Scientist have long known that mutations in superoxide dismutase 1 (SOD1) cause amyotrophic lateral sclerosis, and now a paper in the July 1 Proceedings of the National Academy of Sciences suggests one possible reason why. Researchers from Korea University in Seoul lay blame on a kinase called mammalian sterile-20 like kinase 1 (MST1), which mediates motor neuron damage in mice carrying a mutant human SOD1 gene (mSOD1), according to the research. SOD transgenic mice lacking MST1 live slightly longer, healthier lives, suggesting that attacking MST1 activity could become a therapeutic approach for ALS.
“This work was nicely done and points to a new pathway in ALS pathogenesis that we didn’t think about before,” commented Piera Pasinelli of Thomas Jefferson University in Philadelphia, Pennsylvania, who was not involved in the research.
SOD1, mutated in about one-fifth of inherited ALS cases, normally protects cells by destroying superoxide radicals. MST1, in turn, has been suggested to cause neuron death in response to an oxidant overload (Lehtinen et al., 2006). The study team, led by first author Jae Keun Lee and senior author Eui-Ju Choi, reasoned that MST1 might mediate the neural toxicity that occurs in mice expressing the disease-linked glycine-93-alanine version of human SOD1.
When activated, MST1 forms a dimer and undergoes phosphorylation. The researchers examined its activation state in the lumbar spinal cords of mice at eight and 16 weeks of age. They used antibodies specific for phosphorylated MST1 to discover that mSOD1 mice, but not wild-type controls, activated the enzyme. They concluded that mSOD1 promoted MST1 activation.
Does MST1 directly contribute to the symptoms of mSOD1 mice? Expressing mSOD1 in MST1 knockouts, the researchers observed that symptoms appeared 17 days later than usual, and that the animals survived 19 days longer than mSOD1 controls. This is a notable difference in this model system, they contend. Removing MST1 also doubled the number of motor neurons present in the spinal cord of 16-week-old mSOD1 mice. In addition, the MST1 knockouts hung onto a rotating rod longer, maintained their grip strength better, and took longer strides. “Taken together, these results suggest that MST1 might mediate the disease-related neurotoxicity in the ALS model mice,” the authors concluded.
Next, the researchers turned their attention to the pathway by which MST1 undergoes activation in mSOD1 mice. They previously showed that thioredoxin-1 (Trx1) binds and inhibits MST1 (Chae et al., 2012). In the current work, they found that Trx1 and MST1 co-immunoprecipitated from the lumbar spinal cords of non-transgenic mice, but not from mSOD1 animals. These results indicate that Trx1 and MST1 part ways in the presence of mutant SOD1. Furthermore, Lee and colleagues found that MST1 forms dimers in the presence of mutant, but not wild-type, SOD1. They concluded that mSOD1 promotes the dissociation of MST1 from its inhibitor Trx1, and the dimerization that marks MST1 activation.
The researchers observed no physical interaction between SOD1 and MST1. Rather, the accumulation of reactive oxygen species (ROS) that occurs in the absence of proper SOD1 appears to activate the kinase. In support of this idea, treating mSOD1-expressing mouse neuroblastoma NSC34 cells with N-acetylcysteine and Trolox, compounds that decrease ROS levels, prevented MST1 activation. The researchers propose that when ROS concentrations are low, Trx1 inhibits MST1, but when mSOD1 creates a high-ROS environment, Trx1 leaves MST1 to dimerize and activate.
How does this contribute to disease? The researchers found that active MST1 mediates three downstream effects associated with ALS: activation of the p38 mitogen-activated protein kinase (MAPK), caspase cleavage, and autophagy (Dewil et al., 2007; see ARF News story on Li et al., 2000; Sasaki, 2011). Regarding p38, the team observed activation of this kinase in the lumbar spinal cord of eight-week-old mSOD1 mice, but not in the MST1 knockouts. Similarly, they saw caspase 3 and caspase 9 cleavage in the same area in 12-week-old mSOD1 mice, but not if MST1 was missing. And in 16-week-olds, the lumbar spinal cord of mSOD1 mice contained more autophagy markers, and more autophagosomes, than MST1 knockouts.
“It has been known for a while that autophagy induction is increased in SOD1 models of ALS, and that reactive oxygen species induce autophagy,” commented Ralph Nixon of the Nathan Kline Institute in Orangeburg, New York, who was not part of the study team. “This paper provides a possible missing link that connects the two together.”
What about people with ALS? The researchers looked for phosphorylated MST1 in spinal cord sections from post-mortem tissue of eight people with sporadic ALS and nine neurologically healthy controls. The intensity of immunohistochemical staining suggested more activated MST1 in the ALS cases compared to the controls, implying that MST1 might play a role in human, sporadic ALS.
“Collectively, our findings implicate MST1 as a key mediator of the ALS-associated neurodegeneration, and they therefore identify it as a potential therapeutic target for ALS as well as other motor neuron diseases,” write the authors. “Additionally, given that oxidative stress is implicated in both sporadic and familial ALS cases, ROS generation might be the common mechanism for MST1 activation in ALS.”
Pasinelli said more work needs to be done to prove that oxidative stress is a key mechanism in ALS cases beyond the familial, SOD1-mediated disease. The concept that ROS contribute to all forms of ALS remains controversial, she said, noting that antioxidant therapies have failed in clinical trials (Kaufmann et al., 2009, Graf et al., 2005).
While kinases like MST1 are important in neural function and degeneration, scientists often struggle to study them because few kinase inhibitors reach the central nervous system and selectively block these enzymes. In the June PLoS ONE, researchers from Northwestern University in Chicago, Illinois, describe a strategy to create highly selective inhibitors for p38 MAPK, which has been linked to not only MST1 and ALS, but also tau pathology in mouse models of Alzheimer's disease (see related ARF news story on Oddo et al., 2005). They applied crystallography and pharmacoinformatics to design two inhibitors, one which targets multiple forms of p38 MAPK, and one specific for only the α form.
The team found the new inhibitors prevented amyloid-β suppression of long-term potentiation in mouse brain slices. Given to wild-type mice that had amyloid injected into their brains, the inhibitors improved performance in tests of spatial navigation and fear conditioning. These novel inhibitors are unlikely to make ideal drugs, because they have not been optimized for human medical use. However, they would be good starting points for drug development, the authors write. —Amber Dance