It is famous for robbing Lou Gehrig of his life and Stephen Hawking of his mobility and voice, but just how amyotrophic lateral sclerosis (ALS) destroys motor neurons in the brain and spinal cord remains a mystery. Now, scientists are converging on an explanation, at least for a fraction of the ALS cases caused by a specific mutation. In cells with the mutation, the new work shows, pores in the membrane separating the nucleus and cytoplasm become clogged, preventing vital molecules from passing through and creating a fatal cellular traffic jam.
For now, the work applies only to the mutation dubbed C9orf72—a DNA stutter in which a short nucleotide sequence, GGGGCC, is repeated hundreds to thousands of times in a gene on chromosome 9. Nor do the multiple labs reporting results this week agree on exactly what plugs those nuclear pores and how the cells die. Still, the work is “a major breakthrough” in ALS research, says Amelie Gubitz, program director of the neurodegeneration division at the National Institute of Neurological Disorders in Bethesda, Maryland. The groups worked independently, starting with different hypotheses and experimental designs, yet reached similar conclusions, making the finding more convincing. And it suggests that boosting traffic through nuclear pores could be a new strategy for treating some cases of ALS and frontotemporal dementia (FTD), another neurodegenerative condition C9orf72 can cause.
Based on past work by their own and other groups, neuroscientists Jeff Rothstein and Tom Lloyd at Johns Hopkins University in Baltimore, Maryland, suspected that the long strands of excess RNA produced by C9orf72 cause neurodegeneration by binding to, and thus sequestering, key cellular proteins. The team tested the idea in fruit flies with the mutation, which display damage in the nerve cells of their eyes and in motor neurons. They boosted the activity of 400 genes that encode candidate proteins. Increased amounts of one of them, RanGAP, which helps shuttle molecules through the nuclear pore, completely prevented neuronal degradation in the flies, the group reports in Nature this week. In neurons taken from flies as well as derived from stem cells from people with the C9orf72 mutation, increasing RanGAP improved nuclear traffic in the cells, clearing up abnormally distributed protein aggregations, the team also found.
Compared with the nucleus of a normal fly neuron (above), one from a fly with a common ALS mutation (below) has choppy clusters of nuclear pore protein (red) around its edges. In green, an abnormal cluster of the protein poly(GP).
IMAGES: YUBING LU OF THE UNIVERSITY OF MASSACHUSETTS MEDICAL SCHOOL (2)
Those findings are compelling in light of two “gorgeous” papers by other groups, says Steven McKnight, a biochemist at the University of Texas Southwestern Medical Center in Dallas. Unlike Rothstein and Lloyd, neither group started with preconceptions about which genes might counteract C9orf72's toxicity, he notes. A group led by molecular biologist Paul Taylor of St. Jude Children's Research Hospital in Memphis, Tennessee, and neuroscientist Fen-Biao Gao of the University of Massachusetts Medical School in Worcester, for example, performed an unbiased screen of more than 9000 fruit fly genes, looking for DNA variants that either suppressed or exaggerated the mutation's deleterious effects. The team reports in Nature a very high hit rate among genes involved with nuclear pore transport, in line with the RanGAP findings.
Rothstein and Lloyd believe that what plugs the nuclear pores in C9orf72 carriers is the abnormal RNA itself, interacting with RanGAP and other cellular proteins. They found, for example, that they could counteract the toxic effects of the mutation by neutralizing the RNA with complementary nucleotide strands. But a third study, in Nature Neuroscience, points to a different villain.
Stanford University geneticist Aaron Gitler, in Palo Alto, California, and his team performed an unbiased genetic screen of their own. They genetically engineered yeast to make Pro-Arg, one of the proteins encoded by the C9orf72 mutation, but not to make long repetitive strands of RNA. Pro-Arg is found in small concentrations in the postmortem brain tissue of people with ALS, and previous studies have shown that protein alone is toxic to neurons as well as to yeast. Gitler's study, like the other studies, showed that boosting nuclear transport helped the cells survive. The fact that the typical C9orf72 RNA was not present, however, persuades Gitler that Pro-Arg, and perhaps the other C9orf72-spawned proteins, are what block the nuclear pores—and would be the best targets of any therapeutic strategy.
Taylor believes “a case is building” for those proteins as the pore plugs, but he notes that his own results provide some support for both the RNA and protein hypotheses. “What's really intriguing is that three different groups coming from different directions” all now agree that blockage of the nuclear pore is the central problem caused by the C9orf72 mutation, he says. Given that nuclear pores are known to degrade even in normal aging, “it is enticing to speculate” that their blockage or breakdown could play a role in a range of neurodegenerative processes, Rothstein's group adds.
Because fewer than 10% of ALS cases are linked to the C9orf72 mutation, Gubitz says, the new results may not explain how other forms of ALS kill nerve cells, or why the mutation sometimes causes FTD instead. Still, she says, the abnormal buildup of proteins in the cell nucleus or cytoplasm—particularly of a protein called TDP-43—is a “unifying theme” across ALS and FTD cases.
An important next step, Gubitz adds, is to study cases of ALS and FTD without a known mutation to see whether those people have defects in genes governing nuclear transport. An answer could come from a new collaboration announced this month: The pharmaceutical company Biogen, the ALS Association, and Columbia University Medical Center plan to perform whole-genome sequencing of 1500 people with both inherited and sporadic forms of the disease.
Ke Zhang,1, *
Christopher J. Donnelly,2, *
Aaron R. Haeusler,3,
Jonathan C. Grima,2, 4,
James B. Machamer,1,
Elizabeth L. Daley,2,
Sean J. Miller,2,
Kathleen M. Cunningham,1,
Michael A. Thomas,2,
Richard L. Huganir,4,
Lyle W. Ostrow,1,
Michael J. Matunis,3,
Thomas E. Lloyd1, 4, §
& Jeffrey D. Rothstein2, 4, §