Skip to main content
NCBI home page
ACS Medicinal Chemistry Letters logoLink to ACS Medicinal Chemistry Letters
. 2023 Mar 14;14(4):432–441. doi: 10.1021/acsmedchemlett.2c00530

Discovery of a Potent and Selective Naphthyridine-Based Chemical Probe for Casein Kinase 2

Zachary W Davis-Gilbert †, Andreas Krämer ‡,§,∥, James E Dunford ⊥, Stefanie Howell †, Filiz Senbabaoglu ⊥, Carrow I Wells †, Frances M Bashore †, Tammy M Havener †, Jeffery L Smith †, Mohammad A Hossain †, Udo Oppermann ⊥,#, David H Drewry †,, Alison D Axtman †,*
PMCID: PMC10108397  PMID: 37077385

Abstract

graphic file with name ml2c00530_0008.jpg

Naphthyridine-based inhibitors were synthesized to yield a potent and cell-active inhibitor of casein kinase 2 (CK2). Compound 2 selectively inhibits CK2α and CK2α′ when profiled broadly, thereby making it an exquisitely selective chemical probe for CK2. A negative control that is structurally related but lacks a key hinge-binding nitrogen (7) was designed on the basis of structural studies. Compound 7 does not bind CK2α or CK2α′ in cells and demonstrates excellent kinome-wide selectivity. Differential anticancer activity was observed when compound 2 was profiled alongside a structurally distinct CK2 chemical probe: SGC-CK2-1. This naphthyridine-based chemical probe (2) represents one of the best available small molecule tools with which to interrogate biology mediated by CK2.

Keywords: Casein kinase 2, CK2, CSNK2, Protein kinase, Naphthyridine, Chemical probe

Casein kinase 2 (CK2, CSNK2) is a highly conserved and ubiquitously expressed serine/threonine kinase for which more than 300 substrates have been identified, and correspondingly, many diverse functions and roles in disease have been ascribed to CK2.1−5 Examples of indications for which CK2 inhibition has been investigated as therapeutically beneficial include cancer, SARS-CoV-2, and neuroinflammation.6−10 Since the two catalytic subunits of CK2, CK2α (encoded by the CSNK2A1 gene) and CK2α′ (encoded by the CSNK2A2 gene), have >80% identity, small molecule inhibitors bind to both.11,12 As confirmed via Western blot analyses, inhibitor binding to the two catalytic subunits in cells results in inhibition of the heterotetrameric holoenzyme, which includes a dimer of noncatalytic subunits (CK2β).11−13

Recent efforts have identified potent and selective tool molecules that enable the dissection of complex signaling pathways mediated by CK2: IC19, IC20, and SGC-CK2-1 (Figure 1).11,12 Naphthyridine-based CK2 inhibitors were first described in 2010.14,15 The most widely used and advanced compound from this series, CX-4945 (silmitasertib), has been evaluated clinically for several oncology indications and recently for SARS-CoV-2.6,16 Silmitasertib received orphan drug status in the United States for the treatment of advanced cholangiocarcinoma.9,10 CX-4945 inhibits a number of kinases in addition to CK2, thereby raising the possibility that its success in cancer models may be due, at least in part, to polypharmacology.11,16 Efforts aimed at narrowing the inhibition profile of CX-4945 have resulted in several naphthyridine analogues, including CX-5011, CX-5279, and CX-5033 (Figure 1), that demonstrate improved selectivity when compared with CX-4945.11,17

Figure 1.

Figure 1

Open in a new tab

Structures of literature-reported potent CK2 inhibitors.

The clinical promise of the naphthyridine scaffold, demonstrated by the advancement of silmitasertib to Phase II clinical trials in addition to its acceptable human oral bioavailability, motivated further exploration of this chemotype. Despite reports of multiple exemplified naphthyridine analogues with nanomolar biochemical potency for CK2 and antiproliferative activity, a chemical probe with kinome-wide selectivity has not been described on the basis of this core.16,17 Such a compound would represent an orthogonal tool that would aid in deciphering the complex biological roles of CK2. We hypothesized that synthesizing and characterizing naphthyridine analogues would allow for discovery of a potent and selective tool in this chemical class.

To complement the data that has been collected for published analogues and to expand knowledge around this chemotype, we prepared a series of naphthyridines. We chose to explore the R1 position in addition to the presence or absence of the nitrogens in the fused ring system. As shown in Table 1, either a pyridine or pyrimidine was appended to the quinoline core to form the tricyclic system found in CX-4945/CX-5033 or CX-5011/CX-5279, respectively. Published crystal structures of naphthyridines CX-4945, CX-5011, and CX-5279 (PDB codes: 3NGA, 3PE2, 3R0T) confirm a pyridine or pyrimidine nitrogen within the tricyclic system engages in a key hydrogen bond with the CK2α hinge region. In addition, the carboxylic acid is involved in a water-mediated hydrogen-bond network. In contrast, the pendant aryl ring at position R1 that we modified is solvent-exposed, and the CK2α ATP pocket is not space-limited at this position. Accordingly, we appended at the R1 position substituted phenyl rings or naphthalene either directly via an anilino linkage or using a short alkyl chain. We aimed to explore whether CK2 potency in cells and/or kinome-wide selectivity could be improved through modification of this group. In addition, several analogues (2, 8, 9, 10, 17, 18, 21, 22, 24, 27) were designed to reduce the planarity of the resultant molecule in an attempt to improve solubility.

Table 1. Potency and Selectivity Data for Naphthyridine Analogues.

graphic file with name ml2c00530_0007.jpg

 
 NanoBRET data
 selectivity data
 solubility data
compound X Y R1 CK2α IC50 (nM) CK2α′ IC50 (nM) S10(1 μM) scorea # kinases with PoC < 10b kinetic solubility (μM)
2 N N J 920 200 0.007 3 210.8
8 N N N 570 93 0.017 7 151.7
9 N N M 830 140 0.02 8 109.1
10 N N L 930 270 0.022 9 147.0
11 N N I 21 6.0 0.025 10 54.3c
12 N N G 240 25 0.032 13 209.1
13 N N B 350 51 0.042 17 192.0
14 N N H 110 19     49.1
15 N N F 300 42     121.8c
16 N N A 740 63     168.2
17 N N K 550 110     146.4
CX-5011 N N E 350 66     66.7
18 N CH M 230 17 0.035 14 184.0
19 N CH G 430 130 0.05 20 25.3
CX-5033 N CH E 4300 1900 0.079 32 131.7
20 N CH I 100 22     32.8
21 N CH J 1100 450     202.4
22 N CH L 300 62     187.7
23 N CH B 680 180     236.7
24 N CH N 150 19     180.4
25 N CH A 1400 420     184.1
26 N CH H 220 43     148.4
27 N CH K 260 34     171.4
28 N CH D 1200 340     190.4
CX-4945 N CH C 240 180 0.069 28 277.1
7 CH N J >10 000 >10 000 0 0 80.4
SGC-CK2-1 NA NA NA 36 19 0.007 3 3.7
Open in a new tab
a

S10(1 μM): percentage of screened kinases with PoC < 10 at 1 μM.

b

PoC: percent of control values determined at 1 μM via DiscoverX scanMAX profiling.

c

Insoluble residue noted during solubility assessment.

These compounds were evaluated in the CK2α and CK2α′ NanoBRET assays to gauge their ability to engage the target kinase CK2 in cells (Table 1). Most CK2 NanoBRET assays were run in triplicate (Figures 2, 3, S2, and S3). The NanoBRET data revealed that molecules from this series engaged CK2α′ and CK2α with IC50 values of generally less than 1 μM. Excluding CX-4945, a pronounced 2.3–13-fold bias was observed when comparing IC50 values for CK2α′ with those for CK2α, with compounds being more potent against CK2α′. Previously reported CK2 NanoBRET IC50 values for CX-4945 align with this trend (7.6-fold bias).11 Compound 18 demonstrated the most significant difference in IC50 values. A more modest 1.9-fold bias for CK2α′ versus CK2α was observed for SGC-CK2-1 and was consistent in the pyrazolopyrimidine series from which it was selected.11 Compound 11 was the most potent compound in both CK2 NanoBRET assays (IC50 < 22 nM), while CX-5033 was weakest in these assays (IC50 > 1800 nM).

Figure 2.

Figure 2

Open in a new tab

Structure, potency, and selectivity data related to CK2 chemical probe 2. Kinome tree shows kinases that bind with PoC < 35 when compound 2 was screened at 1 μM. Assay format used to generate data in the last column of the nested table is listed in the preceding column. CK2 NanoBRET assays were run in triplicate (n = 3), and error bars represent standard deviation (SD). PoC = percent of control; NB = NanoBRET.

Figure 3.

Figure 3

Open in a new tab

(A) Crystal structures of human CK2α in complex with compound 2 (pink) in cartoon (CK2α) and stick (compound 2) representation. The hinge region is colored in wheat, the αC helix is in blue, the DWG motif is in green, and the water molecule is shown as a red sphere. The P-loop in yellow was made transparent for better view of the interactions. Hydrogen bonds are indicated as black dashed lines. The insert on the upper left corner is the electron density map (2Fo–Fc) of the bound ligand contoured at 1σ. (B) Overlay of compound 2 with published naphthyridine analogues (PDB codes: 3NGA, 3PE2, 3R0T) in the same color scheme as panel A.

The kinetic solubility of our naphthyridine analogues was next evaluated (Table 1). While none of our analogues improved on the value obtained for CX-4945, all demonstrated good aqueous solubility (>20 μM). For 2, 8, 12, 13, 16, 18, 21, 22, 23, 24, 25, 27, and 28, the measured solubility was reported as greater than 75% of the dose concentration, and thus, the actual solubility is estimated to be higher than Table 1 reflects. An insoluble residue was noted for analogues 11 and 15, and the lowest solubility values (25–50 μM) were recorded for 14, 19, and 20. The series of analogues (2, 8, 9, 10, 17, 18, 21, 22, 24, 27) designed to be less planar had good aqueous solubility (all > 100 μM). Of note, all naphthyridine analogues were more soluble than SGC-CK2-1 when evaluated side by side.

Given the published off-targets of CX-4945, the selectivity of these CK2-targeting compounds was important to assess, especially against the highly homologous CMGC kinases from the DYRK and HIPK subfamilies, which share many key active site residues with CKα and CK2α′. Enzymatic activity assays were carried out at a single concentration (1 μM) of inhibitor to generate percent of control (PoC) values (Table S1). Several compounds, including compounds 2, 8, 9, and 21, lacked activity against DYRKs and HIPKs, while a few others, including compounds 10, 17, 18, and 24, demonstrated potent inhibition of just one kinase in the 8-member panel. This result reinforced the idea that selectivity can be imparted in the naphthyridine scaffold. Most compounds (11 in total), however, potently inhibited more than half of the kinases in this enzymatic panel. While it is not a direct comparison, CX-4945 exhibits a PoC < 10 for all of these same kinases in Table S1 when profiled in the DiscoverX scanMAX panel at 1 μM (minus DYRK3 since it is not included in the scanMAX panel).11 In our limited DYRK and HIPK subfamily panel, more kinases preferred naphthyridine analogues with Y = CH versus Y = N (Table 1), which rendered them generally less selective.

We considered both the CK2 NanoBRET potency and the small panel enzymatic results to select some analogues for broader selectivity evaluation using DiscoverX scanMAX profiling. This kinome-wide profiling was executed at a single inhibitor concentration (1 μM). DiscoverX scanMAX data previously generated for CX-4945 at 1 μM is included in Table 1 and shows it has an S10 = 0.069 and that it potently binds to nearly 7% of the wild-type human kinases in the panel.11 Previously published kinome-wide selectivity data collected at 1 μM for SGC-CK2-1 is also included, which supports its utility as a CK2-selective chemical probe.11 Compound 14 was chosen on the basis of its potency for CK2α and CK2α′ in the corresponding NanoBRET assays (Table 1). Compounds 2, 8, 9, 10, and 18 were also profiled on the basis of their potent inhibition of either 0 or 1 kinase in the enzymatic panel (Table S1). Finally, compounds 12, 13, 19, and CX-5033 were selected as potent inhibitors of HIPK2 with or without CK2 affinity. Exemplars with Y = CH and Y = N (Table 1) were included to determine if the selectivity trend in Table S1 was maintained when more kinases were sampled. The kinome-wide profiling data summarized in the S10(1 μM) column of Table 1 supports the trend of overall selectivity improvements for the Y = N versus Y = CH compounds. Specific examples include compounds 9 versus 18 and 12 versus 19 where the analogue bearing Y = N demonstrated a better selectivity score. Except for CX-5033, all compounds demonstrated an improved selectivity score when compared with CX-4945. Compound 2 emerged as the most promising compound when considering broad selectivity [S10(1 μM) = 0.007]. No additional screening was executed to determine the selectivity of compound 2 versus nonkinase targets.

A deeper dive into the DiscoverX scanMAX data generated for compound 2 revealed 13 kinases with PoC < 35, including CK2α and CK2α′ (Figure 2). Enzymatic assays were employed as an orthogonal method to validate the scanMAX binding data for these 13 kinases. Potent inhibition of CK2α and CK2α′ was observed (IC50 ≤ 3 nM). Furthermore, the exquisite selectivity of compound 2 was confirmed, with a useful 200-fold window between the inhibition of CK2α and HIPK2, the next most potently inhibited kinase (IC50 = 600 nM). This compound demonstrates a 4.6-fold bias for binding CK2α′ versus CK2α in the corresponding NanoBRET assays (Table 1) and a similar trend in CK2 enzymatic assays (Figure 2).

Structural studies were employed to rationalize the selectivity of compound 2 for CK2. A cocrystal structure of compound 2 bound to CK2α was solved (Figure 3A, Table S4). This structure confirmed a key hydrogen bond between the nitrogen at the “X” position in Table 1 and a backbone NH between H115 and V116. Additional hydrogen bonds were noted between the carboxylic acid and K68, a water mediated with E81 on the αC helix, and a backbone NH adjacent to D175 from the DWG motif. This binding mode was compared with those previously solved for CX-4945, CX-5011, and CX-5279 (Figure 3B). The main difference observed was the orientation of the pendant aryl ring in the front pocket. Compound 2, which is shown in pink in panel B, can form an additional π-stacking interaction with H160 that is absent in previously solved structures. The methylene spacer present in compound 2, but not in the other scaffolds where the aryl ring is directly attached, positions the aryl ring so that it can adopt this favorable conformation. It is likely that the ATP binding sites of other kinases do not accommodate this ring in the same way, thereby precluding their binding to compound 2 with high affinity and imparting the selectivity we observe. Many more kinases tolerate binding of naphthyridine analogues with the aryl ring directly attached, which is exemplified by compounds 11, 12, 13, 19, CX-5033, and CX-4945 in Table 1 that all bind 10 or more kinases with PoC < 10. While the single methylene unit seems to provide the most optimal ability to π-stack, compounds 8, 9, 10, and 18 in Table 1 demonstrate that the incorporation of two methylene groups between the core and pendant aryl ring also imparts selectivity versus those analogues with the aryl ring directly attached.

Consideration of the naphthyridine cocrystal structures (Figure 3) aided in the design of an appropriate negative control compound to be used alongside compound 2 in a chemical probe set. To synthesize a structurally related compound that lacked an ability to bind to CK2, we specifically targeted the hinge-binding nitrogen. Compound 7 was prepared with the “X” position in Table 1 replaced by a carbon. The resulting compound was evaluated in the CK2α and CK2α′ NanoBRET assays, where it was found to not bind either kinase at concentrations up to 10 μM (Table 1 and Figure 4). When profiled in the DiscoverX scanMAX panel at 1 μM, a single kinase (PIP5K1C) was bound with a PoC < 35. Enzymatic follow-up, however, demonstrated that this binding in the DiscoverX panel did not result in PIP5K1C inhibition. Thus, modification of the hinge-binding atom on compound 2 resulted in a negative control (7) that lacks CK2 and kinome-wide binding. No additional screening was executed to determine the selectivity of compound 7 versus nonkinase targets.

Figure 4.

Figure 4

Open in a new tab

Structure, potency, and selectivity data related to CK2 negative control 7. Kinome tree shows kinases that bind with PoC < 35 when compound 7 was screened at 1 μM. Assay format used to generate data in the last column of the nested table is listed in the preceding column. CK2 NanoBRET assays were run in singlicate (n = 1). PoC = percent of control; NB = NanoBRET.

Noting the discrepancy between enzymatic and NanoBRET potency for these naphthyridines, we sought methods via which we could assess cellular permeability. We did this using two parallel approaches: the NanoBRET assay in permeabilized cells and a parallel artificial membrane permeability assay (PAMPA). All compounds in Table 1 were analyzed using the NanoBRET assay, while only a subset was subjected to PAMPA evaluation. As shown in Table S3 and Figures S4 and S5, all CK2-active compounds, including chemical probe 2, demonstrated improved cellular target engagement for CK2α and CK2α′ when the cells were permeabilized. For most compounds, a more dramatic boost in potency was noted for CK2α′ when compared with CK2α. Since permeabilizing cells also reduces the concentration of ATP, we suggest that the higher affinity (Km) of CK2α′ for ATP (reported by commercial vendors) results in a more dramatic shift when cells are permeabilized because there is less ATP competition for binding to the same site as our inhibitors. Interestingly, a very pronounced shift (140–760-fold) in IC50 values was noted for CX-5033. A less pronounced fold-change in IC50 values was observed for SGC-CK2-1 (<5-fold), which was similar to several naphthyridine analogues (11, 12, 13, 15, 16, 20, 24). The PAMPA results (Table S2) support the conclusion that, via this method, compound 2 and negative control 7 are more permeable than CX-4945. The PAMPA effective permeability (Pe) values generated for compounds 2 and 7 correlate with human fraction absorbed (%FA) greater than 80% and good permeability.18 Unfortunately, the low solubility of SGC-CK2-1 in assay media precluded analogous data generation for this compound. Since we cannot discern the impact of reduced ATP concentration on cellular target engagement when cells are permeabilized, we cannot conclude that the shift in IC50 values is because the naphthyridine analogues have poor cellular permeability. The PAMPA data indicate that our chemical probe set compounds are cell-permeable.

With a potent and selective compound in hand, we evaluated the impact on downstream signaling in response to compound 2. In a previous report, CX-4945 was found to reduce viability, induce cell cycle arrest and apoptosis, and hamper migratory capacity of MDA-MB-231 cells, a triple negative breast cancer cell line with elevated CK2 expression.19 This result motivated examination of the response of MDA-MB-231 cells to treatment with compound 2. An orthogonal CK2 probe known to inhibit downstream signaling mediated by CK2, SGC-CK2-1 (Figure 1), was included in these experiments in addition to CX-4945. As shown in Figure 5, a 24 h exposure of MDA-MB-231 cells to SGC-CK2-1, CX-4945, or compound 2 resulted in dose-dependent inhibition of AKT phosphorylation with no impact on total AKT expression. The concentration at which AKT phosphorylation is inhibited by each compound corresponds with the trend in their respective IC50 values in the CK2 NanoBRET assays: 19–36 nM for SGC-CK2-1, 180–240 nM for CX-4945, and 200–920 nM for compound 2. The rank order in efficiency of inhibiting AKT phosphorylation is as follows: SGC-CK2-1 > CX-4945 > compound 2. This result confirmed that CK2 inhibition by compound 2 disrupts CK2-mediated downstream signaling.

Figure 5.

Figure 5

Open in a new tab

(A) Western blot analyses of pAKT in MDA-MB-231 cells after 24 h treatment with SGC-CK2-1, CX-4945, or compound 2. Representative blot and quantification of phospho-AKT (Ser129) normalized to AKT, n = 3. SGC-CK2-1 p-values: 0.1 μM = 0.0287, 0.5 μM = 0.0045, 1 μM = 0.0038, 5 μM < 0.0001, 10 μM < 0.0001. CX-4945 p-values: 5 μM = 0.0303, 10 μM = 0.0043. Compound 2 p-values: 1 μM = 0.0072, 5 μM = 0.0286, 10 μM = 0.0149. (B) Western blot analyses of AKT in MDA-MB-231 cells after 24 h treatment with SGC-CK2-1, CX-4945, or compound 2. Representative blot and quantification of total AKT, n = 3. Error bars represent SEM. P-values were generated using a parametric unpaired t test with Welch’s correction by comparing each treatment condition to the untreated control. None of the concentrations in panel (B) were significantly different than the untreated control.

Whether inhibition of CK2 impacts cell survival and is a suitable anticancer target is a topic that has recently garnered much attention.11,20,21 In our preliminary experiments, neither compound 2 nor SGC-CK2-1 exhibited significant cytotoxicity when MDA-MB-231 cells were treated for 48 h up to a concentration of 10 μM (Figure S6). Compounds 7, CX-5011, and CX-4945 similarly did not exhibit cytotoxicity in our study (Figure S6). There are several possible explanations for the discrepancy between our observed lack of impact of naphthyridine compounds on viability and their published effect on cell growth.19 It has been reported that when treated with micromolar concentrations (2–10 μM) of CX-4945 for 24–72 h, MDA-MB-231 cell growth is inhibited. The authors point out that they rely on a WST-1 assay for readout and this indirectly measures cell growth on the basis of metabolic activity in the cells.19 CK2 regulates cell metabolism, which is a process that can be inhibited without an associated impact on cell growth.1,22 Another potential explanation is that MDA-MB-231 cells are moderately sensitive to the promotion of CK2-independent cytoplasmic vacuolization, called methuosis, at micromolar concentrations of CX-4945 and CX-5011.23−25 Metabolic-based viability assay (MTT) results generated in parallel support the idea that this vacuolization impacts cell metabolism.23 Similarly, methuosis is reported to drive metabolic failure.26 In contrast to an assay dependent on mitochondrial enzymes (WST-1), we opted to employ a luciferase-based assay (CellTiter-Glo2) that requires ATP in order to generate the luminescent species. Our assay results are more aligned with one that reported IC50 values for CX-4945 and CX-5011 of 6.4 and 9.8 μM, respectively, when MDA-MB-231 cells were treated for 4 days in dose–response, and Alamar blue was used to readout cell viability.17 Taken together, we suggest that the time course of treatment, as well as the method of assessment, can impact observed viability results.

Historically, CK2 has been recognized as an essential mediator in hematological malignancies.10,27 Multiple myeloma cells have been reported to rely on elevated CK2 for survival. Accordingly, when CX-4945 was profiled against a panel of 13 multiple myeloma cell lines, low micromolar antiproliferative activity was measured via a CellTiter-Glo viability assay.28 CX-4945 has been explored clinically for relapsed or refractory multiple myeloma.10 Apoptosis was induced much more effectively in multiple myeloma cells from patients when treated with orthogonal CK2 inhibitors (TBB, a TBB derivative, or CGP029482) than in nonmalignant control cells.29,30 This data motivated our preliminary examination of the effect of compound 2 versus published naphthyridines (CX-4945, CX-5011, and CX-5033) and SGC-CK2-1 at a single concentration of 1 μM on a panel of multiple myeloma cell lines treated for 5 days. Because it was designed in the later stage, negative control 7 was not included in this preliminary study. It is best practice to employ a chemical probe and its negative control side-by-side in biological experiments. The responses of JJN3, AMO-1 (parental and pomalidomide-resistant), and L363 (parental, carfilzomib-resistant, and bortezomib-resistant) multiple myeloma cell lines to treatment were analyzed via a metabolic-based viability assay using resazurin (akin to the WST-1 assay). As shown in Figure 6A, our preliminary data demonstrates that JJN3 cells were not responsive to treatment with any compound. The viability of the AMO-1 and L363 parental lines was reduced by treatment with compound 2 but not by treatment with the other naphthyridine analogues. We observed 44% viability of AMO-1 and 36% viability of L363 cells, respectively, following treatment with 1 μM of compound 2. The viability of the AMO-1 and L363 drug-resistant lines was not impacted by treatment with any naphthyridine analogue. Interestingly, the viability of all AMO-1 and L363 cell lines tested was significantly compromised by treatment with 1 μM of SGC-CK2-1, which resulted in 9–22% viability of the AMO-1 and L363 cells and 77% viability of the JJN3 cells. This suggests a chemotype-specific response and/or reflects the enhanced CK2 affinity of SGC-CK2-1 versus compound 2. SGC-CK2-1 is nearly 11-fold more active than compound 2 in the CK2α′ NanoBRET assay and nearly 26-fold more active than compound 2 in the CK2α NanoBRET assay.11 Our preliminary viability data aligns with literature reports of multiple myeloma survival being impaired via CK2 inhibition.

Figure 6.

Figure 6

Open in a new tab

Analysis of impact of naphthyridine analogues and SGC-CK2-1 on viability of (A) multiple myeloma, (B) Ewings sarcoma, and (C) chordoma cell lines when dosed at 1 μM. Scale bar shows the color gradient that corresponds with percent viability.

Less has been published about the role of CK2 in rare cancers that occur in bones, such as Ewings sarcoma, and in the soft tissue around bones, known as chordoma, which is a rare, slow growing cancer of tissue inside the spine. The EWS gene, which encodes a ubiquitously expressed RNA-binding protein of the same name, has been identified as a target of tumor-specific translocations in Ewings sarcoma family tumors. A putative CK2 phosphorylation site (Ser325) on an EWS protein isoform has been reported.31,32 Literature links between CK2 and chordoma are tenuous. Validated, repurposed chemotherapeutic drugs have been the focus of most cell-based therapeutic screens related to chordoma.33,34 As an exception, one compound, 5-iodotubericin, that inhibits CK2 in addition to other kinases, has demonstrated dose-responsive inhibition of UCH1 chordoma cells.35 Motivated to explore whether CK2 inhibition is a therapeutic avenue for these rare cancers, we examined the impact of the same panel of CK2 inhibitors, dosed at 1 μM, on the viability of four Ewings sarcoma and five chordoma cell lines following 5 days of treatment. The preliminary data in Figure 6B demonstrates that most Ewings cell lines were nonresponsive to treatment with naphthyridine analogues. The exception to this was the STA-ET-1 cells, which were modestly impacted by treatment with these compounds. CX-4945 elicited the most robust response of Ewings cell lines for compounds in this structural series and resulted in 21% viability of the STA-ET-1 cells and 63% viability of both the RM82 and STA-ET-10 cells. SGC-CK2-1, in contrast, only significantly impacted RM82 cells (36% viability) versus other Ewings cell lines evaluated. When considering the preliminary chordoma cell line data in Figure 6C, a limited impact on viability was noted. The exception to this trend was observed for SGC-CK2-1, which showed the sensitivity of UCH2, JHC7, and Mug-Chor cells to treatment and resulted in 60%, 53%, and 53% viability, respectively. As was observed in the multiple myeloma cell lines, a chemotype-specific response was observed for SGC-CK2-1 versus naphthyridine-based compounds in chordoma cell lines.

Viability data for naphthyridine compounds CX-4945 and CX-5011 has been published. Multiple studies have demonstrated single or double-digit micromolar viability IC50 values after dosing diverse cancer cell lines for 2–4 days.17,23,36,37 Importantly, a change in the time course of treatment (24 versus 72 h) resulted in notably different impacts on cell viability in HepG2 cells.23 When CX-4945 was profiled against 43 diverse cancer cell lines, their responses were variable. While leukemia and colon cells demonstrated greater sensitivity, the 11 included breast cancer cell lines displayed the widest range of sensitivities to CX-4945. The authors identified a correlation between sensitivity to CX-4945 and expression levels of the CK2α catalytic subunit in these breast cancer cell lines.37 These results suggest that for our multiple myeloma, chordoma, and Ewings sarcoma studies, 1 μM of naphthyridine analogues may not have been high enough to see a robust response and/or that, especially in the Ewings sarcoma cell lines, CK2α expression may be lower.

We have described the design, synthesis, and biological evaluation of a series of naphthyridines as inhibitors of CK2. Compound 2 emerged as our most optimal probe candidate. Comparison of cocrystal structures solved for compound 2 versus other naphthyridines revealed a key interaction that may drive its CK2 selectivity and informed design of a structurally related negative control compound (7). Inhibition of downstream signaling without an associated impact on viability was observed when MDA-MB-231 breast cancer cells were treated with compound 2. In contrast, the viability of specific multiple myeloma cell lines was differentially compromised because of treatment with 1 μM of compound 2, which may be related to different genomic lesions occurring in myeloma, now recognized as significant oncogenic drivers in this cancer.38 Ewings sarcoma and chordoma cell lines were generally not responsive to treatment with compound 2. Like SGC-CK2-1, compound 2 does not seem to be a broadly antiproliferative agent.11 This finding supports the concept that CK2 inhibition is a possible personalized treatment strategy for specific cancers, including multiple myeloma, and deserves further study. We recommend compound 2 as a CK2 chemical probe to be employed alongside an orthogonal CK2 chemical probe on the basis of a different chemotype, such as SGC-CK2-1, in cell-based studies aimed at interrogating CK2-mediated biology. As no pharmacokinetic data has been generated for compounds 2 and 7, we only advise use of our chemical probe set in cellular studies.

Acknowledgments

Promega kindly provided constructs for NanoBRET measurements of CK2α and CK2α′. The kinome trees in Figures 2 and 4 and the Table of Contents graphic were prepared using the TREEspot kinase interaction mapping software at http://treespot.discoverx.com. We thank the Mark Foundation for Cancer Research and the Chordoma Foundation for supporting our efforts related to the rare cancer chordoma. ChemSpace provided synthetic support for our program. We acknowledge the Department of Chemistry Mass Spectrometry Core Laboratory at the University of North Carolina for assisting with mass spectrometry analyses. Furthermore, we thank the beamline scientists at the Swiss Light Source (PSI) for their great support during data collection. The Structural Genomics Consortium (SGC) is a registered charity (number 1097737) that receives funds from Bayer AG, Boehringer Ingelheim, the Canada Foundation for Innovation, Eshelman Institute for Innovation, Genentech, Genome Canada through Ontario Genomics Institute [OGI-196], EU/EFPIA/OICR/McGill/KTH/Diamond, Innovative Medicines Initiative 2 Joint Undertaking [EUbOPEN grant 875510], Janssen, Merck KGaA (a.k.a., EMD in Canada and USA), Pfizer, the São Paulo Research Foundation-FAPESP, and Takeda. Research in the Oxford laboratory (U.O.) was supported by Cancer Research UK, the LEAN program grant of the Leducq Foundation, Oxford NIHR Biomedical Research Centre, the Oxford Translational Myeloma Centre, and the BMS-Mt Sinai-Oxford Myeloma Single Cell Consortium. J.E.D. is supported by the Oxford NIHR Biomedical Research Centre. Research reported in this publication was supported in part by DoD ALSRP award AL190107, NC Biotechnology Center Institutional Support Grant 2018-IDG-1030, and NIH U24DK116204.

Glossary

Abbreviations

IC50

half-maximal inhibitory concentration

Km

Michaelis constant

NanoBRET

bioluminescence resonance energy transfer using NanoLuciferase

PoC

percent of control

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmedchemlett.2c00530.

  • Supplemental figures (NanoBRET and cell viability assay curves); tables (enzymatic profiling, PAMPA, permeabilized NanoBRET assay, and X-ray crystallographic data); and experimental details, including synthetic schemes and characterization of key compounds, protocols related to selectivity screening and biological assays, and crystallographic details (PDF)

Accession Codes

The PDB accession code for the X-ray cocrystal structure of CK2α + compound 2 is 8BGC.

The authors declare no competing financial interest.

Supplementary Material

ml2c00530_si_001.pdf (1.9MB, pdf)

References

  1. Meggio F.; Pinna L. A. One-thousand-and-one substrates of protein kinase CK2?. FASEB J. 2003, 17 (3), 349–368. 10.1096/fj.02-0473rev. [DOI] [PubMed] [Google Scholar]
  2. Salvi M.; Sarno S.; Cesaro L.; Nakamura H.; Pinna L. A. Extraordinary pleiotropy of protein kinase CK2 revealed by weblogo phosphoproteome analysis. Biochim. Biophys. Acta 2009, 1793 (5), 847–859. 10.1016/j.bbamcr.2009.01.013. [DOI] [PubMed] [Google Scholar]
  3. Rabalski A. J.; Gyenis L.; Litchfield D. W. Molecular pathways: Emergence of protein kinase CK2 (CSNK2) as a potential target to inhibit survival and DNA damage response and repair pathways in cancer cells. Clin. Cancer Res. 2016, 22 (12), 2840–2847. 10.1158/1078-0432.CCR-15-1314. [DOI] [PubMed] [Google Scholar]
  4. Nuñez de Villavicencio-Diaz T.; Rabalski A. J.; Litchfield D. W. Protein kinase CK2: Intricate relationships within regulatory cellular networks. Pharmaceuticals 2017, 10 (1), 27. 10.3390/ph10010027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Ahmed K.; Gerber D. A.; Cochet C. Joining the cell survival squad: An emerging role for protein kinase CK2. Trends Cell Biol. 2002, 12 (5), 226–230. 10.1016/S0962-8924(02)02279-1. [DOI] [PubMed] [Google Scholar]
  6. Bouhaddou M.; Memon D.; Meyer B.; White K. M.; Rezelj V. V.; Correa Marrero M.; Polacco B. J.; Melnyk J. E.; Ulferts S.; Kaake R. M.; Batra J.; Richards A. L.; Stevenson E.; Gordon D. E.; Rojc A.; Obernier K.; Fabius J. M.; Soucheray M.; Miorin L.; Moreno E.; Koh C.; Tran Q. D.; Hardy A.; Robinot R.; Vallet T.; Nilsson-Payant B. E.; Hernandez-Armenta C.; Dunham A.; Weigang S.; Knerr J.; Modak M.; Quintero D.; Zhou Y.; Dugourd A.; Valdeolivas A.; Patil T.; Li Q.; Hüttenhain R.; Cakir M.; Muralidharan M.; Kim M.; Jang G.; Tutuncuoglu B.; Hiatt J.; Guo J. Z.; Xu J.; Bouhaddou S.; Mathy C. J. P.; Gaulton A.; Manners E. J.; Félix E.; Shi Y.; Goff M.; Lim J. K.; McBride T.; O’Neal M. C.; Cai Y.; Chang J. C. J.; Broadhurst D. J.; Klippsten S.; De wit E.; Leach A. R.; Kortemme T.; Shoichet B.; Ott M.; Saez-Rodriguez J.; tenOever B. R.; Mullins R. D.; Fischer E. R.; Kochs G.; Grosse R.; García-Sastre A.; Vignuzzi M.; Johnson J. R.; Shokat K. M.; Swaney D. L.; Beltrao P.; Krogan N. J. The global phosphorylation landscape of SARS-CoV-2 infection. Cell 2020, 182 (3), 685–712.e19. 10.1016/j.cell.2020.06.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Yang X.; Dickmander R. J.; Bayati A.; Taft-Benz S. A.; Smith J. L.; Wells C. I.; Madden E. A.; Brown J. W.; Lenarcic E. M.; Yount B. L.; Chang E.; Axtman A. D.; Baric R. S.; Heise M. T.; McPherson P. S.; Moorman N. J.; Willson T. M. Host kinase CSNK2 is a target for inhibition of pathogenic SARS-like β-coronaviruses. ACS Chem. Biol. 2022, 17, 1937–1950. 10.1021/acschembio.2c00378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Mishra S.; Kinoshita C.; Axtman A. D.; Young J. E. Evaluation of a selective chemical probe validates that CK2 mediates neuroinflammation in an hiPSC-derived microglial model. Front Mol. Neurosci 2022, 15, 824956. 10.3389/fnmol.2022.824956. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Zakharia K.; Miyabe K.; Wang Y.; Wu D.; Moser C. D.; Borad M. J.; Roberts L. R. Preclinical in vitro and in vivo evidence of an antitumor effect of CX-4945, a casein kinase II inhibitor, in cholangiocarcinoma. Transl Oncol 2019, 12 (1), 143–153. 10.1016/j.tranon.2018.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chon H. J.; Bae K. J.; Lee Y.; Kim J. The casein kinase 2 inhibitor, CX-4945, as an anti-cancer drug in treatment of human hematological malignancies. Front Pharmacol 2015, 6, 70. 10.3389/fphar.2015.00070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Wells C. I.; Drewry D. H.; Pickett J. E.; Tjaden A.; Krämer A.; Müller S.; Gyenis L.; Menyhart D.; Litchfield D. W.; Knapp S.; Axtman A. D. Development of a potent and selective chemical probe for the pleiotropic kinase CK2. Cell Chem. Biol. 2021, 28, 546–558.e10. 10.1016/j.chembiol.2020.12.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Krämer A.; Kurz C. G.; Berger B.-T.; Celik I. E.; Tjaden A.; Greco F. A.; Knapp S.; Hanke T. Optimization of pyrazolo[1,5-a]pyrimidines lead to the identification of a highly selective casein kinase 2 inhibitor. Eur. J. Med. Chem. 2020, 208, 112770. 10.1016/j.ejmech.2020.112770. [DOI] [PubMed] [Google Scholar]
  13. Raaf J.; Bischoff N.; Klopffleisch K.; Brunstein E.; Olsen B. B.; Vilk G.; Litchfield D. W.; Issinger O.-G.; Niefind K. Interaction between CK2α and CK2β, the subunits of protein kinase CK2: Thermodynamic contributions of key residues on the CK2α surface. Biochemistry 2011, 50 (4), 512–522. 10.1021/bi1013563. [DOI] [PubMed] [Google Scholar]
  14. Siddiqui-Jain A.; Drygin D.; Streiner N.; Chua P.; Pierre F.; O’Brien S. E.; Bliesath J.; Omori M.; Huser N.; Ho C.; Proffitt C.; Schwaebe M. K.; Ryckman D. M.; Rice W. G.; Anderes K. CX-4945, an orally bioavailable selective inhibitor of protein kinase CK2, inhibits prosurvival and angiogenic signaling and exhibits antitumor efficacy. Cancer Res. 2010, 70 (24), 10288–10298. 10.1158/0008-5472.CAN-10-1893. [DOI] [PubMed] [Google Scholar]
  15. Gowda C.; Sachdev M.; Muthusami S.; Kapadia M.; Petrovic-Dovat L.; Hartman M.; Ding Y.; Song C.; Payne J. L.; Tan B. H.; Dovat S. Casein kinase II (CK2) as a therapeutic target for hematological malignancies. Curr. Pharm. Des 2016, 23 (1), 95–107. 10.2174/1381612822666161006154311. [DOI] [PubMed] [Google Scholar]
  16. Pierre F.; Chua P. C.; O’Brien S. E.; Siddiqui-Jain A.; Bourbon P.; Haddach M.; Michaux J.; Nagasawa J.; Schwaebe M. K.; Stefan E.; Vialettes A.; Whitten J. P.; Chen T. K.; Darjania L.; Stansfield R.; Anderes K.; Bliesath J.; Drygin D.; Ho C.; Omori M.; Proffitt C.; Streiner N.; Trent K.; Rice W. G.; Ryckman D. M. Discovery and SAR of 5-(3-chlorophenylamino)benzo[c][2,6]naphthyridine-8-carboxylic acid (CX-4945), the first clinical stage inhibitor of protein kinase CK2 for the treatment of cancer. J. Med. Chem. 2011, 54 (2), 635–654. 10.1021/jm101251q. [DOI] [PubMed] [Google Scholar]
  17. Battistutta R.; Cozza G.; Pierre F.; Papinutto E.; Lolli G.; Sarno S.; O’Brien S. E.; Siddiqui-Jain A.; Haddach M.; Anderes K.; Ryckman D. M.; Meggio F.; Pinna L. A. Unprecedented selectivity and structural determinants of a new class of protein kinase CK2 inhibitors in clinical trials for the treatment of cancer. Biochemistry 2011, 50 (39), 8478–8488. 10.1021/bi2008382. [DOI] [PubMed] [Google Scholar]
  18. Drewry D. H.; Annor-Gyamfi J. K.; Wells C. I.; Pickett J. E.; Dederer V.; Preuss F.; Mathea S.; Axtman A. D. Identification of pyrimidine-based lead compounds for understudied kinases implicated in driving neurodegeneration. J. Med. Chem. 2022, 65 (2), 1313–1328. 10.1021/acs.jmedchem.1c00440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gray G. K.; McFarland B. C.; Rowse A. L.; Gibson S. A.; Benveniste E. N. Therapeutic CK2 inhibition attenuates diverse prosurvival signaling cascades and decreases cell viability in human breast cancer cells. Oncotarget 2014, 5 (15), 6484–6496. 10.18632/oncotarget.2248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Salvi M.; Borgo C.; Pinna L. A.; Ruzzene M. Targeting CK2 in cancer: a valuable strategy or a waste of time?. Cell Death Discov 2021, 7 (1), 325. 10.1038/s41420-021-00717-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Licciardello M. P.; Workman P. A New Chemical Probe Challenges the Broad Cancer Essentiality of CK2. Trends Pharmacol. Sci. 2021, 42 (5), 313–315. 10.1016/j.tips.2021.02.002. [DOI] [PubMed] [Google Scholar]
  22. Al Quobaili F.; Montenarh M. CK2 and the regulation of the carbohydrate metabolism. Metab., Clin. Exp. 2012, 61 (11), 1512–1517. 10.1016/j.metabol.2012.07.011. [DOI] [PubMed] [Google Scholar]
  23. D’Amore C.; Moro E.; Borgo C.; Itami K.; Hirota T.; Pinna L. A.; Salvi M. Janus” efficacy of CX-5011: CK2 inhibition and methuosis induction by independent mechanisms. Biochim Biophys Acta - Mol. Cell Res. 2020, 1867 (11), 118807. 10.1016/j.bbamcr.2020.118807. [DOI] [PubMed] [Google Scholar]
  24. D’Amore C.; Borgo C.; Sarno S.; Salvi M. Role of CK2 inhibitor CX-4945 in anti-cancer combination therapy – potential clinical relevance. Cell Oncol 2020, 43 (6), 1003–1016. 10.1007/s13402-020-00566-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lertsuwan J.; Lertsuwan K.; Sawasdichai A.; Tasnawijitwong N.; Lee K. Y.; Kitchen P.; Afford S.; Gaston K.; Jayaraman P. S.; Satayavivad J. CX-4945 induces methuosis in cholangiocarcinoma cell lines by a CK2-independent mechanism. Cancers 2018, 10 (9), 283. 10.3390/cancers10090283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Li Z.; Mbah N. E.; Overmeyer J. H.; Sarver J. G.; George S.; Trabbic C. J.; Erhardt P. W.; Maltese W. A. The JNK signaling pathway plays a key role in methuosis (non-apoptotic cell death) induced by MOMIPP in glioblastoma. BMC Cancer 2019, 19 (1), 77. 10.1186/s12885-019-5288-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Manni S.; Carrino M.; Piazza F. Role of protein kinases CK1α and CK2 in multiple myeloma: regulation of pivotal survival and stress-managing pathways. J. Hematol Oncol 2017, 10 (1), 157. 10.1186/s13045-017-0529-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ho C.; Rice R.; Drygin D.; Bliesath J.; Streiner N.; Siddiqui-Jain A.; Proffitt C.; O’Brien S.; Anderes K. CX-4945, a selective and orally bioavailable inhibitor of protein kinase CK2 inhibits PI3K/Akt, JAK-STAT and NF-κB signaling and induces apoptosis in multiple myeloma dells. Blood 2010, 116 (21), 787–787. 10.1182/blood.V116.21.787.787. [DOI] [Google Scholar]
  29. Piazza F. A.; Ruzzene M.; Gurrieri C.; Montini B.; Bonanni L.; Chioetto G.; Di Maira G.; Barbon F.; Cabrelle A.; Zambello R.; Adami F.; Trentin L.; Pinna L. A.; Semenzato G. Multiple myeloma cell survival relies on high activity of protein kinase CK2. Blood 2006, 108 (5), 1698–1707. 10.1182/blood-2005-11-013672. [DOI] [PubMed] [Google Scholar]
  30. Piazza F. A.; Gurrieri C.; Trentin L.; Semenzato G. Towards a new age in the treatment of multiple myeloma. Ann. Hematol 2007, 86 (3), 159–172. 10.1007/s00277-006-0239-5. [DOI] [PubMed] [Google Scholar]
  31. Kovar H.; Jug G.; Hattinger C.; Spahn L.; Aryee D. N. T.; Ambros P. F.; Zoubek A.; Gadner H. The EWS protein is dispensable for Ewing tumor growth. Cancer Res. 2001, 61 (16), 5992–5997. [PubMed] [Google Scholar]
  32. Paronetto M. P. Ewing sarcoma protein: A key player in human cancer. Int. J. Cell Biol. 2013, 2013, 642853. 10.1155/2013/642853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Bayrak O. F.; Aydemir E.; Gulluoglu S.; Sahin F.; Sevli S.; Yalvac M. E.; Acar H.; Ozen M. The effects of chemotherapeutic agents on differentiated chordoma cells. J. Neurosurg Spine 2011, 15 (6), 620–624. 10.3171/2011.7.SPINE10798. [DOI] [PubMed] [Google Scholar]
  34. Xia M.; Huang R.; Sakamuru S.; Alcorta D.; Cho M. H.; Lee D. H.; Park D. M.; Kelley M. J.; Sommer J.; Austin C. P. Identification of repurposed small molecule drugs for chordoma therapy. Cancer Biol. Ther 2013, 14 (7), 638–647. 10.4161/cbt.24596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Ricci-Vitiani L.; Runci D.; D’Alessandris Q. G.; Cenci T.; Martini M.; Bianchi F.; Maira G.; Stancato L.; De Maria R.; Larocca L. M.; Pallini R. Chemotherapy of skull base chordoma tailored on responsiveness of patient-derived tumor cells to rapamycin. Neoplasia 2013, 15 (7), 773–782. 10.1593/neo.13150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Zanin S.; Borgo C.; Girardi C.; O’Brien S. E.; Miyata Y.; Pinna L. A.; Donella-Deana A.; Ruzzene M. Effects of the CK2 inhibitors CX-4945 and CX-5011 on drug-resistant cells. PLoS One 2012, 7 (11), e49193 10.1371/journal.pone.0049193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Siddiqui-Jain A.; Drygin D.; Streiner N.; Chua P.; Pierre F.; O’Brien S. E.; Bliesath J.; Omori M.; Huser N.; Ho C.; et al. CX-4945, an orally bioavailable selective inhibitor of protein kinase CK2, inhibits prosurvival and angiogenic signaling and exhibits antitumor efficacy. Cancer Res. 2010, 70 (24), 10288–10298. 10.1158/0008-5472.CAN-10-1893. [DOI] [PubMed] [Google Scholar]
  38. Kumar S. K.; Rajkumar V.; Kyle R. A.; van Duin M.; Sonneveld P.; Mateos M.-V.; Gay F.; Anderson K. C. Multiple myeloma. Nat. Rev. Dis Primers 2017, 3 (1), 17046. 10.1038/nrdp.2017.46. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ml2c00530_si_001.pdf (1.9MB, pdf)

Articles from ACS Medicinal Chemistry Letters are provided here courtesy of American Chemical Society

RESOURCES