reply: Alzheimer's disease: Molecular consequences of presenilin-1 mutation

Gandy et al. compare our results1 with their 1994 findings2 that the amino-terminally truncated amyloid Aβ11–42 was relatively abundant in two cases of familial Alzheimer's disease involving two distinct mutations in β-APP. However, four important differences should be borne in mind: the authors compare Aβ11–42 with Aβ1–42 and ignore Aβ1–40, although both Aβ1–40 and Aβ1–42 are generated by β-secretase/BACE cleavage at residue Asp 1 (ref. 3); their data are not correlated with features related to disease severity, such as age at onset and duration; they did not examine brains with PS1 mutations (these were not known at that time); and their characterization was based on the use of size-exclusion chromatography and electrospray mass spectrometry to quantify formic-acid-extracted Aβ, whereas we used quantitative analysis of immunoprecipitated water-soluble Aβ on western blots, mass spectrometry to identify Aβ variants, and immunohistochemistry to reveal amino-truncated Aβ peptides in plaques.