A method for investigating protein-protein interactions related to salmonella typhimurium pathogenesis.

We successfully modified an existing method to investigate protein-protein interactions in the pathogenic bacterium Salmonella enterica serovar Typhimurium (Salmonella Typhimurium). This method includes (i) addition of a histidine-biotin-histidine tag to the bait proteins via recombinant DNA techniques, (ii) in vivo cross-linking with formaldehyde, (iii) tandem affinity purification of bait proteins under fully denaturing conditions, and (iv) identification of the proteins cross-linked to the bait proteins by liquid-chromatography in conjunction with tandem mass-spectrometry. In vivo cross-linking stabilized protein interactions and permitted the subsequent two-step purification step conducted under denaturing conditions. The two-step purification greatly reduced nonspecific binding of noncross-linked proteins to bait proteins. Two different negative controls were employed to eliminate the possibility of identifying background and nonspecific proteins as interacting partners, especially those caused by nonspecific binding to the stationary phase used for protein purification. In an initial demonstration of this approach, we tagged three Salmonella proteinsHimD, PduB and PhoPwith known binding partners that ranged from stable (e.g., HimD) to transient (i.e., PhoP). Distinct sets of interacting proteins were identified for each bait protein, including the known binding partners such as HimA for HimD, as well as unexpected binding partners. Our results suggest that novel protein-protein interactions identified may be critical to pathogenesis by Salmonella.

An analysis pipeline for the inference of protein-protein interaction networks.

We present a platform for the reconstruction of protein-protein interaction networks inferred from Mass Spectrometry (MS) bait-prey data. The Software Environment for Biological Network Inference (SEBINI), an environment for the deployment of network inference algorithms that use high-throughput data, forms the platform core. Among the many algorithms available in SEBINI is the Bayesian Estimator of Probabilities of Protein-Protein Associations (BEPro3) algorithm, which is used to infer interaction networks from such MS affinity isolation data. Also, the pipeline incorporates the Collective Analysis of Biological Interaction Networks (CABIN) software. We have thus created a structured workflow for protein-protein network inference and supplemental analysis.

Spatial mapping of the neurite and soma proteomes reveals a functional Cdc42/Rac regulatory network.

Neurite extension and growth cone navigation are guided by extracellular cues that control cytoskeletal rearrangements. However, understanding the complex signaling mechanisms that mediate neuritogenesis has been limited by the inability to biochemically separate the neurite and soma for spatial proteomic and bioinformatic analyses. Here, we apply global proteome profiling in combination with a neurite purification methodology for comparative analysis of the soma and neurite proteomes of neuroblastoma cells. The spatial relationship of 4,855 proteins were mapped, revealing networks of signaling proteins that control integrins, the actin cytoskeleton, and axonal guidance in the extending neurite. Bioinformatics and functional analyses revealed a spatially compartmentalized Rac/Cdc42 signaling network that operates in conjunction with multiple guanine-nucleotide exchange factors (GEFs) and GTPase-activating proteins (GAPs) to control neurite formation. Interestingly, RNA interference experiments revealed that the different GEFs and GAPs regulate specialized functions during neurite formation, including neurite growth and retraction kinetics, cytoskeletal organization, and cell polarity. Our findings provide insight into the spatial organization of signaling networks that enable neuritogenesis and provide a comprehensive system-wide profile of proteins that mediate this process, including those that control Rac and Cdc42 signaling.

DeconMSn: a software tool for accurate parent ion monoisotopic mass determination for tandem mass spectra.

UNLABELLED: DeconMSn accurately determines the monoisotopic mass and charge state of parent ions from high-resolution tandem mass spectrometry data, offering significant improvement for LTQ_FT and LTQ_Orbitrap instruments over the commercially delivered Thermo Fisher Scientific's extract_msn tool. Optimal parent ion mass tolerance values can be determined using accurate mass information, thus improving peptide identifications for high-mass measurement accuracy experiments. For low-resolution data from LCQ and LTQ instruments, DeconMSn incorporates a support-vector-machine-based charge detection algorithm that identifies the most likely charge of a parent species through peak characteristics of its fragmentation pattern. AVAILABILITY: http://ncrr.pnl.gov/software/ or http://www.proteomicsresource.org/.

Comparative bacterial proteomics: analysis of the core genome concept.

While comparative bacterial genomic studies commonly predict a set of genes indicative of common ancestry, experimental validation of the existence of this core genome requires extensive measurement and is typically not undertaken. Enabled by an extensive proteome database developed over six years, we have experimentally verified the expression of proteins predicted from genomic ortholog comparisons among 17 environmental and pathogenic bacteria. More exclusive relationships were observed among the expressed protein content of phenotypically related bacteria, which is indicative of the specific lifestyles associated with these organisms. Although genomic studies can establish relative orthologous relationships among a set of bacteria and propose a set of ancestral genes, our proteomics study establishes expressed lifestyle differences among conserved genes and proposes a set of expressed ancestral traits.

Comparison of probability and likelihood models for peptide identification from tandem mass spectrometry data.

We evaluate statistical models used in two-hypothesis tests for identifying peptides from tandem mass spectrometry data. The null hypothesis H(0), that a peptide matches a spectrum by chance, requires information on the probability of by-chance matches between peptide fragments and peaks in the spectrum. Likewise, the alternate hypothesis H(A), that the spectrum is due to a particular peptide, requires probabilities that the peptide fragments would indeed be observed if it was the causative agent. We compare models for these probabilities by determining the identification rates produced by the models using an independent data set. The initial models use different probabilities depending on fragment ion type, but uniform probabilities for each ion type across all of the labile bonds along the backbone. More sophisticated models for probabilities under both H(A) and H(0) are introduced that do not assume uniform probabilities for each ion type. In addition, the performance of these models using a standard likelihood model is compared to an information theory approach derived from the likelihood model. Also, a simple but effective model for incorporating peak intensities is described. Finally, a support-vector machine is used to discriminate between correct and incorrect identifications based on multiple characteristics of the scoring functions. The results are shown to reduce the misidentification rate significantly when compared to a benchmark cross-correlation based approach.

A proteomic study of the HUPO Plasma Proteome Project's pilot samples using an accurate mass and time tag strategy.

Characterization of the human blood plasma proteome is critical to the discovery of routinely useful clinical biomarkers. We used an accurate mass and time (AMT) tag strategy with high-resolution mass accuracy cLC-FT-ICR MS to perform a global proteomic analysis of pilot study samples as part of the HUPO Plasma Proteome Project. HUPO reference serum and citrated plasma samples from African Americans, Asian Americans, and Caucasian Americans were analyzed, in addition to a Pacific Northwest National Laboratory reference serum and plasma. The AMT tag strategy allowed us to leverage two previously published "shotgun" proteomics experiments to perform global analyses on these samples in triplicate in less than 4 days total analysis time. A total of 722 (22% with multiple peptide identifications) International Protein Index redundant proteins, or 377 protein families by ProteinProphet, were identified over the six individual HUPO serum and plasma samples. The samples yielded a similar number of identified redundant proteins in the plasma samples (average 446 +/- 23) as found in the serum samples (average 440 +/- 20). These proteins were identified by an average of 956 +/- 35 unique peptides in plasma and 930 +/- 11 unique peptides in serum. In addition to this high-throughput analysis, the AMT tag approach was used with a Z-score normalization to compare relative protein abundances. This analysis highlighted both known differences in serum and citrated plasma such as fibrinogens, and reproducible differences in peptide abundances from proteins such as soluble activin receptor-like kinase 7b and glycoprotein m6b. The AMT tag strategy not only improved our sample throughput but also provided a basis for estimated quantitation.

Global profiling of Shewanella oneidensis MR-1: expression of hypothetical genes and improved functional annotations.

The gamma-proteobacterium Shewanella oneidensis strain MR-1 is a metabolically versatile organism that can reduce a wide range of organic compounds, metal ions, and radionuclides. Similar to most other sequenced organisms, approximately 40% of the predicted ORFs in the S. oneidensis genome were annotated as uncharacterized "hypothetical" genes. We implemented an integrative approach by using experimental and computational analyses to provide more detailed insight into gene function. Global expression profiles were determined for cells after UV irradiation and under aerobic and suboxic growth conditions. Transcriptomic and proteomic analyses confidently identified 538 hypothetical genes as expressed in S. oneidensis cells both as mRNAs and proteins (33% of all predicted hypothetical proteins). Publicly available analysis tools and databases and the expression data were applied to improve the annotation of these genes. The annotation results were scored by using a seven-category schema that ranked both confidence and precision of the functional assignment. We were able to identify homologs for nearly all of these hypothetical proteins (97%), but could confidently assign exact biochemical functions for only 16 proteins (category 1; 3%). Altogether, computational and experimental evidence provided functional assignments or insights for 240 more genes (categories 2-5; 45%). These functional annotations advance our understanding of genes involved in vital cellular processes, including energy conversion, ion transport, secondary metabolism, and signal transduction. We propose that this integrative approach offers a valuable means to undertake the enormous challenge of characterizing the rapidly growing number of hypothetical proteins with each newly sequenced genome.

Identification of proteins in human cytomegalovirus (HCMV) particles: the HCMV proteome.

Human cytomegalovirus (HCMV), a member of the herpesvirus family, is a large complex enveloped virus composed of both viral and cellular gene products. While the sequence of the HCMV genome has been known for over a decade, the full set of viral and cellular proteins that compose the HCMV virion are unknown. To approach this problem we have utilized gel-free two-dimensional capillary liquid chromatography-tandem mass spectrometry (MS/MS) and Fourier transform ion cyclotron resonance MS to identify and determine the relative abundances of viral and cellular proteins in purified HCMV AD169 virions and dense bodies. Analysis of the proteins from purified HCMV virion preparations has indicated that the particle contains significantly more viral proteins than previously known. In this study, we identified 71 HCMV-encoded proteins that included 12 proteins encoded by known viral open reading frames (ORFs) previously not associated with virions and 12 proteins from novel viral ORFs. Analysis of the relative abundance of HCMV proteins indicated that the predominant virion protein was the pp65 tegument protein and that gM rather than gB was the most abundant glycoprotein. We have also identified over 70 host cellular proteins in HCMV virions, which include cellular structural proteins, enzymes, and chaperones. In addition, analysis of HCMV dense bodies indicated that these viral particles are composed of 29 viral proteins with a reduced quantity of cellular proteins in comparison to HCMV virions. This study provides the first comprehensive quantitative analysis of the viral and cellular proteins that compose infectious particles of a large complex virus.

Multidimensional proteome analysis of human mammary epithelial cells.

Recent multidimensional liquid chromatography MS/MS studies have contributed to the identification of large numbers of expressed proteins for numerous species. The present study couples size exclusion chromatography of intact proteins with the separation of tryptically digested peptides using a combination of strong cation exchange and high resolution, reversed phase capillary chromatography to identify proteins extracted from human mammary epithelial cells (HMECs). In addition to conventional conservative criteria for protein identifications, the confidence levels were additionally increased through the use of peptide normalized elution times (NET) for the liquid chromatographic separation step. The combined approach resulted in a total of 5838 unique peptides identified covering 1574 different proteins with an estimated 4% gene coverage of the human genome, as annotated by the National Center for Biotechnology Information (NCBI). This database provides a baseline for comparison against variations in other genetically and environmentally perturbed systems. Proteins identified were categorized based upon intracellular location and biological process with the identification of numerous receptors, regulatory proteins, and extracellular proteins, demonstrating the usefulness of this application in the global analysis of human cells for future comparative studies.

Use of artificial neural networks for the accurate prediction of peptide liquid chromatography elution times in proteome analyses.

The use of artificial neural networks (ANNs) is described for predicting the reversed-phase liquid chromatography retention times of peptides enzymatically digested from proteome-wide proteins. To enable the accurate comparison of the numerous LC/MS data sets, a genetic algorithm was developed to normalize the peptide retention data into a range (from 0 to 1), improving the peptide elution time reproducibility to approximately 1%. The network developed in this study was based on amino acid residue composition and consists of 20 input nodes, 2 hidden nodes, and 1 output node. A data set of approximately 7000 confidently identified peptides from the microorganism Deinococcus radiodurans was used for the training of the ANN. The ANN was then used to predict the elution times for another set of 5200 peptides tentatively identified by MS/MS from a different microorganism (Shewanella oneidensis). The model was found to predict the elution times of peptides with up to 54 amino acid residues (the longest peptide identified after tryptic digestion of S. oneidensis) with an average accuracy of approximately 3%. This predictive capability was then used to distinguish with high confidence isobar peptides otherwise indistinguishable by accurate mass measurements as well as to uncover peptide misidentifications. Thus, integration of ANN peptide elution time prediction in the proteomic research will increase both the number of protein identifications and their confidence.

Toward a human blood serum proteome: analysis by multidimensional separation coupled with mass spectrometry.

Blood serum is a complex body fluid that contains various proteins ranging in concentration over at least 9 orders of magnitude. Using a combination of mass spectrometry technologies with improvements in sample preparation, we have performed a proteomic analysis with submilliliter quantities of serum and increased the measurable concentration range for proteins in blood serum beyond previous reports. We have detected 490 proteins in serum by on-line reversed-phase microcapillary liquid chromatography coupled with ion trap mass spectrometry. To perform this analysis, immunoglobulins were removed from serum using protein A/G, and the remaining proteins were digested with trypsin. Resulting peptides were separated by strong cation exchange chromatography into distinct fractions prior to analysis. This separation resulted in a 3-5-fold increase in the number of proteins detected in an individual serum sample. With this increase in the number of proteins identified we have detected some lower abundance serum proteins (ng/ml range) including human growth hormone, interleukin-12, and prostate-specific antigen. We also used SEQUEST to compare different protein databases with and without filtering. This comparison is plotted to allow for a quick visual assessment of different databases as a subjective measure of analytical quality. With this study, we have performed the most extensive analysis of serum proteins to date and laid the foundation for future refinements in the identification of novel protein biomarkers of disease.