ER stress is not elevated in the 5XFAD mouse model of Alzheimer's disease

Katherine R Sadleir,Jelena Popovic,Robert Vassar

doi:10.1074/jbc.ra118.005769

Katherine R Sadleir, Jelena Popovic + Show 1 more

Open Access

https://doi.org/10.1074/jbc.ra118.005769

Copy DOI

Abstract

Alzheimer's disease mouse models that overexpress amyloid precursor protein (APP) and presenilin 1 (PS1) form β-amyloid (Aβ) plaques, a hallmark Alzheimer's disease lesion. It has been assumed that the neuroinflammation, synaptic dysfunction, neurodegeneration, and cognitive impairment observed in these mice are caused by cerebral Aβ accumulation. However, it is also possible that accumulation of the overexpressed transmembrane proteins APP and PS1 in the endoplasmic reticulum (ER) triggers chronic ER stress and activation of the unfolded protein response (UPR). The 5XFAD mouse, a widely used amyloid pathology model, overexpresses APP and PS1, displays aggressive amyloid pathology, and has been reported to exhibit ER stress. To systematically evaluate whether 5XFAD mice have increased ER stress, here we used biochemical approaches to assess a comprehensive panel of UPR markers. We report that APP and PS1 levels are 1.8- and 1.5-fold, respectively, of those in 5XFAD compared with nontransgenic brains, indicating that transgenes are not massively overexpressed in 5XFAD mice. Using immunoblotting, we quantified UPR protein levels in nontransgenic, 5XFAD, and 5XFAD;BACE1−/− mice at 4, 6, and 9 months of age. Importantly, we did not observe elevation of the ER stress markers p-eIF2α, ATF4, CHOP, p-IRE1α, or BiP at any age in 5XFAD or 5XFAD;BACE1−/− compared with nontransgenic mice. Despite lacking Aβ generation, 5XFAD;BACE1−/− mice still expressed APP and PS1 transgenes, indicating that their overexpression does not cause ER stress. These results reveal the absence of ER stress in 5XFAD mice, suggesting that artifactual phenotypes associated with overexpression-induced ER stress are not a concern in this model.

Highlights

Alzheimer’s disease mouse models that overexpress amyloid precursor protein (APP) and presenilin 1 (PS1) form ␤-amyloid (A␤) plaques, a hallmark Alzheimer’s disease lesion
The levels of APP and PS1 overexpression vary between amyloid mouse models because of different promoters used and local effects of genomic insertion sites, so we started by quantifying the expression of transgenic APP and PS1 relative to the endogenous mouse proteins
We assess the possibility that, in the 5XFAD mouse model, overexpression of APP and PS1 transgenes leads to endoplasmic reticulum (ER) stress and activation of the unfolded protein response (UPR) and that this could be the cause of phenotypes attributed to high levels of amyloid pathology [13]

Summary

To whom correspondence should be addressed

Amyloid plaques, and neurofibrillary tangles of the disorder [1, 2]. A␤ is generated by the sequential cleavage of amyloid precursor protein (APP) by two proteases. It has been observed that interventions that ameliorate pathology and memory dysfunction in AD model mice have so far failed to have beneficial effects in humans, causing researchers to re-examine their mouse models with the goal of generating animal models of amyloid pathology with improved translational potential for exploring mechanisms of AD and testing novel therapeutic approaches [3] For these reasons, humanized APP knockin mice harboring ADAD mutations have been generated that do not overexpress transgenes [4]. We did not see any changes in the UPR at any age in 5XFAD mice compared with nontransgenic or 5XFAD;BACE1Ϫ/Ϫ mice, indicating that 5XFAD transgene overexpression does not cause detectable ER stress, nor do the high levels of A␤42 and amyloid plaques observed in these mice. Our study further confirms reports that APP and APP/PS1 transgenic mice do not exhibit the UPR and, questions whether amyloid pathology causes ER stress in AD

Results

Discussion

Experimental procedures