14 February 2003. A flurry of recent papers is shedding new light on the role of proteins containing expanded polyglutamine tracts (polyQ) and their molecular partners, bringing researchers a small step closer toward understanding the development of such neurodegenerative diseases as Huntington's (HD) and spinocerebellar ataxia.
Knockin Model and The Role of Oligomers
In the February 6 Neuron, Huda Zoghbi, Baylor College of Medicine, Houston, Texas, and her colleagues report that they created a new mouse "knockin" model that reproduces fairly authentically a severe form of spinocerebellar ataxia, which strikes in infancy. In general, longer polyQ repeats cause earlier onset and more severe disease, and in SCA7, lengths of up to 460 glutamines have been reported. The scientists inserted 266 CAG trinucleotides into exon 3 of the SCA7 gene, a procedure that maintains relatively faithfully the levels and spatiotemporal distribution of SCA7. In this way, the model overcomes some of the limitations of existing transgenic models with their nonphysiological promoters.
First author Seung-Yun Yoo and coworkers found that the mice mimicked most of the symptoms seen in SCA7 patients, including weight loss, visual impairment, tremors, and muscle atrophy. They found that the mutant ataxin-7 protein was barely detectable in young mice but accumulated with age, in part because the expanded polyQ tract stabilized the mutant protein. Mutant ataxin-7 amassed faster in neurons of the retina and cerebellum, two areas that are most severely damaged in infantile SCA7. In the retina, small aggregates of mutant SCA7 were detectable at 12 weeks, and nuclear inclusions developed during the next three weeks. Interestingly, however, obvious neurological problems preceded inclusions by several weeks, with nine-week-old mice already having impaired rod function. The authors suggest that the amount of accumulated mutant protein rather than the presence of nuclear inclusions determines their effect on neuronal function and viability.
This begs the question of how important the first, early aggregates of polyglutamine expanded proteins might be to disease progression. Important, indeed, a letter in the January 23 Nature by Junying Yuan and colleagues at Harvard Medical School would suggest. First author Ivelisse Sanchez and colleagues used the azo-dye Congo red, which binds the protein β-sheets that provide a backbone for such aggregates, to prevent the formation of polyQ oligomers in vivo and in vitro, and then analyzed the consequence of that on cell death, protein synthesis and degradation in neurons.
The authors show that pure, recombinant polyQ protein forms oligomers that can be captured on filters. Congo red prevents these oligomers from forming, and presumably works in this fashion to ameliorate symptoms caused by polyQ expansions in vivo.
Sanchez et al. found that in cells expressing a toxic polyQ protein, the dye prevented ATP depletion and the reduction in protein synthesis that precedes cell death. It reduced caspase-activated programmed cell death pathways by over half, a finding that was reflected in a similar reduction in the numbers of dying cells. The addition of Congo red obliterated polyQ aggregates seen in transgenic neuroblastoma cells.
The authors also present data suggesting that Congo red may reverse damage caused by polyQ-expanded inclusions. Added to cell lysates containing aggregates, the dye dissolved the aggregates. This also seems to occur in vivo. Sanchez and colleagues administered the dye to transgenic R62 mice, which express a truncated form of the huntingtin protein containing a stretch of 139 glutamine residues. They exhibit developmental and neurological symptoms, including weight loss, diabetes, and loss of motor function. Congo red given to nine-week-old animals ameliorated all of these symptoms. The mice gained weight over a two-week period, recovered control of their hind limbs, and had blood sugar levels similar to wild-type mice. This would suggest that Congo red belongs on the list of aggregate busters (see ARF related news story).
The role of polyQ oligomers in the formation of polyQ aggregates has been controversial. In a variation on this theme, researchers led by Howard Green at Harvard Medical School step into the fray by proposing in this week's early online PNAS that polyQ oligomers may actually be held together by covalent bonds. First author Shiro Iuchi, with colleagues from Harvard and the CNRS in Paris, shows that in PC12 cells inclusions formed by a chimeric polyQ protein containing 205 glutamines can be dissolved in 95 percent formic acid-a treatment previously shown to succeed as a solvent when the detergent SDS fails. However, when Iuchi et al. examined this acid-treated protein on Western blots, it became clear that oligomers of the protein had not been separated. The authors also detected similar acid-resistant oligomers, and indeed polymers too large to enter polyacrylamide gels, in brain tissue from juvenile Huntington's patients by using a filter retention assay and immunofluorescence. These polymers were not detected in the cerebellum, but were abundant in the cortex, where mutant huntingtin inclusions are readily detected, Iuchi et al. report.
Overall, the data indicate that polyglutamine inclusions may be stabilized by intermolecular covalent bonds. Iuchi and colleagues point out that polyglutamine tracts have been shown to be excellent substrates for transglutaminase reactions, which couple the glutamine residues to lysines on other peptide chains (see also Karpuj et al., 2003. These reactions have repeatedly been implicated in aggregate formation in other neurodegenerative diseases, most recently by Junn et al., 2003.
The Transcription Axis
R62 mice also were the model of choice for Gillian Bates, King's College, London, in her study testing histone deacetylase inhibitors as a potential treatment for Huntington's disease. By post-translationally modifying histone proteins, histone deacetylases can have a profound effect on gene transcription and have been touted as potential regulators of the transcriptional changes that accompany HD (see ARF related news story). In the February 7 early online edition of PNAS, Bates, with colleagues in the U.K. and the U.S., show that suberoylanilide hydroxamic acid (SAHA) can alleviate Huntington's-like symptoms in R62 mice.
SAHA is notoriously insoluble in aqueous solutions, but first author Emma Hockly and colleagues found that the inhibitor could be administered in drinking water when complexed with cyclodextrins. Furthermore, they found that the drug readily crosses the blood-brain barrier, leading to increased histone acetylation in the brain. SAHA improved the motor function of R62 mice, as judged by performance on the rotarod, the mouse version of the slippery log. Mice given SAHA hung on to the rotarod almost twice as long as their impaired littermates. However, other results were disappointing. Inhibitor-treated animals lost weight faster than those on placebo, and showed almost no improvement in grip strength, and, not surprisingly, there was no effect on the levels of polyQ aggregation. Nevertheless the authors suggest that the improvement in motor function validates the pursuit of histone deacetylase inhibitors as HD therapeutics.
Huntington’s and Diet
Seemingly paradoxically for a disease where people become emaciated, a finding from Mark Mattson's lab at the National Institute on Aging shows that in a mouse model of Huntington's disease, dietary restriction (DR) actually prevents weight loss. Furthermore, these mice, fasted every other day, live longer and have significantly better motor function than their littermates fed at will. The results also appear in this week's early online PNAS.
First author Wenzhen Duan and colleagues correlate these behavioral studies with molecular and biochemical data. They found a significant reduction in the number of huntingtin-positive inclusions in both cortex and striatum of DR mice as compared to the fully fed mice, and this was accompanied by marked improvement in brain tissue atrophy. The latter may be due in part to lowered activation of the apoptotic trigger, caspase-1; Duan et al. found almost an fivefold greater caspase-1 activation in the cortex and striatum of free-fed animals. Dietary restriction also improved blood glucose homeostasis; HD animals are typically hyperglycemic, but the fasted animals had almost normal blood glucose levels and performed better in a glucose tolerance test.
With fasting, levels of brain-derived neurotrophic factor (BNDF) and the chaperone HSP 70 also increased. These proteins are both implicated in stress resistance, which, according to the authors, may be impaired in HD. The authors also speculate that "dietary intervention may suppress the disease process and increase the life span of humans that carry the mutant huntingtin gene." (See ARF related news story on calories, insulin, and lifespan, and ARF related news story on caloric restriction and Alzheimer's disease.)
What the Worm Can Teach Us
Meanwhile, work to find new partners for Huntingtin continues apace. Guy Caldwell and colleagues at the University of Alabama, Tuscaloosa, report that torsin proteins can have profound effects on polyglutamine protein aggregation by functioning as natural aggregation busters.
Mutations in torsinA are responsible for the severest early-onset form of dystonia reported in humans, but the exact function of torsins has remained a mystery. Now, working with the roundworm Caenorhabditis elegans, Caldwell and colleagues show that overexpression of the C. elegans TOR-2 protein can dramatically reduce aggregation of a green fluorescent protein-polyQ hybrid, as seen in whole-animal images. Their results adorn the cover of the February 1 Human Molecular Genetics.
The authors used photomicrographs to reveal that TOR-2 significantly reduces the mean size of polyQ aggregates, and that this suppression of aggregation continues as the animals aged. Furthermore, worms that carry mutant TOR-2 inactivated by deletion of the serine residue 368 (which mimics the mutant implicated in human dystonia) continue to amass polyQ aggregates and have exacerbated developmental problems.
So just what is TOR-2 doing? Caldwell et al. localized it to the aggregates, around which it seems to form a ring-like structure reminiscent of similar torsinA donuts seen around Lewy bodies (see Sharma et al., 2001). Since torsins belong to the AAA (ATPases Associated with diverse cellular Activities), a disparate group including heat shock proteins and chaperones, the authors suggest that torsins may work as molecular chaperones as part of a cellular mechanism for the management of protein misfolding. In this regard, it is worth noting that Margaret Pericak-Vance and colleagues have found potential markers for Parkinson's disease near the human torsinA locus (see ARF related news story).
Also with help from the not-so-lowly worm, Christian Neri, from the Center for Study of Human Polymorphisms, Paris, together with colleagues in France and the US, identified new partners for huntingtin. First author Sebastien Holbert and colleagues used a yeast two-hybrid system to find C. elegans proteins that bind to huntingtin. One hit, the protein K08E3.3b, only activated the system if the first 152 amino acids of huntingtin were present, and the activation became stronger as glutamines were added to the huntingtin hybrid, 128 glutamines being almost twice as potent as 18.
The authors then searched the human genome for homologs to the C. elegans protein and found CIP4 (Cdc42-interacting protein) to be highly similar. Subsequent biochemical tests showed that CIP4 binds to huntingtin in extracts of lymphoblastoma cells taken from control and HD patients, while immunohistochemical analysis revealed the presence of CIP4 in the caudate nuclei and frontal cortex of HD brain. In the former case, the amount of CIP4 increased with disease severity, while in the latter case the authors show that CIP4 colocalized with ubiquitin, commonly found in inclusion bodies. Western blots showed a dramatic increase in CIP4 expression in the striatum of people with HD.
The role of CIP4 is unclear, but when the authors overexpress it in striatal neurons in vitro, it induces cell death, leading them to conclude that "CIP4 accumulation and cellular toxicity may have a role in HD
pathogenesis." This work on CIP4 follows previous work by the same laboratory on the identification of a candidate modifier gene in HD, the transcriptional regulator CA150 (Holbert et al., 2001), and the creation of a transgenic C. elegans model of polyglutamine-dependent neuronal dysfunction without cell death that may allow the early steps of HD to be
studied genetically (see also Parker et al., 2001).-Tom Fagan.
Yoo S-Y, Pennesi ME, Weeber EJ, Xu B, Atkinson R, Chen S, Armstrong DL, Wu SM, Sweatt JD, Zoghbi HY. SCA7 knockin mice model human SCA7 and reveal gradual accumulation of mutant ataxin-7 in neurons and abnormalities in short-term plasticity. Neuron. 2003 Feb 6;37:383-401. Abstract
Parker JA, Connolly JB, Wellington C, Hayden M, Dausset J, Neri C. Expanded polyglutamines in Caenorhabditis elegans cause axonal abnormalities and severe dysfunction of PLM mechanosensory neurons without cell death. Proc Natl Acad Sci U S A 2001 Nov 6;98(23):13318-23. Abstract
Sanchez I, Mahlke C, Yuan J. Pivotal role of oligomerization in expanded polyglutamine neurodegenerative disorders. Nature. 2003 Jan 23;421:373-379. Abstract
Iuchi S, Hoffner G, Verbeke P, Djian P, Green H. Oligomeric and polymeric aggregates formed by proteins containing expanded polyglutamine. Proc Natl Acad Sci U S A. 2003 Mar 4;100(5):2409-14. Abstract
Hockly E, Richon VM, Woodman B, Smith DL, Zhou X, Rosa E, Sathasivam K, Ghazi-Noori S, Mahal A, Lowden PAS, Steffan JS, Marsh JL, Thompson LM, Lewis CM, Marks PA, Bates GP. Suberoylanilide hydroxamic acid, a histone deacetylase inhibitor, ameliorates motor deficits in a mouse model of Huntington's disease. Proc Natl Acad Sci U S A. 2003 Feb 18;100(4):2041-6. Abstract
Duan W, Guo Z, Jiang H, Ware M, Li X-J, Mattson, MP. Dietary restriction normalizes glucose metabolism and brain-derived neurotrophic factor levels, slows disease progression, and increases survival in huntingtin mutant mice. Proc Natl Acad Sci U S A. 2003 Mar 4 ; 100(5):2911-6. Abstract
Caldwell GA, Cao S, Sexton EG, Gelwix CC, Bevel JP, Caldwell KA. Suppression of polyglutamine-induced protein aggregation in Caenorhabditis elegans by torsin proteins. Human Mol. Genetics 2003;12:307-319. Abstract
Holbert S, Dedeoglu A, Humbert S, Saudou F, Ferrante RJ, Neri C. The cdc42-interacting protein 4 binds to huntingtin: neuropathological and biological evidence for a role in Huntington's disease pathogenesis. Proc Natl Acad Sci U S A. 2003 Mar 4 ; 100(5):2712-7. Abstract