Proteomics is now an established technique which, when properly applied, can lead to large amounts of biological information. These datasets are being systematically collected to establish databases with the expectation that their combinations will provide new biological insights akin to what was done in genomics. Already, the community has established growing databases of protein interactions/pathways, protein localisations, protein phosphorylations and other post-translational modifications. More recently, interest is growing in integrating ongoing proteomic studies with these databases to better understand biological processes. We also recognized that databases will benefit from the development of quantitative proteomics, more detailed studies based on subgroup of proteins, and protein interactions with other biomolecules. Therefore, we thought it would be timely to focus a special issue on applications and methods in integrative proteomics.
Xing-Lin Yang et al. from the Shanghai Institutes for Biological Sciences, explore the kinases/substrate relationships during cell cycle using quantitative phosphoproteomic applied to G1, S and G2/M of HeLa cells. They integrated their phosphoproteomics results with previously established interaction networks. Interestingly, specific changes in phosphorylation sites were observed whereas the majority of phosphoproteins did not change across all phases of the cell cycle.
Yansheng Liu et al. from the Institute of Molecular Systems Biology (Zurich), apply the technique of SWATH-MS for the data-independent quantitative analysis of N-linked glycoproteins. They demonstrated nearly identical limits of quantitation and similar reproducibility to the SRM technique.
Young-Eun Cho et al. from the Kyungpook National University (Republic of Korea), explore the toxic mechanisms of drugs in hepatocytes. Statins are used to control cholesterol levels, but some have been reported to cause adverse effects in the liver. Therefore, they undertook integrative proteomics and transcriptomics to study simvastatin related hepatotoxicity. The integration of these two datasets highlighted pathways potentially involved in the response to simvastatin toxic effects.
Philip Brownridge et al. from the Universities of Liverpool and Manchester, utilize the QconCAT with SRM mass spectrometry to obtain absolute quantitation of the known yeast chaperone proteins. Using this approach, they quantified 51 of the 63 yeast chaperones and quantification as low as 250 copies per cell. The ability to quantify chaperone proteins is significant as they mediate the folding of the majority of proteins and therefore changes in their levels can have drastic effects on maintaining the ability of cells to respond to stress and other factors.
Fangjun Wang et al. from the Dalian Institute of Chemical Physics and the Ottawa Institute of Systems Biology, study the quantitative changes in the phosphoproteome over time in the hippocampus of a mouse model of Alzheimer's Disease. They identified over 1100 phosphorylation sites of which over 130 were observed to change in the mouse following conversion to symptomatic state. The integration of phosphoproteomics and protein interaction networks revealed subgroups of interest for follow-up studies.
Jun Zhu et al. from the Dalian Institute of Chemical Physics and the Dalian Medical University, present an approach for the enrichment of sialic acid-containing N-glycopeptides using of Ti(IV)-IMAC. Compared to the more conventional TiO2, their approach identified 2.5 times more glycopeptides and glycosylation sites. This approach will likely be key in establishing comprehensive databases on N-glycoproteins.
Yimin Tao et al. from the Shanghai Institute of Materia Medica and the Shanghai University of Traditional Chinese Medicine, study the mechanism of huperzine A neuroprotection against AB1–42 toxicity in a cellular model. Nearly 200 of the 2860 protein identified showed significant changes following huperzine A treatment against AB1–42 toxicity. The integration of these results with previously established pathways and protein interaction networks revealed interesting new avenues including the down-regulation of p53 which might be involved in the neuroprotective effects of huperzine A.
Hongbo Guo et al. form the Donnelly Centre for Cellular and Biomolecular Research (University of Toronto) and the Fred Hutchinson Cancer Research Center (Seattle), study the involvement of protein phosphorylation in the fate of human hematopoietic stem/progenitor. Using a series of enrichment approaches they identified over 15000 unique phosphopeptides in over 3500 proteins. The combination of this datasets with previously established pathway and protein interaction networks highlighted possible kinase-substrate involved in activation cascades.
Finally, in their technical brief Ruijun Tian et al. from the Samuel Lunenfeld Research Institute (University of Toronto), report on a mass spectrometric method to study gangliosides obtained from cells and their interactions with proteins. Very few approaches are currently available to study these interactions. Therefore, this development opens the door to exploring new biological questions and to establish databases of protein-ganglioside interactions.
Clearly, integrative proteomics is growing and leading to a better understanding of biological processes. As well, much better biological hypotheses can be derived by integrating different datasets. One caveat of integrative proteomic is that it is only as good as the information in the databases. In recent years, standards have been developed to ensure quality control of the published proteomic datasets. Unfortunately, there are still many datasets published under the banner of proteomics that would never pass the review process in biological journals due to the lack of biological repeats, proper experimental designs, and controls. As well, efforts in technological developments are needed to increase the type, quantity, and quality of the results deposited in these databases. Finally, more bioinformatics tools need to be developed to facilitate the integration of proteomic results over multiple datasets.