Haploinsufficiency of Cyfip2 Causes Lithium‐Responsive Prefrontal Dysfunction

Abstract

Objective Genetic variants of the cytoplasmic FMR1‐interacting protein 2 (CYFIP2) encoding an actin‐regulatory protein are associated with brain disorders, including intellectual disability and epilepsy. However, specific in vivo neuronal defects and potential treatments for CYFIP2‐associated brain disorders remain largely unknown. Here, we characterized Cyfip2 heterozygous (Cyfip2+/−) mice to understand their neurobehavioral phenotypes and the underlying pathological mechanisms. Furthermore, we examined a potential treatment for such phenotypes of the Cyfip2+/− mice and specified a neuronal function mediating its efficacy. Methods We performed behavioral analyses of Cyfip2+/− mice. We combined molecular, ultrastructural, and in vitro and in vivo electrophysiological analyses of Cyfip2+/− prefrontal neurons. We also selectively reduced CYFIP2 in the prefrontal cortex (PFC) of mice with virus injections. Results Adult Cyfip2+/− mice exhibited lithium‐responsive abnormal behaviors. We found increased filamentous actin, enlarged dendritic spines, and enhanced excitatory synaptic transmission and excitability in the adult Cyfip2+/− PFC that was restricted to layer 5 (L5) neurons. Consistently, adult Cyfip2+/− mice showed increased seizure susceptibility and auditory steady‐state responses from the cortical electroencephalographic recordings. Among the identified prefrontal defects, lithium selectively normalized the hyperexcitability of Cyfip2+/− L5 neurons. RNA sequencing revealed reduced expression of potassium channel genes in the adult Cyfip2+/− PFC. Virus‐mediated reduction of CYFIP2 in the PFC was sufficient to induce L5 hyperexcitability and lithium‐responsive abnormal behavior. Interpretation These results suggest that L5‐specific prefrontal dysfunction, especially hyperexcitability, underlies both the pathophysiology and the lithium‐mediated amelioration of neurobehavioral phenotypes in adult Cyfip2+/− mice, which can be implicated in CYFIP2‐associated brain disorders. ANN NEUROL 2020;88:526–543

Altered presynaptic function and number of mitochondria in the medial prefrontal cortex of adult Cyfip2 heterozygous mice

“…In addition, mitochondria-independent mechanisms, such as changes in presynaptic actin dynamics, can be involved in the presynaptic functional changes of Cyfip2 +/− neurons, as previously shown in Cyfip1 +/− neurons [4]. Furthermore, in our previous study [8], we showed that the number of docked vesicles in a single presynaptic terminal was lower in Cyfip2 +/− neurons compared with wild-type neurons, which is also in agreement with our current finding of an enhancement in the presynaptic short-term plasticity and an increase in the pared pulse ratio in Cyfip2 +/− neurons.…”

“…1c). Considering normal densities of dendritic spines [8] and presynaptic boutons ( Fig. 1c) of mPFC L5 neurons in Cyfip2 +/− mice, these results indicate that Cyfip2 +/− neurons have more mitochondria-free presynaptic boutons.…”

“…In our previous study, we observed a decrease in the number of presynaptic docked vesicles in layer 5 (L5) neurons in the medial prefrontal cortex (mPFC) of adult Cyfip2 heterozygous (Cyfip2 +/− ) mice compared with wild-type mice [8]. Therefore, in this study, we further investigated the presynaptic functional changes in the mPFC L5 neurons of Cyfip2 +/− mice by measuring the short-term plasticity induced by trains of stimuli [see Additional File 1 for methods].…”

“…Specifically, in the synaptic compartment, CYFIP1 regulates presynaptic vesicle release [4], excitatory postsynaptic dendritic spine morphology [5], and inhibitory synaptic assembly and transmission [6]. The roles of CYFIP2 in regulating dendritic spine morphology and excitatory synaptic transmission have been investigated [7][8][9], but its roles in presynaptic functions are largely unknown.…”

“…Notably, a recent study has revealed changes in mitochondrial activity and size in Drosophila Cyfip mutants [13]. Therefore, we also investigated the number and morphology of presynaptic mitochondria in the mPFC L5 neurons of Cyfip2 +/− mice by re-analyzing electron microscopic image datasets previously acquired for dendritic spine analysis of the mPFC neurons [8] [see Additional File 1 for methods]. We found that mitochondria were observed only in a subpopulation of presynaptic boutons contacting dendritic spines of the mPFC L5 neurons, and that the ratio of presynaptic boutons containing mitochondria in Cyfip2 +/− mice was significantly lower than that observed in wild-type mice (Fig.…”