Abstracts

Rationale: Early-onset Alzheimer’s disease (AD) is associated with variants in presenilin 2 (PSEN2) and there is increased incidence of seizures in patients with early-onset AD. Just as in patients with epilepsy, seizures may also contribute to impaired cognitive function in the AD patient population. Therefore, effective antiseizure drug administration to this patient group may provide a valuable improvement to quality of life in an otherwise difficult to treat neurological disease. PSEN2-associated early-onset AD is known to be associated with a biochemical loss of PSEN2 function, thus we sought to a priori establish the extent to which loss of normal PSEN2 function and chronic seizures additively impact cognitive performance of aged male and female mice. Methods: Male and female PSEN2 KO and age- and sex-matched wild-type (WT) C57Bl/6J mice were sham or corneal kindled at 6-months-old. The duration of the kindled behavioral seizure was recorded 3- and 17-days post-kindling acquisition to define whether loss of PSEN2 is associated with any differences in kindled seizure severity. Fully-kindled and sham-kindled mice were then challenged 1-3 weeks after kindling acquisition in a battery of behavioral comorbidity tests: non-habituated open field (OF), T-maze spontaneous alternation (TM), and Barnes maze (BM). Two hours after completion of the BM task, animals were euthanized and brain tissues collected for immunohistochemical quantification of the expression of the neuroplasticity and neuroinflammation-associated proteins. Results: Both male and female PSEN2 KO mice aged 6-months required significantly more stimulations to achieve the kindled state (males: KO 15.71.0 vs WT 12.70.7, t=2.41, p<0.02; female KO 11.30.9 vs female WT 8.850.5, t=2.59, p<0.02). Behavioral seizure duration was, however, no different in fully kindled PSEN2 KO vs. WT mice (males: F(1, 41)=0.298, p>0.5; females: F(1, 40)=0.507, p>0.4). Behavioral assessment post-kindling demonstrated a significant kindling x genotype interaction on percentage of total time exploring the center of an OF, suggestive of worsened anxiety-like behaviors in PSEN2 KO mice with chronic seizures (males: F(1, 37)=11.24, p<0.002; females: F(1, 34)=11.75, p<0.002). Performance on the BM was assessed by the number of holes explored to locate the escape chamber. Male BM performance was generally worsened by seizures (F(1, 44)=4.68, p=0.038) but there was no effect of genotype F(1, 44)=0.554, p>0.4). Conversely in female mice, there was a significant kindling x genotype interaction (F(1, 37) = 5.99, p=0.019), with kindled PSEN2 KO females having the most errors (p=0.007). TM performance was significantly worsened in kindled female (F(1, 36)=5.52, p=0.024) but not male F(1, 40) = 0.054, p>0.8), PSEN2 KOs. (The impact of chronic seizures and PSEN2 KO on molecular indices of cognition will be presented in an accompanying 2019 AES abstract by Barker-Haliski et al.) Conclusions: Consistent with our prior findings in young PSEN2 KO mice (Beckman et al. AES Meeting 2018), we now report that loss of PSEN2 function, as may occur in patients with early-onset AD, may prevent the formation of an epileptic network in mice aged up to 6-months-old. Despite resistance to kindling, hippocampus-dependent cognitive function may be more severely worsened by chronic seizures in female, but not male, PSEN2 KO mice. These data align with gender-related clinical differences in AD burden. Funding: This work was supported by the University of Washington Royalty Research Fund (MBH).