While important information has been gained through the profiling of transcripts, it is important to keep in mind that the end-point for gene expression is the protein, as proteins represent the actual scaffolds, molecular engines and communication mechanisms utilized by cells. In addition, the development of most biomedical interventions center on protein endpoints. Large-scale efforts are underway to analyze the proteome: the total protein complement of the genome.32 However, the proteome represents a complex dynamic entity due to the many forms of a given protein that result from alternative transcript splicing and the numerous post-translational modifications that often define functional protein states.33'34 A core technology developed for the global analysis of proteins involves the electrophoretic separation of proteins along two dimensions: size and charge, using polyacrylamide gels (2D-PAGE).35 Comparisons of gel profiles from two different cell states (e.g. normal versus cancer) can identify differential protein expression or protein modifications. The identity of individual protein spots can be determined by immunodetection or by calculating the theoretical locations of known proteins based upon charge and mass. More recently, microsequencing and mass spectrome-try have gained widespread use for characterizing protein spots of inter-est.36,37 The 2D-PAGE technique is theoretically capable of resolving more than 10,000 proteins and peptides from complex mixtures. However, practical applications of 2D-PAGE have rarely identified more than a small percentage of proteins comprising the cellular proteome.38 Despite this drawback, 2D-PAGE has been used to identify androgen-regulated genes in prostate epithelium39 and prostate cancer-associated protein alterations such as NEDD8 and calponin.40
Mass spectrometry (MS) has evolved to become a key technology in proteomics research. The basic mass spectrometer is comprised of an ion source, a mass analyzer, and a detector to record the spectra of ion intensity versus the mass to charge ratio of proteins or peptides under analysis.36 Currently, surface enhanced laser assisted desorption (SELDI), matrix assisted laser desorption (MALDI) and electrospray ionization (ESI) represent the preferred methods for ionizing peptides and proteins. Comparing the mass spectra of complex protein mixtures such as human serum offers the potential for identifying proteins or protein fragments that are present in different abundances and that associate with disease states.33'41 Several reports have used SELDI mass spectrometry as a bio-marker discovery method to distinguish serum profiles or fingerprints that reflect the presence of prostate carcinoma in comparison to serum protein profiles derived from normal controls.42,43 A major drawback to these profiling approaches involves a lack of reproducible quantitative accuracy and a difficulty in positively identifying the peptides and parent proteins that account for the fingerprint differences.
Several strategies have been devised to provide MS with the ability to accurately quantitate differences in protein levels between two cellular states.44-47 One technique employs Isotope Coded Affinity Tags (ICAT) comprised of three components: a biotin affinity tag, a linker with either eight hydrogen or eight deuterium atoms (generating a light and a heavy form of the molecule), and a SH-reactive group capable of covalently linking to cysteine residues (Fig. 2).48 The proteins of one cell state (e.g. normal epithelium) are labeled with the light reagent and those of a second cell state (e.g. neoplastic epithelium) with the heavy reagent. Equal quantities of labeled cells are mixed and the proteins are separated, digested with a proteolytic enzyme and the resulting mixture of peptides are passed over an avidin column to isolate the cysteine labeled peptides (about 90% of proteins have cysteine residues). These can be fractionated and analyzed by tandem mass spectrometry (MS/MS). The first MS analysis gives the areas under the curves of the paired isotopic peptides (hence, their relative abundances); the second MS analysis provides a peptide fingerprint that can be used to identify the parent protein. Thus, the ICAT method dramatically increases throughput by reducing sample redundancy (only cysteine-containing peptides are assessed) and
Fig . 2. Relative Quantitation of Cellular Proteins using Isotope Coded Affinity Tags (ICAT) and Mass Spectrometry. (A) To facilitate the quantitative analysis of proteins in complex mixtures, peptides are labeled with an isotope-coded affinity tag (ICAT) reagent that consists of three parts: an affinity tag (biotin) that is used to isolate ICAT-labeled pep-tides, a linker that can incorporate stable isotopes, and a reactive group with specificity toward thiol groups (cysteines [Cys]). Two forms of the reagent are made: heavy (contains eight deuteriums [d8]) and light (no deuteriums [d0]). (B) The strategy for ICAT-based differential protein quantitation involves separately extracting proteins from two different cell states (e.g. normal and cancer) followed by labeling each with a different (d0 or d8) ICAT reagent. The ICAT reagent covalently bonds to each cysteinyl residue in every protein. The protein mixtures are combined and proteolyzed to peptides and only ICAT-labeled peptides are isolated using the biotin tag. These peptides are separated by microcapillary high-performance liquid chromatography (LC). ICAT-labeled peptide pairs are chemically identical and easily visualized. Peptide fragments faithfully maintain the ratios of the original amounts of proteins from the two cell states. Relative peptide/protein quantification is calculated by the ratios of the d0- and d8-tagged peptide pairs. Periodic scans are devoted to fragmenting peptides and recording sequence information (tandem mass spectrum [MS/MS]). Protein identifications are made by database searches of the tandem mass spectrum.
retains sample complexity while allowing for accurate relative protein quantification.48,49
The ICAT approach has been used to identify secreted50 and androgen-regulated proteins51 expressed by prostate cancer cells. In addition to identifying genes previously not known to be regulated by androgens, comparisons of global transcript and protein abundance levels showed that for most genes (>90%), protein levels were concordant with transcript abundance. However, there were distinct outliers that indicate multiple levels of gene expression regulation (e.g. post-transcriptional, post-translational). These results suggest immediately testable hypotheses for characterizing mechanism(s) of gene expression regulation. The results also demonstrate that to fully delineate a gene expression profile, measurements of protein levels are necessary, since, for some genes, protein alterations would not have been predicted by transcript abundance measurements.
Was this article helpful?