Regional and progressive protein abundance changes in the AppNLGF mouse model do not reflect changes in human Alzheimer's disease

There remains a lack of effective disease-modifying treatments for human Alzheimer’s disease. Early intervention is key, so there is a need to discover targets present in early disease stages. Obtaining human tissue from patients with early disease is difficult, so mouse models are commonly used to explore (relatively) early events in Alzheimer’s disease development, such as the response to amyloid deposition. Here, we investigated early changes in protein abundance in the widely used AppNLGF Alzheimer’s disease model using liquid chromatography–mass spectrometry. Tissue was collected from the hippocampus and cortex of six AppNLGF and six wild-type mice at 3 and 6 months to assess regional differences and the impact of ageing. Whereas age-related comparisons on the same mouse genotype (3- months vs 6- months) revealed relatively few protein changes, comparisons between AppNLGF and wild-type mice demonstrated extensive alterations, with the cortex being more affected than the hippocampus (763 changes vs control in cortex at 6 months vs. 304 in hippocampus). While many of the changes at 6 months were also present at 3 months, distinct pathways were affected across the two regions. We compared these data to a previous analysis of human post-mortem Alzheimer’s disease brain, matched for the regions studied. Comparison of the disease-associated pathways in these data displayed significant divergence between human Alzheimer’s disease and the mouse model. This comparison suggests that there are both regional differences in protein abundance in the mouse brain and substantial differences in the mouse response to amyloid deposition compared to that in human Alzheimer’s disease. Models should be well characterised and selected with care.