Fig. 5: Glutamate-induced excitotoxicity and BDNF treatment alter excitatory/inhibitory (E/I) neuron and synapse balance.

a Schematic illustrating timepoints for data acquisition; 00:30 indicates 30 min glutamate exposure (orange arrow) before BDNF treatment (green arrow). BDNF treatment persisted for 72 h. Pre timepoint is DIV 14. b Summary schematic of findings, illustrating that BDNF treatment ameliorates some aspects of injury with glutamate. c Representative images for E/I cell death studies (square tiles, merged and separate channels; scale bar indicates 100 μm) and E/I synaptic studies (rectangular tiles, merged only; scale bar indicates 10 μm). Hoechst stains nuclei; anti-GAD65/67 immunostains inhibitory neurons and synapses; anti-MAP2 immunostains dendrites; and anti-VGLUT1 immunostains excitatory neurons and synapses. d The percentage of excitatory neurons in injured networks treated with BDNF significantly decreases compared to control networks (p = 0.016) and untreated injured networks (p = 0.002). e The percentage of inhibitory neurons in injured networks treated with BDNF significantly increases compared to control networks (p = 0.016) and untreated injured networks (p = 0.002). f The E/I synapse balance increases significantly for injured networks (30g 0B and 30g 50B) compared to uninjured networks (0g 0B and 0g 50B; p < 0.05 for all comparisons). Treated injured networks are also significantly different than untreated injured networks (p = 0.041). 0g = no glutamate; 30g = 30 µM glutamate; 0B = no BDNF; 50B = 50 ng/ml BDNF. For d and e, p values calculated by the Kruskal–Wallis test followed by the Tukey–Kramer multiple comparisons test (*p < 0.05, **p < 0.01), and n = 18 datapoints with each representing a field of view. For f, p values calculated by one-way ANOVA followed by the Tukey–Kramer multiple comparisons test (*p < 0.05), and n = 7 datapoints with four fields of view averaged per coverslip. For d–f, N = 3 independent experiments. Gray boxes represent 95% CIs, and solid black lines represent mean. N indicates number of experiments, and n indicates the number of datapoints.