Search

Summary: Endeavour is a tool that detects the most promising genes within large lists of candidates with respect to a biological process of interest and by combining several genomic data sources. We have benchmarked Endeavour using 450 pathway maps and 826 disease marker sets from MetaCore of GeneGo, Inc. containing a total of 9911 and 12 432 genes, respectively. We obtained an area under the receiver operating characteristic curves of 0.97 for pathway and of 0.91 for disease gene sets. These results indicate that Endeavour can be used to efficiently prioritize candidate genes for pathways and diseases., Availability: Endeavour is available at http://www.esat.kuleuven.be/endeavour

Microarrays to examine the global expression levels of microRNAs (miRNAs) in a systematic in-parallel manner have become important tools to help unravel the functions of miRNAs and to understand their roles in RNA-based regulation and their implications in human diseases. We have established a novel miRNA-specific microarray platform that enables the simultaneous expression analysis of both known and predicted miRNAs obtained from human or mouse origin. Chemically modified 2'-O-(2-methoxyethyl)-(MOE) oligoribonucleotide probes were arrayed onto Evanescent Resonance (ER) microchips by robotic spotting. Supplementing the complementary probes against miRNAs with carefully designed mismatch controls allowed for accurate sequence-specific determination of miRNA expression profiles obtained from a panel of mouse tissues. This revealed new expression signatures of known miRNAs as well as of novel miRNAs previously predicted using bioinformatic methods. Systematic confirmation of the array data with northern blotting and, in particular, real-time PCR suggests that the described microarray platform is a powerful tool to analyze miRNA expression patterns with rapid throughput and high fidelity.

The intramembrane-cleaving proteases (I-CLiPs) presenilin-1 and -2 (PS1 and PS2), signal peptide peptidase (SPP) and the Site-2 protease (S2P) catalyze critical steps in cell signaling and are implicated in diseases such as Alzheimer's disease, hepatitis C virus (HCV) infection and cholesterol homeostasis. Here we describe the development of a cellular assay based on cleavage of the transmembrane sequence of the HCV core protein precursor, releasing intra- and extra-cellular signals that represent sequential signal peptidase and SPP cleavage, respectively. We find that the SPP inhibitor (Z-LL)2-ketone (IC50 = 1.33 microM) and the gamma-secretase potent inhibitors NVP-AHW700-NX (IC50 = 51 nM) and LY411575 (IC50 = 61 nM) but not DAPT dose dependently inhibited SPP but not signal peptidase cleavage. Our data confirm that type II orientated substrates, like the HCV transmembrane sequence, are sequentially cleaved by signal peptidase then SPP. This dual assay provides a powerful tool to pharmacologically analyze sequential cleavage events of signal peptidase and SPP and their regulation.

The human genome encodes seven intramembrane-cleaving GXGD aspartic proteases. These are the two presenilins that activate signaling molecules and are implicated in Alzheimer's disease, signal peptide peptidase (SPP), required for immune surveillance, and four SPP-like candidate proteases (SPPLs), of unknown function. Here we describe a comparative analysis of the topologies of SPP and its human homologues, SPPL2a, -2b, -2c, and -3. We demonstrate that their N-terminal extensions are located in the extracellular space and, except for SPPL3, are modified with N-glycans. Whereas SPPL2a, -2b, and -2c contain a signal sequence, SPP and SPPL3 contain a type I signal anchor sequence for initiation of protein translocation and membrane insertion. The hydrophilic loops joining the transmembrane regions, which contain the catalytic residues, are facing the exoplasm. The C termini of all these proteins are exposed toward the cytosol. Taken together, our study demonstrates that SPP and its homologues are all of the same principal structure with a catalytic domain embedded in the membrane in opposite orientation to that of presenilins. Other than presenilins, SPPL2a, -2b, -2c, and -3 are therefore predicted to cleave type II-oriented substrate peptides like the prototypic protease SPP.

The 3Dee database of domain definitions was developed as a comprehensive collection of domain definitions for all three-dimensional structures in the Protein Data Bank (PDB). The database includes definitions for complex, multiple-segment and multiple-chain domains as well as simple sequential domains, organized in a structural hierarchy. Two different snapshots of the 3Dee database were analyzed at September 1996 and November 1999. For the November 1999 release, 7,995 PDB entries contained 13,767 protein chains and gave rise to 18,896 domains. The domain sequences clustered into 1,715 domain sequence families, which were further clustered into a conservative 1,199 domain structure families (families with similar folds). The proportion of different domain structure families per domain sequence family increases from 84% for domains 1-100 residues long to 100% for domains greater than 600 residues. This is in keeping with the idea that longer chains will have more alternative folds available to them. Of the representative domains from the domain sequence families, 49% are in the range of 51-150 residues, whereas 64% of the representative chains over 200 residues have more than 1 domain. Of the representative chains, 8.5% are part of multichain domains. The largest multichain domain in the database has 14 chains and 1,400 residues, whereas the largest single-chain domain has 907 residues. The largest number of domains found in a protein is 13. The analysis shows that over the history of the PDB, new domain folds have been discovered at a slower rate than by random selection of all known folds. Between 1992 and 1997, a constant 1 in 11 new domains deposited in the PDB has shown no sequence similarity to a previously known domain sequence family, and only 1 in 15 new domain structures has had a fold that has not been seen previously. A comparison of the September 1996 release of 3Dee to the Structural Classification of Proteins (SCOP) showed that the domain definitions agreed for 80% of the representative protein chains. However, 3Dee provided explicit domain boundaries for more proteins. 3Dee is accessible on the World Wide Web at http://barton.ebi.ac.uk/servers/3Dee.html.

The receptor-binding AB loop of recombinant human interferon-gamma (IFN-gamma) has multiple contacts with the extracellular part of the IFN-gamma receptor a chain (IFN-gammaRalpha). We explored the possible length of truncated AB loops and their conformations by molecular modelling. Deletions of two amino acids at the tip of the loop were tolerated in the model without van der Waals collisions of the AB loop with helix F. Based on these modelling results, two deletion mutants were constructed by overlap-extension PCR mutagenesis: des-(A23, D24)-IFN-gamma and des-(N25, G26)-IFN-gamma. Both mutations were tolerated by the folding pattern of recombinant human IFN-gamma, as proved by CD spectroscopy. The stability of both mutants against cosolvent-induced unfolding was equal to that of wild-type IFN-gamma. In contrast to the biophysical similarities of wild-type and mutant IFN-gamma proteins, the biological activities of both mutants dropped significantly. Antiviral activity and human leucocyte antigen (HLA)-DR induction of des-(N25, G26)-IFN-gamma was 10% that of wild-type activity. des-(A23, D24)-IFN-gamma had only 1% remaining activity. Receptor-binding experiments confirmed that both deletions had a negative influence on the affinity of recombinant human IFN-gamma to its cellular receptor. We conclude from this combined molecular modelling and mutagenesis experiments, that the reduced flexibility of the truncated AB loop abrogates the possibility of the formation of a 3(10) helix in the receptor-bound state as observed in the X-ray structure of the IFN-gammaRalpha-IFN-gamma complex.

D-2-hydroxyisocaproate dehydrogenase (D-HicDH) from Lactobacillus casei is a homodimer with 333 amino acids and a molecular mass of 37 kDa per subunit. The enzyme belongs to the protein family of NAD+-dependent D-2-hydroxycarboxylate dehydrogenases and within this family to the subgroup of D-lactate dehydrogenases (D-LDHs). Compared with other D-LDHs D-HicDH is characterized by a very low specificity regarding size and chemical constitution of the accepted D-2-hydroxycarboxylates. Hexagonal crystals of recombinant D-HicDH in the presence of NAD+ and 2-oxoisocaproate (4-methyl-2-oxopentanoate) were grown with ammonium sulphate as precipitating agent. The structure of these crystals was solved by molecular replacement and refined to a final R-factor of 19.6% for all measured X-ray reflections in the resolution range (infinity to 1.86 A). Both NAD+ and 2-oxoisocaproate were identified in the electron density map; binding of the latter in the active site, however, competes with a sulphate ion, which is also defined by electron density. Additionally the final model contains 182 water molecules and a second sulphate ion. The binding of both an in vitro substrate and the natural cosubstrate in the active site provides substantial insight into the catalytic mechanism and allows us to assess previously published active site models for this enzyme family, in particular the two most controversial points, the role of the conserved Arg234 and substrate binding. Furthermore the overall topology and details of the D-HicDH structure are described, discussed against the background of homologous structures and compared with one closely and one distantly related protein.

The prediction of protein structure in insertion/deletion regions (referred to as indels) is an important part of protein model building by homology. Here we combine cluster analysis with data base search procedures. Initially, data bases of representative protein fragments are constructed using two different clustering algorithms. In the HCAPD (hierarchical clustering after preliminary division) approach, all protein fragments are divided into classes with similar anchor region structures (a protein fragment consists of two anchoring regions and a central region). Within these classes the fragments are further clustered using a hierarchical cluster algorithm. The DCANN (deterministic clustering by assignment of all nearest neighbours) approach is a variant of the k-nearest neighbours cluster algorithm. Only geometric scoring criteria are used for data base searching. The main advantage of a non-redundant data base is the ability to provide structurally different fragments during the search process, which leads to an improvement in structure prediction. Both methods have been tested on 71 insertions and 74 deletions with lengths between one and eight residues.