INTRODUCTION

Laryngeal cancer is the second most common malignant tumor of the respiratory tract after lung cancer (Global Burden of Disease Cancer Collaboration et al., 2015). At present, the main treatment method for laryngeal cancer is surgery. However, radiotherapy or chemotherapy resistance and postoperative recurrence and metastasis are the main causes of laryngeal cancer-related death. Therefore, there is an urgent need to find effective treatment strategies.

Cancer stem cells (CSCs) are cells in malignant tumors that have unlimited proliferation, self-renewal and multi-differentiation capacities. They are considered to be the source of tumor cell proliferation and metastasis (^{Clevers, 2011}; ^{Yu et al., 2012}). It has been shown by our group and other scholars that laryngeal cancer has heterogeneity in differentiation, proliferation, tumorigenicity, and resistance to radiotherapy and chemotherapy, and there are CSCs in laryngeal cancer (^{Prince et al., 2007}; ^{Wei et al., 2009}). These CSCs may lead to recurrence and metastasis of laryngeal cancer (^{Li & Zhang, 2014}). Therefore, to find effective targets for CSCs in laryngeal cancer is the key for precision treatment of laryngeal cancer (^{Maccalli et al., 2014}; ^{Deshmukh et al., 2016}; ^{Ahmad & Amiji, 2017}). It is reported that in the tumor microenvironment, CSCs and angiogenesis are inseparable, and are related to tumor formation and development (Yu et al., 2007). As a glycosylated protein, human CD133 has 2 extracellular loops and 5 transmembrane domains (^{Miraglia et al., 1997}; ^{Grosse-Gehling et al., 2013}).

It is one CSC marker in Hep-2 cells (^{Zhou et al., 2007}). One in vivo study showed that Hep-2 cell line with positive CD133 expression had higher tumorigenic potential than those with negative CD133 expression (^{Wei et al., 2009}). In this study, we used CD133 to mark CSCs in laryngeal cancer.

Integrins are a family of transmembrane receptors that facilitate cell-extracellular matrix adhesion. They are highly expressed on the surface of neovascular endothelial cells and various tumor cells, and play an important role in tumor angiogenesis, migration and infiltration (^{Sun et al., 2014}). They have two subunits of a and b. Integrin αvβ3 plays an important role in tumor invasion and metastasis, and has the ability to recruit and activate matrix metalloprotein-2 and plasmin, which degrades matrix membrane and matrix interstitial components and promotes tumor metastasis (^{Huveneers et al., 2007}). Studies have shown that integrin αvβ3 is highly expressed in CD133⁺ laryngeal CSCs (^{Lu et al., 2011}; ^{Li et al., 2013}).

RGD peptide is a widely distributed tripeptide of arginine-glycine-aspartate (Arg-Gly-Asp). It is the recognition site for integrin αvβ3 and its ligand, which mediates the specific binding of matrix proteins to integrin αvβ3, thereby regulating the cell-cell interaction and cell- extracellular matrix interaction (^{Ahmedah et al., 2017}; ^{Katsamakas et al., 2017}) Studies have shown that RGD peptide can directly activate caspase-3 and cause tumor cell apoptosis. Exogenous RGD peptides can competitively bind to integrin αvβ3 ligand and reduce the expression of integrin αvβ3, thereby inhibiting tumor angiogenesis and reducing the adhesion and infiltration ability of tumor cells (^{Semenza, 2007}; ^{Cook et al., 2009}: ^{Shin et al., 2015}).

Although the role of RGD peptide in inhibiting tumor angiogenesis has been widely recognized, its effect on CSCs and tumor angiogenesis of laryngeal cancer is unclear. Here, in this study, we investigated the inhibitory effect of RGD peptide on laryngeal CSCs, and the expression of tumor angiogenesis-related molecules including VEGF, VEGFR2, STAT3 and HIF-1α. The possible mechanisms were analyzed and discussed. Our study will provide a new perspective for laryngeal cancer treatment.

MATERIAL AND METHOD

Animals. Nude mice (4-6-week-old; 28-32s body weight) were purchased from Beijing Weitong Lihua Experimental Animal Technology Co., Ltd. They were kept in standard conditions. This study was conducted in strict accordance with the recommendations of the National Institutes of Health's Guide for the Care and Use of Laboratory Animals. The research protocol with animal experimentation was approved by the Ethics Committee of Gansu Provincial Hospital (Protocol Number: Not applicable).

Cell culture. Human laryngeal carcinoma Hep-2 cells were from Qi Biotechnology Co., Ltd. (Shanghai, China). They were cultured in RPMI 1640 medium (Hyclone, Utah, Logan City, USA) containing 10 % fetal bovine serum (Clark, USA) and 1 % penicillin/streptomycin (Gibco, Grand Island, New York City, USA) in a 37 °C, 5 % CO cell incubator.

Fluorescence activated cell sorting. FITC-anti-human CD 133 (BIOSS, Beijing, China) was added to the single cell suspension and incubated for 30 min. After washing, samples were analyzed using a Beckman Coulter MoFlo XDP to sort CD133⁺ CSCs. The cell purity was measured with flow cytometry.

CCK-8 assay. CD133⁺Hep-2 laryngeal cancer stem cells were seeded in 96-well plates and cultured for 4 h. CCK-8 reagent (Dojindo, Japan) was then added and the incubation continued for 2.5 h. The absorbance value OD450 of each well was measured and the standard curve was fitted.

The sorted CD133⁺Hep-2 laryngeal CSCs were seeded in 96-well plates at 10³ cells/well. After culture for 24 h, 10 µL RGD peptide (Abcam, England) at different concentrations (14.4 µM, 144 µM, 288 µM, 576 µM, and 1440 µM) were added and incubated for 24h. Finally, 10 µL of CCK-8 from the CCK-8 kit was added to each well. After incubation for 2.5 h, the absorbance of each well at 450 nm was measured by a microplate reader.

The cell survival rate was calculated according to the standard curve and according to the following formula: cell survival rate = [(As-Ab)/(Ac-Ab)]x 100 %. As: experimental well (cells, CCK-8, different concentrations of RGD peptide); Ac: control well (cells, CCK-8, no RGD peptide); Ab: blank well (without cells, RGD peptides, and CCK-8).

RT-PCR. The sorted CD133⁺Hep-2 laryngeal CSCs were seeded at 5 x 10⁵ in T25 flasks and incubated with different concentrations of RGD peptide (10 µM, 20 µM, 40 µM, respectively). After 24 h, the cells were collected and lysed with RNAiso Plus (TaKaRa, Japan). Total RNA was extracted from cells using an RNeasy kit (QIAGEN, Germany). The cDNA was obtained with SuperScript III Reverse Transcriptase (Invitrogen, Carlsbad, USA). The primer sequences were shown in Table I. The PCR Master Mix (SYBR Green; QIAGEN) and the Real-Time PCR System (7500 model; ABI) was used. b-actin was used as internal control.

Western blot. The sorted CD133+Hep-2 laryngeal CSCs were seeded at 5 x 105 in T25 flasks and incubated with different concentrations of RGD peptide (10 µM, 20 µM, 40 µM, respectively). After 24 h, the cells were harvested and total proteins were extracted. After SDS-PAGE, proteins were transferred to PVDF membranes (Millipore, Bedford, MA). The primary antibodies were rabbit anti mouse anti- VEGF (Proteintech), anti-VEGFR2 (Proteintech), anti- STAT3 (Proteintech), anti-HIF-1α (Proteintech) and β-actin (ZSGB-BIO). The goat anti-rabbit secondary antibodies were from ZSGB-BIO. Enhanced chemiluminescence kit (Pierce Rockford, IL) was used for color development. The grayscale of the western blot protein band was analyzed by Image.

Tumor growth. The mice were randomly divided into four groups, with 10 mice in each group. Before drug administration, mice in two groups received right axillary subcutaneous injection of CD133⁺ CSCs (3x10⁵), and those in the other two groups received right axillary subcutaneous injection of CD133^- tumor cells (3x10⁵). When the tumor diameter reached 0.5-1 cm, the mice in CD133⁺ group and CD133^- group were further divided into PBS control group and 10 mg/g/d RGD peptide group, respectively. RGD peptide was administrated every other day. The administration lasted for 30 days. Mouse body weight and tumor length and width were recorded every other day. Tumor volume was calculated according to the following formula: tumor volume = tumor length x tumor width² x 0.52. At the end of the experiment, mice were sacrificed and the tumors were dissected.

Immunohistochemistry. The dissected tumor tissues were fixed with 10 % paraformaldehyde for 12 h at room temperature. Then, the tissue was dehydrated with ethanol, embedded in paraffin and cut into sections of 0.5 µm. The primary antibody of integrin αvβ3 (Santa Cruz, Mexico) was added and incubated overnight at 4 °C. After washing, the sections were incubated with the secondary antibody (SP-9002; ZSGB-BIO, China) for 15 min at room temperature. Finally, the sections were stained with chrome solution for 10 min, counter stained with hematoxylin for 20 min, and mounted with neutral resin. The sections were observed under optical microscope. Integral Optical Density (IOD) was calculated using Imagepro plus software (Media Cybernetics, Maryland, USA).

Statistical analysis. The statistical data was analyzed and processed by IMB SPSS16.0. All data are expressed as mean ± standard error of mean (mean ± SEM). Multiple comparisons were performed using one-way ANOVA followed by LSD (Least Significant Difference) (for variables with homogeneity of variance) or rank sum test (for variables without homogeneity of variance). A P<0.05 was considered as statistically significant.

Table I RT-PCR primers. 

RESULTS

Sorting of Hep-2 CD133+ CSCs. In this experiment, laryngeal cancer CD133+Hep-2 CSCs were sorted by flow cytometry. The results showed that the ratio of CD133+ CSCs to the total number of cells was 1.34±0.87 %, while CD133- non-tumor stem cells accounted for 95.0±5.76 %. The sorted cancer stem cells grew well.

RGD peptide toxicity by CCK-8 assay. The effect of different concentrations of RGD peptide (14.4 µM, 144 µM, 288 µM, 576 µM, 1440 µM) on the proliferation of CD133⁺ CSCs was examined using CCK-8 assay. The IC50 (half maximal inhibitory concentration) of RGD was determined as 29.11 µM. As shown in Figure 1A and 1B, RGD peptide inhibited the proliferation of Hep-2 CD133⁺ cells in a dose- dependent manner. These results indicate that the RGD peptide has a significant inhibitory effect on the proliferation of laryngeal CSCs.

Fig. 1 Effect of RGD peptide on cell proliferation. CD133⁺ CSCs of laryngeal cancer were treated with RGD at different concentrations as indicated. (A) The cell number was shown. (B) The cell survival rate was shown. Compared with 0 mM, *p<0.05. 

Inhibition of RGD peptide on VEGF pathway in laryngeal CSCs. To examine the effect of RGD peptide on VEGF pathways and downstream effectors in laryngeal CSCs, RT- PCR was performed to determine the mRNA levels of related genes. The results showed that the mRNA levels of VEGF, VEGFR2, STAT3 and HIF-1α were significantly decreased in a dose-dependent manner after co-culture of RGD peptide with laryngeal CSCs for 24 h (Fig. 2).

Fig. 2 Effect of RGD peptide on the mRNA expression of genes in VEGF pathway. The mRNA expression level of VEGF, VEGFR2, STAT3 and HIF-1α were respectively analyzed by RT-qPCR after treatment with RGD peptide for 24 h. *p < 0.05, ** p < 0.01. 

To further verify this, Western blot was performed. As shown in Figure 3, similar to mRNA result, the expression levels of VEGF, VEGFR2, STAT3 and HIF-1α protein decreased gradually as the increase of RGD peptide concentration. These results indicate that the RGD peptide inhibits the expression of key genes involved in VEGF pathway.

Fig. 3 Effect of RGD peptide on the protein expression of genes in VEGF pathway. The protein expression level of VEGF, VEGFR2, STAT3 and HIF-1a were analyzed by Western blot after treatment with RGD peptide for 24h. Representative and quantitative Western blot results were shown. *p < 0.05, ** p < 0.01. 

Inhibition of tumor growth by RGD peptide. To investigate the effect of RGD peptide on tumor formation of CD133⁺ Hep-2 cancer cells, we inoculated CD133⁺ tu- mor stem cells into nude mice. The CD133- Hep-2 cells were used as controls. The gross tumor volume in CD133^- +PBS group was smaller than CD133⁺ +PBS group while that in mice treated with RGD peptide was smaller than that in PBS control (Fig. 4A). Statistically, compared with CD133^- +PBS group, the tumor volume of CD133⁺ +PBS group was significantly larger (P<0.05) (Fig. 4B & 4D), suggesting that CD133⁺ Hep-2 cancer cells are more tumorigenic than CD133^- Hep-2 cells in vivo, and have the characteristics of stem cells. After RGD treatment, the tumor volume was significantly decreased (P<0.05), indicating that RGD peptide significantly inhibits tumor growth. Moreover, the body weight of nude mice treated with RGD peptide was significantly higher than that of the PBS control group, indicating that the mice are in good condition and that RGD peptide has no obvious toxicity to nude mouse at IC50 dose (Fig. 4C). The above results demonstrate that RGD can inhibit the growth of tumor stem cells in nude mice.

Fig. 4 Effect of RGD peptide on tumor growth in vivo. Nude mice were inoculated with tumor cells (CD133⁺ CSCs or CD133^- CSCs) and then treated with RGD peptide. Tumor volume and body weight were measured. (A) Photograph of removed tumors was shown. (B) The tumor volume in each group on day 10 was statistically analyzed. (C) The body weight of mice was comparatively analyzed. (D) The tumor volume changed with the time of administration. *p < 0.05, ** p < 0.01. 

Effect of RGD peptide on integrin αvβ3. To detect the effect of RGD peptide on integrin αvβ3 expression, immunohistochemistry was performed. The integrin avb3 was stained brown. It was mainly expressed on the cell membrane and some was in cell nucleus. It was observed that RGD peptide can significantly reduce the expression of integrin αvβ3 in CD133⁺ Hep-2 CSCs (IOD was respectively 11.33±0.65, 110.54±4.58, P<0.001) (Fig. 5). Meanwhile, the RGD peptide was also able to significantly reduce the level of CD133^-Hep-2 integrin αvβ3 (IOD was respectively 13.64±0.69, 53.56±3.31, P<0.001). These results showed that RGD inhibited tumor vascular endothelial integrin αvβ3 expression, suggesting that tumor neovascularization is inhibited, thus inhibiting tumor growth.

Fig. 5 Effect of RGD peptide on integrin αvβ3 expression. Immunohistochemistry was performed to detect integrin αvβ3 expression in tumor tissues. The integral optical density (IOD) of integrin αvβ3 was comparatively analyzed and shown. ***p < 0.001. 

DISCUSSION

Laryngeal cancer is one of the most common malignant tumors of the head and neck (^{Global Burden of Disease Cancer Collaboration et al., 2015}). Although the 5-year survival rate and quality of life of patients have been improved, the recurrence and metastasis of laryngeal cancer after treatment are still the most important cause of death. So far, there is still a lack of effective treatments to prevent and treat tumor recurrence and metastasis. In recent years, CSC theory has provided a new perspective for the research and treatment of tumors (^{Greco et al., 2016}; ^{Wang et al., 2017}). ^{Zhou et al. (2007}) showed that although CD133⁺ cells accounted for only 3.15 %±0.83 % of the total number of laryngeal cancer cells, they had the characteristics of stem cells.

Tumor growth depends on its angiogenesis. VEGF is an important signaling molecule involved in tumor angiogenesis (^{Ferrara & Henzel, 1989}; ^{Kim et al., 2002}). It binds to VEGF receptor (VEGFR), thereby activating multiple signaling pathways, promoting cancer cell proliferation, migration and invasion, and angiogenesis (^{Rafii et al., 2002}). VEGFR is a single transmembrane receptor protein, including VEGFR1 (Fit-1), VEGFR2 (KDR) and VEGFR3 (Fit-4), all of which belong to Protein Tyrosine Kinase Receptor (PTKR) and participate in angiogenesis. Among them, VEGFR2 is mainly involved in the process of angiogenesis (^{Zhang et al., 2010}). Signal transducers and activators of transcription 3 (STAT 3) is a transcription factor (^{Darnell Jr. et al., 1994}) and is highly expressed in a variety of malignancies (^{Germain & Frank, 2007}; ^{Al Zaid Siddiquee & Turkson, 2008}). Moreover, STAT3 plays an important pivotal role in mediating signaling between tumor cells and other cells (^{Yu et al., 2007}). ^{Zhou et al. (2016}) found that persistent activation of STAT3 was associated with the development of laryngeal cancer. HIF-1 is a transcription factor produced under hypoxia. When activated, HIF-1 can promote tumor angiogenesis. It is a heterodimer consisting of HIF-1α and HIF-1β. HIF-1β, also known as ARNT (aryl hydrocarbon receptor nuclear translocator), is stably expressed in cells, and its gene is located in human Chromosome q21 region. HIF-1a localizes in q21-24 region of human chromosome 14, and is regulated by an anoxic signal (^{Pugh & Ratcliffe, 2003}). Thus, VEGF/VEGFR2/STAT3, and HIF-1α signaling pathway, synergistically promotes tumor angiogenesis (^{Chen et al., 2008}; ^{Xu et al., 2012}; ^{Wang et al., 2017}).

Integrin αvβ3 plays an important role in tumor proliferation, adhesion and angiogenesis. It is highly expressed in endothelial cells of tumor neovessels and on tumor cell surface (^{Sun et al., 2014}). In the extracellular matrix, the RGD peptide competitively inhibits the binding of integrin αvβ3 to the receptor, thereby attenuating integrin-mediated tumor cell adhesion. In this study, we sorted CD133⁺ Hep-2 CSCs and co-cultured with RGD peptides. We found that RGD peptides inhibited tumor proliferation. Meanwhile, the mRNA and protein levels of VEGF/VEGFR2/STAT3 signaling pathway were decreased by RGD peptide, indicating that RGD peptide may inhibit CD133⁺ Hep-2 cancer cells by inhibiting VEGFR2-mediated STAT 3/HIF-1α pathway-related genes and protein expression. Furthermore, we observed the inhibitory effect of RGD peptide on tumor growth in nude mice, and detected the expression of tumor angiogenesis- related protein integrin αvβ3. The results showed that the expression level of αvβ3 in the RGD peptide treatment group was significantly lower than that in the PBS control group.

CONCLUSION

In conclusion, RGD peptide may inhibit tumor growth by inhibiting the proliferation of CD133⁺ Hep-2 cancer cells. The underlying mechanism is that RGD inhibits tumor angiogenesis-related signaling pathways, thus affecting the tumor angiogenesis, and decreasing the progression of human laryngeal cancer stem cells.