Postsynaptic Protein Shank3a Deficiency Synergizes with Alzheimer's Disease Neuropathology to Impair Cognitive Performance in the 3xTg-AD Murine Model.

Synaptic loss is intrinsically linked to Alzheimer's disease (AD) neuropathology and symptoms, but its direct impact on clinical symptoms remains elusive. The postsynaptic protein Shank3 (SH3 and multiple ankyrin repeat domains) is of particular interest, as the loss of a single allele of the SHANK3 gene is sufficient to cause profound cognitive symptoms in children. We thus sought to determine whether a SHANK3 deficiency could contribute to the emergence or worsening of AD symptoms and neuropathology. We first found a 30%-50% postmortem loss of SHANK3a associated with cognitive decline in the parietal cortex of individuals with AD. To further probe the role of SHANK3 in AD, we crossed male and female 3xTg-AD mice modelling Aß and tau pathologies with Shank3a-deficient mice (Shank3Δex4-9). We observed synergistic deleterious effects of Shank3a deficiency and AD neuropathology on object recognition memory at 9, 12, and 18 months of age and on anxious behavior at 9 and 12 months of age in hemizygous Shank3Δex4-9-3xTg-AD mice. In addition to the expected 50% loss of Shank3a, levels of other synaptic proteins, such as PSD-95, drebrin, and homer1, remained unchanged in the parietotemporal cortex of hemizygous Shank3Δex4-9 animals. However, Shank3a deficiency increased the levels of soluble Aß42 and human tau at 18 months of age compared with 3xTg-AD mice with normal Shank3 expression. The results of this study in human brain samples and in transgenic mice are consistent with the hypothesis that Shank3 deficiency makes a key contribution to cognitive impairment in AD.SIGNIFICANCE STATEMENT Although the loss of several synaptic proteins has been described in Alzheimer's disease (AD), it remains unclear whether their reduction contributes to clinical symptoms. The results of this study in human samples show lower levels of SHANK3a in AD brain, correlating with cognitive decline. Data gathered in a novel transgenic mouse suggest that Shank3a deficiency synergizes with AD neuropathology to induce cognitive impairment, consistent with a causal role in AD. Therefore, treatment aiming at preserving Shank3 in the aging brain may be beneficial to prevent AD.

Altered cerebral insulin response in transgenic mice expressing the epsilon-4 allele of the human apolipoprotein E gene.

Apolipoprotein E epsilon-4 (APOEε4 or APOE4), an allelic variation of the APOE gene, not only increases the risk of developing the late-onset form of Alzheimer's disease (AD), but also influences the outcome of treatment. Indeed, data from clinical studies show that the beneficial effect of insulin on cognition is blunted in APOE4 carriers. To investigate how APOE impacts insulin response, we assessed the effects of an acute insulin injection in APOE3- and APOE4-targeted replacement mice that respectively express the human APOE3 or APOE4 isoform instead of the endogenous murine ApoE protein. We evaluated cognition, insulin signaling and proteins implicated in Aß transport and tau phosphorylation in the cortex and brain capillaries. We found that a single acute insulin injection increased Akt pSer473 in APOE4 compared to APOE3 mice (+113% versus +78.5%), indicating that APOE4 carriage potentiates activation of insulin upstream signaling pathway in the brain. Insulin also led to decreased concentrations of the receptor for advanced glycation endproducts (RAGE) in brain capillaries in both groups of mice. Moreover, higher phosphorylation of tau at Ser202, one of the key markers of AD neuropathology, was observed in insulin-injected APOE4 mice (+44%), consistent with findings in human APOE4 carriers (+400% compared to non-carriers). Therefore, our data suggest that APOE4 carriage leads to an increased insulin-induced activation of cerebral Akt pathway, associated with higher AD-like tau neuropathology. Our results provide evidence of altered insulin signaling in APOE4 carriers as well as a possible mechanism to explain the absence of cognitive benefit from insulin therapy in these individuals.

Impaired thermoregulation and beneficial effects of thermoneutrality in the 3×Tg-AD model of Alzheimer's disease.

The sharp rise in the incidence of Alzheimer's disease (AD) at an old age coincides with a reduction in energy metabolism and core body temperature. We found that the triple-transgenic mouse model of AD (3×Tg-AD) spontaneously develops a lower basal body temperature and is more vulnerable to a cold environment compared with age-matched controls. This was despite higher nonshivering thermogenic activity, as evidenced by brown adipose tissue norepinephrine content and uncoupling protein 1 expression. A 24-hour exposure to cold (4 °C) aggravated key neuropathologic markers of AD such as: tau phosphorylation, soluble amyloid beta concentrations, and synaptic protein loss in the cortex of 3×Tg-AD mice. Strikingly, raising the body temperature of aged 3×Tg-AD mice via exposure to a thermoneutral environment improved memory function and reduced amyloid and synaptic pathologies within a week. Our results suggest the presence of a vicious cycle between impaired thermoregulation and AD-like neuropathology, and it is proposed that correcting thermoregulatory deficits might be therapeutic in AD.

Altered cerebral vascular volumes and solute transport at the blood-brain barriers of two transgenic mouse models of Alzheimer's disease.

We evaluated the integrity and function of the blood-brain barrier in 3xTg-AD mice aged 3-18 months and in APP/PS1 mice aged 8-months to determine the impacts of changes in amyloid and tau proteins on the brain vascular changes. The vascular volume (Vvasc) was sub-normal in 3xTg-AD mice aged from 6 to 18 months, but not in the APP/PS1 mice. The uptakes of [(3)H]-diazepam by the brains of 3xTg-AD, APP/PS1 and their age-matched control mice were similar at all the times studied, suggesting that the simple diffusion of small solutes is unchanged in transgenic animals. The uptake of d-glucose by the brains of 18-month old 3xTg-AD mice, but not by those of 8-month old APP/PS1 mice, was reduced compared to their age-matched controls. Accordingly, the amount of Glut-1 protein was 1.4 times lower in the brain capillaries of 18 month-old 3xTg-AD mice than in those of age-matched control mice. We conclude that the brain vascular volume is reduced early in 3xTg-AD mice, 6 months before the appearance of pathological lesions, and that this reduction persists until they are at least 18 months old. The absence of alterations in the BBB of APP/PS1 mice suggests that hyperphosphorylated tau proteins contribute to the vascular changes that occur in AD.