SubCellBarCode: Proteome-wide Mapping of Protein Localization and Relocalization.

Subcellular localization is a main determinant of protein function; however, a global view of cellular proteome organization remains relatively unexplored. We have developed a robust mass spectrometry-based analysis pipeline to generate a proteome-wide view of subcellular localization for proteins mapping to 12,418 individual genes across five cell lines. Based on more than 83,000 unique classifications and correlation profiling, we investigate the effect of alternative splicing and protein domains on localization, complex member co-localization, cell-type-specific localization, as well as protein relocalization after growth factor inhibition. Our analysis provides information about the cellular architecture and complexity of the spatial organization of the proteome; we show that the majority of proteins have a single main subcellular location, that alternative splicing rarely affects subcellular location, and that cell types are best distinguished by expression of proteins exposed to the surrounding environment. The resource is freely accessible via www.subcellbarcode.org.

PathLocdb: a comprehensive database for the subcellular localization of metabolic pathways and its application to multiple localization analysis.

BACKGROUND: In eukaryotes, the cell is divided into several compartments enclosed by unitary membranes. Such compartmentalization is critical for cells to restrict different pathways to be carried out in different subcellular regions. The summary and classification of subcellular localizations of metabolic pathways are the first steps towards understanding their roles in spatial differentiation and the specialization of metabolic pathways in different organisms. RESULTS: Integrating the subcellular localization of enzymes and their pathways from UniProt Knowledgebase and KEGG pathway databases, we present the first database for subcellular localization of 43014 pathways from 80676 UniProt entries and their pathway annotations from UniProt and KEGG pathway databases. To extract pathway localization across organisms, we defined 889 superpathways as clusters of basic pathways with the same pathway annotations from different organisms. Over eighty-eight percent of superpathways in the Swiss-Prot dataset occur in cytoplasm and mitochondria. And over seventy percent of UniProt superpathways have multiple localization annotations. We summarized four common reasons for the multiple localization of superpathways. Based on this database, we also discovered 88 potential transport systems between different steps of multiply localized pathways and 45 duplicated genes from 17 pathways, occurring in parallel in several locations in humans. CONCLUSIONS: PathLocdb is a free web-accessible database that enables biochemical researchers to quickly access summarized subcellular localization of pathways from UniProt and KEGG pathway databases. As the first effort to systematically integrate pathway localization, this database is very useful in discovering the variation of localization of pathways between organisms and also cross-talk between different organelles within a pathway. The Pathlocdb database is available at http://pathloc.cbi.pku.edu.cn.

Automated recognition system to classify subcellular protein localizations in images of different cell lines acquired by different imaging systems.

Systemic analysis of subcellular protein localization (location proteomics) provides clues for understanding gene functions and physiological condition of the cells. However, recognition of cell images of subcellular structures highly depends on experience and becomes the rate-limiting step when classifying subcellular protein localization. Several research groups have extracted specific numerical features for the recognition of subcellular protein localization, but these recognition systems are restricted to images of single particular cell line acquired by one specific imaging system and not applied to recognize a range of cell image sources. In this study, we establish a single system for automated subcellular structure recognition to identify cell images from various sources. Two different sources of cell images, 317 Vero (http://gfp-cdna.embl.de) and 875 CHO cell images of subcellular structures, were used to train and test the system. When the system was trained by a single source of images, the recognition rate is high and specific to the trained source. The system trained by the CHO cell images gave high average recognition accuracy for CHO cells of 96%, but this was reduced to 46% with Vero images. When we trained the system using a mixture of CHO and Vero cell images, an average accuracy of recognition reached 86.6% for both CHO and Vero cell images. The system can reject images with low confidence and identify the cell images correctly recognized to avoid manual reconfirmation. In summary, we have established a single system that can recognize subcellular protein localizations from two different sources for location-proteomic studies. studies.

Fast automated cell phenotype image classification.

BACKGROUND: The genomic revolution has led to rapid growth in sequencing of genes and proteins, and attention is now turning to the function of the encoded proteins. In this respect, microscope imaging of a protein's sub-cellular localisation is proving invaluable, and recent advances in automated fluorescent microscopy allow protein localisations to be imaged in high throughput. Hence there is a need for large scale automated computational techniques to efficiently quantify, distinguish and classify sub-cellular images. While image statistics have proved highly successful in distinguishing localisation, commonly used measures suffer from being relatively slow to compute, and often require cells to be individually selected from experimental images, thus limiting both throughput and the range of potential applications. Here we introduce threshold adjacency statistics, the essence which is to threshold the image and to count the number of above threshold pixels with a given number of above threshold pixels adjacent. These novel measures are shown to distinguish and classify images of distinct sub-cellular localization with high speed and accuracy without image cropping. RESULTS: Threshold adjacency statistics are applied to classification of protein sub-cellular localization images. They are tested on two image sets (available for download), one for which fluorescently tagged proteins are endogenously expressed in 10 sub-cellular locations, and another for which proteins are transfected into 11 locations. For each image set, a support vector machine was trained and tested. Classification accuracies of 94.4% and 86.6% are obtained on the endogenous and transfected sets, respectively. Threshold adjacency statistics are found to provide comparable or higher accuracy than other commonly used statistics while being an order of magnitude faster to calculate. Further, threshold adjacency statistics in combination with Haralick measures give accuracies of 98.2% and 93.2% on the endogenous and transfected sets, respectively. CONCLUSION: Threshold adjacency statistics have the potential to greatly extend the scale and range of applications of image statistics in computational image analysis. They remove the need for cropping of individual cells from images, and are an order of magnitude faster to calculate than other commonly used statistics while providing comparable or better classification accuracy, both essential requirements for application to large-scale approaches.

Calcium-dependent interaction of calcineurin with Bcl-2 in neuronal tissue.

Calcineurin, a calmodulin-dependent protein phosphatase, regulates transcription and possibly apoptosis. Previous studies demonstrated that in baby hamster kidney-21 cells after co-transfection calcineurin interacts with Bcl-2, thereby altering transcription and apoptosis. Using co-immunoprecipitation and subcellular fractionation techniques, we observed that calcineurin occurred as a complex with Bcl-2 in various regions of rat and mouse brain. The calcineurin-Bcl-2 complex was identified in mitochondrial, nuclear, microsomal and cytosol fractions. In vitro induction of hypoxia and aglycia or N-methyl-D-aspartate treatment markedly altered both extent of complex formation and its subcellular localization. These observations suggest that Bcl-2 either sequesters calcineurin, that calcineurin dephosphorylates Bcl-2, or that Bcl-2 shuttles calcineurin to specific substrates. Calcineurin also co-immunoprecipitated with the inositol-tris-phosphate receptor. This interaction increased after in vitro hypoxia/aglycia. In Bcl-2 (-/-) mice, interactions between calcineurin- and inositol-tris-phosphate receptor occurred less frequently than in wild-type mice under both control and hypoxic conditions. Experiments involving cell-free systems, as well as brain slices treated with thapsigargin or with N-methyl-D-aspartate suggested that calcium and calmodulin activation of calcineurin leads to interactions between calcineurin and Bcl-2. These data indicate that during times of cellular stress and damage, Bcl-2 targets activated calcineurin to specific compartments and substrates.

Subcellular alteration of glyceraldehyde-3-phosphate dehydrogenase in Alzheimer's disease fibroblasts.

The regulation of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) has been implicated both in age-related neurodegenerative disease and in apoptosis. Previous in vitro studies suggest an interaction between GAPDH and the beta-amyloid precursor protein (beta-APP), a protein directly involved in Alzheimer's disease (AD). New studies indicate that GAPDH is a multidimensional protein with diverse membrane, cytoplasmic, and nuclear functions; each is distinct from its role in glycolysis. The nuclear functions of GAPDH include a role in apoptosis that requires its translocation to the nucleus. Accordingly, beta-APP-GAPDH interactions, altering GAPDH structure in vivo, may affect energy generation, inducing hypometabolism, a characteristic AD phenotype. Because GAPDH is a multifunctional protein, pleiotropic effects may also occur in a variety of fundamental cellular pathways in AD cells. This may include unique GAPDH-RNA interactions. We report here the identification of a high-molecular-weight (HMW) GAPDH species present exclusively in the postnuclear fraction of AD cells. The latter is characterized by reduced GAPDH activity. The HMW GAPDH species was not detected in postnuclear age-matched control (AMC) fractions nor in AD whole-cell preparations. Each is characterized by normal GAPDH activity. By definition, the preparation of whole-cell extracts entails the destruction of subcellular structure. The latter findings indicate that the dissociation of the GAPDH protein from the HMW species restores its enzymatic activity. Thus, these results reveal a new, unique intracellular phenotype in AD cells. The functional consequences of subcellular alteration in GAPDH structure in AD cells are considered.

Subcellular distribution of calcium- and calmodulin-dependent myosin light chain phosphorylating activity in rat cerebral cortex.

The subcellular distribution of Ca2+- and calmodulin-dependent myosin light chain phosphorylating activity in rat cerebral cortex was studied. The activity showed a high degree of association to nerve endings (either crude or purified by discontinuous sucrose density gradient centrifugation). After osmotic shock of the nerve endings, the activity was largely membrane associated and could not be released from membranes by freeze-thawing, dilution, 75 mM NaCl or 4 mM EDTA. The association of Ca2+/calmodulin-dependent myosin light chain phosphorylating activity with synaptosomal membranes suggests a role for calcium-dependent myosin phosphorylation in events relating to neurotransmission.

Heterogeneity of human whole blood platelet subpopulations. III. Density-dependent differences in subcellular constituents.

Structurally intact platelet cohorts of differing densities can be isolated from normal subjects by the use of isosmolar arabinogalactan density gradients. Using platelets separated in this fashion, we have studied the density-dependent distribution of four subcellular organelles: mitochondria, lysosomes, dense bodies, and alpha granules. Mitochondria, which are not secreted during platelet release, demonstrate a slow decline in monoamine oxidase activity within the gradient. Lysosomal beta-glucuronidase does not vary significantly with platelet density. In contrast, dense body number and endogenous serotonin content decrease significantly with decreasing platelet density, primarily as the result of differences in the number of storage organelles. Platelet factor 4 content declines rapidly in comparison to lysosomal activities (P less than .001 from bottom to top of the gradient); but beta-thromboglobulin, also an alpha granule component, exhibits considerably less change than platelet factor 4 (P less than .001). Thus, specific platelet subcellular constituents have different density distributions. We postulate that these density differences may be due to differential in vivo loss of selective biochemical constituents from unique subcellular compartments.

Abnormality in calcium release from skeletal sarcoplasmic reticulum of pigs susceptible to malignant hyperthermia.

Two fractions of sarcoplasmic reticulum, one light (LSR) and one heavy (HSR), were isolated from gracilis muscle of control and malignant hyperthermia (MH)-susceptible pigs. Part of the gracilis muscle biopsy was used to compare the contracture sensitivity of the muscle to the calcium-releasing effects of caffeine on isolated SR membranes. Gracilis muscle of MH pigs was more sensitive to the contracture-producing effects of caffeine than control pig muscle. The caffeine dose-cumulative contracture response curve for MH muscle was shifted left of that for controls. The amount of caffeine-induced calcium released from SR is a function of the amount of calcium preload and this did not differ between LSR of MH and control muscle. When LSR fractions were optimally loaded with calcium for caffeine-induced calcium release, no difference in calcium-releasing effects of varying caffeine doses was observed between MH and control LSR. At calcium preloads below optimal, the MH-LSR appeared to be more sensitive to caffeine-induced calcium release. The HSR fractions could not be loaded with calcium in a manner similar to the LSR fractions because of an apparent calcium-induced calcium release phenomenon. Therefore, calcium threshold for calcium-induced calcium release was compared between MH and control HSR fraction. The effect of caffeine on the calcium-induced calcium release was also studied. The average calcium concentration threshold for calcium-induced calcium release was markedly lower for MH vs. control HSR; 20 vs. 63 nmol Ca2+/mg, respectively. Caffeine decreased the threshold for calcium-induced calcium release more in the MH than in control HSR. Under all conditions studied, the amount of calcium released did not differ between the two groups. Ruthenium red increased the threshold calcium concentration for calcium-induced calcium release while it reduced the amount of calcium released. Increasing concentrations of Mg2+ increased the Ca2+ threshold for release and the amount of Ca2+ released but did not significantly affect rate of Ca2+ release. Results of the study suggest a defect in the mechanisms causing calcium release from SR in MH-affected muscle.