A comprehensive two-hybrid analysis to explore the yeast protein interactome

Protein–protein interactions play crucial roles in the execution of various biological functions. Accordingly, their comprehensive description would contribute considerably to the functional interpretation of fully sequenced genomes, which are flooded with novel genes of unpredictable functions. We previously developed a system to examine two-hybrid interactions in all possible combinations between the ≈6,000 proteins of the budding yeast Saccharomyces cerevisiae. Here we have completed the comprehensive analysis using this system to identify 4,549 two-hybrid interactions among 3,278 proteins. Unexpectedly, these data do not largely overlap with those obtained by the other project [Uetz, P., et al. (2000) Nature (London) 403, 623–627] and hence have substantially expanded our knowledge on the protein interaction space or interactome of the yeast. Cumulative connection of these binary interactions generates a single huge network linking the vast majority of the proteins. Bioinformatics-aided selection of biologically relevant interactions highlights various intriguing subnetworks. They include, for instance, the one that had successfully foreseen the involvement of a novel protein in spindle pole body function as well as the one that may uncover a hitherto unidentified multiprotein complex potentially participating in the process of vesicular transport. Our data would thus significantly expand and improve the protein interaction map for the exploration of genome functions that eventually leads to thorough understanding of the cell as a molecular system.

Structure of melanoma inhibitory activity protein, a member of a recently identified family of secreted proteins

Melanoma inhibitory activity (MIA) is a 12-kDa protein that is secreted from both chondrocytes and malignant melanoma cells. MIA has been reported to have effects on cell growth and adhesion, and it may play a role in melanoma metastasis and cartilage development. We report the 1.4-Å crystal structure of human MIA, which consists of an Src homology 3 (SH3)-like domain with N- and C-terminal extensions of about 20 aa each. The N- and C-terminal extensions add additional structural elements to the SH3 domain, forming a previously undescribed fold. MIA is a representative of a recently identified family of proteins and is the first structure of a secreted protein with an SH3 subdomain. The structure also suggests a likely protein interaction site and suggests that, unlike conventional SH3 domains, MIA does not recognize polyproline helices.

High-Molecular-Weight FK506-Binding Proteins Are Components of Heat-Shock Protein 90 Heterocomplexes in Wheat Germ Lysate1

In animal cell lysates the multiprotein heat-shock protein 90 (hsp90)-based chaperone complexes consist of hsp70, hsp40, and p60. These complexes act to convert steroid hormone receptors to their steroid-binding state by assembling them into heterocomplexes with hsp90, p23, and one of several immunophilins. Wheat germ lysate also contains a hsp90-based chaperone system that can assemble the glucocorticoid receptor into a functional heterocomplex with hsp90. However, only two components of the heterocomplex-assembly system, hsp90 and hsp70, have thus far been identified. Recently, purified mammalian p23 preadsorbed with JJ3 antibody-protein A-Sepharose pellets was used to isolate a mammalian p23-wheat hsp90 heterocomplex from wheat germ lysate (J.K. Owens-Grillo, L.F. Stancato, K. Hoffmann, W.B. Pratt, and P. Krishna [1996] Biochemistry 35: 15249–15255). This heterocomplex was found to contain an immunophilin(s) of the FK506-binding class, as judged by binding of the radiolabeled immunosuppressant drug [3H]FK506 to the immune pellets in a specific manner. In the present study we identified the immunophilin components of this heterocomplex as FKBP73 and FKBP77, the two recently described high-molecular-weight FKBPs of wheat. In addition, we present evidence that the two FKBPs bind hsp90 via tetratricopeptide repeat domains. Our results demonstrate that binding of immunophilins to hsp90 via tetratricopeptide repeat domains is a conserved protein interaction in plants. Conservation of this protein-to-protein interaction in both plant and animal cells suggests that it is important for the biological action of the high-molecular-weight immunophilins.

The human formin-binding protein 17 (FBP17) interacts with sorting nexin, SNX2, and is an MLL-fusion partner in acute myelogeneous leukemia

We have cloned a fusion partner of the MLL gene at 11q23 and identified it as the gene encoding the human formin-binding protein 17, FBP17. It maps to chromosome 9q34 centromeric to ABL. The gene fusion results from a complex chromosome rearrangement that was resolved by fluorescence in situ hybridization with various probes on chromosomes 9 and 11 as an ins(11;9)(q23;q34)inv(11)(q13q23). The rearrangement resulted in a 5′-MLL/FBP17-3′ fusion mRNA. We retrovirally transduced murine-myeloid progenitor cells with MLL/FBP17 to test its transforming ability. In contrast to MLL/ENL, MLL/ELL and other MLL-fusion genes, MLL/FBP17 did not give a positive readout in a serial replating assay. Therefore, we assume that additional cooperating genetic abnormalities might be needed to establish a full malignant phenotype. FBP17 consists of a C-terminal Src homology 3 domain and an N-terminal region that is homologous to the cell division cycle protein, cdc15, a regulator of the actin cytoskeleton in Schizosaccharomyces pombe. Both domains are separated by a consensus Rho-binding motif that has been identified in different Rho-interaction partners such as Rhotekin and Rhophilin. We evaluated whether FBP17 and members of the Rho family interact in vivo with a yeast two-hybrid assay. None of the various Rho proteins tested, however, interacted with FBP17. We screened a human kidney library and identified a sorting nexin, SNX2, as a protein interaction partner of FBP17. These data provide a link between the epidermal growth factor receptor pathway and an MLL fusion protein.

Dual-function regulators: the cAMP receptor protein and the CytR regulator can act either to repress or to activate transcription depending on the context.

Studies of gene regulation have revealed that several transcriptional regulators can switch between activator and repressor depending upon both the promoter and the cellular context. A relatively simple prokaryotic example is illustrated by the Escherichia coli CytR regulon. In this system, the cAMP receptor protein (CRP) assists the binding of RNA polymerase as well as a specific negative regulator, CytR. Thus, CRP functions either as an activator or as a corepressor. Here we show that, depending on promoter architecture, the CRP/CytR nucleoprotein complex has opposite effects on transcription. When acting from a site close to the DNA target for RNA polymerase, CytR interacts with CRP to repress transcription, whereas an interaction with CRP from appropriately positioned upstream binding sites can result in formation of a huge preinitiation complex and transcriptional activation. Based on recent results about CRP-mediated regulation of transcription initiation and the finding that CRP possesses discrete surface-exposed patches for protein-protein interaction with RNA polymerase and CytR, a molecular model for this dual regulation is discussed.

SPINDLY, a tetratricopeptide repeat protein involved in gibberellin signal transduction in Arabidopsis.

Gibberellins (GAs) are a major class of plant hormones that control many developmental processes, including seed development and germination, flower and fruit development, and flowering time. Genetic studies with Arabidopsis thaliana have identified two genes involved in GA perception or signal transduction. A semidominant mutation at the GIBBERELLIN INSENSITIVE (GAI) locus results in plants resembling GA-deficient mutants but exhibiting reduced sensitivity to GA. Recessive mutations at the SPINDLY (SPY) locus cause a phenotype that is consistent with constitutive activation of GA signal transduction. Here we show that a strong allele of spy is completely epistatic to gai, indicating that SPY acts downstream of GAI. We have cloned the SPY gene and shown that it encodes a new type of signal transduction protein, which contains a tetratricopeptide repeat region, likely serving as a protein interaction domain, and a novel C-terminal region. Mutations in both domains increase GA signal transduction. The presence of a similar gene in Caenorhabditis elegans suggests that SPY represents a class of signal transduction proteins that is present throughout the eukaryotes.

Clusters of charged residues in protein three-dimensional structures.

Statistically significant charge clusters (basic, acidic, or of mixed charge) in tertiary protein structures are identified by new methods from a large representative collection of protein structures. About 10% of protein structures show at least one charge cluster, mostly of mixed type involving about equally anionic and cationic residues. Positive charge clusters are very rare. Negative (or histidine-acidic) charge clusters often coordinate calcium, or magnesium or zinc ions [e.g., thermolysin (PDB code: 3tln), mannose-binding protein (2msb), aminopeptidase (1amp)]. Mixed-charge clusters are prominent at interchain contacts where they stabilize quaternary protein formation [e.g., glutathione S-transferase (2gst), catalase (8act), and fructose-1,6-bisphosphate aldolase (1fba)]. They are also involved in protein-protein interaction and in substrate binding. For example, the mixed-charge cluster of aspartate carbamoyl-transferase (8atc) envelops the aspartate carbonyl substrate in a flexible manner (alternating tense and relaxed states) where charge associations can vary from weak to strong. Other proteins with charge clusters include the P450 cytochrome family (BM-3, Terp, Cam), several flavocytochromes, neuraminidase, hemagglutinin, the photosynthetic reaction center, and annexin. In each case in Table 2 we discuss the possible role of the charge clusters with respect to protein structure and function.

Isoform-specific interaction of the alpha1A subunits of brain Ca2+ channels with the presynaptic proteins syntaxin and SNAP-25.

Presynaptic Ca2+ channels are crucial elements in neuronal excitation-secretion coupling. In addition to mediating Ca2+ entry to initiate transmitter release, they are thought to interact directly with proteins of the synaptic vesicle docking/fusion machinery. Here we report isoform-specific, stoichiometric interaction of the BI and rbA isoforms of the alpha1A subunit of P/Q-type Ca2+ channels with the presynaptic membrane proteins syntaxin and SNAP-25 in vitro and in rat brain membranes. The BI isoform binds to both proteins, while only interaction with SNAP-25 can be detected in vitro for the rbA isoform. The synaptic protein interaction ("synprint") site involves two adjacent segments of the intracellular loop connecting domains II and III between amino acid residues 722 and 1036 of the BI sequence. This interaction is competitively blocked by the corresponding region of the N-type Ca2+ channel, indicating that these two channels bind to overlapping regions of syntaxin and SNAP-25. Our results provide a molecular basis for a physical link between Ca2+ influx into nerve terminals and subsequent exocytosis of neurotransmitters at synapses that have presynaptic Ca2+ channels containing alpha1A subunits.

Receptor-G protein coupling is established by a potential conformational switch in the beta gamma complex.

Receptor-G protein interaction is characterized by cycles of association and dissociation. We present evidence which indicates that during receptor-G protein interaction, the C-terminal tail of the G protein gamma subunit, which is masked in the beta gamma complex, is exposed and establishes high-affinity contact with the receptor. This potential conformational switch provides a mechanism to regulate receptor-G protein coupling. This switch may also be significant for the role of the beta gamma complex in regulation of effector function.

The interferon-inducible double-stranded RNA-activated protein kinase self-associates in vitro and in vivo.

The interferon-inducible double-stranded (ds) RNA-activated protein kinase (PKR) exhibits antiviral, anticellular, and antitumor activities. The mechanisms of its enzymatic activation by autophosphorylation and of the observed transdominant inhibitory phenotype of enzymatically inactive mutants have invoked PKR dimerization. Here we present direct evidence in support of PKR-PKR interaction. We show that radiolabeled PKR can specifically interact with matrix-bound unlabeled PKR in the absence of dsRNA. The self-association activity resides, in part, in the N-terminal region of 170 residues, which also constitutes the dsRNA-binding domain (DRBD). DRBD can bind to matrix-bound PKR or to matrix-bound DRBD. Dimerization of DRBD was directly demonstrated by chemical crosslinking. Affinity chromatography and electrophoretic mobility supershift assays demonstrated that mutants that fail to bind dsRNA can still exhibit protein-protein interaction. The PKR-PKR interaction could also be observed in a two-hybrid transcriptional activation assay in mammalian cells and consequently is likely to be an important feature of PKR activity in vivo.

Interactions between the retinoid X receptor and a conserved region of the TATA-binding protein mediate hormone-dependent transactivation.

The retinoid X receptor (RXR) participates in a wide array of hormonal signaling pathways, either as a homodimer or as a heterodimer, with other members of the steroid and thyroid hormone receptor superfamily. In this report the ligand-dependent transactivation function of RXR has been characterized, and the ability of RXR to interact with components of the basal transcription machinery has been examined. In vivo and in vitro experiments indicate the RXR ligand-binding domain makes a direct, specific, and ligand-dependent contact with a highly conserved region of the TATA-binding protein. The ability of mutations that reduce ligand-dependent transcription by RXR to disrupt the RXR-TATA-binding protein interaction in vivo and in vitro suggests that RXR makes direct contact with the basal transcription machinery to achieve activation.

Mapping sites of interaction of p47-phox and flavocytochrome b with random-sequence peptide phage display libraries.

During assembly of the phagocyte NADPH oxidase, cytosolic p47-phox translocates to the plasma membrane and binds to flavocytochrome b, and binding domains for p47-phox have been identified on the C-terminal tails of both flavocytochrome b subunits. In the present report, we further examine the interaction of these two oxidase components by using random-sequence peptide phage display library analysis. Screening p47-phox with the peptide libraries identified five potential sites of interaction with flavocytochrome b, including three previously reported regions of interaction and two additional regions of interaction of p47-phox with gp91-phox and p22-phox. The additional sites were mapped to a domain on the first predicted cytosolic loop of gp91-phox encompassing residues S86TRVRRQL93 and to a domain near the cytosolic C-terminal tail of gp91-phox encompassing residues F450EWFADLL457. The mapping also confirmed a previously reported binding domain on gp91-phox (E554SGPRGVHFIF564) and putative Src homology 3 domain binding sites on p22-phox (P156PRPP160 and G177GPPGGP183). To demonstrate that the additional regions identified were biologically significant, peptides mimicking the gp91-phox sequences F77LRGSSACCSTRVRRQL93 and E451WFADLLQLLESQ463 were synthesized and assayed for their ability to inhibit NADPH oxidase activity. These peptides had EC50 values of 1 microM and 230 microM, respectively, and inhibited activation when added prior to assembly but did not affect activity of the preassembled oxidase. Our data demonstrate the usefulness of phage display library analysis for the identification of biologically relevant sites of protein-protein interaction and show that the binding of p47-phox to flavocytochrome b involves multiple binding sites along the C-terminal tails of both gp91- and p22-phox and other regions of gp91-phox nearer to the N terminus.

Functional recycling of C2 domains throughout evolution: A comparative study of synaptotagmin, protein kinase C and phospholipase C by sequence, structural and modelling approaches

The C2 domain is one of the most frequent and widely distributed calcium-binding motifs. Its structure comprises an eight-stranded beta-sandwich with two structural types as if the result of a circular permutation. Combining sequence, structural and modelling information, we have explored, at different levels of granularity, the functional characteristics of several families of C2 domains. At the coarsest level,the similarity correlates with key structural determinants of the C2 domain fold and, at the finest level, with the domain architecture of the proteins containing them, highlighting the functional diversity between the various subfamilies. The functional diversity appears as different conserved surface patches throughout this common fold. In some cases, these patches are related to substrate-binding sites whereas in others they correspond to interfaces of presumably permanent interaction between other domains within the same polypeptide chain. For those related to substrate-binding sites, the predictions overlap with biochemical data in addition to providing some novel observations. For those acting as protein-protein interfaces' our modelling analysis suggests that slight variations between families are a result of not only complementary adaptations in the interfaces involved but also different domain architecture. In the light of the sequence and structural genomic projects, the work presented here shows that modelling approaches along with careful sub-typing of protein families will be a powerful combination for a broader coverage in proteomics. (C) 2003 Elsevier Ltd. All rights reserved.

Gene duplication and hierarchical modularity in intracellular interaction networks

Networks of interactions evolve in many different domains. They tend to have topological characteristics in common, possibly due to common factors in the way the networks grow and develop. It has been recently suggested that one such common characteristic is the presence of a hierarchically modular organization. In this paper, we describe a new algorithm for the detection and quantification of hierarchical modularity, and demonstrate that the yeast protein-protein interaction network does have a hierarchically modular organization. We further show that such organization is evident in artificial networks produced by computational evolution using a gene duplication operator, but not in those developing via preferential attachment of new nodes to highly connected existing nodes. (C) 2004 Elsevier Ireland Ltd. All rights reserved.

Three distinct molecular surfaces in ephrin-A5 are essential for a functional interaction with EphA3

Eph receptor tyrosine kinases (Ephs) function as molecular relays that interact with cell surface-bound ephrin ligands to direct the position of migrating cells. Structural studies revealed that, through two distinct contact surfaces on opposite sites of each protein, Eph and ephrin binding domains assemble into symmetric, circular heterotetramers. However, Eph signal initiation requires the assembly of higher order oligomers, suggesting additional points of contact. By screening a random library of EphA3 binding-compromised ephrin-A5 mutants, we have now determined ephrin-A5 residues that are essential for the assembly of high affinity EphA3 signaling complexes. In addition to the two interfaces predicted from the crystal structure of the homologous EphB2 center dot ephrin-B2 complex, we identified a cluster of 10 residues on the ephrin-A5 E alpha-helix, the E-F loop, the underlying H beta-strand, as well as the nearby B - C loop, which define a distinct third surface required for oligomerization and activation of EphA3 signaling. Together with a corresponding third surface region identified recently outside of the minimal ephrin binding domain of EphA3, our findings provide experimental evidence for the essential contribution of three distinct protein-interaction interfaces to assemble functional EphA3 signaling complexes.