Altered balance of glutamatergic/GABAergic synaptic input and associated changes in dendrite morphology after BDNF expression in BDNF-deficient hippocampal neurons

Abstract

Cultured neurons from bdnf-/- mice display reduced densities of synaptic terminals, although in vivo these deficits are small or absent. Here we aimed at clarifying the local responses to postsynaptic brain-derived neurotrophic factor (BDNF). To this end, solitary enhanced green fluorescent protein (EGFP)-labeled hippocampal neurons from bdnf-/- mice were compared with bdnf-/- neurons after transfection with BDNF, bdnf-/- neurons after transient exposure to exogenous BDNF, and bdnf+/+ neurons in wild-type cultures. Synapse development was evaluated on the basis of presynaptic immunofluorescence and whole-cell patch-clamp recording of miniature postsynaptic currents. It was found that neurons expressing BDNF::EGFP for at least 16 h attracted a larger number of synaptic terminals than BDNF-deficient control neurons. Transfected BDNF formed clusters in the vicinity of glutamatergic terminals and produced a stronger upregulation of synaptic terminal numbers than high levels of ambient BDNF. Glutamatergic and GABAergic synapses reacted differently to postsynaptic BDNF: glutamatergic input increased, whereas GABAergic input decreased. BDNF::EGFP-expressing neurons also differed from BDNF-deficient neurons in their dendrite morphology: they exhibited weaker dendrite elongation and stronger dendrite initiation. The upregulation of glutamatergic synaptic input and the BDNF-induced downregulation of GABAergic synaptic terminal numbers by postsynaptic BDNF depended on tyrosine receptor kinase B activity, as deduced from the blocking effects of K252a. The suppression of dendrite elongation was also prevented by block of tyrosine receptor kinase B but required, in addition, glutamate receptor activity. Dendritic length decreased with the number of glutamatergic contacts. These results illuminate the role of BDNF as a retrograde synaptic regulator of synapse development and the dependence of dendrite elongation on glutamatergic input.

Figure 1.

TrkB-dependent upregulation of synaptic terminal numbers by postsynaptic BDNF. a, Real-time PCR analysis of BDNF-dependent Syp gene expression. E18 hippocampal neurons were dissociated from bdnf+/+ or bdnf−/− mice and cultured at an initial cell density of 75,000 cells per square centimeter. After 5 d in culture, cells were exposed for 16 h to either control medium or medium supplemented with BDNF (100 ng/ml). Cells were then harvested for RNA extraction, and the transcript levels of Syp were determined by real-time PCR. The chronic absence of BDNF reduced the levels of Syp mRNA and the addition of exogenous BDNF increased it. b, Immunofluorescence images of an EGFP-expressing neuron from E18 bdnf−/− cultures after immunostaining with an antibody against Syp. c, d, Counts of Syp+ terminals in contact with an EGFP-labeled neuron. ROI size: 100 × 100 μm. Typically 5–15 neurons were evaluated on each coverslip. Error bars indicate mean ± SD in each coverslip. Note the absence of significant differences between coverslips in tests with ANOVA. e, f, Histograms of the data pooled from c and d. The mean ± SE of these data sets are shown in the first and third bar in g. For further details of statistical evaluation and meaning of symbols, see Materials and Methods. g, h, Quantification of results obtained from sister cultures from BDNF-deficient hippocampi. The data for wild-type neurons are from a different set of experiments, but the cultures were prepared in the same way as the cultures from knock-out mice. Note the increase in the number of Syp+ terminals after addition of exogenous BDNF (50–100 nm for 16 h), BDNF transfection (expression time, 16 h), and absence of BDNF effects in neurons treated during the expression time with K252a (200 nm). ∗p < 0.05; ∗∗∗p < 0.001. Prob. occurrence, Probability of occurrence; ns, not significant.

Figure 2.

Enhanced glutamatergic, but not GABAergic synaptic terminal numbers in solitary BDNF-expressing neurons in a BDNF-deficient environment. a, c, Immunofluorescence images of transfected neurons after 16 h expression of BDNF::EGFP. Pseudocolors show EGFP in white, VGluT1/2 in green, and VIAAT in red. That the synaptic terminals were also stained with an antibody against Syp and visualized with Alexa 350 is not shown. Boxed areas are shown enlarged below the corresponding image. b, d, Quantification of results. Both the tBDNF-induced upregulation of VGluT+ synaptic terminal numbers and the decrease in VIAAT+ terminal numbers are prevented by incubation in K252a (200 nm). Cultures prepared from wild-type mice displayed an appreciably higher number of VGluT+ and VIAAT+ terminals per neuron and ROI; however, the respective cultures had to be prepared separately. ∗∗p < 0.01; ∗∗∗p < 0.001. ns, Not significant.

Figure 3.

Enhanced ratio of E/I synaptic terminal numbers in BDNF-expressing neurons. a, Immunofluorescence images of transfected DIV6 neurons after 72 h expression of EGFP (top panel) or BDNF::EGFP (bottom panel). Pseudocolors show EGFP in white, VGluT in green, and VIAAT in red. Syp immunofluorescence is not shown. b–e, Quantification of results. ∗∗p < 0.01; ∗∗∗p < 0.001.

Figure 4.

The presence of tBDNF in the postsynaptic neuron increases the fraction of glutamatergic mIPSCs. mPSCs were recorded from cultured hippocampal neurons in bdnf−/− hippocampal cultures at DIV8. The controls lacked BDNF and either were not transfected (n = 11) or they expressed EGFP only (n = 18). The test neurons (n = 14) expressed BDNF::EGFP. a, Sample traces of mPSC recording. Two populations of mPSCs could clearly be distinguished by their decay kinetics. b, Single Gaussian distributions of log τ_decay in the presence of the GABA_A receptor blocker gabazine (3 μm) (top graph; n = 4) or the GluR blocker DNQX (20 μm) (bottom graph; n = 4). c, Double Gaussian functions could be well fitted to the experimental data acquired in the absence of DNQX and gabazine. The point of interception of the two Gaussian functions corresponds to a τ_decay of 7.4 ms. This threshold was used to discriminate between AMPAR- and GABA_AR-mediated events in recordings of mixed synaptic activity. The probability of erroneously attributing a postsynaptic event to either GABA_AR or AMPAR activation was calculated to be 0.31%. d, Effect of transfected BDNF on the frequency of mEPSCs or mIPSCs. The control values obtained from nontransfected bdnf−/− neurons are indicated by the dashed line. e, BDNF-expressing neurons display an increased fraction of glutamatergic mPSCs (frequency of mEPSC/mPSC frequency).

Figure 5.

Altered dendritic morphology after BDNF expression in solitary hippocampal neurons in bdnf−/− cultures at DIV6. Expression time was 16 h. a, Immunofluorescence images of transfected neurons after immunostaining with antibody against EGFP (green) and MAP2 (Red). In the BDNF-transfected neuron, all neurites except one (arrowhead) were MAP2 positive. b–d, Quantification of results with BDNF transfection. e–g, Quantification of results from experiments with addition of exogenous BDNF (50 or 100 ng for 16 h). ∗∗∗p < 0.001.

Figure 6.

Differential regulation of dendrite elongation and branching by GluR activity and block of TrkB signaling. a, Negative correlation between number of VGluT+ terminals and number of dendritic trees exceeding the ROI. This correlation is significant at p < 0.01. All data points are from one experiment. Similar results were obtained in a total of four experiments. b, c, Effects of the tyrosine kinase inhibitor K252a (200 nm). d, Addition of GluR blockers DNQX (20 μm), MK801 (1 μm), and (RS)-α-methyl-4-carboxyphenylglycine (200 μm) during the 16 h expression time of EGFP or BDNF::EGFP does not prevent the tBDNF-induced upregulation of VGluT+ terminal number but does produce an additional increase. e, f, GluR silencing prevents the tBDNF-induced suppression of dendrite elongation but only weakly counteracts, if at all, the tBDNF effect on dendrite branching. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001. ns, Not significant.

Figure 7.

Clustering of BDNF::EGFP opposite glutamatergic synaptic terminals. a, Immunofluorescent images showing the distribution of BDNF::EGFP (red) with respect to MAP2 (white) and VGluT immunoreactivity (green). Large boxed region is shown below each image. Small boxed region is shown in b. b, High-power magnification of a triple-stained thin dendritic segment with one glutamatergic terminal and the respective dual line plot. c, Averaged dual line plots from 15 synaptic terminals from 8 neurons in 2 cultures. The individual fluorescence values were normalized to the peak value of any given line plot and are presented as fluorescence change (ΔF) with respect to a peak value of 1. The presynaptic and postsynaptic lines were parallel and of equal length, coinciding approximately with the axis of the dendrite. Position x₀ corresponded to the peak fluorescence value of the presynaptic plot. The distances of the postsynaptic BDNF fluorescence peaks x₁ and x₂ were determined as illustrated in b. The averaged postsynaptic line plots show that BDNF preferentially accumulated just opposite the glutamatergic synaptic terminals.