November 1, 2013 (Vol. 33, No. 19)

Vicki Glaser Writer GEN

Protein kinases, enzymes that phosphorylate proteins and other organic molecules, are increasingly important targets for drug discovery.

Inhibiting these enzymes can disrupt signal-transduction pathways that play a critical role in a variety of disorders, particularly cancer and inflammatory diseases.

At CHI’s recent “Next-Gen Kinase Inhibitors” conference, presenters discussed recent advances and ongoing challenges in targeting kinases for drug discovery. Examples of the innovative work described include novel screening methods to quantify protein-protein interactions in live cell-based screening systems, approaches for selective targeting of protein kinases, and design strategies for generating potent kinase inhibitors.

As a target class, kinases have “proven to be highly druggable and a rich source of clinical agents and approved therapies, some of which have been identified using relatively new approaches such as fragment-based design,” said James E. Dowling, Ph.D., principal scientist at AstraZeneca’s R&D Boston facility. “These accomplishments are quite impressive given the early views of many scientists who felt that the conserved ATP-binding pocket would present a challenge to the identification of selective inhibitors.”

As with all drug candidates, balancing factors such as potency, selectivity for the target, off-target effects, bioavailability, pharmacokinetics, and other drug-like properties is a challenge in the development of kinase inhibitors as therapeutic agents.

In the early stages of drug discovery, potency and chemical diversity are perhaps most important. “We also monitor parameters such as solubility, in vitro metabolic stability, and lipophilicity, and we use these parameters to help rank emerging chemical series,” said Dr. Dowling. “There tends to be a willingness to embrace emerging series with potential selectivity issues as long as there are strategies available for increasing primary target potency without simultaneously enhancing off-target effects. Another early objective is to identify early probe compounds with sufficient selectivity to enable successful validation of the target of interest.”

Screening on µ-Patterned Surfaces

Julian Weghuber, Ph.D., principal investigator at the University of Applied Sciences, Upper Austria, Wels, described a strategy for evaluating protein-protein interactions in live cells for drug discovery. Dr. Weghuber and colleagues are using microstructured surfaces, called µ-patterned surfaces, to analyze the interactions of various signaling proteins, such as tyrosine kinases, and membrane-bound receptors. Detection and quantification of the interactions in the context of a living cell is achieved with total internal reflection fluorescence (TIRF) microscopy. This µ-patterning technique has the advantage of being applicable for high-throughput screening.

To illustrate the mechanism of the µ-patterning assay used to detect protein-protein interactions, Dr. Weghuber refers to the membrane protein of interest as the “bait” and the fluorophore-labeled protein as the “prey.” Grids of µ-patterned antibody specific to the membrane protein bait are printed on glass surfaces. In cells grown on the µ-patterned biochips, the bait in the cell membranes will arrange on the surface according to the antibody µ-pattern. Introduction of the fluorophore-labeled protein prey will result in a homogeneous distribution of prey if it does not interact with the bait, or in co-patterning with the antibody in the presence of bait-prey interactions.

The researchers are applying this µ-pat-terning technique as a characterization tool to analyze medically relevant drug targets, and specifically to target the interaction between epidermal growth factor receptor (EGFR) and growth factor receptor-bound protein 2 (Grb2) to identify new drugs that affect EGRF-based signaling. According to Dr. Weghuber, the researchers “performed an in-depth analysis of the interaction properties of these molecules and then used different pharmaceuticals to modify this interaction,” including monoclonal antibodies and kinase inhibitors.

Although the current methodology is not suited to large-scale drug screening, Dr. Weghuber cited several refinements his team is exploring to automate the technique and increase throughput rates. “These attempts include µ-patterned surfaces generated in 96- or 384-well plates, customized software, and the development of a TIRF-capable well-plate reader,” said Dr. Webhuber.

Scientists at the University of Applied Sciences, Upper Austria, Wels use cell signaling to determine the selectivity of protein kinase inhibitors and their use in drug targeting. In recent years protein kinases have become the most studied class of drug target. Many protein kinase inhibitors have now been approved as anticancer drugs and many more are undergoing clinical trials. [Gustoimages/Science Source]

Selective Kinase Inhibition

Piqur Therapeutics is a Swiss pharmaceutical company focusing on the discovery and development of cancer therapeutics targeting the PI3K-Akt-mTOR signal pathway. (In this pathway, PI3K refers to phosphatidylinositide 3-kinases; Atk, to protein kinase B; and mTOR, to the mammalian target of rapamycin inhibition.) In particular, the company is working on overcoming the problems of dual PI3K-mTOR inhibition.

Mutations and amplifications in the components of the P13K-Akt-mTOR pathway contribute to cancer cell growth, survival, and proliferation. The main challenge in targeting PI3K kinase, as in the kinase field, “is to obtain highly selective inhibitors, which allow us to understand the consequences of pharmacological inhibition of a particular P13K isoform,” said Doriano Fabbro, Ph.D., CSO.

In his presentation, Dr. Fabbro discussed the current understanding of kinase inhibition based on the knowledge base gained from studies of small molecule binding to the ATP pocket of kinases. In addition, he described an example of how selective kinase inhibitors can have an effect on their overall structure.

Despite the central role kinases play in human biology, “and their sizeable potential as therapeutic targets, only a small fraction of the human kinome has been functionally annotated with ‘selective’ small molecule inhibitors,” said Dr. Fabbro. “Thus, the kinase field appears to be frozen in targeting the obvious. This is true with respect to target, site of inhibition (that is, ATP binding site), as well as indication (mainly oncology).”

In light of the cancer therapeutics currently approved that target the ATP binding sites of a handful of kinases, only a handful of these enzymes (about 20% of the kinome) have been exploited therapeutically, noted Dr. Fabbro, who added, “only a limited set of chemical probes is available to functionally annotate the kinome (including the untargeted kinases) and to stimulate new drug discovery efforts to address unmet medical need.”

A key challenge is to understand the “disease dependence” of a target kinase and to anticipate the potential for resistance following treatment with kinase inhibitors. Resistance can develop through various mechanisms, and minimizing the risk of resistance in the target kinase as well as the reactivation of relevant pathway(s) necessitates the use of a comprehensive combination of inhibitors, explained Dr. Fabbro.

A variety of tools and techniques are available to probe the kinome and define the functions of a specific kinase. These include the use of small molecule inhibitors, RNA interference (RNAi), and genetic knock-out or knock-in models.

An additional challenge in kinase-based drug discovery involves targeting kinases for which the cellular-signaling pathways are poorly or not at all understood. Deconvoluting the unknown cellular signaling as well as identifying the pathway in which the target kinase is embedded can be laborious and time-consuming, noted Dr. Fabbro. In essence, the various approaches, which include genetics and phosphoproteomics, are aimed at obtaining the relevant substrates of the known kinase in the unknown pathway.

Technology advances that could accelerate research on signal-transduction pathways and the identification of selective kinase inhibitors would, for example, “provide a common framework for understanding the activation of the kinase, disease causality, therapeutic modalities, and resistance as well as selectivity,” in Dr. Fabbro’s view. These include structural knowledge of the inhibitor-kinase interaction (x-ray, nuclear magnetic resonance, isothermal titration calorimetry, etc.) as well as pathway analysis combined with genetic and pharmacological tools.

The ultimate goal is to obtain a comprehensive annotation of kinases with their pathways and the understanding of their involvement in diseases. “The availability of selective chemical probes to functionally annotate the untargeted protein kinases will stimulate new drug discovery efforts to address unmet medical needs,” said Dr. Fabbro. All of these technology gaps will likely require a collaborative effort by a network or consortium of research groups.

Designed for Potency

Dr. Dowling’s presentation highlighted the importance of developing an early understanding of the physical properties of kinase inhibitors that may influence activity endpoints such as mechanistic or phenotypic assay readouts.

“We identified an early chemical series whose activity in cell-based formats was shown to correlate with the lipophilicity of the compounds,” said Dr. Dowling. “In a situation like this, if the lipophilicity is relatively high and the activity is modest, then further increases in potency are likely to be achieved only with a molecule whose overall drug-like qualities are poor.” Dr. Dowling’s group identified other compound series in which significant gains in potency could be achieved in a property-independent manner “and rationalized through x-ray crystallography of a ligand bound to the target.”

One of the challenges in identifying a new starting point and novel chemical series against a particular drug target is the limited knowledge of and experience with these new chemical entities. Structure- and property-based design approaches can generate data useful in establishing a profile of a newly discovered chemical series and assessing its drug-like potential. This information can help guide early-stage screening and prove useful to later project teams that may identify hits in subsequent screens using this series.

“Structure-based approaches offer a visual and conceptual framework for generating novel ideas that can be tested experimentally, initially using in silico methods and then in other settings, such as test tubes, cells, and larger organisms. In an early discovery program, they can be applied to enrich the diversity of your screening output by facilitating the design and evaluation of additional, potentially novel chemical scaffolds,” said Dr. Dowling.


Interest in drug discovery has driven much of the research on the role of protein kinases in a range of signal-transduction processes and in disrupted cell-signaling pathways linked to disease. In turn, this research is contributing to a clearer understanding of the many physiological and biochemical systems in which protein kinases play a role.

For example, in a study of cytoskeletal signaling, Izabela Michalczyk and colleagues report that a role for protein kinase Cθ is emerging (Journal of Leukocyte Biology 2013). Also, in a genome-wide analysis of kinase-chromatin interactions, Jonathan Göke and colleagues report that the extracellular signal-related kinase ERK2 co-binds to a DNA promoter with a transcription factor that is essential for the pluripotency and self-renewal of human embryonic stem cells (Molecular Cell 2013).