Sage Journals: Discover world-class research

Abstract

Two different signaling pathways lead to the activation of the transcription factor NF-κB, initiating distinct biological responses: The canonical NF-κB pathway activation has been implicated in host immunity and inflammatory responses, whereas the noncanonical pathway activation has been involved in lymphoid organ development and B-cell maturation, as well as in the development of chronic inflammatory diseases and some hematologic cancers. The NF-κB-inducing kinase (NIK) is a cytoplasmic Ser/Thr kinase and is a key regulator of the noncanonical pathway. NIK activation results in the processing of the p100 subunit to p52, leading to the formation of the RelB/p52 complex and noncanonical pathway activation. Because of its role in the development of lymphoid malignancies, this kinase has always been considered as an attractive target for the treatment of certain types of cancers and immune diseases. We at Takeda have pursued a drug discovery program to identify small-molecule inhibitors against NIK. This report provides an overview of the data generated from our screening campaign using a small fragment library. Most importantly, we also provide a kinetic analysis of published compounds and chemical series developed at Takeda that are associated with a slow tight-binding mechanism and excellent cellular potency.

Keywords

NIK kinase NF-κB noncanonical pathway time-dependent inhibition residence time

Introduction

NF-κB proteins are a series of nuclear transcription factors that regulate a variety of physiological processes such as cell apoptosis, inflammation, and cell proliferation. This group includes RelA (also called P65), RelB, c-Rel, NF-ΚB1 p50, and NF-κB2 p52, which play key roles in the noncanonical and canonical NF-κB signaling pathways.¹ Each of these pathways is activated by different stimuli and involves subtle differences in their activation mechanisms that ultimately lead to distinct cellular responses. More specifically, the canonical pathway plays a key role in the innate and adaptive immune response. Its activation is rapid and transient, and will depend on the activation of transforming growth factor-β (TGFβ)-activated kinase-1 (TAK1) and the IkB kinase (IKK) complex. The IKK complex is an ensemble of three proteins, IKKα, IKKβ, and IKKγ, whose activation will lead to the phosphorylation of IkBα, a negative regulator of the NF-κB pathway. In its resting state, IkBα binds to the transcription factor RelA/p50, preventing this heterocomplex from migrating to the nucleus. Conversely, the activation of this pathway leads to the phosphorylation of IkBα, which dissociates from RelA/p50. Once free, IkBα undergoes proteolytic degradation by the proteasome. Its disappearance allows the RelA/p50 complex to translocate rapidly to the nucleus and to trigger the transcription of a variety of genes involved in the normal inflammation response. Alternatively, the activation of the noncanonical pathway plays a significant role in lymphoid organogenesis, B-cell survival and maturation, bone metabolism, and other cellular immune functions.^2,3 Unlike the canonical pathways, the activation of the noncanonical signaling pathway is slow-paced, will last longer, and is activated by a variety of ligands such as TNF-like weak inducer of apoptosis (TWEAK), B-cell-activating factor (BAFF), CD40 ligand (CD40L), lymphotoxin beta (LTβ), and LIGHT, also known as tumor necrosis factor superfamily member 14 (TNFSF14). The binding of these ligands to their respective receptors at the surface of the cell activates the intracellular NF-κB-inducing kinase, also known as NIK or MAP3K14. When activated, NIK phosphorylates and activates IKKα kinase, which in turn will phosphorylate p100 protein. On phosphorylation, p100 is ubiquitinylated and undergoes limited proteolysis to form a smaller p52 protein. In return, p52 forms a complex with RelB and migrates to the nucleus to initiate transcription of specific genes distinct from the canonical pathway ( Fig. 1 ). The action of NIK is critical for the activation of this pathway, and under normal conditions its activity is tightly regulated. NIK binds to the ubiquitinase TRAF2–TRAF3–cIAP1–cIAP2 complex, which exerts a rapid and negative control over its activity through targeted degradation. On stimulation, NIK dissociates from this complex and is free to phosphorylate IKKα and subsequently activate its downstream signaling cascade (see Fig. 1 ). Unlike other kinases, NIK is not activated through a phosphorylation mechanism. The 3D crystal structure of NIK suggests that the enzyme is already in an open conformation and adopting an active-like conformation.^4,5 This mechanism allows the enzyme to rapidly phosphorylate downstream substrates for the activation of the noncanonical pathway.

Figure 1.

Overview of the noncanonical pathway in its activated and basal states, with emphasis on NF-κB-inducing kinase (NIK).

NIK is a key regulator of the noncanonical pathway whose role has been well described during lymphoid malignancy.^6–8 As previously published, we also propose that NIK inhibition could be beneficial for the treatment of some cancers while preserving the ability of the canonical NF-κB pathway and the innate immune responses to fight against infection from external pathogens. In this study, we will highlight some of the results generated during our limited screening campaign, and exemplars of a few chemical lead series associated with the slow tight-binding mechanism and potent cellular data.

Materials and Methods

Materials

The NIK enzyme used in this study is a truncated recombinant protein (amino acid 319–947, 96 kDa) expressed in baculovirus with either an N-terminal glutathione-S-transferase (GST)-fusion or biotin tags. Both enzymes were purchased from Carna Biosciences (Kobe, Japan). The adenosine triphosphate (ATP)-binding site probe, tracer 236, and the europium (Eu)-labeled anti-GST antibody were purchased from Life Technologies/Thermo Fisher Scientific (Carlsbad, CA). The ADP-Glo kit was purchased from Promega (Madison, WI), and biotin full-length IKKα (dead mutant) was purchased from Carna Biosciences. Most of the studies described in the Methods section were conducted in 50 mM HEPES buffer containing 10 mM MgCl₂, 1 mM ethylene glycol-bis(beta-aminoethyl ether) N, N, N’, N’-tetraacetic acid (EGTA), and 0.01% Brij-35 at pH 7.5 at room temperature (~22 °C), with the exception of the ADP-Glo assay that was conducted in 40 mM Tris, pH 7.5, 20 mM MgCl₂, 0.1 mg/ml bovine serum albumin (BSA), and 50 µM dithiothreitol (DTT) at room temperature. The potencies of the NIK inhibitors [see the two-dimensional (2D) structures in Suppl. Sec. S1] were also determined in a cell-based assay by measuring phosphorylated IKKα (p-Ser176/Ser180) using a SureFire kit from Perkin Elmer (Waltham, MA). All of the concentrations reported are final in the assay buffer unless specified otherwise. Most of the equations highlighted in this article have been described in Ref.⁹. In addition, and due to legal and proprietary limitations, we are allowed to reveal only the 2D structures of some of our best compounds (see Suppl. Fig. S1 ).

Methods

IC₅₀ Determination

For the determination of IC₅₀ values, increasing amounts of inhibitor were added (from 0 to 10 µM) to a fixed amount of enzyme (5 nM) with or without pre-incubation prior to the addition of the ATP-binding site probe (i.e., tracer 236, 100 nM) and the anti-GST antibody (2 nM). The decrease in Förster resonance energy transfer (FRET) signal [λex = 337 nm; λem = 620 and 665 nm; FRET signal = [(fluorescence signal recorded at 665 nm) / (fluorescence signal recorded at 620 nm) × 10,000] was recorded using a PHERAstar instrument from BMG LabTech (Cary, NC). The experimental data were fitted using Equation (1) for the determination of IC₅₀:

\frac{E_{f r e e}}{E_{o}} = 100 (1 - \frac{1}{1 + {(\frac{I}{I C 50})}^{n}})

(1)

where E_free and E_o are the free and total amounts, respectively, of NIK enzyme in the reaction mixture; n is the Hill coefficient; I is the free inhibitor concentration; and IC₅₀ is the measure of the potency equivalent to the inhibitor concentration that leads to a 50% occupancy of the total enzyme. Each data were generated in duplicate; in addition, two IC₅₀ values for each compound were determined, as illustrated in Supplemental Section S3 .

To evaluate the time dependence mechanism, we used an ADP-Glo assay developed by Promega. Two sets of IC₅₀ values were determined for each compound, with or without a 60 min pre-incubation period of mixing the enzyme (10 nM) and the inhibitor in the assay buffer prior to the addition of ATP (200 µM, ATP Km is 40 µM) and biotin–IKKα dead mutant (1 µM) as the protein substrate. The reaction was quenched after 120 min by adding the detection reagents from the ADP-Glo kit. The reagents were incubated for an extra 40 min at room temperature. The luminescence signal was then recorded using a PHERAstar instrument, and the experimental data were fitted using Equation (1) for the determination of IC₅₀ values.

Thermal Shift Assay for the Determination of Compound Binding to NIK

The assay was conducted by adding 1 µM of recombinant NIK with 10 µM inhibitors to a 50 mM HEPES buffer containing 100 mM NaCl and SYPRO Orange (1×) at pH 7.4. Melting curves of the enzyme were recorded in the absence and presence of compound using the ViiA instrument from Life Technologies/Thermo Fisher Scientific. Briefly, the detection of a melting point (i.e., Tm) for a selected protein is achieved by gradually increasing the temperature of the medium, which causes the protein to switch from a well-folded state to a denatured form. The transition between these two states is measured by adding a fluorescent dye (SYPRO Orange) whose quantum yield increases when in contact with hydrophobic regions of the protein and whose enhanced fluorescence with time is recorded. A protein melting point is determined from the midpoint of the fluorescent curve. The binding of small molecules to the enzyme can be measured by their ability to affect the Tm of a protein.¹⁰

Determination of On-Rate (k_on) and Off-Rate (k_off) Constants Using a Probe Displacement Assay

The same GST-tagged NIK kinase, Eu-labeled anti-GST antibody, and fluorescent probe as described above (see the IC₅₀ Determination section) were used for the determination of k_on. One potent inhibitor (compound 13) was selected for this study. Each of these inhibitors was available as a 10 mM stock solution and was used in two separate experiments for the determination of k_on values. For the purpose of this study, compounds were diluted 100-fold in DMSO followed by a twofold serial dilution prior to being added to the assay buffer. In parallel, the enzyme, antibody, and active site probe were mixed together in the reaction buffer for a final concentration of 8, 100, and 8 nM, respectively. After a 5–10-min pre-incubation period, the inhibitor was added to this reaction mixture. The decrease in FRET signal was recorded with time using the same PHERAstar instrument as previously described. Initially, pseudo-first-order constants, k_obs, were determined for each inhibitor concentration by fitting the experimental data using Equation (2):

F = F_{m a x} + \frac{V_{z}}{k_{o b s}} . (1 - e^{- k_{o b s} . t})

(2)

where F is the FRET signal, F_max is the FRET signal at t = 0, V_z is the initial rate for the displacement of the fluorescent probe from the enzyme, and k_obs is the observed rate of association between the compound and enzyme.

The k_obs values were replotted as a function of inhibitor concentration in a secondary replot, and the data were fitted using Equations (3) and (4) to determine k_on, the overall on-rate; k₂, the rate of isomerization of the intermediate complex E–I^*; K_i^*, the dissociation constant for the first step of the inhibition mechanism (see Scheme 1); and [I], the inhibitor concentration.

E + I \overset{K i *}{⇄} E - I * ⇄_{k_{- 2}}^{k_{2}} E - I

(Scheme 1)

k_{o b s} = \frac{k_{2}}{(K i^{*} + [I])} . [I] + k_{- 2}

(3)

k_{o n} = \frac{k_{2}}{K i^{*}} w h e n [I] < < K_{i}^{*}

(4)

A similar protocol was used for the determination of the off-rate constant, k_off: An excess of Eu-labeled streptavidin (200 nM) was mixed with an equimolar amount of biotin-tagged NIK (200 nM) and inhibitor compounds (200 nM) in a 96-well plate format. After a 30 min pre-incubation, an aliquot of this mixture was diluted 100-fold into assay buffer containing an excess of active site tracer 236 (300 nM; 7.7-fold greater than its K_d; see Suppl. Sec. S2) in a second 96-well plate. The transfer of reagent from plate to plate was conducted using a BRAVO liquid handler (Agilent, Santa Clara, CA). The increase of FRET signal was recorded using the same fluorometer instrument as described previously. The experimental data were fitted using Equation (5) for the determination of k_off, the rate constant for the dissociation of inhibitor compound from the stable enzyme–inhibitor (E–I) complex (see Scheme 2):

\begin{array}{l} E * \\ ↑ k_{\deg} \\ E - 1 \overset{k_{off}}{\to} E + 1 \end{array}

(Scheme 2)

F = \frac{k_{o f f} . {[E - I]}_{o}}{(k_{d e g} - k_{o f f})} . (e^{- k o f f . t} - e^{- k d e g . t})

(5)

where F is the FRET signal, and k_deg is the constant describing the unfolding process of the free enzyme in the assay buffer. The detailed differential calculation process to generate Equation (5) is described by Capellos and Bielski.¹¹

Cell-Based Assay for the Detection of NIK-Dependent IKKα Phosphorylation

The assay was conducted using a stable double-transfected HEK293 Trex cell line (Invitrogen, Waltham, MA) that constitutively expresses the IKKα kinase domain and contains the full-length human NIK gene under control of the inducible doxycycline promoter. Prior to the addition of compound, cells are grown in a modified Eagle medium (MEM) with 10% fetal bovine serum (FBS) to reach 70–90% confluency. After a trypsinization step, cells are centrifuged and resuspended in MEM in the presence of doxycycline. Cells are added at a concentration of 5000 cells/well to a 384-well plate containing compounds in a final concentration of 0.5% DMSO. Negative control wells contain medium only. Plates are incubated overnight at 37 °C and 5% CO₂. On the next day, the medium is replaced with a lysis buffer provided by the vendor. Plates are left at 4 °C for several hours in the dark. Detection of phosphorylated IKKα (p-Ser176/Ser180) is determined shortly after by adding donor and acceptor beads from the SureFire kit to each well according to the protocol provided by the vendor. Fluorescent signal is recorded using a PHERAstar instrument. Experimental data are normalized against DMSO-treated cells and after subtraction of background readings (i.e., wells containing medium only). Calculated values are fitted using a traditional IC₅₀ curve; see Equation (1).

Results

Correlation between Biochemical, Biophysical, and Cellular Assay Data

A large subset of compounds was evaluated by measuring their potency (i.e., IC₅₀) for binding NIK using the LanthaScreen FRET binding displacement assay and a thermal shift assay, which measures the ability of an inhibitor compound to stabilize a protein against thermal denaturation. Using this assay format, a large subset of compounds with a wide range of potency (i.e., IC₅₀ values ranging from 10 to 0.01 µM) was evaluated in the presence of NIK. A plot of IC₅₀ versus Tm indicates that the potency of these compounds is directly correlated to their ability to stabilize NIK under heat denaturation stress ( Fig. 2 ).

Figure 2.

Correlation between data generated using thermal shift and Förster resonance energy transfer (FRET) binding assays (see Methods section). The data were fitted using a simple linear regression analysis for a final R² of 0.81.

In a subsequent study, we also determined the cellular potency of these compounds in a stable-transfected HEK293 cellular model that monitors cellular NIK kinase activity. In the cytoplasm, NIK directly binds and phosphorylates IKKα ( Fig. 1 ). The modulation of NIK-catalyzed IKKα phosphorylation in the presence of an inhibitor was quantified using an AlphaScreen assay (see Methods section). Because of its direct association with NIK, the phosphorylation of IKKα can be used as a direct target engagement biomarker for the inhibitors tested in a cellular environment. As expected, the IC₅₀ results from the FRET binding displacement assay correlated well with the ones generated in the cellular HEK293/NIK-IKKα phosphorylation assay ( Fig. 3 ).

Figure 3.

Correlation between data generated in the biochemical Förster resonance energy transfer (FRET) binding assay and cellular assay measuring NF-κB-inducing kinase (NIK)-dependent IkB kinase (IKKα) phosphorylation (see Methods section). The data were fitted using a simple linear regression analysis for a final R² of 0.66. Despite a good correlation between these two sets of data, our compounds are, on average, ~100-fold less potent when evaluated using a cell assay. The reason behind this shift of potency remains unclear; nevertheless, we suggest that the presence of a high concentration of competing adenosine triphosphate (ATP) in the cell could explain to some extent these discrepancies. For the FRET biochemical assay, the Z’ factor for each plate varies between 0.7 and 0.8. For the cell assay, the Z’ factor varies between 0.5 and 0.7.

Kinase Selectivity Panel

The control compound 13 was submitted to the kinase panel to evaluate its inhibition profile against a panel of ~300 kinases (see Suppl. Sec. S4). Based on its published crystal structure,¹² NIK kinase displays a unique hydrophobic back pocket near the ATP-binding site. This back pocket is a rare structural feature among the enzymes from the kinome. Most of the compounds published in this report were designed to take advantage of this back pocket for increased potency and, more importantly, improved selectivity. Despite the lack of co-crystal structure of compound 13 and NIK, we can certainly assume that it interacts with a similar mechanism as reported by de Leon-Boenig et al.¹² with its alkyne motif extending through a narrow channel and its aminothiazole moiety occupying the back pocket. As a result, inhibitor 13 is potent against NIK and is selective against the rest of the kinases from the kinome (i.e., only ~3% of the kinases of our selectivity panel exhibit a pIC₅₀ ≥ 6.5).

Time-Dependent Inhibition, Residence Time, Off-Rate and On-Rate Constants, and Binding Mechanism

Compounds used in this study were evaluated for their time-dependent inhibition mechanism against NIK. The study was conducted by measuring the potency of these compounds with and without pre-incubation with the enzyme. A few of these inhibitor compounds (I) showed an increase in potency after a pre-incubation period of 60 min with the enzyme (E), suggesting that formation of the E–I complex is slow to reach equilibrium ( Fig. 4 ). Time-dependent inhibition is often due to slow on-rates and/or slow off-rates during the interaction between an enzyme and an inhibitor. For this reason, some of these inhibitors were further investigated to determine their kinetic constants and the residence time, τ. These compounds were selected using the following criteria: pEC₅₀ > 4.5, good cell permeability (0.8 < cLogD_7.4 < 1.5; Mwt < 500; TPSA < 120; not charged at pH 7.4), and good solubility (>50 uM). The goal of this selection was to evaluate a potential relationship between cell potency and τ; therefore, we wanted to ensure that these compounds were not impaired by any solubility and cell permeability issues, and could reach the target in the cell without further interferences. A comparison between the EC₅₀ and τ data has been provided in Supplemental Figure S5 . To determine the off-rate, the assay was set up by first pre-mixing high concentrations of enzyme and inhibitor to allow the formation of the E–I complex. To initiate dissociation of this complex, an aliquot of this mixture was diluted into a large volume of a buffer solution containing an excess of an ATP-competitive active site tracer. On dissociation of the inhibitor compound, the tracer probe was able to bind to the enzyme replacing the inhibitor, leading to an increase of fluorescence as a function of time ( Fig. 5 ). This increase was followed by a decrease in fluorescence at later time points, however, which likely reflects the instability of the enzyme throughout time under these conditions. Therefore, experimental data were fitted using Equation (5), which takes into account both the increase in fluorescence due to probe binding and the subsequent decrease in fluorescence due to enzyme lability. This allowed for the determination of the dissociation rate constant, residence time, and inactivation decay rate of enzyme in assay buffer. As shown in Table 1 , these compounds could be rank ordered by their residence time values ranging from 100 to 1 min (see Table 1 ). One potent inhibitor, compound 13, was selected to further explore its kinetic profile and to investigate its association rate. The assay was set up by first pre-mixing NIK with the tracer fluorescent probe to allow formation of the E-probe complex. Next, the inhibitor was added, and displacement of the fluorescent probe from the enzyme active site was monitored by measuring the decrease in FRET signal with time ( Fig. 6a ). Initially, the experimental data generated in this study were fitted using Equation (2) to determine a pseudo-first-order rate constant, k_obs, for each concentration of inhibitor used. The k_obs values were replotted as a function of inhibitor concentration in a secondary replotting of the data ( Fig. 6b ). As previously described,¹³ the secondary replot saturation curve suggests that compound 13 interacts with the enzyme according to a two-step mechanism in which the inhibitor first binds to the enzyme to form a reversible intermediate E–I* complex. This complex then slowly isomerizes into a more stable and reversible E–I complex, as illustrated in Scheme 1. Furthermore, the fitting of these data provided values of the K_i*, k₂, and k₋₂ constants, which are reported in Table 2 .

Figure 4.

Time-dependent inhibition (TDI) of NF-κB-inducing kinase (NIK) inhibitors. Most of the compounds tested using the ADP-Glo activity assay gain potency (i.e., the IC₅₀ value decreases) with time when pre-incubated in the presence of the enzyme. Each data were generated in duplicate and exhibit 20% or less in standard deviation.

Figure 5.

Off-rate determination of NF-κB-inducing kinase (NIK) inhibitors. As previously described in the Methods section, k_off constants were determined by rapid dilution of the E–I (enzyme–inhibitor) complex in a large volume of buffer containing a high concentration of fluorescent-labeled adenosine triphosphate (ATP)-binding probe. As illustrated in this figure, compound 7 (•) is slowly released from the initial E–I complex as compared to compounds 9 (■) and 12 (○), respectively. Control experiments were conducted in the presence (□) and in the absence of enzyme (▲).

Table 1.

Off-Rate and Residence Time for Selected NF-κB-Inducing Kinase (NIK) Inhibitors.

NIK Inhibitor	k_off (min⁻¹)	Residence Time (min)
Compound 1	0.010	101.5
2	0.012	81.6
3	0.031	31.9
4	0.064	15.6
5	0.074	13.4
6	0.133	7.5
7	0.137	7.3
8	0.209	4.8
9	0.312	3.2
10	0.384	2.6
11	0.712	1.4
12	1.092	0.5

Note: All the data reported in this table have a standard deviation no greater than 15% of the original k_off values.

Figure 6.

Mechanism of inhibition analysis for compound 13. The mechanism of inhibition and the kinetic constants of compound 13 were further analyzed. The rate of association of this compound to NF-κB-inducing kinase (NIK) is slow (see Fig. 6a ) and follows a two-step mechanism (see Fig. 6b ). The study was conducted using a fixed amount of enzyme and several inhibitor concentrations (from 0 to 1 µM).

Table 2.

Detailed Kinetic Profile of Compound 13.

	Kinetic Constants
Compound	K_i* (µM)	k₂ (min⁻¹)	k₋₂ (min⁻¹)	k₂/K_i*(min⁻¹, µM ⁻¹)
Compound 13	0.057 ± 0.01	0.244 ± 0.012	0.075 ± 0.013	4.28

Note: This compound displays a slow k_on [i.e., k₂/K_i*; see Eq. (4)] and k_off (i.e., k₋₂) rate constants.

Discussion

The noncanonical NF-kb pathway plays a vital role during the immune host response and the development of lymphoid organs, the maturation and survival of B plus dendritic cells, and the T-cell response to pathogens.² Conversely, the dysregulation of this pathway has been implicated in certain forms of cancers and severe autoimmune diseases such as lupus erythematosus, rheumatoid arthritis, inflammatory bone loss, and liver inflammation.^14–18 A detailed inspection of this pathway suggests that NIK is the main gatekeeper responsible for its activation, which provides a unique therapeutic opportunity for the development of potent small-molecule inhibitors (see Table 1 and the compound 2D structures in Suppl. Sec. S1). Because of the therapeutic attractiveness of this target, we describe published and novel chemical lead series against NIK that were systematically profiled using a combination of biochemical, biophysical, and cell-based assays. Overall, the data generated with these various assays (see Figs. 2 and 3) correlate well with each other, and further analysis shows that some of these compounds bind tightly to NIK at the ATP-binding site, are selective (see the kinase selectivity profile in Suppl Sec. S4), and significantly downregulate the phosphorylation of IKKα intracellularly. In addition, the data reported here revealed that some of these lead compounds exhibit a time-dependent inhibition mechanism during the interaction with NIK (see Fig. 4 ). Time-dependent inhibition occurs when the interaction between the enzyme and inhibitor is slow to reach an equilibrium due to slow on- and/or off-rates. To further understand the causes of this mechanism, we have modified an existing fluorescent binding assay to investigate the kinetic profile of some of these compounds. The data generated from this assay show that some of these compounds exhibit a combination of long residence time and slow on-rate, as illustrated in Table 1 . The concept of drug residence time was introduced in 2006 by Copeland et al.,¹⁹ who demonstrated that in vivo efficacy can, in some cases, correlate well with the ability of a drug to remain bound to its target. To this day, the residence time remains a useful kinetic parameter for the optimization of lead compounds that need to achieve durable in vivo efficacy. It is also important, however, to emphasize that the impact of the residence time for a selected inhibitor can be limited by the clearance of the compound in the systemic circulation and the half-life of the biological target at the cellular or tissue levels.¹⁹ As for NIK, published data have reported a half-life of approximatively 3 h after pathway stimulation in normal B and T cells and malignant T cells.^20,21 NIK is quickly degraded by proteasomes, which effectively shut down the pathway shortly after enzyme activation. This fast and transient activation mechanism is well controlled and allows the cells to respond to any immune stresses with limited negative side effects. During a disease state, it has been demonstrated that an increase of NIK activity in the cell can lead to malignancy.²² This increase of activity can be due to either increased protein stability and/or overexpression of the enzyme protein level. As such, it remains unclear if the development of small-molecule inhibitors against NIK with slow off-rates (i.e., long residence times) could show efficacy at the cellular or in vivo level. The efficacy of NIK inhibitors is probably context specific and may not be a one-size-fits-all paradigm. Long-residence-time compounds could certainly be efficacious for cancer cell lines that exhibit deficiency in the TRAF2/3 pathway, but not in cell lines that display only a higher NIK protein level (and no impact on the enzyme half-life). In this study, we are using engineered cell lines that overexpressed NIK, the protein half-life is probably not affected, and therefore it is not surprising to observe a lack of correlation between the cellular efficacy (i.e., IC₅₀ determined with the AlphaScreen cell-based assay; see Methods section) and the residence time of our compounds (see Table 1 and Suppl. Fig. S5 ). Based on these observations, and as a strategy for lead optimization and for certain cancer types, we suggest that greater pharmacological effect for any NIK inhibitors could better be achieved by lowering the equilibrium dissociation constant, K_i, and/or increasing the on-rate, k_on, as opposed to decreasing the off-rate, k_off.

In summary, we have described our in vitro pharmacology process to identify hit compounds against NIK. In addition to published inhibitors, we have also reported some of the compounds developed at Takeda and new methods to investigate the kinetic profile of these inhibitors. We hope that this study will provide useful information for the development of more efficacious inhibitors for this target against a variety of disease states from cancer to inflammation.

Supplemental Material

Kinase_selectivity_final_version – Supplemental material for Biochemical and Cellular Profile of NIK Inhibitors with Long Residence Times

Supplemental material, Kinase_selectivity_final_version for Biochemical and Cellular Profile of NIK Inhibitors with Long Residence Times by Petro Halkowycz, Charles E. Grimshaw, Walter Keung, Paul Tanis, Chris Proffitt, Kim Peacock, Ron de Jong, Mark Sabat, Urmi Banerjee and Jacques Ermolieff in SLAS Discovery

Supplemental Material

Supplemental_information_with_revisions_V2 – Supplemental material for Biochemical and Cellular Profile of NIK Inhibitors with Long Residence Times

Supplemental material, Supplemental_information_with_revisions_V2 for Biochemical and Cellular Profile of NIK Inhibitors with Long Residence Times by Petro Halkowycz, Charles E. Grimshaw, Walter Keung, Paul Tanis, Chris Proffitt, Kim Peacock, Ron de Jong, Mark Sabat, Urmi Banerjee and Jacques Ermolieff in SLAS Discovery

Footnotes

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 received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Charles E. Grimshaw

References

Napetschnig

Molecular Basis of NF-κB Signaling. Annu. Rev. Biophys. 2013, 42, 443–468.

Sun

S.-C.

Non-Canonical NF-κB Signaling Pathway. Cell Res. 2011, 21, 71–85.

Gerondakis

Fulford

T. S.

Messina

N. L.

; et al. NF-κB Control of T Cell Development. Nature Immunol. 2013, 15, 15–25.

Leon-Boenig

Bowman

K. K.

Feng

J. A.

; et al. The Crystal Structure of the Catalytic Domain of the NF-κB Inducing Kinase Reveals a Narrow but Flexible Active Site. Structure 2012, 20, 1704–1714.

Liu

Sudom

Min

; et al. Structure of the Nuclear Factor kB-Inducing Kinase (NIK) Kinase Domain Reveals a Constitutively Active Conformation. J. Bio. Chem. 2012, 287, 27326–27334.

Paul

Edwards

Pepper

; et al. Inhibitory-κB Kinase (IKK) α and Nuclear Factor-κB (NFκB)-Inducing Kinase (NIK) as Anti-Cancer Drug Targets. Cells. 2018, 7, 176–196.

Annunziata

C. M.

Davis

R. E.

Demchenko

; et al. Frequent Engagement of the Classical and Alternative NF-kappaB Pathways by Diverse Genetic Abnormalities in Multiple Myeloma. Cancer Cell. 2007, 12, 115–130.

Gardam

Beyaert

The Kinase NIK as a Therapeutic Target in Multiple Myeloma. Expert Opin. Ther. Targets. 2011, 15, 207–218.

Hiromi

Kinetics of Fast Enzyme Reactions: Theory and Practice. Halsted Press, John Wiley & Sons: New York, 1979.

10.

Huynh

Partch

C. L.

Current Protocols in Protein Science: Analysis of Protein Stability and Ligand Interactions by Thermal Shift Assay. Curr. Protoc. Protein Sci. 2015, 79, 1–28.

11.

Capellos

Bielski

B. H. J.

Kinetic Systems: Mathematical Description of Chemical Kinetics in Solution. Robert E. Krieger Publishing: Huntington (NY), 1980.

12.

de Leon-Boenig

Bowman

K. K.

Feng

J. A.

; et al. The Crystal Structure of the Catalytic Domain of the NF-kB Inducing Kinase Reveals a Narrow but Flexible Active Site. Structure. 2012, 20, 1704–1714.

13.

Bieth

J. G.

Theoretical and Practical Aspects of Proteinase Inhibition Kinetics. Meth. Enzymol. 1995, 248, 59–84.

14.

Liu

Zhang

Joo

; et al. Review: NF-kB Signaling in Inflammation. Signal Transduct. Target. Ther. 2017, 2.

15.

Kajiura

Sun

Nomura

; et al. NF-Kappa B-Inducing Kinase Establishes Self-Tolerance in a Thymic Stroma-Dependent Manner. J. Immunol. 2004, 172, 2067–2075.

16.

Shen

Sheng

Xiong

; et al. Thymic NF-kappaB-Inducing Kinase (NIK) Regulates CD4+ T Cell-Elicited Liver Injury and Fibrosis in Mice. J. Hepatol. 2017, 67, 100–109.

17.

Willmann

K. L.

Klaver

Doğu

; et al. Biallelic Loss-of-Function Mutation in NIK Causes a Primary Immunodeficiency with Multifaceted Aberrant Lymphoid Immunity. Nat. Commun. 2014, 5, 5360.

18.

Shen

Xiong

; et al. Medullary Thymic Epithelial NF–kB-Inducing Kinase (NIK)/IKKα Pathway Shapes Autoimmunity and Liver and Lung Homeostasis in Mice. Proc. Natl. Acad. Sci. USA. 2019, 17, 19090–19097.

19.

Copeland

R. A.

The Drug-Target Residence Time Model: A 10-Year Retrospective. Nat. Rev. Drug Discov. 2016, 15, 87–94.

20.

Qing

Xiao

Stabilization of Basally Translated NF-kappaB-Inducing Kinase (NIK) Protein Functions as a Molecular Switch of Processing of NF-kappaB2 p100. J. Biol. Chem. 2005, 280, 40578–40582.

21.

Saitoh

Yamamoto

Zahidunnabi

; et al. Overexpressed NF-kappaB-Inducing Kinase Contributes to the Tumorigenesis of Adult T-Cell Leukemia and Hodgkin Reed-Sternberg Cells. Blood. 2008, 111, 5118–5129.

22.

Thu

Y. M.

Richmond

. NF-κB Inducing Kinase: A Key Regulator in the Immune System and in Cancer. Cytokine Growth Factor Rev. 2010, 21, 213–226.

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

0.13 MB

0.20 MB

Biochemical and Cellular Profile of NIK Inhibitors with Long Residence Times

Abstract

Keywords

Introduction

Materials and Methods

Materials

Methods

IC50 Determination

Thermal Shift Assay for the Determination of Compound Binding to NIK

Determination of On-Rate (kon) and Off-Rate (koff) Constants Using a Probe Displacement Assay

Cell-Based Assay for the Detection of NIK-Dependent IKKα Phosphorylation

Results

Correlation between Biochemical, Biophysical, and Cellular Assay Data

Kinase Selectivity Panel

Time-Dependent Inhibition, Residence Time, Off-Rate and On-Rate Constants, and Binding Mechanism

Discussion

Supplemental Material

Kinase_selectivity_final_version – Supplemental material for Biochemical and Cellular Profile of NIK Inhibitors with Long Residence Times

Supplemental Material

Supplemental_information_with_revisions_V2 – Supplemental material for Biochemical and Cellular Profile of NIK Inhibitors with Long Residence Times

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

References

Supplementary Material

IC₅₀ Determination

Determination of On-Rate (k_on) and Off-Rate (k_off) Constants Using a Probe Displacement Assay