Divergent roles of p75NTR and Trk receptors in BDNF's effects on dendritic spine density and morphology

Abstract

Activation of TrkB receptors by brain-derived neurotrophic factor (BDNF) followed by MAPK/ERK signaling increases dendritic spine density and the proportion of mature spines in hippocampal CA1 pyramidal neurons. Considering the opposing actions of p75(NTR) and Trk receptors in several BDNF actions on CNS neurons, we tested whether these receptors also have divergent actions on dendritic spine density and morphology. A function-blocking anti-p75(NTR) antibody (REX) did not affect spine density by itself but it prevented BDNF's effect on spine density. Intriguingly, REX by itself increased the proportion of immature spines and prevented BDNF's effect on spine morphology. In contrast, the Trk receptor inhibitor k-252a increased spine density by itself, and prevented BDNF from further increasing spine density. However, most of the spines in k-252a-treated slices were of the immature type. These effects of k-252a on spine density and morphology required neuronal activity because they were prevented by TTX. These divergent BDNF actions on spine density and morphology are reminiscent of opposing functional signaling by p75(NTR) and Trk receptors and reveal an unexpected level of complexity in the consequences of BDNF signaling on dendritic morphology.

Figure 1

BDNF increases dendritic spine density and affects the proportion of morphological spine types. (a) Dendritic segment of a CA1 pyramidal neuron that was volume-rendered to illustrate individual spine geometrical dimensions and examples of different spine types. (b) Representative examples of dendritic segments of CA1 pyramidal neurons maintained in serum-containing media (SM) and treated with BDNF (250 ng/mL) for 48 hrs (scale bar represents 2 μm). (c) Dendritic spine density expressed per 10 μm of apical dendrite. (d) Proportion of each morphological type of dendritic spine, expressed as a fraction of the total spine population. *P < 0.05 and **P < 0.01, after an unpaired Student's t-test.

Figure 2

Role of Trk and p75^NTR in BDNF's effects on dendritic spine density and morphology. (a) Representative examples of dendritic segments of CA1 pyramidal neurons maintained in serum-containing media (SM) and treated with the function-blocking antibody of p75^NTR, REX (50 μg/mL), and BDNF (250 ng/mL) for 48 hrs (scale bar represents 2 μm). (b) Dendritic spine density expressed per 10 μm of apical dendrite. (c) Proportion of each morphological type of dendritic spine, expressed as a fraction of the total spine population. (d) Representative examples of dendritic segments of CA1 pyramidal neurons maintained in SM and treated with k-252a (200 nM) and BDNF (250 ng/mL) for 48 hrs. (e) Dendritic spine density expressed per 10 μm of apical dendrite. (f) Proportion of each morphological type of dendritic spines, expressed as a fraction of the total spine population. *P < 0.05, **P < 0.01, and ***P < 0.001, after a one-way ANOVA.

Figure 3

Role of neuronal activity in k-252a's effects on dendritic spine density and morphology. (a) Representative examples of dendritic segments of CA1 pyramidal neurons maintained in serum-containing media (SM) and treated with k-252a (200 nM), TTX (1 μM), or both k-252a and TTX for 48 hrs (scale bar represents 2 μm). (b) Dendritic spine density expressed per 10 μm of apical dendrite. (c) Proportion of each morphological type of dendritic spines, expressed as a fraction of the total spine population. *P < 0.05, **P < 0.01, and ***P < 0.001, after a one-way ANOVA.