Global Analyses of Protein Complex Localization, Oligomerization, Composition, and Dynamics Using Quantitative Mass Spectrometry
Proteins perform essential processes in plant cells such as photosynthesis, protein translation, maintenance of metabolic flux, and signal transduction. Many of these functions could never be achieved using individual proteins. Consequently, many proteins oligomerize to form complexes that act as tunable molecular machines that perform work and transmit information. Cells contain thousands of protein complexes, yet the composition of the vast majority remain unknown. Previous high throughput approaches to identify protein interactions have relied on binary interactions or tagging individual proteins for purification. A new set of label free correlation profiling methods were developed that extracted native protein from Arabidopsis leaves for separation by liquid chromatography and mass spectrometry quantified the elution profiles for thousands of proteins in a single experiment. One new method expanded protein correlation profiling to membrane-associated proteins, which are normally discarded because of their insolubility. Hundreds of novel membrane-associated complexes were predicted based on their high apparent mass compared to the monomeric form. Dozens were dual localized proteins that partitioned between the cytosol and cell membranes: a small subset had oligomerization states that clearly differed as a function of localization. Protein correlation was then expanded to include an orthogonal separation of native complexes by charge. This added resolving power allowed us to predict protein complex composition. By creating a reproducible data mining and analysis pipeline, over 200 cytosolic and 120 chloroplast complexes were predicted. Validation included the accurate prediction of known protein complex compositions and novel subunits to known complexes. Using reverse genetics we discovered a new protein complex subunit AIM1PL, which appears to broadly affect protein complex assemblies that are involved in translation. In the last chapter the SEC-based protein correlation profiling method was used to broadly analyze how the proteome responds to a stress condition. Hundreds of interesting examples were discovered in which soluble and membrane-associated proteins are predicted to change in abundance, localization, and/or oligomerization state in response to metabolic stress. These analyses uncovered interesting biology that likely underlies the post-translational control of gene expression with the mechanisms by which diverse cellular activities are integrated during plant growth and development. Collectively, a suite of new methods are created that enable high throughput and broad analyses of how protein complexes in the cell and how the proteome responds to metabolic stress. These results are being used to generate testable hypotheses about how cellular systems respond to metabolic stress. These technologies have broad application to agriculture because they can be applied to any plant species with a well-annotated genome^
Daniel B. Szymanski, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our