mTOR inhibition alleviates mitochondrial disease in a mouse model of Leigh syndrome

Abstract

Mitochondrial dysfunction contributes to numerous health problems, including neurological and muscular degeneration, cardiomyopathies, cancer, diabetes, and pathologies of aging. Severe mitochondrial defects can result in childhood disorders such as Leigh syndrome, for which there are no effective therapies. We found that rapamycin, a specific inhibitor of the mechanistic target of rapamycin (mTOR) signaling pathway, robustly enhances survival and attenuates disease progression in a mouse model of Leigh syndrome. Administration of rapamycin to these mice, which are deficient in the mitochondrial respiratory chain subunit Ndufs4 [NADH dehydrogenase (ubiquinone) Fe-S protein 4], delays onset of neurological symptoms, reduces neuroinflammation, and prevents brain lesions. Although the precise mechanism of rescue remains to be determined, rapamycin induces a metabolic shift toward amino acid catabolism and away from glycolysis, alleviating the buildup of glycolytic intermediates. This therapeutic strategy may prove relevant for a broad range of mitochondrial diseases.

Fig. 1. Reduced mTOR signaling improves health and survival in a mouse model of Leigh syndrome

(A) Survival of the Ndufs4^−/− mice was significantly extended by rapamycin injection every other day; life span more than doubled with daily rapamycin treatment (log-rank P = 0.0002 and P < 0.0001, respectively). (B) Body weight plots of Ndufs4^−/− mice. (C) Representative forelimb clasping behavior, a widely used sign of neurological degeneration. Clasping involves an inward curling of the spine and a retraction of forelimbs (shown here) or all limbs toward the midline of the body. (D and E) Clasping in vehicle-treated (D) and daily rapamycin-treated (E) Ndufs4^−/− mice as a function of age. A total of 15 mice were observed for clasping daily for each treatment. Age of onset of clasping behavior is significantly delayed in rapamycin-treatedmice (**P<0.001 by log-rank test). (F) Ndufs4^−/− mice show a progressive decline in rotarod performance that is rescued by rapamycin (*P < 0.05, **P < 0.005, Student’s t test; error bars are ± SEM). (See also fig. S5, which indicates replicate numbers.)

Fig. 2. Rapamycin reduces neurological disease in Ndufs4^−/− mice

(A) Representative cerebellar staining for neurological lesions in 55- to 60-day-old mice. All vehicle-treated mice showed glial activation and lesions at this age, whereas lesions were not detected in age-matched daily rapamycin-treated mice (n = 6; scale bars, ~500 µm) (see also figs. S6 and S7). (B) Representative olfactory bulb staining shows activation of glia by GFAP staining and neovascularization by laminin staining in vehicle-treated knockout (KO) mice and a robust attenuation in rapamycin-treated KO mice (n = 6 per treatment; scale bars, ~500 µm). (C) Western blotting of whole-brain lysates from a separate cohort of mice shows increased GFAP in vehicle KO mice and rescue to control levels by rapamycin (*P < 0.05, Student’s t test; error bars are ± SEM). (D and E) The percentage of living mice showing neurological symptoms is greatly reduced by daily rapamycin treatment (D), as is the number of mice showing neurological symptoms at the time of death (E).

Fig. 3. Rapamycin does not substantially alter mitochondrial function or complex I assembly

(A and B) FK-506 delivered at the highest tolerated dose (see fig. S8) failed to enhance survival (A) or attenuate disease (B) in Ndufs4^−/− mice. (C) Rapamycin has no observed effect on respiratory activity or complex I deficiency of mitochondria isolated from ~50-day-old Ndufs4^−/− mice; n = 4 to 6 mice per data point. (see also fig. S10). (D) Native-in-gel activity assays reveal that rapamycin does not influence assembly or stability of complex I (see also fig. S11). (E and F) Complex I subunits (NDUFS3 and NDUFS9) are significantly reduced in Ndufs4^−/− mice, and rapamycin has no effect on their total levels (F) or subcellular localization (E) in brain. Levels of other mitochondrial proteins (cytochrome c, the complex IV subunit COXIV, and HSP60) are independent of both Ndufs4 genotype and treatment (see also fig. S9). *P < 0.05,**P < 0.005, Student’s t test; ns, not significant. Error bars are ±SEM.

Fig. 4. Ndufs4^−/− mice exhibit mTOR activation and metabolic defects that are suppressed by rapamycin

(A) mTOR activity, as indicated by phosphorylation of S6, is increased in Ndufs4^−/− mouse brain. Total IGF1R and S6 are decreased in Ndufs4^−/− mice, suggesting feedback inhibition from chronic mTOR activation. Rapamycin potently inhibits phosphorylation of S6 and rescues levels of IGF1R and S6. (B) Total body fat progressively decreases in Ndufs4^−/− mice but is maintained in rapamycin-treated mice. Fat mass differs by sex in control but not Ndufs4^−/− mice (n = 4 to 6 mice per data point). (C and D) Liver fat droplets are deficient in vehicle-treated Ndufs4^−/− mice and partially rescued by rapamycin (representative images, n > 6 stained per treatment; scale bar, ~100 µm) (C), as are free fatty acids detected by metabolomics (D) (n = 4 per treatment). (E) Accumulation of glycolytic intermediates in Ndufs4^−/− brain is suppressed by rapamycin (n = 4 per treatment) (see fig. S14 and tables S1 to S3). *P < 0.05, **P < 0.005, Student’s t test; error bars are ±SEM.