Before they were amyloid: understanding the toxicity of disease-associated monomers and oligomers prior to their aggregation

2015-07-10Many degenerative diseases are associated with proteins that misfold and aggregate into amyloid fibrils that are deposited intra‐ or extracellularly at or near the site of tissue degeneration. It has long been established that the protein species conferring toxicity in each case is most likely an intermediate on the pathway to fibrillization or an off‐pathway oligomer rather than the fibrils themselves, which are inert and could even exist as a protective mechanism. However, while the current understanding of the toxicity of amyloid intermediates remains rudimentary, research upon the conformational behavior and possible toxicity of amyloidogenic proteins in monomeric or oligomeric (but not aggregated) form is still more severely lacking. Such research should prove crucial to the development of protein-targeting pharmaceuticals for treatment of these diseases; this speculation is based on evidence, part of which has originated in my own research, that, depending on conditions, these proteins as monomers or oligomers can severely damage membranes and/or even in solution form unfavorable structures. My research focuses on the cases of α-synuclein, islet amyloid polypeptide (IAPP), and huntingtin, amyloidogenic peptides associated with cell death in the substantia nigra in Parkinson’s disease, pancreatic β-cell death in type II diabetes, and spiny neuron degeneration in Huntington’s disease, respectively. ❧ Structural studies have shown α-synuclein to assume multiple membrane‐binding modes when in contact with phospholipids, including an extended amphipathic α-helix and a broken‐helix conformation consisting of two shorter α-helices interrupted by a short linker region. Other experiments have shown α-synuclein to induce the formation of highly curved membranous structures upon binding to negatively charged phospholipid bilayers. I reconciled these findings by probing the helical structure of α-synuclein bound to membranes made up of compositions corresponding to the latter findings. Using an environment‐sensitive fluorophore, I found the α-synuclein binding to these membranes to be in the extended α-helical form. ❧ My work with IAPP was more extensive. Using a variety of techniques, from clearance assays (to measure light scattering loss by IAPP‐affected vesicles) to leakage assays, from fluorescence microscopy to electron microscopy, from circular dichroism to thioflavin T fluorescence assays to ensure that the effects I was gauging were exerted by non‐fibrillized IAPP, I unprecedentedly uncovered a mechanism by which IAPP damages membranes by remodeling their architecture, causing them to form more highly curved structures much like the curved structure formation induced by α-synuclein, suggesting that this could be a common mechanism among multiple protein toxins. In addition, using circular dichroism on IAPP bound to vesicles of different sizes, I discovered that this peptide is a membrane curvature sensor, not only an inducer. In vivo data on immunogold‐labeled IAPP in rat and human pancreatic cells showing IAPP to bind mitochondrial cristae, which comprise membranes that are by nature rich in high curvature, suggest physiological consequences of the peptide’s membrane curvature‐sensing ability. ❧ My studies on huntingtin, an essential protein for a number of cellular processes that is only known to cause Huntington’s disease when the polyglutamine (polyQ) region in its first exon contains an above‐threshold number of Q residues (Q-length), are covered in Chapters 4 and 5, starting with circular dichroism studies showing huntingtin exon 1 of abnormally high Q-length to adopt additional α-helical structure in solution to the α-helical structure that is normally present in this protein. This structural effect of increased Q-length can be exaggerated by decreasing the temperature of the sample and diminished by raising the temperature; temperature effects are reversible and are therefore typical of conformational transitions. Mimicking phosphorylation at serine 16 with an aspartic acid mutation structurally alters huntingtin exon 1 protein of higher Q-length such that its conformational behavior closely resembles that of its normal Q-length counterpart. Chapter 5 focuses on my electron paramagnetic resonance work, which clarifies the location of additional structure formation in huntingtin exon 1 of extended Q-length: it occurs in the 17-aa N-terminal region as well as in residues of the polyQ region that are closer to the N-terminal end. ❧ Altogether, these findings emphasize the importance of studying the structures, functions, and mechanisms employed by amyloid‐forming proteins under conditions wherein they have not yet formed amyloid