Predicting the structure of biological protein assemblies

Many biological processes are controlled by large multi-component protein assemblies, the size and complexity of which typically precludes the determination of their high-resolution structures. Yet, for these systems ample non-structural data is typically available providing a wealth of lower resolution information. Our goal is to use high-resolution structural modeling techniques guided by constraints taken from lower resolution experimental data to generate structural models of important biological protein assemblies for which high resolution structural determination is unfeasible. In particular, we are interested in a subclass of protein assemblies, homomeric assemblies, which are produced by the repetition of single type of subunit. Homomeric assemblies typically adopt internal symmetry, which provides a crucial constraint in the molecular modeling. More specifically, we are working with two problems: I) Modeling the assembly structure of the type III secretion system and II) predicting the high-resolution structures of homo-oligomeric proteins using limited experimental data from Nuclear Magnetic Resonance (NMR) in order to speed up the process of structure determination.
 
I. Modeling the type III secretion system assembly

Our goal is to determine the molecular architecture of the type III secretion system (T3SS), where non-structural as well as low and high-resolution structural information are available. T3SS is an essential component for the virulence of pathogenic Gram-negative bacteria, such as Salmonella and Chlamydia. This macromolecular assembly comprises more than 20 conserved proteins that form a series of ring-like structures and a needle-shaped protrusion, which allows the transport of proteins into a host cell. In addition to low-resolution models of the intact T3SS core from cryo-electron microscopy, several high-resolution structures of monomeric T3SS components have been determined. Other sources of experimental information include X-ray diffraction, biotinylation, sequence mutation and deletion, solution binding and chemical cross-linking data. These low and high-resolution data sets originates from T3SSes from different species and can also be collected from other structurally homologous systems, such as the flagella. Our goal is to generate a high-resolution model of the structural foundation of the T3SS by modeling its homo-oligomeric rings and filaments. Such a model would provide an unprecedented view of the molecular architecture of T3SS by visualizing the placement of its components within the assembly as well as interactions between these components.

References:

1. Andre, I., Bradley, P., Wang, C., and Baker, D. (2007) Prediction of the structure of symmetrical protein assemblies, Proc Natl Acad Sci U S A 104, 17656-17661.

2. Spreter, T., Yip, C., Sanowar, S., Andre , I., Kimbrough, T., Vuckovic, M., Deng, W., Finlay, B., Baker, D., Miller , S., and Strynadka, N. (2009) A conserved structural motif mediates formation of the inner and outer membrane rings in the Type III Secretion System. Nature Structural and Molecular Biology, 16(5):468-76.

3. Sanowar S, Singh P, Pfuetzner RA, Andre I, Zheng H, Spreter T, Strynadka NC, Baker D, Goodlett DR and Miller S. Interactions of the Transmembrane Polymeric Rings of the Salmonella enterica Serovar Typhimurium Type III Secretion System. MBio. 2010 Aug 3;1(3). pii: e00158-10.

II. Inference of oligomeric protein structure from limited Nuclear Magnetic Resonance data

In NMR structure determination experimental collection and assignment of distance constraints necessary for structure calculation is a time-consuming and laborious task. The process is further complicated for homo-oligomeric proteins; spectral symmetry renders intra- and intermolecular distance constraints indistinguishable. Specialized experiments have been developed to distinguish the two types of data, but these techniques require the preparation of mixed isotopically labeled samples and typically yield only a few intermolecular constraints. To address this problem, we recently developed a method to predict the structure of homo-oligomeric proteins using only limited sources of experimental data that can be quickly obtained. Our goal is to extend the size range of proteins that can be accurately predicted using this method by constraining the simulations with additional experimental data in order to develop a general method to rapidly determine the structure of homo-oligomeric proteins. 

References:

1.  Andre, I., Das, R.,Shen Y., Wu, Y.,Lemak,A., Bansal S., Arrowsmith, C.H., Szyperski, T.  and Baker, D. (2009). Simultaneous prediction of protein folding and docking at high resolution. Accepted for publication in Proc Natl Acad Sci U S A.

 2. Sgourakis NG, Lange OF, DiMaio F, André I, Fitzkee NC, Rossi P, Montelione GT, Bax A, Baker D. Determination of the structures of symmetric protein oligomers from NMR chemical shifts and residual dipolar couplings. J Am Chem Soc. 2011 Apr 27;133(16):6288-98.

• Print version