A PopZ‐linked apical recruitment assay for studying protein–protein interactions in the bacterial cell envelope

Most bacteria are surrounded by a complex cell envelope. As with many biological processes, studies of envelope assembly have benefited from cell‐based assays for detecting protein–protein interactions. These assays use simple readouts and lack a protein purification requirement, making them ideal for early stage investigations. The most widely used two‐hybrid interaction assay for proteins involved in envelope biogenesis is based on the reconstitution of adenylate cyclase activity from a split enzyme. Because adenylate cyclase is only functional in the cytoplasm, both protein fusions used in the assay must have a terminus located in this compartment. However, many envelope assembly factors are wholly extracytoplasmic. Detecting interactions involving such proteins using two‐hybrid systems has therefore been problematic. To address this issue, we developed a cytological assay in Escherichia coli based on PopZ from Caulobacter crescentus. Here, we demonstrate the utility of this PopZ‐Linked Apical Recruitment (POLAR) method for detecting interactions between proteins located in different cellular compartments. Additionally, we report that recruitment of an active peptidoglycan synthase to the cell pole is detrimental for E. coli and that interactions between proteins in the inner and outer membranes of the Gram‐negative envelope may provide a mechanism for recruiting protein complexes to subpolar sites.     

Drug–drug plasma protein binding interactions of ivacaftor

Ivacaftor is a novel cystic fibrosis (CF) transmembrane conductance regulator (CFTR) potentiator that improves the pulmonary function for patients with CF bearing a G551D CFTR‐protein mutation. Because ivacaftor is highly bound (>97%) to plasma proteins, there is the strong possibility that co‐administered CF drugs may compete for the same plasma protein binding sites and impact the free drug concentration. This, in turn, could lead to drastic changes in the in vivo efficacy of ivacaftor and therapeutic outcomes. This biochemical study compares the binding affinity of ivacaftor and co‐administered CF drugs for human serum albumin (HSA) and α₁‐acid glycoprotein (AGP) using surface plasmon resonance and fluorimetric binding assays that measure the displacement of site‐selective probes. Because of their ability to strongly compete for the ivacaftor binding sites on HSA and AGP, drug–drug interactions between ivacaftor are to be expected with ducosate, montelukast, ibuprofen, dicloxacillin, omeprazole, and loratadine. The significance of these plasma protein drug–drug interactions is also interpreted in terms of molecular docking simulations. This in vitro study provides valuable insights into the plasma protein drug–drug interactions of ivacaftor with co‐administered CF drugs. The data may prove useful in future clinical trials for a staggered treatment that aims to maximize the effective free drug concentration and clinical efficacy of ivacaftor. Copyright © 2015 John Wiley & Sons, Ltd.     

Split Nano luciferase complementation for probing protein‐protein interactions in plant cells

Deciphering protein‐protein interactions (PPIs) is fundamental for understanding signal transduction pathways in plants. The split firefly luciferase (Fluc) complementation (SLC) assay has been widely used for analyzing PPIs. However, concern has risen about the bulky halves of Fluc interfering with the functions of their fusion partners. Nano luciferase (Nluc) is the smallest substitute for Fluc with improved stability and luminescence. Here, we developed a dual‐use system enabling the detection of PPIs through the Nluc‐based SLC and co‐immunoprecipitation assays. This was realized by coexpression of two proteins under investigation in fusion with the HA‐ or FLAG‐tagged Nluc halves, respectively. We validated the robustness of this system by reproducing multiple previously documented PPIs in protoplasts or Agrobacterium‐transformed plants. We next applied this system to evaluate the homodimerization of Arabidopsis CERK1, a coreceptor of fungal elicitor chitin, and its heterodimerization with other homologs in the absence or presence of chitin. Moreover, split fragments of Nluc were fused to two cytosolic ends of Arabidopsis calcium channels CNGC2 and CNGC4 to help sense the allosteric change induced by the bacterial elicitor flg22. Collectively, these results demonstrate the usefulness of the Nluc‐based SLC assay for probing constitutive or inducible PPIs and protein allostery in plant cells.     

LuTHy: a double‐readout bioluminescence‐based two‐hybrid technology for quantitative mapping of protein–protein interactions in mammalian cells

Information on protein–protein interactions (PPIs) is of critical importance for studying complex biological systems and developing therapeutic strategies. Here, we present a double‐readout bioluminescence‐based two‐hybrid technology, termed LuTHy, which provides two quantitative scores in one experimental procedure when testing binary interactions. PPIs are first monitored in cells by quantification of bioluminescence resonance energy transfer (BRET) and, following cell lysis, are again quantitatively assessed by luminescence‐based co‐precipitation (LuC). The double‐readout procedure detects interactions with higher sensitivity than traditional single‐readout methods and is broadly applicable, for example, for detecting the effects of small molecules or disease‐causing mutations on PPIs. Applying LuTHy in a focused screen, we identified 42 interactions for the presynaptic chaperone CSPα, causative to adult‐onset neuronal ceroid lipofuscinosis (ANCL), a progressive neurodegenerative disease. Nearly 50% of PPIs were found to be affected when studying the effect of the disease‐causing missense mutations L115R and ?L116 in CSPα with LuTHy. Our study presents a robust, sensitive research tool with high utility for investigating the molecular mechanisms by which disease‐associated mutations impair protein activity in biological systems.     

Feature selection and classification of protein–protein complexes based on their binding affinities using machine learning approaches

Protein–protein interactions are intrinsic to virtually every cellular process. Predicting the binding affinity of protein–protein complexes is one of the challenging problems in computational and molecular biology. In this work, we related sequence features of protein–protein complexes with their binding affinities using machine learning approaches. We set up a database of 185 protein–protein complexes for which the interacting pairs are heterodimers and their experimental binding affinities are available. On the other hand, we have developed a set of 610 features from the sequences of protein complexes and utilized Ranker search method, which is the combination of Attribute evaluator and Ranker method for selecting specific features. We have analyzed several machine learning algorithms to discriminate protein‐protein complexes into high and low affinity groups based on their K_d values. Our results showed a 10‐fold cross‐validation accuracy of 76.1% with the combination of nine features using support vector machines. Further, we observed accuracy of 83.3% on an independent test set of 30 complexes. We suggest that our method would serve as an effective tool for identifying the interacting partners in protein–protein interaction networks and human–pathogen interactions based on the strength of interactions. Proteins 2014; 82:2088–2096. © 2014 Wiley Periodicals, Inc.     

Endogenous biotin‐binding proteins: an overlooked factor causing false positives in streptavidin‐based protein detection

Biotinylation is widely used in DNA, RNA and protein probing assays as this molecule has generally no impact on the biological activity of its substrate. During the streptavidin‐based detection of glycoproteins in Lactobacillus rhamnosus GG with biotinylated lectin probes, a strong positive band of approximately 125 kDa was observed, present in different cellular fractions. This potential glycoprotein reacted heavily with concanavalin A (ConA), a lectin that specifically binds glucose and mannose residues. Surprisingly, this protein of 125 kDa could not be purified using a ConA affinity column. Edman degradation of the protein, isolated via cation and anion exchange chromatography, lead to the identification of the band as pyruvate carboxylase, an enzyme of 125 kDa that binds biotin as a cofactor. Detection using only the streptavidin conjugate resulted in more false positive signals of proteins, also in extracellular fractions, indicating biotin‐associated proteins. Indeed, biotin is a known cofactor of numerous carboxylases. The potential occurence of false positive bands with biotinylated protein probes should thus be considered when using streptavidin‐based detection, e.g. by developing a blot using only the streptavidin conjugate. To circumvent these false positives, alternative approaches like detection based on digoxigenin labelling can also be used.     

Fluorescent protein recruitment assay for demonstration and analysis of in vivo protein interactions in plant cells and its application to Tobacco mosaic virus movement protein

We describe a simple fluorescent protein‐based method to investigate interactions with a viral movement protein in living cells that relies on the in vivo re‐localization of proteins in the presence of their interaction partners. We apply this method in combination with fluorescence lifetime imaging microscopy (FLIM) to demonstrate that a domain of the Tobacco mosaic virus (TMV) movement protein (MP) previously predicted to mediate protein:protein interactions is dispensable for these contacts. We suggest that this method can be generalized for analysis of other protein interactions in planta.     

Iron–sulfur protein maturation in Helicobacter pylori: identifying a Nfu‐type cluster carrier protein and its iron–sulfur protein targets

Helicobacter pylori is anomalous among non nitrogen‐fixing bacteria in containing an incomplete NIF system for Fe–S cluster assembly comprising two essential proteins, NifS (cysteine desulfurase) and NifU (scaffold protein). Although nifU deletion strains cannot be obtained via the conventional gene replacement, a NifU‐depleted strain was constructed and shown to be more sensitive to oxidative stress compared to wild‐type (WT) strains. The hp1492 gene, encoding a putative Nfu‐type Fe–S cluster carrier protein, was disrupted in three different H. pylori strains, indicating that it is not essential. However, Δnfu strains have growth deficiency, are more sensitive to oxidative stress and are unable to colonize mouse stomachs. Moreover, Δnfu strains have lower aconitase activity but higher hydrogenase activity than the WT. Recombinant Nfu was found to bind either one [2Fe–2S] or [4Fe–4S] cluster/dimer, based on analytical, UV–visible absorption/CD and resonance Raman studies. A bacterial two‐hybrid system was used to ascertain interactions between Nfu, NifS, NifU and each of 36 putative Fe–S‐containing target proteins. Nfu, NifS and NifU were found to interact with 15, 6 and 29 putative Fe–S proteins respectively. The results indicate that Nfu, NifS and NifU play a major role in the biosynthesis and/or delivery of Fe–S clusters in H. pylori.     

Application of quantitative immunoprecipitation combined with knockdown and cross‐linking to Chlamydomonas reveals the presence of vesicle‐inducing protein in plastids 1 in a common complex with chloroplast HSP90C

Knowledge of the interaction partners of a protein of interest may provide important information on its function. Common to currently available tools for the identification of protein–protein interactions, however, is their high rates of false positives. Only recently an assay was reported that allowed for the unequivocal identification of protein–protein interactions in mammalian cells in a single experiment. This assay, termed quantitative immunoprecipitation combined with knockdown (QUICK), combines RNAi, stable isotope labeling with amino acids in cell culture, immunoprecipitation, and quantitative MS. We are using the unicellular green alga Chlamydomonas reinhardtii to understand the roles of chaperones in chloroplast biogenesis. The goal of this work was to apply QUICK to Chlamydomonas for the identification of novel interaction partners of vesicle‐inducing protein in plastids 1 (VIPP1), a protein required for the biosynthesis/maintenance of thylakoid membranes and known substrate of chloroplast HSP70B. We report here a robust QUICK protocol for Chlamydomonas that has been improved (i) by introducing a cross‐linking step (‐X) to improve protein complex stability and (ii) by including a control for the correction of unequal immunoprecipitation and/or labeling efficiencies. Using QUICK and cross‐linking we could verify that HSP70B and CGE1 form a complex with VIPP1 and could also demonstrate that chloroplast HSP90C is part of this complex. Moreover, we could show that the chaperones interact with VIPP1 also in membrane fractions.     

Dynamic map of protein interactions in the Escherichia coli chemotaxis pathway

Protein–protein interactions play key roles in virtually all cellular processes, often forming complex regulatory networks. A powerful tool to study interactions in vivo is fluorescence resonance energy transfer (FRET), which is based on the distance‐dependent energy transfer from an excited donor to an acceptor fluorophore. Here, we used FRET to systematically map all protein interactions in the chemotaxis signaling pathway in Escherichia coli, one of the most studied models of signal transduction, and to determine stimulation‐induced changes in the pathway. Our FRET analysis identified 19 positive FRET pairs out of the 28 possible protein combinations, with 9 pairs being responsive to chemotactic stimulation. Six stimulation‐dependent and five stimulation‐independent interactions were direct, whereas other interactions were apparently mediated by scaffolding proteins. Characterization of stimulation‐induced responses revealed an additional regulation through activity dependence of interactions involving the adaptation enzyme CheB, and showed complex rearrangement of chemosensory receptors. Our study illustrates how FRET can be efficiently employed to study dynamic protein networks in vivo.     

Improving protein–protein interactions prediction accuracy using protein evolutionary information and relevance vector machine model

Predicting protein–protein interactions (PPIs) is a challenging task and essential to construct the protein interaction networks, which is important for facilitating our understanding of the mechanisms of biological systems. Although a number of high‐throughput technologies have been proposed to predict PPIs, there are unavoidable shortcomings, including high cost, time intensity, and inherently high false positive rates. For these reasons, many computational methods have been proposed for predicting PPIs. However, the problem is still far from being solved. In this article, we propose a novel computational method called RVM‐BiGP that combines the relevance vector machine (RVM) model and Bi‐gram Probabilities (BiGP) for PPIs detection from protein sequences. The major improvement includes (1) Protein sequences are represented using the Bi‐gram probabilities (BiGP) feature representation on a Position Specific Scoring Matrix (PSSM), in which the protein evolutionary information is contained; (2) For reducing the influence of noise, the Principal Component Analysis (PCA) method is used to reduce the dimension of BiGP vector; (3) The powerful and robust Relevance Vector Machine (RVM) algorithm is used for classification. Five‐fold cross‐validation experiments executed on yeast and Helicobacter pylori datasets, which achieved very high accuracies of 94.57 and 90.57%, respectively. Experimental results are significantly better than previous methods. To further evaluate the proposed method, we compare it with the state‐of‐the‐art support vector machine (SVM) classifier on the yeast dataset. The experimental results demonstrate that our RVM‐BiGP method is significantly better than the SVM‐based method. In addition, we achieved 97.15% accuracy on imbalance yeast dataset, which is higher than that of balance yeast dataset. The promising experimental results show the efficiency and robust of the proposed method, which can be an automatic decision support tool for future proteomics research. For facilitating extensive studies for future proteomics research, we developed a freely available web server called RVM‐BiGP‐PPIs in Hypertext Preprocessor (PHP) for predicting PPIs. The web server including source code and the datasets are available at http://219.219.62.123:8888/BiGP/ .     

Optimized distance‐dependent atom‐pair‐based potential DOOP for protein structure prediction

The DOcking decoy‐based Optimized Potential (DOOP) energy function for protein structure prediction is based on empirical distance‐dependent atom‐pair interactions. To optimize the atom‐pair interactions, native protein structures are decomposed into polypeptide chain segments that correspond to structural motives involving complete secondary structure elements. They constitute near native ligand–receptor systems (or just pairs). Thus, a total of 8609 ligand–receptor systems were prepared from 954 selected proteins. For each of these hypothetical ligand–receptor systems, 1000 evenly sampled docking decoys with 0–10 Å interface root‐mean‐square‐deviation (iRMSD) were generated with a method used before for protein–protein docking. A neural network‐based optimization method was applied to derive the optimized energy parameters using these decoys so that the energy function mimics the funnel‐like energy landscape for the interaction between these hypothetical ligand–receptor systems. Thus, our method hierarchically models the overall funnel‐like energy landscape of native protein structures. The resulting energy function was tested on several commonly used decoy sets for native protein structure recognition and compared with other statistical potentials. In combination with a torsion potential term which describes the local conformational preference, the atom‐pair‐based potential outperforms other reported statistical energy functions in correct ranking of native protein structures for a variety of decoy sets. This is especially the case for the most challenging ROSETTA decoy set, although it does not take into account side chain orientation‐dependence explicitly. The DOOP energy function for protein structure prediction, the underlying database of protein structures with hypothetical ligand–receptor systems and their decoys are freely available at http://agknapp.chemie.fu‐berlin.de/doop/ . Proteins 2015; 83:881–890. © 2015 Wiley Periodicals, Inc.     

LuMPIS – A modified luminescence‐based mammalian interactome mapping pull‐down assay for the investigation of protein–protein interactions encoded by GC‐low ORFs

The GC content is highly variable among the genomes of different organisms. It has been shown that recombinant gene expression in mammalian cells is much more efficient when GC‐rich coding sequences of a certain protein are used. In order to study protein–protein interactions in Varicella zoster virus, a GC‐low herpesvirus, we have developed a novel luminescence‐based maltose‐binding protein pull‐down interaction screening system (LuMPIS) that is able to overcome the impaired protein expression levels of GC‐low ORFs in mammalian expression systems.     

Non‐interacting surface solvation and dynamics in protein–protein interactions

Protein–protein interactions control a plethora of cellular processes, including cell proliferation, differentiation, apoptosis, and signal transduction. Understanding how and why proteins interact will inevitably lead to novel structure‐based drug design methods, as well as design of de novo binders with preferred interaction properties. At a structural and molecular level, interface and rim regions are not enough to fully account for the energetics of protein–protein binding, even for simple lock‐and‐key rigid binders. As we have recently shown, properties of the global surface might also play a role in protein–protein interactions. Here, we report on molecular dynamics simulations performed to understand solvent effects on protein–protein surfaces. We compare properties of the interface, rim, and non‐interacting surface regions for five different complexes and their free components. Interface and rim residues become, as expected, less mobile upon complexation. However, non‐interacting surface appears more flexible in the complex. Fluctuations of polar residues are always lower compared with charged ones, independent of the protein state. Further, stable water molecules are often observed around polar residues, in contrast to charged ones. Our analysis reveals that (a) upon complexation, the non‐interacting surface can have a direct entropic compensation for the lower interface and rim entropy and (b) the mobility of the first hydration layer, which is linked to the stability of the protein–protein complex, is influenced by the local chemical properties of the surface. These findings corroborate previous hypotheses on the role of the hydration layer in shielding protein–protein complexes from unintended protein–protein interactions. Proteins 2015; 83:445–458. © 2014 Wiley Periodicals, Inc.     

Improved sensitivity of firefly luminescent intermediate‐based protein interaction assay using Ser 440 mutant with lower adenylation activity

Protein–protein interaction assays are important in various fields including molecular biology, diagnostics, and drug screening. We recently designed a novel protein–protein interaction assay, the firefly luminescent intermediate‐based protein interaction assay (FlimPIA), that exploited the unique reaction mechanism of firefly luciferase (Fluc). Using two mutant Flucs, each impaired with one of the two half‐reactions, namely adenylation and subsequent oxidative luminescent steps, FlimPIA detects the proximity of the two proteins tethered to the mutant Flucs. Here, we found that introducing a mutation into a residue in the hinge region (S440) of the mutant with lowered adenylation activity (‘Acceptor’ Fluc) further improved the response of FlimPIA by lowering the residual adenylation activity. Mutants with bulkier residues showed greater inhibition, probably due to increased steric hindrance at the adenylation conformation. As a result, the FlimPIA with S440 L acceptor showed the best signal/background ratio for the detection of rapamycin‐induced FKBP12–FRB interactions.     

Scoring ligand similarity in structure‐based virtual screening

Scoring to identify high‐affinity compounds remains a challenge in virtual screening. On one hand, protein–ligand scoring focuses on weighting favorable and unfavorable interactions between the two molecules. Ligand‐based scoring, on the other hand, focuses on how well the shape and chemistry of each ligand candidate overlay on a three‐dimensional reference ligand. Our hypothesis is that a hybrid approach, using ligand‐based scoring to rank dockings selected by protein–ligand scoring, can ensure that high‐ranking molecules mimic the shape and chemistry of a known ligand while also complementing the binding site. Results from applying this approach to screen nearly 70 000 National Cancer Institute (NCI) compounds for thrombin inhibitors tend to support the hypothesis. EON ligand‐based ranking of docked molecules yielded the majority (4/5) of newly discovered, low to mid‐micromolar inhibitors from a panel of 27 assayed compounds, whereas ranking docked compounds by protein–ligand scoring alone resulted in one new inhibitor. Since the results depend on the choice of scoring function, an analysis of properties was performed on the top‐scoring docked compounds according to five different protein–ligand scoring functions, plus EON scoring using three different reference compounds. The results indicate that the choice of scoring function, even among scoring functions measuring the same types of interactions, can have an unexpectedly large effect on which compounds are chosen from screening. Furthermore, there was almost no overlap between the top‐scoring compounds from protein–ligand versus ligand‐based scoring, indicating the two approaches provide complementary information. Matchprint analysis, a new addition to the SLIDE (Screening Ligands by Induced‐fit Docking, Efficiently) screening toolset, facilitated comparison of docked molecules' interactions with those of known inhibitors. The majority of interactions conserved among top‐scoring compounds for a given scoring function, and from the different scoring functions, proved to be conserved interactions in known inhibitors. This was particularly true in the S1 pocket, which was occupied by all the docked compounds. Copyright © 2009 John Wiley & Sons, Ltd.     

Detection of membrane protein–protein interaction in planta based on dual‐intein‐coupled tripartite split‐GFP association

Despite the great interest in identifying protein–protein interactions (PPIs) in biological systems, only a few attempts have been made at large‐scale PPI screening in planta. Unlike biochemical assays, bimolecular fluorescence complementation allows visualization of transient and weak PPIs in vivo at subcellular resolution. However, when the non‐fluorescent fragments are highly expressed, spontaneous and irreversible self‐assembly of the split halves can easily generate false positives. The recently developed tripartite split‐GFP system was shown to be a reliable PPI reporter in mammalian and yeast cells. In this study, we adapted this methodology, in combination with the β‐estradiol‐inducible expression cassette, for the detection of membrane PPIs in planta. Using a transient expression assay by agroinfiltration of Nicotiana benthamiana leaves, we demonstrate the utility of the tripartite split‐GFP association in plant cells and affirm that the tripartite split‐GFP system yields no spurious background signal even with abundant fusion proteins readily accessible to the compartments of interaction. By validating a few of the Arabidopsis PPIs, including the membrane PPIs implicated in phosphate homeostasis, we proved the fidelity of this assay for detection of PPIs in various cellular compartments in planta. Moreover, the technique combining the tripartite split‐GFP association and dual‐intein‐mediated cleavage of polyprotein precursor is feasible in stably transformed Arabidopsis plants. Our results provide a proof‐of‐concept implementation of the tripartite split‐GFP system as a potential tool for membrane PPI screens in planta.     

Prediction of protein–protein interaction sites from weakly homologous template structures using meta‐threading and machine learning

The identification of protein–protein interactions is vital for understanding protein function, elucidating interaction mechanisms, and for practical applications in drug discovery. With the exponentially growing protein sequence data, fully automated computational methods that predict interactions between proteins are becoming essential components of system‐level function inference. A thorough analysis of protein complex structures demonstrated that binding site locations as well as the interfacial geometry are highly conserved across evolutionarily related proteins. Because the conformational space of protein–protein interactions is highly covered by experimental structures, sensitive protein threading techniques can be used to identify suitable templates for the accurate prediction of interfacial residues. Toward this goal, we developed eFindSite^PPI, an algorithm that uses the three‐dimensional structure of a target protein, evolutionarily remotely related templates and machine learning techniques to predict binding residues. Using crystal structures, the average sensitivity (specificity) of eFindSite^PPI in interfacial residue prediction is 0.46 (0.92). For weakly homologous protein models, these values only slightly decrease to 0.40–0.43 (0.91–0.92) demonstrating that eFindSite^PPI performs well not only using experimental data but also tolerates structural imperfections in computer‐generated structures. In addition, eFindSite^PPI detects specific molecular interactions at the interface; for instance, it correctly predicts approximately one half of hydrogen bonds and aromatic interactions, as well as one third of salt bridges and hydrophobic contacts. Comparative benchmarks against several dimer datasets show that eFindSite^PPI outperforms other methods for protein‐binding residue prediction. It also features a carefully tuned confidence estimation system, which is particularly useful in large‐scale applications using raw genomic data. eFindSite^PPI is freely available to the academic community at http://www.brylinski.org/efindsiteppi . Copyright © 2014 John Wiley & Sons, Ltd.