This article is mentioned in:

Introduction

Kisspeptin is a protein encoded by the KISS1 gene in humans, which has been reported to serve a diverse role in various physiological and pathological conditions (1–5). Kisspeptin has been reported to suppress the metastatic capabilities of various types of cancer cells, through the activation of its G-protein coupled receptor (GPR) 54 (5–7). Kisspeptin has been demonstrated to activate various cytosolic protein kinases, including B, C and D in several types of cells in vitro (8–10). However, kisspeptin-mediated signaling pathways are complex, and the molecular mechanisms underlying the involvement of kisspeptin-induced intracellular signaling in the regulation of the migratory and invasive capabilities of cancer cells have yet to be fully elucidated (9,11–13).

The eukaryotic translation initiation factor 2α kinase 2 (EIF2AK2; also known as double-stranded RNA-activated protein kinase, PKR), has been revealed to be ubiquitously expressed across numerous cell types, and may be activated under various types of cellular stress, including viral infections, hypoxia and nutritional depletion (14). EIF2AK2 activation has been reported to suppress tumor growth (15–18); however, its expression and activity levels have been demonstrated to be increased in various types of cancer cells, including breast, lung and colorectal cancer cells (19–21). It has previously been reported that invasive ductal carcinoma cells exhibit increased EIF2AK2 expression levels compared with adjacent normal breast tissue (21); however, the expression and activity levels of EIF2AK2 have been negatively correlated with the metastatic capabilities of breast cancer cells (22). Furthermore, the double-stranded RNA-mediated activation of EIF2AK2 has been implicated in the regulation of breast cancer cell motility (23). These results suggest that EIF2AK2 activity may be critical for the suppression of cancer metastasis. Conversely, EIF2AK2 has been reported to be implicated in hepatocellular carcinoma growth and migration, as well as gastric cancer metastasis (24,25). Therefore, it may be hypothesized that its anti-tumor or oncogenic role is determined by the type of cancer cell.

Previous studies have proposed a role for EIF2AK2 in the regulation of cytoskeletal dynamics (23,26), which appear to be similar to the roles of kisspeptin in the regulation of cellular architecture (7,8,13). However, whether kisspeptin may inhibit cancer cell migration and invasion through the modulation of EIF2AK2 has yet to be elucidated. Therefore, the present study aimed to investigate the role of EIF2AK2 in the mechanisms underlying the inhibitory effects of kisspeptin on cancer metastasis. The results of the present study demonstrated that kisspeptin activated EIF2AK2 in several highly metastatic types of cancer cells. Furthermore, kisspeptin failed to suppress the migratory and invasive capabilities of mouse embryonic fibroblasts (MEFs) following the deletion of the EIF2AK2 gene. An in vivo experimental metastasis assay revealed that kisspeptin-induced EIF2AK2 activation attenuated cancer cell migration and invasion, thus suppressing distant metastasis to the lungs. In addition, kisspeptin was demonstrated to promote the interaction between the Ras homolog gene family member A (RhoA) and EIF2AK2, resulting in the autophosphorylation of EIF2AK2. The present results suggest that kisspeptin may suppress the metastatic capabilities of cancer cells via activation of EIF2AK2. Therefore, kisspeptin treatment may be one of options in preventing cancer metastases.

Materials and methods

Cell lines and reagents

The following cell lines used in the present study were kindly provided by Professor Barat B. Aggarwal (University of Texas, Anderson Cancer Center, Houston, Texas): wild-type MEFs (MEF-EIF2AK2+/+), MEFs where the EIF2AK2 gene was deleted (MEF-EIF2AK2−/−), human breast SK-BR-3, prostatic PC-3 and colorectal LoVo adenocarcinoma cells. Cells were cultured in Dulbecco's modified Eagle's medium or RPMI-1640 (Cellgro; Corning Incorporated, Corning, NY, USA), supplemented with 10% fetal bovine serum (Cellgro; Corning Incorporated) and 1% penicillin/streptomycin (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and 5% CO2 humidified air at 37°C. Kisspeptin (C-terminal amidated kisspeptin-10) was purchased from Tocris Bioscience (Bristol, UK). The Rho-associated protein kinase (ROCK) inhibitor Y-27632 was purchased from Selleck Chemicals (Houston, TX, USA). The EIF2AK2 inhibitor C16 was purchased from Santa Cruz Biotechnology, Inc. (sc-204200; Dallas, TX, USA).

Short interfering (si)RNA experiment

3×106 LoVo cells were transfected with EIF2AK2 siRNA or negative control siRNA using Lipofectamine 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.). In brief, 1×106 LoVo cells were cultured in DMEM medium without antibiotics a day before transfection. Then, 20 µl of Lipofectamine 2000 reagent in 500 µl of serum-free DMEM medium and 10 µl of each siRNA in 500 µl of serum-free DMEM medium were gently mixed and incubated for 30 min at room temperature, and the mixed solution was added in the cells. The cells were then incubated at 37°C in a CO2 incubator for 24 h and then used for the assays such as the migration and invasion assay. EIF2AK2 siRNA and negative control siRNA were purchased from Santa Cruz Biotechnology, Inc. These siRNA products consist of pools of three to five target-specific 19 to 25 nucleotide siRNAs designed to knock down gene expression. EIF2AK2 silencing was confirmed by western blot analysis.

Western blot analysis

A total of 3×106 cells (all cell types used in this study) were lyzed using radioimmunoprecipitation lysis buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 1% NP-40, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na3VO4, 1 µg/ml leupeptin, 1 mM PMSF) for 30 min on ice and centrifuged at 20,000 × g for 10 min at 4°C. Protein concentrations were measured using Pierce BCA protein assay kit (Pierce; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. Equal amounts of extracted protein samples (30 µg) were separated by 6–12% SDS-PAGE and transferred onto polyvinylidene difluoride membranes. Subsequent to blocking with 5% milk for 1 h at room temperature, membranes were probed for 1 h at room temperature with the following primary antibodies: Anti-phosphorylated (p)-cofilin (sc-271921; 1:250; Santa Cruz Biotechnology, Inc.), anti-cofilin (sc-376476; 1:250; Santa Cruz Biotechnology, Inc.), anti-p-myosin light chain (MLC; 3674; 1:1,000; Cell Signaling Technology, Inc., Danvers, MA, USA), anti-MLC (8505; 1:1,000; Cell Signaling Technology, Inc.), anti-RhoA (sc-166399; 1:250; Santa Cruz Biotechnology, Inc.), anti-EIF2AK2 (sc-6282; 1:250; Santa Cruz Biotechnology, Inc.) anti-p-EIF2AK2 (3075; 1:1,000; Cell Signaling Technology, Inc.) and anti-β-actin (4967; 1:1,000; Cell Signaling Technology, Inc., Danvers, MA, USA). The membranes were then incubated in secondary antibodies conjugated to horseradish peroxidase (anti-rabbit; 7074; 1:10,000, anti-mouse; 7076; 1:10,000; Cell Signaling Technology, Inc.) for 1 h at room temperature. Bands were detected using LumiGLO chemiluminescent reagent and peroxidase (7003; Cell Signaling Technology, Inc.). Active GTP-bound RhoA was detected by a glutathione-S-transferase (GST) pull down assay using a human Rhotekin Rho binding domain (RBD)-GST-fusion protein.

Scratch wound assay

An in vitro scratch-wound healing assay was used to evaluate cellular migration; 3×105 cells (from all cell lines tested in the present study) were cultured in 6-well plates containing DMEM medium at 37°C in CO2 incubator for 24 h and scratched when cell confluence reached about 80%. A ‘scratch’ was created by scraping the cell culture in a straight line with a 200 µl pipette tip and cells were washed with PBS three times to remove debris. The cells were then treated with kisspeptin at 100 nM for 24 h at 37°C and then random cell migration was evaluated via counting the number of cells that had migrated in the scratched region. Experiments were performed in triplicate. Non-treated cells were used as a control in the migration assay. Cell migration was observed using a Zeiss Axiovert inverted microscope, and images were analyzed using Zen software version 3.00 (Carl Zeiss AG, Oberkochen, Germany).

Cell invasion analysis

An invasion assay was used to evaluate the invasive capabilities of cancer cells; 3×105 cells (from all cell lines tested in this study) were seeded into the upper chambers of Transwell inserts (Corning Incorporated, Corning, NY, USA) pre-coated with Matrigel (BD Biosciences, San Jose, CA, USA). The upper chambers were filled with DMEM medium supplemented with 1% serum, whereas the lower chambers were filled with DMEM medium supplemented with 10% serum as a chemoattractant. The cells in the upper chamber were treated with kisspeptin at 100 nM for 24 h. Non-treated cells were used as a control. Following incubation at 37°C for 24 h, cells on the top of the membrane were carefully removed with cotton swabs. Cells that had invaded the lower membrane were fixed with 4% formaldehyde at room temperature for 5 min, stained with 1% crystal violet at room temperature for another 5 min, and then counted. Experiments were performed in triplicate. Cell invasion was observed using a Zeiss Axiovert inverted microscope, and images were analyzed using Zen software version 3.00 (Carl Zeiss AG). A total of 4 fields were randomly selected and the invaded cells were counted.

In vivo studies

Cultured LoVo cells (3×105) were subcutaneously injected into the back skin of 6-week old male nude mice (Orient Bio Inc., Seongnam, Korea). A day after tumor cell injection, kisspeptin (100 nM) alone (n=6), or in combination with C16 (200 nM) (n=6) or control (saline, n=6) was subcutaneously injected every other day into the tumor site for 14 days. Tumor volume was measured every other day, according to the following formula: Tumor volume (mm3) = 0.5 × (tumor length, mm) × (tumor width, mm)2. Mice were sacrificed 14 days after treatment, and tumors were removed for immunohistochemical analysis. For the experimental metastasis assay, LoVo cells (1×106) were treated with kisspeptin (100 nM) alone or in combination with C16 (200 nM) for 1 h at 37°C, and then injected into the tail vein of the mice. A total of 2 weeks following the injection, mice were sacrificed, the lungs were isolated and metastatic foci were counted. Mice were maintained under a 12-h light/dark cycle with free access to food and water and diet at 24±1°C and a relative humidity of 40±5%. All experimental procedures were approved by the Institutional Animal Care and Use Committee of the Kyung Hee University (Seoul, Republic of Korea).

Immunohistochemistry

Tumor xenografts were removed, cut and put on cassettes, fixed with 4% formaldehyde at room temperature for 24 h and embedded in paraffin wax. Tumor tissue was sectioned at 10 µm, deparaffinized, and antigen-retrieval performed with 8 M Urea, H2O2 was used to block endogenous peroxidase. The sections were washed with TBST (0.1% Tween-20) twice, and blocked in 10% normal serum with 1% BSA for 1 h at room temperature, stained with an anti-p-EIF2AK2 antibody (ab32036; Abcam, Cambridge, UK) for 1 h at room temperature, incubated with anti-rabbit secondary antibody (ab205718; Abcam) for another 1 h at room temperature and then incubated with DAB for 2 min at room temperature using DAB substrate kit (ab64238; Abcam). Stained sections were observed under ×20 magnification. Cell invasion was observed using a Zeiss Axiovert inverted microscope, and images were analyzed using Zen software version 3.00 (Carl Zeiss AG).

Statistical analysis

The statistical significance of the differences between groups was assessed using unpaired Student's t-test or one-way analysis of variance with a post-hoc Tukey's test using SPSS version 22 (IBM Corp., Armonk, NY, USA). All in vitro experiments were conducted at least three times. Data are expressed as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

Results

EIF2AK2 is involved in kisspeptin-mediated inhibition of cellular migration and invasion

To investigate whether EIF2AK2 may be involved in the mechanism underlying the inhibitory effects of kisspeptin on cellular migration, wild-type MEFs (MEF-EIF2AK2+/+) and MEFs where the EIF2AK2 gene was deleted (MEF-EIF2AK2−/−) received a scratch wound and were subsequently treated with kisspeptin for 24 h. EIF2AK2 expression in MEF-EIF2AK2+/+, and its absence in MEF-EIF2AK2−/− cells were confirmed using western blot analysis (Fig. 1A). Treatment with kisspeptin was revealed to inhibit the migratory capabilities of normal MEFs; however, in MEFs where EIF2AK2 was knocked out, the inhibitory effects of kisspeptin on cellular migration were abolished (Fig. 1B). Similarly, kisspeptin failed to suppress the invasive capabilities of MEF-EIF2AK2−/− cells (Fig. 1C). These results suggested that EIF2AK2 may be involved in the mechanisms underlying the inhibitory effects of kisspeptin on cellular migration and invasion.

Figure 1. EIF2AK2 is involved in kisspeptin-mediated inhibition of cellular migration and invasion. (A) Western blot analysis was used to assess the protein expression levels of EIF2AK2 in wild-type MEFs (MEF-EIF2AK2+/+) and MEFs where the EIF2AK2 gene was knocked out (MEF-EIF2AK2−/−). (B) A scratch wound assay was used to evaluate the migratory capabilities of MEFs. A scratch wound was inflicted and cells were treated with kisspeptin (100 nM). The number of cells migrated into the wound area was counted. (C) A Matrigel invasion assay was used to evaluate the invasive capabilities of MEFs. Kisspeptin (100 nM) was added to the upper chambers and the number of invaded cells was counted. Data are expressed as the mean ± standard deviation of 3 independent experiments. *P<0.05 vs. control. EIF2AK2, eukaryotic translation initiation factor 2α kinase 2; MEF, mouse embryonic fibroblast; WT, wild-type; KO, knock-out.

Kisspeptin-mediated EIF2AK2 activation is crucial for the inhibition of cancer cell migration and invasion

To investigate the effects of kisspeptin on the migratory and invasive capabilities of cancer cells, SK-BR-3, PC-3 and LoVo cells were treated with kisspeptin for 24 h and subsequently evaluated in scratch wound and invasion assays. The present results revealed that kisspeptin significantly suppressed cancer cell migration and invasion (Fig. 2A and B).

Figure 2. Kisspeptin-mediated EIF2AK2 activation is involved in the inhibition of cancer cell migration and invasion. Human SK-BR-3 breast, PC-3 prostatic and LoVo colorectal adenocarcinoma cells were treated with kisspeptin (100 nM) and their (A) migratory and (B) invasive capabilities were assessed. (C) RNA interference was used to silence the expression of EIF2AK2 in LoVo cells. The (D) migratory and (E) invasive capabilities of LoVo cells were assessed following knockdown of EIF2AK2 expression. Data are expressed as the mean ± standard deviation of 3 independent experiments. *P<0.05 vs. control. EIF2AK2, eukaryotic translation initiation factor 2α kinase 2; sh, short hairpin.

RNA interference was used to silence the expression of EIF2AK2 in LoVo cells (Fig. 2C). Following EIF2AK2 knockdown, the inhibitory effects of kisspeptin on cancer cell migration and invasion were abolished (Fig. 2D and E). The results of the present study suggested that kisspeptin may exert its suppressive effects on the metastatic capabilities of cancer cells via an EIF2AK2-mediated pathway.

Kisspeptin-mediated EIF2AK2 phosphorylation requires RhoA activation

Kisspeptin has been reported to modulate the RhoA intracellular signaling pathway (8,12,27). In LoVo and SK-BR-3 cells, a GST pull-down assay demonstrated that kisspeptin activated RhoA (Fig. 3A). Western blot analysis revealed that treatment with kisspeptin increased the phosphorylation of the downstream mediators of RhoA, MLC and cofilin (Fig. 3A). Notably, the inhibition of EIF2AK2 with C16 (100 nM) did not appear to prevent the activation of RhoA in LoVo cells (Fig. 3B). However, the inhibition of the downstream effector of RhoA ROCK with Y-27632 (100 nM) appeared to prevent the phosphorylation of EIF2AK2 in LoVo cells (Fig. 3C). These results suggested that RhoA-mediated signaling may be implicated in the mechanisms underlying the kisspeptin-induced phosphorylation of EIF2AK2.

Figure 3. Kisspeptin-mediated EIF2AK2 phosphorylation requires RhoA activation. (A) Western blot analysis was used to assess the protein expression levels of RhoA, cofilin and MLC in human SK-BR-3 breast and LoVo colorectal adenocarcinoma cells following treatment with kisspeptin (100 nM) for 15 min. (B) LoVo cells were treated with kisspeptin (100 nM) and/or the EIF2AK2 inhibitor C16 (200 nM) for 15 min and western blot analysis was used to assess the protein expression levels of RhoA and EIF2AK2. (C) LoVo cells were treated with kisspeptin (100 nM) and/or the Rho-associated protein kinase inhibitor Y27632 (100 nM) for 15 min and western blot analysis was used to assess the protein expression levels of EIF2AK2. β-actin was used as the loading control. EIF2AK2, eukaryotic translation initiation factor 2α kinase 2; RhoA, Ras homolog gene family member A; MLC, myosin light chain; p-phosphorylated.

Kisspeptin-induced EIF2AK2 activation inhibits cancer metastasis in vivo

Kisspeptin has been reported to suppress cancer metastasis (6,7). The aforementioned results suggested that kisspeptin-induced EIF2AK2 activation may be involved in the inhibitory effects on cancer cell motility. Therefore, the implication of EIF2AK2 in the mechanism of action of kisspeptin was investigated in cancer metastasis in vivo. LoVo-derived tumor xenografts were implanted into nude mice, and kisspeptin administration was revealed to suppress tumor growth (Fig. 4A). Notably, the inhibition of EIF2AK2 following C16 co-administration appeared to abolish the inhibitory effects of kisspeptin in vivo (Fig. 4A). Similarly, following treatment with kisspeptin, phosphorylation levels of EIF2AK2 in xenografted tumor tissue appeared to be increased, whereas C16 effectively prevented EIF2AK2 phosphorylation (Fig. 4B).

Figure 4. Kisspeptin inhibits tumor growth and metastasis in vivo. (A) Mice were subcutaneously injected with human LoVo colorectal adenocarcinoma cells to develop a tumor xenograft model. Mice were then treated with kisspeptin alone or in combination with the EIF2AK2 inhibitor C16 and tumor growth was evaluated for 14 days. (B) Tumor xenograft tissue samples were obtained from mice and p-EIF2AK2 protein expression was examined using immunohistochemistry. Photomicrographs were captured under ×20 magnification. Scale bars, 100 µm. (C) The effects of kisspeptin on the metastatic capabilities of cancer cells were assessed in vivo. LoVo cells were injected into the tail vein of nude mice and pulmonary metastasis was examined 2 weeks following the injection. Lung images are shown on the top of bars in the graph. *P<0.05. EIF2AK2, eukaryotic translation initiation factor 2α kinase 2; p-, phosphorylated.

An experimental metastasis assay was used to investigate the effects of kisspeptin on the metastatic capabilities of cancer cells in vivo. The present results demonstrated that kisspeptin prevented the development of LoVo-derived lung metastases in nude mice, whereas EIF2AK2 inhibition with C16 abolished the inhibitory effects of kisspeptin on the metastatic capabilities of LoVo cells (Fig. 4C). These results suggested that kisspeptin may inhibit cancer growth and metastasis in vivo through the activation of EIF2AK2-mediated signaling pathways.

Discussion

Kisspeptin is a protein encoded by the KISS1 gene, which has been reported to suppress the metastatic capabilities of various types of cancer cells (6,7), and KISS1 expression levels have been revealed to be downregulated in metastatic compared with non-metastatic cancer tissue (28). However, those findings inversely indicate that KISS1 expression level is higher in tumor tissues than in normal tissues. Conversely, the G-protein coupled receptor for kisspeptin, GPR54, has been implicated in breast cancer development (27). These contradictory results suggest that the effects of kisspeptin during cancer development and metastasis may differ across various types of cancer cells (5). Kisspeptin-mediated intracellular signaling has been reported to serve diverse roles in physiological and pathological conditions; however, the implication of kisspeptin-mediated pathways in the regulation of cancer cell migration and invasion has yet to be elucidated. The results of the present study suggested that kisspeptin-mediated signaling suppressed the metastatic capabilities of cancer cells, and EIF2AK2 may be implicated in the molecular mechanisms underlying these effects.

In the present study, kisspeptin-mediated inhibition of MEF migration and invasion was abolished following the deletion of the EIF2AK2 gene, thus suggesting that EIF2AK2-mediated signaling pathways may be involved in kisspeptin-induced inhibition of cellular migration and invasion. Similarly, kisspeptin-mediated EIF2AK2 activation was demonstrated to be required for the kisspeptin-induced suppression of the migratory and invasive capabilities of cancer cells. The present results are in accordance with a previous study reporting the implication of EIF2AK2 in the regulation of breast cancer cell migration (23). Therefore, it may be hypothesized that EIF2AK2 can be activated by kisspeptin, and kisspeptin regulates cancer cell motility via EIF2AK2-mediated signaling pathways.

Kisspeptin has been reported to activate RhoA via a GPR54/Gq/11/p63 RhoA-specific guanine nucleotide exchange factor (p63RhoGEF)-mediated signaling pathway (27). The present study suggested that RhoA may be required for the kisspeptin-induced activation of EIF2AK2. Therefore, it may be hypothesized that the signaling pathway underlying the effects of kisspeptin may be represented by the following scheme: GPR54/Gq/11/p63RhoGEF/RhoA/EIF2AK2.

The results of the present study suggested that autophagy may be linked to cytoskeletal dynamics (29), since the roles of EIF2AK2 in the regulation of autophagy have been previously demonstrated (30). However, the effects of kisspeptin in the regulation of autophagy have yet to be elucidated, and further studies are required to investigate whether kisspeptin-mediated EIF2AK2 activation may modulate autophagy, as well as the association between autophagy and cytoskeletal dynamics. Vesicle trafficking has been associated with cytoskeletal dynamics (31). Therefore, it may be hypothesized that the trafficking of autophagosomes by cytoskeletal components serves a role in cancer cell motility, as well as in their metastatic capabilities. However, further studies are required to investigate the putative relationship between autophagy and cell motility. In conclusion, the present study suggested that kisspeptin may regulate the metastatic capabilities of cancer cells via EIF2AK2-associated signaling pathways in vitro and in vivo. The present study, therefore, may improve our knowledge of the role of kisspeptin as a metastasis suppressor and address its pharmaceutical application in cancer treatment.

Acknowledgements

The present study was supported by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT and Future Planning (grant no. NRF-2014R1A1A1035831).

References