Figures and data
Effects of FAD-mutant PSEN1 on endoproteolysis of C99 by γ-secretase.
(A) Diagram of proteolytic cleavage of APP substrate by γ-secretase via two pathways. (B) Ribbon diagram of APP substrate bound to γ-secretase. The six FAD PSEN1 mutations studied in this paper are highlighted in yellow with side chain atoms as spheres. (C) Standard curves for AICD50-99-Flag and AICD49-99-Flag, coproducts of Aβ49 and Aβ48, respectively, were generated by MALDI-TOF using synthetic peptides, with insulin as an internal standard. (D) Quantification of AICD-Flag peptides from enzyme reactions of recombinant APP substrate C100-Flag with WT versus FAD-mutant proteases by MALDI-TOF MS. Standard curves were used to quantify AICD levels for all reactions. Detection limits prevented measurement of AICD-Flag production below 62.5 nM (the lowest standard concentration) for two mutations (A431E and A434T). Consequently, these concentrations are marked as not determined (nd). In all graphs, n = 4. Statistical comparisons between AICD product levels from FAD-mutant versus WT enzymes were performed using unpaired two-tailed t-tests (*p < 0.05, **p < 0.01, ***p < 0.001).
Alzheimer-mutant PSEN-1 affects the processive proteolysis of C99 by γ-secretase.
(A) Schematic representation of the reaction mixtures and their preparation, analyzed by LC-MS/MS for the detection of tri– and tetrapeptide coproducts. (B) Comparison of tri– and tetrapeptide coproduct concentrations between WT enzyme incubated with light-isotope substrate and WT enzyme incubated with heavy-isotope substrate, as analyzed by LC-MS/MS. (C) Bar graphs illustrating all coproduct formation for specific mutations. For the Aβ49→Aβ40 pathway, blue and purple bars represent the first, second, and third trimming steps. Red and orange bars denote trimming steps for the Aβ48→Aβ38 pathway. Purple and red bars indicate coproducts formed by WT γ-secretase, while blue and orange bars indicated coproducts formed by FAD-mutant enzyme. (D) Bar graphs showing the percentage cleavage efficiency for each trimming step for mutant enzyme compared to WT enzyme. Cleavage events where the precursor Aβ peptide level was zero (i.e., no detected coproduct) are marked as not determined (nd). For each graph, n = 4 and statistical significance was determined using unpaired two-tailed t-tests comparing FAD-mutant with WT enzyme reactions (*p < 0.05, **p < 0.01, ***p < 0.001).
Comparison of Aβ40 and Aβ42 concentrations determined by LC-MS/MS vs. ELISA.
Final Aβ40 and Aβ42 concentrations upon incubation of purified γ-secretase with C100Flag and the resulting Aβ42/Aβ40 ratios assessed by (A) ELISAs and (B) LC-MS/MS calculations. N=4 with unpaired two-tailed T-tests comparing FAD-mutant to WT enzyme reactions (*p≤ 0.05, **p≤ 0.01, ***p≤ 0.001).
Calculated concentration (nM) of each Aβ variant resulting from processing APP substrate by WT vs. FAD-mutant γ-secretase.a
FAD-mutant PS1 stabilizes γ-secretase E-S interaction.
(A) Design of fluorescence lifetime imaging microscopy (FLIM) set-up to detect E-S complexes of γ-secretase and C99/Aβ-intermediates. 6E10-Alexa Fluor™ 488 over C99-720 fluorescence ratio (6E10-A488/C99-720 ratio) enables distinguishing C99-rich and Aβ-rich subcellular compartments. (B) PSEN1/2 dKO HEK293 cells were co-transfected with C99-720 and WT or FAD mutant PSEN1. Transfected cells were immunostained with anti-C99/Aβ (mouse 6E10) and anti-nicastrin (rabbit NBP2-57365) primary antibodies and Alexa Fluor™ 488 (FRET donor) or Cy3 (acceptor)-conjugated anti-mouse and anti-rabbit IgG secondary antibodies, respectively. The donor 6E10-Alexa Fluor™ 488 (6E10-A488) lifetime was measured by FLIM. Energy transfer from the donor to the acceptor results in shortening of the donor lifetime. Scale bars, 10 µm. (C) 6E10-A488 lifetimes were analyzed in randomly selected ROIs (N=40-47 from 6-8 cells), highlighting increased E-S complexes in the cells with FAD PSEN1 mutants, except F386S, compared to WT controls. One-way ANOVA and Tukey’s multiple comparisons test; n.s., p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. (D) Representative images of confocal, pseudo-color analysis to identify C99 or Aβ-rich subcellular areas and corresponding FLIM in wild-type or F386S PSEN1 expressing cells. Scale bars, 10 µm. (E) In the areas with lower 6E10-A488/C99-720 ratios (i.e., C99-rich areas), 6E10-A488 lifetimes were not different between the cells with WT PSEN1 and those with F386S mutant. N=21 ROIs. (F) On the other hand, 6E10-A488 lifetimes were significantly shorter in the cells expressing F386S mutant PSEN1 compared to wild-type controls in Aβ-rich ROIs (N=23). Unpaired T-test **p < 0.01.
FAD mutations lead to stalled processive proteolysis of APP substrate by γ-secretase.
PSEN1 FAD-mutant γ-secretase complexes are deficient in specific cleavage steps. Enzyme-substrate/intermediate complexes are stalled at these stages.
Key Resources Table
Process of installing PSEN-1 mutations into full γ-secretase plasmid.
Step-by-step Ligation-Independent Cloning (LIC) was developed in E. coli, along with restriction digestion of both the insert and vector, enabling the successful insertion of mutations. A tricistronic plasmid containing genetic codes for three membrane protein components of the γ-secretase complex, including nicastrin, presenilin enhancer (Pen2), and anterior pharynx-defective 1 (Aph1) was prepared. This plasmid was created in two steps: initially, Nicastrin and Pen2 were combined using LIC in E. coli and restriction digestion, forming a bicistronic plasmid. Subsequently, the bicistronic plasmid was further modified by including Aph1 through another round of restriction digestion and LIC in E. coli, resulting in a tricistronic plasmid. Finally, Multi-Site Directed Mutagenesis was used to mutate PSEN1, and this monocistronic construct was incorporated into the tricistronic plasmid through additional rounds of restriction digestion and LIC in E. coli. Figure 1 illustrates the details of this process.
Expression, purification, and quality control of C100 substrate and γ-secretase.
(A) Schematic of expression and purification of γ-secretase and C100. (B) Western blot analysis of all components within expressed and purified WT and FAD-mutant γ-secretase complexes, normalized to protein concentration using Pen2 intensity. (C) Characterization of light and heavy isotopic C100-FLAG substrates. The identity and purity of both C100-FLAG variants were assessed using SDS-PAGE with silver staining and MALDI-TOF mass spectrometry. The theoretical masses are 12272.89 for the light C100-FLAG and 12951.6 for the heavy C100-FLAG. Prior to reactions, concentrations were normalized based on band intensity in western blotting.
MALDI-TOF MS detection of AICD 50-99 and AICD 49-99 products from wild-type (WT) and six PSEN1 FAD-mutant γ-secretase.
The theoretical mass for AICD 50-99 is 7286.02 Da, and for AICD 49-99 is 7406.12 Da.
Alzheimer-mutant PSEN-1 affects the processive proteolysis of C99 by γ-secretase.
Bar graphs illustrating coproduct formation at each trimming step. For the Aβ49/Aβ40 pathway, blue and purple bars represent the first, second, and third trimming steps. Red and orange bars denote trimming steps for the Aβ48/Aβ38 pathway. Blue/red and Purple/orange bars indicate coproducts formed by WT and FAD-mutant γ-secretase, respectively. For each graph, n = 4 and statistical significance was determined using unpaired two-tailed t-tests comparing FAD mutants with WT (*p < 0.05, **p < 0.01, ***p < 0.001). (Note: This figure is another representation of Figure 2C).
