Targeting NAD+ Metabolism as Interventions for Mitochondrial Disease

Abstract

Leigh syndrome is a mitochondrial disease characterized by neurological disorders, metabolic abnormality and premature death. There is no cure for Leigh syndrome; therefore, new therapeutic targets are urgently needed. In Ndufs4-KO mice, a mouse model of Leigh syndrome, we found that Complex I deficiency led to declines in NAD⁺ levels and NAD⁺ redox imbalance. We tested the hypothesis that elevation of NAD⁺ levels would benefit Ndufs4-KO mice. Administration of NAD⁺ precursor, nicotinamide mononucleotide (NMN) extended lifespan of Ndufs4-KO mice and attenuated lactic acidosis. NMN increased lifespan by normalizing NAD⁺ redox imbalance and lowering HIF1a accumulation in Ndufs4-KO skeletal muscle without affecting the brain. NMN up-regulated alpha-ketoglutarate (KG) levels in Ndufs4-KO muscle, a metabolite essential for HIF1a degradation. To test whether supplementation of KG can treat Ndufs4-KO mice, a cell-permeable KG, dimethyl ketoglutarate (DMKG) was administered. DMKG extended lifespan of Ndufs4-KO mice and delayed onset of neurological phenotype. This study identified therapeutic mechanisms that can be targeted pharmacologically to treat Leigh syndrome.

Figure 1

Targeting altered NAD⁺ metabolism of Ndufs4-KO mice to extend lifespan. Brain tissues of age-matched (P-65 to -75) WT and Ndufs4-KO mice were collected. (A) NAD⁺ levels (B) NAD⁺/NADH ratio and (C) protein acetylation (LysAc) levels were measured. SDHA was used as loading control. N = 4–5. Unpaired 2-tailed t-tests were used. (D) Survival curves of WT and Ndufs4-KO mice with indicated treatments. N = 10–12. (E) Body weight plot of indicated mice. ^*P < 0.05 versus WT or WT-VEH. ^#P < 0.05 versus KO-VEH. Log-rank test was used. Full blot images are presented in Supplementary Fig. 7.

Figure 2

NMN or P7C3 failed to impact NAD⁺ metabolism in Ndufs4-KO brain. Brain tissues of indicated mice were collected at P-50. (A) NAD⁺ levels, (B) NAD⁺/NADH ratio (C) NADH levels of brain from mice as indicated were measured. (D) Protein acetylation levels and (E) lactate levels in brain from mice as indicated were quantified. N = 5. SDHA was used as loading control. ^*P < 0.05 versus WT-VEH. One-way ANOVA with Newman-Keuls multiple comparison test was used. Full blot images are presented in Supplementary Fig. 7. WT or Ndufs4-KO mice were treated with P7C3 daily starting from P-21. (F) Survival curves of WT and Ndufs4-KO mice treated with VEH or P7C3 were plotted. Log-rank test was used. NAD⁺ pools (NAD⁺ plus NADH levels) of brain tissues from (G) WT or (H) Ndufs4-KO mice after P7C3 treatment were measured. N = 3. P < 0.05 versus VEH; ^#P < 0.05 versus KO-VEH. Unpaired 2-tailed t-tests were used. (I) Table summarizing median lifespan and clasping occurrence of Ndufs4-KO mice with vehicle or P7C3 treatment. N = 10. ^#P < 0.05 versus KO-VEH. NS: not statistically significant versus KO-VEH.

Figure 3

NMN supplementation attenuated NAD⁺ redox imbalance and protein hyperacetylation, and suppressed lactate levels in skeletal muscle of Ndufs4-KO mice. (A,B) Serum and skeletal muscle of indicated mice were collected at P-50 and lactate levels were measured. (C) NAD⁺ levels, (D) NAD⁺/NADH ratio and (E) NADH levels of skeletal muscle from mice as indicated were measured. (F) Protein acetylation levels of skeletal muscle were measured by Western blot. N = 5. SDHA was used as loading control. Full blot images are presented in Supplementary Fig. 7. ^*P < 0.05 versus WT-VEH; ^#P < 0.05 versus KO- VEH. One-way ANOVA with Newman-Keuls multiple comparison test was used.

Figure 4

NMN blunted the activation of hypoxia signaling in Ndufs4-KO muscle via up-regulation of α-ketoglutarate (KG) levels. Protein levels of (A) HIF1a and (B) LDHA of skeletal muscle from mice as indicated were measured by Western blot. SDHA was used as loading control. (C) Acetylation levels of HIF1a and (D) KG levels from skeletal muscle of mice as indicated were quantified. (E) Glutamate dehydrogenase (GDH) catalytic reaction. (F) Acetylation levels of GDH in skeletal muscle from mice treated as indicated were measured by Western blot analysis. N = 4–5. ^*P < 0.05 versus WT-VEH; ^#P < 0.05 versus KO-VEH. One-way ANOVA with Newman-Keuls multiple comparison test was used. Full blot images are presented in Supplementary Fig. 7.

Figure 5

Supplementation of dimethyl α-ketoglutarate (DMKG) extended lifespan and delays the onset of clasping in Ndufs4-KO mice. (A) Survival curves of WT mice, KO mice treated with vehicle (VEH), NMN or DMKG. N = 10–16. (B) Table summarizing median lifespan and clasping occurrence of Ndufs4-KO mice with vehicle, NMN or DMKG treatments. N = 10–16. Log-rank test was used. (C) Levels of SOD2 protein, SOD2 acetylation, H2Ax phosphorylation (H2Ax-Pi), and protein PAR in brain tissues in DMKG treatment cohort at P-50 were quantified. N = 3–6. Protein levels of (D) HIF1a and (E) LDHA in brain tissues were measured by Western blots. N = 3. Full blot images are presented in Supplementary Fig. 7. ^*P < 0.05 versus WT-VEH; ^#P < 0.05 versus KO-VEH. NS: not statistically significant versus KO-VEH. One-way ANOVA with Newman-Keuls multiple comparison test was used. SDHA and actin were used as loading control.

Figure 6

Targeting NAD⁺ metabolism or hypoxia signaling as interventions for Leigh Syndrome, a mitochondrial disease. Complex I deficiency in Ndufs4-KO mice triggers NAD⁺ redox imbalance, activation of hypoxia signaling and lactate acidosis. NMN and DMKG treatment can improve lifespan and health of Ndufs4-KO mice via normalization of these pathways.