Warren J. Goux

Macrocyclic β-Sheet Peptides That Inhibit the Aggregation of a Tau-Protein-Derived Hexapeptide

This paper describes studies of a series of macrocyclic β-sheet peptides 1 that inhibit the aggregation of a tau-protein-derived peptide. The macrocyclic β-sheet peptides comprise a pentapeptide “upper” strand, two δ-linked ornithine turn units, and a “lower” strand comprising two additional residues and the β-sheet peptidomimetic template “Hao”. The tau-derived peptide Ac-VQIVYK-NH2 (AcPHF6) aggregates in solution through β-sheet interactions to form straight and twisted filaments similar to those formed by tau protein in Alzheimer’s neurofibrillary tangles. Macrocycles 1 containing the pentapeptide VQIVY in the “upper” strand delay and suppress the onset of aggregation of the AcPHF6 peptide. Inhibition is particularly pronounced in macrocycles 1a, 1d, and 1f, in which the two residues in the “lower” strand provide a pattern of hydrophobicity and hydrophilicity that matches that of the pentapeptide “upper” strand. Inhibition varies strongly with the concentration of these macrocycles, suggesting that it is cooperative. Macrocycle 1b containing the pentapeptide QIVYK shows little inhibition, suggesting the possibility of a preferred direction of growth of AcPHF6 β-sheets. On the basis of these studies, a model is proposed in which the AcPHF6 amyloid grows as a layered pair of β-sheets and in which growth is blocked by a pair of macrocycles that cap the growing paired hydrogen-bonding edges. This model provides a provocative and appealing target for future inhibitor design.

Chiroptical Probes of Protein Structure

Although x-ray diffraction studies of crystals is the definitive technique for the investigation of molecular structure in the solid state, it is certainly possible, even probable, that there are significant differences in the microscopic structure of protein molecules in the solid state and in solution. Even though it may be unlikely that significant deviations occur in the path of the polypeptide chain it is likely that the conformations of certain amino acid side chains, especially those on the surface, may vary significantly in the two states. Since it is the side chains that constitute the functional portions of enzyme molecules insofar as biological activity is concerned, this is a problem of considerable significance. The recent discovery that lysozyme can be crystallized in at least two different forms (l–3), and Vallee’s (4) results which indicate a conformational change of a tyrosine residue when crystalline carboxypeptidase is dissolved may be taken as evidence supportive of this point of view.