Dynamics of Protein Assemblies by Analytical Ultracentrifugation

National Institute Of Biomedical Imaging And Bioengineering, Bethesda

Investigators

Linked publications, trials & patents

Abstract

In this project we seek to develop experimental methods for the study of size-distributions of macromolecules and particles. This includes methods for the study of polymorphic dynamic multi-protein complexes that elude classical structural techniques, as well as methods for high-resolution particle size distributions in nanoparticle and biotechnology applications. In previous years we have designed computational algorithms for the analysis of sedimentation velocity analytical ultracentrifugation (SV-AUC) data with high hydrodynamic resolution, based on the deconvolution of diffusion and nonideality effects. These were implemented and distributed in the software SEDFIT, which is widely used in research and quality control applications in academic and biopharmaceutical laboratories. For example, SEDFIT analysis has become the de facto standard for the detection of oligomeric impurities of protein pharmaceuticals, as well as the characterization of viral vector and lipid particle preparations for therapeutic applications. It is also widely used to study weak protein interactions that govern dynamic multi-protein complexes, as well as intrinsically disordered proteins, with many applications in the study of cancer, immunology, and chronic diseases. A long-standing effort of our group is to ensure the reproducibility of measurements by establishing calibration accuracy of analytical ultracentrifuges. After designing calibration methods, several years ago we have embarked on a collaborative project with colleagues from NIST to design and fabricate a calibration window to be made available as a commercial standard reference material (SRM) by NIST. While the production run of these windows is currently ongoing, in the reporting period we created software that will facilitate processing of the data files acquired with this calibration window. The new software will conveniently generate the desired calibration constants that transform measured apparent sedimentation coefficients onto a reproducible, absolute scale. Furthermore, we have explored additional uses of the calibration window that will test the performance of the optical system of analytical ultracentrifuges. The release of these tools is planned to coincide with the availability of the calibration window as an SRM at NIST. Systems of interacting macromolecules with complex lifetimes on the time-scale of sedimentation exhibit reaction boundaries characterized by dynamically coupled migration of the interacting species. The sedimentation patterns of such systems have historically been considered counter-intuitive and long remained unexplained. Several years ago, we developed effective particle theory (EPT) that describes the concentration-dependent sedimentation boundary patterns, and we connected EPT to sedimentation coefficient distribution analysis c(s). This framework provides a complete description of the concentration-dependent velocity, amplitude, and composition of all experimentally measured sedimentation boundaries in such systems. In the last year we have developed an approach to expand EPT by including hydrodynamic nonideality in the limit of dilute solutions. Once implemented in the analysis software, we expect this advanced EPT to improve the quantitative analysis of sedimentation patterns of weakly interacting proteins that require moderate to high concentrations. The practical application of EPT has also been limited in some cases by the fact that the predicted boundaries may not always be experimentally well-resolved. Therefore, we have developed an approach which maps EPT predictions to experimentally well-defined features (e.g., total signals or average s-values in a certain resolved sedimentation coefficient range) for isotherm analysis, termed ‘partial boundary analysis’. This facilitates the application of EPT by removing the need to resolve reaction boundaries, and therefore greatly simplifies practical analysis of rapidly interacting systems by SV. We have recently implemented this approach in our multi-method biophysical data analysis software SEDPHAT, and during the reporting period verified its expected performance and utility in applications to model systems. Finally, we have continued our effort to disseminate expert knowledge in SV-AUC. In particular, the concepts of coupled sedimentation and reversible reactions can be challenging for novices to understand and utilize. Previously we published a monograph on this topic and developed lectures for our annual biophysical methods workshops. To make this material easier to understand and to adapt it to modern learning styles, in the review period we have created a webinar on sedimentation velocity of interacting systems that can be widely disseminated on common video platforms and used for self-learning.