Neurotrophin receptors TrkB.T1 and p75NTR cooperate inmodulating both functional and structural plasticity inmature hippocampal neurons

K. Michaelsen,1 M. Zagrebelsky,1 J. Berndt-Huch,1 M. Polack,1 A. Buschler,1 M. Sendtner2 and M. Korte1

1TU Braunschweig, Zoological Institute, Cellular Neurobiology, Spielmannstr. 7, D 38106 Braunschweig, Germany2Institute for Clinical Neurobiology, University of Wurzburg, Wurzburg, Germany

Keywords: hippocampus, LTD, LTP, neurotrophin receptors, p75NTR, spines

Abstract

Tropomyosin-related kinase (Trk) receptors modulate neuronal structure and function both during development and in the maturenervous system. Interestingly, TrkB and TrkC are expressed as full-length and as truncated splice variants. The cellular function ofthe kinase-lacking isoforms remains so far unclear. We investigated the role of the truncated receptor TrkB.T1 in the hippocampus oftransgenic mice overexpressing this splice variant by analyzing both neuronal structure and function. We observed an impairment inactivity-dependent synaptic plasticity as indicated by deficits in long-term potentiation and long-term depression in acute hippocampalslices of transgenic TrkB.T1 mice. In addition, dendritic complexity and spine density were significantly altered in TrkB.T1-over-expressing CA1 neurons. We found that the effect of TrkB.T1 overexpression differs between subgroups of CA1 neurons.Remarkably, overexpression of p75NTR and its activation by chemical induction of long-term depression in slice cultures rescued theTrkB.T1-dependent morphological alterations specifically in one of the two subgroups observed. These findings suggest that theTrkB.T1 and p75NTR receptor signaling systems might be cross-linked. Our findings demonstrate that TrkB.T1 regulates the functionand the structure of mature pyramidal neurons. In addition, we showed that the ratio of expression levels of p75NTR and TrkB.T1 playsan important role in modulating dendritic architecture and synaptic plasticity in the adult rodent hippocampus, and, indeed, that theendogenous expression patterns of both receptors change reciprocally over time. We therefore propose a new function of TrkB.T1 asbeing dominant-negative to p75NTR.

Introduction

Ligand binding of the neurotrophin tropomyosin-related kinase (Trk)receptors activates a series of different, well-described signalingcascades, and thereby controls neuronal survival, differentiation,axonal outgrowth, and synaptic plasticity (Huang & Reichardt, 2003)[for reviews, see Patapoutian & Reichardt (2001)]. Interestingly,alternative splicing generates truncated, kinase-lacking forms of boththe TrkB and TrkC receptors [TrkB.T1 (T1), TrkB.T2 and TrkC.T1](Klein et al., 1990; Middlemas et al., 1991; Tsoulfas et al., 1993).Since the discovery of the truncated Trk receptors, their physiologicalfunction in vivo has been a matter of debate. Interestingly, the ratiobetween the expression levels of full-length and truncated receptorsdepends both on the developmental stage and on the neuronal cell typeinvestigated (Allendoerfer et al., 1994; Escandon et al., 1994; Fryeret al., 1996). Overexpression studies in ferret cortical slices haveshown that the ratio of T1 to full-length TrkB (TrkB.TK+) can act as aswitch between two different modes of dendritic growth – de novo

growth and elongation of already existing dendrites (Yacoubian & Lo,2000).The role of the truncated Trk receptors was initially identified as

restricting the action of the full-length receptors through theformation of heterodimers. Indeed, T1 overexpression studiessuggest a role of the truncated receptor as a dominant-negativeinhibitor of TrkB.TK+ signaling (Eide et al., 1996; Drake et al.,1999; Haapasalo et al., 2001, 2002; Ohira et al., 2001; Lahteinenet al., 2002). Furthermore, T1 has been shown to act as a brain-derived neurotrophic factor (BDNF)-scavenging receptor (Kleinet al., 1990; Middlemas et al., 1991; Biffo et al., 1995; Eide et al.,1996; Saarelainen et al., 2000a), thereby indirectly regulating theaction of TrkB.TK+. Deletion of T1 partially rescues BDNFhaploinsufficiency, suggesting that it may, in fact, negativelymodulate BDNF–TrkB.TK+ signaling in vivo (Carim-Todd et al.,2009).A possible signaling role has been proposed independently of

TrkB.TK+ (Baxter et al., 1997; Rose et al., 2003; Ohira et al., 2005;Cheng et al., 2007). In addition, Rose et al. (2003) showed that T1 isthe predominant variant expressed in astrocytes, and leads to calciumrelease via protein kinase A and inositol trisphosphate.

Correspondence: M. Korte, as above.E-mail: m.korte@tu-bs.de

Received 4 December 2009, revised 19 August 2010, accepted 31 August 2010

European Journal of Neuroscience, pp. 1–12, 2010 doi:10.1111/j.1460-9568.2010.07460.x

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing Ltd

European Journal of Neuroscience

Furthermore, there is ample evidence that T1 is able to interact withanother neurotrophin receptor – p75NTR (Hartmann et al., 2004). Indeveloping hippocampal neurons, overexpression of T1 has beenshown to induce the formation of dendritic filopodia. Remarkably, thisoutgrowth of protrusions seems to require the action of p75NTR

(Hartmann et al., 2004).In this study, we aimed at characterizing the functional role of T1 in

the structure and function of mature hippocampal neurons, and toclarify its active (signaling) or passive (dominant-negative) mecha-nism of action.

Materials and methods

Mouse strains

Transgenic mice expressing T1 under the Thy1.2 promoter weregenerated by the group of Eero Castren (University of Helsinki), andall procedures were approved by the Experimental Animal EthicsCommittee of the National Laboratory Animal Center, University ofKuopio, Finland (Saarelainen et al., 2000c; Koponen et al., 2004a,b).Mice were housed in metal cages under standard animal roomconditions (12 : 12-h light ⁄ dark cycle, ambient temperature of 23 �C,and free access to food and water).Briefly, cDNA of T1 was tagged N-terminally with an eight amino

acid FLAG peptide and inserted into the murine Thy1.2 expressioncassette to direct expression to postnatal neurons. The constructs weretransferred by pronucleus injection into embryos from CD2F1 females(BALB ⁄ c · DBA ⁄ 2) mated with CD2F1 males. Male and femalemice were used for the experiments. BALB ⁄ c · DBA ⁄ 2 mice wereused as wild-type (WT) controls.Expression of the transgene starts at postnatal day (P)10, increases

until P18, and remains stable thereafter (Saarelainen et al. 2000b).Strong expression of the transgene can be detected in hippocampalpyramidal neurons, dentate granule cells and pyramidal neurons of thecerebral cortex (Saarelainen et al. 2000b).For all experiments, the mice were genotyped by PCR with

genomic DNA extracted from tail pieces. PCR was performed withthy1.2 and trkB specific primers (thy1.2, CTCCCACTTCCTTGG-CTT; TrkB, GCCCCACGTAAGCTTCGA (Koponen et al., 2004b).All other experiments were performed with cultures of C57BL ⁄ 6mice.

Cell culture techniques

Organotypic hippocampal slice cultures were prepared as previouslydescribed (Stoppini et al., 1991). P5 ⁄ 6 mice were decapitated, theskulls were removed, and the dorsal halves of the brain weretransferred to ice-cold GBSS (Gey’s Balanced Salt Solution). Four-hundred-micrometer transversal hippocampal slices were cut with aMcIllwain tissue chopper, and kept at 4 �C for 30 min in GBSS.Subsequent cultivation was performed on tissue culture inserts[Millicell; 0.4-lm pore size, hydrophilic poly(tetrafluoroethylene)membrane], four slices each insert, at 37 �C, 5% CO2 and 99%humidity. To reduce the number of non-neuronal cells, antimitoticdrugs (uridine, cytosine-b-d-arabinofuranoside hydrochloride, and5-fluoro-2¢-deoxyuridine) were applied for 24 h 3 days after prepara-tion. Subsequently, 50% of the medium was changed every 3 days.Primary hippocampal cultures were prepared from C57Bl ⁄ 6 mice

[embryonic day (E)18]. The tissue was incubated for 30 min intrypsin–EDTA at 37 �C, and then mechanically dissociated with aPasteur pipette. Cells were plated at high density (105 per coverslip)on poly (l-lysine)-coated coverslips (13 mm) and kept in Neurobasal

medium (Gibco) supplemented with 2% B27 (Gibco) and 0.5 mm

Glutamax at 37 �C, 5% CO2, and 99% humidity. The cell culturemedium was not changed.

Transfection of hippocampal neurons

Organotypic hippocampal slice cultures were transfected at 14 days invitro (DIV) with the Helios Gene Gun system of Bio-Rad. Goldmicrocarriers 600 nm in diameter were shot onto the slice with heliumat a pressure of 100 lb ⁄ in2. To avoid damage to the tissue, cultureinserts with a pore diameter of 3 lm were used as filters.Bullets for transfection were prepared according to the manufac-

turer’s instructions (BioRad). Briefly, 2 lg ⁄ mg plasmid DNA wascoated onto 0.6-lm gold microcarriers. To visualize neuronalmorphology in detail, a membrane targeted form of enhanced greenfluorescent protein (EGFP) [farnesylated form of EGFP (FGFP)] wastransfected at a 1 : 2 ratio. For coating of DNA onto the goldmicrocarriers, Ca2Cl precipitation was performed (Wellmann et al.,1999; O’Brien & Lummis, 2006). Primary cultures of hippocampalneurons made at E18 were transfected at 17 DIV with Lipofecta-mine2000, following the manufacturer’s instructions.

Electrophysiological recordings

Four-hundred-micrometer transverse hippocampal slices of TrkB.T1transgenic or WT CD2F1 mice (P14–P19) were prepared andmaintained by standard procedures, as described previously (Gartneret al., 2006) – (in mM) 124 NaCl; 2.95 KCl; 1.25 KH2PO4; 2 MgSO4;26 NaHCO3; 2.5 CaCl2; 10 d-glucose; 95% O2; 5% CO2. Monopolartungsten electrodes were used for stimulation in the CA3 Schaffer-collateral region. Synaptic field potentials were elicited with afrequency of 0.1 Hz, and responses were recorded with borosilicateglass electrodes, with a resistance between 5 and 15 MX, filled with3 m NaCl in the dendritic region (stratum radiatum) of the CA1pyramidal neurons. The slope of the excitatory postsynaptic potentialwas calculated and used to measure synaptic strength. Long-termdepression (LTD) was induced by low-frequency stimulation (LFS)with 900 1-Hz pulses for 15 min. Long-term potentiation (LTP) wasinduced with a tetanus of 3 · 30 pulses (100 Hz, 60-ls duration, 5-sinterstimulus interval). Data were collected with a program written inlabview (National Instruments, Austin, TX, USA). All measurementswere carried out in a strictly blind fashion with regard to the genotypeof the mice used. Statistical analysis was performed with a pairedStudent’s t-test (two-tailed and two-sample unequal variance). Resultswere considered to be significantly different if the average response55–60 min after the induction of LTD or LTP showed P-values below0.05.

Immunocytochemistry

Organotypic and primary hippocampal cultures were fixed overnightat 4 �C with 4% paraformaldehyde in phosphate buffer. Washing wasperformed at room temperature in phosphate-buffered saline (PBS) for1 h, and this was followed by blocking and permeabilization at roomtemperature in PBS containing 1% bovine serum albumin, 10% goatserum and 0.2% Triton X-100 for 1 h. All primary antibodies wereincubated at 4 �C in PBS containing 10% goat serum. Rabbitpolyclonal anti-human p75NTR [anti-human p75 pAb G3231 (Pro-mega, Mannheim, Germany), directed against the cytoplasmic domainof p75NTR] was used at a dilution of 1 : 500 (3 days) for organotypiccultures and at a dilution of 1 : 4000 on primary hippocampal neurons

2 K. Michaelsen et al.

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

(overnight). Polyclonal rabbit anti-T1 antibody [rabbit anti-TrkB(TK)), C-terminus, gp95, C13 ⁄ sc-119 (Santa Cruz Biotechnol-ogy, Santa Cruz, CA, USA), directed against the intracellular domainspecific for T1] was diluted 1 : 1000 (overnight) for primary neurons.Secondary anti-mouse or anti-rabbit antibodies conjugated with Cy2,Cy3 or Cy5 (Jackson ImmunoResearch, PA, USA) were incubated at adilution of 1 : 500 in PBS for 2 h at room temperature, and this wasfollowed by washing in PBS for 1 h. Subsequently, the cultures weremounted in aqueous medium containing anti-fading agents (Biomeda,CA, USA). Golgi staining was performed with the FD RapidGolgiStain kit (FD NeuroTechnologies, MD, USA). Briefly,2-month-old T1 transgenic mice or WT control mice were rapidlyanesthetized (CO2) and decapitated. The brains were removed, brieflyrinsed with distilled water to remove the remaining blood, andprocessed according to the manufacturer’s instructions. One hundredand fifty-micrometer transversal hippocampal slices were cut with avibratome (Leica VT 1000S). Slices were mounted on gelatine-coatedslides, stained according to the manufacturer’s instructions, andimaged by light microscopy as described below for neurons derivedfrom organotypic cultures.

Image acquisition and analysis

Neurons were imaged using an Axioplan 2 microscope (Zeiss)equipped with an ApoTome module (Zeiss) controlled by theaxiovision software. For imaging of the entire CA1 neurons in theorganotypic cultures, several z-stacks of 1 lm were acquired with a· 20 0.8 NA Plan-APO objective (Zeiss). In primary hippocampalcultures, plain fluorescence images of hippocampal neurons wereacquired. For analysis of spine density in organotypic cultures, parts ofbasal and both proximal and distal apical dendrites were imaged at ahigher magnification with a · 63 1.4 NA Plan-APO oil immersionobjective (Zeiss) and a z-stack thickness of 0.5 lm. In primaryneurons, spines of secondary and tertiary dendrites in the mid-part ofthe dendritic tree were imaged with the same settings as above. For theexpression time-course of T1 and p75NTR, images were taken with aZeiss 510 META confocal microscope and a · 40 1.3 NA oilimmersion objective.

Morphological analysis was performed with neurolucida andneurolucida explorer software (Microbrightfield).

Statistical analysis

Obtained values for Sholl analysis (Sholl, 1953), spine density ordendrite number and length were exported to Excel and Graphpad

prism for statistical analysis with a paired Student’s t-test (two-tailedand two-sample unequal variance); significance was set at P < 0.05.For the Sholl analysis, data significance was tested point by point, andstatistical significance was only considered if more than two adjacentpoints showed P-values below 0.05. All data are shown asmean ± standard error of the mean.

Results

Overexpression of T1 alters the morphology of maturepyramidal neurons

For study of the role of T1 in regulating the architecture of matureCA1 pyramidal neurons, organotypic hippocampal slice cultures oftransgenic mice expressing T1 under the Thy1.2 promoter were used(Saarelainen et al., 2000a; Koponen et al., 2004b). Strong expression

of the transgene has been detected in hippocampal pyramidal neurons,dentate granule cells and pyramidal neurons of the cerebral cortex(Koponen et al., 2004b).For detailed morphological analysis, hippocampal slice cultures of

transgenic T1 mice were transfected with FGFP, by the use of particle-mediated gene transfer. Mature CA1 pyramidal neurons overexpress-ing T1 showed marked simplification of the mid-apical dendrite(Fig. 1A, arrows). Detailed Sholl analysis of the apical dendrites ofT1-overexpressing neurons confirmed this impression. Dendriticcomplexity was significantly reduced at a distance between 330 and430 lm from the soma (Fig. 1A and B). Interestingly, morphologicalalterations were restricted to the distal half of the apical dendrite – adecrease in complexity of the mid-apical compartment and anelongation of the apical tufts. In addition, the apical dendrite displayedstriking elongation (Fig. 1A).As it is known that even pyramidal neurons of the same subgroup

(e.g. CA1) can display morphological differences from each other, weinvestigated whether certain subpopulations of CA1 neurons wouldreact differently to T1 overexpression. Among 17 neurons analyzed,seven displayed Sholl values between 600 and 700 lm from the somafor the apical dendrite, whereas only two of 15 neurons among thecontrol cells exceeded this length. Therefore, we divided theT1-overexpressing pyramidal neurons into two subgroups – thoseshowing Sholl values for the region above 600 lm from the soma(Fig. 2, group A); and those that did not exceed this length (Fig. 2,group B). Interestingly, both groups displayed significant differenceswhen compared to control cells that were, in each group, restricted todifferent compartments. Specifically, group A neurons showedsignificant elongation of the distal apical tuft, whereas group Bneurons were significantly more complex at approximately 200 lmfrom the soma (Fig. 2).As a next step, spine density counts were performed on T1-

overexpressing CA1 neurons. Total spine density was increased inT1-overexpressing neurons (Fig. 1D). However, only the basaldendritic compartment showed a significantly increased number ofdendritic protrusions (Fig. 1D; Table S1). Again, as with the analysisof dendritic structure, we divided the neurons overexpressing T1 intotwo groups, A and B, and analyzed the spines separately.Interestingly, both groups showed significant changes in spinenumber when compared to control cells that were again restricted todifferent compartments, as was observed for dendritic complexity(Fig. 2A1 and B1; Table S1). Spine analysis of group A neuronsrevealed a significantly increased spine number in the basal dendriticcompartment, whereas group B neurons showed a significantincrease in the number of dendritic protrusions only in the distalapical tuft.These data and a previous report showing an increase in filopodia-

like protrusions following T1 overexpression in developing primaryhippocampal neurons (Hartmann et al., 2004) prompted us toinvestigate the spine subtype composition in T1 transgenic slicecultures. By means of morphological analysis, three different spinesubtypes can be classified: stubby, thin and mushroom spines(Fig. 3A) (Chicurel & Harris, 1992; Koh et al., 2002). This classi-fication is based on measurements of the total spine length as well asthe ratio between the spine head and the spine neck diameters, andprovides an objective, observer-independent classification of spines(Fig. 3A). In line with previous reports (Zagrebelsky et al., 2005;Chakravarthy et al., 2006), mushroom spines comprised the predom-inant spine type in mature pyramidal neurons analyzed in this study(Fig. 3). T1-overexpressing neurons showed no alterations in spinesubtype composition (Fig. 3) in the basal and in the proximal apicalcompartments as compared with control cells.

Role of T1 and p75 in processes of plasticity 3

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

We chose organotypic slice cultures for the analysis of T1 functionbecause of the superior properties of these cultures in terms of spinetype composition and analysis of dendritic complexity. As a next step,we addressed the question of whether the phenotype of T1 overex-pression can also be detected in vivo. Therefore, we stained the wholebrains of 8-week-old mice with the Golgi method, and analyzeddendritic complexity and spine density in the hippocampal CA1subfield (Fig. S1). The phenotype of a significant increase in spinedensity of the basal compartment was also found in the intact brains oftransgenic T1 mice in comparison with control cells (P < 0.05).Because of the high density of labeled cells, we were only able toanalyze the dendritic structure of the apical dendrite for a distance ofup to approximately 300–400 lm from the soma. However, thedendritic phenotype of reduced dendritic complexity in the mid-apicaldendrite and an increase in the complexity of the basal dendrites wasconfirmed by in vivo experiments.

Overexpression of T1 impairs synaptic plasticity in CA1pyramidal neurons

The observation of an altered dendritic architecture in T1 transgenicslice cultures led us to consider whether overexpression of T1 would

also affect synaptic plasticity in the hippocampal CA1 region. This is ofspecial importance, because although the BDNF–TrkB system hasbeen shown to regulate positive synaptic plasticity (Gottmann et al.,2009) [for reviews, see Poo (2001)], the p75NTR receptor has beenimplicated in negative synaptic plasticity (Rosch et al., 2005; Wooet al., 2005), and T1 overexpression might interfere with bothsignaling pathways (Biffo et al., 1995; Eide et al., 1996; Haapasaloet al., 2001; Hartmann et al., 2004; Carim-Todd et al., 2009). To thisend, LTP and LTD were induced in the Schaffer-collateral pathway bythe use of high-frequency stimulation or LFS protocols, respectively.LTP could be evoked in slices of both WTand T1-overexpressing mice.However, in the latter, we found LTP maintenance to be significantlyimpaired (Fig. 4A). Sixty minutes after high-frequency stimulation, themean field excitatory postsynaptic potential (fEPSP) slope (as apercentage of the baseline) of transgenic T1 mice was 139.0 ± 4.6% ascompared with 160.7 ± 6.2% in WT mice (P < 0.001, t-test).The overall capacity to induce changes in synaptic strength seems to

be compromised in T1-overexpressing mice, as we found that LTDwas also impaired. Again, as was the case for LTP, the maintenancebut not the induction of LTD was significantly impaired in transgenicmice overexpressing T1 (Fig. 4B). In T1 mice, the mean slope 60 minafter LFS was 93.8 ± 2.1% (as a percentage of baseline) as comparedwith 82.8 ± 2.1% in WT mice (P < 0.01, t-test).

WT Transgenic T1

WT T1

A

DØ Complexity

0.00.20.40.60.81.01.21.4 *

Spi

nes/μm

den

drite

Dist. Prox.Apical Basal Total

C Spine density

9 11 10 810 11 10 9

apical

Basal

Apical

Basal

Apical

0

50

100

150

200

250

0 50 100 150 200 2500

5

10

15

20

WTT1

μm from soma

μm from soma

# of

inte

rsec

tions

# of

inte

rsec

tions

# of

inte

rsec

tions

0 100 200 300 400 500 600 700 8000

2

4

6

8

10

WTT1

*

B

Fig. 1. Overexpression of T1 alters neuronal morphology. (A) CA1 neurons (17 DIV) of transgenic mice overexpressing T1 as compared with WT neurons. T1-overexpressing neurons show a reduction in dendritic complexity of the mid-apical dendrite (arrows) as compared with control cells. Images were captured frommaximum intensity projections. Scale bar – 100 lm. (B) Sholl analysis (basal and apical dendrites) of T1-overexpressing CA1 neurons (n = 17) as compared withcontrol cells (n = 15). Overexpression of T1 significantly decreases the dendritic complexity of the apical dendrite at a distance between 330 and 430 lm from thesoma; in addition, T1-overexpressing neurons are longer than WT cells. (C) Total dendritic complexity and (D) spine density of CA1 pyramidal neuronsoverexpressing T1; overexpression of T1 significantly increases the spine number of the basal compartment, whereas the total dendritic complexity is unaltered inT1-overexpressing neurons as compared with control cells; *P < 0.05.

4 K. Michaelsen et al.

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

T1-mediated changes involve p75NTR

Neurotrophins modulate synaptic transmission. BDNF and itsreceptor TrkB have been shown to be crucial for the long-lasting

enhancement of synaptic efficacy (LTP) (Korte et al., 1995; Lu,2003) [for reviews, see Poo (2001)]. In contrast, studies in p75NTR

knockout (KO) mice suggest an important role of this receptor in

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Spin

es/µ

m d

endr

ite

Spin

es/µ

m d

endr

ite

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Basal

Spine density

Apical

Spine density*

*

A

A1

B

B1

BasalApicalProx.Dist.

Total BasalApicalProx.Dist.

Total

59 11 10 86 5 4 9 5 11 10 85 5 5

TrkB.T1 group A TrkB.T1 group BControl Control

0 50 100 150 200 2500

5

10

15

20

0 50 100

100

150 200

300 500 700

2500

5

10

15

20

0 200 400 600 8000

2

4

6

8

10

0 200 400 600 8000

2

4

6

8

10 Apical

Basal

**

µm from soma

µm from soma µm from soma

# of

inte

rsec

tions

# of

inte

rsec

tions

# of

inte

rsec

tions

# of

inte

rsec

tions

Fig. 2. Overexpression of T1 has different effects on subgroups of CA1 pyramidal neurons. (A) Neurons overexpressing T1 that displayed Sholl values for adistance of 600–700 lm from the soma were considered as group A and analyzed separately (n = 7) from control neurons (n = 15). (B) Neurons that displayed noSholl values exceeding a distance of 600 lm from the soma were considered as group B (n = 10). Sholl analysis reveals that the two subgroups show significantdifferences in the different dendritic compartments when compared to control cells. (A1) Spine density of group A CA1 neurons. (B1) Spine density of group Bneurons. Spine numbers are also differentially affected by overexpression of T1 in the two subgroups; *P < 0.05.

Mushroom Stubby0

20

40

60

80

Thin0

20

40

60

80

% o

f tot

al s

pine

s

% o

f tot

al s

pine

s

Mushroom Stubby Thin

Basal Proximal

II) Mushroom: diameter head/diameter neck ≥ 1.5I) Stubby: diameter head/diameter neck < 1.5, shorter than 1 μm

III) Thin: diameter head/diameter neck ~ 1, longer than 1 μm

I (Stubby)

II (Mushroom)

III (Thin)

I

II

III

WT T1

A BApical

Fig. 3. Spine subtype composition is not altered in CA1 neurons overexpressing T1. (A) Representative image of a dendrite from a WT cells showing the threespine types. Scale bar – 2 lm. Spine types can also be seen in the illustration at the right (the criteria used for classification are indicated in B, and are based on thelength of the spine and the ratio of the diameters of the head and neck) – type I, stubby spine; type II, mushroom spine; and type III, thin spine. (B) Proportions of thethree spine types (basal and proximal apical dendrites) of T1-overexpressing CA1 neurons (n = 5) as compared with control cells (n = 5); spine subtype compositionis not affected by overexpression of T1.

Role of T1 and p75 in processes of plasticity 5

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

maintaining LTD (Rosch et al., 2005; Woo et al., 2005). In ourstudy, the observed impairment of LTP in T1-overexpressing micemight be induced either by its dominant-negative effect onTrkB.TK+ or by its ligand-scavenging function for BDNF. However,the additional impairment in LTD indicated a possible role of T1beyond its function as the dominant-negative form of TrkB.TK+, andour next experiments were therefore aimed at further characterizingthe possible mechanisms involved. In this context, it has beenreported before that T1 might interact with p75NTR to modulateneuronal morphology (Hartmann et al., 2004). In addition, T1 hasbeen shown to functionally interact with p75NTR to induce neuronaldifferentiation (Hapner et al., 1998). Another indication that p75NTR

is involved in modulating the phenotype of T1 transgenic mice wasprovided by our analysis of dendritic structure. Specifically, thegrouping of T1-overexpressing neurons revealed that the dendriticmorphology of group A neurons resembled that of p75NTR KO cells,as described previously (Zagrebelsky et al., 2005), with an increasein complexity in the distal apical dendrite. Therefore, we exploredwhether the structural changes that we observed following overex-pression of T1 were indeed linked to p75NTR. We addressed thisquestion by comparing the morphological changes occurring in WTand T1-expressing neurons biolistically transfected with p75NTR inorganotypic slice cultures. Neurons overexpressing p75NTR or bothtransgenic T1 and p75NTR showed no signs of degeneration in theform of swellings or retraction bulbs (Fig. 6A). Overexpression ofp75NTR alone led to a reduction in both dendritic complexity andspine density of CA1 hippocampal pyramidal cells (Fig. S2A and B;Table S1). This phenotype is in line with the role of p75NTR as anegative modulator of dendritic architecture (Zagrebelsky et al.,2005).

Remarkably, the phenotype of the T1-overexpressing neurons of thegroup A subtype could no longer be detected in cells overexpressingboth T1 and p75NTR, as none of the 11 neurons analyzed displayedSholl values for the region above 600 lm from the soma (Fig. 5).Indeed, a detailed Sholl analysis revealed that the dendritic complexityof neurons overexpressing both T1 and p75NTR was no longersignificantly different from that of WT cells (Fig. 5A and B, apical).Similarly, the analysis of spine density in CA1 neurons coexpressingT1 and p75NTR revealed that the increase in spine number of the basaldendrites in group A T1-overexpressing neurons (Figs 1D and 2A1)could no longer be detected in neurons overexpressing T1 and p75NTR

(Fig. 5C; Table S1). Interestingly, the significant increase in spinedensity of the distal apical tuft found in T1-overexpressing cells of thegroup B type (Fig. 2B1) (Fig. 5C; Table S1) was also observed inneurons overexpressing both T1 and p75NTR.To determine whether an induction of activity known to involve the

action of endogenous p75NTR would have a comparable effect, webath applied 20 lm N-methyl-d-aspartic acid (NMDA) for 10 min inorder to induce LTD, and fixed the slices 2 days later. This protocolhas been reported before to induce a p75NTR-dependent form of LTD(Woo et al., 2005), and we confirmed the reliability of the approach inour system with fEPSP recordings (data not shown). The overalldendritic structure of NMDA-treated neurons was normal as comparedwith control cells, suggesting that the stimulation did not lead to anydegeneration of CA1 cells. Sholl analysis of basal and apical dendritesrevealed only a slight (not significant) decrease in the dendriticcomplexity of NMDA-treated neurons as compared with control cells,which was restricted to the basal dendrites (Fig. S2C). A detailedanalysis showed that the chemical induction of LTD resulted in theloss of spines both in the basal dendrites and in the proximal apicaldendrites, as spine density counts revealed a significant reduction forboth compartments as compared with control cells (Fig. S2D; TableS1).Remarkably, Sholl analysis of T1-overexpressing neurons treated

with 20 lm NMDA showed no difference as compared with WT cells(Fig. 6A). Again, as was the case for neurons overexpressing p75NTR

and T1, cells of the group A type were not detected, as only one of 14neurons displayed Sholl values for the region above 600 lm from thesoma. This observation indicates that the changes in dendriticmorphology observed in T1-overexpressing neurons of the group Atype (Figs 1A and B and 2A and A1) could be completely restored bythe induction of LTD (Fig. 6A). Interestingly, the phenotype of asignificant increase in spine number of the distal apical dendrite thatcould be observed in T1-overexpressing neurons of the group B typewas also found in T1-overexpressing neurons treated with NMDA(Fig. 5B; Table S1), indicating that this phenotype could not berestored.Comparable treatment with 60 mm KCl instead of NMDA to

increase overall synaptic activity, while not inducing synapticplasticity, in slice cultures showed that neurons of the group A typedid not react to this stimulus, in striking contrast to the plasticity-inducing NMDA treatment (Fig. S3).Taken together, these observations show that the overexpression of

T1 has different effects in subpopulations of CA1 pyramidal neurons,with a group of cells showing a comparable phenotype to that ofp75NTR KO neurons. This indicates a dominant-negative role for T1for p75NTR. Further support of this hypothesis is the fact that LTD isnegatively affected in slices from T1 transgenic mice. This is also thecase in p75NTR KO mice. Moreover, overexpression of p75NTR oractivation of the endogenous receptor in transgenic T1 slice culturesleads to a restoration of the dendritic structure, making it indistin-guishable from that of WT cells.

0 20 40 60 80

60

80

100

120

140

Slop

e (%

of b

asel

ine)

Slop

e (%

of b

asel

ine)

0 20 40 60 8050

100

150

200

250

Time (min)

Time (min)

A

B

TrkB.T1 n = 9Wt n = 16

TrkB.T1 n = 21Wt n = 26

10 ms1

mV

Wt

TrkB.T1

10 ms

1 m

V

Wt

TrkB.T1

Fig. 4. Synaptic plasticity is impaired in mice overexpressing T1. (A) LTP(induced by stimulation with 100-Hz tetanus) is significantly impaired in miceoverexpressing T1 (WT, 26 slices ⁄ 13 mice; T1, 21 slices ⁄ nine mice). SinglefEPSPs from both genotypes before and 60 min after the tetanus are shown atthe right. (B) LTD (induced by LFS, 15 min, 1 Hz) is significantly impaired inmice overexpressing T1 (WT, 16 slices ⁄ six mice; T1, nine slices ⁄ six mice),Single fEPSPs from both genotypes during baseline and 60 min after theapplication of LFS are shown at the right.

6 K. Michaelsen et al.

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

0 50 100 150 200 2500

5

10

15

20

0 100 200 300 400 500 600 700 8000

2

4

6

8

10

WTT1T1 & NMDA

0.00.20.40.60.81.01.21.4 *

*

Dist. Prox.Apical Basal Total

WTT1T1 & NMDA

A Apical Spine densityBasal B

T1 & NMDA

WT T1

9 11 10 810 11 10 99 9 10 9

# of

inte

rsec

tions

# of

inte

rsec

tions

µm from soma µm from soma

Spin

es/µ

m d

endr

ite

Fig. 6. Induction of LTD can compensate for T1-induced morphological alterations. (A) Sholl analysis (basal and apical dendrites) and (B) spine density of CA1pyramidal neurons of transgenic mice overexpressing T1 as well as neurons overexpressing T1 48 h after the chemical induction of LTD by a 10-min application of20 lm NMDA. The induction of LTD reverses morphological changes caused by the transgenic expression of T1 both at the level of dendrites and at the level ofspines; *P < 0.05.

µm from soma

µm from soma

# of

inte

rsec

tions

# of

inte

rsec

tions

0 100 200 300 400 500 600 700 8000

2

4

6

8

10

WTT1T1 & p75NTR

0.00.20.40.60.81.01.21.4 *

*

*Sp

ines

/µm

den

drite

Dist. Prox.Apical Basal Total

Apical

Spine density

WT T1 T1&p75

T1WT

T1 & p75NTR

9 11 10 810 11 10 99 9 8 9

T1 T1 & p75NTR

AB

C

Basal

Apical

0 50 100 150 200 2500

5

10

15

20Basal

Fig. 5. Expression of p75NTR in CA1 neurons overexpressing T1 compensates for morphological alterations mediated by both receptor types. (A) Maximumintensity projections of a CA1 pyramidal neuron overexpressing transgenic T1 (organotypic cultures, DIV 17) and a cell overexpressing T1 and p75NTR (anti-p75NTR labeled in red). Both neurons were transfected with FGFP to allow detailed morphological analysis. Overexpression of both receptors compensates for theT1-mediated reduction in dendritic complexity in the mid-apical dendrite. Scale bar – 100 lm. (B) Sholl analysis (basal and apical dendrites) of CA1 neuronsoverexpressing transgenic T1 or transgenic T1 together with p75NTR. (C) Spine density of control cells, neurons overexpressing T1 and neurons overexpressingboth T1 and p75NTR. Note that the significant changes in spine density caused by overexpression of T1 are reversed by concomitant overexpression of p75NTR;*P < 0.05.

Role of T1 and p75 in processes of plasticity 7

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

Mutual inhibition of T1 and p75NTR depends on the extracellulardomain of T1 but not on ligand binding

Trk receptors and p75NTR are coexpressed in many neuronalpopulations. In this study, we performed immunohistochemistry onprimary hippocampal cultures to follow the developmentally regulatedendogenous expression levels of both T1 and p75NTR (Figs S4 and S5).Indeed, both receptors are coexpressed in these neurons; however,their expression levels vary in an age-dependent manner. p75NTR

immunostaining reveals high expression levels in developing neuronsat two DIV and seven DIV that decrease during maturation (Fig. S4).On the other hand, T1 levels were found to be regulated in theopposite direction. T1 expression was low at early developmentalstages and increased after 14 DIV (Fig. S5).To further characterize the presumably inhibitory effect of T1 on

p75NTR, T1 deletion mutants (Haapasalo et al., 1999) were transfectedin primary hippocampal neurons. In a first step, overexpression of T1and p75NTR in primary hippocampal neurons confirmed the resultsobtained in organotypic slice cultures shown in Fig. 6. Indeed,overexpression of T1 reduced the dendritic complexity of primaryhippocampal neurons, as indicated by a reduction in the number ofdendritic endings as compared with control cells (Fig. 7B; 71 ± 3dendrites in the control group; 58 ± 3 dendrites in T1-overexpressing

neurons; P = 0.004). Overexpression of p75NTR alone also led to areduction in dendritic complexity (55 ± 4 dendrites in p75NTR-overexpressing cells; P = 0.003). The dendritic complexity of neuronsoverexpressing both T1 and p75NTR was indistinguishable from that ofcontrol cells (Fig. 7B; 63 ± 5 dendrites). A similar effect wasobserved after the activation of endogenous p75NTR via the chemicalinduction of LTD (20 lm NMDA, 10 min) in primary hippocampalneurons overexpressing T1. Remarkably, dendritic complexity in thesecells was not significantly different from that in control cells (Fig. 7C;61 ± 5 dendrites in neurons treated with NMDA; 70 ± 4 dendrites inT1-overexpressing cells treated with NMDA).The expression of T1 lacking either the extracellular domain (T1-

EC) or the intracellular domain (T1-IC) resulted in a significantsimplification of the dendritic tree (Fig. 7D; 57 ± 4 dendrites inneurons overexpressing T1-IC, as compared with control cells,P = 0.008; 58 ± 3 dendrites in neurons overexpressing T1-EC, ascompared with control cells, P = 0.006). This is especially interestingbecause the intracellular domain of T1 comprises only 23 amino acids.However, coexpression of T1 deletion mutants and p75NTR revealedthat the mutual compensation effect of the two receptors observedbefore (Fig. 5B) depends on the extracellular domain of T1. Specif-ically, neurons coexpressing T1-EC and p75NTR showed a signifi-cantly reduced dendritic tree and therefore no compensation. However,

fcherry p75NTR T1

fcherr

yp7

5 T1

T1 & p75

T1 & p75NTR

A

B

0

20

40

60

80

** **

# of

den

driti

c en

ding

s

20 16 24 11

fcherr

y T1T1-I

CT1-E

C

T1 domainsD

0

20

40

60

80

** * *

20 16 17 22

fcherr

y T1NMDA

T1 & NMDA

T1 & NMDAC

0

20

40

60

80

**

20 24 14 16 0

20

40

60

80

** **

fcherr

yp7

5

T1-IC

& p75

T1-EC

& p75

T1-IC/EC & p75NTRE

20 24 16 14

# of

den

driti

c en

ding

s

# of

den

driti

c en

ding

s

# of

den

driti

c en

ding

s

Fig. 7. The compensatory effects of coexpression of T1 and p75NTR on neuronal morphology are dependent on the extracellular domain of T1. (A) Fluorescenceimages of a primary hippocampal neuron transfected with fcherry (control), p75NTR and T1. Scale bar – 50 lm. (B) Histogram showing the number of dendriticendings of primary hippocampal neurons transfected with fcherry, p75NTR, T1, or p75NTR and T1. In primary hippocampal cultures, overexpression of T1 or p75NTR

significantly reduces dendritic complexity; coexpression of both neurotrophin receptors prevents this reduction. (C) Histogram showing the number of dendriticendings of neurons transfected with fcherry or T1, and ⁄ or treated with 20 lm NMDA for 10 min and fixed 48 h later. As in organotypic cultures, the chemicalinduction of LTD prevents the T1-dependent reduction in dendritic complexity. (D) Histogram showing the number of dendritic endings of neurons transfected withfcherry, T1 or T1 deletion mutants lacking the extracellular domain (T1-EC) or the intracellular domain (T1-IC). The T1-mediated reduction in dendritic complexityis dependent on the extracellular as well as the intracellular domain of T1. (E) Histogram showing the number of dendritic endings of cells transfected with fcherry,p75NTR or T1 deletion mutants and p75NTR. The extracellular domain of T1 is necessary and sufficient to rescue the p75NTR-dependent reduction in dendriticcomplexity; *P < 0.05, **P < 0.005.

8 K. Michaelsen et al.

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

the concomitant expression of T1-IC and p75NTR induced nomorphological alterations (Fig. 7E; 56 ± 3 dendrites in neuronsoverexpressing T1-EC and p75NTR, P = 0.001; 69 ± 4 dendrites inneurons overexpressing T1-IC and p75NTR).

In a last set of experiments, we invstigated whether the inhibitoryeffect depends on the presence of the TrkB ligand BDNF. BDNF-scavenging antibodies (provided by M. Sendtner, Wurzburg,Germany) (Wiese et al., 2007) were used to block BDNF in primaryhippocampal neurons. Remarkably, blocking of BDNF had no effecton the inhibitory effect of T1 on p75NTR (Fig. S6). Neuronsoverexpressing T1 and p75NTR treated with function-blocking anti-BDNF antibody were not significantly different from WT cells(Fig. S6; 56.6 ± 3 dendrites in control neurons vs. 56.6 ± 4 dendritesin neurons overexpressing T1 and p75NTR and treated with anti-BDNFantibody).

In summary, experiments in primary hippocampal neurons con-firmed an inhibitory effect of T1 on p75NTR with regard tomorphological alterations. Furthermore, we showed that this effectdepends on the extracellular domain of T1 but not on binding of theligand BDNF.

Discussion

The fact that T1 is upregulated during late developmental stagessuggests a particular role for this receptor in the adult nervoussystem. Interestingly, we found both synaptic plasticity – LTP andLTD – and dendritic architecture to be altered in the hippocampusof transgenic T1-overexpressing mice. Further experiments aimed atrevealing a mechanism responsible for the T1-mediated structuralchanges identified a functional inhibitory effect of T1 on p75NTR.We showed that T1 seems to play different roles in subsets of CA1pyramidal neurons, with a group of T1-overexpressing neuronsmorphologically resembling p75NTR KO cells. Indeed, when p75NTR

was overexpressed in T1 transgenic CA1 neurons, the phenotype ofthis group of cells was abolished. Interestingly, this inhibitory effectdepends on the extracellular domain of T1 but not on theneurotrophin BDNF. The expression of T1 and the ratio ofexpression levels to those of the other neurotrophin receptors –TrkB and p75NTR – might therefore be particularly used in the adultcentral nervous system to tightly control neuronal function andstructure.

T1 as a modulator of both neuronal architecture and functionin the adult nervous system

Whereas TrkB.TK+ constitutes the predominant isoform early indevelopment, the expression level of T1 rises slowly to exceed theamount of TrkB.TK+ in the adult brain (Fryer et al., 1996). It is still amatter of debate whether this rise in T1 expression is predominantlyattributable to expression in glial cells; however, we and others havepreviously shown that T1 is indeed expressed in neurons (Allendoerferet al., 1994; Escandon et al., 1994; Armanini et al., 1995; Fryer et al.,1996; Kryl et al., 1999; Ohira et al., 1999, 2004; Silhol et al., 2005,2007).

The overexpression of T1 in this study was restricted to neurons. Inmature CA1 pyramidal neurons, an increase in the expression level ofT1 resulted in different morphological alterations in distinct subsets ofthe cells analyzed. Remarkably, both subgroups of T1-overexpressingneurons showed significant differences, indicating that T1 playsdifferent roles in neurons. The factors that might account for thesedifferences are not yet clear; however, the experiments performed in

this study give a first hint that the endogenous expression levels of allthree types of neurotrophin receptor might be responsible for thedifferences in the phenotypes of transgenic T1-overexpressingneurons. It is known that pyramidal neurons of the hippocampuscan vary in the expression levels of the different neurotrophin receptortypes – p75NTR vs. TrkB receptors (Friedman, 2000; Zagrebelskyet al., 2005). Friedman et al. (2000) reported that 40% of thehippocampal principal neurons show no expression of TrkB.TK+. Inthis neuronal subpopulation, overexpression of T1 might, in fact, havedifferent consequences than in neurons expressing both TrkB andp75NTR or only TrkB. Indeed, approximately 40% of our T1transgenic neurons showed a phenotype comparable to that ofp75NTR KO cells, indicating that T1 might indeed block p75NTR

signaling in these cells. Two observations made in this study furtherunderline this functional model. First, we observed a deficit in NMDAreceptor-dependent LTD in transgenic T1 mice; this is a form ofsynaptic plasticity that depends on p75NTR. Second, overexpression ofp75NTR and activation of the endogenous p75NTR receptor abolishedthis phenotype in organotypic slice cultures, whereas a stimulationparadigm that simply increased neuronal activity (60 mm KCl) had noeffect on the morphology of these cells. Interestingly, the phenotype ofthe second subset of cells overexpressing T1 (group B) was notrestored by the overexpression of p75NTR. This finding supports thenotion that this could be a different neuronal subpopulation withdistinct expression levels of neurotrophin receptors. Therefore, thephenotype of the overexpression of T1 might have different under-lying mechanisms in this population, for example blockade ofTrkB.TK+. Interestingly, this would confirm the findings of previousstudies (Haapasalo et al., 2001, 2002). Indeed, this second phenotypeof an increase in the spine density of the distal apical tuft (group B)was restored in T1 slices stimulated with 60 mm KCl, a treatment thatis known to increase basal neuronal activity. This treatment might leadto an activity-dependent increase in the expression or the membranetargeting of TrkB.TK+. It has been shown previously that differentdepolarizing stimuli can increase the expression of TrkB or p75NTR

mRNA [for a review, see Nagappan & Lu (2005)].The question of whether the structural changes observed in T1

transgenic animals are directly linked to functional alterations in theability to undergo synaptic plasticity remains unanswered. The factthat structural changes could be restored either by overexpression ofp75NTR or by chemical induction of LTD, which is known to involvethe action of p75NTR, is in favor of two independent phenomena –functional and structural changes – that could, however, be eventuallyconnected by converging signaling cascades. In this respect, it is ofspecial interest that both T1 and p75NTR can signal to the actincytoskeleton, for instance via modulating the activity of the smallGTPase RhoA (Yamashita et al., 1999; Ohira et al., 2005), and wehave also shown that there is a link between p75NTR and the actin-binding protein profilin (Michaelsen et al., 2010). For BDNF and itsreceptor TrkB, accumulating evidence indicates that this receptorsystem could indeed translate functional changes in synaptic efficacy(LTP) (Korte et al., 1995) into more persistent structural changes(e.g. spine growth) (Tanaka et al., 2008). Whether p75NTR couldperform a similar function for the other – negative – side of synapticplasticity still needs to be clarified. p75NTR has indeed been shown tomediate opposite functional changes – LTD (Rosch et al., 2005) –and, in a separate study, to be involved in dendritic spine retraction(Zagrebelsky et al., 2005). It is intriguing to speculate that p75NTR isactually involved in translating negative functional changes intostructural ones, because this would suggest that the T1 could indeedemerge as a central mediator by tightly restricting ⁄ balancing theaction of both neurotrophin receptors.

Role of T1 and p75 in processes of plasticity 9

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

In summary, we provide evidence for a role of T1 in modulatingboth functional and structural plasticity in mature neurons. This goesfar beyond what has been shown earlier, such as that the truncatedsplice variant can alter dendritic complexity (Yacoubian & Lo, 2000)and spine shape (Hartmann et al., 2004; Chakravarthy et al., 2006) inneurons. We can now show that the precise action of T1 depends onthe subset of neurons and on the T1 ⁄ p75NTR expression ratio. Thismight also explain differences between studies, as the outgrowth offilopodia in hippocampal neurons was reported in one case (Hartmannet al., 2004), whereas Chakravarthy et al. (2006) detected changes inspine number in cortical neurons but not in hippocampal cells. In thisstudy, the authors used sparse expression of T1 that might have beenrestricted to only one subset of neurons.

T1 as an inhibitor of p75NTR?

Trk receptors – both full-length and truncated TrkB – have beenshown to physically interact with p75NTR (Bibel et al., 1999). In thisstudy, overexpression of p75NTR and chemical induction of LTD inorganotypic slice cultures of T1 animals inhibited some of themorphological alterations induced by overexpression of T1. One canargue, indeed, that one subset of T1-overexpressing neurons mighthave reacted with apoptosis to overexpression of p75NTR or chemicalinduction of LTD, and therefore could not be analyzed. However, wefound no increase in the number of apoptotic cells in slice culturestransfected with both neurotrophin receptors.Our findings do indeed suggest that T1 and p75NTR functions may

be linked to each other, resulting in mutual inhibition of the tworeceptors, thus suggesting the possibilities of either a physical protein–protein interaction or of partly converging signaling pathways. Theexpression of T1 deletion mutants in primary hippocampal neuronsrevealed that the intracellular domain of T1 is dispensable for therescuing effect of T1 on p75NTR-mediated structural changes. Thisresult does not support the hypothesis that the observed mutualinhibition of both receptors is attributable to an overlap in signalingpathways, for example at the level of activation of different membersof the Rho GTPase family.Another important finding of the current study is that the mutual

inhibition of the two receptors occurred independently of the ligandBDNF. Remarkably, this is in line with earlier studies showing that theT1-mediated phenotype of an increase in the number of filopodia wasnot dependent on neurotrophin binding (Haapasalo et al., 1999;Hartmann et al., 2004). Taken together, these results point to afunction of T1 and p75NTR as modulators of neuronal morphologywithout the actual involvement of neurotrophins, and thereby broadenour picture of their range of action.The results of the current study shed new light on the function of

T1. We provide evidence for a role of T1 not only as the dominant-negative form of TrkB.TK+ but also as a receptor that can inhibitp75NTR signaling. T1 expression and transport to the membrane couldbe used to tightly regulate the action of neurotrophin receptors,especially as T1 is predominantly expressed in the adult brain, wherechanges in neuronal structure should be restricted to regions ofneuronal plasticity. Although heterodimers of T1 and TrkB.TK+ havebeen previously reported (Ohira et al., 2001), the nature of theinteraction between p75NTR and T1 remains elusive. T1 and p75NTR

might interact – and block each other’s action – directly or throughadaptor molecules clustering neurotrophin receptors (Chang et al.,2004). What kind of interaction occurs between these two receptors –T1 and p75NTR – and what could account for their complex crosstalkstill need to be clarified. However, different models exist, such as

receptor clustering in lipid rafts or interactions via different intracel-lular adaptor molecules (Barker, 2007; Wehrman et al., 2007). Onecandidate could be the ankyrin-rich membrane-spanning molecule(ARMS), which has been shown to interact both with p75NTR and withTrk receptors (Chang et al., 2004). Moreover, it has been shown thatARMS is involved in the regulation of dendritic branching and spinestability in mature neurons of the cortex and hippocampus (Wu et al.,2009). ARMS could therefore provide a link between p75NTR and T1,mediating the inhibitory effect of T1 on p75NTR.Overall, we demonstrate here that T1 regulates the function and the

structure of mature pyramidal neurons, and that the expression levelsof p75NTR and T1 play an important role in modulating dendriticarchitecture and synaptic plasticity in the adult rodent hippocampus. Inline with this, the endogenous expression patterns of both receptors doindeed change reciprocally over time. We therefore propose a newfunction of T1 as being dominant-negative to p75NTR.

Supporting Information

Additional supporting information may be found in the online versionof this article:Fig. S1. Analysis of dendritic structure in acute slices of TrkB.T1-overexpressing mice.Fig. S2. Structural alterations in CA1 neurons following chemicalinduction of LTD or overexpression of p75NTR.Fig. S3. Blocking of BDNF does not prevent the mutual inhibition ofTrkB.T1 and p75NTR.Fig. S4. Expression of p75NTR decreases in the course of development.Fig. S5. Expression of TrkB.T1 is developmentally upregulated.Fig. S6. Blocking of BDNF does not prevent the mutual inhibition ofTrkB.T1 and p75NTR.Table S1. Spine numbers of organotypic cultures.Please note: As a service to our authors and readers, this journalprovides supporting information supplied by the authors. Suchmaterials are peer-reviewed and may be re-organized for onlinedelivery, but are not copy-edited or typeset by Wiley-Blackwell.Technical support issues arising from supporting information (otherthan missing files) should be addressed to the authors.

Acknowledgements

We kindly thank Eero Castren for providing us with T1 transgenic mice and T1deletion mutants and Anita Remus for experimental help. We would like tothank Diane Mundil and Jasmin Will for their outstanding technical assistance.This work was supported by the DFG (M. Korte).

Abbreviations

ARMS, ankyrin-rich membrane-spanning molecule; BDNF, brain-derivedneurotrophic factor; DIV, days in vitro; E, embryonic day; EGFP, enhancedgreen fluorescent protein; fEPSP, field excitatory postsynaptic potential; FGFP,farnesylated form of enhanced green fluorescent protein; KO, knockout; LFS,low-frequency stimulation; LTD, long-term depression; LTP, long-term poten-tiation; NMDA, N-methyl-d-aspartic acid; P, postnatal day; PBS, phosphate-buffered saline; T1, TrkB.T1; T1-EC, TrkB.T1 lacking the extracellulardomain; T1-IC, TrkB.T1 lacking the intracellular domain; Trk, tropomyosin-related kinase; TrkB.TK+, full-length TrkB; WT, wild-type.

References

Allendoerfer, K.L., Cabelli, R.J., Escandon, E., Kaplan, D.R., Nikolics, K. &Shatz, C.J. (1994) Regulation of neurotrophin receptors during the matura-tion of the mammalian visual system. J. Neurosci., 14, 1795–1811.

10 K. Michaelsen et al.

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

Armanini, M.P., McMahon, S.B., Sutherland, J., Shelton, D.L. & Phillips, H.S.(1995) Truncated and catalytic isoforms of trkB are co-expressed in neuronsof rat and mouse CNS. Eur. J. Neurosci., 7, 1403–1409.

Barker, P.A. (2007) High affinity not in the vicinity? Neuron, 53, 1–4.Baxter, G.T., Radeke, M.J., Kuo, R.C., Makrides, V., Hinkle, B., Hoang, R.,

Medina-Selby, A., Coit, D., Valenzuela, P. & Feinstein, S.C. (1997) Signaltransduction mediated by the truncated trkB receptor isoforms, trkB.T1 andtrkB.T2. J. Neurosci., 17, 2683–2690.

Bibel, M., Hoppe, E. & Barde, Y.A. (1999) Biochemical and functionalinteractions between the neurotrophin receptors trk and p75NTR. EMBOJ., 18, 616–622.

Biffo, S., Offenhauser, N., Carter, B.D. & Barde, Y.A. (1995) Selective bindingand internalisation by truncated receptors restrict the availability of BDNFduring development. Development, 121, 2461–2470.

Carim-Todd, L., Bath, K.G., Fulgenzi, G., Yanpallewar, S., Jing, D., Barrick,C.A., Becker, J., Buckley, H., Dorsey, S.G., Lee, F.S. & Tessarollo, L.(2009) Endogenous truncated TrkB.T1 receptor regulates neuronal com-plexity and TrkB kinase receptor function in vivo. J. Neurosci., 29, 678–685.

Chakravarthy, S., Saiepour, M.H., Bence, M., Perry, S., Hartman, R., Couey,J.J., Mansvelder, H.D. & Levelt, C.N. (2006) Postsynaptic TrkB signalinghas distinct roles in spine maintenance in adult visual cortex andhippocampus. Proc. Natl Acad. Sci. USA, 103, 1071–1076.

Chang, M.S., Arevalo, J.C. & Chao, M.V. (2004) Ternary complex with Trk,p75, and an ankyrin-rich membrane spanning protein. J. Neurosci. Res., 78,186–192.

Cheng, A., Coksaygan, T., Tang, H., Khatri, R., Balice-Gordon, R.J., Rao, M.S.& Mattson, M.P. (2007) Truncated tyrosine kinase B brain-derivedneurotrophic factor receptor directs cortical neural stem cells to a glial cellfate by a novel signaling mechanism. J. Neurochem., 100, 1515–1530.

Chicurel, M.E. & Harris, K.M. (1992) Three-dimensional analysis of thestructure and composition of CA3 branched dendritic spines and theirsynaptic relationships with mossy fiber boutons in the rat hippocampus.J. Comp. Neurol., 325, 169–182.

Drake, C.T., Milner, T.A. & Patterson, S.L. (1999) Ultrastructural localizationof full-length trkB immunoreactivity in rat hippocampus suggests multipleroles in modulating activity-dependent synaptic plasticity. J. Neurosci., 19,8009–8026.

Eide, F.F., Vining, E.R., Eide, B.L., Zang, K., Wang, X.Y. & Reichardt, L.F.(1996) Naturally occurring truncated trkB receptors have dominant inhib-itory effects on brain-derived neurotrophic factor signaling. J. Neurosci., 16,3123–3129.

Escandon, E., Soppet, D., Rosenthal, A., Mendoza-Ramirez, J.L., Szonyi, E.,Burton, L.E., Henderson, C.E., Parada, L.F. & Nikolics, K. (1994)Regulation of neurotrophin receptor expression during embryonic andpostnatal development. J. Neurosci., 14, 2054–2068.

Friedman, W.J. (2000) Neurotrophins induce death of hippocampal neurons viathe p75 receptor. J. Neurosci., 20, 6340–6346.

Fryer, R.H., Kaplan, D.R., Feinstein, S.C., Radeke, M.J., Grayson, D.R. &Kromer, L.F. (1996) Developmental and mature expression of full-lengthand truncated TrkB receptors in the rat forebrain. J. Comp. Neurol., 374,21–40.

Gartner, A., Polnau, D.G., Staiger, V., Sciarretta, C., Minichiello, L., Thoenen,H., Bonhoeffer, T. & Korte, M. (2006) Hippocampal long-term potentiationis supported by presynaptic and postsynaptic tyrosine receptor kinaseB-mediated phospholipase C gamma signaling. J. Neurosci., 26, 3496–3504.

Gottmann, K., Mittmann, T. & Lessmann, V. (2009) BDNF signaling in theformation, maturation and plasticity of glutamatergic and GABAergicsynapses. Exp. Brain Res., 199, 203–234.

Haapasalo, A., Saarelainen, T., Moshnyakov, M., Arumae, U., Kiema, T.R.,Saarma, M., Wong, G. & Castren, E. (1999) Expression of the naturallyoccurring truncated trkB neurotrophin receptor induces outgrowth offilopodia and processes in neuroblastoma cells. Oncogene, 18, 1285–1296.

Haapasalo, A., Koponen, E., Hoppe, E., Wong, G. & Castren, E. (2001)Truncated trkB.T1 is dominant negative inhibitor of trkB.TK+-mediated cellsurvival. Biochem. Biophys. Res. Commun., 280, 1352–1358.

Haapasalo, A., Sipola, I., Larsson, K., Akerman, K.E., Stoilov, P., Stamm, S.,Wong, G. & Castren, E. (2002) Regulation of TRKB surface expression bybrain-derived neurotrophic factor and truncated TRKB isoforms. J. Biol.Chem., 277, 43160–43167.

Hapner, S.J., Boeshore, K.L., Large, T.H. & Lefcort, F. (1998) Neuraldifferentiation promoted by truncated trkC receptors in collaboration withp75(NTR). Dev. Biol., 201, 90–100.

Hartmann, M., Brigadski, T., Erdmann, K.S., Holtmann, B., Sendtner, M.,Narz, F. & Lessmann, V. (2004) Truncated TrkB receptor-induced outgrowth

of dendritic filopodia involves the p75 neurotrophin receptor. J. Cell Sci.,117, 5803–5814.

Huang, E.J. & Reichardt, L.F. (2003) Trk receptors: roles in neuronal signaltransduction. Annu. Rev. Biochem., 72, 609–642.

Klein, R., Conway, D., Parada, L.F. & Barbacid, M. (1990) The trkB tyrosineprotein kinase gene codes for a second neurogenic receptor that lacks thecatalytic kinase domain. Cell, 61, 647–656.

Koh, I.Y., Lindquist, W.B., Zito, K., Nimchinsky, E.A. & Svoboda, K. (2002)An image analysis algorithm for dendritic spines. Neural Comput., 14, 1283–1310.

Koponen, E., Lakso, M. & Castren, E. (2004a) Overexpression of the full-length neurotrophin receptor trkB regulates the expression of plasticity-related genes in mouse brain. Brain Res. Mol. Brain Res., 130, 81–94.

Koponen, E., Voikar, V., Riekki, R., Saarelainen, T., Rauramaa, T., Rauvala, H.,Taira, T. & Castren, E. (2004b) Transgenic mice overexpressing the full-length neurotrophin receptor trkB exhibit increased activation of the trkB–PLCgamma pathway, reduced anxiety, and facilitated learning. Mol. Cell.Neurosci., 26, 166–181.

Korte, M., Carroll, P., Wolf, E., Brem, G., Thoenen, H. & Bonhoeffer, T. (1995)Hippocampal long-term potentiation is impaired in mice lacking brain-derived neurotrophic factor. Proc. Natl Acad. Sci. USA, 92, 8856–8860.

Kryl, D., Yacoubian, T., Haapasalo, A., Castren, E., Lo, D. & Barker, P.A.(1999) Subcellular localization of full-length and truncated Trk receptorisoforms in polarized neurons and epithelial cells. J. Neurosci., 19, 5823–5833.

Lahteinen, S., Pitkanen, A., Saarelainen, T., Nissinen, J., Koponen, E. &Castren, E. (2002) Decreased BDNF signalling in transgenic mice reducesepileptogenesis. Eur. J. Neurosci., 15, 721–734.

Lu, B. (2003) BDNF and activity-dependent synaptic modulation. Learn.Mem., 10, 86–98.

Michaelsen, K., Murk, K., Zagrebelsky, M., Dreznjak, A., Jockusch, B.M.,Rothkegel, M. & Korte, M. (2010) Fine-tuning of neuronal architecturerequires two profilin isoforms. Proc. Natl Acad. Sci. USA, 107, 15780–15785.

Middlemas, D.S., Lindberg, R.A. & Hunter, T. (1991) trkB, a neural receptorprotein-tyrosine kinase: evidence for a full-length and two truncatedreceptors. Mol. Cell. Biol., 11, 143–153.

Nagappan, G. & Lu, B. (2005) Activity-dependent modulation of the BDNFreceptor TrkB: mechanisms and implications. Trends Neurosci., 28, 464–471.

O’Brien, J.A. & Lummis, S.C. (2006) Biolistic transfection of neuronal culturesusing a hand-held gene gun. Nat. Protoc., 1, 977–981.

Ohira, K., Shimizu, K. & Hayashi, M. (1999) Change of expression of full-length and truncated TrkBs in the developing monkey central nervoussystem. Brain Res. Dev. Brain Res., 112, 21–29.

Ohira, K., Shimizu, K. & Hayashi, M. (2001) TrkB dimerization duringdevelopment of the prefrontal cortex of the macaque. J. Neurosci. Res., 65,463–469.

Ohira, K., Funatsu, N., Nakamura, S. & Hayashi, M. (2004) Expression ofBDNF and TrkB receptor subtypes in the postnatal developing Purkinje cellsof monkey cerebellum. Gene Expr. Patterns, 4, 257–261.

Ohira, K., Kumanogoh, H., Sahara, Y., Homma, K.J., Hirai, H., Nakamura, S.& Hayashi, M. (2005) A truncated tropomyosin-related kinase B receptor,T1, regulates glial cell morphology via Rho GDP dissociation inhibitor 1.J. Neurosci., 25, 1343–1353.

Patapoutian, A. & Reichardt, L.F. (2001) Trk receptors: mediators ofneurotrophin action. Curr. Opin. Neurobiol., 11, 272–280.

Poo, M.M. (2001) Neurotrophins as synaptic modulators. Nat. Rev. Neurosci.,2, 24–32.

Rosch, H., Schweigreiter, R., Bonhoeffer, T., Barde, Y.A. & Korte, M. (2005)The neurotrophin receptor p75NTR modulates long-term depression andregulates the expression of AMPA receptor subunits in the hippocampus.Proc. Natl Acad. Sci. USA, 102, 7362–7367.

Rose, C.R., Blum, R., Pichler, B., Lepier, A., Kafitz, K.W. & Konnerth, A.(2003) Truncated TrkB-T1 mediates neurotrophin-evoked calcium signallingin glia cells. Nature, 426, 74–78.

Saarelainen, T., Lukkarinen, J.A., Koponen, S., Grohn, O.H., Jolkkonen, J.,Koponen, E., Haapasalo, A., Alhonen, L., Wong, G., Koistinaho, J.,Kauppinen, R.A. & Castren, E. (2000a) Transgenic mice overexpressingtruncated trkB neurotrophin receptors in neurons show increased susceptibilityto cortical injury after focal cerebral ischemia.Mol. Cell. Neurosci., 16, 87–96.

Saarelainen, T., Pussinen, R., Koponen, E., Alhonen, L., Wong, G., Sirvio, J. &Castren, E. (2000b) Transgenic mice overexpressing truncated trkB neuro-trophin receptors in neurons have impaired long-term spatial memory butnormal hippocampal LTP. Synapse, 38, 102–104.

Role of T1 and p75 in processes of plasticity 11

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12

Sholl, D. (1953) Dendritic organization in the neurons of the visual and motorcortices of the cat. J. Anat., 87, 387–406.

Silhol, M., Bonnichon, V., Rage, F. & Tapia-Arancibia, L. (2005) Age-relatedchanges in brain-derived neurotrophic factor and tyrosine kinase receptorisoforms in the hippocampus and hypothalamus in male rats. Neuroscience,132, 613–624.

Silhol, M., Arancibia, S., Maurice, T. & Tapia-Arancibia, L. (2007) Spatialmemory training modifies the expression of brain-derived neurotrophicfactor tyrosine kinase receptors in young and aged rats. Neuroscience, 146,962–973.

Stoppini, L., Buchs, P.A. & Muller, D. (1991) A simple method for organotypiccultures of nervous tissue. J. Neurosci. Methods, 37, 173–182.

Tanaka, J., Horiike, Y., Matsuzaki, M., Miyazaki, T., Ellis-Davies, G.C. &Kasai, H. (2008) Protein synthesis and neurotrophin-dependent structuralplasticity of single dendritic spines. Science, 319, 1683–1687.

Tsoulfas, P., Soppet, D., Escandon, E., Tessarollo, L., Mendoza-Ramirez, J.L.,Rosenthal, A., Nikolics, K. & Parada, L.F. (1993) The rat trkC locus encodesmultiple neurogenic receptors that exhibit differential response to neurotro-phin-3 in PC12 cells. Neuron, 10, 975–990.

Wehrman, T., He, X., Raab, B., Dukipatti, A., Blau, H. & Garcia, K.C. (2007)Structural and mechanistic insights into nerve growth factor interactions withthe TrkA and p75 receptors. Neuron, 53, 25–38.

Wellmann, H., Kaltschmidt, B. & Kaltschmidt, C. (1999) Optimized protocolfor biolistic transfection of brain slices and dissociated cultured neurons witha hand-held gene gun. J. Neurosci. Methods, 92, 55–64.

Wiese, S., Jablonka, S., Holtmann, B., Orel, N., Rajagopal, R., Chao, M.V. &Sendtner, M. (2007) Adenosine receptor A2A-R contributes to motoneuronsurvival by transactivating the tyrosine kinase receptor TrkB. Proc. NatlAcad. Sci. USA, 104, 17210–17215.

Woo, N.H., Teng, H.K., Siao, C.J., Chiaruttini, C., Pang, P.T., Milner, T.A.,Hempstead, B.L. & Lu, B. (2005) Activation of p75(NTR) by proBDNFfacilitates hippocampal long-term depression. Nat. Neurosci., 8, 1069–1077.

Wu, S.H., Arevalo, J.C., Sarti, F., Tessarollo, L., Gan, W.B. & Chao, M.V.(2009) Ankyrin repeat-rich membrane spanning ⁄ Kidins220 protein regulatesdendritic branching and spine stability in vivo. Dev. Neurobiol., 69, 547–557.

Yacoubian, T.A. & Lo, D.C. (2000) Truncated and full-length TrkB receptorsregulate distinct modes of dendritic growth. Nat. Neurosci., 3, 342–349.

Yamashita, T., Tucker, K.L. & Barde, Y.A. (1999) Neurotrophin binding to thep75 receptor modulates Rho activity and axonal outgrowth. Neuron, 24,585–593.

Zagrebelsky, M., Holz, A., Dechant, G., Barde, Y.A., Bonhoeffer, T. & Korte,M. (2005) The p75 neurotrophin receptor negatively modulates dendritecomplexity and spine density in hippocampal neurons. J. Neurosci., 25,9989–9999.

12 K. Michaelsen et al.

ª 2010 The Authors. European Journal of Neuroscience ª 2010 Federation of European Neuroscience Societies and Blackwell Publishing LtdEuropean Journal of Neuroscience, 1–12