20 February 2012. Two potential approaches for tackling Huntington’s disease come to the fore in this week’s literature—an intriguing mechanism begging for a drug, and a promising compound with an unknown mechanism. In the February 17 Cell, researchers led by Tzu-Hao Cheng of National Yang-Ming University in Taipei, Taiwan, used a yeast screen to identify a protein needed for transcribing genes with long trinucleotide repeats. Strategies that stymie this transcription elongation factor (SPT4) could help a broad variety of disorders, including HD, experts say. Meanwhile, in a paper posted online February 13 in Proceedings of the National Academy of Sciences USA, Simonetta Sipione of the University of Alberta in Edmonton, Canada, and colleagues report that brain injections of a membrane lipid can rescue motor function in severely disabled HD model mice.
Huntington’s is one of many diseases caused by genes with CAG trinucleotide repeats. These encumber the encoded protein with unusually long glutamine stretches. Saddled with these polyglutamine expansions, culprit proteins deform and clump together, causing cells to malfunction and die. In the Cell paper, first author Chia-Rung Liu and colleagues screened for genetic transpositions that restore function to polyQ-expanded ADE2 (an enzyme involved in purine biosynthesis) in the yeast Saccharomyces cerevisiae. Of some 200 candidate genes identified from ~180,000 colonies examined, the researchers excluded most because they contained contracted 97Q repeats or insertions in genes (Hsp104 and Rnq1) already known to promote aggregation of polyQ proteins. That left Cheng’s team with a single clone having a transposon in the SPT4 gene, which, along with its human homolog Supt4h, is highly conserved among eukaryotes. SPT4 promotes transcription by binding to RNA polymerase II, helping it stay attached to DNA.
In subsequent experiments, the scientists found that SPT4-deficient cells made less 97Q-ADE2 mRNA and protein, but normal amounts of ADE2 with a shorter polyQ stretch, suggesting the elongation factor is needed for transcription of primarily long CAG nucleotide repeats. Furthermore, chromatin immunoprecipitations showed less RNA polymerase II bound to DNA sequences downstream of long CAG repeats in SPT4-deficient versus normal yeast cells. However, SPT4 deficiency had no effect on RNA polymerase II binding to sequences with short CAG stretches. The vast majority of yeast genes transcribed normally in SPT4 knockouts. By comparison, yeast cells with defects in other factors that regulate RNA polymerase II activity had widespread problems with protein synthesis. The data suggest that cells need SPT4 to get their transcriptional machinery across tricky templates with long CAG expansions.
Does the same hold true in human neurons? As a first step toward testing mammalian systems, the researchers used short interfering RNAs (siRNAs) to reduce expression of Supt4 in striatal neurons from HD mice that express mutant huntingtin (Htt). Treated cells transcribed less Htt but produced normal amounts of other mRNAs. And in a mouse striatal cell line (ST14A), Supt4 knockdown reduced aggregation and toxicity of an 81Q fluorescent protein, but did not affect the 7Q version of the same protein.
“The idea that there’s a factor critical for transcription of these expanded repeats is really intriguing and completely novel,” said Harry Orr of the University of Minnesota Medical School in Minneapolis. Designing ways to block Supt4 function would target “a critical aspect of trinucleotide repeat diseases, and would perhaps apply to a broad number of such disorders,” Orr said, noting that further work is needed to determine if the effects occur in mammals in vivo, and if Spt4 deficiency impacts critical genes.
SPT4 came as a surprise, Cheng told ARF. Initially, his team figured the yeast screen would turn up a chaperone, as loss of these proteins has been shown to curtail aggregation (Krobitsch and Lindquist, 2000; Meriin et al., 2002), and hence toxicity, of polyQ proteins. Given the findings, “it may be possible to block the [polyQ toxicity] process at a stage much earlier than protein aggregation,” he said.
Restoring Motor Function in HD Mice…But How?
Whereas Cheng’s group used an unbiased screen to identify factors mediating toxicity of polyQ proteins in general, Sipione and colleagues took a decidedly biased approach, as described in their PNAS paper. They focused on a compound (ganglioside GM1) that induces post-translational changes in huntingtin, and showed that injecting the compound into the brains of HD mice completely reverses their motor dysfunction.
Levels of GM1—a membrane lipid highly enriched in the brain—are down in HD mouse models, and in skin cells and postmortem brain tissue from HD patients. Since GM1 plays key roles in cell signaling and neuronal interactions within the brain, “we thought that by correcting this deficit, i.e., restoring GM1 to normal levels, we could reduce the symptoms in an HD transgenic model,” Sipione told ARF.
Sure enough, the researchers found that administering GM1 to HD cell models improved their survival, whereas inhibiting GM1 synthesis in wild-type striatal cells made the neurons more susceptible to death (Maglione et al., 2010). In the current study, first author Alba Di Pardo and colleagues tested whether restoring brain GM1 levels could treat HD mice.
They injected GM1, or control saline into the right ventricle of YAC128 mice. These animals express polyQ-expanded human Htt from a yeast artificial chromosome and develop motor problems resembling the human disease (Slow et al., 2003). The mice received a 28-day GM1 infusion starting at five months of age, when motor deficits were readily apparent. Performance on several standard motor tests (rotarod, horizontal ladder walking, narrow beam) returned to wild-type levels after two weeks of treatment, and lasted another two weeks after the researchers stopped the infusion.
At the molecular level, the ganglioside increased striatal expression of DARPP-32, a protein that is downregulated in HD, and restored normal levels of DARPP-32 phosphorylation at threonine 34. In addition, striatal and cortical neurons from GM1-treated YAC128 mice had elevated Htt phosphorylation at serine residues 13 and 16. These modifications seem to be neuroprotective, as expression of Htt that mimics constitutive phosphorylation at these two serines delayed disease in an HD model developed by X. William Yang and colleagues at the University of California, Los Angeles (see ARF related news story on Gu et al., 2009).
Based on these data, “we thought those phosphorylations might act as a molecular switch to abolish disease,” Yang said. “It’s exciting to learn that this switch can be controlled in a beneficial way in an animal model by a drug-like molecule.”
The potential benefits of boosting S13 and S16 Htt phosphorylation was suggested in a recent report by coauthor Ray Truant of McMaster University, Hamilton, Canada (Atwal et al., 2011). However, that study used a different class of drugs—kinase inhibitors—to enhance Htt serine phosphorylation. “This work says there may be more than one way to do this,” Truant noted.
Alongside its dramatic restoration of motor behavior, does GM1 rescue neurodegeneration? It is hard to say, in part because YAC128 mice show no signs of cell loss or brain atrophy until 10 to 12 months of age. To test whether GM1 curbs neuron loss, Sipione said her lab is doing studies in the R6/2 mouse, an HD model that develops severe disease quickly, showing neurodegeneration by 10 to 12 weeks and dying within four months. The researchers are also using R6/2 mice to assess GM1’s effect on cognition, which they did not test in YAC128 mice.
In another line of follow-up work, Sipione’s lab is trying to uncover the mechanisms by which GM1 protects HD mice. “We are looking for kinases that may be activated by GM1, or for altered balance between kinases and phosphatases in GM1-treated HD cells,” Sipione told ARF. “We think more than one signaling pathway may be regulated by GM1.”
Further work is needed to determine whether the compound can improve symptoms if administered peripherally, and whether long-term use would be safe, noted Wenzhen Duan of Johns Hopkins University in Baltimore, Maryland. GM1 was tested in a five-year Phase 2 trial involving 150 Parkinson’s disease patients receiving subcutaneous injections of the compound. The treatment was well tolerated and safe, and seemed to provide some clinical benefit (Schneider et al., 2010).
Joseph Mazzulli of Massachusetts General Hospital, Charlestown, commented that, while details behind GM1’s beneficial effects in the HD mice remain to be sorted out, studies on GM1 in Parkinson’s disease models suggest the molecule acts by slowing aggregation of the PD culprit protein α-synuclein. “It would be interesting to assess whether GM1 can also directly affect the aggregation rate of mutant huntingtin protein,” Mazzulli wrote in an e-mail to ARF. In addition, he noted, “a pathological analysis of the striatum would be instructive, and may provide additional clues into the mechanism of GM1 treatment.” (See full comment below.)—Esther Landhuis.
Liu CR, Chang CR, Chern Y, Wang TH, Hsieh WC, Shen WC, Chang CY, Chu IC, Deng N, Cohen SN, Cheng TH. Spt4 Is Selectively Required for Transcription of Extended Trinucleotide Repeats. Cell. 17 Feb 2012. Abstract
Di Pardo A, Maglione V, Alpaugh M, Horkey M, Atwal RS, Sassone J, Ciammola A, Steffan JS, Fouad K, Truant R, Sipione S. Ganglioside GM1 induces phosphorylation of mutant huntingtin and restores normal motor behavior in Huntington disease mice. Proc Natl Acad Sci U S A. 2012 Feb 13. Abstract