Valproic acid-exposed astrocytes impair inhibitory synapse formation and function

Abstract

Valproic acid (VPA) is widely prescribed to treat epilepsy. Maternal VPA use is, however, clinically restricted because of the severe risk that VPA may cause neurodevelopmental disorders in offspring, such as autism spectrum disorder. Understanding the negative action of VPA may help to prevent VPA-induced neurodevelopmental disorders. Astrocytes play a vital role in neurodevelopment and synapse function; however, the impact of VPA on astrocyte involvement in neurodevelopment and synapse function has not been examined. In this study, we examined whether exposure of cultured astrocytes to VPA alters neuronal morphology and synapse function of co-cultured neurons. We show that synaptic transmission by inhibitory neurons was small because VPA-exposed astrocytes reduced the number of inhibitory synapses. However, synaptic transmission by excitatory neurons and the number of excitatory synapses were normal with VPA-exposed astrocytes. VPA-exposed astrocytes did not affect the morphology of inhibitory neurons. These data indicate that VPA-exposed astrocytes impair synaptogenesis specifically of inhibitory neurons. Our results indicate that maternal use of VPA would affect not only neurons but also astrocytes and would result in perturbed astrocyte-mediated neurodevelopment.

Figure 1

VPA-exposed astrocytes decrease inhibitory synaptic transmission. (a) Scheme for the procedure of plating cells and VPA exposure. (b,c) Representative traces of mEPSCs and mIPSCs were recorded in neurons co-cultured with control astrocytes and VPA-exposed astrocytes in the presence of 1 µM TTX and 2 µM gabazine or 5 µM CNQX plus 25 µM AP5, respectively. Before astrocytes were co-cultured with neurons, astrocytes were exposed to VPA at the following concentrations: 0.3 mM, 1 mM and 3 mM. (d) The frequency of mEPSCs in neurons co-cultured with VPA-exposed astrocytes was unchanged relative to neurons co-cultured with control astrocytes. (e) The frequency of mIPSCs in neurons co-cultured with VPA-exposed astrocytes was significantly decreased compared with neurons co-cultured with control astrocytes. (f,g) mEPSCs and mIPSCs amplitudes were not changed between neurons co-cultured with control astrocytes and VPA-exposed astrocytes. [n = number of neurons used to determine mEPSC frequency (d) and amplitude (f): 0 mM; n = 18, 0.3 mM; n = 15, 1 mM; n = 17, 3 mM; n = 17; mIPSC frequency (e) and amplitude (g): 0 mM; n = 20, 0.3 mM; n = 18, 1 mM; n = 16, 3 mM; n = 16; 8 cultures/4 experiments], *p < 0.05, **p < 0.01.

Figure 2

The number of VGAT-positive inhibitory synapses is decreased by VPA-exposed astrocytes without change in dendritic morphology. (a) Effects of VPA-exposed astrocytes on excitatory synapses and dendritic morphology. Representative images of autaptic neurons immunostained for the dendritic marker, microtubule-associated protein 2 (MAP2) (in green), and the excitatory synapse marker, vesicular glutamate transporter 1 (VGLUT1) (in red). (b) Effects of VPA-exposed astrocytes on inhibitory synapses and dendritic morphology. Representative images of autaptic neurons immunostained for the dendritic marker, MAP2 (in green), and the inhibitory synapse marker, vesicular GABA transporter (VGAT) (in red). Parts of the images in the top row (scale bars = 50 µm) are enlarged in the bottom row (scale bars = 10 µm). (c) The number of excitatory synapses (VGLUT1-positive) in neurons co-cultured with VPA-exposed astrocytes was unchanged relative to neurons co-cultured with control astrocytes (n = number of neurons: control, n = 20; VPA, n = 23). (d) The number of inhibitory synapses (VGAT-positive) in neurons co-cultured with VPA-exposed astrocytes was significantly decreased compared with neurons co-cultured with control astrocytes (n = number of neurons: control, n = 30; VPA, n = 30), **p < 0.01. (e,f) In glutamatergic neurons, total dendritic length and the number of branches were unchanged between neurons co-cultured with control astrocytes and VPA-exposed astrocytes (n = number of neurons: control, n = 20; VPA, n = 23). (g,h) In GABAergic neurons, total dendritic length and the number of branches were unchanged between neurons co-cultured with control astrocytes and VPA-exposed astrocytes (n = number of neurons: control, n = 30; VPA, n = 30). All data in Fig. 2 were obtained from 10 cultures/4 experiments.

Figure 3

VPA-exposed astrocytes suppress GABAergic synapse development without affecting axonal growth. (a) Effects of VPA-exposed astrocytes on inhibitory synapses and axonal morphology. Representative images of autaptic neurons immunostained for the axonal marker, tau (in green), and vesicular GABA transporter, (VGAT) (in red). Axon morphology and synapse formation were analysed at early (DIV 5) and late (DIV 14) phases. Parts of the images in the top row (scale bars = 50 µm) are enlarged in the bottom row (scale bars = 10 µm). (b,c) Total axon length and the number of axon branches were increased from DIV 5 to DIV 14 in both neurons with control and VPA-exposed astrocytes. There were no differences in axon length and the number of axon branches between neurons with control astrocytes and VPA-exposed astrocytes. (d) The number of GABAergic synapses (VGAT-positive) in neurons co-cultured with control astrocytes was significantly increased from DIV 5 to DIV 14, but not in neurons co-cultured with VPA-exposed astrocytes. [b–d: n = number of neurons: control (DIV 5), n = 18; VPA (DIV 5), n = 18; 2 cultures/1 experiment; control (DIV 14), n = 13; VPA (DIV 14), n = 16; 2 cultures/1 experiment], **p < 0.01, ****p < 0.0001.

Figure 4

VPA-exposed astrocytes exhibited reduced Ptprd mRNA levels in co-cultured neurons at DIV 14. (a–f) mRNA levels of cell surface molecules in GABAergic synapses and (g–i) mRNA levels of Gfap and Map2, and Tubb3 in neurons co-cultured with control astrocytes or VPA-exposed astrocytes (n = number of neurons: control, n = 10; VPA, n = 10; 10 cultures/2 experiments). mRNA from cortical neurons with astrocytes was harvested at DIV 14.