Authors :
Presenting Author: Zijie Jin, MS – Boston Children's Hospital
Mushajiang Xierzhati, MS – Boston Children's Hospital
Henry Lee, PhD – Boston Children's Hospital
Alexander Rotenberg, MD, PhD – Boston Children's Hospital, Harvard Medical School, Boston, MA, USA
Rationale:
The enzyme glutamic acid decarboxylase (GAD) mediates most of the g-aminobutyric acid (GABA) production the brain. The namesake reaction catalyzed by GAD also consumes glutamate. We hypothesized that an increase in GAD protein, and the accompanying enhanced capacity for glutamate consumption and GABA production, will correspond to an antiepileptic effect. Accordingly, we tested whether overexpression of GAD by an adeno-associated virus (AAV)-mediated gene therapy approach, suppresses seizures in mice.
Methods:
We used AAV as a vector to deliver GAD overexpression in mice. We incorporated two different promoters: 1) GAD+ cell specific, and 2) broad cell type expressions, in the AAV genome, tethered with a copy of the full-length human GAD-coding sequence to form AAV-pGAD-GAD and AAV-pBr-GAD respectively. To allow efficient blood-brain barrier (BBB) penetrance in mice, we used the PHP.eB capsid to package our AAV payload, delivered systemically by intraperitoneal injection. Saline-injected mice were used as experimental control. Two weeks after AAV (n=10 mice, each construct) or saline injection (n=10, each experiment), we measured: (1) baseline behavior performance by open field test and 3-chamber social interaction task, and (2) seizure susceptibility in a 5-day daily convulsant pentylenetetrazol (PTZ; 50 mg/kg) challenge.
Results:
AAV-pGAD-GAD and AAV-pBr-GAD were well-tolerated; no treated mice showed adverse behaviors in open field or social interaction tests compared to vehicle-treated groups (Fig. 1A-C). On daily PTZ challenge, AAV-pGAD-GAD and AAV-pBr-GAD showed increased survival (Veh vs AAV-pGAD-GAD, **p=0.003; Veh vs AAV-pBr-GAD, *p=0.04, Gehan-Breslow-Wilcoxon test, Fig. 1D-E). The time to reach Racine 1 in AAV-pGAD-GAD-treated mice remained significantly lengthened even on the fifth day of PTZ challenge (median time = 67 s in Veh and 87.5 s in AAV-pGAD-GAD, *p=0.02, Gehan-Breslow-Wilcoxon test, Fig. 2A). Furthermore, the time to reach Racine 4 in AAV-pBr-GAD-treated mice was significantly delayed even on the first day of PTZ challenge (**p=0.006, Gehan-Breslow-Wilcoxon test, Fig. 2B). The overall number of mice reaching Racine 4 and Racine 6 was significantly lowered upon AAV-pBr-GAD (Racine 4 – Veh: 9/10, AAV: 2/10, **p=0.006; Racine 6 – Veh: 6/10, AAV: 0/10, *p=0.01, Fisher Exact test).
Conclusions:
Our results identify GAD overexpression as a potential strategy to achieve seizure control. Interestingly, GAD overexpression driven by either a GAD+ cells-specific promoter or a broad cell type promoter was well-tolerated and effective. Future direction includes testing this novel seizure control strategy across a range of epilepsy models by (1) direct injection of recombinant GAD catalytic domain protein, (2) transient GAD upregulation by mRNA-based technologies, and (3) further tests of AAV-mediated GAD gene therapy for long-term seizure control.
Funding: NIH and SSADH Association.