A Common Pathway for a Rare Disease?

See allHide authors and affiliations

Science  20 Dec 2013:
Vol. 342, Issue 6165, pp. 1453-1454
DOI: 10.1126/science.1248449

Leigh syndrome is a fatal, infantile neurodegenerative disease first described more than 60 years ago (1). Children with Leigh syndrome typically are born with normal prenatal development, but decline after intermittent episodes of encephalopathy and metabolic acidosis, leading to death within the first few years of life. The diagnosis is based on magnetic resonance imaging of the brain, which reveals bilaterally symmetric lesions in the brainstem and basal ganglia (see the figure) that correspond to regions of necrosis, gliosis, and hypervascularity, with relative sparing of neurons in the early stages of the disease. At present, no effective therapies are available for Leigh syndrome, and the mainstay of management is supportive care. On page 1524 of this issue, Johnson et al. (2) demonstrate that rapamycin, a compound that inhibits a protein kinase called mechanistic target of rapamycin (mTOR), delays the onset and progression of neurological symptoms in a mouse model of Leigh syndrome. mTOR lies at the hub of cellular signaling, sensing nutrient availability to regulate protein translation, autophagy, and metabolism. The new connection to mitochondrial disease widens our view of the signaling pathway, with potential therapeutic implications.

Magnetic resonance imaging of Leigh syndrome.

A characteristic of Leigh syndrome, an early childhood neurological disorder, is bilateral lesions in the basal ganglia (arrows showing the putamen region of the brain).


Leigh syndrome is a prototypical mitochondrial disorder (3), and causal mutations have been described in more than 40 genes required for mitochondrial production of adenosine triphosphate (ATP), notably oxidative phosphorylation. Johnson et al. noted that glucose restriction, an intervention that works in part through the TOR pathway (4), extends the life span of yeast lacking homologs of genes implicated in Leigh syndrome. Motivated by this observation, the authors administered daily intraperitoneal injections of rapamycin to a mouse model of Leigh syndrome. These animals recapitulate many of the features of the human condition, including bilateral brainstem lesions, breathing abnormalities, and premature death (5). Specifically, they lack the NADH dehydrogenase (ubiquinone) Fe-S protein 4 (NDUFS4), a component of complex I of the mitochondrial oxidative phosphorylation system. Rapamycin treatment partially rescued the mortality phenotype in these mice, more than doubling their median life span. Equally striking, though, is that rapamycin alleviated the development of brain lesions and attenuated the increase in the concentration of lactate (a by-product of pyruvate metabolism) in the brain.

The mechanism by which rapamycin delays progression of the disease in the mouse model is not clear, but Johnson et al. exclude key possibilities and provide several intriguing clues. Treatment with the immunosuppressive drug tacrolimus showed no protective effect, arguing against the known immunosuppressive action of rapamycin as the sole basis. Rapamycin treatment has been shown to benefit cellular models of disease due to mitochondrial DNA mutations through autophagic clearance of defective mitochondria (6, 7). However, the authors observed no improvement in the function of the oxidative phosphorylation system, indicating that the underlying biochemical lesion had not been improved through increased organelle turnover or other mechanisms. Rapamycin treatment can trigger the mitochondrial unfolded protein response (8), an adaptive retrograde response through which mitochondrial stress signals to the nucleus to increase expression of genes that improve mitochondrial protein homeostasis. However, Johnson et al. did not see induction of this response in the brain. The authors found that mTOR complex 1 (mTORC1), one of the two multisubunit complexes that contain mTOR, is activated in the brains of Ndufs4-deficient mice, and its inhibition with rapamycin conferred neuroprotection. This suggests that mTORC1 activation may contribute to disease pathogenesis in Leigh syndrome. Why mTORC1 is activated, though, is not clear, and determining this should provide valuable insights and additional therapeutic strategies.

Regardless of the mechanism, rapamycin is already approved by the U.S. Food and Drug Administration, which makes the current study quite exciting. However, additional preclinical studies are required given that the drug is not without its risks, and numerous practical questions remain. Rapamycin is used in the pediatric population for posttransplant immunosuppression and treatment of tuberous sclerosis (a tumor syndrome due to loss-of-function of negative regulators of mTOR). The immunosuppressive effects of rapamycin must be kept in mind, because infections can trigger clinical deterioration in Leigh syndrome and other mitochondrial encephalopathies. There are more than 40 genetic causes of Leigh syndrome, with an estimated incidence of 1:40,000 children (9), and at present, it is not known whether mTOR inhibition would be useful in all cases of Leigh syndrome, in cases of complex I deficiency, or only in cases of NDUFS4 deficiency. In addition, Johnson et al. treated the Ndufs4-deficient mice with rapamycin regularly from the time of weaning and clearly demonstrate efficacy. In practice, though, most cases of Leigh syndrome present with acute crises, and it is unclear whether rapamycin can reverse existing neuropathologic lesions, or slow their progression once begun.

Although it is premature to initiate human clinical trials on rapamycin treatment for mitochondrial disorders, a rapid and constructive path forward can be defined. The most important next step is to replicate the findings of Johnson et al. in independent laboratories. It will be crucial to evaluate the utility of rapamycin in mouse models after the onset of neurological symptoms, to more realistically mimic the human scenario. The dose of rapamycin used by Johnson et al. achieves concentrations in the blood that are above the target range in patients, so the drug's efficacy at a clinically relevant dose should be tested in future studies. Rapamycin also should be evaluated across animal models of other mitochondrial disorders (10) to more accurately define the patient population that may benefit. Moreover, rapamycin is an allosteric inhibitor of mTORC1 that does not completely inhibit its catalytic activity, so treatment with newer mTOR inhibitors should be attempted, as they might prove more effective. These studies, which could be completed quickly, will guide the design of rigorous clinical trials to evaluate efficacy and safety (11).

The study by Johnson et al. is particularly important because it points to a link between a well-studied cell signaling pathway, for which existing drugs are available, and the pathogenesis of a devastating but prototypical mitochondrial disorder. Rapamycin has proven effective in mouse models of Parkinson's disease and Alzheimer's disease, and extends life span in a number of species (4). The precise basis of these broad protective effects remains unclear, though they may provide additional evidence for a mitochondrial contribution to age-related disorders.

References and Notes

  1. Acknowledgments: We thank M. Frosch, G. Mitchell, M. Hirano, and J. Avruch for helpful discussions. V.K.M. is supported by the Howard Hughes Medical Institute. S.B.V. is supported by the William Randolph Hearst Fund at Harvard Medical School.

Navigate This Article