The neurotrophin receptor p75NTR modulates long-term depression and regulates the expression of AMPA receptor subunits in the hippocampus

  1. Harald Rösch*,
  2. Rüdiger Schweigreiter*,†,
  3. Tobias Bonhoeffer,
  4. Yves-Alain Barde‡, and
  5. Martin Korte§,
  1. Department of Cellular and Systems Neurobiology, Max Planck Institute of Neurobiology, D-82152 Martinsried, Germany

  1. Communicated by Hans Thoenen, Max Planck Institute of Neurobiology, Martinsried, Germany, April 5, 2005 (received for review December 23, 2004)

 
Next Section

Abstract

Neurotrophins are involved in the modulation of synaptic transmission, including the induction of long-term potentiation (LTP) through the receptor TrkB. Because previous studies have revealed a bidirectional mode of neurotrophin action by virtue of signaling through either the neurotrophin receptor p75NTR or the Trk receptors, we tested the hypothesis that p75NTR is important for longterm depression (LTD) to occur. Although LTP was found to be unaffected in hippocampal slices of two different strains of mice carrying mutations of the p75NTR gene, hippocampal LTD was impaired in both p75NTR-deficient mouse strains. Furthermore, the expression levels of two (RS)-α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor subunits, GluR2 and GluR3, but not GluR1 or GluR4, were found to be significantly altered in the hippocampus of p75NTR-deficient mice. These results implicate p75NTR in activity-dependent synaptic plasticity and extend the concept of functional antagonism of the neurotrophin signaling system.

Neurotrophins form a small family of dimeric, secretory proteins that exert a broad spectrum of functions on vertebrate neurons. In particular, they are well known for regulating cell fate and cell shape (1, 2). They transduce their effects by binding to two different classes of receptor proteins, the receptor tyrosine kinases of the Trk family (3) and the neurotrophin receptor p75NTR (4, 5). This dual receptor system allows for the transduction of a wide array of signals after ligand binding, which can even be antagonistic. Although the best known trophic functions of neurotrophins are mediated by one of the three Trk receptors, p75NTR has been implicated in the death of neurons in a variety of CNS areas, including the retina (6), the developing spinal cord (7), and the septal nucleus (8). More recently, p75NTR has been linked with inhibition of axonal elongation (9, 10). Conversely, Trk receptors are essential for neurotrophin-promoted neurite growth (3).

Beyond the regulation of cell fate and shape, neurotrophins and their receptors have emerged as major modulators of synaptic transmission and plasticity. Neurotrophins are secreted by neurons in an activity-dependent fashion (11-13), and numerous studies have indicated a crucial role for BDNF and its receptor TrkB in hippocampal long-term potentiation (LTP). Lack of BDNF, due to either gene or protein inactivation, leads to a profound inhibition of LTP (14-17), whereas addition of BDNF to hippocampal slices isolated from WT animals results in a long-lasting enhancement of synaptic transmission (18). This effect of BDNF on LTP is thought to be mediated by TrkB because this receptor is required for the establishment of LTP (19, 20).

Much less is known about the role of p75NTR in synaptic plasticity. Blocking p75NTR with antibodies does not interfere with the induction of LTP in adult mice (20). However, deletion of p75NTR has been shown to improve spatial learning (21). Furthermore, TNF-α binding to the TNF receptor (TNFR), which, like p75NTR, is a member of the TNFR superfamily, enhances synaptic efficacy by increasing surface expression of (RS)-α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors (22). Finally, p75NTR was found to mediate a rapid switch in neurotransmitter release in individual sympathetic neurons acting on cardiac myocytes (23). Stimulating the neurons with nerve growth factor (NGF) promotes the release of norepinephrine, resulting in an increased twitching frequency of myocytes, whereas application of BDNF triggers the release of acetylcholine, which has the opposite effect. The NGF-induced release of the excitatory transmitter is mediated by TrkA (24), whereas the BDNF-induced switch to the inhibitory transmitter depends on p75NTR (23).

In view of the emerging functional antagonism of the neurotrophin receptor system in general and because of the previously documented significance of TrkB for the establishment of LTP specifically, we were interested to investigate a potential role of p75NTR for long-term depression (LTD).

Previous SectionNext Section

Methods

Animals. The p75NTRexonIV-/- mice used had a C57BL/6 genetic background (backcrossed for nine generations and starting from a mixed C57BL/6-129/SV background); p75NTR-/- always refers to the p75exonIV-/- mice as described in ref. 25. The WT mice used in the different experiments either were littermates of the knockouts or had the same genetic background. The p75exonIII-/- animals had a mixed BALB/C-129/SV background (26). For all experiments, the mice used were between postnatal day (P) 14 and P20.

Slice Preparation. P14-20 mice were decapitated; the brain was quickly removed and immersed in ice-cold artificial cerebrospinal fluid (ACSF). ACSF was saturated with 95% O2/5% CO2 and contained 124.0 mM NaCl, 3.0 mM KCl, 1.25 mM KH2PO4, 26.0 mM NaHCO3, 2.0 mM MgSO4, 2.5 mM CaCl2, and 10.0 mM glucose. Hippocampi were isolated and cut into 400-μm-thick transversal slices by a custom-made tissue-slicer. Slices were maintained in ACSF at room temperature for at least 1 h before recording. They were then transferred to a submersion chamber and perfused with ACSF at 32°C.

Electrophysiological Recordings. Synaptic responses were evoked in the CA3 region of the hippocampus by stimulating Schaffer collaterals with 0.1-ms pulses. Field excitatory postsynaptic potentials (fEPSPs) were recorded extracellularly in the stratum radiatum of the CA1 region by using glass microelectrodes (Clark, Reading, England). To induce LTD, a 1-Hz stimulus train was delivered for 15 min (900 stimuli). LTP was induced by applying theta-burst stimulation (TBS) of either 20 or 100 Hz. TBS consists of three bursts (10-s interval), each composed of 10 trains (5 Hz) with four pulses each. Synaptic potentiation in p75exonIII-/- animals was induced by tetanic stimulation (3 × 30 stimuli at 100 Hz with a 5-s interval between trains).

Pharmacology. To analyze the functionality of the NMDA receptor, fEPSPs were recorded in the following sequence. After 15 min of baseline recording, 10 μM 6,7-dinitroquinoxaline 2,3-dione (DNQX) (Sigma) in low-Mg2+ (0.5 mM) ACSF was bath-applied. After 15 min, 50 μM dl-2-amino-5-phosphonovalerate (APV) (Sigma), together with DNQX, was added for 20 min to the same slice in low-Mg2+ ACSF. Afterward, normal ACSF was used for washout. For the scopolamine experiments, slices were incubated in 10 μM scopolamine (Sigma) for 20 min before the onset of baseline recording. Thus, slices were perfused with scopolamine for 40 min before low-frequency stimulation (LFS).

Data Analysis. Electrophysiological data were sampled at 5 kHz by using labview software (National Instruments, Austin, TX). All measurements were carried out and analyzed in a strictly blind fashion. Thus, the genotype was revealed only after the measurement and analysis.

As an indicator of synaptic strength, the initial slope of the evoked fEPSPs was calculated.

Western Blot Analysis of AMPA Receptor Subunits. Whole hippocampi were prepared from P16 mice, homogenized with a Dounce homogenizer in 1× Laemmli buffer plus protease inhibitors (Roche, Gipf-Oberfrick, Switzerland) and 1 μM DTT, heated to 95°C for 10 min, and sonified for 30 s before loading onto 10% SDS-polyacrylamide gels (Hoefer Mighty Small). Gels were blotted onto a poly(vinylidene difluoride) membrane (Immobilon-P, Millipore) in a wet blotting chamber (Amersham Pharmacia) for 65 min at 100 V. Membranes were incubated with 3.5% (wt/vol) dry milk in 1× TBST (Tris-buffered saline with Tween 20) (for GluR1, GluR2, and GluR4) or 5% (wt/vol) dry milk in 1× TBST (for actin) or in 0.5% (wt/vol) gelatin solution in 1× NET (50 mM Tris·HCl, pH 7.40/150 mM NaCl/5 mM EDTA) plus 0.5% (vol/vol) Tween 20 (for GluR3) with primary antibodies [polyclonal anti-GluR1 (Upstate Biotechnology, Lake Placid, NY), polyclonal anti-GluR2 (Chemicon), polyclonal anti-GluR3 (DPC Biermann, Friedberg, Germany), polyclonal anti-GluR4 (Up-state Biotechnology), and monoclonal anti-actin (Chemicon)] followed by secondary peroxidase-conjugated antibodies directed against mouse (actin), rabbit (GluR1, GluR2, and GluR4), or sheep (GluR3) IgG. Signals were generated by using the ECL Plus detection system (Amersham Pharmacia Biosciences).

Previous SectionNext Section

Results

Maintenance of Hippocampal LTD Is Impaired in p75NTR Knockout Mice. The ability of hippocampal slices of P14-20 mice to develop and maintain LTD was tested by applying a LFS paradigm of 900 stimuli at 1 Hz (15 min). Fig. 1A shows a summary graph of all LTD experiments. All measurements were carried out and analyzed in a strictly blind fashion. Whereas the WT and p75NTR-/- mice were indistinguishable during LFS and the initial phase of LTD, 15 min after LFS, the slope of the fEPSPs from the p75NTR-/- mice started to deviate from the fEPSP slope of WT animals by moving toward the baseline. Thirty minutes after LFS, this effect became more pronounced. In contrast to the responses observed with the mutant animals, the WT mice continued to show LTD. Thus, maintenance, but not induction, of LTD is impaired in the p75NTR mutant. We found that WT mice exhibited a reduction of the fEPSP slope 60 min after LFS to 81.8 ± 2.8% of the baseline average (n WT = 25 slices per seven mice). The fEPSP slope of p75NTR-/- mice returned to 96.8 ± 2.1% (n ko = 27 slices per seven mice). The difference was statistically significant (P < 0.001, Mann-Whitney test). To ensure that the observed difference between the genotypes is not due to a faster decay of the physiology of the slices from knockout animals, we performed a similar set of experiments in which we recorded synaptic responses for a longer period (Fig. 1B). These measurements validated that the fEPSPs from p75NTR-/- mice remained at baseline values and that the statistical difference in LTD (P < 0.01, Mann-Whitney test) is maintained at least up to 90 min after LFS. Fig. 1C depicts the cumulative distribution of synaptic responses 30-60 min after LFS, further emphasizing the difference in the frequency distribution of synaptic responses. Values from p75NTR-/- mice are shifted to the right, indicating that, overall, data points from knockout animals taken between 30 and 60 min after LFS show a less depressed fEPSP size in comparison with WT slices. In addition, we compared p75NTR+/- mice with WT controls. No differences were found, and heterozygous mice exhibited a reduction of the fEPSP slope 60 min after LFS to 87.8 ± 7.4% of the baseline average (n hetero = seven slices per three mice). The difference in WT slices was not significant (P > 0.1, Mann-Whitney test), whereas the difference in knockout mice was significant (P < 0.05, Mann-Whitney test).

Fig. 1.
View larger version:
Fig. 1.

LTD in p75NTR-/- mice is reduced and cannot be rescued by scopolamine. (A) Time course of fEPSPs slopes in p75NTR-/- (open circles; n = 27) and WT (filled circles; n = 25) mice. (Inset) Averaged (×6) single traces of an individual experiment just before and 60 min after LFS of p75NTR-/- (Right) and WT (Left) mice. (Vertical scale bar: 1 mV; horizontal scale bar: 10 ms.) (B) Long-term recordings of fEPSPs in age-matched p75NTR-/- (open circles; n = 10) and WT (filled circles; n = 7) mice. (Inset) Single traces just before and 90 min after LFS of p75NTR-/- (Right) and WT (Left) mice. (Vertical scale bar: 1 mV; horizontal scale bar: 10 ms.) (C) Cumulative percentage distribution of synaptic responses of experiments in A with mutant (open circles) and WT (filled circles) mice giving the fraction of the data points (30-60 min after LFS) that exhibit an fEPSP slope less or equal to the value shown on the abscissa. (D) Summary graph of all LTD experiments with WT and p75NTR-/- mice with and without scopolamine (scopol). Error bars, SEM.


Cholinergic Innervation of the Hippocampus Does Not Account for Impaired LTD. p75NTR is known to be abundantly expressed by the axons of basal forebrain cholinergic neurons (27). Because the number of cholinergic neurons is increased in the forebrain of p75NTR-/- mice (8), we examined the possibility that the decrease in LTD in the p75NTR-/- mice may result from a change in cholinergic input (28, 29). We therefore blocked all cholinergic input by bath application of the muscarinic acetylcholine receptor antagonist scopolamine. We found that in the presence of 10 μM scopolamine, p75NTR-/- mice still showed markedly reduced LTD in comparison with WT mice (Fig. 1D). The average slope of fEPSPs 55-60 min after LFS was reduced to 77.4 ± 2.2% with WT mice and to 96.6 ± 3.4% with p75NTR-/- animals (n ko = 16 slices per two mice and n WT = 5 slices per two mice; P < 0.001, Mann-Whitney test). Moreover, as demonstrated in Fig. 1D, addition of scopolamine did not change the size of LTD in either p75NTR-/- (P = 0.73, Mann-Whitney test) or WT mice (P = 0.45, Mann-Whitney test) in comparison with experiments in normal ACSF.

Deletion of p75NTR Does Not Affect Gross Hippocampal Morphology. Before testing other parameters of synaptic plasticity, we examined the morphology of the hippocampus. At P14-20, the shape, size, and density of cells in the pyramidal layer in the hippocampus appeared to be grossly unaltered in p75NTR-/- mice (Fig. 6 A and B, which is published as supporting information on the PNAS web site). The only phenotype that we could detect in the brain was the enlarged lumen of blood vessels, in confirmation of earlier results (25). In addition, we found that p75NTR is expressed at the time of the experiments (Fig. 6 C-H). For methods and a detailed description of the results, see Supporting Text, which is published as supporting information on the PNAS web site.

Hippocampal LTP Is Not Affected in p75NTR Knockout Mice. To investigate whether the observed deficit is specific to LTD or whether LTP is also affected in p75NTR-/- mice, we applied TBS to hippocampal slices of P14-20 mice. Fig. 2A shows the summary graph for all LTP recordings (n ko = 20 slices per four mice and n WT = 12 slices per three mice). Both genotypes exhibited LTP with no obvious difference. As in the LTD experiments, we quantified the amount of potentiation as the average fEPSP slope 60 min after TBS relative to the baseline mean. The average degree of potentiation was 130.3 ± 11.3% for the p75NTR-/- mice and 133.4 ± 6.9% for the WT animals. The difference between the two genotypes is not significant (P = 0.31, Mann-Whitney test). We also recorded potentiation for up to 100 min after TBS application and found no statistical difference between the different genotypes. To confirm the TBS data, we also induced LTP by using a tetanus (three tetanic stimuli with 30 pulses each, 100 Hz). The average degree of potentiation was 135.1 ± 10.2% (six slices per three mice) for the p75NTR-/- mice and 139.4 ± 8.4% (six slices per three mice) for the WT animals. The difference between the two genotypes is not significant (P = 0.23, Mann-Whitney test). The LTP values for both stimulus paradigms are smaller than those generally reported in the literature, which is due to the young age of the mice used in this study.

Fig. 2.
View larger version:
Fig. 2.

Synaptic responses to high-frequency stimulation are not different in p75NTR-/- and WT mice. (A) Summary graph of all LTP experiments with p75NTR-/- (open circles; n = 20) and WT (filled circles; n = 12) mice. (Inset) Single traces of an individual experiment just before and 60 min after TBS of p75NTR-/- (Left) and WT (Right) mice. (Vertical scale bar: 1 mV; horizontal scale bar: 10 ms). Error bars, SEM. (B) Frequency-dependent changes in synaptic strength in p75NTR-/- and WT mice. Summary of the mean fEPSP slopes 55-60 min after stimulation at 0.1 Hz (baseline stimulation), 1 Hz (n WT = 29 and n ko = 35), 20 Hz (n WT = 9 and n ko = 20), or 100 Hz (n WT = 12 and n ko = 20). Error bars, SEM.


The Modification Threshold of Synaptic Plasticity Is Not Shifted in p75NTR Knockout Mice. Next, we wondered whether the loss of p75NTR shifts the modification threshold of synaptic plasticity such that stimulation with frequencies other than the established 1 and 100 Hz evoke potentiation of synaptic transmission. This assumption turned out not to be the case; application of 20-Hz TBS did not lead to synaptic potentiation in either p75NTR-/- or WT mice (Fig. 2B). Average fEPSP size 60 min after TBS amounts to 104.2 ± 3.4% in p75NTR-/- animals and 108.9 ± 3.8% in WT mice, which is not statistically significant (P = 0.39, Mann-Whitney test). Similarly, 0.1-Hz stimulation, unlike 1-Hz stimulation, did not reveal any differences between the two genotypes (Fig. 2B).

LTD and LTP Experiments in p75exonIII Mice. Because a recent report suggests that a proapoptotic fragment of p75NTR may still be expressed in the p75NTR knockout animals (30), we examined the functional significance of p75NTR for the establishment of LTD in a different strain of mice. We took advantage of the existence of another mutant of the p75NTR gene in which exon III had been targeted (26), thus referred to as p75exonIII mutant. Although the full-length receptor is absent in this mutant, a truncated isoform of p75NTR is still expressed (25). In the case of the p75exonIII mutant, LTD and LTP experiments were combined. First, the LTD experiment was performed by using LFS (Fig. 3A). LTD values were taken 20-25 min after induction. Slices from WT mice exhibited significantly stronger LTD (fEPSP slope = 87.7 ± 3.7%; n = eight slices per four mice) than slices from p75exonIII-/- animals (fEPSP slope = 98.1 ± 3.8%; n = nine slices per five mice; P < 0.05, Mann-Whitney test). This result is in line with the results obtained with the p75exonIV mutant, although a difference in LTD could be observed already in the induction phase, possibly due to a different genetic background of p75exonIII-/- (mixed BALB/C-129/SV) and p75exonIV-/- (de facto pure C57BL/6) mice. After an additional 15 min, the baseline for the LTP experiment was started, and LTP was measured 30-35 min after tetanus application (Fig. 3B). We quantified the amount of potentiation 30-35 min after applying a tetanus and found that LTP was not significantly reduced in p75exonIII mutant mice (fEPSP slopeko = 135.0 ± 4.9% and fEPSP slopeWT = 131.0 ± 6.9%; n ko = nine slices per five mice and n WT = eight slices per four mice; P = 0.83, Mann-Whitney test).

Fig. 3.
View larger version:
Fig. 3.

LTD is reduced in p75exonIII-/- mice; LTP is unaffected. (A) Summary graph showing the time course of fEPSPs in p75exonIII-/- mice that still express an isoform of p75NTR (open circles; n = 9) and WT mice (filled circles; n = 8). (B) Summary graph of all experiments with p75exonIII-/- (open circles; n = 9) and WT (filled circles; n = 8) mice. LTP was induced by tetanic stimulation. Error bars, SEM.


Thus, both p75NTR mutant mice show a similar deficit in LTD, whereas LTP in both mutants is unaffected.

Analysis of Basal Synaptic Properties in p75NTR Knockout Mice. To further investigate the effect of p75NTR deletion on the properties of synaptic transmission and plasticity, we tested whether the magnitude of synaptic responses differs between the genotypes. We recorded an input-output curve by measuring the slope of evoked fEPSPs in response to stimuli of various intensities. At higher stimulus strength, we observed a marked reduction in the size of fEPSPs in p75NTR-/- mice in comparison with WT animals (P < 0.001, ANOVA; Fig. 4A). Because this experiment was performed with 2 mM Mg2+, non-NMDA receptors are likely to account for the observed reduction of the synaptic response in the knockout. To examine presynaptic function, we analyzed short-term plasticity by measuring paired-pulse facilitation (PPF) (31) and posttetanic potentiation (PTP). To induce PPF, we applied two stimuli separated by different time intervals (10, 20, 30, 40, 80, and 1,000 ms) and recorded the evoked fEPSPs. At all tested intervals, facilitation was not significantly different between WT and p75NTR-/- mice (P = 0.34, ANOVA; Fig. 4B). PTP was calculated as the amount of potentiation within 3 min after TBS under 50 μM of APV. WT mice exhibited potentiation of 157.4 ± 16.6% and p75NTR-/- animals exhibited potentiation of 158.3 ± 16.9% (n ko = 9 slices and n WT = 12 slices; data not shown). Again, the difference in PTP between the genotypes was not significant (P = 0.83, Mann-Whitney test). To further analyze potential presynaptic effects of p75NTR deletion, we calculated synaptic fatigue during high-frequency stimulation in the LTP experiments. Synaptic fatigue during high-frequency stimulation was observed in all genotypes. To quantify this effect, we compared the fourth response in relation to the first response in a burst of four stimuli (one burst consists of four stimuli with 100 Hz in the TBS experiments) and calculated this ratio for all three theta-bursts given. There was no statistically significant difference between the genotypes (P = 0.42, Mann-Whitney test; data not shown). Therefore, the reduction in basal synaptic responses in p75NTR knockout mice is not likely caused by presynaptic changes.

Fig. 4.
View larger version:
Fig. 4.

Synaptic responses are reduced in p75NTR-/- mice; short-term synaptic plasticity is normal. (A) fEPSP slope versus different stimulation intensities in WT and p75NTR-/- mice. (B) Average paired-pulse facilitation does not differ between WT (filled circles) and mutant (open circles) mice. (C) Pharmacological isolation of AMPA and NMDA receptor currents by application of 10 μM DNQX and 50 μM APV in low-Mg2+ ACSF, respectively. Error bars, SEM.


Although the analysis of evoked fEPSPs suggested non-NMDA receptors to be affected in the p75NTR-/- mice, we further specifically examined NMDA receptor function in p75NTR-/- animals. To this end, we measured the slope of the evoked fEPSPs after application of DNQX and/or APV relative to baseline values. In the presence of the non-NMDA receptor antagonist, DNQX slices from mutant mice displayed an NMDA receptor component of the fEPSP comparable to WT mice (Fig. 4C). Application of 10 μM DNQX in low Mg2+ ACSF reduced the fEPSP slope to 25.5 ± 4.8% and 28.9 ± 7.7% in WT and p75NTR-/- mice, respectively (P = 0.27, Mann-Whitney test). Upon addition of 50 μM APV, the responses were reduced to almost zero. This finding proves that the responses after DNQX treatment under low Mg2+ conditions are indeed NMDA receptor-mediated currents, indicating that NMDA receptor function is normal in p75NTR-/- mice. Application of APV alone to normal ASCF had no effect on basal synaptic transmission in mutant and WT mice, as would have been expected (data not shown). Taken together, these results suggest that NMDA receptors function normally in the p75NTR mutant, whereas AMPA receptor function seems to be impaired at higher stimulation intensities.

Expression Levels of Two AMPA Receptor Subunits Are Altered in p75NTR Knockout Mice. The combined electrophysiological and pharmacological characterization of basal synaptic transmission prompted us to investigate hippocampal AMPA receptors in p75NTR knockout animals in more detail. AMPA receptors are the principal mediators of excitatory neurotransmission in the brain and thus are strong candidates to account for the observed postsynaptic non-NMDA receptor-based reduction of evoked fEPSPs. Moreover, AMPA receptors are considered key players in modulating activity-dependent synaptic plasticity (32, 33). To investigate whether deletion of p75NTR affects the AMPA receptor, the levels of all four AMPA receptor subunits were examined in lysates of P16 hippocampi of p75NTR+/- and p75NTR-/- animals (Fig. 5). We found that the overall expression level of GluR2 was reduced by about half in hippocampi of p75NTR-/- mice, compared with p75NTR+/- littermates. In contrast, the expression level of GluR3 was increased ≈3.5-fold in hippocampi of p75NTR-/- mice in comparison to p75NTR+/- littermates. The expression levels of GluR1 and GluR4 were not altered.

Fig. 5.
View larger version:
Fig. 5.

Expression levels of the AMPA receptor subunits GluR2 and GluR3 are altered in hippocampi of p75NTR-/- mice. The levels of all four AMPA receptor subunits (GluR1-4) were compared in hippocampi of p75NTR+/- and p75NTR-/- littermates at the age of P16 by Western blotting using subunit-specific antibodies. Actin served as the internal loading control.


Previous SectionNext Section

Discussion

This study shows that deletion of the p75NTR gene results in impaired maintenance of LTD in the postnatal hippocampus. Furthermore, basal synaptic transmission is reduced by non-NMDA receptor-dependent mechanisms, and the levels of the AMPA receptor subunits GluR2 and GluR3 are significantly altered in the absence of p75NTR.

p75NTR is strongly expressed in newborn pyramidal neurons, and although it is down-regulated postnatally (34, 35), it is still detectable at P9-14 (34). These findings are consistent with our results of immunohistochemical stainings of acute hippocampal sections and organotypic cultures (Fig. 6G). Interestingly, hippocampal LTD is most easily induced during the first 3 postnatal weeks (36), which matches the time period during which detectable levels of p75NTR are expressed.

In contrast with LTD, we found that lack of p75NTR does not affect LTP in young mice (P14-20). This finding is in line with a previous observation in which inhibition of p75NTR with function-blocking antibodies in adult mice did not interfere with establishment of LTP (20).

In addition to the p75NTR mutant generated as described in ref. 25, we also measured LTD and LTP in another mutant strain that had exon III of the p75NTR gene targeted (26) and that still expresses low levels of a truncated version of the p75NTR receptor (25). Both the intracellular and transmembrane domain are intact in this p75NTR isoform, but neurotrophin binding is lost. The observation that LTD is compromised in both p75NTR mutants is important because p75NTR-related protein products have been reported in tissue extracts of both p75NTR mutant mice (30). However, von Schack et al. (25) failed to detect p75NTR-related proteins in cultured p75exonIV-/- Schwann cells. Also, a proapoptotic function of the p75NTR-related product (30) appears unlikely in vivo, because p75exonIV-/- mice have more, not fewer, cholinergic neurons in their forebrain (8).

The nature of the ligand(s) used by p75NTR in WT animals during LTD induction remains unclear at this point. Both BDNF and NGF are likely candidates, because both have been demonstrated to interfere with the establishment of hippocampal LTD, with NGF in a stimulatory manner (37) and BDNF in an inhibitory manner (38). Although these observations suggest opposite roles of NGF and BDNF for LTP and LTD, the precursor forms of the neurotrophins may play a role as well, because proBDNF and proNGF bind with higher affinities to p75NTR than to their cognate Trk receptors (39). Recently, proteolytic cleavage of proBDNF to its mature form has been shown to be essential for the late phase of hippocampal LTP (L-LTP) to occur (40). Moreover, application of a cleavageresistant form of proBDNF to hippocampal slices failed to restore L-LTP in rescue experiments, whereas it profoundly facilitated LTD. However, whether endogenous proneurotrophins are secreted at significant levels by neurons remains to be demonstrated.

Along with the LTD defect, the levels of two AMPA receptor subunit levels were found to be altered in p75NTR knockout mice. AMPA receptors are heterotetrameric proteins assembled from subunits GluR1 to GluR4, also known as GluRA-D (41, 42). In the adult hippocampus, AMPA receptor heteromers of GluR1/GluR2 and receptors composed of GluR2/GluR3 dominate (43). GluR4 is expressed at least until P9 (44). A growing body of literature supports the view that a major mechanism for LTP is activity-induced postsynaptic insertion of AMPA receptors, whereas LTD is thought to be implemented by postsynaptic internalization of AMPA receptors (33, 34, 45, 46). AMPA receptor surface expression is determined by activity-induced receptor trafficking and constitutive receptor cycling. Both processes are assumed to be mediated by subunit-specific protein interactions. Specifically, it has been proposed that the expression of hippocampal LTD and AMPA receptor internalization from the postsynaptic surface is critically dependent on the two subunits GluR2 and GluR3 (47, 48). Changes in the expression level of these two subunits are therefore likely to influence LTD in the hippocampus, which is indeed what we observed in the p75NTR knockout. Surprisingly, however, it has been reported that deleting gluR2 and gluR3 alone or in combination did not have any effect on hippocampal LTD (49, 50). It is of note, though, that in these studies, LTD was recorded for only up to 30 min after LFS, whereas we observe a clear effect on LTD in the p75NTR mutant only after this time point. Recent data also indicate that subunit-unspecific transmembrane AMPA receptor interacting proteins add significantly to the regulation of AMPA receptor surface expression (51, 52).

As with cell survival, neurite growth, and neurotransmitter release, functional antagonism with regard to activity-dependent synaptic plasticity does not seem to be implemented by simply reversing a specific mechanism. Thus, although the mechanisms underlying TrkB-mediated regulation of LTP are thought to involve its association with the sodium channel NaV1.9 (53), the regulation of the expression levels of GluR2 and GluR3 might account for the effect of p75NTR on LTD, possibly by means of a developmental mechanism. In addition, although p75NTR is devoid of intrinsic enzymatic activity, it recruits a variety of intracellular proteins, including the mitogen-activated protein kinases extracellular signal-regulated kinase (ERK) ERK1 and ERK2 (54, 55), and may thus be involved more directly, because these kinases have been shown to be necessary for the maintenance of hippocampal LTD (56).

In conclusion, our results show that hippocampal LTD is significantly impaired in two different strains of p75NTR knockout mice, potentially as a result of altered expression levels of the two AMPA receptor subunits GluR2 and GluR3. This observation is of special relevance, given the role of TrkB in the establishment of hippocampal LTP.

Previous SectionNext Section

Acknowledgments

We thank Iris Kehrer and Volker Staiger for their outstanding technical assistance and Georg Dechant (Innsbruck Medical University, Innsbruck, Austria) for providing us with the initial breeding pairs of p75NTR knockout mice. This work was supported by the Max Planck Society, Volkswagen Foundation Project I176781 (M.K., T.B., and Y.-A.B), and European Union Project QLG3-CT-1999-00602 (Y.-A.B.).

Previous SectionNext Section

Footnotes

  • ↵ To whom correspondence should be sent at the present address. E-mail: m.korte{at}tu-bs.de.

  • ↵ * H.R. and R.S. contributed equally to this work.

  • ↵ † Present address: Division of Neurobiochemistry, Biocenter Innsbruck, A-6020 Innsbruck, Austria.

  • ↵ ‡ Present address: Biocenter of the University of Basel, CH-4056 Basel, Switzerland.

  • ↵ § Present address: Zoological Institute, Technical University Braunschweig, Mendelssohnstrasse 4, D-38106 Braunschweig, Germany.

  • Author contributions: Y.-A.B. and M.K. designed research; H.R., R.S., and M.K. performed research; H.R. and M.K. analyzed data; and T.B., Y.-A.B., and M.K. wrote the paper.

  • Abbreviations: ACSF, artificial cerebrospinal fluid; AMPA, (RS)-α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid; APV, dl-2-amino-5-phosphonovalerate; DNQX, 6,7-dinitroquinoxaline 2,3-dione; fEPSP, field excitatory postsynaptic potential; LFS, low-frequency stimulation; LTD, long-term depression; LTP, long-term potentiation; NGF, nerve growth factor; P, postnatal day; TBS, theta-burst stimulation.

Previous Section
 

References

  1. ↵
    Bibel, M. & Barde, Y. A. (2000) Genes Dev. 14 , 2919-2937.pmid:11114882
    FREE Full Text
  2. ↵
    Huang, E. J. & Reichardt, L. F. (2001) Annu. Rev. Neurosci. 24 , 677-736.pmid:11520916
    CrossRefMedlineISI
  3. ↵
    Huang, E. J. & Reichardt, L. F. (2003) Annu. Rev. Biochem. 72 , 609-642.pmid:12676795
    CrossRefMedlineISI
  4. ↵
    Dechant, G. & Barde, Y. A. (2002) Nat. Neurosci. 5 , 1131-1136.pmid:12404007
    CrossRefMedlineISI
  5. ↵
    Barker, P. A. (2004) Neuron 42 , 529-533.pmid:15157416
    CrossRefMedlineISI
  6. ↵
    Frade, J. M., Rodriguez-Tebar, A. & Barde, Y. A. (1996) Nature 383 , 166-168.pmid:8774880
    CrossRefMedline
  7. ↵
    Frade, J. M. & Barde, Y. A. (1999) Development (Cambridge, U.K.) 126 , 683-690.
    Abstract
  8. ↵
    Naumann, T., Casademunt, E., Hollerbach, E., Hofmann, J., Dechant, G., Frotscher, M. & Barde, Y. A. (2002) J. Neurosci. 22 , 2409-2418.pmid:11923404
    Abstract/FREE Full Text
  9. ↵
    Wang, K. C., Kim, J. A., Sivasankaran, R., Segal, R. & He, Z. (2002) Nature 420 , 74-78.pmid:12422217
    CrossRefMedline
  10. ↵
    Wong, S. T., Henley, J. R., Kanning, K. C., Huang, K. H., Bothwell, M. & Poo, M. M. (2002) Nat. Neurosci. 5 , 1302-1308.pmid:12426574
    CrossRefMedlineISI
  11. ↵
    Blochl, A. & Thoenen, H. (1995) Eur. J. Neurosci. 7 , 1220-1228.pmid:7582095
    CrossRefMedlineISI
  12. Hartmann, M., Heumann, R. & Lessmann, V. (2001) EMBO J. 20 , 5887-5897.pmid:11689429
    CrossRefMedlineISI
  13. ↵
    Balkowiec, A. & Katz, D. M. (2002) J. Neurosci. 22 , 10399-10407.pmid:12451139
    Abstract/FREE Full Text
  14. ↵
    Korte, M., Carroll, P., Wolf, E., Brem, G., Thoenen, H. & Bonhoeffer, T. (1995) Proc. Natl. Acad. Sci. USA 92 , 8856-8860.pmid:7568031
    Abstract/FREE Full Text
  15. Chen, G., Kolbeck, R., Barde, Y. A., Bonhoeffer, T. & Kossel, A. (1999) J. Neurosci. 19 , 7983-7990.pmid:10479698
    Abstract/FREE Full Text
  16. Pozzo-Miller, L. D., Gottschalk, W., Zhang, L., McDermott, K., Du, J., Gopalakrishnan, R., Oho, C., Sheng, Z. H. & Lu, B. (1999) J. Neurosci. 19 , 4972-4983.pmid:10366630
    Abstract/FREE Full Text
  17. ↵
    Patterson, S. L., Abel, T., Deuel, T. A. S., Martin, K. C., Rose, J. C. & Kandel, E. R. (1996) Neuron 16 , 1137-1145.pmid:8663990
    CrossRefMedlineISI
  18. ↵
    Kang, H. & Schuman, E. M. (1995) Science 267 , 1658-1662.pmid:7886457
    Abstract/FREE Full Text
  19. ↵
    Minichiello, L., Korte, M., Wolfer, D., Kühn, R., Unsicker, K., Cestari, V., Rossi-Arnaud, C., Lipp, H. P., Bonhoeffer, T. & Klein, R. (1999) Neuron 24 , 401-414.pmid:10571233
    CrossRefMedlineISI
  20. ↵
    Xu, B., Gottschalk, W., Chow, A., Wilson, R. I., Schnell, E., Zang, K., Wang, D., Nicoll, R. A., Lu, B. & Reichardt, L. F. (2000) J. Neurosci. 20 , 6888-6897.pmid:10995833
    Abstract/FREE Full Text
  21. ↵
    Greferath, U., Bennie, A., Kourakis, A., Bartlett, P. F., Murphy, M. & Barrett, G. L. (2000) Eur. J. Neurosci. 12 , 885-893.pmid:10762318
    CrossRefMedlineISI
  22. ↵
    Beattie, E. C., Stellwagen, D., Morishita, W., Bresnahan, J. C., Ha, B. K., von Zastrow, M., Beattie, M. S. & Malenka, R. C. (2002) Science 295 , 2282-2285.pmid:11910117
    Abstract/FREE Full Text
  23. ↵
    Yang, B., Slonimsky, J. D. & Birren, S. J. (2002) Nat. Neurosci. 5 , 539-545.pmid:11992117
    CrossRefMedlineISI
  24. ↵
    Lockhart, S. T., Turrigiano, G. G. & Birren, S. J. (1997) J. Neurosci. 17 , 9573-9582.pmid:9391012
    Abstract/FREE Full Text
  25. ↵
    von Schack, D., Casademunt, E., Schweigreiter, R., Meyer, M., Bibel, M. & Dechant, G. (2001) Nat. Neurosci. 4 , 977-978.pmid:11559852
    CrossRefMedlineISI
  26. ↵
    Lee, K. F., Li, E., Huber, L. J., Landis, S. C., Sharpe, A. H., Chao, M. V. & Jaenisch, R. (1992) Cell 69 , 737-749.pmid:1317267
    CrossRefMedlineISI
  27. ↵
    Hefti, F., Hartikka, J., Salvatierra, A., Weiner, W. J. & Mash, D. C. (1986) Neurosci. Lett. 69 , 37-41.pmid:3018635
    CrossRefMedlineISI
  28. ↵
    Hirotsu, I., Hori, N., Katsuda, N. & Ishihara, T. (1989) Brain Res. 482 , 194-197.pmid:2706478
    CrossRefMedlineISI
  29. ↵
    Kobayashi, M., Ohno, M., Shibata, S., Yamamoto, T. & Watanabe, S. (1997) Brain Res. 777 , 242-246.pmid:9449436
    CrossRefMedline
  30. ↵
    Paul, C. E., Vereker, E., Dickson, K. M. & Barker, P. A. (2004) J. Neurosci. 24 , 1917-1923.pmid:14985432
    Abstract/FREE Full Text
  31. ↵
    Schulz, P. E., Cook, E. P. & Johnston, D. (1994) J. Neurosci. 14 , 5325-5337.pmid:7916043
    Abstract
  32. ↵
    Malinow, R. & Malenka, R. C. (2002) Annu. Rev. Neurosci. 25 , 103-126.pmid:12052905
    CrossRefMedlineISI
  33. ↵
    Bredt, D. S. & Nicoll, R. A. (2003) Neuron 40 , 361-379.pmid:14556714
    CrossRefMedlineISI
  34. ↵
    Buck, C. R., Martinez, H. J., Chao, M. V. & Black, I. B. (1988) Brain Res. Dev. Brain Res. 44 , 259-268.pmid:2852071
    Medline
  35. ↵
    Lu, B., Buck, C. R., Dreyfus, C. F. & Black, I. B. (1989) Exp. Neurol. 104 , 191-199.pmid:2542076
    CrossRefMedlineISI
  36. ↵
    Kemp, N., McQueen, J., Faulkes, S. & Bashir, Z. I. (2000) Eur. J. Neurosci. 12 , 360-366.pmid:10651891
    CrossRefMedlineISI
  37. ↵
    Ruberti, F., Berretta, N., Cattaneo, A. & Cherubini, E. (1997) Dev. Brain Res. 101 , 295-297.pmid:9263605
    CrossRefMedline
  38. ↵
    Kinoshita, S., Yasuda, H., Taniguchi, N., Katoh-Semba, R., Hatanaka, H. & Tsumoto, T. (1999) J. Neurosci. 19 , 2122-2130.pmid:10066265
    Abstract/FREE Full Text
  39. ↵
    Lu, B. (2003) Neuron 39 , 735-738.pmid:12948441
    CrossRefMedlineISI
  40. ↵
    Pang, P. T., Teng, H. K., Zaitsev, E., Woo, N. T., Sakata, K., Zhen, S., Teng, K. K., Yung, W. H., Hempstead, B. L. & Lu, B. (2004) Science 306 , 487-491.pmid:15486301
    Abstract/FREE Full Text
  41. ↵
    Hollmann, M. & Heinemann, S. H. (1994) Annu. Rev. Neurosci. 17 , 31-108.pmid:8210177
    CrossRefMedlineISI
  42. ↵
    Seeburg, P. H. (1993) Trends Neurosci. 16 , 359-365.pmid:7694406
    CrossRefMedlineISI
  43. ↵
    Wenthold, R. J., Petralia, R. S., Blahos, J., II, & Niedzielski, A. S. (1996) J. Neurosci. 16 , 1982-1989.pmid:8604042
    Abstract/FREE Full Text
  44. ↵
    Zhu, J. J., Esteban, J. A., Hayashi, Y. & Malinow, R. (2000) Nat. Neurosci. 3 , 1098-1106.pmid:11036266
    CrossRefMedlineISI
  45. ↵
    Sheng, M. & Kim, M. J. (2002) Science 298 , 776-780.pmid:12399578
    Abstract/FREE Full Text
  46. ↵
    Barry, M. F. & Ziff, E. B. (2002) Curr. Opin. Neurobiol. 12 , 279-286.pmid:12049934
    CrossRefMedlineISI
  47. ↵
    Kullmann, D. M. (1999) Neuron 24 , 288-290.pmid:10571221
    MedlineISI
  48. ↵
    Lee, S. H., Liu, L., Wang, Y. T. & Sheng, M. (2002) Neuron 36 , 661-674.pmid:12441055
    CrossRefMedlineISI
  49. ↵
    Jia, Z., Agopyan, N., Miu, P., Xiong, Z., Henderson, J., Gerlai, R., Taverna, F. A., Velumian, A., MacDonald, J., Carlen, P., et al. (1996) Neuron 17 , 945-956.pmid:8938126
    CrossRefMedlineISI
  50. ↵
    Meng, Y., Zhang, Y. & Jia, Z. (2003) Neuron 39 , 163-176.pmid:12848940
    CrossRefMedlineISI
  51. ↵
    Chen, L., Chetkovich, D. M., Petralia, R. S., Sweeney, N. T., Kawasaki, Y., Wenthold, R. J., Bredt, D. S. & Nicoll, R. A. (2000) Nature 408 , 936-943.pmid:11140673
    CrossRefMedline
  52. ↵
    Tomita, S., Fukata, M., Nicoll, R. A. & Bredt, D. S. (2004) Science 303 , 1508-1511.