We previously showed that alpha-synuclein, the protein involved in Parkinson's disease, reconfigures in the dangerous middle regime in physiological conditions. Therefore a drug that binds to the protein and increases reconfiguration may prevent aggregation and disease. Nortriptyline, an antidepressant (marketed as Pamelor), speeds diffusion at equilmolar concentrations (plot c, above) and slows formation of the first oligomer (plot d, above). NOR was also shown to prevent aggregation in a series of cell and animal models.
The mammalian prion protein shows different aggregation propensity depending on the species. We measured intramolecular diffusion in the C-terminal domain of two species, hamster (which has high aggregation propensity) and rabbit (which has low aggregation propensity). At pH 7 and different concentrations of guanidine, the unfolded states of the two proteins were very similar, especially as denaturant decreased. Diffusion extrapolated to a very low value at low denaturant concentrations. This means aggregation is not likely when this protein is unfolded at pH 7. However, aggregation is probably originating in late-stage endosomes, which have a low pH (~4). At the lower pH, the protein is unfolded and hamster diffuses in the intermediate range more prone to aggregation but rabbit diffuses much faster and avoids aggregation. When astimezole, a possible drug for prion disesase, is added to hamster, its diffusion increases to the fast range.
The peptide involved in Alzheimer's disease, Abeta, is found in several different lengths. While the most common form in the body is Abeta40, the most common form in amyloid plaques is Abeta42. We found intramolecular diffusion for Abeta40 is 5 times faster than for Abeta42, putting reconfiguration of the longer peptide in the dangerous middle regime. Furthermore, raising the pH to 10 or adding the aggregation inhibitor, curcumin, speeds up reconfiguration of Abeta42 about as much as deleting the last two amino acids.
This peptide, found in semen, easily aggregates and the fibrils are known to enhance HIV invectivity. We measured intramoelcular diffusion in the full length and truncated peptide missing the first 8 residues which is much more aggregation-prone. The truncated peptide diffuses much more slowly than the full length. Modeling shows that the deletion of several charged residues is the primary reason. Without these charges, the chain is able to compact and transient interactions make it less diffusive.
Protein G is considered to be one of the basic test systems of protein folding. Most studies find that this small protein acts as either a two-state (Unfolded/Folded) or three-state (Unfolded/Intermediate/Folded) folder, but experiments typically use one folding probe on timescales longer than 1 ms. We studied protein G folding starting from 10 microseconds using 4 probes (total tryptophan fluorescence emission, trp fluorescence spectral shift, FPOP and CD) and found many different kinetic phases. These experiments were compared to a Markov State Model computed by Folding@Home and found good agreement.
A new probe for observing protein folding is photochemical oxidation, the labeling of certain protein side chains with hydroxy radicals. These radicals can be created from hydrogen peroxide using a UV laser and their attachment monitored with mass spectrometry after the folding experiment is over. We have coupled this method to a microfluidic mixer (mixing time ~8 microseconds) in which radicals are created and quenched within ~1 microsecond. This probe provides information about the solvent accessibility of proteins during folding. We found the surprising result that solvent accesibility in lysozyme immediately after mixing from high guandine to physiological buffer is lower than later in the folding process.
CLR01 is a molecular tweezer that selectively binds to Lysine side chains (a-synuclein has 15 Lys). As more tweezers bind to the protein, the reconfiguration rate increases, making the initial aggregation step less likely because two monomers can't stick together before one of the monomers reconfigures to a non-aggregation prone conformation. CLR01 has been shown to have low toxicity in animals in goes into the brain. Our collaborator, Gal Bitan has plans for clinical trials. (MSU press release, Interview with Lisa on WKAR)