Research

Our research interests are focused on developing and applying solid-state NMR methods to determine structural constraints in biologically relevant systems that are difficult to fully characterize by other means in order to contribute to our overall understanding of biological processes.

In the near future, we will be working on determining the internal structure of the β₂ microglobulin protein in its fibrillar form.

Figure 1: β₂ microglobulin. A) Native fold. B) Electron micrograph of β₂ microglobulin amyloid fibrils. C) Amino acid sequence. (Images from Jones et al., J. Mol. Biol. (2003) 325 pp249-257).

Filamentous fibrillar amyloid aggregates are interesting to study because of the many diseases with which they are associated, but also from the standpoint of understanding protein folding. A large number of disparate proteins have been shown to form fibrils with morphologies similar to that found in Fig. 1B. Amyloid fibrils exhibit a characteristic ‘cross-β’ scattering pattern from X-ray fiber diffraction measurements indicating the presence of β-sheets oriented perpendicular to the long axis of the fibril. While the internal structures of at least a few of these different fibrils have been shown to be different in detail, it is unknown why the overall morphologies are so similar. Considerable effort is being expended in elucidating the intermolecular interactions that stabilize amyloid formation, the mechanisms of amyloid formation from peptide monomers or oligomers, and the internal molecular structure of peptides in amyloid fibrils.

We will be focusing on the β₂ microglobulin protein (Fig. 1). β₂ microglobulin is a relatively small protein with 99 residues that is the major component of dialysis related amyloid. Dialysis related amyloidosis is a serious complication for patients receiving long term hemodialysis, resulting in carpal tunnel syndrome and/or arthritis-like pain in joints and tendons. Eventually, a majority of long-term hemodialysis patients experience the debilitating effects of the disease due to a buildup of fibrillar amyloid material in bone and joint tissues. Because β₂ microglobulin is a small protein and readily expressed in a variety of expression systems, it has been used extensively in amyloid formation studies, but little is known about its internal molecular structure in amyloid fibrils beyond the fact that it contains the ‘cross-β’ motif common to all amyloid fibrils. Since amyloid fibrils are ill suited to study by solution-state NMR because they are insoluble aggregates, and neither is the detailed internal structure of fibrils obtainable by crystallography because fibrils do not exhibit the long range crystalline order required, we propose to apply a set of solid-state NMR structure elucidation tools to answer questions about the structure of the β₂ microglobulin protein in its fibrillar form.

Figure 2: Experimental measurements to determine the global fold of the peptide in the amyloid fibril. A) Multiple quantum experiments to distinguish between different organizational structures. B) Backbone torsion angle measurements at particular locations. C) Proximity measurements between different parts of the peptide.

Dipolar couplings, chemical shift anisotropies (CSAs), and their cross correlations can be measured at particular locations in the peptide to give structural constraints. For instance, distances up to ~5-6 Å between two isotopic labels in a peptide (Fig. 2C) can be determined either by homonuclear or heteronuclear dipolar recoupling experiments. Such distance measurements can be used, for instance, to verify a hairpin, in which a peptide folds back on itself (as is postulated for the P72-M99 fragment). A variety of dipolar recoupling and cross-correlation experiments measure one or both of the peptide backbone torsion angles (Fig. 2B) at a single amino acid position for a direct measure of the secondary structure at that position. Additionally, solid-state NMR may be used to distinguish between fibril models with different organization of peptides internal to the fibril. In particular, so-called ‘spin counting’ multiple quantum experiments can estimate the number of isotopic labels within a dipolar-coupling network. Fig. 2A shows a series of different internal organizations of peptides in a fibril that differ in the number of isotopic labels that are in close proximity to each other. Multiple quantum (MQ) spectra of a parallel β-sheet organization would show high orders of MQ coherence, while the anti-parallel case would show no higher orders since none of the isotopic labels are in proximity to each other. Dimeric and trimeric groupings would correspondingly show either two or three quantum coherences at most.