Anti‐growth and pro‐apoptotic effects of dasatinib on human oral cancer cells through multi‐targeted mechanisms

Abstract Dasatinib is an inhibitor of Src that has anti‐tumour effects on many haematological and solid cancers. However, the anti‐tumour effects of dasatinib on human oral cancers remain unclear. In this study, we investigated the effects of dasatinib on different types of human oral cancer cells: the non‐tumorigenic YD‐8 and YD‐38 and the tumorigenic YD‐10B and HSC‐3 cells. Strikingly, dasatinib at 10 µM strongly suppressed the growth and induced apoptosis of YD‐38 cells and inhibited the phosphorylation of Src, EGFR, STAT‐3, STAT‐5, PKB and ERK‐1/2. In contrast, knockdown of Src blocked the phosphorylation of EGFR, STAT‐5, PKB and ERK‐1/2, but not STAT‐3, in YD‐38 cells. Dasatinib induced activation of the intrinsic caspase pathway, which was inhibited by z‐VAD‐fmk, a pan‐caspase inhibitor. Dasatinib also decreased Mcl‐1 expression and S6 phosphorylation while increased GRP78 expression and eIF‐2α phosphorylation in YD‐38 cells. In addition, to its direct effects on YD‐38 cells, dasatinib also exhibited anti‐angiogenic properties. Dasatinib‐treated YD‐38 or HUVEC showed reduced HIF‐1α expression and stability. Dasatinib alone or conditioned media from dasatinib‐treated YD‐38 cells inhibited HUVEC tube formation on Matrigel without affecting HUVEC viability. Importantly, dasatinib's anti‐growth, anti‐angiogenic and pro‐apoptotic effects were additionally seen in tumorigenic HSC‐3 cells. Together, these results demonstrate that dasatinib has strong anti‐growth, anti‐angiogenic and pro‐apoptotic effects on human oral cancer cells, which are mediated through the regulation of multiple targets, including Src, EGFR, STAT‐3, STAT‐5, PKB, ERK‐1/2, S6, eIF‐2α, GRP78, caspase‐9/3, Mcl‐1 and HIF‐1α.


| INTRODUC TI ON
Oral squamous cell carcinoma (OSCC) accounts for 24% of all head and neck cancers. 1 OSCC is caused predominantly through exposure to areca quid chewing, cigarette smoking and alcohol consumption. 2,3 OSCC not only causes significant mortality but is associated with functional alterations in the oromaxillo-facial region and disfigurement. 4 The main treatments for OSCC are surgery, radiotherapy alone, or in combination with chemotherapy. [5][6][7] However, despite extensive research on treatments for oral cancers, overall survival for patients with OSCC has not improved. 8 Therefore, there is a need for both new therapies and biomarkers for OSCC progression.
Src is a non-receptor protein tyrosine kinase, 9,10 whose overexpression and hyperactivity are associated with tumour mass, metastasis, recurrence, angiogenesis and survival of patients with cancer. [14][15][16] Of further note, studies have previously demonstrated that Src is overexpressed in OSCC, including tongue cancer, and there is a significant association of Src overexpression with progression, recurrence and prognosis of OSCC. 11,17,18 These findings strongly suggest Src inhibition as a targeted therapy for solid cancers, including tongue cancer.

Dasatinib (BMS-354825) is an inhibitor of Bcr-Abl and Src kinases
that has been shown to be an effective treatment of chronic myeloid leukaemia (CML) and Philadelphia chromosome-positive acute lymphoblastic leukaemia. 19,20 Moreover, dasatinib has anti-proliferative, anti-invasive, anti-angiogenic and pro-apoptotic properties in solid tumours. [21][22][23] A wealth of information strongly supports that dasatinib is a multi-targeted tyrosine kinase inhibitor. Dasatinib inhibits the activity of Bcr-Abl and Src family kinases along with additional receptor/non-receptor tyrosine kinases and signalling protein kinases, including c-KIT, platelet-derived growth factor receptor (PDGFR), epidermal growth factor receptor (EGFR), the ephrin-A receptor, TEC family kinases and ERK-1/2. [24][25][26] These results strongly indicate that dasatinib's anti-tumour effects on haematological malignancies and solid cancers are largely associated with its multi-targeted tyrosine kinase inhibition.

| Cell viability and survival assay
For cell viability assay, cells (1 × 10 4 cells/mL/100 μL/well) were seeded in a 96-well plate. After overnight incubation, cells were treated with dasatinib (0, 0.1, 1, and 10 µM) for 24 hours. Cells were then washed twice with PBS, and viability was determined using MTS reagent according to the manufacturer's protocol. Briefly, eight µl of the culture media (RMPI-1640 or DMEM) and 20 µL of MTS solution was added to each well, and the plate was incubated at 37℃ for 1 hour. The absorbance of each well was measured at 490 nm using a microplate reader (SPECTRA max 340PC; Molecular Devices, LLC).
The number of survived cells, which cannot be stained with trypan blue dye, was counted using a phase-contrast microscope. The cell K E Y W O R D S apoptosis, dasatinib, HIF-1α, HSC-3, Src, YD-38 count assay was performed in triplicate. Data are mean ± standard error (SE) of three independent experiments. Survival is expressed as a percentage of control.

| DNA fragmentation assay
DNA fragmentation was performed as mentioned in our previous study. 30 Genomic DNA was extracted and analysed via electrophoresis at 100 V on a 1.7% agarose gel for 20 minutes. The DNA was visualized and photographed under UV illumination after staining with ethidium bromide (0.1 µg/mL) using a gel documentation system (Gel Doc-XR, Bio-Rad Laboratories, Inc.).

| Western blot analysis
After treatments, cells were lysed in a modified radioimmunoprecipitation assay buffer (Sigma-Aldrich; Merck) containing PIC (1×). Proteins (50 µg) were separated by sodium dodecyl sulphatepolyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes (Millipore). The membranes were blocked with 5% (w/v) skim milk in Tris-buffered saline (TBS) containing 0.1% Tween 20 (TBST) for 2 hours. The membranes were incubated with specific antibodies listed in Supplementary Table 1 at 4 ºC overnight. The membranes were then washed twice with TBST and exposed to secondary antibodies conjugated to horseradish peroxidase (HRP) for 2 hours at room temperature, and immunoreactivity was detected by ECL reagents. β-actin was used as an equal protein loading control.

| Reverse transcription-polymerase chain reaction (RT-PCR)
After treatments, total cellular RNA was isolated using TRIzol reagent (Life Technologies) according to the manufacturer's protocol.
Complementary DNA was then prepared using M-MLV reverse transcriptase (Gibco-BRL) according to the manufacturer's protocol. The primer sequences used in this study are listed in

| siRNA transfection
Small interfering RNA (siRNA) of Src (sc-29228) and control siRNA (sc-37007) were obtained from Santa Cruz Biotechnology. YD-38 cells were transfected with 100 pM of control or Src siRNA using Lipofectamine RNAiMAX (Invitrogen) for 48 hours. The cells were then lysed, and protein expressions were analysed by Western blotting.

| Statistical analyses
Results are expressed as mean ± standard error (SE) of three independent experiments. The significance of the difference between sample means was determined by one-way ANOVA, followed by Dunnett's post hoc test, using SPSS 11.5 software (SPSS, Inc.). The difference was considered statistically significant for the value of P < .05.  Figure 1D). There was little or no DNA fragmentation in YD-8 or YD-10B cells treated with 10 µM of dasatinib for up to 8 hours.

| Dasatinib treatment at 10 µM inhibits the phosphorylation of Src, EGFR, STAT-3, STAT-5, PKB and ERK-1/2 in YD-38 human oral cancer cells
Given that dasatinib is a multi-targeted protein kinase inhibitor, we next investigated the treatment effects of dasatinib at 10 µM

| Knockdown of Src causes a strong inhibition of the phosphorylation of EGFR, STAT-5, PKB and ERK-1/2 in YD-38 human oral cancer cells
We hypothesized that dasatinib-mediated decline of the phospho-

| Dasatinib induces apoptosis via activation of the intrinsic caspase pathway in YD-38 human oral cancer cells
Activation of caspases is a principal feature for the execution of cell death by various apoptotic stimuli. Dasatinib treatment led to a time-dependent increase in cleavage of caspase-9 and its target PARP in YD-38 cells ( Figure 4A). To see whether dasatinib treatment modulates the extrinsic apoptotic pathway, we next measured the expression levels of death receptor-5 (DR-5).
Dasatinib treatment did not increase the expression levels of DR-5 in the cells; rather, a slightly decreased protein expression was observed. In further support of dasatinib actions via the intrinsic pathway, z-VAD-fmk, a pan-caspase inhibitor, greatly attenuated the DNA fragmentation induced by dasatinib in YD-38 cells ( Figure 4B).  35 Notably, as further shown in Figure 5A, while dasatinib treatment for 2 hours led to reduced phosphorylation but not total levels of S6, the drug treatment for 4 to 24 hours resulted in reduced phosphorylation and total levels of S6 in YD-38 cells.

| Dasatinib reduces HIF-1α at the protein levels in YD-38 human oral cancer cells
HIF-1α is an angiogenic transcription factor. 36 The HIF-1α expression is linked to the tumour promotion in human oral cancer 37 and correlates with the growth and adhesion in human OSCC cells. 38,39 In this study, we further determined whether HIF-1α protein is expressed in YD-38 cells and whether dasatinib modulates it. Of interest, HIF-1α was highly expressed in YD-38 and YD-10B cells, but there was a low expression of HIF-1α in YD-8 cells ( Figure 6A). Notably, dasatinib treatment at 1 or 10 µM resulted in almost complete down-regulation of HIF-1α protein in YD-38 and YD-8 cells but had no effect on the expression of HIF-1α protein in YD-10B cells. Furthermore, dasatinib treatment at 10 μM greatly down-regulated HIF-1α at the protein, but not mRNA, levels in YD-38 cells at times tested, suggesting posttranscriptional regulation of HIF-1α by dasatinib ( Figure 6B,C).

| Dasatinib induces Src-, PI3K/PKB-and ERK-1/2-dependent HIF-1α destabilization in YD-38 human oral cancer cells
We next determined whether dasatinib affects the stability of HIF-1α protein in YD-38 cells using a CHX-based protein stability assay. There was a gradual time-dependent decrease in the amounts of HIF-1α protein in YD-38 cells grown in the presence of CHX at times tested (as shown in Figure 7A), suggesting that if the translation is blocked, HIF-1α protein is intrinsically unstable in these cells.

| Dasatinib blocks tube formation of human umbilical vein endothelial cells
We next investigated the effect of dasatinib on angiogenesis in vitro by measuring the tube formation of HUVEC exposed to the We next determined whether dasatinib-mediated inhibition of the tube formation in HUVEC is due to the drug's cytotoxicity. There was no significant change in the viability and survival of HUVEC treated for 24 hours without or with dasatinib at 10 µM, ruling out the possibility of the drug's cytotoxicity ( Figure 8C,D). PMA is a known inducer of HUVEC tube formation. 42 Treatment with PMA at 10 nM induced tube formation of HUVEC, which was also greatly attenuated by dasatinib treatment (Figure 8E).  Dasatinib exerts anti-tumour effects on blood malignancies and solid cancers by inhibiting cell proliferation/survival and inducing cell apoptosis. [19][20][21][22]43,44 It had been previously shown that dasatinib at 10 µM inhibits the viability and induces the apoptosis of HSC-3 cells. 45 Extending on this, we demonstrated that dasatinib at In agreement with previous studies, 45  Induction of ER stress is common to many anti-cancer drugs and/or agents. 60,61 ER stress occurs when cells have dysfunction of protein synthesis and excessive accumulation of misfolded proteins in the ER 62 and high phosphorylation levels of eIF-2α, a translation regulatory protein. 63 The phosphorylation of eIF-2α is indicative of its inactivation and global translational inhibition. 64 It is known that the phosphorylation of S6, a ribosomal protein involved in translation, is closely associated with global translational activation. 35 It is worth mentioning previous reports that the regulation of ER stress is identified as a mechanism for the anti-cancer effects of dasatinib in head and neck cancers. 65,66 Since dasatinib increased the expression of GRP78 and the phosphorylation of eIF-2α while decreasing the phosphorylation of S6 in YD-38 cells, induction of ER stress and global translational inhibition in YD-38 cells may therefore contribute to the drug's growth-suppressive and proapoptotic effects.

| D ISCUSS I ON
The transcription factor HIF-1α is involved in tumour angiogenesis by regulating the transcriptional induction of angiogenesisrelated genes, such as VEGF, that contain HRE cis-acting elements within their promoters. 34,67 HIF-1α is currently regarded as a prime target for anti-cancer therapies. 68,69 At present, the mechanisms underlying high HIF-1α protein expression and dasatinib regulation of HIF-1α expression in oral cancer cells are poorly understood.
In this study, we found high HIF-1α protein expression in all four human oral cancer cell lines. Distinctly, the present study demonstrated that dasatinib greatly down-regulated HIF-1α protein in YD-8, YD-38 and HSC-3, but not YD-10B cells. These results indicate that dasatinib differentially regulates HIF-1α expression in a cell type-dependent manner. A further striking finding herein is that dasatinib down-regulates HIF-1α at the protein, but not mRNA level, which is due to an increase in HIF-1α protein turnover/destabilization through inhibition of Src, PI3K/PKB and ERK-1/2. It is thus likely that dasatinib-induced HIF-1α protein down-regulation and turnover in YD-38 cells are associated with the drug's ability to inhibit these multi-kinases.
In summary, we demonstrate that dasatinib has strong anti-growth, anti-angiogenic and pro-apoptotic effects on YD-38 and HSC-3 cells, and these effects are mediated through the regulation of multiple cellular targets and pathways. This work suggests that dasatinib could be a possible regimen for the treatment of human oral cancer.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.