Interaction specificity is a required feature of biological networks and a necessary characteristic of protein or small-molecule reagents and therapeutics. The ability to alter or inhibit protein interactions selectively would advance basic and applied molecular science. Assessing or modelling interaction specificity requires treating multiple competing complexes, which presents computational and experimental challenges. Here we present a computational framework for designing protein interaction specificity and use it to identify specific peptide partners for human bZIP transcription factors. Protein microarrays were used to characterize designed, synthetic ligands for all but one of 20 bZIP families. The bZIP proteins share strong sequence and structural similarities and thus are challenging targets to bind specifically. Yet many of the designs, including examples that bind the oncoproteins cJun, cFos and cMaf, were selective for their targets over all 19 other families. Collectively, the designs exhibit a wide range of novel interaction profiles, demonstrating that human bZIPs have only sparsely sampled the possible interaction space accessible to them. Our computational method provides a way to systematically analyze tradeoffs between stability and specificity and is suitable for use with many types of structure-scoring functions; thus it may prove broadly useful as a tool for protein design.

The Tor1p and Tor2p kinases, targets of the therapeutically important antibiotic rapamycin, function as components of two distinct protein complexes in yeast, termed TOR complex 1 (TORC1) and TORC2. TORC1 is responsible for a wide range of rapamycin-sensitive cellular activities and contains, in addition to Tor1p or Tor2p, two highly conserved proteins, Lst8p and Kog1p. By identifying proteins that co-purify with Tor1p, Tor2p, Lst8p, and Kog1p, we have characterized a comprehensive set of protein-protein interactions that define further the composition of TORC1 as well as TORC2. In particular, we have identified Tco89p (YPL180w) and Bit61p (YJL058c) as novel components of TORC1 and TORC2, respectively. Deletion of TOR1 or TCO89 results in two specific and distinct phenotypes, (i) rapamycin-hypersensitivity and (ii) decreased cellular integrity, both of which correlate with the presence of SSD1-d, an allele of SSD1 previously associated with defects in cellular integrity. Furthermore, we link Ssd1p to Tap42p, a component of the TOR pathway that is believed to act uniquely downstream of TORC1. Together, these results define a novel connection between TORC1 and Ssd1p-mediated maintenance of cellular integrity.

Tor1p and Tor2p kinases, targets of the immune-suppressive antibiotic rapamycin, are components of a highly conserved signaling network that couples nutrient availability and cell growth. To gain insight into the molecular basis underlying Tor-dependent signaling, we used cell fractionation and immunoaffinity chromatography to examine the physical environment of Tor2p. We found that the majority of Tor2p associates with a membrane-bound compartment along with at least four other proteins, Avo1p-Avo3p and Lst8p. Using immunogold electron microscopy, we observed that Tor2p, as well as Tor1p, localizes in punctate clusters to regions adjacent to the plasma membrane and within the cell interior, often in association with characteristic membranous tracks. Cell fractionation, coimmunoprecipitation, and immunogold electron microscopy experiments confirmed that Lst8 associates with both Tor2p as well as Tor1p at these membranous sites. In contrast, we find that Kog1, the yeast homologue of the mammalian Tor regulatory protein Raptor, interacts preferentially with Tor1p. These findings provide evidence for the existence of Tor signaling complexes that contain distinct as well as overlapping components. That these complexes colocalize to a membrane-bound compartment suggests an intimate relationship between membrane-mediated signaling and Tor activity

The versatile coiled-coil protein motif is widely used to induce and control macromolecular interactions in biology and materials science. Yet the types of interaction patterns that can be constructed using known coiled coils are limited. Here we greatly expand the coiled-coil toolkit by measuring the complete pair-wise interactions of 48 synthetic coiled coils and 7 human bZIP coiled coils using peptide microarrays. The resulting 55-member protein ‘interactome’ includes 27 pairs of interacting peptides that preferentially hetero-associate. The 27 pairs can be used in combinations to assemble sets of 3 to 6 proteins that compose networks of varying topologies. Of special interest are heterospecific peptide pairs that participate in mutually orthogonal interactions. Such pairs provide the opportunity to dimerize two separate molecular systems without undesired crosstalk. Solution and structural characterization of two such sets of orthogonal heterodimers provide details of their interaction geometries. The orthogonal pair, along with the many other network motifs discovered in our screen, provide new capabilities for synthetic biology and other applications.

Differences in biomolecular sequence and function underlie dramatic ranges of appearance and behavior among species. We studied the basic region-leucine zipper (bZIP) transcription factors and quantified bZIP dimerization networks for 5 metazoan and 2 single-cell species, measuring interactions in vitro for 2,891 protein pairs. Metazoans have a higher proportion of heteromeric bZIP interactions and more network complexity than the single-cell species. The metazoan bZIP interactomes have broadly similar structures, but there has been extensive rewiring of connections compared to the last common ancestor, and each species network is highly distinct. Many metazoan bZIP orthologs and paralogs have strikingly different interaction specificities, and some differences arise from minor sequence changes. Our data show that a shifting landscape of biochemical functions related to signaling and gene expression contributes to species diversity.

No abstract

How transcription factor dimerization impacts DNA-binding specificity is poorly understood. Guided by protein dimerization properties, we examined DNA binding specificities of 270 human bZIP pairs. DNA interactomes of 80 heterodimers and 22 homodimers revealed that 72% of heterodimer motifs correspond to conjoined half-sites preferred by partnering monomers. Remarkably, the remaining motifs are composed of variably-spaced half-sites (12%) or ‘emergent’ sites (16%) that cannot be readily inferred from half-site preferences of partnering monomers. These binding sites were biochemically validated by EMSA-FRET analysis and validated in vivo by ChIP-seq data from human cell lines. Focusing on ATF3, we observed distinct cognate site preferences conferred by different bZIP partners, and demonstrated that genome-wide binding of ATF3 is best explained by considering many dimers in which it participates. Importantly, our compendium of bZIP-DNA interactomes predicted bZIP binding to 156 disease associated SNPs, of which only 20 were previously annotated with known bZIP motifs.DOI: http://dx.doi.org/10.7554/eLife.19272.001

Pathogens use a variety of secreted and surface proteins to interact with and manipulate their hosts, but a systematic approach for identifying such proteins has been lacking. To identify these ‘host-exposed' proteins, we used spatially restricted enzymatic tagging followed by mass spectrometry analysis of Caenorhabditis elegans infected with two species of Nematocida microsporidia. We identified 82 microsporidia proteins inside of intestinal cells, including several pathogen proteins in the nucleus. These microsporidia proteins are enriched in targeting signals, are rapidly evolving and belong to large Nematocida-specific gene families. We also find that large, species-specific families are common throughout microsporidia species. Our data suggest that the use of a large number of rapidly evolving species-specific proteins represents a common strategy for microsporidia to interact with their hosts. The unbiased method described here for identifying potential pathogen effectors represents a powerful approach to study a broad range of pathogens.