LGR5 promotes invasion and migration by regulating YAP activity in hypopharyngeal squamous cell carcinoma cells


 Background High leucine-rich repeat-containing G-protein coupled receptor 5 (LGR5) expression caused by an inflammatory microenvironment was reported to promote tumor proliferation and the epithelial–mesenchymal transition (EMT) in various malignant tumors, but those effects have not been studied in hypopharyngeal squamous cell carcinoma (HSCC) and the molecular mechanism remains unclear. Additionally, YAP/TAZ, an upstream or downstream factor of multiple signaling pathways, can promote tumor proliferation, invasion, and angiogenesis. Our study was aimed to determine whether YAP/TAZ is involved in the regulation of LGR5 expression in the inflammatory microenvironment.Methods We stimulated FaDu cells, a hypopharyngeal squamous cell carcinoma cell line, with inflammatory medium. The expression levels of EMT-related proteins, LGR5, and p-YAP were detected by reverse transcription–polymerase chain reaction, western blotting, and immunofluorescence.Results The results showed that LGR5 expression and the EMT process were significantly enhanced. The expression of EMT-related proteins was up-regulated, while that of p-YAP was decreased. After inhibiting the high LGR5 expression with short interfering RNA, the expression of EMT-related proteins was also down-regulated, while that of p-YAP was significantly increased. The use of verteporfin (VP), an inhibitor of YAP activity that promotes YAP phosphorylation, did not affect LGR5 expression.Conclusions Our findings suggest that the inflammatory microenvironment leads to high LGR5 expression, up-regulating the expression of EMT-related proteins by inhibiting the YAP phosphorylation.


Background
The metastasis of hypopharyngeal squamous cell carcinoma (HSCC) is an important cause of its poor prognosis. In the past 20 years, the 5-year survival rate of patients with HSCC has never exceeded 55%.
Because of the abundant blood supply and lymphatic drainage pathway around the head and neck, malignant tumors in this area are often prone to early metastasis.
In recent years, studies on malignant tumor in ltration, distant metastasis, and the epithelialmesenchymal transition (EMT) have attracted much attention. Epithelial cells lose their original polarity, express the characteristics of mesenchymal cells, and acquire the ability of migration and anti-apoptosis. EMT is considered an important early event of tumor metastasis and is the main mode of metastasis and invasion of most malignant tumors [1]. Inhibition of the invasion and metastasis of tumor cells can be a new method for the treatment of squamous cell carcinoma. The underlying mechanism of EMT in head and neck tumors remains unclear and is of great research value.
The tumor development process is not only limited to local behavior but interactions with peripheral blood vessels, the extracellular matrix, surrounding normal cells, and related signaling molecules to form a complex regulatory network, namely the tumor microenvironment. Various signals in the tumor microenvironment play important roles in the survival, self-renewal, invasion, and metastasis of tumor stem cells [2]. The major components of the tumor microenvironment include the hypoxia microenvironment and in ammatory microenvironment.
The in ammatory microenvironment and immunosuppressive microenvironment are two important components of the tumor microenvironment. The in ltration of immune cells and aggregation of in ammatory factors can induce the activation of oncogenes, promote the formation of tumor blood vessels, and promote the invasion, development and even metastasis of tumors [3,4]. Leucine-rich repeatcontaining G-protein coupled receptor 5 (LGR5), also known as GPR49, is a member of the G proteincoupled receptor protein family. LGR5 was initially found in the small intestine and hair follicles, is a marker of adult intestinal epithelial stem cells, and plays a crucial role in embryonic development [5]. In recent years, many studies have shown that LGR5 is highly expressed in multiple malignant tumor types, including colorectal cancer, ovarian cancer, hepatocellular carcinoma, basal cell carcinoma, and esophageal adenocarcinoma. Overexpression of LGR5 is signi cantly correlated with high clinical stage and metastatic status of breast cancer, indicating that LGR5 may be a promising prognostic marker for patients with breast cancer [6].LGR5 is expressed differently in different tissues but has not been studied in hypopharyngeal cancer.
The mechanism of action and regulation of different in ammatory factors secreted by in ammatory cells concerning the EMT phenotype of head and neck squamous epithelial cells under continuous in ammatory stimulation, as well as the correlation between the phenotypic transformation of LGR5 + stem cells and acquisition of EMT ability, warrant further study. In this study, the in ammatory cell supernatant produced by THP-1 macrophages under lipopolysaccharide (LPS) stimulation was applied to head and neck tumor cell lines, to establish the in ammatory microenvironment model of tumor cells, to isolate the LGR5 + cell subsets induced by in ammatory factors, and to determine the correlation between LGR5 and EMT by comparing and analyzing the phenotypic transformation ability of EMT with parental cells. By constructing a recombinant LGR5 plasmid to transfect tumor cells to overexpress LGR5 protein, the regulatory pathways and molecular mechanisms of EMT and phenotypic transformation of stem cells induced by in ammatory factors were veri ed. YAP/TAZ is an important pair of transcription regulators that are highly active in most human malignancies. Recent studies have shown that YAP/TAZ plays a very important role in tumor development [7]. As an upstream or downstream factor of multiple signaling pathways, YAP/TAZ ultimately promotes tumor proliferation, invasion, and angiogenesis and metastasis. Normally, in mammals, the Hippo pathway kinase cascade leads to the phosphorylation of YAP/TAZ, trapping them in the cytoplasm, blocking their interaction with the TEA domain family member 1 (TEAD) family of transcription factors, and inhibiting cell proliferation and malignant transformation [8]. However, nonphosphorylated YAP/TAZ tends to be increased signi cantly in malignant tumor cells; they can enter the nucleus smoothly and regulate the expression of several genes related to proliferation, anti-apoptosis, and stem cell characteristics [9], including the secretory proteins connective tissue growth factor (CTGF) and CYR61 [10,11], AXL receptor tyrosine kinase [12], c-myc and survivin [12]. As a type of malignant tumor that seriously endangers human health, the role of YAP/TAZ in head and neck tumors has been rarely studied. Because both LGR5 and YAP/TAZ are closely related to tumor development, the ability to invade and metastasize, as well as the properties of stem cells, we considered that a relationship may exist between the two. Therefore, vertepor n (VP), a benzoporphyrin derivative belonging to the porphyrin family, is a YAP inhibitor used in this study to disrupt the YAP-TEAD interaction [13].
It is important to clarify the complex relationship between LGR5/YAP signaling and the development of hypopharyngeal squamous cell carcinoma. Therefore, we investigated the expression change of LGR5 in FaDu cells under the in ammatory microenvironment. The invasion ability of FaDu cells was evaluated with LGR5 over-expression or knock-down treatment. This study tried to clarify the mechanism of LGR5/YAP signaling regulation in HSCC proliferation and invasion. The data could provide a special target for the clinical treatment of head and neck tumors. were obtained from Shanghai Zhongqiaoxinzhou Biotech Co., Ltd. (Shanghai, China). All the cells were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS; Hyclone Laboratories, Inc., Logan, UT, USA), 1% penicillin, and 1% streptomycin and were maintained at 37°C in a humidi ed incubator with 5% CO 2 .

Preparation of FaDu cells with LGR5 over-expression
FaDu cells were transfected with the LGR5 plasmid or control plasmid, which was purchased from GeneCopoeia (EX-Q0041-M03, EX-NEG-M03; Guangzhou, China). The LGR5 recombinant plasmid was constructed using pReceiver-M03 as the backbone plasmid and was inserted into the LGR5 cDNA coding sequence (NM_003667, ORF length: 2,724 bp). FaDu cells transfected with the LGR5 plasmid were cultured in medium containing G418 for 14 d. The LGR5+ FaDu cells were collected to identify LGR5 overexpression using quantitative polymerase chain reaction (qPCR) and western blot assay.

Short interfering RNA (siRNA) transfection
FaDu cells were stimulated with in ammatory factors and then were transfected with siRNA oligo targeting the LGR5 gene, with a scrambled sequence as the control. Eight hours after transfection, the cell culture medium was replaced with RPMI 1640 containing 10% FBS and the cells were continued to incubate at 37°C for another 40 hours before the next step. An siRNA interference oligo sequence targeting the human LGR5 gene sequence (LGR5-homo-883) was designed and synthesized by Shanghai GenePharma Co., Ltd. The target sequence was UAAUAAGAG AAG GGUUGCCTT.

Cell proliferation assay
The proliferation ability of cells was measured by the CCK-8 assay using Cell Counting Kit (Yeasen, Shanghai). Different groups of cells were inoculated into 96-well plates at the density of 1,000 cells per well. After 24 hours, the CCK-8 reagent was added, followed by incubation at 37°C for 1 hour. The absorbance at 450 nm was recorded using a microplate reader.
Immuno uorescence staining Cells to be observed were inoculated on glass slides and rinsed three times with phosphate-buffered saline (PBS) 24 hours later, and then the cells were xed in 4% paraformaldehyde at room temperature for 30 minutes. The cells were washed three times with PBS and then were incubated with 0.1% Triton X-100, blocked in 5% goat serum for 1 hour, and then were incubated overnight with 1:100 diluted primary antibody at 4°C. Next, the cells were washed with PBS and then were incubated with 1:400 diluted secondary antibody in the dark for 1 hour at room temperature, followed by washing three times with PBS. The slides were sealed with glycerine at a concentration of 50% and then were observed and photographed with a laser confocal microscope (Fluo-View FV1000; Olympus, Japan).

Wound healing assay
FaDu cells of different groups were inoculated in 24-well plates and scratched with a 200-µl pipette tip when the plate was completely covered. PBS solution was used to wash away oating cell debris. Next, photos of the freshly scratched surfaces were taken. Thereafter, the cells were cultured at 37°C and 5% CO 2 for 24 hours and then were photographed again to compare the healing of each "wound". The scratch spacing at each time point was measured by Image J software [R M = (W 2 -W 1 )/W 2 ×100% (RM=relative mobility, W 1 =initial cell covering rate, W 2 = nal cell covering rate)]. The relative mobility of the cells at each time point was calculated with 0-hour scratch spacing as a reference, and the experimental results were analyzed using SPSS17.0 statistical software. The counting data were expressed as a percentage, while the measurement data were expressed as ±S. One-way analysis of variance was used for comparison of multi-group mean values.

Cell migration assay
Cell invasion ability was determined by the Transwell assay. Cell suspensions with concentrations of 5×10 5 /ml were prepared with 0.1% FBS in RPMI 1640 medium. Next, 5×10 4 cells were placed in the upper chamber (pore size: 8 µm; Millipore, Billerica, MA, USA) precoated with 1:4 diluted Matrigel (Yeasen, Shanghai, China). The lower chamber was lled with RPMI 1640 medium containing 20% FBS as an inducer. After incubation for 24 hours, the cells in the upper chamber and Matrigel were erased with cotton swabs, and the lower cells were stained with 0.5% crystal violet. Cell invasion was observed under an optical microscope (magni cation, ×40).

Real-time quantitative PCR (qPCR) detection system
Total RNA was extracted from each group using Trizol reagent, and the RNA integrity was veri ed by agarose gel electrophoresis. The RNA was reversed transcribed into cDNA using the Transcriptor First Strand cDNA Synthesis Kit (Cat.04897030001). Additionally, cDNA was ampli ed using FastStart Essential DNA Green Master, Cat.06924204001, (Roche, Indianapolis, IN, USA). The speci c primer designs are shown in Table 1. The real-time PCR protocol was performed according to our previous studies [14]. The target gene mRNA level was normalized to that in β-actin, and the relative mRNA levels were calculated using the formula: Fold change = 2 −△△CT .

Western blot analysis
The proteins were extracted with cell lysate buffer (T-PER Tissue Protein Extraction Reagent containing phosphatase inhibitor (Halt Protease and Single-Use Inhibitor Cocktail; 78442; Thermo). The proteins were separated using 10% SDS-polyacrylamide gel electrophoresis and then were transferred to PVDF membranes. The membrane was blocked with 5% non-fat milk powder for 1 hour and then was incubated with antibodies overnight at 4°C, washed three times with TBST, and then incubated with horseradish peroxidase-labeled secondary antibody at room temperature for 1.5 hours. The primary antibodies were as follows: LGR5 Polyclonal antibody (Cat# A10545), TWIST1 Polyclonal antibody (Cat# A7314), YAP Polyclonal antibody (Cat# A1001), pYAP Polyclonal antibody (Cat# AP0489), and SNAIL Polyclonal antibody (Cat# A5243), all from ABclonal Biotechnology Co., Ltd, China. β-Actin monoclonal antibody (Cat# ab8226; Cambridge, MA, USA) was used as a working internal control. ECL reagent (Thermo Fisher Scienti c, Waltham, MA, USA) was used to visualize protein bands, and images were obtained using a Dolphin-C hemi Mini image system (Wealtec Corp., Sparks, NV, USA). Gray values for each protein band were analyzed using Image J analysis software, and relative protein levels were calculated using β-actin as an internal control.

Statistical analysis
All the data were collected from at least three independent experiments. Results were analyzed using SPSS17.0 statistical software and are expressed as means ± standard deviation (SD). Student's t-test, one-way analysis of variance, and multivariate analysis of variance were used to analyze data as appropriate. A value of P < 0.05 was considered statistically signi cant. Chi-squared test was used to evaluate the correlation between LGR5 and clinicopathological features.
The in ammatory microenvironment affects the proliferation and migration ability of FaDu cells.
We obtained culture medium rich in cytokines by adding LPS (8 μg/ml) to the culture medium of THP-1 cells (Figure 2A). We chose to harvest the supernatant at 24 hours after stimulation because the amount of tumor necrosis factor (TNF)-α, the most abundant cytokine, reached its peak at this time. We tried to dilute the supernatant in various proportions, but the CCK-8 assay showed that the diluted supernatant had little effect on tumor cell viability. The medium was then used to simulate an in ammatory microenvironment. The cell viability of FaDu cells treated with the in ammatory microenvironment was compared with that of the untreated control group by the CCK-8 assay. The optical density (OD) of each group was measured 24 hours after administration at a 450-nm wavelength. The OD value was signi cantly increased in the treatment group. The proliferation of FaDu cells in the experimental group was 24.36% higher than that in the control group. Hence, the results showed that the in ammatory microenvironment can promote the proliferation of FaDu cells ( Figure 2B). We also conducted wound healing experiments, which showed that the proliferation and migration of tumor cells in the in ammatory microenvironment were signi cantly stronger than those of the control group. At 24 hours, the relative mobility of cells in the blank control group was 50.00±4.243%, while that in the experimental group was 72.25±5.439%. A signi cant difference was found between the groups (P<0.01) ( Figure 2C,D). Finally, the Transwell invasion assay was performed. This experiment indicated that the ability of HSCC to invade under the stimulation of the in ammatory microenvironment was signi cantly enhanced compared with that in the control group. The number of transmembrane cells in the experimental group was (55.00±7.071)/HP while that in the control group was (24.20±7.294)/HP ( Figure 2E,F).
The in ammatory microenvironment leads to high LGR5 expression in FaDu cells.
The experimental results of immuno uorescence showed that, compared with the control group, LGR5 expression in the treatment group was signi cantly higher. Additionally, the morphology of tumor cells in the in ammatory microenvironment showed clear transformation. The cell morphology changed from oval to spindle shaped, giving them a stronger ability to invade ( Figure 3A). qPCR was used to detect LGR5 expression. The experimental results showed great changes after stimulation of the in ammatory microenvironment. The LGR5 expression level increased signi cantly compared with that of the original value ( Figure 3B). Using actin as the internal control, the LGR5 mRNA expression level in the experimental group was increased by 2.27 times for 12 hours and 2.88 times for 24 hours compared with the initial value (P<0.01; Figure 3B). Additionally, we conducted western blotting experiments to detect LGR5 mRNA expression. The results showed that its expression level was upregulated in the in ammatory microenvironment ( Figure 3C,D). Compared with the control group, the expression level of LGR5 protein in the experimental group was increased by 2.41 times for 12 hours and by 2.32 times for 24 hours (P<0.01).

The expression levels of EMT-related genes in FaDu cells with the over-expression or gene-knockdown of
LGR5.
To investigate the biological signi cance of LGR5 in HSCC, the cells were transfected with LGR5 overexpression plasmids. The relative expression levels of EMT-related proteins were measured by western blotting. The levels of vimentin, twist, and snail in the LGR5 overexpression group were markedly upregulated compared with those in the control group. The expression levels of vimentin, twist, and snail were increased by 2.65, 1.65, and 2.53 times, respectively ( Figure 4A,B). The expression levels of vimentin, twist, and snail in FaDu cells transfected with LGR5 siRNA or siRNA NC were also detected by western blotting. The levels of vimentin, twist, and snail were decreased by 1.96, 2.56, and 3.41 times, respectively, in the LGR5 siRNA group compared with those in the siRNA NC group (Figure 4C,D).

Over-expression or gene-knockdown of LGR5 in uenced the ability of migration and invasion of FaDu cells in vitro
The invasion and migration of HSCC cells were signi cantly enhanced after they were transfected with the LGR5 overexpression plasmid. The relative mobility of the cells and number of transmembrane cells in the blank control group were 33.40±4.506% and (34.40±5.177)/HP, respectively, while the values in the experimental group transfected with the LGR5 overexpression plasmid were 59.60±6.542% and (68.20±13.41)/HP, respectively. In the group of cells transfected with empty plasmids, the two values were 29.00±7.176% and (30.80±10.40)/HP, respectively (P<0.01) ( Figure 5A-D). As shown in Figure 7F and H, the relative mobility of cells in the empty plasmid-transfected group was 52.80±8.167% while that of cells stimulated only by the in ammatory microenvironment was 56.60±8.620%. Additionally, little difference was found between the groups. Nevertheless, when the expression of LGR5 was inhibited, the in ammatory microenvironment cannot enhance cell migration and invasion. As shown in Figure 5E LGR5 inhibits the phosphorylation of YAP.
The expression of LGR5 and p-YAP was detected by immuno uorescence and western blotting. The phosphorylation level of YAP was decreased by 2.31 times when LGR5 was overexpressed compared with that in the control group (P<0.001) ( Figure 6B,C). However, after the addition of siRNA to inhibit the high expression of LGR5 caused by the in ammatory microenvironment, the phosphorylation level of YAP was 2.49 times higher than that of cells treated with the in ammatory microenvironment alone (P<0.001) ( Figure 6B,D). The use of VP to increase the phosphorylation level of YAP did not affect LGR5 expression ( Figure 6A-D).
Regarding immuno uorescence analysis, after in ammatory microenvironment stimulation, siRNA transfection, and VP treatment, many cells died, and obvious nuclear fragmentation was observed ( Figure   6A-f,B-f). Two possible causes of cell death exist: 1. siRNA has a strong inhibitory effect on LGR5 expression, resulting in low LGR5 expression, which reduces the tolerance of cells to VP and ultimately cell death; 2. An unknown reaction occurred between VP and the transfection reagent, resulting in the formation of toxic substances, leading to cell death.

VP impairs the ability of FaDu cells to invade and migrate.
We conducted wound healing experiments, and the results showed that the migration of LGR5overexpressed FaDu cells after VP stimulation was weaker than that of the control group. At 24 hours, the relative mobility of cells in the control group was 131.0±14.53% while that in the VP experimental group was 40.33±4.163% (P<0.01; Figure 7A,B). Next, the Transwell invasion assay was performed. The invasive ability of LGR5-overexpressed FaDu cells after VP stimulation was predominantly diminished compared with that in the control group. The number of transmembrane cells in the experimental group with VP was (47.67±5.508)/HP while that in the control group was (70.67±3.786)/HP (Figure 7 C,D).

Discussion
High LGR5 expression has been previously reported to be closely related to tumor proliferation and invasion and has been detected in various malignancies. Studies on liver cancer, colorectal tumors, basal cell carcinoma, ovarian cancer, and other tumors [15] have shown that LGR5 promotes cell proliferation and invasion in tumor cells, and leads to tumor stem cell-like characteristics, resulting in a poor prognosis. Other studies have shown that LGR5 plays an inhibitory role in CRC progression in colorectal cancer [16]. LGR5 expression in different tumors is different while that in hypopharyngeal cancer has not been studied. According to our study, LGR5 expression in HSCC is higher than that in normal tissues surrounding the tumor (Figure 1).
A large number of in ammatory factors, such as interleukin (IL)-1, IL-6, and TNF-α, exists in the in ammatory microenvironment. Studies have shown that these cytokines are important factors that promote the generation, proliferation, and invasion of various tumor cells [17,18]. In the IL-1 study, wildtype and IL-1β-knockout (KO) mice were inoculated with B16 melanoma cells. The IL-1β-knockout mice did not develop tumors, but the wild-type mice developed melanoma, which caused deaths [18]. Thus, the in ammatory microenvironment plays a key role in the occurrence and development of malignant tumors. In this study, LPS was used to stimulate THP-1 cells to produce in ammatory factors to simulate the in ammatory microenvironment. The proliferation and invasion of FaDu cells were enhanced in the in ammatory microenvironment. After the in ammatory microenvironment was used to stimulate FaDu cells, LGR5 expression was detected by western blotting. After 12 hours of stimulation, LGR5 expression was signi cantly up-regulated and did not continue to increase over time ( Figure 3C,D). However, when siRNA was used to inhibit the high LGR5 expression in FaDu cells caused by in ammatory microenvironmental stimulation, the cell viability was signi cantly reduced, and the ability of proliferation and invasion was weakened. Therefore, the in ammatory microenvironment likely causes high LGR5 expression in HSCC, thus promoting the proliferation and invasion of tumor cells.
We con rmed through scratch and Transwell experiment that high LGR5 expression can promote the migration and invasion of tumor cells. Subsequently, we detected the expression of various EMT-related proteins, among which the expression levels of vimentin, twist, and snail were regulated by LGR5. However, E-cadherin and N-cadherin [19], which play key roles in the EMT of many tumors, were found not to be regulated by LGR5 after multiple tests in the HSCC we studied. No signi cant changes were observed in the expression of E-cadherin and N-cadherin regardless of LGR5 overexpression or inhibition (data not shown). Therefore, we believe that, in HSCC, LGR5 regulates the cell EMT process through vimentin, twist, and snail, rather than by involving E-cadherin and N-cadherin, as in most cases.
Previous studies have shown that the YAP-TEAD interaction is an important factor that promotes the proliferation, migration, and invasion of tumor cells [20,21]. Many studies have demonstrated that the hyperfunction of YAP and TAZ may promote the proliferation of tumor cells [22] and the process of EMT, which includes high expression of YAP/TAZ or nuclear enrichment, has been detected in melanoma, breast cancer, and other tumors [23]. YAP can be transported to the nucleus and bind with transcription factors such as TEADs to promote the expression of target genes [20]. The YAP/TAZ pathway is associated with the expression of many genes. For example, studies have shown that VASN (vasorin) in thyroid cancer promotes the EMT of thyroid cancer cells by triggering the YAP/TAZ pathway [24]. TRPV4 (transient receptor potential cation channel subfamily V member 4) was found to be very important for the nuclear transport of YAP/TAZ. By regulating YAP/TAZ nuclear translocation, TRPV4 promotes EMT in normal mouse primary epidermal keratinocytes, resulting in changes in the expression levels of EMTrelated markers, such as E cadherin, N cadherin, and α-smooth muscle actin [25,26]. Therefore, YAP is regulated by some genes and the expression of some genes. We considered whether a relationship exists between LGR5 and YAP-TEAD. Additionally, studies have shown that speci c siRNA silencing of YAP in the human colon cancer cell line HCT116 can affect the recruitment of p300 protein, reduce the acetylation of p53AIP1 target genomic protein, and lead to the delay or reduction of p73-mediated apoptosis [27]. Thus, YAP has different effects in different tumors. The tumor we studied was hypopharyngeal squamous carcinoma. Using various experiments, we proved that the in ammatory microenvironment-induced high LGR5 expression in FaDu cells and LGR5 overexpression lead to the hyperfunction of YAP/TAZ and enhanced proliferation, migration, and invasion of tumor cells.
YAP can be phosphorylated by VP, and phosphorylated YAP binds to proteins and remains in the cytoplasm [28,29]. Therefore, the destruction of YAP/TAZ nuclear transport by VP can inhibit the EMT process to a certain extent[30], a nding that was also con rmed in this study. In our study, high LGR5 expression could lead to the reduction of p-YAP, and the expression level of EMT-related protein was upregulated ( Figure 4B). The use of VP to increase the phosphorylation level of YAP had little effect on LGR5 expression, but the invasion ability of tumor cells was signi cantly decreased ( Figure 6A). Therefore, we believe that the in ammatory microenvironment causes high LGR5 expression, which will inhibit the phosphorylation of YAP, resulting in markedly increased YAP-TEAD binding, high expression of EMT-related proteins, and enhanced invasion and metastasis ability of tumor cells. Low expression of the YAP gene by shRNA [31,32] and interference of the formation of the YAP-TEAD complex by VP [33,34] have been reported to induce apoptosis. In this study, VP was also used to phosphorylate YAP, but many cells died after siRNA inhibition of LGR5 expression; however, few of the cells died in the LGR5 overexpression group ( Figure 6A-f,B-f). Therefore, we hypothesized that LGR5 expression may lead to apoptosis resistance caused by the functional inhibition of YAP.

Conclusions
In conclusion, we have demonstrated for the rst time that the in ammatory microenvironment can lead to high LGR5 expression in HSCC, and high LGR5 expression can promote the proliferation and invasion of tumor cells by inhibiting the phosphorylation of YAP. Therefore, LGR5, YAP, and the relationship between LGR5 and YAP phosphorylation can be used as targets for the treatment of HSCC and have potential clinical value in prognosis.    Figure 1 Immunohistochemical staining of LGR5 in HSCC and normal oropharyngeal mucosa tissues.
LGR5 expression was lower in normal mucosal tissues than in tumor tissues.
LGR5 is mainly expressed in the cytoplasm of tumor cells. A, B: Scale bar=20 μm.   LGR5 siRNA.  Relationship between the phosphorylation of YAP and LGR5 expression. A: LGR5 expression was detected by immuno uorescence assay. FaDu cells were subjected to immuno uorescent staining for Hoechst 33342 (blue) and LGR5 (red). The cells were subjected to various treatments: transfection with LGR5 plasmids or stimulation with the in ammatory microenvironment; inhibition of LGR5 expression by siRNA; inhibition of YAP phosphorylation by VP. The experimental results of immuno uorescence are shown in Figure 6A. Scale bar=20 μm. 6B: The expression of p-YAP was detected by immuno uorescence assay. FaDu cells were subjected to immuno uorescent staining for Hoechst 33342 (blue) and phosphorylated YAP (red). C: Western blotting was used to detect LGR5 or p-YAP expression in LGR5overexpressed cells or siRNA/LGR5-transfected FaDu cells. Statistical signi cance is denoted by**P<0.01,***P<0.001. UT: untreated group; FC: in ammatory microenvironment; VP: vertepor n; LGR5+: stable cell lines overexpressing LGR5; siRNA/NC: negative control siRNA; siRNA/LGR5: LGR5 siRNA.