Fig 1.
Schematic of proteolytic cleavages of APP by γ-secretase complex.
In this model, ε-cleavage occurs first and releases AICD followed by γ-cleavage for Aβ secretion (adapted from Ihara and colleagues, [17]). The longer arrows sites indicated represent common Aβ species.
Fig 2.
Neither flurbiprofen nor fenofibrate shifted ε-cleavage position in a cell-free assay by MALDI-TOF.
Representative MALDI-TOF tracings of AICD peptides after incubation of crude membrane extracts of wild type APP expressing CHO cells at 37°C showed the presence of predominantly AICD50 fragment and minute amounts of AICD51 (A). Membrane preparations incubated with flurbiprofen (B) or fenofibrate (C) showed similar MALDI-TOF tracing as control. Inset: supernatant obtained from membranes shown in (A) incubated at 4°C and 37°C were analyzed by western blot. This confirmed that AICD50 was produced only following incubation at 37°C.
Table 1.
Molecular AICD species generated in cell free.
Fig 3.
Different profile of AICD species from APP-V717F expressing CHO cells in cell free assay.
MALDI-TOF tracings of AICD peptides after incubation of crude membrane extracts of APP-V717F expressing CHO cells at 37°C showed peaks representing AICD49, AICD50, and AICD51 (A). AICD51 showed the highest peak height of the three AICD species. In (B), equal amounts of synthetic AICD50 and AICD51 peptides were immunoprecipitated and analyzed by MALDI-TOF. AICD50 fragment did not ionize as efficiently as AICD51 fragment.
Fig 4.
Flurbiprofen or fenofibrate did not shift ε-cleavage of wild type or mutant APP by western blot analyses.
By immunoprecipitation and western blot analyses, neither flurbiprofen nor fenofibrate altered the length of AICD peptides generated from crude membranes from wild type APP (A, B) or APP-V717F (C, D) expressing CHO cells. Similar results were obtained with GSM-1 in APP-V717F expressing cells (E, F). Results were expressed as averaged ± standard error (B: n = 4, D: n = 3, and F: n = 3 experiments).
Fig 5.
Fenofibrate demonstrated iGSM activity in cell-free assay.
To confirm GSM activity, fenofibrate was added to crude membrane extracts during incubation and Aβ peptides recovered by immunoprecipitation and analyzed by western blotting using 82E1 antibody. In (A), addition of fenofibrate showed elevation of Aβ42 species. Shorter peptides such as Aβ38 and Aβ39 cannot be resolved due to overlap of APP C-terminal fragments in this gel system. Quantification of results from (A) is shown in (B) from the average of four experiments ± S.E. (* p < 0.05 two-tailed paired Student’s t test).
Fig 6.
Kinetics of γ-secretase cleavage was not altered by fenofibrate in cell free assay.
Levels of AICD were assessed by immunoblotting of HEK293T cells transfected with increasing quantity of APP C99 cDNA (A). The amounts of AICD generated from membranes treated with or without fenofibrate were the same regardless of starting amounts of transfected C99 cDNAs. (n = 6; averages ± S.E.). In (B), the levels of newly generated AICD were assayed from cells transfected with the same amount of C99 cDNA (1.5μg) and measured as a function of time. There was no change in the kinetics of AICD generation after treatment with fenofibrate (n = 6; averages ± S.E.).
Fig 7.
FAD-linked PS1 mutation showed reduced γ-secretase activity and also shifted position of ε-cleavage.
CHO cells stably expressing wild type PS1 or A79V or L166P mutations were analyzed form AICD production. (A) Both PA1 mutations reduced γ-secretase activity as determined by the rate of AICD generated from crude cell membranes up to 60 minutes (n = 3; averages ± S.E). (B) By immunoprecipitation and western blotting, levels of AICD50 and AICD 51 were unchanged after addition of flurbiprofen or fenofibrate. (C) Quantification of results from (B) showed no significant changes in AICD50/AICD51 ratios after drug treatment (n = 3, “NS”: p > 0.05 one-way ANOVA; averages ± S.E.).
Fig 8.
Effects of APH1A and APH1B on APP processing.
(A) Comparable expression of APH1A or APH1B in CHO cells overexpressing wild type APP (7WD10) cells after stable transfection as detected by anti-HA antibody. (B) Expression of APH1A or APH1B had no effect on total Aβ levels as assessed by ELISA. (C) Representative western blot analysis of Aβ peptides from APH1A or APH1B expressing cells showing altered profile of Aβ peptides. (D) Quantification of Aβ levels of CHO cells expressing APH1A or APH1B shown in (C). Note the significant alterations in the ratios of Aβ42/40 as well as shorter Aβ species. However, Aβ42/40 ratios were altered in opposite directions following APH1A as compared to APH1B expression (n = 3; * p < 0.05, Repeated Measures ANOVA Tukey's Multiple Comparison Test; averages ± S.E.). (E) GSM-1 increased Aβ38 levels in CHO cells overexpressing either APH1A or APH1B (n = 3; *** p<0.001, Repeated Measures ANOVA Tukey's Multiple Comparison Test; averages ± S.E.). (F) No changes were seen in the ratio of AICD50/51 peptides quantified from crude membranes of CHO cells overexpressing either APH1A or APH1B and (n = 3, averages ± S.E.).
Fig 9.
GSMs investigation, method and analysis type performed on AICD and Aβ.