Deregulation of Mitochondrial Calcium Handling Due to Presenilin Loss Disrupts Redox Homeostasis and Promotes Neuronal Dysfunction

Abstract

Mitochondrial dysfunction and oxidative stress are major contributors to the pathophysiology of neurodegenerative diseases, including Alzheimer's disease (AD). However, the mechanisms driving mitochondrial dysfunction and oxidative stress are unclear. Familial AD (fAD) is an early onset form of AD caused primarily by mutations in the presenilin-encoding genes. Previously, using Caenorhabditis elegans as a model system to study presenilin function, we found that loss of C. elegans presenilin orthologue SEL-12 results in elevated mitochondrial and cytosolic calcium levels. Here, we provide evidence that elevated neuronal mitochondrial generated reactive oxygen species (ROS) and subsequent neurodegeneration in sel-12 mutants are a consequence of the increase of mitochondrial calcium levels and not cytosolic calcium levels. We also identify mTORC1 signaling as a critical factor in sustaining high ROS in sel-12 mutants in part through its repression of the ROS scavenging system SKN-1/Nrf. Our study reveals that SEL-12/presenilin loss disrupts neuronal ROS homeostasis by increasing mitochondrial ROS generation and elevating mTORC1 signaling, which exacerbates this imbalance by suppressing SKN-1/Nrf antioxidant activity.

Keywords: Alzheimer’s disease; Nrf2; calcium; mitochondria; neuronal dysfunction; oxidative stress; presenilin.

Figure 1

Mitochondrial and neuronal abnormalities in sel-12 mutants are mediated through mitochondrial calcium. (A) Quantification of cytoplasmic calcium levels in animals expressing GCaMP, a genetically encoded calcium biosensor, and mKate, an expression control, in the TRNs. (n ≥ 25). (B) Quantification of mitochondria calcium levels in animals expressing mitochondrial-targeted GCaMP6 and mCherry, an expression control, in the TRNs. (n ≥ 21). (C) Representative image of healthy and aberrant ALM neurons (scale bar = 10 µm) and (D) quantification of the frequency of aberrant or health neurons. (n ≥ 20). (E) Representative image of continuous and discontinuous mitochondrial organization in the ALM soma (scale bar = 10 µm) and (F) quantification of the frequency of continuous and discontinuous mitochondria. (n ≥ 20). (G) Ratio of oxidized to non-oxidized roGFP1 in the mitochondria of ALM neurons as a quantitative measure of oxidation levels. (n ≥ 20). ns p > 0.05, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 using one-way ANOVA with Kruskal-Wallis multiple comparison test (A,B,G) or chi-square test (D,F). Comparisons are made to wild type unless otherwise indicated. Error bars indicate mean +/− SEM.

Figure 2

SEL-12/presenilin loss does not induce the mitochondrial unfolded response. (A) Representative images (scale bar = 0.1 mm) of GFP fluorescence intensity hsp-6p::GFP transgenic animals as a reporter for UPR^mt. (B,C) Quantification of GFP fluorescence intensity in hsp-6p::GFP (B) or in hsp-60p::GFP (C) transgenic animals as reporters for the UPR^mt. Paraquat (PQ) was used to induce mitochondrial ROS as a positive control. (D) Quantification of GFP fluorescence intensity in hsp-4p::GFP transgenic animals as a reporter for UPR^ER. sca-1(RNAi) was used as a positive control to knock down expression of the sarco-endoplasmic reticulum calcium ATPase (SERCA) to induce an ER stress response. (n ≥ 20). ns p > 0.05, **** p < 0.0001 using one-way (B,C) with Kruskal-Wallis multiple comparison test. Comparisons are made to wild type unless otherwise indicated. Error bars indicate mean +/− SEM.

Figure 3

Inhibition of mTORC1 reduces mitochondrial ROS and rescues the hypersensitivity of sel-12 mutants to oxidative stress. (A) Relative mitochondrial oxidation levels in animals expressing mitochondria targeted roGFP1 in the TRNs. (n = 30). (B) Survival rate of animals following 24-h exposure to the oxidant paraquat (0, 50, 100, and 150 mM paraquat). (50 animals per strain, performed 3 times) (C) Paraquat survival curve in animals exposed to skn-1 or control RNAi. (D,E) Representative images (D) and quantification (E) of GFP fluorescence intensity in gst-4p::GFP transgenic animals as a reporter for transcription of SKN-1/Nrf2 target GST-4. ns p > 0.05, * p < 0.05, *** p < 0.001 using one-way ANOVA with Kruskal-Wallis test (A,D,E) or two-way ANOVA with Bonferroni test (B,C). Comparisons are made to wild type unless otherwise indicated. Error bars indicate mean +/− SEM.

Figure 4

sel-12 mutants have decreased SKN-1 activity in response to oxidative stress. (A) Quantification of gst-4p::GFP expression in animals treated with paraquat. (n ≥ 28). (B) Ratio of oxidized to non-oxidized roGFP1 in the mitochondria of ALM neurons of animals treated with paraquat. (n ≥ 22). Ns p > 0.05, * p < 0.05, **p < 0.01, *** p < 0.001 using one-way ANOVA with Kruskal-Wallis test. Comparisons are made to wild type unless otherwise indicated. Error bars indicate mean +/− SEM.

Figure 5

Activation of SKN-1 improves soft touch response and resistance to oxidative stress in sel-12 mutants. (A) Response to anterior and posterior soft touch in sel-12 mutants and sel-12 mutants carrying activating mutations in skn-1. (n = 20) (B) Paraquat survival assay in wild type, sel-12 mutants, and animals carrying activating mutations in skn-1 after 24-h exposure to either 50, 100, or 150 mM paraquat (50 animals per strain, performed 3 times). ns p > 0.05, * p < 0.05, *** p < 0.001 using one-way ANOVA with Kruskal-Wallis test (A) or two-way ANOVA with Bonferroni test (B). Error bars indicate mean +/− SEM.

Figure 6

Hyperactivation of mTORC1 is not sufficient to cause neurodegeneration. (A) Western blot of p-RSKS-1/S6k in wild type, nprl-3 and sesn-1 mutants indicating increased mTORC1 activity and (B) quantification of p-RSKS-1/actin in (A). (C) Quantification of response to anterior and posterior soft touch in wild type, sesn-1, nprl-3, and sel-12 mutants. (n = 20). (D) Relative mitochondrial oxidation levels in animals expressing mitochondria targeted roGFP1 in the TRNs. (n ≥ 15). (E) Quantification of frequency of healthy and aberrant ALM neurons present in wild type, sesn-1, nprl-3, and sel-12 mutants. (n ≥ 20). ns p > 0.05, * p < 0.05, ** p < 0.01, *** p < 0.001 using one-way ANOVA with Kruskal-Wallis multiple comparison test. Comparisons are made to wild type unless otherwise indicated. Error bars indicate mean +/− SEM.