Brain-derived neurotrophic factor and TrkB modulate visual experience-dependent refinement of neuronal pathways in retina

Abstract

Sensory experience refines neuronal structure and functionality. The visual system has proved to be a productive model system to study this plasticity. In the neonatal retina, the dendritic arbors of a large proportion of ganglion cells are diffuse in the inner plexiform layer. With maturation, many of these arbors become monolaminated. Visual deprivation suppresses this remodeling. Little is known of the molecular mechanisms controlling maturational and experience-dependent refinement. Here, we tested the hypothesis that brain-derived neurotrophic factor (BDNF), which is known to regulate dendritic branching and synaptic function in the brain, modulates the developmental and visual experience-dependent refinement of retinal ganglion cells. We used a transgenic mouse line, in which a small number of ganglion cells were labeled with yellow fluorescence protein, to delineate their dendritic structure in vivo. We found that transgenic overexpression of BDNF accelerated the laminar refinement of ganglion cell dendrites, whereas decreased TrkB expression or retina-specific deletion of TrkB, the cognate receptor for BDNF, retarded it. BDNF-TrkB signaling regulated the maturational formation of new branches in ON but not the bilaminated ON-OFF ganglion cells. Furthermore, BDNF overexpression overrides the requirement for visual inputs to stimulate laminar refinement and dendritic branching of ganglion cells. These experiments reveal a previously unrecognized action of BDNF and TrkB in controlling cell-specific, experience-dependent remodeling of neuronal structures in the visual system.

Figure 1.

The age-dependent conversion of bilaminated ON–OFF RGCs into monolaminated ON or OFF RGCs after eye opening in mouse. A, B, Three-dimensional images of YFP-expressing RGCs derived from Z-stack images. YFP-RGCs are labeled green, and TH-positive amacrine cells are labeled red. The top panel in A shows the orthogonal view with TH cells (red), and the bottom panel in A shows three-dimensional images of YFP-expressing RGCs. The white frames outline the object in the viewing area. Ortho-slices, labeled with a, b, and c, represent the clipping planes orthogonal to the coordinate axis of the image data set. B, The orthogonal view of YFP-expressing RGCs shown in A. The camera (A, white arrows) is set to a position that is perpendicular to the clipping planes a, b, and c, respectively. The blue, pink, and gray circles mark the OFF, ON, and ON–OFF cell bodies, respectively. The yellow arrowheads point to axons. GCL, Ganglion cell layer. Scale bar/grids, 50 μm. C, A schematic diagram shows that the IPL is subdivided into five sublaminas based on where ON and OFF bipolar axonal terminals are localized: sublamina OFF is within the top two-fifths, and ON is within the bottom three-fifths of the IPL. D, Percentages of ON, OFF, and ON–OFF RGCs in WT retinas at P13, P28, and P50. The total numbers of RGCs counted at different ages are listed on the top, and the error bar represents ±SEM (same in Figs. 2–6).

Figure 2.

BDNF modulates age- and visual experience-dependent laminar refinement of RGCs. A, Quantification of relative retinal BDNF mRNA and protein levels in the BDNF-OE mice. RNA and protein concentrations were measured to ensure equal loading. The BDNF mRNA and protein levels were normalized to GAPDH and then calibrated to their WT controls (arbitrary unit 1; same in G–I). B, Confocal fluorescence micrograph of BDNF immunolabeling shows that BDNF is overexpressed in BDNF-OE retina. C, Light micrograph of plastic-embedded sections of BDNF-OE and littermate control retinas. D, Confocal fluorescence micrograph of GAD65 immunolabeling of BDNF-OE and control retinas. Scale bar, 20 μm. E, Comparisons of percentage of ON–OFF RGCs in BDNF-OE with their littermate control mice (WT) reared in normal light/dark cycles. The total numbers of RGCs counted at P13 and P28 are listed on the top. F, Comparisons of percentages of ON and OFF RGCs in BDNF-OE with WT mice at P13. G, Quantification of relative retinal BDNF, FL TrkB, and TK TrkB levels during postnatal development (N = 4; **p < 0.01 in one-way ANOVA post Tukey's multiple-comparison test). H, I, Dark rearing downregulates retinal BDNF mRNA and protein levels measured by real-time PCR (H) and Western blot analysis (I), respectively. Representative Western blots for BDNF and the sample loading control GAPDH are shown in I. J, K, Comparisons of percentage of ON–OFF RGCs in NR with DR WT mice (J), and DR WT mice with DR BDNF-OE mice (K). *p < 0.05; **p < 0.01; ***p < 0.001 in Student's t test.

Figure 3.

TrkB is required for the regulation of the conversion of bilaminated ON–OFF RGCs to monolaminated ON- or OFF-RGCs. A, B, Confocal fluorescent micrographs of retinal sections immunostained for TrkB (A) and GAD65 (B) in TrkB^w/w and TrkB^fBZ/fBZ mice. C, D, Percentages of ON–OFF RGCs (C), ON RGCs, and OFF-RGCs (D) in mice with reduced TrkB expression and their littermate controls. E, One representative Western blot for TrkB on retinal protein extracts from conditional trkB-KO mice (TrkB^f/f, Six3-Cre/+) and their controls (TrkB^f/+, Six3-Cre/+). F, G, A plastic-embedded section of retina (F) and immunostainings with GAD65 antibodies (G) show that retina in conditional trkB-KO mice looked normal. Scale bar, 20 μm. H, I, Percentages of ON–OFF RGCs (H), ON RGCs, and OFF RGCs (I) in conditional trkB-KO mice (TrkB^f/f, Six3-Cre/+) and their controls (for genotyping details, see supplemental Table S1, available at www.jneurosci.org as supplemental material). *p < 0.05; **p < 0.01; ***p < 0.001 in Student's t test.

Figure 4.

ON–OFF RGC dendritic lengths remain unchanged, whereas ON RGCs continue to elongate after eye opening in WT retina. A, Left, A projected image of Z-stack images of a representative RGC. Right, The RGC dendrites were traced with black lines. Scale bar, 50 μm. B, The means of total dendritic length of WT RGCs at P13 and P28. C, Relative cumulative histogram of the total dendritic length of WT RGCs. The red arrow points to the rightward shift of the distribution curves of ON RGCs from P13 to P28. D, A projected image of Z-stack images of a RGC double-immunostained with YFP (green) and SMI-32 (red) antibodies. E, The means of total dendritic length of SMI-32-positive RGCs. **p < 0.01; ***p < 0.001 in one-way ANOVA post Tukey's multiple-comparison tests.

Figure 5.

The number of branches increases, whereas the mean length per branch of ON RGCs remains unchanged after eye opening. A, D, Comparisons of the means branch numbers (A) and mean length/branch (D) of ON and ON–OFF RGCs at P13 and P28. B, E, Relative cumulative histograms of the branch numbers and mean length/branch of ON and ON–OFF RGCs at P13 and P28. C, F, Comparisons of the means branch numbers (C) and mean length/branch (F) of SMI-32-positive ON and ON–OFF RGCs at P13 and P28. *p < 0.05; **p < 0.01 in one-way ANOVA post Tukey's multiple-comparison tests.

Figure 6.

BDNF and TrkB modulate experience-dependent dendritic branch elongation of ON RGCs. A, B, Dark rearing decreases the branch numbers of ON RGCs and SMI-32-positive ON RGCs at P28. C, Overexpression of BDNF increases the branch numbers of ON-RGC at P13. D, Reduced TrkB expression results in decreased branch numbers in ON RGCs at P28. E, Mean branch numbers of ON RGCs in DR BDNF-OE mice are not significantly different from DR WT controls. F, Mean branch numbers of SMI-32-positive ON RGCs in DR BDNF-OE mice are higher than their DR WT controls. The branch numbers of ON–OFF RGCs are comparable in above conditions. *p < 0.05; **p < 0.01; ***p < 0.001 in Student's t test.

Figure 7.

Simplified model illustrating that BDNF activation of TrkB receptors modulates maturational and experience-dependent RGC dendritic laminar refinement and branching during the period after eye opening (P13–P28). A, With visual experience, initially diffuse dendritic arbors of ON–OFF RGCs (drawn as bilaminated RGCs) are converted to monolaminated ON RGCs and the branch numbers of ON RGCs increase. B, Dark rearing blocks both RGC laminar refinement and an increase in branch numbers in ON RGCs. Light-dependent enhancement of BDNF–TrkB signaling accelerates laminar refinement of RGC dendrites and dendritic branching, whereas manipulations that reduce or remove BDNF–TrkB signaling retard these two remodeling processes.