Sage Journals: Discover world-class research

Abstract

Japanese

Korean

Chinese

The cellular thermal shift assay (CETSA) was introduced in 2013 to investigate drug–target engagement inside live cells and tissues. As with all thermal shift assays, the response measured by CETSA is not simply governed by ligand affinity to the investigated target protein, but the thermodynamics and kinetics of ligand binding and protein unfolding also contribute to the observed protein stabilization. This limitation is commonly neglected in current applications of the method to validate the target of small-molecule probes. Instead, there is an eagerness to make direct comparisons of CETSA measurements with functional and phenotypic readouts from cells at 37 °C. Here, we present a perspective of the early CETSA literature and put the accumulated data into a quantitative context. The analysis includes annotation of ~270 peer-reviewed papers, the majority of which do not consider the underlying biophysical basis of CETSA. We also detail what future technology developments are needed to enable CETSA-based optimization of structure–activity relationships and more appropriate comparisons of these data with functional or phenotypic responses. Finally, we describe ongoing developments in assay formats that allow for CETSA measurements at single-cell resolution, with the aspiration to allow differentiation in cellular target engagement between cells in co-cultures and more complex models, such as organoids and potentially even tissue.

Keywords

cellular thermal shift assay target engagement mechanism of action heat pulse design

Introduction

An understanding of cellular target engagement is critically important to validate the underlying mechanisms of action of chemical probes and drugs.¹ Such measurements not only shed light on compound availability in live cells and tissues, but also link evidence of physical interaction with a target protein to intended, and sometimes unintended, functional and phenotypic cellular responses. A series of retrospective analyses of compound failure in clinical studies^2–5 from leading pharmaceutical companies emphasized the need to verify compound exposure and physical target engagement in relevant tissue to understand whether the clinical hypothesis was in fact tested. Following these reports, recent years have seen a boost of new technologies to quantify physical binding events in cells and tissues.^6,7

One of these approaches is the cellular thermal shift assay (CETSA).^8,9 It relies on the established thermodynamic principle of ligand-induced stabilization, in which physical interaction of the ligand shifts the equilibrium between denatured and native forms of the target protein by preferentially binding to either of these states.^10–12 In short, a ligand that binds exclusively to the native protein, such as a specific active site binder, will increase the total concentration of folded protein when the ligand concentration is titrated toward and beyond the dissociation constant. Thermal shift assays exploit this coupled equilibrium by monitoring how these events change with temperature, reported as a shift in melting temperature (ΔT_m) for the studied protein. If the specific interaction of the ligand with the native protein persists at elevated temperatures, stabilization of the target protein is observed (i.e., ΔT_m > 0), while a temperature-dependent loss of the protein–ligand interaction can mask an interaction occurring at physiological temperature, such that ΔT_m = 0. This limitation, or bias, is well described, and application of T_m shift assays for screening activities comes with a risk of missing true binders unless adequately compensated for by increased compound concentrations or clever adjustments of solvent conditions.¹³

Since T_m shift assays report on binding equilibria at elevated temperatures, CETSA data must be appropriately extrapolated to physiological temperatures before quantitative comparisons are made with other cellular readouts.^14,15 As highlighted below, this consideration is poorly recognized in the literature, and abundant direct comparisons between CETSA measurements and cell-based readouts are reported without consideration of the temperature dependence of the investigated interaction. This comes with the risk of under- and/or overestimation of the relevance of observed cellular target engagement and can in the worst case lead to incorrect conclusions on the importance of the studied pharmacology. Here, we use literature data to exemplify the need to understand the underlying driving forces behind the binding event to support such extrapolations and comparisons.

These considerations are important also in the context of recent developments using mass spectrometry (MS)-based CETSA, also referred to as thermal proteome profiling (TPP).^16–18 It is becoming increasingly clear that live cell MS CETSA offers the possibility to broadly capture changes in the cellular protein interactome, including ligand-induced changes to protein-protein complexes and impact on the metabolome through “metabolite sensing” proteins.^19–24 Such measurements offer the possibility to combine primary CETSA responses with secondary alterations of protein stability that can be reliably derived from the primary pharmacology. In analogy with primary T_m shift responses, such functional biomarkers can be either stabilizing, destabilizing, or invisible, because secondary interactions are also expected to respond differently to temperature changes. An additional critical parameter toward a more complete mechanistic insight is offered by treatment duration, because the relatively rapid establishment of the primary interaction will differ in temporal appearance from long-term changes to the protein interactome (coined as “horizontal cell biology” in a recent review).²⁵ As discussed below, critical considerations in temperature, time, and CETSA amenability space will be required to define a suitably combined biomarker of treatment.

Literature Annotation

The first description of CETSA, in 2013, already demonstrated applicability to several different model systems, with a focus on intracellular kinases and enzymes involved in metabolic activation of antimetabolites.⁸ As recently reviewed, CETSA has subsequently been applied to a broad range of different protein classes^25,26 and now goes beyond soluble cytoplasmic proteins to also include membrane-bound proteins.^27–29 A survey of the literature using Web of Science (Clarivate) alongside Google Scholar and PubMed identifies close to 500 peer-reviewed articles citing the original study⁸ and/or the follow-up methods paper⁹ ( Suppl. Table 1 ; latest update: April 4, 2019). We have continuously followed this literature during the 6 first years, and about half of these contributions contain original CETSA data ( Fig. 1A ). While demonstration of cellular target engagement for select target proteins, commonly by means of Western blot ( Fig. 1B ), remains the primary objective of most studies ( Fig. 1C ), contributions using proteome-wide MS CETSA analysis to support target identification and protein interactome studies have essentially doubled in the past 18 months. This emerging and arguably most exciting field was recently reviewed in detail²⁵ and hence will not be the subject of this perspectives article.

Figure 1.

(A) Bar chart illustrating the number of peer-reviewed articles citing cellular thermal shift assay (CETSA) work between 2013 and 2019. The inserted pie chart details whether these studies included original CETSA data. Further details on the literature annotation are available in Supplemental Table 1 . (B) Pie chart illustrating what primary detection methodologies were used in the citing papers. A fraction is provided for the main detection methodology (Western blot), whereas the actual number of studies is provided for all techniques. (C) An illustration of the primary objectives in studies with original CETSA data. Categories include: (1) Target engagement: When the main purpose of the study was to demonstrate interaction between small molecules and a primary target protein in a complex sample matrix; (2) target identification: When mass spectrometry (MS) was used to understand what part of the proteome was affected by small-molecule treatment—either by direct interactions or by changes to the protein interactome; (3) application of CETSA for primary screening or structure–activity relationship (SAR) / mechanism-of-action (MoA) profiling purposes; (4) selectivity, in which small-molecule stabilization of select proteins was investigated; and, finally, (5) thermal stability, in which the stability of various forms of a protein, such as mutants, was compared. (D) Distribution of publications among journals, highlighting those with >10 contributions out of the 498 citing articles.

Not surprisingly, contributions are especially abundant in journals that cover aspects of drug discovery and chemical biology ( Fig. 1D ), in which CETSA and alternative methodologies for measuring physical target engagement in live cells are becoming standard practice for validation of new chemical probes. Parallel developments in this field have also afforded microplate-based high-throughput variants of CETSA to expand applications to screening, structure–activity relationship (SAR), and mechanism-of-action (MoA) profiling.^9,30–35 Our significant interest in the SAR and MoA areas triggered a further annotation of the literature with regard to their experimental details, with an emphasis on aspects that affect apparent compound potency. This includes the transient heat pulse design, whether compound was washed away prior to heating (e.g., through trypsinization), the inclusion of more than one compound to understand whether observed data agree with other measures of SAR, and whether the compounds were tested at multiple concentrations. Choices and implications of these parameters will be discussed in detail below.

Heat Pulse Design

While applications of CETSA have broadened to new types of model systems and readouts, most studies adhere to the same treatment regime and transient heat pulse (3 min) as applied in the original study ( Fig. 2A ).⁸ This is interesting because the heat pulse, while being essential for denaturation and aggregation of unstabilized proteins, arguably represents the main concern for live cell CETSA applications. This is because excessive heating of cells negatively affects cell permeability,^8,9,32,33 thus increasing the risk for false observations of cell-active compounds that are membrane impermeable at physiological temperatures. In addition, the apparent potency or isothermal dose–response fingerprint (ITDRF_CETSA) of each ligand is highly dependent on the duration of the heat pulse,^14,15,34 because the coupled protein denaturation and aggregation continue throughout and likely also for some period after the heating.

Figure 2.

(A) Distribution of the heat pulse duration in studies containing original cellular thermal shift assay (CETSA) data. Details are not always provided and were then labeled as not available (n.a.). We consider it likely that these authors reproduced the original procedure, in which case more than 90% of studies apply a 3 min heat pulse. (B) Illustration demonstrating whether treated cells were washed (and in some cases detached) prior to the critical heating step, such that target engagement may have been altered because of sample handling. (C, D) Bar chart illustrating how many (C) compounds and (D) concentrations were tested in each study. CR, Concentration response, such that 14CR involved testing of one compound at 14 different concentrations.

The continuous drift from equilibrium conditions during heating results in progressive loss of agreement with other cellular readouts, prompting the need for more effective heat transfer in shorter heat pulses.¹⁵ Such technological developments come with the potential of increasing the space of CETSA-amenable target proteins, but this area is hitherto left relatively unexplored. Instead, recent efforts toward high-throughput screening–compatible formats have seen a development toward longer heating times to increase assay windows,³⁴ although this comes with a compromise in terms of lost assay sensitivity. Based on learnings we have made since the original CETSA descriptions,^8,9 we instead recommend that users invest time in characterizing each model system (protein) with regard to the minimal time required to establish a thermal aggregation curve. If further experimentation is limited to a single heating time, this represents the best compromise to avoid unnecessary overheating, while additional experimentation at several heating times is highly encouraged for improved mechanistic insight.

To Treat or Not to Treat?

Examination of the literature on live cell CETSA experiments shows that, in general, four different experimental protocols are applied when considering the full treatment and heat sequence (i.e., including all steps from initial preincubation of cells with the compound to cell lysis after the transient heating and cooling of samples). Most studies (71%) follow the original 2013 protocol, in which cells are first exposed to compound and then harvested and washed in phosphate-buffered saline (PBS) to remove excessive compound prior to heat pulse. The advantage of this procedure is that it avoids the unwanted appearance of poorly permeable binders (i.e., compounds that become permeable and result in target engagement only because of the heat pulse). The extent of this problem is not well understood from the current literature, although recent cell membrane integrity experiments have shown that cells become permeable to some compounds when heated higher than 55–58 °C for 3 min (temperatures vary somewhat depending on which cells and which dye or substrate are used for the cell integrity assessments).^8,9,32–34 These considerations are critically important, since the intent of CETSA experiments is often to make quantitative comparisons of target engagement with other functional cellular readouts to draw conclusions on the pharmacological consequence of the observed interaction. This suggests the inclusion of complementary measures of cell permeability to validate the observed data or, alternatively, examination of whether responses change as a function of time after washing.

There are also obvious drawbacks with the removal of treatment prior to application of the heat pulse, especially if this includes harvesting cells through detachment and extensive wash procedures in the absence of compound. Equilibria over the cell membrane are known to reestablish on the timescale of minutes,³⁶ especially if the cellular efflux is mediated by active transport mechanisms, such that true binders may be washed away unless interactions are long-lived or the compounds are retained intracellularly through specific mechanisms (e.g., metabolized to cell-impermeable derivatives). We were more concerned with such appearance of false negatives when first formatting CETSA for screen purposes⁹ and instead opted to leave compound present throughout the heat step. In addition, this makes the protocol more compatible with large-scale profiling or screening, because it removes the cumbersome wash steps. This alternative practice has now appeared in about 26% of follow-up studies ( Fig. 2B ), many of which use Western blot–based quantification of remaining soluble protein. Further studies specifically addressing the impact of different wash times, combined with permeability measurements, are needed to better resolve the advantages and disadvantages of these two assay formats.

Our survey identified two additional practices besides these major approaches. First, there are a few studies in which cells are first treated with compound and then lysed (with or without inclusion of wash steps) prior to the heat step^37–40 ( Suppl. Table 1 ). Essentially, this becomes a combined live cell and cell lysate experiment, in which the added compound has the possibility to engage with the target protein in either of these states. The advantage of this format is not clear to us, because it comes with the same challenge as in live cell experiments (i.e., added compound must first compete with endogenous metabolites). This is the obvious advantage of dilute lysate experiments, in which the compound is more likely to encounter unliganded target protein that consequently is readily stabilized.²⁶ Finally, a very recent practice comes from the related field of NanoBRET (bioluminescence resonance energy transfer) measurements, in which a fluorescently labeled probe is first bound to and then competed off a NanoLuc-tagged protein.^41,42 These assays can be performed in a “permeabilized” format, in which digitonin is added to permeabilize cells to remove competition with endogenous metabolites such as adenosine triphosphate (ATP) for, for example, kinases. This was also recently practiced for CETSA,³² resulting in lower assay variability for this model system. This format is also a combined live cell and lysate experiment, but with the significant benefit of retaining cellular structures during the heat step.

One Compound, One Condition—Appropriate Interpretations?

A key limitation of Western blot–based CETSA is that it significantly limits the number of experimental permutations that can be tested. This becomes immediately apparent in published studies, which to a large extent involve testing a single compound or a few compounds (1 compound in 56% of studies, and 2–3 compounds in 28% of studies; Fig. 2C ). In addition, many studies involve experimentation at a single compound concentration and experimental condition (61%; Fig. 2D ). This means there is no underlying information on the temperature dependence of the studied interaction, thus preventing extrapolation to physiologically relevant temperatures and quantitative comparisons with functional readouts. Given that established equilibria alter due to the heat challenge (i.e., CETSA relies on protein denaturation and irreversible aggregation), it is surprising how little attention is paid to temperature-induced alterations in ligand-binding equilibria before claims are made as to the cellular MoA.

Frameworks for quantitative analysis including temperature extrapolations are well worked out for traditional T_m shift assays on isolated proteins,^43,44 including an understanding of the natural bias toward interactions that are increasingly favored at elevated temperatures (entropy driven). Enthalpically driven interactions, in contrast, are at risk of being lost, especially for proteins with high T_m’s, because these require extrapolations further away from 37 °C. There are, however, encouraging examples of such interactions for which the CETSA response persists to extreme temperatures, as exemplified by a chemical probe toward NUDT5.⁴⁵ The NUDT5 study represents the first use of CETSA as the main SAR-driving cellular assay alongside a biochemical assay, because it was deliberately chosen to avoid undesirable influence from nonselective cell viability assays in probe optimization. CETSA profiling for this thermostable protein was conducted at 83 °C, in which cell permeability is severely affected, such that what starts as a live cell assay becomes a permeabilized cell assay or even lysate during the heat pulse.

CETSA is expected to be governed by the same principles as conventional T_m shift assays with regard to temperature extrapolation, although with two major differences. First, in equilibrium T_m shift experiments, the samples are heated continuously (~1 °C/min), thus allowing enough time for sample reequilibration at all temperatures, whereas CETSA is based on the application of a transient heat pulse (commonly 3 min; Fig. 2A ). Second, CETSA, as we currently know it, is dependent on irreversible aggregation of denatured protein within this same time to generate the assay signal. While a 3 min heat pulse has proven enough for denaturation and aggregation of many proteins, the heat pulse duration represents the first point to vary in case the protein denaturation or coupled irreversible aggregation is slow in relation to a subminute timescale. In this regard, the early CETSA literature is disappointingly homogeneous with few exploratory deviations from the original protocol. Looking at this positively, there is abundant opportunity for further improvements, including turning some hitherto nonmeasurable interactions visible.

Several interactions have been shown to rearrange substantially on the timescale of the heat challenge, such as the interaction between lactate dehydrogenase A and a selection of 15 inhibitors,³⁴ resulting in a 20-fold shift in apparent potencies when varying the heat pulse duration between 5 and 50 min. Similarly, variation of the heat pulse between 0.5 and 20 min showed a several orders of magnitude shift of apparent potency for a selection of mature p38α inhibitors.¹⁵ The impact of temperature-induced reequilibration can also be deduced from changes to the heat pulse magnitude (as exemplified for, e.g., p38α^15,32 and SMYD3³³). Similar changes are apparent in the emerging two-dimensional MS CETSA literature, in which both heat pulse temperature and compound concentration are varied in the same experiment, as exemplified by the interactions between panobinostat and HDAC1, HDAC2, and phenylalanine hydroxylase.⁴⁶ Collectively, these data demonstrate the need to account for reequilibration in both experimental setup and analysis. Failure to adequately compensate for this (by, e.g., raising compound concentrations) comes with the risk of missing true interactions, potentially explaining why CETSA is commonly tested at high µM concentrations. Given the promiscuity of small molecules, especially early in drug development, this creates somewhat of a catch-22 situation, because the elevated concentrations can also promote target engagement that is not necessarily responsible for the observed functional responses.

A critical question then becomes: What can be regarded as good practice for interpretation of CETSA responses when there is a significant potency offset between assays? Systematic measurements of heat pulse duration and magnitude dependence require access to high-throughput amenable readouts, explaining the present lack of more quantitative CETSA studies. While the development of immunoassays for measurements of nonmodified endogenous protein requires the selection of affinity reagents and extensive assay development efforts,^9,30,31,35 assays based on tagged proteins are more readily available.^32–34 Pending improved availability of these tools in additional labs, there are encouraging data emerging on the correlation between CETSA data and functional data,^33–35 including retained ranking of compounds even when there are significant potency offsets between assays.³⁴ It is important to remember, though, that this may not hold true when the tested compounds come from different structural clusters or target different binding sites, because the molecular driving forces behind the interactions may differ significantly. Since many studies still rely on Western blot for readouts, we appreciate there are significant practical limitations in capacity and throughput. Our advice in these situations is to conduct the CETSA experiments with the shortest possible heat duration and as close to the thermal aggregation temperature as possible.¹⁵

Collectively, these data illustrate the need to study multiple cell-active compounds over a sufficiently broad potency range to probe the correlation between observed cellular target engagement through CETSA and functional responses, experiments that are feasible also with Western blot–based readouts. When building such mechanistic understanding, it is important to also properly consider the “vulnerability” of the investigated pharmacology;⁴⁷ that is, potency offsets between assays can also result from the need to modulate target activity to different extents (biological redundancy controls what target occupancy is required to achieve pharmacological effects). While a low vulnerability can be expected to negate some of the potency drop-off in CETSA (although serendipitous), cases with high vulnerability may see justifiably large differences between assays without jeopardizing the link between these events.

Further Miniaturization and Single-Cell-Resolution CETSA

A particularly attractive feature of CETSA is its potential application for discovery of new biomarkers and testing in patient-derived cells to support translational studies. While most published studies are based on in vitro treatment of lysates or isolated cells ( Fig. 3A ), there are only a few published examples of ex vivo and in vivo experiments.^8,9,48–51 Particularly noteworthy examples are the studies on RIPK1 inhibitors⁵⁰ and panobinostat,⁵² respectively, because they include demonstration of cellular target engagement in blood cells, as well as multiple tissues from treated animals. These pivotal translational studies pave the way for further development toward the potential use of CETSA as a biomarker of pharmacologic response.

Figure 3.

(A) Pie chart illustrating what types of samples were treated prior to application of the heat pulse. Note that several studies included testing of multiple sample types, in which case they were all counted. Cells denote treatment and heat pulse of live cells. (B) Illustration of the cell numbers applied in each sample replicate, reported here as a composite number regardless of detection technology. Unfortunately, 70% of published studies lack these experimental details. (C) Published studies were annotated with regard to what cells were used, whether these are cultivated in an adherent or a suspension state (details were obtained primarily through the ATCC Web page at http://www.lgcstandards-atcc.org/, while literature sources were used when not available there), and what state they were in when exposed to the heat pulse: For example, the majority of studies involved testing of an adherent cell line while kept in suspension (Adherent/Suspension).

Such application requires demonstration that signatures (in, e.g., whole blood) are reflective of pharmacological response in the relevant physiological tissue, or, alternatively, that primary cultures of patient-derived cells can be shown to reflect individual patient responses toward the investigated pharmacology. One important aspect currently limiting such latter application to a few compounds is the requirements for large cell numbers in CETSA measurements, commonly on the order of 10⁴–10⁷ cells per individual sample ( Fig. 3B ). The lower numbers are reported when the authors used ELISA-based methodologies, whereas Western blot–based experiments require on the order of 10⁶ cells per replicate, and MS experiments require substantially more.

These sample requirements currently restrict experimentation to a few selected compounds or, worse, may prevent testing in patient-derived cells because of limited cell availability. This is a key argument for development of more sensitive assay formats, ideally beyond those that are currently applied for screen purposes based on, for example, AlphaLISA (~10⁴ cells per well in published assays). Of potential interest in this space are imaging-based readouts, because they in principle can be applied to populations of a few hundred cells. These assays also offer the possibility to better understand heterogeneity in response, although at present it is not known whether responses in relevant cell populations can be masked when using bulk readouts in tissue samples. A critical limitation at present is that imaging CETSA has been practically demonstrated for only two model systems,^53–55 one of which is the highly amenable p38α system, and its broader application remains unclear at present.

Additional arguments for alternative CETSA formats come from the present literature’s bias toward testing of cells in suspension ( Fig. 3C ), regardless of whether this is the relevant state for the cells. This comes with a risk of affecting the studied biology in adherent cells because of stress responses following detachment, and subsequent retention in a nonnatural state (as applied in 66% of studies). Given the importance of appropriate context, as exemplified by functional testing in organoids and microphysiological systems,⁵⁶ it will be interesting to follow such developments also in the CETSA field.

Summary and Outlook

CETSA has received much attention since its introduction in 2013, to the extent that it is now commonly requested by reviewers for demonstration of physical target engagement in live cells. While most published studies focus on assessment of intracellular drug binding to a primary target protein of interest, more recent studies using MS-based CETSA also highlight the potential to also pursue alternative responses within the cellular protein interactome as biomarkers of treatment. These include changes to protein stability resulting from posttranslational modification, coordinated thermal unfolding of members of multiprotein complexes,²² and alterations in protein stability due to changes in endogenous metabolite levels.⁵⁷

This opens the possibility that secondary CETSA events could represent more robust measures of response, especially in cases in which the primary pharmacology demonstrates poor CETSA amenability, and when considered collectively could signify a more complete fingerprint of treatment. A prominent example of an apparently nonresponsive interaction can be found in the first publication on thermal proteome profiling,¹⁶ in which the well-established binding of dasatinib to the fusion protein BCR–ABL appeared blind to CETSA at first glance. Closer examination of the MS CETSA data showed co-stabilization of several interaction partners, however, thus providing secondary evidence of the interaction. A more integrated view that considers both primary and secondary responses was recently described as “horizontal cell biology,”²⁵ including also knowledge about the temporal appearance of each of these events. To fully exploit these opportunities, the analysis should incorporate also the CETSA amenability (i.e., the extent to which primary pharmacologies and downstream events respond to different heat pulses). It will be interesting to follow how such developments are integrated with, for example, blood–CETSA⁵¹ and other innovative approaches that can bring us closer to measuring individual patient responses.

Supplemental Material

Seashore-Ludlow_Supplemental_Table_1 – Supplemental material for Perspective on CETSA Literature: Toward More Quantitative Data Interpretation

Supplemental material, Seashore-Ludlow_Supplemental_Table_1 for Perspective on CETSA Literature: Toward More Quantitative Data Interpretation by Brinton Seashore-Ludlow, Hanna Axelsson and Thomas Lundbäck in SLAS Discovery

Footnotes

Acknowledgements

We acknowledge Helena Almqvist for early assistance with the literature annotation and Daniel Martinez Molina at Pelago Bioscience for reviewing the manuscript.

Supplemental material is available online with this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Chemical Biology Consortium Sweden (HA and TL) is financially supported by SciLifeLab and Karolinska Institutet. BS-L and HA also acknowledge SciLifeLab Technology Development Grants.

ORCID iDs

Brinton Seashore-Ludlow

Thomas Lundbäck

References

Arrowsmith

C. H.

Audia

J. E.

Austin

; et al. The Promise and Peril of Chemical Probes. Nat. Chem. Biol. 2015, 11, 536–541.

Morgan

Van Der Graaf

P. H.

Arrowsmith

; et al. Can the Flow of Medicines Be Improved? Fundamental Pharmacokinetic and Pharmacological Principles Toward Improving Phase II Survival. Drug Discov. Today. 2012, 17, 419–424.

Bunnage

M. E.

Chekler

E. L.

Jones

L. H.

Target Validation Using Chemical Probes. Nat. Chem. Biol. 2013, 9, 195–199.

Morgan

Brown

D. G.

Lennard

; et al. Impact of a Five-Dimensional Framework on R&D Productivity at AstraZeneca. Nat. Rev. Drug Discov. 2018, 17, 167–181.

Cook

Brown

Alexander

; et al. Lessons Learned from the Fate of AstraZeneca’s Drug Pipeline: A Five-Dimensional Framework. Nat. Rev. Drug Discov. 2014, 13, 419–431.

Simon

G. M.

Niphakis

M. J.

Cravatt

B. F.

Determining Target Engagement in Living Systems. Nat. Chem. Biol. 2013, 9, 200–205.

Schurmann

Janning

Ziegler

; et al. Small-Molecule Target Engagement in Cells. Cell Chem. Biol. 2016, 23, 435–441.

Martinez Molina

Jafari

Ignatushchenko

; et al. Monitoring Drug Target Engagement in Cells and Tissues Using the Cellular Thermal Shift Assay. Science. 2013, 341, 84–87.

Jafari

Almqvist

Axelsson

; et al. The Cellular Thermal Shift Assay for Evaluating Drug Target Interactions in Cells. Nat. Protoc. 2014, 9, 2100–2122.

10.

Crothers

D. M.

Statistical Thermodynamics of Nucleic Acid Melting Transitions with Coupled Binding Equilibria. Biopolymers. 1971, 10, 2147–2160.

11.

Schellman

J. A.

Macromolecular Binding. Biopolymers. 1975, 14, 999–1018.

12.

Brandts

J. F.

Lin

L. N.

Study of Strong to Ultratight Protein Interactions Using Differential Scanning Calorimetry. Biochemistry. 1990, 29, 6927–6940.

13.

Chilton

Clennell

Edfeldt

; et al. Hot-Spotting with Thermal Scanning: A Ligand- and Structure-Independent Assessment of Target Ligandability. J. Med. Chem. 2017, 60, 4923–4931.

14.

Seashore-Ludlow

Lundbäck

Early Perspective. J. Biomol. Screen. 2016, 21, 1019–1033.

15.

Seashore-Ludlow

Axelsson

Almqvist

; et al. Quantitative Interpretation of Intracellular Drug Binding and Kinetics Using the Cellular Thermal Shift Assay. Biochemistry. 2018, 57, 6715–6725.

16.

Savitski

M. M.

Reinhard

F. B.

Franken

; et al. Tracking Cancer Drugs in Living Cells by Thermal Profiling of the Proteome. Science. 2014, 346, 1255784.

17.

Franken

Mathieson

Childs

; et al. Thermal Proteome Profiling for Unbiased Identification of Direct and Indirect Drug Targets Using Multiplexed Quantitative Mass Spectrometry. Nat. Protoc. 2015, 10, 1567–1593.

18.

Huber

K. V.

Olek

K. M.

Muller

A. C.

; et al. Proteome-Wide Drug and Metabolite Interaction Mapping by Thermal-Stability Profiling. Nat. Methods. 2015, 12, 1055–1057.

19.

Becher

Andres-Pons

Romanov

; et al. Pervasive Protein Thermal Stability Variation during the Cell Cycle. Cell. 2018, 173, 1495–1507.

20.

Savitski

M. M.

Zinn

Faelth-Savitski

; et al. Multiplexed Proteome Dynamics Profiling Reveals Mechanisms Controlling Protein Homeostasis. Cell. 2018, 173, 260–274.

21.

Dai

Zhao

Bisteau

; et al. Modulation of Protein-Interaction States through the Cell Cycle. Cell. 2018, 173, 1481–1494.

22.

Tan

C. S. H.

K. D.

Bisteau

; et al. Thermal Proximity Coaggregation for System-Wide Profiling of Protein Complex Dynamics in Cells. Science. 2018, 359, 1170–1177.

23.

Lim

Y. T.

Prabhu

Dai

; et al. An Efficient Proteome-Wide Strategy for Discovery and Characterization of Cellular Nucleotide-Protein Interactions. PLoS One. 2018, 13, e0208273.

24.

Sun

Dai

; et al. Monitoring Structural Modulation of Redox-Sensitive Proteins in Cells with MS-CETSA. Redox Biol. 2019, 24, 101168.

25.

Dai

Prabhu

L. Y.

; et al. Horizontal Cell Biology: Monitoring Global Changes of Protein Interaction States with the Proteome-Wide Cellular Thermal Shift Assay (CETSA). Annu. Rev. Biochem. 2019, 88, 383–408.

26.

Martinez Molina

Nordlund

. The Cellular Thermal Shift Assay: A Novel Biophysical Assay for In Situ Drug Target Engagement and Mechanistic Biomarker Studies. Annu. Rev. Pharmacol. Toxicol. 2016, 56, 141–161.

27.

Reinhard

F. B.

Eberhard

Werner

; et al. Thermal Proteome Profiling Monitors Ligand Interactions with Cellular Membrane Proteins. Nat. Methods. 2015, 12, 1129–1131.

28.

Hashimoto

Girardi

Eichner

; et al. Detection of Chemical Engagement of Solute Carrier Proteins by a Cellular Thermal Shift Assay. ACS Chem. Biol. 2018, 13, 1480–1486.

29.

Kawatkar

Schefter

Hermansson

N. O.

; et al. CETSA beyond Soluble Targets: A Broad Application to Multipass Transmembrane Proteins. ACS Chem. Biol. 2019. doi:10.1021/acschembio.9b00399.

30.

Almqvist

Axelsson

Jafari

; et al. CETSA Screening Identifies Known and Novel Thymidylate Synthase Inhibitors and Slow Intracellular Activation of 5-Fluorouracil. Nat. Commun. 2016, 7, 11040.

31.

Shaw

Leveridge

Norling

; et al. Determining Direct Binders of the Androgen Receptor Using a High-Throughput Cellular Thermal Shift Assay. Sci. Rep. 2018, 8, 163.

32.

Dart

M. L.

Machleidt

Jost

; et al. Homogeneous Assay for Target Engagement Utilizing Bioluminescent Thermal Shift. ACS Med. Chem. Lett. 2018, 9, 546–551.

33.

McNulty

D. E.

Bonnette

W. G.

; et al. A High-Throughput Dose-Response Cellular Thermal Shift Assay for Rapid Screening of Drug Target Engagement in Living Cells, Exemplified Using SMYD3 and IDO1. SLAS Discovery. 2018, 23, 34–46.

34.

Martinez

N. J.

Asawa

R. R.

Cyr

M. G.

; et al. A Widely-Applicable High-Throughput Cellular Thermal Shift Assay (CETSA) Using Split Nano Luciferase. Sci. Rep. 2018, 8, 9472.

35.

Shaw

Dale

Hemsley

; et al. Positioning High-Throughput CETSA in Early Drug Discovery through Screening against B-Raf and PARP1. SLAS Discovery. 2019, 24, 121–132.

36.

Stein

W. D.

Kinetics of the Multidrug Transporter (P-Glycoprotein) and Its Reversal. Physiol. Rev. 1997, 77, 545–590.

37.

Xie

Zhang

Sun

; et al. Deacetylmycoepoxydiene is an Agonist of Rac1, and Simultaneously Induces Autophagy and Apoptosis. Appl. Microbiol. Biotechnol. 2018, 102, 5965–5975.

38.

Jorda

Hendrychová

Voller

; et al. How Selective Are Pharmacological Inhibitors of Cell-Cycle-Regulating Cyclin-Dependent Kinases? J. Med. Chem. 2018, 61, 9105–9120.

39.

Peng

Shen

Lin

; et al. Docking Study and Antiosteoporosis Effects of a Dibenzylbutane Lignan Isolated from Litsea cubeba Targeting Cathepsin K and MEK1. Med. Chem. Res. 2018, 27, 2062–2070.

40.

Dong

Chen

Wang

; et al. Small Molecule Inhibitors Simultaneously Targeting Cancer Metabolism and Epigenetics: Discovery of Novel Nicotinamide Phosphoribosyltransferase (NAMPT) and Histone Deacetylase (HDAC) Dual Inhibitors. J. Med. Chem. 2017, 60, 7965–7983.

41.

Robers

M. B.

Dart

M. L.

Woodroofe

C. C.

; et al. Target Engagement and Drug Residence Time Can Be Observed in Living Cells with BRET. Nat. Commun. 2015, 6, 10091.

42.

Vasta

J. D.

Corona

C. R.

Wilkinson

; et al. Quantitative, Wide-Spectrum Kinase Profiling in Live Cells for Assessing the Effect of Cellular ATP on Target Engagement. Cell Chem. Biol. 2018, 25, 206–214.

43.

Waldron

T. T.

Murphy

K. P.

Stabilization of Proteins by Ligand Binding: Application to Drug Screening and Determination of Unfolding Energetics. Biochemistry. 2003, 42, 5058–5064.

44.

Matulis

Kranz

J. K.

Salemme

F. R.

; et al. Thermodynamic Stability of Carbonic Anhydrase: Measurements of Binding Affinity and Stoichiometry Using ThermoFluor. Biochemistry. 2005, 44, 5258–5266.

45.

Page

B. D. G.

Valerie

N. C. K.

Wright

R. H. G.

; et al. Targeted NUDT5 Inhibitors Block Hormone Signaling in Breast Cancer Cells. Nat. Commun. 2018, 9, 250.

46.

Becher

Werner

Doce

; et al. Thermal Profiling Reveals Phenylalanine Hydroxylase as an Off-Target of Panobinostat. Nat. Chem. Biol. 2016, 12, 908–910.

47.

Tonge

P. J.

Drug-Target Kinetics in Drug Discovery. ACS Chem. Neurosci. 2018, 9, 29–39.

48.

Wang

Luo

Shan

; et al. Inhibition of Human Copper Trafficking by a Small Molecule Significantly Attenuates Cancer Cell Proliferation. Nat. Chem. 2015, 7, 968–979.

49.

Warpman Berglund

Sanjiv

Gad

; et al. Validation and Development of MTH1 Inhibitors for Treatment of Cancer. Ann. Oncol. 2016, 27, 2275–2283.

50.

Ishii

Okai

Iwatani-Yoshihara

; et al. CETSA Quantitatively Verifies In Vivo Target Engagement of Novel RIPK1 Inhibitors in Various Biospecimens. Sci. Rep. 2017, 7, 13000.

51.

Perrin

Werner

Kurzawa

; et al. Proteome Thermal Stability Reflects Organ Physiology and Identifies Drug Target Engagement In Vivo. bioRxiv. 2018. doi:10.1101/500306.

52.

Daumar

Dufour

Dubois

; et al. Development and Validation of a High-Performance Liquid Chromatography Method for the Quantitation of Intracellular PARP Inhibitor Olaparib in Cancer Cells. J. Pharm. Biomed. Anal. 2018, 152, 74–80.

53.

Axelsson

Almqvist

Otrocka

; et al. In Situ Target Engagement Studies in Adherent Cells. ACS Chem. Biol. 2018, 13, 942–950.

54.

Massey

A. J.

A High Content, High Throughput Cellular Thermal Stability Assay for Measuring Drug-Target Engagement in Living Cells. PLoS One. 2018, 13, e0195050.

55.

Axelsson

Almqvist

Seashore-Ludlow

Using High Content Imaging to Quantify Target Engagement in Adherent Cells. J. Visualized Exp. 2018. doi:10.3791/58670.

56.

Boreström

Jonebring

Guo

; et al. A CRISP(e)R View on Kidney Organoids Allows Generation of an Induced Pluripotent Stem Cell-Derived Kidney Model for Drug Discovery. Kidney Int. 2018, 94, 1099–1110.

57.

Dziekan

J. M.

Chen

; et al. Identifying Purine Nucleoside Phosphorylase as the Target of Quinine Using Cellular Thermal Shift Assay. Sci. Transl. Med. 2019, 11, eaau3174.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

4.71 MB