Mutation-specific amyloid-β processing in iPSC-derived neurons

Different mutations have different effects on amyloid-β (Aβ) processing in familial Alzheimer disease (fAD), experiments in patient-derived neurons have shown. The new work, led by Selina Wray and Henrik Zetterberg, demonstrates the potential of induced pluripotent stem cell (iPSC) modelling of AD.

Aβ peptides are produced by cleavage of amyloid precursor protein (APP) by β-secretase and γ-secretase, which can each produce peptides of different lengths. The most studied of these peptides is Aβ₄₂, but other peptides range from Aβ₁₄ to Aβ₄₉. The dynamics of Aβ processing have not previously been studied in human neurons.

Wray, Zetterberg and colleagues set out to address this situation by studying iPSC-derived neurons from patients with fAD. “We are intrigued by the clinical heterogeneity in fAD, and by what could underlie this at the molecular level,” says Wray. “Using patient-derived stem cells allowed us to generate physiologically relevant models with a wide range of mutations, allowing us to study their effects directly in human neurons.”

The researchers studied 2D and 3D cultures of neuronal cell lines derived from patients with fAD with different mutations in either APP or PSEN1, which encodes one subunit of γ-secretase. The Aβ₄₂:Aβ₄₀ ratio — an established marker of AD — was higher in fAD cells than in control cells, as expected. However, the Aβ₄₂:Aβ₃₈ and Aβ₃₈:Aβ₄₀ ratios varied in a mutation-dependent manner, as did the ratios of Aβ₄₃ to other Aβ peptides.

“We are the first to measure Aβ₄₃ in iPSC neurons, which is important as some of the mutations we looked at affect mainly Aβ₄₃,” says Wray. “Our work highlights the importance of looking at the full Aβ spectrum to understand the impact of fAD-causing mutations.”

Furthermore, the team were able to compare Aβ peptides in neuronal cells with those in the cerebrospinal fluid of the same patient and saw that the ratios were very similar. “A future dream scenario could be that if you know you are in a family with AD, you make iPSC-derived neurons from a skin biopsy, check the Aβ fragment ratios and screen for the drug that best corrects the aberrant pattern,” proposes Zetterberg.

Our work highlights the importance of looking at the full Aβ spectrum

For now, the team intends to build on their work. “We focused on APP processing, but γ-secretase has many substrates,” says Wray. “We are interested in whether altered processing of other substrates impacts clinical heterogeneity in fAD in a mutation-specific manner.”