Experiment-specific estimation of peptide identification probabilities using a randomized database

Determining the error rate for peptide and protein identification accurately and reliably is necessary to enable evaluation and crosscomparisons of high throughput proteomics experiments. Currently, peptide identification is based either on preset scoring thresholds or on probabilistic models trained on datasets that are often dissimilar to experimental results. The false discovery rates (FDR) and peptide identification probabilities for these preset thresholds or models often vary greatly across different experimental treatments, organisms, or instruments used in specific experiments. To overcome these difficulties, randomized databases have been used to estimate the FDR. However, the cumulative FDR may include low probability identifications when there are a large number of peptide identifications and exclude high probability identifications when there are few. To overcome this logical inconsistency, this study expands the use of randomized databases to generate experiment-specific estimates of peptide identification probabilities. These experiment-specific probabilities are generated by logistic and Loess regression models of the peptide scores obtained from original and reshuffled database matches. These experiment-specific probabilities are shown to very well approximate "true" probabilities based on known standard protein mixtures across different experiments. Probabilities generated by the earlier Peptide_Prophet and more recent LIPS models are shown to differ significantly from this study's experiment-specific probabilities, especially for unknown samples. The experiment-specific probabilities reliably estimate the accuracy of peptide identifications and overcome potential logical inconsistencies of the cumulative FDR. This estimation method is demonstrated using a Sequest database search, LIPS model, and a reshuffled database. However, this approach is generally applicable to any search algorithm, peptide scoring, and statistical model when using a randomized database.     

Interaction of gamma 1-syntrophin with diacylglycerol kinase-zeta. Regulation of nuclear localization by PDZ interactions

Syntrophins are modular adapter proteins that link ion channels and signaling proteins to dystrophin and its homologues. A yeast two-hybrid screen of a human brain cDNA library using the PDZ domain of gamma 1- syntrophin, a recently identified brain-specific isoform, yielded overlapping clones encoding the C terminus of diacylglycerol kinase-zeta (DGK-zeta), an enzyme that converts diacylglycerol into phosphatidic acid. In biochemical assays, the C terminus of DGK-zeta, which contains a consensus PDZ-binding motif, was found to be necessary and sufficient for association with gamma 1-syntrophin. When coexpressed in HeLa cells, DGK-zeta and gamma 1-syntrophin formed a stable complex that partitioned between the cytoplasm and nucleus. DGK-zeta translocates from the cytosol to the nucleus, a process negatively regulated by protein kinase C phosphorylation. We found that DGK-zeta recruits gamma 1-syntrophin into the nucleus and that the PDZ-binding motif is required. Disrupting the interaction altered the intracellular localization of both proteins; DGK-zeta accumulated in the nucleus, whereas gamma 1-syntrophin remained in the cytoplasm. The level of endogenous syntrophins in the nucleus of HeLa cells also reflected the amount of nuclear DGK-zeta. In the brain, DGK-zeta and gamma 1-syntrophin were colocalized in cell bodies and dendrites of cerebellar Purkinjie neurons and other neuronal cell types, suggesting that their interaction is physiologically relevant. Moreover, coimmunoprecipitation and pull-down experiments from brain extracts and cells suggest that DGK-zeta, gamma 1-syntrophin, and dystrophin form a ternary complex. Collectively, our results suggest that gamma 1-syntrophin participates in regulating the subcellular localization of DGK-zeta to ensure correct termination of diacylglycerol signaling.     

Mycobacterium bovis strain bacillus Calmette-Guérin-induced liver granulomas contain a diverse TCR repertoire, but a monoclonal T cell population is sufficient for protective granuloma formation

Granuloma formation is a form of delayed-type hypersensitivity requiring CD4(+) T cells. Granulomas control the growth and dissemination of pathogens, preventing host inflammation from harming surrounding tissues. Using a murine model of Mycobacterium bovis strain bacillus Calmette-Guérin (BCG) infection we studied the extent of T cell heterogeneity present in liver granulomas. We demonstrate that the TCR repertoire of granuloma-infiltrating T cells is very diverse even at the single-granuloma level, suggesting that before granuloma closure, a large number of different T cells are recruited to the lesion. At the same time, the TCR repertoire is selected, because AND TCR transgenic T cells (Valpha11/Vbeta3 anti-pigeon cytochrome c) are preferentially excluded from granulomas of BCG-infected AND mice, and cells expressing secondary endemic Vbeta-chains are enriched among AND cells homing to granulomas. Next, we addressed whether TCR heterogeneity is required for effective granuloma formation. We infected 5CC7/recombinase-activating gene 2(-/-) mice with recombinant BCG that express pigeon cytochrome c peptide in a mycobacterial 19-kDa bacterial surface lipoprotein. A CD4(+) T cell with a single specificity in the absence of CD8(+) T cells is sufficient to form granulomas and adequately control bacteria. Our study shows that expanded monoclonal T cell populations can be protective in mycobacterial infection.     

Identification of MHC class II-associated peptides that promote the presentation of toxic shock syndrome toxin-1 to T cells

Previous studies have shown that the DM-deficient cell line, T2-I-A(b), is very inefficient at presenting toxic shock syndrome toxin 1 (TSST-1) to T cells, suggesting that I-A(b)-associated peptides play an essential role in the presentation of this superantigen. Consistent with this, the loading of an I-A(b)-binding peptide, staphylococcal enterotoxin B 121-136, onto T2-I-A(b) cells enhanced TSST-1 presentation >1000-fold. However, despite extensive screening, no other peptides have been identified that significantly promote TSST-1 presentation. In addition, the peptide effect on TSST-1 presentation has been demonstrated only in the context of the tumor cell line T2-I-A(b). Here we show that peptides that do not promote TSST-1 presentation can be converted into "promoting" peptides by the progressive truncation of C-terminal residues. These studies result in the identification of two peptides derived from IgGV heavy chain and I-Ealpha proteins that are extremely strong promoters of TSST-1 presentation (47,500- and 12,000-fold, respectively). We have also developed a system to examine the role of MHC class II-associated peptides in superantigen presentation using splenic APC taken directly ex vivo. The data confirmed that the length of the MHC class II-bound peptide plays a critical role in the presentation of TSST-1 by splenic APC and showed that different subpopulations of APC are equally peptide dependent in TSST-1 presentation. Finally, we demonstrated that the presentation of staphylococcal enterotoxin A, like TSST-1, is peptide dependent, whereas staphylococcal enterotoxin B presentation is peptide independent.     

Rapid assessment of adenovirus serum neutralizing antibody titer based on quantitative, morphometric evaluation of capsid binding and intracellular trafficking: population analysis of adenovirus capsid association with cells is predictive of adenovirus infectivity

Neutralizing antiviral antibodies are typically detected on the basis of inhibition of viral function, such as propagation of a viral infection or inhibition of viral gene expression. Evidence is presented that anti-adenovirus neutralizing antibodies can be evaluated by analysis of cell-associated capsids or by analysis of intracellular trafficking of the capsids within 1 h after infection. Quantitative analyses of these morphologic parameters represent rapid, broadly applicable, functional assays for the detection of anti-adenovirus neutralizing antibodies.