With the increasing prevalence of neurodegenerative disorders in the aging population, it has become more and more important to understand the molecular pathways that regulate and advance these disorders. Due to the high level of complexity of the mammalian brain, it is very difficult to devise improved targeted treatments. The biggest limitation in neurodegenerative disease research being the lack of viable biomarkers for the elder population. Neurodegenerative disorders such as Alzheimer’s, Parkinson’s and polyglutamine diseases, share many pathogenic abnormalities such as the accumulation of misfolded proteins due to mutations rendering them resistant to degradation or over-expression of the wild type form.
In the May 2013 issue of Nature, Dr. Zoghbi and colleagues at the Baylor College of Medicine, devised a strategy to identify therapeutic entry points that influence the levels of disease-driving proteins. They applied their approach to spinocerebellar ataxia type 1 (SCA1), a disease cased by expansion of the polyglutamine tract in ataxin 1 (ATXN1), using modulation of the ATXN1 pathway as a proof-of-principle. This model was chosen for several reasons: (1) neurodegeneration in SCA1 parallels with the levels of the mutant ATXN1 protein; (2) over-expression of wild type ATXN1 results in neurodegeneration; and (3) SCA1’s pathogenic mechanisms are well characterized. In order to identify regulators of ATXN1 levels the authors developed a human medullablastoma-derived cell line containing the transgene glutamine-expanded ATXN1 fused to red fluorescent protein (mRFP-ATXN1(82Q)). Next, to distinguish modifiers that regulate ATXN1 protein levels from those affecting transgene transcription they included an internal ribosomal entry site followed by yellow fluorescent protein downstream of ATXN1 (mRFP-ATXN1(82Q)-IRES-YFP). Their screen focused entirely on kinases and kinase like genes based on the fact that ATXN1 phosphorylation is known to be critical for its toxicity and because kinases are pharmacologically targetable. The authors tested 1908 small interfering RNAs (siRNAs) targeting 638 genes and assessing ATXN1 levels as a readout. Subsequently, 50 siRNAs (corresponding to 45 genes) were selected based on their ability to reduce the ratio of RFP to YFP by 2 standard deviations from the mean.
A parallel genetic screen was performed using the Drosophila SCA1 model that expresses ATXN1(82Q). This model can be identified by an external eye phenotype. Here they screened 704 alleles (337 kinase encoding, including shRNA and loss of function mutations) for those that would modify ATXN1 levels. Based on morphological and histological assessments, they identified 51 alleles (49 genes) that suppressed ATXN1 toxicity in vivo. Additionally, human cell-based screens showed 10 human modifier genes that reduced ATXN1 and it’s associated toxicity, corresponding to 8 Drosophila modifiers. Network analysis revealed that the MAPK cascade was the most enriched in both Drosophila and human, where 6/10 genes in human belonged to the canonical MAPK pathway (ERK1, ERK2, MED2, MEK3, MEK6, and MSK1).
ATXN1(82Q) is know to impair motor performance, thus, to determine the effects of the MAPK pathway on the central nervous system, a motor performance test was carried out in Drosophila. Decreasing the MEK, ERK1/2, and MSK1 Drosophila homologues by siRNA lead to increased motor performance and lifespan. Decreasing upstream MAPK pathway homologues suppressed ATXN1(82Q) eye defects and improved motor and lifespan phenotypes. Conversely, constitutively active RAS exacerbated ATXN1 eye degeneration. In human cells, decreasing HRAS and FNTA lead to decreased ATXN1 protein levels, and decreasing RAS homologues reduced ATXN1 in vivo.
Previous studies by Dr. Zoghbi’s group reported ATXN1 levels were sensitive to S776 phosphorylation. Hence, they determined that of MAPK kinases implicated here, MSK1 would be able to phosphorylate the consensus sequence associated with S776. To prove this, they performed an in vitro kinase assay with purified MSK1 and ATXN1 and found robust ATXN1-S776 phosphorylation in both mutant and WT protein forms. Next, cerebellar fractionation assays of WT mice revealed MSK1 was enriched and had increased activity in S776 phosphorylated fractions. Alternatively, immunodepletion of MSK1 from mouse cerebellar extracts lead to decreased S776 phosphorylation.
Next, they sought to determine whether the MAPK pathway could serve as a pharmacological target for SCA1. Human cells expressing ATXN1(82Q) were treated with a PDI84352 (MEK1/2 inhibitor), GW5704 (RAF1 inhibitor), and a Ro31-8220 (MSK1 inhibitor). Pharmacological inhibition of MAPK pathway lead to decreased ATXN1(82Q). Moreover, addition of MAPK inhibitors to cerebellar slices decreased ATXN1 levels.
Lastly, to test the genetic interaction between ATXN1 and MSK1, ATXN1(154Q) knock in mice (Atxn154Q/+) were bred to Msk1+/- Msk2+/- mice. Atxn154Q/+9 week old mice display a motor phenotype that can be quantified using a rotarod test. Breeding of Atxn154Q/+ Msk1+/- Msk2+/- mice lead to better rotarod performance. Owing to the fact that ATXN1 alterations lead to Purkinje cell degeneration, they next determined whether eliminating one copy of MSK1 could rescue the loss of Purkinje cells in another mouse model of ATXN1(82Q), B05/+. Indeed, single copy deletion of Msk1 lead to partially suppressed Purkinje loss phenotype and double MSK1 and MSK2 single copy deletion (B05/+Msk1+/- Msk2+/-), lead to decreased levels of ATXN1.
In summary, Dr. Zoghbi’s group have devised a proof-of-principle strategy that opens many new avenues for the identification of modifiers for neurodegenerative diseases. They utilized combined cross-species genetic screens to identify novel modifiers of ATXN1, and validated in human, mouse, and Drosophila models. This study focused on an early event in pathogenesis that could possibly delay disease onset and progression for this class of neurodegenerative disorders. The RAS-MAPK-MSK1 pathway’s role identified here (phosphorylation of S776-ATXN1) provides a novel pharmacological target for SCA1 and more importantly opens new avenues for combination therapies for this disease. Neurodegenerative disease research has primarily focused on developing treatments for advanced symptoms of neurodegeneration. It would be interesting to determine what the therapeutic benefits are of targeting the RAS-MAPK-MSK1 pathway are on a more advanced form of this disease and whether there would be at least partial reversion of motor defects.
Park, J., et al., RAS-MAPK-MSK1 pathway modulates ataxin 1 protein levels and toxicity in SCA1. Nature.
Emamian,E.S.etal. Serine776 of ataxin-1is critical for polyglutamine-induced disease in SCA1 transgenic mice. Neuron 38, 375–387 (2003).
Jorgensen, N. D. et al. Phosphorylation of ATXN1 at Ser776 in the cerebellum. J. Neurochem. 110, 675–686 (2009).