p75 and TrkA receptors are both required for uptake of NGF in adult sympathetic neurons: use of a novel fluorescent NGF conjugate

Brain Research 920 (2001) 226–238www.elsevier.com/ locate /bres

Research report

p75 and TrkA receptors are both required for uptake of NGF in adultsympathetic neurons: use of a novel fluorescent NGF conjugate

a , b c c*Kliment P. Gatzinsky , Rosaria P. Haugland , Christopher Thrasivoulou , Nina Orike ,c ,1 cAgus W. Budi-Santoso , Timothy Cowen

a ¨Departments of Anatomy and Cell Biology and Clinical Neurosciences, Goteborg University, Box 420, S-405 30 Gothenburg, SwedenbMolecular Probes, Eugene, OR 97402-9165, USA

cDepartment of Anatomy and Developmental Biology, Royal Free and University College Medical School, London NW3 2PF, UK

Accepted 17 August 2001

Abstract

We have developed and tested the biological activity and specificity of a novel fluorescent dextran-Texas Red–nerve growth factor(DTR–NGF) conjugate. DTR–NGF was found to promote survival and neurite outgrowth in cultured dissociated sympathetic neuronssimilarly to native NGF. The conjugate was taken up and transported retrogradely by terminal sympathetic nerves innervating the iris toneurons in the ipsilateral superior cervical ganglion (SCG) of young adult rats. Uptake and transport was assessed by counting numbers oflabelled neurons and by measuring intensity of neuronal labelling using confocal microscopy and image analysis. DTR–NGF labelling inSCG neurons was shown to be dose-dependent with an EC of 75 ng. Similar concentrations of unconjugated DTR resulted in no50

neuronal labelling. DTR–NGF uptake was competed off using a 50-fold excess of native NGF, resulting in a 73% reduction in numbers oflabelled neurons. Pretreatment of nerve terminals with function-blocking antibodies against the low (p75) and high (TrkA) affinity NGFreceptors resulted in a large (85–93%) reduction in numbers of DTR–NGF labelled neurons. Anti-p75 and anti-TrkA antibodies hadcomparable effects which were concentration-dependent. These findings indicate that both receptors are required for uptake of NGF inadult rat sympathetic neurons. In particular, the results provide strong evidence that the p75 receptor plays a more active role intransducing the NGF signal than has been proposed. 2001 Elsevier Science B.V. All rights reserved.

Theme: Development and regeneration

Topic: Neurotrophic factors: receptors and cellular mechanisms

Keywords: Rat; NGF transport; Superior cervical ganglion; Antibody receptor blockade; Confocal microscopy

1. Introduction less well understood. Since direct injections of NGF in theneuronal cytoplasm fail to elicit responses [41,61], it seems

The neurotrophin nerve growth factor (NGF) is secreted unlikely that NGF alone is competent to signal its survival-by the targets of sympathetic and some sensory neurons and growth-promoting effects. The retrograde effects ofand supports the survival and growth of dependent neurons NGF are likely therefore to be mediated by signallingduring development [18,20,27]. However, the mechanism pathways involving its receptors (for reviews, see Refs.by which the NGF signal is propagated from the axon [6,22,29]).terminal to the cell body, as well as the extent to which Two distinct cell surface receptors that bind NGF haveadult neurons remain dependent on target-derived NGF, is been identified: the 75 kDa low-affinity neurotrophin

receptor (p75NTR) which displays rapid association anddissociation with most members of the neurotrophin family

*Corresponding author. Tel.: 146-31-3422744; fax: 146-31-416719.[44,58] and the 140 kDa high-affinity tyrosine kinaseE-mail address: [email protected] (K.P. Gatzinsky).

1 (TrkA) receptor which binds NGF more specifically butPresent address: Department of Physiology, Faculty of Medicine,Indonesian Catholic University Atma Jaya, Jakarta 14440, Indonesia. with slower kinetics [37,53]. NGF elicits most of its effects

0006-8993/01/$ – see front matter 2001 Elsevier Science B.V. All rights reserved.PI I : S0006-8993( 01 )03099-2

K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238 227

on neurons through activation of TrkA with subsequent conjugate was estimated by determining the molar extinc-initiation of intracellular second messenger cascades which tion coefficient of native NGF at 280 nm and measuringpropagate the signal to the nucleus [65]. Several studies the absorbances of the DTR–NGF at 595 and 280 nm. Thehave now suggested that activated TrkA receptors, either optical density of the conjugate at 280 nm was correctedalone or in a ligand–receptor complex, can serve as for the absorbance contribution by the dye at this wave-retrograde NGF signal carriers after endocytosis at the length and the concentration of NGF estimated from theaxon terminals [5,26,60]. residual 280 nm absorption [38]. The conjugate, 0.1 mg of

While the central role played by TrkA in NGF signalling DTR–conjugated NGF per ml PBS with the addition ofhas been established, the contribution of p75NTR is not 0.05% bovine serum albumin (BSA), was freeze-dried infully understood. Evidence that p75NTR plays a role in the aliquots and stored desiccated at 2208C in glass vials forNGF-mediated activation of TrkA indicates a synergy up to 12 months. Samples of DTR–NGF were prepared onbetween TrkA and p75NTR when co-expressed in NGF- three different occasions and tested to confirm specificityresponsive neurons [2,13,30,35,39,42,46,49,50,52,63]. In and reproducibility for retrograde tracing, neuronal surviv-the present study we introduce a novel, highly fluorescent al and neurite outgrowth. For each experiment, smalland biologically active NGF conjugate for analysis of aliquots of lyophilized DTR–NGF were reconstituted withneuronal uptake of NGF in vivo. We use the term ‘uptake’ distilled water and used within 1 week.throughout, whilst understanding that the processes underinvestigation here may include binding, internalization and

2.2. In vitro bioassay of DTR–NGFretrograde axonal transport. Using antibody receptor bloc-kade we have investigated the effect of inhibiting NGF

Biological activity of DTR–NGF was tested by inves-binding to p75NTR and TrkA in sympathetic neurons

tigating its ability to promote the survival and neuriteinnervating the iris of adult rats. The results reveal a more

outgrowth of dissociated rat sympathetic neurons in cul-significant role than has been proposed for p75NTR in

ture. Sympathetic neurons were dissociated from superioruptake and retrograde axonal transport of NGF in adult

cervical ganglia (SCG) removed from 1-day-old malesympathetic neurons. The data also imply an interaction

Sprague–Dawley rats using established assays for survivalbetween p75NTR and TrkA in regulating NGF uptake.

and neurite outgrowth [56,57]. Neurons were plated at lowdensity on poly-L-lysine and laminin-coated coverslips inserum-free media containing sterile BSA, supplemented

2. Materials and methodswith glutamine and glucose. DTR–NGF (0.01–30 ng/ml;the amount reflecting the NGF mass) or native NGF were

2.1. Preparation of dextran-Texas Red–NGF (DTR–added to the medium. After 4 and 24 h of incubation at

NGF) conjugate378C, survival was assayed by counting numbers of phase-bright neurons over marked areas of the coverslip. The

Standard dextran 50 000 (MW 48 600) from FLUKA,proportion of cells extending neurites over one cell diam-

Switzerland was derivatized with 10 mol of amines pereter at 24 h was also counted. These experiments were

mol of dextran and subsequently labelled with Texas Redrepeated three times.

sulfonyl chloride (Molecular Probes, Eugene, OR) toobtain 4 mol of dye per mol of dextran. Murine 2.5S NGF(0.5 mg, Grade II, Boehringer Mannheim, Germany) was 2.3. Neuronal uptake and retrograde transport of DTR–treated with dimethylmaleic anhydride to reversibly block NGFthe amines [24]. Some of the carboxyl groups of the NGFwere activated by a proprietary method (Molecular Probes, Six-week-old Sprague–Dawley male rats reared in aEugene, OR) to react with the remaining amines of the colony maintained at the Royal Free Hospital School ofTexas Red-amino dextran. The stoichiometry was designed Medicine were used in all experiments. The animals wereto yield conjugates in the ratio of one molecule of DTR to anaesthetized with a mixture of halothane and oxygen.one of NGF. The DTR–NGF conjugate was purified by Using a Hamilton syringe, 5 ml of a solution containingconcentrating the reaction mixture on an Ultrafree-4 10–300 ng of DTR-labelled NGF diluted in sterile PBScentrifugal filter unit (Millipore, Bedford, MA) with a were injected through the sclera into the aqueous humour50 000 molecular weight cut-off to separate the conjugate of the anterior eye chamber. This approach allows thefrom unreacted NGF and dextran. The concentrated sample uptake of DTR–NGF by irideal neurons without damagingwas applied to a Sephadex G-150 size exclusion column the iris or cornea. Care was taken to minimise leakage(0.9360 cm) and the conjugate eluted with PBS, pH 7.2. from the eye. After survival times of 17–18 h, the animalsThe amines of the NGF were subsequently deprotected by were killed by an overdose of pentobarbitone sodium anddialysis against 30 mM glycine, pH 3.0, for 24 h and the the ipsilateral and contralateral SCGs were dissected out,pH raised by further dialysis against PBS, pH 7.2. Storage de-sheathed and immersion-fixed in 4% paraformaldehydeand dialyses were performed at 48C. The yield of the for 1.5 h at room temperature. The ganglia were then

228 K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238

rinsed three times in PBS and mounted whole in antifade radiolabelled ligand was used within 3 weeks of iodina-mountant (Citifluor, UK). tion. Animals were injected as before with NGF doses

ranging from 20 to 300 ng. After similar survival times,2.4. Imaging and image analysis radioactivity in the ipsilateral SCGs was measured using

an MRC gamma counter.Confocal scanning laser microscopy was performed on a (ii) Labelling of SCG neurons by DTR–NGF was

Bio-Rad MRC 600 system fitted with a krypton–argon compared with that of unconjugated DTR to confirm thatlaser and a Texas Red filter set using a BH2 Olympus the neuronal labelling pattern observed was specific formicroscope and 203 Plan-Apo oil objective. The micro- DTR–NGF and not due to DTR. Using spectrophotometry,scope output was standardized using a uranyl glass stan- identical molar concentrations of DTR and DTR–NGFdard in order to achieve consistent and comparable ob- were prepared by matching their absorption at 595 nm andservations and measurements. Confocal aperture, gain and 5 ml of the DTR solution were injected into the anteriorblack level were kept constant and the same neutral density eye chamber. Labelling in the ipsilateral SCG was assessedfilter was employed in all experiments. Neuronal uptake as before.and transport of DTR–NGF was assessed using two (iii) The capacity of an excess of native NGF toapproaches: compete out DTR–NGF uptake was investigated by co-

(i) Densitometric assessments of mean fluorescence injecting a 503 excess of native NGF (3 ml) with DTR–intensity or grey value (GV) were made on DTR–NGF- NGF. As an additional control, some animals were co-containing neurons. The first three fields containing la- injected with DTR and an excess of native NGF atbelled neurons which were observed during the systematic concentrations equivalent to those used in the DTR–NGFsearch of each SCG starting at the rostral pole were competition experiments. Neuronal labelling was assessedselected for GV measurements. Single optical sections of as described.each field were imaged using 32 zoom at comparable (iv) Fluorescence immunohistochemistry for NGF wasdepths within the ganglion to avoid variable signal attenua- performed on 15-mm cryosections of DTR–NGF tracedtion and saved to an optical disc. To avoid overexposure SCGs. Briefly, sections were washed in Hepes (10 mM)and photobleaching, illumination times were kept under 30 buffer (Sigma, UK), incubated in 5% goat serum in Hepess for each field. GV measurements were made using buffer for 40 min and then incubated for 16 h at roomestablished techniques for automated quantitation of fluo- temperature in rabbit polyclonal anti-NGF antibody (giftrescence intensity [68] using an Olympus Vanox fluores- from Dr Jim Conner) diluted 1:400 in Hepes buffer10.1%cence microscope, interfaced with a Photonics Science Triton and 1% goat serum. Sections were then washed in(UK) cooled CCD camera to a Kontron KS400 image PBS buffer and incubated in 1:800 goat anti-rabbit Alexaanalysis system. To ensure constancy of illumination 488 antibody (Molecular Probes, Eugene, OR) for 90 min,during successive imaging sessions, all mercury lamp and washed again and mounted in Citifluor antifade mountant.microscope settings were standardised and checked at the Sections were imaged on a Bio-Rad MRC 600 confocalbeginning of each session using an uranyl glass standard. laser scanning microscope under simultaneous dual-excita-At the beginning of the experiment, GV was sampled from tion for FITC (488 nm) for anti-NGF visualisation andthe full range of samples and camera gain was adjusted to Texas Red (568 nm) for DTR–NGF visualisation.ensure that illumination intensity fell within the linearresponse range of the camera and imaging system. 2.6. p75NTR and TrkA antibody inhibition

(ii) Numbers of DTR–NGF-containing neurons in theSCG were counted in composite confocal images of the For p75NTR blockade, the well-characterized rabbitwhole ganglion. All specimens were blind-coded in order polyclonal antibodies, 9651 [42] and anti-Rex [71], whichto avoid examiner bias. To check the reliability of the are directed against different portions of the extracellularquantification, some counts were repeated by independent domain of p75NTR, were used. The concentrations of theexaminers. antibodies were 25 mg/ml for both 9651 and anti-Rex. For

TrkA inhibition, the IgG fraction of a polyclonal antibody2.5. Specificity of neuronal DTR–NGF retrograde (27 mg/ml) against the entire extracellular domain of thelabelling receptor was used [15]. Two microliters of undiluted

antiserum were injected into the anterior eye chamberThe specific neuronal uptake and transport of DTR– immediately followed by 3 ml of PBS containing 75

NGF was validated in several ways: (EC , see below), 100 or 200 ng of DTR–NGF. The50

(i) The dose–response experiments described above volume was kept constant (5 ml) in all groups within an125were repeated using iodinated NGF. [ I]NGF was pre- experiment. In some animals, antibodies were injected 30

pared by lactoperoxidase labelling of 2.5S murine NGF min or 2.5 h before the DTR–NGF in order to investigate(NENE Life Science Products, Belgium). Specific ac- whether blockade would be more effective under thesetivities were in the range of 55–86 cpm/pg NGF and the conditions. Animals injected with 2 ml of either non-

K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238 229

immune rabbit serum or PBS, as substitutes for the 3. Resultsantibodies, served as controls to test antibody specificityand to exclude the possibility that receptor blockade or 3.1. Biological activity of DTR–NGFdysfunction was due to toxic or inflammatory effects of theantibodies. In addition, a group of animals were injected The preparations of DTR–NGF exhibited biologicalwith unconjugated DTR following antibody treatment. The activity similar to that of native, unconjugated NGF (Fig.effects of antibody blockade on NGF uptake and labelling 1). In vitro assay showed that the proportion of survivingof SCG neurons innervating the iris were also tested with SCG neurons was increased by similar amounts when

125[ I]NGF. Radioactivity was assessed as before. doses of 0.01–30 ng of either NGF or DTR–NGF wereAll data were compared using one-way ANOVA, and added to the cultures (Fig. 1a). The proportions of SCG

where significance of P,0.05 was shown, by Tukey’s neurons induced to grow in dissociated cell culture withHSD test. NGF and DTR–NGF were also similar (Fig. 1b). There

Fig. 1. Proportion of dissociated SCG neurons from 1-day rats which survive (a) and are induced to grow neurites (b) after 24 h in the presence of native(WT) or DTR–NGF at concentrations from 0.01 to 30 ng/ml. There was no significant difference between the effects of the two reagents at any of theconcentrations used. Values represent mean6S.E.M. of three separate experiments.

230 K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238

were no significant differences between the responses toeither reagent at equivalent concentrations.

3.2. Retrograde labelling with DTR–NGF of sympatheticneurons innervating the iris

DTR–NGF injected into the anterior eye chamber wasdetected only in neurons of the ipsilateral SCG. Themajority of the labelled neurons appeared in the rostralthird of the SCG (Fig. 2). Numbers of labelled neurons perSCG and GV increased with doses of DTR–NGF to reachsaturation at 150–200 ng (Fig. 3), indicating that uptake ofthe NGF conjugate is receptor-mediated. Higher concen-trations produced no further increase. An EC value50

(half-maximum of the DTR–NGF concentration whichleads to saturation) of approximately 75 ng was estimatedfor both numbers of labelled neurons and GV (Fig. 3).Comparison of the dose–responses of irideal sympathetic

125neurons to DTR–NGF (Fig. 3) and to [ I]NGF (Fig. 4)showed similar dose–responses and comparable EC50

values across a similar range of concentrations, thus furthervalidating the retained bioactivity of DTR–NGF.

When stored lyophilized at –208C, use of DTR–NGFwithin 6 months of preparation gave similar results withregard to numbers of retrogradely labelled SCG neuronsand GV as seen for freshly prepared conjugate. Numbersand GV decreased significantly (by 50–75%) when theconjugate was used 6–12 months after preparation.

The specificity of DTR–NGF neuronal uptake andtransport was confirmed by the absence of labelling in thecontralateral SCGs. In addition, numbers of labelledneurons were reduced by 73% (P,0.001) when a 50-foldexcess of native NGF was injected simultaneously with theDTR–NGF (Fig. 3c). Injection of unconjugated DTR at aconcentration equivalent to 75 ng (EC ) of DTR–NGF50

produced no fluorescent labelling in the SCG. With higherconcentrations of unconjugated DTR, a slow gradualincrease in labelling was seen, indicating that DTR, incontrast to DTR–NGF, is taken up by non-specific, fluidphase-dependent endocytosis at the axon terminals (Fig.3c). The doses of DTR had to be increased over 100-foldto produce uptake and labelling of neurons similar to thatseen in DTR–NGF traced SCGs. Moreover, in contrast tothe competition observed between DTR–NGF and a 50-fold excess of native NGF, the numbers of DTR-labelledneurons increased up to 50-fold when DTR was adminis-tered simultaneously with a 503 excess of native NGF.Besides demonstrating the effects of NGF on non-specificuptake, this observation provides further evidence for thespecificity of uptake of DTR–NGF by NGF-responsive Fig. 2. Low power composite confocal micrographs of DTR–NGF-

labelled neurons in the rostral (a) and caudal (b) regions of the SCG. Onlyneurons.those neurons with labelling clearly distinct from background wereFluorescence immunostaining for NGF showed that thecounted. (c) Higher power confocal image of a single optical slice in the

majority of SCG neurons were positively stained at SCG showing three labelled neurons, similar to those used for measuringvarying intensities for native NGF. However, the intensity intensity of neuronal fluorescence staining (grey value, GV). Scale bar: a,of staining was markedly increased in DTR–NGF-labelled b5100 mm; c550 mm.

K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238 231

Fig. 3. Uptake and retrograde transport of DTR–NGF in SCG neurons innervating the iris. (a) Number of labelled neurons increases from 10 to 200 ng ofDTR–NGF and saturates between 150 and 200 ng, indicating receptor-mediation of uptake and transport. (b) Intensity of fluorescence staining in SCGneurons with similar doses of DTR–NGF. Note a comparable rise in GV over the same dose range as in (a), as well as saturation between 150 and 200 ng.(c) Comparison of numbers of neurons labelled with two preparations of DTR–NGF and unconjugated DTR, and effect of competition with excess nativeNGF. Note the similarity of the dose–response to both DTR–NGF preparations. No significant differences were found between them. Unconjugated DTRshowed very low numbers of labelled neurons, even at doses equivalent to 400 ng of DTR–NGF (3.961.14, data not included). Native NGF wasco-injected with DTR–NGF at a 50-fold higher concentration resulting in a 73% (P,0.001) reduction in numbers of labelled cells in the ipsilateral SCG.Values represent mean6S.E.M. of 4–12 animals.

232 K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238

125Fig. 4. Uptake and retrograde transport of [ I]NGF by SCG neurons innervating the iris. Saturation and EC dose are similar to those observed for the50

uptake of DTR–NGF.

neurons where the staining was colocalized with the DTR– rabbit serum or PBS produced no reduction in numbers orNGF (Fig. 5). GV of DTR–NGF-labelled SCG neurons (Fig. 6). Anti-

p75NTR and anti-TrkA antibodies had no effect on the3.3. p75NTR and TrkA blocking reduces NGF uptake in number of neurons labelled with unconjugated DTR. Theseadult sympathetic neurons control experiments further demonstrated the specificity of

the antibodies for blocking the uptake of NGF in neuronsAntibody blockade of p75NTR and TrkA resulted in a which express p75NTR and TrkA at their axon terminals.

specific reduction of DTR–NGF labelling of SCG neurons(Fig. 6). Blockade of p75NTR with 50 mg of undiluted9651 or anti-Rex antisera reduced the numbers of labelled 4. Discussionneurons by 85 and 93%, respectively (both, P,0.001). Nodifferences were seen between the three DTR–NGF con- In the present study, we introduce a novel, highlycentrations used (75, 100 and 200 ng). Similar reductions fluorescent NGF conjugate which is biologically active,were observed whether the antibodies were injected simul- providing a simple, direct approach for investigatingtaneously with, or 30 min or 2.5 h before the DTR–NGF. neuronal NGF uptake and retrograde axonal transport inIn contrast, intraocular injection of undiluted (54 mg) vivo. Using antibody receptor inhibition we have shownanti-TrkA antiserum reduced the number of labelled SCG that blocking NGF-binding to the p75NTR and TrkAneurons by only 48% (Fig. 6). However, when the receptors at the axon terminals of adult sympatheticconcentration of the anti-TrkA antibody was doubled there neurons projecting to the iris largely decreased uptake andwas an 87% decrease in the number of labelled neurons transport of NGF. The comparable, high and completely(Fig. 6). Similarly, when the anti-p75NTR antibodies were overlapping effect of blockade of p75NTR and TrkAdiluted 1:20 (2.5 mg injected), there was no reduction in suggests that both receptors are required for uptake andthe number of neurons labelled with DTR–NGF, indicat- perhaps also for retrograde axonal transport of NGF. Ining that the inhibitory effect of the function blocking particular, the results provide strong evidence that theantibodies is concentration-dependent (cf. Refs. [19,49]). p75NTR receptor plays a more active role [23] than wasSignificant effects of antibody blockade on NGF uptake previously thought in transducing the NGF signal from the

125and neuronal labelling were observed when [ I]NGF was axon terminals to the cell body of NGF-responsive neu-substituted for DTR–NGF. A decrease in labelling rons.(counts /min per ganglion) of up to 89% followed p75NTR In order to study the neuronal uptake and retrogradeand TrkA antibody blockade (Fig. 7). axonal transport of NGF it is necessary to obtain an easily

Substitution of the antisera with normal, non-immune detectable and biologically active form of the neuro-

K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238 233

Fig. 5. Confocal laser scanning micrograph of 2-mm optical section demonstrating fluorescence of NGF immunostaining (a) and of DTR–NGFretrogradely transported from the iris (b). Note that non-traced neurons, i.e. neurons projecting to other targets, have much reduced NGF staining comparedto the DTR–NGF containing neuron. The fluorescent signal produced by NGF immunostaining in DTR–NGF positive neurons was so intense that gainsettings had to be attenuated to avoid oversaturating the image. This resulted in an apparent signal reduction in non-traced neurons (a). Also note that NGFimmunostaining colocalizes with DTR–NGF. Scale bar525 mm.

Fig. 6. Histogram showing the effect of receptor function-blocking antibodies on neuronal uptake and labelling with 75 ng (EC ) of DTR–NGF. A50

significant reduction in numbers of labelled neurons was seen with the 9651 and anti-Rex antibodies (50 mg) and with the anti-TrkA antibody (54 mg). Theeffectiveness of the latter was enhanced markedly by doubling the concentration. A similar concentration-dependent effect was seen with the 9651 andanti-Rex antibodies (data not shown). Substitution of non-immune rabbit serum for the receptor antisera had no effect on uptake of DTR–NGF. Valuesrepresent mean6S.E.M. of 4–6 animals.

234 K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238

125Fig. 7. Histogram showing the blocking of neuronal uptake of [ I]NGF (100 ng) by receptor function-blocking antibodies. Note the reduction ofapproximately 90% in neuronal uptake of NGF by treatment with both anti-TrkA (108 mg) and anti-p75NTR (anti-Rex; 50 mg) antibodies, a reductionsimilar to that seen for DTR–NGF. Values represent mean6S.E.M. of 4–6 animals.

trophin. Radio-iodinated NGF has been more or less labelled NGF [47,51] have also been tested. The advantageexclusively used for this purpose [40]. The use of radio- of HRP-conjugated NGF is that it can be visualized at thelabelled NGF has, however, certain disadvantages, the ultrastructural level. The signal is, however, again difficult

125main one being that the handling of [ I]NGF requires to quantify. Previous attempts to synthesise fluorescentrigorous safety procedures. At a cellular level, the signal NGF complexes in order to study receptor binding andalso is hard to quantify and resolution is relatively poor. retrograde transport have been hampered by the largeFurthermore, the radiolabelled ligand is unstable and has to decrease in fluorescence yield resulting from binding of thebe used within 3–4 weeks of preparation. Horseradish fluorophors to NGF [47].peroxidase (HRP) coupled to NGF [64] and fluorescently The new NGF complex was prepared using a modi-

K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238 235

fication of previous procedures for conjugation of NGF carriers [5,26,33,34,60]. Despite evidence that p75NTR iswith HRP [64], rhodamine [51] and fluorescein [47], using capable of independent signalling [8–10,25,32], the role ofthe free carboxyl groups of the neurotrophin. Since a large the low-affinity receptor in mediating neuronal responsesmolecule like HRP (40 kDa) can be successfully coupled to NGF is more controversial (for recent reviews, see Refs.to NGF without compromising the biological activity or [11,54]). In certain types of developing neurons, wherethe uptake and retrograde axonal transport of the neuro- TrkA is not expressed, activation of p75NTR by NGF hastrophin [64], we employed 50 kDa dextran to link fluoro- been shown to induce cell death [28]. The most importantphor and NGF. The capacity of dextran to bind several function of p75NTR in TrkA-expressing neurons, however,fluorophor molecules was exploited to obtain a strongly seems to be to enhance NGF binding to TrkA, therebyfluorescent complex containing a ratio of 4:1 Texas Red to increasing NGF-mediated TrkA signalling [21,50]. Conse-dextran molecules. The resulting DTR–NGF conjugate quently, disrupting NGF binding to p75NTR in PC12 cellsexhibited biological activity similar to that of native NGF, results in reduced binding of neurotrophin to TrkA as wellas shown by its ability to promote survival and neurite as a decreased response to NGF at low concentrations [2].outgrowth in dissociated SCG neurons in culture. We have Consistent with this finding, Lachance et al. [49] foundshown that DTR–NGF is taken up by the terminal nerves that blocking the binding of NGF to p75NTR on sympa-of sympathetic neurons and transported to the SCG in a thetic neurons transiently reduces the amount of NGFsaturable, dose-dependent manner similar to that seen with which binds to and activates the TrkA receptor. Moreover,iodinated NGF. The specificity of uptake and transport of p75NTR has been shown to cooperate with TrkA in thethe conjugate was demonstrated by the strong immuno- differentiation of MAH cells, a neuronal precursor cell linefluorescence for NGF in DTR–NGF-positive SCG neu- [70] and in modulating the internalization of NGF inrons, where the immunostaining colocalized with the mutant Sf9 cells [30]. Finally, it has recently been demon-DTR–NGF, as well as by the significant (.70%) reduc- strated that p75NTR expression is required for the de-tion in numbers of labelled neurons when DTR–NGF was tection of NGF in sympathetic axons [16,36] and thatcompeted off its binding sites by an excess of native NGF. sympathetic and sensory innervation of the meningealThis decrease in DTR–NGF labelling agrees well with arteries is severely perturbed in p75NTR-deficient mice

125data on the uptake and transport of [ I]NGF by sympa- [48].thetic and sensory neurons [19]. In the course of carrying A role for p75NTR has also been demonstrated inout controls for the competition experiments, we also mediating uptake of NT-4 and BDNF in peripheral neuronsfound quite unexpectedly a strong potentiation of non- of adult rats [19]. Using a soluble extracellular domainspecific, fluid-phase uptake of dextran by an excess of (sLNR) of p75NTR and the anti-Rex antibody, Curtis et al.native NGF. To our knowledge, this is the first demonstra- [19] showed a dose-dependent inhibitory effect on uptaketion of an effect of this kind, indicating that NGF has a and retrograde axonal transport of iodinated NT-4 andmajor influence on membrane turnover in responsive BDNF to DRG and trigeminal sensory neurons. An sLNR-neurons. induced inhibitory effect, although not as prominent as for

The new conjugate has several advantages compared to the TrkB ligands, was observed also in the uptake of NGF.125radiolabelled NGF. The most obvious is that the numbers In addition, the authors noted a 40% decrease in [ I]NGF

of labelled cells and the intensity of staining in individual retrograde labelling of the SCG using sLNR. A similarcells can be assessed rapidly in whole peripheral ganglia reduction in NGF transport was demonstrated to DRGusing optical sectioning in the confocal microscope fol- neurons of p75NTR-null mice, whereas p75NTR antibodylowed by image analysis. We have previously shown that blockade (anti-Rex) produced no significant decrease insensitive quantitation of neuronal morphology can be made NGF labelling. This prompted us to check carefully thein whole unsectioned SCG using similar confocal tech- more prominent effects of p75NTR antibody inhibition onniques [1]. In addition, the new conjugate should be DTR–NGF retrograde labelling observed in our study. Wesuitable for in vitro studies of cell surface binding, uptake found that substituting non-immune rabbit serum for theand transport of NGF in cells that specifically express receptor antisera had no effect on DTR–NGF neuronalTrkA and/or p75NTR. DTR–NGF is also more stable than labelling, nor did anti-p75NTR antibodies inhibit the

125the I-labelled ligand, retaining biological activity and uptake and transport of unconjugated DTR in SCG neu-specificity of uptake and transport when lyophilized for at rons. In addition, p75NTR antibody blockade significantlyleast 6 months after preparation. reduced uptake and transport to iris-projecting SCG neu-

There is increasing evidence that NGF-induced neuronal rons not only of DTR–NGF but also of iodinated NGF. Weresponses are most likely dependent upon interactions have seen a similar anti-p75NTR antibody blocking effectbetween TrkA and p75NTR [12,17,29,52,55,63,75], mak- (85%) on the uptake of iodinated NGF by SCG neuronsing the ratio of expression of the two receptors biologically supplying another target, the rat middle cerebral arteryimportant [14,73,74]. Of the two NGF-binding receptors, (Gatzinsky, Thrasivoulou and Cowen, unpublished ob-TrkA mediates most of the stereotypical NGF-induced servations).responses of sympathetic neurons. Activated TrkA re- The highest amount of anti-Rex antibody used by Curtisceptors are also able to serve as retrograde NGF signal et al. in a single injection was 10.5 mg, whereas we used

236 K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238

50 mg of the same antibody, a concentration which is available for TrkA binding) in the anterior eye chamberapproximately 53 higher. In agreement with Curtis et al., gave no further increase in neuronal DTR–NGF labelling.we observed no inhibition of NGF neuronal labelling when It is interesting to consider whether receptor expressionthe anti-Rex antibody was diluted 1:20 (2.5 mg per may be affected, in the relatively short (17–18 h) time-injection). A dose-dependent inhibitory effect on NGF course of the present and similar experiments, by alteredbinding to p75NTR has previously also been demonstrated neurotrophin availability, and whether such interactionsfor the 9651 antibody by Lachance et al. [49] in disso- may influence the results observed. Whilst we have evi-ciated cell cultures of sympathetic neurons. In conformity dence that treatment of adult irideal SCG neurons withwith our results, these authors showed that high antibody NGF over 3 weeks leads to enhanced expression ofconcentrations were needed to obtain effective receptor p75NTR [31], we have no comparable data for TrkA or forblockade. As shown in this study, this is also valid for the time-scales employed here. Indeed, given the peak ofTrkA function blocking antibodies. The demonstration of retrograde transport found by ourselves and others at 6 hconcentration-dependent effects of the antibodies thus post treatment, and assuming that gene expression, ortho-suggests that this is the main reason for the differences grade transport and incorporation of new receptors requiresbetween our findings and those of Curtis et al. and for their at least a further 6–8 h, it seems unlikely that the presentconclusion that the transport of NGF is relatively insensi- paradigm would be affected by altered receptor expression.tive to manipulation of p75NTR. If there were to be an effect, we would therefore predict a

The mechanisms by which p75NTR may influence small enhancement of the dose responsiveness of DTR–neuronal NGF uptake and retrograde axonal transport to NGF uptake shown in Fig. 3. Similarly, inhibition ofthe cell body are not clear. One possibility is that p75NTR receptor expression resulting from receptor blockade mayis involved, together with TrkA, in the formation of have a minor, additional depressive effect on DTR–NGFheterodimers which function as high affinity binding sites uptake, assuming that in the case of both p75NTR andfor NGF [3,4,6,7,39]. Our results support this suggestion TrkA, expression is positively correlated with neurotrophinby showing that anti-p75NTR and anti-TrkA antibodies availability (Cowen, unpublished data).each had large effects on NGF uptake, indicating that How the NGF signal is transmitted across the vastblocking one or other receptor prevented the majority of distance from the axon terminals to the neuronal cell body,NGF binding. Direct interaction between the two NGF and the major players involved in mediating this process,receptors has been demonstrated in brain tissue using still remain uncertain. Different signalling pathways mayco-immunoprecipitation [42] and in non-neuronal cells be activated within different neuronal populationsusing co-patching techniques or mutant versions of the [59,66,69]. The present results suggest that p75NTR playsNGF receptors [30,62,72]. Since NGF appears not to be a more important role in NGF signalling than was previ-internalized in cells that express only p75NTR ([30,43]; ously assumed and support the idea that an interactionsee, however, Ref. [45]), the antibody-induced reduction in between p75NTR and TrkA receptors seems to be essentialNGF uptake in sympathetic neurons may reflect a decrease for uptake of NGF in adult rat sympathetic neurons.mainly in TrkA-mediated endocytosis, occurring as aconsequence of disrupted binding of NGF to a p75NTR–TrkA complex [42]. An alternative explanation is sug- Acknowledgementsgested by Taniuchi and Johnson ([67]; see also Ref. [63])who showed that the monoclonal antibody 192IgG, which Supported by The Wellcome Trust (T.C.; project no.is directed against the extracellular domain of p75NTR, 049001), The Swedish Medical Research Council (K.P.G.),increases the affinity of NGF binding to p75NTR, leading Loo och Hans Ostermans Research Fund (K.P.G.) andto a concomitant decrease in NGF binding to TrkA and a Handlaren Hjalmar Svenssons Research Fund (K.P.G.). Wereduction in NGF–TrkA complex formation, internaliza- thank Dr Moses Chao for supplying us with the 9651tion and retrograde transport. One interpretation of our antibody against the p75 receptor. The anti-p75 Rex andfindings could therefore be that the p75NTR–antibody the anti-TrkA antibodies were a generous gift from Drcomplex, by increasing NGF binding, reduces the number Louis Reichardt. We also thank Drs J.M. Cooper and V.M.of NGF molecules available for TrkA binding and conse- Mann for help with spectrophotometry and Professor Ianquently indirectly disrupts the uptake and retrograde Hendry for helpful discussions and critical reading of thetransport of a separate NGF–TrkA complex. However, manuscript.neither of the two anti-p75NTR antibodies used in thisstudy possess a capacity to increase NGF binding to thereceptor. The possibility that our observations may be due

Referencesto a non-physiological p75NTR antisera-mediated inhibi-tion of TrkA-dependent NGF uptake and transport also

[1] T.J. Andrews, C. Thrasivoulou, W. Nesbit, T. Cowen, Target-seems unlikely in view of the findings that after an specific differences in the dendritic morphology and neuropeptideefficient receptor blockade was achieved, raised concen- content of neurons in the rat SCG during development and aging, J.trations of NGF (from 75 to 200 ng, making more ligand Comp. Neurol. 368 (1996) 33–44.

K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238 237

[2] P. Barker, E.M. Shooter, Disruption of NGF binding to the low [24] H.B.F. Dixon, R.N. Perham, Reversible blocking of amino groupsLNTRaffinity neurotrophin receptor p75 reduced NGF binding to trkA with citraconic anhydride, Biochem. J. 109 (1968) 312–314.

on PC12 cells, Neuron 13 (1994) 203–215. [25] R.T. Dobrowsky, M.H. Werner, A.M. Castellino, M.V. Chao, Y.A.[3] D.S. Battleman, A.I. Geller, M.V. Chao, HSV-1 vector mediated gene Hannun, Activation of the sphingomyelin cycle through the low

NGFRtransfer of the human nerve growth factor receptor p75 defines affinity neurotrophin receptor, Science 265 (1994) 1596–1599.high affinity NGF binding, J. Neurosci. 13 (1993) 941–951. [26] M. Ehlers, D. Kaplan, D. Price, V. Koliatsos, NGF-stimulated

[4] M. Benedetti, A. Levi, M.V. Chao, Differential expression of nerve retrograde transport of trkA in the mammalian nervous system, J.growth factor receptors leads to altered binding affinity and neuro- Cell Biol. 130 (1995) 149–156.trophin responsiveness, Proc. Natl. Acad. Sci. USA 90 (1993) [27] P. Ernfors, K.-F. Lee, R. Jaenisch, Target derived and putative local7859–7863. actions of neurotrophins in the peripheral nervous system, Prog.

[5] A. Bhattacharyya, F.L. Watson, T.A. Bradlee, S.L. Pomeroy, C.D. Brain Res. 103 (1994) 43–54.Stiles, R. Segal, Trk receptors function as rapid retrograde signal [28] J.M. Frade, A. Rodriguez-Tebar, Y.-A. Barde, Induction of cell deathcarriers in the adult nervous system, J. Neurosci. 17 (1997) 7007– by endogenous nerve growth factor through its p75 receptor, Nature7016. 383 (1996) 166–168.

[6] M. Bothwell, Functional interactions of neurotrophins and neuro- [29] W.J. Friedman, L.A. Greene, Neurotrophin signaling via Trks andtrophin receptors, Annu. Rev. Neurosci. 18 (1995) 223–253. p75, Exp. Cell Res. 253 (1999) 131–142.

[7] M. Bredesen, S. Rabizadeh, p75NTR and apoptosis: Trk-dependent[30] N. Gargano, A. Levi, S. Alema, Modulation of nerve growth factor

and Trk-independent effects, Trends Neurosci. 20 (1997) 287–290.internalization by direct interaction between p75 and TrkA re-NGFR[8] M. Canossa, J.L. Twiss, A.N. Verity, E.M. Shooter, p75 andceptors, J. Neurosci. Res. 50 (1997) 1–12.NGFRTrkA receptors collaborate to rapidly activate a p75 -associated

[31] I. Gavazzi, K.L. Railton, O. Eng, T. Cowen, Responsiveness ofprotein kinase, EMBO J. 15 (1996) 3369–3376.sympathetic and sensory irideal nerves to NGF treatment in young[9] B.D. Carter, C. Kaltschmidt, B. Kaltschmidt, N. Offenhauser, M.R.and aged rats, Neurobiol. Aging 22 (2001) 287–296.Bohm, P.A. Baeuerle, Y.-A. Barde, Selective activation of NF-kappa

[32] J.J. Gentry, P. Casaccia-Bonnefil, B.D. Carter, Nerve growth factorB by nerve growth factor through the neurotrophin receptor p75,activation of nuclear factor kappa B through its p75 receptor is anScience 272 (1996) 542–545.anti-apoptotic signal in RN22 schwannoma cells, J. Biol. Chem. 275[10] P. Casaccia-Bonnefil, B.D. Carter, R.T. Dobrowsky, M.V. Chao,(2000) 7558–7565.Death of oligodendrocytes mediated by the interaction of nerve

[33] M.L. Grimes, J. Zhou, E. Beattie, E.C. Yuen, D.E. Hall, J.S. Valletta,growth factor with its receptor p75, Nature 383 (1996) 716–719.K.S. Topp, J.H. LaVail, N.W. Bunnett, W.C. Mobley, Endocytosis of[11] P. Casaccia-Bonnefil, C.H. Gu, G. Khursigara, M.V. Chao, p75activated TrkA: evidence that nerve growth factor induces formationneurotrophin receptor as a modulator of survival and death deci-of signaling endosomes, J. Neurosci. 16 (1996) 7950–7964.sions, Microsc. Res. Tech. 45 (1999) 217–224.

[34] M.L. Grimes, E. Beattie, W.M. Mobley, A signaling organelle[12] M.V. Chao, Neurotrophin receptors: a window into neuronal dif-containing the nerve growth factor-activated receptor tyrosineferentiation, Neuron 9 (1992) 583–593.kinase, TrkA, Proc. Natl. Acad. Sci. USA 94 (1997) 9909–9914.[13] M.V. Chao, The p75 neurotrophin receptor, J. Neurobiol. 25 (1994)

[35] P.A. Hantzopoulos, C. Suri, D.J. Glass, M.P. Goldfarb, G.D.1373–1385.Yancopoulos, The low affinity NGF receptor, p75, can collaborate[14] M.V. Chao, B.L. Hempstead, p75 and Trk: a two-receptor system,with each of the Trks to potentiate functional responses to theTrends Neurosci. 18 (1995) 321–326.neurotrophins, Neuron 13 (1994) 187–207.[15] D.O. Clary, G. Weskamp, L.R. Austin, L.F. Reichardt, TrkA cross-

[36] S.M.W. Harrison, M.E. Jones, S. Uecker, K.M. Albers, K.E.linking mimics neuronal responses to nerve growth factor, Mol.Kudrycki, B.M. Davis, Levels of nerve growth factor andBiol. Cell 5 (1994) 549–563.neurotrophin-3 are affected differentially by the presence of p75 in[16] G.E.A. Coome, J. Elliott, M.D. Kawaja, Sympathetic and sensorysympathetic neurons in vivo, J. Comp. Neurol. 424 (2000) 99–110.axons invade the brains of nerve growth factor transgenic mice in

[37] D.S. Hartman, M. McCormack, R. Schnubenel, C. Hertel, Multiplethe absence of p75NTR expression, Exp. Neurol. 149 (1998) 284–trkA proteins in PC12 cells bind NGF with a slow association rate,294.J. Biol. Chem. 267 (1992) 24516–24522.[17] T. Cowen, I. Gavazzi, Plasticity in adult and ageing sympathetic

[38] R.P. Haugland, Coupling of monoclonal antibodies with fluoro-neurons, Prog. Neurobiol. 54 (1998) 249–288.phores, in: W.C. Davies (Ed.), Monoclonal Antibody Protocols.[18] C. Crowley, S.D. Spencer, M.C. Nishimura, K.S. Chen, S.Methods in Molecular Biology, Vol. 45, Humana Press, Totowa, NJ,Pittsmeek, M.P. Armanini, L.H. Ling, S.B. Mcmahon, D.L. Shelton,1995, p. 214.A.D. Levinson, H.S. Phillips, Mice lacking nerve growth factor

[39] B.M. Hempstead, D. Martin-Zanca, D.R. Kaplan, L.F. Parada, M.V.display perinatal loss of sensory and sympathetic neurons yetChao, High affinity binding requires co-expression of the trk proto-develop basal forebrain cholinergic neurons, Cell 76 (1994) 1001–oncogene and the low affinity NGF receptor, Nature 350 (1991)1011.678–683.[19] R. Curtis, K.M. Adrian, J.L. Stark, J.S. Park, D.L. Compton, G.

[40] I.A. Hendry, In vivo administration of nerve growth factor, in: R.A.Weskamp, L.J. Huber, M.V. Chao, R. Jaenisch, K.-F. Lee, R.M.Rush (Ed.), Nerve Growth Factors, Wiley, New York, 1989, pp.Lindsay, P.S. DiStefano, Differential role of the low affinity193–212.neurotrophin receptor (p75) in retrograde axonal transport of the

[41] R. Heumann, M.E. Schwab, H. Thoenen, A second messengerneurotrophins, Neuron 14 (1995) 1201–1211.required for nerve growth factor biological activity?, Nature 292[20] A.M. Davies, The neurotrophic hypothesis: where does it stand?,(1981) 838–840.Phil. Trans. R. Soc. Lond. B 351 (1996) 389–394.

[42] L.J. Huber, M.V. Chao, A potential interaction of p75 and trkA NGF[21] A.M. Davies, K.-F. Lee, R. Jaenisch, p75-deficient trigeminalreceptors revealed by affinity crosslinking and immunoprecipitation,sensory neurons have an altered response to NGF but not to otherJ. Neurosci. Res. 40 (1995) 557–563.neurotrophins, Neuron 11 (1993) 1–20.

[43] S. Jing, P. Tapley, M. Barbacid, Nerve growth factor mediates signal[22] G. Dechant, A. Rodriguez-Tebar, Y.-A. Barde, Neurotrophin re-transduction through trk homodimer receptors, Neuron 9 (1992)ceptors, Prog. Neurobiol. 42 (1994) 347–352.1067–1079.[23] J.D. Delcroix, D.R. Tomlinson, P. Fernyhough, Diabetes and ax-

otomy-induced deficits in retrograde axonal transport of nerve [44] D. Johnson, A. Lanahan, C.R. Buck, A. Sehgal, C. Morgan, E.growth factor correlate with decreased levels of p75LNTR protein in Mercer, M. Bothwell, M.V. Chao, Expression and structure of thelumbar dorsal root ganglia, Mol. Brain Res. 51 (1997) 82–90. human NGF receptor, Cell 47 (1986) 545–554.

238 K.P. Gatzinsky et al. / Brain Research 920 (2001) 226 –238

[45] P. Kahle, C. Hertel, Nerve growth factor receptor on rat glial cell [61] H. Rohrer, T. Schafer, S. Korsching, H. Thoenen, Internalization oflines. Evidence for NGF internalization via p75NGFR, J. Biol. nerve growth factor by pheochromocytoma PC12 cells: absence ofChem. 267 (1992) 13917–13923. transfer to the nucleus, J. Neurosci. 2 (1982) 687–697.

[46] P. Kahle, P.A. Barker, E.M. Shooter, C. Hertel, p75 nerve growth [62] A.H. Ross, M.C. Daou, C.A. McKinnon, P.J. Condon, M.B. Lach-trkAfactor receptor modulates p140 kinase activity, but not ligand yankar, R.M. Stephens, D.R. Kaplan, D.E. Wolf, The neurotrophin

internalization, in PC12 cells, J. Neurosci. Res. 38 (1994) 599–606. receptor, gp75, forms a complex with the receptor tyrosine kinase[47] M.T. Kasaian, J.W. Jacobberger, K.E. Neet, Flow cytometric analy- TrkA, J. Cell Biol. 132 (1996) 945–953.

sis of fluorescein-labelled nerve growth factor binding to A875 [63] S. Rossner, U. Ueberham, R. Schliebs, J.R. Perez-Polo, V. Bigl, p75human melanoma cells, Exp. Cell Res. 210 (1994) 77–85. and TrkA receptor signaling independently regulate amyloid pre-

[48] M.D. Kawaja, Sympathetic and sensory innervation of the ex- cursor protein mRNA expression, isoform composition, and proteinNTRtracerebral vasculature: Roles for p75 neuronal expression and secretion in PC12 cells, J. Neurochem. 71 (1998) 757–766.

nerve growth factor, J. Neurosci. Res. 52 (1998) 295–306. [64] M.E. Schwab, Ultrastructural localization of a nerve growth factor-[49] C. Lachance, D.J. Belliveau, P.A. Barker, Blocking nerve growth horseradish peroxidase (NGF-HRP) coupling product after retro-

factor binding to the p75 neurotrophin receptor on sympathetic grade axonal transport in adrenergic neurons, Brain Res. 130 (1977)neurons transiently reduces trkA activation but does not affect 190–196.neuronal survival, Neuroscience 81 (1997) 861–871. [65] R. Segal, M. Greenberg, Intracellular signaling pathways activated

[50] K.-F. Lee, A.M. Davies, R. Jaenisch, p75-deficient embryonic dorsal by neurotrophic factors, Annu. Rev. Neurosci. 19 (1996) 463–489.root sensory and neonatal sympathetic neurons display a decreased [66] D.L. Senger, R.B. Campenot, Rapid retrograde tyrosine phosphoryl-sensitivity to NGF, Development 120 (1994) 1027–1033. ation of trkA and other proteins in rat sympathetic neurons in

[51] A. Levi, Y. Shechter, E.J. Neufeld, J. Schlessinger, Mobility, compartmented cultures, J. Cell Biol. 138 (1997) 411–421.clustering, and transport of nerve growth factor in embryonal [67] N. Taniuchi, E.M. Johnson, Characterization of the binding prop-sensory cells and in a sympathetic neuronal cell line, Proc. Natl. erties and retrograde axonal transport of a monoclonal antibodyAcad. Sci. USA 77 (1980) 3469–3473. directed against the rat nerve growth factor receptor, J. Cell Biol.

[52] S. Maliartchouk, H.U. Saragovi, Optimal nerve growth factor 101 (1985) 1100–1106.trophic signals mediated by synergy of TrkA and p75 receptor- [68] C. Thrasivoulou, T. Cowen, Regulation of rat sympathetic nervespecific ligands, J. Neurosci. 17 (1997) 6031–6037. density by target tissues and NGF in maturity and old age, Eur. J.

[53] S.O. Meakin, E.M. Shooter, Molecular investigations on the high Neurosci. 7 (1995) 381–387.affinity nerve growth factor receptor, Neuron 6 (1991) 153–163. [69] D.R. Ure, R.B. Campenot, Retrograde transport and steady-state

125[54] J. Meldolesi, C. Sciorati, E. Clementi, The p75 receptor: first distribution of I-nerve growth factor in rat sympathetic neurons ininsights into the transduction mechanisms leading to either cell compartmented cultures, J. Neurosci. 17 (1997) 1282–1290.death or survival, Trends Pharmacol. Sci. 21 (2000) 242–243. [70] J. Verdi, S. Birren, C. Ibanez, M. Benedetti, M.V. Chao, D.J.

[55] F.D. Miller, A. Speelman, T.C. Mathew, J. Fabian, E. Chang, C. Anderson, p75 LNGFR regulates Trk signal transduction and NGF-Pozniak, J.G. Toma, Nerve growth factor derived from terminals induced neuronal differentiation in MAH cells, Neuron 12 (1994)selectively increases the ratio of p75 to trka NGF receptors on 733–745.mature sympathetic neurons, Dev. Biol. 161 (1994) 206–217. [71] G. Weskamp, L. Reichardt, Evidence that biological activity of NGF

[56] N. Orike, C. Thrasivoulou, T. Cowen, Serum free culture of is mediated through a novel subclass of high affinity receptors,dissociated, purified adult and ageing sympathetic neurons and Neuron 6 (1991) 649–663.quantitative assays of growth and survival, J. Neurosci. Methods 106 [72] D.E. Wolf, C.A. McKinnon, M.C. Daou, M.R. Stephens, D.R.(2001) 153–160. Kaplan, A.H. Ross, Interaction with TrkA immobilizes gp75 in the

[57] N. Orike, C. Thrasivoulou, A. Wrigley, T. Cowen, Differential high affinity nerve growth factor receptor complex, J. Biol. Chem.regulation of survival and growth in adult sympathetic neurons: an 270 (1995) 2133–2138.in vitro study of neurotrophin responsiveness, J. Neurobiol. 47 [73] S. Wyatt, A.M. Davies, Regulation of expression of mRNAs(2001) 295–305. encoding the nerve growth factor receptors p75 and trka in develop-

[58] M.J. Radeke, T.P. Misko, C. Hsu, L.A. Herzenberg, E.M. Shooter, ing sensory neurons, Development 119 (1993) 635–648.Gene transfer and molecular cloning of the rat nerve growth factor [74] S. Wyatt, A.M. Davies, Regulation of nerve growth factor receptorreceptor, Nature 325 (1987) 593–597. gene expression in sympathetic neurons during development, J. Cell

[59] A.J. Reynolds, S.E. Bartlett, I.A. Hendry, Signalling events regulat- Biol. 130 (1995) 1435–1446.125ing the retrograde axonal transport of I-nerve growth factor in [75] S.O. Yoon, P. Casaccia-Bonnefil, B. Carter, M.V. Chao, Competitive

vivo, Brain Res. 798 (1998) 67–74. signalling between trkA and p75 nerve growth factor receptors[60] A. Riccio, B.A. Pierchala, C.L. Ciarallo, D.A. Ginty, An NGF- determines cell survival, J. Neurosci. 18 (1998) 3273–3281.

TrkA-mediated retrograde signal to transcription factor CREB insympathetic neurons, Science 277 (1997) 1097–1100.