The binary protein-protein interaction landscape of Escherichia coli

Rajagopala, Seesandra V.; Sikorski, Patricia; Kumar, Ashwani; Mosca, Roberto; Vlasblom, James; Arnold, Roland; Franca-Koh, Jonathan; Pakala, Suman B.; Phanse, Sadhna; Céol, Arnaud; Häuser, Roman; Siszler, Gabriella; Wuchty, Stefan; Emili, Andrew; Babu, Mohan; Aloy, Patrick; Pieper, Rembert; Uetz, Peter

doi:10.1038/nbt.2831

Cited by 234 publications

(293 citation statements)

References 47 publications

Supporting

Mentioning

283

Contrasting

Order By: Relevance

“…Hence, mapping specific protein−protein interactions is central to a systems-level understanding of cells, and has broad applications to areas such as drug targeting. High-throughput experiments have recently elucidated a substantial fraction of protein−protein interactions in a few model organisms (1), but such experiments remain challenging. Meanwhile, there has been an explosion of available sequence data.…”

mentioning

confidence: 99%

Inferring interaction partners from protein sequences

Bitbol

Dwyer

Colwell

et al. 2016

Proc. Natl. Acad. Sci. U.S.A.

134

188

View full text Add to dashboard Cite

Specific protein−protein interactions are crucial in the cell, both to ensure the formation and stability of multiprotein complexes and to enable signal transduction in various pathways. Functional interactions between proteins result in coevolution between the interaction partners, causing their sequences to be correlated. Here we exploit these correlations to accurately identify, from sequence data alone, which proteins are specific interaction partners. Our general approach, which employs a pairwise maximum entropy model to infer couplings between residues, has been successfully used to predict the 3D structures of proteins from sequences. Thus inspired, we introduce an iterative algorithm to predict specific interaction partners from two protein families whose members are known to interact. We first assess the algorithm's performance on histidine kinases and response regulators from bacterial twocomponent signaling systems. We obtain a striking 0.93 true positive fraction on our complete dataset without any a priori knowledge of interaction partners, and we uncover the origin of this success. We then apply the algorithm to proteins from ATP-binding cassette (ABC) transporter complexes, and obtain accurate predictions in these systems as well. Finally, we present two metrics that accurately distinguish interacting protein families from noninteracting ones, using only sequence data. proteins. For instance, specific protein−protein interactions ensure proper signal transduction in various pathways. Hence, mapping specific protein−protein interactions is central to a systems-level understanding of cells, and has broad applications to areas such as drug targeting. High-throughput experiments have recently elucidated a substantial fraction of protein−protein interactions in a few model organisms (1), but such experiments remain challenging. Meanwhile, there has been an explosion of available sequence data. Can we exploit this abundant new sequence data to identify specific protein−protein interaction partners?Specific interactions between proteins imply evolutionary constraints on the interacting partners. For instance, mutation of a contact residue in one partner generally impairs binding, but may be compensated by a complementary mutation in the other partner. This coevolution of interaction partners results in correlations between their amino acid sequences. Similar correlations exist within single proteins, for example, between amino acids that are in contact in the folded protein. However, the simple fact of a correlation between residues in a multiple sequence alignment is only weakly predictive of a 3D contact (2-4), as correlation can also stem from indirect effects. Fortunately, global statistical models allow direct and indirect interactions to be disentangled (5-7). In particular, the maximum entropy principle (8) specifies the least-structured global statistical model consistent with the one-and two-point statistics of an alignment (5). This approach has recently been used with success to determine 3D ...

show abstract

mentioning

confidence: 99%

Inferring interaction partners from protein sequences

Bitbol

Dwyer

Colwell

et al. 2016

Proc. Natl. Acad. Sci. U.S.A.

134

188

View full text Add to dashboard Cite

show abstract

“…The subsequently solved structure of FliG (42) shows that this truncation would remove about half of the N-terminal domain of the protein, leaving a fragment that would be expected to fold poorly and show an increased propensity for nonspecific binding (owing to the exposure of about a dozen hydrophobic side chains). A more recent, wideranging screen of the E. coli protein interaction landscape by the yeast two-hybrid method did not give evidence of a FliG/H-NS interaction (43), instead identifying several candidate FliG partners without obvious connections to the flagellum (PabA, ProS, RpsS, YcbY, YcfM, and AscG). The same broad screen identified a few candidate H-NS partners, including its homolog StpA but not FliG.…”

Section: Discussionmentioning

confidence: 94%

Function of the Histone-Like Protein H-NS in Motility of Escherichia coli: Multiple Regulatory Roles Rather than Direct Action at the Flagellar Motor

Kim

Blair

2015

J Bacteriol

View full text Add to dashboard Cite

A number of investigations of Escherichia coli have suggested that the DNA-binding protein H-NS, in addition to its well-known functions in chromosome organization and gene regulation, interacts directly with the flagellar motor to modulate its function. Here, in a study initially aimed at characterizing the H-NS/motor interaction further, we identify problems and limitations in the previous work that substantially weaken the case for a direct H-NS/motor interaction. Null hns mutants are immotile, largely owing to the downregulation of the flagellar master regulators FlhD and FlhC. We, and others, previously reported that an hns mutant remains poorly motile even when FlhDC are expressed constitutively. In the present work, we use better-engineered strains to show that the motility defect in a ⌬hns, FlhDC-constitutive strain is milder than that reported previously and does not point to a direct action of H-NS at the motor. H-NS regulates numerous genes and might influence motility via a number of regulatory molecules besides FlhDC. To examine the sources of the motility defect that persists in an FlhDC-constitutive ⌬hns mutant, we measured transcript levels and overexpression effects of a number of genes in candidate regulatory pathways. The results indicate that H-NS influences motility via multiple regulatory linkages that include, minimally, the messenger molecule cyclic di-GMP, the biofilm regulatory protein CsgD, and the sigma factors S and F . The results are in accordance with the more standard view of H-NS as a regulator of gene expression rather than a direct modulator of flagellar motor performance. IMPORTANCEData from a number of previous studies have been taken to indicate that the nucleoid-organizing protein H-NS influences motility not only by its well-known DNA-based mechanisms but also by binding directly to the flagellar motor to alter function. In this study, H-NS is shown to influence motility through diverse regulatory pathways, but a direct interaction with the motor is not supported. Previous indications of a direct action at the motor appear to be related to the use of nonnull strains and, in some cases, a failure to effectively bypass the requirement for H-NS in the expression of the flagellar regulon. These findings call for a substantially revised interpretation of the literature concerning H-NS and flagellar motility and highlight the importance of H-NS in diverse regulatory processes involved in the motile-sessile transition. H-NS is a small DNA-binding protein that occurs widely in Gram-negative bacteria, where it functions to organize chromosomal DNA and regulate hundreds of genes (1-3). Binding of H-NS to the chromosome is not strictly sequence specific but exhibits a preference for certain motifs (4) and for intrinsically curved regions of DNA (5). While the logic underlying the very wide regulatory action of H-NS is not fully understood, generally speaking, it appears to modulate the expression of genes involved in sensing and responding to changes in the environment (6), ...

show abstract

“…Systematically generated binary maps uniformly identify PPIs in the whole gene space, avoiding sociological bias that may occur in small-scale experiments or literature-curated interactome maps that focus on well-studied genes [8 ]. This screening method has therefore proven to be a useful tool, enumerating binary interactions not only for human, but for a number of model organisms as well, including Saccharomyces cerevisiae [9][10][11], Schizosaccharomyces pombe [12], Escherichia coli [13], Caenorhabditis elegans [14,15], and Arabidopsis thaliana [16]. While this method is easily scaled and relatively inexpensive, it may fail to capture interactions between proteins which rely on intermediary or scaffold proteins (such as those between protein complex subunits), those involving proteins from specific subcellular compartments (such as membrane proteins), or those which require post-translational modifications [17].…”

Section: Binary Interaction Mapping By Yeast Two-hybrid (Y2h)mentioning

confidence: 99%