Measurement of peptide aggregation with pulsed-field gradient nuclear magnetic resonance spectroscopy

Interactions between hydrophobic patches in proteins are often a driving force for denaturation and aggregation. The aggregation of the beta-amyloid peptide fragment, VHHQKLVFFAEDVGSNK (beta(12-28)), has been investigated in aqueous solution at low pH. This peptide contains a central hydrophobic patch spanning residues 17-21. Diffusion coefficients measured with pulsed-field gradient NMR as a function of peptide solution concentration were used to assess the extent of aggregation. Following the hypothesis that hydrophobic interactions are an important driving force in the aggregation of this peptide at low pH, a non-aggregating analog of the beta(12-28) peptide, [Gly(19,20)]beta(12-28) was synthesized. In the [Gly(19,20)]beta(12-28) peptide, the replacement of the two phenylalanine residues disrupts the hydrophobic interactions which drive the aggregation of beta(12-28). The diffusion coefficient of the [Gly(19,20)]beta(12-28) peptide is invariant over the concentration range studied and provides a good estimate of the monomeric diffusion coefficient of beta(12-28). A second peptide analog was synthesized in which the phenylalanine at position 20 was replaced with a cysteine residue. The disulfide-linked dimer, ([Cys(20)]beta(12-28))(2), was formed upon air oxidation of this peptide. The diffusion coefficient of the ([Cys(20)]beta(12-28))(2) peptide was measured and used to estimate the diffusion coefficient of the beta(12-28) dimer. Using the monomeric and dimeric diffusion coefficients measured for the glycine and cysteine analogs, the concentration dependence of the beta(12-28) diffusion coefficient was found to be consistent with a monomer-dimer aggregation model. (C) 1998 Elsevier Science B.V.