Hexokinase 2 Regulates Ovarian Cancer Cell Migration, Invasion and Stemness via FAK/ERK1/2/MMP9/NANOG/SOX9 Signaling Cascades

Metabolic reprogramming is a common phenomenon in cancers. Thus, glycolytic enzymes could be exploited to selectively target cancer cells in cancer therapy. Hexokinase 2 (HK2) converts glucose to glucose-6-phosphate, the first committed step in glucose metabolism. Here, we demonstrated that HK2 was overexpressed in ovarian cancer and displayed significantly higher expression in ascites and metastatic foci. HK2 expression was significantly associated with advanced stage and high-grade cancers, and was an independent prognostic factor. Functionally, knockdown of HK2 in ovarian cancer cell lines and ascites-derived tumor cells hindered lactate production, cell migration and invasion, and cell stemness properties, along with reduced FAK/ERK1/2 activation and metastasis- and stemness-related genes. 2-DG, a glycolysis inhibitor, retarded cell migration and invasion and reduced stemness properties. Inversely, overexpression of HK2 promoted cell migration and invasion through the FAK/ERK1/2/MMP9 pathway, and enhanced stemness properties via the FAK/ERK1/2/NANOG/SOX9 cascade. HK2 abrogation impeded in vivo tumor growth and dissemination. Notably, ovarian cancer-associated fibroblast-derived IL-6 contributed to its up-regulation. In conclusion, HK2, which is regulated by the tumor microenvironment, controls lactate production and contributes to ovarian cancer metastasis and stemness regulation via FAK/ERK1/2 signaling pathway-mediated MMP9/NANOG/SOX9 expression. HK2 could be a potential prognostic marker and therapeutic target for ovarian cancer.


Introduction
Ovarian cancer is the most lethal of all gynecological malignancies worldwide [1]. Its high mortality is mainly due to the late presentation of symptoms at the advanced stages of the disease, often once the cancer has metastasized [2,3]. Primary treatment mainly involves cytoreductive surgery followed by adjuvant chemotherapy. However, even with optimal treatment, recurrences are common, By immunohistochemistry, moderate to strong HK2 protein was localized in the cytoplasm in ovarian cancer samples; in contrast, it was barely detectable in benign cystadenomas ( Figure 1A). The HK2 immunoreactivity in ovarian cancers was significantly higher than in benign cystadenomas (p < 0.001; Supplementary Table S3). High HK2 immunoreactivity was significantly associated with a more advanced stage (Stage 4), higher grade (grade 3), and shorter overall and disease-free survival (all p < 0.05; Supplementary Table S3 and Figure 1B). Moreover, statistically higher HK2 immunoreactivity was detected in metastatic foci than their corresponding primary carcinomas ( Figure 1C). By multivariate analysis, HK2 expression was a significant independent predictor of disease-free survival (p = 0.033; Supplementary Table S4). By western blot analysis, we found an up-regulation of HK2 protein expression in ovarian cancer cell lines (OVCAR-3, OVCA429, OVCA433, OC316, ES-2, TOV21G, A2780S, and A2780CP), compared to normal ovarian epithelial cell lines (HOSE 6-3 and HOSE 11-12) ( Figure 1D).

HK2 Increases Lactate Production
We first detected the specific transient (siHK2; Figure 2A) and stable (shHK2; Figure 2B) knockdown of HK2 in A2780CP and ES-2 cell lines, ovarian cancer cell lines with relatively high HK2 expression. We then examined the effect of HK2 on intracellular lactate production. Results showed that HK2-transiently and stably silenced cells had a significantly reduced lactate level compared to control cells, as assessed by the Lactate Colorimetric Assay Kit II ( Figure 2C).

HK2 Augments Cell Migration and Invasion via the FAK/MEK-1/ERK1/2/MMP9 Signaling Pathway
Our finding of statistically higher HK2 immunoreactivity in metastatic foci compared to their corresponding primary carcinomas prompted us to investigate the effect of HK2 on cell migration and invasion. In a wound-healing assay, a slower migration rate was observed in siHK2 ES-2, shHK2 A2780CP, and ES-2 cells, as compared to control cells ( Figure 2D). In Transwell migration and invasion assays, significantly reduced migration and invasion were detected in A2780CP and ES-2 cells upon siHK2 (Supplementary Figure S1) or shHK2 ( Figure 2E) knockdown. We next determined whether the glucose analog 2-DG, an inhibitor of glycolysis [16], was able to mimic the effects of HK2 silencing in ovarian cancer cells. A2780CP, ES-2 and OVCA 433 cells were treated with vehicle (water) or low doses of 2-DG (0, 0.5, or 2 mM). Migration and invasion assays revealed significantly reduced migration and invasion in 2-DG treated cells, compared to that of control cells ( Figure 2F). An XTT assay revealed no change in cell proliferation 48 h after low doses of 2-DG treatment (Supplementary Figure S2).
Reduced lactate levels after depletion of HK2 may contribute to its inhibitory effect on cell migration and invasion, owing to the observation that lactate can enhance the motility of head and neck tumor cells [17]. However, investigating the possible molecular pathways by which HK2 mediates its effect on cell migration and invasion is still warranted. FAK is a cytoplasmic protein tyrosine kinase that is overexpressed and activated in several advanced-stage solid cancers, including ovarian cancer [18]. FAK is a key component of cell-matrix adhesion complexes, which together with other signaling and adaptor proteins, such as ERK1/2, act to regulate cancer cell migration and invasion via promoting matrix metalloproteinase (MMP) expression [18]. Binding of urokinase-type plasminogen (uPA), a serine protease, to its receptor uPAR converts proenzyme plasminogen to active plasmin. Plasmin, in turn, remodels the extracellular matrix (ECM) and activates growth factors, results in increased cellular invasion and metastasis [19]. Vascular endothelial growth factor (VEGF) is a pro-angiogenic protein produced by cancer cells and endothelial cells, which plays important roles in ECM degradation and angiogenesis, which in turn affects metastasis, including in ovarian cancer [20]. Interestingly, we found that FAK and ERK1/2 activation was inhibited after knockdown of HK2 in A2780CP and ES-2 cells ( Figure 2G). We also found that HK2-silenced A2780CP cells had reduced MMP9, uPA, and VEGF mRNA expression ( Figure 2H). A reduction of MMP-9 and uPA protein levels in conditioned media was also detected ( Figure 2H). GEPIA further revealed a significant clinical correlation between MMP9, uPA, and VEGF and HK2 expression (p < 0.05) in ovarian cancer clinical samples ( Figure 2I). In addition, 2-DG inhibited FAK and ERK1/2 activation, and suppressed MMP-9 mRNA expression in OVCA 433 cells ( Figure 2J).  To further evaluate the effect of HK2 and its downstream signaling pathways in ovarian cancer cell migration and invasion, we stably transfected SKOV-3 cells, an ovarian cancer cell line with relatively low HK2 expression, with DDK-tagged HK2 plasmid or control empty vector. Ectopic expression of HK2 was detected in HK2-transfected cells [western blot analysis using an anti-DDK antibody ( Figure 3A)]. We found that ectopic expression of HK2 enhanced cell migration and invasion ( Figure 3B), as well as increased FAK and ERK1/2 activation ( Figure 3C), along with increased MMP9, uPA and VEGF mRNA expression ( Figure 3D). To unravel the effect of HK2-mediated FAK and ERK1/2 activation on cell migration and invasion, and to elucidate potential links between FAK, ERK1/2, and MMP-9, SKOV-3 cells overexpressed with HK2 were treated with a FAK inhibitor (FAK inhibitor 14), a mitogen-activated protein kinase kinase -1 (MEK-1) inhibitor (U0126), or an anti-MMP9 neutralizing antibody. We found that FAK inhibitor 14 abolished HK2-induced cell migration and invasion ( Figure 3B). As determined by western blot analyses and qPCR, FAK inhibitor 14 attenuated HK2-mediated ERK1/2 activation ( Figure 3C) and MMP-9 mRNA expression ( Figure 3E). Moreover, U0126 abrogated HK2-induced cell migration and invasion, and also inhibited HK2-mediated MMP-9 mRNA expression ( Figure 3B,C,E). Inhibition of HK2-induced cell migration and invasion was also observed following treatment with an anti-MMP9 neutralizing antibody ( Figure 3F). These findings indicate that FAK activates MEK-1/ERK1/2, which in turn mediates HK2-induced MMP-9 expression. We found that HK2 expression was barely detectable in HOSE 6-3 and HOSE 11-12 cells ( Figure 1D), whether induction of HK2 would enhance migration, invasion or other functions, thus progression, of normal ovarian epithelial cells could be further studied in a future study.

HK2 Enhances Cell Proliferation and Anchorage-Independent Growth
We further evaluated the effects of HK2 on ovarian cancer cell growth. Significantly reduced proliferation was observed 11 d after shHK2 knockdown in A2780CP and ES-2 cells, as assessed by cell counting (Supplementary Figure S3A). A soft agar assay revealed that smaller and fewer colonies were formed in A2780CP cells after shHK2 knockdown, whereas and larger and more colonies were formed in SKOV-3 cells with HK2 overexpression (Supplementary Figure S3B), suggesting that HK2 can induce anchorage-independent growth.

HK2 Enhances Cell Proliferation and Anchorage-Independent Growth
We further evaluated the effects of HK2 on ovarian cancer cell growth. Significantly reduced proliferation was observed 11 d after shHK2 knockdown in A2780CP and ES-2 cells, as assessed by cell counting (Supplementary Figure S3A). A soft agar assay revealed that smaller and fewer colonies were formed in A2780CP cells after shHK2 knockdown, whereas and larger and more colonies were formed in SKOV-3 cells with HK2 overexpression (Supplementary Figure S3B), suggesting that HK2 can induce anchorage-independent growth.

Overexpression of HK2 in Ascites-Derived Sphere Forming Cells: HK2 Abrogation Impedes Lactate Production, Metastasis, and CSC Properties in Spheroids Formed from Ascites-Derived Tumor Cells
We further observed by qPCR a progressive increase in the levels of HK2 mRNA from normal ovarian epithelial cell lines to primary tumors to ascites ( Figure 5A). In addition, a significantly higher expression of CSC-related genes (NANOG, OCT4, SOX2, KLF4) was found in spheroids formed from ascites-derived tumor cells, compared to monolayer cells by qPCR ( Figure 5B). HK2 expression ( Figure 5B) and lactate production ( Figure 5C) were also augmented in spheroids. Knockdown of HK2 by siRNA in spheroids formed from ascites-derived tumor cells ( Figure 5D) inhibited lactate production ( Figure 5E), cell migration, invasion, sphere-formation abilities ( Figure  5F), FAK/ERK1/2 activation (Figure 5G), and metastasis-and CSC-related gene expression ( Figure  5H). 2-DG also retarded migration, invasion and sphere-forming abilities, compared to control cells ( Figure 5F). These findings suggest that spheroids formed from ascites-derived tumor cells endure CSCs properties, and that this subpopulation of cells has enhanced glycolysis.

Overexpression of HK2 in Ascites-Derived Sphere Forming Cells: HK2 Abrogation Impedes Lactate Production, Metastasis, and CSC Properties in Spheroids Formed from Ascites-Derived Tumor Cells
We further observed by qPCR a progressive increase in the levels of HK2 mRNA from normal ovarian epithelial cell lines to primary tumors to ascites ( Figure 5A). In addition, a significantly higher expression of CSC-related genes (NANOG, OCT4, SOX2, KLF4) was found in spheroids formed from ascites-derived tumor cells, compared to monolayer cells by qPCR ( Figure 5B). HK2 expression ( Figure 5B) and lactate production ( Figure 5C) were also augmented in spheroids. Knockdown of HK2 by siRNA in spheroids formed from ascites-derived tumor cells ( Figure 5D) inhibited lactate production ( Figure 5E), cell migration, invasion, sphere-formation abilities ( Figure 5F), FAK/ERK1/2 activation ( Figure 5G), and metastasis-and CSC-related gene expression ( Figure 5H). 2-DG also retarded migration, invasion and sphere-forming abilities, compared to control cells ( Figure 5F). These findings suggest that spheroids formed from ascites-derived tumor cells endure CSCs properties, and that this subpopulation of cells has enhanced glycolysis.

HK2 Heightens Tumor Growth and Dissemination in Nude Mice
To verify the effect of HK2 on in vivo tumor growth and dissemination, shHK2 ES-2 and control cells were inoculated s.c. or i.p. into nude mice. Significantly reduced tumor growth was detected in HK2 knockdown mice ( Figure 6A). Fourteen days after i.p. inoculation, widespread abdominal dissemination, primarily in the mesentery, was observed in control mice. On the contrary, mice with HK2-silenced cells revealed only focal nodules in the mesentery ( Figure 6B). The total i.p. tumor weight in the shHK2 ES-2 cell-injected mice (0.078 ± 0.031 g) was significantly lower than that of control mice (0.199 ± 0.066 g; p < 0.05).

HK2 Heightens Tumor Growth and Dissemination in Nude Mice
To verify the effect of HK2 on in vivo tumor growth and dissemination, shHK2 ES-2 and control cells were inoculated s.c. or i.p. into nude mice. Significantly reduced tumor growth was detected in HK2 knockdown mice ( Figure 6A). Fourteen days after i.p. inoculation, widespread abdominal dissemination, primarily in the mesentery, was observed in control mice. On the contrary, mice with HK2-silenced cells revealed only focal nodules in the mesentery ( Figure 6B). The total i.p. tumor weight in the shHK2 ES-2 cell-injected mice (0.078 ± 0.031 g) was significantly lower than that of control mice (0.199 ± 0.066 g; p < 0.05).

CAF-Derived IL-6 Regulates HK2 in Ovarian Cancer Cells via IL6R
The tumor microenvironment is indispensable in ovarian cancer-metastatic niche formation, yet the underlying mechanisms remain unexplored [26]. CAF constitutes the main component of the tumor microenvironment [27]. To explore the possible upstream mechanisms leading to HK2 upregulation in ovarian cancer cells, we determined whether ovarian CAF-CM regulates HK2 expression in SKOV-3 cells. We found that CAF-CM attenuated HK2 expression ( Figure 6C). CAF contributes to ovarian cancer progression and metastasis through secretion of chemokines and ECM [27]. Given that IL-6 is one of the major cytokines upregulated in CAFs [2,26], we found that CAF-CM-induced HK2 expression was abolished by neutralizing IL-6 and IL-6R antibodies ( Figure 6C). Moreover, we revealed increased HK2 ( Figure 6D) and lactate production in IL-6 treated SKOV-3 cells (Figure 6E), and a reversal of IL-6-induced HK2 was detected after treatment with neutralizing IL-6 and IL-6R antibodies ( Figure 6D). We also investigated if IL-8, another major cytokine upregulated in CAFs [2,26], could regulate HK2 expression. Results revealed that neutralizing IL-8 and IL-8R antibodies could not block CAF-CM-induced HK2 expression ( Figure 6C). Moreover, HK2 expression remained unchanged after IL-8 treatment in the presence or absence of neutralizing IL-8 and IL-8R antibodies ( Figure 6D).

Discussion
Altered glucose metabolism is one of the hallmarks of cancer [8]. Due to the lower efficiency of ATP generation by aerobic glycolysis, cancer cells take up more glucose and produce more lactate than normal cells. Such increased glucose consumption distinguishes many human cancer cells from their normal counterparts, thus 18 FDG-PET imaging, a measure of glucose consumption, is a common technology used to detect tumors in clinical use [9]. Studies on altered glucose metabolism in ovarian cancer are limited. It is known that more invasive ovarian cancer cells display higher glucose uptake and lactate production [28]. Moreover, lactate levels were shown to be increased in both primary and metastatic ovarian cancer when compared to normal ovarian tissue [15]. Our findings revealed higher HK2 expression in ovarian cancer, which was correlated with shorter overall and disease-free survival. These results suggest that HK2 is a significant prognostic marker in ovarian cancer. We further discovered that HK2 is a significant independent predictor of disease-free survival. Additionally, we show for the first time to the best of our knowledge that higher HK2 expression is present in metastatic foci compared to their matched primary tumors. Up-regulation of HK2 was also detected in ovarian ascitic fluid samples compared to primary tumor cells and HOSE cells. Moreover, knockdown of HK2 resulted in lower lactate production in ovarian cancer cells. All of these findings indicate that HK2 may be one of the glycolytic genes responsible for the metabolic switch from primary to metastatic tumors.
The above described HK2 expression patterns, combined with our in vitro and in vivo experiments on cell migration and invasion, highlight the contribution of HK2 to ovarian cancer metastasis. We also demonstrated a link between HK2 and FAK, MEK-1/ERK1/2, MMP9, uPA, and VEGF in the regulation of cell migration and invasion. Mechanistically, we have provided evidence that the activation of FAK/MEK-1/ERK1/2 signaling, which in turn induces MMP-9 expression, is involved in HK2-mediated regulation of cell migration and invasion. This pathway is linked to the formation and turnover of focal adhesions, which is vital for tumor cell invasion and metastasis [18], including in ovarian [29,30] and gastric [31] cancers. Moreover, uPA and VEGF expression have both been shown to be associated with tumor stage, metastasis and patient survival in ovarian cancer [32,33]. However, the mechanism by which HK2-mediates an up-regulation of uPA and VEGF needs to be elucidated in more detail. A humanized monoclonal antibody to VEGF-A, bevacizumab, is the only FDA approved anti-angiogenic agent for the treatment of ovarian cancer. However, a modest response to resistance limits its clinical efficacy [33]. The present link between HK2 and VEGF suggests that dual targeting of HK2 and VEGF in ovarian or other cancers may be a promising alternative therapeutic approach, which should be evaluated in future studies.
The progressive increase in HK2 levels from normal ovarian epithelial cells, to primary tumors, to ascites, as well as enhanced expression of HK2 and lactate production in ascites-derived sphere forming cells compared to monolayer cells, supports the notion that ascites bear CSC characteristics and harbor enhanced glycolysis [4]. Apart from the well-established effects on cellular metastasis, FAK has been shown to regulate the self-renewal and tumor-initiating capabilities of CSCs in many types of cancer, such as in breast cancer [21,22], Ductal Carcinoma In Situ [34], mesothelioma [35], liver cancer [24], and squamous cell carcinomas [36]. MEK/ERK activation has also been documented to be essential for mediating cancer stemness activities in breast [23] and liver [25] cancers. In line with this, the treatment of breast and liver cancer cell lines with FAK inhibitor 14 has been found to decrease NANOG expression [22,24]. In addition, ERK signaling was found to be involved in the up-regulation of SOX9 by fibroblast growth factors in chondrocytes [37]. In addition to the effects on cell migration and invasion, we believe our identification of the downstream pathways of HK2-Namely the HK2 to FAK to MEK-1/ERK1/2 pathway that controls NANOG and SOX9 expression and ovarian cancer CSC properties-Will be of significant importance to future studies. Besides NANOG and SOX9, we also revealed an up-regulation of OCT4, KLF4, and CD117 by HK2 in ovarian cancer, which further supports the concept that HK2 regulates CSCs.
The present link between HK2 and FAK/MEK-1/ERK1/2 activation, leading to MMP9/NANOG/ SOX9 expression and increased ovarian cancer metastasis and CSC properties, suggests that targeting HK2/FAK/MEK-1/ERK1/2 signaling may be a promising therapeutic alternative, either as mono therapy or in combination with other treatments. 2-DG, a glucose analog, enters cells via GLUTs, thereby inhibiting glucose uptake. Once inside the cell, 2-DG is phosphorylated by HKs to form 2-DG-6-phosphate, which is not further metabolized. The latter accumulates and noncompetitively inhibits HKs and competitively inhibits PGI, thus halting glycolysis [38]. A phase I clinical trial of 2-DG alone or in combination with docetaxel in patients with advanced solid tumors has shown that 2-DG is a safe agent for clinical use [39]. Our data also showed a blockage of metastasis and sphere formation by 2-DG in vitro, suggesting the activity of HK2 is essential for its metastasis and stemness effects in ovarian cancer. Since 2-DG is considered as an effective glycolysis inhibitor that noncompetitively inhibits HKs, there is a need for the development of small molecule inhibitors that specifically inhibit HK2. Defactinib, a small-molecule and well-tolerated FAK inhibitor, has been evaluated in phase I/Ib study combined with paclitaxel in patients with relapsed ovarian cancer (NCT01778803), and in a phase II study in patients with the NSCLC KRAS mutation (NCT01951690). It is also currently being evaluated in a phase 1/1b study combined with avelumab, an anti-PD-L1 antibody, in patients with relapsed ovarian cancer (NCT02943317). In a phase 3 study, Trametinib, a selective MEK inhibitor, was shown to improve overall and progression-free survival in patients with metastatic melanoma, as compared to chemotherapy alone [40]. A phase 2 clinical trial of another MEK inhibitor, selumetinib, also displayed good tolerance and an active effect in patients with recurrent low-grade serous carcinoma of the ovary or peritoneum [41]. Taken together, targeting HK/FAK/MEK-1/ERK1/2 signaling either alone or in combination is possible, and can be evaluated for the future treatment of ovarian cancers.
Besides cell metastasis and CSC regulation, we also demonstrated enhanced cell proliferation and anchorage-independent growth of HK2 in ovarian cancer cells, in line with the well-established HK2 tumor-promoting effect observed in other cancers in vivo, such as glioblastoma [12], medulloblastoma [13], and breast cancer [14]. Lysophosphatidic acid, a blood-borne lipid mediator that is elevated in the ascites of ovarian cancer patients, has been shown to up-regulate HK2 and glycolysis, leading to enhanced proliferation of ovarian cancer cells [42]. A recent study also revealed that HK2 contributes to ovarian cancer cisplatin resistance by regulating cisplatin-induced, ERK-mediated autophagy [43].
Mounting evidence suggests that the tumor microenvironment plays an important role in ovarian cancer progression and metastasis [26]. CAFs transform from normal fibroblasts in the stroma after interacting with ovarian cancer cells, which constitutes more than half the tumor microenvironment [27]. CAFs plays important roles in cell growth, adhesion, invasion, and metastasis by secreting chemokines and ECM, facilitating dissemination [26]. In this study, we found that CAF-CM up-regulates HK2 in ovarian cancer. Moreover, we show that IL-6, but not IL-8, is one of the major cytokines in CAF-CM invasion assays were performed as previously described [44][45][46]. Cells at a density of 1.25 × 10 5 were plated on the upper compartment of a Transwell chamber, and were allowed to migrate through a membrane (8 µm pore size)/invaded through a Matrigel-coated membrane. After 12 to 48 h, cells on the upper side of the membrane were removed. The migrated/invaded cells were fixed, stained, and counted. For drug treatment, cells were plated on the upper side of a Transwell chamber for 6 h before treatment with 2-DG (0, 0.5 and 2 mM), FAK inhibitor 14 (10 µM), U0126 (10 µM), or vehicle. For treatment using the anti-MMP9 neutralization antibody, cells were pre-treated with the neutralization antibody or the mouse IgG for overnight before cell counting and cell plating.

XTT Assay, Cell Counting, and Soft Agar Assay
For the XTT assay, 2000 cells/well were seeded in 96-well plates. Five days after incubation, cell proliferation was measured using the Cell Proliferation Kit II (Roche) according to the manufacturer's instructions in an Infinite ® 200 microplate reader at 492 nm (Tecan Group Ltd., Männedorf, Switzerland) [46]. For cell counting, 3 × 10 4 cells were seeded in 12-or 6-well plates or T150 culture flasks, and maintained in growth media. Cell numbers were counted at days 1 (12-well culture plates), 4 (6-well culture plates), 7, and 11 (T150 culture flasks), using a Luna™ automated cell counter (Logos Biosystems, Annandale, VA, USA) [44]. For the soft agar assay, 2 × 10 4 cells were suspended in 2 mL 0.4% agar and seeded on 1% agar in 6-well plates. After 4 weeks, cells were counted.

Sphere-Formation Assay
Cells were seeded and cultured as spheroid cultures, as above. After 9-13 days, spheres were counted and imaged under an inverted microscope. Spheres that <50 µm and individual or aggregated cells were not counted as spheres. For drug treatment, plated cells were treated with 2-DG (0 and 2 mM), FAK inhibitor 14 (10 µM), U0126 (10 µM), or vehicle.

In Vivo Analyses
2 × 10 6 ES-2 cells with stably knocked-down HK2 were injected subcutaneously (s.c.; five mice/group) or intraperitoneally (i.p.; seven mice/group) into BALB/c female nude mice [44]. Perpendicular tumor diameters were measured on days 7, 11, and 14, and tumor volumes were calculated. Fourteen days post cell injection, mice were sacrificed, and tumor dissemination was recorded. Experiments were performed following the Animals (Control of Experiments) Ordinance (Hong Kong) and the Institute's guidance on animal experiments.

Statistical Analyses
Statistical analyses were performed using SPSS 20 for Windows (SPSS Inc., Chicago, IL, USA). The Mann-Whitney test was used for comparing data between two groups while the Kruskal-Wallis rank test was performed for multiple comparisons. Kaplan-Meier and log-rank tests were used for survival analyses. Multivariate survival analyses were performed using a Cox regression analysis. p values <0.05 were considered statistically significant.

Conclusions
In conclusion, we describe in this study the overexpression of HK2 in ovarian cancer. Notably, high HK2 expression was associated with cancer metastasis and poor patient clinical outcomes. CAFs secrete IL-6 in the tumor microenvironment, which contributes to an up-regulation of HK2 via the IL-6R ( Figure 6F). We also revealed that HK2 controls a metabolic switch, which promotes cell migration, invasion, CSC properties, proliferation, and anchorage-independent growth in ovarian cancer cells. We demonstrated that the mechanisms regulating cell migration/invasion and CSC properties involve FAK/MEK-1/ERK1/2/MMP9/NANOG/SOX9 signaling ( Figure 6F). Improved understanding of the involvement of HK2 in metastasis and in CSC properties will facilitate its effective application as a therapeutic molecular target, either alone or in combination with other treatments.