Aurora kinase A revives dormant laryngeal squamous cell carcinoma cells via FAK/PI3K/Akt pathway activation

Revival of dormant tumor cells may be an important tumor metastasis mechanism. We hypothesized that aurora kinase A (AURKA), a cell cycle control kinase, promotes the transition of laryngeal squamous cell carcinoma (LSCC) cells from G0 phase to active division. We therefore investigated whether AURKA could revive dormant tumor cells to promote metastasis. Western blotting revealed that AURKA expression was persistently low in dormant laryngeal cancer Hep2 (D-Hep2) cells and high in non-dormant (T-Hep2) cells. Decreasing AURKA expression in T-Hep2 cells induced dormancy and reduced FAK/PI3K/Akt pathway activity. Increasing AURKA expression in D-Hep2 cells increased FAK/PI3K/Akt pathway activity and enhanced cellular proliferation, migration, invasion and metastasis. In addition, FAK/PI3K/Akt pathway inhibition caused dormancy-like behavior and reduced cellular mobility, migration and invasion. We conclude that AURKA may revive dormant tumor cells via FAK/PI3K/Akt pathway activation, thereby promoting migration and invasion in laryngeal cancer. AURKA/FAK/PI3K/Akt inhibitors may thus represent potential targets for clinical LSCC treatment.


INTRODUCTION
Laryngeal squamous cell carcinoma (LSCC) is one of the most common head and neck squamous cell carcinomas (HNSCC), and arises from the larynx epithelium with high metastasis rates and poor prognosis [1,2]. Currently LSCC patient therapies include chemotherapy, radiotherapy and surgery, alone or in combination [3]. While treatment prolongs patient survival, metastatic tumor growth severely reduces overall survival rates [4,5]. Elucidating the mechanisms of LSCC metastasis will be essential for identifying potential molecular targets to improve patient survival and quality of life.
Years after initial treatment, LSCC patients may develop local remnant or disseminated tumors. This phenomenon can be explained by tumor dormancy, a stage in tumor progression in which residual disease is present, but asymptomatic [6]. Revival of dormant tumor cells may be important in tumor recurrence and metastasis in cancer of the lung [7], breast [8] and prostate [9], as well as HNSCC [10,11]. However, mechanisms of tumor dormancy regulation are still largely unclear.
In the present study, we established a dormant LSCC cell model (D-Hep2 cells) to examine AURKA expression. We regulated AURKA expression in tumor Hep2 (T-Hep2) and D-Hep2 cells to assess dormant tumor cell revival. We also inhibited members of the FAK/PI3K/Akt pathway and found that AURKA might reverse tumor cell dormancy and contribute to LSCC metastasis and recurrence via FAK/PI3K/Akt pathway activation.

AURKA downregulation induces dormancy in T-Hep2 cells
T-Hep2 cells treated with the AURKA inhibitor, VX680 (100 nm/ml) [21], for 48 h showed reduced cell proliferation compared with the control group (P<0.01, Figure 2A). Flow cytometry assay results revealed that treated cells were arrested in G0/G1 phase (P<0.05, Figure  2B-2C). Immunofluorescence (IF) staining showed that AURKA was located in cytoplasm and P107 and P130 were located in nucleus. AURKA and P107 levels were reduced, while P130 levels were increased in T-Hep2 cells treated with VX680 compared with controls ( Figure  2D). Western blotting showed that p-AURKA expression was decreased almost threefold. The dormancy-related proteins, P130 and E2F4, were upregulated and P107 and Ki67 were downregulated in VX680-treated cells (P<0.05, P<0.01, Figure 2E-2F). Co-IP showed that the E2F4-P130 complex existed in treated cells ( Figure 2G). We concluded that AURKA inhibition could induce cell dormancy.

FAK/PI3K/Akt pathway inhibition reduces D-Hep2/AURKA cell mobility, migration and invasion
We applied wound-healing assays to test cell motility following treatment with inhibitors.

DISCUSSION
Tumor metastasis accounts for the majority of cancer-related deaths worldwide and the revival of dormant tumor cells may be one of the mechanisms related to metastasis. However, the precise molecular and cellular regulators involved in this transition remain poorly understood [33]. To study the role of dormant cells in tumor metastasis, we established a dormant cell model. Several interventions have been shown to induce cell dormancy, including starvation [20], short-term chemotherapy [34] and radiation [35]. Wilkie, et al. stated that tumor-immune dynamics in the micro-environment could inform tumor dormancy [36].
We verified that T-Hep2 cells cultured with 0.1% FBS for 48 h were indeed dormant using the CCK8 assay, which showed that T-Hep2 cells were stagnant. Flow cytometry indicated that T-Hep2 cells were arrested in G0/G1 phase. Western blotting implied that P130 and E2F4 levels were elevated and P107 and Ki67 levels were decreased. Finally, Co-IP showed that the E2F4-P130 complex existed in dormant Hep2 cells. All results illustrated that D-Hep2 cells were successfully established. Notably, T-Hep2 cells cultured for more than 48 h did not maintain dormancy.
We investigated tumor dormancy as it relates to LSCC recurrence. Aurora kinase A (AURKA), a member of the Aurora serine/threonine kinase family [51], occurs from late G2 and M phase, whereas resting cells have low or undetectable levels of this enzyme [52]. Based on our previous study, AURKA expression was elevated in human LSCC as compared to adjacent normal tissues, and was associated with regional lymph node metastasis and TNM stage [3]. AURKA promoted Hep2 cell migration and invasion and enhanced tumorigenesis in vivo [22]. Here, we observed that AURKA overexpression could revive dormant tumor cells to promote tumor metastasis. To our knowledge, this is the first report of a relationship between AURKA and LSCC cell dormancy. In our study, AURKA expression was low in D-Hep2 cells and dormancy-related proteins were impacted by alterations in AURKA expression. The E2F4-P130 complex was observed in T-Hep2 cells after 48 h treatment with VX680. Furthermore, D-Hep2 cells overexpressing AURKA exhibited enhanced cellular proliferation, migration and invasion. Together, these results demonstrated that AURKA could revive dormant Hep2 cells to stimulate malignant progression in LSCC.
AURKA reportedly interacts with proteins such as p53, BRCA1, Plk1 and PI3K. Bolos, et al. noted that FAK interacted with Src to activate PI3K followed by Akt to promote tumorigenicity and metastasis [53]. Yao, et al. revealed cross-talk between AURKA and the PI3K pathway during Akt activation [54]. We therefore studied the role of the FAK/PI3K/Akt pathway in dormant tumor cell revival, and the interactions between AURKA and this pathway in promoting LSCC metastasis. The FAK/ PI3K/Akt pathway was activated in T-Hep2 compared with D-Hep2 cells and was altered depending on AURKA expression. FAK/PI3K/Akt pathway inhibition also altered levels of dormancy-related proteins, suggesting that this pathway might regulate dormancy-like behavior along with D-Hep2/AURKA cell mobility, migration and invasion. Deservedly, there may be other more tumor signal pathways involved in the process except FAK/ PI3K/Akt which deserve us to discover further.
In addition, VX680, TAE226, Omipalisib and Triciribine, inhibitors of AURKA, FAK, PI3k and Akt, respectively, reduced LSCC cell mobility, migration and invasion and lead to tumor regression. Therefore, drugs targeting the AURKA/FAK/PI3k/Akt molecules could be tested as single agent or combination therapies. Drug doses and schedules should be guided by further preclinical trials and correlative studies should be performed to test drug pharmacodynamics.
In conclusion, we demonstrated that AURKA may revive dormant tumor cells via FAK/PI3K/Akt pathway activation, thereby promoting migration and invasion in laryngeal cancer. FAK/PI3K/Akt/AURKA inhibitors might serve as potential targets for clinical LSCC treatment.

Cell cultures
Human LSCC Hep2 (T-Hep2) cells were preserved by the Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences. Cells were cultured with Dulbecco's modified Eagle's medium (DMEM) (Gibco company, USA) containing 0.1% or 10% Fetal Bovine Serum (FBS) (Gibco) with 100 IU/ml penicillin and 100 IU/ml streptomycin at 37°C and 5% CO 2 in a humidified incubator.

CCK8 assay
2×10 3 cells in 100 ul of DMEM were seeded into 96-well plates. 10 ul of CCK8, used to measure cell proliferation, was added to every well. Cells were incubated for 2 h and OD450 absorbance values were measured. In our study, T-Hep2 and D-Hep2 cell proliferation was measured at 0, 24, 48, 72, 96 and 120 h. Cells treated with an AURKA inhibitor (VX680) (Selleck Chemicals, Houston, TX, USA) were measured at 0, 24 and 48 h.

Flow cytometry
Cells under starvation conditions or treated with VX680 were seeded into 6-well plates and cultured to 60-70% confluence. Cells were trypsinized, rinsed three times with cold phosphate buffered saline (PBS) and fixed with 100% cold alcohol at 4°C overnight. Cells were then stained with 300 μl PI/Rnase Staining Buffer (BD Pharmingen) in the dark at 37°C for 30 min. FACS Calibur (Becton Dickinson, USA) was used to analyze cell cycle stage. Modfit Software (Becton Dickinson, USA) was used to quantify the number of cells in G0/G1, S or G2/M phase.

Immunofluorescent staining
5×10 4 cells, with or without VX680 for 48 h, were seeded into millicell ez slides (Millipore, MA, USA) and fixed with 4% paraformaldehyde (PFA) for 30 min. Slides were rinsed three times with PBS, blocked with 5% BSA in PBS containing 0.05% Triton for 1 h at room temperature and incubated overnight with primary antibodies at 4°C. Slides were then rinsed three times with PBS and incubated with secondary antibodies for 1 h at room temperature in the dark. Nuclei were visualized with DAPI (1:1000, Beyotime) in PBS for 5 min in the dark. Slides were rinsed three times in PBS and analyzed by fluorescent microscopy (10x). Primary antibodies included anti-p-AURKA (1:100, Anti-rabbit IgG, Cell Signaling Technology), anti-P130 (1:100, Anti-rabbit IgG, Santa Cruz) and anti-P107 (1:100, Anti-rabbit IgG, Santa Cruz). Secondary antibodies included Alexa Fluor® 488 goat anti-rabbit IgG and Alexa Fluor® 555 goat anti-rabbit IgG (1:1000, Santa Cruz).

Plate colony formation assay
Cells, with or without transfection, were seeded into 6-well plates at 1×10 3 and 2×10 3 cells/well. Cells were cultured in DMEM with 10% FBS for 3 weeks, washed twice with PBS and stained with crystal violet for 30 min. Cell colonies were counted in every well.

Wound healing assay
A total of 1×10 6 cells/well, with or without transfection or inhibitors, were inoculated into 6-well plates. After overnight incubation, 20 μl pipette tips were used to scratch the cells. Then, floating cells were washed away in PBS three times. Cells were photographed under a high-powered microscope (2x) at 0, 24 and 48 h.

Cell migration and invasion assays
2×10 5 cells, with or without transfection or inhibitors, were seeded in 200 ul of serum-free DMEM into the upper chambers of transwells (Boyden transwell chambers, Corning, MA, USA), and 600 μl of DMEM with 10% FBS was added into the lower chambers. Cells were cultured for 24 h and filters were stained with crystal violet for 30 min at room temperature. Cells in five random fields were counted under a high-power objective (10x). For invasion assays, upper chamber membranes were coated in matrigel (Becton Dickinson Labware, Bedford, MA, USA).

Statistical analysis
CCK-8 assay, plate colony formation assay, flow cytometry, wound healing assay, cell migration and invasion assay and western blotting data were analyzed with GraphPad Prism 6 software and displayed as means ± SD. Differences between groups were assessed using Student's t test. P<0.05 was considered statistically significant.