Over the past 20 years, the fight against cancer has been prosecuted with increasing zeal by protein kinase inhibitors. More numerous, more sophisticated, and better targeted than ever, these anticancer weapons roll off biopharma production lines only after they succeed in kinase assays. The ultimate mission? Defeat cancer by disturbing aberrant signaling or restoring normal signaling.
Kinases, which are encoded by genes representing 2% of the human genome, transfer a phosphate group from ATP to an amino acid residue on a substrate protein, leaving behind ADP. This phosphate redirect often results in a conformational and—more important—a corresponding functional change in the modified protein, essentially flipping a cellular on-off switch one way or another. These switches control information cascades in uncountable cellular processes.
If a switch gets stuck in the “on” position, whether from a mutation of its own or one upstream in its cascade, it may cause a cell to become cancerous. That makes the control of switching a cancer drug strategy, and a $100+ billion industry has arisen to find small molecules that can exert that control with high specificity and low toxicity.
Kinase assays are the first step in determining how these switch-like proteins may be controlled. The first kinase assays measured substrate phosphorylation by biochemically purified kinase in vitro, but over time, the assays have themselves changed conformation, assuming more complex structures with new functions. Now, kinase activity can be tracked in whole-cell systems in vivo, through substrate phosphorylation, ATP consumption, or even ADP generation. GEN consulted with six experts in kinase assays to find out what the next conformational change might bring.
GEN: It’s difficult to envision laboratory work before kinase assays. What new developments have you excited?
Dr. Lowery: With several very good biochemical assay methods available, most of the R&D effort is in improving cellular methods, and the holy grail is live cell kinase assays. A couple of recent developments that I like are a bioluminescence resonance energy transfer (BRET)-based method for selectivity profiling in live cells and a method for monitoring the activity of a specific kinase in real time using fluorescence lifetime imaging. A nice thing about the latter method is that it does not involve any genetically encoded elements, which can have unintended effects on cell physiology.
Dr. Schmitt: Assays that can be performed with full-length unmodified kinases—and that can detect the phosphorylated substrate without resorting to fluorophore-labeled peptide substrates or specific monoclonal antibodies—are a particularly exciting new development. Recently, sophisticated ADP-quantifying universal kinase assays have been introduced that utilize lanthanide metal-tagged anti-ADP antibodies, with displacement of a fluorescent ADP analogue by unlabeled ADP measured by time-resolved fluorescence resonance energy transfer (TR-FRET).
The combination of a highly selective and sensitive direct ADP detection mechanism with a TR-FRET readout is an evolutionary, rather than a revolutionary, technological progression. Nonetheless, it has made high-throughput in vitro kinase assays more physiologically relevant by allowing unmodified kinases (rather than fusion proteins containing only the catalytic domain) and substrates to be used.
Sensitive detection of the small amounts of ADP generated (in the presence of much larger amounts of ATP) enables inhibitor screening to be conducted with physiologically appropriate concentrations of substrate and ATP and allows substrate and ATP concentration to be varied, which is essential for delineating the mechanism of action of inhibitors (or potential allosteric modulators). Finally, this type of assay can also be adapted to nonprotein kinases (such as lipid kinases, which are promising targets for cancer, diabetes, and neuropsychiatric drugs).
Mr. Roux: Strategies have emerged to target kinase-related dysfunction. One sounds particularly promising: induced protein degradation using Proteolysis-Targeting Chimeras (PROTACs). It addresses the concern that inhibitors are targeting only the catalytic activity of a kinase while other domains of the kinase (which may still have important physiological functions and roles in disease) are left unmolested.
PROTACs are kinase inhibitors linked to an E3-ligase-targeting compound that induces kinase-selective degradation through ubiquitinylation and the proteasome machinery. They require related assays to screen, characterize, and optimize this new class of compounds. This includes assessment of binding affinity on recombinant purified kinases and assay platforms enabling functional readout (monitoring kinase expression level upon compound treatment).
This kind of assay must be selective for the kinase of interest, allow medium to high throughput on cellular models with endogenous levels of protein to enable screening, and also allow pharmacological characterization of compounds on physiologically relevant models.
One can also argue that in-depth validation of mechanism of action of PROTAC compounds would require cell-based assays enabling monitoring of kinase/E3 ligase interaction and kinase ubiquitinylation selective detection. While expression-level-monitoring assays already exist for PROTACs, the cell-based protein–protein interaction and ubiquitinylation assays remain to be adapted.
Dr. Pohl: With the emergence of a variety of multiplex methods for global kinase profiling, we are on the verge of seeing a perspective shift on kinase modulation and the development of kinase-targeting drugs. The ability to fingerprint the signaling and activation state of the kinome at high throughput and with high specificity is paving the way not only to personalized medicine for targeting kinase activity, but also to research helping us grasp the complexity of pathway activation states in cell and tissue biology.
Biochemical assays allowing high-throughput screening for presence, activation state, and target spectrum of kinases merge elegantly to help us form a picture of the kinome of any given target cell type. The prospect of testing unmodified drugs in a naïve cellular environment, thus allowing a better understanding of intracellular potency and selectivity, opens fascinating routes and perspectives.
Dr. Carlstrom: I am excited to see drug-modeling software improving and more scientists testing their structure-based drug designs in kinase inhibition assays. The more kinase structures that are solved and the better the modeling software becomes, the easier it will be to perform accurate structure-based drug design for kinase inhibitors that can then be tested in the lab with a variety of robust assays.
Dr. Goueli: Biochemical assay platforms have progressed from radioactive to fluorogenic to fluorescent and bioluminescent. They have led to biochemical assays that have led to tremendous success in developing many drugs that are currently on the market. The discovery of mutations by genome-wide screening for kinase amplification, translocation, and mutation has enabled scientists to rescreen against these targets for their potency and selectivity. Increasingly specific screening paves the way for scientists to develop the next generation of drugs. It is envisioned that next-generation sequencing will advance personalized medicine, improve drug efficacy, and minimize undesirable side effects.
GEN: What challenges are faced by those tasked with keeping in vitro assays relevant?
Dr. Lowery: It is not a matter of keeping them relevant, as biochemical assays are still an integral part of kinase drug discovery. Full-deck screens are not that common anymore, but regardless of how you start, whether it is with high-throughput screening (HTS), fragment-based screening (FBS), or virtual screening, you are still going to have to rely on biochemical assays for hit-to-lead development. For example, you will still need to deal with specific absorption rate, structure-driven design, mechanism of action, etc.
Dr. Schmitt: Some of the most significant challenges in designing new kinase-targeting antineoplastic drugs include: 1) development of selective compounds with minimal off-target effects, 2) lack of drug effect or emergence of resistance due to mutations that prevent inhibitor binding but preserve—or even enhance—catalytic activity, and 3) unexpected differences in inhibitor pharmacology in real cellular environments compared to in vitro assay conditions.
Problems of selectivity and problems of reduced efficacy in the face of mutations both stem from the fact that most of the current kinase inhibitors are competitive inhibitors targeting the ATP-binding domain of the substrate site. ATP-binding domains tend to be well conserved structurally across different kinases, making it difficult to achieve target selectivity. In addition, mutations that hinder access to the occluded ATP-binding domain are effective in preventing any ATP-competitive inhibitor from binding.
In vitro assays that help researchers identify allosteric ligands that bind to domains outside of the traditional ATP-binding domain (or bivalent inhibitors that bind to multiple domains simultaneously) will be more relevant to future drug discovery projects. The high frequency of kinase mutations in tumor cells—both across the patient population and emerging during therapy—also presents a challenge for in vitro assays.
Activity assays that screen compounds against wild-type kinases will yield hits that end up being ineffective in patients who may have mutant isozymes. Low-cost personalized gene sequencing (using either blood-based liquid biopsies or tumor tissue samples) will help guide the choice of therapeutics for individual patients, and help identify common mutations across patient populations. These mutant kinases will in turn become new targets for future drug discovery projects.
Mr. Roux: Kinase inhibitor drug discovery began with biochemical assays using recombinant purified kinases but lacking other cellular partners and machinery. While these binding- or activity-monitoring assays were (and still are) necessary for pharmacological validation and mechanism-of-action determination, development of assay platforms enabling kinase inhibitor studies in a cellular context was important, too. Indeed, several parameters can explain differing behaviors of a compound between biochemical and cell-based assays—one of them is the presence of kinase co-regulators.
In this context, it is important to monitor a compound’s effect on relevant cellular models with endogenous expression and nonmodified proteins. These assays must also be suitable for studies on varied cellular models, from classical 2D cell culture with immortalized cancer cells or primary cells, to 3D-spheroid cell culture, mouse xenograft, and whole-tissue systems. Finally, the assays must allow selective detection of the kinase of interest with medium to high throughput enabling in-depth characterization of compounds.
With TR-FRET technology, this type of cell-based platform already exists to assess the activity of several kinases. The same assay platform remains to be improved for monitoring binding.
Dr. Pohl: The challenge hasn’t so much been to keep in vitro assays relevant, but rather to deal with their advantages and limitations. For example, purely biochemical in vitro assays targeting specific kinase pathways or groups are crucial during secondary screening and profiling, but they are disadvantaged by their restricted coverage of the kinome. Often they do not reflect the true in vivo activity pattern.
The main challenge emerging in kinase assay–led inhibitory drug discovery is the integration of multiple assay types into one workflow. Combining high-throughput primary screening with biochemical and cell-based kinase assays in target validation and profiling is challenging, but necessary to determine specificity, efficiency, and perpetuity of candidate drugs. Outcome variations due to the different natures of the assays require stringent planning, good and consistent internal controls, and meticulous interpretation of the results.
Dr. Carlstrom: One of the most difficult challenges for keeping these assays relevant is correlating drug candidate activity in biochemical assays to that in cell-based assays. In many cases, there is a high failure rate of drug candidates due to early-stage tests not fully reflecting what occurs in live cells.
For example, many inhibitors of kinases are ATP analogs which may bind other kinases in the cell. Some compounds that bind to a recombinant kinase in a purified system may induce other effects on the kinase in the context of the cell, such as complex formation with other proteins.
Since most cell-based kinase assays focus on detecting substrate phosphorylation, these assays do not necessarily give a clear indication of target engagement.
Dr. Goueli: Most in vitro assays screen compounds against a single target, and profile the most potent ones against the kinome and other families of enzymes. They are relatively inexpensive and generate lead compounds that can be moved forward for preclinical and clinical phases of drug discovery.
Although this approach has succeeded with many kinases, it sometimes fails because the target either lacks sufficient validation or is present in a different form (or in a fusion form) in the cell. For example, the HER2 inhibitor lapatinib fails against the active conformation of HER2 due to an overabundance of HER2-HER3 heterodimer, making targeting the heterodimer the preferred strategy.
GEN: What has been the most innovative application of kinase assays recently?
Dr. Lowery: Well, this might be a bit biased, but I think that using high-throughput biochemical assays to obtain inhibitor-target residence times is a powerful approach for incorporating structure-kinetic relationships early in a program, especially with the increasing evidence that allosteric inhibitors (which tend to have longer residence times) make better drug molecules.
Dr. Schmitt: One of the most innovative recent technological developments for in vitro kinase assays is the real-time intracellular target engagement assay. This technology is particularly interesting as it enables measurement of inhibitor binding affinity and kinetic parameters (for example, ligand association/dissociation rate) in a native cellular environment.
While the technology does require transient transfection and expression of kinase/nanoluciferase fusion proteins in cell lines (limiting its utility as a pan-kinase high-throughput screening tool), it is currently the only assay technology that enables researchers to observe and measure kinase inhibition inside live cells in real-time.
Mr. Roux: Among the 38 FDA-approved drugs targeting kinases, a large proportion are type I inhibitors, and two of the drugs (Afatinib, which targets EGFR, and Ibrutinib, which targets Bruton’s tyrosine kinase) are type V inhibitors—or so-called covalent inhibitors. Besides the covalent vs. noncovalent classification, binding kinetic parameters (such as residence time and on- and off-rates) are of increasing interest.
Since 2006, residence time has been hypothesized to be an important parameter for drug optimization. Several papers have suggested that on- and off-rates are related to drug parameters like efficacy, selectivity, on-target side effects, and duration of action. Recently, the Kinetic for Drug Discovery European Consortium (K4DD) merged several academic, biotech, and pharmaceutical groups for various projects on this topic.
A recent paper used TR-FRET kinetic binding assays on purified kinases to analyze the on- and off-rates of 270 protein kinase inhibitors with 40 clinically relevant targets. This paper emphasized three points: First, there is an overabundance of compounds with longer residence times among clinically efficacious kinase inhibitors. Second, it is important to know compound kinetic parameters to accurately simulate target occupancy in vivo over time. Third, optimizing compound kinetic parameters (and other parameters such as compound clearance kinetics) may improve target selectivity in vivo. This study affirms that kinetic parameters are important in drug optimization and underlines the need for accurate and high-throughput kinase assays.
Dr. Pohl: Kinase inhibitor discovery has made immense progress in oncology while branching into other fields including inflammatory, autoimmune, and degenerative disorders. As the interest in kinases and their clinical inhibition broadens, demand grows for easy, robust, and cost-effective kinase assays with high-throughput capacities and high in vivo relevance.
Recent intriguing applications use BRET or FRET and competitive binding of tracers to visualize kinase-substrate interactions in cell-based assays to profile and quantify a wide spectrum of kinases in live cells. The demonstrated improved selectivity of these assays over biochemical assays underlines the relevance of screening methods that mimic in vivo conditions as closely as possible. An interesting route to single-cell kinase analysis has been laid out by combining droplet-based microfluidics and receptor-tyrosine kinase assays to allow single-cell resolution of kinase activity measurements. These applications display the ample potential of kinase assays yet to come.
Dr. Carlstrom: Cellular thermal shift assays represent an innovative application that can be used to verify the binding of compounds to the kinase of interest within the context of the cell. These assays show target engagement by measuring the changes in the protein stability upon compound binding in the intracellular environment. As the compound is titrated, it binds and stabilizes the kinase at elevated temperatures. Then the EC50 and rank order for compound binding to a specific kinase can be determined using antibody-based methods. The binding of the compound to a specific kinase within the cell can then be correlated to substrate phosphorylation using cell-based techniques.
Dr. Goueli: The most recent technologies are those targeting the kinases in situ, such as target engagement, immunoassays using antiphospho antibodies that recognize phosphorylated targeted substrates, or the phosphokinase target itself in homogenous, cell-based fluorescent or luminescent formats. Chemical proteomics and chemical genetics for evaluation of kinase activity, using kinase affinity probes and PROTAC technology to interrogate activated kinase in cell lysates, has also shown great potential. The most impressive clinical applications enabled by kinase assays are 1) combination therapies using drugs developed for multiple kinases on the same or overlapping signaling pathways, and 2) combination therapies using drugs developed for kinases alongside immunotherapy-based drugs.