The OncoPPi Portal: an integrative resource to explore and prioritize protein-protein interactions for cancer target discovery.

Motivation: As cancer genomics initiatives move toward comprehensive identification of genetic alterations in cancer, attention is now turning to understanding how interactions among these genes lead to the acquisition of tumor hallmarks. Emerging pharmacological and clinical data suggest a highly promising role of cancer-specific protein-protein interactions (PPIs) as druggable cancer targets. However, large-scale experimental identification of cancer-related PPIs remains challenging, and currently available resources to explore oncogenic PPI networks are limited. Results: Recently, we have developed a PPI high-throughput screening platform to detect PPIs between cancer-associated proteins in the context of cancer cells. Here, we present the OncoPPi Portal, an interactive web resource that allows investigators to access, manipulate and interpret a high-quality cancer-focused network of PPIs experimentally detected in cancer cell lines. To facilitate prioritization of PPIs for further biological studies, this resource combines network connectivity analysis, mutual exclusivity analysis of genomic alterations, cellular co-localization of interacting proteins and domain-domain interactions. Estimates of PPI essentiality allow users to evaluate the functional impact of PPI disruption on cancer cell proliferation. Furthermore, connecting the OncoPPi network with the approved drugs and compounds in clinical trials enables discovery of new tumor dependencies to inform strategies to interrogate undruggable targets like tumor suppressors. The OncoPPi Portal serves as a resource for the cancer research community to facilitate discovery of cancer targets and therapeutic development. Availability and implementation: The OncoPPi Portal is available at http://oncoppi.emory.edu. Contact: andrey.ivanov@emory.edu or hfu@emory.edu.

AKT1, LKB1, and YAP1 Revealed as MYC Interactors with NanoLuc-Based Protein-Fragment Complementation Assay.

The c-Myc (MYC) transcription factor is a major cancer driver and a well-validated therapeutic target. However, directly targeting MYC has been challenging. Thus, identifying proteins that interact with and regulate MYC may provide alternative strategies to inhibit its oncogenic activity. In this study, we report the development of a NanoLuc-based protein-fragment complementation assay (NanoPCA) and mapping of the MYC protein interaction hub in live mammalian cells. The NanoPCA system was configured to enable detection of protein-protein interactions (PPI) at the endogenous level, as shown with PRAS40 dimerization, and detection of weak interactions, such as PINCH1-NCK2. Importantly, NanoPCA allows the study of PPI dynamics with reversible interactions. To demonstrate its utility for large-scale PPI detection in mammalian intracellular environment, we have used NanoPCA to examine MYC interaction with 83 cancer-associated proteins in live cancer cell lines. Our new MYC PPI data confirmed known MYC-interacting proteins, such as MAX, GSK3A, and SMARCA4, and revealed a panel of novel MYC interaction partners, such as RAC-α serine/threonine-protein kinase (AKT)1, liver kinase B (LKB)1, and Yes-associated protein (YAP)1. The MYC interactions with AKT1, LKB1, and YAP1 were confirmed by coimmunoprecipitation of endogenous proteins. Importantly, AKT1, LKB1, and YAP1 were able to activate MYC in a transcriptional reporter assay. Thus, these vital growth control proteins may represent promising MYC regulators, suggesting new mechanisms that couple energetic and metabolic pathways and developmental signaling to MYC-regulated cellular programs.

Application of the bridgehead fragments for the design of conformationally restricted melatonin analogues.

Conformationally constrained analogues of the hormone melatonin with a side chain incorporated into the bicyclic bridgehead core were synthesized based on the homology modeling and molecular docking studies performed for the MT(2) melatonin receptor. The methoxy-indole derivative fused with exo-N-acetamino-substituted bicyclo[2.2.2]octane was found to possess nanomolar MT(2) receptor affinity.

Enhanced potency of nucleotide-dendrimer conjugates as agonists of the P2Y14 receptor: multivalent effect in G protein-coupled receptor recognition.

The P2Y(14) receptor is a G protein-coupled receptor activated by uridine-5'-diphosphoglucose and other nucleotide sugars that modulates immune function. Covalent conjugation of P2Y(14) receptor agonists to PAMAM (polyamidoamine) dendrimers enhanced pharmacological activity. Uridine-5'-diphosphoglucuronic acid (UDPGA) and its ethylenediamine adduct were suitable functionalized congeners for coupling to several generations (G2.5-6) of dendrimers (both terminal carboxy and amino). Prosthetic groups, including biotin for avidin complexation, a chelating group for metal complexation (and eventual magnetic resonance imaging), and a fluorescent moiety, also were attached with the eventual goals of molecular detection and characterization of the P2Y(14) receptor. The activities of conjugates were assayed in HEK293 cells stably expressing the human P2Y(14) receptor. A G3 PAMAM conjugate containing 20 bound nucleotide moieties (UDPGA) was 100-fold more potent (EC(50) 2.4 nM) than the native agonist uridine-5'-diphosphoglucose. A molecular model of this conjugate docked in the human P2Y(14) receptor showed that the nucleotide-substituted branches could extend far beyond the dimensions of the receptor and be available for multivalent docking to receptor aggregates. Larger dendrimer carriers and greater loading favored higher potency. A similar conjugate of G6 with 147 out of 256 amino groups substituted with UDPGA displayed an EC(50) value of 0.8 nM. Thus, biological activity was either retained or dramatically enhanced in the multivalent dendrimer conjugates in comparison with monomeric P2Y(14) receptor agonists, depending on size, degree of substitution, terminal functionality, and attached prosthetic groups.

Molecular recognition in the P2Y(14) receptor: Probing the structurally permissive terminal sugar moiety of uridine-5'-diphosphoglucose.

The P2Y(14) receptor, a nucleotide signaling protein, is activated by uridine-5'-diphosphoglucose 1 and other uracil nucleotides. We have determined that the glucose moiety of 1 is the most structurally permissive region for designing analogues of this P2Y(14) agonist. For example, the carboxylate group of uridine-5'-diphosphoglucuronic acid proved to be suitable for flexible substitution by chain extension through an amide linkage. Functionalized congeners containing terminal 2-acylaminoethylamides prepared by this strategy retained P2Y(14) activity, and molecular modeling predicted close proximity of this chain to the second extracellular loop of the receptor. In addition, replacement of glucose with other sugars did not diminish P2Y(14) potency. For example, the [5'']ribose derivative had an EC(50) of 0.24muM. Selective monofluorination of the glucose moiety indicated a role for the 2''- and 6''-hydroxyl groups of 1 in receptor recognition. The beta-glucoside was twofold less potent than the native alpha-isomer, but methylene replacement of the 1''-oxygen abolished activity. Replacement of the ribose ring system with cyclopentyl or rigid bicyclo[3.1.0]hexane groups abolished activity. Uridine-5'-diphosphoglucose also activates the P2Y(2) receptor, but the 2-thio analogue and several of the potent modified-glucose analogues were P2Y(14)-selective.

Medicinal chemistry of the A3 adenosine receptor: agonists, antagonists, and receptor engineering.

A(3) adenosine receptor (A(3)AR) ligands have been modified to optimize their interaction with the A(3)AR. Most of these modifications have been made to the N(6) and C2 positions of adenine as well as the ribose moiety, and using a combination of these substitutions leads to the most efficacious, selective, and potent ligands. A(3)AR agonists such as IB-MECA and Cl-IB-MECA are now advancing into Phase II clinical trials for treatments targeting diseases such as cancer, arthritis, and psoriasis. Also, a wide number of compounds exerting high potency and selectivity in antagonizing the human (h)A(3)AR have been discovered. These molecules are generally characterized by a notable structural diversity, taking into account that aromatic nitrogen-containing monocyclic (thiazoles and thiadiazoles), bicyclic (isoquinoline, quinozalines, (aza)adenines), tricyclic systems (pyrazoloquinolines, triazoloquinoxalines, pyrazolotriazolopyrimidines, triazolopurines, tricyclic xanthines) and nucleoside derivatives have been identified as potent and selective A(3)AR antagonists. Probably due to the "enigmatic" physiological role of A(3)AR, whose activation may produce opposite effects (for example, concerning tissue protection in inflammatory and cancer cells) and may produce effects that are species dependent, only a few molecules have reached preclinical investigation. Indeed, the most advanced A(3)AR antagonists remain in preclinical testing. Among the antagonists described above, compound OT-7999 is expected to enter clinical trials for the treatment of glaucoma, while several thiazole derivatives are in development as antiallergic, antiasthmatic and/or antiinflammatory drugs.

Probing distal regions of the A2B adenosine receptor by quantitative structure-activity relationship modeling of known and novel agonists.

The binding modes at the A 2B adenosine receptor (AR) of 72 derivatives of adenosine and its 5'- N-methyluronamide with diverse substitutions at the 2 and N (6) positions were studied using a molecular modeling approach. The compounds in their receptor-docked conformations were used to build CoMFA and CoMSIA quantitative structure-activity relationship models. Various parameters, including different types of atomic charges, were examined. The best statistical parameters were obtained with a joint CoMFA and CoMSIA model: R (2) = 0.960, Q (2) = 0.676, SEE = 0.175, F = 158, and R (2) test = 0.782 for an independent test set containing 18 compounds. On the basis of the modeling results, four novel adenosine analogues, having elongated or bulky substitutions at N (6) position and/or 2 position, were synthesized and evaluated biologically. All of the proposed compounds were potent, full agonists at the A 2B AR in adenylate cyclase studies. Thus, in support of the modeling, bulky substitutions at both positions did not prevent A 2B AR activation, which predicts separate regions for docking of these moieties.

Molecular modeling of a PAMAM-CGS21680 dendrimer bound to an A2A adenosine receptor homodimer.

The theoretical possibility of bivalent binding of a dendrimer, covalently appended with multiple copies of a small ligand, to a homodimer of a G protein-coupled receptor was investigated with a molecular modeling approach. A molecular model was constructed of a third generation (G3) poly(amidoamine) (PAMAM) dendrimer condensed with multiple copies of the potent A(2A) adenosine receptor agonist CGS21680. The dendrimer was bound to an A(2A) adenosine receptor homodimer. Two units of the nucleoside CGS21680 could occupy the A(2A) receptor homodimer simultaneously. The binding mode of CGS21680 moieties linked to the PAMAM dendrimer and docked to the A(2A) receptor was found to be similar to the binding mode of a monomeric CGS21680 ligand.

Design, synthesis, and bioactivity of putative tubulin ligands with adamantane core.

Several adamantane-based taxol mimetics were synthesized and found to be cytotoxic at micromolar concentrations and to cause tubulin aggregation. The extent of the aggregation is maximal for N-benzoyl-(2R,3S)-phenylisoseryloxyadamantane (5) and is very sensitive to the structural modifications. A hybrid compound (15), combining adamantane-based taxol mimetic with colchicine was synthesized and found to possess both microtubule depolymerizing and microtubule bundling activities in A549 human lung carcinoma cells.

In vivo screening identifies GATAD2B as a metastasis driver in KRAS-driven lung cancer.

Genetic aberrations driving pro-oncogenic and pro-metastatic activity remain an elusive target in the quest of precision oncology. To identify such drivers, we use an animal model of KRAS-mutant lung adenocarcinoma to perform an in vivo functional screen of 217 genetic aberrations selected from lung cancer genomics datasets. We identify 28 genes whose expression promoted tumor metastasis to the lung in mice. We employ two tools for examining the KRAS-dependence of genes identified from our screen: 1) a human lung cell model containing a regulatable mutant KRAS allele and 2) a lentiviral system permitting co-expression of DNA-barcoded cDNAs with Cre recombinase to activate a mutant KRAS allele in the lungs of mice. Mechanistic evaluation of one gene, GATAD2B, illuminates its role as a dual activity gene, promoting both pro-tumorigenic and pro-metastatic activities in KRAS-mutant lung cancer through interaction with c-MYC and hyperactivation of the c-MYC pathway.

Structure-activity relationship of uridine 5'-diphosphoglucose analogues as agonists of the human P2Y14 receptor.

UDP-glucose (UDPG) and derivatives are naturally occurring agonists of the Gi protein-coupled P2Y14 receptor, which occurs in the immune system. We synthesized and characterized pharmacologically novel analogues of UDPG modified on the nucleobase, ribose, and glucose moieties, as the basis for designing novel ligands in conjunction with modeling. The recombinant human P2Y14 receptor expressed in COS-7 cells was coupled to phospholipase C through an engineered Galpha-q/i protein. Most modifications of the uracil or ribose moieties abolished activity; this is among the least permissive P2Y receptors. However, a 2-thiouracil modification in 15 (EC50 49 +/- 2 nM) enhanced the potency of UDPG (but not UDP-glucuronic acid) by 7-fold. 4-Thio analogue 13 was equipotent to UDPG, but S-alkylation was detrimental. Compound 15 was docked in a rhodposin-based receptor homology model, which correctly predicted potent agonism of UDP-fructose, UDP-mannose, and UDP-inositol. The hexose moiety of UDPG interacts with multiple H-bonding and charged residues and provides a fertile region for agonist modification.

Structure-activity relationships of 2,N(6),5'-substituted adenosine derivatives with potent activity at the A2B adenosine receptor.

2, N6, and 5'-substituted adenosine derivatives were synthesized via alkylation of 2-oxypurine nucleosides leading to 2-arylalkylether derivatives. 2-(3-(Indolyl)ethyloxy)adenosine 17 was examined in both binding and cAMP assays and found to be a potent agonist of the human A2BAR. Simplification, altered connectivity, and mimicking of the indole ring of 17 failed to maintain A2BAR potency. Introduction of N6-ethyl or N6-guanidino substitution, shown to favor A2BAR potency, failed to enhance potency in the 2-(3-(indolyl)ethyloxy)adenosine series. Indole 5' '- or 6' '-halo substitution was favored at the A2BAR, but a 5'-N-ethylcarboxyamide did not further enhance potency. 2-(3' '-(6' '-Bromoindolyl)ethyloxy)adenosine 28 displayed an A2BAR EC50 value of 128 nM, that is, more potent than the parent 17 (299 nM) and similar to 5'-N-ethylcarboxamidoadenosine (140 nM). Compound 28 was a full agonist at A2B and A2AARs and a low efficacy partial agonist at A1 and A3ARs. Thus, we have identified and optimized 2-(2-arylethyl)oxo moieties in AR agonists that enhance A2BAR potency and selectivity.

Molecular modeling of the human P2Y2 receptor and design of a selective agonist, 2'-amino-2'-deoxy-2-thiouridine 5'-triphosphate.

A rhodopsin-based homology model of the nucleotide-activated human P2Y2 receptor, including loops, termini, and phospholipids, was optimized with the Monte Carlo multiple minimum conformational search routine. Docked uridine 5'-triphosphate (UTP) formed a nucleobase pi-pi complex with conserved Phe3.32. Selectivity-enhancing 2'-amino-2'-deoxy substitution interacted through pi-hydrogen-bonding with aromatic Phe6.51 and Tyr3.33. A "sequential ligand composition" approach for docking the flexible dinucleotide agonist Up4U demonstrated a shift of conserved cationic Arg3.29 from the UTP gamma position to the delta position of Up4U and Up4 ribose. Synthesized nucleotides were tested as agonists at human P2Y receptors expressed in 1321N1 astrocytoma cells. 2'-Amino and 2-thio modifications were synergized to enhance potency and selectivity; compound 8 (EC50 = 8 nM) was 300-fold P2Y2-selective versus P2Y4. 2'-Amine acetylation reduced potency, and trifluoroacetylation produced intermediate potency. 5-Amino nucleobase substitution did not enhance P2Y2 potency through a predicted hydrophilic interaction possibly because of destabilization of the receptor-favored Northern conformation of ribose. This detailed view of P2Y2 receptor recognition suggests mutations for model validation.

Computer aided comparative analysis of the binding modes of the adenosine receptor agonists for all known subtypes of adenosine receptors.

Molecular models of all known subtypes (A1, A2A, A2B, and A3) of the human adenosine receptors were built in homology with bovine rhodopsin. These models include the transmembrane domain as well as all extracellular and intracellular hydrophilic loops and terminal domains. The molecular docking of adenosine and 46 selected derivatives was performed for each receptor subtype. A binding mode common for all studied agonists was proposed, and possible explanations for differences in the ligand activities were suggested.

MEDICI: Mining Essentiality Data to Identify Critical Interactions for Cancer Drug Target Discovery and Development.

Protein-protein interactions (PPIs) mediate the transmission and regulation of oncogenic signals that are essential to cellular proliferation and survival, and thus represent potential targets for anti-cancer therapeutic discovery. Despite their significance, there is no method to experimentally disrupt and interrogate the essentiality of individual endogenous PPIs. The ability to computationally predict or infer PPI essentiality would help prioritize PPIs for drug discovery and help advance understanding of cancer biology. Here we introduce a computational method (MEDICI) to predict PPI essentiality by combining gene knockdown studies with network models of protein interaction pathways in an analytic framework. Our method uses network topology to model how gene silencing can disrupt PPIs, relating the unknown essentialities of individual PPIs to experimentally observed protein essentialities. This model is then deconvolved to recover the unknown essentialities of individual PPIs. We demonstrate the validity of our approach via prediction of sensitivities to compounds based on PPI essentiality and differences in essentiality based on genetic mutations. We further show that lung cancer patients have improved overall survival when specific PPIs are no longer present, suggesting that these PPIs may be potentially new targets for therapeutic development. Software is freely available at https://github.com/cooperlab/MEDICI. Datasets are available at https://ctd2.nci.nih.gov/dataPortal.

The OncoPPi network of cancer-focused protein-protein interactions to inform biological insights and therapeutic strategies.

As genomics advances reveal the cancer gene landscape, a daunting task is to understand how these genes contribute to dysregulated oncogenic pathways. Integration of cancer genes into networks offers opportunities to reveal protein-protein interactions (PPIs) with functional and therapeutic significance. Here, we report the generation of a cancer-focused PPI network, termed OncoPPi, and identification of >260 cancer-associated PPIs not in other large-scale interactomes. PPI hubs reveal new regulatory mechanisms for cancer genes like MYC, STK11, RASSF1 and CDK4. As example, the NSD3 (WHSC1L1)-MYC interaction suggests a new mechanism for NSD3/BRD4 chromatin complex regulation of MYC-driven tumours. Association of undruggable tumour suppressors with drug targets informs therapeutic options. Based on OncoPPi-derived STK11-CDK4 connectivity, we observe enhanced sensitivity of STK11-silenced lung cancer cells to the FDA-approved CDK4 inhibitor palbociclib. OncoPPi is a focused PPI resource that links cancer genes into a signalling network for discovery of PPI targets and network-implicated tumour vulnerabilities for therapeutic interrogation.

Structure activity and molecular modeling analyses of ribose- and base-modified uridine 5'-triphosphate analogues at the human P2Y2 and P2Y4 receptors.

With the long-term goal of developing receptor subtype-selective high affinity agonists for the uracil nucleotide-activated P2Y receptors we have carried out a series of structure activity and molecular modeling studies of the human P2Y2 and P2Y4 receptors. UTP analogues with substitutions in the 2'-position of the ribose moiety retained capacity to activate both P2Y2 and P2Y4 receptors. Certain of these analogues were equieffective for activation of both receptors whereas 2'-amino-2'-deoxy-UTP exhibited higher potency for the P2Y2 receptor and 2'-azido-UTP exhibited higher potency for the P2Y4 receptor. 4-Thio substitution of the uracil base resulted in a UTP analogue with increased potency relative to UTP for activation of both the P2Y2 and P2Y4 receptors. In contrast, 2-thio substitution and halo- or alkyl substitution in the 5-position of the uracil base resulted in molecules that were 3-30-fold more potent at the P2Y2 receptor than P2Y4 receptor. 6-Aza-UTP was a P2Y2 receptor agonist that exhibited no activity at the P2Y4 receptor. Stereoisomers of UTPalphaS and 2'-deoxy-UTPalphaS were more potent at the P2Y2 than P2Y4 receptor, and the R-configuration was favored at both receptors. Molecular docking studies revealed that the binding mode of UTP is similar for both the P2Y2 and P2Y4 receptor binding pockets with the most prominent dissimilarities of the two receptors located in the second transmembrane domain (V90 in the P2Y2 receptor and I92 in the P2Y4 receptor) and the second extracellular loop (T182 in the P2Y2 receptor and L184 in the P2Y4 receptor). In summary, this work reveals substitutions in UTP that differentially affect agonist activity at P2Y2 versus P2Y4 receptors and in combination with molecular modeling studies should lead to chemical synthesis of new receptor subtype-selective drugs.

Agonists and antagonists for P2 receptors.

Recent work has identified nucleotide agonists selective for P2Y1, P2Y2 and P2Y6 receptors and nucleotide antagonists selective for P2Y1, P2Y12 and P2X1 receptors. Selective non-nucleotide antagonists have been reported for P2Y1, P2Y2, P2Y6, P2Y12, P2Y13, P2X(2/3)/P2X3 and P2X7 receptors. For example, the dinucleotide INS 37217 (Up4dC) potently activates the P2Y2 receptor, and the non-nucleotide antagonist A-317491 is selective for P2X(2/3)/P2X3 receptors. Nucleotide analogues in which the ribose moiety is substituted by a variety of novel ring systems, including conformationally locked moieties, have been synthesized as ligands for P2Y receptors. The focus on conformational factors of the ribose-like moiety allows the inclusion of general modifications that lead to enhanced potency and selectivity. At P2Y1,2,4,11 receptors, there is a preference for the North conformation as indicated with (N)-methanocarba analogues. The P2Y1 antagonist MRS2500 inhibited ADP-induced human platelet aggregation with an IC50 of 0.95 nM. MRS2365, an (N)-methanocarba analogue of 2-MeSADP, displayed potency (EC50) of 0.4nM at the P2Y1 receptor, with >10000-fold selectivity in comparison to P2Y12 and P2Y13 receptors. At P2Y6 receptors there is a dramatic preference for the South conformation. Three-dimensional structures of P2Y receptors have been deduced from structure activity relationships (SAR), mutagenesis and modelling studies. Detailed three-dimensional structures of P2X receptors have not yet been proposed.

Enabling systematic interrogation of protein-protein interactions in live cells with a versatile ultra-high-throughput biosensor platform.

Large-scale genomics studies have generated vast resources for in-depth understanding of vital biological and pathological processes. A rising challenge is to leverage such enormous information to rapidly decipher the intricate protein-protein interactions (PPIs) for functional characterization and therapeutic interventions. While a number of powerful technologies have been employed to detect PPIs, a singular PPI biosensor platform with both high sensitivity and robustness in a mammalian cell environment remains to be established. Here we describe the development and integration of a highly sensitive NanoLuc luciferase-based bioluminescence resonance energy transfer technology, termed BRET(n), which enables ultra-high-throughput (uHTS) PPI detection in live cells with streamlined co-expression of biosensors in a miniaturized format. We further demonstrate the application of BRET(n) in uHTS format in chemical biology research, including the discovery of chemical probes that disrupt PRAS40 dimerization and pathway connectivity profiling among core members of the Hippo signaling pathway. Such hippo pathway profiling not only confirmed previously reported PPIs, but also revealed two novel interactions, suggesting new mechanisms for regulation of Hippo signaling. Our BRET(n) biosensor platform with uHTS capability is expected to accelerate systematic PPI network mapping and PPI modulator-based drug discovery.