MCL-1 Degradation Mediated by JNK Activation via MEKK1/TAK1-MKK4 Contributes to Anticancer Activity of New Tubulin Inhibitor MT189

In recent years, colchicine-binding site–targeted tubulin inhibitors have become a promising type of anticancer drugs with several of them entering into clinical trials (1). We previously constructed a combinatorial library of 6H-Pyrido[2′,1′:2,3]imidazo [4,5-c]isoquinolin-5(6H)-ones (2) from which two new tubulin inhibitors, MT7 and its structurally-modified antiproliferation-activity-improved derivative MT119, were obtained. Both cause microtubulin depolymerization, persistent M phase arrest, and apoptosis. MT119 was confirmed to bind to the colchicine site on tubulin (3–5). Although both produce potent in vitro antiproliferative effects and overcome tumor multidrug resistance (MDR), they did not show obvious in vivo anticancer activity. In considering their structural novelty and potential new modes of action, we put persistent efforts into further structural optimization on MT119. Finally, we acquired a new colchicine site–binder 2-(6-fluoro-3-((4-methoxybenzyl)amino)imidazo[1,2-α] pyridin-2-yl) phenol (designated as MT189; Supplementary Fig. S1).

Tubulin inhibitors, including those targeting the colchicine site, can disrupt cellular microtubulin assembly, followed by persistent M arrest and subsequent apoptosis. Both M arrest and apoptosis contribute to their anticancer effects (3, 4, 6). Besides, other mechanisms of action and/or signaling pathways might be involved. We previously found that the naturally occurring colchicine site–binder pseudolaric acid B (7) could activate the JNK–c-Jun–hypoxia inducible factor-1α (HIF-1α) signaling axis to inhibit angiogenesis and overcome tumor MDR (8, 9). JNK can be activated by its upstream mitogen-activated protein kinase kinase kinases (MAP3K; at least 14 kinases, including MEKK1 and TAK1) through MAP kinase kinases (MAP2K; MKK4 and MKK7). Both MKK4 and MKK7 phosphorylate JNK but MKK4 also activates p38 (10). However, whether the new colchicine site–binder MT189 could also activate JNK, and if so, which upstream kinases are involved, remains to be clarified. In addition, the relationship between microtubulin disruption, mitotic arrest, and JNK activation triggered by MT189 needs investigating.

Besides c-Jun, JNK has many other substrates, one of which is the antiapoptotic protein MCL-1 (11). MCL-1 is a member of the BCL-2 protein family. It promotes survival via inhibiting apoptosis by suppressing proapoptotic proteins BAX and BAK (11, 12). Antitubulin agents paclitaxel and vincristine that bind to the taxane site and the vinca site on tubulin, respectively, have been shown to cause phosphorylation of MCL-1 via JNK, which subsequently leads to ubiquitination-mediated MCL-1 degradation (11). So, it is intriguing whether the colchicine site–binder MT189 also affects MCL-1 via JNK, and if so, what effects will be caused on its anticancer activity.

In this study, we first examined the effects of MT189 on microtubulin dynamics, cell cycle, and apoptosis to determine whether it retains the characteristics of its parental compound MT119. Next, its in vitro and in vivo activities against tumor cells were measured. After that, we further tested the impacts of MT189 on MCL-1 and the JNK signaling pathway. Finally, whether MT189-induced microtubule disruption is directly related to its activation of the JNK signaling axis was explored.

MT119 was prepared as reported previously (4). MT189 was obtained through a deconstruction approach to break the tetracyclic ring of MT-119 (Supplementary Fig. S1). Their purity (≥99%) was determined by reverse phase-HPLC at two wavelengths of 214 nm and 254 nm. Paclitaxel, combretastatin A4 (CA4), adriamycin (ADR), SP600125, and PD169316 were purchased from Sigma-Aldrich. RO3306 was obtained from Tocris Bioscience. JNK-IN-8 was purchased from Merck Millipore. Z-VAD-FMK was obtained from Beyotime. Colchicine was obtained from Sangon. Vincristine (VCR) was obtained from J & K Scientific Ltd. (Shanghai, China). BODIPY FL vinblastine was purchased from Invitrogen. All the antibodies were commercially available, including those respectively against histone H3, p-histone H3, JNK, MCL-1, MEKK1, and p-MEKK1 from Santa Cruz Biotechnology; p38, p-p38, p-MCL-1, MKK4, p-MKK4, MKK7, p-MKK7, TAK1, and p-TAK1 from Cell Signaling Technology; p-JNK from Abcam; β-Actin from Abgent; MPM-2 from Merck Millipore; and α-tubulin from Invitrogen.

Human cancer A549 (2002), HT-29 (2003), HCT116 (2003), KB (2003), MDA-MB-468 (2004), HeLa (2005), PC-3 (2006), and DU145 (2010) cell lines and human normal lung fibroblast WI-38 cell line (2013) were purchased from the American Type Culture Collection (ATCC). Human cancer SGC-7901 (2002), BEL-7402 (2002), SPC-A4 (2003), SMMC-7721 (2006), and MDA-MB-231 (2009) cell lines were kept in the Shanghai institute of Materia Medica of the Chinese Academy of Sciences (Shanghai, China). MCF-7 (2002), MKN-45 (2003), and SKOV-3 (2007) were obtained from the Japanese Foundation for Cancer Research (Tokyo, Japan). The adriamycin-selected resistant MCF7/ADR cell line was purchased from the Institute of Hematology, Chinese Academy of Medical Sciences (Tianjin, China) in 2003 (4, 13, 14). The vincristine-selected resistant KB/VCR cell line was purchased from the Sun Yat-Sen University of Medical Sciences (Guangzhou, China) in 2003 (4, 13, 14). All the other cell lines except WI-38 (obtained in December 2013) were authenticated by short tandem repeat (STR) profiling in Beijing Microread Gene Technology Co, Ltd. (HeLa and DU145, July 2012) or in Shanghai Genesky Biotech Co., Ltd. (MDA-MB-231, MCF-7, and MCF7/ADR, March 2013; HT-29, KB, MDA-MB-468, and KB/VCR, May 2013; A549 and SGC-7901, June 2013; HCT116, SPC-A4, SMMC-7721, and BEL-7402, July 2013; PC-3 and SKOV-3, August 2013; and MKN-45, November 2013). The cells were also periodically authenticated by morphologic inspection and tested Mycoplasma contamination. The cell lines were cultured according to the suppliers' instructions.

Sulforhodamine B (SRB) assays were done as reported previously (4) to measure the IC₅₀ values of different agents in the cells exposed to their gradient concentrations for 72 hours. Cell counting kit-8 (CCK-8) assays were performed to determine the viability of the cells treated with MT189 for shorter than 72 hours. For this purpose, cells were seeded into 96-well plates, cultured overnight, and treated with MT189 for the indicated time. Then, 10 μL CCK-8 solution (Dojindo Laboratories) was added into 100 μL culture medium in each well according to the manufacturer's instructions, and the cells were incubated for 4 hours at 37°C. The absorbance at 450 nm was measured with spectra-MAX190 (Molecular Devices). The percentage of cell proliferation inhibition was calculated as: proliferation inhibition (%) = [1 − (A450_treated/A450_control)] × 100%. The averaged IC₅₀ values were determined with the Logit method from three independent tests.

Western blotting was performed as previously described (8).

Cells were seeded on glass coverslips, cultured for 24 hours, and treated with appropriate agents for the indicated time. Then, the cells were fixed for 15 minutes with 4% paraformaldehyde, permeabilized for 5 minutes with 0.2% Triton X-100, blocked with 3% bovine serum albumin for 15 minutes, incubated with the primary antibody for 1 hour, and stained with fluorescence-conjugated secondary antibody for 1 hour. Finally, after being counterstained with 4′,6-diamidino-2-phenylindole (DAPI), the cells were imaged under an Olympus confocal microscope (Olympus).

Tubulin was purchased from Cytoskeleton. In vitro tubulin polymerization was assessed by the turbidity assay as previously described (4).

The cells were treated with the given agent and harvested in the lysis buffer (100 mmol/L PIPES, pH 6.9, 1 mmol/L EGTA, 1 mmol/L MgCl₂, 30% glycerol, 5% DMSO, 1% NP-40, 5 mmol/L GTP, and protease inhibitors). The polymerized tubulin fraction (precipitate) and the soluble tubulin fraction (supernatant) were separated by centrifugation at 180,000 ×g at 37°C for 1 hour. Finally, α-tubulin from equal aliquots of the polymerized tubulin fraction and the soluble tubulin fraction was determined by Western blotting (4).

Tubulin (3 μmol/L) was incubated with the tested agents at 37°C for 1 hour. Then, colchicine or BODIPY FL vinblastine was added to the final concentration of 3 μmol/L. The fluorescence was determined after 30 minutes at 37°C with a PerkinElmer fluorometer (4, 15).

Cells were seeded into 6-well plates, cultured overnight, and treated with MT189 for the indicated time. To examine the reversibility of cell-cycle arrest, the cells treated with 0.2 μmol/L MT189 for 12 hours were washed with medium 3-times and then incubated in drug-free full medium for the indicated time. Cells were then harvested and washed with PBS and fixed with precooled 70% ethanol at 4°C. Staining went along in PBS containing 40 μg/mL RNase A and 10 μg/mL propidium iodide in the dark for 30 minutes. For each sample, at least 1 × 10⁴ cells were collected with FACSCalibur (BD Biosciences) and analyzed using the CELLQUEST software (BD Biosciences).

Cells were seeded into 6-well plates, cultured overnight, and treated with MT189. Then, cells were fixed with 4% paraformaldehyde, permeabilized with 0.2% Triton X-100, and washed with PBS. TUNEL assays were performed with an In Situ Cell Death Detection Kit (Roche Applied Science) according to the manufacturer's instructions.

Cells were seeded into 6-well plates, cultured overnight and treated with different agents. Then, cells were harvested, washed, and stained by using the Annexin V-FITC Apoptosis Detection Kit (Beyotime). Fluorescence of the cells was determined immediately by flow cytometry (BD Biosciences).

Cells were seeded into 6-well plates, cultured overnight, and treated with MT189. Then, cells were harvested, washed, and stained by using JC-1 kit (Beyotime). Fluorescence of the cells was determined immediately by flow cytometry (BD Biosciences).

The plasmid expressing wild-type MCL-1 (#25392) was obtained from Addgene. The transfection was conducted as reported previously (8).

MCL-1 expression was knocked down with specific siRNA duplexes from RiboBio (catalog no. Q000004170-1-A). The transfection of siRNAs was carried out with the RNAiMAX Transfection Reagent (Invitrogen) according to the manufacturer's instructions.

Total RNA prepared with the TRIzol reagent (Invitrogen) was reverse transcribed into cDNA with the PrimeScript RT Reagent Kit (TaKaRa). cDNA was amplified with the SYBR Premix EX Taq II Kit (TaKaRa) in a 7500 Fast Real-Time PCR System (Applied Biosystems). The PCR program was as follows: 95°C, 30 seconds; 40 cycles (for each cycle 95°C, 5 seconds; 64°C, 20 seconds; 72°C, 15 seconds); and 72°C, 10 minutes. All primers were synthesized by Sangon as follows: (5′-3′) GGGCAGGATTGTGACTCTCATT (forward) and GATGCAGCTTTCTTGGTTTATGG (reverse) for MCL-1; TCTACAATGAGCT GCGTGTG (forward) and GGTGAGGATCTTCAT-GAGGT (reverse) for ACTB.

Cells treated with MT189 for 24 hours were lysed with the NP40 lysis buffer (Beyotime). Then, followed by centrifugation at 12,000 ×g at 4°C for 5 minutes, the supernatant was precleared with 20 μL protein A/G agarose beads at 4°C for 2 hours. After that, centrifugation was done at 1,000 ×g at 4°C for 3 minutes and the resulting supernatant was exposed to the indicated antibodies for 4 hours followed by the addition of 20 μL protein A/G agarose overnight at 4°C. When harvested, the beads were washed 3-times with 1 mL NP40 lysis buffer or PBS for 10 minutes each. The precipitates were dissolved with the SDS loading buffer for Western blotting.

A subcutaneous MDA-MB-231 xenograft model in nude mice was used to evaluate the anticancer activity of the tested compounds. The model was established by the transplantation of 2 × 10⁶ MDA-MB-231 cells subcutaneously on the back of each 6-week-old female Balb/c nude mouse. When the average tumor volume reached around 150 mm³, 24 nude mice were selected on the basis of the tumor volume and randomly assigned into three treatment groups (8 per group). For 2 weeks, these animals were administered intraperitoneally with MT189 (100 mg/kg/d) or with vehicle. During the treatment period, the implanted tumors were measured by caliper twice a week in a blind fashion. The tumors were measured for the maximum width (X) and length (Y), and the tumor volumes (V) were calculated using the formula V = (X²Y)/2. Then relative tumor volume (RTV) was calculated as follows: RTV = V_t/V₀. V₀ was the tumor volume before treatments, and V_t was the tumor volume after treatments. The in vivo anticancer activity was judged by their T/C (%) values. T/C (%) was calculated as (T_RTV/C_RTV) × 100%, where T_RTV represented the RTV of the treatment group and C_RTV represented the RTV of the vehicle group. The animal body weights were also measured at the same time. The animal use and experimental protocols were reviewed and approved by the Animal Care and Use Committee of the Medicilon Company.

We previously reported that MT7 (3) and its efficacy-improved derivative MT119 (4) were new colchicine site–binding tubulin inhibitors. Both are easy to synthesize and MT119 is also active even in multidrug-resistant tumor cells. Disappointingly, however, both of them lack in vivo anticancer activities, possibly due to their relative weak bioactivity. So, based on MT119, continuous structural optimization has been carried out. Among other efforts, one deconstruction approach was taken to break the tetracyclic ring of MT119 to a new intramolecular hydrogen bonding scaffold. A phenolic group was introduced to the α position of phenyl ring to form a pseudoring through the intramolecular hydrogen bonding between the imidazole nitrogen and the phenolic hydrogen atom to lock the planar conformation and to mimic the original tetracyclic ring space in MT119 (Supplementary Fig. S1). This strategy produced a new series of compounds, in which a compound designated as MT189 (Supplementary Fig. S1) showed an averaged 8.7-fold enhancement in the in vitro proliferation inhibition against a panel of six different tissue-derived cancer cells when compared with its parental compound MT119 (Supplementary Fig. S2).

To characterize MT189, we tested whether it retained the property of MT119 in impairing microtubule assembly and arresting cell cycle. The classic microtubule stabilizer paclitaxel promoted microtubulin polymerization as expected. In contrast, MT189 inhibited tubulin polymerization in a concentration-dependent manner, similar to the microtubule destabilizer colchicine (Fig. 1A). In addition, MT189 competitively reduced the binding of colchicine to tubulin as the positive control CA4 (a known colchicine site–binder) did. However, MT189 did not affect the fluorescence intensity of BODIPY FL vinblastine–tubulin complexes that was apparently lowered by VCR (Fig. 1B). The data indicate that MT189 binds to tubulin at the colchicine site rather than the vinblastine site.

In interphase cells, microtubules radiate from the microtubule-organizing center located at the centrosome in the cytoplasm, involved in the maintenance of cell shape (16). The treatment of SMMC-7721 cells with MT189 resulted in scattered tubulin staining, and long microtubule fibers could rarely be observed in the cell (Fig. 1C). Moreover, MT189-triggered microtubule depolymerization occurred very early (within 0.5 hours) and then persistently progressed (Supplementary Fig. S3A). In addition, MT189 caused an increase in the free microtubulin fraction (supernatant) and simultaneously a decrease in the polymerized microtubulin fraction (precipitate) in the cell, although the levels of total tubulin kept unchanged (Fig. 1D). The data indicate that MT189 can disrupt the microtubule assembly by inhibiting tubulin polymerization in the cell.

MT189 was further shown to lead to a typical, reversible G₂–M arrest in SMMC-7721 cells in a time- and concentration-dependent manner (Fig. 1E and Supplementary Fig. S3B). The enhanced levels of the two mitotic markers MPM-2 and Ser10-phosphorylated histone 3 (ref. 17; Fig. 1F) in MT189-treated SMMC-7721 cells indicated that MT189 caused cell-cycle arrest in M rather than G₂ phase. During mitosis, cellular microtubules are normally assembled into bipolar mitotic spindles to guarantee equal segregation of the chromosomes (refs. 18, 19; Fig. 1G). However, MT189 resulted in multipolar spindle damage and scattered chromosome misalignments in SMMC-7721 cells (Fig. 1G). The data indicate that MT189 interferes with mitosis progression by disrupting mitotic spindle assembly.

All the above results show that MT189 retains the property of its parental compound MT119 in impairing microtubule assembly (4).

Persistent mitotic arrest caused by microtubule depolymerization has been demonstrated to trigger apoptosis (20, 21). So, we investigated whether and how MT189 induced apoptosis. The treatment with MT189 led to a progressive increase in apoptosis in a time-dependent manner in SMMC-7721 cells (Fig. 2A). MT189 also caused apparent loss of mitochondrial membrane potential (MMP) in these cells in a concentration-dependent manner (Figs. 2B and C). Apoptosis can roughly be classified into two types: caspase-8–mediated extrinsic apoptosis and caspase-9–initiated intrinsic apoptosis (22, 23). MT189 caused the cleavage of PARP, caspase-3, and caspase-9 rather than caspase-8 in a time- and concentration-dependent manner (Fig. 2D), indicating that it mediates mitochondria-involved caspase-9–initiated intrinsic apoptosis.

MT189 displayed potent proliferation inhibition in 14 cancer cell lines derived from different tissues of origin with an averaged IC₅₀ value of 0.1 μmol/L ranging from 0.01 μmol/L (HeLa) to 0.3 μmol/L (DU145; Fig. 2E) in contrast with its relatively weak activity in the normal WI-38 cell line (proliferation was inhibited by 35.09% ± 7.30% at 10 μmol/L). The data indicate that MT189 possesses a broad in vitro antiproliferative spectrum and potential selectivity over normal cells. Moreover, the in vitro antiproliferative activity of MT189 was time- and concentration-dependent, as shown in HeLa and SMMC-7721 cells (Supplementary Fig. S4). As expected, MT189 could also overcome MDR (Table 1), indicating that MT189, in this regard, also retains the excellent feature of its parental compound MT119 (4). With its enhanced in vitro antiproliferative activity as compared with MT119, MT189 elicited in vivo anticancer effects by inhibiting the growth of MDA-MB-231 xenografts in nude mice by 35.9% over 14 days but did not obviously affect the body weight of the tested animals during the same period (Fig. 2F).

Notably, MT189 caused M arrest much earlier than apoptosis. Treatments with 0.2 μmol/L MT189 took only 3 hours to arrest the cell cycle in M phase (Fig. 1E and F), but about 36 hours to result in apoptosis (Fig. 2A and D). The results suggest that apoptosis might result from persistent mitotic arrest in these cells. Apparently, growth inhibition in mitosis followed by apoptosis elicits the anticancer effect of MT189.

The BCL-2 family proteins are key regulators of apoptosis by regulating mitochondrial membrane permeabilization. This family consists of two types of members: antiapoptotic ones such as BCL-2, BCL-XL, and MCL-1 and proapoptotic ones such as BAK and BAX. The change in the dynamic balance of those members may result in either inhibition or promotion of apoptosis (24). MT189 was revealed to reduce the levels of MCL-1 protein in a time-dependent manner, but did not obviously change the protein levels of the other tested members (Fig. 3A). Moreover, the caspase inhibitor Z-VAD-FMK did not prevent MCL-1 reduction but did prevent PARP cleavage induced by MT189 (Supplementary Fig. S5A). The results suggest that MCL-1 might be associated with MT189-induced apoptosis.

To clarify whether MT189-induced MCL-1 reduction contributes to its in vitro proliferation-inhibitory activity, we compared its effect on the levels of MCL-1 protein in three relative sensitive cells (SMMC-7721, HT-29, and HeLa) with those in two relative insensitive cells (DU145 and SKOV-3). MT189 was 11.8-times more potent in SMMC-7721, HT-29, and HeLa cells (with an averaged IC₅₀ of 0.025 μmol/L) than in DU145 and SKOV-3 (with an averaged IC₅₀ of 0.295 μmol/L; Fig. 3B). The treatment with MT189 for 24 hours led to reduction of MCL-1 protein levels in a concentration-dependent manner in all three relative sensitive cell lines, but did not obviously change the levels of MCL-1 protein in the insensitive cell lines (Fig. 3B). Nevertheless, MCL-1 protein levels in DU145 cells also decreased in a concentration-dependent manner when the treatment time with MT189 was prolonged to 48 hours (Supplementary Fig. S5B). However, MT189 induced apparently less apoptosis in DU145 cells than in HeLa cells at the same conditions (Supplementary Fig. S5C). Moreover, the ectopic overexpression of MCL-1 attenuated (but did not abolish) the cleavage of caspase-9, caspase-3, and PARP (Fig. 3C) and the loss of viable cells (Fig. 3D) caused by MT189 in the pWT-MCL-1–transfected SMMC-7721 cells. In contrast, downregulation of MCL-1 protein potentiated these effects in the MCL-1-siRNA (siMCL-1)–transfected cells (Fig. 3E and F). These results indicate that MT189-induced apoptosis is potentiated by MCL-1 reduction in the cells, which contributes to its in vitro proliferation-inhibitory activity.

However, MT189 did not lower the mRNA level of MCL-1 (Supplementary Fig. S6). But the proteasome inhibitor MG132 could rescue MCL-1 protein reduction by MT189 (Fig. 3G). Moreover, MT189 apparently increased the ubiquitination of MCL-1 protein (Fig. 3H). The data reveal that MT189 promotes the degradation of MCL-1 protein via the ubiquitin–proteasome pathway.

The MAPK kinases JNK and p38 have been shown to participate in the regulation of the MCL-1 protein stability (11, 25, 26). MT189 was revealed indeed to increase their activated (phosphorylated) forms but not to change their total proteins levels (Fig. 4A). However, it was the JNK inhibitor SP600125, but not the p38 inhibitor PD169316, that prevented the MT189-mediated Ser159/Thr163 phosphorylation and degradation of MCL-1 protein in SMMC-7721 cells (Fig. 4B, top). Consistently, SP600125 rather than PD169316 partially reversed the cleavage of caspase-9, caspase-3, and PARP (Fig. 4B, bottom), apoptosis (Fig. 4C and D), and the loss of viable cells (Fig. 4E) induced by the treatment with MT189 in SMMC-7721 cells. Moreover, p-JNK was shown to be colocalized with both MCL-1 and p-MCL-1 in the MT189-treated SMMC-7721 cells (Fig. 4F). The results indicate that MT189 mediates JNK activation, which leads to MCL-1 phosphorylation that facilitates its ubiquitination-driven degradation.

The JNK inhibitor SP600125 also inhibits CDK1 (27). So, we used another JNK inhibitor JNK-IN-8 (28) and a CDK1 inhibitor RO3306 (29–31). The result showed that the MT189-mediated MCL-1 degradation could be reversed only by the JNK inhibitors but not by the CDK1 inhibitor (Fig. 5A). In addition, both JNK and CDK1 were reported to mediate phosphorylation of the antiapoptotic protein BCL-2 (32–34). However, the CDK1 inhibitor RO3306 did not affect the MT189-mediated increase in the phosphorylation level of BCL-2 (Fig. 5B). Moreover, either overexpression or silence of MCL-1 did not change the cell-cycle arrest induced by MT189 (Fig. 5C). These data strengthen the evidence supporting the conclusion that the MT189-mediated JNK activation rather than the mitotic arrest–related CDK1 activation causes the increased phosphorylation levels of MCL-1 followed by its ubiquitination-driven degradation.

The next question is how MT189 mediates JNK activation. Either JNK or p-JNK did not appear in the immunoprecipitated α-tubulin complexes in the control or MT189-treated SMMC-7721 cells (Fig. 5D). In the control cells, both JNK and p-JNK were observed mainly in nuclei, but they were not revealed to be colocalized with tubulin. The treatment with MT189 did not change this feature of either JNK or p-JNK, although it increased the levels of p-JNK (Supplementary Fig. S7). The data suggest that there is no direct interaction between tubulin and JNK, and JNK phosphorylation followed by the treatment with MT189 must be caused by some kinase(s).

JNK has been shown to be activated via MAP3Ks (MEKK1 and TAK1) through MAP2ks (MKK4 and MKK7; ref. 10). Indeed, the treatment with MT189 apparently enhanced the phosphorylation levels of JNK, p38, MEKK1, TAK1, and MKK4, but just marginally changed the level of p-MKK7 in SMMC-7721 cells. At the same time, the levels of their corresponding total proteins basically kept unchanged (Fig. 5E). However, either intact or disrupted microtubules were not colocalized with p-MEKK1 or p-TAK1 or their nonphosphorylated forms in SMMC-7721 cells (data not shown). The results indicate that JNK is activated via the MEKK1/TAK1–MKK4 pathway responding to the MT189-caused microtubulin disruption, but which upstream kinase(s) are involved in the activation of TAK1 and/or MKK4 remains to be clarified.

Colchicine binding site inhibitors are promising anticancer agents, several of which, including CA4 derivatives, are at different phases of clinical trials (1). These inhibitors display differential characteristics in anticancer spectrum, drug resistance, and antivascular effects when compared with the classic tubulin inhibitors such as paclitaxel and vincristine (1, 33). We previously reported MT7 and MT119 derived from a combinatorial library of 6H-Pyrido[2′,1′:2,3]imidazo[4,5-c]isoquinolin- 5(6H)-ones as colchicine site–targeted tubulin inhibitors (2–4). They are new chemical entities with a distinct mode of tubulin polymerization inhibition from colchicines, but disappointingly, both do not show in vivo anticancer activity (3, 4). MT189 is a newest analog of this series, obtained by taking a deconstruction approach to break the tetracyclic ring of MT119.

In this study, we found that MT189, a new optimized derivative of MT119 (4) retained its property in impairing microtubule assembly by depolymerizing microtubules via binding to the colchicine site, leading to M arrest and inducing apoptosis. MT189 showed significantly improved in vitro proliferation-inhibitory activity and direct killing of MDR tumor cells. However, its in vivo anticancer activity is still limited. MT189 has very low solubility in water, and thus it was administered by intraperitoneal injection rather than by intravenous injection (as taxol) in this study. When administered at 100 mg/kg/d once a day to nude mice, MT189 caused tumor growth inhibition by 35.9% on day 14. At this time point, all tested animals were alive and showed no obvious loss of the body weight. However, on day 17, the animals began to die, and till day 21, among 8 tested animals in the MT189 group, only 1 was still alive. Therefore, demonstrating what are the primary toxicity organs and the related toxicity mechanisms of MT189 is critical for our future work.

MT189 binds to cellular microtubules via the colchicine site, leading to rapid microtubule depolymerization (within 0.5 hours) followed by persistent M arrest and subsequent apoptotic induction. In addition, MT189 also relieves the apoptosis suppression of the antiapoptotic protein MCL-1 by promoting its degradation via the ubiquitination-proteasome pathway, which potentiates its apoptotic induction and thus its anticancer effects. Consequently, persistent mitotic arrest, apoptosis, and MCL-1 degradation collectively contribute to the anticancer activity of MT189 (Fig. 5F).

MCL-1 is an antiapoptotic member of the BCL-2 protein family. It can block proapoptotic proteins BAX and BAK to repress apoptosis. Both taxol and vincristine can lead to degradation of MCL-1 via different pathways, such as JNK (11), CDK1 (12, 34), and NOXA (35, 36). However, whether colchicine binding site tubulin inhibitors elicit similar effects remains to be clarified. In this study, we demonstrated that the colchicine site–targeted tubulin inhibitor MT189 sequentially activated MEKK1/TAK1, MKK4, and JNK (Fig. 5E). Activation of JNK caused phosphorylation of MCL-1 that facilitated its ubiquitination and subsequent degradation. However, we did not observe that microtubulin (either intact or disrupted) was colocalized with JNK, MEKK1, or TAK1 or their respective phosphorylated equivalents. The result suggests that MT189 could mediate the activation of the MEKK1/TAK1–MKK4–JNK pathway via unknown intermediary kinase(s) in a microtubulin depolymerization–dependent fashion (Fig. 5F).

In fact, various extracellular stimuli such as cytokines and protein synthesis inhibitors can stimulate the phosphorylation of JNK (37). Many other anticancer agents with different molecular targets or mechanisms of action can also lead to the activation of this pathway (10, 37, 38). Notably, detailed time–effect analyses showed that taxol, a microtubulin polymerization–promoting anticancer agent, activated JNK at early stages (30 minutes; refs. 37, 38), just as MT189 did (0.5 hours; Fig. 5E). All the lines of evidence suggest that the depolymerization of microtubules is an essential factor for JNK activation. JNK has multiple substrates and is involved in extensive physiologic and pathophysiologic processes and therapeutic responses (8, 10, 39–41). Apparently, it is deserved to put continuous efforts into demonstrating which unknown factor(s) and how they mediate JNK activation.

We have already shown that in MT189-treated cancer cells, microtubule depolymerization causes both M arrest (Figs. 1E and F and 5F) and JNK activation following MCL-1 degradation (Figs. 4 and 5E and F). Notably, however, both microtubule depolymerization and JNK activation took place at very early stages (0.5 hours; Fig. 5E and Supplementary Fig. S3), but M arrest occurred at relatively late stages (3 hours or later; Fig. 1E and F) in those cells. Moreover, both tested JNK inhibitors, but not the CDK1 inhibitor, could reverse MCL-1 degradation (Fig. 5A); the CDK1 inhibitor did not prevent the increase in the MT189-induced BCL-2 phosphorylation (Fig. 5B) although CDK1 was reported to mediate phosphorylation of BCL-2 (34); and the changes in MCL-1 protein expression did not affect the MT189-induced M arrest (Fig. 5C). In addition, MT189 was shown to mediate JNK activation via the MEKK1/TAK1–MKK4 pathway. Therefore, JNK activation is independent of M arrest in MT189-treated cancer cells (Fig. 5F).

Together, this study presents a new tubulin inhibitor MT189 with anticancer activity and retaining the colchicine site–targeted characteristics of its parental compound MT119. Besides its mitosis arrest and apoptosis induction, we show that relief of apoptosis suppression due to MCL-1 degradation subsequent to the activation of the MEKK1/TAK1–MKK4–JNK pathway potentiates apoptotic induction of MT189. We believe that demonstrating how to activate this pathway is helpful to further understand the mechanisms of action of MT189 and thus to promote the optimization and development of its new derivatives.

No potential conflicts of interest were disclosed.

Conception and design: W. Wang, T. Meng, L.-P. Ma, J. Ding, J.-K. Shen, Z.-H. Miao

Development of methodology: X.-J. Huan, L.-J. Tong, Z.-H. Miao

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Y. Chen, J. Ding, Z.-H. Miao

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): W. Wang, Y.-Q. Wang, Z.-H. Miao

Writing, review, and/or revision of the manuscript: W. Wang, Y.-Q. Wang, Z.-H. Miao

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J.-M. Yi, X.-J. Huan, L.-J. Tong, Y. Chen, J. Ding

Study supervision: Y.-Q. Wang, Z.-H. Miao

This work was supported by grants from the National Natural Science Foundation of China (NSFC; No. 81025020 to Z.H. Miao, No. 81202548 to Y.-Q. Wang, and No. 81321092 to J. Ding), the National Basic Research Program of China (No. 2012CB932502 to Z.H. Miao), the National Science and Technology Major Project of China (No. 2012ZX09301-001-002 to Z.-H. Miao), and the “Interdisciplinary Cooperation Team” Program for Science and Technology Innovation of the Chinese Academy of Sciences (to Z.-H. Miao).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.