Amidst the debate around whether pathogenic mutations of presenilin harm cells by a gain of toxic function, or a loss of essential normal function (see ARF related news story), new results with another disease-causing protein suggest that in some cases the answer may be not one or the other, but both. In a report in the March 12 issue of Nature, Huda Zoghbi and colleagues at Baylor College of Medicine in Houston, Texas, provide evidence that the same polyglutamine expansion mutation in ataxin, the cause of spinocerebellar ataxia-1, can lead to simultaneous gain and loss of function depending on which cellular proteins ataxin partners with.

First author Janghoo Lim and coworkers show that in cells, ataxin participates in at least two large native protein complexes. On the one hand, polyglutamine expansion and phosphorylation, two features required for pathogenesis, promote the formation of complexes with a newly identified partner, RBM17. These complexes play a role in neurodegeneration, in an apparent gain of function. At the same time, there is less ataxin available for a physiological interaction with capicua (see ARF related news story), creating a partial loss of function.

Evidence that a loss of function contributes to disease in conjunction with a dominant gain-of-function effect comes from mice, where animals expressing solely polyglutamine-expanded ataxin have a worse disease phenotype than animals that also carry a (presumably protective) wild-type allele.

“Our finding that the interactions of the mutant protein with its usual partners are differentially affected by the poly-glutamine expansion offers a mechanistic explanation for how mutant proteins can gain and lose function simultaneously,” the authors write.—Pat McCaffrey


  1. This report provides evidence that 1) disease-causing polyglutamine expansion in ataxin-1 can lead to increases in ATX1 assembly into complexes containing RNA-binding motif protein 17 (RBM17) and decreases in its assembly into complexes containing capicua (CIC); 2) the interaction between ATX1 and RBM17 may involve phosphorylation of a particular serine in ATX1; 3) genetic interaction between ATX1 and RBM17 in the Drosophila eye contributes to retinal degeneration; 4) RBM17 and CIC compete with each other for ATX1 complexation; and 5) compared to mice that have one polyQ mutant allele and one wild-type allele, loss of the wild-type ATX1 allele exacerbates neuropathology. This last point is particularly meaningful, because ATX1 knockout mice do not develop the neurodegenerative phenotype.

    Taken together, the authors suggest that polyQ expansion of ATX1 leads to neurodegeneration through both a gain-of-function interaction with RBM17 and a loss-of-function interaction with CIC. In this regard, the evidence for the general idea that duel gain/loss of function contributes to the phenotype seems much stronger than the (basically correlative) evidence for the specific involvement of gain/loss of biochemical interaction with RBM17 and CIC. The authors further suggest that the duel gain/loss of function may be a common theme with dominant mutations in certain other genes associated with neurodegenerative diseases, including Alzheimer’s. Indeed, dominant presenilin-1 (PS1) mutations associated with familial AD display both a gain of function (increase in 42-residue amyloid-β peptide versus the 40-residue variant) and a loss of function (decreased γ-secretase proteolytic activity). In this case, however, the gain of aggregation-prone Aβ42-to-Aβ40 ratio may be a direct consequence of the reduced presenilin proteolytic function (1,2). That is, both the gain and the loss of function with presenilin are intimately associated with the protein’s proteolytic activity as part of the γ-secretase complex, and not due to the gain of novel interactions.

    Nevertheless, with both ATX1 and PS1, it appears that the gain of toxic function is the primary cause of neurodegeneration and that the loss of function may simply exacerbate the phenotypes. For ATX1, complete loss of the protein does not lead to neurodegeneration, but overexpression of the mutant protein is sufficient to cause disease in mice. For PS1 and its homolog PS2, among the more than 100 known FAD-associated mutations, none lead to complete loss-of-function mutations or to a truncated protein. Studies in mice show that a single wild-type PS1 allele is sufficient for normal function: PS1+/- and PS2-/- mice are viable and do not display neurodegeneration. Neurodegeneration can be seen in mice in which both PS1 and PS2 are conditionally knocked out in CNS neurons, suggesting that the loss of presenilin function, independent of its effects on Aβ production, may contribute to the neurodegeneration seen in FAD associated with mutant PS1. This situation, however, has never been observed in human PS-mutant FAD, in which three normal PS alleles remain, and the fourth (mutant) allele possesses partial activity, not complete loss.


    . Loss-of-function presenilin mutations in Alzheimer disease. Talking Point on the role of presenilin mutations in Alzheimer disease. EMBO Rep. 2007 Feb;8(2):141-6. PubMed.

    . When loss is gain: reduced presenilin proteolytic function leads to increased Abeta42/Abeta40. Talking Point on the role of presenilin mutations in Alzheimer disease. EMBO Rep. 2007 Feb;8(2):136-40. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. For Lack of a Stop Codon
  2. Toxicity of Polyglutamine Expansion Follows Normal Channels

Further Reading

Primary Papers

  1. . Opposing effects of polyglutamine expansion on native protein complexes contribute to SCA1. Nature. 2008 Apr 10;452(7188):713-8. PubMed.