Along with cognitive impairments, stroke patients often suffer paralysis and other physical impairments which entail exhaustive rehabilitation, contributing to stroke's high morbidity statistics. 2,3 Similarly, while less pervasive than stroke in terms of mortality, TBI caused approximately 2.4 million emergency room visits, hospitalizations, or deaths in the United States in 2010 alone. 4 Moreover, estimates indicate that 5.3 million Americans are presently living with disabilities resulting from TBI. 4 More recent assessments implicate TBI in approximately 82 000 deaths and 2.1 million hospital discharges yearly in Europe, and TBI is responsible for 37% of injury-related deaths in 24 European Union countries. 5 Hallmarks of TBI include bruising, bleeding, torn tissues, and other forms of physical damage to the brain that can lead to long-term impairment or death. Additionally, cognitive symptoms of TBI often involve problems with memory, attention, concentration, or thinking, as well as mood or behavioral changes, fatigue or lethargy, and alterations in sleep pattern. 4 Moreover, prior TBI is linked to increased incidence of other neurological disorders, such as Alzheimer's disease and Parkinson's disease, further increasing the long-term costs and health ramifications. 6,7

| CELL DE ATH CL A SS IFIC ATI ON IN AB I
As noted above, stroke and TBI share some overlapping pathologies, but are distinct from each other because stroke primarily ensues as a nontraumatic ischemic insult, whereas TBI obviously arises from a traumatic episode. Beyond these nontraumatic or traumatic events, these two ABI disorders display similar cell death features.
Primary cell death may manifest as either focal or diffuse, with the former characterized by the demise of cells within a localized brain area (referred to as infarcted core and ischemic penumbra or periinfarct for stroke, and impacted core and peri-impact area for TBI), while the latter presents more widespread cell loss including areas remote from the initial injured brain region. Indeed, the evolution of this remote cell death into secondary cell death after the onset of stroke and TBI has now been recognized to extend outside the brain, specifically to the spleen-a major source of inflammatory response-indicating that peripheral factors contribute significantly to secondary cell death. 4,8,9 Moreover, the severity of this secondary cell death may be influenced by age, as the young brain, which exhibits more plasticity than the adult brain, may respond more favorably via host brain repair after the insult. Additionally, based on temporal sequence of the cell death cascade of events, the initial insult is usually considered the acute stage, while secondary cell death is viewed as the chronic progression of cell degeneration. Although both stroke and TBI have been traditionally considered as acute brain disorders, accumulating evidence suggests that secondary cell death persists over long-term, with multiple cell death processes, in particular inflammation, exacerbating these progressive degenerative pathways. [9][10][11][12] Accordingly, the gradual nature of inflammation presents as an appealing therapeutic target for both stroke and TBI.

| INFL AMMATION PAR ALLEL S S ECONDARY CELL DE ATH IN S TROKE AND TB I
While the central nervous system (CNS) has been previously considered an immune-privileged system, accumulating evidence advances a dynamic neuroinflammatory interaction involving leukocytes and glial cells. 9 This aberrant inflammatory response plagues numerous neurological diseases. In stroke and TBI, the initial insult activates an acute inflammatory reaction to combat primary tissue damage, subsequently triggering the secretion of proinflammatory cytokines from resident microglia. 9,13,14 Additionally, increased permeability of the blood-brain barrier (BBB) allows peripheral leukocytes to intrude the injured brain. 9,15,16 Further, sustained microglial activation exacerbates chronic inflammation throughout the CNS, fueling a toxic environment that continuously aggravates secondary axonal degeneration and neuronal death. 9,17 Consequently, therapeutically suppressing the neuroinflammatory cascade represents a major aim of recent investigative efforts to reduce neurological damage following stroke and TBI.

| Acute inflammation in stroke and TBI
Following the onset of stroke, the acute inflammatory phase is characterized by elevated secretion of proinflammatory cytokines interleukin (IL)-6, IL-1β, and tumor necrosis factor (TNF)-⍺ into blood circulation and the cerebrospinal fluid (CSF). 15,16 While localized upregulation of TNF-⍺ and IL-1β has been primarily attributed to microglia 1 (M1) activated microglia, neurons also promote expression of IL-6. 16 Moreover, the ischemic microenvironment stimulates microglial elevation of cluster of differentiation 14 (CD14), a pattern recognition receptor on peripheral monocytes and a major component of innate immunity, indicating that resident microglia may mediate the acute inflammatory response following stroke. 17 These findings provide novel research avenues for recent efforts seeking to convert microglia from the proinflammatory M1 to the neuroprotective M2 phenotype, promoting the secretion of neurotrophic (eg, TGF-beta) and antiinflammatory (eg, IL-10) factors aiming to prevent further neuronal loss and facilitate tissue repair. 14,18 However, there is an increasing trend arguing against classifying inflammatory responses within the constraints of in vitro defined macrophage polarization phenotypes "M1" and "M2." There is evidence indicating the concurrent expression of both "M1" and "M2" phenotypic markers on the microglia/macrophages, suggesting that the polarization phenotypes cannot be easily confined within this M1/M2 binary nomenclature. Both stroke and TBI may induce a broad spectrum of simultaneous expression responses involving both pro-and antiinflammatory reactions, thereby demonstrating a heterogeneous inflammatory response in the injured brain. 19 The release of cytokines, chemokines, cellular adhesion molecules (CAMs), and matrix metalloproteinases (MMPs) by damaged neurons and auxiliary cells, such as microglia, astrocytes, and neutrophils, amplifies the neuroinflammatory cascade during the subacute phase of ischemic stroke and TBI. 4,16,19,20 Upregulation of MMPs, in particular, exacerbates localized inflammatory responses by increasing BBB permeability, thereby permitting peripheral leukocytes to infiltrate the injured brain. 19,20 Additionally, CAMs facilitate leukocyte adherence to cerebral vasculature, allowing further recruitment of cells to the injured area. Activated microglia and astrocytes prolong inflammation into the chronic phase via continued secretion of cytokines, chemokines, and CAMs, thus attracting more peripheral macrophages and neutrophils through the leaky BBB and other novel deleterious microglial and downstream signaling pathways. [21][22][23] Neuronal loss and cerebral edema may result from this progressive inflammation, compromising brain structure and function. 9,[19][20][21] Similar to stroke, the CNS following TBI undergoes a brief, neuroprotective phase during the acute inflammatory response following initial insult, yet this prosurvival stage is inadequate to provide neuroprotection for lasting inflammation. 12 The principal injury induced by TBI is physical, involving damage to neurons and disturbances of the BBB. 12 Following this primary mechanical injury, an acute "neuroprotective" phase and a chronic "neurodegenerative" phase are the two phases of an immune response analogous to those of stroke. 14 Microglial cells mobilize into a proinflammatory phenotype, and some cells manage regenerative/neuroprotective abilities to combat such injury during the acute stage. 14 For example, microglia may stimulate principal neurogenesis in the dentate gyrus of the hippocampus and extensive cellular generation. 13 Yet, the neuroprotection is inadequate because activated microglia that emit proinflammatory cytokines contribute significantly to acute inflammation, as seen with TBI patients possessing activated microglia two decades after initial injury. 12,13 Additionally, mouse models of TBI display a considerable increase in stimulated microglial cells in both white and gray matter at the TBI affected cortical location and in the neighboring ipsilateral and distal sections. 11

| Chronic inflammation in stroke and TBI
Following acute inflammation, a chronic neurodegenerative phase further contributes to neuroinflammation and coincides with stroke and TBI disease progression, modulated by both central and peripheral immune systems. 9,10,14 This fragile communication between brain-resident microglia and systemic lymphocytes must be coordinated correctly, 22 as disruption of this intricate balance exacerbates central neuroinflammation and, consequently, secondary cell death.
A compromised BBB and tissue damage of the parenchyma and cerebral vasculature is linked with the chronic neurodegenerative phase of stroke. 9,23 A weakened endothelial cell barrier promotes infiltration of serum proteins and immune cells, contributing to secondary BBB disruption and heightening the original injury induced by stroke. 15 This worsens physiological damage by elevating cerebral pressure and increasing secondary cell loss. 15 Ischemic stroke prompts an autoimmune reaction to neuronal antigens that could perhaps escalate or mitigate lasting neuroinflammation. 24 Akin to the case of stroke, peripheral immune cells infiltrate the TBI brain via the compromised BBB, promoting the secretion of proinflammatory cytokines, immune cell recruitment, and microglial activation. A decline in hippocampal neurons and reduction in cell propagation in the subgranular zone and ipsilateral subventricular zone further extend this vicious cycle of chronic neuroinflammation. 11 The toxic environment induced by secondary neuroinflammation may contribute to poor graft survival of cell transplants reported in laboratory investigations for cell therapy for TBI. 25 The secondary cell death cascade induced by chronic inflammation may be the connection between TBI and Alzheimer's disease (AD) neuropathology. 13 Many characteristic indications of Alzheimer's, specifically amyloid-beta (Aβ) plaques and neurofibrillary tangles, were discovered in the brains of patients with chronic TBI. 26 Maturing microglia's reduced phagocytic capability is associated with Aβ42 aggregation and thus, a reduction in microglial elimination of Aβ plaques. 13 Furthermore, TBI patient brains of all age groups displayed senile Aβ plaques for even children through post mortem analysis, indicating the source of AD to be TBI. 13 Beyond AD, TBI pathology is also related to many other neurological disorders upwards of 6 months following insult. 27 Several proteins involved in neurodegenerative disorders have been identified in postmortem TBI brains, produced during 4 hours to 5 weeks postinjury. Aβ plaques increase with beta-secretase, presenilin 1, and amyloid precursor proteins, and alpha-synuclein may be found within the axonal bulbs. 28 Alpha-synuclein serves as a presynaptic nerve cell protein that accumulates to produce harmful protofibrils, 29,30 secreted from damaged neurons and also aggregated in the CSF of infant TBI brains. 29,30 Synucleinopathy also bridges the connection between TBI and Alzheimer's disease to Parkinson's disease (PD). PD demonstrates a similar pathology of microglial activation to dopaminergic neuronal cell loss, characterizing a reactive gliosis which elevates proinflammatory cytokines in the brain and CSF of PD patients. 4 Lastly, microglia are the earliest to react to a damaging spinal cord injury and may sustain their activation for a minimum of 6 months following injury in humans. 31 Astrocytes and intraspinal neurons aid in maintaining this immune response by generating proinflammatory cytokine IL-1β. 31

| Central and peripheral sources of inflammation
Both central and peripheral systems intimately interact in neuroinflammation of CNS disorders, creating a hyperactive immune response that ultimately damages the neural tissue instead of repairing it. 9,32,33 Following stroke or TBI, central inflammation refers to the involvement of local CNS cells in the inflammatory response, while peripheral inflammation concerns the role of the systemic immune response involving peripheral organs, namely the spleen.
Elucidating the central and peripheral sources of inflammation, as well as potential mechanisms of action contributing to this disease progression, is critical in establishing therapeutic strategies to target the secondary cell death cascade in stroke and TBI. Recognizing that the ligand-receptor pair CCL20-CCR6 plays a key role in the chemotaxis of dendritic cells, effector/ memory T cells and B cells under homeostatic and inflammatory conditions, including stroke and TBI, we focus this section on CCL20.
CCL20 serves as a chemokine in CCR6 expressing cells. Using an experimental autoimmune encephalomyelitis (EAE) model, an animal model for brain inflammation, CCL20 operates as a ligand for CCR6, allowing homing of lymphocytes along with other leukocytes to neural tissue, 34 as well as tracking of Th17 or Th1 CD4 + cells which produce proinflammatory cytokines contributing to chronic neuroinflammation. 34,35 The expression of CCL20 in the choroid plexus aids the passage of CCR6 + T cells to invade the CNS of the EAE model, further allowing a CCR6 independent pathway of recruitment of T cells to the brain parenchyma. 34 Further, proinflammatory cytokines IL-6 and IL-17 elevate CCL20 expression. 35 Peripheral involvement in chronic inflammation occurs in stroke.
An analysis of the cytokine profile in mice following stroke reveals polarized T-cell responses dependent on the type of mice used. 36 C57BL/6 mice display a Th1 polarized response, while BALB/c mice display a Th2 polarized response, indicating that chronic neuroinflammation in stroke patients could arise from peripheral or central involvement depending on the individual. 36 Similarly, a lateral fluid percussion model of TBI demonstrates that the expression of CCL20 is increased in the spleen and thymus 24 hours postinjury, and in the cortex and hippocampus 48 hours postinjury, indicating a mechanism underlying peripheral involvement in neuroinflammation. [32][33][34] The expression of CCL20 in the spleen and thymus after TBI before that in the brain, along with reduced brain CCL20 expression following splenectomy, advances a peripheral mechanism of activation for CCL20 upregulation in the CNS. 32 These findings also implicate the role CCL20 in neuroinflammation following TBI. The increase of CCL20 in the spleen and thymus after TBI may imply that a peripheral signal stimulates neuronal degeneration. 32 In addition, other studies reveal that the liver may function in worsening the neuronal degeneration after TBI. Deficiency of hepatic Kupffer cells decreases ED-1-positive macrophage and neutrophil migration into an IL-1βinjected brain. 37 Peripheral immune and inflammatory systems (ie, spleen) function alongside central inflammation caused by microglia and other inflammatory mediators. Altogether, injury to the CNS induces a peripheral and central immune response contributing to neuroinflammation, resulting in a chronic inflammatory state that exacerbates neural degeneration and retards recovery.

| CELL-BA S ED THER APY FOR S TROK E AND TB I
The overlapping pathologies of stroke and TBI suggest that therapies that robustly attenuate cell death in stroke may likely prove effective in TBI and vice versa. Of note, cell-based regenerative medicine is shown to be effective in stroke (Table 1)

| TARG E TING INFL AMMATI ON WITH CELL-BA S ED THER APY IN S TROK E AND TB I
Many instances of ABI, specifically cases of stroke and TBI, have

| Bone marrow-derived mesenchymal stem cells for ABI
Key lab-to-clinic translational factors, including dosage, timing, and route of administration, influence the success of a cell transplant, particularly regarding its capacity to counter neuroinflammation following stroke and TBI. 48 However, the specific cell type used may have the most determinative power for cell transplantation efficacy.
Various cell types ranging from embryonic to engineered cells have been explored as cell donors for transplantation in stroke and TBI.
Although the pluripotency and multipotency endowed to embryonic/ fetal stem cells has previously established them as the yardstick of "stemness," these cells present notable ethical and safety considerations regarding the source from which they are harvested and high risks of tumorigenicity. Due to these logistical concerns, focus has shifted to other types of cells, specifically adult tissue-derived cells.
While most scientists concur that bone marrow-derived mesenchymal stem cells (BM-MSCs) do not fit the exact definition of "stem cells," they do display a similar capacity for brain repair. 48 BM-MSCs' adult tissue origin, robust safety profile, availability, neuroprotective and regenerative effects, and established research history advances them as an attractive candidate for cell-based therapy.
In in vitro and in vivo studies of stroke, BM-MSCs afford robust functional recovery in many models of brain disorders, promoting them as an attractive translational cell product. 48 In animal models, BM-MSC transplantation decreases brain damage and ameliorates motor and cognitive performance. In a hemorrhagic stroke rat model, transplanted BM-MSCs reduce inflammation after intraventricular infusion of BM-MSCs, resulting in decreased proinflammatory cytokine expression levels, including IL-6, IL-1α, and IFN-γ. 49 Furthermore, present evidence suggests that the release of growth factors by BM-MSCs or their neurotrophic exosomes, direct cell replacement, and encouragement of endogenous brain repair processes such as angiogenesis, neurogenesis, and synaptogenesis may be responsible, although the precise mechanism has not yet been identified. 48 Unfortunately, specific study design limitations such as underpowered trials, challenges in patient recruitment within the targeted disease phase, and inferior brain imaging capabilities, among others hinder mechanism-based analyses. Initially, studies visualized cell effects with a reductionist ligand-receptor model to determine the restorative mechanism of cell implantation, but this was later shown to be overly simplistic. 48 With scarce evidence sup-

Intra-arterial delivery
MSCs afford beneficial effects in animal models of stroke [103][104][105][106]122 Increasing translational potential of in vivo stroke models for testing MSC efficacy and safety

Autologous grafts
MSCs reduce stroke-induced neuroinflammation 4,123 Determine specific neuroinflammatory pathway targeted by MSCs

Biomarker, Allogeneic grafts
MSC grafts found to be safe 74,75,77,78 Long-term study is needed to determine any tumorigenic risk Safety measures MSCs stimulate endogenous neurogenesis 69,124 Determine specific neurogenic pathway targeted by MSCs Biomarker MSCs secrete neurotrophic and neurorestorative factors 125,126 Determine specific neurotrophic and neurorestorative pathway targeted by MSCs In line with these findings, further data suggest that transplanted

BM-MSCs dampen phagocytic activity and stimulate polarization
of microglia to the more neuroprotective and antiinflammatory M2 phenotype, thereby improving functional deficits in TBI rats. 62 Moreover, BM-MSCs may also enhance functional recovery by differentiating into neurons and astrocytes, 63 as well as via trophic support of endogenous neural regeneration. 64,65 While yielding favorable results without modification of the naive cell phenotype, augmenting the properties of BM-MSCs may increase their efficacy in treating TBI. For example, the homing accuracy of transplanted BM-MSCs to TBI-induced lesions has been markedly improved by stimulating these cells to excessively produce fibroblast growth factor 21. 66 Furthermore, in other current F I G U R E 1 Summary of acquired brain injury causes, pathophysiology, and relation to primary and secondary cell death. Mesenchymal stem cell transplantation stands as an attractive option for attenuating the inflammation induced by stroke and traumatic brain injury (TBI) TBI rat models, genetically modifying BM-MSCs to overexpress the antiinflammatory cytokine IL-10 has not only been associated with increased autophagy and mitophagy-which indicate the protection of neural cells from inflammation 67 -but also with enhanced immunomodulation concurrent with superior functional recovery compared to unmodified BM-MSC treatment. 68 Thus, although BM-MSC transplantation alone may be sufficient to provide neuroprotection, advancing these cells as a treatment for TBI may be improved by modifying these cells to optimize their therapeutic properties in the injured brain.
Accumulating evidence shows that the therapeutic effects of MSCs in different pathologies may be mediated by the paracrine secretion of a broad array of biological active molecules. 61,69 The secretomes and exosomes containing those molecules can be easily isolated from the cells and used as a biodrug. 70,71 In the case of neuroinflammation-related diseases, the intranasal administration of the secretome has been evaluated as a noninvasive and effective route to reach the brain. 72,73 Due to the translational value of this cell-free strategy, the use of secretomes and exosomes derived from MSCs warrants further investigation in the treatment of ischemic stroke and TBI.

| Clinical perspective on MSC therapy for stroke and TBI
Accumulating preclinical evidence supports the therapeutic po-

| CON CLUS ION
Stroke and TBI, two principal forms of ABI, pose a significant health and economic burden globally, and limited treatment options necessitate a novel therapeutic strategy to attenuate disease progression. [88][89][90][91] Primary cell death directly results from stroke or TBI, and the extent of this, brain damage is categorized as either focal or diffuse. Along with the influence of patient age, central and peripheral sources of immune cells prominently contribute to secondary neurodegeneration during both the acute and chronic phases following stroke and TBI. [92][93][94][95][96][97] Indeed, neuroinflammation stands as the common denominator that accompanies both disease pathologies and closely parallels secondary neural cell loss throughout their progression. As such, elucidating the dynamic involvement of both central and peripheral sources, especially the interplay between the brain and the spleen, is key to understanding the mechanisms underlying neuroinflammation. MSC transplantation as a regenerative biologic therapy targeting this deleterious inflammation has emerged as an innovative approach. Clearly, optimization of the cell therapy approach is warranted that will allow the transplantation strategies to accommodate the variable brain inflammatory responses.
Localized intracerebral delivery of MSCs may be more appropriate for focal injury as this will directly target the inflamed area. In con- Mesenchymal stem cell transplantation stands as a promising therapeutic approach for many types of acquired brain injury ABI, including ischemic stroke and traumatic brain injury TBI. These in- for stroke and TBI have developed rapidly over the past decades and will likely only continue to grow. Significant milestones may be reached upon the discovery of viable strategies to improve cell proliferation as well as the progression of MSC transplants to become an accepted standard of care for ABI disorders. MSC therapy as a stand-alone or combination treatment creates a nearly limitless set of research opportunities. As such, MSC therapy represents a fertile area for future research, and it is likely that many cell-based innovations and translational applications will continue to manifest in the coming years. The envisioned product is a safe and effective cell therapy designed to abrogate the inflammation-plagued secondary cell death associated with ABI.

ACK N OWLED G M ENTS
We thank the entire staff of Borlongan Neural Transplantation Laboratory for technical assistance and excellent scientific discussion.

CO N FLI C T O F I NTE R E S T
CVB declares patents and patent applications related to stem cell therapy. Additionally, CVB was funded and received royalties and