The pathogenesis of Alzheimer disease (AD) is strongly linked to the amyloid-β protein (Aβ) assembly into amyloid fibrils in the extracellular space, resulting in amyloid plaque deposition. Aβ belongs to the class of amyloid-forming proteins characterized by spontaneous formation of the cross-β fibril structure, in which individual protein molecules adopt an extended β-strand structure and are held together by hydrogen bonding that runs perpendicular to individual strands and parallel to the fibrillar axis. Resolving the structure of Aβ fibrils has been for years one of the central AD research topics.

Despite numerous studies, the exact three-dimensional structure of Aβ fibrils remained elusive. In the past decade, solid-state NMR studies conducted by Tycko and his collaborators as well as other researchers in the field have shed new insights into the structure of Aβ40 fibrils. Based on the NMR constraints, Petkova et al. proposed an Aβ40 fibril model with a unit comprising two protein molecules. In this model, each individual Aβ40 molecule is characterized...  Read more

The pathogenesis of Alzheimer disease (AD) is strongly linked to the amyloid-β protein (Aβ) assembly into amyloid fibrils in the extracellular space, resulting in amyloid plaque deposition. Aβ belongs to the class of amyloid-forming proteins characterized by spontaneous formation of the cross-β fibril structure, in which individual protein molecules adopt an extended β-strand structure and are held together by hydrogen bonding that runs perpendicular to individual strands and parallel to the fibrillar axis. Resolving the structure of Aβ fibrils has been for years one of the central AD research topics.

Despite numerous studies, the exact three-dimensional structure of Aβ fibrils remained elusive. In the past decade, solid-state NMR studies conducted by Tycko and his collaborators as well as other researchers in the field have shed new insights into the structure of Aβ40 fibrils. Based on the NMR constraints, Petkova et al. proposed an Aβ40 fibril model with a unit comprising two protein molecules. In this model, each individual Aβ40 molecule is characterized by two β-stands involving residues 12-24 and 30-40, while residues 25-29 are in a bend conformation enabling salt-bridge formation between D23 and K28, which stabilizes the fibril structure (1).

A few years later, Luehrs et al. derived the three-dimensional structure of Aβ1-42 fibrils using hydrogen-bonding constraints from quenched hydrogen/deuterium-exchange NMR and combining these with the side-chain packing constraints from pairwise mutagenesis as well as previous solid-state NMR studies (2). Based on these findings, Luehrs et al. proposed a side-chain packing that differed from the packing predicted by Petkova et al. from Aβ1-40 fibrils, suggesting that Aβ1-40 and Aβ1-42 are characterized by structurally distinct fibrils.

In the same year, surprising discoveries published by Petkova et al. demonstrated that a detailed molecular structure of Aβ1-40 fibrils depends on subtle variations in fibril growth conditions and that morphology and molecular structure of seed-induced fibrils are self-propagating (3). This result suggests that the free energy landscape for fibril formation is complex and consists of many minima. Such a plasticity of Aβ fibril growth is quite fascinating and poses questions of 1) whether this molecular-level polymorphism of Aβ fibrils can be traced back to the polymorphism of oligomeric and prefibrillar assemblies, and 2) what is the physiological relevance of this polymorphism.

In the present paper, Paravatsu et al. provide further evidence of the polymorphism of Aβ fibrils and its role in AD. In a beautiful study that represents a leap from a purely in-vitro to a somewhat in-vivo situation, Paravatsu et al. extracted Aβ fibrils from the brain tissue of deceased AD patients and used these as seeds for synthetic Aβ1-40 fibril growth. Employing the fact that the fibrillar morphology and molecular structure stem from the structure of initial seeds, the resulting Aβ1-40 fibrils were assumed to have the same structure as the seeds; therefore, the fibril structure is equivalent to the one found in the brain of AD patients. The resulting brain-derived Aβ1-40 fibrils were then analyzed by solid-state NMR.

Despite the fact that the initial seeds extracted from the brain tissue were by themselves not homogeneous as they contained different chain lengths and modifications, the final fibril structure of such brain-derived Aβ1-40 was found to be more homogeneous than the fibril structure of a purely synthetic Aβ1-40. Moreover, the brain tissue-derived Aβ1-40 fibrils from two different AD patients were characterized by the same molecular structure. In light of the molecular-level polymorphism of Aβ fibrils, this is quite a puzzling and exciting result.

What is the distinct Aβ fibril structure that characterizes the AD brain? Paravatsu et al. found that this structure differs from the molecular model derived a few years ago by Petkova et al. (1). While the 2D solid-state [13C] NMR spectrum of purely synthetic Aβ1-40 indicated a proximity between F19 and L34 residues, brain-derived fibrils were characterized by an imminent proximity between F19 and I31. This result is important because it means that the odd-numbered residues of both strands, 12-24 and 30-40, contribute to intramolecular packing in these brain-derived fibrils, while in the original model by Petkova et al. the odd-numbered residues on the 12-24 strand are closely packed with the even-numbered residues of the 30-40 strand, suggesting even a different type of bend between the two strands.

The same structural characteristics were obtained using as seeds brain-derived fibrils from two AD cases, suggesting that AD may be associated with a particular fibril structure that is distinct from the structure of purely synthetic Aβ1-40. Because different fibril structures may exhibit different levels of toxicity in neuronal cell cultures, it is possible that only one particular fibril structure may be correlated with the degree of neurodegeneration in AD patients. This study opens up a wide variety of new research directions. How does the toxicity of purely synthetic Aβ1-40 compare to these new brain-derived Aβ1-40 fibrils? Would synthetic Aβ1-42 added to brain-derived seeds form a compatible fibril structure?

Should future studies corroborate that there is a particular molecular structure of Aβ fibrils that is most relevant to AD, it will open up new directions towards AD therapy that will aim at not necessarily prevention of fibril formation, but rather at altering the pathways of Aβ fibril formation towards non-toxic assemblies.

References:
1. Petkova AT, Ishii Y, Balbach JJ, Antzutkin ON, Leapman RD, Delaglio F, Tycko R. A structural model for Alzheimer's beta -amyloid fibrils based on experimental constraints from solid state NMR. Proc Natl Acad Sci U S A. 2002 Dec 24;99(26):16742-7. Abstract

2. Lührs T, Ritter C, Adrian M, Riek-Loher D, Bohrmann B, Döbeli H, Schubert D, Riek R. 3D structure of Alzheimer's amyloid-beta(1-42) fibrils. Proc Natl Acad Sci U S A. 2005 Nov 29;102(48):17342-7. Abstract

3. Petkova AT, Leapman RD, Guo Z, Yau WM, Mattson MP, Tycko R. Self-propagating, molecular-level polymorphism in Alzheimer's beta-amyloid fibrils. Science. 2005 Jan 14;307(5707):262-5. Abstract

View all comments by Brigita Urbanc