Structural and biophysiscal characterisation of toxic oligomers of the Alzheimer’s beta peptide (FWOAL493)

Alzheimer's disease (AD) is the most prevalent neurodegenerative disease in the growing population of elderly people. A pathologic hallmark of AD is the aggregation of amyloid ? into insoluble amyloid deposits. The neurotoxic amyloid-? (A?) is a small protein fragment that self-assembles to form amyloid fibres which accumulate in the brain in Alzheimer's disease (AD) (Iwatsubo et al., 1994). Besides amyloid fibrils, A? peptide forms small soluble particles that may represent a major cause of neurodegeneration that is associated with AD. Mounting evidence indicates that the A? oligomers, rather than the mature amyloid fibrils, are the most cytotoxic species (Lesne et al., 2006; Haass & Selkoe, 2006). Recently, a solid method was devised in our lab to generate toxic A? oligomers in significant amounts in vitro (Martins et al., 2008). We discovered that biological lipids reduce the stability of amyloid structures thereby inducing fast fibre resolubilisation. Curiously, the equilibrium is not shifted towards monomeric A? but rather towards soluble oligomers, which appear to be highly toxic for primary neurons (Martins et al., 2008). These toxic oligomers can be immunostained specifically with antibodies that recognize a toxic conformation of amyloid (Kayed et al., 2003). So far, complementary techniques like dynamic light scattering (DLS), size exclusion chromatography (SEC) with multi angle light scattering detection (MALS) and cryo electron microscopy revealed that the lipid-induced oligomer preparations are not monodisperse. Rather they contain objects with calculated molecular weights of 80-500 kDa (between 20 and 90 monomeric units). Fourier transform infrared (FTIR) spectra indicate that the oligomers possess a similar intermolecular ?-extended structure as mature fibrils. However, the difference FTIR spectrum revealed some degree of unfolding in the oligomers as compared to the mature amyloid fibrils. It has been documented that internal order seems to increase from oligomers to mature fibrils (Kheterpal et al., 2006). Despite a better understanding of the internal architecture of amyloid fibrils (Makin et al., 2005; Makin et al., 2006) and extensive biophysical and functional studies, the detailed structural characteristics of these toxic oligomers remain elusive. Detailed analysis of the structure of Abeta oligomers is difficult owing, in part, to their limited stability, low solubility in water, tendency to aggregate and their apparent transient nature in amyloidosis (Temussi et al., 2003). Another major hurdle so far was the lack of a consistent and reproducible method to generate such oligomers in vitro (Hepler et al., 2006; Martins et al., 2008). The production yield and sample properties of our lipid-induced A? oligomers provide a unique opportunity to perform a profound structural characterization which is the subject of this research proposal. This proposal aims at identifying the structural features that underlie the integrity of the Abeta oligomers. Admittedly, as mentioned above, several factors complicate the structure elucidation. Therefore, the development of structure perturbation methods is central to the proposal to alleviate these problems. In particular, comparative studies on the effect of destabilisation of Abeta oligomers and fibrils can infer structural similarities and differences between these conformational species. Organic solvents such as HFIP (hexafluoroisopropanol), DMSO (dimethylsulfoxide) and TFE (trifluoroethanol) are known as powerful denaturing agents to fibrillar multimers (Stine et al., 2003). In analogy to probing the conformational stability of monomeric proteins using chaotropes such as urea or guanidine, the resistance of the oligomeric species to the disruptive effect of HFIP or TFE can thus be used to quantify their stability. Next to chemical denaturants, also heat, pH or small ligands can function as perturbing agents to induce conformational changes. On the other hand, mature fibrils display a high thermostability and resistance to high urea or guanidine concentrations (O'Nuallain et al., 2005). Interestingly, various small-molecule inhibitors of amyloid formation have been discovered in attempts to develop anti-amyloid therapeutics (Kodali & Wetzel, 2007). The SWITCH laboratory has a hexapeptideinhibitor library at its disposal and can design peptide inhibitors against any self-associating region. These molecules can then be used to disrupt the oligomers and determine their stability. Limited proteolysis is an excellent method to identify disordered or flexible regions in proteins (Park & Marqusee, 2004). This approach relies on the fact that proteolysis of a protein substrate can occur only if the polypeptide chain can bind and adapt to the specific stereochemistry of the protease active site. This approach can also be performed under mildly perturbing conditions to infer information about structure, stability and dynamics of the protein (Park & Marqusee, 2005). In vitro studies already showed that A? antibodies can inhibit synthetic A? fibrillogenesis and can even disrupt pre-existing fibrils (McLaurin et al., 2002; Solomon et al., 1996). These results are consistent with the finding that antibody-mediated reversal of cognitive dysfunction in mice did not necessarily require a decrease in the total brain A? level (Dodart et al., 2002). This suggests that A?-antibodies might neutralize the soluble A? assembly forms. Interestingly, monoclonal antibodies completely neutralized the cytotoxic effect of A?- oligomers in vivo and protected rodents from oligomer-mediated long term potentiation inhibition (Klyubin et al., 2005). Beside their potential to destabilize the oligomers, monoclonal antibodies that bind different regions in the A? peptide can be exploited to map the regions of the monomer that are accessible to the antibodies. A blotting experiment comparing Abeta monomers to oligomers and multimers for their interaction with a diversity of antibodies will directly which peptide regions are solvent exposed. Time-resolved hydrogen-deuterium exchange (HDX) of oligomers/fibrils in combination with any of the mentioned perturbation methods and NMR analysis of the A? monomers, will yield a high resolution image of the assembled species. Assignment of the NMR HSQC-spectrum of the monomers (Hou et al., 2004) and analysis of the differences in HDX protection factors will allow detailed comparison of the structural similarities and differences of the oligomers and fibrils, since the HDX kinetics of peptide backbone amide groups are governed by local stability and solvent exposure and are sensitive to structural changes. Hence, a detailed view of intermolecular hydrophobic association and H-bonding pattern that stabilizes the oligomer structure will be obtained. Additionally, the direct effect of genetic mutations in the A? peptide can probe the contribution of the individual side chains on the oligomer stability. Recent literature hints that the relative ratio of A?40/A?42 is relevant to disease progress rather than the absolute amount of A? (Wang et al., 2006). Preliminary results obtained at the SWITCH laboratory demonstrate that different A?40/A?42 peptide ratios display a remarkably different aggregation behaviour. Clearly an extensive comparison of stability and structural aspects of different oligomer/fibril compositions can provide valuable information that is directly related to disease phenotypes. Selective isotope labelling (mixtures of natural abundance A?40 or A?42 and 15N-labeled A?42 or A?40) will be powerful to discern the two peptides. Interestingly, aggregation of A? will be visualized by the decrease of HSQC signal as the peptides aggregate into NMR-invisible high molecular weight fibrils. Hence the proposed structure perturbation method circumvents one of the main problems for NMR analysis. In addition, since deuterium incorporation into a peptide fragment causes progressive mass increase, mass spectrometry can also reveal structural intricacies albeit with lower resolution. Establishing a global and systematic perturbation method, in combination with time resolved HDX and NMR analysis is a major objective. All perturbation experiments will be performed and coordinated by the Switch laboratory, since the promoter of this project has extensive experience with this type of work (see bibliography Prof. Joost Schymkowitz). Evaluation of suitable perturbation method will be screened using a variety of biophysical techniques which are available at the Switch laboratory (DLS, size exclusion chromatography with MALS-detection, Field Flow Fractionation, fluorescence polarisation...). For detailed visualization of the resulting protein conformations, cryo-electron microscopy will be used in collaboration with Prof. Louise Serpell (University Sussex). The NMR experiments will be performed in collaboration with Prof. Annalisa Pastore (NIMR, London). In summary, the proposed strategy is to dissect the A? oligomers and mature fibrils into their structural subunits. The study of the selected monomeric peptides will be directly related to the overall structure and stability of the lipid-induced oligomers and the fibrils. Obviously the unambiguous structural identification of the oligomers and detailed understanding the conformational transitions in amyloidosis is a key step for designing more potent and selective therapeutics (Temussi et al., 2003). Therefore, the structural characterization of the toxic A? oligomers will be a major breakthrough in the understanding of the molecular pathogenesis of Alzheimer's disease. Detailed insight in the molecular architecture and dynamics will provide clues of how the oligomers interact with biological lipids or other proteins involved in AD and how they can directly damage cells.

Keywords:Applied Biology

Disciplines:Biological sciences