Discovery of novel TAOK2 inhibitor scaffolds from high-throughput screening

Bioorganic & Medicinal Chemistry Letters 26 (2016) 3923–3927

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl

Discovery of novel TAOK2 inhibitor scaffolds from high-throughput screening Alexander T. Piala a, Radha Akella a, Malia B. Potts b, Stephanie A. Dudics-Giagnocavo b, Haixia He a, Shuguang Wei c, Michael A. White b, Bruce A. Posner c, Elizabeth J. Goldsmith a,⇑ a b c

Department of Biophysics, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-8816, United States Department of Cell Biology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-8816, United States Department of Biochemistry, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-8816, United States

a r t i c l e

i n f o

Article history: Received 24 August 2015 Revised 2 July 2016 Accepted 5 July 2016 Available online 6 July 2016 Keywords: Kinase MAP3K High-throughput screening Autophagy Drug discovery Inhibitor

a b s t r a c t The MAP3K (Mitogen Activated Protein Kinase Kinase Kinase) TAOK2 (Thousand-And-One Kinase 2) is an activator of p38 MAP kinase cascade that is up-regulated in response to environmental stresses. A synthetic lethal screen performed using a NSCLC (non-small cell lung cancer) cell line, and a second screen identifying potential modulators of autophagy have implicated TAOK2 as a potential cancer therapeutic target. Using a 200,000 compound high throughput screen, we identified three specific small molecule compounds that inhibit the kinase activity of TAOK2. These compounds also showed inhibition of autophagy. Based on SAR (structure–activity relationship) studies, we have predicted the modifications on the reactive groups for the three compounds. Ó 2016 Elsevier Ltd. All rights reserved.

TAOK2 is a sterile 20-like kinase and a member of the germinalcenter kinase-like kinase family noted for their N-terminal positioning of the kinase domain.1 Its major isoform is 1235 residues and is ubiquitously expressed.2 Activation of TAOK2 results in stalling of cell division at the G2/M checkpoint.3 The knockdown of kinases downstream of TAOK2 has been shown to sensitize cells to UV insult.4 30% of pancreatic cancers have mutations in the coding region of this kinase, highlighting its potential therapeutic importance.5,6 A synthetic lethal screen performed by the White group to find protein targets synergistic with paclitaxel provided numerous potential drug targets.7 This assay knocked down protein targets via siRNA in an NSCLC line then treated with a low dose of paclitaxel in order to determine which proteins, when inhibited, would increase paclitaxel efficacy. The highest-ranked kinase revealed in this study that operates in a MAPK cascade was TAOK2. Additional insight into the role of TAOK2 in cancer came from a second screen conducted by the White group to identify proteins that regulate autophagy.8 Autophagy is an evolutionarily conserved cell biological process through which cellular components are sequestered within a double membrane and delivered to the

⇑ Corresponding author. Tel.: +1 214 645 6376. E-mail address: [email protected] (E.J. Goldsmith). http://dx.doi.org/10.1016/j.bmcl.2016.07.016 0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.

lysosome for degradation and recycling. Both basal and stressinduced autophagy can play cytoprotective, growth-enhancing, and tumor-promoting roles.9,10 Acute inhibition of autophagy in genetically engineered mouse models of cancer has demonstrated promising therapeutic benefit, but also systemic toxicity.11 This underscores the need to develop therapeutic tools for manipulating autophagy in a tissue-specific manner, which in turn requires a more thorough understanding of the signaling mechanisms that control autophagy during development and disease. In order to study the relationship between TAOK2 and both cancer and autophagy, we performed a high-throughput screen using a 200k compound library to discover novel small-molecule kinase inhibitors (Fig. 1A–D). Phosphorylation reactions were performed in a 384 well format against the generic kinase substrate Myelin Basic Protein (MBP) with the addition of compound. Kinase-Glo (Promega) was then added to each well to detect remaining ATP concentration. Thus, inhibitory compounds would decrease TAOK2 activity, which would increase both remaining ATP concentration and luminescence in those wells. After the initial screen, 1645 compounds were identified with inhibition greater than three r above the mean DMSO control (Fig. 1A). The Z0 factor of the initial screen across all plates was 0.77, indicating high significance12 (Fig. 1B). These compounds were then re-screened in triplicate under the same conditions,

3924

A. T. Piala et al. / Bioorg. Med. Chem. Lett. 26 (2016) 3923–3927

Figure 1. High throughput screen results. (A) A rank-ordered graph of the Z-score of each compound tested in the high-throughput screen. The black dotted line shows the average. The red dotted line is 3r. Compounds above 3r were chosen for subsequent screening (grey). (B) Z0 factor for each plate of the screen is shown. The average Z0 factor across the screen is shown as a red dotted line. (C) %inhibition in the confirmation screen vs. %inhibition in the original HTS screen is plotted to show consistency between screens. (D) The %inhibition of each compound is plotted at 1 lM (cyan), 3.3 lM (red), and 10 lM (purple).

with the compounds being assayed at three concentrations (1 lM, 3.3 lM, and 10 lM, Fig. 1C, D). 432 compounds were identified after the re-screen with average inhibitions greater than 10 r at the 10 lM concentration. The Z0 factor of the confirmation screen was 0.52, again indicating high significance. The best 13 compounds from the re-screen are shown in Figure 2. The prevalence of multiple heterocycles in the compounds is reflective of the composition of the screen (described in Methods in Supplemental Material). These 13 compounds were purchased from Chembridge Inc. and ChemDiv Inc. for measurement of direct binding using Differential Scanning Fluorimetry (DSF). DSF measures protein melt temperature (Tm) with a lipophilic dye (Sypro Orange). As the temperature of the solution is increased, measurements of dye fluorescence are taken. When the protein begins to denature, hydrophobic regions in the protein are exposed, and the dye is able to bind. This binding insulates the dye from the surrounding water, reducing quenching effects and allowing fluorescence. Binding of small molecules to protein has a stabilizing effect, and this is seen by an increase in melt temperature (Table 1). The DTm values ranged from 3.0 to 0.6 °C. The three compounds with the highest positive shifts were SW034538, SW163112, and SW083688, with 3, 2.2, and 2 °C shifts, respectively. The selectivity of each compound that had a >0.5 °C DTm was then determined using DSF against a small panel of other MAPK cascade members (Table 1). The difference between the TAOK2 DTm and that of p38 and ASK1 was determined. The three compounds identified as most potent via DSF also shifted the Tm of TAOK2 more than the controls with TAOK2, though SW083604 had significant cross-reactivity (0.9 °C DTm with respect to p38).

The majority of the other compounds also shifted the Tm of TAOK2 more than the controls. Of the 13 compounds, SW034538, SW172006 and SW083688 were cross-verified for selectivity by performing a specificity screen against 45 kinases commercially at Eurofins Cerep, Inc. (Fig. 3). The results show that compounds SW172006 and SW083688 are highly specific toward TAOK2 while compound SW034538 has cross-reactivity with respect to six of the 45 kinases. SW022326 was tested also but showed little binding or selectivity toward TAOK2 (data not shown). We determined previously that TAOK2 is required for autophagic flux in cancer cells.8,10 siRNA targeting TAOK2 elicited an expression response similar to siRNA targeting known autophagy genes, such as ULK1, a serine/threonine protein kinase which integrates nutrient-responsive signaling to control autophagosome biogenesis downstream of AMPK and mTORC1.8,13 TAOK2 depletion also impaired autophagic flux in human osteosarcoma cell line U2OS engineered to stably express the autophagy marker 1A/1Blight chain 3 fused to GFP (GFP–LC3). LC3 (MAP1LC3A/B) is a mammalian Atg8 homolog that undergoes C terminal cleavage, conjugation to phosphatidylethanolamine, and incorporation into autophagosome membranes where it contributes to cargo recruitment and autophagosome maturation before undergoing degradation upon autophagosome–lysosome fusion.8,14 Here we ask whether chemical inhibition of TAOK2 could similarly inhibit autophagy in this cell line. Indeed, four of 13 compounds tested significantly impaired autophagic flux, as indicated by an increase in accumulation of the GFP–LC3 when added to the media at 50 lM concentration for 24 h (Fig. 4A). We assayed p62 degradation by Western blotting (Fig. 4B). The p62 protein (encoded by the SQSTM1 gene) is a multifunctional protein that can serve as an autophagy adaptor by physically linking ubiquitinated substrates to the autophagosome, leading to degradation of both the substrate and p62 upon autophagosome-lysosome fusion.15 Compounds SW172006, SW022326, and SW083688 each increased the abundance of p62, suggestive of autophagy inhibition. Compound SW034538, despite showing inhibition via GFP– LC3 fusion degradation, did not appear to increase the appearance of p62. Compounds SW034538 and SW083688 were both selective in the DSF measurements and effective in the GFP–LC3 degradation assays. These compounds were then chosen for radiometric analysis to determine their IC50 values (Table 1). TAOK2 was incubated with inhibitor, MBP, and c-labeled ATP and allowed to react for 10 minutes. The concentrations of MBP and ATP were chosen such that the reaction was kept in the linear range for the duration of the experiment. The IC50 for compound SW034538 was found to be 300 nM under these conditions, and the IC50 for SW083688 was found to be 1.3 lM. The compounds SW034538 SW172006 and SW083688, which were effective in both the GFP–LC3 assay and was selective for TAOK2 according to DSF, have derivatives in the UTSW compound library. These compounds were culled from the 200 k screen computationally to obtain preliminary SAR (Fig. S1). IC50 value estimates were calculated from the results of the 200k rescreen described above. In the case of SW034538 (Fig. S1A), methyl and ethyl amides in the R3 position appeared preferred, with methyl amides having the highest potency (0.08 and 0.2 lM IC50s, respectively). All methyl and ethyl amides in the R3 position that were tested were inhibitive. Bulky hydrophobic groups in this position abolished compound activity. Amines in this position had much weaker binding (>1.7 lM IC50). The R1, R2, and R4 positions were broadly tolerant. These results suggest that the thiazole ring containing the R3 moiety forms the greater

3925

A. T. Piala et al. / Bioorg. Med. Chem. Lett. 26 (2016) 3923–3927 O NH N

H N

S

H S

O N

O

O

N

HN

O

N

N N

SW034538

SW034513 HO

S

SW166693

O

O

N

O

N

O

H N

S

NO 2

NH N H

H N

S

N H

S

N H

H N

S

N H

N

O O

O

SW109820

SW083688

SW163112 OH

Br O

O

S H N

O

O

N O

O

N N H

S

N H

S

SW083604 O

H N

S HN

O

SW164826 O S O

H N

O

N

Br

SW106178 N

H N

O

HO

O

O N

O O N H

O

Cl

SW172006

SW145091

SW131291

O N H Br

N O O

SW022326 Figure 2. TAOK2 inhibitors discovered from 200k compound library. 13 previously unidentified TAOK2 inhibitors are disclosed. IC50 values determined from the screen ranged from 0.3 to 6 lM.

Table 1 The change in melting temperature Tm (DTm) on binding compounds to TAOK2 is shown. The difference in DTm between TAOK2 and the control kinases p38 and ASK1 induced by the compounds (DDTm) is listed in column 2 and 3. IC50’s were estimated as described in the text, except ⁄ compound IC50’s were determined radiometrically Compound

DTm °C

DDTm °C (p38)

DDTm °C (ASK1)

IC50

SW083604 SW034538 SW083688 SW022326 SW172006 SW163112 SW106178 SW109820 SW131291 SW164826 SW166693 SW145091 SW034513

1.2 3.0 2.0 0.6 1.3 2.2 1.6 0.5 1.2 0.2 0.7 1.6 0.1

0.3 2.8 2.0 — 2.0 1.5 1.3 — 2.0 — 0.0 2.2 —

1.1 2.7 2.1 — 1.2 1.7 1.2 — -0.0 — 0.4 1.2 —

600 nM 300 nM* 1.3 lM* 4 lM 5 lM 3 lM 6 lM 1 lM 3 lM 3 lM 5 lM 2 lM 3 lM

interaction surface. For SW172006 (Fig. S1B), there is no data for changes in the sulfate group at R1. On the other hand, R2 can be modified with other ring systems. In SW083688 (Fig. S1C), R1 can be substituted with a flat ring system. R2 can be modified with small bulky groups including five- or six-membered rings. Three new, potent, inhibitors of the MAP3K TAOK2 have been identified in this screen. These compounds are not only active in biochemical assays, but are also capable of inhibiting the action of TAOK2 in cells. Use of these inhibitors has revealed TAOK2 is a critical player in autophagy, and confirms the observations made in the 2013 autophagy screen conducted by Potts et al.8 Both the synthetic lethal screen conducted by Whitehurst et al. and the link to autophagy suggest that TAOK2 may be an anticancer target.7 The compounds identified in this screen will be vital in further research regarding the function of TAOK2 in cells, as well as provide potential scaffolds for the production of pharmacologically useful anticancer agents.

3926

A. T. Piala et al. / Bioorg. Med. Chem. Lett. 26 (2016) 3923–3927

Inhibition (%)

100

SW034538

50

0

Abl Akt1/PKBalpha AurA/Aur2 CaMK2alpha CDC2/CDK1 CDK2 CHK1 CHK2 c-Met EGFR EphA2 EphA3 EphB4 ERK2 FGFR1 FGFR2 FGFR3 GSK3beta HGK IKKalpha IRAK4 IRK JAK3 JNK1 KDR Lck MAPKAPK2 MARK1 MNK2 NEK2 p38alpha PAK4 PDK1 Pim2 PKA PKCbeta 2 PLK1 RAF-1 ROCK1 SGK1 SIK Src TAO2 TRKA

-50

100

SW172006

Inhibition (%)

50

0

Abl Akt1/PKBalpha AurA/Aur2 CaMK2alpha CDC2/CDK1 CDK2 CHK1 CHK2 c-Met EGFR EphA2 EphA3 EphB4 ERK2 FGFR1 FGFR2 FGFR3 GSK3beta HGK IKKalpha IRAK4 IRK JAK3 JNK1 KDR Lck MAPKAPK2 MARK1 MNK2 NEK2 p38alpha PAK4 PDK1 Pim2 PKA PKCbeta 2 PLK1 RAF-1 ROCK1 SGK1 SIK Src TAO2 TRKA

-50

Inhibition (%)

100

SW083688

50

0

Abl Akt1/PKBalpha AurA/Aur2 CaMK2alpha CDC2/CDK1 CDK2 CHK1 CHK2 c-Met EGFR EphA2 EphA3 EphB4 ERK2 FGFR1 FGFR2 FGFR3 GSK3beta HGK IKKalpha IRAK4 IRK JAK3 JNK1 KDR Lck MAPKAPK2 MARK1 MNK2 NEK2 p38alpha PAK4 PDK1 Pim2 PKA PKCbeta 2 PLK1 RAF-1 ROCK1 SGK1 SIK Src TAO2 TRKA

-50

B

8

Acknowledgments

***

6

p62 ** ****

****

2 ***

SW106178 SW131291 SW163112 SW164826 SW166693 SW172006 SW145091 SW022326 SW034513 SW034538 SW083604 SW083688

0

This work was supported by the Cancer Prevention and Research Institute of Texas grants RP100941 to EJG and RP110595 MAW, the Welch Foundation grant I1128 to EJG.

GFP/ LC3 XPB DMSO SW145091 SW022326 SW172006 SW083688 SW034538

***

4

DMSO

A

GFP-LC3 Accumulation

Figure 3. Specificity screen for TAOK2 inhibitors discovered from 200k compound library. Of the 13 inhibitors, 3 were tested for cross-reactivity against 45 off-target kinases. Kinases showing inhibition higher that 50% (dotted line) are considered significant.

Figure 4. TAOK2 inhibition prevents cellular autophagy. (A) U2OS cells stably expressing the GFP–LC3 fusion transgene were treated with the indicated compound at 50 lM concentration for 24 h before GFP fluorescence was measured by flow cytometry. The median fluorescent value from 20,000 cells per condition was normalized to that of vehicle alone (DMSO) within a given experiment. The mean + SD of >2 independent experiments is shown. ⁄⁄⁄⁄, p < 0.0001; ⁄⁄⁄, p < 0.001; ⁄⁄, p < 0.01; ns, p > 0.05 versus vehicle alone (DMSO) by unpaired t test. (B) TAOK2 inhibitors induce accumulation of p62. U2OS GFP–LC3 cells were treated with the indicated compounds at 50 lM concentration, or DMSO alone, for 24 h. Lysates were collected and analyzed by western blot using the indicated antibodies. XPB was used as a loading control.

Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.07. 016. References and notes 1. Chen, Z.; Hutchison, M.; Cobb, M. H. J. Biol. Chem. 1999, 274, 28803. 2. Moore, T. M.; Garg, R.; Johnson, C.; Coptcoat, M. J.; Ridley, A. J.; Morris, J. D. J. Biol. Chem. 2000, 275, 4311. 3. Raman, M.; Earnest, S.; Zhang, K.; Zhao, Y.; Cobb, M. H. EMBO J. 2007, 26, 2005. 4. Manke, I. A.; Nguyen, A.; Lim, D.; Stewart, M. Q.; Elia, A. E.; Yaffe, M. B. Mol. Cell 2005, 17, 37. 5. Gao, J.; Aksoy, B. A.; Dogrusoz, U.; Dresdner, G.; Gross, B.; Sumer, S. O.; Sun, Y.; Jacobsen, A.; Sinha, R.; Larsson, E.; Cerami, E.; Sander, C.; Schultz, N. Sci. Signal. 2013, 6, pl1.

A. T. Piala et al. / Bioorg. Med. Chem. Lett. 26 (2016) 3923–3927 6. Cerami, E.; Gao, J.; Dogrusoz, U.; Gross, B. E.; Sumer, S. O.; Aksoy, B. A.; Jacobsen, A.; Byrne, C. J.; Heuer, M. L.; Larsson, E.; Antipin, Y.; Reva, B.; Goldberg, A. P.; Sander, C.; Schultz, N. Cancer Discovery 2012, 2, 401. 7. Whitehurst, A. W.; Bodemann, B. O.; Cardenas, J.; Ferguson, D.; Girard, L.; Peyton, M.; Minna, J. D.; Michnoff, C.; Hao, W.; Roth, M. G.; Xie, X. J.; White, M. A. Nature 2007, 446, 815. 8. Potts, M. B.; Kim, H. S.; Fisher, K. W.; Hu, Y.; Carrasco, Y. P.; Bulut, G. B.; Ou, Y. H.; Herrera-Herrera, M. L.; Cubillos, F.; Mendiratta, S.; Xiao, G.; Hofree, M.; Ideker, T.; Xie, Y.; Huang, L. J.; Lewis, R. E.; MacMillan, J. B.; White, M. A. Sci. Signal. 2013, 6, ra90. 9. Perez, E.; Das, G.; Bergmann, A.; Baehrecke, E. H. Oncogene 2015, 34, 3369.

3927

10. Choi, A. M.; Ryter, S. W.; Levine, B. New Engl. J. Med. 2013, 368, 1845. 11. Karsli-Uzunbas, G.; Guo, J. Y.; Price, S.; Teng, X.; Laddha, S. V.; Khor, S.; Kalaany, N. Y.; Jacks, T.; Chan, C. S.; Rabinowitz, J. D.; White, E. Cancer Discovery 2014, 4, 914. 12. Zhang, J. H.; Chung, T. D.; Oldenburg, K. R. J. Biomol. Screen. 1999, 4, 67. 13. Chan, E. Y.; Kir, S.; Tooze, S. A. J. Biol. Chem. 2007, 282, 25464. 14. Kabeya, Y.; Mizushima, N.; Ueno, T.; Yamamoto, A.; Kirisako, T.; Noda, T.; Kominami, E.; Ohsumi, Y.; Yoshimori, T. EMBO J. 2000, 19, 5720. 15. Pankiv, S.; Clausen, T. H.; Lamark, T.; Brech, A.; Bruun, J. A.; Outzen, H.; Overvatn, A.; Bjorkoy, G.; Johansen, T. J. Biol. Chem. 2007, 282, 24131.