We carried out a test sample study to try to identify errors leading to irreproducibility, including incompleteness of peptide sampling, in LC-MS-based proteomics. We distributed a test sample consisting of an equimolar mix of 20 highly purified recombinant human proteins, to 27 laboratories for identification. Each protein contained one or more unique tryptic peptides of 1250 Da to also test for ion selection and sampling in the mass spectrometer. Of the 27 labs, initially only 7 labs reported all 20 proteins correctly, and only 1 lab reported all the tryptic peptides of 1250 Da. Nevertheless, a subsequent centralized analysis of the raw data revealed that all 20 proteins and most of the 1250 Da peptides had in fact been detected by all 27 labs. The centralized analysis allowed us to determine sources of problems encountered in the study, which include missed identifications (false negatives), environmental contamination, database matching, and curation of protein identifications. Improved search engines and databases are likely to increase the fidelity of mass spectrometry-based proteomics.

The current in-depth proteomics makes use of long chromatography gradient to get access to more peptides for protein identification, resulting in covering of as many as 8000 mammalian gene products in 3 days of mass spectrometer running time. Here we report a fast sequencing (Fast-seq) workflow of the use of dual reverse phase high performance liquid chromatography -mass spectrometry (HPLC-MS) with a short gradient to achieve the same proteome coverage in 0.5 day. We adapted this workflow to a quantitative version (Fast quantification, Fast-quan) that was compatible to large-scale protein quantification. We subjected two identical samples to the Fast-quan workflow, which allowed us to systematically evaluate different parameters that impact the sensitivity and accuracy of the workflow. Using the statistics of significant test, we unraveled the existence of substantial falsely quantified differential proteins and estimated correlation of false quantification rate and parameters that are applied in label-free quantification. We optimized the setting of parameters that may substantially minimize the rate of falsely quantified differential proteins, and further applied them on a real biological process. With improved efficiency and throughput, we expect that the Fast-seq/Fast-quan workflow, allowing pair wise comparison of two proteomes in 1 day may make MS available to the masses and impact biomedical research in a positive way. Molecular & Cellular Proteomics 12: 10.1074/mcp.M112.025023, 2370-2380, 2013.The performance of mass spectrometry has been improved tremendously over the last few years (1-3), making mass spectrometry-based proteomics a viable approach for largescale protein analysis in biological research. Scientists around the world are striving to fulfill the promise of identifying and quantifying almost all gene products expressed in a cell line or tissue. This would make mass spectrometry-based protein analysis an approach that is compatible to the second-generation mRNA deep-seq technique (4, 5).Two liquid chromatography (LC)-MS strategies have been employed to achieve deep proteome coverage. One is a single run with a long chromatography column and gradient to take advantage of the resolving power of HPLC to reduce the complexity of peptide mixtures; the other is a sequential run with two-dimensional separation (typically ion-exchange and reverse phase) to reduce peptide complexity. It was reported by two laboratories that 2761 and 4500 proteins were identified with a 10 h chromatography gradient on a dual pressure linear ion-trap orbitrap mass spectrometer (LTQ Orbitrap Velos)(6 -8). Similarly, 3734 proteins were identified using a 8 h gradient on a 2 m long column with a hybrid triple quadrupole -time of flight (Q-TOF, AB sciex 5600 Q-TOF)(9) mass spectrometer. The two-dimensional approach has yielded more identification with longer time. For example, 10,006 proteins (representing over 9000 gene products, GPs) 1 were identified in U2OS cell (10), and 10,255 proteins (representing 9207 GPs) from HeL...

Due to a lack of physiologic cytochrome P450 (P450) isoform content, P450 activity is typically only determined at the microsomal level (per milligram of microsomal protein) and not at the isoform level (per picomole of P450 isoform), which could result in the misunderstanding of variations in P450 activity between individuals and further hinder development of personalized medicine. We found that there were large variations in protein content, mRNA levels, and intrinsic activities of the 10 P450s in 100 human liver samples, in which CYP2E1 and CYP2C9 showed the highest expression levels. P450 gene polymorphisms had different effects on activity at two levels: CYP3A5*3 and CYP2A6*9 alleles conferred increased activity at the isoform level but decreased activity at the microsomal level; CYP2C9*3 had no effect at the isoform level but decreased activity at the microsomal level. The different effects at each level stem from the different effects of each polymorphism on the resulting P450 protein. Individuals with

In this article, an effective method for dephosphorylation of phosphopeptides by cerium oxide is described. The dephosphorylation activity of cerium oxide was evaluated by two standard phosphopeptides and the phosphopeptides in digests of phosphoprotein alpha-casein and beta-casein. Results showed that the dephosphorylation of all the phosphopeptides was completed in 10 min, and temperature had little effect on the dephosphorylation, the dephosphorylation could be carried out at 0 degrees C, room temperature and 37 degrees C. The dephosphorylation mediated by cerium oxide can be attributed to Lewis acid and nucleophile activations. Advantages of using cerium oxide as catalyst for the dephosphorylation include: safe, simple, high catalytic activity, and no precise control of the treatment temperature. The method is valid for the phosphorylation of Ser, Thr and Tyr, and can be used for phosphoprotein analysis.