Inflammatory response in acute traumatic brain injury: a double-edged sword Maria Cristina Morganti-Kossmann, PhD,* Mario Rancan, MD," Philip F. Stahel, Mi and Thomas Kossmann, MD* Inflammation is an important part of the pathophysiology of traumatic brain injury. Although the central nervous system differs from the other organs because of the almost complete isolation from the blood stream mediated by the blood-brain barrier, the main steps characterizing the immune activation within the brain follow a scenario similar to that in other ‘organs. The key players in these processes are the numerous immune mediators released within minutes ofthe primary injury. They guide a sequence of events including expression of adhesion molecules, cellular infiltration, and additional secretion of inflammatory molecules and growth factors, resulting in ether regeneration or cell death. The question is this: to what extent is inflammation beneficial for the injured brain tissue, and how does it contribute to secondary brain damage and progressive neuronal loss? This review briefly reports recent evidence supporting the dual, the beneficial, ot the deleterious role of neuroinflammation after traumatic brain injury. Curr Opn Crt Care 2002,8:101-105 € 2002 Lippinolt Wiens & Wilkes, Deparimort of Tauna Surry, The Aled Hosa Patan, Asta "Deparment ot Sarge. Dison of Trauma Surry, Unies Hospital, Zueh ‘Sntztand snd "Deparment of Trauma ana Reconstructive Surgery, Unversty spl Berrian, Tha Fee Unwesiyof Bern Brin, Geary. Correspondence to Maia etna Megan Kessman, PAO, Deparment of ‘raura Sug, Aled Hospi ans Mona Unters, Cammreal Rel, Prat, Vetora, 3181, Astala; ema cistnamorgant-kossmanognedanonasheguay The wok presented by MCM-K and TK. has been supported by the Sis National cence Feuncaton rans 3136376, 92 31-8249094 ana {31-8248207 ana ty to Feandaone Hartnan Miler and Og Mayan (Curent Oplnon in Ciel Cre 2002, 8:101-108 Abbreviations BBB load ain barre CAM lereaier adhesion molecule + Tal taunt bran ry TOF-p —_vandorning growth factors TNF tinernacnt factor ISSN 1070-5205 ©2002 pinot Was &. Wiking, Ine Clinical aspects of traumatic brain injury A traumatic injury to the brain primarily affects young peopl lustrialized counties. Approximately 25% of these injuries result in long-term disabilities, causing fa- milial, social, and economic burden. Despite the steps taken to lessen the consequences of motor vehicle ac dents—particularly changes to safety rules for road traf= fic—morbidity and mortality are still high in these pa- tients. Even if the primary injury cannot be prevented, secondary brain damage can be reduced. Many scientists and physicians search for novel forms of treatment to neutralize the effect of neurotoxic pathways. The clinical progress achieved over che last 30 years in the man ment of traumatic brain injury (TBI) is based on the understanding of rapid intervention to treat the imme diate sequelae of TBI, like shock and hypoxia, which correlate with brain edema and mortality [1¢¢]. Also, the establishment of various intensive care procedures for monitoring and decreasing the raised intracranial pres- sure in severely injured patients has led to considerable improvement in outcome (2+) ‘The distinction between focal and diffuse brain injury involves not only diagnostic imaging techniques but also pathophysiology. The injuries often coexist, especially in severely injured patients. Focal lesions are easy to iden- tify and may be surgically removed, but diffuse axonal injury is often difficult to detect and treat [3]. The du- ality of TBI and the heterogeneity of injury patterns in the patient population may have led to the failure of several clinical trials of procedures successful in experi- mental settings [4]. Secondary brain injury begins imme- diately after the primary impact and is the result of a number of cascades, which, once activated, exacerbate the already altered homeostasis of the injured brain pa- renchyma, Increased vascular permeability, altered ionic balance, oxidative stress, excitotoxic damage, inflamma- tion, and mitochondrial dysfunction are the causes of edema, increased intracranial pressure, defective tissue perfusion, neuronal cell death, and neurologic impair- ment [5¢*]; these issues are the ones investigated most in clinical and experimental studies of neurotrauma. Inflammatory response Inflammation represents only one of the numerous pro cesses activated after TBI; it has not been fully eluci- dated due to the complexity of the large number of 101102 Neuroscience molecules shaping a sophisticated circuitry [6]. The ma- jor effectors in this cascade are the cytokines, multifune- tional peptides with, in some cases, redundant and op- posite effects on a variety of cell types, including resident cells of the nervous system [7#e]. Cytokines are usually released within minutes after challenge because they are stored intracellularly as precursor proteins even- tually modified into active molecules. Some are defined as proinflammatory mediators, usually the first upregu- lated, which induce the synthesis of anti-inflammatory cytokines, creating an autoregulatory feedback loop that re-establishes a resting immunologic status. Recent studies on the role of cytokines after acute brain injury have yielded conflicting results about whether their action contributes to repair mechanisms or exacer- bates the pathophysiology of trauma (Table 1). One as- pect of particular interest is the relation between inflam- mation and neuronal cell death, which, despite the neurologic improvement of the patient, is @ progressive process lasting long after TBI [8¢e]. A direct association has been found after experimental TBI between cyto- kine overexpression and neuronal cell death; however, based on in citro studies, it seems that cytokines do not cause neuronal apoptosis directly but may enhance the expression of molecules mediating apoptosis, such as Fas and Pas-ligind. ‘These and other proapoptotic or anti- ‘Tablo 1. Dual role of cytokines aftor traumatic brain injury Beneficial ‘Constitutive expression in normal brain Promote nouronal diferentaion and survival Induction of neurotrophic factors Anivnflammatory mediators: naroproteotve (Gytokine knockout mic: higher mertaty after experimental TBI Cytokine ocoptor knockout mice: enhanced tissue damage alter ‘experimental TB Deloterous "Association of corebral cytokine levels measured afer TA! with: Tissue inary Edema BBB dystunction Neuronal eel doath Neurological impsirment “Transgenic overexpression: nouradegoneraion Blockade of cytokine action with anibodios or immunosuppressive agents: better outcome after experimental TB Direct nourotoxity towards neuron in vitro oF asa cocktail in vivo In iro modulation of apoptotic factor synthesis, Both: beneficial and deleterious DDose-espanso studies on cultured neuronal cols: high doses are toxic, low doses are neuroprotective Distinction of early (deleterious) and delayed (beneficial functions in "TNF knockout mice subjected to brain jury This table summarizes the conficting data on the detrimental or protective offects of oytakines in the pathophysiology of traumatic brain injury (TB) refers toa large numberof studies publishod over the last fow years tht were performed on head trauma patients, in experimontal models of TEL, and ater eytckine exposure to cultured Cals iolate from rodent brains, Note the diveraty ofthe results obtained by administrating immunosuppressve drugs or specific ‘antibocies (Ab) ot by using mice lacking cylokine/cytokine receptor {gone expression in animal models of TAL BB, blood-brain barter, JTNF, tumor necrosis factor. apoptotic factors have been detected in rodent and hu- ‘man brains after TBI. Fas is of particular interest because it belongs to the tumor necrosis factor (TNF) receptor superfamily; both share some steps of the downstream pathway, including the activation of caspases, pivotal Fac- tors in programmed cell death. Fas and Fas-ligand are expressed on mouse neurons after experimental TBI [9] and have been shown to overlap with neurons undergo- ing cell death [10], Soluble Fas was also found increased in the cerebrospinal fluid of patients with severe TBI [11]. Fas did not correlate with the levels of 'TNF, or ‘with other immune mediators, which supports the notion that apoptosis is independent from inflammation, This review discusses immune mediators that have been analyzed extensively for the discrepancy between their protective and deleterious roles after TBI ‘Tumor necrosis factor ‘TNF is the cytokine most extensively characterized in neutologie diseases, including TBI [12,13+). TNF has been detected in various experimental models of TBI such as fluid percussion injury, closed head injury, and control cortical impact, and in human brain injury, leased mainly into the cerebrospinal fluid [6]. Its creased synthesis has been demonstrated in brain ho- mogenates and tissue sections, particularly in neuronal cells, Interestingly, ‘TNF levels have been associated with the neurologic deficit in rodents subjected to ex- perimental TBI [12,14]. ‘TNF is released very rapidly alter injury, suggesting the synthesis by either neuronal or glial cells rather than by infiltrated leukocytes, and slowly decreases to almost undetectable concentrations within a few hours after trauma, In vivo experiments based on the neutralization of TNF. action after TBI using general immunosuppressive agents or specific inhibitors have corroborated the hy- pothesis of the potential toxicity of this cytokine, be- cause the animals showed a better neurological outcome and decreased neuronal death, brain edema, and blood- brain barrier (BBB) dysfunction [12]. Deleterious prop- erties of TNF and other immune mediators have also been suggested according to the marked neurodegenera- ‘ion in transgenic mice overexpressing various cytokines, making their concentration an important parameter for their ultimate function as demonstrated in dose-response experiments on cultured neurons {7+8,15]. However, this concept has been revised by other studies on TNF knockout mice, demonstrating that there may be an early neurotoxic function (1-2 d) and a neuroprotective TNF function later in the posttraumatic phase (2-4 wk) {16]. Furthermore, mice lacking either both ‘TNF and lym- phoroxin-« or interleukin (IL)-6 showed a higher mor- tality rate than wild-type mice after closed head injury. However, in the surviving TNF/lymphotoxin- knock- out animals, the number of cerebral neutrophils and theInflammatory response in acute traumatic brain Injury MorgantiKossmann etal. 103 BBB function were unchanged; these animals had a ten- dency toward a better neurologic recovery [17]. This finding supports a beneficial role of these cytokines after head trauma. Neuroprotection mediated by ‘TNF in the pathophysiology of ‘TBI has also been described in mice lacking ‘INF receptors (p55 and p75), because the ani- mals showed enhanced tissue and BBB injury [18]. On the one hand, these data add valuable new information on the role of TNF after TBI, but on the other hand, they confirm that its function is more complex than ex- pected. It seems that both time and concentration play a fundamental role in driving toward neuroprotection or toxicity, suggesting alternative strategies for therapeuti~ cal intervention using anti-inflammatory agents, Interleukin-6 IL-6 is a multifunctional factor triggered by and follow- ing the rise of the early cytokines TNF and interleukin-1 [7**]. Compared with these, IL-6 seems to have anti- inflammatory effects, and its neurotrophic properties are supported by its ability to inhibit ‘TNF synthesis, induce the nerve growth factor, promote neuronal differentia- tion and survival, and counteract /V-methyl-D-aspartate~ mediated toxicity [6]. In addition, the constitutive ex- pression of IL-6 and of its receptor in the healthy brain indicates a possible role in maintaining a normal physi- logic function of the nervous system independent of neuropathology. Increased IL-6 protein has been dem- onstrated in a number of neurologic diseases of neuro- degenerative, viral, bacterial, autoimmune, or cancerous etiology [13¢,19]. In TBI, increased levels of IL-6 and IL-6R have been found in the brain tissue in various animal models, in human cerebrospinal fluid, and in the serum of patients with severe head trauma [134]. Re- cently, the work of our laboratory and of others has fo- ‘cused on the significance of IL-6 in the inflammatory response induced by TBI by observing ic in patients, animal models, and cell cultures. In patients with TBI, IL-6 markedly upregulated in the cerebrospinal fluid and serum was found to be associated with peripheral reactions like the acute phase response and with the release of nerve growth factor, also measured in the ce- rebrospinal fluid [20,21]. ‘These findings suggest that IL-6 might be synthesized in the brain and cross the damaged BBB, influencing other organ functions. “This hypothesis was confirmed later in a rat model of diffuse axonal injury; as shown in patients, increased IL-6 was detected in the cerebrospinal fluid and in the brain tissue at the mRNA and protein level (22]. Its function as an inducer of nerve growth factor was also demonstrated, stimulating cultured mouse astrocytes with the cerebro- spinal fluid of TBI patients containing known amounts of IL-6, resulting in the secretion of nerve growth factor into the cell supernatant [23] Recently, studies of IL-6 knockout mice have yielded interesting results about the protective function of IL-6. Penkowa ¢f al. [24] showed that after cryolesion in IL-6 knockout mice, a reduced astrogliosis was accompanied by the reduction of antioxidant enzymes, augmented in- ducible nitrie oxide synthase, and a number of apoptotic cells parallel with delayed tissue regeneration, Our group also reported a higher mortality rate in IL-6 knockout mice after closed head injury but no significant changes in the neurologic outcome among the survivors [17]. Also, a slower rate of recovery and a higher permeability of cerebral vessels were found in IL-6 knockout mice after trauma, whereas transgenic glial fibrillary acid pro- tein-IL-6 mice with a higher astrocytic IL-6 expression showed accelerated tissue revascularization and repair [25]. Independent of brain injury, previous studies ex- amining two different types of IL-6 transgenic expres- sion reported a higher toxicity in the glial fibrillary acid protein-IL-6 overexpression (astrocytic) [15] compared with the neuron-specific enolase-IL-6 overexpression (neuronal) that did not reveal any neurologic abnormali- ties [26], although both exhibited astrogliosis and mi- ‘eroglial cell activation. ‘This finding enhances the idea of the diverse effect of cytokines, depending on their cel- lular origin. Despite evidence of its protective function, whether as a simple association or a causative dependency, the levels of IL-6 in che cerebrospinal fluid of paticnts with TBI correlated with the cerebrospinal fluid concentrations of S-100B and neuron-specific enolase, which are consid- ered to be valuable parameters for the extent of brain damage. In addition, both $-100B and neuron-specific enolase reflected the size of contusion in these patients, which was estimated on the analysis of CT scans and S-100B levels in both fluids correlated with the Glasgow outcome scale. However, a direct association was not found between IL-6 and brain damage or between IL-6 and the neurologic outcome, but between IL-6 and neu- ron-specific enolase, a direct marker of neuronal destruc- tion [27]. Altogether, the results reported in different studies focusing on the role of IL-6 in the injured brain similarly favor its beneficial function, Chemokines, adhesion molecules, and leukocyte recruitment A hallmark of TBI, particularly after a focal lesion to the brain, is the accumulation of neutrophils and macto- phages, which further sustain the cerebral inflammatory cascade. This phenomenon is guided by an exact se- quence of events regulated by the secretion of cytokines, chemokines, and adhesion molecules, followed by cellu- lar extravasation and migration [28]. There is a discrep- ancy in the role of leukocytes after brain injury, because ic is believed thac they may aggravate posttraumatic neu- rodegeneration and the dysfunction of the BBB. How- ever, the data conflict as reported after blockade of n trophils in a model of controlled cortical impact injury {29]. Similarly, the upregulation of intercellular adhesion molecule-1 (ICAM-1), which is pivotal in mediating the104 Neuroscience extravasation of leukocytes across cerebral vessels, pre- viously associated with the size of cerebral contusion and BBB dysfunction after human ‘TBI [30], seemed irrel- evant in neutrophil infiltration as indicated in experi- ments on ICAM-1 knockout mice subjected to brain trauma [31]. Again, this finding contradicts the better outcome in studies on animal models of stroke using ICAM-I-deficient mice or antagonizing the function of ICAM-1 with specific antibodies [64]. Two recent stud- ies favor of the independence of ICAM-I from mediating, cellular infiltration. ‘The first is based on its delayed ex- pression compared with the peak of neutrophil accumu lation within the mouse brain after closed head injury [32]. The second demonstrates that ICAM-1 expression also increased in the cerebral vessels of rats with diffuse axonal injury, despite the absence of neutrophils in the brain parenchyma [33], suggesting that ICAM-1 may have functions other than cell adhesion. Accordingly, the chemokine monocyte chemotactive protein-I (acting on macrophages), but not macrophage inflammatory pro- tein-2 (acting on neutrophils), was increased in these animals, which supports the previous finding of macrophage/microglia activation in this model [34]. On the other hand, after focal brain injury, both chemokines increased very early, corroborating the finding of neutro- phil and macrophage accumulation within the cortical contusion [35]. Moreover, in these mice, not only mono- cyte chemotactive protein-1 and macrophage inflamma- tory protein-2 but also their respective receptors re~ mained upregulated differently over time. These studies confirm that, as in other biochemical pathways, inflam- ‘mation occurring in focal and diffuse brain injury is char- acterized by independent regulatory mechanisms. Anti-inflammatory cytokines: interleukin-10 and transforming growth factor-B ‘The role of the anti-inflammatory cytokines IL.-10 and transforming growth factor-B (TGE-B) has been studied often in neuropathology because of their therapeutic po- tential for suppressing the synthesis of proinflammatory cytokines and their molecular products [6,138]. Before neurotrauma, the beneficial role of TGF-B and IL-10 has, ‘been demonstrated in meningitis, multiple sclerosis, and the animal model of multiple sclerosis (experimental al- lergic encephalomyeliis), resulting in decreased edema, intracranial pressure, ‘TNF and IL-1 synthesis, and im- proved clinical outcome [138]. Elevated levels of IL-10 and TGF-B have been detected in the cerebrospinal fluid of patients with TBI. The hypothesis of a cerebral synthesis rather than passage across the impaired BBB was offered by the simultaneous analysis of BBB func- tion [36]. Surprisingly, in a similar study performed in children with TBI, levels of IL-10 were associated with higher mortality; however, this finding may not be rel- evant to its ultimate function (37). Instead, administra- tion of IL-10 in animals subjected to experimental TBI clearly demonstrated a diminished ‘TNE synthesis in conjunction with a reduced glial cell activation pattern and an improvement of the neurologic outcome [38,39], Increased production of TGF-B has been described in different animal models of brain trauma by perilesional astrocytes and macrophages, and in the facial nucleus after distal axotomy [40,41]. After fluid percussion dam- age, a raised TGF-B activity with biphasic kinetics was described in which the second TGE-B increase followed the early elevation of IL-6 and TNF, which suggests that the functional dependency of TGF-B from the inducing cytokines [42]. More research on these suppressive cy- tokines will be pursued over the next few years, which will, ideally, further elucidate their beneficial role in neurotrauma, Conclusions ‘This article reviews the work clarifying the function of the most important immune mediators in the patho- physiology of TBI. Some studies offer conflicting results about the role of inflammation after brain injury. How- ever, the differences in the experimental setting, the animal models used, and parameters analyzed may ac- ‘count for much of the variability in the results. In addi- tion, che more frequent use of cytokine knockout mice in these models may suggest the presence of compensatory mechanisms, which may mask the absence of immune mediators. So far, the beneficial use of antiinflammatory drugs with targeted inhibition of protein kinases acti- vated in immunoactivation has been reported in stroke patients (43¢*], who share several features with people suffering from neurotrauma. This achievement may help establish alternative therapeutic strategies for better control of secondary brain damage and neurologic out- come in TBL Acknowledgment ‘Tha autor thank. marta Anmann for er excallat echnical assistance over References and recommended reading Paps of pail ert uichad within the ann pete of oN, re “Or paca meres Ghotitndng terest 1 Bullock Rl, Chesnut RM, Citton Gt al: Managomant and prognais of trumatio bran my) Nevrraurs 2000, 17:440-507 Tin pec ns lh ural geo sn exelnt update on he guides fhe Clhios management of patente wih Read eau as a jt venue of various 2 Marsal LF; Henin: cen pat, present and ue, J Newoeugony S200, 47:846-563 {nis an nierestng overview fhe Westen of esinjred pationts ve rcant decades, 3 Graham 0} Mebtash TK Manwal WL ota: Recent advances in now {roma Neropaol Exp Newt 2000, 893641681 4 Reino Mi, Bllosk R: Clie! tials fn Moad injury. Nourol Res 1998,Inflammatory response in acute traumatic brain Injury Morgant-Kossmann eta ‘Ga RSB, KochanokP (ods): Brain jy. Boston: Kuwer Acasomic Pub ers 2001 ‘hia i an ene book dicated tothe rocent progress in the reach on he ripe Hochorical pathway ved in ate bran nr lacing oeecondary ut. MargantiKossmann MC, Rancan M, Oto Vt: Rle of corer in ‘mation raumatc ain nr arewstedconeapt Shock 2001, 18:465— 17, This or entensvervew ate wth moe dtl thn the present review, ‘Alan SM, Rothwoll Ni: Ojtotines and seutenourodegeneraion, Nat Rev Newose 201, 2:734~ 748 Experts nth ol of ejtokires in neuropathology, th authors rviow the ole of txtlinesia bran ier andiechomin partly: whuchivncl ncvssed nine proeont aril, Bee io the stor at ‘RaghupatR, Graham Dl, Melntosh TK: Apopose after taut brain iuy-J Nowotrira 2000, 17:927-990 in this excl’ speci isu ofthe Journal of Neuro trauma ocosing on apoptosis in bad and spinal cord ama, 0 " ‘eer R Franz G, Schopt Metal: Expression of Fas and Fas gan ater ‘oxpermenta auiat bran nr inthe rat. Careb Blood Fom Meta 2000, 20869-877, Grosjsan MB, Lonsinger PM, Stabe PF, oa: Impacto iano oxo: ‘KnasonFoafFe igardexrosion and epoptoi alle’ experimental closed oad uy label) Newratranma 2001, 185118), Lansinger PM, Marc A, TrntzO, etal: Prolonged invaoca laste of ‘soho Fea folowing evar aura briny ster] Neutenmured 2002, 122-161-168. ‘Shoham, ii | Hallenbeck JM: Dulle of unor necrosis ator aphain Drain iy. Cone Growth Factor Ror 198, 10:199- 130. “Morgant-Kossmana MC, OND VL Staol PF, tl: Tero of inflammation ‘nourloge dsessa Crtent Opin Cet Care 2000, 608-108, ‘Published recor ho present cual the article prowee an veri o the ‘alo efarmatn on nouopatichogis other han TBI such ae male slsone tnd AIDS-related demons, “ 6 6 ” 1° a \oblach SM, Fan L, Fadon A: Eay neuron expression of tance necrosis factoralpha ster expemanta rin ny cote to reurlogio wpa ‘ment J Nrcirmunol 1090, 05.118" 128, CampbolL: Stctcal ad uncoral impact ofthe tanegenc expression of ‘totnen inthe CNS. Arn NY Acad Se 1908, 84089-6 ‘Schobel, Raghspati R,Nakamusa Ml: Difereti cite and cree ‘egpenes of nor necrosis factor deficint meta wpernentltain uy. Proc all Aced So'U'S A 1909, 96:8721-8726, ‘Stahl PF, Shoham, Younis FM, tal: Exparinanil closed ead iru: nai of rourlogic outcome, baod-bain bare dyetnetion,nacran "nauopi iaton, and reurnal cl death in mice defi a ones for ‘rolammatory cytokines. J Core Biood Flow Metab 2000, 20:980-390, Sullan PG, Bruce Kelle A, Rabchealy AG, al: Exacerbation of damage tnd sored NE-kappascation mie ching tuor nero lator {Copters alter rau brain jy.) Noose 900, 10:0340- 6256, enveriso EN: Tho roe of ctchines in multiple scerosiafaoinmune Cophaltisand ter neutloic dsrders In Human Cytokines: Th Rolain Research and Therapy. Ete by Agarwal B, Pus R- Boson lacsoel Se noo; 1098:169~160. “McCisn ©, Cohen D, Philips Increased plasma and vei fd inteveukin tvel in patios with heed Ij. J Lad Cin Med 1991, Thenee281| Kossman T, Hare VH, hot HG ta: Itatecs and erm intro Spd io acute phaseretponse patente wth sever trauma bran urea Shock 1995, 4311-317. Hans VE, Kossman T, Lenanger PM, ot a Exponent axonal ny tig gor inieoutin6 mRNA, poten sythess and rlase ito orebroopina Aid.) Cereb Blood Fow Mota 1909, 10:184-194. ossmanaT, Han V not HG, Introd released in haan cre: 105 ‘rp fui lowing trsumatic brain jy ay ogee growth ectar production in aswoxjes Brin Roa 1906, 19:1 49-182, 24 Penkowa M, Gilt M, Cerasce Jo al: pad repent and increased ‘dav stese and naurodegenration ter ban injury tered ‘etoant mice. Gia 2000, 82:271-285, 25 Swart KR, Liv F, SewolO:Intaukin-8 promatespetraumatic hing in ‘tha cont nero eyetam Bran Roo 2001, B6:86-06, 26 Fate; Lazzaro D, Musa Petal: 8orpresion nnewons of tansgenic ‘ice caute reactive astrocytes and incense in rand macrogfalcale bat ‘ro petal damage. Eur Newrosi 1005, 72441-2446, 27 Pines UE, Morgen Kosemann MC, Rancan Mt: 81008 recs the ‘non olny and ovisome, whereas neural spect aclae a abt Sacetar of nauoiarmaion pants with avers raumat brain rary) Novrorauma 2001, 10491498. 28 Asovsio VC, Campbell: Cherains inthe CNS:plinctonl mediators inciverse sates, Trends Neve So 1999, 22504-512. 128, Whar My, Caos TM, Kochansk PU ta: Noutophia do net made biood ban barar pameablity ea aftr contoled cortical impactinras.) Nevroraama 1999, 161589-504, 130, Prine UE, Storr IF, Kotemane Tt a: Sold CAM in CSF coincides ‘wah the exe of corral damage m patents wh svere Wau brat Thuy. Neurone 1008, 1690-408, 831 Whalen Ml, Caos Ty, Kochanok PM, ta: Blood bin barr permoabiy, ‘rauttopsccimadston and vascular adhesion mlocule expression att ‘ontollad cota mpect rate a poiminary ety. AotaNewroohe Supp (Wen) 1008, 7212-214 82 Otto, Stal PF, Rancan Mot al Upregulation of :8AMP-2 and CAN ‘er sever traumatic bran ny cus by canes ndacar and wt ior, ort krates abaact], Reatravo Nowol Nowoeci 2000, 16188, 833. RancenM, Otto, Hans Vel: Uoregation of CAM and MOP but otf MP2 ond sensosto’ detctn espense to raumatis axonal urn tat] Nevose: Ree 2001, 63490-448. ‘94 CuukaE, Hans VI, Ammann E, oa: Ca aivation and inflammatory se panes folowing traumatic axonal ivy the rt. Neworepart 2000, '35-O Vi Stahl PF, Rancan Mt: Roguton of chomoknes and che ‘ine rocapiorsafor experimental closed head injury. Neuroreport 2001, 22086-2064, 136 Cats, MorgantsKossmann MC, Loranger PM ot IL-0 lve cae ‘roepial ud an serum of patina with saver veumatic bain ry rl fon fo 8, TNFaiphe, TGF-beta snd Hoodman: kon I New 137 Ba MU, Kachanek PM, Doughty LA, ata: tein and inten in ‘cerebroapal fad afer eovretnuatc brain yin chien. J Nowa” ‘wouma 1097, 16481=487. [38 Koblach SM, Faden A: Itekin-10 improves outcome and ars pri arama oyckne expression afer experimental raat bran ny. Exp "Nowol 1008, 159:149-181, 139° BalasngaV, Yong VW: Attn of trol reactivity byinterukin 10, ‘INeweec! 1096, 16:2048~2068, 40. Kole R nahin D,Kroutiborg GW: nee and veeioming growth fectorbata| mRNAs a indsoed fra ctl nile folowing motoneuron ‘ony Ear JNowoos 1993, 6:775-781 441 Logan A Frautschy SA, Gonzalez AM, a: Enhanced expression of ane forming grow actor ota | nto rt brah te ocalaad csetra uy. Bran Ree 1902, 867.216-225, 42 Remar! AG, Leki D, Sorano A ta: Biphasic tansloming growth {actonbea produesionfakang he pec ilaeatony cytokine esponsain co robrl trauma: Neuroreport 1908, 7199-130, 43 Barone FC, Parson AA: Therapeutic potential of antnfanimatory vgs in ‘= focal stoke. Expat Opin ines Orage 2000, :2281~ 2908. ‘hierovow power mporartiniormatin ont pont ute fainter, agonta in aka, which may also ba leant forthe managorant of TB