Inflammation and Neuroprotection in Traumatic Brain Injury

7/26/2019 Inflammation and Neuroprotection in Traumatic Brain Injury

1/8

Copyright 2015 American Medical Association. All rig hts reserved.

Inflammation andNeuroprotection in Traumatic Brain Injury

Kara N. Corps, DVM, DACVP; TheodoreL. Roth,MS; DorianB. McGavern,PhD

IMPORTANCE Traumatic brain injury (TBI) is a significant publichealth concern that affects

individuals in alldemographics.With increasing interest in themedical and public

communities, understanding the inflammatory mechanisms thatdrive the pathologic and

consequentcognitive outcomes can inform future research and clinical decisions for patients

with TBI.

OBJECTIVES To review known inflammatory mechanisms in TBIand to highlightclinicaltrials

and neuroprotective therapeutic manipulations of pathologic and inflammatory mechanisms

of TBI.

EVIDENCE REVIEW We searched articles in PubMedpublishedbetween 1960 and August1,

2014, using the following keywords: traumatic brain injury, sterile injury, inflammation,

astrocytes, microglia, monocytes, macrophages, neutrophils, T cells, reactive oxygen species,

alarmins, danger-associated molecular patterns, purinergic receptors, neuroprotection, and

clinical trials. Previous clinical trials or therapeutic studies that involved manipulation of the

discussed mechanisms were considered for inclusion. The final list of selected studies was

assembled based on novelty and direct relevance to theprimary focus of this review.

FINDINGS Traumatic brain injury is a diverse group of sterile injuries induced by primary and

secondary mechanismsthat give rise to cell death, inflammation, and neurologic dysfunction

in patients of all demographics. Pathogenesis is driven by complex, interacting mechanisms

thatinclude reactive oxygen species, ion channel and gap junction signaling, purinergic

receptor signaling, excitotoxic neurotransmitter signaling, perturbations in calcium

homeostasis, and damage-associated molecular pattern molecules, among others. Central

nervous systemresident and peripherally derived inflammatory cells respond to TBIand can

provide neuroprotectionor participate in maladaptive secondary injury reactions. The exact

contribution of inflammatory cells to a TBIlesion is dictated by their anatomical positioning as

well as the local cues to which they are exposed.

CONCLUSIONS AND RELEVANCE The mechanisms that drive TBIlesion development as well as

those that promote repair are exceedingly complex and often superimposed. Because

pathogenic mechanisms candiversify over time or even differbased on the injury type, it is

importantthat neuroprotective therapeutics be developed and administered with these

variables in mind. Due to itscomplexity, TBIhas provenparticularly challenging to treat;

however, a number of promising therapeutic approaches are now under pre-clinical

development,and recentclinicaltrials have even yielded a few successes. Given the

worldwide impactof TBIon thehumanpopulation, it is imperative that research remains

active in this area andthat we continue to develop therapeutics to improve outcome in

afflicted patients.

JAMA Neurol. 2015;72(3):355-362. doi:10.1001/jamaneurol.2014.3558

Published onlineJanuary 19,2015.

Video at jamaneurology.com

Author Affiliations: Viral

Immunology and IntravitalImaging

Section, National Institutesof

NeurologicalDisorders and Stroke,National Institutesof Health,

Bethesda, Maryland.

Corresponding Author: DorianB.

McGavern, PhD, Viral Immunology

and IntravitalImaging Section,

National Institutesof Neurological

Disorders and Stroke, National

Institutesof Health, 10 Center Dr,

Bethesda, MD 20892 (mcgavernd

@mail.nih.gov).

Section Editor: HassanM.

Fathallah-Shaykh, MD, PhD.

Clinical Review & Education

Clinical Implications of Basic NeuroscienceResearch

(Reprinted) 355


wnloaded From: http://archneur.jamanetwork.com/ by Cesar Pereira on 03/27/2016


2/8


Traumatic brain injury (TBI) is a diverse group of brain inju-

ries thatvary in cause, severity, pathogenesis, and clinical

outcome. As public awareness of TBI and its conse-

quences increases, there is a growing need to understand the un-

derlyingmechanisms anddevelop therapeuticinterventions. Within

theUnited Statesalone, nearly2 million peoplesustaina TBIannu-

ally, contributing to one-third of all injury-related deaths. Individu-

als from all nations and demographics are affected, including ath-letes, military troops, and individuals with unintentional injuries.1-3

Traumatic brain injury is a significant cause of mortality in children

and young adults, and the incidence in older individuals has in-

creased withthe averagelife span.4MildTBI(mTBI)isthemostfre-

quenttype diagnosed, typically resultingin post-TBI survival. Trau-

matic brain injury is suspectedto contribute to a variety of chronic

degenerativeprocesses, including chronic traumaticencephalopa-

thy, Alzheimerdisease,and Parkinsondisease.5 Traumaticbrain in-

jury is initiated by the application of mechanical force to the head,

whichcan occurwith orwithout lossof consciousness.This thentrig-

gers a series of cerebral events that depend in part on the nature

andlocationof theinjury. A major challengeassociated with treat-

ingpatientswithTBIisthediversepathologicandpathogenicmecha-

nisms that become operational after injuries. For example, TBI of-

ten promotes disruptionof blood-brain barrier (BBB)integrity and

theneurovascularunit, which canresultin vascular leakage,edema,

hemorrhage,and hypoxia.Other pathologicmechanisms includecell

death within the meninges and brain parenchyma, stretching and

tearing of axonal fibers, and disruptions at the junctions between

whiteand graymatter, stemmingfrom rotational forces thatcause

shearing injuries.6Allthese primary pathologic mechanisms are ac-

companied by cellularand molecular cascadesleading to inflamma-

tion andadditionalcell death.Thisreviewfocuses onour currentun-

derstanding of the sterile immune reaction to TBI and someclinical

successes in treating patients with TBI.

We searched articles in PubMed published between 1960

and August1, 2014, using the following keywords: traumatic braininjury, sterile injury, inflammation, astrocytes, microglia, mono-

cytes, macrophages, neutrophils, T cells, reactive oxygen species,

alarmins, danger-associated molecular patterns, purinergic recep-

tors, neuroprotection,andclinical trials.Clinical trials or therapeu-

tic studies that involved manipulation of the discussed mecha-

nisms were considered for inclusion. The final reference list was

assembled based on novelty and direct relevance to the primary

focus of this review.

Sterile Immune Reaction to TBI

Centralnervous system (CNS)residentand peripherally derived in-flammatory cellsrespond quickly to brain injuriesand caneven par-

ticipatein the repair process.7,8These responsesare commonlyre-

ferredto as sterile immune reactions. A previous study9 found that

the inflammatory gene expression profile is comparable between

mTBIand severe TBI,suggestinga commonresponse toboth forms

of injury. The acute cellular reaction to TBIincludesastrocytes, mi-

croglia,monocytes or macrophages,neutrophils, andT cells, which

are initiallyactivated in part by purinergicreceptor signaling.10,11 In

the following paragraphs, we describe the inflammatory response

toTBI inmore detail,focusingspecificallyon traditional immunecell

populations.Sterile immune reactionsare at least initiallydesigned

to be beneficial butcan becomedetrimentalin certain situations.

Danger Signals

Pathogens can trigger innate immune activation via pathogen-

associatedmolecular patternmolecules,which are conserved struc-

tureswithina classof microbes recognized by Toll-like receptors or

pathogen-recognition receptors. These innate signaling pathwaysallow plants and animals to respond quickly to invading microbes.

However, it is nowrecognized thattissuedamagein theabsenceof

microbial infection can trigger inflammasome and innate immune

activation throughthe releaseof damage-associatedmolecular pat-

ternmolecules(DAMPs),sometimesreferredto asdangersignals.12

Alarmins are endogenousDAMPs released by cellsundergoing no-

napoptotic death or by cells of the immune system.13 Some ex-

amples of alarmins include HMGB1, S-100 proteins, adenosine tri-

phosphate (ATP),uric acid,DNA or RNA, andinterleukin 1, among

others.AfterTBI, alarminsare undoubtedly released,14andthistrig-

gers a sterile immune reaction designed to restore tissue homeo-

stasis. However, theseverityand duration of injury canfoster mal-

adaptiveimmunereactionsthatbecomeinjurious.Apreviousstudy15

found thatATPrelease anddetectionvia puringericreceptors elicit

an acutely neuroprotective inflammatory response after mild cor-

ticalinjury, butsustainedimmune activationmay notalways beben-

eficial.For example,therapeuticblockade ofinflammasomeactiva-

tionreducedinnateimmuneactivationand severe TBIlesionsize.16

Thus, additional researchis required to better understand therules

that govern pathogenic vs nonpathogenic innate immune reac-

tions after DAMP signaling in theinjured brain.

PurinergicReceptor Signaling

Purinergic receptors are an evolutionarily ancient family of trans-

membranemolecules thatdetect ATP, adenosinediphosphate(ADP),

oradenosine.10,11Thereceptors aredividedinto2 basicclassesbased

onwhethertheyrespondto adenosine (P1receptors)or ATP or ADP(P2receptors).BecauseATP is a cellularsource ofenergy, it is main-

tained ata highintracellular concentration during steady-state con-

ditions. Aftertissue injury, ATPis released fromdamagedcells, which

triggers an immune reaction via purinergic receptor signaling. This

reaction canbe amplified by pannexin and connexinhemichannels

thatpump ATP fromhealthycells intothe extracellularspace. Ster-

ileimmunereactionsgenerallysubsideasATPisconvertedintoaden-

osine through a 2-step reaction that involves ectonucleoside tri-

phosphate diphosphohydrolase1 (CD39) and ecto-5-nucleotidase

(CD73). Astrocytes and microglia each express at least one these

ectoenzymes,17,18 allowing them to dampen ATP-mediated neuro-

inflammation.

Microglia

Microglia are highly dynamic CNS resident innate immune senti-

nels that originate from primitive myeloid progenitor cells during

development.19,20Microgliaparticipate in a variety of homeostatic

CNS functions, including synaptic plasticity and learning,21 andare

often the first responders to any inflammatory event that occurs

within the parenchyma.20 Microglia mediate neuron removal dur-

ingdevelopment via release of reactive oxygen species (ROS) and

can acquire a phagocytic phenotype without an inflammatory

response.20 Microglial expression of genes associated with neuro-

Clinical Review & Education ClinicalImplications of BasicNeuroscienceResearch TraumaticBrain Injury

356 JAMANeurology March2015 Volume 72,Number 3 (Reprinted) jamaneurology.com




3/8


protection is upregulated with age.22 Microglia express a large

number of surface and cytoplasmic receptors, and cumulative sig-

naling through these receptors determines whether microglia

remain in a ramified, sentinel state or take on various configura-

tions as a result of activation.23 In addition, microglia can sense a

large repertoire of exogenous and endogenous signals, allowing

for dynamic responses to sterile injuries and infectious agents

that can be injurious or neuroprotective, depending on thecontext.22

During CNS autoimmune disease, activated microglia phago-

cytose debrisand downregulate cellular metabolism in contrastto

disease-initiating, peripherally derived macrophages, which ap-

pearto playa moredestructiverole by promotingdemyelination.24

These datasuggestthatmicrogliaare notinherently neurotoxicdur-

ing development of a CNS autoimmune disease. After acute focal

brain injury in rodents, microglia similarly appear to play a neuro-

protective role.15 Using 2-photonmicroscopy, we revealed thatmi-

croglia respond within minutes of brain injury by extending pro-

cesses tothe glial limitansand circumscribingindividualastrocytes,

resemblingahexagonalhoneycombstructure(Figure1A-E,Figure2E

and H,and Videos1 and 2). This reaction was dependenton purin-

ergic receptor signaling (P2X4and P2Y12) and astrocytic ATP-

dependent ATP release via connexinhemichannels. In response to

cell death (eg, astrocytic cell death in the glial limitans), microglia

transformed into phagocytic cells that resembled jellyfish

(Figure 1A-E, Figure 2F andI, andVideos1 and 2). Jellyfish microg-

lia werehighly mobile andoften inserted themselvesinto the dam-

agedglial limitansin place of dead astrocytes, connecting together

to form a phagocytic barrier. This reaction was also dependent on

purinergicreceptorsignaling(P2X4,P2Y6,P2Y12)andconnexinhemi-

channels. When these microglia responses were inhibited locally

throughblockadeof purinergic receptorsignalingor connexin hemi-

channels, the pathologic mechanisms observed after brain injury

were moresevere.One of themost notable changes wasincreased

leakage of materialsthrough the gliallimitans intothe brainparen-chyma.These data suggest that microglianot only clean up debris

fromthe injured brain butalso helpmaintain glial limitans barrier in-

tegrity by sealing thegapsthatresult from dead or damaged astro-

cytes.Moreover, our dataare consistent withpreviousstudies25-28

thatlinkmicrogliainjuryresponses toATPreleaseand purinergic re-

ceptor signaling. Although it is conceivable that microglia re-

sponses become maladaptive overtime or afterexposure to differ-

ent combinations of stimuli,29 we propose that the acute role of

microglia in the focally injured brain is neuroprotective.

Monocytes and Macrophages

Monocytesare a multipotent populationof circulatingbonemarrow

derived leukocytes capable of differentiating into macrophages ordendriticcells afterinvasionof an infectedor injured tissue.30They

are also known to participate in diverse functions, such as phago-

cytosis, cytokine or chemokine release, antigen presentation, im-

munemodulation,and tissue repair. Inthe naivebrain,thereare also

populations of specialized macrophages that reside in the menin-

ges,choroidplexus,andperivascularspaces.31TheirroleinTBIpatho-

genesisis unknown.Anotherstudy15alsofoundthatmeningeal mac-

rophagesareamongthefirstcellstodieafterfocalcorticalinjuryand

may serve as an early source of alarmins and ROS (Figure 1A-C,

Figure2AandB,and Video1).Monocyte-derivedmacrophagescom-

ingfromtheblooddonotreachpeaknumbersinthedamagedbrain

ofanimalsand humansuntil 24to 48 hours after injury.32,33 Mono-

cytes are capable of crossing the bloodcerebrospinal fluid barrier

with neutrophils into the injured brain as a result of CCL2 produc-

tionby choroid plexus epithelium.34 CCL2is significantlyincreased

in the cerebrospinal fluid of patients with TBI.33 Examination of

CCL2/miceafterTBI revealedslight alterations incytokine expres-

sion but no changes in lesion size within the first week of injury.

33

However, when followed for a longer timeframe of 2 to 4 weeks,

CCL2/micehad improved functionalrecovery, suggestinga patho-

genicrole formacrophages during thechronicphase of TBI.Similar

results were obtained in CCR2/ mice after TBI.35 CCR2 is the re-

ceptor for CCL2, and deficiency significantly reduced the number

of lesion macrophages and increased hippocampal neuronal den-

sities, spatial learning, and locomotion when measured several

weeks after brain injury. Collectively, the data obtained in CCL2

and CCR2 knockout mice suggest that monocyte-derived macro-

phages play a pathogenic role in the chronic phase after TBI.

Additional studies are required to determine whether these cells

can participate in brain repair after TBI similar to what has been

described in models of spinal cord injury.36 Whether a macro-

phage is pathogenic or beneficial after tissue injury likely depends

on its state of differentiation.

Neutrophils

Neutrophils arean abundantpopulationof circulating leukocytesthat

are usually among the first responders to tissue injuries in the pe-

ripheryandCNS.37Neutrophilsare oftenviewedas a proinflamma-

tory cell population butareknown toplaya vitalrolein woundheal-

ing through their involvement in phagocytosis, metalloproteinase

release, and growth factor production. After tissue injury, neutro-

phils can help prepare the damaged environment for repair. Neu-

trophilsare rapidly recruitedto theCNS after TBIand enter through

meningealblood vessels andthe choroidplexus.15,32,38,39Theycan

also facilitate the recruitment of monocytes.37

A previous study40

focused on sterile injury of the liver found that ATP released from

the damaged tissue induced inflammasome activation in a P2X7R-

dependent manner. This activation in turn promoted rapid

recruitment of neutrophils through release of chemoattractants

(CXCL1 and CXCL2) and formyl peptides that guided these cells to

thesite of injury. After focal TBI, we observed that neutrophils are

similarly recruited in a P2X7R-dependent manner and arrive

within 1 hour of injury (Figure 2C). 15 Visualization of cellular

dynamics and localization by 2-photon microscopy revealed that

neutrophils localized primarily to the damaged meninges (instead

of theparenchyma), where they swarmed the area andinteracted

with dead cells. Antagonism of this response by blocking P2X7R

signalingincreased theamount of cell death in themeninges, sug-gesting a protective role for neutrophils in the meningeal space

after focal cortical injury.

Neutrophils are not always neuroprotective and have the ca-

pacity tobreak downthe BBBby releasing metalloproteinases,pro-

teases, tumor necrosis factor , and ROS. Inflammatory mediators

released after brain injury can facilitate this process by inducing a

hyperactivated state thatallowsneutrophilsto breach theBBB and

enter the CNS.41 On arrival, neutrophils have the potential to in-

duceneuronalcelldeathusingthesamesolublemediatorsthatbreak

down the BBB.42 A previous study43 revealed that neutrophils are

TraumaticBrain Injury ClinicalImplications of BasicNeuroscienceResearch ClinicalReview & Education

jamaneurology.com (Reprinted) JAMA Neurology March2015 Volume 72,Number 3 357




4/8


themost abundantcell population in circulation after TBIand cause

increasedexpression of oxidative enzymes indicative of activation.

Depletion of neutrophils with antiGr-1 antibodies after controlled

cortical impact in rodents reduced edema, microglia and macro-

phageactivation,and TBIlesion size,but didnot affectvascularleak-

ageat 24 to 48 hoursafter injury.44 These data reveal thatneutro-

phils can be pathogenic after open-skullcortical impact. However,

thecontribution ofneutrophilsto a CNSlesionmay depend ontheir

preciselocalization andstateofactivation. Open-skullcontrolled cor-

tical impact is highly disruptive tomeningeal architecture andlikely

favors neutrophil recruitmentto the heavily damaged brainparen-

chyma. These findings contrast with mild closed-skull cortical in-

jury,which maintainsmeningeal architectureand fostersa more se-

lective patternof neutrophil recruitment.15Todefinitively establish

the contribution of neutrophils to TBI pathogenesis, these cells

shouldbeevaluatedinmanydifferentmodelsofbraininjury.Itiscon-

ceivablethattheir contribution willdifferbasedon thenatureof the

injury.

Figure 1. Pathogenesisof Traumatic Brain Injury (TBI)

ATP

ROS

UDP

Glutamate

Skull

Normal mTBI

Dura mater

Meningealmacrophage

Blood vesselsArachnoid mater

Subarachnoidspace ROS

Fluid leakage,meningealcell death Neutrophils

Jellyfish microgliaMicroglial process

extension to the glial limitans

Glutamate

NMDA

Ca2+ Necroticneuron

UDP

ATPGlial limitans

Pia mater

CSF

Astrocyte

Corticalneuron

Microglia

A

20 m

Parenchymalcell death

250 m

Skull bone

Meningeal macrophages

Parenchyma

Meninges

Microglia

Microglia

B

D

C

E

A, Comparisonof brain anatomy in

the meninges and superficial

neocortexbeforeand after focal mild

TBI(mTBI). Theduramatercontains

numerous small vessels that arelined

by thin,elongatedmeningeal

macrophages.The subarachnoid

spacecontains vessels, fibroblastlike

stromal cells,and cerebrospinal fluid

(CSF).The glial limitans,composedof

astrocytic foot processes, lies

beneath thepia mater andformsa

barrierbetweenthe CSFand

underlyingparenchyma. Mild focal

braininjury mechanically compresses

the meningeal space, compromising

vascular integrityand inducing rapid

necrosis of meningeal macrophages

andstructural cells.Leakageof fluid

from meningeal blood vessels results

in edema,and damaged cells within

the meninges release reactive oxygen

species (ROS) and adenosine

triphosphate(ATP),initiating a sterile

immunereaction. B andC, Maximum

projections (5-m wide) areshownin

thexzplaneof 2-photonz-stacks

captured through thethinned skull of

CX3CR1GFP/+ mice(original

magnification20).

B, A representativeimageof an

uninjured mousereveals the

presence of meningeal macrophages

(green)in theduraand ramified

microglia (green)in thebrain

parenchymabeneaththe glial

limitans (white dotted line).

C, Thirtyminutes after focal mTBI,

meningealmacrophagesdie and

microglia relocate to theinjuredglial

limitans (arrowheads).Skull boneis

shown inblue. D and E,

Histopathologic analysis of the

superficial neocortexby confocal

microscopy8 hours after mTBI

(original magnification20). D, An

uninjuredbrainis shownfor

comparison. Deadcells werelabeled

transcranially with propidium iodide.

Cell nucleiare blue.E, A large lesion

consistingof numerous deadcells

(red)(arrowhead). See Videos1 and

2. UPD indicates uridinediphosphate.






5/8


T Cells

Although T cells play diverse roles in adaptive immune responses

andthe regulation of inflammation, their role (ifany) in TBIpatho-

genesis is not clear. It has been proposed that autoreactive T cells

against CNS antigens, such as myelin basic protein, can be neuro-

protective after spinal cord injury.45 After brain injury, activated T

cellsare recruitedto sites ofdamage,46andROS release may facili-

tate this recruitment by activating endothelial barriers.47 To ad-

dresstheroleof T cells inTBI, a previousstudy48 examined the re-

sponseto closed-skullhead injury in RAG1 knockoutmice that lack

matureT and B cells.No difference in any pathologic or neurologic

parameters was observed between wild-type andRAG1-deficient

mice forup to1 week.Thesedatasuggestthat T cellsplay norolein

early TBI pathogenesis. Additional studies are required to deter-

mine whether T cells actively participate in chronic TBIlesions (be-

yond 1 week) and/or thereparativeprocess.

Therapeutic Modulation of TBI Pathogenesis

The pathogenesis of TBI is complex as reflected by the number of

clinical trials thathave failed to improve outcomes in humans.49,50

The many reasons for these failures have been discussed in other

reviews.49,50Rather thanfocuson thereasonsfor priorfailures, we

instead briefly discuss somesuccessesthat pertain to mechanisms

of pathogenesis and inflammation covered in this review.

Theconceptof freeradicalmediated damage of CNStissue af-

terinjury hasexistedfor several decades.51,52Administration of ef-

fectiveantioxidantshas thepotential to significantlylimit thespread

of damageand inflammationif given soon after brain injury. In ani-

malmodels, a number of previous studies53,54 have yielded prom-

ising results withantioxidants thatneutralizeROS. Forexample,in-

travenousadministrationof thesmall-moleculefree radicalscavenger

edaravone at 2 and 12 hours after weight dropinduced TBI re-

sulted in significantly reduced inflammation, edema, BBB break-

down,lesion size,and neurologicdeficits.53 Inhibitionof NADPHoxi-

dasecomplexassemblywithapocyninalsoreducedROSproduction,

BBB breakdown, and neuronal cell death after weight drop

induced TBI.54 The only caveatof this study was that the apocynin

was injected intraperitoneally 15 minutes before injury. Neverthe-

less,the favorable outcome implicatesNADPH oxidase as a poten-

tial source of ROSafter brain injury.

Using a newmodelof mild cortical injury, we found that trans-

cranial administration of the antioxidant glutathione at 15 minutes

Figure 2. Inflammatory Reaction to Traumatic Brain Injury

A

G

B

E

H

C

F

I

50 m

20 m

50 m

50 m

Normal Ram if ie d Microgl ia Hon eycom b Microgl ia Phagocytic Jel lyfish Microgl ia

Normal mTBI Myelomonocytic Cells

Meningeal Macrophages

D

A-I,The 25-mxymaximum

projections from CX3CR1GFP/+ (A,B,

andD-I)or LysMGFP/+ (C)micewere

captured by 2-photon microscopy

through a thinned skull.A, Meningeal

macrophages (green) are thin,

elongated cellsthat residealong the

dural blood vessels in theuninjured

brain.B, After focal mild traumatic

braininjury (mTBI), meningeal

macrophages undergo necrosis

within30 minutes and disappear

fromthefield of view.

C, Myelomonocytic cells(green)

invadethe damaged meninges within

anhourof braininjury. D and G,In the

uninjured brain,microglia (green)

have small cell bodies andare highly

ramified. Focalbrain injuryinduces

the rapidtransformation of microglia

into at least2 distinct morphologic

patterns.E andH, Honeycomb

microglia extend processes that

circumscribethe bordersbetween

individual astrocytesin the glial

limitans. F andI, Phagocytic jellyfish

microglia are generatedin response

tocelldeathandforma film across

the damaged glial limitans.

High-magnificationviews in panels G

through I aredenoted with white

boxesin panelsD throughF (original

magnification20). See Videos1

and 2.






6/8


or 3 hours after injury significantly reduced inflammation, glial limi-

tansbreakdown,and parenchymal(but notmeningeal)cell deathby

up to approximately 70%.15Pretreatmentwith glutathione reduced

meningealcell deathby approximately50%. Thesedataindicate that

ROS area primary inducer of cell death andinflammationafter focal

brain injury andthatan antioxidant canhavea major effecton lesion

expansion if givenearly.The advantageof passing a neuroprotective

compounddirectlythroughtheskullbone(transcranialdelivery)isthatahighlocaldrugconcentrationcanbeachievedintheCNSwithalim-

itedoff-target effect on theperiphery.

Previous studies55,56 have supported antioxidants as neuro-

protective agents in rats and humans, revealing that administra-

tionofN-acetylcysteine reduces brain damage andimproves recov-

eryafterTBI.N-acetylcysteineisthecellularprecursortoglutathione.

A randomized, double-blind, placebo-controlledclinicaltrial55 was

performed toassess efficacy inmembers ofthe military whoexpe-

rienced a mTBI that resulted from blastexposure. Patients whore-

ceivedN-acetylcysteine within 24 hours had significantly im-

proved recovery during a 7-day period when compared with a

placebo control group. These findings were corroborated in 2 dif-

ferent rodent models of TBI (weight drop and fluid percussion),

whichrevealed thatN-acetylcysteine reversed the behavioral defi-

cits associated with mTBI and moderate TBI.56 Further studies are

needed to determine whether this promising neuroprotective in-

tervention willbe efficaciousin patientswith diverse typesof brain

injury.

Many clinical trials have been completed or are under way to

assess therole of excitotoxicmechanismsin TBIpathogenesis.49,50

With the exception of amantadine, all drugs in this class tested to

datehavenot beeneffective inpromoting recoveryin patientswith

TBI. Amantadine is thought to act as anN-methyl-D-aspartate re-

ceptor antagonist and an indirect dopamine agonist. When pa-

tientswithTBI were treated duringa 4-week periodbeginning4 to

16 weeksafterinjury, amantadineimproved recoveryrelative tothe

placebo control. The mechanismunderlying thispositive effectre-mains unclear. Prevention ofN-methyl-D-aspartate receptor

mediatedexcitatorydamageseemsunlikelygiventhatthe drugwas

administered a monthor more after theinitial injury.57

Manipulationof purinergicreceptorsignaling is another thera-

peutic approach worth considering. Use of specific purinergic re-

ceptor agonists and antagonists should allow therapeutic amelio-

rationof differentTBI lesionparameters.A previous study15 found

that microgliaresponsesafter mTBI weredependent onP2X4,P2Y6,

andP2Y12receptors,whereasP2X7Rsignalingwasnecessaryforneu-

trophil recruitment. It might be possible to promote neuroprotec-

tiveinflammatory responsesthrough therapeutic agonismof these

pathways after brain injury. The challenge, however, with puriner-

gic receptor manipulation is that specific receptors are often ex-pressed on multiple cell populations. A purinergicreceptor agonist

or antagonist will likely affect multiple cell populations simultane-

ously. As an example, a previous58 study found that P2X7R local-

ized to astrocytic end feet and antagonism of this receptor re-

duced astrocyte activation, cerebral edema, and neurobehavioral

abnormalities after controlled cortical impactinduced TBI. A simi-

larprotectiveeffectwasobtainedbyblockingP2X 7Rafterspinalcord

injury, which was linked to receptor expression on spinal cord

neurons.59However, P2X7R isalso expressed oninflammatory cells,

and a previous study15 found that antagonism of this pathway in-

creased meningeal cell death after mTBI, likely due to diminished

neutrophil recruitment. Thus, purinergic receptor modulation can

positively affect one CNS environment and negatively affect an-

other.It willtherefore be important in future studies tomap outthe

exactcontributions of specific purinergicreceptors to differentTBI

lesion parameters beforedeciding which (ifany) arebest to target

therapeutically in patients.

Discussion

The pathogenesis of TBI is initially induced by a mechanical injury

that sets into motion a complex secondary reaction mediated by

ROS, purines, calcium ions, excitatory amino acids, and DAMPs,

amongothers.Thispathogenesis in turntriggers a robuststerileim-

mune reaction that consists of CNS resident and peripherally re-

cruited inflammatory cells. The response is designed to be neuro-

protectiveandpromotewoundhealingbutcanbecomemaladaptive

over time, especially if thelesion remains active for weeks.Among

the earliest soluble mediators are ROS and purines. Both are re-

leased within minutes of brain injury and initiate an inflammatory

cascade. Even after mild focal cortical injury, ROS can damage the

glial limitans that separate themeningesand parenchyma, which re-

sults in lesion expansion within brain tissue. Vascular damage and

leakage represent another early hallmark of TBI pathogenesis that

can foster edema, hypoxia, and tissue destruction. After brain in-

jury,the innateimmune systemquickly mobilizesin responseto pu-

rines andalarmins,and astrocytes helporchestrate this responseby

serving as inflammatory amplifiers. Within minutes, resident mi-

crogliaareamongthefirsttoreactbyfortifyingCNSbarriersandpar-

ticipating in phagocytic cleanup. Neutrophils and monocytes ar-

rive shortly thereafter and preferentially survey injured meningeal

spaces if theCNS architectureremainsintact.Focal brain injury elic-

its an anatomically partitioned immune reaction (at least acutely)

with myelomonocytic cells tending to the damaged meninges andmicroglia responding within the parenchyma. Eventually, myelo-

monocyticcellscanenterthedamagedbrain,andstudies40-42have

found that their presence there is sometimes neurotoxic. How-

ever, sterileimmunereactionsare notinherentlyneurotoxicand are

usuallyelicitedtoprepareadamagedtissueforwoundhealing.Thus,

the entire contribution of immune cellsubsets to TBI lesions needs

to be considered before targeted therapeutic interventions can

be intelligently designed. Another important variable is time. The

exact contribution of immune cells to a TBIlesion mayin fact shift

over time. For example, an initially neuroprotective immune

response may become maladaptive as secondary inducers of tis-

sue destruction diversify.

Although TBI has proven difficult to treat, promisinginterven-tions lieon thehorizon. Given theimportance of ROSin TBIpatho-

genesisandthesuccesswithN-acetylcysteine in patients withmTBI,

clinical pursuit of antioxidant therapy seems warranted. The likely

key to success is early treatment with antioxidants so that TBI le-

sionexpansionand subsequentinflammationcan bestoppedas soon

as they are initiated. Because TBI lesions begin to expand within

hoursof injury, development of strategies to rapidly preserve brain

tissueis paramount. Thekineticsof lesionexpansionmustbe simi-

larlyconsidered whenattemptingto manipulate purinergicand ex-

citatoryneurotransmitterpathways, whichengagerapidlyafter in-






7/8


jury. Therapeutic targeti ng of t hese pathways has the greatest

likelihoodofworkingifadministeredsoonafterinjury.Forthechronic

phase of TBI pathogenesis, more research is required to under-

stand lesion dynamics. Over time, it may become necessary to

dampen maladaptive inflammatory responses andattemptto pro-

mote wound healing reactions, which would be challenging to

achieve without having a better understanding of chronic lesion

dynamics.

Conclusions

Traumatic brain injury encompasses a complex spectrum of inju-

ries that tax the neural-immune interfaceand canresult in perma-

nent neurologic dysfunction. Detailed knowledge of this interface

during the acute and chronic phases of TBI will help us design the

most efficacious interventions.

ARTICLE INFORMATION

Accepted for Publication: October 2, 2014.

Published Online: January19, 2015.

doi:10.1001/jamaneurol.2014.3558.

Author Contributions: DrMcGavernhad full access

to allthe data in thestudy andtakes responsibility

forthe integrityof thedataand theaccuracyof the

data analysis.

Study concept and design: Corps,McGavern.

Acquisition, analysis, or interpretation of data: All

authors.

Draftingof the manuscript: All authors.

Critical revision of the manuscript for important

intellectualcontent: All authors.Obtained funding: McGavern.

Administrative, technical, or material support:

McGavern.

Study supervision: McGavern.

Conflict of Interest Disclosures: Nonereported.

Funding/Support: This work wassupported bythe

National Institutesof Healthintramural program.

Roleof the Funder/Sponsor:Thefunding source

hadno rolein thedesign andconduct of thestudy;

collection, management, analysis, and

interpretation of thedata; preparation,review, or

approval of themanuscript; andthe decision to

submitthe manuscriptfor publication.

Additional Contributions: Ethan Tyler, MA, and

Alan Hoofring, MS, Medical Arts Design Section,

National Institutesof Health, helped withtheillustrationshownin Figure1. MessrsTyler and

Hoofring are salaried employees of the National

Institutes of Healthand were notdirectly

compensatedby ourlaboratoryfor their work.

REFERENCES

1. FeiginVL, Theadom A, Barker-Collo S,et al;

BIONIC StudyGroup. Incidenceof traumatic brain

injuryin NewZealand:a population-based study.

LancetNeurol. 2013;12(1):53-64.

2. JordanBD. Theclinical spectrum of sport-related

traumatic braininjury. NatRev Neurol. 2013;9(4):

222-230.

3. LoganBW,Goldman S,Zola M,Mackey A.

Concussive braininjury in themilitary:September

2001 to thepresent. Behav SciLaw. 2013;31(6):803-813.

4. Roozenbeek B, Maas AI,Menon DK.Changing

patterns in the epidemiologyof traumatic brain

injury. NatRev Neurol. 2013;9(4):231-236.

5. SmithDH, Johnson VE,StewartW. Chronic

neuropathologies of singleand repetitiveTBI:

substratesof dementia? NatRev Neurol. 2013;9(4):

211-221.

6. Blennow K, HardyJ, Zetterberg H.The

neuropathology and neurobiologyof traumatic

braininjury. Neuron. 2012;76(5):886-899.

7. DasM, MohapatraS, MohapatraSS. New

perspectiveson central and peripheralimmune

responses to acutetraumaticbrain injury.

J Neuroinflammation. 2012;9:236.

8. Finnie JW. Neuroinflammation:beneficial and

detrimental effects aftertraumaticbrain injury.

Inflammopharmacology. 2013;21(4):309-320.

9. Lagraoui M,LatocheJR, CartwrightNG,

Sukumar G, Dalgard CL, Schaefer BC. Controlled

cortical impactand craniotomyinduce strikingly

similarprofiles of inflammatorygene expression,

but with distinct kinetics. Front Neurol. 2012;3:155.

10. JungerWG. Immunecellregulation by

autocrine purinergic signalling. Nat RevImmunol.2011;11(3):201-212.

11. Eltzschig HK, SitkovskyMV, RobsonSC.

Purinergic signaling duringinflammation. N EnglJ

Med. 2012;367(24):2322-2333.

12. Matzinger P. Tolerance, danger,and the

extended family.Annu Rev Immunol. 1994;12:991-

1045.

13. Bianchi ME.DAMPs, PAMPsand alarmins:all we

need to know about danger.J Leukoc Biol. 2007;81

(1):1-5.

14. MansonJ, ThiemermannC, Brohi K. Trauma

alarmins as activators of damage-induced

inflammation. BrJ Surg. 2012;99(suppl 1):12-20.

15. RothTL, Nayak D,Atanasijevic T, Koretsky AP,

LatourLL, McGavern DB. Transcranialamelioration

of inflammationand cell deathafterbraininjury.

Nature. 2014;505(7482):223-228.

16. de RiveroVaccariJP, Lotocki G,AlonsoOF,

Bramlett HM, Dietrich WD,Keane RW. Therapeutic

neutralization of the NLRP1 inflammasomereduces

theinnate immune response and improves

histopathology after traumatic braininjury.J Cereb

BloodFlow Metab. 2009;29(7):1251-1261.

17. Niemel J,IferganI, Yegutkin GG,Jalkanen S,

Prat A, Airas L. IFN-beta regulatesCD73and

adenosine expressionat the blood-brainbarrier. Eur

J Immunol. 2008;38(10):2718-2726.

18. Braun N, Svigny J,Robson SC,et al.

Assignmentof ecto-nucleoside triphosphate

diphosphohydrolase-1/cd39 expressionto microglia

andvasculatureof thebrain. Eur J Neurosci. 2000;

12(12):4357-4366.

19. Ginhoux F, GreterM, Leboeuf M,et al.Fate

mapping analysis reveals thatadult microglia derive

from primitive macrophages.Science. 2010;330

(6005):841-845.

20. Nayak D,RothTL, McGavern DB.Microglia

development and function.Annu Rev Immunol.

2014;32:367-402.

21. Parkhurst CN,Yang G, NinanI, et al.Microglia

promote learning-dependentsynapse formation

through brain-derived neurotrophicfactor. Cell.

2013;155(7):1596-1609.

22. Hickman SE,Kingery ND, Ohsumi TK,et al.The

microglial sensome revealed by directRNA

sequencing. Nat Neurosci. 2013;16(12):1896-1905.

23. HuX, LiouAK,LeakRK,etal. Neurobiologyof

microglial actionin CNSinjuries: receptor-mediated

signaling mechanismsand functional roles. Prog

Neurobiol. 2014;119-120(Jun):60-84.

24. Yamasaki R, Lu H,ButovskyO, et al.Differential

roles of microglia andmonocytes inthe inflamed

central nervous system.J Exp Med. 2014;211(8):

1533-1549.

25. NimmerjahnA, Kirchhoff F,HelmchenF.

Resting microglial cellsare highly dynamic

surveillantsof brainparenchyma in vivo.Science.2005;308(5726):1314-1318.

26. Davalos D,GrutzendlerJ, YangG, etal. ATP

mediates rapid microglial response to local brain

injuryin vivo. Nat Neurosci. 2005;8(6):752-758.

27. Haynes SE,Hollopeter G, YangG, etal. The

P2Y12receptorregulates microglial activation by

extracellular nucleotides. Nat Neurosci. 2006;9(12):

1512-1519.

28. Koizumi S, Shigemoto-MogamiY, Nasu-Tada K,

et al.UDP actingat P2Y6 receptorsis a mediator of

microglial phagocytosis. Nature. 2007;446(7139):

1091-1095.

29. Zhang J,MalikA, ChoiHB,Ko RW,

Dissing-OlesenL, MacVicar BA. Microglial CR3

activation triggers long-termsynapticdepressionin

the hippocampusvia NADPHoxidase. Neuron.2014;82(1):195-207.

30. ShiC, Pamer EG.Monocyte recruitment during

infection and inflammation. Nat Rev Immunol.

2011;11(11):762-774.

31. Nayak D,Zinselmeyer BH,CorpsKN, McGavern

DB.In vivo dynamics of innateimmune sentinelsin

the CNS. Intravital. 2012;1(2):95-106.

32. SoaresHD,HicksRR, SmithD, McIntoshTK.

Inflammatoryleukocyticrecruitment and diffuse

neuronal degeneration are separate pathological

processes resulting from traumatic braininjury.

J Neurosci. 1995;15(12):8223-8233.

33. SempleBD, ByeN, RancanM, Ziebell JM,

Morganti-Kossmann MC. Roleof CCL2(MCP-1) in

traumatic braininjury (TBI):evidencefrom severe

TBI patients and CCL2-/- mice.J Cereb Blood Flow

Metab. 2010;30(4):769-782.

34. Szmydynger-ChodobskaJ, Strazielle N, Gandy

JR, et al. Posttraumatic invasion of monocytes

acrossthe blood-cerebrospinal fluid barrier.J Cereb

Blood Flow Metab. 2012;32(1):93-104.

35. HsiehCL, Niemi EC,Wang SH,et al.CCR2

deficiency impairsmacrophageinfiltration and

improves cognitive function aftertraumaticbrain

injury.J Neurotrauma. 2014;31(20):1677-1688.

36. ShechterR, LondonA, VarolC, et al.Infiltrating

blood-derived macrophages are vital cellsplaying






8/8


an anti-inflammatory role in recovery from spinal

cordinjury in mice. PLoSMed. 2009;6(7):e1000113.

37. Kolaczkowska E, KubesP. Neutrophil

recruitmentand function in healthand

inflammation. NatRev Immunol. 2013;13(3):159-175.

38. CarlosTM, ClarkRS, Franicola-Higgins D,

Schiding JK, KochanekPM. Expression of

endothelialadhesion molecules and recruitmentof

neutrophilsafter traumatic braininjury in rats.

J Leukoc Biol. 1997;61(3):279-285.

39. Szmydynger-ChodobskaJ, StrazielleN, ZinkBJ,

Ghersi-Egea JF, Chodobski A. Theroleof the

choroidplexus in neutrophil invasion after

traumatic braininjury.J Cereb Blood Flow Metab.

2009;29(9):1503-1516.

40. McDonald B, Pittman K, Menezes GB,et al.

Intravascular danger signals guideneutrophils to

sitesof sterile inflammation.Science. 2010;330

(6002):362-366.

41. ScholzM, CinatlJ, Schdel-HpfnerM, Windolf

J. Neutrophilsand theblood-brainbarrier

dysfunctionafter trauma. MedRes Rev. 2007;27(3):

401-416.

42. Nguyen HX, OBarr TJ, Anderson AJ.

Polymorphonuclearleukocytespromote

neurotoxicity through release of matrix

metalloproteinases, reactive oxygen species, and

TNF-alpha.J Neurochem. 2007;102(3):900-912.

43. LiaoY,Liu P, GuoF, ZhangZY,Zhang Z.

Oxidative burstof circulatingneutrophilsfollowing

traumatic braininjury in human. PLoS One. 2013;8

(7):e68963.

44. KenneE, ErlandssonA, Lindbom L, Hillered L,

Clausen F.Neutrophildepletionreduces edema

formation and tissueloss following traumatic brain

injuryin mice.J Neuroinflammation. 2012;9:17.

45. CohenIR, Schwartz M. Autoimmune

maintenance and neuroprotection of the central

nervous system.J Neuroimmunol. 1999;100(1-2):

111-114.

46. LingC, SandorM, SureshM, Fabry Z.Traumatic

injuryand thepresence of antigen differentiallycontribute to T-cellrecruitmentin theCNS.J Neurosci.

2006;26(3):731-741.

47. ClausenF, LorantT,Lewn A, Hillered L. T

lymphocytetrafficking: a novel targetfor

neuroprotection in traumatic braininjury.

J Neurotrauma. 2007;24(8):1295-1307.

48. WeckbachS, Neher M,Losacco JT, etal.

Challenging the role of adaptive immunity in

neurotrauma: Rag1(-/-) mice lacking matureB andT

cellsdo not showneuroprotection afterclosed

headinjury.J Neurotrauma. 2012;29(6):1233-1242.

49. JanowitzT, Menon DK.Exploring newroutes

for neuroprotective drug development in traumatic

braininjury.Sci Transl Med. 2010;2(27):27rv1.

50. McConeghyKW,HattonJ, HughesL, CookAM.

A review of neuroprotection pharmacologyand

therapiesin patients withacute traumatic brain

injury. CNS Drugs. 2012;26(7):613-636.

51. KontosHA. Oxygenradicalsin CNSdamage.

ChemBiol Interact. 1989;72(3):229-255.

52. Bains M,Hall ED. Antioxidanttherapies in

traumatic brainand spinalcord injury. Biochim

Biophys Acta. 2012;1822(5):675-684.

53. Wang GH,Jiang ZL,Li YC,et al.Free-radical

scavengeredaravonetreatmentconfers

neuroprotection against traumatic braininjury in

rats.J Neurotrauma. 2011;28(10):2123-2134.

54. Choi BY, Jang BG,Kim JH,et al.Prevention of

traumatic braininjury-induced neuronal deathby

inhibition of NADPHoxidaseactivation. BrainRes.

2012;1481:49-58.

55. HofferME, BalabanC, Slade MD, TsaoJW,

HofferB. Amelioration of acutesequelaeof blastinduced mild traumatic braininjury by N-acetyl

cysteine:a double-blind, placebo controlled study.

PLoS One. 2013;8(1):e54163.

56. Eakin K, Baratz-Goldstein R, Pick CG,et al.

Efficacy of N-acetyl cysteine in traumatic brain

injury. PLoSOne. 2014;9(4):e90617.

57. GiacinoJT,Whyte J,BagiellaE, etal.

Placebo-controlledtrial of amantadinefor severe

traumatic braininjury. NEngl JMed. 2012;366(9):

819-826.

58. KimblerDE, Shields J,YanasakN, Vender JR,

DhandapaniKM. Activation of P2X7promotes

cerebral edema and neurologicalinjury after

traumaticbraininjury inmice. PLoS One. 2012;7(7):

e41229.

59. WangX, Arcuino G,TakanoT,et al.P2X7

receptor inhibitionimproves recovery afterspinal

cord injury. NatMed. 2004;10(8):821-827.




Documents

Inflammation and Neuroprotection in Traumatic Brain Injury