Christof Winter

SCOPPI: a structural classification of protein–protein interfaces

SCOPPI, the structural classification of protein–protein interfaces, is a comprehensive database that classifies and annotates domain interactions derived from all known protein structures. SCOPPI applies SCOP domain definitions and a distance criterion to determine inter-domain interfaces. Using a novel method based on multiple sequence and structural alignments of SCOP families, SCOPPI presents a comprehensive geometrical classification of domain interfaces. Various interface characteristics such as number, type and position of interacting amino acids, conservation, interface size, and permanent or transient nature of the interaction are further provided. Proteins in SCOPPI are annotated with Gene Ontology terms, and the ontology can be used to quickly browse SCOPPI. Screenshots are available for every interface and its participating domains. Here, we describe contents and features of the web-based user interface as well as the underlying methods used to generate SCOPPIs data. In addition, we present a number of examples where SCOPPI becomes a useful tool to analyze viral mimicry of human interface binding sites, gene fusion events, conservation of interface residues and diversity of interface localizations. SCOPPI is available at .

Google Goes Cancer: Improving Outcome Prediction for Cancer Patients by Network-Based Ranking of Marker Genes

Predicting the clinical outcome of cancer patients based on the expression of marker genes in their tumors has received increasing interest in the past decade. Accurate predictors of outcome and response to therapy could be used to personalize and thereby improve therapy. However, state of the art methods used so far often found marker genes with limited prediction accuracy, limited reproducibility, and unclear biological relevance. To address this problem, we developed a novel computational approach to identify genes prognostic for outcome that couples gene expression measurements from primary tumor samples with a network of known relationships between the genes. Our approach ranks genes according to their prognostic relevance using both expression and network information in a manner similar to Googles PageRank. We applied this method to gene expression profiles which we obtained from 30 patients with pancreatic cancer, and identified seven candidate marker genes prognostic for outcome. Compared to genes found with state of the art methods, such as Pearson correlation of gene expression with survival time, we improve the prediction accuracy by up to 7%. Accuracies were assessed using support vector machine classifiers and Monte Carlo cross-validation. We then validated the prognostic value of our seven candidate markers using immunohistochemistry on an independent set of 412 pancreatic cancer samples. Notably, signatures derived from our candidate markers were independently predictive of outcome and superior to established clinical prognostic factors such as grade, tumor size, and nodal status. As the amount of genomic data of individual tumors grows rapidly, our algorithm meets the need for powerful computational approaches that are key to exploit these data for personalized cancer therapies in clinical practice.

MeMotif: a database of linear motifs in α-helical transmembrane proteins

Membrane proteins are important for many processes in the cell and used as main drug targets. The increasing number of high-resolution structures available makes for the first time a characterization of local structural and functional motifs in α-helical transmembrane proteins possible. MeMotif (http://projects.biotec.tu-dresden.de/memotif) is a database and wiki which collects more than 2000 known and novel computationally predicted linear motifs in α-helical transmembrane proteins. Motifs are fully described in terms of several structural and functional features and editable. Motifs contained in MeMotif can be used in different biological applications, from the identification of biochemically important functional residues which are candidates for mutagenesis experiments to the improvement of tools for transmembrane protein modeling.

Xenopus Meiotic Microtubule-Associated Interactome

In metazoan oocytes the assembly of a microtubule-based spindle depends on the activity of a large number of accessory non-tubulin proteins, many of which remain unknown. In this work we isolated the microtubule-bound proteins from Xenopus eggs. Using mass spectrometry we identified 318 proteins, only 43 of which are known to bind microtubules. To integrate our results, we compiled for the first time a network of the meiotic microtubule-related interactome. The map reveals numerous interactions between spindle microtubules and the newly identified non-tubulin spindle components and highlights proteins absent from the mitotic spindle proteome. To validate newly identified spindle components, we expressed as GFP-fusions nine proteins identified by us and for first time demonstrated that Mgc68500, Loc398535, Nif3l1bp1/THOC7, LSM14A/RAP55A, TSGA14/CEP41, Mgc80361 and Mgc81475 are associated with spindles in egg extracts or in somatic cells. Furthermore, we showed that transfection of HeLa cells with siRNAs, corresponding to the human orthologue of Mgc81475 dramatically perturbs spindle formation in HeLa cells. These results show that our approach to the identification of the Xenopus microtubule-associated proteome yielded bona fide factors with a role in spindle assembly.

Examination of Apoptosis Signaling in Pancreatic Cancer by Computational Signal Transduction Analysis

Background Pancreatic ductal adenocarcinoma (PDAC) remains an important cause of cancer death. Changes in apoptosis signaling in pancreatic cancer result in chemotherapy resistance and aggressive growth and metastasizing. The aim of this study was to characterize the apoptosis pathway in pancreatic cancer computationally by evaluation of experimental data from high-throughput technologies and public data bases. Therefore, gene expression analysis of microdissected pancreatic tumor tissue was implemented in a model of the apoptosis pathway obtained by computational protein interaction prediction. Methodology/Principal Findings Apoptosis pathway related genes were assembled from electronic databases. To assess expression of these genes we constructed a virtual subarray from a whole genome analysis from microdissected native tumor tissue. To obtain a model of the apoptosis pathway, interactions of members of the apoptosis pathway were analysed using public databases and computational prediction of protein interactions. Gene expression data were implemented in the apoptosis pathway model. 19 genes were found differentially expressed and 12 genes had an already known pathophysiological role in PDAC, such as Survivin/BIRC5, BNIP3 and TNF-R1. Furthermore we validated differential expression of IL1R2 and Livin/BIRC7 by RT-PCR and immunohistochemistry. Implementation of the gene expression data in the apoptosis pathway map suggested two higher level defects of the pathway at the level of cell death receptors and within the intrinsic signaling cascade consistent with references on apoptosis in PDAC. Protein interaction prediction further showed possible new interactions between the single pathway members, which demonstrate the complexity of the apoptosis pathway. Conclusions/Significance Our data shows that by computational evaluation of public accessible data an acceptable virtual image of the apoptosis pathway might be given. By this approach we could identify two higher level defects of the apoptosis pathway in PDAC. We could further for the first time identify IL1R2 as possible candidate gene in PDAC.

Structural templates predict novel protein interactions and targets from pancreas tumour gene expression data

MOTIVATION Pancreatic ductal adenocarcinoma (PDAC) eludes early detection and is characterized by its aggressiveness and resistance to current therapies. A number of gene expression screens have been carried out to identify genes differentially expressed in cancerous tissue. To identify molecular markers and suitable targets, these genes have been mapped to protein interactions to gain an understanding at systems level. RESULTS Here, we take such a network-centric approach to pancreas cancer by re-constructing networks from known interactions and by predicting novel protein interactions from structural templates. The pathways we find to be largely affected are signal transduction, actin cytoskeleton regulation, cell growth and cell communication. Our analysis indicates that the alteration of the calcium pathway plays an important role in pancreas-specific tumorigenesis. Furthermore, our structural prediction method identifies 40 novel interactions including the tissue factor pathway inhibitor 2 (TFPI2) interacting with the transmembrane protease serine 4 (TMPRSS4). Since TMPRSS4 is involved in metastasis formation, we hypothesize that the upregulation of TMPRSS4 and the downregulation of its predicted inhibitor TFPI2 plays an important role in this process. Moreover, we examine the potential role of BVDU (RP101) as an inhibitor of TMPRSS4. BDVU is known to support apoptosis and prevent the acquisition of chemoresistance. Our results suggest that BVDU might bind to the active site of TMPRSS4, thus reducing its assistance in metastasis. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.

Protein interactions in 3D: From interface evolution to drug discovery

Over the past 10years, much research has been dedicated to the understanding of protein interactions. Large-scale experiments to elucidate the global structure of protein interaction networks have been complemented by detailed studies of protein interaction interfaces. Understanding the evolution of interfaces allows one to identify convergently evolved interfaces which are evolutionary unrelated but share a few key residues and hence have common binding partners. Understanding interaction interfaces and their evolution is an important basis for pharmaceutical applications in drug discovery. Here, we review the algorithms and databases on 3D protein interactions and discuss in detail applications in interface evolution, drug discovery, and interface prediction.

Network information improves cancer outcome prediction

Disease progression in cancer can vary substantially between patients. Yet, patients often receive the same treatment. Recently, there has been much work on predicting disease progression and patient outcome variables from gene expression in order to personalize treatment options. Despite first diagnostic kits in the market, there are open problems such as the choice of random gene signatures or noisy expression data. One approach to deal with these two problems employs protein-protein interaction networks and ranks genes using the random surfer model of Googles PageRank algorithm. In this work, we created a benchmark dataset collection comprising 25 cancer outcome prediction datasets from literature and systematically evaluated the use of networks and a PageRank derivative, NetRank, for signature identification. We show that the NetRank performs significantly better than classical methods such as fold change or t-test. Despite an order of magnitude difference in network size, a regulatory and protein-protein interaction network perform equally well. Experimental evaluation on cancer outcome prediction in all of the 25 underlying datasets suggests that the network-based methodology identifies highly overlapping signatures over all cancer types, in contrast to classical methods that fail to identify highly common gene sets across the same cancer types. Integration of network information into gene expression analysis allows the identification of more reliable and accurate biomarkers and provides a deeper understanding of processes occurring in cancer development and progression.

Structural fragment clustering reveals novel structural and functional motifs in α-helical transmembrane proteins

BackgroundA large proportion of an organisms genome encodes for membrane proteins. Membrane proteins are important for many cellular processes, and several diseases can be linked to mutations in them. With the tremendous growth of sequence data, there is an increasing need to reliably identify membrane proteins from sequence, to functionally annotate them, and to correctly predict their topology.ResultsWe introduce a technique called structural fragment clustering, which learns sequential motifs from 3D structural fragments. From over 500,000 fragments, we obtain 213 statistically significant, non-redundant, and novel motifs that are highly specific to α-helical transmembrane proteins. From these 213 motifs, 58 of them were assigned to function and checked in the scientific literature for a biological assessment. Seventy percent of the motifs are found in co-factor, ligand, and ion binding sites, 30% at protein interaction interfaces, and 12% bind specific lipids such as glycerol or cardiolipins. The vast majority of motifs (94%) appear across evolutionarily unrelated families, highlighting the modularity of functional design in membrane proteins. We describe three novel motifs in detail: (1) a dimer interface motif found in voltage-gated chloride channels, (2) a proton transfer motif found in heme-copper oxidases, and (3) a convergently evolved interface helix motif found in an aspartate symporter, a serine protease, and cytochrome b.ConclusionsOur findings suggest that functional modules exist in membrane proteins, and that they occur in completely different evolutionary contexts and cover different binding sites. Structural fragment clustering allows us to link sequence motifs to function through clusters of structural fragments. The sequence motifs can be applied to identify and characterize membrane proteins in novel genomes.

Rapid Changes of mRNA-binding Protein Levels following Glucose and 3-Isobutyl-1-methylxanthine Stimulation of Insulinoma INS-1 Cells

Glucose and cAMP-inducing agents such as 3-isobutyl-1-methylxanthine (IBMX) rapidly change the expression profile of insulin-producing pancreatic β-cells mostly through post-transcriptional mechanisms. A thorough analysis of these changes, however, has not yet been performed. By combining two-dimensional differential gel electrophoresis and mass spectrometry, we identified 165 spots, corresponding to 78 proteins, whose levels significantly change after stimulation of the β-cell model INS-1 cells with 25 mm glucose + 1 mm IBMX for 2 h. Changes in the expression of selected proteins were verified by one- and two-dimensional immunoblotting. Most of the identified proteins are novel targets of rapid regulation in β-cells. The transcription inhibitor actinomycin D failed to block changes in two-thirds of the spots, supporting their post-transcriptional regulation. More spots changed in response to IBMX than to glucose alone conceivably because of phosphorylation. Fourteen mRNA- binding proteins responded to stimulation, thus representing the most prominent class of rapidly regulated proteins. Bioinformatics analysis indicated that the mRNA 5′- and 3′-untranslated regions of 22 regulated proteins contain potential binding sites for polypyrimidine tract-binding protein 1, which promotes mRNA stability and translation in stimulated β-cells. Overall our findings support the idea that mRNA-binding proteins play a major role in rapid adaptive changes in insulin-producing cells following their stimulation with glucose and cAMP-elevating agents.