Towards Non Invasive Nerve Growth Factor Therapies for

Journal of Alzheimer’s Disease 15 (2008) 255–283 255IOS Press

Towards Non Invasive Nerve Growth FactorTherapies for Alzheimer’s Disease

Antonino Cattaneoa,b,∗, Simona Capsonia,c and Francesca Paolettia

aEuropean Brain Research Institute (EBRI) – Rita Levi Montalcini Foundation, Rome, ItalybInternational School for Advanced Studies (SISSA), Trieste, ItalycLay Line Genomics S.p.A., Roma, Italy

Abstract. In the past thirty years, nerve growth factor (NGF) has received much attention for its potential role as a therapeuticagent for Alzheimer’s disease (AD) due to its neurotrophic activities on basal forebrain cholinergic neurons. This attentionhas been renewed by recent findings that provide new causal links between defects in NGF signaling, transport or processingto the activation of the amyloidogenic route and, more generally, to AD neurodegeneration. Thus, the concept of therapeuticadministration of human recombinant NGF in AD patients has a strong rationale, being further validated by recent and ongoingclinical trials with a gene-therapy approach. However, the widespread clinical application of gene or cell-therapy strategies forthe delivery of NGF to AD patients seems unpractical, and it would be more advantageous to have non-invasive methods, thatshould also limit the adverse effects of NGF in activating nociceptive responses. This review will describe: 1) the data frompreclinical and clinical studies underlying the rationale of NGF as a potential therapeutic agent for AD; and 2) the alternativestrategies to reach adequate concentrations of NGF in relevant brain areas while preventing the onset of adverse effects.

Keywords: hNGF-61, NGF Therapy, Non-Invasive Delivery, proNGF, Therapeutic Window

INTRODUCTION

Progressive and inexorable synaptic and neuronalloss occurs in Alzheimer’s disease (AD), leading tocognitive decline. Synapse loss is the major correlateof cognitive impairment in AD, in comparison to theclassical neuropathological hallmarks of AD (amyloidplaques and neurofibrillary tangles) [231]. The dis-covery of the selective vulnerability and disruption ofthe cholinergic system in AD [73,247] laid the groundfor the “cholinergic hypothesis” of AD [19,96] and forassociating the cholinergic hypofunction with cogni-tive deficits typical of AD [96,213]. It has been pos-tulated for a long time [12] that deficiencies in neu-rotrophic factors in targets of innervation cause neuron-selective neurodegeneration. Neurotrophic factors, as

∗Corresponding author: Prof. Antonino Cattaneo, European BrainResearch Institute (EBRI) – Rita Levi Montalcini Foundation, Viadel Fosso di Fiorano 64, 00143 Rome, Italy. Tel.: +39 06501703024;Fax: +39 06501703335; E-mail: [email protected].

a class, therefore offer the potential to treat neurode-generative disorders, by preventing neuronal degener-ation and by compensating for cell and synapse lossafter (or while) it has occurred. Neurotrophic factors ofthe neurotrophin family offer the additional potentialbenefit that they influence not only the survival of targetneurons, but also their synaptic functions [112,130].

Nerve growth factor (NGF) [148,149], the first mem-ber of the neurotrophin family, supports neuronal sur-vival during development and influences neuronal func-tion throughout adulthood. In the central nervous sys-tem (CNS), mature basal forebrain cholinergic neurons(BFCNLs) are highly dependent on the availability ofNGF for the maintenance of their biochemical and mor-phological phenotype and for survival after lesions orCNS insults [89,108,110]. For these reasons, exploita-tion of NGF activities on BFCNs may providean attrac-tive therapeutic option for preventing cholinergic celldeath and improving the function of cholinergic neu-rons by preserving their synapses. However, the devel-opment of an NGF therapy is constrained by the dual

ISSN 1387-2877/08/$17.00 2008 – IOS Press and the authors. All rights reserved

256 A. Cattaneo et al. / Towards Non Invasive Nerve Growth Factor Therapies for Alzheimer’s Disease

conflicting need to achieve adequate concentrations inthe relevant brain areas containing degenerating targetneurons while preventing unwanted adverse effects onnon-target regions or cells. In this article, we will de-scribe: 1) recent preclinical cell and animal studies thatbroadens and further strengthens the rationale and thescope for the NGF therapeutic approach; 2) the presentdifficulties and obstacles facing the targeted deliveryof NGF to the CNS in relation to the adverse effectsto be avoided; and 3) the alternative strategies to reachadequate concentrations in relevant brain areas whilepreventing the onset of adverse effects. We shall thendescribe the two main approaches currently pursued tomeet the NGF therapeutic window, in order to developa successful therapy based on NGF, namely the gene-therapy approach and the nasal delivery of recombinantforms of NGF.

Finally, the neurotrophic option can, in princi-ple, be pursued with other neurotrophins [e.g., brain-derived neurotrophic factor (BDNF)] or with smallmolecule approaches that target the NGF signaling andmetabolism cascade at different levels, and a brief crit-ical overview will be provided.

RATIONALE FOR THE DEVELOPMENT OFAN NGF-BASED THERAPY FOR AD

NGF and AD: The cholinergic connection

In the CNS, BFCNs provide major projections tothe cerebral cortex and the hippocampus and corticalcholinergic mechanisms are directly involved in supe-rior cognitive functions such as learning and memo-ry [96]. BFCNs have been identified as the most signif-icant NGF-sensitive target population. BFCNs expressboth TrkA and p75NTR NGF receptors [110,117], areable to retrogradely transport NGF from their corticalprojections up to their cell bodies [113], and respond toadministration of exogenous NGF with an increase ofcholinergicphenotypicmarkers [98,163]. This connec-tion has provided a first strong argument to link NGFand AD [118,190].

Although NGF availability does not appear neces-sarily an absolute requirement for the survival of BFC-Ns in development, at least as concluded by geneknock-out studies [72,82], an extensive body of com-pelling evidence establishes that NGF prevents neu-ronal death or age-related atrophy in BFCNs from theadult brain [139]. NGF is able to prevent BFCN death,following axotomy [89,109,137] and to ameliorate the

morphological and behavioral effects of their aging-dependent atrophy [89]. Most importantly, the rele-vance of NGF-mediated neurotrophic and neuroprotec-tive actions for potential human applications was fur-ther strengthened by studies on adult nonhuman pri-mate brains. In rhesus macaque and cynomolgous mon-keys, NGF prevents degeneration of BFCNs after le-sions and reverses spontaneous, age-related cholinergicneuronal atrophy [236,237].

Thus, NGF prevents the death and stimulates thesynaptic cholinergic function following a wide range ofdifferent neuronal damage mechanisms in rodents andnon-human primates.

Interestingly, a reduced expression of the TrkA re-ceptor has been demonstrated in BFCNs and neocor-tex from AD brains [57,168,169], and loss of TrkA-expressing BFCNs has been described in mild cognitiveimpairment (MCI) [170] and in early AD neuropatho-logical stages [170]. The decrease of TrkA expressionappears to parallel cognitive decline [68]. Remarkably,the AD-linked decline in TrkA cortical contents doesnot occur for the p75NTR receptor, determining dise-quilibrium in the TrkA/p75NTR ratio that might havefunctional consequences. Therefore, the putative “offTrk” cycle of deficient NGF signaling may contributeto the selective degeneration of cholinergic neurons ob-served in AD [69].

NGF and AD: The retrograde transport connection

NGF is retrogradely transported from the corticaland hippocampal targets, where it is synthesized, toBFCN cell bodies [113,136]. Studies on NGF mR-NA and protein expression in AD patients provided noevidence that failed synthesis of NGF in target areascauses degeneration of BFCNs [99,127,215]. On thecontrary, increased levels of NGF and proNGF proteinsin the target cerebral cortex and hippocampus of ADpostmortembrains were found, together with decreasedNGF immunoreactivity in the basal forebrain [83,84,167,215], suggesting a defect in the NGF retrogradetransport system in AD.

This led to the prediction that, more generally, failedaxonal transport of NGF signals might provide a com-mon link between reduction of NGF trophic support,cholinergic dysfunction and neurodegeneration [211].

Cytoskeletal transport dysfunctions are, indeed, acommon hallmark of AD [128]. Neurofibrillary tan-gles, and tau pathology in general, could physicallyblock transport of the NGF signaling complex. Indeed,defective transport might result from increased expres-

A. Cattaneo et al. / Towards Non Invasive Nerve Growth Factor Therapies for Alzheimer’s Disease 257

sion [227] or hyperphosphorylation of tau and the en-suing microtubule destabilization [6,7]. Alternatively,it might result from abnormal glycogen synthase ki-nase 3β (GSK3β) phosphorylation of kinesin or othertransport proteins [164].

Evidence of a role for amyloid-β protein precursor(AβPP) in axonal transport provides another signif-icant element to the NGF, AD and retrograde trans-port connection. AβPP was shown to function as amembrane receptor for kinesin-1, mediating the ax-onal transport of cargo proteins such as BACE, PS-1,GAP43, synapsin, the “Sunday-driver” SYD and Tr-kA [101,106,128,129].

Increased AβPP expression in trisomy 16 mutantmice has also been directly linked to reduced retrogradetransport, with particular regard to defective transportof NGF, resulting in cholinergic neuronal degenera-tion [64] which can be reversed by exogenous, directNGF delivery to BFCN cell bodies [64].

On the other hand, NGF deficits could directly con-tribute to axonal transport defects, through a modula-tion of tau expression, tau hyperphosphorylation or tauproteolysis. Thus, in PC12 cells NGF induces tau ex-pression [80], thereby contributing to the observed ef-fect of NGF to promote microtubule assembly in thesecells [26]. On the other hand, deprivation of NGF incells [175] and in mice [40,47] induces a strong and spe-cific hyperphosphorylation of tau. Moreover, in agedanti-NGF mice, tau appears to be abnormally cleaved atN-terminally located caspase cleavage site, concomi-tantly with a significant upregulation of caspase activ-ity [65]. N-terminal cleavage of tau appears to be al-so found in AD brains, as tangles become extracellu-lar [33,119]. The functional consequences of the anti-NGF induced loss of the N-terminal 25 aminoacids oftau correlates well, also, with the recent report of Fron-totemporal Dementia with Parkinsonism-17 (FTDP-17) mutations of an Arginine residue in the N-terminusof tau, that dramatically affects tau interaction with dy-nactin [154]. Thus loss of the N-terminal portion oftau would results in a tau protein that is unable to inter-act with the dynactin complex and having an impairedability to support axonal transport functions.

These findings suggest that a deregulation of intra-cellular transport and of axonal transport mechanisms,by different causes linked to AβPP, tau, and NGF sig-naling and/or processing abnormalities, might be onecrucial aspect for initiating a negative neurodegenera-tion loop in AD (see tau loop in Fig. 5), leading to a fur-ther reduced neurotrophic support to BFCNs. This alsosuggests that therapeutic NGF replacement adjacent to

cholinergic cell bodies, rather than in axon terminal re-gions, would bypass transport defects and allow NGFsignaling at the soma [62,135].

The emerging complexity of NGF signaling andprocessing: proNGF/NGF unbalance and AD

Other mechanisms, besides intracellular neuronaltransport defects, could account for a reduced neu-rotrophic support to BFCNs, inducing their loss, or at-rophy, in AD neurodegeneration. An unbalance in theproNGF/NGF ratio is another possibility of increasingrelevance.

NGF, like the other neurotrophins, is translated as apre-pro-protein [222]. Two splicing variants lead, re-spectively, to the formation of a long and a short formof precursor protein. Both proteins are glycosylatedinvivo, yielding different molecular weight forms identi-fied [84,86,147,165,173,181,201] (Fig. 1A).

The signal peptide mediating secretion is cleavedupon translocation into the endoplasmic reticulum. Inthe trans-Golgi network, the pro-peptide is further pro-cessed by furin at a very conserved dibasic amino acidsite, resulting in the release of the 26 kDa mature neu-rotrophin [86,165,166,173,216]. Other proteases, be-sides furin, can cleave the pro-peptide in the extracel-lular space [37,147], in particular plasmin and matrixmetalloprotease-7 (MMP-7).

Biological activity of NGF precursor protein

The pro-domain was originally proposed to assistin proper folding and secretion of neurotrophins [228]and to act,in vitro, as an intramolecular chaperone ofrecombinant human NGF expressed inE. coli [131].Conversely, the mature NGF region is strictly requiredfor proNGF to be correctly folded and structurally or-ganized [131]. An intramolecular interaction of theNGF pro-peptide region with the mature part was re-ported [103,132]. The first available structural data,suggest that the proNGF domain, which has the prop-erty of an intrinsically unfolded domain, can exist in atleast two conformations, an extended one, and a morecompact one, in which the pro-domain folds back on thewell structured mature NGF part of the protein [182] .

Recently it was found that in cultured neurons,proNGF exhibits a higher affinity for p75NTR thanfor TrkA and can induce p75NTR-dependent apop-tosis [102,147,184]. ProNGF can also bind to Tr-kA and induce neuronal survival, although to a less-er extent than mature NGF [87]. Upregulation of


A

Signal sequence I

a 34 kDa

Possible N-linked glycosylation sites

-187 -162 -121 -103

-53 -8

+1 +118

+120 STOP

b

-

27 kDa

Possible N- linked glycosylation sites

-121 -103

-53 -8

+1 +118

+120 STOP

Pro-sequence I Signal sequence II Pro-sequence II

Mature NGF

Signal sequence ISignal sequence I

a 34 kDa


-187 -162 -121 -103

-53 -8

+1 +118

+120 STOP

34 kDa


-187 -162 -121 -103

-53 -8

+1 +118

+120 STOP


-187 -162 -121 -103

-53 -8

+1 +118

+120 STOP

-187 -162 -121 -103

-53 -8

+1 +118 -187 -162 -121 -103

-53 -8

+1-162 -121 -103

-53 -8

+1-103

-53 -8-53 -8

+1 +118

+120 STOP

b

-

27 kDa


-121 -103

-53 -8

+1 +118

+120 STOP-

27 kDa


-121 -103

-53 -8

+1 +118

+120 STOP

27 kDa


-121 -103

-53 -8

+1 +118

+120 STOP


-121 -103

-53 -8

+1 +118

+120 STOP

-121 -103

-53 -8

+1 +118 -121 -103

-53 -8

+1-121 -103

-53 -8

+1-103

-53 -8-53 -8

+1 +118

+120 STOP

Pro-sequence IPro-sequence I Signal sequence IISignal sequence II Pro-sequence IIPro-sequence II

Mature NGFMature NGF

B

FurinMMP

PlasminOther convertases

Neurodegeneration&

Cell death

Survival

SurvivalSignalling

Neurodegeneration&

Cell deathSignalling

Sortilin

Sortilin

p75NTR p75NTR

p75NTR

p75NTR

TrkA

TrkA

proNGF NGF

Final outcome: depending on the relative ratios

FurinMMP


Neurodegeneration&

Cell death

Survival

SurvivalSignalling

Neurodegeneration&


Sortilin

Sortilin

p75NTR p75NTR

p75NTR

p75NTR

TrkA

TrkA

proNGF NGFFurinMMP


Neurodegeneration&

Cell death

Neurodegeneration&

Cell death

SurvivalSurvival

SurvivalSignallingSurvivalSignalling

Neurodegeneration&


Neurodegeneration&


Sortilin

Sortilin

p75NTR p75NTR

p75NTR

p75NTR

TrkA

TrkA

proNGF NGF

Final outcome: depending on the relative ratios

Fig. 1. A – Schematic representation of the major translation products arising from alternative splicing of the NGF transcript: the long pre-proNGFform of 34 kDa (a) and the short pre-proNGF form of 27 kDa (b). The two forms differ by the additional pro-sequence stretch I of 66 aminoacids at the N-term of the long form A. Upon removal of the signal sequence, the two products give rise to two proNGF proteins: the long formof proNGF of 32 kDa and the short form of proNGF of 25 kDa. The arrows mark the cleavage sites for furin, the double headed arrows representthe C-terminal processing site (post translational modification) and hexagons the potential N-glycosylation sites. B – Proposed scheme for thesignaling of NGF and proNGF, involving Sortilin receptor, besides the “traditional” receptors p75NTR and TrkA. The p75NTR/sortilin andp75NTR /TrkA receptor pairs can be expressed either on different cells (top left and top right “cells”) or on the same cell (bottom center). Inthe latter case, the outcome of proNGF signaling will likely depend on the relative ratios of NGF/proNGF ligands, as well as of their relativesignaling through the p75NTR/sortilin and p75NTR /TrkA receptor pairs. Figure modified from Nykjaer and colleagues [176].

proNGF was found to be causally linked to p75NTR-mediated oligodendrocyte(expressing p75NTR, not ex-pressing TrkA) death following spinal cord injury [20].

ProNGF was also found to be a physiological ligandfor p75NTR in corticospinal neurons after axotomy andto induce neurons death through p75NTR [102]. More-


over, microglia-derived proNGF has been shown to pro-mote photoreceptor cell deathvia p75NTR [226]. Inhumans, ProNGF can also be detected in extracellularfluids such as cerebrospinal fluid (CSF) after lesion [17,102]. ProNGF was also found to be the predominantform of NGF in normal brain and its levels increase inthe brain of patients affected by AD [84].

These findings opened a new scenario, whereby thebalance between cell death and cell survival might bedetermined by the ratio of proNGF and mature NGFsecreted by cells (reviewed in [50,55,111]). The dis-covery of sortilin, a member of the Vps10p-domain re-ceptor family, as a specific receptor for proNGF [176],added one more variable to the emerging complexity ofthe NGF system.

The proNGF receptor

Sortilin is a type I transmembrane protein expressedin a wide number of developing and adult tissues. Sor-tilin interacts with the pro-region of proNGF with highaffinity and does not bind mature NGF. ProNGF si-multaneously binds Sortilin (via the pro-sequence) andp75NTR (via the mature NGF), creating a ternary com-plex [156,177,181,245]. The binding of Sortilin toproNGF was found to be essential for proNGF-inducedneuronal cell death through p75NTR in various celltypes known to undergo apoptosis mediated by proNGFand blocking the proNGF/sortilin interaction could in-hibit proNGF-mediated apoptosis [176].

Expression of exogenous sortilin in Schwann cells –normally expressing only p75NTR – renders them sen-sitive to proNGF-induced apoptosis [176]. Moreover,proNGF promotes motor neuron death [79], an effectinhibited by blocking proNGF binding to sortilin [79].

Sortilin is, therefore, a third important receptor play-er in neurotrophin signaling. Indeed, sortilin mightserve as a co-receptor and molecular switch, enablingneurons expressing TrkA or p75NTR to respond toa pro-neurotrophin and to initiate pro-apoptotic (or“pro-neurodegeneration”, see below) rather than pro-survival action (reviewed in [214]). In the absenceof sortilin, a regulated activity of extracellular pro-teases may cleave proNGF to mature NGF, promotingTrk-mediated survival signaling. However, there is al-so evidence, in BFCNs, that the sole presence of thep75NTR and sortilin receptors is not sufficient to in-duce pro-neurotrophins-dependent apoptosis [245]. Atthe same time, the presence of both Trk and p75NTR

receptors is not enough to protect these neurons frompro-neurotrophin induced apoptosis [245]. Likewise,

a recent study proved that pro-neurotrophins bindingto Trk receptors required endocytosis and intracellu-lar proteolysis, thus indicating a subtle equilibrium inpro-neurotrophins biological effect upon different lev-els of expression of the receptors system [34]. Prelim-inary findings suggest that cortical sortilin levels areincreased in AD patients compared to normal controls([69]). Pro-neurotrophins and their signaling throughp75NTR has been proposed to be involved in alterationsof the translational machinery in AD neurons [241].The balance between the TrkA and p75NTR levels anddistribution could be critical in determining synapsestructure and the prevalence of the pro-neurotrophinsthrough p75NTR signaling could be linked to decreaseddendritic spine complexity [241].

Recently, a sortilin-deficient mouse was generat-ed [125], which allowed proof that the association be-tween sortilin and p75NTR is required for death signal-ing, and furthermore that proNGF is the pro-apoptoticfactor in vivo. Furthermore, this study demonstratedthat knocking out sortilin does not allow the normal-ly occurring, age-related cell death in superior cervi-cal ganglia (SCG), showing that the sortilin/proNGFpathway is involved in aged neurons death.

The link between pro-neurotrophins and age-relatedneurodegeneration was further confirmed [1,2], byshowing that sortilin levels are elevated in aged SCGand BFCNs, together with elevated proNGF levels. Aschematic representation of this complex situation isdepicted in Fig. 1B.

Therefore, an updated description of neurotrophinactivity in vivo involves two ligands (NGF and proNGF)and their three receptors: TrkA (predominantly forNGF), sortilin (for proNGF) and p75NTR (for bothNGF and proNGF) [50,112,122,152]. In this view,the occurrence of programmed cell death mediated bypro-neurotrophins, via the p75NTR/sortilin complex,would result from an equilibrium including the regulat-ed processing of pro-neurotrophins as well as the spa-tiotemporal control of sortilin availability. This fea-ture might explain why not all p75NTR-positive cellsundergo apoptosis in response to proNGF (reviewedin [214]). In this view, the p75NTR emerges as a keyplayer [18,177,197,203].

To date, little is known about the signaling pathwayselicited by the sortilin-p75NTR complex, but severalpossibilities exist. Sortilin might simply function to fa-cilitate proNGF binding to p75NTR. Alternatively, itscytoplasmic tail might provide a template for adaptorproteins to modulate or even activate p75NTR signal-ing [177]. Others proposed that sortilin, by binding to


the pro-neurotrophins at the Golgi level, might protectit from proteolytic cleavage and activation [35]. In-deed, binding of pro-neurotrophins to sortilin has beenproposed to be involved in the control of their sortingpathways [56,112,151].

In one scenario, proNGF was suggested to act asa “baseline” neurotrophic factor with lower activity,as a reservoir for neurotrophically more active matureNGF – when survival and/or differentiation signalingis needed – or as a facilitator of death pathways – whenapoptosis is required, or after injury or stress [8,86].

In conclusion, proNGF and pro-neurotrophins aretruly independent ligands, with distinct physiologicalfunctions and, most relevant, with signaling proper-ties [14], widely distinct from those of mature NGF, andnot merely intracellular precursors (reviewed in [112,241]). It should be noted that the distinction, and thediscussion [86,112], between the pro-survival actionsof NGF versus the pro-apoptotic actions of proNGF [86,112] is an extreme, and probably artificial view. Theactions of proNGF might be more adequately describedas favoring, or contributing to, neurodegeneration ina broader sense (see Fig. 1B), through signaling path-ways distinct from those activated by NGF [14].

Anti-NGF AD11 mice as an experimental model forproNGF/NGF unbalance

The distinct, pro-neurodegeneration, signaling prop-erties of proNGF lead to postulate that an imbalancein the NGFversus proNGF availability contributes toAD neurodegeneration [43,44,51]. Several findingsare consistent with this view. The subtle equilibriumbetween pro-neurotrophins and their mature forms istightly regulatedvia furin-specific cleavage and/orvia acascade of extracellular proteases [37,112,152]. Patho-logical alterations of this cascade might lead to thelack of proper neuronal trophic support in pathologicalstates, like AD, by altering the NGF/proNGF ratio.

Interestingly, proNGF isolated from human ADbrain, which effectively induces neuronal apoptosismediated by p75NTR [184], was found to functionallydiffer from that found in control brains (e.g., alteredglycosylation and more efficient induction of p75NTR

processing) [192]. According to this view, in earlyphases of AD, signaling in cholinergic neurons wouldchange from a pro-survival to a pro-apoptotic (or pro-neurodegeneration) one, as the TrkA levels decline.p75NTR levels would remain unchanged, sortilin levelsincrease and proNGF levels rise in cortical projectionsites during the progression of AD [2,69].

A significant demonstration of the functional con-sequences of a disequilibrium in the NGF/proNGFratios comes from the anti-NGF AD11 mouse mod-el [49]. AD11 anti-NGF mice represent a comprehen-sive murine model to study the functional consequencesof an unbalance between proNGF and mature NGF andits relationship to AD neurodegeneration. AD11 anti-NGF mice express, in the adult brain, a recombinantform of a well characterized anti-NGF mAbαD11 an-tibody [48,71], in which the constant regions of therat immunoglobulin have been substituted with the hu-man constant regions, to allow the detection of the anti-body against the mouse background [206]. AD11 miceprogressively develop, from 1.5–2 months onwards, acomprehensive neurodegeneration as well as function-al and behavioral impairments that encompass severalfeatures of human AD. AD11 anti-NGF mice show aprogressive impairment of working memory revealedby analysis of behavioral tasks such the object recog-nition test [24,40,74], radial maze [40,206] and Morriswater maze [24,40,74]. Synaptic plasticity deficits un-derlie the behavioral effects. Long term potentiation isdecreased in the AD11 cortex [180,187]. In the CA1area of AD11 hippocampus, nicotine fails to enhancethe synaptic efficacy of Schaffer collaterals [225] as itnormally does, a deficiency linked to a rearrangementof the GABAergic hippocampal circuitry [202].

The histological analysis reveals a neurodegenera-tive phenotype of classical known targets of NGF sig-naling, such as BFCNs (Fig. 2A, B) and sympathet-ic and sensory ganglia [39,47,206]. In addition, start-ing from two months onwards, and well correlatingwith the behavioral and functional deficits (Table 1),the brain of AD11 mice is characterized by a progres-sive and widespread increase of neurons expressing hy-perphosphorylated [40,47] (Fig. 2C, D) and truncatedforms [65] of the microtubule associated protein tau inthe cortex (starting from the enthorhinal cortex [47])and hippocampus and by intracellular (first) and extra-cellular (at later stages) accumulation of amyloid-β, lo-cated above all in the hippocampus (Fig. 2E, F and [42,47]).

The overall neurodegenerative phenotype of AD11mice can be fully reverted by NGF administration [41,74], showing that it is indeed the result of NGF seques-tration by the antibody, and firmly proving that NGFsequestration in the adult mouse brain results in AD-like neurodegeneration. But how can we explain this inmechanistic terms? The biochemical properties of antiNGF mAbαD11 [48], which is expressed in the brainof AD11 mice, provides an answer.


Table 1Phenotypic markers of neurodegeneration in AD11 mice (modified from Origlia et al. [180])

Phenotypic hallmarks Brain Areas 1 2 4 6 8 9 10 12 15ChAT reduction Basal forebrain − + + ++ ++ ++ ++ ++ ++Hyperphosphorylated tau* Entorhinal cortex − + + ++ ++ ++ ++ +++ +++

Occipital cortex − − + + + + ++ ++ +++Parietal cortex − − + + + ++ ++ +++ +++Hippocampus − − + + + ++ ++ ++ +++

Dystrophic neurites (tau-positive) Cerebral cortex − − − − + + ++ ++ +++Neurofibrillary tangles Cerebral cortex − − − − − − − + +++PHFs Cerebral cortex − − ND ND ND ND ND + +++MAP-2 altered distribution Cerebral cortex − + + ++ ++ ++ ++ +++ +++

Hippocampus − − + ++ ++ ++ ++ +++ ++Accumulation of Aβ Hippocampus − − − + + ++ ++ +++ +++β− amyloid plaques Hippocampus − − − − − − − ++ +++Neuronal loss Cerebral cortex − − − − + + + ++ ++DNA fragmentation Cerebral cortex − − − − − − − − +Cortical LTP deficit Cerebral cortex − + ND ++ ND +++ ND ND NDHippocampal nicotine enhancement failure ND ND ND + ND ND ND ND NDObject recognition test deficit ND − + + ++ ++ +++ +++ +++Radial maze ND − + − ++ ND ND ND NDMorris water maze deficit (acquisition) ND − − − + + ++ ++ ++Morris water maze deficit (target) ND − − − − − ++ ++ ++

∗ revealed using mAb AT8ND = not done

Table 2Binding constants of theαD11 antibody towards rm-NGF and rm-proNGF samples.

αD11(kinetic) αD11 (equilibrium)rm-NGF ka = 1.2·106 M−1s−1 KD < 10−12 M

kd < 10−6 s−1

rm-proNGF ka = 5.9·105 M−1s−1 KD = 1.2·10−9 Mkd = 6.9·10−4 s−1

In vitro binding affinity measurements showed thatmAbαD11 binds NGF and proNGF in qualitatively andquantitatively different ways (Fig. 2G and Table 2). Thebinding of the antibody to NGF was found to be virtu-ally irreversible, characterized by smaller dissociationconstant than of proNGF (KD = 10−12 and 10−9 forNGF and proNGF respectively). Overall, the antibodyis characterized by three orders of magnitude higheraffinity for NGF than for proNGF, with a very differentbinding kinetics towards the two NGF forms.

The preferential binding of mAbαD11 to matureNGF, with respect to proNGF and the different bind-ing kinetics, naturally lead the proposal of a mecha-nism for the onset of the anti-NGF induced neurode-generation in the AD11 mice model. Thus, mAbαD11 would determine, under limiting concentrationsin the mouse brain, an experimentally induced func-tional imbalance between NGF and proNGF [44,49],by irreversibly “sequestering” mature NGF, while leav-ing proNGF free to act in the functional “absence” ofmature NGF. In particular, proNGF would activate thepro-neurodegeneration, pro-amyloidogenic pathways,

through interaction with the sortilin and p75NTR re-ceptors. A first demonstration of this prediction [43,44,49] comes from studies in which AD11 mice havebeen crossed to p75NTR homozygous knock-out mice(p75NTR [146]). In keeping with the prediction, theresulting offspring (AD12 mice) shows a complete re-version of the amyloid-β (Aβ) phenotype in the hip-pocampus [44,49] (Fig. 3).

Thus, a mechanistic characterization of the AD11model allowed to show that an experimentally-induced,incorrect balance between unprocessed proNGF andNGF signaling is indeed able to determine the onsetand progression of AD-like neurodegeneration in themouse brain: the “too little NGF-too much proNGF”hypothesis for AD neurodegeneration [44,49] (Fig. 4).

AβPP/TAU PROCESSING AND NGFSIGNALING: LINKING NGF DEFICITS TOTHE ACTIVATION OF THE AMYLOIDOGENICNEURODEGENERATIVE CASCADE

Results from the anti-NGF AD11 mouse modeldemonstrate that an experimentally-induced unbalancebetween unprocessed proNGF and NGF signaling isindeed able to activate the amyloidogenic arm of AβPPprocessing. In addition to the connection linkingNGF transport and AβPP intracellular trafficking (seeabove), this provides a significant and direct demon-stration of the complex interplay and cross talk be-


NGF and proNGF binding to α D11

-20

0

20

40

60

80

100

120

140

160

0 100 200 300 400 500

time (s)

Res

p U

nit

s (R

U)

NGF

proNGF

G

Fig. 2. A to F – Phenotypic neuropathological hallmarks in 15 month-old anti-NGF AD11 mice. At this age, these transgenic mice show(A) a decreased ChAT immunoreactivity in the basal forebrain accompanied by shrinkage of cholinergic neurons (arrows), compared to (B)age-matched wild type mice. Immunoreactivity for hyper phosphorylated tau, revealed by mAb AT270, is shown in (C) with shrinked neuronalcell bodies (asterisks) and neurites (arrows). Such a staining is not present in (D) wild type mice. In the hippocampus, (E) Aβ plaques areshown in the radial and molecular layers which are devoid of Aβ labeling in (F) wild type mice. Scale bar= 100µm. G – Surface PlasmonResonance (SPR) analysis, using Biacore, of the binding of theαD11 anti-NGF antibody to rm-NGF (recombinant mouse NGF - Thin curve)and rm-proNGF (recombinant mouse proNGF - Thick curve). TheαD11 anti-NGF antibody was immobilized on Fc2 (8000 RU), while Fc1was left as a blank. The neurotrophins were injected as analytes on the chip at different concentrations. One single concentration of rm-NGFand rm-proNGF is shown in the Figure for seek of simplicity. The curves were blank subtracted. The curves show association (time: 100–200s)and dissociation (time: 200–500s) of the neurotrophins, injected as analytes, from theαD11 antibody immobilized on the chip. The dissociationcurves show that NGF dissociates much slower than proNGF from the antibody.

tween the NGF ligand(s)/receptor(s) and the AβPPsignaling systems [107,204]. In turn, NGF deficitshave also profound influences on the aberrant phos-

phorylation or proteolysis of tau (see above). Onthe side of the NGF system, the ligand NGF/proNGF(dis)equilibrium (see above) is paralleled by a corre-


0

5

10

15

20

25

30

Age (months)

N A

β d

epo

sits

/sec

tio

n

WT mice

P75NTRexonIII (-/-) anti-NGF mice AD12 mice

2

6 15

*

*

A B

C D

E

Fig. 3. Expression of Aβ in AD11xp75NTR−/−mice (AD12) mice.Immunohistochemistry for Aβ in the hippocampus of (A) wild type(B) p75NTR−/− (C) AD11 and (D) AD12 mice. Scale bar=200 µm. (E) Quantification of the number of Aβ deposits at threedifferent ages. Values are mean± S.E.M.

sponding (dis)equilibrium at the level of the receptors.Indeed, an aging pathway has been shown to controlthe TrkA to p75NTR receptor switch and Aβ peptidegeneration in cells [66].

A further link connects the modulation of AβPPphosphorylation by the NGF receptor TrkA, couplingAβPP via the Shc/Grb2 adaptor proteins to signal-ing pathways associated to survival [208,229,230].In addition to the AβPP-TrkA interactions, the NGFand the AβPP systems show an additional cross-talkbetween Aβ and p75NTR [67,252]. On one hand,the p75NTR undergoes regulated intramembrane pro-teolysis byγ-secretase, with theγ-cleavage sites ofAβPP and p75NTR being highly homologous,releasingp75ICD to transmit a signal to the nucleus, in analogy to

the ICD signal derived by the cleavage of AβPP [209].On the other hand, Aβ peptide binds trimers as wellas monomers of p75NTR and activates receptor signal-ing [252]. While the impact of such a cross-talk onAD is still debated, recent work provides significantfunctional evidence, directly linking NGF deficits tothe activation of the AD amyloidogenic pathway [157].The withdrawal of NGF from NGF-differentiatedPC12cells induces a rapid increase in AβPP levels and anoverproduction of Aβ peptide. At the same time, thecells undergo a progressive apoptotic death. Theseeffects could be blocked by treatment with either in-hibitor ofβ- andγ-secretases or with antibodies direct-ed against Aβ. In addition, silencing of AβPP mR-NA with siRNA treatment also reduces Aβ productionand prevents NGF removal induced cell death. Thisprovides evidence for a mechanism in which a discon-tinued or limited supply of NGF can activate a patho-logical pathway of AβPP processing, triggering down-stream apoptotic cell death [157].

NGF can be therefore considered as an anti-amyloidogenic factor, that normally keeps the amy-loidogenic pathway under control. Accordingly, when-ever a normal neurotrophic supply of NGF to targetneurons is shortened or interrupted, the amyloidogenicpathway is activated and a negative feedback, toxic loopis activated. Further contributing to this negative loopis the one connecting NGF and tau pathology.

CONCLUSION I: THE NGF HYPOTHESIS FORNEURODEGENERATION

A challenge in unraveling AD has been to bridgethe knowledge of the molecular mechanism of rare,early onset, familial forms to the etiology of the com-mon late-life disorder. Although the clinical and neu-ropathological phenotypes of the dominantly inheritedforms are closely similar to those of “sporadic” AD(save for age of onset), precisely why cerebral Aβ lev-els are increased in the latter remains elusive [219].

Therefore, there is great interest to take into consid-eration the “non-genetic elevators of Aβ” and the pre-ceding discussion highlights the conclusion that neu-rotrophic deficits and/or abnormal processing of neu-rotrophins provide a direct link with the activation ofpathological amyloidogenic pathways. From this pic-ture, we can therefore conclude that the links betweenNGF and AD go well beyond the long established neu-rotrophic actions on BFCNs and formulate an updatedversion of the NGF hypothesis for AD neurodegener-


++

++

D11 MAb

++

AD11 mice

Neurodegeneration(sortilin, p75NTR)

Survival(TrkA, p75NTR)

proNGF NGF

free proNGF

Normal conditions(WT) Unbalance NGF/proNGF

Decreased protease activity

ADAD

++

++

D11 MAb

++++

Dα 11 MAb

++++++

AD11 mice

Neurodegeneration(sortilin, p75NTR)

Survival(TrkA, p75NTR)

proNGF NGF

free proNGF

Normal conditions(WT) Unbalance NGF/proNGF

Decreased protease activity

ADAD

Fig. 4. Schematic representation of the proNGF/NGF unbalance model. On the left side of the scheme, the “normal” conditions, when the NGFsignaling is achieved mainly through the NGF/TrkA/p75NTR system; this is the condition when “survival or neurotrophic signaling” takes place.On the middle-right part of the scheme, the representation of the pathological conditions: in red, the AD conditions, in magenta the AD11 mousemodel condition. In both cases, an unbalance in the proNGF/NGF takes place, with the functional consequence of a reduced available amountof circulating mature NGF. In these conditions, thefavored signaling pathway is the “neurodegeneration/cell death” one, involving proNGF,p75NTR and sortilin.

β

A production A oligomerization SynapticSynaptic failurefailure NeuronalNeuronal lossloss

AD AD DementiaDementia

(FAD(FAD&&

LOAD)LOAD)

AmyloidAmyloid cascadecascade

EOAD (FAD)

LOAD (Sporadic AD)

ReducedReducedneurotrophicneurotrophic supportsupport

TauTau pathologypathology

APP APP pathologypathology

AxonalAxonal transporttransport

MutationsMutations in in APP or APP or presenilinpresenilin

EnvironmentalEnvironmental factorsfactorsRiskRisk factorsfactors

&&OthersOthers

NGF NGF transporttransport failurefailureNGF processing NGF processing unbalanceunbalanceNGF NGF receptorreceptor unbalanceunbalance

A production A oligomerization SynapticSynaptic failurefailure NeuronalNeuronal lossloss


(FAD(FAD&&

LOAD)LOAD)


EOAD (FAD)

LOAD (Sporadic AD)





MutationsMutations in in APP or APP or presenilinpresenilin


&&OthersOthers


Aβ production Aβ oligomerization SynapticSynaptic failurefailure NeuronalNeuronal lossloss


(FAD(FAD&&

LOAD)LOAD)


EOAD (FAD)

LOAD (Sporadic AD)





MutationsMutations in in βAPP or APP or presenilinpresenilin


&&OthersOthers


Fig. 5. The NGF-APP and the NGF-Tau loops in Sporadic Late Onset AD (LOAD). Genetic predisposition and mutations can affect themetabolism of Aβ in FAD (Familial AD) and Early Onset AD (EOAD), while environmental and others risk factors can lead to sporadic AD.Among the “other factors”, we suggest an emerging role for NGF and a functional unbalance of its transport, processing and receptor interactions(see green box at the bottom of the figure), leading to the establishment of a negative loop between the neurotrophic system, the amyloidcascade and tau pathology. The green arrows in the figure mark this suggested NGF-AβPP and the NGF-Tau loops in LOAD. These loops areinterconnected.

ation, whereby the common link is a failure or an in-sufficient NGF signaling, leading to inadequate neu-rotrophic support. The failure or unbalance of NGFsupport could be due to different causes in the overall

cascade(s) of events involving NGF bioactivity: 1) de-creased synthesis; 2) unbalanced or altered processing;3) alterations in receptor expression and/or activity orexpression ratios; and 4) altered retrograde transport.


These events would be “located” upstream of the “amy-loid cascade”, as currently described [217], but wouldbe part of a negative feedback loop that involves sev-eral feedback steps from the downstream process itself(e.g., links between AβPP, tau and axonal transport),also directly involving functional connections betweenNGF and the tau system. Also, the cellular targetsfor NGF actions, in this negative loop, could be morewidespread than envisaged so far. A scheme of thepossible events or pathways linking NGF to the Aβcascade is depicted in Fig. 5.

Within this theoretical frame, any therapy aimedat re-establishing the correct balance between ligands(and receptors) of the NGF pathway appears to have aclear and strong rationale.

NGF-BASED THERAPIES FOR AD

NGF clinical applications: problems to be solved

Having established that re-equilibrating a correct bal-ance between ligands (and receptors) of the NGF path-way is a therapeutically relevant target, also for its con-nections and links at the core of AβPP metabolism, itis clear that the therapy of first choice would be to useNGF itself as a drug. The viable clinical applicationof NGF requires providing a solution to major obsta-cles, linked to effective CNS delivery and limitation ofadverse effects (most notably, pain).

NGF does not readily cross the blood brain barrier(BBB) due to its size and net charge. Ever since thediscovery of the capacity of NGF to rescue BFCNs,the delivery of NGF to the CNS was an issue, and ex-periments in animal models resorted to NGF intracere-broventricular (ICV) infusions, to prove the ability ofNGF to prevent BFCN degeneration [89,108,133,236].On the basis of these studies, the ICV way of NGFdelivery was employed in the very first clinical trials inAD patients [81,179] (see below).

A second issue is represented by the well-establishedpro-nociceptive (or algogenic) action of NGF on thesensory neurons of dorsal root ganglia and spinal cord,on which it modulates the transmission of pain sig-nals [189], and a pro-inflammatory action on inflam-matory cells (including mastocytes [4]), on which itactivates the release of inflammation mediators [23].

In humans, the capacity of NGF to cause pain hasbeen demonstrated in clinical trials in AD patients(see below), as well as during clinical trials under-taken to explore the potential use of NGF in poly-

neuropathies [11,188]. The subcutaneous and intra-venous injections of NGF into the arms of healthy vol-unteers produce allodynia and hypersensitivity of theskin surrounding the point of injection. The problemwith minimizing its nociceptive algogenic effects alsoexists for the topical and local NGF treatment, such forocular pathologies [141].

In conclusion, a clinical application of NGF is lim-ited by a double constraint, of achieving an adequateand pharmacologically relevant concentration in targetbrain areas, while preventing its access to non-targetedareas, where adverse effects, such as pain, are elicited.

First clinical pilot trials

The two first NGF clinical trials were based on theICV infusion of NGF directly to the brain. The first trialwas performed using murine NGF infused into the rightventricle of a single patient, at a rate of 75µg/day for 3months. Positron emission tomography (PET) studiesdemonstrated modest increase in [11C]nicotine bindingand cerebral blood flow in the frontal and temporal cor-tices at the end of the infusion [121,179]. Weight losswas reported as an adverse effect in this first trial [121,179].

A second clinical pilot trial was attempted in threeAD patients [81] using ICV infusions of mouse NGF.Two of these patients received 75µg/day while the thirdwas infused with 16µg/day for 2 weeks and later at arate of 3.4µg/day for 10 weeks. In these patients,a tran-sient increase in scores was observed in some episodicmemory tests and there was an improvement in phys-iological parameters of brain function, such as nico-tine binding, cortical blood flow and electroencephalo-graphic activity [81]. No improvements were shown inthe Mini-Mental Status Examination (MMSE) scores.The trial was discontinued after the onset of debilitat-ing side effects due to NGF infusions. In particular,the patients developed back pain within 2 to 14 daysafter the beginning of ICV infusions and weight loss.These side effects were interpreted on the basis of ani-mal studies, reporting that ICV infusions of NGF pro-voked weight loss, sympathetic axon sprouting aroundthe cerebral vasculature, and migration and expansionof Schwann cells into the medulla oblongata and thespinal cord [89,123,248]. Thus, no other clinical trialusing this way of administration was performed andalternative methods of delivery were explored.


Gene-therapy approach

These results highlighted the fact that a viable clini-cal testing of NGF has to fulfill a dual need, to be effec-tively delivered to the brain regions containing targetneurons, while limiting excessive spread to no targetedregions, to avoid adverse effects. Gene and cell ther-apy methods have been considered as one approach tospecifically deliver NGF to the cholinergic neurons ofthe basal forebrain in a spatially restricted way. In anexvivo gene therapy approach, cells are genetically mod-ified in vitro to express a gene of interest, and are thenimplanted into the brain where they effectively act asbiological micropumps, secreting the desired protein.The efficacy of the NGF gene delivery approach to theBFCNs was tested in rodents and primates, showingthe effectiveness in preventing the cholinergic neuronaldegeneration after lesions in the adult brain [29,238]and to ameliorate age-related atrophy [29,53,223].

These preclinical results prompted to begin an 18month, open label, prospective Phase I clinical safetytrial of ex vivo NGF delivery in humans, which involveda dose escalation design in eight AD patients [239].The trial was designed to protect cholinergic neuronsfrom degenerationand augment the function of remain-ing cholinergic neurons by directly elevating cholineacetyltransferase (ChAT) function in neurons. Patientswere screened for the diagnosis of probable AD of mildseverity. Autologous fibroblasts, obtained from threeskin biopsies on each patient, were cultured, and genet-ically modified to produce and secrete the human NGFprotein. The patients underwent neurosurgery, for theintracerebral injection of their own primary fibroblastsinto the region of BFCNs in the brain, where neuronsunderwent atrophy as a result of AD. Of the eight testedpatients, two subjects experienced sub-cortical hemor-rhages during the stereotaxic cell injection, with con-sequences such as hemiparesis, and in one case, deathdue to cardiac arrest. The patient that survived wasnot included in the outcome of the trial. The trial re-sulted a nearly 40–50% reduction in their rate of de-cline on the MMSE and Alzheimer’s Disease Assess-ment Scale-Cog (ADAS-Cog) scales [239]. In the timeperiod from 6 months to 18 months after the implantof NGF secreting cells, two of six patients showed anactual improvement in mental function, one showed nodecline in function, two worsened only mildly, and onecontinued to show a typical AD rate of worsening [239].During the 6–18 month time period, a “high” NGF pro-duction is expected to be present and sufficient timehas passed to allow NGF to “remodel” and stimulate

the cholinergic neurons. Patients also showed a signifi-cant improvement in PET scans, after undergoing NGFtreatment. The subject who underwent autopsy showeda significant cell growth effect due to the exposure ofcholinergic neurons to NGF [239]. The outcome ofthis first trial prompted a second Phase I clinical trial,based on a direct,in vivo, NGF gene delivery into thebrain,via an Adeno Associate Virus (AAV)-vector sys-tem (CERE-110). Results in animals indicate that NGFtransgene delivery is reliable and accurate, that NGFtransgene distribution can be controlled by altering thedose, and that it is possible to achieve restricted NGFexpression limited to, but covering, the target brain re-gion [27]. NGF transgene expression appears to bestable and sustained at all time points, with no loss orbuild-up of protein over the long-term [27]. In addi-tion, results indicate that CERE-110 is neuroprotectiveand neurorestorative to BFCNs in the rat fimbria-fornixlesion and aged rat models, and has bioactive effectson young rat BFCNs [27]. Based on these preclinicalfindings, a Phase I clinical study to evaluate the safetyand efficacy of CERE-110 in AD subjects, is currentlyongoing [58].

In addition to the AAV-NGF/CERE-110 trial, a new,cell-basedin vivo delivery system has been developedand a Phase I trial is currently ongoing [174]. Thisbiodelivery platform is based on the demonstration thatimmortalized, allogenic fibroblasts, engineered to se-crete NGF and encapsulated within a polymer mem-brane, effectively prevented the loss of cholinergic neu-rons after fimbrial transections in rats [115,116,135].The biodelivery technology consists of a catheter-likedevice that contains genetically modified cells enclosedbehind a semi-permeable membrane that permits theinflux of nutrients and outflow of the therapeutic fac-tor(s) but does not allow direct contact of the foreigncells and the host brain [135].

A non-invasive approach to NGF therapy in AD: NGFas drug for AD

The NGF gene/cell therapy approaches provide animportant validation of the therapeutic potential of NGFin AD, and the outcome of the clinical studies willprovide important insights into the safety, efficacy andliabilities of NGF therapies. However, the approachis impractical to be extended to the large number ofpeople suffering from AD, since theex vivo or in vivoadministration modality is invasive and rather danger-ous and expensive (requiring hospitalization) [240] anddoes not appear to be suitable as prophylactic cheap


Fig. 6. Proposed definition of a therapeutic window in the NGF delivery. The red traces refer to the CNS, the blue ones to the systemic/bloodsystem. On the axis: concentration of the intranasally delivered NGF (abscissa), concentration of NGF that reaches the CNS (left ordinate axis,red) and the concentration of the NGF that reaches the blood (right ordinate axis, blue), respectively. The solid lines represent the increasingamount of NGF that reaches the CNS (red line) or the blood (or CSF) (blue line), respectively. The red and blue dashed lines respectivelyrepresent the CNS therapeutic threshold and the pain-inducing threshold (from blood stream or CSF). The “therapeutic window” (white region) isdefined as the NGF dose region between the minimal NGF concentration to reach the CNS, and the dose that reaches the pain-inducing thresholdin the circulating (or CSF) NGF, after nasal delivery. The left-side, red colored region represent the NGF concentrations that are not high enoughto reach the CNS. The right-side, blue colored region represents the NGF concentrations that reach higher than the pain-inducing threshold inblood stream (or CSF).

treatment for millions of aging people worldwide [212].Therefore, a route for an effective, non-invasive, deliv-ery of recombinant forms of NGF to the brain wouldappear to be more practical in the medium to long term,and this approach should be pursued. Indeed, the ben-efits of peripheral NGF administration would includeelimination of the risks and expense of cranial injection,and the ability to adjust or terminate dosing.

In principle, different approaches for the develop-ment of NGF as a drug, could be pursued, a systemicdelivery or a more targeted delivery (such as intranasaldelivery).

The success of an NGF-based therapy by the in-tranasal delivery route will require meeting an NGFtherapeutic window, to optimize the access to CNS,while minimizing the systemic NGF levels, to avoidunwanted pain effects (Fig. 6). The two critical param-eters in the NGF therapeutic window are the thresh-old concentration of NGF (in the bloodstream or in theCSF) inducing pain and the NGF levels in the targetbrain areas required to be pharmacologically effective.Any NGF administration route must be verified againstthese two parameters. For instance, for a systemic en-dovenous delivery of NGF (in humans), the thresholdconcentration of NGF that determines pain and mialgiais a dosage of 1µg/kg [188], while the threshold doseof NGF needed to achieve a measurable concentration

in the brain in humans is estimated to be≈ 11 µg/kg(see below).

NGF systemic delivery

A systemic administration route for an NGF-basedtherapy for AD is limited by the transport across theBBB. Indeed, the permeability of radioiodinated NGFat the brain nerve barrier was found to be significantlylow [194], with a permeability coefficient x surface area16.1 fold higher than albumin [193,194], a marker ofvascular permeability.

Thus, attempts have been made to modify NGF soas to improve its ability to cross the BBB. Antibodiesagainst the transferrin receptors located on the BBB en-dothelial cells, as well as drugs conjugated to them, cancross the BBB [92,183]. NGF was conjugated to theOX-26 antibody to the rat transferrin receptor and wasshown to retain biological activity [93]. Upon systemicadministration, NGF-OX26 was shown to improve spa-tial learning ability and to increase the size of choliner-gic neurons in the medial septum in lesioned and agedrats [16,52,134]. The conjugation to the OX-26 anti-body determined a 4.5% increase in the percentage ofNGF detectable in the brain parenchyma [93]. How-ever, OX-26 does not bind to the human transferrin re-ceptor, making impossible to translate this technology


to the clinic. Moreover, chemical crosslinking of twoproteins, such as NGF and an antibody, is not easilytranslatable into a pharmaceutical product.

To further exploit transferrin, or its receptor, as car-riers for targeting and delivery of NGF to the brain, tworecombinant NGF fusion proteins were engineered byjoining the human NGF precursor to the heavy chainvariable region of a chimeric antibody, specific for thehuman transferrin receptor [160], or by linking trans-ferrin to the carboxyl terminus of NGF, via the hingeregion from human IgG3 as a flexible linker [10]. Bothproteins bind human transferrin receptor and NGF re-ceptors and retain full NGF activity in PC12 cells [10,160] but they were never testedin vivo.

A second approach takes advantage of the fact thatcovalent modification of NGF with natural polyamines,such as putrescine, determines a 10-fold increase in thepermeability coefficient at the BBB, compared to theone of native NGF, but also a 6-fold decrease of thehalf-life of plasma NGF [195]. Such covalent modifi-cations of the NGF, either by glycation or polyaminemodification, facilitate its permeability at the BBB tophysiological doses that are capable of eliciting a biore-sponse [196].

Recently, a new approach was undertaken, throughthe encapsulation of NGF into liposomes (NGF- SSL-T) in which RMP-7, a ligand to the B2 receptor onbrain microvascular endothelial cells, was incorporat-ed into the liposomes’ surface. The delivery of NGFthrough the intravenous route determined a 2-fold in-crease in permeability of NGF to the BBB, with themajority of NGF distributed in striatum, hippocampusand cortex [251].

In conclusion, NGF can indeed be engineered ormodified in order to increase its permeability acrossthe BBB and the concentration that reaches the CNSupon systemic administration. However, the improve-ments in BBB permeability obtained are not enoughto achieve the therapeutic concentration in the brain,while remaining below the pain inducing threshold andto offset the complications of having an entirely newmolecule with respect to NGF itself. No pain mea-surements, performed in the models in which transportto the brain was determined, are available to permit acomparison with NGF doses inducing pain in humans.

Intranasal NGF delivery

To date, two alternative approaches have been under-taken, in order to target NGF to the brain, while min-imizing or reducing its systemic biodistribution: in-

tranasal delivery and topical application of an NGF so-lution on the ocular surface. The approach to deliv-er drugs to the brain through the intranasal route, tocircumvent the BBB, is well validated for many drugs.

In the case of AD, Phase I pilot studies based on theintranasal administration of insulin have been alreadyconducted, in MCI and early AD patients, showing anamelioration of cognitive impairment [198,200], verbalmemory and modulation of plasma Aβ content [199,200]. Another substance under clinical evaluation af-ter intranasal delivery is the activity-dependent neuro-protective protein (ADNP), which completed a PhaseI study, assessing the tolerability and pharmacokinet-ics of intranasal administration of ADNP in sequentialascending doses in healthy volunteers [100].

Intranasal administration of NGF through the olfac-tory pathway may provide a novel, non invasive, safeand effective route for the delivery of NGF to the brain,at pharmacologically relevant concentrations [41,54,74], without triggering adverse side-effects.

In particular, the intranasal administration of rhNGF(recombinant human NGF) to AD11 anti-NGF micedemonstrated that the neurotrophin reaches the brainin concentrations sufficient to improve the cognitivedeficit [74] and to completely rescue the neurodegen-eration [41]. In AD11 mice, the intranasal route wasalso compared to another non-invasive route of NGFdelivery to the brain, the ocular application of NGFdrops. This route was previously shown to improvecholinergic markers in normal rats [142] and in micetreated with ibotenic acid [78], but the mechanism(s)through which NGF reaches the brain after ocular de-livery are not characterized [142] and, most relevant,this route was not tested in an animal model display-ing a comprehensive neurodegeneration more similar toAD hallmarks. The side-by-side comparison betweenocular and intranasal delivery of NGF in AD11 miceshowed that the latter was much more effective in res-cuing neurodegeneration,showing more specificity andless probability to induce peripheral side effects [46].The intranasal delivery route, while leading to effectivetransport of NGF in AD relevant brain areas, does notlead to significant build up of NGF concentration intoregions where it is known to induce pain responses inman, such as the CSF (reaching a concentration muchlower than 45 pM [54]) and in the blood stream (whereit determines only a limited, albeit measurable, leak-age [91]). Thus, the intranasal delivery route appears tobe an effective compromise to meet the required ther-apeutic window, leading to effective biodistribution intarget brain areas, while minimizing the biodistribution


Fig. 7. Structural representation of the tertiary structure of the “tagged” hNGF-61. (A) Structural representation of the tertiary structure of mNGF(in blue); (B) Structural alignment of the tertiary structure of mNGF (in blue) and the hNGF (in green). In both figures, produced using Pymol,loop III and the side chain of S61 are depicted in red. Loop III (in red) does not interact with (C) TrkA (in cyan) and (D) p75 (in magenta)extracellular domains. Modified from Covaceuszachet al. [70].

to non-targeted districts where it is known to determineunwanted side effects.

Indeed, when administered intranasally to AD11anti-NGF mice, NGF or its derivatives do not providetrophic support to SCG nor do they induce nociception-related surrogate markers, such as calcitonin-gene re-lated peptide (CGRP) expression in spinal cord nervefibers, further proving that it does not reach pharma-cologically relevant concentrations in the bloodstreamand in the CSF [46].

The dose of NGF administered to AD11 micethrough the intranasal route (480 ng/kg), required toachieve a therapeutic effect in the brain, is 23 timeslower with respect to the dose that should be adminis-teredvia endovenous injection (≈ 11µg/kg) to achievethe same NGF concentration in the brain. If thesedosage were applicable to a human treatment, the in-travenous dose (11µg/kg) is ten times higher than thedose that was shown to trigger a nociceptive responsein man [188].

Thus, the intranasal delivery route provides a sig-nificant advancement towards meeting the NGF ther-apeutic window required for a successful NGF-basedtherapy (Fig. 6). An ideal NGF product for intranasaldelivery will require a careful determination of the ther-apeutic dose to be delivered and a favorable profile interms of the therapeutic range (facilitated access to CNSand/or reduced nociceptive properties, with equivalentneurotrophic properties). Indeed, the concentration ofendogenousNGF in the blood and tissues varies greatlyamong individuals, depending on individually varyingconditions, such as binding to plasma proteins [91],stress [5,25] or hormonal balance [25,224]. To addressthe problem of NGF traceability, for its accurate dosingagainst the background of endogenous NGF, we de-veloped an NGF point mutant (hNGF-61) [70] aimedat conservatively tagging hNGF, making it distinguish-able from endogenous NGF. A phylogenetic and struc-tural study led to the identification of amino acid in po-sition 61, as the tagged residue, leading to the synthesis


and expression of mutein hNGF-61, in which residuePro61 of human NGF was substituted by Ser61, presentin mouse NGF. Mutein hNGF-61 shows an identicalbiochemical and biological activity profile and potencyas hNGF, in keeping with the fact that position 61 islocated in loop III of the NGF molecule (Fig. 7A), anddoes not interact with both NGF receptors (Fig. 7B),and with the fact that mouse NGF is the gold standardfor NGF bioactivity. A monoclonal antibody is ableto selectively distinguish hNGF-61 from the endoge-nous hNGF [70].In vivo, the intranasal administrationof hNGF-61 rescues the behavioral deficit, as well asneurodegeneration, in adult [46] and aged AD11 anti-NGF mice [70]. Currently under development, hNGF-61 is the first lead of a series of NGF muteins furtheroptimized for their therapeutic use in AD, including aseries of muteins displaying a reduced pronociceptiveresponse in animal pain models [45].

A summary of the CNS delivery methods for NGFis reported in Table 3.

ALTERNATIVE APPROACHES

Small molecules with NGF agonist activity

In the attempt to overcome some of the drawbacksthat have limited the use of NGF in therapy (pharma-cokinetics, delivery, BBB transport, adverse effects),the development of small molecule neurotrophin recep-tor mimetics, or of NGF agonists in general, retainingthe biological activity of the natural protein, was un-dertaken [150,155]. The approaches towards the use ofsmall molecules in the NGF-based therapies comprisevarious types of compounds, most of them based on thesearch for an NGF mimetic able to induce an activationof the receptors, TrkA and/or p75NTR.

Various approaches aiming to isolate small moleculemimetics are based on structural studies, starting fromthe tertiary structure of NGF. In particular, the iden-tification of the structural determinants of the interac-tion between NGF and its receptors, allows designingmimetic peptides based on these interacting structures.This approach was pursued both to isolate p75NTR andTrkA activating NGF mimetics (reviewed in [150]).

The approach based on ligands targeting the p75NTR

receptor also has the potential to modulate pathophys-iological mechanisms underlying ADvia a number ofmechanisms, including Aβ toxicity (reviewed in [150]).

The recent findings of the increasing role ofproNGF in neurodegeneration of the nervous system

through p75NTR receptor (see previous paragraphs)raise the possibility of targeting the p75NTR with smallmolecules, able to inhibit proNGF-induced cell death(reviewed in [150]). One strategy is based on syn-thetic peptide mimetics of NGF loop I, able to inter-act with neurotrophins’ receptors signaling and pre-vent cell death (reviewed in [150]). Indeed, a recentreport has shown that small, nonpeptide p75NTR lig-ands induce survival signaling and inhibit proNGF-induced death [156]. These ligands, mimetic of theneurotrophin loop 1 domain, seem to activate not TrkAbut p75NTR and the downstream pathways involvingNFκB and PI3K. However, high doses are required toinhibit proNGF binding. The same kind of mimeticshave also been found to modulate p75NTR-dependentmotor neuron death [185].

A different type of NGF mimetics was developedbased on the Loop 4 on NGF, known to interact with theTrkA receptor [250]. Dimeric peptidomimetics basedon this structural scaffold (namely P92) were shown toact as partial NGF agonists, by ERK and AKT activa-tion and promotion of signaling activation [250].

In a different experimental approach, another set ofsmall molecules as NGF mimetics were developed (re-viewed in [186]), identified on the basis of three dif-ferent criteria, namely structural, binding and function-al mimicry. Based on these criteria, the mimetics canbe identified and optimized, also through identificationof target regions of the protein to mimic, namely “hotspots” (reviewed in [186]).

Following the approach of the hot spots identifica-tion, a set of peptidomimetics was identified. First-ly, C-(92-96), a bioactive cyclic peptide mimic, resem-bling the C-Dβ-loop of NGF, was identified as a TrkA-interacting region [21,22,75]. This peptidomimeticwas shown to act solely as a cyclic monomer andwas able to induce cellular differentiation and pro-vide trophic effects to embryonic dorsal root ganglion(DRG) cells in culture [22]. A second mimetic wasidentified from the analysis of the anti-TrkA monoclon-al antibody 5C3, a full agonist of TrkA, by binding toits D5 domain (reviewed in [186]). A small moleculemimetic of 5C3 was identified and termed D3, a macro-cyclic ring with the peptide side chain from the comple-mentary determining region of the 5C3 antibody. TheD3 mimetic can activate TrkA and is a partial TrkAagonist, in its ability to induce cell survival and differ-entiation in DRG neurons (reviewed in [186]). In agedrats, the D3 mimetic is neuroprotective and is able toimprove memory and the cholinergic phenotype ([36];reviewed in [186]).


Table 3CNS delivery methods for NGF (modified from Thorne et al. [91])

Administration route Advantages DisadvantagesParenteral systemic Simplicity of delivery Shortin vivo half-lives of proteins(i.v., i.m. or s.c.) No targeted delivery

Pleiotropic effects (pain)Delivery of native molecules potentially effective onlyfor peripheral diseasesPossible immunogenicity

Intracerebroventricular administration(infusion with pumps)

Continuous, programmable deliveryCan stop infusion if necessarySome degree of targeting achievable

Surgically invasiveDiffusional limitationsStructural barriers (pia-glia limitans)Need to refill on regular basis with attendant risk ofinfectionPleiotropic effects possible (pain)

Intraparenchymal, ICV or intrathecaladministration for gene therapy(implantation of genetically modifiedcells; injection of viral vectors forinvivo gene therapy)

Targeted delivery possible.May use allogeneic or xenogeneic celllines.No immune rejection.Targeted delivery possible.Transduce nondividing cells includingneurons.Sustained transgene expression possible.

Surgically invasive.Need for surgical retrieval if adverse effects experi-enced.Difficulty in controlling gene expression in dealingwith adverse effects.Safety issues with viral vectors.Poor diffusion of vectors into CNS (with ICV or ITinjection).Potentially effective only for focal diseases.

Intraparenchymal administration (im-plantation of polymer-release system)

Targeted delivery Surgically invasiveDiffusional limitations in CNS tissueInability to control release rate or turn offLoading capacity limitedRequires reloading for long-term therapy (not for AVVdelivery)Potentially effective only for focal diseases

Intranasal administration (nasal sprayor drops)

Non-invasiveSimplicity of deliverySome degree of targeting to CNS areasachievableMay terminate delivery easilyNo pleiotropic effects

Enzymatic degradation in nasal passages

Ocular administration (drops) Non-invasiveSimplicity of deliverySome degree of targeting to CNS areasachievableMay terminate delivery easily

Enzymatic degradation in ocular passagesPleiotropic effectsNeeds further characterization

Table 4Summary of the small molecules identified as NGF antagonists.

Small molecule type Target receptor Type of action Active concentration ReferenceAdenosine TrkA Transactivation 10µM [144]

PACAP TrkA and TrkB Transactivation 10 nM [145]Glucocorticoids TrkA, TrkB, TrkC Transactivation 0.1-1µM [126]

LM11A p75NTR Direct binding ≈ 1 nM [150,156,185]P92 TrkA Direct binding 250µM [250]

C-(92-96) TrkA Direct binding 10µM [186]D3 TrkA Direct binding 2µM [186]

L1L4 TrkA Direct binding 3µM [60]Gambogic amide TrkA Direct binding 0.5µM [124]

A new NGF-mimetic peptide was recently identi-fied, based on the analysis of NGF/TrkA binding re-gions [60]. The authors combined the NGF regions in-teracting with TrkA receptor, loop 1 and loop 4, in thesame molecule. This resulting peptide, termed L1L4,

showed a good NGF-like activityin vitro by inducingTrkA phosphorylationand differentiation of PC12 cellsand DRG [60]. However, this peptide was reported toreduce neuropathic pain in the Chronic Constriction In-jury animal model [60], a finding which is inconsistent


with the known actions of NGF in this model and inneuropathic pain [30,207,232].

A recent report identified gambogic amide as a se-lective agonist for TrkA [124]. Gambogic amide wasfound to selectively bind TrkA, trigger its tyrosine phos-phorylation, elicit PI3-kinase/Akt and MAPK activa-tion, provoke neurite outgrowth in PC12 cells and pre-vent neuronal cell death [124].

A different class of small molecules with NGF ago-nistic activity relies on the finding [144] that the TrkAreceptors can also be activated in the absence of neu-rotrophins, through transactivation by G-protein cou-pled receptor ligands. The first agonist molecule shownto activate TrkA in PC12 cells was shown to be adeno-sine, a neuromodulator that acts through G protein-coupled receptors. This finding raised the possibility toapproach the treatment of neurodegenerative diseasesby using adenosine-base small molecules in order totarget population of neurons that express both adeno-sine and Trk receptors [144]. Adenosine administra-tion showed a dose-dependent nociceptive effect thatshould be taken into consideration in clinical applica-tions [104].

In the same line, was the finding that pituitary adeny-late cyclase-activating polypeptide (PACAP) is alsoable to activate TrkA by transactivation. PACAP is aneuropeptide, acting through G protein-coupled recep-tors, and was shown to activate the Trk receptors and in-duce survival response in hippocampal cells [145]. TheTrk activation required a long time course. Regardingthe application of PACAP as a TrkA receptor activator,the possible side effects of the induction of neuropathicpain should be taken into account, although results onnociceptive effects of PACAP in animal models are stillcontroversial [9].

The findings that adenosine and PACAP are able toactivate TrkA receptors raises the possibility that thetransactivation of Trk receptors by G protein-coupledreceptors might be a general mechanism involved incell survival and synaptic plasticity [145], exploitablefor therapeutic purposes in AD and neurodegeneration.

Along the same lines is the recent finding that zincions mediate transactivation of the BDNF receptorTrkB, thereby potentiating hippocampal synaptic plas-ticity [120] (see below). Also, glucocorticoids, includ-ing dexamethasone, have been recently shown to selec-tively activate Trk receptors in neural cells [126], by amechanism that does not appear to involve G-protein-coupled receptors (GPCR), but rather a genomic ac-tion of glucocorticoids. Glucocorticoids are known toinduce the synthesis of neurotrophins (see below), but

the activation of Trk receptors by glucocorticoids wasshown not to depend on increased production of neu-rotrophins [126].

In conclusion, small molecule ligands with NGF ag-onist activity might lead to the development of neu-roprotective compounds for the treatment of neurode-generation, but this will require a new and full under-standing of their propertiesin vivo, with respect to thewealth of information that is available for NGF pro-tein itself. Moreover, some of the small molecule com-pounds have often a mixed agonistic and antagonisticprofile, making it difficult to predict their behaviorinvivo.

NGF synthesis inducers, NGF processing modulators,proNGF antagonists, other neurotrophins

Other strategies to increase the NGF neurotrophicsupport to the brain could, in principle, be pursued,intervening with low molecular weight compounds atdifferent points of the NGF metabolism cascade. Theseinclude NGF synthesis inducers, proNGF/NGF pro-cessing modulators,proNGF antagonists and other neu-rotrophins.

NGF synthesis inducers

The synthesis of NGF could be induced in the brainby several classes of compounds, each of which will, ofcourse, have different activity profiles on other targets,which needs to be considered for their pharmaceuticaldevelopment. In this review, we will report only datasupported byin vivo experiments.

Among drugs acting on neurotransmitter receptors,synthesis, uptake or release, clenbuterol, aβ2 adrenore-ceptor agonist, selectively induces the biosynthesis ofNGF in rat cerebral cortex but not in the hippocam-pus [90], while theα2-adrenoreceptor antagonist dex-efaroxan induces a 2-fold increase in NGF protein lev-els in both the cortex and nucleus basalis of Meyn-ert (NBM) in cortical devascularized rats [76]. Se-legiline, a MAO inhibitor, induces NGF gene expres-sion in cerebral cortex from intact rats and protectedrat and mouse cortical tissue from ischemic insult afterocclusion of the medial cerebral arthery [220]. Sin-gle, physiologically-activeand non-convulsivedoses ofthe three GABA(B) receptor antagonists CGP 36742,CGP 56433A and CGP 56999A increase NGF mR-NA levels by 200–400% and protein levels by 200–250% in rat neocortex, hippocampus as well as spinalcord [105]. The corticosteroid dexamethasone induces


a 100% NGF mRNA and protein expression versus con-trol in neonatal and young rats, but only a modest in-crease was found in aged rats [59], but in other systemsit was found to inhibit NGF expression [15]. The neu-ropeptide cholecystokinin-8, which promotes a varietyof neurobehavioral effects both in man and rodents, in-duces an increase in both NGF protein and mRNA inmouse cortex and hippocampus when intraperitoneal-ly injected at physiological doses and protects centralcholinergic neurons against fimbria-fornix lesion [233].Among hormones, thyroxine increases the concentra-tion of NGF mRNA in the septum and hippocampus inrats after an early postnatal treatment [153].

Interestingly, some drugs that are/were under clinicalevaluation for AD can induce NGF synthesis. The xan-thine derivative propentofylline, an inhibitor of adeno-sine uptake, was shown to increase NGF synthesis andrelease in cultured mouse astrocytes [221] and to re-cover cholinergic neuronal dysfunction induced by theinfusion of anti-NGF antibody into the rat septum afteroral administration [95]. The acetylcholinesterase in-hibitor Huperzine A, currently in Phase II clinical tri-al, increases NGF mRNA and protein levels after tran-sient cerebral ischemia [246]. Finally, xaliproden, anorally-active non-peptide compound, has been found toexhibit NGF effectsin vitro andin vivo, to increase thelifespan and delay the progression of the motor neurondegeneration in a mouse model of progressive motorneuronopathy [13].

While in principle it could be advantageous tohave low molecular weight compounds, ideally orallybioavailable, to increase neurotrophin concentrations(selectively) in the brain, the state of existing com-pounds of this nature is still rather immature. The taskof identifying activators of NGF transcription or expres-sion will require the development of adequate screen-ing methods, such as yeast artificial chromosome basedsystems [15], but the selectivity of the compounds andtheir pleiotropic effects will require case by casead hocin vivo characterization.

NGF processing modulators

As explained in the previous sections, proNGF islikely to be involved in cell death and neurodegenera-tion and to be a key player in AD. Moreover, the levelsof proNGF in the brain were found to be increased inthe brains of AD patients [84]. For this reason, in thedesign for new therapeutic approaches to the disease,one possible target would be proNGF itself. It is stillunknown whether the increased levels of proNGF in

the disease states are due to an increase in the NGFsynthesis or to a decrease in the protease activity of theproteases normally involved in its processing. There-fore, in designing new potential compounds for the dis-ease, one possible approach would be to increase thedegree of processing of accumulating proNGF, by theidentification of NGF processing modulators.

There are several classes of proteases known tocleave proNGF, both intracellularly and extracellular-ly. In particular, the pro-protein convertases are themost important class of proteases, being involved in thenormal processing of proNGF to mature NGF. Amongothers, the ones identified were furin, PACE4, PC5/6-B [216].

More recently a whole set of extracellular proteaseswere also involved in the cleavage of proNGF [147]:these are plasmin, MMP-7, MMP-3. Plasmin was alsoinvolved into the protease cascade of proNGF uponactivity-dependent release, together with MMP-9 [37].

The possibility to modulate the activity of these pro-teases by selective activators would be a chance to acton the proNGF levels in disease states. However, giventhe wide distribution of the proteases and their targets,the modulation of their activity would be totally non-selective and therefore a putative therapeutic interven-tion should be designed in order to tightly regulate theiractivity or to target specific cell populations, affectedin the disease.

proNGF antagonists: Neurotensin

Another possible target, in the framework of the in-vestigation towards the use of the neurotrophin systemin the therapeutic intervention for AD, would be to con-sider possible proNGF antagonists, able to compete forits binding to the specific receptor, sortilin. The natu-ral ligand for sortilin is the neuropeptide neurotensin,known to interact with three receptors: the first twoneurotensin receptors (NTR1 and NTR2) belong to thefamily of G-protein-coupled receptors (GPCRs), thethird one is sortilin [244]. Neurotensin has been re-ported to be involved in cellular functions in the pe-ripheral immune system [244], through the interactionwith the NTR1. However, the effects mediated by thebinding of neurotensin to sortilin are presently not fullyunderstood [159].

The potential use of neurotensin as an antagonist forproNGF resides on the findings that the peptide com-petes with proNGF for the binding to sortilin [176].Moreover, neurotensin was shown to rescue aged SCGand basal forebrain neuron from the cell-killing effects


of proNGFin vitro [2]. Consistently, alterations in theexpression pattern on neurotensin and its receptors inbrains of aged rats were reported [205], and it was sug-gested that pharmacological manipulation of the neu-rotensin system could improve cholinergic neurotrans-mission in the brain and thus modulate and/or restoredeficits in cognitive function [205]. A therapeutic ap-proach that inactivates sortilin on the cell surface oralters its expression pattern was also suggested [172].

It is clear that the therapeutic use of neurotensin-based compounds as an antagonist of proNGF actionsin neurodegeneration will depend both on further val-idation of the proNGF/sortilin target in neurodegener-ation and, on the other hand, on the development ofneurotensin analogs devoid of the unwanted actions ofneurotensin itself.

Other neurotrophins: BDNF

Other neurotrophic factors, most notably the neu-rotrophin BDNF and its specific receptor TrkB, are alsoimplicated in the AD process within the context of theNGF system. BDNF is well known to be a regulator ofsynaptic function and synaptic plasticity [130] as wellas of cholinergic neurons of the basal forebrain [212]and therefore its possible implication in AD is wellgrounded, also in light of AD being considered as aneuroplasticity disorder [161,218]. Interestingly, theanatomical progression of neurofibrillary pathology inAD is well correlated to the neuroanatomy of BDNFretrograde pathways, starting from the enthorhinal tosubiculum and CA1 pathway.

As NGF, BDNF exists both in pro and mature forms.The expression of both forms has been evaluated inmurine models of the disease and in AD patients.In mice expressing mutated forms of human AβPPand of PS-1 (Tg2576/PS-1P264L/+ and TG2576/PS-1P264L/P264L mice), BDNF gene expression is upreg-ulated following Aβ deposition [97]. At histologicallevel, a three-fold induction of BDNF was found in mi-croglial cells and astrocytes surrounding Aβ plaques inthe APP23 mice [38], suggesting a possible role in theregulation of inflammatory and axonal growth processaround the plaque. Finally, in these mice, the improve-ment of cognitive functions obtained after exposure toenvironmental enriched conditions was shown to cor-relate with the upregulation of several factors includingBDNF, despite a stable Aβ plaque load [249].

The presence of BDNF polymorphisms in AD pa-tients has been an issue of a long debate. An asso-ciation of Val66Met polymorphism to AD was found

by some authors [72,88,138,242], suggesting that BD-NF gene variants are significant risk factors for AD,whereas others failed to confirm it [31,61,77,235,243].

Despite the conclusive evidence for genetic polymor-phisms, alterations in both BDNF and TrkB expressionwere found in AD patients. The mature form of BDNFhas been found to be elevated in blood serum of earlystages of probable AD with respect to age-matched nondemented controls and more severe stages of AD [143].No alterations were reported in BDNF CSF levels ofAD patients [28,143].

A 40% reduction of pro-BDNF was observed in theparietal cortex of AD patients [162]. In AD hippocam-pus, temporal and parietal cortex, both BDNF mRNAand protein are also decreased [63,114,191]. Interest-ingly, neurons containing neurofibrillary tangles do notexpress BDNF, whereas neurons with no tangles ex-press it, suggesting a possible protective role againstneurofibrillary degeneration [171]. A detailed analysisof BDNF mRNA transcripts showed that transcripts 1to 3 are those showing a marked decrease in the pari-etal cortex of AD patients [85]. BDNF is also syn-thesized in the basal forebrain and a 50% reduction ofBDNF mRNA levels was found in the nucleus basalisof Meynert [85].

The alterations in the expression of TrkB receptorseem to be isoform specific. Indeed, a reduction of thecatalytic form, but not of the non-catalytic form, hasbeen shown in the frontal and temporal cortex of ADpatients [3]. Moreover, a reduction of the number ofNBM TrkB-positive neurons was found in the brain ofAD patients [210], with a greatest decrement in mildto moderate AD [97] and no changes of TrkB mRNAexpression in severe AD [32].

Overall, these data suggest that the delivery of ex-ogenous BDNF could in principle represent a possibletherapeutic approach for AD. However, the adminis-tration of BDNF is hindered by a series of problems(besides the delivery issue [91], for which the discus-sion above, on NGF delivery, may apply similarly).First of all, the relationship between Aβ and BDNFand TrkB expression and activities, or of deficits there-of, are far from being described and/or clarified. In-deed, while the exposure of SHSY5Y neuroblastomacells to Aβ1−42 determine an increase in BDNF andTrkB expression [178], other reports show that expo-sure to sublethal concentrations of the same peptide orto oligomeric amyloid decreases BDNF mRNA [94]and function [234].

Secondly, TrkB expression in the brain is much morewidespread than that of the NGF receptor TrkA, and


a generalized flooding BDNF in the brain might beundesiderable, and likely to cause increased neuronalexcitability and seizure development [140]. One ap-proach, worthwhile investigating, could be to resortto activators of BDNF expression, such as antidepres-sants, some of which (as Fluoxetine) have been shownrecently to restore cortical plasticity, via activation ofBDNF expression and reduction of GABAergic inhibi-tion [158].

FINAL CONCLUSIONS

NGF has received much attention as a potential ther-apeutic agent for AD, due to its action(s) on the cholin-ergic system of the basal forebrain. However, the initialhopes were dampened by the fact that a practical wayof delivering NGF to the brain in a safe, well toleratedand non-invasive manner has remained a considerablechallenge. The NGF-based clinical approaches, cur-rently ongoing, are still following invasive neurosurgi-cal paradigms that, while providing further validationof the therapeutic rationale, are unlikely to become asuitable and cheap treatment for millions of patients.The attention for NGF based therapies has gained a re-newed momentum by recent work that provides furthersignificant mechanistic elements to the role of NGF inAD neurodegeneration, and place NGF at the centerof a loop [the NGF/proNGF/AβPP loop (see Fig. 5)]that could play an important etiological role in the on-set and progression of the late onset, sporadic forms ofthe disease. It appears that different causes can lead toan altered NGF signaling, and in most of these cases(proNGF/NGF unbalance, reduced NGF availability,axonal transport deficits,etc.) enhancing the delivery ofNGF may help to alleviate such causes. Therefore, thenew rationale for an NGF based therapy would prospectNGF not only as a long-lasting neurotrophic choliner-gic therapy, but also as an anti-amyloidogenic therapyor as a therapy that would bypass transport deficits, byproviding NGF directly to the cell bodies. On the otherhand, altered receptor ratios might be best approachedby targeting them with receptor-specific agents.

On this basis, new approaches aimed at increasingthe (effectiveness of) NGF neurotrophic support to tar-get brain areas have a strong mechanistic basis andthe development of non-invasive drugs or treatmentsis highly needed. The development of the hNGF-61pipeline, optimized to meet the NGF therapeutic win-dow by non-invasive deliveries, has the potential toyield a new generation, non-invasive, NGF-based dis-ease modifying therapy.

ACKNOWLEDGMENTS

The experimental work from the respective authorslab was funded by Telethon grant n. GGP05234, Eu-ropean Commission project n. 037837 (MEMORIES),MIUR (FIRB projects n. RBIN 04H5AS and RBLA03F) and by Lay Line Genomics S.p.A. The authorsthank Drs Sonia Covaceuszach (Lay Line GenomicsS.p.A.), Federica Ferrero (SISSA) and Sabina Giannot-ta (SISSA) for skillful assistance in some of the exper-imental work presented in this review. The authors aregrateful to Mrs. Lucia De Caprio for careful editing ofthe text.

We are greatly indebted to Rita Levi-Montalcini forher enthusiasm towards science, towards NGF and to-wards life, for her inspiring questions and discussions,for her never ending curiosity and for her constant andenthusiastic support.

Drs. Cattaneo and Capsoni were previously em-ployed and/or own equity in Lay Line Genomics S.p.A.and are named inventors on patents related to the field.

References

[1] R. Al-Shawi, A. Hafner, S. Chun, S. Raza, K. Crutcher, C.Thrasivoulou, P. Simons and T. Cowen, ProNGF, sortilin,and age-related neurodegeneration,Ann NY Acad Sci 1119(2007), 208–215.

[2] R. Al-Shawi, A. Hafner, J. Olson, S. Chun, S. Raza, C.Thrasivoulou, S. Lovestone, R. Killick, P. Simons and T.Cowen, Neurotoxic and neurotrophic roles of proNGF andthe receptor sortilin in the adult and ageing nervous system,Eur J Neurosci 27 (2008), 2103–2114.

[3] S.J. Allen, G.K. Wilcock and D. Dawbarn, Profoundand selective loss of catalytic TrkB immunoreactivity inAlzheimer’s disease,Biochem Biophys Res Commun 264(1999), 648–651.

[4] L. Aloe, S.D. Skaper, A. Leon and R. Levi-Montalcini, Nervegrowth factor and autoimmune diseases,Autoimmunity 19(1994), 141–150.

[5] L. Aloe, E. Alleva and M. Fiore, Stress and nerve growthfactor: findings in animal models and humans,PharmacolBiochem Behav 73 (2002), 159–166.

[6] A.D. Alonso, I. Grundke-Iqbal, H.S. Barra and K. Iqbal,Abnormal phosphorylation of tau and the mechanism ofAlzheimer neurofibrillary degeneration: sequestration ofmicrotubule-associated proteins 1 and 2 and the disassemblyof microtubules by the abnormal tau,Proc Natl Acad Sci USA94 (1997), 298–303.

[7] A.D. Alonso, T. Zaidi, M. Novak, H.S. Barra, I. Grundke-Iqbal and K. Iqbal, Interaction of tau isoforms withAlzheimer’s disease abnormally hyperphosphorylated tauand in vitro phosphorylation into the disease-like protein,JBiol Chem 276 (2001), 37967–37973.

[8] H.H. Althaus and S. Kloppner, Mature pig oligodendrocytesrapidly process human recombinant pro-nerve growth factorand do not undergo cell death,J Neurochem 98 (2006), 506–517.


[9] F. Amaya, G. Shimosato, M. Nagano, M. Ueda, S. Hashimo-to, Y. Tanaka, H. Suzuki and M. Tanaka, NGF and GDNFdifferentially regulate TRPV1 expression that contributes todevelopment of inflammatory thermal hyperalgesia,Eur JNeurosci 20 (2004), 2303–2310.

[10] O.M. Andersen, V. Schmidt, R. Spoelgen, J. Gliemann, J.Behlke, D. Galatis, W.J. McKinstry, M.W. Parker, C.L. Mas-ters, B.T. Hyman, R. Cappai and T.E. Willnow, Moleculardissection of the interaction between amyloid precursor pro-tein and its neuronal trafficking receptor SorLA/LR11,Bio-chemistry 45 (2006), 2618–2628.

[11] S.C. Apfel, Neurotrophic factors in the therapy of diabeticneuropathy,Am J Med 107 (1999), 34S–42S.

[12] S.H. Appel, A unifying hypothesis for the cause of amy-otrophic lateral sclerosis, parkinsonism, and Alzheimer dis-ease,Ann Neurol 10 (1981), 499–505.

[13] A. Appert-Collin, F.H. Duong, P. Passilly-Degrace, J.P. Gies,J.M. Warter and P. Poindron, Quantification of neurotrophinmRNA expression in PMN mouse: modulation by xalipro-den,Int J Immunopathol Pharmacol 17 (2004), 157–164.

[14] I. Arisi, A. Cattaneo and V. Rosato, Parameter estimate of sig-nal transduction pathways,BMC Neurosci 7(Suppl 1) (2006),S6.

[15] F.A. Asselbergs, R. Grossenbacher, R. Ortmann, B. Henger-er, G.K. McMaster, E. Sutter, R. Widmer and F. Buxton,Position-independent expression of a human nerve growthfactor-luciferase reporter gene cloned on a yeast artificialchromosome vector,Nucleic Acids Res 26 (1998), 1826–1833.

[16] C. Backman, G.M. Rose, B.J. Hoffer, M.A. Henry, R.T. Bar-tus, P. Friden and A.C. Granholm, Systemic administrationof a nerve growth factor conjugate reverses age-related cog-nitive dysfunction and prevents cholinergic neuron atrophy,J Neurosci 16 (1996), 5437–5442.

[17] Y.A. Barde, Death of injured neurons caused by the precursorof nerve growth factor,Proc Natl Acad Sci USA 101 (2004),5703–5704.

[18] P.A. Barker, p75NTR is positively promiscuous: novel part-ners and new insights,Neuron 42 (2004), 529–533.

[19] R.T. Bartus, R.L. Dean, 3rd, B. Beer and A.S. Lippa, Thecholinergic hypothesis of geriatric memory dysfunction,Sci-ence 217 (1982), 408–414.

[20] M.S. Beattie, A.W. Harrington, R. Lee, J.Y. Kim, S.L. Boyce,F.M. Longo, J.C. Bresnahan, B.L. Hempstead and S.O. Yoon,ProNGF induces p75-mediated death of oligodendrocytesfollowing spinal cord injury,Neuron 36 (2002), 375–386.

[21] N. Beglova, L. LeSauteur, I. Ekiel, H.U. Saragovi and K.Gehring, Solution structure and internal motion of a bioactivepeptide derived from nerve growth factor,J Biol Chem 273(1998), 23652–23658.

[22] N. Beglova, S. Maliartchouk, I. Ekiel, M.C. Zaccaro, H.U.Saragovi and K. Gehring, Design and solution structure offunctional peptide mimetics of nerve growth factor,J MedChem 43 (2000), 3530–3540.

[23] D.L. Bennett, Neurotrophic factors: important regulators ofnociceptive function,Neuroscientist 7 (2001), 13–17.

[24] N. Berardi, C. Braschi, S. Capsoni, A. Cattaneo and L. Maf-fei, Environmental enrichment delays the onset of memorydeficits and reduces neuropathological hallmarks in a mousemodel of Alzheimer-like neurodegeneration,J AlzheimersDis 11 (2007), 359–370.

[25] H.A. Bimonte-Nelson, R.S. Singleton, M.E. Nelson, C.B.Eckman, J. Barber, T.Y. Scott and A.C. Granholm, Testos-terone, but not nonaromatizable dihydrotestosterone, im-

proves working memory and alters nerve growth factor levelsin aged male rats,Exp Neurol 181 (2003), 301–312.

[26] S. Biocca, A. Cattaneo and P. Calissano, A macromolec-ular structure favouring microtubule assembly in NGF-differentiated pheochromocytoma cells (PC12),Embo J 2(1983), 643–648.

[27] K.M. Bishop, E.K. Hofer, A. Mehta, A. Ramirez, L. Sun, M.Tuszynski and R.T. Bartus, Therapeutic potential of CERE-110 (AAV2-NGF): Targeted, stable, and sustained NGF de-livery and trophic activity on rodent basal forebrain cholin-ergic neurons,Exp Neurol (2008),

[28] I. Blasko, W. Lederer, H. Oberbauer, T. Walch, G. Kemm-ler, H. Hinterhuber, J. Marksteiner and C. Humpel, Measure-ment of thirteen biological markers in CSF of patients withAlzheimer’s disease and other dementias,Dement GeriatrCogn Disord 21 (2006), 9–15.

[29] A. Blesch, R.J. Grill and M.H. Tuszynski, Neurotrophin genetherapy in CNS models of trauma and degeneration,ProgBrain Res 117 (1998), 473–484.

[30] M.M. Blurton-Jones, J.A. Roberts and M.H. Tuszynski, Es-trogen receptor immunoreactivity in the adult primate brain:neuronal distribution and association with p75, trkA, andcholine acetyltransferase,J Comp Neurol 405 (1999), 529–542.

[31] S.M. Bodner, W. Berrettini, V. van Deerlin, D.A. Bennett,R.S. Wilson, J.Q. Trojanowski and S.E. Arnold, Geneticvariation in the brain derived neurotrophic factor gene inAlzheimer’s disease,Am J Med Genet B NeuropsychiatrGenet 134 (2005), 1–5.

[32] F. Boissiere, B. Faucheux, Y. Agid and E.C. Hirsch, Ex-pression of catalytic trkB gene in the striatum and the basalforebrain of patients with Alzheimer’s disease: an in situhybridization study,Neurosci Lett 221 (1997), 141–144.

[33] W. Bondareff, C. Harrington, C.M. Wischik, D.L. Hauserand M. Roth, Immunohistochemical staging of neurofibril-lary degeneration in Alzheimer’s disease,J Neuropathol ExpNeurol 53 (1994), 158–164.

[34] J. Boutilier, C. Ceni, P.C. Pagdala, A. Forgie, K.E. Neetand P.A. Barker, Proneurotrophins Require Endocytosis andIntracellular Proteolysis to Induce TrkA Activation,J BiolChem 283 (2008), 12709–12716.

[35] F.C. Bronfman and M. Fainzilber, Multi-tasking by the p75neurotrophin receptor: sortilin things out?,EMBO Rep 5(2004), 867–871.

[36] M.A. Bruno, P.B. Clarke, A. Seltzer, R. Quirion, K. Burgess,A.C. Cuello and H.U. Saragovi, Long-lasting rescue of age-associated deficits in cognition and the CNS cholinergic phe-notype by a partial agonist peptidomimetic ligand of TrkA,JNeurosci 24 (2004), 8009–8018.

[37] M.A. Bruno and A.C. Cuello, Activity-dependent release ofprecursor nerve growth factor, conversion to mature nervegrowth factor, and its degradation by a protease cascade,ProcNatl Acad Sci USA 103 (2006), 6735–6740.

[38] G.J. Burbach, R. Hellweg, C.A. Haas, D. Del Turco, U.Deicke, D. Abramowski, M. Jucker, M. Staufenbiel andT. Deller, Induction of brain-derived neurotrophic factor inplaque-associated glial cells of aged APP23 transgenic mice,J Neurosci 24 (2004), 2421–2430.

[39] S. Capsoni, F. Ruberti, E. Di Daniel and A. Cattaneo, Mus-cular dystrophy in adult and aged anti-NGF transgenic miceresembles an inclusion body myopathy,J Neurosci Res 59(2000), 553–560.

[40] S. Capsoni, G. Ugolini, A. Comparini, F. Ruberti, N. Berardiand A. Cattaneo, Alzheimer-like neurodegeneration in aged


antinerve growth factor transgenic mice,Proc Natl Acad SciUSA 97 (2000), 6826–6831.

[41] S. Capsoni, S. Giannotta and A. Cattaneo, Nerve growth fac-tor and galantamine ameliorate early signs of neurodegener-ation in anti-nerve growth factor mice,Proc Natl Acad SciUSA 99 (2002), 12432–12437.

[42] S. Capsoni, S. Giannotta and A. Cattaneo, Beta-amyloidplaques in a model for sporadic Alzheimer’s disease basedon transgenic anti-nerve growth factor antibodies,Mol CellNeurosci 21 (2002), 15–28.

[43] S. Capsoni, E. Margotti, B. Stefanini and A. Cattaneo, Theeffects of knocking out of p75NTR on the neurodegenera-tive phenotype in AD11 anti-NGF mice, in:Abstract Book,7th International Conference AD/PD 2005, Sorrento, ItalyMarch 9–13, 2005.

[44] S. Capsoni and A. Cattaneo, On the molecular basis linkingNerve Growth Factor (NGF) to Alzheimer’s disease,CellMol Neurobiol 26 (2006), 619–633.

[45] S. Capsoni, S. Covaceuszach, S. Marinelli, G. Ugolini, F.Spirito, D. Vignone, G. Amato, F.Pavone and A. Cattaneo,Design and characterization of human Nerve Growth Fac-tor muteins with reduced pro-nociceptive properties, in:Ab-stract Book, 11th International Conference on Alzheimer’sDisease and Related Disorders, Chicago, IL July 26–31,2008.

[46] S. Capsoni, S. Covaceuszach, G. Ugolini, F. Spirito, D. Vi-gnone, B. Stefanini, G. Amato and A. Cattaneo, Deliveryof NGF to the brain: intranasal versus ocular adminsitrationin transgenic mice, in:Abstract book, 11th InternationalConference on Alzheimer’s Disease and Related Disorders,Chicago, USA July 26–31, 2008.

[47] S. Capsoni, S. Giannotta and A. Cattaneo, Early events inAlzheimer-like neurodegeneration in anti-nerve growth fac-tor transgenic mice,Brain Aging (2002), 24–43.

[48] A. Cattaneo, B. Rapposelli and P. Calissano, Three distincttypes of monoclonal antibodies after long-term immunizationof rats with mouse nerve growth factor,J Neurochem 50(1988), 1003–1010.

[49] A. Cattaneo, NGF processing and Alzheimer’s disease neu-rodegeneration: too-little and too-much pro-NGF, in:Ab-stract Book, 10th International Conference on Alzheimer’sDisease and Related Disorders, Madrid, Spain July 15–20,2006.

[50] M.V. Chao and M. Bothwell, Neurotrophins: to cleave or notto cleave,Neuron 33 (2002), 9–12.

[51] M.V. Chao, R. Rajagopal and F.S. Lee, Neurotrophin sig-nalling in health and disease,Clin Sci (Lond) 110 (2006),167–173.

[52] V. Charles, E.J. Mufson, P.M. Friden, R.T. Bartus and J.H.Kordower, Atrophy of cholinergic basal forebrain neuronsfollowing excitotoxic cortical lesions is reversed by intra-venous administration of an NGF conjugate,Brain Res 728(1996), 193–203.

[53] K.S. Chen and F.H. Gage, Somatic gene transfer of NGF tothe aged brain: behavioral and morphological amelioration,J Neurosci 15 (1995), 2819–2825.

[54] X.Q. Chen, J.R. Fawcett, Y.E. Rahman, T.A. Ala and W.H.2nd Frey, Delivery of Nerve Growth Factor to the Brain viathe Olfactory Pathway,J Alzheimers Dis 1 (1998), 35–44.

[55] Z.Y. Chen, P.D. Patel, G. Sant, C.X. Meng, K.K. Teng, B.L.Hempstead and F.S. Lee, Variant brain-derived neurotroph-ic factor (BDNF) (Met66) alters the intracellular traffickingand activity-dependent secretion of wild-type BDNF in neu-

rosecretory cells and cortical neurons,J Neurosci 24 (2004),4401–4411.

[56] Z.Y. Chen, A. Ieraci, H. Teng, H. Dall, C.X. Meng, D.G.Herrera, A. Nykjaer, B.L. Hempstead and F.S. Lee, Sortil-in controls intracellular sorting of brain-derived neurotroph-ic factor to the regulated secretory pathway,J Neurosci 25(2005), 6156–6166.

[57] Y. Chu, E.J. Cochran, D.A. Bennett, E.J. Mufson and J.H.Kordower, Down-regulation of trkA mRNA within nucleusbasalis neurons in individuals with mild cognitive impairmentand Alzheimer’s disease,J Comp Neurol 437 (2001), 296–307.

[58] Clinicaltrial.gov, CERE-110 in Subjects With Mild to Mod-erate Alzheimer’s Disease, http://www.clinicaltrials.gov/ct2/show/NCT00087789?term=NGF&rank=4,Last UpdatedMay 27, 2008, Accessed on May 29, 2008.

[59] A.M. Colangelo, P. Follesa and I. Mocchetti, Differentialinduction of nerve growth factor and basic fibroblast growthfactor mRNA in neonatal and aged rat brain,Brain Res MolBrain Res 53 (1998), 218–225.

[60] A.M. Colangelo, M.R. Bianco, L. Vitagliano, C. Cavaliere,G. Cirillo, L. De Gioia, D. Diana, D. Colombo, C. Redaelli,L. Zaccaro, G. Morelli, M. Papa, P. Sarmientos, L. Alberghi-na and E. Martegani, A new nerve growth factor-mimeticpeptide active on neuropathic pain in rats,J Neurosci 28(2008), 2698–2709.

[61] O. Combarros, J. Infante, J. Llorca and J. Berciano, Poly-morphism at codon 66 of the brain-derived neurotrophic fac-tor gene is not associated with sporadic Alzheimer’s disease,Dement Geriatr Cogn Disord 18 (2004), 55–58.

[62] J.M. Conner, M.A. Darracq, J. Roberts and M.H. Tuszyn-ski, Nontropic actions of neurotrophins: subcortical nervegrowth factor gene delivery reverses age-related degenera-tion of primate cortical cholinergic innervation,Proc NatlAcad Sci USA 98 (2001), 1941–1946.

[63] B. Connor, D. Young, Q. Yan, R.L. Faull, B. Synek and M.Dragunow, Brain-derived neurotrophic factor is reduced inAlzheimer’s disease,Brain Res Mol Brain Res 49 (1997),71–81.

[64] J.D. Cooper, A. Salehi, J.D. Delcroix, C.L. Howe, P.V. Be-lichenko, J. Chua-Couzens, J.F. Kilbridge, E.J. Carlson, C.J.Epstein and W.C. Mobley, Failed retrograde transport of NGFin a mouse model of Down’s syndrome: reversal of choliner-gic neurodegenerative phenotypes following NGF infusion,Proc Natl Acad Sci USA 98 (2001), 10439–10444.

[65] V. Corsetti, G. Amadoro, A. Gentile, S. Capsoni, M.T. Ciot-ti, M.T. Cencioni, A. Atlante, N. Canu, T.T. Rohn, A. Cat-taneo and P. Calissano, Identification of a caspase-derivedN-terminal tau fragment in cellular and animal Alzheimer’sdisease models,Mol Cell Neurosci 38 (2008), 381–392.

[66] C. Costantini, H. Scrable and L. Puglielli, An aging pathwaycontrols the TrkA to p75NTR receptor switch and amyloidbeta-peptide generation,Embo J 25 (2006), 1997–2006.

[67] E.J. Coulson, Does the p75 neurotrophin receptor medi-ate Abeta-induced toxicity in Alzheimer’s disease?,J Neu-rochem 98 (2006), 654–660.

[68] S.E. Counts, M. Nadeem, J. Wuu, S.D. Ginsberg, H.U.Saragovi and E.J. Mufson, Reduction of cortical TrkA but notp75(NTR) protein in early-stage Alzheimer’s disease,AnnNeurol 56 (2004), 520–531.

[69] S.E. Counts and E.J. Mufson, The role of nerve growth fac-tor receptors in cholinergic basal forebrain degeneration inprodromal Alzheimer disease,J Neuropathol Exp Neurol 64(2005), 263–272.


[70] S. Covaceuszach, S. Capsoni, G. Ugolini, F. Spirito, D. Vi-gnone and A. Cattaneo, Development of a non-invasive NGF-based therapy for Alzheimer’s disease,Current AlzheimerResearch, In press, 2008.

[71] S. Covaceuszach, A. Cassetta, P.V. Konarev, S. Gonfloni, R.Rudolph, D.I. Svergun, D. Lamba and A. Cattaneo, Dissect-ing NGF interaction with TrkA and p75 receptors by struc-tural and functional studies of an anti-NGF neutralizing an-tibody,J Mol Biol, in press, 2008.

[72] C. Crowley, S.D. Spencer, M.C. Nishimura, K.S. Chen, S.Pitts-Meek, M.P. Armanini, L.H. Ling, S.B. McMahon, D.L.Shelton, A.D. Levinson and et al., Mice lacking nerve growthfactor display perinatal loss of sensory and sympathetic neu-rons yet develop basal forebrain cholinergic neurons,Cell 76(1994), 1001–1011.

[73] P. Davies and A.J. Maloney, Selective loss of central cholin-ergic neurons in Alzheimer’s disease,Lancet 2 (1976), 1403.

[74] R. De Rosa, A.A. Garcia, C. Braschi, S. Capsoni, L. Maf-fei, N. Berardi and A. Cattaneo, Intranasal administrationof nerve growth factor (NGF) rescues recognition memorydeficits in AD11 anti-NGF transgenic mice,Proc Natl AcadSci USA 102 (2005), 3811–3816.

[75] T. Debeir, H.U. Saragovi and A.C. Cuello, A nerve growthfactor mimetic TrkA antagonist causes withdrawal of corticalcholinergic boutons in the adult rat,Proc Natl Acad Sci USA96 (1999), 4067–4072.

[76] T. Debeir, M. Marien, J. Ferrario, P. Rizk, A. Prigent, F.Colpaert and R. Raisman-Vozari,In vivo upregulation of en-dogenous NGF in the rat brain by the alpha2-adrenoreceptorantagonist dexefaroxan: potential role in the protection of thebasalocortical cholinergic system during neurodegeneration,Exp Neurol 190 (2004), 384–395.

[77] P. Desai, R. Nebes, S.T. DeKosky and M.I. Kamboh, Investi-gation of the effect of brain-derived neurotrophic factor (BD-NF) polymorphisms on the risk of late-onset Alzheimer’sdisease (AD) and quantitative measures of AD progression,Neurosci Lett 379 (2005), 229–234.

[78] V. Di Fausto, M. Fiore, P. Tirassa, A. Lambiase and L.Aloe, Eye drop NGF administration promotes the recoveryof chemically injured cholinergic neurons of adult mouseforebrain,Eur J Neurosci 26 (2007), 2473–2480.

[79] M. Domeniconi, B.L. Hempstead and M.V. Chao, Pro-NGFsecreted by astrocytes promotes motor neuron cell death,MolCell Neurosci 34 (2007), 271–279.

[80] D.G. Drubin, S.C. Feinstein, E.M. Shooter and M.W.Kirschner, Nerve growth factor-induced neurite outgrowth inPC12 cells involves the coordinate induction of microtubuleassembly and assembly-promoting factors,J Cell Biol 101(1985), 1799–1807.

[81] M. Eriksdotter Jonhagen, A. Nordberg, K. Amberla, L. Back-man, T. Ebendal, B. Meyerson, L. Olson, Seiger, M. Shigeta,E. Theodorsson, M. Viitanen, B. Winblad and L.O. Wahlund,Intracerebroventricular infusion of nerve growth factor inthree patients with Alzheimer’s disease,Dement GeriatrCogn Disord 9 (1998), 246–257.

[82] A.M. Fagan, M. Garber, M. Barbacid, I. Silos-Santiago andD.M. Holtzman, A role for TrkA during maturation of striataland basal forebrain cholinergic neuronsin vivo, J Neurosci17 (1997), 7644–7654.

[83] M. Fahnestock, S.A. Scott, N. Jette, J.A. Weingartner andK.A. Crutcher, Nerve growth factor mRNA and protein levelsmeasured in the same tissue from normal and Alzheimer’sdisease parietal cortex,Brain Res Mol Brain Res 42 (1996),175–178.

[84] M. Fahnestock, B. Michalski, B. Xu and M.D. Coughlin, Theprecursor pro-nerve growth factor is the predominant form ofnerve growth factor in brain and is increased in Alzheimer’sdisease,Mol Cell Neurosci 18 (2001), 210–220.

[85] M. Fahnestock, D. Garzon, R.M. Holsinger and B. Michal-ski, Neurotrophic factors and Alzheimer’s disease: are wefocusing on the wrong molecule?,J Neural Transm Suppl(2002), 241–252.

[86] M. Fahnestock, G. Yu and M.D. Coughlin, ProNGF: a neu-rotrophic or an apoptotic molecule,Prog Brain Res 146(2004), 101–110.

[87] M. Fahnestock, G. Yu, B. Michalski, S. Mathew, A.Colquhoun, G.M. Ross and M.D. Coughlin, The nervegrowth factor precursor proNGF exhibits neurotrophic ac-tivity but is less active than mature nerve growth factor,JNeurochem 89 (2004), 581–592.

[88] L. Fasulo, G. Ugolini and A. Cattaneo, Apoptotic effect ofcaspase-3 cleaved tau in hippocampal neurons and its poten-tiation by tau FTDP-mutation N279K,J Alzheimers Dis 7(2005), 3–13.

[89] W. Fischer, K. Wictorin, A. Bjorklund, L.R. Williams, S.Varon and F.H. Gage, Amelioration of cholinergic neuronatrophy and spatial memory impairment in aged rats by nervegrowth factor,Nature 329 (1987), 65–68.

[90] P. Follesa and I. Mocchetti, Regulation of basic fibrob-last growth factor and nerve growth factor mRNA by beta-adrenergic receptor activation and adrenal steroids in rat cen-tral nervous system,Mol Pharmacol 43 (1993), 132–138.

[91] W.H. Frey, J. Liu, X.Q. Chen, R.G. Thorne, J.R. Fawcett,T.A. Ala and Y.E. Rahman, Delivery of125 I-NGF to thebrain via the olfactory route,Drug Delivery 4 (1997), 87–92.

[92] P.M. Friden, L.R. Walus, G.F. Musso, M.A. Taylor, B. Mal-froy and R.M. Starzyk, Anti-transferrin receptor antibodyand antibody-drug conjugates cross the blood-brain barrier,Proc Natl Acad Sci USA 88 (1991), 4771–4775.

[93] P.M. Friden, L.R. Walus, P. Watson, S.R. Doctrow, J.W.Kozarich, C. Backman, H. Bergman, B. Hoffer, F. Bloom andA.C. Granholm, Blood-brain barrier penetration andin vivoactivity of an NGF conjugate,Science 259 (1993), 373–377.

[94] D.J. Garzon and M. Fahnestock, Oligomeric amyloid de-creases basal levels of brain-derived neurotrophic factor (BD-NF) mRNA via specific downregulation of BDNF transcriptsIV and V in differentiated human neuroblastoma cells,JNeurosci 27 (2007), 2628–2635.

[95] C. Ghelardini, N. Galeotti, A. Bartolini, S. Furukawa, A.Nitta, D. Manetti and F. Gualtieri, Memory facilitation andstimulation of endogenous nerve growth factor synthesis bythe acetylcholine releaser PG-9,Jpn J Pharmacol 78 (1998),245–251.

[96] E. Giacobini, Cholinesterase in human brain: the effect ofcholinesterase inhibitors on Alzheimer’s disease and relateddisorder’s., in:The Brain Cholinergic System in Health andDisease, E. Giacobini and G. Pepeu, eds, Informa Healthcare,2006, pp. 235–264.

[97] S.D. Ginsberg, S. Che, J. Wuu, S.E. Counts and E.J. Muf-son, Down regulation of trk but not p75NTR gene expres-sion in single cholinergic basal forebrain neurons mark theprogression of Alzheimer’s disease,J Neurochem 97 (2006),475–487.

[98] H. Gnahn, F. Hefti, R. Heumann, M.E. Schwab and H. Thoe-nen, NGF-mediated increase of choline acetyltransferase(ChAT) in the neonatal rat forebrain: evidence for a phys-iological role of NGF in the brain,Brain Res 285 (1983),45–52.


[99] M. Goedert, A. Fine, S.P. Hunt and A. Ullrich, Nerve growthfactor mRNA in peripheral and central rat tissues and in thehuman central nervous system: lesion effects in the rat brainand levels in Alzheimer’s disease,Brain Res 387 (1986),85–92.

[100] I. Gozes, B.H. Morimoto, J. Tiong, A. Fox, K. Sutherland,D. Dangoor, M. Holser-Cochav, K. Vered, P. Newton, P.S.Aisen, Y. Matsuoka, C.H. van Dyck and L. Thal, NAP: re-search and development of a peptide derived from activity-dependent neuroprotective protein (ADNP),CNS Drug Rev11 (2005), 353–368.

[101] S. Gunawardena and L.S. Goldstein, Disruption of axonaltransport and neuronal viability by amyloid precursor proteinmutations in Drosophila,Neuron 32 (2001), 389–401.

[102] A.W. Harrington, B. Leiner, C. Blechschmitt, J.C. Arevalo,R. Lee, K. Morl, M. Meyer, B.L. Hempstead, S.O. Yoon andK.M. Giehl, Secreted proNGF is a pathophysiological death-inducing ligand after adult CNS injury,Proc Natl Acad SciUSA 101 (2004), 6226–6230.

[103] A. Hauburger, M. Kliemannel, P. Madsen, R. Rudolph andE. Schwarz, Oxidative folding of nerve growth factor can bemediated by the pro-peptide of neurotrophin-3,FEBS Lett581 (2007), 4159–4164.

[104] M. Hayashida, K. Fukuda and A. Fukunaga, Clinical appli-cation of adenosine and ATP for pain control,J Anesth 19(2005), 225–235.

[105] K. Heese, U. Otten, P. Mathivet, M. Raiteri, C. Marescauxand R. Bernasconi, GABA(B) receptor antagonists elevateboth mRNA and protein levels of the neurotrophins nervegrowth factor (NGF) and brain-derived neurotrophic factor(BDNF) but not neurotrophin-3 (NT-3) in brain and spinalcord of rats,Neuropharmacology 39 (2000), 449–462.

[106] K. Heese, N. Inoue, Y. Nagai and T. Sawada, APP, NGF &the ’Sunday-driver’ in a Trolley on the Road,Restor NeurolNeurosci 22 (2004), 131–136.

[107] K. Heese, J.W. Low and N. Inoue, Nerve growth factor, neuralstem cells and Alzheimer’s disease,Neurosignals 15 (2006),1–12.

[108] F. Hefti, A. Dravid and J. Hartikka, Chronic intraventric-ular injections of nerve growth factor elevate hippocampalcholine acetyltransferase activity in adult rats with partialsepto-hippocampal lesions,Brain Res 293 (1984), 305–311.

[109] F. Hefti, Nerve growth factor promotes survival of septalcholinergic neurons after fimbrial transections,J Neurosci 6(1986), 2155–2162.

[110] F. Hefti and W.J. Weiner, Nerve growth factor andAlzheimer’s disease,Ann Neurol 20 (1986), 275–281.

[111] B.L. Hempstead, The many faces of p75NTR,Curr OpinNeurobiol 12 (2002), 260–267.

[112] B.L. Hempstead, Dissecting the diverse actions of pro- andmature neurotrophins,Curr Alzheimer Res 3 (2006), 19–24.

[113] P. Hintz-Obertreis, D. Krumwieh and F.R. Seiler, Devel-opment of a rapid, highly sensitive, non-radioactive assaysystem for hematopoietic growth factors,Behring Inst Mitt(1991), 99–103.

[114] C. Hock, K. Heese, C. Hulette, C. Rosenberg and U. Otten,Region-specific neurotrophin imbalances in Alzheimer dis-ease: decreased levels of brain-derived neurotrophic factorand increased levels of nerve growth factor in hippocampusand cortical areas,Arch Neurol 57 (2000), 846–851.

[115] D. Hoffman, L. Wahlberg and P. Aebischer, NGF releasedfrom a polymer matrix prevents loss of ChAT expressionin basal forebrain neurons following a fimbria-fornix lesion,Exp Neurol 110 (1990), 39–44.

[116] D. Hoffman, X.O. Breakefield, M.P. Short and P. Aebischer,Transplantation of a polymer-encapsulated cell line genet-ically engineered to release NGF,Exp Neurol 122 (1993),100–106.

[117] D.M. Holtzman, Y. Li, L.F. Parada, S. Kinsman, C.K. Chen,J.S. Valletta, J. Zhou, J.B. Long and W.C. Mobley, p140trkmRNA marks NGF-responsive forebrain neurons: evidencethat trk gene expression is induced by NGF,Neuron 9 (1992),465–478.

[118] D.M. Holtzman and W.C. Mobley, Neurotrophic factors andneurologic disease,West J Med 161 (1994), 246–254.

[119] P.M. Horowitz, K.R. Patterson, A.L. Guillozet-Bongaarts,M.R. Reynolds, C.A. Carroll, S.T. Weintraub, D.A. Bennett,V.L. Cryns, R.W. Berry and L.I. Binder, Early N-terminalchanges and caspase-6 cleavage of tau in Alzheimer’s dis-ease,J Neurosci 24 (2004), 7895–7902.

[120] Y.Z. Huang, E. Pan, Z.Q. Xiong and J.O. McNamara, Zinc-mediated transactivation of TrkB potentiates the hippocampalmossy fiber-CA3 pyramid synapse,Neuron 57 (2008), 546–558.

[121] C. Humpel, T. Ebendal, Y. Cao and L. Olson, Pentylenetetra-zol seizures increase pro-nerve growth factor-like immunore-activity in the reticular thalamic nucleus and nerve growthfactor mRNA in the dentate gyrus,J Neurosci Res 35 (1993),419–427.

[122] C.F. Ibanez, Jekyll-Hyde neurotrophins: the story ofproNGF,Trends Neurosci 25 (2002), 284–286.

[123] L.G. Isaacson, B.N. Saffran and K.A. Crutcher, IntracerebralNGF infusion induces hyperinnervation of cerebral bloodvessels,Neurobiol Aging 11 (1990), 51–55.

[124] S.W. Jang, M. Okada, I. Sayeed, G. Xiao, D. Stein, P. Jinand K. Ye, Gambogic amide, a selective agonist for TrkAreceptor that possesses robust neurotrophic activity, preventsneuronal cell death,Proc Natl Acad Sci USA 104 (2007),16329–16334.

[125] P. Jansen, K. Giehl, J.R. Nyengaard, K. Teng, O. Lioubinski,S.S. Sjoegaard, T. Breiderhoff, M. Gotthardt, F. Lin, A. Eil-ers, C.M. Petersen, G.R. Lewin, B.L. Hempstead, T.E. Will-now and A. Nykjaer, Roles for the pro-neurotrophin receptorsortilin in neuronal development, aging and brain injury,NatNeurosci 10 (2007), 1449–1457.

[126] F. Jeanneteau, M.J. Garabedian and M.V. Chao, Activationof Trk neurotrophin receptors by glucocorticoids provides aneuroprotective effect,Proc Natl Acad Sci USA 105 (2008),4862–4867.

[127] N. Jette, M.S. Cole and M. Fahnestock, NGF mRNA is notdecreased in frontal cortex from Alzheimer’s disease patients,Brain Res Mol Brain Res 25 (1994), 242–250.

[128] A. Kamal, G.B. Stokin, Z. Yang, C.H. Xia and L.S. Goldstein,Axonal transport of amyloid precursor protein is mediated bydirect binding to the kinesin light chain subunit of kinesin-I,Neuron 28 (2000), 449–459.

[129] A. Kamal, A. Almenar-Queralt, J.F. LeBlanc, E.A. Robertsand L.S. Goldstein, Kinesin-mediated axonal transport ofa membrane compartment containing beta-secretase andpresenilin-1 requires APP,Nature 414 (2001), 643–648.

[130] H.J. Kang and E.M. Schuman, Neurotrophin-induced mod-ulation of synaptic transmission in the adult hippocampus,JPhysiol Paris 89 (1995), 11–22.

[131] M. Kliemannel, A. Rattenholl, R. Golbik, J. Balbach, H.Lilie, R. Rudolph and E. Schwarz, The mature part ofproNGF induces the structure of its pro-peptide,FEBS Lett566 (2004), 207–212.


[132] M. Kliemannel, R. Golbik, R. Rudolph, E. Schwarz and H.Lilie, The pro-peptide of proNGF: structure formation andintramolecular association with NGF,Protein Sci 16 (2007),411–419.

[133] V.E. Koliatsos, H.J. Nauta, R.E. Clatterbuck, D.M. Holtz-man, W.C. Mobley and D.L. Price, Mouse nerve growthfactor prevents degeneration of axotomized basal forebraincholinergic neurons in the monkey,J Neurosci 10 (1990),3801–3813.

[134] J.H. Kordower, V. Charles, R. Bayer, R.T. Bartus, S. Putney,L.R. Walus and P.M. Friden, Intravenous administration of atransferrin receptor antibody-nerve growth factor conjugateprevents the degeneration of cholinergic striatal neurons ina model of Huntington disease,Proc Natl Acad Sci USA 91(1994), 9077–9080.

[135] J.H. Kordower, S.R. Winn, Y.T. Liu, E.J. Mufson, J.R.Sladek, Jr., J.P. Hammang, E.E. Baetge and D.F. Emerich,The aged monkey basal forebrain: rescue and sprouting ofaxotomized basal forebrain neurons after grafts of encapsu-lated cells secreting human nerve growth factor,Proc NatlAcad Sci USA 91 (1994), 10898–10902.

[136] S. Korsching, G. Auburger, R. Heumann, J. Scott and H.Thoenen, Levels of nerve growth factor and its mRNA in thecentral nervous system of the rat correlate with cholinergicinnervation,Embo J 4 (1985), 1389–1393.

[137] L.F. Kromer, Nerve growth factor treatment after brain injuryprevents neuronal death,Science 235 (1987), 214–216.

[138] H. Kunugi, A. Ueki, M. Otsuka, K. Isse, H. Hirasawa, N. Ka-to, T. Nabika, S. Kobayashi and S. Nanko, A novel polymor-phism of the brain-derived neurotrophic factor (BDNF) geneassociated with late-onset Alzheimer’s disease,Mol Psychi-atry 6 (2001), 83–86.

[139] S.P. Lad, K.E. Neet and E.J. Mufson, Nerve growth fac-tor: structure, function and therapeutic implications forAlzheimer’s disease,Curr Drug Targets CNS Neurol Disord2 (2003), 315–334.

[140] S. Lahteinen, A. Pitkanen, E. Koponen, T. Saarelainen and E.Castren, Exacerbated status epilepticus and acute cell loss,but no changes in epileptogenesis, in mice with increasedbrain-derived neurotrophic factor signaling,Neuroscience122 (2003), 1081–1092.

[141] A. Lambiase, P. Rama, S. Bonini, G. Caprioglio and L. Aloe,Topical treatment with nerve growth factor for corneal neu-rotrophic ulcers,N Engl J Med 338 (1998), 1174–1180.

[142] A. Lambiase, L. Pagani, V. Di Fausto, V. Sposato, M.Coassin, S. Bonini and L. Aloe, Nerve growth factor eyedrop administrated on the ocular surface of rodents affectsthe nucleus basalis and septum: biochemical and structuralevidence,Brain Res 1127 (2007), 45–51.

[143] C. Laske, E. Stransky, T. Leyhe, G.W. Eschweiler, A. Wit-torf, E. Richartz, M. Bartels, G. Buchkremer and K. Schott,Stage-dependent BDNF serum concentrations in Alzheimer’sdisease,J Neural Transm 113 (2006), 1217–1224.

[144] F.S. Lee and M.V. Chao, Activation of Trk neurotrophinreceptors in the absence of neurotrophins,Proc Natl AcadSci USA 98 (2001), 3555–3560.

[145] F.S. Lee, R. Rajagopal, A.H. Kim, P.C. Chang and M.V.Chao, Activation of Trk neurotrophin receptor signaling bypituitary adenylate cyclase-activating polypeptides,J BiolChem 277 (2002), 9096–9102.

[146] K.F. Lee, E. Li, L.J. Huber, S.C. Landis, A.H. Sharpe, M.V.Chao and R. Jaenisch, Targeted mutation of the gene encod-ing the low affinity NGF receptor p75 leads to deficits in theperipheral sensory nervous system,Cell 69 (1992), 737–749.

[147] R. Lee, P. Kermani, K.K. Teng and B.L. Hempstead, Regu-lation of cell survival by secreted proneurotrophins,Science294 (2001), 1945–1948.

[148] R. Levi-Montalcini and V. Hamburger, Selective growth stim-ulating effects of mouse sarcoma on the sensory and sympa-thetic nervous system of the chick embryo,J Exp Zool 116(1951), 321–361.

[149] R. Levi-Montalcini, The nerve growth factor 35 years later,Science 237 (1987), 1154–1162.

[150] F.M. Longo and S.M. Massa, Neurotrophin-based strategiesfor neuroprotection,J Alzheimers Dis 6 (2004), S13–17.

[151] H. Lou, S.K. Kim, E. Zaitsev, C.R. Snell, B. Lu and Y.P.Loh, Sorting and activity-dependent secretion of BDNF re-quire interaction of a specific motif with the sorting receptorcarboxypeptidase e,Neuron 45 (2005), 245–255.

[152] B. Lu, P.T. Pang and N.H. Woo, The yin and yang of neu-rotrophin action,Nat Rev Neurosci 6 (2005), 603–614.

[153] H.G. Luesse, T. Roskoden, R. Linke, U. Otten, K. Heeseand H. Schwegler, Modulation of mRNA expression of theneurotrophins of the nerve growth factor family and their re-ceptors in the septum and hippocampus of rats after transientpostnatal thyroxine treatment. I. Expression of nerve growthfactor, brain-derived neurotrophic factor, neurotrophin-3, andneurotrophin 4 mRNA,Exp Brain Res 119 (1998), 1–8.

[154] E. Magnani, J. Fan, L. Gasparini, M. Golding, M. Williams,G. Schiavo, M. Goedert, L.A. Amos and M.G. Spillantini,Interaction of tau protein with the dynactin complex,EmboJ 26 (2007), 4546–4554.

[155] S.M. Massa, Y. Xie and F.M. Longo, Alzheimer’s therapeu-tics: neurotrophin domain small molecule mimetics,J MolNeurosci 20 (2003), 323–326.

[156] S.M. Massa, Y. Xie, T. Yang, A.W. Harrington, M.L. Kim,S.O. Yoon, R. Kraemer, L.A. Moore, B.L. Hempstead andF.M. Longo, Small, nonpeptide p75NTR ligands induce sur-vival signaling and inhibit proNGF-induced death,J Neurosci26 (2006), 5288–5300.

[157] C. Matrone, A. Di Luzio, G. Meli, S. D’Aguanno, C. Sev-erini, M.T. Ciotti, A. Cattaneo and P. Calissano, Activationof the Amyloidogenic Route by NGF Deprivation InducesApoptotic Death in PC12 Cells,J Alzheimers Dis 13 (2008),81–96.

[158] J.F. Maya Vetencourt, A. Sale, A. Viegi, L. Baroncelli, R.De Pasquale, O.F. O’Leary, E. Castren and L. Maffei, Theantidepressant fluoxetine restores plasticity in the adult visualcortex,Science 320 (2008), 385–388.

[159] J. Mazella and J.P. Vincent, Functional roles of the NTS2 andNTS3 receptors,Peptides 27 (2006), 2469–2475.

[160] J.P. McGrath, X. Cao, A. Schutz, P. Lynch, T. Ebendal, M.J.Coloma, S.L. Morrison and S.D. Putney, Bifunctional fu-sion between nerve growth factor and a transferrin receptorantibody,J Neurosci Res 47 (1997), 123–133.

[161] M.M. Mesulam, Neuroplasticity failure in Alzheimer’s dis-ease: bridging the gap between plaques and tangles,Neuron24 (1999), 521–529.

[162] B. Michalski and M. Fahnestock, Pro-brain-derived neu-rotrophic factor is decreased in parietal cortex in Alzheimer’sdisease,Brain Res Mol Brain Res 111 (2003), 148–154.

[163] W.C. Mobley, J.L. Rutkowski, G.I. Tennekoon, J. Gemski,K. Buchanan and M.V. Johnston, Nerve growth factor in-creases choline acetyltransferase activity in developing basalforebrain neurons,Brain Res 387 (1986), 53–62.

[164] G. Morfini, G. Szebenyi, R. Elluru, N. Ratner and S.T. Brady,Glycogen synthase kinase 3 phosphorylates kinesin light


chains and negatively regulates kinesin-based motility,EmboJ 21 (2002), 281–293.

[165] A. Mouri, H. Nomoto and S. Furukawa, Processing of nervegrowth factor: the role of basic amino acid clusters in the pro-region, Biochem Biophys Res Commun 353 (2007), 1056–1062.

[166] S.J. Mowla, S. Pareek, H.F. Farhadi, K. Petrecca, J.P. Fawcett,N.G. Seidah, S.J. Morris, W.S. Sossin and R.A. Murphy,Differential sorting of nerve growth factor and brain-derivedneurotrophic factor in hippocampal neurons,J Neurosci 19(1999), 2069–2080.

[167] E.J. Mufson, J.M. Conner and J.H. Kordower, Nerve growthfactor in Alzheimer’s disease: defective retrograde transportto nucleus basalis,Neuroreport 6 (1995), 1063–1066.

[168] E.J. Mufson, J.M. Li, T. Sobreviela and J.H. Kordower, De-creased trkA gene expression within basal forebrain neuronsin Alzheimer’s disease,Neuroreport 8 (1996), 25–29.

[169] E.J. Mufson, N. Lavine, S. Jaffar, J.H. Kordower, R. Quirionand H.U. Saragovi, Reduction in p140-TrkA receptor proteinwithin the nucleus basalis and cortex in Alzheimer’s disease,Exp Neurol 146 (1997), 91–103.

[170] E.J. Mufson, S.Y. Ma, E.J. Cochran, D.A. Bennett, L.A.Beckett, S. Jaffar, H.U. Saragovi and J.H. Kordower, Lossof nucleus basalis neurons containing trkA immunoreactiv-ity in individuals with mild cognitive impairment and earlyAlzheimer’s disease,J Comp Neurol 427 (2000), 19–30.

[171] M.G. Murer, F. Boissiere, Q. Yan, S. Hunot, J. Villares,B. Faucheux, Y. Agid, E. Hirsch and R. Raisman-Vozari,An immunohistochemical study of the distribution of brain-derived neurotrophic factor in the adult human brain, withparticular reference to Alzheimer’s disease,Neuroscience 88(1999), 1015–1032.

[172] K. Nakamura, K. Namekata, C. Harada and T. Harada,Intracellular sortilin expression pattern regulates proNGF-induced naturally occurring cell death during development,Cell Death Differ 14 (2007), 1552–1554.

[173] H. Nomoto, M. Takaiwa, A. Mouri and S. Furukawa, Pro-region of neurotrophins determines the processing efficiency,Biochem Biophys Res Commun 356 (2007), 919–924.

[174] NsGene, http://www.nsgene.com/, Last update December 12,2007, Accessed on May 29, 2008.

[175] R. Nuydens, G. Dispersyn, M. de Jong, G. van den Kieboom,M. Borgers and H. Geerts, Aberrant tau phosphorylationand neurite retraction during NGF deprivation in PC12 cells,Biochem Biophys Res Commun 240 (1997), 687–691.

[176] A. Nykjaer, R. Lee, K.K. Teng, P. Jansen, P. Madsen, M.S.Nielsen, C. Jacobsen, M. Kliemannel, E. Schwarz, T.E. Will-now, B.L. Hempstead and C.M. Petersen, Sortilin is essentialfor proNGF-induced neuronal cell death,Nature 427 (2004),843–848.

[177] A. Nykjaer, T.E. Willnow and C.M. Petersen, p75NTR–liveor let die,Curr Opin Neurobiol 15 (2005), 49–57.

[178] G. Olivieri, U. Otten, F. Meier, G. Baysang, B. Dimitriades-Schmutz, F. Muller-Spahn and E. Savaskan, Beta-amyloidmodulates tyrosine kinase B receptor expression in SHSY5Yneuroblastoma cells: influence of the antioxidant melatonin,Neuroscience 120 (2003), 659–665.

[179] L. Olson, A. Nordberg, H. von Holst, L. Backman, T. Eben-dal, I. Alafuzoff, K. Amberla, P. Hartvig, A. Herlitz, A. Liljaand et al., Nerve growth factor affects 11C-nicotine bind-ing, blood flow, EEG, and verbal episodic memory in anAlzheimer patient (case report),J Neural Transm Park DisDement Sect 4 (1992), 79–95.

[180] N. Origlia, S. Capsoni, L. Domenici and A. Cattaneo, Timewindow in cholinomimetic ability to rescue long-term poten-tiation in neurodegenerating anti-nerve growth factor mice,J Alzheimers Dis 9 (2006), 59–68.

[181] P.C. Pagadala, L.A. Dvorak and K.E. Neet, Construction of amutated pro-nerve growth factor resistant to degradation andsuitable for biophysical and cellular utilization,Proc NatlAcad Sci USA 103 (2006), 17939–17943.

[182] F. Paoletti, P.V. Konarev, S. Covaceuszach, E. Schwarz, A.Cattaneo, D. Lamba and D.I. Svergun, Structural and func-tional properties of mouse proNGF,Biochem Soc Trans 34(2006), 605–606.

[183] W.M. Pardridge, J.L. Buciak and P.M. Friden, Selectivetransport of an anti-transferrin receptor antibody throughthe blood-brain barrierin vivo, J Pharmacol Exp Ther 259(1991), 66–70.

[184] C.E. Pedraza, P. Podlesniy, N. Vidal, J.C. Arevalo, R. Lee,B. Hempstead, I. Ferrer, M. Iglesias and C. Espinet, Pro-NGF isolated from the human brain affected by Alzheimer’sdisease induces neuronal apoptosis mediated by p75NTR,Am J Pathol 166 (2005), 533–543.

[185] M. Pehar, P. Cassina, M.R. Vargas, Y. Xie, J.S. Beckman,S.M. Massa, F.M. Longo and L. Barbeito, Modulation ofp75-dependent motor neuron death by a small non-peptidylmimetic of the neurotrophin loop 1 domain,Eur J Neurosci24 (2006), 1575–1580.

[186] J. Peleshok and H.U. Saragovi, Functional mimetics of neu-rotrophins and their receptors,Biochem Soc Trans 34 (2006),612–617.

[187] E. Pesavento, S. Capsoni, L. Domenici and A. Cattaneo,Acute cholinergic rescue of synaptic plasticity in the neu-rodegenerating cortex of anti-nerve-growth-factor mice,EurJ Neurosci 15 (2002), 1030–1036.

[188] B.G. Petty, D.R. Cornblath, B.T. Adornato, V. Chaudhry, C.Flexner, M. Wachsman, D. Sinicropi, L.E. Burton and S.J.Peroutka, The effect of systemically administered recombi-nant human nerve growth factor in healthy human subjects,Ann Neurol 36 (1994), 244–246.

[189] S. Pezet and S.B. McMahon, Neurotrophins: mediators andmodulators of pain,Annu Rev Neurosci 29 (2006), 507–538.

[190] C.H. Phelps, F.H. Gage, J.H. Growdon, F. Hefti, R. Har-baugh, M.V. Johnston, Z.S. Khachaturian, W.C. Mobley,D.L. Price, M. Raskind and et al., Potential use of nervegrowth factor to treat Alzheimer’s disease,Neurobiol Aging10 (1989), 205–207.

[191] H.S. Phillips, J.M. Hains, M. Armanini, G.R. Laramee, S.A.Johnson and J.W. Winslow, BDNF mRNA is decreased inthe hippocampus of individuals with Alzheimer’s disease,Neuron 7 (1991), 695–702.

[192] P. Podlesniy, A. Kichev, C. Pedraza, J. Saurat, M. Encinas, B.Perez, I. Ferrer and C. Espinet, Pro-NGF from Alzheimer’sdisease and normal human brain displays distinctive abilitiesto induce processing and nuclear translocation of intracellulardomain of p75NTR and apoptosis,Am J Pathol 169 (2006),119–131.

[193] J.F. Poduslo, G.L. Curran and C.T. Berg, Macromolecularpermeability across the blood-nerve and blood-brain barriers,Proc Natl Acad Sci USA 91 (1994), 5705–5709.

[194] J.F. Poduslo and G.L. Curran, Permeability at the blood-brainand blood-nerve barriers of the neurotrophic factors: NGF,CNTF, NT-3, BDNF,Brain Res Mol Brain Res 36 (1996),280–286.


[195] J.F. Poduslo and G.L. Curran, Polyamine modification in-creases the permeability of proteins at the blood-nerve andblood-brain barriers,J Neurochem 66 (1996), 1599–1609.

[196] J.F. Poduslo, G.L. Curran and J.S. Gill, Putrescine-modifiednerve growth factor: bioactivity, plasma pharmacokinetics,blood-brain/nerve barrier permeability, and nervous systembiodistribution,J Neurochem 71 (1998), 1651–1660.

[197] S. Rabizadeh and D.E. Bredesen, Ten years on: mediation ofcell death by the common neurotrophin receptor p75(NTR),Cytokine Growth Factor Rev 14 (2003), 225–239.

[198] M.A. Reger, G.S. Watson, W.H. Frey, 2nd, L.D. Baker, B.Cholerton, M.L. Keeling, D.A. Belongia, M.A. Fishel, S.R.Plymate, G.D. Schellenberg, M.M. Cherrier and S. Craft, Ef-fects of intranasal insulin on cognition in memory-impairedolder adults: modulation by APOE genotype,Neurobiol Ag-ing 27 (2006), 451–458.

[199] M.A. Reger, G.S. Watson, P.S. Green, L.D. Baker, B. Choler-ton, M.A. Fishel, S.R. Plymate, M.M. Cherrier, G.D. Schel-lenberg, W.H. Frey Ii and S. Craft, Intranasal Insulin Admin-istration Dose-Dependently Modulates Verbal Memory andPlasma Amyloid-beta in Memory-Impaired Older Adults,JAlzheimers Dis 13 (2008), 323–331.

[200] M.A. Reger, G.S. Watson, P.S. Green, C.W. Wilkinson, L.D.Baker, B. Cholerton, M.A. Fishel, S.R. Plymate, J.C. Breit-ner, W. DeGroodt, P. Mehta and S. Craft, Intranasal insulinimproves cognition and modulates beta-amyloid in early AD,Neurology 70 (2008), 440–448.

[201] M. Reinshagen, I. Geerling, V.E. Eysselein, G. Adler, K.R.Huff, G.P. Moore and J. Lakshmanan, Commercial recom-binant human beta-nerve growth factor and adult rat dorsalroot ganglia contain an identical molecular species of nervegrowth factor prohormone,J Neurochem 74 (2000), 2127–2133.

[202] M. Rosato-Siri, A. Cattaneo and E. Cherubini, Nicotine-induced enhancement of synaptic plasticity at CA3-CA1synapses requires GABAergic interneurons in adult anti-NGF mice,J Physiol 576 (2006), 361–377.

[203] H. Rosch, R. Schweigreiter, T. Bonhoeffer, Y.A. Barde andM. Korte, The neurotrophin receptor p75NTR modulateslong-term depression and regulates the expression of AMPAreceptor subunits in the hippocampus,Proc Natl Acad SciUSA 102 (2005), 7362–7367.

[204] S. Rossner, U. Ueberham, R. Schliebs, J.R. Perez-Polo and V.Bigl, The regulation of amyloid precursor protein metabolismby cholinergic mechanisms and neurotrophin receptor sig-naling,Prog Neurobiol 56 (1998), 541–569.

[205] W.B. Rowe, S. Kar, M.J. Meaney and R. Quirion, Neu-rotensin receptor levels as a function of brain aging and cog-nitive performance in the Morris water maze task in the rat,Peptides 27 (2006), 2415–2423.

[206] F. Ruberti, S. Capsoni, A. Comparini, E. Di Daniel, J. Fran-zot, S. Gonfloni, G. Rossi, N. Berardi and A. Cattaneo, Phe-notypic knockout of nerve growth factor in adult transgenicmice reveals severe deficits in basal forebrain cholinergicneurons, cell death in the spleen, and skeletal muscle dystro-phy,J Neurosci 20 (2000), 2589–2601.

[207] G. Ruiz, D. Ceballos and J.E. Banos, Behavioral and histo-logical effects of endoneurial administration of nerve growthfactor: possible implications in neuropathic pain,Brain Res1011 (2004), 1–6.

[208] C. Russo, V. Dolcini, S. Salis, V. Venezia, N. Zambrano, T.Russo and G. Schettini, Signal transduction through tyrosine-phosphorylated C-terminal fragments of amyloid precursorprotein via an enhanced interaction with Shc/Grb2 adaptor

proteins in reactive astrocytes of Alzheimer’s disease brain,J Biol Chem 277 (2002), 35282–35288.

[209] M.A. Russo, T. Odorisio, A. Fradeani, L. Rienzi, M. De Fe-lici, A. Cattaneo and G. Siracusa, Low-affinity nerve growthfactor receptor is expressed during testicular morphogenesisand in germ cells at specific stages of spermatogenesis,MolReprod Dev 37 (1994), 157–166.

[210] A. Salehi, J. Verhaagen, P.A. Dijkhuizen and D.F. Swaab, Co-localization of high-affinity neurotrophin receptors in nucleusbasalis of Meynert neurons and their differential reduction inAlzheimer’s disease,Neuroscience 75 (1996), 373–387.

[211] A. Salehi, J.D. Delcroix and W.C. Mobley, Traffic at the in-tersection of neurotrophic factor signaling and neurodegen-eration,Trends Neurosci 26 (2003), 73–80.

[212] K. Schindowski, K. Belarbi and L. Buee, Neurotrophic fac-tors in Alzheimer’s disease: role of axonal transport,GenesBrain Behav 7(Suppl 1) (2008), 43–56.

[213] R. Schliebs and T. Arendt, The significance of the cholinergicsystem in the brain during aging and in Alzheimer’s disease,J Neural Transm 113 (2006), 1625–1644.

[214] R. Schweigreiter, The dual nature of neurotrophins,Bioes-says 28 (2006), 583–594.

[215] S.A. Scott, E.J. Mufson, J.A. Weingartner, K.A. Skau andK.A. Crutcher, Nerve growth factor in Alzheimer’s disease:increased levels throughout the brain coupled with declinesin nucleus basalis,J Neurosci 15 (1995), 6213–6221.

[216] N.G. Seidah, S. Benjannet, S. Pareek, D. Savaria, J. Hamelin,B. Goulet, J. Laliberte, C. Lazure, M. Chretien and R.A.Murphy, Cellular processing of the nerve growth factor pre-cursor by the mammalian pro-protein convertases,BiochemJ 314(Pt 3) (1996), 951–960.

[217] D.J. Selkoe, The genetics and molecular pathology ofAlzheimer’s disease: roles of amyloid and the presenilins,Neurol Clin 18 (2000), 903–922.

[218] D.J. Selkoe, Alzheimer’s disease is a synaptic failure,Science298 (2002), 789–791.

[219] D.J. Selkoe, Amyloid beta-peptide is produced by culturedcells during normal metabolism: a reprise,J Alzheimers Dis9 (2006), 163–168.

[220] I. Semkova, P. Wolz, M. Schilling and J. Krieglstein, Selegi-line enhances NGF synthesis and protects central nervoussystem neurons from excitotoxic and ischemic damage,EurJ Pharmacol 315 (1996), 19–30.

[221] I. Shinoda, Y. Furukawa and S. Furukawa, Stimulation ofnerve growth factor synthesis/secretion by propentofyllinein cultured mouse astroglial cells,Biochem Pharmacol 39(1990), 1813–1816.

[222] E.M. Shooter, Early days of the nerve growth factor proteins,Annu Rev Neurosci 24 (2001), 601–629.

[223] D.E. Smith, J. Roberts, F.H. Gage and M.H. Tuszynski, Age-associated neuronal atrophy occurs in the primate brain andis reversible by growth factor gene therapy,Proc Natl AcadSci USA 96 (1999), 10893–10898.

[224] F. Sohrabji, R.C. Miranda and C.D. Toran-Allerand, Estro-gen differentially regulates estrogen and nerve growth fac-tor receptor mRNAs in adult sensory neurons,J Neurosci 14(1994), 459–471.

[225] E. Sola, S. Capsoni, M. Rosato-Siri, A. Cattaneo and E.Cherubini, Failure of nicotine-dependent enhancement ofsynaptic efficacy at Schaffer-collateral CA1 synapses ofAD11 anti-nerve growth factor transgenic mice,Eur J Neu-rosci 24 (2006), 1252–1264.

[226] B. Srinivasan, C.H. Roque, B.L. Hempstead, M.R. Al-Ubaidiand R.S. Roque, Microglia-derived pronerve growth factor


promotes photoreceptor cell death via p75 neurotrophin re-ceptor,J Biol Chem 279 (2004), 41839–41845.

[227] K. Stamer, R. Vogel, E. Thies, E. Mandelkow and E.M.Mandelkow, Tau blocks traffic of organelles, neurofilaments,and APP vesicles in neurons and enhances oxidative stress,J Cell Biol 156 (2002), 1051–1063.

[228] U. Suter, J.V. Heymach, Jr. and E.M. Shooter, Two conserveddomains in the NGF propeptide are necessary and sufficientfor the biosynthesis of correctly processed and biologicallyactive NGF,Embo J 10 (1991), 2395–2400.

[229] P.E. Tarr, C. Contursi, R. Roncarati, C. Noviello, E. Ghersi,M.H. Scheinfeld, N. Zambrano, T. Russo and L. D’Adamio,Evidence for a role of the nerve growth factor receptor Tr-kA in tyrosine phosphorylation and processing of beta-APP,Biochem Biophys Res Commun 295 (2002), 324–329.

[230] P.E. Tarr, R. Roncarati, G. Pelicci, P.G. Pelicci and L.D’Adamio, Tyrosine phosphorylation of the beta-amyloidprecursor protein cytoplasmic tail promotes interaction withShc,J Biol Chem 277 (2002), 16798–16804.

[231] R.D. Terry, E. Masliah, D.P. Salmon, N. Butters, R. DeTeresa,R. Hill, L.A. Hansen and R. Katzman, Physical basis ofcognitive alterations in Alzheimer’s disease: synapse loss isthe major correlate of cognitive impairment,Ann Neurol 30(1991), 572–580.

[232] M. Theodosiou, R.A. Rush, X.F. Zhou, D. Hu, J.S. Walkerand D.J. Tracey, Hyperalgesia due to nerve damage: role ofnerve growth factor,Pain 81 (1999), 245–255.

[233] P. Tirassa, L. Aloe, C. Stenfors, P. Turrini and T. Lunde-berg, Cholecystokinin-8 protects central cholinergic neuronsagainst fimbria-fornix lesion through the up-regulation ofnerve growth factor synthesis,Proc Natl Acad Sci USA 96(1999), 6473–6477.

[234] L. Tong, R. Balazs, P.L. Thornton and C.W. Cotman, Beta-amyloid peptide at sublethal concentrations downregulatesbrain-derived neurotrophic factor functions in cultured corti-cal neurons,J Neurosci 24 (2004), 6799–6809.

[235] S.J. Tsai, C.J. Hong, H.C. Liu, T.Y. Liu, L.E. Hsu and C.H.Lin, Association analysis of brain-derived neurotrophic fac-tor Val66Met polymorphisms with Alzheimer’s disease andage of onset,Neuropsychobiology 49 (2004), 10–12.

[236] M.H. Tuszynski, H.S. U, D.G. Amaral and F.H. Gage, Nervegrowth factor infusion in the primate brain reduces lesion-induced cholinergic neuronal degeneration,J Neurosci 10(1990), 3604–3614.

[237] M.H. Tuszynski, H. Sang, K. Yoshida and F.H. Gage, Recom-binant human nerve growth factor infusions prevent cholin-ergic neuronal degeneration in the adult primate brain,AnnNeurol 30 (1991), 625–636.

[238] M.H. Tuszynski, J. Roberts, M.C. Senut, H.S. U and F.H.Gage, Gene therapy in the adult primate brain: intra-parenchymal grafts of cells genetically modified to producenerve growth factor prevent cholinergic neuronal degenera-tion, Gene Ther 3 (1996), 305–314.

[239] M.H. Tuszynski, L. Thal, M. Pay, D.P. Salmon, H.S. U, R.Bakay, P. Patel, A. Blesch, H.L. Vahlsing, G. Ho, G. Tong,S.G. Potkin, J. Fallon, L. Hansen, E.J. Mufson, J.H. Kor-dower, C. Gall and J. Conner, A phase 1 clinical trial ofnerve growth factor gene therapy for Alzheimer disease,Nat

Med 11 (2005), 551–555.[240] M.H. Tuszynski, Nerve growth factor gene therapy in

Alzheimer disease,Alzheimer Dis Assoc Disord 21 (2007),179–189.

[241] J.L. Twiss, J.H. Chang and N.C. Schanen, Pathophysiologicalmechanisms for actions of the neurotrophins,Brain Pathol16 (2006), 320–332.

[242] M. Ventriglia, L. Bocchio Chiavetto, L. Benussi, G. Binet-ti, O. Zanetti, M.A. Riva and M. Gennarelli, Associationbetween the BDNF 196 A/G polymorphism and sporadicAlzheimer’s disease,Mol Psychiatry 7 (2002), 136–137.

[243] S. Vepsalainen, E. Castren, S. Helisalmi, S. Iivonen, A. Man-nermaa, M. Lehtovirta, T. Hanninen, H. Soininen and M.Hiltunen, Genetic analysis of BDNF and TrkB gene polymor-phisms in Alzheimer’s disease,J Neurol 252 (2005), 423–428.

[244] J.P. Vincent, J. Mazella and P. Kitabgi, Neurotensin and neu-rotensin receptors,Trends Pharmacol Sci 20 (1999), 302–309.

[245] M. Volosin, W. Song, R.D. Almeida, D.R. Kaplan, B.L.Hempstead and W.J. Friedman, Interaction of survival anddeath signaling in basal forebrain neurons: roles of neu-rotrophins and proneurotrophins,J Neurosci 26 (2006),7756–7766.

[246] Z.F. Wang, L.L. Tang, H. Yan, Y.J. Wang and X.C. Tang,Effects of huperzine A on memory deficits and neurotroph-ic factors production after transient cerebral ischemia andreperfusion in mice,Pharmacol Biochem Behav 83 (2006),603–611.

[247] P.J. Whitehouse, D.L. Price, R.G. Struble, A.W. Clark, J.T.Coyle and M.R. Delon, Alzheimer’s disease and senile de-mentia: loss of neurons in the basal forebrain,Science 215(1982), 1237–1239.

[248] J. Winkler, G.A. Ramirez, H.G. Kuhn, D.A. Peterson, P.A.Day-Lollini, G.R. Stewart, M.H. Tuszynski, F.H. Gage andL.J. Thal, Reversible Schwann cell hyperplasia and sprout-ing of sensory and sympathetic neurites after intraventricularadministration of nerve growth factor,Ann Neurol 41 (1997),82–93.

[249] S.A. Wolf, G. Kronenberg, K. Lehmann, A. Blankenship,R. Overall, M. Staufenbiel and G. Kempermann, Cognitiveand physical activity differently modulate disease progres-sion in the amyloid precursor protein (APP)-23 model ofAlzheimer’s disease,Biol Psychiatry 60 (2006), 1314–1323.

[250] Y. Xie, M.A. Tisi, T.T. Yeo and F.M. Longo, Nerve growthfactor (NGF) loop 4 dimeric mimetics activate ERK and AKTand promote NGF-like neurotrophic effects,J Biol Chem 275(2000), 29868–29874.

[251] Y. Xie, L. Ye, X. Zhang, W. Cui, J. Lou, T. Nagai and X. Hou,Transport of nerve growth factor encapsulated into liposomesacross the blood-brain barrier: in vitro andin vivo studies,JControl Release 105 (2005), 106–119.

[252] M. Yaar, S. Zhai, R.E. Fine, P.B. Eisenhauer, B.L. Arble,K.B. Stewart and B.A. Gilchrest, Amyloid beta binds trimersas well as monomers of the 75-kDa neurotrophin receptorand activates receptor signaling,J Biol Chem 277 (2002),7720–7725.

Towards Non Invasive Nerve Growth Factor Therapies for

Documents

Transcript of Towards Non Invasive Nerve Growth Factor Therapies for