? The proposed research addresses a fundamental limitation of current proteomic workflows, namely the characterization of alternative splice and variant protein isoforms. Aberrant gene expression, including alternative splicing of all kinds, is observed in all types of cancer, and can be used as an early diagnostic biomarker. Similarly, somatic genomic mutations in coding sequence result in altered protein function or efficacy. Once discovered, antibodies can be developed to test for particular isoforms, but antibody development is slow and expensive. We propose to characterize these aberrant protein isoforms by proteomics. Characterizing alternative splice and variant protein isoforms by proteomics is difficult primarily due to fundamental limitations in our computational infrastructure. While mass spectrometry does not preferentially sample known protein isoforms, our peptide identification software is heavily biased towards known protein sequences already in protein sequence databases. We propose to build an informatics infrastructure to search a comprehensive set of species specific peptide sequences from mRNA and EST sequence evidence, predicted gene models, and SNP databases, in addition to the traditional protein sequence databases. Identified peptides will be mapped back to their sequence evidence and interpreted with respect to an appropriate gene model. In order to observe sufficient coverage of isoform sequences, we will design and implement sample preparation and mass spectrometry workflows that significantly increase the sequence coverage of observed peptides. We will focus on protein separation as a key technique, and evaluate 2D gel electrophoresis and capillary isoelectric focusing. In addition, we will explore the use of multiple ionization and proteolytic digestion techniques in combination with peptide separation by liquid chromatography and tandem mass spectrometry. We will use the mouse skeletal muscle C2C12 line as a model system to optimize the protocol for isoform characterization. We will use the proposed informatics infrastructure and proteomics workflows to study alternative splicing in breast cancer by analyzing the estrogen positive non-metastatic MCF-7 line and the estrogen negative metastatic MDA-MB 231 line, for which alternatively spliced mRNA has previously been observed. We will also analyze human brain tumor tissue samples for alternative splicing protein isoforms, via a subcontract with Calibrant Biosystems. ? ? ?
Showing the most recent 10 out of 13 publications
