Detecting protein function and protein-protein interactions from genome sequences.

A computational method is proposed for inferring protein interactions from genome sequences on the basis of the observation that some pairs of interacting proteins have homologs in another organism fused into a single protein chain. Searching sequences from many genomes revealed 6809 such putative protein-protein interactions in Escherichia coli and 45,502 in yeast. Many members of these pairs were confirmed as functionally related; computational filtering further enriches for interactions. Some proteins have links to several other proteins; these coupled links appear to represent functional interactions such as complexes or pathways. Experimentally confirmed interacting pairs are documented in a Database of Interacting Proteins.

Computational genetics: finding protein function by nonhomology methods.

During the past year, computational methods have been developed that use the rapidly accumulating genomic data to discover protein function. The methods rely on properties shared by functionally related proteins other than sequence or structural similarity. Instead, these 'nonhomology' methods analyze patterns such as domain fusion, conserved gene position and gene co-inheritance and coexpression to identify protein-protein relationships. The methods can identify functions for proteins that are without characterized homologs and have been applied to genome-wide predictions of protein function.

DIP: The Database of Interacting Proteins: 2001 update.

The Database of Interacting Proteins (DIP; http://dip.doe-mbi.ucla. edu) is a database that documents experimentally determined protein-protein interactions. Since January 2000 the number of protein-protein interactions in DIP has nearly tripled to 3472 and the number of proteins to 2659. New interactive tools have been developed to aid in the visualization, navigation and study of networks of protein interactions.

A census of protein repeats.

In this study, we analyzed all known protein sequences for repeating amino acid segments. Although duplicated sequence segments occur in 14 % of all proteins, eukaryotic proteins are three times more likely to have internal repeats than prokaryotic proteins. After clustering the repetitive sequence segments into families, we find repeats from eukaryotic proteins have little similarity with prokaryotic repeats, suggesting most repeats arose after the prokaryotic and eukaryotic lineages diverged. Consequently, protein classes with the highest incidence of repetitive sequences perform functions unique to eukaryotes. The frequency distribution of the repeating units shows only weak length dependence, implicating recombination rather than duplex melting or DNA hairpin formation as the limiting mechanism underlying repeat formation. The mechanism favors additional repeats once an initial duplication has been incorporated. Finally, we show that repetitive sequences are favored that contain small and relatively water-soluble residues. We propose that error-prone repeat expansion allows repetitive proteins to evolve more quickly than non-repeat-containing proteins.

Reducing MCM levels in human primary T cells during the G(0)-->G(1) transition causes genomic instability during the first cell cycle.

DNA replication is tightly regulated, but paradoxically there is reported to be an excess of MCM DNA replication proteins over the number of replication origins. Here, we show that MCM levels in primary human T cells are induced during the G(0)-->G(1) transition and are not in excess in proliferating cells. The level of induction is critical as we show that a 50% reduction leads to increased centromere separation, premature chromatid separation (PCS) and gross chromosomal abnormalities typical of genomic instability syndromes. We investigated the mechanisms involved and show that a reduction in MCM levels causes dose-dependent DNA damage involving activation of ATR & ATM and Chk1 & Chk2. There is increased DNA mis-repair by non-homologous end joining (NHEJ) and both NHEJ and homologous recombination are necessary for Mcm7-depleted cells to progress to metaphase. Therefore, a simple reduction in MCM loading onto DNA, which occurs in cancers as a result of aberrant cell cycle control, is sufficient to cause PCS and gross genomic instability within one cell cycle.

Chicken prion tandem repeats form a stable, protease-resistant domain.

Prion-linked diseases, such as mad cow disease, scrapie, and the human genetic disorder Creutzfeldt-Jakob disease, are fatal neurodegenerative diseases correlated with changes in the secondary structure of neural prion protein. We expressed recombinant chicken prion protein in Escherichia coli and purified the protein to homogeneity. Circular dichroism spectra of the 26 kDa recombinant protein closely resemble those of prion protein purified directly from healthy hamster brain. The chicken prion protein exists as a soluble, monodisperse monomer but can be forced to multimerize following lyophilization and resuspension. We analyzed the chicken prion protein domain structure by proteolysis and show that, unlike the mammalian homologues, the chicken prion protein N-terminal tandem amino acid repeats form a stable, protease-resistant domain. This domain probably represents a physiologically functional unit. As tested by both mass spectrometry and circular dichroism, the mature chicken prion protein does not bind copper, unlike synthetic peptides from the chicken prion N-terminus, suggesting that binding copper is not the physiological activity of the chicken prion. However, copper strongly destabilizes the prion protein and depresses the melting temperature by 30 degreesC, presumably by binding to the unfolded form of the prion protein. The chicken prion N-terminus may have evolved to fold without a cofactor, unlike mammalian prion proteins, whose N-termini are disordered without cofactors such as copper present. Chicken prion offers an alternative to intractable mammalian prions for structural studies of the amino-terminal domain.

A fast algorithm for genome-wide analysis of proteins with repeated sequences.

We present a fast algorithm to search for repeating fragments within protein sequences. The technique is based on an extension of the Smith-Waterman algorithm that allows the calculation of sub-optimal alignments of a sequence against itself. We are able to estimate the statistical significance of all sub-optimal alignment scores. We also rapidly determine the length of the repeating fragment and the number of times it is found in a sequence. The technique is applied to sequences in the Swissprot database, and to 16 complete genomes. We find that eukaryotic proteins contain more internal repeats than those of prokaryotic and archael organisms. The finding that 18% of yeast sequences and 28% of the known human sequences contain detectable repeats emphasizes the importance of internal duplication in protein evolution.

Mining literature for protein-protein interactions.

MOTIVATION: A central problem in bioinformatics is how to capture information from the vast current scientific literature in a form suitable for analysis by computer. We address the special case of information on protein-protein interactions, and show that the frequencies of words in Medline abstracts can be used to determine whether or not a given paper discusses protein-protein interactions. For those papers determined to discuss this topic, the relevant information can be captured for the Database of Interacting PROTEINS: Furthermore, suitable gene annotations can also be captured. RESULTS: Our Bayesian approach scores Medline abstracts for probability of discussing the topic of interest according to the frequencies of discriminating words found in the abstract. More than 80 discriminating words (e.g. complex, interaction, two-hybrid) were determined from a training set of 260 Medline abstracts corresponding to previously validated entries in the Database of Interacting Proteins. Using these words and a log likelihood scoring function, approximately 2000 Medline abstracts were identified as describing interactions between yeast proteins. This approach now forms the basis for the rapid expansion of the Database of Interacting Proteins.

Protein function in the post-genomic era.

Faced with the avalanche of genomic sequences and data on messenger RNA expression, biological scientists are confronting a frightening prospect: piles of information but only flakes of knowledge. How can the thousands of sequences being determined and deposited, and the thousands of expression profiles being generated by the new array methods, be synthesized into useful knowledge? What form will this knowledge take? These are questions being addressed by scientists in the field known as 'functional genomics'.

Localizing proteins in the cell from their phylogenetic profiles.

We introduce a computational method for identifying subcellular locations of proteins from the phylogenetic distribution of the homologs of organellar proteins. This method is based on the observation that proteins localized to a given organelle by experiments tend to share a characteristic phylogenetic distribution of their homologs-a phylogenetic profile. Therefore any other protein can be localized by its phylogenetic profile. Application of this method to mitochondrial proteins reveals that nucleus-encoded proteins previously known to be destined for mitochondria fall into three groups: prokaryote-derived, eukaryote-derived, and organism-specific (i.e., found only in the organism under study). Prokaryote-derived mitochondrial proteins can be identified effectively by their phylogenetic profiles. In the yeast Saccharomyces cerevisiae, 361 nucleus-encoded mitochondrial proteins can be identified at 50% accuracy with 58% coverage. From these values and the proportion of conserved mitochondrial genes, it can be inferred that approximately 630 genes, or 10% of the nuclear genome, is devoted to mitochondrial function. In the worm Caenorhabditis elegans, we estimate that there are approximately 660 nucleus-encoded mitochondrial genes, or 4% of its genome, with approximately 400 of these genes contributed from the prokaryotic mitochondrial ancestor. The large fraction of organism-specific and eukaryote-derived genes suggests that mitochondria perform specialized roles absent from prokaryotic mitochondrial ancestors. We observe measurably distinct phylogenetic profiles among proteins from different subcellular compartments, allowing the general use of prokaryotic genomes in learning features of eukaryotic proteins.

Assigning protein functions by comparative genome analysis: protein phylogenetic profiles.

Determining protein functions from genomic sequences is a central goal of bioinformatics. We present a method based on the assumption that proteins that function together in a pathway or structural complex are likely to evolve in a correlated fashion. During evolution, all such functionally linked proteins tend to be either preserved or eliminated in a new species. We describe this property of correlated evolution by characterizing each protein by its phylogenetic profile, a string that encodes the presence or absence of a protein in every known genome. We show that proteins having matching or similar profiles strongly tend to be functionally linked. This method of phylogenetic profiling allows us to predict the function of uncharacterized proteins.

Structural analysis shows five glycohydrolase families diverged from a common ancestor.

We have solved the X-ray structure of barley chitinase and bacterial chitosanase. Structural constraints predicted these would work by an inverting mechanism, which has been confirmed biochemically. The two enzymes were compared with lysozymes from goose (GEWL), phage (T4L), and hen (HEWL). Although the proteins share no significant amino acid similarities, they are shown to have a structurally invariant core containing two helices and a three-stranded beta sheet that from the substrate binding and catalytic cleft. These enzymes represent a superfamily of hydrolases arising from the divergent evolution of an ancient protein. The glycohydrolase superfamily can be structurally divided into a bacterial family (chitosanase and T4L), and a eucaryotic family represented by chitinase, GEWL, and HEWL. Both families contain the ancestral core but differ at the amino and carboxy termini. The eucaryotes have a small N terminal domain, while the procaryotes have none. The C terminal domain of the eucaryotic family contains a single alpha-helix, while the prokaryotic domain has three antiparallel helices.

Chitinases, chitosanases, and lysozymes can be divided into procaryotic and eucaryotic families sharing a conserved core.

Barley chitinase, bacterial chitosanase, and lysozymes from goose (GEWL), phage (T4L) and hen (HEWL) all hydrolyse related polysaccharides. The proteins share no significant amino-acid similarities, but have a structurally invariant core consisting of two helices and a three-stranded beta-sheet which form the substrate-binding and catalytic cleft. These enzymes represent a superfamily of hydrolases which are likely to have arisen by divergent evolution. Based on structural criteria, we divide the hydrolase superfamily into a bacterial family (chitosanase and T4L) and a eucaryotic family represented by chitinase and GEWL. Both families contain the core but have differing N- and C-terminal domains. Inclusion of chitinase and chitosanase in the superfamily suggests the archetypal catalytic mechanism of the group is an inverting mechanism. The retaining mechanism of HEWL is unusual.

X-ray structure of an anti-fungal chitosanase from streptomyces N174.

We report the 2.4 A X-ray crystal structure of a protein with chitosan endo-hydrolase activity isolated from Streptomyces N174. The structure was solved using phases acquired by SIRAS from a two-site methyl mercury derivative combined with solvent flattening and non-crystallographic two-fold symmetry averaging, and refined to an R-factor of 18.5%. The mostly alpha-helical fold reveals a structural core shared with several classes of lysozyme and barley endochitinase, in spite of a lack of shared sequence. Based on this structural similarity we postulate a putative active site, mechanism of action and mode of substrate recognition. It appears that Glu 22 acts as an acid and Asp 40 serves as a general base to activate a water molecule for an SN2 attack on the glycosidic bond. A series of amino-acid side chains and backbone carbonyl groups may bind the polycationic chitosan substrate in a deep electronegative binding cleft.

A combined algorithm for genome-wide prediction of protein function.

The availability of over 20 fully sequenced genomes has driven the development of new methods to find protein function and interactions. Here we group proteins by correlated evolution, correlated messenger RNA expression patterns and patterns of domain fusion to determine functional relationships among the 6,217 proteins of the yeast Saccharomyces cerevisiae. Using these methods, we discover over 93,000 pairwise links between functionally related yeast proteins. Links between characterized and uncharacterized proteins allow a general function to be assigned to more than half of the 2,557 previously uncharacterized yeast proteins. Examples of functional links are given for a protein family of previously unknown function, a protein whose human homologues are implicated in colon cancer and the yeast prion Sup35.

Characterization of a thermostable DNA glycosylase specific for U/G and T/G mismatches from the hyperthermophilic archaeon Pyrobaculum aerophilum.

U/G and T/G mismatches commonly occur due to spontaneous deamination of cytosine and 5-methylcytosine in double-stranded DNA. This mutagenic effect is particularly strong for extreme thermophiles, since the spontaneous deamination reaction is much enhanced at high temperature. Previously, a U/G and T/G mismatch-specific glycosylase (Mth-MIG) was found on a cryptic plasmid of the archaeon Methanobacterium thermoautotrophicum, a thermophile with an optimal growth temperature of 65 degrees C. We report characterization of a putative DNA glycosylase from the hyperthermophilic archaeon Pyrobaculum aerophilum, whose optimal growth temperature is 100 degrees C. The open reading frame was first identified through a genome sequencing project in our laboratory. The predicted product of 230 amino acids shares significant sequence homology to [4Fe-4S]-containing Nth/MutY DNA glycosylases. The histidine-tagged recombinant protein was expressed in Escherichia coli and purified. It is thermostable and displays DNA glycosylase activities specific to U/G and T/G mismatches with an uncoupled AP lyase activity. It also processes U/7,8-dihydro-oxoguanine and T/7,8-dihydro-oxoguanine mismatches. We designate it Pa-MIG. Using sequence comparisons among complete bacterial and archaeal genomes, we have uncovered a putative MIG protein from another hyperthermophilic archaeon, Aeropyrum pernix. The unique conserved amino acid motifs of MIG proteins are proposed to distinguish MIG proteins from the closely related Nth/MutY DNA glycosylases.

DIP: the database of interacting proteins.

The Database of Interacting Proteins (DIP; http://dip.doe-mbi.ucla.edu) is a database that documents experimentally determined protein-protein interactions. This database is intended to provide the scientific community with a comprehensive and integrated tool for browsing and efficiently extracting information about protein interactions and interaction networks in biological processes. Beyond cataloging details of protein-protein interactions, the DIP is useful for understanding protein function and protein-protein relationships, studying the properties of networks of interacting proteins, benchmarking predictions of protein-protein interactions, and studying the evolution of protein-protein interactions.