January 1, 2018 (Vol. 38, No. 1)
GEN‘s Panel Gives Expert Advice on Which Format Is Best
No drug developer imagines that every kinase-targeting program can be advanced by any single kinase assay format. Most drug developers recognize that if they are to advance a kinase-targeting program, they will have to sift through multiple kinase assay formats. In all likelihood, the drug developer will identify several potentially useful kinase assay formats, some of which are bound to be more practical than the others. The essential question, however, is this: Which format is best?
To answer this question, drug developers need expert advice, such as that offered in this GEN roundup article. According to the experts assembled by GEN, the best format depends on multiple criteria: project goals (target identification, candidate screening, lead optimization, selectivity determination); detection technology (radioactivity, luminescence, fluorescence); and the availability of commercial reagents and profiling services. Additional considerations include cost, convenience, and automation requirements, such as those imposed by high-throughput operations.
GEN: What steps, in general, should a scientist follow when selecting a kinase assay?
Dr. Lowery: The most important initial question is whether you will be performing in vitro assays or cellular assays; the assays used are very different. And if in vitro, will you be using a purified protein or a cell lysate? Next, you need to look at what kind of detection instrumentation is available; that is, what types of readouts can you handle?
At this point, kinase assay selection becomes more nuanced. The best choice will be dependent on the specific kinase and the strengths/weaknesses of each assay. What type of acceptor substrate will you be using. For example, will it be a peptide, a protein, a lipid, or a carbohydrate? What ATP and substrate concentrations do you want to use? What is the specific activity of the kinase; that is, what sensitivity do you require?
Dr. Ray: Choosing the best kinase assay to meet your research needs depends largely on your goals, budget, and desired throughput. As a first step, it is important to select an assay platform that will be cost-efficient, robust, convenient to use, and highly reliable (produce few false positives or false negatives). For example, time-resolved fluorescence assays are preferred for high-throughput screening applications, whereas radioactivity-based assays are more suitable for kinase profiling.
Choice of the assay also depends on the nature of the kinase. For example, the assay may assess a kinase that comes from a purified source or is part of a crude cell extract. Some assays are most suitable with high-activity kinases; others, with low-activity kinases (such as lipid kinases). Cell-based kinase assays are preferred for studying kinase activities in a physiological environment. Finally, it is important to consider assay parameters such as specificity, drug and ATP concentrations, etc.
Mr. Fossetta: When you begin a search for the best kinase-screening platform, you must decide whether your aim is to find a competitive or allosteric modulator. Quite often, kinase assays are developed using the catalytic domain of the kinase only.
These domains are generally easier to express and purify for screening purposes. This approach, however, often biases discovery to ATP-competitive modulators, which sometimes lack the potency and specificity needed to support the development of potent molecules.
Choosing to screen with a full-length kinase makes it possible to identify allosteric modulators inclusive of both competitive and non-ATP-competitive molecules. It is perhaps even more important to avoid those molecules that specifically bind in the ATP pocket.
Dr. Treiber: First, clearly define the project goals. (Typical goals include combinations of the following: high-throughput screening hit identification, hit validation, lead optimization, and selectivity determination.) Next, identify and validate a combination of technologies and outsourced versus in-house solutions that best addresses the project goals.
While attending to these tasks, keep in mind that orthogonal assay formats are often advisable. This approach may provide additional data validation and may best leverage the strengths of each assay format, such as formats for activity and binding.
For example, in a hit-to-lead campaign, a homogeneous assay format run in-house for high-throughput screening could be followed by an orthogonal assay—conducted in-house or outsourced—for hit confirmation and lead optimization. Throughout the campaign, outsourced profiling could be used to monitor kinome-wide selectivity.
Mr. Bischoff: When searching for the most suitable kinase assay, one should take into consideration many aspects, including instrumentation. Each lab has different instruments, and the assay readout should fit them.
Kinase assays based on fluorescence resonance energy transfer require a fluorimeter to read the assay results. Other kinase assays are based on colorimetric measurements and require a mere spectrophotometer, which is more common in labs. Yet other assays are based on bioluminescence and require a luminometer. In addition, many compounds might interfere with the readout, so sometimes different readouts should be used in parallel.
The assay format should also be considered based on the research goals. If the aim of the study is finding kinase inhibitors using a high-throughput screening campaign, then the assay should allow scalability and should involve as few steps as possible. An add-mix-measure format is preferable.
Ease of use, robustness, high throughput, and specificity are all important when designing a discovery campaign. Therefore, a homogeneous assay, preferably one that is nonradioactive and readily miniaturized, should be considered.
Dr. Olsen: One of the primary factors to consider when choosing a kinase assay is the level of throughput required. High-throughput screens in a 384- or 1536-well format are best performed using a homogeneous “mix and read” assay platform. Such a platform does not require time-consuming wash steps, which can introduce variability into results.
Fluorescence polarization and time-resolved fluorescence resonance energy transfer (TR-FRET) assays are homogeneous and are amenable to miniaturization due to their radiometric readouts. These assays are also relatively easy to automate.
Another consideration is interference due to compound autofluorescence and other factors. TR-FRET assays use a time-resolved detection mode that minimizes background fluorescence. Adopting an assay with a red-shifted readout may also be helpful in reducing background.
Dr. Khimani: Protein kinase assays measure the activity of kinases either as purified proteins or within cell extracts or tissue. Certain criteria that need to be considered for the best kinase assays are throughput, ease of use, sensitivity, reliability, and reproducibility. Another important consideration is the mode of detection. Possibilities include radioactivity, fluorescence, and luminescence assays. Additional considerations include optimization parameters, cost of reagents, desired substrate, and assay protocols. A number of multimode detection solutions for kinase assays are available from PerkinElmer.
Dr. Zegzouti: In general, whether the goal is to monitor kinase biochemical activity or study its regulation in cells, approaches with minimal reagent additions, such as homogeneous assays, are ideal. For a biochemical assay, generic formats based on universal kinase product detection allow scientists to monitor activities regardless of the substrate chemical structure.
A sensitive assay that can detect low substrate conversion with high signal/background ratios will allow scientists to use less enzyme, which will help them study the kinase at the initial linear rate of the reaction. Finally, an assay that is robust and resistant to chemical interference is desirable when screening for inhibitors to minimize false hits. Large chemical libraries may contain some fluorescent compounds or signal quenchers that may interfere with some assays.
GEN: What specific challenges arise when a scientist needs to select a kinase assay for lead discovery and optimization?
Dr. Lowery: The real challenge is deciding which parameters you want to use for lead optimization. You won’t be able to address all of them with a single assay, so you need to find a balance between the information used to drive lead optimization vs. resources, timeline, etc. For this reason, you need to take a closer look at how each assay works and its capabilities.
For instance, the ability to vary ATP and acceptor substrate concentrations is important for determining inhibitor mechanism, and most assays have limitations in this respect. Similarly, some kinetic analyses require a continuous assay, so that’s another consideration. The good thing is that there are assays available for pretty much any question you want to ask.
Dr. Ray: The vast selection of assay platforms poses a great challenge to scientists trying to decide which cost-effective technology to select. Although radioactivity-based assays are the “gold standard” in the field, radioactive waste disposal, multiple wash steps, and safety concerns make these assays less amenable to high-throughput screening applications. Fluorometric-based approaches, though fast and high-throughput, tend to have high false-positive and false-negative rates. Mobility shift assays require the synthesis of specific substrates, and hence they tend to be expensive.
There could be significant differences between assay platforms in lead selection due to multiple confounding factors, such as substrate sequence, probes involved, library quality, enzyme quality, pH, etc. Therefore, it is strongly advisable to test multiple assay platforms to validate the hits
Mr. Fossetta: Kinase screening format choices are plentiful, and they all having their advantages and disadvantages. Except for the tried-and-true radioisotope assays, all other formats are viable options for high-throughput screening. These other formats include homogeneous time-resolved fluorescence (HTRF), TR-FRET, and luminescence.
Although luminescence, like HTRF and TR-FRET, is compatible with high-throughput screening, implementing a high-throughput luminescence assay can be more difficult. Unintended inhibition of the coupling enzyme can lead to higher false-positive rates requiring extensive counter screening.
The choice between the fluorescence transfer–based formats comes down to design. Bead-based systems that use biotinylated detection components often pick up false-positive biotin mimics in small libraries. Systems that use singlet oxygen are susceptible to oxygen scavengers. The best assay options are homogenous formats that can be performed at low volumes (minimizing reagent consumption) and that are insensitive to light.
Dr. Treiber: Lead optimization requires co-optimization of on-target potency and kinome-wide selectivity. A challenge is to identify assays addressing both goals, while ensuring turnaround times synchronized with medicinal chemistry cycles.
For on-target potency, appropriate commercial reagents and outsourced solutions are available. For selectivity profiling, the use of outsourced kinase panels is industry standard.
Another challenge, often overlooked, is to fully understand key assay parameters and how these impact data quality and interpretation. For example, does the assay report an IC50, which depends on the ATP concentration, or a Ki/Kd, which reports more closely on inhibitor affinity and enables more robust findings on interkinase structure-activity relationships? And what are the potential assay failure modes? In general, complementary combinations of assay formats improve efficiency and reduce risk during lead optimization campaigns.
Mr. Bischoff: The challenges differ, depending on the stage of the optimization. For example, in the early discovery stages, the assay should be easy to use, robust, and optimized for a specific target kinase. In later stages, when the efforts focus on determining the specificity of a compound toward its targets, a panel of kinases should be assayed. It is important that all kinases within the panel would fit the assay format, and that the readout would not be affected by compound or reagent interference.
Dr. Khimani: Protein kinases represent one of the largest target families for drug discovery and development. A number of challenges are encountered by researchers in selecting kinase assays. Some considerations include cost, convenience, automation, and low false-positive and false-negative rates.
The availability of a broad range of technologies and solutions poses a challenge in the selection of kinase assays. Profiling large numbers of lead compounds during the early stage of structure-activity relationship studies can accelerate lead optimization. Fluorescence-based or low-cost radioisotope binding assays can enable this process.
Dr. Zegzouti: During lead discovery optimization, screeners want to efficiently determine the potency of the lead compounds against the specific kinase target, and to understand mode of action and selectivity profiles against many other kinases. The preferred kinase assay allows both of these goals to be met.
Assay sensitivity is important. Assays, therefore, that detect activity of low amounts of enzyme are desirable when calculating kinetic constants such as Km of substrates and inhibitor IC50 values. Universality of the assay is desirable to avoid a need for multiple assay platforms when performing inhibitor profiling against a panel of enzymes with diverse substrate requirements. Finally, the ideal assay should allow the use of a wide range of substrate concentrations to study the mode of action of different inhibitors.