Familial Alzheimer's disease mutations in presenilins: effects on endoplasmic reticulum calcium homeostasis and correlation with clinical phenotypes

Abstract

Mutations in presenilins 1 and 2 (PS1 and PS2) are responsible for approximately 40% of all early onset familial Alzheimer's disease (FAD) monogenic cases. Presenilins (PSs) function as the catalytic subunit of γ-secretase and support cleavage of the amyloid-β protein precursor (AβPP). We previously discovered that PSs also function as passive endoplasmic reticulum (ER) calcium (Ca2+) leak channels and that most FAD mutations in PSs affected their ER Ca2+ leak function. To further validate the relevance of our findings to human disease, we here performed Ca2+ imaging experiments with lymphoblasts established from FAD patients. We discovered that most FAD mutations in PSs disrupted ER Ca2+ leak function and resulted in increased ER Ca2+ pool in human lymphoblasts. However, we found that a subset of PS1 FAD mutants supported ER Ca2+ leak activity, as ER Ca2+ pool was unaffected in lymphoblasts. Most of the "functional" mutations for ER Ca2+ leak were clustered in the exon 8-9 area of PSEN1 gene and segregated with the cotton wool plaques and spastic paraparesis clinical phenotype occasionally observed in PS1 FAD patients. Our findings with the "functional" and "non-functional" PS1 FAD mutants were confirmed in Ca2+ rescue experiments with PS double-knockout mouse embryonic fibroblasts. Based on the combined effects of the PS1 FAD mutations on ER Ca2+ leak and γ-secretase activities we propose a model that explains the heterogeneity observed in FAD. The proposed model has implications for understanding the pathogenesis of both familial and sporadic AD.

Figure 1. Ca²⁺ signals in lympoblast from the FAD patient

a. The time course of ionomycin(IO)-induced Ca²⁺ signals in representative experiments with wild type lymphoblast. b. The time course of ionomycin(IO)-induced Ca²⁺ signals in representative experiments with lymphoblast from FAD patient harboring PS1-M139V mutation c. The time course of ionomycin(IO)-induced Ca²⁺ signals in representative experiments with lymphoblast from FAD patient harboring PS1-P264L mutation.

Figure 2. Ca²⁺ signals in human lymphoblast from FAD patients

The average size of ionomycin (IO)-releasable Ca²⁺ pool is shown for human FAD lymphoblast, data is shown as mean ± S.D. (n = number of experiments done). When compared to WT lymphoblast, the size of IO-releasable Ca²⁺ pool is significantly (***, p < 0.05 by ANOVA) larger in FAD lymphoblast with PS1 mutation M139V, M146L, K239E, V261F, A431E and PS2-N141I. There were no significant differences in the IO-releasable Ca²⁺ pool of PS1-P264L, R269G, C410Y, A426P, ΔE9, tau-R406W, APP-V717L, young (YAD) or old(OAD) sporadic cases when compared to WT. Even though the IO-releasable Ca²⁺ pool was more elevated in FAD lymphoblast PS1-A426P, APP-V717L and OAD there were no significant difference compared to WT lymphoblast.

Figure 3. Summary of PS1-FAD rescue experiments

a. The average basal cytosolic Ca²⁺ levels (gray bars) and amplitude of BK-induced Ca²⁺ responses (black bar) is shown for DKO cells transfected with EGFP and PS1 expression constructs as mean ± S.D. (n = number of cells analyzed). When compared to DKO cells transfected with EGFP alone, the basal Ca²⁺ levels are significantly higher (***, p < 0.05 by ANOVA) and the amplitude of BK-induced Ca²⁺ response is significantly smaller (***, p < 0.05 by ANOVA) in DKO cells transfected with EGFP + PS1, EGFP + PS1-P264L, EGFP+ PS1-R269G and EGFP + C410Y combinations. There were no differences between EGFP alone and the PS1 mutations (M139V, K239E, V261F, A426P and A431E). b. The average size of ionomycin (IO)-releasable Ca²⁺ pool is shown for DKO cells transfected with EGFP and PS1 expression constructs is shown as mean ± S.D. (n = number of cells analyzed). When compared to DKO cells transfected with EGFP alone, the size of IO-releasable Ca²⁺ pool is significantly (***, p < 0.05 by ANOVA) smaller in DKO cells transfected with EGFP + PS1 and EGFP + PS1-P264L, EGFP+PS1-R269G and EGFP + C410Y combinations when compared to DKO cells transfected with EGFP plasmid alone.

Figure 4. Summary of CWP PS1-associated rescue experiments

a. The average basal cytosolic Ca²⁺ levels (gray bars) and amplitude of BK-induced Ca²⁺ responses (black bar) is shown for DKO cells transfected with EGFP and PS1 expression constructs as mean ± S.D. (n = number of cells analyzed). When compared to DKO cells transfected with EGFP alone, the basal Ca²⁺ levels are significantly higher (***, p < 0.05 by ANOVA) and the amplitude of BK-induced Ca²⁺ response is significantly smaller (***, p < 0.05 by ANOVA) in DKO cells transfected with EGFP + PS1, EGFP +PS1-P85L, EGFP +PS1-P264L, EGFP + PS1-E280G, EGFP +PS1-P284H, EGFP+ PS1-T291P, EGFP +PS1-N405S and EGFP + ΔE8 combinations. There was no significant difference with EGFP alone compared to EGFP +PS1-G217D and EGFP +PS1-L420R. b. The average size of ionomycin (IO)-releasable Ca²⁺ pool is shown for DKO cells transfected with EGFP and PS1 expression constructs is shown as mean ± S.D. (n = number of cells analyzed). When compared to DKO cells transfected with EGFP alone, the size of IO-releasable Ca²⁺ pool is significantly (***, p < 0.05 by ANOVA) smaller in DKO cells transfected with EGFP + PS1 and EGFP +PS1-P85L, EGFP +PS1-E280G, EGFP +PS1-P284H, EGFP+ PS1-T291P, EGFP +PS1-N405S and EGFP + ΔE8 combinations when compared to DKO cells transfected with EGFP plasmid alone. There was no significant difference with EGFP alone compared to EGFP +PS1-G217D and EGFP +PS1-L420R.

Figure 5. Schematic representation of PS1-FAD mutations analyzed in ER Ca²⁺ leak function experiments

Molecular model of presenilins [8, 9]. The nine transmembrane domains (TM1-TM9) of presenilins, the γ-secretase catalytic aspartate residues (D257 and D385, yellow) and the site of the endoproteolytic cleavage of presenilins are shown. The positions of PS1 FAD mutants L85P, M139V, G217D, K239E, V261F, P264L, R269G, ΔE8, A426P, A431E, T291P, N405S, C410Y, and L420R examined in our study are shown. Also shown are positions of A79V, M146V, PS2-N141I, L166P, A246E, D257A, ΔE9, E273A, G384A, P436Q mutations analyzed in the previous studies [21, 22]. The positions of PS1 FTD-implicated mutants L113P, G183V, and Rins352 analyzed in the previous study [22] are also shown (light blue color). The effects of FAD mutations on ER Ca²⁺ leak activity of presenilins is color coded – red color is used for LOF mutations, blue color is used for “functional” mutations.

Figure 6. Effects of PS1-FAD mutations on ER Ca²⁺ levels and Aβ42/40 ratios

For each PS1-FAD mutant the Aβ42/40 ratios and ER Ca²⁺ levels were normalized to the corresponding values for the WT (open triangle). The PS1-FAD mutants with confirmed DCP pathology (M139V, M146L, A246E) are plotted as solid circles. The PS1-FAD mutants with confirmed CWP pathology (L166P, P436Q, P264L, ΔE8, E280G, ΔE9, T291P, C410Y) are plotted as open circles. The PS1-FAD mutants with unknown pathology (G384A, L85P) are plotted as open diamonds. For mutants with LOF of ER Ca²⁺ leak pathway the ER Ca²⁺ levels assumed to be increased 2-fold. For “functional” mutants the ER Ca²⁺ levels assumed to be unchanged. The fold increase in Aβ42/40 ratio for each of the PS1 FAD mutants was obtained from on-line database http://www.molgen.ua.ac.be/ADMutations [43]. The diagram divided into 4 quadrants (I, II, III, IV) as shown.