PROGRESS IN BRAIN RESEARCH

VOLUME 146

NGF AND RELATED MOLECULES INHEALTH AND DISEASE

EDITED BY

LUIGI ALOE

Institute of Neurobiology and Molecular Medicine, National Research Council (CNR), Via K. Marx 15/43,

00137 Rome, Italy

LAURA CALZA

Department of Veterinary Morphophysiology and Animal Production DIMORFIPA, University of Bologna,

Via Tolara di Sopra 50, 40064 Ozzano Emilia, Bologna, Italy

,

~

ELSEVIER

AMSTERDAM - BOSTON - HEIDELBERG - LONDON - NEW YORK - OXFORD

PARIS - SAN DIEGO - SAN FRANCISCO - SINGAPORE - SYDNEY - TOKYO

2004

Progress in Brain Research, VoJ. 146[SSN 0079-6123

Copyright I(;J2004 Elsevier B.V. All rights reserved

CHAPTER 24

Interactions between the cells of the immune andnervous system: neurotrophins as neuroprotection

mediators in CNS injury

Rinat Tabakman 1, Shimon Lecht, Stela Sephanova, Radar Arien-Zakay andPhilip Lazarovici *

Department of Pharmacology, School of Pharmacy, Faculty of Medicine, The Hebrew University of Jerusalem,Jerusalem 91120, Israel

Abstract: Inflammatory processes in the central nervous system (CNS) are considered neurotoxic, althoughrecent studies suggest that they also can be beneficial and confer neuroprotection (neuroprotective autoimmunity).Cells from the immune system have been detected in CNS injury and found to produce and secrete a variety ofneurotrophins such as NGF, BDNF, NT-3 and NT-4/5, and to express (similarly to neuronal cells), members of thetyrosine kinase (Trk) receptor family such as TrkA, TrkB and TrkC. Indeed, autocrine and paracrine interactions areobserved at the site ofCNS injury, resulting in a variety ofhomologic-heterologic modulations of immune and neuronalcell function. The end result of the inflammatory process, neurotoxicity and/or neuroprotection, is a function of the finebalance between the two cellular systems, i.e., of the complex signaling relationships between anti-inflammatoryneuroprotective factors (neurotrophins and other chemical mediators) and proinflammatory neurotoxic factors(TNF, free radicals, certain cytokines, etc.). Autoimmune neuroprotection is a novel therapeutic approach aimed atshifting the balance between the immune and neuronal cells towards survival pathways in a variety of CNS injuries.This review focuses on data supporting this concept and its future therapeutical implications for optic nerve injuryand multiple sclerosis.

Keywords: neuroprotection; neurotrophins; CNS injury; inflammation; lymphocytes; NGF; BDNF; Trk

Introduction

~

Neurotrophins are growth factors that play anessential role in the development and maintenanceof the nervous system. This family includes nerve

'This review is part of a PhD thesis to be submitted to theHebrew University of Jerusalem by RT.*Correspondence to: Philip Lazarovici, Department ofPharmacology and Experimental Therapeutics, School ofPharmacy, Faculty of Medicine, The Hebrew Universityof Jerusalem, P.O. Box 12065, Jerusalem 91120, Israel.Tel.: + 972-2-675-8729; Fax: + 972-2-675-8741;E-mail: lazph@md.huji.ac.il

DOl: 10. [0 16jS0079-6123(03)46024-X

growth factor (NGF), brain-derived neurotrophicfactor (BDNF), neurotrophin-3 (NT-3), neurotro-phin-4/5 (NT-4/5) and neurotrophin-6 (NT-6)(Barbacid, 1995; Lazarovici et aI., 1997). These areall synthesized as large precursor proteins, which arethen cleaved to physiologically active neurotrophinsby metalloproteinases (Lee et aI., 200 I; Chao andBothwell, 2002). Neurotrophins, by binding andactivating their respective receptors in differenttissues, induce beneficial effects that may be exploitedfor treatment of neurological disorders and cancer(Apfel, 1997; Haase et aI., 1998; Ochs et aI.,2000). Two families of receptors that interact with

388

neurotrophins have been found so far: the high-affinity tyrosine kinase (Trk) receptor, and the low-affinity neurotrophin receptor p75 (p75NTR), amember of the tumor necrosis factor receptorsuperfamily (Barbacid, 1995), which will not beaddressed in this review. The trk family of receptorsare similar to the receptor tyrosine kinases for othergrowth factors, but are selectively occupied andactivated by neurotrophins: TrkA binds NGF; TrkBbinds BDNF, NT-3 and NT-4j5, and TrkC bindsNT-3 (Kaplan and Stephens, 1994). By interactingwith their Trk receptors, the neurotrophinsinduce receptor autophosphorylation and recruit-ment of signaling protein substrates that bind to Trk.Activation of the Trk receptor triggers intracellularsignaling cascades, resulting in the regulation of cellsurvival, proliferation, differentiation and synapticplasticity (Chao et aI., 1998). Intracellular signalingcascades activated by trk receptors of major interestinclude the Rasjextracellular signal regulated kinase(ERK) pathway, the phosphatidylinositol-3-kinase(PI3K)jAkt kinase (also known as protein kinase B) .

pathway, as well as phospholipase C-yl (PLC-yl),which are all engaged in the promotion of cellsurvival (Lazarovici et aI., 1997; Freidman andGreene, 1999; Patapoutian and Reichardt, 2001).

Diseases of the central nervous system (CNS) suchas Alzheimer's, Parkinson's, Multiple Sclerosis,mechanical trauma and stroke, are accompanied bycell death, and are known as neurodegenerativediseases. Therefore, for therapeutic purposes we needto protect the CNS from neurodegeneration. Threedifferent approaches may be considered: (i) drugtherapy, to support the living neurons at the site ofCNS injury (Weiss et aI., 1990; Zuddas et aI., 1992),(ii) cell therapy, such as cell transplantation, toreplace damaged neuronal structures at the site ofinjury (Galvin and lones, 2002; Okano et aI., 2002;Storch and Schwarz, 2002), and (iii) gene therapy,e.g., expression of anti-appoptotic proteins, toprevent neuronal cell death at the site of CNS injury(Deigner et aI., 2000; Hatmann et aI., 2002; Ruth andMelamed, 2002). Clinical attempts to treat peripheralneurodegenerative diseases (neuropathies) usingrecombinant neurotrophins have proved promisingso far, raising the hope that a similar strategy couldbe applied to the treatment of CNS neurodegenera-tive disorders (Haase et aI., 1998; Simper and

Walsh, 1998; Ochs et aI., 2000). A major problemencountered in all attempts to delivery therapeuticmolecules to the CNS has been their inability topenetrate the blood-brain barrier (BBB). Hence,in order to develop effective therapeutic modalitiesbased on neurotrophins, one must developappropriate pharmacokinetic methods and formula-tions for their delivery to the CNS (Shoichet andWinn, 2000).

A challenging therapeutic modality for treatingneurodegenerative diseases is exploitation of theability of immune cells to secrete neurotrophinsduring infiamatory processes at the site of CNSinjury. The immune system is considered a peripheralsystem that provides protection against foreignmolecules. Its components can be divided into threemain categories: cells of the innate immune system,cells of the adaptive immune system and those of thelymphoid tissues (Roitt et aI., 2001). The term 'innateimmune system' refers to the immediately respondingsystem and comprises phagocytes (monocytesjmacro-phages and polymorphonuclear granulocytes) andnatural killer cells. The term 'adaptive immunesystem' refers to cells particularly invoved in thespecific immune response. This system containsantigen-presenting-cells (APCs) and Iymphocytessuch as T cells and B cells. The lymphoid systemconsists of Iymphocytes, accessory cells (macro-phages and APCs) and, in some tissues, epithelialcells. Lately, cumulative evidence points to a parallelstructure-function in the immune and nervous

systems. The basic unit is the neuron in the latter,and excitability is achieved by chemical or electricalconnections (synapses) between different neurons.The precise connective pattern throughout the CNS isobtained by the generation of synapses between twodifferent axonal projections. Several key rules aregenerally observed in synaptic assembly: (i) theneuronal endings participating in the synapse remaindistinct (i.e., there is no fusion of plasma membranesat the synapse site), (ii) several adhesion moleculeslocated pre and postsynaptically hold the twoneurons together, (iii) the synaptic connection (cleft)is stable due to the presence of extracellular matrix,and (iv) neurotransmitter is secreted from thepresynaptic site. Analogous to the neuronal synapseit was found that T cell-APC interactions fulfill these

criteria, suggesting an organization of immunological

synapses similar to that found in the CNS (Dustinand Colman, 2002). Under normal conditions, theCNS is considered an immune protected domain,mainly due to the BBB, which limits the access ofsystemic immune cells. However, in the diseasedstate, due to infection with exogenous antigens orautoimmunity, cells from the immune system are ableto penetrate the CNS, enabling the generation ofboth immune and neuronal synapses. This patholo-gical state may involve cross-talk between theneuronal and immunologic synapses. In addition toinfiltration of the immune cells from the periphery tothe injured CNS area, neuronal microglial cells canact as local potential APCs (Ulvestad et aI., 1994;Car/son et aI., 1998), suggesting that these and otherneuronal cells may initiate, regulate and sustain animmune response (Becher et aI., 2000). This cross-talkbetween the nervous and immune systems hasimplications not only for treatment, but also for theseverity of the disease. Indeed, in many diseasedstates, an inflammatory process develops in the CNS.Although there are studies implying that the immuneactivity in the CNS is pathological (Lotan andSchwartz, 1994) and is even considered an importantmediator of secondary damage (Car/son et aI., 1998;Mautes et aI., 2000), other investigations indicatesthat inflammation might beneficially affect the insultby promoting clearance of cell debris and secretion ofneurotrophic factors and cytokines. In this chapterwe will focus on the involvement of the neurotro-

phins as neuroprotective mediators in the cross-talk,during CNS injury, between the immune and nervoussystems.

Neurotrophinsin the immune system

Following the pioneer studies of Aloe and Levi-Montalcini (1977), Aloe (1988), Aloe and De Simone

(1989), Aloe and Micera (1999) demonstrating theproduction of neurotrophins and the activity of NGFon mast cells, the last decade has seen a variety ofdirect and indirect evidence pointing to the produc-tion, expression and/O!' secretion of neurotrophins bydifferent cell types of the immune system (Tables Iand 2). Macrophages, T cells and B cells of murineand human origin, which are most abundant at thesite of CNS injury, produce BDNF, NGF, NT-3 and

389

NT-4/5 (Table 1). Other immune cells, such as mastcells, monocytes, neurotrophiles, granulocytes, baso-philes, eosinophiles, splenocytes and peripheralblood mononuclear cells (PBMCs), as well as myeloidprogenitor cells secrete one or more neurotrophins(Table 2). All these cells express different full lengthand/or truncated trk receptor isoforms (Tables Iand 2). The lack of sequencing data on the differenttrk receptor forms present in the immune cells doesnot allow a clear conclusion to be drawn regardingtheir homology to the CNS trk receptors. However,treatment of the immune cells with recombinant

human neurotrophin or native mouse NGF demon-strated that the cells are able to respond toneurotrophin stimulation (Tables I and 2). The majorbiological effects observed are increased proliferation,survival, immunoglobulin production, antigen pre-sentation and migration, production of proinflam-matory and anti-inflammatory chemical mediators,secretion of serotonin and histamine and differentia-

tion (Tables I and 2). These findings strongly suggestthat the immune system can produce most if not allmembers of the neurotrophin family. It is alsoconceivable that different types of immune cells arethe targets of paracrine and/or autocrine activity, asthey express trk receptors. Furthermore, thesefindings strongly suggest that neurotrophins canmediate heterologous interactions between theimmune and nervous systems.

Dual role of the CNS cellular inflammatoryprocess: neurotoxicity and neuroprotection

When a nerve in the peripheral nervous system (PNS)or a myelinated track in the CNS is sectionedsurgically or damaged by mechanical trauma, theproximal segment often survives and degenerates. Inthe distal segment Wallerian degeneration takesplace, with both axons and myelin disappearingrapidly. The debris is phagocytized by neural cellsand macrophages. It has been established that theintegrity of the myelin sheath depends on continuouscontact with a viable axon. Any injury or diseasecausing general impairment of neuronal function willalso result in axonal degeneration and the onset ofmyelin breakdown after the initial neuronal damage.Myelin breakdown and neuronal damage affect the

IN'-D0

TABLE I

Production, expression and/or secretion of neurotrophins, trk receptors presence and their biological role in cel1s of the immune system detected in CNS injury

nf ~ no information found in the presented reference."The presence of trk receptors is based on mRNA and/or protein expression and/or neurotrophin autocrine activity.

Cel1 type Origin Trk receptor expression" Neurotrophin expression Biological effect References

or secretion

Macrophage Mouse TrkA, TrkB, TrkC BDNF, NT-3, NO secretion, TNF-a, Susaki et aI., 1996, MacMicking et aI., 1997;

peritoneum NGF, NT-3 IL-I production Brodie et aI., 1999; Barouch et aI., 2001a,bMouse alveolar Truncated TrkB, ful1 NT-3, NT-4/5, BDNF nf Hikawa et a\., 2002

interstitial cel1s length TrkCHuman TrkA, TrkB, TrkC NGF Differentiation, increase in Otten and Gadient, 1995;

oxidative burst Garcia-Suarez et aI., 1997;

Ricci et aI., 2002T-cell Mouse TrkA NGF nf Ehrhard et aI., 1993a

Rat TrkA" NGF, BDNF, NT-3 Enhanced Ab synthesis, Manning et aI., 1985; Thorpe and Perez-Polo,

proliferation 1987; Bm.ouch and Schwartz, 2002;

Muhal1ab et aI., 2002Human TrkA, no TrkB BDNF, NGF, Proliferation, production Otten and Gadient, 1995; Bracci-Laudiero et aI.,

NT-3, NT-4/5 of neuropeptide Y 1996; Lambiase et aI., 1997; Moalem et aI.,

2000; Stadelmann et aI., 2002Thl Mouse nf no BDNF nf Ehrhard et aI., 1993a, 1994;

Ziemssen et aI., 2002Th2 Mouse TrkB NGF nf Besser and Wank, 1999; Bonini et aI., 1999

Human TrkA, no TrkB BDNF, NGF nf Besser and Wank, 1999; Ziemssen et aI., 2002B cel1s Human TrkB NGF, BDNF, NT-3, NT-4 proliferation, Otten et aI., 1989; Otten and Gadient, 1995;

differentiation, Melamed et aI., 1995, 1996;

cytoskeletal effects Schenone et aI., 1996; Besser and Wank, 1999;

Barouch et aI., 2000

TABLE 2

Production, expression and/or secretion of neurotrophins, trk receptors presence and their biological roles in selected cell types of the immune system

nf =no information found in the presented reference."The presence of trk receptors is based on mRNA and/or protein expression and/or neurotrophin autocrine activity.

w'-D

Cell type Origin Trk receptor Neurotrophin Biological effect References

expression" expression or secretion

Mast cell Mouse nf NGF nf Xiang and Nilsson, 2000Rat peritoneum Truncated TrkA NGF, NT-3, NT-4 proliferation, upregulation Aloe and Levi-Montalcini, 1977;

of IL-3, survival, histamine Bruni et aI., 1982; Aloe, 1988;serotonin and NGF release, chemotaxis, Aloe and De Simone, 1989; Horigome

et aI., 1993, 1994;Leon et aI., 1994; Aloe and Micera,1999; Skaper et aI., 2001; Kawamotoet aI., 1995; Tarn et aI., 1997

Human, TrkA, trnncated NGF, BDNF, NT-3 expression of c-fos and Sawada et aI., 2000; Xiang andleukemic and full length TrkB NGF1-A; MAPK activation Nilsson, 2000

Lung TrkC nf

TrkA, TrkB, nfumbilical cord TrkC increased

TrkA, TrkC NGF, BDNF, NT-4 expression of chymaseMonocyte Human TrkA NGF increased respiratbry burst Kannan et aI., 1992; La Sala et aI.,

expression of Bel-2, Bel-XL 2000; Ehrhard et aI., 1993band Bfl-1

Neutrophil Mouse, nf NGF survival, proliferation, chemotaxis, Kannan et aI., 1991; Brodieperitoneum phagocytosis and Gelfand, 1992

Granulocyte Human Generally lacking NGF, BDNF, NT-4 interaction with other effector cells Laurenzi et aI., 1998Basophile Human TrkA NGF increased IL-13 secretion Bischoff and Dahinden, 1992;

Sin et aI., 2001Human TrkAb nf increased histamine release, Matsuda et aI., 1988b; Bischoff and

differentiation Dahinden, 1992Eosinophile Human TrkA, TrkB, TrkC NGF, NT-3 increased peroxidase and IL-4 Solomon et aI., 1998; Bonini et aI.,

secretion 1999; Kobayashi et aI., 2002;Noga et aI., 2002

Mixed splenocyte Mouse nf NGF, BDNF, nf Barouch et aI., 2000NT-3, NT-4

PBMC Human TrkB nf nf Besser and Wank, 1999Myeloid Human nf NGF colony growth and differentiation Matsuda et aI., 1988a

progenitor cell

392

neuron's electrical conductivity, resulting in itsirreversible functional deficits (Kalb, 1995; Haubenet aI., 2000). In addition, a destructive series ofinjury-induced secondary waves of events occur aphenomenon known as secondary degeneration.(Faden, 1993; Y oles and Schwartz, 1998; Haubenet aI., 2000). The progression of CNS damage may beattributable to the fact that acute neuronal degenera-tion creates a hostile environment for the neighboringneuronal fibers that are still intact, thereby propagat-ing the damage and eventually causing degenerationof the fibers (Kipnis et aI., 2000). This process of'secondary degeneration' is likely that seen aftertraumatic injury to other areas of the spinal cord andbrain.

Several studies have shown that immune systemintervention is necessary in order to minimize thedamage and to activate healing-neuroregenerativeprocesses (Bethea and Dietrich, 2002). The inflam-matory process is one of the main host defensemechanisms that also fight/protect against CNSinjury or infection. This is a biological phenomenonby which lymphocytes and soluble chemical media-tors regulate tissue homeostasis (Bethea and Dietrich,2002) and, by so doing, become an important 'housekeeping-repair mechanism' in the peripheral tissues.The possibility that the CNS is dependent on asimilar inflammatory activity for maintenance andrepair, especially after injury, is an important issuethat awaits investigation. Using the rat optic nerve orthe spinal cord injury model, studies have yieldedinsights into novel mechanisms involved in degenera-tion and have led to the identification of specificmolecules with neuroprotective properties (Schwartzet aI., 1999). In the injured CNS, as in other injuredtissues, activated T cells, B cells and macrophages arerequired for neuronal regeneration (Schwartz et aI.,1999; Schwartz and Moalem, 2001).

Macrophage invasion of the damaged CNS isslower and less extensive than that observed after

PNS injury (Schwartz et aI., 1999). The macrophagesinfiltrating the damaged CNS are less efficient inclearing the myelin debris that inhibits the neuronalgrowth required for axonal regeneration and for therecovery of neuronal excitability (Schwartz et aI.,1999). Also the microglia, CNS resident 'macro-phages', are activated after injury, although theiractivity is not as high as that of the peripheral-blood

macrophages infiltrating the CNS (Schwartz et aI.,1999). Regrowth of the CNS axons was correlatedwith the rapid clearance of myelin from the treatedaxons and the abundant distribution of PNS-acti-

vated macrophages along the distal part of thedamaged axons. The macrophages probably act insynchrony with other regenerating molecules redu-cing the toxic effect of the extracellular environmentof the injured axons. Recruitment of macrophages tothe injured site is an important physiological responseand one of the earliest phases of host defense andof the wound healing process (Popovich and lones,2003). The generation and accumulation of CNSdebris at the site of injury initiates the chemotaxis ofneutrophils and macrophages from the periphery,resulting in phagocytosis and the release of neurotox-ins at the CNS inflammatory site. Myelin debrisand cytokines, eicosanoids, oxidative radicals andnitric oxide (Minagar et aI., 2002), may be toxic toneurons and glia (Popovich and lones, 2003), andmight trigger the release of inflammatory mediatorsthat can cause necrotic cell death and demyelination(Popovich and lones, 2003).

The macrophages are not the only participants inthe inflammatory process after CNS injury. Thesecond arm of the immune system are the T cellswhich, in contrast to the macrophages, respond tospecific antigens and 'remember' past experiences.When activated, the T cells can kill their target cellsor produce signal molecules that activate or suppressthe growth, movement or differentiation of other cells(Schwartz et aI., 1999). The BBB, normally inacces-sible to resting T cells, is permeable to activatedT cells. The latter, however, do not accumulate in thehealthy CNS unless they recognize and are ableto react with their specific antigens (Schwartz et aI.,1999). Some studies have revealed a greateraccumulation of endogenous T cells in the injuredPNS than in the damaged CNS (Moalem et aI.,1999a). Moreover, it has been shown that elimination(by apoptosis) of T-cells from the CNS, is moreeffective than in the PNS, and that this phenomenonis virtually absent from tissues such as muscle andskin, again emphasizing the concept of the brain'simmune privilege (Gold et aI., 1997). The T cellresponse in the injured CNS is restricted and strictlyregulated (Schwartz et aI., 1999); the implication ofthis finding for regeneration of the injured CNS is not

clear. It was shown that T cells specific for a myelindeterminant can protect CNS neurons from second-ary degeneration (Moalem et aI., 1999b). Using the invivo experimental model of partial lesion of the ratoptic nerve, it was found that axonal injury wasfollowed by the transient accumulation of endogen-ous T cells at the site of the lesion (Schwartz et aI.,1999). Administration of in vitro activated syngeneicT cells specific for CNS self-antigen myelin basicprotein (MBP) resulted in the local accumulation ofTcells at the injured site in the optic nerve in vivo(Moalem et aI., 1999b). Rats inoculated with the anti-MBP T cells (TMBP) showed significantly lesssecondary degeneration of retinal neurons thananimals that received phosphate-buffered saline orT cells specific for the foreign antigen ovalbumin(TOVA)(Moalem et aI., 1999b). The neuroprotectiveeffect on retinal neurons was distinct despite thefinding that the transferred anti- TMBPcells induced atransient, monophasic paralytic disease known asexperimental auto immune encephalomyelitis(EAE)-a known animal model of multiple sclerosis(MS) (Schwartz et aI., 1999).

The induction and activation of T lymphocytesrequire at least two signals from specific APCs. Thefirst signal results from the binding of the T cellreceptor to its antigen in the context of the majorhistocompatibility complex (MHC). The second isprovided by various molecules expressed by the APCs(Friendl and Gunzer, 2001). There are at least twotypes of APCs that activate T cells: macrophages andB lymphocytes. The neuroprotective effect of TMBPcells in the CNS might be the result of local cross-talkwith APCs and/or a local effect on the CNS injury.In several CNS injury models it was found that therecovery of motor neurons after facial nervetransection in SCID mice, which lack functional Tand B cells, is inferior to that in normal mice (Serpeet aI., 1999, 2000; Barouch and Schwartz, 2002).Using the rat optic nerve injury model it was shownthat administration of T MBPinduces a local immuneresponse, which is manifested by an increase in thenumber of microglia/macrophages and B cells(Barouch and Schwartz, 2002).

The number of B cells at the site of injured opticnerve is very low. However, upon injection of T MBPthey accumulated at this site in far higher numberthan in the control rats, as evaluated 4 and 7 days

393

after injury. No B cells were detected in the uninjuredoptic nerves of rats inoculated with T MBP' Theaccumulation of B cells upon injection of T MBPwastransient and had decreased by 7 days after injury. Itwas also found that in addition to the augmentednumber of B cells, there was a transient increase inthe expression of neurotrophins, particularly NGF.In this experimental model of injury the T cells didnot express NGF but did express BDNF and NT-3;the B cells expressed NGF, BDNF and NT-3 and themacrophages-NGF and NT-3, but not BDNF.According to these findings, the type and quantityof neutrophins are dependent on the interplaybetween T cells, B cells and macrophages at the siteof injury. It is conceivable that the acumulation of Tcells at the injured site is a regulatory step in therecruitment of B cells and macrophages to the site ofinjury. It is also plausible that T cells secrete certaincytokines that mediate such accumulation, as well asthe neuroprotective effect of TMBP' The substantialexpression of neurotrophins by B cells might haveimportant implications for the survival of injuredneurons. Recent studies demonstrated that also Bcells can have a destructive effect after CNS insult

(Schori et aI., 2002). Therefore, it is possible that inCNS injury the beneficial B cell effect depends onproper T cell regulation and occurs in response to theaccumulation of T cells at the site of CNS injury(Barouch and Schwartz, 2002).

Immune cell-derived neurotrophins as potentialmediators of neuroprotection

As mentioned above, injection of activated T MBPcells, but not T OVAcells, reduces secondary neuronaldegeneration after primary injury of CNS axons(Moalem et aI., 1999b). Although, the mechanismunderlying this auto immune neuroprotection is notknown, there are several findings suggesting thatneurotrophins may serve as mediators between theimmune and nervous systems. We will focus on thecontribution of the neurotrophins to the autoimmuneneuroprotective effect.

Based on the expression of different cytokinemRNAs, T cells accumulating at the site of the CNSinjury may be classified as having a Thl, Th2 or Th3phenotype (Moalem et aI., 2000). Also, coexpression

394

TABLE 3

Differential production and secretion of neurotrophins by immune cells at the site of CNS injury

Injury model

Neurotrophins secreted by immune system cells at the site of inflamation

ReferencesImmune cell type Produced anddetected at CNS site secretedof injury neurotrophins

Optic nerve Moalem et aI., 2000; Barouch and Schwartz, 2002

Multiple sclerosis

macrophagesThIT cells: Th2Th3B cells

microgliaTh1T cells:Th2

microgliaT cellsNK cellsT cells

microglia

Nigrostriatal dopaminergic injuryEAE

Sustained cerebral embolism

H1V-I encephalitis

NGF, BDNF, NBNT3

BDNF, NT4/5BDNFNGF, BDNF, NT3BDNFBDNF

Stadelmann et aI., 2002; Ziemssen et aI., 2002

BDNFGDNF, BDNFBDNF, NT3BDNF, NBNGFBDNF, NB

Batchelor et aI., 1999, 2002

Hammarberg et aI., 2000

Mizuma et aI., 1999

Soontornniyomkij et aI., 1998

of specific cytokines: INFy (Thl), IL-6 and IL-IO(Th2) and TGFf3 (Th3) and of different neurotro-phins enables the identification of neurotrophinsaccording to T cell subtype. By evaluating mRNAlevels, it was found that Thl cells express NT ~3, Th2cells express NT-4/5 and Th3 cells express BDNF(Table 3). Neurotrophin detection by immunocyto-chemical techniques shows not only the presence ofmRNA but also the production of neurotrophinproteins by the respective T cell populations. Usingantibodies selective for individual neurotrophins andT cell receptor, flow cytometry analysis of blood-derived T cells or in vitro grown and activated T cells,confirmed the selective expression of BDNF, NGF,NT -3 by T cells. The use of ELISA has demonstratedthat the secretion of neurotrophins by T cells issignificantly augmented upon their exposure in vitroto MBP. Cumulatively, these findings suggest theexpression, production and secretion of selectiveneurotrophins by T cells in the optic nerve injurymodel (Moalem et aI., 2000).

Contributionof Trk receptors to theneuroprotectiveeffect in CNS injury

Neuroprotection of the injured neuron at the site ofinflammation can be afforded in two ways: (i) via an

indirect effect through the autocrine activity ofsecreted neurotrophins on the respective Trk recep-tors in different T cell subpopulations and in otherimmune cells (e.g., B cells and macrophages); thepossible physiological significance would be increasedproliferation of the T lymphocytes, resulting in theproduction of an appropriate amount of neurotro-phins; these would bind to the neuronal Trk receptor,leading to a significant neuroprotective effect at theinjured target site; it is also conceivable that Tlymphocytes, autocrine1y activated by neurotrophins,produce additional mediators such as IL-6, which canaffect neurons expressing IL-6 receptors (Carlsonet aI., 1999), and (ii) via a direct effect of secretedneurotrophins on the Trk receptors in the injuredoptic nerve. Indeed, expression of mRNA for TrkAand TrkB receptors, but not for TrkC receptors, wasdetected in the optic nerve, suggesting the presence ofthe respective receptors (Moalem et aI., 2000).Therefore, it is very tempting to assume that thesecreted neurotrophins activate the appropriate Trkreceptors on the injured optic nerve and inducesurvival signal transduction pathways such as Ras/ERK and/O!- AKT/PI3 kinase. The final resultswould be the reduced cell death of retinal ganglionneurons, i.e., neuroprotection.

To substantiate the concept of Trk-inducedneuroprotection, a pharmacological tool is

available: K-252a, a relatively selective inhibitor ofTrk family receptors, with a complex dual effect,which may be considered a mixed agonist/antagonist(Rasouly et aI., 1992). In the presence of neurotro-phins this compound behaves as an antagonist andblocks receptor activation by the neurotrophins. Intheir absence, K-252a acts as an agonist andstimulates Trk receptors (Knusel and Hefti, 1992).When applied locally as eye drops to animals withoptic nerve injury, it inhibited Trk receptor activity.This, in turn, reduced the neuroprotective effect ofthe T MBPcells (Moalem et aI., 2000). In the same ratmodel, however, the time course and the severity ofthe EAE symptoms were not changed. The ineffec-tiveness may be explained by the low concentrationof K-252a that entered into the systemic bloodcirculation, as opposed to its high local concentrationin the eye. In SJL/J mice fed an oral formulation ofK-252a, the appearance of EAE symptoms wasdelayed (neuroprotective effect), a phenomenoncorrelating with immunosuppression of the periph-eral macrophages (Lazarovici et aI., 1996). It istempting to speculate that high systemic levels of K-252a inhibited the Trk receptors of the peripheralmacrophages, resulting in less infiltration into theCNS, i.e., less destructive inflammation, as measuredby a reduction in the severity of neurologicalsymptoms (mean neurological score) of EAE modelanimals. It is also conceivable that K-252a inhibited

the activated Trk receptors in the CNS neurons at thesite of inflammation. Since overexpression of trkreceptors or extensive trk receptor activity is onco-genic and/or neurotoxic (Nakagawara, 2001), suchinhibition may impart neuroprotection. Assuming ahigh level of neurotrophins at the site of CNS injuryresulting in hyperactivation, it is possible that underthese circumstances the activity of K252a willpredominate, promoting neuroprotection.

Multiple sclerosis as a potential target forneuroprotective autoimmunity

For many years it was considered that when theimmune cells attack the nervous system, a neuroim-munological disease develops. Multiple sclerosis(MS) and its animal models provide the paradigmfor such a deleterious interaction between cells of the

395

immune and nervous systems (Hohlfeld et aI., 2000).Multiple sclerosis, a demyelinating disorder of theCNS of unknown etiology, it is thought to be anauto immune disease mediated by T cells specific forthe myelin antigen. MBP has been studied as apotential autoantigen in the disease because of its roleas an encephalitogen in experimental auto immuneencephalomyelitis and postviral encephalomyelitis,and due to the presence in the blood of multiplesclerosis patients of in vivo activated T cells reactiveto MBP (Ota et aI., 1990). Immunological, clinicaland pathological studies suggest that T lymphocytesdirected against myelin antigens are involved in thepathogenesis of MS. Research on an experimentalanimal model of MS, EAE, demonstrated thatMBP or proteolipid protein-(PLP)-specific T cellsmediate the destruction of CNS myelin. Inrecent years, studies on EAE showed that encepha-litogenic T cells recognize short MBP pep tides(Martin, 1997). These findings suggest that injury-induced activation of myelin-reactive T cells may havephysiological or pathological sequelae. Is it possiblethat the same inflammatory process is at the sametime both neurodegenerative and neuroprotectivein MS?

Recent studies strongly suggest that BDNF maybe the key element for understanding the complexinteractions between the immune and nervous systemin MS patients. Neurons are considered to be one ofthe cellular sources of BDNF in the CNS

(Stadelmann et aI., 2002). Immune cells are capableof producing BDNF (Tables 1 and 2), which binds toTrkB receptors. Both are present in the CNS and inthe immune cells. Immune cell-derived BDNF isdetectable in human immune-mediated CNS inflam-

mation (Stadelmann et aI., 2002; Ziemssen et aI.,2002, Table 3). Recent in vivo studies show that ahigher percentage of inflammatory cells in activelydemyelinating areas of MS lesions contain BDNFthan in areas without ongoing myelin breakdown. Inaddition, a high percentage of BDNF + macrophagesand lymphocytes are present in perivascular infil-trates (Stadelmann et aI., 2002). These observationsindicate that relatively high numbers of BDNF-containing immune cells are found in the early stagesof CNS lesion development. BDNF also can beretrogradly transported from neurons outside thelesion area (Canals et aI., 2001; Stadelmann et aI.,

396

2002), such transport being upregulated afteraxotomy (Tonra et aI., 1998; Stadelmann et aI.,2002). It is possible that autocrine as well as paracrineinteractions, through BDNF binding to the TrkBreceptor, further enhance endogenous neurotrophicsupport in MS lesions. The neuroprotective capacityof immune cells is not restricted to BDNF. As

mentioned above, a panel of other potentiallyneuroprotective and neurotrophic mediators isreleased at the site of the injured CNS. The complexneuronal-immune-neurotrophin network likely worksin concert to preserve the axons in a microenviromentcapable of producing significant neurotoxicity(Stadelmann et aI., 2002). This concept may beextended to other models of CNS injury: nigrostriataldopaminergic (Parkinson's), EAE (MS), cerebralembolism (stroke), and HIV encephalitis (CNSinflammation) (Table 3). In all these models ofinjury, like in the optic nerve and multiple sclerosismodels, different members of the neurotrophin familyare secreted by the immune cells at the site ofinflammation, substantiating the notion that neuro-protective auto immunity is provided by neurotro-phins released at the site of inflammation.

Conclusions

CNS injury is an inflammatory process involvingheterologous interactions between immune cells andneural cells, brokered by chemical mediators inclu-ding cytokines and neurotrophins. For a betterunderstanding of this process we need to identify allthe molecules secreted at the site of CNS injury. It isclear that some of these molecules exert a neurotoxic

effect while others yield neuroprotection. It is evidentthat immune cells produce and secrete neurotrophinssuch as NGF, BDNF, NT-3 and NT-4/5. Thisprocess also occurs at the site of CNS injury where itmay even be amplified. Since both immune cells andneural cells express the neurotrophin trk receptors,it is reasonable to assume that the neurotrophinsaffect the immune and the neural cells both

autocrinically and paracrinically. The end result willbe neurotoxicity and/or neuroprotection, dependingon the final balance between the released neurotoxic

molecules and neuroprotective chemicals (neurotro-phins). CNS inflammatory responses occurring after

injury contribute to both neuroconstructive andneurodestructive processes in the nervous system(Bethea and Dietrich, 2002). The key to successfulneuronal regeneration is a balance between up-regulation of the repair potential while maintainingand/or reducing the destructive potential (Popovichand lones, 2003).

The cellular therapeutic implications of thefindings presented in this review may lead to novelmanipulations of autoimmunite processes as aneuroprotective strategy for treatment of CNS injury(Schwartz and Kipnis, 2001). One approach would beto neutralize the neurotoxic molecules with anti-

bodies. Another approach to be considered isinhibiting the gene expression of neurotoxic cytokinesof infiltrating immune cells. Transcription factors ofBand T cells, and macrophages, which regulatethe expression of neurotoxic molecules, may beimportant targets for novel immunosupressivedrugs. A cellular therapeutic approach could be theadoptive transfer of activated autoimmune cellsspecific for a CNS antigen. To this end, immunecells from the blood of CNS injured patients,would be isolated, propagated in vitro to secreteneurotrophins and reinjected. Presumably, theinjected cells would reach the site of CNSinjury where they could exert a neuroprotectiveeffect as recently reported in the treatment ofacute spinal cord injury with macrophages (http://www.proneuron.com).

Cell therapy approaches can not yet totally replacethe use of low molecular weight drugs in thetreatment of CNS injury. Therefore, novel drugs,such as Trk receptor agonists, need to be developed.The long-term pharmaceutical goal must be thedevelopment of agonist drugs with high selectivitytowards individual Trks and the ability to distinguishbetween trk receptor subtypes belonging to immunecells and neuronal cells. The recent synthesis ofCEPs (Roux et aI., 2002), pure trk receptorantagonists, is an encouraging pharmacologicaldevelopment which may contribute to research onTrk receptor agonists. Drugs modulating the activityof Trk receptor activity may contribute to investiga-tions into the neuroprotective role of Trk receptorsin CNS models of injury and cross-talk with theimmune system and might prove useful in theclinical setting.

Abbreviations

NGFBDNFNT-3NT-4/5TrkPNSCNSBBBAPCMBPEAE

TMBP

TOVAMHCELlSAIL-6mRNA

p75NTRK252aMS

References

nerve growth factorbrain-derived neurotrophic factorneurotrophin 3neurotrophin 4/5tropomyosin-related tyrosine kinaseperipheral nervous systemcentral nervous systemblood-brain barrier

antigen presenting cellmyelin basic proteinexperimental autoimmune encephalomy-elitis

T cells activated by MBPT cells activated by ovalbuminmajor histocompatibility complexenzyme-linked immunosorbent assayinterleukin 6

messenger RNAlow-affinity neurotrophin receptora selective inhibitor of the Trk receptormultiple sclerosis.

Aloe, L. and Levi-Montalcini, R. (1977) Mast cells increase intissues of neonatal rats injected with the nerve growth factor.Brain Res., 133: 358-366.

Aloe, L. (1988) The effect of nerve growth factor and itsantibody on mast cell in vivo. J. Neuroimmunol., 18: 1-12.

Aloe, L. and De Simone, R. (1989) NGF primed spleen cellsinjected in brain of developing rats differentiate into mastcells. Int. J. Devl. Neurosci., 7: 565-573.

Aloe, L. and Micera, A. (1999) Nerve growth factor: basicfindings and clinical trials. Biomed. Rev., 10: 3-14.

Apfel, S.C (1997) Clinical Applications of NeurotrophicFactors. Lippincott-Raven, Philadelphia, pp. 1-209.

Barbacid, M. (1995) Neurotrophic factors and their receptors.Curr. Opin. Cell BioI., 7: 148-155.

Barouch, R., Appel, E., Kazimirsky, G., Braun, A., Renz, H.and Brodie, C (2000) Differential regulation of neurotrophinexpression by mitogens and neurotransmitters in mouselymphocytes. J. Neuroimmunol., 103: 112-121.

Barouch, R., Appel, E., Kazimirsky, G. and Brodie, C. (200Ja)Macrophages express neutrophins and neutrophinreceptors. Regulation of nitric oxide production by NT-3.J. Neuroimmunol., J12: 72-77.

397

Barouch, R., Kazimirsky, G., Appel, E. and Brodie, C (2001b)Nerve growth factor regulates TNF-a production in mousemacrophages via MAP kinase activation. J. Leukoc. BioI.,69: 1019-1026.

Barouch, R. and Schwartz, M. (2002) Autoreactive T cells

induce neurotrophin production by immune and neural cellsin injured rat optic nerve: implications for protectiveautoimmunity. FASEB J., 16: 1304-1306.

Batchelor, P.E., Liberatore, G.T., Wong, J.Y., Porritt, M.J.,Frerichs, F., Donnan, G.A. and Howells, D.W. (1999)Activated macro phages and microglia induce dopaminergicsprouting in the injured striatum and express brain-derivedneurotrophic factor and glial cell line-derived neurotrophicfactor. J. Neurosci., 19: 1708-1716.

Batchelor, P.E., Porritt, M.J., Martinello, P., Parish, CL.,Liberatore, G.T., Donnan, G.A. and Howells, D.W. (2002)

Macrophages and microglia produce local trophic gradientsthat stimulate axonal sprouting toward but not beyond thewound edge. Mol. Cell Neurosci., 21: 436--453.

Becher, B., Prat, A. and Antel, J.P. (2000) Brain-immuneconnection: immuno-regulatory properties of CNS-residentcells. Glia, 29: 293-304.

Besser, M. and Wank, R. (1999) Cutting edge: clonallyrestricted production of the neurotrophins brain-derivedneurotrophic factor and neurotrophin-3 mRNA by humanimmune cells and ThljTh2-polarized expression of theirreceptors. J. Immunol., 162: 6303-6306.

Bethea, J.R. and Dietrich, W.D. (2002) Targeting the hostinflammatory response in traumatic spinal cord injury. Curr.Opin. Neurol., 15: 355-360.

Bischoff, S. and Dahinden, C. (1992) Effect of nerve growthfactor on the release of inflammatory mediators by maturehuman basophils. Blood, 79: 2662-2669.

Bonini, S., Lambiase, A., Bonini, S., Levi-Schaffer, F. and

Aloe, L. (1999) Nerve growth factor: an important moleculein allergic inflammation and tissue remodelling. Int. Arch.Allergy Immunol., 118: 159-162.

Bracci-Laudiero, L., Aloe, L., Stenfors, C., Tirassa, P.,Theodorsson, E. and Lundberg, T. (1996) Nerve growthfactor stimulates production of neuropeptide Y in humanIymphocytes. Neuroreport, 7: 485-488.

Brodie, C. and Gelfand, E.W. (1992) Functional nerve growthfactor receptors on human B lymphocytes: interaction withlL-2. J. Immunol., 148: 3492-3497.

Brodie, C, Barouch, R., Appel, E. and Kazimirsky, G. (1999)Expression of neutrophins and neutrophin receptors on macro-phages. Induction of nitric oxide and activation of signaltransduction pathways. Neurosci. Lett. Suppl., 54: SlO-SII.

Bruni, A., Bigon, E., Boarato, E., Mietto, L., Leon, A. andToffano, G. (1982) Interaction between nerve growth factorand Iysophosphatidylserine on rat peritoneal mast cells.FEBS Lett., 138: 190-193.

Canals, J.M., Checa, N., Marco, S., Akerud, P., Michels, A.,Perez-Navarro, E., Tolosa, E., Arenas, E. and Alberch, J.

398

(2001) Expression of brain-derived neurotrophic factor incortical neurons is regulated by striatal target area.1. Neurosci., 21: 117-124.

Carlson, S.L., Parrish, M.E., Springer, 1.E., Doty, K. andDossett, L. (1998) Acute inflammatory response in spinalcord following impact injury. Exp. Neurol., 151: 77-88.

Carlson, N.G., Wieggel, W.A., Chen, 1., Bacchi, A.,Rogers, S.W. and Gahring, L.e. (1999) Inflammatorycytokines IL-I alpha, IL-I beta, IL-6, and TNF-alpha impartneuroprotection to an excitotoxin through distinct pathways.1. Immunol., 163: 3963-3968.

Chao, M., Casaccia-Bonnefil, P., Carter, B., Chittka, A.,Kong, H. and Yoon, S.O. (1998) Neurotrophin receptors:mediators of life and death. Brain Res. Rev., 26: 295-301.

Chao, M.V. and Bothwell, M. (2002) Neurotrophins: to cleaveor not to cleave. Neuron, 33: 9-12.

Deigner, H.P., Haberkorn, U. and Kinscherf, R. (2000)Apoptosis modulators in the therapy of neurodegenerativediseases. Expert Opin. Investig. Drugs, 9: 747-764.

Dustin, M.L. and Colman, D.R. (2002) Neural and immuno-

logical synaptic relations. Science, 298: 785-789. .Ehrhard, P.B., Erb, P., Graumann, U. and Otten, U. (1993a)

Expression of nerve growth factor and nerve growth factor

receptor tyrosine kinase Trk in activated CD4-positive T-cellclones. Proc. Natl. Acad. Sci. USA, 90: 10984-10988.

Ehrhard, P.B., Ganter, U., Bauer, 1. and Otten, D. (1993b)Expression of functional trk protooncogene in humanmonocytes. Proc. Natl. Acad. Sci. USA, 90: 5423-5427.

Ehrhard, P.B., Erb, P., Graumann, U., Schmutz, B. and Otten, U.(1994) Expression of functional trk tyrosine kinase receptorsafter T cell activation. 1. Immunol., 152: 2705-2709.

Faden, A.I. (1993) Experimental neurobiology of centralnervous system trauma. Cril. Rev. Neurobiol., 7: 175-186.

Freidman, W.l. and Greene, L.A. (1999) Neurotrophinsignaling via Trks and p75. Exp. Cell Res., 253: 131-142.

Friendl, P. and Gunzer, M. (2001) Interaction of T cells withAPCs: the serial encounter model. Trends Immunol., 22:187-191.

Galvin, K.A. and lones, D.G. (2002) Adult human neural stemcells for cell-replacement therapies in the central nervoussystem. Med. 1. Ausl., 177: 316-318.

Garcia-Suarez, 0., Hannestad, 1., Esteban, I., Martinez del

Valle, M., Naves, F.l. and Vega, 1.A. (1997) Neurotrophinreceptor-like protein immunoreactivity in human lymphnodes. Anal. Rec., 249: 226-232.

Gold, R., Hartung, H.P. and Lassmann, H. (1997) T-cellapoptosis in autoimmune diseases: termination of inflamma-tion in the nervous system and other sites with specializedimmune-defense mechanisms. Trends Neurosci., 20: 399-404.

Haase, G., Pettmann, B., Vigne, E., Castelnau-Ptakhine, L.,Schmalbruch, H. and Kahn, A. (1998) Adenovirus-mediatedtransfer of the neurotrophin-3 gene into skeletal muscle ofpmn mice: therapeutic effects and mechanisms of action.1. Neurol. Sci., I: 97-105.

Hatmann, A., Mouatt-Prigent, A., Vila, M., Abbas, N.,Perier, e., Faucheux, B.A., Vyas, S. and Hirsch, E.e.(2002) Increased expression and redistribution of the anti-apoptotic molecule Bcl-xL in Parkinson's disease. Neurobiol.Dis., 10: 28-32.

Hammarberg, H., Lidman, 0., Lundberg, e., Eltayeb, S.Y.,Gielen, A.W., Muhallab, S., Svenningsson, A., Linda, H.,van der Meide, P.H., Cullheim, S., OIsson, T. and Piehl, F.(2000) Neuroprotection by encephalomyelitis: rescue ofmechanically injured neurons and neurotrophin productionby CNS-Infiltrating T and natural killer cells. 1. Neurosci.,20: 5283-5291.

Hauben, E., Butovsky, 0., Nevo, D., Yo1es, E., Moalem, G.,Agranov, E., Mor, F., Leibowitz-Amit, R., Pevsner, E.,Akselrod, S., Neeman, M., Cohen, I.R. and Schwartz, M.(2000) Passive or active immunization with myelin basicprotein promotes recovery from spinal cord contusion.1. Neurosci., 20: 6421-6430.

Hikawa, S., Kobayashi, H., Hikawa, N., Kusakabe, T.,Hiruma, H., Takenaka, T., Tomita, T. and Kawakami, T.(2002) Expression of neutrophins and their receptors inperipheral lung cells of mice. Histochem. Cell BioI., 118:51-58.

Hohlfeld, R., Kerschensteiner, M., Stadelmann, e.,Lassmann, H. and Wekerle, H. (2000) The neuroprotectiveeffect of inflammation: implications for the therapy ofmultiple sclerosis. 1. Neuroimmunol., 107: 161-166.

Horigome, K., Pryor, 1.e., Bullock, E.D. and 10hnson, E.M.(1993) Mediator release from mast cells by nerve growthfactor: neurotrophin specificity and receptor mediation.1. BioI. Chem., 268: 14881-14887.

Horigome, K., Bullock, E.D. and 10hnson, E.M. (1994) Effectof nerve growth factor on rat peritoneal mast cells: survivalpromotion, and immediate-early gene induction. 1. BioI.Chem., 269: 2695-2702.

Kalb, L.Y. (1995) Recovery from spinal cord injury: newapproaches. Neuroscientist, I: 321-327.

Kannan, Y., Dshio, H., Koyama, H., Okada, M., Oikawa, Y.T.,Kaneko, M. and Matsuda, H. (1991) Nerve growth factorenhances survival, phagocytosis and superoxide productionof murine neutrophils. Blood, 77: 1320-1325.

Kannan, Y., Usami, K., Okada, M., Shimizu, S. and

Matsuda, H. (1992) Nerve growth factor suppresses apopto-sis of murine neutrophils. Biochem. Biophys. Res. Commun.,186: 1050-1056.

Kaplan, D.R. and Stephens, R.M. (1994) Neurotrophin signaltransduction by the Trk receptor. 1. Neurobiol., 25:1404-1417.

Kawamoto, K., Okada, T., Kannan, Y., Ushio, H.,Matsumoto, M. and Matsuda, H. (1995) Nerve growthfactor prevents apoptosis of rat peritoneal mast cells throughthe trk proto-oncogene receptor. Blood, 86: 4638-4644.

Kipnis, 1., Yo1es, E., Porat, Z., Cohen, A., Mor, F., Sela, M.,Cohen, I.R. and Schwartz, M. (2000) T cell immunity tocopolymer I confers neuroprotection on the damaged optic

~

nerve: possible therapy for optic neuropathies. Proc. Natl.Acad. Sci. USA, 97: 7446-7451.

Knusel, B. and Hefti, F. (1992) K-252 compounds: modulatorsof neurotrophin signal transduction. J. Neurochem., 59:1987-1996.

Kobayashi, H., Gleich, G.J., Butterfield, J.H. and Kita, H. (2002)Human eosinophils produce neurotrophins and secrete nervegrowth factor on immunologic stimuli. Blood, 99: 2214-2220.

La Sala, A., Corinti, S., Federici, M., Saragovi, H.U. andGirolomoni, G. (2000) Ligand activation of nerve growthfactor receptor trkA protects monocytes from apoptosis.J. Leukoc. BioI., 68: 104-110.

Lambiase, A., Bracci-Laudiero, L., Bonini, S., Bonini, S.,

Starace, G., D'Elios, M., De Carli, M. and Aloe, L. (1997)Human CD4 + T cell clones produce and release nervegrowth factor and express high-affinity nerve growth factorreceptors. J. Allergy. Clin. Immunoly., lOO:408-414.

Laurenzi, M.A., Beccari, T., Stenke, L., Sjolinder, M.,Stinchi, S. and Lindgren, J.A. (1998) Expression ofmRNA encoding neutrophins and neutrophin receptors inhuman granulocytes and bone marrow cells-enhancedneurotrophin-4 expression induced by LTB4. J. Leukoc.BioI., 64: 228-234.

Lazarovici, P., Rasouly, D., Friedman, L., Tabakman, R.,Ovadia, H. and Matsuda, Y. (1996) KZ5zaand staurosporine

microbial alkaloid toxins as prototype of neurotrophic drugs.In: B.R. Singh and A.T. Tu (Eds.), Natural Toxins H.Plenum Press, New York, pp. 367-377.

Lazarovici, P., Matsuda, Y., Kaplan, D. and Guroff, G. (1997)The protein kinase inhibitors K-252a and staurosporine asmodifiers of neurotrophin receptor signal transduction. In:Y. Gutman and P. Lazarovici (Eds.), Cellular and MolecularMechanisms of Toxin Action: Toxins and Signal Transduction,

V01. I, Harwood Academic Publishers, Amsterdam, pp. 69-94.Lee, R., Kermani, P., Teng, K.K. and Hempstead, B.L. (2001)

Regulation of cell survival by secreted proneurotrophins.Science, 294: 1945-1948.

Leon, A., Buriani, A., Dal Toso, R., Fabris, M., Romanello, S.,Aloe, L. and Levi-Montalcini, R. (1994) Mast cells synthe-size, store, and release nerve growth factor. Proc. N atl. Acad.Sci. USA, 91: 3739-3743.

Lotan, M. and Schwartz, M. (1994) Cross talk between the

immune system and the nerve system in response to injury:implications for regeneration. FASEB J., 8: 1026-1033.

MacMicking, J., Xie, Q.W. and Nathan, C (1997) Nitric oxideand macrophage function. Annu. Rev. Immunol., 15:323-350.

Manning, PT., Russell, J.H., Simmons, B. and Johnson, E.M.(1985) Protection from guanethidine-induced neuronaldestruction by nerve growth factor: effect of NGF onimmune function. Brain Res., 340: 61-69.

Martin, R. (1997) Immunological aspects of experimentalallergic encephalomyelitis and multiple sclerosis and their

399

application for new therapeutic strategies. J. Neural. Transm.Suppl., 49: 53-67.

Matsuda, H., Coughlin, M.D., Bienenstock, J. andDenburg, J.A. (1988a) Nerve growth factor promoteshuman hemopoietic colony growth and differentiation.Proc. Natl. Acad. Sci. USA, 85: 6508-6512.

Matsuda, H., Switzer, J., Coughlin, M.D., Bienenstock, J. andDenburg, J.A. (l988b) Human basophilic cell differentiationpromoted by 2.5S nerve growth factor. Int. Arch. Allergy.Appl. Immunol., 86: 453-457.

Mautes, A.E., Weinzierl, M.R., Donovan, F. and Noble, L.J.(2000) Vascular events after spinal cord injury: contributionto secondary pathogenesis. Phys. Ther., 80: 673-687.

Melamed, 1., Turner, CE., Aktories, K., Kaplan, D.R. andGelfand, E.W. (1995) Nerve growth factor trigges micro fila-ment assembly and paxillin phosphorylation in human BIymphocytes. J. Exp. Med., 181: 1071-1079.

Melamed, 1., Kelleher, C.A., Franklin, R.A., Brodie, C,Hempstead, B., Kaplan, D. and Gelfand, E.W. (1996) Nervegrowth factor signal transduction in human B lymphocytes ismediated by gpl40trk Eur. J. Immunol., 26: 1985-1992.

Minagar, A., Shapshak, P., Fujimura, R., Own by, R., Heyes, M.and Eisdorfer, C (2002) The role of macrophage/microgliaand astrocytes in the pathogenesis of three neurologicdisorders: HI V-associated dementia, Alzheimer disease, andmultiple sclerosis. J. Neurol. Sci., 202: 13-23.

Mizuma, H., Takagi, K., Miyake, K., Takagi, N., Ishida, K.,Takeo, S., Nitta, A., Nomoto, H., Furukawa, Y. andFurukawa, S. (1999) Microsphere embolism-induced eleva-tion of nerve growth factor level and appearance of nervegrowth factor immunoreactivity in activated T-Iymphocytesin the rat brain. J. Neurosci. Res., 55: 749-761.

Moalem, G., Monsonego, A., Shani, Y., Cohen, LR. andSchwartz, M. (l999a) Differential T cell response in centraland peripheral nerve injury: connection with immuneprivilege. FASEB J., 13: 1207-1217.

Moalem, G., Leibowitz-Amit, R., Yoles, E., Mor, F.,Cohen, LR. and Schwartz, M. (l999b) Autoimmune T cellsprotect neurons from secondary degeneration after centralnervous system axotomy. Nat. Med., 5: 49-55.

Moalem, G., Gdalyahu, A., Shani, Y., Otten, U., Lazarovici, P.,Cohen, LR. and Schwartz, M. (2000) Production ofneurotrophins by activated T cells: implications for neuro-protective autoimmunity. J. Autoimmun., 15: 331-345.

Muhallab, S., Lundberg, C, Gielen, A.W., Lidman, 0.,

Svenningsson, A., Piehl, F. and Olsson, T. (2002)Differential expression of neurotrophic factors and inflamma-tory cytokines by myelin basic protein-specific and otherrecruited T cells infiltrating the central nervous system duringexperimental autoimmune encephalomyelitis. Scand. J.Immunol., 55: 264-273.

Nakagawara, A. (2001) Trk receptor tyrosine kinases: a bridgebetween cancer and neural development. Cancer Lett., 169:107-114.

400

Noga, 0., Englmann, C., Hanf, G., Grutzkau, A., Guhl, S. and

Kunkel, G. (2002) Activation of the specific neurotrophinreceptors TrkA, TrkB and TrkC influences the function ofeosinophils. Clin. Exp. Allergy, 32: 1348-1354.

Ochs, G., Penn, R.D., York, M., Giess, R., Beck, M., Tonn, l.,

Haigh, l., Malta, E., Traub, M., Sendtner, M., Toyka, K.V.(2000) A phase 111I trial of recombinant methionylhuman brain derived neurotrophic factor administered byintrathecal infusion to patients with amyotrophic lateralsclerosis. Amyotroph. Lateral. Scler. Other Motor NeuronDisord., I: 201-206.

Okano, H., Yoshizaki, T., Shimazaki, T. and Sawamoto, K.

(2002) Isolation and transplantation of dopaminergic neu-rons and neural stem cells. Parkinsonism Relat. Disord., 9:23-28.

Ota, K., Matsui, M., Milford, E.L., Mackin, G.A., Weiner, H.L.and Hafler, D.A. (1990) T-cell recognition of an immunodo-minant myelin basic protein epitope in multiple sclerosis.Nature, 346: 183-187.

Otten, U., Ehrhard, P. and Peck, R. (1989) Nerve growth factor

induces growth and differentiation of human B Iymphocytes.Proc. Natl. Acad. Sci. USA, 89: 10059-10063.

Otten, U. and Gadient, R.A. (1995) Neurotrophins andcytokines-intermediaries between the immune and nervoussystems. Int. l. Dev. Neurosci., 13: 147-151.

Patapoutian, A. and Reichardt, L. (2001) Trk receptors:mediators of neurotrophin action. Curr. Opin. Neurobiol.,11: 272-280.

Popovich, P.G. and lones, T.B. (2003) Manipulating neuroin-flammatory reactions in the injured spinal cord: back tobasics. Trends Pharmacol. Sci., 24: 13-17.

Rasouly, D., Rahamim, E., Lester, D., Matsuda, Y. andLazarovici, P. (1992) Staurosporine-induced neurite out-growth in PCI2 cells is independent of protein kinase Cinhibition. Mol. Pharmacol., 42: 35-43.

Ricci, A., Greco, S., Mariotta, S., Felici, L., Amenta, F. and

Bronzetti, E. (2002) Neurotrophin and neurotrophin receptorexpression in alveolar macrophages: an immunocytochemicalstudy. Growth Factors, 18: 193-202.

Roitt, I., Brostoff, l. and Male, D. (2001) Immunology.

Mosby, Philadelphia, pp. 15-64.Roux, P.P., Dorval, G., Boudreau, M., Angers-Loustau, A.,

Morris, S.l., Makkerh, l. and Barker, P.A. (2002) K252a andCEPI347 are neuroprotective compounds that inhibit mixed-lineage kinase-3 and induce activation of Akt and ERK.J. BioI. Chem., 277: 49473-49480.

Ruth, D. and Melamed, E. (2002) New drugs in the futuretreatment of Parkinson's disease. J. Neurol., 249: 30-35.

Sawada, J., Itakura, A., Tanaka, A., Furusaka, T. andMatsuda, H. (2000) Nerve growth factor functions as a chemo-

attractant for mast cells through both mitogen-activatedprotein kinase and phosphatidylinositol 3-kinase signalingpathways. Blood, 95: 2052-2058.

Schenone, A., Gill, l.S., Zacharias, D.A. and Windebank, A.J.

(1996) Expression of high- and low-affinity neutrophinreceptors on human transformed B lymphocytes. J.Neuroimmunol., 64: 141-149.

Schori, H., Lantner, F., Shachar, I. and Schwartz, M. (2002)

Severe immunodeficiency has opposite effects on neuronalsurvival in glutamate-susceptible and -resistant mice: adverseeffect of B cells. l. Immunol., 169: 2861-2865.

Schwartz, M., Moalem, G., Leibowitz-Amit, R. and Cohen, I.R.

(1999) Innate and adaptive immune responses can bebeneficial for CNS repair. Trends Neurosci., 22: 295-299.

Schwartz, M. and Kipnis, J. (2001) Protective auto immunity:regulation and prospects for vaccination after brain andspinal cord injuries. Trends Mol. Med., 7: 252-258.

Schwartz, M. and Moalem, G. (2001) Beneficial immuneactivity after CNS injury: prospects for vaccination.J. Neuroimmunol., 1l3: 185-192.

Serpe, Cl., Kohm, A.P., Huppenbauer, CB., Sanders, V.M.

and Jones, K.l. (1999) Exacerbation of facial motoneuronloss after facial nerve transection in severe combined

immunodeficient (scid) mice. l. Neurosci., 19: RC7.Serpe, Cl., Sanders, V.M. and lones, K.l. (2000) Kinetics of

facial motoneuron loss following facial nerve transection insevere combined immunodeficient mice. l. Neurosci. Res., 62:273-278.

Shoichet, M.S. and Winn, S.R. (2000) Cell delivery to the

central nervous system. Adv. Drug Deliv. Rev., 42: 81-102.Sin, A.Z., Roche, E.M., Togias, A., Lichtenstein, L.M. and

Schroeder, l.T. (2001) Nerve growth factor or IL-3 inducesmore IL-l3 production from basophils of allergic subjectsthan from basophils of nonallergic subjects. l. Allergy Clin.Immunol., 108: 387-393.

Skaper, S.D. and Walsh, F.S. (1998) Neurotrophic molecules:strategies for designing effective therapeutic molecules inneurodegeneration. Mol. Cell Neurosci., 12: 179-193.

Skaper, S.D., Pollock, M. and Facci, L. (2001) Mast cellsdifferentially express and release active high molecular weightneurotrophins. Brain Res. Mol. Brain Res., 97: 177-185.

Solomon, A., Aloe, L., Pe'er, l., Frucht-Pery, l., Bonini, S.,

Bonini, S. and Levi-Schaffer, F. (1998) Nerve growth factor ispreformed in and activates human peripheral blood eosino-phils. J. Allergy Clin. Immunol., 102: 454-460.

Soontornniyomkij, V., Wang, G., Pittman, CA., Wiley, CA.and Achim, CL. (1998) Expression of brain-derived neuro-trophic factor protein in activated microglia of humanimmunodeficiency virus type I encephalitis. Neurophatol.Appl. Neurobiol., 24: 453-460.

Stadelmann, C, Kerschensteiner, M., Misgeld, T., Bruck, W.,

Hohlfeld, R. and Lassmann, H. (2002) BDNF and gpl45trkBin multiple sclerosis brain lesions: neuroprotective interac-tions between immune and neuronal cells? Brain, 125: 75-85.

Storch, A. and Schwarz, J. (2002) Neural stem cells and

neurodegeneration. Curr. Opin. Investig. Drugs, 3: 774-781.

:=

Susaki, Y., Shimizu, S., Katakura, K., Watanabe, N.,Kawamoto, K., Matsumoto, M., Tsudzuki, M.,

Furusaka, T., Kitumara, Y. and Matsuda, H. (1996)Functional properties of murine macrophages promoted bynerve growth factor. Blood, 88: 4630-4637.

Tarn, S.Y., Tsai, M., Yamaguchi, M., Yano, K.,Butterfield, J.H. and Galli, S.J. (1997) Expression offunctional trkA receptor tyrosine kinase in the HMC-Ihuman mast cell line and in human mast cells. Blood, 90:1807-1820.

Tonra, J.R., Curtis, R., Wong, V., Cliffer, K.D., Park, J.S.,Timmes, A., Nguyen, T., Lindsay, R.M., Acheson, A.,DiStefano, P.S. (1998) Axotomy upregulates the anterogradetransport and expression of brain-derived neurotrophicfactor by sensory neurons. J. Neurosci., 18: 4374-4383.

Thorpe, L.W. and Perez-Polo, J.R. (1987) The influence ofnerve growth factor on the in vitro proliferative response ofrat spleen Iymphocytes. J. Neurosci. Res., 18: 134-139.

Weiss, J.H., Hartley, D.M., Koh, J. and Choi, D.W. (1990) Thecalcium channel blocker nifedipine attenuates slow excitatoryamino acid neurotoxicity. Science, 247: 1474-1477.

~

401

Ulvestad, E., WiIliams, K., Bjerkvig, R., Tiekotter, K., Antel, J.and Matre, R. (1994) Human microglial cells havephenotypic and functional characteristics common withboth macro phages and dendritic antigen-presenting cells.J. Leukoc. BioI., 56: 732-740.

Xiang, Z. and Nilsson, G. (2000) IgE receptor-mediated releaseof nerve growth factor by mast cells. Clin. Exp. Allergy, 30:1379-1386.

Voles, E. and Schwartz, M. (1998) Degeneration of sparedaxons following partial white matter lesion: implications foroptic nerve neuropathies. Exp. Neurol., 153: 1-7.

Ziemssen, T., Kumpfel, T., Klinkert, W.E.F., Neuhaus, O. andHohlfeld, R. (2002) Glatiramer acetate-specific T-helper l-and 2-type cell lines produce BDNF: implications formultiple sclerosis therapy. Brain, 125: 2381-2391.

Zuddas, A., Oberto, G., Vaglini, F., Fascetti, F., Fornai, F. andCorsini, G.U. (1992) MK-801 prevents l-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced parkinsonisms in pri-mates. J. Neurochem., 59: 733-739.