RAF1 promotes lymphatic metastasis of hypopharyngeal carcinoma via regulating LAGE1: an experimental research

Hypopharyngeal carcinoma tumor tissue, normal adjacent tissue and lymph node tissue were acquired from patients undergoing hypopharyngectomy in the First Affiliated Hospital of Chongqing Medical University from 2012 to 2021. The inclusion criteria were as follows: pathological diagnosis was hypopharyngeal squamous cell carcinoma; preoperative ultrasound, CT or MRI were performed to determine the extent of the lesion; hemorrhagic metastasis and distant metastasis could not detected; patients were diagnosed with hypopharyngeal carcinoma for the first time without complication of other malignant tumors; none preoperative radiotherapy, chemotherapy or targeted therapy was received. The collection process followed the Declaration of Helsinki and was approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University. Some specimens were immediately frozen in liquid nitrogen and stored at – 80 °C, others were immersed in formalin and then embedded in paraffin. The clinical characteristics of patients are shown in Table 1.

Total RNA was extracted from the 10 cases of hypopharyngeal carcinoma primary tumor tissue (5 patients have lymphatic metastasis and 5 patients do not have lymphatic metastasis) using the RNA extraction kit (Takara, Dalian, China), and then sent to Jingzhou Gene Technology Limited Company (Shanghai, China) for illumina dual-terminal transcriptome sequencing. A threshold of 1.5 for the log2 fold change(FC) was set, and the edge software package was used to analyze the differentially expressed genes(|log2FC|> 1.5, P < 0.05). Then the volcano map and heat map of differential genes were plotted. Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were identified with a strict cutoff (false discovery rate (FDR) < 0.05). After the differential gene RAF1 was screened by transcriptome sequencing, the RNA was reverse transcribed into cDNA using PrimeScript RT reagent Kit (Takara, Dalian, China). Then, qRT-PCR was run by using SYBR primescript RT-PCR Kit (Takara, Dalian, China) and specific primers for GAPDH and RAF1. The primer sequences were as follows: GAPDH (sense): CAGCGACACCCACTCCTC; GAPDH (antisense): TGAGGTCCACCACCCTGT; RAF1 (sense): GGGAGCTTGGAAGACGATCAG; RAF1(antisense): ACACGGATAGTGTTGCTTGTC. After the reactions were complete, the CT values were determined by setting a fixed threshold.The relative amounts of RAF1 mRNA were normalized to GAPDH using a 2^−ΔΔCt calculation method.

The total protein extraction kit (KeyGen BioTECH, Jiangsu, China) was used to isolate proteins from the above 10 patients. The protein lysates were separated by 10% sodium dodecyl sulphate—polyacrylamide gels (SDS-PAGE) (Beyotime, Shanghai, China), and then transferred onto polyvinylidene difluoride (PVDF) membranes (Beyotime, Shanghai, China). Primarry antibodies against RAF1, phospho-RAF1 (p-RAF1) and GAPDH were purchased from Abcam (ab-137435, ab-60985, ab-181602; Abcam, Cambridge, UK) and were diluted at 1:1000, 1:1000 and 1:3000. The secondary antibody goat anti-rabbit IgG (Beyotime, Shanghai, China) was diluted at 1:5000. The Efficient chemiluminescence (ECL) kit (Thermo, Shanghai, China) was utilized to detect horseradish peroxidase (HRP) and its products.

IHC was performed on 4 μm sections of paraffin embedded tissues previously prepared, including 113 cases of tumor tissues from hypopharyngeal carcinoma patients (82 patients have lymphatic metastasis and 31 patients do not have lymphatic metastasis). After antigen retrieval and peroxidase block, slides were incubated with primary antibody (ab-60985, Abcam, Cambridge, UK) which was diluted at 1:200. Then, slides were treated with boost IHC detection reagent (ZSGB-BIO, Guangzhou, China). Finally, slides were stained with diaminobenzidine (DAB) and counterstained with hematoxylin. The immunohistochemistry results were analyzed separately by two experienced pathologists and scored considering both the intensity of staining and the proportion of tumor cells with an unequivocal positive reaction. Slides which incubated with Phosphate buffer saline (PBS) instead of primary antibody were selected as negative controls. According to the staining results, the expression was scored as either positive expression (stained brown, tumor cells ≥ 50%) or negative expression (stained yellow, tumor cells < 50% of cells).

According to the IHC results, Kaplan–Meier and log-rank test and COX’s univariate and multivariate analysis were used to examine the association among the clinicopathological features, and to evaluate independent risk factors of prognosis in hypopharyngeal carcinoma.

The FADU cell line (human pharyngeal squamous cell line) was purchased from Center for Molecular and Cellular Sciences, Chinese Academy of Sciences (Shanghai, China). The SCC15 cell line (human tongue squamous cell line) was donated by Department of Oral and Maxillofacial Surgery, The First Affiliated Hospital of Chongqing Medical University. Cells were grown in dulbecco’s modified eagle medium (DMEM) high glucose medium (Gibco, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, Carlsbad, CA, USA) in a humidified incubator at 37 °C with 5% CO2.

RAF1 was knocked down and overexpressed by transfection of FADU and SCC15 cell lines with lentivirus short hairpin RNA (shRNA) and overexpression vector. Lentiviruses were purchased from GeneChem (Shanghai, China), and the lentiviral composition was Ubi-firefly_Luciferase-IRES-Puromycin. The transfected cells were divided into four groups including sh-NC (knockdown negative control), sh-RAF1 (knockdown), vector (overexpression negative control) and lv-RAF1-OE (overexpression). FADU and SCC15 cells were seeded in 6-well plates and transfected using lentiviruses on the following day when the cells were approximately 50–60% confluent. The multiplicity of infection (MOI) of the cells was 10. After 6 h transfection, the cell medium was cultured in DMEM supplemented with 10% FBS for 48 h. Then, the cells were cultured in medium containing 2 μg/mL puromycin to obtain stably infected cells. Next, the cells were harvested for qRT-PCR and WB (methods were the same as above) to detect the expression of RAF1. Primarry antibodies against lymphangiogenic cytokines VEGF-C (ab83905, Abcam, Cambridge, UK, diluted at 1:1000), LYVE-1 (ab219556, Abcam, Cambridge, UK, diluted at 1:1000) and PROX-1 (ab199359, Abcam, Cambridge, UK, diluted at 1:1000) were added to WB.

For EdU assay, FADU and SCC15 cells were seeded in 6-well plates after transfection. When the confluency of cells reached 80%, EdU assay kit (RiBoBio, Guangzhou, China) was used to determine the proliferation rate of the cells. The manufacturer’s instruction was followed except that the nucleus staining dye was changed from Hoechst 33,342 to DAPI (Beyotime, Shanghai, China). After staining, the cells were captured by inverted fluorescence microscope.

For cell apoptosis assay, FADU and SCC15 cells were cultureed in 6-well plates until the confluency of cells reached 80–90%. Then, cells were labeled in a binding buffer containing annexin V-FITC/PI. The samples were analyzed by flow cytometry (Biosciences, CA, USA). For cell cycle assay, cells were trypsinized and fixed with 70% ice ethanol overnight at − 4 °C. Next day, the cells were treated with RNase A and propidium iodide and the cell cycle was measured by flow cytometry.

Cell migration assays were performed using 24-well plates with an 8-μm pore membrane (Corning, CA, USA). FADU cells were suspended in FBS-free DMEM medium and then added to the upper chamber (5 × 10⁴ cells/well). Meanwhile, DMEM medium plus 15% FBS was added to the lower compartment. The plates were incubated for 24 h in the incubator. After incubation, cells that had migrated to the lower surface of the filter membrane were fixed with 4% paraformaldehyde for 30 min at room temperature. The membrane was washed with PBS and stained with 0.5% crystal violet in methanol for 15 min at room temperature. Cells remaining on the upper surface of the filter membrane were gently scraped off with a cotton swab. The lower surfaces were captured by inverted fluorescence microscope, and the cells were counted blindly (five fields per chamber). In order to further evaluate cell invasion, the assay was repeated in a Transwell assay with Matrigel (Biosciences, MA, USA).

FADU cells were seeded in 6-well plates (5 × 10⁵ cells/well) and cultured with DMEM medium supplemented with 10% FBS until the confluency of cells reached nearly 100%. The pipette tip was used to make 3–5 parallel scratches on the bottom of the Petri dish. Then the cells were washed with PBS and cultured with FBS-free medium in the incubator. The scratches were captured by inverted fluorescence microscope at 0 h, 24 h and 48 h. Meanwhile, the closure rate of wound was calculated.

Five-week-old male BALB/c-Nude mice were purchased from the HFK Bioscience Limited Company (Beijing, China) and maintained under specific pathogen-free conditions at Animal Center of Chongqing Medical University. Stably infected cells (including FADU-sh-NC, FADU-sh-RAF1, FADU-vector and FADU-lv-OE) were subcutaneously injected into foot-pad (1 × 10⁶ cells in 0.2 ml PBS per mouse) and the mice were divided into four groups (10 mice in each group). After 28 days, D-luciferin (Beyotime, Shanghai, China) was diluted with PBS to 15 mg/ ml and then injected intraperitoneally at a dose of 10 uL/g. Fluorescence imaging of nude mice was then performed. Next day, mice were sacrificed after injection of pentobarbital sodium. Primary tumor of foot-pad and metastatic lymph node were collected, and the volumes and weights of the tumors were measured. The volume of tumors were calculated by the formula: Tumor volume = [length * (width)²] * π/6 [45]. A portion of the tissues was used for protein and total RNA extraction, and the remaining tissue was fixed in 4% paraformaldehyde. Then, the expression of RAF1 was detected by qRT-PCR, WB and IHC refer to the above methods. All experiments on mice were approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University and performed in accordance with the U.K. Animals (Scientific Procedures) Act and the guidelines of the National Institutes of Health.

Lentivirus transfection was used to knock down LAGE1. Lentiviruses (including sh-LAGE1 and sh-NC) were purchased from GeneChem (Shanghai, China), and the lentiviral vector was Ubi-firefly_Luciferase-IRES-Puromycin. Puromycin was used to successfully obtain stably infected cells. RAF1 inhibitor Sorafenib (MCE, CA, USA) was diluted at 25 μmol/L with PBS and added to medium. Then, the cells were harvested for qRT-PCR and WB (methods were the same as above) to detect the expression of LAGE1. The primary antibody against LAGE1 was purchased from Abcam (ab177947, diluted at 1:1000). Next, EdU proliferation assay, Flow cytometry, Wound Healing assay and Transwell migration and invasion assay were used to verify the effect of LAGE1 expression changes on cell function.

Five-week-old male BALB/c-Nude mice were purchased and maintained under specific pathogen-free conditions. Stably infected cells were subcutaneously injected into foot-pad and Sorafenib was injected intraperitoneally at a dilution of 30 mg/kg with PBS. The mice were divided into four groups: sh-NC, sh-NC + Sorafenib, sh-LAGE1 and sh-LAGE1 + Sorafenib (10 mice in each group). Fluorescence imaging of nude mice was then performed after 28 days. Next day, mice were sacrificed after injection of pentobarbital sodium. Primary tumor of foot-pad and metastatic lymph node were collected, and the volumes and weights of the tumors were measured. Then, qRT-PCR, WB and IHC were performed to detect the expression of LAGE1 (methods were the same as above).

All of the images of the WB assay, IHC assay, EdU assay, Flow cytometry, Transwell assay and animal experiments were representative of at least three independent experiments or staining results. The qRT-PCR assay was performed in triplicate, and each individual experiment was repeated several times. SPSS 22.0 and GraphPad Prism 5.0 were employed for statistical analysis. The results are presented as the means ± standard deviation (SD). Observed differences were considered statistically significant at P < 0.05 by using Student’s t-test or Chi-square test.

Primary tumors from 5 patients with lymphatic metastasis and from another 5 patients without lymphatic metastasis were sent for transcriptome sequencing. A total of 2341 differential genes were screened based on the volcano map with a threshold of |log2FC|> 1.5, P < 0.05 (Fig. 1A). According to the heat map, we listed the top 20 genes with the most statistically significant differences in expression between the lymphatic and non-lymphatic groups, and RAF1 was one of them as indicated by the arrow (Fig. 1B and C). To further confirm the expression of RAF1 in clinical tumor tissue specimens, we assessed RAF1 mRNA and protein levels by qRT-PCR and WB assay. Results showed that in 10 pairs of hypopharyngeal carcinoma primary tumor tissues, the mRNA levels of RAF1 were significantly up-regulated in lymphatic metastasis group compared to non-lymphatic metastasis group (Fig. 2B). Meanwhile, the protein levels of phospho-RAF1 (p-RAF1) decreased sequentially in the lymphatic metastasis group, the non-lymphatic metastasis group and the normal tissue group (Fig. 2A and B). We increased the number of samples to 113 cases and confirmed the expression of RAF1 by IHC assay. It suggested that the positive expression rate of p-RAF1 was significantly higher in the group with lymphatic metastasis (66/82, 80.5%) than in the group without lymphatic metastasis (9/31, 29.0%) (Fig. 2C and Table 2). In addition, the expression of p-RAF1 was positive in most of the metastatic lymph nodes (71/95, 74.7%). Kaplan–Meier & log-rank test and COX’s analysis indicated that RAF1 positive expression and lymphatic metastasis are both independent prognostic risk factors for hypopharyngeal carcinoma (Fig. 2D, E and Table 3). These results suggest that RAF1 expression level may be directly related to lymphatic metastasis.

RAF1 was knocked down and overexpressed by transfection of FADU and SCC15 cell lines with lentivirus shRNA and overexpression vector. According to qRT-PCR and WB assay (Fig. 3A–D), the mRNA and protein levels of RAF1/p-RAF1 were significantly decreased in knockdown group (sh-RAF1) and increased in overexpression group (lv-RAF1-OE) compared to their negative control groups (sh-NC and vector). In addition, the protein levels of lymphangiogenic cytokines (including VEGF-C, LYVE-1 and PROX-1) were consistent with the expression trend of p-RAF1 (Fig. 3A, C and D). It suggested that the expression of RAF1 might be related to lymphangiogenesis. The results of EdU assay indicated that down-regulation of RAF1 expression can decrease the proliferation ability of cells (Fig. 3E, F and G). Due to flow cytometry assay, the apoptosis rate was significantly reduced in RAF1 overexpression group and increased in knockdown group compared to the negative control groups (Fig. 4A and C). In RAF1 knockdown group, the proportion of G1 phase cells increased, while G2 phase cells decreased. Meanwhile, the overexpression group showed the opposite trend (Fig. 4B, D and E). Due to the weak invasion and migration ability of SCC15 cell line, the experimental results were not ideal, only FADU cell line was used in the Transwell assay and Wound Healing assay. The results showed that RAF1 knockdown significantly reduced the cell migration and invasion ability, while RAF1 overexpression greatly enhanced the cell migration and invasion ability (Fig. 5A–E). In conclusion, reduced expression of RAF1 can significantly inhibit the proliferation, migration and invasion of tumor cells, and promote the apoptosis of tumor cells.

We constructed a nude mouse xenograft model by injecting stably infected FADU cells into the foot-pad. Nude mice were divided into 4 groups according to the type of injected cells: FADU-sh-NC, FADU-sh-RAF1, FADU-vector and FADU-lv-RAF1-OE (10 mice in each group). Fluorescence imaging was performed after 28 days of feeding in specific pathogen-free conditions (Fig. 6A). Results analysis showed that luminescence area and average photon flux were reduced in sh-NC group and increased in lv-RAF1-OE group compared to their control group (Fig. 6B and C). After the mice were anesthetized and sacrificed, primary tumors of foot-pad and metastatic lymph nodes were harvested (Fig. 6D). Meanwhile, we measured the volumes and weights of the tumors (Fig. 6E) and counted the number of metastatic lymph nodes in each group (Table 4). The results indicated that in the sh-RAF1 group, the growth of primary tumor and lymph nodes was slower than that in the sh-NC group, and the incidence of lymphatic metastasis was lower. On the contrary, in the lv-RAF1-OE group, tumor growth was faster than in the control group, and the incidence of lymphatic metastasis was higher. RAF1 expression in primary tumors and lymph nodes was subsequently validated. According to qRT-PCR and WB assay (Fig. 6F, G and H), the mRNA and protein levels of RAF1/p-RAF1 were significantly decreased in sh-RAF1 group and increased in lv-RAF1-OE group compared to their negative control groups. The protein levels of lymphangiogenic cytokines (including VEGF-C, LYVE-1 and PROX-1) were consistent with the expression trend of p-RAF1. Furthermore, the result of IHC suggested that the positive expression rate of p-RAF1 was higher in lv-RAF1-OE group (80.0% in primary tumors, 71.4% in metastatic lymph nodes) than in vector group (50.0% in primary tumors, 25.0% in metastatic lymph nodes), and was lower in sh-RAF1 group (20.0% in primary tumors, 0% in metastatic lymph nodes) than in sh-NC group (60.0% in primary tumors, 40.0% in metastatic lymph nodes) (Fig. 6I and Table 4). In addition, the expression of lymphangiogenic cytokine LYVE-1 was verified in metastatic lymph nodes. Result showed that the positive rate decreased sequentially in lv-RAF1-OE group, vector group, sh-NC group and sh-RAF1 group (Fig. 6I and Table 5). It suggested that lymphangiogenesis might decrease with RAF1 down-regulation. In general, in vivo results indicated that increased RAF1 expression promoted both tumor growth and lymphatic metastasis.

Primary tumors from 4 mice in sh-NC group and from another 4 mice in sh-RAF1 group were sent for proteomics sequencing. A total of 159 differential proteins were screened based on the volcano map with a threshold of |log2FC|> 0.5, P < 0.05 (Fig. 7A). According to the heat map, we listed the top 20 proteins with the most statistically significant differences in expression between sh-NC group and sh-RAF1 group groups, and RAF1 was one of them as indicated by the arrow (Fig. 7B and C). The expression level of RAF1-correlated proteins was analyzed by NetworkAnalyst. Results showed that LAGE1 had the most significant expression difference between groups (Fig. 7D). Then LAGE1 was knocked down by transfection of FADU and SCC15 cell lines with lentivirus shRNA. Meanwhile, RAF1 inhibitor Sorafenib was added to divide the transfected cells into 4 groups: sh-NC, sh-LAGE1, sh-NC + Sorafenib and sh-LAGE1 + Sorafenib. The results of qRT-PCR and WB showed that the mRNA and protein levels were significantly decreased after LAGE1 knockdown, while the mRNA and protein levels were further decreased after Sorafenib addition (Fig. 8A–D). The results of EdU assay suggested that down-regulation of LAGE1 could significantly inhibit cell proliferation, and the addition of sorafenib could aggravate this trend (Fig. 8E, F and G). According to Flow cytometry, we found that decreased LAGE1 led to increased apoptosis, while the proportion of G1 phase cells increased and that of G2 phase cells decreased. Sorafenib made the difference in apoptosis and cell cycle even more significant (Fig. 9A–E). In Transwell assay and Wound Healing assay, down-regulation of LAGE1 significantly inhibited cell migration and invasion, while Sorafenib further inhibited these functions (Fig. 10A–E). Above in vitro experiments proved that LAGE1 down-regulation could inhibit the proliferation, migration and invasion of tumor cells and promote apoptosis, while RAF1 inhibitor Sorafenib could further enhance this trend. It suggested that the effects of LAGE1 on cell function might be regulated by RAF1. In vivo experiment, we injected stably infected FADU cells into foot-pad subcutaneously and injected sorafenib intraperitoneally to divided mice into same groups as in vitro experiment. Fluorescence imaging results showed that luminescence area and average photon flux were reduced when LAGE1 was down-regulated, and fluorescence indexes decreased more obviously after the addition of Sorafenib (Fig. 11A–C). Measurements of primary tumors and lymph nodes harvested from the mice suggested that decreased LAGE1 inhibited tumor growth, and sorafenib helped to aggravate this trend (Fig. 11D and E). According to qRT-PCR and WB assay, mRNA and protein levels were significantly decreased after LAGE1 knockdown, while the mRNA and protein levels were further decreased after Sorafenib addition. Meanwhile, lymphangiogenic cytokines (including VEGF-C, LYVE-1 and PROX-1) also showed the same protein expression trend (Fig.11F–H)). Moreover, the result of IHC indicated that the positive expression rate of LAGE1 was higher in sh-NC group than in sh-LAGE1 group. After sorafenib injection, the positive rate was further reduced (Fig. 11I and Table 6). In addition, the expression of LYVE-1 was verified in lymph nodes. Result showed that the positive rate decreased sequentially in sh-NC group, sh-NC + Sorafenib group, sh-LAGE1 group and sh-LAGE1 + Sorafenib group (Fig. 11I and Table 7). In conclusion, down-regulation of LAGE1 could also inhibit tumor growth and lymphatic metastasis, and this inhibition was more significant under RAF1 regulation. It suggested that RAF1 could modulate lymphatic metastasis by targeting LAGE1.