28 May 2009. Addressing the longstanding question of whether protein aggregates are a root cause of disease or simply result from ongoing pathogenesis, a study in this week’s PNAS Early Edition suggests the answer may depend on their three-dimensional conformation. Motomasa Tanaka and colleagues at the RIKEN Brain Science Institute in Wako City, Japan, coaxed huntingtin proteins to adopt distinct conformations in vitro under different temperatures, and showed that shape in large part determined toxicity. Furthermore, these proteins shared key structural features with huntingtin proteins isolated from the brains of Huntington disease (HD) mice, and these parallels correlated with differential disease vulnerability in those brain areas. Besides pointing to new therapeutic strategies aimed at blocking formation of the more toxic huntingtin conformation, the findings tie in more broadly with emerging insight into different structures of aggregates in a given disease protein, and their seeding behavior.
A striking pathological feature of HD is the clumping of mutant huntingtin (htt) proteins due to exon 1 CAG repeats encoding long strings of glutamine that make the proteins less soluble. These rogue proteins can also aggregate differently depending on other conditions, such as temperature. Using two different temperatures (4 and 37 degrees Celsius) to polymerize bacterially purified htt proteins containing 42 glutamine repeats, first author Yoko Nekooki-Machida and colleagues produced htt aggregates of two distinct conformations. Though both formed β-sheet-rich fibrils (i.e., amyloids) and had similar morphology, the 4-degree conformation had loop/turn structures that made it more fragile and heat-sensitive, compared to the sturdier 37-degree form that had more intermolecular β-sheets. The researchers further determined that polymerizing at 4 degrees Celsius induces a misfolded htt conformation with exposed glutamines, whereas the glutamines in the 37-degree conformation are more buried and less amenable to detection by anti-polyglutamine antibody.
Using a lipid-based procedure, Tanaka’s team next delivered these in-vitro htt amyloids into mouse neuroblastoma cells that themselves overexpress a green fluorescent protein (GFP)-htt fusion containing expanded polyglutamine. With this experiment, Tanaka’s team showed that the in-vitro amyloids were able to enhance aggregation of endogenous GFP-htt. What’s more, these endogenous aggregates maintained the conformation of the “seeding” protein. That is, GFP-htt proteins seeded by 37-degree amyloids were more heat-stable than those seeded by 4-degree amyloids. These findings indicate that the in-vitro htt amyloids can enter mammalian cells and drive endogenous htt to form aggregates in a manner that maintains the conformation of the “seeding” amyloid.
The clincher in this study was a set of experiments showing that these distinct htt conformations conferred differential toxicity. When introduced into neuroblastoma cells overexpressing GFP-htt with expanded polyglutamines, the 4- and 37-degree in-vitro conformations induced aggregation of endogenous GFP-htt to a similar extent. However, the 4-degree amyloid was more toxic to the host cells. Furthermore, in R6/2 transgenic mice expressing mutant human htt, the researchers found region-specific parallels to the temperature-dependent structural and toxicity differences in the in-vitro amyloids. Htt amyloids purified from striatum, a highly vulnerable brain region in HD, were the most heat-sensitive and had loop/turn structures similar to those found in the 4-degree in-vitro amyloids. In contrast, htt amyloids from R6/2 hippocampus and cerebellum, less affected brain regions in HD, were more heat-stable and structurally resembled the less toxic 37-degree conformation.
Evidence of conformation-dependent toxicity extends to other proteins besides huntingtin. As a postdoc in Jonathan Weissman’s lab at the University of California, San Francisco, Tanaka had shown that three-dimensional structure plays a key role in determining the infectiousness of a yeast prion protein (Tanaka et al., 2005 and ARF related news story). A recent finding that transplant tissue had developed Lewy pathology after a decade or so of surviving in the brain of Parkinson’s patients has raised the question among some scientists of whether pathogenic α-synuclein conformations, too, can “infect” healthy neurons and induce corresponding changes in them (Kordower et al., 2008; Kordower et al., 2008). Moreover, in a fruit fly model of Aβ accumulation, changing Aβ’s aggregation capability led to different patterns of neuronal death (see ARF related news story), and there is growing awareness of different β amyloid “strains” in Alzheimer disease (Rosen et al., 2009). Finally, other recent studies suggest that mutant tau can seed a conformation change in wild-type tau (Frost et al., 2009 and ARF related Keystone story), and that the toxicity of tau proteins derives largely from their ability to adopt a β-structure and form aggregates (Mocanu et al., 2008 and ARF related news story).
The current study beefs up the idea that three-dimensional structure may figure heavily in whether aggregates of a protein are pathogenic. In the case of the huntingtin protein, “it is possible that flexible and exposed polyglutamines easily interact with and sequester other functional proteins into htt aggregates and thereby lead to cell death, whereas the limited dynamics of the polyglutamines buried into an amyloid core exert only modest toxic or non-toxic effects,” the authors write. This argument broadly agrees with a hypothesis advanced by other scientists, as well (e.g., Chiti and Dobson, 2006).—Esther Landhuis.
Nekooki-Machida Y, Kurosawa M, Nukina N, Ito K, Oda T, Tanaka M. Distinct conformations of in vitro and in vivo amyloids of huntingtin-exon 1 show different cytotoxicity. PNAS Early Edition. 2009 May.