We assess the reliability of isobaric-tags for relative and absolute quantitation (iTRAQ), based on different types of replicate analyses taking into account technical, experimental, and biological variations. In total, 10 iTRAQ experiments were analyzed across three domains of life involving Saccharomyces cerevisiae KAY446, Sulfolobus solfataricus P2, and Synechocystis sp. PCC 6803. The coverage of protein expression of iTRAQ analysis increases as the variation tolerance increases. In brief, a cutoff point at +/-50% variation (+/-0.50) would yield 88% coverage in quantification based on an analysis of biological replicates. Technical replicate analysis produces a higher coverage level of 95% at a lower cutoff point of +/-30% variation. Experimental or iTRAQ variations exhibit similar behavior as biological variations, which suggest that most of the measurable deviations come from biological variations. These findings underline the importance of replicate analysis as a validation tool and benchmarking technique in protein expression analysis.

We analyzed 10 isobaric tags for relative and absolute quantitation (iTRAQ) experiments using three different model organisms across the domains of life: Saccharomyces cerevisiae KAY446, Sulfolobussolfataricus P2, and Synechocystis sp. PCC6803. A double database search strategy was employed to minimize the rate of false positives to less than 3% for all organisms. The reliability of proteins with single-peptide identification was also assessed using the search strategy, coupled with multiple analyses of samples into LC-MS/MS. The outcomes of the three LC-MS/MS analyses provided higher proteome coverage with an average increment in total proteins identified of 6%, 33%, and 50% found in S. cerevisiae, S. solfataricus, and Synechocystis sp., respectively. The iTRAQ quantification values were found to be highly reproducible across the injections, with an average coefficient of variation (CV) of 0.09 (scattering from 0.14 to 0.04) calculated based on log mean average ratio for all three organisms. Hence, we recommend multiple analyses of iTRAQ samples for greater proteome coverage and precise quantification.

The unpredictable behavior of prostate cancer presents a major clinical challenge during patient management. In order to gain an insight into the molecular mechanisms associated with prostate cancer progression, we employed the shot-gun proteomic approach of isobaric tags for relative and absolute quantitation (iTRAQ), followed by 2D-LC-MS/MS, using the poorly metastatic LNCaP cell line and its highly metastatic variant LNCaP-LN3 cell line as a model. A total number of 280 unique proteins were identified (> or =95% confidence), and relative expression data was obtained for 176 of these. Ten proteins were found to be significantly up-regulated (> or =1.50 fold), while 4 proteins were significantly down-regulated (> or = -1.50 fold), in LNCaP-LN3 cells. Differential expression of brain creatine kinase (CKBB), soluble catechol-O-methyltransferase (S-COMT), tumor rejection antigen (gp96), and glucose regulated protein, 78 kDa (grp78), was confirmed by Western blotting or independent 2D-PAGE analysis. Additionally, iTRAQ analysis identified absence of the lactate dehydrogenase-B (LDH-B) subunit in LNCaP-LN3 cells, confirming our published data. The clinical relevance of gp96 was assessed by immunohistochemistry using prostate tissues from benign ( n = 95), malignant ( n = 66), and metastatic cases ( n = 3). Benign epithelium showed absent/weak gp96 expression in the basal cells, in contrast to the moderate/strong expression seen in malignant epithelium. Furthermore, there was a statistically significant difference in the intensity of gp96 expression between benign and malignant cases ( p < 0.0005, Mann-Whitney U). Our study is the first to report the application of iTRAQ technology and its potential for the global proteomic profiling of prostate cancer cells, including the identification of absent protein expression.

Electrostatic repulsion-hydrophilic interaction chromatography (ERLIC) has been introduced recently for phosphopeptide enrichment. Here we compared ERLIC with the well-established SCX-IMAC for identifying phosphopeptides in EGF-treated A431 cells. The ERLIC approach detected a higher number of phosphopeptides (17 311) than SCX-IMAC (4850), but it only detected 926 unique phosphopeptides compared to 1315 in SCX-IMAC. Only 12% unique phosphopeptides were common to both approaches, suggesting that more comprehensive phosphoproteomes could be generated by complementing SCX-IMAC with ERLIC.