Semaphorin 4D promotes inhibitory synapse formation and suppresses seizures in vivo.

PMID 29799628


We previously discovered a role for the extracellular domain of the transmembrane protein semaphorin 4D (Sema4D) as a fast-acting, selective, and positive regulator of functional γ-aminobutyric acid (GABA)ergic synapse formation in hippocampal neuronal culture. We also demonstrated that Sema4D treatment increases inhibitory tone and suppresses hyperexcitability in an organotypic hippocampal slice culture model of epilepsy. Here, we investigate the ability of Sema4D to promote GABAergic synapse formation and suppress seizure activity in vivo in adult mice. We performed a 3-hour, intrahippocampal infusion of Sema4D or control protein into the CA1 region of adult mice. To quantify GABAergic presynaptic bouton density, we performed immunohistochemistry on hippocampal tissue sections isolated from these animals using an antibody that specifically recognizes the glutamic acid decarboxylase isoform 65 protein (GAD65), which is localized to presynaptic GABAergic boutons. To assess seizure activity, we employed 2 in vivo mouse models of epilepsy, intravenous (iv) pentylenetetrazol (PTZ) and hippocampal electrical kindling, in the presence or absence of Sema4D treatment. We monitored seizure activity by behavioral observation or electroencephalography (EEG). To assay the persistence of the Sema4D effect, we monitored seizure activity and measured the density of GAD65-positive presynaptic boutons 3 or 48 hours after Sema4D infusion. Sema4D-treated mice displayed an elevated density of GABAergic presynaptic boutons juxtaposed to hippocampal pyramidal neuron cell bodies, consistent with the hypothesis that Sema4D promotes the formation of new inhibitory synapses in vivo. In addition, Sema4D acutely suppressed seizures in both the PTZ and electrical kindling models. When we introduced a 48-hour gap between Sema4D treatment and the seizure stimulus, seizure activity was indistinguishable from controls. Moreover, immunohistochemistry on brain sections or hippocampal slices isolated 3 hours, but not 48 hours, after Sema4D treatment displayed an increase in GABAergic bouton density, demonstrating temporal correlation between the effects of Sema4D on seizures and GABAergic synaptic components. Our findings suggest a novel approach to treating acute seizures: harnessing synaptogenic molecules to enhance connectivity in the inhibitory network.