Mapping protein-protein interactions is an invaluable tool for understanding protein function. Here, we report the first large-scale study of protein-protein interactions in human cells using a mass spectrometry-based approach. The study maps protein interactions for 338 bait proteins that were selected based on known or suspected disease and functional associations. Large-scale immunoprecipitation of Flag-tagged versions of these proteins followed by LC-ESI-MS/MS analysis resulted in the identification of 24 540 potential protein interactions. False positives and redundant hits were filtered out using empirical criteria and a calculated interaction confidence score, producing a data set of 6463 interactions between 2235 distinct proteins. This data set was further cross-validated using previously published and predicted human protein interactions. In-depth mining of the data set shows that it represents a valuable source of novel protein-protein interactions with relevance to human diseases. In addition, via our preliminary analysis, we report many novel protein interactions and pathway associations.

An evaluation of the proteolytic labeling and quantification of proteins for diagnostic purposes using trypsin and 18O-enriched H2O is presented. We demonstrate that comparative or relative quantitation can be performed effectively with this approach. We have developed a protocol that allows the conservation of the labeled peptides in natural abundance water without fear of back-exchange providing that pH is sufficiently low to quench the catalytic activity of trypsin, but not so low as to promote chemical back-exchange. Because the labeling efficiency depends on the nature of the peptide, a simple linear relationship between the relative 16O/18O digest buffer mixture content (x) and labeling efficiency (y) does not exist; rather it follows a probability based y = x(2) relationship. As such, the extent of peptide labeling using 16O/18O digest buffer mixture ratios may deviate significantly from that expected based on a linear relationship. The evaluation of the relative Ziptip efficiency indicated a loss in sample recovery as the peptide concentration was reduced using normal conditions, suggesting that there is a limit below which there are diminishing returns. In addition, the adsorptive losses due to Speedvac dry down and recovery indicated modest (20%) losses that may vary widely (0-50%) from peptide to peptide. The in-solution digestion efficiency of standard protein mixtures as a function of concentration revealed a linear decrease with decreasing concentration. This is consistent with enzyme kinetic effects and emphasizes a potential quantitation error that could arise when evaluating differential expression based on peptide detection. The results from our studies demonstrate the power of 18O labeling as an optimization tool for proteomics process development.

The ability to quantitatively compare protein levels across different regions of the brain to identify disease mechanisms remains a fundamental research challenge. It requires both a robust method to efficiently isolate proteins from small amounts of tissue and a differential technique that provides a sensitive and comprehensive analysis of these proteins. Here, we describe a proteomic approach for the quantitative mapping of membrane proteins between mouse fore- and hindbrain regions. The approach focuses primarily on a recently developed method for the fractionation of membranes and on-membrane protein digestion, but incorporates off-line SCX-fractionation of the peptide mixture and nano-LC-MS/MS analysis using an LTQ-FT-ICR instrument as part of the analytical method. Comparison of mass spectral peak intensities between samples, mapping of peaks to peptides and protein sequences, and statistical analysis were performed using in-house differential analysis software (DAS). In total, 1213 proteins were identified and 967 were quantified; 81% of the identified proteins were known membrane proteins and 38% of the protein sequences were predicted to contain transmembrane helices. Although this paper focuses primarily on characterizing the efficiency of this purification method from a typical sample set, for many of the quantified proteins such as glutamate receptors, GABA receptors, calcium channel subunits, and ATPases, the observed ratios of protein abundance were in good agreement with the known mRNA expression levels and/or intensities of immunostaining in rostral and caudal regions of murine brain. This suggests that the approach would be well-suited for incorporation in more rigorous, larger scale quantitative analysis designed to achieve biological significance.

No abstract