Selective TBK1/IKKi dual inhibitors with anti-cancer potency

Cell culture and materials

Human oral cancer cell lines (SCC-9, SCC-15, SCC-25), human prostate cancer cell lines (PC3, DU145, LNCaP), human breast cancer cell lines (MCF-7, MBA-MB-231, T-47D) and RAW264.7 macrophages, purchased from American Type Culture Collection (ATCC), were maintained in respective medium as recommended in protocols. Myc-CaP mouse prostate cancer cell line, which was isolated from a c-Myc transgenic mouse with prostate cancer, and can grow as tumors in immune competent FVB mice in an androgen dependent manner, was kindly provided by Charles Sawyers, MSKCC. Myc-CaP cells were cultured in RPMI-1640 medium supplemented with 10% heat-inactivated FBS, penicillin-streptomycin 100 units/ml. Lipopolysaccharides (LPS, from Escherichia coli 055:B5) were purchased from Sigma-Aldrich Co. (Saint Louis, Missouri).

TBK1/IKKi dual inhibitors

In counterscreening studies of our in-house 4-phenyl-pyrimidine based JNK inhibitors, we discovered that some JNK inhibitor candidates showed strong TBK1/IKKi inhibition (Patent: Scripps, WO2009032861 and Supplementary Figure 1). After structure modifications and SAR (structure-activity relationship) studies, we successfully generated some potent TBK1/IKKi dual inhibitors (Supplemtary Table 1 and 2). It should be noted that we realized recently that our compounds are structurally similar to those reported in the Dominex patent applications (Domainex, WO2012010826, CAS#1356225-62-1, and CAS#1354224-35-8). For cell culture experiments TBK1/IKKi dual inhibitors were dissolved in DMSO, for animal studies TBK1/IKKi dual inhibitors were dissolved in a mixture of 5%DMSO:10%Tween-80:85%H2O.

RNA interference

Cells at 50% confluency were transfected with 50 nM scrambled small interfering RNA (siRNA), TBK1 specific siRNA, or IKKi specific siRNA (Santa Cruz Biotechnology, Santa Cruz, CA) using TransIT-TKO (Mirus Bio Corp., Madison, WI) according to the manufacturer’s instructions.

MTT assay

Cell viability was examined by using the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazoliumbromide (MTT) colorimetric assay (Molecular Probe, Life Technologies, INC, Grand Island, NY). Briefly, cells were plated in 96-well plates at a density of 3–4 × 10³ cells per well, cells were treated with indicated inhibitors or vehicles for 72 hr, followed by addition of MTT for at least 4 hr. The absorbance was measured at 570/630 nm using a spectrophotometer (SpectraMax M5; Molecular Devices Crop). Cell viability (%) was calculated as follows: $({\text{OD}}_{570}-{\text{OD}}_{630}\text{of}\text{the}\text{samples}/{\text{OD}}_{570}-{\text{OD}}_{630}\text{of}\text{the}\text{control})\times 100(\%)$.

Apoptosis analysis

Apoptosis was examined by using the TUNEL assay - based ApopTag in situ Apoptosis detection kit (Chemicon, Temecula, CA) according to manufacturer’s instructions. Apoptotic cells were counted and compared between different groups.

Western blot analysis

Cell lysates were separated by SDS-PAGE and transferred to PVDF membranes. Immunostaining was performed using antibodies specific for TBK1/NAK (D1B4) Rabbit mAb, IKKi (D61F9) XP Rabbit mAb, Phospho-TBK1/NAK (Ser172) (D52C2) XP Rabbit mAb, Phospho-IKKi (Ser172) (D1B7) Rabbit mAb, IRF-3 (D83B9) Rabbit mAb, Phospho-IRF-3 (Ser396) (4D4G) Rabbit mAb, Phospho-CYLD (Ser418) Antibody, CYLD (D1A10) Rabbit mAb, Cleaved Caspase-9 (Asp330) (D2D4) Rabbit mAb, Cleaved Caspase-3 (Asp175) (5A1E) Rabbit mAb, Cleaved Caspase-7 (Asp198) (D6H1) Rabbit mAb, Cleaved PARP (Asp214) (D64E10) XP Rabbit mAb, AKT (pan) (40D4) Mouse mAb, Phospho-AKT (Thr308) (244F9) Rabbit mAb, Phospho-AKT (Ser473) (D9E) XP Rabbit mAb purchased from Cell Signaling Technology (Danvers, MA) and Cyclin D1 Antibody (M-20), β-actin purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Immunoreactive proteins were detected by chemoluminescence using the Pierce ECL Western Blotting Substrate (Thermo Scientific, Rockford, IL).

Reverse Transcription (RT)-PCR assay

Tumor tissue was homogenized with lysis buffer (RNeasy mini kit, QIAGEN Science, Germantown, MD) on ice. Total RNA was purified using RNeasy mini kit. Reverse transcription was carried out using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems by Life Technologies, Foster City, CA). The sequences of the primers were as follows: for VEGF, 5′-CTACCTCCACCATGCCAAGT-3′ (sense) and 5′-TCTCTCCTATGTGCTGGCCT-3′ (antisense); for β-actin, 5′-CATCGTCACCAACTGGGAC-3′ (sense) and 5′-CGATTTCCCGCTCGGCCGTGG-3′ (antisense).

Mouse tumor models

For SCC-9 xenograft model, eight-weeks old nu/nu mice (Charles River Laboratories) weight at 22–24 g were used. For Myc-CaP allograft model, eight-weeks old FVB male mice (Charles River Laboratories) weight at 22–24 g were used. The mice were housed with a regular 12 hr light/12 hr dark cycle and were kept in a pathogen-free environment. SCC-9 cells (1×10⁷ cells) mixed in Matrigel, were injected into nude mice subcutaneously, Myc-CaP cells (3×10⁶ cells) mixed in Matrigel were inoculated into FVB mice subcutaneously. One day after cell implantation, the mice were randomly divided into two groups (For SCC-9 xenograft model, N = 5 for control group and N=12 for treatment group; for Myc-CaP allograft model, N = 6 for control group and treatment group). Mice were administered ip with TBK1/IKKi kinase dual inhibitor 200A (100 mg/kg) or vehicles once a day. Tumor growth in these mice was monitored and measured every 3 days. The tumor volumes were calculated by the equation V (mm³) = a × b²/2, where “a” is the largest diameter and “b” is the perpendicular diameter.

Immunohistochemistry

Immunohistochemistry was done on SCC-9 xenograft tumor tissue frozen sections using the VECTASTAIN Elite ABC Kit (Universal) (Vector Laboratories, Burlingame, CA). The primary antibodies were mouse monoclonal VEGF (C-1) antibody (SC-7269, Santa Cruz Biotechnology) at 1:200 dilution, rabbit polyclonal phosphor-AKT (Ser 473) antibody (Cell Signaling Technology, Danvers, MA) at 1:200 dilution.

Statistical analysis

All cell MTT data were from at least three independent experiments performed in triplicate and expressed as the mean ± SD. A P-value of <0.05 between experimental and control groups were considered statistically significant. ANOVA with general linear model repeated measures were used to determine tumor volume difference among different groups over treatment time, followed by post-hoc Tukey test. The Student’s t test was also used for univariate analysis. A value of P < 0.05 were considered significant. Immunostaining was expressed as the arithmetic mean ± SD and each evaluated with an unpaired t test. Apoptotic index data were expressed as the mean number ± SD in each tumor area, and nonparametric comparisons (χ²) were made for each treatment group compared with their respective control. A value of P < 0.05 were considered statistically significant

Both TBK1 and IKKi are essential for tumor cell survival

TBK1 and IKKi have been well established as regulators of the innate immune response via their ability to phosphorylate IFN regulatory transcription factors 3 and 7 (IRF3/IRF7). Recent evidence indicates that TBK1 and IKKi are also involved in promoting cell survival and tumorigenesis. To establish whether TBK1 and IKKi are constitutively activated in cancer cells, we checked the phosphorylation levels of TBK1 and IKKi in a number of cancer cell lines. We found that IKKi was expressed and phosphorylated in all cancer cell lines examined while TBK1 was selectively phosphorylated in certain cancer cell lines (Fig. 1A). The expression of p-TBK1 was very low or undetectable by western blot in human oral cancer cell line SCC-25. However, knockdown of IKKi in SCC-25 cells induced both TBK1 and p-TBK1 expression (Fig. 1B), suggesting that inhibition of IKKi leads to a compensatory expression and phosphorylation of TBK1. Consistently, although IKKi is constitutively phosphorylated in SCC-25 cells, knockdown of either IKKi or TBK1 respectively had only minor effects on cell survival while knockdown of both TBK1 and IKKi significantly inhibited cell proliferation (Fig. 1C). These results suggest that both TBK1 and IKKi are essential for cancer cell survival, inhibiting either one is not enough to inhibit cancer cell proliferation. Thus, simultaneously targeting both TBK1 and IKKi is necessary for efficient suppression of cancer cell growth.

Identification of selective TBK1 and IKKi dual inhibitors

To demonstrate that the dual inhibition of TBK1 and IKKi is an effective and safe way to suppress tumor growth, we generated highly potent TBK1/IKKi dual inhibition compounds which are based on a structurally rigid 2-amino-4-(3′-cyano-4′-pyrrolidine)phenyl-pyrimidine scaffold. In counterscreening studies of our in-house 4-phenyl-pyrimidine based JNK inhibitors, we discovered that some of the JNK inhibitor candidates showed strong TBK1/IKKi inhibition (Supplementary Figure 1). After structure modifications and structure-activity relationship (SAR) studies, we successfully developed compounds with significant reduction in anti-JNK activity while keeping a strong TBK1/IKKi inhibition (Supplementary Table 1 and 2). Compounds SR8185, 200A, and 200B exhibited low nanomolar activities against TBK1 and IKKi in enzyme assays and had good polar surface area (PSA) values as well (Fig. 2A and Supplementary Table 1). Counterscreening studies demonstrated that these compounds are selective against ALK, CDK4, CDK5, CK1d, JNK, IKKβ, IKKα, ROCK, PKA, p38 (Supplementary Table 2). Remarkably, SAR studies demonstrated that there was no need for a heavy halo-substitution (iodo- or bromo-) on the molecule in order to obtain high potency and selectivity, and this heavy halo-substitution is required in several published structures for TBK1/IKKi inhibition ^{17, 18}. The inhibition profile of a compound against cytochrome P450 enzymes (CYPs) is also an important property to evaluate in drug development. As shown in Supplementary Table 1, inhibitor 200A showed an activity against four selected P450 isoforms with IC₅₀ values estimated to be > 1000-fold higher than its activity against IKKi/TBK1, which indicates that 200A is a promising lead for further drug discovery efforts. The stability of these inhibitors in human and rat microsomes was also assessed. Among these three inhibitors 200A had the best stability.

To investigate the potency of TBK1/IKKi kinase inhibition of these new compounds in living cells, we tested whether LPS-induced activation of IRF3 in macrophages can be blocked by the treatment of these compounds^{20, 21}. We verified that LPS treatment induced phosphorylation of IRF3 in RAW264.7 cells as expected (Supplementary Figure 2). RAW264.7 cells were then treated with SR8185, 200A, or 200B at different concentrations for 1hr, followed by LPS treatment for 1 hr, cells then were collected for western blot analysis. We found that treatment with these compounds reduced or completely blocked LPS-induced phosphorylation of IRF-3 (Fig. 2B). It was noted that compound 200A displayed the best inhibition. These results suggest that these compounds might exert their cellular activity via inhibiting TBK1/IKKi kinase.

TBK1/IKKi dual inhibitors suppress cancer cell proliferation in vitro

To test the anti-cancer potency of these TBK1/IKKi kinase dual inhibitors, we treated a group of cancer cell lines including oral cancer cells (SCC-9, SCC-15, SCC-25), prostate cancel cells (PC3, DU145, LNCaP, Myc-CaP), and breast cancer cells (MCF-7, MDA-MB-231, T-47D) with different concentrations of these compounds for 3 days, and the cell viability was determined by MTT assay. We found that these TBK1/IKKi kinase dual inhibitors have a potent anti-proliferative effect on all cancer cell lines examined although the sensitivity of cancer cells to these dual inhibitors are various among different cell lines (Fig. 3). These compounds displayed potent cytotoxic activities on cancer cell lines, with IC₅₀ values in low micromolar range (Supplementary Table 3). Among these compounds, 200A presented high anti-proliferative activity in all tested cancer cell lines, with the lowest IC50.

Cyclin D1, whose activity is required for cell cycle G1/S transition, is a well-known marker for cell cycle progression and proliferation. Consistent with the anti-proliferation function, TBK1/IKKi dual inhibitors reduced Cyclin D1 expression in SSC-9 cells in a time-dependent manner (Supplementary Figure 3).

TBK1/IKKi dual inhibitors induce cancer cell apoptosis in vitro

MTT assays showed strong anti-proliferation potency of these TBK1/IKKi kinase dual inhibitors in different cancer cell lines. To further investigate the cytotoxicity of these dual inhibitors, SCC-9 cells were treated with 1.0 μM of TBK1/IKKi kinase dual inhibitor 200A for 72 hr, and then the effect on apoptosis was measured. We found that the treatment of TBK1/IKKi kinase dual inhibitor 200A induced significant apoptosis in SCC-9 cells (Fig. 4A). Consistently, the expression of the pro-apoptotic proteins was significantly increased in a time and dose -dependent manner when SCC-9 cells were exposed to TBK1/IKKi kinase dual inhibitor 200A (cleaved caspase-3 and cleaved PARP: P<0.01) (Fig. 4B and 4C).

TBK1/IKKi dual inhibitors suppress tumor growth in mouse models

Based on the anti-proliferative effect of the TBK1/IKKi dual inhibitors on cancer cells in vitro and other biochemical characteristics, we chose compound 200A to test the anti-tumor growth efficacy of TBK1/IKKi dual inhibitors in mouse models. Nu/nu mice were inoculated subcutaneously (s.c.) in both flanks with 2 × 10⁷ SSC-9 cells. One day after cell inoculation, mice were randomly divided into two groups. One group was treated with TBK1/IKKi dual inhibitor 200A (ip, 100mg/Kg, once per day), and another group was treated with vehicles. Tumor growth was monitored and measured every three days. Tumor volumes were calculated by the equation V (mm³) = a × b²/2, where “a” is the largest diameter and “b” is the perpendicular diameter (Fig. 5A). Thirty-four days later, mice were sacrificed and tumors were collected, weighed (Fig. 5B), and photographed (Fig. 5C). Notably, the treatment of 200A significantly suppressed the growth of SSC-9 xenograft tumor in nude mice.

In another mouse model, we employed the Myc-CaP mouse prostate cancer cell line. This cell line was isolated from a c-Myc transgenic mouse with prostate cancer, and can grow as tumors in immune competent FVB mice in an androgen dependent manner. 3×10⁶ Myc-CaP cells were inoculated s.c. into FVB mice. One day after cell implantation, mice were randomly divided into two groups. One group was administered ip with 200A (100 mg/kg, once a day), and another group treated with vehicles. Tumor growth was monitored and measured every three days (Fig. 5D). Eighteen days later, mice were sacrificed and tumors were collected, weighed (Fig. 5E), and photographed (Fig. 5F). Similar to that seen with the SCC-9 xenograft model described above, the treatment of TBK1/IKKi kinase dual inhibitor 200A significantly slowed down the growth of Myc-CaP allograft tumors in FVB mice.

It should be noted that there were no apparent signs of toxicity in both Nu/Nu and FVB mice treated with 200A, as evidenced by body weight and other life symptom monitoring. Therefore, TBK1/IKKi dual inhibitor 200A showed a safe and potent anti-tumor effect in tumor mouse models. Thus, the data indicate that simultaneous inhibition of both TBK1 and IKKi might be an efficient way to suppress tumor development.

TBK1/IKKi dual inhibitors suppress the expressions of p-CYLD and p-AKT, and angiogenesis

It has been reported that phosphorylation of the tumor suppressor CYLD by IKKi promotes cell transformation ²². To examine whether TBK1/IKKi dual inhibitors suppress CYLD phosphorylation, SCC-9 cells were treated with different TBK1/IKKi dual inhibitors for 1 hr, and cells were collected for western blot. We found that SCC-9 cells treated with TBK1/IKKi dual inhibitors had less p-CYLD expression than that in controls (Fig. 6A).

AKT plays a pivotal role in fundamental cellular functions such as cell proliferation and survival. Recently it was reported that both TBK1 and IKKi phosphorylate AKT ^{14, 15, 23, 24}. To investigate whether TBK1/IKKi dual inhibitors affect AKT phosphorylation, SCC-25 cells were treated with different doses of 200A for 30 min and cells were then collected for western blot. We found that treatment with 200A decreased p-AKT expression in a dose-dependent manner (Fig. 6B). Similarly, western blot and immunohistochemistry analysis showed that treatment with TBK1/IKKi dual inhibitor 200A led to significant reduction of p-AKT expression in xenograft tumor tissues (Fig. 6C and 6D). These results suggest that the anti-tumor effect of TBK1/IKKi dual inhibitors may be via their suppression of AKT activation.

It has been reported that AKT1 regulates pathological angiogenesis, vascular maturation and permeability in vivo ²⁵. Since our results showed TBK1/IKKi dual inhibitors reduced AKT activation in both cancer cell lines and tumor tissues, we asked whether TBK1/IKKi dual inhibitors affect VEGF expression and angiogenesis in tumors. We therefore used RT-PCR and immunohistochemistry to examine VEGF expression in tumors from 200A- treated and control mice. We found that tumor tissues from 200A-treated mice had much less VEGF expression compared with untreated groups (Fig. 6E and 6F). Altogether, these results suggest that the anti-cancer function of TBK1/IKKi dual inhibitors may be mediated by their suppression of AKT activation, leading to impaired angiogenesis and therefore the inhibition of tumor growth.