Abstracts

Rationale: Mesial-temporal lobe epilepsy (MTLE) comprises the most frequent cause of adult epilepsy, while it lacks basic remedy. One of the difficulties to block development of new treatment is absence of appropriate animal models of MTLE, which enables in vivo therapeutic test. Animal model of MTLE ideally and reproducibly involves hippocampal sclerosis (HS), limited extra-hippocampal brain damage, and verified spontaneous hippocampal onset (1). A spontaneous epilepsy mouse model was obtained by ablating the function of Girdin gene. Girdin is a filamentous-actin-binding protein expressed throughout the nervous system, of which molecular function remains unknown. Here we describe unreported phenotypes of Girdin knockout (KO) mouse to test whether it serves as an ideal MTLE animal model. Methods: Global Girdin KO (gKO) (2), and Girdin flox / nestin-Cre mice (cKO) (3) were generated as previously reported. Two network cameras (TS-WLC2, I-O DATA, Ishikawa, Japan), a 4-TeraByte network-attached-storage (HDL-XR4.0, I-O DATA) were combined to construct a long-term video monitoring system. EEG electrodes were placed at (3 mm rostral / 1 mm left, reference), (3 mm rostral / 1 mm right, EEG 1), (1.1 mm rostral / 1.3 mm right, EEG 2), and (1.8 mm caudal / 1.8 mm right, EEG 3) from bregma. Immunohistochemistry and ߿"galactosidase staining were carried out following previously published protocols (3). Results: To compensate the poor nutrient absorption and the possible masticatory dysfunction in gKO and cKO, that were causing complete pre-weaning lethality (2,3), soft food was prepared with powder agar (Morita Sh_x005F_xDC92__x005F_xDD6E_, Aichi, Japan) and powder rodent diet CA-1 (Nihon Clea, Tokyo, Japan), and was placed on the cage floor. Since 2012, gKO (n=36) and cKO (n=6) were successfully weaned. Once weaned, KO exhibited stable viability, which occasionally lived longer than two years. Surprisingly, all weaned KO exhibited early-onset generalized tonic-clonic seizures (GTCSs) without any epileptogenic induction. GTCSs of gKO began as early as 21 days old, and persisted for life (Fig.1A). Overall EEG amplitude of cKO at electrode EEG3 right above hippocampus (1627 169 V, n=8) was significantly higher than that of control (704 60 V, n=3, t-test p=0.0052) (Fig. 1B). A video-monitored GTCS and an EEG burst were synchronized in a cKO (Fig. 1C). By GFAP immunohistochemistry, bilateral hippocampal sclerosis was observed specifically in KO, which was filled with numerous activated astrocytes (Fig. 2A). ߿"galactosidase staining revealed severe granule cell dispersion (GCD) (Fig. 2B). Conclusions: Loss of function of Girdin gene spontaneously caused early-onset generalized tonic-clonic seizures, bilateral HS, and GCD, all with complete genetic penetrance in mice without any epileptogenic induction. This mouse completely fulfills all the aforementioned requirements for MTLE animal models (1), which enables in vivo therapeutic tests. Furthermore, considering that loss of function of Girdin gene unexceptionally causes early-onset epileptic seizures even in human multiplex consanguineous family (4), cell biological function of Girdin will become a hot research subject to explore the fundamental mechanism of epilepsy. References 1) Felix Rosenow, Philippe Ryvlin, Hans O. Lders, The Mesial Temporal Lobe Epilepsies, Montrouge (France), ɤitions John Libbey Eurotext, 2011, 43-56 2) Nat Cell Biol 2008; 10(3); 329-37 3) Biochem Biophys Res Commun 2012; 426; 533-8 4) Brain 2016; 139(Pt 4); 1036-44 Funding: A-STEP from the Japan Science and Technology Agency in 2014 (AS251Z02522Q) and in 2015 (AS262Z00715Q), a Takeda Visionary Research Grant 2014 from the Takeda Science Foundation, and a Grant-in-Aid for Scientific Research (C).