Class Of Protein Kinase Inhibitors Research Articles

Systematic generation and analysis of counterfactuals for compound activity predictions using multi-task models.

Most machine learning (ML) methods produce predictions that are hard or impossible to understand. The black box nature of predictive models obscures potential learning bias and makes it difficult to recognize and trace problems. Moreover, the inability to rationalize model decisions causes reluctance to accept predictions for experimental design. For ML, limited trust in predictions presents a substantial problem and continues to limit its impact in interdisciplinary research, including early-phase drug discovery. As a desirable remedy, approaches from explainable artificial intelligence (XAI) are increasingly applied to shed light on the ML black box and help to rationalize predictions. Among these is the concept of counterfactuals (CFs), which are best understood as test cases with small modifications yielding opposing prediction outcomes (such as different class labels in object classification). For ML applications in medicinal chemistry, for example, compound activity predictions, CFs are particularly intuitive because these hypothetical molecules enable immediate comparisons with actual test compounds that do not require expert ML knowledge and are accessible to practicing chemists. Such comparisons often reveal structural moieties in compounds that determine their predictions and can be further investigated. Herein, we adapt and extend a recently introduced concept for the systematic generation of molecular CFs to multi-task predictions of different classes of protein kinase inhibitors, analyze CFs in detail, rationalize the origins of CF formation in multi-task modeling, and present exemplary explanations of predictions.

The Binary Classification of Protein Kinases.

In an earlier publication a binary model for chronic diseases classification has been proposed. According to the model, chronic diseases were classified as “high Treg” or “low Treg” diseases, depending on whether the immune response is anti- or pro-inflammatory and assuming that regulatory T cells are major determinants of the response. It turned out that most cancers are “high Treg” diseases, while autoimmune diseases are “low Treg”. This paper proposes a molecular cause for this binary response. The mechanism proposed depends on the effect of protein kinases on the immune system. Thus, protein kinases are classified as anti- or pro-inflammatory kinases depending on whether they drive “high Treg” or “low Treg” diseases. Observations reported in the earlier publication can be described in terms of anti-inflammatory kinase (AIK) or pro-inflammatory kinase (PIK) activity. Analysis of literature data reveals that the two classes of kinases display distinctive properties relating to their interactions with pathogens and environmental factors. Pathogens that promote Treg activity (“high Treg” pathogens) activate AIKs, while pathogens that suppress Treg activity (“low Treg” pathogens) activate PIKs. Diseases driven by AIKs are associated with “high Treg” pathogens while those diseases driven by PIKs are associated with “low Treg” pathogens. By promoting the activity of AIKs, alcohol consumption increases the risk of “high Treg” cancers but decreases the risk of some “low Treg” autoimmune diseases. JAK1 gain-of-function mutations are observed at high frequencies in autoimmune diseases while JAK1 loss-of-function mutations are observed at high frequencies in cancers with high tumor-infiltrating Tregs. It should also be noted that the corresponding two classes of protein kinase inhibitors are mutually exclusive in terms of their approved therapeutic indications. There is no protein kinase inhibitor that is approved for the treatment of both autoimmune diseases and “high Treg” cancers. Although there are exceptions to the conclusions presented above, these conclusions are supported by the great bulk of published data. It therefore seems that the binary division of protein kinases is a useful tool for elucidating (at the molecular level) many distinctive properties of cancers and autoimmune diseases.

Targeting the Transcriptional Addiction of Leukemia Stem Cells By a New Class of Protein Kinase Inhibitors

Cancer stem cells present a major therapeutic challenge - their effective eradication depends on therapeutic targeting of the relative vulnerabilities of the cancer stem cells vs. normal tissue stem cells. One feature that distinguishes cancer initiating and propagating cells from their normal counterparts is transcriptional addiction, providing opportunities for novel therapeutic interventions with the aim of curing cancer.CKIα ablation appears as a promising means of activating p53 and killing leukemia cells in MDS and AML (Elyada et al , doi:10.1038/nature09673; Kronke et al, doi: 10.1038/nature14610); yet, with no selective CKIα inhibitors for in vivo use, the therapeutic value of CKIα inhibition cannot be validated. We succeeded in developing two different classes of CKIα inhibitors, one inducing CKIα degradation, and another targeting the catalytic pocket of CKIα, with no degradation. Both inhibitor types activate p53, yet we found that CKIα ablation or degradation was not sufficient to selectively eliminate leukemia stem cells, due to robust expression of anti-apoptotic oncogenes driven by super-enhancer activation.Through extensive medicinal chemistry, we obtained CKIα catalytic inhibitors with profound killing capacity of primary mouse AML cells and human AML cell lines. We demonstrated a strong therapeutic effect in a mouse model of MLL-AF9-induced leukemia, with >40% cure rates evident by transplantation of bone marrow from treated leukemic mice to lethally irradiated normal mice. All transplanted mice regained normal blood counts, exclusively derived from the donor bone marrow. None showed any evidence of residual disease for 6 months, demonstrating a successful therapeutic window, distinguishing leukemia stem cells (LSCs) from HSPCs. Parallel in vitro experiments, show that the inhibitors did not affect normal hematopoietic progenitors in a CFU assay, but eliminated AML progenitors at an IC50 <9nM.We found that CKIα inhibitors possessing potent anti-leukemia activity are distinguished by their capacity to co-target CDK9 and suppress the RNA Pol II elongation factor P-TEFb (CDK9-CyclinT1 complex), evident by blocking the phosphorylation of RNA Pol II C-terminal domain (pCTD2/5, see Figure). Co-crystallography studies elucidated the structural basis for co-targeting of CKIα and P-TEFb, the latter controls the rate limiting step of Poll-II-transcription and functions as a gatekeeper of super-enhancers. We found that both P-TEFb and super-enhancer activity are vigorously activated in AML cells. Therefore, the efficient P-TEFb-targeting activity of the inhibitors enables a selective blockade of leukemia super-enhancers (SE)-driven transcription, demonstrated by ChIP-seq analysis. As a result, SE-dependent transcription of major leukemia drivers including Myc and the anti-apoptotic oncogenes Bcl-2 and Mcl-1 was nearly abolished.Another remarkable feature of these dual CKIα-P-TEFb inhibitors is a rapid hit mechanism, by which, short drug exposure (10 mins in vitro and 2hrs in vivo) results in prolonged (24hrs) SE suppression and leukemia cell apoptosis. This unique property, which is at variance with the current occupancy-driven pharmacological paradigm, likely contributes to the dramatic therapeutic effect of co-targeting CKIα and P-TEFb in leukemia, whereby combination of rapid p53 induction with disruption of oncogenic super-enhancers eliminates leukemia stem cells without compromising normal HSPCs. The ability to disrupt super-enhancers and eliminate their associated oncogenes within minutes of drug exposure permits short intermittent therapy, which minimizes generalized toxicity. Indeed, at the therapeutic dose, we did not observe any significant adverse treatment effects even at daily treatment for 45 days in mice.To conclude, we developed a new class of small molecule inhibitors that co-targets CKIα and P-TEFb. These inhibitors have unique pharmacologic properties: short-term kinase inhibition results in long-term disruption of SE activity. Shutdown of leukemic super-enhancers in synergy with robust p53 activation compromises leukemic cells and stem cells addicted to SE-driven transcription. These features explain the powerful and specific anti-leukemic therapeutic effects of this new class of inhibitors in-vivo. [Display omitted] DisclosuresBen-Neriah:BioTheryX Inc.: Consultancy, Equity Ownership, Research Funding. Vacca:BioTheryX Inc.: Consultancy. Mercurio:BioTheryX: Employment, Equity Ownership.

An aptamer to the MAP kinase insert region.

Targeting functional, non-catalytic domains of protein kinases or other proteins is an emerging field in chemical biology research. Non-ATP competitive kinase inhibitors allow for the investigation of active-site independent functions, e.g., the biological role of protein-protein interactions. These inhibitors also serve as a starting point for developing novel therapeutic strategies. However, the identification of such inhibitors by means of conventional low molecular weight compounds represents a great challenge in modern drug discovery. Employing the mitogen-activated protein (MAP) kinase Erk2, we show that RNA aptamers have the capacity to be a novel, promising class of protein kinase inhibitors that can be applied to target individual subdomains and block domain specific functions without affecting kinase activity per se.

Abstract B246: NVP-BGJ398: A potent and selective inhibitor of the fibroblast growth factor receptor family.

Abstract The mammalian fibroblast growth factor/fibroblast growth factor receptor (FGF/FGFR) family comprises 22 FGF ligands, 18 of which exert their action through 4 highly conserved transmembrane tyrosine kinase receptors (FGFR1, FGFR2, FGFR3 and FGFR4). This highly complex signaling network regulates fundamental developmental pathways, physiological processes in the adult organism and a variety of cellular functions including proliferation, differentiation, and survival. Owing to such prominent roles, it is not surprising that different types of FGF/FGFR alterations, such as abnormal expression, mutations, and amplifications, have been found in diverse tumor types. A growing body of evidence linking deregulation of the FGF/FGFR system to the pathogenesis of cancer makes these receptors attractive targets for therapeutic intervention by a small molecular weight inhibitor approach. We have established a new and non-conventional strategy to mimic documented lead compounds which consists in the replacement of real rings by stable pseudo six-membered rings resulting from the formation of an intramolecular hydrogen bond in a planar conjugated system. More recently, we have reported an example of the application of such strategy to the design of a new class of protein kinase inhibitors featuring a N-pyrimidin-4-yl-urea motif as an alternative to the pyrido[2,3-d]pyrimidin-7-one core structure of a well-known class of protein kinase inhibitors. In this context, we have described a prototype compound displaying submicromolar activity against several tyrosine kinases including FGFR1. In this communication we illustrate the process, guided by the use of molecular modeling and biochemical and cellular assays, that led to the identification of NVP-BGJ398, a potent and selective FGFR antagonist currently in phase I clinical trial. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr B246.

ChemInform Abstract: Entry into a New Class of Protein Kinase Inhibitors by Pseudo Ring Design.

AbstractChemInform is a weekly Abstracting Service, delivering concise information at a glance that was extracted from about 200 leading journals. To access a ChemInform Abstract of an article which was published elsewhere, please select a “Full Text” option. The original article is trackable via the “References” option.

Entry into a new class of protein kinase inhibitors by pseudo ring design

A pyrimidin-4-yl-urea motif forming a pseudo ring by intramolecular hydrogen bonding has been designed to mimic the pyrido[2,3-d]pyrimidin-7-one core structure of a well-established class of protein kinase inhibitors. Potent inhibition of a number of protein kinases was obtained with the first prototype compound synthesized to probe the design concept.

Protection against glutamate toxicity through inhibition of the p44/42 mitogen-activated protein kinase pathway in neuronally differentiated P19 cells

Excessive levels of the neurotransmitter glutamate trigger excitotoxic processes in neurons that lead to cell death. N-Methyl- d-aspartate (NMDA) receptor over-activation is a key excitotoxic stimulus that leads to increases in intracellular calcium and activation of downstream signaling pathways, including the p44/42 mitogen-activated protein (MAP) kinase pathway. In the present study, we have demonstrated that 1,4-diamino-2,3-dicyano-1,4-bis[2-aminophenylthio]butadiene (U0126), a potent and selective inhibitor of the p44/42 MAP kinase signaling pathway, prevents glutamate-induced death in neuronally differentiated P19 cells. In addition, we show that differentiated, but not undifferentiated, P19 cells expressed zeta1, epsilon1, and epsilon2 subunits of the NMDA receptor. Differentiated P19 cells exhibited specific NMDA receptor binding and intracellular calcium responses to glutamate that were blocked by the selective NMDA receptor antagonist [5 R,10 S]-[+]-5-methyl-10,11-dihydro-5 H-dibenzo[a,d]cyclohepten-5,10-imine (MK-801), but not U0126. Glutamate treatment of differentiated P19 cells triggered a rapid and sustained induction in p42 MAP kinase phosphorylation that was blocked by U0126. Pretreatment of differentiated P19 cells with U0126, but not other classes of protein kinase inhibitors, protected against glutamate-induced cell death. Post-treatment with U0126, even as late as 6 hr after glutamate application, also protected against glutamate toxicity. These results suggest that the p44/42 MAP kinase pathway may be a critical downstream signaling pathway in glutamate receptor-activated toxicity.

Identification of novel classes of protein kinase inhibitors using combinatorial peptide chemistry based on functional genomics knowledge.

A discovery approach based on an intramolecular inhibitory mechanism was applied to a prototype calmodulin (CaM)-regulated protein kinase in order to demonstrate a proof-of-principle for the development of selective inhibitors. The overall approach used functional genomics analysis of myosin light chain kinase (MLCK) to identify short autoinhibitory sequences that lack CaM recognition activity, followed by recursive combinatorial peptide library production and comparative activity screens. Peptide 18 (Arg-Lys-Lys-Tyr-Lys-Tyr-Arg-Arg-Lys-NH2), one of several selective inhibitors discovered, has an IC50 = 50 nM for MLCK, inhibits CaM kinase II only at 4000-fold higher concentrations, and does not inhibit cyclic AMP-dependent protein kinase. Analogues of peptide 18 containing conformationally constrained cis-4-aminocyclohexanecarboxylic acid retained affinity and selectivity. The inhibitors add to the armamentarium available for the deconvolution of complex signal transduction pathways and their relationship to homeostasis and disease, and the approach is potentially applicable to enzymes in which the catalytic and regulatory domains are found within the same open reading frame of a cDNA.

Molecular basis for p38 protein kinase inhibitor specificity.

p38 is a member of the mitogen-activated protein (MAP) kinase family and is a critical enzyme in the proinflammatory cytokine pathway. Other MAP kinase group members that share both structural and functional homology to p38 include the c-Jun NH2-terminal kinases (JNKs or SAPKs) and the extracellular-regulated protein kinases (ERKs). In this study, we determined the molecular basis for p38alpha inhibitor specificity exhibited by five compounds in the diarylimidazole, triarylimidazole, and triarylpyrrole classes of protein kinase inhibitors. These compounds are significantly more potent inhibitors of p38 compared to the JNKs and ERKs. Three active site ATP-binding domain residues in p38, T106, M109, and A157, selected based on primary sequence alignment, molecular modeling, and X-ray crystal structure data, were mutated to assess their role in inhibitor binding and enzymatic catalysis. All mutants, with the exception of T106M, had kinase activity within 3-fold of wild-type p38. Mutation of T106 to glutamine, the residue present at the corresponding position in ERK-2, or methionine, the corresponding residue in p38gamma, p38delta, and the JNKs, rendered all five inhibitors ineffective. The diarylimidazoles had approximately a 6-fold decrease in potency toward M109A p38. For the mutant A157V, all diarylimidazoles and triarylimidazoles tested were 5-10-fold more potent compared with wild-type p38. In contrast, two triarylpyrroles were 15-40-fold less potent versus A157V p38. These results showed that the molecular basis for the specificity of the p38 inhibitors was attributed largely to threonine 106 in p38 and that methionine 109 contributes to increased binding affinity for imidazole based inhibitors.

The protein tyrosine kinase p56lck regulates the serine-base exchange activity in Jurkat T cells

Different classes of protein kinase inhibitors for protein kinase C, cAMP-dependent protein kinase or protein tyrosine kinases have been studied for their effect on phospholipid metabolism. The results show that among the compounds studied, only 4′-aminohydroxyflavone (AHF), previously described as a specific inhibitor of the protein tyrosine kinase p56lck, markedly increased phosphatidylserine synthesis in Jurkat T cells. The biosyntheses of phosphatidylcholine and phosphatidylethanolamine were not affected. Also, the synthesis of phospholipids from tritium-labeled fatty acid as precursor was left unchanged by the p56lck inhibitor. The decreased phosphatidylserine synthesis induced when triggering the CD3-TCR complex was impaired by AHF, suggesting that p56lck could be implicated in the regulation of the serine-base exchange enzyme system. Direct evidence for the participation of p56lck in the regulation of the serine-base exchange enzyme system was obtained by using p56lck-deficient Jurkat cells (J.CaM 1.6) in which the basal base exchange activity was markedly increased and on the other hand AHF had no effect. In addition, transfection of J.Cam 1.6 cells with p56lck-cDNA allowed recovery of the AHF activity.© 1997 Federation of European Biochemical Societies.

A Chemiluminescent Microtiter Plate Assay for Sensitive Detection of Protein Kinase Activity

A chemiluminescent protein kinase assay using biotinylated substrate peptides captured on a streptavidin-coated microtiter plate and monoclonal antibodies to detect their phosphorylation is described. Assay conditions were optimized and validated for sensitive measurement of protein kinase A, protein kinase C, Ca2+/calmodulin-dependent protein kinase II (CAMKII), receptor interacting protein, and src activities. The newly developed chemiluminescent assay has several advantages over currently used radioactive or colorimetric methods. It is highly sensitive at low enzyme and substrate concentrations and high, close to physiological ATP levels. It is fast, simple to perform and amenable to automation and high-throughput drug screening. The assay is also robust, exhibiting minimum interference from solvents and test substances from various sources. Overall, among the presently available methods for the detection of protein kinase activity, chemiluminescence was found to provide the highest sensitivity under conditions most closely mimicking the intracellular environment. This assay is expected to be useful in both academic and industrial laboratories, especially in identifying novel classes of protein kinase inhibitors.

Mechanism of Topoisomerase II Inhibition by Staurosporine and Other Protein Kinase Inhibitors

Topoisomerase II is an essential enzyme for proliferation of eukaryotic cells. It is also a target for many antineoplastic drugs that promote stabilization of covalent complexes between topoisomerase II and DNA. Topoisomerase II and protein kinases both catalyze the transfer of phosphoester bonds from nucleotides to proteins. This similarity suggests that inhibitors may affect both classes of enzymes. In the present study, we have examined the mechanism of topoisomerase II inhibition by three different classes of protein kinase inhibitors. We report that staurosporine inhibited the catalytic activity of topoisomerase II by blocking the transfer of phosphodiester bonds from DNA to the active tyrosine site, a mechanism of inhibition not previously reported for this enzyme. In contrast, other kinase inhibitors, such as methyl 2,5-dihydroxycinnamate, most likely inactivated topoisomerase II by alkylation of essential amino acids, whereas the mechanism of inhibition of bis-indolylmaleimide possibly involved a direct interaction with DNA.

Stimulation of Cell Elongation in Teleost Rod Photoreceptors by Distinct Protein Kinase Inhibitors

The rod photoreceptors of teleost retinas elongate in the light. To characterize the role of protein kinases in elongation, pharmacological studies were carried out with rod fragments consisting of the motile inner segment and photosensory outer segment (RIS-ROS). Isolated RIS-ROS were cultured in the presence of membrane-permeant inhibitors that exhibit selective activity toward specific serine/threonine protein kinases. We report that three distinct classes of protein kinase inhibitors stimulated elongation in darkness: (1) cyclic-AMP-dependent protein kinase (PKA)-selective inhibitors (H-89 and KT5720), (2) a protein kinase C (PKC)-selective inhibitor (GF 109203X) that affects most PKC isoforms, and (3) a kinase inhibitor (H-85) that does not affect PKC and PKA in vitro. Other kinase inhibitors tested neither stimulated elongation in darkness nor inhibited light-induced elongation; these include the myosin light chain kinase inhibitors ML-7 and ML-9, the calcium-calmodulin kinase II inhibitor KN-62, and inhibitors or activators of diacylglycerol-dependent PKCs (sphingosine, calphostin C, chelerythrine, and phorbol esters). The myosin light chain kinase inhibitors as well as the PKA and PKC inhibitors H-89 and GF 109203X all enhanced light-induced elongation. These observations suggest that light-induced RIS-ROS elongation is inhibited by both PKA and an unidentified kinase or kinases, possibly a diacylglycerol-independent form of PKC.

Protein kinase C inhibitor chelerythrine disrupts memory formation in chicks.

Chelerythrine (CHELE), a specific, potent protein kinase C (PKC) inhibitor, disrupts memory formation for a one-trial peck-avoidance task. Three predictions were made about how CHELE, injected into chick brain near the time of training, would affect memory formation, based on previous work with two classes of protein kinase inhibitors (M. R. Rosenzweig et al., 1992; P. A. Serrano et al., 1994) and the in vitro inhibition of PKC by CHELE: (a) CHELE, injected into the intermediate medial hyperstriatum ventrale, would significantly impair memory formation; (b) the amnestic dose would be approximately 10 nmol; (c) CHELE would not produce amnesia for about 45 min after training, but significantly impair memory by 60 min. Experimental tests confirmed each prediction. This study adds to evidence that PKC activity is part of a cascade of neurochemical events initiated by learning and that PKC activity shortly after training is necessary for long-term memory.