Regorefenib induces extrinsic/intrinsic apoptosis and inhibits MAPK/NF‐κB‐modulated tumor progression in bladder cancer in vitro and in vivo

Abstract The aim of the present study is to investigate anticancer effect and mechanism of regorafenib in bladder cancer in vitro and in vivo. Human bladder cancer TSGH 8301 cells were treated with regorafenib, NF‐κB, AKT, or mitogen‐activated protein kinase (MAPK) inhibitors for different time. The changes of cell viability, NF‐κB activation, apoptotic signaling transduction, and expression of tumor progression‐associated proteins were evaluated with MTT, NF‐κB reporter gene assay, flow cytometry, and Western blotting assay. TSGH 8301 tumor bearing mice were established and treated with vehicle (140 μL of 0.1% DMSO) or regorafenib (10 mg/kg/day by gavage) for 15 days. The changes of tumor volume, body weight, NF‐κB activation, MAPK activation, and tumor progression‐associated proteins (MMP‐9, XIAP, VEGF, and Cyclin‐D1) after regorafenib treatment were evaluated with digital caliper, digital weight, and ex vivo Western blotting assay. Our results demonstrated NF‐κB activation and protein levels of MMP‐9, XIAP, VEGF, and Cyclin‐D1 were significantly reduced by NF‐κB (QNZ), ERK (PD98059), and P38 (SB203580) inhibitors. Regorafenib also significantly induced extrinsic and intrinsic apoptotic signaling transduction in bladder cancer in vitro. In addition, regorafenib significantly inhibited tumor growth, NF‐κB, p38, ERK activation and expression of tumor progression‐associated proteins in bladder cancer in vitro and in vivo. Taken together, these results proved that regorafenib not only induced apoptosis through extrinsic and intrinsic pathways and but suppressed MAPK/ NF‐κB‐modulated tumor progression in bladder cancer.


| INTRODUCTION
Many epidemiological studies presented various environmental risk factors such as fungicide, tobacco, metals and motor vehicle exhaust, and occupational exposure to aromatic amines are associated with development of bladder cancer. 1 Carcinogens-induced genetic alteration result in conversion of proto-oncogenes to oncogenes and silence of tumor suppressor genes. High expression of oncogenes and inactivation of tumor suppressor genes cause hyperactivation of signaling pathways involved in cell growth, survival, angiogenesis, and metastasis leading to cancer formation and progression. [2][3][4] Impact of genetic alterations on dysregulated signaling transduction has been found in muscle invasive bladder cancers (MIBCs) and non-MIBCs (NMIBCs). 5 The p53 and RB (retinoblastoma) tumor suppressor pathway which controls restriction of cell cycle progression is altered and negatively regulated by mutation of tumor suppressor genes and expression of oncogenes. 6 Hyperactivation of Ras/mitogenactivated protein kinase (MAPK) pathway contributes to tumor progression. Overexpression of Ras/MAPK pathway is modulated by Ras, HER2 and HER3 mutation, and EGFR amplification in bladder cancer.
Ras/MAPK pathway has been targeted by tyrosine kinase inhibitors (TKIs) in clinical trials. 4,[6][7][8] Regorafenib (Stivarga) is an oral multiple kinase inhibitor which targets several critical signaling molecules including angiogenic, stromal, oncogenic receptor tyrosine kinases and already approved for treatment of patients with colorectal cancer, gastrointestinal stromal tumors, and hepatocellular carcinoma (HCC). 9,10 Regorafenib has been indicated to diminish tumor progression through suppression of MAP-K/extracellular signal-regulated kinase (ERK) activation in HCC in vitro and in vivo 11,12 . In previous study presented regorafenib induces apoptosis and inhibits metastatic potential in bladder cancer. 13 However, effect of regorafenib on tumor progression in bladder cancer is ambiguous. Therefore, the aim of the present study was to verify anti-cancer effect and mechanism of regorafenib in bladder cancer in vitro and in vivo.

| Culture of TSGH cells
TSGH 8301 human bladder carcinoma cell line was obtained from Professor Jing-Gung Chung's lab, China Medical University, and routinely tested for mycoplasma contamination. 14,15 TSGH 8301 cells were seeded on 10 cm tissue culture plate with RPMI-1640 medium and containing 10% FBS, 2 mM L-glutamine, and 1% penicillinstreptomycin (100 U mL −1 penicillin and 100 μg/mL streptomycin) and grown at 37 C under a humidified 5% CO 2 and one atmosphere.    and subcutaneous injected into mice right flank (N = 20). 15 When the tumor size reached approximately 100 mm 3 , the mice were randomly divided into two groups and received the following treatments: CTRL (0.1% DMSO) and regorafenib (10 mg/kg). Tumor volumes were calculated from the following formula: tumor volume = length × width × thickness × 0.523. 12 Body weight was measured by digital weight every 3 days. Mice were sacrificed and tumors were finally extracted, measured by digital weight on day 15 and lysed for ex vivo Western blot.

| Statistical analysis between different treatment groups
Results are presented as mean ± SD. The significant difference between regorafenib-treated and non-treated control groups were analyzed by Student's t-test. P < 0.05 and P < 0.01 were both defined as an indication of statistical significance. FIGURE 2 Regorafenib affected tumor progression-associated protein expression in TSGH 8301 cells. TSGH 8301 cells were treated with regorafenib (15 and 30 μM), p38 inhibitor (10 μM), ERK inhibitor (10 μM) and QNZ (10 μM) for 48 h. Cells were collected and total protein was determined for SDS-PAGE gel electrophoresis, as described in Section 2. The levels of MMP-9, XIAP, VEGF, and CyclinD1 (A); p-p38 and p-ERK1/2 (B) after regorafenib treatment, p38 inhibitor treatment (C), ERK inhibitor treatment (D) or QNZ treatment (E) were estimated by Western blotting. **P < 0.01, significant difference between regorafenib-treated groups and the control as analyzed by Student's t-test 3 | RESULTS

| Regorafenib enhanced bladder cancer cells cytotoxicity through blockage of p38 MAPK-and ERK-mediated NF-κB pathway
In Figure 1A, regorafenib markedly enhanced cytotoxicity of TSGH 8301 cells by dose and time. We further investigated whether regofenib may affect the activation of NF-κB by reporter gene assay. The activation of NF-κB was markedly suppressed by regorafenib with time-and dosedependent manner ( Figure 1B). Quantification results, Figure 1C, demonstrated 20-50 suppression percentage of NF-κB activity by regorafenib.
Then, we further investigated the upstream signaling transduction of regorafenib on TSGH8301 cells by treated with different inhibitors, such as NF-κB inhibitor, p38 MAPK inhibitor, ERK inhibitor, JNK inhibitor and AKT inhibitor. Our results suggested that only p38 MAPK inhibitor and ERK inhibitor showed similar NF-κB suppression efficacy as NF-κB inhibitor ( Figure 1D). In sum, these results indicated that the toxicity effect

| Regorafenib inhibited tumor progressionrelated gene expression through blockage of p38 MAPK and ERK signaling transduction
After we confirmed the toxicity and NF-κB inhibition ability of regorafenib, we further investigate downstream molecules that mediated by NF-κB. Here, we found that invasion related MMP-9, proliferation related XIAP and CyclinD1, angiogenesis-related VEGF proteins were all decreased by regorafenib (Figure 2A). Moreover, the phosphorylation of p38 MAPK and ERK were both reduced by regorafenib ( Figure 2B). To inveterate the tumor progression inhibitory effect of regorafenib was through the modulation of p38 MAPK-and ERKmediated NF-κB signaling, we performed with Western blot for tumor progression related proteins evaluation by treated cells with p38 MAPK and ERK inhibitors. As showed in Figure 2C, all NF-κB mediated tumor progression related markers were decreased by p38 MAPK inhibitor. In ERK inhibitor treated groups, we found similar inhibition result as p38 MAPK inhibitor ( Figure 2D). We finally validated whether NF-κB inhibitor may demonstrate compatible results as Figure 2C,D. Figure 2E specified that the suppression of NF-κB was certainty reduced the expression of MMP9, XIAP, VEGF, CyclinD1. In sum, regorafeinb may reduce tumor progression through diminish of p38 MAPK-and ERKmediated NF-κB pathway.

| Regorafenib-induced mitochondria dependent apoptosis and extrinsic apoptosis of bladder cancer cells
In cell cycle process, subG1 population has been recognized as apoptosis phase of cells. In Figure 3A, the population of subG1 was significantly increased to 20%-50% by regorafenib, which represent the apoptosis effect was increased. Caspase-3 was known as an inevitable marker in the process of apoptosis. We found the obvious induction  Figure 3B). For intrinsic apoptosis marker, mitochondria membrane potential was noticeably loss after regorafenib administrated ( Figure 3C). Additionally, extrinsic apoptosis marker caspase-8 was induced by regorafenib ( Figure 3D). In annexin V and PI double staining results, we found that the early and late apoptosis were both induced around 20%-40% in TSGH 8301 cells by time ( Figure 3E). Finally, we further confirmed whether the death receptor and ligand were both modulated by regorafeinb. As showed in Figure 3F,G, the activation of Fas and FasL in regroafenib treated group was found. Furthermore, regorafenib not only enhanced extrinsic and intrinsic apoptosis of bladder cancer cells but also induced the cleavage of PARP1 ( Figure 3H), which also demonstrated as an apoptosis marker. Figure 4A indicated that one representative animal tumor on each treatment of mice. Tumor size was markedly decreased after regorafenib treatment. After animals were treated, body weights and tumor volumes were measured once per 3 days. Figure 4B indicated that regorafenib significantly reduced tumor volume when compared to non-treated control groups. Figure 4C indicated the tumor weight from each treatment and demonstrated that regorafenib significantly reduced of tumor weight. Figure 4D indicated that regorafenib did not significantly affect the body weights when compared to control groups. After tumors were removed from each group of animals, proteins were extracted, assayed by Western blotting and results were shown in Figure 4E. Figure 4E indicated that regorafenib significantly decreased MMP-9, XIAP, VEGF, and CyclinD1 when compared to non-treated control groups. Moreover, as showed in Figure 4E, regorafenib inhibited tumor growth and tumor progression through dephosphorylating of p38 MAPK, ERK, NF-κB signaling transduction.

| DISCUSSION
Death receptor Fas (CD95)/Fas ligand (FasL) interaction initiates extrinsic apoptotic signaling pathway through triggering caspase-8 activation. 21 Loss of mitochondrial membrane potential promotes release of pro-apoptotic proteins from mitochondria leading to intrinsic apoptosis. 22 Caspase-3-modulated deoxyribonucleic acid (DNA) fragmentation and cleavage of poly (ADP-ribose) polymerase 1 (PARP1) can be activated by extrinsic and intrinsic apoptotic pathways. 23 Decreased expression of Fas and Caspase-3 is associated with poor prognosis in patients with bladder cancer. 21,24 PARP1, an ADP-ribosylating enzyme, participates in repair of various forms DNA damage and maintains chromatin remodeling. 25 Overexpression of PARP1 has been found in high stage cancers. 26,27 Liu et al. 28 presented PARP1 inhibitor not only induces DNA double-strand breaks (DSB) but also sensitizes bladder cancer cells to radiation. We found regorafenib significantly induces apoptosis and increases activation of Fas/FasL, Caspase-8, -3, loss of ΔΨm, and cleavage of PARP1 in bladder cancer TSGH 8301 cells.
Nuclear factor-κB (NF-κB) family of transcription factors are composed of five subunits including RelA (p65), RelB, c-Rel, NF-κB1 (p105/ p50), and NFκB2 (p100/p52). NF-κB p50/p65 heterodimer is critical mediator for tumorigenesis and cancer progression. A number of tumor progression-associated proteins which modulate proliferation, antiapoptosis, angiogenesis, and metastasis are encoded by NF-κB target genes and upregulated with active NF-κB signaling in bladder cancer. 29,30 Constitutive NF-κB activation was observed in high grade bladder cancer. 31 Many studies indicated inhibition of NF-κB activation suppresses tumor growth, anti-apoptosis, angiogenesis, and metastatic potential in bladder cancer in vitro and in vivo. 32,33 In our results   34 NF-κB activation may be regulated by AKT or MAPKs cascades. 11 We used NF-κB reporter gene assay to evaluate effect of AKT or MAPK inhibitors on NF-κB activation in TSGH 8301 cells. The results presented both p38 (SB203580) and ERK inhibitor (PD98059) significantly inhibit NF-κB activation in TSGH 8301 cells. Harb et al. 35 found high expression of p38 as poor prognostic marker was correlated with high grade stage and distant metastasis in bladder cancer.
ERK and p38 MAPK activate cell growth and invasive ability in bladder cancer cells. 36,37 In our results presented protein levels of MMP-9, XIAP, VEGF, and Cyclin-D1 were significantly reduced by PD98059 or SB203580 treatment, respectively. In addition, protein levels of phosphorylated ERK (p-ERK) and p38 (p-p38) were diminished by regorafenib treatment.
In conclusion, regorafenib not only induces extrinsic/intrinsic apoptosis and inhibits MAPK/NF-κB-modulated tumor progression in bladder cancer in vitro and in vivo ( Figure 5). We demonstrated regorafenib can be considered as a potential treatment, which offer therapeutic benefit for patients with bladder cancer.