SARM1 activation induces reversible mitochondrial dysfunction and can be prevented in human neurons by antisense oligonucleotides

Abstract

SARM1 is a key regulator of a conserved program of axon degeneration increasingly linked to human neurodegenerative diseases. Pathological SARM1 activation causes rapid NAD consumption, disrupting cellular homeostasis and leading to axon degeneration. In this study, we develop antisense oligonucleotides (ASOs) targeting human SARM1, demonstrating robust neuroprotection against morphological, metabolic, and mitochondrial impairment in human iPSC-derived dopamine neurons induced by the lethal neurotoxin vacor, a potent SARM1 activator. Furthermore, our findings reveal that axon fragmentation can be prevented, and mitochondrial dysfunction reversed using the NAD precursor nicotinamide, a form of vitamin B₃, even after SARM1 activation has occurred, when neurons are already unhealthy. This research identifies ASOs as a promising therapeutic strategy to block SARM1, and provides an extensive characterisation and further mechanistic insights that demonstrate the reversibility of SARM1 toxicity in human neurons. It also identifies the SARM1 activator vacor as a specific and reversible neuroablative agent in human neurons.

Keywords: ASO; Axon degeneration; Mitochondrial dysfunction; Neuroablative; Nicotinamide; SARM1; Vacor.

Fig. 1. ASOs targeting human SARM1 rescue human iPSC-derived dopamine neurons from vacor toxicity.

(A) Schematic overview of programmed axon degeneration and the mechanism by which vacor acts on it. (B) Representative image of hiPSC-DANs plated as spot cultures growing long axons. (C) Representative images of axons from SARM1^−/− and isogenic control (ISO - CTRL) hiPSC-DANs treated with vacor or vehicle. (D) Quantification of the degeneration index for the conditions described in part (C) (mean ± SEM; n = 3 from 3 independent differentiations; two-way RM ANOVA followed by Tukey’s multiple comparison test; statistical significance shown relative to SARM1^−/− + 100 μM Vacor). (E) Representative immunoblots of hiPSC-DANs untreated (naive) or treated with SARM1 - ASOs or a non-targeting control ASO (Ctrl - ASO) at the indicated concentrations, probed for SARM1, TH and β-actin (loading controls). (F) Representative immunoblots of hiPSC-DANs from 3 healthy individuals untreated, or treated with SARM1 ‘A’ - ASO (this ASO was employed in all subsequent experiments at a concentration of 5 μM) or a non-targeting control ASO, probed for SARM1, TH and β-actin (loading controls) (Line 1 = 067; Line 2 = 156; Line 3 = 053). (G) Quantification of normalised SARM1 level (to β-actin and TH) is shown for the conditions described in part (F) (mean ± SEM; n = 3 from 1 independent differentiation; ordinary one-way ANOVA followed by Tukey’s multiple comparison test). (H) Representative images of axons from hiPSC-DANs untreated or treated with SARM1 - ASO and Ctrl - ASO following administration of vacor or vehicle. (I) Quantification of the degeneration index for the conditions described in part (H) (mean ± SEM; n = 9 from 3 independent differentiations; two-way RM ANOVA followed by Tukey’s multiple comparison test; statistical comparison shown is Ctrl - ASO +100 μM Vacor vs SARM1 - ASO +100 μM Vacor).

Fig. 2. ASOs targeting human SARM1 prevent metabolic changes caused by SARM1 activation.

(A) Representative images of axons from hiPSC-DANs at the indicated time points after administration of vacor or vehicle. (B) Quantification of the degeneration index for the conditions described in part (A) (mean ± SEM; n = 11 from 3 independent differentiations; two-way RM ANOVA followed by Šídák’s multiple comparisons test). (C) NAD, cADPR, NMN levels and NMN/NAD ratio in hiPSC-DANs untreated (0 h) or treated with SARM1 - ASO and Ctrl - ASO at the indicated time points following administration of vacor or vehicle (mean ± SEM; n = 9 from 3 independent differentiations; ordinary two-way ANOVA followed by Tukey’s multiple comparison test).

Fig. 3. ASOs targeting human SARM1 prevent mitochondrial dysfunction caused by SARM1 activation.

(A,B) Mitochondrial respiration in hiPSC-DANs untreated, or treated with SARM1 - ASO and Ctrl - ASO (A), and in SARM1^−/− and ISO - CTRL hiPSC-DANs (B) at the indicated time point following administration of vacor or vehicle. Oxygen consumption rate (OCR) was normalised to basal respiration of vehicle treated hiPSC-DANs within each individual line and shown as a % change (mean ± SEM; n = 9 from 3 independent differentiations; ordinary two-way ANOVA followed by Tukey’s multiple comparison test). (C,D) Quantification of basal respiration, ATP production, maximal respiration and spare capacity for the conditions described in part (A) and (B). (E) ATP, ADP and AMP levels in hiPSC-DANs untreated (0 h), or treated with SARM1 - ASO and Ctrl - ASO at the indicated time points following administration of vacor or vehicle (mean ± SEM; n = 9 from 3 independent differentiations; ordinary two-way ANOVA followed by Tukey’s multiple comparison test).

Fig. 4. Prevention of axon degeneration and improvement of mitochondrial dysfunction are achievable even after SARM1 activation has occurred.

(A) Schematic overview of the experimental design for experiments in (B-E). Fresh media with or without NAM was administered 4 h after vacor treatment while simultaneously removing vacor by replacing culture media. (B) Representative images of axons from hiPSC-DANs following treatments outlined in part (A). (C) Quantification of the degeneration index for the conditions described in part (B) (mean ± SEM; n = 9 from 3 independent differentiations; Mixed-effects model REML followed by Tukey’s multiple comparison test; statistical comparison shown is +100 μM Vacor +Fresh media at 4 h vs +100 μM Vacor +1 mM NAM in fresh media at 4 h). (D) Mitochondrial respiration in hiPSC-DANs following treatments outlined in part (A), at the indicated time point following drug administration. OCR was normalised to basal respiration of vehicle treated hiPSC-DANs within each individual line and shown as a % change. (E) Quantification of basal respiration, ATP production, maximal respiration and spare capacity for the conditions described in part (D) (mean ± SEM; n = 6 from 2 independent differentiations; ordinary two-way ANOVA followed by Tukey’s multiple comparison test).